From cf87dae47dbe2e2d965805babc53243c49da9e55 Mon Sep 17 00:00:00 2001
From: Nicolas Hug <contact@nicolas-hug.com>
Date: Wed, 4 Oct 2023 17:38:39 +0100
Subject: [PATCH 1/5] Copy docs from
 https://github.com/pytorch/vision/pull/7994

---
 .../auto_examples_python.zip                  |   Bin 0 -> 102610 bytes
 .../plot_custom_transforms.py                 |   121 +
 .../plot_video_api.py                         |   346 +
 .../plot_scripted_tensor_transforms.py        |   136 +
 .../plot_repurposing_annotations.ipynb        |   205 +
 .../plot_visualization_utils.py               |   462 +
 .../plot_custom_transforms.ipynb              |    97 +
 .../plot_transforms_e2e.py                    |   181 +
 .../plot_scripted_tensor_transforms.ipynb     |   133 +
 .../plot_custom_tv_tensors.py                 |   119 +
 .../plot_video_api.ipynb                      |   298 +
 .../plot_cutmix_mixup.py                      |   150 +
 .../plot_repurposing_annotations.py           |   211 +
 .../plot_cutmix_mixup.ipynb                   |   133 +
 .../plot_optical_flow.py                      |   198 +
 .../auto_examples_jupyter.zip                 |   Bin 0 -> 144297 bytes
 .../plot_custom_tv_tensors.ipynb              |   140 +
 .../plot_optical_flow.ipynb                   |   176 +
 .../plot_transforms_getting_started.py        |   266 +
 .../plot_tv_tensors.py                        |   224 +
 .../plot_visualization_utils.ipynb            |   464 +
 .../plot_transforms_illustrations.ipynb       |   511 +
 .../plot_tv_tensors.ipynb                     |   237 +
 .../plot_transforms_getting_started.ipynb     |   154 +
 .../plot_transforms_illustrations.py          |   320 +
 .../plot_transforms_e2e.ipynb                 |   158 +
 .../plot_custom_transforms.ipynb              |    97 +
 .../plot_custom_tv_tensors.ipynb              |   140 +
 .../plot_cutmix_mixup.ipynb                   |   133 +
 .../plot_optical_flow.ipynb                   |   176 +
 .../plot_repurposing_annotations.ipynb        |   205 +
 .../plot_scripted_tensor_transforms.ipynb     |   133 +
 .../plot_transforms_e2e.ipynb                 |   158 +
 .../plot_transforms_getting_started.ipynb     |   154 +
 .../plot_transforms_illustrations.ipynb       |   511 +
 .../plot_tv_tensors.ipynb                     |   237 +
 .../plot_video_api.ipynb                      |   298 +
 .../plot_visualization_utils.ipynb            |   464 +
 .../sphx_glr_plot_custom_transforms_thumb.png |   Bin 0 -> 26794 bytes
 .../sphx_glr_plot_custom_tv_tensors_thumb.png |   Bin 0 -> 26794 bytes
 .../sphx_glr_plot_cutmix_mixup_thumb.png      |   Bin 0 -> 26794 bytes
 .../sphx_glr_plot_optical_flow_001.png        |   Bin 0 -> 150277 bytes
 .../sphx_glr_plot_optical_flow_002.png        |   Bin 0 -> 200828 bytes
 .../sphx_glr_plot_optical_flow_thumb.png      |   Bin 0 -> 94362 bytes
 ...x_glr_plot_repurposing_annotations_001.png |   Bin 0 -> 45288 bytes
 ...x_glr_plot_repurposing_annotations_002.png |   Bin 0 -> 208073 bytes
 ...glr_plot_repurposing_annotations_thumb.png |   Bin 0 -> 105726 bytes
 ...lr_plot_scripted_tensor_transforms_001.png |   Bin 0 -> 171473 bytes
 ..._plot_scripted_tensor_transforms_thumb.png |   Bin 0 -> 77760 bytes
 .../sphx_glr_plot_transforms_e2e_001.png      |   Bin 0 -> 278328 bytes
 .../sphx_glr_plot_transforms_e2e_002.png      |   Bin 0 -> 205620 bytes
 .../sphx_glr_plot_transforms_e2e_thumb.png    |   Bin 0 -> 85362 bytes
 ...lr_plot_transforms_getting_started_001.png |   Bin 0 -> 274415 bytes
 ...lr_plot_transforms_getting_started_002.png |   Bin 0 -> 285810 bytes
 ...lr_plot_transforms_getting_started_003.png |   Bin 0 -> 284637 bytes
 ...lr_plot_transforms_getting_started_004.png |   Bin 0 -> 264577 bytes
 ..._plot_transforms_getting_started_thumb.png |   Bin 0 -> 127385 bytes
 ..._glr_plot_transforms_illustrations_001.png |   Bin 0 -> 120009 bytes
 ..._glr_plot_transforms_illustrations_002.png |   Bin 0 -> 79983 bytes
 ..._glr_plot_transforms_illustrations_003.png |   Bin 0 -> 71477 bytes
 ..._glr_plot_transforms_illustrations_004.png |   Bin 0 -> 80624 bytes
 ..._glr_plot_transforms_illustrations_005.png |   Bin 0 -> 92704 bytes
 ..._glr_plot_transforms_illustrations_006.png |   Bin 0 -> 127303 bytes
 ..._glr_plot_transforms_illustrations_007.png |   Bin 0 -> 80795 bytes
 ..._glr_plot_transforms_illustrations_008.png |   Bin 0 -> 222987 bytes
 ..._glr_plot_transforms_illustrations_009.png |   Bin 0 -> 110388 bytes
 ..._glr_plot_transforms_illustrations_010.png |   Bin 0 -> 45718 bytes
 ..._glr_plot_transforms_illustrations_011.png |   Bin 0 -> 251321 bytes
 ..._glr_plot_transforms_illustrations_012.png |   Bin 0 -> 126316 bytes
 ..._glr_plot_transforms_illustrations_013.png |   Bin 0 -> 109936 bytes
 ..._glr_plot_transforms_illustrations_014.png |   Bin 0 -> 55859 bytes
 ..._glr_plot_transforms_illustrations_015.png |   Bin 0 -> 51565 bytes
 ..._glr_plot_transforms_illustrations_016.png |   Bin 0 -> 48231 bytes
 ..._glr_plot_transforms_illustrations_017.png |   Bin 0 -> 29575 bytes
 ..._glr_plot_transforms_illustrations_018.png |   Bin 0 -> 29575 bytes
 ..._glr_plot_transforms_illustrations_019.png |   Bin 0 -> 56979 bytes
 ..._glr_plot_transforms_illustrations_020.png |   Bin 0 -> 346468 bytes
 ..._glr_plot_transforms_illustrations_021.png |   Bin 0 -> 127886 bytes
 ..._glr_plot_transforms_illustrations_022.png |   Bin 0 -> 117059 bytes
 ..._glr_plot_transforms_illustrations_023.png |   Bin 0 -> 123801 bytes
 ..._glr_plot_transforms_illustrations_024.png |   Bin 0 -> 55715 bytes
 ..._glr_plot_transforms_illustrations_025.png |   Bin 0 -> 55869 bytes
 ..._glr_plot_transforms_illustrations_026.png |   Bin 0 -> 66340 bytes
 ...lr_plot_transforms_illustrations_thumb.png |   Bin 0 -> 205458 bytes
 .../sphx_glr_plot_tv_tensors_thumb.png        |   Bin 0 -> 26794 bytes
 0.16/_images/sphx_glr_plot_video_api_001.png  |   Bin 0 -> 894322 bytes
 .../_images/sphx_glr_plot_video_api_thumb.png |   Bin 0 -> 84921 bytes
 .../sphx_glr_plot_visualization_utils_001.png |   Bin 0 -> 239651 bytes
 .../sphx_glr_plot_visualization_utils_002.png |   Bin 0 -> 247376 bytes
 .../sphx_glr_plot_visualization_utils_003.png |   Bin 0 -> 202016 bytes
 .../sphx_glr_plot_visualization_utils_004.png |   Bin 0 -> 9826 bytes
 .../sphx_glr_plot_visualization_utils_005.png |   Bin 0 -> 9191 bytes
 .../sphx_glr_plot_visualization_utils_006.png |   Bin 0 -> 192949 bytes
 .../sphx_glr_plot_visualization_utils_007.png |   Bin 0 -> 185983 bytes
 .../sphx_glr_plot_visualization_utils_008.png |   Bin 0 -> 153833 bytes
 .../sphx_glr_plot_visualization_utils_009.png |   Bin 0 -> 242568 bytes
 .../sphx_glr_plot_visualization_utils_010.png |   Bin 0 -> 189893 bytes
 .../sphx_glr_plot_visualization_utils_011.png |   Bin 0 -> 144896 bytes
 .../sphx_glr_plot_visualization_utils_012.png |   Bin 0 -> 146862 bytes
 ...phx_glr_plot_visualization_utils_thumb.png |   Bin 0 -> 61829 bytes
 0.16/_modules/index.html                      |   800 +
 0.16/_modules/torchvision.html                |   770 +
 .../torchvision/datasets/_optical_flow.html   |  1160 ++
 .../datasets/_stereo_matching.html            |  1893 +++
 .../torchvision/datasets/caltech.html         |   906 ++
 .../_modules/torchvision/datasets/celeba.html |   858 ++
 0.16/_modules/torchvision/datasets/cifar.html |   836 ++
 .../torchvision/datasets/cityscapes.html      |   890 ++
 0.16/_modules/torchvision/datasets/coco.html  |   773 +
 .../torchvision/datasets/country211.html      |   727 +
 0.16/_modules/torchvision/datasets/dtd.html   |   769 +
 .../torchvision/datasets/eurosat.html         |   727 +
 .../torchvision/datasets/fakedata.html        |   736 +
 .../torchvision/datasets/fer2013.html         |   744 +
 .../torchvision/datasets/fgvc_aircraft.html   |   783 +
 .../_modules/torchvision/datasets/flickr.html |   835 ++
 .../torchvision/datasets/flowers102.html      |   783 +
 .../_modules/torchvision/datasets/folder.html |   986 ++
 .../torchvision/datasets/food101.html         |   762 +
 0.16/_modules/torchvision/datasets/gtsrb.html |   772 +
 .../_modules/torchvision/datasets/hmdb51.html |   820 +
 .../torchvision/datasets/imagenet.html        |   887 ++
 .../torchvision/datasets/inaturalist.html     |   910 ++
 .../torchvision/datasets/kinetics.html        |   916 ++
 0.16/_modules/torchvision/datasets/kitti.html |   827 +
 0.16/_modules/torchvision/datasets/lfw.html   |   924 ++
 0.16/_modules/torchvision/datasets/lsun.html  |   836 ++
 0.16/_modules/torchvision/datasets/mnist.html |  1217 ++
 .../torchvision/datasets/moving_mnist.html    |   762 +
 .../torchvision/datasets/omniglot.html        |   771 +
 .../torchvision/datasets/oxford_iiit_pet.html |   794 +
 0.16/_modules/torchvision/datasets/pcam.html  |   799 +
 .../torchvision/datasets/phototour.html       |   897 ++
 .../torchvision/datasets/places365.html       |   839 ++
 .../torchvision/datasets/rendered_sst2.html   |   755 +
 0.16/_modules/torchvision/datasets/sbd.html   |   792 +
 0.16/_modules/torchvision/datasets/sbu.html   |   778 +
 .../torchvision/datasets/semeion.html         |   760 +
 .../torchvision/datasets/stanford_cars.html   |   790 +
 0.16/_modules/torchvision/datasets/stl10.html |   843 ++
 .../_modules/torchvision/datasets/sun397.html |   745 +
 0.16/_modules/torchvision/datasets/svhn.html  |   798 +
 .../_modules/torchvision/datasets/ucf101.html |   799 +
 0.16/_modules/torchvision/datasets/usps.html  |   764 +
 .../_modules/torchvision/datasets/vision.html |   779 +
 0.16/_modules/torchvision/datasets/voc.html   |   893 ++
 .../torchvision/datasets/widerface.html       |   860 ++
 0.16/_modules/torchvision/io/image.html       |   933 ++
 0.16/_modules/torchvision/io/video.html       |  1084 ++
 .../_modules/torchvision/io/video_reader.html |   967 ++
 0.16/_modules/torchvision/models/_api.html    |   946 ++
 0.16/_modules/torchvision/models/alexnet.html |   788 +
 .../_modules/torchvision/models/convnext.html |  1083 ++
 .../_modules/torchvision/models/densenet.html |  1117 ++
 .../models/detection/faster_rcnn.html         |  1512 ++
 .../torchvision/models/detection/fcos.html    |  1440 ++
 .../models/detection/keypoint_rcnn.html       |  1141 ++
 .../models/detection/mask_rcnn.html           |  1256 ++
 .../models/detection/retinanet.html           |  1568 ++
 .../torchvision/models/detection/ssd.html     |  1351 ++
 .../torchvision/models/detection/ssdlite.html |  1000 ++
 .../torchvision/models/efficientnet.html      |  1800 +++
 .../models/feature_extraction.html            |  1232 ++
 .../torchvision/models/googlenet.html         |  1014 ++
 .../torchvision/models/inception.html         |  1147 ++
 0.16/_modules/torchvision/models/maxvit.html  |  1501 ++
 0.16/_modules/torchvision/models/mnasnet.html |  1103 ++
 .../torchvision/models/mobilenetv2.html       |   929 ++
 .../torchvision/models/mobilenetv3.html       |  1092 ++
 .../torchvision/models/optical_flow/raft.html |  1616 ++
 .../models/quantization/googlenet.html        |   879 ++
 .../models/quantization/inception.html        |   942 ++
 .../models/quantization/mobilenetv2.html      |   823 +
 .../models/quantization/mobilenetv3.html      |   906 ++
 .../models/quantization/resnet.html           |  1153 ++
 .../models/quantization/shufflenetv2.html     |  1096 ++
 0.16/_modules/torchvision/models/regnet.html  |  2240 +++
 0.16/_modules/torchvision/models/resnet.html  |  1654 ++
 .../models/segmentation/deeplabv3.html        |  1059 ++
 .../torchvision/models/segmentation/fcn.html  |   901 ++
 .../models/segmentation/lraspp.html           |   847 ++
 .../torchvision/models/shufflenetv2.html      |  1077 ++
 .../torchvision/models/squeezenet.html        |   892 ++
 .../torchvision/models/swin_transformer.html  |  1702 +++
 0.16/_modules/torchvision/models/vgg.html     |  1180 ++
 .../torchvision/models/video/mvit.html        |  1566 ++
 .../torchvision/models/video/resnet.html      |  1172 ++
 .../torchvision/models/video/s3d.html         |   888 ++
 .../models/video/swin_transformer.html        |  1412 ++
 .../models/vision_transformer.html            |  1533 ++
 0.16/_modules/torchvision/ops/boxes.html      |  1086 ++
 0.16/_modules/torchvision/ops/ciou_loss.html  |   747 +
 .../_modules/torchvision/ops/deform_conv.html |   864 ++
 0.16/_modules/torchvision/ops/diou_loss.html  |   763 +
 0.16/_modules/torchvision/ops/drop_block.html |   824 +
 .../ops/feature_pyramid_network.html          |   919 ++
 0.16/_modules/torchvision/ops/focal_loss.html |   727 +
 0.16/_modules/torchvision/ops/giou_loss.html  |   745 +
 0.16/_modules/torchvision/ops/misc.html       |   989 ++
 0.16/_modules/torchvision/ops/poolers.html    |   996 ++
 .../torchvision/ops/ps_roi_align.html         |   759 +
 .../_modules/torchvision/ops/ps_roi_pool.html |   739 +
 0.16/_modules/torchvision/ops/roi_align.html  |   943 ++
 0.16/_modules/torchvision/ops/roi_pool.html   |   741 +
 .../torchvision/ops/stochastic_depth.html     |   735 +
 .../torchvision/transforms/autoaugment.html   |  1284 ++
 .../torchvision/transforms/functional.html    |  2286 +++
 .../torchvision/transforms/transforms.html    |  2829 ++++
 .../torchvision/transforms/v2/_augment.html   |   994 ++
 .../transforms/v2/_auto_augment.html          |  1304 ++
 .../torchvision/transforms/v2/_color.html     |  1049 ++
 .../torchvision/transforms/v2/_container.html |   849 ++
 .../transforms/v2/_deprecated.html            |   719 +
 .../torchvision/transforms/v2/_geometry.html  |  2116 +++
 .../torchvision/transforms/v2/_meta.html      |   711 +
 .../torchvision/transforms/v2/_misc.html      |  1090 ++
 .../torchvision/transforms/v2/_temporal.html  |   697 +
 .../transforms/v2/_type_conversion.html       |   761 +
 .../transforms/v2/functional/_augment.html    |   724 +
 .../transforms/v2/functional/_color.html      |  1378 ++
 .../transforms/v2/functional/_deprecated.html |   693 +
 .../transforms/v2/functional/_geometry.html   |  3049 ++++
 .../transforms/v2/functional/_meta.html       |   943 ++
 .../transforms/v2/functional/_misc.html       |   946 ++
 .../transforms/v2/functional/_temporal.html   |   696 +
 .../v2/functional/_type_conversion.html       |   694 +
 .../transforms/v2/functional/_utils.html      |   810 +
 0.16/_modules/torchvision/tv_tensors.html     |   700 +
 .../tv_tensors/_bounding_boxes.html           |   774 +
 .../tv_tensors/_dataset_wrapper.html          |  1323 ++
 .../torchvision/tv_tensors/_image.html        |   724 +
 .../torchvision/tv_tensors/_mask.html         |   710 +
 .../tv_tensors/_torch_function_helpers.html   |   743 +
 .../torchvision/tv_tensors/_tv_tensor.html    |   803 +
 .../torchvision/tv_tensors/_video.html        |   708 +
 0.16/_modules/torchvision/utils.html          |  1266 ++
 0.16/_sources/auto_examples/index.rst.txt     |   279 +
 .../auto_examples/others/index.rst.txt        |   113 +
 .../others/plot_optical_flow.rst.txt          |   380 +
 .../plot_repurposing_annotations.rst.txt      |   445 +
 .../plot_scripted_tensor_transforms.rst.txt   |   267 +
 .../others/plot_video_api.rst.txt             |   637 +
 .../others/plot_visualization_utils.rst.txt   |  1092 ++
 .../others/sg_execution_times.rst.txt         |    21 +
 .../auto_examples/transforms/index.rst.txt    |   151 +
 .../transforms/plot_custom_transforms.rst.txt |   233 +
 .../transforms/plot_custom_tv_tensors.rst.txt |   253 +
 .../transforms/plot_cutmix_mixup.rst.txt      |   291 +
 .../transforms/plot_transforms_e2e.rst.txt    |   364 +
 .../plot_transforms_getting_started.rst.txt   |   428 +
 .../plot_transforms_illustrations.rst.txt     |   837 ++
 .../transforms/plot_tv_tensors.rst.txt        |   439 +
 .../transforms/sg_execution_times.rst.txt     |    25 +
 0.16/_sources/datasets.rst.txt                |   182 +
 0.16/_sources/feature_extraction.rst.txt      |   166 +
 .../classification_quant_table.rst.txt        |    23 +
 .../generated/classification_table.rst.txt    |   124 +
 .../detection_keypoint_table.rst.txt          |    11 +
 .../generated/detection_table.rst.txt         |    18 +
 .../instance_segmentation_table.rst.txt       |    11 +
 .../generated/segmentation_table.rst.txt      |    15 +
 .../torchvision.datasets.CIFAR10.rst.txt      |    13 +
 .../torchvision.datasets.CIFAR100.rst.txt     |    13 +
 .../torchvision.datasets.CREStereo.rst.txt    |    13 +
 .../torchvision.datasets.Caltech101.rst.txt   |    13 +
 .../torchvision.datasets.Caltech256.rst.txt   |    13 +
 .../torchvision.datasets.CarlaStereo.rst.txt  |    13 +
 .../torchvision.datasets.CelebA.rst.txt       |    13 +
 .../torchvision.datasets.Cityscapes.rst.txt   |    13 +
 .../torchvision.datasets.CocoCaptions.rst.txt |    13 +
 ...torchvision.datasets.CocoDetection.rst.txt |    13 +
 .../torchvision.datasets.Country211.rst.txt   |    13 +
 .../torchvision.datasets.DTD.rst.txt          |    13 +
 ...torchvision.datasets.DatasetFolder.rst.txt |    10 +
 .../torchvision.datasets.EMNIST.rst.txt       |    13 +
 .../torchvision.datasets.ETH3DStereo.rst.txt  |    13 +
 .../torchvision.datasets.EuroSAT.rst.txt      |    13 +
 .../torchvision.datasets.FER2013.rst.txt      |    13 +
 .../torchvision.datasets.FGVCAircraft.rst.txt |    13 +
 .../torchvision.datasets.FakeData.rst.txt     |    13 +
 ...ision.datasets.FallingThingsStereo.rst.txt |    13 +
 .../torchvision.datasets.FashionMNIST.rst.txt |    13 +
 .../torchvision.datasets.Flickr30k.rst.txt    |    13 +
 .../torchvision.datasets.Flickr8k.rst.txt     |    13 +
 .../torchvision.datasets.Flowers102.rst.txt   |    13 +
 .../torchvision.datasets.FlyingChairs.rst.txt |    13 +
 ...orchvision.datasets.FlyingThings3D.rst.txt |    13 +
 .../torchvision.datasets.Food101.rst.txt      |    13 +
 .../torchvision.datasets.GTSRB.rst.txt        |    13 +
 .../torchvision.datasets.HD1K.rst.txt         |    13 +
 .../torchvision.datasets.HMDB51.rst.txt       |    13 +
 .../torchvision.datasets.INaturalist.rst.txt  |    13 +
 .../torchvision.datasets.ImageFolder.rst.txt  |    10 +
 .../torchvision.datasets.ImageNet.rst.txt     |    13 +
 .../torchvision.datasets.InStereo2k.rst.txt   |    13 +
 .../torchvision.datasets.KMNIST.rst.txt       |    13 +
 .../torchvision.datasets.Kinetics.rst.txt     |    13 +
 .../torchvision.datasets.Kitti.rst.txt        |    13 +
 ...rchvision.datasets.Kitti2012Stereo.rst.txt |    13 +
 ...rchvision.datasets.Kitti2015Stereo.rst.txt |    13 +
 .../torchvision.datasets.KittiFlow.rst.txt    |    13 +
 .../torchvision.datasets.LFWPairs.rst.txt     |    13 +
 .../torchvision.datasets.LFWPeople.rst.txt    |    13 +
 .../torchvision.datasets.LSUN.rst.txt         |    13 +
 .../torchvision.datasets.MNIST.rst.txt        |    13 +
 ...sion.datasets.Middlebury2014Stereo.rst.txt |    13 +
 .../torchvision.datasets.MovingMNIST.rst.txt  |    13 +
 .../torchvision.datasets.Omniglot.rst.txt     |    13 +
 ...torchvision.datasets.OxfordIIITPet.rst.txt |    13 +
 .../torchvision.datasets.PCAM.rst.txt         |    13 +
 .../torchvision.datasets.PhotoTour.rst.txt    |    13 +
 .../torchvision.datasets.Places365.rst.txt    |    13 +
 .../torchvision.datasets.QMNIST.rst.txt       |    13 +
 .../torchvision.datasets.RenderedSST2.rst.txt |    13 +
 .../torchvision.datasets.SBDataset.rst.txt    |    13 +
 .../torchvision.datasets.SBU.rst.txt          |    13 +
 .../torchvision.datasets.SEMEION.rst.txt      |    13 +
 .../torchvision.datasets.STL10.rst.txt        |    13 +
 .../torchvision.datasets.SUN397.rst.txt       |    13 +
 .../torchvision.datasets.SVHN.rst.txt         |    13 +
 ...rchvision.datasets.SceneFlowStereo.rst.txt |    13 +
 .../torchvision.datasets.Sintel.rst.txt       |    13 +
 .../torchvision.datasets.SintelStereo.rst.txt |    13 +
 .../torchvision.datasets.StanfordCars.rst.txt |    13 +
 .../torchvision.datasets.UCF101.rst.txt       |    13 +
 .../torchvision.datasets.USPS.rst.txt         |    13 +
 .../torchvision.datasets.VOCDetection.rst.txt |    13 +
 ...rchvision.datasets.VOCSegmentation.rst.txt |    13 +
 ...torchvision.datasets.VisionDataset.rst.txt |    10 +
 .../torchvision.datasets.WIDERFace.rst.txt    |    13 +
 ...ets.wrap_dataset_for_transforms_v2.rst.txt |     9 +
 .../torchvision.io.ImageReadMode.rst.txt      |    10 +
 .../torchvision.io.VideoReader.rst.txt        |    10 +
 .../torchvision.io.decode_image.rst.txt       |     9 +
 .../torchvision.io.decode_jpeg.rst.txt        |     9 +
 .../torchvision.io.decode_png.rst.txt         |     9 +
 .../torchvision.io.encode_jpeg.rst.txt        |     9 +
 .../torchvision.io.encode_png.rst.txt         |     9 +
 .../torchvision.io.read_file.rst.txt          |     9 +
 .../torchvision.io.read_image.rst.txt         |     9 +
 .../torchvision.io.read_video.rst.txt         |     9 +
 ...rchvision.io.read_video_timestamps.rst.txt |     9 +
 .../torchvision.io.write_file.rst.txt         |     9 +
 .../torchvision.io.write_jpeg.rst.txt         |     9 +
 .../torchvision.io.write_png.rst.txt          |     9 +
 .../torchvision.io.write_video.rst.txt        |     9 +
 ...xtraction.create_feature_extractor.rst.txt |     9 +
 ...re_extraction.get_graph_node_names.rst.txt |     9 +
 .../torchvision.models.get_model.rst.txt      |     9 +
 ...rchvision.models.get_model_weights.rst.txt |     9 +
 .../torchvision.models.get_weight.rst.txt     |     9 +
 .../torchvision.models.list_models.rst.txt    |     9 +
 ...rchvision.ops.Conv2dNormActivation.rst.txt |    10 +
 ...rchvision.ops.Conv3dNormActivation.rst.txt |    10 +
 .../torchvision.ops.DeformConv2d.rst.txt      |    10 +
 .../torchvision.ops.DropBlock2d.rst.txt       |    10 +
 .../torchvision.ops.DropBlock3d.rst.txt       |    10 +
 ...chvision.ops.FeaturePyramidNetwork.rst.txt |    10 +
 .../torchvision.ops.FrozenBatchNorm2d.rst.txt |    10 +
 .../generated/torchvision.ops.MLP.rst.txt     |    10 +
 ...torchvision.ops.MultiScaleRoIAlign.rst.txt |    10 +
 .../torchvision.ops.PSRoIAlign.rst.txt        |    10 +
 .../torchvision.ops.PSRoIPool.rst.txt         |    10 +
 .../generated/torchvision.ops.Permute.rst.txt |    10 +
 .../torchvision.ops.RoIAlign.rst.txt          |    10 +
 .../generated/torchvision.ops.RoIPool.rst.txt |    10 +
 .../torchvision.ops.SqueezeExcitation.rst.txt |    10 +
 .../torchvision.ops.StochasticDepth.rst.txt   |    10 +
 .../torchvision.ops.batched_nms.rst.txt       |     9 +
 .../torchvision.ops.box_area.rst.txt          |     9 +
 .../torchvision.ops.box_convert.rst.txt       |     9 +
 .../generated/torchvision.ops.box_iou.rst.txt |     9 +
 ...orchvision.ops.clip_boxes_to_image.rst.txt |     9 +
 .../torchvision.ops.complete_box_iou.rst.txt  |     9 +
 ...chvision.ops.complete_box_iou_loss.rst.txt |     9 +
 .../torchvision.ops.deform_conv2d.rst.txt     |     9 +
 .../torchvision.ops.distance_box_iou.rst.txt  |     9 +
 ...chvision.ops.distance_box_iou_loss.rst.txt |     9 +
 .../torchvision.ops.drop_block2d.rst.txt      |     9 +
 .../torchvision.ops.drop_block3d.rst.txt      |     9 +
 ...orchvision.ops.generalized_box_iou.rst.txt |     9 +
 ...ision.ops.generalized_box_iou_loss.rst.txt |     9 +
 .../torchvision.ops.masks_to_boxes.rst.txt    |     9 +
 .../generated/torchvision.ops.nms.rst.txt     |     9 +
 .../torchvision.ops.ps_roi_align.rst.txt      |     9 +
 .../torchvision.ops.ps_roi_pool.rst.txt       |     9 +
 ...torchvision.ops.remove_small_boxes.rst.txt |     9 +
 .../torchvision.ops.roi_align.rst.txt         |     9 +
 .../torchvision.ops.roi_pool.rst.txt          |     9 +
 ...torchvision.ops.sigmoid_focal_loss.rst.txt |     9 +
 .../torchvision.ops.stochastic_depth.rst.txt  |     9 +
 .../torchvision.transforms.AugMix.rst.txt     |    10 +
 ...torchvision.transforms.AutoAugment.rst.txt |    10 +
 ...ision.transforms.AutoAugmentPolicy.rst.txt |    10 +
 .../torchvision.transforms.CenterCrop.rst.txt |    10 +
 ...torchvision.transforms.ColorJitter.rst.txt |    10 +
 .../torchvision.transforms.Compose.rst.txt    |    10 +
 ...ision.transforms.ConvertImageDtype.rst.txt |    10 +
 ...vision.transforms.ElasticTransform.rst.txt |    10 +
 .../torchvision.transforms.FiveCrop.rst.txt   |    10 +
 ...orchvision.transforms.GaussianBlur.rst.txt |    10 +
 .../torchvision.transforms.Grayscale.rst.txt  |    10 +
 .../torchvision.transforms.Lambda.rst.txt     |    10 +
 ...on.transforms.LinearTransformation.rst.txt |    10 +
 .../torchvision.transforms.Normalize.rst.txt  |    10 +
 ...torchvision.transforms.PILToTensor.rst.txt |    10 +
 .../torchvision.transforms.Pad.rst.txt        |    10 +
 ...torchvision.transforms.RandAugment.rst.txt |    10 +
 ...n.transforms.RandomAdjustSharpness.rst.txt |    10 +
 ...orchvision.transforms.RandomAffine.rst.txt |    10 +
 ...torchvision.transforms.RandomApply.rst.txt |    10 +
 ...sion.transforms.RandomAutocontrast.rst.txt |    10 +
 ...orchvision.transforms.RandomChoice.rst.txt |    10 +
 .../torchvision.transforms.RandomCrop.rst.txt |    10 +
 ...chvision.transforms.RandomEqualize.rst.txt |    10 +
 ...rchvision.transforms.RandomErasing.rst.txt |    10 +
 ...hvision.transforms.RandomGrayscale.rst.txt |    10 +
 ...on.transforms.RandomHorizontalFlip.rst.txt |    10 +
 ...orchvision.transforms.RandomInvert.rst.txt |    10 +
 ...torchvision.transforms.RandomOrder.rst.txt |    10 +
 ...ision.transforms.RandomPerspective.rst.txt |    10 +
 ...hvision.transforms.RandomPosterize.rst.txt |    10 +
 ...ision.transforms.RandomResizedCrop.rst.txt |    10 +
 ...chvision.transforms.RandomRotation.rst.txt |    10 +
 ...chvision.transforms.RandomSolarize.rst.txt |    10 +
 ...sion.transforms.RandomVerticalFlip.rst.txt |    10 +
 .../torchvision.transforms.Resize.rst.txt     |    10 +
 .../torchvision.transforms.TenCrop.rst.txt    |    10 +
 .../torchvision.transforms.ToPILImage.rst.txt |    10 +
 .../torchvision.transforms.ToTensor.rst.txt   |    10 +
 ...sion.transforms.TrivialAugmentWide.rst.txt |    10 +
 ...forms.functional.adjust_brightness.rst.txt |     9 +
 ...nsforms.functional.adjust_contrast.rst.txt |     9 +
 ...transforms.functional.adjust_gamma.rst.txt |     9 +
 ...n.transforms.functional.adjust_hue.rst.txt |     9 +
 ...forms.functional.adjust_saturation.rst.txt |     9 +
 ...sforms.functional.adjust_sharpness.rst.txt |     9 +
 ...ision.transforms.functional.affine.rst.txt |     9 +
 ...transforms.functional.autocontrast.rst.txt |     9 +
 ....transforms.functional.center_crop.rst.txt |     9 +
 ...rms.functional.convert_image_dtype.rst.txt |     9 +
 ...hvision.transforms.functional.crop.rst.txt |     9 +
 ...ion.transforms.functional.equalize.rst.txt |     9 +
 ...vision.transforms.functional.erase.rst.txt |     9 +
 ...on.transforms.functional.five_crop.rst.txt |     9 +
 ...ransforms.functional.gaussian_blur.rst.txt |     9 +
 ...ansforms.functional.get_dimensions.rst.txt |     9 +
 ....functional.get_image_num_channels.rst.txt |     9 +
 ...ansforms.functional.get_image_size.rst.txt |     9 +
 ...vision.transforms.functional.hflip.rst.txt |     9 +
 ...ision.transforms.functional.invert.rst.txt |     9 +
 ...on.transforms.functional.normalize.rst.txt |     9 +
 ...chvision.transforms.functional.pad.rst.txt |     9 +
 ....transforms.functional.perspective.rst.txt |     9 +
 ...ransforms.functional.pil_to_tensor.rst.txt |     9 +
 ...on.transforms.functional.posterize.rst.txt |     9 +
 ...ision.transforms.functional.resize.rst.txt |     9 +
 ...transforms.functional.resized_crop.rst.txt |     9 +
 ...sforms.functional.rgb_to_grayscale.rst.txt |     9 +
 ...ision.transforms.functional.rotate.rst.txt |     9 +
 ...ion.transforms.functional.solarize.rst.txt |     9 +
 ...ion.transforms.functional.ten_crop.rst.txt |     9 +
 ...transforms.functional.to_grayscale.rst.txt |     9 +
 ...transforms.functional.to_pil_image.rst.txt |     9 +
 ...on.transforms.functional.to_tensor.rst.txt |     9 +
 ...vision.transforms.functional.vflip.rst.txt |     9 +
 .../torchvision.transforms.v2.AugMix.rst.txt  |    10 +
 ...chvision.transforms.v2.AutoAugment.rst.txt |    10 +
 ...rchvision.transforms.v2.CenterCrop.rst.txt |    10 +
 ...n.transforms.v2.ClampBoundingBoxes.rst.txt |    10 +
 ...chvision.transforms.v2.ColorJitter.rst.txt |    10 +
 .../torchvision.transforms.v2.Compose.rst.txt |    10 +
 ...sforms.v2.ConvertBoundingBoxFormat.rst.txt |    10 +
 ...on.transforms.v2.ConvertImageDtype.rst.txt |    10 +
 .../torchvision.transforms.v2.CutMix.rst.txt  |    10 +
 ...ion.transforms.v2.ElasticTransform.rst.txt |    10 +
 ...torchvision.transforms.v2.FiveCrop.rst.txt |    10 +
 ...hvision.transforms.v2.GaussianBlur.rst.txt |    10 +
 ...orchvision.transforms.v2.Grayscale.rst.txt |    10 +
 .../torchvision.transforms.v2.Lambda.rst.txt  |    10 +
 ...transforms.v2.LinearTransformation.rst.txt |    10 +
 .../torchvision.transforms.v2.MixUp.rst.txt   |    10 +
 ...orchvision.transforms.v2.Normalize.rst.txt |    10 +
 ...chvision.transforms.v2.PILToTensor.rst.txt |    10 +
 .../torchvision.transforms.v2.Pad.rst.txt     |    10 +
 ...chvision.transforms.v2.RandAugment.rst.txt |    10 +
 ...ransforms.v2.RandomAdjustSharpness.rst.txt |    10 +
 ...hvision.transforms.v2.RandomAffine.rst.txt |    10 +
 ...chvision.transforms.v2.RandomApply.rst.txt |    10 +
 ...n.transforms.v2.RandomAutocontrast.rst.txt |    10 +
 ...sforms.v2.RandomChannelPermutation.rst.txt |    10 +
 ...hvision.transforms.v2.RandomChoice.rst.txt |    10 +
 ...rchvision.transforms.v2.RandomCrop.rst.txt |    10 +
 ...ision.transforms.v2.RandomEqualize.rst.txt |    10 +
 ...vision.transforms.v2.RandomErasing.rst.txt |    10 +
 ...sion.transforms.v2.RandomGrayscale.rst.txt |    10 +
 ...transforms.v2.RandomHorizontalFlip.rst.txt |    10 +
 ...hvision.transforms.v2.RandomInvert.rst.txt |    10 +
 ...vision.transforms.v2.RandomIoUCrop.rst.txt |    10 +
 ...chvision.transforms.v2.RandomOrder.rst.txt |    10 +
 ...on.transforms.v2.RandomPerspective.rst.txt |    10 +
 ...sforms.v2.RandomPhotometricDistort.rst.txt |    10 +
 ...sion.transforms.v2.RandomPosterize.rst.txt |    10 +
 ...hvision.transforms.v2.RandomResize.rst.txt |    10 +
 ...on.transforms.v2.RandomResizedCrop.rst.txt |    10 +
 ...ision.transforms.v2.RandomRotation.rst.txt |    10 +
 ...n.transforms.v2.RandomShortestSize.rst.txt |    10 +
 ...ision.transforms.v2.RandomSolarize.rst.txt |    10 +
 ...n.transforms.v2.RandomVerticalFlip.rst.txt |    10 +
 ...vision.transforms.v2.RandomZoomOut.rst.txt |    10 +
 .../torchvision.transforms.v2.Resize.rst.txt  |    10 +
 ...ransforms.v2.SanitizeBoundingBoxes.rst.txt |    10 +
 ...chvision.transforms.v2.ScaleJitter.rst.txt |    10 +
 .../torchvision.transforms.v2.TenCrop.rst.txt |    10 +
 .../torchvision.transforms.v2.ToDtype.rst.txt |    10 +
 .../torchvision.transforms.v2.ToImage.rst.txt |    10 +
 ...rchvision.transforms.v2.ToPILImage.rst.txt |    10 +
 ...hvision.transforms.v2.ToPureTensor.rst.txt |    10 +
 ...torchvision.transforms.v2.ToTensor.rst.txt |    10 +
 ...n.transforms.v2.TrivialAugmentWide.rst.txt |    10 +
 ...sforms.v2.UniformTemporalSubsample.rst.txt |    10 +
 ...ms.v2.functional.adjust_brightness.rst.txt |     9 +
 ...orms.v2.functional.adjust_contrast.rst.txt |     9 +
 ...nsforms.v2.functional.adjust_gamma.rst.txt |     9 +
 ...ransforms.v2.functional.adjust_hue.rst.txt |     9 +
 ...ms.v2.functional.adjust_saturation.rst.txt |     9 +
 ...rms.v2.functional.adjust_sharpness.rst.txt |     9 +
 ...on.transforms.v2.functional.affine.rst.txt |     9 +
 ...nsforms.v2.functional.autocontrast.rst.txt |     9 +
 ...ansforms.v2.functional.center_crop.rst.txt |     9 +
 ...v2.functional.clamp_bounding_boxes.rst.txt |     9 +
 ...tional.convert_bounding_box_format.rst.txt |     6 +
 ....v2.functional.convert_image_dtype.rst.txt |    10 +
 ...sion.transforms.v2.functional.crop.rst.txt |     9 +
 ...n.transforms.v2.functional.elastic.rst.txt |     9 +
 ....transforms.v2.functional.equalize.rst.txt |     9 +
 ...ion.transforms.v2.functional.erase.rst.txt |     9 +
 ...transforms.v2.functional.five_crop.rst.txt |     9 +
 ...sforms.v2.functional.gaussian_blur.rst.txt |     9 +
 ...orms.v2.functional.horizontal_flip.rst.txt |     9 +
 ...on.transforms.v2.functional.invert.rst.txt |     9 +
 ...transforms.v2.functional.normalize.rst.txt |     9 +
 ...ision.transforms.v2.functional.pad.rst.txt |     9 +
 ...rms.v2.functional.permute_channels.rst.txt |     9 +
 ...ansforms.v2.functional.perspective.rst.txt |     9 +
 ...sforms.v2.functional.pil_to_tensor.rst.txt |     6 +
 ...transforms.v2.functional.posterize.rst.txt |     9 +
 ...orms.v2.functional.register_kernel.rst.txt |     9 +
 ...on.transforms.v2.functional.resize.rst.txt |     9 +
 ...nsforms.v2.functional.resized_crop.rst.txt |     9 +
 ...rms.v2.functional.rgb_to_grayscale.rst.txt |     9 +
 ...on.transforms.v2.functional.rotate.rst.txt |     9 +
 ....transforms.v2.functional.solarize.rst.txt |     9 +
 ....transforms.v2.functional.ten_crop.rst.txt |     9 +
 ....transforms.v2.functional.to_dtype.rst.txt |     6 +
 ...nsforms.v2.functional.to_grayscale.rst.txt |     9 +
 ....transforms.v2.functional.to_image.rst.txt |     6 +
 ...nsforms.v2.functional.to_pil_image.rst.txt |     6 +
 ...transforms.v2.functional.to_tensor.rst.txt |    10 +
 ...ctional.uniform_temporal_subsample.rst.txt |     9 +
 ...sforms.v2.functional.vertical_flip.rst.txt |     9 +
 ...ision.tv_tensors.BoundingBoxFormat.rst.txt |    10 +
 ...rchvision.tv_tensors.BoundingBoxes.rst.txt |    10 +
 .../torchvision.tv_tensors.Image.rst.txt      |    10 +
 .../torchvision.tv_tensors.Mask.rst.txt       |    10 +
 .../torchvision.tv_tensors.TVTensor.rst.txt   |    10 +
 .../torchvision.tv_tensors.Video.rst.txt      |    10 +
 ...hvision.tv_tensors.set_return_type.rst.txt |    10 +
 .../torchvision.tv_tensors.wrap.rst.txt       |    10 +
 ...chvision.utils.draw_bounding_boxes.rst.txt |     9 +
 .../torchvision.utils.draw_keypoints.rst.txt  |     9 +
 ...sion.utils.draw_segmentation_masks.rst.txt |     9 +
 .../torchvision.utils.flow_to_image.rst.txt   |     9 +
 .../torchvision.utils.make_grid.rst.txt       |     9 +
 .../torchvision.utils.save_image.rst.txt      |     9 +
 0.16/_sources/generated/video_table.rst.txt   |    19 +
 0.16/_sources/index.rst.txt                   |    69 +
 0.16/_sources/io.rst.txt                      |    89 +
 0.16/_sources/models.rst.txt                  |   573 +
 0.16/_sources/models/alexnet.rst.txt          |    28 +
 0.16/_sources/models/convnext.rst.txt         |    26 +
 0.16/_sources/models/deeplabv3.rst.txt        |    28 +
 0.16/_sources/models/densenet.rst.txt         |    27 +
 0.16/_sources/models/efficientnet.rst.txt     |    31 +
 0.16/_sources/models/efficientnetv2.rst.txt   |    26 +
 0.16/_sources/models/faster_rcnn.rst.txt      |    31 +
 0.16/_sources/models/fcn.rst.txt              |    28 +
 0.16/_sources/models/fcos.rst.txt             |    24 +
 .../torchvision.models.alexnet.rst.txt        |     9 +
 .../torchvision.models.convnext_base.rst.txt  |     9 +
 .../torchvision.models.convnext_large.rst.txt |     9 +
 .../torchvision.models.convnext_small.rst.txt |     9 +
 .../torchvision.models.convnext_tiny.rst.txt  |     9 +
 .../torchvision.models.densenet121.rst.txt    |     9 +
 .../torchvision.models.densenet161.rst.txt    |     9 +
 .../torchvision.models.densenet169.rst.txt    |     9 +
 .../torchvision.models.densenet201.rst.txt    |     9 +
 ...terrcnn_mobilenet_v3_large_320_fpn.rst.txt |     9 +
 ....fasterrcnn_mobilenet_v3_large_fpn.rst.txt |     9 +
 ....detection.fasterrcnn_resnet50_fpn.rst.txt |     9 +
 ...tection.fasterrcnn_resnet50_fpn_v2.rst.txt |     9 +
 ...models.detection.fcos_resnet50_fpn.rst.txt |     9 +
 ...etection.keypointrcnn_resnet50_fpn.rst.txt |     9 +
 ...ls.detection.maskrcnn_resnet50_fpn.rst.txt |     9 +
 ...detection.maskrcnn_resnet50_fpn_v2.rst.txt |     9 +
 ...s.detection.retinanet_resnet50_fpn.rst.txt |     9 +
 ...etection.retinanet_resnet50_fpn_v2.rst.txt |     9 +
 ...sion.models.detection.ssd300_vgg16.rst.txt |     9 +
 ...tion.ssdlite320_mobilenet_v3_large.rst.txt |     9 +
 ...torchvision.models.efficientnet_b0.rst.txt |     9 +
 ...torchvision.models.efficientnet_b1.rst.txt |     9 +
 ...torchvision.models.efficientnet_b2.rst.txt |     9 +
 ...torchvision.models.efficientnet_b3.rst.txt |     9 +
 ...torchvision.models.efficientnet_b4.rst.txt |     9 +
 ...torchvision.models.efficientnet_b5.rst.txt |     9 +
 ...torchvision.models.efficientnet_b6.rst.txt |     9 +
 ...torchvision.models.efficientnet_b7.rst.txt |     9 +
 ...rchvision.models.efficientnet_v2_l.rst.txt |     9 +
 ...rchvision.models.efficientnet_v2_m.rst.txt |     9 +
 ...rchvision.models.efficientnet_v2_s.rst.txt |     9 +
 .../torchvision.models.googlenet.rst.txt      |     9 +
 .../torchvision.models.inception_v3.rst.txt   |     9 +
 .../torchvision.models.maxvit_t.rst.txt       |     9 +
 .../torchvision.models.mnasnet0_5.rst.txt     |     9 +
 .../torchvision.models.mnasnet0_75.rst.txt    |     9 +
 .../torchvision.models.mnasnet1_0.rst.txt     |     9 +
 .../torchvision.models.mnasnet1_3.rst.txt     |     9 +
 .../torchvision.models.mobilenet_v2.rst.txt   |     9 +
 ...chvision.models.mobilenet_v3_large.rst.txt |     9 +
 ...chvision.models.mobilenet_v3_small.rst.txt |     9 +
 ...ion.models.optical_flow.raft_large.rst.txt |     9 +
 ...ion.models.optical_flow.raft_small.rst.txt |     9 +
 ...sion.models.quantization.googlenet.rst.txt |     9 +
 ...n.models.quantization.inception_v3.rst.txt |     9 +
 ...n.models.quantization.mobilenet_v2.rst.txt |     9 +
 ...ls.quantization.mobilenet_v3_large.rst.txt |     9 +
 ...ision.models.quantization.resnet18.rst.txt |     9 +
 ...ision.models.quantization.resnet50.rst.txt |     9 +
 ...dels.quantization.resnext101_32x8d.rst.txt |     9 +
 ...dels.quantization.resnext101_64x4d.rst.txt |     9 +
 ...ls.quantization.shufflenet_v2_x0_5.rst.txt |     9 +
 ...ls.quantization.shufflenet_v2_x1_0.rst.txt |     9 +
 ...ls.quantization.shufflenet_v2_x1_5.rst.txt |     9 +
 ...ls.quantization.shufflenet_v2_x2_0.rst.txt |     9 +
 .../torchvision.models.regnet_x_16gf.rst.txt  |     9 +
 .../torchvision.models.regnet_x_1_6gf.rst.txt |     9 +
 .../torchvision.models.regnet_x_32gf.rst.txt  |     9 +
 .../torchvision.models.regnet_x_3_2gf.rst.txt |     9 +
 .../torchvision.models.regnet_x_400mf.rst.txt |     9 +
 .../torchvision.models.regnet_x_800mf.rst.txt |     9 +
 .../torchvision.models.regnet_x_8gf.rst.txt   |     9 +
 .../torchvision.models.regnet_y_128gf.rst.txt |     9 +
 .../torchvision.models.regnet_y_16gf.rst.txt  |     9 +
 .../torchvision.models.regnet_y_1_6gf.rst.txt |     9 +
 .../torchvision.models.regnet_y_32gf.rst.txt  |     9 +
 .../torchvision.models.regnet_y_3_2gf.rst.txt |     9 +
 .../torchvision.models.regnet_y_400mf.rst.txt |     9 +
 .../torchvision.models.regnet_y_800mf.rst.txt |     9 +
 .../torchvision.models.regnet_y_8gf.rst.txt   |     9 +
 .../torchvision.models.resnet101.rst.txt      |     9 +
 .../torchvision.models.resnet152.rst.txt      |     9 +
 .../torchvision.models.resnet18.rst.txt       |     9 +
 .../torchvision.models.resnet34.rst.txt       |     9 +
 .../torchvision.models.resnet50.rst.txt       |     9 +
 ...orchvision.models.resnext101_32x8d.rst.txt |     9 +
 ...orchvision.models.resnext101_64x4d.rst.txt |     9 +
 ...torchvision.models.resnext50_32x4d.rst.txt |     9 +
 ...ation.deeplabv3_mobilenet_v3_large.rst.txt |     9 +
 ...s.segmentation.deeplabv3_resnet101.rst.txt |     9 +
 ...ls.segmentation.deeplabv3_resnet50.rst.txt |     9 +
 ....models.segmentation.fcn_resnet101.rst.txt |     9 +
 ...n.models.segmentation.fcn_resnet50.rst.txt |     9 +
 ...entation.lraspp_mobilenet_v3_large.rst.txt |     9 +
 ...chvision.models.shufflenet_v2_x0_5.rst.txt |     9 +
 ...chvision.models.shufflenet_v2_x1_0.rst.txt |     9 +
 ...chvision.models.shufflenet_v2_x1_5.rst.txt |     9 +
 ...chvision.models.shufflenet_v2_x2_0.rst.txt |     9 +
 .../torchvision.models.squeezenet1_0.rst.txt  |     9 +
 .../torchvision.models.squeezenet1_1.rst.txt  |     9 +
 .../torchvision.models.swin_b.rst.txt         |     9 +
 .../torchvision.models.swin_s.rst.txt         |     9 +
 .../torchvision.models.swin_t.rst.txt         |     9 +
 .../torchvision.models.swin_v2_b.rst.txt      |     9 +
 .../torchvision.models.swin_v2_s.rst.txt      |     9 +
 .../torchvision.models.swin_v2_t.rst.txt      |     9 +
 .../torchvision.models.vgg11.rst.txt          |     9 +
 .../torchvision.models.vgg11_bn.rst.txt       |     9 +
 .../torchvision.models.vgg13.rst.txt          |     9 +
 .../torchvision.models.vgg13_bn.rst.txt       |     9 +
 .../torchvision.models.vgg16.rst.txt          |     9 +
 .../torchvision.models.vgg16_bn.rst.txt       |     9 +
 .../torchvision.models.vgg19.rst.txt          |     9 +
 .../torchvision.models.vgg19_bn.rst.txt       |     9 +
 .../torchvision.models.video.mc3_18.rst.txt   |     9 +
 ...torchvision.models.video.mvit_v1_b.rst.txt |     9 +
 ...torchvision.models.video.mvit_v2_s.rst.txt |     9 +
 ...rchvision.models.video.r2plus1d_18.rst.txt |     9 +
 .../torchvision.models.video.r3d_18.rst.txt   |     9 +
 .../torchvision.models.video.s3d.rst.txt      |     9 +
 .../torchvision.models.video.swin3d_b.rst.txt |     9 +
 .../torchvision.models.video.swin3d_s.rst.txt |     9 +
 .../torchvision.models.video.swin3d_t.rst.txt |     9 +
 .../torchvision.models.vit_b_16.rst.txt       |     9 +
 .../torchvision.models.vit_b_32.rst.txt       |     9 +
 .../torchvision.models.vit_h_14.rst.txt       |     9 +
 .../torchvision.models.vit_l_16.rst.txt       |     9 +
 .../torchvision.models.vit_l_32.rst.txt       |     9 +
 ...orchvision.models.wide_resnet101_2.rst.txt |     9 +
 ...torchvision.models.wide_resnet50_2.rst.txt |     9 +
 0.16/_sources/models/googlenet.rst.txt        |    24 +
 0.16/_sources/models/googlenet_quant.rst.txt  |    24 +
 0.16/_sources/models/inception.rst.txt        |    23 +
 0.16/_sources/models/inception_quant.rst.txt  |    24 +
 0.16/_sources/models/keypoint_rcnn.rst.txt    |    26 +
 0.16/_sources/models/lraspp.rst.txt           |    24 +
 0.16/_sources/models/mask_rcnn.rst.txt        |    27 +
 0.16/_sources/models/maxvit.rst.txt           |    23 +
 0.16/_sources/models/mnasnet.rst.txt          |    28 +
 0.16/_sources/models/mobilenetv2.rst.txt      |    24 +
 .../_sources/models/mobilenetv2_quant.rst.txt |    24 +
 0.16/_sources/models/mobilenetv3.rst.txt      |    24 +
 .../_sources/models/mobilenetv3_quant.rst.txt |    23 +
 0.16/_sources/models/raft.rst.txt             |    25 +
 0.16/_sources/models/regnet.rst.txt           |    37 +
 0.16/_sources/models/resnet.rst.txt           |    33 +
 0.16/_sources/models/resnet_quant.rst.txt     |    25 +
 0.16/_sources/models/resnext.rst.txt          |    26 +
 0.16/_sources/models/resnext_quant.rst.txt    |    25 +
 0.16/_sources/models/retinanet.rst.txt        |    25 +
 0.16/_sources/models/shufflenetv2.rst.txt     |    27 +
 .../models/shufflenetv2_quant.rst.txt         |    27 +
 0.16/_sources/models/squeezenet.rst.txt       |    26 +
 0.16/_sources/models/ssd.rst.txt              |    26 +
 0.16/_sources/models/ssdlite.rst.txt          |    27 +
 0.16/_sources/models/swin_transformer.rst.txt |    32 +
 0.16/_sources/models/vgg.rst.txt              |    30 +
 0.16/_sources/models/video_mvit.rst.txt       |    27 +
 0.16/_sources/models/video_resnet.rst.txt     |    28 +
 0.16/_sources/models/video_s3d.rst.txt        |    25 +
 .../models/video_swin_transformer.rst.txt     |    27 +
 .../models/vision_transformer.rst.txt         |    28 +
 0.16/_sources/models/wide_resnet.rst.txt      |    25 +
 0.16/_sources/ops.rst.txt                     |   103 +
 0.16/_sources/training_references.rst.txt     |    29 +
 0.16/_sources/transforms.rst.txt              |   665 +
 0.16/_sources/tv_tensors.rst.txt              |    29 +
 0.16/_sources/utils.rst.txt                   |    20 +
 .../_sphinx_javascript_frameworks_compat.js   |   134 +
 0.16/_static/basic.css                        |   930 ++
 0.16/_static/binder_badge_logo.svg            |     1 +
 0.16/_static/broken_example.png               |   Bin 0 -> 21404 bytes
 0.16/_static/check-solid.svg                  |     4 +
 0.16/_static/clipboard.min.js                 |     7 +
 0.16/_static/copy-button.svg                  |     5 +
 0.16/_static/copybutton.css                   |    94 +
 0.16/_static/copybutton.js                    |   248 +
 0.16/_static/copybutton_funcs.js              |    73 +
 0.16/_static/css/custom_torchvision.css       |    35 +
 0.16/_static/css/theme.css                    | 12482 ++++++++++++++++
 0.16/_static/doctools.js                      |   264 +
 0.16/_static/documentation_options.js         |    14 +
 0.16/_static/file.png                         |   Bin 0 -> 286 bytes
 .../FreightSans/freight-sans-bold-italic.woff |   Bin 0 -> 39560 bytes
 .../freight-sans-bold-italic.woff2            |   Bin 0 -> 31812 bytes
 .../fonts/FreightSans/freight-sans-bold.woff  |   Bin 0 -> 32396 bytes
 .../fonts/FreightSans/freight-sans-bold.woff2 |   Bin 0 -> 25672 bytes
 .../FreightSans/freight-sans-book-italic.woff |   Bin 0 -> 33944 bytes
 .../freight-sans-book-italic.woff2            |   Bin 0 -> 26832 bytes
 .../fonts/FreightSans/freight-sans-book.woff  |   Bin 0 -> 31612 bytes
 .../fonts/FreightSans/freight-sans-book.woff2 |   Bin 0 -> 25120 bytes
 .../freight-sans-light-italic.woff            |   Bin 0 -> 29304 bytes
 .../freight-sans-light-italic.woff2           |   Bin 0 -> 22720 bytes
 .../fonts/FreightSans/freight-sans-light.woff |   Bin 0 -> 26908 bytes
 .../FreightSans/freight-sans-light.woff2      |   Bin 0 -> 21012 bytes
 .../freight-sans-medium-italic.woff           |   Bin 0 -> 19420 bytes
 .../freight-sans-medium-italic.woff2          |   Bin 0 -> 16000 bytes
 .../FreightSans/freight-sans-medium.woff      |   Bin 0 -> 32072 bytes
 .../FreightSans/freight-sans-medium.woff2     |   Bin 0 -> 25460 bytes
 .../fonts/IBMPlexMono/IBMPlexMono-Light.woff  |   Bin 0 -> 50680 bytes
 .../fonts/IBMPlexMono/IBMPlexMono-Light.woff2 |   Bin 0 -> 35916 bytes
 .../fonts/IBMPlexMono/IBMPlexMono-Medium.woff |   Bin 0 -> 51872 bytes
 .../IBMPlexMono/IBMPlexMono-Medium.woff2      |   Bin 0 -> 36648 bytes
 .../IBMPlexMono/IBMPlexMono-Regular.woff      |   Bin 0 -> 50664 bytes
 .../IBMPlexMono/IBMPlexMono-Regular.woff2     |   Bin 0 -> 35536 bytes
 .../IBMPlexMono/IBMPlexMono-SemiBold.woff     |   Bin 0 -> 52936 bytes
 .../IBMPlexMono/IBMPlexMono-SemiBold.woff2    |   Bin 0 -> 37592 bytes
 0.16/_static/images/arrow-down-orange.svg     |    19 +
 0.16/_static/images/arrow-right-with-tail.svg |    19 +
 0.16/_static/images/chevron-down-black.svg    |    16 +
 0.16/_static/images/chevron-down-grey.svg     |    18 +
 0.16/_static/images/chevron-down-orange.svg   |    16 +
 0.16/_static/images/chevron-down-white.svg    |    16 +
 0.16/_static/images/chevron-right-orange.svg  |    17 +
 0.16/_static/images/chevron-right-white.svg   |    17 +
 .../_static/images/home-footer-background.jpg |   Bin 0 -> 38907 bytes
 0.16/_static/images/icon-close.svg            |    21 +
 0.16/_static/images/icon-menu-dots-dark.svg   |    42 +
 0.16/_static/images/logo-dark.svg             |    30 +
 0.16/_static/images/logo-facebook-dark.svg    |     8 +
 0.16/_static/images/logo-icon.svg             |    12 +
 0.16/_static/images/logo-twitter-dark.svg     |    16 +
 0.16/_static/images/logo-youtube-dark.svg     |    21 +
 0.16/_static/images/logo.svg                  |    31 +
 0.16/_static/images/pytorch-colab.svg         |    24 +
 0.16/_static/images/pytorch-download.svg      |    10 +
 0.16/_static/images/pytorch-github.svg        |    15 +
 0.16/_static/images/pytorch-x.svg             |    10 +
 0.16/_static/images/search-icon.svg           |    19 +
 0.16/_static/images/view-page-source-icon.svg |    13 +
 0.16/_static/img/pytorch-logo-dark.png        |   Bin 0 -> 27763 bytes
 0.16/_static/img/pytorch-logo-dark.svg        |    24 +
 0.16/_static/img/pytorch-logo-flame.png       |   Bin 0 -> 1010 bytes
 0.16/_static/img/pytorch-logo-flame.svg       |    33 +
 0.16/_static/jquery-3.6.0.js                  | 10881 ++++++++++++++
 0.16/_static/jquery.js                        |     2 +
 0.16/_static/js/modernizr.min.js              |     4 +
 0.16/_static/js/theme.js                      |  1128 ++
 0.16/_static/js/vendor/anchor.min.js          |     9 +
 0.16/_static/js/vendor/bootstrap.min.js       |     7 +
 0.16/_static/js/vendor/popper.min.js          |     5 +
 0.16/_static/jupyterlite_badge_logo.svg       |     3 +
 0.16/_static/language_data.js                 |   199 +
 0.16/_static/minus.png                        |   Bin 0 -> 90 bytes
 0.16/_static/no_image.png                     |   Bin 0 -> 4315 bytes
 0.16/_static/plus.png                         |   Bin 0 -> 90 bytes
 0.16/_static/pygments.css                     |    75 +
 0.16/_static/pytorch-logo-dark.svg            |    24 +
 0.16/_static/searchtools.js                   |   531 +
 0.16/_static/sg_gallery-binder.css            |    11 +
 0.16/_static/sg_gallery-dataframe.css         |    47 +
 0.16/_static/sg_gallery-rendered-html.css     |   224 +
 0.16/_static/sg_gallery.css                   |   342 +
 0.16/_static/underscore-1.13.1.js             |  2042 +++
 0.16/_static/underscore.js                    |     6 +
 0.16/auto_examples/index.html                 |   743 +
 0.16/auto_examples/others/index.html          |   705 +
 .../others/plot_optical_flow.html             |   904 ++
 .../others/plot_repurposing_annotations.html  |   944 ++
 .../plot_scripted_tensor_transforms.html      |   823 +
 0.16/auto_examples/others/plot_video_api.html |  1227 ++
 .../others/plot_visualization_utils.html      |  1299 ++
 .../others/sg_execution_times.html            |   700 +
 0.16/auto_examples/transforms/index.html      |   711 +
 .../transforms/plot_custom_transforms.html    |   827 +
 .../transforms/plot_custom_tv_tensors.html    |   810 +
 .../transforms/plot_cutmix_mixup.html         |   859 ++
 .../transforms/plot_transforms_e2e.html       |   895 ++
 .../plot_transforms_getting_started.html      |   949 ++
 .../plot_transforms_illustrations.html        |  1077 ++
 .../transforms/plot_tv_tensors.html           |   925 ++
 .../transforms/sg_execution_times.html        |   708 +
 0.16/datasets.html                            |  1011 ++
 0.16/feature_extraction.html                  |   846 ++
 .../generated/classification_quant_table.html |   790 +
 0.16/generated/classification_table.html      |  1497 ++
 0.16/generated/detection_keypoint_table.html  |   706 +
 0.16/generated/detection_table.html           |   744 +
 .../instance_segmentation_table.html          |   706 +
 0.16/generated/segmentation_table.html        |   734 +
 .../torchvision.datasets.CIFAR10.html         |   731 +
 .../torchvision.datasets.CIFAR100.html        |   715 +
 .../torchvision.datasets.CREStereo.html       |   760 +
 .../torchvision.datasets.Caltech101.html      |   738 +
 .../torchvision.datasets.Caltech256.html      |   729 +
 .../torchvision.datasets.CarlaStereo.html     |   745 +
 .../torchvision.datasets.CelebA.html          |   744 +
 .../torchvision.datasets.Cityscapes.html      |   756 +
 .../torchvision.datasets.CocoCaptions.html    |   753 +
 .../torchvision.datasets.CocoDetection.html   |   736 +
 .../torchvision.datasets.Country211.html      |   731 +
 0.16/generated/torchvision.datasets.DTD.html  |   735 +
 .../torchvision.datasets.DatasetFolder.html   |   783 +
 .../torchvision.datasets.EMNIST.html          |   734 +
 .../torchvision.datasets.ETH3DStereo.html     |   769 +
 .../torchvision.datasets.EuroSAT.html         |   728 +
 .../torchvision.datasets.FER2013.html         |   726 +
 .../torchvision.datasets.FGVCAircraft.html    |   745 +
 .../torchvision.datasets.FakeData.html        |   733 +
 ...chvision.datasets.FallingThingsStereo.html |   763 +
 .../torchvision.datasets.FashionMNIST.html    |   731 +
 .../torchvision.datasets.Flickr30k.html       |   726 +
 .../torchvision.datasets.Flickr8k.html        |   726 +
 .../torchvision.datasets.Flowers102.html      |   737 +
 .../torchvision.datasets.FlyingChairs.html    |   743 +
 .../torchvision.datasets.FlyingThings3D.html  |   748 +
 .../torchvision.datasets.Food101.html         |   733 +
 .../generated/torchvision.datasets.GTSRB.html |   728 +
 0.16/generated/torchvision.datasets.HD1K.html |   740 +
 .../torchvision.datasets.HMDB51.html          |   757 +
 .../torchvision.datasets.INaturalist.html     |   762 +
 .../torchvision.datasets.ImageFolder.html     |   719 +
 .../torchvision.datasets.ImageNet.html        |   734 +
 .../torchvision.datasets.InStereo2k.html      |   750 +
 .../torchvision.datasets.KMNIST.html          |   731 +
 .../torchvision.datasets.Kinetics.html        |   778 +
 .../generated/torchvision.datasets.Kitti.html |   758 +
 .../torchvision.datasets.Kitti2012Stereo.html |   757 +
 .../torchvision.datasets.Kitti2015Stereo.html |   760 +
 .../torchvision.datasets.KittiFlow.html       |   739 +
 .../torchvision.datasets.LFWPairs.html        |   733 +
 .../torchvision.datasets.LFWPeople.html       |   733 +
 0.16/generated/torchvision.datasets.LSUN.html |   729 +
 .../generated/torchvision.datasets.MNIST.html |   731 +
 ...hvision.datasets.Middlebury2014Stereo.html |   774 +
 .../torchvision.datasets.MovingMNIST.html     |   731 +
 .../torchvision.datasets.Omniglot.html        |   731 +
 .../torchvision.datasets.OxfordIIITPet.html   |   737 +
 0.16/generated/torchvision.datasets.PCAM.html |   732 +
 .../torchvision.datasets.PhotoTour.html       |   737 +
 .../torchvision.datasets.Places365.html       |   738 +
 .../torchvision.datasets.QMNIST.html          |   744 +
 .../torchvision.datasets.RenderedSST2.html    |   734 +
 .../torchvision.datasets.SBDataset.html       |   745 +
 0.16/generated/torchvision.datasets.SBU.html  |   729 +
 .../torchvision.datasets.SEMEION.html         |   729 +
 .../generated/torchvision.datasets.STL10.html |   734 +
 .../torchvision.datasets.SUN397.html          |   729 +
 0.16/generated/torchvision.datasets.SVHN.html |   737 +
 .../torchvision.datasets.SceneFlowStereo.html |   771 +
 .../torchvision.datasets.Sintel.html          |   760 +
 .../torchvision.datasets.SintelStereo.html    |   762 +
 .../torchvision.datasets.StanfordCars.html    |   726 +
 .../torchvision.datasets.UCF101.html          |   758 +
 0.16/generated/torchvision.datasets.USPS.html |   733 +
 .../torchvision.datasets.VOCDetection.html    |   734 +
 .../torchvision.datasets.VOCSegmentation.html |   733 +
 .../torchvision.datasets.VisionDataset.html   |   719 +
 .../torchvision.datasets.WIDERFace.html       |   741 +
 ...tasets.wrap_dataset_for_transforms_v2.html |   778 +
 .../torchvision.io.ImageReadMode.html         |   699 +
 .../generated/torchvision.io.VideoReader.html |   828 +
 .../torchvision.io.decode_image.html          |   712 +
 .../generated/torchvision.io.decode_jpeg.html |   728 +
 0.16/generated/torchvision.io.decode_png.html |   711 +
 .../generated/torchvision.io.encode_jpeg.html |   715 +
 0.16/generated/torchvision.io.encode_png.html |   715 +
 0.16/generated/torchvision.io.read_file.html  |   703 +
 0.16/generated/torchvision.io.read_image.html |   723 +
 0.16/generated/torchvision.io.read_video.html |   718 +
 .../torchvision.io.read_video_timestamps.html |   711 +
 0.16/generated/torchvision.io.write_file.html |   703 +
 0.16/generated/torchvision.io.write_jpeg.html |   705 +
 0.16/generated/torchvision.io.write_png.html  |   706 +
 .../generated/torchvision.io.write_video.html |   711 +
 ...e_extraction.create_feature_extractor.html |   798 +
 ...ature_extraction.get_graph_node_names.html |   739 +
 .../torchvision.models.get_model.html         |   712 +
 .../torchvision.models.get_model_weights.html |   705 +
 .../torchvision.models.get_weight.html        |   705 +
 .../torchvision.models.list_models.html       |   713 +
 .../torchvision.ops.Conv2dNormActivation.html |   711 +
 .../torchvision.ops.Conv3dNormActivation.html |   711 +
 .../torchvision.ops.DeformConv2d.html         |   708 +
 .../torchvision.ops.DropBlock2d.html          |   711 +
 .../torchvision.ops.DropBlock3d.html          |   711 +
 ...torchvision.ops.FeaturePyramidNetwork.html |   764 +
 .../torchvision.ops.FrozenBatchNorm2d.html    |   716 +
 0.16/generated/torchvision.ops.MLP.html       |   708 +
 .../torchvision.ops.MultiScaleRoIAlign.html   |   749 +
 .../generated/torchvision.ops.PSRoIAlign.html |   708 +
 0.16/generated/torchvision.ops.PSRoIPool.html |   708 +
 0.16/generated/torchvision.ops.Permute.html   |   713 +
 0.16/generated/torchvision.ops.RoIAlign.html  |   708 +
 0.16/generated/torchvision.ops.RoIPool.html   |   708 +
 .../torchvision.ops.SqueezeExcitation.html    |   719 +
 .../torchvision.ops.StochasticDepth.html      |   708 +
 .../torchvision.ops.batched_nms.html          |   715 +
 0.16/generated/torchvision.ops.box_area.html  |   708 +
 .../torchvision.ops.box_convert.html          |   715 +
 0.16/generated/torchvision.ops.box_iou.html   |   710 +
 .../torchvision.ops.clip_boxes_to_image.html  |   709 +
 .../torchvision.ops.complete_box_iou.html     |   711 +
 ...torchvision.ops.complete_box_iou_loss.html |   724 +
 .../torchvision.ops.deform_conv2d.html        |   736 +
 .../torchvision.ops.distance_box_iou.html     |   712 +
 ...torchvision.ops.distance_box_iou_loss.html |   724 +
 .../torchvision.ops.drop_block2d.html         |   714 +
 .../torchvision.ops.drop_block3d.html         |   714 +
 .../torchvision.ops.generalized_box_iou.html  |   711 +
 ...chvision.ops.generalized_box_iou_loss.html |   724 +
 .../torchvision.ops.masks_to_boxes.html       |   712 +
 0.16/generated/torchvision.ops.nms.html       |   720 +
 .../torchvision.ops.ps_roi_align.html         |   727 +
 .../torchvision.ops.ps_roi_pool.html          |   722 +
 .../torchvision.ops.remove_small_boxes.html   |   710 +
 0.16/generated/torchvision.ops.roi_align.html |   730 +
 0.16/generated/torchvision.ops.roi_pool.html  |   721 +
 .../torchvision.ops.sigmoid_focal_loss.html   |   716 +
 .../torchvision.ops.stochastic_depth.html     |   714 +
 .../torchvision.transforms.AugMix.html        |   735 +
 .../torchvision.transforms.AutoAugment.html   |   742 +
 ...chvision.transforms.AutoAugmentPolicy.html |   699 +
 .../torchvision.transforms.CenterCrop.html    |   724 +
 .../torchvision.transforms.ColorJitter.html   |   765 +
 .../torchvision.transforms.Compose.html       |   737 +
 ...chvision.transforms.ConvertImageDtype.html |   727 +
 ...rchvision.transforms.ElasticTransform.html |   738 +
 .../torchvision.transforms.FiveCrop.html      |   739 +
 .../torchvision.transforms.GaussianBlur.html  |   753 +
 .../torchvision.transforms.Grayscale.html     |   732 +
 .../torchvision.transforms.Lambda.html        |   699 +
 ...ision.transforms.LinearTransformation.html |   730 +
 .../torchvision.transforms.Normalize.html     |   738 +
 .../torchvision.transforms.PILToTensor.html   |   700 +
 .../generated/torchvision.transforms.Pad.html |   753 +
 .../torchvision.transforms.RandAugment.html   |   732 +
 ...sion.transforms.RandomAdjustSharpness.html |   725 +
 .../torchvision.transforms.RandomAffine.html  |   756 +
 .../torchvision.transforms.RandomApply.html   |   733 +
 ...hvision.transforms.RandomAutocontrast.html |   722 +
 .../torchvision.transforms.RandomChoice.html  |   698 +
 .../torchvision.transforms.RandomCrop.html    |   785 +
 ...torchvision.transforms.RandomEqualize.html |   722 +
 .../torchvision.transforms.RandomErasing.html |   763 +
 ...orchvision.transforms.RandomGrayscale.html |   726 +
 ...ision.transforms.RandomHorizontalFlip.html |   740 +
 .../torchvision.transforms.RandomInvert.html  |   722 +
 .../torchvision.transforms.RandomOrder.html   |   694 +
 ...chvision.transforms.RandomPerspective.html |   750 +
 ...orchvision.transforms.RandomPosterize.html |   725 +
 ...chvision.transforms.RandomResizedCrop.html |   792 +
 ...torchvision.transforms.RandomRotation.html |   751 +
 ...torchvision.transforms.RandomSolarize.html |   725 +
 ...hvision.transforms.RandomVerticalFlip.html |   722 +
 .../torchvision.transforms.Resize.html        |   775 +
 .../torchvision.transforms.TenCrop.html       |   739 +
 .../torchvision.transforms.ToPILImage.html    |   710 +
 .../torchvision.transforms.ToTensor.html      |   705 +
 ...hvision.transforms.TrivialAugmentWide.html |   730 +
 ...ansforms.functional.adjust_brightness.html |   712 +
 ...transforms.functional.adjust_contrast.html |   712 +
 ...on.transforms.functional.adjust_gamma.html |   720 +
 ...sion.transforms.functional.adjust_hue.html |   724 +
 ...ansforms.functional.adjust_saturation.html |   712 +
 ...ransforms.functional.adjust_sharpness.html |   716 +
 ...chvision.transforms.functional.affine.html |   733 +
 ...on.transforms.functional.autocontrast.html |   714 +
 ...ion.transforms.functional.center_crop.html |   716 +
 ...sforms.functional.convert_image_dtype.html |   726 +
 ...orchvision.transforms.functional.crop.html |   718 +
 ...vision.transforms.functional.equalize.html |   715 +
 ...rchvision.transforms.functional.erase.html |   714 +
 ...ision.transforms.functional.five_crop.html |   722 +
 ...n.transforms.functional.gaussian_blur.html |   732 +
 ....transforms.functional.get_dimensions.html |   705 +
 ...rms.functional.get_image_num_channels.html |   705 +
 ....transforms.functional.get_image_size.html |   705 +
 ...rchvision.transforms.functional.hflip.html |   712 +
 ...chvision.transforms.functional.invert.html |   712 +
 ...ision.transforms.functional.normalize.html |   716 +
 ...torchvision.transforms.functional.pad.html |   744 +
 ...ion.transforms.functional.perspective.html |   729 +
 ...n.transforms.functional.pil_to_tensor.html |   711 +
 ...ision.transforms.functional.posterize.html |   716 +
 ...chvision.transforms.functional.resize.html |   768 +
 ...on.transforms.functional.resized_crop.html |   744 +
 ...ransforms.functional.rgb_to_grayscale.html |   720 +
 ...chvision.transforms.functional.rotate.html |   732 +
 ...vision.transforms.functional.solarize.html |   715 +
 ...vision.transforms.functional.ten_crop.html |   722 +
 ...on.transforms.functional.to_grayscale.html |   718 +
 ...on.transforms.functional.to_pil_image.html |   721 +
 ...ision.transforms.functional.to_tensor.html |   707 +
 ...rchvision.transforms.functional.vflip.html |   712 +
 .../torchvision.transforms.v2.AugMix.html     |   738 +
 ...torchvision.transforms.v2.AutoAugment.html |   745 +
 .../torchvision.transforms.v2.CenterCrop.html |   714 +
 ...sion.transforms.v2.ClampBoundingBoxes.html |   699 +
 ...torchvision.transforms.v2.ColorJitter.html |   753 +
 .../torchvision.transforms.v2.Compose.html    |   765 +
 ...ransforms.v2.ConvertBoundingBoxFormat.html |   705 +
 ...ision.transforms.v2.ConvertImageDtype.html |   722 +
 .../torchvision.transforms.v2.CutMix.html     |   727 +
 ...vision.transforms.v2.ElasticTransform.html |   741 +
 .../torchvision.transforms.v2.FiveCrop.html   |   738 +
 ...orchvision.transforms.v2.GaussianBlur.html |   735 +
 .../torchvision.transforms.v2.Grayscale.html  |   709 +
 .../torchvision.transforms.v2.Lambda.html     |   713 +
 ...on.transforms.v2.LinearTransformation.html |   717 +
 .../torchvision.transforms.v2.MixUp.html      |   727 +
 .../torchvision.transforms.v2.Normalize.html  |   726 +
 ...torchvision.transforms.v2.PILToTensor.html |   704 +
 .../torchvision.transforms.v2.Pad.html        |   740 +
 ...torchvision.transforms.v2.RandAugment.html |   735 +
 ...n.transforms.v2.RandomAdjustSharpness.html |   714 +
 ...orchvision.transforms.v2.RandomAffine.html |   749 +
 ...torchvision.transforms.v2.RandomApply.html |   746 +
 ...sion.transforms.v2.RandomAutocontrast.html |   710 +
 ...ransforms.v2.RandomChannelPermutation.html |   698 +
 ...orchvision.transforms.v2.RandomChoice.html |   727 +
 .../torchvision.transforms.v2.RandomCrop.html |   770 +
 ...chvision.transforms.v2.RandomEqualize.html |   710 +
 ...rchvision.transforms.v2.RandomErasing.html |   753 +
 ...hvision.transforms.v2.RandomGrayscale.html |   706 +
 ...on.transforms.v2.RandomHorizontalFlip.html |   726 +
 ...orchvision.transforms.v2.RandomInvert.html |   710 +
 ...rchvision.transforms.v2.RandomIoUCrop.html |   728 +
 ...torchvision.transforms.v2.RandomOrder.html |   718 +
 ...ision.transforms.v2.RandomPerspective.html |   743 +
 ...ransforms.v2.RandomPhotometricDistort.html |   727 +
 ...hvision.transforms.v2.RandomPosterize.html |   714 +
 ...orchvision.transforms.v2.RandomResize.html |   742 +
 ...ision.transforms.v2.RandomResizedCrop.html |   782 +
 ...chvision.transforms.v2.RandomRotation.html |   752 +
 ...sion.transforms.v2.RandomShortestSize.html |   734 +
 ...chvision.transforms.v2.RandomSolarize.html |   714 +
 ...sion.transforms.v2.RandomVerticalFlip.html |   711 +
 ...rchvision.transforms.v2.RandomZoomOut.html |   729 +
 .../torchvision.transforms.v2.Resize.html     |   762 +
 ...n.transforms.v2.SanitizeBoundingBoxes.html |   744 +
 ...torchvision.transforms.v2.ScaleJitter.html |   736 +
 .../torchvision.transforms.v2.TenCrop.html    |   719 +
 .../torchvision.transforms.v2.ToDtype.html    |   726 +
 .../torchvision.transforms.v2.ToImage.html    |   704 +
 .../torchvision.transforms.v2.ToPILImage.html |   715 +
 ...orchvision.transforms.v2.ToPureTensor.html |   699 +
 .../torchvision.transforms.v2.ToTensor.html   |   715 +
 ...sion.transforms.v2.TrivialAugmentWide.html |   733 +
 ...ransforms.v2.UniformTemporalSubsample.html |   706 +
 ...forms.v2.functional.adjust_brightness.html |   694 +
 ...nsforms.v2.functional.adjust_contrast.html |   694 +
 ...transforms.v2.functional.adjust_gamma.html |   694 +
 ...n.transforms.v2.functional.adjust_hue.html |   694 +
 ...forms.v2.functional.adjust_saturation.html |   694 +
 ...sforms.v2.functional.adjust_sharpness.html |   694 +
 ...ision.transforms.v2.functional.affine.html |   694 +
 ...transforms.v2.functional.autocontrast.html |   694 +
 ....transforms.v2.functional.center_crop.html |   694 +
 ...ms.v2.functional.clamp_bounding_boxes.html |   694 +
 ...unctional.convert_bounding_box_format.html |   694 +
 ...rms.v2.functional.convert_image_dtype.html |   694 +
 ...hvision.transforms.v2.functional.crop.html |   694 +
 ...sion.transforms.v2.functional.elastic.html |   694 +
 ...ion.transforms.v2.functional.equalize.html |   694 +
 ...vision.transforms.v2.functional.erase.html |   694 +
 ...on.transforms.v2.functional.five_crop.html |   694 +
 ...ransforms.v2.functional.gaussian_blur.html |   694 +
 ...nsforms.v2.functional.horizontal_flip.html |   694 +
 ...ision.transforms.v2.functional.invert.html |   694 +
 ...on.transforms.v2.functional.normalize.html |   694 +
 ...chvision.transforms.v2.functional.pad.html |   694 +
 ...sforms.v2.functional.permute_channels.html |   717 +
 ....transforms.v2.functional.perspective.html |   694 +
 ...ransforms.v2.functional.pil_to_tensor.html |   711 +
 ...on.transforms.v2.functional.posterize.html |   694 +
 ...nsforms.v2.functional.register_kernel.html |   700 +
 ...ision.transforms.v2.functional.resize.html |   694 +
 ...transforms.v2.functional.resized_crop.html |   694 +
 ...sforms.v2.functional.rgb_to_grayscale.html |   694 +
 ...ision.transforms.v2.functional.rotate.html |   694 +
 ...ion.transforms.v2.functional.solarize.html |   694 +
 ...ion.transforms.v2.functional.ten_crop.html |   694 +
 ...ion.transforms.v2.functional.to_dtype.html |   694 +
 ...transforms.v2.functional.to_grayscale.html |   694 +
 ...ion.transforms.v2.functional.to_image.html |   694 +
 ...transforms.v2.functional.to_pil_image.html |   711 +
 ...on.transforms.v2.functional.to_tensor.html |   694 +
 ...functional.uniform_temporal_subsample.html |   694 +
 ...ransforms.v2.functional.vertical_flip.html |   694 +
 ...chvision.tv_tensors.BoundingBoxFormat.html |   704 +
 .../torchvision.tv_tensors.BoundingBoxes.html |   729 +
 .../torchvision.tv_tensors.Image.html         |   724 +
 .../torchvision.tv_tensors.Mask.html          |   715 +
 .../torchvision.tv_tensors.TVTensor.html      |   740 +
 .../torchvision.tv_tensors.Video.html         |   714 +
 ...orchvision.tv_tensors.set_return_type.html |   730 +
 .../torchvision.tv_tensors.wrap.html          |   714 +
 ...torchvision.utils.draw_bounding_boxes.html |   730 +
 .../torchvision.utils.draw_keypoints.html     |   720 +
 ...hvision.utils.draw_segmentation_masks.html |   722 +
 .../torchvision.utils.flow_to_image.html      |   713 +
 .../torchvision.utils.make_grid.html          |   723 +
 .../torchvision.utils.save_image.html         |   706 +
 0.16/generated/video_table.html               |   762 +
 0.16/genindex.html                            |  2479 +++
 0.16/index.html                               |   856 ++
 0.16/io.html                                  |   800 +
 0.16/models.html                              |  2459 +++
 0.16/models/alexnet.html                      |   710 +
 0.16/models/convnext.html                     |   719 +
 0.16/models/deeplabv3.html                    |   718 +
 0.16/models/densenet.html                     |   720 +
 0.16/models/efficientnet.html                 |   736 +
 0.16/models/efficientnetv2.html               |   716 +
 0.16/models/faster_rcnn.html                  |   722 +
 0.16/models/fcn.html                          |   717 +
 0.16/models/fcos.html                         |   710 +
 .../generated/torchvision.models.alexnet.html |   760 +
 .../torchvision.models.convnext_base.html     |   752 +
 .../torchvision.models.convnext_large.html    |   752 +
 .../torchvision.models.convnext_small.html    |   752 +
 .../torchvision.models.convnext_tiny.html     |   752 +
 .../torchvision.models.densenet121.html       |   751 +
 .../torchvision.models.densenet161.html       |   751 +
 .../torchvision.models.densenet169.html       |   751 +
 .../torchvision.models.densenet201.html       |   751 +
 ...fasterrcnn_mobilenet_v3_large_320_fpn.html |   768 +
 ...ion.fasterrcnn_mobilenet_v3_large_fpn.html |   768 +
 ...els.detection.fasterrcnn_resnet50_fpn.html |   817 +
 ....detection.fasterrcnn_resnet50_fpn_v2.html |   762 +
 ...on.models.detection.fcos_resnet50_fpn.html |   794 +
 ...s.detection.keypointrcnn_resnet50_fpn.html |   842 ++
 ...odels.detection.maskrcnn_resnet50_fpn.html |   806 +
 ...ls.detection.maskrcnn_resnet50_fpn_v2.html |   763 +
 ...dels.detection.retinanet_resnet50_fpn.html |   791 +
 ...s.detection.retinanet_resnet50_fpn_v2.html |   759 +
 ...hvision.models.detection.ssd300_vgg16.html |   791 +
 ...tection.ssdlite320_mobilenet_v3_large.html |   771 +
 .../torchvision.models.efficientnet_b0.html   |   752 +
 .../torchvision.models.efficientnet_b1.html   |   786 +
 .../torchvision.models.efficientnet_b2.html   |   752 +
 .../torchvision.models.efficientnet_b3.html   |   752 +
 .../torchvision.models.efficientnet_b4.html   |   752 +
 .../torchvision.models.efficientnet_b5.html   |   752 +
 .../torchvision.models.efficientnet_b6.html   |   752 +
 .../torchvision.models.efficientnet_b7.html   |   752 +
 .../torchvision.models.efficientnet_v2_l.html |   752 +
 .../torchvision.models.efficientnet_v2_m.html |   755 +
 .../torchvision.models.efficientnet_v2_s.html |   755 +
 .../torchvision.models.googlenet.html         |   752 +
 .../torchvision.models.inception_v3.html      |   757 +
 .../torchvision.models.maxvit_t.html          |   752 +
 .../torchvision.models.mnasnet0_5.html        |   752 +
 .../torchvision.models.mnasnet0_75.html       |   754 +
 .../torchvision.models.mnasnet1_0.html        |   752 +
 .../torchvision.models.mnasnet1_3.html        |   754 +
 .../torchvision.models.mobilenet_v2.html      |   786 +
 ...torchvision.models.mobilenet_v3_large.html |   786 +
 ...torchvision.models.mobilenet_v3_small.html |   754 +
 ...vision.models.optical_flow.raft_large.html |   943 ++
 ...vision.models.optical_flow.raft_small.html |   808 +
 ...hvision.models.quantization.googlenet.html |   806 +
 ...sion.models.quantization.inception_v3.html |   813 +
 ...sion.models.quantization.mobilenet_v2.html |   840 ++
 ...odels.quantization.mobilenet_v3_large.html |   841 ++
 ...chvision.models.quantization.resnet18.html |   807 +
 ...chvision.models.quantization.resnet50.html |   879 ++
 ....models.quantization.resnext101_32x8d.html |   879 ++
 ....models.quantization.resnext101_64x4d.html |   809 +
 ...odels.quantization.shufflenet_v2_x0_5.html |   808 +
 ...odels.quantization.shufflenet_v2_x1_0.html |   808 +
 ...odels.quantization.shufflenet_v2_x1_5.html |   810 +
 ...odels.quantization.shufflenet_v2_x2_0.html |   810 +
 .../torchvision.models.regnet_x_16gf.html     |   783 +
 .../torchvision.models.regnet_x_1_6gf.html    |   783 +
 .../torchvision.models.regnet_x_32gf.html     |   783 +
 .../torchvision.models.regnet_x_3_2gf.html    |   783 +
 .../torchvision.models.regnet_x_400mf.html    |   783 +
 .../torchvision.models.regnet_x_800mf.html    |   783 +
 .../torchvision.models.regnet_x_8gf.html      |   783 +
 .../torchvision.models.regnet_y_128gf.html    |   791 +
 .../torchvision.models.regnet_y_16gf.html     |   855 ++
 .../torchvision.models.regnet_y_1_6gf.html    |   783 +
 .../torchvision.models.regnet_y_32gf.html     |   855 ++
 .../torchvision.models.regnet_y_3_2gf.html    |   783 +
 .../torchvision.models.regnet_y_400mf.html    |   783 +
 .../torchvision.models.regnet_y_800mf.html    |   783 +
 .../torchvision.models.regnet_y_8gf.html      |   783 +
 .../torchvision.models.resnet101.html         |   790 +
 .../torchvision.models.resnet152.html         |   790 +
 .../torchvision.models.resnet18.html          |   755 +
 .../torchvision.models.resnet34.html          |   751 +
 .../torchvision.models.resnet50.html          |   790 +
 .../torchvision.models.resnext101_32x8d.html  |   785 +
 .../torchvision.models.resnext101_64x4d.html  |   754 +
 .../torchvision.models.resnext50_32x4d.html   |   785 +
 ...entation.deeplabv3_mobilenet_v3_large.html |   757 +
 ...dels.segmentation.deeplabv3_resnet101.html |   761 +
 ...odels.segmentation.deeplabv3_resnet50.html |   761 +
 ...ion.models.segmentation.fcn_resnet101.html |   763 +
 ...sion.models.segmentation.fcn_resnet50.html |   767 +
 ...egmentation.lraspp_mobilenet_v3_large.html |   763 +
 ...torchvision.models.shufflenet_v2_x0_5.html |   752 +
 ...torchvision.models.shufflenet_v2_x1_0.html |   752 +
 ...torchvision.models.shufflenet_v2_x1_5.html |   754 +
 ...torchvision.models.shufflenet_v2_x2_0.html |   754 +
 .../torchvision.models.squeezenet1_0.html     |   752 +
 .../torchvision.models.squeezenet1_1.html     |   753 +
 .../generated/torchvision.models.swin_b.html  |   752 +
 .../generated/torchvision.models.swin_s.html  |   752 +
 .../generated/torchvision.models.swin_t.html  |   752 +
 .../torchvision.models.swin_v2_b.html         |   752 +
 .../torchvision.models.swin_v2_s.html         |   752 +
 .../torchvision.models.swin_v2_t.html         |   752 +
 .../generated/torchvision.models.vgg11.html   |   751 +
 .../torchvision.models.vgg11_bn.html          |   751 +
 .../generated/torchvision.models.vgg13.html   |   751 +
 .../torchvision.models.vgg13_bn.html          |   751 +
 .../generated/torchvision.models.vgg16.html   |   785 +
 .../torchvision.models.vgg16_bn.html          |   751 +
 .../generated/torchvision.models.vgg19.html   |   751 +
 .../torchvision.models.vgg19_bn.html          |   751 +
 .../torchvision.models.video.mc3_18.html      |   755 +
 .../torchvision.models.video.mvit_v1_b.html   |   759 +
 .../torchvision.models.video.mvit_v2_s.html   |   761 +
 .../torchvision.models.video.r2plus1d_18.html |   755 +
 .../torchvision.models.video.r3d_18.html      |   755 +
 .../torchvision.models.video.s3d.html         |   758 +
 .../torchvision.models.video.swin3d_b.html    |   789 +
 .../torchvision.models.video.swin3d_s.html    |   755 +
 .../torchvision.models.video.swin3d_t.html    |   755 +
 .../torchvision.models.vit_b_16.html          |   823 +
 .../torchvision.models.vit_b_32.html          |   751 +
 .../torchvision.models.vit_h_14.html          |   791 +
 .../torchvision.models.vit_l_16.html          |   824 +
 .../torchvision.models.vit_l_32.html          |   751 +
 .../torchvision.models.wide_resnet101_2.html  |   789 +
 .../torchvision.models.wide_resnet50_2.html   |   789 +
 0.16/models/googlenet.html                    |   708 +
 0.16/models/googlenet_quant.html              |   708 +
 0.16/models/inception.html                    |   708 +
 0.16/models/inception_quant.html              |   708 +
 0.16/models/keypoint_rcnn.html                |   711 +
 0.16/models/lraspp.html                       |   712 +
 0.16/models/mask_rcnn.html                    |   716 +
 0.16/models/maxvit.html                       |   707 +
 0.16/models/mnasnet.html                      |   720 +
 0.16/models/mobilenetv2.html                  |   708 +
 0.16/models/mobilenetv2_quant.html            |   708 +
 0.16/models/mobilenetv3.html                  |   710 +
 0.16/models/mobilenetv3_quant.html            |   707 +
 0.16/models/raft.html                         |   711 +
 0.16/models/regnet.html                       |   763 +
 0.16/models/resnet.html                       |   725 +
 0.16/models/resnet_quant.html                 |   712 +
 0.16/models/resnext.html                      |   716 +
 0.16/models/resnext_quant.html                |   712 +
 0.16/models/retinanet.html                    |   713 +
 0.16/models/shufflenetv2.html                 |   717 +
 0.16/models/shufflenetv2_quant.html           |   717 +
 0.16/models/squeezenet.html                   |   712 +
 0.16/models/ssd.html                          |   711 +
 0.16/models/ssdlite.html                      |   712 +
 0.16/models/swin_transformer.html             |   730 +
 0.16/models/vgg.html                          |   728 +
 0.16/models/video_mvit.html                   |   711 +
 0.16/models/video_resnet.html                 |   718 +
 0.16/models/video_s3d.html                    |   708 +
 0.16/models/video_swin_transformer.html       |   719 +
 0.16/models/vision_transformer.html           |   725 +
 0.16/models/wide_resnet.html                  |   712 +
 0.16/objects.inv                              |   Bin 0 -> 18596 bytes
 0.16/ops.html                                 |   858 ++
 0.16/py-modindex.html                         |   688 +
 0.16/search.html                              |   683 +
 0.16/searchindex.js                           |     1 +
 0.16/training_references.html                 |   706 +
 0.16/transforms.html                          |  1648 ++
 0.16/tv_tensors.html                          |   721 +
 0.16/utils.html                               |   710 +
 1350 files changed, 577786 insertions(+)
 create mode 100644 0.16/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip
 create mode 100644 0.16/_downloads/163bc3a8b6281b990b458e3eef431d01/plot_custom_transforms.py
 create mode 100644 0.16/_downloads/175e3bb1b03a25b015d65842a3b8a883/plot_video_api.py
 create mode 100644 0.16/_downloads/1c7d3f45724a1171e328ce5cdd7bc79a/plot_scripted_tensor_transforms.py
 create mode 100644 0.16/_downloads/2753c5fd8349a28b023c919d8ddf3573/plot_repurposing_annotations.ipynb
 create mode 100644 0.16/_downloads/27d6228101da5d268ad41f8f06010df8/plot_visualization_utils.py
 create mode 100644 0.16/_downloads/3f3cbe5cc8b5758610d0ab95995b0b8c/plot_custom_transforms.ipynb
 create mode 100644 0.16/_downloads/40e082c3b2415f11f987eaa06c4d7513/plot_transforms_e2e.py
 create mode 100644 0.16/_downloads/4333051365715be174a783856419631f/plot_scripted_tensor_transforms.ipynb
 create mode 100644 0.16/_downloads/4890c38ddf15e56f86ebbdd480cb7ade/plot_custom_tv_tensors.py
 create mode 100644 0.16/_downloads/4b3151295a1f6b75ffd88837fb31f719/plot_video_api.ipynb
 create mode 100644 0.16/_downloads/5a9ea474844f53cd409e782bd87efb13/plot_cutmix_mixup.py
 create mode 100644 0.16/_downloads/6159c523793cb1afbb503259bd141f91/plot_repurposing_annotations.py
 create mode 100644 0.16/_downloads/66c906e53334aa968837db20de95f3b2/plot_cutmix_mixup.ipynb
 create mode 100644 0.16/_downloads/6acd2f50f8be085d0a8ac5874560f268/plot_optical_flow.py
 create mode 100644 0.16/_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip
 create mode 100644 0.16/_downloads/8b062cf0fa8a991b8cff18fa361c8ffe/plot_custom_tv_tensors.ipynb
 create mode 100644 0.16/_downloads/8c66a0169c2c41937d8db4238842fe5b/plot_optical_flow.ipynb
 create mode 100644 0.16/_downloads/8faa1256268baabef0c4b947064f34f4/plot_transforms_getting_started.py
 create mode 100644 0.16/_downloads/92763696a071eab76a5ea62a1d9f8c5c/plot_tv_tensors.py
 create mode 100644 0.16/_downloads/95e7320166df5d4ddcbdd5ea64a5c98b/plot_visualization_utils.ipynb
 create mode 100644 0.16/_downloads/a8ef8e39164e980a89a9afdfc51d961e/plot_transforms_illustrations.ipynb
 create mode 100644 0.16/_downloads/bf35a573bbfe2773c19cd458bf8508c4/plot_tv_tensors.ipynb
 create mode 100644 0.16/_downloads/d1e0b3022ec2c789e633f49bb9dd0cf4/plot_transforms_getting_started.ipynb
 create mode 100644 0.16/_downloads/f9ef41d32e46834ba2f93a1d3c4b4568/plot_transforms_illustrations.py
 create mode 100644 0.16/_downloads/fd385b39c48cf3b5eb72e901830b8621/plot_transforms_e2e.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_custom_transforms.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_optical_flow.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_tv_tensors.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_video_api.ipynb
 create mode 100644 0.16/_generated_ipynb_notebooks/plot_visualization_utils.ipynb
 create mode 100644 0.16/_images/sphx_glr_plot_custom_transforms_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_custom_tv_tensors_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_cutmix_mixup_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_optical_flow_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_optical_flow_002.png
 create mode 100644 0.16/_images/sphx_glr_plot_optical_flow_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_repurposing_annotations_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_repurposing_annotations_002.png
 create mode 100644 0.16/_images/sphx_glr_plot_repurposing_annotations_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_scripted_tensor_transforms_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_e2e_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_e2e_002.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_e2e_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_getting_started_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_getting_started_002.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_getting_started_003.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_getting_started_004.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_getting_started_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_002.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_003.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_004.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_005.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_006.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_007.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_008.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_009.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_010.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_011.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_012.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_013.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_014.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_015.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_016.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_017.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_018.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_019.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_020.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_021.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_022.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_023.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_024.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_025.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_026.png
 create mode 100644 0.16/_images/sphx_glr_plot_transforms_illustrations_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_tv_tensors_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_video_api_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_video_api_thumb.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_001.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_002.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_003.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_004.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_005.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_006.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_007.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_008.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_009.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_010.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_011.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_012.png
 create mode 100644 0.16/_images/sphx_glr_plot_visualization_utils_thumb.png
 create mode 100644 0.16/_modules/index.html
 create mode 100644 0.16/_modules/torchvision.html
 create mode 100644 0.16/_modules/torchvision/datasets/_optical_flow.html
 create mode 100644 0.16/_modules/torchvision/datasets/_stereo_matching.html
 create mode 100644 0.16/_modules/torchvision/datasets/caltech.html
 create mode 100644 0.16/_modules/torchvision/datasets/celeba.html
 create mode 100644 0.16/_modules/torchvision/datasets/cifar.html
 create mode 100644 0.16/_modules/torchvision/datasets/cityscapes.html
 create mode 100644 0.16/_modules/torchvision/datasets/coco.html
 create mode 100644 0.16/_modules/torchvision/datasets/country211.html
 create mode 100644 0.16/_modules/torchvision/datasets/dtd.html
 create mode 100644 0.16/_modules/torchvision/datasets/eurosat.html
 create mode 100644 0.16/_modules/torchvision/datasets/fakedata.html
 create mode 100644 0.16/_modules/torchvision/datasets/fer2013.html
 create mode 100644 0.16/_modules/torchvision/datasets/fgvc_aircraft.html
 create mode 100644 0.16/_modules/torchvision/datasets/flickr.html
 create mode 100644 0.16/_modules/torchvision/datasets/flowers102.html
 create mode 100644 0.16/_modules/torchvision/datasets/folder.html
 create mode 100644 0.16/_modules/torchvision/datasets/food101.html
 create mode 100644 0.16/_modules/torchvision/datasets/gtsrb.html
 create mode 100644 0.16/_modules/torchvision/datasets/hmdb51.html
 create mode 100644 0.16/_modules/torchvision/datasets/imagenet.html
 create mode 100644 0.16/_modules/torchvision/datasets/inaturalist.html
 create mode 100644 0.16/_modules/torchvision/datasets/kinetics.html
 create mode 100644 0.16/_modules/torchvision/datasets/kitti.html
 create mode 100644 0.16/_modules/torchvision/datasets/lfw.html
 create mode 100644 0.16/_modules/torchvision/datasets/lsun.html
 create mode 100644 0.16/_modules/torchvision/datasets/mnist.html
 create mode 100644 0.16/_modules/torchvision/datasets/moving_mnist.html
 create mode 100644 0.16/_modules/torchvision/datasets/omniglot.html
 create mode 100644 0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
 create mode 100644 0.16/_modules/torchvision/datasets/pcam.html
 create mode 100644 0.16/_modules/torchvision/datasets/phototour.html
 create mode 100644 0.16/_modules/torchvision/datasets/places365.html
 create mode 100644 0.16/_modules/torchvision/datasets/rendered_sst2.html
 create mode 100644 0.16/_modules/torchvision/datasets/sbd.html
 create mode 100644 0.16/_modules/torchvision/datasets/sbu.html
 create mode 100644 0.16/_modules/torchvision/datasets/semeion.html
 create mode 100644 0.16/_modules/torchvision/datasets/stanford_cars.html
 create mode 100644 0.16/_modules/torchvision/datasets/stl10.html
 create mode 100644 0.16/_modules/torchvision/datasets/sun397.html
 create mode 100644 0.16/_modules/torchvision/datasets/svhn.html
 create mode 100644 0.16/_modules/torchvision/datasets/ucf101.html
 create mode 100644 0.16/_modules/torchvision/datasets/usps.html
 create mode 100644 0.16/_modules/torchvision/datasets/vision.html
 create mode 100644 0.16/_modules/torchvision/datasets/voc.html
 create mode 100644 0.16/_modules/torchvision/datasets/widerface.html
 create mode 100644 0.16/_modules/torchvision/io/image.html
 create mode 100644 0.16/_modules/torchvision/io/video.html
 create mode 100644 0.16/_modules/torchvision/io/video_reader.html
 create mode 100644 0.16/_modules/torchvision/models/_api.html
 create mode 100644 0.16/_modules/torchvision/models/alexnet.html
 create mode 100644 0.16/_modules/torchvision/models/convnext.html
 create mode 100644 0.16/_modules/torchvision/models/densenet.html
 create mode 100644 0.16/_modules/torchvision/models/detection/faster_rcnn.html
 create mode 100644 0.16/_modules/torchvision/models/detection/fcos.html
 create mode 100644 0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
 create mode 100644 0.16/_modules/torchvision/models/detection/mask_rcnn.html
 create mode 100644 0.16/_modules/torchvision/models/detection/retinanet.html
 create mode 100644 0.16/_modules/torchvision/models/detection/ssd.html
 create mode 100644 0.16/_modules/torchvision/models/detection/ssdlite.html
 create mode 100644 0.16/_modules/torchvision/models/efficientnet.html
 create mode 100644 0.16/_modules/torchvision/models/feature_extraction.html
 create mode 100644 0.16/_modules/torchvision/models/googlenet.html
 create mode 100644 0.16/_modules/torchvision/models/inception.html
 create mode 100644 0.16/_modules/torchvision/models/maxvit.html
 create mode 100644 0.16/_modules/torchvision/models/mnasnet.html
 create mode 100644 0.16/_modules/torchvision/models/mobilenetv2.html
 create mode 100644 0.16/_modules/torchvision/models/mobilenetv3.html
 create mode 100644 0.16/_modules/torchvision/models/optical_flow/raft.html
 create mode 100644 0.16/_modules/torchvision/models/quantization/googlenet.html
 create mode 100644 0.16/_modules/torchvision/models/quantization/inception.html
 create mode 100644 0.16/_modules/torchvision/models/quantization/mobilenetv2.html
 create mode 100644 0.16/_modules/torchvision/models/quantization/mobilenetv3.html
 create mode 100644 0.16/_modules/torchvision/models/quantization/resnet.html
 create mode 100644 0.16/_modules/torchvision/models/quantization/shufflenetv2.html
 create mode 100644 0.16/_modules/torchvision/models/regnet.html
 create mode 100644 0.16/_modules/torchvision/models/resnet.html
 create mode 100644 0.16/_modules/torchvision/models/segmentation/deeplabv3.html
 create mode 100644 0.16/_modules/torchvision/models/segmentation/fcn.html
 create mode 100644 0.16/_modules/torchvision/models/segmentation/lraspp.html
 create mode 100644 0.16/_modules/torchvision/models/shufflenetv2.html
 create mode 100644 0.16/_modules/torchvision/models/squeezenet.html
 create mode 100644 0.16/_modules/torchvision/models/swin_transformer.html
 create mode 100644 0.16/_modules/torchvision/models/vgg.html
 create mode 100644 0.16/_modules/torchvision/models/video/mvit.html
 create mode 100644 0.16/_modules/torchvision/models/video/resnet.html
 create mode 100644 0.16/_modules/torchvision/models/video/s3d.html
 create mode 100644 0.16/_modules/torchvision/models/video/swin_transformer.html
 create mode 100644 0.16/_modules/torchvision/models/vision_transformer.html
 create mode 100644 0.16/_modules/torchvision/ops/boxes.html
 create mode 100644 0.16/_modules/torchvision/ops/ciou_loss.html
 create mode 100644 0.16/_modules/torchvision/ops/deform_conv.html
 create mode 100644 0.16/_modules/torchvision/ops/diou_loss.html
 create mode 100644 0.16/_modules/torchvision/ops/drop_block.html
 create mode 100644 0.16/_modules/torchvision/ops/feature_pyramid_network.html
 create mode 100644 0.16/_modules/torchvision/ops/focal_loss.html
 create mode 100644 0.16/_modules/torchvision/ops/giou_loss.html
 create mode 100644 0.16/_modules/torchvision/ops/misc.html
 create mode 100644 0.16/_modules/torchvision/ops/poolers.html
 create mode 100644 0.16/_modules/torchvision/ops/ps_roi_align.html
 create mode 100644 0.16/_modules/torchvision/ops/ps_roi_pool.html
 create mode 100644 0.16/_modules/torchvision/ops/roi_align.html
 create mode 100644 0.16/_modules/torchvision/ops/roi_pool.html
 create mode 100644 0.16/_modules/torchvision/ops/stochastic_depth.html
 create mode 100644 0.16/_modules/torchvision/transforms/autoaugment.html
 create mode 100644 0.16/_modules/torchvision/transforms/functional.html
 create mode 100644 0.16/_modules/torchvision/transforms/transforms.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_augment.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_auto_augment.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_color.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_container.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_deprecated.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_geometry.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_meta.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_misc.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_temporal.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/_type_conversion.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_augment.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_color.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_meta.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_misc.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
 create mode 100644 0.16/_modules/torchvision/transforms/v2/functional/_utils.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors/_image.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors/_mask.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
 create mode 100644 0.16/_modules/torchvision/tv_tensors/_video.html
 create mode 100644 0.16/_modules/torchvision/utils.html
 create mode 100644 0.16/_sources/auto_examples/index.rst.txt
 create mode 100644 0.16/_sources/auto_examples/others/index.rst.txt
 create mode 100644 0.16/_sources/auto_examples/others/plot_optical_flow.rst.txt
 create mode 100644 0.16/_sources/auto_examples/others/plot_repurposing_annotations.rst.txt
 create mode 100644 0.16/_sources/auto_examples/others/plot_scripted_tensor_transforms.rst.txt
 create mode 100644 0.16/_sources/auto_examples/others/plot_video_api.rst.txt
 create mode 100644 0.16/_sources/auto_examples/others/plot_visualization_utils.rst.txt
 create mode 100644 0.16/_sources/auto_examples/others/sg_execution_times.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/index.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/plot_custom_transforms.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/plot_custom_tv_tensors.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/plot_cutmix_mixup.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/plot_transforms_e2e.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/plot_transforms_getting_started.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/plot_transforms_illustrations.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/plot_tv_tensors.rst.txt
 create mode 100644 0.16/_sources/auto_examples/transforms/sg_execution_times.rst.txt
 create mode 100644 0.16/_sources/datasets.rst.txt
 create mode 100644 0.16/_sources/feature_extraction.rst.txt
 create mode 100644 0.16/_sources/generated/classification_quant_table.rst.txt
 create mode 100644 0.16/_sources/generated/classification_table.rst.txt
 create mode 100644 0.16/_sources/generated/detection_keypoint_table.rst.txt
 create mode 100644 0.16/_sources/generated/detection_table.rst.txt
 create mode 100644 0.16/_sources/generated/instance_segmentation_table.rst.txt
 create mode 100644 0.16/_sources/generated/segmentation_table.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.CIFAR10.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.CIFAR100.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.CREStereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Caltech101.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Caltech256.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.CarlaStereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.CelebA.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Cityscapes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.CocoCaptions.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.CocoDetection.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Country211.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.DTD.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.DatasetFolder.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.EMNIST.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.ETH3DStereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.EuroSAT.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.FER2013.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.FGVCAircraft.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.FakeData.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.FallingThingsStereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.FashionMNIST.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Flickr30k.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Flickr8k.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Flowers102.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.FlyingChairs.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.FlyingThings3D.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Food101.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.GTSRB.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.HD1K.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.HMDB51.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.INaturalist.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.ImageFolder.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.ImageNet.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.InStereo2k.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.KMNIST.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Kinetics.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Kitti.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Kitti2012Stereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Kitti2015Stereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.KittiFlow.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.LFWPairs.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.LFWPeople.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.LSUN.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.MNIST.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Middlebury2014Stereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.MovingMNIST.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Omniglot.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.OxfordIIITPet.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.PCAM.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.PhotoTour.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Places365.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.QMNIST.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.RenderedSST2.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.SBDataset.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.SBU.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.SEMEION.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.STL10.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.SUN397.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.SVHN.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.SceneFlowStereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.Sintel.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.SintelStereo.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.StanfordCars.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.UCF101.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.USPS.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.VOCDetection.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.VOCSegmentation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.VisionDataset.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.WIDERFace.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.ImageReadMode.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.VideoReader.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.decode_image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.decode_jpeg.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.decode_png.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.encode_jpeg.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.encode_png.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.read_file.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.read_image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.read_video.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.read_video_timestamps.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.write_file.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.write_jpeg.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.write_png.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.io.write_video.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.models.feature_extraction.create_feature_extractor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.models.feature_extraction.get_graph_node_names.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.models.get_model.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.models.get_model_weights.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.models.get_weight.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.models.list_models.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.Conv2dNormActivation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.Conv3dNormActivation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.DeformConv2d.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.DropBlock2d.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.DropBlock3d.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.FeaturePyramidNetwork.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.FrozenBatchNorm2d.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.MLP.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.MultiScaleRoIAlign.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.PSRoIAlign.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.PSRoIPool.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.Permute.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.RoIAlign.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.RoIPool.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.SqueezeExcitation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.StochasticDepth.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.batched_nms.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.box_area.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.box_convert.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.box_iou.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.clip_boxes_to_image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.complete_box_iou.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.complete_box_iou_loss.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.deform_conv2d.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.distance_box_iou.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.distance_box_iou_loss.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.drop_block2d.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.drop_block3d.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.generalized_box_iou.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.generalized_box_iou_loss.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.masks_to_boxes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.nms.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.ps_roi_align.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.ps_roi_pool.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.remove_small_boxes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.roi_align.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.roi_pool.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.sigmoid_focal_loss.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.ops.stochastic_depth.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.AugMix.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.AutoAugment.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.AutoAugmentPolicy.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.CenterCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.ColorJitter.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.Compose.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.ConvertImageDtype.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.ElasticTransform.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.FiveCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.GaussianBlur.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.Grayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.Lambda.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.LinearTransformation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.Normalize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.PILToTensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.Pad.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandAugment.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomAdjustSharpness.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomAffine.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomApply.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomAutocontrast.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomChoice.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomEqualize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomErasing.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomGrayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomHorizontalFlip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomInvert.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomOrder.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomPerspective.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomPosterize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomResizedCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomRotation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomSolarize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.RandomVerticalFlip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.Resize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.TenCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.ToPILImage.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.ToTensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.TrivialAugmentWide.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.adjust_brightness.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.adjust_contrast.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.adjust_gamma.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.adjust_hue.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.adjust_saturation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.adjust_sharpness.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.affine.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.autocontrast.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.center_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.convert_image_dtype.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.equalize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.erase.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.five_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.gaussian_blur.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.get_dimensions.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.get_image_num_channels.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.get_image_size.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.hflip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.invert.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.normalize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.pad.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.perspective.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.pil_to_tensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.posterize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.resize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.resized_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.rgb_to_grayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.rotate.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.solarize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.ten_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.to_grayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.to_pil_image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.to_tensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.functional.vflip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.AugMix.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.AutoAugment.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.CenterCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ClampBoundingBoxes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ColorJitter.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.Compose.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ConvertImageDtype.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.CutMix.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ElasticTransform.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.FiveCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.GaussianBlur.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.Grayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.Lambda.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.LinearTransformation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.MixUp.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.Normalize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.PILToTensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.Pad.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandAugment.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomAdjustSharpness.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomAffine.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomApply.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomAutocontrast.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomChannelPermutation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomChoice.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomEqualize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomErasing.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomGrayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomHorizontalFlip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomInvert.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomIoUCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomOrder.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomPerspective.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomPhotometricDistort.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomPosterize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomResize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomResizedCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomRotation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomShortestSize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomSolarize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomVerticalFlip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.RandomZoomOut.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.Resize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ScaleJitter.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.TenCrop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ToDtype.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ToImage.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ToPILImage.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ToPureTensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.ToTensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.TrivialAugmentWide.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.UniformTemporalSubsample.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_brightness.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_contrast.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_gamma.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_hue.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_saturation.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_sharpness.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.affine.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.autocontrast.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.center_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.convert_image_dtype.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.elastic.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.equalize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.erase.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.five_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.gaussian_blur.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.horizontal_flip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.invert.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.normalize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.pad.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.permute_channels.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.perspective.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.pil_to_tensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.posterize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.register_kernel.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.resize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.resized_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.rotate.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.solarize.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.ten_crop.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.to_dtype.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.to_grayscale.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.to_image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.to_pil_image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.to_tensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.transforms.v2.functional.vertical_flip.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxFormat.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.Image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.Mask.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.TVTensor.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.Video.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.set_return_type.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.tv_tensors.wrap.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.utils.draw_bounding_boxes.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.utils.draw_keypoints.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.utils.draw_segmentation_masks.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.utils.flow_to_image.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.utils.make_grid.rst.txt
 create mode 100644 0.16/_sources/generated/torchvision.utils.save_image.rst.txt
 create mode 100644 0.16/_sources/generated/video_table.rst.txt
 create mode 100644 0.16/_sources/index.rst.txt
 create mode 100644 0.16/_sources/io.rst.txt
 create mode 100644 0.16/_sources/models.rst.txt
 create mode 100644 0.16/_sources/models/alexnet.rst.txt
 create mode 100644 0.16/_sources/models/convnext.rst.txt
 create mode 100644 0.16/_sources/models/deeplabv3.rst.txt
 create mode 100644 0.16/_sources/models/densenet.rst.txt
 create mode 100644 0.16/_sources/models/efficientnet.rst.txt
 create mode 100644 0.16/_sources/models/efficientnetv2.rst.txt
 create mode 100644 0.16/_sources/models/faster_rcnn.rst.txt
 create mode 100644 0.16/_sources/models/fcn.rst.txt
 create mode 100644 0.16/_sources/models/fcos.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.alexnet.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.convnext_base.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.convnext_large.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.convnext_small.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.convnext_tiny.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.densenet121.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.densenet161.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.densenet169.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.densenet201.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.fcos_resnet50_fpn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.ssd300_vgg16.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b0.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b1.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b2.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b3.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b4.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b5.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b6.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_b7.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_v2_l.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_v2_m.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.efficientnet_v2_s.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.googlenet.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.inception_v3.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.maxvit_t.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.mnasnet0_5.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.mnasnet0_75.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.mnasnet1_0.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.mnasnet1_3.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.mobilenet_v2.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.mobilenet_v3_large.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.mobilenet_v3_small.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.optical_flow.raft_large.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.optical_flow.raft_small.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.googlenet.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.inception_v3.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v2.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v3_large.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.resnet18.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.resnet50.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.resnext101_32x8d.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.resnext101_64x4d.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_x_16gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_x_1_6gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_x_32gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_x_3_2gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_x_400mf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_x_800mf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_x_8gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_128gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_16gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_1_6gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_32gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_3_2gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_400mf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_800mf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.regnet_y_8gf.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnet101.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnet152.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnet18.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnet34.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnet50.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnext101_32x8d.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnext101_64x4d.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.resnext50_32x4d.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet101.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet50.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x0_5.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_0.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_5.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x2_0.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.squeezenet1_0.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.squeezenet1_1.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.swin_b.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.swin_s.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.swin_t.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.swin_v2_b.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.swin_v2_s.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.swin_v2_t.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg11.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg11_bn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg13.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg13_bn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg16.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg16_bn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg19.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vgg19_bn.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.mc3_18.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.mvit_v1_b.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.mvit_v2_s.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.r2plus1d_18.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.r3d_18.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.s3d.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.swin3d_b.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.swin3d_s.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.video.swin3d_t.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vit_b_16.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vit_b_32.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vit_h_14.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vit_l_16.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.vit_l_32.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.wide_resnet101_2.rst.txt
 create mode 100644 0.16/_sources/models/generated/torchvision.models.wide_resnet50_2.rst.txt
 create mode 100644 0.16/_sources/models/googlenet.rst.txt
 create mode 100644 0.16/_sources/models/googlenet_quant.rst.txt
 create mode 100644 0.16/_sources/models/inception.rst.txt
 create mode 100644 0.16/_sources/models/inception_quant.rst.txt
 create mode 100644 0.16/_sources/models/keypoint_rcnn.rst.txt
 create mode 100644 0.16/_sources/models/lraspp.rst.txt
 create mode 100644 0.16/_sources/models/mask_rcnn.rst.txt
 create mode 100644 0.16/_sources/models/maxvit.rst.txt
 create mode 100644 0.16/_sources/models/mnasnet.rst.txt
 create mode 100644 0.16/_sources/models/mobilenetv2.rst.txt
 create mode 100644 0.16/_sources/models/mobilenetv2_quant.rst.txt
 create mode 100644 0.16/_sources/models/mobilenetv3.rst.txt
 create mode 100644 0.16/_sources/models/mobilenetv3_quant.rst.txt
 create mode 100644 0.16/_sources/models/raft.rst.txt
 create mode 100644 0.16/_sources/models/regnet.rst.txt
 create mode 100644 0.16/_sources/models/resnet.rst.txt
 create mode 100644 0.16/_sources/models/resnet_quant.rst.txt
 create mode 100644 0.16/_sources/models/resnext.rst.txt
 create mode 100644 0.16/_sources/models/resnext_quant.rst.txt
 create mode 100644 0.16/_sources/models/retinanet.rst.txt
 create mode 100644 0.16/_sources/models/shufflenetv2.rst.txt
 create mode 100644 0.16/_sources/models/shufflenetv2_quant.rst.txt
 create mode 100644 0.16/_sources/models/squeezenet.rst.txt
 create mode 100644 0.16/_sources/models/ssd.rst.txt
 create mode 100644 0.16/_sources/models/ssdlite.rst.txt
 create mode 100644 0.16/_sources/models/swin_transformer.rst.txt
 create mode 100644 0.16/_sources/models/vgg.rst.txt
 create mode 100644 0.16/_sources/models/video_mvit.rst.txt
 create mode 100644 0.16/_sources/models/video_resnet.rst.txt
 create mode 100644 0.16/_sources/models/video_s3d.rst.txt
 create mode 100644 0.16/_sources/models/video_swin_transformer.rst.txt
 create mode 100644 0.16/_sources/models/vision_transformer.rst.txt
 create mode 100644 0.16/_sources/models/wide_resnet.rst.txt
 create mode 100644 0.16/_sources/ops.rst.txt
 create mode 100644 0.16/_sources/training_references.rst.txt
 create mode 100644 0.16/_sources/transforms.rst.txt
 create mode 100644 0.16/_sources/tv_tensors.rst.txt
 create mode 100644 0.16/_sources/utils.rst.txt
 create mode 100644 0.16/_static/_sphinx_javascript_frameworks_compat.js
 create mode 100644 0.16/_static/basic.css
 create mode 100644 0.16/_static/binder_badge_logo.svg
 create mode 100644 0.16/_static/broken_example.png
 create mode 100644 0.16/_static/check-solid.svg
 create mode 100644 0.16/_static/clipboard.min.js
 create mode 100644 0.16/_static/copy-button.svg
 create mode 100644 0.16/_static/copybutton.css
 create mode 100644 0.16/_static/copybutton.js
 create mode 100644 0.16/_static/copybutton_funcs.js
 create mode 100644 0.16/_static/css/custom_torchvision.css
 create mode 100644 0.16/_static/css/theme.css
 create mode 100644 0.16/_static/doctools.js
 create mode 100644 0.16/_static/documentation_options.js
 create mode 100644 0.16/_static/file.png
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-bold-italic.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-bold-italic.woff2
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-bold.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-bold.woff2
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-book-italic.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-book-italic.woff2
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-book.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-book.woff2
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-light-italic.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-light-italic.woff2
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-light.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-light.woff2
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-medium-italic.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-medium-italic.woff2
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-medium.woff
 create mode 100644 0.16/_static/fonts/FreightSans/freight-sans-medium.woff2
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Light.woff
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Light.woff2
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Regular.woff
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Regular.woff2
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff
 create mode 100644 0.16/_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2
 create mode 100644 0.16/_static/images/arrow-down-orange.svg
 create mode 100644 0.16/_static/images/arrow-right-with-tail.svg
 create mode 100644 0.16/_static/images/chevron-down-black.svg
 create mode 100644 0.16/_static/images/chevron-down-grey.svg
 create mode 100644 0.16/_static/images/chevron-down-orange.svg
 create mode 100644 0.16/_static/images/chevron-down-white.svg
 create mode 100644 0.16/_static/images/chevron-right-orange.svg
 create mode 100644 0.16/_static/images/chevron-right-white.svg
 create mode 100644 0.16/_static/images/home-footer-background.jpg
 create mode 100644 0.16/_static/images/icon-close.svg
 create mode 100644 0.16/_static/images/icon-menu-dots-dark.svg
 create mode 100644 0.16/_static/images/logo-dark.svg
 create mode 100644 0.16/_static/images/logo-facebook-dark.svg
 create mode 100644 0.16/_static/images/logo-icon.svg
 create mode 100644 0.16/_static/images/logo-twitter-dark.svg
 create mode 100644 0.16/_static/images/logo-youtube-dark.svg
 create mode 100644 0.16/_static/images/logo.svg
 create mode 100644 0.16/_static/images/pytorch-colab.svg
 create mode 100644 0.16/_static/images/pytorch-download.svg
 create mode 100644 0.16/_static/images/pytorch-github.svg
 create mode 100644 0.16/_static/images/pytorch-x.svg
 create mode 100644 0.16/_static/images/search-icon.svg
 create mode 100644 0.16/_static/images/view-page-source-icon.svg
 create mode 100644 0.16/_static/img/pytorch-logo-dark.png
 create mode 100644 0.16/_static/img/pytorch-logo-dark.svg
 create mode 100644 0.16/_static/img/pytorch-logo-flame.png
 create mode 100644 0.16/_static/img/pytorch-logo-flame.svg
 create mode 100644 0.16/_static/jquery-3.6.0.js
 create mode 100644 0.16/_static/jquery.js
 create mode 100644 0.16/_static/js/modernizr.min.js
 create mode 100644 0.16/_static/js/theme.js
 create mode 100644 0.16/_static/js/vendor/anchor.min.js
 create mode 100644 0.16/_static/js/vendor/bootstrap.min.js
 create mode 100644 0.16/_static/js/vendor/popper.min.js
 create mode 100644 0.16/_static/jupyterlite_badge_logo.svg
 create mode 100644 0.16/_static/language_data.js
 create mode 100644 0.16/_static/minus.png
 create mode 100644 0.16/_static/no_image.png
 create mode 100644 0.16/_static/plus.png
 create mode 100644 0.16/_static/pygments.css
 create mode 100644 0.16/_static/pytorch-logo-dark.svg
 create mode 100644 0.16/_static/searchtools.js
 create mode 100644 0.16/_static/sg_gallery-binder.css
 create mode 100644 0.16/_static/sg_gallery-dataframe.css
 create mode 100644 0.16/_static/sg_gallery-rendered-html.css
 create mode 100644 0.16/_static/sg_gallery.css
 create mode 100644 0.16/_static/underscore-1.13.1.js
 create mode 100644 0.16/_static/underscore.js
 create mode 100644 0.16/auto_examples/index.html
 create mode 100644 0.16/auto_examples/others/index.html
 create mode 100644 0.16/auto_examples/others/plot_optical_flow.html
 create mode 100644 0.16/auto_examples/others/plot_repurposing_annotations.html
 create mode 100644 0.16/auto_examples/others/plot_scripted_tensor_transforms.html
 create mode 100644 0.16/auto_examples/others/plot_video_api.html
 create mode 100644 0.16/auto_examples/others/plot_visualization_utils.html
 create mode 100644 0.16/auto_examples/others/sg_execution_times.html
 create mode 100644 0.16/auto_examples/transforms/index.html
 create mode 100644 0.16/auto_examples/transforms/plot_custom_transforms.html
 create mode 100644 0.16/auto_examples/transforms/plot_custom_tv_tensors.html
 create mode 100644 0.16/auto_examples/transforms/plot_cutmix_mixup.html
 create mode 100644 0.16/auto_examples/transforms/plot_transforms_e2e.html
 create mode 100644 0.16/auto_examples/transforms/plot_transforms_getting_started.html
 create mode 100644 0.16/auto_examples/transforms/plot_transforms_illustrations.html
 create mode 100644 0.16/auto_examples/transforms/plot_tv_tensors.html
 create mode 100644 0.16/auto_examples/transforms/sg_execution_times.html
 create mode 100644 0.16/datasets.html
 create mode 100644 0.16/feature_extraction.html
 create mode 100644 0.16/generated/classification_quant_table.html
 create mode 100644 0.16/generated/classification_table.html
 create mode 100644 0.16/generated/detection_keypoint_table.html
 create mode 100644 0.16/generated/detection_table.html
 create mode 100644 0.16/generated/instance_segmentation_table.html
 create mode 100644 0.16/generated/segmentation_table.html
 create mode 100644 0.16/generated/torchvision.datasets.CIFAR10.html
 create mode 100644 0.16/generated/torchvision.datasets.CIFAR100.html
 create mode 100644 0.16/generated/torchvision.datasets.CREStereo.html
 create mode 100644 0.16/generated/torchvision.datasets.Caltech101.html
 create mode 100644 0.16/generated/torchvision.datasets.Caltech256.html
 create mode 100644 0.16/generated/torchvision.datasets.CarlaStereo.html
 create mode 100644 0.16/generated/torchvision.datasets.CelebA.html
 create mode 100644 0.16/generated/torchvision.datasets.Cityscapes.html
 create mode 100644 0.16/generated/torchvision.datasets.CocoCaptions.html
 create mode 100644 0.16/generated/torchvision.datasets.CocoDetection.html
 create mode 100644 0.16/generated/torchvision.datasets.Country211.html
 create mode 100644 0.16/generated/torchvision.datasets.DTD.html
 create mode 100644 0.16/generated/torchvision.datasets.DatasetFolder.html
 create mode 100644 0.16/generated/torchvision.datasets.EMNIST.html
 create mode 100644 0.16/generated/torchvision.datasets.ETH3DStereo.html
 create mode 100644 0.16/generated/torchvision.datasets.EuroSAT.html
 create mode 100644 0.16/generated/torchvision.datasets.FER2013.html
 create mode 100644 0.16/generated/torchvision.datasets.FGVCAircraft.html
 create mode 100644 0.16/generated/torchvision.datasets.FakeData.html
 create mode 100644 0.16/generated/torchvision.datasets.FallingThingsStereo.html
 create mode 100644 0.16/generated/torchvision.datasets.FashionMNIST.html
 create mode 100644 0.16/generated/torchvision.datasets.Flickr30k.html
 create mode 100644 0.16/generated/torchvision.datasets.Flickr8k.html
 create mode 100644 0.16/generated/torchvision.datasets.Flowers102.html
 create mode 100644 0.16/generated/torchvision.datasets.FlyingChairs.html
 create mode 100644 0.16/generated/torchvision.datasets.FlyingThings3D.html
 create mode 100644 0.16/generated/torchvision.datasets.Food101.html
 create mode 100644 0.16/generated/torchvision.datasets.GTSRB.html
 create mode 100644 0.16/generated/torchvision.datasets.HD1K.html
 create mode 100644 0.16/generated/torchvision.datasets.HMDB51.html
 create mode 100644 0.16/generated/torchvision.datasets.INaturalist.html
 create mode 100644 0.16/generated/torchvision.datasets.ImageFolder.html
 create mode 100644 0.16/generated/torchvision.datasets.ImageNet.html
 create mode 100644 0.16/generated/torchvision.datasets.InStereo2k.html
 create mode 100644 0.16/generated/torchvision.datasets.KMNIST.html
 create mode 100644 0.16/generated/torchvision.datasets.Kinetics.html
 create mode 100644 0.16/generated/torchvision.datasets.Kitti.html
 create mode 100644 0.16/generated/torchvision.datasets.Kitti2012Stereo.html
 create mode 100644 0.16/generated/torchvision.datasets.Kitti2015Stereo.html
 create mode 100644 0.16/generated/torchvision.datasets.KittiFlow.html
 create mode 100644 0.16/generated/torchvision.datasets.LFWPairs.html
 create mode 100644 0.16/generated/torchvision.datasets.LFWPeople.html
 create mode 100644 0.16/generated/torchvision.datasets.LSUN.html
 create mode 100644 0.16/generated/torchvision.datasets.MNIST.html
 create mode 100644 0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
 create mode 100644 0.16/generated/torchvision.datasets.MovingMNIST.html
 create mode 100644 0.16/generated/torchvision.datasets.Omniglot.html
 create mode 100644 0.16/generated/torchvision.datasets.OxfordIIITPet.html
 create mode 100644 0.16/generated/torchvision.datasets.PCAM.html
 create mode 100644 0.16/generated/torchvision.datasets.PhotoTour.html
 create mode 100644 0.16/generated/torchvision.datasets.Places365.html
 create mode 100644 0.16/generated/torchvision.datasets.QMNIST.html
 create mode 100644 0.16/generated/torchvision.datasets.RenderedSST2.html
 create mode 100644 0.16/generated/torchvision.datasets.SBDataset.html
 create mode 100644 0.16/generated/torchvision.datasets.SBU.html
 create mode 100644 0.16/generated/torchvision.datasets.SEMEION.html
 create mode 100644 0.16/generated/torchvision.datasets.STL10.html
 create mode 100644 0.16/generated/torchvision.datasets.SUN397.html
 create mode 100644 0.16/generated/torchvision.datasets.SVHN.html
 create mode 100644 0.16/generated/torchvision.datasets.SceneFlowStereo.html
 create mode 100644 0.16/generated/torchvision.datasets.Sintel.html
 create mode 100644 0.16/generated/torchvision.datasets.SintelStereo.html
 create mode 100644 0.16/generated/torchvision.datasets.StanfordCars.html
 create mode 100644 0.16/generated/torchvision.datasets.UCF101.html
 create mode 100644 0.16/generated/torchvision.datasets.USPS.html
 create mode 100644 0.16/generated/torchvision.datasets.VOCDetection.html
 create mode 100644 0.16/generated/torchvision.datasets.VOCSegmentation.html
 create mode 100644 0.16/generated/torchvision.datasets.VisionDataset.html
 create mode 100644 0.16/generated/torchvision.datasets.WIDERFace.html
 create mode 100644 0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
 create mode 100644 0.16/generated/torchvision.io.ImageReadMode.html
 create mode 100644 0.16/generated/torchvision.io.VideoReader.html
 create mode 100644 0.16/generated/torchvision.io.decode_image.html
 create mode 100644 0.16/generated/torchvision.io.decode_jpeg.html
 create mode 100644 0.16/generated/torchvision.io.decode_png.html
 create mode 100644 0.16/generated/torchvision.io.encode_jpeg.html
 create mode 100644 0.16/generated/torchvision.io.encode_png.html
 create mode 100644 0.16/generated/torchvision.io.read_file.html
 create mode 100644 0.16/generated/torchvision.io.read_image.html
 create mode 100644 0.16/generated/torchvision.io.read_video.html
 create mode 100644 0.16/generated/torchvision.io.read_video_timestamps.html
 create mode 100644 0.16/generated/torchvision.io.write_file.html
 create mode 100644 0.16/generated/torchvision.io.write_jpeg.html
 create mode 100644 0.16/generated/torchvision.io.write_png.html
 create mode 100644 0.16/generated/torchvision.io.write_video.html
 create mode 100644 0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
 create mode 100644 0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
 create mode 100644 0.16/generated/torchvision.models.get_model.html
 create mode 100644 0.16/generated/torchvision.models.get_model_weights.html
 create mode 100644 0.16/generated/torchvision.models.get_weight.html
 create mode 100644 0.16/generated/torchvision.models.list_models.html
 create mode 100644 0.16/generated/torchvision.ops.Conv2dNormActivation.html
 create mode 100644 0.16/generated/torchvision.ops.Conv3dNormActivation.html
 create mode 100644 0.16/generated/torchvision.ops.DeformConv2d.html
 create mode 100644 0.16/generated/torchvision.ops.DropBlock2d.html
 create mode 100644 0.16/generated/torchvision.ops.DropBlock3d.html
 create mode 100644 0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
 create mode 100644 0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
 create mode 100644 0.16/generated/torchvision.ops.MLP.html
 create mode 100644 0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
 create mode 100644 0.16/generated/torchvision.ops.PSRoIAlign.html
 create mode 100644 0.16/generated/torchvision.ops.PSRoIPool.html
 create mode 100644 0.16/generated/torchvision.ops.Permute.html
 create mode 100644 0.16/generated/torchvision.ops.RoIAlign.html
 create mode 100644 0.16/generated/torchvision.ops.RoIPool.html
 create mode 100644 0.16/generated/torchvision.ops.SqueezeExcitation.html
 create mode 100644 0.16/generated/torchvision.ops.StochasticDepth.html
 create mode 100644 0.16/generated/torchvision.ops.batched_nms.html
 create mode 100644 0.16/generated/torchvision.ops.box_area.html
 create mode 100644 0.16/generated/torchvision.ops.box_convert.html
 create mode 100644 0.16/generated/torchvision.ops.box_iou.html
 create mode 100644 0.16/generated/torchvision.ops.clip_boxes_to_image.html
 create mode 100644 0.16/generated/torchvision.ops.complete_box_iou.html
 create mode 100644 0.16/generated/torchvision.ops.complete_box_iou_loss.html
 create mode 100644 0.16/generated/torchvision.ops.deform_conv2d.html
 create mode 100644 0.16/generated/torchvision.ops.distance_box_iou.html
 create mode 100644 0.16/generated/torchvision.ops.distance_box_iou_loss.html
 create mode 100644 0.16/generated/torchvision.ops.drop_block2d.html
 create mode 100644 0.16/generated/torchvision.ops.drop_block3d.html
 create mode 100644 0.16/generated/torchvision.ops.generalized_box_iou.html
 create mode 100644 0.16/generated/torchvision.ops.generalized_box_iou_loss.html
 create mode 100644 0.16/generated/torchvision.ops.masks_to_boxes.html
 create mode 100644 0.16/generated/torchvision.ops.nms.html
 create mode 100644 0.16/generated/torchvision.ops.ps_roi_align.html
 create mode 100644 0.16/generated/torchvision.ops.ps_roi_pool.html
 create mode 100644 0.16/generated/torchvision.ops.remove_small_boxes.html
 create mode 100644 0.16/generated/torchvision.ops.roi_align.html
 create mode 100644 0.16/generated/torchvision.ops.roi_pool.html
 create mode 100644 0.16/generated/torchvision.ops.sigmoid_focal_loss.html
 create mode 100644 0.16/generated/torchvision.ops.stochastic_depth.html
 create mode 100644 0.16/generated/torchvision.transforms.AugMix.html
 create mode 100644 0.16/generated/torchvision.transforms.AutoAugment.html
 create mode 100644 0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
 create mode 100644 0.16/generated/torchvision.transforms.CenterCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.ColorJitter.html
 create mode 100644 0.16/generated/torchvision.transforms.Compose.html
 create mode 100644 0.16/generated/torchvision.transforms.ConvertImageDtype.html
 create mode 100644 0.16/generated/torchvision.transforms.ElasticTransform.html
 create mode 100644 0.16/generated/torchvision.transforms.FiveCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.GaussianBlur.html
 create mode 100644 0.16/generated/torchvision.transforms.Grayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.Lambda.html
 create mode 100644 0.16/generated/torchvision.transforms.LinearTransformation.html
 create mode 100644 0.16/generated/torchvision.transforms.Normalize.html
 create mode 100644 0.16/generated/torchvision.transforms.PILToTensor.html
 create mode 100644 0.16/generated/torchvision.transforms.Pad.html
 create mode 100644 0.16/generated/torchvision.transforms.RandAugment.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomAffine.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomApply.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomAutocontrast.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomChoice.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomEqualize.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomErasing.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomGrayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomInvert.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomOrder.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomPerspective.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomPosterize.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomResizedCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomRotation.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomSolarize.html
 create mode 100644 0.16/generated/torchvision.transforms.RandomVerticalFlip.html
 create mode 100644 0.16/generated/torchvision.transforms.Resize.html
 create mode 100644 0.16/generated/torchvision.transforms.TenCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.ToPILImage.html
 create mode 100644 0.16/generated/torchvision.transforms.ToTensor.html
 create mode 100644 0.16/generated/torchvision.transforms.TrivialAugmentWide.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.adjust_brightness.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.adjust_contrast.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.adjust_gamma.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.adjust_hue.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.adjust_saturation.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.affine.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.autocontrast.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.center_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.crop.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.equalize.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.erase.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.five_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.gaussian_blur.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.get_dimensions.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.get_image_size.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.hflip.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.invert.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.normalize.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.pad.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.perspective.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.posterize.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.resize.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.resized_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.rotate.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.solarize.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.ten_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.to_grayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.to_pil_image.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.to_tensor.html
 create mode 100644 0.16/generated/torchvision.transforms.functional.vflip.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.AugMix.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.AutoAugment.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.CenterCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ColorJitter.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.Compose.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.CutMix.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ElasticTransform.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.FiveCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.GaussianBlur.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.Grayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.Lambda.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.LinearTransformation.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.MixUp.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.Normalize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.PILToTensor.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.Pad.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandAugment.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomAffine.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomApply.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomChoice.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomEqualize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomErasing.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomInvert.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomOrder.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomPerspective.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomPosterize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomResize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomRotation.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomSolarize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.Resize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ScaleJitter.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.TenCrop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ToDtype.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ToImage.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ToPILImage.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ToPureTensor.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.ToTensor.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.affine.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.center_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.crop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.elastic.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.equalize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.erase.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.five_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.invert.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.normalize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.pad.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.perspective.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.posterize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.resize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.rotate.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.solarize.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.to_image.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
 create mode 100644 0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.Image.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.Mask.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.TVTensor.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.Video.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.set_return_type.html
 create mode 100644 0.16/generated/torchvision.tv_tensors.wrap.html
 create mode 100644 0.16/generated/torchvision.utils.draw_bounding_boxes.html
 create mode 100644 0.16/generated/torchvision.utils.draw_keypoints.html
 create mode 100644 0.16/generated/torchvision.utils.draw_segmentation_masks.html
 create mode 100644 0.16/generated/torchvision.utils.flow_to_image.html
 create mode 100644 0.16/generated/torchvision.utils.make_grid.html
 create mode 100644 0.16/generated/torchvision.utils.save_image.html
 create mode 100644 0.16/generated/video_table.html
 create mode 100644 0.16/genindex.html
 create mode 100644 0.16/index.html
 create mode 100644 0.16/io.html
 create mode 100644 0.16/models.html
 create mode 100644 0.16/models/alexnet.html
 create mode 100644 0.16/models/convnext.html
 create mode 100644 0.16/models/deeplabv3.html
 create mode 100644 0.16/models/densenet.html
 create mode 100644 0.16/models/efficientnet.html
 create mode 100644 0.16/models/efficientnetv2.html
 create mode 100644 0.16/models/faster_rcnn.html
 create mode 100644 0.16/models/fcn.html
 create mode 100644 0.16/models/fcos.html
 create mode 100644 0.16/models/generated/torchvision.models.alexnet.html
 create mode 100644 0.16/models/generated/torchvision.models.convnext_base.html
 create mode 100644 0.16/models/generated/torchvision.models.convnext_large.html
 create mode 100644 0.16/models/generated/torchvision.models.convnext_small.html
 create mode 100644 0.16/models/generated/torchvision.models.convnext_tiny.html
 create mode 100644 0.16/models/generated/torchvision.models.densenet121.html
 create mode 100644 0.16/models/generated/torchvision.models.densenet161.html
 create mode 100644 0.16/models/generated/torchvision.models.densenet169.html
 create mode 100644 0.16/models/generated/torchvision.models.densenet201.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
 create mode 100644 0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b0.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b1.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b2.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b3.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b4.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b5.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b6.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_b7.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_v2_l.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_v2_m.html
 create mode 100644 0.16/models/generated/torchvision.models.efficientnet_v2_s.html
 create mode 100644 0.16/models/generated/torchvision.models.googlenet.html
 create mode 100644 0.16/models/generated/torchvision.models.inception_v3.html
 create mode 100644 0.16/models/generated/torchvision.models.maxvit_t.html
 create mode 100644 0.16/models/generated/torchvision.models.mnasnet0_5.html
 create mode 100644 0.16/models/generated/torchvision.models.mnasnet0_75.html
 create mode 100644 0.16/models/generated/torchvision.models.mnasnet1_0.html
 create mode 100644 0.16/models/generated/torchvision.models.mnasnet1_3.html
 create mode 100644 0.16/models/generated/torchvision.models.mobilenet_v2.html
 create mode 100644 0.16/models/generated/torchvision.models.mobilenet_v3_large.html
 create mode 100644 0.16/models/generated/torchvision.models.mobilenet_v3_small.html
 create mode 100644 0.16/models/generated/torchvision.models.optical_flow.raft_large.html
 create mode 100644 0.16/models/generated/torchvision.models.optical_flow.raft_small.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.googlenet.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.inception_v3.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.resnet18.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.resnet50.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
 create mode 100644 0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_x_16gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_x_32gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_x_400mf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_x_800mf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_x_8gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_128gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_16gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_32gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_400mf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_800mf.html
 create mode 100644 0.16/models/generated/torchvision.models.regnet_y_8gf.html
 create mode 100644 0.16/models/generated/torchvision.models.resnet101.html
 create mode 100644 0.16/models/generated/torchvision.models.resnet152.html
 create mode 100644 0.16/models/generated/torchvision.models.resnet18.html
 create mode 100644 0.16/models/generated/torchvision.models.resnet34.html
 create mode 100644 0.16/models/generated/torchvision.models.resnet50.html
 create mode 100644 0.16/models/generated/torchvision.models.resnext101_32x8d.html
 create mode 100644 0.16/models/generated/torchvision.models.resnext101_64x4d.html
 create mode 100644 0.16/models/generated/torchvision.models.resnext50_32x4d.html
 create mode 100644 0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
 create mode 100644 0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
 create mode 100644 0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
 create mode 100644 0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
 create mode 100644 0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
 create mode 100644 0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
 create mode 100644 0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
 create mode 100644 0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
 create mode 100644 0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
 create mode 100644 0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
 create mode 100644 0.16/models/generated/torchvision.models.squeezenet1_0.html
 create mode 100644 0.16/models/generated/torchvision.models.squeezenet1_1.html
 create mode 100644 0.16/models/generated/torchvision.models.swin_b.html
 create mode 100644 0.16/models/generated/torchvision.models.swin_s.html
 create mode 100644 0.16/models/generated/torchvision.models.swin_t.html
 create mode 100644 0.16/models/generated/torchvision.models.swin_v2_b.html
 create mode 100644 0.16/models/generated/torchvision.models.swin_v2_s.html
 create mode 100644 0.16/models/generated/torchvision.models.swin_v2_t.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg11.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg11_bn.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg13.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg13_bn.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg16.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg16_bn.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg19.html
 create mode 100644 0.16/models/generated/torchvision.models.vgg19_bn.html
 create mode 100644 0.16/models/generated/torchvision.models.video.mc3_18.html
 create mode 100644 0.16/models/generated/torchvision.models.video.mvit_v1_b.html
 create mode 100644 0.16/models/generated/torchvision.models.video.mvit_v2_s.html
 create mode 100644 0.16/models/generated/torchvision.models.video.r2plus1d_18.html
 create mode 100644 0.16/models/generated/torchvision.models.video.r3d_18.html
 create mode 100644 0.16/models/generated/torchvision.models.video.s3d.html
 create mode 100644 0.16/models/generated/torchvision.models.video.swin3d_b.html
 create mode 100644 0.16/models/generated/torchvision.models.video.swin3d_s.html
 create mode 100644 0.16/models/generated/torchvision.models.video.swin3d_t.html
 create mode 100644 0.16/models/generated/torchvision.models.vit_b_16.html
 create mode 100644 0.16/models/generated/torchvision.models.vit_b_32.html
 create mode 100644 0.16/models/generated/torchvision.models.vit_h_14.html
 create mode 100644 0.16/models/generated/torchvision.models.vit_l_16.html
 create mode 100644 0.16/models/generated/torchvision.models.vit_l_32.html
 create mode 100644 0.16/models/generated/torchvision.models.wide_resnet101_2.html
 create mode 100644 0.16/models/generated/torchvision.models.wide_resnet50_2.html
 create mode 100644 0.16/models/googlenet.html
 create mode 100644 0.16/models/googlenet_quant.html
 create mode 100644 0.16/models/inception.html
 create mode 100644 0.16/models/inception_quant.html
 create mode 100644 0.16/models/keypoint_rcnn.html
 create mode 100644 0.16/models/lraspp.html
 create mode 100644 0.16/models/mask_rcnn.html
 create mode 100644 0.16/models/maxvit.html
 create mode 100644 0.16/models/mnasnet.html
 create mode 100644 0.16/models/mobilenetv2.html
 create mode 100644 0.16/models/mobilenetv2_quant.html
 create mode 100644 0.16/models/mobilenetv3.html
 create mode 100644 0.16/models/mobilenetv3_quant.html
 create mode 100644 0.16/models/raft.html
 create mode 100644 0.16/models/regnet.html
 create mode 100644 0.16/models/resnet.html
 create mode 100644 0.16/models/resnet_quant.html
 create mode 100644 0.16/models/resnext.html
 create mode 100644 0.16/models/resnext_quant.html
 create mode 100644 0.16/models/retinanet.html
 create mode 100644 0.16/models/shufflenetv2.html
 create mode 100644 0.16/models/shufflenetv2_quant.html
 create mode 100644 0.16/models/squeezenet.html
 create mode 100644 0.16/models/ssd.html
 create mode 100644 0.16/models/ssdlite.html
 create mode 100644 0.16/models/swin_transformer.html
 create mode 100644 0.16/models/vgg.html
 create mode 100644 0.16/models/video_mvit.html
 create mode 100644 0.16/models/video_resnet.html
 create mode 100644 0.16/models/video_s3d.html
 create mode 100644 0.16/models/video_swin_transformer.html
 create mode 100644 0.16/models/vision_transformer.html
 create mode 100644 0.16/models/wide_resnet.html
 create mode 100644 0.16/objects.inv
 create mode 100644 0.16/ops.html
 create mode 100644 0.16/py-modindex.html
 create mode 100644 0.16/search.html
 create mode 100644 0.16/searchindex.js
 create mode 100644 0.16/training_references.html
 create mode 100644 0.16/transforms.html
 create mode 100644 0.16/tv_tensors.html
 create mode 100644 0.16/utils.html

diff --git a/0.16/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip b/0.16/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip
new file mode 100644
index 0000000000000000000000000000000000000000..95d7e2eed994907da16be950030c9e9f182ddb61
GIT binary patch
literal 102610
zcmeFa&2MDqzV8)@D<OG=fC17KVr_?m%f0P#+nwIY_Ko+!>7<iBXQ$KZc&E?Crzw|R
z%XX#9RduV%?sh`91PmB3ff3A@AR#eCNKD{wK#cC}&h8lg1U}#2?|EBQWw$$>eaV1U
z(rvHxtmpOj{`Nfk`+wt~{JS6M&*6Xe`HM&Y{r~k}|Ec=m1O5L`c{ZAz*OTeicrcoE
zM&nukq&w)G4o0ur<LlMc)s?-U^JnE7@2GlAN4wRtN!{z8%=*K#>SA<NU(~}{^}0Vh
zuNb;|zW?}u$Gv*+GsjwKx2xf3R`2evRQx%ZTvwxEb#yWs47$hFr{}ZTc)Gi_#qZtY
z_N1QH-O0&$`)o8i8`SNS(Z$voQ@K3e8eh*wyxF?yPy3_c*70C;ymfZIIqsg-)2)kc
zf4J2-tB3WZJF9!0{`h)$+|jVdqtUBn+h;BQaMZC$jV9IZq(0p}IvZ6p{$t(suvdLL
z9iPAHoDC+O-stskFzWU?-OJgiQ@`n6j0g3!<A>8(b~t}HQcv+a>+<w+FsQ<})d^eL
zUP&jRBdz*VYd7moUsa>iYP{rJj_cX$x*k@u*Q2U`p|x*Rm($DcU~sKxD<`Ajv_83<
z^{?vcbke=3r|PL&UG;nQs9nW_x8dwm8M6DTF8Zsw)2crlU(PnF?y$EKCP5eaX4e;D
zlN|yyY_`w2*Po6D-4h2MMy#)D4tfeq4eDy#e^U?WR3ERL_9xSs&)r6-F1lkzoN2rI
zvnd-h0Sd+=x(6W3Pb*f&WB;~oz&l;(PU@<)vr&DuQN38-?Dc_O!AEyc&G>0FVSiqO
zL9CEN^|8Be*yubA3$P=PORam=`DoJrVKki43wuzp7DK$MT;-J0J5Zkh^60hZ8jrR=
z7+g+g0>mjjy{3EM*1@(q=?<&onj>QrhK;+cy`8RHYQL({<%A^yUqZz0jK$~(HT{f>
z?dnAhjsj2J!F0sN)L|snHPm!%t3&S$W&LZw=U;vn;D(7$01LrR{|6hUy&{UT!q_7^
z7+zkCuLTUlaeQP8iN9QQXM+1d{}@uFpVW3dn8o+Ss@rVFaLRZW)As4*5K?0YMEl1p
zD=U1{o}4`E0<F`-)oJ&tKJA~ikB>)hR$o?o)#|K&c0OBO`LF^T`$O;%U<bH5P=$-*
zVYff%FxBIFLfaj>>(yWm(`ofD&N%#g_xcx8PNDPab$4<$U3V(fKNU>^;rhertUElZ
zt@YvkmmAffKb?hEqE{bQUkg9F)lRQEoAi5v>|w?C9LS_zn^xa**41`38ND`AqHi=Z
zvpLkiFIVg_Q|!=0j|cV8Mq5uWpnc^Frs3~8AjliN(H2cFkF~66Yp8zqVuu^mP@miH
z`eCD*{`1Sa{-NG`3~1Eek&c?LI{n_84gDZotm(n1rBBB9MYC69dh*6Bq1l+Pi``NC
z?i*&vJP*?d;p>-aL)1(??NyK4(9bbvX2eTF^PshQGQM10_qnRA{&e58{p4E1al1S1
zPA1)JwO?O6)5bV_H?3!_H(=~5SW3;_9CEDJbHCgB;+1|}_p{P&8_;zI-Rsfiti|&W
ztH1dh{(4?_jVyg9ayqU4WC(kZ@tY=_Z}~%$h`UaXCiEF8G=xDUU{_eOd-RtQhx#M9
zBOJ&m`B9s0_6M_iG6Z$6YN-EcSQ~YLv>34%)~~DGlR<YnUF^uFueH>ZqiX-zld9D}
zrNd%4@uK6){$K_d8C_IUh|)>Du2DtvY!21yK!P-)kyyStoj{5pS9)e&9*t{=|5Tiu
zj}yCp{Tc}-y@9E-pIhC{i|%j}K61?_Y<4%#;9r}g)6HWTVm&(s?>8TA|9)rt;m2EW
z?hX1o5sQRrINv^c2JskFpLd7-!S&^<8%O(($7;9P3+?C&TAt`ueK9`m4`5VU__#Ye
zhs22>izq(hSATIi8Hn^vYOu}tnZ4)fq@GRs5D9v;Rdsj<;IBqOCyE46P22DGXmYkS
zL#&weyMry?6E>@3bL_-9cJdx{&U@{P@kgtnd(DvkHM7<%%)Nc_s;8e@>+984wVJ>6
z4?DCkX4*>g-0+iS?Adt+`nN-4#iB1KLl*@cI9&LEfL-?bqbQ0<pa9H(po@Ce?R95e
zn8tK;(wESdWlrm%T|I&E!3$5i(;6lYSPcR5VJ*%XVUXT@ARv>UzY?R5U}=Q;Ld1I~
z@>%!}U>!s#d%8KT2la_YA9oQ<1iE47bT<V47&6FFt?m2wH>&OX+v_G2w6-7W*N5Bd
z?dpr}$$4PXz`gf0Ct!5xPq>XZWOA3wSiqfuS%G<Nwktqe#&=ptmfNUc2|B&}riG*t
z4$w|*gFTxey>-NHyR*I3gGXPzSmiJwo!dasjG68^rng~zC$CzE-uEHPgHFa@9^QWm
zJ>;XEw|ul+ezdKR;ONeN!+05zL_4w&Q*}>H>M_`xg%Xo%oGhNzCd9<u*EX?*cYsAn
zY)Fv(AB+xONa{fnymS$;KS78KGUAl}BtG`~63&h}nd9r~lXmqKwE+<8q6MJ`ycrMr
zC;b_4Aewe`)E&-1)P8sB4EE?~1Bg25@{;am=g>g@GqLY{#0Npher+P?1U04ZwJY{v
zsLs`+U^&Z25Zz&K066xCvypWufR+R#sk|CZAlX_GqTP5hI;p3Qb0)AGSh8|d#M(sZ
ztL^~eq~;7lXChX8C?=~o+}s9uUmhK4pn5{fWuWMjZ<l&!gw5yO(^=<_h*Wjwh0kwg
zB~Em&T7297{ENr?fBNIWO0m|xxU@n*6ti_1F0D{pKY-GP#E1ZqQItzBf$PW8AUXNQ
zCwj1V_;3ew{r&F*<8w#oa!2lXuAeX*g4Lb+aZ*?uX=5&V6s>dl!Wcgv@6BJ2Cmdbt
zbai^(9oKyGWA3eO&wkpddb8_=clPWjkpw`(q?#|LGsH1705lXt7e<7zw0w(L!)qG7
zC`1C$ZbSAtW#$%4UIK6Ji2!XNKm*fWpZB3(u5vl#PoawpETchx1|=7sqgV*5BoXyM
zhe`j~5Xv4N!Dr9<(htXM+a${f-N~E&l`-S)@pNnF{{08-?VbA%es9u&t;|#{uukmB
z`dUVeZd}6j0{P`Am{LA8;R;-iOM+%t17;-pb=sYRmBQS~<q*j0vWQd25VB#nI!0NB
zPeDB|C*70lcSPL0pF001rxhZu1V<EuOKIrLBr39Kp55xpXMY0gWD!E!M0M>}()E0T
zZ7}t!e<J8xJ-O_4S4}aJINa8+?f$gWy+T9lLR?$x6>?xrLuCDx727Ai%hPEE^09X_
z@z?+{MrWXSAinET9()SxsYmO%T|?4DOQg<qMyF;olPYCz3#2M~D2Yq)cpbxlV_HNe
z{o87Eye@o-%~0^pOeXtLs@_~XrV!Lsju42&Okh$8zX@h-dcw_sng>me>AZbvN~#Gf
z^D4gEI%>%ySPwA4aGVl7T2D8sXeFs!3Sh+fuZ`)=3og|e3dXQnO_P-_?#3LSE~Xe%
z#*2|yDPmn8WfsWMYQiVjqEJ(WndK@u%t4K8M{n5P9NP8K&%#g>{UjyQn|Qj(ipJWI
z&NKF(I_A4!jmK^mX)!FO4nEWP7bE0(N3NN=gN#g3Ix({}B+Dr0+c&6LSQ8Mr<HqRJ
zT_)Y17DT9fjDZyfWCPpT1T&QjzgaZNHm$0~=BDFgJ}=fyMYltA^Pa6@J=nR5v!6Rj
zEv{+VqPL~CEuE9~<K{OPQTe*C-36(lX_KSBEIetrc>vQM13IhL?`m~9>|<t3yDyW>
zP+}_<sal%C>-MBQYugSNTN8lD&7bXTwo5qPKsh9xmkBi!KTH$zkJ^bTMkG#Lb-tyu
zR?s+|2`5ru{{`!V8P9*l<JLMH`NGU4i?6$HcwI8(?}BMgG(UCb0L#wY!i{-*kS!9T
zm||GY5|?o-jK%2Zf}sw!8QayPN)P)pY|0WsqfIe_Eutx0DvRx?YG8~RA}`0%gHf`i
zXcW*X%Yi}EY7Yt2EPyZzrxF@Eu@|htV-Qm#1i~Zj0E^OSVj3a9LRWtFup>T&j>x_|
zbrhQRFW@(3!NT}!NR5h#sjDaHUNGsLhXG(__&eBNbvQ)4^Dol`rzWU}_Rvvw5{Ust
zK<rF24960B4W5&cSyC8B_6lR>h6ornYF1$A!C>RAOv;y<sb2T`r%*sSUh+|=c`&^I
z^44YCwQNVPT4y7H2~$#MgZgylI%;TI+Rf=Kw7wi;&skJI4IHAB-+&4l3rG+%#Yh{k
zT4~yjPsYFwm`gaQa||x*H*6%FFQj=?1tkPo2!R4BC%w_6P!U>$bgjlH>Y+64mxZ{@
z_=m>CtKU}W<6Efs>Oc~(jxtNam|x9mOM97PQF`JUUfLJdNq)GluYc%|3jra&oUTJi
zq}r;-%%VRV4KJs=)gyFroBWqg9xr3fZ2T=qTfU;@6UU4wh#A`u`>}mA-X@#2H%eRY
z#6hrvX(L!O=*u<wcC}w&Rb*d<)D>6;APZWe#Q#`WXt-e-ENxsKj0huPa7cFmDH!aA
zBB@e68R4+sfUhz+M(Yo@Gr_+`n44g``b=mWC6no}560`PQMlzEk>W3z6zjyKFbnlT
zN7M^UWAcm0%3+<$kG7zJ$9)57k85EDsD-c%4KR`R0`Uk%+z5vH2|}*cB0+a_Ie-9R
zIAXQ4(Fl3haZCHrhpXG3vcBa;o7WTMfX?^hT4cq5as%_KfgXSF7=&fpyI?KacB1KJ
z_vJde?T2OscZRx~=LONq%0bxK1!@ZvBlTcb1{V4_(l`{{Jc)Uz=R6~DiGCn%vL=)>
z)k5{B!A@zNMd(6GcHy~rupYinr*D*j&l}x|JLB&NZ6j{d>NfIXV!!(g!iUM`HIRJ-
zZDeb9G2U90`{qw$?2MA^0ER#y(I3@}%>m9}73U*#n9x%`zE+$~VN$d2ep{WcHdod8
zvHw&1`|(*JxPOCF$pBQ(Z8eb=N(PdbtA|s<Rj<Tg1HBV|21J6WNqmQ<1lPdn=>?$W
zysmDawE+~~ho_qyx17Q|)usWHe2~?H`_*PYBcUBb=Wlv8ug$?|G_E$cZG!DHya3Nq
zG{Ond2&cpTfBEgd|G@|P|38|GMgV3mC*zTfnw{<t+08YXcL_{bd7k@#GfZE}$i?c9
zM_9pxw>Sao>D>YkR(?H!1~=`@FCKGn(?H|vJBA+|K%Q@RV2glD><MaWQs-Cn!h<GC
zIyl0Y|0#blK4jt&IY}h2nCHs0KFf~n;><_K-y;`7Y%yO3hcw(JftB@>#U#Z@OjlaM
zfNW|L#CVSwR!!cE?p@<k<L=NbndmHe3M+Ub&PTnafG7K`bra@>cI%?6+!y)?0m@wF
zxA>-mduFxW-r~RDyWT=1t!LA%VqZ!W%C7D9czBjw@}nst(EB;(V+ZA#WsB?*@BIF^
z-+poMZRhhR&%bzd@Xhl-i@357UjS3Pqxarzx)NH%VQzQBFZkf~1>**Px>(X_A4m2w
z-?&`hd&$Foqcgm3>`KC3X2r$4Gz7mgCV#hO)HpfEn54)0<F6K7*|~+>;mEB0mpAyZ
zm$iHkuQeRN47pz$Rs=zlhY}-=st)?N!QPGxQcp#M#iAr=nWRZTg}fFn>*-B+;QEyG
zBD_d29r<IiymqUL%fYPQ2*okyedXBNt)5<9Jd?QvhdvGsK6rL*3TrThV91EUXPGpa
z*6qB`&!D9M2**p@V&~?9!2Z+k$#W@|_`+P$MJoxMB3jw}7RZalLL)xrVo?)gqQR}#
zYW5wPvOO_O$;1jI&J%n@j4P|NcjtDoAd3PFLELa(jSYc3?{g}w8#_B6Rc|W&EqPQ8
zP9>x)&Q%k&ZPB2i?{IvHY^DKj+sThG4et0~^6=B$el?U;*A-r@0Cev?+ZH0HGxoI!
z>xiS+@Nl;$8qaQugNGO=Ff0;So=!&t^gv{rE-rHK*SCr>8-(AW)+B9%>qj2cB>HY3
zt`R<qGQwvjBivcLikcQfJ_h>I@n}1jnMYt|ZCkvoBPNd=Wri5{z!-%)>|&@>+_6P@
zhTRLQ&atFU`<7#NxL{=kn^D{)o`FluxwLp}UV-P2o<23$cwVD#cAwTVG$zlBMe_x%
zTL9$emMnNj&+6f0O!GscB-mtBtwu=330l_WxE>DMb?<WPFW*nk`-CL)I_I;CfnqK=
zl0}QboWt4%H{X`U5kN#o)3(@P`+KYut$BvBQLWm9p4Gkk_wPUa$XFiMd<fCc`o5zt
zLnltC%~@`<67LDY4Qj>Iig#&5RC_km0M7yn1Q~kibP=A70|FHkK<7-LUQfTQo=EBz
z`N8gw+TezOG0U9+>vF>Ch`x|jJ_I!4WFc;==Y3?c#_U|sGwbMnP+pIruV)j1yL!Qb
zB-}I@G7`Yn;*c)f+NSksLIx$aP`rEiWZ>53$<U*=xk-t+@<-xx%_nLUjT{bN(Ib!5
zmY^3XAT~MX$|f)mJUgy9gZY}4<8{^Vv2=bWnoKO*#E!Oi&`Ki&vohMZQ^UMqGrLb}
zg(aKnoqGyD9d7Sp;*tD?xzYx<X>G>cwbh$W(M9YtzZnt1_#I@4;z+}eru}3F(IDZ=
zc|DrE(h^_SSTj8AF93tN^zE?285@I5TK9~;Z3vcav)BU>d?OBbH#o*20HeQc{swJN
zv5GW-4QF4Ntw*zJx(Dp8?^xm*5@bxlRjf?)z|Or_BRdDPZ)SN>2Ebs&NHB9T^as_r
z9^s7fZX){&Ls^Wo#RsS191)x^+E~VRfm-v<sz(A107+<>mQ#EOBheE>My<stu+SpT
z%e(<|i=K+NIL^frD9opZP;@0*@DMpM3|h+gzW>d!3sjgcf&6I%{ST`@0ydXZ^l=ZB
z6;}BIZ0Rxx0-9w&vDnnY@CIqvugr}K$+vx2#&6sl-<xz_cgjwTtW7<roOkwl`96a4
zB6)5YbZ_Vv@tjVFJ_rDN_;N*hTfiRg+apAac5RvJhrQMfeTvw+KnaZ)jL*A!_uHRr
z1hTQWdJI`q%;$<J1x2@c=MP4-PT|Z6Djchp7zzYYciG})EIo0^QZte^>uM}L<Krm9
z+)NY@#vC2B-r(KAnqz;z!Na4UyKmNsu3;CUy-;57Kl^0}s)6OZHfIY;;>F5}ckN_j
znw}fI1TLLmTbnt+wP$(z_MMu#Co7Nx!>@$)K9Jcx@3;;lKR|DYV^2t7LGss87Bu$t
z7T}IHo+wiVj}sdmJXVT<T4G^vaQ@W(`2%sVi#xjr@XfIzu*YpS_-!_E*tG^3VzD1%
z;48p&ZoeVS?2PxS{TV{e-HC&E90g*WPz&;%sCNMwY=Oxg@S+~9&xL_*kp8OYo09*`
z28+y02rpjjn`7gw-Bc>9TQ1seJf55khaE`ju%11<-#Hx*kCyD?{I?NYZ|Gv9kWV|$
z>nVaMeLsHobXkB{7G;+<P0?b$PRzk3`Yud6#f*g5k%a(}CE|@mpb0M}kc97=0cb8Z
z$}D2d1Ptpdcq1eL;oAKv{Cv1-XKh&u_EUQ9ap(5Jf8;Y4eH2X4a!hvQy}!OeaDlm>
z_q+*&?ROuo6Im0EtP_Ji3d=hpR0#8L(Lr_SlT+*^z%#EheOs8jlGqQM{5*~jAM$O~
zj^5>8SGQef2WA5z%Vj}e<VYz<AsI+XE#l9Ss7oZvmbgmsz>?8u(Jhc9JT~H<2Ei8=
zQB2JA=AJl3(5vjvKuW!R6C&n=;};~26HAuDVoK>e;yl$w#8a?6`kXuhhxtwFO*fD`
zua8|tM#I;IGM0|wAVh*^y=+;W_;d>>sV8`~6*|3DP>)UwC2vb-bCu1q0ICuU4o2zl
zKlfd4Iq(_nY-_IV@FIj+Dy^o|=@YQg>9h!%K2;*gNYNWn;f+Q;*=N+Ct@i-F{z;}Y
zdyzj8UZn3+BN+pf6CsJ)^ccnCqiK-NWgb>UNO|9i_9?(O+9o7~Vn;s1aFv?b*x<78
zUQVc%`m1{_)KmQmsn(BKQ}y+&qTiW%H$_@0a#XB07e(H!ySn$aG!O61Cg*z#bx{OS
z+Fp`%icFiZC$=BR-%+zqi(_3&)#Y>&=-GKC?4?dfn{miE5DvGhRVM1ay`#8>(2s3H
z=qU9>=Q@4mtk|pz<+~iL*+xAD4SlYo$E1*$Ea<&;jAS~WE~%#}WX<I;!t#$bSd;dK
zuhY;bPyTGZH!d@4tpUkQfsXobY=sxkLptuD2gh%6&r!8BeK@{;AH6vCVQ0%0>fh=X
z9X7Ui`R69#wb)s(7$k*Zn3WmESQ(v4EpFqS3VBwx-g4HNx+r;eoS#2i<lXb{7kIcQ
zhp8E607IWly0i?`ZPVv-CY`mF5|rjK_nCvQzlC?lDTEMm5Ga0HxMh2|W&fxD=YRO0
z|Brv`gAero`kY%<@y=t&TE{I0ojl2#J+mtZ5<mS*tc@1Kwan}KEAp}4vdmxH&3emN
z@8D?#M52t)FEWIWj3)RHE6EsJGX37@WV%H(3{huW-xGim7<)V@nfqi^v3``vG8(P~
z$82J))u$N@aLp10b70xg&R?-j1UDw4qsb>?g7ax_(RGMEE2GC^;%{tV6KVX?9wLH)
zA;X+ySLi!=YHD<~y{OSGw=loO^|)p^oNFsK9ZdK`Oj4LsVEEgnBYWuk8Z?MFY1wR$
zJ3ZtHA%Y#kkO&RPK^ie}&o<lrpT>vTdt|xudVrT)36Mkv#aurxU7*~@`SJxb@+7@#
zL39bn4Vzw1SNPxLkm|RXV&rbkTeOO0#)b&v1rvf!WF?r^pY0!f1;KvC8MfAvuXe3a
z4!%~1bM7-iQUQwYykh1a;Zd~4&S%aY79I0TeT(rLe5-UwOOQG;lb%~ci<XU));xpu
z)lU17ja47*qMwh5<7>zc%5p20w^gmFk%Cy(5X$&@=Pn=Y#1EQjNK5A5GHQIa-F}X&
zHM)32>}Lxv+4@HF`B$k@z~cd-Tw`op4`{<)XQGj@FlvFqVt?Ptd%!mqU3jR4Ay_3k
zp1yDCjaVM5y|Ka0dZ`9de1p9AA>d8@4Y~I8XJ1k1DV)%KS7{K3U=frfCrwUL3Jv+_
zDD<QJ$SK}m4B7p5O&|sztzK=f7Qo^mv{yUDqclM1<)&eP2@Ggy46{^g1gC8T5nCi6
z_K(odr&rsZ(WG;=1Az-sj%H`gD?<_VfG?}P5qIN&KARIPLz$hd`e6`fNVK5)aZx`9
zgXRj_#j(ov2Mxj1xkCv86Kw9?%bQ?SWexoXuQ6l^)l>CGF2|umv!BVR+=@`J4qcR0
zy^lKy1Z!8g0bnQ>xX^KCn9FsFx^u2P#pF&3y<z9UBcBsrwTZD`qhK|@B6&isaEMZ5
zd$@O?hRwvS&|B=CXhHOQuSUY$q>+Hc3C5hdO8nRAJZpDR$grp$&8Kky%J8>L4gqUA
z=rCua)O?W(Nle+<LaWwCx4={p<I(^xSskuM6nG$26cg*ze0HqagH<7UGUv(omn&^t
z94m^{Rn&yXF;I&AAQVClTrCnL6r7IX8|Ej$5aQW&VWtJ_Fe~OVIwq4w02Gpj(}B??
zVO74-xF*(QBvgO~O^=SAd1lLKLMPr3q?AsakwCj0T+J>5=;U}fWO|AU?sKk~Gfd^h
zgcp_oEL-TS?{mStZH38_?Iy)-%=~Tu-2NmR|5n@#k*3&!UwmzgLCrO)x?p1{^k1aX
zdeT~NCmTf8@rtxlvpVKMlNnJkp+Bw-o85E!6Jr=Mfo^~DRy$(MC-ZJXN5!r#wIt>|
z(}T$+r8v?e13&~j${T!f+R?VlWb4)D9~uv`LSb|vFso7LY|`zuiY?OIREuL?$#|2u
zN=>diIMaHuT|BYYT&tjic;hy<PaGO8Jb8N(o&c5Tuq0Tk1i}QSf~w(YxS1YA51+-x
zHzAd_0r<#QGug{Rh|F48im6f3RnDhn<0`7`XXYPj4=8$tlM6{H{tSE-359X$wqLQW
z?|wn)Q%8+&S3JiqGW?Kf@0?xi`(?8=r@%Z?S1tQ`eTB!JI13$E6bTj@#shibW>M^|
z{ZobayE!a&jAoFwFmz>fyKzndp)sJ*sj&)n8w3r4=J!+7ixmSz!a|3w1u2tcj5Q;P
zAXdjv9;;!me$&>MYviSnz1YQBJMGA-=dhhreYI)J_}41h0B|5o)hCW#3zv-_^)_zC
zl7M=)4eMmFcZ#PM=u2v+iif_JDWlPkHE`eB>FP6w2yK{#{(N*sdbM&*e%h^m^r;<6
zBvlV>S?#E_@zT<_0%m;MkMZBOpRy0c2_r0dqC<|I>QQTnDKQ!qrU*9xcd#ByFiZQc
zavoNe#P>e_+J4IE^bUAAnsg`EkJ(ZZZQ0{fUvV5X!*E-%=dIJW1gl7rZ1}XJxh*i4
z;oXiUey80mceoRlu!roRwJhl@0(Cfq*3!=<^k<k~ZzDY@i`RBFho;xqK>qxH{^$Se
zzuf%b1O5N+&DlUK38m8=_vcL>E8nTmj7r{Q^Tw~psFCOWi`y`A*Y9t#nBvrlgoLe>
zn^}Y=O9i|nolTL86R%POKF&tiQ8D?sJ5z)uSVJsZ*a~J4Hb}{#UUaYoGdFp~rG4VV
zYFmkbvk9Vu<q_`sSCR|M6eJ9yiYFH?X&^ioQt=r!L^UGk09Nnf=eo5meLXhwEs`P(
z*oF={97uBn;;=ys9%hA0&lPMvo2o_)MjV(dd5XgsUA?Ylmv9MDmQfSCULoZX`!)UO
z{{6T`@<AQoP(#cn(?iL$5>%~3j^YSSF0?gc3$U%Q@i0DgFVxbL>_!Sf#E~9I*CW*}
z8HTyTxMui97}eFs&FaOUemLuW{AhdpVqevS_)32^Ly0I_+3)@=&Ulk4r-lZJ&30dH
z?))zH<X?XM?29iqKdSb>dj!!Z4~)7y>s70Jb%OX&j&Mrx$ZF^J?caZ#`nQTh&HmL}
zwZH!;R3mCJ?{2dnL<)%;N|@QLKKkVEtFL}Qm5EHxK)dlvu(K_Q!@<i(6s59ir9r7|
zR5Tr_7L?ngf>MpU(+hf0Was3Sf@xMAQROzNC!SCAc_=;^xv?f{tuT9pvXu1e28&c?
z>SoiERnD}6nS#ogieMd1x8_dwZ;1-pn<2cey4_>79e<c>xQfZcXD-trd}Fme-xtXO
zkvSRwdK8%hsl|Fo*YOlwy#43mz3?&HdO%rXo@W4hXrDHN6hqi5<+c>}Wy%r+M^%DA
zrlTWX0OlPPYuF7x_j~9??x1bF=P3-D@vSdIl=R-dQPB&?2ToVRIFJ@+6s@5^hzGHb
zk+eG}L+NXcPVaJY!1qbNv=BK9(<@L>gX8xlW`25tX+}Z?t1x<^?_9!;Zm3i`OOlB~
z!;4@zI|76?U$2hAo`hOqk(~$NqoUk*Vu#NS)`}uh*p_VX!jZNaxf4fC+&EX;k|SOf
zD2)r)cpi36iBj)p3VDbLZKB7t_p)IAT5ElDNjUvb){B&36h5xBg*x7o!6kK94%cji
zwT)`c`dE9Z&n>jpYrTwhB03$&mgyQQm8va*P=sw!k0Bc&^ycZ(T@@ryXlrQKms--w
zQH=@#n7~O8)d=J9l^Qg1=vEu=NeIKl{KRRyS>Bs5A?fq5zwi^b8>LAIm*XY{JHBZ*
znxYaapv>@5?XZ9>RC^Lf0D7*Cerx#{TRbV)DZ8x?m=ZinUKXVFu$L+6(c#R(4dBnl
zvBGL?c=2)Vj*FmP)Gg$^n#SiA2T`Y#-s@?r$pVuYhp?|*eWrM)(=KcEc?7-EqN2f5
zCVzA^tl!K6_^Ic5BI}tUq*S!HW08O7REE=jJy3kHEO`g4G4$<^WiNya@M@DGM3;|_
zcFVrHLAG^2S!hQ&vV~?I<4cJo$R%t-)Ww2r3%s*Epr}o(YPaXRFpoI~3o1|9p>&Vn
zz2ke|S~XY}xevRk5KLvS!lV_-kn;N_yxVz(cJiT80%Hbb#7|UF%{C9^L~-h7kgD?Z
zp+a5|0P~{zyR{-K&fv=sU9q1;ux!e%OZbp6>eGHZT+_y`h4=wYW9~JN^frWHtv);W
z))XoSywjpe@hWN^pC9f%q`9kMs|Vzi4vKsopTKTKTcvpUuUP*vTTMo9I>S@;ARfVC
zBo9oHGD#}3v!+br>fW*@)|HsQwTa{JTfcxwa~fK*#r7*I+;Ed3aR_`iu{4*o3XxZn
z*UV&LjeO?JCdPm<tHz3eMnkvHus6S!a1AkGEJN}rv>6hfpwG%qsN>~WEt<*!G{^{l
zYq&uVq9JIUJrR$Pvg7ilr~|fiY?%%z0E4&+EUd0yfr!!2>Iq<KeTvl@Af!`+nqUb>
z%oQRBEcDpZ8$i%7DB>&B)b>Nr768*x`b(=jWaDT;O=~p6n5hye0bl7k|Dc9*f=WqX
z$>CrIsza?z2)!hiACZh{B4%LcGy{pL14dXy@vK@T+PT#-@^1SlKAFcLMNn9bv@?JX
zg2~gN9R+m67Xn?Chb~mz@PFXRQGQHQae-+mIWVmI+2nxrzYsLkuh>(jr|3A?4yJAY
zCv3lSCaqh3l0N<e<&})A0K@kExE3TDlH(IiFf`RsrDag%OJ6zxXEpV!)(%c81oj)r
zMkWY%Me&Fb-v9@O3UixpJHEqGnEyirP+zGa;m$ju#z40hW>onbFCdv0YBeE{24LA?
zmuSvMmI??)<SYxvZ_fN?f*()Vn<nB7Slxl%`^iNn6Smx~WC@k>`6wmKrj1eU;<(qX
zFgU+CTr<*KeY#gY3}s($KCevXBWb+rxg~PQ3R@3h#T3A}x0E?X!S3bZ8cJP*4s(Q}
zUQT7@q=4GrKx{);gHoB6QCj621k$>LY$9ue3+z$<gI2pjmdGCMZ&m!?<0V~(rCH2u
zfHydpC8T@N7a+}zA|kv-XZ-LZRdzb;GLZ}#YGx9o$~yD0ey=HbZ-J9XN9rMzkj|7H
z-QP-X2Kq)|cx)7;GPhbtEkKUmBe5Axb<Z=Ha@##QIb?U`V-siWK>#X{Rvu2|5`E@`
zL7KiETCtDrthL%7o<j5VI5hv%N6gI$=7m&!r<A%cCX>;`m0o?JKdb7^IpMh#ae?1i
z<wzA-0)TA#)?fpPYF4e<SpF?J<*WJq-jDY4PbLS)UiW_Fr=Rjiz>>bT^XMczHifl~
zL3VacY<AHCd^0qU3(Py0*XMotcHhbv(xm;42g}$=&r132?qq<)#rn@_0?=~Xh-g4h
zL)fr$vR)|)#)iHimOj_kh)?!Qao&aeQ~ja3pZ7#vE!|^gM)_jj_>GFaww*Q5yY)D{
z#nR)jS9$2S9*4CLFspecC?J7Kzw1w8olUO!HmdwrIgYgS_d5>nCN0)Mp36WM4kK!N
zGBboTL5x&M@bp`>JIt^Vr|BU-H1>Yp0s*00&Vl5iSNSF||H013w)fVNA@akW2iO@+
z77y~f&c+5^xgwyw%m2`z5m{>}av<wX1To6?SEnHZOW%ecGlzLla+sVTfGeJWA}k%w
zbCS8cY6TG;28U>Cq6{8pH(U$1LjGzrs$x@>4f8$t&%1oeNRN=8;Q@&G1VJFEXu6Ud
z6U&j>c48(C7>~));gzf$VFfZ*!1Sh<XWU(b7U7|Hs=_5oWtH{fT6H+CXj`2j#XO`k
zpG_&rIi!gQZpV#XnDQV(OAcw;_E_;f%eV-{1|``DZkK)pp&)b?<#(#EkS3tzkuflB
zVdxAEP)m)|c1Ew9_1(~JYx~YEcNSZ+38or0tim??>0;nynAvUS=>HFF$Z^ALBUYDK
zl7IO*J>NfgzW?Zp&TH%^ozu?nqB9(CLxVb8G}U?Val3ofU$k|X^E7h4dWUXg3IFBi
z|Lr&b@bG$nd$9A7d0lRqD<I=@o%`gkGq0=8w<8RblfMI>2Z++0vz-T>ll$9m+0Zn}
z+rIzTHsg=pbH?d=%{b!cS0iZStGXVWf`MS_d2w<(2Cq;q=j60$NeX7dUj%$vZ^M@&
zK3olAEMWoZW1}KENq&?d;}dl;?0H-UnTSxZAfQ{MhDD(q_AEBQc&GVgpmy9TKtZ;N
zwU=;C(8IK&2b9E=j1#_YE2AnDO7hSTRg~>1DCx%&B0dyzKrKV-nsAQzNP<zW!xZ{j
z|H*m|5=3~YZ!l_Gq1xV?up|iwPO7IZ&)LEZMcfc3zv&+IS(l}d{Unei>B^-$?$IcF
zVuH<$Q9mx8h6$8g9!4oU5a~~2<{vXS!1Ef4ec_x92WcXTxz@R_rGG+mcL&GH0jwMM
zuyb4UJQW4C9~*dl?ht`60=RX1NDwmCrTfXSSF`_c+2qSTIyxb0xr68}qc{JkZaK`b
z%jWdQIu@7lfN2@g?U$TTqHhwA*_rySZ`O>DW8FzBi=^*eq)VSPzn8nuxiH2XMUkr{
zBLnB_1LgPCHpY-yt*{=Cs2(A8U%o&$Xf_hh-4W{<jZcDv5%$T$nC%qh%+T#z)OkR3
zv0~*Gw<>2`kOH(L_>+W#7!-qz&+MNrV1gBLyvudB5<(rHo5&GHTTHN=n@)gC0AqQg
zr7g9ir3Te|?l{8JxlImF<2mnb|1OLpwDq9M#xxrdmWF}WP}zb3$H0|&j0>ulTIyOV
zRd+Sv&BKxg72$GhJ%liIC<R|fXX{xs3#g~!y<eJUagAXTS#`9hjkft0*9g^_wThd-
z3gVAxLe^kzM`<_rq}hCKt7$&>Bu9J_w#p8K+|J?zv`u{>8Z6+fwt&!fo`2C$i=YlD
z0lCgBg7%Omr1x^Y$BZ#1;|<$h-qgEZ*uC7XL$PcAjb3>{XQtyVcO3h$!PA@%MqNj5
zjBm}nJ4`L&AKy9K?(E9)IoU(3V2UZqC#nxWB&iuAx5!=e{h48+F@FMAmG+}*PjNbj
zVY1j7o6ez4BJKt15j+THT=Gq{Pd3-;)}yrTxRN>Ci`M<N6@9lUG(XVU``T_Wr@JxT
z4KOHP<;3P7Lz_|VU{vpyLA5LC$)wwylRApNNw<*L41=|}v9D~t3^3?zz%Nu2f2sxR
zHX8W}#xO#+*pw4ByNw2oe;^DtX>XKxMYhSrhOy^mt=*>3T$xu<00@3;Oi<me34+&l
z9e)<z+DN>%e_gnald!Knf0G9O!;%(ugfy^!m%PW;!`outUG|BiT+7cL?)6IYrF+M^
zHfix}n_UWnF=-L!G65=x379q7$r&ER4OpxGY`Pp&ZHM3enIcWA7_wD7YPM>(rIPz=
zXQMms*^Bi^AFz2@7vhD!z?Noh<xW%z5y>V7)P>cE=_X(?0y1kq_p6{<+uLMI@fRVw
z?q1I#j0jsyJZq+6g??$!V2^X^eqv!auw+A?JX{n9bJ$oRo<zC*b5{CC6?U1dgaz+e
zfewGx3KJt2Nj2SE9$;-NdM@Y=q-~b~0%Bw$<k>F{X7;0_s)Pvu@9^bM=HT;~b%>FV
z>o_;Zp=74Dc8~;NjnyhQ7~n6>A^CA44QjML?1xwZed5i1iZADyho`iU@zn`2{WC3Q
z{v&&K^C!ik=f89uf2nV^DTH$ePnQOZ$M#|MQQMB9`mVe;7kGWUC5=X3__fiFtf80~
zNmH_oCg<?Y$+gOM<zig0xy3&w2DS{afoqSC&tgOp^7K)-IJQNFPyVwgcBhPb8Ln9M
zGfG@hrE52kyxT9iq;qm{IjdV_B!Wfv8A~UpPj!uQYn4b}`Kg0Z>X9nzU5-uhu!R}2
zY7a49?k+06({pLXvXgm{KcUyWJxC>&)T%wXAiSwAF>t-ei$5nO{9pZ#|J#4_Pab{n
zf&O1zhzT#QI_{`La%^wAL!DUh;)_Ckzv0UDUsW7=G4rt2zd}U$;vm0lY<aSn08a``
zyD7lj=OH_b>t@i#SRB@>QAl<w>I`{B3vrQG^|d<!Qh@o5cT9`eL0*^F@kTBti!)R$
z_x;W_ymy*KyloW`69^fYzhwJn3J{Dx#Woa4xkXx|G~qj3tR3^?-S=x{%~TI2NS0D(
zagom5kbS4p+*Gaj|4Vbf3KvR8${2}>Sp)qhafCL9*S9bbULRn{X_z&DILuo?dNA9R
zE~UvX{TS|r4bmJ*@s1YyaxUm7l?rPX;i$pBff&FIOmrG^jfIV<S$2h}1xvFOu1OA;
zF^Q{L9UkmFm*@kZ5IF`MTBa$&S(bRpnMaW)Enu<T($Wvsn=o>rMujr$$TzdPQwl_|
z#n>IW&uHE#hoD3?HIF3$l$Q5ypQdU8eA-b7DF}pR@2eY6)7MY=4x+S9W%An>p`Kv#
zMxhc`nw0ETNa1CWo8^_e8C;gsJIr-Oy?l?1al?4De(#RrCsdpbNQ{+#JrRza6IET;
zX<Lz8{yiny**L%bAX!C<f`Hh&mn9VV`k9q8jdw@TS+W>ZfmnpdXl2_WzIAx08=LLc
zX5Hk>6`x4RhdbN+Y<D?7(5sIgaPjfdqENnS9kJIg#iV>qQQ*;QN^%fsON2A;mQ2U0
zHaJ8CZsa}f#cj0h+j|&5!Mj4FV#hqmR#zv=8fnQrQj7c|QWc(ULWfeDNQwgEb=6tH
zEmQ3}`t<xxkUG__BS8CKTb0Q${>@iL1~naOVpY$5wqAsVVh7AW-4*F|AYxjvisgn%
ziA#%uhGnOni12emVaYobvn;MVzH2qhhK9Pd^dy3TS(kyb+_=^AD<91p6DbE&#Zyb)
zqc~>l!2XRX?kj!-QljlSMXEHVs1Sv>S?2eY0IP#6;^3&~`|OaY;9Va<^vMb!TOjk?
zCzexe*Q66rNys~HdrYR{Q^tuxFfizGj<W-nLaNN9M$8INm~Q(M6@r;|)RUK$6}Fnf
zasyr#Xu0PLJ6y8`5Pukdw{ZwaBr0icNa%#)F6BC<6-iar@fh{@&qd~R$5%YHc$XC0
z`7)G}ZO%CFYsm!`<ySD_cBPXBw@VGSC1uC4#B>TR6s;kUCdd$Us85#3(u%$|AZcTe
zrP31lg3V?Uwj-v}PdREJNTdvr*$|71*!^zYVoyv^aqp_KTRi5RK9p3h^3#?z1bg8T
zt|G{Qk6+LO&jMLKF@sCRP5!61{ISj>=weEa6+pyhv3};N3@+PHlH>8Cr`2;SGqEYA
zV)m4tHDst|oZ07r!8hwE?bDN7qxYz-I^^_hb`jBcZtfxyNSDk>H3IPhz(G0nn<Ly3
z!E4D#i53<ELI58<&9!-7EZ*_H?0qV@k%<>5xP;Pyv`7Wnn+4e0#AsbE`>l03(pSFT
zc<ZxUvYKY+w-VI#K>3)1^%fG=57*e)wRnfD@3LR89+A0CP8g*l@bFLvAV1WVuE}`C
zIT<(KPe~&JB#ZI&#Hi-+m4M*0k&`T%*d=Vot6vbJY_+zR8wx^?-u;d0s}1g^Jpy(t
zKhATmtQ%E)$T#geT{E{a(|ql=*YsN8q+2*K@gE)eO`OQ~K%jeDNiG<j&Tz7D!{n(G
zY{I+zQIUzk_ce8OkRaV>st{g@j2N&dqvP(eUpFhKF>!=al9UmyYm1Z;=r_SKw{&|3
zlktN8^4G?jV$BBaq+pIIxyo-=gnSig;!&d8!WoEku!f6Xg{O>q%!MBQLaCLMXe~Ye
zK#O7<2x}nDLEp6a!D1WcDpN!(kVENkl~vYf6c4MXt1k(4SV`O6OGWryc%@2CB*@(X
z0T#B`Er%B5)KHN{{Dn!w^}6J0s=h4qa@IHBv#!xnP=>4`&7~Z&j+Y+Lp@oEYVqxx`
zl<hY%vS0{QJHAqzmb5A5Woiyon?pb`qr&3UGLHirJgPsYDLN4VM#)z>#WE^A0wLEa
z96k9lcp)y?C*Tejx*dmka8NqLd{_<%2RvIwaoqy=Gnk;n7G*b)C2vJht$x9bYZL?J
z<xtP;nsH-^DuBjv;t_`B>U*iq0WR&nq<GVZJ@XZs<Yb5G$p9gDv_KhEYcfp=#H7FN
zm%{b~&*7QtFyM91a{^LW2q5j?3@8CJ+vrYqKT1dE79puM4@9vdQ2KOT1yvRR%0MC|
z^bujN!mYxXg+ZL6{uDH?<xR}Kpeic{SBDq6w3?t}7mn?s;Dv6Ld&s^R=@ldpqZX)c
z$OC6rLMYDB(Zbg;4fR+RTPD|I<QhL1%E-~H<Rb-_N(g|I3Hs8uR)nxjuE{}p42F0L
zak!=$Xs~Uu_IBC4lR0&$;HJM_jlzsBa8cPlM9UYJc2t&y&}gL2>V65kE<H$ZM^vJe
z!;f|O1va90)&&_4D8e=>VE{@dF`YFLa&@?jFL)9)FAX=XD<Z$hcZ4#x!m<NdDlW#d
z%)iMM;HiE`6J_-}Li#N?P*kSq#5i1W697NJse~&&rA`IXJfr>$9ELSS9pgg9buyCD
zUlUCUiCVNhEgBk1w(J{j#VUi!5wWFEITk<gn!na`Pk+O9!{sO+XDz}?|LU!fyQx1c
z0O=A<!g|kC_JwnF+Y3u=G@VL`pf&_99v9z#gUotSUmUyE<uum`vPV~-&UGa#BVk_$
z(&gumx+as3fv>7GIQP}`b@&Eth7QK?ReaL1NWtB7XBj(UB=@j#+>5;+%0S=0zYK*_
z673HtWm1HT17g<ptBp^Gk8oeXL4-`zBb!2jW(_1W+N_PBJ$EgE$F@KH;gxQxBOk`J
z4;5+*sq#hrKL}|Q{!ME&bEt2mN8qt`B@^mvcVv4?q0?WJ*<kxwT!ZG#W3v;*L#{q&
z`P+#a&o|Qr%C=fUGDXAhdvkU(ih}9W+|P0f?lK2~f{I2eCbzzDXH_hfm3wrNRHQpU
zTO{knyF||$eJz`|_qg!phTa$3rIEBVZXZx1!xJWagA`C$?yTjW&3&Qwh{5gO1<tfM
z#sV%%_y@Ix9wsKZCcy8|MKg}<1O&5N#x%d7-Pq+AD`eqQKs{fT=3x;c)55#ZLLzOD
zK!9+r<LCym8gvRUaEPd=Vd=~Ure7%UPDDdPY)!L7AutAfIhl+=Ft)>%;oObVOs0R*
zn^TT+CBIv_2~iJFS|y+yH3RP`JMMDj=vn2ByhCv^I?KEaAp^8haa3EV2}aIXAYP*@
zsScWMu`$+82#)rw%x}{>Lai3MK03PXaCO1l(Gm2j*t?={UkRNvCVly@W)P5l;!SnG
zR7dPWAV+_riK-^6gF=iz(1oXZOos%?Xihm`DJt9{0u`#`FpQO<bsE*6Ti=H${9uyc
zTL?mkgvMf4(unPq1vp+tfI{KN^reY;cO(+=db^0-N%&hBc@d8nn<Nx=yOz`Av^?#P
zj@>2-aI)3WbQH^BO<#8jX`U@z_u_DB-R1J}mb?R?FNpL@aCy;)Xezvg$uG?37A6lo
zn0PDeSG2J$Sihc7t*EP#grz{@D0(OB*KOas6sPJ<XT&_Bzgr!jc~c&tjub_3G(k^T
zVk9(z;WeJjNNkoi-4PLr!$n!n5MNHFiTMavfBbUJMAjX42iHGDijWN~UAskg(=JlA
zPupZAQ$H<`LFOpcPNjP(u1nnz)wD3Tr-7KGrtA*bMl&mp;N~9$pkr>4;O3T&<xUui
z=yuxb55k(QUSd3yQw1CuN3ebC0uE>prraNfuG~Z7A-x@luz`nxhDz0UV)TPv;jD#@
zn>en6^fFvNVCanPCN#kXx1s>x1*dWCACmSk>B7}cM`pOm(G@zXI3+2<yGM;O@Zt2F
zOUSDoC$@0i7j#&{o`w(oxz6pAki2LsFnX>~g;%lJGUZ`m*c)6y@8nL>&<oSvFnANV
zWrOMs_o98x_cu1Yvvn1E6`=H;{76eD|8q^S0a~Li)#uZ6B4w$_5%(BnS}FsyTp4uX
z{%pG*RNjRsFq!D=MpHY(fEX}Hhl>PA6gYxLM4C8V_hR%&wbF2aoT8!KNbNMmR0)e)
z0LE<E!G0P$60eDfX|7fYBc&%6^@Ato9rqv2Z$xhb-YB^^{tIobSbfHCdLK>Bx<l;a
zv^KHDtqOi;Ya|1D1Bjy@u@X<+VFZjADJ(P@ei1GvR91Fz(>|HIy&M6=OM=^S9m(L7
zEMJ3I5iXNyKunAm?jXa^=l*}0{DP0nzg#o?TPv;3TKjZ5I?-)qp4e>WGwV-RoMXs9
z5sDj339gDHUDd)g>lOvbDoTE`azsZW3DuijVE|QxNJ{m5kdHZwn_<N+DillGDH<EG
z6m^ZYafXUjU<93v`!a(YOzcOqx1Cr<PzsA;gobdrr7b@i`z5^Dyxp1;6i%$;gf#2K
z{g`YI4PcX*QR{>a&9wf^T_al!T+uWLOej-4b-qdBlfJ+uOqZ?#9W)V#8x$^Q_}vf%
z{RgR*ruEo-q$)`f4e!E1N7qPU-pijXigv|2m+1(kFqsGk>MEWT1}POt^Q~YP#KSQE
zMh=#;=<-*B1})Vmn}J-+I=dO%W*Qw08zx9(bHawi4Ynot9#HW?0%P`80+OFJ0_H8D
z1?Oc*;I?0Mw+O;IISj<~GuN`bC|KR9vxrC$=pzi6k`N9=qXC%F3yAvYMn%U#lq8m(
zhl*z&r;}`W1umcvxMXf7O3fq4)FT4;?AMezn7eUxH0X1?m}0$)!<(HOhc?5Pw#&X(
z9B~7Rd<knWWvdOc>=Sv5zz5k>h<nj&oYXtX=0H@`5g|smRRoL)q}o}@8D61tnPwrm
zIs^sF6efH8)K;MTEofW{%Wf=C&Y*iNMDmM^#LeOwRSoR)Bo4+A=OItP<e$=+Ipq*~
z7cFl|NH!TKdm8}9)vvO1OwcKgY3ec^-U6zo9R2MojyjQbxBo)2W4@y?A_fbyXYbWr
z;pZ?PU|h9NJRnDo!r-XfWNU|Zzx}X4JcaMHWNkaK8XXqKM0T2%CN87=E;)QsXBdou
zFT8_|C=dE{lBAhr6V4L4M8wA2=K);jDOK+2269z2NO~xXIS?JkDK7=ygx{=>a>-Og
z??B2C>?3@5u)CB|LxkLWB(V8OJKpmuCDi?C7=t9li#=M!KHr1FST)xvM;t87zsGCG
zw39c)aSA<)yKsRU6Q0>I5DVMBqng+?*hiNx`peZZz$z;L9I6d(iOH7=X>8H&k;ee+
z!3iKOWa(tU3CoS^7G;+eqLMQhE9o|wrs;^nez@_01KT(3@=jtJBqPWh1;tS2fo7da
z1vufu-{471(jVVQ05U=k%e<7}qi%dLg_J0wj%#yD@u(5EThy<mtjfr1qCwO%ZQ4c-
zBsd_?)1ByXu&xQ(jY$V(%lfkQM7y~(MwBp%f=aj`4%7blA$<wvT+qsKfVOIeN_0L*
zGi=)ONGW<sjy~s>(Ot~?4<ln&4Afy1?v@2WEy1v8-QZJEX8u8*kTaOM{5%UvfdY3j
zo(OqJLvggY#c)y}g4oHLpZ8jj9s*Y>7-sMoO3ehGaex8+g;~(GU=B#dqT!S21c3sb
z&)k8E1(}*1DnnZnug@4nRi4mW<h_nyM#2eKjb(|mjAIzQc1+&<Ik#-QMyM_)3KSG(
zqNrd8LCyoD%6PNbK#2WIgqfxOZj4Z8Ad4<bE#<JYjhMFx!TO+8LXzNq;nO1674-c3
zkk(-ao+9>YsSh!Tzlj>$VttVO70oja7XgPAizV{r!0y6ekXWQlEP58_2W6#JC_b)!
zi3y=x)N(VzLX7>=*!dL~n-ru7w~_mLK1g0UE@*4@Y`VHmko>BDU~%%)&*1VGj|JoF
zx6bKZEEZv2UTvAsEqqW@=;p<=m1oT8&a?KCb!lO%-p^o>!(I=|38feYKVrf+frTQV
zOSVd3aDadis8Oou65ugzbrtd>Ksv~+O(ln2B}LvXgci{iU@qha!{7|$%q=QIol-Cy
zP2{S~6+B{>+xq?snNZ%<2*7u4u2^aWI5O23d)$#r@A}D^vVMwSJjJYt$s_&FtuP1#
zQ!Hv>QXZ0ayJ?_l8HhSkGz`$wlWZASunUx`eiCRJ3d73ED^GL}CyT~=xRw?`+@H@G
zXfGT1L*54IM14eM$lqlM(Mgw>GOx;*F+j?_xKJebb83y2d8*qBqbd`P6DfQnsxs!w
zT(m#vQ+J~2LOl%iUTqYqEa=P*?RH)QWNyJbC6sDnHxV#2CWT1l?6n=FhnaAjt>{Q7
zm>Z-#M^njQ3eUPx5DTN41lAZ5I_p}3oMuWqA@(tadQ13Ja2Wm!!Iw#odE;C$X2M!p
zY)ahWaCCWgF3>|ol+ZcfE9iG05zD$H0ba3;9J!%&f(QDeaQk#jHD#0cf;GT2QFyF_
z*qpk8tmQS3+;3+0PjIoLFah~#{*(2UJQ!3CxU!4Y6QISE#SC9xviTV1wNWhE=!1?Z
zcXR8U1184S!uFIECIRQVB2w_0brZv*65;avvgaB>;m#ho8L1Saj;0ZP9uX!B_8b%5
zrNqQBArmP+4%Xh31Jrf$&e*WBv-M|OnQ2wmDcS=o`y|Q=OPK9q7*85_MVUKNx*F3d
z7|urCMq)Zg1Iml<#q=Mp-|Jsa_a5ZLoqQ%}*qlY=3TZqQFGf$1E803^gHQUClL3=Z
zsUcgQ+c8-60b6`BIvM3Dm>xmItbv<>w*^+wi;g56n{|^T(Zmbd|A_mSzVe>4YL8P|
z9ioz`K-Qo>optJK`<WQV^Yn|q=uXrqywERZwDYe}OsTrbg*B-~`NY2;9Fr=Q-q-$X
z(Z_2p-JInH=~w&uyg$yb?3d!pSLoIGZT!u?>keNHig)2xsUv}Xzfi<Q-{_mj1)yYx
zC$<zW@%<=Y8<oz`E9!$L{0AkdF>33IQ(*q|a-gp)*HYAhtCQuzo1rSe`BgiL+Ew}~
zkYN2>V2(z!qf0|A7V1BOo(<LC6&P?t@sCi_OWo|drp+BwTpDua#rz}5zk)u~H){nf
z{b4jlq+vr%R3@R&YS};yx+HBgsuvTf&M&bdS5xA$>zU1-W&$8xTIuI}UHsggV7b5T
zkGt2o*^l}HAd(F4v9|<*>N<hlt^KJt)t+xrs(_rFvNxMs1j;;FL1FsMKrRoU-X5wq
z3Z#Fme?Q)+e&6iXr&V?vCK~(oU(1fw>+jUt?>4GW^e>mCQY+skyL^&>dzzwvJN;tN
z9OkP@ZtHSr%AN2Nmr(Q51AZc}Q2TJxOhf1XI`;0r#qKpzDvYQ@6;k02Y<{AY1XU)i
z>leXe9dnG`D+S2tW{O}H0YD;D6DV)GDqigX%vLik{M?D4mFQCfk$AWUD$KzSnSbLk
z#f^=}<r{2t!jF!&zS;bcwkt0#?^|A`?;!2tf(y!lg-8*G;)L#LF`*l5(UY(o!_$nW
z0hffqO-u)W-<20-IzT3={gU?XCRW9vdBR$QXXSep52&G0_m6(N*81QB{m<2kA7njj
zOLA5&W;5+j22K`wF0r93JcoE|IgO%f#tLQ3EXN~!=%$T4P<6GFuV*a2YyS8vs<)A+
zzOX01NZE~||F@q3*Ply>Dld(LF%1xxyf(7J$q-d7^ff+1geTBe7vLb>(H2qIuAPzA
z5c$L_Np|$ise2Y0JkpH5fcpx@&}tW%fA^@A_42%RnY*stR4%y!V(pqU1)3-vkLoa3
zE-hTcgddOMvQQ$iNH=zx3eoLh8CHU`MIh>nc%x!wvs<L-WaD_M>|x6KV_kS9mxn@q
zStKd|m#*JfPN1x~5leR5Sf4{RaV96v#F3|mtGcL(Iz8=U6y?>I5VKTnW5G<5cPW~I
zfd7S7%Z)j%Qoxd_KY~-B?%{WkY~#wJE|ClswmG=TrNj%FsgP0d?RAemuUcm62DrHK
zCyM#>c3ux8b9xBZ`E16ei(6Ya-!Lebz?R)Gcm^<$2C?PKiQR;46)3Z-XKc<gGt^o#
zLt)RPRznv6#eTxgnct7k)-XiHxWXTGW0}J02~Mo=6#vMNZFAEUzsKOi(qAkA+iI;g
z&78*6^zNWYp)U^UxHcWjpigeJvr8_SsCLaX8{og4?lwOnA8g1L>dqvb*dveE0A;wm
za8j?6HfE7|Get&7$BkAm?%5G@;b<%}g9|I91JhAzAuKTCIxdp*kQHC;w4b96j4mGG
z$!U3^sGS{1KYy`nQfUk8n(Z6PC1=r6W|*Zf7Jii{9)ce)ol?-d#p2j%6N)mg+s(xB
zt?g~FWS9}_L{hhTU6yt%-!CkNji4DC5>@k!X5hp0qhsXeRB(ijxf;rn8$3(aD2w`0
zl1JJpseNqKqY;+ZY29*NDDB;I%TMhs+d`2ej%(2GPQA>#D?;hhugEO<LBh}AF(qSL
z<2@Xf>#5Db=yL;CM_e;@K0w@{!1AD;>+WgH?P*=$&e%J=-~Q+m#1{T}h|t16_kTwQ
z`gGRQtDT+S>zN(2{^1(6jTO#u?zpFKJMV90u1TBZN<`XGG#(~Tgr^i9zg<wOEYM5@
zPx?reU-DJ9K_2p0#LvAF*tGgI)C8OU0efNHCP@A0=!pR9X+3LEc+4(iinzcHq|p|v
zY}SCE=7}TZVTLKdj7(x&OhsB8K7+?0+*Hmokr!S~WB>wbKvt?q>>i0kLHSOE{-5SC
zoPo-pICR66gQt|X3!JW$_{FU(O^$oRj7TDWU145@Rgtf5p^lg)FcruJz+XOCV6`kg
zkJJhDkr8bVLZGWT053BSE3w}BWa#tJ<?QSJ8=C|F{%NdrP{Ha==WH<H4jzhc$5BhM
z%|wWk%h^T$O^5$4#|UnJ2-~_CkD572cc6W|W;X}d66;|j;p8)&EGkj~23OnS`8rGe
zLGw&|C_%#aWr_?TkaMwFa>Uh62Egg3SA-i)I#)Z)uzG^?HFRdQfP3snvC||P-x#rw
z%p<ea1$Bb@$4D7bR*AEc!QAusG&%FmMp5SkhcNMm>ezYJZw%%>kE;9zWe)Q3Y)}r6
z8h%YC5(!@8-^5w8%p15%umONiOLtX*)DaOITt+$WszkPmIGgXtrKscjS~tBdBht20
zYzaKYJjV$R2}ij!dnUb^T_37atPkd!rK{lG#+G6A{le{=F@CdOpR(!EP)txq+_5Bm
z+9|~Gd^Cj7d1)tXdo{c2grbdp8Ky+X7<$R^K#yFM8^Ap$1Sp70*a(-?q6kpxr2Sb~
z=w~B^_M2+xgkK<VxGkj+E%L#~QYp774`eOu-Ok6neDIOKSa`erK%ad?6Due55#-lC
z4L!(>)t&9F{==XD;m;VVIDxA!x7=fc-;=CjvdQ7*&aU`wI+ec~gL-y8nh`0>UB@S%
zV{%i;{_!5B1Do;9Y)mbjq2MrdIUyBU$&>c<MuVa#p&7sPC$MwLYA_}LRPIC<J<9J;
zx7DIK%6b+`0Mi3dGb%5xe4ma^pqtoHF|_&B?+N(Y-1HWWe9A{M7^ag}LuT{h9qkNq
z?XS^Gjy$_KblE@)a3IEL2@MO35f_LBdEYjO(T*c*Dv(F_cu2aW!@yr5LZnSNb<Eo=
zJ0t72BwUdx3l$`g*qP&J*Jg&7_7=hzO=9g<r_dcTF$t`LSM9ovC!C=REHnq-9asft
zxR^K*vYzE$1#A9=>tTK!sWbhUyYDjU7Kf(TQS>U08q3rDTUYy76^vZ55k6p`b#`qi
z5i2&U6sbUCrL<_g4s-0}gYkF;v)pR(Y1*>PHt22m#mR15oN<T7$+vykN08EE6DF0z
zdcdycyQtF``kG<l_l-n`Jj1rhGDqTkZ)@2gNq73fz|s!m4&6I=1(_*rfm3Ms#A#hq
zo?l6}rp`oVCKBCE9n^Jo?;fkZcQ2F<H23uvV!$Sw(?lk8{AtdJOgo1zj|UR3;__YI
zjT>!)A<A*1MGKLRb;vkQR&ii^?Rzje#i4ci4vG%hH<tB!G<hWq@1{dXq=!y)dSfXy
znb6z*hbtEl9`RhwWGCE_<lMrjXFu>4cxRg8p3l!*Nk){ot$^#yR)1_3z+@aqm>6?_
z8B;k^7_8dcWaiaRzH=(iEQ5w>kF<O}DoW@AcLgfCaw9ntKxdmd4yXh_8<I)8kCL%8
z$z-r+L(>~J>&%L01w~R_q?hRQb_aN5*?M{Jo*@C}?StrajG{UL;F5Yosw}eEeVpjx
zQeaItOM!{Hv)sIf1o1iq{uzqYNRX<`s+4`iE@~IprX>Sg=9};c#ApLzT2Lme60gr8
zR0<e`N)}SP<eXS6j<!MCyqk1fK+^a+h{i^5povd{W-bps<nAV-BidtxXWtXviT0<!
z6($;_X8in7Fy957ky%V_fdfRPsu|jWj%8!9jDEY8uC!FGQ=Y0Efy;V4CaIQ}K`^8h
zeKo}}QdLslbG%o$l>n{e)KMFI!c8$|;K<D)IWXdZxoy(5927B#RF{5w@17G{Tb)U!
zcm<YZ(^r`31{U=QB6w$g1WnSA6@-^UYp&586$KwvLe>aTF)HQb7@JJPJ7HnOqP#({
zsUR}Q2WDJ3@IN}U&y!eD?1@9BTWox;n0=(qLM_mRNxjn?*QG4`QvCot1BBOR4Hxds
zO_27^RAsrajyQ@RO;B4kiE4E>NbLU1gK>E1G%DorTgxD2-j`iK678z+s-yqTNKLGJ
ze8e^MVRNsMQh($=)Np?Q?^b`a`qTQ#PqHmOuQ&6};l+Uq^sJUj-x#@D3r#p{%M6iW
zDy<}c>_Fu&%z$}gA?A?2wO4)AkR1I}N^?rGr=-%n3_1TEX)@UlHM#J9ZEjcAUY4{O
zPgg=vgGlL+V6EBa!nrFs+|kg^(<3@VHj}s{PuWAQTv`Fp>mwTw1EdMf0-Z$VHa*tp
zKM3nAAe{^(TD<`|23P+~5GMMId`H)|`GBB!xGA`WwN2mLyC-pck<hzYBcG0Xz+O2|
z&F?)1Z~z||h4F`An_zNgvV(KOM4fe+)h^nMHiIcfILVHPP#<ekKs33~%@l%5UgB${
z>nTJ`f<+fnW?G)#Hx;7bKpi;~Af-e4>ea_2!!r+8-6cut=)%aI^ZHWZ2K^J4B^S*e
z_wJz^!3W8opRhqTQ37gZTdKJ4oahMXPh>?#b$uCN@Tr2`r?2{idWaS?8T~={Hcl%W
ziZeq!eWpvimdi+SPdRx=bG!_L8#t{DqD~iLA{t|9j&YP})H{GK|8dh|6v)di=oP7u
zPcUSu!ijb{rSP&stXkp?`s0F~TY0Pqfx(jiUh0JMqB$`O1qvU;h1@O`yG$c#2GDj?
zxyHCHyPU5l^rkDC6AUs=%uHxtiHEru1{4u5qcbgc6gm>co{&dozyT;DYF1-yyo;m7
zow3E%CSIu=QnPhHIDKDuU?Ov%oDM?TuzR8U1ziIs^l-NgiTze@ZST^2ZqH2O5YL|c
z(Y21CdjK*a<saA)vKxV5yOE_sfep4GtcbIA&Cldar*R2OL!0K}hS={+Oq@&RE+i+D
z!4czQ?)P1nmu8L{Jp`3BdB+_cTAFbjz7)rg5=$=uCn8GH%3<qPPm8@s-PxWXOiI^4
zB$(R+ua>-O(^!SVB6U~(n^04RuJP=A^EJ1Vrw*9`I?}W|Sk35eWCYMKC7N1j6fgy_
z!vST%qG8!^1Lwj9_<>oVzn!jmWnf1*Hg7^4$F-#Z5PFL+6DYYlK8g{(?5%9;&5fY~
z*Z^f;c6FCM!tzB7A+Nrd<yjC<Xw?JKv*7r3%!+5)Bp`_=V*G4>OJ?*8%UM%z{!J(@
zCOV(Hj(21t<S4Ey8iOr-(6RO)^=&jf`6m|Jv#@2hh2_so@PDpt5i20%Po-Lo5eIoL
z>Tl5vf(HW~JO^q@mq|#QnS5SayHXAyt<Xq}PJ1<6vwy(?`F0}GE)Q@atYwoj%W;iJ
zcPGy_0YOb=rl|j>a|RMaKZL@*#%s(ZDdG_)?Qs1k@n1^P6y2y72DqW5u5C_sD;O#5
z5&IUd@0j!FIfrU0jjTD#^?%2t7)(<7awpW>UcaNe$>1XFq~&U{=Vv2d`a#}Lw@(BU
z2=xJ?qf2Z!7!Nak^Q<%%dgwzo6&Xbf=E&F)j>?nK=h2+5B9E4s8DC55;b&1FfsNfm
zZSm2;PVm;)3}QDfk)BL_j1`=c8g8{s=$E9xP!5HX^uuI>F(oDLG6J}7&OR1O{4-D}
z6{A<^D&g907)-O~x$9m<Uz!5XbytGca>0@!;zln{)7t|*rG3`YLe%lnU3@+1jw4m*
zFr|`MC`2X3At$j=a?j@RV07|omkZ+1B_+@?CH~AiE$os}L!c1g&^R)9&aGp|Uc=s?
z=QTbzr)U<3rh6CH#oP(R(~R=Bw~rMs5euIwz#3Wt+{6z2P^S4*T#9y7PvjsTBZI#k
z)jcqsn`Ij`#AxEsC>6XS`QJkRw2u^Y3rMmAI!&`QIYl6ima=yB2oc*9>hvnJHDD!q
z4kboK15qG`NqXsc5btORNPw;8#|3}8x;4s9>cA{E(#mAbF;6lEPkDshmc!Idu}Ky&
zCXU2wuHuRAZox~2km_)@S?#-Chj%&RQQTbe*^Z7niYrrn=}rg3D88HX5u*EX&D^5(
zLiCfpF+d2akR(^}+sork6V*bKc*<g#-2b=lwtU+bmohM_ST$SQ{>*YxWQA?O7nJ3M
z$AI0=*K9d1J~=BwcyrE?^pCKlc5!Q$A`ZX$%fI=*{EPMnAL#$z-X3w7;?(57cx$}j
z%99w@ru`NrOx*sK_`+XjTw!zNf9<Hk=4ij{tzIg%!`<k`VHAEA<PaB21j;1^&qs$4
z2tFFODncm~z_7YX>0D*y@%N4PBeNddbdgscCEwo_uJ1BxkCG)DH!jo4w41kv4&E3o
z7{du|jSK$oB8(!AM@%3Fue*g9VkUO;NfF4P;6PPTfvb{eCDt0;3E%|5VuR~VsOq3E
zeLiCeUdUJufFO7Ai13jq>d|0pw&nB{J&1+h!<K-@DE3wNe&graE`t?`d*`gOEch!A
zR#eSknK5C`cKmHzRKwOY>`@ChOU*?qYMB;+`ehwl?}e$4cgxfW9_QA+;b}WPFw3}?
zNk5>Ajtc&RFvbP$%3h1){wxjYhdEOL95o8w0P#V)Y)5)B;=hF>Ao8JP$&uZUbtcTK
zO|YDqcf~uND8t_><znmrcq}Rih(~B+eng4odeCW-z7DVLDSf=mD?^GSQ?-v(Pn=AF
z9@NAv;L#TC?Dtt$MykL3<uCkwK-;$|j<hE%IszjfanCbygeqAqOHxZMr)BZ|F`Y<c
z;$IewyyA;Fwlo=;4zNp_a%;<rrCnBNPRy9#Mj;)C=&;PDrKg<GUifz%X6?Y`0SDo+
zYy<LFT3CvLPwsPxtw&76_Ajg7YDRGg!zcqak74v}y8BM~SMR@BB-Q~ywqO3E8N_VI
z+RD1)Wn)i#T#r}P%lQq_Gv5u{25kvH#7(eUTbFK4IaJ=8M*wv_QFv$CAGYYO(3HP^
z|9v{KnILbD#<p4d!6mZbwKc;|AiLhUy;6J>VJgq*m*2y3xV(pt@hZ>lAe(s?Y>Z#N
z|L&bKhqvvW6bGbu6^|QDhVmGjiV4-wy0}`6jJ?sgy+XLaxT6#-1X_NwBq)Szs=kdb
z!t?M;c;(w72R=9**AybI4LbR2y@2B-h<o_+#>}x$)pKLt&sY|zh~x!I0}v=T{@n3g
z7W#Vs15(~Vn3&VfdV<nIy&-5ZVu$6jhWD91OLd6csci)-LX)iWqIwje?#1$3#3vwB
z$pc2Y0Yz2cZxi%Egj*YFqnJ)2$mO#{`0L3WLGFn7Flp+df3%d)l--EdXGHOBeLnJX
zO$+DqPKOhFdH<7{Xg+~J+mD$*J1Yw(#F7Uj40-d$uE)8wdNsGT)zdvo)%!%3M0@K@
z;KSGNx~F9~^ZU7fdWuVDZa3d)FTIEN-^s4cWGQ(2tD7L(-$tOu(^$f^&o1w+LQFrt
zk7L_lbimoM-))i>@3)oh>9EcdOA+|^A%_sUkAla-XzzL&MSuA)zTlb#guMfx#mcqJ
z9f-)#9S>!`r}saY+UFKGcyeK9<mf2JQk=8%TZ}wo?iL)K;VeH*E^V;hgcj!-NYJ`r
zlV4aAB4l)uZNBC_CYu)5R#V7M#TpkzkLNqTP<$xk^X}fx!}}spX;y`Nw4|$>4{v8F
zd7AGwpP=u@94Ei?spYTV|GdP93n3`70CN>bhJh(I=mBaC(JgUCi--((N4dT`pMVEy
zXkt?M;^wVePTB4I`TcTv)iQ*Zfc#xy-gmF)>-{$|3cQ;2T{klYtSDqolD1=nx{z%>
zU%;k!QG7SKYwiGwm4tIpcGH6$v4kM=+ir4bhyEb;@~(%{=q?{l5#0WE4O<a<Rr!T0
zbO}}7qGe-+>Ytj@D|Hpx5yR?()#Po8_OZ#41hEltjbWH!m7HpYK}^@Cz=DbqOQgoR
z)tnU+2qR81*-)Y(=d9p5A{{i;8nLSXg|<(B?)=7Mq&g9@5=iUMR)Q2<<f`@;Z%%M-
zzg5Hja?-sv9z+nLwj)Plq>1~SA-V{DGCFvn9(SfE%aTRFrB7%TjrVR_4g|O;7s@hp
z_p(*+wX1~ifT!;nGDg2jkMb3zsnUBAlll2Bs!RIz{SlS)th2OfvMacAKoL_hP7~Zq
zO;+?%us)%0q=~d{)-cE#4d0L`cyiGl@2%+pm-8{2*$5t~q|M*;DQvAMjkFnervAnE
z%?<HWhRHfyIz3qq@Kd|sS^_(H2bUJ*i5CPIn%sa#?UnCs#!mT*DYWtcdy<3bZ4con
zG%xt6fB&u|$G-~Q<URb7OLeCG?(nn0<=iITl?#+TG`I2Rn5oXfd^*QWs411kSCw96
zENSXndN+xDwVYb(RgL~ko;gflPZczNFS)aScF_%z=)*R2-apTxf9URAH@oa6@3r&G
zCstm$39xtF>#~RU-)roj6neHzP7v?U0tdB{QRaJsCKeW*_KG))`&>?|wVqxUFdN&y
z>qf@!7U0&0d`3z-w&EY=HL1JH?Q{LS|E6MN_9->@H9hvN4aQgHqau<=V|TvYfiQ6e
zp}_i#H5gkpY6J*lRET3V<j1rbEq%SWgTWxS%K)7oP2Tk=(wE-ZJXn7_qL9F{lVBNl
z;Z1%Jv6uHhhiRBk^BCgOxl?d3(=o_0URuCC44E}ZEJ#KC+AFc6#tFpPw^GBs?ceW6
zFN)3cLBuETd=Bx$<;Sq!Q%KUc=iSLzve_J&Nj&nM4rQ*-_dlDiO|pZp#%I|)BY*2B
zl$tE1Lg{VCziHB=xoNj@Yo&XPCc+SVrZ~oAP|K}ipYMt&%|7SxWS`9TC@(KPop(E$
zqPJh*T*}@g)J;s>MG$gX9pSEMA>~R~(O?W^gSq1<K7~m%8s_b9Jbq`~XnegKLB9Cs
zs2A@|r(!Sfzvo=eW%Fs`Kz!<0K%&8Hgz(Enuom9d-16AAL5ku8+3EN`Z*qL{&iflb
zoZs60qTs__G*_vvVhT6vSp>};t6-->WwhuEw4}b3TLL7#va*!0+T(b#`iE0?oK;y<
zMCIUj)oC9H3^V21L<{e%FBIQ5&s|^_yvD*6-wcnVRG}B#!14)%Q&Z#3(8hK2@o3OL
zQIa8*8Q$&<E@tt}KJ+LrVPa1CwJxn#bNoq03pB5VuiKBFJl=mU7wyuwPrly&@{6Zm
z9Nh5w+wZ=5`Vw5Wf)eshEsWCg)dk0}O<2x|IIWk;fVIyovFk8(bvxCy5JnC=HCLbM
zmnk(eC!^OLqRxoAWJ*{C;yl_@0!3zoJgluBZf^V4E<VNPwA@d97=5&IRMVve)ahTd
zte#C#P18Vo>8B>m-}l}5Ss*8=D13hu(lRVOLbYra3~0f4;74(+?R*@rV-|f#I1k&M
zHtY6{UUb(t-9rZxsswfiVU91TCDL4Hz%m{R-`^JLu-Mz;s#8#+BhUQeKU8eFa;ubg
zgeXqQ1QE?hHcZBg7K%(&$dtm4HD`Q@%`A~)MCmqG;qmBH&T<c6>Ng||1{}9yiWX9l
zS^|Rg@Qfe}7xEWQX>sYl{9$oQl-fxLbjL&SuL_FC#-ieYGytJ4=AFG(VsHegUWCg=
zoX)~}z1LrUje_<@MA$(*`nS}$iIK54GbKOsBG}9&AM&wu&yN?ACf;?dyXydV!XEkn
zl{exs*ocFRJZJVDYK;X{c~(Tl{EcaL3NTL*<)Y>yJ@CK~sT1;FZf>4H5q4whr7=iS
z-4xVI`U9E`Zwx5ad(%{DxFTz-$4dJqe#H!OF|Y(S1>@w$I|9+N9t)77m!D%TIzOdK
znMqEV^4&Fmo+e%CrI0^bB#B~Ilv8BH#i_Mdtfx7DKy&?}7}`=7SN!^2X=!n0rhqgV
z$LQia>@$i*Q!%!s`~Ob6U-tI?dw;dK_fTCAoqLy^w|?e!zMSc8`yRjjrT2Yb0fEg8
zf2W<c@0(j3iAPqY$ealU6?qqA+FKyhRj`JTZ^N__5KZ?a;W?%~%Fe{!Q9ki^R8K@!
z@p^tsK{#~({yUi~^8T6l_%GW3ZS`OM(+@t-|G&9CKEB?m-x?L)tfG6vZ!4MPBh>Cz
zUkrPj$lR8nFjrk^tC7kSQWtNDufJi$=Hp*y5WcqeuN`{tM|j5~zTeE9g+;bC&cmz1
z$yX><@XAutW1XJemD{V#)x$1@^QKk?U+L}%OU!3zspsf4R{L&)+r+$vT&PE3GqFNj
z^R<Q*NUmF4&CLZsMYu|SzUKTVp@_IXanW(*ph&y(%N;B@zqy#0Kck{3%{RDhx!IdM
z8pNF2cYrHwp~s=jk(Dfl{hQbklKmayxsf_v<wcD94e5SME?@P+$7O}$kXxV}g+c#V
zrH#C_pBCKsC0?UJn&HjW^L(LHI10&#IA&9Yq`RgZSboBin0?D+*4<iE5K>No#xTvU
zOdU~{JvMKSTKa=(Jp}Tg@&kx`UPRL}JXP5*6h1Q%&XTG8n<Q@`ZHUQfC~5k?wzPrg
z-+Xg`fEo^uAaBwF=I_$d9UQAGW)b+d^KAd%D?X5BZQpX&mWLWq4r>4D({B#;51xGU
zba}gw+&^u9&mE|=Hw;)-U^%&ZKX)i#+iP7LA-fQpxFqqwq<Bi5Sao(z?o>%+I_l_S
zS*n6YAe&26jld_tJQSgewWcfy9(^(0ZN+p#;vqC6^vC`z@3~cM;zl*UT}Xi;2`+uF
ztBg^+L4sIF$6Mq&Fjkw8CRUQ`ABSF7+FLJqM1Se2^~eKV9yiS;y;zWW*-L^ISEF-K
z{6)xh5c;IR3nQMI4Kz|Y8J6WQjZwvC>6G)3{*k)E9(K;nG+Zgj)v~&|^R$bx&*Lz>
zeo)c3ihYoqgPVrjJ?WlwIG<pv-ko|VM?=3Q+8^^Dp>xniKGXtq{}5$CL=_kEa2vV&
zamwx66I62|u3ZAp4BrLjv~Xxm8F}ERNI!Xx0myWNTCN}^N+qFEwW7mhOAx$$+A*&-
z!kBK`Au2W$CF1UzP@jpPj5XSIdVd`_HHd72W2$##aPB*D7;baVpsf0w@tse1ZTU-E
zL3mfxuY+*JsxacaHV@ke7;d_*M%2L5LLHJRk6<>Y0IXIQi^DP_NxT-lU428$LOc^)
zMi8@<DVy#n?Eys`z0NZv;tF{Eu3|cr9OU_hF*lfM2ckoB6($kC2+AacO6^0L6X%2n
zRwqRj3v;nZs_Ag60Zn@_C>8zDr<X%a0n@3fo(T#<wm6dk@2@NY(idx`!?ae<*s~O9
zRYEk5Sy(Wv6Dy76Orq28QKT!T8GEe_>XnKIt2bR1&cBO=jht&PRCQ0q2^Q)tASHzg
zIjZ=h@*ni8KLLh0rTK%V79suyyVXHEXq1lnd^FT)IKUfqqB;QbLg<gGD~ZY|S8+-K
zHnAMtP2ict>9uh$ZvN7>C$`B{gYyNG%`A6;3cznr4N$lB&aoYkK2(arl3@k4qVDLu
zg7|K+4WI^DPa@Rew%t<Bgq^O(FPnBT+j%dEEHJg2Xq+U_%T8<}#c0mPOw;$&vj=9H
zgw|^#A}uqVanE57Q@l$E%1gO4hHM4Tvb61%s(6R8@5w<L#sC`he_IK6Nyiyk9Tb^G
zL_V_SV!WQ$ljg=|YIRaF+swh`61_-*<)t?f$pm*Rnb7_m=J!YBi+Vy5Z`Q2k(m)x1
z6e{$)bHn}?^zr6k^h8QXD^#=#PoJCh@>zCaEfzU$wERy-ql<4YDfS5@?){i0WUr${
zRlMyTCE#sy`_uM)U4FFcrGhAmcmJnlLq8e)$xH{!-hGu!YqCcbb(h1GmR7f!Ti{)<
z5O}$Q6cmpvi9yq*fOfkW1(GJ%DXZ`=>-Oic##pa!32Q7;oc675F+4q91VI^Cc>yR^
z26>P^7Rx9&84hujhGEbfPudoAS|*qx_irjil@?2qd90)5)mv0>1;)~!iU=Y^z+lY0
zRZ+Ae(0QZeFr`!--+2Y)*=&kP8~z}14z6jA)X$mH8DSLUqoHa;3r~p#M%yWJ+&i++
zTBUx<Ho#{^TV<2hLL+iDq^*2-TF+A0!oucd(sE`vf?T+up1dBL5stD2^`3m!S{%~c
z>wWQ@3Nf<09Mh$TOqw7@ciZ?IChFI$X=_Y5^eT0D2&nlgBOvTHR>(Co6-H}FZ@hdj
z-eK|2Bu;aXP&e$Xy16AB<H>})W_DAo>kgagR3@N0rLouti9Jqmo%Wla<5}~O=~-#~
zP+ZH@S}}#-^)c>~YHWyE+A?8B$QomSpKBdBZ0r%aHbEy<!`!(u(*kg;I+xv89P>&Q
zrWi#I=P+LFD<sp29sdBGlwZUyGK6osH$`S@L>`^(0l3x#6W3#S?6QE>4@)p9IjN{9
zS?wcdXivnFB><gXg5)4Tw>L`4g{!1=VHeB!W{Z0y^;E)2e67uv$$yE8TRbvmZN*+$
zv4l8xPkxQ`vS}BHC!tb4kXTc%VWTolDavu;gu!A&f2eoSZe-FAkT)ZEfojR1j<V0)
zg2dA$b|V&}vnQLLf}SVMl2K;Y3u<yl%DF`fxs1A$wfb!bfgN}W*khG(_GsQ6e`5EW
zy9gQ!^$<6SlM-V=s*Pk6SAoC6IEgIt^{tFYe;5@=FA^_Zi_n)?lPbWN9hb}QNd;5L
zg~3u#V?%FC<^dH(LuTiE*#O7ME0k|d34H>mdQWK(q4w%DuF%i(RHb|b%z@^b?P`7@
zoN0YRG_^0Y2rYHlE6mI#FqJi7Hqn)9lQ0vh(8^>^T{gvrx9u}uPTe^j?hU#Z$35<`
z&`XN_snT)l2ZV!r_GBF~0zPgqhCt+6d0Qog?Z=i$;R~0bcdmBO4#G|CdrwIuNxyYe
zq4t<3_ISN*=z<e@1<e8J@(h2gt5dc;0+;|_yEgH`XgalO#&I$#-J8Ez51aA~EROYq
z{1!+r5F}Q(Vr`N-s%xi0b@VU=or`+ay=khyE8!?NFVt5Li3!@D5X@ky@twNVbtz~A
znmMYVdQYoptEv(v8(&;wAbqv_>CXM1evDu4wjZ39^?(near$|E%DSZREvGj|WF8kc
z5=Fi3<h(wiqJ1dqdt`v@mEYjOMW3I(%dKnq%Rd}-f*Q<#GO>~}E1Dwc#P}-oNYru)
z2h>gIuz!Pi&`QfQy(K`MEn}fFu1H)@C}&Oc;9|2@I7`;0LH$NLgFNZ*9G?vnF<PS|
zGFXwK$Jm?Z9#Gv&leiKRI(-f`_9XO>er0>4`hS9I_&(AW@*W@@Z?Z^=KGnBj4(XFZ
zvd!1&@HKZ9Io-Y5L9fT4<|XftI=ZLqht;ODYR05~S&tUl@~Wm}R$Q8`s;qK#h!^i)
zjeqms{onu22OsGF|6qw1Z|*`SD`wR6i8X#(Aw)-{Pe-+I4OF9L^Os+dW392`zlvL}
zF~XgzSlvPT^Bs2&(jA)H<y^FuEo3yd>jE1>ANmAwKxgoe71OIFpW*tMU}2rBhAENh
z62A%74sP&lqy=ItrvFr%Is<xC!J#O@Sq}CeWFK0gaW@!VykULio^zs_a{XdvZa#6-
zv~EDq)$=m$B#!IM-F!TT!m*q!_=~YeccY1}m}H#lHeT$4U($C(nvRYn)CH3O{yXRc
z^0h>EXQoLOSc&x-mpIXkmM*%vf&>I}bKlxyyjL1+u_cERVhUng)!<rr`lqiukN&v-
z?YCci3wxr(NAj>dYg%`r%CVd!NCY){`ob<m6&}kRZ8gk~#Sh#{!_d-Py?!4>yqj8r
zY*o7UeRRT93R_9K%@A66FmkmsS`ldG%FA<4$SuMg(lwEi?f<HG9O^nRs8_iEP`u-i
z@Y!v@HH#U}-F-+|gN3^fm3zkBhrC+2`w*GQ21tf3LxCsSN$wZaEmnmABLaZuD|4Pq
z1W~VJTTW4n;=BuU$X?iYx<XPxxae|eI)@Oz3!~u;ex<kFPs;n)S0QQuL*G!p++o)4
zI*~O*TL|v)Tv|AMbr_<#14rOuhstP=lb(P-!b$p@vqeT|jW_gCQA)T-*?+q%SAmLX
zh!Xh`5Q0gQ_jHtTwOy$guW(iOqMkx;5ck-$z{ib03cvFg#zB(^5)?MSpWf7GE2w}*
z&Y033z2>}D9#$9^2qHL{6Im;q;}LW7h<@}*W6CeG^QO>?J_5+jFHcY7<(yGU29l6@
zgM|a=?Lk4jPZuPUv>Iq32`yuVQU4T=cbe}5Y6!UArQaVEW$8~#!rj;`Q^o#Z^6@cM
zPhW8w7^eMibU6vzQAh}IfC%JUN25XuUi{H$(|#h!d7t=jS+!SduR5}BbQk!9Gm_Y4
zqq|z?G!8Hcg@LVrH2`eM@R<oPOdiu~pzHT=aOZ-ta37@|7uEzpmT$K~RX7|8mzOAi
zx(Y^ZT%?30+1+UoB&FhIavxPzX+SVIfGeLbhnVMeN0>RW%QgdvwmUmpH&B@Ml?&a-
z=}m3ZZf@zl>UPZH8z=3L&PUwi&^q)clzmjEB73f2-4%L2^8&Mej$1aJU0tuN?=$?~
z4AR~*;mIC01-o@LPTB9sNr==`^c?nn*4RM@eOJ7Z!|bWZX*Aje5Hh<}^UBP+$L)Ed
zNw~F)@C5xGFtm@5Q|QdGzLn|82G$9=on1RIVaf<Y*`#DXXdnkhV-s-30sPvORAS$&
z@<Vwy#R@cpbk=OL5glbhWoc~PQ60sn?x0Os4%A~6L>r|mulBRhCFDwCSL&B8m(ha8
zGb0Soa2p!KqMNi9P<RP_ab|`jYqC(L6Is(?)I?L<=#Di5&2DP<vCM$Z#1Jv|_C}ax
zB%#g}r<9u{uU1Cq1^1OjA3!vU_y+AXPl#%2EF|p|lUu%5Rm)UkosJqDJ-ap;ocZ|$
z#8Jbi$ha}F(d2AP@3ydO5r(%V64XAQT?{_V#Aq{^kv30<QuXQDE<j@rrg(E-3zB3y
zHVQu9&LxAztrL1h*yTz1UZyGs(_kwJh1=&*`rce8;Tver`w&qV$#nf62|Ig~lnGOm
z{Vf0Z&IIlqXqiZZn=9C5d0pm-nQ~w7HYcv6gbuZnojXB?9eoD4-@H4e;%v?Lj7@J7
z$Qbsz6B*@ouNuUi6gnC#a`W)Sa1oovDHqnebsDT<eo5O1uJIft<JHU|bq4oGN0DSR
zic?3fZ`As11{iLQNf>%M<u(@=_J3-fvlsaLb%SG$Q8J4^z-m`GU?;l3&OAQ{E-E#n
zXe_NR?BTXhU2~j$)R4VIR&8S9_^^JD1o4~&55GB_zFbvEM!WjsX!I(&jKH+Zd~so-
z4e|C9)n7*dZn(w8OBV~ndwc~KJXnxHiayBdc-q_f=#5O^GI?VWY?k(3)RXG@Y@(<^
zq9R9r3QOkav@7V|h%m;55G=y=ak8(bi{cP{dM#w#2i%Og!@v9$yw|I4Ql+JTCkyFn
zP>mX&<6BkvzlOHtg+e~skPTZO>d&ew<aW%3uCO5bu$$;avCr%h5YS65K9*b)8|;_1
zuiEE7>Haw*NsUD>qe`~e!#MeuhiS?$=|M!%)#Y1du_(8My~R$CiZ>0vt%p&wmPfCL
zYueSdmkrs-p8GDYy*ykCdsj9>0mu>mW-DC%-~Wq$@o!TRpa1{ErM5!d!;;I|y}cBE
zG6F`s;cMKTb+)=TZq&F-@cysLjyS&?e<fq${4jU6C>}`bIJ+cLo&adGenA)9e(UE@
zh(xBxn`eLV>iV)Gekpx?w^Qvudva$RSX7F0tV<yl#A~p#HWUnHX8a1w$JBhgB12ZI
zqa#GgudO7%m;rQIp`jBk*Cd@IDCKL|6fgbi3aDBF$e;IvF|sE^;~Wr0Ld>(7LA+TG
z26W#<QRHX9sD<l&Bsw}(gT$(J2Q5=Fz7nszlx(gC&ROkH%tet-${Cw$iVadr00Eta
z38<0?5}t^+DMNFHJYxZ6Att+iEC*;|?N#kHD>g|VG5jkrTsMt!0Dga_Q7AEFo3(i_
zv`(3mZ6Xv4=^-*&g0ZdUXyIr~dGJx~6PJAzcW7HqJfXf{xC_VERegQ^$gX!eh|m+R
zU{3IX?C8ORVg(W|189)BI5<UuN~W|%Lp8=6K+LFx8^_K)Ua#m-?Sfc3tY0e~5V?^b
zY+vI1WiZ?IfFZRv+$?I)VQkW#g`mN3TTMQ7g9Hg1!T2EG7W055hk2xIh3~dY6s4Yx
zR0d2%&196+Ga+|aEfwn>V}dH|oBZq|Q|@-B?6wS;@OWp>K7E01s&Dz91Ce(DXlB}l
zKW?U6AUGOy$U4tLV2+v*W=W88DNw;GQLGV#G60D>{*uTKXPAf{#`Eq396FI3z~b2a
z?13q-?+PLiOA!;0+k*IMX3h}Tp2RFPa+F_NeiB05JePjesHdw>)-Lkanz$t|jI>EP
zJxov6_1@0hf2v$2(Wm-zx@w!~la!R@Km9}+fo%hie#&3Cr9<D{2-aNihV@Qfdl53*
z>L>;(MurRb5_wAN2(WK(-`FmMo(yO~_6cE??kZSB+#s?o^%k_UkO;YdCr80{?sv|{
zF9(4BBidGW{X;4n>p!ekGF*MNQN3W|@!obB-TBouWs_bJUDbyAa~_Y;xDNiakev3J
zZU^>A82{7_F|9-MJ}|n9c<g8jhY$Hl|9@=%6ZY#Otm{~NtN-xlfB5sNKZ3kob*G^!
zp3h-@B_8?%6E~d{AqjTkvFPX6CLfKH@z}=RtA-E?Uo<Fxb2*c$=WuC2iDa(MGGzYa
zFb~8Y*<0^Y&{z&E2SecKMWDFA65DYfZX<0?m|(CSJ?T#CF{(h!jEjgPG_ID37q6an
zFSu*bYMNO~5qa;+h!IMh0*fL`2p9jv6t&LfP$8vb#_t^0&|`w;Cj{Oz<8M8K(Qq{0
zSnVW0p_!KiHefNQ-JN8MC`uUn_Zh{Dlc6up2aL$bdhD?aiTc~^_49HA1KYN^N^J5_
z{7|U;&ENYLL}sj2*mk%qfjXYnz?x!b+G9Mb#jVJ`^~P>9*)rpwVimCc5pp5Eq@PK`
zo9buLavMt`fttB*hwMd#(mIRZmvT(Rt*UYw>C^h`dPmONOjaR41*tPbZnC)1+GE09
zc45GDsrdX6+13P_Wt&LAO|f|94X_H@TY^7LJ#v0FNtQ|+=2%D{qlX7Iw{Xz8fiwzG
zidIYFlS@tNPq9{_?(+hax1b<Lg06+D=93Hte@h2OQ75y+lv*|#t_ZRb>DBHK@x$d7
zp0}Eq9@<FA;gA4(flWtz90VxuX@@SZlIu4ZfxkbMqlf(}Zbn!VaNe4qF633(Hs19<
z+a#kBzJsEX_ev6k?y!$d7v3M5W`E1XiRRiPn!cEOBx(mhSb4U-E#l_8P^`s~){1RJ
zr@r5~q`kxuYzgKbSIu?@{6I-`HMwJ{Vvf?-Vz<1x3c4};1t>8`B#1BZ=O{?ecZkR0
z(#Yd2T@1Kotj3XMD-AW>y4u;wDm-cXSa5|cs88Kq7pWgvkog;$OMMB6cTYfE03Tsl
zC_}_`%>;GbHPRi;FNb3Sq3oT5qQ|^L4OoFyOYr#CBh1X)*}#CeEhX5Zx8~NeWf7)Z
z9q+@ZZsEYfDOE~ItHfih#BpXC@RgsLNFYC{8TgEVEWh#zAczf6qAD@1^p}c%>HWQX
zCQ7Co>tnB`+DY$=JPHBbu8Ma4zFFjI5mn9QbEZfH;oi5;=Jo$JcQ(I~W>*=nAq1or
zLP%th4UsW|R6FfzCOr{KOC9ndI&DRWf$?M*wMO;Y?rPiMahWQ6Ivy!VEtZKQmh9NH
zgv2t6ltqNZ3W*hBh1jq|Vu_SLfc&24+^_eos<smmgfo@4tKPaF=bn4+InVjHbPKav
z-!o<HK;0#iP*Hjefboa{WVJBXxDL{>s+7i}D``TSZWU8^^9VJbppDXR2~}g29f}w7
zPu3|MsX3xy52T>p4U5ia6}mk>Q+YHLAtWpN;_8s{XN>5hUFtIWK$*`8Z-pP~g<6oE
z|1AAgCOaG_wpINeQ!j#tr+75n@SGA#pUa>l>_`m^h1|n;+}aK;eF%b5>M}I?hDHis
z;>bj_MoeJ$_0!8ISBKh*^VjlQ+9dw+lKn6YqWjYIxBuyXJ~JBW^&4w|v5IJ{D}!;T
za>wJZ{Ba-U`hSFHHIw+c09j4@w??oU(utAx9bH)xS*#1<IFy)0OPT|gvvaKG+*Dv;
zUsh=j5UZsj8ozv9G;L{#$K;-s^(yrYuNK-X0F$^6%ibsL&dy%Sm!`1WGw!fEnjtmb
zJ*S?9!5?X2!rru5EMj}-0AipCNi8EVT;t~1d6cVSucw{pI98uF6if4dGTbNkE%W0)
ziX$AHl52vnezB``xijcQ9-dqCc9RoC;ha<ewCS!5o|{DvCy&(byoVa@uv<!R@I85h
zQhO2@_$w%V<=P8n1`-L}PjY&sTm2Y~vX$7S{?;a7g4`jM<J!AKBwvg8N^y|($Y7se
z5mOAU>{iv{djc_tMO=-*mGlZ~lXFG#?KG|umk(2SCAS)zwJtk|o^kPC#AitHlRY9C
zSZ1*X=_JrC61E0KGip>&(pnRmSvLW;ak7EllhiYNe=`_-`TKmd#+ha@Tt?%sd^}ss
zfz~dN{=>ZiN2d3N25YuvRZ>tmL(gX2(_7>C=2{KS-+_Mi7RP{#cIMBVT{*p=yaa^`
zXxjs8-YgS;SL|aFpUic!FvN!2*dgLS7L}9f#tXz@ET?~Y+3>}KiXDMbgtHqnj<5o<
zUcjMp%mJho=V)O=Q?Y~Fq+QclwhEUWQs+WdQLu2NktrLX&*OpZ?QF>6Dh;LlK8A7g
zqu*4K8zp_;{AdqqxZPKjYZQj9IZ(>co_2VG<uapfV8(_&2YeZdD+2TR3~^%!$xzs0
zcI+7BV2CBjgkJES9hMv+ncl`4Y|Gzd{U}i%uHbN|Zl7a$a}0FLi(TMuhz=0Cu~2De
zrYp58Nt(%y-Xy8<fbWISbCium0IlbVtFVPt)mtG1A&;CgdOU{*4^W($ik;@16VtNu
zxp_txg&Tn2fhPc=5Pc6e#s(6K!1VGEZb|#NAYCk78u6ri0WtXxuht)+pp{+{impHv
zP7DYYOA*&fX?w6xh^VYRUvV3QHYc0xV8HE@eQ)+o(r|-hdCCvi)|&#mFv@=S1AeG?
z&$Y3a;p9i?WV*c*^^LTh<6@2+ypl=C4!p1nrZ;nN5~YfJIwDHfJ=1oLl1yl;#Hi_a
zo#|b-AUzDRkAro=J*(O|#k;Q+bf_>W*XG#_CWF1I&K?C%rbuII&9>b<+FxAF@}!#{
znyF)N>84ik$-M!Q<3yPhhFa#7VNED+iGD(El|J6|Udlow`+4&B%7K;}&1WLV$wn&F
zFZpKTT7Bp~DYA?CIfY^-e|B-Kx+Bd(2cj+S3&hQ#r)BBoa>?6(D9s>Wp$Qp8pcNRb
zV$Q~MuQS8<;=-Zve%TRZi}o-ZddidDM6AA0&4W-JmiUc+ROs4f838atW}4h&pdW6I
zEfF349Oa<oxa-(c8N|gA1LNW+TxJhuN|lN-BGTzdykOL2;yNl5VLV$Mb~7n^#S-*L
z`z{iAI^i6>W4|}mfU2K9b@2giv3&3r!5MFT@eA6sVs~Et=YtP==yb!8O>^L6P-i0K
zdKI{iwKt&rbocWQRXZR!B*w^wNFp>u91^^;vS3;Xcpk~JZOy+Hckd^6q<LNeFIvI1
zH!$~`uLXg`P3jPl7t%4MpVE1fgqc~Z@|_e_7jc#LoSk<8?!u^1Mz8Qlxp^A-0VH&C
zDov&&6W|HL(seWv>1@e{C@W30lDhC|`vT_(<H{q}ToXc<TvElZzauwgWe2NtF}<-C
z`t!VW|KnTKoEkjZr_;E$gQpC?*Q30Cs|a8FXtSCP#{jQpxUW7uX|{^g%IKhdx9|!Z
zrEU_#thE_=sbd2h@L<n&T{NX5K^ZV2<I();OkrLwIs+iaW9S@|FJtdXWsNFS4mD^>
z-ZW?YIwwLXh<jlB3VqA$;yzWcMIVrOmCH5A5qEZhTXmf!&?sLkznbB{+LYr8RfmIw
z(-Mu^P}ak4lg!KGLPaur%AsMyFd^p9K>&Rm^c)I#r5X%3=!rgGnPjlKwE51->GR99
ziV8M+5SO>uy`#eaTTgsIzt`uxXqt-dIJj<Lt#$T^jN4|Ql`U=<Y}VrVe&fWqkzJ5o
zLyp{4iUmpFFB%?;nKaI#Z)y>WJ4NE#s3_D(rl$QJyPRP?qnNMPeJ+R~+cJo=i01PG
zB@Z+xNHka-8Ik~IcWEY}%!$h**sXXJw#HuBu1iW>(wrF@Zza=|L#nE^UuC!Kfq-CP
zxBLvuD7IZAi}PDOnN0Rt60Gh%N!Ni;6^B^1DH730BU8$QqO)?!&q@4&LOdRT^G_}4
zY0UMk+Yrle%t^NxngUr7?Fby%$f=dh`lhxIu=?TulH_CpX4kKr8HtJNtHCDH;BR^Q
z0g_B($4{+0)2fBwG42JFFX;XJ^q6?qOG8=NqJF_ew&21GUwIPVEJ;qDzMOqaK+ZX;
zo{F{*l#UeC1(1q0mq2v{c?622_;IA+!R1nfwME)UlfmG|+nE!#d0Km^5*3O!GUEo3
z@j7{Bn!a~(9mfY}NSw#;qKvf6+5Lab=Q}%SK)P@oB0&H=Ld>1&nWVK&ZZD1LBZY7V
zR5V0;h+e=PQvo*{wg?*7^`rTjFWOWf*H0_=L+Bq1NLl2#51Q$wQ(HFEEQ6*~-Yw3Y
zO<4pIoXh+00J-vzED{RJpVZ<de=ZzRFpsWTI?HO*dAj&`Yx$u(Eei2Ffp@ZggyYNt
z2o5^<XilEIwXN)U7MNfut!xZs@u7+802jaBM%Ly!9Q4B8p|R-08dVyIR7nNQNOeF{
z2+hGhZf1h-`$LHhq^*LsDFFOgou^IwTR6XWcJ`5H`F;?c98vM<!z)xkYI|tREH0^Q
z6Oa<|lk2@Si|tS;nkKz6U$1J>ck9#MM}w*{X@E=Ei#d5}<~Pe;1ON=&uviX_4?!WZ
z+|v%2M2v`P33B8-G~J8OoITWajvq-tO{yk%vyQSSi?P$b!|N&zNC}5&^0j#4%}S1@
zjfiqJKoJZQ-xF~a<LTwHn-e(&&GmUwq(f)tvo_l)JXfj$b)^D8hci(`PwLdYY5&EX
z0+m>zVabdGbs@Eby+dm@4n)P^m?m(B5BX`85tt3)R=-*VbE)icRN|h^99{!H_e}LG
zPz#GvqDmcftgll<c_C$qInbU%K%%Q;zg+XkF)4Y%D6Cqh*y6;m39$>_=hTVv8pVr;
z5QMko_GIhg&f_kc=+|Dj6r7Wjla9gTuMQ_fJ>Jj0;hxX#{_j0_^NW_Yac!_h#8|5|
z_GkO{t4i&{Ei!w7|3_i29zA35q_?Z$zP+`$f@f*wp#mea#+)Fo$q}ad43lB)hnI8T
z6_P#+C@8e5iTt-mCkcL9`{f=~H0?Mp@c2sji2w3QXQm2P*gJQ{t||<?82Q0e_N(hq
z=O;i)Q)8l57|V81!)8*a5BgELUVtVgW-JHrfyXYNJwpN+k`QYHp`RUaq?J=Ad67Z@
z!p0i5&#KqqN2`z^hQH46z)2n@h{&UV^X`OdkJB+%0Md3m0N`3RWMh-NF)SfsVr2)k
zHae;|TgdFvjarkklGaIZ3mi_kRG<acpY8C9LVJdu)IP+!mRk>m>fr32wN#p_i8txk
z<{F)qU>A?RXL@*G83XahOmp>%Wzc-NwR3DTtub;oc9fd7{qBckT1USVgd+8s(i5e}
z#*tH=ZQdKCl>BJ+t9$R0PNLWAQc9}kb#cADz1??`Z-}SckW-QwKlG*BwIok&vh6HO
z@p_^_bx!d+tlABK9~@3}!X7B!|9CzpxXV&01k_MCIez8%WcPyOY8gQ}<ngH3;-&7b
zytqC;oT}SHGQyVkRsMAv!66k%SpIx&epv11);@V3G?7)pe!))Tz+=xzrRuw5&5Wv|
z3kE#n21KqFSg0<sf|DfCa2>1aL(WY}sVR2`DMJ0_jPo%aG_D<Koht{!hwn<_ENgeY
zTN`@mMw6|SXD;jH^!$p0u8k&@f$W1$mdc&5`>2OVHN>;U6hM1EO^VF2MpG@PUc<&x
zaZa-s8j*knk-fm$vn$Cj?7O&jp=^egM^rU!O>vlH@j*+=^^zfj@J58$51xHIa^zT<
zJ3Z$#lmQm5YLN`jIK#Y2QOOI^MU!X_kvt&R{EQf2g^0^3OXsxahzij%#}q{DYU`a(
z%CST!+iUKu0rZ7sC=5k2Q3EunwpkKjckIeDq!R+z&M7i~(BuS;?@Bf!8J&2Ijuz>b
zW}03_;7!_L53EYl)Y;IPl8PwsNQ%+3Svwg7uqB<P^2hz@VG0rFmv2Qx8~94eT*$N@
z*RM?Sd&$%_NJ*i->*+JqZ}`LO{P|e!HBy!zfZh7jf{&_PhKYviY!)Z=f$fNE6~!?{
zr{z|H#tu-loyB%z1vXai&)Y7Jr4;*6nKrhs_GdO9!4jN9_FJ}UG%EL%xusB?)sAc6
z6E2?Y>FkSvc|s)@hvYqCFKg-YoZwYhVLJsl4hE6wk&Vfuz0#Z%)3vUG+|80mmA0e-
zer_uH&lcpF>fU;aODlMs%{k19!9Y|EU#LlRjte^!&Q5ccP}M*gdA`uiccX37U{Sd=
zbFExm`n+gTPx5HiPusMAotR;6SuQn8h9&qIHvRHr4#VDL$MvB{+rC?kdXR;ADQ>iU
zREK%jNwt1z3Wt`*#%9-hTLcEE#N*D+pA%(uEF?&WbNGk0RmfUF`oEd#*vOKgPNlZN
zMAc!(yKhFiOQ@}N5KCVIG3NOMxcNJ%S<uuV&%VN0b6^wfcD;W2i<EYIb}JyY04A2V
z4;nzY9aAkSG|a9WOc>l(41jaM$xunUr&La@XQWXTSkexPDPK|6Bcw^3S9Kflp?>xv
zK2lh!cC(z|y;Y=R1TiQC0zz|Q^KpjLV){P##nri!C=Sl@G7F?u*fX}iei>{p8p=CM
zn^0=QT!(*ZFi;bF=PZ!ishViz#15wwnH)QKj3=^!J(aHe@?h{S3qGEhyY6uz!30lX
z?8FuEmm5eDF!r9skO0%5(u&A(914Tzy2{YdAP2Hz?wGIoA~}?G1jn=S1BCMz@U8NH
zxW%VDmj)Z$Z|f#-TsAS}(@<LXC0(ySse5ASPv~m?VLuwb@g~5mE~g7GfjH>=l!r)*
zo6X%EY|v}9u71#0(>B;l81lnADa(Cb*RDM8$zBepxyq$TOeQUBCNEGu1gvdVNuLE?
zqLW=Q^-GCW8ku^4P^ps;5WCL;qgH^+CSn5?))s0{sjbjF&<Qd^s8TTH;6cR&CoUur
zZEaB4r@%?W_B+q4Ux#4`m(83@1IJ|g<4E}=y{Y(`(so?I<`8r`lnCL1e7iU@tE`A)
z5=;F``<!j(G@P7fN;=ggGcX3UNC&Y1mKEpY&-O+}XWHG)$rMTo>Y>bQi4BcRl0(W^
z3O-L3IC2Lu%7|1n@<``}V2V6^L61=L_=<_MDiXJ-jFK<Krwivh^P1Rzx+0q5i@9)P
zblWISkCL?DR~6}P>W0hZfkpyI2c~jly<_X>`UvB&f+#OEyr}A02Lc<0xNb&kV#j(g
zXt9#2nqh#qqb-0?tW+%d9=IrqTucv60_KW!L>E_n!@XW26|D;IP{ZJ`0puXT3P~8Q
zNR5f`vq*$W#jL^8UXFms3TLaJKu=o&iil49eXxwW8=F*wNXI>&|HZ-ojb;GH#Cae_
z?7F_dhIF>RSW=37?{G|I+0b>KKz`s7qs$pTQr*pk8<#)}V4~zRDFrCxJqSP9uOZS#
z_IwUUcx=s)eG)>=rx~nCsT<+bG@HD|u@Y@3LhzjSwoV^`sUQNu)XURjeB1zhFsht*
zaGh0#<hC)4-i(|9nmED*!LU7mRAc~$LXWqDdmGM=J`*eJ+#m1Hj!w>_vJc%8Ca2}d
zY9elHCI8fG8^S2XZ~gW+-uQ)2ed1FjaEwN(ck;Er`Pc8fyEPi=#U-cZ=haL78rQ7^
zDl-QrMh}=0tM>vct!?OthMu3F{?$7qhUoP>rJ2#_)ywZ~^&27;#di}j;%SpjHfZx1
zZT{VV|Lx!Z!>^4-dfn}{`P*OWw;2_*0h9u0ZQzJOvoo4~|6gAI;6HXpBfUP|YxeFR
z^_%q#o701f(?KgIwDO;?e)!kF__fhUukBtdkG?%@WeF^L`mtr=G_DBf6L6pQW~bff
zK06+b^kQJ;=he%X|2P_zb2VXP^>0_j5iLnVSYVY8hN`BY{N*oy|KBDQwc+)-(y-R%
z$N%-(ri-;E2f?7>b@$`uwqL#c^-sU?x-rjuTW-rk>k4((=KlS`YICXlnF}lq#<s4K
zPi^dzPu4ZIT4H86ykQDdZSgZtS6eKm+>me@+z=D4P5tS|t4*y#D&PLbCx1!b`W627
O&z~5Le*E1x`0FPy-zC-n

literal 0
HcmV?d00001

diff --git a/0.16/_downloads/163bc3a8b6281b990b458e3eef431d01/plot_custom_transforms.py b/0.16/_downloads/163bc3a8b6281b990b458e3eef431d01/plot_custom_transforms.py
new file mode 100644
index 00000000000..898c2cd0bea
--- /dev/null
+++ b/0.16/_downloads/163bc3a8b6281b990b458e3eef431d01/plot_custom_transforms.py
@@ -0,0 +1,121 @@
+"""
+===================================
+How to write your own v2 transforms
+===================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_transforms.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_transforms.py>` to download the full example code.
+
+This guide explains how to write transforms that are compatible with the
+torchvision transforms V2 API.
+"""
+
+# %%
+import torch
+from torchvision import tv_tensors
+from torchvision.transforms import v2
+
+
+# %%
+# Just create a ``nn.Module`` and override the ``forward`` method
+# ===============================================================
+#
+# In most cases, this is all you're going to need, as long as you already know
+# the structure of the input that your transform will expect. For example if
+# you're just doing image classification, your transform will typically accept a
+# single image as input, or a ``(img, label)`` input. So you can just hard-code
+# your ``forward`` method to accept just that, e.g.
+#
+# .. code:: python
+#
+#     class MyCustomTransform(torch.nn.Module):
+#         def forward(self, img, label):
+#             # Do some transformations
+#             return new_img, new_label
+#
+# .. note::
+#
+#     This means that if you have a custom transform that is already compatible
+#     with the V1 transforms (those in ``torchvision.transforms``), it will
+#     still work with the V2 transforms without any change!
+#
+# We will illustrate this more completely below with a typical detection case,
+# where our samples are just images, bounding boxes and labels:
+
+class MyCustomTransform(torch.nn.Module):
+    def forward(self, img, bboxes, label):  # we assume inputs are always structured like this
+        print(
+            f"I'm transforming an image of shape {img.shape} "
+            f"with bboxes = {bboxes}\n{label = }"
+        )
+        # Do some transformations. Here, we're just passing though the input
+        return img, bboxes, label
+
+
+transforms = v2.Compose([
+    MyCustomTransform(),
+    v2.RandomResizedCrop((224, 224), antialias=True),
+    v2.RandomHorizontalFlip(p=1),
+    v2.Normalize(mean=[0, 0, 0], std=[1, 1, 1])
+])
+
+H, W = 256, 256
+img = torch.rand(3, H, W)
+bboxes = tv_tensors.BoundingBoxes(
+    torch.tensor([[0, 10, 10, 20], [50, 50, 70, 70]]),
+    format="XYXY",
+    canvas_size=(H, W)
+)
+label = 3
+
+out_img, out_bboxes, out_label = transforms(img, bboxes, label)
+# %%
+print(f"Output image shape: {out_img.shape}\nout_bboxes = {out_bboxes}\n{out_label = }")
+# %%
+# .. note::
+#     While working with TVTensor classes in your code, make sure to
+#     familiarize yourself with this section:
+#     :ref:`tv_tensor_unwrapping_behaviour`
+#
+# Supporting arbitrary input structures
+# =====================================
+#
+# In the section above, we have assumed that you already know the structure of
+# your inputs and that you're OK with hard-coding this expected structure in
+# your code. If you want your custom transforms to be as flexible as possible,
+# this can be a bit limiting.
+#
+# A key feature of the builtin Torchvision V2 transforms is that they can accept
+# arbitrary input structure and return the same structure as output (with
+# transformed entries). For example, transforms can accept a single image, or a
+# tuple of ``(img, label)``, or an arbitrary nested dictionary as input:
+
+structured_input = {
+    "img": img,
+    "annotations": (bboxes, label),
+    "something_that_will_be_ignored": (1, "hello")
+}
+structured_output = v2.RandomHorizontalFlip(p=1)(structured_input)
+
+assert isinstance(structured_output, dict)
+assert structured_output["something_that_will_be_ignored"] == (1, "hello")
+print(f"The transformed bboxes are:\n{structured_output['annotations'][0]}")
+
+# %%
+# If you want to reproduce this behavior in your own transform, we invite you to
+# look at our `code
+# <https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/_transform.py>`_
+# and adapt it to your needs.
+#
+# In brief, the core logic is to unpack the input into a flat list using `pytree
+# <https://github.com/pytorch/pytorch/blob/main/torch/utils/_pytree.py>`_, and
+# then transform only the entries that can be transformed (the decision is made
+# based on the **class** of the entries, as all TVTensors are
+# tensor-subclasses) plus some custom logic that is out of score here - check the
+# code for details. The (potentially transformed) entries are then repacked and
+# returned, in the same structure as the input.
+#
+# We do not provide public dev-facing tools to achieve that at this time, but if
+# this is something that would be valuable to you, please let us know by opening
+# an issue on our `GitHub repo <https://github.com/pytorch/vision/issues>`_.
diff --git a/0.16/_downloads/175e3bb1b03a25b015d65842a3b8a883/plot_video_api.py b/0.16/_downloads/175e3bb1b03a25b015d65842a3b8a883/plot_video_api.py
new file mode 100644
index 00000000000..ac9eb0ba27d
--- /dev/null
+++ b/0.16/_downloads/175e3bb1b03a25b015d65842a3b8a883/plot_video_api.py
@@ -0,0 +1,346 @@
+"""
+=========
+Video API
+=========
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_video_api.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_video_api.py>` to download the full example code.
+
+This example illustrates some of the APIs that torchvision offers for
+videos, together with the examples on how to build datasets and more.
+"""
+
+# %%
+# 1. Introduction: building a new video object and examining the properties
+# -------------------------------------------------------------------------
+# First we select a video to test the object out. For the sake of argument
+# we're using one from kinetics400 dataset.
+# To create it, we need to define the path and the stream we want to use.
+
+# %%
+# Chosen video statistics:
+#
+# - WUzgd7C1pWA.mp4
+#     - source:
+#         - kinetics-400
+#     - video:
+#         - H-264
+#         - MPEG-4 AVC (part 10) (avc1)
+#         - fps: 29.97
+#     - audio:
+#         - MPEG AAC audio (mp4a)
+#         - sample rate: 48K Hz
+#
+
+import torch
+import torchvision
+from torchvision.datasets.utils import download_url
+torchvision.set_video_backend("video_reader")
+
+# Download the sample video
+download_url(
+    "https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true",
+    ".",
+    "WUzgd7C1pWA.mp4"
+)
+video_path = "./WUzgd7C1pWA.mp4"
+
+# %%
+# Streams are defined in a similar fashion as torch devices. We encode them as strings in a form
+# of ``stream_type:stream_id`` where ``stream_type`` is a string and ``stream_id`` a long int.
+# The constructor accepts passing a ``stream_type`` only, in which case the stream is auto-discovered.
+# Firstly, let's get the metadata for our particular video:
+
+stream = "video"
+video = torchvision.io.VideoReader(video_path, stream)
+video.get_metadata()
+
+# %%
+# Here we can see that video has two streams - a video and an audio stream.
+# Currently available stream types include ['video', 'audio'].
+# Each descriptor consists of two parts: stream type (e.g. 'video') and a unique stream id
+# (which are determined by video encoding).
+# In this way, if the video container contains multiple streams of the same type,
+# users can access the one they want.
+# If only stream type is passed, the decoder auto-detects first stream of that type and returns it.
+
+# %%
+# Let's read all the frames from the video stream. By default, the return value of
+# ``next(video_reader)`` is a dict containing the following fields.
+#
+# The return fields are:
+#
+# - ``data``: containing a torch.tensor
+# - ``pts``: containing a float timestamp of this particular frame
+
+metadata = video.get_metadata()
+video.set_current_stream("audio")
+
+frames = []  # we are going to save the frames here.
+ptss = []  # pts is a presentation timestamp in seconds (float) of each frame
+for frame in video:
+    frames.append(frame['data'])
+    ptss.append(frame['pts'])
+
+print("PTS for first five frames ", ptss[:5])
+print("Total number of frames: ", len(frames))
+approx_nf = metadata['audio']['duration'][0] * metadata['audio']['framerate'][0]
+print("Approx total number of datapoints we can expect: ", approx_nf)
+print("Read data size: ", frames[0].size(0) * len(frames))
+
+# %%
+# But what if we only want to read certain time segment of the video?
+# That can be done easily using the combination of our ``seek`` function, and the fact that each call
+# to next returns the presentation timestamp of the returned frame in seconds.
+#
+# Given that our implementation relies on python iterators,
+# we can leverage itertools to simplify the process and make it more pythonic.
+#
+# For example, if we wanted to read ten frames from second second:
+
+
+import itertools
+video.set_current_stream("video")
+
+frames = []  # we are going to save the frames here.
+
+# We seek into a second second of the video and use islice to get 10 frames since
+for frame, pts in itertools.islice(video.seek(2), 10):
+    frames.append(frame)
+
+print("Total number of frames: ", len(frames))
+
+# %%
+# Or if we wanted to read from 2nd to 5th second,
+# We seek into a second second of the video,
+# then we utilize the itertools takewhile to get the
+# correct number of frames:
+
+video.set_current_stream("video")
+frames = []  # we are going to save the frames here.
+video = video.seek(2)
+
+for frame in itertools.takewhile(lambda x: x['pts'] <= 5, video):
+    frames.append(frame['data'])
+
+print("Total number of frames: ", len(frames))
+approx_nf = (5 - 2) * video.get_metadata()['video']['fps'][0]
+print("We can expect approx: ", approx_nf)
+print("Tensor size: ", frames[0].size())
+
+# %%
+# 2. Building a sample read_video function
+# ----------------------------------------------------------------------------------------
+# We can utilize the methods above to build the read video function that follows
+# the same API to the existing ``read_video`` function.
+
+
+def example_read_video(video_object, start=0, end=None, read_video=True, read_audio=True):
+    if end is None:
+        end = float("inf")
+    if end < start:
+        raise ValueError(
+            "end time should be larger than start time, got "
+            f"start time={start} and end time={end}"
+        )
+
+    video_frames = torch.empty(0)
+    video_pts = []
+    if read_video:
+        video_object.set_current_stream("video")
+        frames = []
+        for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):
+            frames.append(frame['data'])
+            video_pts.append(frame['pts'])
+        if len(frames) > 0:
+            video_frames = torch.stack(frames, 0)
+
+    audio_frames = torch.empty(0)
+    audio_pts = []
+    if read_audio:
+        video_object.set_current_stream("audio")
+        frames = []
+        for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):
+            frames.append(frame['data'])
+            audio_pts.append(frame['pts'])
+        if len(frames) > 0:
+            audio_frames = torch.cat(frames, 0)
+
+    return video_frames, audio_frames, (video_pts, audio_pts), video_object.get_metadata()
+
+
+# Total number of frames should be 327 for video and 523264 datapoints for audio
+vf, af, info, meta = example_read_video(video)
+print(vf.size(), af.size())
+
+# %%
+# 3. Building an example randomly sampled dataset (can be applied to training dataset of kinetics400)
+# -------------------------------------------------------------------------------------------------------
+# Cool, so now we can use the same principle to make the sample dataset.
+# We suggest trying out iterable dataset for this purpose.
+# Here, we are going to build an example dataset that reads randomly selected 10 frames of video.
+
+# %%
+# Make sample dataset
+import os
+os.makedirs("./dataset", exist_ok=True)
+os.makedirs("./dataset/1", exist_ok=True)
+os.makedirs("./dataset/2", exist_ok=True)
+
+# %%
+# Download the videos
+from torchvision.datasets.utils import download_url
+download_url(
+    "https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true",
+    "./dataset/1", "WUzgd7C1pWA.mp4"
+)
+download_url(
+    "https://github.com/pytorch/vision/blob/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi?raw=true",
+    "./dataset/1",
+    "RATRACE_wave_f_nm_np1_fr_goo_37.avi"
+)
+download_url(
+    "https://github.com/pytorch/vision/blob/main/test/assets/videos/SOX5yA1l24A.mp4?raw=true",
+    "./dataset/2",
+    "SOX5yA1l24A.mp4"
+)
+download_url(
+    "https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi?raw=true",
+    "./dataset/2",
+    "v_SoccerJuggling_g23_c01.avi"
+)
+download_url(
+    "https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi?raw=true",
+    "./dataset/2",
+    "v_SoccerJuggling_g24_c01.avi"
+)
+
+# %%
+# Housekeeping and utilities
+import os
+import random
+
+from torchvision.datasets.folder import make_dataset
+from torchvision import transforms as t
+
+
+def _find_classes(dir):
+    classes = [d.name for d in os.scandir(dir) if d.is_dir()]
+    classes.sort()
+    class_to_idx = {cls_name: i for i, cls_name in enumerate(classes)}
+    return classes, class_to_idx
+
+
+def get_samples(root, extensions=(".mp4", ".avi")):
+    _, class_to_idx = _find_classes(root)
+    return make_dataset(root, class_to_idx, extensions=extensions)
+
+# %%
+# We are going to define the dataset and some basic arguments.
+# We assume the structure of the FolderDataset, and add the following parameters:
+#
+# - ``clip_len``: length of a clip in frames
+# - ``frame_transform``: transform for every frame individually
+# - ``video_transform``: transform on a video sequence
+#
+# .. note::
+#   We actually add epoch size as using :func:`~torch.utils.data.IterableDataset`
+#   class allows us to naturally oversample clips or images from each video if needed.
+
+
+class RandomDataset(torch.utils.data.IterableDataset):
+    def __init__(self, root, epoch_size=None, frame_transform=None, video_transform=None, clip_len=16):
+        super(RandomDataset).__init__()
+
+        self.samples = get_samples(root)
+
+        # Allow for temporal jittering
+        if epoch_size is None:
+            epoch_size = len(self.samples)
+        self.epoch_size = epoch_size
+
+        self.clip_len = clip_len
+        self.frame_transform = frame_transform
+        self.video_transform = video_transform
+
+    def __iter__(self):
+        for i in range(self.epoch_size):
+            # Get random sample
+            path, target = random.choice(self.samples)
+            # Get video object
+            vid = torchvision.io.VideoReader(path, "video")
+            metadata = vid.get_metadata()
+            video_frames = []  # video frame buffer
+
+            # Seek and return frames
+            max_seek = metadata["video"]['duration'][0] - (self.clip_len / metadata["video"]['fps'][0])
+            start = random.uniform(0., max_seek)
+            for frame in itertools.islice(vid.seek(start), self.clip_len):
+                video_frames.append(self.frame_transform(frame['data']))
+                current_pts = frame['pts']
+            # Stack it into a tensor
+            video = torch.stack(video_frames, 0)
+            if self.video_transform:
+                video = self.video_transform(video)
+            output = {
+                'path': path,
+                'video': video,
+                'target': target,
+                'start': start,
+                'end': current_pts}
+            yield output
+
+# %%
+# Given a path of videos in a folder structure, i.e:
+#
+# - dataset
+#     - class 1
+#         - file 0
+#         - file 1
+#         - ...
+#     - class 2
+#         - file 0
+#         - file 1
+#         - ...
+#     - ...
+#
+# We can generate a dataloader and test the dataset.
+
+
+transforms = [t.Resize((112, 112))]
+frame_transform = t.Compose(transforms)
+
+dataset = RandomDataset("./dataset", epoch_size=None, frame_transform=frame_transform)
+
+# %%
+from torch.utils.data import DataLoader
+loader = DataLoader(dataset, batch_size=12)
+data = {"video": [], 'start': [], 'end': [], 'tensorsize': []}
+for batch in loader:
+    for i in range(len(batch['path'])):
+        data['video'].append(batch['path'][i])
+        data['start'].append(batch['start'][i].item())
+        data['end'].append(batch['end'][i].item())
+        data['tensorsize'].append(batch['video'][i].size())
+print(data)
+
+# %%
+# 4. Data Visualization
+# ----------------------------------
+# Example of visualized video
+
+import matplotlib.pyplot as plt
+
+plt.figure(figsize=(12, 12))
+for i in range(16):
+    plt.subplot(4, 4, i + 1)
+    plt.imshow(batch["video"][0, i, ...].permute(1, 2, 0))
+    plt.axis("off")
+
+# %%
+# Cleanup the video and dataset:
+import os
+import shutil
+os.remove("./WUzgd7C1pWA.mp4")
+shutil.rmtree("./dataset")
diff --git a/0.16/_downloads/1c7d3f45724a1171e328ce5cdd7bc79a/plot_scripted_tensor_transforms.py b/0.16/_downloads/1c7d3f45724a1171e328ce5cdd7bc79a/plot_scripted_tensor_transforms.py
new file mode 100644
index 00000000000..128ce7778f3
--- /dev/null
+++ b/0.16/_downloads/1c7d3f45724a1171e328ce5cdd7bc79a/plot_scripted_tensor_transforms.py
@@ -0,0 +1,136 @@
+"""
+===================
+Torchscript support
+===================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_scripted_tensor_transforms.py>` to download the full example code.
+
+This example illustrates `torchscript
+<https://pytorch.org/docs/stable/jit.html>`_ support of the torchvision
+:ref:`transforms <transforms>` on Tensor images.
+"""
+
+# %%
+from pathlib import Path
+
+import matplotlib.pyplot as plt
+
+import torch
+import torch.nn as nn
+
+import torchvision.transforms as v1
+from torchvision.io import read_image
+
+plt.rcParams["savefig.bbox"] = 'tight'
+torch.manual_seed(1)
+
+# If you're trying to run that on collab, you can download the assets and the
+# helpers from https://github.com/pytorch/vision/tree/main/gallery/
+import sys
+sys.path += ["../transforms"]
+from helpers import plot
+ASSETS_PATH = Path('../assets')
+
+
+# %%
+# Most transforms support torchscript. For composing transforms, we use
+# :class:`torch.nn.Sequential` instead of
+# :class:`~torchvision.transforms.v2.Compose`:
+
+dog1 = read_image(str(ASSETS_PATH / 'dog1.jpg'))
+dog2 = read_image(str(ASSETS_PATH / 'dog2.jpg'))
+
+transforms = torch.nn.Sequential(
+    v1.RandomCrop(224),
+    v1.RandomHorizontalFlip(p=0.3),
+)
+
+scripted_transforms = torch.jit.script(transforms)
+
+plot([dog1, scripted_transforms(dog1), dog2, scripted_transforms(dog2)])
+
+
+# %%
+# .. warning::
+#
+#     Above we have used transforms from the ``torchvision.transforms``
+#     namespace, i.e. the "v1" transforms. The v2 transforms from the
+#     ``torchvision.transforms.v2`` namespace are the :ref:`recommended
+#     <v1_or_v2>` way to use transforms in your code.
+#
+#     The v2 transforms also support torchscript, but if you call
+#     ``torch.jit.script()`` on a v2 **class** transform, you'll actually end up
+#     with its (scripted) v1 equivalent.  This may lead to slightly different
+#     results between the scripted and eager executions due to implementation
+#     differences between v1 and v2.
+#
+#     If you really need torchscript support for the v2 transforms, **we
+#     recommend scripting the functionals** from the
+#     ``torchvision.transforms.v2.functional`` namespace to avoid surprises.
+#
+# Below we now show how to combine image transformations and a model forward
+# pass, while using ``torch.jit.script`` to obtain a single scripted module.
+#
+# Let's define a ``Predictor`` module that transforms the input tensor and then
+# applies an ImageNet model on it.
+
+from torchvision.models import resnet18, ResNet18_Weights
+
+
+class Predictor(nn.Module):
+
+    def __init__(self):
+        super().__init__()
+        weights = ResNet18_Weights.DEFAULT
+        self.resnet18 = resnet18(weights=weights, progress=False).eval()
+        self.transforms = weights.transforms(antialias=True)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        with torch.no_grad():
+            x = self.transforms(x)
+            y_pred = self.resnet18(x)
+            return y_pred.argmax(dim=1)
+
+
+# %%
+# Now, let's define scripted and non-scripted instances of ``Predictor`` and
+# apply it on multiple tensor images of the same size
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+predictor = Predictor().to(device)
+scripted_predictor = torch.jit.script(predictor).to(device)
+
+batch = torch.stack([dog1, dog2]).to(device)
+
+res = predictor(batch)
+res_scripted = scripted_predictor(batch)
+
+# %%
+# We can verify that the prediction of the scripted and non-scripted models are
+# the same:
+
+import json
+
+with open(Path('../assets') / 'imagenet_class_index.json') as labels_file:
+    labels = json.load(labels_file)
+
+for i, (pred, pred_scripted) in enumerate(zip(res, res_scripted)):
+    assert pred == pred_scripted
+    print(f"Prediction for Dog {i + 1}: {labels[str(pred.item())]}")
+
+# %%
+# Since the model is scripted, it can be easily dumped on disk and re-used
+
+import tempfile
+
+with tempfile.NamedTemporaryFile() as f:
+    scripted_predictor.save(f.name)
+
+    dumped_scripted_predictor = torch.jit.load(f.name)
+    res_scripted_dumped = dumped_scripted_predictor(batch)
+assert (res_scripted_dumped == res_scripted).all()
+
+# %%
diff --git a/0.16/_downloads/2753c5fd8349a28b023c919d8ddf3573/plot_repurposing_annotations.ipynb b/0.16/_downloads/2753c5fd8349a28b023c919d8ddf3573/plot_repurposing_annotations.ipynb
new file mode 100644
index 00000000000..9ea709ee226
--- /dev/null
+++ b/0.16/_downloads/2753c5fd8349a28b023c919d8ddf3573/plot_repurposing_annotations.ipynb
@@ -0,0 +1,205 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Repurposing masks into bounding boxes\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_repurposing_annotations.py>` to download the full example code.</p></div>\n\nThe following example illustrates the operations available\nthe `torchvision.ops <ops>` module for repurposing\nsegmentation masks into object localization annotations for different tasks\n(e.g. transforming masks used by instance and panoptic segmentation\nmethods into bounding boxes used by object detection methods).\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nimport numpy as np\nimport torch\nimport matplotlib.pyplot as plt\n\nimport torchvision.transforms.functional as F\n\n\nASSETS_DIRECTORY = \"../assets\"\n\nplt.rcParams[\"savefig.bbox\"] = \"tight\"\n\n\ndef show(imgs):\n    if not isinstance(imgs, list):\n        imgs = [imgs]\n    fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)\n    for i, img in enumerate(imgs):\n        img = img.detach()\n        img = F.to_pil_image(img)\n        axs[0, i].imshow(np.asarray(img))\n        axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Masks\nIn tasks like instance and panoptic segmentation, masks are commonly defined, and are defined by this package,\nas a multi-dimensional array (e.g. a NumPy array or a PyTorch tensor) with the following shape:\n\n      (num_objects, height, width)\n\nWhere num_objects is the number of annotated objects in the image. Each (height, width) object corresponds to exactly\none object. For example, if your input image has the dimensions 224 x 224 and has four annotated objects the shape\nof your masks annotation has the following shape:\n\n      (4, 224, 224).\n\nA nice property of masks is that they can be easily repurposed to be used in methods to solve a variety of object\nlocalization tasks.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Converting Masks to Bounding Boxes\nFor example, the :func:`~torchvision.ops.masks_to_boxes` operation can be used to\ntransform masks into bounding boxes that can be\nused as input to detection models such as FasterRCNN and RetinaNet.\nWe will take images and masks from the [PenFudan Dataset](https://www.cis.upenn.edu/~jshi/ped_html/).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.io import read_image\n\nimg_path = os.path.join(ASSETS_DIRECTORY, \"FudanPed00054.png\")\nmask_path = os.path.join(ASSETS_DIRECTORY, \"FudanPed00054_mask.png\")\nimg = read_image(img_path)\nmask = read_image(mask_path)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Here the masks are represented as a PNG Image, with floating point values.\nEach pixel is encoded as different colors, with 0 being background.\nNotice that the spatial dimensions of image and mask match.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(mask.size())\nprint(img.size())\nprint(mask)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "# We get the unique colors, as these would be the object ids.\nobj_ids = torch.unique(mask)\n\n# first id is the background, so remove it.\nobj_ids = obj_ids[1:]\n\n# split the color-encoded mask into a set of boolean masks.\n# Note that this snippet would work as well if the masks were float values instead of ints.\nmasks = mask == obj_ids[:, None, None]"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Now the masks are a boolean tensor.\nThe first dimension in this case 3 and denotes the number of instances: there are 3 people in the image.\nThe other two dimensions are height and width, which are equal to the dimensions of the image.\nFor each instance, the boolean tensors represent if the particular pixel\nbelongs to the segmentation mask of the image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(masks.size())\nprint(masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let us visualize an image and plot its corresponding segmentation masks.\nWe will use the :func:`~torchvision.utils.draw_segmentation_masks` to draw the segmentation masks.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_segmentation_masks\n\ndrawn_masks = []\nfor mask in masks:\n    drawn_masks.append(draw_segmentation_masks(img, mask, alpha=0.8, colors=\"blue\"))\n\nshow(drawn_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "To convert the boolean masks into bounding boxes.\nWe will use the :func:`~torchvision.ops.masks_to_boxes` from the torchvision.ops module\nIt returns the boxes in ``(xmin, ymin, xmax, ymax)`` format.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.ops import masks_to_boxes\n\nboxes = masks_to_boxes(masks)\nprint(boxes.size())\nprint(boxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As the shape denotes, there are 3 boxes and in ``(xmin, ymin, xmax, ymax)`` format.\nThese can be visualized very easily with :func:`~torchvision.utils.draw_bounding_boxes` utility\nprovided in `torchvision.utils <utils>`.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_bounding_boxes\n\ndrawn_boxes = draw_bounding_boxes(img, boxes, colors=\"red\")\nshow(drawn_boxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "These boxes can now directly be used by detection models in torchvision.\nHere is demo with a Faster R-CNN model loaded from\n:func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights\n\nweights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT\nmodel = fasterrcnn_resnet50_fpn(weights=weights, progress=False)\nprint(img.size())\n\ntranforms = weights.transforms()\nimg = tranforms(img)\ntarget = {}\ntarget[\"boxes\"] = boxes\ntarget[\"labels\"] = labels = torch.ones((masks.size(0),), dtype=torch.int64)\ndetection_outputs = model(img.unsqueeze(0), [target])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Converting Segmentation Dataset to Detection Dataset\n\nWith this utility it becomes very simple to convert a segmentation dataset to a detection dataset.\nWith this we can now use a segmentation dataset to train a detection model.\nOne can similarly convert panoptic dataset to detection dataset.\nHere is an example where we re-purpose the dataset from the\n[PenFudan Detection Tutorial](https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class SegmentationToDetectionDataset(torch.utils.data.Dataset):\n    def __init__(self, root, transforms):\n        self.root = root\n        self.transforms = transforms\n        # load all image files, sorting them to\n        # ensure that they are aligned\n        self.imgs = list(sorted(os.listdir(os.path.join(root, \"PNGImages\"))))\n        self.masks = list(sorted(os.listdir(os.path.join(root, \"PedMasks\"))))\n\n    def __getitem__(self, idx):\n        # load images and masks\n        img_path = os.path.join(self.root, \"PNGImages\", self.imgs[idx])\n        mask_path = os.path.join(self.root, \"PedMasks\", self.masks[idx])\n\n        img = read_image(img_path)\n        mask = read_image(mask_path)\n\n        img = F.convert_image_dtype(img, dtype=torch.float)\n        mask = F.convert_image_dtype(mask, dtype=torch.float)\n\n        # We get the unique colors, as these would be the object ids.\n        obj_ids = torch.unique(mask)\n\n        # first id is the background, so remove it.\n        obj_ids = obj_ids[1:]\n\n        # split the color-encoded mask into a set of boolean masks.\n        masks = mask == obj_ids[:, None, None]\n\n        boxes = masks_to_boxes(masks)\n\n        # there is only one class\n        labels = torch.ones((masks.shape[0],), dtype=torch.int64)\n\n        target = {}\n        target[\"boxes\"] = boxes\n        target[\"labels\"] = labels\n\n        if self.transforms is not None:\n            img, target = self.transforms(img, target)\n\n        return img, target"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/27d6228101da5d268ad41f8f06010df8/plot_visualization_utils.py b/0.16/_downloads/27d6228101da5d268ad41f8f06010df8/plot_visualization_utils.py
new file mode 100644
index 00000000000..98089c54dbb
--- /dev/null
+++ b/0.16/_downloads/27d6228101da5d268ad41f8f06010df8/plot_visualization_utils.py
@@ -0,0 +1,462 @@
+"""
+=======================
+Visualization utilities
+=======================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_visualization_utils.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_visualization_utils.py>` to download the full example code.
+
+This example illustrates some of the utilities that torchvision offers for
+visualizing images, bounding boxes, segmentation masks and keypoints.
+"""
+
+# sphinx_gallery_thumbnail_path = "../../gallery/assets/visualization_utils_thumbnail2.png"
+
+import torch
+import numpy as np
+import matplotlib.pyplot as plt
+
+import torchvision.transforms.functional as F
+
+
+plt.rcParams["savefig.bbox"] = 'tight'
+
+
+def show(imgs):
+    if not isinstance(imgs, list):
+        imgs = [imgs]
+    fig, axs = plt.subplots(ncols=len(imgs), squeeze=False)
+    for i, img in enumerate(imgs):
+        img = img.detach()
+        img = F.to_pil_image(img)
+        axs[0, i].imshow(np.asarray(img))
+        axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
+
+
+# %%
+# Visualizing a grid of images
+# ----------------------------
+# The :func:`~torchvision.utils.make_grid` function can be used to create a
+# tensor that represents multiple images in a grid.  This util requires a single
+# image of dtype ``uint8`` as input.
+
+from torchvision.utils import make_grid
+from torchvision.io import read_image
+from pathlib import Path
+
+dog1_int = read_image(str(Path('../assets') / 'dog1.jpg'))
+dog2_int = read_image(str(Path('../assets') / 'dog2.jpg'))
+dog_list = [dog1_int, dog2_int]
+
+grid = make_grid(dog_list)
+show(grid)
+
+# %%
+# Visualizing bounding boxes
+# --------------------------
+# We can use :func:`~torchvision.utils.draw_bounding_boxes` to draw boxes on an
+# image. We can set the colors, labels, width as well as font and font size.
+# The boxes are in ``(xmin, ymin, xmax, ymax)`` format.
+
+from torchvision.utils import draw_bounding_boxes
+
+
+boxes = torch.tensor([[50, 50, 100, 200], [210, 150, 350, 430]], dtype=torch.float)
+colors = ["blue", "yellow"]
+result = draw_bounding_boxes(dog1_int, boxes, colors=colors, width=5)
+show(result)
+
+
+# %%
+# Naturally, we can also plot bounding boxes produced by torchvision detection
+# models.  Here is a demo with a Faster R-CNN model loaded from
+# :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`
+# model. For more details on the output of such models, you may
+# refer to :ref:`instance_seg_output`.
+
+from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights
+
+
+weights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT
+transforms = weights.transforms()
+
+images = [transforms(d) for d in dog_list]
+
+model = fasterrcnn_resnet50_fpn(weights=weights, progress=False)
+model = model.eval()
+
+outputs = model(images)
+print(outputs)
+
+# %%
+# Let's plot the boxes detected by our model. We will only plot the boxes with a
+# score greater than a given threshold.
+
+score_threshold = .8
+dogs_with_boxes = [
+    draw_bounding_boxes(dog_int, boxes=output['boxes'][output['scores'] > score_threshold], width=4)
+    for dog_int, output in zip(dog_list, outputs)
+]
+show(dogs_with_boxes)
+
+# %%
+# Visualizing segmentation masks
+# ------------------------------
+# The :func:`~torchvision.utils.draw_segmentation_masks` function can be used to
+# draw segmentation masks on images. Semantic segmentation and instance
+# segmentation models have different outputs, so we will treat each
+# independently.
+#
+# .. _semantic_seg_output:
+#
+# Semantic segmentation models
+# ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+#
+# We will see how to use it with torchvision's FCN Resnet-50, loaded with
+# :func:`~torchvision.models.segmentation.fcn_resnet50`. Let's start by looking
+# at the output of the model.
+
+from torchvision.models.segmentation import fcn_resnet50, FCN_ResNet50_Weights
+
+weights = FCN_ResNet50_Weights.DEFAULT
+transforms = weights.transforms(resize_size=None)
+
+model = fcn_resnet50(weights=weights, progress=False)
+model = model.eval()
+
+batch = torch.stack([transforms(d) for d in dog_list])
+output = model(batch)['out']
+print(output.shape, output.min().item(), output.max().item())
+
+# %%
+# As we can see above, the output of the segmentation model is a tensor of shape
+# ``(batch_size, num_classes, H, W)``. Each value is a non-normalized score, and
+# we can normalize them into ``[0, 1]`` by using a softmax. After the softmax,
+# we can interpret each value as a probability indicating how likely a given
+# pixel is to belong to a given class.
+#
+# Let's plot the masks that have been detected for the dog class and for the
+# boat class:
+
+sem_class_to_idx = {cls: idx for (idx, cls) in enumerate(weights.meta["categories"])}
+
+normalized_masks = torch.nn.functional.softmax(output, dim=1)
+
+dog_and_boat_masks = [
+    normalized_masks[img_idx, sem_class_to_idx[cls]]
+    for img_idx in range(len(dog_list))
+    for cls in ('dog', 'boat')
+]
+
+show(dog_and_boat_masks)
+
+# %%
+# As expected, the model is confident about the dog class, but not so much for
+# the boat class.
+#
+# The :func:`~torchvision.utils.draw_segmentation_masks` function can be used to
+# plots those masks on top of the original image. This function expects the
+# masks to be boolean masks, but our masks above contain probabilities in ``[0,
+# 1]``. To get boolean masks, we can do the following:
+
+class_dim = 1
+boolean_dog_masks = (normalized_masks.argmax(class_dim) == sem_class_to_idx['dog'])
+print(f"shape = {boolean_dog_masks.shape}, dtype = {boolean_dog_masks.dtype}")
+show([m.float() for m in boolean_dog_masks])
+
+
+# %%
+# The line above where we define ``boolean_dog_masks`` is a bit cryptic, but you
+# can read it as the following query: "For which pixels is 'dog' the most likely
+# class?"
+#
+# .. note::
+#   While we're using the ``normalized_masks`` here, we would have
+#   gotten the same result by using the non-normalized scores of the model
+#   directly (as the softmax operation preserves the order).
+#
+# Now that we have boolean masks, we can use them with
+# :func:`~torchvision.utils.draw_segmentation_masks` to plot them on top of the
+# original images:
+
+from torchvision.utils import draw_segmentation_masks
+
+dogs_with_masks = [
+    draw_segmentation_masks(img, masks=mask, alpha=0.7)
+    for img, mask in zip(dog_list, boolean_dog_masks)
+]
+show(dogs_with_masks)
+
+# %%
+# We can plot more than one mask per image! Remember that the model returned as
+# many masks as there are classes. Let's ask the same query as above, but this
+# time for *all* classes, not just the dog class: "For each pixel and each class
+# C, is class C the most likely class?"
+#
+# This one is a bit more involved, so we'll first show how to do it with a
+# single image, and then we'll generalize to the batch
+
+num_classes = normalized_masks.shape[1]
+dog1_masks = normalized_masks[0]
+class_dim = 0
+dog1_all_classes_masks = dog1_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None]
+
+print(f"dog1_masks shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}")
+print(f"dog1_all_classes_masks = {dog1_all_classes_masks.shape}, dtype = {dog1_all_classes_masks.dtype}")
+
+dog_with_all_masks = draw_segmentation_masks(dog1_int, masks=dog1_all_classes_masks, alpha=.6)
+show(dog_with_all_masks)
+
+# %%
+# We can see in the image above that only 2 masks were drawn: the mask for the
+# background and the mask for the dog. This is because the model thinks that
+# only these 2 classes are the most likely ones across all the pixels. If the
+# model had detected another class as the most likely among other pixels, we
+# would have seen its mask above.
+#
+# Removing the background mask is as simple as passing
+# ``masks=dog1_all_classes_masks[1:]``, because the background class is the
+# class with index 0.
+#
+# Let's now do the same but for an entire batch of images. The code is similar
+# but involves a bit more juggling with the dimensions.
+
+class_dim = 1
+all_classes_masks = normalized_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None, None]
+print(f"shape = {all_classes_masks.shape}, dtype = {all_classes_masks.dtype}")
+# The first dimension is the classes now, so we need to swap it
+all_classes_masks = all_classes_masks.swapaxes(0, 1)
+
+dogs_with_masks = [
+    draw_segmentation_masks(img, masks=mask, alpha=.6)
+    for img, mask in zip(dog_list, all_classes_masks)
+]
+show(dogs_with_masks)
+
+
+# %%
+# .. _instance_seg_output:
+#
+# Instance segmentation models
+# ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+#
+# Instance segmentation models have a significantly different output from the
+# semantic segmentation models. We will see here how to plot the masks for such
+# models. Let's start by analyzing the output of a Mask-RCNN model. Note that
+# these models don't require the images to be normalized, so we don't need to
+# use the normalized batch.
+#
+# .. note::
+#
+#     We will here describe the output of a Mask-RCNN model. The models in
+#     :ref:`object_det_inst_seg_pers_keypoint_det` all have a similar output
+#     format, but some of them may have extra info like keypoints for
+#     :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`, and some
+#     of them may not have masks, like
+#     :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`.
+
+from torchvision.models.detection import maskrcnn_resnet50_fpn, MaskRCNN_ResNet50_FPN_Weights
+
+weights = MaskRCNN_ResNet50_FPN_Weights.DEFAULT
+transforms = weights.transforms()
+
+images = [transforms(d) for d in dog_list]
+
+model = maskrcnn_resnet50_fpn(weights=weights, progress=False)
+model = model.eval()
+
+output = model(images)
+print(output)
+
+# %%
+# Let's break this down. For each image in the batch, the model outputs some
+# detections (or instances). The number of detections varies for each input
+# image. Each instance is described by its bounding box, its label, its score
+# and its mask.
+#
+# The way the output is organized is as follows: the output is a list of length
+# ``batch_size``. Each entry in the list corresponds to an input image, and it
+# is a dict with keys 'boxes', 'labels', 'scores', and 'masks'. Each value
+# associated to those keys has ``num_instances`` elements in it.  In our case
+# above there are 3 instances detected in the first image, and 2 instances in
+# the second one.
+#
+# The boxes can be plotted with :func:`~torchvision.utils.draw_bounding_boxes`
+# as above, but here we're more interested in the masks. These masks are quite
+# different from the masks that we saw above for the semantic segmentation
+# models.
+
+dog1_output = output[0]
+dog1_masks = dog1_output['masks']
+print(f"shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}, "
+      f"min = {dog1_masks.min()}, max = {dog1_masks.max()}")
+
+# %%
+# Here the masks correspond to probabilities indicating, for each pixel, how
+# likely it is to belong to the predicted label of that instance. Those
+# predicted labels correspond to the 'labels' element in the same output dict.
+# Let's see which labels were predicted for the instances of the first image.
+
+print("For the first dog, the following instances were detected:")
+print([weights.meta["categories"][label] for label in dog1_output['labels']])
+
+# %%
+# Interestingly, the model detects two persons in the image. Let's go ahead and
+# plot those masks. Since :func:`~torchvision.utils.draw_segmentation_masks`
+# expects boolean masks, we need to convert those probabilities into boolean
+# values. Remember that the semantic of those masks is "How likely is this pixel
+# to belong to the predicted class?". As a result, a natural way of converting
+# those masks into boolean values is to threshold them with the 0.5 probability
+# (one could also choose a different threshold).
+
+proba_threshold = 0.5
+dog1_bool_masks = dog1_output['masks'] > proba_threshold
+print(f"shape = {dog1_bool_masks.shape}, dtype = {dog1_bool_masks.dtype}")
+
+# There's an extra dimension (1) to the masks. We need to remove it
+dog1_bool_masks = dog1_bool_masks.squeeze(1)
+
+show(draw_segmentation_masks(dog1_int, dog1_bool_masks, alpha=0.9))
+
+# %%
+# The model seems to have properly detected the dog, but it also confused trees
+# with people. Looking more closely at the scores will help us plot more
+# relevant masks:
+
+print(dog1_output['scores'])
+
+# %%
+# Clearly the model is more confident about the dog detection than it is about
+# the people detections. That's good news. When plotting the masks, we can ask
+# for only those that have a good score. Let's use a score threshold of .75
+# here, and also plot the masks of the second dog.
+
+score_threshold = .75
+
+boolean_masks = [
+    out['masks'][out['scores'] > score_threshold] > proba_threshold
+    for out in output
+]
+
+dogs_with_masks = [
+    draw_segmentation_masks(img, mask.squeeze(1))
+    for img, mask in zip(dog_list, boolean_masks)
+]
+show(dogs_with_masks)
+
+# %%
+# The two 'people' masks in the first image where not selected because they have
+# a lower score than the score threshold. Similarly, in the second image, the
+# instance with class 15 (which corresponds to 'bench') was not selected.
+
+# %%
+# .. _keypoint_output:
+#
+# Visualizing keypoints
+# ------------------------------
+# The :func:`~torchvision.utils.draw_keypoints` function can be used to
+# draw keypoints on images. We will see how to use it with
+# torchvision's KeypointRCNN loaded with :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`.
+# We will first have a look at output of the model.
+#
+
+from torchvision.models.detection import keypointrcnn_resnet50_fpn, KeypointRCNN_ResNet50_FPN_Weights
+from torchvision.io import read_image
+
+person_int = read_image(str(Path("../assets") / "person1.jpg"))
+
+weights = KeypointRCNN_ResNet50_FPN_Weights.DEFAULT
+transforms = weights.transforms()
+
+person_float = transforms(person_int)
+
+model = keypointrcnn_resnet50_fpn(weights=weights, progress=False)
+model = model.eval()
+
+outputs = model([person_float])
+print(outputs)
+
+# %%
+# As we see the output contains a list of dictionaries.
+# The output list is of length batch_size.
+# We currently have just a single image so length of list is 1.
+# Each entry in the list corresponds to an input image,
+# and it is a dict with keys `boxes`, `labels`, `scores`, `keypoints` and `keypoint_scores`.
+# Each value associated to those keys has `num_instances` elements in it.
+# In our case above there are 2 instances detected in the image.
+
+kpts = outputs[0]['keypoints']
+scores = outputs[0]['scores']
+
+print(kpts)
+print(scores)
+
+# %%
+# The KeypointRCNN model detects there are two instances in the image.
+# If you plot the boxes by using :func:`~draw_bounding_boxes`
+# you would recognize they are the person and the surfboard.
+# If we look at the scores, we will realize that the model is much more confident about the person than surfboard.
+# We could now set a threshold confidence and plot instances which we are confident enough.
+# Let us set a threshold of 0.75 and filter out the keypoints corresponding to the person.
+
+detect_threshold = 0.75
+idx = torch.where(scores > detect_threshold)
+keypoints = kpts[idx]
+
+print(keypoints)
+
+# %%
+# Great, now we have the keypoints corresponding to the person.
+# Each keypoint is represented by x, y coordinates and the visibility.
+# We can now use the :func:`~torchvision.utils.draw_keypoints` function to draw keypoints.
+# Note that the utility expects uint8 images.
+
+from torchvision.utils import draw_keypoints
+
+res = draw_keypoints(person_int, keypoints, colors="blue", radius=3)
+show(res)
+
+# %%
+# As we see the keypoints appear as colored circles over the image.
+# The coco keypoints for a person are ordered and represent the following list.\
+
+coco_keypoints = [
+    "nose", "left_eye", "right_eye", "left_ear", "right_ear",
+    "left_shoulder", "right_shoulder", "left_elbow", "right_elbow",
+    "left_wrist", "right_wrist", "left_hip", "right_hip",
+    "left_knee", "right_knee", "left_ankle", "right_ankle",
+]
+
+# %%
+# What if we are interested in joining the keypoints?
+# This is especially useful in creating pose detection or action recognition.
+# We can join the keypoints easily using the `connectivity` parameter.
+# A close observation would reveal that we would need to join the points in below
+# order to construct human skeleton.
+#
+# nose -> left_eye -> left_ear.                              (0, 1), (1, 3)
+#
+# nose -> right_eye -> right_ear.                            (0, 2), (2, 4)
+#
+# nose -> left_shoulder -> left_elbow -> left_wrist.         (0, 5), (5, 7), (7, 9)
+#
+# nose -> right_shoulder -> right_elbow -> right_wrist.      (0, 6), (6, 8), (8, 10)
+#
+# left_shoulder -> left_hip -> left_knee -> left_ankle.      (5, 11), (11, 13), (13, 15)
+#
+# right_shoulder -> right_hip -> right_knee -> right_ankle.  (6, 12), (12, 14), (14, 16)
+#
+# We will create a list containing these keypoint ids to be connected.
+
+connect_skeleton = [
+    (0, 1), (0, 2), (1, 3), (2, 4), (0, 5), (0, 6), (5, 7), (6, 8),
+    (7, 9), (8, 10), (5, 11), (6, 12), (11, 13), (12, 14), (13, 15), (14, 16)
+]
+
+# %%
+# We pass the above list to the connectivity parameter to connect the keypoints.
+#
+
+res = draw_keypoints(person_int, keypoints, connectivity=connect_skeleton, colors="blue", radius=4, width=3)
+show(res)
diff --git a/0.16/_downloads/3f3cbe5cc8b5758610d0ab95995b0b8c/plot_custom_transforms.ipynb b/0.16/_downloads/3f3cbe5cc8b5758610d0ab95995b0b8c/plot_custom_transforms.ipynb
new file mode 100644
index 00000000000..fecad9844a1
--- /dev/null
+++ b/0.16/_downloads/3f3cbe5cc8b5758610d0ab95995b0b8c/plot_custom_transforms.ipynb
@@ -0,0 +1,97 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# How to write your own v2 transforms\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_transforms.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_transforms.py>` to download the full example code.</p></div>\n\nThis guide explains how to write transforms that are compatible with the\ntorchvision transforms V2 API.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nfrom torchvision import tv_tensors\nfrom torchvision.transforms import v2"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Just create a ``nn.Module`` and override the ``forward`` method\n\nIn most cases, this is all you're going to need, as long as you already know\nthe structure of the input that your transform will expect. For example if\nyou're just doing image classification, your transform will typically accept a\nsingle image as input, or a ``(img, label)`` input. So you can just hard-code\nyour ``forward`` method to accept just that, e.g.\n\n.. code:: python\n\n    class MyCustomTransform(torch.nn.Module):\n        def forward(self, img, label):\n            # Do some transformations\n            return new_img, new_label\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>This means that if you have a custom transform that is already compatible\n    with the V1 transforms (those in ``torchvision.transforms``), it will\n    still work with the V2 transforms without any change!</p></div>\n\nWe will illustrate this more completely below with a typical detection case,\nwhere our samples are just images, bounding boxes and labels:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class MyCustomTransform(torch.nn.Module):\n    def forward(self, img, bboxes, label):  # we assume inputs are always structured like this\n        print(\n            f\"I'm transforming an image of shape {img.shape} \"\n            f\"with bboxes = {bboxes}\\n{label = }\"\n        )\n        # Do some transformations. Here, we're just passing though the input\n        return img, bboxes, label\n\n\ntransforms = v2.Compose([\n    MyCustomTransform(),\n    v2.RandomResizedCrop((224, 224), antialias=True),\n    v2.RandomHorizontalFlip(p=1),\n    v2.Normalize(mean=[0, 0, 0], std=[1, 1, 1])\n])\n\nH, W = 256, 256\nimg = torch.rand(3, H, W)\nbboxes = tv_tensors.BoundingBoxes(\n    torch.tensor([[0, 10, 10, 20], [50, 50, 70, 70]]),\n    format=\"XYXY\",\n    canvas_size=(H, W)\n)\nlabel = 3\n\nout_img, out_bboxes, out_label = transforms(img, bboxes, label)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(f\"Output image shape: {out_img.shape}\\nout_bboxes = {out_bboxes}\\n{out_label = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-info\"><h4>Note</h4><p>While working with TVTensor classes in your code, make sure to\n    familiarize yourself with this section:\n    `tv_tensor_unwrapping_behaviour`</p></div>\n\n## Supporting arbitrary input structures\n\nIn the section above, we have assumed that you already know the structure of\nyour inputs and that you're OK with hard-coding this expected structure in\nyour code. If you want your custom transforms to be as flexible as possible,\nthis can be a bit limiting.\n\nA key feature of the builtin Torchvision V2 transforms is that they can accept\narbitrary input structure and return the same structure as output (with\ntransformed entries). For example, transforms can accept a single image, or a\ntuple of ``(img, label)``, or an arbitrary nested dictionary as input:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "structured_input = {\n    \"img\": img,\n    \"annotations\": (bboxes, label),\n    \"something_that_will_be_ignored\": (1, \"hello\")\n}\nstructured_output = v2.RandomHorizontalFlip(p=1)(structured_input)\n\nassert isinstance(structured_output, dict)\nassert structured_output[\"something_that_will_be_ignored\"] == (1, \"hello\")\nprint(f\"The transformed bboxes are:\\n{structured_output['annotations'][0]}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "If you want to reproduce this behavior in your own transform, we invite you to\nlook at our [code](https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/_transform.py)\nand adapt it to your needs.\n\nIn brief, the core logic is to unpack the input into a flat list using [pytree](https://github.com/pytorch/pytorch/blob/main/torch/utils/_pytree.py), and\nthen transform only the entries that can be transformed (the decision is made\nbased on the **class** of the entries, as all TVTensors are\ntensor-subclasses) plus some custom logic that is out of score here - check the\ncode for details. The (potentially transformed) entries are then repacked and\nreturned, in the same structure as the input.\n\nWe do not provide public dev-facing tools to achieve that at this time, but if\nthis is something that would be valuable to you, please let us know by opening\nan issue on our [GitHub repo](https://github.com/pytorch/vision/issues).\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/40e082c3b2415f11f987eaa06c4d7513/plot_transforms_e2e.py b/0.16/_downloads/40e082c3b2415f11f987eaa06c4d7513/plot_transforms_e2e.py
new file mode 100644
index 00000000000..6c58b4a5a9a
--- /dev/null
+++ b/0.16/_downloads/40e082c3b2415f11f987eaa06c4d7513/plot_transforms_e2e.py
@@ -0,0 +1,181 @@
+"""
+===============================================================
+Transforms v2: End-to-end object detection/segmentation example
+===============================================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_e2e.py>` to download the full example code.
+
+Object detection and segmentation tasks are natively supported:
+``torchvision.transforms.v2`` enables jointly transforming images, videos,
+bounding boxes, and masks.
+
+This example showcases an end-to-end instance segmentation training case using
+Torchvision utils from ``torchvision.datasets``, ``torchvision.models`` and
+``torchvision.transforms.v2``. Everything covered here can be applied similarly
+to object detection or semantic segmentation tasks.
+"""
+
+# %%
+import pathlib
+
+import torch
+import torch.utils.data
+
+from torchvision import models, datasets, tv_tensors
+from torchvision.transforms import v2
+
+torch.manual_seed(0)
+
+# This loads fake data for illustration purposes of this example. In practice, you'll have
+# to replace this with the proper data.
+# If you're trying to run that on collab, you can download the assets and the
+# helpers from https://github.com/pytorch/vision/tree/main/gallery/
+ROOT = pathlib.Path("../assets") / "coco"
+IMAGES_PATH = str(ROOT / "images")
+ANNOTATIONS_PATH = str(ROOT / "instances.json")
+from helpers import plot
+
+
+# %%
+# Dataset preparation
+# -------------------
+#
+# We start off by loading the :class:`~torchvision.datasets.CocoDetection` dataset to have a look at what it currently
+# returns.
+
+dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH)
+
+sample = dataset[0]
+img, target = sample
+print(f"{type(img) = }\n{type(target) = }\n{type(target[0]) = }\n{target[0].keys() = }")
+
+
+# %%
+# Torchvision datasets preserve the data structure and types as it was intended
+# by the datasets authors. So by default, the output structure may not always be
+# compatible with the models or the transforms.
+#
+# To overcome that, we can use the
+# :func:`~torchvision.datasets.wrap_dataset_for_transforms_v2` function. For
+# :class:`~torchvision.datasets.CocoDetection`, this changes the target
+# structure to a single dictionary of lists:
+
+dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=("boxes", "labels", "masks"))
+
+sample = dataset[0]
+img, target = sample
+print(f"{type(img) = }\n{type(target) = }\n{target.keys() = }")
+print(f"{type(target['boxes']) = }\n{type(target['labels']) = }\n{type(target['masks']) = }")
+
+# %%
+# We used the ``target_keys`` parameter to specify the kind of output we're
+# interested in. Our dataset now returns a target which is dict where the values
+# are :ref:`TVTensors <what_are_tv_tensors>` (all are :class:`torch.Tensor`
+# subclasses). We're dropped all unncessary keys from the previous output, but
+# if you need any of the original keys e.g. "image_id", you can still ask for
+# it.
+#
+# .. note::
+#
+#     If you just want to do detection, you don't need and shouldn't pass
+#     "masks" in ``target_keys``: if masks are present in the sample, they will
+#     be transformed, slowing down your transformations unnecessarily.
+#
+# As baseline, let's have a look at a sample without transformations:
+
+plot([dataset[0], dataset[1]])
+
+
+# %%
+# Transforms
+# ----------
+#
+# Let's now define our pre-processing transforms. All the transforms know how
+# to handle images, bouding boxes and masks when relevant.
+#
+# Transforms are typically passed as the ``transforms`` parameter of the
+# dataset so that they can leverage multi-processing from the
+# :class:`torch.utils.data.DataLoader`.
+
+transforms = v2.Compose(
+    [
+        v2.ToImage(),
+        v2.RandomPhotometricDistort(p=1),
+        v2.RandomZoomOut(fill={tv_tensors.Image: (123, 117, 104), "others": 0}),
+        v2.RandomIoUCrop(),
+        v2.RandomHorizontalFlip(p=1),
+        v2.SanitizeBoundingBoxes(),
+        v2.ToDtype(torch.float32, scale=True),
+    ]
+)
+
+dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH, transforms=transforms)
+dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=["boxes", "labels", "masks"])
+
+# %%
+# A few things are worth noting here:
+#
+# - We're converting the PIL image into a
+#   :class:`~torchvision.transforms.v2.Image` object. This isn't strictly
+#   necessary, but relying on Tensors (here: a Tensor subclass) will
+#   :ref:`generally be faster <transforms_perf>`.
+# - We are calling :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` to
+#   make sure we remove degenerate bounding boxes, as well as their
+#   corresponding labels and masks.
+#   :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` should be placed
+#   at least once at the end of a detection pipeline; it is particularly
+#   critical if :class:`~torchvision.transforms.v2.RandomIoUCrop` was used.
+#
+# Let's look how the sample looks like with our augmentation pipeline in place:
+
+# sphinx_gallery_thumbnail_number = 2
+plot([dataset[0], dataset[1]])
+
+
+# %%
+# We can see that the color of the images were distorted, zoomed in or out, and flipped.
+# The bounding boxes and the masks were transformed accordingly. And without any further ado, we can start training.
+#
+# Data loading and training loop
+# ------------------------------
+#
+# Below we're using Mask-RCNN which is an instance segmentation model, but
+# everything we've covered in this tutorial also applies to object detection and
+# semantic segmentation tasks.
+
+data_loader = torch.utils.data.DataLoader(
+    dataset,
+    batch_size=2,
+    # We need a custom collation function here, since the object detection
+    # models expect a sequence of images and target dictionaries. The default
+    # collation function tries to torch.stack() the individual elements,
+    # which fails in general for object detection, because the number of bouding
+    # boxes varies between the images of a same batch.
+    collate_fn=lambda batch: tuple(zip(*batch)),
+)
+
+model = models.get_model("maskrcnn_resnet50_fpn_v2", weights=None, weights_backbone=None).train()
+
+for imgs, targets in data_loader:
+    loss_dict = model(imgs, targets)
+    # Put your training logic here
+
+    print(f"{[img.shape for img in imgs] = }")
+    print(f"{[type(target) for target in targets] = }")
+    for name, loss_val in loss_dict.items():
+        print(f"{name:<20}{loss_val:.3f}")
+
+# %%
+# Training References
+# -------------------
+#
+# From there, you can check out the `torchvision references
+# <https://github.com/pytorch/vision/tree/main/references>`_ where you'll find
+# the actual training scripts we use to train our models.
+#
+# **Disclaimer** The code in our references is more complex than what you'll
+# need for your own use-cases: this is because we're supporting different
+# backends (PIL, tensors, TVTensors) and different transforms namespaces (v1 and
+# v2). So don't be afraid to simplify and only keep what you need.
diff --git a/0.16/_downloads/4333051365715be174a783856419631f/plot_scripted_tensor_transforms.ipynb b/0.16/_downloads/4333051365715be174a783856419631f/plot_scripted_tensor_transforms.ipynb
new file mode 100644
index 00000000000..2509ad25532
--- /dev/null
+++ b/0.16/_downloads/4333051365715be174a783856419631f/plot_scripted_tensor_transforms.ipynb
@@ -0,0 +1,133 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Torchscript support\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_scripted_tensor_transforms.py>` to download the full example code.</p></div>\n\nThis example illustrates [torchscript](https://pytorch.org/docs/stable/jit.html) support of the torchvision\n`transforms <transforms>` on Tensor images.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from pathlib import Path\n\nimport matplotlib.pyplot as plt\n\nimport torch\nimport torch.nn as nn\n\nimport torchvision.transforms as v1\nfrom torchvision.io import read_image\n\nplt.rcParams[\"savefig.bbox\"] = 'tight'\ntorch.manual_seed(1)\n\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nimport sys\nsys.path += [\"../transforms\"]\nfrom helpers import plot\nASSETS_PATH = Path('../assets')"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Most transforms support torchscript. For composing transforms, we use\n:class:`torch.nn.Sequential` instead of\n:class:`~torchvision.transforms.v2.Compose`:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dog1 = read_image(str(ASSETS_PATH / 'dog1.jpg'))\ndog2 = read_image(str(ASSETS_PATH / 'dog2.jpg'))\n\ntransforms = torch.nn.Sequential(\n    v1.RandomCrop(224),\n    v1.RandomHorizontalFlip(p=0.3),\n)\n\nscripted_transforms = torch.jit.script(transforms)\n\nplot([dog1, scripted_transforms(dog1), dog2, scripted_transforms(dog2)])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-danger\"><h4>Warning</h4><p>Above we have used transforms from the ``torchvision.transforms``\n    namespace, i.e. the \"v1\" transforms. The v2 transforms from the\n    ``torchvision.transforms.v2`` namespace are the `recommended\n    <v1_or_v2>` way to use transforms in your code.\n\n    The v2 transforms also support torchscript, but if you call\n    ``torch.jit.script()`` on a v2 **class** transform, you'll actually end up\n    with its (scripted) v1 equivalent.  This may lead to slightly different\n    results between the scripted and eager executions due to implementation\n    differences between v1 and v2.\n\n    If you really need torchscript support for the v2 transforms, **we\n    recommend scripting the functionals** from the\n    ``torchvision.transforms.v2.functional`` namespace to avoid surprises.</p></div>\n\nBelow we now show how to combine image transformations and a model forward\npass, while using ``torch.jit.script`` to obtain a single scripted module.\n\nLet's define a ``Predictor`` module that transforms the input tensor and then\napplies an ImageNet model on it.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models import resnet18, ResNet18_Weights\n\n\nclass Predictor(nn.Module):\n\n    def __init__(self):\n        super().__init__()\n        weights = ResNet18_Weights.DEFAULT\n        self.resnet18 = resnet18(weights=weights, progress=False).eval()\n        self.transforms = weights.transforms(antialias=True)\n\n    def forward(self, x: torch.Tensor) -> torch.Tensor:\n        with torch.no_grad():\n            x = self.transforms(x)\n            y_pred = self.resnet18(x)\n            return y_pred.argmax(dim=1)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Now, let's define scripted and non-scripted instances of ``Predictor`` and\napply it on multiple tensor images of the same size\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n\npredictor = Predictor().to(device)\nscripted_predictor = torch.jit.script(predictor).to(device)\n\nbatch = torch.stack([dog1, dog2]).to(device)\n\nres = predictor(batch)\nres_scripted = scripted_predictor(batch)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can verify that the prediction of the scripted and non-scripted models are\nthe same:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import json\n\nwith open(Path('../assets') / 'imagenet_class_index.json') as labels_file:\n    labels = json.load(labels_file)\n\nfor i, (pred, pred_scripted) in enumerate(zip(res, res_scripted)):\n    assert pred == pred_scripted\n    print(f\"Prediction for Dog {i + 1}: {labels[str(pred.item())]}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Since the model is scripted, it can be easily dumped on disk and re-used\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import tempfile\n\nwith tempfile.NamedTemporaryFile() as f:\n    scripted_predictor.save(f.name)\n\n    dumped_scripted_predictor = torch.jit.load(f.name)\n    res_scripted_dumped = dumped_scripted_predictor(batch)\nassert (res_scripted_dumped == res_scripted).all()"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/4890c38ddf15e56f86ebbdd480cb7ade/plot_custom_tv_tensors.py b/0.16/_downloads/4890c38ddf15e56f86ebbdd480cb7ade/plot_custom_tv_tensors.py
new file mode 100644
index 00000000000..bf5ee198837
--- /dev/null
+++ b/0.16/_downloads/4890c38ddf15e56f86ebbdd480cb7ade/plot_custom_tv_tensors.py
@@ -0,0 +1,119 @@
+"""
+====================================
+How to write your own TVTensor class
+====================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_tv_tensors.py>` to download the full example code.
+
+This guide is intended for advanced users and downstream library maintainers. We explain how to
+write your own TVTensor class, and how to make it compatible with the built-in
+Torchvision v2 transforms. Before continuing, make sure you have read
+:ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.
+"""
+
+# %%
+import torch
+from torchvision import tv_tensors
+from torchvision.transforms import v2
+
+# %%
+# We will create a very simple class that just inherits from the base
+# :class:`~torchvision.tv_tensors.TVTensor` class. It will be enough to cover
+# what you need to know to implement your more elaborate uses-cases. If you need
+# to create a class that carries meta-data, take a look at how the
+# :class:`~torchvision.tv_tensors.BoundingBoxes` class is `implemented
+# <https://github.com/pytorch/vision/blob/main/torchvision/tv_tensors/_bounding_box.py>`_.
+
+
+class MyTVTensor(tv_tensors.TVTensor):
+    pass
+
+
+my_dp = MyTVTensor([1, 2, 3])
+my_dp
+
+# %%
+# Now that we have defined our custom TVTensor class, we want it to be
+# compatible with the built-in torchvision transforms, and the functional API.
+# For that, we need to implement a kernel which performs the core of the
+# transformation, and then "hook" it to the functional that we want to support
+# via :func:`~torchvision.transforms.v2.functional.register_kernel`.
+#
+# We illustrate this process below: we create a kernel for the "horizontal flip"
+# operation of our MyTVTensor class, and register it to the functional API.
+
+from torchvision.transforms.v2 import functional as F
+
+
+@F.register_kernel(functional="hflip", tv_tensor_cls=MyTVTensor)
+def hflip_my_tv_tensor(my_dp, *args, **kwargs):
+    print("Flipping!")
+    out = my_dp.flip(-1)
+    return tv_tensors.wrap(out, like=my_dp)
+
+
+# %%
+# To understand why :func:`~torchvision.tv_tensors.wrap` is used, see
+# :ref:`tv_tensor_unwrapping_behaviour`. Ignore the ``*args, **kwargs`` for now,
+# we will explain it below in :ref:`param_forwarding`.
+#
+# .. note::
+#
+#     In our call to ``register_kernel`` above we used a string
+#     ``functional="hflip"`` to refer to the functional we want to hook into. We
+#     could also have used the  functional *itself*, i.e.
+#     ``@register_kernel(functional=F.hflip, ...)``.
+#
+# Now that we have registered our kernel, we can call the functional API on a
+# ``MyTVTensor`` instance:
+
+my_dp = MyTVTensor(torch.rand(3, 256, 256))
+_ = F.hflip(my_dp)
+
+# %%
+# And we can also use the
+# :class:`~torchvision.transforms.v2.RandomHorizontalFlip` transform, since it relies on :func:`~torchvision.transforms.v2.functional.hflip` internally:
+t = v2.RandomHorizontalFlip(p=1)
+_ = t(my_dp)
+
+# %%
+# .. note::
+#
+#     We cannot register a kernel for a transform class, we can only register a
+#     kernel for a **functional**. The reason we can't register a transform
+#     class is because one transform may internally rely on more than one
+#     functional, so in general we can't register a single kernel for a given
+#     class.
+#
+# .. _param_forwarding:
+#
+# Parameter forwarding, and ensuring future compatibility of your kernels
+# -----------------------------------------------------------------------
+#
+# The functional API that you're hooking into is public and therefore
+# **backward** compatible: we guarantee that the parameters of these functionals
+# won't be removed or renamed without a proper deprecation cycle. However, we
+# don't guarantee **forward** compatibility, and we may add new parameters in
+# the future.
+#
+# Imagine that in a future version, Torchvision adds a new ``inplace`` parameter
+# to its :func:`~torchvision.transforms.v2.functional.hflip` functional. If you
+# already defined and registered your own kernel as
+
+def hflip_my_tv_tensor(my_dp):  # noqa
+    print("Flipping!")
+    out = my_dp.flip(-1)
+    return tv_tensors.wrap(out, like=my_dp)
+
+
+# %%
+# then calling ``F.hflip(my_dp)`` will **fail**, because ``hflip`` will try to
+# pass the new ``inplace`` parameter to your kernel, but your kernel doesn't
+# accept it.
+#
+# For this reason, we recommend to always define your kernels with
+# ``*args, **kwargs`` in their signature, as done above. This way, your kernel
+# will be able to accept any new parameter that we may add in the future.
+# (Technically, adding `**kwargs` only should be enough).
diff --git a/0.16/_downloads/4b3151295a1f6b75ffd88837fb31f719/plot_video_api.ipynb b/0.16/_downloads/4b3151295a1f6b75ffd88837fb31f719/plot_video_api.ipynb
new file mode 100644
index 00000000000..ffba78fd53f
--- /dev/null
+++ b/0.16/_downloads/4b3151295a1f6b75ffd88837fb31f719/plot_video_api.ipynb
@@ -0,0 +1,298 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Video API\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_video_api.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_video_api.py>` to download the full example code.</p></div>\n\nThis example illustrates some of the APIs that torchvision offers for\nvideos, together with the examples on how to build datasets and more.\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 1. Introduction: building a new video object and examining the properties\nFirst we select a video to test the object out. For the sake of argument\nwe're using one from kinetics400 dataset.\nTo create it, we need to define the path and the stream we want to use.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Chosen video statistics:\n\n- WUzgd7C1pWA.mp4\n    - source:\n        - kinetics-400\n    - video:\n        - H-264\n        - MPEG-4 AVC (part 10) (avc1)\n        - fps: 29.97\n    - audio:\n        - MPEG AAC audio (mp4a)\n        - sample rate: 48K Hz\n\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nimport torchvision\nfrom torchvision.datasets.utils import download_url\ntorchvision.set_video_backend(\"video_reader\")\n\n# Download the sample video\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true\",\n    \".\",\n    \"WUzgd7C1pWA.mp4\"\n)\nvideo_path = \"./WUzgd7C1pWA.mp4\""
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Streams are defined in a similar fashion as torch devices. We encode them as strings in a form\nof ``stream_type:stream_id`` where ``stream_type`` is a string and ``stream_id`` a long int.\nThe constructor accepts passing a ``stream_type`` only, in which case the stream is auto-discovered.\nFirstly, let's get the metadata for our particular video:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "stream = \"video\"\nvideo = torchvision.io.VideoReader(video_path, stream)\nvideo.get_metadata()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Here we can see that video has two streams - a video and an audio stream.\nCurrently available stream types include ['video', 'audio'].\nEach descriptor consists of two parts: stream type (e.g. 'video') and a unique stream id\n(which are determined by video encoding).\nIn this way, if the video container contains multiple streams of the same type,\nusers can access the one they want.\nIf only stream type is passed, the decoder auto-detects first stream of that type and returns it.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let's read all the frames from the video stream. By default, the return value of\n``next(video_reader)`` is a dict containing the following fields.\n\nThe return fields are:\n\n- ``data``: containing a torch.tensor\n- ``pts``: containing a float timestamp of this particular frame\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "metadata = video.get_metadata()\nvideo.set_current_stream(\"audio\")\n\nframes = []  # we are going to save the frames here.\nptss = []  # pts is a presentation timestamp in seconds (float) of each frame\nfor frame in video:\n    frames.append(frame['data'])\n    ptss.append(frame['pts'])\n\nprint(\"PTS for first five frames \", ptss[:5])\nprint(\"Total number of frames: \", len(frames))\napprox_nf = metadata['audio']['duration'][0] * metadata['audio']['framerate'][0]\nprint(\"Approx total number of datapoints we can expect: \", approx_nf)\nprint(\"Read data size: \", frames[0].size(0) * len(frames))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "But what if we only want to read certain time segment of the video?\nThat can be done easily using the combination of our ``seek`` function, and the fact that each call\nto next returns the presentation timestamp of the returned frame in seconds.\n\nGiven that our implementation relies on python iterators,\nwe can leverage itertools to simplify the process and make it more pythonic.\n\nFor example, if we wanted to read ten frames from second second:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import itertools\nvideo.set_current_stream(\"video\")\n\nframes = []  # we are going to save the frames here.\n\n# We seek into a second second of the video and use islice to get 10 frames since\nfor frame, pts in itertools.islice(video.seek(2), 10):\n    frames.append(frame)\n\nprint(\"Total number of frames: \", len(frames))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Or if we wanted to read from 2nd to 5th second,\nWe seek into a second second of the video,\nthen we utilize the itertools takewhile to get the\ncorrect number of frames:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "video.set_current_stream(\"video\")\nframes = []  # we are going to save the frames here.\nvideo = video.seek(2)\n\nfor frame in itertools.takewhile(lambda x: x['pts'] <= 5, video):\n    frames.append(frame['data'])\n\nprint(\"Total number of frames: \", len(frames))\napprox_nf = (5 - 2) * video.get_metadata()['video']['fps'][0]\nprint(\"We can expect approx: \", approx_nf)\nprint(\"Tensor size: \", frames[0].size())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 2. Building a sample read_video function\nWe can utilize the methods above to build the read video function that follows\nthe same API to the existing ``read_video`` function.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "def example_read_video(video_object, start=0, end=None, read_video=True, read_audio=True):\n    if end is None:\n        end = float(\"inf\")\n    if end < start:\n        raise ValueError(\n            \"end time should be larger than start time, got \"\n            f\"start time={start} and end time={end}\"\n        )\n\n    video_frames = torch.empty(0)\n    video_pts = []\n    if read_video:\n        video_object.set_current_stream(\"video\")\n        frames = []\n        for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):\n            frames.append(frame['data'])\n            video_pts.append(frame['pts'])\n        if len(frames) > 0:\n            video_frames = torch.stack(frames, 0)\n\n    audio_frames = torch.empty(0)\n    audio_pts = []\n    if read_audio:\n        video_object.set_current_stream(\"audio\")\n        frames = []\n        for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):\n            frames.append(frame['data'])\n            audio_pts.append(frame['pts'])\n        if len(frames) > 0:\n            audio_frames = torch.cat(frames, 0)\n\n    return video_frames, audio_frames, (video_pts, audio_pts), video_object.get_metadata()\n\n\n# Total number of frames should be 327 for video and 523264 datapoints for audio\nvf, af, info, meta = example_read_video(video)\nprint(vf.size(), af.size())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 3. Building an example randomly sampled dataset (can be applied to training dataset of kinetics400)\nCool, so now we can use the same principle to make the sample dataset.\nWe suggest trying out iterable dataset for this purpose.\nHere, we are going to build an example dataset that reads randomly selected 10 frames of video.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Make sample dataset\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nos.makedirs(\"./dataset\", exist_ok=True)\nos.makedirs(\"./dataset/1\", exist_ok=True)\nos.makedirs(\"./dataset/2\", exist_ok=True)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Download the videos\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.datasets.utils import download_url\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true\",\n    \"./dataset/1\", \"WUzgd7C1pWA.mp4\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi?raw=true\",\n    \"./dataset/1\",\n    \"RATRACE_wave_f_nm_np1_fr_goo_37.avi\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/SOX5yA1l24A.mp4?raw=true\",\n    \"./dataset/2\",\n    \"SOX5yA1l24A.mp4\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi?raw=true\",\n    \"./dataset/2\",\n    \"v_SoccerJuggling_g23_c01.avi\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi?raw=true\",\n    \"./dataset/2\",\n    \"v_SoccerJuggling_g24_c01.avi\"\n)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Housekeeping and utilities\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nimport random\n\nfrom torchvision.datasets.folder import make_dataset\nfrom torchvision import transforms as t\n\n\ndef _find_classes(dir):\n    classes = [d.name for d in os.scandir(dir) if d.is_dir()]\n    classes.sort()\n    class_to_idx = {cls_name: i for i, cls_name in enumerate(classes)}\n    return classes, class_to_idx\n\n\ndef get_samples(root, extensions=(\".mp4\", \".avi\")):\n    _, class_to_idx = _find_classes(root)\n    return make_dataset(root, class_to_idx, extensions=extensions)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We are going to define the dataset and some basic arguments.\nWe assume the structure of the FolderDataset, and add the following parameters:\n\n- ``clip_len``: length of a clip in frames\n- ``frame_transform``: transform for every frame individually\n- ``video_transform``: transform on a video sequence\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>We actually add epoch size as using :func:`~torch.utils.data.IterableDataset`\n  class allows us to naturally oversample clips or images from each video if needed.</p></div>\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class RandomDataset(torch.utils.data.IterableDataset):\n    def __init__(self, root, epoch_size=None, frame_transform=None, video_transform=None, clip_len=16):\n        super(RandomDataset).__init__()\n\n        self.samples = get_samples(root)\n\n        # Allow for temporal jittering\n        if epoch_size is None:\n            epoch_size = len(self.samples)\n        self.epoch_size = epoch_size\n\n        self.clip_len = clip_len\n        self.frame_transform = frame_transform\n        self.video_transform = video_transform\n\n    def __iter__(self):\n        for i in range(self.epoch_size):\n            # Get random sample\n            path, target = random.choice(self.samples)\n            # Get video object\n            vid = torchvision.io.VideoReader(path, \"video\")\n            metadata = vid.get_metadata()\n            video_frames = []  # video frame buffer\n\n            # Seek and return frames\n            max_seek = metadata[\"video\"]['duration'][0] - (self.clip_len / metadata[\"video\"]['fps'][0])\n            start = random.uniform(0., max_seek)\n            for frame in itertools.islice(vid.seek(start), self.clip_len):\n                video_frames.append(self.frame_transform(frame['data']))\n                current_pts = frame['pts']\n            # Stack it into a tensor\n            video = torch.stack(video_frames, 0)\n            if self.video_transform:\n                video = self.video_transform(video)\n            output = {\n                'path': path,\n                'video': video,\n                'target': target,\n                'start': start,\n                'end': current_pts}\n            yield output"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Given a path of videos in a folder structure, i.e:\n\n- dataset\n    - class 1\n        - file 0\n        - file 1\n        - ...\n    - class 2\n        - file 0\n        - file 1\n        - ...\n    - ...\n\nWe can generate a dataloader and test the dataset.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transforms = [t.Resize((112, 112))]\nframe_transform = t.Compose(transforms)\n\ndataset = RandomDataset(\"./dataset\", epoch_size=None, frame_transform=frame_transform)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torch.utils.data import DataLoader\nloader = DataLoader(dataset, batch_size=12)\ndata = {\"video\": [], 'start': [], 'end': [], 'tensorsize': []}\nfor batch in loader:\n    for i in range(len(batch['path'])):\n        data['video'].append(batch['path'][i])\n        data['start'].append(batch['start'][i].item())\n        data['end'].append(batch['end'][i].item())\n        data['tensorsize'].append(batch['video'][i].size())\nprint(data)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 4. Data Visualization\nExample of visualized video\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import matplotlib.pyplot as plt\n\nplt.figure(figsize=(12, 12))\nfor i in range(16):\n    plt.subplot(4, 4, i + 1)\n    plt.imshow(batch[\"video\"][0, i, ...].permute(1, 2, 0))\n    plt.axis(\"off\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Cleanup the video and dataset:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nimport shutil\nos.remove(\"./WUzgd7C1pWA.mp4\")\nshutil.rmtree(\"./dataset\")"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/5a9ea474844f53cd409e782bd87efb13/plot_cutmix_mixup.py b/0.16/_downloads/5a9ea474844f53cd409e782bd87efb13/plot_cutmix_mixup.py
new file mode 100644
index 00000000000..d26b027b121
--- /dev/null
+++ b/0.16/_downloads/5a9ea474844f53cd409e782bd87efb13/plot_cutmix_mixup.py
@@ -0,0 +1,150 @@
+
+"""
+===========================
+How to use CutMix and MixUp
+===========================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_cutmix_mixup.py>` to download the full example code.
+
+:class:`~torchvision.transforms.v2.CutMix` and
+:class:`~torchvision.transforms.v2.MixUp` are popular augmentation strategies
+that can improve classification accuracy.
+
+These transforms are slightly different from the rest of the Torchvision
+transforms, because they expect
+**batches** of samples as input, not individual images. In this example we'll
+explain how to use them: after the ``DataLoader``, or as part of a collation
+function.
+"""
+
+# %%
+import torch
+from torchvision.datasets import FakeData
+from torchvision.transforms import v2
+
+
+NUM_CLASSES = 100
+
+# %%
+# Pre-processing pipeline
+# -----------------------
+#
+# We'll use a simple but typical image classification pipeline:
+
+preproc = v2.Compose([
+    v2.PILToTensor(),
+    v2.RandomResizedCrop(size=(224, 224), antialias=True),
+    v2.RandomHorizontalFlip(p=0.5),
+    v2.ToDtype(torch.float32, scale=True),  # to float32 in [0, 1]
+    v2.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),  # typically from ImageNet
+])
+
+dataset = FakeData(size=1000, num_classes=NUM_CLASSES, transform=preproc)
+
+img, label = dataset[0]
+print(f"{type(img) = }, {img.dtype = }, {img.shape = }, {label = }")
+
+# %%
+#
+# One important thing to note is that neither CutMix nor MixUp are part of this
+# pre-processing pipeline. We'll add them a bit later once we define the
+# DataLoader. Just as a refresher, this is what the DataLoader and training loop
+# would look like if we weren't using CutMix or MixUp:
+
+from torch.utils.data import DataLoader
+
+dataloader = DataLoader(dataset, batch_size=4, shuffle=True)
+
+for images, labels in dataloader:
+    print(f"{images.shape = }, {labels.shape = }")
+    print(labels.dtype)
+    # <rest of the training loop here>
+    break
+# %%
+
+# %%
+# Where to use MixUp and CutMix
+# -----------------------------
+#
+# After the DataLoader
+# ^^^^^^^^^^^^^^^^^^^^
+#
+# Now let's add CutMix and MixUp. The simplest way to do this right after the
+# DataLoader: the Dataloader has already batched the images and labels for us,
+# and this is exactly what these transforms expect as input:
+
+dataloader = DataLoader(dataset, batch_size=4, shuffle=True)
+
+cutmix = v2.CutMix(num_classes=NUM_CLASSES)
+mixup = v2.MixUp(num_classes=NUM_CLASSES)
+cutmix_or_mixup = v2.RandomChoice([cutmix, mixup])
+
+for images, labels in dataloader:
+    print(f"Before CutMix/MixUp: {images.shape = }, {labels.shape = }")
+    images, labels = cutmix_or_mixup(images, labels)
+    print(f"After CutMix/MixUp: {images.shape = }, {labels.shape = }")
+
+    # <rest of the training loop here>
+    break
+# %%
+#
+# Note how the labels were also transformed: we went from a batched label of
+# shape (batch_size,) to a tensor of shape (batch_size, num_classes). The
+# transformed labels can still be passed as-is to a loss function like
+# :func:`torch.nn.functional.cross_entropy`.
+#
+# As part of the collation function
+# ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+#
+# Passing the transforms after the DataLoader is the simplest way to use CutMix
+# and MixUp, but one disadvantage is that it does not take advantage of the
+# DataLoader multi-processing. For that, we can pass those transforms as part of
+# the collation function (refer to the `PyTorch docs
+# <https://pytorch.org/docs/stable/data.html#dataloader-collate-fn>`_ to learn
+# more about collation).
+
+from torch.utils.data import default_collate
+
+
+def collate_fn(batch):
+    return cutmix_or_mixup(*default_collate(batch))
+
+
+dataloader = DataLoader(dataset, batch_size=4, shuffle=True, num_workers=2, collate_fn=collate_fn)
+
+for images, labels in dataloader:
+    print(f"{images.shape = }, {labels.shape = }")
+    # No need to call cutmix_or_mixup, it's already been called as part of the DataLoader!
+    # <rest of the training loop here>
+    break
+
+# %%
+# Non-standard input format
+# -------------------------
+#
+# So far we've used a typical sample structure where we pass ``(images,
+# labels)`` as inputs. MixUp and CutMix will magically work by default with most
+# common sample structures: tuples where the second parameter is a tensor label,
+# or dict with a "label[s]" key. Look at the documentation of the
+# ``labels_getter`` parameter for more details.
+#
+# If your samples have a different structure, you can still use CutMix and MixUp
+# by passing a callable to the ``labels_getter`` parameter. For example:
+
+batch = {
+    "imgs": torch.rand(4, 3, 224, 224),
+    "target": {
+        "classes": torch.randint(0, NUM_CLASSES, size=(4,)),
+        "some_other_key": "this is going to be passed-through"
+    }
+}
+
+
+def labels_getter(batch):
+    return batch["target"]["classes"]
+
+
+out = v2.CutMix(num_classes=NUM_CLASSES, labels_getter=labels_getter)(batch)
+print(f"{out['imgs'].shape = }, {out['target']['classes'].shape = }")
diff --git a/0.16/_downloads/6159c523793cb1afbb503259bd141f91/plot_repurposing_annotations.py b/0.16/_downloads/6159c523793cb1afbb503259bd141f91/plot_repurposing_annotations.py
new file mode 100644
index 00000000000..b1617cacd99
--- /dev/null
+++ b/0.16/_downloads/6159c523793cb1afbb503259bd141f91/plot_repurposing_annotations.py
@@ -0,0 +1,211 @@
+"""
+=====================================
+Repurposing masks into bounding boxes
+=====================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_repurposing_annotations.py>` to download the full example code.
+
+The following example illustrates the operations available
+the :ref:`torchvision.ops <ops>` module for repurposing
+segmentation masks into object localization annotations for different tasks
+(e.g. transforming masks used by instance and panoptic segmentation
+methods into bounding boxes used by object detection methods).
+"""
+
+# sphinx_gallery_thumbnail_path = "../../gallery/assets/repurposing_annotations_thumbnail.png"
+
+import os
+import numpy as np
+import torch
+import matplotlib.pyplot as plt
+
+import torchvision.transforms.functional as F
+
+
+ASSETS_DIRECTORY = "../assets"
+
+plt.rcParams["savefig.bbox"] = "tight"
+
+
+def show(imgs):
+    if not isinstance(imgs, list):
+        imgs = [imgs]
+    fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)
+    for i, img in enumerate(imgs):
+        img = img.detach()
+        img = F.to_pil_image(img)
+        axs[0, i].imshow(np.asarray(img))
+        axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
+
+
+# %%
+# Masks
+# -----
+# In tasks like instance and panoptic segmentation, masks are commonly defined, and are defined by this package,
+# as a multi-dimensional array (e.g. a NumPy array or a PyTorch tensor) with the following shape:
+#
+#       (num_objects, height, width)
+#
+# Where num_objects is the number of annotated objects in the image. Each (height, width) object corresponds to exactly
+# one object. For example, if your input image has the dimensions 224 x 224 and has four annotated objects the shape
+# of your masks annotation has the following shape:
+#
+#       (4, 224, 224).
+#
+# A nice property of masks is that they can be easily repurposed to be used in methods to solve a variety of object
+# localization tasks.
+
+# %%
+# Converting Masks to Bounding Boxes
+# -----------------------------------------------
+# For example, the :func:`~torchvision.ops.masks_to_boxes` operation can be used to
+# transform masks into bounding boxes that can be
+# used as input to detection models such as FasterRCNN and RetinaNet.
+# We will take images and masks from the `PenFudan Dataset <https://www.cis.upenn.edu/~jshi/ped_html/>`_.
+
+
+from torchvision.io import read_image
+
+img_path = os.path.join(ASSETS_DIRECTORY, "FudanPed00054.png")
+mask_path = os.path.join(ASSETS_DIRECTORY, "FudanPed00054_mask.png")
+img = read_image(img_path)
+mask = read_image(mask_path)
+
+
+# %%
+# Here the masks are represented as a PNG Image, with floating point values.
+# Each pixel is encoded as different colors, with 0 being background.
+# Notice that the spatial dimensions of image and mask match.
+
+print(mask.size())
+print(img.size())
+print(mask)
+
+# %%
+
+# We get the unique colors, as these would be the object ids.
+obj_ids = torch.unique(mask)
+
+# first id is the background, so remove it.
+obj_ids = obj_ids[1:]
+
+# split the color-encoded mask into a set of boolean masks.
+# Note that this snippet would work as well if the masks were float values instead of ints.
+masks = mask == obj_ids[:, None, None]
+
+# %%
+# Now the masks are a boolean tensor.
+# The first dimension in this case 3 and denotes the number of instances: there are 3 people in the image.
+# The other two dimensions are height and width, which are equal to the dimensions of the image.
+# For each instance, the boolean tensors represent if the particular pixel
+# belongs to the segmentation mask of the image.
+
+print(masks.size())
+print(masks)
+
+# %%
+# Let us visualize an image and plot its corresponding segmentation masks.
+# We will use the :func:`~torchvision.utils.draw_segmentation_masks` to draw the segmentation masks.
+
+from torchvision.utils import draw_segmentation_masks
+
+drawn_masks = []
+for mask in masks:
+    drawn_masks.append(draw_segmentation_masks(img, mask, alpha=0.8, colors="blue"))
+
+show(drawn_masks)
+
+# %%
+# To convert the boolean masks into bounding boxes.
+# We will use the :func:`~torchvision.ops.masks_to_boxes` from the torchvision.ops module
+# It returns the boxes in ``(xmin, ymin, xmax, ymax)`` format.
+
+from torchvision.ops import masks_to_boxes
+
+boxes = masks_to_boxes(masks)
+print(boxes.size())
+print(boxes)
+
+# %%
+# As the shape denotes, there are 3 boxes and in ``(xmin, ymin, xmax, ymax)`` format.
+# These can be visualized very easily with :func:`~torchvision.utils.draw_bounding_boxes` utility
+# provided in :ref:`torchvision.utils <utils>`.
+
+from torchvision.utils import draw_bounding_boxes
+
+drawn_boxes = draw_bounding_boxes(img, boxes, colors="red")
+show(drawn_boxes)
+
+# %%
+# These boxes can now directly be used by detection models in torchvision.
+# Here is demo with a Faster R-CNN model loaded from
+# :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`
+
+from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights
+
+weights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT
+model = fasterrcnn_resnet50_fpn(weights=weights, progress=False)
+print(img.size())
+
+tranforms = weights.transforms()
+img = tranforms(img)
+target = {}
+target["boxes"] = boxes
+target["labels"] = labels = torch.ones((masks.size(0),), dtype=torch.int64)
+detection_outputs = model(img.unsqueeze(0), [target])
+
+
+# %%
+# Converting Segmentation Dataset to Detection Dataset
+# ----------------------------------------------------
+#
+# With this utility it becomes very simple to convert a segmentation dataset to a detection dataset.
+# With this we can now use a segmentation dataset to train a detection model.
+# One can similarly convert panoptic dataset to detection dataset.
+# Here is an example where we re-purpose the dataset from the
+# `PenFudan Detection Tutorial <https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html>`_.
+
+class SegmentationToDetectionDataset(torch.utils.data.Dataset):
+    def __init__(self, root, transforms):
+        self.root = root
+        self.transforms = transforms
+        # load all image files, sorting them to
+        # ensure that they are aligned
+        self.imgs = list(sorted(os.listdir(os.path.join(root, "PNGImages"))))
+        self.masks = list(sorted(os.listdir(os.path.join(root, "PedMasks"))))
+
+    def __getitem__(self, idx):
+        # load images and masks
+        img_path = os.path.join(self.root, "PNGImages", self.imgs[idx])
+        mask_path = os.path.join(self.root, "PedMasks", self.masks[idx])
+
+        img = read_image(img_path)
+        mask = read_image(mask_path)
+
+        img = F.convert_image_dtype(img, dtype=torch.float)
+        mask = F.convert_image_dtype(mask, dtype=torch.float)
+
+        # We get the unique colors, as these would be the object ids.
+        obj_ids = torch.unique(mask)
+
+        # first id is the background, so remove it.
+        obj_ids = obj_ids[1:]
+
+        # split the color-encoded mask into a set of boolean masks.
+        masks = mask == obj_ids[:, None, None]
+
+        boxes = masks_to_boxes(masks)
+
+        # there is only one class
+        labels = torch.ones((masks.shape[0],), dtype=torch.int64)
+
+        target = {}
+        target["boxes"] = boxes
+        target["labels"] = labels
+
+        if self.transforms is not None:
+            img, target = self.transforms(img, target)
+
+        return img, target
diff --git a/0.16/_downloads/66c906e53334aa968837db20de95f3b2/plot_cutmix_mixup.ipynb b/0.16/_downloads/66c906e53334aa968837db20de95f3b2/plot_cutmix_mixup.ipynb
new file mode 100644
index 00000000000..d4eba2316ac
--- /dev/null
+++ b/0.16/_downloads/66c906e53334aa968837db20de95f3b2/plot_cutmix_mixup.ipynb
@@ -0,0 +1,133 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# How to use CutMix and MixUp\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_cutmix_mixup.py>` to download the full example code.</p></div>\n\n:class:`~torchvision.transforms.v2.CutMix` and\n:class:`~torchvision.transforms.v2.MixUp` are popular augmentation strategies\nthat can improve classification accuracy.\n\nThese transforms are slightly different from the rest of the Torchvision\ntransforms, because they expect\n**batches** of samples as input, not individual images. In this example we'll\nexplain how to use them: after the ``DataLoader``, or as part of a collation\nfunction.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nfrom torchvision.datasets import FakeData\nfrom torchvision.transforms import v2\n\n\nNUM_CLASSES = 100"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Pre-processing pipeline\n\nWe'll use a simple but typical image classification pipeline:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "preproc = v2.Compose([\n    v2.PILToTensor(),\n    v2.RandomResizedCrop(size=(224, 224), antialias=True),\n    v2.RandomHorizontalFlip(p=0.5),\n    v2.ToDtype(torch.float32, scale=True),  # to float32 in [0, 1]\n    v2.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),  # typically from ImageNet\n])\n\ndataset = FakeData(size=1000, num_classes=NUM_CLASSES, transform=preproc)\n\nimg, label = dataset[0]\nprint(f\"{type(img) = }, {img.dtype = }, {img.shape = }, {label = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "One important thing to note is that neither CutMix nor MixUp are part of this\npre-processing pipeline. We'll add them a bit later once we define the\nDataLoader. Just as a refresher, this is what the DataLoader and training loop\nwould look like if we weren't using CutMix or MixUp:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torch.utils.data import DataLoader\n\ndataloader = DataLoader(dataset, batch_size=4, shuffle=True)\n\nfor images, labels in dataloader:\n    print(f\"{images.shape = }, {labels.shape = }\")\n    print(labels.dtype)\n    # <rest of the training loop here>\n    break"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Where to use MixUp and CutMix\n\n### After the DataLoader\n\nNow let's add CutMix and MixUp. The simplest way to do this right after the\nDataLoader: the Dataloader has already batched the images and labels for us,\nand this is exactly what these transforms expect as input:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dataloader = DataLoader(dataset, batch_size=4, shuffle=True)\n\ncutmix = v2.CutMix(num_classes=NUM_CLASSES)\nmixup = v2.MixUp(num_classes=NUM_CLASSES)\ncutmix_or_mixup = v2.RandomChoice([cutmix, mixup])\n\nfor images, labels in dataloader:\n    print(f\"Before CutMix/MixUp: {images.shape = }, {labels.shape = }\")\n    images, labels = cutmix_or_mixup(images, labels)\n    print(f\"After CutMix/MixUp: {images.shape = }, {labels.shape = }\")\n\n    # <rest of the training loop here>\n    break"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Note how the labels were also transformed: we went from a batched label of\nshape (batch_size,) to a tensor of shape (batch_size, num_classes). The\ntransformed labels can still be passed as-is to a loss function like\n:func:`torch.nn.functional.cross_entropy`.\n\n### As part of the collation function\n\nPassing the transforms after the DataLoader is the simplest way to use CutMix\nand MixUp, but one disadvantage is that it does not take advantage of the\nDataLoader multi-processing. For that, we can pass those transforms as part of\nthe collation function (refer to the [PyTorch docs](https://pytorch.org/docs/stable/data.html#dataloader-collate-fn) to learn\nmore about collation).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torch.utils.data import default_collate\n\n\ndef collate_fn(batch):\n    return cutmix_or_mixup(*default_collate(batch))\n\n\ndataloader = DataLoader(dataset, batch_size=4, shuffle=True, num_workers=2, collate_fn=collate_fn)\n\nfor images, labels in dataloader:\n    print(f\"{images.shape = }, {labels.shape = }\")\n    # No need to call cutmix_or_mixup, it's already been called as part of the DataLoader!\n    # <rest of the training loop here>\n    break"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Non-standard input format\n\nSo far we've used a typical sample structure where we pass ``(images,\nlabels)`` as inputs. MixUp and CutMix will magically work by default with most\ncommon sample structures: tuples where the second parameter is a tensor label,\nor dict with a \"label[s]\" key. Look at the documentation of the\n``labels_getter`` parameter for more details.\n\nIf your samples have a different structure, you can still use CutMix and MixUp\nby passing a callable to the ``labels_getter`` parameter. For example:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "batch = {\n    \"imgs\": torch.rand(4, 3, 224, 224),\n    \"target\": {\n        \"classes\": torch.randint(0, NUM_CLASSES, size=(4,)),\n        \"some_other_key\": \"this is going to be passed-through\"\n    }\n}\n\n\ndef labels_getter(batch):\n    return batch[\"target\"][\"classes\"]\n\n\nout = v2.CutMix(num_classes=NUM_CLASSES, labels_getter=labels_getter)(batch)\nprint(f\"{out['imgs'].shape = }, {out['target']['classes'].shape = }\")"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/6acd2f50f8be085d0a8ac5874560f268/plot_optical_flow.py b/0.16/_downloads/6acd2f50f8be085d0a8ac5874560f268/plot_optical_flow.py
new file mode 100644
index 00000000000..3ab14493417
--- /dev/null
+++ b/0.16/_downloads/6acd2f50f8be085d0a8ac5874560f268/plot_optical_flow.py
@@ -0,0 +1,198 @@
+"""
+=====================================================
+Optical Flow: Predicting movement with the RAFT model
+=====================================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_optical_flow.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_optical_flow.py>` to download the full example code.
+
+Optical flow is the task of predicting movement between two images, usually two
+consecutive frames of a video. Optical flow models take two images as input, and
+predict a flow: the flow indicates the displacement of every single pixel in the
+first image, and maps it to its corresponding pixel in the second image. Flows
+are (2, H, W)-dimensional tensors, where the first axis corresponds to the
+predicted horizontal and vertical displacements.
+
+The following example illustrates how torchvision can be used to predict flows
+using our implementation of the RAFT model. We will also see how to convert the
+predicted flows to RGB images for visualization.
+"""
+
+import numpy as np
+import torch
+import matplotlib.pyplot as plt
+import torchvision.transforms.functional as F
+
+
+plt.rcParams["savefig.bbox"] = "tight"
+# sphinx_gallery_thumbnail_number = 2
+
+
+def plot(imgs, **imshow_kwargs):
+    if not isinstance(imgs[0], list):
+        # Make a 2d grid even if there's just 1 row
+        imgs = [imgs]
+
+    num_rows = len(imgs)
+    num_cols = len(imgs[0])
+    _, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)
+    for row_idx, row in enumerate(imgs):
+        for col_idx, img in enumerate(row):
+            ax = axs[row_idx, col_idx]
+            img = F.to_pil_image(img.to("cpu"))
+            ax.imshow(np.asarray(img), **imshow_kwargs)
+            ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
+
+    plt.tight_layout()
+
+# %%
+# Reading Videos Using Torchvision
+# --------------------------------
+# We will first read a video using :func:`~torchvision.io.read_video`.
+# Alternatively one can use the new :class:`~torchvision.io.VideoReader` API (if
+# torchvision is built from source).
+# The video we will use here is free of use from `pexels.com
+# <https://www.pexels.com/video/a-man-playing-a-game-of-basketball-5192157/>`_,
+# credits go to `Pavel Danilyuk <https://www.pexels.com/@pavel-danilyuk>`_.
+
+
+import tempfile
+from pathlib import Path
+from urllib.request import urlretrieve
+
+
+video_url = "https://download.pytorch.org/tutorial/pexelscom_pavel_danilyuk_basketball_hd.mp4"
+video_path = Path(tempfile.mkdtemp()) / "basketball.mp4"
+_ = urlretrieve(video_url, video_path)
+
+# %%
+# :func:`~torchvision.io.read_video` returns the video frames, audio frames and
+# the metadata associated with the video. In our case, we only need the video
+# frames.
+#
+# Here we will just make 2 predictions between 2 pre-selected pairs of frames,
+# namely frames (100, 101) and (150, 151). Each of these pairs corresponds to a
+# single model input.
+
+from torchvision.io import read_video
+frames, _, _ = read_video(str(video_path), output_format="TCHW")
+
+img1_batch = torch.stack([frames[100], frames[150]])
+img2_batch = torch.stack([frames[101], frames[151]])
+
+plot(img1_batch)
+
+# %%
+# The RAFT model accepts RGB images. We first get the frames from
+# :func:`~torchvision.io.read_video` and resize them to ensure their dimensions
+# are divisible by 8. Note that we explicitly use ``antialias=False``, because
+# this is how those models were trained. Then we use the transforms bundled into
+# the weights in order to preprocess the input and rescale its values to the
+# required ``[-1, 1]`` interval.
+
+from torchvision.models.optical_flow import Raft_Large_Weights
+
+weights = Raft_Large_Weights.DEFAULT
+transforms = weights.transforms()
+
+
+def preprocess(img1_batch, img2_batch):
+    img1_batch = F.resize(img1_batch, size=[520, 960], antialias=False)
+    img2_batch = F.resize(img2_batch, size=[520, 960], antialias=False)
+    return transforms(img1_batch, img2_batch)
+
+
+img1_batch, img2_batch = preprocess(img1_batch, img2_batch)
+
+print(f"shape = {img1_batch.shape}, dtype = {img1_batch.dtype}")
+
+
+# %%
+# Estimating Optical flow using RAFT
+# ----------------------------------
+# We will use our RAFT implementation from
+# :func:`~torchvision.models.optical_flow.raft_large`, which follows the same
+# architecture as the one described in the `original paper <https://arxiv.org/abs/2003.12039>`_.
+# We also provide the :func:`~torchvision.models.optical_flow.raft_small` model
+# builder, which is smaller and faster to run, sacrificing a bit of accuracy.
+
+from torchvision.models.optical_flow import raft_large
+
+# If you can, run this example on a GPU, it will be a lot faster.
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+model = raft_large(weights=Raft_Large_Weights.DEFAULT, progress=False).to(device)
+model = model.eval()
+
+list_of_flows = model(img1_batch.to(device), img2_batch.to(device))
+print(f"type = {type(list_of_flows)}")
+print(f"length = {len(list_of_flows)} = number of iterations of the model")
+
+# %%
+# The RAFT model outputs lists of predicted flows where each entry is a
+# (N, 2, H, W) batch of predicted flows that corresponds to a given "iteration"
+# in the model. For more details on the iterative nature of the model, please
+# refer to the `original paper <https://arxiv.org/abs/2003.12039>`_. Here, we
+# are only interested in the final predicted flows (they are the most accurate
+# ones), so we will just retrieve the last item in the list.
+#
+# As described above, a flow is a tensor with dimensions (2, H, W) (or (N, 2, H,
+# W) for batches of flows) where each entry corresponds to the horizontal and
+# vertical displacement of each pixel from the first image to the second image.
+# Note that the predicted flows are in "pixel" unit, they are not normalized
+# w.r.t. the dimensions of the images.
+predicted_flows = list_of_flows[-1]
+print(f"dtype = {predicted_flows.dtype}")
+print(f"shape = {predicted_flows.shape} = (N, 2, H, W)")
+print(f"min = {predicted_flows.min()}, max = {predicted_flows.max()}")
+
+
+# %%
+# Visualizing predicted flows
+# ---------------------------
+# Torchvision provides the :func:`~torchvision.utils.flow_to_image` utility to
+# convert a flow into an RGB image. It also supports batches of flows.
+# each "direction" in the flow will be mapped to a given RGB color. In the
+# images below, pixels with similar colors are assumed by the model to be moving
+# in similar directions. The model is properly able to predict the movement of
+# the ball and the player. Note in particular the different predicted direction
+# of the ball in the first image (going to the left) and in the second image
+# (going up).
+
+from torchvision.utils import flow_to_image
+
+flow_imgs = flow_to_image(predicted_flows)
+
+# The images have been mapped into [-1, 1] but for plotting we want them in [0, 1]
+img1_batch = [(img1 + 1) / 2 for img1 in img1_batch]
+
+grid = [[img1, flow_img] for (img1, flow_img) in zip(img1_batch, flow_imgs)]
+plot(grid)
+
+# %%
+# Bonus: Creating GIFs of predicted flows
+# ---------------------------------------
+# In the example above we have only shown the predicted flows of 2 pairs of
+# frames. A fun way to apply the Optical Flow models is to run the model on an
+# entire video, and create a new video from all the predicted flows. Below is a
+# snippet that can get you started with this. We comment out the code, because
+# this example is being rendered on a machine without a GPU, and it would take
+# too long to run it.
+
+# from torchvision.io import write_jpeg
+# for i, (img1, img2) in enumerate(zip(frames, frames[1:])):
+#     # Note: it would be faster to predict batches of flows instead of individual flows
+#     img1, img2 = preprocess(img1, img2)
+
+#     list_of_flows = model(img1.to(device), img2.to(device))
+#     predicted_flow = list_of_flows[-1][0]
+#     flow_img = flow_to_image(predicted_flow).to("cpu")
+#     output_folder = "/tmp/"  # Update this to the folder of your choice
+#     write_jpeg(flow_img, output_folder + f"predicted_flow_{i}.jpg")
+
+# %%
+# Once the .jpg flow images are saved, you can convert them into a video or a
+# GIF using ffmpeg with e.g.:
+#
+# ffmpeg -f image2 -framerate 30 -i predicted_flow_%d.jpg -loop -1 flow.gif
diff --git a/0.16/_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip b/0.16/_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip
new file mode 100644
index 0000000000000000000000000000000000000000..c2b311d77358eebd3843e05186c0eab6ef91da60
GIT binary patch
literal 144297
zcmeFa&vP7CmhYzsztvm&IpT#KjvVk!puhnN1W2~j96~b`CAK)zq8JpqHDiELNK_?&
z5(`x|l~n*CS#RsD?WLEFcn(K+>!l;S@Yde?kMQ31!W)0S=iGZUv#LN6LN{7aH$x(U
z%zJO%ALpLmKVJUhfAp{Z<B#;G`(J+k?f>*|KK~#8kGmgzr2qeOo{gp_#iZOE4@T3@
zXguwA^FilmFgkDd#}~uHAO3Zgt#pgQpj>&B?eR;N>5=}}<IePAT<~h;G@rcbjn0QF
z>+$tzG0l7VH0SLPKc=VUXg2AphYw~0?GNu{U;C%CXMFlm_Ht76`rT=Nc$}S%&Wh7w
zIL*%c)02$JX0M(++vRbu80-)ChmU*xv#dMF%W`LbB_9-%X=Z<K^oK{I{gqE2pFH^V
z#b{bQ-sFeJ<4<=d7ujf-?R7_kL4NqUbuyif%SW4={GK1SCq-H0lkQ3Tcr-d56z%Tl
zbn}=G&ki@o7t;}MHqZKHe>B`Y9E=V(k54wn`EgNhp631GX6LvV7L$Bh^g6bAogv>n
z9F5*wy1i@rL+!w5k{uk6vMK+uzGB$R9+%^jx1Hm`q@zt8jPhP5pG`-d;%$C99u#HA
zkFB$KY~zbh4>VHjXG0#%27@g0&AMz|`|;-Z)5n|a-KW~bxcfTztY2E6X<oj`Mn~Cr
z$ypp0)AOPjX4CUg)<4w_tY@=wmJbFOdbU68j)rB?olW~^MRqjFPm5AR<=I)kSB%=}
zwC#I#lzcOPQ`82`@-piW$Fu2rmJfUT!!QlTP^U+}ESv5SP+`-3*1di?9^_pI89rH@
z6&!d8Bn^se+<#jP7*(D3he!QMIrZt=H?q@w%r~dnz5cXhpC*9Ac*OX?VZCq0+IZ}J
z+Xl1c{xF{uS!;Ve`*J<|c5S2A2WkZ=`5>F}(`dpTopaY%B!}!D&)>4y)puBefk0Yq
z(aTOolm7Rk;gn(6hm7Ue9nI<sE!#T8T@4FVkIpsUc-sBJU{+2Al_g`H^96v|VL0pN
z!|br&<oFI>j$5n^W<4|QTE=~4oxq$fVLoL+I!P^|x`*xT+k*QEjOBxJ#O@T~Q><;M
z8T*EZ@%fm2L7#l}o6iCOG4%=HAt361Z{J~2D*>6jwtoW_S&o1N)fMt7-Wv3chMn$c
zHk^WQhFpP!h3dE|d_o9KC}GTI(ENy%6ya5cy+*Ss5NWvZbsB6LN@K@HEPFUR9bX8N
zhvWTWcxDHXeml*ly2FG1VS9X`pVWFhn5G7`^J;U9!;*bEE!#)4A^6dtiw@6p{y@!O
z+MaY@=3KaPZ-1rC&x)h|ar^La^mc#cb+(i3uT1;LC)52EZ{I7p-h8vwKRpII@80d7
zmcU5o&3QgKF4rD8cIqExATrQGe^^fQVYjg6d-q<iXM=t@jm?C}?qpx-7U$V^FFT&}
zd%AwZi~)d`NwHdH-vNu+RyG-(``GHFk2AMD{rg%I(?fQyLo2-)6hr%BK$d#LN!8yl
zm+-Q~O@6Ca+KO^^sMVFNp+?$?L#}6_r-A>jAJ((-UuH$|eX;XQhaN@}IMS`t@4a2u
z4?-ISLl}bh`Goz^bk&-%yfq7Hwx)Z1m((GD%OsiVUiwDp&JO5({G+jUvS)2@|G3Zi
zLbm!k4_f;x-SKRHWi8A&cFEti{Ty4vaXT;b$t1r}m$k(cZhni7Wif5N1;M<57#8f&
z9*2A}_q)BXz0!|N!N=0R8;o@Z`32XdwRVL&d>tm@&YkR4ksGT1y@*Vi{m$^$ZiVmm
zhY?LV_MOl^(j<_eu-zlwt49a_zJBZaBiI56u+grA_Wtn6U|LLu;Ha|#dN&#thDgDp
zh8Bm#dG^Q{lErasW}0R(Imn*8e4e%XNBhIt$3%P&XZ^txsyRB%9JQ?RO(A!iKz1Hb
zt-fdEgf2%DP%da$&#dF#7z{Wl#lKW6&iVOyyLt{I!!~W^8>jhj1Dbik`Zw~8V~F*}
z=xE~*@?T63A?X_rw|>37_3-1(wRH%)xB-ZxvnzWqK{12u^L*GJT+H5F`l-JktKCK~
zw9|raKXRWIr{kmkK%9imYMf6`K;t5oLj5oKHN2Tk2Ey-?0yJYBhc)2oq?k_nph(BY
zb{-ubbK&<_B3~N$ILL#+NqaOo-kib<PWt&^(~px=?${1@;tqGJ<LjLC+Na|O!ukFM
zO_Wb)f~~lQ_UW6Rer~O;Wt)7x>K^-c=u(?<E6sM@;gShVPj2e6zyF$M?Coqabf(be
z8u(BM#H`mJh2NaF6gLxD0_bl!>h{IOS0aZE(9WL2`@jx$^Rj@K<HilS|HDGuQ@ox!
za<RJx;a`f|i}%$?->JyscI336-(2^=OWV_pvKSOyeS4h4p6X8fcwD_9<MO@0PPex1
z-CNJL?rp6ZU)b7us9zs$t+lf+^6p7s(!d8k&>WMQ@57rKPi)+`bK1B2R@x=3q0%;o
z9PVO;zT@$%XK)Gv>FQMrE;&H8f$}<MEpD$PhAW>UUG6^p@>|HQZY$(;i#s-Dqj#L8
zZo?aO-?aApc=k9QNN4=z;l0-)k@VVrPp_?duPy5(zRr2H@X4FHWuiG1lb_|?ZZQV+
zS3JA%{Z31d3uE#lbm}nnhaYfFCDR~f^}jdbc`BIzVQJ=4Qh$P&6U3+>U?H)>h#n%$
z9bROgv@?nJ+`b(B0Fv|D@u1)BPq`(cga-%ta0-^}=cTiD2M6n1w{Ff$sO|}blmAR8
zIvLSE2%_gEqfL<Li(Wfp6NVaF1L~GnJUE1U*c)(F`%rL=r@JopT7oMwJw!f8RpZI1
zTa=EAOs2If%d%0n6eeq)<pXGgnlsG)P^6*{rDX|w8(ZA#*9QmsOfjM1FYKE7uD9#D
zr+9B)<ww)bZ;?fc&bPj^x_fbxce2Iz?a#k>_T+cJ-Q_rvti%lsU1|bU%dYYM&_tj~
z5l%>Ka=2vcl1svEjm&%XmNB16Sp?iR`sl&V-otHh@vna+3@`_Qd{u0EHT2v-;e&`I
zA2yIvxf)P5W>bA8thQP}eR<3Q(fmDIpq(F2fWg)glI2N0F6j0{HDKGG{kWd>q!OHa
zXU~2Fcix;Z@I^UASTYr2L%ZdCzR+4F4(M)+up6?2X4I7(mt(RefZLu3Vh4g)P|f*C
zANt@b8Nd7`BwUK3?n!?NnHQQvdKc!96ww1AO!|k0QS9L!Ox|%{8l*A%J*n91eDbz`
zW}IMtSZ;3LyLZ36wSDjYuf<q@%b3!_aGqjYY@lU`_|oZ@r!f5oQNt=}1WXoJ1H!cT
z;5MP|qr3zu32#kiL!dTi0Y{)<M5#PGM3)8w1KpTS^6tf-58kSvK)2>j_WUSA7M7HV
zDq$w|Bik4ibSTvF>^Coe$BmSd8zmYlxjd69$aj~0>lJ7Ht}YcK|E!nqub56nl5JbR
zxBF!$Kg;_F?SrDVmLd5SYAWr3W?)B6?`mGHfDv~tN1^M2pkr`cAf9VdIrv1tCUHpn
zw_QM%9HUAj)EOO_2|;=odtXCJwL$7AsP>q_6P}5vME|xLL$3*&)LNn>IySN92dRtZ
znlcYGH-`W!;K<EG8)~dKzv=GT9B&Re?efVQTNQMQrb{!1e_l=MO5;LIJt|nV*#Hmx
zd+Wt|7PZZpv!bkLzO%7L^UQyCjOt<tgrvO?oVZ@xcLSw18m-e22&#u2r;oY~IMrr_
zF6i~qfCz!rHxZZ_6i7$nV2&IwsLxNpW0_<p30baJH%ZGFASCDtK^{4BbPENKU9a5Y
zn`m8~hh!1R+>V2$j0U1D(S(5@8-iIB#chpPlhAh}4~vo00t8QnzMhmdES(=>n}WpK
zXYEvirr!m}igm5b0$Gb~O(zGkUNhC99iQ7sYzb?@Qcyeexii$Pnx+B7`(?HPosSLU
z1`L;y=v=dY1h%tj)z}|athGKi4_Mk>c4vRZ#swqIhJ9=`Y3rpj9!hdTonK(TId4ze
z)3)tqZDRuPxDCm6_zE1t&r9W>HYnk~1{Q*kN41+UTjG-Dy0sIlRBN2rt@*~zbAjFx
zNMU+vBF0k=2}W(cCy!feaBHWgh+cf1zvcDK;Y75zzYn$_@eXOWexp%;R$5$<B{4S6
zH!b6AFgv39)J5u`HaFNh$n>y3#XRZONwZ4Dn~VGgH&Y3=XnJHtmMLpCma+~ljb#1W
z4Fy*1R_(9!5Y)^RBdYA^LZ42g9IN6O)TB=Yyek13ZU1Ou8YOOwGy^iCLiy<l5|y2@
zbSzT#Poe8(UBHBBSPcyoc8H!NVBJh7#f%T%fVG_cGh7}!rumg7%!cOB5cUzK5?NVX
zgXR}@Jq#Mm?jtFv`HD2b#`5VGsf!DyZzzIAk8@ciXPT#8_xeX*e;MtogG|!^1v}Zb
z5#x?kw%Dw7JQ55rA&oUCj;5~S3Qg;_SE!&jo{ew0PI*ClWk4JS;RY;=qq7GHUt+Nu
zFI#EH9Sxb_cSl+b#xfW?K@|_EK>q_|N_~+9H3K>g*iLAaR8A&5iAo~G>MV>L?n%uE
zCaWYrrm$xmUiLQ`+Q?1xn;Jn4bRnvSX52HgAU$O6L$eVV^4j{?FxA5~_5Z#<uJMo6
z%W_TlRGKjjd4rPWXCqAGkFuwz4sC~i^ZeN|zHNW_+_6Dq^o?B**I_4M9F454-Y6ZL
zV?0r96v9CV9JQA1>`8`!iUScI%RnsLUR{zX;XV`MF??XEE!w!a5s`#s4WNgc3DR=C
zy;Kyv46_azZ!nasS5aM#QA3wrADEk7JNrzSDsZCZun&q8h@gg(Pl1$y5_nLFPHKAe
zJ{Sn;6l;Zi38Y@Nu|>w`^>T5~KvRc>umRWvDHNn<;?gP1FzPwOkQyckP+EzEsL^Zy
z!bQ~%>Kcs@r5v4TFZ#~M-C8c=4jNX@Wx1aFd;-(j`EFc50cfa8xQ>a@E*gJt+kL8x
z*t%K4x{x&0k6y1y+jYl`$#)FFKdSZ=3R6k^0VWnmBbOE)9n&f>+{bwaaKfo0QgR=;
zniJTAfuKdwkk;qwqO<1F;G<|?$&@v*(WS}S12CU|E5WW|MEW<z6JLh^Kx`cG5zXF(
zS4|j>_V?H~*$bpFVSxRW&FSfQ6TSe)h~JHIhLbi1kPf5`&5#<hO<Wcz{K*I*KMq!%
zWh>69W|6Rtf0Ln#Yp$;IL;uJ2cjM!m<J`W+A;hk+K5Kgx35_HKxuEoLA#m+S-5p58
zL^v#98(<Zg7E~Gx@kd9efU&bcMf<o7Oo9dd;RbLeTDF~S82HM$DZ76!+vrz#%Fep;
z*FBrn#$YrWXB%5Kz4oze?hEE;e0|KNXC)c}<+o@KKr3&s_QQ{ji*9;`w~#wvU36rY
z<N3<Vi|Gl->3-#eG{}d?Gv=wcSZ{e-y^Fp4`F+Pn!O>_XcI?Q9vHne=%G3U2GMaRB
z32LvHP#hB9!^XFG1-C0#G}_-^@M)3>8^1#NsCknWOwH9qPcgtmC(kL(Hs5Z{jh{Ze
z@cb~ie=)qc=wh;LC&m}YA_rw>fXjll2n(p~WHtYl`|VHKTc3peVllR7E5k#u9qL{_
z@Ia1P=|KlUXQWZ@{p%lJCP&yFj<EmD|MS=X+yAik(MS59{~C_4Af(x3Jd%mDlMfMP
zU14;?KC!Q=S-`86Z)8{E${ga+W_~WXbrqP!Zj}%0J9ie8Eqo6yxQ;vQWuN#-p0Q$_
z-|bW>_DF(M&?zQ8tcy(tbxJ+Lzw|MG;i1d*ogsCJM%U(xRlcy>V1v~UeRTL8QYNen
zCd1$#3FDR^(*~-<F8S^P87*PX$~YuG%Haruhl!8TqtAH2aXvI79D1N?4rqD{ut{$z
z?8zu<<Am9vDs)j;-W2L2q3qVOTETf>#72f6Q6m3jcg{3O9Uv+r<#LD4C*OSY#qKwq
z&!4~g;_2?!um0JXigtSw4L~t1UAg~LoIWLYh}U|LkLRL;r}{97=&PyQdep~3a+%Ym
zd@Fa7^XBDVmhh+y3vPRdf2A7is99U;=<bP(4f?qL_`5|H$yyURS|YdhU$>WgMlNmo
zlYJqt0P)xKc>K!mxdct44}(7LMeo1VQu~RVh(nV!X0j-jaCtnqJg7I}fvcIMFT%eC
zZU#>zY?gU;IvY&;4L5i5G|e38<k^ea=}T$h@jews&$E{oro{_7a#Y+gprgXCrq481
z&yTjnjq5ma!A<wXd;l;C^n19PiST`E4v@7*l1PP+c@)e&0ZB@7G!$yi8ERrIWN?;h
zHHQvO*_#*vU_3iQX&2uTvmt~gZYK%SDgYK(d3W-#FCgvvlxkY<?d=EI+f08;VpW49
z2>^>z)jVxaOlf9PLO8jE`qTLD*u@7}injePSCc)-hO#HR1`wqvB8ZP`+lLg-hc8T&
zN8-bvgLh{lQ|zRekPtS&t&IeNr{!pXatWS3$FI@H^<Cd5hNVWi3|gTLSE3Hryem&V
z6hTDLu8mglS!AU?b5`mZfSL|do4-uG3;}Q3Zb`drJlH|S0K`o*%4lvu(GMl=@FEMx
zZU@E5A>yC)1SA}Wm}HHr!t<jSE@zkmY4OaA@2{S|cwun#sz4o;zbK^PK+R&NWDBYZ
zia)`cpK>w+?Y%69&oCelapC1mifQFyAZOt&8k^a;7!KP-Z?^gGzbjArc%1e+C)3lx
z<}Xd*I_EBEn<fZ_yqIq*^k-DI<r>)j4h3~<o;I%|@7rF!EPD6u-Fx_;Jw_8RBegdD
zrw!3j*EEh3I`dex*@=R?+pT|t8P;B=uilhu)Vy7FX&lHzLSSuiAV9hUX#lX$fET~X
zo=bKY7RO$XYQvSNW43e(-jEv{@cV_V-67-(R|x_(dOk$O<C+CUF;yDpUQiPrLZgl+
zy44JVbwXIqZkJ(F4i9(0nopsrs5BkBK$<s`$IwIPE1{;Jn<bLHY#Hd7gPH<F-0%p%
zwC5hF*;OFq=e5*3QYc5?Q!LPU*h953tpOMS9H)*ajO&A;SOS6|s`PsTGkzgtM;xtj
z<!v9qw$Atql&Ef)La*#pyaF)uA(#}3Pj#(wHCXs*Z|jknJnm$~B%+tr=WTKu5ugp?
zf*uU)D{v%U6c-8^XW-D7qU&HI5HpLYZTD53k0x)l(sN>TL!ec-f9A$&_bXg0V+@mC
zcB;YcBdp%$u>*qh{n&f7j^=mBCDq@rZ-}3Lfk8oA<dYKmz=bUAhDT5`<6Ro~nIwKn
zl7p#L<(Hq`x2xPM5RHS0xZ(vPXN>iR59tH^e?J=+BivSen8+9Umn_QGf>P7hoNIP|
z(LQC|C}_x>{WKuHyC4||w-!_D$KmRUp>9@UyoS(RoRv9G%&oYrP-r~enyuAcFNU%j
z50T5lC+p(7pX}V1e-WsyT!zTXt6;M0@RPsgvd&6$*B-Ab`r!v_AR((HIH4ZeE4Fjt
zU~~l|rv}?+;IWM3bccrCBtP%e$1v`o^k{G1S@-&Tze#-K79aS6wX0zlzhq0aV!ayr
z4uRnIK*Z9u8KRTSBJ3LCH)}21^3@pjT9=F~hRbDVsLNn{lJDGWf3hA34gn&Ea4&Ft
zMmkKzQXBfFAai%bd{WSLLMM(?m*|f_5CoSAJxPqHtF)x!bmb-n@G0o43d1s4Noe-q
zp!F8LJa!@b`z=mL`Z<5QMwkP~1P}HLgIeuMP$Jq%B+*fz>ignUxJ+-13n>%pW{u`?
zo;|!N$a>-&kUOIV0&6rY5mH}>S>ZqQkfG-wzL1C@p;OCqa1k{R#^JvQFGizC6v%_S
zml$937E_c?c2EaYM<kE!pHCIm`^(0X%?+wBWCT7#dCf+S?yNx#$>D@yQf&>v-6$A3
z#Q3xYH$&J)cP>Lo71dI>#x1B##3oj5gdDgDtZ>xWRPhFu7?}s1O7?0)S`~9Nz~(}{
zHonD|hO}9;QkrSEeze_3V$vNBJBYu-V*2o2=V&}U_+`P^{IVm0yJYq2HGNg*RZ*g-
zVq2cQe6c*Xt5R+)ZCnL<`61!1VcM{TO;3>yA>v^ntYN8a?D1nFMG_8|hmW~BMkX9<
zr`rOy#^p)!C@!}@#$QOqVc_aHIQ+-Zj#r*(!Qc7Mbaru6zUnQf1CIIct*zs`ZZ5(*
z-UN#Ns|S2G0beJ^6V&`<kZVJVixwkfdp^h4H$^yb?yV_bzo}UlquNoT%jfa>#D}^q
z-@1)}OBxWme31y@94Vs^UI9w&qj)V15ZRu*$!IGFm1JvL<Ta!a9)kv|Z=5?N5=Id_
ze)&KgI@qK>oIxRjeI2qv1n*l=Qk+&Yuhu4%PA#jZ6R{GgEqcLu1PHU66dUfC;Icv-
zDKcXF!<s@jZRc)C7EtMSan!Kbp?WHTauZN0CioRA_I<O4DLXOSKwEm>TdwxC#2i^2
z26smZ@;-M^Y5DFLXVBIjx5JAV2dbda4$e9KX{Xa7V*f}PVn&L7h|2zCSd|@1E!t|p
z#nwNS>C)a*opkZj2vw^)MuN;2)077&Ku7^5{S#i-a&44ru51l%G?TYkilL6S98;$Z
z`o~!Ks&@8y(3WtVoZ?y<vFNqXbL&^As(!3g-oD8F6|}gf#Ki)=#-F(XUDUYZh(+%!
ziDYp=AJX?3bgEdKwAEzfi3gzthuyKAK{*qR9Mz;);8N{-Iu>c}YfEBdbzs_;J-!uF
z3+Y>0+6Ro4788cD9f@N#CMJkbZW)_>U7xS7>5sZL8>Kn0@AG3lBjJn5em-EwD220e
zb)sQf!P>Efk1X%4mra@zyH{VorW)3lisR3F)UmPMmDIAaQ#E+hat8HmY;Y|#Z5%|q
zrR(HddYMMGesa#qUMl_7H<^o$tOYSNgFW1VCqAs2yo-NO$hMMpUBW<AgETmpxGsOV
zl#^L#vdGcY*NcsO9__Ir)U~E;4mz1QCYKPb8d<8jeh%3Fws9FkRD7-ToC+Hc!M`|r
znm_2DfwO0Uv*+V~_22zh|Lwp1=p+5VKIiPIp!+douj59BPDO%m*xz$kOqtIn2A=8S
z+@cJp@7u~ib5O)y$HDU*U-^kXo*<yeYxisUGWU!T^v^Blo_nKixk;D{5p<j15nEy$
zay&el^FZ0KW+~<7{_vmzuk3O4GcKVxkMF{_u{qAmAWh7nEl)Xl8}eV8L?Xs<AtPGy
zgG9+IJIRvFm#K+uT=Tc~5V;F-!M$j{P+FkTu$JLP&Ca$II(2EhOvQR>wD&qvu3BQl
zRm6Iep2(Yrp%04;9P^gZpP1=k6)H`jzVYTU1Pex-1IGLdb*us+Vs<x!6Zh&g78bCP
z>Gz}<;90Eveyky8g!6pFO6@nFlXyxRKu<+{OyCyM@n+hi@&Y$j{xWVp`x`7G`ztt-
zRse_sBJ5TCXwX;dkjQ!Ho~SRM?0yM%^OBQottLm*)tk~Je5F|aYR>7R)$XZ#<gq#K
zzF%vNjn<sj$V(N{Gx|B=-t>)9^I5JarCEy6OI5>0KFI3(?X&InQ~OeJU|x&2_P+K;
z1Zy^A5oJTGANpptsz!>ZS(V7gFWVo|V>|V5oto#brRS2(Z#ffsw$*-x95*_BN`!I?
z&qXDk6i1j|e3_CWKN}FlHpUKjUybY*R^%bc(&=-;(f+<w9i#QKNYXtm5uq({1Zjkp
z-iUQ(*_SrhUb``!+NDeZ&V0wkBqTonHlL8z<2dF%f1-pHDFW496w?YrU?KGC;K0si
z0izzEz}7J+(5S>482^xdyCA@TKKm=++LhXE^I-F{?b@UC1s~?J&v2_S1E!CdlZHNV
zK%BjWj-rkC5gwmybqIwz+lG3Em{>ETS8GB`=dpG8k3mnp+b)AOCsvR3cffQIAS{*W
zQupe`NHrKVX5OwrlPs6ucU)h$$)jxe<?h{T?S}DJ6+utdC&t0Hg`#3cR%mUg*&gIC
z(29_^2CU1Gq>o1&h^d_s!3~pgiZ3<ZLFR#5qRlx`s=MS8=YwH!z%#qbE2*uE!y0WV
z@OP7&Ysip|{R5{dw4<g-RVbDPpZH*LY=SdUyL!@hW+p8QxG!#K$ZrHW)yg~}<{~d^
zF)Er(WABxeSXo}Q59xrzgoAQRG0!2{b;+hy+D`R(2Ncy}DjEPKP4L->?9}8L#4oPI
z<FJ_y--DZ>7SaUE&Q+x|KRs03u`y`X>Dq?5?g_tmIq7PxB+*vhdZ@a22ce|)++$(J
zHRxe>%xQE;q8))y$jq4zjS-0zvfak-WK{+$x$4vM;NYd_z8Osz!`pEkrSz-rfVMn1
z6FYmU)8iqxK@sGAjuu_>bGboclL_+&mT!Bg^aA5&CzI^(ZkgM{f5VMYTc6YoQDiz+
zO^H~EVl-*NCVXWFA?0JZ9rJqNx?YmhA;qM%)=r9I$czA1e}X{?vNZqg%C6-tqxRU$
zh+Cf+#g#>I>y!5ywo+EE#z-~5SZdg+5*F@GW=e)$Ej7S&(0ISaiO%?77h2Y`jZYg7
zYMLCwZ-L&AI>(c|*Q)Q7=BIS%^Xi?qwH;HNi;jYwW5alUt@&m_`{S+K=#FqKY2k;r
zvvpH!DxQplPSvk!y34wI<e=I}52E*<rI2XeXWIG|CFPe%4;FG`)~>po7{yE_7qq;Q
zI<wo$iC!<&T?mIqh(=0@xRF~}&s`Vy-oS}6GuGI_vqsU=NX^c;>nuzA>x`qg!`Jie
z0aPZFhHF`WK%3{CI28ePWIE?tUg0RO<I?!88$U6s_K%cR#4Sg0-pJMo&cZ~E3uwS{
zK(N@F;Pu6J0Fk#_E($X6cjbsXMG`ZaqY3sMwiYB{^4_c(TL5zI812A7*el+))pwN)
zMrG!D@s6x_0XZgpy&ARUXws@RGUN-=wJaWj3i&Rer;L!YB%KE^g)OoMSeDI0$WzJ%
z($Z^y3{UUIK#t;5ZnBqtZ*=7P`sbr#61^(9&yS=;@cHdYJk%I%ndZnE|M~`#g?-bH
zIVTL@WDtp4f<)peXsWoi9znaTnZ=N?0bh9#FlKZKKerlB2FO&>elJ20pff&bzu@xq
zc6mLT<dchM03De@fx@EzdvVJuEZA1Ij@D6IqE|)uZTk|;Rjol;70xu!O+5u~bs+47
zGuT1Y&|xeYEn&p~r>&*^rJ<;h=2{zhW1+yvUu-+WKc7S%C)jrW&lmsUKfL(+k3Q1>
z|LL4<$5L5!@^Sx$O*y|;Z535`5stZ3SM&2LQ8%|=U;*7_h`@F1HPxVhSo2JY7g%IU
z*lM{ai)pE>ExaTTp`xoNq4){ZaXP}G1Vg%OUBspY(~qUNwQ2x_H!JIPbjk`|B_zXb
zUx~544mXC++?LWqPAABmru=y1lafxPo~8|F1|<d&S<^ajN^*}ND7><gOcC#iHL@r+
z6dH(F);eTeLhC@36nn-lgn@7$bQQWYaO4xfiiHqG#>pXefb&Ae4Oe8y0l|dqH)taW
zwJaapyBGJ2o!Z4E50wzPnUok-+7zplqJU;oU+4oe5!u}BYuGRHQ?>MrgClw2G0=@|
z`csvG7>2#UsyD?6ie(uKyODkSyYG*CA3xn1fBVFW;@H*L2sN4lQPpkP$l`u%upxef
zHhj=Vt1maUfAzp$d$+G%e({@)2icR~Kh0W-@}j!STGq<Xy0R{*iT-S4N7Sp#wtwCJ
z^~Z5+t7y|0TSH|}o;(f3Zdw!o${T~gnoE#W>MT+wee#dlm)~onue8-auTap)5b63S
zvsVonEU^Rz|4D+}l`%H@FGri4N-+gnUr>ePM(S0-DB{w6qd+0l-oo^_d8?+E`Nhj_
z_jxGVX%{POwKbI8B-Ha{^kDC(ialL!+Ou+_R%KKdHY9-Av@y%gImrAwLZ5b~Xm9pc
zg5d;PU+w3)=1L<2Z1_U#LMnA#dw$gYl^YU7zcDlb(go2ts<7Y<=|w#iE}q^e5>8-d
z?R;sa^!T1#(nD*t@no+=FrdPgVk1ql3LT&-4fyQffal;ED_{F4{M_%MLv>f%#(SP3
zVeoky%CJ>>Z+(=kAEb-_EXZmXtcDXFjX~QG&R8`^dSJ)hsk73mF}R$$kXuOWX(6N*
z#acs$`ZgZCWLhb^m~SL=1%$3{u4Rf5(4^wX1=J?DDWym)Sk%rJByo$cxE~aPN4R!^
zTw7L@MZ!I$V&Cs|EnnHKtQB>_a%blr!M)AKb>gNiND}XIx3~^%HxF;9oT67$E7jN*
zUxy`&p{TA;vjp461c@T>wD;m={u(%VIwM2PP*$LncP#YH1YLsr?qEh0kiAv=!0LLo
zYD28PrnhBT@TwzY!K&aEl4q(&mG%|%Kx@LbYrv4*Ck*C!nRBInSKxhU*GI5!RpHR`
zai74KAcGT?<IA*5%59J{`STFHjRB8a;2tyHis{HA0;?8&!T_eELn2&pgMy2V={_|<
z5FSheJfAjKcQ2G(60gFgx-iy4Up=yW7w2qi<EH7BO3$ME*TdYX{G5LK%$*)q(m1KG
zQv0}APVI~<7ChSV>ZX(Hz>^z=M1oSjUQg?qOl66s5z=dCpD8-(C}%Hyr9oqE**9SK
za8f!LlIk&B91UDdWZ*P3sKVUwKq9YM<sLff7XwSJ;FvIs>`#T!4FPR{fQ6<G4j$D<
z%7a3|{qXjOYDU4ag;t)_fbu?2Lf9A=0qpQn7IItvt;Tf`D}mbVHq4=_fcR9fwL^Iw
zA%TuVB|M4U@W6}&NKgp0eC0tqqN^#zx>P;}?V+7KjJY0cyOh@6M@8T4=u!A7nTwe_
zvugh#U7nO31Gf8y>qsm;emDoI5$#8jIa^NPR{g6n?Am_YTNP-mMz?#-WbQqW>~)OX
z!h3)B8zXssJCABr$6<Ri4YBv=p+<BaC#C#I={cxn3G;b8g0P6B4ntaG`MF>sos8ag
zhDU5iJd?eMLzye3jZ{o#RY`=iyUUu`*kZJ;O*MxnHZm~ZoQhU&*>IH9!_XMnQxV#H
za%noX#fZ$A+-24uZ|4hQPBCzdxz!hTL%iH)xKEr*DhIK%8uMF^!kr<kE+VR|-hw8e
z$P!Qbg%0`2|4xX60ff&{>h^@lLq;-}LWQdF3&Ae$a0%#%^MT7MiZ>vw=se(w?bP}R
zCk!XL2BYrRfha7Thk~&RSic0`1I_W4kTu(DL3X$W+#gaITZtVb?V3*0xq`2l-5n;@
zljb5KD}$)=)=0*-)ecm<Ub%_%jy&c_C`{7g-P}TB3EYgriS}4&QMBJyw8uy7cWo-q
zxP3v#VHB3}bq36w@$3}9xV{PqoLmh<jRpVbcJquuEKGa&PQ+=b=~{?W(%-j&<l+qw
z!Tu_U5cUOU6aOawk8?OWH+d}eapqB`s*&X~+PW7H1BpQy!4vv4Tqa0M%c1m_x;m+b
z8We9@+qkKrIk+6df;uWJ^r4vWh7iWrTopqVa}a7^xy^R5M-R~%`EGJqK6tZf%jNf4
zo~i}o6-TR`w1NS!tsYDoBr4oMlmilcQDqSdwBlBC7tT^FWge<K;0sM`0BGYwSO<sf
z$h8EB)Mh0pZUV4U6xz09`aC`C<r$78Z}(P>$Y+mtvWKC-^g95$#xh(Tu$v^l^$?kh
zLJ4=5a$Ql1dcC)b7H*E71BoJ1POVj(1HR#Mq8(iQm5psF#a_E9XB9Nf+lT~Gc)NZg
zss`aIT-%jMsNFWL(SotV1%fHE{Lo(BgzM?La}ipjEe;Qn3On6(c>%?=TAA#I{$5vG
zzZY+%c159baG(Kv4mCn<H0=uL=;x)TMDU8l*tCYeydY}14qjRcB<~$E7NcX_@l0&h
zR_w?zOICjN#yEX@5U&T=Q?8}(>*^PXfE34iYDF#bDIx-<C!=$E6S_S1&(^vS<qmQB
zy^=D1p`rspU#UL&v%jJ-opU>}nkz79R{T_Dpa2tFhBa76;+WE*(qn4?Rg%?8t;x<0
z_VbS>NXLP9e&8ou3#xubywus?s*`Y(X#(4ryQdc<Iq>%01WaN$%%8E?YAb7N^|pQh
z%cUx7kv7fdZ>8RAgWJQptW(kH_n|q~84z7ac<jX7_9QuRXng(gRnLh<4YA4F#a&_x
z<{BG%4cnx)Y)ef%$Ue>P)%HdWmTot*qvpBKW<Ark`z9M$zWZ#v$?~(YceU@o|7@(s
zZ9AK^1=l?rA876>bh+n?ClggYX0?c~>-{Mh<lD{BATrH*{ZWS>;VB!Z&{Ul2+{O7W
zXC~!Bt%^)whx^+fo1kuE*-d}AeIGNJiLXKG6`x?=g;giBLeBp(l#_T3?KYP!h%yV<
z6X<jC{_MzwUgn&~A8(MF-mg>BDv1XU*@U#9sdf6Atjk#|h_-MH2!+Ps;-u?_T-|9U
zFF|UEMn!Cjtn?}JAanwu74LL7Go#8N>qJns6<t%6LRL*S46md;4Z}i7#XirDsXYi1
z_J}pr(-PXUs(|tPIupyW8Pbo?Q(9~@>o%A|C;r9;xrH%~V3M#fFDWh4PM9^CC07y;
z3pLo_16MLsTt|QU6&z?dmAErk{QA#}#Rwv=3RjT6LuXGWj}oHa-VCQ9HJ7njozWZT
z9xiRVNm$Q^H{M=sd<9PJI)~UWxf%$u{40Ub&+8Buq8Gx^z0+9zr!p=_xN>0GUs-Np
z{$tGU)sx*<Po93!IY*D(IqD2gJHzo7oJZ$)H0s>{xSgN%mzu*HGj^tBf8_%P3&!$4
z#!|ld`kx+NJlPs-KQRB2cTZavElzu`;~#N~XPs|G*wiNf0Ld62qIHh9?{~WQw%)(Z
z_4&SQl$)LWgFndR(<nCi8^U*Aj!+!FDT=WP?Z{!CAkEypZZX=c(XQ~La|^=dm!m}V
z39{6N9-?@iNa+@O#a6XIC^?eo14^SuW-q+cdhWtc`aMWz6s^sx--0>zdgdf;tgJbw
zKxa!>FlEeA2Z>X-#+MyEP@P&yP@#KUNw#ABA`cW$mFk`%pnm8MN-HQx%}pD;PvXt+
zNLsV1c4ruj8r?Q_VC%zEb;6Ks)vtSRtCfjqckZ)nc`mP1_pjU$e}J2q2sl51HxZ}e
z$~(;wnQ3#AG*}QO!%XUD5WZEPlQ;^Qd9j?8JJpXj#ihgbHn>)IGcx8m@~cwKn}g{g
z<!`z)cMr&rvYXw-A=0^R^YocvR1K>AX9kR)JLm3Jr@RMSFUu3@A4cc8vA9;?We=Mj
zy$cTxx`eV*1QFx=Bgx{&3N^v?lC$iOb@(o=xmnzPt}@lAOM+(u9NCU8ybFC}5mps1
zl2zSBxzwZCU7l6Wgrj9wUIdvSuI|OGJ&wW7T7NMfk?dajPWffI?XHp3*S+vO;okEg
zfP`JNz;lBic{q&D)nDpDOn%s36?81msGzvie&NC*bb$3NShUR2=TawkQ=X{!Q8lA}
zuYkCJUaX>T%qx?s!iFuTFr3@bc(V=RElXu+$JNns1qI|B$c3kKkQAOqP~O@4Rs4`J
zpohdZCfbaTb!28um4YweD$vb5=1Q6-wY`%)5l|T1EBu9kmqA7(A27sKtH$Kut5q%0
zkZP=212P!m31C5~VIl#=shvG;G_HQpZBZNjv37AgSV;VFuEpHu(s1s{T>H6gr~TZM
zIEQe)Y(>bQSU(R*^?jihXzX#}ORbhsZ>go+$$o<-!pU(Em#bN+fqX}CD2mzIyk*<V
z8eg`L)xdr^T>;x1mp3fX6uiqF{QSeoDX?}-8a{e$_$Dp*(WpuM<C|z(U%9{d#Oxuq
z4Mh~h&5hmekWdtRuSiB9e?CNW8H6D#S&-5>#wn46$3`dl&b%wh2FW0F2O_`_U%0ea
zv_s7;+V$$hOBq&BVH*znv~{no`UO6ZxgG)rU*skB<+?%|)l+?P^Pr-QsqbY}a9St(
zNJYC`NtsM|bF<P&(Z9)+#nwI82b&wrF-RYc_6nGWI^a)canj9Jew;CQ%{AEdxw;-(
z-;IC38#Zxom8fP-K^<Ghv6nS>wd&mRJXtdk@k4Wx8gNx$z52+n)soltG2Yw1K^Tk^
zvHpG&Y2iOCYhqwR3;TCj1I!Y<uAM+&IOB&d<nM>W&2&7QUV(sKN7Lh;HM#AJI3!zt
z^6)l608co)ui?p19k-*;Vc_h1CpVl$&PWgzmqKKPjQ=5GLr=q#Mw@oK-5g>28Y9?X
z$*Ou>;OnY*84^pwJzCy*RfU9ij6phK4NTz{%XsMq8IGN{Ulmr9thKd8Y9;<6npnaf
z{BeaPL;zy0t@+eVVN+=Gs=))zD?G)3zKf%o`{Y%Gk)IbsSP9BCo;u3Cp95OIRY?QL
zSOBpdOS$e3TVXdzeWW}GzF(XPvN0VEQ$?Ef2>($0$s-Q2UmRc9kB)qGv|x}u0>sS&
z#Iq+rV*J|yiOz8%V!W&&cwR{Gb&^DNEj5wki4jRlqxoJxx{p@veVVW5+J~ofh>`5o
zb@uOQMf1Jv**p5wmOkIrvG!|qHmH}6shh3zxvRk3b#N$79}n7g>e=r@4J-4WBqo3$
zLYasRv4-kaBzuy*@Kz(8q#L?C>!cDu9`%nQ8!i4Z(Whl735?{xTVs|h$wuAE*`Xzn
zqsT7*S+ayc>Gd`gopundv3x2?y`Dh?UbhL=Je^I87TM=Og7@@6fv$Sz_`{%_A(j=5
z%ni}lr>a3X8=KO?&fdt7AQWpPclH)jpG6I0c_Q7ZS-`xWqnXMUY#1Dgw`@;Ni6SbR
z2DSN<`AZrC)9$T2%Ki*eOpi!I@VEc($$$NCIv;(c|NnX+4S{1)!?7KUHoftL1hpA7
zZ3VX~Jwa_1^?eNJ5)t-#i3%2n|Kp`BNCq3Q>^t)o_@tyYc8w;iS*%-b#RDXaTi}ET
zKzIoPi4AXx3-_r(Ubq6`_<1q20biT?C&hK+jXfsjkwnm>(EI?>ULPdrkSZypmM2m}
zJ`;I^DTDM*x-X@tFVWK|PztV~eaM!3v8I#`TCJsLvx;ac8xqwa(lq1tNM>r{fphfe
zs#t(mm5QsxHlp%+3TM5ie-VB3h)-cUc&o=X{-SK*k3J}O#2?#-IC?bn1xngy<S<p%
z9_uW<9=ye=f|6@SBkNLATd2W~nzaeiq!ewmN@)y#_l(48opFCq6(|XxP!p!NCl&_Q
z`P9}v9Jli_pG@)#%`WmV&8F}Vt+!-Vc;ktrb|flX%>8cf%j%VWoTsmUtl(TY<H!-4
zWi>mVpnilK_Pgg{jSag+fG5Z5X7^H;D)1<VE@7`V=dO?xBggyM)A$9~s1R*b3vvwx
z49+mOX|%J<(!lCdXc|cmOF$m)6j(mMNMr3aS1a~{oo9v~>=Rt4fojVD3Ldz{n4l7C
z>s!vjpz$Z}NmgI&{1QWP-?H1PmHGbKK80qIFP9{#S3W<?B380Vj@H(tg<ffS<~KD@
ziLR9e@M?XlB-!-s2t5cL#5s7@X^jYS*dH1|s$NW3uToQ6A7fBQ=Pfn(6J3`kVM>}U
z#W-DlXO@(ntcn_Yc^ug!S-PD<TLD(`ca&qKtg5l#yJHl568k-6l?g@B`S*xGauZ<e
z2MKmugk5_2;4%es$Nv1P(i+N&_h-o|2->RhwC-W=IIFd{_Yl;t|F`b(*Y-Us3uSxT
zTY8|U_x0C<`}bb+cv<NwKQlpjESoI1gA1zIkIt<W0ho>WbQ&y~mnCD@eGn(~Rw)mC
zxSa?I1_3({<Aw0CZt6<Bh(>AijuoI<dU5Fo<`_5SXd=54WP!{GHTxFP`Jt5P{?K!3
zg840Lgh$}tT)k5Jy61zkxKi;6N|nQX)XfX#3n)%G!?^Er_39R+Dp!hAtYtFPA4CpF
z1<XT<%3oT|-EdOr?Bd0`q%i3R)k9S<u+qGX5B%VYs1;{RujRE%11fa|JT^kF%Fybq
znGxMwnj%#(nqzBYv`g0V-dxlHjv2ursvy^dxJ{HUW;)VD><x^goHlsnO40E8rtYq%
z(y*&4_;&i>C+fFUvE04{QK56H1$j{WTo+3e3x`L_9%WmA>%ampInP+R5E>3zI>A>6
zT6Bw<GFxG1dJ-1HIkrDhmaF5kqfPL7+&tZL(c}_rUPI<Pe&l<rw)w<Z#^3Er1YHuG
zH1|<xjAMLN!j-robAsU4NQCznh5w#fArY+C1M(860pcdV%+30yfNoCYSQO^pCv@1R
z*1L5~2XLtf!ugOQY?;;oyzUpUjY|Usju6b6&7@X{xSU;U=p9t}pv+E`Q6$!lAacd;
zx5cWQha00DOjAn`0v<wiCZRI*<jG=WSAjqxc4YbI?dq9hf4qHoRK)5G7D-RO39A`g
zWr3t%!n3C@vR9Tue?xqUnVx#qkQzS<O?`CGDDmT;!Qr7&tF({0RW<K}cIKGE%txZ$
zgVBfp9h{A%$!SnV{pJ|sR^Rgl;cLkrC5l(uD#+Z^7gY)GZ=IZJD{7Q&NsFow!u!Y>
zm<Nu&GVB^zvDtxk>Zf)MDrKe{hI1a6Qzz_kubS9$!pTGe?uo?BhFRNN<?$*=r%vM9
zr3!RLtkfo97Fp|w!y>tM^^iga@!?G&rza^5jc%F+d3(~Rurzo-L3sVDIJnEfE_&cd
zCt>STYALtu+Ig)jtwMCQUeCT<&%PzkR6Dc8WS)-7Mp2geA^uCb(l+3iE2?UqwJBLj
zdaK!-f}+H99!Nd4^%?~TH`F~&QI?F3rno@0vnNMRlbO6f!v6&fC5D~u1*Kt-FWqlc
z5zrDSaXOid4)a6OM@%WofL|?9mQoWbZS0``rcffb>{#@j)9t^d5DfEYDu;!o2sEc8
zd+qKghVSNh4p-lrPRLN{VL{T0#Da*jX}=*=!GlE@IXsoDrc^m{{=VHtx{`X&b$@ZL
zo_ynRi^or1t8MXv`Nq#B!H?{&eB}@=IiBK}Vqqk~U3o2M`Tb#nl%7+`bgL=w60;QK
z1HAMNOj-`+2(eZ`L<5L}VAJ7ja*_vzdcT3anIBv#UHSG)Mc-V}9?N+R4v6eZXb4aR
zZ%pZUBeN1!saiSx)TpI$Y)a>+X{<UEQ?W5~6TCrW;ps184n0T1+yPSH9N|MCJq$V`
zr;!PUz08(W<pi{zN|I!uo*beAXuB)Bfc`Rq?aJ4y^Fd<UxuDAXi$YDMn<d$fLQXjZ
z+)4kK#8O;ii6yFO##$nj5W}FvT<VYVHQZa>b>xPT4Cp?jug(Mq`+_IR2q_I~2t|dM
zObrFr+8+vS@jcH7G&f$rQa#U?k%9_t_m%hS&Nbsg;&B(2+|d<?m#j5^y<&Br)OSs_
zEf%htU6m^Ro6=5**b1pEe8%bUk3nm-yoo8aRPZ8PxV_W1<-Mc8f^MkDPXvK#{x@d9
z&>bN-FA(dH(H4*iK`sXe3u}fz8ks#*@r=pE7_Q%emfCq_Kmw4ZHkV&5G*Hwhfn0|5
zX>y?iHqThNC;c?K2Pl_TF>=*9oT_lA6Ub0m>+cjS_Y<N@EJ0H}3%TY)zRJWM1li|+
z;;A7({oJw-h?ol9l86Kyrp*esekG|fMPuR+G~P0P!?U9D(r}taBDjm|56N$ZZ3#3(
z6=FOqAh(;0iIX#KwuT)|lt^<f?1W-byo79XL{9LV$Rk{K2{(*EdUV8_<x<OgkRZ%Q
z)KZ)_(Hwkg1JB)Z=3zlpS=4ja*;NOaK-q)VB4V7n)?%@bteo<vhi>fXZ&=d7_#~$q
z>Bpdk8fEuJLJLh5#KKK<IzFe2un*i63IEnPV1pD>`I^#7Sr;{p49Ip`_*V$jr^V@^
ztF@e<JGJ)^2-x%-K^+~!MWC#HNl=H$-eMSADZ8Ef)zG>jb|Vd^7`v@{rH+Y|d;*R@
zSt5mE?-KNNHx7an0qgMHEEF<GY(+#%#=|=|E@s(&weOQg1M@Y^iIw_$YBQ)|Dt{gj
zl~I0eL&B7{JN@C=XmBQW2i66&gg!8&2(FclgiwP!lTOgAwZ4lU9>~g-mrkAC2i&vM
zIe9mkQP6p7JveGI9`tk`Q#--k+kzM7<hByApKnJl(?B|isuh!fqrP<j6ru>3I#(kt
zXAxYE$G1aqPqq22?LAtz#996~UyM>ZOhYxO3$tPls%MLgL+xFSV{oBA4!3ObKIp=m
zOFppJuJ$pPR!kAAc7{oq_9cvR&8}uGcWea<ABbqK{j10T(ayEK=_5CcuB4j)y9$LE
zr|yaheuu7F(%J2RgwA0KZ23dNVnob^d4y5;MWy|^Z;mH&uJW*pxdxKDgA4!`I4rt(
zpvpnx0g@UbLg=fIRMPruROt40#8ig*?6Va=#QfyM(VdJ)xy=6Aw48^c0E}r@qV?oN
zd4A0_DC`Q-gT)ERD{vrn?3JIYDuFlh@nn!h$C20}IAsuz3J=>hm|$XJ7N~u6CcTE{
zTH6$BCIm=f1>7!h6X&N;*+8nDcR|hG)+0n3uz79UYC}5?W^6tg?zv-3DTv=@_iSEH
zER^yn$Ua6dRHIo(2%n082~*~w3zgH3*P@nNo3M8%Bf)7a(~;wEsdmKu8czQ%WVi6|
zYYRX)Az0m}aceQLZE`@L$XE+Edl`BT1yoa46R|!Je<m!NkX{nJ7QVfR-WHqGFdNCn
zI2fn3>3DR!Ixm|nw+cv!hKE>us65Y!rssYxjLza0p0(9H$Hkx}O(N*d5b7U<Ni^*F
z0X*Wuoc{DYqFZR52dBm*x?i)<yz9M7I`UjQCNdQ*s@1TVw;B<TN-}Cg2*~tHtWm~r
zdfgW@#hNil7(SHoit<iC{z4LRR0#Mqai7VGlLHkO-$%q(X}I(3E9k_=tEVqsII3^M
zBoL_Qn4SqredOjZ5|ppX&CHzh3>e*@=33wy_Y(orI|T5w>J`?jVt*W)8REIBt2WHM
z?TbWVD2GZuqMgzWdFR@8X%Lpv2jT^!2OgN+f&5@qEy*K-9|yuf*@;fZdWEMU1k9vZ
zovoMa@lPdr=rotx4yd<QzU`C)y6}D^1DWKY<)e}53o0LGp?Du;nF26Cf>TGbinQaj
zCf#9Wx9c2@hmasi0s6WR=N1f~f&>YhzELuH>b?oBKIo&DxJMk!iTt;gek@4V^V<&h
z6`V24Hw!zzZnsF6sImt>d-<YXcD=F*>cU;x^uvvVhO)yA#0lYjCOaW`TNA`UAh588
zYuFIOm5a7OpA_?}TLEv1ZFWZL5VfA<Zm6abnz^5siO0@F2aeh#P&fMd-uO!3r0fo*
zJ4g8s5Nz}gDA%0rLb{TYfLm5;c$Uj1V_chmZ%F)j^6&ftU-UGNXnI6l<|rqAweDw%
z2+tF9{`4QsJP3j3mPg*!u&2-(Rfg*LkFiIS<9ujlMSg*ub=@YWrW%^m8*l*%29|+g
zhTJ@2cWq<LkeQ5@Freh!2^(iUEP;GErUU@I2}ucLU1YI!`CsgaeZp$%n4SRih5poC
zU@-EyYpcvX!3)Nb*s7ta)w;oyb(iI++sE+bsYK1TVZ*7o6q-^juIY@>{E~xHr5(-o
zK;5E@qLWYu<O)Ll7)&Y{Z3P5GdYk*{Ty!e(QsgsXxz3Ck==k)uYmG%nw4<>M!q6)_
z>P+0rN@`|^lSsD4v|U(+bZc792nFtBM3NvDXcKk<W0Q7OanRLZ)g_X?nz3!4foZb;
zXpzwFDQmCTk;sPxKElLG^s*&RG`>MWf|TiDkxfbL#;w-km@XjUF#o}HGWNoDQkn&+
zaKg&~ohVTj+5f83BHv$o=M2AGgPm66+O@xeZv<bet6t1o0-FtX{rNX?nYh8)<X`Hd
z2+n;&s>`ptUpWuBtuML{tXGK81od_CuSDm&Ljhtb$bs&a6M;T}X+h`$U^e6+%w%vG
z@3z1iv!*t`(NMNnN?I63(x@6M;+KdF;*B=(I-I#a1{!tBAx7J+J&u`BEU@nCOYuI$
z$t2@R4M!MGzO}57gS6Ydw1q(u9bpi#UsI;HY71QV6p-d4GavdWd++jE*s~peZ6M+&
z4fr>%nIFOG>l?E$K9_q8h*XMfiuP%}&KNjF09aK#5aEFAei4+6EJStxF?8@mCPip8
z04mS;nhhS`L)E?c35NcCN;tVmK3G=#8k!@1Q~+RjCs$nj;QW=;HFNtL*~jJfgOIpr
z=tyYZE?wn3z&+hx`Lgl_(!CXTWw#pocljMQjRmF#+>*p=u)9{Bk4n!TO<;@*UkkHm
z=S5CyU$()6<<KV{utw#Y%Q?CE&`#`L`(X_s;jG`1b)##_K?_(QahJv?8BJeFujIRt
z#!}(NTWpcAQ*Ssevm3#jpyRI;<x9vJ{>O9uFC}o*SeH<_T7#Ny%Tno!d%<;8N=*}q
zD<xNJYb_$~z`!8qNAUIx2BzUMBvE}KQs?SPI|Pq1ppMhcj(7%3xED|8;p+B{nrUEe
z*k}#7*Z%d5)vVb~7zaXhYQ)i`cFqN~gWf%{rov53*cn1OFbuHM$r%aFJ^V3g9MDx6
zuQ@J8D73*)?#T;AgQ9f@Tmd=txq;J}@2FE5jFpBmSxOb|Ld`b#KgT+^+a1R#Q-jU~
z(c{L*jFOP(Fe~~3D)nQz#G5QhnDmjLWt<Znx@1d@;%8hFM8zcGK3C!uGU{=M&ZSJ0
zHP*zCs0g$hP(^c<Z#)}Ju<!^;8s`=CUN)5NRU~(^5}^;1*J_1A#jv$M2J7geyS!Az
ziP@6TKP0{>h)tZw?Z&*e2@T9^dr<j%Ip3V(ML4>>C`Tv~WG@Q9%fdjGVdkh|r>|HM
z{xMC1S$5gYpQy^OVNOtI{6=yI(U+Ecdf4*l!uS7_ULLz#1~F-?ej`?cxWcprc{L_I
zxQ|IM3L<oIzh=R6r7eOt0T8ga3C6j68fXsW#n_onI21g$CSDdXXbpIxkPy$7;5q=~
zj0h^tLQUyPdNI7edPsuelhrjCB-wkaX<qf7Tqm8D3?hxR0a5!5aoJ@b-lNAzt=|{1
zlN!o@%eHLZ_#tPq5o8HVr+b`;>zs5?JYpV1_m3ebh}kk~fiO$wIIrG)hgO~Kj@AYT
zpMVV&7$frO7tf{CATHPm-C{2CgSFxD*cP<O>y6{ML@U1VndR9GU_wpeQqUW$o;)_^
zvlbv;v^P-?7E|2e*T^Wlzv6u?cL9%#4XkwVby!(Cxhv}7!vwv4H^DSj;Au5=n?h8<
zU%GMER>M~Tr+f8TRZhYVujNfC<dOQMP(-aFq!=zs@DZ;opQVMtyk+Wh?vId5Xl%-6
z7$n#V*HF9N(A3j@IYSm#I4ZlF<}fESosx$Hw;7n@&{@Ms;mwF!bT^g!Rh8S&1=GQ6
z5Q&kLQLU8QxFKU##~OFw`V0MJ^unISlP@tTGR8sty%U<mDS434yH(yzjd#>>MTPhN
z@Oh>1PJTo1WTx+~tFV($5y+&gG;XyZ{Kj)C2Vh5?ySEy5nn+uC6!EOOEA<xsP_Lul
zt6{jfE;IP+o4Po<0smO1X>*z=GXhAdFLoz#7IRuDK;S6D*FE`12a`nd36J#K{N@&{
z4(jPnRrrICaA@$#s7Q^*?@R2ED$D>`x4WE8j)-fSDDs;RbMGUOw70h9CT*DQ9z-6B
z6$c^I7bsE4b1v`^3AFRkHQ}Qs2-Uv`m3AeM+^Cd7g&@5Yy&VTa`=)DWvTe0#uJL8Q
zT3BpKj2vN($0sZX-LNF$`9VP?ctk(65lOMEs5`*P5M2T`hKovRle%0l?Q3A<Tg)Xm
z)@X=4+6~xTEBWG{RObTIR=91S1_8C|!`-TOYA4fm+oOB^2F6rxH+AoR16<<BAs_-E
zjGzd5t7*RbK#iWn#-b0C!eS+?EIs#*h`9@3f`|#iuSkELE{_;lkG}*g1j~pTnnd)(
zj90bB^~NN~(rm>~LRV547Ywt-!>4zQoZNd1Ol5Oa#AnnlTVAcmfZuB9TwpF2V<pnI
zXrWFG3=gK~t-wLSc&=9fkdpFL34SJduRkky?gP#y{4&^0foqe$N>f*vZj?;2`&;fH
zbb{Ufq&ooMR8htbKazP-n(nAOs+MRr8DfG1dIu?5gb#xVc_xUqQL2zBu~qH;{aXz-
z+K1{c!5@S`nf(<?P8ME6YET?aJH<u#jR2#Q+OPgHpLo;o(tdSy%KnPA08FSHOR93K
zO6AR(&wQZ4q0&alP8@~`4;q7=PuTf-tJ(t_=A=KazxH1nL%c!FUT;$UZUf}QH{{`~
z%_BZQ;d2E`=XF#O-^%5^Z(N0_v_-NfpNH>8URJh&ssB!51Q?kAunHHrj;Np#`1EK-
z-0P5tKrp?C@>rVAO2sOVYZFF*bYpuYm{dqv-<qTG1*NoX=%`pdfQcBQRO+9BT@Uai
z4@I#U<%#Rwv(cdf=sdG&WU|g+i5QxNLVEmRG=HS8hGN3ZL1uHT>2@`x9O0W@A0j)M
z;jNLCgzXnoo4TUIS>wj1NI?#~)zA3^-<Ws(aTj6gCzHclD6SeT!M;XIu>Q{BG}tzo
zA6jGsS5}*9Fq>Jt1J(2d@aZ=L>}q@s_E3X8T+cq%zaOt>zitfLyjitr#$mldeOwLt
zD-HIm_3RV<`-yVF*{I8>$aP8+)V)c+*d+<G)fBh3G~5>3xOLxty3bEa<_7sOEK*;+
zKM7;|FL7}HEsm~vQpZ3etIWg$`{@BcDT<|1Ud8O3F8u;4Bs@bSX`v8+Mx-1hA~5wM
z9!mj+sVAm!u5itNUU_!IGCJ`(R7gDmc7#a-a5mCllYb+4MTm%C=8)kC)M1Xs4krAN
zPP#gn>d31j^24Y7Ry?d5L#0S=Xci2Kxto`1L_i_JpJBfYbyb)vAQ&OFCh{}jxW(Xc
zHO8;P>f~g!UlLqh&KGjNp2q844B}t%>i-$#)!!wr{@?$8@4x@&dmnwI|M~BuN^86+
z^+Z)7sO(TBP>f2);yJ*_{qe={@CP7frAuPbl3E0N{E}sQq(An!GrbrWDpPPBu1g*p
zRL3(mDSX{koz-l!Gb%-0J>kJF?!@Sh_7G3dlfG`9OsC`W(dH(<=Z9@>CE`o>q>a%T
zEpEFzI^8@bLBQ;Ab9`YU@TOaSHV+4*!_DK9jj_@zZk|HaHrX#PIEA4=JJrE_R^T1-
zru?v-g>GsW;RCOlbidk_&il^r51W5q;VG!2&c!=QZ5P?bH&P*h*nuFgg@z1gs%kJj
ztXFZKJwomF=%B7us20&a+djyiynL?A#0ui>n(RpYiy|Z`nSvSL14&Gj^yoB;dJ@O%
zrhjv5$a{E?C>^YYgsKbd?{^|471zCq>|FECfvjAC<G2opKkH9Q<a>{kWDn8u%}Bm~
zrSJc|I3Ck|EY1HNqFC&-6_pTTS}rRcbd^#83ksJCj8G`qhlD5=nq6k}vpKerV`lk4
z<t2N;nd^>XhXL_bgg`<PjDsCzlbK{PfFG3N7{#;oN{be2+B?5vR>G)t6M8!-22yl*
zOkTu7j%y6<%;6ikRymqE=F*WQ2EesN#c{%7b^o=+szk8NX1SV7me}|=v(+5g4g1$@
zn$oSi)d9coAB4%R@&IGu4gLe8w#}`s_InIKy+NOFd;_FAvO>}p8L8iPE316X5Q?yB
zs#LrCasxe{4Ny-#GAD?De%d4|`-;5LgW`ZRXk5S}#BhAy6;jT<1Y?~_LBB?_G;^ew
zDi%aq{%CGpcea)zw@8$A!+I)55E}>snV-z<c415F;rP$C+pkc{jZUB9BG~fUUfbJn
zoctxTw+s?Bi?l(Jkek4>r7Uiigz?5)-vF8Cp3&r7{)Ey{tAqixdyaZ&j=^|k_S~+N
zKIZYZw7HzMK?SJOX=WwDol4<$?C=>v?-O;xb$MJ^HYr7Dnw`pEQ*O<_5a^J36YzLC
z0{fOlD`+wkz`U~bqP=BsBht<78uasWXLm9y)<Ol<)b~sBDt#{%!r&Q2$y;MG1EGg;
zwuwf&qt6XxISx0si2De>Wj7xb^8>z!39eeFWR~06yVrj32@*8_JVbuxpL@R|4RSf{
z>DBi3ul39TT>m_Lt*^?tVGhpAKQ*9#BO7MAsIKNqAcsbh{UJJB3Q39L(yY5YesGXF
zslLhNcqWHsIXfBoI{4g6)s@-fP_(Z66bFN%P5zF9gXg+PFDOJq={@!(jkJ?~DwT#*
zcWW17iPsdisVVj!v$%?f1j$OdiucvC5TY(0PSG!3kwE151PlsaEIym7+Xd}Et_J_q
z%3H|0l@x(;3iS&sHR>tLpfo)nly0>f3}495q!bmmNUZ~fL1Ty-sH62hD7<DPi><!V
z31pXV*{VZCyt$&zD(~*v>YYi=4xY}YU-jP_>F3|y!Q(wH$0u()$Ad`+x;yH`&(%dQ
zW8=HC>1qFMhyQ2ewitKY!m9?YSA0u8(2icPuLEOXU|QIC4$v-4C<0S089+GO5+5FJ
z7|P+ar9z<xB`m$b&0$1b{n>T}eC6XaLf0mpvu!4xJ?GfNMTL1GX;ev?JAJ*;#qcP4
za^RD*gO}_d!tzJSAx=lRp<u;=xTL%=QSRvCrVLKkxXus#)#J_ar;j&#{j*O6e$au>
zV=ahva$;8pw~oIdH8~SMq19Z%Qv^NyW!om{<$ySoQ?&ni7jdVYjg1d=$>p%PAa->r
zHL(q0<!}<_x}1&z{|R#FKxpLGYXv&$G7D^6CC>!kYJ^CuGg%tQ#PH;Pal~dt?+9Fo
z;$C2h^)jWl)$wcYwqYY&K`8ne3MDo84sel8&pM&1Y?j@@?jbodhD8HEph}m_1~7XQ
z!W~55ZF6P@A|7K4hm(eBe-_T@Go4V&fin<4iS>J1QqZE(z5lWFyIT}^@iz8y`(xhT
zf8cKxUvJ%4#|N~rk}waTK;AhHLVzHNAa?dw{^_6p>7QjW1BlM@(zAGrZ#O=DZ)5wB
zs8$0@{n4oP%ahTRi1z8E-~Aj*A?2c=*w`X`=9d$7YZ8bnE|<<Krosp5QW&k}WVGF^
z$;B3<|NaQxO!5m9+WWdUrZb?PX)w04Hq8&M-56`S3$BW{&$Z%5qb_6xi#dLDbLxtA
z_L=<d%^a!hRd(A6HEReo5s_D*`4u{o15c(JMr_~@kVTBZYidJ89AiEB_kF|A=K)qA
z$b>sYN%G=coX-FylX}J}MRkO-hq8Ff`YEJs0kY&2Fr4!lcpWv8^h)B9!CWTG<=GLq
znWTmUOhUjPx%Nr-lMyf}?f!n(>cBzTMWkW$m8C2Avz)qm$P`v$hivl|;s}pm(%Ys>
zYA*T7cv(&M?_3=vFU(TT`Tn^;4PYVUD{lr|N4i%)nb*-H8BsGUwYDHF9xF85R@@4s
zs|_G&`z9UWJ$D$=J6Yg<mY-3PC>Pmn^d?1W5Skm69cuW`xWEH$_xl<)5aow-Fz~g5
zd;j1Lh6X>?{GH-Kb&rZ-;BufzT~QwpDZz%$^dh@^m#w^eHw%v$><h_uChE~b4gq-O
zIaQZzuOc+#fmqYHbR#%vi|t!5z~zkU0|(?);8rP*XW$^(ci|U2$1N|+Nzusv`DpUS
z2ahYc-4Y@gw$mF+El^=51Dcj^7wQstoT<#8+Y-WBm@4fD{sOO<=4r>9n!{K?=o9sk
zx~f?hlutncq!kM1K%%wGlT23{E}zo8Qj4xmFV(~KUiL`G7Y4bZ%e~vcXs)Y}48u(a
z3WdGo-5S(XHB=>Cl{-(d1vP39pnAE4psCUb6@m+aN|8U*%e!~&)^izL6$b1YNpdvW
z>%-#KL;$?MPzOy+K?eWrnNVB^&AZH0ENrMs^Uq-QMyjsNYzCDNuFG%{@>#Ri%e^y4
z6pufR`hCOhs8Y{Q!1Y1DQ<Db8Pshq>q?!3Z;j0Fu;|rI(kRA-tLQ<Xwou53P34k+f
z`^3}h!b_6k1=TUq6+Y5Yx_iV`R+0w^x{$~yemCO<kj8C1G`MYIXAcgR>KByuQ66#2
z20R?8R%3`}QY(oJE<8A^L{6VH^oMd+L3n>Yq+pqJqcxC;!i{hy86PLPrRz~b9(M|w
zVN$z$*GY>l&BPnrJXFm~;1EJ3SW$)nWMHA20Jx7O_a@SLiaMuAjSI94bp?XxvEyd{
zit!NX<Bm`HUM-Ds(8j_{$(6}CmcSk`C*?4&<f3Y&4szze5N3+DSc(s}1APg3+%&^A
zDO7x&w)dVcCm2%g!p=>P=FV~4Zrtu$0CN1nxqF;c#h3462(}*r(~lcsj{lfFcibFu
z23w;bq}(&@n#7|06<zs`{yT<f7CnvE+Fyy_aUn_P2g(mjC$sJpP}BSUmA~3w`4NfW
zN0a!oSH*_cs+mJ}x<Qu1zi)h`8)A$Dhu}+UND@I>Tm=m-K==z2sOoZsKUO79-vib|
zDzC*6>n$YjRCGy=!zinBTd?~222CyB))1B5PN`nNUb5@PTcD(~j*qm#A+S1R!4jjy
zQ>Q?ZQMt%rH+5UMu+c`Mlxd#n(X^8_y}OMX4;V7UV0_2!*x%~5Nl}R)z{qxZTu1S6
z7EC@j%H6w?2p4D>v;|P8&Q=3p@|h!8cb@@(fFYF7C}2?U0eWTtfOCTnanfuyfT&^J
z2ntLgCw4=`=+Na6`GcA+5qVBRP9$9^`bunN4mvlzLlMjihB84SvV}uU;PQLLAsKd=
z#a{{FukI3Y_b%c&WS*qF6E?@DDc1$#atHDd1Pe!6YkTF3Kw3MJR93#}<Ml0a<aZwd
zO}1??ZtgyOy2>xPotLo3aX;Z|AWp-Wvt49HbskMeZJ5>);)QRNQ7%<#BvW;5n6oX(
zPisd>&b;UBkMhZDhf`KemWMJJ+Nh~HNAJ2&OKNOzBrun}DWHTpND<AkseNSf-z^9^
zvnI9$fcE=w!L5-R+ZW@5Q9h%ZFJFl%HyT&No!vEqkWP=oMvZr@wZh(MQX*Ls@^%?e
z&=3hD8*9&9&LhLRWuk!))^;`V2}(3*n;5k-{BHKIfOm~J=`(jj3T*}knCbsvn~_vq
z?`!VjTWZcUVvm>4f7>=`!T+YjN(^fUWVADe5!h%HdGJ7NfR^u@!x_6uz&K}Tz0$q(
zCe4KnN9nuoIpvr{_5EPsOH0#6w-Pd%hwD<GZd`O-emO|2OhQ$)yOA_CurpTEg?)&N
zx9xyER&`YXU~c_h33-3TF75sb_!RvHRMamqn=3u>czUvN-dFlFK>_<Y;c86Ed_WAe
zaE_iC;s7d4xk5o8%1|=Fa5@!B)lhc14ZuHDZGfMar691AHA5|whZy|J?u*~BWrrUI
zi{KLQw#Ge(p|WhP*z)GaP=;neN+{W$0@Grv6z)Uw<Xc%k38Jhl1#fy5y_B>!@j#o<
zl!zZTgZ4M1&72bJ*i?`F)qH&$)A#JAo`#Q;b<V;QqUFS5taLc9v<M+;Vslhg(@11s
zJEQCv&3f46K;2caS=8A%PaCOkwZ?X){9GvZEQ37Q&dZOQ_7d`B7ZKDex0xCjRyk`R
zkR8qlV%z{uG>568>6=zYkN_hP<amyg4YMPMb)2oip30m4PO=1PkZU7pAg#W#e;G#W
z5l3;n)coQ!qA8TJs!F=}2R`na#O3DFxxKuadDIRv?3AUvVNZ`ow!qzLyWB7ogf4id
zXqFWf465Dgrtz$y=an9+(x4e`3A*Aq4p#f_=<{e5Qo%IKi56c=?BZtz(e)|bqSYr`
zNYX#N*QXDkKiq6=rBR)b1boZi&9Ee!3XC8O{jjElgeQ(03}0;<^91cNhC~pB>b)y`
zjLCQ!w=`v*yXaRN%J@A><K=h+^2t-YH0G?0O{=81SkD4NYj)5}+%Da~^GQDLgpYTa
zP~E62x5gMHX%zL~-~czT6ERK;4r*xu!Lh4xG9dYPen)U020$ijbTuP=#d1(->7q7A
zq>!rLyT8#24e4@+^zEv;h2-o28MeHv_8i`n)e3pV%bM<V=u^9_QPJPL?S{67!UZzc
zC6jN+Kj^~+)*L)2wHFhrtE3{VMLk+Odnyi7_#nNi$OMOOEVm#Gh9qi`vV)`<>Ub#E
zXvnd0*UWi<+lT(#EdaZw?2)ogiXih{Kus(s1XlwwWs?Vzs*EiJgPM23q(r819?w8!
zBYWZsKl~ZuHHaIBL&d>CNAB7xaNFsiPR8tZGJ@+pESOnTerTZvz?lp!5+KC)iEhmf
zH%!kLTEt^iT0%bTb`skRr_@|~E9q!^UeSW&+$8Y(0=H?HFs`PP8)sLI2XPVs|KoT)
z@$F$fe00GJ_Ag-ye@0;n|4f*||NQ&^@Bf<p_M?yVfBW5G3h-hOX0H%f-f)n@^V+xV
zS|-L!KzvnKou5Z=!jk<2$BT;vTl@h-6Pn-o!v-eQieY3*jz1!gMh#I_Mo!qy1H*}R
zYF@R3F$b?3*HUu^Gf%85n1FtUE-{GH(G|v|>cD}#%Y7dQ6KOdI{X?A5^pmu2`05MN
z1CwH#xP+OhHZ*?Jqj#?2>3>Il4@NR8CoehZD@{3k964FU%t1dnPSt$IIeBog0sSBJ
zrM;<L%sWZTBDIJZm(a@;Gtod{mXy>@qBfNGE}A0@!*Q^p2VY~0ZiEbY*zWBL8Bobm
zsl4GAUCn|;df0l2{;~C*n1L%?gzGpuz&QS<2>wJhz489Adg|wwPRA{e#<&>+FTg;*
z1_Oq8dW0y2$}Dm`$7yKSz)h7>UX6k&4GvJHj?BD+fNJhkh%CtKa+2wzgSM}k#|(bY
zW>bKXd5*6@Dv1Utn%^ZC3891bc;Z&YZPOHe6<%UaD7T)r1u=wT9(|O-2IhFIQJ<{_
z-8Kon>x}lx92Ms-tA3yQfrj|dMyST2NSV{j31M0Om(;MVS9l*YhKv|Fk=)Y5_%aSH
zPk@w9D6$PJ+rmJ~mJECetGE7+n8KeIB4w%g7VFZ;=Y0<y4*S^BQ=n=m{JRE^?KjIE
zu)<^M#pK{%9)WWEySIi*pIeb)hu7KPpp#19xT)MO64BgC|HFXC+g=-fFcfo}1Va9k
zsh16i44$umWqw{z;;U(ZiSL+)&1li{FrsG;AO;zN5`Kta(Y7v2p{4$<n*^7q;J;#`
zn2rwvnX3J@!KQ1UMid{71v~oJ<<;QmV!Go%-@Wba2B+G3Bi5;Q^N#>h_3zyzNImPH
z6(0^zvGujV>4?w12biQ>LgAyDPNNAC-$GPo9?Y~?6KJ}C@ODEPE_vgQ^3I^V;U`P|
zf;Nd-Qlr!GJp2+~Ihe>j4l)=|BLoH-_TsO#8lGE5>-K)yM{mqk+X$lx7B4w<c?|<B
zABON=&3C&tpelGcOu^W$VuGfR+6^#qNIsTCExfK6JE`0vLwmd7v$pz5dyTrW7MWT*
zqZW>eNNI8tA>pl}0q?bOFeMtHEm5sLSY+nv&l0~|Oy-!mj;J&{PSx`Vm+)`ZF&U$H
zlPJ`I?$yXeldAw~>~n1>o#^Av)Y(rJQ|A&vfL3;pY#X;)81R)gATaJ?t5+PWGRlr+
ze)<GF2#;&qQE`Q#`<3CVKF-a<){~<nTu!b6S>EH?aFyGKmmxlX^Gn2ezOsn!aMr40
zJKxs|(iAZE@jV<54S5Z?gg{@LgkbmDN&@FtkS7-N{P9D9WEekUz=h9U88&MJ-Xv&#
zF(9c$zq<==DMb@M3_Rz1T^mFTpGG{d$*G{#Q8K<5iKe))g6(j8GQviN=i*WFFod|6
z0PGwlimBerLYNh$RYV7-jdKonvSC3c1(n6Dtl)wH?e=I@7WORy`uqvupEEwmceWqi
zGb)~Tw5F0-GS=l#yuF-)&yRbPpzPrC;bKtL{@UOx8Kx4bBiCsDjFHD2jJlP+@z&~R
zv{1OcIx^LUN7-dt2^P0M5rR}plet5e%R%W9X3z$>dH8X4lcn7FFkI+hRO@zaXo`v{
z?Z5{@Rq`oU^o~?sjwv;Qym3M#lcGqzzwNA}Ezh(LbAVVof&h<7XmWqsI7sW&vu*~J
zzDseKD+6c!h;A&8II8Aan0UjiQ-AKZgoKF=>i`v$(M(p8nqdcnZU4wqbjjjVg`o$#
z-+OF3ho;gX?1%_rd~x`y617=Ab2IOd+gJdHU8No#Y;%H8V}(pMN0K8~wn|*>$SD>w
zq$@97-MmE6MDyCUr2UgKS*28Ky|PNB5Eqk@zIfZkCHK8%o!?CI3*+<%7?bg<N}8Uw
z(q}v|r2#fBFG=Apz{xCXAxo(T#@^_8ZQeYAEvc0?1(K+?iff*dgAi39zunZG55^co
zsb+0CQmmx8ei({M&99A`$iQ-(_DNXY9m>RW1sfLFCbhWLGOAV)Y`kdc(iXPfp_e0@
zS?EMMr0!`x-dWXy)tjVZPhpHF|Inw7J*j#=#GF~{&;SnybGUE6mQGNXU-#1L1WRK?
z8iAQbCE|C-B_gf>3Pi=@IRo+S)?R#PyX!nqZTDK$=IlsRWG8lzANPmezQ$?$gDVrA
z#yB?#Q@_b)W!cY%pABXo2vqeR4XFAl#`5Fv`OYEVZ>TZin@n#iwE4?X-D+Kv$Zyv-
zwe_aJoI=)37`h#$Gyk;|75(GWJg6=nw#D~&|4MTf#=A1;)yKI>$aCiUxm8X7FmzV$
zac#)M9R=hikW1Lm9|iVWxYj&h#4yxssJ9!x`4S`4)i<=oPQ<r0d=?*F8LVQ*FxJgO
z)=LtX;@9;3mDIAS)3xCX=O_Ps1H{r(hqB-sq*QdMw4x6dg#`~eA{1am{}T?t55_|p
z>uod}^l-c3FzrJ_-;j2{rG@Wo1Cps>h3L|w$(4aJb@kD19w@&dz4}#QGWNMPNS5K_
z4Ty{n=K#{LuOL!BY;|8X4=?wql2qaEc0*&_g9b3hZEvMT?ri;f+f=NveFc;8$<;wJ
z_WV;p<ddGl0lzuPCu52FSE0mnU9SzFxlN^mPpQwP0jK`bp)|D+j{HT@=d`kpueV95
z=Qg}mwN^fw(Zn#OJyghXa>%Tj#X+x(IGcmsJl=dlg4n2ox)O-i`duGB>x0!9M^|*A
zDXBG(ZuS=in)Suafn{ykVj~+ZD==(4zB;08bpKOf%P;-~-Q^W|%Gl@H@K=P7H}GP7
z=$MhzD$@~)sncH^W{m4>(CE0eg1XqC0=W3(>aZ4j-ZYqHPii@|8&-hbcd4z3+CEf0
zTeNg0vYw)L?kldf(l(>hE<6&o(xnqqF|LdPrqfrjkMVL+ML*$QhY5Wc&aHK50K2Yk
zi1R?{nEM(%YG3662;8vTkuGJc@kVIl29EJ)(C;c2>NiAVyf1{KHiwti&x4YMimPZh
z8xh6j?Pgo!Do|G{=we~z?WfP5J$a?biluL#fA!=yU%dEY_mbD&{Qk=qIM;`hkcDaC
zGc7-v5YS<}El0pOFM3BO8jqXYOM|_K0a1{pLdjEm4V!*}NoDBC=)6OqH&D*JFbl=W
zv`Zr0rUYxPuI+7Xy|!_Do^x~aAd|fqPM|0g#<ceO*PKp{OVs`8OFQYOo92L?_?`K%
zYbdo&=s<cPGvHW^mnvQ>mfu3^;hbv#bGt7>f(%#O*Vb=Ztq_-#TD=GO7K4uj#W#84
zyA#Sk=7VsC-;$93!w!D2-{SeFBnysP7r=ltHcGLk3}h&<$gM3A3l*ukkeNOpngc5;
zgKZNV@tk!tHna3jBka6N8Dj%PznL^_$T=9)tX2)v;^e4Q<X+rd?_zjN<eaO<eo@f6
zcxbm1Xn|JhdVCn5dfOV00JNbk1x<sZO0soK&y_be-q>EmT8&WDUlbJhj@-tR&xCFp
zG3L7wU05occXF<NziaX7fd`av`7l4U)LkTX8IyTeiPeZ->w{^$5s$(5xbBh%*wm4@
zoQT>HL;BaIC07!pl=Vmqph!J@a)`+dsnRZQp70Y4s%h%QM3V{8@=X#RPyM;7oTKOL
zPhAXs%DcsqQC1N%UowZPbkm337T@4z&)VJjX<^DIN0g3|B#8~*BW6z0)H1yj;*YK$
z<XKYOa`afK9nS5(gtiHDzGlsR4#l;AuL#sN#ais%iaexq!talUnIG&XnG1)e-^+6Y
z5LN-Pt3YAB-?d@zYyk%0JUku1p9=)q7R*6leUIM>f3fo|guf?B0&x|<vknc2i?k#7
z84B|o>sV)4Ar`+(%;hTQhBC}{sFhGlxg({3-}a}tY6ZvHUp-jQeno9c<XW#AxE!^G
z@&DkE^_K)@e?|$+_DEp%fB*0Qw){{2`lFBZ|G#^80<&VfxZ$*AyOF4q<8b>?_QkNb
zf%4H(3(i%X+H7PH4Y_;e1baJko29LW0sO;fG}GumDUaFLbGvRC1sXf=#RKJCE`9fz
z<z0^zh)56_N(sf%U+H+sUq&{_VyJ>nW%eDpZ<WkF;Yz6`Ji-iRsMSA&)t|2yf|x^!
zEvl{?&udwpV-9Df&+4McHeDOw6QL-1h&R_Tp+=HEsV<%l7)n@6ZaX8da;S*iTu{t(
zsmdqKH~9KfPl7cv%lB+UJNp7QE0l4vn(~M-mZCG{vv=Iz!{_jn5?B+_Y2>KoX?3X(
z4#U>WuWM(-Cf(XlsF1}iB!SZ{Z{~HY|JJs{YU^awa6W%6U}Z%w{d}n26fzYPh>)s3
zE%s_B$fSx&RlQy3ph8advg|R(n!)2>q^wg%$^)klnJ&AYUQoH<@qq$P#U!Z?Lx5n_
z2SDcb;;@zgTDfJta@`%5qAg{scQTer0=2t<+qXFHpJSr8S6_d<i<Tf9C5frX<3_Ib
z_NL15;as<(>hJq2UAznSSIAQ)o5(kvmrr)Tq?fD;Hh`KqexSqt@X3o8U++HIeg5@}
zOPYqjkh1+<Nm65VxvWP074FD2lg51>QW6tdGp5>%jD-SNN}H6FB2i9h{0&tVHCNg$
z^IceKYl{#(mm@nuwtywnCyWKOyviU6edX{h#H-X+%4&Vf?QdZx0F>o>(yBqzdNvO<
z0ELxVlASh<uy@amW}pWvrX(P=6vgOiQh_(C=ZQ_ii?Hnvf3OGMVc9bVtsdL3JZZlv
zF3KCP<5nYbanC>^1dvT)CO@IA6y#;}U#o+#f(fTq3PGx36|;syL&CvAX(UX?CwOhQ
zvu{QMWvUwGSi(K#i?r#DYFv<V)zws45^^(3=e(HxT!TB}R{lZ#R^esvorrcef|g=`
zS02KgPf;yJR5aDmT9!93&bR&U#Nt(wbjpFNzQ$FqqZSQ%uLt-zfB<u;h!9tm&E~cP
zf>m8H(FO_9r6t_H0)qN^<(mt;T-~kmJ^LLKxyQTL(GA&wT7>dd>kxrhwo5;I(b*Nn
zV*F7rvfbF`bzQ&)eztq7Hr3VIbyw-%_|hkT*;`vqe7SN{8cE1ctSTgq)+@+q<6&h*
zrbyt9vl)Qu+0p&NQ@q3MlHpEUaXCdV(A10M^@r{3Yr=*i45)r?AUQz1e?|3S2+ZV?
z2=qfLdfjnP$%*F!LdHQ%h)*pJk+BXBI~C{l>4B=F5X?%`csrd%+qFVFB>J!F%2I^s
z+hUcLJ+v_@6s4ZCAvS`tRJlN{+AX1)1z7RS(lz@cNjf$o!*EFbYf1lAI@Q>?nqkAn
zuwq()Vd?ZK7UvZ`s){WLF=7oFZfa=ksxqvX67kANv04`uLQ=(zg$g1PoxJjr%1Eic
zjEkM=63N?Q?)a(QdF1kzS1S^g7%GjdA!4VBhT5rn0gvkgT3x4LtX4|olEQ}KW_l``
z(oEBD98jqPDEtH{ve<sSIsWwVX0LztsV+4c|K*-EAh*wYHYlbP7?@*iK%Z5?cva>|
z1j2{E-6+&OA;KksXGr49-q!0I%r@6>gxeqf)(%2f+R7FJ{oHh}%LdYjPSUv7O7hR1
zaLt;u+iR>MvzuT`XnGF<80CM&eq(?Ql^O$Ln9+@X6AJpBA@bU_*-alY2!&6PMFqWr
z7f=$*qVEFNsbK@F`Q@!DjZoPs#+nnF3m3t+iV4Bt6eGf+uf9GPGpz_qC`avRrxHfn
zpTkN07G}Pfc-7WhOUSF;gO;PxeoBoRWY|{lDf54%#NyrV=((clTd{syc>IbZ9-Y3d
zoE@q3$A(R@u52xB`d^Gjr(e&grwnG<`JqDiUUZG(n_JuW$%nG_v2xNrAU<<{#e$-(
zqUXIImyY>-^gGkTFMauCGQP?pw`5*R%VN#1b9eN)%L1k*I6{`2Dp7`Kx!Rq~iui4$
z3}ag&flLBO*%a&EEy93)6J+mHKQi)?ao^4CE@!@~%Al7pUpIz@o=`Jg8S+6c$RZGN
zs*=kei2{m{7irHzc^W72A$fwKB2a)=8PH9JNrCAW%tn6U9i`9=mZ^zuT2U3Df?4U8
zBBAi`FfEd#Gb#WD4#Ic@_#Dr!j!UgGA(V@>`N8@4gH-E&Em6*cV<JXK*$VQ4gdUOU
z1Fox9>7gSp0W0^g0TW{zAXKOIeZV@F!Y;NvtJzppEkJgf6sHJOAa<V*{)!7)SaPTr
zVz~^t_F-Z&<)4K+7>`spoq~{7TDQX7;OY<B?Mr4^Rg5$Zm=(*!4~4?o6o`{wo0sIK
z4wy5gDib(WPE-zH+#ehF@VAm(IXpO8w57A4RAu5566CXyVs$=XA03SaJVLQP_CU<D
z5hJIHrgHIG4(N59HPN$B?avf<BA3Cl%9xZiOYv5=#=mea1qGB5@Fs7>IIeioxQr)M
zwGwLM?GHburzJbuRgXPO*)N*2!X+_&8;|u!(>--gWZ9kF@PdmWEvp#T?;$*LH&WSp
zl_qp2C3!<2F>uhkV3X#Ra2*Mc1Dm1>jtLh_%BU6a&7J_G0s^tBk{^y{U@@+G-Ww&|
zg{uP@!i*FKTm%D%ok+?-@$<0(50&nGe`r<2S&&2=_w@V<L1W|9(-$u)SxX*8F>Ox6
zC}Mh+Amxbd3HAww10O^yg)#r!qbV^06Fg)Qp<tTaUcu8PMha$%3`UiS%gq$G*ia!@
z>_}s40a@>00RyuX$TG$)0qJ2r?Vfm=iS6*{j<CB+m$AkqQRjGN3Y*26eMat-l0wK&
z9PYen$cMiZb!Wv}Wnh-7iMX0rC^5jlL6WOFlLktG84x>>0aM_`;g-w<GKg-FgZH%q
zy4^P@H8m-81b9*Rlxr$f;V#o6UGzUvg*sieK>W;HGQSo;vgnef#8+Ai&37>&&dsF-
zl~`gv(SOyZA%4{TR^3I=cAZ<L6Yv!koulE-AU{3q<$fkqgPoxkZha3Yc-PvmiE%TI
z&F%?|u<Iq;6uax#kLam{04Ckxumdq17So6KI!EJShYCX|0M4n|eL_tA3zBxF-#V&U
ze#jGhye6wte<&YR>=mFNxxGhanC(*y+#nHJz+hCC9n+tL8L1$A)rq0hisWYCn<##}
z?bGYvi80aV)T&Z&?=eW5kkheJb=f?0N*eh!D_7n3g&qwRxH<(Tk1&RT*IZXM8KPZ6
zuGbRK`REgAIc-%CL_Hps*_@vGJ$k%-@5div-$(8HM>klG9(24LZOE^b{Y{t3oNb>?
z$axOfsf4)IhyvesPm1mv(Gt5BB$vm8=mv7wt=Rc>>ts3|myb3#k0B4U!!|nYO^R}=
z@us)l#0kIHRQblusvj(*zNU}KVn`Zf@U7q;Q=+b3q7o&%&X`ngFnM;FxUyP;3GV@x
z1qcg8*oFLiy3N+I`moq3t2D_(xhrp39BVaGlh0)z#SY|82W^li4G>@6Fgv~KEgLhP
zZnx8|P*GNFB)Ri!bI`i!n!w$`NW;i;BfN<Ada9OLt{D+dJu3QJ8jzJ^t9IEs+fuvz
z;n_C2Kx`CV;T48EKVmN+si-+I_YTO~X9HU3#G8U5lW|>!n?W86H0v*ZRexrFRqg-v
zfBBb{|LmiW?EfWxRo&V2wEwol|FiK8dsPuaBv95he>$5I6k@>4zrQm+@v7S9KVdb<
z#Z3&92c&Y{3XqMlf0DP<2S|**|G_Kh1KQgm(Hx9E9*rfaHbq7B4n*i2!{i9dBL+*z
z0cA}jwR`E@BLrGw7ZE&Wc$6q`*W{LZmQdHBlmjs(=o~7{85Nt5CW+Uc-eb2SS$sCB
zoEogz1+FY&Gets-Wh%<McQt@Z!EQhvVgg<_t8a}vbWYSfV2mh4r!FXh%nA%U5TP=u
zI#s_ueFV}r(-gGs;6Tz*P`2Wl1#}>qOW4;0^1$Yq!$stw*_2zWkPRM5#v3g0T?#@5
z9HV<KJ;NPI`?I8Jb(XQQ*TdW&zWCi&ou|Kj^36A2d;`lz(VQ#n)^+4|SgV&!Q$p0?
z2#UWINEl+1-P}q?W0uQ;=ASi4#W!~iV*gvNRKaTq^GZ_#zvpZ?dUyDu-H~e-lTJE$
zpTF`9%{3?V?d=Ea8GoTTmtamxjlA5c>!(xS1-&@&UQvl{`u_JSyd{1>je;^<jN~2>
z@mpxX#nm&$3u+@+QjeDM{VK=egHNm+#Dj+v$KkhozfuCmX-}`Vw|}i?wgCM@#fWwL
zLNfjs274|@cu|0jUYj@-?gZNxZ-cwz1K3Oksj_2GLv55A;MdgaJJ`3h|27zc5Mqn8
zoQxMn#yYp)6hJ%k`&rL^kU-Uwj6{zg!|yI&`R{SG8x&%`9?Iq>=SV<6>^M9(=Y&R`
z=Th!MU1H9Pz(x!~tVW@nKn46Nq^-3#41n?0nay45;+~ZzI|Un)TaMim?l$c$0!gqV
zBDU-gg{d5qx3hmBGnOzpamd9H1PMupGYNu&vE-6;svgGYm{?<aXw!?Gk7k1&g5~H9
zPBr*tqUq&}=aSjb<GY?LEHAF^c4G>s;AViTtdSADS9>LND%efhtLjx{3#w_TKp5r8
z?C6N1RfZ_k-<1Q>uY~)Mv=lySklriBLLl`EkY5IE!=iG*jR}5t0`GSxdt7Hu8s}}Q
zF?<JvS`H^g{^sV$*0&~y2fjPp64bN~TiPv4>5_PsYP<FB1uVNeIOw7Tc$99=26Anj
z6FhDN{&KphczoDhur%=%H&nku60`k+gkwI@Z5psKJq%4!KO&*4i64yJpJ2Yx+0LY~
zGYw(5bFgNzdbv1Eza~o=^^Af77`@wljjnq)x+ebxdTmlz>k={pSf5Ht^tN{3?`Z1d
zc19ByExnT)rk<XR`rV?n=S`?bu2G%-UG=7ZR?s7gkDE3pOfFZszTP(Bg&ow{CJ@Jo
zqo}MSj&QdVs(4}-<*9AlmA~)z|L2G&UWmFz`(U_fqD5{<N$hyhd*oCK%53;ND5kls
z&MAzJu&S7XwpHQibrkRND0Tgq$zBUh>bwq0uw9Yd+GnHjw(?a&JC@vOHpCa`3(~t*
z<_OVMykjL=JvuO{q&*zgOL9{y1s$)7l8(j~?hRweu#(Y)gBHyT!i+C*Rx{Rw7HOg&
zu_cJq*%s`sCK*(SvQmUNHZyQ`V?DZegx@mn$*D-fL*#tmQF<fn2J%pzDxRZi65WO>
z48GtWWDdh1)|yAMHjyVd&NP;hw!pQvi+hl@Oy{b7*Cy?~yfA5;*>%fGX&f}g?a}0T
zQ*So0#1Lw=DJ<4LnVt^rRQPNoSUon5h6XTL(kE=RDRXmdn<y2c$!V>vBe((${du_{
zzmpXT#;s1+4J{Oos5lC%mr=;?Ry0Dm1M31FyE`{f=wy<@wOA~dOYn0cA3e#Kt@Mld
zW?8Gkr1{alb5-<yB}{%Ndx2OWT|O`<uao&LQ0gIEGZSB-)G=yDkD3}r&|6&@(sGDL
z`#IuhR1jF{Hn>4B$P@VjC~$!-CWW401jE4*dO5d@gIic-(m({4kIKd0tusg*cK;6!
zBH7y?I`&^v$Y5ZYu(}#*GB(UfpE(HOS#TUY8wCxu&<gV<zzkC|G3+p>7>Va+PzGZr
zmLF$OIfk;pKdL7MS@7=$>m{;b^Mdb#|8vJ|&pZ-*@At~r`zvylZfC!hOofoBg9Ycz
zPC@Xld5jqH;J_Ez!Agp}*KocR$QsJQTSs=_wzk$Cn<q(U7`%Ne^<>)E%DhI6#s9y(
zvx$-G%IbJc29S^ts{|21sHTUJ%RTPuxb0{(CeGNI8RJ$+CQLeZtZu8Sy1U$6q+Qji
z>bBh(vha!xLco$G${-{bXe2gBNGxW{2ni%uu|`5-Nm#Jq_dn-;y;uF=vZIch#7QOP
zs#o{jcklT=|8s7L356v~X0sKjXx*hhhyp^A5{glj9y2?a-yslrw(VFm53mw{<P0ZR
z`j(+2T~QOv-E868fu_tsp?IlhZr6mi@sGr^6zg%KBsGlsJP!h}PRGht!Y6)$#^no^
zJ=bsbXKii*#HM3#O_g|Pn6<g+)o)Q6Y;lNWcf|80RUW9{f4ny2-zcmr1a_#rwgmvl
z7$4>?@6!_yw9!9OdX8;9)>4nwY@lZ_1H^zP;cw@|-ODf#<F205f}gvaJzT;+_PI~!
z+}+i4VLIhS<k)L!_sb097c9g0=CxO^HeM#R$m`sp8AdSF_w%`~lg}s3_z(=G;SY%C
zk>Y=_RM@xTtuvKwJak2aZ;HZt_MBt=22YiETv0e?YoA65O1ddKMq!L6eSJ3yN6F&2
z$|nk^fqIq#jP>LA967UBY+LizQuF#d%k494`iI<?J>PWVD~3aJVZrn@?P3wbXb+ZZ
zGWodEjGr+JX0h)LB6ToVj|?Xs^;?{~ywjJ#i1DVn*Rd!-=2y3iIrR`@l`p}_ZjS)D
zU`-9pG{o>hTj#*JEO<e5poyeyW!{qu0;W8Ytu!(4QZ(p4pw`FT(`0_d%&^<WMq>u@
zNo(eYt49PY))WEwwPCouPQJi&l*YGcP%U+5HXcZ_m~i$nL$Z%3BnfVvK1pIz7G5kN
zO6IiZfms`2;)3dyxovYc{14M$afV(<$Eg>axW?-BYL7C#rZODEJ#FUh->ceBtKDwH
zVw?RdSDHBE+g{u+d&QJ{^R0vL^01@(EAe4kZaSrHjb5gHA}=&-D2KZ3+{3}jPY{CY
zx=w#U27_9%0_Zrj8R4PA&o63{KvBUzZrnp1OMz?28#7;?N<>Tk6o+$?Q_)feYkk8W
zwAeNy54jj2@SG))TXazzHE&(azXI&#(Gn7d)mAHs9;i_z`KPM%RDbBsfO$!I)KVjv
z?31}C^tD%Kf2~?{cw6!j7U9K%2yOw3>3J0pb+Fr>+TLLLA#&|5+5gL>2Vy$BD0jL9
z`A;|qhB5Mq0nkI><Zf7=ZO3Cqx<7*E*A3Vkx%Z0b9cLQ<19#Io8M<OcLl=Q0pa_&I
z+ozq&u8T*P^WViB_txf?&w<&D%wxV{>YWUH#X_i-z_qs@<E%`LU{v5!m8<3NK3=0W
zgzY4sk=V_@3I>I4e<b_1n>P(b>bOWFnCQyF7yDZrX^Y@DS(XP5Ty7j*Svhfe5$IJf
zc{1<c4V8QWM=&j;icJ3xQ97ghC6u6*9Y(lmHWWds`kT)G3mu?;sdiP)fAu0@0pSH$
zG|y$+ywJSO{Fj6d;6G{(w8h`EE84=D)kXXw?Mc@3?KejIR;cvG=uv+XfcgGo51hAF
z4WPgr)VhC36Ip$gcY1xxUKHQL+D1JV6}q}M_x_#t?|}36Pthq4dXsQ~WS9w)iL`J@
zdw~pcNvt<<uk=1%l_+xJ`YT>z-`AbP;|!7KD<+#VJU{S;Ag=h0wFonWRdHuo43t-g
zEw-<H9gMlk<vK;=#H=38Z97t1N=d@SWJPRi;MM_o2DF@69oj51cgi?S_K|B|qlcL4
z(F<%~XV=+<G7EetQ%*$mg!v1$K{ru*XMd==&s}cZS?{BuX=yO-Ruvxc&Ud|qo5{X0
z_>8goA-HXqBr0;$OiGEJa63PW&I*@Y_nTN&MwzgnEP|-<WkH&}N~2*w20P1cgAm82
zjlYv=`hz$9@uFF#x!I$-O-a->k2PRsY4rdSx;bK<5&YwtMPgYj!M;gi0bMtri1$=A
zPxKxxwmZo*)K!jHp&6TMP1{<A>xzI<k*dR)DZg=)<of=AD)C`+rSPe6q%t3;<PuBh
z1?Fx|s@262i*Cyjq`ECU0o&rinKV40@N`RLRY-W&NBK5@qGb7Q;JSujW`Sv<I}HyQ
zjcAn<B^%mJXNFT5jGnJ<MEeE&I!&Cz>-M;5Dk8!>|9^o4NL0pIM3xAshEf&1l)81n
zzc$w!k}(VhIA67@QafEHgB==!D@F4i8JzHEXRt-soz6=7B}#8?ZktkD8l#~+h<O%M
z3A=7u+%@wQQ19aTR`COyJHtS0#BKq`4%ly{H~<V3m-)E#nV`CvtHZTZ7Z5@{=?SP5
zx;pg`N%*fIL=s~_Hoi=<<gUnSn!e$=VqpoHzH;{z<$WnnZ~+Yo$Z@g3F9>3lGN~ZK
z6$MMrX2OxfJVT7k<hg;-kdX2URweLr9{1?tpY@&5x)=YLt&&QDLbH|0;)A8d91EoV
zr?wB7+S}~mBjY47a}$Pg^+UD=@rH!Jy_9I7h`W#bPo~M_6iDkKWiubxcTV|UJCDd@
z?$+tq(Q~W#yHq9}`w+2`srD3$hEQAV_z0S`0I6QVR4gJuK^!hkGRQREGlw9b(^^AE
z_>6)RZb04S!sNA4o2de#aLARuo}V|@#L?-+pcem7a3m^c^DGd8R&KOVG!mjpdw=3i
zcW`!n-E3Ce6fTm`oY<0D)@>nFfpI%!>ImB?{}oeTR&2@nN!&?hw4OD(BmK?u>7nK4
zFu+h0sEQ~yS8L_!LGc?XP|t&>SyK}F)b`M{7alIyF(DgNH=r2O?*2MQd@-K~3!G$W
zs&+J@Hq!zg+XL!>eZ^UH(~NtZIO6~cMRV*U9y^Hyzu1wNC{Q(YiitN!kjB?4-iO%w
z0$_SG7}~{+cKhneDzatgCi^HJFFITOox$GC{dGO|=vXQXOkqDp?@8L#0(M^}QNCbF
zl*YgP-RmzcG#Yv}4@sgVOZl|3Ct1kcxc+;h)LFfX&#^44zG$c#Im4epf}|MvsWKxJ
zI^A0F@=yy4`TNvx#T+eAXwu$-vw%xLJe#9&V$<do1p#AiQf@dMVd%)S%B@}d2B@E+
zh4PS0BptcYo4w&yf4nyL`fJ;lU%w3~3u*Y;BJW?@eSLm@Ikes6*tpGs3XV}=?+t0E
zZNYlEOv?R8cAJ{7SRtsSq6#<7(*o?icW466`$6*r%@FZEf(pCLw|tWjgDeC8(%iT=
zpf@!p!@~tN%GzN)X~5;fnI;r2Ma4QCU1$+iX`DDN-Y_nv!OgCit5~})AsT)s2u>VO
z8VnE##|K$8e|iWhss25x-dGf2lhuC*&EY>gDstRbKXN?9Jq&~c<<8XdLe3u3y|+&t
zHS1-FKhcRjLYT(*z_p_?NIQK&URa&}z;mC%*r}^=>13IwWC7{T!5A-{$+@d0j*(tY
z4^nA<SJPW(C%zL>WfUZNuP%wUS(9-yJp40e*>1~@lwFl};X9ge`kQ_CQGYxlGqj6Y
zW}>AKCuciEA0DS3JB^W5#JWf4xF9u4_g!uIPq`vFiG@RHzk&&@2*pLPypiB|3ioaO
z+B>`S_ohOM;cZ=u4^N2#-jkFuF5DiCFreW_gc9T+F$L#iP+y@rLO?aUd0FIzB}F8D
ztpAm<Hdtbk?6fuIQ!zK8H!#n%${~vsOHaLSM&G(yWf)y8<RQ08%ie|Pc2%rJ$Vz^r
zYRRf{&`5Fqh&EPf9N#|rqMm%^Egi)0L|0ZK>QD3wbtd9M6~#1-2j$BxOZixMlul_u
zjn~-GU1!|ddwmp+WkoE{GwyF$v9uG?q`BZ*IX};4q9?=0?X}_eb8DmMt%ZeUyYV-A
z>cFLKHy<&=*`0>Fo>M|`kbP<?m8>4FN+bp~)1KF7v})@?`mXb=L6V7o29FP=4%vMi
z!IR>&u7<4h-PQ=I>doA+{aM7LZO%;sx9rL?G#=tlEKz(8+Z-YJ-!q7%NIG4nrD3^y
z@It&iq{0QKSG+fh(*@f>aRHlFxb@_#jRoa|rv314eeJLzy2T8l;`z#i94T8@`9q%R
zq7qL;0bjYxALOS8&+J#Tq<@};1m5)%NF>b+ui1>%2L?<b#DWOgHE2m?=ZX8Pm#Xcv
zh@#uI+E+g{g&+e}N_wyGRJS{(DHWCxFv@Wc{0h$*VaYR7`}4KGR`E=ZPh};CXwd0R
zChRJ)e%1A6>K<BIpjdnvBB_-1DUHZ2PKzX#k&&C5I2xnjg&J2z{a9Bvw$V|bR@-gn
zOHCtqMj{fAkUjBd$({@fj0m+`(`VZPJCZu`^C~P>9+|?)DR{7)3%}R?{oO5c169%B
zsZ$|_%*+x&ofc`lD-==`_7fa2W9&pObfb#>8zoS1wghN!D{C~y5aX?Xo|FhJ7c0rD
zOb<tqQPmWQb5P}#Bow#d{8`yzmbX`ms%U0P3hbkp%~O=NY-;Ko)#`RJRAO!)_z{v0
zz}zI_@Iuoq?_e)qnF?7dt3uaGvn<1^*rUoe?8}Ev;a;0(6%d5ca)y;$NbFSPUuBPC
z8|hW+<qV|b(q+mKE`R$~X+LQ}UiQDccSH8f_wH)pk`x2pu#IgCju9>|Htec1g@FxL
z7nZJo{!h%~6Fn_f$wm{VV-|wS)yuHdmfs2BMAM~&AdEB)JI<u``+RJPxJe^oKDhxL
z8b#vPSr_EQt2wF@!XKd4ESZ%O_@3S;9PbCj#R8vAI*S<FB4x;Q*?|rjp9yjb*=2=C
zbcPGTsYL1&CK0I0hZvAY2-6j>h6HB-ah2JNGi&b&a+%q|d}H61mgszm*Ck$;IV?!6
z^U7;$1d_i57YEAiYM|ZhV&~j$d2K-`1;>C8NI%S1w+ptBhwis1Kb>J8p2Y-3hGw;4
zNQ68|fq7lDTgnS*-tGECWnQW?kSJ`!I)F&mK=`4bNU|}_w2e^ea2$s@KFh#aRoNHy
znGGixi@qQq#y1|c0H?RJOF1JAc#9*C+zWIIIeU{wX{~sZ*lV1bTEsLUKX6KgGw~aq
zZG$HUfii{qh7o(oR9>e}Hl+FP_Tb)L#0iisyhPjVM^s-aMzfA{(!6PA=It8q_~^}V
zHVt$)efLaEN5!iYwJu^?l^+~6<x}E~BQpsDTC;>a!rNq(V-mq)_X)3HN%bC0Ha7?=
z`DvapkdeX|BH#q5Qeezd{uT?Dls3rI0y|KNv-9gL<&}AekuOx4*KHbc2goit7#lBR
znKZI7w2W7e35bt_wHRSV&6Y7mx9e=7DO)po=91W{d%hA>pUIAT7NI5e{s!V_nO-eq
z2udkJ;@*TsRPVG+3Od>z$2Edd9g2Fy2T{qd1-n$kY6%FID1IqZHN<8{^haX7x!R!_
zTFIGAhP5FD2Rd|(tq$k}2l<K3vZd0A{(o0x3o|Y2^W~#wZpK9~WkuzSLNu9$mcRUr
z6!0m@i|ad<K=X5Zof8#zeL~}s%od%kRYuYgV`8MjioNC*z$9EdXz<=(3uNe?<sg+n
z6ZJjZ9)Tdl(IL8X-%*455m)c^A1V7b#7%{7>2iU(9UzI+zCv2r+EO?rhOQK--QM2h
zj8mM&w+n)*tty^=;_JCs6O`$#Mx`g{v(H&QNebTH21qO$jsLhmfcUbw@%%i#T2tqm
z$YoP2dq{xeT4@{FJ~6Nq*9aoCK)nLW$wDgfbJP$ZM+r42dkjJLV6^i<LbM1kJ*i>l
zLH#WycgC>@V{N$F=~&FJ6BS*jlW|G``Zm>?1-qg@S%eAbY|U)EOH2AICs@<@+q~1!
z-iUS|(MKB!yC4>t&@^$!$^f@vo#=g091~B3>pd9J)7o6Qh-_}F5iC2Q);4N?M@I|w
z;S;kyi7Jns__+yk1OB%=h|N`vsA-7j)%#qwH@3E$-iDzTz<`9g^a#pay&pI*`y7UD
zQD?{W)IN5F0ee|gK|eOveNqW@_Sb~jQJSE&$?S-Rl)Eo?*lMn<BcaYUjp{N+5>MW0
zGdl#qXLg)<M;(_xmXVmLVFVu}Tii2t*Kql0#8uktJc7-KhUNmoU9lBUh-@~b+XY0N
z(n?;%hGafz&dUn&bzvtUs@i#Vl0tOSi_uSXJ?Zy}jY2*ViV~bpG<0x5>#7P;+b2VY
zBAtU+5AM0g0r;+L2j#dwoyN0Izv=b9Rznxcm^*1g{y?4@Q2m2qW8{SBE(5!$JrD5)
zt)NlLvzWh7Aj_T+i(Y1>^SAv>`+nbwHApjVs{1^t!DI%j$3C;t&pFL|EZ=|!WIQW7
zFk3c+OAfeVi|&XDma}xJhjF0dka%={KnZ(m`~;6g976HJlCq`XBp(cvAmg@%Z^Z5b
zKyvb-0S<;-{MMz5->#Q^e7*VoN2;LJ?8JHS9rPnP3b=>Uw)*K?&A@KGHkS)Z=mRsR
z9+`<;5t#i-*EuWX=Gq%XM*Fzohkm&?>Jpv{UwOX6&G@#g&W-Jm9Xj;4m3&C2i`t-p
z4AU3gMuIkKGa^|Y+L20G0oX>w>%zHA-7bNxHhGPXrFG3cNlc`1jd)eEVpj?}L}FBO
z$3z2=VrTsRCL+nAzoK~3`#T&wP~O@`c6kVOGiQq-oNXK0L*tbQudy+8boDeQ`po_F
z_#6AbFZ{*xZH`&&w3h6muX_T63U$aogP~c|1z%Jp55<Dk{ipDh?AES;WH#9i6hGkF
z9I}DZ;Y~`5JgREWrXkM!=r+qm>jrH7fsl1%GI~1~%mvd}xgSw5KA^N|Q_OB%iEG&0
zOsC`Dc_%1)*y-P<sn?<s7UMW1B60M+!q}}eqM~_r`S~{DUZ>53OVSku4kuk)d{%Q-
zzJyeN!64QD^vD0c`01}T8hRbqmnM53gWZGS`i~%#xee->CR8{1Np1Q_f9zv(D*5V}
zj(rn%3bKN}s`aIb65l~6l0!3_4>B_8UT(ej*0X`n-+SwAnOr?g_V(6jL>*n5P66R4
zpZYOARs*xOy|9bzHt~`kT1e^mM+)dVwpJZE2kSaFa8YLfcf87^IQZII{lL3gK5UQA
zsxeL8v!{Yzn{(lfX=JBF#IpR0Dw@{hn9lYim)-B(CylXgHK#h=NZK6iuR4dv_BNrr
z4N>uGv!Uvyy4eVrSKDtTwB~|LwY^Ec4kk}gt6Aw>)NY0rxQbK^hWprfWm&G`k!Iv6
zY=KoU+1-BB+1eR*NXw5p!Lpy^daOyu_ObB$y%B0nmYzA7ubFPsy>n17!YURq{po88
z9A&)x)cWf{S!Ri){yOGCh7%=Br=$9TqM5Td7-APIGUe?=u?AI6b}|$p5@**%4*r98
zDiQLbPhq=SI0!q+z11g#+li_oK;BEHM_QVIGY@XF%n=06$U2kYHB8)XNbZRSuv}_-
zxZ;5}dNd6pd>}}|g3KK33)PydrLT8$H1;VEwm!;Vt|V!~Gs?<ZO*8IqW+6FqVPld+
zi=Fk*uz@#^R3m71{h8$)-;K8pB6+k5p%6cdfj>`LN?F}M=xnM(c=0=71L+ETN%>;?
zdWI!=mgI(d)f;yZaWEN3>lP$7@L8PdP!Aa&#7W<ZU@a_#K!)bm;hx+C$RWS-vQiAn
z#EQeovh`2yDA?phLMikqQy55dvO7f<N`we`(?fR%V%Q-HA?O%UL2#4@GGLqNYFG((
zy5Ufz*f0n#uRnoi8-sm1Wl!9z0E``_K|dzPYID*YKKr2wn0TKF81k}JHb7>MM(>a5
zjHwBGvuwmwy&DnXG@UxlQc{0o&B7DAhewst)igEp@!%uW7!%b&w(3Na!=XI6_AS#@
z+%m0ooXz1+?JeTYaP}wnhmO$2a$BNSD)yPU-wko&+VI*9EWXNCweofJgW2|#F5Ev(
zn3-J7%M8=1<&j<2fqjz{Xc;eGXaZf=NuKlbAJEus%hDb)_8j=RHg`iu5@VHoEuF@x
zND-%iN>H?Q_twJ2U?6&MqR_@|OJnhakiW7Yofh_KMWyAjx(kjxIGjV00<g&%(gSTr
zpqm8c$<TS!q4iQQ7U*_si%GIr>;asED3thC4ZlJwMbWzV_EO5zfFd0(R(Cy?+Zl3J
z*J0TUnXP#?@=u`W3hgi-G&-mf=Aws0U||?@J2&tNk<(jWak}14rY)t})s29lvfk80
zQX!2bxnzuwWK#Qx9+)?1nTc?_sYWQJ+B`-NZD!^HleQIc75R0$*J^Vw>HbFB==nmk
z-EIS8t0mv36h|B>=nfZLCc>4L^wp{gT11XjX($kvd5M*X%`DD#D9!k=s^vX6id@=M
zTdP!>1!s4-Ff6#sjMHM^>u3hZOM6s;6OXcpOPSjKArCUTUu{-Wz(wG0JJlK<B=|2&
z6_0x+1!@E+dazVCy%f$iMTEZ8hb>!c_o@WE$YAeu31C+7ciV_XVav$z*&>r1Ca)2R
zBw8PVz6_(3f8;W#FmQV4$1EF_IbK0vcp^E_C~3|-K)z9sgKA|iuO+57bs$KdjEaLZ
zC5a?01Z40SY(Av=FbLE<2}?PnI^v<`{v_2|C}jjJqVgbHWS%fABxKD+s0mLZTO5rT
zOJk(If<|&9l{~_}6bAC!dQwI6nx<iU%-K>?<8Z^6>?viv1glxGc$tcLOJ`^}sST}9
z=nx{K5H*3)0U_i^eQwA;QQjloQu$3DdTXD_!i5U4V;xJmB-ABrZ9TI<HRpw6S`SCc
z^x_|n`}a|YxD;dF>b!3x4h^nk;sAy;*gW_oQZyT8s5dtbHi)_9q|8Gm=cha8X+SBy
zXvJJ@d3*(#vbJ02ZxIPE=NgkszO*Pp@~tFZ@p=5w*lZGJzV0~lYCXf*)J0^8F_nGL
zuq1jtD}>HU%VF-iA`DL>>GA@`5C9SV&}lV#X4<yM1T3XR64W3d5Ma+JVg~o|`2ukr
zXGfp6mZ8JZhk!lAT4&#UhLPMh5_@*LRSN*BGFui}=TMD#Ot&c5x?R_d;etJa+3GE*
z2!rwFk;Ez6(g-G+X8~6orAY*nKy}`aIiRR-nE#3BaA|C29sJ5EC4vSpXuy9ujh44i
zy9hP2JIri{*H{4hEllW@X6^UjUIAKQMl+!UPR0NX#k7$vvhkIA1wLT658b1Iaa8!-
zdAg7We#^k%aDQv1zp*{^16jB^#c<k)=%)cgRkJD@gy`JK6zV%PX!T|4><f@O`^pP1
ze)UB<4>cMafAGEAfBn+;{_1D*jfP&oX`lLkK6$MFTaj(Z`HTQ&-~x511AXY}o_F5N
zV{J0lpMLs{pZ?|fMnf<DO@BUl{EOeLjz#AzpeiGo9i5_)-t@42hJEel{`?RBj58<n
zx{wBKG(LI!=}L9j$;NoFtE82~7>abq^vK(c{I75S_{VQwYc%xw^~%WqJgknau4JQ5
zZ+9>~(1?M4`+v6o<fRT>op^ntGSHpBsSh+EbwBvX%!G~~h-xXeaxqsJ^4hJ{zqs>C
zqoEi7razxN{>d+YrIAd+usQkfcKVCfm1Z?#)?#|XcNqO|Z~gCo?*DG1p;tSNtu=lB
zJI9Wm;(h|@Opm?F*#G^HFHe4jRnY6VDr3L=N5_ucU+Nz|V*4Nd{=>Q7Xf*6q8S%rP
zOph22cC^zb4<E07l1Uu$lgIz^v%|)#ACfV(VS4O(XRkcyul~pMpeej|#3I&{+Ih4O
gUVLF{`;W-SzVqhK|AKz_OZ@X+UxhIjU*K>52PYWqW&i*H

literal 0
HcmV?d00001

diff --git a/0.16/_downloads/8b062cf0fa8a991b8cff18fa361c8ffe/plot_custom_tv_tensors.ipynb b/0.16/_downloads/8b062cf0fa8a991b8cff18fa361c8ffe/plot_custom_tv_tensors.ipynb
new file mode 100644
index 00000000000..8e754f6c721
--- /dev/null
+++ b/0.16/_downloads/8b062cf0fa8a991b8cff18fa361c8ffe/plot_custom_tv_tensors.ipynb
@@ -0,0 +1,140 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# How to write your own TVTensor class\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_tv_tensors.py>` to download the full example code.</p></div>\n\nThis guide is intended for advanced users and downstream library maintainers. We explain how to\nwrite your own TVTensor class, and how to make it compatible with the built-in\nTorchvision v2 transforms. Before continuing, make sure you have read\n`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nfrom torchvision import tv_tensors\nfrom torchvision.transforms import v2"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We will create a very simple class that just inherits from the base\n:class:`~torchvision.tv_tensors.TVTensor` class. It will be enough to cover\nwhat you need to know to implement your more elaborate uses-cases. If you need\nto create a class that carries meta-data, take a look at how the\n:class:`~torchvision.tv_tensors.BoundingBoxes` class is [implemented](https://github.com/pytorch/vision/blob/main/torchvision/tv_tensors/_bounding_box.py).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class MyTVTensor(tv_tensors.TVTensor):\n    pass\n\n\nmy_dp = MyTVTensor([1, 2, 3])\nmy_dp"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Now that we have defined our custom TVTensor class, we want it to be\ncompatible with the built-in torchvision transforms, and the functional API.\nFor that, we need to implement a kernel which performs the core of the\ntransformation, and then \"hook\" it to the functional that we want to support\nvia :func:`~torchvision.transforms.v2.functional.register_kernel`.\n\nWe illustrate this process below: we create a kernel for the \"horizontal flip\"\noperation of our MyTVTensor class, and register it to the functional API.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.transforms.v2 import functional as F\n\n\n@F.register_kernel(functional=\"hflip\", tv_tensor_cls=MyTVTensor)\ndef hflip_my_tv_tensor(my_dp, *args, **kwargs):\n    print(\"Flipping!\")\n    out = my_dp.flip(-1)\n    return tv_tensors.wrap(out, like=my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "To understand why :func:`~torchvision.tv_tensors.wrap` is used, see\n`tv_tensor_unwrapping_behaviour`. Ignore the ``*args, **kwargs`` for now,\nwe will explain it below in `param_forwarding`.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>In our call to ``register_kernel`` above we used a string\n    ``functional=\"hflip\"`` to refer to the functional we want to hook into. We\n    could also have used the  functional *itself*, i.e.\n    ``@register_kernel(functional=F.hflip, ...)``.</p></div>\n\nNow that we have registered our kernel, we can call the functional API on a\n``MyTVTensor`` instance:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "my_dp = MyTVTensor(torch.rand(3, 256, 256))\n_ = F.hflip(my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "And we can also use the\n:class:`~torchvision.transforms.v2.RandomHorizontalFlip` transform, since it relies on :func:`~torchvision.transforms.v2.functional.hflip` internally:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "t = v2.RandomHorizontalFlip(p=1)\n_ = t(my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-info\"><h4>Note</h4><p>We cannot register a kernel for a transform class, we can only register a\n    kernel for a **functional**. The reason we can't register a transform\n    class is because one transform may internally rely on more than one\n    functional, so in general we can't register a single kernel for a given\n    class.</p></div>\n\n\n## Parameter forwarding, and ensuring future compatibility of your kernels\n\nThe functional API that you're hooking into is public and therefore\n**backward** compatible: we guarantee that the parameters of these functionals\nwon't be removed or renamed without a proper deprecation cycle. However, we\ndon't guarantee **forward** compatibility, and we may add new parameters in\nthe future.\n\nImagine that in a future version, Torchvision adds a new ``inplace`` parameter\nto its :func:`~torchvision.transforms.v2.functional.hflip` functional. If you\nalready defined and registered your own kernel as\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "def hflip_my_tv_tensor(my_dp):  # noqa\n    print(\"Flipping!\")\n    out = my_dp.flip(-1)\n    return tv_tensors.wrap(out, like=my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "then calling ``F.hflip(my_dp)`` will **fail**, because ``hflip`` will try to\npass the new ``inplace`` parameter to your kernel, but your kernel doesn't\naccept it.\n\nFor this reason, we recommend to always define your kernels with\n``*args, **kwargs`` in their signature, as done above. This way, your kernel\nwill be able to accept any new parameter that we may add in the future.\n(Technically, adding `**kwargs` only should be enough).\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/8c66a0169c2c41937d8db4238842fe5b/plot_optical_flow.ipynb b/0.16/_downloads/8c66a0169c2c41937d8db4238842fe5b/plot_optical_flow.ipynb
new file mode 100644
index 00000000000..ce2b6ef7b7c
--- /dev/null
+++ b/0.16/_downloads/8c66a0169c2c41937d8db4238842fe5b/plot_optical_flow.ipynb
@@ -0,0 +1,176 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Optical Flow: Predicting movement with the RAFT model\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_optical_flow.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_optical_flow.py>` to download the full example code.</p></div>\n\nOptical flow is the task of predicting movement between two images, usually two\nconsecutive frames of a video. Optical flow models take two images as input, and\npredict a flow: the flow indicates the displacement of every single pixel in the\nfirst image, and maps it to its corresponding pixel in the second image. Flows\nare (2, H, W)-dimensional tensors, where the first axis corresponds to the\npredicted horizontal and vertical displacements.\n\nThe following example illustrates how torchvision can be used to predict flows\nusing our implementation of the RAFT model. We will also see how to convert the\npredicted flows to RGB images for visualization.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import numpy as np\nimport torch\nimport matplotlib.pyplot as plt\nimport torchvision.transforms.functional as F\n\n\nplt.rcParams[\"savefig.bbox\"] = \"tight\"\n\n\ndef plot(imgs, **imshow_kwargs):\n    if not isinstance(imgs[0], list):\n        # Make a 2d grid even if there's just 1 row\n        imgs = [imgs]\n\n    num_rows = len(imgs)\n    num_cols = len(imgs[0])\n    _, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)\n    for row_idx, row in enumerate(imgs):\n        for col_idx, img in enumerate(row):\n            ax = axs[row_idx, col_idx]\n            img = F.to_pil_image(img.to(\"cpu\"))\n            ax.imshow(np.asarray(img), **imshow_kwargs)\n            ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])\n\n    plt.tight_layout()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Reading Videos Using Torchvision\nWe will first read a video using :func:`~torchvision.io.read_video`.\nAlternatively one can use the new :class:`~torchvision.io.VideoReader` API (if\ntorchvision is built from source).\nThe video we will use here is free of use from [pexels.com](https://www.pexels.com/video/a-man-playing-a-game-of-basketball-5192157/),\ncredits go to [Pavel Danilyuk](https://www.pexels.com/@pavel-danilyuk).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import tempfile\nfrom pathlib import Path\nfrom urllib.request import urlretrieve\n\n\nvideo_url = \"https://download.pytorch.org/tutorial/pexelscom_pavel_danilyuk_basketball_hd.mp4\"\nvideo_path = Path(tempfile.mkdtemp()) / \"basketball.mp4\"\n_ = urlretrieve(video_url, video_path)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        ":func:`~torchvision.io.read_video` returns the video frames, audio frames and\nthe metadata associated with the video. In our case, we only need the video\nframes.\n\nHere we will just make 2 predictions between 2 pre-selected pairs of frames,\nnamely frames (100, 101) and (150, 151). Each of these pairs corresponds to a\nsingle model input.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.io import read_video\nframes, _, _ = read_video(str(video_path), output_format=\"TCHW\")\n\nimg1_batch = torch.stack([frames[100], frames[150]])\nimg2_batch = torch.stack([frames[101], frames[151]])\n\nplot(img1_batch)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The RAFT model accepts RGB images. We first get the frames from\n:func:`~torchvision.io.read_video` and resize them to ensure their dimensions\nare divisible by 8. Note that we explicitly use ``antialias=False``, because\nthis is how those models were trained. Then we use the transforms bundled into\nthe weights in order to preprocess the input and rescale its values to the\nrequired ``[-1, 1]`` interval.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.optical_flow import Raft_Large_Weights\n\nweights = Raft_Large_Weights.DEFAULT\ntransforms = weights.transforms()\n\n\ndef preprocess(img1_batch, img2_batch):\n    img1_batch = F.resize(img1_batch, size=[520, 960], antialias=False)\n    img2_batch = F.resize(img2_batch, size=[520, 960], antialias=False)\n    return transforms(img1_batch, img2_batch)\n\n\nimg1_batch, img2_batch = preprocess(img1_batch, img2_batch)\n\nprint(f\"shape = {img1_batch.shape}, dtype = {img1_batch.dtype}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Estimating Optical flow using RAFT\nWe will use our RAFT implementation from\n:func:`~torchvision.models.optical_flow.raft_large`, which follows the same\narchitecture as the one described in the [original paper](https://arxiv.org/abs/2003.12039).\nWe also provide the :func:`~torchvision.models.optical_flow.raft_small` model\nbuilder, which is smaller and faster to run, sacrificing a bit of accuracy.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.optical_flow import raft_large\n\n# If you can, run this example on a GPU, it will be a lot faster.\ndevice = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n\nmodel = raft_large(weights=Raft_Large_Weights.DEFAULT, progress=False).to(device)\nmodel = model.eval()\n\nlist_of_flows = model(img1_batch.to(device), img2_batch.to(device))\nprint(f\"type = {type(list_of_flows)}\")\nprint(f\"length = {len(list_of_flows)} = number of iterations of the model\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The RAFT model outputs lists of predicted flows where each entry is a\n(N, 2, H, W) batch of predicted flows that corresponds to a given \"iteration\"\nin the model. For more details on the iterative nature of the model, please\nrefer to the [original paper](https://arxiv.org/abs/2003.12039). Here, we\nare only interested in the final predicted flows (they are the most accurate\nones), so we will just retrieve the last item in the list.\n\nAs described above, a flow is a tensor with dimensions (2, H, W) (or (N, 2, H,\nW) for batches of flows) where each entry corresponds to the horizontal and\nvertical displacement of each pixel from the first image to the second image.\nNote that the predicted flows are in \"pixel\" unit, they are not normalized\nw.r.t. the dimensions of the images.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "predicted_flows = list_of_flows[-1]\nprint(f\"dtype = {predicted_flows.dtype}\")\nprint(f\"shape = {predicted_flows.shape} = (N, 2, H, W)\")\nprint(f\"min = {predicted_flows.min()}, max = {predicted_flows.max()}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing predicted flows\nTorchvision provides the :func:`~torchvision.utils.flow_to_image` utility to\nconvert a flow into an RGB image. It also supports batches of flows.\neach \"direction\" in the flow will be mapped to a given RGB color. In the\nimages below, pixels with similar colors are assumed by the model to be moving\nin similar directions. The model is properly able to predict the movement of\nthe ball and the player. Note in particular the different predicted direction\nof the ball in the first image (going to the left) and in the second image\n(going up).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import flow_to_image\n\nflow_imgs = flow_to_image(predicted_flows)\n\n# The images have been mapped into [-1, 1] but for plotting we want them in [0, 1]\nimg1_batch = [(img1 + 1) / 2 for img1 in img1_batch]\n\ngrid = [[img1, flow_img] for (img1, flow_img) in zip(img1_batch, flow_imgs)]\nplot(grid)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Bonus: Creating GIFs of predicted flows\nIn the example above we have only shown the predicted flows of 2 pairs of\nframes. A fun way to apply the Optical Flow models is to run the model on an\nentire video, and create a new video from all the predicted flows. Below is a\nsnippet that can get you started with this. We comment out the code, because\nthis example is being rendered on a machine without a GPU, and it would take\ntoo long to run it.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "# from torchvision.io import write_jpeg\n# for i, (img1, img2) in enumerate(zip(frames, frames[1:])):\n#     # Note: it would be faster to predict batches of flows instead of individual flows\n#     img1, img2 = preprocess(img1, img2)\n\n#     list_of_flows = model(img1.to(device), img2.to(device))\n#     predicted_flow = list_of_flows[-1][0]\n#     flow_img = flow_to_image(predicted_flow).to(\"cpu\")\n#     output_folder = \"/tmp/\"  # Update this to the folder of your choice\n#     write_jpeg(flow_img, output_folder + f\"predicted_flow_{i}.jpg\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Once the .jpg flow images are saved, you can convert them into a video or a\nGIF using ffmpeg with e.g.:\n\nffmpeg -f image2 -framerate 30 -i predicted_flow_%d.jpg -loop -1 flow.gif\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/8faa1256268baabef0c4b947064f34f4/plot_transforms_getting_started.py b/0.16/_downloads/8faa1256268baabef0c4b947064f34f4/plot_transforms_getting_started.py
new file mode 100644
index 00000000000..c61d1cc1be0
--- /dev/null
+++ b/0.16/_downloads/8faa1256268baabef0c4b947064f34f4/plot_transforms_getting_started.py
@@ -0,0 +1,266 @@
+"""
+==================================
+Getting started with transforms v2
+==================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_getting_started.py>` to download the full example code.
+
+This example illustrates all of what you need to know to get started with the
+new :mod:`torchvision.transforms.v2` API. We'll cover simple tasks like
+image classification, and more advanced ones like object detection /
+segmentation.
+"""
+
+# %%
+# First, a bit of setup
+from pathlib import Path
+import torch
+import matplotlib.pyplot as plt
+plt.rcParams["savefig.bbox"] = 'tight'
+
+from torchvision.transforms import v2
+from torchvision.io import read_image
+
+torch.manual_seed(1)
+
+# If you're trying to run that on collab, you can download the assets and the
+# helpers from https://github.com/pytorch/vision/tree/main/gallery/
+from helpers import plot
+img = read_image(str(Path('../assets') / 'astronaut.jpg'))
+print(f"{type(img) = }, {img.dtype = }, {img.shape = }")
+
+# %%
+# The basics
+# ----------
+#
+# The Torchvision transforms behave like a regular :class:`torch.nn.Module` (in
+# fact, most of them are): instantiate a transform, pass an input, get a
+# transformed output:
+
+transform = v2.RandomCrop(size=(224, 224))
+out = transform(img)
+
+plot([img, out])
+
+# %%
+# I just want to do image classification
+# --------------------------------------
+#
+# If you just care about image classification, things are very simple. A basic
+# classification pipeline may look like this:
+
+transforms = v2.Compose([
+    v2.RandomResizedCrop(size=(224, 224), antialias=True),
+    v2.RandomHorizontalFlip(p=0.5),
+    v2.ToDtype(torch.float32, scale=True),
+    v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+])
+out = transforms(img)
+
+plot([img, out])
+
+# %%
+# Such transformation pipeline is typically passed as the ``transform`` argument
+# to the :ref:`Datasets <datasets>`, e.g. ``ImageNet(...,
+# transform=transforms)``.
+#
+# That's pretty much all there is. From there, read through our :ref:`main docs
+# <transforms>` to learn more about recommended practices and conventions, or
+# explore more :ref:`examples <transforms_gallery>` e.g. how to use augmentation
+# transforms like :ref:`CutMix and MixUp
+# <sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py>`.
+#
+# .. note::
+#
+#     If you're already relying on the ``torchvision.transforms`` v1 API,
+#     we recommend to :ref:`switch to the new v2 transforms<v1_or_v2>`. It's
+#     very easy: the v2 transforms are fully compatible with the v1 API, so you
+#     only need to change the import!
+#
+# Detection, Segmentation, Videos
+# -------------------------------
+#
+# The new Torchvision transforms in the ``torchvision.transforms.v2`` namespace
+# support tasks beyond image classification: they can also transform bounding
+# boxes, segmentation / detection masks, or videos.
+#
+# Let's briefly look at a detection example with bounding boxes.
+
+from torchvision import tv_tensors  # we'll describe this a bit later, bare with us
+
+boxes = tv_tensors.BoundingBoxes(
+    [
+        [15, 10, 370, 510],
+        [275, 340, 510, 510],
+        [130, 345, 210, 425]
+    ],
+    format="XYXY", canvas_size=img.shape[-2:])
+
+transforms = v2.Compose([
+    v2.RandomResizedCrop(size=(224, 224), antialias=True),
+    v2.RandomPhotometricDistort(p=1),
+    v2.RandomHorizontalFlip(p=1),
+])
+out_img, out_boxes = transforms(img, boxes)
+print(type(boxes), type(out_boxes))
+
+plot([(img, boxes), (out_img, out_boxes)])
+
+# %%
+#
+# The example above focuses on object detection. But if we had masks
+# (:class:`torchvision.tv_tensors.Mask`) for object segmentation or semantic
+# segmentation, or videos (:class:`torchvision.tv_tensors.Video`), we could have
+# passed them to the transforms in exactly the same way.
+#
+# By now you likely have a few questions: what are these TVTensors, how do we
+# use them, and what is the expected input/output of those transforms? We'll
+# answer these in the next sections.
+
+# %%
+#
+# .. _what_are_tv_tensors:
+#
+# What are TVTensors?
+# --------------------
+#
+# TVTensors are :class:`torch.Tensor` subclasses. The available TVTensors are
+# :class:`~torchvision.tv_tensors.Image`,
+# :class:`~torchvision.tv_tensors.BoundingBoxes`,
+# :class:`~torchvision.tv_tensors.Mask`, and
+# :class:`~torchvision.tv_tensors.Video`.
+#
+# TVTensors look and feel just like regular tensors - they **are** tensors.
+# Everything that is supported on a plain :class:`torch.Tensor` like ``.sum()``
+# or any ``torch.*`` operator will also work on a TVTensor:
+
+img_dp = tv_tensors.Image(torch.randint(0, 256, (3, 256, 256), dtype=torch.uint8))
+
+print(f"{isinstance(img_dp, torch.Tensor) = }")
+print(f"{img_dp.dtype = }, {img_dp.shape = }, {img_dp.sum() = }")
+
+# %%
+# These TVTensor classes are at the core of the transforms: in order to
+# transform a given input, the transforms first look at the **class** of the
+# object, and dispatch to the appropriate implementation accordingly.
+#
+# You don't need to know much more about TVTensors at this point, but advanced
+# users who want to learn more can refer to
+# :ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.
+#
+# What do I pass as input?
+# ------------------------
+#
+# Above, we've seen two examples: one where we passed a single image as input
+# i.e. ``out = transforms(img)``, and one where we passed both an image and
+# bounding boxes, i.e. ``out_img, out_boxes = transforms(img, boxes)``.
+#
+# In fact, transforms support **arbitrary input structures**. The input can be a
+# single image, a tuple, an arbitrarily nested dictionary... pretty much
+# anything. The same structure will be returned as output. Below, we use the
+# same detection transforms, but pass a tuple (image, target_dict) as input and
+# we're getting the same structure as output:
+
+target = {
+    "boxes": boxes,
+    "labels": torch.arange(boxes.shape[0]),
+    "this_is_ignored": ("arbitrary", {"structure": "!"})
+}
+
+# Re-using the transforms and definitions from above.
+out_img, out_target = transforms(img, target)
+
+# sphinx_gallery_thumbnail_number = 4
+plot([(img, target["boxes"]), (out_img, out_target["boxes"])])
+print(f"{out_target['this_is_ignored']}")
+
+# %%
+# We passed a tuple so we get a tuple back, and the second element is the
+# tranformed target dict. Transforms don't really care about the structure of
+# the input; as mentioned above, they only care about the **type** of the
+# objects and transforms them accordingly.
+#
+# *Foreign* objects like strings or ints are simply passed-through. This can be
+# useful e.g. if you want to associate a path with every single sample when
+# debugging!
+#
+# .. _passthrough_heuristic:
+#
+# .. note::
+#
+#     **Disclaimer** This note is slightly advanced and can be safely skipped on
+#     a first read.
+#
+#     Pure :class:`torch.Tensor` objects are, in general, treated as images (or
+#     as videos for video-specific transforms). Indeed, you may have noticed
+#     that in the code above we haven't used the
+#     :class:`~torchvision.tv_tensors.Image` class at all, and yet our images
+#     got transformed properly. Transforms follow the following logic to
+#     determine whether a pure Tensor should be treated as an image (or video),
+#     or just ignored:
+#
+#     * If there is an :class:`~torchvision.tv_tensors.Image`,
+#       :class:`~torchvision.tv_tensors.Video`,
+#       or :class:`PIL.Image.Image` instance in the input, all other pure
+#       tensors are passed-through.
+#     * If there is no :class:`~torchvision.tv_tensors.Image` or
+#       :class:`~torchvision.tv_tensors.Video` instance, only the first pure
+#       :class:`torch.Tensor` will be transformed as image or video, while all
+#       others will be passed-through. Here "first" means "first in a depth-wise
+#       traversal".
+#
+#     This is what happened in the detection example above: the first pure
+#     tensor was the image so it got transformed properly, and all other pure
+#     tensor instances like the ``labels`` were passed-through (although labels
+#     can still be transformed by some transforms like
+#     :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes`!).
+#
+# .. _transforms_datasets_intercompatibility:
+#
+# Transforms and Datasets intercompatibility
+# ------------------------------------------
+#
+# Roughly speaking, the output of the datasets must correspond to the input of
+# the transforms. How to do that depends on whether you're using the torchvision
+# :ref:`built-in datatsets <datasets>`, or your own custom datasets.
+#
+# Using built-in datasets
+# ^^^^^^^^^^^^^^^^^^^^^^^
+#
+# If you're just doing image classification, you don't need to do anything. Just
+# use ``transform`` argument of the dataset e.g. ``ImageNet(...,
+# transform=transforms)`` and you're good to go.
+#
+# Torchvision also supports datasets for object detection or segmentation like
+# :class:`torchvision.datasets.CocoDetection`. Those datasets predate
+# the existence of the :mod:`torchvision.transforms.v2` module and of the
+# TVTensors, so they don't return TVTensors out of the box.
+#
+# An easy way to force those datasets to return TVTensors and to make them
+# compatible with v2 transforms is to use the
+# :func:`torchvision.datasets.wrap_dataset_for_transforms_v2` function:
+#
+# .. code-block:: python
+#
+#    from torchvision.datasets import CocoDetection, wrap_dataset_for_transforms_v2
+#
+#    dataset = CocoDetection(..., transforms=my_transforms)
+#    dataset = wrap_dataset_for_transforms_v2(dataset)
+#    # Now the dataset returns TVTensors!
+#
+# Using your own datasets
+# ^^^^^^^^^^^^^^^^^^^^^^^
+#
+# If you have a custom dataset, then you'll need to convert your objects into
+# the appropriate TVTensor classes. Creating TVTensor instances is very easy,
+# refer to :ref:`tv_tensor_creation` for more details.
+#
+# There are two main places where you can implement that conversion logic:
+#
+# - At the end of the datasets's ``__getitem__`` method, before returning the
+#   sample (or by sub-classing the dataset).
+# - As the very first step of your transforms pipeline
+#
+# Either way, the logic will depend on your specific dataset.
diff --git a/0.16/_downloads/92763696a071eab76a5ea62a1d9f8c5c/plot_tv_tensors.py b/0.16/_downloads/92763696a071eab76a5ea62a1d9f8c5c/plot_tv_tensors.py
new file mode 100644
index 00000000000..0cdbe9d0831
--- /dev/null
+++ b/0.16/_downloads/92763696a071eab76a5ea62a1d9f8c5c/plot_tv_tensors.py
@@ -0,0 +1,224 @@
+"""
+=============
+TVTensors FAQ
+=============
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_tv_tensors.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_tv_tensors.py>` to download the full example code.
+
+
+TVTensors are Tensor subclasses introduced together with
+``torchvision.transforms.v2``. This example showcases what these TVTensors are
+and how they behave.
+
+.. warning::
+
+    **Intended Audience** Unless you're writing your own transforms or your own TVTensors, you
+    probably do not need to read this guide. This is a fairly low-level topic
+    that most users will not need to worry about: you do not need to understand
+    the internals of TVTensors to efficiently rely on
+    ``torchvision.transforms.v2``. It may however be useful for advanced users
+    trying to implement their own datasets, transforms, or work directly with
+    the TVTensors.
+"""
+
+# %%
+import PIL.Image
+
+import torch
+from torchvision import tv_tensors
+
+
+# %%
+# What are TVTensors?
+# -------------------
+#
+# TVTensors are zero-copy tensor subclasses:
+
+tensor = torch.rand(3, 256, 256)
+image = tv_tensors.Image(tensor)
+
+assert isinstance(image, torch.Tensor)
+assert image.data_ptr() == tensor.data_ptr()
+
+# %%
+# Under the hood, they are needed in :mod:`torchvision.transforms.v2` to correctly dispatch to the appropriate function
+# for the input data.
+#
+# :mod:`torchvision.tv_tensors` supports four types of TVTensors:
+#
+# * :class:`~torchvision.tv_tensors.Image`
+# * :class:`~torchvision.tv_tensors.Video`
+# * :class:`~torchvision.tv_tensors.BoundingBoxes`
+# * :class:`~torchvision.tv_tensors.Mask`
+#
+# What can I do with a TVTensor?
+# ------------------------------
+#
+# TVTensors look and feel just like regular tensors - they **are** tensors.
+# Everything that is supported on a plain :class:`torch.Tensor` like ``.sum()`` or
+# any ``torch.*`` operator will also work on TVTensors. See
+# :ref:`tv_tensor_unwrapping_behaviour` for a few gotchas.
+
+# %%
+# .. _tv_tensor_creation:
+#
+# How do I construct a TVTensor?
+# ------------------------------
+#
+# Using the constructor
+# ^^^^^^^^^^^^^^^^^^^^^
+#
+# Each TVTensor class takes any tensor-like data that can be turned into a :class:`~torch.Tensor`
+
+image = tv_tensors.Image([[[[0, 1], [1, 0]]]])
+print(image)
+
+
+# %%
+# Similar to other PyTorch creations ops, the constructor also takes the ``dtype``, ``device``, and ``requires_grad``
+# parameters.
+
+float_image = tv_tensors.Image([[[0, 1], [1, 0]]], dtype=torch.float32, requires_grad=True)
+print(float_image)
+
+
+# %%
+# In addition, :class:`~torchvision.tv_tensors.Image` and :class:`~torchvision.tv_tensors.Mask` can also take a
+# :class:`PIL.Image.Image` directly:
+
+image = tv_tensors.Image(PIL.Image.open("../assets/astronaut.jpg"))
+print(image.shape, image.dtype)
+
+# %%
+# Some TVTensors require additional metadata to be passed in ordered to be constructed. For example,
+# :class:`~torchvision.tv_tensors.BoundingBoxes` requires the coordinate format as well as the size of the
+# corresponding image (``canvas_size``) alongside the actual values. These
+# metadata are required to properly transform the bounding boxes.
+
+bboxes = tv_tensors.BoundingBoxes(
+    [[17, 16, 344, 495], [0, 10, 0, 10]],
+    format=tv_tensors.BoundingBoxFormat.XYXY,
+    canvas_size=image.shape[-2:]
+)
+print(bboxes)
+
+# %%
+# Using ``tv_tensors.wrap()``
+# ^^^^^^^^^^^^^^^^^^^^^^^^^^^
+#
+# You can also use the :func:`~torchvision.tv_tensors.wrap` function to wrap a tensor object
+# into a TVTensor. This is useful when you already have an object of the
+# desired type, which typically happens when writing transforms: you just want
+# to wrap the output like the input.
+
+new_bboxes = torch.tensor([0, 20, 30, 40])
+new_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)
+assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+assert new_bboxes.canvas_size == bboxes.canvas_size
+
+# %%
+# The metadata of ``new_bboxes`` is the same as ``bboxes``, but you could pass
+# it as a parameter to override it.
+#
+# .. _tv_tensor_unwrapping_behaviour:
+#
+# I had a TVTensor but now I have a Tensor. Help!
+# -----------------------------------------------
+#
+# By default, operations on :class:`~torchvision.tv_tensors.TVTensor` objects
+# will return a pure Tensor:
+
+
+assert isinstance(bboxes, tv_tensors.BoundingBoxes)
+
+# Shift bboxes by 3 pixels in both H and W
+new_bboxes = bboxes + 3
+
+assert isinstance(new_bboxes, torch.Tensor)
+assert not isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+
+# %%
+# .. note::
+#
+#    This behavior only affects native ``torch`` operations. If you are using
+#    the built-in ``torchvision`` transforms or functionals, you will always get
+#    as output the same type that you passed as input (pure ``Tensor`` or
+#    ``TVTensor``).
+
+# %%
+# But I want a TVTensor back!
+# ^^^^^^^^^^^^^^^^^^^^^^^^^^^
+#
+# You can re-wrap a pure tensor into a TVTensor by just calling the TVTensor
+# constructor, or by using the :func:`~torchvision.tv_tensors.wrap` function
+# (see more details above in :ref:`tv_tensor_creation`):
+
+new_bboxes = bboxes + 3
+new_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)
+assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+
+# %%
+# Alternatively, you can use the :func:`~torchvision.tv_tensors.set_return_type`
+# as a global config setting for the whole program, or as a context manager
+# (read its docs to learn more about caveats):
+
+with tv_tensors.set_return_type("TVTensor"):
+    new_bboxes = bboxes + 3
+assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+
+# %%
+# Why is this happening?
+# ^^^^^^^^^^^^^^^^^^^^^^
+#
+# **For performance reasons**. :class:`~torchvision.tv_tensors.TVTensor`
+# classes are Tensor subclasses, so any operation involving a
+# :class:`~torchvision.tv_tensors.TVTensor` object will go through the
+# `__torch_function__
+# <https://pytorch.org/docs/stable/notes/extending.html#extending-torch>`_
+# protocol. This induces a small overhead, which we want to avoid when possible.
+# This doesn't matter for built-in ``torchvision`` transforms because we can
+# avoid the overhead there, but it could be a problem in your model's
+# ``forward``.
+#
+# **The alternative isn't much better anyway.** For every operation where
+# preserving the :class:`~torchvision.tv_tensors.TVTensor` type makes
+# sense, there are just as many operations where returning a pure Tensor is
+# preferable: for example, is ``img.sum()`` still an :class:`~torchvision.tv_tensors.Image`?
+# If we were to preserve :class:`~torchvision.tv_tensors.TVTensor` types all
+# the way, even model's logits or the output of the loss function would end up
+# being of type :class:`~torchvision.tv_tensors.Image`, and surely that's not
+# desirable.
+#
+# .. note::
+#
+#    This behaviour is something we're actively seeking feedback on. If you find this surprising or if you
+#    have any suggestions on how to better support your use-cases, please reach out to us via this issue:
+#    https://github.com/pytorch/vision/issues/7319
+#
+# Exceptions
+# ^^^^^^^^^^
+#
+# There are a few exceptions to this "unwrapping" rule:
+# :meth:`~torch.Tensor.clone`, :meth:`~torch.Tensor.to`,
+# :meth:`torch.Tensor.detach`, and :meth:`~torch.Tensor.requires_grad_` retain
+# the TVTensor type.
+#
+# Inplace operations on TVTensors like ``obj.add_()`` will preserve the type of
+# ``obj``. However, the **returned** value of inplace operations will be a pure
+# tensor:
+
+image = tv_tensors.Image([[[0, 1], [1, 0]]])
+
+new_image = image.add_(1).mul_(2)
+
+# image got transformed in-place and is still a TVTensor Image, but new_image
+# is a Tensor. They share the same underlying data and they're equal, just
+# different classes.
+assert isinstance(image, tv_tensors.Image)
+print(image)
+
+assert isinstance(new_image, torch.Tensor) and not isinstance(new_image, tv_tensors.Image)
+assert (new_image == image).all()
+assert new_image.data_ptr() == image.data_ptr()
diff --git a/0.16/_downloads/95e7320166df5d4ddcbdd5ea64a5c98b/plot_visualization_utils.ipynb b/0.16/_downloads/95e7320166df5d4ddcbdd5ea64a5c98b/plot_visualization_utils.ipynb
new file mode 100644
index 00000000000..f8d08a978a0
--- /dev/null
+++ b/0.16/_downloads/95e7320166df5d4ddcbdd5ea64a5c98b/plot_visualization_utils.ipynb
@@ -0,0 +1,464 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Visualization utilities\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_visualization_utils.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_visualization_utils.py>` to download the full example code.</p></div>\n\nThis example illustrates some of the utilities that torchvision offers for\nvisualizing images, bounding boxes, segmentation masks and keypoints.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nimport torchvision.transforms.functional as F\n\n\nplt.rcParams[\"savefig.bbox\"] = 'tight'\n\n\ndef show(imgs):\n    if not isinstance(imgs, list):\n        imgs = [imgs]\n    fig, axs = plt.subplots(ncols=len(imgs), squeeze=False)\n    for i, img in enumerate(imgs):\n        img = img.detach()\n        img = F.to_pil_image(img)\n        axs[0, i].imshow(np.asarray(img))\n        axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing a grid of images\nThe :func:`~torchvision.utils.make_grid` function can be used to create a\ntensor that represents multiple images in a grid.  This util requires a single\nimage of dtype ``uint8`` as input.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import make_grid\nfrom torchvision.io import read_image\nfrom pathlib import Path\n\ndog1_int = read_image(str(Path('../assets') / 'dog1.jpg'))\ndog2_int = read_image(str(Path('../assets') / 'dog2.jpg'))\ndog_list = [dog1_int, dog2_int]\n\ngrid = make_grid(dog_list)\nshow(grid)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing bounding boxes\nWe can use :func:`~torchvision.utils.draw_bounding_boxes` to draw boxes on an\nimage. We can set the colors, labels, width as well as font and font size.\nThe boxes are in ``(xmin, ymin, xmax, ymax)`` format.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_bounding_boxes\n\n\nboxes = torch.tensor([[50, 50, 100, 200], [210, 150, 350, 430]], dtype=torch.float)\ncolors = [\"blue\", \"yellow\"]\nresult = draw_bounding_boxes(dog1_int, boxes, colors=colors, width=5)\nshow(result)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Naturally, we can also plot bounding boxes produced by torchvision detection\nmodels.  Here is a demo with a Faster R-CNN model loaded from\n:func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`\nmodel. For more details on the output of such models, you may\nrefer to `instance_seg_output`.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights\n\n\nweights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT\ntransforms = weights.transforms()\n\nimages = [transforms(d) for d in dog_list]\n\nmodel = fasterrcnn_resnet50_fpn(weights=weights, progress=False)\nmodel = model.eval()\n\noutputs = model(images)\nprint(outputs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let's plot the boxes detected by our model. We will only plot the boxes with a\nscore greater than a given threshold.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "score_threshold = .8\ndogs_with_boxes = [\n    draw_bounding_boxes(dog_int, boxes=output['boxes'][output['scores'] > score_threshold], width=4)\n    for dog_int, output in zip(dog_list, outputs)\n]\nshow(dogs_with_boxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing segmentation masks\nThe :func:`~torchvision.utils.draw_segmentation_masks` function can be used to\ndraw segmentation masks on images. Semantic segmentation and instance\nsegmentation models have different outputs, so we will treat each\nindependently.\n\n\n### Semantic segmentation models\n\nWe will see how to use it with torchvision's FCN Resnet-50, loaded with\n:func:`~torchvision.models.segmentation.fcn_resnet50`. Let's start by looking\nat the output of the model.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.segmentation import fcn_resnet50, FCN_ResNet50_Weights\n\nweights = FCN_ResNet50_Weights.DEFAULT\ntransforms = weights.transforms(resize_size=None)\n\nmodel = fcn_resnet50(weights=weights, progress=False)\nmodel = model.eval()\n\nbatch = torch.stack([transforms(d) for d in dog_list])\noutput = model(batch)['out']\nprint(output.shape, output.min().item(), output.max().item())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As we can see above, the output of the segmentation model is a tensor of shape\n``(batch_size, num_classes, H, W)``. Each value is a non-normalized score, and\nwe can normalize them into ``[0, 1]`` by using a softmax. After the softmax,\nwe can interpret each value as a probability indicating how likely a given\npixel is to belong to a given class.\n\nLet's plot the masks that have been detected for the dog class and for the\nboat class:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "sem_class_to_idx = {cls: idx for (idx, cls) in enumerate(weights.meta[\"categories\"])}\n\nnormalized_masks = torch.nn.functional.softmax(output, dim=1)\n\ndog_and_boat_masks = [\n    normalized_masks[img_idx, sem_class_to_idx[cls]]\n    for img_idx in range(len(dog_list))\n    for cls in ('dog', 'boat')\n]\n\nshow(dog_and_boat_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As expected, the model is confident about the dog class, but not so much for\nthe boat class.\n\nThe :func:`~torchvision.utils.draw_segmentation_masks` function can be used to\nplots those masks on top of the original image. This function expects the\nmasks to be boolean masks, but our masks above contain probabilities in ``[0,\n1]``. To get boolean masks, we can do the following:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class_dim = 1\nboolean_dog_masks = (normalized_masks.argmax(class_dim) == sem_class_to_idx['dog'])\nprint(f\"shape = {boolean_dog_masks.shape}, dtype = {boolean_dog_masks.dtype}\")\nshow([m.float() for m in boolean_dog_masks])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The line above where we define ``boolean_dog_masks`` is a bit cryptic, but you\ncan read it as the following query: \"For which pixels is 'dog' the most likely\nclass?\"\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>While we're using the ``normalized_masks`` here, we would have\n  gotten the same result by using the non-normalized scores of the model\n  directly (as the softmax operation preserves the order).</p></div>\n\nNow that we have boolean masks, we can use them with\n:func:`~torchvision.utils.draw_segmentation_masks` to plot them on top of the\noriginal images:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_segmentation_masks\n\ndogs_with_masks = [\n    draw_segmentation_masks(img, masks=mask, alpha=0.7)\n    for img, mask in zip(dog_list, boolean_dog_masks)\n]\nshow(dogs_with_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can plot more than one mask per image! Remember that the model returned as\nmany masks as there are classes. Let's ask the same query as above, but this\ntime for *all* classes, not just the dog class: \"For each pixel and each class\nC, is class C the most likely class?\"\n\nThis one is a bit more involved, so we'll first show how to do it with a\nsingle image, and then we'll generalize to the batch\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "num_classes = normalized_masks.shape[1]\ndog1_masks = normalized_masks[0]\nclass_dim = 0\ndog1_all_classes_masks = dog1_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None]\n\nprint(f\"dog1_masks shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}\")\nprint(f\"dog1_all_classes_masks = {dog1_all_classes_masks.shape}, dtype = {dog1_all_classes_masks.dtype}\")\n\ndog_with_all_masks = draw_segmentation_masks(dog1_int, masks=dog1_all_classes_masks, alpha=.6)\nshow(dog_with_all_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can see in the image above that only 2 masks were drawn: the mask for the\nbackground and the mask for the dog. This is because the model thinks that\nonly these 2 classes are the most likely ones across all the pixels. If the\nmodel had detected another class as the most likely among other pixels, we\nwould have seen its mask above.\n\nRemoving the background mask is as simple as passing\n``masks=dog1_all_classes_masks[1:]``, because the background class is the\nclass with index 0.\n\nLet's now do the same but for an entire batch of images. The code is similar\nbut involves a bit more juggling with the dimensions.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class_dim = 1\nall_classes_masks = normalized_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None, None]\nprint(f\"shape = {all_classes_masks.shape}, dtype = {all_classes_masks.dtype}\")\n# The first dimension is the classes now, so we need to swap it\nall_classes_masks = all_classes_masks.swapaxes(0, 1)\n\ndogs_with_masks = [\n    draw_segmentation_masks(img, masks=mask, alpha=.6)\n    for img, mask in zip(dog_list, all_classes_masks)\n]\nshow(dogs_with_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n### Instance segmentation models\n\nInstance segmentation models have a significantly different output from the\nsemantic segmentation models. We will see here how to plot the masks for such\nmodels. Let's start by analyzing the output of a Mask-RCNN model. Note that\nthese models don't require the images to be normalized, so we don't need to\nuse the normalized batch.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>We will here describe the output of a Mask-RCNN model. The models in\n    `object_det_inst_seg_pers_keypoint_det` all have a similar output\n    format, but some of them may have extra info like keypoints for\n    :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`, and some\n    of them may not have masks, like\n    :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`.</p></div>\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import maskrcnn_resnet50_fpn, MaskRCNN_ResNet50_FPN_Weights\n\nweights = MaskRCNN_ResNet50_FPN_Weights.DEFAULT\ntransforms = weights.transforms()\n\nimages = [transforms(d) for d in dog_list]\n\nmodel = maskrcnn_resnet50_fpn(weights=weights, progress=False)\nmodel = model.eval()\n\noutput = model(images)\nprint(output)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let's break this down. For each image in the batch, the model outputs some\ndetections (or instances). The number of detections varies for each input\nimage. Each instance is described by its bounding box, its label, its score\nand its mask.\n\nThe way the output is organized is as follows: the output is a list of length\n``batch_size``. Each entry in the list corresponds to an input image, and it\nis a dict with keys 'boxes', 'labels', 'scores', and 'masks'. Each value\nassociated to those keys has ``num_instances`` elements in it.  In our case\nabove there are 3 instances detected in the first image, and 2 instances in\nthe second one.\n\nThe boxes can be plotted with :func:`~torchvision.utils.draw_bounding_boxes`\nas above, but here we're more interested in the masks. These masks are quite\ndifferent from the masks that we saw above for the semantic segmentation\nmodels.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dog1_output = output[0]\ndog1_masks = dog1_output['masks']\nprint(f\"shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}, \"\n      f\"min = {dog1_masks.min()}, max = {dog1_masks.max()}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Here the masks correspond to probabilities indicating, for each pixel, how\nlikely it is to belong to the predicted label of that instance. Those\npredicted labels correspond to the 'labels' element in the same output dict.\nLet's see which labels were predicted for the instances of the first image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(\"For the first dog, the following instances were detected:\")\nprint([weights.meta[\"categories\"][label] for label in dog1_output['labels']])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Interestingly, the model detects two persons in the image. Let's go ahead and\nplot those masks. Since :func:`~torchvision.utils.draw_segmentation_masks`\nexpects boolean masks, we need to convert those probabilities into boolean\nvalues. Remember that the semantic of those masks is \"How likely is this pixel\nto belong to the predicted class?\". As a result, a natural way of converting\nthose masks into boolean values is to threshold them with the 0.5 probability\n(one could also choose a different threshold).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "proba_threshold = 0.5\ndog1_bool_masks = dog1_output['masks'] > proba_threshold\nprint(f\"shape = {dog1_bool_masks.shape}, dtype = {dog1_bool_masks.dtype}\")\n\n# There's an extra dimension (1) to the masks. We need to remove it\ndog1_bool_masks = dog1_bool_masks.squeeze(1)\n\nshow(draw_segmentation_masks(dog1_int, dog1_bool_masks, alpha=0.9))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The model seems to have properly detected the dog, but it also confused trees\nwith people. Looking more closely at the scores will help us plot more\nrelevant masks:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(dog1_output['scores'])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Clearly the model is more confident about the dog detection than it is about\nthe people detections. That's good news. When plotting the masks, we can ask\nfor only those that have a good score. Let's use a score threshold of .75\nhere, and also plot the masks of the second dog.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "score_threshold = .75\n\nboolean_masks = [\n    out['masks'][out['scores'] > score_threshold] > proba_threshold\n    for out in output\n]\n\ndogs_with_masks = [\n    draw_segmentation_masks(img, mask.squeeze(1))\n    for img, mask in zip(dog_list, boolean_masks)\n]\nshow(dogs_with_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The two 'people' masks in the first image where not selected because they have\na lower score than the score threshold. Similarly, in the second image, the\ninstance with class 15 (which corresponds to 'bench') was not selected.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n## Visualizing keypoints\nThe :func:`~torchvision.utils.draw_keypoints` function can be used to\ndraw keypoints on images. We will see how to use it with\ntorchvision's KeypointRCNN loaded with :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`.\nWe will first have a look at output of the model.\n\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import keypointrcnn_resnet50_fpn, KeypointRCNN_ResNet50_FPN_Weights\nfrom torchvision.io import read_image\n\nperson_int = read_image(str(Path(\"../assets\") / \"person1.jpg\"))\n\nweights = KeypointRCNN_ResNet50_FPN_Weights.DEFAULT\ntransforms = weights.transforms()\n\nperson_float = transforms(person_int)\n\nmodel = keypointrcnn_resnet50_fpn(weights=weights, progress=False)\nmodel = model.eval()\n\noutputs = model([person_float])\nprint(outputs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As we see the output contains a list of dictionaries.\nThe output list is of length batch_size.\nWe currently have just a single image so length of list is 1.\nEach entry in the list corresponds to an input image,\nand it is a dict with keys `boxes`, `labels`, `scores`, `keypoints` and `keypoint_scores`.\nEach value associated to those keys has `num_instances` elements in it.\nIn our case above there are 2 instances detected in the image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "kpts = outputs[0]['keypoints']\nscores = outputs[0]['scores']\n\nprint(kpts)\nprint(scores)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The KeypointRCNN model detects there are two instances in the image.\nIf you plot the boxes by using :func:`~draw_bounding_boxes`\nyou would recognize they are the person and the surfboard.\nIf we look at the scores, we will realize that the model is much more confident about the person than surfboard.\nWe could now set a threshold confidence and plot instances which we are confident enough.\nLet us set a threshold of 0.75 and filter out the keypoints corresponding to the person.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "detect_threshold = 0.75\nidx = torch.where(scores > detect_threshold)\nkeypoints = kpts[idx]\n\nprint(keypoints)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Great, now we have the keypoints corresponding to the person.\nEach keypoint is represented by x, y coordinates and the visibility.\nWe can now use the :func:`~torchvision.utils.draw_keypoints` function to draw keypoints.\nNote that the utility expects uint8 images.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_keypoints\n\nres = draw_keypoints(person_int, keypoints, colors=\"blue\", radius=3)\nshow(res)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As we see the keypoints appear as colored circles over the image.\nThe coco keypoints for a person are ordered and represent the following list.\\\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "coco_keypoints = [\n    \"nose\", \"left_eye\", \"right_eye\", \"left_ear\", \"right_ear\",\n    \"left_shoulder\", \"right_shoulder\", \"left_elbow\", \"right_elbow\",\n    \"left_wrist\", \"right_wrist\", \"left_hip\", \"right_hip\",\n    \"left_knee\", \"right_knee\", \"left_ankle\", \"right_ankle\",\n]"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "What if we are interested in joining the keypoints?\nThis is especially useful in creating pose detection or action recognition.\nWe can join the keypoints easily using the `connectivity` parameter.\nA close observation would reveal that we would need to join the points in below\norder to construct human skeleton.\n\nnose -> left_eye -> left_ear.                              (0, 1), (1, 3)\n\nnose -> right_eye -> right_ear.                            (0, 2), (2, 4)\n\nnose -> left_shoulder -> left_elbow -> left_wrist.         (0, 5), (5, 7), (7, 9)\n\nnose -> right_shoulder -> right_elbow -> right_wrist.      (0, 6), (6, 8), (8, 10)\n\nleft_shoulder -> left_hip -> left_knee -> left_ankle.      (5, 11), (11, 13), (13, 15)\n\nright_shoulder -> right_hip -> right_knee -> right_ankle.  (6, 12), (12, 14), (14, 16)\n\nWe will create a list containing these keypoint ids to be connected.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "connect_skeleton = [\n    (0, 1), (0, 2), (1, 3), (2, 4), (0, 5), (0, 6), (5, 7), (6, 8),\n    (7, 9), (8, 10), (5, 11), (6, 12), (11, 13), (12, 14), (13, 15), (14, 16)\n]"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We pass the above list to the connectivity parameter to connect the keypoints.\n\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "res = draw_keypoints(person_int, keypoints, connectivity=connect_skeleton, colors=\"blue\", radius=4, width=3)\nshow(res)"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/a8ef8e39164e980a89a9afdfc51d961e/plot_transforms_illustrations.ipynb b/0.16/_downloads/a8ef8e39164e980a89a9afdfc51d961e/plot_transforms_illustrations.ipynb
new file mode 100644
index 00000000000..b885564fff7
--- /dev/null
+++ b/0.16/_downloads/a8ef8e39164e980a89a9afdfc51d961e/plot_transforms_illustrations.ipynb
@@ -0,0 +1,511 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Illustration of transforms\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_illustrations.py>` to download the full example code.</p></div>\n\nThis example illustrates some of the various transforms available in `the\ntorchvision.transforms.v2 module <transforms>`.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from PIL import Image\nfrom pathlib import Path\nimport matplotlib.pyplot as plt\n\nimport torch\nfrom torchvision.transforms import v2\n\nplt.rcParams[\"savefig.bbox\"] = 'tight'\n\n# if you change the seed, make sure that the randomly-applied transforms\n# properly show that the image can be both transformed and *not* transformed!\ntorch.manual_seed(0)\n\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nfrom helpers import plot\norig_img = Image.open(Path('../assets') / 'astronaut.jpg')"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Geometric Transforms\nGeometric image transformation refers to the process of altering the geometric properties of an image,\nsuch as its shape, size, orientation, or position.\nIt involves applying mathematical operations to the image pixels or coordinates to achieve the desired transformation.\n\n### Pad\nThe :class:`~torchvision.transforms.Pad` transform\n(see also :func:`~torchvision.transforms.functional.pad`)\npads all image borders with some pixel values.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "padded_imgs = [v2.Pad(padding=padding)(orig_img) for padding in (3, 10, 30, 50)]\nplot([orig_img] + padded_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### Resize\nThe :class:`~torchvision.transforms.Resize` transform\n(see also :func:`~torchvision.transforms.functional.resize`)\nresizes an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "resized_imgs = [v2.Resize(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]\nplot([orig_img] + resized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### CenterCrop\nThe :class:`~torchvision.transforms.CenterCrop` transform\n(see also :func:`~torchvision.transforms.functional.center_crop`)\ncrops the given image at the center.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "center_crops = [v2.CenterCrop(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]\nplot([orig_img] + center_crops)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### FiveCrop\nThe :class:`~torchvision.transforms.FiveCrop` transform\n(see also :func:`~torchvision.transforms.functional.five_crop`)\ncrops the given image into four corners and the central crop.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "(top_left, top_right, bottom_left, bottom_right, center) = v2.FiveCrop(size=(100, 100))(orig_img)\nplot([orig_img] + [top_left, top_right, bottom_left, bottom_right, center])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomPerspective\nThe :class:`~torchvision.transforms.RandomPerspective` transform\n(see also :func:`~torchvision.transforms.functional.perspective`)\nperforms random perspective transform on an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "perspective_transformer = v2.RandomPerspective(distortion_scale=0.6, p=1.0)\nperspective_imgs = [perspective_transformer(orig_img) for _ in range(4)]\nplot([orig_img] + perspective_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomRotation\nThe :class:`~torchvision.transforms.RandomRotation` transform\n(see also :func:`~torchvision.transforms.functional.rotate`)\nrotates an image with random angle.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "rotater = v2.RandomRotation(degrees=(0, 180))\nrotated_imgs = [rotater(orig_img) for _ in range(4)]\nplot([orig_img] + rotated_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomAffine\nThe :class:`~torchvision.transforms.RandomAffine` transform\n(see also :func:`~torchvision.transforms.functional.affine`)\nperforms random affine transform on an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "affine_transfomer = v2.RandomAffine(degrees=(30, 70), translate=(0.1, 0.3), scale=(0.5, 0.75))\naffine_imgs = [affine_transfomer(orig_img) for _ in range(4)]\nplot([orig_img] + affine_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### ElasticTransform\nThe :class:`~torchvision.transforms.ElasticTransform` transform\n(see also :func:`~torchvision.transforms.functional.elastic_transform`)\nRandomly transforms the morphology of objects in images and produces a\nsee-through-water-like effect.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "elastic_transformer = v2.ElasticTransform(alpha=250.0)\ntransformed_imgs = [elastic_transformer(orig_img) for _ in range(2)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomCrop\nThe :class:`~torchvision.transforms.RandomCrop` transform\n(see also :func:`~torchvision.transforms.functional.crop`)\ncrops an image at a random location.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "cropper = v2.RandomCrop(size=(128, 128))\ncrops = [cropper(orig_img) for _ in range(4)]\nplot([orig_img] + crops)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomResizedCrop\nThe :class:`~torchvision.transforms.RandomResizedCrop` transform\n(see also :func:`~torchvision.transforms.functional.resized_crop`)\ncrops an image at a random location, and then resizes the crop to a given\nsize.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "resize_cropper = v2.RandomResizedCrop(size=(32, 32))\nresized_crops = [resize_cropper(orig_img) for _ in range(4)]\nplot([orig_img] + resized_crops)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Photometric Transforms\nPhotometric image transformation refers to the process of modifying the photometric properties of an image,\nsuch as its brightness, contrast, color, or tone.\nThese transformations are applied to change the visual appearance of an image\nwhile preserving its geometric structure.\n\nExcept :class:`~torchvision.transforms.Grayscale`, the following transforms are random,\nwhich means that the same transform\ninstance will produce different result each time it transforms a given image.\n\n### Grayscale\nThe :class:`~torchvision.transforms.Grayscale` transform\n(see also :func:`~torchvision.transforms.functional.to_grayscale`)\nconverts an image to grayscale\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "gray_img = v2.Grayscale()(orig_img)\nplot([orig_img, gray_img], cmap='gray')"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### ColorJitter\nThe :class:`~torchvision.transforms.ColorJitter` transform\nrandomly changes the brightness, contrast, saturation, hue, and other properties of an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "jitter = v2.ColorJitter(brightness=.5, hue=.3)\njittered_imgs = [jitter(orig_img) for _ in range(4)]\nplot([orig_img] + jittered_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### GaussianBlur\nThe :class:`~torchvision.transforms.GaussianBlur` transform\n(see also :func:`~torchvision.transforms.functional.gaussian_blur`)\nperforms gaussian blur transform on an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "blurrer = v2.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5.))\nblurred_imgs = [blurrer(orig_img) for _ in range(4)]\nplot([orig_img] + blurred_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomInvert\nThe :class:`~torchvision.transforms.RandomInvert` transform\n(see also :func:`~torchvision.transforms.functional.invert`)\nrandomly inverts the colors of the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "inverter = v2.RandomInvert()\ninvertered_imgs = [inverter(orig_img) for _ in range(4)]\nplot([orig_img] + invertered_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomPosterize\nThe :class:`~torchvision.transforms.RandomPosterize` transform\n(see also :func:`~torchvision.transforms.functional.posterize`)\nrandomly posterizes the image by reducing the number of bits\nof each color channel.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "posterizer = v2.RandomPosterize(bits=2)\nposterized_imgs = [posterizer(orig_img) for _ in range(4)]\nplot([orig_img] + posterized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomSolarize\nThe :class:`~torchvision.transforms.RandomSolarize` transform\n(see also :func:`~torchvision.transforms.functional.solarize`)\nrandomly solarizes the image by inverting all pixel values above\nthe threshold.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "solarizer = v2.RandomSolarize(threshold=192.0)\nsolarized_imgs = [solarizer(orig_img) for _ in range(4)]\nplot([orig_img] + solarized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomAdjustSharpness\nThe :class:`~torchvision.transforms.RandomAdjustSharpness` transform\n(see also :func:`~torchvision.transforms.functional.adjust_sharpness`)\nrandomly adjusts the sharpness of the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "sharpness_adjuster = v2.RandomAdjustSharpness(sharpness_factor=2)\nsharpened_imgs = [sharpness_adjuster(orig_img) for _ in range(4)]\nplot([orig_img] + sharpened_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomAutocontrast\nThe :class:`~torchvision.transforms.RandomAutocontrast` transform\n(see also :func:`~torchvision.transforms.functional.autocontrast`)\nrandomly applies autocontrast to the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "autocontraster = v2.RandomAutocontrast()\nautocontrasted_imgs = [autocontraster(orig_img) for _ in range(4)]\nplot([orig_img] + autocontrasted_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomEqualize\nThe :class:`~torchvision.transforms.RandomEqualize` transform\n(see also :func:`~torchvision.transforms.functional.equalize`)\nrandomly equalizes the histogram of the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "equalizer = v2.RandomEqualize()\nequalized_imgs = [equalizer(orig_img) for _ in range(4)]\nplot([orig_img] + equalized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Augmentation Transforms\nThe following transforms are combinations of multiple transforms,\neither geometric or photometric, or both.\n\n### AutoAugment\nThe :class:`~torchvision.transforms.AutoAugment` transform\nautomatically augments data based on a given auto-augmentation policy.\nSee :class:`~torchvision.transforms.AutoAugmentPolicy` for the available policies.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "policies = [v2.AutoAugmentPolicy.CIFAR10, v2.AutoAugmentPolicy.IMAGENET, v2.AutoAugmentPolicy.SVHN]\naugmenters = [v2.AutoAugment(policy) for policy in policies]\nimgs = [\n    [augmenter(orig_img) for _ in range(4)]\n    for augmenter in augmenters\n]\nrow_title = [str(policy).split('.')[-1] for policy in policies]\nplot([[orig_img] + row for row in imgs], row_title=row_title)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandAugment\nThe :class:`~torchvision.transforms.RandAugment` is an alternate version of AutoAugment.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "augmenter = v2.RandAugment()\nimgs = [augmenter(orig_img) for _ in range(4)]\nplot([orig_img] + imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### TrivialAugmentWide\nThe :class:`~torchvision.transforms.TrivialAugmentWide` is an alternate implementation of AutoAugment.\nHowever, instead of transforming an image multiple times, it transforms an image only once\nusing a random transform from a given list with a random strength number.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "augmenter = v2.TrivialAugmentWide()\nimgs = [augmenter(orig_img) for _ in range(4)]\nplot([orig_img] + imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### AugMix\nThe :class:`~torchvision.transforms.AugMix` transform interpolates between augmented versions of an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "augmenter = v2.AugMix()\nimgs = [augmenter(orig_img) for _ in range(4)]\nplot([orig_img] + imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Randomly-applied Transforms\n\nThe following transforms are randomly-applied given a probability ``p``.  That is, given ``p = 0.5``,\nthere is a 50% chance to return the original image, and a 50% chance to return the transformed image,\neven when called with the same transform instance!\n\n### RandomHorizontalFlip\nThe :class:`~torchvision.transforms.RandomHorizontalFlip` transform\n(see also :func:`~torchvision.transforms.functional.hflip`)\nperforms horizontal flip of an image, with a given probability.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "hflipper = v2.RandomHorizontalFlip(p=0.5)\ntransformed_imgs = [hflipper(orig_img) for _ in range(4)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomVerticalFlip\nThe :class:`~torchvision.transforms.RandomVerticalFlip` transform\n(see also :func:`~torchvision.transforms.functional.vflip`)\nperforms vertical flip of an image, with a given probability.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "vflipper = v2.RandomVerticalFlip(p=0.5)\ntransformed_imgs = [vflipper(orig_img) for _ in range(4)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomApply\nThe :class:`~torchvision.transforms.RandomApply` transform\nrandomly applies a list of transforms, with a given probability.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "applier = v2.RandomApply(transforms=[v2.RandomCrop(size=(64, 64))], p=0.5)\ntransformed_imgs = [applier(orig_img) for _ in range(4)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/bf35a573bbfe2773c19cd458bf8508c4/plot_tv_tensors.ipynb b/0.16/_downloads/bf35a573bbfe2773c19cd458bf8508c4/plot_tv_tensors.ipynb
new file mode 100644
index 00000000000..8a6bcd96b54
--- /dev/null
+++ b/0.16/_downloads/bf35a573bbfe2773c19cd458bf8508c4/plot_tv_tensors.ipynb
@@ -0,0 +1,237 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# TVTensors FAQ\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_tv_tensors.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_tv_tensors.py>` to download the full example code.</p></div>\n\n\nTVTensors are Tensor subclasses introduced together with\n``torchvision.transforms.v2``. This example showcases what these TVTensors are\nand how they behave.\n\n<div class=\"alert alert-danger\"><h4>Warning</h4><p>**Intended Audience** Unless you're writing your own transforms or your own TVTensors, you\n    probably do not need to read this guide. This is a fairly low-level topic\n    that most users will not need to worry about: you do not need to understand\n    the internals of TVTensors to efficiently rely on\n    ``torchvision.transforms.v2``. It may however be useful for advanced users\n    trying to implement their own datasets, transforms, or work directly with\n    the TVTensors.</p></div>\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import PIL.Image\n\nimport torch\nfrom torchvision import tv_tensors"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## What are TVTensors?\n\nTVTensors are zero-copy tensor subclasses:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "tensor = torch.rand(3, 256, 256)\nimage = tv_tensors.Image(tensor)\n\nassert isinstance(image, torch.Tensor)\nassert image.data_ptr() == tensor.data_ptr()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Under the hood, they are needed in :mod:`torchvision.transforms.v2` to correctly dispatch to the appropriate function\nfor the input data.\n\n:mod:`torchvision.tv_tensors` supports four types of TVTensors:\n\n* :class:`~torchvision.tv_tensors.Image`\n* :class:`~torchvision.tv_tensors.Video`\n* :class:`~torchvision.tv_tensors.BoundingBoxes`\n* :class:`~torchvision.tv_tensors.Mask`\n\n## What can I do with a TVTensor?\n\nTVTensors look and feel just like regular tensors - they **are** tensors.\nEverything that is supported on a plain :class:`torch.Tensor` like ``.sum()`` or\nany ``torch.*`` operator will also work on TVTensors. See\n`tv_tensor_unwrapping_behaviour` for a few gotchas.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n## How do I construct a TVTensor?\n\n### Using the constructor\n\nEach TVTensor class takes any tensor-like data that can be turned into a :class:`~torch.Tensor`\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "image = tv_tensors.Image([[[[0, 1], [1, 0]]]])\nprint(image)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Similar to other PyTorch creations ops, the constructor also takes the ``dtype``, ``device``, and ``requires_grad``\nparameters.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "float_image = tv_tensors.Image([[[0, 1], [1, 0]]], dtype=torch.float32, requires_grad=True)\nprint(float_image)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "In addition, :class:`~torchvision.tv_tensors.Image` and :class:`~torchvision.tv_tensors.Mask` can also take a\n:class:`PIL.Image.Image` directly:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "image = tv_tensors.Image(PIL.Image.open(\"../assets/astronaut.jpg\"))\nprint(image.shape, image.dtype)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Some TVTensors require additional metadata to be passed in ordered to be constructed. For example,\n:class:`~torchvision.tv_tensors.BoundingBoxes` requires the coordinate format as well as the size of the\ncorresponding image (``canvas_size``) alongside the actual values. These\nmetadata are required to properly transform the bounding boxes.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "bboxes = tv_tensors.BoundingBoxes(\n    [[17, 16, 344, 495], [0, 10, 0, 10]],\n    format=tv_tensors.BoundingBoxFormat.XYXY,\n    canvas_size=image.shape[-2:]\n)\nprint(bboxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### Using ``tv_tensors.wrap()``\n\nYou can also use the :func:`~torchvision.tv_tensors.wrap` function to wrap a tensor object\ninto a TVTensor. This is useful when you already have an object of the\ndesired type, which typically happens when writing transforms: you just want\nto wrap the output like the input.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "new_bboxes = torch.tensor([0, 20, 30, 40])\nnew_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)\nassert isinstance(new_bboxes, tv_tensors.BoundingBoxes)\nassert new_bboxes.canvas_size == bboxes.canvas_size"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The metadata of ``new_bboxes`` is the same as ``bboxes``, but you could pass\nit as a parameter to override it.\n\n\n## I had a TVTensor but now I have a Tensor. Help!\n\nBy default, operations on :class:`~torchvision.tv_tensors.TVTensor` objects\nwill return a pure Tensor:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "assert isinstance(bboxes, tv_tensors.BoundingBoxes)\n\n# Shift bboxes by 3 pixels in both H and W\nnew_bboxes = bboxes + 3\n\nassert isinstance(new_bboxes, torch.Tensor)\nassert not isinstance(new_bboxes, tv_tensors.BoundingBoxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-info\"><h4>Note</h4><p>This behavior only affects native ``torch`` operations. If you are using\n   the built-in ``torchvision`` transforms or functionals, you will always get\n   as output the same type that you passed as input (pure ``Tensor`` or\n   ``TVTensor``).</p></div>\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### But I want a TVTensor back!\n\nYou can re-wrap a pure tensor into a TVTensor by just calling the TVTensor\nconstructor, or by using the :func:`~torchvision.tv_tensors.wrap` function\n(see more details above in `tv_tensor_creation`):\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "new_bboxes = bboxes + 3\nnew_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)\nassert isinstance(new_bboxes, tv_tensors.BoundingBoxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Alternatively, you can use the :func:`~torchvision.tv_tensors.set_return_type`\nas a global config setting for the whole program, or as a context manager\n(read its docs to learn more about caveats):\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "with tv_tensors.set_return_type(\"TVTensor\"):\n    new_bboxes = bboxes + 3\nassert isinstance(new_bboxes, tv_tensors.BoundingBoxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### Why is this happening?\n\n**For performance reasons**. :class:`~torchvision.tv_tensors.TVTensor`\nclasses are Tensor subclasses, so any operation involving a\n:class:`~torchvision.tv_tensors.TVTensor` object will go through the\n[__torch_function__](https://pytorch.org/docs/stable/notes/extending.html#extending-torch)\nprotocol. This induces a small overhead, which we want to avoid when possible.\nThis doesn't matter for built-in ``torchvision`` transforms because we can\navoid the overhead there, but it could be a problem in your model's\n``forward``.\n\n**The alternative isn't much better anyway.** For every operation where\npreserving the :class:`~torchvision.tv_tensors.TVTensor` type makes\nsense, there are just as many operations where returning a pure Tensor is\npreferable: for example, is ``img.sum()`` still an :class:`~torchvision.tv_tensors.Image`?\nIf we were to preserve :class:`~torchvision.tv_tensors.TVTensor` types all\nthe way, even model's logits or the output of the loss function would end up\nbeing of type :class:`~torchvision.tv_tensors.Image`, and surely that's not\ndesirable.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>This behaviour is something we're actively seeking feedback on. If you find this surprising or if you\n   have any suggestions on how to better support your use-cases, please reach out to us via this issue:\n   https://github.com/pytorch/vision/issues/7319</p></div>\n\n### Exceptions\n\nThere are a few exceptions to this \"unwrapping\" rule:\n:meth:`~torch.Tensor.clone`, :meth:`~torch.Tensor.to`,\n:meth:`torch.Tensor.detach`, and :meth:`~torch.Tensor.requires_grad_` retain\nthe TVTensor type.\n\nInplace operations on TVTensors like ``obj.add_()`` will preserve the type of\n``obj``. However, the **returned** value of inplace operations will be a pure\ntensor:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "image = tv_tensors.Image([[[0, 1], [1, 0]]])\n\nnew_image = image.add_(1).mul_(2)\n\n# image got transformed in-place and is still a TVTensor Image, but new_image\n# is a Tensor. They share the same underlying data and they're equal, just\n# different classes.\nassert isinstance(image, tv_tensors.Image)\nprint(image)\n\nassert isinstance(new_image, torch.Tensor) and not isinstance(new_image, tv_tensors.Image)\nassert (new_image == image).all()\nassert new_image.data_ptr() == image.data_ptr()"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/d1e0b3022ec2c789e633f49bb9dd0cf4/plot_transforms_getting_started.ipynb b/0.16/_downloads/d1e0b3022ec2c789e633f49bb9dd0cf4/plot_transforms_getting_started.ipynb
new file mode 100644
index 00000000000..751326092f5
--- /dev/null
+++ b/0.16/_downloads/d1e0b3022ec2c789e633f49bb9dd0cf4/plot_transforms_getting_started.ipynb
@@ -0,0 +1,154 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Getting started with transforms v2\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_getting_started.py>` to download the full example code.</p></div>\n\nThis example illustrates all of what you need to know to get started with the\nnew :mod:`torchvision.transforms.v2` API. We'll cover simple tasks like\nimage classification, and more advanced ones like object detection /\nsegmentation.\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "First, a bit of setup\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from pathlib import Path\nimport torch\nimport matplotlib.pyplot as plt\nplt.rcParams[\"savefig.bbox\"] = 'tight'\n\nfrom torchvision.transforms import v2\nfrom torchvision.io import read_image\n\ntorch.manual_seed(1)\n\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nfrom helpers import plot\nimg = read_image(str(Path('../assets') / 'astronaut.jpg'))\nprint(f\"{type(img) = }, {img.dtype = }, {img.shape = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## The basics\n\nThe Torchvision transforms behave like a regular :class:`torch.nn.Module` (in\nfact, most of them are): instantiate a transform, pass an input, get a\ntransformed output:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transform = v2.RandomCrop(size=(224, 224))\nout = transform(img)\n\nplot([img, out])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## I just want to do image classification\n\nIf you just care about image classification, things are very simple. A basic\nclassification pipeline may look like this:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transforms = v2.Compose([\n    v2.RandomResizedCrop(size=(224, 224), antialias=True),\n    v2.RandomHorizontalFlip(p=0.5),\n    v2.ToDtype(torch.float32, scale=True),\n    v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n])\nout = transforms(img)\n\nplot([img, out])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Such transformation pipeline is typically passed as the ``transform`` argument\nto the `Datasets <datasets>`, e.g. ``ImageNet(...,\ntransform=transforms)``.\n\nThat's pretty much all there is. From there, read through our `main docs\n<transforms>` to learn more about recommended practices and conventions, or\nexplore more `examples <transforms_gallery>` e.g. how to use augmentation\ntransforms like `CutMix and MixUp\n<sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py>`.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>If you're already relying on the ``torchvision.transforms`` v1 API,\n    we recommend to `switch to the new v2 transforms<v1_or_v2>`. It's\n    very easy: the v2 transforms are fully compatible with the v1 API, so you\n    only need to change the import!</p></div>\n\n## Detection, Segmentation, Videos\n\nThe new Torchvision transforms in the ``torchvision.transforms.v2`` namespace\nsupport tasks beyond image classification: they can also transform bounding\nboxes, segmentation / detection masks, or videos.\n\nLet's briefly look at a detection example with bounding boxes.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision import tv_tensors  # we'll describe this a bit later, bare with us\n\nboxes = tv_tensors.BoundingBoxes(\n    [\n        [15, 10, 370, 510],\n        [275, 340, 510, 510],\n        [130, 345, 210, 425]\n    ],\n    format=\"XYXY\", canvas_size=img.shape[-2:])\n\ntransforms = v2.Compose([\n    v2.RandomResizedCrop(size=(224, 224), antialias=True),\n    v2.RandomPhotometricDistort(p=1),\n    v2.RandomHorizontalFlip(p=1),\n])\nout_img, out_boxes = transforms(img, boxes)\nprint(type(boxes), type(out_boxes))\n\nplot([(img, boxes), (out_img, out_boxes)])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The example above focuses on object detection. But if we had masks\n(:class:`torchvision.tv_tensors.Mask`) for object segmentation or semantic\nsegmentation, or videos (:class:`torchvision.tv_tensors.Video`), we could have\npassed them to the transforms in exactly the same way.\n\nBy now you likely have a few questions: what are these TVTensors, how do we\nuse them, and what is the expected input/output of those transforms? We'll\nanswer these in the next sections.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n## What are TVTensors?\n\nTVTensors are :class:`torch.Tensor` subclasses. The available TVTensors are\n:class:`~torchvision.tv_tensors.Image`,\n:class:`~torchvision.tv_tensors.BoundingBoxes`,\n:class:`~torchvision.tv_tensors.Mask`, and\n:class:`~torchvision.tv_tensors.Video`.\n\nTVTensors look and feel just like regular tensors - they **are** tensors.\nEverything that is supported on a plain :class:`torch.Tensor` like ``.sum()``\nor any ``torch.*`` operator will also work on a TVTensor:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "img_dp = tv_tensors.Image(torch.randint(0, 256, (3, 256, 256), dtype=torch.uint8))\n\nprint(f\"{isinstance(img_dp, torch.Tensor) = }\")\nprint(f\"{img_dp.dtype = }, {img_dp.shape = }, {img_dp.sum() = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "These TVTensor classes are at the core of the transforms: in order to\ntransform a given input, the transforms first look at the **class** of the\nobject, and dispatch to the appropriate implementation accordingly.\n\nYou don't need to know much more about TVTensors at this point, but advanced\nusers who want to learn more can refer to\n`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.\n\n## What do I pass as input?\n\nAbove, we've seen two examples: one where we passed a single image as input\ni.e. ``out = transforms(img)``, and one where we passed both an image and\nbounding boxes, i.e. ``out_img, out_boxes = transforms(img, boxes)``.\n\nIn fact, transforms support **arbitrary input structures**. The input can be a\nsingle image, a tuple, an arbitrarily nested dictionary... pretty much\nanything. The same structure will be returned as output. Below, we use the\nsame detection transforms, but pass a tuple (image, target_dict) as input and\nwe're getting the same structure as output:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "target = {\n    \"boxes\": boxes,\n    \"labels\": torch.arange(boxes.shape[0]),\n    \"this_is_ignored\": (\"arbitrary\", {\"structure\": \"!\"})\n}\n\n# Re-using the transforms and definitions from above.\nout_img, out_target = transforms(img, target)\n\nplot([(img, target[\"boxes\"]), (out_img, out_target[\"boxes\"])])\nprint(f\"{out_target['this_is_ignored']}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We passed a tuple so we get a tuple back, and the second element is the\ntranformed target dict. Transforms don't really care about the structure of\nthe input; as mentioned above, they only care about the **type** of the\nobjects and transforms them accordingly.\n\n*Foreign* objects like strings or ints are simply passed-through. This can be\nuseful e.g. if you want to associate a path with every single sample when\ndebugging!\n\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>**Disclaimer** This note is slightly advanced and can be safely skipped on\n    a first read.\n\n    Pure :class:`torch.Tensor` objects are, in general, treated as images (or\n    as videos for video-specific transforms). Indeed, you may have noticed\n    that in the code above we haven't used the\n    :class:`~torchvision.tv_tensors.Image` class at all, and yet our images\n    got transformed properly. Transforms follow the following logic to\n    determine whether a pure Tensor should be treated as an image (or video),\n    or just ignored:\n\n    * If there is an :class:`~torchvision.tv_tensors.Image`,\n      :class:`~torchvision.tv_tensors.Video`,\n      or :class:`PIL.Image.Image` instance in the input, all other pure\n      tensors are passed-through.\n    * If there is no :class:`~torchvision.tv_tensors.Image` or\n      :class:`~torchvision.tv_tensors.Video` instance, only the first pure\n      :class:`torch.Tensor` will be transformed as image or video, while all\n      others will be passed-through. Here \"first\" means \"first in a depth-wise\n      traversal\".\n\n    This is what happened in the detection example above: the first pure\n    tensor was the image so it got transformed properly, and all other pure\n    tensor instances like the ``labels`` were passed-through (although labels\n    can still be transformed by some transforms like\n    :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes`!).</p></div>\n\n\n## Transforms and Datasets intercompatibility\n\nRoughly speaking, the output of the datasets must correspond to the input of\nthe transforms. How to do that depends on whether you're using the torchvision\n`built-in datatsets <datasets>`, or your own custom datasets.\n\n### Using built-in datasets\n\nIf you're just doing image classification, you don't need to do anything. Just\nuse ``transform`` argument of the dataset e.g. ``ImageNet(...,\ntransform=transforms)`` and you're good to go.\n\nTorchvision also supports datasets for object detection or segmentation like\n:class:`torchvision.datasets.CocoDetection`. Those datasets predate\nthe existence of the :mod:`torchvision.transforms.v2` module and of the\nTVTensors, so they don't return TVTensors out of the box.\n\nAn easy way to force those datasets to return TVTensors and to make them\ncompatible with v2 transforms is to use the\n:func:`torchvision.datasets.wrap_dataset_for_transforms_v2` function:\n\n```python\nfrom torchvision.datasets import CocoDetection, wrap_dataset_for_transforms_v2\n\ndataset = CocoDetection(..., transforms=my_transforms)\ndataset = wrap_dataset_for_transforms_v2(dataset)\n# Now the dataset returns TVTensors!\n```\n### Using your own datasets\n\nIf you have a custom dataset, then you'll need to convert your objects into\nthe appropriate TVTensor classes. Creating TVTensor instances is very easy,\nrefer to `tv_tensor_creation` for more details.\n\nThere are two main places where you can implement that conversion logic:\n\n- At the end of the datasets's ``__getitem__`` method, before returning the\n  sample (or by sub-classing the dataset).\n- As the very first step of your transforms pipeline\n\nEither way, the logic will depend on your specific dataset.\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_downloads/f9ef41d32e46834ba2f93a1d3c4b4568/plot_transforms_illustrations.py b/0.16/_downloads/f9ef41d32e46834ba2f93a1d3c4b4568/plot_transforms_illustrations.py
new file mode 100644
index 00000000000..95ab455d0fd
--- /dev/null
+++ b/0.16/_downloads/f9ef41d32e46834ba2f93a1d3c4b4568/plot_transforms_illustrations.py
@@ -0,0 +1,320 @@
+"""
+==========================
+Illustration of transforms
+==========================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_illustrations.py>` to download the full example code.
+
+This example illustrates some of the various transforms available in :ref:`the
+torchvision.transforms.v2 module <transforms>`.
+"""
+# %%
+
+# sphinx_gallery_thumbnail_path = "../../gallery/assets/transforms_thumbnail.png"
+
+from PIL import Image
+from pathlib import Path
+import matplotlib.pyplot as plt
+
+import torch
+from torchvision.transforms import v2
+
+plt.rcParams["savefig.bbox"] = 'tight'
+
+# if you change the seed, make sure that the randomly-applied transforms
+# properly show that the image can be both transformed and *not* transformed!
+torch.manual_seed(0)
+
+# If you're trying to run that on collab, you can download the assets and the
+# helpers from https://github.com/pytorch/vision/tree/main/gallery/
+from helpers import plot
+orig_img = Image.open(Path('../assets') / 'astronaut.jpg')
+
+# %%
+# Geometric Transforms
+# --------------------
+# Geometric image transformation refers to the process of altering the geometric properties of an image,
+# such as its shape, size, orientation, or position.
+# It involves applying mathematical operations to the image pixels or coordinates to achieve the desired transformation.
+#
+# Pad
+# ~~~
+# The :class:`~torchvision.transforms.Pad` transform
+# (see also :func:`~torchvision.transforms.functional.pad`)
+# pads all image borders with some pixel values.
+padded_imgs = [v2.Pad(padding=padding)(orig_img) for padding in (3, 10, 30, 50)]
+plot([orig_img] + padded_imgs)
+
+# %%
+# Resize
+# ~~~~~~
+# The :class:`~torchvision.transforms.Resize` transform
+# (see also :func:`~torchvision.transforms.functional.resize`)
+# resizes an image.
+resized_imgs = [v2.Resize(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
+plot([orig_img] + resized_imgs)
+
+# %%
+# CenterCrop
+# ~~~~~~~~~~
+# The :class:`~torchvision.transforms.CenterCrop` transform
+# (see also :func:`~torchvision.transforms.functional.center_crop`)
+# crops the given image at the center.
+center_crops = [v2.CenterCrop(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
+plot([orig_img] + center_crops)
+
+# %%
+# FiveCrop
+# ~~~~~~~~
+# The :class:`~torchvision.transforms.FiveCrop` transform
+# (see also :func:`~torchvision.transforms.functional.five_crop`)
+# crops the given image into four corners and the central crop.
+(top_left, top_right, bottom_left, bottom_right, center) = v2.FiveCrop(size=(100, 100))(orig_img)
+plot([orig_img] + [top_left, top_right, bottom_left, bottom_right, center])
+
+# %%
+# RandomPerspective
+# ~~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomPerspective` transform
+# (see also :func:`~torchvision.transforms.functional.perspective`)
+# performs random perspective transform on an image.
+perspective_transformer = v2.RandomPerspective(distortion_scale=0.6, p=1.0)
+perspective_imgs = [perspective_transformer(orig_img) for _ in range(4)]
+plot([orig_img] + perspective_imgs)
+
+# %%
+# RandomRotation
+# ~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomRotation` transform
+# (see also :func:`~torchvision.transforms.functional.rotate`)
+# rotates an image with random angle.
+rotater = v2.RandomRotation(degrees=(0, 180))
+rotated_imgs = [rotater(orig_img) for _ in range(4)]
+plot([orig_img] + rotated_imgs)
+
+# %%
+# RandomAffine
+# ~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomAffine` transform
+# (see also :func:`~torchvision.transforms.functional.affine`)
+# performs random affine transform on an image.
+affine_transfomer = v2.RandomAffine(degrees=(30, 70), translate=(0.1, 0.3), scale=(0.5, 0.75))
+affine_imgs = [affine_transfomer(orig_img) for _ in range(4)]
+plot([orig_img] + affine_imgs)
+
+# %%
+# ElasticTransform
+# ~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.ElasticTransform` transform
+# (see also :func:`~torchvision.transforms.functional.elastic_transform`)
+# Randomly transforms the morphology of objects in images and produces a
+# see-through-water-like effect.
+elastic_transformer = v2.ElasticTransform(alpha=250.0)
+transformed_imgs = [elastic_transformer(orig_img) for _ in range(2)]
+plot([orig_img] + transformed_imgs)
+
+# %%
+# RandomCrop
+# ~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomCrop` transform
+# (see also :func:`~torchvision.transforms.functional.crop`)
+# crops an image at a random location.
+cropper = v2.RandomCrop(size=(128, 128))
+crops = [cropper(orig_img) for _ in range(4)]
+plot([orig_img] + crops)
+
+# %%
+# RandomResizedCrop
+# ~~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomResizedCrop` transform
+# (see also :func:`~torchvision.transforms.functional.resized_crop`)
+# crops an image at a random location, and then resizes the crop to a given
+# size.
+resize_cropper = v2.RandomResizedCrop(size=(32, 32))
+resized_crops = [resize_cropper(orig_img) for _ in range(4)]
+plot([orig_img] + resized_crops)
+
+# %%
+# Photometric Transforms
+# ----------------------
+# Photometric image transformation refers to the process of modifying the photometric properties of an image,
+# such as its brightness, contrast, color, or tone.
+# These transformations are applied to change the visual appearance of an image
+# while preserving its geometric structure.
+#
+# Except :class:`~torchvision.transforms.Grayscale`, the following transforms are random,
+# which means that the same transform
+# instance will produce different result each time it transforms a given image.
+#
+# Grayscale
+# ~~~~~~~~~
+# The :class:`~torchvision.transforms.Grayscale` transform
+# (see also :func:`~torchvision.transforms.functional.to_grayscale`)
+# converts an image to grayscale
+gray_img = v2.Grayscale()(orig_img)
+plot([orig_img, gray_img], cmap='gray')
+
+# %%
+# ColorJitter
+# ~~~~~~~~~~~
+# The :class:`~torchvision.transforms.ColorJitter` transform
+# randomly changes the brightness, contrast, saturation, hue, and other properties of an image.
+jitter = v2.ColorJitter(brightness=.5, hue=.3)
+jittered_imgs = [jitter(orig_img) for _ in range(4)]
+plot([orig_img] + jittered_imgs)
+
+# %%
+# GaussianBlur
+# ~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.GaussianBlur` transform
+# (see also :func:`~torchvision.transforms.functional.gaussian_blur`)
+# performs gaussian blur transform on an image.
+blurrer = v2.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5.))
+blurred_imgs = [blurrer(orig_img) for _ in range(4)]
+plot([orig_img] + blurred_imgs)
+
+# %%
+# RandomInvert
+# ~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomInvert` transform
+# (see also :func:`~torchvision.transforms.functional.invert`)
+# randomly inverts the colors of the given image.
+inverter = v2.RandomInvert()
+invertered_imgs = [inverter(orig_img) for _ in range(4)]
+plot([orig_img] + invertered_imgs)
+
+# %%
+# RandomPosterize
+# ~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomPosterize` transform
+# (see also :func:`~torchvision.transforms.functional.posterize`)
+# randomly posterizes the image by reducing the number of bits
+# of each color channel.
+posterizer = v2.RandomPosterize(bits=2)
+posterized_imgs = [posterizer(orig_img) for _ in range(4)]
+plot([orig_img] + posterized_imgs)
+
+# %%
+# RandomSolarize
+# ~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomSolarize` transform
+# (see also :func:`~torchvision.transforms.functional.solarize`)
+# randomly solarizes the image by inverting all pixel values above
+# the threshold.
+solarizer = v2.RandomSolarize(threshold=192.0)
+solarized_imgs = [solarizer(orig_img) for _ in range(4)]
+plot([orig_img] + solarized_imgs)
+
+# %%
+# RandomAdjustSharpness
+# ~~~~~~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomAdjustSharpness` transform
+# (see also :func:`~torchvision.transforms.functional.adjust_sharpness`)
+# randomly adjusts the sharpness of the given image.
+sharpness_adjuster = v2.RandomAdjustSharpness(sharpness_factor=2)
+sharpened_imgs = [sharpness_adjuster(orig_img) for _ in range(4)]
+plot([orig_img] + sharpened_imgs)
+
+# %%
+# RandomAutocontrast
+# ~~~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomAutocontrast` transform
+# (see also :func:`~torchvision.transforms.functional.autocontrast`)
+# randomly applies autocontrast to the given image.
+autocontraster = v2.RandomAutocontrast()
+autocontrasted_imgs = [autocontraster(orig_img) for _ in range(4)]
+plot([orig_img] + autocontrasted_imgs)
+
+# %%
+# RandomEqualize
+# ~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomEqualize` transform
+# (see also :func:`~torchvision.transforms.functional.equalize`)
+# randomly equalizes the histogram of the given image.
+equalizer = v2.RandomEqualize()
+equalized_imgs = [equalizer(orig_img) for _ in range(4)]
+plot([orig_img] + equalized_imgs)
+
+# %%
+# Augmentation Transforms
+# -----------------------
+# The following transforms are combinations of multiple transforms,
+# either geometric or photometric, or both.
+#
+# AutoAugment
+# ~~~~~~~~~~~
+# The :class:`~torchvision.transforms.AutoAugment` transform
+# automatically augments data based on a given auto-augmentation policy.
+# See :class:`~torchvision.transforms.AutoAugmentPolicy` for the available policies.
+policies = [v2.AutoAugmentPolicy.CIFAR10, v2.AutoAugmentPolicy.IMAGENET, v2.AutoAugmentPolicy.SVHN]
+augmenters = [v2.AutoAugment(policy) for policy in policies]
+imgs = [
+    [augmenter(orig_img) for _ in range(4)]
+    for augmenter in augmenters
+]
+row_title = [str(policy).split('.')[-1] for policy in policies]
+plot([[orig_img] + row for row in imgs], row_title=row_title)
+
+# %%
+# RandAugment
+# ~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandAugment` is an alternate version of AutoAugment.
+augmenter = v2.RandAugment()
+imgs = [augmenter(orig_img) for _ in range(4)]
+plot([orig_img] + imgs)
+
+# %%
+# TrivialAugmentWide
+# ~~~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.TrivialAugmentWide` is an alternate implementation of AutoAugment.
+# However, instead of transforming an image multiple times, it transforms an image only once
+# using a random transform from a given list with a random strength number.
+augmenter = v2.TrivialAugmentWide()
+imgs = [augmenter(orig_img) for _ in range(4)]
+plot([orig_img] + imgs)
+
+# %%
+# AugMix
+# ~~~~~~
+# The :class:`~torchvision.transforms.AugMix` transform interpolates between augmented versions of an image.
+augmenter = v2.AugMix()
+imgs = [augmenter(orig_img) for _ in range(4)]
+plot([orig_img] + imgs)
+
+# %%
+# Randomly-applied Transforms
+# ---------------------------
+#
+# The following transforms are randomly-applied given a probability ``p``.  That is, given ``p = 0.5``,
+# there is a 50% chance to return the original image, and a 50% chance to return the transformed image,
+# even when called with the same transform instance!
+#
+# RandomHorizontalFlip
+# ~~~~~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomHorizontalFlip` transform
+# (see also :func:`~torchvision.transforms.functional.hflip`)
+# performs horizontal flip of an image, with a given probability.
+hflipper = v2.RandomHorizontalFlip(p=0.5)
+transformed_imgs = [hflipper(orig_img) for _ in range(4)]
+plot([orig_img] + transformed_imgs)
+
+# %%
+# RandomVerticalFlip
+# ~~~~~~~~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomVerticalFlip` transform
+# (see also :func:`~torchvision.transforms.functional.vflip`)
+# performs vertical flip of an image, with a given probability.
+vflipper = v2.RandomVerticalFlip(p=0.5)
+transformed_imgs = [vflipper(orig_img) for _ in range(4)]
+plot([orig_img] + transformed_imgs)
+
+# %%
+# RandomApply
+# ~~~~~~~~~~~
+# The :class:`~torchvision.transforms.RandomApply` transform
+# randomly applies a list of transforms, with a given probability.
+applier = v2.RandomApply(transforms=[v2.RandomCrop(size=(64, 64))], p=0.5)
+transformed_imgs = [applier(orig_img) for _ in range(4)]
+plot([orig_img] + transformed_imgs)
diff --git a/0.16/_downloads/fd385b39c48cf3b5eb72e901830b8621/plot_transforms_e2e.ipynb b/0.16/_downloads/fd385b39c48cf3b5eb72e901830b8621/plot_transforms_e2e.ipynb
new file mode 100644
index 00000000000..6c016414b4f
--- /dev/null
+++ b/0.16/_downloads/fd385b39c48cf3b5eb72e901830b8621/plot_transforms_e2e.ipynb
@@ -0,0 +1,158 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Transforms v2: End-to-end object detection/segmentation example\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_e2e.py>` to download the full example code.</p></div>\n\nObject detection and segmentation tasks are natively supported:\n``torchvision.transforms.v2`` enables jointly transforming images, videos,\nbounding boxes, and masks.\n\nThis example showcases an end-to-end instance segmentation training case using\nTorchvision utils from ``torchvision.datasets``, ``torchvision.models`` and\n``torchvision.transforms.v2``. Everything covered here can be applied similarly\nto object detection or semantic segmentation tasks.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import pathlib\n\nimport torch\nimport torch.utils.data\n\nfrom torchvision import models, datasets, tv_tensors\nfrom torchvision.transforms import v2\n\ntorch.manual_seed(0)\n\n# This loads fake data for illustration purposes of this example. In practice, you'll have\n# to replace this with the proper data.\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nROOT = pathlib.Path(\"../assets\") / \"coco\"\nIMAGES_PATH = str(ROOT / \"images\")\nANNOTATIONS_PATH = str(ROOT / \"instances.json\")\nfrom helpers import plot"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Dataset preparation\n\nWe start off by loading the :class:`~torchvision.datasets.CocoDetection` dataset to have a look at what it currently\nreturns.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH)\n\nsample = dataset[0]\nimg, target = sample\nprint(f\"{type(img) = }\\n{type(target) = }\\n{type(target[0]) = }\\n{target[0].keys() = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Torchvision datasets preserve the data structure and types as it was intended\nby the datasets authors. So by default, the output structure may not always be\ncompatible with the models or the transforms.\n\nTo overcome that, we can use the\n:func:`~torchvision.datasets.wrap_dataset_for_transforms_v2` function. For\n:class:`~torchvision.datasets.CocoDetection`, this changes the target\nstructure to a single dictionary of lists:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=(\"boxes\", \"labels\", \"masks\"))\n\nsample = dataset[0]\nimg, target = sample\nprint(f\"{type(img) = }\\n{type(target) = }\\n{target.keys() = }\")\nprint(f\"{type(target['boxes']) = }\\n{type(target['labels']) = }\\n{type(target['masks']) = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We used the ``target_keys`` parameter to specify the kind of output we're\ninterested in. Our dataset now returns a target which is dict where the values\nare `TVTensors <what_are_tv_tensors>` (all are :class:`torch.Tensor`\nsubclasses). We're dropped all unncessary keys from the previous output, but\nif you need any of the original keys e.g. \"image_id\", you can still ask for\nit.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>If you just want to do detection, you don't need and shouldn't pass\n    \"masks\" in ``target_keys``: if masks are present in the sample, they will\n    be transformed, slowing down your transformations unnecessarily.</p></div>\n\nAs baseline, let's have a look at a sample without transformations:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "plot([dataset[0], dataset[1]])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Transforms\n\nLet's now define our pre-processing transforms. All the transforms know how\nto handle images, bouding boxes and masks when relevant.\n\nTransforms are typically passed as the ``transforms`` parameter of the\ndataset so that they can leverage multi-processing from the\n:class:`torch.utils.data.DataLoader`.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transforms = v2.Compose(\n    [\n        v2.ToImage(),\n        v2.RandomPhotometricDistort(p=1),\n        v2.RandomZoomOut(fill={tv_tensors.Image: (123, 117, 104), \"others\": 0}),\n        v2.RandomIoUCrop(),\n        v2.RandomHorizontalFlip(p=1),\n        v2.SanitizeBoundingBoxes(),\n        v2.ToDtype(torch.float32, scale=True),\n    ]\n)\n\ndataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH, transforms=transforms)\ndataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=[\"boxes\", \"labels\", \"masks\"])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "A few things are worth noting here:\n\n- We're converting the PIL image into a\n  :class:`~torchvision.transforms.v2.Image` object. This isn't strictly\n  necessary, but relying on Tensors (here: a Tensor subclass) will\n  `generally be faster <transforms_perf>`.\n- We are calling :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` to\n  make sure we remove degenerate bounding boxes, as well as their\n  corresponding labels and masks.\n  :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` should be placed\n  at least once at the end of a detection pipeline; it is particularly\n  critical if :class:`~torchvision.transforms.v2.RandomIoUCrop` was used.\n\nLet's look how the sample looks like with our augmentation pipeline in place:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "plot([dataset[0], dataset[1]])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can see that the color of the images were distorted, zoomed in or out, and flipped.\nThe bounding boxes and the masks were transformed accordingly. And without any further ado, we can start training.\n\n## Data loading and training loop\n\nBelow we're using Mask-RCNN which is an instance segmentation model, but\neverything we've covered in this tutorial also applies to object detection and\nsemantic segmentation tasks.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "data_loader = torch.utils.data.DataLoader(\n    dataset,\n    batch_size=2,\n    # We need a custom collation function here, since the object detection\n    # models expect a sequence of images and target dictionaries. The default\n    # collation function tries to torch.stack() the individual elements,\n    # which fails in general for object detection, because the number of bouding\n    # boxes varies between the images of a same batch.\n    collate_fn=lambda batch: tuple(zip(*batch)),\n)\n\nmodel = models.get_model(\"maskrcnn_resnet50_fpn_v2\", weights=None, weights_backbone=None).train()\n\nfor imgs, targets in data_loader:\n    loss_dict = model(imgs, targets)\n    # Put your training logic here\n\n    print(f\"{[img.shape for img in imgs] = }\")\n    print(f\"{[type(target) for target in targets] = }\")\n    for name, loss_val in loss_dict.items():\n        print(f\"{name:<20}{loss_val:.3f}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Training References\n\nFrom there, you can check out the [torchvision references](https://github.com/pytorch/vision/tree/main/references) where you'll find\nthe actual training scripts we use to train our models.\n\n**Disclaimer** The code in our references is more complex than what you'll\nneed for your own use-cases: this is because we're supporting different\nbackends (PIL, tensors, TVTensors) and different transforms namespaces (v1 and\nv2). So don't be afraid to simplify and only keep what you need.\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_custom_transforms.ipynb b/0.16/_generated_ipynb_notebooks/plot_custom_transforms.ipynb
new file mode 100644
index 00000000000..fecad9844a1
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_custom_transforms.ipynb
@@ -0,0 +1,97 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# How to write your own v2 transforms\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_transforms.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_transforms.py>` to download the full example code.</p></div>\n\nThis guide explains how to write transforms that are compatible with the\ntorchvision transforms V2 API.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nfrom torchvision import tv_tensors\nfrom torchvision.transforms import v2"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Just create a ``nn.Module`` and override the ``forward`` method\n\nIn most cases, this is all you're going to need, as long as you already know\nthe structure of the input that your transform will expect. For example if\nyou're just doing image classification, your transform will typically accept a\nsingle image as input, or a ``(img, label)`` input. So you can just hard-code\nyour ``forward`` method to accept just that, e.g.\n\n.. code:: python\n\n    class MyCustomTransform(torch.nn.Module):\n        def forward(self, img, label):\n            # Do some transformations\n            return new_img, new_label\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>This means that if you have a custom transform that is already compatible\n    with the V1 transforms (those in ``torchvision.transforms``), it will\n    still work with the V2 transforms without any change!</p></div>\n\nWe will illustrate this more completely below with a typical detection case,\nwhere our samples are just images, bounding boxes and labels:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class MyCustomTransform(torch.nn.Module):\n    def forward(self, img, bboxes, label):  # we assume inputs are always structured like this\n        print(\n            f\"I'm transforming an image of shape {img.shape} \"\n            f\"with bboxes = {bboxes}\\n{label = }\"\n        )\n        # Do some transformations. Here, we're just passing though the input\n        return img, bboxes, label\n\n\ntransforms = v2.Compose([\n    MyCustomTransform(),\n    v2.RandomResizedCrop((224, 224), antialias=True),\n    v2.RandomHorizontalFlip(p=1),\n    v2.Normalize(mean=[0, 0, 0], std=[1, 1, 1])\n])\n\nH, W = 256, 256\nimg = torch.rand(3, H, W)\nbboxes = tv_tensors.BoundingBoxes(\n    torch.tensor([[0, 10, 10, 20], [50, 50, 70, 70]]),\n    format=\"XYXY\",\n    canvas_size=(H, W)\n)\nlabel = 3\n\nout_img, out_bboxes, out_label = transforms(img, bboxes, label)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(f\"Output image shape: {out_img.shape}\\nout_bboxes = {out_bboxes}\\n{out_label = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-info\"><h4>Note</h4><p>While working with TVTensor classes in your code, make sure to\n    familiarize yourself with this section:\n    `tv_tensor_unwrapping_behaviour`</p></div>\n\n## Supporting arbitrary input structures\n\nIn the section above, we have assumed that you already know the structure of\nyour inputs and that you're OK with hard-coding this expected structure in\nyour code. If you want your custom transforms to be as flexible as possible,\nthis can be a bit limiting.\n\nA key feature of the builtin Torchvision V2 transforms is that they can accept\narbitrary input structure and return the same structure as output (with\ntransformed entries). For example, transforms can accept a single image, or a\ntuple of ``(img, label)``, or an arbitrary nested dictionary as input:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "structured_input = {\n    \"img\": img,\n    \"annotations\": (bboxes, label),\n    \"something_that_will_be_ignored\": (1, \"hello\")\n}\nstructured_output = v2.RandomHorizontalFlip(p=1)(structured_input)\n\nassert isinstance(structured_output, dict)\nassert structured_output[\"something_that_will_be_ignored\"] == (1, \"hello\")\nprint(f\"The transformed bboxes are:\\n{structured_output['annotations'][0]}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "If you want to reproduce this behavior in your own transform, we invite you to\nlook at our [code](https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/_transform.py)\nand adapt it to your needs.\n\nIn brief, the core logic is to unpack the input into a flat list using [pytree](https://github.com/pytorch/pytorch/blob/main/torch/utils/_pytree.py), and\nthen transform only the entries that can be transformed (the decision is made\nbased on the **class** of the entries, as all TVTensors are\ntensor-subclasses) plus some custom logic that is out of score here - check the\ncode for details. The (potentially transformed) entries are then repacked and\nreturned, in the same structure as the input.\n\nWe do not provide public dev-facing tools to achieve that at this time, but if\nthis is something that would be valuable to you, please let us know by opening\nan issue on our [GitHub repo](https://github.com/pytorch/vision/issues).\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb b/0.16/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb
new file mode 100644
index 00000000000..8e754f6c721
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb
@@ -0,0 +1,140 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# How to write your own TVTensor class\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_tv_tensors.py>` to download the full example code.</p></div>\n\nThis guide is intended for advanced users and downstream library maintainers. We explain how to\nwrite your own TVTensor class, and how to make it compatible with the built-in\nTorchvision v2 transforms. Before continuing, make sure you have read\n`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nfrom torchvision import tv_tensors\nfrom torchvision.transforms import v2"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We will create a very simple class that just inherits from the base\n:class:`~torchvision.tv_tensors.TVTensor` class. It will be enough to cover\nwhat you need to know to implement your more elaborate uses-cases. If you need\nto create a class that carries meta-data, take a look at how the\n:class:`~torchvision.tv_tensors.BoundingBoxes` class is [implemented](https://github.com/pytorch/vision/blob/main/torchvision/tv_tensors/_bounding_box.py).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class MyTVTensor(tv_tensors.TVTensor):\n    pass\n\n\nmy_dp = MyTVTensor([1, 2, 3])\nmy_dp"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Now that we have defined our custom TVTensor class, we want it to be\ncompatible with the built-in torchvision transforms, and the functional API.\nFor that, we need to implement a kernel which performs the core of the\ntransformation, and then \"hook\" it to the functional that we want to support\nvia :func:`~torchvision.transforms.v2.functional.register_kernel`.\n\nWe illustrate this process below: we create a kernel for the \"horizontal flip\"\noperation of our MyTVTensor class, and register it to the functional API.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.transforms.v2 import functional as F\n\n\n@F.register_kernel(functional=\"hflip\", tv_tensor_cls=MyTVTensor)\ndef hflip_my_tv_tensor(my_dp, *args, **kwargs):\n    print(\"Flipping!\")\n    out = my_dp.flip(-1)\n    return tv_tensors.wrap(out, like=my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "To understand why :func:`~torchvision.tv_tensors.wrap` is used, see\n`tv_tensor_unwrapping_behaviour`. Ignore the ``*args, **kwargs`` for now,\nwe will explain it below in `param_forwarding`.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>In our call to ``register_kernel`` above we used a string\n    ``functional=\"hflip\"`` to refer to the functional we want to hook into. We\n    could also have used the  functional *itself*, i.e.\n    ``@register_kernel(functional=F.hflip, ...)``.</p></div>\n\nNow that we have registered our kernel, we can call the functional API on a\n``MyTVTensor`` instance:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "my_dp = MyTVTensor(torch.rand(3, 256, 256))\n_ = F.hflip(my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "And we can also use the\n:class:`~torchvision.transforms.v2.RandomHorizontalFlip` transform, since it relies on :func:`~torchvision.transforms.v2.functional.hflip` internally:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "t = v2.RandomHorizontalFlip(p=1)\n_ = t(my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-info\"><h4>Note</h4><p>We cannot register a kernel for a transform class, we can only register a\n    kernel for a **functional**. The reason we can't register a transform\n    class is because one transform may internally rely on more than one\n    functional, so in general we can't register a single kernel for a given\n    class.</p></div>\n\n\n## Parameter forwarding, and ensuring future compatibility of your kernels\n\nThe functional API that you're hooking into is public and therefore\n**backward** compatible: we guarantee that the parameters of these functionals\nwon't be removed or renamed without a proper deprecation cycle. However, we\ndon't guarantee **forward** compatibility, and we may add new parameters in\nthe future.\n\nImagine that in a future version, Torchvision adds a new ``inplace`` parameter\nto its :func:`~torchvision.transforms.v2.functional.hflip` functional. If you\nalready defined and registered your own kernel as\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "def hflip_my_tv_tensor(my_dp):  # noqa\n    print(\"Flipping!\")\n    out = my_dp.flip(-1)\n    return tv_tensors.wrap(out, like=my_dp)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "then calling ``F.hflip(my_dp)`` will **fail**, because ``hflip`` will try to\npass the new ``inplace`` parameter to your kernel, but your kernel doesn't\naccept it.\n\nFor this reason, we recommend to always define your kernels with\n``*args, **kwargs`` in their signature, as done above. This way, your kernel\nwill be able to accept any new parameter that we may add in the future.\n(Technically, adding `**kwargs` only should be enough).\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb b/0.16/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb
new file mode 100644
index 00000000000..d4eba2316ac
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb
@@ -0,0 +1,133 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# How to use CutMix and MixUp\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_cutmix_mixup.py>` to download the full example code.</p></div>\n\n:class:`~torchvision.transforms.v2.CutMix` and\n:class:`~torchvision.transforms.v2.MixUp` are popular augmentation strategies\nthat can improve classification accuracy.\n\nThese transforms are slightly different from the rest of the Torchvision\ntransforms, because they expect\n**batches** of samples as input, not individual images. In this example we'll\nexplain how to use them: after the ``DataLoader``, or as part of a collation\nfunction.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nfrom torchvision.datasets import FakeData\nfrom torchvision.transforms import v2\n\n\nNUM_CLASSES = 100"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Pre-processing pipeline\n\nWe'll use a simple but typical image classification pipeline:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "preproc = v2.Compose([\n    v2.PILToTensor(),\n    v2.RandomResizedCrop(size=(224, 224), antialias=True),\n    v2.RandomHorizontalFlip(p=0.5),\n    v2.ToDtype(torch.float32, scale=True),  # to float32 in [0, 1]\n    v2.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),  # typically from ImageNet\n])\n\ndataset = FakeData(size=1000, num_classes=NUM_CLASSES, transform=preproc)\n\nimg, label = dataset[0]\nprint(f\"{type(img) = }, {img.dtype = }, {img.shape = }, {label = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "One important thing to note is that neither CutMix nor MixUp are part of this\npre-processing pipeline. We'll add them a bit later once we define the\nDataLoader. Just as a refresher, this is what the DataLoader and training loop\nwould look like if we weren't using CutMix or MixUp:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torch.utils.data import DataLoader\n\ndataloader = DataLoader(dataset, batch_size=4, shuffle=True)\n\nfor images, labels in dataloader:\n    print(f\"{images.shape = }, {labels.shape = }\")\n    print(labels.dtype)\n    # <rest of the training loop here>\n    break"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Where to use MixUp and CutMix\n\n### After the DataLoader\n\nNow let's add CutMix and MixUp. The simplest way to do this right after the\nDataLoader: the Dataloader has already batched the images and labels for us,\nand this is exactly what these transforms expect as input:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dataloader = DataLoader(dataset, batch_size=4, shuffle=True)\n\ncutmix = v2.CutMix(num_classes=NUM_CLASSES)\nmixup = v2.MixUp(num_classes=NUM_CLASSES)\ncutmix_or_mixup = v2.RandomChoice([cutmix, mixup])\n\nfor images, labels in dataloader:\n    print(f\"Before CutMix/MixUp: {images.shape = }, {labels.shape = }\")\n    images, labels = cutmix_or_mixup(images, labels)\n    print(f\"After CutMix/MixUp: {images.shape = }, {labels.shape = }\")\n\n    # <rest of the training loop here>\n    break"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Note how the labels were also transformed: we went from a batched label of\nshape (batch_size,) to a tensor of shape (batch_size, num_classes). The\ntransformed labels can still be passed as-is to a loss function like\n:func:`torch.nn.functional.cross_entropy`.\n\n### As part of the collation function\n\nPassing the transforms after the DataLoader is the simplest way to use CutMix\nand MixUp, but one disadvantage is that it does not take advantage of the\nDataLoader multi-processing. For that, we can pass those transforms as part of\nthe collation function (refer to the [PyTorch docs](https://pytorch.org/docs/stable/data.html#dataloader-collate-fn) to learn\nmore about collation).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torch.utils.data import default_collate\n\n\ndef collate_fn(batch):\n    return cutmix_or_mixup(*default_collate(batch))\n\n\ndataloader = DataLoader(dataset, batch_size=4, shuffle=True, num_workers=2, collate_fn=collate_fn)\n\nfor images, labels in dataloader:\n    print(f\"{images.shape = }, {labels.shape = }\")\n    # No need to call cutmix_or_mixup, it's already been called as part of the DataLoader!\n    # <rest of the training loop here>\n    break"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Non-standard input format\n\nSo far we've used a typical sample structure where we pass ``(images,\nlabels)`` as inputs. MixUp and CutMix will magically work by default with most\ncommon sample structures: tuples where the second parameter is a tensor label,\nor dict with a \"label[s]\" key. Look at the documentation of the\n``labels_getter`` parameter for more details.\n\nIf your samples have a different structure, you can still use CutMix and MixUp\nby passing a callable to the ``labels_getter`` parameter. For example:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "batch = {\n    \"imgs\": torch.rand(4, 3, 224, 224),\n    \"target\": {\n        \"classes\": torch.randint(0, NUM_CLASSES, size=(4,)),\n        \"some_other_key\": \"this is going to be passed-through\"\n    }\n}\n\n\ndef labels_getter(batch):\n    return batch[\"target\"][\"classes\"]\n\n\nout = v2.CutMix(num_classes=NUM_CLASSES, labels_getter=labels_getter)(batch)\nprint(f\"{out['imgs'].shape = }, {out['target']['classes'].shape = }\")"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_optical_flow.ipynb b/0.16/_generated_ipynb_notebooks/plot_optical_flow.ipynb
new file mode 100644
index 00000000000..ce2b6ef7b7c
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_optical_flow.ipynb
@@ -0,0 +1,176 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Optical Flow: Predicting movement with the RAFT model\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_optical_flow.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_optical_flow.py>` to download the full example code.</p></div>\n\nOptical flow is the task of predicting movement between two images, usually two\nconsecutive frames of a video. Optical flow models take two images as input, and\npredict a flow: the flow indicates the displacement of every single pixel in the\nfirst image, and maps it to its corresponding pixel in the second image. Flows\nare (2, H, W)-dimensional tensors, where the first axis corresponds to the\npredicted horizontal and vertical displacements.\n\nThe following example illustrates how torchvision can be used to predict flows\nusing our implementation of the RAFT model. We will also see how to convert the\npredicted flows to RGB images for visualization.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import numpy as np\nimport torch\nimport matplotlib.pyplot as plt\nimport torchvision.transforms.functional as F\n\n\nplt.rcParams[\"savefig.bbox\"] = \"tight\"\n\n\ndef plot(imgs, **imshow_kwargs):\n    if not isinstance(imgs[0], list):\n        # Make a 2d grid even if there's just 1 row\n        imgs = [imgs]\n\n    num_rows = len(imgs)\n    num_cols = len(imgs[0])\n    _, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)\n    for row_idx, row in enumerate(imgs):\n        for col_idx, img in enumerate(row):\n            ax = axs[row_idx, col_idx]\n            img = F.to_pil_image(img.to(\"cpu\"))\n            ax.imshow(np.asarray(img), **imshow_kwargs)\n            ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])\n\n    plt.tight_layout()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Reading Videos Using Torchvision\nWe will first read a video using :func:`~torchvision.io.read_video`.\nAlternatively one can use the new :class:`~torchvision.io.VideoReader` API (if\ntorchvision is built from source).\nThe video we will use here is free of use from [pexels.com](https://www.pexels.com/video/a-man-playing-a-game-of-basketball-5192157/),\ncredits go to [Pavel Danilyuk](https://www.pexels.com/@pavel-danilyuk).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import tempfile\nfrom pathlib import Path\nfrom urllib.request import urlretrieve\n\n\nvideo_url = \"https://download.pytorch.org/tutorial/pexelscom_pavel_danilyuk_basketball_hd.mp4\"\nvideo_path = Path(tempfile.mkdtemp()) / \"basketball.mp4\"\n_ = urlretrieve(video_url, video_path)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        ":func:`~torchvision.io.read_video` returns the video frames, audio frames and\nthe metadata associated with the video. In our case, we only need the video\nframes.\n\nHere we will just make 2 predictions between 2 pre-selected pairs of frames,\nnamely frames (100, 101) and (150, 151). Each of these pairs corresponds to a\nsingle model input.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.io import read_video\nframes, _, _ = read_video(str(video_path), output_format=\"TCHW\")\n\nimg1_batch = torch.stack([frames[100], frames[150]])\nimg2_batch = torch.stack([frames[101], frames[151]])\n\nplot(img1_batch)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The RAFT model accepts RGB images. We first get the frames from\n:func:`~torchvision.io.read_video` and resize them to ensure their dimensions\nare divisible by 8. Note that we explicitly use ``antialias=False``, because\nthis is how those models were trained. Then we use the transforms bundled into\nthe weights in order to preprocess the input and rescale its values to the\nrequired ``[-1, 1]`` interval.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.optical_flow import Raft_Large_Weights\n\nweights = Raft_Large_Weights.DEFAULT\ntransforms = weights.transforms()\n\n\ndef preprocess(img1_batch, img2_batch):\n    img1_batch = F.resize(img1_batch, size=[520, 960], antialias=False)\n    img2_batch = F.resize(img2_batch, size=[520, 960], antialias=False)\n    return transforms(img1_batch, img2_batch)\n\n\nimg1_batch, img2_batch = preprocess(img1_batch, img2_batch)\n\nprint(f\"shape = {img1_batch.shape}, dtype = {img1_batch.dtype}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Estimating Optical flow using RAFT\nWe will use our RAFT implementation from\n:func:`~torchvision.models.optical_flow.raft_large`, which follows the same\narchitecture as the one described in the [original paper](https://arxiv.org/abs/2003.12039).\nWe also provide the :func:`~torchvision.models.optical_flow.raft_small` model\nbuilder, which is smaller and faster to run, sacrificing a bit of accuracy.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.optical_flow import raft_large\n\n# If you can, run this example on a GPU, it will be a lot faster.\ndevice = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n\nmodel = raft_large(weights=Raft_Large_Weights.DEFAULT, progress=False).to(device)\nmodel = model.eval()\n\nlist_of_flows = model(img1_batch.to(device), img2_batch.to(device))\nprint(f\"type = {type(list_of_flows)}\")\nprint(f\"length = {len(list_of_flows)} = number of iterations of the model\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The RAFT model outputs lists of predicted flows where each entry is a\n(N, 2, H, W) batch of predicted flows that corresponds to a given \"iteration\"\nin the model. For more details on the iterative nature of the model, please\nrefer to the [original paper](https://arxiv.org/abs/2003.12039). Here, we\nare only interested in the final predicted flows (they are the most accurate\nones), so we will just retrieve the last item in the list.\n\nAs described above, a flow is a tensor with dimensions (2, H, W) (or (N, 2, H,\nW) for batches of flows) where each entry corresponds to the horizontal and\nvertical displacement of each pixel from the first image to the second image.\nNote that the predicted flows are in \"pixel\" unit, they are not normalized\nw.r.t. the dimensions of the images.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "predicted_flows = list_of_flows[-1]\nprint(f\"dtype = {predicted_flows.dtype}\")\nprint(f\"shape = {predicted_flows.shape} = (N, 2, H, W)\")\nprint(f\"min = {predicted_flows.min()}, max = {predicted_flows.max()}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing predicted flows\nTorchvision provides the :func:`~torchvision.utils.flow_to_image` utility to\nconvert a flow into an RGB image. It also supports batches of flows.\neach \"direction\" in the flow will be mapped to a given RGB color. In the\nimages below, pixels with similar colors are assumed by the model to be moving\nin similar directions. The model is properly able to predict the movement of\nthe ball and the player. Note in particular the different predicted direction\nof the ball in the first image (going to the left) and in the second image\n(going up).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import flow_to_image\n\nflow_imgs = flow_to_image(predicted_flows)\n\n# The images have been mapped into [-1, 1] but for plotting we want them in [0, 1]\nimg1_batch = [(img1 + 1) / 2 for img1 in img1_batch]\n\ngrid = [[img1, flow_img] for (img1, flow_img) in zip(img1_batch, flow_imgs)]\nplot(grid)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Bonus: Creating GIFs of predicted flows\nIn the example above we have only shown the predicted flows of 2 pairs of\nframes. A fun way to apply the Optical Flow models is to run the model on an\nentire video, and create a new video from all the predicted flows. Below is a\nsnippet that can get you started with this. We comment out the code, because\nthis example is being rendered on a machine without a GPU, and it would take\ntoo long to run it.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "# from torchvision.io import write_jpeg\n# for i, (img1, img2) in enumerate(zip(frames, frames[1:])):\n#     # Note: it would be faster to predict batches of flows instead of individual flows\n#     img1, img2 = preprocess(img1, img2)\n\n#     list_of_flows = model(img1.to(device), img2.to(device))\n#     predicted_flow = list_of_flows[-1][0]\n#     flow_img = flow_to_image(predicted_flow).to(\"cpu\")\n#     output_folder = \"/tmp/\"  # Update this to the folder of your choice\n#     write_jpeg(flow_img, output_folder + f\"predicted_flow_{i}.jpg\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Once the .jpg flow images are saved, you can convert them into a video or a\nGIF using ffmpeg with e.g.:\n\nffmpeg -f image2 -framerate 30 -i predicted_flow_%d.jpg -loop -1 flow.gif\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb b/0.16/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb
new file mode 100644
index 00000000000..9ea709ee226
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb
@@ -0,0 +1,205 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Repurposing masks into bounding boxes\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_repurposing_annotations.py>` to download the full example code.</p></div>\n\nThe following example illustrates the operations available\nthe `torchvision.ops <ops>` module for repurposing\nsegmentation masks into object localization annotations for different tasks\n(e.g. transforming masks used by instance and panoptic segmentation\nmethods into bounding boxes used by object detection methods).\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nimport numpy as np\nimport torch\nimport matplotlib.pyplot as plt\n\nimport torchvision.transforms.functional as F\n\n\nASSETS_DIRECTORY = \"../assets\"\n\nplt.rcParams[\"savefig.bbox\"] = \"tight\"\n\n\ndef show(imgs):\n    if not isinstance(imgs, list):\n        imgs = [imgs]\n    fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)\n    for i, img in enumerate(imgs):\n        img = img.detach()\n        img = F.to_pil_image(img)\n        axs[0, i].imshow(np.asarray(img))\n        axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Masks\nIn tasks like instance and panoptic segmentation, masks are commonly defined, and are defined by this package,\nas a multi-dimensional array (e.g. a NumPy array or a PyTorch tensor) with the following shape:\n\n      (num_objects, height, width)\n\nWhere num_objects is the number of annotated objects in the image. Each (height, width) object corresponds to exactly\none object. For example, if your input image has the dimensions 224 x 224 and has four annotated objects the shape\nof your masks annotation has the following shape:\n\n      (4, 224, 224).\n\nA nice property of masks is that they can be easily repurposed to be used in methods to solve a variety of object\nlocalization tasks.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Converting Masks to Bounding Boxes\nFor example, the :func:`~torchvision.ops.masks_to_boxes` operation can be used to\ntransform masks into bounding boxes that can be\nused as input to detection models such as FasterRCNN and RetinaNet.\nWe will take images and masks from the [PenFudan Dataset](https://www.cis.upenn.edu/~jshi/ped_html/).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.io import read_image\n\nimg_path = os.path.join(ASSETS_DIRECTORY, \"FudanPed00054.png\")\nmask_path = os.path.join(ASSETS_DIRECTORY, \"FudanPed00054_mask.png\")\nimg = read_image(img_path)\nmask = read_image(mask_path)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Here the masks are represented as a PNG Image, with floating point values.\nEach pixel is encoded as different colors, with 0 being background.\nNotice that the spatial dimensions of image and mask match.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(mask.size())\nprint(img.size())\nprint(mask)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "# We get the unique colors, as these would be the object ids.\nobj_ids = torch.unique(mask)\n\n# first id is the background, so remove it.\nobj_ids = obj_ids[1:]\n\n# split the color-encoded mask into a set of boolean masks.\n# Note that this snippet would work as well if the masks were float values instead of ints.\nmasks = mask == obj_ids[:, None, None]"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Now the masks are a boolean tensor.\nThe first dimension in this case 3 and denotes the number of instances: there are 3 people in the image.\nThe other two dimensions are height and width, which are equal to the dimensions of the image.\nFor each instance, the boolean tensors represent if the particular pixel\nbelongs to the segmentation mask of the image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(masks.size())\nprint(masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let us visualize an image and plot its corresponding segmentation masks.\nWe will use the :func:`~torchvision.utils.draw_segmentation_masks` to draw the segmentation masks.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_segmentation_masks\n\ndrawn_masks = []\nfor mask in masks:\n    drawn_masks.append(draw_segmentation_masks(img, mask, alpha=0.8, colors=\"blue\"))\n\nshow(drawn_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "To convert the boolean masks into bounding boxes.\nWe will use the :func:`~torchvision.ops.masks_to_boxes` from the torchvision.ops module\nIt returns the boxes in ``(xmin, ymin, xmax, ymax)`` format.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.ops import masks_to_boxes\n\nboxes = masks_to_boxes(masks)\nprint(boxes.size())\nprint(boxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As the shape denotes, there are 3 boxes and in ``(xmin, ymin, xmax, ymax)`` format.\nThese can be visualized very easily with :func:`~torchvision.utils.draw_bounding_boxes` utility\nprovided in `torchvision.utils <utils>`.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_bounding_boxes\n\ndrawn_boxes = draw_bounding_boxes(img, boxes, colors=\"red\")\nshow(drawn_boxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "These boxes can now directly be used by detection models in torchvision.\nHere is demo with a Faster R-CNN model loaded from\n:func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights\n\nweights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT\nmodel = fasterrcnn_resnet50_fpn(weights=weights, progress=False)\nprint(img.size())\n\ntranforms = weights.transforms()\nimg = tranforms(img)\ntarget = {}\ntarget[\"boxes\"] = boxes\ntarget[\"labels\"] = labels = torch.ones((masks.size(0),), dtype=torch.int64)\ndetection_outputs = model(img.unsqueeze(0), [target])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Converting Segmentation Dataset to Detection Dataset\n\nWith this utility it becomes very simple to convert a segmentation dataset to a detection dataset.\nWith this we can now use a segmentation dataset to train a detection model.\nOne can similarly convert panoptic dataset to detection dataset.\nHere is an example where we re-purpose the dataset from the\n[PenFudan Detection Tutorial](https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class SegmentationToDetectionDataset(torch.utils.data.Dataset):\n    def __init__(self, root, transforms):\n        self.root = root\n        self.transforms = transforms\n        # load all image files, sorting them to\n        # ensure that they are aligned\n        self.imgs = list(sorted(os.listdir(os.path.join(root, \"PNGImages\"))))\n        self.masks = list(sorted(os.listdir(os.path.join(root, \"PedMasks\"))))\n\n    def __getitem__(self, idx):\n        # load images and masks\n        img_path = os.path.join(self.root, \"PNGImages\", self.imgs[idx])\n        mask_path = os.path.join(self.root, \"PedMasks\", self.masks[idx])\n\n        img = read_image(img_path)\n        mask = read_image(mask_path)\n\n        img = F.convert_image_dtype(img, dtype=torch.float)\n        mask = F.convert_image_dtype(mask, dtype=torch.float)\n\n        # We get the unique colors, as these would be the object ids.\n        obj_ids = torch.unique(mask)\n\n        # first id is the background, so remove it.\n        obj_ids = obj_ids[1:]\n\n        # split the color-encoded mask into a set of boolean masks.\n        masks = mask == obj_ids[:, None, None]\n\n        boxes = masks_to_boxes(masks)\n\n        # there is only one class\n        labels = torch.ones((masks.shape[0],), dtype=torch.int64)\n\n        target = {}\n        target[\"boxes\"] = boxes\n        target[\"labels\"] = labels\n\n        if self.transforms is not None:\n            img, target = self.transforms(img, target)\n\n        return img, target"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb b/0.16/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb
new file mode 100644
index 00000000000..2509ad25532
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb
@@ -0,0 +1,133 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Torchscript support\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_scripted_tensor_transforms.py>` to download the full example code.</p></div>\n\nThis example illustrates [torchscript](https://pytorch.org/docs/stable/jit.html) support of the torchvision\n`transforms <transforms>` on Tensor images.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from pathlib import Path\n\nimport matplotlib.pyplot as plt\n\nimport torch\nimport torch.nn as nn\n\nimport torchvision.transforms as v1\nfrom torchvision.io import read_image\n\nplt.rcParams[\"savefig.bbox\"] = 'tight'\ntorch.manual_seed(1)\n\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nimport sys\nsys.path += [\"../transforms\"]\nfrom helpers import plot\nASSETS_PATH = Path('../assets')"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Most transforms support torchscript. For composing transforms, we use\n:class:`torch.nn.Sequential` instead of\n:class:`~torchvision.transforms.v2.Compose`:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dog1 = read_image(str(ASSETS_PATH / 'dog1.jpg'))\ndog2 = read_image(str(ASSETS_PATH / 'dog2.jpg'))\n\ntransforms = torch.nn.Sequential(\n    v1.RandomCrop(224),\n    v1.RandomHorizontalFlip(p=0.3),\n)\n\nscripted_transforms = torch.jit.script(transforms)\n\nplot([dog1, scripted_transforms(dog1), dog2, scripted_transforms(dog2)])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-danger\"><h4>Warning</h4><p>Above we have used transforms from the ``torchvision.transforms``\n    namespace, i.e. the \"v1\" transforms. The v2 transforms from the\n    ``torchvision.transforms.v2`` namespace are the `recommended\n    <v1_or_v2>` way to use transforms in your code.\n\n    The v2 transforms also support torchscript, but if you call\n    ``torch.jit.script()`` on a v2 **class** transform, you'll actually end up\n    with its (scripted) v1 equivalent.  This may lead to slightly different\n    results between the scripted and eager executions due to implementation\n    differences between v1 and v2.\n\n    If you really need torchscript support for the v2 transforms, **we\n    recommend scripting the functionals** from the\n    ``torchvision.transforms.v2.functional`` namespace to avoid surprises.</p></div>\n\nBelow we now show how to combine image transformations and a model forward\npass, while using ``torch.jit.script`` to obtain a single scripted module.\n\nLet's define a ``Predictor`` module that transforms the input tensor and then\napplies an ImageNet model on it.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models import resnet18, ResNet18_Weights\n\n\nclass Predictor(nn.Module):\n\n    def __init__(self):\n        super().__init__()\n        weights = ResNet18_Weights.DEFAULT\n        self.resnet18 = resnet18(weights=weights, progress=False).eval()\n        self.transforms = weights.transforms(antialias=True)\n\n    def forward(self, x: torch.Tensor) -> torch.Tensor:\n        with torch.no_grad():\n            x = self.transforms(x)\n            y_pred = self.resnet18(x)\n            return y_pred.argmax(dim=1)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Now, let's define scripted and non-scripted instances of ``Predictor`` and\napply it on multiple tensor images of the same size\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n\npredictor = Predictor().to(device)\nscripted_predictor = torch.jit.script(predictor).to(device)\n\nbatch = torch.stack([dog1, dog2]).to(device)\n\nres = predictor(batch)\nres_scripted = scripted_predictor(batch)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can verify that the prediction of the scripted and non-scripted models are\nthe same:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import json\n\nwith open(Path('../assets') / 'imagenet_class_index.json') as labels_file:\n    labels = json.load(labels_file)\n\nfor i, (pred, pred_scripted) in enumerate(zip(res, res_scripted)):\n    assert pred == pred_scripted\n    print(f\"Prediction for Dog {i + 1}: {labels[str(pred.item())]}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Since the model is scripted, it can be easily dumped on disk and re-used\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import tempfile\n\nwith tempfile.NamedTemporaryFile() as f:\n    scripted_predictor.save(f.name)\n\n    dumped_scripted_predictor = torch.jit.load(f.name)\n    res_scripted_dumped = dumped_scripted_predictor(batch)\nassert (res_scripted_dumped == res_scripted).all()"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb b/0.16/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb
new file mode 100644
index 00000000000..6c016414b4f
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb
@@ -0,0 +1,158 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Transforms v2: End-to-end object detection/segmentation example\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_e2e.py>` to download the full example code.</p></div>\n\nObject detection and segmentation tasks are natively supported:\n``torchvision.transforms.v2`` enables jointly transforming images, videos,\nbounding boxes, and masks.\n\nThis example showcases an end-to-end instance segmentation training case using\nTorchvision utils from ``torchvision.datasets``, ``torchvision.models`` and\n``torchvision.transforms.v2``. Everything covered here can be applied similarly\nto object detection or semantic segmentation tasks.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import pathlib\n\nimport torch\nimport torch.utils.data\n\nfrom torchvision import models, datasets, tv_tensors\nfrom torchvision.transforms import v2\n\ntorch.manual_seed(0)\n\n# This loads fake data for illustration purposes of this example. In practice, you'll have\n# to replace this with the proper data.\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nROOT = pathlib.Path(\"../assets\") / \"coco\"\nIMAGES_PATH = str(ROOT / \"images\")\nANNOTATIONS_PATH = str(ROOT / \"instances.json\")\nfrom helpers import plot"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Dataset preparation\n\nWe start off by loading the :class:`~torchvision.datasets.CocoDetection` dataset to have a look at what it currently\nreturns.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH)\n\nsample = dataset[0]\nimg, target = sample\nprint(f\"{type(img) = }\\n{type(target) = }\\n{type(target[0]) = }\\n{target[0].keys() = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Torchvision datasets preserve the data structure and types as it was intended\nby the datasets authors. So by default, the output structure may not always be\ncompatible with the models or the transforms.\n\nTo overcome that, we can use the\n:func:`~torchvision.datasets.wrap_dataset_for_transforms_v2` function. For\n:class:`~torchvision.datasets.CocoDetection`, this changes the target\nstructure to a single dictionary of lists:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=(\"boxes\", \"labels\", \"masks\"))\n\nsample = dataset[0]\nimg, target = sample\nprint(f\"{type(img) = }\\n{type(target) = }\\n{target.keys() = }\")\nprint(f\"{type(target['boxes']) = }\\n{type(target['labels']) = }\\n{type(target['masks']) = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We used the ``target_keys`` parameter to specify the kind of output we're\ninterested in. Our dataset now returns a target which is dict where the values\nare `TVTensors <what_are_tv_tensors>` (all are :class:`torch.Tensor`\nsubclasses). We're dropped all unncessary keys from the previous output, but\nif you need any of the original keys e.g. \"image_id\", you can still ask for\nit.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>If you just want to do detection, you don't need and shouldn't pass\n    \"masks\" in ``target_keys``: if masks are present in the sample, they will\n    be transformed, slowing down your transformations unnecessarily.</p></div>\n\nAs baseline, let's have a look at a sample without transformations:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "plot([dataset[0], dataset[1]])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Transforms\n\nLet's now define our pre-processing transforms. All the transforms know how\nto handle images, bouding boxes and masks when relevant.\n\nTransforms are typically passed as the ``transforms`` parameter of the\ndataset so that they can leverage multi-processing from the\n:class:`torch.utils.data.DataLoader`.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transforms = v2.Compose(\n    [\n        v2.ToImage(),\n        v2.RandomPhotometricDistort(p=1),\n        v2.RandomZoomOut(fill={tv_tensors.Image: (123, 117, 104), \"others\": 0}),\n        v2.RandomIoUCrop(),\n        v2.RandomHorizontalFlip(p=1),\n        v2.SanitizeBoundingBoxes(),\n        v2.ToDtype(torch.float32, scale=True),\n    ]\n)\n\ndataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH, transforms=transforms)\ndataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=[\"boxes\", \"labels\", \"masks\"])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "A few things are worth noting here:\n\n- We're converting the PIL image into a\n  :class:`~torchvision.transforms.v2.Image` object. This isn't strictly\n  necessary, but relying on Tensors (here: a Tensor subclass) will\n  `generally be faster <transforms_perf>`.\n- We are calling :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` to\n  make sure we remove degenerate bounding boxes, as well as their\n  corresponding labels and masks.\n  :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` should be placed\n  at least once at the end of a detection pipeline; it is particularly\n  critical if :class:`~torchvision.transforms.v2.RandomIoUCrop` was used.\n\nLet's look how the sample looks like with our augmentation pipeline in place:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "plot([dataset[0], dataset[1]])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can see that the color of the images were distorted, zoomed in or out, and flipped.\nThe bounding boxes and the masks were transformed accordingly. And without any further ado, we can start training.\n\n## Data loading and training loop\n\nBelow we're using Mask-RCNN which is an instance segmentation model, but\neverything we've covered in this tutorial also applies to object detection and\nsemantic segmentation tasks.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "data_loader = torch.utils.data.DataLoader(\n    dataset,\n    batch_size=2,\n    # We need a custom collation function here, since the object detection\n    # models expect a sequence of images and target dictionaries. The default\n    # collation function tries to torch.stack() the individual elements,\n    # which fails in general for object detection, because the number of bouding\n    # boxes varies between the images of a same batch.\n    collate_fn=lambda batch: tuple(zip(*batch)),\n)\n\nmodel = models.get_model(\"maskrcnn_resnet50_fpn_v2\", weights=None, weights_backbone=None).train()\n\nfor imgs, targets in data_loader:\n    loss_dict = model(imgs, targets)\n    # Put your training logic here\n\n    print(f\"{[img.shape for img in imgs] = }\")\n    print(f\"{[type(target) for target in targets] = }\")\n    for name, loss_val in loss_dict.items():\n        print(f\"{name:<20}{loss_val:.3f}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Training References\n\nFrom there, you can check out the [torchvision references](https://github.com/pytorch/vision/tree/main/references) where you'll find\nthe actual training scripts we use to train our models.\n\n**Disclaimer** The code in our references is more complex than what you'll\nneed for your own use-cases: this is because we're supporting different\nbackends (PIL, tensors, TVTensors) and different transforms namespaces (v1 and\nv2). So don't be afraid to simplify and only keep what you need.\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb b/0.16/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb
new file mode 100644
index 00000000000..751326092f5
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb
@@ -0,0 +1,154 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Getting started with transforms v2\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_getting_started.py>` to download the full example code.</p></div>\n\nThis example illustrates all of what you need to know to get started with the\nnew :mod:`torchvision.transforms.v2` API. We'll cover simple tasks like\nimage classification, and more advanced ones like object detection /\nsegmentation.\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "First, a bit of setup\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from pathlib import Path\nimport torch\nimport matplotlib.pyplot as plt\nplt.rcParams[\"savefig.bbox\"] = 'tight'\n\nfrom torchvision.transforms import v2\nfrom torchvision.io import read_image\n\ntorch.manual_seed(1)\n\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nfrom helpers import plot\nimg = read_image(str(Path('../assets') / 'astronaut.jpg'))\nprint(f\"{type(img) = }, {img.dtype = }, {img.shape = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## The basics\n\nThe Torchvision transforms behave like a regular :class:`torch.nn.Module` (in\nfact, most of them are): instantiate a transform, pass an input, get a\ntransformed output:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transform = v2.RandomCrop(size=(224, 224))\nout = transform(img)\n\nplot([img, out])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## I just want to do image classification\n\nIf you just care about image classification, things are very simple. A basic\nclassification pipeline may look like this:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transforms = v2.Compose([\n    v2.RandomResizedCrop(size=(224, 224), antialias=True),\n    v2.RandomHorizontalFlip(p=0.5),\n    v2.ToDtype(torch.float32, scale=True),\n    v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n])\nout = transforms(img)\n\nplot([img, out])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Such transformation pipeline is typically passed as the ``transform`` argument\nto the `Datasets <datasets>`, e.g. ``ImageNet(...,\ntransform=transforms)``.\n\nThat's pretty much all there is. From there, read through our `main docs\n<transforms>` to learn more about recommended practices and conventions, or\nexplore more `examples <transforms_gallery>` e.g. how to use augmentation\ntransforms like `CutMix and MixUp\n<sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py>`.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>If you're already relying on the ``torchvision.transforms`` v1 API,\n    we recommend to `switch to the new v2 transforms<v1_or_v2>`. It's\n    very easy: the v2 transforms are fully compatible with the v1 API, so you\n    only need to change the import!</p></div>\n\n## Detection, Segmentation, Videos\n\nThe new Torchvision transforms in the ``torchvision.transforms.v2`` namespace\nsupport tasks beyond image classification: they can also transform bounding\nboxes, segmentation / detection masks, or videos.\n\nLet's briefly look at a detection example with bounding boxes.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision import tv_tensors  # we'll describe this a bit later, bare with us\n\nboxes = tv_tensors.BoundingBoxes(\n    [\n        [15, 10, 370, 510],\n        [275, 340, 510, 510],\n        [130, 345, 210, 425]\n    ],\n    format=\"XYXY\", canvas_size=img.shape[-2:])\n\ntransforms = v2.Compose([\n    v2.RandomResizedCrop(size=(224, 224), antialias=True),\n    v2.RandomPhotometricDistort(p=1),\n    v2.RandomHorizontalFlip(p=1),\n])\nout_img, out_boxes = transforms(img, boxes)\nprint(type(boxes), type(out_boxes))\n\nplot([(img, boxes), (out_img, out_boxes)])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The example above focuses on object detection. But if we had masks\n(:class:`torchvision.tv_tensors.Mask`) for object segmentation or semantic\nsegmentation, or videos (:class:`torchvision.tv_tensors.Video`), we could have\npassed them to the transforms in exactly the same way.\n\nBy now you likely have a few questions: what are these TVTensors, how do we\nuse them, and what is the expected input/output of those transforms? We'll\nanswer these in the next sections.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n## What are TVTensors?\n\nTVTensors are :class:`torch.Tensor` subclasses. The available TVTensors are\n:class:`~torchvision.tv_tensors.Image`,\n:class:`~torchvision.tv_tensors.BoundingBoxes`,\n:class:`~torchvision.tv_tensors.Mask`, and\n:class:`~torchvision.tv_tensors.Video`.\n\nTVTensors look and feel just like regular tensors - they **are** tensors.\nEverything that is supported on a plain :class:`torch.Tensor` like ``.sum()``\nor any ``torch.*`` operator will also work on a TVTensor:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "img_dp = tv_tensors.Image(torch.randint(0, 256, (3, 256, 256), dtype=torch.uint8))\n\nprint(f\"{isinstance(img_dp, torch.Tensor) = }\")\nprint(f\"{img_dp.dtype = }, {img_dp.shape = }, {img_dp.sum() = }\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "These TVTensor classes are at the core of the transforms: in order to\ntransform a given input, the transforms first look at the **class** of the\nobject, and dispatch to the appropriate implementation accordingly.\n\nYou don't need to know much more about TVTensors at this point, but advanced\nusers who want to learn more can refer to\n`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.\n\n## What do I pass as input?\n\nAbove, we've seen two examples: one where we passed a single image as input\ni.e. ``out = transforms(img)``, and one where we passed both an image and\nbounding boxes, i.e. ``out_img, out_boxes = transforms(img, boxes)``.\n\nIn fact, transforms support **arbitrary input structures**. The input can be a\nsingle image, a tuple, an arbitrarily nested dictionary... pretty much\nanything. The same structure will be returned as output. Below, we use the\nsame detection transforms, but pass a tuple (image, target_dict) as input and\nwe're getting the same structure as output:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "target = {\n    \"boxes\": boxes,\n    \"labels\": torch.arange(boxes.shape[0]),\n    \"this_is_ignored\": (\"arbitrary\", {\"structure\": \"!\"})\n}\n\n# Re-using the transforms and definitions from above.\nout_img, out_target = transforms(img, target)\n\nplot([(img, target[\"boxes\"]), (out_img, out_target[\"boxes\"])])\nprint(f\"{out_target['this_is_ignored']}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We passed a tuple so we get a tuple back, and the second element is the\ntranformed target dict. Transforms don't really care about the structure of\nthe input; as mentioned above, they only care about the **type** of the\nobjects and transforms them accordingly.\n\n*Foreign* objects like strings or ints are simply passed-through. This can be\nuseful e.g. if you want to associate a path with every single sample when\ndebugging!\n\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>**Disclaimer** This note is slightly advanced and can be safely skipped on\n    a first read.\n\n    Pure :class:`torch.Tensor` objects are, in general, treated as images (or\n    as videos for video-specific transforms). Indeed, you may have noticed\n    that in the code above we haven't used the\n    :class:`~torchvision.tv_tensors.Image` class at all, and yet our images\n    got transformed properly. Transforms follow the following logic to\n    determine whether a pure Tensor should be treated as an image (or video),\n    or just ignored:\n\n    * If there is an :class:`~torchvision.tv_tensors.Image`,\n      :class:`~torchvision.tv_tensors.Video`,\n      or :class:`PIL.Image.Image` instance in the input, all other pure\n      tensors are passed-through.\n    * If there is no :class:`~torchvision.tv_tensors.Image` or\n      :class:`~torchvision.tv_tensors.Video` instance, only the first pure\n      :class:`torch.Tensor` will be transformed as image or video, while all\n      others will be passed-through. Here \"first\" means \"first in a depth-wise\n      traversal\".\n\n    This is what happened in the detection example above: the first pure\n    tensor was the image so it got transformed properly, and all other pure\n    tensor instances like the ``labels`` were passed-through (although labels\n    can still be transformed by some transforms like\n    :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes`!).</p></div>\n\n\n## Transforms and Datasets intercompatibility\n\nRoughly speaking, the output of the datasets must correspond to the input of\nthe transforms. How to do that depends on whether you're using the torchvision\n`built-in datatsets <datasets>`, or your own custom datasets.\n\n### Using built-in datasets\n\nIf you're just doing image classification, you don't need to do anything. Just\nuse ``transform`` argument of the dataset e.g. ``ImageNet(...,\ntransform=transforms)`` and you're good to go.\n\nTorchvision also supports datasets for object detection or segmentation like\n:class:`torchvision.datasets.CocoDetection`. Those datasets predate\nthe existence of the :mod:`torchvision.transforms.v2` module and of the\nTVTensors, so they don't return TVTensors out of the box.\n\nAn easy way to force those datasets to return TVTensors and to make them\ncompatible with v2 transforms is to use the\n:func:`torchvision.datasets.wrap_dataset_for_transforms_v2` function:\n\n```python\nfrom torchvision.datasets import CocoDetection, wrap_dataset_for_transforms_v2\n\ndataset = CocoDetection(..., transforms=my_transforms)\ndataset = wrap_dataset_for_transforms_v2(dataset)\n# Now the dataset returns TVTensors!\n```\n### Using your own datasets\n\nIf you have a custom dataset, then you'll need to convert your objects into\nthe appropriate TVTensor classes. Creating TVTensor instances is very easy,\nrefer to `tv_tensor_creation` for more details.\n\nThere are two main places where you can implement that conversion logic:\n\n- At the end of the datasets's ``__getitem__`` method, before returning the\n  sample (or by sub-classing the dataset).\n- As the very first step of your transforms pipeline\n\nEither way, the logic will depend on your specific dataset.\n\n"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb b/0.16/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb
new file mode 100644
index 00000000000..b885564fff7
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb
@@ -0,0 +1,511 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Illustration of transforms\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_illustrations.py>` to download the full example code.</p></div>\n\nThis example illustrates some of the various transforms available in `the\ntorchvision.transforms.v2 module <transforms>`.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from PIL import Image\nfrom pathlib import Path\nimport matplotlib.pyplot as plt\n\nimport torch\nfrom torchvision.transforms import v2\n\nplt.rcParams[\"savefig.bbox\"] = 'tight'\n\n# if you change the seed, make sure that the randomly-applied transforms\n# properly show that the image can be both transformed and *not* transformed!\ntorch.manual_seed(0)\n\n# If you're trying to run that on collab, you can download the assets and the\n# helpers from https://github.com/pytorch/vision/tree/main/gallery/\nfrom helpers import plot\norig_img = Image.open(Path('../assets') / 'astronaut.jpg')"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Geometric Transforms\nGeometric image transformation refers to the process of altering the geometric properties of an image,\nsuch as its shape, size, orientation, or position.\nIt involves applying mathematical operations to the image pixels or coordinates to achieve the desired transformation.\n\n### Pad\nThe :class:`~torchvision.transforms.Pad` transform\n(see also :func:`~torchvision.transforms.functional.pad`)\npads all image borders with some pixel values.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "padded_imgs = [v2.Pad(padding=padding)(orig_img) for padding in (3, 10, 30, 50)]\nplot([orig_img] + padded_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### Resize\nThe :class:`~torchvision.transforms.Resize` transform\n(see also :func:`~torchvision.transforms.functional.resize`)\nresizes an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "resized_imgs = [v2.Resize(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]\nplot([orig_img] + resized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### CenterCrop\nThe :class:`~torchvision.transforms.CenterCrop` transform\n(see also :func:`~torchvision.transforms.functional.center_crop`)\ncrops the given image at the center.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "center_crops = [v2.CenterCrop(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]\nplot([orig_img] + center_crops)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### FiveCrop\nThe :class:`~torchvision.transforms.FiveCrop` transform\n(see also :func:`~torchvision.transforms.functional.five_crop`)\ncrops the given image into four corners and the central crop.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "(top_left, top_right, bottom_left, bottom_right, center) = v2.FiveCrop(size=(100, 100))(orig_img)\nplot([orig_img] + [top_left, top_right, bottom_left, bottom_right, center])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomPerspective\nThe :class:`~torchvision.transforms.RandomPerspective` transform\n(see also :func:`~torchvision.transforms.functional.perspective`)\nperforms random perspective transform on an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "perspective_transformer = v2.RandomPerspective(distortion_scale=0.6, p=1.0)\nperspective_imgs = [perspective_transformer(orig_img) for _ in range(4)]\nplot([orig_img] + perspective_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomRotation\nThe :class:`~torchvision.transforms.RandomRotation` transform\n(see also :func:`~torchvision.transforms.functional.rotate`)\nrotates an image with random angle.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "rotater = v2.RandomRotation(degrees=(0, 180))\nrotated_imgs = [rotater(orig_img) for _ in range(4)]\nplot([orig_img] + rotated_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomAffine\nThe :class:`~torchvision.transforms.RandomAffine` transform\n(see also :func:`~torchvision.transforms.functional.affine`)\nperforms random affine transform on an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "affine_transfomer = v2.RandomAffine(degrees=(30, 70), translate=(0.1, 0.3), scale=(0.5, 0.75))\naffine_imgs = [affine_transfomer(orig_img) for _ in range(4)]\nplot([orig_img] + affine_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### ElasticTransform\nThe :class:`~torchvision.transforms.ElasticTransform` transform\n(see also :func:`~torchvision.transforms.functional.elastic_transform`)\nRandomly transforms the morphology of objects in images and produces a\nsee-through-water-like effect.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "elastic_transformer = v2.ElasticTransform(alpha=250.0)\ntransformed_imgs = [elastic_transformer(orig_img) for _ in range(2)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomCrop\nThe :class:`~torchvision.transforms.RandomCrop` transform\n(see also :func:`~torchvision.transforms.functional.crop`)\ncrops an image at a random location.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "cropper = v2.RandomCrop(size=(128, 128))\ncrops = [cropper(orig_img) for _ in range(4)]\nplot([orig_img] + crops)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomResizedCrop\nThe :class:`~torchvision.transforms.RandomResizedCrop` transform\n(see also :func:`~torchvision.transforms.functional.resized_crop`)\ncrops an image at a random location, and then resizes the crop to a given\nsize.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "resize_cropper = v2.RandomResizedCrop(size=(32, 32))\nresized_crops = [resize_cropper(orig_img) for _ in range(4)]\nplot([orig_img] + resized_crops)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Photometric Transforms\nPhotometric image transformation refers to the process of modifying the photometric properties of an image,\nsuch as its brightness, contrast, color, or tone.\nThese transformations are applied to change the visual appearance of an image\nwhile preserving its geometric structure.\n\nExcept :class:`~torchvision.transforms.Grayscale`, the following transforms are random,\nwhich means that the same transform\ninstance will produce different result each time it transforms a given image.\n\n### Grayscale\nThe :class:`~torchvision.transforms.Grayscale` transform\n(see also :func:`~torchvision.transforms.functional.to_grayscale`)\nconverts an image to grayscale\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "gray_img = v2.Grayscale()(orig_img)\nplot([orig_img, gray_img], cmap='gray')"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### ColorJitter\nThe :class:`~torchvision.transforms.ColorJitter` transform\nrandomly changes the brightness, contrast, saturation, hue, and other properties of an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "jitter = v2.ColorJitter(brightness=.5, hue=.3)\njittered_imgs = [jitter(orig_img) for _ in range(4)]\nplot([orig_img] + jittered_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### GaussianBlur\nThe :class:`~torchvision.transforms.GaussianBlur` transform\n(see also :func:`~torchvision.transforms.functional.gaussian_blur`)\nperforms gaussian blur transform on an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "blurrer = v2.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5.))\nblurred_imgs = [blurrer(orig_img) for _ in range(4)]\nplot([orig_img] + blurred_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomInvert\nThe :class:`~torchvision.transforms.RandomInvert` transform\n(see also :func:`~torchvision.transforms.functional.invert`)\nrandomly inverts the colors of the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "inverter = v2.RandomInvert()\ninvertered_imgs = [inverter(orig_img) for _ in range(4)]\nplot([orig_img] + invertered_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomPosterize\nThe :class:`~torchvision.transforms.RandomPosterize` transform\n(see also :func:`~torchvision.transforms.functional.posterize`)\nrandomly posterizes the image by reducing the number of bits\nof each color channel.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "posterizer = v2.RandomPosterize(bits=2)\nposterized_imgs = [posterizer(orig_img) for _ in range(4)]\nplot([orig_img] + posterized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomSolarize\nThe :class:`~torchvision.transforms.RandomSolarize` transform\n(see also :func:`~torchvision.transforms.functional.solarize`)\nrandomly solarizes the image by inverting all pixel values above\nthe threshold.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "solarizer = v2.RandomSolarize(threshold=192.0)\nsolarized_imgs = [solarizer(orig_img) for _ in range(4)]\nplot([orig_img] + solarized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomAdjustSharpness\nThe :class:`~torchvision.transforms.RandomAdjustSharpness` transform\n(see also :func:`~torchvision.transforms.functional.adjust_sharpness`)\nrandomly adjusts the sharpness of the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "sharpness_adjuster = v2.RandomAdjustSharpness(sharpness_factor=2)\nsharpened_imgs = [sharpness_adjuster(orig_img) for _ in range(4)]\nplot([orig_img] + sharpened_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomAutocontrast\nThe :class:`~torchvision.transforms.RandomAutocontrast` transform\n(see also :func:`~torchvision.transforms.functional.autocontrast`)\nrandomly applies autocontrast to the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "autocontraster = v2.RandomAutocontrast()\nautocontrasted_imgs = [autocontraster(orig_img) for _ in range(4)]\nplot([orig_img] + autocontrasted_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomEqualize\nThe :class:`~torchvision.transforms.RandomEqualize` transform\n(see also :func:`~torchvision.transforms.functional.equalize`)\nrandomly equalizes the histogram of the given image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "equalizer = v2.RandomEqualize()\nequalized_imgs = [equalizer(orig_img) for _ in range(4)]\nplot([orig_img] + equalized_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Augmentation Transforms\nThe following transforms are combinations of multiple transforms,\neither geometric or photometric, or both.\n\n### AutoAugment\nThe :class:`~torchvision.transforms.AutoAugment` transform\nautomatically augments data based on a given auto-augmentation policy.\nSee :class:`~torchvision.transforms.AutoAugmentPolicy` for the available policies.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "policies = [v2.AutoAugmentPolicy.CIFAR10, v2.AutoAugmentPolicy.IMAGENET, v2.AutoAugmentPolicy.SVHN]\naugmenters = [v2.AutoAugment(policy) for policy in policies]\nimgs = [\n    [augmenter(orig_img) for _ in range(4)]\n    for augmenter in augmenters\n]\nrow_title = [str(policy).split('.')[-1] for policy in policies]\nplot([[orig_img] + row for row in imgs], row_title=row_title)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandAugment\nThe :class:`~torchvision.transforms.RandAugment` is an alternate version of AutoAugment.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "augmenter = v2.RandAugment()\nimgs = [augmenter(orig_img) for _ in range(4)]\nplot([orig_img] + imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### TrivialAugmentWide\nThe :class:`~torchvision.transforms.TrivialAugmentWide` is an alternate implementation of AutoAugment.\nHowever, instead of transforming an image multiple times, it transforms an image only once\nusing a random transform from a given list with a random strength number.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "augmenter = v2.TrivialAugmentWide()\nimgs = [augmenter(orig_img) for _ in range(4)]\nplot([orig_img] + imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### AugMix\nThe :class:`~torchvision.transforms.AugMix` transform interpolates between augmented versions of an image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "augmenter = v2.AugMix()\nimgs = [augmenter(orig_img) for _ in range(4)]\nplot([orig_img] + imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Randomly-applied Transforms\n\nThe following transforms are randomly-applied given a probability ``p``.  That is, given ``p = 0.5``,\nthere is a 50% chance to return the original image, and a 50% chance to return the transformed image,\neven when called with the same transform instance!\n\n### RandomHorizontalFlip\nThe :class:`~torchvision.transforms.RandomHorizontalFlip` transform\n(see also :func:`~torchvision.transforms.functional.hflip`)\nperforms horizontal flip of an image, with a given probability.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "hflipper = v2.RandomHorizontalFlip(p=0.5)\ntransformed_imgs = [hflipper(orig_img) for _ in range(4)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomVerticalFlip\nThe :class:`~torchvision.transforms.RandomVerticalFlip` transform\n(see also :func:`~torchvision.transforms.functional.vflip`)\nperforms vertical flip of an image, with a given probability.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "vflipper = v2.RandomVerticalFlip(p=0.5)\ntransformed_imgs = [vflipper(orig_img) for _ in range(4)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### RandomApply\nThe :class:`~torchvision.transforms.RandomApply` transform\nrandomly applies a list of transforms, with a given probability.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "applier = v2.RandomApply(transforms=[v2.RandomCrop(size=(64, 64))], p=0.5)\ntransformed_imgs = [applier(orig_img) for _ in range(4)]\nplot([orig_img] + transformed_imgs)"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_tv_tensors.ipynb b/0.16/_generated_ipynb_notebooks/plot_tv_tensors.ipynb
new file mode 100644
index 00000000000..8a6bcd96b54
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_tv_tensors.ipynb
@@ -0,0 +1,237 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# TVTensors FAQ\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_tv_tensors.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_transforms_plot_tv_tensors.py>` to download the full example code.</p></div>\n\n\nTVTensors are Tensor subclasses introduced together with\n``torchvision.transforms.v2``. This example showcases what these TVTensors are\nand how they behave.\n\n<div class=\"alert alert-danger\"><h4>Warning</h4><p>**Intended Audience** Unless you're writing your own transforms or your own TVTensors, you\n    probably do not need to read this guide. This is a fairly low-level topic\n    that most users will not need to worry about: you do not need to understand\n    the internals of TVTensors to efficiently rely on\n    ``torchvision.transforms.v2``. It may however be useful for advanced users\n    trying to implement their own datasets, transforms, or work directly with\n    the TVTensors.</p></div>\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import PIL.Image\n\nimport torch\nfrom torchvision import tv_tensors"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## What are TVTensors?\n\nTVTensors are zero-copy tensor subclasses:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "tensor = torch.rand(3, 256, 256)\nimage = tv_tensors.Image(tensor)\n\nassert isinstance(image, torch.Tensor)\nassert image.data_ptr() == tensor.data_ptr()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Under the hood, they are needed in :mod:`torchvision.transforms.v2` to correctly dispatch to the appropriate function\nfor the input data.\n\n:mod:`torchvision.tv_tensors` supports four types of TVTensors:\n\n* :class:`~torchvision.tv_tensors.Image`\n* :class:`~torchvision.tv_tensors.Video`\n* :class:`~torchvision.tv_tensors.BoundingBoxes`\n* :class:`~torchvision.tv_tensors.Mask`\n\n## What can I do with a TVTensor?\n\nTVTensors look and feel just like regular tensors - they **are** tensors.\nEverything that is supported on a plain :class:`torch.Tensor` like ``.sum()`` or\nany ``torch.*`` operator will also work on TVTensors. See\n`tv_tensor_unwrapping_behaviour` for a few gotchas.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n## How do I construct a TVTensor?\n\n### Using the constructor\n\nEach TVTensor class takes any tensor-like data that can be turned into a :class:`~torch.Tensor`\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "image = tv_tensors.Image([[[[0, 1], [1, 0]]]])\nprint(image)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Similar to other PyTorch creations ops, the constructor also takes the ``dtype``, ``device``, and ``requires_grad``\nparameters.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "float_image = tv_tensors.Image([[[0, 1], [1, 0]]], dtype=torch.float32, requires_grad=True)\nprint(float_image)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "In addition, :class:`~torchvision.tv_tensors.Image` and :class:`~torchvision.tv_tensors.Mask` can also take a\n:class:`PIL.Image.Image` directly:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "image = tv_tensors.Image(PIL.Image.open(\"../assets/astronaut.jpg\"))\nprint(image.shape, image.dtype)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Some TVTensors require additional metadata to be passed in ordered to be constructed. For example,\n:class:`~torchvision.tv_tensors.BoundingBoxes` requires the coordinate format as well as the size of the\ncorresponding image (``canvas_size``) alongside the actual values. These\nmetadata are required to properly transform the bounding boxes.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "bboxes = tv_tensors.BoundingBoxes(\n    [[17, 16, 344, 495], [0, 10, 0, 10]],\n    format=tv_tensors.BoundingBoxFormat.XYXY,\n    canvas_size=image.shape[-2:]\n)\nprint(bboxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### Using ``tv_tensors.wrap()``\n\nYou can also use the :func:`~torchvision.tv_tensors.wrap` function to wrap a tensor object\ninto a TVTensor. This is useful when you already have an object of the\ndesired type, which typically happens when writing transforms: you just want\nto wrap the output like the input.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "new_bboxes = torch.tensor([0, 20, 30, 40])\nnew_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)\nassert isinstance(new_bboxes, tv_tensors.BoundingBoxes)\nassert new_bboxes.canvas_size == bboxes.canvas_size"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The metadata of ``new_bboxes`` is the same as ``bboxes``, but you could pass\nit as a parameter to override it.\n\n\n## I had a TVTensor but now I have a Tensor. Help!\n\nBy default, operations on :class:`~torchvision.tv_tensors.TVTensor` objects\nwill return a pure Tensor:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "assert isinstance(bboxes, tv_tensors.BoundingBoxes)\n\n# Shift bboxes by 3 pixels in both H and W\nnew_bboxes = bboxes + 3\n\nassert isinstance(new_bboxes, torch.Tensor)\nassert not isinstance(new_bboxes, tv_tensors.BoundingBoxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "<div class=\"alert alert-info\"><h4>Note</h4><p>This behavior only affects native ``torch`` operations. If you are using\n   the built-in ``torchvision`` transforms or functionals, you will always get\n   as output the same type that you passed as input (pure ``Tensor`` or\n   ``TVTensor``).</p></div>\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### But I want a TVTensor back!\n\nYou can re-wrap a pure tensor into a TVTensor by just calling the TVTensor\nconstructor, or by using the :func:`~torchvision.tv_tensors.wrap` function\n(see more details above in `tv_tensor_creation`):\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "new_bboxes = bboxes + 3\nnew_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)\nassert isinstance(new_bboxes, tv_tensors.BoundingBoxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Alternatively, you can use the :func:`~torchvision.tv_tensors.set_return_type`\nas a global config setting for the whole program, or as a context manager\n(read its docs to learn more about caveats):\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "with tv_tensors.set_return_type(\"TVTensor\"):\n    new_bboxes = bboxes + 3\nassert isinstance(new_bboxes, tv_tensors.BoundingBoxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "### Why is this happening?\n\n**For performance reasons**. :class:`~torchvision.tv_tensors.TVTensor`\nclasses are Tensor subclasses, so any operation involving a\n:class:`~torchvision.tv_tensors.TVTensor` object will go through the\n[__torch_function__](https://pytorch.org/docs/stable/notes/extending.html#extending-torch)\nprotocol. This induces a small overhead, which we want to avoid when possible.\nThis doesn't matter for built-in ``torchvision`` transforms because we can\navoid the overhead there, but it could be a problem in your model's\n``forward``.\n\n**The alternative isn't much better anyway.** For every operation where\npreserving the :class:`~torchvision.tv_tensors.TVTensor` type makes\nsense, there are just as many operations where returning a pure Tensor is\npreferable: for example, is ``img.sum()`` still an :class:`~torchvision.tv_tensors.Image`?\nIf we were to preserve :class:`~torchvision.tv_tensors.TVTensor` types all\nthe way, even model's logits or the output of the loss function would end up\nbeing of type :class:`~torchvision.tv_tensors.Image`, and surely that's not\ndesirable.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>This behaviour is something we're actively seeking feedback on. If you find this surprising or if you\n   have any suggestions on how to better support your use-cases, please reach out to us via this issue:\n   https://github.com/pytorch/vision/issues/7319</p></div>\n\n### Exceptions\n\nThere are a few exceptions to this \"unwrapping\" rule:\n:meth:`~torch.Tensor.clone`, :meth:`~torch.Tensor.to`,\n:meth:`torch.Tensor.detach`, and :meth:`~torch.Tensor.requires_grad_` retain\nthe TVTensor type.\n\nInplace operations on TVTensors like ``obj.add_()`` will preserve the type of\n``obj``. However, the **returned** value of inplace operations will be a pure\ntensor:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "image = tv_tensors.Image([[[0, 1], [1, 0]]])\n\nnew_image = image.add_(1).mul_(2)\n\n# image got transformed in-place and is still a TVTensor Image, but new_image\n# is a Tensor. They share the same underlying data and they're equal, just\n# different classes.\nassert isinstance(image, tv_tensors.Image)\nprint(image)\n\nassert isinstance(new_image, torch.Tensor) and not isinstance(new_image, tv_tensors.Image)\nassert (new_image == image).all()\nassert new_image.data_ptr() == image.data_ptr()"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_video_api.ipynb b/0.16/_generated_ipynb_notebooks/plot_video_api.ipynb
new file mode 100644
index 00000000000..ffba78fd53f
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_video_api.ipynb
@@ -0,0 +1,298 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Video API\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_video_api.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_video_api.py>` to download the full example code.</p></div>\n\nThis example illustrates some of the APIs that torchvision offers for\nvideos, together with the examples on how to build datasets and more.\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 1. Introduction: building a new video object and examining the properties\nFirst we select a video to test the object out. For the sake of argument\nwe're using one from kinetics400 dataset.\nTo create it, we need to define the path and the stream we want to use.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Chosen video statistics:\n\n- WUzgd7C1pWA.mp4\n    - source:\n        - kinetics-400\n    - video:\n        - H-264\n        - MPEG-4 AVC (part 10) (avc1)\n        - fps: 29.97\n    - audio:\n        - MPEG AAC audio (mp4a)\n        - sample rate: 48K Hz\n\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nimport torchvision\nfrom torchvision.datasets.utils import download_url\ntorchvision.set_video_backend(\"video_reader\")\n\n# Download the sample video\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true\",\n    \".\",\n    \"WUzgd7C1pWA.mp4\"\n)\nvideo_path = \"./WUzgd7C1pWA.mp4\""
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Streams are defined in a similar fashion as torch devices. We encode them as strings in a form\nof ``stream_type:stream_id`` where ``stream_type`` is a string and ``stream_id`` a long int.\nThe constructor accepts passing a ``stream_type`` only, in which case the stream is auto-discovered.\nFirstly, let's get the metadata for our particular video:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "stream = \"video\"\nvideo = torchvision.io.VideoReader(video_path, stream)\nvideo.get_metadata()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Here we can see that video has two streams - a video and an audio stream.\nCurrently available stream types include ['video', 'audio'].\nEach descriptor consists of two parts: stream type (e.g. 'video') and a unique stream id\n(which are determined by video encoding).\nIn this way, if the video container contains multiple streams of the same type,\nusers can access the one they want.\nIf only stream type is passed, the decoder auto-detects first stream of that type and returns it.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let's read all the frames from the video stream. By default, the return value of\n``next(video_reader)`` is a dict containing the following fields.\n\nThe return fields are:\n\n- ``data``: containing a torch.tensor\n- ``pts``: containing a float timestamp of this particular frame\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "metadata = video.get_metadata()\nvideo.set_current_stream(\"audio\")\n\nframes = []  # we are going to save the frames here.\nptss = []  # pts is a presentation timestamp in seconds (float) of each frame\nfor frame in video:\n    frames.append(frame['data'])\n    ptss.append(frame['pts'])\n\nprint(\"PTS for first five frames \", ptss[:5])\nprint(\"Total number of frames: \", len(frames))\napprox_nf = metadata['audio']['duration'][0] * metadata['audio']['framerate'][0]\nprint(\"Approx total number of datapoints we can expect: \", approx_nf)\nprint(\"Read data size: \", frames[0].size(0) * len(frames))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "But what if we only want to read certain time segment of the video?\nThat can be done easily using the combination of our ``seek`` function, and the fact that each call\nto next returns the presentation timestamp of the returned frame in seconds.\n\nGiven that our implementation relies on python iterators,\nwe can leverage itertools to simplify the process and make it more pythonic.\n\nFor example, if we wanted to read ten frames from second second:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import itertools\nvideo.set_current_stream(\"video\")\n\nframes = []  # we are going to save the frames here.\n\n# We seek into a second second of the video and use islice to get 10 frames since\nfor frame, pts in itertools.islice(video.seek(2), 10):\n    frames.append(frame)\n\nprint(\"Total number of frames: \", len(frames))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Or if we wanted to read from 2nd to 5th second,\nWe seek into a second second of the video,\nthen we utilize the itertools takewhile to get the\ncorrect number of frames:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "video.set_current_stream(\"video\")\nframes = []  # we are going to save the frames here.\nvideo = video.seek(2)\n\nfor frame in itertools.takewhile(lambda x: x['pts'] <= 5, video):\n    frames.append(frame['data'])\n\nprint(\"Total number of frames: \", len(frames))\napprox_nf = (5 - 2) * video.get_metadata()['video']['fps'][0]\nprint(\"We can expect approx: \", approx_nf)\nprint(\"Tensor size: \", frames[0].size())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 2. Building a sample read_video function\nWe can utilize the methods above to build the read video function that follows\nthe same API to the existing ``read_video`` function.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "def example_read_video(video_object, start=0, end=None, read_video=True, read_audio=True):\n    if end is None:\n        end = float(\"inf\")\n    if end < start:\n        raise ValueError(\n            \"end time should be larger than start time, got \"\n            f\"start time={start} and end time={end}\"\n        )\n\n    video_frames = torch.empty(0)\n    video_pts = []\n    if read_video:\n        video_object.set_current_stream(\"video\")\n        frames = []\n        for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):\n            frames.append(frame['data'])\n            video_pts.append(frame['pts'])\n        if len(frames) > 0:\n            video_frames = torch.stack(frames, 0)\n\n    audio_frames = torch.empty(0)\n    audio_pts = []\n    if read_audio:\n        video_object.set_current_stream(\"audio\")\n        frames = []\n        for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):\n            frames.append(frame['data'])\n            audio_pts.append(frame['pts'])\n        if len(frames) > 0:\n            audio_frames = torch.cat(frames, 0)\n\n    return video_frames, audio_frames, (video_pts, audio_pts), video_object.get_metadata()\n\n\n# Total number of frames should be 327 for video and 523264 datapoints for audio\nvf, af, info, meta = example_read_video(video)\nprint(vf.size(), af.size())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 3. Building an example randomly sampled dataset (can be applied to training dataset of kinetics400)\nCool, so now we can use the same principle to make the sample dataset.\nWe suggest trying out iterable dataset for this purpose.\nHere, we are going to build an example dataset that reads randomly selected 10 frames of video.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Make sample dataset\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nos.makedirs(\"./dataset\", exist_ok=True)\nos.makedirs(\"./dataset/1\", exist_ok=True)\nos.makedirs(\"./dataset/2\", exist_ok=True)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Download the videos\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.datasets.utils import download_url\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true\",\n    \"./dataset/1\", \"WUzgd7C1pWA.mp4\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi?raw=true\",\n    \"./dataset/1\",\n    \"RATRACE_wave_f_nm_np1_fr_goo_37.avi\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/SOX5yA1l24A.mp4?raw=true\",\n    \"./dataset/2\",\n    \"SOX5yA1l24A.mp4\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi?raw=true\",\n    \"./dataset/2\",\n    \"v_SoccerJuggling_g23_c01.avi\"\n)\ndownload_url(\n    \"https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi?raw=true\",\n    \"./dataset/2\",\n    \"v_SoccerJuggling_g24_c01.avi\"\n)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Housekeeping and utilities\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nimport random\n\nfrom torchvision.datasets.folder import make_dataset\nfrom torchvision import transforms as t\n\n\ndef _find_classes(dir):\n    classes = [d.name for d in os.scandir(dir) if d.is_dir()]\n    classes.sort()\n    class_to_idx = {cls_name: i for i, cls_name in enumerate(classes)}\n    return classes, class_to_idx\n\n\ndef get_samples(root, extensions=(\".mp4\", \".avi\")):\n    _, class_to_idx = _find_classes(root)\n    return make_dataset(root, class_to_idx, extensions=extensions)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We are going to define the dataset and some basic arguments.\nWe assume the structure of the FolderDataset, and add the following parameters:\n\n- ``clip_len``: length of a clip in frames\n- ``frame_transform``: transform for every frame individually\n- ``video_transform``: transform on a video sequence\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>We actually add epoch size as using :func:`~torch.utils.data.IterableDataset`\n  class allows us to naturally oversample clips or images from each video if needed.</p></div>\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class RandomDataset(torch.utils.data.IterableDataset):\n    def __init__(self, root, epoch_size=None, frame_transform=None, video_transform=None, clip_len=16):\n        super(RandomDataset).__init__()\n\n        self.samples = get_samples(root)\n\n        # Allow for temporal jittering\n        if epoch_size is None:\n            epoch_size = len(self.samples)\n        self.epoch_size = epoch_size\n\n        self.clip_len = clip_len\n        self.frame_transform = frame_transform\n        self.video_transform = video_transform\n\n    def __iter__(self):\n        for i in range(self.epoch_size):\n            # Get random sample\n            path, target = random.choice(self.samples)\n            # Get video object\n            vid = torchvision.io.VideoReader(path, \"video\")\n            metadata = vid.get_metadata()\n            video_frames = []  # video frame buffer\n\n            # Seek and return frames\n            max_seek = metadata[\"video\"]['duration'][0] - (self.clip_len / metadata[\"video\"]['fps'][0])\n            start = random.uniform(0., max_seek)\n            for frame in itertools.islice(vid.seek(start), self.clip_len):\n                video_frames.append(self.frame_transform(frame['data']))\n                current_pts = frame['pts']\n            # Stack it into a tensor\n            video = torch.stack(video_frames, 0)\n            if self.video_transform:\n                video = self.video_transform(video)\n            output = {\n                'path': path,\n                'video': video,\n                'target': target,\n                'start': start,\n                'end': current_pts}\n            yield output"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Given a path of videos in a folder structure, i.e:\n\n- dataset\n    - class 1\n        - file 0\n        - file 1\n        - ...\n    - class 2\n        - file 0\n        - file 1\n        - ...\n    - ...\n\nWe can generate a dataloader and test the dataset.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "transforms = [t.Resize((112, 112))]\nframe_transform = t.Compose(transforms)\n\ndataset = RandomDataset(\"./dataset\", epoch_size=None, frame_transform=frame_transform)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torch.utils.data import DataLoader\nloader = DataLoader(dataset, batch_size=12)\ndata = {\"video\": [], 'start': [], 'end': [], 'tensorsize': []}\nfor batch in loader:\n    for i in range(len(batch['path'])):\n        data['video'].append(batch['path'][i])\n        data['start'].append(batch['start'][i].item())\n        data['end'].append(batch['end'][i].item())\n        data['tensorsize'].append(batch['video'][i].size())\nprint(data)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 4. Data Visualization\nExample of visualized video\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import matplotlib.pyplot as plt\n\nplt.figure(figsize=(12, 12))\nfor i in range(16):\n    plt.subplot(4, 4, i + 1)\n    plt.imshow(batch[\"video\"][0, i, ...].permute(1, 2, 0))\n    plt.axis(\"off\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Cleanup the video and dataset:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import os\nimport shutil\nos.remove(\"./WUzgd7C1pWA.mp4\")\nshutil.rmtree(\"./dataset\")"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_generated_ipynb_notebooks/plot_visualization_utils.ipynb b/0.16/_generated_ipynb_notebooks/plot_visualization_utils.ipynb
new file mode 100644
index 00000000000..f8d08a978a0
--- /dev/null
+++ b/0.16/_generated_ipynb_notebooks/plot_visualization_utils.ipynb
@@ -0,0 +1,464 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n# Visualization utilities\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>Try on [collab](https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_visualization_utils.ipynb)\n    or `go to the end <sphx_glr_download_auto_examples_others_plot_visualization_utils.py>` to download the full example code.</p></div>\n\nThis example illustrates some of the utilities that torchvision offers for\nvisualizing images, bounding boxes, segmentation masks and keypoints.\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "import torch\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nimport torchvision.transforms.functional as F\n\n\nplt.rcParams[\"savefig.bbox\"] = 'tight'\n\n\ndef show(imgs):\n    if not isinstance(imgs, list):\n        imgs = [imgs]\n    fig, axs = plt.subplots(ncols=len(imgs), squeeze=False)\n    for i, img in enumerate(imgs):\n        img = img.detach()\n        img = F.to_pil_image(img)\n        axs[0, i].imshow(np.asarray(img))\n        axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing a grid of images\nThe :func:`~torchvision.utils.make_grid` function can be used to create a\ntensor that represents multiple images in a grid.  This util requires a single\nimage of dtype ``uint8`` as input.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import make_grid\nfrom torchvision.io import read_image\nfrom pathlib import Path\n\ndog1_int = read_image(str(Path('../assets') / 'dog1.jpg'))\ndog2_int = read_image(str(Path('../assets') / 'dog2.jpg'))\ndog_list = [dog1_int, dog2_int]\n\ngrid = make_grid(dog_list)\nshow(grid)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing bounding boxes\nWe can use :func:`~torchvision.utils.draw_bounding_boxes` to draw boxes on an\nimage. We can set the colors, labels, width as well as font and font size.\nThe boxes are in ``(xmin, ymin, xmax, ymax)`` format.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_bounding_boxes\n\n\nboxes = torch.tensor([[50, 50, 100, 200], [210, 150, 350, 430]], dtype=torch.float)\ncolors = [\"blue\", \"yellow\"]\nresult = draw_bounding_boxes(dog1_int, boxes, colors=colors, width=5)\nshow(result)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Naturally, we can also plot bounding boxes produced by torchvision detection\nmodels.  Here is a demo with a Faster R-CNN model loaded from\n:func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`\nmodel. For more details on the output of such models, you may\nrefer to `instance_seg_output`.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights\n\n\nweights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT\ntransforms = weights.transforms()\n\nimages = [transforms(d) for d in dog_list]\n\nmodel = fasterrcnn_resnet50_fpn(weights=weights, progress=False)\nmodel = model.eval()\n\noutputs = model(images)\nprint(outputs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let's plot the boxes detected by our model. We will only plot the boxes with a\nscore greater than a given threshold.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "score_threshold = .8\ndogs_with_boxes = [\n    draw_bounding_boxes(dog_int, boxes=output['boxes'][output['scores'] > score_threshold], width=4)\n    for dog_int, output in zip(dog_list, outputs)\n]\nshow(dogs_with_boxes)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Visualizing segmentation masks\nThe :func:`~torchvision.utils.draw_segmentation_masks` function can be used to\ndraw segmentation masks on images. Semantic segmentation and instance\nsegmentation models have different outputs, so we will treat each\nindependently.\n\n\n### Semantic segmentation models\n\nWe will see how to use it with torchvision's FCN Resnet-50, loaded with\n:func:`~torchvision.models.segmentation.fcn_resnet50`. Let's start by looking\nat the output of the model.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.segmentation import fcn_resnet50, FCN_ResNet50_Weights\n\nweights = FCN_ResNet50_Weights.DEFAULT\ntransforms = weights.transforms(resize_size=None)\n\nmodel = fcn_resnet50(weights=weights, progress=False)\nmodel = model.eval()\n\nbatch = torch.stack([transforms(d) for d in dog_list])\noutput = model(batch)['out']\nprint(output.shape, output.min().item(), output.max().item())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As we can see above, the output of the segmentation model is a tensor of shape\n``(batch_size, num_classes, H, W)``. Each value is a non-normalized score, and\nwe can normalize them into ``[0, 1]`` by using a softmax. After the softmax,\nwe can interpret each value as a probability indicating how likely a given\npixel is to belong to a given class.\n\nLet's plot the masks that have been detected for the dog class and for the\nboat class:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "sem_class_to_idx = {cls: idx for (idx, cls) in enumerate(weights.meta[\"categories\"])}\n\nnormalized_masks = torch.nn.functional.softmax(output, dim=1)\n\ndog_and_boat_masks = [\n    normalized_masks[img_idx, sem_class_to_idx[cls]]\n    for img_idx in range(len(dog_list))\n    for cls in ('dog', 'boat')\n]\n\nshow(dog_and_boat_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As expected, the model is confident about the dog class, but not so much for\nthe boat class.\n\nThe :func:`~torchvision.utils.draw_segmentation_masks` function can be used to\nplots those masks on top of the original image. This function expects the\nmasks to be boolean masks, but our masks above contain probabilities in ``[0,\n1]``. To get boolean masks, we can do the following:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class_dim = 1\nboolean_dog_masks = (normalized_masks.argmax(class_dim) == sem_class_to_idx['dog'])\nprint(f\"shape = {boolean_dog_masks.shape}, dtype = {boolean_dog_masks.dtype}\")\nshow([m.float() for m in boolean_dog_masks])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The line above where we define ``boolean_dog_masks`` is a bit cryptic, but you\ncan read it as the following query: \"For which pixels is 'dog' the most likely\nclass?\"\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>While we're using the ``normalized_masks`` here, we would have\n  gotten the same result by using the non-normalized scores of the model\n  directly (as the softmax operation preserves the order).</p></div>\n\nNow that we have boolean masks, we can use them with\n:func:`~torchvision.utils.draw_segmentation_masks` to plot them on top of the\noriginal images:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_segmentation_masks\n\ndogs_with_masks = [\n    draw_segmentation_masks(img, masks=mask, alpha=0.7)\n    for img, mask in zip(dog_list, boolean_dog_masks)\n]\nshow(dogs_with_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can plot more than one mask per image! Remember that the model returned as\nmany masks as there are classes. Let's ask the same query as above, but this\ntime for *all* classes, not just the dog class: \"For each pixel and each class\nC, is class C the most likely class?\"\n\nThis one is a bit more involved, so we'll first show how to do it with a\nsingle image, and then we'll generalize to the batch\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "num_classes = normalized_masks.shape[1]\ndog1_masks = normalized_masks[0]\nclass_dim = 0\ndog1_all_classes_masks = dog1_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None]\n\nprint(f\"dog1_masks shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}\")\nprint(f\"dog1_all_classes_masks = {dog1_all_classes_masks.shape}, dtype = {dog1_all_classes_masks.dtype}\")\n\ndog_with_all_masks = draw_segmentation_masks(dog1_int, masks=dog1_all_classes_masks, alpha=.6)\nshow(dog_with_all_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We can see in the image above that only 2 masks were drawn: the mask for the\nbackground and the mask for the dog. This is because the model thinks that\nonly these 2 classes are the most likely ones across all the pixels. If the\nmodel had detected another class as the most likely among other pixels, we\nwould have seen its mask above.\n\nRemoving the background mask is as simple as passing\n``masks=dog1_all_classes_masks[1:]``, because the background class is the\nclass with index 0.\n\nLet's now do the same but for an entire batch of images. The code is similar\nbut involves a bit more juggling with the dimensions.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "class_dim = 1\nall_classes_masks = normalized_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None, None]\nprint(f\"shape = {all_classes_masks.shape}, dtype = {all_classes_masks.dtype}\")\n# The first dimension is the classes now, so we need to swap it\nall_classes_masks = all_classes_masks.swapaxes(0, 1)\n\ndogs_with_masks = [\n    draw_segmentation_masks(img, masks=mask, alpha=.6)\n    for img, mask in zip(dog_list, all_classes_masks)\n]\nshow(dogs_with_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n### Instance segmentation models\n\nInstance segmentation models have a significantly different output from the\nsemantic segmentation models. We will see here how to plot the masks for such\nmodels. Let's start by analyzing the output of a Mask-RCNN model. Note that\nthese models don't require the images to be normalized, so we don't need to\nuse the normalized batch.\n\n<div class=\"alert alert-info\"><h4>Note</h4><p>We will here describe the output of a Mask-RCNN model. The models in\n    `object_det_inst_seg_pers_keypoint_det` all have a similar output\n    format, but some of them may have extra info like keypoints for\n    :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`, and some\n    of them may not have masks, like\n    :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`.</p></div>\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import maskrcnn_resnet50_fpn, MaskRCNN_ResNet50_FPN_Weights\n\nweights = MaskRCNN_ResNet50_FPN_Weights.DEFAULT\ntransforms = weights.transforms()\n\nimages = [transforms(d) for d in dog_list]\n\nmodel = maskrcnn_resnet50_fpn(weights=weights, progress=False)\nmodel = model.eval()\n\noutput = model(images)\nprint(output)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Let's break this down. For each image in the batch, the model outputs some\ndetections (or instances). The number of detections varies for each input\nimage. Each instance is described by its bounding box, its label, its score\nand its mask.\n\nThe way the output is organized is as follows: the output is a list of length\n``batch_size``. Each entry in the list corresponds to an input image, and it\nis a dict with keys 'boxes', 'labels', 'scores', and 'masks'. Each value\nassociated to those keys has ``num_instances`` elements in it.  In our case\nabove there are 3 instances detected in the first image, and 2 instances in\nthe second one.\n\nThe boxes can be plotted with :func:`~torchvision.utils.draw_bounding_boxes`\nas above, but here we're more interested in the masks. These masks are quite\ndifferent from the masks that we saw above for the semantic segmentation\nmodels.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "dog1_output = output[0]\ndog1_masks = dog1_output['masks']\nprint(f\"shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}, \"\n      f\"min = {dog1_masks.min()}, max = {dog1_masks.max()}\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Here the masks correspond to probabilities indicating, for each pixel, how\nlikely it is to belong to the predicted label of that instance. Those\npredicted labels correspond to the 'labels' element in the same output dict.\nLet's see which labels were predicted for the instances of the first image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(\"For the first dog, the following instances were detected:\")\nprint([weights.meta[\"categories\"][label] for label in dog1_output['labels']])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Interestingly, the model detects two persons in the image. Let's go ahead and\nplot those masks. Since :func:`~torchvision.utils.draw_segmentation_masks`\nexpects boolean masks, we need to convert those probabilities into boolean\nvalues. Remember that the semantic of those masks is \"How likely is this pixel\nto belong to the predicted class?\". As a result, a natural way of converting\nthose masks into boolean values is to threshold them with the 0.5 probability\n(one could also choose a different threshold).\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "proba_threshold = 0.5\ndog1_bool_masks = dog1_output['masks'] > proba_threshold\nprint(f\"shape = {dog1_bool_masks.shape}, dtype = {dog1_bool_masks.dtype}\")\n\n# There's an extra dimension (1) to the masks. We need to remove it\ndog1_bool_masks = dog1_bool_masks.squeeze(1)\n\nshow(draw_segmentation_masks(dog1_int, dog1_bool_masks, alpha=0.9))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The model seems to have properly detected the dog, but it also confused trees\nwith people. Looking more closely at the scores will help us plot more\nrelevant masks:\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "print(dog1_output['scores'])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Clearly the model is more confident about the dog detection than it is about\nthe people detections. That's good news. When plotting the masks, we can ask\nfor only those that have a good score. Let's use a score threshold of .75\nhere, and also plot the masks of the second dog.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "score_threshold = .75\n\nboolean_masks = [\n    out['masks'][out['scores'] > score_threshold] > proba_threshold\n    for out in output\n]\n\ndogs_with_masks = [\n    draw_segmentation_masks(img, mask.squeeze(1))\n    for img, mask in zip(dog_list, boolean_masks)\n]\nshow(dogs_with_masks)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The two 'people' masks in the first image where not selected because they have\na lower score than the score threshold. Similarly, in the second image, the\ninstance with class 15 (which corresponds to 'bench') was not selected.\n\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "\n## Visualizing keypoints\nThe :func:`~torchvision.utils.draw_keypoints` function can be used to\ndraw keypoints on images. We will see how to use it with\ntorchvision's KeypointRCNN loaded with :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`.\nWe will first have a look at output of the model.\n\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.models.detection import keypointrcnn_resnet50_fpn, KeypointRCNN_ResNet50_FPN_Weights\nfrom torchvision.io import read_image\n\nperson_int = read_image(str(Path(\"../assets\") / \"person1.jpg\"))\n\nweights = KeypointRCNN_ResNet50_FPN_Weights.DEFAULT\ntransforms = weights.transforms()\n\nperson_float = transforms(person_int)\n\nmodel = keypointrcnn_resnet50_fpn(weights=weights, progress=False)\nmodel = model.eval()\n\noutputs = model([person_float])\nprint(outputs)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As we see the output contains a list of dictionaries.\nThe output list is of length batch_size.\nWe currently have just a single image so length of list is 1.\nEach entry in the list corresponds to an input image,\nand it is a dict with keys `boxes`, `labels`, `scores`, `keypoints` and `keypoint_scores`.\nEach value associated to those keys has `num_instances` elements in it.\nIn our case above there are 2 instances detected in the image.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "kpts = outputs[0]['keypoints']\nscores = outputs[0]['scores']\n\nprint(kpts)\nprint(scores)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "The KeypointRCNN model detects there are two instances in the image.\nIf you plot the boxes by using :func:`~draw_bounding_boxes`\nyou would recognize they are the person and the surfboard.\nIf we look at the scores, we will realize that the model is much more confident about the person than surfboard.\nWe could now set a threshold confidence and plot instances which we are confident enough.\nLet us set a threshold of 0.75 and filter out the keypoints corresponding to the person.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "detect_threshold = 0.75\nidx = torch.where(scores > detect_threshold)\nkeypoints = kpts[idx]\n\nprint(keypoints)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Great, now we have the keypoints corresponding to the person.\nEach keypoint is represented by x, y coordinates and the visibility.\nWe can now use the :func:`~torchvision.utils.draw_keypoints` function to draw keypoints.\nNote that the utility expects uint8 images.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "from torchvision.utils import draw_keypoints\n\nres = draw_keypoints(person_int, keypoints, colors=\"blue\", radius=3)\nshow(res)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "As we see the keypoints appear as colored circles over the image.\nThe coco keypoints for a person are ordered and represent the following list.\\\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "coco_keypoints = [\n    \"nose\", \"left_eye\", \"right_eye\", \"left_ear\", \"right_ear\",\n    \"left_shoulder\", \"right_shoulder\", \"left_elbow\", \"right_elbow\",\n    \"left_wrist\", \"right_wrist\", \"left_hip\", \"right_hip\",\n    \"left_knee\", \"right_knee\", \"left_ankle\", \"right_ankle\",\n]"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "What if we are interested in joining the keypoints?\nThis is especially useful in creating pose detection or action recognition.\nWe can join the keypoints easily using the `connectivity` parameter.\nA close observation would reveal that we would need to join the points in below\norder to construct human skeleton.\n\nnose -> left_eye -> left_ear.                              (0, 1), (1, 3)\n\nnose -> right_eye -> right_ear.                            (0, 2), (2, 4)\n\nnose -> left_shoulder -> left_elbow -> left_wrist.         (0, 5), (5, 7), (7, 9)\n\nnose -> right_shoulder -> right_elbow -> right_wrist.      (0, 6), (6, 8), (8, 10)\n\nleft_shoulder -> left_hip -> left_knee -> left_ankle.      (5, 11), (11, 13), (13, 15)\n\nright_shoulder -> right_hip -> right_knee -> right_ankle.  (6, 12), (12, 14), (14, 16)\n\nWe will create a list containing these keypoint ids to be connected.\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "connect_skeleton = [\n    (0, 1), (0, 2), (1, 3), (2, 4), (0, 5), (0, 6), (5, 7), (6, 8),\n    (7, 9), (8, 10), (5, 11), (6, 12), (11, 13), (12, 14), (13, 15), (14, 16)\n]"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "We pass the above list to the connectivity parameter to connect the keypoints.\n\n\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": false
+      },
+      "outputs": [],
+      "source": [
+        "res = draw_keypoints(person_int, keypoints, connectivity=connect_skeleton, colors=\"blue\", radius=4, width=3)\nshow(res)"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.8.18"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/0.16/_images/sphx_glr_plot_custom_transforms_thumb.png b/0.16/_images/sphx_glr_plot_custom_transforms_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..8a5fed589d17fe6ee62d6a9788519dba471fbd65
GIT binary patch
literal 26794
zcmd3N^;cV6&~<SuCAeFl(4s|(J3)&T2rk8;P~2Tha45yKc!RqHEp7#hLvXj?ZZFTf
z*8BYp-w#<U>z<r@Gv~~mJ+tTR2sITsTx<&LSFc{-D#(M>U%f(#fBBGMqQ6{mHpumQ
z^~%~w0VJvEnR)nlooKF`yb|H}xA|C<^cUE&K&!B{#ar*%RrUI{fOnuSCL$03!--#|
zXbLuXBIIR=G00EgrNBb5!{Wg0L?-(}Cu?Haf+0F7%3dlH!I7q$mL$Yecj9<`jSzfT
z^D5Zgt0)_Z#Ecth&8aB^q#^tt+jp*g;+l3Kuta%Af{32Sylxr^V1hg&jmnR}mkXfv
zAbI2;{|#XPvE=`E0P<S?^9AVTE1Z82S>eA)C<&;HFH;EcKm_ss4Tt~~BwnThPbC<U
z{y+Qy-yssCdl=>g5(+AS2SjXlQXBFG6FYU#Fiz_d=HGDi{5xv>!;RVHaotlES;<Vs
z54)3=h(~4As6dmO)fSx8D!+i$VC>Y{+4GrXjSnu(wiwX$)Rm{msq2XKT(PDE-xEgi
z_(VivIE9l($F=w;5%X502cyd9S;o)?@*@IUt%-#RKi)is3GxWf#e9B1a0p}-K^GkH
zE#RcLH-hEomo}wtS-`xST`geg^-vL1i@Xk;hL(*ihF<qw7%c*?Q+E}k)1pi-WN4&J
zrc9QIvTOJ=TWorN19X#w9YvE^0xfTy!~GL#58!=LRoJfOBeNadJQGEW^>VL8OfGro
z7;=ZwLXX%~LDV&-I)%f^Zc@iBUPr)DOdL&SY8Fe4jRr0M^?OY#0}B$nzJ&L1In9Al
zQOK9T4G@xxBY?1&=&}M<$uJ+xfTKBW>PRAN(VR;B>we1G_F=pv`)g>`Z^uYRx_jEF
z<Ml1t4a&-D$31j+QCSBStgUGDG#OX))Ly@#cs>LgunT}hWZL=J)o}-tSc-H43?kvu
z2VGc#wT9ESOGB&rE@vy9Zz3|&#<T%q7(yR6b>)3Ae*}ua?Sht~cbom|GLpPlTSm&X
z8g90~9$!qLFeLh!xc`P0$(R$Ja&h&6EezoO__ACMkpyr<B+{;|AT;h98iu@OkGbHl
zGYQkX4NM%Rey;N@d@DM~2kNR!mU_AYWtf#)e}ui_>*-p^zc6Sxuga|EOJ~>i$53|i
zk3uPR?4WZH-H$~ya%J$=ipQkS-owRa4tw(P{%YPDc|taynH>AT=I|^o0}49wU~C{J
zauK2=H85m@3Fzliq*hcd5LX*{uUE#$CpBIVU20AE#`cp-cs73fa2V=^W|F4HJO04v
zh?+`+`1w}s-C)?|2Q&Bi)%tp}=My{T_L}AL?X!5%*(mp1+HX=ICa0VpGHL53#n?|w
zrsVN!1h2aM0663#mGa+wr{K=E&1QL;2u+|ml4!Vcv|H~i8eectf>jNDUkq^angTdW
z)#m!O%sOuC%7f^!Qht!qCw#tfe81zkyyL#c-mQq;u_=hU8Y&X0P3ZX%SuC=j%cBp_
zr)IOhDL^juo$ys}!om$%Q`tjo6J*I-hLpYq;KY(EwP$7&><+R6Cs-uf#l&-#?&}Uc
z2}fsB6#K32O!O58@ksbMFteP}G2(2lnU(Xx0EQ{|6Fyl`z1ErflfT{pGkLUjTO->;
z)k=c&e_nqG{-=f;xxac&YcIobydWM*>-WCPDRj}<`M?J$>U-1W=e@wPjOqn`WB6h3
zK~jXSxO+anw%MV<(bYq5#QL@?&$9}E9mOt!yYPhfLyOPk<($)-DALZ(Z<${Ys2e!I
zEf@cqbht`&4<v}sP#}Lx4n!!n0Yy~ga^ziw<gSkSH*`h*QC-LW26MvjQwQ^Vr$sI=
zO3>b(^c=S@?tyU^-%7nsg>47yyUusdaMN|<O`K4>9?WoX_ry|GMNc@rE*&1ZWvhRi
z_S3m-n`bk(@O+$m8Kp&CbTqx#!3Ro^Y+#WD0wD>iyApzndvKsR&ejCNHW0D$L?Y{&
z-+!(ex(HR2t+z8ezJ13p%e#_pEzeK4BMasjYdnwJ?s{s>RYi;(GFI&e<aS()?XjWO
zoYY<N7r&duY31K@BufxfDguxb7oZWLurjz{vg84UbI9BdjF6-Y{<Q7%utyW)+c$+0
z<MkF<9?8fp7m*z+oS$&oEZ>z1i>3C&Ca}f(#daZAeX;DyOc))i?10TL3cyU9coSu%
z)362xbDc4*fSbB%;E}Qhj(t3Y;h0y44|>PJ{Wnh8tcJHxpgfb*cWZ`3JyR%I1eX@D
zhJN6ES7k-g^nBRA6+Lz{hr;f=v9U%{mDsRr?D`_~Bl#>!q2hGpQrdVLN8GXyT@?Oz
z+7-##Pu6db%j-<<)!r^-I*w+`UXXo=8fYr1;~kdxDdbuG7#=k3gn<y*eN*f1KF$Bh
zuM!f73PNVVWu(3KcN;%3OXZZG_KI`Pw|e|IE$cJPXsj;kh%dz$%0ABb0gso)Q9gp2
zNtf=PDrDsl^fanC@8EI&q(M1hW$#)lTUm1o_)})e6B!%S2>wVoy*Pse*8?CP{EM4h
zidh=Lh5n)*Bd=X5LAxl_{94L7yp5$zMl^#3{EGwW`CdAsJ<I9#=;R3}l<zuf@=^R;
zn<3z5TIkm7d*S{Q&!Lv#!b=h$$&Th_>ZjClo-A{sUr5=r94uSOqPi<ojUL#vQoR0D
z9@tomD=TNeq@pL#T4Ehm8}_x9rr7iysgJ_9?J!1H=qLr{xNdprpE-Y{97Xs^)JR-0
z+mCbAPM4RgJda+_w1ZhDF<f#6KMK?RRs9%DGo70zPi6-^cvsZh<Pp*;9q$hp1HB?1
z1Thj8DEHa%3MWR`2XbC2FZ(&PJ%6R;$UL|(@;7M#``OoA{*nKc(^Xao#r-E)ODh-A
zOQX3n*Lvs2@mM-*O*5azC5U{B1MKB77dxh5mKt#I>$Bx2XqwJ0Xj>oqW`wvg3e07J
z)B$YP_^e2>9eu-`H$Ilmv=_G3H|}OW)AT=Y3o&4{OAN>c@maT9IRH=16i5c13XIbq
z?_;?AkC)0lxBghxk!|c14fTiYk`u>n>GBPA8g4GyLuofED1?CZ_ogTPNTn=*7Rz>`
ztYLiEcuec^&_!PYFq(!2r=O|z#Ft@!biey30#@bucj&Ux#$iwD(4dukXbV1&1KhNt
zqjE7f0?G2qOI$YC4rS^$^olhr+rM>-n6e_o=%}lSws$wIa?!a=?|&NpC0eU;zsWc{
zKWScWoG(_FP6kwvxFXl0Buq*6*?NA&t1N%CNhYQxAHE>%%=x@2fFKMkfr^ZzsGxgw
zHOYm3bg4i(w(q;Sv}7U`6Pvh6Lm7Ptpx?d8)x>^1RQ1|vuG<vm`}&L2^y60pzYXJ>
zD<jpjZ7a;%a|D&NxnAb?nXhfne1x>|M?Wx8nh#9<4K<gGSUSEv7K#rQi7l9~Uj!&y
zKe5Cei&k^WbCf;y%2w7n*r#WSHtKp+%roO8#6_LwL1~<?Pvf`t?S*N^=01T?r|(|z
z$d^E$%(J*;d8ErD{ZEFps9f+WKOzsubk@s$VbOjhU+t{w@7U3KK%cp%uNDtZ>3^p^
z+tzr;24w)=wRU5Hn4~bhvixzHRwBbhqo*XVUm->z)SC1la7x*{$-z(_ak+M#+ZP-g
ze<IxXQ4aEqNb_=<ma`Qe!{XjvZ9eq5Ib;szIj%cDn`nu6F$klktHN)8IeDy8S5qM2
zC%E^-H3a_+ex7p9xz-wfTD14=;J)uV*XXUj)pBRk@Pnh8CqmCZ&iJ1QUo^Z)B@vIg
z(FlGVdndb1<*-47yzX?oP5~QapP+;@NEQSQ)AQC1_Ty>JQP$k`X`41*d?oW^%4JZH
z`2Ld2K6XjjT<v??>~MILbr{_xL>OfMd9;zSl*--#DwbWDM4Mp%J~N_%)nv|~hmH9@
z)4zT59opehPLJ^KJ6=VVO6Ho)jzT3gR98(NX*61GGVm56{>rj2M18)d^HhPG&iNrA
zw3~jo&)!0v{GJE>iXI;ylLdla?z?dM+T(hzjWMg#Y7nngVIY8XR3%u9|L|1IzJP2%
ztaEj&!l5_dWLvR#((N&`TQ%Z_aD4=0dIY1oPwr0rTMp@~krDj&qrOo>N7s>&+3e2#
zj!pO%Q}P>}{K5VvO(@IX)|C%9o-~d!1M>GPa<ndYmNv;aFk~!<is==b3_pJ<30VRo
z#OB^{wUn=&24gx1tf_VnB;eG=J!0z~bgcJJX~-Ya3?h@$yfOIL)G*(evb?a%VX!~f
zqp6{~#WEH-Q1?9^ryjP0Z$$eQAyAnI&7&uR5P|1uwXt}Tyv%;tYAc#KvBk|6x%}OM
z9w~hzs3Y5-^N5AaGZ~c=G;%D?B(WRA1c1N1778yvhoaD%3rgBv|7EatdJt$>$u9}~
znp>y9i{nXjS1MjmdxDO!=)mKBA}-}mBp?edtY>D#G4~`p;ncc6ZCuu^9162{V`}LD
z1V6Z{i6Rj#_U_ZbKJm)mqBY>vgDo5E5h<Emio&5$E@AF?G3Le0&bO!{+=wz|)7zrU
zih#Jkm+X}@+r2~Gu{6>b5`g#UiRW{W{;Yv3BoGE+jQunVS36Np2enN*G1C~E`p6Rz
z{j*O=Hr16UEyfQtZjl0gr^mQuLw~P`!;^`G!)B$eL<AKUJuf^<NAv;d9AiDUw_1$1
zJPo3x5S=Jc8|9~ObacN3AY*5Q!3T7T+fx>?HKwu<G$dMGGeKe#-WS@H^V?x)LO8fs
zqmwi>JHg*Av%bEfp+&rI#S6h3Du-ZF@PZN*czynUs7CxrIzi<4CKW$DVUD<li_Y|i
zG17HQvoz)>Zp4t@iHd(8;)$gN|7tcwXZ*}bVfvjelr52^Z_wbOI56$NP(NgU%iVHE
zF4%~Q^N)t*gCUrPDQc6o=jm9zod2D$s=kAl9(4KKJVtQbESUEoxkjYn&#4)gw9Hf$
zxkomp*C}$oszKqQn1bcxZ?n$jcvSPa-%+!QHfnAo#hP0@4Eu7#;_sn^YAFm}a5$kj
zR>ve*17@Gze8vcL$w+E@`m5(=Qy*Q~>OI;1*l|C1v1PMWOw3eL9{)(8=9b%IM}6!U
zMcHwGU&b-gd%E#WWQ1W&us{G=3v*Gczbd>nU~XA+RYmPvAK8xKIjItpke-|6DHLM?
zV^m(Trt~H3n8Djvl-rxNw5%%KO97+CBR|c6-1xk!_MSCj{y-=Tl)@CO+~o2vmxWhz
zY)tX$O0oEU#QKS|IP#<h`b6RAJ-6|QMRwxTv+S{FC=#_NXFVnwajvlc^v;Fe8pX9A
zhf<6M-uxR-uogNQ8hEK)IoW(Fw~$RK7de(kPizF2`4M=?^(HIkpV8g2>9wssiTL8{
zJD#*5d^it313T$)w|h}CIq*lseFB}(#~)6PjENhpg?^8Xm3F>+oFgAQ0$j`746Z8!
z4T=vRkY=AQ@V*?NvGxTdbw)Lbpeb#Hwn$jEkRl=kfxy_mFdnIZZ|v?Bk47e(MA6iu
z=n4|<iHvkUa;Ob(KQEpZ&J@0)KduZPmaKWtq0k?3_hJD2Gm*qyzI97E{Lc|!PSUoE
zK1%~C&UNiRN)zQZhN-r>Yw5OvgD)#Z`yIX8TL)kRf#j_X&X!Xre9^FbrGeT&ua~Ez
z;L~rWd6P4b4BJHLCtjp5IQbK@Yn5nt(<W&i`^gZ*x8>u(A%E*th>PWT9zmg#$i405
z1u)h5>|%j;2S6YL!d+3OTS>AcGo{_X_PRbtC^cZ~OZKy980j48DF6mKe0<jN+^Q!d
z6_+wYh)0B~28opBG018{`&UQC-Z&pbLmX%ur<=#gp5NWH^gbnTA4PSVe6dX3@YPg6
zDr(C&J6Z51?||W#t8Zxn-kdrXW+H?^p#W}niWn)GH&MG$2$2AjQmXqq)5=*Mm#Z+(
zS55!)cU_L+xQ?>`ShOy)%)$Gpx&xYS;yMa=?0Luop+4>{HHg<+3oxDAZ>#%N{Z}PF
z*y2V1MAwfF%&zQdI+R~N;zar?o(NK19$nQ&-1jXxMS(~iDLANz^Z7LflHO0Vw(S<{
z-&<r5eM|PgPITIz&u6(UmdqwH`L~qt)9ZR|&VG<JT1tSay74fpHXwRH=;&6EH^5|0
z+LuTHxkiqxO8upJPHvd7O7}8jG<V61p27Jt)N=R14)#js<10eRO&x|CW~Yj>6fR1v
zwYJ=7+&AQJkPZmGr9o>Y;3Dk51T1!Wny3eOhpSrVHP)r8j)n->0clR9O!IoX`MucX
z^Zwiy9^o(E6WZ-Mwy{?YbRia8Hzm{^kdB^~ULyM{?&zV&+;rnvab{}4=iyZAU08O5
z@MV3_6vpe|{fU%r_^adQ-fn4Eqe6TqJ&VNr&47I)qv~~M_?mkg8<^CFz(S?YiNc&J
zF8sSlt6VZ89ZVI=m8J2(?|e_Ih#V2_^euv=$5qy@pDZ&CU`?d!{!Xjfsm*EywtUo{
z7M`bz>Lm*<Ce}MtHYm{7r(J5)5q}n1Vl~^$H^%&tC52`G-pRCoZdtm7PEwjEjW?3{
z11<BF;^@L~F7Cc<j;XTn3-~YX91YzpO}BG$c$6|oZbk!Z-y5|EV+OKNAcMg%n<sVj
zl-e-={H)#pfrJC{pT?h4SGInUS*~;n1&|2O@-f=eNKvRZ4R^(dzj~9cl6p)6{Kv$q
zqIam>;&#qzeUCS&pM&I|_VXXi>@ekIJ%6OP-srY6;NEm+rQ*Vm2SP=dpg`9v=8of0
zhOkW8VtL8ieYY31ECO7=oqO;@Z;1Kzm!dv@X`5}nJ?l8#I_`)#C(=D2%+*Mp=89va
z5suRsE0+)2#jpPQZ^r2U4e8Tn0Artnx(_PB{~E_^f3atL)FN~J89#@h6>YEJ)SkOU
zkVI-6t3OWeJfD)pik~F|nR+pV9<?<+S1*Gk>ikPTcq<UXdw_)K!lcpOk1iX@!U|Th
zoEhV5Sl$g0E!rygh}-<LV39X5LUDVTxY1S(aK&!Kr*rX$uZujy4v|hnhJ2+@m<J{n
zzX?5@^gCzR)(UN#6V?@^t({?Bx@QIn1n)vheiuiVE^S;DDalwPZ+w=+Bbf|;LoX&9
z8F2XSC1R0>m+4>0E|mxI>`U=jt6gulFW~I2NBxSomP5IND~h{YMxG>1blL@4eF^{6
zk#UK;tX9dPQL#jH$V)FC!y~DdwEKH`o<?o9|BR|2<^vN!e^m`uTJGR!jYdHl+#*lk
zs5uVJf+{6hxY=%{b1<y_plLd2Uui24dyK=?>^W&`SzLFoKY%L61$Od-y6mNtoA!97
zb$UD;2B^}_laB+luIKy^Uxf-;n)u=5i3a`beLsWhyI-)zg6gb&zCf*CjB)ffyJji%
zQ#_YU?1e#Jdx!n7Hh(Ev_|i!PMx4H_`%&raZn?l8`iT!XKN5a0UHxpS!xGFhTNf+u
zxpL3g?fDyTpaY9;=4WvgLgnhu2&s^lz|K2HyQC6^rpuT~EyYofcy)zpZBfmg;sx&O
zRGn#3W%*3V<n*|z)Qkf>l+k#uA^Ks&Nc5APO_6`soi1jM35BHy;tl(LIhwN&a^a`%
zJuR6vsJ;uQlY9qsuOj!Xa9BQ;(SH{l%NmW8PkWZ6_$5iF;!BonGJlTSxyBnademO^
z=BG6}B=bz`T7Vp#E}gK#fA=$pv=-a?p8t<XnSlE>oJvm3oLf8KYg@E~;oi^ZJM)2>
z@1e%vtsSw=PX_(a0OjVZhHr*B{g*Z;W~=k%vO2VPhKypEAzx81pO<(tom_IjrLqcK
z@zif`H$2E)oZ<bN=C9xVdlaU6(E3^#$Kr3w>V2<j@7jJovp;i8k@<L1`K44d!+7h^
z;j!IeS*PpL6VpURLOzQr98v;{UR7Gn4&D`3s(EoL{Hb3^v2>FE-3#8Buc{2P?839y
z#~{t0<hQk=<*T18>*VWy@2eH+)q@uti{*C6u-6YLDb<8RHk){z#wL@!tX{(Ia1#tU
zO(0i6^t)Ueq*UJ0(u~F1x&}Xc`0a@7^&Z~y#d7ER`@Xo6>+bhoC<@-^DTsWJz$8O@
zE6q6A|0WUD9M91pzApN8$^s8ZUZ_$??^mF<Akg}=Aj6dC0d1XBWe6+yIu{JYbdiY4
zRV#EJzDK4PtaJ~0363$VH|~2P9v-U1O5Y`@`zw}ofRT_oR8hMLibAhfS|zpBh(>i=
z)P@l=?Y7Lm!mt}6Km%zR(K8vzVwM*Cph$R+Xh-GcA2iO3omIER78s5{b=pAhf9(?R
z=jTLCaD?TmqKr8O<U~<+4)>I!4VC|q!DCP=dfL9nazg#oyW^g{NeK5d*yo|ECkDoA
zntQ_k_T^y$F%++7ZZFRcq6{D9EFXT)`{k$vq&&@tPha9Z|9HA1$islU8DzR5%*4N_
z8DoCrp?%1ld;gLWA)zb^F16gFV+?4T=W$cQrr1dXp6_NT3tcXsDig~m+Mf7k+r5{4
zaX`hIACv4>RGTPbo0)OUixoJ^mShV!I=(iQq{0#r8*6Kg8efTj75)X$kJM+F@I}UK
zUT6J1kRe_tI*P!@&+7A20@t?a1+@I_mNWPu$Azzu(L~LJHNe5}vc>A0`ZhB2cvGWA
zAP_fmED`+(u^vqY)Gl~k>@lkPE7s{N6Gk9+#3TqD@!RmIUBVFrL?mpY%{jeTz>Ul#
z^|pRz2~Co5>*5Vwzu~?e8!^P_RU&lWU@N*MZCSemvk~1wkWh+dwBQK1rWMVqOV=%v
zyWG(np@yt5?UDMjtFamAdo2)+Vqc~!LSR~pYD@3!kZ<Na$LxL#pj0rAjQqD|-&1$j
zy#(nmP~x*E^M5Xg`E%7amTX+r3i*l@v57pn?=H|ni%Eevpn^1ymfX-wy$foPXGw0Z
z1mw|CuClWHN>umq_zz#BBQJsnet2!E?A{p`ONN@O<Arp+Ey-6;LrrroraY~_4`3M{
zSw}L+i>d?Dfmf(G#Pe1^hkYDJ<6WdPOs0PO-=ay0Yv^)oC(`N1#}Xo#K?6jVUH7&Y
zt>ZCkQuHIG`z{jqel&L+J&lkI(N??MwcMz<!|5hMT0kTAH8^+~^b&)pnX*vkErQ2~
zPu@;et;%C0=sU%49+J^q3X1pBn*(_u(Z-0MEB<C}&nqd24lK5gXP&vny^1Gezm2Q_
z+4}k|da748EB;qKA4<mdW-dcMU?lj5m76+sff54`Km3OuVccWpy=)w?d#3TqxE`h}
z6l9ih_x!f;a;+y-2LrFARpjx26^wNDnukYzyUV0&I13UvG2=?cXJ$kgj4+>EWwuRE
z#tsWN%OvjLGsAyYLoMu0+RDt)d<{d{D_#C6oX)J(P(ujhKG$ds$ObrB+81&6HEUZK
z6)Ny?G2BiOc@i>AX>0zY@kpJT=631_c&}0lHb_-=K+_#y@r3`Ov!QI~A~_f2I%EKS
zJ7emxmhe7MD5eWcXU=kDb2?}pJ=Z2ixgDeUC)Cr0Dgk+Xip@@Z?fW>sIrSfsd3iKl
zA^E2N1so#Rx-^RoLChMgcI*of<7Fc`lfSl9eUto$^I#>M(!@{8#<(NM6gWSTnNcQu
z#F(n@U$X%u^XVuNW%;9#eG{+J3*S0eDbzOK*JU^!PFZiL;WG-ghu)JglFs-`?gx{0
zg^83973w0~@wfomO4tTCaGoAw5ySDvK+M&*B-%3F6upN6405V@Led1c)Ncz<5n^(<
zPiF~QP0i%xt-8+Ans2bE6p^{j(LYwh4VMN%1BUMdM~<t<Q@y^cy+VT`F&9XO-UmwV
zNQSFdQ9@(h9NntM?`(8xNGQ6_Nxr;hVsb>=lxfRktQVj7Tbz4GRfJ5l**g7#AP{Ll
zg26O4>0syPBWE01LOh{Vto8R|+hzwxjMgaa+4w4znI|@APC<L_p3o7U4Yvm^QVm_5
zOVTvNz*$GCz`?SqIMSzE7St(9XFgH5M$3pE98RZ&THMK=&w-ai#*K*W-HmlvcPGZx
zE|2{<GyASOzW{5A0ac5FB?W6uG!rdIrEI7rRrbvHHSzD#WAq@ZMp<5G1U+3Ot38^0
zkWUd%2mn+s+69nZC|8j!jec@6cB^Hb=$2uG2_CMALSGUF22lMX_H9x|#-j&=|NN3R
zm3MsbUbp3kXQaeGa7qqy#C07rvSnB^KY)88=v<VG3e-?(hSrV*cU%;r4bEmZXYHnj
z=YJxI7X?am{GN_kF@k$GTAc#_;Ns8cBGR(=(_bTbRFcwwQMtQurq!C}McDx^cmimi
zLAKxbb3<jnqHw?7Ob(Pc29cn_Of$5{HnwD@ef{P(d^BGojR$WAu=r_gjLP$CQG%I$
z%4-rTB?K(yt0^=2p#hk#Ngr`76lRn^!7j&$f_FU~%2Iu9H{5n#WSYWDG%R%8>}4r-
z_j-3kt^W!$v8L)O)(h**BQ`G7xTp5~*cIxL$Cbd!4BDqq1iG#K=~<?S695vr6nLB_
ze5lVfgCMMs$o<^jHKS+&C+0>9AS9TqsXP_zV8f-mKeyL`)N8UjxE6ah@%CHBRjiYY
zIAN$UE$g3x=1lf{!xt1ZRgVcz_({~_;ON4pN2KG(x>7`@%V6tAaX}iC*iEyNuh{Z<
zP4>B7WT23ha(Y<Qv)`n5lm-6SK+9jP-Q=@WZuFraml@kW?^>zwOadZ+-><%ECjXix
zzBw)Eg|m*luIPw7+s=KKGbMdlsp~TSU^^P?aC>~6uI9b%gV=r|dD=H2?R4lgxOouS
z4y$=lE(0JSD=3f=cism*huq^ph>fl#eruAl&Qgel9&OAcjR<Alj3|8+Aa?D7u|RnI
z4cuE_e%8kAT`@G7_haq-o0?hACI10@peGt_iAj_kTckd!D#<{VT33EZe@1Y-4A*a^
z_LEZcNew!=aM4z|te477Rk)Kg#_|Dfz~k%pOW!YGvYaJ4-%?DS6ZBjA&d8C~ri<ER
zgh~TqKZmiqZvv5nGo+gIxf6*Rwk_>k)uVWD_8t)s7{ATQEaE!nr^~>g=%(4pS8+Bu
z5=G_(yS2XmS6Y<<>th8PL}|H=vh_+`rF++VgzfHwF?)POOi$iLaTu-X)KFZf9*u?T
z?)}Bqv}%JriAp(<V<mQC+Jv+oFd~ryN0m6QSZ>#fdiklAyZx}<fs<VM%#R&iP`U8S
zOCWg^;(=)Eo+T0tz=;EEkv+9t1OqTPasza(9LhQ|<YBtAV9Rl<t4SIz+SeaeFy7sb
zJmT+NaXYej@T-2JltU&JP>iL7=i?Yhzm53*70oyBuUiOR%j>5$SfU;Qg4%7qnxL^X
zxI<aYkVGyg07x>~FR3wjB?`nEIEh!feQ?EB5fxl_Xbb9SubMyRuRRP*Q$EOTP&drs
zY?gCRJmNK?Yq1V9fe#~p=K6_U2~;O55mu0P&seC38CLy+tv~Wp-x`q#x&>ZM7-Q3m
zW4eD3)V})uA3+zGj}#kgL19Tb|B00v_HN*#OGe|6oD8pH=^tCuMly2~^7SO4!t)?g
zy6@oxDL?L3C~3=371z7zURp$Z`m{P~37?4aXWtd&fX+CBi)^dH8Pk?Ah@fkZ()!`)
z1X0mZu{ka}gkauYDfc*`#Qvobr##rtiB{U4PY809iCvoe)JrGm!V8g_?{9}sjEN0=
zJ7<CK)qfT-JtwkE7tN=pto<cUp>$GABFPts7amvF?Wxg3FuLN{L~fQO2Ji_C{H?Zr
zfx1Mn>%xiT^c%1j?7^kj&;Ai(5oLtxTKUAKq*cME;>S!-zWFU`$Q5Qp+Y6LFt-d-*
zI|Yi&1D5sozyVt6Io*3Y(QbMUH6Zt&!WxBbM#)2%rL4sK&J$q7R-gKw`9Ij&^fvO1
zmLCl&t)><H@>E;}-txS})rj3hexI>&3B9Lw&HJr;em;dp7)6hFS<O(ccpsv>x)BpE
zDm*>V;wiw)K?<mVEhhW;@iA7j3ILNZMLoPF&x{hn&$8>Qc**$GE$ELU6tzKjmkL)s
zu`UM#;VDh!tL`#Wf4#+=lYGrdZfsUx$JbO|r4oTwXfXkrg4`=3V%Izv8Kj(cTDGhf
z58eQ=^xq<-^r*F|u>!{;_Em2^J)jAy{&s94JOl#v*YHux@HFrJNuq_KMj{cd4B!GI
zyQy%L+U#GslYhy*L&uRMJLVUA4D-9ylSREQjKgNu|8-2>jqtiW+McHya%Ket_zkuj
zm3$1|$Ks(pmrP4EkMi}jCyvG+`NH5bpZ;|q0Y&){coHggK%;uf=ygA9Gjg@-z<06?
z?Ix4=xt4ntSo@+9oy6cjU5elzD4ub<xZ(T}<I6I7?f4SEItGZCqH^<VV-82;is}N|
zCJ#Iem)VB}DuIbW^(ifRZW(vT?%x(jWuuHGjsQB<=Ca$*?Z803l=!O9dzPD^C0fKr
z2m~N1WJ$ToKe1jEO_ejA(9L6h#ABVRAa=9%1Jyi$K$hq6Ap+kBGrPmDUB7Lrbp7}D
zT~(ys+@NI%xgYT&n2Q#8OCR|4nQ!zGoaEPRz8-8C*(dO+(WTLQqg~S|jv4N|PBNb0
zjg))lyh7kNehT(miL}8oD{hw%^MCcSz(ND4@xi<I#8^W?mgCIKl)sq&d`dbMmE%a8
zZ%Hd$rWX!{XmXV0v02hdx+S*&R$ekJ@^R2dY1h%ElrES8zk8r$p`pt`@N(yoa$K8(
zjhK6o-347bbur0FPyBz2^m{=@pz-rv&bObZwz1EzCUD*?!EiX3xSDA}=rb9C8xwGC
zNrz1eL{|I#PEL6#J$&j5zf+699C`UqL5~2qV!2i!hst1Z=Yiy{?8%n+gz3Y*e1piL
z6O(<SfzEwKP5Di6F}9!EG|9*jG3lry?fcTF&yqX}`=^~vl)#HXi_dK&J=(Y|Z8<*}
z30@Mwy@}{(<;t;;0%r<52#kSC7Bu51jNyy-t#x>WSy9ggSr6AII%4s~kygw*$EfUf
zCY$#?+H2MPU7Dk3nKJ$X=vT#P&+<xyH8g+ku0O7RnV>45%*v>a)aJH8>pmXiNa08~
z#<?i9?rpIxfwN-`_yM8u7UP^;zQna2AKg3DlMak0P;y?5eH=PSdr}C999F+r^0L#3
zC^_JpLv#E;X%i<g;;`yFp%_PWU!s&&mSJg0X@WSs0r!qKjk}nBa(sb>uSsyGl@a`R
z?7#3vi8E>bKryMKl1od9W<sxngnJ;(4%kmLe@X$yzeNlN(!>2}*q@jjedoHB;K~H?
z;h{>mbHAOQ?yR})5R|r$vm|$Ig$ZnnL#t*8KK5}E+@@Kv8^(9yG?EfT5PU1;lin{4
zLSUjS%c8>N;~$k2tlzG=zm*BM6jW_WNrM*T)#Gz|3{UBi_BFO`gWIeuu7`n7qJ7!+
zoB49j8dB$kFYl%CkKkQAyfMq60G`QsWATY0(n^Qq0MqRiwat%l{@qJH_O!Wp|2ASV
zp~0AbnKEINW-+Zod_NkA)8QUU3OqKkv6{WUv7umiWmN6w4y`-SB|;BV;-wz#^2nX)
z8^1f-0CkHspK8Awr2)AoVa;dY`-(cwmj-#a^qWnY@txS1{hg<?>b=G>2m7{0c%!)H
zbs=y!>P<LW&QMLJHU2(FX@TL%w+0?R_@3OuR}@{Lg&|edk7qocJn&~JH@eW7osFA8
z${K&8_u3o`BS;wP(94`+cY0*KMQ^Pn!zqW=({HA?)YFG$Dggy=&6wNiOg1+<AFBE@
zZw48Wylw=jke|06tUZ<MZt;wTJkCAv8q;y;Hw%GG<FW<SZJt=!e<_wD0%!MN`hJPm
zeHrurlz8sVi+qX5ww`QtsWJQ+HeWESuU#cztrb5u?LN_a$E%6jk1*~Pa%#{$x4pRQ
zkb}<7AKLw!Ny9$X$MW3rj_>=MRpDu#0WZhGe0hmuS$N;@)mS|*4E~<ZFs$_6CKXO<
z1gc-%wS@=}LjMsBXkO0zzU&v~<KDHjs^+$;w(=)gvt-zC>72mu{YNq`RZI2l38VG;
z&9cBZbIPwX$TSBPKN?k&W?cZIRHxaR)9Ch}tUXnw4yeb9w9Y0HH=?rzg6qU0_kXX&
zikwQyBcGjclOqS5tHEx2(=0Eyz4ua<7wa614t|PXbnX@1^gG5j4}b>F?#wG?KKQUL
z>D3%@JC#ak1x48sS1KpB=kC_yJ!7{-np@4coMInIt48ux_LJd7=zo%8N)@-)@Mqn?
zCB{TDMLie03HEs~sahW1-369Q7k;$b8PLlKc|e60hl^hIV7!27_os@JOqQ{&^m@It
zOU+BY&Kd1I<4B8XNEOGQ1fxpfLKZXxc`gJZVp(4YLUrlS#O3!<O<<Li8^|ViP~L7q
z$w48`7Ze*B>F;mrX!*VfOSO6<2k<HrvugZWKTT3;LK9K#TC2QDfk??*7a5I^N#_yO
zlk5BFkcI9a+tRGJPA{oW`Bx4JhUDGs&@j*6IQ?(=Va@q^QQIdWfsS)iGG_#=im~(I
z#@c8z_?^~$u5IHC!|pN^Wg8orei*{_?;yp&F8Qjk*|OneM3YNAYj1+?PD$;rZwV=n
zzRaTq6k(wsa2;IUu}FF`nav--XUmtN>839IE#ij3B6`Z33p3V1pDGW)z||*)hzf4|
z#5U$te~Ik~F3J|qKe`<wNN1Otc$I~u5`XtwL{mgdIA4xb^cB@R&G$qUX|DV&YaYo%
z&L(UQ+fbU3rKLh<U@cfzC3O-2z?eQ_-(cwL{I(P80|tK|D1sOdPghb|TlSh~KPvqa
z>0sb25<>Dt`sQv!b)&q;?!Q=CJ{$~>URp+ihe!Y><Q%ep+4I|OL&zo!K@3$9_XK}G
z;>>%EQdotYO8c58TuoTLmkppGqebiz(tbA&68SL)Ym2PBh`wyUU1P7^o-V&xxQ%&I
z1`AGU3bH%$YNuf^IXQ&0!Tdb$I7(3)vaau0-GLq!gUc*E{8a*|5H8aRMZy`Da}U;Q
zk0R&2Y^ra(MQ4s*7$;THElm5vNno7%&0ev8lfW&rjt8_-k;O9IROA2H67|DhXL*}G
z{w0f8xhpks8E$F1j1p#xIWiXZjBfJ5^IaMxafGqW`53v4sVk*MQwPMFnyS<{K}hw@
zd!wvxfN`sRTTVb-x;RGfyoA$6%2mq}{0gGHfQRm0M9P^IUVT{HFSh1`bO^0`(macn
znI;iM@Bp<><Rc~X?fKpyA~tHnvI!PdWl@Ff{c(Mz{#_*v)u754f48v>s_2kCwF=JY
z&PSrL)N>3YjsOL_gI3lipNJ2XrOul1Wkw+f?&ac$$3qPNi@V_?<6r^H*bU;p$q=>>
z1HvZjJge#_N_vj=kpT^tx;Xz>wFlov3AYhs1rZwWA82QuRk~@988N#nN1&YOzQHa3
zdf&p06k#ycU~rClDyg|%s8er0F|}lCP@)Z{v-VzuK5T>5<=`+i$=M^z<d&;ru`drv
zD}9p7>g~GidW{KUn8}&u^k7nksXyJs5{jRl`tHwI4aOs5*?ya1z}BK`rSbW4Osuwk
zzP!7-44zPQ)BpsgLeCxhiNT!Za_>?f*CXJ;7qQ8#<^K=Qzj13VkbE1||LKI?N#Dgz
zIi7fmv}!*%-MEm8-?F^f%-T%3{*>l+mq_<~HYo0Jb>M|QdT}F~I?YSYm@dv}bQ}pW
zFz@r0I@tf#7ANC?Q{`j!@F@?F3h&0m;LY#5vfPc>xE`I4U`@X&vyP-}ERVw+47+d*
z8XbB2sGh5Hd+{W*%UR)T5f#vG^F8`sWtR@~8)B#C?QQLfjP@T}EoZ+c4rjM1ZjGme
zhKp}sp}+!o1JpBW<=)Fa2TtF_-3=QXVIC2VH8rTfxH&^ZfK)EC-4LZ7vc)gt$j;&B
zE7W-upbGnrXAHR~IRB-K{$r25*hy70%I93yIb>rw9p~$V7vkmchY;zmv47Gi0=ZcH
zG-@I9F(-ru^j;GAl76&@L6#uC9|fE$kv7K>zz5BI7}0z>OgtK##ulGp-w47^V_Q0#
zHuCjpJ4->8o$~7oo7i0@Y29LCCw36}NAQbYir!Nzr3tMo>~3U^4m{^j8Rl_h?YAiE
zyhGp#8%ummkQ_Q>=-qTpeH?NTNV6{LaTjc#b_*38J~BuhbgN0&^W0t-1FuH~#hhC1
zMn{OGvG&h*`xt>KB~YjZM>i)`Nsmwxb)F$$r<YPHwV!0V&ufqKc|0#JGL-x$z7jU6
zO_Psw{GVRRH@I>XEa__@Kz+APMRe*11R#?(&0!_hWIB<((OfR`TY$wQ>BK1O&x$6u
zr8b-zG`lss-&MNJ<t*)>83H{XmrzrxLCXP3zx616s_fK!t~=CSlew@PbyMR61D!a|
zsLmv7%{S${Gp_IQYMXWX0v;SV?H^Pci3At^G0=Y~46@<M7=4?8k}(Y00h#Mf_oLZt
zEaKFb;z6mASb~Eguu=$Q19<su2WT2XS;fbTS~>dK00jSvyO&zMzs`y%Yr2nYu6#9Y
z*<ebMtaZPY1pYJH@gdD`b5my{U|!8nF|1wUcOG>YTghKFZCC2=D}K1OtgE7My;a;3
zs<??3)@Os1_A7_?q*<m<poqUE`vs{gNE#3&pb~Jtpqr7P??e(>O4r0%Bicd7q4#W5
zm+Xvgm|hj6LwVVX7t5zeMTO<OPj`cE&#GepIHe@n2rhx=mL&(@5j6PcH#(beQ9WN4
zyIpJsyi|#V-!H)R;a@QlA^Sv^gHLsmemx4LH>VcpsaPSw)$>PxMkvdb`kyB?G;aGW
z#kZjiTUtq)U|p}b)@9@V<bL(e^H4Y^m}z1uIIpXRTP|1tiM}ZzqJ;t#i5fFPFcqfn
z2z=owWH=(tAaZh0Ynt=ie?x=L2_;a!FftKYNz>#PZn`{gD1Td0g|}}0wF!xU=e0hy
zHyG}uiLnRu=sn7Bh;pjoG`}Fdv!-#0vnYegTi{ni^fepVyOqj1R+r<N-dlesR3mD9
zm6yuO68TADe<cb6Yg!{id<L<`O<<}=d~#(&Cbq{P4PwJd2Q!&O0rFb3C?c_?WF{aG
zZ?hU~C0biaE*z>>Tla9W>4-K)UAhR`x5(nWp0izSdW$kHF1DHoWrOr2;ljn~@&VXv
z!;to}?n^_JyK<p97NU<|6OUgi6*^+*3?Ag-1|ge9mf(w0fQT!Dt#yUCTC~%ww`fXn
zYe%j1#Dvpwgl2}sXR*)Bi@5Pam3~zMMZ4~y|1|3I$4Tu}z&RUh&RMNk?j%v)?=jcW
z>~TjZwrA8#KSX8dc!BRrSBlPOK7p;OZB1NX&jQUu>3+zhyr%DF>>dQ*Vj_VXb;zX(
zIo_JSRUrlqS!=a;rg_c(E}I>bRiUcdPKf);b;qbq#k^^jYshOCbOAWZUqIUnQ9<E+
zRok82*>fUKBEK$&)hXv1&qdgv{1eugEQ%@G{a2OG^*wXuG3b~$Zv(2H3pToh28)cr
zYev{e<VQ3a$$qKuIJ2AsLNvmpKCza{4-AWhV(06z`y6=Ho!wS{5X#=ko8yJb{f;7i
zLZulxI-bl`%@^msa=c$EaYC^O>l;D$^5k5A9V}#~b=H}vTS^YGz0)O@*21Z`O0#~2
zm-{BE`O_B3P}xZhi&+@oFrY>(h+qjmQno+gN&jLJZ+iW2Lp5>J!M6k+iEQybA<igG
z$gTS1+Cf9P^nkLzr`p<6KqN`O)ppZ(S;pYC?O=*vBqr!*TnwJi-ek!Z-fC;Q`)J=X
z!!_TEG3s|BpO9!}p&gL_)2w|}bh)RXA>pKr0+#up4UogAeoM;JMHEB#{VVG0Br(~t
z;o<{@RWdi)*lYX)(Ellg;`^QWQ~uKqQPcgOc8HPgMHiIdN%#N6MK-MwOPxW=U$AsA
z6hD+|C?>Sb{SdlDXTqSEA{YPm#r(u&eVOc1!~;E`A@Ii?TO5jMV6eX7U#3^=(KbJw
z)X$50!WpsTP5Ivi;dG`1(Q1Mn-iO$?#VW{dUm|NOB$ja@QE>~Ph#-mNJ6c1`0#bL8
zb^;qmJmr$E@2Yme^?T-@OSlKhpe$oGY@`wSL5|#*;yAw8FZzoULOkIY8Pjn%67i)}
z{l}6W>Jt9?%jS@lTQ|d>jTNm#PJF*w6sHB{j<*k-*5Fdj_2a$#24~eH=jC%yHg2!H
z2%3x{t~SIdKK>vjhfqvt_1%^QjO=(dj%>Ar?0bGs(A!D$zPa)z8Xsv49exN|7Z+O6
zQRUPUKl+X}fXYb+dctaPfkRo?R%^7-e7>~5?UUu~$KcRYt<o{R+o1Mwy?Wcy$&MqM
zX6)=qi6@i8KsE9mLTjqme|l?11n=)oV8n8bC54`TgdpCv2(x><>l?;Xf7hVW&f)4q
zKa7KyF0KN+?^GbTpU)om1euxNw5TZ^rg5oVeTlQIZ`^tJb~Uv$0s|b;q%9c&;>51e
z`c$)xs>`*eJ`x%qvrd4^9+drc4!I@NZrA_@HNnBsw{F1nVkF8z@ec`yiSEx14LDqy
z(CcM#uu}3QSi0JRtqHAV&yzQlJ+KM6c}f*i3M;DrmHQUjI^~nLX$wx^x;4UaN$FVt
z&Pmm5(XnW;jlm;3k}hAJAQFW@KVom}CA#H78GhaYQ}k-~GuPz3;-Eyv;kR5brwy=B
zfR<wKis`>_?_yPB`{N!wjgu_j5w7@JOz2OZGu;OL@xz{Q;*pimlR!y=pCB{#+t5AL
z32+inpvA+yi-KRXMGWN<G-5CAPp+`+`51>P^XPZAp}vroYP!sQG4E;+EVla%vHfn|
z#|EaX2j^1(GZJzEND26^KBVP~l=+=c^IXE0_g9ivk#n7S?_xastD0^LpRc#-hdV_0
z{hMS2L%wspFLfHM`AppRfmS04pWmTbkzSqy*eo0l65!CmA7^`)_Abq90Nw2}UpR@b
zjs1LC*2a%m*M;*QVbo&1a^QSx7#|CBpliq@NG4%{O(l;%KbSXgE&km>;pUK}O#LY7
z=hKdHY$ZPGW{US;_!1p?0+d#mw!Ifo*>HlFQT4NYKK3Xx=%8w}&z^)%lHyL5sH{iV
zQlF5F9w7Upp0rr!ed-P?%I&>cfx#jb4Y6>zdKeJ+1*oh}oS%=)jNbSgBl(?h{7G(h
z3mkpT=d=<%WB8h)flTzOV9O;(hB8mL9kMwruOktKJSwJobi9HGFIkOasl_Se(vrzI
z%21W_{aWwfkgZ?Y-lndle*~}utqX+WJJEhlYJk+i+X|(+ms~(05=ByP)FJUvJ{*@t
zA0UuB_vEcydLk(s6&v~4(`<o4;R*T{&oB-8e<%$goyxpQ?J7FYYc+<99Q)lgt?@xA
zX-rc?;u#A1GRWL$gJsFtWG893Ku3h=i@qaH*m7rH?^fI3dBIlxoT$9-J?M?bJblm3
zNSf4ot>h))Cc`&Un_zxPiYeT<0))7e6|JO%<o+Dy>vqVKb!nHdzRPQar^O`$jL}9z
zcyBLNT;7G_UWPng87D7{$!@~X$jtVVDcw-^!O--QX8KHDsde0RJpzB=Dl8CGTWVX;
z@z}4;-=5e^MKO4ZMW5ZX8QNt)v*|Dr=8Jk=B(Ie^;aW$=do+aqxqxHb|3JavGWra5
zNAiX(4t5fh7us?>RPI#)^Zbo>R|As>Q{=BSGd7fK<Qw7V=POlq(v^hD5W`{}-=-TG
z?Rppmv{;`!1mS<Pydzg9Su&G;Nr$7{sv4_rXpAg+zV<=?5*r72EPpGHSIds~KTiFc
zid5uOgrDV_@}54LVWCG9hyESCUBCM=!p6trMfDSKx}yhwN`2b)sPH_?H#Gy4Z&@}?
zi-J1yh5g_uJ$K5DMwa$r8nKfr6gUS~qXm$K^ji+|s8C<D(KjLf!J^i-ww6}Zeuc+O
zkGmNBW(F-)pGO_sA2+=IcFk4BTCtor(T-?tqd2Se$e|)bBICmVjHoi24i@iUlXxRC
z7h&F0_niYUl;%?K$<?)UzeUsIwS96wnf^rnhCSqZ)Lc~7SjHyJYjSj<h})i~cqo;B
z<!FNZ;9|poC*VEd-lpWNvmJMdm~s|4<!`2F@Jj=GBLwAG-OWF0xtFuLL4ct4!Ryp{
z8M<8OnMs@=SZPLy79X_Zri#Zc*~nA5pZxm|HO>Nt-!1Ld6HSL-vHePI$O@TM*k}5G
zq^A`yH&o-#vlckF*uR?APTJx>E?Zt)SPk|+!z72N4GVeicTyS;X^jnzEuc?(yP3P>
zlN08Op$ef~naLYyYId-keedgm&K9%#VT>(izue;kL4<Y4r#AKvbVBM_M%d~z9I`8}
zO<4Xscx?5Wq<N?A*?YIla({jg0-q0{zTX#?(mZ?QNe1#qNwf@mHiT$g$_#1|qw2FO
zRE3Qn>3eqt%)>!v%o88*#7LwDY|8_^$tgu9%gq<`i_mHPkDUVII;Qtw_RB+^7)5mC
zCO)~}!GYsE*TkmRB3s_G>oXw8BYs`AY5sg;!geCmo1pHIex)aefr2!DvA#Du%ja7g
zDT(m(jrA81Ja0NuFR4?npk#SW=Tgxj>f>|w4$m(>3*rxA$Nt%CR|a@Y`UTlmzk=`{
zlJa_LK&$aMpOzv$X4@i{)S2$*P^*VjVyi`d><LTua(@O35dB1<y&Gk`LK{T(DyY8R
zt$d1V@;WlCe!`me<h|mj>`~!zR2>ac!W2?GYJ3BA%brp-`1T-|#o!YMKf7{>+eEb0
z!+5L{^jYlE?$bB4;fJkXuVVT8R&evJ-b|1^3j?y}_D<b~_Hzv8%bqq0jXutN%#N{S
zp56+LREhdWa*YoBWmMqM5huJdnuW}*xjc4<-!<yQkD?eJD{Q;XM(<yX?`p=F2*o!?
zn`kWJt`8|cu`SQ-s0u%(tB!h6I41m=3A7kNZA{j^E^;t3FxiVyKC@h@$qNeCkas6a
z!XmbjC)uAKdbkR<=q8Q0-r#r^_XywwUeXdv7D!0{LJ)HGp#a@e=Xhq8Jq07}zaP#Y
z<Ld?88FYt#3t2=VVcl3z6mP#O-&*edewH>NB@Qmhb~hIpqim9aAH5+>c)Y#7rpKts
zZ(aG`MU5}{u>`^kuxaA4s219<A|G{NOKDv^1`iL59{>GVYtPiXZkO2mN)&g2(3*85
zekL#c#|vwqsPkYpbh%Qbr(NbNcExY?J_YDh8Z^w7#{*G;43yA?l`OC2K^dy!{v6SH
zps#MZGnK;JmJV?nmva<nJXxc&4RYOlv9SV(r;Q)tIq$Q{pIz%CD_fh5NcdZ}s#nkU
zL<X#+F*O7vLH-~Sny2P=>Y@bw2t4~(UEcWEs7l%MB<!}dh)QKkNlS`fL!tZ&X$I*W
ze{f==JFtWVr>_4?CHKX#Q!hWaXl6K6-TZ5uZ=I^?gOjBz|5}r&69B4}z_Tzld}wGy
z?b%1(T+-Xdvq{HW@~S}iyt|<G&wq{s>ILpxD5FIltHoy%AD(m<*-9OrjE#=`&No~h
z&a5kP)IFE8-cJKe!KPQ3n0;~FRcZ(VX?#sEJ~^5!nlJ&J+T5a$c0~u#e_WY-)+gKH
z{YZ+(zrN=FbSpBh_D&uHpO)o?oyPXOrEcIeidy<KdwN!Dq%<0}?*eotcp(%c^=+zA
zKA8yv?l3uQ#A&X(-ptDd5i*?>H1w<d%ovWkm7jj9q7B2GFIt>R_`PpE?ueekOtlSq
zD?*Um=U0_Q!WP0HmHr;0IDH($;EznIm)1y;*Uw?!R{{+X&=}A!?<5WRid`iRdZFm&
zs+(j3E{U~_ZlOKp%p!IEGLh?yA26aZ^Jr+5KW(&=>FBfk$ha?qWhR^MGpKL!pMKr$
z7H_L>>D%j8#_jS#U#seThtMgb!S$HNFps~D>K6`Rg~$(tc6ZusJL_?`wc1}j9<yCO
z9scI(xa-mJZ;=@q_fCBw#;Z&(Mp`&azZuk&vx=6taMInqJ}D>}iJbfaut)2sEk--+
z=k|Oi_(_-E>o54eBAm;1UF`l>4h<9QdG{*Q0M!SAvrL_28w3V~MH0;eJIaZKp0v29
z&8L=>zYp8bg30W+;VutS>>Hof2%kjM&0lXT{-?6DY-;oS+IEp5#jUsnEAH+Nh2lYq
z6{p2DxVr_~QrtbbyA>}E#e;i+;BI$*|C#3lJTLPmbIoM!wXeO`ah$Pjy|bz4_Md;~
z1pC@Ph0t7v#4(dWv8uN^bN0A5tHRbFqD*Fu_d=(_{>f6i@-qE|xqU<YN8oj_yB@#2
z2jz*7*GY`t=-?u|()ju^Uq1RQDLlhs;{PqgoZc2bT?(S7!J(Y5g@B@6x8G$f^tX*F
z%9EGK)onwTK7{&cZ+-XdI#tJ*rt9yTF%A4-|HMO<fT-?D9c;wwDAKXSuPee}g_}?|
z`V-|jdWZ=>7m}YZ4OB1(@)vk7l|OabbzUsXrP!g8cW9xZUgV;}&7FRAIGETMG4Qrc
z@Rcqr_DE;dhCVJEZw@J!9ZD-9!<^q9?2uEH5}_nyZc=qL*#3$=-TZa@m4Q!GdR6EU
zY4dEtdO*_8qNu$V?|#l`$KVS~E7m$SZ(prQt+X8QNkep)GiGJi&2G26{L^daEZycT
z0URxTOx4*qHtXF&(amaJTQ(5U6R32nkJcNKyGF-aYJi~Z%rhUGfjt8)G*Wh_n4ye|
zZ5=c1H%zCvqvVahp$pm$Mk{QSf!7M&wdCw*V-Q<*DX?w&RYHw_PTRtKRD@@Cq(zii
zt?u$wxSHhJF;K4=a%UZOB&L8KibDYJ`ufhIDyMtYH6BaRStSfkIWkwNG4l&t(SxYJ
zQLQF`xN_vv{wyCaZ3~tpf^SRmTuiH&-0V_x6!%xw6rX|$3ILYPtUHWr;FSzYX&~<H
zx%C-1bj$?Gwtn3H8oNz*VQ+ZFS^gczqb6EX5kJu%>@n6+zWdPH@-*=_8Us~~T#}hK
z@N*_gzQ<t_Jq<NQ<OvytK1Fhc%ZHBvS{3PC+x9Q0{t(?bPj~v+kMhM7WcVRoKnBg%
zAgDwd4&-yx61pBK92Zv+Y|mm1x4pD&7t-tL`2FFqX0wYHeK=s~xP965x!8l_CVVZ{
z_K0x8tkMprNd6zfnKV`j*+}r$kg@GK<UJqBgw+E+_F;;cO$h*q;#NIBAn<O(*>2b^
zx|WPHq$4pL3fGtHr0?`l)0fV5ivu;yzAajO`U^J&U;4RYQ!5P{U%oY7%`9|Wu8HJR
z(-Gzn&XqZwD{+HyJ~^i-OJi@j#631hjddE()`yHmt57~FEo>vS_;W7Y*7Lb0Gi9*k
zn6ht=38plebV1|SSqhBbd_+mA=y(3l&F!asQT$W6(CKy$(Z6pXFhQ1N`5w<2w)15j
z9odK{jC;(c*JjE&C%`l3cN!Le0fv>-!1OqU$tH*A(E5nr5y;}!?O8NyN;W%q`bp56
zU+>u<+@g=_-?kcDzZ^l_9i(>K7B!WJfC<RD0TbF$YULk$o|4$ZTK3Tp7Bn-Juf5i#
zmk(N6<AO6V26B^qw`9MpSlP(*ur(oGT-m-STm930KOTO)9_R>p&eEp4?|?8Okn+GS
zT>@*2p4(bjyL&ZTzw2()&PmDGE;^QvR2&Le553=b5P>jB87lho$Vtn?l+lg~B_KxR
z%i|BP(Tzs6D+sv*OVrpQ)s;aY6pdCX&P7H}O~-L`b=Z(tWH<g$;qF0h&_~1n^QHl8
z_BwTu6izC#KH)!)2mW5RbF#3bP8sYsmHDkGCsde^HRK$tEa~}XD5;G*ImVKpg>AO;
z0SjPieNi@v_gnd&TD;`@&{MWwPoa7#qokn6N#O3+-VD&~dsROYw)rA29lkd|5CR?R
z(!8zwUPV>1syn!rUoQL%UwMJqj8@iI;X5^}PCrMF*a9p@%>$|C_N5b0vDn?GY5<j=
zMB_yoaMQ1J1@AT`y;2t2BfLGpLzPcoLNBY?9E_@=a(V1uWt&(_Az^HzglWJX(l{ng
z$AM$i^3LRJ17&=0UXtUTfB3l6H6k6;KD2&cDj{lWFj(iizhctLD{<n+R(xrEne$F;
zWTMc3`1>L>x`NI|S=4kN8UhC-(u8?-#qWPk>GvY!hTsQ3>$elCyZ1wiF<g4WLy_WL
zLuv5dgHX?Lhd<wLJ-5e=HBN`ChC~3Nb*tamSY<%&m6m>oM=6^zt53mj_xLU<jsFs$
z+7!bTtfWnIX*qrRZsl2C2PpNKBcSH+B|k&mL3diwt&@f;dm{2!9RBko{UfgEzOVK;
z(u!RB<EG$+pVCLuT%|5@R9oMot)jSED)vTrioPIz0Oj^W@lj%)0jowMC7G1x&|!bS
z8jo#;Vag_P&tS-zAD@%yT2%EL(=3kPHW|{d7tW{;cS)YAeMP7hX1nE0-Y?`Xo2WM8
zo{ZDZD03qs41DNAjs+};*>qaDdBI*2;&ndxKcbao6C<K_tbZ{)FZ!d*E6D;Agfm<I
zGB3Qk*WQj_vvgTHBP=3o(PfIm$7y|iLDe!(AEBRH7_-Rhzc=T&CRvmwFn?dQF<b0M
zqWP<U7P0kRz5>nTY5RGuXUdiv!~|-l`MzV}<lMlqzS2Ole~~-c_V|0Z(#wT@^wsI4
zr@qG&5gW_55RCLcq4&nPY3=THxl!`IDoRpT;zpQWv;xxXSwG^@XoKd}L*FO%<)zDK
zqovItLa9eyc^}s^NBSOwpo5eECJF{8SYO;@WB_kXji669J*RJCWvLGTJAQ8B)wqTg
zfj)0_8h|8EoDnVQE9Lv7`ZI%mq#5+An@W{tNTARN6$1|QPnbPSYg9La+9fs9C|ese
z<Jqn)R>lS+#5;z<cYc#G0-dtWAGWtLrFQu4*f_jDF0ESpgBr@q^&$cOh*M~p<b1(t
zd3&C#1)A@3@(@!!eh+7LW7_s(zzB-u`IC}@y}RQk*MGf!3C>J-IVr*U%&TC*u|K0^
zf}G~(aVqCOX;9?~8S=Gm%SmpysOc;p#n2FFL+_-qi<KKvX}`DTGb7dw{e4R;uXAZD
zz`RwG@dtyo=+V7FQdcM+GWiBILj12{8vS42NyalL3#a>6T)j0v^vhY97Ye|3{<X+X
zA^wT)?;DEPpj^N&TI5R6kd8pieoY)&ny`o373Sh>p8cC3X>-o}cd&TX9WRcHtwxh2
ze~ZRP9M;=crn5a~#rp#|M{`;hwOYO^_%dziUVV)v8hoIco9tODo1YCDXl76J4dkHs
zJH|uIgFy6&y2vc=8okc|G~xbf@a2A>Qf0xX(+!2Zk80(AxF-L5UPvX<_+d%5TpcgA
zqP!(O(--EWZ=7YKE7&jl`82e-2$#Mk=OIx_xb!V}a)k5)RX6E|4p?EoHZNl`{meSf
zElFi(<2+}sSK$)C+7E*{(ZZzzZS(WOzBm*IT!+MBAvn<76h#v<gae?u?iK`gXeH4w
z+w#++lyySVTbEkiMGm?8*PW8(ujv2EWNZHjS3(u~Qne`+c9FSIXXJ?32E-%c{sy8P
z>fH7~=|ALqjKjTES@TEU0!YgbU~Rgb@O-&BENqDu=QN24{s@~={eReNU6E4=aQBoJ
zzI1Eo9rrzlHmvAQ_pT8W)21an2KAa{)?y|A9Zk|x!u5+hSi>x1juF3}FZ0^t#hMmC
zTk?1Rl`0FN@ZeBs%}*q{pq``Yw4`h#`F;oqqW#>abcnmkW{@bmgRNqTdHaKWnv?Tp
zi<)ShZ8BM%r9)_kR^mH7aE)~=<p2}_W(~ZNto7dgXzZZ3@BFjj4l}$Bm-7pX8Zs;1
zVW<vNMM@6fb*Ofzkv8Ct6&aPW;4k}yzGqr(cM0_C!BnDq=CuT+8q3!Y%mVLPn31l@
z5D*ZL+{5qvoB$1cx$<~~>mmKIomZ5(N1(F&ceUC9UU;u$?qz?Ai4JImr~!Cdmrn(%
zu2>#`Z36A`a6A`pZ(gfeo5zzY_B*8_>1=Lqr=v;|31BZsE;nqdPa^0o;xzjW-b08E
z47SBlhPGEj5|*c<lUoLJBZ%KRwC2qy9n<FWPk}ESn>#eE35?#=v_aeN<$j4l4(6aT
zV&CYNpDouTGUyjCi1~8B)6iSIvL3A4GX(q={PrkFOw>mrUyHOm+EBK(61DJlxa}Y=
zvpU3;A<G}b&EnI(YrI^Jjk7L51TBb8@jBSlTD^pj6zBWX-?)_H=C|6h#PAzI{WZvc
zM8v3O7;-6BI)Y;1_KG^HJ^EIRfvGe0h6tmQR)T*>D4pDx3XhMir#gllFyHPVBc%u<
zIh0{{Q4)<XW<r7YkbD6v<Tr`O`W4~hD(!zRffL^7dSQ2s`ExQjHWD0VyIM_yK0H-=
zbo%E82StvEbIzRlX^k6~3FTU1%ikrx)V^effszye!*-1NzCUcV?sxLX`kq=r*3*K1
z--hVr-4w-*5H-<v{8JH6toHLrqbHDQ{%2EJce3OpokkDSO`2=cMEx5L_SWVOMGqgj
zQ*U-5=F~=YXAts;T&lq3;1O%5UclqG{dM6E)qg2n1#ni+!>H6;%620zIBt{BW8Mru
zG0#P>RK_T=)G^~KMfa>G_|GE;A7Ur_*1uo1drGF38p_<^QW3E@9}!zoGx1zVLFb-K
zHy13$*N*>Fr3O2qj}Vp>9|2}$wy{d$r|=3x*aCzS9Hh&~kS6BdE*riDGG1ni$r0zd
ze<-U=Y|AifYm)6-Eo)bqks@)j6Y`?jS8~tbvd~zfuO)y#p99oYEMYrOrVxuTKXkq%
zdQ(gA)Yawm!!m77y7@&vM%W&6u2*-YH5aIZ)MSx_SwZiSXxb0S_J@S_V~BL@<_2(%
zkVMalhm5@ruCOY%ErBRG&ASCDhgov5@S83%31k&=CA#qzFb$Y8gbK&gCJP?ur+n6j
zScb?LO@*dr>ImjWXMO>#pm{~#CnB(8MdF)1=^*bI3CS#>z+%>Byr~@v=zrpf04!XI
zpK=_lh{0<a@q{Vtodsz!L(DvKdJZx+V!W@|g;cR$L)M5^Rh=0?@U7$<iDdlR8w^B_
z))@5<WAaIwGXqPSGv-Z4{V3cM9}`LO9OfH2v|3}IgWr5^TfyfL<ZJkhGKn<Q5{ZG4
zijwVMEA&0qp3HMKORcTAZ-2sWLndMvc`%rT2aYLFG8n)GK3{BsgxKJ9JI#;B!Y0CO
z9|!qmJY|3-k{d`EQ-CObw`(gm7UEO-^|=?BZ#i>lECh@VP*o-AD@r)(1tWldjgD#k
zK}+*BMMMV%5!m_}!(R44G`0JHo9DMNDUT~4yOG|iC?^WL^W<<c=8!Nx%F~?e4B{K}
zr&n+)6g$d-ckU_+S~R%75W#gkO%LlC>wRcpCpbZ&tw1EYJ#W=mWT|tW?TCwVq%jHJ
zZ5KS~_XlicZ_m(Ff<)Eoc0DP3&ouUna3KCA;!{(JC%^V#j@Ed2+jk$ox}cAWLT!99
zV{S6`Sf5E!pbkhSHnkJeY92(9pB&KAeQ|F$V7FrHKUySb>|r7t@3da`;o2QjYJ2-E
zl0eP3xq|9^GmNV)WVzfw{1vw<x<UN8+a4;={(xMLiS|whm7MPX8G9BjL99`5gKU$x
zM#@>%PvvYEccJ>Gs@sPY-40sv7BDqlLtb5#lERvjv2ldvj|V@y1W4oI5q?6foBgkH
zDYz9~77WRA3H2zcWY}o{$s6S{99wPto*QrJpn(T6L<$P;6w-mrj^)^~3HEc<$@Ar#
zu1BGaCe<E?A;(<Ir41`B7_b<oOAFJQlVN9{7URam>cg=ii9d_l>Il^PFOmd`>1Qb~
z3pRU>UF}qc89<MvaT{YR_+a8OG#G;wZj!;{F;X<~G?Jngx)%R(o8Pq{&G$ks2@6;o
z_5%^nIT7igk8d`?7dk<&){Hm{^Y6MEmpLb6NAb?RV#L(-e3a}1$K-nHSQ56BE?SS-
zzKRfliD8b@0NCiaPLNSckiejZIHIcS>%|)49<R_GW>*uDH*dNz!cq)(u{KS#VmiE!
z#z$qe#Kt&K+kHG1Um>#0Piq}wKO=`<t%W;Yi=-RhpG{6`<y}m~$1DI|ve8&mlVPxt
z<i^|K)%Eu6gpIP3g$X##eakK9BP71_oI%LonY2<1uF7D~HA|i*C{WE74%)kA$@Eg!
znPYM-mJl@_i?{w9eb8vzh22s(Ln_;cIm9g8_BT?FKvXFv_*g_Mq-vpyfK$8>$y!#L
z5HC`{Dt5pvJN{Q(&OJYw_BM@;#1qD%k!V==&6QG%_nQBu7E_mv2C|uqtcIY9)u)?u
z5sMY0Mfw`Z@GsxhOV(Q6|9WU)zdO;8+(8lH8ZB*5YiQ-W8*FdZMDP=0dd^~WiJm*=
zt~Pt=;N_Vj$9W{M!4vHQ`Or;(Pp*_=$=ODDBkM23%0^kMV@2tJ)=W7|KG%?Vh5QM~
zBj=eVy1zJvbxCgG*l|2zr?j|uzMwrE+R?2008;tk7UB*)p8qn#u<j3QIHQbisSbu>
z=9A~A=hKKPEt8%I<crIv&?w|Bx@rMVvw05G9j&YqoDfGX2N3UVk*4Re0)z!uK7{Q~
z%e*iDlLGaxgh05VyjW)Kcw=ha_FGJwCasT<KAGY=HIc#g|7{h8^phzC-ap6Ma!1hH
zQ%Y&DVJ<|?tXgs931W}lWc75o*s!b7=x^NVm(!D~{OQC3<%dNk9-VxnJV~JtTESTO
zW)RiOr?*#OQCIlepLkTYLlW*!QnDx6k6QpY*W{yGaQ4Kn71sTcL4D(nF^h&f+YT{l
z-7@pp$mycb-`-UkUD+^UJRdqXH0!Fg)&M_~hN#9Uv0QeI+P7CjDaYO%`l8wO+81e@
zBsMvTeoX&kJa{IN%;DYWI$O)%Vhj@%uJcS1eUI;UN_&pK=Kq$xDt~~U3*~q)gagvR
zW0jxubv<$6<7IOYFm~Em-N37TJlCi@Lq1ZC$Y41xK-Mg@!D2-d-s*W+^1uh5P1xDZ
zx=_}nJIER;$ZyW544R0)mYXA+dc>Z3FM0FhuPIo5<s3|8CT<7%HQIFS*5I<=jE`v-
zt*y%c(_sab#6Bua`t5F+J29Xc!iz}rJBD4eW>z8pGu~ccK>W+co^5~{Mp86<2`P0(
zkvNV<88MPGlZyLr@fBNZEM3R{2<F#O$f4xKHvI=2HyQXOm?Pz9B9|^DLR@F~7txrk
ztMG2{+*9c3D!3Mmx+#>{76PN$I^<Y09(%7jq2*Yy_^^a+C}@a18J3tsGkV5DnXsyi
zEcNyR^XQtj`Ja{I`p=t91rkwir<v60+##uzb}>S*wI1vKvJB^%mra3@5vK!Vs5)Qz
zKxPI(x%&x-3=2WCh;6hn0N5$<-9|5mWU`)a+#XO$g{YP!XE_8PCpi{~AdTh5f5M_O
z<*Dk791XVh6f68NJVB?GLtFtlxQp?<{#8(I>vpkiIj+uXRt@t;heRS#P;ovb%XpWa
zdJ30Pu{=JH(2$iHQWjNI824mVzT<d%o;=Ow$5ARFddw{z1HYikruCQc99Uf^(w8cP
znh4FD*JVlT(2hF04f&DoQ*w%$mIOYl?3^rcGe>e)$?T-J1fKX*KY)`TS+`t;+iqol
z;mIsxo+5kX)byt1*akU;m|zZjd(20OcPS>rCAaW{cUnf6CRwPR2=+ewrIo2y7W-VF
zFS7;ROoKQKJNuVuOz-*04Uk0oNVFL3Gs&2L8Gh=7oE5ir@4{k%K!#b1l0~(2u0FP#
zRfF1t6rL-;Vpogd0$)qGX)Z3?W*d+_err<Xe<3W>bl^3cWsSyyFJ2TWMw9<}NIV~*
ziQ$ZD{ys+0Bst9nhuif+JFy#Wo^0?g<R5{X!!vmxcbu`8f1~fA57SGp634IDCaAp;
zza;?If@S_G7n7&jxf$PbT%M!)+WYxSU-f=|?El<Ec)H)a<ftaW={W{aH{+IGk6!PI
zZrOQ%q!d?k_J9}CdGlP8$vd+hds~W^51*e-gH5G9J|H5Bm7Lv{T%3_J4PLlb)}87m
z>5Y3&j?IsouNsM;Z`kKNO1HtFxceUu1Kd~|^xTRhcc#Vb;5k~SsfFAZZxddhLo0G}
z3XQ~ZBoc-(SrF)~O4ph4UCZZ*ti_wXrY-J2jmOlhiis*74U^XGDF#o|;_$zr)y@`6
zw8?qBzn>RP5rBm6&UEAVOJ<k({k5PV(j+1pV|zP)`lHP(b>scQxOX$9`!*+qW8l9X
zl5jG0%>Q@h5l6c*)llUKpMT03e3@C%=U3Jh3r`HSr_QCkk)ozTX`djmh^}FCtR8ZG
zR+$m09T<-8{%9MBmt0E=zEEQ%PkR|m(%sfdJkBxhXuIFF-ULoL!_6H$65z-;kZK!_
z1a9-*N96h_x>)h%1)TQ=r(d)K=^~jl86(6e6q79lPbq~DD9ExIoj{%4*L6b|9@16X
zgV|}qEQj`+s8Tw1M}1<0ERG>reUIxh_04_#vknTAnY0br@_k|l!!<*`sA<3Lsk}U!
z@UkCvjE;sa=?oLjDp?*_9iegDH>;F$iE8D6Z{!H^dd(}?mLaV<>&K5c>C$U<rDF(+
zBwcSGhT&W4<auN1z|zabZ5Q3#{B2P4xZm_@d~R9Ch&V3ZFk%$`cpJb|x#HEPw)vvE
zGgX3h+a`dSulU7{PgMB0whKp|ZaYOSA)@=XU&+F~d<!&h{^TFpJZ8{4F7u0)dEhqN
zo|=NMK0%0h%=<cW^!b!8Oz$#x;lD-nKT|YZfbqiS<%O17baB8Nt9%Vbm%X^F<TNts
zizBYZSMI9Kn*{T95B4?m%I#GyPd}~tn^@Z$K9Ab|=a>%^qoD!jRaumQ)@)(XA$Dkz
zT(^4T&z~#)WDnRDwi%qp4jcf1$$LhSF%#wxlY8}q2Bw6@OQdw8t9G^Y4RJ+9nT-Y6
z%O1u2;bw>3cGDkqMzqvv&h@`@i&zkP*uSDfu@mtLPZfv!NzXCNJ2|zh{CQRV-Lr3M
z;Wdz@&=+h5z76>Bb5e|ZO-N`7ZT+cNYk$-(xkE(rpgMI)(I^#%)(-tkDB*fTcPXhp
zWvU_bZ*%x?;7BvzwN;Zn+nzX3P)`);;cv~*ed&MKE?<KJ5O)c_tKazVES4`8si5tz
z+7p^lO;7(^@D+St&<@1AwW_~+536oHW3f=Io{~7{%e%j{O{KgMIwcP0jnB2&BU@Vj
zp?$AsV#utAEP2q<t_=jfmeOEJ^BDj@8mN=3AX7vR<P;3va5gGRbWT2`KxH5p+cye~
zRF=A>5$&W8?2<cu;5QUsDT0bA%-=1sIb}*UN|5Zo=)*+hQO{{8Gr2fMt3>sBO-h#U
z`lPO986`baZ#KaN@%x(30G!*1gOlu~1wG&M#uUBg056Bpl^@=B7zSsPAADrwOX|h?
z%==Mq`?o=+f0Y>lA=aeN0Z!%Q&y!LnBGVHP0*irT+o@UU`+OPh<7o+Ou`%iv#h|kW
z>rLj=$8^D8fw7SK=;7J-+X&I~J{6!$SfdwNi{UOzQ2vT2(#g=~yxi>SjQ?|_q{{q5
z`(FbAoJULXX$^7Sl_^O-myl|I3$tfb>NGU(2xK?Z=KSN7f(@{Lx>MY=MO;G7F>K!z
z-Dcl#iD@YGCE^3Y!`V6p4%P_b5{`a{<?iyr!J2-7uSaIUey-h3p^~S5z3M+%s~4pq
zIY<6Qd;}M^w(_a?_bFVylVt%PfVXye{5K|Z0p@&m)Z(9p1M%+WXmR{YgmIqK?Vj(J
zM@J18CP9C6S;N!?LIboiXQk8h%yi^JOSc^DLFzs_t*b%hjtIu`hNp_6<pARSbhFJx
zC75=&1U$fPwvXy)TLIptk;wKMxzCBGj}g1B8IG#62jnTUY9M5DhKHSC7{CSIiVuV`
z7}*eQvhMuL9yqT#{g28k!A~b_<f1q0GB>7g`dB{IWl$iz2o|JuJUm%mzTSZ+2dV>L
zQEWu8EL9JzW@1$jlXLDdjKH{?Wk)<WPD;mL)282uUGMLdZbn*lZ@hyCYld!x;PGS6
zVL;Zal-I?kDwC3Cu}!P}#4u;*NJ2y|QchLSk}M}fGkZd>I-;(;o8Q~5Pi!XcEv5Ii
zLotVmMB92J%aVoMyogWLL7dnwcfP$mi?8L{hk_~P7fPHD^@5_@hHP|J4WZt}On=i9
zQ%CstnpM(7#dutJlU61^2=elpvTXWQxmp2+<3lTr_E$E4ROWJGg9^c%@OMmNBHc<{
zU%|z;JF8#oEw%<P$|>{2c-Eoc7O|@&k5f)`zriTO=`BbF2evcx?_>0iIKAimc}4GL
zC#NJXniI{}<U}FT!QNgwquM$>M4=wMb|iedDvu{l?4W@C2L*l}13n4ks6cvT6Er}i
ztbj9P%Qkq}1%O40*57v0tMK*wI{r_$dkv)k{N=2RD><Av^gv(ZM!PY1C<Y#8Wol_F
zfFUAtw1H`LOE&ih2)RJ*nI?Hw3TkTK&-W?cSnI3NA8sn*?l}sd=wTfa?E%!34h(<)
zYUtu*8K0#2S#f`OnJMUg==iT43Wc5&YO)WL8+M?5zOc7AI4#z)Q^|&IS=1_SRQDGz
zKs8D!fgpfy|JJNUv8leI3B&Sp+}>8_LDw5s?hpI>D|x;Z$>2{`J`HFhlE-+GF!<ga
zXs+*pceI)d+AeoFk92AQs>3m0x2SB_RP1;euJ+`AzN7xNlP#=50gBf~U_7@K{1mI2
z%l-*CsG7lZe?0Vc%nxxy7ubgq_LWae`ohJ+c_|(vIpgQvEO`-UXFKqDq-cE2M|)>m
zzL*-%68p9`aaS7UeRc*GKG#jPo!trmsL6+%57dvN1a;4YFHWv`UX~7PJMQ~!&v`9c
zGe<(GfF#s6E!3dzZ|fvA`9y;)nQSTiLg$wuS5BF4t$Uh#)CWugKj!wJ13Z+|pW#7e
z^IqR_I?SjjS)ji8b#ONdmf}3o-ZU98-y1XE%b4fkvuLpn{y-rujITk4K-N1!D&I^F
z2X}VEH>}(xtMXATDh5|O(Fs|QK{`!DR^1N#C|^e$_Z0Zdfx`AJE6KSJY7yNoHOZ#p
z+*+C0Y`^QJwY_hiW#DU7{6@s;)ZvIdz*H~r&gZx~PI>-;B+u*Jh?*!*bw2wDO8D&<
zis|h{r$<V&<)b0dlm@sjj>JM8TD5)7v*=M{=h%6<w(iGEE=*>>m8Skx;#@my@P{oN
zedyIcBCmG;^A9>O{?BgRP`{gm>^`q)nAc!Rc55tlI8Sw_CQ9AfAR%hYen|OCA`+(a
z<Nm7L4oA68QB{RKw>gl=SidR`qiAW#<cd1C<8n;j0;VtexTBPv#qkqd+kKQNt_!b)
z^@7Ro&DnF#&atldqj}vo9C@~6-X|m=POO|_HbrSycrLE`kqMMR3(2mJj%KeV<!|S(
z9@-|^dk(t|M0n=K*qXKFy@$di#ll|izQ*%zC)p+4Z?8PRE?Y|`jAW<I#`I3z-o{!Z
z7M%r((@i-yZqSh~kfL|Acmb>GpGdr~@r(DyH_FVyW_>VS+bq!R?fz%R7suiHW@xz{
z8CIvW51-`xL?0(B#qw+|BTTtYii*ml&&Ra*-%ZV$i@9Kl?{#I1i$fnj1<|;AOjut6
zg}a$irou3fbij@gEUR6u`zQr;Mlk=ATDe`mlpPbqRV{iLog(BP2|V{2AFmSLeT!x%
z;<J<be%5Bnk8=E1;ua$<RO&Nt=?lx*4)(;tw%qzyKr`|7tqOW2%W~d~Uz>Z{TE*7G
zKtgA&5zq=-5h;Mal<JS1b<fI{rIsLd*<*!0_410^+b%l(JGFqB{8|TN=tCRj>$0cS
zR=;iPz&p#?@PR*1kXBO`<B0^zI**Pro8pY6UZY(Z)`O3{v&4h}q{ASmcLX@_`l6!B
zj(Eb9hjA76dtr~f(m=f)4T0s}S~549VNA{f1Hxz$MCFSSsfW41f5x!W+f~7*ux@hq
z03_71w;$3;VMG8TSkd44!M2SBXW>ZKU*9etEeQ<aalAmX?Mbs4IjgCDy?7tXGSBJI
z9|aF~>~p5yf^s1rwyc+V@2!>n4x#3^on?vQ1R;{;!(WHM2(v8>E>0~5u=l*uyjU@q
z0%VeEVryW-eGNHWQ07XoAFmroP^W$4C<;n=9t*ihNb9PdN<O6?znqWe_OD7loRnQX
zFtCedt^__N!ZEQ9HkoVA!-|)wMUfyqrbBklp-S@D-baf>Y7cj^j(A(`FcgoePUj9>
zdPOl%N)P2ev>9A5M;(aUkGf6<o6aLFJlzt{%toU$5{Uw_4JO2Jn$H35OJ_1A3-6-<
zORcGmDkF;BrRlGKxR_43cd|a!2R3UB+F9)?s$8~ah_`pO04wE4Ccmfw6;}O`lLTN6
zcn%?g{hheUP+C+?(1J8&m0{CC(8(z!UV+-r538MWMPO}!dvH}dXw^zwM9=tmS+#!H
z#!Z;-dw9G8<NhFj!)M`cY(LLQ8y1&!I63827DQCT;T2(^(0H$||2+6QJi|&fFsq(I
zws#p9&B9aG8fRvPS?{8wAU%jn?XiH!M5ENX>8?n!gQC(&)6yz+2s6er7&7QHkT=k@
zAv#aC?b@<x-D+}YSQqn6)((6o{jC1f9QIUC<1cH(AD5`bd7GZy3n5RIrCJsE^>|MG
z^jL1#?NMpCTavL<2H;9ASBLzgA1s8Zeo+NTZ~!b4z#CdL$as|leV<&2rR+s9rnq%f
zpiRyOAiutW1s4O*G_ZdG@}d0w_6%fHH1??TDDZjgPM!K2(Q9K{mezFk-yxGhT1sp`
z5+&=+J{6R)wQnqC2r(nus{Sv0+nn~AgGglM%nf9-B#N2}e><&ezl8`obK*`0VG>(y
zY~Yt=VPZb%07yc<uCB8v7Ut!Da$`o%`yE=47M4q!P6%MJ>|V|lcEi7*>)|V$PmwLm
zbN=#(XnC+0j572(e}1m50%Ul<`{n9eN1T9T826Ck_}2{I*~|#(O)p0^#{8It5^pFF
zb&?q-ai=CnK|F1gD3hD_k3kJHU7j{SUn^007#_v0QO<9CFrq5_?qTZMRT<7-YHpJ)
zT8+_pVKgI(^@+=(wal$=_oA4hN}w+#fJt(Z3=3aOpeXlQEDx_v`w1?9h+wX`m+ivb
z+)ZTY>t}c9n-fNE`Jm+73>4iCCV59~xd%SAO!1LyGDEF>;QIW|_rDAl-l`WA1Vg57
z^xneX7r@l_5@Qrcs`J8^7UDb$-WeK4m+XtJ3&KZ7S6+^$NZ>!&X`PU!KTz#P_GAY}
zUxpQ}?a#t%EUg?CV!M87FVE7`zVtM28Ws)S<jPCEvBkrE^BEaA=sg0$|NlpWAAg<J
xk>A`7AtOq~2ECaAz`qv_?F~C3qSUCUM{7shv@F07{vSkdl;zdrs${-|{2xkbS>yl!

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_custom_tv_tensors_thumb.png b/0.16/_images/sphx_glr_plot_custom_tv_tensors_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..8a5fed589d17fe6ee62d6a9788519dba471fbd65
GIT binary patch
literal 26794
zcmd3N^;cV6&~<SuCAeFl(4s|(J3)&T2rk8;P~2Tha45yKc!RqHEp7#hLvXj?ZZFTf
z*8BYp-w#<U>z<r@Gv~~mJ+tTR2sITsTx<&LSFc{-D#(M>U%f(#fBBGMqQ6{mHpumQ
z^~%~w0VJvEnR)nlooKF`yb|H}xA|C<^cUE&K&!B{#ar*%RrUI{fOnuSCL$03!--#|
zXbLuXBIIR=G00EgrNBb5!{Wg0L?-(}Cu?Haf+0F7%3dlH!I7q$mL$Yecj9<`jSzfT
z^D5Zgt0)_Z#Ecth&8aB^q#^tt+jp*g;+l3Kuta%Af{32Sylxr^V1hg&jmnR}mkXfv
zAbI2;{|#XPvE=`E0P<S?^9AVTE1Z82S>eA)C<&;HFH;EcKm_ss4Tt~~BwnThPbC<U
z{y+Qy-yssCdl=>g5(+AS2SjXlQXBFG6FYU#Fiz_d=HGDi{5xv>!;RVHaotlES;<Vs
z54)3=h(~4As6dmO)fSx8D!+i$VC>Y{+4GrXjSnu(wiwX$)Rm{msq2XKT(PDE-xEgi
z_(VivIE9l($F=w;5%X502cyd9S;o)?@*@IUt%-#RKi)is3GxWf#e9B1a0p}-K^GkH
zE#RcLH-hEomo}wtS-`xST`geg^-vL1i@Xk;hL(*ihF<qw7%c*?Q+E}k)1pi-WN4&J
zrc9QIvTOJ=TWorN19X#w9YvE^0xfTy!~GL#58!=LRoJfOBeNadJQGEW^>VL8OfGro
z7;=ZwLXX%~LDV&-I)%f^Zc@iBUPr)DOdL&SY8Fe4jRr0M^?OY#0}B$nzJ&L1In9Al
zQOK9T4G@xxBY?1&=&}M<$uJ+xfTKBW>PRAN(VR;B>we1G_F=pv`)g>`Z^uYRx_jEF
z<Ml1t4a&-D$31j+QCSBStgUGDG#OX))Ly@#cs>LgunT}hWZL=J)o}-tSc-H43?kvu
z2VGc#wT9ESOGB&rE@vy9Zz3|&#<T%q7(yR6b>)3Ae*}ua?Sht~cbom|GLpPlTSm&X
z8g90~9$!qLFeLh!xc`P0$(R$Ja&h&6EezoO__ACMkpyr<B+{;|AT;h98iu@OkGbHl
zGYQkX4NM%Rey;N@d@DM~2kNR!mU_AYWtf#)e}ui_>*-p^zc6Sxuga|EOJ~>i$53|i
zk3uPR?4WZH-H$~ya%J$=ipQkS-owRa4tw(P{%YPDc|taynH>AT=I|^o0}49wU~C{J
zauK2=H85m@3Fzliq*hcd5LX*{uUE#$CpBIVU20AE#`cp-cs73fa2V=^W|F4HJO04v
zh?+`+`1w}s-C)?|2Q&Bi)%tp}=My{T_L}AL?X!5%*(mp1+HX=ICa0VpGHL53#n?|w
zrsVN!1h2aM0663#mGa+wr{K=E&1QL;2u+|ml4!Vcv|H~i8eectf>jNDUkq^angTdW
z)#m!O%sOuC%7f^!Qht!qCw#tfe81zkyyL#c-mQq;u_=hU8Y&X0P3ZX%SuC=j%cBp_
zr)IOhDL^juo$ys}!om$%Q`tjo6J*I-hLpYq;KY(EwP$7&><+R6Cs-uf#l&-#?&}Uc
z2}fsB6#K32O!O58@ksbMFteP}G2(2lnU(Xx0EQ{|6Fyl`z1ErflfT{pGkLUjTO->;
z)k=c&e_nqG{-=f;xxac&YcIobydWM*>-WCPDRj}<`M?J$>U-1W=e@wPjOqn`WB6h3
zK~jXSxO+anw%MV<(bYq5#QL@?&$9}E9mOt!yYPhfLyOPk<($)-DALZ(Z<${Ys2e!I
zEf@cqbht`&4<v}sP#}Lx4n!!n0Yy~ga^ziw<gSkSH*`h*QC-LW26MvjQwQ^Vr$sI=
zO3>b(^c=S@?tyU^-%7nsg>47yyUusdaMN|<O`K4>9?WoX_ry|GMNc@rE*&1ZWvhRi
z_S3m-n`bk(@O+$m8Kp&CbTqx#!3Ro^Y+#WD0wD>iyApzndvKsR&ejCNHW0D$L?Y{&
z-+!(ex(HR2t+z8ezJ13p%e#_pEzeK4BMasjYdnwJ?s{s>RYi;(GFI&e<aS()?XjWO
zoYY<N7r&duY31K@BufxfDguxb7oZWLurjz{vg84UbI9BdjF6-Y{<Q7%utyW)+c$+0
z<MkF<9?8fp7m*z+oS$&oEZ>z1i>3C&Ca}f(#daZAeX;DyOc))i?10TL3cyU9coSu%
z)362xbDc4*fSbB%;E}Qhj(t3Y;h0y44|>PJ{Wnh8tcJHxpgfb*cWZ`3JyR%I1eX@D
zhJN6ES7k-g^nBRA6+Lz{hr;f=v9U%{mDsRr?D`_~Bl#>!q2hGpQrdVLN8GXyT@?Oz
z+7-##Pu6db%j-<<)!r^-I*w+`UXXo=8fYr1;~kdxDdbuG7#=k3gn<y*eN*f1KF$Bh
zuM!f73PNVVWu(3KcN;%3OXZZG_KI`Pw|e|IE$cJPXsj;kh%dz$%0ABb0gso)Q9gp2
zNtf=PDrDsl^fanC@8EI&q(M1hW$#)lTUm1o_)})e6B!%S2>wVoy*Pse*8?CP{EM4h
zidh=Lh5n)*Bd=X5LAxl_{94L7yp5$zMl^#3{EGwW`CdAsJ<I9#=;R3}l<zuf@=^R;
zn<3z5TIkm7d*S{Q&!Lv#!b=h$$&Th_>ZjClo-A{sUr5=r94uSOqPi<ojUL#vQoR0D
z9@tomD=TNeq@pL#T4Ehm8}_x9rr7iysgJ_9?J!1H=qLr{xNdprpE-Y{97Xs^)JR-0
z+mCbAPM4RgJda+_w1ZhDF<f#6KMK?RRs9%DGo70zPi6-^cvsZh<Pp*;9q$hp1HB?1
z1Thj8DEHa%3MWR`2XbC2FZ(&PJ%6R;$UL|(@;7M#``OoA{*nKc(^Xao#r-E)ODh-A
zOQX3n*Lvs2@mM-*O*5azC5U{B1MKB77dxh5mKt#I>$Bx2XqwJ0Xj>oqW`wvg3e07J
z)B$YP_^e2>9eu-`H$Ilmv=_G3H|}OW)AT=Y3o&4{OAN>c@maT9IRH=16i5c13XIbq
z?_;?AkC)0lxBghxk!|c14fTiYk`u>n>GBPA8g4GyLuofED1?CZ_ogTPNTn=*7Rz>`
ztYLiEcuec^&_!PYFq(!2r=O|z#Ft@!biey30#@bucj&Ux#$iwD(4dukXbV1&1KhNt
zqjE7f0?G2qOI$YC4rS^$^olhr+rM>-n6e_o=%}lSws$wIa?!a=?|&NpC0eU;zsWc{
zKWScWoG(_FP6kwvxFXl0Buq*6*?NA&t1N%CNhYQxAHE>%%=x@2fFKMkfr^ZzsGxgw
zHOYm3bg4i(w(q;Sv}7U`6Pvh6Lm7Ptpx?d8)x>^1RQ1|vuG<vm`}&L2^y60pzYXJ>
zD<jpjZ7a;%a|D&NxnAb?nXhfne1x>|M?Wx8nh#9<4K<gGSUSEv7K#rQi7l9~Uj!&y
zKe5Cei&k^WbCf;y%2w7n*r#WSHtKp+%roO8#6_LwL1~<?Pvf`t?S*N^=01T?r|(|z
z$d^E$%(J*;d8ErD{ZEFps9f+WKOzsubk@s$VbOjhU+t{w@7U3KK%cp%uNDtZ>3^p^
z+tzr;24w)=wRU5Hn4~bhvixzHRwBbhqo*XVUm->z)SC1la7x*{$-z(_ak+M#+ZP-g
ze<IxXQ4aEqNb_=<ma`Qe!{XjvZ9eq5Ib;szIj%cDn`nu6F$klktHN)8IeDy8S5qM2
zC%E^-H3a_+ex7p9xz-wfTD14=;J)uV*XXUj)pBRk@Pnh8CqmCZ&iJ1QUo^Z)B@vIg
z(FlGVdndb1<*-47yzX?oP5~QapP+;@NEQSQ)AQC1_Ty>JQP$k`X`41*d?oW^%4JZH
z`2Ld2K6XjjT<v??>~MILbr{_xL>OfMd9;zSl*--#DwbWDM4Mp%J~N_%)nv|~hmH9@
z)4zT59opehPLJ^KJ6=VVO6Ho)jzT3gR98(NX*61GGVm56{>rj2M18)d^HhPG&iNrA
zw3~jo&)!0v{GJE>iXI;ylLdla?z?dM+T(hzjWMg#Y7nngVIY8XR3%u9|L|1IzJP2%
ztaEj&!l5_dWLvR#((N&`TQ%Z_aD4=0dIY1oPwr0rTMp@~krDj&qrOo>N7s>&+3e2#
zj!pO%Q}P>}{K5VvO(@IX)|C%9o-~d!1M>GPa<ndYmNv;aFk~!<is==b3_pJ<30VRo
z#OB^{wUn=&24gx1tf_VnB;eG=J!0z~bgcJJX~-Ya3?h@$yfOIL)G*(evb?a%VX!~f
zqp6{~#WEH-Q1?9^ryjP0Z$$eQAyAnI&7&uR5P|1uwXt}Tyv%;tYAc#KvBk|6x%}OM
z9w~hzs3Y5-^N5AaGZ~c=G;%D?B(WRA1c1N1778yvhoaD%3rgBv|7EatdJt$>$u9}~
znp>y9i{nXjS1MjmdxDO!=)mKBA}-}mBp?edtY>D#G4~`p;ncc6ZCuu^9162{V`}LD
z1V6Z{i6Rj#_U_ZbKJm)mqBY>vgDo5E5h<Emio&5$E@AF?G3Le0&bO!{+=wz|)7zrU
zih#Jkm+X}@+r2~Gu{6>b5`g#UiRW{W{;Yv3BoGE+jQunVS36Np2enN*G1C~E`p6Rz
z{j*O=Hr16UEyfQtZjl0gr^mQuLw~P`!;^`G!)B$eL<AKUJuf^<NAv;d9AiDUw_1$1
zJPo3x5S=Jc8|9~ObacN3AY*5Q!3T7T+fx>?HKwu<G$dMGGeKe#-WS@H^V?x)LO8fs
zqmwi>JHg*Av%bEfp+&rI#S6h3Du-ZF@PZN*czynUs7CxrIzi<4CKW$DVUD<li_Y|i
zG17HQvoz)>Zp4t@iHd(8;)$gN|7tcwXZ*}bVfvjelr52^Z_wbOI56$NP(NgU%iVHE
zF4%~Q^N)t*gCUrPDQc6o=jm9zod2D$s=kAl9(4KKJVtQbESUEoxkjYn&#4)gw9Hf$
zxkomp*C}$oszKqQn1bcxZ?n$jcvSPa-%+!QHfnAo#hP0@4Eu7#;_sn^YAFm}a5$kj
zR>ve*17@Gze8vcL$w+E@`m5(=Qy*Q~>OI;1*l|C1v1PMWOw3eL9{)(8=9b%IM}6!U
zMcHwGU&b-gd%E#WWQ1W&us{G=3v*Gczbd>nU~XA+RYmPvAK8xKIjItpke-|6DHLM?
zV^m(Trt~H3n8Djvl-rxNw5%%KO97+CBR|c6-1xk!_MSCj{y-=Tl)@CO+~o2vmxWhz
zY)tX$O0oEU#QKS|IP#<h`b6RAJ-6|QMRwxTv+S{FC=#_NXFVnwajvlc^v;Fe8pX9A
zhf<6M-uxR-uogNQ8hEK)IoW(Fw~$RK7de(kPizF2`4M=?^(HIkpV8g2>9wssiTL8{
zJD#*5d^it313T$)w|h}CIq*lseFB}(#~)6PjENhpg?^8Xm3F>+oFgAQ0$j`746Z8!
z4T=vRkY=AQ@V*?NvGxTdbw)Lbpeb#Hwn$jEkRl=kfxy_mFdnIZZ|v?Bk47e(MA6iu
z=n4|<iHvkUa;Ob(KQEpZ&J@0)KduZPmaKWtq0k?3_hJD2Gm*qyzI97E{Lc|!PSUoE
zK1%~C&UNiRN)zQZhN-r>Yw5OvgD)#Z`yIX8TL)kRf#j_X&X!Xre9^FbrGeT&ua~Ez
z;L~rWd6P4b4BJHLCtjp5IQbK@Yn5nt(<W&i`^gZ*x8>u(A%E*th>PWT9zmg#$i405
z1u)h5>|%j;2S6YL!d+3OTS>AcGo{_X_PRbtC^cZ~OZKy980j48DF6mKe0<jN+^Q!d
z6_+wYh)0B~28opBG018{`&UQC-Z&pbLmX%ur<=#gp5NWH^gbnTA4PSVe6dX3@YPg6
zDr(C&J6Z51?||W#t8Zxn-kdrXW+H?^p#W}niWn)GH&MG$2$2AjQmXqq)5=*Mm#Z+(
zS55!)cU_L+xQ?>`ShOy)%)$Gpx&xYS;yMa=?0Luop+4>{HHg<+3oxDAZ>#%N{Z}PF
z*y2V1MAwfF%&zQdI+R~N;zar?o(NK19$nQ&-1jXxMS(~iDLANz^Z7LflHO0Vw(S<{
z-&<r5eM|PgPITIz&u6(UmdqwH`L~qt)9ZR|&VG<JT1tSay74fpHXwRH=;&6EH^5|0
z+LuTHxkiqxO8upJPHvd7O7}8jG<V61p27Jt)N=R14)#js<10eRO&x|CW~Yj>6fR1v
zwYJ=7+&AQJkPZmGr9o>Y;3Dk51T1!Wny3eOhpSrVHP)r8j)n->0clR9O!IoX`MucX
z^Zwiy9^o(E6WZ-Mwy{?YbRia8Hzm{^kdB^~ULyM{?&zV&+;rnvab{}4=iyZAU08O5
z@MV3_6vpe|{fU%r_^adQ-fn4Eqe6TqJ&VNr&47I)qv~~M_?mkg8<^CFz(S?YiNc&J
zF8sSlt6VZ89ZVI=m8J2(?|e_Ih#V2_^euv=$5qy@pDZ&CU`?d!{!Xjfsm*EywtUo{
z7M`bz>Lm*<Ce}MtHYm{7r(J5)5q}n1Vl~^$H^%&tC52`G-pRCoZdtm7PEwjEjW?3{
z11<BF;^@L~F7Cc<j;XTn3-~YX91YzpO}BG$c$6|oZbk!Z-y5|EV+OKNAcMg%n<sVj
zl-e-={H)#pfrJC{pT?h4SGInUS*~;n1&|2O@-f=eNKvRZ4R^(dzj~9cl6p)6{Kv$q
zqIam>;&#qzeUCS&pM&I|_VXXi>@ekIJ%6OP-srY6;NEm+rQ*Vm2SP=dpg`9v=8of0
zhOkW8VtL8ieYY31ECO7=oqO;@Z;1Kzm!dv@X`5}nJ?l8#I_`)#C(=D2%+*Mp=89va
z5suRsE0+)2#jpPQZ^r2U4e8Tn0Artnx(_PB{~E_^f3atL)FN~J89#@h6>YEJ)SkOU
zkVI-6t3OWeJfD)pik~F|nR+pV9<?<+S1*Gk>ikPTcq<UXdw_)K!lcpOk1iX@!U|Th
zoEhV5Sl$g0E!rygh}-<LV39X5LUDVTxY1S(aK&!Kr*rX$uZujy4v|hnhJ2+@m<J{n
zzX?5@^gCzR)(UN#6V?@^t({?Bx@QIn1n)vheiuiVE^S;DDalwPZ+w=+Bbf|;LoX&9
z8F2XSC1R0>m+4>0E|mxI>`U=jt6gulFW~I2NBxSomP5IND~h{YMxG>1blL@4eF^{6
zk#UK;tX9dPQL#jH$V)FC!y~DdwEKH`o<?o9|BR|2<^vN!e^m`uTJGR!jYdHl+#*lk
zs5uVJf+{6hxY=%{b1<y_plLd2Uui24dyK=?>^W&`SzLFoKY%L61$Od-y6mNtoA!97
zb$UD;2B^}_laB+luIKy^Uxf-;n)u=5i3a`beLsWhyI-)zg6gb&zCf*CjB)ffyJji%
zQ#_YU?1e#Jdx!n7Hh(Ev_|i!PMx4H_`%&raZn?l8`iT!XKN5a0UHxpS!xGFhTNf+u
zxpL3g?fDyTpaY9;=4WvgLgnhu2&s^lz|K2HyQC6^rpuT~EyYofcy)zpZBfmg;sx&O
zRGn#3W%*3V<n*|z)Qkf>l+k#uA^Ks&Nc5APO_6`soi1jM35BHy;tl(LIhwN&a^a`%
zJuR6vsJ;uQlY9qsuOj!Xa9BQ;(SH{l%NmW8PkWZ6_$5iF;!BonGJlTSxyBnademO^
z=BG6}B=bz`T7Vp#E}gK#fA=$pv=-a?p8t<XnSlE>oJvm3oLf8KYg@E~;oi^ZJM)2>
z@1e%vtsSw=PX_(a0OjVZhHr*B{g*Z;W~=k%vO2VPhKypEAzx81pO<(tom_IjrLqcK
z@zif`H$2E)oZ<bN=C9xVdlaU6(E3^#$Kr3w>V2<j@7jJovp;i8k@<L1`K44d!+7h^
z;j!IeS*PpL6VpURLOzQr98v;{UR7Gn4&D`3s(EoL{Hb3^v2>FE-3#8Buc{2P?839y
z#~{t0<hQk=<*T18>*VWy@2eH+)q@uti{*C6u-6YLDb<8RHk){z#wL@!tX{(Ia1#tU
zO(0i6^t)Ueq*UJ0(u~F1x&}Xc`0a@7^&Z~y#d7ER`@Xo6>+bhoC<@-^DTsWJz$8O@
zE6q6A|0WUD9M91pzApN8$^s8ZUZ_$??^mF<Akg}=Aj6dC0d1XBWe6+yIu{JYbdiY4
zRV#EJzDK4PtaJ~0363$VH|~2P9v-U1O5Y`@`zw}ofRT_oR8hMLibAhfS|zpBh(>i=
z)P@l=?Y7Lm!mt}6Km%zR(K8vzVwM*Cph$R+Xh-GcA2iO3omIER78s5{b=pAhf9(?R
z=jTLCaD?TmqKr8O<U~<+4)>I!4VC|q!DCP=dfL9nazg#oyW^g{NeK5d*yo|ECkDoA
zntQ_k_T^y$F%++7ZZFRcq6{D9EFXT)`{k$vq&&@tPha9Z|9HA1$islU8DzR5%*4N_
z8DoCrp?%1ld;gLWA)zb^F16gFV+?4T=W$cQrr1dXp6_NT3tcXsDig~m+Mf7k+r5{4
zaX`hIACv4>RGTPbo0)OUixoJ^mShV!I=(iQq{0#r8*6Kg8efTj75)X$kJM+F@I}UK
zUT6J1kRe_tI*P!@&+7A20@t?a1+@I_mNWPu$Azzu(L~LJHNe5}vc>A0`ZhB2cvGWA
zAP_fmED`+(u^vqY)Gl~k>@lkPE7s{N6Gk9+#3TqD@!RmIUBVFrL?mpY%{jeTz>Ul#
z^|pRz2~Co5>*5Vwzu~?e8!^P_RU&lWU@N*MZCSemvk~1wkWh+dwBQK1rWMVqOV=%v
zyWG(np@yt5?UDMjtFamAdo2)+Vqc~!LSR~pYD@3!kZ<Na$LxL#pj0rAjQqD|-&1$j
zy#(nmP~x*E^M5Xg`E%7amTX+r3i*l@v57pn?=H|ni%Eevpn^1ymfX-wy$foPXGw0Z
z1mw|CuClWHN>umq_zz#BBQJsnet2!E?A{p`ONN@O<Arp+Ey-6;LrrroraY~_4`3M{
zSw}L+i>d?Dfmf(G#Pe1^hkYDJ<6WdPOs0PO-=ay0Yv^)oC(`N1#}Xo#K?6jVUH7&Y
zt>ZCkQuHIG`z{jqel&L+J&lkI(N??MwcMz<!|5hMT0kTAH8^+~^b&)pnX*vkErQ2~
zPu@;et;%C0=sU%49+J^q3X1pBn*(_u(Z-0MEB<C}&nqd24lK5gXP&vny^1Gezm2Q_
z+4}k|da748EB;qKA4<mdW-dcMU?lj5m76+sff54`Km3OuVccWpy=)w?d#3TqxE`h}
z6l9ih_x!f;a;+y-2LrFARpjx26^wNDnukYzyUV0&I13UvG2=?cXJ$kgj4+>EWwuRE
z#tsWN%OvjLGsAyYLoMu0+RDt)d<{d{D_#C6oX)J(P(ujhKG$ds$ObrB+81&6HEUZK
z6)Ny?G2BiOc@i>AX>0zY@kpJT=631_c&}0lHb_-=K+_#y@r3`Ov!QI~A~_f2I%EKS
zJ7emxmhe7MD5eWcXU=kDb2?}pJ=Z2ixgDeUC)Cr0Dgk+Xip@@Z?fW>sIrSfsd3iKl
zA^E2N1so#Rx-^RoLChMgcI*of<7Fc`lfSl9eUto$^I#>M(!@{8#<(NM6gWSTnNcQu
z#F(n@U$X%u^XVuNW%;9#eG{+J3*S0eDbzOK*JU^!PFZiL;WG-ghu)JglFs-`?gx{0
zg^83973w0~@wfomO4tTCaGoAw5ySDvK+M&*B-%3F6upN6405V@Led1c)Ncz<5n^(<
zPiF~QP0i%xt-8+Ans2bE6p^{j(LYwh4VMN%1BUMdM~<t<Q@y^cy+VT`F&9XO-UmwV
zNQSFdQ9@(h9NntM?`(8xNGQ6_Nxr;hVsb>=lxfRktQVj7Tbz4GRfJ5l**g7#AP{Ll
zg26O4>0syPBWE01LOh{Vto8R|+hzwxjMgaa+4w4znI|@APC<L_p3o7U4Yvm^QVm_5
zOVTvNz*$GCz`?SqIMSzE7St(9XFgH5M$3pE98RZ&THMK=&w-ai#*K*W-HmlvcPGZx
zE|2{<GyASOzW{5A0ac5FB?W6uG!rdIrEI7rRrbvHHSzD#WAq@ZMp<5G1U+3Ot38^0
zkWUd%2mn+s+69nZC|8j!jec@6cB^Hb=$2uG2_CMALSGUF22lMX_H9x|#-j&=|NN3R
zm3MsbUbp3kXQaeGa7qqy#C07rvSnB^KY)88=v<VG3e-?(hSrV*cU%;r4bEmZXYHnj
z=YJxI7X?am{GN_kF@k$GTAc#_;Ns8cBGR(=(_bTbRFcwwQMtQurq!C}McDx^cmimi
zLAKxbb3<jnqHw?7Ob(Pc29cn_Of$5{HnwD@ef{P(d^BGojR$WAu=r_gjLP$CQG%I$
z%4-rTB?K(yt0^=2p#hk#Ngr`76lRn^!7j&$f_FU~%2Iu9H{5n#WSYWDG%R%8>}4r-
z_j-3kt^W!$v8L)O)(h**BQ`G7xTp5~*cIxL$Cbd!4BDqq1iG#K=~<?S695vr6nLB_
ze5lVfgCMMs$o<^jHKS+&C+0>9AS9TqsXP_zV8f-mKeyL`)N8UjxE6ah@%CHBRjiYY
zIAN$UE$g3x=1lf{!xt1ZRgVcz_({~_;ON4pN2KG(x>7`@%V6tAaX}iC*iEyNuh{Z<
zP4>B7WT23ha(Y<Qv)`n5lm-6SK+9jP-Q=@WZuFraml@kW?^>zwOadZ+-><%ECjXix
zzBw)Eg|m*luIPw7+s=KKGbMdlsp~TSU^^P?aC>~6uI9b%gV=r|dD=H2?R4lgxOouS
z4y$=lE(0JSD=3f=cism*huq^ph>fl#eruAl&Qgel9&OAcjR<Alj3|8+Aa?D7u|RnI
z4cuE_e%8kAT`@G7_haq-o0?hACI10@peGt_iAj_kTckd!D#<{VT33EZe@1Y-4A*a^
z_LEZcNew!=aM4z|te477Rk)Kg#_|Dfz~k%pOW!YGvYaJ4-%?DS6ZBjA&d8C~ri<ER
zgh~TqKZmiqZvv5nGo+gIxf6*Rwk_>k)uVWD_8t)s7{ATQEaE!nr^~>g=%(4pS8+Bu
z5=G_(yS2XmS6Y<<>th8PL}|H=vh_+`rF++VgzfHwF?)POOi$iLaTu-X)KFZf9*u?T
z?)}Bqv}%JriAp(<V<mQC+Jv+oFd~ryN0m6QSZ>#fdiklAyZx}<fs<VM%#R&iP`U8S
zOCWg^;(=)Eo+T0tz=;EEkv+9t1OqTPasza(9LhQ|<YBtAV9Rl<t4SIz+SeaeFy7sb
zJmT+NaXYej@T-2JltU&JP>iL7=i?Yhzm53*70oyBuUiOR%j>5$SfU;Qg4%7qnxL^X
zxI<aYkVGyg07x>~FR3wjB?`nEIEh!feQ?EB5fxl_Xbb9SubMyRuRRP*Q$EOTP&drs
zY?gCRJmNK?Yq1V9fe#~p=K6_U2~;O55mu0P&seC38CLy+tv~Wp-x`q#x&>ZM7-Q3m
zW4eD3)V})uA3+zGj}#kgL19Tb|B00v_HN*#OGe|6oD8pH=^tCuMly2~^7SO4!t)?g
zy6@oxDL?L3C~3=371z7zURp$Z`m{P~37?4aXWtd&fX+CBi)^dH8Pk?Ah@fkZ()!`)
z1X0mZu{ka}gkauYDfc*`#Qvobr##rtiB{U4PY809iCvoe)JrGm!V8g_?{9}sjEN0=
zJ7<CK)qfT-JtwkE7tN=pto<cUp>$GABFPts7amvF?Wxg3FuLN{L~fQO2Ji_C{H?Zr
zfx1Mn>%xiT^c%1j?7^kj&;Ai(5oLtxTKUAKq*cME;>S!-zWFU`$Q5Qp+Y6LFt-d-*
zI|Yi&1D5sozyVt6Io*3Y(QbMUH6Zt&!WxBbM#)2%rL4sK&J$q7R-gKw`9Ij&^fvO1
zmLCl&t)><H@>E;}-txS})rj3hexI>&3B9Lw&HJr;em;dp7)6hFS<O(ccpsv>x)BpE
zDm*>V;wiw)K?<mVEhhW;@iA7j3ILNZMLoPF&x{hn&$8>Qc**$GE$ELU6tzKjmkL)s
zu`UM#;VDh!tL`#Wf4#+=lYGrdZfsUx$JbO|r4oTwXfXkrg4`=3V%Izv8Kj(cTDGhf
z58eQ=^xq<-^r*F|u>!{;_Em2^J)jAy{&s94JOl#v*YHux@HFrJNuq_KMj{cd4B!GI
zyQy%L+U#GslYhy*L&uRMJLVUA4D-9ylSREQjKgNu|8-2>jqtiW+McHya%Ket_zkuj
zm3$1|$Ks(pmrP4EkMi}jCyvG+`NH5bpZ;|q0Y&){coHggK%;uf=ygA9Gjg@-z<06?
z?Ix4=xt4ntSo@+9oy6cjU5elzD4ub<xZ(T}<I6I7?f4SEItGZCqH^<VV-82;is}N|
zCJ#Iem)VB}DuIbW^(ifRZW(vT?%x(jWuuHGjsQB<=Ca$*?Z803l=!O9dzPD^C0fKr
z2m~N1WJ$ToKe1jEO_ejA(9L6h#ABVRAa=9%1Jyi$K$hq6Ap+kBGrPmDUB7Lrbp7}D
zT~(ys+@NI%xgYT&n2Q#8OCR|4nQ!zGoaEPRz8-8C*(dO+(WTLQqg~S|jv4N|PBNb0
zjg))lyh7kNehT(miL}8oD{hw%^MCcSz(ND4@xi<I#8^W?mgCIKl)sq&d`dbMmE%a8
zZ%Hd$rWX!{XmXV0v02hdx+S*&R$ekJ@^R2dY1h%ElrES8zk8r$p`pt`@N(yoa$K8(
zjhK6o-347bbur0FPyBz2^m{=@pz-rv&bObZwz1EzCUD*?!EiX3xSDA}=rb9C8xwGC
zNrz1eL{|I#PEL6#J$&j5zf+699C`UqL5~2qV!2i!hst1Z=Yiy{?8%n+gz3Y*e1piL
z6O(<SfzEwKP5Di6F}9!EG|9*jG3lry?fcTF&yqX}`=^~vl)#HXi_dK&J=(Y|Z8<*}
z30@Mwy@}{(<;t;;0%r<52#kSC7Bu51jNyy-t#x>WSy9ggSr6AII%4s~kygw*$EfUf
zCY$#?+H2MPU7Dk3nKJ$X=vT#P&+<xyH8g+ku0O7RnV>45%*v>a)aJH8>pmXiNa08~
z#<?i9?rpIxfwN-`_yM8u7UP^;zQna2AKg3DlMak0P;y?5eH=PSdr}C999F+r^0L#3
zC^_JpLv#E;X%i<g;;`yFp%_PWU!s&&mSJg0X@WSs0r!qKjk}nBa(sb>uSsyGl@a`R
z?7#3vi8E>bKryMKl1od9W<sxngnJ;(4%kmLe@X$yzeNlN(!>2}*q@jjedoHB;K~H?
z;h{>mbHAOQ?yR})5R|r$vm|$Ig$ZnnL#t*8KK5}E+@@Kv8^(9yG?EfT5PU1;lin{4
zLSUjS%c8>N;~$k2tlzG=zm*BM6jW_WNrM*T)#Gz|3{UBi_BFO`gWIeuu7`n7qJ7!+
zoB49j8dB$kFYl%CkKkQAyfMq60G`QsWATY0(n^Qq0MqRiwat%l{@qJH_O!Wp|2ASV
zp~0AbnKEINW-+Zod_NkA)8QUU3OqKkv6{WUv7umiWmN6w4y`-SB|;BV;-wz#^2nX)
z8^1f-0CkHspK8Awr2)AoVa;dY`-(cwmj-#a^qWnY@txS1{hg<?>b=G>2m7{0c%!)H
zbs=y!>P<LW&QMLJHU2(FX@TL%w+0?R_@3OuR}@{Lg&|edk7qocJn&~JH@eW7osFA8
z${K&8_u3o`BS;wP(94`+cY0*KMQ^Pn!zqW=({HA?)YFG$Dggy=&6wNiOg1+<AFBE@
zZw48Wylw=jke|06tUZ<MZt;wTJkCAv8q;y;Hw%GG<FW<SZJt=!e<_wD0%!MN`hJPm
zeHrurlz8sVi+qX5ww`QtsWJQ+HeWESuU#cztrb5u?LN_a$E%6jk1*~Pa%#{$x4pRQ
zkb}<7AKLw!Ny9$X$MW3rj_>=MRpDu#0WZhGe0hmuS$N;@)mS|*4E~<ZFs$_6CKXO<
z1gc-%wS@=}LjMsBXkO0zzU&v~<KDHjs^+$;w(=)gvt-zC>72mu{YNq`RZI2l38VG;
z&9cBZbIPwX$TSBPKN?k&W?cZIRHxaR)9Ch}tUXnw4yeb9w9Y0HH=?rzg6qU0_kXX&
zikwQyBcGjclOqS5tHEx2(=0Eyz4ua<7wa614t|PXbnX@1^gG5j4}b>F?#wG?KKQUL
z>D3%@JC#ak1x48sS1KpB=kC_yJ!7{-np@4coMInIt48ux_LJd7=zo%8N)@-)@Mqn?
zCB{TDMLie03HEs~sahW1-369Q7k;$b8PLlKc|e60hl^hIV7!27_os@JOqQ{&^m@It
zOU+BY&Kd1I<4B8XNEOGQ1fxpfLKZXxc`gJZVp(4YLUrlS#O3!<O<<Li8^|ViP~L7q
z$w48`7Ze*B>F;mrX!*VfOSO6<2k<HrvugZWKTT3;LK9K#TC2QDfk??*7a5I^N#_yO
zlk5BFkcI9a+tRGJPA{oW`Bx4JhUDGs&@j*6IQ?(=Va@q^QQIdWfsS)iGG_#=im~(I
z#@c8z_?^~$u5IHC!|pN^Wg8orei*{_?;yp&F8Qjk*|OneM3YNAYj1+?PD$;rZwV=n
zzRaTq6k(wsa2;IUu}FF`nav--XUmtN>839IE#ij3B6`Z33p3V1pDGW)z||*)hzf4|
z#5U$te~Ik~F3J|qKe`<wNN1Otc$I~u5`XtwL{mgdIA4xb^cB@R&G$qUX|DV&YaYo%
z&L(UQ+fbU3rKLh<U@cfzC3O-2z?eQ_-(cwL{I(P80|tK|D1sOdPghb|TlSh~KPvqa
z>0sb25<>Dt`sQv!b)&q;?!Q=CJ{$~>URp+ihe!Y><Q%ep+4I|OL&zo!K@3$9_XK}G
z;>>%EQdotYO8c58TuoTLmkppGqebiz(tbA&68SL)Ym2PBh`wyUU1P7^o-V&xxQ%&I
z1`AGU3bH%$YNuf^IXQ&0!Tdb$I7(3)vaau0-GLq!gUc*E{8a*|5H8aRMZy`Da}U;Q
zk0R&2Y^ra(MQ4s*7$;THElm5vNno7%&0ev8lfW&rjt8_-k;O9IROA2H67|DhXL*}G
z{w0f8xhpks8E$F1j1p#xIWiXZjBfJ5^IaMxafGqW`53v4sVk*MQwPMFnyS<{K}hw@
zd!wvxfN`sRTTVb-x;RGfyoA$6%2mq}{0gGHfQRm0M9P^IUVT{HFSh1`bO^0`(macn
znI;iM@Bp<><Rc~X?fKpyA~tHnvI!PdWl@Ff{c(Mz{#_*v)u754f48v>s_2kCwF=JY
z&PSrL)N>3YjsOL_gI3lipNJ2XrOul1Wkw+f?&ac$$3qPNi@V_?<6r^H*bU;p$q=>>
z1HvZjJge#_N_vj=kpT^tx;Xz>wFlov3AYhs1rZwWA82QuRk~@988N#nN1&YOzQHa3
zdf&p06k#ycU~rClDyg|%s8er0F|}lCP@)Z{v-VzuK5T>5<=`+i$=M^z<d&;ru`drv
zD}9p7>g~GidW{KUn8}&u^k7nksXyJs5{jRl`tHwI4aOs5*?ya1z}BK`rSbW4Osuwk
zzP!7-44zPQ)BpsgLeCxhiNT!Za_>?f*CXJ;7qQ8#<^K=Qzj13VkbE1||LKI?N#Dgz
zIi7fmv}!*%-MEm8-?F^f%-T%3{*>l+mq_<~HYo0Jb>M|QdT}F~I?YSYm@dv}bQ}pW
zFz@r0I@tf#7ANC?Q{`j!@F@?F3h&0m;LY#5vfPc>xE`I4U`@X&vyP-}ERVw+47+d*
z8XbB2sGh5Hd+{W*%UR)T5f#vG^F8`sWtR@~8)B#C?QQLfjP@T}EoZ+c4rjM1ZjGme
zhKp}sp}+!o1JpBW<=)Fa2TtF_-3=QXVIC2VH8rTfxH&^ZfK)EC-4LZ7vc)gt$j;&B
zE7W-upbGnrXAHR~IRB-K{$r25*hy70%I93yIb>rw9p~$V7vkmchY;zmv47Gi0=ZcH
zG-@I9F(-ru^j;GAl76&@L6#uC9|fE$kv7K>zz5BI7}0z>OgtK##ulGp-w47^V_Q0#
zHuCjpJ4->8o$~7oo7i0@Y29LCCw36}NAQbYir!Nzr3tMo>~3U^4m{^j8Rl_h?YAiE
zyhGp#8%ummkQ_Q>=-qTpeH?NTNV6{LaTjc#b_*38J~BuhbgN0&^W0t-1FuH~#hhC1
zMn{OGvG&h*`xt>KB~YjZM>i)`Nsmwxb)F$$r<YPHwV!0V&ufqKc|0#JGL-x$z7jU6
zO_Psw{GVRRH@I>XEa__@Kz+APMRe*11R#?(&0!_hWIB<((OfR`TY$wQ>BK1O&x$6u
zr8b-zG`lss-&MNJ<t*)>83H{XmrzrxLCXP3zx616s_fK!t~=CSlew@PbyMR61D!a|
zsLmv7%{S${Gp_IQYMXWX0v;SV?H^Pci3At^G0=Y~46@<M7=4?8k}(Y00h#Mf_oLZt
zEaKFb;z6mASb~Eguu=$Q19<su2WT2XS;fbTS~>dK00jSvyO&zMzs`y%Yr2nYu6#9Y
z*<ebMtaZPY1pYJH@gdD`b5my{U|!8nF|1wUcOG>YTghKFZCC2=D}K1OtgE7My;a;3
zs<??3)@Os1_A7_?q*<m<poqUE`vs{gNE#3&pb~Jtpqr7P??e(>O4r0%Bicd7q4#W5
zm+Xvgm|hj6LwVVX7t5zeMTO<OPj`cE&#GepIHe@n2rhx=mL&(@5j6PcH#(beQ9WN4
zyIpJsyi|#V-!H)R;a@QlA^Sv^gHLsmemx4LH>VcpsaPSw)$>PxMkvdb`kyB?G;aGW
z#kZjiTUtq)U|p}b)@9@V<bL(e^H4Y^m}z1uIIpXRTP|1tiM}ZzqJ;t#i5fFPFcqfn
z2z=owWH=(tAaZh0Ynt=ie?x=L2_;a!FftKYNz>#PZn`{gD1Td0g|}}0wF!xU=e0hy
zHyG}uiLnRu=sn7Bh;pjoG`}Fdv!-#0vnYegTi{ni^fepVyOqj1R+r<N-dlesR3mD9
zm6yuO68TADe<cb6Yg!{id<L<`O<<}=d~#(&Cbq{P4PwJd2Q!&O0rFb3C?c_?WF{aG
zZ?hU~C0biaE*z>>Tla9W>4-K)UAhR`x5(nWp0izSdW$kHF1DHoWrOr2;ljn~@&VXv
z!;to}?n^_JyK<p97NU<|6OUgi6*^+*3?Ag-1|ge9mf(w0fQT!Dt#yUCTC~%ww`fXn
zYe%j1#Dvpwgl2}sXR*)Bi@5Pam3~zMMZ4~y|1|3I$4Tu}z&RUh&RMNk?j%v)?=jcW
z>~TjZwrA8#KSX8dc!BRrSBlPOK7p;OZB1NX&jQUu>3+zhyr%DF>>dQ*Vj_VXb;zX(
zIo_JSRUrlqS!=a;rg_c(E}I>bRiUcdPKf);b;qbq#k^^jYshOCbOAWZUqIUnQ9<E+
zRok82*>fUKBEK$&)hXv1&qdgv{1eugEQ%@G{a2OG^*wXuG3b~$Zv(2H3pToh28)cr
zYev{e<VQ3a$$qKuIJ2AsLNvmpKCza{4-AWhV(06z`y6=Ho!wS{5X#=ko8yJb{f;7i
zLZulxI-bl`%@^msa=c$EaYC^O>l;D$^5k5A9V}#~b=H}vTS^YGz0)O@*21Z`O0#~2
zm-{BE`O_B3P}xZhi&+@oFrY>(h+qjmQno+gN&jLJZ+iW2Lp5>J!M6k+iEQybA<igG
z$gTS1+Cf9P^nkLzr`p<6KqN`O)ppZ(S;pYC?O=*vBqr!*TnwJi-ek!Z-fC;Q`)J=X
z!!_TEG3s|BpO9!}p&gL_)2w|}bh)RXA>pKr0+#up4UogAeoM;JMHEB#{VVG0Br(~t
z;o<{@RWdi)*lYX)(Ellg;`^QWQ~uKqQPcgOc8HPgMHiIdN%#N6MK-MwOPxW=U$AsA
z6hD+|C?>Sb{SdlDXTqSEA{YPm#r(u&eVOc1!~;E`A@Ii?TO5jMV6eX7U#3^=(KbJw
z)X$50!WpsTP5Ivi;dG`1(Q1Mn-iO$?#VW{dUm|NOB$ja@QE>~Ph#-mNJ6c1`0#bL8
zb^;qmJmr$E@2Yme^?T-@OSlKhpe$oGY@`wSL5|#*;yAw8FZzoULOkIY8Pjn%67i)}
z{l}6W>Jt9?%jS@lTQ|d>jTNm#PJF*w6sHB{j<*k-*5Fdj_2a$#24~eH=jC%yHg2!H
z2%3x{t~SIdKK>vjhfqvt_1%^QjO=(dj%>Ar?0bGs(A!D$zPa)z8Xsv49exN|7Z+O6
zQRUPUKl+X}fXYb+dctaPfkRo?R%^7-e7>~5?UUu~$KcRYt<o{R+o1Mwy?Wcy$&MqM
zX6)=qi6@i8KsE9mLTjqme|l?11n=)oV8n8bC54`TgdpCv2(x><>l?;Xf7hVW&f)4q
zKa7KyF0KN+?^GbTpU)om1euxNw5TZ^rg5oVeTlQIZ`^tJb~Uv$0s|b;q%9c&;>51e
z`c$)xs>`*eJ`x%qvrd4^9+drc4!I@NZrA_@HNnBsw{F1nVkF8z@ec`yiSEx14LDqy
z(CcM#uu}3QSi0JRtqHAV&yzQlJ+KM6c}f*i3M;DrmHQUjI^~nLX$wx^x;4UaN$FVt
z&Pmm5(XnW;jlm;3k}hAJAQFW@KVom}CA#H78GhaYQ}k-~GuPz3;-Eyv;kR5brwy=B
zfR<wKis`>_?_yPB`{N!wjgu_j5w7@JOz2OZGu;OL@xz{Q;*pimlR!y=pCB{#+t5AL
z32+inpvA+yi-KRXMGWN<G-5CAPp+`+`51>P^XPZAp}vroYP!sQG4E;+EVla%vHfn|
z#|EaX2j^1(GZJzEND26^KBVP~l=+=c^IXE0_g9ivk#n7S?_xastD0^LpRc#-hdV_0
z{hMS2L%wspFLfHM`AppRfmS04pWmTbkzSqy*eo0l65!CmA7^`)_Abq90Nw2}UpR@b
zjs1LC*2a%m*M;*QVbo&1a^QSx7#|CBpliq@NG4%{O(l;%KbSXgE&km>;pUK}O#LY7
z=hKdHY$ZPGW{US;_!1p?0+d#mw!Ifo*>HlFQT4NYKK3Xx=%8w}&z^)%lHyL5sH{iV
zQlF5F9w7Upp0rr!ed-P?%I&>cfx#jb4Y6>zdKeJ+1*oh}oS%=)jNbSgBl(?h{7G(h
z3mkpT=d=<%WB8h)flTzOV9O;(hB8mL9kMwruOktKJSwJobi9HGFIkOasl_Se(vrzI
z%21W_{aWwfkgZ?Y-lndle*~}utqX+WJJEhlYJk+i+X|(+ms~(05=ByP)FJUvJ{*@t
zA0UuB_vEcydLk(s6&v~4(`<o4;R*T{&oB-8e<%$goyxpQ?J7FYYc+<99Q)lgt?@xA
zX-rc?;u#A1GRWL$gJsFtWG893Ku3h=i@qaH*m7rH?^fI3dBIlxoT$9-J?M?bJblm3
zNSf4ot>h))Cc`&Un_zxPiYeT<0))7e6|JO%<o+Dy>vqVKb!nHdzRPQar^O`$jL}9z
zcyBLNT;7G_UWPng87D7{$!@~X$jtVVDcw-^!O--QX8KHDsde0RJpzB=Dl8CGTWVX;
z@z}4;-=5e^MKO4ZMW5ZX8QNt)v*|Dr=8Jk=B(Ie^;aW$=do+aqxqxHb|3JavGWra5
zNAiX(4t5fh7us?>RPI#)^Zbo>R|As>Q{=BSGd7fK<Qw7V=POlq(v^hD5W`{}-=-TG
z?Rppmv{;`!1mS<Pydzg9Su&G;Nr$7{sv4_rXpAg+zV<=?5*r72EPpGHSIds~KTiFc
zid5uOgrDV_@}54LVWCG9hyESCUBCM=!p6trMfDSKx}yhwN`2b)sPH_?H#Gy4Z&@}?
zi-J1yh5g_uJ$K5DMwa$r8nKfr6gUS~qXm$K^ji+|s8C<D(KjLf!J^i-ww6}Zeuc+O
zkGmNBW(F-)pGO_sA2+=IcFk4BTCtor(T-?tqd2Se$e|)bBICmVjHoi24i@iUlXxRC
z7h&F0_niYUl;%?K$<?)UzeUsIwS96wnf^rnhCSqZ)Lc~7SjHyJYjSj<h})i~cqo;B
z<!FNZ;9|poC*VEd-lpWNvmJMdm~s|4<!`2F@Jj=GBLwAG-OWF0xtFuLL4ct4!Ryp{
z8M<8OnMs@=SZPLy79X_Zri#Zc*~nA5pZxm|HO>Nt-!1Ld6HSL-vHePI$O@TM*k}5G
zq^A`yH&o-#vlckF*uR?APTJx>E?Zt)SPk|+!z72N4GVeicTyS;X^jnzEuc?(yP3P>
zlN08Op$ef~naLYyYId-keedgm&K9%#VT>(izue;kL4<Y4r#AKvbVBM_M%d~z9I`8}
zO<4Xscx?5Wq<N?A*?YIla({jg0-q0{zTX#?(mZ?QNe1#qNwf@mHiT$g$_#1|qw2FO
zRE3Qn>3eqt%)>!v%o88*#7LwDY|8_^$tgu9%gq<`i_mHPkDUVII;Qtw_RB+^7)5mC
zCO)~}!GYsE*TkmRB3s_G>oXw8BYs`AY5sg;!geCmo1pHIex)aefr2!DvA#Du%ja7g
zDT(m(jrA81Ja0NuFR4?npk#SW=Tgxj>f>|w4$m(>3*rxA$Nt%CR|a@Y`UTlmzk=`{
zlJa_LK&$aMpOzv$X4@i{)S2$*P^*VjVyi`d><LTua(@O35dB1<y&Gk`LK{T(DyY8R
zt$d1V@;WlCe!`me<h|mj>`~!zR2>ac!W2?GYJ3BA%brp-`1T-|#o!YMKf7{>+eEb0
z!+5L{^jYlE?$bB4;fJkXuVVT8R&evJ-b|1^3j?y}_D<b~_Hzv8%bqq0jXutN%#N{S
zp56+LREhdWa*YoBWmMqM5huJdnuW}*xjc4<-!<yQkD?eJD{Q;XM(<yX?`p=F2*o!?
zn`kWJt`8|cu`SQ-s0u%(tB!h6I41m=3A7kNZA{j^E^;t3FxiVyKC@h@$qNeCkas6a
z!XmbjC)uAKdbkR<=q8Q0-r#r^_XywwUeXdv7D!0{LJ)HGp#a@e=Xhq8Jq07}zaP#Y
z<Ld?88FYt#3t2=VVcl3z6mP#O-&*edewH>NB@Qmhb~hIpqim9aAH5+>c)Y#7rpKts
zZ(aG`MU5}{u>`^kuxaA4s219<A|G{NOKDv^1`iL59{>GVYtPiXZkO2mN)&g2(3*85
zekL#c#|vwqsPkYpbh%Qbr(NbNcExY?J_YDh8Z^w7#{*G;43yA?l`OC2K^dy!{v6SH
zps#MZGnK;JmJV?nmva<nJXxc&4RYOlv9SV(r;Q)tIq$Q{pIz%CD_fh5NcdZ}s#nkU
zL<X#+F*O7vLH-~Sny2P=>Y@bw2t4~(UEcWEs7l%MB<!}dh)QKkNlS`fL!tZ&X$I*W
ze{f==JFtWVr>_4?CHKX#Q!hWaXl6K6-TZ5uZ=I^?gOjBz|5}r&69B4}z_Tzld}wGy
z?b%1(T+-Xdvq{HW@~S}iyt|<G&wq{s>ILpxD5FIltHoy%AD(m<*-9OrjE#=`&No~h
z&a5kP)IFE8-cJKe!KPQ3n0;~FRcZ(VX?#sEJ~^5!nlJ&J+T5a$c0~u#e_WY-)+gKH
z{YZ+(zrN=FbSpBh_D&uHpO)o?oyPXOrEcIeidy<KdwN!Dq%<0}?*eotcp(%c^=+zA
zKA8yv?l3uQ#A&X(-ptDd5i*?>H1w<d%ovWkm7jj9q7B2GFIt>R_`PpE?ueekOtlSq
zD?*Um=U0_Q!WP0HmHr;0IDH($;EznIm)1y;*Uw?!R{{+X&=}A!?<5WRid`iRdZFm&
zs+(j3E{U~_ZlOKp%p!IEGLh?yA26aZ^Jr+5KW(&=>FBfk$ha?qWhR^MGpKL!pMKr$
z7H_L>>D%j8#_jS#U#seThtMgb!S$HNFps~D>K6`Rg~$(tc6ZusJL_?`wc1}j9<yCO
z9scI(xa-mJZ;=@q_fCBw#;Z&(Mp`&azZuk&vx=6taMInqJ}D>}iJbfaut)2sEk--+
z=k|Oi_(_-E>o54eBAm;1UF`l>4h<9QdG{*Q0M!SAvrL_28w3V~MH0;eJIaZKp0v29
z&8L=>zYp8bg30W+;VutS>>Hof2%kjM&0lXT{-?6DY-;oS+IEp5#jUsnEAH+Nh2lYq
z6{p2DxVr_~QrtbbyA>}E#e;i+;BI$*|C#3lJTLPmbIoM!wXeO`ah$Pjy|bz4_Md;~
z1pC@Ph0t7v#4(dWv8uN^bN0A5tHRbFqD*Fu_d=(_{>f6i@-qE|xqU<YN8oj_yB@#2
z2jz*7*GY`t=-?u|()ju^Uq1RQDLlhs;{PqgoZc2bT?(S7!J(Y5g@B@6x8G$f^tX*F
z%9EGK)onwTK7{&cZ+-XdI#tJ*rt9yTF%A4-|HMO<fT-?D9c;wwDAKXSuPee}g_}?|
z`V-|jdWZ=>7m}YZ4OB1(@)vk7l|OabbzUsXrP!g8cW9xZUgV;}&7FRAIGETMG4Qrc
z@Rcqr_DE;dhCVJEZw@J!9ZD-9!<^q9?2uEH5}_nyZc=qL*#3$=-TZa@m4Q!GdR6EU
zY4dEtdO*_8qNu$V?|#l`$KVS~E7m$SZ(prQt+X8QNkep)GiGJi&2G26{L^daEZycT
z0URxTOx4*qHtXF&(amaJTQ(5U6R32nkJcNKyGF-aYJi~Z%rhUGfjt8)G*Wh_n4ye|
zZ5=c1H%zCvqvVahp$pm$Mk{QSf!7M&wdCw*V-Q<*DX?w&RYHw_PTRtKRD@@Cq(zii
zt?u$wxSHhJF;K4=a%UZOB&L8KibDYJ`ufhIDyMtYH6BaRStSfkIWkwNG4l&t(SxYJ
zQLQF`xN_vv{wyCaZ3~tpf^SRmTuiH&-0V_x6!%xw6rX|$3ILYPtUHWr;FSzYX&~<H
zx%C-1bj$?Gwtn3H8oNz*VQ+ZFS^gczqb6EX5kJu%>@n6+zWdPH@-*=_8Us~~T#}hK
z@N*_gzQ<t_Jq<NQ<OvytK1Fhc%ZHBvS{3PC+x9Q0{t(?bPj~v+kMhM7WcVRoKnBg%
zAgDwd4&-yx61pBK92Zv+Y|mm1x4pD&7t-tL`2FFqX0wYHeK=s~xP965x!8l_CVVZ{
z_K0x8tkMprNd6zfnKV`j*+}r$kg@GK<UJqBgw+E+_F;;cO$h*q;#NIBAn<O(*>2b^
zx|WPHq$4pL3fGtHr0?`l)0fV5ivu;yzAajO`U^J&U;4RYQ!5P{U%oY7%`9|Wu8HJR
z(-Gzn&XqZwD{+HyJ~^i-OJi@j#631hjddE()`yHmt57~FEo>vS_;W7Y*7Lb0Gi9*k
zn6ht=38plebV1|SSqhBbd_+mA=y(3l&F!asQT$W6(CKy$(Z6pXFhQ1N`5w<2w)15j
z9odK{jC;(c*JjE&C%`l3cN!Le0fv>-!1OqU$tH*A(E5nr5y;}!?O8NyN;W%q`bp56
zU+>u<+@g=_-?kcDzZ^l_9i(>K7B!WJfC<RD0TbF$YULk$o|4$ZTK3Tp7Bn-Juf5i#
zmk(N6<AO6V26B^qw`9MpSlP(*ur(oGT-m-STm930KOTO)9_R>p&eEp4?|?8Okn+GS
zT>@*2p4(bjyL&ZTzw2()&PmDGE;^QvR2&Le553=b5P>jB87lho$Vtn?l+lg~B_KxR
z%i|BP(Tzs6D+sv*OVrpQ)s;aY6pdCX&P7H}O~-L`b=Z(tWH<g$;qF0h&_~1n^QHl8
z_BwTu6izC#KH)!)2mW5RbF#3bP8sYsmHDkGCsde^HRK$tEa~}XD5;G*ImVKpg>AO;
z0SjPieNi@v_gnd&TD;`@&{MWwPoa7#qokn6N#O3+-VD&~dsROYw)rA29lkd|5CR?R
z(!8zwUPV>1syn!rUoQL%UwMJqj8@iI;X5^}PCrMF*a9p@%>$|C_N5b0vDn?GY5<j=
zMB_yoaMQ1J1@AT`y;2t2BfLGpLzPcoLNBY?9E_@=a(V1uWt&(_Az^HzglWJX(l{ng
z$AM$i^3LRJ17&=0UXtUTfB3l6H6k6;KD2&cDj{lWFj(iizhctLD{<n+R(xrEne$F;
zWTMc3`1>L>x`NI|S=4kN8UhC-(u8?-#qWPk>GvY!hTsQ3>$elCyZ1wiF<g4WLy_WL
zLuv5dgHX?Lhd<wLJ-5e=HBN`ChC~3Nb*tamSY<%&m6m>oM=6^zt53mj_xLU<jsFs$
z+7!bTtfWnIX*qrRZsl2C2PpNKBcSH+B|k&mL3diwt&@f;dm{2!9RBko{UfgEzOVK;
z(u!RB<EG$+pVCLuT%|5@R9oMot)jSED)vTrioPIz0Oj^W@lj%)0jowMC7G1x&|!bS
z8jo#;Vag_P&tS-zAD@%yT2%EL(=3kPHW|{d7tW{;cS)YAeMP7hX1nE0-Y?`Xo2WM8
zo{ZDZD03qs41DNAjs+};*>qaDdBI*2;&ndxKcbao6C<K_tbZ{)FZ!d*E6D;Agfm<I
zGB3Qk*WQj_vvgTHBP=3o(PfIm$7y|iLDe!(AEBRH7_-Rhzc=T&CRvmwFn?dQF<b0M
zqWP<U7P0kRz5>nTY5RGuXUdiv!~|-l`MzV}<lMlqzS2Ole~~-c_V|0Z(#wT@^wsI4
zr@qG&5gW_55RCLcq4&nPY3=THxl!`IDoRpT;zpQWv;xxXSwG^@XoKd}L*FO%<)zDK
zqovItLa9eyc^}s^NBSOwpo5eECJF{8SYO;@WB_kXji669J*RJCWvLGTJAQ8B)wqTg
zfj)0_8h|8EoDnVQE9Lv7`ZI%mq#5+An@W{tNTARN6$1|QPnbPSYg9La+9fs9C|ese
z<Jqn)R>lS+#5;z<cYc#G0-dtWAGWtLrFQu4*f_jDF0ESpgBr@q^&$cOh*M~p<b1(t
zd3&C#1)A@3@(@!!eh+7LW7_s(zzB-u`IC}@y}RQk*MGf!3C>J-IVr*U%&TC*u|K0^
zf}G~(aVqCOX;9?~8S=Gm%SmpysOc;p#n2FFL+_-qi<KKvX}`DTGb7dw{e4R;uXAZD
zz`RwG@dtyo=+V7FQdcM+GWiBILj12{8vS42NyalL3#a>6T)j0v^vhY97Ye|3{<X+X
zA^wT)?;DEPpj^N&TI5R6kd8pieoY)&ny`o373Sh>p8cC3X>-o}cd&TX9WRcHtwxh2
ze~ZRP9M;=crn5a~#rp#|M{`;hwOYO^_%dziUVV)v8hoIco9tODo1YCDXl76J4dkHs
zJH|uIgFy6&y2vc=8okc|G~xbf@a2A>Qf0xX(+!2Zk80(AxF-L5UPvX<_+d%5TpcgA
zqP!(O(--EWZ=7YKE7&jl`82e-2$#Mk=OIx_xb!V}a)k5)RX6E|4p?EoHZNl`{meSf
zElFi(<2+}sSK$)C+7E*{(ZZzzZS(WOzBm*IT!+MBAvn<76h#v<gae?u?iK`gXeH4w
z+w#++lyySVTbEkiMGm?8*PW8(ujv2EWNZHjS3(u~Qne`+c9FSIXXJ?32E-%c{sy8P
z>fH7~=|ALqjKjTES@TEU0!YgbU~Rgb@O-&BENqDu=QN24{s@~={eReNU6E4=aQBoJ
zzI1Eo9rrzlHmvAQ_pT8W)21an2KAa{)?y|A9Zk|x!u5+hSi>x1juF3}FZ0^t#hMmC
zTk?1Rl`0FN@ZeBs%}*q{pq``Yw4`h#`F;oqqW#>abcnmkW{@bmgRNqTdHaKWnv?Tp
zi<)ShZ8BM%r9)_kR^mH7aE)~=<p2}_W(~ZNto7dgXzZZ3@BFjj4l}$Bm-7pX8Zs;1
zVW<vNMM@6fb*Ofzkv8Ct6&aPW;4k}yzGqr(cM0_C!BnDq=CuT+8q3!Y%mVLPn31l@
z5D*ZL+{5qvoB$1cx$<~~>mmKIomZ5(N1(F&ceUC9UU;u$?qz?Ai4JImr~!Cdmrn(%
zu2>#`Z36A`a6A`pZ(gfeo5zzY_B*8_>1=Lqr=v;|31BZsE;nqdPa^0o;xzjW-b08E
z47SBlhPGEj5|*c<lUoLJBZ%KRwC2qy9n<FWPk}ESn>#eE35?#=v_aeN<$j4l4(6aT
zV&CYNpDouTGUyjCi1~8B)6iSIvL3A4GX(q={PrkFOw>mrUyHOm+EBK(61DJlxa}Y=
zvpU3;A<G}b&EnI(YrI^Jjk7L51TBb8@jBSlTD^pj6zBWX-?)_H=C|6h#PAzI{WZvc
zM8v3O7;-6BI)Y;1_KG^HJ^EIRfvGe0h6tmQR)T*>D4pDx3XhMir#gllFyHPVBc%u<
zIh0{{Q4)<XW<r7YkbD6v<Tr`O`W4~hD(!zRffL^7dSQ2s`ExQjHWD0VyIM_yK0H-=
zbo%E82StvEbIzRlX^k6~3FTU1%ikrx)V^effszye!*-1NzCUcV?sxLX`kq=r*3*K1
z--hVr-4w-*5H-<v{8JH6toHLrqbHDQ{%2EJce3OpokkDSO`2=cMEx5L_SWVOMGqgj
zQ*U-5=F~=YXAts;T&lq3;1O%5UclqG{dM6E)qg2n1#ni+!>H6;%620zIBt{BW8Mru
zG0#P>RK_T=)G^~KMfa>G_|GE;A7Ur_*1uo1drGF38p_<^QW3E@9}!zoGx1zVLFb-K
zHy13$*N*>Fr3O2qj}Vp>9|2}$wy{d$r|=3x*aCzS9Hh&~kS6BdE*riDGG1ni$r0zd
ze<-U=Y|AifYm)6-Eo)bqks@)j6Y`?jS8~tbvd~zfuO)y#p99oYEMYrOrVxuTKXkq%
zdQ(gA)Yawm!!m77y7@&vM%W&6u2*-YH5aIZ)MSx_SwZiSXxb0S_J@S_V~BL@<_2(%
zkVMalhm5@ruCOY%ErBRG&ASCDhgov5@S83%31k&=CA#qzFb$Y8gbK&gCJP?ur+n6j
zScb?LO@*dr>ImjWXMO>#pm{~#CnB(8MdF)1=^*bI3CS#>z+%>Byr~@v=zrpf04!XI
zpK=_lh{0<a@q{Vtodsz!L(DvKdJZx+V!W@|g;cR$L)M5^Rh=0?@U7$<iDdlR8w^B_
z))@5<WAaIwGXqPSGv-Z4{V3cM9}`LO9OfH2v|3}IgWr5^TfyfL<ZJkhGKn<Q5{ZG4
zijwVMEA&0qp3HMKORcTAZ-2sWLndMvc`%rT2aYLFG8n)GK3{BsgxKJ9JI#;B!Y0CO
z9|!qmJY|3-k{d`EQ-CObw`(gm7UEO-^|=?BZ#i>lECh@VP*o-AD@r)(1tWldjgD#k
zK}+*BMMMV%5!m_}!(R44G`0JHo9DMNDUT~4yOG|iC?^WL^W<<c=8!Nx%F~?e4B{K}
zr&n+)6g$d-ckU_+S~R%75W#gkO%LlC>wRcpCpbZ&tw1EYJ#W=mWT|tW?TCwVq%jHJ
zZ5KS~_XlicZ_m(Ff<)Eoc0DP3&ouUna3KCA;!{(JC%^V#j@Ed2+jk$ox}cAWLT!99
zV{S6`Sf5E!pbkhSHnkJeY92(9pB&KAeQ|F$V7FrHKUySb>|r7t@3da`;o2QjYJ2-E
zl0eP3xq|9^GmNV)WVzfw{1vw<x<UN8+a4;={(xMLiS|whm7MPX8G9BjL99`5gKU$x
zM#@>%PvvYEccJ>Gs@sPY-40sv7BDqlLtb5#lERvjv2ldvj|V@y1W4oI5q?6foBgkH
zDYz9~77WRA3H2zcWY}o{$s6S{99wPto*QrJpn(T6L<$P;6w-mrj^)^~3HEc<$@Ar#
zu1BGaCe<E?A;(<Ir41`B7_b<oOAFJQlVN9{7URam>cg=ii9d_l>Il^PFOmd`>1Qb~
z3pRU>UF}qc89<MvaT{YR_+a8OG#G;wZj!;{F;X<~G?Jngx)%R(o8Pq{&G$ks2@6;o
z_5%^nIT7igk8d`?7dk<&){Hm{^Y6MEmpLb6NAb?RV#L(-e3a}1$K-nHSQ56BE?SS-
zzKRfliD8b@0NCiaPLNSckiejZIHIcS>%|)49<R_GW>*uDH*dNz!cq)(u{KS#VmiE!
z#z$qe#Kt&K+kHG1Um>#0Piq}wKO=`<t%W;Yi=-RhpG{6`<y}m~$1DI|ve8&mlVPxt
z<i^|K)%Eu6gpIP3g$X##eakK9BP71_oI%LonY2<1uF7D~HA|i*C{WE74%)kA$@Eg!
znPYM-mJl@_i?{w9eb8vzh22s(Ln_;cIm9g8_BT?FKvXFv_*g_Mq-vpyfK$8>$y!#L
z5HC`{Dt5pvJN{Q(&OJYw_BM@;#1qD%k!V==&6QG%_nQBu7E_mv2C|uqtcIY9)u)?u
z5sMY0Mfw`Z@GsxhOV(Q6|9WU)zdO;8+(8lH8ZB*5YiQ-W8*FdZMDP=0dd^~WiJm*=
zt~Pt=;N_Vj$9W{M!4vHQ`Or;(Pp*_=$=ODDBkM23%0^kMV@2tJ)=W7|KG%?Vh5QM~
zBj=eVy1zJvbxCgG*l|2zr?j|uzMwrE+R?2008;tk7UB*)p8qn#u<j3QIHQbisSbu>
z=9A~A=hKKPEt8%I<crIv&?w|Bx@rMVvw05G9j&YqoDfGX2N3UVk*4Re0)z!uK7{Q~
z%e*iDlLGaxgh05VyjW)Kcw=ha_FGJwCasT<KAGY=HIc#g|7{h8^phzC-ap6Ma!1hH
zQ%Y&DVJ<|?tXgs931W}lWc75o*s!b7=x^NVm(!D~{OQC3<%dNk9-VxnJV~JtTESTO
zW)RiOr?*#OQCIlepLkTYLlW*!QnDx6k6QpY*W{yGaQ4Kn71sTcL4D(nF^h&f+YT{l
z-7@pp$mycb-`-UkUD+^UJRdqXH0!Fg)&M_~hN#9Uv0QeI+P7CjDaYO%`l8wO+81e@
zBsMvTeoX&kJa{IN%;DYWI$O)%Vhj@%uJcS1eUI;UN_&pK=Kq$xDt~~U3*~q)gagvR
zW0jxubv<$6<7IOYFm~Em-N37TJlCi@Lq1ZC$Y41xK-Mg@!D2-d-s*W+^1uh5P1xDZ
zx=_}nJIER;$ZyW544R0)mYXA+dc>Z3FM0FhuPIo5<s3|8CT<7%HQIFS*5I<=jE`v-
zt*y%c(_sab#6Bua`t5F+J29Xc!iz}rJBD4eW>z8pGu~ccK>W+co^5~{Mp86<2`P0(
zkvNV<88MPGlZyLr@fBNZEM3R{2<F#O$f4xKHvI=2HyQXOm?Pz9B9|^DLR@F~7txrk
ztMG2{+*9c3D!3Mmx+#>{76PN$I^<Y09(%7jq2*Yy_^^a+C}@a18J3tsGkV5DnXsyi
zEcNyR^XQtj`Ja{I`p=t91rkwir<v60+##uzb}>S*wI1vKvJB^%mra3@5vK!Vs5)Qz
zKxPI(x%&x-3=2WCh;6hn0N5$<-9|5mWU`)a+#XO$g{YP!XE_8PCpi{~AdTh5f5M_O
z<*Dk791XVh6f68NJVB?GLtFtlxQp?<{#8(I>vpkiIj+uXRt@t;heRS#P;ovb%XpWa
zdJ30Pu{=JH(2$iHQWjNI824mVzT<d%o;=Ow$5ARFddw{z1HYikruCQc99Uf^(w8cP
znh4FD*JVlT(2hF04f&DoQ*w%$mIOYl?3^rcGe>e)$?T-J1fKX*KY)`TS+`t;+iqol
z;mIsxo+5kX)byt1*akU;m|zZjd(20OcPS>rCAaW{cUnf6CRwPR2=+ewrIo2y7W-VF
zFS7;ROoKQKJNuVuOz-*04Uk0oNVFL3Gs&2L8Gh=7oE5ir@4{k%K!#b1l0~(2u0FP#
zRfF1t6rL-;Vpogd0$)qGX)Z3?W*d+_err<Xe<3W>bl^3cWsSyyFJ2TWMw9<}NIV~*
ziQ$ZD{ys+0Bst9nhuif+JFy#Wo^0?g<R5{X!!vmxcbu`8f1~fA57SGp634IDCaAp;
zza;?If@S_G7n7&jxf$PbT%M!)+WYxSU-f=|?El<Ec)H)a<ftaW={W{aH{+IGk6!PI
zZrOQ%q!d?k_J9}CdGlP8$vd+hds~W^51*e-gH5G9J|H5Bm7Lv{T%3_J4PLlb)}87m
z>5Y3&j?IsouNsM;Z`kKNO1HtFxceUu1Kd~|^xTRhcc#Vb;5k~SsfFAZZxddhLo0G}
z3XQ~ZBoc-(SrF)~O4ph4UCZZ*ti_wXrY-J2jmOlhiis*74U^XGDF#o|;_$zr)y@`6
zw8?qBzn>RP5rBm6&UEAVOJ<k({k5PV(j+1pV|zP)`lHP(b>scQxOX$9`!*+qW8l9X
zl5jG0%>Q@h5l6c*)llUKpMT03e3@C%=U3Jh3r`HSr_QCkk)ozTX`djmh^}FCtR8ZG
zR+$m09T<-8{%9MBmt0E=zEEQ%PkR|m(%sfdJkBxhXuIFF-ULoL!_6H$65z-;kZK!_
z1a9-*N96h_x>)h%1)TQ=r(d)K=^~jl86(6e6q79lPbq~DD9ExIoj{%4*L6b|9@16X
zgV|}qEQj`+s8Tw1M}1<0ERG>reUIxh_04_#vknTAnY0br@_k|l!!<*`sA<3Lsk}U!
z@UkCvjE;sa=?oLjDp?*_9iegDH>;F$iE8D6Z{!H^dd(}?mLaV<>&K5c>C$U<rDF(+
zBwcSGhT&W4<auN1z|zabZ5Q3#{B2P4xZm_@d~R9Ch&V3ZFk%$`cpJb|x#HEPw)vvE
zGgX3h+a`dSulU7{PgMB0whKp|ZaYOSA)@=XU&+F~d<!&h{^TFpJZ8{4F7u0)dEhqN
zo|=NMK0%0h%=<cW^!b!8Oz$#x;lD-nKT|YZfbqiS<%O17baB8Nt9%Vbm%X^F<TNts
zizBYZSMI9Kn*{T95B4?m%I#GyPd}~tn^@Z$K9Ab|=a>%^qoD!jRaumQ)@)(XA$Dkz
zT(^4T&z~#)WDnRDwi%qp4jcf1$$LhSF%#wxlY8}q2Bw6@OQdw8t9G^Y4RJ+9nT-Y6
z%O1u2;bw>3cGDkqMzqvv&h@`@i&zkP*uSDfu@mtLPZfv!NzXCNJ2|zh{CQRV-Lr3M
z;Wdz@&=+h5z76>Bb5e|ZO-N`7ZT+cNYk$-(xkE(rpgMI)(I^#%)(-tkDB*fTcPXhp
zWvU_bZ*%x?;7BvzwN;Zn+nzX3P)`);;cv~*ed&MKE?<KJ5O)c_tKazVES4`8si5tz
z+7p^lO;7(^@D+St&<@1AwW_~+536oHW3f=Io{~7{%e%j{O{KgMIwcP0jnB2&BU@Vj
zp?$AsV#utAEP2q<t_=jfmeOEJ^BDj@8mN=3AX7vR<P;3va5gGRbWT2`KxH5p+cye~
zRF=A>5$&W8?2<cu;5QUsDT0bA%-=1sIb}*UN|5Zo=)*+hQO{{8Gr2fMt3>sBO-h#U
z`lPO986`baZ#KaN@%x(30G!*1gOlu~1wG&M#uUBg056Bpl^@=B7zSsPAADrwOX|h?
z%==Mq`?o=+f0Y>lA=aeN0Z!%Q&y!LnBGVHP0*irT+o@UU`+OPh<7o+Ou`%iv#h|kW
z>rLj=$8^D8fw7SK=;7J-+X&I~J{6!$SfdwNi{UOzQ2vT2(#g=~yxi>SjQ?|_q{{q5
z`(FbAoJULXX$^7Sl_^O-myl|I3$tfb>NGU(2xK?Z=KSN7f(@{Lx>MY=MO;G7F>K!z
z-Dcl#iD@YGCE^3Y!`V6p4%P_b5{`a{<?iyr!J2-7uSaIUey-h3p^~S5z3M+%s~4pq
zIY<6Qd;}M^w(_a?_bFVylVt%PfVXye{5K|Z0p@&m)Z(9p1M%+WXmR{YgmIqK?Vj(J
zM@J18CP9C6S;N!?LIboiXQk8h%yi^JOSc^DLFzs_t*b%hjtIu`hNp_6<pARSbhFJx
zC75=&1U$fPwvXy)TLIptk;wKMxzCBGj}g1B8IG#62jnTUY9M5DhKHSC7{CSIiVuV`
z7}*eQvhMuL9yqT#{g28k!A~b_<f1q0GB>7g`dB{IWl$iz2o|JuJUm%mzTSZ+2dV>L
zQEWu8EL9JzW@1$jlXLDdjKH{?Wk)<WPD;mL)282uUGMLdZbn*lZ@hyCYld!x;PGS6
zVL;Zal-I?kDwC3Cu}!P}#4u;*NJ2y|QchLSk}M}fGkZd>I-;(;o8Q~5Pi!XcEv5Ii
zLotVmMB92J%aVoMyogWLL7dnwcfP$mi?8L{hk_~P7fPHD^@5_@hHP|J4WZt}On=i9
zQ%CstnpM(7#dutJlU61^2=elpvTXWQxmp2+<3lTr_E$E4ROWJGg9^c%@OMmNBHc<{
zU%|z;JF8#oEw%<P$|>{2c-Eoc7O|@&k5f)`zriTO=`BbF2evcx?_>0iIKAimc}4GL
zC#NJXniI{}<U}FT!QNgwquM$>M4=wMb|iedDvu{l?4W@C2L*l}13n4ks6cvT6Er}i
ztbj9P%Qkq}1%O40*57v0tMK*wI{r_$dkv)k{N=2RD><Av^gv(ZM!PY1C<Y#8Wol_F
zfFUAtw1H`LOE&ih2)RJ*nI?Hw3TkTK&-W?cSnI3NA8sn*?l}sd=wTfa?E%!34h(<)
zYUtu*8K0#2S#f`OnJMUg==iT43Wc5&YO)WL8+M?5zOc7AI4#z)Q^|&IS=1_SRQDGz
zKs8D!fgpfy|JJNUv8leI3B&Sp+}>8_LDw5s?hpI>D|x;Z$>2{`J`HFhlE-+GF!<ga
zXs+*pceI)d+AeoFk92AQs>3m0x2SB_RP1;euJ+`AzN7xNlP#=50gBf~U_7@K{1mI2
z%l-*CsG7lZe?0Vc%nxxy7ubgq_LWae`ohJ+c_|(vIpgQvEO`-UXFKqDq-cE2M|)>m
zzL*-%68p9`aaS7UeRc*GKG#jPo!trmsL6+%57dvN1a;4YFHWv`UX~7PJMQ~!&v`9c
zGe<(GfF#s6E!3dzZ|fvA`9y;)nQSTiLg$wuS5BF4t$Uh#)CWugKj!wJ13Z+|pW#7e
z^IqR_I?SjjS)ji8b#ONdmf}3o-ZU98-y1XE%b4fkvuLpn{y-rujITk4K-N1!D&I^F
z2X}VEH>}(xtMXATDh5|O(Fs|QK{`!DR^1N#C|^e$_Z0Zdfx`AJE6KSJY7yNoHOZ#p
z+*+C0Y`^QJwY_hiW#DU7{6@s;)ZvIdz*H~r&gZx~PI>-;B+u*Jh?*!*bw2wDO8D&<
zis|h{r$<V&<)b0dlm@sjj>JM8TD5)7v*=M{=h%6<w(iGEE=*>>m8Skx;#@my@P{oN
zedyIcBCmG;^A9>O{?BgRP`{gm>^`q)nAc!Rc55tlI8Sw_CQ9AfAR%hYen|OCA`+(a
z<Nm7L4oA68QB{RKw>gl=SidR`qiAW#<cd1C<8n;j0;VtexTBPv#qkqd+kKQNt_!b)
z^@7Ro&DnF#&atldqj}vo9C@~6-X|m=POO|_HbrSycrLE`kqMMR3(2mJj%KeV<!|S(
z9@-|^dk(t|M0n=K*qXKFy@$di#ll|izQ*%zC)p+4Z?8PRE?Y|`jAW<I#`I3z-o{!Z
z7M%r((@i-yZqSh~kfL|Acmb>GpGdr~@r(DyH_FVyW_>VS+bq!R?fz%R7suiHW@xz{
z8CIvW51-`xL?0(B#qw+|BTTtYii*ml&&Ra*-%ZV$i@9Kl?{#I1i$fnj1<|;AOjut6
zg}a$irou3fbij@gEUR6u`zQr;Mlk=ATDe`mlpPbqRV{iLog(BP2|V{2AFmSLeT!x%
z;<J<be%5Bnk8=E1;ua$<RO&Nt=?lx*4)(;tw%qzyKr`|7tqOW2%W~d~Uz>Z{TE*7G
zKtgA&5zq=-5h;Mal<JS1b<fI{rIsLd*<*!0_410^+b%l(JGFqB{8|TN=tCRj>$0cS
zR=;iPz&p#?@PR*1kXBO`<B0^zI**Pro8pY6UZY(Z)`O3{v&4h}q{ASmcLX@_`l6!B
zj(Eb9hjA76dtr~f(m=f)4T0s}S~549VNA{f1Hxz$MCFSSsfW41f5x!W+f~7*ux@hq
z03_71w;$3;VMG8TSkd44!M2SBXW>ZKU*9etEeQ<aalAmX?Mbs4IjgCDy?7tXGSBJI
z9|aF~>~p5yf^s1rwyc+V@2!>n4x#3^on?vQ1R;{;!(WHM2(v8>E>0~5u=l*uyjU@q
z0%VeEVryW-eGNHWQ07XoAFmroP^W$4C<;n=9t*ihNb9PdN<O6?znqWe_OD7loRnQX
zFtCedt^__N!ZEQ9HkoVA!-|)wMUfyqrbBklp-S@D-baf>Y7cj^j(A(`FcgoePUj9>
zdPOl%N)P2ev>9A5M;(aUkGf6<o6aLFJlzt{%toU$5{Uw_4JO2Jn$H35OJ_1A3-6-<
zORcGmDkF;BrRlGKxR_43cd|a!2R3UB+F9)?s$8~ah_`pO04wE4Ccmfw6;}O`lLTN6
zcn%?g{hheUP+C+?(1J8&m0{CC(8(z!UV+-r538MWMPO}!dvH}dXw^zwM9=tmS+#!H
z#!Z;-dw9G8<NhFj!)M`cY(LLQ8y1&!I63827DQCT;T2(^(0H$||2+6QJi|&fFsq(I
zws#p9&B9aG8fRvPS?{8wAU%jn?XiH!M5ENX>8?n!gQC(&)6yz+2s6er7&7QHkT=k@
zAv#aC?b@<x-D+}YSQqn6)((6o{jC1f9QIUC<1cH(AD5`bd7GZy3n5RIrCJsE^>|MG
z^jL1#?NMpCTavL<2H;9ASBLzgA1s8Zeo+NTZ~!b4z#CdL$as|leV<&2rR+s9rnq%f
zpiRyOAiutW1s4O*G_ZdG@}d0w_6%fHH1??TDDZjgPM!K2(Q9K{mezFk-yxGhT1sp`
z5+&=+J{6R)wQnqC2r(nus{Sv0+nn~AgGglM%nf9-B#N2}e><&ezl8`obK*`0VG>(y
zY~Yt=VPZb%07yc<uCB8v7Ut!Da$`o%`yE=47M4q!P6%MJ>|V|lcEi7*>)|V$PmwLm
zbN=#(XnC+0j572(e}1m50%Ul<`{n9eN1T9T826Ck_}2{I*~|#(O)p0^#{8It5^pFF
zb&?q-ai=CnK|F1gD3hD_k3kJHU7j{SUn^007#_v0QO<9CFrq5_?qTZMRT<7-YHpJ)
zT8+_pVKgI(^@+=(wal$=_oA4hN}w+#fJt(Z3=3aOpeXlQEDx_v`w1?9h+wX`m+ivb
z+)ZTY>t}c9n-fNE`Jm+73>4iCCV59~xd%SAO!1LyGDEF>;QIW|_rDAl-l`WA1Vg57
z^xneX7r@l_5@Qrcs`J8^7UDb$-WeK4m+XtJ3&KZ7S6+^$NZ>!&X`PU!KTz#P_GAY}
zUxpQ}?a#t%EUg?CV!M87FVE7`zVtM28Ws)S<jPCEvBkrE^BEaA=sg0$|NlpWAAg<J
xk>A`7AtOq~2ECaAz`qv_?F~C3qSUCUM{7shv@F07{vSkdl;zdrs${-|{2xkbS>yl!

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_cutmix_mixup_thumb.png b/0.16/_images/sphx_glr_plot_cutmix_mixup_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..8a5fed589d17fe6ee62d6a9788519dba471fbd65
GIT binary patch
literal 26794
zcmd3N^;cV6&~<SuCAeFl(4s|(J3)&T2rk8;P~2Tha45yKc!RqHEp7#hLvXj?ZZFTf
z*8BYp-w#<U>z<r@Gv~~mJ+tTR2sITsTx<&LSFc{-D#(M>U%f(#fBBGMqQ6{mHpumQ
z^~%~w0VJvEnR)nlooKF`yb|H}xA|C<^cUE&K&!B{#ar*%RrUI{fOnuSCL$03!--#|
zXbLuXBIIR=G00EgrNBb5!{Wg0L?-(}Cu?Haf+0F7%3dlH!I7q$mL$Yecj9<`jSzfT
z^D5Zgt0)_Z#Ecth&8aB^q#^tt+jp*g;+l3Kuta%Af{32Sylxr^V1hg&jmnR}mkXfv
zAbI2;{|#XPvE=`E0P<S?^9AVTE1Z82S>eA)C<&;HFH;EcKm_ss4Tt~~BwnThPbC<U
z{y+Qy-yssCdl=>g5(+AS2SjXlQXBFG6FYU#Fiz_d=HGDi{5xv>!;RVHaotlES;<Vs
z54)3=h(~4As6dmO)fSx8D!+i$VC>Y{+4GrXjSnu(wiwX$)Rm{msq2XKT(PDE-xEgi
z_(VivIE9l($F=w;5%X502cyd9S;o)?@*@IUt%-#RKi)is3GxWf#e9B1a0p}-K^GkH
zE#RcLH-hEomo}wtS-`xST`geg^-vL1i@Xk;hL(*ihF<qw7%c*?Q+E}k)1pi-WN4&J
zrc9QIvTOJ=TWorN19X#w9YvE^0xfTy!~GL#58!=LRoJfOBeNadJQGEW^>VL8OfGro
z7;=ZwLXX%~LDV&-I)%f^Zc@iBUPr)DOdL&SY8Fe4jRr0M^?OY#0}B$nzJ&L1In9Al
zQOK9T4G@xxBY?1&=&}M<$uJ+xfTKBW>PRAN(VR;B>we1G_F=pv`)g>`Z^uYRx_jEF
z<Ml1t4a&-D$31j+QCSBStgUGDG#OX))Ly@#cs>LgunT}hWZL=J)o}-tSc-H43?kvu
z2VGc#wT9ESOGB&rE@vy9Zz3|&#<T%q7(yR6b>)3Ae*}ua?Sht~cbom|GLpPlTSm&X
z8g90~9$!qLFeLh!xc`P0$(R$Ja&h&6EezoO__ACMkpyr<B+{;|AT;h98iu@OkGbHl
zGYQkX4NM%Rey;N@d@DM~2kNR!mU_AYWtf#)e}ui_>*-p^zc6Sxuga|EOJ~>i$53|i
zk3uPR?4WZH-H$~ya%J$=ipQkS-owRa4tw(P{%YPDc|taynH>AT=I|^o0}49wU~C{J
zauK2=H85m@3Fzliq*hcd5LX*{uUE#$CpBIVU20AE#`cp-cs73fa2V=^W|F4HJO04v
zh?+`+`1w}s-C)?|2Q&Bi)%tp}=My{T_L}AL?X!5%*(mp1+HX=ICa0VpGHL53#n?|w
zrsVN!1h2aM0663#mGa+wr{K=E&1QL;2u+|ml4!Vcv|H~i8eectf>jNDUkq^angTdW
z)#m!O%sOuC%7f^!Qht!qCw#tfe81zkyyL#c-mQq;u_=hU8Y&X0P3ZX%SuC=j%cBp_
zr)IOhDL^juo$ys}!om$%Q`tjo6J*I-hLpYq;KY(EwP$7&><+R6Cs-uf#l&-#?&}Uc
z2}fsB6#K32O!O58@ksbMFteP}G2(2lnU(Xx0EQ{|6Fyl`z1ErflfT{pGkLUjTO->;
z)k=c&e_nqG{-=f;xxac&YcIobydWM*>-WCPDRj}<`M?J$>U-1W=e@wPjOqn`WB6h3
zK~jXSxO+anw%MV<(bYq5#QL@?&$9}E9mOt!yYPhfLyOPk<($)-DALZ(Z<${Ys2e!I
zEf@cqbht`&4<v}sP#}Lx4n!!n0Yy~ga^ziw<gSkSH*`h*QC-LW26MvjQwQ^Vr$sI=
zO3>b(^c=S@?tyU^-%7nsg>47yyUusdaMN|<O`K4>9?WoX_ry|GMNc@rE*&1ZWvhRi
z_S3m-n`bk(@O+$m8Kp&CbTqx#!3Ro^Y+#WD0wD>iyApzndvKsR&ejCNHW0D$L?Y{&
z-+!(ex(HR2t+z8ezJ13p%e#_pEzeK4BMasjYdnwJ?s{s>RYi;(GFI&e<aS()?XjWO
zoYY<N7r&duY31K@BufxfDguxb7oZWLurjz{vg84UbI9BdjF6-Y{<Q7%utyW)+c$+0
z<MkF<9?8fp7m*z+oS$&oEZ>z1i>3C&Ca}f(#daZAeX;DyOc))i?10TL3cyU9coSu%
z)362xbDc4*fSbB%;E}Qhj(t3Y;h0y44|>PJ{Wnh8tcJHxpgfb*cWZ`3JyR%I1eX@D
zhJN6ES7k-g^nBRA6+Lz{hr;f=v9U%{mDsRr?D`_~Bl#>!q2hGpQrdVLN8GXyT@?Oz
z+7-##Pu6db%j-<<)!r^-I*w+`UXXo=8fYr1;~kdxDdbuG7#=k3gn<y*eN*f1KF$Bh
zuM!f73PNVVWu(3KcN;%3OXZZG_KI`Pw|e|IE$cJPXsj;kh%dz$%0ABb0gso)Q9gp2
zNtf=PDrDsl^fanC@8EI&q(M1hW$#)lTUm1o_)})e6B!%S2>wVoy*Pse*8?CP{EM4h
zidh=Lh5n)*Bd=X5LAxl_{94L7yp5$zMl^#3{EGwW`CdAsJ<I9#=;R3}l<zuf@=^R;
zn<3z5TIkm7d*S{Q&!Lv#!b=h$$&Th_>ZjClo-A{sUr5=r94uSOqPi<ojUL#vQoR0D
z9@tomD=TNeq@pL#T4Ehm8}_x9rr7iysgJ_9?J!1H=qLr{xNdprpE-Y{97Xs^)JR-0
z+mCbAPM4RgJda+_w1ZhDF<f#6KMK?RRs9%DGo70zPi6-^cvsZh<Pp*;9q$hp1HB?1
z1Thj8DEHa%3MWR`2XbC2FZ(&PJ%6R;$UL|(@;7M#``OoA{*nKc(^Xao#r-E)ODh-A
zOQX3n*Lvs2@mM-*O*5azC5U{B1MKB77dxh5mKt#I>$Bx2XqwJ0Xj>oqW`wvg3e07J
z)B$YP_^e2>9eu-`H$Ilmv=_G3H|}OW)AT=Y3o&4{OAN>c@maT9IRH=16i5c13XIbq
z?_;?AkC)0lxBghxk!|c14fTiYk`u>n>GBPA8g4GyLuofED1?CZ_ogTPNTn=*7Rz>`
ztYLiEcuec^&_!PYFq(!2r=O|z#Ft@!biey30#@bucj&Ux#$iwD(4dukXbV1&1KhNt
zqjE7f0?G2qOI$YC4rS^$^olhr+rM>-n6e_o=%}lSws$wIa?!a=?|&NpC0eU;zsWc{
zKWScWoG(_FP6kwvxFXl0Buq*6*?NA&t1N%CNhYQxAHE>%%=x@2fFKMkfr^ZzsGxgw
zHOYm3bg4i(w(q;Sv}7U`6Pvh6Lm7Ptpx?d8)x>^1RQ1|vuG<vm`}&L2^y60pzYXJ>
zD<jpjZ7a;%a|D&NxnAb?nXhfne1x>|M?Wx8nh#9<4K<gGSUSEv7K#rQi7l9~Uj!&y
zKe5Cei&k^WbCf;y%2w7n*r#WSHtKp+%roO8#6_LwL1~<?Pvf`t?S*N^=01T?r|(|z
z$d^E$%(J*;d8ErD{ZEFps9f+WKOzsubk@s$VbOjhU+t{w@7U3KK%cp%uNDtZ>3^p^
z+tzr;24w)=wRU5Hn4~bhvixzHRwBbhqo*XVUm->z)SC1la7x*{$-z(_ak+M#+ZP-g
ze<IxXQ4aEqNb_=<ma`Qe!{XjvZ9eq5Ib;szIj%cDn`nu6F$klktHN)8IeDy8S5qM2
zC%E^-H3a_+ex7p9xz-wfTD14=;J)uV*XXUj)pBRk@Pnh8CqmCZ&iJ1QUo^Z)B@vIg
z(FlGVdndb1<*-47yzX?oP5~QapP+;@NEQSQ)AQC1_Ty>JQP$k`X`41*d?oW^%4JZH
z`2Ld2K6XjjT<v??>~MILbr{_xL>OfMd9;zSl*--#DwbWDM4Mp%J~N_%)nv|~hmH9@
z)4zT59opehPLJ^KJ6=VVO6Ho)jzT3gR98(NX*61GGVm56{>rj2M18)d^HhPG&iNrA
zw3~jo&)!0v{GJE>iXI;ylLdla?z?dM+T(hzjWMg#Y7nngVIY8XR3%u9|L|1IzJP2%
ztaEj&!l5_dWLvR#((N&`TQ%Z_aD4=0dIY1oPwr0rTMp@~krDj&qrOo>N7s>&+3e2#
zj!pO%Q}P>}{K5VvO(@IX)|C%9o-~d!1M>GPa<ndYmNv;aFk~!<is==b3_pJ<30VRo
z#OB^{wUn=&24gx1tf_VnB;eG=J!0z~bgcJJX~-Ya3?h@$yfOIL)G*(evb?a%VX!~f
zqp6{~#WEH-Q1?9^ryjP0Z$$eQAyAnI&7&uR5P|1uwXt}Tyv%;tYAc#KvBk|6x%}OM
z9w~hzs3Y5-^N5AaGZ~c=G;%D?B(WRA1c1N1778yvhoaD%3rgBv|7EatdJt$>$u9}~
znp>y9i{nXjS1MjmdxDO!=)mKBA}-}mBp?edtY>D#G4~`p;ncc6ZCuu^9162{V`}LD
z1V6Z{i6Rj#_U_ZbKJm)mqBY>vgDo5E5h<Emio&5$E@AF?G3Le0&bO!{+=wz|)7zrU
zih#Jkm+X}@+r2~Gu{6>b5`g#UiRW{W{;Yv3BoGE+jQunVS36Np2enN*G1C~E`p6Rz
z{j*O=Hr16UEyfQtZjl0gr^mQuLw~P`!;^`G!)B$eL<AKUJuf^<NAv;d9AiDUw_1$1
zJPo3x5S=Jc8|9~ObacN3AY*5Q!3T7T+fx>?HKwu<G$dMGGeKe#-WS@H^V?x)LO8fs
zqmwi>JHg*Av%bEfp+&rI#S6h3Du-ZF@PZN*czynUs7CxrIzi<4CKW$DVUD<li_Y|i
zG17HQvoz)>Zp4t@iHd(8;)$gN|7tcwXZ*}bVfvjelr52^Z_wbOI56$NP(NgU%iVHE
zF4%~Q^N)t*gCUrPDQc6o=jm9zod2D$s=kAl9(4KKJVtQbESUEoxkjYn&#4)gw9Hf$
zxkomp*C}$oszKqQn1bcxZ?n$jcvSPa-%+!QHfnAo#hP0@4Eu7#;_sn^YAFm}a5$kj
zR>ve*17@Gze8vcL$w+E@`m5(=Qy*Q~>OI;1*l|C1v1PMWOw3eL9{)(8=9b%IM}6!U
zMcHwGU&b-gd%E#WWQ1W&us{G=3v*Gczbd>nU~XA+RYmPvAK8xKIjItpke-|6DHLM?
zV^m(Trt~H3n8Djvl-rxNw5%%KO97+CBR|c6-1xk!_MSCj{y-=Tl)@CO+~o2vmxWhz
zY)tX$O0oEU#QKS|IP#<h`b6RAJ-6|QMRwxTv+S{FC=#_NXFVnwajvlc^v;Fe8pX9A
zhf<6M-uxR-uogNQ8hEK)IoW(Fw~$RK7de(kPizF2`4M=?^(HIkpV8g2>9wssiTL8{
zJD#*5d^it313T$)w|h}CIq*lseFB}(#~)6PjENhpg?^8Xm3F>+oFgAQ0$j`746Z8!
z4T=vRkY=AQ@V*?NvGxTdbw)Lbpeb#Hwn$jEkRl=kfxy_mFdnIZZ|v?Bk47e(MA6iu
z=n4|<iHvkUa;Ob(KQEpZ&J@0)KduZPmaKWtq0k?3_hJD2Gm*qyzI97E{Lc|!PSUoE
zK1%~C&UNiRN)zQZhN-r>Yw5OvgD)#Z`yIX8TL)kRf#j_X&X!Xre9^FbrGeT&ua~Ez
z;L~rWd6P4b4BJHLCtjp5IQbK@Yn5nt(<W&i`^gZ*x8>u(A%E*th>PWT9zmg#$i405
z1u)h5>|%j;2S6YL!d+3OTS>AcGo{_X_PRbtC^cZ~OZKy980j48DF6mKe0<jN+^Q!d
z6_+wYh)0B~28opBG018{`&UQC-Z&pbLmX%ur<=#gp5NWH^gbnTA4PSVe6dX3@YPg6
zDr(C&J6Z51?||W#t8Zxn-kdrXW+H?^p#W}niWn)GH&MG$2$2AjQmXqq)5=*Mm#Z+(
zS55!)cU_L+xQ?>`ShOy)%)$Gpx&xYS;yMa=?0Luop+4>{HHg<+3oxDAZ>#%N{Z}PF
z*y2V1MAwfF%&zQdI+R~N;zar?o(NK19$nQ&-1jXxMS(~iDLANz^Z7LflHO0Vw(S<{
z-&<r5eM|PgPITIz&u6(UmdqwH`L~qt)9ZR|&VG<JT1tSay74fpHXwRH=;&6EH^5|0
z+LuTHxkiqxO8upJPHvd7O7}8jG<V61p27Jt)N=R14)#js<10eRO&x|CW~Yj>6fR1v
zwYJ=7+&AQJkPZmGr9o>Y;3Dk51T1!Wny3eOhpSrVHP)r8j)n->0clR9O!IoX`MucX
z^Zwiy9^o(E6WZ-Mwy{?YbRia8Hzm{^kdB^~ULyM{?&zV&+;rnvab{}4=iyZAU08O5
z@MV3_6vpe|{fU%r_^adQ-fn4Eqe6TqJ&VNr&47I)qv~~M_?mkg8<^CFz(S?YiNc&J
zF8sSlt6VZ89ZVI=m8J2(?|e_Ih#V2_^euv=$5qy@pDZ&CU`?d!{!Xjfsm*EywtUo{
z7M`bz>Lm*<Ce}MtHYm{7r(J5)5q}n1Vl~^$H^%&tC52`G-pRCoZdtm7PEwjEjW?3{
z11<BF;^@L~F7Cc<j;XTn3-~YX91YzpO}BG$c$6|oZbk!Z-y5|EV+OKNAcMg%n<sVj
zl-e-={H)#pfrJC{pT?h4SGInUS*~;n1&|2O@-f=eNKvRZ4R^(dzj~9cl6p)6{Kv$q
zqIam>;&#qzeUCS&pM&I|_VXXi>@ekIJ%6OP-srY6;NEm+rQ*Vm2SP=dpg`9v=8of0
zhOkW8VtL8ieYY31ECO7=oqO;@Z;1Kzm!dv@X`5}nJ?l8#I_`)#C(=D2%+*Mp=89va
z5suRsE0+)2#jpPQZ^r2U4e8Tn0Artnx(_PB{~E_^f3atL)FN~J89#@h6>YEJ)SkOU
zkVI-6t3OWeJfD)pik~F|nR+pV9<?<+S1*Gk>ikPTcq<UXdw_)K!lcpOk1iX@!U|Th
zoEhV5Sl$g0E!rygh}-<LV39X5LUDVTxY1S(aK&!Kr*rX$uZujy4v|hnhJ2+@m<J{n
zzX?5@^gCzR)(UN#6V?@^t({?Bx@QIn1n)vheiuiVE^S;DDalwPZ+w=+Bbf|;LoX&9
z8F2XSC1R0>m+4>0E|mxI>`U=jt6gulFW~I2NBxSomP5IND~h{YMxG>1blL@4eF^{6
zk#UK;tX9dPQL#jH$V)FC!y~DdwEKH`o<?o9|BR|2<^vN!e^m`uTJGR!jYdHl+#*lk
zs5uVJf+{6hxY=%{b1<y_plLd2Uui24dyK=?>^W&`SzLFoKY%L61$Od-y6mNtoA!97
zb$UD;2B^}_laB+luIKy^Uxf-;n)u=5i3a`beLsWhyI-)zg6gb&zCf*CjB)ffyJji%
zQ#_YU?1e#Jdx!n7Hh(Ev_|i!PMx4H_`%&raZn?l8`iT!XKN5a0UHxpS!xGFhTNf+u
zxpL3g?fDyTpaY9;=4WvgLgnhu2&s^lz|K2HyQC6^rpuT~EyYofcy)zpZBfmg;sx&O
zRGn#3W%*3V<n*|z)Qkf>l+k#uA^Ks&Nc5APO_6`soi1jM35BHy;tl(LIhwN&a^a`%
zJuR6vsJ;uQlY9qsuOj!Xa9BQ;(SH{l%NmW8PkWZ6_$5iF;!BonGJlTSxyBnademO^
z=BG6}B=bz`T7Vp#E}gK#fA=$pv=-a?p8t<XnSlE>oJvm3oLf8KYg@E~;oi^ZJM)2>
z@1e%vtsSw=PX_(a0OjVZhHr*B{g*Z;W~=k%vO2VPhKypEAzx81pO<(tom_IjrLqcK
z@zif`H$2E)oZ<bN=C9xVdlaU6(E3^#$Kr3w>V2<j@7jJovp;i8k@<L1`K44d!+7h^
z;j!IeS*PpL6VpURLOzQr98v;{UR7Gn4&D`3s(EoL{Hb3^v2>FE-3#8Buc{2P?839y
z#~{t0<hQk=<*T18>*VWy@2eH+)q@uti{*C6u-6YLDb<8RHk){z#wL@!tX{(Ia1#tU
zO(0i6^t)Ueq*UJ0(u~F1x&}Xc`0a@7^&Z~y#d7ER`@Xo6>+bhoC<@-^DTsWJz$8O@
zE6q6A|0WUD9M91pzApN8$^s8ZUZ_$??^mF<Akg}=Aj6dC0d1XBWe6+yIu{JYbdiY4
zRV#EJzDK4PtaJ~0363$VH|~2P9v-U1O5Y`@`zw}ofRT_oR8hMLibAhfS|zpBh(>i=
z)P@l=?Y7Lm!mt}6Km%zR(K8vzVwM*Cph$R+Xh-GcA2iO3omIER78s5{b=pAhf9(?R
z=jTLCaD?TmqKr8O<U~<+4)>I!4VC|q!DCP=dfL9nazg#oyW^g{NeK5d*yo|ECkDoA
zntQ_k_T^y$F%++7ZZFRcq6{D9EFXT)`{k$vq&&@tPha9Z|9HA1$islU8DzR5%*4N_
z8DoCrp?%1ld;gLWA)zb^F16gFV+?4T=W$cQrr1dXp6_NT3tcXsDig~m+Mf7k+r5{4
zaX`hIACv4>RGTPbo0)OUixoJ^mShV!I=(iQq{0#r8*6Kg8efTj75)X$kJM+F@I}UK
zUT6J1kRe_tI*P!@&+7A20@t?a1+@I_mNWPu$Azzu(L~LJHNe5}vc>A0`ZhB2cvGWA
zAP_fmED`+(u^vqY)Gl~k>@lkPE7s{N6Gk9+#3TqD@!RmIUBVFrL?mpY%{jeTz>Ul#
z^|pRz2~Co5>*5Vwzu~?e8!^P_RU&lWU@N*MZCSemvk~1wkWh+dwBQK1rWMVqOV=%v
zyWG(np@yt5?UDMjtFamAdo2)+Vqc~!LSR~pYD@3!kZ<Na$LxL#pj0rAjQqD|-&1$j
zy#(nmP~x*E^M5Xg`E%7amTX+r3i*l@v57pn?=H|ni%Eevpn^1ymfX-wy$foPXGw0Z
z1mw|CuClWHN>umq_zz#BBQJsnet2!E?A{p`ONN@O<Arp+Ey-6;LrrroraY~_4`3M{
zSw}L+i>d?Dfmf(G#Pe1^hkYDJ<6WdPOs0PO-=ay0Yv^)oC(`N1#}Xo#K?6jVUH7&Y
zt>ZCkQuHIG`z{jqel&L+J&lkI(N??MwcMz<!|5hMT0kTAH8^+~^b&)pnX*vkErQ2~
zPu@;et;%C0=sU%49+J^q3X1pBn*(_u(Z-0MEB<C}&nqd24lK5gXP&vny^1Gezm2Q_
z+4}k|da748EB;qKA4<mdW-dcMU?lj5m76+sff54`Km3OuVccWpy=)w?d#3TqxE`h}
z6l9ih_x!f;a;+y-2LrFARpjx26^wNDnukYzyUV0&I13UvG2=?cXJ$kgj4+>EWwuRE
z#tsWN%OvjLGsAyYLoMu0+RDt)d<{d{D_#C6oX)J(P(ujhKG$ds$ObrB+81&6HEUZK
z6)Ny?G2BiOc@i>AX>0zY@kpJT=631_c&}0lHb_-=K+_#y@r3`Ov!QI~A~_f2I%EKS
zJ7emxmhe7MD5eWcXU=kDb2?}pJ=Z2ixgDeUC)Cr0Dgk+Xip@@Z?fW>sIrSfsd3iKl
zA^E2N1so#Rx-^RoLChMgcI*of<7Fc`lfSl9eUto$^I#>M(!@{8#<(NM6gWSTnNcQu
z#F(n@U$X%u^XVuNW%;9#eG{+J3*S0eDbzOK*JU^!PFZiL;WG-ghu)JglFs-`?gx{0
zg^83973w0~@wfomO4tTCaGoAw5ySDvK+M&*B-%3F6upN6405V@Led1c)Ncz<5n^(<
zPiF~QP0i%xt-8+Ans2bE6p^{j(LYwh4VMN%1BUMdM~<t<Q@y^cy+VT`F&9XO-UmwV
zNQSFdQ9@(h9NntM?`(8xNGQ6_Nxr;hVsb>=lxfRktQVj7Tbz4GRfJ5l**g7#AP{Ll
zg26O4>0syPBWE01LOh{Vto8R|+hzwxjMgaa+4w4znI|@APC<L_p3o7U4Yvm^QVm_5
zOVTvNz*$GCz`?SqIMSzE7St(9XFgH5M$3pE98RZ&THMK=&w-ai#*K*W-HmlvcPGZx
zE|2{<GyASOzW{5A0ac5FB?W6uG!rdIrEI7rRrbvHHSzD#WAq@ZMp<5G1U+3Ot38^0
zkWUd%2mn+s+69nZC|8j!jec@6cB^Hb=$2uG2_CMALSGUF22lMX_H9x|#-j&=|NN3R
zm3MsbUbp3kXQaeGa7qqy#C07rvSnB^KY)88=v<VG3e-?(hSrV*cU%;r4bEmZXYHnj
z=YJxI7X?am{GN_kF@k$GTAc#_;Ns8cBGR(=(_bTbRFcwwQMtQurq!C}McDx^cmimi
zLAKxbb3<jnqHw?7Ob(Pc29cn_Of$5{HnwD@ef{P(d^BGojR$WAu=r_gjLP$CQG%I$
z%4-rTB?K(yt0^=2p#hk#Ngr`76lRn^!7j&$f_FU~%2Iu9H{5n#WSYWDG%R%8>}4r-
z_j-3kt^W!$v8L)O)(h**BQ`G7xTp5~*cIxL$Cbd!4BDqq1iG#K=~<?S695vr6nLB_
ze5lVfgCMMs$o<^jHKS+&C+0>9AS9TqsXP_zV8f-mKeyL`)N8UjxE6ah@%CHBRjiYY
zIAN$UE$g3x=1lf{!xt1ZRgVcz_({~_;ON4pN2KG(x>7`@%V6tAaX}iC*iEyNuh{Z<
zP4>B7WT23ha(Y<Qv)`n5lm-6SK+9jP-Q=@WZuFraml@kW?^>zwOadZ+-><%ECjXix
zzBw)Eg|m*luIPw7+s=KKGbMdlsp~TSU^^P?aC>~6uI9b%gV=r|dD=H2?R4lgxOouS
z4y$=lE(0JSD=3f=cism*huq^ph>fl#eruAl&Qgel9&OAcjR<Alj3|8+Aa?D7u|RnI
z4cuE_e%8kAT`@G7_haq-o0?hACI10@peGt_iAj_kTckd!D#<{VT33EZe@1Y-4A*a^
z_LEZcNew!=aM4z|te477Rk)Kg#_|Dfz~k%pOW!YGvYaJ4-%?DS6ZBjA&d8C~ri<ER
zgh~TqKZmiqZvv5nGo+gIxf6*Rwk_>k)uVWD_8t)s7{ATQEaE!nr^~>g=%(4pS8+Bu
z5=G_(yS2XmS6Y<<>th8PL}|H=vh_+`rF++VgzfHwF?)POOi$iLaTu-X)KFZf9*u?T
z?)}Bqv}%JriAp(<V<mQC+Jv+oFd~ryN0m6QSZ>#fdiklAyZx}<fs<VM%#R&iP`U8S
zOCWg^;(=)Eo+T0tz=;EEkv+9t1OqTPasza(9LhQ|<YBtAV9Rl<t4SIz+SeaeFy7sb
zJmT+NaXYej@T-2JltU&JP>iL7=i?Yhzm53*70oyBuUiOR%j>5$SfU;Qg4%7qnxL^X
zxI<aYkVGyg07x>~FR3wjB?`nEIEh!feQ?EB5fxl_Xbb9SubMyRuRRP*Q$EOTP&drs
zY?gCRJmNK?Yq1V9fe#~p=K6_U2~;O55mu0P&seC38CLy+tv~Wp-x`q#x&>ZM7-Q3m
zW4eD3)V})uA3+zGj}#kgL19Tb|B00v_HN*#OGe|6oD8pH=^tCuMly2~^7SO4!t)?g
zy6@oxDL?L3C~3=371z7zURp$Z`m{P~37?4aXWtd&fX+CBi)^dH8Pk?Ah@fkZ()!`)
z1X0mZu{ka}gkauYDfc*`#Qvobr##rtiB{U4PY809iCvoe)JrGm!V8g_?{9}sjEN0=
zJ7<CK)qfT-JtwkE7tN=pto<cUp>$GABFPts7amvF?Wxg3FuLN{L~fQO2Ji_C{H?Zr
zfx1Mn>%xiT^c%1j?7^kj&;Ai(5oLtxTKUAKq*cME;>S!-zWFU`$Q5Qp+Y6LFt-d-*
zI|Yi&1D5sozyVt6Io*3Y(QbMUH6Zt&!WxBbM#)2%rL4sK&J$q7R-gKw`9Ij&^fvO1
zmLCl&t)><H@>E;}-txS})rj3hexI>&3B9Lw&HJr;em;dp7)6hFS<O(ccpsv>x)BpE
zDm*>V;wiw)K?<mVEhhW;@iA7j3ILNZMLoPF&x{hn&$8>Qc**$GE$ELU6tzKjmkL)s
zu`UM#;VDh!tL`#Wf4#+=lYGrdZfsUx$JbO|r4oTwXfXkrg4`=3V%Izv8Kj(cTDGhf
z58eQ=^xq<-^r*F|u>!{;_Em2^J)jAy{&s94JOl#v*YHux@HFrJNuq_KMj{cd4B!GI
zyQy%L+U#GslYhy*L&uRMJLVUA4D-9ylSREQjKgNu|8-2>jqtiW+McHya%Ket_zkuj
zm3$1|$Ks(pmrP4EkMi}jCyvG+`NH5bpZ;|q0Y&){coHggK%;uf=ygA9Gjg@-z<06?
z?Ix4=xt4ntSo@+9oy6cjU5elzD4ub<xZ(T}<I6I7?f4SEItGZCqH^<VV-82;is}N|
zCJ#Iem)VB}DuIbW^(ifRZW(vT?%x(jWuuHGjsQB<=Ca$*?Z803l=!O9dzPD^C0fKr
z2m~N1WJ$ToKe1jEO_ejA(9L6h#ABVRAa=9%1Jyi$K$hq6Ap+kBGrPmDUB7Lrbp7}D
zT~(ys+@NI%xgYT&n2Q#8OCR|4nQ!zGoaEPRz8-8C*(dO+(WTLQqg~S|jv4N|PBNb0
zjg))lyh7kNehT(miL}8oD{hw%^MCcSz(ND4@xi<I#8^W?mgCIKl)sq&d`dbMmE%a8
zZ%Hd$rWX!{XmXV0v02hdx+S*&R$ekJ@^R2dY1h%ElrES8zk8r$p`pt`@N(yoa$K8(
zjhK6o-347bbur0FPyBz2^m{=@pz-rv&bObZwz1EzCUD*?!EiX3xSDA}=rb9C8xwGC
zNrz1eL{|I#PEL6#J$&j5zf+699C`UqL5~2qV!2i!hst1Z=Yiy{?8%n+gz3Y*e1piL
z6O(<SfzEwKP5Di6F}9!EG|9*jG3lry?fcTF&yqX}`=^~vl)#HXi_dK&J=(Y|Z8<*}
z30@Mwy@}{(<;t;;0%r<52#kSC7Bu51jNyy-t#x>WSy9ggSr6AII%4s~kygw*$EfUf
zCY$#?+H2MPU7Dk3nKJ$X=vT#P&+<xyH8g+ku0O7RnV>45%*v>a)aJH8>pmXiNa08~
z#<?i9?rpIxfwN-`_yM8u7UP^;zQna2AKg3DlMak0P;y?5eH=PSdr}C999F+r^0L#3
zC^_JpLv#E;X%i<g;;`yFp%_PWU!s&&mSJg0X@WSs0r!qKjk}nBa(sb>uSsyGl@a`R
z?7#3vi8E>bKryMKl1od9W<sxngnJ;(4%kmLe@X$yzeNlN(!>2}*q@jjedoHB;K~H?
z;h{>mbHAOQ?yR})5R|r$vm|$Ig$ZnnL#t*8KK5}E+@@Kv8^(9yG?EfT5PU1;lin{4
zLSUjS%c8>N;~$k2tlzG=zm*BM6jW_WNrM*T)#Gz|3{UBi_BFO`gWIeuu7`n7qJ7!+
zoB49j8dB$kFYl%CkKkQAyfMq60G`QsWATY0(n^Qq0MqRiwat%l{@qJH_O!Wp|2ASV
zp~0AbnKEINW-+Zod_NkA)8QUU3OqKkv6{WUv7umiWmN6w4y`-SB|;BV;-wz#^2nX)
z8^1f-0CkHspK8Awr2)AoVa;dY`-(cwmj-#a^qWnY@txS1{hg<?>b=G>2m7{0c%!)H
zbs=y!>P<LW&QMLJHU2(FX@TL%w+0?R_@3OuR}@{Lg&|edk7qocJn&~JH@eW7osFA8
z${K&8_u3o`BS;wP(94`+cY0*KMQ^Pn!zqW=({HA?)YFG$Dggy=&6wNiOg1+<AFBE@
zZw48Wylw=jke|06tUZ<MZt;wTJkCAv8q;y;Hw%GG<FW<SZJt=!e<_wD0%!MN`hJPm
zeHrurlz8sVi+qX5ww`QtsWJQ+HeWESuU#cztrb5u?LN_a$E%6jk1*~Pa%#{$x4pRQ
zkb}<7AKLw!Ny9$X$MW3rj_>=MRpDu#0WZhGe0hmuS$N;@)mS|*4E~<ZFs$_6CKXO<
z1gc-%wS@=}LjMsBXkO0zzU&v~<KDHjs^+$;w(=)gvt-zC>72mu{YNq`RZI2l38VG;
z&9cBZbIPwX$TSBPKN?k&W?cZIRHxaR)9Ch}tUXnw4yeb9w9Y0HH=?rzg6qU0_kXX&
zikwQyBcGjclOqS5tHEx2(=0Eyz4ua<7wa614t|PXbnX@1^gG5j4}b>F?#wG?KKQUL
z>D3%@JC#ak1x48sS1KpB=kC_yJ!7{-np@4coMInIt48ux_LJd7=zo%8N)@-)@Mqn?
zCB{TDMLie03HEs~sahW1-369Q7k;$b8PLlKc|e60hl^hIV7!27_os@JOqQ{&^m@It
zOU+BY&Kd1I<4B8XNEOGQ1fxpfLKZXxc`gJZVp(4YLUrlS#O3!<O<<Li8^|ViP~L7q
z$w48`7Ze*B>F;mrX!*VfOSO6<2k<HrvugZWKTT3;LK9K#TC2QDfk??*7a5I^N#_yO
zlk5BFkcI9a+tRGJPA{oW`Bx4JhUDGs&@j*6IQ?(=Va@q^QQIdWfsS)iGG_#=im~(I
z#@c8z_?^~$u5IHC!|pN^Wg8orei*{_?;yp&F8Qjk*|OneM3YNAYj1+?PD$;rZwV=n
zzRaTq6k(wsa2;IUu}FF`nav--XUmtN>839IE#ij3B6`Z33p3V1pDGW)z||*)hzf4|
z#5U$te~Ik~F3J|qKe`<wNN1Otc$I~u5`XtwL{mgdIA4xb^cB@R&G$qUX|DV&YaYo%
z&L(UQ+fbU3rKLh<U@cfzC3O-2z?eQ_-(cwL{I(P80|tK|D1sOdPghb|TlSh~KPvqa
z>0sb25<>Dt`sQv!b)&q;?!Q=CJ{$~>URp+ihe!Y><Q%ep+4I|OL&zo!K@3$9_XK}G
z;>>%EQdotYO8c58TuoTLmkppGqebiz(tbA&68SL)Ym2PBh`wyUU1P7^o-V&xxQ%&I
z1`AGU3bH%$YNuf^IXQ&0!Tdb$I7(3)vaau0-GLq!gUc*E{8a*|5H8aRMZy`Da}U;Q
zk0R&2Y^ra(MQ4s*7$;THElm5vNno7%&0ev8lfW&rjt8_-k;O9IROA2H67|DhXL*}G
z{w0f8xhpks8E$F1j1p#xIWiXZjBfJ5^IaMxafGqW`53v4sVk*MQwPMFnyS<{K}hw@
zd!wvxfN`sRTTVb-x;RGfyoA$6%2mq}{0gGHfQRm0M9P^IUVT{HFSh1`bO^0`(macn
znI;iM@Bp<><Rc~X?fKpyA~tHnvI!PdWl@Ff{c(Mz{#_*v)u754f48v>s_2kCwF=JY
z&PSrL)N>3YjsOL_gI3lipNJ2XrOul1Wkw+f?&ac$$3qPNi@V_?<6r^H*bU;p$q=>>
z1HvZjJge#_N_vj=kpT^tx;Xz>wFlov3AYhs1rZwWA82QuRk~@988N#nN1&YOzQHa3
zdf&p06k#ycU~rClDyg|%s8er0F|}lCP@)Z{v-VzuK5T>5<=`+i$=M^z<d&;ru`drv
zD}9p7>g~GidW{KUn8}&u^k7nksXyJs5{jRl`tHwI4aOs5*?ya1z}BK`rSbW4Osuwk
zzP!7-44zPQ)BpsgLeCxhiNT!Za_>?f*CXJ;7qQ8#<^K=Qzj13VkbE1||LKI?N#Dgz
zIi7fmv}!*%-MEm8-?F^f%-T%3{*>l+mq_<~HYo0Jb>M|QdT}F~I?YSYm@dv}bQ}pW
zFz@r0I@tf#7ANC?Q{`j!@F@?F3h&0m;LY#5vfPc>xE`I4U`@X&vyP-}ERVw+47+d*
z8XbB2sGh5Hd+{W*%UR)T5f#vG^F8`sWtR@~8)B#C?QQLfjP@T}EoZ+c4rjM1ZjGme
zhKp}sp}+!o1JpBW<=)Fa2TtF_-3=QXVIC2VH8rTfxH&^ZfK)EC-4LZ7vc)gt$j;&B
zE7W-upbGnrXAHR~IRB-K{$r25*hy70%I93yIb>rw9p~$V7vkmchY;zmv47Gi0=ZcH
zG-@I9F(-ru^j;GAl76&@L6#uC9|fE$kv7K>zz5BI7}0z>OgtK##ulGp-w47^V_Q0#
zHuCjpJ4->8o$~7oo7i0@Y29LCCw36}NAQbYir!Nzr3tMo>~3U^4m{^j8Rl_h?YAiE
zyhGp#8%ummkQ_Q>=-qTpeH?NTNV6{LaTjc#b_*38J~BuhbgN0&^W0t-1FuH~#hhC1
zMn{OGvG&h*`xt>KB~YjZM>i)`Nsmwxb)F$$r<YPHwV!0V&ufqKc|0#JGL-x$z7jU6
zO_Psw{GVRRH@I>XEa__@Kz+APMRe*11R#?(&0!_hWIB<((OfR`TY$wQ>BK1O&x$6u
zr8b-zG`lss-&MNJ<t*)>83H{XmrzrxLCXP3zx616s_fK!t~=CSlew@PbyMR61D!a|
zsLmv7%{S${Gp_IQYMXWX0v;SV?H^Pci3At^G0=Y~46@<M7=4?8k}(Y00h#Mf_oLZt
zEaKFb;z6mASb~Eguu=$Q19<su2WT2XS;fbTS~>dK00jSvyO&zMzs`y%Yr2nYu6#9Y
z*<ebMtaZPY1pYJH@gdD`b5my{U|!8nF|1wUcOG>YTghKFZCC2=D}K1OtgE7My;a;3
zs<??3)@Os1_A7_?q*<m<poqUE`vs{gNE#3&pb~Jtpqr7P??e(>O4r0%Bicd7q4#W5
zm+Xvgm|hj6LwVVX7t5zeMTO<OPj`cE&#GepIHe@n2rhx=mL&(@5j6PcH#(beQ9WN4
zyIpJsyi|#V-!H)R;a@QlA^Sv^gHLsmemx4LH>VcpsaPSw)$>PxMkvdb`kyB?G;aGW
z#kZjiTUtq)U|p}b)@9@V<bL(e^H4Y^m}z1uIIpXRTP|1tiM}ZzqJ;t#i5fFPFcqfn
z2z=owWH=(tAaZh0Ynt=ie?x=L2_;a!FftKYNz>#PZn`{gD1Td0g|}}0wF!xU=e0hy
zHyG}uiLnRu=sn7Bh;pjoG`}Fdv!-#0vnYegTi{ni^fepVyOqj1R+r<N-dlesR3mD9
zm6yuO68TADe<cb6Yg!{id<L<`O<<}=d~#(&Cbq{P4PwJd2Q!&O0rFb3C?c_?WF{aG
zZ?hU~C0biaE*z>>Tla9W>4-K)UAhR`x5(nWp0izSdW$kHF1DHoWrOr2;ljn~@&VXv
z!;to}?n^_JyK<p97NU<|6OUgi6*^+*3?Ag-1|ge9mf(w0fQT!Dt#yUCTC~%ww`fXn
zYe%j1#Dvpwgl2}sXR*)Bi@5Pam3~zMMZ4~y|1|3I$4Tu}z&RUh&RMNk?j%v)?=jcW
z>~TjZwrA8#KSX8dc!BRrSBlPOK7p;OZB1NX&jQUu>3+zhyr%DF>>dQ*Vj_VXb;zX(
zIo_JSRUrlqS!=a;rg_c(E}I>bRiUcdPKf);b;qbq#k^^jYshOCbOAWZUqIUnQ9<E+
zRok82*>fUKBEK$&)hXv1&qdgv{1eugEQ%@G{a2OG^*wXuG3b~$Zv(2H3pToh28)cr
zYev{e<VQ3a$$qKuIJ2AsLNvmpKCza{4-AWhV(06z`y6=Ho!wS{5X#=ko8yJb{f;7i
zLZulxI-bl`%@^msa=c$EaYC^O>l;D$^5k5A9V}#~b=H}vTS^YGz0)O@*21Z`O0#~2
zm-{BE`O_B3P}xZhi&+@oFrY>(h+qjmQno+gN&jLJZ+iW2Lp5>J!M6k+iEQybA<igG
z$gTS1+Cf9P^nkLzr`p<6KqN`O)ppZ(S;pYC?O=*vBqr!*TnwJi-ek!Z-fC;Q`)J=X
z!!_TEG3s|BpO9!}p&gL_)2w|}bh)RXA>pKr0+#up4UogAeoM;JMHEB#{VVG0Br(~t
z;o<{@RWdi)*lYX)(Ellg;`^QWQ~uKqQPcgOc8HPgMHiIdN%#N6MK-MwOPxW=U$AsA
z6hD+|C?>Sb{SdlDXTqSEA{YPm#r(u&eVOc1!~;E`A@Ii?TO5jMV6eX7U#3^=(KbJw
z)X$50!WpsTP5Ivi;dG`1(Q1Mn-iO$?#VW{dUm|NOB$ja@QE>~Ph#-mNJ6c1`0#bL8
zb^;qmJmr$E@2Yme^?T-@OSlKhpe$oGY@`wSL5|#*;yAw8FZzoULOkIY8Pjn%67i)}
z{l}6W>Jt9?%jS@lTQ|d>jTNm#PJF*w6sHB{j<*k-*5Fdj_2a$#24~eH=jC%yHg2!H
z2%3x{t~SIdKK>vjhfqvt_1%^QjO=(dj%>Ar?0bGs(A!D$zPa)z8Xsv49exN|7Z+O6
zQRUPUKl+X}fXYb+dctaPfkRo?R%^7-e7>~5?UUu~$KcRYt<o{R+o1Mwy?Wcy$&MqM
zX6)=qi6@i8KsE9mLTjqme|l?11n=)oV8n8bC54`TgdpCv2(x><>l?;Xf7hVW&f)4q
zKa7KyF0KN+?^GbTpU)om1euxNw5TZ^rg5oVeTlQIZ`^tJb~Uv$0s|b;q%9c&;>51e
z`c$)xs>`*eJ`x%qvrd4^9+drc4!I@NZrA_@HNnBsw{F1nVkF8z@ec`yiSEx14LDqy
z(CcM#uu}3QSi0JRtqHAV&yzQlJ+KM6c}f*i3M;DrmHQUjI^~nLX$wx^x;4UaN$FVt
z&Pmm5(XnW;jlm;3k}hAJAQFW@KVom}CA#H78GhaYQ}k-~GuPz3;-Eyv;kR5brwy=B
zfR<wKis`>_?_yPB`{N!wjgu_j5w7@JOz2OZGu;OL@xz{Q;*pimlR!y=pCB{#+t5AL
z32+inpvA+yi-KRXMGWN<G-5CAPp+`+`51>P^XPZAp}vroYP!sQG4E;+EVla%vHfn|
z#|EaX2j^1(GZJzEND26^KBVP~l=+=c^IXE0_g9ivk#n7S?_xastD0^LpRc#-hdV_0
z{hMS2L%wspFLfHM`AppRfmS04pWmTbkzSqy*eo0l65!CmA7^`)_Abq90Nw2}UpR@b
zjs1LC*2a%m*M;*QVbo&1a^QSx7#|CBpliq@NG4%{O(l;%KbSXgE&km>;pUK}O#LY7
z=hKdHY$ZPGW{US;_!1p?0+d#mw!Ifo*>HlFQT4NYKK3Xx=%8w}&z^)%lHyL5sH{iV
zQlF5F9w7Upp0rr!ed-P?%I&>cfx#jb4Y6>zdKeJ+1*oh}oS%=)jNbSgBl(?h{7G(h
z3mkpT=d=<%WB8h)flTzOV9O;(hB8mL9kMwruOktKJSwJobi9HGFIkOasl_Se(vrzI
z%21W_{aWwfkgZ?Y-lndle*~}utqX+WJJEhlYJk+i+X|(+ms~(05=ByP)FJUvJ{*@t
zA0UuB_vEcydLk(s6&v~4(`<o4;R*T{&oB-8e<%$goyxpQ?J7FYYc+<99Q)lgt?@xA
zX-rc?;u#A1GRWL$gJsFtWG893Ku3h=i@qaH*m7rH?^fI3dBIlxoT$9-J?M?bJblm3
zNSf4ot>h))Cc`&Un_zxPiYeT<0))7e6|JO%<o+Dy>vqVKb!nHdzRPQar^O`$jL}9z
zcyBLNT;7G_UWPng87D7{$!@~X$jtVVDcw-^!O--QX8KHDsde0RJpzB=Dl8CGTWVX;
z@z}4;-=5e^MKO4ZMW5ZX8QNt)v*|Dr=8Jk=B(Ie^;aW$=do+aqxqxHb|3JavGWra5
zNAiX(4t5fh7us?>RPI#)^Zbo>R|As>Q{=BSGd7fK<Qw7V=POlq(v^hD5W`{}-=-TG
z?Rppmv{;`!1mS<Pydzg9Su&G;Nr$7{sv4_rXpAg+zV<=?5*r72EPpGHSIds~KTiFc
zid5uOgrDV_@}54LVWCG9hyESCUBCM=!p6trMfDSKx}yhwN`2b)sPH_?H#Gy4Z&@}?
zi-J1yh5g_uJ$K5DMwa$r8nKfr6gUS~qXm$K^ji+|s8C<D(KjLf!J^i-ww6}Zeuc+O
zkGmNBW(F-)pGO_sA2+=IcFk4BTCtor(T-?tqd2Se$e|)bBICmVjHoi24i@iUlXxRC
z7h&F0_niYUl;%?K$<?)UzeUsIwS96wnf^rnhCSqZ)Lc~7SjHyJYjSj<h})i~cqo;B
z<!FNZ;9|poC*VEd-lpWNvmJMdm~s|4<!`2F@Jj=GBLwAG-OWF0xtFuLL4ct4!Ryp{
z8M<8OnMs@=SZPLy79X_Zri#Zc*~nA5pZxm|HO>Nt-!1Ld6HSL-vHePI$O@TM*k}5G
zq^A`yH&o-#vlckF*uR?APTJx>E?Zt)SPk|+!z72N4GVeicTyS;X^jnzEuc?(yP3P>
zlN08Op$ef~naLYyYId-keedgm&K9%#VT>(izue;kL4<Y4r#AKvbVBM_M%d~z9I`8}
zO<4Xscx?5Wq<N?A*?YIla({jg0-q0{zTX#?(mZ?QNe1#qNwf@mHiT$g$_#1|qw2FO
zRE3Qn>3eqt%)>!v%o88*#7LwDY|8_^$tgu9%gq<`i_mHPkDUVII;Qtw_RB+^7)5mC
zCO)~}!GYsE*TkmRB3s_G>oXw8BYs`AY5sg;!geCmo1pHIex)aefr2!DvA#Du%ja7g
zDT(m(jrA81Ja0NuFR4?npk#SW=Tgxj>f>|w4$m(>3*rxA$Nt%CR|a@Y`UTlmzk=`{
zlJa_LK&$aMpOzv$X4@i{)S2$*P^*VjVyi`d><LTua(@O35dB1<y&Gk`LK{T(DyY8R
zt$d1V@;WlCe!`me<h|mj>`~!zR2>ac!W2?GYJ3BA%brp-`1T-|#o!YMKf7{>+eEb0
z!+5L{^jYlE?$bB4;fJkXuVVT8R&evJ-b|1^3j?y}_D<b~_Hzv8%bqq0jXutN%#N{S
zp56+LREhdWa*YoBWmMqM5huJdnuW}*xjc4<-!<yQkD?eJD{Q;XM(<yX?`p=F2*o!?
zn`kWJt`8|cu`SQ-s0u%(tB!h6I41m=3A7kNZA{j^E^;t3FxiVyKC@h@$qNeCkas6a
z!XmbjC)uAKdbkR<=q8Q0-r#r^_XywwUeXdv7D!0{LJ)HGp#a@e=Xhq8Jq07}zaP#Y
z<Ld?88FYt#3t2=VVcl3z6mP#O-&*edewH>NB@Qmhb~hIpqim9aAH5+>c)Y#7rpKts
zZ(aG`MU5}{u>`^kuxaA4s219<A|G{NOKDv^1`iL59{>GVYtPiXZkO2mN)&g2(3*85
zekL#c#|vwqsPkYpbh%Qbr(NbNcExY?J_YDh8Z^w7#{*G;43yA?l`OC2K^dy!{v6SH
zps#MZGnK;JmJV?nmva<nJXxc&4RYOlv9SV(r;Q)tIq$Q{pIz%CD_fh5NcdZ}s#nkU
zL<X#+F*O7vLH-~Sny2P=>Y@bw2t4~(UEcWEs7l%MB<!}dh)QKkNlS`fL!tZ&X$I*W
ze{f==JFtWVr>_4?CHKX#Q!hWaXl6K6-TZ5uZ=I^?gOjBz|5}r&69B4}z_Tzld}wGy
z?b%1(T+-Xdvq{HW@~S}iyt|<G&wq{s>ILpxD5FIltHoy%AD(m<*-9OrjE#=`&No~h
z&a5kP)IFE8-cJKe!KPQ3n0;~FRcZ(VX?#sEJ~^5!nlJ&J+T5a$c0~u#e_WY-)+gKH
z{YZ+(zrN=FbSpBh_D&uHpO)o?oyPXOrEcIeidy<KdwN!Dq%<0}?*eotcp(%c^=+zA
zKA8yv?l3uQ#A&X(-ptDd5i*?>H1w<d%ovWkm7jj9q7B2GFIt>R_`PpE?ueekOtlSq
zD?*Um=U0_Q!WP0HmHr;0IDH($;EznIm)1y;*Uw?!R{{+X&=}A!?<5WRid`iRdZFm&
zs+(j3E{U~_ZlOKp%p!IEGLh?yA26aZ^Jr+5KW(&=>FBfk$ha?qWhR^MGpKL!pMKr$
z7H_L>>D%j8#_jS#U#seThtMgb!S$HNFps~D>K6`Rg~$(tc6ZusJL_?`wc1}j9<yCO
z9scI(xa-mJZ;=@q_fCBw#;Z&(Mp`&azZuk&vx=6taMInqJ}D>}iJbfaut)2sEk--+
z=k|Oi_(_-E>o54eBAm;1UF`l>4h<9QdG{*Q0M!SAvrL_28w3V~MH0;eJIaZKp0v29
z&8L=>zYp8bg30W+;VutS>>Hof2%kjM&0lXT{-?6DY-;oS+IEp5#jUsnEAH+Nh2lYq
z6{p2DxVr_~QrtbbyA>}E#e;i+;BI$*|C#3lJTLPmbIoM!wXeO`ah$Pjy|bz4_Md;~
z1pC@Ph0t7v#4(dWv8uN^bN0A5tHRbFqD*Fu_d=(_{>f6i@-qE|xqU<YN8oj_yB@#2
z2jz*7*GY`t=-?u|()ju^Uq1RQDLlhs;{PqgoZc2bT?(S7!J(Y5g@B@6x8G$f^tX*F
z%9EGK)onwTK7{&cZ+-XdI#tJ*rt9yTF%A4-|HMO<fT-?D9c;wwDAKXSuPee}g_}?|
z`V-|jdWZ=>7m}YZ4OB1(@)vk7l|OabbzUsXrP!g8cW9xZUgV;}&7FRAIGETMG4Qrc
z@Rcqr_DE;dhCVJEZw@J!9ZD-9!<^q9?2uEH5}_nyZc=qL*#3$=-TZa@m4Q!GdR6EU
zY4dEtdO*_8qNu$V?|#l`$KVS~E7m$SZ(prQt+X8QNkep)GiGJi&2G26{L^daEZycT
z0URxTOx4*qHtXF&(amaJTQ(5U6R32nkJcNKyGF-aYJi~Z%rhUGfjt8)G*Wh_n4ye|
zZ5=c1H%zCvqvVahp$pm$Mk{QSf!7M&wdCw*V-Q<*DX?w&RYHw_PTRtKRD@@Cq(zii
zt?u$wxSHhJF;K4=a%UZOB&L8KibDYJ`ufhIDyMtYH6BaRStSfkIWkwNG4l&t(SxYJ
zQLQF`xN_vv{wyCaZ3~tpf^SRmTuiH&-0V_x6!%xw6rX|$3ILYPtUHWr;FSzYX&~<H
zx%C-1bj$?Gwtn3H8oNz*VQ+ZFS^gczqb6EX5kJu%>@n6+zWdPH@-*=_8Us~~T#}hK
z@N*_gzQ<t_Jq<NQ<OvytK1Fhc%ZHBvS{3PC+x9Q0{t(?bPj~v+kMhM7WcVRoKnBg%
zAgDwd4&-yx61pBK92Zv+Y|mm1x4pD&7t-tL`2FFqX0wYHeK=s~xP965x!8l_CVVZ{
z_K0x8tkMprNd6zfnKV`j*+}r$kg@GK<UJqBgw+E+_F;;cO$h*q;#NIBAn<O(*>2b^
zx|WPHq$4pL3fGtHr0?`l)0fV5ivu;yzAajO`U^J&U;4RYQ!5P{U%oY7%`9|Wu8HJR
z(-Gzn&XqZwD{+HyJ~^i-OJi@j#631hjddE()`yHmt57~FEo>vS_;W7Y*7Lb0Gi9*k
zn6ht=38plebV1|SSqhBbd_+mA=y(3l&F!asQT$W6(CKy$(Z6pXFhQ1N`5w<2w)15j
z9odK{jC;(c*JjE&C%`l3cN!Le0fv>-!1OqU$tH*A(E5nr5y;}!?O8NyN;W%q`bp56
zU+>u<+@g=_-?kcDzZ^l_9i(>K7B!WJfC<RD0TbF$YULk$o|4$ZTK3Tp7Bn-Juf5i#
zmk(N6<AO6V26B^qw`9MpSlP(*ur(oGT-m-STm930KOTO)9_R>p&eEp4?|?8Okn+GS
zT>@*2p4(bjyL&ZTzw2()&PmDGE;^QvR2&Le553=b5P>jB87lho$Vtn?l+lg~B_KxR
z%i|BP(Tzs6D+sv*OVrpQ)s;aY6pdCX&P7H}O~-L`b=Z(tWH<g$;qF0h&_~1n^QHl8
z_BwTu6izC#KH)!)2mW5RbF#3bP8sYsmHDkGCsde^HRK$tEa~}XD5;G*ImVKpg>AO;
z0SjPieNi@v_gnd&TD;`@&{MWwPoa7#qokn6N#O3+-VD&~dsROYw)rA29lkd|5CR?R
z(!8zwUPV>1syn!rUoQL%UwMJqj8@iI;X5^}PCrMF*a9p@%>$|C_N5b0vDn?GY5<j=
zMB_yoaMQ1J1@AT`y;2t2BfLGpLzPcoLNBY?9E_@=a(V1uWt&(_Az^HzglWJX(l{ng
z$AM$i^3LRJ17&=0UXtUTfB3l6H6k6;KD2&cDj{lWFj(iizhctLD{<n+R(xrEne$F;
zWTMc3`1>L>x`NI|S=4kN8UhC-(u8?-#qWPk>GvY!hTsQ3>$elCyZ1wiF<g4WLy_WL
zLuv5dgHX?Lhd<wLJ-5e=HBN`ChC~3Nb*tamSY<%&m6m>oM=6^zt53mj_xLU<jsFs$
z+7!bTtfWnIX*qrRZsl2C2PpNKBcSH+B|k&mL3diwt&@f;dm{2!9RBko{UfgEzOVK;
z(u!RB<EG$+pVCLuT%|5@R9oMot)jSED)vTrioPIz0Oj^W@lj%)0jowMC7G1x&|!bS
z8jo#;Vag_P&tS-zAD@%yT2%EL(=3kPHW|{d7tW{;cS)YAeMP7hX1nE0-Y?`Xo2WM8
zo{ZDZD03qs41DNAjs+};*>qaDdBI*2;&ndxKcbao6C<K_tbZ{)FZ!d*E6D;Agfm<I
zGB3Qk*WQj_vvgTHBP=3o(PfIm$7y|iLDe!(AEBRH7_-Rhzc=T&CRvmwFn?dQF<b0M
zqWP<U7P0kRz5>nTY5RGuXUdiv!~|-l`MzV}<lMlqzS2Ole~~-c_V|0Z(#wT@^wsI4
zr@qG&5gW_55RCLcq4&nPY3=THxl!`IDoRpT;zpQWv;xxXSwG^@XoKd}L*FO%<)zDK
zqovItLa9eyc^}s^NBSOwpo5eECJF{8SYO;@WB_kXji669J*RJCWvLGTJAQ8B)wqTg
zfj)0_8h|8EoDnVQE9Lv7`ZI%mq#5+An@W{tNTARN6$1|QPnbPSYg9La+9fs9C|ese
z<Jqn)R>lS+#5;z<cYc#G0-dtWAGWtLrFQu4*f_jDF0ESpgBr@q^&$cOh*M~p<b1(t
zd3&C#1)A@3@(@!!eh+7LW7_s(zzB-u`IC}@y}RQk*MGf!3C>J-IVr*U%&TC*u|K0^
zf}G~(aVqCOX;9?~8S=Gm%SmpysOc;p#n2FFL+_-qi<KKvX}`DTGb7dw{e4R;uXAZD
zz`RwG@dtyo=+V7FQdcM+GWiBILj12{8vS42NyalL3#a>6T)j0v^vhY97Ye|3{<X+X
zA^wT)?;DEPpj^N&TI5R6kd8pieoY)&ny`o373Sh>p8cC3X>-o}cd&TX9WRcHtwxh2
ze~ZRP9M;=crn5a~#rp#|M{`;hwOYO^_%dziUVV)v8hoIco9tODo1YCDXl76J4dkHs
zJH|uIgFy6&y2vc=8okc|G~xbf@a2A>Qf0xX(+!2Zk80(AxF-L5UPvX<_+d%5TpcgA
zqP!(O(--EWZ=7YKE7&jl`82e-2$#Mk=OIx_xb!V}a)k5)RX6E|4p?EoHZNl`{meSf
zElFi(<2+}sSK$)C+7E*{(ZZzzZS(WOzBm*IT!+MBAvn<76h#v<gae?u?iK`gXeH4w
z+w#++lyySVTbEkiMGm?8*PW8(ujv2EWNZHjS3(u~Qne`+c9FSIXXJ?32E-%c{sy8P
z>fH7~=|ALqjKjTES@TEU0!YgbU~Rgb@O-&BENqDu=QN24{s@~={eReNU6E4=aQBoJ
zzI1Eo9rrzlHmvAQ_pT8W)21an2KAa{)?y|A9Zk|x!u5+hSi>x1juF3}FZ0^t#hMmC
zTk?1Rl`0FN@ZeBs%}*q{pq``Yw4`h#`F;oqqW#>abcnmkW{@bmgRNqTdHaKWnv?Tp
zi<)ShZ8BM%r9)_kR^mH7aE)~=<p2}_W(~ZNto7dgXzZZ3@BFjj4l}$Bm-7pX8Zs;1
zVW<vNMM@6fb*Ofzkv8Ct6&aPW;4k}yzGqr(cM0_C!BnDq=CuT+8q3!Y%mVLPn31l@
z5D*ZL+{5qvoB$1cx$<~~>mmKIomZ5(N1(F&ceUC9UU;u$?qz?Ai4JImr~!Cdmrn(%
zu2>#`Z36A`a6A`pZ(gfeo5zzY_B*8_>1=Lqr=v;|31BZsE;nqdPa^0o;xzjW-b08E
z47SBlhPGEj5|*c<lUoLJBZ%KRwC2qy9n<FWPk}ESn>#eE35?#=v_aeN<$j4l4(6aT
zV&CYNpDouTGUyjCi1~8B)6iSIvL3A4GX(q={PrkFOw>mrUyHOm+EBK(61DJlxa}Y=
zvpU3;A<G}b&EnI(YrI^Jjk7L51TBb8@jBSlTD^pj6zBWX-?)_H=C|6h#PAzI{WZvc
zM8v3O7;-6BI)Y;1_KG^HJ^EIRfvGe0h6tmQR)T*>D4pDx3XhMir#gllFyHPVBc%u<
zIh0{{Q4)<XW<r7YkbD6v<Tr`O`W4~hD(!zRffL^7dSQ2s`ExQjHWD0VyIM_yK0H-=
zbo%E82StvEbIzRlX^k6~3FTU1%ikrx)V^effszye!*-1NzCUcV?sxLX`kq=r*3*K1
z--hVr-4w-*5H-<v{8JH6toHLrqbHDQ{%2EJce3OpokkDSO`2=cMEx5L_SWVOMGqgj
zQ*U-5=F~=YXAts;T&lq3;1O%5UclqG{dM6E)qg2n1#ni+!>H6;%620zIBt{BW8Mru
zG0#P>RK_T=)G^~KMfa>G_|GE;A7Ur_*1uo1drGF38p_<^QW3E@9}!zoGx1zVLFb-K
zHy13$*N*>Fr3O2qj}Vp>9|2}$wy{d$r|=3x*aCzS9Hh&~kS6BdE*riDGG1ni$r0zd
ze<-U=Y|AifYm)6-Eo)bqks@)j6Y`?jS8~tbvd~zfuO)y#p99oYEMYrOrVxuTKXkq%
zdQ(gA)Yawm!!m77y7@&vM%W&6u2*-YH5aIZ)MSx_SwZiSXxb0S_J@S_V~BL@<_2(%
zkVMalhm5@ruCOY%ErBRG&ASCDhgov5@S83%31k&=CA#qzFb$Y8gbK&gCJP?ur+n6j
zScb?LO@*dr>ImjWXMO>#pm{~#CnB(8MdF)1=^*bI3CS#>z+%>Byr~@v=zrpf04!XI
zpK=_lh{0<a@q{Vtodsz!L(DvKdJZx+V!W@|g;cR$L)M5^Rh=0?@U7$<iDdlR8w^B_
z))@5<WAaIwGXqPSGv-Z4{V3cM9}`LO9OfH2v|3}IgWr5^TfyfL<ZJkhGKn<Q5{ZG4
zijwVMEA&0qp3HMKORcTAZ-2sWLndMvc`%rT2aYLFG8n)GK3{BsgxKJ9JI#;B!Y0CO
z9|!qmJY|3-k{d`EQ-CObw`(gm7UEO-^|=?BZ#i>lECh@VP*o-AD@r)(1tWldjgD#k
zK}+*BMMMV%5!m_}!(R44G`0JHo9DMNDUT~4yOG|iC?^WL^W<<c=8!Nx%F~?e4B{K}
zr&n+)6g$d-ckU_+S~R%75W#gkO%LlC>wRcpCpbZ&tw1EYJ#W=mWT|tW?TCwVq%jHJ
zZ5KS~_XlicZ_m(Ff<)Eoc0DP3&ouUna3KCA;!{(JC%^V#j@Ed2+jk$ox}cAWLT!99
zV{S6`Sf5E!pbkhSHnkJeY92(9pB&KAeQ|F$V7FrHKUySb>|r7t@3da`;o2QjYJ2-E
zl0eP3xq|9^GmNV)WVzfw{1vw<x<UN8+a4;={(xMLiS|whm7MPX8G9BjL99`5gKU$x
zM#@>%PvvYEccJ>Gs@sPY-40sv7BDqlLtb5#lERvjv2ldvj|V@y1W4oI5q?6foBgkH
zDYz9~77WRA3H2zcWY}o{$s6S{99wPto*QrJpn(T6L<$P;6w-mrj^)^~3HEc<$@Ar#
zu1BGaCe<E?A;(<Ir41`B7_b<oOAFJQlVN9{7URam>cg=ii9d_l>Il^PFOmd`>1Qb~
z3pRU>UF}qc89<MvaT{YR_+a8OG#G;wZj!;{F;X<~G?Jngx)%R(o8Pq{&G$ks2@6;o
z_5%^nIT7igk8d`?7dk<&){Hm{^Y6MEmpLb6NAb?RV#L(-e3a}1$K-nHSQ56BE?SS-
zzKRfliD8b@0NCiaPLNSckiejZIHIcS>%|)49<R_GW>*uDH*dNz!cq)(u{KS#VmiE!
z#z$qe#Kt&K+kHG1Um>#0Piq}wKO=`<t%W;Yi=-RhpG{6`<y}m~$1DI|ve8&mlVPxt
z<i^|K)%Eu6gpIP3g$X##eakK9BP71_oI%LonY2<1uF7D~HA|i*C{WE74%)kA$@Eg!
znPYM-mJl@_i?{w9eb8vzh22s(Ln_;cIm9g8_BT?FKvXFv_*g_Mq-vpyfK$8>$y!#L
z5HC`{Dt5pvJN{Q(&OJYw_BM@;#1qD%k!V==&6QG%_nQBu7E_mv2C|uqtcIY9)u)?u
z5sMY0Mfw`Z@GsxhOV(Q6|9WU)zdO;8+(8lH8ZB*5YiQ-W8*FdZMDP=0dd^~WiJm*=
zt~Pt=;N_Vj$9W{M!4vHQ`Or;(Pp*_=$=ODDBkM23%0^kMV@2tJ)=W7|KG%?Vh5QM~
zBj=eVy1zJvbxCgG*l|2zr?j|uzMwrE+R?2008;tk7UB*)p8qn#u<j3QIHQbisSbu>
z=9A~A=hKKPEt8%I<crIv&?w|Bx@rMVvw05G9j&YqoDfGX2N3UVk*4Re0)z!uK7{Q~
z%e*iDlLGaxgh05VyjW)Kcw=ha_FGJwCasT<KAGY=HIc#g|7{h8^phzC-ap6Ma!1hH
zQ%Y&DVJ<|?tXgs931W}lWc75o*s!b7=x^NVm(!D~{OQC3<%dNk9-VxnJV~JtTESTO
zW)RiOr?*#OQCIlepLkTYLlW*!QnDx6k6QpY*W{yGaQ4Kn71sTcL4D(nF^h&f+YT{l
z-7@pp$mycb-`-UkUD+^UJRdqXH0!Fg)&M_~hN#9Uv0QeI+P7CjDaYO%`l8wO+81e@
zBsMvTeoX&kJa{IN%;DYWI$O)%Vhj@%uJcS1eUI;UN_&pK=Kq$xDt~~U3*~q)gagvR
zW0jxubv<$6<7IOYFm~Em-N37TJlCi@Lq1ZC$Y41xK-Mg@!D2-d-s*W+^1uh5P1xDZ
zx=_}nJIER;$ZyW544R0)mYXA+dc>Z3FM0FhuPIo5<s3|8CT<7%HQIFS*5I<=jE`v-
zt*y%c(_sab#6Bua`t5F+J29Xc!iz}rJBD4eW>z8pGu~ccK>W+co^5~{Mp86<2`P0(
zkvNV<88MPGlZyLr@fBNZEM3R{2<F#O$f4xKHvI=2HyQXOm?Pz9B9|^DLR@F~7txrk
ztMG2{+*9c3D!3Mmx+#>{76PN$I^<Y09(%7jq2*Yy_^^a+C}@a18J3tsGkV5DnXsyi
zEcNyR^XQtj`Ja{I`p=t91rkwir<v60+##uzb}>S*wI1vKvJB^%mra3@5vK!Vs5)Qz
zKxPI(x%&x-3=2WCh;6hn0N5$<-9|5mWU`)a+#XO$g{YP!XE_8PCpi{~AdTh5f5M_O
z<*Dk791XVh6f68NJVB?GLtFtlxQp?<{#8(I>vpkiIj+uXRt@t;heRS#P;ovb%XpWa
zdJ30Pu{=JH(2$iHQWjNI824mVzT<d%o;=Ow$5ARFddw{z1HYikruCQc99Uf^(w8cP
znh4FD*JVlT(2hF04f&DoQ*w%$mIOYl?3^rcGe>e)$?T-J1fKX*KY)`TS+`t;+iqol
z;mIsxo+5kX)byt1*akU;m|zZjd(20OcPS>rCAaW{cUnf6CRwPR2=+ewrIo2y7W-VF
zFS7;ROoKQKJNuVuOz-*04Uk0oNVFL3Gs&2L8Gh=7oE5ir@4{k%K!#b1l0~(2u0FP#
zRfF1t6rL-;Vpogd0$)qGX)Z3?W*d+_err<Xe<3W>bl^3cWsSyyFJ2TWMw9<}NIV~*
ziQ$ZD{ys+0Bst9nhuif+JFy#Wo^0?g<R5{X!!vmxcbu`8f1~fA57SGp634IDCaAp;
zza;?If@S_G7n7&jxf$PbT%M!)+WYxSU-f=|?El<Ec)H)a<ftaW={W{aH{+IGk6!PI
zZrOQ%q!d?k_J9}CdGlP8$vd+hds~W^51*e-gH5G9J|H5Bm7Lv{T%3_J4PLlb)}87m
z>5Y3&j?IsouNsM;Z`kKNO1HtFxceUu1Kd~|^xTRhcc#Vb;5k~SsfFAZZxddhLo0G}
z3XQ~ZBoc-(SrF)~O4ph4UCZZ*ti_wXrY-J2jmOlhiis*74U^XGDF#o|;_$zr)y@`6
zw8?qBzn>RP5rBm6&UEAVOJ<k({k5PV(j+1pV|zP)`lHP(b>scQxOX$9`!*+qW8l9X
zl5jG0%>Q@h5l6c*)llUKpMT03e3@C%=U3Jh3r`HSr_QCkk)ozTX`djmh^}FCtR8ZG
zR+$m09T<-8{%9MBmt0E=zEEQ%PkR|m(%sfdJkBxhXuIFF-ULoL!_6H$65z-;kZK!_
z1a9-*N96h_x>)h%1)TQ=r(d)K=^~jl86(6e6q79lPbq~DD9ExIoj{%4*L6b|9@16X
zgV|}qEQj`+s8Tw1M}1<0ERG>reUIxh_04_#vknTAnY0br@_k|l!!<*`sA<3Lsk}U!
z@UkCvjE;sa=?oLjDp?*_9iegDH>;F$iE8D6Z{!H^dd(}?mLaV<>&K5c>C$U<rDF(+
zBwcSGhT&W4<auN1z|zabZ5Q3#{B2P4xZm_@d~R9Ch&V3ZFk%$`cpJb|x#HEPw)vvE
zGgX3h+a`dSulU7{PgMB0whKp|ZaYOSA)@=XU&+F~d<!&h{^TFpJZ8{4F7u0)dEhqN
zo|=NMK0%0h%=<cW^!b!8Oz$#x;lD-nKT|YZfbqiS<%O17baB8Nt9%Vbm%X^F<TNts
zizBYZSMI9Kn*{T95B4?m%I#GyPd}~tn^@Z$K9Ab|=a>%^qoD!jRaumQ)@)(XA$Dkz
zT(^4T&z~#)WDnRDwi%qp4jcf1$$LhSF%#wxlY8}q2Bw6@OQdw8t9G^Y4RJ+9nT-Y6
z%O1u2;bw>3cGDkqMzqvv&h@`@i&zkP*uSDfu@mtLPZfv!NzXCNJ2|zh{CQRV-Lr3M
z;Wdz@&=+h5z76>Bb5e|ZO-N`7ZT+cNYk$-(xkE(rpgMI)(I^#%)(-tkDB*fTcPXhp
zWvU_bZ*%x?;7BvzwN;Zn+nzX3P)`);;cv~*ed&MKE?<KJ5O)c_tKazVES4`8si5tz
z+7p^lO;7(^@D+St&<@1AwW_~+536oHW3f=Io{~7{%e%j{O{KgMIwcP0jnB2&BU@Vj
zp?$AsV#utAEP2q<t_=jfmeOEJ^BDj@8mN=3AX7vR<P;3va5gGRbWT2`KxH5p+cye~
zRF=A>5$&W8?2<cu;5QUsDT0bA%-=1sIb}*UN|5Zo=)*+hQO{{8Gr2fMt3>sBO-h#U
z`lPO986`baZ#KaN@%x(30G!*1gOlu~1wG&M#uUBg056Bpl^@=B7zSsPAADrwOX|h?
z%==Mq`?o=+f0Y>lA=aeN0Z!%Q&y!LnBGVHP0*irT+o@UU`+OPh<7o+Ou`%iv#h|kW
z>rLj=$8^D8fw7SK=;7J-+X&I~J{6!$SfdwNi{UOzQ2vT2(#g=~yxi>SjQ?|_q{{q5
z`(FbAoJULXX$^7Sl_^O-myl|I3$tfb>NGU(2xK?Z=KSN7f(@{Lx>MY=MO;G7F>K!z
z-Dcl#iD@YGCE^3Y!`V6p4%P_b5{`a{<?iyr!J2-7uSaIUey-h3p^~S5z3M+%s~4pq
zIY<6Qd;}M^w(_a?_bFVylVt%PfVXye{5K|Z0p@&m)Z(9p1M%+WXmR{YgmIqK?Vj(J
zM@J18CP9C6S;N!?LIboiXQk8h%yi^JOSc^DLFzs_t*b%hjtIu`hNp_6<pARSbhFJx
zC75=&1U$fPwvXy)TLIptk;wKMxzCBGj}g1B8IG#62jnTUY9M5DhKHSC7{CSIiVuV`
z7}*eQvhMuL9yqT#{g28k!A~b_<f1q0GB>7g`dB{IWl$iz2o|JuJUm%mzTSZ+2dV>L
zQEWu8EL9JzW@1$jlXLDdjKH{?Wk)<WPD;mL)282uUGMLdZbn*lZ@hyCYld!x;PGS6
zVL;Zal-I?kDwC3Cu}!P}#4u;*NJ2y|QchLSk}M}fGkZd>I-;(;o8Q~5Pi!XcEv5Ii
zLotVmMB92J%aVoMyogWLL7dnwcfP$mi?8L{hk_~P7fPHD^@5_@hHP|J4WZt}On=i9
zQ%CstnpM(7#dutJlU61^2=elpvTXWQxmp2+<3lTr_E$E4ROWJGg9^c%@OMmNBHc<{
zU%|z;JF8#oEw%<P$|>{2c-Eoc7O|@&k5f)`zriTO=`BbF2evcx?_>0iIKAimc}4GL
zC#NJXniI{}<U}FT!QNgwquM$>M4=wMb|iedDvu{l?4W@C2L*l}13n4ks6cvT6Er}i
ztbj9P%Qkq}1%O40*57v0tMK*wI{r_$dkv)k{N=2RD><Av^gv(ZM!PY1C<Y#8Wol_F
zfFUAtw1H`LOE&ih2)RJ*nI?Hw3TkTK&-W?cSnI3NA8sn*?l}sd=wTfa?E%!34h(<)
zYUtu*8K0#2S#f`OnJMUg==iT43Wc5&YO)WL8+M?5zOc7AI4#z)Q^|&IS=1_SRQDGz
zKs8D!fgpfy|JJNUv8leI3B&Sp+}>8_LDw5s?hpI>D|x;Z$>2{`J`HFhlE-+GF!<ga
zXs+*pceI)d+AeoFk92AQs>3m0x2SB_RP1;euJ+`AzN7xNlP#=50gBf~U_7@K{1mI2
z%l-*CsG7lZe?0Vc%nxxy7ubgq_LWae`ohJ+c_|(vIpgQvEO`-UXFKqDq-cE2M|)>m
zzL*-%68p9`aaS7UeRc*GKG#jPo!trmsL6+%57dvN1a;4YFHWv`UX~7PJMQ~!&v`9c
zGe<(GfF#s6E!3dzZ|fvA`9y;)nQSTiLg$wuS5BF4t$Uh#)CWugKj!wJ13Z+|pW#7e
z^IqR_I?SjjS)ji8b#ONdmf}3o-ZU98-y1XE%b4fkvuLpn{y-rujITk4K-N1!D&I^F
z2X}VEH>}(xtMXATDh5|O(Fs|QK{`!DR^1N#C|^e$_Z0Zdfx`AJE6KSJY7yNoHOZ#p
z+*+C0Y`^QJwY_hiW#DU7{6@s;)ZvIdz*H~r&gZx~PI>-;B+u*Jh?*!*bw2wDO8D&<
zis|h{r$<V&<)b0dlm@sjj>JM8TD5)7v*=M{=h%6<w(iGEE=*>>m8Skx;#@my@P{oN
zedyIcBCmG;^A9>O{?BgRP`{gm>^`q)nAc!Rc55tlI8Sw_CQ9AfAR%hYen|OCA`+(a
z<Nm7L4oA68QB{RKw>gl=SidR`qiAW#<cd1C<8n;j0;VtexTBPv#qkqd+kKQNt_!b)
z^@7Ro&DnF#&atldqj}vo9C@~6-X|m=POO|_HbrSycrLE`kqMMR3(2mJj%KeV<!|S(
z9@-|^dk(t|M0n=K*qXKFy@$di#ll|izQ*%zC)p+4Z?8PRE?Y|`jAW<I#`I3z-o{!Z
z7M%r((@i-yZqSh~kfL|Acmb>GpGdr~@r(DyH_FVyW_>VS+bq!R?fz%R7suiHW@xz{
z8CIvW51-`xL?0(B#qw+|BTTtYii*ml&&Ra*-%ZV$i@9Kl?{#I1i$fnj1<|;AOjut6
zg}a$irou3fbij@gEUR6u`zQr;Mlk=ATDe`mlpPbqRV{iLog(BP2|V{2AFmSLeT!x%
z;<J<be%5Bnk8=E1;ua$<RO&Nt=?lx*4)(;tw%qzyKr`|7tqOW2%W~d~Uz>Z{TE*7G
zKtgA&5zq=-5h;Mal<JS1b<fI{rIsLd*<*!0_410^+b%l(JGFqB{8|TN=tCRj>$0cS
zR=;iPz&p#?@PR*1kXBO`<B0^zI**Pro8pY6UZY(Z)`O3{v&4h}q{ASmcLX@_`l6!B
zj(Eb9hjA76dtr~f(m=f)4T0s}S~549VNA{f1Hxz$MCFSSsfW41f5x!W+f~7*ux@hq
z03_71w;$3;VMG8TSkd44!M2SBXW>ZKU*9etEeQ<aalAmX?Mbs4IjgCDy?7tXGSBJI
z9|aF~>~p5yf^s1rwyc+V@2!>n4x#3^on?vQ1R;{;!(WHM2(v8>E>0~5u=l*uyjU@q
z0%VeEVryW-eGNHWQ07XoAFmroP^W$4C<;n=9t*ihNb9PdN<O6?znqWe_OD7loRnQX
zFtCedt^__N!ZEQ9HkoVA!-|)wMUfyqrbBklp-S@D-baf>Y7cj^j(A(`FcgoePUj9>
zdPOl%N)P2ev>9A5M;(aUkGf6<o6aLFJlzt{%toU$5{Uw_4JO2Jn$H35OJ_1A3-6-<
zORcGmDkF;BrRlGKxR_43cd|a!2R3UB+F9)?s$8~ah_`pO04wE4Ccmfw6;}O`lLTN6
zcn%?g{hheUP+C+?(1J8&m0{CC(8(z!UV+-r538MWMPO}!dvH}dXw^zwM9=tmS+#!H
z#!Z;-dw9G8<NhFj!)M`cY(LLQ8y1&!I63827DQCT;T2(^(0H$||2+6QJi|&fFsq(I
zws#p9&B9aG8fRvPS?{8wAU%jn?XiH!M5ENX>8?n!gQC(&)6yz+2s6er7&7QHkT=k@
zAv#aC?b@<x-D+}YSQqn6)((6o{jC1f9QIUC<1cH(AD5`bd7GZy3n5RIrCJsE^>|MG
z^jL1#?NMpCTavL<2H;9ASBLzgA1s8Zeo+NTZ~!b4z#CdL$as|leV<&2rR+s9rnq%f
zpiRyOAiutW1s4O*G_ZdG@}d0w_6%fHH1??TDDZjgPM!K2(Q9K{mezFk-yxGhT1sp`
z5+&=+J{6R)wQnqC2r(nus{Sv0+nn~AgGglM%nf9-B#N2}e><&ezl8`obK*`0VG>(y
zY~Yt=VPZb%07yc<uCB8v7Ut!Da$`o%`yE=47M4q!P6%MJ>|V|lcEi7*>)|V$PmwLm
zbN=#(XnC+0j572(e}1m50%Ul<`{n9eN1T9T826Ck_}2{I*~|#(O)p0^#{8It5^pFF
zb&?q-ai=CnK|F1gD3hD_k3kJHU7j{SUn^007#_v0QO<9CFrq5_?qTZMRT<7-YHpJ)
zT8+_pVKgI(^@+=(wal$=_oA4hN}w+#fJt(Z3=3aOpeXlQEDx_v`w1?9h+wX`m+ivb
z+)ZTY>t}c9n-fNE`Jm+73>4iCCV59~xd%SAO!1LyGDEF>;QIW|_rDAl-l`WA1Vg57
z^xneX7r@l_5@Qrcs`J8^7UDb$-WeK4m+XtJ3&KZ7S6+^$NZ>!&X`PU!KTz#P_GAY}
zUxpQ}?a#t%EUg?CV!M87FVE7`zVtM28Ws)S<jPCEvBkrE^BEaA=sg0$|NlpWAAg<J
xk>A`7AtOq~2ECaAz`qv_?F~C3qSUCUM{7shv@F07{vSkdl;zdrs${-|{2xkbS>yl!

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_optical_flow_001.png b/0.16/_images/sphx_glr_plot_optical_flow_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..e09fe3a2a02e22e7470207eb37aeeb75b55d9f43
GIT binary patch
literal 150277
zcmY&<bx>Tvvo0Yxgy30VA-KC`afe-G35#oRcZVP$SQdxHHGu^}a1HLT!QBG{4;~=6
zz1;hIa^D~4oT;Aao>N_2Q(u464Sx$&et}1YhlYms0s>Zmp`oGs{2RC8VEudlZl7-X
z_Yw6{)c4YIwej*b_pnA&H}`ULa`keux1jT}_VBcKbrIxw!z0K=XXoYR<|)R@>->Kg
z@VI)|@{Y;K|NdtL*9~moiH7#l{J)-OU!_az(Vn59K@?=Q{c;W;cGImO0Xy)AkxI9t
zFN^{V);ekX`y?xI6n0+E?Y@Qr8uSU%1&0lI#uIJ*32!3quH3H8O9$)7x(KN3%+Q1s
z9KF1~(<@af7zDkVqGC8lVqYI}bOuW**yuhFL%8Z7Th|=tXN2D$*36r}R|i-UjBjO1
zAZ*Xn@&*O{S;n`d-zSBvQTqKG{SUE*avpK(|6B8KCWPrMc0uxg=w;qBCfW8K{|}|{
zt=Dh~MxX!B&=YuzvgZB&G=8JZF#c)fzqkHBy3!>6N{9dZcm5-LO#1XbOUBTVR>DuF
zWLg&I)P8sGKQm)o1AyBHDQaqJuz#cIfne-^8^4!NtNXQF`e`K$`(=~g>4w(e&;RTY
znB{SGY{FyQ<W_8(XZ|0dG`8rAj@;-4Z!sFQ;FEWcdNU5fz&-r|HVl{wke&Dg5nWL`
zOZbmRWOXMoIz6-85@k(_0In)+TF?2Gd*Z6b@v}u&_VcxVX*$(bfbyw7KQS-qeAxyW
zPGZqrkwmI=1ezw@g1<2x-h6bv!42L^tU~&bhl;1J`QV#_a?^m&9{?aq5r=O;;1(3G
zsTm&4NPwj*hGWiW{$u@^;#HNgIT@u8$(}!s{qPpk$>d!6;85zeBFoAA!rgY8#SazB
zDe31}Vf>dbEhh{S5kgr<Q7=WQfbKU#<UrJ59_9qw1Ue(rQj=d^!mg?ni5y#h!s8|H
z*XHCm!KVjwtl6GAxfCao$MymJm+r8pX=TIfN5G;OPlds@(1668=l0#=oh1Eh6y)}&
zyZsU$mgS7X8f02pOl;$O$laAdn^+8T{66OCZ0&8?Y*KvKCJ1cKub?7uSkq}8$r_*d
zwo8&-tDL@<kDZyJTLkaOgCb9T9r{tS-)k6<y2<ljej3b6N2j;a&7?Ff`T&Jt*`MNR
z#=Z(?w={XynK822_r5+AVVcpzn(T<4gk)v^{;hg#Q#|C9duddx^HltaVEYz8pBQhs
z$@e*~@d!elG6``0edHo|NaJ?eyh}O2A*e(GzN<=Uia2GU*<{%_1}yV%4C&1vQ|>;>
zF!Ir1ut0c8E%^q?5rI|vT9;)?8BF!h3dzGv6I#E!Oy|BcyZMRLbCKM(WGkU;!G~j>
zh!vZJQTU<rZtT10!s-@v=fnH@v@{uTxSLJdmj!8I(a4eg@BO>mvDd}>5FF}OVt|mw
zu!WjcoP=twz`|HhmBEEs>{#%w`I(QVK}YGFlan&XJMLec|ALq?_(CsaQ~TP7!y?^g
z<SGb10GEFFQDDM6YsPwudk+n_CL66)PC3~k3y}8oZ3=uYs(1+fbO0jJ^XA6x-OOiS
zeU2b*9dg?WlbJoYN4XPU=V`PqB3GP9zM?-<6>}qt42PKovOZ_j!?7xih?g5sRvzK4
zB67s7%x{aa;*<T=!qYgMJe!|FvIoYuQBo7`sn;mc7RF=&f1<hhRn{;7z*l>3WQ0TS
zcTtz$@4mtm$RYO3hOg;EjNZAs37l+cE(KPfrM}-`N!?oRGFHOuUc&5H!pxhX>H#G^
zG59ot0AdFjdR#T_0rlcN0vq#<5{GLHY>zmyg(Q^POU7p$yd5sY;x~Wc#OMg@y{`RW
zq-dOkX~EhrM3K^)?p_Y4#9NE+?KrDLt<TN8Gf$Sh0aE5!wYP-(8|ZU&%c@<*aO4bU
zn%_D>k6H4V={*^6ji?ea`E>a^PA1E5yu`Q}3|M?{GX?4DWvS(nWZofR30IA$;pQE)
zOcdyGH2|<&!1R^QS1nW$We091f=u&Ah;?Qil~CFYb06Ur2Gj=DH%1xM`&er&-~adO
zYH>B7ub(sBz|=^2b3q#u(u$$C(+35F@}tjt+X+#JkBY?=Ym}`#Ti@C|RcOR}02ZI=
zr+t%%>A5&8ojPa1x9$2A-b%ZcugKp~9&8h#w!xE%{ONOks7_i3=}8SqyE`9gDQ$D8
zrM|Ibxyg&qEkWID5w~*5TPG5KEN{j~cs)uOlYdTy)N%FuQ;_%=uYS&&DS`##S|^T7
zv3+<;n=n$TCn=pOo#b^0vmLiLF=a%B7809<E8xKSsn}^!bTh#8z-(JfuDdq_bg|ub
z@;S)4y!_quT=9`nUOHsD-DI*$k{3f!3bUz&$0fYRO9Ks{`5WkAk(TLQLzXfnVa{jB
zSdU%-`nkpmOlb1DA&K8r_wq^#%wlq>AI%=IIi>)fcbk0np2Fg!da%buL8!{7f0RIW
zJ{yXsP3KN;?{Q(2XB<PWhRsWE<!)UfGA6(|y7hAqWAt1p$EhZaB?<qP7&T9mF3_M8
zcNP&H_i5%^Mh}5zj!5wrrJT}w!9EnUI#TcYHrPj0+;hrhbo}!Oq%CEJr-nN<NZO-)
z`S^9o(2EL#eMUd4Wtb<3VAU(JkZ~|e>)P9LjX=7GnuU*lwhYEp1!ARc&ywLNW@&3`
zS#c^*5wVm45o=lHopMPr(s+8`z$Te&DOKr;Wf_HGK_ahzS_WC>-fWY5sr*j%LRLF3
znk8$>SSY~Q*}{c{vW0=D<Zc0TaYDUsL&iK$R;Q{Uq0Q!c>2%!Xa-nD;?gy^jp-9z%
zFR}8ASM=c(@xV?GZsW^9+z_p{vVGfr+i$BT4I;W~WOGbZ2H9y9FWo>DIq-#|;jHO!
z?ru1#3>*7kc-l!hvNAec$3*%#bMwbE8Dw}Um8Vz#qp(Rw_#qC${ufE4z+zX#j}~90
zJ%x*1uh~VR7g$APDrRsaV4=L~03TJM#RR_@^(O(WJwK2fUuWCM&cj_!XJEvSuLNx{
z=O})b={Ab0qOfo5r2gP~2g;+mF8#SJCIv0d{^LtVr9$FYJSIICf!V>dRBGnU6lm6P
z+eYK0wMoM-I*)hmgNr4KKcI*m`*UNL*8@31>5`%=by5hGn?;ZOZ!PJKK?<Q6qd%0J
zAGEid!5$4YAsQTVIAx`oq3W}JoG2{Ko`z|nhPvdI&5gWoCuiQG{-dkl0B?tnN&QK7
z$}uO6pRMaUZD-80SgiaE#*9wuuDKXky(7$BJq`sMPpG*ei`ft?x!ee0yIiwutU_TG
zlvj&M<H?$}1d#4P-D|0_mQ20z8(sBq^|n;j=}dWEF*B<oi`I{YTOzCJUt0AK^`<Ph
zq0yj)WouX0D9>HqiDRG1%p|-BUyx=qB*!N(8CqcPlZ!i5r0OrA<DM&5yz1=zIdeR}
zSp2mzp;poj(2e2M;@>VcsU!}D=p#RhFV9)XWlK^vdx|%mAB`3U2Px5x;X*!_yv+K|
z+$%L9ihLZSSmS4J@G)$+8~l!#1oZD;sBKZw3nUUpL^3Fx8dmC7#3l>|juGgwjVsvB
zN3un8@spw#k+}mI!)3~FzF5jwkK@vJxyypHa(UbV#PI_{t@wT5dNu^Upyq4DbH^5m
zZ%4IU4i?3kk#DG2c92-%bHKwRJ|F=oWRD?!w>b5K+CH}?sGvbMLqU+^$kkbZ#`G*F
zPA_}Vf19g7`w+Q(``|z5ej|2tVKB^~kzB%RP$|iNeD*<q1}{Z8v96wcd(o4;CXnhx
zu)m0G`Wt_FxMKf%ZOfs}`_br4<bhi(t_nJg^pr_NN;FSCX#&?=&!$$VPfm>0+<<?i
z{Eqa7$HK`+@QDHM#g7aHM6|3WNnCE%(pHg>iftd%F)rE?$V9!%>v1rrZW?eQtX=$t
z6TDP|F_$-rG;D;ea!}w6AMy3}3mue9jEvs$#LA+GTg8q3zI<u0w(xm%A9*`)NK@=B
zBpJD-Z2-0(jQ1j%%A5+!)wS6mLwJj2-u`XZzz(%h#g*j{d~1w1Du)L*VV%@^Fh5(I
zPW}3gM$ME|%j$$lG<(41W6E}28V8J7)e0vxmnH?xjiQEQK!?RjXFXrUulNQ!{3f%c
z=Wk3smsQ<&#4{%npv{ZR{`FS7O^eup-#8V_ER#*_MSO;q^-2ZKw9}ZFdrKPg{q8|&
z^So}=b?Uz6#NkY_mi0A_ujIC<%{YY7+06j0Xc=0v)vB-Ozy`LB8bf2nn;G4c#D(h_
z=jl9~H&SAhY*u(md#&J&0XDwR^<5bo9$I40p=;B7V8dCbgV)h&&;Kk6IY_&*?_M>D
zRR}amIS+<|A`jJh9;L}4%N8xDFW_T=d-jW6f^i$OC9Y&)^AbHnXvAVMlg|9N%GXhj
z_Saj(WbxxYOyG%OV3g}L%-dT9Uqtge!$Sx|<e{&_2^&zHr34Wkt6-xvibMalhlxqa
zBDP>L)AC@FqOg$Gv`kUW1(CDc|2eJ}-)z8PF*Xs9<HgT1=+fb&JDZ>pquxJ$<{c}G
zGm&I*H_5U&`RV)>@5VfJqLWVsOOAbB@;&~Y30e;m^?IxVwLDOn^e900@!OE+VS8zy
zB#o%n3Pne{yxv-!rJ7-R{W~ib#R0nF!x39MXN@1WnSJV=A>yl?&Yz^YEwC}G+3mc)
z*)(QaahF!N{szE&seXxbejdfs@@@y%W?gwK7z?I*h$BKRv1G{2(h1QnpI|w4Oz>$l
zoohg=8j&*bfZwkcTw9;yINZ{G3H5PByjQ4G_cp32Z(^MN3Xzv@URQV2bn9cXF%pAY
z1-%^x-i~Q3(Vv~^*D{k<*G2XWupu!Dg}VL1HM1__ULbcDBbH$+M$1^CL<7fQAv|)_
zXVTUlFI}Cl@9)Z27Tj-bp8Pz1z88DZHA2OeaIlbHoFhb+FT!g}Kg2pZQugU*xdf$*
zt>6pC=o_TP-3SEHPU2oR^$|V@|E*0G!MA<jCI-_ZfA^^TQMJexc1VKz?NiQ-&UW3k
zf%j3<`|Xnwc;|g<69F4cG6(!An;5`6(c>~HU|3kRQk1i3SuB!a_@zPaGe3GgqyF;3
zukMHUZP!!7c8JDNFS;}zvvd|gc_qhKGI*R}a#FYoh0!6`$2Y1_@gyzHzRO<q0y?vY
z9!%p-m%VJrLEjuqmf1zRq3g$1k!_14k<wBeSsLYX=Qm$iQ{I5Vg>FKtk42|c<CX@C
z;`FiyZJANYkL6eX!g7_fW6If!q~iwGJlSb#ndwM(z}uxtFlEKKkVsD40(oedJq!lv
zO5U}y$Gv2l^45KiFu1afC~Yz2k+-2{fEBv3Kunfu>e5Zxq@MrHAVlHLpS7`sPXGoA
z(U4)gKhQN(DB5CI(AMvy@^&{{Xs1s3hKQ8pV<*9S6Kk&pB=#5Mxq-+{me!H&K0<j%
zeeqmrOEq+WK6fs(@tk#-BF&Ct;DIq#U}SWBF^;6DWeoexKhR`UwPcq9v$H&s1!gu`
zu+J2tr9^A!MA_$!%PFuNJhAgd!E#FP<!8no+RpTZtMlvs8nS%S;0!^n_w_w6M8ex7
z^X5!lhK#!S{dC%b8F+@QEOcV^%wELR61RWSoV9XpX@9+T$q6x3Vd3f=`edMExagaa
z(?&l0E(aVPxHyLs+I4I^r?h?9wfEWnkVPP6Y`@pc&PV+4$i;K!VAHpS4HlQb5=_s|
z^=|1O3&B|5p6y1@-k0LAUq0-{y11k$+ch>P&zFV&ikDvEdo@>1B&QAClSO!j-+NL+
z4tjT*m+YRg=|z8vx{tKCIktE7xgR$F{;V@|uNh5UUOU`AiEbf7IqsZ^q)Y41EsWL@
zkEI7{y*C%fDrPz<_x573*hLx*W~zUMWj5&&^!^oOwcp7LUr607AA0wQomTOxmkj>X
ztF@9U1y3b>M8h~>8?>QB%ke%>PIQU@VEKt5Vh#Kwp3IJ!AGQ>P_2HoDx+8(;BO4xW
zR2W9R5Kb7Uu|osZmCKsV<_0D@c>&E!)P<X3_XW5X&cN8pOgHpK-wuYXtFvvr^LFzC
z^8stx0tn}?xan9GD#80sH;CKGB`T#?>CiaLIVYbi3B+iz277|)zSIl<EUQvHP+oD$
z!LNE%!Me(Pp)yIH*QEC6Ubc5AbE9bNTJ&C6tm?V`$O)urfZv|h*OltMM}Y|BCkY_4
z<xS>zrgw5AZ-tU}QXPCG<)DF~0)DTClv{^yIqXZ59{wX7jOTm18w|DBy8vD)zkV!d
zG6zu@brELb>FXPB9Iq_5-S0d4Fl4o{I&&@HC}^Ur*Me##Y{KguHp10jQCsK^OnuHt
zmt;jlEiRc8tb@!P&!yr5;<g)Jh(lQ2Z^CD{a6CLy+qAZvv`UWBj5#x-y%Zu#`wjE=
z9qVqJi82ixEoybR-e`Q!r)AwvGv4kfTJT}T)ivLcK2^cDQ8vg;(KDKIyb&bJV{g2$
z$XZ&k$?T|Av(&qmH~@GJnAF^?)zAr98|Uz>L3MBKarF6ra(VbH<<^jrrz}xh1<l<G
z$bCqP7%a|EE9216f?<T5DZvN~F)RCbG1crE(I(P_>=_fMdM_lA2Nv}zV)fnb7_e&`
zq26`SWR}5eP`o8g1-cz7(pxg2uvdORmrz9+fQc997Kl&L^LG2SQ|)?D44||TYN_4#
zb&Zt@v%IpB(J|PjDMg*Acr|83Eb4-DTmM=ZtH?5C3XI};jbmRuPzIqXAwAF{cx#*V
z?aGwvsYLM>il~}49{>EnZxfV&c$}We(Rsl^`6tT@my1%s=AvIPdh-V|pNwD82B6u8
z$6kJikLP^Tq;A5xz5uzhQGc6ICeJ?IrnmWP<nISu@s{YO;XRGO@2~JW%OLa!4W<M#
zwmb;yQiL<UoE530vVOK`k&Ja0J0cWjB=JV<N!8*FmQb=`9VX19sZ-U_B~L%>#~=2J
z79`C#h4Uy`5IV_I#N2GMv!J3e7_J8`DxYkBp(O-lPYx{r^qgiKRVddlJ5!{m5mjR2
zma49JkPTW8hWkr8AX+^j!X;)CFh*q0t6zAiTgx2T55=Ck4?$@XX}Xnvoo=pb#wk!7
zEO4EVOqn7gtOhL3hKRkC<t=_u9@hY{W!B6HvlKzx#g`1y`$cQf1C?n?xJk#&Iw&f8
zc7ChR>Dt}4nsp`Lo3}l0t>jmZVDlT<zj*VU94*l^>V5B~#%QnybgrC&rzGIz<Hd0k
z7m^My4(O%hqiKxI>RF=Y>nD9Eb;TgTXk6?bz*ovRJ3B#_KD3kj_1t#wKK1DON(=%P
z>Cm373k*yT%WPYv{rYIY-}m8#AGRCFfotEDPz>)rW;us!F}+wElWE{|Q;Wq@FX%_W
z<!w?&zV(-ahYheFHdiO7Kb=em*l@&4)LPQ-+1quGee)OJTNG$6Ye{dgFCIi~f@f`T
zGp~2meB+D9+UjU>KEnUh{b73urg){UbHy66`yjO|iMY|=(cLi9&4*-onf-$*TriQc
zK%c`b^?RFjTTN^2!2EvSoQ+z@TzPnXMj=T7%ix2*jS+h?oSXFdS8%<{$5hYVg*M-W
zSK_&Svj}JGcsIT!x05-V;LE!2&AxXcV!Gv3m4BLf1M`OUsy5~w=Yrs$f64RElY9WT
zky;_%CBdo~Y7m%5Z}WCf&AN`qD2{xxn<Ersy^`+`h@rLwWLS80`J0A(QUvWyv;|s`
z_r`1C7t)M;%yN2xAsQcaNHx-ytrYc70f-rk_Z++-Y0To!6;i77V)aPf(}b%K=34Kf
zORcL*aS;7RhZ+fA6hMWoh($E7a@&g;cmBF=M~Y9mSQx|P9#n3lr<DwW{dJ`I%&VqY
zRk5E^kJHbMF2r+9Y<-9q>txXF3}q@-ySz;Z<{s5r2y0+;NN)@-U^%ZhJcXwxmCTkI
z=U|6PWa4<N?!hjm!Sf*8ouE>t3{Xu?VTBh9TW(XIX&P)}Eo${vZkOt|8`;<pfL6F9
zO);S^np`frl10oAje?MtOgS+URE<|}M&A+q2X>CZa+k(3F;spDDrh-0IDpWMEs_15
z$2O`)eR;0_B(znD_Z#E#LG87~h2IUEg+>}wjxD-x9O-w7(Xl^dKi0>CD_1P9YT?Kc
zvyhx8hp^IKIL(!lW>2rX#w}GpWm1=lC!jc&;ymPP%Wg7dDzX}3y-{CRZYdLF=JOHm
zv5Qa%3@pmVn4%~oU4HkEl1p)8UXq~yD!e?iuTFU7W1ue?H9H`tVBVJ>e4c$j`z|PZ
z^3NwT2`mC>(vQBums-JBSHbDL!P~=4qby8FM~|CIiC-w3FNk~cxE<1Y?NS~Hb;0Q;
z^G?+LPhbUf{&<HwrF`Oyo8?tH20odB71fu5RaMptr)4#y*B}p9jA;5+kmL)s3<vqT
zp3xTGZ-(`$wLVojHZ{i68CtZ<HmG-*RhbMEJ6kQBbHm{~Z<?J@UDJb^4-4vOT0SFV
zf0InDeOcw0>VC3T3yU5<{(T31tXmVPC^rb_!8|2d<Oqtk8qjN#O8QPYo_gT1Zz{Jx
z1!6;muCTeyNN@by7MrS3O=nNiuIaH;1TMz!gP<Nl(BK(pfo?C8>0-kqs1h3|QaR{8
zVDz>Jfeyq4;TqI@vvgE}{aJM;nP*v8Y3Vgha1*j|_s6X=t*g{c_Odp|Z5%4KT$+T*
z*;XxXTcU_5tJQB&g1VltNnMR(c9_ZOLwQ|MU2<)S`o_1jM?lg-g5UsuOipgfcLw@A
z#pCF4?Fi1zzD<>AM;&En>#UX()6uS0kGs5xXZ~%7vDJzUf&?q~2(C>T(KznSMT~f0
zn7nlkWKO#legGe*m}Y#+>6-fXC&F<M;V&EAIc=v3%0sZe*Gg*AA(?791@}-Qo{CPF
zpVSQ7Atue`<<5Wpbkn5Gi;$$VWl294N^gumZDn8j)w0Pss^>_whBS?%P9DQwRXJ4$
zK9ba~FOlP(VQOX<s{wo*V$_~>DnOy6=wBOTBk0ZfidS4VXQ_4h(bZ_T^;wknPPAik
zUBPvm^0&cfAfyPKJv6(iZ|MBd4`atqlqatwbASu@wI^<)3U&JJtx_4@-y!LEgCI6k
zBF|F^_7VW2@}3NCg6F14-5?;FTtMJ(>6)fn1@NH|EpIUmCj5f%d_0R!e7<qIHEEi@
zNYYb&w(HKXh*me2N>p*?S%)U@+~V7_8a8tx0V;w&G)aeJBz=Q7WhC!MVkJr6(jgrO
zQy7yy=@BhZCqbBcI#$*ZVN|>PT()-#nDE^khkv&f&f>|-g{8p0oiKGDAD`b>?-k5t
zkQM`l#NBQ(t&0{e!Fm1L%(}YcH66iuaSg&d6gw5)J(GY<ap0d4#l!dvHQ-Qf@$}K^
zx)zOKPWS}Y7Xqly?^^_-iUa+z$z2j6l{X-$V?M5OIi^CU!QflhlX@PIe0eOi{#uzU
zFLR5Rm6V~TjFe&4iE^2OQw@&=4*r;0I-t%p!+vnJC+|;XChR4+0xoJbt!pFi;cMkh
zftzk$S%O!bjEDHS9t={ConWjg31s!;r0!cw9EA<JIV6Q%;c~xzSu6dL%RXe2OCp}r
zCeHw+sjrr7w)m|Ojy?kaxqw1%{`yaE!u7E99fc;}fHyEDT2sk(_aHXx!-{eOXeqF(
zClDJ|6`JTHubAw*%tA_y`U_%{T~B4D>aY?hOSPB~S73iH7O56^ESj0$Vw`F)GqjzW
z3=_}<`z{p;*qlZ4M{Glii$2Q)Ik9&Pdo-5yNR=YpwVfBl;(z8U+RV-xFpZXsbd($k
z`ZC&%CQGSPx6X*D;MH<pEK<TnKWm3hZe>;@!mIUvEQXIsY2~y-SMr;07&NNxjYW?v
zto2CvS*Qz%7-~w#S{=n|#Ms9eGCW{QSg?fs^A#R<p#e|vx!5^7KD=teXnv+8ZTMyj
zq--$m{n0P;+QlN1@rKC0s3=K_GzvH(v#3u7dHs7c;b!t>NZ-x_Hkk42`~g=*Wp!=+
zkEt_~8ra|Wl43^`12SYveH=SuquGIf$KKaddk60wyxSbY*E?ONi9KNLS3zrSveNu2
zUFiG(wwD#8@v9dBJNtG~T7TK)LORX0e<Re!g0o`gX=G32isT_0kM$(q8&3Jas8Etd
zYMm;AZ({yR#)&19UdyA4364dpSG#^RhG=!6rrL42>YPQEKVeZe>2K(n6bsNoO|u7n
zmEP^0Ff?;blrfZML+e_Crvtt*xuoi62c`vU(Qz))rO%r4clY(<V6o>~HP8M6DVFoG
zF-`s6vZC>bWC-tvtuv9DVso3B`)A(yx(K}K?DBm&Tkh+n?*T9reM*(f2GwvP!jmkn
zK94_+QKVm-f3|RQ802}FZn%-i=wj=C93$*zGjk4lQ8h<rLZMF1tffqEYbg(FioM>k
zjD##@R*VKF8!Ng*V*xY}#_A^0icKXEzHS1MIETsin+wO}43fkL;Y2qJx?l(sh6M9^
zDZC#s(`}h6L9e*99*xxpH<v(cad9WIUGB`d7pvK{+e<`Llgea<0tZPTgldS8*%GM{
z^yWQB18zN(yoa<w^eA|Re>`RQ9fYd5=hZRm9ugaXZJvlc(wDM*>Sl}hhw-A5V1`fL
z9kw)ttlKKfxsoTkHZt_}m_${&th6*=HgFhCw1I_|{coOrN&oq^8e|&uAlz}v*{Sk<
z`pfLmgk8he*}xvYt<S#Tve|(P38WhdiG9$Xxlp&;Cp`Jft4pGGTE|zsHVUJF*7I2^
z2BcVPI34ROceo5l_kb-{$v!ULGInyYcG2Bq2?`^xY}0vg-xeB|R+kZXG%TU7EMfND
z7J}CG8G5&Kj2N7|v9Qy|WN-~Gov4{XTfWFK`H8@_>R6i)+g3d%#z}ui`!v4Bo+GPC
zJJ}JOT9Ezt7+VhBQYi>oWI3LP;Y<r$VnI8{;K3@4mVJEKjt!^PKn~xloH*X{2+sWZ
zh@?@3E_MXiMuEb44k(&XvG%VJ&XQ>A91Od=8XTX441JBr{mnuS=+A0zmz?0A>YIDH
zn(6F>=?<*@-mVD5DzB*8oOOVkoFUbVM$i8Olnt}UCIl(U&YAV;6WKrO&b7pC+FKg)
zH^67g2D&vM@r?1s;GHS{Je1Khwqn6^e_5X%t#y*Yz0ARn=Fn<qV>4LTqSnx2ngvAM
z2zm(aVyjzJU(#H9ZoO`CsiQU)tw<t*IZ}Q_f&Y$4*<knMhQ|+kuyH^T$Eu}iFrG0(
zXA&8Eh%9STYvx{Y@NsdnMz@q?2U7&pQC#obF<cBP0_<ZWFVMyT7|(Z~n;4IefxV~p
z;)v`jj&`1B0#TMguZ;-+{hw_0$X?=6XcQ=ZkA{}y^~LqcWBSvJiwMjDePTuac-QYM
z2~ms%Y*yaZnAUVKEMRu!?ll<I-A$Jj*=shU5pt{5;(0#8%}dO#ShCpjsV~uBd71+j
zS8$=4I_C{@B~dL#SPo0f=gwS`GkI&@G9iy4MbQ#8*g{XDKT6S@1_U+a{shx*ja7zI
z-~h>PZ;_2$Pr)Ri+3J_!iCW{dSE8wDU{<7BQ&|R%vT5<XX;y2>4Eebkb{b+xyH7US
zeaXeKqOETW_BjL1yST;KhYWzQ9<O8?2Nb@osl_0~R$ka7so-R`eA&%^<G*MiFkm*t
zs1geu$`xlh9;^B@iG4D8a-oTpQe*n0*jR6l&l;i7&;MYW&~5(mdpkh~l<I8_#x_H7
z8Ohu|6CrmgLasW%g=W64LrUG^bQ_HCXAiXZCVPm+>jGq=cvVz-XP*65$X#+Hq4VZ?
z0x+<2?ADOsnn<~%=RJs;t-9N1^hp`zTTD~rIA?m_-arnAmIM+-S6=zweIGuzm%7jf
zvJO&e<niJDQNR6n7I|Pf9>9zyI83V0d&umpD6YH~EQ(}wjEl;3KOAvQrSnI9d!GU-
z2FvnwvZ>Gs;U|_=ZI>&OC8~accuLdDfMqL8dryt;TN2<e%t1ZBi3;Vlf$JdJ;EKr2
z_dbhalOHmj&}!=R$Q5x=-R^CHN^}-PB=*K$z;#icO1?I~Q}ngRgr}sgD|=op{x{#d
zZg0E;$NSGEmX1Uw-NL7Q(kldrBI(=1<OwgYnnSJ#=c|Hx+TZ(X8lHCEtUb7%2EHbZ
z?Bjs?+!cupDll`yICAZL>%4J78EKG?Z--><)yZN7KX85zo&Et?&tpM*kZ1AMVfs4_
z-UPj3(b=<ko^kzmFqg=}LrdY~3q`fBOKkixP2|JEsbjeD(=p>?eF6qOH5=_hV+DDf
zy+_)fX$ybViN9(ks%O%#VMJOljdBebtLhc_kT1Bvb*c19=cdRx58U)QQ(&N}eoqtQ
z`I@_SzD%WeoYavFElk7Na$;$}i<Y=C;t%-ej|w($z&<l98auawVXPI;{b=lOu_dX>
zh@tlWRU<}<E%q<M_|j%*4v#ikc|7s)apobkYEJuBhFjquR1D}Q2TS5W^sSd=B0e8;
zc&IIbLZ!1aTSiHjBg#4TwJ{D2drSG(TpiOM8TUauYfl|PA!EsBJFOm4+$C>ruC#^}
zZl=?p<KRhoO>-6-pAPVtF+aL{)v9<K`j?NAZg)-tJ%SABB=%gt_1K_k%w7p(s-}C}
zO1g%Wtl>SP1m%>s&r#c@ATU#`wODRXkU4MBlRkFe)KWFviWLDH;t{r!ZJ$Kyx_&Gq
z5|d*USEFIIjwT_ew6JtjsIA`vhik&4^%;z#*3H1J%f*`KO@8iS3Bk7_377AI@k&C5
zG?AMh^!xqtR10Ti=rn5@#Zd)-P{~sUJoyc_o(tf8KyFp=gJav-;`kD?5j+2q13u!M
z!|4T+pna}}OI@^v?SkAALZpX}iJ-8jy!iSUij}z8Gr}OI5r!l6!mTWN-F_;^K;fX=
zeC$?@+?;c6{%>*22EWulsybFhU3EmOtHMBE7y5lam^_|YI=&y3)3=Q=d-=r*la3KK
zyO8>*-keSxVX&*Pn%v32k~L^XDf*SI^I_!Fje<qKz}X$iEIlP==xg$xEMrp4eZ)nh
zy2f49euw0Y;fUXGAek%Ii7ipCI(QvDhEtUwA>M65KPb(*ya&fvc>_Wh4;P9J!x^tC
zR1o;^ih~JNkFlN%_YtC<TwW{DEt@>{snFWmP?EbNOLW+^lKn!*G6nYSG91S&^_is5
zDG_(k+K6J8aP9it=dz;zU02YpG_c{?rEJx5_9VnE+(kFI_z!B!+b1Ar_~FG8sHF21
zq!G*lDHZ3v%gi8=sSVvofq3d#_(o^$BS5rLj!i=CZ9QkZ3Cqse&$KX!pb`k35Em~)
z^`+u-V{+Fs#wW|#gyPacGPmJz7C!nSZ&?x)KZovbkX%2Wa)8h>jqy982*#B7WHFBl
zl{LtwH)R4Tg20?ej<KlbSTbUz!qBw=rvQhY8M~KpNAF)jT+7eqP6!u<SoB*>HL~!(
zgfE@T<72u4AO`ySR!$a3pCs6Xlkm+&$eGdot?LtyX}dkHZEdB)5r*d@hr-54(#&aX
z4v<a_#w+)^zH#`fUrYisLxSf=P6Uor+~3;W%se%{em1{0=Q~uf$&WjO2Ims^71wax
z_WjoUUkpE~N*wxR%Ui3t?;1V%b*p7(*6GEPWk=Bd)QWH<>FnQ3IpZxP)MnF7Wk0&r
zzaN8ebVx5R9xVN%-W^j;uSGH5W_mm8V!2pnqNfD6xa`Z~FWbcUu5;sIVNw#th>MUt
zyf@FpGLs~im6Mq_wg3X57!~Ce?(N;#?`{RMua4=fSzh-5TBYT1_AA5Ex76{we{NV<
zkSdGey^8#M--oYSP0(%kV||q{*C!k5E44_>Rwp(TZKxkV>?|#+<k7=|WKLxeRWL;^
z`>T#;ho4#>6$7nty;xASl+Au}<#$nZiudd!lI%8IhI-|I?I3xkpVfWX3QO$X1*yXg
zUZpmo&Z=0UnM;y_Mq`a+!%STg#{?p&h!2$j8;U*7M;<Ay3#S@6+azB4<f4q3#CjPf
zD1-O*U)GM2T!Bq5Og*lxU+L%`qIp6ox^pgl?!XN`76SAFz8*l>rg(kpi^wqAZo5w@
zziv%Uc`%g3!0f-rMw4<W)<0;@*|suv6H~?72nk`xcHM6X@OEArh`g4t@bt~^4Ga1R
zV$v2{%vt}8Sgd(>o?(<zC_cZvKNQnsntT3=$L*r%mHy&_%$&nq5u7M<F)C3&oNIi{
zVLCuv84KZD4|yestg(3(+^ejZc9@qskvqG3)O5|d=le|F_E|TsKuo<l;sGLR!@OO}
zWt6AzYsf#lR;zR(fc8p_SUbjC&?2c_9)2)#RuvMEIT3G#t7AF1`KXin1gh18*M0g~
z3`HWk0{;xAzgLh7{$!P`JvC*0%&QQ867fB0U6-6c<0~}@x~<LkNr*t^)v0|f&&KA~
z-KzyzzF5WqO}kSWyhB{j(y;{vVI^bvxrH^K7)?r1MS@kTOma(;K$--lW8>KkZ%C5H
zYDhZu>tHiy)!5J5qBzf7YGc-Ly7vdLE{-LdpatAonV$Z`n|LJzcItef0$;a(Rh31H
z_JwstUaVzC&q=eS<u9{X4+KKwWz45De}u&R-KK8+ob3$5P&G_th$gQQ@zXi^KJl0k
zl8=hI8`pu|*Id{Y|M5;nX5t2N-d^Pb9o1|oTw-|^t_{%baw$uu69R=q@dKpiDi;i7
z{xZqvrF;__y0Qs1yEWEX*s>_TRD&DD3uAuZkVZu0GI^tj`!-6BGG}Roh8-0H7lSb3
ziNa+SPO76=Yiu9-p8*sL*TcLf9Qrg}Uy@@rRa&ly-Jn|nF{B86AUz-fM*$M-O$Q)$
z;2I<?Kv;8&U;E#^ZLFN4qf7ivk-w7WCwJq^*Dls^0&xdrt{np*53Y}#7DWao;wUq+
zUw)Uy%d&g}uSnKg23s*@9Wc}^4!NqO@~31C&Ff_Y=)$2kg6PbLrGWTwr8=oP?$iUn
zxsz$MDKG5NhnEm`!Ojo;W71`Uc`1go57Rz}0)r!!=|k5{3RP-}3HZJ_oh@||h%lS|
zmz4|YeJK?qo1cC#u&}C=J0F0IzxCw*HJrthCl4;}<b;^)lbODy`^+$YreI7))yHIG
zt!JH;{|%4*;aR33&Do&c?Vr;0U5~@u%3+BLCtC*?kGufAP$VdRVP%;y*5baUG4jt8
zR5t?4{O~wL{MY`LNixW~dMtK&RDkJfn(ydXqVr+=QS|9#m0tYi>jK6yVi7r`WeIaT
zZ^LqF--z2YJQbFvtcCHU`gLtan|hmx6G0%5|IHhX!j;-Q-0G95;PHbuingKh9bb^z
zQF|oH0kqCT6ySNEsTN_<&u^J#I2O5`Y(ze6-j(y)6sSL0rohkA)fH-oQB_>Xj5OCz
z)h9z_^+f~7lM105e)eW*$&TE`fy#Zs2UE<qbvd~C-h(3TJ|tfx+;M#@W`=1ZmmJi5
zKEb~Uj7fdC1DQ4EBkkm76S=LA&x35*sh9s<=<=^UtzP;4F(ZA(RxzQZSvH&4^JURV
zfc7O$7{7aY>6-%~G48BU3gOO(gE==K_IO?8+e3`yU;ealN^ujOlmziOO8kYIeaD5)
z1)BK}N9qNBu~Tz&%{u}04`SxP#mv_fGkGD$cq`4KLLbL|O7KTD3BR7d08tSV$$d^y
zIi#)-w6`qB;-$%%qH&nBj;>Z6qB0(rEko|9mer=gHbkM;?Wt?i1}CXSYWrt5GLNgn
z*&v(#U<y2j@QmNbsK}sLzkF?k9*b$+aftv8&ucp<lyy?g3h|Sb)$?CsdZ#y#Wk9O=
z1G|eO*|QNM5HZf+f_xh588g;;qgYUF<$vaT+i&H>FWyFX_eMjom7`CglP%p_U}e+D
zTWL$-vGtk~g;!$UV5g!8%8Ma_T4XzNV8di<lei$_D<qo7^KxJwli>J}Ti%fTqa){4
z|GN^2r9c@V=5@#Tz<_`7_IyWojB6WVWy09l5nHjXbZRsuEe$b2)Ce!vTUQ}L?wpbR
zVW@-RtvG*HI`kOEWul(CjR$Fv&bS{%_G#ecXme&rCfbQS3s-1?Gv47<OmGEC;Jf7M
zKpB!l-oYX~1#@YsCt4O(V>rq8K(Jrf$e|axL=q-%>y~X-1pn-^iov-41%#$JWqEu^
zagL9uM%^@o)x79#k-*KvSA`ffZ2#(AczasN&iK|AKVu9-2*0ie->^&u>>746)g^})
zeo*`U`}o)x4@aZa0g{t4K{+<k662$89kgem+LnRp;3E?Rf|r!oqHs~xK!Xeq+Z0uB
z504kBuzCItEP-Msh>YzaR_fI}`4rGDJAMAOneaDGa#8r>G0(iTUwZ$x&L;+jio|+K
z!gxM&nUz$~)Y6$KMACqoA{P*~7zk0`Kn|DHC>;R8vlr-@r1D~w4rbQRyo@BFc@%s_
zR#5iT7rK`@%Rfan-L<|kB$d2)nUmo&7N;OWD#`6cyZb}5U7B;!>BHbh)(!x~S@SHQ
zV$40ylpG#FQp(_3(qI2=H#9Rtl=G<$H<vxIioxA#FNwHd!*=#gfb8M-_PR5b5>iZ%
z|MXZvMJ1dKG$xrjQ*@n?J}nhx_N}6L%a(-yygpWB#Cw--i7qybFSwv<3Xq!N!{(e0
zBjxm>4_{Hp+7>tHfNqFnicHY~qwv#t>Y7>xcpvtwCd5Gsn9mvlaxZ@ZJ6ttdCgk4E
ze8-(~veO3gOYD=zJ>L;GDR_J8_VmRYF(u7__q>M>Eb?IyX)<5<52HGX?A05MaE}PS
z>69s4QEC~($j|<|$#hZ>sPzecJDlL8HWIIv^q1!85!!Oe<dJo46Cn!ieX1{t^;8_=
zYbG-~^lTPsScpHD$SUVWqI|R9w!i-VzQ<6lsq1o0GrX)f6FrFS7eGs^T0CICCdV}i
z=6?B{eS!vi3Spn_v{CC=VAhsj)~23DD4Iw<!O85o7vi6@Ep+zhLO1(qC*)ymk)~RO
zxC~H4X2{&=qp3iq^cBPs+q3olQ?zADG<@Te8<0rRokT=|Q@&j8PCR}xN!9^EhUt}!
z?<WWI43b+bqMZj2V1eSh)O~gxoaY6B^NRMPN7>&2moMFe5(*3HdE4&=dfb~luL$w+
zW6JZEyWCB(Dkv%~UyfDII)LK)yo2tQvTe%XdS&!Xpm>v7oh4fL;#iDLLybQUBX)d!
zRdhZYMnfkBc&2Isy;;IB;*Me(RE^QU)guO<YdrD*K4?sqQ8vPL>@|wmr+nT;o5&Ew
zQW8dawVV&`MErgHOO$poOSv6jJm|i0%JQ|pHlOSgq=kS;=Od)HJ#3FJSFIjn{MmMJ
zrwXH4h7AVvO#Gl$R^|-HThLQq^3t-tIs$akmbA7_^fz1s+NU3<cq843mg*|`K}sK!
zk*(09G`SZ;OOF=iE4e?G;xkpcne3^P0Co7Xh12F_v8TofsSV_<ZP^jy@KT7n%9N9X
zyadqmZiidy=C?ep<WH-RlMjWm`MsvRqTIbpD&q0I&2gRt%JR2SQE=Ah6S$(z4TeJ0
z>O3YLBiZ?Pz^cw0tRojx@(_~4S=6d{BrUC<h1AyEWEva>7WN+GQ7zi|nZQ1a*eccI
zJ2fpC99oOYN%NDcn1$c_Mr(*F{}`mzPO9<R)X)i@vUV@?g7SfCLUr=zcHlrax}$qr
z10~SW;dh|2caEBIdXb3*Yp4rVV~$I6+xE#UG689lPFMQLWUsRr$89#OyS%<AWrUlj
z2$E#R`_XDwi~i-J;NY+H9j;|Q&50d0YYn%FA&0KmXwikm^;_i=#sWY+OFZN8YHP%9
z<D{lc;i}ouy(l_nOw2H{w_#JI$4mjpiJbqnb9AS`!GeRXRG6{+{e(F3S>B03NPyvX
z>xZIx!Z<{c#X4AGUfzDxjqmfS-BfON$<DmHB(`49<Rjte?z_+WL3Lo10a)vDKxN$O
zSZf?GUpghuGB%C=%FUs?0SSH;OkVI~m(IXwOpfKKV~DwH#>4w^Wor({K&>=w;$In@
zugh8&K>i;In?$SgIfFI56%#DqZ~Nz5NJ}C51<G=e%a{|YHokLQde|v=dNAlEb>=wj
zmljpplg5xq0imkHI?HieewpOtGxTt;KA^sRa@O$kD;=|*Xxb}N*-KYVBipF48C21O
z6fo{j(CZ@dAq+@}us(P70Kj_iUBScmi+61L^pk4xc67|$1K$4<1AZ#n$s?c1QrW=d
zRxdOP7T+lHgy+GRo8PYwT%X9Vj(^I7^^FZT_GfpS1Gn)SY;v1#F!x6(E(~AHpW6v#
z_o>Kkc$qq?lZ6vt$$y;uWLkyNtU|3`Qv5sE^PBq>$D~6r_nMJDf%XGKt}>s(!);VQ
z_`nUz`UdOx>1?hV3^q46RVc-!UT}>TNt}i{n5hlw-mJPSyjAN<r(i$QLLj;A^*>tq
zUY;)4;Em%eG%jw`dnm=xM2vw%V{$Mm7{H4LUG}nPejZ?Y7ftY{6&-YwYrq7`n)>Ze
zf3;$mW)gFrq@=JXw^~4gde#1wb|J4AJ4YJh4XdgNazZO!T)1Vd-#Z8X%m5E`NHdk2
zkD84J#Wn>WSw@CEOXoHDyaOR5j0c6B3Ur<sb;=h?Mv35QtCf>u{h@h<a);L|;YF?6
zg&%<WPBYl>;HeCN1WN2nUw{&Ztc(O9fG~0{HD<0)v(F5~6fLCGdwkg_DWZ;(6-O0=
zZ}QWSyCO7`&fNy4M<9(Tmyi@?U#pugpR;4GuOQRggCGd45Y~0bcZnzUstst#Y|(Be
zQ6SxLb8w+~`opHa^<}aEyw^P!9i^6bpX{aefmDtUK=7O9@Ep;3N6*`KEwDKCjsnCU
zq8VdX{E8zl5{E>OQf><c&)wra+AQgBBb&C-FrFTtwy;j(B8R`rJhVP(30E#nXknls
zE%M?xs3eg;kLHdqWpFM~BP=d(Q?z1#wfQ4t-a%^AC&sXX&@`7%o6S1NFsrWtE;+~o
zhb+CY6L80bN#vF3+bVHl2)J7ixYWspguEn=!Wqv$<Q**G$E!Hp+7=JFaPGKx{lqvf
z8It-hhj@B1-uVapJm&nmLj!>2wX@3yPJw;3>X%nR<(=dja2nN$P#HRZ*?o^wF=77W
zq1E?haM`t{x<&1MN48*ecy8KS!e~|FHZh6gxb5_5&}*6~NWMf~ir(<1`Fe@!&JZl;
zkNTzs&$jjW^ygvRO}XL`)eZ58Iqn0hJwTrG{Y`zE1u;B2-Kx~;OrMW^+Io<q=Z#~d
z5#dXF)@;Ie>MSLj<)&lYro&*hK^NCK8AC!rV(3;_Du_P*o5ZLJ<5ONRKeJm7c?ZCC
zmdna8+FxJ+!HDtEPH5*LzrA-StW_vW_+Rq_>s(gV{n?xApL!iCOf)}jR(O2Tgi-X-
za`<#|zr-7JXC$JO+Lgp)w78UU7p9F70BIxT!of37U?D?mz0(te;Sr%g4W&e-Q%Ndr
ztrM>r;-Z9(wO?2iL_1$PJe+$oC9`gEmL6t&%Tc^+D7COH!pJnv@vk>(V3YnLfF>5@
z2Kv?AE?#fMO?dfI-$Xo$(PZb<PEx$y#ADIUN+cf+APzNiARM^X?|$g+iyspMiqFn(
zft43pEVHL@q?Xjs@hZ58vqX2Yx<wP@7so4w_dM)%p5aPv`yHOof^-$jzrk0%GKZ{O
zDG0ZnJO7F@KAd@yy5oFq+A`?+Yq=&%)$zxGxiY0GUY-oq)S(bC74Vyvi<iooXv&4q
z4XL0a-u-!@a{4iK4nbjN5#Hqxk11s6O!z)1EiIjN$_~|)vb7jo@YcL<r^4v|@=b86
z-n;j?b5kS@Hu&@2&qAYx-R(4RNCrdV6G55WmayG{tbDK~P9$UFo7Wz9x872>za%Sz
zJmOL>Jnxk}qTj0gW=pgF=$0~A_ahGyR8Lib7@R@M?`ym~853aUv^1DKFB#o+&w{*P
zqmX{-n{32IuTu7^=uWXHp_`7B`mliJvPR^i)@crxtu&{II_P;FB+NA9UXs!#pt+l(
zpbZ`>cfDtDRG!<23e^z}ym^0`X^>y4W_RF}Bse&3T#2>jVW@MeG1JUdQAXa?ybv;t
z1&c4c_9Eu1J1F`ihcgjrCuH;U#QkW+cX8=wAHoOg;EmY*dB`L0LesqjUM7Lx07==_
z4M_L8Lmr5l$rUqnbm-8t`#pPMIHE^2Mp)}o^PAM09?v{qfkXV5y+<PVFAMI1&}0B`
zMVp*RH}qwS36<^Wl3G~WuK(fTAs_AEX5Ryfamg#-9d_&CiNvqDI7zA7-|}<1dKvHZ
zH-6qKng(vGQ^kqcJ5yCGtOoyhfASD}we~*$&ASl9{F+}F8-d?u+_?Mtz2ZRvj|q>U
zz3I>rs^p5NctahMQH||lJr<;)&kbg+n$Bkahrfee8cj~^S>UtYNm({(Jici)j+eGs
ziOlFF^j<orNdL@F)qHQqd&-2iFp9|wpZ&vq784wD{lo+AvUqAauQO~8=6fN=(xDu7
z2U7vHw12=X0!q{vgmNu7^phX-9RJnNFc8Wkz24fVLx4a_g3Z$m^D5NMg^twi-tzs*
zRcFH5G2MgMzkGCOch}xqth~H(qb09FRhNqX$00K8G^5Zozju`$JL>MMeqCLA@A|4#
zO*EgKn5XA`UC1e~eZ5hm9E+ckeH*xPk$*5*CM?&B8rbuXg?l~$c~|AfI_+8#*ol`t
zbW{L5uFqh>Uym1T@XKnsli<<$Lv55|a73%3pDJvk<^L|fCveMM#m5G_zcFk$e1S&e
z>)9XFxJX&gXzhx)OO!$8ADVO}9n;})ogfNpj~kd3ouBcI?J}ME6VGbaeV!w9n(XK-
zh}Q0w%^`U#Fxo3lHtL!5?eP_W=9rO>jo$yXKmBHQ?M8A(Z|*HPpj}FZGcDcGXNL_#
zmS|ywO?~u&Xn-jG`F_O?d{1}qFtn_|!bU<`)#{(Y<{=Oiy&BO_=ntFiQ^jGu30#Va
zwSARTU!|P_Q_u|G1s}UeZe1Q3IXRURW77X~@!i!4`ac>VKQ~tyqdOxeJ+EZ)H$%8W
zX(IK!)DElk<Fn=EYz5iAitN{(yO-V_e<FBKZx37<kTE@<b=3Yvv4il`8)3(tz@@@f
zsLoDGA2J}0VUm4)r2g{g^x-{4Z|blBe&^Wj$X1=`;&8~T;jBjGq}f5Zw;mxvR-1gV
z;@Gn}GSObqi=MWyqUz0jgZN=;v|u22G8Vf@TbAq5e*Nq(`$IG5*x5W8Xl28{W}rlX
zLr4C$nhN9}hqoqeQR$d29XAEA)*}m_%wX$!m35J7?L)DHt4zQA@(pD~G?<;C!v6a*
zTLo?pQDKAg;(!&jsDf!Tga2AMpH0`NH;YsBAM^ZlyY@J~*c8|wfXJUB+vYJI@=G}Z
zVIyV4=8z0Ha@rPhf^pu~0=mrhK^h(PC6ll72Smk4B@W08RL(0>Hh!7CM!9F?zsqRS
z>nkvTE|gX3G^>d6ye{RwH&6~kNpSkWHU;sKRvU944X$iJXA&=n#ytLt#w_Gw#IpwF
zVaXh&w{hmORCqi<2XxfYtwVb<dyK~afQE<m#B@bp6j=~nT>>r$PmK?7WCDac?i=1&
zyV$2{5}AdomMohKQSk!_N%KADYzlYgLHk$kCjU^5fOOEdsW%lYeF$<a3=N>23f;Ws
zW9{d6uF~uX$@%u=XJI!z#k#Y%%f8KwF78Pl8%FHfb}Ht3vTA<b9h8EnsiJxA0KoKI
zA=Cu?IaPlBC#JJLl!yq^ay8_{<kH8t^hQ{z_XD(b_yOB@?==m2R&GVq;p^k+eII+F
zEZmym>8hb+_n`gq4SkbQGKwNb3vy_#O}ez&dvegIO|8XZLT}CH4^v@6^|{gqGE!%4
z<iMLu?1#aYtbw+kJ0h}ADxwv80U8h!cOHIH#yH8HVL-Np!*Xe=x2HvYEi0I*gU?a;
zr6rjrUnBMZ;^?~Lq5l7PB4jJ-;%ph&5{E-%6H)da+3Uz&WzVy-sg6|k-g_KbnP;yv
z5@#LG_WOK)e|YqdKk9nlYdqJhmq6Ja<Cw?&3H=hG)y<{w9i#59huw*bl<IQJ$-Vd*
zFV`z10oSrHcV1xxI(m!;y(3wO(0KHdIQ*KbI)5!e#on*EXO^|%+9cpr_LTcZTR-{g
zcVrP$YlCLD67AQT!Rda8XjP|!U09qw{c{6T&K>pd!`{F4$Rm_{qhFi34_R?kD^I})
z_O}FnHglsqqgb2Id$0j7^=~7zzonRTz}c#B(kUPB%Yu>M-F<9z{!LiAnQ!L#xo?%O
zDYJd`8uyaGZ~G{V?N1~}K~*_0tUL+TK%9>H%_G*##ew61uH*jfmN)O;@R-U(Pcqn2
z0^T_rJX0O;Y*9Z0ZPAyf52oL-tIa<5n$x*cc-*n+uu_`revxLu)e)yWxyezybf5f{
z&O-HX<F-eUc-PlbbT26N=0qGZ{G>cvR4(zI772!}5j)d~<8z9>d67F!AS>eTM8ndE
z@oBeNY>O4v!#HSEbrVj?rqK`_0OsUlP<NcwC{ppeSYNkS10zrWLac<pElKXeZkJ2|
zayp&gWV00)6cWQ0CIki_+lgk&A5sfUX2mJJ0;39b%dw8_7mh8!r$D7(#pooxezd_O
zl1;w1)Y?5g&Gg?>bHBNgXumOPAHKQ@xSJ!&q!2eKRX6IwlfcHB>!z9n=F#pTXTfev
z7jS;h7r&uh`^BukjeDlYsk8bU71N%ugFN*=ESK+!X&_`irm0g&;+J{-pzWsw{`iyk
z7gF&6<kb-Lz;{O5cLG`MZ{w92(|m44`>>aU+r*9CaDIYo4Kifh)sWYLW4ECC@~y)G
zRprg|Jg)xz2cf&rWre?M{3<CT(D1D$FyroX)ZUjG7BV?SB_(+@wU!{D`K+H<)u4IK
zm;eF8g&+i#9OGW|As<ZGvp~Dl)}VzRrl05rpP+cdQjJz!TVK!2w0QEiv$XPqg{jAa
z+Eih9@D#uY$as7dj4|pz-M5{s59AD{l>4}Eggj1ko%!f+6!_@8#?9$blD4W49XWBx
zN8Rdh9tH{u97ICm%ak1zGRCX<Rq0X>c4q>{PFvs^m&z}pE#FDZErQW4u2M%P)g4{`
z6on#!zN>3|Z_Clh>zA!m?w<$DC5&Sev+op7#~(hDZ%WMwwa}oMKzx5d_x5h3NPLP-
zVRcu2gvbo#+dC(guQa`PpVFwA+kSuI^2oN~Ik6(1H7S#LlDa?!w1wGVkME;tQol~&
z2RiL{I%cwhHc&{HHv_xZBh>DPA1{nR-1_>a9PkpWVZYW0v`-w;s;15ssfz0KG`Xsw
zE3DvD1Ha`=NWmH4)TYvhE%L)?H2U<fa+^uH_XTtA$=x;%tgf=eE3%>5B$9J;r9{Q{
z-Cdm18a3wgtk?YFA^~BPUl_HqiK(p!qaf(mz}Q%pxO3@7TFhkuXlj7AHmH1bvDy0&
znx1ZULpCadgSrOYc<}O-YaAnM_)EQP+zr^uw>ZlkSqLEazYhdNt8<8cf6uD|;~A@E
zZyR<W=OllKSch4clT_OT%FOCgs8u9ySV}h8N>t|jxKA>g87*?uLX&>{i-VQcnz?tx
zI9F^+NkLA6WcEi8Q=#ny^j>F1Xy~gygh>2Ct3P;hNT0U+nd@io1lj50sZ=dc8l9j|
zZeYO&_S(3f>6;-Nm6Xc^HBn3AaO#*`j8a%P3k>v}F<yK@=A{yg1e`d8y1SEF7R&;D
zyxkPn$8xC@c;FC6K?x&}hY=GRbNlT5roHJ6J~R`ouTe%!bzlSkozQi<^;})C)_Tpm
zv{9YT$l)t1g=pyxe8xP|yq{Z}Z)IuB+AOA3FZahu^e|AycXVnmL4!v#<xego=ur=h
zp)b-35S+{`&TU?uW5K)5)$R2Ik<AG+^Rt^>;jz4Y`RKJYJv;BmDV)y3y;}W?Mq&+K
z7&d;hw%=2aO?$i<KI=`}i5h{omiI-hrg(AuI(6L>Z8Cg>xe6o7yUf{!_a)6WuQX(-
zUV|baWwGe#io8^3{0}6}{c(cl2HD)7Lhd^oDBi%&)p)F@o$7sL1k|-ZM<0LpUex&9
zu6(5$#!v1_GP;qW=<BV&ig>w6@P)b;bNt;cOx>Qt|EgJ{t&%~>t((~#O2?Xli%9&i
z_1F!Xpl>dwHA(kybTWImz$siY`%0u}=s9>xuQ;pa_13a_7&qDbpDb(l-2s5zIo6#{
z*c&`>6Ui<|+_cDV0h${YJdtQMO#m#8dyCEYI&vkweUJZA`(OIk%=7UZw{#WfU-K3_
zIvTgR)1q8wJ`I4v!o*~7guypk!5aGqf<`XmbAR@6KG@;W!z-)+8&s?~LHJX4iP>A`
zNlnwGoQ{o$A<wka%XDWR8g<Me9xoJ_M_bTomi4AbY17N_)xBRXRIYeWn5lM9k^!$U
zV6M*Q(bH9Y=|R43t*87!RwvE)$7R4bZsC-qmfIu?Ke~<GAZT{Ycf-Ol$olDt(be+p
z=q>(<&7e!7@=&nxGV&w<TCZ3%0SxY|K4EN6Re2oZCE|5sZHjZxdXHfNOf1G=<X}!w
z=u3Mw;=)N{=Rb25t<|A_s3NFC#}e(J8oP4%7om7bA*C24VwR{rbQr%ON^@8@bTO0~
z6dm)T{mQX@Bt2-`@$||C?>ASCt}^nKBn91$HAL#2I!QDQ1&0)ysJ+#!coS&XqH#bc
zDL=w$M<}q84sWu>P@{&a`r$Q{c#&62^=qwSaYiKN9=_Y^l{g@&e+TxMoPV_=Bq@ml
zK7U5c2sn2SO8l7Y*#ENScEo6*-9I*{T!|*mx5hIe-RzvVy!F(t?zy3T+RD_SlK#lE
z3AD!1y&`ZEiF@7f!;sz7@ZZjtAiqeWFII5Ez^qb^SkhQ9IibA8(A+;?>tU6be{MDg
z8@FrMMyk8W&f-?|T12Jr!@pf;UH9jTeODXQfxYfY3KaOR#m;p9u&~50SK32lp1@8R
zl%pE_&ogIj4rXp}4h=bU>v4lHE10sG2l)y;dE>-DTN@qTuQ_Cv27c@^T{>{_bG+a^
zLV1;LsEXcK+uSc_OXfAJ9zn9fb`GIkoKQnq?2g3lJ|LrGI{LfH5qy&#tO1S)*i0Ke
zI4Qp-k@3BBzS$kUNnJHhheZ9CTCuZ34}oizPIoX9hf*`O-7<zEq>FiYT5jeX0VmbX
zn!rs6GqJv@!N&9aK!$At6crMqkSP)c*j^Giz1*`-6CB3VEI!i=upFIg(zunLmBfpk
zr@U{;c)k05KPmI&Mwrk!AGS?fW8O3@XY%!Dl<I6ifwf`civP_)Pb&B2=Os^$1HRk5
z<OJb1oz_T$cFg4?uJ>;&h9a4Qti@P~)pATwY}zrMcJed?#M}T=(g5$FW}V;WGTxCH
zc|9`hWlV%Tw4bB{CI-jgN4;^Ocbv)MBV2co=7(BF5BS2F>4<(n?uy9LurlN`#|c^x
zJhLgAC0J!<kS!YPAmEFeZpcI0ng&rL*uu~T*r!$l^QWsU*jTkjm4^?UY2fKT_XWrE
zl9O_Ho+YPSD03*ViuX!Bc^f}H^H8|o?BiHNnIu-~3c69^t9;pvhUlJuC68CkjgR+~
z*3h_gvrcU&5>i(B<5b%S!_w_+<#(6Y3--BfO?v{^exRu6^nZ1MaF{c{WpraJ?-hxx
zAfx-Fr2Mpl2pyk8ZHoQGgy~Ne*jCMvel}-8s`9Qq(e}6#jlx2sXDoySL;<&}r~j|u
z);o{(8X4=GqZgiOH=R;#5(NBm!Xz_ghQ9eh*v2ZW*4dDO+VvNH8(wpmuTky&;Qq5B
zD}i@*n;Lh17K9mWLHS!Vw{?4rJooj#Wmx-sm;~}n@IHk~Il%1gY5DrG5(ua2G6rv{
zb~95Ao-ob(t`;u`RZM6r)w&C|ShcD5wm9=e-SRf1azxuy@XbI8$YV->;}*A1qd}WG
z(pM`|jLy!_K@|FihCN^00@F99o*H5vQ8G!dU4?I{>;!}3d=JAP1E?rWaPxBcQhI(~
zQY4!%FZkw}*;$vs%Ek0w-|6fJ11rw3=i+|b_BoKGK7kY&W)JP`m9$j42X9GN*NaJz
zNi|o7!rYG(vYu}TS^FJ97?hG)XwzR7k81|&^@;@w^p5oq+#?BW(SH?wU(kB4lSb;d
zVg9dU4Tm)3s(#kj^TPK*SlkWw7=$_Im4AUHh?>A^2P#QS+?B{hcE9`2Mcu@ykI>f=
z0d)kqD*ZP*r*iypVv2D2CsqsyP)UMDsi5I3g8U(0UX0b}&8=;(CLg(R2hpgmJN9JV
zu$N%Mzxu09rl&Vx7JCW)%6+R;%fM@JwpiK21o)47BW08VcbbzlmKdzkZ2HRbQX%;E
zV|cG`$#=t)qox`dF?Xpj`xqTPI8KWgpDt9`OF{XrAt+T>m;Q_ZDSQ*C9uqH=iE*Wi
zoQ|2VcZFf*h;o+ox5hWHE!Qbb=0h1=G20Aq;=Yhr5Bk%nVZk0xDSnftFhM*RxVp6&
znYLB-$3bjndDilBm8~rBNN7~bI~*`y(KiH1jioZCH>O}sd2uC`7b@`Y0YfI0cS~s`
zB_)B%AdySaY1gy2$1gcgIrgjz2X^dvhDp+~04SG|YS^JT-7v-YA-79Y#<L$Zv$KKK
zHkFNLKU=zf*|u$Xm=@!?+L)58{`zMvv9NACg%$6AeK(Nyr=&BZ#zi3b=b$`yL@OhV
ziskh$Ax{HCgHLj%nySy+MH`)T8`R3Ee+Yv3q_Q8ZTzOwC>}q_r-9CdK+T%~yM_m*p
z?WI6vW;aXHuOMVU8!41p8Ar|9hNd^A{=G<-xt?a4aH&5uN%ty>7R^dadjWh9s)(sB
z(C}&}UUqgptZ=0>)W79&NphhD7I`xkb>1j`bJBiO-WYH${-MhHUBG2;%Q`)`_STf9
zzfBJMRg+Uz^5Y!@<jG6YF#c=Fy<ODtsP!mQ`<mc2=U0)#pMG__d|&=m|MORcmaJzr
zqAS=}YYPYLV(O=yi%DcOx@s)X$8n4Qyv}|L+^Rk*QqHt%qOr$+^=&|BF7Kyvg8Dmo
z3we-24<POfC+H+*=#9hocku#+#Ua`Hx>@SW%s45O<X1sy4eBs(<e^WGas8{J#Sr~f
zl2ASD0XRC<&F#le&nz;y;;jCbvPwxQPrg@kuMhYL&2&RtXr;JH%u>o=f-o(Tlu;^@
zL|%=Pn3i__`~3`~8*9YW7Q5pwV&q(aUh|hO+Zj`LIs~KecFz~(RO7VmKILYp5fu{r
zw)@7+3}<CUaO$jZ_D^(GaYic<ciM?m)wRN{OqP8pA}^Z>R-<~u`9o#N+g&u-w=p>4
zK73KpEah$3<W6<Pq&)MVIk)Rg@()!7m#zCj38CHi0izqek4+1vO8$#*Ge+|L{V@?o
z6Gcb!s5*I8>z=qIY)u%^V#~5@yFi=mh@zmNXz<?W2bPMUpgh0UKCRotRktYgOB(or
zF%b%uSY|Ew*8^BX714<&@oFWdW+nYT6dDR_>sQhPP7;|qk2=$Oik~M6#I8U8^Xo&*
zcJAkzw0#ELb-{t$`rrc{{4N!D!@E9o$qos2RU^_wN`X6nRP+@x(f%K{HZdKkd}9UH
z;>!+ZC4X~#WHs87bB*uYt5rh$zgGyo|7M)pMQ;_KUv2Sz?#?PN!=}nQyZf=dQ=T$f
z%Eenp3vQGwH6WIAU3Nc2h!jgofMy9kXSzvT#PFF0p6HaBT!xvpk6vHmlC&%8>ov<!
zwyv)E;o*#EW^<RIq@Z@B+4Y7QZ2QW0deqhJd>lJ@^$2M0wk2hy6;(|!)U4P7GtZ+D
z-u*qzTDZb^TlLeoR$~1mFCo(|uX9|4bvOdKS!(cj#b0|Teh(97i>Eo@7jn?dFYT%@
zSk}y%OuKL6Iin(+4mU8M<n^3I_6}CB+1Yd({W*xK8Gn$YvN*3n_2xm#L+(0u4q26#
zxhsi(vS96_lsBiYFYoneWz3t=`A<90M3NEVQ+U+_eAdGVp}ehc;Va?kreB&Xl<S*c
zB4*`!W9Y?P?}T`AYgKl6mm)tx)SZsM5hFxY6J%9^t})?LG2k&Og{-AqA|mr|_Gd0h
zkY;%b-(e0bFfD(Jg+fR3YF5u*fKl(k1Q56yfd4&Tu0=7Z(2&h8&%U>}Slx&5p$`ie
zT^=e>8*+nD=n52K%H{ctM3?8zLX&6u;v&8mt}b(b1iK;7Hu7&y0${L&gIc3^4EcUc
zXNj2xhss<aWGsVk?~Hm<C(Dz9k1n#!R<6#sWL1acUuWh}l3I*9773m_OKPlMn=Ba^
zwqLIl@wy{Rb#Soyih|-45Lqqbs0Iv@+w|Cot&EAl$X-v!FpK`vQ?{B<YUySAk&6m%
zs97@^Qm!?q3)J^^ad(AYQISRd=)tp=cwTll+u%9+mzz$0byx1DKVby-xV7YYjIV0`
zZ$q;4vYJ@{kSHkg;frA-R8k5*{;KQx#?|}T5og9^M&ZY-UWud#-j>0*FWX`IG)p_|
zG+TyJ@fHC-J%9fahgk=~s8QYzAGmpioYvUnA$zZDCVX;y2KR4xiQKY#^rit%wHAoI
zd2rfy^_(Y8(aL4HO^HSc0=EwOI@lQh{#}WdR>z5^G;^GlX)u-)cymY1?U()~`SUMT
z)3@?uRw!>*Zq<qnUiGpVB>)iMhfc_U3w75PH$#v#Ew!)^o{EV?Ve#F@x2L>6sGg^5
zKPO9J_5Umf60$(#KvVL@vL0}<sMO9OlbU|l@JJWM_>oblr#?RyJAD6gtmG+j-BNlq
z;%yP9w>YEvnml)#h?zm#zz?&!xX)0VWR<UZkkyVviTlzJNXK6gQM?pY??+Jrv=Fd>
z?e03wz7>eO%Y8sc-b5M;y&WHHIhaN5a>Of=7KEl!w8wFfvC<85w3$`&zpu%KkVoM6
zbql>1MX#8rFXlQ!ku1b_=SoG(P#q^w-BB}E;=5gfL()1l&*|<RUn(`g)kU<jxZvV+
z6nE^V9U$6k4G&Nrb|O19^yG0g8521c7pWDy_PthtG}AMSvp`Ifx69%t(zP{w-#=hm
zjj!$CT;QnG^;)Rj6^QHdk*^*D9$PHey)wtqhUsUUbDXG|dH;?=wRGR@!gK>e<F)eJ
z&2nO-m9F0CVOMPVM>p*$x#=~sd+;PmlrImT?8G+bw2j=S3&@G5qGeHvOtFJB6#7zu
z8b<RVPH8fL37N!2YZU5;WFxaG_}~VP6eNKaF20Br%A~IP1pcd?{PxZ>13^^{(yWm_
zf>ZOYL!Y75PDTk+p#I=-KiEpBOV6hx8?^ZK>=dSAQuovJZ}g?8^4KHZ=0W*bahB3w
z{!_Tvg0Z0LpeOe%1Z_khs%5jfbD7!Q#ZzMs*NYwAWYl>Ma%J@|yV8Ui-=0~O1)kxN
zdU$|O{ft`z!qcU(!oyw@4lY$^g;TwB!}BsXk~de`!E%CN)-ZFJ{U;#4HQwy(pTL+Z
zG?eW-R_CcE34hTO-rKKtEGk<+e_9vmQC0=w%S0P)29yH!{g+wbuZ)ovZXyL^b`|3W
zUs3z8Nha-bUjl>>O8Pu1chLl(DD{nu-g|FRqof-n_I<%8m%u16Z`^)(A8p>Vw?wgg
zwaGM&`EcG_eoIF>$P<1wJfi1YzdP~6RgzsoJf1q_U8!j}gHEt4)rucu-q!nwePLMC
z-ES4XPK5j{Bw0<{ABxLz!*n)me;?{dNm4&jeu1uX6F{zic+9PjSu1di|8=a{r2F)|
zl}3Q)H{Y)~fn;j2%gKk0gQ6G<A}8mPrM&}lCKY!LPby&$7}iify0k~3MPtG4b0XoP
z#$)Wa_Y_D6^Bgg2lPewT__B>l0DPozA$Q7`SjiM?Hs?>nNZH4Uy9B$+#WGxLVL9@b
zqGHI44W18k&i?S?)D}jbmL}I1sXs$)tdbeN&h1^jR?d^k<24sy6rv-0l7I~BOf-jD
zLK)|!cO^vV$T!o|-5hS-#p|Fhx7Y2DdneQ-I$cP#w1%=>TLw<2ImBFytwiW5TU&eG
z?7Za!1O(h%0?C~#;A72oW;+lYt84J%k#d&8Ra@8g;d$5A-hbo5!1$3?T+W<hXaW5#
znm&IhXmV1dd^Y}FH}`fzy=0;-pGazH_Xlr+X8{Pr2yI9Z@@n=(m~z6c)*!5LG;`QP
zDI$D8n+9h9yhI}m{+GthH_`%4{Az<+d;#ChbpeZiLD`+GOu__SC%^JVmodqO!eF15
z^^%(ZjNTTs*1n5)o75B|As}O9A|rmT@|Bav#I#4XOZ<suQkW2}%hBZr#v&}?xlO3$
z+oAMt$Zot<yC}~>ds}y`+90ujXOw|ms?LT$z@v1f@h8W7)1tu#w--#lI845IoTa_R
zCq(B*ulnxt01QvSNqW1YmBl^2d*4^I0<kxNH%zyvTl#s*dbh&Z{bs1&aTJmFe%0k)
zmfNfLD?Oh@aqZDn<N!KkWScNYlk<i%$sqD|zNjW4sURhj(~%pNx-}?)8H`qqw5c9h
zFwO?_WQ<F}{XR4g6tZ%2O(Fod$fjY-$uwp3w+zz9{A2qJN1d1Z2f)umWL9tQQ$++G
z<(^DO^KNBLjfj<b-UzCQrIBNp`Ms=1z}-xq)5b_ju|x2{@88dcy(ABJG6hOG?yEyn
za{q+(x4ohk`z7SyWtI<QZ-Kw^pdZs{G^teTt>bHoyYNh@ZBE&5%(&%@2R6mA`UmM|
zEI1X|-PNY{zDVE1r|SEt+hg&O=N^+;bKmL>BYLd=V%US>v)G6SP~UtuPEh*G4Ij-M
ztsdxsB%)^lJ#(2F5+|!1=k<gFDU$UXy8O!Spr!6aEnPi~oXVlsxQM(gmg-AtU&Gxr
zA^Xe?vHBK=KU{ygCyu_07PH1nb>fxQGIST#B6-l8=)}gitAjOVtb<sCi3!7p;Vjyq
z|3<Ff@OyrPYCd`EM!^?IK&8{VZ56i2<#B#Qdu2!zDi<y!=4yh_^6))7pk@lb!lPt@
zn)~}lo`?J1Jif);#&~R%a}-I^s=uu$sL)Rpx!u;rz&3$6WZ|j2Om^6|<498~<n48i
zFZZKe;oZ$m4mg@BGC~m{TQu=?d4FQ(<j-$&WBMh29T0SsEshMxd`%!EK(NI{%PQye
z;NEQRm)!As`0``vXsYq=)H@`h`%PpC7ek)S5Fkp<GTnN@<uiY%Iz31;?YRn~KNybR
zd!;?_D1rdgyFMrlCtm-UCO{eW&i@VZ^ZNyj!$vy6ybkxj8kbLRB?(5D3!007HxEe$
z7S^=aq?Sf3H6^S(EkB!`Ey+OEZ;yk|WZW;&jy)3W(?SV#te+o0HUR9%KNB%VVh}S|
zmaN>ks9%C>GFP3q36Ydc4+w=6?=cg+7?ri}C6c*(LeDc8a4vRp|C{u-*}gsX$GC@j
zpOS>^O5MP?na=H*vwT*qOuCi4KeWmh8tLijN!n|^FDwh+s7~xtHhN(=0ES;0``fm`
z8+B15Ne91=$jDukf=5_|CAQpLZ0Q?FB9Zw4{c{%*j>TJNd*1B;2f9j{e+!kEo^(7B
z=@r2Zee->YkDB<^-ji9!>o90@R-fbPa<bNSm#7)Y1alg*o1~5&^MvWJ()YzpjTE}P
zk1hLr^p#&Gr&Sri?TdQ-Q!O(~Py-S}gNDp^^z+lYW&$?es}Cz_2mG4mc!xGjW|L*K
zJ=Huv-*yyl(@5^8;ZgQ<c{@rH<ILU4$N7)@HdRiLkrD(xnO&aW+QZCGLW%4<k{-0r
zZU&z&-$>3yJO5A^ljq@4pSCtZ@#XkBA!Q3`;3AL*VQk42x+)=+P4X-SQlz5XdY}le
zjX{XY;xO~ia2OZQPp(tu=KIK*E;cu1PFI6-rqDk+f%%Eiu+&|hA(G$+hJ`+&U<w2M
z477Q0B&CcW6ZgG*-xHmnV+7N<<nGtnZ%YTWOS9h>_o~|4V=eO2ref`^d~UCCKT*a2
zK;N}7sAu<rZg&q;)lk#3Hts&|#Nm-LI~xrZP$`yVK_O|mFy5(4dhRMi|0t*2H-{ho
z7vQz=k5sP_c1?)S0-}h_OzGe)K~pJEmpLX7(2)_w_2JbT#^+)J*UqH8ym6WNW@~GO
zTd;reIRwP>74PS$?pUPeHul7D!GEIYi8;;So+blp3((lX*i$En2&}|SDxo`<VaL(}
z+EYFSNE*A9?{gGta1@JYf7Y3Z{EtZq{G6`mW<trmV4dmqAX1(rLM4>8pB+v>Rh_B+
zY33=VZh4{JPy{KeR+mf(KP#S4SPR%?NEAj_tLfy<n|dkcc_TvsxfLuqi{Z~B-OmG&
z2zOK4!lY%fzAy2sGS>t3XX~Mqpadl@XE$gw{_fbR=#MDLdgHmg?td5eb}ol1zkI-i
z;u;TEkzbskzwHF?GH?&Q^mvE$b=$G>+G_{GNd!XOG-vOBZ_@O9Lee?@o|=BHA|Ldc
zG7`V6?M%2#??3ZOPu_c7#qn7TklPazX1gXGPN4(4g$VUk)YkTPb+MLu_1xSr>K1Q|
z%Sfz^o${l{=FQ2_%<S)|<FgW|DdsvxPDS(dkXXln%_)@c{xz;K2t%j^280^r&p~kg
z3^TbVX%0vG?+3mCjaYPko;v$VRDd@3bN}Xr%8Q!sa2f)ty8)x`cUJK?A3qYV7#ZF@
zx^d#nQfX6I72wnR85i=i8lNk<-QN0{B0I;C=HbBICtOHgVNHB)gJ;D?9PXLvg*oBT
zoia~7-}@WCZ4&i$J@S(+7BU!$S=));vETVM&iV*DF9|UZ_|S86DmrXPzFcb~GY)IK
zI^y||uKQ-AS|@h>-i{UDNp^alW1Nb{>T5O8pV!e{cTlylxbYbeShwJp`P}S);ItVn
z)777Cu7fhIgGVaLi<PS+PbBuDt2Ja0g4P~8EV#YJwX1V-KqPmzJ@1M=SY=DqR>s3<
zgl?4?+r0PvWK=J00j;Zx)-&_TG`pA+5JKf5lPFdT)vh+ouDF51W69v|R-qsTUiEvO
zH9P{OYj)v&I}4+2)_fq(Xx|Mvw1gAL=)~tz-Cf4{?)~;~lA2=$kw>$>E=P4^9inYU
z{FYA(dT>t{8%4W2GX%9EWvd4jn(Z>y{<g7$>HL+jRVbk(f?AH_S8#K;PA|3OeD<!3
z%(p2AVKBw){Gt`$^KaKhOgXSIM7ntADIzq6%Mm&~xsvtgXXu1R0KR8xx#SP+{0M~j
zRX~vR`ImRm%!-Z*2wkgN^Zczm|4$x|`C-@#a=0InanpxS`@y%a;{TWq#?=HSoa!1c
zt_IJBMU^R&$Ysu%|AA(fmTc^OlOzEYG02Lce2maIZaOX^Fkn={F$LVYj#f^9!2h0{
z!oVs;^TVTdG;|0u$!5y!8pEEXwG>CD%Ia2?Tep0kQIl_DPKQ@hdiGwA@GQg%6GM6r
z?_{ewlv>to{;f!W%LymC-y<E2&Gg0^z}^kd39qpdtU^{Umv5yduNk@GQy(x9txA9x
zuy(dYW_>wkCaZ2c(rU4!%E(^3ZeR}((zb`P+^_5zx$_&uy1~paFn4sfHY?qpLF4I9
zx&YRIZy2s*ajVzGl$3Ri-nW4LGu98$yKn{FZ^7c9;_BSOR?PSL<r)BV%hV5$!(qbT
zPEAZC3McV@F85)b91yWL9LLSri5eIht|2I8zRlWEDdOdAttmSuh=Kr-<Es+dPG)OC
zFm<m`scvrUWB#hDrap5K>-~d|fQTb4LHg=rX{l0<k&$hXMUuzPmX;x>wdJ=D_8K;u
z`}WacyE!|@pq_WdDaj(-?nD?41M!z3D)42;`Ya%>ucEFk;@Fs=PMuf<+?&zE?NqX>
z!bmK(QyzYut8#UZoQ1GZ&i}M3(VA25aYj5<?*&)6;5s?W_Sf^)^^zIruP$6zR3U6D
ztEubot>T<&|4~knsKT3Y=6cw}0^$#I(ROs2jUrE?`BSG0PP7BOSeYgc=OW4dC7Y=t
zd(e2E%L=NYp&iPRJOXWDu$*WJk4hocq(f~9VC;)VM<KYrO<>j}!dW0JBQP2vMzU>@
zY&$E;0&f#BNig%%$c@l-o<nx=DrQMbrwG$+Uj1W8Ls~pjA3J~WZ@iI5W?7MqjpBO?
zZ3FW%f!T*XAAz}@%f{Y5I&JE`D4_4G2#(qUlCYGZ3kyA8>_r2OQ3reeOoyA<W-sqB
z1)cFT1st;wwZ00m8_K^flCgBgK+Uek%hew=TQ0r+jpy#1Ex6u%Vf&Y27sN*WbKq@e
zA`^B;#9cH#x4*TlUXmeTnNuD7lFo`a<PMTLk@xaR$!7sFbTr|fG7-wd4Gf3~@Pcs&
zCP2>?lsOzpIlH*{)wCRT()bt#PH*;hIlSJ#CeOnU8ZKPkv5M>PmXcT$owIva^7H)l
z=g*(duO7GchbkpBc<yj`?=P$(D3^+YKlFr?ZapDWitA%zbP7u8G$#OkPX7#nLf0|%
zL0!Q43%#>W%Ao$e5sLilA)bJv!^p{h0H*%3de(Alj7=9cE$+R_8m|~>g9xVrf*tl-
zar}Vb+-0$4U~L=<Fm)8!TLNx3Ywsy2`~X~ILUO@2s0>hF;Q)b6%hfm!_(Un-Xn^zh
z;*ytl1iMBmP+d`>PQ~5iK?-O^4|4afXdeRv6fnr*_7`6vsdK$jjKpLt0t3^4Og**o
zwoYU3C*a2{#$3+U8IxD9O{>xEvI;_>-HbFg8MTI?BUjM0oy38OiNUc6dkQ#wW^VE6
z17@(wAndT<WiJU5hu0nC+AqlzpzaX_pWfKApO)y+>bHaR&?YzS%waHHRM{XacGV{#
zTs^M!nP#SyYWCGfS>K>GRY7r?`u|i-T!4t~ai{+?WEmOlSEuzeOLGPD_8y08@9E+x
zA|so@;9WH9AFZgZ<}i!&Lpx<eYsS+<5f$UKcX-WIf}Q<p!fMre09shlQZ+R9TL3_T
z1MBMQe~A49RQ`UgBWXN(`uh4S&CAz}s`R|%1FwaqQK&}meMZ-y)mUio$+Jb5NP4j{
zmV))w=|^)NwL4-y+i9Z}K{t16om515R--p$B;S{E712P(AMmmeXs>^#)uMs<BsRS2
z$oo3)Ao^3x@vIg=8r{I@y49cdo6p)J$AluU8F70{t80a^3#}fJ7$r*T2dObWSZuzn
zr)MsrldDpvZjb-bqjakAr^Kvbe;V=z7yauS19v1`TW$!4btC-{u^&pB<E~z0moOA@
z@N6sALg0tSlH9|y2~Xb%=MvD=K%Q(pY*hUEqgHLF-&~M`6$C_n6fFHP(tDuc9zuLx
z-B`>%uQ=|XcE&pXPL)7Qy^JxjtReLS-P`N0Zh@!5m9(WVT0`-_je%$h8qVR=#>=+o
zqcnD=@Ts&XhfkbA2_+^B`h2F%i#8&3XjQejj<#2{M#x4|QY;rO8X}E`y=)!Ji4hZ<
zW#I{4BHHx_rY0>&6{zkFI$xMyM>UP%*vh&a6dFSda7RlcjG?j#f?zap<oUUHjp0hM
z+4VD-z#UF=8l_NaSt=mA;Oq>In_UP{Zf=fBjo3lStcFr9b>g79defkU2q#xeMMHyX
z@go1bMPA7+IlQn;JIBns7QX_soicF{^5vraFBuE{dfdHBd;I7_5K9R)f=Z+6DwDp)
zfxNpDB90vWlJ<S76yZ7K1OE7FzCO6CR@24J$0sA#<}i288{-?O2!t}Re3|%(sHue!
zcSkXkQENX3M=Z3~54edi>t^L&9Xn@X9$j}cGDoGZAvDVjN?##l!Gy$4L^~&a$8b4C
zbl}Ya<xOIfnH3b|R^k0I=0zwX<K%<_K=?Lpt^vd_c1Jv?-7Pado|KoDf5Oh|tXk&!
z;lCG=ev3_B;!wy+uJ;mR%z+wts-LA;8ljb?&+P!{RJB`fmx#F18d7KG(MzqD<^-3K
zfi-mkwOX#ssWvVyaIQE(Lu!SX&u-Cd-NQyGpPFl`&_SN^mcG98t?iNF6z+2c7>D$^
zbnJtBEZvA8Lh=W)&61j>OTXoczUWJrPu%3sKluLfYpu=Y*rT3l|K{u#GyC1DCbtr-
z5t24cYv|^N1ts0q2cR99n^0(jf5Pew3qK*s?q8Wz`c?c(`r_9=r*?oz4%SLnse)JB
zgs0RAp*IGz0N*dsc$@fDH)U@Z_~N3}MJr>xUkX$EN#N$D#4*L?LR(-$DaN^gtufd&
z(h;>+9pn-p9xe`O{vUFS462a^?2)><-CR8d++7M*is+ACRX@>NvU|@+rD!WIj#S27
zyw66KuV~x$5jEd+i?O8b2flGa0Vl;~?*%~zH<wHYc8>JiU#5Pn`cstZrZ5x#zW^7P
z2j>ytnB1_No&cQ$fMx(U5Q|$1Okfu0Ge24IV?h+9M9c2oUk5<oVIuG7Z;6E)w7Pnz
zH@_=tPonZ~j`{jR*`TAHTnZpUz<d!#^`Ds&9G^J;-s&2vMj8t$oSF6OAz}*X$7GA0
z9khDS1e|?zDT?1|l-GLMUY}s*dhmCQVsJ3_hyJJWfYVx(?`mQCIl}ekz9Kh31<?i_
zcki<lthw#84BKR*OfR9aYFl<SlKn$%6BM)0;BFOzev*>&yhFp-vB-VwKgaySsl0At
zS(<yhF0K=~JHm_r5cm(MML^Yt#6X}6oVwI%YJ(#5&Qx%z7sLSf$q3jB3pW(X%-Tm5
zCF^H=kE0qJMb&uJfI}!p<2<fwQ(3d{OYkfAqomi}&!VPN2h_xklh)tWL_UjO{n-Oj
z>;~QtUia#q!gt>*31n?uJ~j5-{BrPh%6g;v=Sx=xVwGB!>LYc{9AB#Tz&7Q;-n9Oh
zORZ3!AR|$<-u=hRxFQlu^}QG+Q0VyATxI*>kBxy_9GyQj!I3RzYdq%QRME+7zQiN-
zlOuUA?>EMU>cVD0E9zrg_QX4*E-=`AZVLzvi2NGCvfqZoQ@Y0c&rT|HK49c!RfLBm
zk7ny_y~LgEc98(8j+*-VhJbDH>CdRMQ?uI_rMi)$fj8tWhUCbTDl!doCoP`zf26A6
z)U;W4Z;b<FzAhv*(;dzP?NU%u{KZ`s*s92_>Ce0@-9(^)%V8EJRDFy&ykQ-%UH7l(
z#vibur6m@n&_qNhXcQZTo}r*iQhg?m7N;=ej;mAr2lSK9zfqrEnC~^fYcC}w=$Hw5
zU}STD?CZBrF1ZQ?6=ym{-vI?dC}NDu8Zu`Q1cMKz3wU7rUIVXEKO^5tgpsA7+YSVT
zX^KijAbLK(;U30FW&?uz8tTa_2^hHI@fLFWqg25hCIR5MnaM!?LU;X-qj&s{0ichN
zC2#9F<2^D$JXMAnAOh0vN&Dv49AIC@??bSZ+AyTeP%tvY+r5y?X|EVkVhVl<QV2L+
zjSN83w|DCSQG;&;HS%&^+YI;JXrwZ%x(!vxgbkG`3d^*AL_-FCzDQXgx%m1Ps7j-P
zeFsN_`MoDQ@><rTrXRq6BABi*-scy3!AU@XWf{|AN6G_)h@4>&ybomgJ~>_}S)=+j
z!lpF@hr?Mw6@L+%g?tW~Nds+AITq9C8yHz_tkI^TE&t4f{^1xT{yZ6YwJPbg{YDwE
zzqa5yHQm_b`lkDem-jt<W~{t%iJi<H#NFjL{{)@`pf!ckM=8%<;10_|OT0IVCFU-_
zQ_0v~%@Z=hfUCnuU^;p+J(6qdy8#UvMwY*J`FG~quO4EQZ2!%a;rmn-e$vkDQ+MvC
zVy;W_j)3~`PL1K~fYEy*iyA4Ay16Uw<-l~;0Cc_Gq$4mc{F-uEZqT422)M!K6LsJM
zOwV^6MY6qEwLB~Ir&iw72>WZF1Eu{f5b#$6q%w?SyzX~>{LimNI2o7#SnxrH-Vb0U
zRNBuODFHHnjBDiv(`?+}ljM)vbHCnnj-xuJIGv@Qw|l?q&g&&!kCJ31ekCaQ7iV89
z0cHdr9E?2zSc?&WF8-s6CZ1VdeAV>FMHCpB*35%};7Wx@>3Z8E0{_VgS1w;bM>j5N
zHG$qr38}C9-5r_z15yy|(;qr4$;)5nHX~WEEXhvR_d$Z8&Z`bq*i9;6tiaviyCd4V
zghv|dQIao-ockv}hw)XEl`madY0YmP9vNW+j>>h6E}cI?tWu38oxSKB{|)6C<RP~3
zx%uaKcJh6S*Cwld?R)fgy{LU;hDiz8+t1u`Nm5x>sOUWog5y)1V3Mb+<pYn1)Q@Fm
zO>suW29Y&If6e9OTD8!tp^-zQqZ1j$XdnJ=>lDLQQziD(?ctt{Xl+2deLv+vO=W`(
zOwZO6u|^!aI)ZU@a+NVJo3OCApdBs8-be-`eFEf%y&*G;@+6Ua3lH*rF|o`oCGOJk
z+nbZ}bs5S@qr~psWjYp`%H5#ukr9f#9bpt^98*IaBGk6La<dueveY&_JzXay&IyMZ
ztCWz%`vGG+<B1<B>SuLRQzy}W3o;@23@NXA2&7hjS`}QsT617|l~R|uaiwcxG=tE?
zZt)0W>@U~Wg}i%1YM;_4AaxnCoO0ZfO`sBROru}l7urcj={&S~Tql-*ICPAPkh&EC
zTLX{%Ie@lU`{ICWJG*;(<C4&tyoNM62?`*xW6UnDN+W<uIba#k&{G2mQWAFWL1ZQ6
z)qbs?M6JFlKa7)GgUOffX%(Yh&Y7r}>c(=#{AD^BGpzQWvD&XtDzOv{Nq>}1j7X#x
zbNPe$0D|`cN)t#0J2Aols5ai-`HG0lhM;Kbu3xi`ZILuo@-Q1FQIfla-9p-`gb1Nw
zKL&0mPSl(r<2_=QXsc$d8#r#;Yj?a{*v)OGX!9~)58K|6uzW!sioi4Ek-i1E&8rQb
zsIJ*xy6W2uoAKNrB4~pL(?h^>jQ>slEmn#kq&HllAhGeq>$`DxEr>cGD@rnF6nFJa
zjV(786ui8>0dnc0(c`Qvc1t;RI$(pRy+3!}o0m(a8T*F5!%nU+X}r^1-)R1emMdPw
z^&-Q!L#kzJ5qd_Bk5Sh1?SNEor4}VK*fl4fs{E<E={o4>kmw4t9F#5F9=Ua4zA`of
zH1WBYj9`j8oj08<hTU9oZ&VMxW&G$_N>>5jdDbvtZC2#|CBP)-YNSJ^cQJbp39LS+
z>)MIZ6CeOy)Dj8?aO*JiJKJ5MbqM(i+KzB=RGmC1+RHpRI3M`8UCo*3>V7_73B>_)
z8q}|**G{A$coJDYB}iyl7@bzTXlQ8Ifiscz!7=o$M$x0npMxp>QwYicW6D|X^0|qr
z&+7XYL#fbFb*r~o?4xw}d3|?Y{G}8$L{wg{ncsHQJqPjtO%!GA#76BDfHe{vF}WkH
zO(GH@s2vv)bu&h_j1ym2E}F#{qIa1&w6mD&B!2*|86-9lj#qdeUjcLqAg3s-FR*I<
z6r<-V1OQy0pipKqw}nSUk&?%L_<IYjq{=OZp@7Zj=;+zj^eABQkdh9#H5tj3R4PXy
z0B*ItUg_Y_GoT!fRs}9)y;$xxk9rrj<qI%|WhU*P#=q29LKrL0%~NbzM=V@F#*nKz
z(wz{Nv5B|C^889w_78|zp60p17C8`F{eZ$?eWcXL4t>xoaQ%QZws<mmqrbr7b$k6Y
zZK9~;<)#588(ioIq)g&@pk!q*XeRKOcx^`Fl^BNyWBiB4<1W(Z^p6usK*UqoU|%@)
z^OvXmElrCq!WvOM5gH*$%)Djd8kVRZnDeq{x)8Z<ABF&2&A>1RJy)w4g-+ZL+%JC;
zf5AI6T3J`sx3)%t7^5~a$`)~uMvqZ@cmRNcLa&vi!Q;lk?TIwBQ?f;W+|$DXhbqT}
zSFXmR@5>~S0s*U}S7S`qEN%M$fZj1OB5mhZpa$yWD!kv7s88P|_<gUjgumQ<dz_Xa
z`4F<^k$-y+jNHie4hH6QY)NS;pjP`+Dd!_SE2L%LrkS*wqC8(D`)_3^Xjl7^xN~lf
zE>|NGNc@714%%Sa_MZ~|+17Or<+=A_xi-k7B1_DQnT)MC{P`a{k%o%}tk)y9-K6(K
zL_}PLQE@Nc=&DjjP-|J%l-AmakQ_iMQLw1BjI1qy5fr3&_?RX+{9GKY<UT1=Rjs<r
z7Kedn3X=KMoPUhWrraI3YVPhkC-;k#0!%uwI~M^TFcQOxxr)^)G{E%*@<)#z1DY%#
z4<%lscmx%+{rIqjaQS)(2TV}Fuo8i$LRi3HnSjr5_436x2nPwGhqq^rS$zEBycp#;
zH0lXT2<_oO;Pp}Qgr|B<{BHDcTxbZI|CP!R*?v9Z{YRqb#G(w44~>19{-HO@`Ws>G
zJ_ciOS#0VY2_<sEI6v&8Hbq3B8+daiZJl{0>Rn5)T2>+e$)9NIn$JuOnOR07U9U5i
zZ(2zw9bYV69BM*Zyz@s}PM;n6$S-2g<Cm|<0Z*Hhw9hR*SVWsxgpTc{quONif1*|A
zB(IahgDCy*$$$iAibJXNM<mpatcBw3m3@!fubvj@z0$nXo_SV=0F}}JV*_wAueMf)
zf-(UD1y!}d;6O8Hvjz&cJWV8|Ota*jJI4{f<Cg9Nf1zUM_-`42z+!yuE)mhCPnor^
zFD=U#N!#-CpalKpzG82ndmo8xoaMp*{+IPHS1FKlP*r=9Xb%2n*=uK)z7MqDe&9jK
zT#Zp`sdB1qL<HRua}yx6Iv~V9P(X;vN~`y>6_l~Ev2g(SwN9U&T1P<4P-n2*5TXt@
z2zgih7R;U@m=G|xkzlES(Bf3~ipXzw`-CZVwxoT?(`+mXiaRigP8F)uCXkIX!osA|
z6%n6e?!jp0Rlbk)0DZ=^+Y<^`WQz@+6I3_us>$+H?ChB9m<j-tyuQXte`EmZ0LkB>
z+L_r|+1@PRv9E?YKWd6~!Z@*nvB%t1yp%d!2n7BUot-zszL*xnE>BIqaouHyvppdK
zNDT%@gy3(H(#kQ2_TfZ+pdIFmR05=GQ&^FkfRIJ%&s`LZIG33F|7K_CGQt#_m>0S;
zH9IT5e$^-5w0JM*AaoynDV-m1CDeXmbeoE5JJ={m*C=D9sgEzn=Z@Nl{iS35%gOt5
zQ$E1BI!XBf!Q0P<b^W`Bb?EB2GTW63#TVw;C`!?R7V6*MAN}*37BJ3o5G!Kw?V^SZ
zILOWAmA<l+gh|XScH&B}ZD``p{vy2X>M>f@Cw`$uVy;@)9r?=5?gVM;pte|TA*Ns_
zPL*+5=-eE7bNCW)-cHzwG<sqg+c*KCDRgINOJAm47DPd9hzNgGJ)7tAFrP5*-T4Qe
z1UXGPS6!5@?-}#A&8gYt#cojNMI$-Ci60#Bk8l7WDYcC64qumGj|agxUi)5UTP3j2
zdAlug6!R4CIy-)Y+{LSRS+aI@O^1-3@9yMV*?V?hT<}<Xam35Ojt&|`PLJI=EK&Qo
zXfzkrz?K0#_xj<s_x#V@GEEe%eA0_!lq!%y9Ir^OTo)i2;!@u`W<GRfoNf-~fU_?7
zpviB5lu1zox!SaXI2K~Ioxd~KlwA7<g}mGcQVDJ})V<Vls_&enJdEb?!-a$vT|%?A
zgk@xqrOnx(Qa6Uk+9}#NEnv!-a1fc9ugsF;CQzjJyJouxmG&G(daw_FNBT?xf0nV?
zi~Iv1%s);el`Snp!`Wg$?_ewy=}HIE5bZ=b7~|Yb1d59$RSKv=nhqfn{}Asw+#pqh
zsS;H#CagvYa>QlQ!D)hTa*Sk@1Q`$4Vg(KdIGL)v8=`l7OG--q<nFu?S>dhka7SRF
z361CcGB-=!m?xp}5<$%>_0N$I_}TVo(C=c`79r|UAIMrtg3M}ZeNg|+4TahusYsS|
zcjregasA8e+e4XyQzIE0?4FdV|H%tepX=(Y=3w0Z8SeIG*#m<}xYI1D`*dr3`k!=C
zOr!S}pq9Pl6`_wqvZ#hCQacvE_~lOlr<vSgxGB0#S++e~AT)&NVjX|0`JCcv6J*vT
zgRx@w2SCrom@4ao&0naFD>%-~E+J-J1WDw+b%NqN$^Jni(x!AMDExs=&d%upfisi6
z${OaF1Hs`D#}`tk6cmThfur9fdS(IPz|FaN^xM2->rhmYQ?=fLG0d$3***N)82P@a
z{S`<NILiE=B?M3}HrDF2U8D(M5lq)nTrr3AFV?s?KuQ4dRiujrQQ)^-rH#6|Tti#C
z@d>Y`U4!;KrLPWN0YHyT;3b~~M!d-*)=TMwg6S@B(ouTYV1I3m^(0FZLJ^f#MbxGA
z^{L(5TGD<~gD2qD7$an#CQsFk0iKB$`?D%TaoQ(@*DQS>w+#8psX`9T!xFs|tOOyJ
zK`wdS-E34Lt*s0ft$Dd;s-0Y9V@Vt!V9{19p8al$dpN(40jMlMMy2vJfbcCntyX7k
zChLm`-D`(h2_m9MVtb*Q&E_yTdkmXpAzrft)sy(nFG#!>%LrYp>@&dXL)6MvGY09i
zfPjSQ=;%BX0C|255CY~^k25eZ*c=vX>oB`LF{5G^Ep=@>G~hR#sW5W*CE{uAA3)lm
z^W9I>_}tqe)|De}l3c7Tu)n_T1bMaTdh^iqVrXab-vm9?FKY;t6+{7|y!bpBV;tPl
zP^IeV;<C07GvT<G41VlYT2!z%)_r-kjs&K+ALwSrdY9!d-VNk(Lg09Pnfj5XCF#)4
zjMi301T2aw8Th-neyO#Al2liJPNM|+P&e$UB9E@_$e5joZ$;z4=?G&UYGt93w8X4I
z4xy_ASt^1*eYnA2_6urJ<@MGk*AZ3r9F=P&0-Vtvrv-Fz|7Al0FC@A<!#ntt;+bxZ
z1kmJ-fQ2RUl4zk37YO9$<v}P)Ufa`DHdVz<$4J(@xX(yhTll7Jc+mHr_o|zU8YEP^
z+0W6e*gNix$nX9KX#rnzyV^P9X0OnoeL<s`*Q4rjI!-1opm_8d1b!kUMuy)p5|yTX
z)}~Z!pYYFaIxs>>3cz6?^=%`kr*(_>Ga=sUn6rYKY6WDH9Rnf>RB(^O;>VOp$RUV<
zdpz)M8QOh&?Wvbn9S}(878bfCVGL~_aRaX^tI)(}I*P$6*j)rc<4%)QRkFc?RdCz*
zPOkW*&kdpu{Iwgt;PSq?IdLE@ao8qwanYn^`ZxhjHO?sT0efl!Ms{X!u@#>Nn8>{C
zB|_f`?ZfWE8Bu8s{5_Rbl7%5c7~Sk*Ju_>2`*5b~Ga`sKfLPuOVP5MqGhw4(V_azO
zo=P`GESBc}oB#l>*sC+i<&A01{4btEV_FG*&Iw{|XHS7`$GIC&@4l$tUs#(iH6kNo
zgQD0-<O(VW5rcz+0jp|eK;@BMKh2#FQ+oB^AzwAufX_g_>p}bN{;mIi0*@nLEgjpz
zSdsDa^14^Fz77tiw_hn*K6eZ{tPf5Q2*#xQez+7hqTG96MY(q>l|I$TZ+8A$;2?ao
z^^-<|K-M4S%}{x;B2bXsuP;3N&_b$?jLw)lx}*WvO+K~MR7(-Mr?$Dzj8TQA<$zsg
z$!~|Vq0DxeoEeBfh_+HF)vZp+6SQ$@1Llnfxi0D9y$V#x%w&fFnI}n{mEfD27oOEU
zu-(p?{M)Oo|2B?>;9Kdt{}E_y`_+xUqb_!IL|ph9B?E!&2CvV6II4NqYX%(S!q%|Z
zQo*AAA;vX0yRc9}8rLXxzS5052XF|t^WQqtqGdDWc*y>u$~YkznS_ptrmC12z-GD7
z|2gF?j`5lHw3g`f-$ft(IW9}u)ux%(W_D1N68)zbXX!MG4A&7YoQXr3JUw9Bwh;uw
z+<acr*e~b{GsZt5vI4Gi59`LB_xhfC<3QvQT-=SmIMKYVKX#(TO0OJo^-Zef-tzm(
z6F*Ry`8z<))Va~Z^1|jQV7I%o6W;-YhVf*Y>93D0`Zu&izWmx9HTn1pn%EfmN>ybM
z2aJ;8g-=f@{2}mH)jFJU|0TnpGY;FNX!z$zdNZwX*k%Xe{Fjxu6{@vtojt>Kn-yAT
zzqA-JWtf^|7}_YmklwC!a3?aaS^N|X0K%u!Y-?P=p1HJo)vGr7FFu+3_c{Uh;(>QX
zGpSbpkLCJ?{FkwxlhyRhUN)HKZ6S9fdrcDiKtWdk0Xi}o1gMp&>Z%mWQT26JFLqX$
z>15XJL^_z?(0%LA;=DdF3rx4EzTd?^#m~pT;zv;#BlLSkj4m!Tu@u86kt@12nqU9W
zjDb2KG7b*68|G8Dtsq8dMu)xElId0XsoWnxGi!AfBDhk4+PbEcHV4j0&wIPLUrsxJ
zgM6FEtoW^{4%mrU-+n>rwH<kGHt`u^ME_k8?ziCHPJa5f=u!ntHN9%1l~qk}Hoiw`
zfuHi;j2CJ5A(Ng{H!D#xY2&17gyWtwS0uObqF^Qe=#mQ!n(s4{V0K}kdl0!52)C98
z%$NyrZ)_}Tu~q8P*5d<cf;q^>Uf*?*AI^3t6^ZDK`p-i$^ImqFI4)Xzmq8F1Qe<6U
z7VIW(wff_zyBv_I+CV%8aYE}N1CVbZXSc@nZR`LOhjN^-oBT_ubbLV<KDEoo$M<a*
zWK|Mj@9P_G<jv&vX9tE`VoM$W2plo;sAx}61X4p^eq_rXdL-mL`E%=HlG%iN*bSn{
z?aaV0Q1|x(Ch2uT52M`wI6Ch~s{b#Je@J%rikm&N_qqt#B4lUp$d-HU%(An$GP7mx
zo4wtTD|?f9adB<mci$iS19ac7^Lm}<c|ISf2e>Xufx!s*U<3g*T#zXKa;&1}*f1Ei
z<H*+HcW&n#E4}e-LWnBQ;_aq}#TEXV`+v$zMLs^Kr`$EK^RowsawMKUGOUq>U+%+z
zm0J~2(T_~3r1CbVcD)nhZJnT>`2)8`Z?700fPhAAQ(LSM{zh9vI@MJpAUJfLEWm)^
z5c?D<kttg2>fsjHPZ=-9aC~Gcr%-s68hpWjby{<ezjCh~+*2G;AyFOn>4~$37hub)
zpkeM=V+wL6VhUVH#b4MBvm865o4s20g@qwjCFqqo<FOUNgT)3F^$iU}fD;nv7VRrA
z?l7zb7zq#7F1jp6#X2BeeO_kQcUJa6md|__npVEgLXS@E!V^J944zk#2?B)20sEzb
z=|Y}qD;lyHPxTJH_NSB9No~u#7tEv%iBt}O)d~ZXmM>~nO3iCf0K0=WoA}7fW}nKu
zYT(lGg?CX#{ywYl+urUq*@gCnjw>KvuqO-HI!S@)oEAl9WVX1Q`_9EPdKsWw+Mm6;
z3qE&q^4%sIzp%}Hv)x)dd!V*?v?Se9JsM+ZN*JZs!&xU*q6VWJF4Ct9+n<jR9rRG0
zz^~u2^$$I9>~n|Mq7r<L_a!a}=O-P*H&plX5|H?NsAs^!kH!zl_EpNw(>*>_R0BR(
z#7E04(ac~t8-h00R)U^fi7AgFHWDzaN$E(G7}o&AP#`k2ay5Z=zmuxouWN7D@O91W
zhNx0-T@Vw|;9&THR@+>g)<XpyQzbstTmc7@5#;3#3HkK_?FYhXFnvNqtXN<)c{nlY
z>f_kBz{Ag9*Vx(5f7@etnGL%l$nqiSabH_kGCBNBZ8WtUqyER{V7eTc4hL%$3IkL5
zll0&dhqvt)q#h#nR<3z#I{6XB`r(OcDQX?NSORRm1N8pi*E7~uhq{n2HCRCDF@_eZ
zTEbZqyG}Qb)jt{ngM;1D1#lsk?lC6g8;%mzs5P<pjk;R*`2E?Mwbu8_HmPT7Y81{X
zQvkmponLP7V#2%*@0EuLEh8{B>$O5Ax}rf#M_?=9Erty`LWZTO>evK%+&P^zYQM5S
z8Tp-?u+w(t7xkM?Hbz6^Z3pBt@H^JFUWHWU@?LyE3(o&?`Tb?8cvjBZwM)aHYjm=9
zu*!6{6}qpruH6JjAgILM7FY9cFOYQ&4WSzl$})J`w2SiS$OUotfGap&V1QmYS^vXu
z7r>_o03a1?Rum*EfRa-8?PvK_Dg|Mr&7Np`uM3b!s&15_Nx%do5@&IH$L#2(_s;25
zW`kB>qQohtcmQ(ISG^5+`P|;g<74k~$g*W(3)N?gc0EJf{pnJ3^e(#U*#ZQ_);@hL
zve($Bir^z}&hmL%s#1CA=~}bmVD&A+yV9Z{in#fg^6|WL&6<QP;BO*I=d;+fxT^^6
zXLy(tZ}r$`C+nTzhEL1w)n6T3m3(Zt_*m#N>Q}nZ<MVZefic_`I75Lr#Wn#er8Cun
zH1vQCR0XzT<Kgj<^Uc4|#P)3PUB3HT5B8E)_KpRF67Ke<sdpnp1^465KbS#LZBR`B
zEZ$wFK>&d36r}+2PLVrqK}cgGrIZ)%(c+wy&-)1F3Lw@z<>*TiZ-rV54F6SG$$N4e
z1GfRSSNQyqTSW$)`SOhU3Ml_*`w-vQ_yoakwwix`k3@|>oIDIK2H^0Dnr1h*UiWx;
zx540B8q}4MSWzau4(NMcYgRyY@}O8nyq((hks3lLT;C4o2PEg_@)kYv@_PRPeKD#2
zEus$v_Uokg^73-5N>m`jLL7ZJM|*%i>q<%Da}H4Z0PdZ+71$Ri(`SR-C{V!4Mg)*P
zTsm5Bq0c>H9#<w;`LpCa1t*@@N{>6>&Sdw!s<^&NlFI%ddrTMw=%HQx7YHh#wHaK5
z1t;9y1r|tX#ZEHeGq+WV(K+w@L7wk25WS!U=G7G}K(C=p7w-LihVFOar>ZL6qLduz
z=17Aq*li5VN&>XjN(8%{;b$rCrcpZDucZC2;`?bbC4CAaHjHuM(Bk)LQP47Y_AxoI
zrTOb5u66NAAhnIidE0w_<`ezYfgg1Znw8AJvk})P_TYDh#a{s;7I$z~P~37iUyOzq
zNK%oA_~H|bQm26;#w6!Yn$KZjBEv<$ehg|>Rr0L2h)Q`q16o7F{NOJtrVFtuHld80
ze;>4$^2Lll6rpoh;Y)BoG@XX>%cIOcUtVBXetXT$1#CQeSL8aLAyVi{M^}FZLKVmj
z{4r{_6<gwc$Es8Caz;72@f$wpfj=`^hQ-F>$nS%1K6TL1xGF_7R&;@g*VyZ)J}m14
ztvi<<!M9g#8+b<YNky6i<VCuF%KfmE0P?Asq$zHdsFMjPR?>%xUGO~^L)nI_tQ^Ef
zxdWFO_+c~18J#1(a?REml<&0zM-BID|Cw+3EE-u=^?pLuVC_bw^RR1`_X^xKU-L8n
zXr=o;1iccRP+?fZ^lKBUMWF-~?Re8i-=kWpOvkp7R#p9PvH%;H6#r0+wX29TNyGdz
zDg%QZ(cPN(W($t}tC=ZVR~&u)4iq9g;P@Wrjh&LQQx7Lo4Ze3;!G%n0#9g2Y`BUw|
zzf)r|I0Qvd?4I?jq!ZilQyK7<)vq&Zzar!Au<`fh%r;<JptcOHpOR_IdVRDnMPI?=
zE=4UKo+R+U&pw?E2$ssM`*9Kiyo$c;NFt2d=w$Nz_SE>mXhLa!|BkBxx=p8#Nbc|y
zzRIt^&u(O5BT+l_y(r9*X*_DV`wZ&5;Jmft5N)}u|H3ynhrN85CmtR^7ABOY`m}n!
z#<Ko2Hg97yC@zCS5R}FSAwJMLJN<&VJpM?9mZb_YTDIJWY|5lz+fYH;80gDk*3^yJ
zEvOb~(8&2+0+SB)GYt9FFKpx~I0c?z!9tmgcH|*i&5wFsxhk~KEjjk}eNoZbOo`B|
z!)J14;A1d*6S6_J_gvZ8VMj2xh;V?*4R$VOMMJL4r0krQ@T_p*kSvnPxnd9qzRt;Q
zH~>)r>2~Qi#Y*KWcnu1w1BHS*Ab>iPsYw?8+L4#jbJSZ3$mcaoNI*{|2DN#-?icmI
z54QvXm-nU}C)duZrXKeP&`^OaUkQ4uCjm#~o^Srj(lz99ob35mq$sF%H;(3?#ux^V
zt?Q<5R5<6z+|X=HrdV{`)27RIeo9Qb+27|Z2B@0db;vbn^t6Mv*I{ykU{g;9rWH>{
z4)IE0*uybLaH!$Je{Ug2n5A&pZgJSjvk{@4AEo(4<@j~rt@S9R$J*77q|dd46$>*z
z_w9Fi5^<wVo?Z6@bwjK;K6U(9pKza2_IlQ7uEMntPQpLZG#gy`{pR^=kiJY|2gp(D
z4rC{{SrQDk6v`B!ZSAj`b(5waaE~SV&Auo=8=o0l%EQMel&d8;a7AEm9J-?cAZ7+n
zPsu6<4LCJMiGwd4a^xOZ;j(D`Tb76=7W6hQ|6kbo33uQZ_4@YC{y~PomlbpCJrO&t
z-n(RUbaVhpbRZc>wE#Vp%dw?RSiw|C@xreAdLL~R4o=31`{^cLTP`2qd)HP=;-cm^
zw*_ZfS6Tb}6BHMPk=(pLT7r!RUrq;)Yw&mW*d)k>F^r8xKa}sIwwQf`mcA8ztrW?o
z%YJEPM9@Wl)WeW<D>cggkx9+V<T*B${#J)ngs7ugT}#W)qmSetxr+ZR&gb~vfr~j~
zqIYfA+rlCS8cbBb?OF4Eeb<kh^HYs(*fy)6s@pKBg>=#d%X0CSwfJl0^J=qpc^c)n
zoqjQlxX0jp*U~{tftP-+0Sj%8KAfRdHHS6K0!I6b1gfuff9(%6ERg1%Cgv5CuTCs2
z{4+>`=#L5my7%UVf*2ud+;Fe`0k-QqS?)mueN$Kba-&84uLqwRxg6>J>FQ|#zfL}m
z`24Rc`#(DC<qPM4y9E)?;45Z$)SRg`aNy$RvHp8kWqWMbR#S@pi`GC34{DAfGL@X;
zcms0ZK%Xmf2trf^U(UKp(r1B9ZD{C^ydLh&ltoK*Rr`;QWb5}l!1R*jLCzT^TnP1%
zI#yiF(4QQmy1-v(u;P%;_YE`Q7lGjX>rJ-l;S}?l-XFz}4Ghwa-oCA*Tz(p4y!Lnx
zrapTl&b=p#ENenN@Zt|<2JMxVwn8_Q$G6~u20!ZS6&F8A`W}IR?eLGBkHd>dn`q*3
zXx<OODN^i)t%E;q^(*B;4!~9<zg!!S1Nz6E(;Z)u#*7Y<o-!fPlddN_;~EcJJE#29
zOE<a^X2$xL!eM{lAQbWQxBE+FhN6gdikz;f)8dU)HAy%;IqwB^d{Jd%w`Ey58Pj7b
zmAyv9E0eAo&Jj=ZgH*eS$w=4I&wanyhb)V`2`HXC)z!z|<yN;jE-L21k9=fE_-=!<
z*2Ec8u4E2(PFf<ze<rnn@i?_!;wQAa*eQNTv2@l-h_BWpx%jaH<n~SuFuL;bdOqCC
z5Kc_ldD+B--2wZ#58d4NFP;F_&fRmVlw$iN*2^wEd8IGfSw)<?lP=02Q2f(ZWb=pE
z#wLjlSpxX2{%*NJuvfovCA}^LrE+C`UCnj?x2~FwBTriQ8X&t_*x7B-aA<?Ebx#bD
zfr}Gxlh>;8Q@5jan|Ekg03Y$>)OKijz^n=3R^QMul9|ly;>KTJ-#iEmcHm%?NVBW?
z7T{sD@i}L$f!$*lo}>undZH1p@?1n)0w3tzZcjyNtE3dPy0639K?g^k9kW{(siPSM
zKi@poH@2<|=Vs^_Jzf>jsZa`b%CxEA_=3G|kZ&u*3>PY>jl+2})%$lUklooB<{zac
zZSTGcN#23y?%ZEv|L1$OxYZrzQ^!*}=kibU1_mruCWLR2fF(!>T-(mtY}&feCA0Lz
z^;>cdTxYzF#BR%<{6QzrH3rjg3#lyb_GTqLPC-@_ccR*}l@%z98fo#cYFV`Z`6w`h
zT)8@g^)|!qo(V((=o>M5@Yf;F5;fi4lr*QL+S-PpI~?L;%ml_Wqrd1+1)sD^dx2`v
zx8Mc?ycT?UJ021=g^hNBfyo~@N@m}aM#^$m48ES)aCgg(yYU4+R*&#c+YYxCSIzl=
zuIWcp^<#eALR~i8y!&dxc%O&UUf^}Hs5&I2_1kaCkNhWKC1ZoLHZ!s2i?<XsTA#uw
z8>D&5>%Q`M7D!ayNk364lcXX1f>|>6`*R<d_%l=Dgc#Q`CDf|yyJ|AU3Uvc5Q(HMr
z0{dpf+F~o}>?Y;_d01cH-aBg!&D<E&K3;k_GXv~cxPke?4sf6bj4$8cgOt8|M+g!}
znB{(m;ymUm4`qc5q*@B(e0{%_B)U$LO;st^(tGT#Y3!7ip5E%eLk|nMbLcp}CFqO$
z<dq;ZK&}lAJKON<yOV;P*T~AY_Yq+9WI==Kt87-y15O?*XrD6NA80!@vZSR^mz}!Y
z^BGHRjT9$b0wX=hT0}bJDtl!>o#C@>umg^1s%Cwew07yw-U11qBd~!{t*aUN>EEjW
zdalFzsKCupU;+s=?{1!3Ba}eQJ=LR{*ePobI;9gFYY)!zJjl>nbaO)%RI!bF1!zn5
z-j!!f%NIykN`Q1m$}h-Mae4HOe;s~+9^7sm92_wfB_Jd6+tVH%CYXY@Dp)w%6x&RA
z+m^&u?k}1J^nuqHe_egOo|~#JARb1AVc30>$9)qg1%3W?dRK^+7?^lrOD`;d&wuMt
zC=JXwr~f^_4!EPU*ts~&UWECE6+AaKmX1sDnHsI(t)pLJyla-~dSg|}9Vcop`zd~V
zM&?#q*C4CFo!%Xxpudp9xQH|@#tmR#v%9OY42Jv2Y2H}^p%`xqytTz&F-bqsCmr*)
zE#AY_T?`WRDM-nQ7~b6D7tUa9olQwLqKRj8aTN3l(ep1v8Tp0MeJZe^I$v11a<v|<
z_!_Y4w%`=;sK<Wf@xAR%O|^?G<Z{XaMvB4t_7%e$333zv{&D2LeQDGJ3rYY$2(ZC2
zz0&Q7@W!L-r_S4v)m=!Pw8tKcJt$UVog}-6<d$@$_dz@v>|Ato`xJJ{!A_rpU~mX{
z>!1}yQwV4z59D;&%@RSLpC9fS?h?FjDiXki`OR4L9?6iR@!;KQZ77h@)4^|Q629(a
zU$z_)8PC)#iRG|KIW-Ed1>lAj5{5pkI5;fPy3_B)hEdT8a<`u%`hjA$V(sge|NZF~
zcD<e<`4a+<Y&vSWqA}gt*ISat?<I9?@uqD<8TwzwVQb5xK!aL2-)2mUmMpr{iK=U2
z<^5c31UKgO?Iy0g9R9dsRo2KcX2$O}WFH_IbW_+?ZFn`=$pH^(ycnDO-;Epko0~^i
z_;6~G4Q%7QLT>Tpr5#&W>XSdMUbh#%UERRxk54-O)5-z4Epsq7mV8AbCa6HqRTT8Z
z*;bYzVB5z^ZpR`1b}`+EK@O=Rho0R(AGRh{myC(w*@}|8#SnLUTU!2{iaLE5mD3p(
z2%PYx!#2H!$VtbZ8-{?Fc{{+Q3Lbn<P%=#j%$IEr)^G_ft$r%NRP^m)2e>wd!-pM5
zeiSPi7#c32KHsG(-MQ@_r^Hlo`>3U+WVB^OMIk5@)Wb2ur!Au7KOwfjo;vE1;%PJb
zA5SUk6T2FQ5^ds;(C}|6G+3IYqJAML%s7PvPW+*|SZ4k`LR(x#H5$mK-5R5fGbh2T
zHRo_qNZz`8%JFV7H)N#Kw|TakMzLnLc=A(*R^UXAWbeaO{=MEmE?ewf*rO;_L}ttU
z*tWOoG**Ll;WKFwMbmP1Mx`MLn2A#a@utBqFfC6aMmHK6aDx;A#mzz7(-UL^5?tGd
zBOLNouGSRki~j4bDent@pYD1;qu3To5-)eNoLys69xCjhHEm|ZI?4?I$ZVE__IZhj
zh-tLnex$f0)u`Y&xg{yo7-0s&qWM7kkg;PP@a6lQQo1A{NOkW;vu0WY%UM1^t@`%L
zl8p%M%<Evyg9@;iH_?pmz<wq9X0viP8HkzTRiOEhz5n!~LMe&eyZsrknD_*&23ske
z-51Q%G<r)2I4~5|)bzI@_|)3=PxSR$SZT~A9UJ~Ov>zN2{i?$RrOU))0wl$qiQLk;
zzwQ^B3FdAXRT`(>(F~F2cO)>+35zolm3#CoeZC4?&}K1g{XW?Ku1-Qk+3>Ncsp;Fc
zGp^a{6-%jCnOa`-(!TLE_lQxj5e$x)-~(vJ9Q>zkAC8CXawgresDQ`Y&e`1bVw*?u
z`1lh8LjxldsR3Y%t}u33U?pY1hAFIf{zMPg-k2-@P5@>sh_<_gSZrTDA&gcs!)hTK
za4=0$rW+YivQ=Bd!Z4w$eB~&jz%2K@zTkZ>H-T=M<nR8R`(sXdedjP8Ew-2trzEm|
zj45&o6U8m|$u)P)w6v7gk`Yl=`8;YL$7Ur{4bY{dX+cVil26GOJ|f<D7Od6{90LsF
zEV3vjb*?Ormw9X{#WAdSm<i;A+O@#~mIAco{jVnSqzC_9sHqz1Cn=24f$F&)`v{-N
zUd6U0BDJJs_NUk}6xdHy=W~gmdBV9xtWFA`D4MU|+mk{wFBJvAG^kw;>Z|?}1{8oE
zA?jVra0X;g0Z{Fa#^H8kmqt5zL4}g-Qq_$Rd{<}XTO7`QkyGfKP-U9BcJaP-VHy_!
zHK9{o=^zq3tJP^ou}rN&)-taxbWs8M^<waG$cZCBneQ016p<$$gpB0QS;GcT8bx1z
zDXe6BrR!ZrI&UrT7vFoAMM)m-z^%HzrR7B$_Dhci-da%?TWE$CcZ#i*)tAFKe+ng~
zv~-HCkb>_3ER^C=>0X{4y>=;{dh5|H;d&<J@}PEQ-H#tiU)n=<afrnx=L=F)bZlMT
zySi_*q0&w}6Ng1MrnULMbtG;K!r?Qfu<F|TgVXcBPp&!)`wLeddPJAgGhWIHyIlqS
zOxBj|t}+!@O-e!Sz)#jS_<`}ypM`~u`Sbm1a(>%ay);&_Yu(#aKoL<c>}R;WM_Ruc
zc9-z9DXKKGSn2yaB4ab2!5ItW2TLX$0Ywei7^nE@!E<{ect(ec?+RNu#@SEeWa6>4
zc`hO2f2;!%+#I`VJwv6>2^1`VW(RnVOx%!ioJ!T{zDlNe4`SAN8rML@1k#o~@{y1L
zNv;`lnuQy$O6&Qu4$wxI_l>gYe8qj2`l0Mei4e%td>eLyyeC!4lh*l!{IVsPO-D=<
z7a4G~uwsALEw!)!?2h2Ob_TyX4j^U%&4IPOGD;fepJ0axLY>r-bHEh>^ToBGc|53j
zZOd;P=oSqecB#*2cAJ8;-|SqS*G)SLk+Su_!KkQkZZL_lD1cn%JV_*Rkulnt$-l*t
zAKnVezF)`Gmbn8Fbh3AOdgYzwd@Y-GMx8{td0mf54iw_a-3(W_xXAuAg186g>g{0i
z1)m9S+&zotDI5}dyZpm+AbW`~y~d2A#W#Pc^S$iVaUd{K(w0Msz)-%y`A%u1*rYnY
z%ge-6MpCa(SI11er=h!$mw#_SAbP1s*RC-(qXZy&BFW%;5Ta*df_{3pNDHT~FD~Od
zDQ4seB=ozJlCJs%nixPxoZSRYiTH>a9Ev6HhA_hcUnVj1YO4m2W_sdGXMtS&_P)i#
zGm~8U?<<S(yD=0#nA-vU7HA@i1O4W?vFVfqSxAET$}L-AH<OwWn;Kbcgn4r>L}WXe
zSk7LAK6z6TYHS_L<QiVx9z%WyxT!BLF8(a6Yz@TGfe4ko+X9yZA6TZmqf<h)nj_w%
zt*y~_E~~#cJBK=;SGx4XrGdwe?;GBoH`Xdvc<+d!V@MQ@Dl<-2=u0Red09A(+}gTn
zwC@@D&kmru<f9Hqyd$@i;;@$tNr7y3#yXpi^>@kmWLqH-Yn=UKEl}K9==b6764wPo
zB_036D=n^U`wZQ>+F$)>nW@UJoa#NC1?4=(X;bgQ|7+xb^TId{n9Ju1#7lf|`4|?)
zG?-Ac2Bn{^S@~K?xcNBWyyNewpxcrq6#AL*wJkJK0F#RXQ%#40v-lVsOr)>!QQKHD
zFI*IR3yRgnipyMAx$01JpOVZ0uWfA|$6@Gz%iTCUPy@pFC<JWna|m9!x<F34d~pA^
zqQU@E!TFl)>+-%2y6F&SCj$Cn_gGggD~@Vd#Kyr_HH(T+8PBrp-Sj$`|Eac~J!M;e
zk}cPg#QJjBE7|KyG6$`k&C{8EBd*;GsgC2ZdT&}m{jAsaPfK_(j#I<uX((-eDShN%
zgWzl9768(Ws7_wOOjgIxi;$i676G$_nVqgvW)AJjllgtQL8HM*P#pW)R?q&cs)$#i
zFTNF-QdE>Pj^dl}=HDx%@3I#1aJw#dywazw^vSO;7g-&+n6l0%caGt-yBe4c0*2F~
zROyJliFj*Q<ZT3X9lS>hCS}!v1RZY5EaE=fbq1dX<x@|dM@$TuobnV8>Bciy6wf*f
zg9vv&NP|!MHh>olP=H6r>BM)><~2*Xl=PWXSzY@;rf+-Dx#yE{bcVzGn`uEc(&*mU
zf(!|wGH*4Oc)Z?3*HUd5Y-rN*8g|IS??tlwWqiL7m0h`UEMz7{?+RoQxg1=gZ{|1&
zf0R%e;9?k9d+)Y*^mwhu>guH>>9&i|+OT137)x#i=xq0ye@dC)W77~WI>W<cC<g5i
z6;8GlaAMIYQ7Cy!KjBI04FkZ5bhjt4TN}V8V{tQYDEYx96G)IJv+5y`D;udb!C|SX
zHh?lR`#U4*=}K4W_i{{lIx60-kkm^eLja`xZbc_OzEG5+*j_~I&&tZqW|>#OR)0F=
z_A>{t%mxs<^r$2s*0dkOdwXBe2j2<?A<+;1>&k;>_3#gEvNt(nQ9r7Jm0#OFm3x?j
z1JB@>UtZFU-!Bsx*1YXAQlfY~@$AJ<z^gyJ32+C+of?5R#$kPXyNW*5!6863kai23
zdE0u1XC`MWMgP&G<P!UZY6{2Iu0;TG05K~e5js<oIm!E<xocxLym3a<9J=KT07sO@
z$H2O9nKo3Cu23$!cN_2X+)2t;|2MwE=de3b6aI}Gk8&ivu(_}nOtXzD)TX)Gw&7~K
z;bf@@uddDR?r+Y7G~>(5d>)`FB(Fd&(bZ)k+StV0-f2rB_!#Hmla>P>ms}}fE$p;E
z!0ggzu;$S>2iSS-&$|QPuAjdzGwrP@lsb-kVnA|1Ih@BssMugs&4Xy+aE8b@P-Aw+
zGKCMG(F=hRNR0fJ!~k1|mkaVph=%@OGro}^(~a&z%NQ6K0w8Ye8-JG5TPme5lSd-E
zA51HsS#0g4a%(EH@PQxnweo^)h4U}ms9#;alJc_<g08PI2n~=PCQw->wDb}<JO^v=
zzE;<9QmMQE6PIDx8GAq^y1x#MWO<46Qzg6ztyj>+2eTc^L|B;Os1hJT-jFe|L3geF
z?FBx6e~%9?+e4?=Dn(RjntgvhSjsJ$YahlldW@Mb;`OfFzFj_R6YHBwA*m2;!t1<*
zL`*r{p=c%p^;3~<6d}O!3`gG2p=shuRpE+s-QC@ZtgjrKtuZZh^+O|qqB_G!m5R&Q
zfi~@ztN4-@w<FmgA9)jv*0;#sq=uiCpD~GTo==H5xCH|!(BU68F@V?lYmv$-6pT8X
z&XIn=8X?|O*q<d-@Yw9mMr!XrIPgMszVX@`%QrQ~V5#$RMJJ84DB-SE+K03nX4i+4
z#%J)drc!mBZ~F(_nAW%lSbb@k*^Hp(XKWbQBXf$2SuCg4GE2i<3xO6sN##0DEl5h*
zg=iWZyqp^RO41Wjt%^z!n7XtgFiV5vY3JE>JY2McTW&&_d57wgOq<;b<6`ip^ySe;
zI9yn0<okj>Sa6G6<;GG@{*t!|V)kWyW8+ra_r_-Hn(5Vp%#9qWK%)1IYQhI=7=!Uu
zi1uFoQ%i)LaI@Htv4Ck=hMroLk~B$Iso+3m_6P3k5R2oKMWmhaNW%T?&B2jN?Jp!V
zNuNioVB~bEcKxq?tBRyS`rue(&_fjP5s|7e>2j6Vtwz+mU9}8OAm>1|c<(+><pb(Y
zX?{?&Mt;!V?c+uz!^p614B$=X5e($<0uN15J(q5yw5yZw0M>#VuiK_r4+9PpI@ZWK
zPSPHq#qTWtMR=A;MfH+G*TkyZ%3mnxr#fyF<fM@DopdMTIvuBI$OsrP>c22{To?FC
zF~2pMmB^Y&f;I71;tc#;PbU?S0hE@femq|P{;`ztjOA1vm5U6>Rt9<PAl?V8u*56G
z^0t0B$DiI!u^&}>ZhwBuoU<=9ZchGDV_r3xOC1WxK2B-ES(><eV~2Ime|=`u@CrIX
z85+O+>$Scq+0@jeO6IWPalR+|Z`lbI70~h`zrMnN5n0y^-UN7AGu-8gW&ul-;u7TU
zL%^d$_-a{pl^|1lFXx1?FbsXX!pV#gn#f+RIOW(=@KH5T99=gU{K#xpN4B^ZgWdCs
zKQKr!4SQquObE+HyrxKH=hDR_opKd-s(iNL4z_m)&R?@tfX$YSm4dG>lPjz9Lgsr<
zyu|GkUQ@9F7Q7s=2c|Y&Q?qZP>ijrS;Ti=Kp*%$*D_*nflJ~E`)#-QA{9Si<q+R%L
zClN?#^UO0H^Pi3q)#CG#T{}P|%wlcxgmO6$osRVpeuYZ#*c;n8yV1bi!hA&U$|x6A
zBJ6+X<=g@5`s2UZ8UH${OJd}BPTc3QLj6qf+)<)~UeIMPvhGK{d|tZIA*Yrva_KSC
zUvEE64x2i#4XXPQQRJgM$gNu2;bS7=!Ns<~sH)s&CCOMeywkQLfHkZ$D4p^6Q%^>B
zcgSJYlRUo*g+P4kiO#!M(mui2sJV2$gZLh@fj=sQ=W^0`>KN+7COUlYR1wy(?6``j
z9n5N;<x<!JxNn1)5&a}G*7U~YnSsC$&axB2_c*SEAUC1l124)zbVa>A{<FJM{ysbH
ztPVx!3ogLgXJ<Sy(>P%ACRgG03?{yI2okTMY13GlMww=V;{CK!=qW{&0eLa&H~c?y
zOIF~_f;f*fTEr?&wC7vkh=lYNzkU-B*Gpj=G4y$vrEyRzN|-_|X|?$+8B`1DN~nI$
zt|Sk-d80o}9CZ9B&;Lj=$y+_m0etnF9DH|{?CpOzZ*;X*Ah!_L{t9_;u0mY{m$hx1
zir;6)D%mz9Qf3ja-AC7{;I??lRnDy>_CP<XDVe8qoNO=jc_fo-j5ZW6IB^WTHlWP=
z@n?P?P80T;Y^~OlBu$9_$IrEHrs<Q!qmkk3devLbasnscr4_%sbWZJBSo<L0!hxRd
zmU50npV;;sEqB%L3t+DhXK0dz{$z2*;^962GGYT(|4(IrPeTut_*Po-I8Z1zdFSzf
z5q22UfO=l%u_=~LOhF}X&tI$2dyVn8O-;KVco+(*_QtgN;)nPUTgCIC8>3&MeY>uM
z=@MVs7T)jI6fzbq?0Q~|Iceoe4RcV-S%KA{hm95h6id)<cvufuf|lZt7Zlp<h=m(;
zBfVlD?3S==3q0S6I)h+M{Gr{JL)&5hI2##S9KihzgeN6}?|D4xhlQ#6B-*2u_X%>2
zf}b)-Z(vevBd?6hKa+&2w;R3WGK;k;B(;1lJ|T>iij+DLcU@*0_Vj(mYq^9_qlq~K
zHyfCQ_2Io<=X^QavC6}*Q&2FwuS=E1!!3Z&&5TwC0eZ#B2F;enPnJ`e=Vn4OJb%WT
zj7|P-82LwX<Uw-Eg#Z;)&q<?<TlV%cFks6IIq@pKil@+L$i68`)Rmks)iaxx4S?=}
zeOnP-9E{P!sH6s-<CKC)N5+(5PS2yo)?qXZZFWESu=X&pU(-1Ngv9$s`GeyOE$$?Z
zvo@eHK+27u{I^2%^+-_30?N4azPGBh8ei7$L8jWQRWpe-C6yi5G-aT!)&JAW+GgL7
z#)+0?qp+`*S=*{?T=*O$lLKYC#Mp|;8-fVhuB%h-zvTr(sHj^HF1((MT~;;?f{~F#
zJ3N9InaD7~RMJf-UK2AR7@)m&vkpsmTnnZd#<v@dTR@Mn;_Q==fG6g7EecuN?mM)6
zA99!**tR4)f>LF-DcV`*ukzS8FyHik6I18zkvU<mGTzB3vo(!GKYcP7AH<4fr*pj9
zpWxKT?{Tpxsg)-+>S*GQyl20Q*a$yyNWHE<zpHodslic(4Qb2x)xiG3??sU!+=Qu{
z$HzbIs-iI97LdHoP8mQ+_V>prGTuJ@S+%`g1)Q(`RmL?FpRf@}1i}DvRl$z}wvo9@
zNU*xTd_zkQ{?q5!>*Ix^-7^b20fIOel0I!YbTbc)I}j|7+m2b-Uu5vyhWOr}?#XC<
z`m2ZYmz*Ic*lYftU9UBmPQb+PpATv$CG|(hCbQaQQJoVFPqaTt$oan0M=);xh%NWm
z^t#ZSRNW#cEj@Lrn`i9P6fE}LBU<ND8!|>Vue<h-BvogBiw}F8Uo-<jx{x3ET#qQy
zKJZ%$Wc-I&_@rw`kAhNHGO0m~vQQ^ptX3}L_0yX~lwB??m7+QsKoUZ~aWxQ3iw$(o
zvY(l8yjE=pG72B?P;FmgR1*T~mw&q=FwP;363Kpr9rXyfuVz141AsSR-|2Su0QMk)
z>U(*QvCEjw&$)@wr9vK!lhhrd<999dMlqXAHL9~_l1xj1XJEiam#J0i)jc-pdRQ3q
zO%J<NO@Tl93&KYtmcBly`73qX<6#u#fJu^VC=%**Y5w({#FBic@cHtags1fMb@tDT
zyqUI@bO6WUE=eAISGe@*szPq#4!yFmm2R;;)>{8kehA=i00tt4`+kc=J4rQwnNGht
zrA|bo{8DFVcbY@OcN=6_ZMJcJE)CILkHYBMnJHPI6z-iD;cxM~!-geqHTYFru6-fe
zrOxPFEEZ0&>jsKgE98WwWSR@Gehf1BhbD7qPamEOVv{r)m2E!&*RvOT!fe**f)e0H
z<w*L|!V`kFT5p}7tt9;9-?ddw(wtkIXC4&>_Lg17Fx(~rN~&p&9GZq9@J!IKdU>@t
z-F*FgHh6g8+N|fb$LVh_?=32dDjul$G5UU);dr%MD&M?qj1CqMWl*Y=wuSo*QhOC5
zT&gYg@Di`4<J2cF@WS}NJm>}HKr}pWvG>P0l>gxgu_>@#YugH^Vfi;0w9i<qp#Cuh
zQRX1{SDmG>jS(MQ$D9jyDhMjEm1%M)J>n@MoDcoE>aW$oRF|X0&8!9z_ee^y<hct<
zX9ZHUfa{`PWPEs}qsjhMwLZLDsAxLeo&~Gosv8Z1pYQw9#B)Fzzuin+RQo4rY0}_M
zRa$>ru|iwEC=->Mp-|GQr%ZNW)0LO@+*=UIwSI!j#RcR5^L{xFa8iD4v~uu`>OLl^
zB0wPGJ+9B35C}vIFs-4^gXYzM(L#9)Hav{uy%AlJ#tx|8J32VgJX^WqE2|?j74h0e
z2C;)>?G9vso5;!PZB&DKA(mwfSp{p}8&vg|MogQ>cM)u~9;(d?g=4w2D<;0!3aYmP
zn@87D$xiXqG;N28yQ~<zoYDAf-zHrwW7n?FomtgJy1Ox84^a=e;b`~giS?laab+_z
zqIrJnPus?cvQ4vJ#<9D>?!*Ab3)V}kPska%$gzF^ixef21w?g|J2<%-0liL4$(-3!
zQRhfUKT?7yvB&fJw?`UP`<dUfiTvQC{;drjz8dBtwBf{K;+T#(83{jr`sG3vZKB{i
zTL;*X_hmn^bFL%aW$(&EAd#<cOW;ks-s!`{%G#=BKHy-d2Iy*YS1a3WIy7Io729Xn
z;d<=Tc--8hZ&Jzaus>=i*|JU(Zpl|HF%n_6RblJm0>0sSn%}|CMv!g+0c!ENnaLRr
zC&DV(!bdAlX9vsvbGRjBvR6tZ2+Y@dwx*}L*stlulem+*;E<(4{mnSNvRuSKzGmrb
zIX;l~$_QO^4{Z!JuTg}pSh)#=(h3dWYud$c`T%YdGiJ$s+~@M<DVq`OqvQcA4n}c*
zn{0}rzmG(-`d}wHnTNRBk2<AIz7zYy0F1&t#f`_c{jpQj%5{ZjB-J%@Y%WCx1@;5*
z>l$*9h+f9;YJ(FWzAd?a<<1v)<7j3bXDOuS%(>`6M14bGJIy6`calH@B9f4hSWVVm
z-T|z1^3ZSOg@4*!PiJ*I{&W;r{XIMl$x}%yj>V3$%su_^0hTGh)@ZSMVBYenL29tM
zsrrhCT3!94!mq=#TBwe8xz@>B+sD3cOAhayJ~F;p)d+yA;2#(FSbiTV?+O~DRzeFH
zx>T?1EDFQPnpyc)-wM(a0|K$C9bge+h26Bn;3qR19S>`Oemo&=enlG<7Fd$Zq(fsF
z2V_tIX-hHICwgbCmK83l`@Ximvp-o|4lrWy^>$M*@y@4+WbvMYneeSkKGzpPuXjs7
zZzy*rf|iQm_{&jBNEY&UX5KY5D?4%G<{NQ9wODok^d>%mR-73){%ZHf$U6+gr`F-^
zlZ~CrzZ=?@8*=wHO}-y6G)N%cZLhlS4&6=?fRI=}>d|bKn8nxYOv(p@C;H;q{T1wT
zhcGHivxJWp?x>hove*HQDa{Xz?7Vvv&AhWW!wu|63m51c7egavpAwNj=fx}9>O?Q^
zgb~Yrm+lA^;m<fa3TxC%;!@}tHx+jhdz-fl2uVPI0yg?@QaM+jAf25tKLxOy$(2d1
zXb{<4vy=ztehF{)VZ%(R%<+$Su$?oq^%?Z<ANd@lW${z!fNY9iW=Go#ds^e!iz&p{
zu#u`_v<Xc-(lY$(cS2YY!ioQGrnB&q>ed<#=6mnlkG!wxA@~iLh1``hs!nCzA9tcs
zpljcsf9xR+RJUpMN79E)0ziBfUAmYvR;b{&=@0hX@%t{Hdp?P9@4b0qypT^5W&7p`
z>e4`~$zHa9H1j(GJp2ps%Mt4z>mnxqAyj$sr?u)do(1&&A5pHKJVEZL08xulvF$qf
z*)KZrD?G|4j~h3-V0GZ!8Q_-+(!~C16!g}S`1Ychn_VqlUIK5XQrkVeRv^3Lzi?e1
z-e<s0jNaf+{A~Wx(u8Q@VN|HV**l`vee)wElI$guosPg$#aV(OZgIl_EAm|D$SV@(
z$54aUyV>BA%julJTPK`R1Uzk@>Tnu)(Gqly4}+YSy12MVEZcpl4|WeZ1#8M|x98F;
z(;#NW|9;4(fDRlV|F=RsIiPgMre=<me;z#*Hw%X21*(xia-@CJyms>X{)7~W5UP;-
zb>7!_htcz%$!Y{b&YVw%XgOqe_EnIFg@cpTtE39U)YaGY_nj_MTN}hHTiw^sqz>7+
z2Mr{e8e)?GAn}A3a@h>i;^dM%0mbmJ;agUY4@Wz`#NbG%VJy07pRKWQaQ2Uxc7R(q
z*b8O;7irIJ4_L)})^SKj2#)Vo4JS5c^LBRvzVlB<NIze`79tzspZjs6U!8^^KGgo&
z^~DEAfpe?HM?XS0l=JLMUj3PBJto)NAp6xDtS;oL9w?Xq!Wu-d-{vxHt4<MV>+~6^
z5&FwPM17J9n`KsM<g{mNEX|MT0b7Nlk#B&o!Kdqk!VAzKt6Kbyzmd%+1%~i%lem+q
zde`wBPjw|(Y<HoYk7XwN90v;p5!MX5Vr@O1Qm^p9gYUVX5AxE+d7)8Mv1Kw>v3+Pt
zsIhL!{2|sxhzM8CshV1Nx$pfHSA4I)&o^li_n5!{`P&>}qOO1h=opAHJ~mD}X*3L?
zTZ!p5X=)a`f<IirTR&__@xzYEI1(5*G}NojTY=*=ONt1|q;z*}6j#Bv&MLu@UtK+7
zsXiB;S#hy#ajq1w8(zN0=px~1&VJ3k`P|Yu_~#mAWJ+NH?_J8*U1)<Mj@~#V+4XJf
zQSA;C@-d<3c=d=^OYVwi%8U4I#bgg{93MAd(5YYyL155+F*W3(XPOWj#|5Bz{P?kU
zfbo1B0=BZVOFePaOw*r0D62)+F=k?8se4|n+l>@_x9QHoXqu#qxudTEoD_>)J~0Gv
zi`8L{d|+j5G+2r8b&x>Ri`*)WjW0OYj)wKm3C!>25-vw!w|Gj`46;LH?8&js1IWVB
zoOJV+pUcZcqk{%tbWckk?zJJumW5&&5#^P?^!uIVg5%PI_bDfGf&nBz;_bPaG<q>!
zS@C`bcDvFVjO3%zsF*2QK?>5wah+mvQ@VAc_|&qbeDZ#C!fR{)G8U~Nn4nJGZ5dq!
z?)K&CpWCLh$J=wMaAlNWD@<_i*fOY&e!oAmQ>Aozi{ZX7c*s`BR_fV5DaY?l$^3fH
zaBgYY*4ZJu-XZ^czF1g)z{JkAXB5dq`y?vBA?ZHTe;X=G<3JDAm=`qIsDm`<&8BAw
zq@1<A8u5HtY=*J<(iXcohfUqsq{#i59NSM5acz?d&Wg~Ha82y8de6EkyPbE6+(ajP
zL}EUZe{tyN9d8+5K#%eg*tw|syVr)e0zE)FJSmhA-SH<GF+Hy-;yput(3mmxlE4i)
zDZf>ag#)8!!56SuR=9dfA}80=c=FS7Tc<JtWz;o)@aM$P#*M*G1tE8xk2Li?&`0k9
zUoRy|x67Zqktgo(YnMd?Hu-@2;VE0t^>n_c0F5rKqCQ+3lXE2L7uw9E?>6xKi&n*V
z<Ycd$qvr6$5g$X~B`NTOBaG+|4u!`(+q#%9o+aG;sCU2%=s%e*y*!)Jzrbc`io~&n
ze*O_UDe>(&ep;HJqaH9m{yO)rzP_=$OmFA7u;h_i<<B=+hQRFmLsRq66j!Ns{x<r4
z75SqpC?B9I^O<#x<+h_=NbgNuOkj*qV_B)fR@ly{pHZ@M&n?N>g4G35tG@+m&_otF
z?R8jw0b+wzB0zEUdUHv0usC!{0DgkXpKfE$)fz(O7__U=MB<jYf6A3SMS)stnj%?j
zJ13!#-uLgB<)YkSBuPIzj(sT@IWhpdlvskCN8X2r$Fg^uWO=3ks7282IS0w$Bc8$8
zv={1#_?p^g@vo#xK%)7SnI&40it5_CO@@dFJ646oJYY>J=xSh}#(8Ko+@M}<(S`ZV
z&+hwA(+-+4%?xZL;8@vP{x1|~bL()E(fj3P>ifHup`*?En(_Ij`oA@E`|4ah0wI0^
zT-ui2{kAzXWWnxFtFyq|dqjxdVQ3^ugX_e({fli@($3>DvhQi5FLNnVt^V4?S_!Fv
z+-hWrxpLRXP-Y$uADL@@DU(B@wc0=bpnbXYC1oP*1@7lFRwezlDAFhUVeaHkd8yv*
zuQ1^puQUcxr;l*D>E}MG?jDfb3Kad0jAUReMrC-8sK02-4WNm5UFwU=_C<aD%Jm~~
zWrDdKMQCY~zX3#wPioVI?|Elk7fYt&CsRwN_X|=6-38XefM)DuFyG$E>Ph9+_BkD(
zQ!lVW4WbZe$)>t7AkwB<AFEn+bu=ZD!6(uxtAjRqG9GF52{8|CHP%t2MrErQOF@JO
z%kD~`TDaBRxV8vgipOYdmbX0QIVOMI*UmgbU$o>_c$g^)8}_v_cRU>gScU?igunb^
z0t3bge@6sBM39?9amk>7)F6kCbvyudLc4iES$%km+Yltqfx8{?PKQDy>&ZYeDR8#*
zDRoG~5&^(J$J;Wd7KwPs^>As6*FIIJYZP_YDcNDu0>;i<%F@y^tz5|jU9Es0K@aph
z6LRsVw<>wh%=Cds&@mz#wz|AznxYgT=o2h&^sD;qa9#9*IP&E8M#4ASpSIw&8Ndd9
zG4X*{)pImoFmZ<sP!)jGnsa3?_#;vYmpk{z9;r|1=vYGnctT(QH2;1&AK0v>54@+p
z0}B(vYei@UyI?&R6JpIri~r@MzAV$|;ba{Z)?}XyInAAOetjb+)tD#Dluwzc+j8{h
z^b~VRD=aQfCg6c7r>1ecRPGMX_)DS#*QiDeP}@=GZUM;)&90~Ea@PcTo$eo%K~@#f
zb<bZsqzjn(kT(cx`l{<VJJd*%rjrqyi8;g+HS(1Uyy8|KPKk*4*IA2jWoVwx&sJ|=
z({dEK_t4h%6uf#|TzUBU?MW&3xOdV=?;lahrG4L2>UUY}l(Nz(isk7>ftg8hP`=3M
z3&*0rb>bPXH!1D5@SEnqyor2LHACNt^omXe4TQ(*u5z43H(LkhD<|%oRE^0lN!9WY
znZrzEYC29199C<q?tRuUy+T0%fZgJC!(MZLzyXZn-61W5-D_e5F+hr*p}indcC^mY
z*(*7spDnA3%BXjBHL>0I$bENO16YEZyXLql<dv|<4@Utcyn~~ImR57F23g${IQ6nI
z(^{LEE*GC!>qxz}tbFE<5uW_-pGH9IOETcTwa%6YIaSl*sUqnlLG>D1IAo>JRMg_u
z`Q>R$hZ;!R3oI6jq!IH44V~2QNcil5dY0S^IJeP3NaRT}aOJyo1`-r){@HuJGm%&P
zT(|3VvEJpqivr{DDaiTf@07yBLz$kRJGZUV|0a1z#IvbQSq~7YxOI5Zwxc0qj*-Bq
z`q5JGi?r02hXa*SxBDs&4*2-M2S|T}8rX-eo^ELa!_p<!(>`+T(<u#D|Bo5bK|2xJ
z6=Iz!77g8H=*}}Saj{=*Mbn4w7mLumpljoF0t;Ak+4~^-V6;`B)s@R6YlycL+c;Sg
ztCdjZ>J9zFdhk{L)%C2!c5ymvz)}M6urr)qWFucwJg#{-^{M1?YE;N;b^pkCOp_3m
zBj=lz4!Z#O44XG{ml%dME1P|v$q7|}pyKwD?+RNyGayT2rq=`jMmYT-G&(3DB7#jT
z2P~~pvh8rG^!hJs-Q;0LsVbX*Kj;+iVSD9vf&X^dqP$kGkz;<W<HLtBfTm#MSmbS=
zXJ<@;^{$J6E<ukP@~oe83p8Kf;2z|8{?#8BDiNRc!t#kG2cGaK#tQnxS>J#JOZ7_|
zr+}ohSv1r`Qm<~Vt=-ms+0SQ9SD)5~oAwD#s|h$@MR7lASERS#QxEpE_a{EluaZp-
zpBg7H0z}jW2LJ`QES5<v=gFrRwG9ox!qgYIz7KJvfZMFDf%#Rg5{^RcA7rxq=Z#&g
z376*CL$XUUn-P;wOt>ANB+^ps<kM1VSa!YVRWZWkLx1AQDVz5IP#k{ngDGT+<5fwD
z(s1{^taAH}mjo?p2LP8zv^D|%*_N#fUg+@&<?^3s@cNLv#^DDCdNfrelL79!@)%(J
zlCbzMoP;mDgY$cAH!Y5sDBTlp<HXr@TyCtO!jJ#ejq%dwg@34tqLyN)k{o~JKqh5+
z98+Lv<?*G3)TWmTAu*{@{49108f|C}jR2^k94W@~g_E19HLx)tDI|9{r#<w}qSw48
zO3<GENg69&SB-#yUTRin@0d?k0ZEuVSAGE}(6>xGXn7ruP|`~WXWezaH#sEQN6+s1
zw4Jp9!6Pp*EV~J5x6baVDWT4YViw{`D<S&hVZfVnJ~F7up@=iP-f^?t0owm6ZT57a
zkaS^Tp>LcWJ=^>=u5<zs(YDz{75<X>@^3P+;3oL$&vcHAS6;*xWw<V2RVTq>T3S|E
zdIrVyf!!Rn-yW38d4b5ZCy{6!bq`1j+)>2Z$0t>7{nMOQh7O|pA`+F5YwLhaG<dI?
ztEp!aC#$6MJ6cC0tAyqd--k_RUE=GslK>6j<a|<*(^9|hI-u|_`y*v0a=;(?SpgrP
zO;&M=T|cmv2~+WJ#2b_8ccPMkduQpH6SBea6+86Qxkf9GP7>`ct18r|MSKG0csQa1
z!$ml^1T$?>3ZothOQ0fdMJG^7SxV6uex(?nvBGCRa~~}Xgh`KpeU_~%yq8wP$l;gY
zh|l(Oo2RdY()S7}<VhrCGsrH3!o?;+i@w{7d{%Md)grXQ3(Zh_6(i`^^s#QpWUZoc
z01cY40uS$1H`QATny@%mJkdi%SEUviA6e>iL#~?5x0iEG-{{L1YaG?A0II-X@TlA+
z&I7=TLW#-@2<2+F2HicU<`3#QT_=N~Z-Q4nO>u4l9@<wnMpE}Fl=>k;#aUi{{kM`o
z+N|AtshP08F|@=D@GRl3eO>W8KYzsV(M$9-W@I6gVYPYj?7J8G1wTEK8}~nHMJl@J
z9IsB8-MXf?6C>0#2G_hsbNegAwbZh7ofHK@!ZKi6A#h(FSTE-5;w@z)M(_i}Eh%v6
zf(-dAlreGq{c)LEfEStNFe+uM7+`B$01M}NMggB?ZRq(q?}sK|iNI|_`PGy`WkVnx
zM{(6)LVBo`Kj}F1nY^OovWGyW&70WJ)l~WwxN=LjH?#iVOyKGs4I9lhZ64Yep^`gp
z@!NWCSRFKiyw6>cSBpqYD=r46k%O1i54RTo2)+f4G4xLgLxAwh!|ubS_T_+hFpjD3
zF+Gvmj^K@P3;Hi0`n2Do9<CnxCytt&161WnumBIg$IXNc(7-9{PysBlusY7q8`#r2
zp*sYAcT{ej=X||u6wbVWk>y|J6<{nRl@LbRL>6+#XPc;e6<bdPIIQH6q=jqYzY-+9
zuE7ddjun%3N8jZ~6xV4fQpH~CoVmNuOFqDr*PhAHD^w9DJjIgE)Vc!v$B+K)a%4~5
zGz>z|J_AbUzKBBI+jbmiP)~fAm1b3FDn1j%h#rdgZ>92EX&E)o=#f>H*IjT&p<*7N
zCEW*?EG6Ej)Du)M9L3kdtwI{Vf)hl>Gc-%QdOCGws|-e!Rvp8B%zy{R^yq0`WtL)D
zy<l;bit;FQwyuy?aVAt3{Y5*F%(Zlcsg#$z!$qS3x^BoKP2SC@6YH9ppW5|f4P(;Z
zUH$KKk0CuY3qhRLeT{*5kyJuqH%#aPU=fBPx+rH>T-<;JAW$f9>P)JjRyxh7Ei_Wz
zhah(=owh?-GMKd`9tOcwbR6lW(X86wdyy$sZyk?X{})2z8SGDs%Sf1297pG-9p?W0
zvHme9e_gLMwiH(2O)9Rdetq}2ygWy-cEDu*kjgj-k^J4?$h3;eI4f~WYn+>;A^XXI
zu}L;>`-BV2Ysd)4E4|7Yh&tmPo9(Jq`4r{e8{GFhW{)az-PW$dd<^m-p~@sy;V(=N
znW5#;_NBd7gZZ`|Rv+aa?t#!rqWJP9o2m2y$kccMMOkr234E)W?0OB8MNW}O4XX5Z
zKoSS_v#Kd)*{2j|wzuPSsU*chH{$Y>cK}cP&^a9g`2a@evXF7o@x(&Sx0KDTnn1e3
z$LOO)TOnCXD=X5(H~QVc4l4lvLn~fw2B4zz73flMl!C$^VWC?85G#sL5Qf;=*eqP_
zjfbcThT+Rd<zm}?LG2Fa*ETnf170r3(G|7v0YB;l2T9uhs_S}0-06eV-|`-s-YPor
zf7)R>;IyLvpj#R&@gAYY?7C7n;UeQZ`SQ3*d4`u=8a&_C568MS;zVh=e(Gsmfi5Si
zyAkXQRy4nM0b|d+rZS$#LC>5|u`}m_&-KVux7|YaLTss?vJdI9qu3uvZU2_WLv&5h
zdBs0sA*oYVQ0DG|7qOYinFHgWLa_B`y=BS>J*C-cQIFmlCrV+W@<Mf1H3iX)%5_MA
zuurQlJ{MM(r+Jp$R?06c3g+;Iuh6`x{*#!h#+N@5lq}=CLu*ZP_ivb>=D+sai#|`X
z4$=f}-Y!MG)@9<!nEHx2XiU`c9~^yRt6bK|{p;PQ>CZ27EdEE)IR)0Wg+VyBt){VU
zoY=O}*p1oPwr$(C?UOV%8@EBzw0GavJf5t*@XyRQRhpzyYW!#=9`}YRR%7urY9!rU
zq#{{fY>zD`fTb{mPL)fay_}wICIrZ}JRE=i;Qx10_~%P>u2GZo^#tnQ{eRp4W^DLQ
zeq62s9hFOf(~5xPaiRClwM!(&H$fi-fp$*3D}Im%&q1mbv()2MV4MJh#rhX{q*KP*
z^%%m0O;{3)>C|@(TEZWWluh1nTp6^azHz(n*HOTO-XmTl@Ixu4MOn2agUs3LOpG9e
zHd}F$DwV$Yvs;tb&{^m^=xHa`1BVa=4c+_3TI95Oasi;0NMD!lALYO7nYtZA`V<HZ
zRC(;aC~r_BS^6kb#x#wjSlQ-7QFaB~2n?{Yb%;+Pe4#Iys@&B-8R1-1vz<p~?Hk(K
zqOZe>n}n`wZxa*>!yrLJyJ*{H!9AHGaQQ-rhW_$;${M)PZogrNuCjChA`U;DNhjti
z4FZC}8HYBa1r>Gsy^BC&)(<k5>jp2&9sNINxlZ>-v0tbi+Oo@&Ky4hZoCUSCq{4~f
z1T%dHi+_KrbbB1r;7N3@maIEcl#S%TiUPmo{c+(RYO#Y}>oGgCeQO{U*TcjQB1z7)
zS|FC{>84MhNIUPnrE^?BEGx%^4+HR(>1mcTgYRB}Bmer>kA7YPC<&)jMH&<t%7X5D
z5H8>imFw)3B^j6LDUtbc5SKl8(gUn+VSbK%WMr`O-L25yll>3D24lC0JM`YQz>kM=
zq~3p*fxiR)4>SSeF)3RV!1#KZQbKpt_-kkQi)?3Z3{;`^sI^P8OzE<PzklYkcii1O
z-RW`yahh>-3X%wdP3N{_CYKyR0^HMxGI8wigIXDmg`V=dvuaU;*F?diEcQk>r-lJ`
z;O<FTn)@diTl=T!NL=><{5E~&5;n<m+!f?128oRMIGgg&K0Bzu##IU={|R%2f=o9A
zbvgVc&|Sycu7bT`=8|BG?!4z&NH%tVmIVAr4~o|Q)7hWi8ZMjoV>F-?GsEs2)-oPj
zkHDN=6EHXcO>nQ%Dgncb0guQyUD4a<o7;j#H8qqQ>;G2)m)i%-*IobaE?eBfo5=DJ
zF0TGLx^LQ%%99}Bf|JtFE75A*(E|6COMVUX*_`eg%HWk$FXMoIaahZv(&M?1CuejN
z=)rk@I2I1L=owu>x9`PK*s*EpYz3xUr1rJ;b^C*npZP;Cmu<!@I6T7rGfF`4(~`+5
zt(Z-1o35!%X<-A2z!Aw0T>8U><LPRAyr@Hik&y>oFC~6|Sl;gX4?RCa|Nh{~6!uT-
z_2S*!G;Dn1v2LaV91C+_{$4ElGzAorsj?s`A?fEwIXK06Xc3KsG=cz|H?2w@87sp#
z<k;5$(To$vh$cLP?Itpqg7XF)4(2SP*#ox2y^pT|PmY3B=pX1tJJiv-J;D|%;@ubU
zaMB09l@<b%1|X<|tJ|xEe}D!TQxZj`L|+@zM|ZW5dn_hjtvfhqfQ-5WSF?<0JBCzR
zTo<O1G#j*(8b_l6P$|O5axKc_q}7Q2EaAe+r5vrxA*mz5jbd7=g*g9pR+ozlA2yN1
ziTEE`Hb4K*V`2Uym&8gt|KZ}~dZL}!+ObdIZ=I-s0o5r#-Q3gC7FGmgkE>6<Hs5fp
z=8DjE*Sp}h<kd_UiUX6_#mMd~h&BdbeQqE~sc93~PXj-vEwCJ^^v;ToOQ>$Smf-4K
zc00(T?)~)&-Iyg?=F+<hzC1w#qOnx(cEPrynitFHkXPS6lgtK13{WtFlR}sFDoj-!
zYnS3$(U6&@75-?1H9Gv4h4TxaeZN5Th`E5$eIG_GGtl0yNK3C}>2q(A`oGO&QYS-A
zOf(1A5p&YTtK9tvN~SZ#ElckJ#+g$^Y!SW5a{{t~7?uy(&0tj_%$4E0G|uv!=Y732
zq5@ArQaK@O|2@>dw_BkQvLV^=bYkA?o?qmeHuh>WCV`@rt5mvZo4cPlrWG7_yT><V
zBDs1Mbj`cnA1HwE`$oV6a#+$&{NH(eJ|UUL-=anS9lijyRw_XHJ3L$}LSkb)v2;@n
zT9KQzXq1c$i6X~kh-&^?3~4eW(dBkWZ_oGI$k^`^R=I#rPL2P%?(k4};C$8As|oD-
z^jixcCfEw>9&kSOoBI0ZfBtuhPKe@}&zhGl^cl_+R$M$oYv;@U$;pEjAVxTQPu&O-
zJAUN}N}?mFE3|;kY&1Mgjr|OOPjsS)=fA~uPbBqLagIMGJ7zBTJzp*UAJK0HyUb+b
z;U^aWj&SV+=H`GqG~kK+H*j-%*JW4tl7WsM2x&)Ac#V+lza({WZ>5s<tY~Uo+Ca;e
zsgj}Gc+R~&RT7XSocH|+9hLZdGx)aQ_jLW`c<*DI07YSTHjm#cCsCv)N;4*vR@pS(
zS32CenzQv+!yIx9N9A|N?q67uaoy`!7$NRgCTD;bU_2K8ZSQ4Z(S)t+yF*C32h17%
z{|SQ5wm$LNT>t}ti_x;M;Qm%O(e)_}cJJSTL|KXl97P_#N62}hFI<>RXSZ_cdRYy_
zOv}OR1NMmz(DsRBN&t}S%mIr{I%6h~I7A16=45iNfY*s4x3@Q~qHIODq(c=bhWhv*
zFKU`dq2d?AaP>PS2>zUO#xJF{<{U$|_(;Uy8N5F)4PBl_$);dUR~KI{@9HN)F-!AA
zg<#G#Z-srzIMY%_hdh%r^kY4?Rs&M7IzDYJDKgZ&MjiV#g=>B%3|jZ_k=puuDTMY~
z@B4SNpwNiNzM4Z>Xl|M5EIqTxs#d1_API!TY4@Wdqf^hT8_(RCtR4@RreD?tm|*Lr
zo#lciO+fS7^;ruvuZa2uYE+)Rw_`aRg<XDN9Q8RLefbdhvGZ2*eWfJq-A`WjUQ{-b
zE}@Dz!>0av5SB9+k=)Q!IJYjALI!rhKvhnTAk0D!Nm~_{-zVAi80>)tM+w?LAD-ap
zS&>wWyCZ(=a~Au)BF3eqSL83op=8^cOGZ_ycGRFputsVQ>&a!wG4JCX2<XtCXN>^M
z{Pne<y8v@pn5}2%^~PO_B`k1*gqxSnWz=bKI|idp93$FReBTu;TStl}##F;tvD4JO
z#;dJa;Tnn3?wtmMS!`FVJbn?l;M||2s<@p0g|5k26TU8@`;|nA`-*94iwOf0Agg8*
zQ5yCu-b<Eo{&5AX*TA6f<@P<zXZWhwJk1h7-A%ker!a-<E2^G!`|gb-(V@sy5MFYX
z?p`LHBKj0pw^5@(cN*Y_C^w27&{xO}|NRTG_yj~llFe(;9ECdi05hrC=_W@x+@;R~
z$p)~i;~5{A+J|1~xyIEO%ardB3<L&-hJIj<iY=_H9R2@9fAN+THxbUb!DK0cv|bNk
zWEi0v(a%IlJ(rq?b4AwXV_Z>GLQYPp*?V0-sNHVDgJp0&dFh){`}KGAX$&N2KzCm^
z>7ApF7m;pW>)`mc_f1A%z`Xan2_F$>cE!TuGy@QBE>K5-uihr0wCU6j1DdC}%1TCF
z%x3zEb{#lLm9r2Hl0wqUsd>CQNb>5qS9AcXF*~)T7I~OV4qxD+6(vo!GQ0FTsVi=(
z+zi(=5PqP?O?8>*9Q*|c_w!Dx2<9_P=gw?YavW**O_~<rK&2BSpf}X?l>PuQa2sPR
zebEJ=?=Mpk>MNI?`O_X+nBkDNIm-@QRc%j!*kBi;boH@#xYXq(awR;!FUxmc4_7E|
zDu^x1IJKy}jyq*g6GQ>esT($)zv*sNYqJS^EBy>vXI1eq#@TlVM13H0(K~$J<G(x_
zeEEkKfZ=vH;`X!mlkndYiQC=YSa#gnT1Ma$SFuD9BQdQzHX|h=RX;12Kq?7hL#e9S
z5$j(KvCTPp!*Zy9ZjS&Of|cv)L)P8(*|&3FE}83^27&qqoavLKP$Q|*^|PDK{i0pl
z<x6HZ^EkbF!i48;mtR=G8)t;T%Ri>YyDT*V(nt@AwS<lVyzv=>S09+@&XYTPjGF;g
zGkyx2fIpcc2an_EuQNKv2S2mje&Q#xlb`^*<k?B}YM2YWHRf@TcBj+cf$#nYr%n*!
zu=rpiJ{_>`dwYFJM4IIZz5(`!P(U-#GqSYzg<ed31vZec;lB>?`jsQOJcw1u)9k}+
zl#I`~+s7!vXcC~5^Ytlb)zRQ3B83^y>#WS5An*L?_&r?X(E&J4{jL#>+)QPm-^|q(
zQNM8{8v^1eoXbwp6eEz$e;9=zXA(3p`1yneU}ylVLDOKMY>LG~uM*$l`l0jGB`0mS
z&lAn+<=7=g-L)g0)N}ZGGpp97t<gG+4KJ+SO^-X8C-+m#<Jc9FkDdub>}T190-@uR
z-Hp2<`vn<2oG>*2P+Ycg4QOhKRC0QQKvo)8s$ki1@^wI*-llWQ;m7R)a?}+1S!!Qv
zxHHYLJUx(4_C8H-bTRuHZ_5D;>jcx2lF&HX_}2A|>U|4xMyYJIRZkkQE7BHC@`6-<
ze$${yrT+*Zrp_-KLXsN1Lt&UzU;SbzNom*80&BG5z}4glY0oT|f;EX6LPmiyX;T^<
z0zsW$HRjXVO(89J4oeePUY9=ULb9>v1s9q>N-h5_EnHgG;4WzAwvVYc$GmE~g}z=p
zw<SV7y^?K9JS>*cb*)$?L2pt`x0CE*i?pJqCX!8U-ZvL_@b<-`UYf`4*N-_oDf{@l
z_g{|NHwB-5T;Jw5qoFx*8BiUEv1umo(ioXq+ph$kz)PAcmabdy{Xh3BZXeM|*{$~g
zjQ51``LGGC&=?ln=p!$FI&?%k6;0adXFkDqg!=Uau^M07bCe=VQOb^uaP*W@kTMeU
z4=!m7Xc7F{8rUja*`-|yxu&NdWAIu1!dtBisIwKLN0SVAol+{Wl}SVv4u_+LO~otQ
z$DS5NW*@K_=M>vM{VX@apX4i)GSW8u{S&KFg@>=BYi7o&@Ia6oRG$_*!~s?%wC;;z
z-|6oiP7M4*^bs)E?RLII5K}jVln4X>U`(07Hbn6>fPgASd|qEwM#hT7)7iNb!UVue
z7_Xq<`7`w7E`9R&{d0V;Up56v1faTIh^za;c0VK%n3>gW6>4{>??d%Bk6(qQ^p-C@
zFIj&ByVD)Kq*&DXo4443gK&&l0R#u*t#$skI`Qr9{+TnEw5}W<y7|9(bb1FN8Xu}v
zEegA3)p>=wSeT7&9t$bRsI7l967*jAzduNQ;YWKYZ+L`TOK{3sF)%U0NKNd^>xl2v
z<?HB;h|m}A8tSZK)Fxd3vG(3?WQz8Fx7E?Tf4XA<pH(5{8}vn&)xh}$dODcyi&{Fa
zDGo56txxgDlht~c&1m1!S4BR%?q)&(NWGCJyKv2P3Xe>cNeMxi(xiH4I3GT02fhso
zUN5(Bxz~X)b=A<Lhf!_7)S74y4=R?E3tv1%Q6mo1W~X(TGlFRvgAX&4J7YYcP%TFd
z2X@*bE-A#SRIcEZ%2pjeIAH=?26NRy-FywnsJeUQ3q3kLK&L2)jrsM<Lx48nX>{X@
zd0x8vH=&AAp_rM6$uLcK!B24wbTfHF=m00Xa@lSr${eU52L5+wU*3OrM){%RhI_1S
z1?}D_C44eCp%r<<l+ttkbAkwTamR#nN~0OtsS3{{vi<yqI-`E{PX7KbL3_bbQ!uM=
zG_@=Rct##T<6tm!8@;yx3PK35$Qs?=j$}v#J4TLX5IZ`+-F}zm1>YX^%6<~YOZWZz
zD`@!Vr$iIKUIvCVyV|IA4j%ejoQGI_TG{pu2SMCLo39NeK|IaE%3RF1r(-hb{g;bh
zJtsa3H`~1Ye4EE()vF+{L)er{AerkmwHh{A_${H1ZrO79?;h#DtH6Kezzzx)uiP&G
zZfE9V){wrATa$xkf~rfoxwBcZk^x%OyYMosWVx41O<|fv{Idi*h76z$)b(l6+7?)9
z_MKXrmQGEaL<xBFPcxj4=9v_}Z|__4k(RWZcv%<T<QsuR4u;R*ME*Q|z6W4ZZ(EFC
z02|!i!Sj4B=y?;Hsga2@TiNGno;Mk(jm3_!Jm#YxQp(njrM_B<BiF*`EQguTZgSM&
ziq+GvdW0ogQ8u+Kg4@WetR{v3oMDkOP5mcc55d<gMfIHcD1fx$a{H4IzzLE$=6n-v
zyL8t_iKK*yphEy=il%&7&VC->Ej?+XpE70@;B5(1Wud+tMq+_<EXG`9InfBwD!EFt
z(#NGdJH5QmrfGEz3}}A(AMw>h@1eD%7er!m${IjeO;8m2^6F7;Zf%`k{mQkuOt!)u
z_^hT4TaW^}^=&}FNhx8fZabSsX5}Z10t1G_%1=P*rd>LPipn+^S9ohAR$VonzEYVK
z2ys7xBy?bV80cE^;_JWSB&lFWXy7~6DUA=C@SXG8nJ0-mtHyF#+!R~Ic?eqDCr@Ov
zMT%)!FZ5gmT0s{vTGu|->lhC=@;k^FO%l6EHg1n=2<KY?;n~9T`ObZD6W}?!>>e__
zu$~Pkm!spx+LxY=3l|Ca>%<OxW!VSkeo<%%GzXQAs#@@CqGEVh7!(L9m%DrtHo||h
zR8w6C1A)KXDFzzH7eX`|7*fA~@FFJGFG&yHiaQQ!>(WbL0fg-r|4L<?kz|+fx>X9M
zRNE^78?~X^y|#!JCNSUXNl8_fvid}QT35{>0kV`<k!GA`iXX`8eRx2FlW4xkc!#?^
zsIsMwnHA_40YV(73ZXqg>11}HSKqKHT#Psm=^X)DwKEsY;n7oE^%d-NBdmZU3WDLt
zw0UB_yBut(geHc7ih#XX??P5P79-4v20|_71oL0-lw(?O)l6Ou-G!$B7f}PJenyUM
z3P&~#Xmb-V+$?Wlyb2R>Mr!!6TDiF;QZe6aMPY>0-x!Us6|LJH7WvXTxEj*i738$a
zb4+yci66$0DxjXP=jD~uO|=5n+4v!6w$0xhtKR`U&p)rh(Nq;7rPlA#_Vc+;1LL9a
z*Fg8zOrMF9Q=#n{Xd6U6J)6g->gMgZx7L&~Y;pvHP+kx#UPVqsPa?yziaT(whC@A|
zzA3M9MqRHBz5g=B$;%VlH|9djIKq~PF9b2aiMz9vvlZJ~$Lkj|q!Ab6w=nL$V<DBC
z?&zj#m*(DA$|dIlgTxzBrfzdApvJPljoa(8W5Om7wgcWJjD8I13G&NY{6x2i8zygR
zFBXH5a#L^g?CF`b;v$nuRlI$(VO-ws64(6W1|4Cv3h#)0f6Cx^fiTRCToo_AoTFio
zUx6gY!+?5Cw|*ACv1Z1l_sUio970*uPO_!<OP8r~QDlNu_;1*Ju)?<OR*oVIjC5Cg
z5GK*3g2Uy6n--o$w#xb%PWjn${`>#w*>fNt$Y+Q73YkkxOf{W`F8l|A9Jk&x5_rV0
zXVd7~>l*2i|MX9>6~};)Y*snwooeo5;_KM42FbC*rBHeGi4&T(Tm3s|;*k?BqvP!8
z;E`L&TlNk+7i+Y8{OjukP=sbrXEWdO_xCF1xSa+*@OT{!twDqUuoV2#U$%@uh?1v@
zywh&SO;lIUbc~3uA?9S6-1|a0w;Vy3clWpb2}&qi`96K}`8G=MS`cWdlj(ajU`@8B
zx0$uzBb(>g^L<wWl70SwpoJI;K@R(2>~(8t2YJf45k=3q0-9XY))$vg`#I${fA(Au
z8leC{)S~?|9kp;^T?8pQNP|B&`8r=niuI5!x_Ww{s~Icm>7}?!mJ%NO#DG`{P^5K3
zk4w!ELLumJSLMXh@qtWvbPWRnXuMoyU&1gM$5JYTKiv&OprN^*Yar!BN>0n>JUCWL
zZio+=8aR?e?}6#cpOL4{Ooktza-iK;u`=Og-?1IADKy>^NScO?m(0e5I~E%JX2CD$
zABtu5iw9sP49KN&;tB}BwbrIXaL2c4*o4bduy^iebKvn$t6~K-P-hhxiaGFk<?coG
z`dat6Z2iU)e-XO$>wN1f)oNI{qipBuVL!m<{@4@9cgzK*1QFJS0V$rnQY4W!8YYa;
zGr1}WtqvkOxWpJ(_9h-4*=&#9tpn5jz8e>wc*y|}+99e4rzI|baxU&dDt9T$C5N8R
zgr1+#Y{u9X6c{z7O_KojAPBZZdoy3eqagj1Md1TY9tkcKy5URz%y=T}(_v*MVH_<B
zi&B?ZY!&t&yW45CDOm<BUr_h7ihS^$AYc}JdT6U?X>dC|fC;@rR8%XeYjw;r!mqc-
z_c~S#hu)uj0sY=(Sskk;C<S70A5qIx#dMEv6h}U!K5A>-C3xhYe;?Tjx!rqi?B#4t
z?`mAMP{rk`Iq0RodN3Jc054~to>LU3+Z}fQ1*p_JlO)l=b#%#A<cMi0;WAl`YhlVB
z&)2hm&>ZsS8n>~AVJ@d$WT<lyTrNt9i-eubP+kJs4)%IwlKGDv_R+s@3y$edKeS_O
z{&JKD$2i!dd;~;tWvj1IU~6buT&`gb)|XEqKeinh?7s>~m46o^F-vfmZd1l+my#xt
zI<XZGTeRRa`~Sq52(4#k$_N+)V*Wwur1Cw~Gf$upwxb`}XalqWP?B#-kG>A<Q3=h9
zDgNP0q~@iUevbVc^7tW1ECu*RcO5fFM<F|%cBKr$MO~epMb+?s8KbrnBhMg3NK9(X
zw1m*Lc+j{F15ioRojeJ>iG^34jO>{uT?02$O;nHWhcbI4XuP(01kgoW4jK{D-?Jmc
z{njjoal6C^=PC6@u}+TyzJ%!(5}-NpZqKd2VxlDN-LQDSs0KCEe-U>{UH029T7Lm{
ziKYi=KhB9E%z4=}tLMoG^9^}nR+(?r%*e+5l1F%6I<*G4?n!9zV;^gmN)brXnPw|<
zf=-Wp7O+*<G$Va-6om%qRZ2Vn$8?$=qBPlkm3z!}L2v90y%<M*@%JrG#|~A-YBsnL
zvLdE<Y@WF#67q8>O!!11YjR5?u$P{e>^?Py5-nw2I9rU}Vce{9$3zKe|2zMuOKrMC
za4}I)!RV7CWicBVsK;>avx5vXmDS3SRhhpeQ~P+SYnL?TWAz*6Ms_cM5Gq{9rX}cm
zh8y@?RNvrJS0LeiH4iU}AuwA;rGPo(_TbbrqJRGJ_H5wva7cRxR2O`FS}5?@sh)?2
zetv$@HP$;GuHRNx1Wxl+cbWi~TxS*enLZeM+1512UI_tQ#TF83Dz(~64qv6Q!^5l=
zy6=00E$!{GNT3mE;15f{unvII(ISB0A}Rm=-2Z3r?1^^&F(kHOD^}3kRw&Im_5{N?
zJBx7GO{1W8q%aOxwhl~0J2$M@^01`X0n!sr=x|br```2hu5U54ume~&h9#*>)ZpJ-
zxk5lEhvTiQ^F4Bb|6iY>miF%W#=O^IeX+GD;Ju0flWHNr>#m#%Lk(N6z3T6jqKcOT
zaGx@))5^#7@#krpm{{6%g~62bT%*frpxJxqM2*V4Pn*-J_9#F)Czlx{QZ|Jo{sl9v
zxD-+kp>Hah{Q9>4oL**q_(1`_6y{WJq-pr-ian0IOfLc)0b;c0k}WdVnN+Nx#}m`y
zXDnm8w^K_5D)U%=GVR^Qk^%yIuHr;OXc?m<s;R(8O(^#;u9ub?K!-+vO4RO+$B5|!
z`u=Y6Kh4N%kl-;fGbdZ}+or736h^GCQ-gr;0;xf#r%eYD;P*;-f7!q+DcZ<}#gouV
zYcSV68&eBP0o({I$o(w4#*(e&Q;(SAjx&jYedD=KeZ!wW4gq(ehDs2YBfhSRI2fr@
zt94=5kAsvvZJXLx9sv0&^haf~<Q=ZY>U<X@XzuEYJNI=9tSO;Mt1JMBvV00D0fI0_
zu6JO8;Ll4xeqtI=xnr0pN#Vc)s95-1udYvYzjw1zX~9E8W%QR`<_m@TmY{qyd)NC6
zdOACVofo07mZ;522K#qJ@@T*C48F%fZW{ol8M>A;PYRHV$Pl7NqlQ{#VWnlmmbfto
z@x%#2tJ2M-@pUzdK(VaDsd%=iD>h|K?m3t~ITUphsg9=XT2dqtZkyT+IvB47iL_J&
z&iP+5)u^Tz<L^O!*F`6EanmhvGi%O{$%k*k{d?j@1PH%lXUl4+SS#J%6&$_*thH_?
zsW0?UBF=)dTl2fRFFm0~d6&%wQ#d2CXr5ST4&r)xpnVTi_iC~3hVajHz+70x4GkW;
z5~~yMlwWr?HK_W&KdG*cUsXKGI+`)3zMLulz~;`RaAxG}84d27I`b5^^l*rI1=y!I
z#B^ZQZ~!3StWIMJK5i=2cu_VLeXlP(-6?sHG>PD0aC@Fmcf<qrdt{_8`^1rJn%JCH
z1m{A|a)K8}u58^hR|#DFIzq4|hjLbehIA>6NZhrQtXX>~N{nrc#|>uX9EnnsG+FvN
zld$|Gd|jO>fZmca-7Rs0wJNWTB&?bEUK%DAR%1;c>pPDOHL1W{%P+!>P=;;%fttAN
zJXfxeY0Hyy!62j4kj1;jmfe%&94yHp!R$w?n6!{my&PrBL!1_VjTBkD*v(n_I<YYr
zv79k&p{R%FU66=V!(wLt5azKQ${8rA!tbb&%&!SSPcJ|EcQ~kJO*8%4@NFO$JKlpm
z%r-~t>(4#APCY|osZ1hh6trhOBO_B=pX8c4*;S}&9H2se;sC=4PtrxQ7BkgKqlJm`
ze%HsUJXB+rsggnd@v&H62Z643CYsmb^x%D~Vs!0g^ZUiCuCq3q^fCFTa>*`D67sOp
zC_a+J7N!S#%O=qSIdoIzfwYEDi+t9Ss`DaofM3s1QD4r|VED3~DpT)Z7Z#y6_8~D6
zkWrSkvSh=JP~z-(UeaDrs-abmn0`J+cX08ki-iNv;k0OwX*LQvarcIKg+hX$@bkyG
zN|*mN0B72KxK!0FcS-6vFOHX}3K2k4r;tr0p)Oj!P^(@MYYzxwC&W`Ai6r@<!oNmo
z{I@<+&%j6u>Acu(NmeD<ioaLTSJ&t<GVtE3s!hy3mKNq*xiSH%1XZVTk8GX+FoXg_
zRKQx$)8%N$+rXp{^DULK%z~r9`E-Ul;H6y0rZY$O6ed`GCMU!^uXAq>rFYHmr8jWF
zgwO0|gubQD%*O}S92H<IPCx_@w0a(;<ubNUfMEyOGk}ezj?J4?vbbay<ytzfd@3*L
zJerpDcr;Ns8-G?_|Isgo7S*)u{trz-%zcjVq>c_?OA7dk*xJ$;@fb?^GYLTi<y^Si
zQmwo1n^haC%&MsTGrYv#6P3S0dfF~mWsB1qllNsNV$oY=J8l_b1r!K=Ir9<NpXPat
z(cEft8O;(*Z0hAw!v3BAdmQvH_W7ZJ>v=c$4{v*$xPj=!f5kz4w$Z<3=jN)io$bJ+
zNv`vAbFXjaM&Mg!T%{mFHip!#_z%bZiW;Q4g{Be8dzj!tbH$PISf}FnVm^{&s`oe<
ziIS<uKW^4b<rNH!Wi3GWqoItvIv)SblRf<_+M;lK1ctD!ep7pY;(4ZF_ixBdmch`8
zC5N$Bd9^fZOUwI^rS>23$7wzTccd>p<#$uh!nE4{%5@Dugz^Bn(w&>A9O~Ss5_$3~
zA}Gz>DtYC1-E$%(NQ<>52cDRBDmL#@b@}!uU%Tgu<-ttKN2sqXx!g+HG3|RD>7Bq)
zYYLU#1c144!-JJ)@VXJ=z6;Ycw^xWp2(B)TESgdPQXL3Wu_e#YDEFZ8$0Lf~++LIy
zINh|-_h8!3f^wCN^p7N8<bZg=$~9u&)(2=d?{!EYA4fzffgPCorzCjcMO<hYP2RD`
zDof?}Qf!LPcV2-9LYCtF!s<zEHT7GVJfhLEuh}Syr`#NOx}@Ds3F1$WgQ)CyFuE4T
zWbXPC*K#24OOrfnjjWG)j!L+;-pXt>o>rMAH0c(H{esUAq|J+#T`Lw>>`Le)<tG;f
zclOC0+vs9QLMi(Vt~EX7d)g3V$BupkK<!)d&TM7jkttyd|Ng0QWR?31K9{@fX6i^<
zFgf-P9=KfI2*z1)DbdZ$&_6Jdp_D#k$>r}dhCg`$3b|+LG}rW|0!^LBCIw@Sk8?+R
z!xJBNJ4}0}j#dv$0vk${MB?VleHm5#xKIt=PjB@usYa&!`Q*wWf|!;pYt~-G>FNtB
zs#n4S>z|`jeD;zz%++dmbxpqrz-c|=8~2QOEhS__H^#FqWRGixQwKhSE<b<$0p3!F
z&%*dDAiz~(#z#`5Ai?1RD$K6_t%y+$k~;9wtZ@Fcnse^Ur=+W|d-D2lywmNQh;2Wq
zXKY-tZr9x5iL#&i`YVhDw3schAxA9^io33L%II|x2PCvi79(@60uCZtT^v(_WPIFR
z<4h(Gu!ORZ6R>7D&nd#3!}lkC3(Zg*PScV?o69{Sm8s;Mg7)2`q;Xzl5#yyLdp}{*
zMdM@u?dZ~@m>Uhol-U(F`9<yHy1*+cyZF7KmxikJGOW*i%>Ak%r=j%Fz5ElefP+b!
zl{a6`lceR%Y$fC5=RaEttgde}mND|j8^XD}j*}h~_D<_oI8Q%TRUXd_oQ@*0pHzRP
zjSvL?+}&q+G~2T@Fw#4{KNzBRwz)YAyyj?&XL~IkIq0$Sz&BxW%_Y3HGbiz4jLG_*
zrfa{f4Zs7|YA&ZO#EtF1cTT`Z7P>rP<V&7SLnw9GlMV(JQ7K1J+_k;}wM|Elt^p8x
z1`1k9xU1v~`NBaepUS0ubSb;O-3W{>1Id_QO%C<^Y<l{Ka^t5ejUg$T=~DQxn;}GE
zGIGPIiTRc5Hq9=9`E^XYn>elu)f4LPc=|I;lP^##{@<dlC`-6-rA}XmO$ip`yThH|
z?!iM>%~IMK?-v}D)mICQ=<(uA725BDYWeM)4ASE>TS!{z8X1KB%CUS2O@Mwu4G*sj
zYI`_<wA4F%FAi0zC9$W{XI#PQaNHk7rl?yc0VY)gB8#;h2OKrmZ#Y9X5*U1uF!hXn
zi;c{lbAIzu2q{$LND?J%6tk`X@(vN+wpUz7Uj~w`L^`o5GNYM$?Z)hv)l^C~|F0*q
zAXaa`>g@Z8d`6f8d(Lw$IsR$qt8x9mhGk)5QI94$KUS|{E#Alr1Ae2yDPYK_wS*0n
z)L@n9uCKghk8qAML>WNU*;g8v>*_jS!3E4Cy7qGI>!fM%>sS(Dr&_|g{Mah{mWL&|
zbla{g0CtZZNFo!DsTtCMimSi{@5zE)bkbb;%i&@hb|g%ib(1+r?yx7>S@MiiPepC3
z>lmQ`LSlc%F&g1@u*aZ%nJ67yTSbkMkxl|{(nxh^Era>_y%nm6cLU>z&!#*j_KRvp
z4+YA#iB~ozV62qHe#P4tJUWWMCZ#J+7Vqj}5)sug{-kG7Rg+yZOOZ}K|BNs~ILF-z
zV&@`GF;-E>b3EiS8yPVR1Ls(c${fT1-fT}7_98k%OePFv-62I5R`wo%DRbu;D2CDX
zvK}#G<B(G-gDBo3+h8g%aRAk<R1PSkr@n2z%su^RU|phkJfD=(*s%WmuQO&-8Y&Iz
zOW{&$pE9*;hi!5=<`vveDWd-@ozmOYUpx~5Abv|NZ18os7aKwMkMSiLp9NV@fv=@u
zliqo2Je;UASH<E!#T#xwHS@tcur!jb^uBf<kQEK#qe#D_&_rhG-yg|EKI)7>;ifNB
zMQ=0lCls}cx$yV>;%<j)d*7Buis>5xu<6<Cd?0{F9wRS!h$>SLqWSfvDxuT<6(vRb
zWc8ZszV154Qc8Izt_iRxZ45p6>CfXGKV(gFk;h3K@i{kJ>6^z}q?B89KlP@HoBAuL
zX*CdZz}X+`8<u{?3R8gpl%sEBeyvQ$S9->kP3G{UE@%d-mZDO%j?@b8pHjkzuDtmY
zc;$<(g+Tt&bqBz&p|VYQ#P9OG8_3P8hr3>O<OLKp`EKADh7I)v6DB786cz|H532yq
zoEyhX$KH9MV#1?mKFVAn_O?|kV}@@VGjre=4@Yl5v4!r6r66}y`O_CmkW#nX9c1qI
zds=$&3Y;|3m+mZDYZixmc2{kQvY8g3CUpS}eY@8q#hDOo@F@<)C_@%TFNFdCEE18I
zTNZDael1@-GRyq?of8nTr6b>|&KF!?{=#*8PUBCAC$AUyj*wHxuVHA_r)+=`d7DPQ
zx?h=%B9l;8>|gV>T2M!aV6;)IBAId##4qFa-2UJ%Rtv{+w4F;y6Q#%VHJVrrlNW3=
zwesA(^2o2(P?9aYOqrd*2q*PnCgaqS%PA+QRV$XHL++@O;%32`LkB53Vx$O6ZXhm!
z78`l%66`ba%ZXKF%d|AXDQD$RN{m=-9xRTj$;a8(TYUR2_3||%-@nQ+0={VP0<(Y4
z>Z5-M)ix<&4B{p=Mnn&0ldM{YspMo=<H(|x+V=||K@RzIfc8_PuCXrfNggK{xn=&f
zNd%Bx9$w#80hl!zr`N4GJwv@*D^48{k||-cvQwZCDVd^0P<t@(jT{}uIQ7jF_J`Yd
z%iu>q0DVSeV*O=1ksu*dVC8Lk0#@QwXj-`-I!xAH+e=ERnbW=RL7|mM;TnI+`Pqx?
z;N%2kvIXQ}PsX!+UwGN!N|tCFXL?jqZ5V;>!(fg3$&q3)EkJiS@v-z6;OFwtPej&c
zb4xbTvMu{u;`iAS7SHGjY((pwQwD$_AxsHXCV;4ADX37{X?PjI&uMRsZ7$5W;Y;1T
zg!jHn4=SGZjRmIeL+6S4xo-+tU4M`rcS;gV+18WakQ*s)*B^!b7yI(c@!A^>aTR|1
zBz--GDxrv`U&^{zkt{Mx1o4;&3TE!7bF9g?W*d79>IjOvo{1wxEL@B|yB+rF@`y8C
z&jb1Nu|gZ4lCa2`i|=9t!zf)BAF=ZbWHt=Z#A-jS7mgo3x6qUrT;o-qxq3A?<PW!T
zQdGa|p?~f3;lAY~6K;LYC)63!ml1tAnG2OSN67m@gy3%=IMfT9D1Bh~BYa3_FZsz!
z28Sm<QTyviU3`O&@S`lgI0r-GM0dB?uETF#Tyw<ymU;ZKW=fX@CV>80zE!(0#_ctW
z`JiCb-S{gGg84c!<68(AW~LIT%t>)kBMTcwh61T@5RfcInF-9}x!^RrUh$b$M&D}+
zx%he;D`qQo`MwD32FUgvuOk_v)5FNqwbAq6GOH27uVz?EY1P8-dOSX)Z8OUk)7eLH
zBjjJ&_pU-gYN_!9U3D0afQoL{D6RsimRo$|1q~=Zw<hUjJoxiZTkRxbI1xY*kTVxW
z%m}}xRz?CZgJCUf{cKJq^|evESq>FI6C#mPW_HY{GPa*weh6K{Z453loy&XTjJ1_u
z)1gtX&B*U8Ip#Wj%gP_cp)TK+gn<jCaRiJx6DH~@r-VLQqV#Ml)oh<uu1j9N^`n^*
zuz~VSl5x7NKt$`=>5d=~tPiT4ESDK)xs*B2_f`mBiR^PX55MGdSrm1C`KkS%Hl5}f
z83K6%9(l2aeu`Q>i(P$vQ{DnrTtrKb@La^qH8o=wSmDK(aJpX+1)1ZPWe5K!ot-A7
z(&9CP%~KQ!P|)@Mt)edE8J_;Tj~rAXnY<17HpBeG&17kw*)o}vC;<aO{>EgQX&%r?
zjWv?TG7LIdyX#J@W<6lT+-rUd#G4qh#swUhF3a9rFphpOqj~b$x(jIHYNq<5MMg5Q
zG6U1E`DpCPY1-Z&Ux`X0?t1RhX%;7b_Cn@A6Uyw1I(ze3xWU8IGK3pE!l<_`Rk)5P
zG`F{>jpWoA&vDsRRHX>3o$!<gBPEHiIE+aqmtgw72209rC^bNdWvm}>B7sh2&rJ_{
zIHWGkzQZ7i+}n%^%U$@f1FS+gz%PZ3`HfWU07d|yO`l7O?nFKKi(xSQX>q=9aqX@{
znmkOMz-<_Cj<Lm^A`)Y`Pcv`>X?Zo)Vj19i95UMQ0Dy2wL@8VWO8>B`yW-OCFzwC$
zz3XP6>?B-RA4@hG>I8XsEj)f73RQ8{GD5whxUqIZ@V)<`KrWap&XkYOd8q|5mI_5^
z#XaQQ7su!FV0JocH&iHMlzrP5-sTtu7~h);<R8J4-e|`v_bMfq1O76g>iPS&2-U1Y
z%ZSJA?96=eK#I4qXyf*1PK<d!OYjD*T*Z%DRE)oy9(qqBgzgGeaWqKgO-Z!b+qwY1
zl51XC(VklxVcb29jj$ZJjkO6Fi7vcFr%ro=q1{OO9@?ZkV6cKypyH3cRIsea?o2#r
zSEgd($lO`bq={RYa`CMl`&$8HHNiN+S4{9%8~IegLweC>Mk5#8kwM|r*2=YLv}1h0
zNf>{g|F?z7-V5;<%Qs9zX3=|F>aBCFId&w7?x$U-0)L@ly|)eFA^pBb_?ewhI9tOO
zs~xv3L6`>#fo3Zm(V;ivK#tOLLnx1ZxBw&i#zeYDEV%vo7fFkA_P2z03YaK?8X5c0
z`yBuyaXUg-f3EI84coEzY5SWs#mVTsJL9fW`pB>_r#(9TX&x_|+xkmdEYO3hp(ogB
z4#{R2Ub)jzMTgY+l%h<nlaA&oiqz)moR-9CxF-gY&Vmp!<gx<ublrHxW~Cu>fKXNY
zLAZ7)d^~5^MdvE>9dn?!!UL$Uh`hTbs%avKTW_EYqrUHpMB=o%+KSw8IqN)kP$%8{
z?DbDn_vWZs9-qGAsioi^MNg><EC&h4paHzPw;g+719I5v;B@Wz{BKcIL_WC6k#<6K
z-0A_=?%CsMqTxyGdV+{Le&8H3kz5YGE+bv+iodqK4Hw0PN3o$+`gW7eF-3*lE7zi1
zOo@AVk8^Bl!_NLPI2zt6yqzbahHhz#tehfwp{%2+;+9Bvgsd!0Rf{lQmb{z>4SI!%
zS{3SW;}Rz5yn8;O9evWWmYQ9_(~#r}H99_jN_0bPy@qAaz9BYySWsP{%ZUD5&6;`M
zDz>J^P#ie}q+pMGq?ksz3>VTQ)q8wYmgelNoYBjb25Ov&*RayoFXbGy5GbkA;LM9|
zEJd~PUvq8^05f@uuk8t<r4)R1w~7RV`SA{lsO`|!M;+&!0?JnX>k!{Ajyv<yqmxji
z;o~acaeN}uGg{BJ*~=#`FziuHf2>9s1Z`WK=*8gI{O13>dN&zVvVs16pI99yg6Z>P
z>qdZL0kv`(*bZur{9_6ob4Ex<Gos0}?Zq##rVeO6GR9cBe(ifrU*$eQmFNE!k%C3(
zFu$97`XnF|R5!KL>Q{4&#JK-9pl6O)8brsy@6|AQWTr?Nqq+jS+x@mo>41<kZ)r%@
zyS*7N<9~G4>Q+loO_eGqPQrcfkBC)6_{TBqSjX(V`OERc7<Dm$gur}~Y}4)e5#Pa|
zH%pWeLQRF!FfHlQ`e9AL4#V%`8oz2SU~@c1dF&#NGx-!X#<VE6vh(K1pvTyUOhr1w
z=T5Dl(MgZ1{PsPD0$?B5PRPk?Wn|45u{smm($3_}bGi=jyFTTFGKD*XnJtU&sbncJ
zDaKM$luK{gzR8EU=ibU2@Or(G3Ow80mKW`RRZSyAtxsN|MS`Ab8PXx0*AAkk_R%i(
zm^d`atf}#t^*M}T+?Oz+k0nazfGP)Nvc}RN7;d1{5++9Wy)6g^wYD{juZcl;gpL80
z;u&v_38GQ(E!yZqSj)F8PqR3{BDoRm9TuutsbT%GmO$zG{0!LALvKEy<3<_sRma={
zDhkdq;mFl6QmhGAH8Fpa&a;s{f~@n5G!YDW7j>|H{VTs&EMSsreHR75$n(2*@x;m5
z>00YXM*UdGkSGvX1kRlC78UUQaw@++pIa9Es$$Qcrs70HR397o6=Gb|(&8E&<=faB
zxV=1j*Ck^bfz`VyNstRs!^lZ_+!MIZKZuLEia|hpLZTk03!Z!5BF4~ods|a4_^W)t
z!MuW!N7mF<H;cC5++>j4Hrl}bbC2x{zC*Z&N<6ptnjY4|Mob>E9b=X>w_Yp-j(m1r
z?fVnL&>SacF#Bp#4TJ03DPeRYz5HRMe6mBSGoBsw%I;5;wSR3BbjSi6rAmdQI<2&^
z&aBMsHwH0YXx(_v<lvAk63*fdQzZ`-Kd%G{$tbXL5%}NlbUyU`PvadF8HJUq3W;9i
z_ak?I2JD^OzkTiY8)`tc!9z6*6M6a#v-DhMyX~&QZ=EAg5}LsKO*UToxyVE0wJG*R
zYUmDoE|emuZ0$XIC^l4pRC?`Wn$%#~E<ALo8e(SFw^WMW*s?SvDir!aT+)57if!l}
zp<-k-Y~fSQ9*c(X;XE8>P*20xrhBVz-!Eib(Dqa}gzgl^gjW$f(CE8AN+97JqVIVG
zO%Ra>{NDKsThrk$4}{>Z59iAd0BFEsk^a3w;LpYb9D<0j>o?NI!B^OUryHkY&x*ED
z6010p!-!LVDi6DFtltN`6egU&4VZpp?JI(7XlSIHYwAN7>!>A>s$RMV?ZH7#C?e3-
zccIZ^#PB1gd3??uEJWL|<a>IrwOQLSQR%vSSf^Vk)a8GDB^Ejt*X|_x>|~;b(HOaV
z&gl)cB$?>_i6??T)q)*=N}N(y1Xim9*WUL=E$S4oR<%rSw@0VGm~`EDOL3QOBCXDb
zY#RL7Xbg${&Y1o^;syn)J6o$AndTpg6u9d)qUeoKFSUyvUA^VVO~|S-*;F9ZPyUYx
z^yILfgE8!Adv3we<XLib+1P%PNRai6q@;`9Z^N8BJ5*swOYI}RCZM>u8d*Z9VcR&H
z_Tml33EQxz5Hg%K#vpA@ad+J-p)vtaZ~}Cx97CyG2Osfg=)FGReZsZ!)@>;pV`9XY
zx<0d+THyz-pd{-OFF#(tj{NxK61*7u0=YoH*JtaTz>Gx>hd@HQ@MUS*p`xK7=Kb9Y
zKMiEdba=4n7x35vMNw6L2pS}x*B3mN88e-S-rH{-X#o#R_NQT~K8a^Xdh&U^9<7fn
zYWR85Z!SjBAG-2Oz!1RA&Y{fxc~U-HCKoEM5Am2L@jkeJ99Mujzn|tcHd)YdUtL9|
ztbOha^|)GLYi8#nhiThuUO?V$d;d68ZOATOZ!g_|`<m~Qt4e|f-S8ALc^${?RmE6m
zgz{tI?-}o4xGfS4#n4Lv7CIV>#0h_(5#cw5wFGNz=C3kW5^-I|*6%)XQwkFv-H$K!
zNaqxfTWU!8trAyw!IIDl+zicYjPm(6%C{J57-1Xv-K8!QL&RUZfDe@rXw`OGv}rU{
z6ZqHf7)VQtc3a?;k8V+nLOWzC`+9JTH_RP&uu0^>H}W2WF5ETDhxU4LiinPG&17FO
zAn>hjwx#!Of4|UY0v=upf`oFoVPlrGL&Zn7m5zxW8dVJQGbqowKsOV2<C<{!1ZEA6
z;oFU%=n7FV3LW35W-9Ai$hxrLdPLeQ=?hqugD{tQI{3#TYv7Q7{V>Zp+8ysM@+EQb
zr$0jly;6B|sw9wLaB<NESCn&avlzrd*p*mX%^Ff8s<Euxv$5Lx!vPO5=nIa7Hqw`s
zG$^81h|?Li@gmaQ?``#-E>JXU6`uxi1q^jV)4N@?-9ENptDiHmAcUP~$@@)0WLbNu
z_hEsM|0H}dZ2yYwrf(#^8%2x^#^Niw?}MNy{?#qjm8|fvVB#4ue_+;xDhM4;OQIf$
zm$ZC47Rpl<xodOStc{e_ab>06Ak#4lPr6q$Pqnn<2kejbTXx8Ko*E_W_D|18edJ=I
zoqd16(GL6&9!AfH=db<b5RecMa*|@27O_~HuG4+|mXPXTaOHeqA_L=cM3@B@1m^hS
z-@ml7t5N%ZMzJFRg2)1X6%PUdKl`E%agk6Sq4evVg+$6*Hor?R1el*dOTSp3X@NrZ
zb5#sE%t_7*Qr9w_-$2$U(Qq#$?<S>DIP#To9t5bQET%UQXO=qs4xWlqNmQs#iI&zZ
z3wWPSAmGwb`NF@p;&3tLOym9K8|Qe{)<I<inNtssaNzd+*{A*auQpEgrKfWsc2*3P
z52J-~6dn?WWrFmhP*IvlJ_LQYvLm<28&jAuY}i{#_L^>F(_10BdBVO`*pfM#+U--#
zb&{HS_mbTlGZfr>oj;W#Z`+M-Wgkw1C|@z*y*vBqwR6w)=C$m2Ei(FRjzHg7FIJKn
zW4CbSW$cX)C5CK58_<!GQNE^~VkcA2%kxi%@k!cXNX)?6+xK2)&;~)1gnnK<Iq(aW
z4^^<>>T-5Y@_ruXEW3F76CRAvMk~OoO5%4!{ne3V@GF&>p6>|kI`LPQt*=Q~L+nNT
zQiHq_CqCaP3n5e?&<92h)4CD(4_p26A{wV8VT{JW8O-&ohEun;)!{<Cx1Qorb6Y^<
zqJSc9>EY;&M8d_(YjZJ3Daz>|`WDW$-otNv9k}n?Zm)ly@wONp(I6{V5v7R9I?EPB
zQCn(`9{eU;%x+R!UuZW%Jabj;EZq{_F0_KJDajm2)KVS&$`%z9Bn3J>@4Gw4=Y!8m
zOchnXhg<|33kHR?aeV&iZsG0JfskJxa5IfqhjBC2T~b3)4q2|fmbaWfSeTfsKpjMY
z<rmcz=q<|5Gle|Q(C)lxW&OJPu`xW{ogQ<28iGMNhPmEY&f&|^m^RS{S-xh}D{T>X
zo;{^QVGrZr0C~`F)_~vlP2(W1I7zH{z*GQk?g@9cq-ze@mE~TH%{`G%@?-_maOjZl
z@cWZsox;TNTg>EBoKN?GwMg%XFN<|M7jc2oLb+Xcc*tZVyh&=_dIT65+&AJ=aQd5{
z6Jvj0q-cEskKj_V=$`a$uV3tI+#!D<tK!<l+EOe=F>`ql*dYROT-)+>Ho$-WEILAe
z91XDZprC{8FzNPwLT(#Kj74u}h9qO{VV6JfK(vD~G{o&K|D73`r1-c(e`Bsv#^~SA
z#aEYCbug{<(~advGT1kRp!l}P!xs##b`A|zg2wHm{V?6nC2~GY7r%1Er_668Z7Sd=
z<yx~)y`K(c${AWSuz+kQX%YM*Z>mh}Yl#oyH?JF2^++tTsXy##c(?3SdV*-pT*9sA
zF}zA2O+ihB;6`1*j6T7b{6d@$A0~C6dWBz|*ml$JwYo<bV>akni*fz@@fDwad3d}-
zUQFIytz(a$)hBj5xrx#sfUIA~P9_eLfk_;{Sk_Chtluk&J*#+()B%%3nYYXL4W*yE
z@9*W8qtd&-Jj#=TuNTb==t16$(4Ax&{;q`4%d~N}HoErXeM>)&`yw?1&K9Z;_nrcZ
znfgm@cezbh_$-oUAaZ_HWvDv~?ENUN+<rO-3X48nJ^#S97g0Op)YcbswcU_km(KB-
z#r+X6Q{Gi9LBknx<kzpiD!cy=TZ#=8IwOys{&hxFx$Lo;X5qW~gmq(X_9W$uXzRo?
zVF{%;nedOrM)Z~qCaKS_FHq>?busF%h7kvs4vmC+P_+irbw8hm3*;;(f9TI!(wiKR
z5p$Q2Z!GgObpvIhyuWL8#-WZdN+l4s$!=c%@)`^%>ip=JL6NlSuu89&voz;?{TxX^
zP`}y^c`$w<#pD$KAiI7PU=kYX<-T8yr53rAMc%T`rS|#6OwtK=l<lD|c*v*7A^@)(
ziv0J7+sp8%1g3gZ?AD%EDoI<)_k5-wV5^GaTP}0Pa3=W^!Am2Ewo)$&;?Xn$h8vO`
zX2lDVwmd1iXpJg%6o~89wwBdiFQN!P=Z>Tx|CJtcQivi8*1QrA^FT!w%HX&tck=Be
zxn0jwvi4o}ZJ|})P|slP#Fhn=k<N*PQ35qEVPW7ilD$##dRu#eniA54=&3gQ802*y
zg=At9h3l;7oi5mJ7C*BBCzKs=4jFZCG`vzb_JUG>wZDMHx<ufZO?5l`nX29Px!{R4
zj$PU<h$nRxg4v5=kA{&T60*im$KG}d=xen=fSA-uOM!GRBVcN=62nLP@apvH%CqKh
zAHMyGK|aAmr|7Fq3`G<|K74$UY&7ZeIK@rg6=c&al@_fl+L-@O3HasM_UK^Z=!;(>
zT!@+*INS72D<H0r$YOFeh7N~J?{ojc>Qmoz*CB&FhCPg%Szz0tHcn%>8}YUlfn)nr
zvD<7A8WKwR-5EAVvNvQszoOOt+eC(Z(gy52mGn<zcXzcl==|Bc{r*Vc-o}hp<z+y?
zSwze^W*xB)J($>Nm;6uR%NP45#&#3;iFY#z{<sjgH85mE3ZJzm1ts0V*p4t>eA9<0
z&&*N)E1&W2ZX7f%Mo)eE@32U{B)wOA<uMC)Q$8`TzLn1<Tb?3uDw(Hbq2%&sP~2%z
zdd6X)f;xf@<XHwf45C|Ukt3_YjWh9Wo^XQVZ;!)d-dbc6ZKZBu72~QZpbwOTs&;7A
z{>Y61bf29nelDkP6rUCQ1TnrNM^JHbBYC^o25(z5537(NbWnS4gC(d=RNP|pNed^F
zR<O%a!jVbQB#E+oB7}ItV!4->ZeiAKq$wcQdW|yUc)u7PjtuEyxO5qF-0q7|WW+g%
zSP!1Ep{OpocyiN%*%DrF(FYP$2L65WyMLJU{`Izae`OsQn>hHi{d!^`GPM8iLgg)l
zU{{CHt*+y$*rzyyZhY$R_(k#vYGj#msSE;XyVkZi{}88Qhi)DL=IphA8+|oSB(H+5
zlXzw+LBFgJactF#K;5Ioe*m38V!vb<hO{3dJ_Obk8Tz`@)~2n?nT}oc0L($Q4ZLwk
z+?YMZ^U?7F@;+jmJg&<O2L`69Q8LY5$`}IFIdlsVcGecaTSRDNfk&<db6NS>Z@Sdh
z*gNcr>gtBJC@&O%plH*M+WFo!?dX=f7*qNR(i|Y#tt^#D1@{5RXrZmz%KpwNMw(zs
z-G|>_<?E{hnkrcJ=RJmTL~tIZG5WzdBGIH@qMdI%hzFNQg_`i)36ZWblA>$MKwl_L
zRxZ-HC!S|!aGvyFO6WrK6eUQ8Ya6;%b1_p|GEtLq80yL)y(<P!p?CDGEE2$~6xprZ
zA;_iB<>wr_IEug>sH<v<+8J-<MUaMR&+>x+>K$s{=pyDdqIVl)VSUhM3{yq)f{td(
zZ!oQD=OSfGYsm$@GD*NWk1lp*_HGL^HE7ZuR4a!i!0rKXNLK7E1+~|m5!it2Vc;Fm
zRViT+pw|Zm*DGE4IcLP4@ZrZV$j}6`4P6BuL{Q-rD%8fIE8{GjGZb;Ve<&ex&Yl}5
zkj_E9^#&HJbZmX+t#$(d=x**^F{DK8x|7{77;NG103ZY=98CWm0`Ei9#F1rl2tHWW
zo(&AbEV`y04j$W!+gq`|)ktu_hH?-79?f*E!#M0&ZfaWWuZG*pCLIz%F4UC=3qtOO
z@&$V~Ni7$2?Y7L5Lf*REq1)X`fS?=j((L_!a8UPl_8aB{RUqezHCH}Q1_Y@B+2kNO
zQ&(MTMa~(e@G}4eth4O9XJDOIthJ+95>D$|uzIJKuF}*Zf_(vTU`q+q26H<X-2T_G
z#1E7HY@itNsEslFM;h$JHi!_O&<`h^&?%1Rv{|)5gcIk6WNky0s-n(y?+YzPjrDHY
z87D?vF$0Y5LTboVndj7cXF|R>h!Un1ArIZrj7qZ2*zEt>w-=xjc>uRmIy4(vEqqBk
z9|G?@@|qEnr&Sv~n4CB?PaV*vQ>q)vXcuwVVhT#xwOJeWwH+*g7?R}-Du24@lc%FI
z*MgLy+1}YD8v?7NHngT#B@w#{Gg3%4=~Gh$a3CK>vy->B!`^}pCjs6>`jBmNZ2O?y
zd&6q=I>Ew^yHOuiJ3Iu@g#9pzvgBq54#PKF6=(yNrQ`LN@A2>c_|Nd!XYZimP<H)=
zz-h<7`Q6{Z;pP^>Q&-h0|6Z+kBCx7>|Lz^uTJZeCkMYTOejT5E{!?7$dwleb@8O3(
z{Bzs<Y$y{!s|+OW;P0UwI35l>tT$9vWY2VKty_tCL$75=a)X&_dRNrUBzqqte?E2?
z_1-X8l*9J(-?S0EGn9DKK6+agmSshZem?-WpQ9HOdF$4bYBZ|Q03v`5jvO#bp}ud3
zY5%>dFsXghNTCBUex;hm^Y^W-L(EREN`s4d29T>6Yie)E7FV;Dg4Nrz*BPxr&D9pH
zNu6K=0A<vH*a4+V3M!bfN1UK#cB!i7oowdp=HTq_9zB2F)ua-=bu;JKQMoCae}%8L
zW;2YqJ-dt4yr4<rITy%S5b0t#ZK=&JPP+bcqj_X&h;cTci_!fGVXtrDFcgI6qXZt{
zPH1k4KA}QQTEQWLt~;K8o3x_W%FEe#(=fW7a&&>0*@ke`-qEVU1wK38c?97sO#!<i
z1-~yF$#IDjTbLCv&80aA9d(8sL;%H*ej;Qrw#L|tW|y}BO%!ADMnMT`Er^aaXrk6l
z?dgQAz3*MI92o8VZ3@(m(kiJibFvEr-OOjw)y!xf!6!J>CpHNj?rQl;54y7DLq*WQ
zF417E7Kjd8<yvozNSG6hQ=s9a5shq}kr~dh06xSBs59rhgEl3y<`N)%zreSEdExeA
zG7z#C1PWk*)?9)@3?{|sa4<uZ1j-vgiIaEUBo6{_(IZ5e4T*BB?W@fOVkdAK&2Z2$
zq+t5yjOZM^Fzh}$K`Se|b35`Tw%~}cX`;6VG#U?5@TiTFg-xTVy`$H<4{Y8kf&ktz
z?j#LHt}Fli-~+^J=fo+bQ~p9tr=geK8l;2|HMQ#=DUtyhV}i&4wX<BdaL;YH0?$J)
z9Go|OqQWF9t=W0V^RiUlsnztz+hDGn=Did$AZmJDy~UbW#5iDHEJmgA*TJ3xvF~T2
zfm>_)pZCjreXYT+i_*nf+s-iyU>|1Psa5}#G;&q3!Hwa9{AQn`6<~BO&lC@vmG-|J
zwIQDL0|4;>g2orbv?FShEZvGi=cx^C%u=&}za?WFBUGA2v;@;M84U!7m;yqK=(WKK
zwON}`eEWX4*b#22f@DxXH$?u8YGzhiYfM|;0$5`TW<O^#Jk%lQ70FVqRXeh`@4@Iw
z{)~;@?urWjUaj+TX|d5F-72D|CE7qYcuyx>Gx~WO;JkBanGa|04Vo;D1Y)i*9(biS
zJkB#>aLAQEQ#xNsdDL1sn3>U|R%-T!VZ!_OGroLvkIVgvmVw9fiZ34?@v~ptqpGt^
zHQfiwP0Q$fz%oA~uN0KGS}{(OZ7smV!y}%5_yPXnPyQo*?+^bKKK=f$<IUp*(1x|t
zeSi?aA~j?`EIFhr*18ojFNh(cOGjxd(vVD8>k;7T(1e+RL{M5o$pvazTwOcHBkOEv
zkq2O?$@Yw>k}{u{8ELSTblZ5>yuv%``n;!cg2NGik2E|DY27re-a9x$e|zV^)q7)s
zIshk|Tes5)q#drB!&Z1H<x+RFc|#<<xnW5+Dzw-I*@3JLwX-g)^~OuLJ5y8Al;I7T
zZiR#W_FNE{rcNqXE^raBuG@|@Siop`x^P+3=64i;EejO9M?=24^5XNJBK}R4^{ag4
zT&S&+21p?t39FeBzEAmD&3Vsfc+E5y`O3dz@MeNCsRNr|_*{75pi$a}5Hv`PP;EeA
zV@+?B(uNL{o_Dl_>;fPb!Q@<kb42;014`#*Xh$lwS)ICN?H$oYYSZitqtoBfoMnTa
zxBxh!hF^Vtdk3>8ZRuE6#_7Zmu!poZp37bwN@-BbNRvRMm$Dr@I}G(JwP2%GyEeGS
zi*D1B_O3v0Jm-xG>7_!vBT5SfQYo-Y7tmQpR$H@k+;bQgqs_M)N7Z6090wtjj|eC`
z@xuTYDG}1zp4cIDyK9(+z)<nV1<T$CZ^+uN5o?t2hz%Ml9bGH(Qs6Aqnx+UdQniNQ
z9cm*=wrRy-cR@@JwHB=FYRf)l9?@}OY-TWmNiAx9qMo-ukF$^NnnF<<gg3UTQvo61
zF&%xN-n$ZMQ_~5;0d4!2jG1ZH6a>)VAu8yla?vq>!`r!9N`4A5-i;;zs-mIpKt-z-
zf93=}8oFo(#qF$*!9yFM;PR}EOSe?&>z()TVl5|ds4F$GF$AW2v?u*#uFK9U4r$;(
z-zq#5&I`sAC@DchcZ#Zta2%k$qgB3ZF$SAI6bcTX-~7f2^Sr<@du`V~pU)SbH~#6}
zYvwYB2<Cagzy0@r@k^a!UvIG6IA?Rgc)M`;k8PVXei&ef=k|ly5WZ+MV<`nO#{GXb
zOWG$qPFR-kXV_^@?N4dUPGoP2*bboziuHTvz#+2fHh9T7p(d}IAaO$t`9lO+WhPLx
zOryc_^nfQ3<>opiiY@51v(y*eP#U&6Y=^{}D_rPMiv?+|nh64Er6PogQrX+ME{h3G
zD-NE{+|@F>#&Lu+TC%Up702U&1IfBFwxcP|mn&NYV%W3loEu1mcfpt@_~31ra7g25
z`8{XqSX#E9yS=2!7jMZcTJwgCa<JR%{lRS7{woh-2odM|89(~VUtl>mNG077RB-eO
z(Rt)b;koey`!n|Co41%heF+#vzsw7%uptpmXiU?u+HgoC{`xQe1poA({}Eok{1`v|
z$&XNT<=~GB@53&--4P4N@XPI`3MrzrhP?2yUdsBFY(Ah{k+}jy`;?T52tONcbOp&m
zh`0EZE$nxHKLkj_$p2obW7*(+=SlHEwuOhzS?@ir*Vzu{w$tu)&FGtpw0BbDj$zMg
zMhI$HgO#eg{oSnH+^e>oZ~uH4iQ@I`9c$g_HX_fhvAS&g%$)NWox`%M5P>0$MD}tn
zPx@JV|C_Y^=P?WedaYl{v18qp*u|{&{fTUGIzX)rMM;x0=$S)H>p(WK@s7Cs3SYrn
z)|sJZVT-R(>fVZFb~^D4CRg2&k7`jIJq84~nTI%fK+QB}L8+S_u{Hi|=XKrRqux6X
z<FK2zLh`7ro`85h!?iUcY(`T??`U9XHXMRM(*owkg~6MbhqPDkK<*8-ss+D(#eAiv
zxsxq3lx#;`8y&pPqZ@5DZ-cm@s4&Fe2ZtC3<Yh%^g}=_TN=c1wa)F3QX$9Ii6gfNV
zTW^*TX_=d~K|CwP!Z5JHQp^+}c26{F<2JgxwgwM^4(#Y8O_+v2w3xaI8ri(tQD6aZ
z|9C+fkC=`VzYpzD>F^-Mti8b}N*Tm);NK`h*u6v~yMksKKppjfrFT0^t+k{zlQA)e
zQZ^>4a?x(unGLrrcIJltIoUEeyYih^7vYr8RWA7-4EHh7NYd<GX{F%iaNMa>agNU{
zKSmk>Rm>d4C#u$J=U2s$Mz~<^+-{30KSkcnj#T6@41kK6u_`X>f)pdfG_#wWq&GUt
zw`2wYJN*d&8cZhPd4@Clj6+%Jq^!|Q)}-wH8N7#fL60U`F+{N|pp{0zk{gy~we!{6
zlE|i`O(qZwDHg|frIdyY4#vbIh<c*~UX7X-vY4XXkYvlLR0r>X)(nKPWQoH>^tHD_
zCPMl9Z|J+1f26pRoqp9V#=$;xI|SWMJId}n+_VuCAQF{vUolQZa}0UMp3U0m&YPwQ
zP={949{80ZsXpu*MP4(t3#JDuwOV9JMU3p6-Gr;h=|HI0Q?tRpJfZAbE0g<GS&?L+
z%2j!Y86vwsXw@!oRYlDmwKNQ4LUIv;j&VrTr#X)lJJy;Zffn~AuZV$thqtE_)UkUs
zIFDe1gZEFZ8(pcJ%WGjMX7D(jP9_3Y9H)W9tSS!U$cx5=*L&WeI3A43m)dVbv34w1
zZlg`u(y+&z*ymj<x5sP)L-Ya1<6-~&wyn2V#!hPuF(}@?eusK#z}(EmNKIr^#c3Ka
zpRf2|Kc=dRpMLflm*4sg9OD3pq6aFrssc9?1>&_-{PgdChJX3L`!{&?>T~?3KmND)
zy?^z8!0qYQ769xNAt_~|6Wc=IT?D{9x279~Uoo!>^^76xTT`<fFiH$8BXFbsU=B(J
zL@JiN5&;mv+eQd5gD}L1(s+<)uDnx}RK&7a?UH1a%G$Ty8d8i{mjxkO3G_yTE_5Dx
zGwQ(T;$CUuQB|z#0)a#CTo6<{w@$IhkuK(-&44PF$4Aqc(l`<~x32SOU9pzROVHlw
z6eABU2vHJM^hWcDh(}r3+1QLORdHDJij-JiR$7Oeoxjw!pC`4^F-?sOX?|XOj8TF*
zgb<-!v6jW67l;&z`_(JOoX392_~0Q`_V)ooM5X0|<_sYeI-a@2K3$H`ZA<6>?!1RK
z)~of-OSmbAmQ5S%z3dm66GnE3x8ZU_R~5V7m5Tt+i(K2@KH|YMtrz951>m@Yz=_d%
zaFnJfjT{m@=DhOHb7IJ<nL_N(BC19jSA@^?wOxod#Z9I`O>gR9mvXd1cENQTbh;{H
zil&cZo|<@SjGF?%nUqFJBZ?Ei(+v+XM9)Bg=wwHPp^Dx)R@7b~MxQxh6h(@GONfF4
zi&!?BQ*?DW-rOLjh?*-N&KLafM?b<Je)A9U{P{QW=Izhngcq=bz$fPLaUd3m2KMpR
zYSBEv>2$(8uc$Rcyr9|=>pTO~ofEj=EQgZA83(-%IA`%L4Jicn#7-mQxl-D*C__km
zPlZe07HVC}ieXF;&_GvOMax`DwkkG>z-77OFpMaRQQ4+ha?b5>8(lX<RGyGJ=P;LS
zZ53*9O&u^v!J2uljMHTA$&)U8)0%Gvq7(*HYKL%hbFlNK;C#7Y7zfkYc6b*NBK6O;
zRSZ@k3efG#;Z&ZCJ{GhN#|O&=&+kPyL~^SPqXxTd(uqzP&L(pRG4P(S_b0{x7aX|5
zTQ!(IA#bWr@7cOOfk<M2#*^-v&v>a8&1HZV;*+m{W;N4jehm%r*F)^Oy;QcfYww(7
zyP;JkCIq&CMlmZpg`)i{&KI!=hz%WpIZKBy>_b$DM9wx8NCb9q8kJ};Y^xQ!Pa*P9
z@TOnOB?H=VnXeRST1Tq3W*RsXfEyh!_+)N3HE6%lC?MQ^)pR`01gFdhBelEahUn=y
z-5g;O1ko{vN1=dA=#_u=<N1+ja_>8>XZ$rO&dx302fx>w9l`<QFrqakRjYX9#+)z+
z;dvN!w7DIp(Nn;TW|~EBlwyXD1AL;;wD-j#7GT>tFr*P<8urXWC^bT@H{>;A7zV3l
zDXe5O-H-N;npd>k_Me&Bq5a}!#JlTd&p?yyv4P-PI{w37{tz#ppRlYM=j)6yczpWJ
zm(Vcc_VxxY6qH`@<(s$ojsNbS;cx%$@9_6O`Fni(JKsXIVW@Q)6uk3rN*xxn&k9Og
z;9S7t<0GcSh*k^6abhs90ND_s-34jG3+vCk9Z(zTVuzkLO=k$EodmY*!o%@ETHIb(
z=@gawV3~gApoTmdi)txY*KBbs-5gPBl(I+b46_Y3fQpwLH3&Yyj~;n7`pYip);mh;
z6n}e)&{HDP7`MYv`5xw4_XAa2z`Cx6PVm<)*M;;<N(`Res+Kx0ueNQSMts%Y8O*yZ
zh>^qNVklMX9ft#dMmaN|j+!IWe4?9M5!k{omUX3(!kZS?B8-0JFA)rBz)~{A!+M2z
zwn=A$P1AxFnEkjl%}_%F#3+nEGGmqxPXkN`;4K=$J0K;VY2xI`0HC^y!A1C-h<G$5
z`n0hSyBOaG%ZRH8ftQna4(dGMnZJn60nwQ}Var2gz%Xlz#Cv!ju{8jRDuR06P`1Ep
zS98BRLxlPJY)nWOi+*Vh4P4}c=l2*LFYzeMxbzCtR^el$bfxoMg-J59(Z}BVr!%&8
zEIsp_9uspel@x3mEg-n^eT@i+K9J23huC<R_>DTYm?QB4&IjbBA`K&MZcn&8Fjwx+
z|Kcz4!8bp`@Bhm`!e9KA;PcOZ3>Vb2-L}9bo{P0KXa&mb@bIqU^_Oq(`@j1VpFDek
z`|Eq;K4YswYpOs9=&oa47l@e_y5)tEMlM*^m5oBj8+HqafQo`394u7{W-M#om#}Ck
zIQ71Ob}ob3sdrpv>wNYeTlhS~uRF#4F#6j}b<WPfZqo{%KWFVBo$hz<_t+@y6;1e=
zK^Y#O&B+Ux6C5y9wOQV2;q%ivuIoZmmo~~`Y_<q)GhEn$;~mysQMw^`=8&#Ukt8C<
z1a-i&&K#B$=(SJ=V>NVMJZi5*7Gd|61K>T9IBYNB_L*sKkjArrlMg*5V06=A@}D&`
zb%?p^OE1vYOuKD{I8g(~up@%=N@r{g2@TAp+<>vCsH^|-*VkTgOR0d__?;)pMii?<
zl}KB&gmF{fAb5dGk(;(5E-?n?LRCcw4iIiK8w*CqOEtuA3qjqGq>f_N^OPcjBkce9
z@Bjd44Vaf1x3{+lG4Amj**4u)G7qlS>~OXA0^5$4IotO@5o8BrOkO;{H42z@J-Kn%
z+N!#QPyLf9!8r<uO%LRpV2S}PIF!s`f-YRb(N!sc7CWrOI}9li;Tr}sVgQeikA|jI
zT<2xiZ4D_IHL4hgagQb`wc>I)Q#WE1ubSp3S;uTC)gCxHfnk)~m=bOd6AuQ^RoZJe
zDwl4&)9sDvNdjGIrSPI|uhDzm-A~a+h<A7v2Hd2Cby>giuyVp{5d7$~*ZARQuTagQ
z{NddlZl4L>{_qvv-2WwBe)tR@y}ZM>t6-ihe)G5f8ParP$O;FYk~5ui!C{`S$kv6)
z3_e9ID`0^n!!VdeZN)f_h(7FFbSYb}D-97MfcR+I5FToAh`jJy!{KlM`d0tqu(_Z&
zO-u5goqqzb|F!S`&K7gy98=3hsHUp9n3jnLGEo>_n!Pn}xulk7+p2?!d`W*r<|*1n
z>20gt(6ZXp&>4qPs52q@9Z<WV){J#!$n-RFkl!wq*c(ONaG1~4&}<e~2OkWXQPk3L
zKF?TYniCGk5qah1<()?wEVD_~7D(*xpi1BWJr=EREgKdpZ3+AJzkzw8b()rTr-NyF
z1kR)=7CU1`g)Lc7ji%|KCQ3ny4%XPPUAT&7)SVr4vc+6%5ktDOAPNg`M?nWLFN>k7
zGy}yzA^Eyyuy8#Qb@2Ro^xl#CLK93k%2OF5Q7w4YS%lPPBHEDC=$$$HXyB=>!dY*n
z^UNz_-WmH}yDbe{-I7?lL$w7VYz8l*OJPV574(AZdc`=5h=B-k&6Itt>uQUWv;JxR
z6tCBXnSDmr=4@y+0B2gU-Wt|oL9xmpb2@JYB3rhj!fD;LRmam2*UJ?Sg2#tP{Mn!W
zDL(xt-@&i{`fuU0&wt9+9ss>Gu$*4ZXjk#-v$uHj#Rcc{ids8d`y2e;@BA9VNYh44
zK+dbZFVry)<3OZcJbW6knj_suK^hX!fpHjjR18X!)G!PX@q05--<}-+aJ-{9=+b?z
zz~1S-GgoQ=h{9-XhJbImoLdqDAc(_Y%Y4B!j6@;%%l2iCLz}~v!`eFKCAD=*LOQ8@
zg<_pE6rhYrw|6Ar;e5fCxk`kLL*y2|UW<kFnuf7<g=<7@yP5gO-qGlFooA@<@7}eW
z?vvlRc4c}*H#*hWxn3DF2FPyK+TPc7Suv#1z8B_87Ca?CY^Wit>ozSW|L=`a>5T*N
z#;$2scneAH;NS5&8|`k)(<Y<&Wxu}mV7Iw6)CQ1E_u~voudSfuic*>_dkTR^&X%5U
z7x)wMLDbGW*$<V?c(G}SwgD@}v=1i5s-0>pkfxSfw-ru1NC%c>VJN6`dn>?}pC<s~
zFwjlqC<?BvV@RGfgjJ0=N4m<J&Sy29gTev^A0FW#xV^g}oLe#$Lwv?@<Zxjd&X@*v
zOL_(>2ML_$cb+$Lo9McW$Jvid-OA>i-F{AdIr&0Q15vdQ<DS#kl!I?q#ogU4V8{l8
zfeCG+8%;M1s0+8bO(LwlV=V>CvT|4#-E;n!1IR;d&Y8pWb*A0ER%ULQpIc2w0ds+R
z0j4DIOGUHlAc5Nlo~0ISN$n~fOD%wlI2;E2`44}BzxmncZ~^##{4alkk6(U>Pd@ns
zzxf-#fsa1AMOibvc%)&1fV!CwB4W;ll6h;Cc*?b6Zd%6}lVy)>!goRUPtRd31u;d_
zw^#&$Y0EC_h4h+jaNe=>d>hy{L2$KYV!On1HiJlKHX{uJM8LLo5$6Ocaf1|bXe+w}
zeekH-F~o%FaHO87vzNDag{s4vX{rI}&fNwQqG28DnsJ?1EK7#>k(9a(Omuqn{ax^m
zo8-DM_fhRI-;lCg>K^Z~NsTrgUCs?pb63Urx)|kYdLDVobaeJJ-Lssq4SE(8U`Q_j
z4n{DFa{Fh$!dELD8PpN+v-D39B91Be4Chs&4lO&*{v$?}^6cPQ=S`D|&gC`O;lI&d
zWG|Sj$oI>#{0wCYraN+5=y+$i@z$K<ouhWGHFKr%_t-9@rXAAxa}AO7RW;g{T1!K4
zmBGVgI{MzQ(M#h%svw$b+BEHydlYYp24>C)mc>>rpOE@#`vI#%sk&D$mRhVNet?R@
zQmW}HmGxZ~htjm6l!hTiUfxQr(!8wDrtpcZgg7FJQF&Q!NK<be>q;G!cK+;2t;n@v
zN&{Nw1A90gF;0(I=LKE_9*RHx|NIC15C6yi6MpU2{}6xk*Z&r^vZCwZi+lY2k6)uz
z{wzxAnCF7l`QHBU?|%U=4j;T2@%*D(#G3%|WPMokMJ=ND0ZN59;xL#wjqj@yhs%7~
z?>qn)6SdV_4yzg;<o%`v>!^(jTC?{zjjW;Eeogd&F(!5nZJf!b*(){-#*pBg1=-gv
zYmMRQ<2Z5v>;wg3c+&rI$t3GJja;0YojD>v9MThYuWUV9WJQW2x;Df_i59?nd07^y
zA;ZCY#1tua=yj(;w`K{g8OxGUm46>`z?zHA6xMaWA*7;a_Gs25#dq}PcqHpniG*&|
ziqql9MKxDjBsbLB?{{>EBDkl8Z!Hz@G@<p4(sdrS7L={BtZzfP1Js!ou<z*KCUN*>
zzrNOB7Xn4>jH~+!5+SlvLcKTsJNA-4iJvK!reoEJEOR;W*Sihj+(qu$ObFh4#29y7
zSMMFwDUQbx(_y42xOa@>SKX_0%Te4-6vO+7Y!p`tk<_n?mSsd8KsSMQ2;LxO!BIx!
zHRF7_@WFSEn`x~$+#H!ZWdp82FijK0bV}Xzoyvgm<huqy74wBPCet|4scXTX)awyJ
z+p_qUdFEhhIciX#b4kDg>#KEJ$Z_2RUt3o^JX|^8SZ3I=t{77UI;&pB(eG6;&Ne;(
zc=zrd?$2kebAd+SbUH#o3%fXnOIdKaT=D$wgdrsem;oiNA|yecNm;FH#@#UBzyBBi
z9A7?uiI@39{Ot7?xLgV@=Ys2;aeseBt)1>SfJ3eTX-JGBaSqY*V9r(X<@+lh&cDFl
z{^%F@(?9(S{QiIU9sKU^d}`um(oaMT*wGt(uz}TvcJHB<#}%Vdo|VI^bIy7o&7`0#
zE9W87_65h8EMZ6qy;Jwuje=|-I<e_8*V1sEiMkx7k%D*O<vA}aS|c^eY)D?dLr93p
znO=kWV<89*encsmpP>TR>lL{)b5YS@IxhwDno+eO#e`Z42<Y?qTb5<F<pT~~*S+h}
zI5BiQ4HjiBsMY8iyEt3fT(pMR+RvqLqI=yg3*Ai#yH<BIb#!)>Y@L$4XPRkFjJhyt
zTp;KwJ2DB#KCnJbd;j%c5?J8`M_P2-WE@P3=9m$us)!*V+UDPz4Ol`nYeU_H(5&Tm
z1^7U%STpzMvdoAfGJw_c$6$Zw=p!9?!W_ZPNKo8UmSzmmhQJDlLx=-lW)xTi&!+ce
zi$pWweBBZ8F*(eaE3{J|y!FeT4I#+2U>Fi68#;U-m745(qU0l3ty!_*mhVmNY-X9~
zD^g6D=ZxALQi$jv{dv7Ch>1S~CqS|eVu_R`Fnu2gA)<E53${?#-~(D?yNM4DU8!eV
zY*xvcOJWK{IJ+UC%wkjPjSXOgU`JEn_Sqf2xXze{5oirx{QNU~{<AOeYrppUxPSK<
ze)7}5L0K!#j|<+voe}(CLC=6HFYL>@k|p@@Pw~OsZ{hXN-{QN!`E5Mr2XwWJ!w#bt
zD;>d3Ai~cPo2*ILO5%M$NRfKyR#Ec`@4~Km-4>qa=)x|oVHhSF!3Lg9JL21$ElQij
zLF~EQ-vY!TS?p7Vv!SWk_w55~<+gmaU}xtR@wMj76hH~pS{CL%x&TowAx<oa-kz21
zes_*(1Z?!_<}fw7(-yjIDQCDeEM38dho)F-!LnwENo%~lgWJO8eY98~K11q=);0^h
zYrPTm^}G|g<iT1d_)bvgs&+<fM^8toD||>h7|<*htYyWTiw(qy=T@`E6iZmxOi?-4
zy_0;Ku>jfc#b4&@YijH)zQY^eUn@H3hjy}suJU5`okP*)dVzA|$k=EC)-~ld!wJit
zy?027AI5&EI)|;y{b^ufSZPc|PP+YMnV(~*BKV*lrU43y7h7$ZM*iPk1Sv$Lb2`E_
zSWb{ps%of|04YSI4N-I+rB{B)M#sVP6luo4v(<noNeT(?@81KC4FJyG9t!t?+$z@A
zaajwRvNO*6fWt5#h=po8hvVU38aiIQJ@bWd7N!d@0M}dA-MjZp!akob`10);=a~oM
z;c&n(I^3Qn-0LHzG~#r3gQXNA9~-{T>4xHbc%;E0O%dqqB;4|Wdh2+0dt%^k45Vp&
zKpYf@y9tjED=L7Zz}gyq^Y?xe_;JB=9Pzzh|Jb5&h~~U~bHPu4`X&DAZ-0u<UO$>1
zQ_zZqM)s}aRJSmGrtU}5dPDH=c3_lN(ew6ska~<3!_({~-1`3j#j+N(+HiY1VH08x
zgI%Uh(3;=%U{FQ@IE&RmS8T5w6Wr-^Lt{P`=@JiU!(p6ITEqE##Sp_TGe|=sf|eJf
z_qg&dbw(+Ilp+T^(@hS^IwO_2i?u1{wc`GPMckztJ-V!HbvO+erz2V`D7hiV#P`oS
zT(?8Ta85hS`hIA2?P$R=CY>{i#e%~7_V4k)6=8h!_MD4=8qZGM+*Ti@j_(bDdk)MS
zd`*F|1&u<nb^ZbZF+uVExCSDS?Y{*TYi@07gtlb)4qhLd)0CmRPzJE_8S%C>@bcqT
z)1kG)kHfA<YL%K}wM!kK=Ik=h(B6<lb`2ebjBu;nVnR$y83f}r(A}?e+|vC|4QaFO
zziLB-wIp;RwRx@75j7x4z>+gorE}9eBDSp(jj5eW?g*kNt#v#;UXX?XDMr92v{sR?
zSHuu-oDS%fj@Rqug7Gln#%it}A0Lewv1lVz>N9hu++^z$SHN{a2wXm@vSlYYkB9Tw
z&`d5O(1v9#h(4klvM2&L&<HZv@*W%)leH8?M05p~TmWrs5rRYSbdq+kzc&sE<B)*D
zq3hL8e~j<{;CuM+!*ApB&tD-wyvO-|#ogfsUIQ-IE78r)49wCyuBGAUU%th^|I@$0
z5Iu(D0na`@;#$sVS{T$^n}z%HjA*R^UQu&1<3vP=5h*bp;#v!V3@0ePB3BNl!Fe)9
zcAhC<_W}nMK2L&DSaIn+k=Mt^M~EZ4<P9b2s#x=i>2O4=&7AdH5U--nivb}|9o$1q
zP!XcAEBkpNj9{si4AMAI+Hk$Hsc5TA6O$vXOC~xNn7|NYVrz<PPw^y;_uO^4Ab6$#
z_yD7&X{2zT@5fpy89nC#3D$<fpGR#*pBfbp5X~&LLcz*3aWsD{%S@K5*;%*Wk)1{#
zM<zu)HL10-*H~>ZZ(Xrun!iFw=$$4kfbXrS0G;P8k2dQ0JItKvc(1SZAB~FTNxE&Q
zd)$KCuOd)3;S8C23h~__UuG{EJnucGVdNHRF!N$rT&*=rL$^Y6<LSi?c>uUxW;kJd
z2T>h|t^K0bhNUgsLZ=a}6gZG;hnXi@p>|0u%c+4wK}hD_62jJW?=XO~HCBQJM{|br
zbn=z0YG)Gwt3sP%$unVZZK#c|QZY)T*j85x9h&2qFdkU5Qfq~Jzqb&~r65Kka3vQw
z2@H$1J~<y4B>*cEzGm9{yL0d%;KL7&bQ^jH6`~>I5HSqtA84~S1Qm`Usxc;%2Z}g-
zb%=rATdx&E8ZZq+Jcr;BokNiZMTd_LcQ0=7+0Whrj+fVUneq1RJACls1H69!1qZs$
zq1}t(!<Pqq_~G~PgMadE{Ez?B|IA@j7{cs=Lzy$8#hz>_+#D7foEN-&dB+$6=ip^P
zbG!sn47=8ljxL%ZHUv0Wc{?BR&FO7OjQukoBODZQ)Ag7OYpZG+V>Bl;hn{s=F{Nnl
z2me{*2}BXSShP=P3Rlyo!e>=QFoAk&tkLOR;m3&5>K-<#P@r|JYr(_g#VA$&cR5$Y
zl$aG&D)KU;mvyJ|=j(zsbC}h>fAG5)dK>7rc%orQ7>5JI>26eT4k(*YxR`dyQ3tx!
zrEME_@PYI~?|h~T5Ak7$Xq6WU0LyWD(kAljNry^PkL4xSYP!S#@6&&i96KMpA-x=!
zx-ynS(9v^;OT0Wv=6Mo!a{-{m+=dnz0dV-u>qajE4y@4=VOZ=G1Dv;`)sUAMlW8Zd
zYD^W&yby5;4$cwjW5BN4jSWN__*|+2=!$DzFb)yPdkk^dwY#h<BPEwhp-e)7c`=vt
zx?mWRHTF;{5X57eCO#M5Ljus6$Cwfaz0M2D$ADo-sI_2Tt_UIE?sNn11IBSe$qilk
zPD|h7qnf!{8@w>h076lDG>0Zb;RPj=qGTqcHG9~SsSeh$Nlk4Go-RhWxfYE*iXPBF
zEpamhW;3seVmgip!CC&C$NByZ9^U=}LrD1M%TMul|M7=d9t%VTIWHDtB|B|qsjFDJ
z;HPiy5xqnD%OB!jzxbWKZj!pc0PV8>`wAk((=?*hW;K6;yjtBOHrX2ypKWm1;<Nw`
z|E+eEy6p7h7S*)XuJLU5xO;ZXoH@JW960<OEMkeKAUd?Y4-oK<-OefqBWJWP$eEw1
zw`I2raIyH7ob#To8GJNWPT&l<XodQ>jt;NJRwM>*S2NP^oVIXcwfwwNQ56Ug=(VHd
zyl24X%*dBo85<Q0^@}0kGBZ;&rGbA(%lzyLLv<Z`QIxXmcS0>pFi2zKebGIZ+Ta|K
zKk5p}P}SUMcoUexr65F)MfvyfPR#Akv!+!`Ug(WYE>JG))%35nnEbL|U$0WEu=qPp
zqkU6bb{Pz<Q?mk75ZPKW8^1H?m(#6w>NkRM@k30|ft6(JHQdCVY64z)5}YQBJL#5n
z<pk4|jN>Q7OYKR+7L@Ej>LvvWd)wwdG9A>GyH_gpZ|l4=!$u7*?nHMs$64s4*aYO-
zJAh<q|7d+5Xj|`)NWpti=96uCYMbLS#l%Zq1ma<uFWDnRyf~)XM~W<i_fThAonYBx
zZGemXyQp##`PMf-Krn}z`}UL0O`JfT!;-VPptwb;Y3*Q?7l7T$?HrM)VTcUmEEz+L
zh#?}k3fFblgRl{UXzcOkK%s@9D+=K4o45G%`yb=gm!Bg<hapC+=8hClZ!{*`-uWvL
zV<a8u9H!|2g}J^=AB5iV@rTcF_xuLE&!`qRQ3^v~ua^cN0*wf@VVN_gX~cA}@JK@$
zwkC)tM=Z5&&T<mjy27e6Y^Dh3kkUZ7y>c_w9a-FvlxErvUcBA3R-FPzA~}82$xs96
zg^oxK0W~w)rrIGZ0M>cgFTNoLeE9M?0JlH0>yqIjv%RQ$5?fsQJ}~kU7xYFtq11w0
zi1zW~7qkZIEY||$3}(mC09E$b4Vr{V{dZEk+D(p-b{)+&<WaL(?dR>V)&DpfRyR%G
zcIGf>+!=C5o#_?_008PpEp4IDztWcy>D;o_03dn3+tCLox(&xP;yBoFu5|ut>BuWx
z+bIPbA_lZpF{QL?ymF;TeoBc(pG^;&D|%*T+xEG2&<sWNiAzZr?C$-I_H@o6idbMW
zmrL2-zm7GtO0HE4lnxOwIzT*8Y@^zuB@>AEyFsFLhQ*q-3cU(^_xs{=xy<O*7Sw6P
z;V|NOIG8rj&=gfPt*nOY7L`+;-Y@TkTDztQk$GZtT@y7w94A2kVf#cgy_ffZD&5<0
zNIb7n+|S~?uBeSm1s&%LQX1gF_9Ru(v$ZfEFAW}RS>R=$ws5)PCx7=h`1o7j!`si_
z<K3G_h<Y?-^ag=uB3i+DIIVC3WL3<SQ9<jvFw+c{(Wng$M3QP{reHIHf2-6%NN0*c
zZUhdlE8Xauq33XzP)a6}Rr~JL-cV){Af*AN6jJP9R1CYzIkavvDUq*6hwyAkA@QAT
z#>_d2bb>tPmX3+{P>Nhc1b|k>yj&4u0z`4TIRVm9bA?kzQYrY%x6VO3h9d{pUU74n
zP&@VSAu>-cc(Ql3Z(2Ibqw7S|ip4>pGyIw?iARi+eMXcJgqUyuJJW0B`Oh<{89ncS
zF1XGamkW~zhG|6a9gmM!#1t)x%JOQhFu;jJY1A#((jZFLZEY3J-0)kwhscv$Vbjhj
zPZx;kj(OH^Kj$Z{^e^-E^)z-qV46m<bfy(!h>&gfG3*aG|9xuJn3q-dOU2oNw1o|q
zS{VB2#4-nkFuC$_vy&8CDC_oXlOZizpyjsUs{&JscF-%*5UIZ?%}fcbh}o*1h}Liz
zpoGq2Emr<+Dc)*2qBJDSf#Yx)+*9vu&5avg^l<8sOQFCXq*Ru5We=llc{?7hu*%GP
zPqA>|U>jn9cMRL4j-+DF8Qw+Qo{osYS>!{*I3|=<&?{56lWE<YGqI-QP|`a^_e}wH
zj8aiM<h3h&pY}jY5e~k`;{jQZ=-P3e7hEnE3`0a#AVqdwdN1sE)J&S&do0T%^;zCq
zPCh@2&wlm`eE-)!#j-94-2)C-EVe_rE<uHb(oojD3^5|F3pHXk+~jKRI&^r~@ms(7
z9i$}a)x)8oRm;?~dYyUBSWCuQ3R)ASG@)e0bzU(q3!DJYp5N_3(STw|19~qY`cBuD
zn6_jSC4V)Xxikh>Z*@m|G=$M);sTPe<I{l%O)iC|4$l`ECgNUuMhpQdMAMfvZ24zS
z9L6;4LD|7J@W#yn01yWY_!PctG&EIfJaGG<m`g>kVuM{JMFD|GU|v+^_F5{to5k}`
z79XvOhX<B(!E&(px#e7*C|CRQxA5aV7qImG{R`gG^BaZO+Bq<__7q6IX>3#kYUmCC
zZvEPf+WghNO0Edvn9~=DVs8#)!Pd*m3r@Rb8#-s9*4*m5iK-d?%P`z2jR;N<Q$VTN
z4)cy>nGGePxLSmDQ(|JnZE&L2PweLvA&#zR1&Hnf2+kwdY|c_!0xXKgOF)A|@DVWv
z1l{@?Xms($#IUkXG0zL;cMFckfzksgQW~&{Ae{o`dPS*RWRKItK3R*tU<O*VK=BOm
z4&BlU#xww)kv<MW8ZGb)34z1pkOpdtP1oFO!*n>n!G<5@z|sw&51#MaI1VV)3~{Yf
zV&WP?jBt*_$>A{3VQY6-Ev$C4{S%_>!0tlC%mG+0Gk*Nne}}uf7dV}6FpeWGk5^mN
zh*op(Huk3}mO=ob_J$Nk{O<4n9-e*h49mM$Q0<Jt(k;Zi0_st6fe-u~YvIq&c}8n#
zU$Ewk!#H7>A_Rgm&sdfjH#fHk0!-6^f%3i5SQBU<b0Qd$Ag>wgx+28^(Isr{Q85Ix
zlFdv=<ccx`qIpqhFx*N4aw)s)U^9#b=gmw;nZ-Dcyl1tew`z~M?0Q{QMXwxk1-y&5
z{Lq*pV6|)A0E^ZHAtVl->QNg2WhUu$t|mpu@PTHGycU$&aF{zvVbjOvz$Kd+j4>md
z(|boq1TVVDS<dH$4M^tx=@$8>rj6T%vTizJN>ez4XwOgooovo!TeMWr)XtZ#hC;%$
zde|3Ri?4V>YJbVEuh|s~(gS!ZFt*tey;t;Z1;=Q+@nA2#SKVn@h8_Y9bXAeo<wbco
zPS`5Pq6@Sj3sH@=RSYAI8Pb8^BY2qBT^qNB&Tg=~_sp${YMY4<9JD~Ow>a4^lwj3i
zrB-U`N<s9=grufzvloF|0byuw7n*_<P;{LtrfJ$KYaVor2nfMr@Z55o_qblJJP?B*
z>Q^-+eBtP%oZw3#rKN&p$&5YN0-n_)cwVslbIP7GwLKW6R)iQa`Uq8CGPP8M7;t%9
zfJSZ9R!bzH2*2bE5#TsYP_>v3ah7`}gp09>VA%^gP7|_J#1ye!IV3y#c?~JycsO7!
zE0%R-#ZB-A#akwn>9!!GPIAtB;PUo-*HVx?@a<1u;*)Pa$9z3=8-};PCyf!>l>LC&
z10|r=PC;@fbr^>Th~T;|Sk{F-kvsihWzv?l$c!}xbCOn!Pm6__A+%9vN)(O<R$i-7
zTLY$HFq%@IC}03v?9Q6mND>Wk+YZnq)UAj`EHI&+&-$_ukx<)an>nku>Zn$pT&@sj
zmiun{JjR}E8rH5x?`X4wh-k`o5ce*7PwS=ygF0lXPbiw{W@M8k=#!R)D3G&7kk8Y)
zmi?R{^@?3UEzUu<K25*|Pi&HDyU0C}+F$W2gk+|gW=J}h5~5$m25@~by{obx)bTIq
z9D~u}1<`vf3-w67Rh+LY#%X|$g2QpdR#M+PXaHC%dMiNhNJvBgEWFu?gNr7=Xo_6U
zdl;*T!(mK_Q-Ft1C)XOPHu#XN7SGOJ!a$I%_D*TWB1+0ZAccS!CY)|23!<j9rK3Z=
zLp7aN@PV3WRq9wBheYOiDXxzTFF);2$KU((>=wbZae>143hyJeT_#BHh2nSTNmpy9
zbZ5<Fr$>ij;xh}No6*KcE`&aKv|RSEdf5M@d`5Oj3zwy4LBKrkwljx8z7?XZ%FAm;
zTW0+I4}XMoIO6+1_zwQ$KmIv-qn;{yhoEAuJPygwReI(sysH?dfQrowL>IF6kw~fM
z*?Sl!Ha?6DfL`VWAqFx#E|`0>?NSo$9mk1vTrqlRU6H(?i8JT4<ImEO8S^e+O3Z$A
zz3;3<so822nQ)Oamq0FETUa<{H@#El#KW1Ws`EC`=9R-Pl?*7Izen^Dy)<0r%g*YE
z2ySjp%sLb>|IpaKnk(OlrpT3fh0!@kUQvo-%@x<{f^{WRbe$JQh16m_y<~Hob9lIr
znP}Fh3^QrfvT5!3nNX&*?E2nX`ko&PGdnp?y(XI>Y!(oKf@g(e>O8~ig-bp~tuWS#
zjLOy;^5pr_od(|#y#Mua_K!8#wOS#G8xaDXdh@(MM3Dx%IjPa)W-z3LwiblIY1TVB
zn7pM$d&lF(yTF$X0oOTWT^70^L%?8(>;TX@@bItzLP4zuH+tsYGzhG08BdKD9n=A_
z$%jrwZ!oeqq^@-aSGLA2mXxKls<S7Z%SI0eZvl<116{CeE}-6FE#3^X6^A07MbX}4
z5HOa+@nY=&;04d0Gm50v4b`N}u{Sl5ySI(Lqq~k=?S;4Km+N%~M9m^jGeT(<p?H>y
zLv|gRm_T_`b!?eOhv|T2i=V+)R)?;Zp{5o_4Y-&%q~%Tb$4mF-Z~)C8G#bQ1#Odyc
z%lj3kD6^WRx}zT6x?Aav<aWERR*=0*tgKe0Zb|so!{~q*8vf}&{Vsrvd7Y6<B~m&J
zP-S~VseGAhZx*u4rR2ISh%up6Qp@~tIF!oaBbqyUjLDp2+lER6$D6>L`W%LC7F!;i
zxkQ~alZ&rNLnIoqFr0ZU1;a4zl(~rD{_!5Qc07A_2dH8gM+TKT>!mck$wqN)?%iP=
zF<Okzvf9NMxENh8GlOj5pqUw4^SoNXwZl5EsHWSgwSe(NM!N}zx@H4oZx&6XmhlIB
zAN<z$Y)hP7TzjpWe5FI$9D%_*sBYcE&RnzIMB2h(Ch(@+4!f-jy4FH8R$<v#@^#L=
z|37D3M-zpQd|m;uj57-+)lJFAn2N1?wA&Kogq2l32&(f~<`v?29=PC9YGbdaz}gRp
z>~GJs_Kw3iLhR?6#>Dr7_XsH-`?I?$9v-h8rcWoNG%$nH^qEsixPQ1nTf?)v6PS0_
zFkcw%974npnQ5o0_I-OIJ%$Ve3hC6)b{EhcFb#n@h}E8b0fv-_=77UHy5KNQMk$eY
z3(=u>Mp;Dr85H|FYj6A6a2!XYXSry&6iv2c^mHUDNE2JJcYxX^MqYGl2uOezsAr~K
zZN)mW8#)U`&D<2wH3HiyrON+3j?lJ1d_Y47yu-Wq_vjDr5C%aS1KKs)v#&s_7>8iU
z3?0nLIe6dkgCBg1I05e-?(IBhP&n_OZdH8*Ovr$-W^6i)J0k21Id((br3le`EO|w#
z1!Y}v7)EP>XnaR}g0s7qVdq;^5brPoAFR#Fg{YS^RBlKFG)iHZeXAGv;4n=iQW~J0
z=k)uBM@-WQF+-$x5xo_R!@vj{FF>d1=X7)6B4g5>`Fh>|E?b4+V)>R`sZnLa3^kU$
zDb}@c`0E|gi8*fCwr-1INMV;|v{KNUA`JtyZDj_+=&JrCDT6U6G|vfZ0fRSrjREIM
z(7IXOA<0CZ!#fjUtV8R%f3D6xKT_eA!Kc<4ZIT|be~;&%-NgTuhuYU0>{_EBF|WWB
zLAPnZbQmBYMZwyyh~A?G?^#i`(K6UvCE8JI$CL)l3kQ`nBuk?w0=O;(Ie46I4)DQQ
zi8APrY!wX_k>Fs|7j+z)D@TOr4(-_#I6LQ1GWl?4{Jc0$(XFwxSD_Z6ptcFI)AoA3
z;&?dVe7<5D2MR>plP0YT4`XV(92x{`&d8-<90!Ep_6H!B%t>GH4|2|0YX~l497c+r
zE$^l_=164=^&Pev8WWW|YZBO-L7GwBpo6s7u+vQ+mA3jaLw;gNpnrAC5aef`QpCK7
zg<F<gSEa%OC|lsAcE$buJ!u|mPiU><ba%un_m;v{VH8R(nS-%sR+Y*Y8^X(<;lWS5
zdWsRp!x3|{@JbuBhr#2&|AT*uPe1(-f{f$w27~=98Hh@1`e4+;_Oq$AqGXz2Ac7DN
z*qqzzvf_G~0dXkN!+SyYmC+*Oz$`@NVD3AM%e|)~%~eHbcoG9vJ2geMHZ<*M=L^OZ
zY~XSBoMd$OM3+v}5zh0qoR<Y}-`yjn#0$Q5lt!0rEd`}=z$0SK0cp({%bJN>2CJeI
zhjqQ~8IWt`&=wONchNh{YsPdqSW>%S7@m4u(~#gC4{`4p0gxgi6!;ACV9sj=S2d>R
zw-$EcuiLWN#KGqRZMwf#l#%e)H88WbmO{-Cw#XFSwRSG>v+P>-_eSua*>L|WzI*`E
zz%~cjs)w8{es=i?1i2PCQ5=ucP7@DC&j>oiDN^tl;*m?GH_A_eT08PO6NL$;FLYv&
z52P2hb*yz|0)1&f2abm$FPBd0XqX0k_~8rd8m?IK@+46ZE*ax6AnxrX&dz8KMqa|6
znKljs6b)*PH4rmYIRPA(@1hMc`2CCsK0-TVPNJ2=Gnku;X-sY(Bti@(5O2uK3Kx?t
zoxDW3s0orM1-o@4$2h=wk7c<sq&U#YJ&o4yDpn8JE0!grRNf2YG~(@>dvt9Wj|u4P
z{w&I&;yPb(dvnA%jPT-6+eUX%y3&;Ik03IhBYK2^Z3eA(JbU&GE?6suvk>~m>{dm`
zGzIjac>Z$0_kZv)zWcpzKwUxUe2(=i1`Bq>TDwE%z`(n#8x_ZQNJT7e$TBt?&&}h_
z4J-PTzpiKg8dg`5qSxM$ZK>PZMVeUPx_Fdak&7ApLgf9IxWs7ZvUr^-Q3>Y@rXj&c
zk7Z@hdtOQXjsqo50^oYNVi*%bG}BWf_~M;o@UK;K*4p4=z`Ax^mj$_0tR+LhxUz9f
zIGs-XIW)!b$n#*$E1(Po-`Kb<8xI?>vt`RQo;`;#SribkmK7plm!=fcBa=BW3k7#j
z>bU`IO%Wd3HDrc21OtT_9pjX1f}q2JGJ#Tx<ySs27!WhdS#m=&?cozV_g^>MeXYUn
zba%vYny}`;7pyR8J_fgUqXzFSDMOxw;`W2<othv~AOrz~De8w|M6Rqgnx36(AR&F=
z4QYcD1P`>FF$|oLo#m2w$LNzSUo50RYw`wZLia`p>WD}qluVKKG|}qq&2sM}sZ+6F
zOID|Oq!@9%%>4B!!r{p+7Gp%uE6($bVH|L^3_bwd(w7A>(7+KK1^8W`zQ-+JC<Kq*
zpMsxT+jLwGP9oee7`(p1M~Z#vB;$dx)kle9M;;Z_I{?^Y1-uQJn{d8sg_xj3?4V`K
z2DfqA!g<k^{qSc^+~IHlT1SjWzz|QOpAEOq4)8(H8@JKBXLsB#D+4eu^9%sEy}L!r
z1+_N3e>mgCi|6>otGBym(;0%(3Ilk(6WmMz-~aA6se`iwaSq0tyUCs+u|Q_e$deQt
z<S^~IMlj5!tHKW>rZk`yHc@m1BnG6xT$fJZC18jN>pGhtmKRY-5nG>euna+l1q0{D
z6|Jq9rU9p$5m6ko!HyG=*h(EpE*1GQ!#juZkZ^N*hms4H>x!lwhvSj@7U6TwfrcTM
z&J(4`x!OYIP#p)|G$d+Sy~pMK1<rfC{NM$~7*WfBhD!9ZC8!E=rt@`KX4>#O>89gx
zLN0XE4pTJIaAYT<dL$pA0LE#=yv)eCpf(<mT6n3%;8B%M>uo6$=a7cTiX|0f01e$x
zJjTJ~ocFk;F2SfQXN!>~=>Lje;;EVI%IY_O%Z)l`;mr)#9>>9k5gz9ALJi_HO_-(w
zf}=~X8(ZLlNAPx$T5JP{C%-%Yuh%OCKpF!8Z9$U0ro%*p7Ya#++AH|EdykuVWF?cK
ztZQB^&WBlATbQx4EH4OSm^$?x)CXIbY%~GAZc-W38&&OG=1RxqIwOq(?w;Lpf)+;n
zI1!*#WJBxJ*pI_uvZ^y87weiKLb1HUV{M8#&(?xr&w@HE>lN?r&$!GPF;cqFdWClz
z!rS4UFrN;*;Ke)iV(~unig))9xP9>qYhG;dIUq*i8Rk3`9YYvMk*Pyh!IxjY!SbLu
zO?N1r6d4h%KzI2N5XOYwfpx8zjyHhw5YOzp+KR~_J6^mv;a~jge}T{6e2#Pwc(2f2
z&>KHzHH1bXfOK1W`8;z8WOIvmlru@&qL(_V@SUiw?X3w9?=M)_6)#`DL{;9GQ=-_P
z&+^vgsx&KDF6w4Xl<oN#B8GHAt{J^}qa2B_pKc?jF&QwzXT6t-(Vk5%EV@ffqGv0{
zGG`1eqV-Jl)jN#Si0H|Xz-a#KWx-|1csyURmcAors*G7#Go=D-B;kUZx4}+H!TLR{
z*_cCH2AxpH>w<&x3auUDJ%V`j!i0xrpLa-4S-HiiTNk%BDgF;1JW}V9$z`TMVHJA-
z2K)X1Ml*@$@IGMx-MTAsCdE4(l4*&nL4o{rTaDXODDhW1tAB~FuQk|h6+QySHJ}%q
zPA7;MbS)ms!VuIg66GoGf$Es8<U0D%4rL4W^o%X>gF_mU<@=byp##mbmQwJ$a9qXP
z51i@mrBTZgV<NQwm8g`mp*8I(>~PEAxz4j$xj99%vSlP!-urI+7{>vpn+drUOvfW?
z$w1@gMe5DtcsNj_qfZeHHIX(;fyv-`W@vJXkzt4yhoS0$)=`z!Eoor%2^=>=G9#q9
zh8WSHXepW=ts}T#s9nbp(*7iFqW;z!IxM2b%nA(ZWpCah)wai{&zubw)+PWB4&EqT
zhHEy~6Dc6V!7LXlk;iewFhxwqgiFrwLUT;B%Br;#Y9%fgw5Eub3)f9kvuUopclJVy
z4NDyKUc5NrbV%@C0T)cf?~puWa%!)n`aPF|+8Syudo)cN5``e(lF?d&I66YTDDG}=
z(4`wiX;nP3m620V*{;eVudk@NVj9{1c|PAm95>nbj~Bds`v?(_=`di3-^AKyJb!+|
zvSeH?XN-xFCcP=9VPuw5FGz{;6w)0A7vPlCLP!zQFpwfJ3Zm79MKhkAZj2IZIHViu
zMVcB#op5(^G;Jv_y2W>>QVvlHbMd05D{f3j(|}Qr(mH=O5(I(@XA6ah;^r{yn$j@|
zqMvX9&@};0QSD;pP-HoGnbbK1I36d2KsuDQQ?QInrr*~-*i*^(cUkyMtm}$gDmJ0<
zukzIjP#m3u6y!4khVd$m$gGd16Yl6{n(*Emb#Kd6w+4p3p)1sw0Uh3?8c+^-&U0u?
zH%HQ-k>?3(imYl;6cCS)4iIgS&57q7j^l_i1=5txqj$ktGNT}J#o7weNJOso0yxH;
zJYLTT-t*aO9bdkGhv_h4G&u_td#~wmJYmf%zI=O6U7VTE4%38WjX%B9sXL|twape(
zYq?%*U&6W;)HcHf&w%c(Xaaox>J8o4>hS9IJ6z{tMu7o`L&EK|1Ev(=9q{b-2IDxv
z35W1nD|!&rTyeZP;&5|B90x#IG#{f!t(nWXW1CN{)GG%=1#j;Z^?IRJ8?wJocc&wK
z5Deo0b<Bb53jFz>{}qvOWxiOmO2Lo5@dE$n|Lgw+zx8{+jX(a6e~i~(zCvEk5YHjq
z+jCEfb7fZDkl5k?(<uVz4p7rpldAKO&AbG@f7`G>j3Z=|B8V`^*knk(S%qDU2&x!D
zG6sj0UaeC%Ii>-<vLD&`fFg7Wvn!gxi`f?2Ar44kfNDnr5#Q1}v_}Xi4H>t16;DUE
z2y+z~j-OHGsfSiov8<Ga#K07dU`<wCftCwGqEssk9JH%Zxs<`sU?+|mWy-pEr#Vm4
zB`Z*!B@vjmt^(Nt*0<;=4!s-D(rvinpXD9HnOo&~6~2<9uv)b$9ov<9>_VMq%hix}
zLk0`)$=-p`H#3Ka6HhI%V~dV+jK*sHDb)R!{rXyioeu-cqBWTaTX4BtnA)!m5x~4w
zEc43raMLK&S}{!{U>jSE6pS*Yu;B4}Ma~(=X=Kh)S9o=B4H$+4;5wGPz{kMD0Kj3K
zuoOFB8WWw>ti&!b)Yivf5ei#UG<J?L=e$Q*h;vx&|1LRW7!tAx_<N_zv{l8tu#~;5
z6(toDRz`?})rs+<@!N->ioC9n7_crgdI$)B>5pvqtqm^O0}7!j#06Ao7-GU=Y2n&C
zhLHzwuazHqe?t328MwSrjQ}sju5|8hgPDjRH?|IJ(H-M3K-9rU>P|L|or2r96t`2Q
zqKfEf7^t<PIc{dn+{QpREg4bPm;yfh=p~l>&p9-Pgz4rP=B1$2jM^&FFd+>CyivQ$
zTJYxWJD4<eAMm?5`~XrH&rUZCzLmCzME1@+A}=ht4k19alb#O|NC6=j(xlEDu!1x(
zJw6SIn$a;~U01ASMejh0asRz(;sAOV43W-NbVX=%hB+TF9wy6s18#3lcz*jF({w<q
z1&49OvH-7NzsI~TSl1PICnm)QbC-!1)U~3_SF8&!xHQmpc=89RdvY?yKy6SM1~wS<
zr=Z_$*aa+WO=;NarfVtHvXmN8g1oM9&SMxYrUXP=VxYEd7{<L{v}wnZGoSh3Fr^VO
z3UXmhS}lrc^Z*?5#hMksdoDE)MQ<K?<zPAtf^j^ctfbX8hkS@Mu^hsP+B)WWu^FyF
zB;asJ2!%OZzv|b+<zk|7BJwd3x!?g0x=ize=$P%du@2&h6qMW$Q^L|36e|%rCGA}c
zbyHazTmzt;4!qtGgAjF3p3Fw4rm}S!kl5$xY_YcLuE1Qe5IlP0ASNb}7*d2A0z|zn
zcMZ8v*q(f3JdFhxUoKZTA2Ch`_z-qIGCF&2p}=xo_&zGI<cga(V7eKtkCoJZS0FcP
zwL;+KoD1a}Ar6@51+7s3>pXBgOqiE!T2YS}90u?4;_is&cf3erAj0N@!!$<3=n<Tt
zH_~|Q=jC9vQC!Qq;^A_E_YRW<r~4qteZ{&K9HxmzfY#82&U-p?6P<_3rAM{xcbLGT
z_6`>!#8rI$>NU>SD`H>{tHglkFYoZL|A&8rPrv_bxMYvt`n`XR?=4rnd;1ycdc{xv
z?k`c+i-qP2hG{}lYVDdKOk%@%?Per1^T8%j%B?^}F|P{_<HY6z%k<kcG@={D>QJ|Z
zLKM=0^D;x)lWEHbzK2d6YITst30#HrHXzVi4E^%dtE-~40+^n$^@i49S+fDrIWt!(
zf*0gGn;B};69N^LIy)Pdo;^El#0i`Pn#k4#A(7!CeXZ!iHWLw=m#QIqwelU!))8I`
zwaBFuWQ$J{xSd5T&0>>!-}Ri%Q!>C@LG1a5;seVSO0}jAqMvgk;1i6cSgew>_ihp=
zx^xw0;xzeL>x^vCu4FL`ThLN02Ih%~r3|HiZ94XA4R-I}Jz$z5O3t{wxgqVk=`p}V
zCk!6Xh7-{ZtG<dUVj3s28=DZbKSAoHGz`PYyq;X4mOdg-%;z)Gz`U7KGW>zlf2kdB
z-@IdHkr!y|hH43l852CXJrl;{LF&Ba=XGvR%(1DMmyB1h-lEisZ-4p?9L9m-cP458
z7J-1S%(H6jRqU<83&q50j>9Fddqv7HjC>dh%hr7i$gKd08%Y|*ol;Q|taGMo2-E^D
zGijB>;lKb$5eiooXmvJ3jStTCy6n+1DF!nR6vP;rpgqqvWKXEtjY_gE#>R(-qGD$`
zN!XpWPS|;;sunX~Sy+*66dO!D3<+wvc&t0}`1W_cg+UJ}Ys07C{dF8}PB<QKAdUyu
zd@W1^?{skv!&eSiu$E?vW~kO3>YT&v=?J$e5|j}e>ym+^g=9ieiW=Hsp|-7JT~{0q
z2is=1=nsZ{uJepiSDf@@W{WKXhr=r!)3zTd;JTTHjBcvk!MOslA*eUN2~8tIa#&hN
z$qVLr!F9eG{S~0fR)A)X)t)=xsnP64e-9~PUS=*T+HpD@0dx!l^ShjPh!E5QLzalF
zI<S^WoR|iWgVA9k`dro-X^6Oac7t3dJ17fsUT{1d(E7wfxmHRrYR7fvXLfT#ed6^p
zTR=E)nkH&q+=Sq7IdnTZ)@$bgNi<*KQo!lRgK;whX;%xe6}u2BMELn?uNWe!TQAfq
zQUiFv<vM@8)9*j?bu*4|4$xNMVz8VwYJ)=vmX*n+jk+6i{@bswxno(EUFS1RBb*9c
z;gGN_i)C^;#1x^qp|y;gky(`dxzN>DJG)lZ5b#zv&966;ltR$atM1y~Tq-;qhH*gc
zg}HZ@lNeJ(bP3S~L)}JJMu`Q&s>51Y{gguF8Et4|8x!U=qt%WnMjXaOQ^-K{gnHXf
zp}Gj>IpcVoP*veVKy4W@QqD3d05sej4vgHX1utHla2N(?>-gluXQaZx0Y#0fFL}l0
zB=<%MZZZwlZ7CNG&sGXL6V=Y)h*~qIVYJX~qx3risG|?Kv=x4%E<2?p&NKd{w#tE5
zxZr&I(@zkb$2c9~yvIl1{0JX@{1S)bfFJ(ohxnb}`va7s2=Ry?{MtW*0Qkno-@zaM
zFaI+hAKq91wXr?H!wUygMzoL$XA}<g$_|2%My&P@)z&Z!8xhtZKB8GOfv8%@ccrFw
zS+NDGt2P`C6Ivm2(E$u`KuiHDPqgiZ=4~<w>42I<N424>D;?8d7_yofN|fQeC1=DG
zAl@M+>ML`}>@i2*vj(**dZ`f4&(KE)t%^8A9EN}jCNS`wV0Ild=`3EPHT0f&kDDvG
z)o!gD%x4sfe5#EDsCd7JgtuNz2W;}OMt5py6~!`TOPlS!<6tTRp81Vh@79VTzsZI-
z8Z@?WVSq!g5-%~6hZ#Nin_;VGQ@hL9oW)CK18~01*VhXut`%dA)MmJ?VrdJQq^Nx@
zE5b<o^)N={Tz7buaH29qGI->b)AsrCfd%Nn;o0pCLh#%!O&1W-0c)-{F}FQs`_-3k
z@a*mmlVvT{l~Zsn2Jg;v+zbN;oM+EYM3z9k$(;Gn1RtRh`1F%+qIbp3aX>Tb_4ar`
zUNYXkzXy82yevqbl-M+mblp})f<S@md_^sa!@*%~+1#ie$HNWgWoA@^b6A!ISp<3o
z;GMNcL{hLZ;qiQC<yqgtoFmYHX(AHj%%NHf>6q)hasbtgd7g25Iw0qaG!0m5L4GWJ
z5ydVdRj6Z#?~)sm8zG$sO9v4t*w!k-a%Y%>FIz$MvMd;f5#unBE(c+24`!WRE;G98
zkeYBj-e8qXI<GN=l5~J#p0CW7^3m#&Mh=&@%8bLv7Xr*YbRvks;biIXfZKI{H>V@K
zb696;!^UyMl2`uuYlaU2#~3YkhT4gkqG^ACx>TSsaJ9CEs&#ikU(Z|S81}}F+W2>R
zIG@p5!v}|FxSTJzEDS(>czA>}3ifiP)jwk3W^Dnz+ACV^c)YR(?s%Ap+}L4I3a$N*
zmn#m30nTDo?jO%Q-@M@6+xK|+!3&F=ptXN~d;rAZ{BXgL2BeYb!m_N;4y+3=-;X|c
zj;(k*7X}v}k0VaU!G165O*AGvK3;LTuDHxA)|}z|<L(w`=&fKF5>jxuIeMg&Se*vN
zVG3AtK^-DWp*(;ZEQ;<~3&i&ZVkWDmSeL8C<cvr@AWRXh6~r*?uK2JGT)*PiI3x@y
zp!JGm^=-;66C#_ljZU~$8<X}O7Y}Cm*|r~uk~U=~n^U9IWrW}UT_MGqhJ@hBY>@hr
zy*jRAlfSrK+EH>r4-C~6(AgP$G|2_&veF8(+r+_x@h!o7v|cTyitcjCH#&mw=R2f;
zoEw&F2Ar^^pgkEuc!q$f|BV_vht7&F=P-<k=vRz@^BBiynT-l*YG|%A4R*m=_}sN#
z;KN`sIf@jSzThnjvYCc^7zRMexG2n=Asu6Ii~tE9&rF}#v>TVVwTL_+Cl+3Sx)eOj
z1;HguV?tiHB~@u`>&+x29i25002)p=6TbNezl(2w`fWU3XG>mCeD>LAc>e4e{?otz
zf8%^!@XhajAJ2aK_aHvu#W#K(|NM^}{_X$k|AtnIp-r-%v)EV_2(_Fp0HAqqT`R7a
zE0%RZt{hrp956)Br45BrkZEG;N8`XkSrS1A01W%B6^IdWNLX_rinrQ8p*)+}2oxAE
z<~(xUBvsUlCNl^GaR{S-ibrY;EEY$D(H6PXdciOT#KG@UAo}Yaa><NXsSRr@4A52@
zB3kD=p$$ZPD-mKMoRkCY5h5y2DH%{G+P)|`@crxHKn;L^hE_IW*zlwZpo>LG@k~(r
z|N2)d3rk@Hiw_>&ns7)fKQ-oT?HrrrOKfm%`ecYGh9=U0M|ZWDtDWDE?sxV-n#PWQ
zIR*;0UiUw+lxBQ4t>^4p?5Ncn)#Z$C@EFDcH-`2RUCMaz{2Au!jNUpP9?$q7CJe(w
zN9KG%DHS(2M`k(+C*5UP@xcc#aeI4%S~;}jWkppQI#}~n&5RHTy?3yu&NE1nusjim
zvR=4B)B?1KRtp~11)~WkPq)Wi<Xmz=tsSeY7{`$=-9k4}k3bN`I<MRkbH<o%a6O+f
z9S-}4cr!+vFIU?zf%o@UjN^dAG+KWr>8I1(9a7*x`S|XUd1N_b7zgCiaC1B$x7D0y
zW`hTI$1vge0w{Gor8d->k%nQfG22A`EHd|S2s?uK@OZ|%$9rlVRB=9^p-RokJTGWn
zaXw#mb4(holdpC-H_$=ZqP@akIB#%}K;wq0;&_^HJlyaeZ?*t9bFqS*lCX^`=LL*S
z2^dGxDIo?NrxCfVq%T@UZ2}1Kyf8$}76~v-2RfE(K`U%h2r7H*36xoi3LrtD4wmD`
zfhD+rVH}WB#C5qM_l$9fXqCmzeH$j%OpVatV6>yrqTVByg0(D2Mt2Uw35PM^a2&BL
zGY*Fl-Gq)Q4X9P|?%{%UEm+<!xV=51)&;e!cyV`!VH)u4`7I!7V(*A0XB<Mnx)vPo
zPRMJ)`^O8`QZQd<+})m7VK)tM;i*dK{_%`guis!u5#Rd8#~6o%Uag76IUH{s(m3H6
zXy8F-4%AktzOKcD?M!`t{rVjk?#!RTbzVqoRkng`{g9;=)b@ZSXM_+zjeme>LrM{+
z;{o>%4<@$XC=uZNI795hU*=!Fa_9foSE~)pJ35gq^tRSQQxSV6(0etizr!od1Kv3_
znA3ku2#$h!4~3U4=doe%!6?1}h_X_Ok{s&a1W3`NRiZ{Bg0$szt_zx_^)I>XvJw$z
zrm(_Ti^Nu(-Wm!lSIsAEQ8LTApqty-hk%+hQ~ez;$=nK>y|LG21-jbVQP4YeL#z#q
z)C%@SM!?QN5g;!!)@6nF9^*J51}<uhA+btFVXKW3!EiXBG&Of<2YSO&3fAj_b!J7_
zFedc&fFVTWk}<C<ocFl7xxuntpbA9qpsrzQ#ojq;BhTj<<6-2GrZlLGDIyIcx>J;^
zxkZ&yO$T<3w_G42hofGod)7b~^8NjLly%1O=7{%?SG2}_%-8eh`112tI7~;}{r2ym
zw*qki0*_}ezl(2w?|1NbfAb%>G{}ZJb)(f2Vw#v=c_1<uLqLiIQNrwz3m2jJa)nIe
zE-6}Tfv;hY7IGFIyfq!=QW)SX9*Z#n;%IzKDKUY;WC%|M0zfT!ulk~fjqjr~bPxIz
zxSt|hplWGwwP6S}=%7;`Q#P3ph)Rp6T&U(^s52Qn-Sl@5I0<N;_hyuUNiA2g6QNsc
zX3g9d?r!&IZN=)&Y>+a|nkeN^G%?kko+bKg@mm|32K%*!s%`&aQ2Z9}vkhvt@mTYR
znC31B85B`sFVti9-CPO59xc>6?=9*+Y&<na?5l(2f9dP%Y3!V{!-kk5>uMYne9#7K
zWtfH$n^9t%MgVFWh7FyIxV^ol5VbcP4g+Ge>MV;zC~Jmubn!esJ_2Akuy>?>=|au8
zURQVz%=t<Swou@|<*M_eR!rjo_Ye1oDNr*Mc|ntk>kJO{4j2M%P7|EJ!TEef>kY^0
zh<d%kIpF4aYg{^qOYb61#}jI<nDdHp@R)7~T(4K8l<@xjd!sTv9v{wFYsF!l&<2LX
z-rYWj4<6Ucf){spxH}z?*A<VKgxU;|6OZeh5&VRl6|Fh6nL(A0eTI74#9gTD#1)4x
z-@Zq0iqq{84QhS0U`cpYygx61Q>1yp5IqnrcEEWHfOStqKP;Co1W52`r9r%hf|)zc
zIehx*w{W_>!NcVduV25x!}$SP6u<cC8yw|`AAJ06h&WuAE3@h>NP1ltq!{pUevffD
z!U;2|*ylu%{9c9Duv4)yc}#~1mw7=bX2?kaF-BY;9&uicmR5#uI_n^n%1DG-sLRQ@
zVqM;2N(r~8BYG_eL!=8<6~&(GHLq}9kYYk!d2p28(Z#V>b4*Ba!2SI_LW=m{gAXjq
zX6riSi~085dw3r(FIRl?laCRtVZ?jEe7)juWLkT%5Yqei4|wzTg3rFV$HU`{nk(*}
zop8EIcz!n_9~3^S<(dhuONGyhd0kN}jYJ{=OJ1<#4(B4CKRdvChs#{>?%@>#Z2u79
zKok6g)&!q_{$S&H$C_6x>jGyLaH@iFV2JKKFSuT>@F60lf$v{ee0l#4Ij>L%VoKC1
ziJ;~c<1pYbPPj>?6%}gpy!Vz5#|*;Y=^XUl;pWH&lv*3Q9?-jD91aLRVi-r#@W0|0
zGqHeNmCXo_x;Uo_)ru4uwkjeR#%a$?)6U^;I3`3-6U8VK7t!8PS~r~>(A7fO9pF1P
zut6O(2*hpKfiU_sJ-L!io7^fJUrMFRF#3Sj3I*`J6MZ#f09h3x2@uvP#lS_0CG;KR
zI2qzeH}If>^SokRXGAZUMn<}{CRi3~y?hGrt--@GxGHH+HuHGMVM19+1&T1*rj`sb
z1TG{ZK5V$5IKOp98!<M51Fm)jLco*;q%lFYVHzeB6R<-B^E@Ml#DHo88dMkn8)Lwj
z5^g@ag$SE3oF@$)0)u$l+R&;{_iRy03V!CnjBt#gsp!p9;2#C0HeAkUJU%|6cEvPK
zh?ZCoJcFL|^^7uK5M5x$wIA_=-~1!|<R^cJw%$M8tv*0I@NoZ#Id6gcz|HB1$IFE}
zIz!Bu@*sOa|Byzw5TK6B_;p#}L%`|g29M`Q91hG*Jxmi?&1i+jkenIebTb~{M5sk;
z9g;ktLUUjxtveYjgD$KYvjl=_15E23(Rs{y-Q!VSzkZG9&z|9UbD%yxBuH^)e59t7
zX$68|9ErBMfN6BQwzbyAWzl3XT|3h(I*?0aKlZv>&ZG6T_NUHP24oZcgWV%U4C6l-
zs$)#M%m!ffSh9b{=o8E16E1(Gcl&OP+DstM;?KJEOao?wFvP7|QGxz%21Sb*gM(V#
zBBRm7>f1K7_?P?oT7%tm7#S4-ZgBiSiPClFaC0~!#b6G$YR<I<F$NP4vtM!y-0o^E
zm<;i921PC_d+(e8-etnE4XBO0c}lkJ3-x(1If{yFgZJEUI~3F5fOqfjcPR66eFqkX
zM@-`p)0j|7L5R%GD~%ej5CSgO1+{ZSnAe6{FDQkYf}ATJ9?ukq=ZeRN3x|Maw2Z~T
za5!CYy1Qf0G3aKjtph&b!GamDufN1}n4k`L`~DuaRy>|BnAZY^NYY0A_T76dxuREA
z^q?y~cz%b&VWdSn2+ESt3Q@>qt+-ql<VHd0-6`TaXXKLcaDT?K7L4P77cXvs(s39Q
za$ik+OD&0W0U-?7S{U*&Bj=1ECW>ZjLAGAXAOHAAc>dx8JbU>OzI?ObvtK;m&8zqL
z{O51*y^lY|x1WCk0GO{Aw9cW{Ifvi}3^5tSThVI9t5;v_3E@1$_(Bnl>j+-(?CuB`
zJRZ&uG}ib8Z$Yq2&Up9k0U-uFdv?c{s5g|lqLc?XCs@~t*Iz!MRm(thijO{ifzxq7
zNFBM-)Ud8AoOrn4DD2h3NQH;jcz?NYoAwURpFaos0`EN5bHTg&*O;aW<Ct){TyQv0
zP7uR@aU76*M41~N-aNuVcJz)-5sa_6&I^K%SRPm8^9tVtf`H}zipd8&>H@D1boe>C
zs+iPIG)B9!JX`_Iisf;pjA8<grxOqZrg*>@Np(jHi++E9kGs1&3~8heG_iW9R2q_6
z>(mcwLs=L8_rhnN`Xw<}tD5uHnmNFqHRDqF+-zB1aDvCjM?5?}>~uvf9oJ>Xl2=@=
z*S*nTT{B&Ts+KwDQ0t0Z*ROM&{b#=@0m&H0i8@mYjaTrHZ5@V$kg{spSWo51K!g}}
zH-4+tDznsdK@QipGFVk$z1hGJ0GL6;Ifq^f^>AB^l=rMKRKN#zj%uUss^tnvVuL~3
zFPBnQXk&P*h-3Gt2-Y=2yCTH^iEMmv-Vnb`8OS=XklHc&2<H^&73u^r2t*^)!i&Z6
zb0(l6$YljUx98piEDm8{9L~HhP*paslx*!l)Mak5HYIcE789_GDsB!FARW#rk<z+Y
zhc=%_wvseN5lHe_mldi!CqEdU0lh<8;dALdm($vzwXu*scxdhD!k-_<IuLx>LbdHV
zov$<C9AX-bT61`J|A4pe-r;-S{VtA&gu$}`BhOd7dG!k796tQkcaaXafRC6CcliB3
z_}}A;pZ^u!zxe|9_is^Zwt6j?YS=0$je!`a5$l!4CTC&Z4Z!)j(m8$@kaIDFlzR1f
z$%w;%rh;S*G%+Rw4+gm7hE%g<tJPBYe)o>Y#|!V^;4zLg@ojp$tzIxVf-$ACx4O0f
zQAU)AbNJwcml%ctUOeVy#<|f%ku1qUK%_5($b8&VAqeQb@=mT5^E_k8TY&ji@!8>E
zuGQYyf-;PIc=c+4PmGZZ89MT+7}8`>F@@)p0*7PRzo%wN9<8@0-Jl2nRyL}=w<2IQ
zC>%lzs9NCNCM)UoTo@8q`Q8JY&Q96O%)^gAD{NI`bnh~ImNKc$R2eLb_m}<pT7w<=
za%PC#pD!4PfzH3mLp!h8c<aUkR|Lo7ft6X#qpM&#9-&U~c)c*0Ts-d2kI1FLSqbwr
z4t6-PKMx>MBVK^o5qyI4z#&8qR*8X%xe}`M!C@>7506(UIu6qTwRFo*d&F^?P)o!4
za>be{Ag!(9_V$RT9rLpAVY|$5j>tr78M#y}%ZfCNtd??2kG_BZ0N|;>9k2?f-Z8|4
z6r<IZ@TDuI;Ib_6J^)Q|ofpiO+w_y4zsBn~@93_J5idV@fiwh6Hz&kKebmj}9j=!P
z)CXLajHnUEG*SyDG;9D4rC~2EChg2J^XYJd=?J{Pf5f}@S41y(eE%6Hcd)HM?C(Pz
zU(OBnamFx=7>65ZYd8!8JR)XUAn=&y3;yn}e}o_Z-~Syx|KdGT8ljEux|9M+V=~$K
z{D{1)c=e07JiL|}(UJ0M+L6nO`-geg6A%-ogLt!+2k#KP;BWrwZ=t;+goJgyhX`;u
z9&w##oKCkmjuWo$XE^6^z0SyMMjGNSY(K;SORb1LVHyJli8#gs9^RbsaK7Tji)Tpw
zh`dy^Y?{g{cy@Dx)A0pf%rjyLI86LZ&yN?}9!~%SAKrbuU;6o`Sb#P)L88n?R29VM
zfdXWs-eZWQ4}@_B{JYXj(;Ab^Q=kyMw1V0TPRAQ~Q0L*yLF`NzJ`95$#uYhJvQSF~
zl<h5>=)9}$14pr5;cjM@5Sd#kLd2jObs-Sb<4{9F@i7N-XLS1Z88TU&gUji7+<%W+
zE9UD=V@3)nt>W*0{u%C`J%@<H+xrK!4!nK)02g?@o^DQPjZ5R@O6H)}iZ%c0RO~P&
z9Hs&2Vr>XSLe(<$ngY@(*srz>c33B}Dr%$FR;{8<JE-^S4bF2g>|mZHW0xFir#T7U
z+F1MmCxG+9K;71Mmt)h;3cMz8fkD;1H<Vtn^%-v(TxZoyfI!7A2@$|`)LzgR-39P4
zRJFE-!*l@i@XG!SstQWY7{;(ub=S*+B`<IxV9uG+jv*o?4twGqmg@?!<{c4F-J)=q
zI$tk1-Q7TiEkvy}I?R=_oet3C<+)^vX@g722x)*fr}oXw3Fn71#)103)9DCNK65!|
z?hTGcwQLbArSSXShKsHZYNi$KL`*|WhTt}%%1GlDL#)N|!Y^yZ&G7{99oFjw&Ge&n
z&G_Np`~_Zr@fqHH`3f(-`7PW&{{YA13w-qPxAEc27kGI4CEkAd3V;3=e+=~*hhxA{
z3QDcG6>}+rT0b{VhN`k=t9C_;l-77*U$3`H$5JxZWkt3y{%Z_y#`5J{aGVb4;PB`~
za2Q9Z=QHEI!?WiD^15IxD-qw`kdi}b3=VfTU=3+NNCTIg&L#|^7!55isE+;Js%(^K
zt)n&~sm>hbwN@<4!YC31^v2&J=Ymp6UAoP%)*4DJd)&%LAAW#bDy|Fl+G+3@4*}C5
z@yvo+|8RqL!7{H<1k=PdoNf-NO1d^E8%&bp8C*&w{ah<4dix%<!@gHCn6=m4*($~p
zGJ}wR&gOh~$*5mDYqwrpwZsl5-t>r`tsiFItELJ3I$=&<Yp{F!_7R8Uh#^JV!9mOL
zCh}iP$HRqvohf-<giUdNoZ)>z?HwT{b7^&~MG&qHd8ONNZ~^cRRTUmUaEuwbTxL2M
zjn)kz!Uu{<m$~8mIOFcw4RXy4sBIoi7r<N8g}G!?n(*-O0Ou0ky}uf?OPct4U9l|H
zf{Tmwvl6Z!$An>s5D+~~gNHg&z320VA;7BUHe7JM(!4N^Bd#-{`qS}fn>`O>B9q)a
zh$MQ1P~m;RhtFT4nJ(n@t9SgfouO#np-I7EpvI1wa0Tb{6~PD0*DIp;I2<MnX~1|h
z;PLT{hld%f9lE!t6N7fEU|wb{=$M8B#=|Dwgt?A7mh+6Z0C_HmHzSHP)RJ-SGY|Mi
zq4(r(P&{xOuekfcEr!XuFgtL5bH=;#JB%@5e)Ty{Kf1-IAAJicCd@NKM5B+i&<k)F
z-s8{y{B!Ee6hLk0!O<Jife;=3=%4)ozWuG2aH4RI+OBD&*uR*Ia2UoX@u^*K!9rY}
zCsf|+7Fwz>gEt4O-LTZryE9slA<|IOU?>OIXed@G4K=TZvIgp?986eG@on(IpnXF6
z4zMaFHL-J~?k^_PUhHS6d%TU>-$mj8_{rR%4Cme&OcW7j;j<uAA|Jycm_2{AUPeW^
zF8jc|<tk3&fGr+?6?*)2d#|%yT3c?@)@iEUF+RvnIS1J(>rE}JDCnIux!7eH<A^Z^
zynT3&^V0Ce+Y7#Udqr=RREY3Erqes%zCKb5R_OM0R(ZGPU!98WrJ{J!;{uPz^A!(|
zj}{BW?9G(;xk7AsSy%Mh0P#4T4!k6sL$YqoJ+NIB!^oPYWu75mi~}=C)#y$Ri_T0v
z+q1b<tS##dFF-35uKV3(=L0X)QW~04`&-p|w80@!ASk(_gNWTQkoGLKBDw?>L1W(_
zy18hXl}6;(`-IjCYFUWvi@+s10jF`mVH~W%qT7X3EmKX=E4z}Z$*gFt<9IwWzDU&i
zQ0eT9Hb8_lqAV4Pj);W3u87WK94B-roH{xxooC&i#jK!Fv*gw7#+{TB9xqoKESzIV
zc4?4SxYX1}RQ4+(dhdzKdMBolVwPMGBB-j0<UF1o5BTgCzrZ>qpcl6CI%f3MR&lNc
z^W_0weDMo>|2KXE<98h(1DY!0aKLxJ{~H(&5&!Pr|37h=`95->Yk<Y*cpuo(1IE^*
zVZ<>+`)89SJXF97YAXmsKyZR_5cJA-<C<5T?rw1R{D87Dk|=~kD%=cb#c2Q1nl-~v
zN-~-eCnblaEy$Ijyz}{t)6ETCyUr0*IZi0^3hxEi>tcgw!|mOjA!-{FLg3(VJVJ%&
zu_$v7bIF)?9e1Ol!lGXW>*tQaBj**@Wx@M*_c)zyaP#5?v^Ug&QUq_yW@;We7dS+V
z*x}+_Dghg!fQ~Gn8H9qTUao6H@Mc00pasqhO5rU;xcsu=XEj+2jJ;qx*Saq3oOQy6
z6L7I!7X~-0ZkdCQ_j)bnRiJTi%TfJhzrNOBw=NaYRcIBY6tT_=YGwy!ZA7zrGhkl0
z(Tr)J#b0?4J0JF(nP<;#(HgfxK;hkjG(@P*Jgsb#EVW`?3odcBZdpQ-xfV8Gq=2#(
zI4=k>Vrdn%URYz)J5pjfb_xMgoKTmFd}WdQGz>V#2|=$w1+*z#cNkoRG;Rw^&hQ7f
zPwwYT{oLU&;r7J|r{fWq%Z1KZ=b#@jPiGtk9Hs;M2OT*xM@=nuV=HTqfd`}a4mqzx
z_)<d71?@w{mQo+A%IEg>25X+}pLZM%Hxy-$5$iIO&R$le5O6$97>5z-$}rB}Du!Xe
ze7!<Nas2MD!FjsJob!0~*=u|;za^bx)Xyg$eS(i)Jcq(U8_lfHtc1E-4p7MjZ{NIw
z6yWEtKF8%UW62rE;{hRh{Nl^c;NXFv__g2tC-{wD`zOfDg5Wq51s@oO3c)AesQ9bD
z{u%!Er=J0UQS%CQg5DdR-Q3~_-}w}W(Hp=hI38|H>(p>G=N!!w{1VS0#hZ{=&@IQR
z?Nrj%85qO(ghUB9DYu2dsb%(=hSPgyE_p|%Sn?CL^-VzMm2=$Gf<>!%&pe;fd8v3|
zMO?9}uPzLjt(E2$v0h14n<gv-*onuJkIo}S*4DsTO_i{=%sC2$gAb_SK<9(?BRUgF
zYe%s(|1O*Bi<{@#v;CbN-oJlOeOXFS#a=U)OW79`R<;$pko@i#E0qDI5XP@;CZTRI
z0&gDP<JH@@m~%Ht37w1`94=cDw91lsg`%n8@q9rV280kW|2HsCh<HOb2{-0a;LJF(
zrQcsy#W*F5R_iyU1UI-{E6B5nvXkC9%mvS`<IClYX=Ke;YZcwJQd>{1b77}s#Y|_+
zl>~C9i6^E=G^$lLOIX+LVDgf6ZK%CtKA&+o9N++~nPJkyIH20aa-OedYKv$Btm}&V
z^A*E5;&`0kYd7>wFt00OXqMOKtPq~aQm+h!O+$i!1KnB*&^uBX&_to8)|~UgXsF8f
z)~AS4GA@R;uFHz?Fk%`av`{xl+S=phaD%)ucQS?mwN{Vb70LU(`DOdLltyjn?ePfh
ziZLaW=1o}77%UNi7l+`1qE?T{C&3mp(yCj~z5>)%DI15@hX=HI!SiWCIiDfm`|PYL
z4xWO^D2yUnE@!;{>=(FPGfI9$FDpKHegt%ds3KRIhqQE*&if=-EQlzx)drJBv|=(5
zVN((y%p)}&x&#OHg6b;7B?f4VN6kbO*>_E3IysNwbc5@{s2=aQ1cxJQ?9%39b%Mj;
zh{x-KxAzy!*Nk;#n0rM<9J!EIW<<e7Hmf|pIbqG&+I$*f^mzN`JrrzlLGSRO3%6H=
z7s`L!!E~d&qvpVYKLieJ0M@Sf;Mp^L^6`gUK+3|QEHY0N8*|XRnNX|%!QM><;gi0N
zE?^obb|>Sh^T7{;#S3+$5DYmxV9jNxb_EPc9!BQ!F=LN|XpA()I1#HzWX|6-L>xxm
z<E1o<(_*|*^>oKq3-c%P=zi&Q>}w5ncjJKps^Ss4gLKy93?bq9hacj4xuSOt1+DQ%
zGsXnm@z;0la4}ku?}BG{Cjf2lrrS3o=MZdLaN<~pCES7?fZLlJEX#s4Oh#cj)Uslj
zNP7and|41;#E=GRQk0@%x?t&V%j=5ia6pWSNz|<(L>7xX%aI|HLL_9%hzpUriy@6z
zR|>(e(*>u~2@j7CINsdg@$qr*S0(+L?f>4wW7Q}VUh**niw_}+bv#b|pEoC5uU8zV
z2@mH7Jb(5KU%q~W)9Hl!hX=fT`I7(r;edI$V45a8K0YEE#WqbN7QbK`4_FuK4x+gT
zTc_qrmAPn#!~}CfRs7G8My$P<pxc{UPS~X=0pi%f7sh~3J{fVkxxoj|pW&xJ`vorZ
zg6p!PYQb<wNJGMKj7SH*T&)ydhWi%77jQ}*|MH*zE`IUqHI}tlIfTQIB8HUk(TisY
z0`zJQI!7eahpn>7@_-EP&N)N6`-*KGfMo|ld9ZanrF%P3Y>gu^B@S@Tf-XD}xNLQF
zQ0&_@0HPkPS%zIR>XTVvdCj;k3(`0mJ?Yu4={?@RKl2#{>GRKDeSvwt;=`9O@U3rt
zgF{+wdrZKF?v+Ltcx$|5OUb5V<Kex`3x<@;%}!zQ<#O54ylQrPfUy}+3W~E`v>e`u
z`cNC04Ml<lcoX4i+IAFY%W;B5EVNbu8$`GM=GIAtsL~x>T1C;0fQUon&sDn(b|$P}
zt{IE8{jy6#!Z;n-EK?d{8o$nQ_MiRoJ|gFgYnzem&yZlKmkPSFzR)>}-9t#&!oO8%
zMhM<x9OwdOIB@`kIbFdA^qRq5Pipgs##S3x84%l`(9Sp!@5E442X$f{su8kM6C1rp
z>kVsO7<dlB`*-&YxjsMQco=P|cJN7|P0>q3UKoD7r5Lo{a5~+fwnj&*&|mLd!0AK@
z247M|^n&QqlY_5YJV<5M=3$ujpJ{S_ufV!F-SC;n>^)vrRxp~<-CX*7H>B}0R#oIB
zLlavEHmw(6b#$#GuPc^X7}D;DREqPcrQ>{=v2}Ep#?Nw-0%+&RX+S(+r?WSN)R}{R
zi?Gnf3%lmRytm3~I96IJ(C8GU>oY^5L-QVWF1Ypw{QU8P&-{-OW5i(!IGsj>4E+4d
z&ru5Xh2Aq&U<hdsn1@q~jiG+c32EkKVe5(y0n;=A+Ic69BckWPQc8v@5hyD%LkN~X
z*hwF@nvuo<!w?KO5!77JlmN=PQByZgBVNCKhx>;s&ewvxHoS`t<&zH)z1ScJTrV>&
zk7wLHzr|q~@o;$r8qqH?Q{=Zk{e-&X7@)mEYh`w1E!0g4a2Tl_6a|+j-MUph&>N)@
zh7<xoZ;`rurL=3!6<2E;s+B|B#-~z&+A)q1cc&wbht8q3z|2fz$%1nTA!40pkdqZ7
zk{TWE-nx^cQuc)gXuvWvJl;8hKQQ~1^f)`UQy37!Xb$CuHFMdhl{)!xVD?@v>~P=o
z?7z&{*Bb18^Vh$JF|n?ROXUBTwm)mSY)P&IvE>o5_deq{WZt|}jX(ia1rz}g99nJc
zR$I4dw3bN^GSlzWi(Vu>$V^6=NP^7)APInlns3z|a{Q(<>?tB#5AGhZzpMc_S-^pz
zGH-t8J7@3MF)Uxce7QssUV9+&<%Nhdozs+3f-jmab#;cM%2JXB)y}C0m6PV$tZV_4
zVFREVqlBDO>Cbw(zm~}wv?GAusk6R_GPiA07>xaTCney)!&;c-7_nV<lDc&aPr%HK
zpp`nUHaKqE26IkWtu|WeDwZN*b27pQLgX1?)gZf!(0Z)aE#}x_z3y;4%xIb*?^X7~
zjG4U)Qc`cebese<4oRDXXK;>)^9b6m;ni$(wU#LM;$nw5k63jLhF-wm>2$<yx5YS)
z*zI>1=L!3Z3moo`*lf2r9FN%TcNorRDF&G*@k>%qypm&Du%$T*_?{-IT0EZztX3=Z
z{ZKT6UJX<C7kg~CTm0Vd{T{|~!rkpXe)!`b<2;Rk2c}`bJWQCH37k~_!<zm{A|3@8
zPcAn2+S6;id~*jPM&^WR0$h&xqwoAan#O~(FhtDK$2Lz{Qau`NF#|zUpq!%GG)O_p
z9JN@+d+9&x9a5UqlTq{?eLrBD1ek-R-Hdp>%^3Ot<^u*29n>%k5|tVgl@r#Tq=#9}
zGY`ij=2+P+(+;L6#>p`yq%8U^fbsJ47x?K9eyZs)vHEcf&?K4Ktf^+Hv?49Xw~8zM
zfijcOa(wv@spwbpOaj1CmM@(Ita?iuW1F`eashU>w&ne~uzy(t1KlV9K-&+=)5%)h
zELDgC`aGbb`-;w`@=ns2i9IrcMB6dp6i?bCUDE359a0s)?llYpLgU5WC&^*jHDa1g
z?8<j!^&N8{#LSqp^gC^X_#37iMbktCM$#@C4VV)$OK-@g5x_UkGrYDHW7bxTNh5S3
zP!mK5-$>iMah%aOsa&)Gi9jM2?|dn4nR7%&LK7T<Ux7(j<Tf<Oa0o7x>buMltB%0r
zFpVR?!lX29qrqsFtU(ejsCOhFiAZ&Rj2MLp?m{E&%vf~MtSd4lt?YHmF*5-W){~s1
zqB2X*<+f=Nls%osSuBinzIg(+`>j;StNAiWvNh6WMZi!}mageKVck-eVhfNMGkfvq
z<0xPkoRsM%HJla3$7xXCTjoiN*9h`nG3_-Sl&n8iD~l)3o*)Op?cpBn*kawT;Jm}-
zvuo^EEBKT#oqF87x|6H`$+~KSN6f+yx!}cgm?gC<-i9J}=-M?lTWLR-b3)g)@JY1X
z;~<?+t`tlQxXo&0EF{T1N4U@+G}3yDKtgULDL|a&j8(TP=ia$L<ND$fyY&t+3jn>m
z*kX*61Z#5!1qX5s+VvKofPNSO5RQkVFlZj)Z|EFOryds<d(mbFY4O-L0f8Mz{V`YD
zwb&3PE`ekdJdh+sotTjzngn*Du}4w^en=M<R=x3Ry%7`WQHnvPDBieDaJp9m0HGhm
zK1@1vhy>;GVv9Hpm^1^i-M8}nDGK|Z<=LfKJRPsDHW+##YrM7`4ObgD3UYopDaa$1
zcAD!}#-t5WP~<3}Y1hb*wu!^|<)tR?1ng3r;o2Yuo8XeV5>c*{MF$5sWW-q(K~#N^
za~^FgEyO}_XhJIygp<;cq?N8gJ2qHLE-+(C1AOB|kf{?iM~QG#5@unZS$yO4z--%A
z7Pz*8a$dZ$n$TdL=c4~%NR`_>4iI<y0o&~kecxlV+2VXUA+)XN=kz3GmP30Sd+aYR
z@NhVyZCc#Cxkc9hn`h~7H>q*4F&UgQeAB|aq&`><ec!9#Rr>7CQ$$Xp`<TWVoEhiS
z87XJnKRn>#;sS@m0auq-I2}${uQwR`LAnC;GlHAp$V(Mu)8KG>z<#&Kx$m)CuW`J4
zz}3YChEWWLhoM#_7r+jwY6}z)e22rsQ9eJ9Sg%%?2Z^Fi4@aSxol_4h)!ryO_TJSM
z+TZ^+e)N-{;pWW^yeI744m>0rZXcv#SOv7w)j|%h;c&zzIDGBB_u%Fc@BzF13jh3n
z{O|FP{^TFv)1UlEG*v7hEN2F%glQZ_gBCQ^m04TMOppfdVw*BWX=xU8a|Zw@J7i{%
zDQB&6A{Z)4u!CLx^yEs)m6?%A*iZXB+x&sr0*)1oRj?&LHxTA5jmshhI80~R@bx{S
zvv(Q{3HxBzMf$yR*suzEsjT5#&|H0qdyz1wVwP$eLWxY71c+s})HD?jChO-5$Y)p=
z5J1!U2C_+5M%MJU3HCWl&%_Kalu1_b=U6s9Z+GUT#Z{;F3c|8NfgKf(tl#G48aWi(
zHYn2yW!xC!FoWF!$P!_kM(HCOrL~Yr6l5HQ6Y#DneY3q*&U-JxtT_URKwR43K|QjZ
zcJ7I@=v~FaL70qTk^-X(>;+sV(fH14akJ_=t>_C_bz=5AjWe>7acn|^h*&ZajTc6l
zlN2ym(CnN5#$gyRq=;3!($ubuMNSd$0!V@r34z>SKtl~OC5W4e_^Czh3=w<)A&Ab{
zIS^-Tw>!ifF-#+7o&XnQzjR(em=n|9X&MD2fVJHhVY}PGJ7Db3IxkV!duYJnycF=v
zvlN<0aZw9uJRXK~2|(n$q9T6@L0G#<D?@Y62+oz5C2>ZMB6KHcB{-yF`-6`^z^n5;
z#{LHE2sE))%dgR&d%Sx23Sa%oM|g7i9u7AzFsCGeHZ7{DbCUHAt*9i)5=d5s5y%0~
zBT>dU%ov9m;FGK{-H&t1qFY<9kn<#aln2_TlWq;1)z7X`V`!0|1Q3R4!Zb^2@Zw^N
z5WH%m143v4%IKoQ{lfu6-%FP4JmS#zc=5$+eDLf&v5i`Z#AUl#W4#tYKTndf%`6>A
zq$uZvtII2-oTR<yG$V0TgKq)W>(w5v5!u3Y?lsVqH6V<p_=OS(s}S(shnGk>V;Dya
z(}=Di?6zW=G|W+58c029lLT21Ia!SIZ__-Z4IWM7B|W`v)tXEGB-QAhhf&h8^MrM~
zD_Ue|G2JkX=(NCSn1o4hy>z-F0^o&w;>04ui5W4D(n0A<zTOGg_2&V*uG7v0PIYsm
z^c-ES^gv>D_Xe7#!#F0K&L`~ldn6_#0^&5urr$79idxF6+9fM5C+P{-_dUiat#vN8
zJ88m8K%NtL);_M8(KU_e{aV?S%$ztzNnhPR9I#riF{weZ^P)@eA>iij4o%ROV9Yq4
zPH5X!A|4h>z8?p)O^0Ea(0C^mmZ8J({)qFjM_vPO-rOQ(F@gNx!w*4xM(aDoF`+;8
zk~)kT9*qDnACO!^o-^iQ2G5NCp~rX}@x|eD5vy_lMc{Nk!g+^&ID=-<!yM8*&c`#(
zcLy9Ej+jpq91-5!y+Ye{82S;L%^D962i%1lFy!Ey<{8i9FW{Vr(#7&+hVNF!G=m|V
z)TclF8D9V5=hEhBQ~*fgw<Z1dxafd#5|d2;T1l7XfCC{WU`x{5^3VU}zZTCVWrd*K
z2Mo;TELX}YiD&-#j6?)94*%o7_#e<X;4l8&p96`(B;O-0{M9S0uv50N15z>%S%8_O
zvWZz6b8}6_a?#l^thqKTLzW4J31gOGhQeOyX;M`>qW{VEotnB8w1)zSvZl8TRGZ*l
z5fBug>7lOH1e~Om&=OdcgG{`W%6Hix`d(=lSHYcuH@j~Gt)yTp74QVTn7Y9%0iM0&
zn(F-l%X`hLYax&}=_UdwGnRE<@^b}%%vknIbfmTiTp3@NYZdNq_0n;h$EXD+!lI`f
z(R4vl!h=*=H;pzz&(d$VYg?_-_DC$P2F*f&NZQp6Ll)g69K2mGN!24Vm1<_+2&kH7
z>mU(|X)1%@e1P+WQ-78&B2z-U>M)Em`ax23Z7YRYO>ojyDl^)4CA~MvW0-m|v(|QW
zSt|kCt^)(o&Bh7q<+LX>vJ^VZ5FMm=3R-v~>HTRQ(U8Y5h>m%+YO!9gz$_Zslp>s_
z_!0wwfW|p=ZHGK0G!#Tg?#0gxfsi<%A16#o45Tb`Q%u-p4w%+;nkV=+Adpnk&RMIx
zllcEh=N4C3(ni_26Xvph+qRX|CKz!ZaXOs@1UpyfNtxRcxY*w4n9+ETC-1)pf4M`7
z2l#aZPXR~{=V`>B|EIsk@!^Eip~ojbe~oW{@Aq)Ay~IQRfH*~Dma5a^=>YN$n{JI&
z*J0{=VLhQtA>`aS^+#-XJM4Bl>^D2f0%U3XH;$9^+Ai})*tpZ_gg8YUPiN6?2Fa+M
zPbblS$BfW6xH}v%#|hV$S0d+WrEO$?dVnX<NJdUL9;I-vYde)YIb2>`gV~{J#NRST
z@x2UQ%#eqE6!Iqk+l_eHWg@^6<~d_J513~uqS{~Vk@JKp&giERr{f9U2Moi6i;FAx
z0JID#Oau$->%>8Ua|H8@Rp-#PJ4niNLfZt4Q$p7T9FHgT=MmE^b}yMT+D?qQNsCj~
zT`P=YL%8x6rM;b3GiWR7EZWQjfE@ZkH1`jOUe=A;#2t@koK6FpR$8U{Ag&e!!Vo8%
zkAs@ruZuk9%e>wR*mX^doMf{V!N1py&|w;r8gmk+IZ5T*EQ9p&^&Nik`Adw`2=a{m
zW{qc0p5V##HA0Y__~GFWUDsh!U7h&XCA@iai-+R@*Vk8YNuu5R;{p547V|J-x87jA
zTH|==i^#HR8cBVr56t0o#O?78P1}M1Y&To18flgC{N+m=jt_YD^r`x)357q8GXfDV
zFZPHGoK9yPjt^*?7HtsC(lm~0o-F0b<0yKv{eFvd7)79-#TV{uEv}ljRGJY<iphle
zDT}RgiqhsJuq&xX^OCcQ%Eu>>=nOuGlT@S%ea%IiA+)rey-7ic(_hcbNFZ#kuR%?R
za_r}X{&>X0%`Kc7JrYR;l~sq5dLT)8rvXABvKd(Gzv#K0IbwnW2|c(uYtKv8A~2Tp
zsY_Wa21V#gobky|e~N$f-~CfOdHNJL&tEDCA_Z?^hhREEB4d%1eInz=vIx4he_}#b
zc1gMim<HZJlC^iq>dR(qT2^LD@Qxfb)=yF%idGpjp(KbE=m1thslFmB{9|E9E&4A{
zl*L1hHT6)~K6{g<Y>M7Z;bct>*~n92`-C_oWgV0qH0_Xr%hFEC#u2bu5V)*hy^NdO
zvpL%XfYZYxD<fhBD4O3Dz^2y=7Wa4ou)=gd8U$;r1DPkmDeG5&w=s~Y3WDGAMVZkL
zgEUfyUiUbP697v?^sO59HqId>R0Hjtuxc7KO~5d*niq>uI;Sk=#$6-!3&X6o4=upj
z(ul+=CPxNb#5@ZKp5vg|3DM_9W%EM-R-IJv&S?hJ|IWdo@gB}e>$~ARs5z}X(>O*X
zlJBHB0?y-nIEyrfq+&NKU~&MU&CWNgPRx0|L#uvg2~s6IO%u9Ss<BgwXncU{<nV5&
zMI40<%9-Io65x_#D`>Ssw`%d`<`(3{B%8DJ#2$uT^vRnw+S~w*^{`gX*K0p;W-WY~
zu<F*LA=WCx5FH|CjN>c;R*D#-bdhM=PP%*~*1`i)WBe95C-mnt=1B?)R;v~{u@;dj
zt70s>vNELeVz=t>yWjl|=>04F=*1^^@BL>OdjY)jX~r*p@CnYxh%^&k-MzpsUcAOv
zzWF{r{>J-&V;s&W>@Th)|C8leo%;#5xA%DV>=_;oM~q{S4?g%1-CCN)ZZ;bg+(#g4
zB{~ZbXD2d)<N010WiR4?lIqQ2m@$vy=+L&JolMB^1TpfTL_5q39PbY3$FWETHfyOC
zPf4uBeDLtWYbTu%S*wlv!voSh<N9ikVU)`1)p~{7hX-WNIQIkAotS#x-royAn_2vC
zW5l_at~;UeQXSrSAOmwsxV^svw8dCv@oRLBakbyz`f`U52+ns1O^5r#8TSt-TwU$q
z9Amu>80U!9szK9;HOgk&A!pHyl00Ax^MuW2jcFJ~7t0<o5zeO(LA$*S=TS2&BDcxf
zWg;=~>dgt~)1dZg5}*(R>vn_8!1;7Wx034b)oP9YBvPP;nnm#Y<zDXu?0)jYPjInb
zVYOPp1xceia)qssw39~WF^jk`lQfUkRF!j#!)Zo7&bWVZi<}b1X;Kd^C(%F_Z#?Hj
z4|Ta&Bi{G;*%!CcP&!83HB##BE$zqY(RGw;n9>~?j0q<&2uIxM@5!OqA>i=*POe)G
z28l%Tar?z<=!pbJgq{cVc7ia_uz<3hdNfQ391!m(93G|ufDOO^zzJgkb@sCE3MCz;
zXVnZsB+;({0!YX}5{!k>$~iJ%Bf?*=9RuniZGc>~8VWf8W_SiXd3Xk9<D1x8rLek;
zYf4hVhFOS^0Ux%p(}P;AM2g~9wJKP+yPFqK;m0y52$OVVqjWSBY~39W_=~^zPx$IL
zzK(DI;dk+8FJFn_k%Gw#sK%y_tC39L%Dx#BWY<lFYnD+pcF4Ipmy%Qi=)enb^iHf4
zQeHX?v05f5I#kxIFri$A<B>O31$i=_tY7RFhew&u>TWJME@c@rm-{lUPR;_L+=A5r
zDkE3Mh$>SE;^Na*OhL=x#rm1T8UbS93g*(Y(HOR@^dVW%i59I0u(P?J)f1|6bxW>P
z<_XH5IxP!`UH*nKwZH9)96(O`;}VIlUDt_5hXJX4Ya6UrE2NZh>IcL*VY885%V4n_
z8Aoa5MZ`!+bhOhfhPt*^m>F@BVhhTQ2r(dcPK=&g*GPsy`*J%1`eDZFS2q}^h>t#c
z4_za3F?a_Lpe0{q39FT`0J8%9l#*!7ycDGQMnvNC9I;mKy4`LshR-Qs=tt={-L{yf
zNh<WF5mQQ95CgQWi~~XZ+}5iV9Er{^0eJc94c>e99@gs){Wv1D4%^KRhr>~IQ_?8i
zInk7PW%{HSEHKRzR_j*!5<9UQN^?evQXy(7QPID)SZ}wb;36_>+rTKSxnr!?9n$16
zg`}As>2Dgf_(|?50FW~%&C-ss3DO@u&9k<z>y=Qq;76)Ry>r;CI&8Na(0YU8xko?r
z5<cbxv<}-RTYURZeiuLf!OwB?`iwL?98NR-=D+?Nx3BN<5C730Vtu{AY3eagGgj*s
z1Sveay1YV88T*SpeD6S>u-k3WG!APmxD(_5wb;Hm(XE0ATera~$y)WU0bzu9gm%-x
zk#y7m82jx8=TnbyK4W$Xo*WuwIhm!+TAm`jZ?M}b)8rjCYA;oSEs~;=?QRW*q_oFj
z1Zlv*U^pMJyWFY%_=IymVZGj9x88!O$1ptL;_?F5R~Hz^5y3e*2gzZaW{lG;#Y`l6
zy~DA`93^Y`!3WQ<+i&1nkvNQF#C$&D)th^~zPZQC*Ejg+!}qXWi4ULnh_-0jjJ_Wc
z=Y-&d40<?7!Pz`U%wq=Ug#P>nqnhD2K486BOF-y6;w;+wpdCfXd${CqxxK);6RAfC
z4x3I2KF4Vmb90@$nG<46qQxU8q*qJ;{{5HQ$KDCpHE_sN#1PL#BV-4O0hQ7k!y<S%
zG@Qj1oWu;05-=VIjK@Jdot={+i!4pm)k+B>w~jatQ}N5nYQ~tN6j(&EgH0BycT_fg
zDt|iF=~OW}L$o|X_SiK8$czlNh>#PnuoxMZszX`KT$m46ywVEz(zFTd-<_cZQjHZv
zdsrQ%4g9*I0R$u^E#D#E!w|7Np&q(rE$KLO0dHK=+@|ZX);ZpZR)82t><YC{`e$T2
z>`@MXsMuW|uK-{n#JDhAb_)C8k;*-hszfx9)RkXA4Mfwz3aE(*o32AT_ehZOniFGA
z9*5%*fAGCO#DDn@e~$D05mrm!K?T@ZlXX@=Y2X&*Do`MMh7~Z`^Rsa=@ULW-_>tub
zkR<hjav$}*#x^a$Agal-0UWjhg91wQ`GHFc){#?BKfNbAP!u5w2zB<`0<iUSDig>B
z5ZV1L(2>qBDz+Q;J9!Z$5>(%+sZp8ERxnY4p^U#?5985ynzMg_XCx1gn2{fSlR#}&
zfXS*aE6;ZMR@C|Q+rGqbUDG&?%$s2tMK0ojdCmYQoX-#N!QtGCS?xT{(hFOK|9vm*
zi9`-EBBhA7?J$i=J+Ocfg0KP^*zYd1-Idq2YO)Tx;Kh7Z*%!(Y?_FPEnr5t<295Rx
zzS!*$oal3J@9weNZna-^2S-vcu<AODlNh<)-rmCdfa7V9%4`re>osofAF$g<+ApVs
zt`#q_rU{5K<Kf`}<RPAdZHwEx*T^|zzuRG+Mr3lhzdzvI57=K^U<LpmaPB2rv)=B|
zpT)3#M#9DZ0@FO<^_w?hLJ#RpD@>)aC@s$CGl)dH+_o#kX$E=1G|b|m2w~Dp|Cw_J
zDWK_EZIkFcj^|!vJn)#3c6(qcs#3KDn&3fB+z>d6MN-=|ngLO>@kTODPS>8LX?NQ-
zo?KrdY*uJ4FYxof{?A|*dzCrPnBoXWC!i&K^v$OTtALkZJRtTNlnBpXJm7!(w||Xq
z|H0R=d$z^~YjhzLkH)6+;3UOXQqU05kJ54F{^1@(z-GHv%c~asB${X<30ing*sWJc
zamKps5YvoVy*Gz*59ffnm13c8YTz4(wq1!IT}n8gdwlreGi~)X6hr=)GRSQ(%@NZq
zg)ZJZ>@P0Rc!xPnAWj&^bMg1goX~b1a*P<x6Z$^l{-MX`FYb{!;)7>T1e8KFk8LAH
z$0<rXx0D?+d2F^j5P8Jp;2S9t3TmtN^36S>Rh^EL7CvsUZny9*;MI!<WIY?_X~x6h
zfTrs(&l!yiSalM}ZJITX$1~!bvFX-g;(T=}mN35;u+mOEZPUQJfV=y9T<o{#npV>P
zp#id(ev8WhA#&7SyNr}&pPQ4|T%FEC>E$kMDMzIF%Ucw`6R<<hV5gorRE|ItR;c8!
zP*OPto}in$q8tDQ{;2$$HO?&n1$dO2Dbaj6q@kaq|3Nh!<zdH@s~><R=oF?4ECTr8
z%nBnTf~9Akepk6S{cj^^jCmnqP1D)6EDxW`VieG1e}}+D7!U*mSmvY$T0N+2ELbbI
z9^bz*Z;P~T0aLy*K2kmG>nAE2VL-+4<rGFv8C`2p8kOglQ&LYKd1KDnRlv@4(v7Zw
zXjb6L3r0`g>{zbD+8!ATL{5Ns6T~;~I7LJTD2v$i)ytRohkyDf_`^T`9{%Fb{=HOr
zQDuFI7xRVR0#nKwCL5<RN2sJdq19p1@k06p3P_h{Re-b3cN0_tT!4%{mx`Zk&emsY
z@TSZYs{iT?z)>wy%1Nsx^?C?uwW1ZfwgJjos`E|eqkghoH`kvTz^(I^u+$~ixl9^#
zDaYyL>7|^J%hHz$Rb#kpPS8TfSO;Li)}m??m%)Kvs04)CW3i<FdHMdug7__7($k!9
zI-Z5y;e^e)Lr%bH5C)}f#o}atu}8*)6jQN18s`YYh`aj(LhG^Jh`kamtYtD1+O9!=
zI3Y)A?=<!!<}~5SlWQ?=^}0T@FeKfo!)mj_H1u%Ph&Ii8%=3h4?vdt*-Ace%iZh0(
zN82?RrcqiUIRS1hOI3H<Gze{5TKf%i#27Q2Z*hM(<9Ik@x7nfz9%&X|OmG%oN-u5A
zQq0mnc8-|u4{$^{oZjGgIwLd<j_M-7oJBC+HMqY!04!<4X`Db#4!J<WK8g0N!SQ$o
zIVUU#2u)~k>PI*yRnc>dI1VRy1`PP1=_>C4CN!;>#KsuGNwNz9d<c#SZMVYR!vp4N
z#&)$s>(?b1YC;1Khd7H4n8<0y!j&L7v{I-iffs=^{TPBn<cMeQzmKo|&hO*%*ROGR
z_Z;LMTHj(Adt{cv6biumA725U@%*zJ<X&2oy?AwpH@C0xPyg8;;lq!gVV)<n+cnPP
z8K;Mnw6*gRluuF8%T3#$^$yNK8A6ZaNkG)?{Q+$wHZeJg7bB4b9U!R`?$3kRNfk|5
z20}oyTFL!39_?lY=Lq)?Vg}r;TRjV@fV@02UcP*drfG3J9q`q!eg%Hr!NH^L8tg9}
zT=4KNAZKCeZf_pY_Ytpej<~%)Atyo;9QtFzs`KctdW`43bkZR0&=6G?LfYf}a5!R^
zW*ko95OHyFf!2%We!bdZxASUww#Mz<9h~!c_T(dN!?wn%?J&)=0#6}VTJJGVeJTD}
zbsJ?U#kVx4q@;<o{4#mZhtJ*<Fh58^(>%;t?JU*BW1NwloU`LJ<Fn5{hjR`WmzNl(
z5jQt)01#HIPTe0IrYU{tee9ipo$RGXK366OkP0|18kx#E2%AHyeS)U$DyU;v+DCtf
z06SkD7OAzj02cHh5wKxtKDic#d<Lr-Y7w&+HVc|=lZ(lm5tFb&7PaM+%XQhye)kb0
zsp&dWF)k5l3c4~W<q%?M+QDf$O;c%HZc*N!+0HZBZ1f=eC2jbEbs?hCd7xga0nW$w
z0TQ;!suTt5xT=ZDb%IL40hPm_lp!@=G)vW%@2u3JoV&aL#GLAVDoDr*z{p@Yky_-)
zXIyMIn9>n3W^q-I2|xJ35AhHF=nwIy|MZ{WZ~n_)qd%IVb%hD`TFwhF)L>P<!zwu0
zkf4Dj@{h*atPH%jl&z(BrC7uf6(-BMIu6A4kQ6*?D&2H=Qpn&GlvU7W41iP&Rx1Uv
zph{E6j!1hhLN$W&MH#AZ)X8e>V*$QU@Xc0@iW+$7&-0R5Q&v#QX;6w~h{fF1f;j65
zYae1dHThk&RMDt}RFqj#;N}lyK}#KcETtvpX=v-;BO_<|y_~=N#`kMqx=;0|9&wKF
zt+YN`X?6YXVuzG6&gT=Rc|zB<=wmNlT*`1xaYVaru-<M!S$Z2cZHsB1aPFl;j3=>c
zU>1{d%29evdyl^FktP8@!#Kcs@vWO<)XoHw7R@PXR~o|MG@|h`3qJbzLu~~mUU{d}
z2{B1U{bsWUbHY4D#F)@F9`iJzbq<H)0nSUcV1RgCK0Msx`ubY$ErN1F*R<%T9uMaO
z?vH0Qs}9L$eDK~Act_Z6HW-EpS67$VZP%Em2|0K?+&;iJ0o(l+pML%vSJ#(F(~MQO
z0$jkHq*wPGMV>PZz1TjeEyOU*SgnPvyS~0c-}eX<&~7$3o=*tgl{DX`TO(&FVlZ+4
z1D|m22W(a=_&|92>>2>zbA#!W5IN!FuY834Cg8=3=bFvQNSqNm@!G3hA+%q*dhNN^
zqw50NRf{L@J;4WG`541A<Gl|*#$SE^KZv8ltbNb}0WqRmcQ_o+*j#S#UW$1A**z#b
zBqp5C34ilfpW&+?eG}w8Ztri=uDn)h&X~qYfaENdob6ifH>hbmal+{|AY}o!^L&Sk
z-3HrrgLFROV!uNMV@}d#Vt09o@jQZ^Lyj2_hXWqYJ@RV^uyBNrKl~8u%^rO2aXL@9
zyMMrXyM=QO4~G*rn;n>?(ssY!<6^f*o*1tm?vaw{Kc&CCq~4{VCm|0R`}H1q7tp!2
zdf@^@8DDw-BY-E2=ZuTXr?}Yc&@>$+cm;rjIM3Mb_VC_g7$=-gJ$Ab-hzOf*B|%K^
zYPv3G+&91aRY~J|DSik}y(tx-NENG`9jmTE%CnjpM=83QBqi&d%!fJ7%1<%86XR!T
zd)K1N2{9#{wS{T3>fm$0dFb)t^$mt;LbF=qbUNYq@)csrIG)aMj_}Z*(KQV=>opF?
z<CosX-U-;11LP5vhQ*;^zE*EV&bhJ&awMrXkGZC-Y)3a_rHn(e!|v^iSsX$J&`N3w
zmc}AIys;1;<1xrtKmlPnwDQS427~1q!CDz5<FHw+kyBJ)m8H44rm2V^=8|M=;A3!V
zHJT+grVZ2Ca(j@&T(F7W8`U(LiFZpCBe_S<MJnus<?*Wj5GK#d_2&!<LD?ZvV@fV*
z+fwm}n(}=FpNtvwQgi}{7A<Qe3#tqc>7iY(RjP&xNG=DpSUp$;mvaug^%}#&0Sqy-
zy}h}^U;XF5!vFfe{vYv;-~AT;<}d#m1aW*fgW!}C{DNJx>lHnw{I^%z44VyH7++Vd
zIvDVPcvI;db^1JnE#ne&R85~26mXqGmO5X~ZPK7Z9<4BL1^i%H*kC$IC%V4!h^rp5
z%+bZ1T?SV~lB&+63_}T?sQCWbcjc59%h2Qp_AXMeDL*&wIxdtT%HNG0t)ByscA2qp
zk@;IIYn{_;qY5Iykbad4L^G0Dyr_G;jKSZH-3bBMY}ROeK<0!wMGRw)uJyRM*n>HW
zSC|uzNruPe<zCVzYpDjzS%lurDga5WDpp;G%n94g8Vv=IC*<JOQ*cHT93m6W=QC1f
zDL@;f-O)TvNU${_TC}d~K*V@*eSy#f+}+$_n0kEh>;nPiqs;Z-o!B7_Qvpd&pFBaD
z66Ta}>QA`3yg=8ju{*Bef<r$H=+8a6PIN=(e!w`*IG;}V;Js(a8QAUi`0(SeBBqQN
zuU`W>;>pzo`u>RB?h^M8M@++rW6sDq;zHYpHQh?uHLljk$zd2G2odY;8gZI1rGz<3
zR}t?5UcWg2M7X-Vl+GnX4^9d0aKJQ091cAubwFqvk0v<mb{nKQ;dne^v)N*^?$8fC
zeB*GnzSP!NfNuz=;~-<32~F2xce%xz*Uxc2AF*F=F^wafn!3k%Mw}<C8Yz?m5kl8s
z-8Kkai}h}ck3RkoAAR)$^l8F|8LM^$f<r%!I6T~A)vmDiE#{ojk26-A9T+3JZNk-4
z#^HVhGZ15oH*XI3^Z)pL{NsQ8hq%}=e)0Sl7;?n@!x86U#N~dE{bmj4I`n7h{B%5?
zvDs}g^b^jf5w~{-ghuQ*ZrUS$=NlhjwQ6wpFyPep;+r`q{Pf44qX`Xym;UbL2;+Ii
z?cD*_SC<G~gP;BQ3-qTmE-o%`b#;Y?SIFm#Zr$O7%MUP4BR1O$oN>mcS!;jj0JoE_
zI4R}g0qdQd$7ZP`1vwp@6AdWm3?DqEaS$VPK)_ng(*7;y;vZ-XTcd@T`C5vQIA=`L
z0AlgIY=cA09_&Oy)P@c@CuvJ|I-zMBblqAiUDZf^nnuj?2;T$*=Y?P(0nn@jNKy-e
z6eZ)AD1l-E`B^LC<@~+8*uetjVVD3oG|dq|`0>xMySP9G<9IsbbQXC8b4Isn(Y6uy
z$1}$M@5#qbDWuAPVR5+QmkR<4dt@3iWd`1UpQR0r&6BUH2eL5_CM@Iy%Vmt%@{lww
zpL+BQ*w9MRa&3$262hhyFgQEC?Y7G$l#&Cmw&9B7Sd1Nsgl_a|wk#F~!k*N$Ba}I+
zOqiv=7$^lG%nTm_h-7Ry=gLAFcvetRh&$$bPgGgS<yexw&spTD_iL$N=iI{RRX)o*
z2hWz?6H#)V5)#$San>}TEAzq?ux<YhF)S79B4KnCBrUa%F<17C0IXWsSO+z6n&ylj
z{O~9E-XDFA-~aA+@Hc<?*Dw&qSxkx-WM)ZG=bVdisM7)j)-<Z=90jQ5Qti5YS-?={
zp9U&fpMg_-Cl>;+vZqcTiUAb?iLCREAZbZ}^J8|^t`~2y&CXeI%wnCO0y}#ZPrRIw
z$pR?{=RKSSLAgRZA`<ZR7O<0k$WY^CnHvi(TsyVs99RD~Vx<a8z_!%v**wn**nq^b
zu(@IrB48$QM$C-TEucJGF1kPlKU%Z@t*@pHh;c%<Y6M()hZsGAI^r`&kYj8%Ys^!U
zbeC_my^!ccV-nkp7!xQ-!Pp!loR<-6T44yrIIE_VF=b(^eDH{grSpuMh`;yb8sd33
zmLfDs%?wg0JV$h)K~6~;)I-?Iwr${vuv@RO?KVi9F$_KWbC1jYm8A1BFi(S~djRJN
zt4>;$ktYoOfN4&+xjTReIGltTJNIYY9v;xN9Wo-OdBAxbu-<O*`t=R&A5O?wOpe>G
z!Qs^#Jb&>5>&+fSLD;1*V%;_vrxOn6GrV^=^gZS%Yh=F_=Yxmi2{99LO1Qq31MuP4
zW1JISyuJm2aIxP>ikwKsl^J84akzPd>&pv(GGd-_dJuoN{cZyu5>f&h3UJh5&Jn-(
z^i!NqxA3`Em9~upzw<m}9w*VFd5^Yh@$~&Cc>2Lp01nqrpJ98k$7vq0fBG@n)ecSD
zOVjrRtTtEh&Y>S-k;U}q8Jr2-Y7KYoG0pcto-j>PG5(_;eFDx2|M;K&A&#dWcc&ir
z_Yc_b_lS}4;>8<yBE&hP?<WvB<f+HQ;ee~FYjkb|Wx{&hV!a9oXb{k0IF3jo<D-v0
zMo>?@rfEgn8E1eP-}&7=PG?EO3&3bf1+fIbtQt7cQ%^CX(KenwcqvTDP<?Dte?V7s
zkX}>Vp$UjtiU%BdtuUAKwQ(ygNRaa?t29PT^Fs-?ZC?<7zgc?~C-~4vN0gj#J`Sp3
z74OEBBFH&(>kT&B9db@EjdDaRxI{ib7vf}^2IM4OkUj`twF>y)L`#>Zh{*CgnUkJ9
z>Ia&D7=gd}$xra<^H+F%bHMZ0hjN`J>uK8-O>*du6Oc3dzc0HJdp%~npkU9y5ZOQd
z+E-<+9{-y|Wg859m4gKUCr#IxNvN@T&Z*}a<ghSsWCyDbumL#xT3TGFF)hpM0R+?-
z;~~bB9+oc>M$xMM1e%eT9*;69LPdr8UB*-zv-qpVQ9!3ymq4u&<RuAjDl~Cc(4wuK
z+)}et@6$j{Cetk{8>&CgHtv}bqv)l8jO<iv<_InW?8GqF?#GxZW217C%1ICvR&&Xg
zRc)bZ)P%B@Ha}+h!38MUJn(>Y0-5Fl81C*K@RJ|?2!Hg)e}uMcFb|`4T!5zjb-Xo|
zX$1vN8RC?+T6qy@ESceA52B2yRf#ef$t>&iolcoPdw!*m1dr#bS+Nu^ELWLa1M{r5
z3C7T|&In`u7E`na8Fn#>p(GU`Rz#+2%+^AuxZ86LgfyrU{g<+5Y;2kU$;=D$VY^vl
z)eH<L^E4BtF58&oTJ#{Gug=lO#U%nr5=%jWVqlrOENrz4?vZrE^rg1>{<g0<B@x@}
z*aY9e`yhgFFgT0z|1@VzqjZ5t`GBSoAUF&oEX97jzr$v=0d+0XFe0?lzj7ESthKM{
zFwT%zB*Vw7U1}nj9k$yIHeHM3;Q`x=y$+K@jFT`K%wisXJ|ct_;1ZY#p=m**IN=ZD
zjPv;c^CZ?E<0SpK4@WUH_d$DIvSb#9{*0yx*ljn`$-)J6%y6m=>NYEkeUEmt204$@
zxt9uA5MI2xLDMuC=Y%OsALrvRBcFP-T?^l?@NnwY%S$VdIpLeX^D$hM!0+Pn68+F4
z_<&W@q6uQYI}9UI%xK#Yn@xwN3&=F#{SV$l+jKY|Pts@8H<)H&2`Mz#ZMX1U3r-0a
z`wr78BE}3RM$@%eQ$QL4o*kIQHx(>EWsb9C0=&aG^cecHbS)yr(4R4mJvQ48-Pb<E
z`(ODGoBb7H%-FA<pxHeIs6m`({P6of#p!%TiU}XQ{|q5`93GCyIbqeUq<66M*zI;`
zKNL%`*RSsI=Jh@1Iph0(^Ak9i@V$TdM|gdE3r<-8`^8TF?j=}DNeuZNoV1VgL0WYZ
zl?v_5Ndk4(pdUuGU8}`v;wjm!TOegIcV|Wy)}_-(+cubdM2Z>HFd;#nU)y%@t^s>1
zBFu1*4hlrI#h^)eYzqsG1dx)__9``*fh^e$(G+KJ$_Q=1JWmKs3$JT3Mae9s9EAbr
zB=Mq?D&7zR&V4UzTL@^jZRvUyXOS^514tR;Fp92DWlz(T5WM&W&T~f7IE?co@+D-%
zn9-l5CvnaheSa#|;Yqt1&GRg+$w;hbX6=<bj1l9Ua5$ZC=toR36~Z9v`RCMo4@}w}
z#=n!)<lQuO1QkhF&|p7cpm?@PJGOdQS_+o!D5Apa0I&_3A8pk3`9)(_*fJ`HpMaFP
zJ7;Wxo)1n_zQ+2PT?o~xPoq{!dd#~jp111;ipLyW1qD`R$vP$#DJrPcKBPq2^cM@w
z(_0MM+b?NJV@#xqLrY4|m^z3_XOg~qb`KJ@RB?P>{H|&6f|ush7HQQKv1FffVjzgm
znWVqWYfo>{Nh=y-xyl(28xz%$>2(tMM>Z%+W-dX{BAMVnK{bD-53JYUH32lwHiVex
z89)2U&+zTv`#ot9MgkP-+_K5Q<-5&nnuvr(*Y^kKP=s90EdY;fFl69MHzltHK8CX7
zS*tFMr6GHnV)nU$R7-J1Vft(?6()+MrJfTZv{P^=<L1iu88fDaHVV`<)vQ)Su%BV6
zAGNf)JSA(zRV+|QD}gzeYcks#DA;8M;8s{6>(3#RM!XDw-j-!Np6?b&*?MG_ws6h|
zWMqU)U^rw3QkFHJ7Az|1+;&vgzNL?Uo0n)Vq4pZUI8PvWG@VCiT7X-C2;(%vIjP8N
z8qqxgEX~K^FwGMJd5PS807$AlH|q^JI{+Tts>AU#0!*ObkT~OfIH74;P|P5j;lU!1
z?+5f+JTT8A*6S8=p7HAD1~;#7aD91=^Z6(&T@Z;t2$G6T%xJns+8Bla@oN-g+x15J
zSwHa&Vv1k@clWngw}Vs)ri}GwZBGotIEv5bJcC@uG-qr!Tio5>=^4=D_U?e2n|rj~
z3fs*F``rehi0yul?WISzJz=-q;e0s50s5H<tG2^9&H%*nBqcyQr%CsLm~lAtcsQKl
z8^UI{R=-ryt{o2ttT!8^G}R1^6Kz%F9WM8KkQ1}{{(Q!6yTNw7hRY5uNiW`<MBu*a
z0_Jm%RU^H0&!>}EpLj3*b(3@idG*Ch9B*&%<oXi(%N<tR4R%l8$5+1dC)ixAKq=v$
z{K-GTJV`q6`sxxy5|F1D#VlDfA~8m+HyfE><BY%h%OBt;Klug5k?`XA8~pFqJAC}X
zN4USgL!M_eo%BX_Kru=ek5s7&%@(NTC}zV=fTS+PtF`MoVZ@XA&T8;GXYEOyGlCP%
zvAi#XQ^Iz$QD#nLJ~1<LOywGY8f1AdzK1aiGf5881}`N<C2FQwx&VomRP4;`3~NG*
ziD$$FxZn`81QE#Mzc)we!z~3AgelER3`jx4cD+T+a^1`sjSmP;#%7)->8-53xrgHs
zp>2fi9VSdu#Ov2Lcyn`utIJC`?=Vj@hG{~a_4<{th*2%wvh+QV+F#iE`ZA>AKIe>>
z#VkAJgq3p$fpni+xKMz>i7^~|Bqb=`rDftf0Xq@g=W=M->ycppSN-B*1DCVZ-e3K-
zNErb9hKrYjG8dK(sEmZMeat98;aBbRmEqG0T)lSZoU~5Tiat5us>t4ivzCst`yvGZ
z21FLrjeMW0V`EIzBb{CGDYG%M_h#`Eb5vR<5v00}!I-Rlmw`8x{fTl-21^QEq+zxh
zyT%07nMriw8?Ti;kQAw}ZL@@}%3?!61ulUA^%K%(LR{KrErFR_L;R(2rd%sZb4tZV
z23*tOq`z-)UV2B4nAM0lv%~G(9gfF?n#FoVZF|LUfr%vxY&m@I1$>s4VN{=s^}U4!
zC7P}^cFcZepxBChd|}m;wGxzB7C%rmmKh6O9-Gmz85sjPM{u&WBk6K!Ii(#Ks8;dT
zd6^Lb)fRGD3({vbNl}t|_c3P$>o#6Bs48jb#l2Qpgr%F+EL-;ksnXM6#2C57I1rWD
zQV^JAYHM2CzQ<U_V1+x@;9HiVv~7#h@<-T2fSmX@{<bf25RgU=?P>+Y8ACtfcy~tE
zwjfB0t$CWIV+rJ-WMnj5z;?6N^yd!U6oE{b=Ly5uOF%MZcn9g&GY&YNd&HEnT6Ho%
zNGv&;;L*(<pcP!`@NhU|QZPBqGul-P-!>Sh37h>+IxWN&iL>gJ8l*HKGGk6+ui{$|
z#|}e3VV=b=Z<-T89<Ofhg=z5)T@!GgCbUflB1x%1ZI+s*!!U^77_$V6!x+(YYpKi(
zqUrna!w-=Y;r9Lkb7C|Z;4z|~&IrNbbUeTlfie(W1CIb=hr2fqGLI9ZA8v4UeTmJc
z!^OtIlf(VP38(WJP1m45oN=+=W4qnpVzmX25ot`wDPgzS2(uGfobp*TlRe||LO{t!
zAAf+H64slQv}wr>r+Gkfq5+L_lma(L7>5yQ77Hd$tadmZI1Bja?2*QZ`Pc)~gz0=j
zyI$er#v@R~_rCjGJpc5$Xuzf!ZQDQ+8nV{BYWT)cbXDX7+{)um|MYuUZ&vuLzy2YH
zG2-)2KE?Zg@`vEvMq0YDQvySnZt|jAa~e#87^G}F959U1@xk^;snCqt*I2rINU$sW
zSX3(xX$$9(lN1&9{aI7#!m>%xphL_uf3mXj^E8)~vkyUu5!TE>W!$ADByjg|gVkn@
z-~-wa5T}Ubv$h3g5PQTaf*c{ogh&}_mf{Jq$cd72J)hCGEoMz!4#R}on;ZBb!SFms
zwGfd;0GSzao~6e(v!-#g2IWF>oDPHRL7A24kp0XuOn_xK45tXdBNA_4UYtt=?FH)y
zObHDe0R;8+<dXeJL`XVrIemEz@jDk!(62c(i_n{7^Y!!?@F{IjfDINulY^D(;j#R8
z`!;?&832^((VWD7r7|gU@L=)&qA|1YH6!4{hH)(xAcFHXJ+%~zo3iNb&jrk5uIk*f
zGF_I=#bTTapjA*Izc1HLdU#rXpHU|gU@WF9U_k5)#QB;R;~{zhMvij<Pf{r1?2{yN
zND0giloGPCFXf{2EKXV=V)Nh;Q{t9sFjAeE90brlquzT3EM~c)ep65uHs_+7BtZ0q
z#_&0Z6cbwSF{xi+&W!WWOO;aan6p(4lcby#9Gj+V3Er4l)6Edzq%@#`l){k7j#<;^
z#sE64d}d1-s~v*#3O1n8tb9e*v~LA{b&kKy+~&+>ZW^OVaN5hUE^Jl+Z_yxF0jSOa
zW9i6Dp9|ZT)t|3^j$TY*0xk7fR-Xkh%k_fCg*(zOS_28>0$v5LsSd{(!$?IN=Sk+9
zNdM%srTQ$F{U&op+jeLhNu`Xl7?Vcb{9+W}ta<+O8kAr6BIl(I-~ACU4mVhDRv4!l
zo;+6F8cAD*iKZ>vpokn`o-s|*{iN#v7d)n6!Z3}vx_*j>!-F(&&jcbbEfKp7e2OZ}
zj&O~~dL7UYlJ*<Mh<ToIK2HMB^MrAjkU`k2SD2OExV=APz22x^Z^mXL9VudzgX;3~
zN__H&L7tJvQ7k9aY&da7Ka5~xJh{F^o+IXwaeI4@)vCdEzd@X5ym@(x7_$OHN#BiQ
zuV%*+cAGW2&;ZWid>()doQDb1k<bQ7LtkEBV72aq5ejl>yGHi0)pjEw#Rbe;@oFP5
zKKke>_N%REo4OVHp;vZiMt|;+Q^NPY^=-h(JZrtj&`<E8Ma*JkKaV4JyRGOUeNe_a
z!TSa;U%ml&!pqmMvEQw+^%-5`#1uAX)gv1V4QY>+5{Llb1SBw~d4i(`-KxXk`Dgen
zM|cIfw=Z6UzzAJif*gk0mVhwMGssKV5eH@7!1&I0zJ(0nhu{AJe*A-<;9Fn&I-Y&>
z3^xbq%#ca??*fcGjd0F`ltC8IG@=RO@8`%NW-(;;-eVebS=Y0At2qX|Z@|b1kh5kS
zM(KuB0(4e=n(U`d2$}`Spx{x4i3#`jM}#Jz2?0$LFb;!Q&;S@m@lRxCeDUHXp1*j3
zd7kj$hach_U;i4?G>Rec95KfPPr|;YoP_0#3A1|BF>A-Bln~>rL3iEHYDQ@x1j%xt
zmQ-O#Yrm|_i0sW3B9#E#_D)VPnF`dr*$53`OBuuxfPxt=$oXI?W@NEy(M?{8@&XXx
z#0n)da(0+B(0-S|^&N{R^crZ%x={fUR;eN>q?!)3L6de&OF*XsD5F%t9^PXA9vyyE
z!69d1OBB&sTF4&4^1myKWSXUAimSfre^)lgEq;F~-7eQ_Oq`;3(kE|iyeOxl&Crn$
zpDIWb?Obl%+CfPr%@SN$B&Cl4Mm`B31(0U(KDLKn-l$jWs5FkZ`<$j(^mWW&%Hj*8
z3le%TXJP~r;i{$V2-+7>KZn4@!b9|cwg8NU$>q5-Ln@2(mCdF^XrOZaR&6Okfv`?Z
z8-RH%5#=yO9M3)4^$Lgk1FXwO&Pi=RNbOWCE#Q@v)$diH2eqdGP}pAcB?JRebt(4r
zQ&jDvUNe)JHnWXineS9KgUqZI&Q)XiC`e(s=iuc_9Xn`K6It|Z!9@*3toVY|z?euj
zW~nmGI=;17#D1S>G0&OnJkc&87625`2o0LZiD21)voh4qxpKdCy_0T|wt;eeu9-E}
zpE2fHSb>zZeUgB2vti*ffe%`_?GZwd)>DmiS%8Y3YxC9r_SY~Zc$bhe(1Z@03F~$(
zg6Jg7<UGyjc01hPKOo11^E66ZCFc;2+5~-O_#pmveLsNx16J)CISLChM+P#B>G7Ds
zo)AKVG)tACgP0z7p~G;R5ZVqdCA1zGPb2PMA90)S;hProEQcH67S1_r)-AjvtXCb*
z=M#t=UVQ!<>-8FK>#<&~uwAXOU2UY3#55u^<JtZS<1D^>`}H2{%?2O6_dfdb2|hS+
z7ML@<52ahg#eR?T`HbCuC+TS-tlJh{x56+;E3VyckDE7d@cy&+@o+p~y;-Aa8k~+t
zbZv*>e8zk4J;m{K#CE%pG?`XRJ{%v=wGHNJ#P!7`&c^|5+hW_b$jsPow>Y1BY`1G1
zk4J3RYveg1&l%g*1`NWk+h7=bY&X~FhXLE&PWo;#W7TwOd(q<gt6Ti}Uw$7SK6`>c
z`kk+$b-<h@U=c?FISZT`y?5(1n8l_hIRWw^cxg*H&p6(^#>Gb;pj`zqYEM#3V9Y*w
zlA;6EbInnJB01^sVL|Zke&Z{M!x2CK>8JSr{p){=fBt{?U+~_CAK=r^ej#Z~M+n*#
zLl~zS<h<yiNwnsPxfIJ-5ks6Mpl}d<VNQ}_ZyV7V3UhAzjp)#t5a1AyW5k>p<2YlU
zX3?mz?6HleGp9LXI1hOJ`ZYvrC;idqI3s5%C>r`fefJ#ZBx%uK{P<IR_Os7GoXT0m
zSqP7$te{~A(vu|%nv%UNb2(E@Bdo{k<KWu<AOPB6K$toKS(zA_<Km6-KdgPT2~{_4
zLAZgt$5}WXLltM+T*{drcl|LU!1{LUXFL7x<{gT|-U-+V&!WJj(254^%$QbL4IPMj
z_$%Th0+VD8mt7*tfdwe2KmmKAb<*fUKLd~!E)~Sq1JATKKuz;ly0?O}S}j?NISP%V
z7jF+s9!9BXwC}SA4PM;4j*BC;0LjIjf5||W@hYHD?pGR|%lqb(;mGSaG2jia>10mU
zi&B1<)uT-+Ta!kFPD@ZxElXH*71@eP3d5xnn<7F807?JaoTIRIR2S4NK}zY6!OH45
zzht45EpkHBH$H%s`Lh(dF`(A<!d{jxD4h+6(0Yf~c}!}@!I^RM`VB5GF7fKcO9hAW
z`_jF`A&^*nl<y+|W#KFs$Yn=%P|+Zk-l$|u7OCeSmyJGWDRju%(#@h;V+*|#!)3c>
zW)yH##!BhRlF|ms?+exzsQ`0jdgbSpW#+}b6B+X=KQnK<wiQ8eVcMN;-n!N@xaux5
zi5|p?A94ai(iG--C>B4Db&+N<Wcoh<Qi03%3~FJG85<Wn8LmMiE8AB#P?m@~O{uPP
zC*E%%1hw(`?OvzzjA_o&9fD>4wVG-LjQ|0`6Z-RvlqF?%Iu9bG)_VzlLTDN_@lBYf
z0cm6${28mZMU0kGodl@*1`p$ai~UxLN5%<_ci5~q=uZP&BJen4)i$`?Z}8ERCzysw
z^gFHvAnlq$@K|*##Cev^0_zppu90zJ#%i?!AYf=325i<F98M3yFt#g9!-Vy^!|{B=
zX15deV@?R}12AXo_B#yy0PBrA^gULqHO{LQF0U>n$Z9)uO@qVz1HSg*hj@6n$Npl6
zZ6}WW>-GuG{TWYoJH+!Dm)k7@B&+cB@(Rb(30GHFcsSf+x9$*Vk8v2V-CrQh3D?)x
zI3ADKZMQfNJ%(O-n;z~DxVYS-?@w5-R+w`Hh_LEb=!XHTZjIyVh_+p!?+3Ah2?6Kh
z8SCvH5674IPv8F`4r7A9JLCDy376}Di`_~CLeZL$2EG}Lrj>hg4p^^(RPTEa@&O<L
zDj<_62ooLWh*jHxnMJ#nrMLMoj6fE@!+Dkp(Uc@WU9CF&@gIK=F-Cm)v(NCa{`J4b
z|MbuQCtO`#;o<&HeCnd4Q&Ws$vD1lcplv#&6u~)(mtJOgAJoS%f?5yGjA@9tJ3nCP
zrLXWj&Dv93*4;Ra7<%bhJWUgbfMFPMK1mOBJ5y$D!Io7b!3!N)*$Wm@mkP2HOmLCd
zb~OzKmnQ=O4)Q#ZB;(`Cxo*-X`PBllpwb>#hKx$EUA~5Lp=_j&IqC0ffNVqqNDe-`
zB(4%4Be&!_tGtXzQurl9WqUlCvw`k;8pxVoC~G@MpEFp`u`m02=c?GH_`Uw4GIolR
z8H~in5<E86t<Z_>b{?^K+T@@d_I4OEL$zA207m=#C1A485sMJL)<w0PUwFNhc3?oE
zSBoA@EhVTNvbh|FAXY|A56}fWXlPp=s0z&W`keRR^tiI6{M&+EEI%Uxkjgc%MK5-r
zPUJ@tq^M?$RS>T<GgxVAWmBbJsZB6f?v07SDPbmK*W^#H77iXju)~-$GE^+fiz-gV
z+;ZTvE)bLhFV50_$+|v}9uNjfc)=u@h#j^msIl|9Z4kAM6Oq_rT)+1o@(+JhtEDwE
zN=kJb(ZwvjXPE_Dcs2X9Yqv4M;(41(d!6?NK6tShtsRkV4OX<{Dzl|?pZ(Gv#bKGk
z4!)*C39bMa17)Nj%=w_#R4shuek<TH&^5=Tn=GrZ(<F8>8vVkaXU@v>EykZ^6RSX7
zeVTM`6%fjg$Hz2#%g-vCs{Xg#w{2Dy{K>?3s9um%4nX^Kd)Z-PT{q@b*ZF_zt3UQQ
zrb!P(hW7-_v*?eM6^&B@dk>HV8h*W%K9Am^YZ|QdaBeRy#PiNObX|jS?9sLjE-x?8
zwkw=Yr_zh{V!ua!K8hi37NPsq<)s96-eVYgG>ymg^@a4?Br$sRuGPb~*Fp%jfyj(;
z8nNAOaCkUiyIGflnm~j}vjS}h=t7I8X)&G$tlAaMryiK4q5ENcz<#&G-NUU`I<`{H
z*{(4UGq&3u=2_O&W-Yx+H|rHLWkgPBS1rakpk1%<aC^Yj<ps_?qaTFjI~*Rc*>3P~
ze}~n2jn%5d@pQy~x5LBz0o&aUX-Z-^+%kq~z{TYr_jmW$?RGewPGB#o=EK7sF7_8V
z_h)pS=ne0W57?|X7{&>^?GE=32L$icpNz2Ei>1fUKYM|@(|}A45B-3jJb#64aQMcD
z@8NpC)wWy?fUwZ>Jb?*l*DW?`B|-qEl%$0f7;y-|n*(<HGYDc|;Y*<(X;H|8oJTm8
zAoqMcpy^s9lK!-z4LF~riuiZG@hyDui&yyRC%?e?|MCCAKl>m4JAC;5GYtKJ+q(yx
zP6Ot7!q^W;WIbU6rhdS~{T)(F(&u&%+ng99a*}puF$yS)+VtK6&}7y0K&gtibt%~%
zlD$NM4|L_*VL-X=2v%H?k_j@5q^X%8Tiz*G>oVy6Yu=`;eh!wgik6)KKq;5Afyk~a
zS4Rwj?xRkbKp{Zb&Op5f+2e%BF_V7VdrTg}wDeG@q>}zp$wno;CrhQ)-r}~m1JLub
z5LBcYKOM{Ep7~{7?*!~b*`;Y3JJeXyP)i`D7bj^_s^W4Jf1|c+(pe${W;dd@=g7bf
z;F?BJBPm0u))dRCP-Ehp83<P<p&ZC0o?G@R3<52g4|Tg|L+UcpHH|4unUy#i_)M14
zBbf%m;!|<_XFxn0^m~Nqp(47p0u8%=GYQV9(T!8KNN@xrk%3?bN51F_EOK!wsP#mM
zi8UH>h|qqF>O4NC2x<pxF0PP<jLmw5X$0JUho)%|1nVjYbch*<8kH4~KBLAJSQmQ?
zt9W)&l3(;{^cXN#Fr`H-1xOP}3kg<Dz&OpAvOE*`7T^5Nckpli^}j9JLZORv5iWp3
z&O+^T!5kHEnb+VlJ#QLG`+a+EGW$p?o*>}I5gOO3*``%t`@$$e*8sG|TL=M9pFYL$
zc$C8lGS`ZUHVE^Slrfbm*F4Xd7wOIALN6%+o$F<TVw6fN0>QDwxCTuhzm}km4Wq(}
zGjycjC@lV)AAdw)VRc<7J85Bb5fD>(d;PvL_um3<3dl=1$aW>wT=`LT`fu_2`bQsP
z)ds9rEn*b5I)nz$in_MzKm<&2f)4?k)f$cnUE3n2D2BkR7IB)eUTqO)@waQ5219=q
zlVbq$IOAgd1n2V!*Vj*wIb+*6Y<D|hm{u#*wMrp`=Y-vEgPgKx{MrtuQ;*H6!!(cR
zng*T;=b=Zt-(eVqJtLOG)xiNNiX8^Iq7QRgDGr9Tz3Mt#>@U!C(l)K}joM>K0NXST
z4#ysC(_-j*Ow(2iDn?vh>~TCCkaL!<8PkC6c7xlSTQp54g$UDx-F}aUhdVsIeu8Nj
z;Xru$^gZ;wv<N%j++wwobngBAJuY^8+~3~g^7;xP1dQW^%gaj~4`M)k|8M}oVYl1k
z;qZWqiwm4iCxos+(}?B8X0yTZaKgoQk74N1HXXXI#p!g$@f7j>AN>@G35<*>GoIhw
z<JoS9(=a2)d+gRDF83E$HF8#n^@OlrAvChDW$7+KvqwxBpoBwsz|ZC%Ax+jTB7hUn
zXeX0~0^Z!cL7e69U|?=198V_+Zk*_M=V`_~NnyksGv3_X;m`i;-{Bwq@gL*alP9?7
zws`aTEBy4QKgH|UuMuS-iGu_$0L!8~k^CDK`3bZ;j4h;`t-z3L@J3jqIjwD=(`<zD
zY($U9l(Lp}4<f~H%&blWbvLQ31r)@w;*46nXh6x%Qu}2KtK}T=N(yMOYODiu)yC?%
zYotzT<+%t{ZJY?zZJ|npPy+wQZ6A$+s~T?`mkI?d=qr-7^1ZS{vx3pgwsXF<*Zzjr
zy8%1W1@5R+dYE1fL`74E1!C4AE5J!iYhAe!J2(oUvGat)#jDtskfoaB4N{Ll<!G_S
z?gk*4mu!H3o~Pns4LhifX)6Gu9uzXiHC<|bEfv8tqxj1i_|#escm|lvKTI@mBCfCL
z*>Z&GiNvsWOu~kcP+tOyWI&Y8XHEAB6Ik{13`W8<Yr%p8rj!wrxVfh}VVotlnR3E7
zkC;_-ZnU%y0hi|ilrv6Ui`F^pb3~Vo2G)aLa+|t9Dsaur;3OVKHZME@$O#F`u!6P7
zz*4Nbpp$u>$+(oXngMeqT7x5e<KwTQj|olJ;^VJ>6BqA)0N;oZoVECXR8wY*AOJY_
zg;|xR?&94QRKQGxlA81(s8MbzzhfImNf|1LOblfqg#k-*5`i#9<k`}QSu0yz1z|bk
zcYgPG@#&|Z;*+2MT)d-7Wi@PZ6!1do5oIZa0;C_UQDvBQEFBb_md#u@6?>lc-@szM
z)&R$JeD#`5<7(`$&OLjM5?(JhY59%?7pPIF0Xn@+V_Q*yUSS`3vG(;lj5RRU*+RSD
z_VwYjYi!yEpe%lXQoe@Z9Ogl)3dwsUULkZXIAwH}KGW7;<1k>oS}P+Gv0kmw_rlf*
zQ<BiNErw~tX0t)&jMMq7AZ<n)S_BuQo5iM)_pep}c?`oD>-9<sDCUS5CmauVxW2m5
zz!Kodh*QLNwH3XQGORHsgr+U*it`@RG^6WQWv?3JjMLC#O+eEG3`38L{SHkK8wwv9
zM6Jkg+7?4Uz=syYSwLpW38%vm>&*slZXU4T?g0iK?(Wev9p2pB;De7ofNKLz;~CqF
z9bUb9g=ZhUhcslIPA7D$4sUMW;OgocGtU_29-E5|Zf<XId2xl~a0KTh>Dc~^_nuti
z=H?cDx5LH7B~Hf^_LrA9o{n(NW4+qoa5&@X)f1eKM|jr&#5f*L$Q<y0`@jD`@%-kY
zrobT96USk|;oKvvS{zf4ahS2`Jl5+LZPy@95$AI+)yhx@gA8!Wh;sr{#CY==#_=b`
z<E-z;(t#nS2(zS0F@a|}ykiyyS`2ed7p5{0Css8N$20C84tV|Q7XRXZ{-3dF96o;U
z66@w0_&2w&(f6VabBncFI_l^-S#8`!Hd5GdwzH2kyJP>%DDqtrLLwuhx~Eo#i7aT>
zv$tep6dx2*fBh}i4b^@RYG3jwc-0~_1Eg8AP^9N43}_ZhtIGP<px3UCO3{P!DEn@)
zf}%PXY%eP$QIRlJpjG55I_i>X5T66<o(0G`0UHO@`*`$4#!kY{9LZR{mF)YCuXpSf
zYmd*iSbL;gs7x|8&wd9Ax=5lSYouILZVI{-O?ziZ+w$D{urd|!Xdj$)5X!+#buv^9
z6&F^Z`f)7|Mk0;A?E6t%_7zX*LzUk%z`zS6UamnWnH_vt5jZ=COhUybWghA=<^+qg
zvKgdh(x?+LC1ow3<<`1cryLPkn75b{Vq(n7?nJH{FJt$d0v_)Jf^V>HJG4zeqv^gt
z4vrOOIzk}A`R*Q*8aD=S9(IOm>LFZ<Hk7S#uDma2(e7Dwa8_VJUOawC=ZIHPvVe2B
zR!c()>Jro4i~SWc#NOn5=&`-J#@D{}Eex;Uz^MDYxTdUWTq$p#YwIax3B|NhL0PJ)
zOK1>K7OkNWU4<o@gt35{#@2~b^%OE+{Nmr7^Z4TVbKKqE<6GbQ7XI)L{s2Gv(T^|;
z!vfgU$gZ+kT41x-O3Rwim!;AZbLn3Uy^pNx(;`)4_m+CU#{f?P76;MW!RBf?1MK%*
znbmJsxR-ddMBSBjE23d~ENp&!O=T4H<GBWrf5)%R$%edIuPX!Sgq3c#YYfAP)(5nq
z!+GqnU9V-+jne+-csME06Ogsr?GV$1rfsoWtuTxux^|6m68nU1)gh&X^=c&*ro)8G
z{TkCGW~>at@pQ&cnJETiwIZC3M_gWBYE`E+ML!%5*zNW*&ypN)hr@$-b2&yo_UN`9
zrhbA46d2B!Q^XV}jMIpiBbYL#G$Ax%(wjIVv>nE2LgyUXwgUs$Y&JNbd#u(Q5J}I{
z-Fk%-6B^w=`tu2!%^LT&57=$@@J+z&?JW{VynOW%zw?c6VtcX0;cyhd+K=eE79j-O
z-QHoh+u-*07Ei99VBM{8dN|<n@(Ryiyujrp(D)YT!x>k*J?`%Bv0iV`x&|*leSz(6
zgX8%C=T68Vyc%8soN<45hu|HiX~s`Kdx;mf2c);y=M3O9&G`AN8%&oMczP*4f~WJ0
zV=tAa&)$E6@BYzu@q?dyf`9XG{vEdK00Tk%z72L=1DazYD4-SYwvR}PQ}x5PZ`j&?
zbf&10csbf_AIb0x5FDnIaDV%NKmU*a34i*%-^ayfg^%8UhCluX-^0K9cYmS4%K}mL
z=VVp}j*G9Q?JtHmtT(x)Fc#;&yqA@&v+S3|f3*r;f@?`JI#gy)H~O*%7NDi;-_AfH
z3F;XvIy1coOO+Cg&|831-ID}RQ6YZh3{+4>73A5liwvdWhny20iN{FRkv$2rM`|0A
zfdm$-s-!;<9!u`WBx3-I^MjoA_ILf0?L4vTTv|^pXXoGkdM99)SSlx7rjjpy#J1TX
zX22-}mYOpZ>=q1)UT^tbGXS<o++3=SEphbt(5d**4$xep3|7Qj)2ELQ3(JMe1g;DK
zNR0$*@FM9?o&Q3j)q@SF%pj;PVrExtCcsk7C!kIGATn#)Bof2IIVH?d0h#uwj7g$<
z5wNFH^kcW~yZ|Ndqz|n39$nL5+jeL}K;wmBbt=LXrpguI3}yv#5Ks=3-l>vC)3z_H
zX2pwz(RlrQD>~4X;)54Kc+&#LP-ztAP$(V+YBeI`m{9a+EEX2liY2S=DA~eoI&9Y$
zaGY^}bBBJMaQ&4J@E^YaSE2*V!n9c=nAvn~T=bMWJ5Y9(MO0^-oVB|ukJ><}UH<~y
zIh`9z_RGe~cFbj|{4`G>BAiYq{N-Q%B|iAz1AO<p-^I^={&U>i+?07;Jl~jL56c2O
zXH{zkhEOlIFliS3StRNlpp|i{-r1wq608?sWpmH&X;Halzhk=7U;aFl9j$D$%Yc9$
zkK3a(NMYv~X0ZYS5|LD|1$C)^2mQ%!dNF_yf?8L2#F(Vb)~dxY4#+H4H&cvg+78F#
z30>1-wdycUBi5@Gx~9W4PS|WVYSj?YbsZiKN36OP7>scoadmMa9$d;ApHFA3w>wO6
zMn4YNwR`E=8ai~L!5n6+)+;#Y(6lX_C-lc2T_+xRDJL+*M7U`JGJx|qU>CMx&!dtE
z?;OHvQ}){?cpOeW?hp5vW~qD(x_>lnK;(#VnlQ~Xwwnz`>tNu0v6u-V;B-2nZ33oo
zfM}EnuixC@^6DDo#W{bwTH$;=VZYx2YsUTSd+fGbjQs!#Yq&@_4`=kJ698c4TkM(*
z?q1!(Q^0T@@o;lQj1%r&y#W$qo+Fqu;v6v!(rZ=BV3X<!v$FNV6p5@PVdjj_?jGdu
zcgvGg18M;GL&Om8ahN7tU+mD3Lqm*}_t*>xKl$tx{@?%V-{bk~8$7+3@H^K}u!0ub
zaR##55OGE#hfE@6Ve^9om+DA^9A%|2m*p*|oBYHpkXkh@8t3uZFFpfJXZ(|Y^e5Qv
zHn_Un%lb~afG*MxVf95N2n$r%EMvP!s!?s-GRpoU>r>7hAvYF#8X2XL!K|4Rg6itX
zjzJi5;f3{?rD#p-tf&EUG}nCt%l!t6^O_&;bEd<GBVpHbDtn%Nj*%`>m__B@mV}N;
z!z>$Et^<#8OZ6l*uQ~(H3i1Tu%ln;XAj)1X*Ie#L@9$Cdu8z;I*jj(>>)nbc1X&3Z
z)<e3m1x8fp2-y@h;i0f;4o1E#P?-!cKl&%IF-%sqi3;qBrVXi9=MohyK}ntQ;yd(z
z81+CRXA{YQ-{WLX(R*PTwY+t(>m(s3V)T?VVwTN6LrS`*l=R<)O`DVaX+R=pom4u$
z0RTCm^Fe@G2<V!&FlLRWbVC#132+)Ac@mRcBD+s$H18A$0OX~5&hF15IR*;56P(99
z3zcu4ZFA1(+D=R<4Gb#aWYq!2WUW*pFpScynV^0=xtI<!xClsXj7^_u52N0Lfp4dU
zAwc#gY`0f%&SO0H_~gew!XNzUAEIj#e*9m4fR~>=7u}yRmdtPl;K>&BVl#vG8D?G3
ztQ!q#B@!FU1!JcyovxcoL0olXwrGoN<|w$Ufd~q_{Ml!p;r{*}-}uHiu-onM{P}ZC
z)1)0T04^zd@uoA)XHp<4pTz}?76w`6qe(XsB9tN%$8vZrgQdqzTU`gFPTl6u^vmzD
z->vzA3X~te*EaA9AT4!|l0GRch6ZYNUepiEfY^uV2qFLSb-nA08Df9YHW-FU##sHb
zE-v>-5s34QZY_2mU|Flj(?LL8*J3!I(fAfaf5tRVc=GfK&gV0l;1RsT@!=lZ?G_Bi
z>2${B<prj3Eb@T+yF2Xnd*qZboX>c2{S^J!t6jtz4~GMG7duQT0^X^w-za^U9b-x}
z+O9)Pl6q}cYi+wEU@vMLs$q&aoCd763DeAI!Ul(j69@_W?G|I`q#630;n`t44|qLA
zsk&W9cps3*8U6hcE(1O}>2#4C#>0S^GG4!WjW0fV4hP_VzXy<!=7^V{d@h0P6s2_+
zkyLhqv=Tm-0}o*BA4}StJ(shGU<Tr;fCDfU-%M~?>>~hFy9+|u=S;Q$EDmfvcQZ2j
z#Q5Uw0k@|!8t1TS8+`j~U&s4j`(6C2fBzTw?9Bt19Ojsj6N9};ION*Xg0ZS+_kxkv
z{RSF%CIcO|PL1{B3dTwaAd9~G`O8=M@lQWN(>Az%@*bW%d4d-&U(3D2_E*1fA9%c$
zB-5pbU<s~NI%99rGd^oh&^rgkfVw|v2CuAb14w%Bi@;O9Q;H9Z(K#)uVeOZ6tca_=
zA0+#=E98$dgo_W)tWqK4_m+hTc9xS>=t2v}2;`)Tg>}Npyg&rSMS-F1p}@&Tv*H8W
z<7<YezOwDjs93_TdYP)O&&MXF`=!tNO|N&Xij~dRlT@l1;+1Qml8$wHEnZGCouG1I
z;tysOlr>LY(od+!)efO$zf12b3kPEDoBGQ#a5bVS{x&6@XK)-`sWw_v{1h-(k$1|e
zc(_?>m3bDgrCD2pWJrsVD0>c4PG}kt9yg6gXdA42i&fi7F+vb`?chXs-8Noyi-`rG
zdD+CwhQmAYjmvRH&eBFL<{5#ci$*ELP?Q*Rk~E6`H*0BloF<GD>$qzfevBn56u<-3
z`_%=XG_4xEucpg3os$B<oGbgumUh;cjUUuc$<?A3r_XTNnZuC^e;t+Q<wy!PK7aKR
z&p!B?un8&QFaP2{VGKRK_KmONpZ{O~FF3rq#sBvI`Tye07hfP}*=~|;P>T(`M2kdl
zwwQDiFmPlB8U|TPi98H=S_?J%=sMRN#F<}Mc^Mn~?Wzso0xA!O1HS+L@8j!V|2n??
z?Qi23zxV}?$5UxlQ&J~7jX5)j_1=odkzT(=idGfOzTQYlm1YDj#;F>($3U+>zA6~5
z2kzsamqV%qf40_8!Fq|fb?w?`yT#gEM5_d0dKua|SC=ibXlj;U@pr-5&GQT&I-E{Y
zSxE%MIby#T@3R~;x~{`C3NzHS9nM-InZY>sXKA;@8RI;N9Z26J=Zx)UjiDc~+ig{!
zcE*0cS5T9%+3s*YosmHprV(I9vtFT}2IM3<ySu{?$8(PtH*bo3!jr>#y#fO`oz6HP
zPk8To1*D8PM!b6d3OBcJaD8=!)596_C@I*R+k4#I-ebG{6y6bren6ZiJbiKroD<GB
zXE-9n_&HKcIGs+2DWPjxc}6CA0i+ULX0w2sA_LL+niB3me`Wtt4z&y`Qj*UxeJ*ot
zFmG7|XfW=bLvUt^1vnBjUprKn^HdmmQg{QkB$Bn465~8fzZ|CUg00a2Edn@BGduxv
zX5`fd_v4JP-Xby3xCT&fGI**QZ5W{Bg*>E!xkrb)1_@N|nTqxpHkyS}5)dX^E&=fP
z!H<55Cr{tQ`aGd+R$ylBG{F@F6&VeKv~?iJK$f)^<LbEyd6s4*PYXaQx<Lm!2}qxy
zY04ZeI?=FX7>qS8scHjti>F}eXJOuI<v*7yLlm}I{5I>5NM?t%zim@8KFf5dOr?da
zq>MioOE$n5QhQ#ctgf23^OC7tv@~S~Wh)P@xTZekO1s$}Zs4WP1se}d*E8dnn6JO(
z^=`n9EuCeDpj@7fR{>lFm72pm>Bgr>8+w=`OD<d~Ok?Lz5V){7-g`YPt(}<Hc0<tj
zK>+}Y`69yuBrQ@JAIkMWQ<ihim}aRUjp|rEO_SI`q>Pz?N%bYtMntdK*#Z%xP8SFr
zq<3qhtXa_FlJ#nh;3SH*KCIqZ#ZU%E719o83?ySKDFmBaq!C$t=s>mKkV;qsb6PFC
zIF!W_A!+(8mqlt-PzmwT1_)FJP2M4Z#}mPaS|v&lFfJ8VPW5!Tu7QOZv=+T+WJLhy
zhy?2~Dhm@_*(;f9R0|(iHkAa4em{Nv8Xp~wc(UJvbH>AqSGXH*@ROf>g8#7jYy9v3
z`+tV-eD9C(ufBLLbBTaJz6KId%VM%1qTahB(Kx=nhy<)IOV$-Uk)+vE&Sf1et9zlp
zXP<A;Fe}KF$}^QVD9HZV&whqy&z|91-})9_y?Ti+o<9eYRch<9C#$NKi>rpt)k_lP
zS&HVw7N>0#j>s(f76-=T4N#tiMXwg=R?DEJR))&^7dLJHZO@bp1X4Xb7l)p$FJ~KK
znMXk7H4IdvL?XWmjQ+R0=A7W>8S9m>RsArbYdcI+L`Fu_1k5>N&Kctz@x|*`IGoSe
ztUG-0-cxK=YeWXdX%vItp~pN$T<$NVmDdn~>~Od}mNsI?`vWoqcl`ry?(T3ppE1u9
zHk&n4O1Ri=%W%$d#;2cu0TA%vhaaN#0giz;!z=W|0D#d1kJrr$<RtxRnWbgY`Q^Q=
zBg-trh%E)sNqVbxMCho&c;92Z>&wHlz{6w;k;fvd$P2J;&vSW(*)tM=E_=0iwiZQd
z;fXX*1ZoRJ0W`%B9RN&o%gSCUILTZ=fNes|qMI(3RvN@QCl$^)XUsX_G{;}bKraIs
z1ELa85vH6GorCiY{^)z(!(aZ_zm{Svs~|S;<^;$<_x=hMNC_BOM_H{!peftheTEnN
zr;aZfl_KY+;G4Q9{l~xf&miY8PN|ynqgKP3MjUz;i=#(IUZh{0bBjP%KSL{Yq2Ebm
z{fcCVxt`;4!w|;`9XK|S&B~71KBMgE@*aiOl4ocPE_800p3e3>u;|&mC#|Stl{0Zk
z<67V@K}7Y{153a1MMzxOHvPo{WHV=BCkrCgRChM9h<<{eA+Ym6!Jhy}4KVD!-<oR-
z+Z$2p3(9jc<o_jK?*!~R)m>OB50E~8OFuI2JscE_fd$xT%aOwPc!yQD!gjO9G>Y>%
zGY}eykmos~X<B5p56D?6&)^)$iNHQ3VVUL_Ra+vrJtt{DGE5QUJYyK7|JyXr80NY7
zUd^*GUy=%CeO`><14034k~;C^(Rf1Bcr@NCW2xfpVyhu~G~^6ND!%W}vJogavFGm>
zhmiqVYuK!3H>{Do9Jm_IDmazYHGv%=W!Vg!^MI=&e}XW6Wu3`BgRF~9mx*XjtjC~L
z2Lt7InQIzPK3jT>TD7S44z8`NmK(LD#Yv`GlKzVpNyZozgmW<rou&zvN?&%#us`RR
zFr<Y4`qLlaD^K6U9D9&w_?5%yG~n=hz`s2GOMLfhUqKyA*u?`y&jOVHEno|E(?DHt
z#wZd=8zqTW0cL9}W&uZKmzSUG?YU_bXNRG&mBt)eI+~gB;>8Qx-{0ZuU;i3@=bPWa
z&p-VI#&MR7lUdj+4Q@OU(jq;Dx;Mh9aWd>X?f;=sw(S+RApvEBLHSWlamlD@MEfXB
z!R6p6-7a7SbSlg)n^}FHk9B%jY@GJxC;>@*WSS`VqNiNWk>B=3a9Fo%P-Z+FPdE<)
zf(vkt&~+^i@q}?4RVO12=r_K_#<e)!9&t0@p=nz@Jlx~S)fJBS_t<TA7#{}dKN_VP
z(#F`pbxg4|2{!<II*tG`=9^;)*qIr7*Mb@7U)^J}>t>W^KMxZU#xnknq>uqa?Q-Mw
zUdtNRq0O0LeWlBzh0YaQNU|jrG5Zqub2cj|Lbaxni<k46!8yU&^aXEBvI0o3YP{s&
ztZNA<#;`1vqwXzf1yECHU<snL`IT|i?}A7>Gl;~5%G}ejd@H!A$F|LTk&=A&i%)U=
z^tzk{zVT`Z&&5)!Xt_uvF7{>243bK{QY9-W(!mAYcU7lo-kM3}JI*QN7;`B;lg>6?
zd&t%?F-9n8VH%jV1s4h{^cXy%xE%nG-y^nJni+zOfmcTX5UV|x-n33AN2CD1eun|P
zQpm6f@YwdYGVf64Xi#Hg%igi>=M@;rdh*_B@FLkS=B&jrQ~+Wv(9mlV#-Rif_Ib#b
zMwhG?9BW`fFfXinZs-orT+R^~pILyJ&8=#9{7dtV>xu)QV{A3DU+(oz!0zLZKNfE}
zomd6}02IoTnURxF#G<_tqf`b;Q(Ok{2jBe|-~8q`@o)d^&+z8uEBOq#f|zqQ2FHZ+
z_8FQUWyU;<rpqEwE4Cm-)KE4>0H-W0?+B~VV82=k!{(?&S<Y#+0>E)i5pw22H;WsF
z(ar=WT_g+uNe^cSDw9o~yhLakO#*rV>eRIfPx^f%V`ovDP5#2#F03J3`I&(_L-}HH
zBGE!+SG{xE*Hb39*QU@F+{nc^@5&}^|5i*S^&xt%m9LB{04`DGW0t4fhfO9sh**T!
zc7+CXjB1yNxlVkqoj6|Jyuzjf)+=#WpOZ9$_ESVUKH%p+JcIJ0_<<Cd5LuwCV1;2C
zF4lVxok(SY?0vHN$f4tvIY^6^Lk6fMSg=D$f&c~(4an%`?wu=|8C&4?eGFijCj9s(
zKf!zNJ;gV^_I12?@e&UYhtdnTTpP5ai)72S1TtyF337#%c%&P(^<rgS3>4NwpG{$K
zNwyZz0HyLXyN0*_zSJ!}PJwWpBMZQ85vd(YWuquq!al5S#<rRs9sa-N>;BCHhKDnz
zaX^0_HS#xoFc;t>r-Oj@uIq3-jTnbXI1cZ?IpOB?5;-x3`=OYPntsPPavv5<lYXq%
zdvOY4nG9CR_mSSL$t$2NhgF%&bCvm^l#GnKbC1TzY?CroecJMw*mE+}Q@|ZjF*>in
zKtP;xPJ%WADyYnkB&eK%mU@6%8n4b50T!iIod#XjIR|7Kx<!x)Yl9V}b(hW$+fSBz
zoOMST;)M8C9I<@;s$fA!2lF^$$O*sr^eCMeQU+G703Z8u-Q&nf>pNqZM24W71rme+
zAsdBqHaLSgx<3V#a+p;}P+*(5N*N45=*ZW=h~)dd?nmjZ0E!fd6D0UAS(}RZ;Rp!6
z0U&y6`RA<TtPE4R=UVVsx;LO;eUaKM_C}C`(9C5n!byv^IE#yct;PCWWKOZIk>&Yi
z=UEYGP^rL|HDLRO?R_Rj(Yb8@u4k(S#~{m;=y?Luh`Zv_VL^LMFB@33{TQWb0o2_7
z5x65Mkg?w_?BY8Vh`kfA`}EV#&^C&6v)mZzMLAcSj6M2a0<tn#<a~+hUc9+Qat%It
z{t|Z&hoV<AKb@R~S}wp5T)&Hi`D>h1V75M>Z5`U+5u8Kg1A+&F<1)#D^YF?zcw6iy
zyd*-SkzL}f8Uqy2vuv>>>y;F>o8Vp_ux?ISr<>FF$>XcJa+%l;$sT>#OtYp}QNf!X
zTF&X1MB85BnTq*q$}B#8kW_bqrYCfQ)u<C%u_YTKd9SHc84KAQ6|_45oOSULJlQ5}
zng>a%<eW<{RV)teT)-({pwglgVX|s8<6wt{-5pDjv!xD*KweVZtIp%&kKV_6)u8nr
zr}G)(JflT}wIhTiV)BCTb%<qMQlad#0YBB$Ikmd6|HzQdiaF_XWH4*D4$75fQ<lnP
z0D#OLGaF{u!CFC*({U^F0yQ0Jz?k*9zI^!#$HNitzyBdtt2JJ|dZpl8K!h<FL^5}6
zTenng*=B5>W!v`U**JC!=lHsQ$)1-T+>q}Gnnn^2M3e)<){=nNnp$`aLK%;WnwFr?
z)<OZ-^6yE|Ckjwd;47b9Kj&AR!|(ih^Xd&6)pj}c<4l0|dA3n=5FKLzyMQ6~AQkq*
zIZ(`M3uKkqc3{^%!}g67g?aLf>J>@4!dTXSQe86i05MQyv5Hd^SX%p)2MDPUhcOBA
zp{07zIW<(RV@=A`6PII1<I%;KFP~chcMQOx@&3^Plv9#1DF9LSr5Z=bo~UX`In<O*
zD*&%XVg;aymudyYwpJ{CDqqEA{S#U6$tDWHDbAP+1bl}V%i8$PcfW&fwZ;#B_6d%C
zFVBYoJ*_Gr)@yW*7J(n6I#>5s=R>i|kl->EHp=cv3I^)68iC+k?YbhvTvvTrn^F6Z
zJ2A2+B2Z4!j~eu7Ed_`MTEfI!Kog*6hJo(P+Q5Ne9&&cp*gkGOxvjv^=6(4&P|m+4
zdnK4RW7acTY*n&CB+~12fOkl!X6?3wKq#GhjP3UdunBb?8X+Q5Znx;;niiGz^6ccs
z1VoCwOxLMKvDS0f2prQ?b72lOrCa6(N)T_Lv{<<1T)=UbjQQUmuww>tmK!p^Xse{j
zSx_~DF~wXCiUdZ?!muIfhTwqS{{-D<pX0^L*N91&uHYOR7X*Y6&;(aBYOV7Kp3rKQ
zXDMmc2_kG8>O|&TKokQ&N@9%5!YGnPJHZE7-%k^kCf5FQnbiZ$>w}VqP#%s&v}A{*
zMhJz`gC0n=V9!IYAw>mVi_=JgCRw~rI~1g(n?tU&t=XK|Vyp*d1-_nHfS0densk!c
z!S9`~sWMcMRu-<_521daJ!5-WmkZMr$OamvO4m6=wK1{VEQJyT1TP!2q?48Ech0uO
zq5-jemxIIp=^pR%i1pP5FP^_bjtSdtjgUPyK48}ccvv;7Ob9cbEL->@va^1{fC2Oh
z)Dw$#kr}nZN*Ap|)m>UD-@qNP1Vk0sEaz(hctoXAs%%br2rhxv>D1$sPd>r>AAEqX
zeD$k%{>2v<#!<&d0kuxF%vlaopl+PD5D2x<qHMAnb%P4C_lQk{P1ADvOOU0X;WVm!
zoKjE__UIs50=iu0!}7jsdbXzFYV`eh?)`nd>`++@7nm8r1z20NGN*{NLQvkFNjC7T
z@&xEltc*M_0ClkmLY6*~<fJ_l$&<*shhzvCAsYv<+E<|ZJOWz4C6&&qI$n&W#WJ}7
zyrhT6^1dN=Rs$gUL3xkOORGqfTeE{(#zm~6lrf_!W3b;h!MXEJWhQFB&9)!PUgyXy
zQeAl=Q&7l~wd)RRpr!{nk+F&>`<{VP(Msn6G{<?C3=pbHiWU6+M$IOR9TSMq4?P%?
z>TDZdK&pd|iKT$`(Mk(S$#g`LpwT&xoFcS5NSgq7Fqvb4NQ9V7a#X`~lM^*M1{qe6
z$i<A<K(x%p0EXx+>wakWWx%l7j45WT=Yj1>x{olkW-F{h)zX<s_D;PY`x?6@fTaDV
zpetd#xZIYeYjw37ur%CZV6Rw+)%`<f459@u3~g^JmD2{c?Yyje8bEN8<#8UY85xi^
z9Do5|2liaA4RZb^lP?s&q0%btW>U5e5SVHj*>FTzg9Y5%v#a}XDlIqu_SZWByBEhn
zY&}@|N=9W3jKUR)MG`??cYf1KtwTWDiFR$(wfHB0@()1l@Yj>*NW20&bE<ZLXe1c$
z+8fhB%JdnA!qs1mH6k#_X`KWDVk}~-hpDh{%Z?MeD6*x=^#7F_PTEi!kn|(6D}aV%
zZ-TKR)@n?HT&?kv^q-}cOA};$U{1xt5-sGCB7^=u+X5v6a$2!!rz^>ZUTJPg8<7{V
z<N#y`<Y@G|lh!r>tm?=XZ8=Px={is*m|d&<dqN1I><r|n=oIB&q|m~v=aIDEG8QRU
zI}%p#ym)pNl?JseU`RFW=5hzOU*S4l;53c6yx1X+86g3iroqMst(;tJJ{g7ew8gkA
zW+<t{T=j|03y^nui88Pv>ZXPGHf2U~rdHGnGp+u%F)@OMoFrCqgdSX#(Zb@-G6rvL
zq^!*G^B2!?eSM9OzVb2d@9%NAf5045-JHFHhaP~=Hb~VXxyMnT%+Xweq>?Ns-}!hW
zDa9H(Bn8mCbq^)ZE!D+(0Nd{_K6g3q1R%0Ly86r%XXCBQ`FnX;<-7G&wj!IDqOKiV
z>$CvqH5d@>OifY38~|Xz2nsqR;PK>&x2H|p46*0Qwwelvrjo{A?yX9Pz)p`cuM{lV
z_>->Hx54oe#ML0E20s=IIemuNu2TaVXZxOfh|X;)zo$2~Sl_IHi(IdBwTNIjMw0Fl
z*1|gXShc316=SSxdW<#CXSzWnA~JmNavgA(a>ns=2H-S!(+w1LA9yr2Hdf`ZXZ+{~
zKfq?c$N79l8v-_(?T|Guy8<tQrDCd`C8%A3CIAZXM62zXC0$Bpnrm#nJ`rQk)gRCp
zFh_($gelp)qoUEuoC>)21hMvj%BVC@uxvp})>#PUJS}S^G04?flyZVs5Kd5u4h&5i
zQL@;FONhqB>$aYCH78{Azrt?a56W3%fUdraO_ym4C#%MpS<=uU=rurtIQd3P36lbG
zZ@^Wr$u=OFcD3I*I3pDRSI>P0fGqze@(o!yMx+p-6o}RUj8QJGd@ppLAhARHlCO6H
zcIR0fw`~(%HI16%5?bdFyr9ociw+#g1aee2Zh&yH-{a#a*ErlCv2r!dW04h-L{(Dk
zX+AfS4iQ1E9*&BByiRT`eC#x&fK~AnGseY1QbgWbS;RS+46D%kayaW|M2o?8sx{QQ
z_H6o_Bp}dRginAqEjC1vi^q{pt}Oi@CEX(Nw&+VlM+1$DvH~PCFO&^IM9E$cKI`tl
z1&rx}v%r8PHC=p`s0vXX)#fm>whb!j;YC^wsC{g`ull|cF%jwS=wi!2%n58&T8@xY
zhJ)9!;tHtj`8cU0c1{;j2F6p5IgK*$r-XDEvGyIdzCq`lxaku>Sw`HLOk_OLn3ZQ=
z7*k_n<?WoZE=Z~RWd>pfJXi`}<daIe3>qbi1BY%b1g-2dmSu6c>!dT*YRTZq_*Ayd
zHfQ$Yx0F)C>o;$3I2`cRuYMIzuCDRfXP=|*hq9qD1kgb@t7!KH6wp*QkpfioW)^_c
zpo;yhm}Y9=RSiKcuv(;abZweWtg4=V^~;_~jf5YA-g2lFboUq-|6RR=@j?kwOxMMW
z=WfT7fwy}2b6N9NUiJtWXynEA>(JC7L-r|{D?l!KWUr}L7+7a$niQq@s9dXktX`*8
zoi7<gsF>V}3`!tuz`%=Zfb(7oI&v1xnn@ybP3DCR2S8yC<$3r53>U*zfEEED(=EX~
zAf?F40&Z(R$XS_AXKP9VR_B5os=3}vaK2cuG@%h2CxhTVfSJ*T2F!#hPdHCAMA>gS
zg5NYl>oW`R*tDGhU6#4hl0zrf0gHf}K)JG|;yoEqHx5@Ue`G&h$b(n_k^>=vh27z*
zWfLY>nM$(sCBqm{c^0OT6c8Dl1p3L$=aOoLf%5E359zhbkLb%5;EtSDo9cn0Y+6<q
z15YY3u|1G%+6i%k?6hT?$)t2numC@A_Dgx0c2>EIo_Qv5LD`F_WK^8lJLz4SU6@H0
z3Nev0;M9^T*Nzdg&p|bj+Ltz079S+qRXexME@mOzF7LCs?yP9WLG_})e$QHFJh+_4
zgkR?$`s-frRy?uap$kp%+L08AE=maaAqcow(xPA#@h|{v)*U!U91jlwEK1`7F36b}
zU2e!qwVkgKi`}quwo$#6%96jpyn5_Vm8j67*m`g{y>2+F79sjAveBT>`qD{4QwbU^
zi+Q3$HqC!II5oo56hUwfajrTAOJUiTTy#yUVUUPc8JCz8Ag};ut%@z*W5{1`-wIA_
ziiKcc%)w;^bXmVk=e6jEWKoKCE&v7?xPoIF=dyX|qlO8%p_-R$tc`9$8izSA3(5{3
zRvQ;0@#Qlhl#^`K!8yb+B1FPGPuR1^)#d`h1@JsWB|h@|jzJ2VEH!3}gS3h)C}U|=
z(zZcby3OX4u^h62Nn~q<5(aQ=K9(|jPyx}8v>&?B+4Ej{7#ZxTfXdd3{+9vjAaiD(
zXMFO>CwO}O1Rs9*0p7fMgY&u9gHF~{Nr~yEXn_H84QwosAxDxD)N3d(Yyq^K24Qlj
zKvRbm_?3BK$i=8pW=TISHm2p~w$$KRI^(e}?=jFUpsvipC7bcvz3O*VBrKo30K)tj
zur1d_StELolut9j3kwNUZeZZiewebf)6*!uu$Zt^KNjO7zf+pA+g`&4KV_bkpl)$}
zRuu?H^=RW8gwTM95M#tV&(dB@_eQ5Izu7+-dtvh<!`kUdI#e{XnsO!yl2kHR=DbyO
zT7Pbnh9H|DzG@QzIUqTSPhLhs@bqY&lwv;pyESHKJ-ChG%$YGUW1uP#sbKx_>k&BB
zF>RZG{eF)?z_A}d1Z<jswez*WqX1t`&)eA8%w^jHHI=B(5N=r{pcybHeSWRTZ!VRf
z34pSMx))giL(ps&^mB{}XHYr+Bu$J0(90Mu=7*$Cv*2F^7cMvuoa*6|VHOamnlYC(
zw8&iK1ZFYr_oZNqb*!{{l`~SRAl3>VviYidwR0j{NAg_ANgFq(6|wr&QdFsHH!_1U
z%UP|2LDE3Lg8b5gjO1sTtIP@jNK<3RHj@P)ws)}*S>m5tr;R6H%x;z~A?=SPX0bV^
z(!CO#R95_Nf4vj1YrI1f%;lTmT__h(1g^H}*x^tj+8WKTH){~2k032Tnx$`-6V;Pb
zHjA`6j0-rk*K%<eT}SzD0Hg~pTY5k%Jn5|gwGw1{s}O{hs?~I=*XWOk#Bza!Nwi+4
zaxGQNYEMtB4TR+q#!c>Txu`5Ml7`;ei!ZBwCpfhVAWad^a6TY1BPO*KVHIkk>P^fL
zTJ!*F2I>HgN{1))CQ_Ev7FFr#%a!G{>$J3qffO6doJ?ya!<udFDhRH_&9zWL|IvQe
zG#NUOupWNI1Q!!_p_RiVWl$C`J|d`*W-1#oY0%(<90>OBIg2r`rHftg$PCOeiIxQd
z49nb9w!%?Ot$Xi5DHTfs)&;xl1)NzoFf8<anv!<TAvi{gw$W+BXB{}Jwlxv6fk#W@
z0y39gzJ7(n@rd`|e;@132KRS&VyB@g4t9ho>Sx(Cl+8oescC==a*%1-^x^4RlS(!P
zFpIQo71}=rVZ|JI@meyIkJ1mn=Jhy`SUT0$byYv>H-6NYdR1fD3bgecSzL>bB%3ti
zH$>nt`2%3KrH-;~m9pM6D1+*2IVZ7MGW#6;EbAB|tb#Fo%8pUpYc18Kn`yPN;YAWz
z$1$gf?RJNGp3#qISgcyosj{#n_)&%f1<I+^LZu_l_BEgcteH*M#LDOhxN<B%%T}Gq
z5)?C`6olD2Dk=V?mKRvC$BXBZIhXx(`6A~K^m9mc)2~lwOc`KRmfE=fnul2bLImFg
zgdoLxub;nwa}F1)4y_Y5E2vi4)^ko$fM6PCr>R5%yE2DnWnarWCsL}frmIO2dI+It
z8YSCO*&)jo)PRy68=JEvI-Tv;3=LvY=81)rRt_PT;uLgiYfBck%I2E2Yy?CGWK13b
z70?$DE%)g4oUjUUM_SXbQW(x+Ce4syvy>B3QcX8orU@YLP|DQQ1})p1g4|!$=U(<X
zSG!h-wy*9ftkzK|>$Z-q%(*Q0x)?9{SD<)&TA-(&PX!N1w+cy1SN&(5W|et=9Q@iT
z^5tIdRy;v+(+(zzwJlD=NLzq}LV?zxT^}|W>ve}XNA&$r(5{uXm+yjR(2dm~(k=?s
zV@)cfbQ0#rmK~NxG~6H?stuBekF3H+n2WTC1W<|$oF2r+_5n~T8H&!GMzac_fMO?N
z3>mW=$d=!?SB$gh)KC|qOl~^>Yc)~M8I;Qu0~VGuP7ge3ktUUH%QOouHali4fSKsN
ztWd+4v+85DWGt(;&OS$8Rt-v8#gkPCY<30m94!57&Elm6n>vKU!7CG#Qd051YzV<s
z#!V3?au&UWSUWI`#>ga<NftGe31C?r*zJmwcF9m$(38_b6P+tX_rXQ%-KamK*ejR_
zn*f%v^Q!gs1c2lktTMS2yDtQ}41kjs%IMrC9anY$yBeXDd7!}#=|*jHdl(0N_W5Uc
za{UBf`^wia4n1Ou$T{P1J|mucBnt?z1RM28BV&dNg`wu+Bxx2|in>9nhRo|4V7Z*f
z(WgeQWT>Vb7xne_0sQFo>(1B`Rom?RyMMiXT~?p+%NIq9b+8O{<ef<?R}GL^imiZK
ziG1sUj1o$!0k%mbjQ!DfEUu*lvjx~`b|h0t_Z!#~Hm|mPBa-5=IL#u>(9D3(;uB{9
z45xBoHEWUUp)DF~*8X%TRfsaD%=KT#QqxyvKwj=d^`EjlbQDx0y{rUPl@bWzjF56%
zfCDZ;%Q6@(&r}1iIVHTge?UEwAHOQANmzF)IC41lr!wyQ-42;En)L>6Zr)(s1YB)a
z@QxR2b{PcNzRLh-0rKR~grMGq_I<7vkg<{v3-Ic^FXIQDlODD2tVUA+tSAtg>bK{w
zOrE~Sc_-}#7s05_3;Q_9zSQ(?a8A4<$x@c`UF?*_1A~g6DxstV0NL0eZVqBnEQpV7
zaxV>hA@C`a7QxumXGV;voRK+axa2e-QlbKe1EM@*C!~;!juz-BU@SmdAC(-#3TE}^
z8IR`?>1c^P7L_$di}049zK*)ilLfnE@+_U_8rW2e94bFI7SwwwAkE!x*e~CDy<<x;
zRIM5eL6SzfRjgVOytzH=<^%EGX`2=dmR^hYxrT6A4ttOaX3Hs9RS;~f6mv;E6>zwW
z<Q9vHolZ~HH!K+_(&)6%_r}n{q7Xaps%bE3UsY+rl8eFbqtDcPi6t7V+B9ZH(=??B
zVX24lcF^lWwFy&2R~p&nteO}BUNO$)kXT-a7;`QT9qZshs0RpYYS1e7gk`JGo=7HH
z){^cFHPQ4<u|~1+3aZtar&OLqWhYfK<4CkNIq4y^Jg^zw2ME$8v~7zN)g_)tHI;x1
zu4rA@IV4V+cCeXfQ5qvaZpSH$Hl<npuoQHWE7mVgb(^v-bwdV~Vu&S`9#a+%F=rlp
zR_R*u@-n_D!h^I7-zwe+MzeEPs~TfAxmv^+tL8MQF~DztH|LDkZ{DCkopF7Ajd7lF
ze}7-*Q2`QV7m>eP89wK{5XVc}84F-3ahb3ikD3UT9Vu6Ew>T?H6l|bmF_%hEwE&Y;
z_5v;&Im)`%^`QIDvK8Uq<!i|V0kYmR8CQ#Pjr7+2*bPfCXvWZssM|(BZ4eeelAjeu
zP75L!MT2GUcN&Z^oTf6Je$ZnOQ739DDR?3-X<pWK=Pcj?!5NSOqYbKQBB?$#>!u>H
z(LIxZqOmo08?H<k1z56Y+1db<%~pE|dxo63>>Dvh1&nnsGoWN_$sA)DJh9UPX3j+-
z<49UFn*P;VUBwhJ#)!&|0Lw{hSsw?&d)H5pIO8;&OG@wh`U>MPV7=YobUNU}iybzB
zq+1Fp(?y1Vvwf%lIymH%kvNG(l@SM^LNpLjwe(p`5RoLIkPHPk79^&At}Nt>wbV5K
zLW9wHfno?+q-bMNn0P(=9ExsD_Du%so+RglrZ9DlGiPZrNdh{H<p`^0S>ci+4_1&#
znGvI;V4X)zk3y?_*|L8XFhw(mFP~xigC)ClPC)@0K@6c;uu}pfDM<w^3N@(4wp{+5
zv_pkF7n2~_J|&$(mMc*I?hHeq>}>+}xO|WJ=XwM$eWI-msf}BX+Ojk!^pn|<|Lw1L
zv^%LOx3a0}M3x0>5mPPz&<=BZpmtE*Kip%Q=f#KDjZGLDSH4H0-p3l942EBT!LlF<
z)9`pxAP6&+Q@xMnc~J7&Ssu3er<KnG5Sj*@Gv=aQlgiDq;Vr;asYJ@P*PU(t2I7`c
zz%q^eHYhWVpJeAAgHhS=Qi<Y9w6o9~I*YQ(cgZ1a{a|g=EZ#(9_pXrw7Cu2_I>_=M
zBn7Hq*n9)Y5$1V@BVVGG0_vQC(uG*MwCHli#-cuZl1A*1Ss}?2SN*bLPNHMcew73u
z#j2Au4PmKDlYJKrLM=oIsy*`#<QX|7t&T;}EOO4kQb%Lg6HG^BW3Lghrh9F}A)ttH
zrHh4I5M%@jK`Yc2y1ep!n?J1MU8|XwK;v}oG0zcCpFY9+&z|Ac>(_{Jf@p<nL~L`f
zR3hqU8^bC7dMYZmXqq1#e~Ul4{7qPf`eV|Ar}6=Hy}q4FgMAkk!1>W@k+xW@oxhJ)
zW#}witp}||((?WIB>-x_>!`jDM5XfE7*Bn>+!q6s^~3Yyd$lp9`U_>ht!Z>UR85ad
z5ZQ*SXj`72c16gQEe0TgbGcnts&;L$xaBj-)s7_Rg*FbR3o<s(!WP}L9?c8VF;XL2
z=Kzm#aEirLMiUw}yst8qm?E4H8i>^czV3@g4w%l&)}bEX#Auu^<BSqW$i;2fYplB!
zZtm{z{MAdHzw!#+<7$7QK9v~(8Bf<Mcw><bOgIPU>;7%d5yI{)x^1YQkQ5M`9aiRy
z9IJDK6*&Mv+qRgd88MmdjHG#OB5;CcGDx!wCNn}|qGXDuGS)V?EX7*@JOnsB>euQ3
zGo&V+SI#-@ev_-eDOf<B!6qj4oU_6NExxnD5VJgCk`|l*q>G2nEmoLBQ~+72kXC!3
z5)_so#Ig=1L6F}wdlDm~NdqRc?}7a(Hw{~{0!m#0eZ?2gbdTkIa_+V&&mt$7z=wdg
zX+Ttq&&Fv+iU~1B4D$&oDl5yne~@Sw=OteHTVC%3?Cf9`A0uBk69Xx>HWU;*+K8Do
ze;&Y`(Vu%H^#YQ;O&+{WC{l`2qBtplh$+EqgY7`Iy$FCuRh#8niVu=phbJdAu|0I=
zGLb!~bW$K?_3GxWepNXO%a-Lc9Slrric&Xs@0_O09!Hx7^sx9@(XlOF2C}T0%^os?
zREt#)KQkDWjYO)#NWb4c*HWTZzAlS2FBk&5bS@Llv>yzK5_66SKByQ~K>^FgY~DmE
zGqRISRj{KPXu`k)L<>l2wj8Z@F2A)oJES;+TqEh{675$IU{y^895@2UR8rJl7n4Py
z2G+{~$+CzY0f|Y|qI%%uI<KJ5SN0o~1L&;IjAFl&QbbA--uY79OC)Xeqz#R1UWqfT
zzi-t()_H2oHlg}6F)DyrGE@e>Z5@gpi!jU+zIgEx7yBLFfBysA+}+^*c&Z|NM$vBp
z>hn}i+{L}?rpOk`!ZyBoh=MC{1(mmz&$4T_`}kGt+HxJ+N7=?fikxH*WIdSa(dYbK
zGj{cZA%1ct(6J|Ffvr9ZRvEYM{9>bO^Z&6OOab8vc5R*Noxx=*$I<)Olfw3*Dml@#
zYnkJg4ti_sG>`&ODR^-p!26;nWZUZ&&w=1osH9Sba=!*{mEl*wZ+o98bgW~IT#WVl
zv_Z7ME|oKsOO?KdfVFP3!Z;2Xr@0)II*4-r8e~aP1fdC`tPN%}`-a*qNrioI9(F`O
z+3j$9e?(*n%9ym+$~k=VqYtrNZ*Y5mhbh*>ln7Y0EiU$ZGyq<_e1+@n1{bYIur#9H
zPgFLsXf|0hI`vK@D*zSg+k9im8U+GomAwS3Tve8T7^hO0Ql>sD_#+QTUe5&?uf~ZS
zWJ(E1!F?G+T7VK;%XXew8kQOJlq}f|P!eG0Y;X0&@+&0;-z8{Ue4h;nD(FgamQcfc
z`AlARvQRHRl9Z-XR*v*J8IW7{zZDWj9V#GaXpXiZBnz?3=w(5rGOu)&6PM{+_A?9S
zNP}NT1m8$I)h3AMGAniT82|uET1iAfRD?#f(?RUnpw+!GOMBC4oYD6^tO!qlVJ_<k
zT#JxOHL>o!46V@plCO78#oA_Jdp()Ub*Vg+Ap<1$Ze_$)<+<H%(X}n^Z|^F)T~w)(
zKA2LrY!y{(Iy|*ESyYb!i_KkU=G)8_pt=>#YL2NJlpP$JFDHjb2EaMzYAZleF(&pT
zp@Ju+O)bp|J1`eox8?Yi^eE~<z5F+e@=E0m)y%!wuo#PHRZ*4-o;96k_x3ih^cszO
zuR6m;|3O<M#A1+|>_DO{zIPe$E+FQtLTN2jP-ce~V@A1}r~1$+AR>Sy1(Z39!+e%M
zy;oZY)$j!8L8`&i-3LmI!Z`;|D8NU4Tn%mIepIu`PMF%vT7d{S7a$F$RRdwmR?i+T
z%q0pKXPuvRVAW^FdYEfEgDma9)yAbDg2lmE=EP$LZTa^=A|f`^W&;T3r)R;C?8^1t
znWee)o4W@bk7qo2_7v@UgPXe>%+o9zT3HtbR!X3dm)Bx*w*K8TL|F|?79UVwv<T^V
zahZ$zUV^bI5qJz->&vc*%DOZl`S<Z+7&cl4cN(qRzc8yDhV*At?wL&U38~<;3c!pd
zNUq?#Y=qEpEBoNvw@D??0Cd{{%Ax6$eX;6q*0f}1OSyUb6o7@fODd~iJu%I$N2D2g
z$T_nqH}FQ~u;C~{zPxJCY(T+SX$zP@j5$RC1R<2HfrHoklKta+KGm~9T@tK9-jM)-
z%%WoktM$v`(9TvFi43-a4Yd#PaJbxUu{{qs%~9AR9hfKk4Ziu&`}pknD}C;ZAcBGI
zYJ=b%F7_ArFaPzg@NBn%%K&++y07O?*?Y5g2~h@9Je5s0R2f&(63e3?v{p0-luiUl
z+E!DaVfA4Z=~6~g&77Vy%oyegAn8fVY6>s&R=X@HurAo!5h9zGPj$;A8KbO$s{EnZ
z3N_3wfhK7RngC9$nR!k2>i)r6V4)M`5hyocwv-_Vizgj&{34*2VNltKJ%0-TZQoFg
zLa<_!MQWE>niuGjBvA45^WMX|fDi(F8_)!gMk~j2W-VHbn5P+YoH2}}-~*OoO4~GT
z9W#VLU}nHO(dT9Pp~+gLSW=!(@v-Exn||rnyKC%hFc!O!6iRSexQ42&`-CF`sff1g
zTEsbG7{<jWuc=K-QR?6AbBbd->0C%j(s89UyDlV~D{nj6ONvW!_$FEp>Rip75s_<j
zwXFCf+r;E$B6sK_%o-^<{WoH9#AYn$YkvIqUzK((-aNpg`>|ZU>9yVlhTi#FZKmm8
z6CvxT={;n?F)~}<Jp!_U95ZT^0dORXSJ!XZXbhal<4*8UhEEsGGHq>e$)1scBL_|E
zIP<C!syd|v3a$hI%wh*ntsKPv$a_uW5(#UIUk1=AFZ=8ou+5?o)Au;%z~oCcmw_=@
zL52=sUOX>v^CgREb(|xbM*Omf2wl@?)GF+<Dd<3wM^7w3GNlMl4&K?dOQm^MfKY63
zAoINd*u`8n8;pAGi-zg2^}?9qj4z(Q#MSi$KKkGTym|cwLoYp0MWAgZYk<^X8y5v=
zObLb+9?AF^;1MT(F3)@k76JQ^3OLI@muS8Q9jrj<*S!a*dw?B&e;-WUqm=OC<ClAj
zd~M;|^xCP6<1c~c1#kn^tc2{DvRbSVOb%&F8DjA)3=fuYSofd|7d$8gp<Yc6z*&43
zRR<N7(Y2>1YeU%q1Mnb(MlvFE7A=r~lFArK6Q0WOSzPgGoQO1|v-Iy2nS<Stw9Enk
zrx_Xj0GG?YY=?W$E*0#mp{op^tPx9XCtZ6sKg-%HfP)ZI#xPCb9P!b`7KeUBKhBtQ
z#K#}MkFS690k*3Khtr{Gb(gOv*Vou>w&=PRJoVUbuR$yo^&ryf*IH~Q`d`+2vhzc^
zAu_g1*;fM$*8WTP)2PplbUjHA_6i(x7IrOX)0HW>u^<$BCJ4F&GOO<?i?(@ji_1(A
z7-9jX86!tt_HHer%wTYmbgK1(&oD#atVLJCoC!OY0IJ5@tQQx209bTyj_Mi9oQiXT
z$zOz^wn}l?=@L*Yv30!^-&dIY#H0iX=94IWjU_eez4T2d@<oCJ#TY3jO!I_?(;0J|
zky8?%R8;b#tmqQtDd4pGR3Hrn_E;3GSRnMckgZx6p|1EZ{d#v`XB!`?4vtwO7QhN6
z0Fe5W(L!TEDuPdpz8~!a>&@y!TOwWa-vL`Cr2V8W*0-6OGFbXr>fGe_P_zPu<$=c)
zpGBufwWwmbVO!(nWfc^lI)PlUcJWer`_M}%mCt;esk84a8W(-6-v4rkh?Ek%RvV?5
zQ0x&j)yAb&jeucj8bpjwX)(qc6)(ZC>bP=NeNfqyt1S#zSU~dxa;o)kj>^5KoJAKh
zPXhSJTEl4CRmVNXR6wfv-a)TXs$CbUR|TxgO-Z804Cuyc9*D*mh@MaGA!{mE?v>TV
zS{GiS_E{_Q3`k^<u^!~mv=UJVl2m<)3FM8zfmA0u@#V9PR9dZh)nDnnVFo!b76LgV
zIc3Idj^(<R>Io<l)j3y0;bkq@ri@yVeS33*dEVgq`Wh#H!u|1}X0E2aLy2S^8NgTR
z1q@{AD$5EeVH?-409jlvqa0jIu(o8Z^@0A%SN+_&weh15|9g7r_trErl}o5w?E-wt
zfIrebLiavn3~V2(pP-;_0h)Q43THNwdLba!-*pC_<bdGCwdEvPiB#Ak30@2PqaUAh
zmSBP`?S;jDRHY&oB$=KHEP%r)bhDri5LS#eYf&w*tPq3L5{&>K;3$YjF4w|0yFh{z
zKBSzbI7vs%d_J=VV#e?gImBeaxey5+RspKgjx1|9YdPn}JFIsbU=OU<E&k8{r+<O}
z`hy?hVHk1f`x@}*enkviKe@){Vu#b+9p2k+(UL1?l7OIuY{Gs9jN?sN3s+br895dq
zeV&E%$SgqLlwZyPS^d^B07{Qj+P3a8XQVkHvjUb9ebgYO1l~?k&%~%106km8{9gf-
zp2=)up3R?D!@X4Zg-SvQ$SaJD&2@dOuzS>dus++&3{GrTa<C)bhyv>lVRI6;B1u|S
zT7uiLv3yee*0wF26V|QkT7*W5{4z_`cg`7Uj+my1aU3zvr{Z{EDP{dCxs5E|park!
z*cq5MR=P0vzmlLKSPEF&F1W5?{VbVJ3n?2}Aok^7?+onZiUkqLgSWJk)AXo&oHEr7
zT+p${X1$guWR7J6Tn4mq(3tx)7auRypDA-GLuwnSMHFtakt`St`COxAB}#JkYx=;X
zZ$YJS0TjXRvY%-269Q18Zv!BDz+1E9C6i*Tl#NLUL99KBW~N49^3n7N=ozJNroQE8
z&e%kWQhks#wQ26=c`jpW`Ff{%6ab~FI4x4nK>?XtEG(z($qa1Td22#a;*8*AaWfO(
zy?U_;aLPF<xY0vHpG($0!Q!EpbQ6&Ob3}-7E>T^17JB%QcZk{{Bj=1JNaSa1F0^XU
zbWWC%HPJ99(}zO6mI@+5*&B5nvf(9?h39=JjmjNa8lKgs(7`pN-d`vmMkHAbXY+}T
zX%tpP46d_oWVy%#V89d>wb%J<#+~}F07PE{1+oUj3Bc)a!ufc{)zuZ=d+!NuZ>50A
z(i3{Iy1_fuAPO&)bjP(dYXH4$*2@92C%gb{WQVq`_r=<*@{ZrMb-hcx`gixT;D_{}
zvD9lhILm1Ksx-Tv4Q>Gv$t-J@5eal((zKjg{<E@NY7Oyd6!a33^v#1~E6tOjO*+5G
zY3bQ?M0Skm!Kt59^izOpbPYsj1raq3YN=*xnr>j9Sp#Mr6Nc%rT>%@#swOko2?L)~
zLe#bBnBk<~V3|J51a^!RBb=`N<$-U7TQ)chE`R|bBwOzmTQ33;=}GSKWWT}6d&FTv
zN{rJmf+2-croSU1Twm_7U$61`4}O5_)w(c6ayS{O0a#_QD_T~EsIn8Ous4mE+C?dd
z%2Xb!ytHDOVVNs_+#8jlR#J};QGvANTya@o&T!h*Nx+&yGX<DM?w~<8fJf4ktk)~&
zeSjnRZ_YA5<V-DaPXh{=gU6yUk2P4A`I#)RR{f*NXqFa31~la?_3**Lxq#3%XxkPc
z1axhyQ&ITDlrpAq#`)aia5y5SBsNgY`e)YSKV7Jd>Yh;%Nj-lROd1LGb}(B=5ys+N
z#^QZ-?JZpb<a6y-HAqyiLj!d00_uQw1a>)Rz=ta9k*n5_QX)ujO~GQ}44!jF)3tB}
z91llro~;6WV@}`-cr$QfFN^35h!6?6=NxGsU+G?$?-P}Yv;;|Ja45Qp$0;sL*@*$9
zT%Uouw>3cq=BzZkY}Up`EI{896xz!)aW*z~jUg<|O`UUaCQ-NU-#4~xP0WdHJDJ$F
zGvUOU*vZ7UZQHhO+c^2|x%Zyawf^d=y{oIMy7yjdJ^T4ZCw0fRU_Ij~X#aFIgwI`)
zZn0EjDaiOITPK{3(Pc+_jv@d8V;*;WT-)duE3Ru5NH(fo9o|}%^wZ~9TM<53<u2A%
zz|bZJY@0K9nhj(z;z#w{4-kthbq7bY6P0#1at-C;mG>^ZZC02BPc<>n5?|<W{R0gt
zAe;rVJcKXNiA_<rSpTGT?I?JgP~6X1rXWAVj?TE#1=+YrQB%o04i!epSV&Cf%b)z|
z`P}UU#f=Zcx<o!x={=H7K<bPa{fM}yU5Dqx2di^>`s@7>)Rp@v@qMzAe<sG}wlAt=
z{rObL!bu)ZrkYlRgXs0NAT)xmwKI8@MF)izd-r2i9@7l`Y4;yERqq+?1YTW!)wfqF
zl4`Rh2N(A=RaxjUN!@;DtJ%HZrptt6=MtP)Z~`bJ#Kw@<h79#IV}s^rPO~d{uMJ!b
zv=_2ODe?ji=S`eYCjRAEW$SjP9Cnb3vLBLGb*Zw$?~Csy(yX8=sksv~p%~q97(H1R
zS(+9n!Z$O_WkiG^AS3FIAcm<s87}PrQjQf&BNSv$Qv<4%5%;|e;XV9sM`<JHG~Q(1
z!JP;yF<P8Hzi#yg_|C^3=IaxJlS>SIk*e_+-?h13D2XCJbS)E!22XQM8Q}z9S!Qn0
z;KC`9!Kzm&P2!w$ZXto;W#w4f&=&{z<{WF=!}3=_e4oS!FJAGBx>LS67N7-N1;Nrs
z#^={XhgF@)kOO{mt>u-|q?a?h020wW5s_x%09JHC$jzUc_W@MquGJkVpA-gEW`n;3
z29o)7ac;;U)6r{QdPnCvDx8Er)l@TJLxhx*vEld72sga0+|PB|sI5#6NB{Y<m9el<
zQGW3b=bcUA)`&M_oa=tWk}JUV<cI_wkE_*Q57TQ>*q6`Q+}@v=bSSJ{wGxyzh`|$0
zdOp3Q)7q?f^f1b$do*c&tura?wXVh~xjv>q|2}f+X5O$VpySz`>L(!H{iFlP6TXk|
zj=Ep^+9hBs>%Hx{wv|<Mjj(yYxsmE81zC>Cppdo5RJi0x$7~b31kv1CD@OGhY|Z6m
zA~ZzyTzlXKdeUs@0(3<!x|jkAps;~tW$(Y#M<Ys#n;(f+*?xdpH(+47_G9Rc(jO>x
zKzE3TQdZ26{uuHfmVo}N>4LDFyo#6SLYlmb2c}2Q9c&Qd!YB(t1o!&P^ZeqV)8GM<
zuhtXRo$Czhi{@P<ZEapbzRR8ecL=KAZG%eIT!pS84*Q3~{bmt<$k1ExZ$&Fe7nq|>
zj;cIb*bscE{vXKN>_js;$0b63ctAm^({9qv?cd=JFiWSwd0FIx#PJ+pM1~zUR(~7v
z>#}YH3Ln#+pR~YY{}S}wVSgLdf4k9}mIp~g(-WP8-4r6`@&L*jD$+!SN}{qCni)tK
zWd)XZw)*FI_^uvhfTAE|dkYKjmse`-9|27@->!<c1JYG8&U<;_NXoW$-XZGQL9tkE
zkcBYO?irmhWS^#D@nX(yJIR_9#~}>_U00iR2raH{a@gdLZ`g-2(b@A)yy2kFp2Y3@
zWn}4vH~I>^7<v>)?D84iV~d!|i6dPBfG`8Z%94F6C7>tPNFD_OL@D**1Q-jZl*<G?
z2;|U;YQeXzYN02{VhK7T7~K(<GcEFo)hu!gwRoktg7b+rCB%k}SYPx?Pu7==M9yn^
zc_?(>wUE|*naejoTuKD&6a_n*RJ2GLeQm;cDNi-~H|1oPGI)w2>tj1OI@?@e5x)na
zK<zFF6giZttLpI8>-NUe^?gRuO_R@~kD=05k-4urQwlbCUM$u^zI#2RsLR3S<z&m?
zB;p+{czPI+t9YIJUgo<d$%2uCy{(gW^E)SuY%@d9FZ~G0<K8K(`)}Ed3Q;O&pMo;A
zdF$g{2WwtZ{=WWntJN`9{j^Je1J7=J-kG@)jWe?y-DqP9lRWfV9rON(Qqi@wW*PjL
zrAodXuSejM_dUVqq1`72|JSn`xkC9MawXQJ=*c`XcT<sD1*dwYLx?vU)9;P<Y3%Y?
zUW}6?W?qR%uZ@glw)c*=y4P;+=8fr3-uJy_zI&%m1Az0FJh1H?*J<Bx#lPIagq&xs
ztV&x56-Gak4yS>ZW;n8XM9Se=3n&m4Dy{f38q@YgufWQy3c1f0Po5+gD9e1R8-SWn
zhYR5k(FjJ$5faiZ5k!D+mdF23laoX2CIHhD&)_V!J9vg24v|?>hzi_&l)SiD2g(&J
z;!pI7JP76zBgwH-6<tA~>M8Q`&)wr`OOtC9w+h6(=!RDDtEF{oiu=ex&TCF19FI(t
zMIlDZGq+V4$^y54m4WmbHz9FFmew8W+W}NRLp||HGoby+>ENt?<w!a)j#`S&ofwt!
zC`~a9p>xWb9xGMxG24c_2T*wq!B<>*wUNeaKL)1J9WlZ^I$a*y8nd{gp|JuAe39`g
ztM*uDwXBh&Io7WOn6iU$UH0yP!F|A~&QBb>jt4M1es}-0HZ2-3pdl}z6qBY%J>*@8
z!JOM5*pkuwMCL#<y-%H@2Ud^gwU#sKah-s`(}!VeNa^prbwTnjny7&uK=cyciKT0`
zzf7+g*wd>Y!1~dh3XiGZi~}GNzQ(F(D8EBfRylMWqGLkCYbh^a+0L<A1Q(la&3!dG
z>z4syXcb-`HiFOT;-$#mjQlqbA1&;iL?C2mgg&v%aCcZdC1qXfOi?SslfZITt)RD_
z{O_cT;2lYV5Rn=z<MDvS9|?xW>fv~utFp2r4xe`@6OC9Adp3D>1e%nQVe01qMDGY~
zTwVUITeJ$^7}`M^(^5_4phW*d+VvQ>Oip=$+B4~q^uYuj`RR`KezRk<Xd7lwh57=(
zbgDXoosW$${HE5dDL8&<qi#a($-A}u7A#!7O8DZz;hPEo!=42wQ#*=D)?bU`Ta<dS
zE*}qlUP~mX)N9(-W82inGo=7rY!LWW;{iPQmRv2dv#T$mHzZJ|*SXiVeBEAk^%rG%
zhw;Cj$lvUb7`>mHeZjkq0V^JM8E!azujo7p_J6cNTq5ajrz3j?UNb%gNw82sd8sum
zjVry3K5fj2rwdCgBKb*ylJdd2<Ka&-=fOiy<2w?>^tpN^-I$iB(>I&OqCWD$?@S!-
zs|2}Y4>;<_knMb4Zocl;APKZ$ni-7X{0TVc0r<D&&o|U4dE2aOFNQ5sYsm-?=AvFA
za~;X|V&&e|=ZnwIN5(CNMT*ywf5jta%A3l>E5RuTQR=oMZbj(MfU?Do362hQ`Mkd;
zd&&!5Gv<gh6}B4Pz!B(GG&H&53Rt+{mxYYxfaY1G3I$udTRUpa3rflprl}WU+mEKJ
zmnMisX3OO_1xX8yV5HrF*?qgyouF+(Nw2FcG*kavxwC_x<1x}uTSBZ|Lbqh?9*`Tz
z?b2{eIAq4MMs^Q^dI#J;lp&I#$$+)o1)meiCD%6VoK*eD4n3}S7P$0I=O){aYc-dW
z0{om#-3X>;IiMsk;Ovab^a7BaSRl`WIo_hxiW0ehI~EkE{`q(ZpL8|*&+`2ZcP7x&
z)#sj4U-VC^(%rdZ=<N5_P`6*Xy3pbozDbFk1R{n9SK%*l2wt));mXM0Cp~XZVVk@c
z<?%GNY*`i$**#Q6ZZS<PNG|lDLrCORTZTV9;|D93esXSa-diE(Pf{UNQ#edN<wB+%
zk@2STsrKEzqV#9{<nMpyGN&PuOH?cSXL}w7xDh*N92x<(yESk<wY`28dVy}w2xL*T
zQqeo*|9RYel~W;}u(AGG^v97{2V^k!nm2ZLMb^+v?w&p}_#nz>w0N@a=cbl?w2t0Z
z_96%!&1NUZP>%LqXR`OnUbyu8C#3w%*5|8P%&K2ohN}sO4;@Hr%KFWMcOzMM8BgS)
zzDn-6Q~~AbwwdDivML;qgVe$rX4~S-nCM@ykHtV{Gu*Jc(aNglaO|C3NTW5vJ_Wy7
zQF^T7&y2=&rYY|8Nf(YaYbM#cib&d<P7|G)uCAe4+UAyyo>q*6QynHvM35O5F_sz?
zJ5)2OeJZo2p)kzv*FE+5K6Mix&DY!Y(e#Dt{TbfXm!!wb>`l0ofO?G{QUAJ<INa9A
zd9l4brO5R;9b;E#52A2<5P1gWbrn?VYebq^Q<J$+Wl=3F#w)F8gLl*oP<XNHJd+i`
zzaH!`#1~l|b=KcTlSo>P0o#LYnHlr8msfsdDc!B6-T7Cti=}TaoJw9TSlg=h4@KxW
zZXE6Gl2Oc3{ns{CD<It~kIYUDB9_t=%TN9HW*=6AzYiubsgw&~pkf%XXlf&f#f!W7
z6(bGV<%-imuw%c~xqv{QJ)9zS{sGWiLPz?m@?_<90O%nc?>p3{E}LD4g{=z7Z1|A;
zeby<!tpJP=$-<8*NT;P!ii{>6p}vV=+-QmLEjUH-AnW$m)2$H)z>7`_I4G7Jq7YNf
zvVp-s*@GTI+;C!ad25xSQ>Ew^y*@C(00|W2KJDbJF+g%mA);-PBv|zgqF*rA*L7DY
zcufe2v{Jk)C!u9~+HiV=P~w$iCD(`3km^OvAQVUwYwLsy29K&aL~s-=MQv1}`szhw
z@MvLH@SaPPliZwK!%cX%D=4C4vn7%+n>qN0R&2xTHZZb$Zhll&+GQJh_nDiX<Fq>I
zLsSIReDb_@4SCtarE(#L^Pl*XeyIV&$l<C!_<S>fHW1{muhG^5+Ic;61TyEzy@p_y
z2aM~wyi^FguU^YZr6FpicgF?ef!w&ERI+(xWLVYyR~Cxk{_>|KvPuRh7N?yqN?j`n
zSS$8Jf_MNS#WhQ0;!00~Fi_o};`U+{WqJ)oTl0Bg0lXq{+vH3amNU=jTttQ=s592-
z$f_-q17=pHrfRrf<NL8s5j=RC-L)P!fyusAzgvKSEu)cUttnj|aSQZdlO9{<J;)=o
zX6|zJ(eJ<Zh`8HT#nAIu*4Gl{d<YuL65+U<g|YDZV_Fd8@|%hvmuTpNNUHx3{=4?Q
zQ(v9r={$~Syy#;jY~{=j6_s@+?iAP_)1h$!CIr+o)0}ZnOak-a1%~oZKoj@YH)c<;
zfq3fhm@pCam)}S2_Twi)Rb^#0E&bOH8j4tZvd*@!jrP>_am5$UYOy<7Dtb4(Dp~Km
zlUVP@1_A%q%ht|C)7GFk{|i(!Vb@<huXiQ{U80a!kK8^UuIfk-As@QD3E!LGv?^E!
zzA9*Q>1|X99VHBwS@%J&b$<F!lD_CPJv0?+bKG+Wu4Zrj-uedJ?!Q-jo3w_I52ph1
z*xd*vr#r8O=q1?6&F$^o1GZ-9oU7Fd=xY_ogr4`K8icki<ywF|Wr1bUnvsmclDo!|
zyWMX#U!}Npad3(3Um;8QK&d$LJj#mSwTymd50Q+8cco+MA|{D0oA`in&ODCiJy&KF
zv6$}!Pw|!viJOpeTNmCJY`$T@Y`<Ln>~vE{zC-cyuYvAnzz}q~y5GMxzU5cyQr!yt
zKIG?YXXw1J-AvOsW1Hp#R@v_Kj+g?)x{>yhT=W%0!P5ocru{l1h!yJ|B7d1@6UOiH
zWJvk;qB9Lf)PbK!#h%sZaKAzCSU^jM<3)c;Qz^J@^_Z5-!U?EFvL@*mg{qVaBuRz-
z`UwS<%IlO>w=kDViZ75|skhV;P-DuA-lR$p8&VAaCvxzQ5GY4{no62`dLvsMCU>*D
z;suzM8#*CasqTCn%3{OL%3ad;((?8Jj3#7|U{t&mkEV}RYOD|CCKFVEf_szIM~8sl
zpu!%KP$U8z5<g8arH;fFJ=&bCZP`v=uJL!eS6m0gbE`Dp?0OB&{852)6kc@;EKR&%
z)PhovMR;<?lk#=SaO9{+(5TG?@BcIdH;<c#{b%lkMFn|6$ve81eY4`r9r&wHO^UIa
zaeOWA<tefD^#dgYe$^7;R?CShPoycGidlb@St`#(qp1Od=UT!`SpasYL;R}TNpL~e
zTzq1V3>&6L3FdJiJu^4#p!y1=>r?$)(D={$35A=3Q@E-xS=i7Kx3+ELFx3uXR%i4t
zL;;QN#Xo^H+HX2otNm7@5oOX88;O2b9XvhHZI{{4FZnjk#qjL|D<q8g@;28(=IJ(5
zHT>uJ<-ZX;R%M1q!h{PgdTrbKT-y5DQ-}d<Ymtz8Ja4{1-@N8;pY8Sa&Fz5&=>(w{
z27X*n1kY{be&*80_CFtHk~;EeC<vVu;&0f0-8{KiULKNbg-Yla-6|eeWNA7*9^s}c
zGk5;8U-Y@F`eOBdxEXT$+#RTv2U6PD-tn$9F^5kjBO8y&yxI0(x$y5c4>UaK%W@RY
zhDSMrMW4#GWl6(lhlq3Zcqbd;aeZ=@FI{nM%3Ss%myr{2Z4S`^eH|9FVJhI?+C1?t
zFOfdT<Mo;4Ikqk$z5pENC;R=2>4DTb!^j}j2V$~dXd>_M=DZ7|*sL}zb``A7MZTJ8
z7|ish3T=VGkvs<yoKCkR_y(8wVBP`}pwrE>91oPTN6~-~v?C~<q585`=ZK$DeIY^W
z9NCRH3pEfV?XD#Xy21b^dMQ*;%)<FiSURHexC<OdMc<tw=e&1;OHqm*(u6(BDW-Bd
z-?7bUtvEM6@~&)yc3F}l>Fz+}qZ#o4dY3H>9ji^b@hgoK@<#VE1Wnu)DnOzmBPZ?y
z=+EEC+e%iDJ#k34qPx0A8u06))!7>LOwuAE=h&+cnz30E4++fuoN-_`e<t(h;l=3|
z(%jzKxi<WCU64|vNq+6uuGCE&iwZ}4#!w8Hp?i3rt_#S|fcTeokEty+w{_HEcuUHd
zlKg0Y*k_uYba{R~J8ITSM1VsRx6=4GL=aW^gd3$=p0VenrC;dN3w1Pj<K@@lfJ7Vh
z<b3bB7S#7~dn7a9VvEPm06zfOh186qOy;IC*oVH|kqZF3u7l}C;4`{>{($cBQnE1L
z7g{y`3m0W`qw9X^9mgx*b=ZIweZS^WzghkGcp3#JFTEN$9fWr@iCCY`oVu88M%Kk1
za^c6b;+=Xs*rIBRSSlS-3=qvgiI?H(r>2QaC59ow!_V_OUkYn$Fzl*S*Xmovae;1L
zp43IB8D60_?|R<7x^eA#M=yn<pQ0zOCh_Beo-BfXu#bG&dLtaCqQSsZI7xxyN(BfB
z$L#Kz;%Fk>q_NevdWhgVG?sBi^Zxkd3geC-(n{EE+V)VYxL4D0H@UWK-IU$^J9Qjd
zZ^_?8V><N#yQ8B`xJ9$vwI1ftgcUa%A~}~ku$QAFbCqTC@%Rr>u38PCk8`Zkbnwn9
zush%b1kg|(2_CG;Av1Jj%03G7cG(Kj^?E?z_VNxx=ygv_prgBe$M8uItVt(4<V60s
z8*_<<H9%e|i8t@QpC;EJ=9RHUZG%o4y-{N9(b%{rgtAuExzsOhy3UVhBYSQHm%zT&
zBu;e29mOs^z(zDt*aW_HZmlJpsY5H%c;ahQKf&VZ9Kof5Frf(zjpF#Tbw-9Mlr)Lf
zchGV@=&+dM0O8j;k!_=W25#aUm|qbLXoqWbh$5lbq0(f|g(?Wr>l?%-9>mTt1Pte6
zz<Wp@Te}_4a<f1&C>DeOumV1AUZA73GwV%bZXT8|FqjdDm_ElWfxVxe4eAqRp(BRY
zf{h|L98edj;{J#z{iG01)g9nftrkr^omzD2iO!gpzt?p_Ik9OOp=(gj>6ne_MXqz8
zwh*#wk;!y4a!FA+)PkeZcZXJWJszh3^f85akLf>LTf(6^hg!z&$-tL!Y}2qD$Q!P-
zwLLjHoDA`B5Yxjwf1J^b^ud&`V}t$2VK>eCxxUyi2nB-&OL>^CxDkJd$H~x(xYX(T
z4bnK}iaShqGa>!TOYD4oeUUzI#Ql!)VBSi*D|?3*sH@H=`cCf~kft|bJ8sNa91z)A
z+9>7Mt!k=Oc)mz3KbXi9HzNDUcVWD)D<j5J=?!rwft1k`eQZm>7OLcqZwje7q{J-R
z8JZ>DPJ)xj%;1G66>KY;ffZ3qA@T>)&}fEs7Hge&kS_}f8lT;`NIxFA!ka8r^Xa45
zMk0-T!=SG}x-Bb*p8QUC^endbH8bumWEM=cWE(1oe-O?jXP5l1>U!40L`4U;z;cx|
zbj582*cg;V2-*d{C-0UaCD_L?{{}SL1`QJ$P%xzNHk<s$Vt!7$sq64){+@xCzpjEv
zv<+&{YIMnkffuF-jFKLR%9T2BQmtJ6WJM9roQK_G*0Dj%!=xJ?o0G%JgU$}$fIlWX
zwq+aF<%`Be|C)qsHQKYV@6>$l&9l9^)nAn+Sv_1w*1lc19@3V<${4m(ZDDZr8uK^N
z+*4r#;TdwZNSnbXt&|NTyby3Y>XGkdSV1+01Q5k`n+W~;IfK_%h#`X*Q@R3F?K^*Y
zw6ndw?<;yfk^h3q^7*vV?(i@H>645Q#*P&^O>RBfKZ1T@Qvoi6lOH?LnGP`It>K%F
zCG);6=A^G#Qj-Mzo4|H#XPGq-^W^m2O<Qhb2X)c!ul{?4s3Px)EiQTD0k>(avX>>x
zxp(<F=|(^n*nRuH?{~us*FcfK9Ra>I5((NvMB$HITcc)3*vJU_rxmjo+K4hEoA&-h
z{W(^NB^4pl-$&pO1VdVD!W)&UhO5kzc;BhU%_zYjhdAQ+F9WvZ3a54ll*qr>5uAle
zY68hS*?6$z!<UdFF+Q9t?#`qo(J30&=+HE~^u2`I0Lv|?>-zAJ2TgS8kBOS%U<$#Q
zN-|_zq~eN~Bbr*{_<W#Y--4>2^#HnzQ@C`hIf)x9k(rC+hr&Lm0ohe$T|eUsT4o0F
zI6hnfo)XAPhk55rt1+3~myu$NMQW~+eT8bQSVceJ+SdZKHI}{80Kuuy@DC!48G{La
z7_Y~D;|)Gf_ZbWZ#EL;2FZ_k3mUjS#BxQDU^8~73ke2R^gakN36S#`sSU5AFd*d5D
zGuqtVERagB-O<VSljA-l$KT)b^^ZYar$w^i6l5<peEy%khddj1EZ{qzSia%~;m&YK
zE+)gW%Iaz;XWbcV6y|~K^PhC-Z+t}d5v661RZ+SKiRh-8--)IO@SBbhQ+n7hWdw#a
zJgho?kTuJrNu;n(Wu!4q-vF`nXYWZPp}fa03;$#=TE*(kIH$*%1rwfNLD`i;)cRsh
zqcG#9YG+jzA<(gr*ZlD(P(mys7=K(=me0fj-pZz#znmi>%u7Qtl@l$`rH~)NYJ&ng
z%^11zVvrIh9uma+c@wcFt;y}?{fEokIRJuEpdLW0k~2chb2JgrcE%mhRGN&cycu!W
z?2>+W4)*BfKAP;<=w$!1o9!TCV+A1w4WX57FFVOF`=Xk*xwCVywcLE+UQ(kNW2)<+
zCD5rerL-s>ZM9b0%+hMw`J-9zgFcQMQEL69!x?hgWL9}LHW)!=cXqCATmXonX!}d)
z+9^8SI3$o6peK*Td15~VWB&%mf*!(xHyRf0^8o!HE7fSeS13DPcmFBAFFdOf41<R-
z)l2w?Y{Zs&?WFoQ48&Tqt0V{rn`RyHbA9iCG4z#PF&3(AFnDF~E$PGidnZ*l_|h7?
zFXEV|=ty=D7Z$ifrbchcU6VuG=`;k8y2&CgZvHxekc~meFeIop4hWwSm{fx(s14~<
zt&aS-rL}47q^B=Z&KQzKNlZkDwc>&Y7|m`s4PtowLV&p{k)U0sN{25t&|@NP0-}JG
zB1ewE?l^Y<Owm0R5DgKv(6YkL-UBU|=a+^<=8hA`=C+%FV-PmvYlwMv0fRoIBPiyr
zEW)_b(S4NSo5UmT2oR@7HVk+MX49=^%n0*Z7A*4qRX0co+4d9IAPaE_<Ty&`tf4k#
zz_SPHMq^^zIb}wSM<8UDSZ``R98vlbBBa)1qI+r*hpJV+@Z7^M_52lO{FC8lc@e^9
z|3@+ZL`(SYI1-(?xXMX6)(p1B-i-yj7=9Hb1Md>ta)L(WClR?j-d~Jk_>~E!lSw~?
zZ{crwg5olC1+?weqxx9bSUBu_Y4i61K?p&3f^gjGneCT5Gs_Q&I5RMV**uW-u^|Cu
z@hjNZG5zb#s~%PxMGh@l$@#i3mk@TMkmablbe}n8fkwh}97zpD^j}At09)k(pUNzV
z83ft`uzD2V6v$t+RLy<nyqVfva~_40=~}FS)qYh8jU_<o$_a=|OJZVCJm|vkWHB40
z<?*XcbpWMoxtlni)cA6<C`7u}ASLCu>=7j2`mBU1(XhRKR&_loOt?bbsz(z>s*HRA
zq`$QZ%1A6oKOW2}vnI{V`Yv$7l&#Z_ZAf`KhMl?6&W=$yh8>&#^LXg_%IAaGdgWCS
z9XvD6YH_2mheY^0+rNCrW6<037n+F!OLGs{cmcmn(76z>ckzyzXlH`d085*QSWB}q
zS1C6c*!Un=Vqa3NR)i~OS|bZuwEhhi+zv07MuRwQs$g3}Tx+(SssvkZ+!gg(N8o58
zU32asp!u`(ib*Rfqi#Hkc%_ai(ec3j>`{p4HPBXAg~w<B%?pK0-8VlmGo$Wd?_JvQ
zu*IdBz~;ui^@F%;gWAmQyK!>UGgtH+&;QB({JbZB{eGYVX2T=1dxq@r_{<jNRh&7A
zda*DJ+<-Kj-7M?mZ@qxxMqz6xvKfiEKDp>d+!(>_5^o4-XUDcHG&<7#$A*>Fqb%f)
zwkakIzSdHmMFcoMj4bH19N?Maecg%YgTY<{dlnKqn)(skAv19>p9gI0JokmL4v^Dj
z<=z&Z_J)xdEf+_o{q!;|jp&{N#!m#;JoX-A_P{Ggt-~#NO#Yx4$0BYP^aA~Bx|xlt
zW(9lYo8BT)g5*rEeLVS+u4ukIKV_3QbktKBkXA-u6M}HL+!+2R^)OhP;2D8DgAu9E
z3qM{SL|ufMSu-Vl>*z!_ZjVzO|Gc;C>D&OKNlbQMnU^<(%-t9cS>?wycxn-mY`PLo
zdlFIaa|Cj3vzjPcf4x`>GRM<6+r>V~LLm}E4khBGPh<N(r&&marO4p?c@w>CF}6j6
zA{rieYe>a<vs^XobtMEo$hV+VE~dnv*S6>rEi_aCQlQTMftxhYN6^J4=W6G+KfvDl
z{7tBuIk@|5U86n>29z+;l)D=$&r@5A)_(Uxy>goOZ#xa3U7PpzCg+I{oxGe@2X)!`
zic=H_0!li&kkMWDKYK6GB3TZiic<!bN~=jG5y*iGf)wLVe&S(4`|&rU@740#K)s4n
z^o%yq8%rzKrMYS^F6<W^H?b97-kV{V(9=K0F~l1TNSKPM4=%EiR4ptlvf>=zM2~mw
zb-R9nxf3nrVWs{iI`&<v<pug(7_<w%5-JRqjz>IHb8~}qnNGd7snrwMX%@X~Q^~dJ
zzyx(x%Gjn>d?1SV?VFEpRhG}+?~PK^rWY*X=jYX4)#sQVQHfb-KN=oDkQ>h$k-TT(
zfv&Prcjp9B-og6?EGa44+FKiW=X?F&Kd-?<Xx^%7?(FC>@afi!R7)07z@TaORok6g
zWIauH?|}`^2ixV9_6#z&))>Jme`Y@1)E5JgAcy2UlIyRaxo}tO_gZ(iVqiLu+eO&)
zBYDD4mPPX-h#^3Wq^Dhg$HFy{T_CUN8f187rF=ktHYROEe@JwLq!XBi*U=Q5|N1dt
zK=*twl(l=hK=bE$2h#2S8JX|poSF6vMo_3Ctk)|hnf1JfM5&fm_EZqn))lJLz@na;
zp}wN#B7JD(0?lwzq6zA4aLQAEO6-lzf6&LeU4wejM((Z@{V-3fE<_^p=e(jD&a)a4
zv+Y@}z6M6OpQHixPv6)_EfuRQA$yZ40l(ru`}z36fh9A8k8BasScRIhO1+WhR=m56
zajZkXXo33m!MQa-sB6+E#M*k+ISzeqV~K;X(Qx>o%uR#~wGD1VFz^tpp`*w)Bpfru
z@T=`#)@6!F0-54Jg8hnlv~N4@u`0&OX9EO&eQ$9S*(%SQmSv*4H}PP3576+)R#R5d
ztV)vCkI~d}5kKhlu4TlBe)lS^q9G)bsV^;v&~Hei@vDpW2ddYRloa6%aJZ5q8vKb{
z%|#pWUFLy@kRyQ^h1dPiEV2Np7uDdyOyzZC;3(GF6o^tH(2mFHHYpMAyJ9I$eXHB%
z1>z{mZdzHOa~@<8powS(1-g{o-g;Kap*>pVf1M5CQ%VbJjhMsZ>N{Pn)YZPVwRC(}
za|TuhIA@7>yt`a)Uyt?rvi9utqJBawgHY#IZ%#h`y=GvVdel7ewS8VkUhfFqc?%v?
z=MXKU14uc2P(AEIo8Kd1P~=f9Vb+V^7fk)79$CT3PAc&fJ1=#)o`iz>^A#Ros^S(K
z6iCNROEqmjpX_J8qb<NYd!cfc`dyajnAQ2yLltqim<~#jnTamQf%l8l++l)h5)>V<
z5;$zdY}xCr<It7-XSf!lT(g5zgkQ7bx$m>{H=rqlxzT-)O?G(__0Xabs#M)I0**r`
z2c!f<^JDuQpe@R1=jwJgf-2)Qs08yl^z|Zd|8@%F^Zn!QwyhJKME$O2wWr*SecO03
z@O{*C%a*!5o^vXVKEK~hjV~sFKp_0V@FasDImX)kDHa!Vi{?l%dL(KTS{Vs(FN5kk
z4fe|)Dr}9-4WkU+YWTI;MO)Thx-#o42!YjC6?R)?5}WfE#BpHvlB5wS2<rqlvjKig
zjk&-t$nDd+D`W|+fJaVZ{sr~e(i@LrUcG)@k)&q-rf_EuVVi}X_6kz;o+XA#``;fp
zSKk0o@7JnNbiTKDCvBaM+PcA`n$QX2Ilu(XQpvG+aLCaQ)=M|ujY)oIO$+8WQR739
zQXpE(&_6bX=&K}zLRm@vw9YX%;V7L8E)f>p>-rgE50P@P{e>G>vihXO-5h4aghDDZ
z&ZudN@d&5C)O{ThnJ;87hB>Cw4v3;z-3dfL|L)`w?_<5#lHr2lFH{NJt&On0IZ73G
z0@ruo>7fl{(}?qr8G5km)dr)GSqLuS0_w8;mN*=DLEt<CopsdW^FwpN;Jp`P+a(=2
z;Rv)z05nuN<a0Yyg_TzM0irw}$Ed|ZcBsBZtxhi9q6c7PR}Pd8Qws1K3~V)OY8FN1
z0ip*|BHZkAc#*vpJ0x_eo(C0-#t@TTA{reW>P)uUoQ`Cvn)A%zp>%b9p(<S1*AoOO
zhetF^kyY-0rNgdY@&i?YVlza$Die!sD4{%vq4Q-)6ssPSsPhk1&daDhUoup=Lz2vz
z%r<?r8}7&qC<Y8l{+?QdGDk!RUlJ8Vs{-e)f3rcF*RE<3{U=<)zu|!2SSKTg4*|Y3
zvdM=0fCB)eaK3-26oLA_KLHTKM|Swf^@|hVbl$R#(oPfMJH1{9kD+iOfN|1lz08d`
z0QUnS4zioif3t@Z%i;%Va<r6{a|+y~2p-s|dCPN8mY`|;&`pI0YW1VrCqDLWLI8$7
zw$yZK#$J3shZJr8^;J=1wZydby|3jhK?BvIs(mr|TkW=j8lGjJQtBM6+ti2sSX^~%
zASW~%l@Yi6vMe*(T?7?{F+iifX+jEGS#Gn1*dUd{n%pW2JUL?M<%RRx9r?nAyruxY
z0GL+iB+)qDZ|WNKl;!RF^^WNiy0%p1@cX-@Y3ft}|6~R3z8Vp7afKLD*xa>&OA$3w
zlF&io3Mi6fWa@x+A}XSc0s<JSsaelM81T5y{=myAC+*=Nx{leh2*^|hKyzCyEtku;
zRUnZ0so)5=uP$WeaJX-6$F5N-NBC|r_vlNlfaeLqL8)Ge&qUjqI%#@kv3nO*c{UBf
z2&9UnH7#S<A`V4KDu~_<yF12UVNjWnT}4l{7)%@M!18@PU^aJj4tVl=U?v@5KJSiI
zDUNZaBVRfX#u_NPY294{{ubOW-8EEn6|G5Z#jh$CBR}k4^?ApJOMg`<r~2lQV1{@-
z=?vjMhE2?T&8HKC5-H0SSA08J<Gi>r`}=R0?vYuFmf&~C&NU_qeuAM?Bx$su>qA%u
zduMhJOFwh)N#-z-@r04_joJ13i66yB>W5aD{9@#W(oZnX-3btHic^=eabwKsj`GGD
zL&As4;A7u)KaKX=Kpm8WM~0R;IhN-hFaLJt=$Js|H1-vPDQd*A!j74y40|(<h2k}H
z@!0l?)aH01l4drVG&6o0MDg*C0*?0?4DiU;K##7Hyb4I-e4kT%J#zm?$RAGU24-1e
z7)_o}X0TeAUSr98wK0dd56R|A-3%?rpoSdf?;eJ)p9Ux$J6v`MT3y!#hiVd@&TZh;
z6rN?Ln3z>Rd<k{Cn&=0Ck*;GY$!V^flqR9lv`SL5h)l;zVpnjicx5JGtntqvsL4AT
z-*T;*7sUI)U9QU^Ab5_vRcA%D{JfK*)VovdjP$d)j8}FJ9`%X&cYVE{kgEpaA354C
z$XHG>N5nDzD;qq{L@3Qrw^mwHrHe0;-TvXOqTT>4Fz)L<+gm=6`<2vvv*_62{q5{U
zh<voFMnpe(g1<60yp9_M2?0(peei>_HVI-VWBvm8L;&9ljoU3pxj|56?oVlzrZvd(
zrJ(O-UUenMy#J5kXH^~pj&7I))0&Rv5mYv(L2`MF$TW)c!-$Rgo&?nI_mojix8Ty|
z*9&tuIXP9c=zr1P!(&D}=C3qP%za+*<z@r_ZQ8<thv4@4&M8%wnU`wTe)p-|{&v|G
z!?IK1y4myiCtRNI3+(05em^E&rGVfs3UJhtWdAo^C;y22zicw>N$DLpQ5Kj)I{&eS
z?yMSg;UC$W9y+O9=7h6fkqdq6*^JZ3f0-J|ccwUgsItbd9ZDfrg6W?kXRbiK<q9_C
zP&Sg22C5fy^>SP~JJ9xurA>Gl4*07U|A^InTq29~%tl>Ig=O)qfk1(UwP9!C79GO{
z6cHYLls|cyFd^?fv$DqSs8!iIIJrhY`P>V?Kc6W7M}VQLqbHn;BuOH*)Kb}RzwrY7
z(Rp9Z82<<dOy1F2v-aMq67R`JUUVSzk>8u}&A>s-Ss5YV{ETUsCN?afcZ1%3w}upu
zCFW`w#LJJ5Eg0iRs$TK4PBIIsP;+X$&VBV`_OG9usXN6EPgGG`nG_eW0n;M!|M0$w
zMS833tL^tZoG5fHGl!6=<0M|L8tX<DKCbKWBk-)h^&XBa+jNXvwUYTE?oq26GUX{%
zShuJh_hWWaJHoKwVqbMdyQgPvUm&Lz>Xs6v7@=@J(yL^lTZ9TjR%#_z=5oTy@QlE(
z)Zw&0&z0<*St#*aEyS$VTFrLGesY)D@z$46el4gnZ7E195KVpV?0m`$uB3y9?u+`*
z@7IQaV|-79U4{4a0a!&rz`r0y61W4J9;5@9bpb78ImqSxif4Bim3yfw!K)BlJt7^6
z72XYcK1=;*T`OOCkg0Rz>-irJ=bM!5J&zU!5P6!_MZeEMBlqHZX{;j%B9lkxR?VGS
zm3Pj(yP5*r+C=%%s=Kya#V2SFV&P8~s!#}}tWgQrgN`#KLWZMG7Lmyy<nM_=-s!}a
zV{Wd(w*P%M;jEZ0hwxQQk<NQt%`)Qz(uhLL_#7YosTwQ~08tLbz;jHP>>|E`AxdkT
zm!mMDSfMRsbHRH^qz-jdTG_SZaY+~ZjCPdy#Wy9~#J;e$KH!_lKqp4|stjXe59wnc
z4_TdxQ)czX`0@lQ{gi9-a>mZhzz+gF?QO%l0hdVuo8@$yoC<^6h`ZC?oeH%127Nl3
zL=(XZ^r@=K>cs^TdQ~MtG7mpau^Ng3aV{yVW-YaLCl|k5|A|YyuiNNPB)g6~sIO0-
z7v*=ouM?l(%5l8=A|vkMMGHCTx%QsABL-3>DP6p6HUeHD&<i|9Sr=Ub@rnZOQ%+Gu
zXl88MRTS4{in4cMjm2e#3XN=GugLskFI&vGza6Ehg(FJ@Ktft`gD_c8g|dt<TutYi
z$#q3%GLyHND`@lz&THh)k$XzzIs$Wv#0RQ#`LiQarzaA;@w^Y~_K`fas#cfuD@Cze
z0QXg@avYKW?~3R1I=#L3b2xeaXFxRl)XjjJUQ0;3t7$A+Jw0e9{<(%jSW!KGv$s}I
zHYmIB>*<yIFeG_*k@8cF8YJFYv)$X19gf|iyT3LKRGgC>H7VV%8BAtXlZqP2!+D?v
zl&KU20R=WxK9=1yhIK_4k3(#-02y$9ICAdBPTTo3+&e~;n7^c)=5S&VfC67~4@J<g
z9Bo@aBE09=)TnzchalHW$l-E+M->M6KDAEioLW2e!|)>tq!I0>sius|#5R*LKVS_>
zC`yUUEvyk1lr#m$)5>0drtgO8z+7-3(&AMF!;9ITQsLkzB(`~HE62AKIk)a?=pdnY
z8=^;7V;~)QH`B8J*boE4J(K76-)T*l4i2BhKZ~4NFoA=m2{=^WhsLDJ^xqh8Pz&u!
zYK*sx2AjeC{)BPL%T#q!_ok}-egXmYCubH(3^QF-jz%z79l!O%<-iLwV-!(-XI;T`
zz>Qd>w8o1Ht*uS{OBOPPr$=Gmo?}VmXs&|!Z_Piiv>rif2AQC#6ktVH`~<%?fzkF2
zXxFK*(Z5R=1hqGtlHaof%WjVf2^{R1?OAKWX4iGi%JEr++LJLFFBZtA5dq5~wBC?J
z$(e(V{|9LC#CeQT!~iZKWl?=CW$1FLmW%PC2rDw^haZQCQ4nn<Hy$ZL5=Zd;!dy`S
zmttLs`DoSeDsR@q$eReUs()!%!AXljsa~_LNn>y3+1i;o04L!04jj8<=HdGm=$Hn(
zOQ|DxW@D^xZZ3Ni!@n&t+<0(i(}D*#WLq+OF5$2+QCmF1Xop{d9T`)D*E1Lm5sY|y
zWj5MOxGBFWU9XhkYHM$7@1CPvy<aok((f!y@eO2sR{4ajB~@MPeMR#(K}Ge>*-yTs
zaj_+We=}r%bLg6Cm2Z_9gQ5t8hkaDa7iK@xliu5gO#F$ZNAX-)lqx~jut7NYf~ruG
z6q^GQcR+$_Dg@yW%=%|<%#ia6)n~a%q7<%Nj&W?-J~dO|a+4907$4SaN@4k$c^u=B
zTDu;Hy_#vV#6J#irtucQS5vQ(q{5tjyS^b>2}PNcasq&EWti7JEuL&|*z9khpKlC4
zeodJ?z0uzH!tEYjk!MTng*Q7Trh6i6EL-$8fX;k=rV|W5h*h}MDB|C<UR^VMk?Kcj
zE@wFzp2|rNky>F3s!T^K2yaR`@!6K5afkSj<+T#zkbnG%1RI5jn8lvs#7&MvOP|a^
zrRowjI%F~Y(-}f91ct3FqgL^!p%LA-jw2?UYZ{%g2jVVLMbJuEufWX}%eI@5?_bDp
z3G)mx&DxqY^Q2UPm|r4*0?kND+DXMS?_;fm!T~EW;Ni`6<G}oGTd|L@J&kvDO_@Nv
zwSGxTh(mp?Aivp&T78R@LNdM9WagG@jKPfim)Jsc5m;H{sIIIuVt>4q;&m=PMhXt2
ze0%}UpOQsObbC%*!M5`!q!B?~bjf0&sw$5ksXAIEjX7z!1*<R@<*`-+B$Sce+;0ZH
z1oTx6hK{vNh{$2~w=GsQE(+JacGjC0xJE?nLn~KP0|JQ!W~g}8F{mV_EUQH7QX$0%
zlO>;qVTa{-ge1Qr#gsk4Fj0v9{i_8d2NJCB(Km3p-S<gtTjhWrGkLhc4b6wk!_9Af
zr!=FelT%Z%aogUUc<98DK`}=80uiJGpfjWhr+rU@6sKv73f^~E_&k=XY$qP6AJk}A
z&c#AXf(s;*m-B+{)JPLFB;eTl*senoWSM+ma-iR!fF&3ef?*zoW;uMoD^n7$Aq9#0
z!XlT=xDqu6OZn&fcKbMtARqV8%bvKY=fJD;Pf=@aj_^#L*7p8%=4^6ll}f$6gSx8g
z*(%YDH8Y+Wxh<w?h!ErQf|SO5HhFOmscB%GsLYRZ`1aDog<hm39Gk^H<uXp%!T>6D
z;CeYagRSk&o!&eC;r#U?H~0I()F+($*8n0=<C^fZ>C1P@`%!wm)6GT)iIN|T^+G7x
zx+)m_>FG)L;CvKrQ;rGjg_Lg1-!i{175RA;BPk6<RjIG1gHw|;-SG-ek<RefMxt|Y
zpCSfrcZ|MEK<aYjX7H+JK)}BF$cYQmb)wT{wfVhf04^z{G41>A)7uXb`R!&6)$~{K
zdI+VfG)<h<H-5oiJWxuq2|-C@2;~-S*$#z#?N7Eo15@nq%0@tN&wtM;4-0je9Q<p$
zA5U9Dwsiw_m<*9u#fx6z_<?Hm<_?^hL%XqdHhH4~Ya$l{3_sj%(!bcwv5wT<+Ig|W
zXLz@9AFtRW*{!G!DNOC|={v$CD_$XGD-M!_<Gw$q4|gd_D<->cJBpSX!}@64zw8OU
zX{;ps`?}B_U_eB}=zJ<t`WDa9T^xS&WcI|cicC|ma@NQYeaASk$QMdH>zA*2QjAh8
z<pL^f%A6<JTPqRNhNp-6t6R`iKsXN+QS<Lum&Yk-ix4*Q&;&e@>T+;uXla{fZrWzf
za|d78nZ=+4igbDC=tiw?KBzScqjIVUHx={CT?72u4-=Gg#@*IY&a@k0*bNdtwYFE_
z7)fXXYANFn1K$cX_Me97Sc-cW`w2wY5)*lhBeV5X(7#aXalXF0(4_*z!<d4&##|r?
zJ&lfufvuI~`{}Q*an>Z26{>e`6FQ(uVii{7{D|g!;mPa^5~R+Pe*=kGWAx}M3UFfA
zxjftnVxeR9NsA)FFk(4C;}Xj`z8w&Tr)Q>+jBiP=_k8MD_hSX0N<r~TDRl9Ska(_L
zeb;7z=b|cj<#xvt6Ra+6`1Dg#4ozK6^q9wxAqZwcNE1TQm&9<aK}$DgTZUjhb5mQ+
z(-P5Nhpkc4(TP48kXHO(hwX1ISdNk*_;cYzVl7)t1E|i>CgLcbPYD<JBd{&U`Umb}
zsc`FbnWNe&ipqnjXcL}<O%+eIj}alIB^VOQ^(KC3dtCC$y=*4d<P2D#j4gZB?cNr=
ziZgREJ!22<yNwdM??Kisb~f;wu(R{HJ<2d(oe*!#?3-#Xty1|->LMcGw~M6m5gZkg
z4h>y>odnmnuh^aKtNKqXVXi(`(3PLB(%RO9J6j(&S-E%^uv|4zRh;nve@+t~zOL?%
z2^~GO)$HNJdusQF@pmjn>VRXu4FB}F-y@QbCz9m?)Pgb!+Q33^ngZ*5vv_g_539((
zZYw}-`5S5fyCs<t-xTbzQDftjbf3b?((cli9nRid+xcl!J_TUD<n{LWx`z1}C3T`K
z=~?$r4jr8D2KWB?6x7UEKM0*F3|skEPbCD>tgZZjKkqD@^H0YmODErBt3m!g-%4<t
zO9WW&hmj(;r{JsYC+{oq=hL#P8hY?pF598Tg$V&|t7V9-A|A<58ViK+%xaqGd%f+o
zi&VaP)~<nPx$Y=E3Dys=F;O!~0_f5WbBm;QB%5ZAwOnd@SEqosj~?p0K*ZHtA{zAr
z3JGCUV?X^-DxT0F2x4NsqsbYo3rdxL%{$7s1ID_p8kWAr?x{jcu#EJfy=>}AL_hAx
zc68~ps1fgo(i(#-YRMN_*cvKokW-?JqJ_`uj<j1XdUL@<Eg0>gA*-_g@@vx=xRS==
z#jp)%B0k?PShcS*G}FNc8^(pDqM3tZ`{#D&r8{t&eM`2W39i_R38;Q8Lo_{dp^j=Y
z$zyBMRjg_wO8vg3wqmYPwn?5lKXDA#r5^yytdn`9fovW*s7rIEv`dztH>`eCw66Z<
zP}dxj`rfOdsvBt4FaJ5(tTA?sBtZ-iSn#fdPhQtRqVU62TqMoEfA6TkOwCO|PaH8d
zUTF)I@Xt+lxNf1@P4FlF$1i)xlRzX#(|yXW65bx_USaflO<7S!0|?zs*Gwyz{da;S
zHAhC&w(Vf0umqhi2<g!u2{P(?f!hdrC#pY#KuLfSuy(zEy?td=HAk=_r+!9vYpu^I
zj~#4svE*Q^b*#PocY1muZwb${0nx}Q0waFyc@4nRSyT@*U;W{Q`11$DNO`}4C=Y_8
zAhadPkF?feYh^oDNI{f0?&iw@gaQ@ZLyKM(6PY`Rv<qyuHo8OiLDRZ{mly0?<COTe
z58lg1Pg<*ig#{YzZ$#=_!M1{oW`8bH=Ta)?dP&MgifDC;nYJ~X+fSSE7j(f!6B7gi
z9vey!x-i~@6icuGVvY8gut%Ip8kS#UmvF!7Fj^D`luEk>EqFCMXneUDChC<Z`f=Ht
z(MhkAL$Ft~%P!D=UyCr-X77@-|M7alnWNteK=cXRI6(N;d0X*#-uz&*2zc7^Cj2@{
z`a-VK?GUpxM7~ZBUeqSv^?ccf?tFg5PBdIHSIE!7;7;qe5kd|xU9`m467eIaA0YwJ
zkv29P)$}v0)To0NE}_5dQe{ZD56pL_EN{^F{aReWfueEbAX%uaNr@9a054GtVZkn<
z_rO?!q+|FPjpJfwS#vVd5_1tjX=<@~hzMiBoDbIP^6D(XDZZ;60alsv_xrP8Q08-Z
z9tAfpk{Vn>2rn(Tje<P{t?~eucR885je4M#JD8!sw!;~^dIw&>y{`iZ^;a00J$~Di
zLU+b;R&iPXUacT9ohgCtU&`PDO*l0JDqn8ZD{durm}}#fieKP<)eyhp>puL4@Y{#T
z|1Qpd&LX$>9bacRCq-GiL;RmNxE)8vd;fhCaj$K;>D!o0)9U%5$|q7hnxI25^XdPN
zsL1*5xOWcD70E}3m3EGCtTfMYM9w`Q@oe^YdQs@R;uJajPkg-(rL{Et?>j{|Q!eSg
z8y4w<|39vVIe=~dndJY)R;&Yv%l{tKMY;!e{(qkBzaN6dTYM0DIpcq(uy~J(K&|Hf
gpDE^^@v*<yMegL8lQ&fR0N<CSsGLZpkiP$a0Te96W&i*H

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_optical_flow_002.png b/0.16/_images/sphx_glr_plot_optical_flow_002.png
new file mode 100644
index 0000000000000000000000000000000000000000..1e9a644669ae52d69a1a9c03602077f0bbb63cd2
GIT binary patch
literal 200828
zcma&NV{~Ot^!FL7Z>$@$WAny0M#ssoJGO1xb~?7ZW81cEr(-*r{O6f@W<Brb#aU;q
zI#v7BI;h%ve?L21QC<=m2_Fdp0s>iDN=z970?Gvf0x|>v?yKd2Y25qk#N#Bc=A>e4
z;^eCDU<@Ir?__6X>ttnaK<Z-b;An1Z!_LIb#Lhr!=Hz7O$ji)Z{l7af**cgqPl82{
zzl<Q-NvS(RK%naX&jDE~RALSR2>~H3Camg~dG_v}NXe7FiTD8_aQ3h!a4|eQ{P*V1
zpYhSw)<1t=Y#gMxXzBS~ei^xj^0Dda9qlO^HET^K@U5rWIHdPg+PM7sWt`=ngN}!u
zX77>d)3#QG>-P@*$Hd~$;?RVcIBlij+QcI@jqVBx3G5F)^{pLk5ZGtv&JcT0!~c8e
zMvZvP@R^VKCUUa-|7_&JcL$t(UHt!g`iSncR(t**1Cis0x-b5>o&V{|Ek+F8_P@;j
zKO?u${{I+4cTE6mSYB*){3w_KBV3o>-u=(a+!<6NSwl?fMn)2;;~iRgPWJ@=*X@jd
zo4+)L`)zErx$-z3j>jAQzq2SRudH-<xjQM_<-gebzb*LU8YO&Klj|*+aX>uTeq0Nc
zsM7674UtRG+82r;`C~z`NYTbAlLsBYOY8em!1Us^Qh$H=`I4dK^|Av*U=DE#+hpML
zB|0w3T8i%y%8Y$$(Dzo)2Ff!}%-OAf*rM_HZ7ji(yE1;Jd8f35>6i3PeZKe}y$U^9
z<0uG7mm|;Gd)P;@QrZVZP$bE)00~b>gL=f!+cGRO1wDU*5jikU+!4%kMot!r#1PmO
zO@M)RXO19=0G37uPabx?7X`ywA!xWj)(6zNPbP>0EYypQ@A3xnS%T4fppIq;d{_y+
zIN;u{hhR2yym}brPL}Q)%oe8Cz2+9YX*CaS@`=*09Z~Ez?s&mj)5!qbOjwE_y>&9y
zy;*$hDWGGiTi>JSy(d_3P`VR27S#|UdN_D&*0w?8?V9A1yW_(Ge`Hsy%{fxSFr=Iv
zbZkDjF(ovTs!aUGjjM>|ML)|Nfj{ogGkQn7&?EZ3E=y2_I|t09l(4cEA@o;n#Ox{j
zCnvZXwnY`~7*<Xo(G?f~)lr=L6j6yhFwyzQbl@91P5b>N+%)WOv<_7;5I_^MXSCJd
ze3M!a8+$A|*q`0VcxHRKSC5Nxlc$`fGc+UhF+oDTqPe<m>F_cB9R5GwfcE0aOQ%Aa
zs)P*BQwkIx*)r<CS(c6-MU)K!D7%90wdPkmMjoa~8i8lGK3+fKE4*ewC@CEOwoYR9
z7h>8c!($Jo*g+s+_oK}Gpa1?X(B(A>#19`(hg0OEMcI=roFIjl&Vy-NC#b#+#b8<@
zR)!w?^JM(q1NmJbzpQY6%N7RKREdtflmY(W8#;IwNegEac{Th5T7&5qFSkmx%IrSo
z=cK5;<!k+A3ZkT)J%NPn%7{zYQhC;f3K$j(NXA=5o23+lwh4#)KT)8lB$5A7CF^_7
zL6q%O&t;N>Xqz)sFIkd+kA!_MwbB@=K1;S7z`#ZPihC(-_(S^WIkRy#c6N3sJ?wsh
z$y-Wu>=5=q$Pj1T+<ty0G(!qm`dAEz80oREw<1i5E=E>b=;}=pH&%)I$>lc_I@vpX
zDhYdbGsTLtjNAyn*PI#*o1pyN`u*LNbW7(fD}T`$?TUCQL9!{d0K#P;55>YU6xK!j
z<l;0LCF+{xB%QTOqKJ1Ii~4wr3d$vIe}y>>3Th>g0`8<x^c6wTlwB3Zq8P%^9geBQ
zRM{>iI9{AV6DjDX^|&bGhvWsk-X2){a$s!hw6FOtn6RSGS6eR#j1bO>Pe!R>u9}ZN
z%Vxf^Z4xB`oCaHAi6H!8M_#;GFJ|S7y&A*O3;+0G8$~9apvp}<OaW_T-U4Z~Ee|14
z?46D0wNMAi*AdDJtU#B@OAI=Xb|!=zlFtz-l+_l@J}*$&SjB0{f+?)Kxob0qsSAI1
zgZp<ICz<6#e0H}gC{VdOeCvb;9N!{AX}c5vj-G-@)tsj8{CqyrA)r^Kl6ZPM^u*Qa
zX$Jecv(Ue<bU97~wYnLwkvl?D`k-3^YXzlesi6Nndm&W+^!K70xSo@@_FeWWmc0lC
zS&<RvXL?NOKP>1gZ1^OhoT$|^7j)>UU-Hq~5jfi>96j@^RZPY5tTBt<^Y*}a`~xN)
z?6a_8SFcT_-kZ+U4p|{Xw?GlG89pJ6|AAeC`*6s$z*6BttR{dR41v^zJ45p}83zoE
zfKy=kis8}fPz;-~*c5-k9ERK`hD&?$k%^=F(K1}gB*0WK0@*bJV_1118GaN4K$Y+!
zY+d!^@1Cwaulw$*i;)O_RcMooZEvUH!@cLMxYDqC3lH>J@Byt$c$&ElIV0#AiT$i!
zYmsm(%8){Kuf0}=4|67D0Xl7Y9qWI8mI`?$M<vws=>M~20S&i<WI7SRZoot4&o)ye
z0i)JA7S{q`U;b;s4MX+zI)n$z9Cx{^Dz7{H)u+%%p+d%x&rb>eIQylc5jRCgkY7pw
zF0HT>+8;kof|O7~HjKG_Jp{7@5<QDFh_i#94Ld@zWp2-r8+I@4Gt?XR2Ls8TMIs#(
zp=Xem2K<IldOM&gqHokGN;@x{12qW_uLDpT|IJX&V+EhrBPuKD_baA@ny;=T`WP2V
zbF|^L2P_ZN_Pu(04}bhY8@K;xAULtj25D|Y=+Q2{ZwE;o2rWUMKuNWT6a=OfCmfO@
zo8}chUJe*fn!r%!k&<p%>^3{*SQ}@oKqH8UYUi&mY_hTc^t1V_Io)SKATT-g7>|%f
zu7QJrtH8>zv}~>%8x7w>IJx63C>Lr1MAd^ijS-_^D&$6oi!x(J<n%%#&0v~e^H<u<
zZxvy7BlgD-tR)}?_*jpt8ZuqHH*ul%W13g6%GOK;F-7m`B#aDYH&KX>2c}I04|~{X
zMOhQ1iNlN?^cdc?RhKBPrhNt|Ae5C%s4Y~AK^@_PcZk^(W7ClCc<MRsaBM`{;}mBX
z4@VAK2Cr*NGe(R{E-2%D`pzWAPw!%}cpf1ieDrXYa`Y<fE*+fCsaQ$(!eAT|2=IzI
z?;SYfERECik^Q|En2YV-VeeZ2v0Zxlyal}_=|@1uzODDNuGyHEo9pv<N+a)B56no}
z^x%tDLe%&n+sHaF&0Zh_9<}ODlG1uT@c`+BuQ()U-E*2GAWSjGQz8}By_^MJl9T2b
zmn=~RgMoomYovKKe##}jE^=8h<msMrgiD6@#_rv3t4kHHn6NM+W>H);j4Tqxbt^j8
z_cuR~kh;#;Dr3H-HA?~vQ2N%9hqA=rfIOc1yBxv54s%M)uC$(f$)6gMvca-RQ|N^g
z_0OO`WVRS+zwWg5vYk^c=_k*sL)I_*+DSTKd!9Sbx~1ky5@Y$y@!^9Ah+l?+nKevJ
z53U$@Xtt`PsbOQ4;WE}(q%s*U3ibQLP7C$~w9FN_36W;jAk%A{3%Gcfw_7aPBKh!D
zqfMG00_z~RLFEP8t27(pJ2!d`pYM6Xa=>qQ`=o&!7%a8AL#3AwA{rMfjz!-_^J&Be
zkfH)vp=b$%&~r~}NJXRwhX$Pay}F`~o=MB`%9X1eA1AE|6La?<Dlnpcb_oBn?1)tw
zIU3Phx>OHfoMJ+PVIzuDQK^(cXwx1o!WbL#R$R)Uxr1oRGX4hU#gK%O(65(}3AM7k
zh7h}Gh<`U0VnF?tgaTy}st*&o=ZzjUUy3+ekWhXRIv2XjM$DA+L1G5refLrY3>>LY
z?D}<uE^z-_?fb_^vre`dzM~sjxx@^`BpZLmvGU{fGv3-OcX$H00pD;aed;vJI|*GZ
z$@|Vr&J*E!NxFfG>C2Yp6HTNpX$gd54u<YSBc;GKUGC={(N*#g@OH2>9O!J14NvjD
z6f(78)V7tR-~ZM`Q^Eh2a~y0^!wz&nYaaVJpXgN99s~y1nBjr80*s?Y5W*M3VMSx|
zl&`P(+alEo<mQnTd3$p0gZqJfFM;q$7XJZ|X0?k%WcbYf`D|b_u9Coz;X!beVGq>0
z(1n{@Un<i6SH_7CJ--TW8%{>?o53&Q)HuOvMj|AOGx(%Wq5M-#7^_SSyFBvmrqrot
z2CA<BZdfwDseBwR1fhq5ggGX){<YUQVHf|gU9yJ`a5~ncZ;y4Z*1fIKc(!YoYVIuW
zD5sX|tcTxX7ivCb-afLD+rw%O>Yo|e0agcYHBcP?ph9xK<1Yd}-J2QF>7_iE*Cm!h
zt2L5^{0;NYnC~C0jtS@z_G1zB5(2bL0M#?Gx)Ix^`Sv(ob8TyDYn9G#O)ixl&fL9&
z6MQdkjI%c6g;P#jPbUAHGZp<VrKP1n9wZ73p(&C|=ns(={)2%rb-S#kyJvJlxw?Gs
z9<#TzOGboi4Ko|xA&0fXwDMLQaW)s1rpH)N`d?Ix;R6>J?eq6P(Qv&DOlq0)kDES<
zj4Z2ahiHcWU8e>KPAp4siG~uM3?@Gn;TK1CLJsp?KB|4zwjB+QX0M@QDk*r5)2Q7m
zy`HhH<5KtW<K(+LpEjm*{u1`aIH$JeY)0`t+i9`f58|VV+wSr~#89~h<fGq0Xt^4(
zaRLc&-Rf&a%3la*o?UJR(yU3W$xwxcl#02E+Auw$q~qxm^?x%N9w_f=eFhtc5oO#k
zlXM+>qMI#qgT|~%N=~4It<JTO;@}{}gtLc^45vr6@6P&KJ-X(u7LRNzUmAo7EIk~w
zNV?AUno(rFdn;0Z#|YaCVgY#iNn4)6PX&k>%4K!-jjt1G5)d#D@yVB-nE)!-mdG(E
z>YZcC?mK(xOfh~4XB8;{1}RHvjBGtaWr}*o*U1Ot7ytY554WmRFq#!1?IMA5F((@b
z$uLP9zLX>8ba|2T9AA*kn0H3tCB;sQNZ}hLF`TNP?32H<Iz{)f64ZEs5vim<ADP&c
zk!{YDe0B`Xy0NOB|DO~8Ns2gTJ+Lp<dRVX>#cWzVvK*4#>;qf!K09|*AsmF!#ip3*
zIM&o}Tw1uHfSTwlZqo!6RSn7DIn=&hk_T;2^Cg<3g&QcTEm6rz9aR=tSExJhvh-B^
z<<_2xb?5NjXzTpQ%<nhzyNDvUlkHO*Y~3q-dZ3RfbLB|4v56faf(cpFubb;1#z9!e
zNrR)UGN+%v48J^#z@!-1pUVxnI4~M^2gyXeq`7kz%9ZowW0?9Ye6@^htJyRLP0JB^
zD4ZKa88yxtdHG=*=hk!wZ0xtU+XzuL###QY-ZsXyQ%>$-u+dy(l9u&cClUbjd*tnn
z>#9Bu=nuPWl(BIl*uSn=;O|xXNuAzPqhxs56izt_!Fb;;tj0Il1!v#cO^z*NR9_4z
zfBldQ0SW2OgB<j-?s$sX@`$0<iAo(4_zA_+8|SzTX4xgiQ*~imyRg_MI|X=S`>f40
zrhged;{-qRl*_8%N>Zyp@ge!I3$k!R{4=t+z!E6FB(*c^nFGE@!KFQW2JT3)CTyt2
z!Mi+H<D||)mjqkMg~T{4!CgFp)md2xpVNvWb4z~SL8AK2ri)pQCey_jqAodT!$+lf
zY;hlhU}|f}VtmH_Xe6qKCfDSaYUJp<dIrjrS>bU*uI3A|VCbK{@!l4du(_aUX5+AF
z*p>9gb$mUrWTg8Q9y_l!Vlmb+KDChdjrw;_jFV<5#<_XrF0T${?qH$^{cbI#cK%}?
zsU!bx@6mRzO*Tao!j7_Lgi_=Z&$z8teF68ntkhC^5@th$HfFU5hk0LHVjYx%J$;!n
zH_&7ZftT5Wcq}_|Gy7}jz7KJLQa$s$(X|KqjnJ#%<jkm@rO+tXJ&OTSC+@Ga(oFeB
zVv>ck;~zbz3cfc%M4599W-B&|aPSss@K!w4i36eU5X_1Cahx#=1TcvZWpi3-{(qA`
zP9Pg(hkU#{>gTHaYEi>#gtUaJ29_e>d}gL(kpy469*5BT_4^NB;7lidVJTEPd({$~
zu%MzO9$or*9~jgni%su}nS7%mlQ!Pd0~%!-CLbQSzWw;t&nrwiLnXXzFpXe{C#+A&
z8*}2)>cmHsPeY0%x=TqFL;ld%Xf|)bP*ILHgt+hmIFP}fa<-r4e{lIZ=c;C@yJ97g
z{%CEoaROq~{E_^E#PQIg1jl{v50V?=Mhe1){43dlb0MYH!pcibb-4$z)O-b>^Vq6;
zBy5fx*sEl<sU%ItM1!FBumQw8(x3G#ym<!f?u}c(3fO!W{F)LEd=EmDdx<JowIQBP
z^X5jG9I~tKH7h_fcd^af|KX6CeNr!oPit^Q>@8EEOk0^gGFGiNhx;w6#PF=}7?RHe
ze&7gF*~R8;x<+@!*U)bTLgiO=8PrSSZxL`9SZWSF3;Jmo>zEEgy~P>O2VuAE2y89_
zWJsY$=&|U`Jo?{-g0;~zOogL2niZ>lve1dOqOAG}>nkGYPZ><e7K`vo4T*#BQe~D1
zi#rcJP7e#a|Eivxi3e@?IaEZ*c+Vmh#3KaAhb}Dq$E4fdA3arB)7BR1M45fW#>W@v
zM7gH9)*N$|MYo&cl{Cxo$2r^gLA~>R<FeZB4LE2&J$%>A0yfL|sZmR{C5bnCYp$Xf
zUTyhL73o6NB45tmU}js;qB{o*fe;qcb-A8|YoY+kJJGK(pQ8l!_guBBE@c}JsbQ)A
zRMz|J+u%4wlGRZcIC8#FF!Qo6s!m`;oUnx2(zattYn{B<#z@^XoG|}N@?h#rH_;6!
zt{ZY#-L)*(6+Uo$#N2wT7;Pis&S5U=eG_z+_Va!kzZEbZ&E2p%v!te%YG~u34-Y2#
zjU9wly5FrmId6V{2_KY(71&R{XnUoj1K0hNR}OdorpN5C=J>Vy5^Fv>GG}8(W6zx!
zD!jj@bGN>S#uM(Z5~9d5QC;QHN>+Ko-AXMo5=J7ME?jw_J_XGZzEi6l-9kiz9a1#-
zb|g6R?a9^$AYykhuHN~mxn=ykjP#9uZ2pw9&}n&nGk^bC)#AT3I6um8*1_pb%i5-f
ziOCBTVJx<~X0Wvl8LXjtA6Krk#e>|y^k2-C+dvvLTAJ@lC#+_#Jujt~q45P!XmiKU
z{NOP={Mdjr27^VE))eg=I9ZeHP8J@Bbz^0${YWm4d&m1@2euw3KVnl~J<f+k|EOZ#
z5-QDcw79XFAd0d<s9b6>-WsOnPnHA%=;E-XS67Yx&*L>Qorf-pf}2P8F5y@xH|z-w
z(5zE2$H1ZgbZ&tGSuIf-g=|raBpP(NCfqy{MCve2Mh``$*q_ppzwrMaplP?ki8g`^
zqB|ZtHQN=Y9wW2(P}l{jEi<_`!!T>F7VuoOFSxL1!AcsG0ic*!uDlsKi83@phO1jI
z)TO%OWh^epmFKF(<gj=bl|Uu9X@sFZ=3gf+ZBGkG(u8vs=U0PMx<@1*9YVWjDhm45
zd^YbgX$~zx6qH%=OT5G`MD^&P*&cgBts!Blk=dHPTtmfl1!f%sZ^uAfoo8Nrn&<D+
zh;W`(sWTR(<zg?stA~Zv+f!yy%FUH??2ktWG!sst0=_O9&$BFwhWjjfc7$L8XJ%h-
zAVdswBr5$%8~H;lhh*_5O)rj#Ij`o?5!IYP9fVRnvIzzw>Kau<s(^4FNo)N0YLN87
zsbRh(d&?jXQ;#$|i+SJhvywnis(=09k<t96k36vs@r-4M!Xph{@}bhBi3%hi8=N&n
zYB({(NE=E`xt2-e#KtMBmFVGWRP^b|C3JVMSJ--uj)C1a=O7j@Jrh(Wo}r0Y*YqPK
z8x5RYnoZw_=kpm3#4zL)9WWwA$Z=7SzAU^|i*Ya6>rXWPhEzNynS`Vnc76f^?f2}<
zy<{yy^Qr$9%FyTHcnKEqvnUaUgn0##f|ZaIrKUuw<2MlL!Qy1PJdPoY@2^^jP=D$i
zlio0H9sF$w16h;CjQ?~h7?84VBZNz1Wyz;d$}exK9gQ9+G<%(}xn8axeqK#|(8|#F
zveXXc@R*a~XK2RTM5CRYaeI5nR>?_gQOc}9MeK#0B}KN&A`PP;-Lm+d(#(Fg8$A(u
zsJ@^=UxnOqGG;txPzCngG(WJPgMC3fNN>x13<BSI?l`O;2GJP{5E8ULE#ZRz(?#BQ
zcAA(54uzz82JlX>Jz`T8s!D?-5)XXYJgb{#>Fqu|FFo(dNmb;{e1G&03=gh`JLC5D
z6jTyISF!m3@`W@m(ri}5M2oC|cB7uv)_A>gT+`3zs`ob(9TF30cEuFYpTXC4DXpmW
zY^f>P*n9zLd}zmrGX-#Lzfl~<4N>Jyt^t*nJ+(3zS|@Q3yy2`GdC7UIe-!C{p)pA%
zE;~E<#i0}evg85bqcIl_<!xvc9#iDnDL?;88x{*!?B0&6C|v1&ocS7BSBVGdx%lKQ
z4J6SUn;e<gY3G0<!ZF3MBuYWWb0@PFTqpIKX35)bOjG=y*g8H3r%CkKURzBz>h+^a
zlyHGi6#nbsxB)uuQe<&4(6eMQ2EEmhGq3*IjY_7OQPBUv`$zO0XHt#(^Uh95LRwS7
z%7;R#?E#yV9&1ly6Dr1jpey1T-@lR@1Ah@#8pTxdqt2?j`H~nXDydM6=SrqiI2d-L
z#StvB1NJ@a9ra~DLhzdJ72U^AKKP}U3(`;s&pj;$!FhqkpBZn*9`S)5(SaBI?fms5
zZAvi_k}b4&T#<)W$k264<|wChttli34<|VJM*&5TD6x>H0m8r8kh2f$NUQb)t&7>g
z+~p-z6X`gwx$<$@?RqZ<MM)C+?+diaxk(Z-u@)8nzZ?|*1THG2hXn2+WN<-6i%$Rd
zlpSge@7(&ERlpJFN<5gQ5wNY(G;tGdY=Vp1V*kpd)HX3`lSmUYTgS@@v<nJE+m`BC
zKV#UC$E`pXu^c_X(Dl5>{umK@#pO6Wocox|4$5n&o_v&gbwbqBZVIbsfExw<s<Ow>
zjN_&(06QzCn(yo16JhhEFemFrzmi>u7AB5eQnqQM))Cua;8L&L?=-kJ62-L+yFXRd
z4l{664iXocn`+=eCQEMeWD?#x8R-49XefS1bZ0*<9Rke`jln^)rwUFxX7c<=h#nc3
z^11xRq0S{C&iN7_;D2J=%tB1%T7cy8^zwn^TOYO8H;S8>K`U1im}`v>=0?K8@g~9U
z%!$5c*ld1X0o$QvP<g_2_nbb1+^6kHOUEEvK8k#x<t|ZDy&q4|7h|>IiucsegUO_^
z4C@E)W#&!)2I+^Eu0iVC2`mXJ{PsQBmJ&8JS+vOolNLb^5&IuX1@>PkBL~KTX2uzr
zZpXjr>fav{y)_^=YRVa8#X-b4v7CIsGJ<)bs?PS7WEgVzL*kf*Vu+5)OmlVys(hmz
zyx|y)NO}-GJ%W_dA~F4E-o2F<%d>x{4If?)4yJ4;dneEoHtjt^fPcZzM67lT+fzf?
zzjwoj5aBf)x1tpcrJ&o#{>xkpyhBbIvun<0Bf%~~*gp51n)^*&*zX)OD1I%hXQRcB
zg7+Ai+_(S45@w;rdouZH#i+j6fWxZSC;ctLKS{6sxz6=3Sbr11cfoUe5`$zixRI$C
z1O%DB(ufnPj|ni0GC|oU7CAO63q*{J?3iK|7=#Dd!Vl5??mLelg8rk&QH}rj3(FUy
znj0~f0D&5YoX>)R{lXZtg1FRBstNc)fVv+6&W)x+64qwcfy_O9m)Iti>P>a|8+|F$
z_KgdXm8yriWzjTC%Iw@aJ_s`VNUyNkL=C)48f5v3AY|AA=v9LsQVzKMQ5ZxyOBpUn
z9iga^Z>I+GiE62^=9QKur_{70c$N2dJ15ZHQ}~z>>Sj(><8cX~S#<kOGWB>eh|Y^t
z*2*L@dgsOqX*8ZC|0*5sENsOsZDwd!*wZR5J=KB+72WOp9lJ}zEn%QK3fgXt#T@$j
z*p_Y-rR1_DBhOTEwaPHTDa1!%_5JYjaTkk!tM8>bqo0G<%a<Iig=Tu?dN>y6^Qkc%
zOPCzh(1{JhAGVHH(k6xSsA}>##f)bEpq30x(!7MnKC|tvNFXRBUiA(LC4h1jC!l#S
z4LSMdUh8VNb)BHn&-i<t@_d??SiHdFsp&BHpAQhv8bQ#*vM2XW;di=NG1q`ej(Ffc
z4IDE`2||6e$>M`9fwMzTKLlHB-E96S#D7yN>C33gdj05Xum`oD9>4WQp->hSm^Z^B
zFh%arL`vok&YLz3TwPK9Kru5CN|j0Ry*t81B7YlxV8NML)BHE8`tLuw1^mumea`%a
zNuWx;?aD!CA{pZxn3UW#G5}VVOaI{@m-MvJdk?QYh8_z@S^4^hs#Xpt9J1{|7{Tr1
z`N;idgjyQFz`#`3(H6P)>Xt1@e+(5@3iGR(RJmlNmsdu@=r0Y!a-wo>w8$x?F($0c
zN(CusHo--CyX8VrJDR3Xbe#aV@mGa;+q}jjSX)pwFFRa?Fv1b6Q!>Mc;wFzJv@}WW
zECUFIPVy#daSYGnX06+bm4^f0O+Rs#Jv)<>-Nm`5QBHW@uplZFzUk36G7w+I+#OEi
zk@AN*6HxSBM&Z4X>*+me`LJp#_;MuzXUQD8>YPH-I?YdWOTj_MM4ZnD3HIAI>w&m}
z`+^e#C(#)3udeBC1p@%HOrsFHs|GTHL5i~3GXD+AqdH=kMXH7PfsrAip~17y0{B@t
zrw6sxosE)#+)5JeCA{ztQz3a0{)cXZSdHPrPx7$L3ni-!h&}4_YL~qi<Vx+1ke^5=
zAW$VG0#tyF-yUeupA{~O^FKs$w3tZFUr3o)%ZOww_(?xOF~)CNQxq}=s^z7>e}M1_
z$-;V3gy#o<qK>Yy%4H%O%z0Yy1(@Gv3v)Xf2JRJ@>#FOcsA1`1uC;{cQ51z!X2$_B
zXi{RaHSDWuB_nVwC<y*el*+sykpxN?qP)FUV1beZ`D`M4RNnN)h_D{x6i6V-$RkPF
zJcA`Hvn^y3T{?8$JGC94MS!^(=ELWureZBvwdbUctJ^0v4rSnJrTv+;Tvg(Cz~u3Z
zScEN?uabVML>b`*i99WIN<%+g1(sO+`LI$5$&P=6Gd=(aj<^fL+o%E>I*>Aac1TGq
zs6Ox!xqZWHgLYs9_Tiw^6%{|da<JuF3jz2z1?=!kPRvZuQ1wNDWrkYxAX)r^jV?mK
z&&YkH@wTsD0|&Bnr@=v-jTVB?<mqNBa=GiQ={+brtr-6JW$DG9{jeP9hVpN`@<6NQ
z?&65s6e9HlhW*sm!A;R9?M(aPGeyDi$rGnj>+db|hwUc@wx(piJw`zSOc7>23waFr
z*fy<PLjRW>uEZ)s%YmP~Y$r{KIaIUJk#2@TggJz+L^yLEV&3f6k~nmw9a{T~{Dx9z
zf7aX;4M>QnY{5Yq*IXIO7GjyBQ`FEDee?<=-2YsXBLYHsyx26f@yR~eqqftEl_@gF
zq9%|lQ<YH{i&ZX?DEirV>+LxkZumpeQqBlV<6D%ybdtz-$SfCl-Feg@KRHO4khIn}
zQ!1*gGLi*WolrzRCZ8RfIo4~B;)8(GM7G>lZg8CCwRh(<LYOEYPZ_6)vgv%F{!Pn;
zLRd%=Nj%mKnmc(3X_-><$q#D7tKmGzV;GJ@O50!;uY?t2sdsCnigVYBv&r}QaCh6I
za``R-J|47|%lDu;meoq4MpTR0t2^OUmUP_RE3j1(A<pOBx}X3*nwt^Kh1kt=7tx7X
z&MRkTiV~Wus!l9izn}~cmSKuFv%%KQ;~-==Ih;*W%-Q3%bI8q%>DnN>-<@D2#D<-Q
zxI01TBDEPDe7Z11+=pevT$WL;lU7PHCbg$u%_ks@Ht~a1yYN8Rhg5DLHEE`93)6Tb
zpC&mM^K{R~(mR?UM_Yml(JW9IvFY+F#N;~$<!V({JKO-{F}~KNNATZ0seG1P0_y(o
z?YSwN1R8NH$+NSD=d-%ge(9hi_YMwEW76+m$zyk>>`P5w=j=%3bej^<7Ogx%_Q=cU
z-`Ytbiw|%MCD|u|M~IL^lnL;=X?K4;l6Ky%*sM1tLzuxpalP>vQx9LgvvLkgp|nXI
zshS+6u~cdHq^b{BzOYr@z>j7?2xW{~hdVObR{%_S!diKImgSf^7AO$Be_RNEf%I~1
z4_7Mk+ZN_u1_UaRa`A|E&$%=kdrURPUH~@z6x?h-BPxUW+E8Mu|1(<{$H-ld|9;&=
zynh=p&3v<cSQNAvhA4(8mN4P{*0Ytm_yhNq-TQ~5Z(8d55ewm1hbpvwq}ZMC5m{kr
zv>-2t4LR~ZuTLeBqRtB+?fsH}uO-#*Wu`Sz^eZD<9p;<Q=A))EgE53_$ys9>Fq=1*
zft+%3qs!6op=p8+U9Y|ci?`7xI}gyqJ9f`gy`C+%#&c-F3{oGfrFS7+xbGLyXizT{
zRmZ5{=i=E&&A87@nXOwmI3Kdy1#{M}`k)eQdo=s|y0M@QloI*uZdbjgm7S52>_a}G
zH*br&TfZJk)*59XAls0LQgW1+@B&U-Ye~2)j53@N8dqFQ$!w*Mh8!TeYXfdupo%cW
zNRI1A%(8($p{wVzvbD)q4ZsSpSTVI}^w+Hf2+Z`2fm=}h^UWc5Rn4QDxp`$-3pk@B
zBJxf5ezOKnnFy!nHf1A^?MqRF?*W&br4oWbF|;XTK+<sRm?Yq=id<BWeU=W&c`ZCU
zi@dqv6rGJL@0pGD&1ReJ?Kt}s@^4p@QF&ieFq0wXt%460v}m!T-WgGX(&JzQQr+d)
z2SN@ieD&u$^LURigl)ihjQ@e*a@@OpXz0#aI$o=l!+Q>1^Y1>34o|ofcPZ^!E4^vR
zj&IUT=w#jYJ@?)5D`N(i(kt*LBCA3}a`5CsACtDlGGFQlCGBge@ON*nS+xhHeWD6e
zzMuvuT2=uFWtKS*xKrzGw!WnA0=46FBf}}e2&of!kVh+r?V<xeg<$5jsp@FQ(vnv%
zRGT0;*R|A(SWi0be3%K!t3ZvP6-zKu7iJo0?~PU(&31HHW!x1RLonuxqPvC5l*MF{
z6$JthfqvU?ERS^>1+H(`oqx;8!Ir0ic4rF#MZ|-b|Jm*Ip{jY>rC3yqceJ^p#q>S;
zUZT2RCohE+1!~e7MhyZ<w(P^mPd`y0jfZL~iqDkjQqDY@Gv{`6d_o77FSdYzGpjyy
z!HY=AW^I8#>60SL<}F;_H=JU<?9}g=u>$T3WRAMM5P%A?fDKIXEMK^X7ih_!*sp6E
zz9Q3(0#KwJB4+6mI2$%!0!GxeHg6V}>rY$ds-G+Cs|Lo;z-(OuAAQJ^m-lB$=B9Mg
z<5`=gvIUDIe%up;sF`tQrtManQT_}AG2T1XG{Ru^FeMguknBu{{rRO?2^1>xK$Ib*
z5s{=E3LvET_-02O>+sO;4KR8Ikrhh?VGa=ajx&F5EsXUbiUhAk!p}_LbUYzbyVVgT
zW5<8x_3a-aDj_@tWpNXS^c`EFXO)Qsl0;$!`?-J1czL-&m|>KyGYQ9;kk*XCY9P)*
zerY30rl8kJKfntIdb+1f)c$Q3RA8ZS7I=#KiX>+mG4U}QvSB*?0tPr-3STmRWB1Zh
z!!dQkgzmUX3ttGq<2Vsv9UiNEyb*BfIQA5r1P!9VrS<fR;CFH)HRZn{hhBQ8y#KjK
zQ9$CT6{I<bjU2ar4oP@N)?$`f1F8aN*H;h!4UyD!bcRq9722h0@=lVx1^N!(T+dO4
zBn`r*trP^4%43ZBASP3~#YA$YE00{-Isla_1MzC6zK4!7GKOoV>I_PD0^p%awXz?y
za}<X-YIT?NOq)r}uzR8djX+*CD1UWvwN3*&+gs-B30t#J&JUTE1Uts-rUb_x7OC3S
zP@7*)yeoA-qlyN;o0uGde0CY8y6)(bM`f8&Wd_!}>PCYjNv=l*BHM57f4T1rq)7F~
z*|xth%IEM}F6k5t;=KN@um6iJS$ATbw9ogj*`^HMF<xLB$B!QUPybv_h7B;HB=Fp0
zLglE~U4}&xpH6`lXns!S!p%SrN4x9TQ|0%|V9PTxx^o4F2K9xoYRYCfs)E=jWiDy%
zV}dOLW${a!Q;8B#R3ZzU2s|aZb#_8q_XN={@=Md#yRYGXl1-1PDwsR$454zux-$&B
zE;1kb$c_rg-wYR|^`4wQ#!)K_V=YH>m!J&Nnk<G0FAA4PV?8$#G)>yhKbi|~(8Q&L
zk9XVncjn)P%rVYp!8<@NodmW34gu`UwYUz7J8Iw*Z*DYa-uCw~U9=|6MgHyP_M1`e
zr;E#X$b~c>3<|S`wSN2Y-TbAw3`)exBw)prHfjNfK9+LXPgL0^DHOU#a<EIB59=Kd
zPbmSMqyf#o8*koaTeX?kk~xqE>md>|9$BZ1rYI~Q!GGr>OVi56=Bj^Xj?!MNWW&Vz
zTu_4=MmXdA+&O9{lgBe##u$f;Qku~2dRNlx=|c1D%qrmLdh-YEQl-w}+SUg#bH^JJ
zLTJBP0*G81<Ie(}y5(dJC3`b%-<jnnDn+np5#Aqj6+4Bf^jMOLC_~r?x1>Peuq318
zYC<eRAA%pKUN%ccp!{1mbKiJAtcf}=9b96%Lc_%`F28C?g56{W!_aC5ZQqNuKPx#W
z?ZT}?ZYG%bo3xlb;srVTKj;tr1nnIJ;J0x5a8b%U)4xa08O-O=dI|rVGb3P}%i<SC
zp(KJhy8&C15)ZUlN@_FDKC+Cj>kDr*bKI7%u#l$0OrI~MdDFTHT4^*&15SX|TH)UA
zv<dlMl=l6TgRRaAyL<?Q?J`+azPD{j7IpQ!+Ch7>W*i88abHrbhs<|N8|w>BdqViB
z+bgZAC6mD-0Xo|PwRGvdM;~^iYPrfiJxnq5Cq7{5LC+rgP9jCTzGP4K(M~II%-+@8
zt?tgt{onO2*I3G$wld2DI<wTAjDih|>y?sUUqLXOKWU7}#d>?Va)d-d&%+`Ubxe+t
z^F39?1C##p&ci>V^mBojEPn9<)*qdNAzQ&dn{bS9sQ2x%p#&k8LcWYux1`hG2We9k
zBW~X9fM%qjm_Y)g^(8WYSZeHfaDYaZxbXmnKY^=&Bs%XbjTQG`m-F;yD|DU3MM7LH
zlD%ZcxlrRpS+2FogZ{90v?;**8vV6u7arYbp!#A#^85JA5p!ZgK_`!+S9?A;m+OWs
zhKp}1`ucwA9Z7|3i#z#Lt-2f8EdfRCr<jNK10?C4|75e_&|sSSP(>)gfERML8tl<L
zPC(1JlkAI+v7ek#<mhAQWlwoIAr=3wh1$zY#HBIcH~8#$A4fbcsTlaU0b{&iNKZb^
zq#wkU^}Lqsb@N9!_HP(UZE`ElXG%t93H{q8@?pu!{W2;Q|0#_!#kdMIu}st$vrzda
z4sN-=qq!fC^@#egx@8zuu1swx)yPZI>H~L+2&~R{1|k*~*1WEIJj=_=uX=BpuXZFa
zbEC--DOS(@WQs;-N0II|vVJ<EcB$rRmEEj$y^H$1|IOL+NN_NZPD@kA!AWVh+kwi_
zX&DTBedKZdK>R{w<tl!jCJ*^sV&g$D*-V8dD>Z1rj7N~824U{)#c*^)NtqqnqcXw^
zYy~x+_@NcBy`Z@?Q(*WD|AFZx{Z;DnvwVy*>WBhat%+8?)gIF0bQ8j^8erXOPp<ic
zdX~$Gtz#c8Lk|dr5P$6wxQ#+ZdwbF=Pa*l&7QrF9>_xQ55~+)A`p+Y0+b)i-kxCjD
zL1N=sU=td$MK+bf9AsO-Muo$16I!oL`Sc?H?r`VIPZ{$by+NJ>$h>+z#fQ8(y;St_
zHINBYqSCnfj%|j~1@m0Wx0H7rgX%1!7Mwl^r)oma#~Tv&t5jRZkWDyD9xPwrHu4x{
zZe?kEK<<zE^Ibj+f4CjRvZd=W>?pKxMuyzPR7$0!QEbAY@8K0){6o%jDvX3$IUjri
z$khVdEht`-ckG|}M5;f%wO5*?!RfG#9lEG352s@Y{~&m?<VWXQeg`u?17SQp+uSPm
z=cbr2#cMY9O<}U+!%qLZdPmauBINQnQ-vW=7m2HvtQ@|Gq=yF=b?uIU!$YbAhu|u&
z1N!!>-bk<8wJiht=jCDPFk{YOffHy=Qx{V2r!R`C=914WH`ikjKYegTwxmn^5`Eyw
zE;K%#)G<AYN~9}Rs#pvpk43s`wtXFSn*Yuw<ADIqNB5h|BHRo$F|UP1E<htpwdxh5
zD<6L6ixb%~L>UuqxGRO&-e-G|;{g{(QzLa|^tir&lS4;N3ya>4E%kX-%B8e%5cs5P
zNGI<Gt;xWa!I8NeHapMX+cl5@l;OShKR}g<gAFeXEN8Zr!&U!bVvGY7Q(Qx|U~VER
zANw*0Q!xY@ze77h+rx<R>Mt8nIkx4%sdqTx=SUts$F!8J4c5c!KiV#%6r(y*o7-YS
zs?jjuqKOGh0iEVn3MaG-ZBT|BrDt#K`Gu;`mL5Hg)51r5h!ctkUIC&x-Y2ffY{5?6
z<tD(15aQGU^x36H7QMSMhA;TH-#$ngvGUaJagr#3vUu*o;om-ORJ@hVnf>F1_MZi$
z68X%X2HTFovN=<A9iH&L<ha=$pj>8uCf_5><K;$?Mn!%@&#hMiB^iQ<&yj0a&rOZW
zw4)-w848zZ=V>h@#i?(M2$g`M9w|KeW5=;ie0I-;u@S~A@t7=#BZ?+ga`C1ZSVe4N
zMnCGC{hf%9`^VlGs6I;T<Ikyn;WyfXVmqr?tJ<=ir<V~mfw9$aR@vF()87MlA+hj9
zZ^j9Nt-kwLCg7f@+F{W{h6_7irOX_2N1hk02-y%z=NpYoe&#RK0<G4qB7-%ZCCcpL
z93xmeN-1gVdwsgTsVDr@J&*GXVT_X}R>T4Y9Hh#oLKdM>R|Hr)*4PkGo#<V?pdfii
z$Mxxs20lAo{@n0+fK;y<hhyrt`-|bu6*CUgF7)obU=QI(fNfGP572KCZJML<B&uim
zVgQU{03-Oz^?FuGVOrXNSdYM^TjSqLLhIY1yl4JZp}9&_apu|1tDsmq=EUqcvZ6vQ
zCdbDMqSg(jiA7=(pF1)oASqGo*5|v>ZeS7W)11(&?(*tpk28$rsio8j8+TWLZlxhi
zy~{(h`#XM?OJtqtlrmj1P0CpN{i3E*o3?y5|G;%(3Hl19xmXpon9JSrb{+p&jhT9a
zz}SlDKYUIG5`;ITi`WQHpgZtlzxqS~$-5vs@6n5bPDu|;y#k(&J^G~tg5FHM8DtY}
zjWz|Uq1N}AXNyLatus@YWoQv&QyYhL%D!v@LWs8lkZuK)oC(iWP!An=H@>ti`XVtT
zVM1{MOu3S^gH_X<8H)=vCq9mU|9z=Im!4nP#P5_8HJ$XK`y=$;fVOtqwss#fnEk~g
z<gH8p);q?r<w~XJb5<*J+IueEDE-F>*_OT0LS}c9%4Hk2%u2%1cMzKTT^o%;L1i?&
zGgN)vngu#6WZVm5zy858Viel@Mq_}td*)$jnSPD^t(?0Xra$2-VS@BlxQ%_f$9~6@
zN-eLqY9%j90sGjLu!o0<f;CI{Zc9ndj;rqz-`mSlm!7L>hU=y~!B9LXCbx!VtW1Sk
zJyo~+am=j}-cJUjUdd~(;n$3GZx&zE^YfH%Plu=F`1tr$Ti2_n*Zb*K_x*ZjWMVJB
zDJHM-R=6beF&iyML3F&5N+du9Ye8nFnyn`IIBcJ$d!u@!5kV6Bo=x&QCgfEf1XLi$
zYBq=d9XQnBP3GUK2~Di~p|x3dD)Z20YwO)ExDE>dg1V$7PR%OBzjdEvhmo9?k!Tt!
zNF9~UxG9+9n@Nmfn3POj>Pxn@NKB=hoHrdXk<|vme3iE-S1qOG<P^-ERCjg`PCih@
zRo6AdpbcK&FX5A6{SxdLKB+rO%hJqOt_WQ2>T7uP9ByPon4R4^`JvZ!^pAE5=ArlT
z64^I~ZRxNN*>`G)ah8wnt+Tm!D=OlKg^@mGYG~UWW8iu1h)aaJ-o-^pDWHCXH0FrC
zse1Wx_$Eu!)`I`v3OkAsXzE$FrOvAkBsxVq2YTlv{!EaCb|Qu|lf!r<{e7oY?5#%h
zDqPxaCRcQJd%<Trx4KqW68A)W+4)MPp&mLTJbf_P(dM<Mxph!hk|U01?Y*?o(I-jL
z`D5kLtRX|I#Vaj3nn>fEi@mO{{cWL&wW~U)@u6IpoI>cS_>m+n^(16UXL%h#2qb`)
zM9z}Z<%5>PzmOPfz>ivB*}Grd!G#CRWSYq5fV(rL77CkZl>Vn7*7Duy1M7GGlPDs}
zk#DNf=;x%J?xjSh=qWm?0R&n^dq6mniLxJ;fMRA4E#qV$5Z$KsZcv;d^NmR2Gj(Kt
zMv6o*c0w`_i^!cz2W4Ox&v1Elb+;!FUaQr~-=RgeV21SA1t9=pE2R|ll`%{5BB6!-
zikRV1FTS#x{O{|Ls&vt0ek=&pvPio(REt*DI*pUek<`J}okY<RQs&$=BT@IXa00=4
zY`7+>PdjfYp#;BN^yq)?OZX|}+FW_E4hYdA5W)lNrq05OLE+{X=hK?><lVPFR8Q}*
z!~%^+YMsOTh{PT|^_Vg?iT)(!?P9ceinhQ4BnnSaqqQH|u4F^|L9qb>o=G0;8+a=B
zrL!X>8$}H!o}Q6|vVks+u@x1_8WmJi{O<svkL%d*T?GGJohrkqdqf;3-t|^@qN;tz
zw|=}YT1^5mhB0T~QdekVjJa#n^Rly#+2{Jl;h_ml$`2ZBOlqVl=JUT(1x5!{Oo6PA
zAwN10MCClrOK;iADnqS#B$X%<1A4ypUY33|39|Y|L$O95dGKy?qiF^y^y{`t%s<X&
zh4^oL>#-#pMyS)swAiWh@z-IpQDfU;vzL~VbuO=2ZmD18C*3=tDgH*C<*n4P3#bKO
z_%OyYWMPT0RI}p|*M|Jxp}6O4+a7Fr-ea}8@I4TveedQZSA#{g#YkU#zK|gX)U0+5
z7f~cB8<o6~Fne_hiiz?vwmC|`!~}9ajghf2-bsqbopJU6J(L3581z{icgzG3wLd~w
zb!*?kBgfnOnQijo-anpk%mBh&QLtG5y*Kwq+XAH~2SZS<R)zAP`PuQJ?|gAn?b*H0
z^Z{6w-~3rZKr!Z=8B)-_@lQo;MGhma!z1K(yu{ylH9h5++Fyw?P#3SfVjYii_UF^)
z=8q0pMK6p|5<+3cF#Vq>;I#KvmpSZj*%0+cf9tGw`NqZyd67&xf<EVYY^pZ%HZBb6
z$yt$h5MqclNsnRH+Z-b;mjnkOgr2gtOnrR9si$)G-dak`TzNS%_}1H<k%#=VvKTj?
z?x)N-Oh7(23?=e8d@Qk&)&WG*S+5C}A(%s_UEc>tE!G?C+2{xi_logoBVMKQZj|zd
z$eiIvC`S;Bj;%A%XsaEylI}tlmkzMf1N1ck3(myrc!4+}OptkDaU#l`|7j7zowZOc
z%@KsPRlB0@qBLWdj)R^3J45}UEG6t|2^(GD{WC}2HTmVzWKjlYu_bt$Ln2oyqzQKQ
zOLWxN1NX0-+2zV<{j;>;lUeO#23C|o^~(K02#88B*{9^eqD=(FRv9o^XkT=r3wIgk
zA<Zv*S_}G3N@3LDgHH$j<2KDQRXQg=xkj$VFCQr6PDWsGYJV5`82;5EMt`q%>vhb4
z<Cx&XrY+_J4$*_p?)C(WubAi&g`~eXdeSvV39}=f9@>Rym$bW3^F5k%oQYvO>CNd!
zuNi!+I86FvX9;#G=0Nf{+~*VBPtMcFONZTM){FIyp|U#Dr$C_>?@0k~;u2M9l}k$`
zUdE4&&(qJ5p-b<u%T~WHXP`trYajLFIM(fCVTomO{-d!ZL3*MxF~obhzIcIi_$@sg
z$HUa#)b!}-Z1%fgheL43A(IktS1Qz^VSHq+1fegCugSKqp(S!A{foS7bmiiC?N_hV
z?dX4PlS@$B;bHnxM-Dq4w{{+~_z)<T*VcCKb3TIhhGK7HKSnUk;&0D5sbUL$6A*?I
zDsm+{eZmK)xgg*r7ePA%X*4!D3Ec^wGTjvM#4n@%u~RMZiQgEGX#qVGU<PgA#uzXw
zW7!h!N^gA{9g!YaNTUQE8<u{~P0CwNc)Q0ctbgOz%a8)HB)GGVIr2{={v$+@zYJRj
z{56L?{gr!9DyNC$LaAy3pr1k<t%K)Z_mf2-jVRge%eCR9y3eCR4vl%UQ%RPz#5`8v
z8ThLd`xC`DB6=@Op-M}bHh3}p`PscN2}U@@1`X{LCaN%#w|HSS8$064tns1Q{#IqQ
z0iT+V6Glg7Ib4u!nBC>Mt6mDG<&u`=);o41%lfPQj(#ZS(JyZy5uUAJY0h|R`Ll<V
za}zq#aQ9BFxiiAbTg}IX`ev$4LGnS9ZJiy*ziC|kulMA;e5!Txx>N1r%H^M06I>VG
zubg{FR#9(_;DDBJBYGY@v}CgWZ{{=FFm<)${<kOQkBF}4Enl&wj!?FU!waV9LBK?X
zoQ?Kd+;i<%FXj=t4PKzLyMFyHu8?OI_m*3v&EIcuT1$CnZDtLl3chy~hku;qg+BQu
zcnfZw`vI$0x8x6(i<xRe6b6rQX%+7o->J}{DZ{_$)kasBV>n1*AgO=CQEg4l<oGxt
zbLSHV>cy6+t!qlwhCn{(E4c)@#>Y2LfZTaMvDJIae<9Qy>O`3B{-RFI?*?HepTj@0
zyGwri?c0eb8?iDVu$8%2l`kB??pjR(7O<L#WeXt!viQnHb$ORYMM)j8yp+Ip7xaDu
zmj)5TQbSO%=pZIGT>(yi*kV~{50fK8hKIQ?d8S0}Lir=GG#klBwFOzF61B?re*1?<
zXvAE}TJ4IfaoXa9oSWB1-|Pz+BVbFzyx(KeX^ef~HTv85!S}o)H24pd0IjnFx=FJf
z;mtX4#d{zA?yK|)JVVSiv%~$KxAjMKB+oBvF+D8}FTf_{r(-OobZSOMA5au62;wO_
zmJ5H$o@F&Ri7Lu~f9d;Kh*f2U64Sy}J6)aQt<;WMX*^1t`_K>so3%l2;gN_;$&t4x
zf_TwBEBBn9y?xI`Ic8i(J6()jVZR=W#_m(m2RpW`x+kfurDt5nC2Dn8NAA1t>=+Lf
zlX0U_WpLW&d*|F?KHFrEP)46lwU14-5@_~Bz;FvyH<n`Tz}i<7Uk_)FNe)8kE7n*$
z-(df$zN_V*wgy;iF?ZC;!0Q;sB5n6IZ5>p7ya~<Oa7ougnM<X6QrgQ7Xi`~R)V6p8
zrag<VyH}-jAHKuC-P!uc&Lf94h**l2V(Am251<<ZE<gU2j59GR3c2Lxt`Clw$>jZ#
zfc7pqS;}IiR1%v$K^s87{qaNUBtKbRZZz?5$IM98bF?ajFG;5yuS1o|(dq!?IYG6y
zrb!EJrsi2aFgVC`l<gKkieT>AN-449NsNF?M)bAg93)rK|CG<%PCLprHMFd2s{1%G
z%{DN0$D<X{T55vc>E<YPGrcW<_$MATS6JSErRK10%l_n<LueAfBC{p@pKoe$jjy0*
zH30j#FxCy?nw#yd6EsMwDjS<F>Ni8Pr0_cPZXWfw1d@~-kg|_0h?XK1)VrOJUUn}G
ziT|DBG9%V}7|Aq|{a8wmBZzTXhiKG}i9!jNtEw-&wflO4bd|yF@XM#`KiDbzEF2{{
zv}OS2p;VnYo;ji>J}Jo_q9m00XC);lm;Fy_{71=#L|G6hqS^DY(%E*~od$aj5qce{
z^-I}Sb=i^&u{YJCX!PAKYhpq`@bye(GFxDXe(MQk))~Cqz^@!ZJQqCk#XD8c6{|j+
zxbU8xUDUMK2PAz_dv*e-Q}(neGD$y-S<+n`Q`%fFYaFbC|Ndf{GjB5Ea#dH;*gmN0
zM4A;ZgPC$RcU>K4Ue0R$iU%oUO8@(z)pCP)YKi)Vh2YYp{AFNZSaXC_p^`hi{^h|5
z5Ro~+I5$c3=s=2r?}apU&6J5pc`|o5phO5<d;o~V6OnE>ev<uPES+UkRPEQsrCVuf
zN$KwHR$4$Bk<JkqknZk~5|EY$hVF6bknR{l8iW~Iy593&>;1}?nKj&JpS`dByRKVD
z-9h_QuC5U}ezhQxhj0cY<y^h?G2{bolF_lYT(oiDrUVyp{6~BU?5!@vza}*!fz~#w
zo-qn(oA-;o64FnjZyqtO7Z=q&Arwb0XaXk7<_|pX0d=9~Q@M%VZNIb~H37hhy83SK
z%Q87Ng9shPm(5-HF;&$gG7M<mx3((>?)u{X8Wh=m(0>)ki@Lht=#Um6ln1?%GCSC~
zr^}N0nw)rf->>`o_Ck9*>p$#+x&Wlmu0~>g$Qib6=>+N6ZaEVySv|CXAJY{8AF<eJ
zBM7;8cnFmlcE996j%b&UUK-V!4_upNOuL+jfXbNFiJdNS%k**v25iA#pMxj{P;ndR
z5fen?dAyn~mNZIYKe^&lMktUt;ard^g**KAxFiH_;P(YoQ81bp+P*Ef8d`_(ZnX40
zk{;~rf+x-_DgFKFXH%Sk_LvrV?2DpA=?mdj^g-;=rOO4c*JU3vbiN%3S?%<jkuU0m
zWmF1EJ6Ao%6R{C9wD*3{%1zJiS7rs9=Nl>>>~vD2sWa;o^_(exyV%<2Y75}2Ibkio
zkTa|c9*+7uw3K{=<(JEwx`TxPLEOY9`bL<!{nHApJBag_*J59oE!yrb@pH^`nC4L{
z=|<=+%!eX6*dV$ZHl_U_3<+tAG0#9CZy;7aCY$Z$(jWN70t6v-`>*jEic9NRK;gix
zazaM6Y}@3D`*U2~Y^lMOW9zb#u2QpswH=uJKrKUni|2ou(kK$i<&SU<c$mgAgZX7@
zG>@!F`WS!wSU&6OdHsi@s@1(7x`J-yRxaPzJ`_u0e{*~t+^rpSP9n0f;CnEze|sA4
zbEcV=h}CLrY`ithS_mNP>W7|qVl&DTFBL-c;>IBa|LqDCT6B&b2(sSZgkSsK9aMaj
z0H?9(A&-vhI>E_)u#vkv3E;bY7X#t4o4Nh@HLwP4&CYi|UG?PUo6Tn8CvQ4FIzAt@
zgiVmxYk8jhp~{kJm!ko*WtmY43Io!F?=JQFF0$SJ>>(~Pp`CyM?Y#{}0z4ijYBYcY
z&0EQ4HV;Sj?bHqgf9q8JXD$6$RP-}_a?EyokIu+_^#_C$TV7eij6wZVqz7#;+up(&
zRy7w_-IqI=O#O;87%qqUlG%E<a99I^c!<9!q@OK;NGaC#*p7VfqZAnb;P-Uxb~kqZ
z`)1FJS+D#G{+8LeZR|ymL+c85iSK>BrN`r-J=fT{$KIIuNG9fDjV0sXrk~$50>Hb9
zO@?!b#xWr~HlstE(Pv#vaM<uqx6f39<@^3)6O(8DUH8xjc;O>u{*QQsIPyqXeJdu%
zFW(6=vK)@3q(Q_=#loOP3L-rl!<%F*E#;RSodwo@y<4ck7naX)d7|!mkOg4j$2@L_
z=VOytwoCRs5W&KV;*-4451M-C&#Y<S;Z7CiY`o75s$h=|&mO7=y~6gQh;jvnfL#d=
z*pst+byfPGSQ5CB>W}Ywvd7FnB?;1nB^mIcpQ5Fw`{`$Fe&cgs9h3f0BUD=5L*bq-
z=dqt2ar>s$I3tRW&$8e4kJ1<A{<T(8O`~ptaVMTodsjVf;`rY-_20U`V$<K9;I1sH
zSwWu(=oCF^6yD~iywoNem|ENNO)ucWn?;rUds{y%S2bTeThFcMoJe@p?J;yz_%1Zn
zYZ;6diO@*kH{fK!&t;2)!~g8<*LF=LQn$7R?u*D1Zrt*4jt+rJKfC)xA3+E-Iu#*%
zvu-{;*kRx-dR)RJhQEVmzix0?lit|72Igz?SSsKxtu4!Is8^Hb)5|$3mznMTSrcyJ
z7n@=D+wXFM^tgT^Z#4uQb3~FWl0r@&uD~X3ou9#W1@W(WLED2Iz@%wXG68SY2A$7Q
z9wuTAsy)pD|F_IEwBFNW|1LWp|M*w30fq2L4Dj_8c6dnI2?M@Y_{$o^c`qKX$L4lC
z@~GKEI$8xbva9*p14eAR^4@H*V&#Zf8O6tat=HS^cpVz8MNH)l)oeC7I{gFfvI~5a
zxg3-NCW=P<0))WEqn(3@6K@%>&+^B9YbAg8*2VG&fR!QOX!hf^t8bqSdRfamDp3Tk
zdfz$^w2#Po5V4U&wi3hlPB^y~{$kB~e{F`b@7~vkJ4R}Rvsdo>0M#ut+adKy?B@Im
zxvebCx&17*OlP)qdoDZ)nS5|#uR_7zAAtBNN9sq?WB&z=7D2JE)`R13BY}yeYj|&v
zywC!sVAl}z{22dTJTS@$(mStRYhy4rGlM!Qb1f9FIVzSEJsl?&d#1VTXwj&so?5b5
z5+#m@hiBH|5vo_&zkEm4=-CMD5AXm9*dp+f8d$@2-5CMp46cGJKkRuXou${V1hd#?
z4ZbINI(R;N27(Q)+rMs8`5dpxvkz!n2Gc@FQF|6G@87Q99lKBeoP3ul$`ECGz0akn
zo@{gb7Lpa-z1dGxkx?Q<CM|XC(X{n9!lh4%m$fE7_0M0mzz5N`3t{SDVTz%;@weQh
z5*S`qIxu?TNX0OEhK*2%#?2Z*(_9vpt834lXUs$SuzvyyD{sw05DnF#MXe~1pWL}b
z>#X1K@%Iv5sE+0ZqcEy*-i)52W4*T?9R25v)WiG%sC%ug(IS|W-u#2~W&ITIhp%-;
zM2^G^9ZG=M8th2IFcnx>N*JGki@PWs?s&1$^HvX+*mm3n;S457K9OikV#a045O64~
zTts+D=Kn*-#TX=H#qo!{g&3xfClXLK3T6Z(8$#b#BSH2x+r?(#@5ZG;)yAePKg_s2
zxB798*P!u9B*1_qFZ#oPg9IJe_qm$*n}b_T9b74Y`m8^Om_U*c-+qePwfozX<+#o8
zVHUF$&k~%v$##3wBGxA_Uxez#IW_nCpO}9JNZEDHdgyF<`{M4^(Pfh!gmU!S+^L!B
z$S_``{ax8k2~F(2zmj$DF*q-sRr}%zdNM_z$dQS;es}fC_sDde#a%lM?NV&XIHJCp
zs!7r-*R`P;L_EBTlnUCDd24!-UGQM7*ZBAGz7?fsWCW*l+ayaA?l-1%wg+gsnv=4V
zW+n2i^MjtuDpeL0gq)G2Vf1k!m_MtNtPnkb)9N?#AW3cmr>;tv0JF-5MA+aMS3K*a
zl~nBz*A)Pni9*N#)@4I%16zIS;PeU0ebX<D4?8C>Zy!+pgiN_}p{^pWJ}JZ&UoWXw
z+rWuy`Q?FjWv5cfM@Q05nYiP7JVS=$sYy!I<HyH5S-h~xhX(=R8I9*;#hhGtq+v$$
zQyn12WpEIpa&dLl<;e7qPFAtOWI#Ux=V85~qc7W8s^1pME8G-MyKR?1jbvhRhuxA}
z24AiijM!=wNGO+Wd2TO%$XE!thOU>bxivx?M?EK{RbdPC)Ep`VDz|vjZ*M*Srr<#z
z`G<pe!q=E|o@^=qMA;HAc&(>oqrE}Zl-qe+q->2C77@T#B^Qr(PJ)l9Jts5^GNU6I
zi2B(z?Q@_qb+g9XE`vLjO-3*C(xn?M(KhXtY_g5Wq~)?cjmLJcC@zBJb9P4W*><fQ
z8oS=Tj1!eINm%gC5D=gBGZydP!?}Mv&X;V;eV?%40MWO~I;R|@=Vx<OYbr<j|IUD&
zNyP$P)2doMY*xWJ=7U0+-k@i7`_bI6ev=6|usxjCD9KRx9rI<F6<e`bspvx(LSuxi
zP^0-B4T~E4(bgS(KZpi6QR9Zu%nRfXk?)<FiDy7D=@bXphKAp=3kz+$FUhS*i$m?~
z$;yN{bid1ZM(t+Wtaf<En>4Fbtue$fnH;@v#U~5S&&k1*l!Ug`Xr?hAxGAlC6~hi#
zWQnsOB!&n}Lf^Nv<gr)H7wz~t#F~4#qpN>67IgpQAhH51YwW0nSEkjX?Ha@P+XGJg
zWj97DvX<a*mLw)EI@y6(n;|Y?8LQOh+CO|h(MB7NU0hPVI=m#zGep{w;t9iccgbkj
zb=yUz{1cny2Z(|%jGi&rF@jG{o^9;*vX+*X`n}D`_x6C+)}<9xx-)NNxy$>!r7Hbe
z{A;5E1?MTV2Is_gzc4><OcWuXme>wg8llXz_$1;_+r>!mN3&NJ0SC(8v<j8Wtm+CU
zzf-h<j&R){4!t|_MUuRLGVS69VWO2!W9Ay03{R<sK!^r%VQV}loNi{&$ZZLSGIip_
z6G`7r)pmpO*(1Fg10sc)BXpoO4F6L34rW;5>5{B(Ur;O~sXcH;I5F+*jfK+c_`*CW
zeBdvi$hckA@aihfYDet0+8_TD=``leiF|0&iYpilX3IPpWM^jqk)nj<z$ug7nC|t!
zpcFy?*HX52P*-2yL+$>n$EWMw^na(j!k}`Qz~C{%pXu;tP0P{`kQgN${;N-`Z{J_=
zjm<YzuV)d(kPTapxKr&8aKkrP%s)@5%j?S89vMe3!3SgMa*U>?@KgNMTW!qf)fGZ3
zsSnR-MvMErzoL2Qh1gc)K0-VzXW7trygLpCV=a37?jvRb+E{+_F3Z5OUZrFH{`hVz
zB`*9$pKfELboZ#&i<6%Tg(A}YS5HrQjtu)BSJ7s2tH)F-p48S+%F)4wE|=ru<GPgw
z4z0OjNj8#a%!3z5D#;SBWGFf62+J#1B#nwAfeK+%s(E}gP-Q{Kv7UgI#TKWwl?{VQ
zZ;3I8fWPg&eOX`f*Go}qQrTIN5WTG?e!qh`b<jsTZK*EHi{E71?`5+IS?G_o!wvD5
zdK9HhY?%SQd^;0IE32@;Xfdsy8KX3k-~_{}f$;lRfc@+K)dG8_nbf*!WrQC@p)beU
z)iwXi|G3UU=XKvqH$u;YbK49(;FUVKAVQH{XlZ3*6YkWkL@_|&IysJx_(f^MmKn&g
zOYG?>=mlkXMOUp~X{g7SN*~dGDP_2q)$V_dqD!;&mF_jKMd=!R;L^$(luE*?r2{b>
zK3&|PXzwb*LH-EZonZklECAM3<`^404e!$JI^PGeoD#?1FCsBMR8=kFbZ{55mA1Ez
z0{fuJ>)@N{3{4h59GB7lE;XFPI<bYz_LVO9K{a@c^)(NG7toeAN#=i8K6A8LQG%IR
z`_$s}(m5=wt}!H`a@ZCs4J|Y=pfg9QQ5ur(`yS=hZf0^W2<2IprUmyT+5xgye4!a*
ztj)KqZw{@{YKS|(tnARl6%R$SR;y+)afFu|apX)~RiSwaziA)E)Upm_VOpYS%~8L0
z{x5hXqG$;Dj&_|B^fX)nhr?A65zSaW9uHLZlfvb}^Mk*2@7*|sO!vD>7UIN`fNH>q
zYa=yTqU%my_Swy?zz;UD=70S+3WWu!@NaERT9RESiuvG&LGRJL>96}gl5ClEEmtqr
zFAl(vzOB7KFN4|zngoB**Q%cPrw@mst*?^_U9x04QjYhLw{~x!b1S*9Qtd2frz}8b
zI*<8n5SVr^F}X?OL{K9U3w-9uFNr?Dx);Y5+YpdxB6*F!u|k(z(m78e4h%8gZoETm
zIvhVFUb%fm$Fsk(%)?}pb^>4NdRV+-RR+J9zW2L%ARos}(!gY8!WuT{=E0Uwvu3Nv
zuRXbM!tVzqXyr&Qf?g!<e-%rTtBg>Ap$-y85tXGycQ-g;|1{L}3eS$b#Te;K){V=u
z3`n(nx?7ZnDPSjvDCs6_tCb=J1yxq;4*X@*m)=wUm@9S0f7@awquRTrK(Oweos+-0
zcPL%VhiQ{tn2#YODd|5pcL?;et@D_!$@(4FRs=(21`osqOxbz0B&tM7O<K{?lLHsh
z<mBIQdQ(LI_=W@3H34(-0wi4shK37Z4xRbovl^#EgsrJfj>t7sn@njn7pt;OAR081
zz!%C$yVx&e#3nBIdPT{X+h|{C<$Z|>nRr|CKOg((YVWEQukajQ6xNwWS%l?zThKOD
zy#`-R{F(WK69G-MX{+0pb0qvbojlDzfZwgaN!Rns>yveBaEp93V+SBsw~j8{Nt`^M
zzm~e3`&wghoAJ6ZI8|^`YM|?(>sj;Z^7*FrtH*Vcuo#{-D=Mngy0TgiBv169#bmCE
zOd|i~vN1{2%r~+czn3-Wi?!#h&wt-|91dSN9GFDS;p-z%t%XEo?h}$_A7Wszm)4|H
z**UhZfh1*CV7J)jW^NHtA2y>=uMU*@83%mzPw!**3;l9OvudWCS{+65teVz->IwrU
z3-y&%$h@SK@p`{E1~2S}*uEejsZsm8qC{FrnoQr7gRXLr=fK1jc~w;vFt67)H!G*H
ziHI!R9%}M3(MIc6Xz85?%G~ZSE>i%MIlFQzHR$<|pWkN+VBx;#y9c-b>hUb|d^`Ef
zR2z6fZPw-y>hE*%BQ3%``2JqVZVz&?6r?h^cM<>3+9%s2`|X1}shd=9bk4W4DU~0z
zZ`7BPGgUI)_VoN)C?!ak6yL?fJIBY3RKOc>yOr^Ip#4MIdh34}{atPE*7eprXz!Q2
zev`k1Yb`Us8aj^(vlq6`U<fJfJ;goU%M>1iy_Y<A>GyetNpbRbX5Z-EWk(M)Q*)m`
zlRHr9nf&Ukf8&t5U&z%pJC8KzKF)7lFE**<lN1~s(B`R!HAnI<NM%AQ^*9c4mIKxY
zDAb3q|29D>Y_Rs}V@{>AOvY8|D~(2*@+8gv!1tEUwq4v$rrerAG!^?+^B|gtjJBlk
z?WJaK1qljfT^CC3Nl)w!b10#taCk?-(8vgXi%$y|l$BP4C7FaZiAaUxGOMD}kVA|D
zI7W~}0^rY1%?eOzMp~>AArJ*a0FH|OIm;+sqwlT&t5!i>_j}Fb)pkWvW%M;31#A;;
z=Yj^?68^ys;Z(gXG4N^QME;TEt0boE{Jb6za;dx-C{BQ-L$~(g;ez+e){djh!(Zuy
zC;->OkUJmDYn|3;Y;V^hL^nW`v6~TLPsn-~hz4H?=K4dy)4?=TZazy(t5)8w;Wic8
z43X(ct+|NBTF~wY*}~#Nj~ww3ILnz7<g!aq8-&O+dqm=%z@BA5Rej^mM&j4V*<UwX
znh&y1(a)iUPyblWx~|wEh9qbJjX~U!ixj#h968;r4IXyxdXE0}Faiu<@DE@Sx|9?Y
zx#WRC8q45a+s<d$x`J2@hS#UyWUdfq^){bpx;McOMAnB+Hm<IHJXkR{JeD01B2j#C
zQx03ka6wJlEWKgYB&I)p2-i}tl!H|Jf_;6SEPc#cmBw~eQWYUPjxpY`8Hc%#CKg48
zh4PMxd=f(ebe+_J%oSYkzVT_W1UzOM-Rx{Bi-?M(D0l$ik4D-5RxR?*;f5o*=jNho
z@c(pT%Xf^hmuYJP&T(4ZYR@zG3v7ad3M?PC_UD|03j(hw-B#NscC}?TR@1c`+rHqY
z=5><|XAQ8Ql|(Of-Jbs*)sfpW;&K*6(hdD1D1$xV+rJ(9s~#Y=X`N8cNmHX|E$CN3
zYgEUDG09?|EiC*8B(vFKT%j1l)IC(B86EN>e52vlV1rn!na7Xejir1}!djP}wHFh=
z@mVpZJ3WXVk-?N9e|q!`oRS-u-vm~^!$7*N{ZfZtPc2Uod`h4+BTTt%CpqZZ)Xm7X
z`=CI>nULBf$&f_ac2JJ3+F`&!mmLB#?4CI08I9MxDNtRC^~-NvpaSMO3ExgdDDBJv
zCb<JecnVcS`MlfA&qcU17W0U}7s(Z<H0yAzo|3NG@HNoemReMy;p%_MaWt4x*&N}R
zcglfjW!XW~YY6@sn|)W!dr>EMb#rFxw$?d0U$qEGif{25!|F84^sGHSNfLh;@PreJ
zY2ixfV-nX*Iv}mG+hY&FicKOnoqZ1hnR>&v;BtTm&0~Z3K3*2qcleQ%F#%j?ln4=k
zO$&70T~6xYYUN9$AIES5IVtoTOR*#YC#}JGmJdhC{1WnNKMkDzSGInRjhlLK{%z+K
z_c`XO{dMl$(kh~s7Sf)BY1`G0MB>kYjKzb#FLpk1TRa2aWQ4@n*qkm~!v2?s*v5zW
z=h0`fGKFGhC5>24bGJZn{`13|C)%Dt%r~94n)taV51Vf;j@EA(WN-7jKJOygov#+7
zY&0(d5T><TV!1z8`l}{>D0eh+AAN|Z3&J>A?HFEh?@~`=LkH4*giLA8hI)82bMo`8
z0|RL%uPY5l7ZNp=-s*j0xRQDUIf}xW^zopL){^`&OAy28-}EhGLnGR$eN1q_Pzih3
z;!uwjyJOC*wfp^;4hxbby~ge)pvTN`=<9|MQ0Rcx9Yo==f8`2Du=0)b#d?*eiSEBf
zfV4;e=}b)Fk)L8nAZrwC{zExAenoe^c<lLfa8X^&1v*;(;0F1_sIK7D%t4(;y$bb7
z&WE@H=e=kBS@k)?&qie54E}BCw^U=WhfKq{nr!kH?wQcLm#Is-JD$0hI+i<D(y(dG
z6fv1jBL#HOf8BR=@sIzlr3h=;zY>RMKup=O60`*?b-%qND4NMYXPrZRmJ3d<iTyhg
z(;s^lF_gQsF|_5~;`k+eR7&3uFb9r`nWC;0cRgsYh`0;=#0Klj`dr%lJ99X8clq;*
zLBDmH2;YACT9>1fg0jh_Y~iZsA?{H=zHFU?)n7HPTu{|!B7cUI+PX=;v_h==N4tuK
zBbMw?&jq?JO%X97i8+ZRt-t1{p5Bxb&In8LvRMv&lk;9g2pHOzNG-}0p6RDC_AXNa
zn`V&#pRu82yc94EuxCndG24N`QSEkFU_wI9a1G=#eb>&`Jr9*~<xYL;>nS(yzfFpz
zz}DVbuX0JbNiBRE*C~^@hLjCWgB2K|_Q?NJ8zKaw9k2O^%tY;Xopb5kn05Tau`d+s
zD|{F(JY5XFv}^|<$ccT=V_6-VmsejYLQTamI*+s}uFV7+Ka@roR=Fl;If6mvp))d_
zC%T1@DS})xSjWKC6_4kIr<Wv3*Tb$h0Fd0G0S`$Jatwb`>Yg~L-DIhF5Cx3^z3fYD
zPL71$Qaa{?9@)X|26mVm9s}goTn(#)Oj`;R6*(YoXymR}C?1J`M#x8NKtjE&Z9EBd
zXe8_V1t+BwLFgv9%j2{%cc)Byc1~>hb-%tO?Cx$$T|Rg|z=9vNZxGM_D(37FUcKaq
zYoo7s?G-uswP*cs!Gm6_SU~@-S(9gA%ake>>*f|GN59O{1f;Xk<e-0+Z9dl`NlZC8
zm5uE%n?-I^U{j=3p;)4;su8;XRcV^;Md!(8hZb&=bcV($=%bn0C;NRri6TO^p@xB)
z{nMTF{mMoAl%&zIG1NpqG*#SDMm+His<&uv%APAvjtnKqUH%4)GpM9;fFFD!|3iz2
zi0?D3_5Uen-83uw8<TBux>3CNpGu9eR@Ewak6QTt$FB4;jrLZhVt&E+?GCe>8`Kx9
z+69;!?oWvI88@F(v*?q6Ye{f;rfvug-r-wHsm^M>-+HvbqSONalkQj95wD>G83+3D
z8fSr5D316H$OxT(D-ZGqI4CAdWmF6Btex)>WKw3(&l1fP@_iW&Khv7AeQ|8HPO<jx
z1Bfkm-(AxTbr5hc#C#<r;KNN@LY08f$5wa#B=EmJnlSFFm-S3WoqNv`U45f_dF8<R
zYP+y*Xn}kjHl(r^?0ITVGk{O`sIS%jRe_Kd7qT4~{vJGMOf%%mK3bH)h|ib+t~40w
zS5KH(kcW65a{{tLrFer9=@11v1Wa89aaQ%G`ha^_+J7BT%m<Va<Jp>-t0r7kJ&vyi
zBQo+8v0n0Te<YT}o%j)RM0$LZD=!%Sg3%)C_SQ7Y6xxx*stmDuzBTcLFE34S)`6)>
z{iLmlnWN-$!nRnGsW~_o<>c1|-L9T6?E@LBc0OQ8E4X)lCHD24^ch+B<Qa5#rwv~+
zZY@8Y>(8+K{>P=I(jdU8Sps7Cnt>)G4L!7Y9``ix4Mbt$yYz6S;E-7GlemTN6AN%K
zU4tG+3pdq$4<Gbz$~0<iujJ7pUG1l&BV4U&5F6P3lj_<%Z2@PfU3a_M#wOo)Z_c)*
zKvz;wD0k3Pe4*!NuPQb}Tj0G&4Fa^aLun|5H3Sm-dWfDS>-)x)c*U-1@Mp^P`W*{E
z1ziUZcoyk#U-ZWF0wdn!`S@aIct+U8q-?YyzctPI*}eP4gC_4;KS>4qB8O6aFQ?Jz
zs4&e?(aLJ?2i#hTa3(oCtbDhYX_HyERz2Amy1W%2acig-i4Otiu}<V~b6Vcw0tBXr
zsp(47%EH4*JKe<v!9$v=XDx;t=>P?RKPVn3SG~QcM__6jd_h5i-YE$r%;$s-IHena
z4yjbXS^IP@3oKD*19k_DJYb!*)uA1fzU{2m=XCk_Tkk^EiI3MTc1vE#{?MyA{ii>f
z(jVu>`s2^fX}6a4FSoeOPC5S9$HarU4ET+L1X?t*gA)tnCXPU5c@JCI0Jx+KAn!dg
z{<Wj9DFst|@Uu)3sR0L>-9A<IpZ4M4hsNjbs@kBu3vKD>yaHUlKmRzyh<#XRf3jJz
zmXG(ywLW3YzE(|r_-%biBxA||6%c8n{QZ-#*kgQ7E`P6MT1K(_?ChsTv~uW?sQ!W~
zg`a%~6`BGL$Dpz}&m_9`m!J!&w@pK_mlTYMf#FZ@P-ajTvGR?jB|?vRZZ6GKlIqM}
z^>Zh;y^NoEGfz!NcS9z%iT$l!=T?byFXiL&4~KgN*<U~hwI$Z$ogett_-~^;<E}sn
z8)F>l^3umNM7vl4y*jC$38C7uZziy9i_jpY3e~7)pDMlJgbq$L0QRwx*(`=6k|u$%
z@r4u$n#log7O@2gZt-XNl6}HTbZrF+738t;!aGN3(+6ke;o7O6nLyL6;^^E!*1O^V
za*a=cB2-|(;P7@TJFjRyVJfzG_Ao5<;m^q<&jp_6YG)v|D?Z0Gzy&4)++XK4?BvVE
zp10dt#{kkLoO!xQV!<}{1u#{Bp_(O|I#nOB2A~iT!OQ=I&%h)=nvJO2nq5$lJj7!2
zL4Q)9t8NT}C_LMv?qXI)W_UVpqb#}yjnvz4ewVpyG7lFqyi?)AQg^paue1N!`M6*T
z@T2h)7K!&&i*26Ug5Qj^zU=J?Tpx2cv^NiIZJ}z*UXT_B`K4a}@sA!DAPc&Tx2#`u
zN{=viv*;S}OvGJu`?Isyb%!ypf@4Ff;^YRgu)pY`(wxTyrd`rrLNe8#%*C^go(Hg2
zIFK^+1vc?8Lp+pJsN$e#YZvO)`koXsLsOW>dYs9`mInjh#|T*b7%jyv&vCE-5B%sF
z^wkf7#oG0atD_HL^rePX7xT42nZPJd7{##M?*-*oPn%j9wnHYlK?ta+sTT_K5Hn8A
zM<4$Siak1V4SKp*{3za{(7wFkZ262?p(U+esV83dQpTMX8ZJ28<0cFh6%_1y6?7^I
zD8uVR4g-L|Bh?mM$cBI51qFBW=V}r1b+!FI%$IEI)T~4{^2w^Un_0Vozhzkl#R3F-
za$+I@u=;oQe_aM_^no?7VRjy5J1ZcG)=~v`cyF1?$IW~@mbIW?a}vyh>XXjL70r=-
zM)9ST#;#iU^P}Av3@uGc0G@TND>Q=c7gsFn@~aR3>-~Q-a+C6)yS38aPBbwTF7~L7
z%*4ZtbvtYUOu7o&xH)$iWnE_?@f8>Xp}wm020*em5;l=7<IK%pyv`HR6g*RJ_Ys0T
ziDu7Yq#R(qJb#X<+)yvrrr*FSX%NHsLER~KDR^=sM^~^`FtPby&^?h$1PT6cj1n|h
zZi_$=rip1`_sv}viZItRCVV24z=&Mvso`YlJlSqbOSD@#u6~a<KxZLx^H=?Q9kuXL
zlIh|=-&8C#r2E1w!SJRt+KWp_H&%f?a$79ITFIsQVxwG_-4sim*t$uZH1fs*XEHOw
z`d#bnKwjnM*Ovu|@;t5v+U_X6I>~^^@8ZJ?S9C{O(OnN!Zvc(f_leVT^8m2R_YS>Q
zON*JG&wJR2mfiVRsb5|5hY?#Lfv8g55Kh7#(Q|Fy%lFA{p?rmasY70AnDgUC6{(3L
zblw{91X|BoTRX-ARH8(Fh@kx>00ulC$UdslRj=>-JFo?Ebsl8-TRn=i2S(K=n<h41
z+-+jLIqkuELqFATods-nl3Fzz=gZ)3vresYVykW5@qk<#n~t6E*~T+V%y4WT5h4Fu
z^(46ap_j`@_CBK-kN`5%xceGZMgKM!<Za#!yi)HaED=_xAdUiKnNoNGApOlQ?ygg$
zeSv>!roMgUBsafcYZ#}z2H+L-pQuxtT%C#%L!R%DRD7)RTWt?~O*S+pa_HiEL<od(
z;9H}Q+r>`;kCzQ;Swr%EyGyk@Q$+X)bdFkB;)opN;+JP0L8zpQ-YuvtJ#gK5|Nc(^
z;&ofn&R8X@?_v9EG0+x{#;%0P@1{BZ!5>D;Xqk1-9TaW4Qf1%kqABJ#FE4~imzKZ`
zv{(mX06`i2xEMUp>|9txJaAsWU0v&5SKq8mqY8kEeBapD;P7qOwG5E%lAX70FZ+y<
zuazPp;#SuTizZqap9ID)-PN*L*G4E*v0{F=|I7xgs6ndp-<+(Gr`IcoK)RMSoHvLI
zbYG_33y$>Y6x}5otC3G%jGCpf#BGp+)>%?hqkrr|H#tX&aQgMO)f1g}Uihy%cTg$q
zEi3)YX}8EqL6u~M)u7OLe3wblCt^1DD1%)aUyX*|=m#9+oyi;|Poxjey^Q1bOnv(e
z&D%=SE3<ThB2fi*NaA*$YW7&C>RaUvof`CZ1*E*(&0<Y)>2pWHXLm~PQzm3WmG)05
z9xVDp@61Cg+L-;F3m=D_jc1-4E92o?=6Y0iJtGE@=l@D7GXfIoClZ@m-hEF=6HBW6
zFp7;7FVfv1@`4&@0D-%6oVLYvn&0~Yv)ToX;Wu)iE6#=RBq7dbMtLuqYXcKo*8}j_
zi_9)#(s_((Er2dO|Mpf~;wS`Mktn}`9nJ)BGaq{SN@rVCO)@p<F#Q+GHH1(T_f4Y*
zBu(GTJpHx#M*bYOew#1*+223;TJ|ml3*kIw*)<3S@c&f>G6t^Sv_iTBA}>+!P22q5
ze%hGGU-4RYE#*ut?jPKv(~v~EoJ<`?DaF2xVMIrrJkhB%`0E8lUyD&<H)$9(gYPdK
zazIxk18$zM0Jx&4g`j{bMd6E!W2_D55IwFuOT^lXe1`}D2O$X*+IBXE*wFmud#+k8
zk|A>qTjwk_af!R=n-DKrt?yERGAXg<rc^asHX9fs0(=g32}S18upH?SFwV>r?OQFD
z(hq>p<Kr9qbz}JB0Cm`Hz4Lt<o7oUUoWk+i$KA$BJK*$+bsAwF03gSEj!cQ6q{UfU
zZ#(8Xy1mJ(X9}C*@w`mjzMXBY=`!2!{Q7jEb>6cXO<vH&9zjd1kC^|S(0ag8UXzJG
zK<O9ay_R9ny<+xcAqksu3o=<>`dkYyHFQ43T7Ne8BCfBmSBH`iDiis*t!s)_w9Pqv
z_-|gq)APc&U=oO}y@X@*?@Psz)jOmUliTzKYW;QzaTHGo1O2tb-8;s^D8~4Nbc539
ziG)9{qf^oa4N;-C%%poXsma|ue$rsIBPLY&AsFR0XUUO93V_hgSXM&JC~fmW7BZs0
z9&M|}_yt;|%JW%hhpU~=sU+e$zT+<K`iwe4{8cq~De20PAHOc7aLhD=9*IHbSPVzG
zm_s1!@NA`(bgYI6p)O;Iu9w_#Hl=lV)BIOs(`+#wkw9chJJNG=i|pzORq|H_8Zo<$
zR|DEsB(u7x1<=eH6jw`<uiLuSgd)Cr0oiZrKj)p%<v+cOa5un4RV>phGicyio_Ahk
z%hUvdX6m3Tr53foPXs_@W{g@w<<#e!(Qm#y4tfLl=_gxToXK=_UAeTS^Eu28bA-c3
zY{Z0e5&JEVbw-NPf^Ll$jalovN&^O5l0UxNi&D2cf~#F!gF|nE`#Ea^@&OQY+R;bM
z=U7M<&|hYLPQn}Bp(ryH<z3+yFK(A!mJvWDq3G5LYRvN*iQuYu5myP?feAOPM0JEI
zp%Om?Eb7WdzJaA1^V856=3nUZ=P{&WqF{Swcd{E~Y+}5H;O&wdxX@X9$$>wu9+m-4
zPtb{#_MCPwAl=r$Neb`ZxF?h9a!TgYkK;Q;#jO3@Q!IeVz&shWi$VT+?{*{Hr*gsW
z^pO}?vw=2ymA901dG0etd1^Rg80FLY(err{b11)kBX?+)=m+&)I=8BQO5`uz7hAbz
zZu8jc$F<k4)q&gvTvpXKG<l^B?2?Gp*yk5AdV2Fs16SCfMfp91a*PUo?W>4GY9p<+
zl8?Mhn)09-qjZ;kIY-=m=jX)xld`e<?Y}Pf@&XQl)-IDwq_qXR82bR5Ba?$ubM)JI
z-|0c7wnMo+D=#-UkS3{a0FZ>mX52sq;Jikp31skL6*WG76#L94IhU<El-I#(>|9{@
zCL=VpOVQ>gLty92?~tjx?o>WYa9_Wlx&qp%AlTfvm|oC4j<hVNA60Os3ne5=+o9h%
z_txUoRdwU1TI(E`Dkd_hgi3yz`C8QICFmTTy^d*Z*B-M5?SM=1N8*9*LelmU{1?Ai
zO*}$9%&q^<jwF%!UEs91R{89|>~4HDJ4r%KTj<4t_jj-H!XA)WgctU9gfZezy%M80
z@74}Ei&r0Z(_q$>+}AQ;4N8r&28`qy#j5`qWp!$nN~|?}Z&<Q?wKu8m8#-!K2?wrx
z8~D1tg|c|fH)gfn0U4YqDZT<73Vq)pADyFuu}CI1!za38<q1`%;3OuSB9IU|6b_^B
zzh(hsPG<F(67=w1ka-wiPfSP};>Y{274X+OaAEtcIKXxc_V#k=VIMYg@OC{209|FM
z7mS9*=-m-Eio!~`yB|VZkAhGcr+$W6_Z?vk10gy;^GE@9@#Ti!s0)_ow<vQ>cJ+)M
zn2!vA3v`aaZH994^LB4v7pN!`nJB9v-JB1;Qk3aRU|ORjX$i%~0Sck9ijdEBOmR${
z*R&Y6we}|@ra3gc*?`Hxhb@a!$j9dfzWx+JEm|{sfa$OS*hw5*OlM5^B_*R%!shdQ
z*t3v1H7j|Nz6ILF!~v5P@WUkm{;2WyPzE{m$P$+MhS6aXPN~Z=&$ggQW5*V7>qirb
zC_-X=k1f?6)6Afq$FqY@7O*9&a~#j*C4PIuYm!~-rcOTmo0YCx&MR=MnExH=d_(>!
zuq9_L3@8}5{c1-gZ1^qfmRI&p({0^fuGw%PHSNdB-oeE_9KN=NgbSK={$p1_ZnpER
z-z|uKJz4X?3FBd`V~uWF<_-ha<goi+4;n4(y#dETh`6{wD)RqgF7Jy!Z>8l;F^k8_
z)Srw-Y`)3ap{rrFNtr`&5&+eCOe=0}6i0en6fC>AW}rBeq9G3bx;#G~@PlsS)a1Ww
zA^AVM%^%7(v~Fq67<Lhg6B{0nGU=m1J1yB~1@%#omS{F93L6r{3`e@}ICSad0OXq&
z4K?54lWjhMf2U&mPFSj!ggh@dv|QWxH4Q*q?35&~t?KaHGpmQ4WPzmBo6U-;nXjzA
zI?jI63*lpqmA?O=CO_<b7D&C=p`@<Xq`(==gcNVppftnM=(JhHiDF>sis-)kv2rBB
zo*8hMnCt!9?7SV=jvOUjvhieipKP)4)N2@jcc(A?F^d!Xp@WAj7DTJwZia%s+u%N`
z?1*^k@!=C=gH&d4DuI-$$OHsU2eA7JpKl8<`bqYe*4BC`#&a6&0l|c3W_ant=nT&6
zo6qL~It!jVMV4<-N4ukRGmiaQUclxZulq`&MS{bH|Ai7VJ0K<oQ~Nb8j_=>-KnyB7
z=z^;GRHo*m_;9(cw6yQ}mg&v0N;2~Is1(2FUbTpdI-VTy1#cxBaYgn6*`hR}-f2}|
zx3juFySqn2#<{(SFCO1K0Ikd@Kv~Uz+CdkX>VDrQ_ZD~x>}id7xyAuwH9xM|;(pH^
z)e$&^zl9d3+CS4xRL#5`yQKx}&;pgZ{1>0W+1RfhwttV!&U(xy0ESa2lSX)<N!FiA
zA^NEqR9?1xyv6%VoBUaiKBNEYh8N-b`P@SD^AY1^BO8&gMQm;0<%j?6bb_8n3mfa(
z2Y^@B&H{v5NgnHfLPUg28`zj(A@V37xBkJ;CnS+5D=RCjQ2#ff<|kAGAlww*3uNPD
zLu>2Mc;4bi3zMCJEDK-~JwI{}UPRr-d6CcVA>dDugBpfr`D^$xK25!Qt*Ly8iGM9u
zUJk~a{8l>?P6$^>D%k0Pz=K_l(RIk^3B=f;tAZt{D#l?Y(Q96^)O{f}qFGdfdEy??
zFK8<y9aMJ?>fg~TV~C=cV$~!EoJG4QII)B;-fy2yj}}o#ve7YP`V5-1SuNHdcv;m~
za5FF8HIXO?$jWLquCW}Tf0NeFMLx{DOHhZ?-~b>BI|R+r70F{BvS$$uZ>J=GZRgM;
z>A-MiR_oIZH8(i<LF~Ggh%HaUnms2Rl@TU3jdXF@7L!p6H6%X3wzK1OaTkU(s+Gwr
zG{SpQuXbX@;{FOTg&EJfc<U}tS*E$=Xr?=WeN268R4LfQjLRgsgzLkM;WkT6(@OQW
zf8{prUwV29cdVrVSs~%~44~>*#(}W}ZO_`i%uypYBdM?-qmjuno6DC7t~Z6O2@(bz
zK~Ci3u_!h|lKzr&93P~B(HxrE=N1F{O+j2XAELker9EN)EAr@wpea;p<HS<*2Lrw$
z0Ius%Pv!Fc{qj5^y%&&WQWo)#CSY^wsKsNJX`9B##8OK_a$c!8lkugnO5fGqTh<Q=
zOw!#;g1)-G7C1wfH_M#0rx%5=3Dzu`&3^~8=(_#0+WGbXp|*avS9KMU#K)Y#=>N7_
z4;$afHSY_2|6e7^|5#c0t;B1Yj)*gJv&kXvGVh;IqKY4ApE5m8p3qG4Lzc$`W-bo|
zC+uR>r~}*^>}tFeJbki!{`iH*zm{dhXF9x;*q<!ik9=*hHFwZ>wWbF5SZq#D!vhGK
zvtEo|wo-$0F|bRHd%%UTo&D<<fUN*clT9TIYKEq(-g}hSy0mJtzXvfzFOtT)0ZeKe
zSY3~ST7@#>_rN@wsa<=Xe-aRVy&hbMb$y~!m1%5Z@)n!8m`#DY?fk$Vpmi^rBmp<~
z%Qrz@sT%V&?&!mh%~;Ip83mZ0-xn?J-wq`g{+M>~xck>m!z<puROY4fsxBxeWG;)H
z_Z5RGYEx%ZTGHI|lKPmK-xt&zGI(1S#JNa2n>w^biz=n#4f8#s_M`oVG*TF`YX8cL
z3h%g7_TVCD_;WFQy5!xz5Uc2k*-}19Q81lICBMKf6%=V)Dsws`VJm<x%{q`XN+Mr4
zo_gR&0*CtodN4aAr_t>E0F!8$!e1fM5G~(VgpHEvzzaDTW&#OY;Lp^Eg~50r;244T
z<a_oGV{@@Sxxn=~#Ch)M4hHV@QxtixUG48?QoX_m%NPt9SSBlG9k<)b@Wc2LaC$@f
zmy)_K%rhE-Zq0^l8|O>mFCk~uM)p)C7#b)hY0P!7Mth%mFw?yS_0-4`!5eE-K-r@M
z73-@Ql(7nmwbd}|f{ol-N$Ojj%`=DO0mxq#2zfbwg1XN-`uLt&C3ZcY1@A`7KA<@@
ziwjK1B%$Vxzpe5y1VwDz*8U3Vu7kmbVUn%0J$fq>Y?qwrmim;+MCxfv=>i!VD{^lF
z;8CcI=B)b^*{);6K2QUo1`8{gUL7sZn|?R8D7E0Vp-C+$!qjulFqHNYqMUl&TT>{g
z;(0Q`Vc)u#V=D1d3?RN?FxS!)>`mR;54W4{&msGTkGMb@1_}Tl8#c!B;{bCGYKi(*
zHpx&@V@pfawmL9$t|aPEEr<fAY!%pq?>h+n7qR)$WT>AXTen=dwxO-Ae|ec|NW1#<
zuxkB>31r#lbDNQxPnf-AQ=<rhqdtF>w#-w^o5#PrZC?&>*XWfr{4vk&-vo58oc!u}
z^VD;q%buA^tg(x`Ul35xbravU6MVn^^j&MH#`&LAv<YmiO!Ljf`cvz(^nAx9JX`!u
zMomd#3U)D=_plOt@BTN+9Oy}IBi2&H;(Y5`Cgu^MfFUeZ572H-NYH_h6@f%Jrq9Wm
zeJ$mW#PE!`DkItL<>sn`Ws_(z9BeQI&4CP4Os@{w>dJ~=GnpLx(CPpY_dX{D+z1~{
zKB}-OrZG$TdK5arCRQJ$bSWBIu0r(MduL`ETwD9bMj3XB2D8I{r<?e`rPX}*Qk2H9
zyk0{3-_8A3roQ^}&w?=p=Ylmv0(z5jxu`jzZJYi%IEEpqn7Swx%xV(%F?~TTQMl4w
z@ggTgmEDLXrc9<r!=z6m<vKw^U)kfNi=}DQC7&KFqSz^+)j8LB(1jLdiKM?Md^vww
z=xzZFanPG(x8ZVU{b0}wN~9oGnKLh6Uj8;gI-M93{szi9+4wHRuq~<7;G3Fd8Z$0|
zS9pveRFl*ERhfPx-5`DblbgjNC^hLaT(yIG^3rVih^J=1xVhZm3sJ95)dzs<jZ_rH
z1<^!nv`6qqIW@m4LjnlbGWIAB{*8Lw?(~rt`>l+EF9rIpUaXy<+`-s76+5+!sIQXl
zcemmflibse(LfMM#zF=s%T`_demgyS&2|d*z`~x0JM7VUZ=V_|;sb=-|GJ;Zi?Hr~
z^LSRMQ#BFFzxoRl{yg~%Y(P9uc-Dh1g|+A;L%mpD%RXhBR*9Y!?F;E&8H-7d3oR2y
z5*3`lOJJqQP|5phxG3V)E40{|Rddr4W?GFz2|dTA2E|i5ibgn~@nj(7my)dNP4<>t
zXotwyogG*57Y1c&_m@co{v&HyINAPyxXHxS6Pt7PZi<U7Nhy3rdIQX>wXI|1`qR^O
zo7XO}d*?0YiliwQH@AET3Dk^7A7Th!St-+FFG&Q7c&@+sf1Q)|Mm5Vga?3^EMp?`j
zt*wKSvMl@x*78Xn7OuM{&$cT7o1oVUH}Fu&2<GL=5lF7AjEzlP0=I3t3SgFyu9zYo
zko&=-K;e#Yy=$Q_nL~d%$fVAI6`qHm1V>{zUmKg5m9M#7%oaXzLkuMYv6LtAgEmXz
z0dJCkW#H}y!f<Hud#6up+N*G!(Pn3(+gHyQT@~<EMYaDKhZ`FO0c3k&LBzoaE$DHN
z8W{1v%n$LlV-HoBaTnnuy+09e(<jjD;;0v&Nx!l|FVn*rIxtx<GZyr2MuJ>yWAoR@
z6#(KG2o0qcbvu=P?0V*YMp{020%n)@X-d`W!QI2N_P4-0)MI<uUP*myR$ErP@b%+E
zq8_Z^6#EY87!#&CSRTmmZe*Vs^LuRfvsaz<=f_u;Bvv$pwtP-g_>$erwuMY@aZyv~
z#IV{2A0*jSG3kPtY|o!uHGEhP?()Cq9d9xXAWdGUu&dRV&y00t7p)M$SB}C`fb{6-
zj<py*HB1X}JYrWgAKFk>S~GTaARvk{j)`N}$xkoD``na0YszH$0NvqE5_}|Hfyxf~
zXo#5evM04&FEObi93m>%FP&uYFoR~Q7sW8WZ#U=}^X*svl>f>(G=+y^luNuu6|(m!
zLe6>K|0OhOvLMG<M{N^G^mLp%>W~S~AHS`Mw``!sDi#~WX3LTSpd*u^u`waF4B(Ck
ztG#MV>Gdi?bd(y)Pqp(b*r-s`Mzfxc6eSdD1U1sdl`%0~H`m#y7Qki$dzm+3$ug+y
zKq*3Zc6Mgg*z_-ue#r{?WF#@CK?YowClQXI3w#rPc@*6vu}2yh;$GU#^PPmv)uAdP
zP1=Jzkv=W#H=D?n1ni)~ERDv}15|2Wl7r%jP%`+`RJ4-?SQ@x^c!B7>AYs!Y7x>Q6
z5%=+W*JvW5a8m>D_f0%(Np)SWtZ%RR{Mm%G{wbb41}?J&EYU1+j5NTthGxraDLHUV
z5Ge`paVPp;Xg@>qYFLE;2`T8rCpdXA_*}ohcME^siFUoV4%&zzH%0Jsa?Fa)fy{Gv
zhBe?Q(_M#S-|v%6Om7Ini1z|v`Is@@d{1G^H`p+o!*59;PfC5fW^NzoN|M=qtd%}$
zRMXhp{Jyy1(i;<kVNFVM(*N+!Ue?;$Hnr^5CgY*u(vO2h#ngwHQTmZt&=D1Re0+NH
z{McpVX@eGCH2w2>ge}PiSY!_GEnu!Jwwy}68H)Tt&B*5qhzB;TDoKbRB=4KeTDU+o
zil6?1kG|kpBUJ!aN6h;$+uVp^K<!|e1!V+vP7`Pf*B|nN$5-Fw&$4QRZqHf%yOfZ9
zh9e$!DrEN%wZD3E((V7Yn6f+HU?g~LImj<8g&tkd9YqNUa=Sb+V(qVfO6hT4_}o}X
z&Y2NjAa>c2g6E@yFDX0b@QVSpp8kuB{Vn*dCjF{e?tPZWdmep*IVPr>elPg*7qqUv
zmWpYCBiyaZko<iyau)(h>4fQ0EOmNX==cf<NmVbFbSnM&UMCF(BVVI0HL2SCN_PZ~
zrdRXxW3VsRc>NsCteeYBYo^JP{AIZ25;vSMqR#o0#t|;M$TFRiEUli_Pmi9|<UF*s
zbZ72q{yLKx?2)$c@ju(M?hC0j=1)Eq?}uv{%M_uF@e7!$ONxJO$Yu`ET~3;SyN~np
zDySpSbq~+cdS<p5si_f4DmL%JPW*uL&?0gjvn7T%II_hEc-9XvNkV6GcLy-Gm|v9e
zDG+<vaLiMN$-x`FGR@P0OW!JH3E+5&2!VmrHb_Wq@S|-I0=jMkvZiGJ{?BEVM_!cP
zU}17S&%JYCcE9lR<&|rJjOSpkfO-zCwKkc~NQcQsW8=@xp0m080dJT9ToRb{Vy7PM
zBxPl_p=*G)&9^`~0$d3w-}=6kOl2jvX3maHQ@wG%-ywD%`>>JVdm6^rOntdu)1dlA
zDT#8$3JOZ*7fRjydO{hDWDUx_*W@>wuY$8)$@nHqiW<R}m-{c73j-H+pO(>pdrpIj
z-c1nzA!vZ;UwI%}WOnw~%Ou;&^@rQh$2v}K#hiW|Q&ZE`Hc%!XQm5(;NVN?2D^Y2j
zFS4kbc5)-KaQD0$O9$@jy8aUw|34t$CU&-z5I>W-dJWpAdIB5HpQmmzp}n`r84G+T
zA>Vr(M$%gl8*p+gZ8o_-V&CfEjVe@rLi8tu6VTkV0KW75<f-C0^<uwZ3s|0Q=ImQ2
zNzi5S1oS`jGFE68_MJT<G2SA<x3jX3`aG7M3H(uC%6eCU<QYNI$fgzU(N0z(Jqh%+
z5GCOFZ%BAt#p*XfoiiTvBy{l{4G>D!*fV!0@!M_o|6K(p=j3n%4cj`G06{+e>MrY=
z>ovLw>Dl+!$DtNQ?Ttfg_dXxjm+6zp3QoRUUvUIqRXpRx$7=>0Ia?|lXv})qM+&Td
z)pB8!reAGRS_<jp^&^#J{xbe%R+xZ9{Qb?DA7(S9{2@1kvOX-I$j)-i+Wgav#w+ow
z6O+965(0Le>VEIOyhNq?SX%!qZGX#`b0F*<MfR;;QXr>zs_T2iu}3~FkA2@%rSWKA
z=?9@TimbxEQ2-cj?R><oYCW!BbSYBv;Vt&ZTCv}6534KJc0DzFGf9>auqc6Th?$>D
zmzdHYYhK4&y7nPMTF*|VAt_S6LZ2GCFvfz-=tNe?7YE?k*vth8D<LkfvBvgBz@G$$
z(NU>tM9it#&{AfNuyAAw9D(s7dc4ihQPY3^%fL~j8BmHXH>;d=_2FJAB1lxpG?};e
z;icF8#?C)TN4mbcqPchkr`1dH(5d&1LUtV3#fvduEBcpLG6?r{L|D5}5&(CvuopGY
zmYQ&ycKH>RPoKv0{qB(0t)!-4hWc}7G`v0ZgQ(Cr=0eSR%zyPZkBWNGMB7!3?`jqj
zN8!ZQCnaZE_Fj&7bugHeaitPU>r`h*Tf8rSSKjj@<IuI>=i(u=Zo>3}Ssnx}wm5_h
z+r=AKjuVK6xZs>(Q$r0{WW}%EUN?4#!0Yvl4X0R4Z^=R&bqG7;r}Xo;y4QZ$I<U{F
z9YZ4zVWgjqYcRFGy-~Ly3d;k!?}v2zcHd2#1rbr-r^tN}P3XxzU~?zL@U<;Gc1v%F
zXor79{~$GKZ(|b;mXzbP)~=sIxirC;Bw2;|$Fm`4+(bXZ0y8xry)#?tbgR&DpiBtT
zIQkw`t<%SD{#naw0*j`yYJW=j>~5b2Y<IR2Egxwzy$;aLBhWH)M~^M6d<5{US1>o)
z(jo^iTCPnD5P}(_WLIF`PkMHv*icg=To1EoljgYTT*<2Ir4=`T!{Fu#I>(umxx`;F
z%>$%XV1gc0{Mb@8yIBEvnE&Uk&seyr0oWX=&V7_zKbThK=Sjb$;%qTP=8+8v7UE1R
zn&C&Ryrx_fYbzfK{~9H1<zXKk6<vrS+wIFWEjdZ-Gbt><OaT#Hg1c;WpdPgS6k|X}
zwE-7p5=1pOdq;%wB{=$^mNK2$6ym+H?DaOc2)eDRuC4XFzsm2sju5A$+mDl23OC$;
zhkE`}nokB_(vwsGhv6)x<f%kJ*qPzFy`$i7!lnhGen8+)J#IFH<yaUF6Q5HI>Ubl#
zZ#br>YwY7$*8c(HKpeltjn3aKnefG#s%*c=7jd*q1?>L2|L(u%jbFUZ*7i2qK%6A$
zQCAk+B}&ujbh&Z!mSuaC!I6T~Cs%m#+!^Y%S#I9ALpQW*)#5nDJ@*7b5HZteFf%*L
zlTV&QYs1ygKI7vH7bur2#7V;X<`&(^f+4W$<CIdpokJ?!dAqk~eZ!#)o^a7dqtiwA
zuIrJg$mUHO!Z5~YK^Vl8W=fR&3Sl=!NQZ7WMrnH*1z5+%S}A<br%|sWgdmRX(PLnr
zQ<5IL1}Te}nq@`mk|F`^8anMRN+sz@xT0rr)FxvrtwUSSbCJ@zhe%>goH{g4Vr4y$
zr`|3zeka@Q+YRgv+V|f3h&SGN8wV(r%lLkYX44*A??}h4YeR7F-aX<dwt6yCMJm}*
zaniA7SX<_zcdQ)7STLVDt2Wu@3#@CD9^g-prOI@hbZwr=T(R5#aXk;|IV5e>ODo&F
zk%`utAP8u++Eix)N)>CGE_o%p!_636DcRfGWou`HcBjSd+jnTp&2stDr)+F&Q7ZW~
z8Z$)c5pk~TQK^=2grQn35yuHj3o}$JWr83k3=_NCS!;BnXf@lk+aar~_gP$;;n%->
zom#cQ^&7V-mwb*NU*h)aHQs&iV>UOoxcb=@PM>*#o!u6{`sFY9_~Q@R-q|4x+LS8}
zuIEy&__W(mdKRQLJ69_ej;$<HtJLXsV(N`5Qo5A<GLm-heSma)Fp6%cL)R!=$EVxb
zWqYT?^=r4--q>WW(O`aYnT?HgX69zuT6@4>{pd%0_~8Y7&%<>bX6iL6<#NviuG8Jy
zVHl<-OW4GMZiMGsW`1wCOBC7nLmQ}8eNG-*Ac-Sd?SNLNh0sv;U4*p$DwmfQ*xcAA
z4rAg(<CW~BP^X(cT-ODoiIRl)|Hs^)J;{z-X`bLuHvnel?st!!WH87~=3a~B)HW^C
zJu`pt$jCf&OTCQLNOf0LcV<;(Rwj#?kz_=WvD~<eUw2yo=r#{W0fTM6#EqzGrSOJ1
zpin3P1@zPJobw&;N0$p#{BmPT_?Y<RFaMrT|NgV@1qp2Vm7CY*nVToqyR7SP#X+}R
z&)b5Z@7?ruveI3l@m<SJJy$1wm!kD{SDx0~DjD3Td$}B2?7Wi$4|OI-uCU~Z9Rgu?
zc*6<6n~ufFuEVSUKMBZtTi)z$u<P(FM-)A8=)-o)G>zuX=lwpwTR!-8inIMb&UQZ^
zJ0m7S=L8hQwZ|CE+1X;0G<o{?5zDJ9P?;C6-cVV?hws0~@BjSgJpbku^Vy6KKKOu5
zyGAK^_ucnM3F@jMrpP<bp5T3t|Iz!*W|jvR4_K{R{_;QmHGlv4XCxB657b83q{Z2S
z7vEmus=7o>Dn5MweSZ4Wk2#y4akac+vsnwkONr<MWRaJMLF7`LbD$DcRWq*}ybq#v
z*VMeZTJhk)LqdqmW;32Ven3@QkuU*fvpGt6K$UhBx=3(T&X7{TXpOa!2X?OQ=&Ps-
zGikJ9vz9O^0Wt}cXxVA4MDGrd(TW)32p&}t!8BTATtFglqZZTnASn@b=FM`+JI`if
z>dOgT{MD2a%|Jhxh4-Ji77#Z?!5Xbc?YgFEFvb-54J*2C1;h>8K!BKYb~LdMvMxqx
ztg~YPjj@`la^xJTDlv;YSB*e-{LEzufN|E2ruM;9>Z*!wzWS}OnbBjd<>jlFY`UKH
z^A~s@&`OgMgpde67LjZ5X7;@&R)*Db$;HJ3>R5Ai)uOa!aW*68M2Lz$D4N+hDH*>0
z`Vwt2o3p@by`k^L^#3Si7H1DgQS<5Fe~Rw|pZ>#dBrRT6q-e-AT%14V(c=f?oLRl!
zVvHkr@wGHc%$BQF$K~ZUfAruNJbCnp|NH;+Kk@3#8)6I?ZAwJBFF-GmLSVh@_=n&8
zjz<sQ;iDhD&yz<NgudsuUw%awGfh>=yP7qheDVnyg%6Q=Qxjt3i{E`&f{x_gSnGIn
z{s5GSZm*Ube)8iV^ZeVdxn8$2o=CKPiw`Y}h2_cPhh%L)8=AV|ooCO`#!7T=Wg){)
zK2~Vsu+~vIM_tt{&d*Rv7E1H^T-a^4Y%)qRw*1@w_22Wi|K;z{Xg>bw$KT--c)T3@
z`0fL#Qy<{BVe|HnxvlOG#cbVH-ri*TeHmTvS&Wfv4f0R<=5BQ$PnYk9nNh_AT;$-T
z<q5LLyOnpg?To!NG1dNeD!1K{?*ixlTV2i%!L#i`v)ivEmw$hVYrYq-TQ8R^mm+gO
z5~jxIz-F^4EBY3#3?F^;KC9Kbtl%|2`{ZXld-9Y|fAtwvEdY)z5vNygF8Sd74|w_N
zHQ#*mf-isf4NYwcv8OHI@A`UEoJ@szb5?xmlvczPnJ;F{=L`Bi(9|{O^En^?_$T~}
zfAP=AInr%}A=6ruQW1hhG}?w`w<^rnAfznJlj8Zemwf&F1^m8Y)AclUO$a?&!;ga}
z^kSCkdXIIE4}S8A1X#qx>dk8&KYGCL|K88pY&L8*J+5lVO7rm16Z&q$kAM6TIR#$7
zy5!<wPKuEj0$tx@a)yxD1Objp8LpR0Vqs}g5EFCfeF14Q@uk9~id>2ozrCD^MM5Q~
zvck++f;GC9Pk#JkW>qczZ+hz!QoBmD?wn&@)hs^vfX_bvyqJ_dZQtPwixZ1E)`v_8
z8FDHq*;)~XMoy~{GkwmKd!2|$O!0m1A=cDY!^*G8-lMg~nBse)gyAG5Dpyg>YJU6K
zZ+Q3F1CpJIPTpq8|MI{5YhJv3%@<#Mg|QBlAw&s$Q7VfGvAkNn_b9CiA=CDrS*od<
z3SAi2lq6?uGq343fj4i~eDu*z`S~Y5D~pLti~(if{SSV|>sN1h`QkM{{mGy3^G`IP
z?+LxIT|fTON5qu5e09Z}S4+;CbKZUW0p}0TNh$N>v3%vdk96LT?^y_e-+ulD-@Yh;
z3&!&7$x|dd$JRCza~w^$Uw!or=JT(}G0<%`eA#Ve9OTUMv(LpZvM8n6swM;4WnxbB
zNv_8^Nr~3fmH1VrL`<3Mt0hV?ror=G^5*&w#NC2OjGD{K73;MGMzq=U<yX%!Miq8B
zm!K;7TPZEiHiLFn@<T_F6B2#JlCZt}-^ax3%WGUMOx}xE-@e7C@UWZ$Xm4!>J97NJ
zVa|UGbL@%E<h|eO$e#V~jrZ-X`@BQ59q-f)$7F4`y*sdbo;c52m)#4WX2?4R?uPCA
zjNSMF1ycM!s}b<_@?91=Z*#8wbL@}OV_<q(?=Fk0`}gXHE%yR;F$%*a@+Fq0p3(b&
zb56j+P%i}DfA2k>KYz)`Klzw{^iTerS>5p9@o)J0o98I4dH=%?S<D-L{p(-z>(72o
z2$`3cZ>X%}AN}bcalL#6IdQ#OqqLZ+6bg+*+b+(VB2p}qgEN{}*DKCvXFPfQnE&wa
z{vF>w|At2oFZkKdKH=v-`zh9#(S)ds1#3X*arHL5H$!BvLh<9P4PSlz4WEDgEx-Bf
zYwGz^T+W5*tB6q|TT_gL82PBG#GIKDFTVMj_3||z|Kxq~FU*>l3<=Ho*#c`a#)!nj
zk3M>z#oSU^VW58d#TUGI{w;s`m;af+|IKG?`asH&*mYZTCX#gOYPpmfFL~YoSZ`Wl
zVNeu``Rq*m8nb*ve2O?LrfQ=#Kl}J6#CAi@TdhA3HqcokzKvS*$tNH4<?p^=zBnVq
zj1PgjsfAfW)AoVoYBTB^K700zFTebf-~(0V2%%)BKp4g3V8EFvikyndnvepsdBb|W
zMnTCptWm~bv}V@KNI}f;eZS%5*VjCI|3hBA{1%lHzxvguym)oVYTe@MimvNNX%?R&
z^O<G6-jGp*AX<UOIV8WW=i+RJGmd9ZpHNi~`1+f#34JCfjkTIjKm9dV*DKz6`i{5&
zmH+{iBKCpv#TkG1%U{WIL4of&g73+yMEOQ|@14i|?597a^PbDgE7ZhP9YUbQuTm*!
zyOsw}&iVTJ*CJvaA}?OOCdR<o*%{{-4+@(qLqRJIDf8!l{wFx2C1SOip_C?-fEID$
z&f}sWk;Z+_d*<^QZ!TYR_2vy1&z|z=(F4}&HP63!QG|~ZpMLr&eF)^7vCc`<c~)4Z
z>Ds`{7niIzJ+pd7N}1kwJbC<xciwv!5Howsi5y|j*-0spyPgmOMj4#mnwulSIE#!>
zOZ0&!4<C?n&#PCj@AEmkwVe8}-}b@ZnYd1XxcjqqM<#<C7@gfp-!CgL#UOWU_ogd8
zJ;`l7txwdsb$;B3mPn+2S=0{O)FA6L{C&^#d$$9X-S)&+`OpM=^sH(Nil_Bb;PAHI
z0XV-EPkLwtJ-+Q8O$?W?->>Q4ZwrH_x9PC0yBK0mAAR##@3#G>H{h-{_sjiEu{q<M
zqwQLf&pdnj4(nD-Pa()ibJmfF%$f$H6`y?ab1u#w^RNH)|IBZ{{F3#iW7W1ic=!;s
zWx3j*w4&>Js@kEI;^Cu*ynOYXvx{>cJbJ|C>o;7S&#0Zel)-yUG2Mku9QV7n#b~&=
znDfp%Px#wk{xz4Em&|5!e)X%r<H_TP{HuTY&-wY!eo9QBq6e=f+7L1A7-I+_vR<!H
zM&o;rQHFVA_~b{=&^hqgufO1W-SP0@Bk>0-J{uv3uU2-#YA7wt;;d=-=Buw+uU5<#
z=Tx&Zl&R4f6cOtz%jJ^SFJJQTV#WvWJ?5QfPx#4)@1vsUlb`&EzVCVY`cgnhOsE*e
zA5RN&>8v4DmQ)0AC6L5olmwIvLt7g|(=;@*hSg?`vyR49IGYKv14I==gs8VZbtwc=
z)`Z@REKFT-HeYbHT%ilorZqfz@PJpZ-=K29XrTBZijJVBTYHSvXsfY7`l+=5WDqmO
zx(%4hk-Y%G`FugUX-Py0eo2}0^G7^;`Yv_TppD_;;X_t0f6Hcb&H00eTz~mWa`2WQ
zmYgyg1A5S1On4t5YjOltwBGvGlY^mMt$FjZC3vvzlB$~FyF`i<xf6%84V!h(a@`V2
zsyrD#9t~IxA=R{fAm)V1iS>F3g;DFe7T*Vc@}m#&K5)HUlAz=*4qU#z1TB7{4=&CL
zeJ44RUB%~Le8F<P;l=CMK<442$1L7lVXeg|LsdyVog?o)eFP@)=IRZ4(}EIKO&ejz
zybs0HZb&f{v$+$d(tDbEhIuF%HoC6o^7S>Je)<{K%>YddX_Q=vy&osX7=%qr#YG@1
z@WB&fCZbqvTI#w6io{DSvO&g>5f-#-19c^tI0pgrw(GIZg3_cc_ebPbEEnhJ_xYS1
zEz=2i`Z-tA4U_FhHt%da1!?ZHed-HzzsY4fX??#efl>n!QKJQfvTYiSum!e<AnW)Y
zxHAoRqU;{C1EEI7OO2h|?ZNcv-_)<~-X8#E4xe(KRNZ0m6uY9t2qsmU6FI2=JKAQy
zhjKfV`7R*q`x^b<?cD!8%DsTy-~atDdGX?F`jUV0#Y<SPuQA5b_ZzIW^dS;sk8z2!
z^NQu_70;jlj^*k#ZF?omOIEyo`8ChK{)~2WMdd1t%6#zAyC~pq{^l>ay1vF$j>nHK
zxVnChQY-%87atIOEWTHX2j>gYqT$^q7o-@_<|(zSh~X9g_#gcgvSGDq`R4i8tX8kN
zIJ5ly-~DH@RxIWXRvUa2Q>w9!7z3gA0vwGgW?Mi(<!Y2xG<Jz<BD(J}RHN^hQ<*+`
zd=~-H=$ae2I0ur(*EPljvbiR+=9@2mMelo@F`S=Y@a)}p(Z1v5SFc&V`T}xhz1i^Y
zI}0uz{6c(;v}V0&`Q7inqW2MHguM*0m|9D)0a^U~w9;rL!3;4<z(h(3Z47PS;he)+
z$E>ON@PlW({`M7q-7;hhQDokEe8JCu`W`t$Q#syyHs@y_KV{a;S*>T1e>Vyf^zgw0
z7LB6qJUQ1u;=x=q|L_UMRV-$jm;%8E=2eX`24fU++emPP0&4|$8LfyRk&y^i-*-6U
zh$(V?eZ?1Fe8!WHKW32>O>;)`&bug`XzDqC_p5(T_5q#XylFT)JChuaDidNXCTc^<
zV6>#%iYZ`?BIQWC0pCaQajYT67Of(cu_)`%IkUWa!+eo={puw#B|6`aUtps%?Ro_m
zVg{o%O<j{yL>Du6PKiy|@~cljrJBw8^=Ds@5K-dZdqOW`u&x?5t0h`xw1&$!*OD?X
zA-;Mf`WG*-mv6+R3C!nn>ZWF1*F3s7XS3<~=J^XYKFA^_i{Nuri;tvp4(A-IfJ$qv
z_&XZI+1Ua+o74G@=#wPDR0gdTb>)a5NJO+&=#+}@q)hdF$>d^;F6T^RFih9=BQu<n
zxD})nG0v39OW0e`7`;_-T5UGCC}X?xp2~<2a%~;6S@YH(#PO4@cHzfOQ|K{M&M9~@
z1!t$2p3_WW-rJaW>fOX}?S%#$DT9sT=KVW8oPf6Rypsmhu`ApGBYXXvHoJ4G+j*QV
zfZLJd7RafF9v*o((~}<sq4%B4^W^`>i?VrVd>x!M@3o*N|1H1GEN@x7oyMo=aOYv{
zKc4%SW4Gdaz>(h#Cj9^aR>6IGbKfP__Sdl^Z{t#L9#i|}Ucl}@{KdbOm3&MXtGQae
zB9~NCDP?>V$9QAlo9CbL`00%2-~O8a_y6Y$Qi?Qlg)O4Ry?=>PiYJd9MuRpM-+fC4
zB9Zwbqm5^|dO_Vpe*Ex>ys}9y(i%m~VnTDy6_xwUtgdh+^3K&YPv2Sa{)ZQkGsd-q
z+!K8uukD!VJcNKMlW*k=)+VwF!Zaxfaj;sP)1N-A(P~EM1Csx$qfY^24Be|M9$d_c
zDWJ9JedPkSj7uby@iAbO=Gi+6r5Zv^q_pJnx)aG4EpM-LmKelbI`|S+u3}-~=fC$Z
z+DIN-Yc1WTrKu}aIaZjqOkFo<g@_m%k%+15nv}&qPAM(I%Q>TUCWo5m>rIJdl~wu2
zAHU0!cMa`&O;uMc&K5j)^e3bgg?$N$#bQnfzL<JHWW8yribh&0(Z4Y!v@U7bqCB?{
zBhEVP;7^qjs`uo!Lure)imR(7i`jz8)x;Q>FT|1FI1!LeAy8KheX7Wz<ISsA{O<El
z`QW{WJXtpqxMB=CE9Q$C%k_p;=ZoY{#5sdDij+0y=QUMrL?bW@!)cA<`udu#H&m4X
zo)j~$Uk9uStloT0%p%7VqL{+P)FSzOl~OqCxayX8@8!7_@$DD`fAe?$RU)5zFCY6P
z6H!W_qQZxc_2wEy692UXDXZ}tuTaF`i9v#2l!D&3RO^a<KId=#Ys+f6<a)Wnr!>Y;
zOX|Op^+hY3t3)m+XSC8RnuZ4#GtO&E)QYZc$+bnP!1=rqpUc^d^?FU)Y3ioJT0_@~
z3qa1AP2baXUYOEyZ)A~?@ZMvL1kR+G2nAH?EYaa9rNS(FvMD~CG1G2VC}vn=NTElW
zdNStUKF5-1o!wq<|MR@h=pFw|4=x@z_TZKZH+|gq+rPUx=rkL+3&akmrDK5c7)y8?
zK$7b|==%{ME1RdkcQDVvf2T*<95@syPM~5rW=AIf;NRWWwv(;e*>?N^W<(Ck7zGp2
zmCR(!ZHdtB{&JeZcUD&!Hy<`myEM7-V<6@G!a@h_94<k=|MIq-KepJ}wtO0^ewS-G
z#l@fcdkXvRCq#Uz+|S1@_W~dgu#r<jJ7Hm}Mof@eiJwf25=@{~f*deLqqM?Wg)^Cu
z1I{{(HYL)nOjufBtX$|2X{)Fpi;0K~IZIGL_M+RCON1pRf=YPbF`LyWt1ud>$`;Kp
z&<0de!tepMGPuf#|CI{FEKc>}CYL1J0sw)mhKN*!_^Zofa3<6Hmh*Fi)t*{Qw5P~@
z4B0F+F(+cO<xQ;$h#Q2Il}zTvw5tG>Em}(?a)TW+V{F!)n5~V09B}6j3R5DiTm}b9
zb)=k`)nF{l=fY^@tdYDnGSNyu68(DntOwbWS);&e&d=ub!J|z}J(I7GRvN7>AW@zQ
zF`HdHu;oLfNH{`@kd-+6W??w9%4JN5@UBsYK6t`4EU&InI$~`_PLbDdUU4>Cu$a#=
zTJiYlJDfebU~#^{SStYxaY5U6eEH>PJbBvi@X@={uGVD{t5GTu^J7Aa_#n(LT4RhM
zrodv>Fq>I2lJd$IFtXV^B6!c)#TnWtq8CtWlqusR6Vo;=oDV~Esyx#eB52KKvq9@(
z{#0UqR9fSG<a)U*fe0FkzGEmsFRDbm4gozerjmLt(D$K4E{jY^jGmmq_a05p)k;jo
zw$`Y)eXhdbL=*|DEh%hR_faCYvm%8xIaTz%=oM<Ec<|s1r4x^zF6ezk+eB79G4*6+
z(Us+5eohLBoWQQ({A^BLRjig9Y*4J$8{WLR6c)UykcCwo$DFuEBvO(^m9?gr^HlNC
z^YmSZD^b|ddp0DjGh4y$w<!DT;$!RU>9zsT-Mrr3he90qaUFvL_XlY=_xA7{_hF%S
zTfD9QX|Q+GxewXAQ`f%*aubcXy}nKVmTS-GqC7C(j66JxSn4Lm%n(N*B4fQ%*P)I+
z&D~N|nGRYQ@Y$5T+r6kvbCFRt6#TYjA;-JE5Asg*^Y(HVFm;N7x(%e<#!&6=yRHBC
z1D8AZ{J^9={R*50?7Q6Xv>&J5m%A^t-nQHe*v;#Px)#katpun^pn)r*s-U&SYDg)F
zgoQDvlu%Zb<!kH2Tv{?ygd}D+r3DtavOT4UuF9(`C`TN^vqq_0QcDRdl*^JPV=zXS
zC`vOvuoy-7Rx8O)5o0X*D&>|MqedBz@pVuGaD$<uGNz<TGh;&cLF3CABN2?{I-vkb
zZA!$Sf{+Us6IFe!Y)M)_2pkKD3?Y)q0uHsoq(tv~g70uu#bPcgzH|Xa!VHQjYtXVI
zi{_fu8nnV{P0V73?E9`TM~>kuGGsQOFds6RCLjVLAnlCCXo~zu5sg;ce7>NFA;{1!
zZ0L|)P3a6Nj#0T8Xggu=q8C6BLm+P=IVOVl=#+W*;DT3|mt0@Jq3>FrK6%1NAAZOO
zKl=&USaM4AC35oR%dh#%zx?0mLr+MtFmiT`woEagv@1f%l2|^3NQ{Bme8zfhP#Hq>
zq8Av4k31-&S#Q>o?yclz%~6g`34BZ#ZK<lNFi=sVVS`+ovDFya>AkRXm8+<mhOXUU
z@EDbu&E{;{4V6_iPJpjQVNE^qF<OgH;CizrgfOPk%2|SVe7P=XD^y`a6-8q)2~(()
zmYmDh74{aIW=04>vc}{@-?x*6mjV@u*;6%;bw=yN<7W@J2#LP;#4G~B>&>9M7-^cC
zb-QM>>4f2p8D+#2yj}|s%afUGc<+YoaRAB1u_)(Eijl6=4ZcU^Oo*P~!+kzyC(FS~
ze)2uv2X))u`QX^Yzqc$@o*Xkh@6F(D+Tu+;Ki0#$9CuT>+vIba`Ma&H>C|}o_(N81
z3jp_l-n8v4ftAZ3LufKZvD~C`ucv@*dzQmza178*f9(R(Dc6}xmYzI$G<nk8xH(U*
zt8A<mndVV{(7&DYf9J66ZEyEs^>J`=4EXNC)MLlxqir1P$7J9Bi+J!mPk!x#tI2OQ
zxo;EntV+MOU!L38|3k1}zRvq+-0R=5h1Xrn-2vdefSuL~ql7^U$%_mD8Kot<(kd+|
zJ&P$RB}pOVj7+Z9I(i>TsUJh%`xsF{CJv)D(PT+%@5Rwxw9q7Xt3r{ZT%55+G9qX#
zeo`cS^hF9p5qys|MITL@k}@ie0MXPIjri_Zt%*q-{8NYp6fz3%hLng|ViYsb^*xGh
zHVpYz$VVcEfHe|QFXqLJF?JN$bylP)qSA%&5?~jDq%yOnu%W|hGAo?6po^I=gHmRk
z9DEeUMQcMW{gXuM<=z;Fb22&W2+CSa9sqes{iY-vNZ<J)L!w4WiHt~23|TVrkejsV
z)Wwt}(i4eXRY(wlGGcC2N#ldZxr%HDAFDFdi#FL1yrNLFuQ$AW^Nj$nO^308hw}xg
znbU{B)%6vB^B?~c#=uYh=+A_utR0){E57*bA9(lK6GD~<;=%Mc=9~qTvL#arMI4!8
z0+-P+m;sG3MP@@7E^TztRTIBI1(@;yO%ValPTCG3pk>C8#R!t_Y*;A7=t(irG&6bj
zNx*E%X?*^^nDDK&DDn>Ap?o}I6wQO-V<95lqBj>pB<D!X2{-Bq=7|~9j0=hwhg{BD
z^s%%W^$z8jAaXTE(#?esiiV)DXemR8LE7#kT^BIc(Dz<U)FI;gju^q0c2HvS4k3!*
zb}rF^F%_mx{3+Yc6I01It90>^%ws=NigIs_LS=oQ&)M-ZA!=^pU8a6(+qx${zB>)Q
zEy&u|y@eM#y5$&fI^<98-^aTF&|~b#&1L@@55e7C+shD3IOzcHMqVk6kNyze90%>E
z`*%2bPuZ)Y$1dOSJn8l11Y^+dq$zDOXYYEH4kb0(;hA@V<mqx77q-hf-UP%C&$GSO
zyz@>B@aZ@tI`)@rk;ilhmBx1RA-vzkfhQI>>R9g&%S|I{dcVtTFxh@v>$_LGdq?+2
z;cY$HE%yR;eJ297gG9zK?qU$hkC?>lswPVjr6a8lW0y1n00uxa6kXKlyV#3UeE~*3
z%OtB4#WyLJ6+FZgG1^Fu*vb;g#dph>bXyv0D3i5l(-|DOXp;^4U`<_%o>yPYqA>}B
zXpAHzS3~~4(oZb0#G7lSKo7uS2)2*`CVdIy*h}<Y7V}+95vOfQV<!DG+A<DIrAaXe
z1K5b8c+|o`$~{k5Z3teX7*kAmFTQM&J5-i{DMiVXi-1DW<ILq_qR@t-As5A`O&1>!
zEo@cJnc#b3@Hi`ejB$*b6o8PSJOfjhH%b~einF|tj2wNBuzPy+9UKO&DLIHKso(VY
z#wiGL@2h4;H4|nk#K5oq=C8qMe)ebQ*!crCn-w`nT%{o@)|<Xui=?JXddrMjD~T3U
zjKXM^u_~0a<cL)&VN78|^LF78v?mwFa%L-1N+lv(%>349xrQR5J7npIA(reU3NZA(
zha9j*#%&^(NM!L@Rar!|lMkpeu7zM2K3*B3n1lzAo{?P5e!^TC5~}PcCp4K^QyAA2
ziU~c9!q-L{$VwJhSs2@x0`sOu8N^i}8-h>8A5Xw*P>N&}jkDC1qN*!Wsz_u)@KpK?
zr6s*sP70JMz)X<!pckmiif?dJ@G;?IF8Y!Z{1QYn2k#>>>k@dQXg3{y^UHs@#~1Ce
z<nq5mz;=ft6(^573|=_yZ+E!uX1}~GP*Xcc+zd#MgS*qU?^f?v-*0d4aKb+I_p#sm
z=CI@UbHYCjNzD(scG%bb$$9^t-3IWEPtIywluvQn(YrkKQ&JRR3}_om6MLxd03h9+
z2|BEKm*;j0yySze$L_-JmEQz|PT%wKx0(XT{pHnhbFez{>r}bh8*um}cj4->bMGI0
z4ES#M8{&~Y$IJbE>{7%BCAsPl-!_owyf8^MWt1)xbrv9}5OXGAjmVS?Y!wVn?2eR0
z*jFiqu@)af$v_}Z=cf1(r4(_l5+*4Ibk-y?KE@)TD%l>Kb!_^Qe^o<H2{(YCf$cM9
z^f@wPnk)pU6f<kg!dMNHwHvcXWI~yc21dmhBc`unCQT^{@XL`tc)XWD3#}Eg?}`CH
zUPPrd+4saEN0Cxub0z8522eR@vCZn5TqZe@oQNaff{ZJAZz(5QA0;JN5<p0S7b$1-
z2o{QdSW5EL@)j_tVgfVPqSF=xz`&X?GC)IsftVe^YJ;&x&XDCEnM&FVF^Lm=E=;A;
zgyLUC@k?~hioAuD`)aJEnKvvRE?8f`;l*!%$+I8-A$7Gt8|d0K89hpNXsx&GV=)hV
z-<9)5HhrdTCD=n7gVr+6<$0*lAzvv)E)mEfdK5yUrWmo#k&`gSeb*C0;B2uN$F~4?
zRn95F0b<rK5zWQC9g{pyr>hYhj%o7>;1NbO<D5KS=VY$RWY%pf{&I!6OEF@VM6Y%w
zxB|r_J)69f!y-!?i^>$A!$ejR(QB*~e^e|fOM*SEfm}H6oF&g|lyU2{AtcE!thBK5
zp#(h{qX}Lj=e0GE6VZ3$yQ2ZLTxakR4OB`|LU2$j(KH4yq-5#(KuU(Hs#xi}eS+^&
z_W6qAfGr=e0;lWl);o5dV{IJPyDJlRQ{B^mb(d*5+<O{i9_#zfAZ{qP0i;uZ@86F*
zP0_mtm(APj)$BgMW6$eu=NfA7!Wvodvh}gcTY$II=A;v1`GDjhM~S+hRysavZhrP#
zC~+Ki9qY>B-^T#U?%g}x&dqo5F86o8AIHJu-68I6i=n%h({I3SfO5aL)2HL^(Sv&d
zJMTTsOq}*}p&Yf=C|xE1BW!{zla`plN~VFqUri122TCTnm<kwFV<e*Zf%RfOW$+D3
zRL+g&&yW~RW}qSF0#r0wiC>W}OhhJ8*$Q18?HeUd+v5aRm=x<QDGV=&r00_U8Z=vm
zQP?$#IIbTjMx~(7;=iObP$C>bQdl`x6&6d_msHp@rPLTSp_RtgHDm}Tp~k4=R8ls{
zdAhzEO?4=Rtt`ffcHJm6oAYR%Q)JKzpTuCR7({=CEh0v=DXYnp$R!1vH3oD|DrwA;
z0%Z->h*P_Bj*J@3mRbWUk+T*PUGb}<09UOPUEgzc=~1SlYu704Sg)^n{pt$Sb-X*T
zFjiC7H413Ej?J}*FxPd>2OoXJkALr{Ts(fvs~6w!&2RsnXCM9^@4WXTzWid1kCN(6
z7}rqz+=Q)byPgzfEV<fIS&duFvDT5ZL}EHC09int8jtPzA#z!+b=7Vpe{jiS5z1#I
z<RqrlBJVK-z{HXdI2ZGzM8?LV=QaR7P49bI7!-C|WrfPAJa;eemXSarT>v3vF_g%K
zRm(UlAgiitl(THQR%BK-68UVM6Q4JQFv?oe*hy5kCMWTIRZ7z3SuA}FY<iKzsZB*j
z6-`BjF_i#)CH|ejR^u+pwS*+wVu~0q?{8H^y;TYRQ8^M7WEG2~i!Rq^BzUphL#;H<
zRRk~T0kwL2Q*2h5eX|do0&I6>sSb}j)%LLEu58|Y&vo~58Ytcj-fm|MhjJ5uyARmA
zJ2wR*xEs5z#_!=i19trWADgNVpJxQbrXwLrAsYgQjG_v9YmSA(Ad2#18sh;Fd+<w(
zc^FQ*(e01Xo7t+Jvwo*%c6$41fOpq&x955@aC<vYcL>Aodo?$;dtdN+9~Smrzz#%b
z%h(D^Mgz&SS6IBVf|rRjk&7G&Stc8;WPfa9iEu4J1X@d0g&503QdlCy|40USF{c)P
zDV1UgrpV%84Z@5mG3OX7;pE~MBzaL2fJ#^;K7m;wQI0v0VjwGt=98#JL*Mr}XCxXG
z5C&3{%mmab2>_umYPqmb3WfK*MDJ#hU<zT=2C!g^E`bPYjDSozQyKBq65llOmm2c0
z4*6G|lYF&4idi$}gzwrBl;pA`%vnsARV^%?_<;3ggl4od7^jPqd||iJ*54$iNFN80
zVS_RPGE>gvm@x{REds+z2B7cdImWEXXo44|bl)dly}ssu|BL^^@4kEq%HU%bQ)C29
z;(z$FKVbIG6P$tG51A$UEhCvTtLtk%|A*i3kKcKZ_kQ#V-#q`4b$iYGAN`!)e(?{a
z<be_(kcCkm`~kI==-pW}D}fz(1hURrDpwI=WYe}IV`~bCwIzx-rHt={aZ=W>={qvT
zml6THTB{OdA)rR(OrJ!^dn+|kgm}9SYxS5FWdOoj>+M(_GD28GR+>%MkMEr>?_P+3
z&;}}375_I)QjyhWg|@}qoifhVBap&oMj>h?-Z42PND);6adNqa5PSL<h$Zl(*K%J)
z0!e&r2Q9^n_`r%wtlU2}1UY5N&udMI0M{Z0F5=OC{CuUt1m-fHoe`6ID48x|l=l@>
zS%}=-RB^ft{K;L|ncIQRF&^b^X3)DxqugBYHgI>-^&JOi`QX>#wVY-h2dJ?N%68$$
zcV+K(|31!K?etZ_pfMuAZhIf6p#AnTc<9NQdSHd7#UOn1u-K5mKMmwf-UW8UKSu`3
zE`U??SDM+wLtwlAEcgEl`=j8tk#(&0ZSd#R-tWqH{V?_Ke*EDZa@TX+bjA1S@AUn>
z&vEwxc1dM?%EkXnjW2{aTZhpcCw@u7Flb#gvr;DI0ENLAKF$LrEo_t?EVd&(u0d4s
zOT?-&mWXKSS!YCmx8$VCg~6~^z|s)u8ly0GecRK_DiV>DBgSqgB8glz*vbf#5h7YS
z0aHf&&vIE>8DmQq^JqfKg{>JK%)K8~`K@)LtY4xueeA@P3y?K^-&5C)RH9ZpV}*T*
zi9F~l4QxabW+=vPWPzd+2BZ%WXC#9}9>_?oaK;jSC?-K+Ka^4=9}9~V$xauOo*FZg
zWiovXl8(w6R1lMR3=zrJAg0(1Bl9SFf}$;VdAZ`{izTbeme9j;+4B7IiqF4%Nyw1c
z%2SjnUzi{3XgAkIpjLc1o7rp}*;m)sJb3z$U;dZ>#Gm|=f5rzN|D4O^nw-O!Hy0rv
zM<6U9N|k3PV$wN_Kb^Cdq#`~>s;UxSGiR}YRYO3E+!P-p!3S~%XB{Xpi;wI60mK(j
zH<+YDh_qcRtSzcUTttab)`rxVg@O?VP8G(rs%l|$(WCs2F&gcv5(F}cE=Rgff_jt^
zmNLbRwPKFfUKmVWf>e4Rw}RG)@1wBa5@eyY0C$B52>TrINiv?q6pOonF6IFNSq5FA
zZp9BM<t)H116rds^g+gl&4$j)H^5ajIV<|UFJoO5*A7F@5|tfeM5`bHCRW@EV#zUV
zj4k<TzdJjyErS^Hq1f*}Yc~ZcyZd%O#E0|sA!s@dgbqR5DG+(;_|s*dcRbDz%J$>n
z?euw10j^``InM6gz0G5PpI{qhqB(Ma$e`@&7f!*<0LZ5;C*Tsr8iPkOcv)w3;a3Kj
zxc!~UgtV)A2V2&TJQ+6wCK|RW>HZXa?vIX}Se?5-touNi`>}U<GM=Q|&L-Yo)IqEF
zJ^n7`{yg1H_xdh!IQ!?_KJ&Yk`vJQ$`b1hoq5=zP28jP#&Jm@=hB@f2$pj=ONLL)!
zeaT)i{5KZOI2b0wVs6p4fQI7tRag{_q_R?39ONm?nMRYMr1%nGDIhb3IQqT$Vpg=y
zjrT8S2z{ha9$QJ8A~BI|Z?O2d6=(U(Xu4{O8Bc2kNzwU!oP_%>5Uj8X3JL5$NrY1h
z32R)@9<yvV?dV50Gr|&fT~{&?)TjjBN7mhj#e630T-QmqjFecdH#lR3%}e4FSk;x>
zyQHWqNgj`HF-lW4Gg5?gyA%^{3}q!bd<12}Zu<@!B~n@It<;NvT0OE_a)<!gtIG|)
z{;yxrE(0n6Sz!_^jKdj2?_-JV+6^mz{`_0k?>|PDHrw?EZLCP!7!k8=yOuzvHilpQ
z&0q10fBfe>dHOED`}{Zf-jj1ABY_FQdlZTo`@)h6STH3SxOJ8gBYi7OTHA>f%*Yy*
zG;@Qwa`4lWY%4t|@CeBBp~$)9!rG3JYqIpVt|F$4DPq?{5Yhma2cJ4&L**W?*K4#k
zgcyn+q9V1z5URpnYpc;(GIZn;%%PBAByF_36NU8(1rUkL0ps?%-ntin09jQ`puP8m
z7+GyLP*``PHMQ1sZ6_dFDNyo$^d)P?@Xd%Z5DTDGT8T(@xjrZvG%~0O8#nlhit9oy
z+a*_Z+xqd^V^Z{4GzoqGedyS28N2D0{mDq3*f#|*$M<bpz5D7v?;U#^h}rGKe%bbM
zckgND?*!|0vTgZwf41#zoYHOWoZNn*;x^dtG&mk6>goP#-~|>5NgAM=&=I2=>5vy}
za(rsGEZx-4ZUArtz#Nr1GTXqMv8!d#wX+Dh@iKR=WBd2vI5|GnZ*KE0+mHQy8vNY7
z>@UFX-b&v2IR)}>Rc>iwfVt|(){`agj?>%9y?~wgYRUgh5$IJ~2A5G<eDCtapGxa7
z_h?q84CEx{KVym$xE%v722HV)5>3+-V37ftEUb}p3^%0fddUN-HMy`w+DNts=Pa%g
zGiFMWvl3CMwI-Fm7|f_iN10q=h(+2$M3hB`ZeXMs1!hwmG|417ubEX;+5wTe$X;k|
zsH_uHrZspU>3Wfh&?wH%7UG~@NkE3N=Osu%j~aP4Ngw-8^43~wxTwyEF_3d)F>A)m
zAM?5qKdr)u4w`$0A|{<B6NJ;Ie15WI21pQlQDjd!?J%IBFgvLPP1r0ctA@Y;QppnF
z10cY}S;ebwUlEp~OP9jXHn^O4(9~?g>LA7^<;)jfzhwD`nX{@Qlkp~@NZ1VLGs!v;
zg6Fqi{Eq+2zxwCAeDRY1{1^Y8Kl|tZH_p$`G1`u4u~G?w&?S{x8N?VR0?HwHV#+1B
zCDC`SWJAajT?z=(S-_7etQ`fArkqQ#52&Q+`hEl=L;5TbpF(3Sp#*<qAH;;3A?8fi
zwG!#w_joVD-YG^Z=jgkBG(m>~9J{WgZX~Nlj1fW!%m{g8mb<Q%g;cg9YgpG68Ic^x
zc@njqGwt$P{99uzIfm`%N)c10>pdAIezZzcS&IZu6xKFNe%#)Blrs3P!&z5Y<xEo0
z_bpZBsH&Pi6abz@?{0K!h!JaC@!14xMc>dEQ{-v1EQXTk?8Oq<n^FROZ_korU%B$p
zJq6mPe;<BmZvt#L0l#B^A8+CG`A_xf*s+{A*M1-OkDV5E@b7f{F?Q})z0>>m&wq@K
zx+};VpX-36_V(>(vfxOq{n9awP2F;O>c}WaY_DQ{4$T7cHKQM0O1nUA1bAiM8GD)~
zVt&}BkM#2{Kx}^u-2GYHt<GKH(J44{f97tlOp)^bnU0+B^g`)70KL3B_K!ihBM04D
z=Z9nL5~3M3INhgIe2$U?M`X%K72%B(i@>TbQY9&l{EYaE2_Q+xsMcCxZ9QIrD`vBl
zCCPV&F=|$6T}(ZR6g<Pc?Ucq?BmPfX7se*DYTFS^N?$~-!xlZRj5w}Wg;C0-k0AuK
zvBFLk(5R4!KcrMxNlAy)rji1TTv8Kdl$FTP45XZBbVUrYq+S}bTx2R&5li$geaSSC
zB(<2e0zM_ebS8OgmH6;!t?+%o_dQA*VFZB?icE}7=tk0S#Z<&zm1L*rx(;h<DqE9t
z(H$K0-VoowSu>}Pvr*nA6+g8YBfgguWnCwFWOZF(lpf_)Lf?|QSlSe`yVIJ-XAQ45
zYuc3d-ZjOl5B$a7e#)bZ1)H{Gx!Q0xtN7@>X9VRqzc|C$ikLiCS4)2HkN=F{eEtQ$
z``wp(@Zoz{=LlgVQaQn+RWaA+jLPD-=Zl$dxm+?|%!t9$G&Aut1p2Nm=1*1RIAq*a
zbycD|2RV_l_EHK|k<&2-XN%5Y(F2^#W@18hj0*yzbcr4=`gUc^52oY3fMsKCNjoQZ
zNc`}uh$8nr##)FHeQJ!%63$rc+~PYAnR2W!=f3C^nvy<DqRzv}6pOi@V$x2;!kTw|
zPbzJtoaj<Z&XPSM#8e`dOV*Q|2+`w%xLzm}F)6w>v080#PJmk5wxc<DHlLGortcG4
z`NCr4BIcUueJ@}X8HaLgkarRRc0PU|K6ayM&;d(x3Ycvt$wR<)UodwFy7H0j``c~;
zZKvVMO(xizCYD`rwhxF-0lmBYeF(1hdv+Teb!=O{tx_H!*|B~fySu|ZQVgJQD~h}C
ztyjQbp0IQp`e&dcBm>bwyac<F#nG^4R^^|Z1@sN=h-1d|UhR*#Qx`hDw%cIqO~>A2
z`yEbmnsJ-fI6ls9E&1Tb>9ZWIa+mvb({cj#v&?LdxM>ME?B7n1KNw>-03Ea`dR?NW
zHkv4Nq-kd2@70S5Oeyj6sq0#t%TvVnVt&m@7$N5@7)egARGyewwlWntmB(NH7U9%L
zRXMZ~^Ul@P6)8oG)wJD)v$JzaXkI{IU-ZUe6e$MhtWex)n=jTG%VyJ!=jpx2$4KKG
zPaZ!MP?wcBks~aaR%#@2h83LXl2Ilc@k6slqn#l+LyVH(J%u#NjzlAZvMMXE7KNdz
z+!%qXjRb*QUta^llx*6brm4}+QCE%>o%n+_C8{z-LQJgJD-qo-09h8U31=j*C59qG
zJbWZ_CTEErZR$#ZTF&&LCxyaBHWiA@`NbKEV&;sJ&T2lNlhe22Hy9;~%ULI;LDz6z
zIoeIf&fr#=h~gi<dJbPvO#7L49zEvKBh9N{Kj+JD|B5G19`fwj1rOf+h_lDn{P^SF
z=j`GUF?Qs#hV@>Y@wL@#x{j{vCGxoddEa*!<50-hiBZ6DGn>&=^(c((iyxS-2IGo=
z@FAc@8$;#X$Z9rCQ#96u;TrtdMmZQOPSi5KVqxL(IJSoXipn}rMJPCVYUjw5V3$Eo
zL~8-`N>M3Wq-z)jmxB?Lu`-qr!>9)*ZVfU}`w;29T#L~b@4*+-r&gAb6MestKqBjC
zs;cDDmAuEUA7o?<{j$RckrL{CkF_NeN6Pp<vRbuttppS`O@sHIw(ZJ2+|c(PSum<%
z)|WAu$#i{R0<i|kqwTjLl*L~79)c%Mn_|`E_wL?ftlkvR-381Yf~#YoYrm~ypl^Eg
z;qk}HaX@qU_c73OxcxXdyF0i$*1PHco7s=!;O?eCi0$*<0*qsKav#KlD>OwO>uy1o
zxU-b)xxl}nVgm0#J%*CRhIlOjW>JADnsvqqt3QA{27|#-fO~3*@ck5ZqRG1~Al|+l
zYdaroKJH)mw*9&bfINMU-TQWI33Z=#e;{BdjD%=f3A`0YW~DVw4eIux_~~p#MsrC|
zBv~RDB_MKfgw{nWMQb&Jx`ACA0v1B_)OC$D5)CSj^~K3MWt2*sEoL(LX-!Hp>3QFa
zNUd|D)}FPJI?7m!j|G^fM2?ZhRm9@pJvguHD7tvP4-yTjHE7XZ+pN~2QC10n^g(i2
z>Y}5U62dkrXlzAH0c8zjCuZIfh=BskU@j1`XRVAdGFj3<dG8s-V;O){dEB;W#Whut
zoQUE}RXa<}aDIM{V#p|vi78-B1xc9Hx~}Ng5_q8s>y}ey)3&TPouuDV!q%FsP*Ef{
ztTD8kR<z4P5Dmj946Deeh}4bJnyiGuwF(w9!(|u5kvNyg<&x!k$hrWUtTTOzRI@pC
z)9}mRe9mvb{WbaTGXLg3{56jrJmj5c&-nSze#(!2^aOH;QHHvn6(&~tSUD#wnUQ;%
z5>RrhLM9S1RXx^Gg)thm871;FQq>}rqjjcst_bQv>wD3q3o=$j_LtGrKlogBp{J@U
zSp?*a@5Nu!hd^D|XeFZ9q2!SjL1lw;Lxi0ob|sj=NU%j+Qj^ImDPRAX5;<ws-qQs!
zNrw=MFRNw9DA9F2ZRhDbF`@Qe-rZFEE5(#rGF&J{Wo!DrE4grGtPB}2e9^O%?J}$f
zlYQTdFtJjK+y|CxF<T2k98jufG+_+-f5o3RQzpPrGL_^hY}&H-w<Q6(TN(1`?wexs
z&Tn@1Z2`?8ce0z{=oGIq{bb*jmD^{qcH77&>K;y1r-7UNIt}!00&&OBb`#jUjgh;p
z&Rve(sXQPpwg3EX)^r>8HHbS)+J}FIRTC+o;naED7nO*aYYVCe0%K4_!Eal<sfq+_
z5!llvVN||_T{?vG$Hv*QEqCO>P91(r8$T5I+q=5&^)%mCW4o;5jnz(__8wii7qGLg
zFi1*FdqZ{xMbV~1KE_q;#t6J2(RnNY${NXCs!*V`Bn;05Od#NH0CnC=E>UYOvsqK5
zLlWeK(wW7qp=lb7lANR>jiSbTXso60tMa0WzfSKxUALij70x!o7UoR064{zm6p53{
z7C91y!qIg-UDpB#tA>G?sOK}xC_^GXNKK=QxUQ2Z))>kpEfbRn(@Nyxa=D_eYUcBq
zXsA`fWQreHCKVqqltkJpCCpEXN$y=qKbBL%72iHGBP-f<o;F0@EZ11$sGFLM<YKKV
z15pBC5~^s1sges=M%e3DuUPfh^j$zX&Dr@(q+?2!f~$VRa=GU5!w1Y7x!2BU2;gc%
zxAw#go3`ialQaIO|L6ae-@f>QkNk(cy1u6G6W7-**UOf#zkbc~x+RvySXN@H75ze}
zlosYHXHtT77x6FO@Z!}qUw-)&|Mbs3=1>3lBO($hs*7w)Oc7I}mr;ro0<rjd4*7&#
z<wo$O3hNg`C}T}@<ZKCYm;#I8KGn_<lPoxfT&xiZqR>i2dDm@6+lfZqtf{FBz>G1`
z`9RDW=R}-17oXGGRyeD%*3h+mdFK>!+eq3xEA9cC&4$hg`aY0iruTui3vAks$d*Zz
zxlG(A+P0mf=rR&z8n@m+@*hRz8d4I|yU%?IE>XpN4@w!dQD|$hNv64}Ph-j=Oq`4V
z-!SFtr12pY^YkDeMDe|wFm^eWZ-dNkMQCZ1*GVyMRbSt_$kaMD#TNLu1Js6^hv}w{
zMgQ#{v%7Ei@9nmI&~_X|P0xM|<Q->r4)+bu<PhZD9JP8H0Pf409RGb&AMXM(ZqxQU
z3etuLJN|e3q}4%{Lr!YgSCMff!d{&VI^GK%360l=7DrH;Hfvg}EE|gZg-VcREwwe*
z9^45)3mB}LynD8t1InHO^_6S98B8B3C&Ahs-mf2$xO>}^{x|R2iDlPK58+fFGxz$w
z-OtA^irGUeQNE8c;FCxjR7eD(_}gi0VKY=tvP41j)VvpzH3S#LvI-x3$*eWSf6S9r
z#%P209hGxrXNl2|44+bp;3Iv}BNIP8Mz)6_DWfD}&=GxL-c%w;oChC10i)K6NNQHa
zhp(_^3aDL0WlRxc*7PZoiV1DdM(cY|C_Y_AYkZQV<kncOm)9txnbq|O1VRXelu25#
z>Uyp=EvjTaFh(=4YSHZ~DW(>S1y!R&?+eit6JLH+A}k>!(UmjS@$IWOC={#Jny;T<
z(g&G1o2H?vH4n~bC>1%IFSvMcUIbwU(5RAC0If0B;qnY)5>;(UvBy<ak%y6qdj8}w
zl@nGhr;ODGHB&4eEx3HWCMpr^Ty-0seejI6dBa((dH<uwW1vdTuvvOuzFP9vfBQLq
z^Q+$yie)W^Ku(EuF+0Nc+m&L(RSxe1XR{iu1oWjWepv&nXtaRu6cb4ao0jshs1On4
zlp>48nLMMcXlyM37$s%g;2$Th2POSnxd*2Ft{aK8?uK_ECA5*;#<Qws*385wEqH<#
z?Lws#^O=CBlsqw%e8$Gm_bn-E`d;C^<XLsrQ8`-zlb~xo>y77nwc+Y|g^wb$qqQsS
zvBFx>=}Xa<=w^9GQ!325QX*1q6<SHOYK*CTgAzt7T;niytIZ!$CKi{7A@43lM{<ZD
z9wbbRQDS0EvIrc$Q%WN`hiTv{bYc2s`c$ecxJsrNlp@6~lMh({<s=i%+mx*-_GZTJ
z7`WU2d;fTL{=N-}<(*cKPZ7tWWv5V!8-E|xJq5;Yj&R+d^pBtG^zk=syS>iw^By}!
zkzs#~3|L5QTifMUPEV!=m=S21N*pU%6m?G2F@la{<;RwKWw<gmuR_ggrAeJ4MPa0e
zpos0r9@kdozZqATZ^qFpKEhb-Yq%V%f1h$AQ~3Ry;=g%0wb~lC-RT;p6>p&3?%dv9
z$sCT|8_K;+vE+;@QHH~$HfUpsFOfwRe<GzsV@g){d2}w<K(e`*#Y85<_uUwjA(5d5
z>+C2(EYXZ0&e#G3`nDtIP9#(6ipm*L5HhMn%*w=CHx0h;um#v8gbDC{FVS}lrYDKg
zb51fiOwG1rCFzTD!qpCIZAmdE6JW}TrmDnDUEWBUkP?e#MxVswr;B*8v6i@Li6P*<
znAV!6;%wfCA6QvQdM|zI-dq;pTTS0~Vor;R6awA{VveYw5%E`BOos5Fx}a(1k~zQ_
z0VyhzQlN5LbkcN&wEd_nrG-r`;P&e3nyRvBlc}67rfSKUP!*x%AstjmQDkR~=E0+L
zey6{cRijpH+Lo6uUh?G8Q@(ofJ5mbBL@qKkO{4hm!v~z5|1kg-LFvA~{`_06R~=b_
zj{#d*Qs|j27O1PMGERn}5osF7(`OG!DU|FZZuHl*)(}Hr-HOA1Oqt9!=dCfqt|}$q
zK2ZR3l89i7MQIC2CJ<{a&TZWWve0~lB3UBuMM5iwfD-{K&ZHuetuT54h!13plD2Mm
z9hFiTWN{I4qONK((9{xjD)J+#h=NPB?55q2VxX=nYGnLtYjLXhN@~zb`rcHAuIuo<
zN0;}FEUEwcP;%ks3}{rzvf-S=+Ci*%>l@g6Pl~x{8>(^KC|CRwqug(6jQ~0&zO7x~
zW38sD9I4LYYgckeYqib)nu=&S*%a||3F--6+$Ca^g_Sk-d-1W8xArtEcMPZ<GI{&^
zjxl?O^T+<NyM5xw_T92|=HK6T3RoQja>t}IZU$}BV~75J7l3#dkha^}EhU%xeB5ri
z*W}@SAFk=xGaK&947l8bflH;iG!^f}qANM+OGBil<J={F*)QmOnXhvpoOM|rjVdYA
z|7)mcd;H*d>Qi^)E-2#INW9_LgUz>`<4*PO#3lZ)CGY(n9{8Pr*#5!$<t?vonC$LY
z?w2dpM?4<aNGF8>5vc>okE)6;R{={UeU(H}W{eS;l$=Dr%asqeF&Z~(aFsB2DWx&B
zn3_12XEL>GO5m-4CSxs>C{63EC<-YsjgcS(#0;dhMAXHQB<Mj8eqZvrQ3x|+tP!A8
zUg$msF{g&YOj=6{5}`V*uEcaH05wI)%c_mS2kBE|40W#XAreC|OQlR@9q!zTKblr#
zRWbk+CI8^P$iSo_IXkGVm{kqhYH~F(c^NH18aYAAJbrK{av0vDpXyOJOerlv5HVtn
ztQH4pSJ4_&Ba21O%w`RbxF|r#VznlvfOVB*bI1uoq-$I8M;x*zWN7A10qcxQ;uqh(
z;Ac;MLf>{&wWF>qZP$(21(eoB4yGKIVb(Nda7z?sEDHpcA;%0U^5c)5u{f`ZVNHnQ
zFWGio$v6=ul2m0eS9YHB#TiOv`remuq{aMR{MtZX#oh<nEGzCIFKjH&85m9O5=@W^
zurF{x!5}O=ELbv`*eA*OKqAFRNOAJcsv=Yj&KMeH@xB*k+DrBkQ}o_ad5<b*c=qIx
zfU4q`7DB{WS627JocqyTL6g(G<XM%hA1Y<K;OY7(<21?`^dZs5K<5KU{CZ(8^XRjh
z%g4`_XkbvpkVJRS3M)7WCWjEkT%01lR}9%y6vD(w>bkJ?t@Qz=hxbd7QYa?it-!L@
ziIl{z^zBP|R%|ElT|g)Q9)i7n@U{E*A^4ltWzyR5wg3L_?cck`!b1=>S+E@(e+c@H
z`OO_?*Ny?KL$mE}8^=KF>2n;fb9!<+iQgvQv+Ybh<;Aw`AA{ego(Bv86eYhl!^p^7
zE6qDMT!&Q|g5-ZCv*Tm8kF$n9n?=I8IGNzT3kbWd>>qGjyZ1Q3jlkqC?Wp4e<-0Io
zcPY1Z_Xc9_WMfmns3r{@#e!;gK>Sd^&N@poYZ#Kl$Cwbo(nRljs%i@8<kdH3(5&-C
zJzk9xDSZ-9)NMqOKgNKoOaX!ct2EV+<pJ?+(?vH=D@ENj6CfeX#E?_fpbK0PvygR$
z91}zVN45aCF_&iyViGK=ltP&xt#d?`!HO`evKlv}M=LTCrue;um`IcqWvNW;B_M>P
z#@w<>2~(l8m{$8<0z9&k2w|l()+wBIGKnGnUR~d#&^$Ol$GIX7t0jMGA4<jrt#S1L
zW|TPHYK8A2%hj4q+oFvFqeYe@3DAx{WAj|(B#IWH(FSb{7-8|2n+-XJBIyEkRpYD`
zT`@$*Eh};M?Yf?r1^j#BI1#+~=3D;o&wdY-2sy_rsnaSa{&^`6>{S{;iBXEK<Ympp
z$ITZXKWi0dbHgA0;wKoB(Wb)0Tuh*1&J0QP`r5Xm3xOCiy^m-s!2)gDlF2-Jbb;1J
zf>aRqfIRt#x)PZ>q*NQCI8EP+&Rmj@t|?|}Tl`Ej;>KW&rM4Dy8gmE-z)k{%v^HdA
z=zTk~e+miCNouPjS?NjuAp(ZI55yAS5lhr)+O)!c#gYQ77olOaE;(T{eIH4w<gzW1
z(L<0+jDZ;B1CvQ&e%6Ydjl7XrayG{pF-o?jEcaf%4dr`O%(WoQy#!+nK6AoYYGsNg
zK>9x%vjuMxhPGs33Aq4~7|>add1>FK$d^sV|CYHs4%&8ExZ8p&j<s`m?6#<*f9-Q#
zyFhFT)TVtnC1869`c8qr!@ivMq1*rabla!Poq%4xTg?&E4G8K0)NO-hcFXR)J@)Lx
z@uDn@QKSO$dK9RH7;Gm874}}CAQfO&{OpE)CbrDoa2{<l4<47wWB0xXM00BsH&i%4
zp|`a2*4;hQz`fg?!t_)7@7R^^dOhQVy`_Pj{rQIjcH~Sf8eT&XgbyN=Ym5k=io1?1
zk!(@zA0{p5s$$jzlrAO~(b5uT#Uqa0a;{`V|4NbRgQp8(a<#?m7h^8IZ2?rqXnFZU
zl4S127*X}l2^%#bcq+RUKdmZPbh!%4XT{OI_hF388$^#&D!D+FWVkR&K&cPKK_bh<
zS2-&TgVt2q36Qf&#BhUv5og8Jn~{iF@k0`TIDng+Gb!dVs<NsoVY8egrpWc{D-n{d
zD%R^(Cf*V};(L#_2F<LPbmbbYb3C5UF@^oGR`TImXIWi$#1=G)Q0|BLFa}g%(=O(7
za+b^nPTSEO?F+Dx=O{pX-puH_NHwe3EY}QEgx2EkHk&oHy#%Aw#rH53(O=*D;>0h-
zdyn_N2nQR64`WV|tRg@E`OmmGcU)e+q7MR`XU&XELY1^`XpM0eqd_}G))MjQoEri7
zYSXdVbUe7YC=sN&xC59HC7tMc&w91NRj&B1N!qy_B}JW9reuw<C4HNG&|>O^@pKxM
zWemlXshWCZ8no72U0<_lmn;@@Dl2}f*=LdlS5>1sgw~qeOJs74kq|=}$Fk6h1!SQW
zeTw9)QMLvGDT;Z!>pi^}1}(<oo)A14U0eV}$at!e7gL68Msq9_pE;#Q0p*-C%E&z%
z7<Ulqqm)X-Z!y8kJ0>m@sBPLl7^EPK<cB7Y?1lVJC38hCk+YKs#J4JgDK?LPN$=VF
z`xHZWTX4m+?x|^H3L|a;Y|}Qj?9%R0$62`P??c{hx39M`Z0g|O!}IOdzYWHgGGT>I
z2TANc^pOnNtD%OS>iBWT+uLpB7%;7nAPT?tT=YHw8#^HNVp6hb8Wt;QShTPcL@|t>
zvxkY<tmIKXnBdj4@(tH>k8)=ak$3)cW1Y9?&d%}oV#;pl%T4?D2hpj1d@tXeJC+{`
z*tMGutt`IpaFrReBRD6fNAa;yB{fhiCZoaOyl8r5iB=s}<W>*hPuQKyH%q#<qp2&L
zwK%P*>#9g+Xx8hV5WGam7JnNN&lMJAh@6|v=Xl>2tC)P+s*)ls#3;UdB?CrG5mP`;
z9OVue*^(i}QH&8?*G2AyQI=&8Xr)8|CLhGdEem7Whk(+C*~|zi&8h@XNUEyYOazC;
zTveE)zQ<}OPV`j~YEFsGW`lx~zqZi-gPFRnCCkFZ$x=jj#oRU4FX_fo6sLGHoS&Vs
z^zCHI$z%VGw#AIDiAl~KV<MJ3t6FPFi4+5U%wrY-tuzlGpL6wMi9?eqV<Lva-inDl
z#j@f}ViXQB?wHlAwKR1@Xhl*)72sJp%{xyP{P7=sf-y+?G-L7J6H+9GzOZ2ePP8#N
zCBAUJ?+H<!yAPhaX@pHnnd{}61k|;|S<R3~RTc^c&|-c!s`ORN++E)i`<BX8<mA!L
zkx3=UCX$e>8QyzhO0=B_JUde&v$F)_)Q$LAN)`uIbPn|h@|BY4+~LC$i|c|`iWns4
zZCi9Cvk&+pH4<Z_>w98C7`~V!aySI~7)u`8q7zwC04jyN-&)=qqeN7^GOlFzNaLbP
z0!p-4Oq9r|DpMGWB7H*!8zVX9QFvPtzLvbsq+(7kZ(#|niD?K{(Ku%bqyLil?d1~4
z^tNSS>;?neF+f+|4(`I(9RqvAzQeLLpHBa}n+Wg(K<~71*!o@ByJP*l{rrc=9=3Oj
ztVuq0e@?f*-MiPXV-(xgF?Z+aeZEe?-ZpjQml#B963ZePEO~Ok5e2OcIlgyJmkB}P
zeF^4HTOx53B4*%?!PfBbQOg)U%YgjufN<{3>fHsX?bW@pem?T&efp#h|J+5rZ&6NN
z>jWqt8>sg>|8x-E^qjc;ao*$mb}wMJSk%<DD-n_o(6pNkvw2NUB4=W(74h0ieBfk4
zDzAwU2*uf4UT$YCIR*NzgIrNNhbs#9A&Nhmq^=OD3ljp~OY~;e0_w~lErOV6`@To1
zjH`qwH&&0pQyW7_fvhz?L^iRfDN&W$*wHNJeL!0=#SM|VZM$K$j)czh;NlFG!RLfk
zMj|sqpzS+QijYN@tL-}aE;5@nRF(LT8C1y>AwGU-(D%xcQ&uO<yqSw$T^5J=oDw>L
z%1W|c1Tn+;AmYDORTaj>3#;Y>i}{>1<}4M^D`rejT~#G$K^VeRoIRAHL>9~cYwJi!
zo=FISe$YZI%vm1=tPOODXs%VH6xp;rXBV^LQ>akNaPi=R-uqD=$9s<|Y!n$(jI7sd
zs=8s_Zy;x^(KJoXx)&d@p`Wu_@h|_ypE8>}`p`+h4NzBd?_w@KW=auaMrkYBf;o$C
zo>HV70bwFjhQ5=pt;&kk)e2N*KA#a|7_&8KG)5UxFQ!Xl4KYX-6P*Y7A9;2mM%qpo
z&Cai=D@WxDNYonZYytHp$EtP2kl1W`P>HVdXr)93M^+of|7~EY5`EXBP_%uZ?|TvR
zwGM48oAm}`48D(8YiQ;(F0ZcfQN~A8)tt?0`p{EVH9ppW!Fw5tUDsi4$!CmZHk+3`
z%td=J=h3C0YyFtsuBt112n6pX*RB%~=Zqutk}<{^O-`{0bs>wlBoW3wim#tidR%yc
zWbTNFSvRKQ+CgE{zYo(Crx?3K5O))x+h^sr`wjuZK7dnf!P$PDX}xJ1#{u0g(7B6v
z@ZF+n4?)ne-@E5Neh%QIZ`_UG?`Yk8bW&8)NfR{zJI7}3={-7xj{Ef$nK|$+0~nJh
zU5%%Z@66anq2xj+(Xa!ki(5c9;?B~>lV?4Z^GDL59lzd#>$&r5aSVjrQs?Nd9{_0H
zbJyFQ;sjv*KA6IruK%WXGd~=#vz4W3ELumbF|5}c&dwG^Mxo?kjft3LWj+Kdq+|fm
zqZ<5(n%a#F!>lkAIR>hEjWK%E$4XH?`~vhNDF+raH~QLS7)_rkN6^sBXLuhAt0pV?
zti)Wj={Dpjj@?6KZyy4#M0fT<(kfZ&P`UVg_2RHyy9NdH+F<I66cbK~Kg7WNMWe;R
zzvyb6oiFe+c_0Jpm2>9&{9FK4xwk`vF(t||<*iJrHU?YQ5^1T0EgI6eg&4#)s-!c^
zCC!_g8!<VLgRyA4X-!?#7&n9`<idgptHmV3St;>7%qfY$Y!qF&!Q7S+;pD}9fidEP
z7J`W0rWC2Git`6xYmN89>NIsNzJ^`^;^pNPA$T4=enjj&D8**e@$lgT-Yi>&(Vo0$
z(GAhNR%;$TnDg*rj#3G2had$QY#O2msIgXKYym_;d_9eGC@4|afRXGe$~v6VbX`wX
z*C?lj?Nb(Qv;a9}sa=f^9bMZKLn>J)<Xp~FlD;bjk?T;JZ=S!zhs62$f_I)iLS@fl
zF-KcX*S1(AW4iCWu+T;N=fP}F-+9{2nhY!!3-KYXEmjL~?Mm)c>oT!vXxpwZUKul=
z3$Sk#>&*sIz$!R<aK_o$yof4m7RuqOnh<JYNW~PJ2{T#rq@)NbLk=iJuEmw$8J!F%
zc`B#T7?e}Y7c<(nWs_wQpfm)P$tkdyH%0D9o=Zvz;|$HLVYOO`GQYJV`=S+<lZ+{C
z-^t=o8}SK@5<JrNJuwdM>u*=`=nP-p%RN9hY&%?s@B8B&W~4ZbTHOV3YH#LHN9s(=
zE_gczNN!`{jz_@W1*jduv%7-2a*pq7QXa}F)@5h^1kjBSaEqc2yR-}Zrek1wmU%fk
z7Pd~j?Do#3CjH5o;31cJvK+M2`T*{%%bcC}A{(~f_ZzSONTvG%wSy{mse4Q3?$Z`<
z{5js(fxEZ2GbZlQhogtx($2}<`Ii0Dsy`60Go~UdjW!PJ991o3aqj~$1gff{Z3BJ3
zK@T4sCC=5(S+u5z7mK;n7=urd<$6=nStV*58p-8qHDE+$A{NooMbqGlk6fgH$t3Fi
z(MNnt)J+XBiJzGkowvcUIs?{9lJ!thW2GGFd@;{v@qIE@wCV;4l$@mR^I3x`&fb!j
zRZP#?7;-OpVM!UMMPeiQRJ6Yk$q1zgT~FI~!m{}e>)g(F#-Itl#~EQ9t&xBV*9e%*
z#q^iRDD800VRFV<sUKq;?|;q|zbWxk(OERxbR05%h%TBlj*Z<$#crMXCA$G2`FFFW
zqLydpy)cg{m&i`>d5k5*D=Dy*;o`v=uU1!R3?XLFML5~FSmyvmO3*Zoq__%+Z(qFH
zJ})C#L{f3^S18DlCyy_f&nmzRYZeR3=o%CX6N_IT5WSZmmy$I*bi!H=3k{_;DwoU>
zDsgdfPD+v1x3a)U0^Y2Z$Y7;K7cN`LPSSNPo$qm#ft*;k;#YWmwPLyINHMc$W|Fm|
zWZT%Zo=vxrh+^lc-Ar=QI)|}_u85p=y=1EpKT8F^M3ox_Q4x)yF48TVO-Fq;m$YM1
z{9$E`Lw>Npl$2^I4(z9*^`_(|jzr(%y+jcMn97PjZ5f~fZZnkkx(}h4_!IL+K(_dV
zW~^z7uaSr~$0XBTRaq)G10_+f?WQFIO{J;ag0AiHPFPb7)Q$M9p6goDlNDdSZX?%c
zbS1fMzZYni7;;z~18u{<r-h00&awM|ZI^+|li$0?9<q4ba*Dk>1p;>g*7V<FZ6Eu6
zcL|icw08{D-OA401l*1ud}5}|Y0G>lgh8XAsDpN<BVhkT!zjqqKt<vbGa6(I_jXW!
zXd*^Mj*ue~l|I31Q*bw1B%BEoz;uF5Vsg-_a%UhjsduO9$LgNC|Mx}o-@^*vF8#Vo
z1NT1k1PI@$at~nrLos&iwJ<2b`=VPVrZi*Z^WsAwDKVeg;#?2Hs5Ny>Okorj_CZ*`
zR3aQ>jGUh@NJ$)}t+At%fpr$&dYr8=TD0x7F|4ka5&)4SDi2JX#Ttw6Jt`Xp36Buu
z<q#h!Mvc25D!+7n;Kj>J+P3BW_nt9pYGh?ua)TDVduclmdM`|u5>qfpwg+9Z6AUsK
zAp~OQ>3vUc3{{D!ozEA5Vm@0)WayBsAfyp!Xr)=LSH-;PNU;E{5(ylO<b_faiJN;d
zae5!9td-0YUO>T+bUlPX<s82F_f@W<t{gFB7<JjyXudVZki9(9AzjpvZ&ho}d_FH?
z&C>6tF{1{P(FT<bAAj;Oy8j(1z|;4B%!5ZyK>d=cYUq7OR$`{x#DJI=Mb|9nVR0bx
z9wb%3ibOI$y11aY2m<2#-}e!XjK3jkiBc4QK0RumRj$Hl!+NtJrJky((8fs4<JhxU
zNQAdGMMPT(BPsVNkP$ZnYYoo2l957KYF9ZfE-pwZGn*}V7KN#;>zd7ag*BS1<%ac}
z4Q<~OLg4A+1|JiZbtS5nv46n_bX`#)G9S&X#)ly81!T^un!Xp8g35`Hp;DT;oiSuP
z8FC@}AX<#%MB`j(zo!}^fV1Q5Of>eq_gHHPLw1N1MN>~{y1pYR$GixF&*w9&HaMrq
zISON*6issmC;_T{UjShkS?5_*akg-x4_Ty9STwX<U+%j`>5_LDRpe5-BBI=5onyXe
zNV%r%`ulv&j+R`2-4vLW(KY@$Hr?;Dbd1LyvUKd!-Q9ou7}>VZ9UVJnce~p8TYeu0
zWQU+{y8WhOjvsf}zw$dD?mt@Y!_1tjoA=MM2aYBW_*ku72rqx{o@%(?nGf+JoO6It
zakpOHJ|B4p)UJQs5Ij-v;BcPBqC{AX8}}~u6a#h}qjVdCbjQudpTkkac|$wXH*LSt
zt&sn)t?vN~9+sQ>apZ(2uk+ZScRc=wV(ga7H6&rMn)3?DrI|+{khA!S)r~~II%kPN
zIx|G<dGCoqqI(RAP2aOxZa6zXW3^htY=){0tJMmHB>Zlgimne_EnChOGsprWzWVwb
znpwl*Y$ka;T}4kxb{;cXNrYupJG3%XbrD;&mW0A6VeO)%DqvPOG*!c_p5d%zHmjM{
zj&9R)y<FjpWqG}nWcCzxPRX^A<wNYdp1zOF=X3gYP1DQ-WEn^AdrVRLzq-E086$aP
zl`3C6&1Th7RTZ1IBY2VNFgjq)h2^U2TFk~N5kpVkmlRJ*lY=McOxw21W|Er9I5S2R
ziXWo<w+MHTl4sp|ta2qkY$Rh#-d^#yGsfVoE`E<iP9rC}PJ$Eas=^k>cBM=DrI<+}
zgO1dV<HL_WqHjIT{G5y{rf<-8Mht5}768s!&d(NP#F?HPaaK&*Dod0yO462{EgB@j
zlrGr_U=R!5w9;nIiM8*jsz%sFC4O4TBtn!$L(Xah{C(G#pc{n`o{)kBylBm8xt8=;
zR*zCBIvct!vf8ZizUSeCbC%aD`WUIIhBud&qy&q_9J5x$n5e5pf=^;%K3|Mcsmtpt
zv{E#6g))HE#S97|Ze%fJMViD|hgM$XUlc4?D;_?0AbM_Df+aTH28E{YdYl!Oz4PLO
zM^e--QQ1m@PE@98YWl)@lM~S?nx-bkJZjtZePF#_vD&nB-b+-kQgmHc1iLef`3$2K
z^VyvFtPv6Glu$|pki8F*L8VM)DMezAG?goXLmp!?XhTlWwUQOY09j=oJv=7_&wOUZ
z+-{vPuIl^Ju}kArGo922kTYfJ;BZ>nmfaSPn_~0M-f6qj?Kk_?4ZV;p(``q=)zRbk
z&wB{hPBU?LWbbYTb!u-yyKRO((#mas=4k4<fo0o=$orhz{s`Fea_7L0*{~tg;TWZf
z_x<_G_bH_zL(U}Y1lzv{xf5doYRIg#GbneL9mj#dE%$GyiF+_d_X2LGK+CabaN|ku
zR*tp}r-9?J@7T6`_T$!jc*{My_3$4E*vUjI(T+nN$5e7dia(YY{W<i=I)va6XY4gv
zYi2VslR0aP{6j_Xk?S{C7$*tujj>eK45eVbUP)$y%At)BCLm|}(BZ0@vx^HLF{^93
zO<yuOh&EQ&y}?>b-8&#r*AHpiwJ>TZ*6R(k*_^7<gg)}{;*8)uR%<rvHNJ1DpFBdV
zjCB%(U`)yQ5W&_Cr8I53q1$v48QFQrn$5byTFrX3l02P6D(lFR;tXHY_nx-hV4UU6
z)wP(nt)pG8*lafVdL}tVouzAg+R$-+mS|JPSWDM?YF879Y&M-}pyh;*k$H*Ytm>Ne
zW<%e3=CekkF|StwXjH;FLkON)Ln`-WNF<e2X0_R{>NX`$Z)UUENE)ib04^2_y4K@^
zFZWmiP?FCiW$-al#hD0KBl0QQ7?gp~2l;qeMN`H_E{^e|@GXo2z!HqoqL4+yd?u08
zk|-h5`9!rC0wIK1bI#)G3Q+hIXzGTu`Hbaih1L?8UAvmaY>w|cf(BF9q%0<I<HV#6
z!fsk)v97S8eJdJ-R(jslwTN@)p6lx++H3%kMDfAX`^;=6If45qGB|D5mNZ*6i^Yr(
zBlWCCQV}PbW=8J=*UOf6xuL2o=Vx=)n+>mDzvg_tplT}GzUO+mq^TWMU5h`OmcX0N
zd)8e~N(vtX%cYFTMN{$Q@dIYFO1@)S5!#M#U%cXMG2`*W2NH~w1KMelv-psx>KfzZ
z9I40(*_j!Eb>AiG%A&Mkwc5}b$-7L3P1gYmA%BaUj#6|!vDvHzbQXA$0hJT6)bU=j
zolG_HHk%eGfgj!Xn|b1pui5@PHN74OK<dc8eGr!?f9->+;mG6L_xBz5&)WxJ`+dwe
z);WIuW4})^eR7JMz#fx&_orHpu|>zq-u)jwEIT0S=0=7Qk#}lsW2lF15(lA(Xe1Hr
z9snE1MH)ezjPn8#3${%eS!~r4I`5#Xr>_4F+xIsez4ROEpW1k*=Xe5$e&=!%FgjeK
z9UgP|ejXjTcj&^+WAd&bcUnqcGCv%!%Sb|dt%^CtNQMFL{ph@3yPBqwMDr9tmQ$kl
zk`R5hx~8cs>bfpISsNZbctAg!(Fc!pGI6FDsVYZZKVZ4MW<Fowi^H`7OmLWDwgOG;
zWyK$j#u!V?nYyk@)(P<=TP{}ue0q;54%mIykEzVG%G8x*F>g3qG_2Mu5{da@4qBnK
zp{WEYfnvGb@bKXS&KKvj8%d~c<`Q|iUd!Zpb$vx;EnO!Olzj};uA;6TJ&}hGACoe?
zxm@w!?40xYj2Hr|rU7MWstV^E>-7eWqwggDW?hxsy($xOBq_wy){8mw&DBa6p~bA2
zwhcOnFCl0`km%SZi*_9)Kc~?arA*QBQ*1VC*6o`4*<2FN7p)iL92JU~h1q-f_%Q&>
z%{A+`r)ztxnQ`@MLlpzhW{=TIizEngAAE`g?{UuJlc%y3BF$n@DisL;0kOjiApU|P
z(^A!r^?FU+RAtd1K3G@FHLH!sSwrkvQi@Pav0f!st7~EqlXKtqT>2Gnu2#%yM+iOh
z`JBdCG0Npkw1%$lM9)oY2_muL!|A9nS_yloG@JE`*=)uKv-vg?NEVRR_q_lSrD?kj
zPoF&zHq=#Um07P=m<YxxA{nbCBJ}!t!}FKdym_<2cb*3i&Nx4>xj1jAn;El3!)(?7
zl1eNj$VC{{IZIW|S*<o~`hd}vM~`OIwL>eoyx#C;xt7RlWJ2ls#Ab6%3gXrv;@B-Q
zix{%x)fN9ss~xNBCA|++O)Ux!O4Itl^3As_mrG(0zgKI_$h<10sOyTpE17F@E&(W}
zm}DdKrokCSb>Udd=UlIs<2>oTCmZP7fU_0z`5bHN`+Uw$eB(|rb};ZW)9q@f<~}QM
z2+odyxM?}$jP?Q4J{Y@6-ekJ}?sbmMXQ!fN<?myC<!-)o`PiNtCc2}$ZkvJkC;rLb
zYWQ*3aNhg3$foa{Es*HTcv%A-mWU>(B2a&<zggy*{d48*ZV3`HV2$5yIJlmDz<T7G
z54PV>ZUY&&fy9%2xoO|@k9~l3$8ro*9j`Fibcb{7fWApN+RYPV?3Qlq?$~Ew_r4YP
z@K3xKu$#|jR91vx2S@Ih0(C7?81vZ-Ys8!uQy`P+yAESDRb@vVHf>~b@O@9yRKy%`
zl_e`p%#rKW5^^SIi7sqgPuITz1yxm-pa;!n(-K0$IV~Qj&WUf5k<1R8wi92Z5<%B>
zp80%+%9;7BA!li?>pK#wG3qo9!4l03?-MHb^gd8mVsaXy4*Ry_`g+aTVvYhr3MFDM
zgE7RMIX}O^I){&uH&<`ibe@Zg3)XE%*LLXDhO_f?Y~^U%mU&aN+-z8`+L56!M)S?L
zFC<bJ@pV%f&gL>f+se?zo|r`D<ZQ9v>gtNF^PKk=SZi3!XSDqi?>%oWui3Oci?ccB
z=M9as*hPa6S!}Om;#Fvx*@8A)p|ohWDXp<a6QdWOJukj`N^5@d@yC4f_z(E>AHLwT
zU;mC@fBH3_{^|=pdi((o|KJfQ#b&+2hky?ffPzXGZ6#Ahgd79UpFfxEB_&Op0vT&j
z(qb8{xp*+Ao;6%9-_XB)g{x**4Ruqq={&D4uP{b)esL!997vGM>o*15`^@F#iWr48
zat59~d&s<TR85u?TYb-_-B8&Iv@Sxumd&Q+)#Vjeo3)tswc^>cXE<X~x@NT&CbpRg
z>%3fEGhdt$LSoZ)B^R(J_K8j3V$_DJcK8r6+7f$cyPh{!SEK}&uh;zU>(_(`-bb!q
zuXyz6jOBSn>@&gl5<Q+3&ei34D25bmAtqMK6<14PJI~J-oXuvU+n5rwW<iXiS7)@s
z#^NLB6QNI3^^Ce{==u)VG%Tt_T{jr3$uVHwv*N>N43(?JpU_nj!K}q6GlYO5qm8_`
z)<~qQEF@%MaecjHvst6G_`HT7>DIcw<Mrhg))~l|^}4058#0M!&z_ddHpikGzf0Nm
zu>%f((LRJ5w%iQh4rha7AnrJL8(KUD>JCBG?b0a6>l}i`X*mT7Pqdj&*PAqV(}p|k
zy0OAAIUYru(+M+iJoCi1RwA<^@Pa^xWZrQisZK{m#J1*iI5I|-RYj<Om$fV1Kt8pk
zom>)PhAro{Het1wx$VekJUY=I-L`kV$1d=;op-Hwr*qyFfaN<p`@LiD-OsxMCv{>V
z-*l?O=YG(Koj-4>dk+Qn`(>UWqqM;}$?xf^N=&dJpt6`4t+9kC@)KGqnwdCu8)NB0
zPqR1^)8q1rrfx`RuCK3&2~e7@6LDf|48cco5@A`ZEHMXC7VR%>V5T+grU#{n!IO%E
z>ZT9O7IR)*UZE8<%}fMogC~Z>d^V$P1MBsMKD;Ir<^IKDjy0L~@)~U{p=hkt&Ju#h
zIT2G<N)x=~1x-rv)r(g}sM!kOEuojed*;myXGO-tRt;yaMjImlw-VFrdb44@Y3aS?
z#p~bl?YA%KLuNjkaeltQIYTv@({(+YwuLMvUS%|W3b^Pvf3P5gh;xamsyRD<KuA|y
zU9G69iky`MNhHPm?3`wiSuR(!ZI<ZY{u;cMp*RLkSW*;?HF^Q9m21!_7F9*<X2t&r
zt}n0nxBupU<%_Rg^PA6~qm_dQ{U%_n#fKiHBCGX^zUz7U{Bn#mZLF23Vm5TX<IUAl
zz_DoEiYRW8e<3TZRn(Q{w_p4g-*?0m>ADvfqnOVZELSU<W=1osx%~EeTm-ajS0Z60
zXlA+EuwHrUx<NG--v)F&<K^==ELUsh1+c$qHzXuyZK$Af4rgXKC3=pF*^Kpd$Ljhu
zj~+h4=!!}yVw-7Z4fEPyvgUkNb1~Oc))a}F%zQQ{7yrOnbw*W*FJ;xtd3^ScM3eu_
z;C-g|@(i4HqWze9V)Q(`IH$Ii`2T6iEvvNzCscJ^To^p7b<3=-C3T-MG<A)wD#_8B
zizqR$Y1`s<U?dRAN)`>J^eA8~QVj|3#hZu{Oe3;g^6rn{jPg?!T^9XBd~HJrgb--g
zEm?!LicP!b<?Bn%&L5DK;^oVi^geQRb&YlswZ53m@jB7>9eKSGZOY)q4dN}!E@L<4
z!j1!^?Uv)<>^Pu1WT&Pe?-)0BC;)uQKaa7E-EI4=?gF)4=I*9zJ-mkTS3aF9b}R>T
z(+#ubQE+gK@tlI4BaKX%mz|Vi!~VQmZ>Y-F^eVp-J&~7GbclP0CZkC>z~YElK(pYK
zTR?`aL~t7<O25#6Ia4II#K*3lCDb5vd>01jfI(7|>)Nlg4=|1ajw4|5&H(W?5OTva
zJPFb!*L1Jb-2fP;K9G08P3ng6DiwCyQ^U5qw0}$6Ka?r<^39sr%!&K~nL1^11XDyp
zSL=>9aYfFNx^^^8Lr$61b<4(hLdsN4L%VFnc{~R?-_f>-ki?XutpTlxDvP!kaJ_5^
zaYbEMtX925zFLiQHSH!4Q_K0;jOC^knG|R78I+dD#x8`S4OVfrT9PpM%N4zs>;v8h
z`aaP0N!SH1+HlS}w1LIEkzBJu5q#$CY=+XJ!*;b?iHQnD+qc9w;wN-|Hm9j8UcI?y
zvk|>NV~XG5a?PdyB9anOVq>Xm1=;fG;WL5{ELUsZyk65Zm5e6^R#zAeXHTCHVrJ7x
zMD2R>3au6GrlqNB>ZYbPHBM!g*Gt~Kd_!G3>ZaoS>|9vWRmbMKMU!c&hDVFXh>jhg
z5=tf3*DJ1Gt;kw&aq$47WHN^(3UThY)?ssv%YholFFyGx^{hq#zV)P)XW1=rTJ!Dm
zZ+P&{0~Xae&Q`3JOH|I7Ea0!vj`Mj&sV^7=XI08%wbd50#_)$f{{<hu{|H5*at+=G
z5nVPKA0%RVHlLFU%jknh;)uUq!Wdn0tAbG$;}l9~s*(oJ`zS0j8HGV>i#3uOPM36e
z!W@f9IK@og2XYX9Fl(IzwdfMDoMm!0MP4T-5%5+}at_OT@0=w@iS7<5VvLqWla@fD
zT>M*APGl|Z8&z_wMmcv(4O$Dk<(v~Ur}tdU7o&_#ijj=Sz<?T4-Q}FA&d;PylDf9A
zYLfF<#F6{n7s(Yn$rLd#y^J5RRPxqR03k?3u`yPBX`RKKRVbsmUa!ce;;R=|nAdAk
zjN&VZ;`(|;<)nTJ8**Ncxrr&6PkeZLAG=`!-<Ew?v|nxlaVI9BgVs;|K1`*x1z^XR
zy6t}tLEWi{-D%wc>>Y=$yJfGve9JK>Z^PS8(vxi`cPX}dBAEkik;2#s#@dd6Kqip!
zBm#++NJC<VXG29S?hcAfS2#T3+Y~jKtTVg#uAsM67($gxo@St=R!)FPpj75;-a>5m
zuIG*{+5wh124L>Oa&cIHzul5i=Ahh!2i386<)%K|k*(V4>)tuk{(iP&;%FNu`Z2U~
z6y!}82}jzRcI)W&6HB9<e>h;bUUibjs}lK*)pbt{VlE8P)Avif4`n4H{#tcav0iV;
zWJ!^vB^{Obfd>yRAje`hN*J@jISYBskj{&&=z2-u-c*%DqZ-jx>wQn(iQ<3vZ~>;I
z6iXi1kV_h~x~Z^eiB8ou-Nv(CuF*wQ*H}xxUa@?=7Riw~IPmu<4XevF^Z86<Br>et
zv^eLO&1YOR7hIg3v01N0A8uw!nyEmAV}7<EG`^%y6Gp7deaZVt$<mO}S`oabnKe{Z
zMYrjAdj7BkaiwxUHD~7wLg<7&$xu}dIVD^rX~<HF^u0tuKD;=mt{eJJQd8x^DhBU~
zDf0Ute2io@j3{M!{o<0BFW!*J;&+$fy{GT-{?n(TO;%S$F2&-57YUXkbJA@(UcP)q
z+9aO8en|+Pw(mGwoRKQ{?5odkx+3Ysdmq2cKl<Z8rfWB7#EfRNEvy{8S+;!o>DT=3
z+t)ym>Ok0{7$X-K=lsbpevk9B8lw~$MJj+oNh%FtO^0+oit!~BsTg6KQi@_qE`cFK
z%C)8zvEwwZl$At4XK>1n+J;tZs=6knD1jE0o@Dxv^JFqwSwadW^_{|5Ii?LQ<c!f;
zBxp(=W~Eih29in^mOzT4eK!bwD@5{Th)6eDf@OmD#V1jrjO2XPN{Iwa0gEZg{m$ef
zxh40t$Pc2-z$g}Jq8JK5HY}G*6ve!nh3O6@RoNgn6hpxGB@<1_85>hB|CNuE4_?M=
zOvEIj*WbQ=$>(2x!?N`<{~`f2iIOw9mi(H&msIO9C6?>81ala}dbPRF=j=ooOtI5R
z;WTfx|K+|3sBHn)KFFG|o0B6C*|*`}eE_yAaC`{t@=onj{X7K%IoLm*cJf!=-+RLZ
zx!3w#=IlGo>o<U&5kDLO+c`m&3_cOb``VLO66lF6@dQL)RHJcZ6g6VA(FG8yo#ntA
z6kdZQT{^s*4;L-YF7RF*Vy*<%!NvT#M8xi0$N{)JfuK%;yW`al79>Z3@;!j~k?THv
zj5_-BMo@WUn@8T;eGqpPEKU{xrx4hveR%5n_V3-^m>U-x`+d8Cjn1d~d%WBW*i}lC
zS^+K5Bl#*xXq$O_{($9jNp_54-Wh0Y#q6vCL@HyD`-ma%@WDALiLP1<L6|_TBvHJv
zV@_Am&dN9=FVXqpjIL{Ol}IsYqlmE=mazEFZC0&B#MU)k*C7eb6N|;7q`*q}zGuEz
z;9Mo5r!f{k9x%#`QKl*FPExiA4%hXJzUyh~iq&ex+1Y~Sa>Z;hXT4rgS7P?6>!LF@
zYb0@c88f1sZ%3#6uIs3*qw6|m^M<bNS)85Gwk>tjuv%SnadE+`m#<hX&bYq5=E;+%
ztT$`s^EquR0UOuLYhh|aW<GD&Y&I-r3o%93O_4*X@!nHc4WT4p_sgDF>PiM|5u;no
z&j|er+sx#eZ4vb?X|d1(qU!US`E0?1^9w%z{BwT$<yR=>Xs?PNTJ10n&My|6J#1)Z
zH7Nv2YymK2fk-^QSnxmmvwy<>>wo_L()%Dz$V$;v6}78)@986&%92Bp<l9=Iw8IYh
zNR>im30Sb!iWG;FtQ=@fNRrb-T)PyZ_hem|Q*BU4Uf$kIw56C9<(WqnOA4)kLg70v
zpd$ofq_iZXuj-1NBCardMj3(+Z2DF*S^&m4f)9{2>vbnAS<b9B9p8TYg3V^l(<hI3
z_T({I8-h>8KTrJUjLjt{uCR5I(W5Kk(GjH-o6UyCRVCl7BILyN^)<i<1P5l+_wqZ&
zNZ)nkJ0cp8Im`RI*|ddu))VG06AB1cS`&OA#qDz)qKcI=P_o6uD2oe{<d_X9(q{pe
zjkaZhT9o{AB6(Q*p0$eNnqei{xvFNwC?L41X7~A=ohW5&l)rX`gm;VD`Twzcd8f|w
zxZ#LH@Fsiq!P!3eI?mGVGjPYi-1LaNw{_o-Z_C8&1Gr;F9qc|b@f~AGj+ML5VRtxg
z!p;nfy-7u-GO}QTtYpy;kkoZ)@?XihVn`$`D<X+NMZl4Z6gLR)_6qu(sZkRwxjkoN
zdmf&3yuO?X@0Yh^5DJ<m(zx~rOLGix?K5xpW#abRJ;h+%*6tJ-?o}Ly+9`0dlbPdG
zA5Wb8z>GYl%Z~Kruz_25zg<yBUWEx@1Gc`HlXDJ(VZ59lh_U<AKl}x=S|Si-Aj)Vo
zMv275JI`>|VYILVIZ5>5a7{yog{~K^u=z~%u7(8lvba%2U{%`GMPekA$T^m*6d7$F
zlCv5Uk!vlvCUX{tg|t{uIdR%H4;Ex46Pf7gK@JhE&Ly3ZHOva*HLy_5Rdj8Ot1G;h
zv}c>PrKu%iudZuC=&=@(O4RcTWi_rgEEWwZd1fU=k{f*VtQAQSD?Ub6o5H%}65X4r
zYDeccxXR-D9<n5}H@X&`xfp5cS|Tye=A_uuR2JX0G<8F}*>JX4;6q^kV2<~m^NS1C
z%Qekn&T6?5u((-M*A-pgV{KKyK^g&_Dz;xgB$F=_zADehT9Mdjni}sz$tIF{^!O3y
z7w7!(AOEog8N9mW*T4A<Z3-x(2%%@QTGP4NHeYMZXe|K1Swqf|cOG5vqbHAe{^}B?
zMJurHdrXe};qQHdGo}Q)NPvY#lf3w&4DtgCRfK~Rq3gs98YR<2h=G`jNOllLG^B*q
zf;4^aY5SflzZ9vR;!A3srT0Bu*UQSffSed5m0It6HeE+fA_p`tI&V3PT*sTMH-!O9
z5@npk+~$L)Yg@bzgqZ2P<nI04nj#pyHEZguiX*LjFj5jH?m>b`e740eF{L72lP6LQ
zqm<0>p)O`pr6i|s`Ld);F{2jwnbGlF0;8t6SqI;{R7|*HeH}htseHJ^RGi7A6vw$v
zDUH@(%l*Wb4`GsAyCElH79pIBFyg-VB8^hOXH0352<&ajc$aQuxTZ|laNHC~?Si?y
zw@vMV$3yUT4CoE#IDO1bz-|{#=CSXYEsQ*rXF1ME?ADrG;$FE42%G|aKXBQ{wF9Cm
zMV2W7?!~Ao@#PBe1?FWG6&;#2ftrqUI_4xyxvLq22bCyVOnWDnI!`VFzv?nLg;A5&
z1XwIQI`?dE%5dzEdu;#y9RX$s_j4D_?;LZ~#Cr<ho!lU8AAq-fV*81v4)EA51LIi4
z@SVynfNrnqcQ5<Pu;JWOFq;o{ADlDa-PPoNawjQZRE1TLl_cI)i1bO$VrsHlVN@o^
zzO3?P6>qf0Xh%qqy0SF2!xpocOnfq`C0j)<<|80wV(3L>KT389r8UuuG)hXs1`OJ2
z0`fF<Ei9pRcH9<X7}*ttn7d+5SmW@0hf<lY+b~}&Xg3>Xvl*N98e3JO*@Xar9K{^g
zwk_uu=UiPc#WcNo4FW+4!DA~cT5t0OL-q>+jgF`!2Wu`mcw$2B;OvaxMJLbufHhv2
zo|QQ8U$2%d77JFZ6&L3Pur-ZjG$<xft3@89U$zpc(Qa7G&)BpZW=+HD>Y9u5^Af01
zv)M>mvq7gYgP8cxSmV$svsrG6Us{q>XnjEC%=&62xqT6{cUF|#vl)lx!*`yss2cwM
z(|_Rg<t5rGF6=q9iPf8>$Z14LLzQx(uAEHV>ka3XVfOSX#`h>=IA5IcKmN-<=b!%R
zAM@g?-_ms*WJT|b*lh^-;&-KtF85nRmwVr%&=^69bbU|nJMp=TTY!_2uu$?0MU+>S
z^cFeg5j3X39IDEKBnq&KO2Ew+%?o2Nim;R0<DrNu57Tz(hDOGI%973Hb^E3)+yr3D
zelg9XL~uDy!9!*QQ2YK;B38W{v#EwfnpQ9{aYA2|?@KPG=@J30)aa`>URxfnYXE$z
zq*odMWjRJdyGJmQl05f8<Qpjdlv#2xmy1$rRJNnivbc)FJxfY7`{LXC*lm5+rod;|
zw_o=(fSVqB2*P$5!BhLUg2sp9#fL!ikZn^t&wt3zI;pLO<<Inb*^}rv;&C^fNRK1G
z>FqeIbf?Ms=H0iJlgCeI%<X$malbKom|*4ZRarAHF|Xw9`yZ6}-9bkruwlh()zC5{
zD#IBqAqSbf$270ytxM9DH3=g42*GT-)OmO>m!OFqmSz`(9tB#*Fx5~L$B!9z9@#$y
zS*O~)(~f(9sDozia_l~s{SJt2Z`>cP`(4K6bP+M#sP@~N0=;AX`+<O6*LIxMwPb%R
z=FHCb7%LHW3ee;tW(&^NbX{P%T5*1UPRtN;E_pi@SVxTFt0e-*k!qMUl)}fzX0xG>
z9-|fK3(0kv5HVeRk8`yM7&}MwVN7Edig4;6jdFFpqHb!sUNQ+-qexMlxL?0{LuJJQ
z9bmm)V{Jv>c8~<9tlJHht7$hKl{NHjC)#*#mTXonsmZ)}`HGYR-uJxw-g}VNRHh-c
zfz7HV`a}u|gQeeitg)mN@tvg0%00B#8~Rnp*VkWRP}pQpL9u+jL>W!jwd5|*wJq9b
z*2^_3ea-56Nj#U7N-r*-i;PR#F`Lb}zPc83cLBlPd%iA!rR#*T>w1rJwTv}Ym>&{f
zeEMsypMOn`Nal<pJdLJ^5@%uhfS6)|B;p4t>D**$N+EHk96D?M^S}9b_~^x(si<^@
zD1iv1Ow0+NfEck#lQ(Ok(p0mG|Hr@jAHl$1{BQpUIivy@<i9HCadI1kKIM-RcBWLx
zp{qy=Qqfw}$f|k(DpTo<aat^^tE3n~O)ln5kcEX(6q9Zm0Z&T9WIKp@Gx%4JjMo^&
zo(qU7&tT}m@NG@yBOejx?ifmP+?>bjRi&K~tf>i8qZq;7bh=IeD(`@^@yvxOBdfx~
zZ8f+u2vCFZJo0nMH_W!)c&#EmB*7Sx(L-fWV;;l;HVXI~0BSA)C?Mu@T^Mc!WG$e3
z$I0nCmwi+0VcCMP{h4z6%%*MZw!P1?onqgPGjEfpx!1qEQ+EV;S?wIR9VKcDG}t#!
zPG6oA$Zd;zcMl!Q6e#RZo+oC<n<mX;tjtjp+|J9icfJgvEB(&f{)Yf-`Mcif`*5D&
zer<>O4hg3e;K|vO)sk2iA~peO&CWf{Y)QAiV!gf?Gf?Crec}^*{yWE?ydN<`^o8Q^
zdWZIo0=6S%dVB6t_t@{d9y3``9RK?s?VdVdAN+miV{fT{3M}ql+R3JlR2)H_nv6xY
zcZ}M7?mtwnSgS21qm!^rM$$}~tcp;mMH|teQd!~qmadOHfAN|xzI;yCcW9G2U(9*>
z=rK<oJ;E4E-+8XDuc+&Wt~mOKAm*x<uU@fSt$6h4A$Unde|3G$`QnVwcAU)?G<D4~
ztc6i2QL$t(8O9j7UM{&@y}?zIaDK5^(AWyA4c~nGE!WE{o<4aZ*%7ps;Eq0Ewc_IJ
zjF2+x&4$%#iM5W}Riv2cx{lRyg;L`D-uIr6J&VN*|3(<Df&@nGyV{AVTIO8Iiz=q=
zT#~X2C`~?Kl*Sr6f(VJOl!rS`t>u|3t>pt#{Ef6W#V<_0BxAIq65q2iIE7435Kzn?
zK7y)d<O*V9eSOX4i<ju)*ERS_r7SrNWeUq%@uf1g_?(OIb0Sepe+gnn^1!NG#<40<
zp^1=7DxD;0rc_Ln1O-?l`iZ~(^{4!kKmT){K7GRTFTW;bHG(@uD&S5)oGRdH04gJ}
zQ6c~=;=u}38jpzs_)-I~QANIH0B?DG*K!u1G6I*fZ2*T*?iq^FK&%F-kUY*g0~p&*
zKLfZ?Bg2(O^jG*5MHxF=RxML<;+Aj4kU;@ez~Ll-B$fW^GA>f-%s9Tr`#2I9n54%%
zv!F)Prye3^vzV_?lD%YzmX*mbmunqI9pqRP<7D+JKzlnEPXp&PO4#zW1)wsz$`4&M
z_<X;t00961Nkl<Zw%@vVAW+UYx#P$D+R{#)Ez>sqf4fh!U3Bb%FE!Rxd+ji3YybSy
zZPW8k@9|iBG9L|Y6_6*)-1Npx!Rg+nouc-(PX_QlkYTUhjqToW%(43W+lH6-;FtlC
zrR|?X)$-zQ@5c>5bO$JxUE@MXew&?R%btAzoMBOIFgDY!l#p{1Ec)uwvGxt;ZaKNG
z6W4xl4|Z?C?WI({wf;@d^;G9LyodKibf<vTLA^T<s5_kN9_`<``EkG<9(?p&D)sZJ
z2YtHS3)ubU?>^&v)<}vhtuRKFWZvZij~Z9=X)v!PMceygq7h%X<vI|rI$nSIf}CRU
zE0O>OThd3R!nA3nIG@kxU$6M=t5>2~6(et~BSrDK6F;>$%0|e9I83amOl%3^vlMII
zpcQB(J`q(mTz~yq`khMRaRn(xF4GGrU_fa_$Q!x=s33kp<V5rvOqS@w@TO;Z)sGW%
zR@={HU{fY7HF=+M(Et;sa`;Fm6hZmo(2s1Hd{~A_JEcS}NvBm7`5A+fkJy;2Moic_
zCG_xNFM~#9sFXz)Qv(d3rG);cOfD=7dVmlE(4#og%e~WsE=$g&kP15`7pjW`$>5tQ
zV5cC$LOG=}s5J_wbup7l`kmKr-teFP@;~$8kAKY1{^0ld_UmsT2v`~?y^^|0u3_R=
zmIn|zEV4^#JU|VTcm`A&_1s32WuioLmWE4!hBjhyjyo%<T*ltyC00nlgDNK4@~#T-
zR^ta<`dIEKa^hiJjI9rVS*Zf1^ElpzA*BkqlZfg(0<QsVrtyU39u>fw%F0ot;TS!R
zHGr(MWJ#Jp4_(aE5-B`kJaYksS_|k*X&m=k$dD+T+1}kejPwzJ49AwcyM3nPmNLwb
z$`9PPbSTyZjBWqj1yYRnW!OH{+67vw>>EJb0J?VHVYRpCu>Iq|r+{!An$t_k$HAG}
z`Fk=>PtF@R>^}x7j+GlH%zJFVwQd<5BOAF_XGqFAonwbp+7RV%6EHeDxewp+{e9|q
z{jzChH+=9;$EUxU5g0ABU|`cKe)sAjzj$|v8n5Z5>we4acd9pRybr)Ta_mU}c84}j
zKAStXc>pqRt9xR@ce#Y=i+41l_SjgNfV_L$xBD@6C{iD2!<tl*DGw9Czy>HaO!no&
zkHVD+MG|#`3ebm+o(@w$jS(b6jDCFgKzyB6oi8#Kn&iEtGz%WG5}zgwIZ7CwHZqZ@
z$pBDVp%7iKVOkK0jDZu+C8L&tlu3l*C$<G5xlGu?vXy5r0!W#p({P>R1T(NnWkMTX
zeoeW5l6-y?b1lm@RRC>D!i)?|o`&M@C6lm1%#W1kKAdBiT!&^=k$o`?BC?7h@IVa_
zq*;%09|9~yuyzD`N<f!N1uT|&+iNXs%{cZH#u{RAN7l$S4fkW1`o`Z%VU@6sF$Q#2
zESj3Q-bgB+obWN@lVP=5@kf91C;aVy{41_sUCIZsq&m}uwi%cr0Z;>THh|DHe)~tW
zVaj7#wQ+*YTkt;&53NO96eOjaQff@tJ-ma%d1Pqq&bWgf+yHP6S$Z`(P>-9HE}&2*
zpK+lu__vjJXaKCFVw#=+ojm@pT;YH&Cq*k&Qpt@^eW!||uzkb303(BzMrO1enTr(5
z_NvPBn7o5i6U4w^P5^O0j=yfk2*kA(Hh#h;Zi`&e_XFxg{HJ94EaM>V0;s+J?(W<E
zR(myf|2}4tRXc5R(8eiXCxdQ}qnk`%NAAn93F{_McM8nhb&~uc0m%Wd%-bDDuXFM~
zmT&lB{XJl~)6S&+w4rhGE!%d0-k?Cd+n6TtM<2iB>u*2gYBg9GkoiF?{`QLp{Os8`
zINcpSgZ+Cy`R?Ai+yy7x2=I=aGv8EiZ`+-4%4uLMzrPE@x@o-M)Q#@~2+KQo07j1?
z)MI0DQs)QDJds3HF{W7hT0Sp%;z+G_KfGcp*_y?s*D<g=gV{&)+O`wa4qq@#zU8;f
zx&>e$fMA%qw9*(|CijvDb`-lECv!D2Oa>)-XvGW{$l@rT_pYsgypf<$L+|qFV<nU6
z)*oo_Y09IY*U<lR<gk6X)Z)CK)plZ5<B4*aw1)|w5zq}FTiVPcv!f@^Bc&t~C40{e
zgptZ==(m`(O;uqATUYr|#UcqpVHZ=%n8I*U=IJ3qc3@lvMm1$6AcYJnV=~1QH-O1<
zABz9f5H)($O8!#?w4LXxFTdo^{?VWCNB{Uw`QQJ4{}$yFL!fAurYCx&Teel9MD3s`
zzmFgz7fvM&2j(%8N={-*9xsqBa2}r7$hHoUWnjLh07?}+yaTqTufqVu^Z0HI;9Tyf
zXzNY;qX45t2Qb7SZ^HNv4#RBA*2%U!_Gtuu<G7u?yMwt?ZNX<ACN1Pzj1i_Pq>`(6
z3f|IiZz=Ep?sywv$>i@5+D#`|7+%`3#vrva+}CZ@?*rUT&8b`FZEH3i0uuK3ZGTUD
zF$H0#{~p>qjC!4(=ivCvWKW)KJGi9dzfW#clkLYQ*6D<FViLL!n7e7;ecGA!cb~DF
z?i;_sM~+p;els~>9~7ME;K4oLyDw$ilf^6k@b@0_fBcsPU9WZq$JehMUtB)nr;olI
z$JkDPjy|_rfY5E_u66GU%I?+PcS2hy{{&6|s{1uNZoG5%_A3s@PTs4dFYEqzo&Ly-
zAL2Wd`$fg>d^}~~4}PUGk#5Z`+isQVE31)-8`!b&gElZ2#a}4pZP@+DLKRRonlQ^m
zB0wt^wyA(Gaxp_iaxoPRlh-)OP1OEV0Zb|7k@=wn1SqXU6?|kz7+E2p__B$Xn>NB=
zg(7D`&PA3Zk3JM6$(Jc0L+yP_GD@3?*=o24<-Aj1$Q1BN|58DmYBZyzku4nWPr1A_
zOul(ApXw5wIGDYLn<mF;MFE?6Jl7ViCx&SaRhT_BG_$=|;v-m?)gqZgF7*pbq?N)B
z<6jjsd`zr*0WQHu{_a<Q#~=RTFZjp*^q=snzy2GpUcV{I?E|n%Bk&nn#I)6|8(2Kb
zSd&17?bx!Itvh+nB6>VL_rlU<hN#oMu~AqrIWCW2E)C-xwZjw+Y%1jOSqb=N3^LdT
zV{H8eb$K@gjOXp&84+tPKzsm_VtU?2${M}BelcURT@aM(o|@c;lx~?~`X{3T<k3{D
ziut#UA*GARwHzNu-_R`xg2Dd~P#KjA+dln%l*Ps<QzREN0&Sg#cOBvo-TYRi49c-=
zQGl9kKLyyf`?nx)8m+o(rajh19{1j~eY-DjYaN{@_OInwojYvX*?w#Sn*4FsHtz%G
z?$ky;?BlTs^5~5^()Ux~>{L0}J$W`TnV7e8sKWCHAAG^z|Mus6{-wj1^2yDhg5s}!
z{g@v;_y(uK<k%x1<2z%b6JYQ@ZQccN+;;!I6R<mI;P%a@K-(RgySd(N2i-Ep)ZVs(
zvHSx?#g6_l2WWHnJy2O@_qYPK6oWUD8U<g6ft7PEjCDCjIcAt>23eFLN{+!Pd&&Y1
zKm&PYv^Ek!OTrpcCeU&nrW`ud&)czO^&lBunl;=040)XeC~9VPjqi&;T}Xr!$rGP1
zJu$0p0p->dH+<lAL9bdX7=S_1_ZqG<m&tV?fsCF=QvgGb+TmFY`9f0>>NO=ugOU=9
zoKc7=I4f0{p(31Iq+?LUcXD!H(|FIy^D$adU!e-~H$-w3FrNpr<2LX^DLqa~Mh|0Q
zaOTgG>(`vkW^B4EQihl_mzP)k^<VuLe*B}4`S_Ec^H+cImnfBs{D&gMSR!->(_$H=
zX&Vrr6)3TZ7j|>Nf5W{jA3Oyni!_c9ki{!d6@A5l`5T`P<QTWWwX`*`&oa()0e6G^
zNFKpx&Wu2AYf>M;otBId6LZrz^e13wI`%Sb&Gp5Eu7|N#UZKdg|1b^IkY_vw_Y{^)
z8$CwDj>k<ViqXV8qK`7xltQVkuEWTDjt=eR8&SY;xd>v>#mi9uVnQTRM59o~a|RLr
ze#@dp___sWyZfi`XaJ{&0B;{csU!Qge;)^Pu#H#UIVK-&yLAG)%LMl};OlMcy=|ZN
zfx*f1?%lj{@5cdqAVG1Dz8=&$)(JJ)E@Q{34Zr{6D?b1F@feOXbnokz72jMv;KTE8
zWShF-8i)4p)5rb8e~|q*-P0pr=UylMj^!Tg*WJ2s$8j^(+Xwq_()>S~On#uKSW15;
zIBK%W&m*Hj*}gUPsDqEs7EtBUcWM04?VE*=@ocOf-``>QS}U4u1H+((cJz?!C5wW6
z(Xt~$E)%C7R{o{YsV~;}sa4x^<qSi}p@4xbx@jqmrc;IF?u6n8R{*F?qJuBcWHQ(R
zql2)kv~P!uSOd!_t-{3jkGutud2;VW7jtVy-Zg<y?ztZCrTE&(J=q2d$i!xiB(WE8
zlXsrW&_*aZQU}nU^Y;D?&r&O4lQaWv)?`Bgr&4v)GgHHO$dU@q8a7e5Js%SP@aboK
z_St9r{1?B+U;oGd1WB?Zi2RfIMGef@IL?5<K~iZk%TB@5cGAp+mD(-i`#AJ54fm^<
zOqCv^RVBZxo-BxWLBhatshlxdkue20Ds}=UG7|<;jXH4x+S0gf+DGwaEenQB0fxm-
zQyXbZY4pyRn)s1U7Byo(ixf>NDch#!%Z%YC<Kkz$S>?qSY?sDcHw?U-^O(dd<~ZKF
z5Hl$bzI%y~izzwfVf`m#t$=x@z}Vdd<^P|(KY6+($@0Un-*Jz7zv<06Rc2)k-Bs1q
z^8hpj0}K=ifFJ;pf*D1_LK2~kcG4f9h5iKXq=lATz)UOdWd@QU2v7vtM0BIO(T!$z
zRb^FH<(zMR-*j)dA1%%~?j9Z<aqoTa%c|}~oXqdt7<>*Net!H6FGrk!mQ7g0Bi7;g
z&5LjjL(MS7q&<pT-<Nq+&uv@{i~MT$Et&*sdxD3X29C4O_5pOes9aEobLD6;ZYA2x
zXV{PA(YUOHh@XGsGra!bE*?JZ08+jM13eRd^yogmdi&#X+c)#B3qtM3dxdXBY5<ab
zlHC<&GQYV5Uy%+;GL^6To|EbA2vdOjCP3%mEr2i1+(|lit#<~>!eRqSl#jZyoLG?t
zAb=#{Ktj^qv#@DlGk{nsD#(JtF)K8H8P4g^j=IQVRgNHLbyo$AZ%1X&<6kLs<ZOS;
z7W|IH#Te^O&g^ahvWNwBuqjn%V+8>qIB(N2@-r%{W+2Ae5h?FdHp@6oPUX<ZDc3Fr
zbm}Q1^dWIGvzZJ|`NSxTqa&x~%}2KaQauo(II{N+=p%xQK&J~Qo8kaq=gI~oO_4*G
zCoIdfc@p6aFiis!MCe&KY~DNcu?MkJS6QJriWh|7iEy;?xHKQWOnCnM1%B}T@8h?B
z`!D18XpO6&K8n5k6WF+3mVZnpGl;x!R_%D!yTplQ^#lSvzGjtcOKGKkqT_%|yKd4Z
zrP&SR=vlvEHuE3~0`2&mC4RK;LHRh{G&9S2F)4k34dtX!>B*@rNd@7KlDw$IHQ*Yx
z4<mW0VC9;j08wU`IeigkfD3LYLw(x@1)etfYq?5e%&g$F*wEN!&_TGm!75^Hriw;H
zfa@Gq3S8}I9hzP!0O!k6(m<SeJ2?8RfZ8zR^t0Tz%ek?>2l7#Fou-0qyDa0+0-Jq+
z!9=*$qV+03?ZyD%0cDZS+YVS(xlX%Ia2BAn;?&@-&WLd0U*K1N{u%z+AKXV$Fw<zC
zy*NdT4xYEbV6#}Hzc+BJ%P<pu60FWAJ$%O#&x@O8|J#xGtI$4Uq6{}`@2PZo7Wbtv
zbI-VLlLF*@VZaVq)`hAU42YZ)>*89F97PQyB^eV1JeFn&>IhK#Y!b&7d^R~Xl10{=
zYc5E`dav?5$acC0oEys1rM)BP;MF00lAUtmsUU_ZfT%+Ppjobn(#K?7=(8*!r3K7@
zqH$16VFa#N;E?qZQy!d;+c+j9EQ$q)qD4#&&S$eTB{wvgN+DAY$XqpXHXpzts;eq3
z8+7pv4E9vny@R?x%SVMER<Nud0|;PN5T>2M<kzBpU(>t|vDeK)7Vys;Mp@7qjP+^-
zcfCbWpt$XOy#L<&_@!U?1wE!$Zl5Bw1dfq624Fk(CHbz}PG%Sx0LWW81Q2=W!Dhcr
z6#_cr4C(hVu4U2|CEYNHMnN)BBLE9Yk%QHoiO8kpJ`mJIjsZkgMIJLN!=QXTsRK}s
z`bE~!A>_@F!Z)+GN9y8;A-2^7*m0Uq1owN`@hSuA$}Gz*%K8}dkswKbiWG`@t(+Ch
zZz|&_rgREr1D3w!psSj93IwH%LX2A0sC^k=19=4idNtzew>4AO&!%VdA($IKFYoL3
zxg6sY4{BvM3!ZH8-?1?70~l@^(7kkA-Y>mDowyMI&>hG`K_>EUzMCn~<SQ8i-+1j2
zj@I{ad7VEM6L7ha4YfB`cydBt)r++rd(m)wPL^+<cffSpO|?x6xa`bq;(NDfCvL6B
zZzIwahjbLCk$yea(E4VS){sHuK^HBa#V<(4E`gT1$hMGHHn-AIEX$5Qaq$%7<;7R=
z3Jn{vjBmJ*qjEnCs2N8DNY?1BaOE-EpqwO}7-0&yL0DUCKxV!pC<nuK&&9E33E+?w
zGRk_0oL2^w_G|fyuT$ZkNO)O%qq?YqbIN%X5a{&y&;St}NnF)Tr2+^dI(1`J$hRpc
z%QL}uPB?X(!{&1|1!7U;1`V9E*4rCcBAsWT$AeKI0~Ab{e^4`|OC+GmD8!iEp0X)<
z@SqqB!qIAtmAevmUo8GopI<!3)%7KuBdlD9K8lkcRy%Hh0Tuz6jF9T@ynw7=9A6z{
z27-x{107YwAUaTI;pA&&4-En=JMUF9R~1kafYY)>vF(UZ0v%0H6*7YMg9P?CcPHaC
z+xHcmo&l^hnQ{2C(nWx9<yaC<eOHtztbOuWWR3$7R_SMkna%el0~}QwoD^V*pWno3
zPLWH1%>xi?78;2h$om9DNx`>w4vbDc$~e)$1tSCjwjoCJz4)#Ly}FEDz3{iD_0fq%
z5q1j5Vd8lW?ppVxi+Ngp28%{9N0#N<xe)L1tX|mm1z?luy-K`^1$6JUc;A4#)0_kt
zQ(&(S*3RqB>N#G!x5W=XTH7I*O#2=g*e9+~-5z@&o?T_$r|fg->;#5&`_w>ZN2Ys)
zZS;TZ!gJrVwK;3`yE?JizS*j8BW#*aC(^LEUl_1M0k)6<lhYl0scC8g+JiXOC+mq*
z-$w;#!+bPqI!4NAaTHGj2oaI8rl5r{0hB$n#l<*c!^JgQCGT997HBcoF)l}zg)-10
zHohpcTI9G$!HOx4B-t~T#WDG8BqZ%$M;UY)Fc6@{@ZQ5tYB`#z6|0LAc`CBw(QISP
z4Apw)5Mn^sik7O_2qSd;5!ibq^D#P!c*)qD+sLBpJjjd%I&mjv3qdv%U6?&l&K+|=
zK<2jE+7UxcLneVg1OHBB(N<lDt6l(Yh!N)(SLj2JZneT?IERe}QBY7VXMJkGB?8i4
z0(9f6;4k$BZ8I~-tlULAhE2dOm6_Bop=AW&GR2^qF_vRr?dR72ZG&Qc(mAn@=-7~)
z8|-sTzEN7A2X(TpWc|$ga>;cw<?vDgh>OW2Yol^1ciHz!A%3wdm;0#|lt5T@vM4#<
zX`M6hl{#_;tj@tfILFK~hNA+t5CS+x0qZfM4*^F<$EpO?3wPAC&$WHS_GW=y>W__3
zt!Md-#cwPYjkvA*!84vO%S!{OS54z(pB)@;1~Bc7(`JCd!R@ohLcUj-hR+<X@wL~U
z<A)zVP?>E24!G*Y&gg0Y=y(`^A7HnuEc?8(05-VmZ`1Db7`8uhdF%QlYM`-E9qSAy
zGT&Jri&t)9Ci)x|?@Ml*fivR}lwnG6##LDdY1gB?+O&O9cI-%PM+o;ObGj^lU5-E`
z-VX@InLrM!0R;>y2rJGN%&J>jgj4l$gV>@(q+q2GYN`QYTKF^_jw(1$Lm~Afo2K`X
z3)w1A(m(}(Bm@pQ7LxKRrztWh8u0Ui)M9`)!>$IFoClFOs<*1i5vjGO9IX<Zq^+ng
zFXrYJssK-s0g|I>xGeXE7(tHE!KKANgoqG&Iq`zDVI(9k1Ozb;9T*@8x3y<6+u=dk
z&K?WE;RLFsJ)(9P5k*&%9Aw6k1>jlvq~8`t&dgw!K_7(4h1J@Dt^qrq-EIROJ${1Y
z(-S;?_-Md!Wwo!z$fjMKcOWNeg%Gt(1vC!0u#lr|$)DBUp|)GJBVQ7@!%>Z6i<`A9
zi<gf#ow|@K=PK*-<(MFjnS&K=gJR>wC+?$h&1EB^+*;=#vU5Y_mQwV<j^@**NBf$M
z1@a!Q^Qn({*QNGzs>SRaso5H+XPKBb0#MM#%m`6z>xB>zLs0EwMZe?#Hjt;_m_zz@
zff=E(5MNzir^(Mdht+C@)vA+Y+7J(#_@+rF;xX9TV*xpU#>y1qG;4rI7zH^1#I+VG
z8oe5)S1ovRdE5x7<94{01&6@IB0Z7MC{6=lG~?|^Z#J$xBQU;t{~7%1fygpDncgp>
z$2Fbc7*`m8gi%@d&2!JP?tIVVKqcU7*0ao;r^DsdU>vu+Z+#xB`)i#>UH_sCP2k+B
zWA%QoF)=7)<9+q6zW$<Q?3}e@$vtBQN{qCC0OlD%C#MALRCY{fsN``$V?ntJS>YlB
z83SaQ1E*Z0BK&}xW^u|7X8A{33<oTc!LgPj0-K2ybK{kcbWQG01>BR;g#ez&5zadW
zZw3SzUAF>rp`Zam*L7(z1dzCDrY$H&hE*EMWywIN9s;~=>tKtjl&@DUQNY?SB5)L7
zq9#xPIX;hIKxMr)Td^~@T6z5*43qjda1&!t7h^~OUyhAMzEKS6sw|Y&A%*}@L{wX2
z!E6uO&Qw2z8G7c(U?P$8V}Y7$?{Tu~aJlKzoN#e|f!nw5;sg3H^*3f#KfQ%F1u{;S
za=TR4E-I&yv|kE0l_ai{b~7zg^A<v6K2wXMRt2@T5R!Fx9XAG`BAZqxCHXq-po2|`
zQ@^OvHk8|uso><w=ElSz$cZPD&U-Dh%ce^R$%bOkdL*+o+8-jKf+XRVU4`px$e9^2
z_8>6&!d@bC_P~)fw<SY?ozizaQ1tx_dIRY=*SXR9FUd-$n@A*1pN*3cBf|A{9`DZU
zaqX*8xmeZUYe>@HgoE1Tq@^%*TuHZ;-c%axqP75L1;h>kNUs1&_6oZfIBPt7sWRJQ
zyNip{0y&Kg_5qDEMHxS+7j`M{n<o$P=Dn}tqsQx{ta#3ZN1J2ZU0Z%F@S!xh=hzo~
zG@g&Z6OY5}N^lYkHGpazaS*qvd#u6^!eH25qx^RDjq*V`ruK=m>s_l~HBoAQ^hE)?
zfstS!RWQp{4Y?Jj2}mf8YqgYlvZGXX<E~jvoHMQ-fjY6=Sb`#-YTQbP0E(Flr`#nu
z=F8xW%vWg#Ewkh&$C)bA6ayG8bKsEe>Io+5V2c$yvqU8EK-oAgoJ)p}t>i=mv#@Hi
zg9gaNmR!hOEjaOo3RP%81Zmy?;Hgs^ZauWJASkF}0ZJjr;v2njO$1DZ5Wwyl3=dLM
z9~>Fo*)i6ul`hQC#oLMb2_S(R$CR{nk({}!B1!^U!p-Z0x{;u)k(|F1vX9!-*971>
zPdGVRVYBU_I<Dvg<C{PCP5kj6|J!5?m?0cVK)G?ckzlP?RB$F*%annghUH*32W@kK
z0cTe*M?M1Of_rV@4fK~LPe^SPO8;@_I{97%WKNkAxiaTE>mV7JtGpw#)UWS*E0J=^
z0{ZP1{dTMO5q*f-o;o~YKxC(IB$IsCP;jMnOjoYagp)0pf_V7tIF+Uti>eum@>>?8
z*5Z}6`)tEl3=)YCZ4a2OJ&nszIE4NF6W=`LNk7d|TH>kWReM+4y|Xv1wCrw}egy<e
z4a&V151I};2Z);Yd*r<<pMCO9&$#jI&@>tWch<ELV6V2G3pDe(2@K@f3YYlH-}nUo
zpJ(qPFsUm?;NkWber{dItpL_k*BpVWiTJz9KLt<*K&chP@{UFMHQoStO}EQl&-eO{
z<+J)_)qqoN)ojaFoC>Jz=q{7(Yd>F<9XszFI&uZ5F-<!?+9b7RMYe7WFmqsCtW2xV
zG;1}irH8Zz==<yg(~Pa6;GQLjXA3V&I-QehMFTc!*NmKlc}YnALLoj)<(Nw9X&Qv-
zc(6!EZ$(B+IS_WN78HaD*G(!BSzTR422oIp|HUCQlN?^Ei!<vAB1|p<=>XJ2feO?|
z;hqIBqJk)|$F<w7a#K357?QAs9P3_hHVEX<$B3(bgUz<brth%{J+8J}bn7)9Ja`>p
zyT$40F**?LeC17aekCr~V$5}9#wMCyUnzCmifo@5|MgBdTg2k*oz%O9X*vsdlcPLh
zlNB5u-JABTw0XqZcer=!HZHe4*2inS{`R|g_gmkF>sE+Y*WwD!z>tDl0i<Rl4+{RI
zOdv=*I|r|MqH}QGtC9zc(u!^>Sly`say)Og`4~XgDJUbK?Ivm;H<>z+*!%+#-hKC7
zfCzu~XMc+4&z~c*0Biw{iubiYnjhk*Tu|$u*7gd=G8auJ0a51Q0v|9a3p{A~O51Ij
zxSVut=y74(+PL5PPGTz%G`Oo6jTO{cJ;VxM0dOcQ<|L5?jUS?!XUX6wZweJCQp)-j
z$=YK2{p13MROBHzYrNYheD;~$NUO=MD~h=D*-ZTU{mX#2sb~9ux<i22%Yejv<II&|
z7x<e^qgA$Sn(JHlKEPYAy@3xNAA=nbK={GaTllSG)r_Ty{H;%nK5YeNnm})&O<TvG
z;No|xKIOAkUz+M$JHG8?9uv=+Nta{T>LByrB$%5_w$%@&Hbk!g@R+|iU<V*^mB_`M
z4?!V@*^A@EX^=@e2ndWa``Y4@xHTbLw>K`BnR}olH^Z8h{5o>UbcQvAL-uH492XPN
zH%^i*R-|Ew95~3arGQa}^IkYNw*Cn+D~@g{J3lvv>yE&jwG3JP>N12$7eavdUdu8B
zc2F}G)(@B&<f5$XX`!~o5vlF$xHiUsK1TGBalP#kIAR+kHhsW01kpUzUjYgd*E>Sj
zd3fh=>-0oy_MM;O$Pws5Wc40?^bt<F4&I3yD&v+INQxxrLqHEJCs92u^a11@LXhJ-
zpX~aXmM-ZLFjV{03?@_dxS?O3x=!sRGTwRnZIE9h5@EeQ#p#3BaQ}_hv3>SJxV&~0
z3FHlfbBG}%!F=C!;z0p4G)%2WejtE@0NHIXnz*s=5#mOVWH}vaM>exIQupMI!njtN
zzR5BgXYTXQKf^m;eHY*U_IL1uAN&BHefsIZwxG;6x^XbhHL{j%1Z$r+dDaZ7pvSP@
z<&Bp0*O*38HXV}13``cUlqr?=h-zPjphwy+K}@SC^1vx4MKM5}Ik2L90=BeC>zDi0
zkaMWuP3HvyrxSo!a`SG0GCwhc=HWsx4ch8(Wmu}h>;it}`;lprrXN%rER9uRulN(;
zCqdAGVP@k^EO?XPWRJM%**<B{Jex~1JsW@bz0#oKLGj%>T;tZ+6|T>Z?H~-FTpVG8
zHBPY2@mp<RoRJYYt3X&298ZL)xopEo(?FyonjSUdwA-)u3O~i+n+(&aN2`uI%Tqqz
zH3sd5CZ0ELt4;PUItbVW<=7B8J9tftXOhD~<~l`XIhfrl0dD3Lz9_IQCTTiT;c(`(
z=tLG5S=sO{C#}XTfF4s=bKRAFq-HKaDa#~gSezlh?1n2DSrz^Q`OHz1g&QfY{AUJR
zg<lc%*wv}4E>g`$E(GIQH7ccT8BxJ--}mTy0Tr7N(5s+2Gvu0OQ63m&Rk=aOK-aCX
z@*XGM8s3S_(#kvd&WoEfAo*oA8-tF;pKF2;9N|L5Roti?lYYy{@V-l4K7=bHOT71;
zfGgJtymJ7s!YZ15L8mUrJi7G?P_8b5a`N<hhR`SQYR8JEpy9g?jym+A$K%f)!gXuB
z{<U}U{=fM*a6J$=8*KYNfv>>&&7%5bRREFY15vDtC6EYUT?D!4H_RBAuu<FtGbuw_
zPV3j^$`1DJi}~$`@4bhI4?oA(zy1x}y>kcezyCfq*Bh`ZLP#Da#l@+Q%CPZjpx2V;
zoOjJ_W0Nf;IOXI94CVT2rHfQ(87QftSQn*ItGwj2)*OS@lthDe<5d0<&<Pi>=woJB
zT)#nN26^2;9w2`*PX_9WvZ8&&ItFcJ#m5}6DIk|u<f<4guG)xGSAx$5aTTs!cy@uf
zW?D1P_6jq#fb9XorU1#_Vaw-JK%-u8TjeX$ZhiBquv2B$KA>scy|o2@js)z^Lx;!x
zDNed)InPmAT*#cY%2!`E$~jrLiEpzPz+&Vh1KY&S^xH|$v@c-Vm0qjNGe~W2TparD
z(DIfzX&=z4bg<_MV`5ncaxINiJsy#rrUQ{^ReGnk)@T^VH1taOIZ_V*CKL_`2d@kl
zsq_>f0L(ZDwzgS*wf{>Aq}&lijSKY3Y#3<C63|<GU5pF}f`LAPQ{m)nS!CIuW<~mb
zt9H+Xi?->7!xl_#DyTdZM`>2|3gDEx>72vLd3f)z@(!o#HN5v&uUBv+o_U?~=*Xc<
z#+l_vDcSsEk#%z`pS6)0Nil>eA9s6QK%Em%$}B?g$0sLZC>SEFZpX^9$jWhHVBror
za@cHpX@d|1+^AAT3_UiR4fygxIAJPhmc9*}!bIU;*STcW_EsRgasN#?W?VdfflojA
z2*3DSzk<&{{t=#h{1BVV4TxDiS#UCHY&tInzv}spweLCOl9~zMbc|?6>`1oOSj$Y>
z7ofnLIB#qW<dqpFnY9@4^vP5F(X)SpcfR^neEU1!!3RJ237$QBCT$yKZV%w31R{pO
zRrw-yNT}JGh!UVlwo#8}2}B;I$?mjWP>jbKSf#G+NPXraaWqGmBeA&C&=@TY_Kbn#
zfx?c>DW4bTGEr3~-GrodC}{!jn1J^r`O^C)lpA4Re$75yI!aG#PGkerjVAjdM!ks4
z0;ZXFt@~Z^2hS!zLHT|%ek1M_V5{Gi_s#f?9mrf-qvzbHLj{BeX_xudlm;;X!fF+h
z@&rU28h^Zb7Z29YaTG5AFsSJizOFAf+HvN3P|mvCQ}vm6UbjbDca4#<W7JH6pt@g8
z6lFTSnZUc!nHz6&@%Jus;xJU<xRLiX9d;C}zF?iBy!fYgHUU%Rq?*-gzD}8mlaP1X
zX}7NnI*|y=lXmK;8km6@@Y!SstJeu2AK^-ya1uU`VBrQ;@1=ao4KOF~MOEd94gd*f
zBir9n!>GL+FK+vYfT*CY$7U->qFduQnZtKS@NOkXpT2|hE3A$@PRL=sUSYlJu=29#
zuAD>HIjmg=-^sD!%6B5?7PbIK_|7A;aN$VKX%m2R-XW53_Uzb?gMiM!B5)tHO=1K#
zn+-w;=2;gG5nDA?fj-OjU0q#?3>gfNs|gEH^F$x}G+=4q89~0StwZD$Fmd9?AaPz<
z%iv@eK}1k(7ZAjZ#^;Y8;=yZg!O;rW*Bkud@BSW6UOU2H`+I*2tKQ>}{?#AghyV6Z
zaD9FWgZl==M9bOuUqf#U>BKWZJo8i`PlApuGT||(m;p$R1hxGP$PtiSbwQdaZKypJ
zN(T(Y(BsGNzmGe2@8D}+`#SEuc!7^U{utX$ul+7+9h3^NZk#t8Sq7G592P5s-ipmb
zAj{fWpSPsP6K>kfXn9NSnvE&e$;UWtVl3Lg$IzURas4cAcI{UbB$ZC;=B$4DENcCs
z;s7RB_-d5rr+Vfs0oRzO6esoxHwG9~T&FODiM_f{-+C8_YTUm{SfEjkR{V)D^@2AC
zu6XWwD?U%>HJK0ZF4Gjqm<v}vqB>ndVnZ_!x)c2H*`3@A8DJ*-;gfrKxczOsdwhYf
ztsmkY_ddFCnUWaR!>)Ovj64iKYQu^4ZPmN=Y-gTiTo;scP$RMZ`qWJHo#j3>jaMnd
zWd6mbw$*mK@~Y@CU>6zW%+Rhb0A^($wiM<fDlU=<ot((_m=^<AY7oi-$uL8(tP5ko
zwIcwqU5qMfFsXenU8qUTgV<t5qLu49l|v#AI5jZ>wPmKw)rTJ2zQ=aE6%9K8o9kXU
zZZTpTVw&<p9}xNgs4Y4NM%NMi>IkPt9-Z10yLEPkqtyy0M{9K6!?D9^)hU>iMR2>_
zDu84liVRo`s)^h8h#RTf1}1o}8m`*@4p41RQf+8Xb%Gvg+7YKMeg>cn>t=tu8L*i=
zl+!!_Aok)%HiFSP@iHK$DFCxm2ZCs@79F6}d<02;0p`e_(K!Jgq2k=7qebga4D>PJ
zy^lV{TVMM&?wp(=a=?$@`w<@f$s@RRhp&J2=kZtn%3sBUyRYG2{?mT~4pD)q?h<1@
z$^)}#sj`B6GJm+{4lPeO##Kt*d^jSKwUPLywP9zQ6|G4229i8tpx7jc9HrO3{^ao!
zTwYw_8{hl}zWa;6h@brUeLQ*g3>*N<oQch3VH#-0sS`k_N41Hyg(NNj>m3xOENr3&
zTzH!pP9SNsWg8;OqdaTG@<0Q&<N{hDx#Gx~0#uSG1d?o@DP?59PpQAENTzgY%Ay4O
z^UbD-R06R$1P0O)TXWU&^EwqEgYaPpiBqOCm@m`Ya)PfxUAyGFS2LQpE=oBEYDQqF
zxR`ur_fx>YuIJuQEb>#}uDqOjKB`{}fK&yV`vlGJlb~5u?y{{11qA?H(J?-{T4PIT
zTP7RuE#XJc*ZAPs8h?bl_$v?A_>J|y#-N9B)v#RtR?o^Z%z?SO9R|;KwBsmlwfPuu
z@5=F3QM2t^<(r^gLt_`X+?9@{(d-8Mq_Zo`{-6#x5ZHCji4eUnCpocqmX1rEz&@|C
zaGjH5JK-FuCZXEeI$Ez}tqlR>9J<vJ9ErTyYPAN(h|bHxdvvr`&X<@Sk>fi;ANnkm
zO~TdV7;(Lg*ha?nW{UvAHi!mj*apPwEv~m4Y_?m35QJl;U`-Z4h*NqkuZJiOtUHgD
z_wXyzh!k+@KyY5b4#fyZ!cD~X8rWWAdqLO~i=rJE8yAzvE5}g5BNuXeY>NyA3p2AQ
zc?1P><i!(&9XmS^Jc$oo0apR+VzSRhPz_~{AhkW0j=zb-KAw(CgNk;n=LiTKbn#^c
zHjGHD8oQ=Gq3>PROeJ+r?`fE)BLUpo?M5;*Q#Vw|HbS_h&o|fjw?F&=-hKUbgt)<V
zxB_<pFCIU`=N~-7;}0L<+i$;>g#7^n!D3uD4HU}ltU{n!t_>N`kh*Ok0M3h6YzzR$
zUcaNXccdZ?G6%{T_|>#bc_nHY5<t<iY`0r{|9juZy?giY)py^)$=NAB`Q&5ty&ZeA
z^dU{xkr-W;jhD(bDhCp*3{~Z*C%HJ{bnI_~*<8JqvUApUrgd)hF_SbU_Zzy2lkb&*
zO$w1j09f^c%JK|3g#+#$bCF8Q4W_461JPHboEHY*ZcNjFEypteINb8qfRL-u#dx;`
zc7VgCK-@%Gn~U(=g4{@c_T2>Vn#_AH?8IU{2^MDZpDb4caO?$Crph#t<~UHhUIT4R
z!t;2F^WLF%*#>3+fROLE3~=_iJYAM4PUNVqr`n*^hI8##=2=~*M2hgqU8aLDV=!)`
zLG&GCeb-cQ=v2`RZ{@GY+srta>qm)7fL=yDe11C!*nRD5Uq|O5F0CM1a!z;3Vw6_3
zE{zkWW=70#XyA~HtA6V{zkqk&eHXv~d%uT=pM4I1!L$`wtr)>9_Scw$YVt9lny}1_
zZ69#G-D2DK=tDpsqWT@x#V?BFTv`YSYI23J@*P&rVeRGbWVJ%KUc)&;=RG{h@m=S=
zE{^7cOiWlf$`f%m#0!A?<Ow6P`CiR(dP(3O`UzPkaj-6g=AhjGLOB2v*FyPNL~3V^
zQ2OtTsx%TgAoc-S=w1|8ymx9WSh!eXPK6V|v2SSHv{EqLIB3@YT;W=()ILr}!fJq*
zVH~r3RAxXtajg^xnZMbg5#D@c$x#5s`yc%n9L{mHVw~JM0Y^fAy#Z~uc=pN1_`$PB
zaH_p)!@`d0$b7g)*1oFCag5r`3R<kN3<208I4A8x4uID*#iN7ltkw847hX&3n3S`D
zWCOF>jgXWcG!16PqsNbMe(?ftzw<V}{<W{;qYppA<>gfZyoON?im)?DQ5h$;@gNdF
zo4zp*O^RghqI?enp(xtERAex%;B?AbkcBN}xH*-OQeShf0&a{1;5dcN^L)wGOKU<Z
z6AkO}O2ok6h&q=WsYFQEG<eR5<5mLOGIRl5M{g|;s_w^Otl?`eTTx^STQ5LUz-sFH
zmBZCQsTFT#K^y^)xp-6Iii_DaMxA*goyNntJX&!^?UA<tPr@l)Z1aZ9j!ZZLD<^B)
zS8u<-caOeTd0UyN%Lv@@M0sZbTI+r?pW?dwE@c_(!OR4Z*9z~!J=+KC%mPGI`RtKk
z3#{#mKKZ<SF~O;v8(%x<AYk{=haY0qQCeVaj<F*<GsQ4ef&ijT^IlB~$B6r{KLM?e
z@q-UO!jq@Zgmb3$s>Gg`YJ9R;ozKFg#u4<+sp~RA=NwK}9XjW*?mDczL+8Y=s&fvV
z_vrMH)_X6b#B5`cxZ47W6LwOLi1R*a*d{Lu<{y<6C>Xa8wg4$$fO46W-&C87S&#fg
zsbDxCan?3A<Y#PZaw51MVjAdPE%Y0ZB~Qr4ffY>Yv8o*b$`Pr6s1T#nF$5&PxL!D0
zY$^tn<;-$tR1TIJ`HeBctH}@Hf{BoS<Ot{j&M_tpPCH^Zu3&CMGlODzsY#0v)vE$w
zMJpU1dE9yZ9&X<|!^#t`uC8$L;u;-yI3mFH*EoN2iKNs3<ZH^V<kHC?<ok+e6P5RL
z>JE*?pKcCkw&_@YqRKF;cD40OGcA%x!95tt304sT13ok~(rX91*P9LA|Iz!nd*?3R
ze&?%r{P-~*KY9c$qRPE07coXrhyc=9wDo@g&=~+{Sy)F--Gnhp;4t*FY~mNJGKLO4
z+XhY&$|td`9$2*9b@OK)SWKq2<hJQ4IbZ-z3&+r7SNoC$-btGu%kxG!cM(HiSA#lf
zA1hsXTsw7io5{XR=IT)jW0fk>*auMYFo)sfUo9BrVzy7(FCA`Xftmtfvw*8gY@hUI
z-|>$8TIFwnnjL92pOqkcyts3Bg!isq!+BJtxT=UQGJfg)GyLs4zl)Rb3}s&%mGhMW
zU8|j{JnDM10No_WD-w+aP>{LIY0arhdye)AHX0jsq<SpPBs%w9NAD;4=Tu^YF<pJ9
zJ?S7|*Y^QL9fTvKd8XqX^%2U9t%5UVk)_&3IYJYY6S9b1xIe}EhwtOVk3Pb-?-OUE
za~(P-#$k_-WRdKg!`geSyu-S43f6=}Vy?^_0p1l;8Aph3RsB9O!rM_}RE<U|7Fd|w
zF6C&Dfj5B6C@p7gJ_8BDc@n?~ZHIIQ21?%7?UQ(w2@E|;y48-032HA+Y7&ACT;r-i
z-C@}Rs$4$g1=8%YaTH)ns^nl?DkJk>TiP&_3#eaL{jTcUv+iH%s8J11o6Ka4y}Cl9
zq@<yx(>$aPbj|~2C|HmUl0jlED?1k1e*nC4IJ<X-x55?9Ub}}+KK>L4jN{cB9Xw8)
z$L*sI9YhWd3a|`|1J&@;GMJ2*$p{Kpl4N6{HU<-CkEGs2oO}XX9+87;iDyp~rFN-J
zRHSW3Q1da`ht0*8Z33)6MrJ&I`~)vvoZ~BRe+9=U@8HvqKfz`zHmwXeIc-DfgJuj|
z`hMa9TOTqY1yUo?BCA%aACzE;a|N=G*`SM!l|6wniBU*L4*Qp;jiT=XI#jfcBA5LV
zX$;*6cAF&|yLLA|jxC=J-&?%lqA}#u6`ipT!x=ITY}bvh8w+{)ynY8@H(;cRMZCV|
z;leWo;zs$;r9BaD3V6(cyV<-N3+xovul0U1{XOFB^A5lV^alQipWMNQ2`k7<;@I4I
zGJfsBQ~ce#zl-DkDQpuzYK&I9wLsn=>O{OIxSOeO4dO<5&BcY*DFD<VnK2lI$9^;!
zj@spD0=<cH4&uzcZzb6pFB6sA)$TKN5U~5~;u_9v#pftT^vOP#$`i@ipn^1c<<#Ru
z=RDSH4+p~W(Gh<AyWfF>Lw9)&a1^KB&Z%OC0X5m#JKw3PlPIpKpyo@M1^DQZCtPXo
zOM!uw-C=n2laBpZk6EG4%?)=j<XAEt1C!RrgCNh%Wt!!~T3@l`PYEz)&sDPMnj?1U
z_xZSMrJfpCWe&lBV*#MBx)8%;D`bI<iGgJG_RhmG175skcoMD6*c%{q@Li|<JGVpG
zK3=)2L522>yOM?5g`;V8HTPV}(Yv(;!~n7LKDCKcvo54yAgU(~10KdHGVN0)pa<jR
z?k)IR$9NF$;-bIC-8;7tHvzr}PF#mu-AZK7m|=qYguD(V;FQG5OgV?r5ibXZ@4U9J
z9{;KlX;OtB)=f%?Ky>nrh@F0yY&VgdOGn$}9Gr5V&4;alt#Vu^^FISWvJ!5)-Qp)7
ze1O+pe;x0<`&B%8{1_L{Ux<q~qyd-AwWL2d<+Q1vSj!7gPNOI`m?stU4l&?>v2um&
z>yiN$)fO-Mu}T9dcRTmX;k1<R%I{uk5SQ<Cq~MFEA;*q#jhoNL$FBPA2W92-hEkk-
zKU!1<>vRp)^rHEE?!H_sT3ovcfY!Wj0i?O-Wx03dF_%Uif2Pd4;@0u&#eQendjOtM
znB82RfBon#J~^k9oGDI(j`1J9^C|w5Tfd8Sf1z_{UFPC`0IbUEUGM8Kje6{zFE`%T
z?b`TgApq!wbLTo+2&QQtyp#HUEAR}c8??#H#;G+o)$KWPJ4$D2%<NN7rGtRov+GN&
ze5c2Aq{<ZHxnQkI6^;mN=f$2H5jr9`HA-wUM|W=D#?O7_O}u!1j$2)qTy^D&L=cpd
z=0pfTDhP_k0aDHz!09*<EDN%6I*eT>p2W&swtR-Ph&T^G<oTFoaSc%o7we+$%oYYk
z0a02Xr?PM60fK}0hjp$~cAo&T#F!vX{$By8jW)n{9Y7I|voBEwB{L<?$x*B)FY-~*
z3?GX&tkb^fRl`)wkQ4w%IW-_T-V7?o1)`);=_uO)+-`fA)-#c~+@hG34Va8eONiW$
z@-RQC1kp<EW6*YHmG+Mc3<=)JH}jo1?6vRiSOAe6hevELu5iIy=}X&y@Zt(5{s^aj
zg<~&fSeTGNu$kWgBiVZs@c=!RG}%NoIGwoAy>MM)5RFx4pl2~r;?ww5c{cryG0G8n
zZyK?L>~0I3+(RW8p#4gbcH&T#$`??VW&{E|NvH1vKK<kqoSdBCov(fsc<=xpe(({l
zuCKG)q9iGs8Im$tzaPzECjW<2KxpAH49<xQU&`v)QERG-C7}}^X}<_ICV?SfOvT9Q
zjn_+S6BG+UmRr18VxTI)5QGh%jVAnbKRjw3#vjI^)IT!JOTi`qnLV%a9xcFAz_1Lr
zCrE5AFs(f8x?cQS;pXB`g=sC6GkNXIuZ%ml2-kpTu1t+Ud-T<Bo<7CVnsK$2d+&hz
zrvZQG?H}W>t^Wu<To&Bi7MPm4o&#HR@g~YYRfbV{s(kBo=JJUFa6RBwdOUl;#%}T(
z)oF|-#3}(ovkt8^Mz8A)+uZl3c9=Bg#`Dq~9R%#Ydh0fhRx6N`3EPts4-Lfd5b#4J
z;3Lb~$+Pk};`H<chymx%pTUlBrR-*?*r_Wd0azkTPascrvI*u04@xe~(qau1M_+0_
zg`*xnQSwKcTTOtChZEa;3UtUN-Bsl}%0ex&XfC<evd-cNDr<$o_oPDR%-SZXMz3=Y
z(V39F0#B3ivU*7(a1MwHI0y{ixpa&t?JNMeTo?nFCPT!1Sd3x=&*^cwvP1;{GD8eX
zo1M7`ave%jC?+2TWVyKXWf@{;P~(RfK?<74Orb!INy&>`m-%1?>>G*AJp)1<#4&y6
z9Qvy*+_p#DZg9pPcTex4a~^T4K5Lz}n`4l9J@IP8&%l#%LOb6f5)gYeo~-ASfmx3;
z)hv%HmjFCaEI|UvfN`t?t0xl(z_l-t<@=yE=5l{WfyGM7enPd4wc|VQ!HBrJy26is
z_(Qz*;5B^Z?XTd`!-u%IxRf?eN4Y`#z&H~x(C?s~Ch8Hb0C#{xR;rMmApP6=lgS?z
zFcD@3Ce;T8;7Rtn7~s5VD8rWHHKxr*26|=qlEY=wTpjLJDW_Zx<PwA-jnxvUwY6s@
zZVM15P4@!)0bIEcD}kI|aQ$4%H(DI)IIZ-`G%yS|3S-x;#eF8sWLc_rQ)Mde2XT2i
z|K_uOfXdAMWFD<B6KynC7Jcx`{yqF(zkP{M;(gqS7dYWdJUF_--S7m|i&EAo?-r;k
zuj}7txK_DZZQQCGH?HUMs}i02E+GJ+2mD&OcZ2O?gDPM)+b=5MMb&*OOPxracQr1@
z8IAkNXxx_X>+v^JmkBxu*sVK<b*JD6;)l|)$qK3YP@R-hVW^n4SZ>eGP5=Ps&z~1S
zK{z65QrAVZBQHt;lm+_5?wW^b{K=#*91szJ&oWN3u6b#cXxs!!H+4S|?pme4#!)mI
zRU^rH<E$uH<cJue_=4?(i^j@@v;|(;h#<$#cHE{45O##6F2g`_gend-31r8W9SNt)
z*|Q$bcj!4{%R#`fa^3(dpJ5Pb0C5Y(ve@?_9jlTvP8JfthM8euq+>e4JpnOrvKbgt
zTeyOwDs7kSjTwj4FgK~U4@UY3b{yKN2Mp^pjzmM6nBe<}Tise-1>s7%#C&6EckOc0
z<^Wbcfklw@A6Z<TljHRmfj&fZ-jo_LSIWS)BZAkD>b(c4Yc~Ouq>M;W%t_sfAaan%
z^>lh%+Brt(10pynTv)wB6tHO;yi!_D?Ew&Ci1_r=PjP&5j5pqR11G0vc>e4uwtcUg
z5ZTy-J|EW`N7cp*QkzLGX10ZNLMtF1NPQnmoMIYd0VpFNYhF!?wkS7N`zSmN?dR#H
zL;K`GGu_@)nrbn?rX6Ph^5l8|%8D<A=5>pTMPY6c9mM9weR)e0ugXRARl!^Fpc!Td
z@Y)%_RgPBpMjx16%y)s6x-A<wlj-kK-d4mStnaV!^Y{^do`0Cja!n%)G6qmn`OX2X
zM*LBFt@fzvHI-gn#zq;Mc{JX|0Qi;s4K!zqzCIWdtuaxLr|N$14Psf3iE)_TO>~w)
z8&oqxYy2HJ-yP^u%!JI1Q)X+ZX-(?Rn~?pZ+IL|C7mVYRWAxh|7Z+DDxsx3ODiM+q
z*0e}C5yS`SaifBhq(xZRM2jl_W}2RnkesCh3izB)es5z8V9K&R1Y2~>yopPT5<uu<
z_WkKd-uq5Iq9Y)L9v~zR85EfLZrI#?*rN0DENMhavI#7yfiroZ7DYj9L|Rv(B-`i6
zsltMS3gu>&oVFN)EX*Xz3*P$-+@PE({Z<U9T%)kUL%&VWQM9#FV8{?J3c``}U_d#4
zF10s<#5;zXHF5I65m|f@2Sp9c@~v6Ft)eRdb0EOv^6W@j*>+@HA4kbtP3N^NAVfsv
zW@ZpiE=$J|>oKN!4S^%9oXCedAi09GV@!Zmafjf&D&jyjqe)-_10eGfV*ND4WV<fL
zn9NyOo~9&2Y5b5HTb8G$*?V<$g%3Xb5cluh$5-F^Dz2`su-R-7W5A2cOZ0q|`U&*w
z5uoD*shq~aC2c3CZEk*(Z4+STX}pesR7kxZZ0z)2M_%4YSvZvUfcfSfPqMiT+1B9x
z6Ww9NMUKPYp*AZQZ!I28T$gFs9~GS0LQbz-Fh+}{JufH!8purmigIylfxi(zn7E$>
z5_Oms@R&-gv7nXlM}S}^tyY?|<!U50u_ll2>oQH$W8ZSJO$0Weif7F@6LlGUo7S`X
zx{=nXY^}OXrNPy+xpbzki*z^uvCrVn0G_RhljSB1$}}Q<U+=tTr<ZDMbN#$dV4^?G
zrNNCbd(vT^CxGOoK&CDImyqra97I`(06<+K$pYSak5$(p#DMF~b=r-WlQfArc>ySz
z3(UnBWHItY>cJq(+?fqDD4=or(M`KDf*~SoSym{yGy^F}5v2j2>SW;osAmjf79n+~
z-V)UIo@s}cN(SER(W`bo15qIath$x_u2zsmkMaVnCN>y)Y)h<m@tks31PD1tswGV5
zdlAeqPGVU{v4N-BwCZD5I7|#i$XpOoJB6fjV{s@~<+~2d0cPMfvT!g>6WF*Now|PO
zw*jnP13dZ6_h1%zH4f5OTt2E}UG2Q<<batRy44Ck^m5dwK$ltME1h;g=fqZGRMs$A
zzar_IPVb#6Wo-Hg@?Jfbs1Mx`z=(i%i7RD_K@0*H3nJ@a4Iu)ZOUgdVriD5D0HjZF
z--~%3_6kbH+>IEM7Fj~49AJ|b^xCIQoWL}9qk&wIlr2U)diV&J=NGvD;650PC(oX$
zCVMKd{6qQ|Q78aKX-6QN_(^a~5A@k&CiSQuq#FQ&9i<OWF4A1aQr}`8DK}&#=Wc)r
zbsVSm*7Do@leEA0-f8bcwYd|iUsnAnci~xaBJ=pm`#Cu~4AL3T+tzqf0I!T!FAUXn
zGh7V@_PE|F++4c#g8eFSC(@dFHkH<B!QBCDCe!EX`_{J|JTE{;`lWHW*>dnW+~Aux
zo{hpZ(rVq0(wM1F8Gb5WvmA}~V*p{RvUgp@y{j~6C+h;#V|bWvjqb_>8yVE*h6d@@
zbs80JzcgNAemjg~mnPq04@+XF&LzziS!j|lIAx^?u|Oe{`Lk!wu-$CR#8pO3k0g<8
z<B8~$OYWDfECf!IzbzDIW<&Bc7iIeCJ&DHv&HywuKik5Txj3vG2J;-ST6JJ%Y@#X)
zu*hmT*h1w}e5YEJrF$vU?js_2RYJ+ry=exLhRfQ&a<ojz2WV6?98eAlQ?G!d()?x1
zOO6`B0Feia9dm@(r!v97pm9UAlXu=DBFa%HiFXY1$N``*VWZ_IQWtOz{dNmS3Y?%E
zMi{3kyEaS89M+EBj`b;RBtkDfZhi4xgL*FLV-V9VY8abXWarqp#^yOfIf}xO16Ez9
z3IR|vC4lQX@xK~FlBHunfDCK9TYuH><<x~-h_(^o)Ye3zxlQ_QdIm@lqto(8SVFhz
z5JMEh+MK#Nz-ot3b`{C-y>X>2Z35x)@(SBeKEwU{_weQ$Z{p#jhd95wR0SIaLI#e=
zz8$5^Oxw<E4hH3Zf-ztk6g)W8vU8+$0&Fuq)QC-*uGM^E$~-loD<?_YeDL6ic4{(y
z8RFv4@-8R!{p9KL5o|pv7eJ5m#xbSXn&is`X9}RqDnLvul3JeKU~THUwcxkPuq#|+
z@!RM5D4(hOR`?y@1!ym*JKJ^+tTYl!--D}XJS^*M9#d%**MqX`k=`6o9i>;7ccMPK
z(w*#AbsVulD;k6y;8pgnK5nhsuQjC%o*RRr3Of#nay}TPF`2+rf0`J_b@&4S-a)`F
z3psP*{^*$=3?IlEYVM$rff!^Tyjrb61YBNTCesV-eN|K(QM+a14vo9JTad<ryK8WF
zYuriW?hsr81b25Df;$A4KyY^ngplFC>(0!5pVxV*hdSrf!&$pdt@`%2_m;NOmDgBb
z<Js92n)koZhyRE0;RU9MR|^HP>Kqw|DSTiYj-%=|U3Co=B6K`&E|e<$hz?MtK<KJt
zb=o!u2Ri)o9&fKTVoXN(G1N+Hg>)27nl@2MnIqpBg36!$tukn300d2J9o~<pu=g*7
z6ooL@)81&1l(+?nWZi#CCqC#p);8p=+BDuTLaWWOkMX9ft)i#PTdk1uJO^xBC)oVZ
zrk7!%j$Tl*qI<GzV>(5G=>X)P7*;!d4dzXgP=X8*_#=!ILrcEJBNi$o+2eCH5fvu_
zrDhQ425L4?miuarj*Cn*plzQOM!!)@-v?pU|D|0~*qm;}%=l>qqz#)2i%GynXvBOk
zDB{C)B3q`qTKKLzk&G<1^Zz9%41NU7sH(aBdvp9`TCg{tR<3}X^FwgSWRkeZjeBOT
zX$cJxXNRrT7x%h|rsu~^-0S4aMLb)*(*j*4v)%?jn|#rSY`rE=mvl>8ULBCc%8u$}
zBzJoq#^pp--p?{m2MEkQN?+%y&W<apxH2(kfx(B|t0)EKzFfBN+OPgM1<bQ1CvA{X
zZGBQ`xlUPS%ae+0i=7&sCR|A;f0lCQ=n47)+nr;pPIbc5ShLRCAz#wo^nvl=HrY1%
zJvNwnavOAK6jJq*evPBGgJbk%rm_^#Woe{>TTOYn9^0y^pG<I=JP^;&og-l{UC_7<
zIDyWQNApg1gn#z-j(u>5n11j=w1MQzXKpnpk@1bCI8jo`*$sBy{-l27cGCx-cK{BB
z9L!EA{H+i{LY{|A`N&}6`-be3xd^%u2vpy&ePs`-)2>Jr)&pmOr=ql>08oh2G(>rv
zM+lFE|5W-S#;#7Cb=G>DHJ(oT0{^ZY&<*}()KfItvvv<pO^G7Mzb3CK)itPrcSV@j
zVW7SvrwUS^3Bev}GsGkHqn35^2Feh_q4F1*N$kTkM<HG5A#;~Th*XJwypl$2{6S8b
z=6@G`>XVuBiatp!1Xne1@?QQyXpLx=FeC}|XaGzD_l0rxqxH`J$$%TtH{M`fkk7-K
zgo%MN6mad~R^PdBLIRNxzhY@vvGm#%rLuH4V3I7m6^r_)n`|U=y$i#w8OvU4IS(x5
z{fbr9&EmNjcGluN@=T3(FT(<S?`qqxe~;6?v%?t9LJkEZ^rR3r%Ovdb&ZV`R`bX#L
z9{`8WJO;r`q>2uY;nnH9`Cr10C3u;w^q)3vWhL?ttUpVEGe4~S;H-K=Jml#q8PJn@
zv*ggM5H~V7KSd}W<78LnU9w}xsU5a(fXO!eR7Nm!=<ze<Qmn{g+viVYJ}k|D8e7f+
z86cC|Le?DvS4JcWA_fokZs$2Sq;5y23$HE-fs;Zi%hqx~Yi;>IM>0WYtgc@D(JyiM
zAe(?={RRu8goUxa&Y<n+q3!}b>wVpvsCgxT@Z_gf5{dQ1Vg<Vc8o4ei1E%b(l{3D&
z`u7OQ4hzN^^dsXY!%XF<hE6(<Q+t}&3rRT(MmA=|eUzojEk1tA(za-XR{Is+EaYlc
zfJ{7gF9j|SFHkb)OLJ2w8xwUbD2~WKi}xNrD?Q)~$UHrVmE12hZ7Ls3((4W<RR^cB
z8#_^w1u8wk&DzY+J3!&wB*l8DWwnaqO@ZM465#$6Qu|cBKQc|=Q*J(f5OSd14H>vD
zfFd8Kdk`(b$GK4t=-=@+IT-US*2Ke#rGChkF5`YjFTM;SOrN{<cnohVhG^c)lslbl
zeKy37ASy$Pz-*H)`k6$ys}!Aoz=kwghmy98*2N=}5TAa-&EiP9McnB(g3}+!()w4u
zS?1~)I6A1vP;!sIpmz~dEH#)S2>E({hg{P%dg>$>ltIYN;mV{;hQ2SU4=rDseZlkW
zLBN?gJSzM+xK@kX^E`6)?vYDe@G&3+C<9_<;Kc8$WTA?a)FnPCGb&Xd_A5&A`);<k
z`l3|OV&@*rGU}|UO?<y$PSmZYzbROu#H?h(@A}Ok6D=%}6}>uh!0?MY*-uL#Ehz!#
zIK+f|Kp;c|4WO)T&y%nlNqTkqfxR~0%(>2}7J)S3gAZGHJ^D2Zmv+D7H(tWeI6yv}
zY^Q$&g6n(1z;@-5lwjW482@zhT<2!F-t3JVS-<v$tSYtn!f~iE@Qx`zN0zu$g~)c1
zhNNBPZ*Vf$vWDNj@J)N`{PZ91z2*e&7yHyDb3{MxezTd`I~O6!4IFG$HC1scBV%J$
zF}!4JG<|&($-^Qj0Yxhcl6VWcl4pLRd0akkh9Y)qg?Wu<d}&$+ks<*WUOq1ZtoT&8
z@4;^gLMnx^^9v2?p7D9X2*6s38PB{pi#8{8sv~-#lzxR7aC$g3)xN6XCgiUA)D##F
zT+sM>M-h!mwh&?|ru)Gye2jaXBQ?7wWWtiUDBV(@5lRkVV!BR$@nsB_lG5Nn64)W|
zuEy-yqT-B4bTD%G+{no2(kI<FZ8ccaiBdMKOt-@wMxki7ou=lE+`i!?LOf+bGd$ue
z#Hf3P_JuJ$7wy@FN&^r}&G31222`U%>a$cUt!=hZ-EVT%lq!!7W6>zLaMi`piL7|J
zDlei1rY>|4-2j#IBnn~H%x=ucjegCR<(?t<MyoXP97nk7QU^Yo5f$m^AKWWfbL(9z
zZA?wq(1765bXefDxk*hi8$;5PR#K5cy{W?bn&Z^_BLQe>TBR}S0t&?<=0(a%AzyK%
z5d(GGNXEXR#_euXx2qPm5$w(vhnOGauJBTUIZXcMel!uT9e1B%Ulkrq5SQP3cIskR
z=fSSa-X??Cf7?L*xF*Zn+lCTpv#T0;bn3$p$h#%GsgYL<4mcW4@C}O-eUW{r+233n
zmmKf)$alXAEa*UNK@wM*pWw0<3fFiQDx9}1Jkcs?%9wZR+e*TO_2!-!9I0xZ;8K-a
z)>L_%`}E;Ov1Fin#<KNv&b2L?nju*}y<F6W4ycoJ3K&YV=0Qt!YNIAL^4f>0wrI&D
zi`Fx1s5n!dHj?w^>6&(cC~Tzpn3<?3(@2uJ4GmDpNCxXW!YX^y;lmMpx0wYLr&6uz
zNfC3b%(<Km=GApN1rS_joNAdn#*MHC9a<GBa4UF;nP3VIM<%*C!Z4jPfwYpVgP<s#
z_S80+FZQ$z6mYob(T31;mR9i4MD=q#)1}$aIT{QZ_6tUje>E@(na;*&(YL_^9xDwO
z-2yf$A0lcKeI)(s&}+R7#{Ye$+M-tj+yKx&*<aRtXh?>DeLH!55l^&K$lb~zkKZE!
zq2d`Ai!V+JoS$VLV>7l|T~?iTwcuBtr;g_#8{Y0d1~-=Chy<<l_R~(JZK@WM2u5*&
z)hC!qyd5q<K|FLRJ|W=_i<x%~2z*;(6cp}`SlRZ)m@VvuJ`YELmubsgNd}ly;QWA+
zJW5VQvb7tadJ}qecD!uEgwSfF3c_TRoWsG4&88rqZBD5JcmuU#Vx#pxQ`2eS2MxA`
zToUx+<l1UNv;b{RAzm5P6)lMUp+Em-pPBv2gHfjgSiL7t5!Y_T%kyJ~Dx8&aMstQ~
zzHT!|LR<A+{f~l|FAsn5I&$HHYI}U#c?@q1%?+NHXAO#M^gh>a>)9{5myYz7m;g+C
zYBsxj`Y>_roC=(JlKPBkmC>kXMyl^vCXz$@UN*fOZs8?bd5ImqNcvQBPsT)q<O?u+
z#$%s7pe7=7a+pxv2MKA0t493<=(s<~toN;@D_HRvVoOJkcMz(AY;EEu1L||qnL0$)
z5Wn#L@j@wTfQ3|yriq%7am?h@a_96L08owA(1|kTWE1B)Ghp+R7r5>)Z72}LVeNG}
z3<qR9bn!*;#z?>l)(`;Vkx-VY3GBGQP7P>Wd0fCgz89c;IP@T;ZRl!Xa^~>7%E=&3
zW02L<;z7qvRzo)L`5{X3PSEcJkax8RZG}`WIV3nA>zsT?r!ef-;JT2`7KeU<kpDq!
z#U5M#>78dMX8zUX*C!bB+mKk3zu#>*zKt`(LZXz!BYT)70T#F6Ukz}%WRRW_ExJ<z
z5$}HnaIX~SnUB8MuXT(8p=FPcEsWpOV3yI%hzO$0nzF68@y_V6xf|Tc^{u21!W0Y^
zvfgWU2psiwZtsT+S$^K`NgCs-QyU;=4bS<x)Q=?qu{8}@cxJ`Yn`?-qc=_9oP#DY|
z;bf`#>N2IOha%b(Mj<a?&zIKO;vC<y0&wqu=Y^(E(cH#(%`=H}tDiUZi<R**$m;CW
zXQ_GMuGnZZTHqVIuz~bhM5U29DYVHX&-RDQ{t_&v)`<W}3HM-IcoqdZ7r3J6fq6W&
z&zGS}>|*=31G{hR4hR-NOJrQH&}r9wDG6lgn`iph#G!TILT?t7iMR8x+gdCP`<k*|
z$P%VilIW2;@~ey^f*x<rsjR+urrPhv*yNychNChPjBq&_Mw$*!fVr7!UAh(ZxW7fB
z5zpyqkKFcCN8eAV)oBeZ4Ix7Uy}OYZXR-PWyu^N<X9$CM>-~AOE}^Y1co(1_F9Z2G
zB?)lYCyHN1GDddh);08^>jMv8H|lwi)eOFR$B%VIH;1G2l1Z_wjN5~5=PSp5#*mrw
z>%(O6g!ti`Wkf#xWZVe!UOed~-;M$Hr+<}G*Wyt9fr7#%gb>NnO+;g{eWuYC2-h4%
z7)s<tU^$4oAM1`aH82!v8BIk}i2645Mne6B-%Oba(?%5FT_A39yXsOeGo$%;fd*NR
zJsDtkTV@4p_U54tal+dciZnYdoXZ;?Cd!S$1^8cx!tAG^q!OXv!4*;PG%1q~xjM$L
z?ISja3b5C^!{z%Wm^Hu1Drgz1D(MSus|qIO-ysLV$En}}hQSO_uQD@uZjTCF#q~{S
zOfycjFC6FxjWxL?fOp^-pFbD!9+>!RW~<wGFJa~8_b*x_W7Ck{#3Ldzxdn?U{lJ%4
zZp*ZBE!u<dQe_w_IG%}y$UC`;WgcGIv@>qRAdJN_4Bm$qYlQ&4;fQARhYH~cR(uxs
zEM0TQWC2b?y?)0KcBQTmD{pp^P2VMW1YG;Xx+B22a$eX%<(W=YuD&|s*}%owyl{Nw
zo9wjEbSWr%79MO5lQ(WF(^bo*Wxx2$Kl$SY)8pihOPTeJ>b>Z<^j_Bc>G{oEKEh&{
z;8T8ds7yW~grCbXr201hKu^`-;^t#dz9R$$^68T7spva|oGGMWe=kFO$2cef=XU4{
z2bdh;)@owjy=77Ics3;x=Et7AzrV^gbiTdRq1DvDa)h!b5<Nb#*2pfAbeK0HAcKk%
zU`ZmG1k&U3YcCB!kv;0(CzwiPPix0;2S{rS_Gp#LOr^kzWP;=r2QSXF!&;S|Vb`P;
zw2Npotcg#gIL(GNP9YYUG9gwlJ$d6c%*gExgDPJ`*oddtP~jKH>y?bUrS0>Vg;MPo
zWrtQ7?|4*ug$0jbFe{y>o;d6ca}@BKC*@A%^u*ijI0F!hXh^eLCG}Nmqgi<4#?w_3
z;fB>c9tC%M)JPV_@$rX?>rtOrB|oHxh>?G0*rAl!*hnL!U9k~JL=O47DGeKX+sLx4
zaax|N0{6YI_HY)M&B|ywnF2=9x2D8JVsrbR2Ue<7FJ{{G%#JrUyF<tmRkd0ZeF&Pu
zRx1=4SXx4G$=@0rKS7a1|KvkX1~PJO_kG(X%UoGGR-ukUp5kD3TJ4}izqd%~n2WC=
z;^F4Tm@dO}_kl~yZ%J}Pv(_shjDZK5X4ibqc-jlaI$A@&tU&!3t<PDE(y(;pplZ^4
z-6m+qKG#62+G{5;FgIz^4uDGg$vyW=Z@zS~jUJ0$xfz$Q07yKne-eK9^5o$aj87%N
zvZtKSxKZ@b!{JjcGVGgkR&vt!W@4SA15aTb|0i{38rje+fZpBbCmh`wc^=ZH;?-nn
znV^qQVk+4%9yGUdWydeyGv)HQ`}|}7JC_M7=j%;{NwdM=R9fRqWN9uGzF<Y_6k=i4
zebIQ)KKNCGao^v+bW}VKl1H0iiHbcF2L>f4jE7=mO&0I&=-9BdW7)$Io828AvcNmF
zZ|t{*jI~J7_tSd)Tk@)&%CMx{>+%LVJnp4#uv+1g*Fcx{58`oc<r+t*E!&1M6c{Gr
zd2nwVs~O9KqL!>;*h%$CaZm|>xAUAuO;)F2UPbp7XhCj^$`ie~*A-=AG_(y_2GmO<
z7kOI3K5V44SwXFU#`_dLNH3M;uf7O?;6qxqy|5hBHi}-f&W|558B6e4aCt9WDBKnY
zfxF9}0$}qxyxq-+hCR`;<LaX^^V#|;Pms9~u-z5LCDtRc6ZUMIoKCZ#{DhD!S8CYk
zZ=(CFxPCuizW+t-@&_^4W>K4++$B;KY><V7k=;Eq%#LZQYP|t*S+t4X!zzK25_;jZ
z2Xs|5F)az5ewO3{;ge?=b@lIuPX0!R#~*Ybm#vYtac-lPU$>{IAK%82e1HE&;SZYY
zppN-Kw{WS)BAospQ1~5jj^{*f2#k(C{dLhy{~vo_e_+C2#y<NAxkl^6XY}HNxjPS*
zf=p;qZ2t!_Bzx+pI>P#S;odA*s)xa4^hM-$*51PhBai{w^`cj=?-jDvVLxk<jlK<F
z9QpXe!Uk);ZWgh+CS37aPt}>wrBRmdZ=3w`b?+6Ax?87L{xdz373T$}0CMkZwxns1
zukZbX9V*IM(k=)!v(gmx4Pw+E23kq?wM^9(YXb#AI!k2TaI2mO<j1VK1T~r;r>tD1
z%?F7j20<WEgOWRUHX@jsq@l-8+fj7)GuS9}LU0)IR>mJC1-U0smAz~T>SD>Q4QT09
zx1b~WZ&K)44MFR_NTDgi1LF3JJmOo<o%_tNupiJYM>hJ9K6M1n`1#^+<Ys%o5E(P(
zNi_1UoMAhQT`D&P#KVWiOp`?n<0fbTR-Ni}zG$#;8u8dOG`kT{zu8M|$Jvxk3~<~2
zz_FnTyI&X7M0|};8><Wo|FfGP=%i**N8bREYjEe+k&q_hwqTH5pAN}e;btW1Ms#l#
zkN%tdeH0n>mRFwf>PIL=s;~g4@4YrhJdYr%X<MV`DBfA>jfb9iEQKYlx!spa3D7B4
zS08d%H3*?+<Y@MkO_aK%N%0@#GQ*|m(F0sc&krOP)~X3}-pCY6OEhQS#kh-{j_uTf
zF;KE;v(jUs;_CWx!Y|h3?Gt`+#flh+J4*$>_88R~*PP~_U~;FqXJ%XmV0`ylk*@o`
zA0<&(Zag8cd!{sxG(FIBjAR7*<-(04S~RBo`XGk0%Ifo*31_wh^mh%hlOV&7Z{j2~
zr-6QcC+R`|aMc>3N1w~#khn<&C<FOD*2s6a&LtGI4qo`Sv(D`Qsf_6FabX_`7$FCw
z#T-rh617?5iYxE;S8sYi!m6PjPk$Eu;L~H9S{HJ5$y7t3vg;++wfCo<r|#~;TSC)^
zPql@76ALU+39W<flDsD9;{7nXqtKTx;0M57Q_G}~uU=Gjr?U?1pBbHrAwOcUPlHlp
z1)89g^(FE|3LFtO8IhyAps?7$gz&p`Mls=~ARm<wO28EPGH6wbFyoiGzqY@sZCzY!
zM}cEo7msY|aVs>ya_{OkW_^IFD-+{<K77rZXHb_MsAOoimL*l^diyKqC30*DJI+{O
zp=oGN3B-4@@Rp&+WkJ$dI+Bc=8AA<_NV1|uf}k*r5QKrlloM6O{O#lCje>@);Oeg0
z%@Fb>2FM9V(+;co!y&8H0;pcph`_*1l{zlo9qq#u=@gjLpn)@PHfAP1sgXm*Hu!lm
zsTY15E5V^69>*%aacpUBkMH8guZn!&r3*3%!J;myh9=we9bgvrAOU&yR2hWhHogxF
zc>@y7gcz*ASE<jBh**?&w*!>cm%?th%1-(<e)dEa>XH8v@U8T#Ghoe%-SRl1BzJ$J
zRO($JAPy~~SIouUqbs|<#fM?i>mgLJn2efkjz~0C94K)q%lo~I;lL&gJ3Gti=G8mA
zc7lDmeI)_p<;-X7`J2&xy+4jC67XVlaCs|Y@UE5_87u3KzPUL~Co@*cwh(ka>_#Vd
zO)b^6yvz$}bC(m<ZXkl~awu+REeIAcimx6vcQMa}h4+X%u0!y?YB5CPCW(7f?y(7G
z2RF*Sl>E^Z6NXHg?EJD`aQOzB$AmT{dI=dtR9?wl+0<b<NA&`X{iYM#F+jV29%{31
z6Lv}Pn@7it_53J9&OFQl*f+Z$PSkU~TQpb?5vzV2dFL5>;KLyZ_ow`P^UNX_<`2;4
z7^ew|;H7Fh5*tU)@Arkl$4I5!oN3`Ae-Xn|h-DY3N3%rMhg;f@4)iqRP562m2LC{L
znxS=oynjF}kpu)e7Q`}qgN#GjA`?UC48Nqqjt!L)tMgLjp8+jqltkAC_mNb<n<1&U
zVdI|GI}DLd`Sz}<Vb2k*wh*$5cW)tu9c$RknsP``VF)*Gm#Xzovoc#r#g#NAmwIXS
z?E_kAMERiFoGHuYqKx3P`%haf&a`ol9tKmfkEwJjb+OiIrDVi+xsIeTKX4EvHY7mu
zhn1m2mI}M44!+(R3IA-w>9<iPrqff2kJC&eeJs>-K)I{jVKdk^Vo|l4l^CH8BEt0~
zRq*%J5TZ}D!!_~tOR*siTi~Ir^^rv7h-wD+avFi0>F})VA~AR`Y+Lsjs6(sd_R~T=
z?qCw}K6f=zs;bI&^#zLZVpGG;{pc-v`fQxv<BYUMt^wh;qHP<Hk`nwpxR<`sD#N|Y
zt?uJ4d<wB37P)2wb>12CNTEC21?UH$EzrT4rCw>R*WCXb^S4|Y(!Uo+6Mq*OwZmCq
zOc4Y0=Pk4c#}&G70x^x0Lqo~UpA>Ab?imOU^?pHBzKLeMF>M(iRKrZ7cW{BPSx#1{
zYB{dx)N6m#iEI09IX!jy-wIw<)Fj;JKC6{x-qwrVYEy*WlN`9VY0*+ZUbi|r>dz#<
zUvoz@Ts-0k_Q82FVq6;OfH`g*BG+WB2bAh)_O4sPG51zO0D=udd;$l^cybam1OfRK
zGH)?%Jtmf|2ESfjgvfm)w}rI?8!V0YoRV>;Rg57tq(_7Mp2&DGf&?m(wz2{A)0%!E
zMQEK4<3c#>D!l$1Pt@bW#B$SqnN#|L4>2q;`s9<kDG^TZ9hr7PbSKHr_0w^k<JR0r
z(^`PoCyo#U{d#;O@l#mD==@LvFX6)__{a~jc9B(&VDa!MC6}#*L8Vfjkb_XC#}{p>
zFUWNzp?AXy-I<o@mg>$(+{VRaqEkCf{uRevn1^_wF@TEd4@>(APG*{Y0Xtubvs5pT
zWcjuhMomDLj~HmUjNDY>9gP$(&Uj<p6*hVcnrw0HJ`_Q&20x#fh#bn3$7ylG&yL9~
zw&UuJY_4UUUYwV73m?{qiXAMguQp9WXg)bJjV#(KGl!B@v-E)_^m~hjfY<`0S(nz)
zBN`mP72topx_z~f9Lbs_b0vqb68~C^i+v(Y5=|rhmjHSf=V4^zx7p_v`7lqCJ58`b
zIP2ya7#iSnxwAHmz`;?E2uN6id@QlmdK?RK7AeO|@dii^Jk=c3;<LHLcKl8wV3sf*
z@10D4$B(Z}R?7CuZdtH^L9`)R@2StJKjW%hgd-cB2igTwe)dz^&I$hDqO*ZdRPTU1
zaUCKSJCH}uq(J_&oJpE>&EyBIs7`RLkJbTeA;v;Q;nIWa!R(%omx0!U{ecH^yEo64
zP(txGE*vR>@2F2m39prawTuYMp4&}B7AIzQK~6w#<p7sj6pZkJ8AN_lo_Q3K|9kXY
zsD3jhnGg+@278%x?b+1C6=#PB!a&B*$}QzSwc=OrF4?tA;58{6plH;aDmC)O`&7;1
z5ub&sFt47~XY{q6WChZ}CHG5bAu`g2KAbyq>q~Yn8<1K&F-!PrE-KK|&M%w@6(o3&
z_i3G_{4fYfJe}#>?n_aViAfJ(RwgW$?a||UBn0fNd&)f+GDjP2$mxEauhzCW)K)<B
zpfcK%{F2tNde5ZM8DEuc;I8}C+8kEW4c(4GVQO3-nYcqHGV1~opIKpj2%LMy?|H;%
zEI_>1n4p9n%Ma9Xb?pD%K>2->j<BA81??qcMeCR`qjHB_4asBWt`SA;)2;w076H6P
zCgSOqY^cZ%0$lL(rOOARRUyvxZKv{oIEnzmO?Fnyqr4=98~l{0MlZmHq3@`%-^c^#
z0^G!Rry7H?)QsIJ38!--c!eVtj%cJH!7L<bOaiLk?%p+ZKaP$B(MAPDMF$QJ=uHCO
zKt}DU(}^(>$EB9`jni;lC#=r=hcUB=O~b@2;^-^UY@`kbrw1LKAe@7kV(XL;LzG*^
zD#i!&fBW0F#0yVkBkYy%9<>o(G=D0=-9NE@PVWmm*R%pHEvW138Fw!gyH!8Chtw1L
zw48NT_jvUj{vj8)s6tP3BRN<{s!2E%(7`NFv#m0UDqE^K8~VHX<m${K`>ZmB6xKA`
zQr?hkr4ol7opj(I6noR*>UKf|_iJ0@w=RMK0{i#x3}e~8Qtu6Zexk80m1}H^q=W<e
zP14ztu%M%zUtoXtDl=Jt*|sgA!rDx(Ukxx}0h*5fj{#E-CN*!u4&^v?@1r3RP284<
zoI5<Kg*xK|+zfqllRZoZ38=LeN#{cY0zGm&N9R237|7{Q(V;}s*S#4Z7Br*^S}lY$
zn|~^|PgZ~##st+X<pT8yyioTFZT?xS`SahaAWkve>?QAlEEG4?a43x&Lr(R&#wV*M
zoR0aP8Zyk=>6bY)WHSwylpn(uzltSswP27d-K=#xJH+lkWn-K0nbPiPY`M>?ul`K&
z<x>D;CEVer|6zFco^Xq=x%Fk4qmL!pP0~^R5zQ0rb4a3uA#jFW$1F0$1kvjFR0VIr
zcTn?y(HduKrqE&Sy{TMj3}TU28%fZIDRN^X{<=bIJ>mT@CC83qx7emdEWRomTf0pV
zD^;3{+~is^+TA<GmD+|>l;@e+Mz2Ci^$UOm^M)?Dh#?UVZ6-YG==!<4Z-;Rt!ofKW
zANxoP;K-o&(2TkhJIhk6&0Y7akeB$PjNJ!tL%d{rOK`vV<XA#{!pq{lYw2QqsEOI>
zeSqx*hkn{l!JhS8GDFz>O3nts;Mk_x63Uk)E~_()1G6TLdm0qMT=;*u%kupw-~S2x
zG+u&IR(4qqm8o_T3Dh%9TCiVoW=Wv=yVmo_G>RA1l&$vHsd$Ef-+7;l+@*>m%_|F8
z>nD1F&SvwPn333upP2`BEWUzeTV9rppm*5Q!GRb6i>UPi#<QsY(?PO*2rcgzZ3yj^
zFc=v@5AM`di(s$eqH)V*IHX%nQ<;fS1c0(1_kf7lA$X|`wnoz%Kfq5m_VtXm*dGi)
zau?q>*><~iff|vlWeg7s?}f^jx9k}N9QMe~S(@{E(A)IMWKhBnRdh3puCw=R#y=km
zT%E~`%rO(DQUmzz9i~<j|7eHl3QD$hH2-Exct=Z-Fq{lvcQjq+EtZE+rsGRlBa2a1
z;*b>lB1Y!(d;)jz9XS-BL4mK@S+08UjC>m;@K?7Zt~94{%7H^hT)XXirT{8>$_bCK
z50x2d+d_Yl4VX-OMCEj7>V;hy8*XfZ7&9my_6_ES@k~^3Tv)S-*4CG$%2>$WRtL86
zdWK6+ESQhf%Ud~bpulb?2M0M;tN>ElR;=YpG~(8>cWNcT8D5G~u(hP%IJ$Z^?=V#c
zUPYwV4l_7VQRtORRu!4gs?%r<{pLb_(e|xpWyLVuIVmj-ce5)f{{7SC)V49CGIK4<
zu?EPYe~%c2%^T%FWfp3)*LZ84mezEzz=QIC{y2Wb(BX3+$R1*VKx8vOO8Br`JwLdZ
zvsrzs=5}U~_P1`h)@#dv?g^!^&QQD*UEroKzP~x|PpO0sISJaXRWKjaZg?ZZr8>{r
zL~A7={lt3VBa=~yNWgIY#M(r^QQjGs$4!oX%fvDZug=xsg3HqQ8#8p{5LU2@erE=v
zQcXdhRYHSNwDU7#uWu7Fb^pE7O-Lf@Io7&W7EAOCxM?FvqLss0;;U0Ni|h^PJ)n0a
zmMGV`utR!;@1%=^m=%$Sag@c{AbV|EQ&+PgCgaCVwwtgW;67Ev3|*msxolX1hvy1y
zhbdSUCTnn*-|5>W*?i?fxQ}XQI*S^DZh|3QJ(-PE>zNP?E1MAsFY|65Zg+6{M_&ia
zqs7R4j(Z!jOy`^N_hX=~X2nwAmox$&WAT@ql<<pj`pfD-xk4y8c7u)<k-Jr<^X3}2
z_u`(dxkVF+cH(j17b*1lj_P=^EHZ2JDSen+h-q{$h`sO*6FrJ3z*n^F=^UR-|D$-S
zeteL32x2JHdIQBc3!#9coDvDORULhc%DaA1yHQ*U?e{6|3)O9MVjh4%*nj!fMvF5Y
zkCI3ovy_RDoA*VGMeYNQkRs+Um9+t@g{&sh-qLwg1=>@Mj;C<D+pv9TBa)QJH@bRY
zBDE|6!hO>Nt>`knjAj}4c^*Vw3(VaegSgekZ4nTh_~4(gS3+B(N6t}K7jDRepllYc
zl;GNwq~&BK4>7RN=2;+9hoxLAI1oOG)Bixj>T>?^MkR$J{%A(KchW}X)|7!lwDAaG
zh&-~E5U=k0w0f^EOjc&f-fM7!{32Nb;;1VlV;!mh=_7<|(^o*z(nVwQZp-MU_9@I(
z7&>Rw!SbsTEIcrV30QjNmkDilUI6G%cP$+>i1BG<YV>qXCG^;CxYPW-{cCcP{){+)
zzQv23f5RF-%FRVIS>jg&@H$znCh=fCRW1}U3l_d+(zU8V5Bs*&LJN<1j3*8e-qN3h
z^=UDvTtR(c8jvTefHv9XXokS0Qd=aVYZA9xWljgA^s^FDhzDtqn0=6q0M`Q8>nZGI
z34Tyl*RhI8^r14uA;Dk;_2bVF0puoXEgn>Nh60gLF(tLHzifUT&g!?Pp5kykoNF~$
z$&PK!qQ61$%--|&LMubXqK@S@gCjU$S{DR&sl)TVp&u+h7&d>E1H*gfv*c$rbh{%u
zC)->`Yh6)=i`ih7fV-R8fk-K}_YR~k%_zrR8GtBYDhTN4S0M*PiW*C%b&h0`l$Mpl
z6#WEpa3>=GReaVxz-x5jq8c4IbeXOWH2YG^TUh_1LPN`HQ9(>ui1D17{UruClTFc&
z3c*QYpdLck6n^*5R0B#h(ZbWn`%rV!Tp;X+l~CZa3UDiSebY~VWgI^FLm^EuBYY4n
zbkd#kb#Zi{!<rvxloK8)EDQUsjr(VAl@=XG=lG~gU47%4%X`~;Xj+<<k*Tre57!h%
zK~)cFrFnC!wuqEOFJl-shlb_UN?vZ*JB+^$R=~EroU<7JoLje=&E5}KDb?FSgI@%u
zhie;gR5^UTh^ydp;~Z|xPXDq++ePFnRljo7%D5McSg+UK&PF(YF=yyEHgiP%v}iu@
zhd?}cvOb&e3cF0TzebPtoQeGBmRPg(>U?5z5b)wDjjAKqbn^+u_okCbpzm)N5f8<E
zR4iI%mMS|~KyPNmhut(tgG?9+M@2@o+YN<iZ1?P*7|G*=ZZkY>&k_J#df>|%%IUMD
z>u-XLwaEPaG=3iy%2VF9Ci2X-S4KuTx)!9rOH(EzMKI~oD%pSWOPDdQW{%K^DFkI?
z06UG2gM8MdjTD>as{K>_N>xM-#!c4@%=|C^01S5uRl-G}e@28Bac}7N9aLN*?nkh3
zW+KM$y;AvT+HHK+T&Jqjt362JVjgs~lpg(T<8}QLD)(rflh_3+eWT6O`c+YRAPO_8
zDhS!#5P&FP?B(Qg6KJte!^4wz&vd9AnnRl7E!;B^Q`m9eg+5;jJh}v*B=V+^V46~B
z1?4Fx4617;*kP$c9m;w3qc=r_!ST@(#CWy(OG5~YMoADZ!@-M+Ej(1{;kiktn!_Re
zJ<PGTqrt{}<tG@xc)I}gXD!XKy8gT`!ymp4JZCKngLaV!a~V=!$)DldE({r5K7H-k
zG0s#qTWZ6YDv1S;17yV}@S>*-7_@UWWJN8@rsF)RjPeTq^Y?mujQrE5qYr0I*hcrC
zofOv>?p(Km;X`t!?^LobYOn`d3GU=_1&j-UVA@g&CHIL+vgba)&~TbR38mYKEJ*|I
zQ>~7}KTyzU=n4Q7w&Wk02GEW(3A+<8D6+GGaV{0sDJ1~p!iv^@K!SwW>@Kb(412N6
z>Vl2{pZ@mg9}3NoH8{MR^W`&B31qJ9^dU5Kkh`ulOS}y~F#W6a&dGx7$XJ#(<-Ak0
z=bbZ!oBfE-%jC0-xXd`l<FhoRt(7CBI75yVxk2TG3*~P5da&i$pv&gzhJzg>sm^%P
z=P{mC?o-@gb_<p>W{fMQpyfsr7=4{)fTndAX=1I6!;)qH>QQ}R%Zr774~>MOT|4DV
zrYqC>MCeAM`X+6W>)5Kf;p)<$E>8q)!Ba?@wYo`bdh$0SwU9T;FEXM;w4AiKL!UzZ
zNmmP`6}lS!F3HkkH;9AcOG1YK?hW$dJjTCpemo1(kO!;(v9aNZNJ)<Cx}}+(z#}Hw
zC7DW5dUKvu(T`quo?fjXlS1RVW*fbgu0h2@kCZ|_#8mC*C}n*F*m+3H<%cdKA0ed5
zlfvekQaiQD!~PRr{zO0V762o-J&>}<%QN|8rm5=4SSAlWCZ@VYUN%KuQ)Ay3;$b|X
z)Xa^m**1Pr<v0D#7TJO!T^32NC|mJOr8G!^M?KNxa0XSL_Sb5Y8<tm|{l7be-%5K8
zupioVt(`4ZcQDook#-dSXvgb2u2~6uJ$dK;){3MG3I#JHAR7-hVPf!P^+{m&z<AaN
zBz6uslv+@Ia#fn}MYEKOqbhZ(GPNz(eldzmpq4^D+fwf%Wg8T^jb3rq)4a%E?xTEj
zyatC@)S~>4+Bnfx-{q^$2QbYP9Tii3E{d9Zue_S{Mx3EW+jqRI%EC;)ncnIj_UY9R
zJsFo|L<LUUh*{Z_u6mm<{&+scQ$4hsO#PkSOgujyrV-tIjX1PtaZf}vBh^TkarWB<
z`{N}-r=SN~++ps2Fm~=KQj_#29X~i#x(VNiB0glJG$Xwe5?|({AQ@LfK1@XCA?#1@
z9}H5xgr2N_*QYz=WWYbQ6kgUU;(Ng{Y$;$3C#3!@WPpN8ZA&QnK^K!6{rGkxf4mzA
za6l9L#dHDeg{qjHCX&InLvwee1J`^d_2r+XEHjYULVC8h3pLHM--J(>JiNrZN!<p>
zFZqFgSeyqgu_|Io(3?Y)x0yXPiUi6|z%x2HO~cf#(CM3!-x0i<1+wZQLi7GXtG!nx
zzHZl@@tW`MBTWnki)lXgz2n;g=HVq9$`^Z9KgOVu^P`ZE)+=dUEqxMhcnzx?SolBA
zSQR79jmb=I*^6@Ayjf%x>ccPukhP}`&v-&(WrGoJ_}>{wi*b=1eQ7TU7qf$YiwfdN
zGXnq_L=Vm-U1b|6YY6*B5!Ebd*(UB%LEPQox7HF<7(*6ALR4!&RyiV0K*~4+H!Z_u
ziT$TZGj0wq9&vQKEBvp!Lkg|%Yq7-&$_$Fp$n!6u%Lyr+wq>iZXp=Pyh!&T}!^8CN
zCh5&TaDA`9enI&?3;WETf)duiJmy@MR59GDe21S*`*`2BO!CL_U<AbTwlXL1Dn)bb
z4+Ni41SA*zj+>W_Og>hJNOa}zDDy2R%<nFv{|MwV?Hl*RRQBM1<p5gt^8!jTJh#NA
z8)|#X#8s-DgvIB@ew9IHkjYdI1`-;q{e@3?P&P@d9eQ(PyIpev+`_A`OVXkbXFT%6
zKJ8y{P{6GaLVE$5!77PPn#t66Orq6~E`G}_XY+RdbPdlXqU?uJS#W1a<x+Aj%gIC{
zG_lQW3y2^t^r|bpKFGZjq9-pWM6grZ5)vOAd5@;(xe%QMM1`)<(x?s{`E9Jxjm@T~
z!n0*i*mg?qD1OG_gpIZO-Fvs7A~k@uE%1I2&SdK$<Lr&Lky?4J4U4&TF;#q?x2m5>
zoMP+A!KIcwDsBBV5JyXAjZuHANq;1#+>Pw3*!Bmto`m#47rZ5B^lqNg*g1lTEfI9v
zRrfvHN6@g|EwE)|J!Lfn4xY<lmirl`+wDwq@5*DGx2YauA<HRuG?Q*$#9##9LPR*>
z$4P-54OKl|e;OP|#Y@vT6hki0?D1uflSntp)B6mSMmDcLR1$%O>vbq1Ihsn`Q^6MR
zs_D>2+o(m4EB^c*d$F#%FYGmn2(v9=)F^Xd84FI=n$UNX1eFgCXkQ;9<H;zIQ5PKn
zY|7V{!_7wLR3RaqbIr6m%~*uM=0WrmH|8eq77WE@|1U<My&lZ;WT*n-LgI#g=hTUB
zCZ9~cEnA?_vYv~rgXsiU^uVf%fbU!J==cq{OMd>&?z>i^FO$OJj;th0=`LQ+ed+GU
zqm_3vgQ{X|j$`GunV-gu;m0GIAr4Gq)-9nVAs5!*JspS5%ExTS_DY|FD90h<P6vnd
z`e&C>0jItKSbCy&?_LPgUJ_LGtP8(iz8gPS{fq4C%i3b7DC!TUS-RC6rax4i!Q-9}
z!O5e_eJTLKl*@`KUVDsN=WoWw%@KRAxfKXKr%3!|M8)cjFa@INa_yIJTN)V@7%r^*
z!5SPR_dXjpqPBLNX0~a^x(VNekmqjp#i?P*aE>s5VIAm@cir$lG7R+j8Lc&l9*6C*
z{<;8uQP3THBB+rVm_`K}8e>1v$chU4{gsigV!}Sy&EfXDUzE7KQQ8H=V8G4qhEov>
zbzD0<9Y^(xjz*67#0C^<7~Erv%X~EkjiJ|3)F7)Yv?id;pQU()IGPb%<(P(SC{VzW
zlz?^!7V&^-4*OTv6;hmDd7kL;E};dy%m}~J_4p54U+><5IF+H+@m>(38w3tW)Nvtl
zStGRB?}OBywDPGko<gd_;U`G&-5o*a=sktNWZZ;QPBb-v3N@O}`6u%c!$`CXJx>gk
z89q-k3{~54DKE@epVc2CVDVVw!)A3&HW{hwALk~qudjOHW5@Hf(z5r(8B=DSr-ZtQ
zDMX{zkxV8;NQ<Cre!6CsuU8Sx^lWwe%kNB_yb1RrpYd)50yYLDrzGT2=QjNu>qQj`
z8ZJT?&qgQGk4%aNwzA32Tm6{(-l4;wL_!m1fl@r7IZ+n2<JpX_!@x}k&CZZ1-5yf#
zU_ghzxg*1M<k4a~1Zb~F2v8t;4xkf~wfzNk=`iLT=8rzuIJRXVdGQ;D9cHdTjJ<`j
z?k2sjBYE?-9=baHiRWzHOhmd<r>UL9%06&*5A-_QTo7aJHQqwMrJ<wc1jikzs4Pwq
zJ8|X^wB7U#H?kHk`SDHKaym&F_BPz-jDE7tt_sx31=Fh4?F!tImuh9u_f{{c&5}|T
z^-&tn?r*RK&=%VCCCCKUiKw_(<-2T^d65+$3*h#@K86L4T=<r_Pxt0l8g6}NH!_aw
z?Bit$d*j1iKo<~A;6|KkLL@5k*x4oVy^+jdSsp8@`;FikWrR02fgN|&l*h-tw|0<W
za#^tb2<tst_z?;4+Wz^zHz<^|Wlj?9@6_@0-BCxt)8&90V5hzR@$)~L*#8W$d(yVd
zzRLC>y~c6xe#iPabg|Ly`X2N1HG;t4_%J&A=Jna5=Wa$=LD}5Y;IQLAcV~IYdHOF9
z&It_c{xj=8mzO)WC}&5)|8c`Z39Z5K|Bk25!Tyi=eRM5=?*Cr;zJ?Qp9qRJGA7yZ?
zB>DfB{@1qt|84qji2c7oVoNN<BL4UFcPzI*kIuk(LgTXkf?in6{{c7nvO)BWXO#W@
zh4@_XPVqT&nmzrBqTGGA_gqiF`!8r7$CXn4`#tLWwB>^p+dSol_CDDZe%&^B18sGy
zFNhRy{q(Q>pD-CEdDweOH$@^q4s!-Oj~1BnQ#jefUk#8VQbL9)O(k9k5_CTSLGTKy
z0^&Ztgq4Qv5`PF{xow~J<;7589QMK7|3DrSu06yOv>*;*hY*#x6X}aR&hSv%AriEp
zrcFwu#vV9x5_kCD!vEX-{M!}$=N$g?-qkB(^cHp&@=U<H1k(ea7w+A)nr0;Wy{TmP
zb(}Wv?@?dGBoI22>UGNr@*I9@zCz91z!+NKBza)<A?DKy2TGVh$UgbY29Kk+ou2UV
zVr@jU+(uI^&;7^jbgg9c_i-53_0EF;`07DGVoPxn704}g1;Z6hlo&FGxss{(yYS!W
z=h$+rk0YP2+y8OQn0$tl#*_|j!qjB|MEk2rhaH6p5#DSi-m~?D#{h6k>A9S$GfUsi
z7Bp)<U);`!+S;%Sm;WJSc#8YwQQD_Z5H`S*%+l00-j6+7Sh)A`73-gFPCX`d++F$A
zU!r!`?STqVa^&tYh@`MRc+sgd^S6_h21A-d&#J%%Xyl;J9$f7f3vUOM4WP!_NUAdB
zf=eQPRX_YDfl1JQEVsT<D%OXjfkg#4_9&GdKd_yHvl+}7U^t!933M5%P0bH9?c3*}
z!ZJyi7|jbidBp<mhG~iXn{bBT=Vq#P^h$LU$orzTfDMrAof|VDEqtg~a&%P{INW!7
zo)CC+Dtk-&3$C(%ugwXi^^Pjjc1uNpk~sEnpYLpc4;TJhH@Vv|fzB{&my}r6Bqh!J
zaf(l2_jtfH9ONl|Moiy9@4=AJLv$n_S`x287r^&EB?lyzz>_#8Tf3dw?lZysXnI3w
z__kIIaKiwU&pD!>RyCDExf;09KQc*W7i3TmuMD)W?b!7Wp3IZ@xa6CkB?04H!^Wbc
zdCZF01~-6i?kyuv1%a5chMV19Xof%<B573!NvW~oj;i$a&$Es2CkBc8298J0qmyg^
zl$u}G1I5$9EOTAqCo1?NK;-S5xIn}&yzt*lyW7zKshK;#am`a|mz=p`aY0)0h+K76
z6<DVVcK44rY<Gd%*P{Gl#FgdM@#dV~h{&sJ!!nq#OL#KTZUfY6R8>|Z4;y#q_ne7v
z@{-XH{^Jdjt>z$J#5hxTt-;K#0g3KgEL$3P6hE2=Jn_g?5*$5V2zPiKiJPqy>y}(S
zQQO#KQuUqh@)BRlLbis|U>Y_`*d84r;l!(gTuC8Y+K19jeqr08Ot7nF=<y656tLff
z(gr9TQBTwW-5PiWV>YN<%e=GNF9;RQPq8(-YbeJv`7jNWC1RXvO)$0B)E}~ueU&!s
z1vP&wLu1`@5P^&#`C>$()&2G*pb2mQOgB$j4E)i{pUv*fHGcuO;2Ut^e~yk!98b)j
z2gB3&MP-P(phST1ufT+X>2pdvs|$eVL!k!laVtQUy#J6!`zATms&0>iN>_{(76V|+
ztNc*IVd<Zy7u1`8+#|szXf%b6iOr8^9lU>!2K*U##{T;IFXe7`|2^w*HXr4|*ArE3
z$u4tS`fShf$U0Rkf^rNfq9~1VtDq5mlno&xI@x<SQqspi=t-1fLo0nZUU^%cQOLEb
zDCe!gb8iG09&*T3qk>{$p;u(c94eBPY+0u1hkXYXZ+zk`1sD(<z+&7Q#fR3PYZd!V
z{CPz-IYYIZ+_H<U9Q`u-%av}nI=L^w%cdT_Bv$&Nf+yYZHc>?Xww4eAP~tn%Ol&k9
zh6@8aeLhl5iDv2kab#aWAdQ9rvYtZ1z?Wr;nhBb@riOpL^=Pq2ZYFG1!Y6fBJAH6>
zFca=8VVk9k#{TT&+MFD69Z_-VJY{W*Xa_(f3Igz*OxLl4^$e$4*C2k~awQZ7_ss7!
zEJ!p**@)d8&P)IS>NJl>_G70BU(PS;jc;II8`J=&E=0oif-^usq$@6W(g*dsg~6{_
zifcbGe8c>yC9I6gM)2F0>tYUEG)y$WUADy@2Ba<jJrz$BLKNzhYavb9F4*gZhV}6c
zI9n(>kP`Z}7(P`u9(C0!l#IM`0I=ABMSRdiATIu|)%Ibq5Nbbw&(djt$Tj$gTKM^e
zxvq}mddW4ATyC`7drsJ{o%%$}IxOYMOzyfGbD!*->m8~LIZ_~L-DP8QO$wwk(}lX2
zK?VvAeuM*j!sa$(ttC^0^9(bmsao<3$3*KWw?F8KotA6`TI*Tig6FmMil*?NUYgW<
zRWoOgDr3wrCubS5u)(=w#T|@@K+zqdBFr6=5LM$`ss6V~wsDakWVkzS;l*Wj<rCV@
za&ZoAVB4449Lt7fDvzHF2GWBWaRC3}lxmh>6M?9m<8K0Z0uixqWDAK0p3+fn=PasG
z(JdFF8Xg#PJUdhKLygCYEn2mtH<++q^vLAlv)!?Q)8Dxe^e`uyo3=^7iFJ(7&1fZC
zKlPcf(TTT}fBmn-&7{ZNGCXL;zjQc6yYG)TeHsDo{JMS=HvOKo{h6}=H>(e}!n2oG
zeUprd^@l7^0i%Hee;7OV)ko#jJR~nZ`pf*0i+YZm6GqAq%OPRYza+2o+#SDZBInPV
zz`Lpr@{IOP9sn{$L4h1GR*oe+jN9sM6{GgSw-0uLe{qe5xpNItq!Be<bn7^<I)Vc-
z&36N^lKh!Nx^m&YwLLeBNlpf4hRYM<zW49-)jl-C7DaJDJ4eYhEhvmMsy8m6v2bR^
z0%XMddH{b&`wtls7_q6(i=9m$LiAJSRHr8Kv-07*_9}jucaU6VGeof+&`owIcaoHR
z%(pPi4JD8*M0;9jIfC>wa(kIcrWr%tIodExIB#`s|EL-^jS!AM4dO7RPUc%M%WPA5
z=f-#S08uY7>vb)toB9mN>YBE;D5XfXfVjv%Y{a749P&xakpan1QjZcZy53(*87=CE
zD9#mC2?u7lqLY&yq^0Qqs<+1Z1r|Y`p|(jMM=y}-&Cy^jk!P}LvOXp5Q5&Tl8*wne
zX=~Tjt)$22wMqniB=|x}g#m<2AJB_EU&Ga`u=d<6wg2kkQ>P>*rPAR@3NbosSC#u>
ziA5;-LNxCx5P&h4nT`Gk64?Z8_rosnaMT7yK{rFz1U>7Hj~?0hq1}IPce+;~zKRTu
zJrwSjbp1e?XInx-u&D#gnFeKXAPB!c_%jn9;d%_16E5R1!9&X{FA~2yCM!9~(s^@6
zCTkKBQgPBrJLYz7bYk~<Y;8wKv^Sp1vN;JpNk<%Q`2K+yb?R{-ddXG#OK$bSm8Sc!
z&FGsV7`duScPP6;CC%n&vT&H=*;YJ?e3-?7?qDY!Vh_EM9Io7eIW{2U(_`j5|Kl<=
ztu--A4Ui{v=NU$|^_sY%r(yWP>Vi1S0Aq&F88#o|<0Xb`z94!#-(iVFaG>zKH`NLk
z3o{f?1Sg0%&p;1}xgObY-P<dCMfSP+8TRh)#eE*Vn4<M7XeLaPHxBQYc4P!K)Y^sZ
zA~iF2cM)Tci-U9K=0uM2*YXK7Mb8bxj&Oj^xpyShCIfG>6lTre(AwsJWDomp=J9>&
zZ>{o|&1RSbnumgI$7uO+RpZ{fUL0tXS&foQlD$?HCYJ5u{z1;$?yq&}-+0BbQb?^e
zj^kKdpgQaI$_cHn){p?+>CbWmnYmwMtaH3SV^?t~j%&lR!rlOu0d=;TITO{9J;Gw=
zM#;S|L)CyT(c1=Zp`sEs7{MwAJja2SOM?^>8#d*ogSi`-utL(W#<@kA-(VTeWsoR`
z2gONwDp<ESx<hoGP})$7JedS1t%<_QNUjqWyu#@?1;)a7H>3j$Rl6~2WUU=NTa1e^
z%an_u)%y;r2Qoz|%90$Rt)is`s#CcDOjWO0TBIQ&t(<RxB)p6u2}Db;!0?hit#_rN
zZO6am4cIrM@DE>p#9;e@8(==q%OeO=!OmklU><H-n0|E4el^P=g$GEF+}WyK@6R7~
z_*`*ZZhfvzR%q^}3+eoAb1;)_$K{NLR3>%QJyfZ@Z3VOmT@cv75X>SLU?p$+*jbzh
zZP;E$WsQk5a(O|sGW^V`#2WULSMT=sw1GYWpN=R6rVfS{Tiv6>G{gH>Bx12^$ZH|0
zTS<p7>WvoXo(j+sOrtGUZ{HPM72;YEN?e4gAjRLhk8Y>rMoK)5`GCl^SNWvnb_Ctp
zyo>QgsID2Z@c3T<9zo&0o4i?kT28S=#BzR}`Zje92f(jLW5jgiD$YduhtgC!yUITo
z|By|8s?$ph)`Njv0&HqV#GYr47iF#4;%k(kuLZ_R09GqRw8Gl+vTQA|m)=_gRlt(c
z!#3b-{g!DO_hu^0L_-z{<io-#5LpCG&^<#vdtNN441gTGU%|s+Ta1pZQ79ZZRxX{a
z*Cd=bFd`G~J$Qhh{O~>8d+jwi?-AL7NKKKL@<b|>)8h>wCRGK3i}Fol<Ys+F0F}yJ
z+zht|0Aj?`$4~J5$s@e}#v3@hb63BI9m^>gDRUe)q;MgrY)PH;)igrq37*^#IAaW2
z2IXvA?=$Fs8Z-c<Ro^Y2+i$6;P~zfp!fmAuxZ*p~0HozS2K)R$n<m*${VU_n9qHZ#
z?aQBeX}ZbPXDa=de?3dP@a{hG4=;yoHYqh+F_!BM^+6@k$%Y&N5AZg>qcXL6+f}wg
zz0G^s54#UYv%ym%n?ur`hARf_>YB~9X$|&56~0_0O2}M}{E6v)*w<z{UR}G%@=jF!
zP<;NXV_;u;l{^jtcHWa3BhCheGuUDoA5_>q8*#^WbLo<yVyoOl{MP%q`?{6#^WC7%
z09?MO#NCpCup0=bYchb@=1308mY*OOLCede$A!gU)PI;6&XgrMpbtPKg^%nIBLK=W
zQAs|j;H*RdZheGXckkfSk3PVg@4O2J!zl;Lk=Gqn#Ea)o@bvLhAV%CiJ;l8T_s|_3
z0gezEAeQ6I5(E(e#A0edHjS#Jz<`+`kn2(kO@;6QaOm;$@nc+^zrdSc`3jCt&oqr>
zl%)CwtnXZrn!ryFfT#d?%!tGQ1$Z#Xd3fiF9aQ7HVrKfhNlp?l=#L|^<K?;db(uMH
zvXMYjeGJMeylwJ_Wuo!J#?&+{HmCIShQlc{gmmLkmSJ3nVV}CecK)Jg?w3q|gr|QR
zD@>ydH<9h3=}35m1_u@(Z^)(+UIP?#W19=d0C?ljf3imZoZycc?pBZPUIe;)%5gZ)
zs0ufVwqtu(hfnuA@}78IgSzs0`(`@eL720go)3)sMeU7#R-5jdq{BIOYM6MyX>%f5
zL*$Z7ow3K7T)7r_qTO<JWjTiL2;`8sDw*SA90~htfJB1GP)*cVuVm+8UMRRI`wXaS
z0=)@PH`O*dK$%;fpfX_0hGk_z6s{R_L@*YMa-NKn1t7`Z4cRc6lXw61H}K&PzK_>l
zyFzzzs^H#($l=qEKgK`#@BTad+duw8JbCyKh!OYi+`(`B=5OLZ_?v$NZ@%+3xRc{x
z6!{r=1jocT14R(L3h%}QI7<6f*7mXA;_2hZxW2l?8*jdW)zJwe#L}F}(Ku4A))6<y
zkReG}u9PbVlPIANDyUdA`rHohrY40skTScb5zW7Eq)~EFQ#~3LQMlJE0XLqdIeXVp
zG=}n$DX(%Jo7n_Oe;v*dUo@pw7$L2n9sM+ma#P&PgU5?bRk9HyBxBt(QE2d?$`@rR
z9DoksO?t-6=s!NfvwwPm>mNGAt!UG_6>$1X5qJK^HQe1TvM|REK2Ob>YK{xz%9Y{j
z@^d?@S#YTLgnD2hZP#cRY^rEA!?Xnm%bZ;iXE4gGdNhvnh0wm3`wOJ0&vw&vpz?$p
zd2+~qP!lNm*a~O^w867I3=mHPK4sD*;Kneo0t0+$-EZ(hI+Pvdw77)85V$dD;QDFq
zck<buz%IqLq9b9uF}SO)%W)rL5a0#{rp4|u7TScA%F0QJ0wB26F&@137C!s*Q+)mB
zz6I8A|K9h%hkx{6{a1MS>8I#JkL%49A{fsv&T;uWmsoYc-};aLBiwoIP0(spv>^t^
zqpTbzVg{>rT3*x{K-P9-0NnSuzPP}u^SJZoTX5@D0`gX`d{C{E|7b!j+djv$djchc
zoqgJoOQ9l$^8*zh;wI*dQNL(~xq*CI`Auo50&U<{Q<H%(@xBac9J+leZ)%45to&4r
zFNOvXw~cp4;Y<3W0N5Ty`Ki&(t)m6%NRMQ!4(dCLNGgukD2&(nLgt*!_fGKmzdptG
zA;3|`I1|_b7r*Bao}S?Df9%n{eFdU@Yq|scHN)`4bsemFHyDI-X%nzV`I}8^Ym@N?
zihwqCrY~0{uxE`1>)mAdiOxQF%)_!xg`3?V%)I-;>A-<>5U_)SCvpX(Cs{z#k)@V`
z>_~RT`-6cqVS%xxUxFJ+Ln2uo`PessyQ0O(3H+i0z>GO`GyqqvvPsn_`--LMW7kPg
zSO7U71Ia;?PI30i_b8Y~MV$b{srgz)m&nO6MFb8Ah;Ss67TNYG^>U{^0qqFW?2oAl
z5a)64jW_VYk6xhfd#sO+@$k_@{J;OvKf)&;`~==Pgl&%y89}Riae0o9-uoV2eE415
zIzGbct_QpIu)Zu4DO0YKe)80RY=V{DbHsMD!RG1;-aD+<M<7oHh>OTQaE_-|zEZ92
zB$2Hf14wt2p66j&vr|*!tI}}nG~PGDe^E4&W5gKx?v>77&l-z$`L^IBW&j=Yd?S-b
zD)UWp>|O;Nz04B7cV5M!ITv@X{ChEb4k**U=^a9cm1VlN)Vdx$<cvb#z=#Ns)_DA1
zpJ4lmo-Rn&vnc}cl5qWf#?ybi#)E&*gHE?OtyT?MT(^C$N5HI=)-IX@ij?x%)sO7D
zC0|=6H`O<}NUR1~Q+>$VyH&38w+3f4@vcQL3(0S47T!4zyu`wD&`juD=CF|hB>Qz@
ztvGSv5=g_C!$t|n@tC`&tEvH(61?U6F<=}3BsUD7eb=^uRe7BNT&o_a@+tBy{o_^{
zD8OqSvrZqqCJ?D@BK*P<)D`6{z-`8}vK)sP5M$8Zu4VKDkQ}FuF@m%&xeng1;a4lT
zqa!?c^G$sC(FY)o_!s~DpW%ZazK7UvvF*3mhAm?3)oe<{)pd`DkDlVmCm#Wu7eK!q
z>?Vy%SMZ1OpLG)C{^W>oeR+Y0pMDBQz}c<a=vD$09nsKPbxN0uM$!J3IC>VIb33MI
z!($y0JxnR4_NU^zzAV+7f~qA1b~pO?P5nvzhTc5t^2h}Iic=WvaDg__2G0JpneEzp
z2k)KA`wr{!rK7q-{FLcsK>bbRIscB*oM;Szy;Do`K*^k7h|Dnq)B%og`8!AGKNYh+
z6d?{z1AwSWYyiF_Y(ECV=N&SXm};@9Z4wthRhSm2Yu-;}S?0x4WwZBDPfhaUK@=W@
zE|-|}2?n2`@lDJ_1sA`ifrCToWdYg03-t0@u?l=L&}$r@0)XQJ&>4rW1atK<EUG_7
zc3q|&sF=+60cz>qfK&|x%d%S<ni&}frvVD<Je#0S|I3p|4pS~l)hi$YIfEN@(uJ$D
zTtVSDsexio9vlg=kD#?n?%SNYVR8&4C)x}ob1*K~C58Y7aCCBtho66jho60pKmOOh
zkB%b(Q8w+#D(+4Wk%9BeYdn4O1dbyp1{uJFgGRAv=`vnCeS#;C9^(GJ*KqsJU6Aio
z*#jUEu7piVt^QD&bxCr2Nj($VAN4S>YSKEbo1l^twJXd`=DUmPer!kPT$SNv0Mst<
zC0tyC*+X}ItcpIZ<F3*a%1t7leN(!rTvIwX_^DsDYz_GFv%r=q|2%3MeW+2ef!(^g
z?~tWhKlH$rNB<`)Py|@PoFNf1!<?pLaZU)2Ji4zzPO`X_Zdr@Q{Rq^}B(WDvnF4|{
z<txp%@I>FK9u87zb&SR{!-(}#Fk0WrJc>Q~UNeuC|7AMsAmhKdbl_2J34}5?!G4A8
zn1oZf#Pyot98BK30&D>TZd@+b5Kz|7Y7Wa(URFn4V}wgaKt9@?29C{zxB#oR%Va7#
zU9or}fi%stz#YSM6GE)=WXnP{U4tki5EH;2&adEI2j>_u5}ad%sK5ud;K~S+liWDy
z;{?Xt`>*5s-+vEJo<2o{XuL8r90Ng@WUzq8?RJaLKK~ro7Z>PmZvhS%*pkB>hSIWx
zfTxci;qv?iUVHs@oSxkl?-(#pr{6P`FoU}xIB~@cVG{~#8``It0#ZEOJii3x?_1oL
zztr$FGAhRxLsNMgsH;6wB#_rQvOOffwo(fv4+FVaoHmKMKkD}`G<a<<c2iWAZdbBD
zYtVBObkpnDzR>2C(j?okjXA>kNmeS$sT6<qj<UB?zUaVD2owP70ak9E?Hx$xadZH$
zfUEC$9Q`T*ehN^{)r7~-crM%&?fafF?%ix|Ty~@}oho5G4hE~$6!80L0jLA>`e_!e
zq?_6(9>}r7K-s}l`A%-+NHQOpjyZBQAQ>M=)%NNV0NN8=mB+#|5bek@AE&1H#ga}W
zw)f=N695{DAh0y2L7n9*paTUU&VbGtXTsvKE^Gzb8-op}J5K|PgRi5rjZ9=A^ay=`
zrw;3Gg|6%19Vz@_aERa-5JCVecaWLY7M<ECOaO?%@Hjd-0|VGxU4a=0Dg##<*dQ=g
z)S>epans}3qsLg?e+&Mk3^N1?%*LH$M2>I(V&CKW)2G;8UE%fD-#~Y~R#Y!$NAwb`
zmXoZ%rSD&qC+`vo%n{-}SK8A!JvN{Wv@o&QuZtn~^O;dsgSGli>1Vv`Z>X++THLvP
z@**PC`UvS*x-9$XXnnuDGbc3}HxX_zvU?(seKX@kL)RhgwH(1w`_2veiY@7tfZG9)
zZ)#HiqUfhpwpKsdmG|I%;?CLpTJ}yxI*z64IUoR!9Vh}WD5%>)j$%zgUNhsB{DntA
zGLuq*$U#L~wQ7L6U33uIsMD|Os~iN`*~>Uq?iitHMxz7fzG4Y%N7bxHFZRsyP-07K
zNBdE6G1w3M1*4ZTcOrJ|bRjjaTEQ`^IcP0Fx6^Oy3}>yyX&b+T04O)lZOiud47i~P
zyF0Jizol`JEWMncm8r-B!(KmnJ}3)d#|bn_doZLRTn;IN>l`{-q3b*x$>PT`BF2Ca
z`ot+B1;e6M?B)9~00lzCfO8$*{rcB%a(0S~7cbyAAQ>!C_2kYuth@k%%{Jos`32ta
zUD77aN1z%nDMzp_;EcZC;Kh?C2qECX>kr^pM?jL1H1j0Z7A}3qR=W%Ov&l=2l_|;0
z1w{7Z;J!J5FkIn&Zv0lL7E~2*I1ZnHV;%O-A+^diuo+i81I@cc8<S~aSNC?U-#L9_
zm!&AJpNQdN1>u@+_!&>aqwH&j%hV#*h;m4;n2U3h+{8nH=RW)@!j5iZAgY;&fs<B<
z2{4loOU1D(!qN%D9YHvD`o|G!{}UOMIDi9V^_3o275-+qOmkNYOiUEu3C>-8KZ8~V
zxQ>g4ZU9Q=rmWoHd)Iii>I_soGabDqpc_Y>$Z(i7Rrk$6bN74D&QGT_heuy^thKQ1
zp&RGJ%$=x$K+9cNiz_$fLBigZ&r&5CK&p;ghO_G?SFjO2y@z!$0U@^3OBh*5GPuJa
zueov=zQ_}X#W$XjefZ%5X<%;G!e4R(^Bdh}>6ec&>X9#lyu+$n2^W>49GfzMToytX
z&Q%oKcP#p`L?mnzCV-s+JmA*dd-&Dg`c3?g|IL33uoK=_1VWU-eB?b=#6XM)0Pftq
z2iL8#Ej>WEJ<Q5E8`{(hn`=CI^cmLO3iltp2G^-pYVplV{*bKsUcuJ{0<)n(4z2(q
zF^X3M^Qk+=G6FK3L*l4buSb1jl4#s_{-RM`_R`O<OzKm!@^DI$P_;`jc_Vq~3d5Ac
zsl7$>LLsIXGYdtI=Wn2GXI7oPC+`~on1_O>zbG_Xdv{=x)+#4ciJ4_8*=Sq@E5I9o
zF?Ws|08cYG1J>UP*u3Y#K}AzcX`(Dw1iHJx>0iA7QLJ9J3O`lLo8FPs1+ChGtDdau
z(?}1cY|@F&n7|wjn{X86FG^>ASuptLj1C8O3g`wObOL<5%3qz<=y(>>fKL8gGF4&=
zPG+WM?-=H?6^Tnw0ZT69)<#e{ap4Mtmfn@|43;F>xe^dGK;8J7Qd%el{Zppd<O~DY
zZnxO<qJr<881H37u!9^i@(R&A#GpG-xLgheaSi}GCle?uxN#l+y}$O?@V)=#Pw?^k
z?*pL+2srIL91%JPIA$;rUVHOR9N)PIc&}W)m=<iAoE6vs05djMmw5W<5sr_JaOdt_
zQ0KuUUQS>OeyVf?v&i`wv#};^6_{^GW;@E3{3J<MkeQJk>Gz8<VA}`uF=EwV+oexF
zk@;T~?%b;qDNr{WffB3tfV<rWxm0v)1O0N!MXDrWP|r-^D6r$3nP6@C+x4ZI>)oCO
z*cTw~SW^!Qe^Ws5#sDXeFKPU09Nacj2|SblO#>vhO9Rv;4S7p!oG=k^^mjM7{)1!0
z4=5>5P{FaYlh@N<-@?CPuE-`T*t*|`HAIadX~V<A0L{j$fb49`a&eL|25EyLRKP7+
ziyEwHn7C=A{SpKBrK7gm^d<BD^GpYsJ1LJD4JA#1vMH|Cs{p;4>oy7M%nS?E90m!1
zb7UqH<lb-`7btZM9*2MRbBV^o2_p&h7VSXwG4J>~*G;%+WgWBJV$N4fTs<RjjM(-)
zK7aHG*PE?sCI_+c=b+lm4&;350=Z4~9ZOnbYJvqMdk1n3uYcuj{GI>oKf~Pz_p$0a
zti8wD6TFM?PN3S!*$LkI#@BKB`WpydB(*p##=;$xNjL&A;PU)Ao<4en)6+BDxpyDv
zIz*^U9XT)@7#_h|ksYY8F=xXqb7TtunDR5#HraklSOtC6_aFvB2nbQ-@miyqP~_95
z)@J3)LZfkK&-I-v{-}yKC3hR6$j^lk)qpl9lL)mm=V&ALQ>pbUP-8(EEw_#Jx8}Jt
zcrf#6HqzO<yfbAU{PFk?^)nRD8p&7iY!qhjj0jw8(z5DmOyhfC#$*5hEqwVqEI7Rt
zarR$cVf}gp1%?X%Z2?>Zya9MC^WyO@^=S^-`K1Pp6Z5mVuiEU?T>TaeYtOj+-396d
zAQQ(nkD{O<L!$Vz^%mJY5!GRezhR=SI<~19|B!TU7}On-+^%sum)~r?8+5RY-9CR)
zfU7)yolAQzzn$eM6*piDFBKiI#TgvhxXZs*4o>0ndmi-Yv}>`l7kL&5O>%k0_psRQ
z_Oe{k?ny+@EziF1!Ep_M5#t8jbpnDS!Q_(i*e|<VR?x6I4ZaBg=RCgqOTUW$&42!1
z;3q%%G5+`u{tccyeuNOV=(-NCz4ki(`v3ah$9MnIUj~j&5KRe1a#I0aLWmJD1YBO6
z<Ko40+`fAkr>AFN`+j8g;o<jCj-Pkgh278#Eb_-lrZCEVsC+*n5`~h4(1G&Ig~`GD
z4&=PHZ}kl(z}goH;)rhE>HSSms&lcRq~)h7Z;Yh|FsJdAgVA6-+O!s<aD3TZ&Vf*x
zvFQ94*X~o4HUwWp!h0q<6?86%`rQ{%k9jDu-}CZ4mp2JW0U(uc8&#+PK&dNDLCRq%
z2tfCn8{GZBUE%5<ALIIWj<EUAAzlHl1KjJ3TYq<h^{-wH%hF7@@qV(xRF%CQWTxl^
zs-8=yE{_7cD*d`Y<n^Wyrl9frT|{-s@^mm6ey``Pwz(ntY0oaNlHV%>yE#{2ySQ(M
z#GhBT5b$M-LE0^h@-(Fl14QL-4eD(1A6mK39k&EKc>zi;3`tzDWlhN49J{id_6Ow_
zwRQ?b6O%WIPs96n@9Cygj$XxNx@r+dfCFbiGeY95$Zi;bsyw&^YP51-bgLD9?pxo+
zJKy*Q{^sBLyLj~Z=jgAmaB_Nv`>(x*)3Xye=MahDstJ=J_VHrh<NW+NwpZ7<f9D<|
z6F&d!A<oawadmNl(D&&3t$<VogPrfNTCH$&bcEB>6WqFU8+Y#B#nI`Ra2kvD8Rp4I
z95yx}i~?*c?Bt=pS>al?HzJ4Jzbegfd3VuI69|@N8!8s@AWZJ-rX<0Gcw+#|$yal>
zTZciM8=uJ!puKXY!O!mahFkXd+@noL^?ctvchzH%4(|@njhj)NiNa5nFZU+<=CJ}-
zlmiU_pk6oKXH_!>>&*9dz{&sdIZpnIOY}cpAv|Ze+d%hr1l`uTqFJE|SaJ1RUr!aK
zVi$}aO#;cH9J3%Vw<yhNl~>?xv@s}(*7}^Sk{v<^>~rQb-Xot@GE?vNP2*+gRod)u
zb0;&PPQn}p?DpaM<^1b+2Iz`3VW7??L>`pg!i~XQ9004jzGc%U4v%Gq4DK}L!F_s5
zIb2R>c%92^Xs+<0If&mF@OH_-_4NiIzys*ufgo~jBBZ3+4j~|h0I>T1g|$DAa<Igi
zqRbOSIK6up*N-0J?A94>-@awjgG?nPVFW<Q_e#@lWNa?4@#7!<2;cwSpW)HN&++)-
zBRqQi2v46qL*L60oOg2c`{rA3;^Oig&!0b+@)HD%f^hrRE!@6y2S4|7KaXGg^<T#?
z{mL)n`0Nzq)Jd8N_Ny&B29gFXfkd;?ksMQ7-PlPp1@YR!YzA`MQ_D2ynp;+XdbBSP
zd{qEA2Djt%m}{GJV}Mrvwc^`%WrzcKNrez@LV3dS*|8|R)FNYPADGTdEozN4CW`ZN
z>Fm)}rWU4pRE*k#Gf*hs#o|}^!9D6tK~AOI4HsaJ>JMAra{;&_5D5eioH16vc#YM_
z<_VBYhO|Id>v(ho1}ntN)lEW8I&j$BY)++70J{49fXX~nd4d>ha^)7R#FLGVGdI1Y
z)vB)Y&vr^|($8*Nc=L3yuUH^+u=eG&wSe8+aqT!fCjt5K-&EdhfR;gB3G$>~`B&nP
zeNPs<8^Vwq$C8n86>3mw;nH;z)LHt3X?u0UbcQeVFqXu|-1*fd)~l7+s*Bf|bHWkg
zhc$qi3;8_~Ez>%E-J#_<FiKrl9d6&bjjM}u+`fI=ym-PqtL6=}nX_PQFR$?jzx%uR
zpZ@v(7e9Xg#~=WA@889@zx6HrrQiN-tWS<`cI!6QN5?oiIzqQz<JRq4*lv5ou!R|4
z-d=BTaV}iF^Ye4Oc=ilGe(wi3>Ktz0y^FWM`E|IilTsuTE9174-KVs-w4EFkhZxX@
zJ}G!iEw8Qb{3=v)s&1O|H>BZIHl3H$pGUxHZYmu1U8|B)ZKWY-%ZsboqMHJlk+ZUm
zl)!-97eFnHUftdl;M$eWO>YtAI1_sXJ|O56yqJX=RfR|KSOX<lT~r8z2l1yXQGu;8
ze2QX)C<9dU!C)CF!68>^73G>l+Z7Ph4d(!>kq45M$+~qK`3<Pu78K+tmK)U?+!R1>
z&_s2USJ}0I{J0Ri0_>({nJ32Y&x*ciTzeSD4oEUdEmvZ$@vH^5lKj-pKO~Oa@F;c{
zsGH^tmJ801OE!*^Y`m2dMpC2A|5BAtv+YJeGUuB%mjw__Th|&44hBghEv2KQHBQgY
zaCCeObjR3i147sen2Rx)H?hpfiM*MkWWqQDLCQy*AqEgcQ<WTWa(apvj~}DoZs6AI
z!JI-cc`A{CAqM{WfB#SL-~Z!(jBozj&*T5_5C2QN`;D*R?t^<cJ~~Ff-Qw)dZ7?}i
zT#&q-_j)WEbll6td2xP;^=ggPdIgFB{pA&Y^8Wky-QWA4@i*UifU~=|hcge4cbCir
zxCocLi--WWW5#;50@T)D3t{gDU55?dRv-D|srD@~Mdk5mi%Zg?8bnw3#*7%rK+dNq
zU0nek*NfT>(jLj(G{7^W)=O#^sg?J>Eir>#UpkMy09%n}4e<xhrT`uzgSu=bbGopz
zr7H@S_wMpfnpbylyavxoU^dgLQ;kmqq&cSmoP%TB9(goWKyCmoC&mYt_mn^~$=-Ip
ztcSneNzLj{gn5;`cEQ}aem(_QCuydj`^|PQPlq4HDgYwAcM9?n)JVXqJ~G|?=Ugjn
zPA8{7Eps<4Zwu&+>r{ii8W@)80r`UvXlI<(eag2y@+^SrBr&Hv(|s##23Fo<<vq52
zk8VwHs}&+ggwP`@sN=|Ru7mSl6cL;XvM1Rn`A0yW);JQt0Yn7vI)DhrXJ>f!{5f8G
z?R7wn$C7gbnsh=6Az*uXjn6;$0Dt@6|Lgeq@B9L8z5Y6m&(5&wRsci`wcl*vd<Q$u
z_T=DI%FhnCnSpJ;Mci!g<k<^6eewi9e(yc}>7V>De)QfC@r`f&Je>3Dk$}=Dq9#}_
zXL1F(DE#L#14-C_Xps7*Idz#10VwsNbI7b*huxL0fxsND<q8cLg48*8Y2{-E;NZYj
zIa(ad^sHRZf0}r>qFsx_tE5-bK?~0w3GEB!cJ+*X_(Clhv`;$rBEyu{2EzuqUeA7n
zJ-$$lZn$+*q9ER*F?WP%<uwT3g2Gj+=80?kZj@dTo-$0F97lDZ(du*c{#Ex^T$ka7
z_G>TO{|*{ge@^IN<q2CL9m$a!+1gz=rLV!9aq`+Sd2@elL&m$n*A(|IQ|Ui2gSrwx
zmg%;CZ7}Qxf>Y^P{fiK=g?b7!Od@?Q-7o>(dk`2gY!Uh%?7a9{?URlvh#b6I!>_zT
z8FROll}6GBhPrWza@gv;2a!VrusS-y`HORGH`fC8tkXLZ+enddb#;jsPoLncUwH$6
z@~?jn|I<JFXE?fj2kch3efJLT-n)mh+qZD<-UH##d5;(wj(`|?^xG{q*Vni>KgaWD
z&+z2wV}!tX^zaeRpFKxpagFxISKh*Z`uG1nj!#de?1KhtErI3*Ar)G{EPHIXJ+7~>
zGeB&7)8Ut?(=MOaR0ErdUH>d)?+mrU-~GyPI2piaW>sootX8Wk%=E^FE0#@p-&Tm2
zr<<N%_icHjIiS^ixF_nH8soKFy2*TB28mHUG7cjiU9@pTYYyA>tcEFqNA$|LTee9{
zbNc}_6=wjHnlFd%(#Ciq|2b?nmAq}<(i~;$goT@%J9r}8e(!f}SZ+|PwVC%+Ki?}`
z9{iMsZPN9Wy|-*gNhodo59wCB8t*_lB<sTr6en(7=E~_jKj5Uv?|w(700!~OzbQGq
zlID(GcXHnL4qI{jO4nYP&!fK15A(Oqc}#~lZdEEn^_5c{AaY7XK(`XVy&(wO-fYO(
z&~VOUb-aRe00%ju4IE(pRmqMgwLFf1>wXJj#?g9}O6Hu$?b~<o;>9!ExpN0#@kJYZ
zhRO3`dyOLpJh*cQZ@lq35P=Ur`4pF%9v}Ru$Ce|`PS3FI12+9ethy5b=E@G5wKrq6
zK0+S@R*ryfjbHq=U%`L=5B>q}zxk$m%@|GrCXYpWXA@-E+(c4Qr`*o%c8eH8^<5@*
z!_E6Lul(+2xP7nU8&%3vni!G)_nSWYzQ^aEe}?sXjRy~28-ceGXRUZQbw3C4My03h
z@#1H>c|H)Y?Y$~Zt_i#1&osuab}IcD51#LcHw7qLBV_JbnO7ao@KsuvnSl_HSxj2^
zk$Z@m${s`Q$2&{76Nzb1t{zRrEowbAtqj5uWCLn-MsA=~lq!MNc&bDoc->9+5jtR;
zwZQAk(gA6dd*reidx);x23g&vHjib`ES(jQJW?I$y@mS%vFs!r+LVU@J96mM3xG2@
zxnzy71a<mL1S(ySHGms123Y%;K(Tm6Q=o4yTpKX90igU^2n7Vn^(x=hMAc4Ej;*uj
zh*8>9#}|N-_GtP~E1ZQ*T7{6KOn}7TYlz?wAYItQdyk`|H8}RzY%di6B7x4#p;ehf
z0!~j)z#+hnD-Ck3S4X%8aD91+&U=6ekpsfEN8D}^Hdl!K7N0-*98aG=!>zlwv3mRj
zOdSGWA%+0&SGc;q28V#7ZY2Oyft@$IgwBKD5IKSzq4OTEz4;dY*5CVk`1k(mUjaFf
zsBJTx%c^5+YquT4v=2QZ6r55$6};^B?Wae~jO|wJ3vM>oAQ09^$LcFso=3&{&buT#
zI|5dwv_OQ-^)((peuRJXhrf?s{k31kU~w>|hNj7qVYyPj8kj|aD)hKPT1_IVKk_S6
zTIvo=_Ek%<3#-k7yVjaLlh#!JH8!w$VBfmUZE<$Nt*N*)8GvQAcI6JaY?ubZEN=$8
z<GS_dN;p$X>yP{N%a_`b6l>I|VK@g#`^bqhbKN#tW>@p6#<!Et=89Vu;Sf}`=O)Ff
zx&g3dz`XW406ha&kQHx(wnVWt8Dsca?EXFh_b~2R2LZg7iuCgH%7MU+)LA?P&K3ZG
z)b$e)C3Ccc>kLf*i(_1Z8BkYJ3zW64OKsLMII}#9zd^ZJ>qUA^Kyq~tji7F><iQBM
zjfl#lRb9|n^DRMV8D;QqQ9>R>8F5DJ)ponZ_4O9cb%1vWA)@cE5&8|h7x_2W0jrL2
zeYru{UV{+}psq%nQ>vZ51K>nUOZkJ~*n`o-Ifo~YAL8DfJLtC&S65d!fA$R9%S&vo
zFA-yp&U>))cy@7#=a*NA>~Oy65g4G&M!aFTi0cqQ#PE)Adgm5Sj!&>YKElz_F-}iT
zv0fkH?XSLrcfRoreD{}r8TVg%4H1mZcB6)B4OrN>DW&uZZxuQRGgAsL0FtmiTH)mM
zSo;(2G<&iy-3+%5Bj7X#q@F!{g6(FDv(qyKx5cAJpW*4#&#_*0*lxDCd+#1@-@cFI
z)7wQ))~SsdvF~xcxx#9FEOK!D7OT$V^yCz+*{U3GEe>9bqg}r%<F@K0dmQ=Jv1X3E
zB+W8RS}$nuvol?rD(ZQjcO){xa>Y76nI8}H-m9Qx&{msLRb~utX2D#ACq@e?>B!_E
z_awdN%vp(Lg0cwZ>zx%TfLG%7HR8?o2Cl%_L>hYmxJG~!uS9BPl%n-UW#9&=n*yu3
z9oV^1$eWp|UOu7bIB8r50G~wQf|2}5@0w;?O8zNLedm$8Pn}-fe`{)H-`H<VmGf2R
zv%_TUoGY|pk+^aK=o&|<`pg$7D#eAF#Sp+}|4wn{_Tjp<-c9{xgT{b-Q7S;6c6iDF
zA?J;O!P?S9iG7MGPT#OCso;8PJJ{TmfFDNML38a;+F!c5y29t5eTIt{&*4u_5o18y
zTmhknb8B>62ZSxQ7cX#qeu22%AjTf#32+48bz)&1%9SJqKPZe9J5o)27W;M-@cHMT
z;o{kI#1L_Hd4=aspCN2F;Lu~Uxx)F2=h*Z;HX))rTI2l1C7zwX09I>U#emKMSJzhn
ziw5>@|E0f#|NTGwhq!h34*a?UNI1urFV1mxb{j{x&d~V|k<3XvqjMhMpe~0L+>w4G
zCNSwShSno*I!cf&&(BWJu<AM-%))8VO#!+xttl?`;F)vSZnyaClaKM^t1Eo_+uz3H
zho9p4v(NDHCqKsXr%&<Ln{VRwod@{#cYhQ2U;he-oC03P`LieZ-v9E)_>({RxA^XN
ze;MEX#b3gz^Z3@czJuGh?~Z`kT)kW6Yk|Ewy}3NfUU?HVIZ;oojrOz0=bb$~-?6K2
zEbo&6Czta*SM8|<tHwxE*FXkhQ`y^n0(qRTV<RvZ@6(tbEzUM!)<~0!k5o*m29Q#P
z8!b0A_G`3S5>b9XK;(^<Eg#j_DNb33iLaY9dwC{IE581^MtdGet$tHKYk{<8wsp0u
zc<Gm=zib%@2Es6I#t~`2=4ULrkE8MHeQ4kDy7#y~fDTrkaK_0gK}f}MbM{>2>_F<H
zlt`_G)W9lp_~wA!{5wh4*){U}#GzBLHuNTu*ZlmxT8LOdsg|_dwU(c}wScv_Ynh8_
zkhTg_%mq}%ZRJ%wOoRLCdV?Q+|2_P_{-6I3tdEbe>N+^*<@>Bwh>_9x4!3UIf_H>|
zdyURH^s&eK_!uW=XE;4Q!}0Mcj*gG9K01Qyj&QU(!twDDn8lyxX0riv#QMk~Y&Qsf
zz$Z^W1&4^f54b!($MxkULcc`_TfBJj9M{``Pd@z=6oJk46^=U(_72`V+_`fLpFZ4x
zqX4+S@gMzdy!*{>B6`^d1U1(Z0)smbhC>9w>7?Ct9f&-T+)i=+2I*6oghkYTY1}w-
zr-lR+*4;XPt(OF5Uj(4rhnPj)l4Hd6<t0A(_#-@ca39;v6|OEXu-$I3xw^o1dyXeh
zKf{x!kMQ92ui*ZJH^eV696tTz$N0Vf-@m}ci>C<NEBxdq@8QQE`~d%}zw+1d^>2Iw
z1P|tjO}_;Zq3g0D#T*B*uG7mJN45#v;XudN6I3pVgS8!dOz!m>-bi~uG%=~qWK=JP
zGgvbLAu1wJBj6w%YNy*;j0br&^0ntw{MLBOK%8MN3mc2F)p({X>mbo+AE2mZ&ACbn
z7e%;KVY{CIMI)MMjlDWc6|MxDBapdA-gD@r8NO&#130#MY7{Ts)9`+-R5uM)Ed9aP
z;gW+p27Nm}Mct<wbjiRGLDraR4DHiHcD4iK#+6VGJBoD~KoKyoW#^Ow<!l$_D1kWu
zIZTtYrke!||1!X?G+<m7nzELz<<;iuVenhNllqqV7Tv3u8}fCPW_}`d9dL#+F)C0u
zv4~G`BGa|&(5>Xi?DBex7cVZ-!s)w}XnSrpxW2vu$B5I@6LeiCI+jt479-g!bg2zJ
zt+47=Sgl0JHiRgKgE^vGiA<DtKFM-<a#*icpvWL*FaoxrM})_A6A(8WIA+{FIs%-?
z{M|n}0iT{BhJZVF@8T;D?qPHO4F337IK~_iwp%=X^bq$RJV1AHn*7CXw;Ob-;Va%Q
z)DScBcVz0)cNBxWqG-aA5jnzn`9@6A7H+jxxyYX_9R}#~bmN?SgQN8tckbN6+1UyH
z>`(s$j~_q8_0==<{T4?@$LP8hj*idp{U7`pzW&W$z>)6|IpVWVKE~$y5^ucm27dn!
z{tdqR^{?Thk3Ycm^#!)qmw5Qu=eWGQz#DIW6~`xM`)xp~G9KCvbzB*H^%m*n$VPUZ
zR;#qT7Pd0I1Xd`Aam5rem<Q@PY^Hnzz*PaNx~$1pJ&oh0{E9TnPdD|oN-$Sy9S+iM
zL>{yYq;1$1IL;h9M#g3qh6p&)6_&MP>FmQYHI7-y!L$2;W;PAWS2lbS#o0xT3R0EN
zA!)Q^;OvJR;SIjj+Oc)m@?#BvE`S(8%+-2ybKq=WtXonmeBHmxCK+w)c=uS^MG?Sb
zNZ)&tUT*VvlkK=K9e5NAB<@!!Q%3|x(U4gJSt(wF3syc$fUmHJXW{DbEl`)0FiNmh
zbLGZzd%KQob2v+jbMk~y*nBAi#1-7Vsk{emQr%C5t>aY>@b<g!;{Wu2{=XphJwoge
z`hd^}Twh;dyV>G;bB*ikYizEsvAN!0v$?|c^#*;vLBH8xbF~pWaZ&8vZMPc%e7%4q
z(H8B|haUZA1B5LuV<@}=0GsUw@4WRE?%qCw3&82g8r#h^t~MJyzqmx~uMwHC?p6uN
zdjP!IVoikZXpPry-@@<z&cDQ+KfMQkw1Qu+(Dywy*Vnjr|30ptKSg-!3hUz&c-P_L
z@*H>X-4*2tBDG)2Dnj5?_aVZ>20#Kv9Roc_1UQ5!-7rFcUWmPb3E-s$=c{t-UN-*a
z<t5hZH2|56t}ZX|`@jEtxVk*YFaPo{;@jW;CjQmG`j-e{gQriQ;b^_V{rerh^__3w
z>hc0a9aifjeCIpAfUC===sJ&||Ms`AUaxWY&RzV`AN*_l{_p-PeDd+f_^rS6Tlo4n
zzk^k`hM3ElA=~xNaOp~!yynreaTaQTcLSW!0Yj(Gw+`EQR=+nu7a)S%<xx@E17pmE
zum!~S6xx*~Zn0REfGmZ>xDHMiB}Z*QnKz+D`OE3Fs%V)<3D737NA;`>Z@?}>+Q;&)
z>yf}810VSnp1t`1o&kyx6azQ{9t?7XpmBt!FB}(Qh6@qY_aFv3W;ig&5(X=?jt7m?
zK8bBW==z+xiDv2tRfijNz$(4L*>($!7V$>ewm{zy;1Tck6?+Ikn;jpABsAE)RM8C6
zP9<kmeVM7DNcRR#seh_;uuCyn4`oqI95Sbir^;d(his11mXa)=k3bmGqE8?cL10I{
zi9=X@);Unu0-5D~yL@&(9j?4C^^0;I(hGORoqE^0Z$0Z)D?E7f%`!md7|KsFA;Qpw
zREHq}qCJa3yEX<m211PJV?^wGY&RR6KYfaS^*g_V&DA-suP*f1Pc$00n`^8(5APke
zmzQ|)>@h+RkhJ!Udw0)pcKa@#KRcJnJupJ(;f^|NFBq$%BV29+zW?42vHi0RUR*o}
z>%@C}e2k;xBdm^&ap&HB+`4@S_wL@u>FEi6^ur%wwO-@a?c3;9YjI1~t#RwlZ5$sR
zVZB~~oWuI)7~Xg2LyuLrLU(k8)zOi-MdN_WXHRf-aS0!fl39&?=%$WcZz7$|W|K6O
zPfky8=k{%c{u&n-7x>=yet-}O*Bb_~#%2Tb!J+d<00^tq8U%+IFJ9oqix>F855A9Q
z&t9P4ZgF{ai95G%;n#lUSMb(v{04sGH+~!ImAY>O74xtm1JZ?~V4}=aZw}PW5z$Q?
z#vaaD7?nd4i4Z_{&A=uCTL7CHfK_>xpbr2%W#DJ6(b;AZ*{Kwe*Zl;TiVQ5Bj`FoH
z(lKsc!ZXD}R$~Qj{FTZpEi6+n^09~$ihBcOgY>w_Befsl(s%OC4;Oa39<(C>Qjpi_
z^A)rm0Jvm%Kx}7r09Fn-3R|2#eF%!N%Ajr_6yT{s0R!2$!HXijSb(|~g0Y6+MgzB~
z1@0Q<N!JnXdINuQjDS~=Se*fE^{Yd$-CTqS8Czw4FTTVp(;C2?FA~*#0cjr&?3}sy
z5)PIlQzjq<VVDMWmQ3+183oof^06w*aUgh*rW8Ycvf7M<3h*^RTp1TtIm-J%b!PzK
zAdYyvM`{LJ;8Hxh*G8PlZ|Z)Un>f)cXn4WU^%c^lgGo3ST*5h@ViG~>2B0ItS|r-w
zBEu8m!E5*7;+cRla;nuPCN^H2KgaW@PqE!xqVKN}S!~{QN5}ZBU->oMd+jy6`|dYz
za&`tHhadjvJzQUHaPQuIIT{TSLA5b2F3xfO;ss(KaCLbBAmaS|9Op02aqsqB{P7?D
zq55|XTCYlD6w*&Sul-F``_C^g!2oWb-NLQgXE-`O!D=P4c301yWBvBqV7?i^@Y5n9
z!pX@AVvM-Hyn=It_uhLC@4x>(Hrs3b+OPd4wwo<}@{<qnqaS^Mdw1^O*M9B0__bgE
zO?>5@uVZy|0zg*F%gYPA^Ugc?{`bF+)6-KNA06Sd&pyTF#Rd8h@XNpQ>o`5TD+(!?
zn0syQpy`Rc?8m!IrL^MQ6qTT$eAcQLHNv6@*E4Vtf%BNnFbo~p7jbA9->Pm#;4MEH
zhe({hdIGP3U0UE2Om{p6c9xGP1v$lES=Z9<h~+cj(v_<z=S!<^4Gx3W)6y{@Y(Ox{
zP8Gn;>XOP;fvZtF7T2x-ELNWcW(x^<Qt*4^0FNH&%OwNIoGV;{rKqjuoZk09*Z@%t
za4-Yp8iRl`xU;#~?m-a0u&x4PWi1b&vfTBNY~RDzJ(>nT!|SQxH@9gki_r?Vt2#B9
z*30W+Q!q8I4xTM^d^~{Iva<wQV}7ov>E@-;^5Jx#uUH2npE+pqPyk(FrUaM<b_qxj
zfxI5ws<nS2r{iLd+g9u-Gak|3`6FLgd5d>t*!p^o%Qtvtd7}VzC?*}`W88ed8^rC*
zx2_ZlMFE}4#%Yw`mh!)p!5$HhJ7r3f!H@~)+0&;W4kDYuVk7VT@(K?>`xFl!evXT)
zbF5cukOR7-W4!U^JNV^a{|$WeTi?Om*IvW(7te8WdWv2*tH1HqTUd2#gcz~vR&eCO
zYHWG4-J)|I$Hzx_@$4y{K6!*&XJ_d99;>dyix=lutykFe0o&~c+ij1{W`nK%U0q({
z;{VUxe|AZdq<6m9&xDImsY`Z=)@o*YL=Og7$X#Id@{(uo8-4|T2Yv{?<{=Lc&po_%
z@8a$*fn6-r9n1^{qkE>Ob#--DRd?yE)EUZ^3BE9o$jHb@RaM;p_|(bD$Z&Tvb8~aE
zzx^LI98eV{MNyDv8CjO$go1jv-tTr9^twcmqS<aTw>VGSYM$sVJ`Kot8i3^JGQt>C
zO3@$maaD;Dk}S)~vH`{zzWwdr;P&m?n99=c_2_oD+1=eGPP&}C*a4s_a~|G*NPn=+
zn{U2JuiGPu6Rb7#dwtHGyTszsS=<CuHd-jdl612BoK~$K&Di<+-2H!t^G{UW(J}%O
zQxow4P$%{#_-PixdB{C<Eo3#QJD7}Z|DrQV-avi+2y6HW05Va}qzZSsos*2%WLFh6
zE(sf*x~%$dC2BSch{WDpJFZtuwQolKqw+=vNA0x}e@}wK$u^F@n+A4*Jpee#Ebgy#
z@7HP8Pe1E-{roka#8x%nngDYUzB|zQ9$X1UNyz0oJ%FZx*(A(7q+z4u(`@vJP#D!a
zdF}y7K3(?2(;E$R;Xz5<<XSOZ!g0-<amq<S(98l3d63iC&iL@2+1aNbSKNVWC#4Cr
zXBF@`bIlA;dKx*+Wv7l~Cr3=%7`gZwWKDyh{fyk`*xI8<w3<!k=I83NB%=%mqMmJL
z0mRYgk?M5}@W%fiZBMt?u7eB4kzjBi5IR`K-Xs%1)t;TbsK47E^ihbZgw<2BuY7-W
zTlZuym+9=k_g_06Afs>~6R%%5Wn!_c{~Jve9uaKmydxX-iKL*n+hu2Go1N`#^5GC?
z4Kj+EUp~wE%g^xgYhU5g)fbsxT0zB0U5=%JB#Jb`g)9=;ppS8$C{FQ0;9Qu@<2a_7
zrhvc+C=o23zd*Cms5iEb#=;_T5>Kt}Q_p*}f_Pth)>@1)SYvRuqB2#mF)RzL4NfG@
zPMfsTIrMoQ{!E^(Ofp!<*B8c~_kNsF<nraq{MK*&CLdk<fT}9_?sxy1ycp7%YjN(}
z3QH@?teiQ^{QP|ApZfW}`s&LJhkYKc-si!+d)&J98B0rx{Qe*O5wCsqn;@g{!cxEW
zgi&&^O;bGa$;NhFwx9f4>^<iFtlpDvi6_CuBuF^0VEIv-_i7%!aRU#Jsy(&qX<Ic0
zj1!elmf`#!u-xYdFj3bC82PF1CeN2s*Gj)<OC%@3<c!BMs!NTY#RQ0(taCJ-Os8<#
zXHPeE-53K++%Hq-OoEQd&!akK0gmYzakAchYp!6d@zkyi1caYx-SkPaX6Z!5*v`Pz
zcj_$rnYLMjV$^wJ@8+30K4$MG5BnMCPd}{D039o#*$dTMgFsQEqG1oQIKS_K9|h|E
zUS-r3K6tGb9Av?tR1`4BUYAe){<7t?_jM{@=cTAk3nmTFXMi0MRHahL1E5F>*j#@^
zRb-qwcOI$L7%U>EcL{+9+fzr#u|vY>XkGB3OyOkRvwly?QL5MdpTi3-kb9wS$KZB)
z>KTR6IM|=F7dElSs>uoHdV=dm;F7(o#%-HCXYU?hWPLIV(1`>0z4tV$G9w=j7!C&f
z>X&bY39MGniYVrv|GR&~%Wr(0_TmCcX(or-2nGllKw?Nw=17v5-Q6x)Ym!D%_ihvx
zG4BcgBvLD)G{M*XW30zZJ$Bch?wc`i4G*biBqf??+^RkIh%n)ef)S59??@KucscUA
z11rm@I<fDZaqk~9yZG9c$C`>XO~{5NfAv>?i4gGOOE0psv&Hr6p9HUHEH`i6LP*WG
z-~1i&d_bHe<ay3de)=PlICv}H{_G~(TbuNHJ?0nYdGV!}NgM46uol|8w`mxyVdJMg
z(Z0!hJb{dDKxZu2;|<XM_>zw+BVadc{hS1Gum@U3zeZrHYXf6<+C^YroinZ-10C4&
za6RPwY22_6G>M6FCQchUXpG*1ak-ONW~$B+0Nk6C_v)JjQ={_sHi!oS;3TM<eIrwH
z1mK*W@}@!EsOZTdGaYgw#-qFq=+*oEohJn(2{ckrb$;xkge-^Q5QMBjodY+rpJP9=
zo_PcQ;63|8j*SuUL@$cz@<%UQ4KT;b_L_whc04Z<N7(T^cd%(+giMe5(~2XgVV^X)
z;w(uB_I0O4a>s(<FQIQw$=HopJ3j_=V#K;l0HQs+Ii}7NG#d>zA3X$TIdk^xcub9$
zW-;*(u?KPl)IohmvgZ&m8igYm873OBca9KD`ZbD)#aj4@S|>s3#C?m2EvaB+WGfK0
zN>Z@*ZtMZ4$tlg(8@X|Z900s~pBdl#9#GrA>hGPtZ;4^Qx|fOTCwlC_pz{0L9r|&v
zT?YV3eZPZVmqD+`=K4AsL7c{6$1gR@m#*^WZ~rbLO~-aKQtU0xf-uKKQAAmm<0SII
zHY`P};2|3T-K5d-UgCM-sY9rEvTR7R(HhU_GuyqdzxMjk@9V-)fS;Q2k5&yHE+;*I
zZz4E2_GZb+CPL57?k-h6q~F`+-1)P#+bvcft@86<yoIxlC{nDPSwzYX=gyvCb@eXR
z3C>p)ODku2{`nW^wA(B%ukh_R-{kFIy+wccL)vp~bR@7g!@1yS5o+`kwKE2HV(Qol
zn!Zma-+{x*nlKJ|B5{5{Lo{lRU~kPF51EPAX#|QUEVA}pyVuUy-%pl%;P276X8#DD
z1TnJ~NI&)6<T)p<d-6Je@VMW1!hV?qc9YwtS#pxmF{>Y^f1jk>5r7-9p0nx^!qbp8
z*xm&<`H2Zb;3JJ+m<!>cMu<1k5ZB?6hB=hQL`9PT>@j544Ib$C8KY|7fU8p@j=!%5
zm6Z5K13NdjPn~(x+jQ8SoOUzwq%vAxrpNkW<NP$@3GZ_}<EBn?pZiyD4tCy*armTt
zcsgK54eV+LYQ)No02sKvAh;9A-%C%3g(hjW8aQj{bXrJNljf5S5Oq#Y@2S&v$&f)n
z?D>Z6+4qU@`YQwo@5ZXy0}e=|woLwA|F;KhCKuKT8$#Up9oY9HCc)s|^tD&s-n1)%
zV|`tY7&}!>1IRsKIWnX@2u7ywX{Lz%ALMA;IwAEGweR%NeKGNS|3$&GY<GJHn`h*`
zE@={@l@EDorRLSId=-@>SSIgr|IiKnCW)eGY%CwVY0JRV=?EeHc=ta7Y*ks14f-r}
z+Tg)Cixh$=)>F+q=zY9**L$CZ2vFU#r#?KumOX}?7zBFu9PH}ha_9tbs`_DJex94_
ztNh>x-$zM75=T6I@Q`-9jgB;1n;UeyJ<cpIaR2V-^tw9~Mb0<B^&QfNqTlaw=FC}!
z!yymu-{zxh*QiXz+WIPQz4asJ=a+c)`PZ4BUtz2&K8lbS4Z&mkX7q;7h5-iv+*4dh
zDuT_<NXAD06CiF9OpSg<3!?Fd11*>yeS30jk4(&GlT+I?(3^enesDFb&dJZyi{fG6
zXHupy%V@6klhe*IH#w^f(;Ydn$DVzS-~a%hHSf(@8>ZXRz)n;oCOS|+n&Ol?FlUG~
z#*2A`j-W5V)=b$6#{BsAfgQvljXVxr?CW@c@H<hb5o<JId5J1*jPq>Ixbo03{iJee
zxyN<c#5kA)c28?MD1afs&h8{KJ<gw2J07RcPX+9h2<gFu2x_#DHr9`sB2!f$71n$5
zJR^!C8jS>@R8ZqeMI0+Sb1gb^ZG`Zcs$kITGPkh6+M`Dd`+cMm<VD6yFTYH)Ifu0#
zrRr<~9#kANia~}4q+r<Zvik5oSFT(kNt?*pvvV}diZSCBz8t9)M__6c(qhEOc}6ee
z9z!?ssu~X_zdu&qUWI;q9izXT#d}vLT9<gHZSNTOWB5I(w3sn3h&{M84G#C6HolMj
z%b^$hT`0uF^~c#hdFR+#-@xQKw#bO}USG;6;^K49;l-p6+QBxC112klwHBp<Q;yOS
zFJPDrNt2MLHLwL9V=c})mKGOr&LWg1&+~D-wrL~$F{GRVdE<6@@Hm9fdul^*3Q#sf
zPJQhR@H2DOI%OM1eXz5=&F1<BTiY8{Re=_Yl`|_md-V#hzWOTfzWp}CL4k86!{Lx-
zGv)rhTdc1=VlXUOT3Td%ZIxfY^D8#DHi@EyA}@LCtzYoc%dhg%D_`FaMyFQVeWR{M
zulqp8zD0YqYS$xjdV$_|UVZFISdbY~`JhIHz}$^M-e^4^fw<9HJz7x0Ms%-?QM->@
z3{L}|#$7XUjhOm9z1WR^KgMD+3pbpmw`bY^9aaBfkMh9OcgXc()_XIBo)0{eLqM+-
z5P3U(?t3Q+ycD=3#^Gm@1uBU^wga2zJq!ki-`p(mQx9v*QcR=V>HDkOI0_V<Wn9yJ
z6vYMUF6nNN&e7dU*8oWokPt*VhIE&7OG-(M6p(Hh0y4T|Dlxj_`9JS?!HaF5U)^)Q
z=U!GRu)~{jprcZ_E_I#g_f_^ns~%Oo+ARvw{T)N7_R1qmV$5v9R~8jo+WjP!C=^At
zdR5vk6U_3);dD}qz>VPh8h*k|0~9H{k3TvjU}k2AVhQhX!;6;pJ+^0c7EI0A^l_NJ
z&Dr6nW$^Iurew024O?`5OlO^8ZFskU^mw@Satl1gMobj~y{D!;T5vEfNZ`iE#-_lG
zyDNr|dj~jv8K2Jr51p+Xb`K<iPvCV#zTYj{2c{smEU_PZ{)HhNOI<C4cdI8R`Adqe
zb&<pWbO=GEQB|Cjc$iJOuaDNdr*>avYNUz?fYwqkglAMky$H%!72lWQo%|SU?2*ta
znO2T|_B&iRq@)VR#VA!wzooTxam93dv+GX89DDrb%TIgcrm~*IU!CQJZo0DqL%;B!
zv0{;7VgS#7c5nRIXCJoCh=;YAZv^aq?WYh%nc7{tuo8=y(Ph92mNAR3=0pcEBykrR
zE9Tsy<VXERS1e*@P?{phY*6iZTdI<qO}q;zIPE<MQdwov@|Qg=Jud2g)#o4bj@d?g
zMtUL&m;|cmD@(Rbil{zP%ZvY=I1ynlR~qaObRa-Td!$*rMnDTLE?)bGp?m;-rNI5k
z51tRVqNgkR85l@9K+b2^Fz>uwng>>??ZxzS@;!v#ZiI>NzDc)Y8CPBBElHe~7OPWO
zSA!IOO~p~bKZxMZcTY1Qf{)S!Sh5`-3dO#KZ5-F{_h>uTypLn$>D_?j%u$a1OjyC*
zmx^u9mB1sU%KcIV(}6(3X~o~`d}6Q~*}(ruCO{tb=y80P4GjaCK03BUv;EPF&mxw`
zHTkShNhhKafWaQ{)k(UD1z#gnoyMA{ZBv2nsYdF|J^5qJQ`Y6ci{eXfJzuVqTK1O3
zK6WluHOl>Oc}ifdtv9~&pU|ABPg_ZCQOFe!a6MQa4t`1$&spV6yq`_6yB6uFrplBZ
z(sy?RTb1YGEIa_$g2MuXb-cZkUkok<f2zHNd2Vh^om^?kfCvq?N`XNJn^O*I!5rol
zk)}RbB;mSWRJBRr?sSRTO3-mguYjGWYeE#^kzKrv+jk{PGE?V2-W_DS9a&?)T*REj
z%+EVyWIHaXfb)U6JNW7iYp5C_-^op_xpITNH8EC8tC*>!#8kE?!pV!=6f6;^i^GL}
zg<P{coPNGk%T8#*K%&uGG8a`N_haI8Oah}HWp$OVMO++gYG&=~nxI<JyQ`I{G0P5U
z@Y|dHm`!F}Q!;n7h$*Ymxg5fn!{)s%tM19YJr+KzpA-DSC!XFut<(C+1F2={=*F<T
z?bgRp2KUFUGfq16W5JdrT>h0cZIAURY@@@S%?F}?tc-TZl`yYY^UYS$T}OQ7Jd}8M
zAmk8ko!ZCEQzPm_f+Jm{Kf!vMztl<Lz$YCPbV{|YVH@`>=2V2o7YJskmPEjbhYYbW
z^__A=&BfiGzF9J4sf61F1mqtbkx#0Ism`#++O7HBRlEgLaM7CZ$;HXfI<~am-d7hD
z1Jlcn__W7o5~v0=SOp|*q=#5P)xX~8+@4;dc9_bu&l2StC$QIJE_>(tKnz^XS}>lv
z`i_n`uMAL3bPeMSKE1T*?qKE`qw6Q%%SDQIpF2Ke=NY}9@$nE4QXGpGafMQlEg&5n
z%d_LN(+u4U7CZiiaRd=>9C9v-CByxV1W%w-+egWZkH2wy3?-rj-wOAx?hH@sk{Gy3
zs}k7R)B7;x7HAIaTo=rhbG`ec`^<;vZYs@bLOAH(x2(~CJg;t4^>8JB23m$oQN;}-
z_utgeW-siLoe0J7XJeXbMWV{YX|dMZV$($T@tDSi0;s+vSL1iqxD}$D5lOZdBz7f~
zG#qaw%V`G56cef5#(hhm`u9p*_WHB3V)#nMLBqtXHZSJi@US3htpW@zN$v8&e6opG
zw}Khl)H2h%vqFvQ0VEQm7g1et9DLTA+UN&It3NSe+h_+z-tAszP<ik$v8Glen0VnO
zB4#_S@gr*`HGiO=aWcgVnp|_B)T%WvoObm2R)ym%GbDLxOH=tmgO+$OXDmFV&EBTL
zJG!C1V|iOm?p9Eb_O#$K;CXoT$XP1uk^58co0b;ulz)<n>N$sgFn}5pc7T0+?DGST
zajtIveT4Mcgo<}Zhhy2FqodPJwfX36GA1nUpX%H0hGGM{<4}~sVE-?@0%}+a6|*}q
z1n?<DDDiNrxWC0lU?N_!hPq_U?i0i?M|kA;as0XRzyDXu;JLXaQ>y9v@Y1Sc`hf9b
zgl?7Lne!KBf<U<(Gc>5c&{z5%yD{}wzTEUr@7?O;LaUhfuNH-iiLuJsZ@1#?KFob-
zdM>R;0{wVNOP`diAcMO}g~jyCjC2@VKYUa$_QlCUIiEm1EIgh9X4cXw?j&kpY{H<Q
z9a?yQjp_2C&hwpLjMkLJ*J;Ow4k+yZwJoyw?u6Kh1lr_Vpp{vu!u2AQD{j$Z`f=H}
zv(kk&+@}$%7p0<Qcu;gLDmc&Cr_Fa0wflZ|Du#6OJ}?TXIJgp;GcFQ%d%jShuRbPk
zGAe;|Iu<wE9SVn~dM~|`pa{OkhM%J)7or92pRxxBRbT~Dn28$+o&4_)-YO*?g_R&T
z#}OFpB}uQH&b!l8_C4%#dgUfXXXj<K!6p0j6-kNyW7~EmP98%I;JR%OkoyY!$4bzG
zf$vfue67Zf@Zop#s48)}*aQ_@5}rzeCVq$}7nQU_0*ZlxeYt6?M)=Xb75&?1ZY|S&
zrTBSirqb<+0($W;QZ=#Si447#m5dX^DhVIH*^?p?6ek^trRCHbvK>&?wz2AnWn~W9
z<>2oS$<7tSeSb<MuP^@YCij}7QvC{!{~Y9K{?5gn;FXuoC*caCK{GWHwX8;EtloC0
z!{Xv%Z;tgJWLJqCWih%4H9of;QG^<YpeD|Y$e^i$5mgf$ga|Whoh#<#cCq|jqN7Xp
zV#uG`NmGnccZZx#Mf&c*$ZRrA=Q~7EY$Dq#3pj>wk<#`@ra|=zHPoN(I(oGxCF#Gr
zx!*yh)XQ|uBI;>L+}1i@82amPc2@AboGDe_;Ht(c3#&~%$pC5kB3F~R#!%_l$m}&T
zK5W-ka$uqJp+&q-%`&d;?Z+fkPmN}cNij<;eeLZd9R5rCYDEZg)!(g98SMjk5bM4R
zfDBYy-3`&p+sQo_mx)QRSx1}o$#PxcnR7OJaUSr-td>A8l^%o)dGiKeIj_N`w$?5-
z)CI&CGA1BY0@_<C8z0UP#ohSp%P%qW?YnA;;G+}+!M~ol4U5q(slsU&F~M{u4H9~3
zU-=r+yr`>nNYZY)3cZZk?&AaI!GQ})^A7Nb#oe_*0p!T9RDb1%A8zf&zc~lri@D5p
z_Y>*0bi;@qUxkwCx3{j49Kn48&tu=z{H0QLHVL0cX(#Hl>ek?1*dE{Y#-RU6pK#}6
zdYwB0_Qoynmg|~?t52UTeP^c3VPS3Yt8i;y`b!gV4zm8oZAerSz0d@PNDe-Lm{3o3
zUE_G-zHDas{phWOyuGqV^^1MD{=F;k!}j!gV3VITIC1^b&b0cu%c91(%wC7xc((24
z(1uf0Pqw{wpu`y+ky@)~`70~owI0~FRtz~ac~XN4;@o%!iu;BHc-u+&hbfu>tATK0
zuZpk=9(MHDXXLBVPcKAQ$>7<I({Ecvb;#M$G!3xZ`Ex@HkBX?o`O1fCLB~wbqzp2N
zV`S#`-%eMcctwq-@cjNV!g_r+9BC&T)NT+oS4N<JZ?6s3S<tIM_0Df6NMQXsg4=mj
z>hyfOMUlH=X0waMIHzf@=o@#<RdmixPrwW-^Axv*y`Fy5x2p$~J|(h2R*i7a0wI2d
z-wOI&Ul-p`BORK}VE_MuRbyM>6;gk;xsz(~^f{LfC8!g!5^8yy;H4Uhdd8H_m8W?i
z{Td#AyZGL$R8<y`*;{a8i6U2YlPV6Us}VAs)z31;{sny!!`^_TqFOIE(`!-;@MoXi
za_6kww>QZO-Q#^*v@pu>=*6W48$YKQ@t?P^*MjeOF~XSiO4~o~f?@e@s$@*A!hTE}
z6kr|oJ!YSrT#gj3-iITDkPrmn;|8JMX4!nSLr4cW6(4gaOjbn<>vQziwN@N*6PQqU
zXXE^sQdZyg+a*x2Pr#R}^#`^D5vCXtAOt<QwDhJlRC*kE>Ga>;vfDeETCKm!+I1|u
z$zOk<`yTWuA()@$gS3$A^+tbE%yw4egx}tlL~1Fq>CP6lxLXkHbj%MX=NTcQL8K4)
z%J`Q0;2@x|u-*)1K%=kOdEV5_tel72*cf(le>NrDUEdSO-=*9!w+T<k1=%?^8IY4N
zi{xS(flp@(@3U@h>?MCr=63*3`Guls_3L}ML*V%)reXO2>iYd9^Hy?`^BsJO^kREW
z|5Qa%gO(i%A9o_RJ6(YspW5@OR(x~*fh0H_uN1oKL*y0x16ye)e+ZySe2`~<L4SA<
zPVN1v_8C`Y+E7wH-sRqIjQpE;W(Gz!W(`#`f6m#+1rOO+#u-z(hB6ZPr_k)2hwg8~
zh(#!s5oypzo^Mj`elavNNKW5<9Esfd=3IDtPMfjmkw3k2wXNx3>AmpPZa-k@Be!*H
zV`C^uD7>BB&+`8a^DBFK-DR`S7l~C(pbzeA+f76MalH6u!;W-SuZqzSXy5U$5n5jw
z>IKwTVn&OZU{Xtm<Up;;Go^Gdgjg`OasbgL4{#-Te<|B#YY}T*FkO~cOPvBLghbH<
z<6M_t6|-*B>IVL+C_&qft!8;r(1{NZU)J}IR-^9E9pp;p-lLOGzc1D2;o&<u-c;J6
zCc7QmvJj9p5&~&(vN?(7lxhHm5eWmt|9Pz<D6@0qWUZvBc-k?Q-IZQuspl@4dcQwO
zmYp9bCm|f_F!k~BfNDnddAWgKFN(LdoR+x9lCPA%``>I4yD@0I<%g&XHA1R$@Ai{X
zHH)@uJde~@Jf&arG*mTy!#Gy+=>m;XFZ^vVrIsc%N9QQ46bA(_u`QdhC*2+*wkhQ9
zuo-?niGjte{hT5!TgNY9V%#`3KDS+|kM7YIi8GY3eaYZ{^ABCGtLMJXkAI7|xB015
zve=V_<4jlUjcG^cyzh(`_4<HHjpRjBF8=8npS@>dlPJsnzz)s*!iTe(068qj0%bIt
zPHMYn3VP&eDuaZ<$3c8n?R{N$SEfPNL`-P4-npFQ%9^PNs#e*KK^iN_7hwnTu&`ga
zQ=Vsg9F*p#eQ{G?O1a{+URwM(r8WEO5ya6k{Fm+RI+f1u8iP4Oc75TWm{(T!WWy73
zs(vdR2cOL(<R3iN8@4ZhK>yJt^o;J49$30>YM7h$sOIPoU`fBPiKc6}Odz(_l}6#l
zx%5?xwsv;FEBW>2=OSa0tys<5WsHtau+lDW4{uMe^$K+DmEeA#(wMZmy0pS};<t~t
zQli^=92x#}?{Q6c7}l%(xvj4@9-pVBc3$6~nL2-~s<%cfe_zV(2}0<aeeEO!Ecye>
z8mKZc*<f<#2p|e=4yFjPs?iffIhLt=DZiX+n(;f5wWyeW<;ad5hwFKus+%5Z%4Lj{
zE8@gyG6o^P#dGj`Zn;Efgfi32ydr%CPG<MOiZxHS=zheO)YO;_zi1ZCsU-;G>Gz|=
zi^7A_X6~w2DoX7y4p8Ch7r)gV&M<;q9*M0PV%0XZSlfIZglSXk)1IB-mTH8(gAQgx
z$6b&vXaB0}PA=<$wR4o!u_^dvl>*!dC|_gWaeR5eec;jcIle}5*FD=?&ZW7(R)dUl
z6k4#*ock`kpP$$g3c#+~7rGSuK;P+LcR}1zMS1DJxyyi}Lw7O(or)6R5MB!<#oVWR
zzF0(laCiiqmMa}kA*?YPsZ7;LaAHNMNbAU|luvSTFC*?A3QVJF%%;}Cr(y~^#kc)X
zLWI<IYVq+&L0zg9TXoCymMU8<=$BF}C247Ci3R{xpwut?Du}!A@cc^&JEV-&d^xwt
z3_l|yQzGDqR@^OktfDZuZ#BSYacQLwtOm_SB@VNRJZN&9@<}#3zHtx-2Q{e{I2?TP
z0l#^(yQIc&I`uTzy_G#a^|d7JVJx|XE!>u>`SzOa5!nwT06kk8u1N&#vO>_D&y`&R
zVn4cg2v_FLl9TLkHm~WaOxQM7oSb~Me^u#8C}sG;!6^|c@>}jg=?Zm>v-eu^U>>AQ
zj@f7&5B0}NHbb9UQpar3*6bw0D}C3m>3>w@x)PHk=L`nzzVlo+H(mcH`(j_CKn7aa
zJ29@k*s}dISNFZ#1@BkXJqZbS+VvT^kc8#RkbF;!q+-tM6Gh&#fu#*8q0?mCn!H*i
zBQ&FYQG#qo_e54>Q#&H&XXScr@lXet?W@tfJq%D(sz9rNnjTentm6Q7*Ae+kEzk3S
z=RjqQ+M`8MlZtn?aZ-_t#a|EqTBfWH^7O{8v?b3ZSD(cI?q+K{&o~8$dIrU{zsJAJ
z+k=CHMEsLp^vl<k))bBw^?Cvrb}v&o>p$fila?ui*3+}9<X;Ys^LI^bZ3%(JcJh7J
z-?;@B%y*x@a9e5X9a?ftOLIXUgtw@=FHO4>2~2tyJzH2@+FbQn-|f0u>u-;39B~rO
zJ#ay~Z&JmjR_zt-&j*F}p3%G5FD$O~kfEy>#d8RApdDV)j#Kc_Pv3)QXAiiXX>rP`
zu5WZ8eNx@QQH|d?Un-H6y;b?_De;41v(v%TGwxl5M)!KZu}$V-C;pc-B1SYx{euIH
z;XHL`+Kl#g0()@7tz&*8Wn)*@WH$8W1V7(#-+0}x0;?PuaI})F+}AuwD0&qCyU=<K
zvo}E_`OAeV2nj}0Q;VJ5OslPny3VnGC2E})aPu^k7&wmf`JK|(SbaSsXKd=r%sT6M
zw(-dOTa)gsxlAkeDGZCJ>*t@yJWsM6QZ$2^q?gxezR`?g7oN>hqgq``x--j*-`rq@
zh;Yru-FaNR;w01ljj7Y*25a*@eaTiNNY(AO0}dI+Rj^_=6n%fj{B~s8wp6)Nl-0`t
z9nM!OkfzC~Th!l?%Cgv)`kAzM1(9|YTzn^|@O3SWuGxyVsa1;WTdwS(vTLr<FTsl2
zN<#7>0#@jEzf_+wlpw+wtbYq7d&>~Vk3S|z^=;#Kdmk`aJ>orEU`-^yz2Rm}Ggw%-
z=mneZ>gHm{SS~LpyEatZOf3PqH-G}67_u}MRjrN<^ZS+BSThp&ym2$R@#BIw?j{?|
zl-eo2nP3{VOmqkvqn~!4E`>xS7e8l`QcsG<xwam5Hd-O>D3Y_%=9kME8(o~G9gw)X
zdgMEO)%;^Yl4408>8#4tu8LQ7RZv{xj@m@TGa>%xlYSZw_ZuV|5aQo_xRL{wDqZd7
zK|9-)=}EKr&mqX&>^=`4Ut?Q6t&3QB<b(|_a91EJ&%d33quWKq<%So99ZW4uccx43
zM9f})BJ+5>moN;P^3~~HxAzRn$DOiZ)}HwRL=mq>FoPO{84#Ch_K>}QFH(&?hkEMl
zWNt>!9fhroREtE=3cL@FTs*FhT)n+pxME4Zk~#0^Ztr{h__)Jtx*P;(ZRyg|*<zWu
zIGM+VWFN2XQd!wknHXwgS!Or~!8*Hm9a3+|$@f6e#)+G=mo+tcOxm`mRjgv<bd~(O
zD#5h6t;3btnq6&J3VJ(l{r<Nd`X+Vc2F<j<JAX_YfHk{fkqd;?5<3dl(WN!HXTAZ@
z${@9tNArNh@ZcFmvi|>3`WX@CSXTYMS{jDgS2R3)g-YsO;@R}hG*<w{!&?0~AY@l2
zdHy%L0vb*0eVF+(n$~F3PVAjD$16cswSs_w6N7Okd3)nGG46jnhGP7+$ra3PKE2Wo
zKill0`OF9o`>Cfno&pT$BuN^a^UN($o`HsbCAC9D8~D^_y??TElUAi~EJ<c^YkGIE
z`Y_RWxpwgozet*}Z6LieUHPTxy2);v(?voOUb_#{!rvg|+OMCynE0WMoyVXCC^><2
zH{G0ou>I$?Nz?-Iw&Q^!hG<R$IhTn2rU0BsMd%ri0r1M3dWdqqD8h7BlBo5@H89GO
z?5WeIwxd;X5?EG95F5c%)(f8@Ceh(R_aGrvi!H{nF^B}Kl3_2^qM1|quhRS3w12z+
zlT&0atSs2tdFHQQ!#=nlt@u6O6+Za1w6r>kPy;2k=L*EVaI&`6!%g3r7xrQs{q6Tz
zrUVvE13{jTd6@^FFim+hieqvG?JYmvT~@e!xDrg??2Eu%=?X|HeBAkP^Uvg`X2<Hz
zO{V2mS*E4ZV6UG|8s?cH$_I`|1_;M9hZ7i68Y)wY{^`*Cn!NvT;lmB2WK{!fVOGSA
z6Z+2}*<R*5<eCrp^UnF_WwZG8<%dTx%RzP4T<86^D%-4!lixA5`=(kHbr$$2QeF6@
zVR@qdN&XL)5D2h5gCA!L-4C~&Ma+5Qn|nzm#;4618ymw#GajvbvV|RvLGXDoL6}Wk
zsJiovNUvF38VMcOjL5ryfGADssw&0mBC(e8w;nbr4uwIV4{PbdV}EfI#Jhe{_be=y
zE_l6))(`f{FZ^T4RJ7_QD7U`Z8oV!FIhN-<|NX@AkXRopz7i0?8k@Ub8&P&XV`I}u
zEjCoHkWp*x=yzy!t8e((dp0Gf1Ab@mw4;=fk#P+d@9sU6Dm=TXO&X1jKclE)w*OPp
zj{=t+oRXvVBXPg@-7Lhx`G?i(HEJ!tmiFzA&Y-4;f24aOiQL6*UQ@Il<7SLv7ts#)
z&=0doNmK>-%|Q_RimU*u{APpVMhXZiXes2b3ljX|Bv~uX6}Gd0`&o#pS^d@exUv#Q
zMRgF@HbF`d^bwmnz<L?!5NPt$;&rF9jHx!MJG=3nu(T(2eLpjW7L2TPny}j;3)}s~
zHCS4Fo;YSmX}%d)U-A6dk~v(STVrJJ#SwF&(!gCZu3=H>GdfgsDf6$)5!H$};B1zF
zM0KrHQ{rLbej5x$Cak}0)K1DQ38S%j#Wm`-35KB}dZi8JoLb4BU0=Rb(=%gLNSB<5
z0vc#~G1oeYQRhi?gu-)x3vBY;m^G#6%?#V)Fysk&Q}=Yy4cui0VesTT_zyELo-k6>
zmf}SxT`CQ*WR)hP%EXa&5=hh+7%C^``*<_1>@k6ST4R_-UbP3^kt4gWocESqTS5-2
z|A6pA^w30$vuvQBIJLz0rBD`&rluhGz|27R-7$Nh{I%u@JYKuI-%*o$7vK>0DBsZ3
zCePyPak<6zDQ%;cX&j>lP+*h@t!|cRO7m5!n77vu;vVI1L9}QN*=7m!Y+Nsa25+5Z
z6S#0=hK8o5ki<`}C1Qdq6O;#cN4;T;3#h@@j{y@tlAi)(_Vy%QTwRG(Qh{2jp|fcK
zc0}Mo0&Q+5E=b~vX)2w5bpo?_4G8dXYU+eE5hvSf*E|7VRx`Cyr-8mcW|QyI<9+u+
z<8cV5rX{i)A|$$AUifc$osD1aH3q%J)!kVqPU|P5*sYK20czdzC7hAKmd#A63mQnE
z5&wG?g?B5#-*wGw74%Ro-3ci)c1{E`Vab^i8nk)2r_vSk$P#*2_oI82|4W{WeTK7*
z-XK_^+|*Vd^ZHZoI>7j#s|O<qACu#zEV8qDDl&72&402Rlmz`f1(Eg0F`u|IzT1*h
zz7b;hOkOWP2)aumJD|!}U}ttI65-sJIefd)Z+e9Ouxi)ScTpBKhAChIIx$3Dm&K?0
z%%T@2rJ<{qghGD@N#dh+#T)o75fxOxw_U&KO&OAleyM5wM==;Nd%n9Z5FkT%wj}}|
zxZ7=-9TjPlt_2*Sbo&S|0t?;h*E*fO&W3ZAzKjul5?V$^y-Fepl~sAH`pTMLOBvzZ
z-*AGWJ9gc%z8<3mFd(EAapIv&ByA>+=y7LsVy)hlnsG5QDk)#si8~%T*XSh18kM;m
z2Xa8K>w)@_7A6TRkmyG~Jt7Ur(65eGRn&5%$J-!CJ^*S0atu%O)k((fkSAGBPuIMw
z^V|MpvsdBQ{wkI|j1{YHjnLVWZ#)l{k(IRY45o@I<QD|`_{%t+TAq`1@t?h8i)Skx
zXH=Fey;(g~u5KUPpB`NS188h(t3615%Td+gUlwNg1`u7nzeWZ}8)ujoe#*)%L_}m-
z*?rF!n0JbBp_H`s@JJ~u^9w$iW%zipNb%>$`)19jd+K(GMc(}iM)zHs;^c@Qq5*OK
zA~eMr7g}VERaqz*7uR|9i1}@Q9=z^5aFHZiH0yM>v|o1?06C62k9tm>Kwb-%mRu_2
zvgJQYrG&g8M+dC7QjNa7Aq;z?85fwx<%H1kvQ(a&4}l$TN5`AkVtM$<Mb2beX-bIu
zxqp!2zXn<+GV|Q!gQH7FAIUU(ja1f8OT3X`ZyH@j>Vhw{em)M(gTWHx4nA@f^ZO_H
z$a`5&T&<9mr^tL;Vri}bA=khl3+$-OeXg@SLO>BMd^i|B59lCMQP3j`u?`L{sGWTM
zJ;*&_^@#7mg;GpuZT3i{fj~M#+aYB)o$W}jZ3@1JQMh)UmM}?YRx7b~yDLRzZ>^1>
z*aH?u^W0sYL;&{JbEew6sQP$1{@M6(9*KB1KU(i2L*MtQToPML9QGask<xUI5MAx@
zjeHAf&l%z*eM0RR($t?xoXW2Ca_Jul80+k))R;pCQf8PJK4aI^QL(e%Znq^$ygKoi
zaZV(jD)8YEWf7eWISk<V$FI76Q|b8*r6Xuq$CuDf7xDkOoF-~58!Jki5HZ`C3#Hs3
zS-{9xfVN82#>0Pr1Ln5JXwb7{D}jY>R37fpcm6|(<KRc<6$mlXNW#Q=XdK6xN@<XQ
zV$Jf}1I}1F03?CnX&~e;#qB&>fe~FZJwqn9ybIK*3L$brN@rcO;`9N-0%rc0Pw_L<
zwXEDT3iy9{?e&=QSY8`sWFm+a8!jF5FD}&XPayY66O&UXx$8Fs6tKc=z{}?v4~-&Z
zIIb06e+wWwcUMTgj9pM^%2feq#?*v(u*!H^5;(bv4`vSuO|uG1c(l++oWoX4TBWSL
zywV-`WX8wqehvvwSrpGBp^YCiTtFL;yX~hQ_qIXLDY@b9Tazc2NjR#m+;NHgpf6ft
zL1)qW;b|8#IP=k3K4)aZr5-vtsZ(b_<^G*n+qN@`M%>K<an=FUV?|zl@JoVKHwp*5
z+<ltP3uvJ}xkTaWb@&aaY<)lUsyRT8kriehd>_`Ws!Xhief#hxwsA@dNdOM|RLC>c
z=KjdseYOl4xrs9w=EiGy>W;Dc>}6(_H@rxjRIz$g4-{9@{5KbM>neZifiZFy+%W1E
zG4W$iW1u{kdaSRgR8b?L8N)fOYe&0X3!1HO#748Pw1<$&lL{yR%%xVhAVo_&kmn;w
z_n8z#y`(+FP}njB3_TT$U$c>FIZ^iApkf%FqYD;r6}jGPUE7JEu#$hN(b_S|@HG6C
z?pCMMEobfEkO|n>gQjX32rO*RS^@yL<F0m2`g7p}$osS=b<)FyPdA0okMAO;WIHA=
zjWc<=W3D1^A|%Mkmn?qV;}x#{(2%7)UNY$(O|Py)Gl;JkXrdcll5tf<mX^s7MCWEa
zpQmrGeL_uM%EW(Gm=<xZSbF4%!ZY&z!(6`dRo@rf$zPiHBV>L}^7GdCmszEX2bK22
z>pGBv<{llnr~bX$3Z9aX!O*TY?=R0>)8u@9V1IdU{w`m5@E~C<8?7JezW<HELD2h4
zw;hTn2?hbi<MFwb@ZD3|RtTNd^77~xl!|QRe5S~?&fgdNp@W>gy+F6B#M3pq{C=sZ
z0Rv+;{Xs87f8x}5X9Xi;>7@nMDJ5@gXP$2xPVZ|K#;v0qPI?J66==Eg8ki(nqd9Sv
zs;3z<Ul@UHTqoQGST&fN^hHwHho}I>Ccb|u*7zlUWo5nN3j>L%?mM5SM9!JO^Jz85
zGDk7eJpT5}Y0OU@9TQXN$r%n31Ppo1d1W&5sGC#-LZ>`R8X6thsk>~emWAr*N!I&#
z{a@`sBR9QQPm%9fcGFF|hk*MQn7Afe2nmkDBs@Bsj$){9T6qUL*xqO2e)k%GyC(~i
za5JoHAqjH*T=wQ%5<nKHUK{V^NqHu&S*O?5($#!kvsSm~T{f37t7X@J+0r^c(#YH?
z;^nrURQCn}cPR9@Sy9>Z8mnK&yax?eq~&ySj#h=L`!6|#3BQbUIr0WCFBFOYf!6=p
z-0Y97O~A~PgAf~%xg4viW|J~2yZ>5TT(WU;!nVVK!{=i@y<%}nX62eOWOV{Mx@^E}
zbCpxr;js}n5=8aw(`@im?~dNFF&x4pIr(qN9Ucx9acQwtcm+lL-S;?C7SiK&>y~O<
zDiy-R5oZ)mB1=s3+rBONU3j<CS6}vfnMjWtIvRfvZ4zF&$Qu7+9M&iq%~lQTEgC0$
zy4Pv<L=xCLMVFvOZNnW}lQmOW@#T|useoTD*Ucvr@m)5lHTLex#P4_m1<=!u+!*%1
z)@bs*kyNXmV~iE;SAfCbV(_%i`=xzSvYLPins0YoNImqsGSmWo>l4}6#2oGO6>|(F
zAIr!6XE=hY_@uX1<J}Jy6X$tfY;+YN>{i0fyh)0B(bTJ-3gQXH4Yd1z%h+e5mtPJ-
zs~LV{MN3>(9EMf~<PP;2cqKgB_|`f78?qpZ`7B>fna;dbOL#CeSYhiklfp$csFx+D
zIS_(!n8Q}#S#UfHR}9`V&Hp$i(Mj5j39L`FZ3`@LXpiY<;~ewC)puDDM0j6AWEPMr
zbf{(8h^-oqSjAI2JG_1y?HZYc8v;Gr6nk`dN6ui}pR3zaz54Ahnlml~rZlE`%@?mT
zC=fBt11~VR&>n(^mX_Fu{2Mzr9ORy`%~Wb+JEz*u92$%b$Rt!mW9K>cf6?I7l2%nI
zJ0%biPz;lmedT;viU;N@W-YGLqzC@{s{86qMkeg^tneWoI8cxIpLV-<06B|gN|Ssq
z^Clm1I;FmS`SbC`o$TwH{5u#i*-bLT<3>Q>dG8$_#&d2v4Ql+Fuz>3Q0ZCiepbFS~
zQu#Kb7Y{ZB+~-hFlqP`kTRNQHzamEE%kPS=#WMoKEvrMwL~>m*`a@lGb*a?AZ+ZA&
zvq%E2gbGg0<4BRj_REu~n(wt|2e@?~FWNJS^tD~$uT+hz=e7!;;vl1U$gW91EYPl#
zDqZhZs-EkU)UqTS(0G-jYD#K43&0KwaLO895BL{~RJS^sEsm-zZHL6#+SQE^CN`L{
z#oZs7Jw$Mrc=v-V-<x3aeRP!<sk7C3>-6%|EE)dy%9mX))YQr(UtJb#&nD|>0f`I1
zL;2gpb(;Dic-!36`R|4QN~d2OuuHd0H#>I$D<x~JzT?YgUWrhoiKSnR_<G<M<ijN4
z!pa(JqMf%7W=-9P`2J^Q4SNw3-y4@#d$X!8qvaH62(d>3^>wczK6kU2>{N;!qd6aA
z^GR3T_a`P)%yg^0KWK85JzAPw8LKQ9MbkYi85`a#=%dh9oKj;9Ty5Qw4}0f2Kwo{w
zU?)K5(;bya`i}5I1CdPJA+}ULX?sLa?ZVKZ<~pvn5vlqNl5<Eyv3=F*{3C(o*}srb
zj#bqH3m)C;WV@?22e$i7yA9SUm#3)N(Z3=2$;v5Jh5pY_SOmoNv%@F2NXl4K2)sX~
z7^e*}rz>Jb88VR<)SAnViBg*)1=3V|;}Sf4(KP)_vD&TDJ({Qi{3=C+`mH^=OZi^T
zTR#<O%FI#IA41_8E@G@1)~|AZY-%gUaJMpgr24OZUWu=?dE|BqIdU}^1853Ij3_QH
zZ!s`~%+}^>|Lq<@I&@(|pA7xbJFNKIFSBG;;x)fmTS{6Lqk8^>t0<AcSTZK}qoE;;
zF3U{qicz5U9?uqjY4P(@Oh>C6g62c#DAdGK_0JN4D{AlIpau@lJkQa$kw}ZXjpuHG
zAD`Rt!X;Z;{h7=oNlLkTo!hDmmWUc7He@U{(z4>!*a1eXuKN~eTnm?{Ns6rF%`2n1
zvwV3hbbTP!`r03y$cK;XiQi}wBU(_^#HyiLvHG=2^JwLf)`1Q%=(IAw6R{=e&2k(K
zGZ7EyDb^TRLN2^u1u{u9Ij!rDUR_jJUjWf7s*@Aw_bLbnO5N8a96US|!V^OA9xkqj
z`WcZn@BUm0*IayB!cPlD6n2c<c#h{uK>NnV(8<ZCJdd{Qi#a7y{b)qZ;!R*B=Nz-c
zgqh$F_fDvcsP=i9imH));BNDL`^m|$^q)GR_3Ui`U7$8!HV3wBWEhI)nWu}J8?iV^
zCoa{yS4NG?%WN9V06L)oPH!^fy?tQvvm_dfg;~(>vW??<VsU{jE&bvnWmRqh(1e~C
z-;#S>q}}c~HD(QW@j`c;pn)tY%)Z};NNn{2Zj%lU{%H#Dj;KU(V9ujfgZelx8OmPC
zeHrau{oOEH&spMp^+pcgp93-woXmOOFnjR%@^wq~xxl#<L4CCvh75!4@*w65Z99q4
zYVrMR54=5%-A|ScW~l?j^hrLtWP@3ixR!NHdzr>eH5#<ThC?L7g7AQvI>%D{(zF2$
zWhaYh6^?&RJj?=Ypl|VIr<<lr(P3|Lo4xR)G4+{P5^BNh&iI-V%)IsK!-nW<;L=k@
zUPW|_hPmh%QaW2WTj(7jeR3ufhs&<gD-kNL*s8Z4-y?`ZtD4BtIT8&TJLMf?62m0*
zs-C^=Y9=dF7^dET*@<5EX(U*XM*!B?^!uGNFrxV8iyV`N!H-qC8r~bA`O4r!fKPVe
zKI86!ziUzyIV%6XY`NP9AF9W-#g&~dpEVl?pUgGu=)rl62Bz=1ygXqxvh;bM*XifO
z&zXePlaskvT}6Cvc1@CjL2C{H4w*I?vo36$Ux^f_)k#Z*Eu^6XZa(rURGlJZ>4TtO
zLd^RVh8{7M{w~N*dO}-+6qcLKb-|>{bYxUI;GLDu6RwwWS{5Cb#K`VD=bC0p#GlI_
z=NK|upuQ!6@+}aP6Z?)mEMgFfcl#!?KVE1mMN(07u&6L$oKOl|r*6GE?5&AKyMMZg
z<t;aScp!mYMUP13q-|eKD1X&X&Rpc=UxLEbxihF?KA$Jhs-0zUm{Y3%4@#js@d^Du
z#!XZmG1}uV48et~`S^TqenzL|Xm=zCMw(a!rQ3*Yv^(%aKko5B?uVaZ#weSNv5;+n
zf0^s1x=~SotTKfu6s8L!faf>Zm_fGm8_pJ}e15gJJr?0x!1Ge?qez)Lf%c7ojF6pi
zAZ-gQfT8ci#f{!$kZZR-+b`08w_zM&Rxt_En2@{-cU_w@k(2Su<CA&6DUEMxhze6F
zV6%y?Dg0FlcO}WtUn%Z?O(OPpOnh|ZLx0v31n1#|&2C_n${v2O<&X!OJbEqkK6ps<
z%UVf^vzojcE9rvY2Ns#_udw$>M;zXVf`si%Hi`P(A{Hj<*W0AS97=<<ZNo|!hqBDR
za0xY?GRz&r9s&kKXFaln?T&;WV64woUuZcrObtso`#Ny=Y4tHM+(0~knrL`nts_tp
zi)WuTM~TjjancC5O_<R2L&@k(hLy1w>7<8DTUr7f={7?QRk09G;s#B+y~|BUSk6kc
zU7$n{m^Gb?%N45B{N(%eQc7qz96Ou>4;s&!!1+?k(zNEnI%p(_hO09K6ubY6zu3F}
zxQ54zGd5P;DPGV8iN{r|f~BzJ=zg|`AD))i?T%r`9Sj7mU0mJ%M+Sdx{!B;<sH!<B
zIu0S)@r#5Lo2j8-fV8p^E7v=uk9$BQtzRwaTV6jLu`Tmci06~!r;axVf3oMkzV8dd
zbcoPMFKH;4tmR$vyL$88qBB}BGJu8#@SoyDYE4=4?fg7qAQw}Q(YTrXCnwE1*XOYR
zJC}zvegv_0bTJv}0USBA*~{2IqGlRti>|5kYW|rV^BMyVoYLf#4$sV|#|p@p`!9re
z@R?NLcJBHHEPwgjza#K;wBr8su(dbf$0~i7B}1hWB^`p!0rvfPA|lQglfkkBw)al_
zV5C`*J4PSi7Ux=$64jh(YGf@CK9D;SAvX+Nh$7@tCUObj1}96R2uDuicXw4XGEE#s
z%++5cDEy!_&qDL^0PmUF{qlH~Dx#>Ke)|f+qd0Pl$O1!bRbN~rVV(w#+#&CCfc1X$
z_*ah2jrIP31CT$EXY-H;6X~maPeXf!>h$FDScH{tMy|LlVkZ4%=XKv10D+xs^gajj
za~i2x07(pk5#XMA0*N}wGuQtSV>7d@t--h^GhEY`N|Fj?m!dFeD~@!6bldF4l;I0x
z=L5mhP4_*5gD;>kuc$J~OZ@3c<x*Ny!nIXeLp<@jD7_{kCNRRCyk3<1gz;nNb(+N4
zlTGisN*jvSR=`T~lQ_4fZ|G9Rj?>In9nRT>-hL3-7-v)57Na}OC{|~@%bIE{`+lJk
z9q+&Wci#O>AwvcMgQw>m?j%h*&$YCoT{AnL6{Rge8E7vOzC%>CMOZG*N?C9Wc4L*?
z6p3zw5oAp}T*5lxS#G^rpXy$)48hEzjF}hUe{WS;B8HR2Z2h}Ji#hf5h31`R-9-Oh
zMEW8cnq}S`hl&Phg<@IOEB-6GW%2z@Zm$A18yhQWE?uVk^~x@%wtj>VnmQBy$+}}M
z#FXQ0KZ>L=<#Tf>Dtj9x>6Vz^U|*u<TDUuwzCB3V&oUej5A`^-s{Y(1^7wPZ{*QQd
zwH+gciu@VJq0iqlLuU+q(WIoFlOxPGWxEiI)h2B>f4zb)##=|?;-d(WC)2|p&TgI8
z75_HT>!MmA)>l_ntQ`V!$4$PI{KYa)2X=<R|In0qtL*ksR}|i<%fZ6N8`GV1oSH16
zP*dlcD@sz;7*=1}?dLpLQ{d!s{8Z4^ERDE(c^{1y%WL%uRN`d(s0})Cxu3A=x}T9F
z_*0hcV^izhst{cp>|RLRSTRvr#Mf1)`j&=CM*i#Lr6}pEt#N6oc?F^+ib28dt2>B%
z$J_>@`&vTo=C2Odcq&Cf_gG2Pm8oT(-S@dZIEpf67!wTbvW7?eZ#l8Op2sH8)gQ76
zM5~f2xhX;|&4CxU^(;8_eK*EyOEmKVWyT}{&86`VKRCKp>0}?yB_VeYKpyN-ru1Cd
zV*qH*al_I5U>>-8QKxTAj^IsDjbe3kKO4e-1lWbp<_k4nB%ow&1A50!(&MpJeSI5=
zt%io^rr*&pa^Zltc9>@TB1z6R$SxtcPUpifKuSHGhn!^IuQA;70U<#Eh!~p+z7!UN
z!!>3x=4F3I5=Gv$lRHcmIs^qJY`V*X(;3k+`K)%~2%qbn+1xxSpTt$mOx2P}hr(4=
z=Rsi#{>E}P2V>YN^6~H~X0Ysqq-R|7$MvKqisqO>4e{K75Wmx(gaPaY`IE07xA>nz
z3!e@Uc(-As@Kdt4siUCLNtvN4&jU8KJOR%H(9V4~zr~4)AcRBmgCbX>6RW4B?V1A?
zm^kC;%`pSh>YaKw&UZ{%wHQ7e6;W`ZucGZaw$zpfpWMG}9jGn3Zrg)pMQl-G9vV?G
z1}L470OJd{3AzaSH{!aU-bEL78}XBf6O;|p(5vrEzt2?0YP=I^U#qgHXxDo;KAOiY
z_4Xk@OTDJ!9%OiXJ}_|eg*jxU|2gheH>Y(JMQ__$1h?9I({FQ3eJo|gJTE)X@IJqL
z`TaTR%NkJ-0@zJ90G*)7Spx+*6CX72_xYt6516r(RaEu4X=;8>6aib33}UZI&ak2)
zL)Bgq?4v)}d{!e;A)Zu?CHnPOL{C%Su*I9iy_ZQjo}4(wN0xH_r2+{D@ID(9Oh)>J
z>Q~TI%R6Hexcude&10PgVTJ|Ey&WRremhl9Uds~i`NuR2vz8fXO$KPvo;Ba`!9nNA
z-M7_G+F}zxtDAL|D#|3pbidK#74lJZz78@m&i`{H!V^K|KRmR=^4JX&tpJHBlNLKN
zP7+l|l`8NSjo#dWN<9?~L1&~F5l{BpB^K-1)N>7z?Hb(6&ta$tTEKPR9czak)YOIf
zG;O1?F(xv<N(rl6ShH|o;Wu41t!QHUbvoj-q=J)X?HQ2E%r6*W@8FfbI_WiV#y|hu
zRF{kj$bD?Stm}s1^8f6)y_H-u`_Zv%Xfz1BBmg8G!$_@6zLZ@m{cn<K*oNtL0wVy#
z?@69XLF`*qRm2(Uzic+x?m2GJe$SHxr$6aCn>rn$S;w-2MJ&V>ztb8Q;(!w~<&BGc
zp)h=syPaM?rGPvl4i3DBo%Wb{{hEHj`lGHEea}P6236v@+T1s}4`D)2Q-EHWk4Z?s
z8AOTHt7!|k5=)u6Cw%JfCai2nd$mO+Xc<#}m9lwvwH6fh^!MrE32<D8*@<8Qw-2&c
zM>0FGL#dit8CwSjnld&x4Yz@y_<FIh`0y|T1n}n+$?W^%tsYKIIBOf=df;J%53384
zz%!I*&oq=343S%%)>-MCDEZ!BI$Dv8>3eIdn4M2+nD_gThYw*~5xi6cA8VSMq<Xd5
zfzXGVkhq*&@2T+5s1qyUcb9@ax!2m!8w!xa66gET!h4FNwe~O{>y=r(u{{5;ffEro
zG0*1Yh$0aJH;uhpPXBT+2-7KQq((by<kJ}O+!<CJHW)(qd!&w&t0^iEZl)g}<yR|E
zy=<Ry?^VWZxaYwSVmIF8vd`j1#r)rg>|UW#KmT)!OqC>G8`yI&G=~1%1oJHK9EkBo
ztm8Tf1wB&m3WFzZhhGQMKlB8Li2Z#bJ!8*#8B)(4(yM$Rm>Zizbzc9TP&RBEc*ntv
z-RkU{UrD}FM{%M}p-8N1bgUeKrrt!5m$bnZ(Q^3_w0;G4fs7m+243;MfArmLB<oVs
zCW58`YNNBZ|8ulJY7<UDiljc}HkX!G$0{?wRyY;`CD?Mtr4)qVjG}^pOAVW!tFN9Y
zGQJle+o!bS;f_^M-vZ%mnz1iYs>HG;a-J4n?$j>Z(mGKa7?W?K>yd26_2fyIBQh_J
zr_~BKTt0gxszN^-5}W5rbzU6>bl#590O>jw>6Pnq(%@?!2!#;pfj8JR3o*7k4m=R2
zKCG9|(?tTNW@MA4fl`m(XAc6sbh|oBcQ2g!C6sdG3X}9+zc}}Rs>?E<DQ1U%#G1Eo
zX~;NEjo>|C=#wKKws=Jn$4&Up9G=b~SeuTtzIYxv@G0Ih5e#rZFZDJRdE~OshqED|
zGvzOmyxrU3dEYyP+%^CdyxpAu`**yN{c&kS7v~zqGal^~s>W1Sp3-s*Q$S1S<*W+b
z)FoES$v=EUo{8Xk^<pUXn1`3QIK6K<_QTH;jP5I)Cn}E>MaRJ*|6{DK<B1nUp#XO(
z5wORjx1_ASIG=7mgo*6#Z+1pPJC^%%x_kgmAQ@2jzYxmaN3Y+)S_R5?j`DL|T)U@+
zRMi9ag8tx|OS<@s{GY~p9R2Vl<^Yo+WtTzrd7cL6skOmtgmcKz6X)&-ulb+;%1t9L
zk`?Ql8r2=XN2n$9k^@|gAnn`Ev;9`5mK?0V$FP|+Y+$;;rvN342h{zrgna>;nfFb<
zG`G4~V@N05$wXEFe;5~$smE+9Fd;Yubmt2!G_D>M?|$_94ms7M+xdMqDc-GAH}4Q+
z7ltT+IN1^}%axhioveMeeWTWsVwWy-k-gpI1Olr3^@k&$o8|5YrU&`zuw$&dHjmA#
zo$(>93Jm$D9+Xu%W9%CnSD9a`VYR5@ST1zvmg?RE44Z);L&~I@r3Vq$Dj|+K)@x&S
zux$1eR+QhbTgMHAEkwga{aY*Q3XKhLPz}WK|EUgl^XN+7@|ORkX|`*+A{T}c*CU69
z7=IyABsA(*!-!U>t(PMvs7nIdtSIYo-F^uhvTkQ!c8CO;h7@xuV9`iBcH!bBd2rTV
zRyu?~s})^a@Zgcg!kC7mMMb**>;cpK4xJ3I!g&v-<C0F5emR1x1ziLXbcHe%U5%Dy
z-m%6aPCOMSAP~i=sjHb{-@o+CJb6G+ue7oXCr&{tfh0A3*WTOvhkUJVYG}K6Zt;jN
zZ^9V?8#-Xr_Au3Rvi-RbqV~|No+kL59f7CKX+}DT@9iTLj^-@m_YTx?ipSq|Oubm&
z)4?0t`PS2urk3;*nu^oS;?(-q?J&cvqZqONv}4P$FEgQJBIlSio)Fc7GTmRI$f3K9
zuND}|yLoiWiwOtqP4G1P&sT1Ex71FG!Wzeu<eI*>>Af^csjNU6<@dki{iN^EIlAO>
z?6F7_UCkI09A}=eRi=FHFP@00lRwks51l)E2{eJh+ma@hzW7{Z4T=BTP;$2&-Tk0p
zK>8X692mLMMYof}$JwW$2O9Cw@mOI5Z2HiPD^3U$h?O*x`mRJFgJQ>CV^%!+z59;E
zGletHv(B5^55UE8iRe7O_XHWFQvJE~=>;!Mimj!AP>{fc<`2Pl%RuSpS5rI`kXP3h
z{16m;-3obKw_Ix{)Gjl}#h72jE%RVGbh`vN!oM7w+8;-GBmF+tK0LC0#`?xR-alvb
z_^<9U9>^?UOvzrMP8!p^VDri<2nD$^?mT88ZmfGCkK0cp|N8GXWB^%eV#;TvG5#?B
z4J==({}M*<H2!odmtwCZ=}8yMUHU3@&9(}Vb+yRUl-1J=)ypd5(HXXM6QLs;&MywA
zGBU}1kKOT|!zjKsM|R(icLQNGfXcH226&9`R-0s5S!^I5FOi^tBk9ZnzsRbQdib<U
z%P&cS9;1*-&B(=&h?|?G$9m?e)1HB`XCQ_(NYBua<W%=Dt=!G1PMh~3wFfPfNaeG_
zH*SpnO{a2bw$Nl8J%`SA9x4Asg=tqe=lxaFSP=EDbXSWn37%s7M*+&?{SA1}RWF4v
z%Icr>`pGAoZlwdJXUNhbyNoKN$4*S|>cHSn3_X=MM<P&AEg$Wi%)c%V^ZQ3{2}Rnr
zXc+-vZirC807&{f;XK~`re&R#NL%h8d{WYBo&CH%bcp6mhAN*QyJqK)tX8+fEJ*8%
zM_!)xi}5DSw@;4lzV0ZDTrX&udu-yr)^JCH)m6&U7!yTmjj?pLWSVcM(BZEz%KLs;
zKc9E3Xy%^$EnyVYl_l9}SM!hVEgQU76FNGyCpJV~no;WXsyaIrAptfHS9z`hogSc8
zN<lD{QH83e!d+}Ja6aX;8)EB-P4v~qJ~JanAQt{&l%pYI>=InX>iHbk<5qq#y!|BU
z=_?)9`;6;(0J#cw%D;x@W~CffAP543g?tFw7B>Y;jvwujzhUxgYJs^Y>!mXVTd|`>
zaAg-F>wo^=$7<NDUUp|jMqF;{2u9!=Rb$ZnCUB3++uIv}12jXLGa@1YMv;U6WjK)F
zNq1>k#=GsXE9XoU^~@L3*Z5RAjoRsDs4^OW>{4qKnBuR4j8*s`g<I?U;70_y0XP8)
zz}J#cvd6PW_Y*7?7FVMku8~>VdGJ}cciY5EFpfGy@c&Y`EzJO`btXD^;p5-JaQXR^
zcWySHo4|V6senffCQmccD9N6297r2X7BSk%wsB~sP(1u?*`(QX2<*}2SXGh~#GCI{
z0ZExw57Z<;*RJc0d@J|IJ~m_6n~hZ;$mq-|Di8)%#EKD(Jgmk4ZuOKr7gY#tZ-|`{
z)T<f6fqpPYC`Xf04s)sKHFPYOjL-qza$<Ay?Rr4qVQsFMzT_ds|GxL(Mbyd<k8BFd
zK%le=PU!G)^F&<oTKZj7=XG53s3-Vf7-ZZJ7CwltL69;Q`Cxk+51Z2Hy2VyJJMPoY
z9*Gf8VDs_m)i|~S8IV^sFB)G=uhMNr^!$64rc%l|7FvaKRW33yB@VU(TemKk)r68T
zyOlCw`3S!CIwYPn9)rWhI=X_RjU*@=XO{=tQne`fPR6S1&Uyoo$Bq}oR*-EP0DhTe
zI0uSQh++xxnfnD0l;-PN3(3#7v?eL*Gkk&d?|%AueC2y6Q52Ni?p1X5HJQD?$xmbw
zzZM}dG@ak1Ky%K@5HeY;N)IiF<XnU<3H4x%@U0nxrr+^CcGtgrO!A!{isQgylvOGq
zZ_DK)O*HX3+W$da6i()-7EVNb$1fc}oipfWgofc(>Sjn<wJ2Dh`^-;bczHTVuWq;o
z;j!X#_q$<sm?Ntdye@cgrxAHk8M3;Vj-ElDsWC%l$Ls9ue0~?yq;F>o^G(V`twi$!
z6Ah;p3#nH&uq*8ssnQU9EKP1pxvC;-rf{J%H60>T`8-qA;(>D`KVDqA_)2{i%{fw?
z&alwlKI~RJmUEnhD|A*t__dP=Nhw*HJ(Z~}zpi3Ydhz;y03Si%zTIUMYTJ8HlEm}|
zy|JgPF^)JA^!okUFg!diDI~tmvf&lAy3UiP4K_BrBuUI*n4@HbRtZ^=Q<Wx62tuF}
zS*txEFH7>WL<-5<Z@<TP|LSk>-jO5;S)S9GTc9jTyz@BWDT<1__g85&8e{u1heJ4l
zGX`(NzR8M_#hs@pikf{1xp#w|lEsa_69Nb!Wo-RSgb;Y=g9^3I1MID#EK97lG}|3^
z2Lpcg>-X>o&Yn9*Wh%<58iU`&#UZs@S<<p8y>5^8{5*4=IX2eUxc<qf{FC4LU6N=H
zCnDOHFY@lYZ_{ix0LRwW2A#QivSG&7_BN(6tgNgsKR3s&55b4yShBR#LCN4)v9r@>
zI4na}jJ%|B6$^76n#~5@IjnbVZ*8FzeEiX`G1jxT_K5Yh2W)RYB!36i*H?M(-M9JH
zH@{7!Q?6gX&JVu-H>_{0;(bM47!o0wn{Uw{<OK7spvaH3W_f9z^DAc=4sxEo`aHKj
zyF*z78`I!>UHi!jQG4uadi?=A+dJI8d!P5OeZn&re#MK=Jx4wmV9JuJ7}D<#czFLV
zAH4qoKl$-b!%M6+%gal|aSXzdrU^P;rmPAoV?&N-W6ARhtrARGVH>c$(<3V@^32q!
z@?x|ScxP!eo5V>7-Ll&okR%al8p5UQZuikqN>v$p{T^u}K}Cw)ZkMf{9lU|TpwDi%
zPZULDSx%l8<ozLoexJK{Z`0e|<;JHUk~UKEvSfQ}8zF;vgL9rTw+sh;^1^_SK{YJ}
zjWi7&#kOXhYu0dXu0x*ZJgJN()ak`y-*-pMvIo>IUnT&2%JL)&ra16n|G}9I({X)u
z*a|nTzdLDJ^?3qN-U0?hN$`2R4I;y9$piI&rRV?k*nDDEwWGEj+Gk%dIG+X9MxSSm
z<pan3en5BZu$?M&8k|l)s_DN+ju>v@f@wfBYR{pC`~#8G0I*pM;9=Ly0?8+pQvthw
z^DqB^uYdhpL{UsxR{Z2AKVfHkhq?I<*)ZeNPj7MW(b`z;XRJY4MI%jvvw^n=8R5O7
zEDcCStJxsZQOFu#EwYZo<q2wERY$QG&eKktC?$xb9(&<>AhVVvP02D7@_^RBB#vX!
zBt}cgAj<;)s#TQ1-pgTXd%e-XQd(l1LDebDycZbjD2g&H$n{oDYfV{I2uWZQ@n|4P
zli)02g1#-Nixp+fCP=BV&Y_jYny_d&JkraM>Z-CB4+udwA7YGUUMakHXeH|{j=+1#
zpqC+~L`o1|5Jj3uE3^(?l|qlaD?<nwNmQqS6DXxoN|MGg*10;pR8q4)4V(?2Fo~1B
zfhIx>L05%E_40Gi^W#7NE1Xn7V4#%NXxMWAP!OpI>rCC(P?l9-ldNZJYlpx7tH0pl
z<qN$2`qx=rI)k;AhpP|R+S+90?0G)FeV5)~5LEQukCWP0Rav(?qAChRtSGCJ<;8h~
z?~&ypZJ0n&*^=QPV>rk#l|cZjtE&hl@d9?cJv!|s&32Rf_c!QuGj4x=hadj!_gG)u
zU~OXqYYkOtS=-n|N{^5VB^8}k6Kla3hclK=qe(Mv@Y&7VEG^HY^#*vs^70CPa|dG#
zN+~=Z;|$JNvcUkU1X`zLgOaS=roTHR)d`PQAF{JE<XgY-8&qY<&wu<QKK=9taimd=
zh&YaEx0>}k5lk8KqF|8q$%_J8$1AihY=F{4V+u#tgdomB2s)h>D@z^Pos`|~5RU`r
zh+{>&lj3pAbvis+U1xWvN0hcORfV?}Z$d!V;UK3fLw;vO9k#`Vc#c~eTjYa`+js7w
zwPdap;#pc#Qk6Mzq%f7k+W?QH@N`-&@B_RTSg)u|6=G8g5Gpt&HBya~l4i3>tJON?
z^mkDF_`f6MuvzvWlAiK})qJ7r<TMI8M1^-E)%WzOf10|Fy2()kgauI%tjg9vtl>pL
zrq0o`zl-zrx-<da_O*Lzb>COz3EKRm&*3Q6?*Q{W%^ZoaE}DsRXW^-R<sU`=O`LyN
zTOk0qk@LYMD4q<JvhNyyu*xr7ju|_%K=kzQ)5xr6_=PieH*VZub@dTTOUvwTZv}Oe
z7C2K;RV7Iy1s_zY!JbRwokK?&TUEqyM61;Z-XK!aYBhQNwO6_P%rlggq2KG_y$fJP
zXdo<7#u}Wnfn|#{!H8Qk7-kg10;v>IYx2V2k>q(!nskUGO_HiQV}(JTZPILo>;u+1
z=I7_J2y7KXU{qyJtJR`31$fA_iYSWcbmpkcHd2-dC5faYiWSzD#PtI02&`g}XINt?
zD?=Q`R9K`4Az+*XOUT?1$1#J!FyzS<VbKjXfq|#4sw%_^BmmK>s_KPs9A3BvHzRN<
zz^T9xSQlOx6PTf@D)81|jlnBfX8>UgjEyx!wf$CE2IAZKI_K|%JE#YWEXP!icB@5a
zex9NVj{m{qIke5Wz&09FA)F^}#0-ZS{oxQHWt|lx{8UxN?aw~r)@PqmW`=fqfg3kI
zWw+a-C@odx>Gg*g8=P)TRZ*4(C3Q&I;yguG(1;Xin&54YR+2`VG8~jh1ns#N-Q8U>
zS#j~=Gqjs6(m2N05}^c98l&T2vo{<HUViDTT)6lGQc89<x2ejUmBj@rW686EPHTZU
zj(P3%mx<yC?*Js$Sd4L)q8K|lR8`6CJCAtt&2RDgSH6xhRe07~;jAOeGrY6lETu6R
z<N5VFACQ+Nue|y-(ngE3X9Ju4;Rhe$oF{G0@!D%&#acsA8l;j$afGv;&u@QD<wCFl
zDI5#)Ere7^Az9zt#W{HH*(-c{;|7%pcA!dWwzs!PV$b|y!r8OS>~=GZR9wCCJdI`(
zt%6B|H6HU4C>4Q+NJpg2X2`r!2NzG%q-JbGMnwq-DcJ1|`0n5QO?ZCKwE4=buOLMT
zj4cpjHdD%=_HQ;CAwMUw-jId3mW<|znn|`c{GH`F|MMUJ=_#kcgK`p}ap0Jkeg4eJ
zz5Ze*oA)KITF2h!(?)*B-P<gh=ZmP<@2|=OR!D-<ZwTB4F$6In=VdC_+6OF->$krz
zQt1FFn*pRpzun}(oTUyt5!yaj{-J8#Biee{doxSEs39^_PVgX4`iks1zU{wGOcgNQ
zACu?FQ{3z+ZR{c*K3GSqm^4ip^!t>hWjM?*)`f*!1RJ|ZN9af)grss6QiQZ#NfNQW
zwaendBFoFm4Eh6Jdhtc(=jQM@w3b9#qw82+La7iyZ*+%J?!91jd^hhsNX1}S@b0_s
zQyGu3BQM-A@weJ7vMj?Gi&mPd3JlUKI@qB(?|5+cHl23LZ~o?YD2kF{R)P%KEh4Rn
zBE@{C&D>m@Vb%w4LyR})*xcO28q3D|I$IlC;3UR5jIj)|9OL}h2vUd;+pest@Iuvm
zw8JwP4k_{itt9QVg|(J^n2kLyk+RN98&rN@|Jql1@bCe~I${+OE6L(~CwQkSi9pb9
zG&z4}nU6pIBurv8s2bTRO7?>7+S*utd1;B^U{L4m4c#w1<XMLIjwsQ%2|p$80~mF+
z<BRtJRF1qHwbE3s#9N2+!S1UnEAnARPgqK0`SsiHa{t}~%CaOYON3IC#!*!PFjiFv
z+#BM~#TXmH-^59SqBJy2ur<?KQx^RYco`oW=LLCQ;=QEZnJ3b5z2ovA1!vC8v$e5F
zHmr!E21`rLz@C-?pnKtoq@vSm<81I2Eru0YR)UbVGm57y3ruCgI7MJfOI|wG);4K1
zhoC~8-urPN>fpEokxr0UnrVUrvf%(*6?H#4e3s*D69ejgjp`6K#yYIEL~%qcH9|U?
zjfhBVAgKMfHh07ODZtlE8Xn5hP#H(RUosq6c6J`&MT8KJwT*3(ct9L!5I$@Qq(&-*
zRvIN0ZgZ0)3E4WTvLcG(Fb?XO%+>1rZof~SmvtP;AvZs}j}jiCL@?11A&@k+sX$=W
z!;I+aeuA*!fM8!|t*bXpjyx|;H=P|LGuO5wR`XGq%YAYxba=AMPICM-GG0s$QJc=Z
zVEUYC5OE05JyuS5(G%VOB-2uV$h=}f^bzdh#SM@dhzd*Nz|BnC*Bt;S)0E!dgN!^W
z$1=spO|p#pOy#G3@4K%NAU>k=XMwK60N*Tj@VN4(+x4Y^yjfD;*mS)!+cD!N_kp|m
zoLNlY82X%|X{Q2q{b9~-zt2VB{Dq6GKD<w)G&{SyI4?LOm$`fYVes;F7OfN7os^;+
z5=AkEB-o1iz_^_~v&;t{e8_+JU;miJ<z<!@=J}_8@cZb77^_RY@Z$x=Bftk=8X+a<
zzz&TV!a1$^+E-ua(b^`T-u#?yFK2Oig~E75;gPCABz>@t6AoXNq;ZI|=cVV_=U-rL
z^*$SG>&$oNSXx-b2*q%avoybmjwG?x^twaRH09jc6)v1zq}@uf*0Hs@$y;x|g{dk&
zzw>~~d9)5G&qAGbl<GRxpY<-Nozy5jjiks!S|%lBZEvWks)|OVL0MJ^DM{l5Q&c2T
z#PY&Cs}JtsoWTgl3hG#F7L5ciaHZkWg$rzNtkWB2SPRZt;z;A2BI%^;?(Wo}EdYij
zZL+)DBZ)O-UZ6ZUV;E)woUo`w;iUs#5keaaiY!NI6;!4Y%A&#!ef@kwh6@Kc>&fyE
z#26iu8BbL?cDlii(E-K*-m$&CjdQjRBcehO=&Zt+!1hTcD2kFe)?|gLRn2hj{8f_1
zfYsFp*h-*eL>$H3zI_|-VR3N@QwPOXGN`&eg2lx}Hr6(AQeYi;TTx^^5FxBje|Nyb
z!V<5(`ZBw_U9!^P0t!JmhqQ`VNTMi0Yt5k7W4_a1YvU2s{2c3BTkLeZNTs-N;Uck)
z5VZxdmI?{H^un_wad_@UX^7HhV1;WhW$#^W*eu4tM+m5@oXb}(vAw&+&dv@i%V#;W
zyvXkEF6(QXq-ldk>l=LV;l~)`n46oYC@YMqKq!!!&puyeI4nsU4T_?`x>cIZCNIDA
zLLE*BoHuw3-seDVBkRGsFb+nh6&{E1KB#%DWou)d<;4Yr6buJ_Zr<EPN1C+RMycQx
z?Y+l%TZ3O1?S@A0IWB<W(T2|Z(dLe^2k#dpCjl#uqkfr`FB8x`Re9R%-Z51@-2w$A
z7W_lJyp31)@xo(pWPbSg6+Y&5Ptui>VCgW%?6|$gq$+XxnpwwYo^%3v8uvL37@y|Y
zI7}^i*om_M*-7h|+*nO_qle>j>Q2e2fL&+4O=qsjtFOGw=bznRXKS0KrA01Yy1@PW
z4=9QOX%bYd#u`kKlPoomXuNfFnkmj&8fWI|%*C{tg88`w=S%KAxXYb;ce(l5=Ul#U
zmVf)N{yB|ivrgV1@ggkvh+x1P7!t?s&JI#4tgTQ=Gwcsp>?C~c)fc#Z_Z~NH-Uz9k
z=H~18SOFm$8yqn*WKabuf%g^48-#`Xw{PMQq^%CdLe<-5m=9T8m}hNmjZUY<otuzE
zu&~ggSp%<pxI;667p|Ve5&&m^kg+aCDbZBzKp|;H5yn*@0}B)Bm{=*C3+gNDt6(E0
z1dT?6!7wc3jU+)LXf{%mbfj^J<E6AFNn?a4aCwDuRS5TCE4+2Yu_h@6g?C6CN`~+#
z?RJA~*bfYn2n?dY(rm;ekwVq#*s3b1Ooj7~B#nt8fs_uV0zlQ$AtQyXJ)gZ6C<W46
z5(U!B+9AcDa2Qd0I1^Y?C4#qfmggi%5|~>d5lMonQ?dyu1L#GPBx#797&RF5`}DgR
z)(ec?B2Cs&Qig=}LZFaDN|R?HfUa{pcqxO5Ua1fp(L3_N0Ou^m37iM51%v($QU%s4
zA7oV4@XYW$JG(n<@Aj$c3?IYcfL^bM){0K6MW@rD$Og>Mb=c|l`27AocDsEBSzZU(
zF7QAcYm|zZYqjytveapDW_g~X$XMU%5v4<{3&16gVmy}0S&}4}K2&)@n#MR=fiX0b
zlz90JN@(&y$=3QVw?4Z|t22jp!7Ce}-ygA>fENt<c@2(2N<BO{69SS~&f^Uzt;2D^
z*a8(rA<wXN7*hskCjp=k2+FELM=`={v`A<s4XiJT=3A77rYIdMitsp8q$r9CsT4{n
z`u$;G@ua|5kE{Wmsp??Tb(RX}1DltZPY>8lX{u&v^G?R(9di9Fc}#NeROcLeo;bld
z#~l>MpK_Yw9{0RcNxiZbNW54>$op7P0)h&UspsvZK!`Z-;mPDl@$oDbpAf;@^e~ku
zOo6&X*yjT*(~L5vJ8KrBb)2U6F*Lyk+A>||<H&Sd546kAZtGF6$WdVNIF|FIaw=d~
z47X@B8@&C_&*=Ah#8J$>hke#JHARtAR5{N+cNT9P)>y1HXyI5`TBwaq6=jhjBLP%w
zZrtPS*$zTz%F42_vBO}n&30GtgCG43*>HfAl6Jd;bC$de4hU(S;(TCnupS-9IB(Hf
zV=F_NB<LtXNX6BQO&kI(i`uzB2gd~uRTW{3MT&?dj?lV8USuSZMqXLOI){p5vaF!Y
z3(9gAXAKK;j;qfu(`uwB;X*E4A#q;QY@g+o*Dj9(%z9vFdp9sTBB+*xg#I9-kv7pf
zs0)n^eKzu{3@n930K+UNZ8kC1G3;lx>Q*w>YSHdAIeV^+^FdXqltxDiXH8&DtjAQA
zD_1VEuoxCvV;y-`)|oK^o7jK$3`J=|h#FC+j;lIo5z$FocpQ0_jhT~nvq`&=@cawU
z&>Lh_Wffu*)vSwFb?Q0qiQ~Ao4-BATWRK^ZCyGMgXQ@1{%8=gDNHkh0=G${fA5?Qr
z1QnchKB&gy2x}X7Z&6B-L~~T82u?Xd(M(f}t&m>OYzA;zl_f<nq}$tJr@I4Ek!2aq
zL!=cFPf=vS*2|09I6uXCLmVeKXYtnXXnl=$-g_Tw1*>c8crOCrwGQt*N(-`VKyT2;
zTg%4Q78_gJ<YgV&rv_>FA3g}GZV|GObUO2-QA9J2==KIYc=(WqYa4;tl);8CilgA=
ztrTg}Ad;FQ&xsOEq$H`5w3{8GMk6>8)qq$^iSUjnQiv*em})6%M<a`r5@X6bz_iEJ
zKz9`OCeKTRsEDFC06u3a$|5-2;PEyD9FOCOs<gGEjSu#&jb@kteej%j-i7hgZ18BK
zOJ0;1TM<VI!#oEOoRbEF45c(u$q>Jiz|d#IEQ~278Koa{&M|xnaqNJDJ8-d|sW>qC
zPR-&^lR0=A4BnyVoCeH2Nj*<k$I}LM0>&%lNX6w*Ah&UH3x@$527@DK6kPxBOHQ+>
z&pLM>7@JVP&Qfd6QUyMa%)0)ljnZNAL?HLD&N!tU2iT?=wpr(#MD09X=98$IPX+AS
zO^uE%UO;C)MmT{~5dx3Yiut8wBCT=8)rQR>6_a%?*r;iZbt%>b8>u9U0$9m%R?aN2
zwA4UKjrSIza~f@l_ZBAy2no$dU@HT@q}fanN(L2Stw!_SLL%@!$5B${!TCW+8I0C*
z$IypGNXIeTyE`<}gd~a?^ec3o;C2YLYO6pfg|Welb#X}o8A?0u-M<zXq(C**ETTdw
zfxy)%)EqjB!eXyNW(X+-NgCswV=&kviZn_E7R5V{@Df2>=j073y^5m5i$RF<hec@}
zSPEh34|YfshjAW32pSwGF{UblbAq*`iNu;s<~jmvBOH#R@U&V@;wU1|a?YHM$%_iD
zqu}gfL)tazC9ag%p&`$5@B)coabbZpPWZ;xUZpTO&RK@ToU~Cp$KVk)8;bXV6>}lJ
zU0_QKA|2PM-2#RvhB-<q%F19XgLj_6V2BWcv=KARa|U@qSvrJr=w=A0D+MespCOI~
zN(KfhIIxuDMR2@{RLE-3&ss<wGZ<u8=a_HDSnKN)YXY@6N0#LXAW2fnvZN>rysK(Z
zEAW)uzki!xyA}Y)$X3zeu@<_$f?j_&gs7=QtbjlWhps)7U1hL7*q<2@yhEFf21V8#
z2iz@<!__e&J%ki>QjP<2r%`RsCZL(7tju?4rW)rIMP865DN0!u<{PAG%v@)g;b2H*
z9PL&cA$(B53xRWvZf{s`c7nIKtkuM2S)#Q@>j22DHDl)&Tid#pWr@;~kki>y76sX0
z5Ljx$21ZDMH^KJOdx5haFM{{1w;pRkS+;VNMTK)AShkL$ad;C`ncB{daS(0qeHfol
zDC5cW5cSHe^S=-#=@>bdQJYrX9Mj0tsEtl~{%Oj6T4fz}-DJD>f$_SZqz5m%7`hl?
z5LgUgd=F2BK%e;jPtwjWLTh~-6EtGtCcit3u{)~#SwQAQD$81qgUGY`W_m+)SRJQr
z%aL*vO?$}cU!<I-{8IrtDT9hm09yLmORcu{kvc4E-V#N%omCzF!r20$!TR9kSy_X!
z0c<IykTNg;IszqZV6SSAOz$0~4FQDHBqd5CTvcIQ876uUUJAVRb+F=KKZgJ(Do`P|
z9$MC3o&vnY<7mw_Nn(u>&}cX7;K*UIkph%z)Uo*h@9Lnt9(<hwt<GwobOgeObX(Sj
z0E<co)mg0TT(Kb#a<kC@U=$KU$e?y@h|yvPqNWJqsDV^r5f_LMg`d!t@#H;D5f&c3
za4y)MB`Hy?vCfWN^y66L2sVEUofN4RUO<{Q!`1aJ&U*)52R6h)nF+S~S@6L1prvAY
zVF_;yY3zs-MdhK{>7b>qlO6|~u)(0Dk*44SC=ozuecd)gtb#qEcY;p41;S%}5|~&m
zsj9h<5km>Aca%0b3K(lqN}}r=#O-z)p&UYvoC7=%Jg23AxlXHgS}}Ox0~1s`FIZFC
zYdJ?{EY^8+6k(0QR0iv6;3va#bk>I~4KiZ4*QK-uoI?;Cd#cKY@IAG0za7OD^fef%
z*)1tm2nuW*&Uz|isH{PHhs2XMW8xYZj@YWv7#Ffuce@f-=SVFI%XYuQxFJ%4R+{<w
zd4x91Ei|Y~6Y{xwOH~zEACtxr3yn6FwMe1Sk*3vYQ<jFoAfwyQ$nqX{?gTs8DAJT=
z$PnXk;n~%jD<K4FKV_7budFJJDJiNFsX}1owaqX`3<d#EcX#_(3272j#xecwE|u}b
zNld@rud|C7(sZ=x3jorXlCmtvz)s2#TIgxWVNBdi5eFB;S*oYUkrTIJRy{MWdm13?
zNdV0kdVFftKkOMCC__-&FA1EeaD?=g;Dg;Aat{<8)rLb39LC<9hOs&kz7YGF$mz-t
zYyYID=1JOhR6F(^o5cp7Ku_uk&tsb5I*PrT^=$pD6AzK20P1mal51vd2u=e&AE#ZX
z^ok9PlvcI5d>uH{I*UTo#@%(e0u?})sj7WBNxgR@QB<d|QbbzOYBcaz>eRRaVAWw5
zoO2j&>-0n6_M<2)u-*qdD7+7LV1jYGbe-M8TN~oC)hWZIf~pL8Vk;Bc5K?PdRHOo%
zqQiAs%AlI6XINL;V~s*BD5<dcv8u>c6_E}=jv9>AidLk8Mk5YtH>t4R;e`x<K!ofW
z4quOIKUUF65p;J_Dx5KOh6ynS39|k-+If2Ki6Ry3_Pj&L&_C8Vv=(S7LwF7kf?9<P
zwVKn6Gk^fKXCqjvDDz=3viF`SlC&BzRT)xdRj#CP70v~fqLL%-3QALvCJo9w*k47l
zrnDj0a#2(@VDTssjP4O50Aq=?M(6;howY<d!dilg7pRJmZ6J<fHn(@i6Eb+bap7L-
zK277m4pyf2<n^`vn1`aQK!linu8v#itfMRo5H&;U1X4(fGDqM^;t)S>m}NNOQBpA9
znPWJ}gX2)7aNePHL{#TdMuyEmRait6fVPe`LIideBC?_&QZaE7*UvZDB#xrOdlzCR
z#&JSX<U~3mi8RiISdK-R;fxEJUj#TZyl_N`Lx{i}Uu-X7OfYSzs*FM3l4T`jVXzhk
zc}89o<XJJ&#)32yrl80K&ee8#1g8x#u_Tn)ppN$)mP7A7r7>ghHOvNh=P}0OZP?H`
z;hAf-!3);c*YG05RLruxW{xdom4^+PA9=$DukOhyb2M(haQULP<|m;f9;5mv1ym#O
zbJB|TE#fm)wd2YXlOIm70v-iu4+Ys1<q82-2nsA#RAKXrBc}+w-3a{EApSI~@~NKH
zAt#T(&S?sr31DXwJjvi4SH-bIz#o3j<R;~~a;Cj*r_Y}Rpwquc&&N-le^^;F+j(63
zPI%sAc_)0ek1gk<aw=daq(DhoGgra0Qfo~T$7rRfOo*u^rK*96!22kG3|yVXK!p6I
zIxu`n2Jkb=XKKA8QZ?|40{vYWgQ%~Iq(ll&8byKa@WBI9)%At#mB5r0kx~eu5I#63
zc!Kww@*d$GsnR&_(JBgRHWvcMDiy#fqK>guhjpl|#gD?4)WKklH^gWZf+$MI0MY|U
zi75lHl}aLoL^w~Ks3Wvl>N;i|f&IW?oB<IQd~ZXhQs)e=W-F8w!HH1@Lv>$+zftgH
zBSIDmYMUzRMOAo-HGVu${K()7IR=yvV(bC6x=Uhx@Q8I)P^&i6-~{8GWpQZ^<0_nz
z^<o=TwcdJ=;AOBCOdB!EOS}?9Hpbatw>DzRl!{PNg{&A3loU8;sB@SB3RkP@6d~WP
zi5<S}N#Yo#M`1ETh7whe1z%eYYblYaunAE*IN~@To+rKzvlCHPre<Yp)u{>~Tnb%B
z({wm%h_oWoft7VNcvd0=EDmg)Ax@GyWQPPFK&GgB-&%+8!QK)fsj4Du7NQWwWn^C&
z=~15Fy>K$Jw!y4QgOUoZL;14y4yWE^dF!yavDdBlA*P}fs`gY3_g|JKq?Qv3TUp97
z0Lh}L)%m4G$T~z%QBl=dUTSCHpoUgKMO@@1rSZY58BvEi31h7)D=J$C)ps4iI%okv
z2#>GXW$%N1Utq<<PslJ)XHszxK&7a|>QLX`(+74>4a13<A5UHBVTVrp{h`MWYv(bI
zo&_W(KOa}tq3u8A12}0v9E=|(F@VM5a99i$hsS_$^?bR%u#+@x5~QAXQ6B*a!9lQd
zSY0#oicURKKk>&1FrEagoMhY{i$AA2;Q-J&Dd<0<=*f2YiGzMd@iSS>!zw(aZKoNc
z(~oD?<2vlZ)6{b+U{{)Q%w8z1@j_A*6>%)^Ca^<+t*MQ#MMx@doUMb7O0<+A42BIk
zDT4ybBhU!p(REl8;|#v?Xd#J29aL8Zb>}G6mC_nvYx}L5tr1dT>hL1Cki5x_g2E~p
z$aSg00x0&pYeyLzj4>gXs}gtxcoA~id5;$o=RHxK4Ff#U$V<0o-#~;#NUEA~s_lxr
zM+r$tG%Z3rHwQ{6@FK8H&WFWYfD@K9)<ktU7Gn&N)WnfSIIuN)hzP*L5?GoMV3bmV
zHDDa24FP>UK}86n&aW$j%2H6P_F@F4R+|bC5@$W8GULF-!b7Xs!g`N!!P7PtB7oB<
z#BLLHA5=B~avi5Zp<#mutaTRWVx+1a2ebw+gH56e&#kHhH&ZYCzP8gFshx#)A$Y9=
zrS{kwU#kF>Qs^kEK~bGI*99ijivXY{fju=Qj8&y0jJH@*r+%Y0HxRN;cpt`&@FBRc
zR07pVG0sv|1}$X`C<V@IthYg}T1S@@a>Qywq-99`<~`n8;y4O`o{C5jK(Tj7s&SBL
zNpQfaqkhn=0XW`=bb11W4h*dG;DkftQBerz;%o?iBkJrWN(Vd0Bvv$Sgt3M+4yFRs
zCJAAdtn=E|?6`3@u*4#~FV2SdOlei;Z4EY`5ZYyo#W+V*8LY8&Dm^i_m$TL(2$>|D
zZBteiQJj!vjM=%M&aO8jwa9T!>9Fuxf1)Y!$>b!U<T23TQv;AEDDPB&<O{0ji9y6;
zo^u$$bV5-Q6OuFfS|RK4wQU_71cQ@*9tS8-#PZB4&>vm?%=*M$eY4((1AX9+2l>-7
zzMlzTIDN@+WzOuAQ-J1KZ5n@`YS#g<H+}q&a`t^U)mJB(ZDs+<S%04x;GIxT=M`I)
zVTUb1R8@%nFGR?aAi@G$4g-sT2w+9jDTb7iD5bC}<b!okdyP6mRxcZv2j{4)t8LwE
zt%|cy+FIpl0^8@omm@WA9f($fbR*yrY`vtA2w8ibjxf6nDlX@INOR@G0&Y!UG?dge
z*b584a3PLfqynJRDgc;0k34~lLTWH2BzaK~#UX&}NbhNF@K6;(Vk#RLE#W{*Dr@SP
zU9~4|4YtZEs6K^~IB$_Qs4WFGcyJ!DNU^u@x+;WjF_lFS03T3$JsNO6fNkr7lZ%ju
z;4=~c%}B{s+m!`V*8qM4d+0-8+9W~65!O_A0&uS(J-${Y$C1J~i}%4(6+rmVUa7^{
z1J_y;Vz|k0ox@^%ux}GJL#%4n*V>Q^S1XCJ9)-kD0Cl89M^OOic}`TT{jKwX?XAa<
zbv0;Hst%eL)PhP$P#Q5(Gx{)wMnHJ<23D@Z+0jcL@(wGZ>a=(ws4%U?Nm+;g@U=ap
z!g_<K$B!`<;RUX&KzO{80qhc3s=`#nNjlC=EJOg#!g{PXd+Ga(#ftFW1Rxj2oAp8U
ztaV5?=N&j7+TooE>Ut9ZV0eapJjP5_fEQ?`Yu2pJbL+vF5`3^9jN=fTSmN+fpd(c~
z3zd}0<7{9<v~J?9!<cZLu7RMn;0QTlFJ7D@&kKaCv!l4+JkswCYP-l$Y9~MTvIc>*
z6@y{HFMf5J8SRwvG!fBb6!SQ&+>z&=f;l@)U5{JFV_hc>`E$l9c^E(o9~Bu(sG742
zGpoiD2Lqw{8b85u?JM)7Z9nG3Y3GKSDB=Vt;8er#kUl)Bis}0~w!TxMmM4CVM(vuR
z9-aoj)2iE9-8=qnR>Katw@KjoIG}AB=bwhXJ{7RjN{1k!_<F%rwfZ#JxK+-ebqqpc
zePD;A2#bL!E2LDx3slyNhk#MGIIR@M2irYsT>vw-!bFxl&vD+NR4|@yBneunu`QDc
zx>1A*wr{3(V1TgDMM@%7a3lyJ66&lJ1l5_2LVPv625)f=N@p=9+?S~;lDGz{zE0sT
z_kf-EV4SHJ^#EF{8k9*9LQ0I-1)L4GW;&)U3My+b){J9!1-mwfR5}Ftwbfq!OdkMR
zxDOS;o)CBwz?_mAmBd(61vaI&F$;0pf@iOj!7;@c8*H$=qsTKnf#r;n5DrGwjGdGU
zrNi$K1_XBc1EQo!qnVD|;e|t?!#U0c&{scWS!W9oqK1vmV_gV`LmfxW*BgeacK#3|
z0BLJ&oiEr^l{pBB@B(9kV$*s@UQ`rCt;$rjI<#i$0^SWq(B5mLmsnd-Rw3P3>U>~B
z<DizeRYg@+SX0qzG=qA$c2Kdk^NmL!wW?LAfqf0fLtx#)7?LPP*7ku_eGdVWNn~xe
zsdR|NC!lt;Fbu1LNJU`X7|1yr@+?O>sO_B#%<;(969kbK<FF<Ui|{Tmu0m7Qp=GGY
zkE|z(L@B%zxF`U3WN@r-KG-N~rKt)_Q5DEKGlaDvTuD^Fi$XYbZKo-aR5qw;1CXtQ
z3?oCZY!g_=5eFjuXv1Lf1OO+HC<)e;W2X#&Rx=77)b)E7?!}MQ&jN@}X)%1POfRy>
zobv@#`{dO;0RwW{`kw?8p6b{spUX@ZY4W%Sgu~LI-rCeBXkdsC1;RI}x70_9nAD%b
zer9N@O~<lnM;$l@V4PO%dy*=Tk|VUg`@b0V#}q&wpEJ!Ko&dxid9Px|i8Ia_S2Mwo
z9tXw_1ADVb*U`DNfaa{GPG0voayqZrTIFS}MeCqKtSzOXws*0v49)>{OI59o^>q$e
zTgSGr*4H;^YwtjZaMWN>LLGk#0V)c+CMjg?jp)Z=BDB&WS7b<-Iu5W3A#;G1lG0kJ
znGJ=es;Z#ots~@zkPaa$gi;buy-<aOm&iy{m<p6Y3qffOwz7K<L)6(LeC>gY2x+lg
z2zw%RP@%cnb}RC_29ge2R#@ke-i53S-qn%&ZLo*)Vw_FE5!~{%4vb95Od&w2pyCyQ
zX$x(ENQUPkMk-gr!fJelthb+|KGnLm#q&OR^$Jnj%z<};R;NuA$Jnx>+wF1l)@OYC
z!nasnxx~iCU4#hql@uY*XJC;`sNbu=QW{HH$8wvSo1@5scOwdnHvtH%kO@Kwfz)zr
z@8?{FR5}FL6$I~9>uRrBK{HLE4vSHiL4B!pZ96$uk2=b-!c+xu5+hI;YXi70?8JCb
z2;mtN8O<c6%<CY=-b3w`8jSkg80-d66~?cSf^3w-*M-<^)&=k<WQgwvkfv$P4LFRm
z;02M6@!o~N+*%UrB&apLr`>MiyeBU!2Hh@6oQxSeDK$zfj0>sqVx?=7fnZNsKb)Xe
z_71Rn3y}B7T7^4$r@gBkK7<?R_oW73!D~9~AeGkG8dymQaimeBG<~)6jjKP0u@p9Y
zN=k~N0D=dy5W$9Z1WvtE;eEEDouf?@h_PyRl)f%7*ADkIGEa!To$*N+qNgE87Aa;N
zpTQU%dH&OA#}iz4*zp4k_$kVLoH}O$JZhFqC@Q{#;IHv)BM>!LcLyu~672t(;M_M=
z$OKSPjyh9P39<s;92Nf5!1|G3V;ampP2JuR0BIHgISQbT&p)J7jso)2^`8p-Orf=z
z;B6Xw&v@5PL!Ryfb?gD!lRU@iXLy?Na=JaPfH+PmO^J|TYQ_jVVjRMb+UTGf451b3
zWb+=PNCXv)3$gHY?d52l51xE7Oxj9@ptixWPK+0*ut-=4WJDb=E2t^)t`5JV!$Kt$
zQtNtg1sw(F4kaaCX{-<Rc@IK`n0o>O15<mdI)WXS7Xn#_Gzm@eC@-N_C5Dr0n?Dgi
zoKkYk*g5ZUwF=Fez;H?t0D`W6q?9N^|0pl0oT=><8D~d;Fwlys9J4Q4hn%g>IB+FG
zOX65#DpT9rN#Zm@MFMMrIxs?mqE4kH&^{RBE2SBCu6GC(;=~o+5XA}Y`3{5K9!|hV
zAAiikxn;inJKy2c8-lxcZy<H8+O@t0<T8YdDGj!QasWG91)E#jeEn-noL^jIH{YSO
zIo1WI1q&c0PCByU|Hs^)#mcg!`F+@LjS-Q#hcn!v?!8rARb4#~?3S8t6B5ZbK$?aq
z5wc)IfNjWp@Pqs!!v<^t2CN4UFa#U20n4Uk!;oPD0t85gA<8eb4ap`=so6ByUEN(>
z#j5J+8t-(*y?18DuqHozD{|-FJ9D4C&pnr7T%5ZzV_Xp#5$nJH-}n9hBeR4^iNfj_
z>j)`{wqGuo<r1p|n8qkta$Va>u|bsea?SCSYGp)vqewwYs09=x7<%y<wYADLQ^;+{
zRz*&+1yVVQnDk8JE2zJ#FsE8)LKK+~T|^wX6e+&3&O%XyDKe5!N<g+TQfQH4l1gx+
z$Y~NrGiBC;_yors1!R|8!^6T{Zb7V8OgR#=!dOMoka9$b%Yw5FSrz(WQy*(Zj+sKW
z@IwIlltdavDUB{gfZ<$)jRGa<2a>;+Ax<?1QdqQ@szgw&>ZU-V=FA~!2nqr9B}pD$
z&VmHh3_4ZW`bPI-pk#%$PIUJ49G(_-u<^~kUT1q>ar*$qEpOD<0yT$vh&$|WUg=I?
z>9#i9arwQL?VQ)W!Rlvh&z;Wk7Jwj`)W1S1KZ{ns1Axb()S8L^9Z7$HPJf<qeJ5&X
zXL_JDzs*Daf5z|$7yKnI(mzkmUj=1u0<XCV_W)q`0)zYRv(L1{%Wby1yzRF>fO=l@
zkb95655OJ*do!@wuC#CY^?B;r>gDEouL5`1+jT1#-FXjOVFT|8((eWAQc6r=tRYEx
zQHwaQ-iA?><Yx=uD<DiBC@r0UF7ka_O{Z(F6##k5*vtvrIcJ>F71Y&gRmxUIp&;`&
z<0zrblyRmOE@(we5)$>%5pt4Yl@`D;#R%F^Dj)^$jjy3VwPj|Y>eFdmJ8q3p$U0V6
z?K(vQlrB<JHH~N`E)t@RR>g`7TjnO$#B=KuF-@ZHW)0N5qEeZre(bBztR(SEF5H~l
zF5=METB&%XFpQIcS#2m)I93^j=<g|!qKTO)1UXKY%2Lro6A|6nSl;A9l}Y31_?UjR
zK>_QF3%>oWZ}G*ievzO4+As3qCqE%fE9tyLLSaa$pp|9pCqDl08PA`s={J(!_Pf)!
z_|;$f8JZ;+?{MA`!X(wpQW}YGJUL3<qV|o1Pp!duPuI4nBA%?m4hBG1?Ldz<b7gjW
z^(CN?P>ZS{J4~PyZQIu0S_**Xu>HJotdbovND*){hZ2nU#1siJ&^9eG3{{}}vZzCA
zRf}gdF=ld(lv)Hd%K%M5Sio_Z<lbrAw$jsD&-rR2d4GD=)>7yy_?)v8H9)BY6(AK-
z7LVeTGe#+LmVrS?iOn#ADl|<aEMaX63QCIGkd#Oj8PJepVi-qIx@t8BtdUM@8pS=L
z<Ww!-YW`?WP&F)N@MS8PmA9l)Vizb(T_mKi&4Cn#ujUdWd9eL3$^gYV86?(R+4!ca
zVz0e*Z+j8nnK5`%l+8U4@p){s-EJQUzQa!66!^RqG#^;!IzZU|DGHqzWc^=6mEQ#w
zC@2gbr`9O_Z<6xQL3(dz2L+`Elsd<&@A63hb&kv5LUTqT)2iQRY5pBvn16{r{Z-&0
z^3+~y!mCdHGca%aLWTRLyWHP<#vynA;oc;Bc>S<DU*=TnJMG|(;O@%&`r$5M?jVqR
z)y8k&THO=aWt6TVzY-;cT63zjmXNEY@)QgZkua6;5DHZ;8IQTT&D0KfX56f<@}@Ox
zS+YhuLrEg8nn>G3)?7niN`Y^j%u6*#2v7yQnH9qeuDTTDniDuDXV2`H05(mWMOU+g
zpLNxKGgUJ?#)7hjG)Cf(rJy7irZJ$ctKc)EC{iJq1SIH~(Ar9V*cio?BMF$INR@0#
zl1nwlNp#SR#oDT`Hc4lzpvp*y$a5S;e!@A6S95-uB20;##Dh~C$p<q=!rl~NN=qH`
z4C5e`&b31gF%VOtYZu#Sd%Ni9mJ9mT3a1T*%r}1j>-^R`{|vwI3x9?``o`a87&hdT
zdHLj=Kl$TN$WbbRb5`_ykSg0;`2Kp0GlqBHKH}u@lI}rE(ZU?%EaJDi=1is{5(KVk
zqzI-IVg=u07y-2{G-(@89LEY^4W%+l5^~mcK#)Lb(NfEWrfr0AuDN5@I0;{u9q<A5
zo)sZs3N@Kuia%nCGU&+(vZ8I9s*fn}$XSi8V%)-tQduTf!<<rjVy#6lM9(o-5IIeO
z5F*ZcDFUhYB)}9C<1p63AdyG0*46t-iX(EVg&GNR9@+F`ty-09452dMB7|+w<S2QQ
zS{q^1i^#8NtynA;q6wHX&N)ek$ceU-w1|)rG0kCkq)?+Ik&;1S=Ngn!b5V`1Y`^FM
zDrHD{BursD7*Un^rWQ&`9$;YBwG(0T*Lw-RvFSg<<G&7*b%Tk#)}ibLQ*W5P{48wZ
zz7<}V?NenIx&NmL_Gd8sQ3ZBKG<ZBeqSyZ}ar{@Hh=PN%&v{_~En2>b<0mLy5R?NR
zAZn8lt)6jO{u-E1>C3-{DxKWZ_1Jgm!>?cS`xJkF)?4xh2z?Vff;S5Et^m$!7xq=`
z*!&*4%XFPZ+YPDiW&3uHU#_pS3r2Uo?gh^WzTY<8Q^(F!tuZ;w%vdO>>`19P-9=cl
z>Ijjl*zZ>EK_tiOOPrdk<HU|ji6v9cnV}!?-hsEID4Ka-YW1Q;7fEMzR#({+$tx4t
z1Yzz_qTdA)!V!l6$$PTSW2#hw6kUk?LL37j6;jGHz9EcaG1WE=(<H|2F-5f2gz7G_
zXgdLBF=CV!Q}!YtRPv4-DFjN1B8_1TIZE!;X0xGf8dmF`m=cY1prw*=69<~Mp>e`w
zS|dKHlCuXXi5!XmmD!6_Iy<8cWJ3%YQ?yiLs=y|T%*HeYl-777GAg18Xt2tH(pW1U
zz1E6cBB{X1gH!s=ny&Q_BF{elgrEHILw@cTe}xy%e#Cdb_f5i-SX~T^n~1jyD1{Uf
zrD#G7tj2+1nke7<J}1jxWV3nB*N(nSEYr3kIOj|lL)Bc%7^8q7p1VT32?)<(1=q9<
zDMiVDtOC?pSF!7QPf8x9Fk5Zgp{ybT(-cHx*ht=2G!lxR#!++*y~F5B2HLn9%1xp_
zm}&u$^&W2>DJDt!*8(oH(6kKGBq7VnW?Sc&#-5ZT);P&St-5*w>ZHiTIajN@HC9m_
z7SxtGv|1{1CEu^q`>ckT>)@&=L`J6O1X}CJB&?B&-<+iq*=UE-NN6|4t+ULWPnb(W
zC8@lfj~Ao(Hm4n7Yo&zw6bm|GpQU=*Z5?IADos^69RbTt<LZmH)Ud9?TPfPgIWta?
zahj;cHd2|dCEOhI8|W?6l|KiU|D&4j0)}o~=RUyaz{1xh8($Y>y>k9N&S~HB!|N+l
zKBkm^7WlUGO;n3V6<O;4fR+Cx()5=os?fUsgd_F0C~QbXbQRbKc~zA9e^jO^e}h5&
z8m0V;wU4}8+xNEewvFB}GxoVn`_AD&+YbSs+jhnLy1o-~zWMdaX63*#b@lrez?Iv7
zhU@#k(f(_o5AJv#JJY>GvBfBis>KijkpN*kG|l#9oKiyNQkf;GI8-r9rwC9IaY{Jf
z5OX9<0i))ytVpOx=d9U`W7W?sRYF9rT*)HRnyY}Svl>*!IFHh$f;fe$6`fNETdFpZ
zQ9+XQDYGZ)RI6w$-dKilLZNAz4h>aaL9^&uQpq%4oDxz=<d9jdH#BXF@0@ss=FDas
zX}y<ft1J(BoJR1Dqve7^L2U}oI@(r}#kEpcCwXqoqQMv^E*LRUvL<J-s7_gcXUT=L
zi*t<CSnmY%rN9&uw%TCDSfsjCYlbP(c~9eY1<Ud)>dImb<0JyYH9ssSC818G@U6$U
z4XyX6lwd48`{;+HuYH+EkKW_><4^hY*~hHUMot&U5EbjqraD@PU0f_M#=<8r*ZjRd
z_%^L~n8x$|R~|9Np~?+pQp{WV5@S>?j1gVE7^QH+dN07fWQHM1o|=*KNI^V`t+8mO
z2^GX=YMvr&L!}C}wW!h+AgnhVv^Mxsg|(58-!P3dO-G6m@2rIOD~KA0sj_@(8`^Ds
zBSk~1kV2I3u~Niwl){P_X&Y7B5U}1=zgC1j4UusWuU6+QAw=4y5r#Y`(U(+~sfJ?1
zI!F;lN|M@OMPR#BG^C9%ohe6>HI(9*achWyO|0x38W}Q4aZpN;QlvUK43Vx|Fic{S
zo)hGxBsD{6sJXDFR_!VU&WVlKG)1{y3PLOlW0hnvmMI96n6tD%0Wnk%n6ro~m+D9p
zGcYsxg<%Nyd;1-l_Vg5Y4!M3#$G^j(*V^Tq14W;KvAYMLyw5r}+J3zkzV@-qbWHy9
zSo43w@dHeakbnghSh|0kC*~oB3l{p{Be0=n24l96ZL5-5JMjW4(3Zc>z+VL3dUfai
z8q@W#>n<326=8OxJ>}YZ`zWb<POm!0LyY@&_8r5lSAOC<FolQMw(DQ^fvoGGb3SI-
znMb+${Z&Bq`k@ta@7jCZB=?g$@fKw?V+e?zZpL|sLUh@*l2BfV6Gm&iHQPKR$<xuO
zD7xNCDG7xU$T<nKp{li3E?C#lwGAlg<mbFQYc1AjO3a{y6-kk;->6cW5X46oeR*L4
z$z4lj%Y<ywz_M#;ti@VGM?=v9Vzf3G(@MUY65!KMBU))To1Ug?#IQjtbk;PTr|UWa
z1}G_3$cDD@(y2xPlg1iOPfti$OzN$*93L%F+Ta>T3Z*h@z|4ZEh|(TYLzCKQR4Em-
z7?e>g7o8NeR3Y8gH6)X9)-zNGffy!GnlK88ZoI4cQ3AkYln|P+25Y?JXhrF?jaHKI
zuAo~qU|`Atn+@yp=RE!R2YBZ>Zch2V?|+@J5mqpc;+RpCFt-J%x<x6*44+(VXsl-O
z-S6?<`zN(3xsZ#(c?VfE`AXK?(NNpAq!3X;Su(^)q)JL<x+(y()>05)G~a*d8m_N-
zS^}0_Z5t%2Q%2M9;K3>5I5FGb0cd0J-jh<sxQ1LZVG8xWv^e9osRr5P%5=#(MJXxR
z$-;=GxxzPON*PHZ;hdKO3nOC3MTr-&YSfk?%Q=b2bC{(+3fdIBwxXG*)Yd5_)s%)f
zC1QwJD{OB}k!hMp$uNwP52_0^O~W+ySnDZ9p3z)&{^IP$fjF3~*XNQ-VKpH{Hfwn<
zy!WI4IYpF;+v@6+MbtWsk&>jt8-ol$v=MNeQj$Sb7QyY9Qk8p=-!z5l!Z8aY*Yu85
zvV?y><Mg=^9tW7f&)T8CE~9tfwp{Nd?*zII^p>x~0KV4peeJ!&j#PO-=2ubtXj=zc
zY8sAHAM>{PH?W*zIA`~ct6AoNrqnGJW-h%#^An8v2BrLYPzR(_4%~ou0crOE@2{M<
z@3;CaH|uMh*R3nxa)`UI(%W6^L>=z}vUWdQZ}V<QcXwZ9_n3Rwyt3!>mi<$g;Ld$s
zz%ErbL7=nR+HPqj8dr0ow{!MVldZA@C1=rBE0A))deOmh1?iZS<bjnec{e&|yfwrW
zx9nJn(kW=Gs=I<XlFkCEQdOIg+DJtx@;GW$iye%CQdx_lYe+4T3k+kbOqX<=xyp<<
zCm~M-IOj06;|?*C`%#k4OW9T@QpEB~8F?9`l0|1OD`JZH#&dkMq-`2v7J#d2Sau3i
zbvcOjdP~7ZE(xt;WoQaksjYm*5GJW8b_PYM{<<}Hu_RIqG|r$~M~uQGE*1*`4R!nw
zCdsFBu3;QT(9rr0TZN{n*Kode&Rat0@kmH*>U;k9Ti@cVKlh8gd~(iaHA2zC2-Tt#
z2`y_>(o8CCa~-Fg@Jh2-F7Vzm#-Scko&b><5uiJTNQs59pG1l#OX0{EMo^0N(BqA(
zR&-)XHw?XK5ys4HB3~%=evwBg1aU%1C4-Vse<=f0px8F;Htab+c8a+m!iu(CWt4I$
zwW?FXz#^R@A=!l7>uE|%(<H#Pz|mqsE|OlLiV_c7GTPZ%C9WhTL5XO0dwe8=E3L5F
zNnT&gXDd+(7epmQ1b_=v#<ZY{6wZ{aND10@QJGL`K@pVVNy!jmCWIv6=_J~T&RJ5J
z7>2;bg)p&ACsumvbuT%B+7KtPKa)X&+&3X+!W2oVh!)`tP>Yf_=d$(WoHY=073~%n
znB^^5*?U#1b8``w;MtwxGfl74d3{b)-kWI3K0tFHVD*{Sz2%snh55V7@$Pes-Of{9
zJBq|tQMkHBs`r{qj*ozh@Z*==vdFhfIJVTIY(?8ipe9uLV+w!%aNDjRh1(b3<vjnf
znO@_(Zr#@&fOMaIY=PXX1}nFqy&H?v;nmBv^|pG4d)icXKJTt#J}+Qbvb5Z~tEBT*
z5x!i4g`DJ3QA%T~xM?;Zo|#T^pENk@(2}l=Z#+|+u+G$ue2{%(B>b2~AXpp8!%0$Z
zTSr5ZKjn;<mwc4`wo)qyjWy&rp(_g!OAx6KQ}y>ySgmo5LzU{+S`@lShnZ5udV|%5
zVHg?5fkoHI4^WFWvQ|)3WoX1r!WhHRqQll4AzQ0BH=CZ63d1;Jya16g1iZC4UpsUw
z(h=GyrVyDzV7=MkT_ZP#n(f;%ZQF?@T8eVbloSs5dM%Bay=<lWF{jLW7=)dxKpd@j
z`|Y=c5v{*v42e977Memj<fidtEuh(;7*-oT-fVd5?RR<S-M4u3_yM1M^kiEMVw7Cl
zkRnbyI_v1e#DFj~-Y@v2U->0Y9-eUVbS*%+O4&#b*GP)B;<*vsv}u|chf&&Y<@z<I
z!CMh}HcByV)=)BSdrWZ(@1#&67XcJTi@n&K4;F${wtClyNU$o5d93%dFi%QpDZVHo
zuc4K~Yq@?gORm`%$68e_GCiica;OT7l@_f(ZL5m{1*Z*tm}+`~u!NdUz<f$jQq?D{
zK&dD~5JtC9Fe+EJP(wDX`;i<)1bG+-a)B_8pbX<Q%KZs6XOrwFi9?L41w^GN#z~gN
zl;k(Dx=OVaLdZ;GprlMW-_*)?LrUYN4=#ek#Kta*o?X?il(Ym@vns@{Pn)s@ZKW!z
zNTfX1R1CE(U@4L`c=>MP4YF<jnD&&f0!MdYS6-LRyYISpJ=SLec=uosZ#(u3>ZIbl
z-b>}n6grBU>hPh|zAabt$~w?s0Kz~$zZcx83hv~e4DC(z|H&0dq}y}kcH4H_dY=Qe
z{EsZ%;4bvoEz=Ch)Qy?<Z^mt#zv?%*!Pwo>j@^*<?taxB@8QDzz}h_pit7lb>UBY>
z3H)iB>@AZ>p@_a5s4-MpQA;6#6Jg=?dd<jycb*smK6u)N!+Xb^@ID0*ZcQ<XYeUUl
z%GC*=ij1H!45L(CrV6y?5YXxIQu516Atedn=(&P7gR&y1n@W<rt}G)otqckgl+_lz
zH8`Viw&D1w75R!HOhF=7>nOdoyejotQBn;BLW+T$6~id(oWd{-fu?mF9W7|Q#~G1A
znZ^lKHP^JYq!?K)mQX5N<RsT<ngTIJtkT=?VoaG71;99?QAUws5UsEfX}p)PoYjP!
zq%cD(rZ7s*Q<dU~QzlOqrG!l?fH4lqp^_p6DL%-2|BwF!*EM|U%kT5-=`+S5iglV%
z=u~Kp73M}|tP-*3QUZ&%P;}sYvy!`~x`RkHrO0tuWdT)lxD=Cfv~44@7gj=`GiXgo
z7(R`cv2w{cZ6L>LnI*0c7|YQ2SZ8Q`OGK)WMRG;P<6?Og#eE}7C9J9i9Vv^b^f*W%
zjj?9y29Z-Cq$p3TuhrC5mV~0P&ViQH0;3E$M5#zNn#E#))`qS!g|l=}Mm%(<F*3$P
zh>3n2L`p{+O5QLGfiZ|5YYb9pJ&dDd!d6C6@&%<$tIbr^_9aXzIjCh@6f##yYhxH=
zX38S9lVT>sRPFU<U*^j$FxT8Xg<Ow<!qjIW<x-`OW?e{m)=QD?Bzl}HgQ}{no2nh^
zfoZN_{+y=I0$%R4&s*50y8xokvi@!J?z-*4`PbNq`+~!}F2AwPEgS5+Rx@iS^Ax50
zEJ6J=mT!red%g2Yg={lhw+rd=_LHJ)89rV05frif7CHTM=)0cKZKreB?Qe_S&R;9g
zdR3M5eGh%>bjM>|pM=EO&z9Y3;{%}bb<Sz;bD=H|^~!axe|`nByUT!Jp6&(g;xyrO
z4Vf0Ps>;>rA=ld(6e&jmEvi}=Da|;AEu&OWFc(57P)6glFm0=KkI@6xIIOkeBA|qA
zp)wss6=IH(dsM-h))s}B&ZlX@I!h{|9VUL0FpLpv9ZE~5vtDmdDWi`UxJKHt=?7wn
zc<1p>QBsl$P$k`pDjBWBFIFk3V)WLLHDEMq*8Zxjn9^FRT&vDlE`<~er^m|*Je7FY
zYPoI#D#dzQdk^Sp4iB#JP;<pf5$i4Qys$x9WMg6q^#&07Kc`H;35;VS1uRBmtzjGn
zrWmkA0Ixz4+O3fg>a2lhjT2B<J#QsfN~3LuDg(+GiWV)pei~R$8??2!#)uE9GBu=`
z#ND7)Oj@JqyyoHKQ-0$&egU_D)gY29Ifq&}TIz3?DFn$?O!M6>!nE3msO}WUI`Xa#
z5Lxc8P_Owg4lJ5hfLe?aZr4iO6^by2#%Qs}D%70LjH}_8VF;4<rK{tGck+(N5|$sQ
z5v`!>TDdQ@B1I{BXtU&9Hcbl}VhWgAb&QdmKc!*Od2)t0MM6x(Y1&qs=B$V*(+^|S
zk*n8Ni<MrfEM87k(=As2VI>SJsOlO~w$310Nj_<c8Sk3yfJiFDrBKD%*)|Yy%qYU5
z8mC3*xK@wP;ZkJ;w1tqWe2Ed+nVM%CD*HDVuu65jk;hAGVc?6TV@$Dn6I&;pVG;Yb
z`At_o2OxK!m*>?DcRrW>^WOw`*|*-ELERl$ygTjdj_sLe+y?;P`8vH~#k(Fu?Q5Sp
zA@VQc`Pb2mb+9r6Ik_+64ocX?3w5}HMv<_T@;-sTOy-|QspD5R{IlLC*Ta6V+QC6)
z>W*OP4!8cRr#rHf>e}C1Ry}--*LcQjH*s@cJ0D;7-0w7Kxo4a21?+rhXj%`dpmTMJ
zFyinKVvsy60G091;hn1?)d0p}w4w11g|HCH*HEceTLIEi94ST7*HTIxD3I`s){2lK
zMvG@AF%Uz<C?lz=O7iTMM;+Q|mQ{;v97oPC&V_l2SwfvTvRJgVz5zt)B9rLahU^?h
z$)(X&wC~WOK~}Ql;87$D9g~F4vMzLsu2wS2KgslyFkUXAy_T}%A~~mP7)q|0swVR#
zgYtfxdMhF2f@>s<xEV%<G2nfJbCzjLY&ILLoxNsL)%%h>FY7!;i<HK=-mo4fmd8t0
z>lI4D$%7*StB6d@X1(EJb-~H;3CGJN5+WWMC$(klW=9K!!8nI@2H!PkW0|Ib58ith
zZNH7N1*J1F1d5TNyLE~$ec@f+P7_CuI^KEbgyq7K(kMW{=xYC_(58wkBl&$P)#~e7
z<*bz6x)kW@E1N5@j#VFU*7{p678oVfv)((o^=pVYrbNjib(2#PcLyVFu|^ol5(T8F
zA{al6le9sb%f$Us=!YKftbpN^YbaGJvQx}8_YTG?sO|ivYX*pWNw;W;LGH1rBhFaX
zn~A>fnX)1lVabLm(6tTTiS1ep0b>*%#V`&ijD%q0RP`bSq-Ch#T?rE#FaE5~*jnAI
zn7gWyXInXNMx;t+TftPlcvVJ(zmWlo6p6U4?HWoPFqAe}uWO~eXxqtsCm^*bq*$X0
zW_w=+i08pfhEis)XPK@fEWatBcTe{0(DrYH**LK8eP3S(bbS^!?oI6b;PE~KXuH1B
ztB$L#tviR*GCxn@AzGbRcJ9)5q9z1v|F5>YP-F~>_i*LU(U<=miJwPt3QAq(4{z7z
zRy6V^n555jQn%cLWoPDRc{KN(&>LbC_kpoP0PTjJ;PvIVH|^lGZy<EN5#{q_>?F)3
z9ke9TI|<>Gye$N<S|k}tNyJ#q(W`_<+gO~_IAc+0@oYtec5~IV)5?;|rC~Zqa{D%s
zz4JJ4X%Ue~y%33&f+8lp^?OYvOQmhj6lbwkqiMHnfaL6C&M(gKji=L|q9hN(S}9XE
z#?ZQk5CWTNkYxCJ(ZdvoF%lDmAZ!#mkYbc`Qb@SdXhKOMcB~EEvcc65o8+KXJvk+^
z6uBhU<3x^u<K;0C*sLeQ#lU8mc>eN&^Yab4K-YSFYdJYticX-)94%T*lAgE}$>A#r
zsftXQFinKmNXXY3I^Wcgs>W!G(}uR`@Kt!$8q3qoQ&wjar^gSgM<-}iXuB5cExz+8
zX9$yo|IW`=qzqYWLdt}kDJjz~9KZHY{0hJI7k`Vdf8*<naSbWfViASP5sKt&iWS+2
z^B!ldFjm#IB*uU$3g<jpDH`vILD*Wfgb<w<(|B<}sQH+PFR3tk0yuK1R%)ei+%2J$
ztyW>K4CTUma2eOM4AV%eT8SxTl$F85!kY>>^HzJ0YEmC!7Iv<PgN3&q@4aXi+Kw0|
zVosvF2US2w7^zJ^aK75mkAc1)2}$l>V=Y=+E>`PW^((;C7)Oke5GIizDIy?loF&Fs
z^AU9|?9jC^!^<<PaxFm6ki<zMMFDWS>hWzQXOi6KMJbHeSZ%7<ab*RyA*Gro5OcNL
z1Er*cj3HyR3=%=%8&8Ufv5bsS+)11>J1!sw9kXOCCE>-hVjJ2npBIDohSt3g=(&xN
z`Yi3a4Fv8x=B<Eemqpp>h^}`&ZxT4Za=TkLc-6H~JM&bfp$d*7C1TtE@-`GJ62x<v
zgxtmQZQ>y<f0>5=GV6T8B_&t|;dcazw+-0CO5O3){xh9!ZP#r#+-H4M?*aDiyPLaS
z>)Y<-`!r?mJzW3pV&nVV1NQ=U7psAoN)=2kI4fa75iX@#j3#ku(2{3RaztrUE8Ge(
z33H{jA*RaQ6{((+!qS@OR8rNYQ&{7~H3C43Sgk@yC84c_97^N7g=(D?W2EHFqHQSN
zR%;?fE+So_a*;4n)ghb4i6IUc<1k8#&!^U0AcY|p{hH-+iFehzv_7QDXi6%KAu)s%
zA%tpA6|lo3P8v$WXwHfJMoNK{V(s`MN;?to&54i`zHzjT=lS_LN*O+V@<gh1t>@|U
zv&vcsAc<)p=fv^Rg2vTK<Ra{nHX4(qO4fV1Jr*&doF&D8MUjReovSH~aiVdawwlY&
zStLhC$EYBTloZKiN>2D@iSy!g5VLrV=A!uB-}`;eS1Yu3Omh{aQT)QM{xbi}zwj^M
zkCyzhU-0C~$2@!b6LOpwHW!jxSOhf86{w+l-s;+8I3vtt3Q_FEj3Gw32V?b1jVUtA
z5`(ZmB^R=m&cAFE__GuQSYhHaQpGI}DrKv2sFdU<<|Mjxp;m-zYwG(tV{0lw&bsCx
zip0wt^4;2>Sw~alRSXK1#qGjbN2;RS5GsoxVbmj~=oX25dV<0*O%qcP4_Q@UvFIcu
zqP2v^i)ix2C~S60GVqEuZ6c*C3U!5AND+i7E14J*Ax6;~OlgOO7ac>2<Z$}hURC{C
zV-(%WS$-uDD@d%xN>SLoS=c*9VI*@F_lumUgND5ImQEFvE2V0+I@>}e$vKf{YuCx<
zGf$Uf<<)t&?fz8&Y+u679rDgzHGLK!xjXZ9n6(@1!47ltMwz=;t#HTv?AzCFd#|*;
zo{K6PMT?=-{7xBo%|I`2!?2fsN8V;a5&7Q|`H#T-42s{P@GHPkJ;GdVd#!Qhj=$UW
z`EMFDy<&PJAoF$Ut-I{=O--Nkb-YV^U-vTI3)mU0$SE?_5p;D{65Fb(S(jp!lkkn?
z803sba)E};DE44Qbkmx~lXJ#vDV9hniF8Goa86j7QW9O`ur=12GlWT)hXSG!RAnH@
zO1xHU1)n%C<f^BnYO#b^ixm@So2G`*GEo<-U(&dyYMP0ECPfo-;beKl7$#IPlCz}5
z*EMCSa2>}<90+1&(RCc3o>U99Bvp`E0PXqe0!m@I?3mJs^-|$0ot7|C!=@L1TokDU
z-hKCNoR;hR{yPt=)>nP|)Wr4}M0#ZoVTDm#4vLbnW(<jt#OHI<542s!ao1t3NLXwO
zU6hiLb#(_YS~Cs<eIMz(FrK*@+G|}UXKH1pwFc`P%jJ^i&(CDvC!Pn7PI>=}U*P9{
z{%d^v^cf#4PI>3uFZ1y2FQL@0Q)1wI-~Kv(^4&jVng&n^dzr-OT^j*j<RrqwIZLiw
zE^MkqhBlTVis0|M-|*<cgKd~kn5#r6vF0OEnZE>+JTf^KQVeumhtsBp|HO|}8$*t%
z=5v<nqgw$+mM}I&)m*qlm?pgS!epmd^%WsblCIFFUP9+qq+C)e41FL~*%Irl*p?yT
z`PNy?qN!SQhMZcde)f{TTC%XHG1cM*ZRH-As8z+_lp{=4JF*l~j!Z%9#!`yp97(ZM
z?MIE)23M<NQ3`DpsuX!1Qa~f@d@f3C^QJ<~Q)SMS_#hW~)~t8c_ce=#VNq2XNZdeF
z5owrW3PmMK7H`-so~o>)WEd2n0@x~g4b%dRH#qH#nAOhrSFC=gg}3eRl`PvHF!ox^
z+$)ydb)CK6?+6%Q3*a4^UcnUJ)`nd$x7&p(EU?(Fv-@vOvym-?-;XR9`68Nck@+DT
z^$i>in*S4ue*(pS2g*MM%IzHI(!pta+&u$}SDpVppmzui-1|v+1)KPprdRFo^9EP9
zP1oA?s_XgM`GvPk+q*>V8BE<W-3!<)7aihJFMg&#wX>=WM%C`xhF^`w)m}7LR<5Lo
zLgXD(Nm%1BRB$pzu>q?imlPtZN+6^VtF%L9CUOC-8AGZgdI`P87^_bwP#J(yB=I{;
z5oaAu>(N$;SZWxFIn#Dst;7^|NJ4E<Kv@cS*AeMzIJ%IP<QJ-vczAMzs_q6_DbNzK
z(pti5eHh3^Ts3l%a3IQHT|<o?qP*YQfEgn>V<8rbDo_&3uA#(KLx*1Uy(-wtr4Sue
z=T4X?Z6r&tC{5!$u_+AGM7QX8`yI($(ty_n=N;oT(6vt3tC~wVO%pj}j=BYT5lBG_
z8AMxb4p$WE#5FYsCGo-g@A6Y$`hd6IdWVb6hQ-klP3QUa$*07WSadBfUamNL^q4Py
z<!dNq`Q=}KK+`Pv_Ba23q!`6QPD`$usZ39b5>8GzG4wqJUB!h#>BKk<QuSLD{V<`a
zc6LS>(XQ(#N*KGGB`JTJhFVBqXxdgl1X4*mO%aAk{9P+EB<>5sT>7RFYpy8FmC+iD
zt-`#})DW-SvkH<trp`HHii|O^=sHwRG@T<v;R3W#w64L_bOmiSDML|;u?ijMoEg@l
z-<M(*;2t7lh@>J+;}nAYUa3fEeD>ig1+*ilOiCijAkskMp$n3xkScqYazRZ}5H!pA
zl=>_tF^8_K8I_fjWm-W<iD@I@+c{iXYFkRJ(iPTC*Zoy3K#)8`WI4Ll$roWpRNhHt
zuj{{YaCdFG(RuD;;;yg9;q~rwyt|&+U0J-_0NKIici7i!0l|IW_gvq*0PNfLy?6eV
z->lqNZV9z2R+d&39z2=^1R@y`UXQFW{0@qjh#Q3?Vo^+JzK$*b8jAlBioe8Fn7F5H
zhwi0Y&g(ww-RnuY<-XWXpBDhTPecC^GL`!d^(Njvhh7z*D`Qtu!Z=Uk9NtKUYD!5Q
z7?hEbb`ci|a>1xlof0I$JjJq=43QUjqOHl)B}bNRE0QMGh@h<zwS7%ZaHrtCkphEy
zx^2^vb72T0sv4~i!z9@cwKCOJVc(`Enir~;n@YlPDYdOKO;O}9n#QBFX6%JMD_V}f
z=_eAxD1~XFq=a>jFpVNBqQyE)LYC6GExV4GstJ0@^qZAfs5vVDxq9d3vSXPwiPe&T
z2USB?wbLt_ehQ$Z(+*QW8HFZ@&fYjMRv*}=ZA1npOQL++wHPD1Z8=5OL(lLs;HnEn
zDFy2_t@k3EGK$8TLF6#VGFsJ23-Gp;F-kQouEbB&DuXJ4cOIYe<oQd^UIzMM!?0TO
z@sECl^DV#n8@~xTf%7P}<SRe>PxAEX$2|SyM+m+bU?w2h(6-%{Azm&PRp!GoO-VwE
zN(=ZUGxiflM+?Ff2|3blHsn%RE*IO1%oK&0R7j}2BoV8g(j@?yVyKQEhLRJYMfA7k
zn~mcrfZT|^mvfFq*P&~HLFa`PresVb#RoA2C=TnqTt@=Vh`@B7tIg}ZQ;@v8lndh&
zM02jNWd~=K4`CP$C~1en)<P7KywO^da;f>M5Yix?wizjk(IWKA%$^mh!1Se?ha?E8
zt@rIbPt~v~RFURc&cqlDY76>AC-3qtrHnySSvy3yh@=Iy)tpF_xM+ZY)A_kVu@w{F
z>PWtcaO};#_VB82V7)gD-gZ}b9q{!g0Nr8WbkF18k%c=r-NJO;oWBF=)O@dJH2pQU
zP7tR88pMuqB3GTd6%c`l!Xw?Qr(no9k~%|`{}q}qqtwrf$oif(-M!7X&%bSYWB1zY
zHsbRDVV~!?Zr>nd|Dw;d>)m(7?(0llTlT8yUce4*$VIAtLN3)A0aPjEG~vA!YcW7M
z&#c{+qv&}#YiOLoTFqQvq?BT|w1NV)@Ie)YvIbvF)zm5&BcZ>!`q61ug@C1V$T3&Z
zSILUdM!ZmC3}RYd<N>tSR1eEpt4o+V??rf*l2KLWdC@LF%?=2Orft}afnk_1y3jh&
zQp-8fhX4w+H39}e^0ca9LEAQ>>((^V?v&B4g3}Z!%3-y_XiHIwwiPB+nHugk8kG~;
zfOCd^2y8BdacUPWF(-WMX_^M(>{b><M4T&g*-xYr$cQDGb1l|1Xrr0N0c{M{dSQaH
zB**8<Vo~P1O<r$CSN*b-$Z^D(P6`>4R3iW6$3GGPpmde2DD;~RR%^PZ!zqYkFGlXl
z<NPT<|I7ar&tLovR1&~MDwUCi)oLRijHOVJT*URTA!o3SWoZ@|JJC-gDQBipLZU@$
zd?z`PFVD~LZOiHL37ho>U&W!V&g2*<SuqSf(=@T@7Btpit>nZ^r4|7=PfVdYd{}%V
zuBy`*g#DU<R&^tpomt3I#1S)m`t&KsM@KA=7c{MCBI*!`Ib8;LG~)GYJd&<v=vvXs
z6I(IyOASdtWG=$qxg`3JQ;3XH5RvIDz$hf~*)6ag_@tNwMQRDR8f`!srYRDWXdFsT
zXO;LVR7N`qqnzcQM@ey&sxCD1F)m+BA_FtCh}szGd@CDDl`@$t2PG9lo=2c|gi;-^
zXN0KaSr;$a*%{;?^XU#R=(_^GdjqXo=D#+GJHRfzDFAtM{+?vCS4^*BzAhQNU4S=W
zSTVbz&w0Cw`aVTraKfxJFiW<}&{wQ5<-6I=FX{=2|09<FF^VHZiTg09dtJuvb=Uc<
z*yzp-(}5M$9k#to8{ar?ynfJf`RBf*;`Y?*$?iTwhd|zar+WdraTv%+<9v%Z;)pQI
zKS;T0s!7$deg>2haa?0G-aA~CAR$xpgVHvXm8z&~CSz(y$ZIg#3L}CfbUSai8eOJV
zBATj6HcWlhP7`ld+ge)JQfl(JQix}&7JWOdBv%imgn^<a>R0WxSxTZGMy$0A{Xhz$
zWoJ+*B?Ta@cWgF2x#m-$gw@kZiXL*B>Q$7EscpQxDvV+rCx&sLC}^FRe5WjWcUnu~
zfkKPZfmWnR3N2KUFt9PYx>QKMS<DuJQHsrGP1m;M8n&Iru$7l_R>XLn5nVHH9m*Q2
z(lXUHfKp#mTRW~g6gHLNn_q>)rUz@V&R3@i&1&8A?8OVd^uY&oT?5HTWpPTpeEJ06
zw4A<eFkVr#<>BKm@}-~p6`p+bZN`4h6vr9@O$`0S*?L_a7&6P{l9(e++t9C8lngNm
z%cG5@A0wJbw_Ma~ZD^cD8_T*EVPBOpjk7o{xnMcLqLcM=$+!wc)|+1P2}@zUzL0yP
zaWsuX7e&lLDxFJaoJ6gDs#awg$tP9H2+LGpd3=m-T9h&j(+DVG1Z!G@sm!l6W*ZU}
zz=Q}w=R`k@^h1yUc#$xEOc}CbngW~kn$0-jTmyh{imZhFR@z_<6swp>A}eGxDMq|=
z6cpYoN*(aX0IGW5CSjzNo;$*-YpGS$uLFfRn#4Nr%30Vvt#xGzGjW=>1H4jwZS$;T
zqH2h>K(0wgMw@MzTXG%c&ePS|V_WF)S*Dqqc|!>^cLgl3v%fdS)TqPz{VbmkF2B;D
zmA&g7Z0jxa@7kTK8~>l*V0qgfk-UF~QD;z%!3?w4iL5j8bb*9$`G(7S*SYv?zM=FE
zH2eo-{(Thx9E!J@#lo+r61)yLzx|xA&ASC$-FLdnZ@2f6+Oy(~S+9Z8KI?OKXJC5I
z?GFu#%AV!d_jR9hx)-o(T1(S-QL0Y~>l?HdVP6Tehzx--nsJ(#f`s^tk?O%b1|}4#
zYI>CdO(PY87()o6Wz|&B6G8!#phVCjiJ++#P)0MwOe!L!f>LA!-aAT`&T$G8<1k2W
zgi<&w@+>G3b)Dr<!YDvW8)4q8HBw~Z9J590?9;kv7bIW9gH;Dj*WAA2<0Gtflnlc#
zlBP^Hn$7x*VHiZ7#5<Cb7e|;P&Wk%jOhxvEXnGCf$nnVu%G6w?DT;s`1?L)(e=y>V
zA^l!P1LHK|8c#{3Ui)J<>kX}UA_VPPDFjdgz7h#krfWOKX(E@x*bgMDX<UnQUaX_$
zJSwqM#aKsCT4XwEtyqi`VVrP{__F2<tBVcZiD>WCZ^$u{3Y>?)cfU38<jE&|@RhIf
z@ZI-lk52IZh+q8GKh4km?9cG*(+~OZCqLl(KlnD*X&yaZusm5Z)rx0ht+2a*UvyPw
z1jZO7?5ztqN2YPC71tsu6JlT-rY%5ey+<p#wjo3oZ6}2YF<`2|bXHXc#~7B&V?r32
z;)F?saoor~tHqo?W=R8ZR>Ir4h)aX6>{iK{90J8T;shZUF4pG&QoLbp4P8n;>o|_$
zVo?k^t7@qhw%T~0FwRg&gcxwf1BJKWe#9_FViqB1V^w`7MA%vco&}7F{7~Fhqn48C
zx@E0=Rg}(()?O)F853h`io}fX=G@C771A+8rZ`<0oaY20B5f!s;hnAS7S)n1mueL#
z6|W**C%>Jwt^#UL&WWj3L+dK0y=%z+nWqCR)OPXCg4gXtJ}<WIAJyl*ZMbpWZUQ%T
zb%kr|UKS<X*!RAJ-dOK0DB$KX)b)0%g5}$2dTIw(;TbacLX0pIkw`$^=xz($1SV$Y
zq44F1K~tDW{M!`%Lr{MT&0oQmUq=-uZM)^(yD{$`_u;|)9oX>J{ar&F*9$-n?eBF!
z%4bGtyVD)oc^4M$(0SbV{PwMXD9`Rb)4hOQP9kN378Y*PkE}N@u-4Lek1>X(ae!3B
zX=~`mK|-X)a<uFuq^LE%ZMOT>T2qu_*><3cbi_u?#Iurc)U1G?tNxpHs#X<>jEUBo
zaU4(-a?F_Ok*cZ#hRP!IFosCeG^8ROk5Wp!T5D&8l3X3<EEE{WfYXkm6|rOq0~t+g
z{g$zsMv-*MDKW%Rm`H#s-_f=WUE9(%4TZdQQ^*<VELCCAcFfg()>szZqE-|utkrZ)
zgGfzOCdoIfFV3a%R9VJp#Az$x<a$kV2pH|~&etkKDNgZD#G*r-u*TrMXT8~M85Qqs
zt#pi(5^E);supf?t63|RrdmnJQ5A(s8k;4IUUJ4N!*Sd3>8GC%mK`MpRF%lUIW{p8
z$DWtZKjrArBfk8#pW*#4eHEn}vUVIlc!xjr_WK;4p770Y{sS?-2O2Mf32jZy1<R5G
zpcFX<jutJ}%7CHdgi%JSQ)kx)XE1nT6rj4<3`}9<=;R3Bdd4)NFg5%Mu}C3_Nn#t9
zQr=2-<TT)|6wYMCfg(}JMN#^J<D(^|6uIu+;f*Iw0cSMBI5Gjl6gW9PL1|406aCPO
zWC~(ErfsHcT8uK{(HaBZdl4*-5ml=6kI`u5Me3#kBdy{5`E%N?;jMQb({+~EMe)<s
zY@JZTB$cki7=bK>KSe^ZX{;W(V-U%fQW<h%$YT}(?5e}(oao=>3dT@UoTE_U?Q30q
zzNC1=d&f8qQZ%JiEq=)~P0KWlgsNFM7p#ORU^k~Eg36jI<8l>jz4mmc-t>N(?_2gJ
z`>MNuzdL{411hcpxpHv%p|<P-w_SjC9sKR?|JD0CG~EIUx8H7_gM2T{oSY)j@O4?Y
zs45sQws`{R6ynC6@mLHFi$T}9Qc{tSr>WF`)s=_<j1=`f6n{)8f0If5*I1N)1%tiX
zuK7Ccp}Ss(#k&i$c9&&00Og^T-vAr<MyEH>fN~%cU>_jd1#bH=rMkA>?!4RPziPT)
zj$P4AQ^Z-JriW1~N;ME-W(tXZ8fY4Eoe*}(p{ywWOYXH|3W1QsFuiFTtTB~E8R)tO
zjf6`T5{8v9Z4@>8>RnlistyH(SRiKsWzIXYii9kkQH+VE@wh5to0C)(t~MLmrh$TV
zj(~4GF=v#~^y5S>5{fNJ6Z(mZexPYv+Qy^Fc%)-ok3HUs@U5a2P^457@}&~jdl6KQ
zF-o;!N@C+wD_<3baT;o_V8OTI<S>o{aT5Aw7)H89OVe0%&Q(tkw9$0!f-sFJrSVd_
zEESC@;dIpkbdHz>%!ZVRC3DntP!wI$N*KH*<!e>Lfd#Y@dooonZ&PtHsEPJSzEnY&
zJzcVZkj)xB4m@ZY#+T=4RYAC`q4cE4N(6EV6H~wD;^i~?J~ECg(zxODxMSG}CDaeS
zFrq+6qa@!OQxo_Fj4LH9U7TX|`mH|0MVK9{B^OL9P2)8&NU@I(7H=FzDFH3oV6}x@
zXq=_7mT{U$RpZAwhpEp&V;x%0;@qBfKXP_4FboqZ8;%yOw6B4r60szdHIyJD{l{-T
zt}-rB1~FJ(ym(0|@a{YBV3fvNsaW<F>?GjcnhvF9fRbWj3IXke9ZwX7al$*tm%sD@
z)@qTop<tcJ(tuQ_Mg*`oO-mP(fUA`1{UddMRv8wZ6`Q;|$O34+Lu&&?e4ukKl&)cl
zX&c&Ah*<a3Nf=xJamFqhtkrnmNEAR7VMdLxq;a9ir6UAk$YV^{tO-+w2||?@0&yer
z{(i5sgVSB!pf}ce6`(l0qpQH^Fs1NX&AK-4wOF`&?PCvUQdgJV-0nJic5`|i5W|&Y
z&O_3xUFB79DgT69TjBWOl_RK9({GS>2qkkw!x1Mnt4``;Clv`K0+GJ*c_L>l$I0~w
z5}GX(wSw~ZfdJL<=r*0eS6_qqH`os|?sk3EEz4fD?Hv~$qNd(>g52GvJF{(jF6$vs
zw|^jW;FvpkdovPz-Ai(>H=(r+<2W!)k;Yr;9E_5DwsfgmrL+MQ*7&Lg2Pqb&DPoPK
z>Dn4TEI98`N_<eo3p5}G?2;2^j3VbKG90R4l%W7IXR#hrNWNN8pew`SJZc<KTHzhi
z>BXTkIEClWpHoU<oJN+5j(!-?MvL?WqU=A80q3j~NiZ=*kvAE~NvxZ6CN&;o9mhvU
z6=Vv~v^C+Kn&?jfr4=y*6pGfj^@mpYrZPVz(Jfj)k%H)-<uuW>&DJ0DwDA<gTT>UR
zkaHo767u#<Q*)mTnSwPQT?%0wx7CnhDj>F8(~#H<6PgP2l7zMtQmm2?fH&tl&68HD
z0w6<5VVfJ0bM;}3ksK4L9@94hkjT|Dw&q-Erzyser$o5u`SG{D!-wDbzBqognxmyh
zxnrI_`$V+i=Bm<|L~pQM<_9X|R<z!sl_5;Re8rl-*)|O&N%d^oHaKG#hOy?W35%?e
zP^l`46sm3=nHVy@6_!`rB@}TO$VljQlh-(7Y1)PtXXm`USaWthFb<K%8Q%NgU92g*
zZ608BWY`RB)@zPVj%eDJei|SvVw})M;l1MvUwD@(M$w0yM$$CW*abNSw30%Q)_cM<
z(RzdSo?KcABH~<fB4tFb$H{LPLn0=**OWF)Au@$1)!<1Iz2{;DtrSHS7R!bQW=Wa?
z+8T0_972>u8z~yG)-n#g<WAOHqB2BsiPBFgu&&T`3t=vbW8CyuE6?|$YssZBjT7S-
zC`HqBo@KjWm?jFD5NoQ(6j&}=+NMa#!W8P8Pke<x^K|9yTlUPmjiGuIJ<=iY^ZK7}
z&bt)v-ktwSHb-4Q#v$hG`uE%N=njLr&%)T<7mTPKU<aUihT%C^e^U_}_sXI(*wWyO
zk)j?|Kuh$cDMo;uvSdaIY0KK}LfqMsOk;8UCJKpSDYfsK+~gg1&v*Yhms_skO;mO0
zUcMizcMo3ab<z4827dPmpY9FQz9tKJeP45t^e!0OZQJKo6*~;QjP|`n)v%qhj>dZ=
zbXBWhOLZwg3jnKmI@UQ6IVRO)XFx?tsdYysk&CzvIBRPs9?@C^c5{iOng?X6>c8`j
zF~sUXkeEWM0<#KdoB$6L);k)l#qiuW#GJ^8*4%R0g3>jQ$D)fudvPStT9ZqpRT`9G
z90nB7wyw&(NZ3j%VJ&R6X$m#;m!@1b;Y7sQTPu2OIe{)B?V@JEW(Nw4DcE8z`xR@d
zAaB4o4UKbHWtoC#^Mw>eiUeVs7K@IQGNDvgkeWBBr02ycRhc}orP4~~S!=i9z{VI#
zERc{&;au4`gsn41O8pg57&8nbC1u8O6pcZsA@VABp-Yw4kZ?0Ga{BNw*(uJ?U*a~7
zMbn~;=HcT995*c<MObZk`pI*t@P%6JVMwLmt(W_t6m%(DU|L8}6y7$pjku!}CBJ(r
zk)~-GM+rlB$4w1c$~8$zEWu(-#1OI8V=CjUl~gDjtyp-^I3`-(;;LQOFb;H0gI1Q4
z<wHs?93L-;rCNSXSwe>ylyMZa_|O(uovnd_HUfl=F>E#i>iCFm;c>>JP$*~7Iiadf
z;lg`qcdXBuQUdIf_|8fSH(QN4ov8M90=BcWl#=n@({>9=$_(QGN(wMy%2LVR&94eY
zKd5cRxl$rmbb2h-Xj7E)X%})1GT_US!chuS407EcE}4cA-x!RRL4rP!JjrpGFs8}|
zQLq{st5HfU1D%t~abr7*sxBKTGL7Lrud@SFVSmEVZN15X*v!tPc3#nk*s527n!B#Q
zdxZU<?SA3T83^31eK)r|yzFxMtHI&C?rl)wer%_@vcH+Fo59vI6dUHy?ByzjCU}||
zZYeJBmm(~jRZ8BIMH5v^DT%<DXlO#!#gijhz|jCd&n(=%zs*}6iG#~;Ki8`Z4zqmR
zxQw^o+SgmP|2KGbtM4)G8))t9=sx@0e@ESK-TUrip47GZw=#bBobCnev@^7Ai}#K=
zh03&v2jn<T6t$x<RbZS3nzn&-S<watblDoH8|`ZNPm<w{k_uR@HPbYqwIRh^D<MH!
zPf2-O4I4xKXC+mY0)5y3aJ)Q1G|CFYB#fPPj>b0va+GD!HH0XgLe{0`33-f)umwAo
zX__D>8YfN>*19UcBA_PcMC%%IQG_u_L21sQ#qmO0!x)2vFtwJ>SLa&cC!V@-oKRN6
zg+{q*=_EEbZPmgX!bl8J!d6K-a&H~Qh|@-xCf4hXFo#z3!A?$2&<dtu07?y?Qk^3Z
z%|WBdKn!tPUD&i9vSu1K%(buD$TrSN6|S=qCZEP(>q24mY`X@jNLX*SI#WtIYu{Mj
z|Ec$RHk|WhJfmoZ9(%fO!J=C*^dlcWdCCVLe1X%GM_jykB6@5}M2l5QThL&TT*m4_
zAlh^l#CmH*XDw&OVPYI4e{L2D(pBs?#Hw9LB$kXa;B7-piAaK!an2K?NSYXJ@Qr2+
zfs`Xk!|CY}#u$cSKr4$i9%V!r`uuz)nuyLb^?_AC@bt+`-hTXu5GJhiK$cw0PMFzo
zjJ2yN0#0*b=zES&j^#RASN)a~#wy5~Fik8MEei3<ora!NsJ3@w&Al^Bqg3K{&hhB<
zM2cp{kzpEXoaSh`5beB>YT*afrcA>7S_{yMDH5hcvI%E2ZR<gSZzWA&w(?U*^>;s5
zjB~^|(fQ@}c~HptjKfIR3LqV)h&F~si(HOU3hktrh+HtbNJ>W>F-ByRKJ%n5SsR)5
zfwuXTdx&ki4Fuh<E4w>@+WmYRYjp|x_HB3BO&>mo!`r{QP5alqZQ1o{KMQqpU;Dbr
zL;IZfyBm&uj$*nzUV&M&ah81AF4*?Ks<Ql|_~3AK<q=E%0!KXKY59-n)ki=^M8h&I
z5l3bu3gag}HtgPam*b3^zst7md+yQe9Neu%Dc5FR-}2zPH?C`aF3O=T4l}g-TXr1^
zUElsH&Gkw!bPWJrS$EGZa|nnX+{Y^pedw;d4)XTRyAQ~_XO10E6k!TdSxLruOUx1H
z#XE8w2b9vZ?Q*MLueGjzof1+_C5p^Ql|Z0~l*%w>+Qy^MnE4&8MTnGI**60~trZ~z
ztaB)3tMq}~k=j_4R;>Gth|GqB^`3DYVVLl4!a3QOH->Q>rSjETF*8RqPO|S|lrU|o
zc|KB{px6wfxC_)ABU70vt1Yo=#>Hx@w59izVhkfz8KxniG&HTFYg>v_HIx}p6hfLv
zS(r(!wG=?40jDI-$v0AMELK*Dewvt4z*;d)mpj4MP%7KbID~-l7ORA{2|0+I$doY7
zNq9}^tt^ExrrNg&&=rwrFD?$Hkix`zy+$GOCs*!FsF7&3b;;1B;Emz&yKiG3921|f
z(5<ailA852@O!`iO;+bC))xcc|KU@9<`;jOlf^0L;{|aH<RZ={tIZm%HH)Uhd&e|P
zRh~vyU*N*iXD{f74U6TH<#Nf<;s~V_DTjK$j>355xq7Y8E;`nm4Ka+YR+}x8+qAIR
ztZ7`srZQq@s}&&y9-N#?MY5O5>wbMM;5<&mk~zOv;f&$&gGVSpp*dQfP>RNwMjSM9
zVi-o8wWORF`-xPbA4TLkrK)kS6k|+`AroRC#K`f{5jh9OII$UeHk%%;4dWO%J~_o|
zC`!=}BMPakwnk&CKQkq;Xf17XL_w2Np!J3r1IlSus||fW5<})<y%yo`#u2J7Z;X+y
z@ig9Ho#IrV;GL7oaVPe0&f8iwZ!74Ndt(?Q-aFP8Jrof}R!Xy8Z`f=mtQZ{&<Lw(%
zfpG|IhJmi_aIU4~!qM{RUazzLQ%3Zg4!r>PeZFnp?y~D_-R|c@z~izrdlkEO^YhIP
z`EH&2t3ciD?47!$&9}AXwtXI&w}1ZKz{6!bcR@q}B90$OiA1RYX<n3(8&~hrU)Qgx
zf}$zPismc7fagoR%U|Fn{{a{L-%+e)wN(Nm9sdWiI!5uEDEtl$Rc2ZcC|3@p_8kBE
zAmY`)_%LYO9jNTz|INAE-*nsl>xX|GZfpO%>jQ>;?K!lluyffCtX^&a+^dJX6>QaQ
z%Pq&av3)tbPsJ>XbGPYUz)qFIF!apIdP)J42otLUB}NuaE7nR{Y>)&fh(=fIJ<b}2
zsZ?7k$<@j^<DC_-Vl`n96LRmoglDT2Qc;B*OSSA0kJk_fj5SrFK^z2POgN`#y~Sve
zw<tPF!n-n9Beq?(@eKVy+cu;iW2HDsLVL;ul|-Fr)Aux9Tmpu1Btuophmysowrf4E
zGL6nz(H_f^P#&hR6~c|NhMBXl9Zq{v$XhuTqa^IA(e&d$Lxbr&&tIH@QZ!z4?wo5F
zhMo{7nzq9iOR*9H(B8|9U?mhLaX3w!CfcSIa9XQVZ4}9pu^`o>F)+q}Z#>R<QYs=x
zQ6$7{3^cwWrA)tB6T>8IW$VeQx+c`h=h^B>YrVCcv)XXfHT?81{sKHcVe|3#dF!3G
zn1%_b4bwXCqwoBH^+g~B#gpe7e)!=hy!VxN_~6U$f-YQaRvaCl(iN#nOF7dIffp~%
zdHe0RSzoM}rh#|g{Q_;*qEwOmNi6_Gl~O%~#onzH#V`$QUc8h_Yc%6H2_x+s!x)HR
zpznL}2}ak!iRAQ^sciE%FD`n9P<?tcEZSCNbW+B3ON=#GYgsHcLq8C+xQ$$_M^cQO
zo-P@tIymZDUS6D&OJ=hfX_^+J6)(@v=*Ni^s!K@<Y&Jcor^h&F7^VU1Bo{EG#LKgn
z<Sf;^F^N=)F>rd+@$kWs<b>J=V?8g=H=JLrc<|sDtqW~u8K=Nv;b<C5NQuQ!L(0|i
zud=^moahz{!Z6DD7DGrd^druR_FTUiDJXK5>yuL=CJE8^{W*O<(E5gyC!94^AF`tq
z==&b$Eg=NjrejzS4C91D(gzd@!kZ8%-ox1Os?qY{PUw0^d!OkL`*wGLw-12br6a#>
zzjpy+uWs|9bGys;SIw7Yw_odB@X59<Gi!H7!`GS18@B7uz)nfOfSOyOwkV+}>CqFK
zh+$G+pzuBhhvFR`@>gkj#D>34<Yi4tDirk{EdO%~|2+!-1r(O*?ZTDcaSzyd)$elq
z_uE06y78&(op*hU!+_@y5W9w$ZfjIsN8LF7zV_VQ(GJ0Sa5~7Y?c({nwDlDjvs|08
z`&(Y$wZ4m{Or7or>=d;MPaG#)<MCbqiK?KZv5vNF$vLyy^u#IBEjqCYN`+W5<CwN#
zlNcrGIfR697M*8Pc?i=aR!aq9j^rS+Db6~Kc2Fji7Eq_q48z3Jr!R;x@rC!^1}Dn@
zA_*cnce8a=+jb}gL5YM*DVd@S%1F3bPnKy4q$ul>GmGUCW3*_TjgolptRn&`!D<|F
zDObipicL}yFWWFhx~8RQMTk-v_4L^b-hSt8+RigfBi0+1i;j!)6}gJI3V_yxBn1!a
z&9DWrA&#_N!xTo0an*jSpi-?6t$kFC3FjTna)Fz*`?9!&<Prs_8co+VM6H>uI2D8-
z8f9@-oL1s^Ay#HmBta>7Yo(AP1e&H{==<#rqd*uhV=%^W)HW=S7O1x4<!VJgY$#dS
zDq|Fj2OU5AYd_7OeCJ0zd$J~sifJ9V_~s9J{`8Dr`}JSsorfLkX%JuFM)di#HavP*
z^V&|1=m+t<T`oI}mAu2v(35hc@ogn#1W=<COT8eaB2Ej&qR<$naGgij!GKb5yj;+4
z2Bx7WsEjunjFqamf;fo8gwdA8QMc7xTy_fqL3QY*3i#G?a&#n8C0a9#gRsDjr%-4X
z4M(R-VoY4DHVi|c?K+n2lA?Of#&eEOk9qv?LDlM0P#jtb<2;6dZ$$Bav+h}K24XB&
zYk2hdG0R0OPA@7m^ns8jo;^R~`O6EQJb%G^@4iLbcv1q;IIC-F#2`9?&Os@hU#ul`
zT1H|>sFK;No-$4&FJG>3*0Ef4TStp=9BCV0t^Pn6OKTezrwdx&(lid|MB>L=EBTO>
zO_nHt<U)=$oGhAriq(2^pV!%eX*V2q+d5aLLttwEI(PfLe;+qH;@yt+)%)Ggw!I#3
zyse#w*T3yr-EF(uj(NDJy$P~!R>c-Fj(>>dqdi;BbWa6#b9Jnmi6|vbJwQ~tg1cYF
z;-LD;DjW;`3Z6%-_`lNgM-mM|Q}`j8|0(bS_{%_h119a}1@0R(Tn9UM2YA<yvAfO<
z2(o9ro1p6O@3j{)9ay{{w9W1`SNPU)^M1a(3Wx8D!!WR1S!eIv^a`-GyUqM&*=^g+
zYqWd)%9S+@cOSUk3)rPpNI}ApbB@(8jJWE~FpZIt6zk0y+6d!Rie?-GjWaafW36Me
znJP%Iq%;s?!26bIOibe>)<Z^<vVsi9%VYX+5TRo&x^6i{oRtLgc`_>V?&C*<80nnD
z))3}{<72FnMD=ISU(g*NGxP(#X;3v^$~6*p%_Z}q7ag(HdLu3ZTC-RzcyX~}(KTY5
z6%vha7=}SuxsX_`SLh7_B%SBQ%M~eSjuuN&48*KBKfhqJ=~*5hGv-WInv07K#wyx{
zXS43H)|2wc$?-AcH1YJ=v+5CB!&kKeux%aAc{ZC3F-?RJX<S1HVi`6?5wR_(tsPWO
zq7$a7g;eWZbqFz>uQue6Bs>r&$!#o#*`OW77EVzpNs2;@)*@+BiDIoa);r17D{PrM
zqcz9Nj)xCVamx<>-~k_g<M$*tE+;|?OktG$SjD?veuV2BAOGY9!#bmq;?qyg`JLbX
z20!!jAMoJ)6PBu@v9|th31h{*B*lnxmMKo6&!!aTt1}sVbPdxu;*2AN31uW7PZcO3
zvg|r?Wu8MAiD8ludOu(^OwQn3)dKYLI~y$w)G%y#=dH(}*ye`<j8mX(TBaBn#)&9M
ziIbBfoKeJFAV)$NG1fB1iIgLbhMW|e^~h>7a(1?%AD&WF;@!6&Ng+}eM-Xd-Ax}{e
zqA+106|8G9+A@xb#iACSI8Q$WR;yn8j!R~mMq)~|?E-IFta5z%;Y(7Aq6ruitMwY+
zwxlGLw~MC1H;zZg57=x*G?tSik2RW;<0Gt<L0D5oxSM*PoSmJqTrTilD&dWlP;H1o
z?oVVOSE~4N>LmOeQy|8`re8~ybY-4H5Sbur-aw8WnJWqNw*tJ~=`h&4i=4;x`S$@@
zm-F_71V6)m@3Z`MK<mwAhe5&J$&l;&IM^lcEStOB1PnC%T@;b4Fbe6!HS!WBRi<O6
zS!B?bfus}(MZ}<ZM4^YOe<zd1^5=NKFX8yVu;KrSVL-%BUSjz-Q2da@e-n5>J?KvB
zU!B8wJ=aiOS$6Yk+`M<L=3Cy}Zr=JPpu4%w^-uR>u4THJ9&zIusCnO4^GM6~kn`_(
z8(`Zv9pvw>wPpW#UH#?iUv(W5*1=>w2eobIB^=c1a{_iDB+!QQvy~*)lQ}+G5U0!#
z)`TFrb^xajPEk`NhQKrqMAy*QFz5Nj3g37ZU57DJEt_(p@1<PadCRo!$w@TM`qhS*
z1CJj+B8&mCWR<6=q)gK`G@WPcC22jQu%<dstk#Ufz;V}#!oDi>(+2Mx(==8?bLiR@
zUvui56DgNwl7#IsBsRSO&a?9s2F>wuDV3fvgA*OCW!DN@G7f|i8OOvJg5>xa!^L`F
z(@W?$#K_Un0!k)?NG_V^&(5)0)3z<^O^>aiU2nu;L2HB7mh+1hxfFckp_B@i2hj+t
z>iwJbn&aamls0HBV9Yur1tLmA%~4cEL&`iqKLZL@8Jf1^VznYnk!5>?QzFA+jiK=k
zMyZ;#U$#||#z>CcIE|7omJ+JeLI!KZpVul<iM;>b`+V^$U*p54PkHg;V=x+TJpIt4
zjW|3Q7kTu~F~wv){_!*78gel_eSXgKzx#}T^3VJVU;gR`48w+_Wy=&tUY?zayor%8
zxi+f0n|Q1fJGNmM0L}SojkT8bi!+?_cqiVcImKG-tx3h;m1P<OP2*{shLj^oK{f*F
z#uz~rx<!jNn$>2-V!1$DQ*-$WLmX**OG%MWK6!$3j`e2EPkrf&*tS6zgKI2V$!}$>
zCFji9%QId)J7btKPoJ%LdA6YxsU}~aO&lGy4C{e$JxEwLie!manxcd$B!f0a#Ee5?
zy&34*mgAFSTHnz4mbPgyMshQ|w&i?vLCNs=;X|Btw2e5B#0v6?5+H1qq8~PdIMTG8
z6eXdt&frY-bWXVzLy7jDLh<<FgPKFw6GCK|M#;gmo@tu473CpDrYZ8##~*`I9G@IB
zj1w<jJ`=r0?<rYQ8)P!~d7T}Y_J?BU>2N1-I2?Oh=W!o~?$z60??UeqHoffH-vj`B
z9Z+|3-*>&{x3}x^{135L>dHF8s6EH?51HLzx63Mq6O2_rNiJqlT3W0M32{G?>~n>v
zMxJr=o}zZ#Z*k1OjN_*n`S&S&ghYD>9RCik{0S5PT{2&pDJphB-F4u#{~TX+54`GH
z-zJ%IlL@;Dny>HmFhG3ubQ?=Pf4|Cb?GCJpngOxe=Dl6#Z1)~>3l{GfALjbmUj07b
zb2AXV3b^xj+2y%hdMuhf;P%jTFJR}arJ&hNBCzXP&!(S5;z4U3JUk`m%=yJRo6TCp
zf>Xk|R+t;3nWmA&QAfAvC@EGCS<e_orhb$WywYeaXlQ+dCWCH7du`f~N@2NJuo-)i
z(oirA6PsRS8gi1Hr>+xo`+n#}vcX#3{lYs;VUl73BMI6jWnP}2akN}wjAGafq?qtt
zI{RhQaIs!ttYNWO;FP87TGpEtr>7@UDHunLvb3%tj)}e>={F<BI*9P-!6{mcrBmOJ
zEEWq+PL4@2P^{tX>>S&8x<$v2Kl+Hrj~|gj#A=JN7E-`iC&$W#)oP743X0fwg%D^v
zVMiZ6c!=wJR53KZW!-PEMiHarEiRgt7$c94PMM~$Ccv*4L&UZgZQ-qV9+6XqWSG{X
zYxdQzei^SbAAa~_sY;HClv9=0kRljo1*CiLQK{gaK|3o&B8!&yzW6?m-g%47F!JE7
zx0!OG#Kbggq-xoFQp(uIF{Od!=@D<G#FHPrKqW|7GY;_WZ~TNWzV{V6f6UY8PtZ<6
zp&=#u&4Blg##yEim{RpBt)kdk88(|*;aM%aj+Y&e9xTb@!12)$A!kC8JkGXV5T_u5
zxG8gXamLHlW`0eeD7^RX7Z81V$LeC_*^3t<P17`NHUqH{rnPGqoSvNU==33R$UHqe
zC#4BvMM9=3#jsw>;G;Dirw<pLc8?g|iln*fn9SR6KO)8fQs#sA-sPxWV6EhW<}3_E
zjDc>k#9B!$7^gtnwxpEs#?n@S@F`5>ERr$rzw-_uh|Ca`l{H2z=CqdLhf)+2uvSyl
zLTmw%4sRR+LlzcM8AGUExJHW`hSC~k9qsa%4q=~%O;1dTqmyGonAi*hpFDrYs_!vP
z%X+is;?pODm|3mYBHf}Dt@pHT!`Kgu)0;3A%k>3&bGkdII|Rn=2DA=sbCrL)3KtJC
zGB??h{mbtoqq6^4yEa|>G|~1kKkC}|-OsnNM0cBa_x0y6>_o#Kpjk0%%%Q}Scm{R5
zbv0#Pj@&g-l`Wu4+q>w=4Zv}=9Y?hM72q8v{w)e0Z(Gyxw>c^wvQqylWBJoN6wkb$
z+d$pz@+fx!+J_ms!^hm;rhWJQ&HLf@A<S(6^CpP8{_XPa+@h%+h_(Nj+68o1&+5=n
z?S6y6-J!#!>&bp#r{MDgb}=TFM;-p{$D|Y)!^k)dEV_o}Vo50x*BIXNr-UHM(Z|O}
zgfP){tw?H=mepp1bsDq)?LZR0#c3nO3ngL~E#8O-ZOBn*Frp;Ar%2y#0P(h5Z$?Uy
z%17~x%uw^kT-(rD#aj;_V4dapvuA`Dc=YHYA&MZYF-j`<>;r~=plcU6=Que!!aNi}
zaIw0;9UtLcL*u2eqaR0Bs|%D8UAm{wUNDBhIE=ja&bv$@ptR<__ul6VO7Zmhb8-wE
zA9aK<(sjo*e=Tx;et{yhXj{owY!^%+u{e6bI4QyqNeQ}z%$Lqj64D#*`1q6OD6M()
z@FBwxn5u<VE>a2n{KYxLFiP0V8oV<cA1|23iOqV0b&iwcCH>f=t>U;{)?h(`A|~@*
zi!+69Jh3F6eDV=}zvjW>m?=obqSHp0mT9WR2pNUOwVFlOVx4FC;Dj%I>3vQfo-pRb
zk<xf48gj!l(vK6ZYbj`AE{vOzwq26)K(k!())zD{pRJ)JQgpn0x#o9%`|JGXfBx5a
z^x$p2^Zjo#mduNn=S(qja&pX3+X66L46IihrjS@39dUkf&NM`xK6{BVQWUVd7-+o5
zd&BDD8LQ0(=R74BUcUGdZPU;+j-%xQZ(BNBc>dyylj9>6UB~lJ&zXjiqva6~9z3Ld
zyhLe*a}N999i}PJEjs#5FN~y-!Hd=gSGD9)%5~Xn31+sEG|qXnwzW_wVzeQ~Ngny?
z-cgcNZkKBPma9HpLAIZ!K*=IbT(aoZr7)3Wt!UPY#z<O)6fQxC!nivn@-IqhDDrxb
zaT2!_V^CTPJ7>j%S)N^Cbd8Z}>=dgIw9(=j91}_lL)*4~>v*D2Od(@T!w-M>J=#S_
z0oHxbrXL7Va#-8eGft70=POdk^uu_c*V%#TCcAQI+Sh5_#@t<QbNl+YGa9=c+`eP&
z-}=?nus656AG>oG*68*&><1>dweaw@x*O2j<>V5IGZg(clW88bp`;qsC`C2}N|s26
z7SZ7ZJ51eWFkir7kH9{?J<?6MY_a@l;J?kt|DMbf3F%UB>bsnj{~0gT|B}eh?hHEa
z%F<mecG!3BSGRlBU2%92aC7aO;BtR^Zyw?1biKX%!PQmJxPJ(9bKceKwF}_pbG<o?
zD%Y;;&Hdaw-qmM8QD!;!on=>n<a|B??Cw6@3)r=t$Tv7=Nku9lQ%azr9|lU1kd-ka
z+&c`TNG?o4{7bb+dZ4B1&|pv|v)S|*D;ed=h=TxH6LMy#Va%p!g&i0Jr>92@{RZbO
z%Lk`8@92kqn>?9IqVLz_7#YTZHy)*h6;N7ZRFz+;GAgYTJFREWo@2CO(Je5>5JM0v
zBPY@%+Q6c1MAo5cSgluB?|6Q8K`Dj4de!#Bz(v2=n$*WIalTp8HZ7k%d(LXzlaho|
z8}E3z?)m7Wk8r+&VhAzeY(wKTU28EHYpgR2HLNSbzXoe8=c_dZ0kjVvoM4?}44KUs
z7-B*hN6HG{bTo~mU#5}>B@sfwjz%oaH1z!j1uS((%CTw>X0&olG4Sz6A2IajI3+F<
zl(Mysn<jDO82TO*G;PZlK6saRzxXanTTUK4Mq9&FGAEDU!!}1W%>~047^jIC#igT+
zL2}@nqhDVLV9$zn(cw1}B^9Pgk+R~)Klz04xBni0=AZs`N-?~6cE-iUnv;_gyz*SU
zT#Ix_NQ8}mhOmm5pn3P<J9N53D@D^f7M<sGv7~ETrZKQ|r@Zy{TQv`C)>bT}5J)M|
zb_+JcgwdMD+iF=S71^pNjFS6BD_v6!B(E{0Bo)v}9$U$mtktC@qY*J=Tg9}kw<79W
z0Ut`Tb(9h|tmLXKSfFTwxO-*EsYE<vl@gslXDu-b!;nhBH?0_zCt=x)))dP2ud4cS
z;)z_ls**{iY%RsCv7kjdsAS2JO{vrhaZ8xS%IaFF_zsyk1yU)L0+~#VJtb#SOoWga
zr^G+}-uL*?N1t%9?kVc*a{OB4xJ~0FpKuIN3L#~}n;`6VY}+rVeL<D$(_QD?zU*dl
z{63azciwduiJJ_~KGy9}d-n4?ykhRFK;AyK?lypMbsvY0dtc`6=J79&r3fe{67Q2a
zK{H(8pA>WX0~D1}O04oEuTF1y-b+w7pTU4|c=pCoFIhKGcr3q(V$I0EOW_lOQcx6a
z`7tN)uW@ewRg(Hb-K20c%z70V+&u5H^Y?y?=<fV|aB7#)x(%q^Y}dX!`|xz=KD)Zc
zE>N03m7P}JWa#z>S`S^%`FFbvu3jAw?#j~K1a=B3mC-wYFI#?n*mn8dyz=h8obKs7
z5$G1J<X2UFHmwaQ<yirRDJ1fk$Tj@6T5s^)lX8*mav_z%YO`k9Eu~rxBrjDdQkZay
zPGlxlJxb}Crxv$4I%W(kyM{&UdHLcQN5?0mB9>(_1}I4`p7vZ^oZ+1(#l$#?Ovp54
zLh3PE)Au7`ifAJa1<pCv{Y&w3Z5+8|md*)qTaQpOo$u&-$<;De^Z2btP{4afm?oOV
z0;LV(AjKFd!zWK(V2owmk8Cz0);U}(ygXas8%K9^QrU-dylZ&za)lxD!B4%5E(Mbl
zr>Cb(Az-zk@lKc?ZMayiDXCzMqw6H!ut9P9K-0K}p&#jok&q<8{Gy0&VbeOMEVt&O
zJ4UIIah!O0zF~RP;<ZI73A+{87Lr)22{E7*C{@_3MSOP_i5-T4Fim{;gN)II_rLr_
ztdVfW;^+asJpyGI)*F8K$3J15CRBlU-+c$~EU{0FaiVP-36G=bP8YoQ_IrTl$*0fx
zlRy3uoAtzpA3foB|K9KO=l;T<<){0Bl0<u|U37>bvM^sUCeZ}M)It)`!ZQj=Nm5Kw
zN)2yoxz1KX@ni)`(KH_2LdjXeGmRCe60LB~q6v5>xt_TsRMu#xp-43+S)^l1o}V2p
z`IRZj>p?+QBG#>}MOj3+SWEFr2mxk)-fCedIdn3Bu-3}6q@0AtRjFF40?2YMvONmz
z9OE!xy~k*9MjTOcDOK}OLCRa}zmy8a$T3CEMr2T)J$uIS$uShjDHCE8S(TXUbq{Pd
ztF6O^*qQ}WjD%WkJxvo+n3$%Caf*bH7-D1!;viCzRLbWf42jk-ril^@+G=qP$qLk)
zD4v)@BKt@4``5qu{SbiIKks^%cDQ>xoO4%pI+*?NZ9h=FT>B>Ix(YCN6Xo}_K!-rV
z;Wk}e=T-YTy!^KHZ@(5Z$O)+G;u?8~;>mU$$tAvQrQ98bOsr7PY&&IHz8hp4xs*-O
zT)n3AcU3!LPs2Y;$FDK+H%a`56t%*W@Y8SeVEA9~;`o0+<P@_l#ZXs%nR(7`2=8*<
z?ZE1C`7Xq`dT(66Q|<;e4i7r+GhKgD_J6+qeC(zOT;0?B{~LE_k^iafzumLA+WP95
zRe!~TE<13&?1Axn9$$6sL)m^-S9aj;tO0mWU}u~rl>{i-u49-+N`Wb58fO6!WsE7H
zl~fP5o%qWZ6wW)GsSJd%lngm0hJJ%KhSSprOkrTtkGR$|j1wtkDH2eSi-_$SG*gJ;
zVqh%meqghaYSL0N-9kJxpFDrYdbQ@^$pbDf&auv*jmTRVBOU7$CK}&VRxwc~P{wgF
z^!S;X&{mS>b0USvIEVyFF5tYWRfVGGw%PQ=RH``aNZYpbV-Qf0BJ1^rmoHZs<FU^2
z;K4E8cQi(^(2AqeW4!Z(ae|Z?gM>KSreVG6$)&Ig8?-homR%J`7I1xWzUJwZ=QIn?
z@_2!_mMM)4eb2?&IlgHqWzK7Cc=mL~YPF*Aj^jl~+jWH6>5iMo(Q*lKqBX5_u%+Ov
z!+S&6^emf(7$W_;CnsU>Y}LODQ(|+`^Cy4w$2|Y>M;t9XnngqR@G)Qc)j!SYJ71#n
zmcQ^@e}R70lXIbM9M0~%uCx|`WF(ZQwXjt0z4IQAA3o+AfA|l1`SOC7XBVU~^5yqG
z;Ns$(7y{ObmY=C?747^W>#7-;$yjonP{nR-#Z+|)X_|(ZGdf0$)`XZaMl+5A%2N<A
z=u!$PWz<Bq=CaUw$^9&)zFzY-`8wr<O7-<vrFcZ7TaNF&BgVM(JspNYbQmd;vXB#(
z_IHhhz+=HWO^6YcCQkw9#L>lIE69^nic+Lfxs*)nJ3^Eq230{|&IJmLF%T;l8^)fq
z^K+Dz0*i5&n8wJ(#hOju(>4-PPbo=SM9S6LO<q$uB~ld-&xo(I0+iO(<wR8$Fm1ID
zt(I#OvnCbD*^tYe`6QLyQ$LdGbw#Uf74m1Bc1PZDv!l3u-eF*~Z`pNbVgGbDR%L&i
z_Q9!}K<uvJy4yhO;m`Y~+ko$FpzZc+akUG6bvHQw))|OEpfC`akihCyR!r@zEq9@+
z{;wjFa@nQp+TB}J^@;_*r+tbU8nB&X(0q+W`Im|ODuKTS{0QX|cKt^z<KO3{{S8nj
z%-y%zg1B99e|Zgd_kQ!bUA;&44G#7V0&bhLZ&G(y?=a(cXrI@f)N<LH-T8W+Q!b(3
zyzTBWRo-FGN|}{xu9AFL4t)Ewb*D4f-Jr~~RQ+78oXBj|d6(&4z^>GbGX>GeQbx1c
zY)CP&TrQX^G*gflx7N}zX`|3i<Gd%8%wo|AkP2Dy-K=9620r=pQ=F}ARAm@qN~md)
z-^5r#h=efVtpU9NG^>lA5F>dAgb;abPH3BkDFj^GvI!Ha9@{!c+jXQgLV-pbtn=8~
zIjvSbeLq4`Ok)HEPoBLXW$8SR77K<Dnc{|Fm}pwby(^_upV)@J?@3u?Mpo+q*GQg|
z_l`#o9`NAw5yKETKfmDRi*trek9Uss(9^Uc%52d%V^PZBl&6Hk*|ReMQik<r!}4gs
za@ld*F7OK{temZOX+y@eS}dxTj(1NV$*ZSix~2g%-gn|zn==MDIXR&pdX`5=jN>3q
z0kJUjBNw43Bt=Fu^dnX)V$S3cp|Y(MC0@L)A*Z2Mml})`Ey6NI!ubn6{r)#mB+@kU
z(_i^AKl%QTW&2D8SaMURBJK{e1)4&1q8e}b@|WModCNce-9Kb=vEqk+`~%*9>oE^c
zjyYScx5@p&^ay*DlIY7#<Al<RQqp#GHVgySJ7JEi*J>(>5CTzad}VEwHkEnKl?@b6
z)2zglUMZ2CF~&$p)!3SDChYD!K1?B@R7O{SUTckr4aZnJ3pwAEBU6YxdG-{o70cz4
z<D(@xq&j%=#1J7%9%`mAjT1R0#$hB(6BHz^AO+UzHKsmCIVQ#_&~JK_uI);hoRiqV
z6}A91rOKF9sT~=O8cHc#j2kFoA6Jmy5V8Q<DlRVj%Jto<h$Ranh_H#OvXV2fGfGl3
zw5j|<eVG=-MJ1O^(pD;zYn!Gw5{oTdedTlQGxfLG{ag;Ncla1L_j73fw@v$V=We!r
zUl+6=B$d6(_jmI9_kRdL?BD*-_kB5F*SqrTpziW_xZ1tH>Uw8p)~@g)5>HShDcakz
zF|zgUogw*krGAoo{*nWl+i;HJyJ&v%+IdKSdKFO50C66I{#iWlV)$QE@;6a-q$_`y
zf%ga;?ObQc3f$$Zcf))8uh-S@`xv?Fzv8R*b$Hpn{q5Wr<>uU19cSC#?Y5=9ldkQr
zl*@lyn}1`Uw>*t|AMIUA?(SFPa@L`9FPDF=1HU^?_XBoCSG~D4;}lsOEkyppE81>R
zg;o{q(UF+KmrN=d=RG+^&d)a(t(aoMH=d*t!!&SmdMXJ?JJ%@$P})}~2Tx9kqBYG?
z%h}lraY>jm>&;pO=anYL$i>A#$x6)OD~k|9pluhFoM?Q&`+=r$#E=MK;^^pz5T&3Z
zhDaC#i%u&2Vwg}_QDR};Z%|1XLu(abjD%sJTeidyI2&Gy6Gag(*|u%y`wh83iV4ib
z(W0YiJ8XjGqeo)io&sId(kjR4gHsmER>Xg^R1qE@EeSEOTr3G;0<>6kX$cW{XE<3N
zOHoDJvfga)UKpEcny^~aEfzGbC&!30Vl`!5BjINw9<y2-9-Ta(A4eWMJ|V^lYa2e9
zo-$1<zWMF%@#w)3?>{<5StTB-N{s0xVPDMLt#jBc*(Q=#m(bIQ#?f{S{fp;(nEr^P
zx8CBdx4y(jKm3V2kXocXN(Bg}7Q;w@L>w}#z05Rt|NSqJOW_Z{{`-9QyWivecfP>S
z{ld?3d}LUyS5in&Wb0|OL0Q8*qRt_bl2m?LV=&fLAW~39i~EO?O7Lk4ByD9t(KO^d
z7X~PXeyjxx1#1;1Ngk;-m5EB3oTOOBH==(?5pU(y#hP&%SuPga{L>i3hHldj48y=U
zj(qy$DQD-B$NAvF13viRJxUCcI~XD%CP*+%BdH`}%#;egX5LIyHm>AY1;DGH?>vev
z5^|cyv?X7TZs$;Z++oy81!S`bb=mG?euit<wz{5_%J@R9P_9Rk0YM!xD@7`*hO9HH
zNLQC;P^YRatWl_}w(XuPWGQbd6#FX1Zr}Vv-><*a55JucgQxw=@5*HD-^Wd`bG1{y
z8#}i@ymknb9X|HHeOzzT_3+~L=_Xfm2-NQGcLrgZ#7B($E^vXS><q9j_q_`l=K_hc
z1!QokIXC}{z&Ek{8k*+H)xGS-XEVbY+))vWM)5U*{_oT9KLPzCRQLv`XwKb#g~ZQE
zI?T>F-TZtzNW1y>W+vLL*SlPmU)cw{_lALYF;PL>fT~t&&GEUi?iG!a8%Nk#@!;ci
z+p^sQ?K|7u4G&$7L(lv5?RRJGta~fao42_;vbz_s+w_T?M}|Q{J{M;joYBOR32BY7
z267^Ih{Qt(Tx>QpjmH`3RK0H*Crvs#XB<X`VM8n$tqk7SZOE-Cs+&P#eZFD2=;Q(D
zV9|7_0{v!SSWOu3DJ9Z5OKBR$3n`?~){(O$tM{WM(uY{ceLycehQ6n1Tl%v#$0x^9
zjB&caC{5RP4E>sE8pV0QN|8W6^qicYisVBs<Rn$Qi>?!0I20*ojMB7i$9lEq<n)x)
zW<%GuIBkeAVvS`OdQMJGc>equj~=}xzNMwGTy#8t{+x#o9&mAX&VvW1;#sRD1U`mI
z1e0Grui!P}v_?1Lnfcbkhpaan9y~Z@v*|0yi<A^;y{BLI94(ItG2z>eDU8*RQ!Lq*
z%Oyh%v4xn4vGA=w{sG_l;ZOL&`VpsXhv^(83$Qb~0?opf#p)IvAPyzQYud)sI4dE-
zQV2s2B#hIbkx=>!?v+AYO{%tYITuna)gM?av?PB{^Uhlj`SoA>WxoD<f56}QPk$Td
zTYm0WeqOv@W0VTbl;~<`HJ608nvxJHno?7u6@V6osJfqMV^BrpWmHTecFY`pm5R<7
z0&$As24XGKW&)L2g*EH7RB;ZQzA|7UH-i?y7pA~wy%AP0g=#}qh*_LJQp)l$D}_S_
zDw~&UzVq_i+ZSr7OsPVcD{Z70hIIKd&Gn@z6|FN#nYJC1s#Qm2=f$!0UY5%uVcH6O
zX9hJ<^9HZHbfi7=OH!82K=9hgc@}sr@*I_V??|Zm5@?p4mwHyEiZV-^b*cHi3bb+e
zd7T}Y<}UkH>)iZ)6Ifhd#+A?N!18_5UHO39I^*jd?A?HxdgV6zfWmE`Z!WvpalU%q
zzH{6M7_WA?`?e9TxToj;OXA095`oJQp4?v7f89%N4K9a7m!C`hoGCmZ^8(Emwr8>b
z_uK~{6ooI7+<yy`|8*2U!i=ACT3)dBe--#y3U<5CtHAE&QFikwul#0PpjEceRA1Tt
zmixF|S>@_Bd&fvu=U)eW^SkRt^KSz^`&YUb%XM>uJAkwsd2e?)pt`*12RFM3lRr0L
zcYd~JrGcVINnY^Yh&gx6lXQl_h2B6-PLETj#K=%4jDkUp0!E@Fj2olKt4Pt2=)`u*
zI>*`hIR(XHvBWklS{KG)*y_(cJbj3F4dtXT#fheIyz|y0F-nhOxn*5b`;Ek6(TV+6
zjFf8ob$oJ6UxCjIkea5W@7Hu~NADfpTAcA*tS&e`IpJb;LAO|fRt%e-Mcc9NSDY-5
z8OOP}z_Qt_Ia)3-C>C9{hEa;;V#&*wFE}|`qDfe*c=z3RnFe9=&d$&9W=ZEg(=gzi
z<>KOuw;sRE#l;0yYa%diHk>SvS+CYSdia>lrl)Be0UG1PvTZqg@q(k16UNOz(>jzA
zrf6}rWYhO77DtTJxXn!*N3jIktX8yL%Q%i`)9}fY7kuZ3A2Joq)AJ4A|KvGGR`K53
z4+U@+c?Bp1xj;+-V;s&nyw@xj9&aT7DIo3_X^3=VCI-=wlaQjgJXq_QV!(SRoq0-F
zV=LIww6>L(XBL3<?z`{u(_dTj%|HBy{GI>w@6yMSpZU43vTQu-&4wX}AL}$pia>Hy
zeMQgM2l~xg!qK)`&gEQ1mj#4xHa)AA<RcoTNihoWDpFifa%P+YMyr||R|r!i#k_S=
zC|Rm4by1)tq)e)B(^?cTzf8Bhm#PnVVH?KG+jpJ-zEVS;0#3D*#0#2}_SM%zE}(U}
z@*>KmhCyfGr*-vytx}mYI~V6nB5Dp|W$U)@o9g|H<jmy?MyX+3!Ecw0o%EQhhHU4Z
zNX4t<BT0^uL}Lqv9SwG+K+lzBm;6XBMVQOaJY9F5I5h18WH$lcO{V1L%lUHstIK!0
z_N)84Ol&_?MZ3Sf`#Zzi@}>?Yyk7-%*V}gqgRXA<Y6+A5`G$9qK6w>DzRE1^hIKWC
zk>6$H8z?3UnLxp8*WHC<WxKb+)vJ9S-bu}OqL8=GH`k7R+2(R({*`-3DYE$ujQLF{
z|E3Il%cmUWiJ<@M1obm09_(x($Cv@v{H*M|Kek>NJK*dlDBl0|;I-Zh=3cShE0^CB
zK<xvW+x=W$_$rWi+py~Jb9L)BWnWuv0_m%peNMn`;TlR7J+2rMknv75vAkc1rz%-q
z$Vyxapx~+`M&leOM~hkvun_r$B#GYB(6d-dHEd2*0cf*fv0UJtXV|Q1yOt?LeB-3j
zHfEf)vfiS_H<qzl4TUM;eFs1cBfhb$Hye(ZOU7YDYlqXCK%upkSS2MCk{4vH<W-5T
zo*`$cI(61#jX@hjR1=N&tk(sj4Q<mhjsw<MvE$N~qBLXGlDoJN;Y+0zW147>IyReL
zn7z#g?;qg3=ltTFqvIv3ixo#lM{G75hA^@`Uh?wnC669GWOcD7>&&9<=vO^QC&#S&
z4a=h=HtRJxXHHH|Sg(5?oIYf=Iuq?QZP=_gbju|#&tI}!E^D%TqH&J%muGZswcYZb
z5DVY^lOHiwf7?w6{P9Ph@_5nVyk{IIPL7UQb}fobDH2XpM&rAN<+8<@8fJ`vDOayg
zEb-%KC|bw@T1qZhFRm3`+YzRb7cXAYR&UpW_!AGKxF=X^nR>}-dvyF5qaA1G7yQoe
z{5|@9#jpIrFYs`A!bRTj;~#&>ix)3gou4!2B>8A5lcvZtj@#!b<(eCo(l)dRA{!!D
zN`O|*@;mBUO`9u-f%+n*K&g@=NN5qNWJ12&PpJ?2%sQp|eOEte4%N(IlAJR|D^w9h
z^;x8wlEt13YTm7|Tj<KLrCbR>dCliMgBWRxN^_Biw3|HV5M8!IKxRRyQeQN>j_hZC
zLx5b_zK1RW+$E4xl$5UI5^AN;MT#g&%~?`a)<>$4=i~2y8bO+<ddzeeJod^-QsVb9
zP#m0hdy&C)fcC2OFF#)wFx_|jeI3<&&#3LUtA7u5DA(8DzwdpZ?&{pv180};^w7E9
z_4l~_81r*<2|v&1`Hv|CW_Xsizsa0XKLgXtC-btEb1+uGoaEfi{}h&YQ5>=VT3$W_
zTX0`wpL)l2pwN)re-%Cbdnol2l=>9P|A)jx&zETUi`eo@SXy+c&%}Jb*MI(N?br|9
zx6Qd533Ywie+%4y8~6;3xa&a<fveB5&+F42*uTACPTjir&~z_g_mwYxfzzX<gh{gG
z7@-u6b7IA$Bw2o#2HHi3)n;q{o?^f{@r2Z!$6HU|_naPelGNThmTgN+iN-r(&d_;|
z79GPd&~+_s+hTNNvFI2^VcGz^w|M7-`SA{lB8(H>dM?%%EW4H*g~`h)6Juc6Eiq_f
zNR(Wd#z@<Wb=EAc5XOLCc&XmA2IK6`^;NY4)Z(@fBgR<9X{7O9Docj}XDsX0ipG0}
zp{HqDmdj&d;of>g-#^D_i}#+FXJ<To{D7C|FEQ58b{(trng<V0IX^$cxnrESoS&c5
zv>i{MJ>~7kZ&6ZVwOZl5=jqd@JbLheah&8DILGs6PkH$0A;YjiYfaP2eiw@c!!Y5T
zLt}|k<nhBttTt=frel~QfBQfAdwll?KdK|nLM*VECR%Ihhk#C!lQ?t>jus;Dt2EWa
zGth4o!LS7+3PsKeWfYqi8@}`OTa;WR(Laj3MXumaDUre$hY6jeqmM}>ZH8fB7WEbD
zDujtGh1GiC@Bf3}=j_=zzxJ!Yf>(+MUBmg4i4TAHLqbuP@+L)9uUxgQ9^LJlgm|jo
zO4&vU0LZ1XOtP%Z52B*xk=4-ceA6h|VV3gKx_>p}s@mZwikicgmw>9&R?W}d%z%{&
zM7IF4Hf<Iuo_9WnV@vHgW?edILzz#s<{8%_2=rE8uN0}qou_gw1fgqt^88+??40Dk
z>1~@*1+_*g#H>Upg1LQg31o9ol~Js}#XQj6<^{^XjD)OT;m3WaN!^*T+ui0S5V=j3
z<m$TnLEz=OhdZ^K49s<4wx6Zh|NS~!bO>PW2XOl(Q?AdsJof%&yUPy)N_FG^H|Ou;
zT4ta#vsaR>8@b^BMc||D8imVh*Greyt3bKzT%SvJuCndLpjo0>VtJn>zfOZemB?K8
zD$*Eg+b~sjSkmqNe*GH!6zG2m%Kt`=ozF1l@6(nKL4AwN2l(<|!lXZYB}MJprRC=D
zfAhY%xy^y8>~4Q(c{#97oxfdY-}D-w?$*^`1>EL~e;a$YXCJpehFq_3y<1!APSnol
zJ5Kikb`OpgSY4_Jv7rFQSyayCDAj=&&BMh5#xP=yge8YTES`pbpyW)~E!gxUUDtt<
zYP#c-6Q(JE661`i=8>hGw#T_xtvEh9CQgA|6v`NeO;6XhtoxA=C!B9sU#!J=yd-i8
zICo6nkK%Mt6md+bPO;f+M7K=A7)IKq7{6-;Mr%@vq>@QhV=JUUC<zq<jaIDJ7c3U7
zh|@;t_^feQ=Q*jGb8WW-Acjce9R2Eolao`)%L<XB#e&VnisRD<tTt;l!$48+<BxyB
z`|rNbs$a1lHdyaityVbaIJ-Clr8qiX^5og4y!H4k`t^WDalAa@$&)9%_14==qvQv=
z<${;bU-0nZL;7Bn(2v?B{V=d>m#j7$juuNO@X5!YvRo{9cJ_=X&sY4;@Ba~-5U<=)
z893_)K6?I=FPt24v@oR2hE2bw@rJf(I5|D0d;662df@xt`2pTKw1PASrXi4XCRTCW
zR6$JMGG%~Lb+A%j*I8b4g;}nn-td(z%OGXgML_!MV$JWJeS<~2;4AOF#mS=K7k}=n
z^!=J|fA1%fD+a`x+^?hoK-m^P6sWA7Vwa68rAXcxS6RBPPTlsw)>{z-0aOB@OcnIi
zrPPYrIxw4$bz>BN*>SQ3f|U=LfoT0Zm*`i?x-G+?U>j;xAV3%qP%cGeSg6Z&^|+O5
zEjfp?aPlsoE6j}L4sM+x>dre$!qIciT)hrRq0Aieof%AF-(<k3>atv4{BoPjpl%k6
z6$W(s^L^&2?EJf&_5rV(ujO)W{yt!Ib^gtH*E^KMCv~;EoJCmo?f)*vyY3QDuATGz
z|Lf<tn|-zq{#}<`DX+MOx6eDg-<#{J!VWN-?@zTAdetaKzR9`zk10g1OA8BJyB_ww
z`m0=CS%boHQhtfk{O57yODLLJFmX-lq+H6A+-UXQpK1qesQ(6(A3*tSC>QuJknD!Q
zh*nRT_z|6Y56aJ6BE+(1Usn%s?aJ-D7jK)&tzU1<EC=Qv+V`EF>+5LhP}^_Y+0HiQ
z`Zc-hKHz5F-Jwlhbz<u3YWH}*-3!>MTv#kSaZ`v%bl}F3wsD{#gh=O?Ok+T4L&}*v
zPB`aDAy6_LEshz6k+$nF#<J<xc<)(nL>^<&HC$YrakN~ZwNxjzZ6oZLZ$K&fabR(@
zBurvcM1~Y2O(Tq3%88~GQ~9P9gY$kIdHC=lsTOCT4Hv5wUDpyz-ey6RlxeJolu0Rq
z%8b**`RWX16*)!<8RrbvTe60t6d^`j<I!3XV~~8dVUT=03d7K2tzwFSY1}|bJb&?m
z<?@J7T^zc$;q3Wyx<y9{nTv}HmfeE&#hOQt9?_2*@|1Aeaq;|&M~@$|Y!>vZ4UKK-
z&o>;MoM3T$`q8I&>xh%IW#~tqfAj>UG^>jhrR1tPmQbo@lavfOlcM;d{^;oohAGwr
z{mZ*67YHTu;n|wi82Q4ZV@?(whJu|Ec0FO5mg9%-@VEZ{|C>Mh)^~Vxa?BSWoYF9Z
z2|*xMYjnxPoJB{dP=S=HKu6Xq6$}(0#rkShmm}r^l$3Bqf`isKXcT|&dw&4q*ZAsB
zeZb?#5Bb_xzsh&N{}W=aU~sO&+<As(SB3f6ngO0LWxH(7CCFd~QbujH;C5k|oNo~U
z<!ZN8R8^pw8MFB~wUZX95CM(3%<?K`3wY+x<~GkR%R#l8!=JL>Qi~0W5?2?k4a(Gz
zZn<=jkV{ddh@*nmxEnUEU{CGrUx>LK0HYLT{%Ok2JFWsWDb|>Su=6=o22$5skez*2
zP{+<ac%?E|4q7|gHxY~7^$ysb?)q}y2MG6r#Qpo30n%qdsrvxlp*9@q)UU63Gn9ID
z`Jw#5tNXlu?zh3S8HmlxXOOl#U-B%~_OAkT5!Ys!Kc_`pw&s#uw5-%05?Q0Rk|&q{
zW>7a@%U$S@xhhPe5HHW;@*8ydPocI3Q2Wkq9}qh5n-?zU7l4PL{)bTh4vIgZ<WFeB
zL~$S1^F0IgH=+DA+ajhLyWMwhT?a<jkNX<aUAMdU{_k*2?+pmyD&>EDfOX69?{t0d
zI^A~MyYKgYawkf$-c)_F1R;!g=h(z*y54kb`X28aa?T8+NI8Tspq1j}<cMjQ@Ww*U
zT%1cTS;~^I?wf|y>RcF^TsU8yb9{UxU@B$Y!gEnss}v*a)ryBF4**HLZoFffMlljE
zX&WXxyEqqs76ztqq^Ux$+Bsa)QedkmrmJXbh?#LpfMFN|r5Ij3Kf_r~iU~I-(N{gN
zFiiv_qQtZ1SDXT&A92RX-$g4%_G6@TEgydT2~YJ4tk!sCp%hGn@#O}l%+r^jY!lNz
z{lUjjL^dN#Qh@T&_kT>zg<+f|M3)t*q?*qqAhmYC;`kx4FN`*zR89Ub=prwnQp8S5
zkqHGhtJT#a#@&0N6hg_o=m+{w29DYmZ!MiQeBt3knxjYjjsM5r<`2I8J*He}Lm<Zj
zT7(5;hpIBN6*TFBDteAmUWOyJ+Is5c8tT*)@YEYqUbA_A%~C6w4RM<Iqi=k(%FHwz
z9iMP?bi|9Z^LlPY{7EG-bo+9jfm6BmFwMGdm!ql6<ulwcR=SutvyoJmkqo8k>eUM>
zxaP9+nJY(UF=h#iUIIS>9l|K88t&ZY0BN-aXQm1*rz%PE|8MVIdnQS)GqLAHWLCYm
z?w;<R8FGdk-i8#t!a|ll*j`}03tx8O6=44m|2=;);6Gq!VF@Bd?MftvB01Nd?&<FL
zT9ug*#~;o)5s{gZRrPjrVVS5Pd)~^-$jG>#C(e~A^V7`V9g$hBs_-f8<7%AXq}QRK
zkVveSt?!fGThR@aDPabv95XQ2_GF^TAUFZT07Tr&S=5|}3y;cVHmhL0N22ri_9=^E
z$6TU1r)e&RT?R3Sr)3~G=UkQV)QGcO*En7^usY?^?GlNW&!<Kf`QH>wO`ma*XAIa#
z*Yflkig(@Q9=PZuoijdliMCnqM{IHHe~YM5Ud2nN@s)Gb0sCh?k)faE2>?2XL*iCJ
zU9k{GZ!OKy%Cpq<<^2fY9f1B<0RIlezX8RsaN7SP(EJ?iz5>%P09=CjVJ>Lz^m`jH
za-Y}d`{qIU5B=EAVt18AlOZSyy0_))+w4O{eU=^_knRQSx~{|7`6+sJO1oaK)G$CI
zAat8foZpg&J5Sc9;=w-jI6pnX-R2fCFdEn3_D0;GFf+Pti-#BI0+<?y^NR~~+bzz{
zPT{@B)y*x=&M!pYs_)URT6BGfcGY6wfUetOwO*kQ18z23@a9fLOF@WU(==$>Rz2s7
z8w=S09PYLqZf>s7G#<~Nzr>(UiEnQ2@Z}fJ@a*fC@ZRD2>IS#h*SI)8!IMYtfbRlU
zs}-Jo`W1#D;N{C#==&ZQXA)-?%ov6NFFyZTtRrIpb3niCB?3T<Y9t~Yl24tTu2^)p
zlyebdgiCWyw>D;YFF+)+D3%jL+=V#d$}sc<OCnPC<lu=wYCsWUgrh8GG<~heQ6n&T
zY`S62yU>kO3<M$q12;WzH{8I3aK2vSJKy~Q==1@uyNFEyylcQw#Cy@3hz-f0z|O~V
z-$ZV*sYhmzIvaL{5kT+&LLojXbhJoAXoY@&Bf=0FfBW-a;K8GJaDMRs4<A0ptJiXX
zFbhsr9W8TJK;#98spyoWL`KMftN>%A%TR%tL|!q{qy&JCIFO1%$3)meK$&xHOq}c;
zh(b3(fR&o#=CU{mI#a0xf_KQ^OvTpDiNS+$CIt``h8YDn5DLNsY+626cPLhUQisnm
zaVFs%5Uf;}j4c=dGbOGqYo8@}VEcY7Z47f~Bw$_CPnf$hzurDA>DV25lCITZcfIDI
zYYqu%^1H@ESaOt{jX_F?l>=Ydv(l+;+G}<^M|yt@@a90Qyf?l^aevwmwfp(sxUIGZ
zroeOtYD#*I8*K0l;?>chT83@q3uAtrMHDXkHDDa|bMmoz1(x&g1t<jI450rCz{ha>
z7vTQC2b%u_(ftaZUjldt({(*Kz0DFj2vjX)s)5hiOh4osOQ|et0j%+kX9s;_ntv%R
zf%qL|91Mo<3G4z37p`qQPS<NyycZuj8x7nZ!PsoJ7-GaXFJ9wjyTQqN1!iE~u5hwi
zq3?UNrz;F0;>9=L;K9WM3I2V%#p(GOE?-{3dnc|pE^k5XvAyf@&8t_q+idXFm(Or|
zdV&xl&QDL_K=9t<ZgYp<fA%>dBHn-ReYDP_bq+B^TwY(ok%*igoLzuJ#I3qdxxT)_
z`Pn(zwng6$*mfQ6Zp9bg%d3qlx{KR|e%RovS6`%#j)g@h2)M?pWYxqv77P`SK}A=f
zJZDKZ-UtG*6Tk*JZ88UVICygp0Vf1hUCjiJpa3+cljM|J#ag1?_rA%TAC@?t012>C
z-(f@!@E+iR5F)yPalP%+wpL4}n1vz!&o?om_a0qfeEj1d;~zfx1g&pDu7M}Y?l_E#
z#GK*^8@nd~mm-<<XN;Wt&UEX*2}oI8(i0NV?9J^4fBiRqi>7Vy>he~4(>b-swKkZM
zf&gYjs3F6k!_5FW1<a~<M(S3?c@RZa(-Y9`KnX00$dgi6CQ+tV*bEMsWo#pJ;(l5E
zoLHqGk~r@zxfLKIuc9LunI!}b1KtssJcbxFI)qa$C6bXv1ZRNHQPhA`=PnV<BpTQG
zW<V{n_|7ZjuOUQasdXlQ4VkW<5X-`8iuQ$_9FTx=5;@!V6htC{Q<}|}H&1h-UbaDg
zO+f|Ea+U5e=D=+Z-7;9=;yGdb-JI9%*n3VvMhzU55lXqZUjtnwFta0ty;<g;KFjxO
zfUL}0ws-t~S;G=6mfy^&MtqkDas?nTW1uSxxIuY@7z1b3Go5`Ks=6by*QeITYAN%y
ztDc7>#`y5z0i1*IQ(*Yt;pr0q?<39SYBi?)wNTf$J}d12bbHG_^y5q4*wvr+olBMX
zsz7vdx!8pJj`{u4y?|Y0;G{i4(>UDTZgIEi(0C8N-k@n4aJ)m;bzpKBLcrbat%`>l
zY;QVjxWj8lxW2x^)5nkT`sFKJoSp;y6<+lNcnEkNzD~Ztyz_`dK-c%^x&hrFI(6iP
zA>fE`*)$;2Ap+1Sp}zj&MT*!b1R4MsTV~wh5*|X#K!i02!{se{&O>@aUN`~jB-Rx6
zqgv6Xsl-{VBw$Gh8^f#{)yb}cK&O_4CgwBpPEqx@f;f_8jv|grfKX0|#uo1TArUh&
zV#~te8lnVoca9U`=E9tlNUa!2^+3o8A=cmI2%(SIMzgalNQoH_NaJTjCbERdy1Kl=
zJ5Qfr)p(F2G;IUdIFOE;%qP(^BS&bw0Sy(ic@Y=K7!ih;mO<l1hbw|X-X)@+^wFG$
zW6{GikZd8vp1*jNzDrbatduK9u+UjDR}yrMdaV|skxBPe6(o{~=Ut&IXW~%nA5kwL
zfOWR$7zn3JiZ_sI`iCZ>Hoym1-<i%FB@w14>kG=db4e$Sw2cI2G2Dp6@T7Cp$SB!J
zy5z6U`Uh#AQwkGi960DYVV|)bIV6r;WQ|Z_V2}%4J;;H8wKub&M4or0<7DeBg9^^c
zDtF`%=?~+_ZtU(P1FiBPLzBN5tmNNgP*(!BG1xOeN3(J=(3%6MG2ol$nR3NSU{$_Z
z17amun*zG>O)lzZBQn32%bV6?Y?v?wj#EHKqdH7={>rf#*K!Qh^IRI}H3Dv7K*`K|
zQ2Z-t^K`G{(K8-fu}*bZMg)EdI4ps%@z=*~B3o>F^$_4cFFv{p9PI)r2ZE}%27Yhy
zjY?lo?OCmI3;n*-zp02r0NC{K@CNTpzgbS<d-V0afZg`G!&P^M?RJCfs~hyg03t%$
zG;mJBcbQRz)9_Gj+o0=j@cN4v7(&#rN*y*Yt}yfguklj3SrE%dKsY%>@Qur+ZaQ`=
z5)qGA0uX}$9D;B<JV>k~2Qm9&63&!}AO&zTk!pEyaT=3fCiVo&??_OfB_dM*6H$<D
zOS;jJ42C&l6(FQrmb!xpAtZt*#7&|FnBJ0Yst%-{{j)esJ<X=rnxrhjB#IzPz)7qd
zrBj>M17HPCQHZ(|kQSnd*-W1=GGQAA3_EnuOJbXq_{Gh~?ad8dY&UrJ{43EpY#I=+
zM&WO)&Z6kgslPxLW=h&i3Xp`(lj!Kh!UYBaCMJxmz>)~$MXYC{$8u2+GN_0^a(!a;
z0mg1kq|T~`s01JpC`EqG5n8nrmsq9}ztljqHQG7hkVaCGIZuQ-G+qOMfR`Afx;jyE
zsgAIRz85huO4kgM%-3Ed8CSukdf4H79La_n+A96DHzOA?lDWP)+)DsVeaw+UGYdr)
z1yD-AW|uDp%9F7Qv+6bFZ?kp;!=(w6tdCPq+Pv5P;(X|Nn`F)|mVuVts{y=e-f0M|
zk~2q>GHbsHbAXj$MGf#R{hfb<7V?(Btp2?G>bNaj)ET4S<2<O8VRy>%%Q3G3x3WcZ
z_%&V)a}cw|b1*s>|BWiZE>ni_w`?58eMum0#-MImjahwbc@6=6CB>&vHo4@5_YO>)
z{&UZF7ShtvJm?ckptig<1@TMoPoFQncfdDy=0EVAomr-1cldbR3)p@B>1SAZ3H<Gn
zFEC)h1||w}a0oPDAPKl0x*muO2akq4csrnBLI^#~jiP`W1qLh%^njf7oy*$;D`jv5
zwPX~%8gg>Kal-Y1#mOYG0EB^-7>n*F{e}_<UA;j;135w*OlL^fOCf?v#8&|bLx>Mq
znQ0tUqy~r9d&Fp51p$hrU;r@!>K89F#JJd1gyK6#YPBd^i&M^@L^jA=AOk#L<BUbs
zYf>foq@N`ll-eE0E5x|!Ht0+7;SQRaEJk&Vy!ZJ24}XBBUEx1}@*CW2HacfuqnOHM
zCRNfmC3|tH;bkC2FVMNf%>s#w68n~IsZqhI0bLU{Is%5E9HvBC3f(RS5TMHU9-ib-
zLL3sQHvKyQAlK*|Q2%KJ1v9F5MgTbnA0r~^IHcJt+*m-AzRk+i=2qH@leI4rkr3k&
zC^(Yw?z;{Q&f}^kCIH~c1C2vORIZ+u9+5zOP6Y48Xhpzq)OnMEi5%2i0&6mlbu2x>
zMQN2CP^`Yf-8O@VJV(W8vh;x~;?`!DtXkcI!^dDCN~8<}%L8W)tT|_Xo0R?7je%SK
zJ>|yDxpUC#`bKO#@GSPv<=fvW*J{5Q6vh4gZmq2OQl10Xvb|G4Rs(%YV%h2M63o@g
zEWz0nG>!XmTF$iYC9cW$xOmLCrQadYkoG_u8#9Wqaf+i+ngjj}h8s9OLw(%K8csmy
zEc-rSSjk_-usdG7?^3DeKLqTQv~XYEdwqQJ9#`&7-(Lb>`#@Rc`MxpR*`7W1vUy6A
z`EidiyPs8TWCY?ILryp2RkK0}CQz{hgF5=v>i~vtT2RcoASNCKbaU3_<-F4=5!K(A
zY7;52vLzSoW4h4`QP%}e*~eC7RT?J}MSLZqCZ06KZ-V8I%Bs_Cm2HWWR)JF6)8<x)
zLXGIyCKr|KA!0t$2~u%rhzi716h@hA1XD&=-7W$F$w6)Q^q=&l3jHV|5)q>LWMi`=
zly67n{+VuIFlW13_I3zRd(}B1*rm@D8T~Ne;=uy}I86iZJz6gc^ie^L^sz%^HArxf
zHbt?JWD;W&@4NyU)gL2*Cl3RHAi#Tx{R#llkTB=6TknN)3rgZ}2bn9`%GCkn!Hnd?
zZPrXe-sTnP#2827U{Yti<eYLFMFS_ssPj39PNTNRJLRN8V>4P=(0VbnzmN?!z#4S|
zfN?u@U0ArU#`!eP1{|$9LBTzN5-4H?a|R}5Z_zbC3Y3|_j*^pMB=^fK9AqZID?rn=
z?S1x%2k1QKkahYH!O=LfNNo^Kbxi$X7P2MCPP&oI2m#_t@~smCiGNyxvoX*cgS<KL
zr~$h<Aeufa#g=1mgv$LqqNUE-6wsCBk42xk+;Mr!Ky3`>#^0H~W6#Ebt_~1)mo@Gi
zq;f{d=Ia#XOcUFGA~y6n^5IbD60-Q!C{0;l2|05Psab;yH2kE<Pxar6+@)f~7TaJ8
zW$E3@i?h5tioS2Emc64cWfq>~-G{Gc<xewAzDo=DYZc+?y?vF~2i*37`!ef7-r4=b
z2k4$Uc9|<-;!gnRhh7#XF&H7~^d!aS#4U+QVc*_2(Nl6zsvAFugnL3(mNYDq#I;La
z_Q{x98bQD|Iq$MHqb&s0m5RYOWd%oEa8{J4m2<+J&k_TO#;`9Y1Dh9d>mSp#iz*s5
zQJ`{a$ciTGmegO4#tYZ0&$H~&U%OACfB=yo;Lr(o%k<QIBi@{h17nMa3(f`+As~uA
zS`Yx)$B3KTTQEyNc7_4wtdlmjoXm*;IA;9%m%qa4`8jT`Z_s#;v(_iDDI=s|t&}TC
zKoZsZ#`Nv9{S3IqBZeqr%5uOUCWI)44bF*zmw<DrmxGAC!Pf6G@A?q555=&Az#45t
z$4kp2K*i>iz_;m2M(fK-Kg2;e$kxL#1|36foAUyC*ye1sfzYX>q=XcF*+E6;1VKF4
zoA9U{0Rw5qAw!yHx_zl`Xaa3^NYHudRcBE`pk>1puLmK~A&T$>RRy5LO;c7cm>^-n
zSmwVS4w8EkT{gz840gCkz650-At#4B(;-ZvSgm{M7}}AlEtu=Dw@Rf|Yzf}R0B;K1
zGC;sA-xNa0uQ7<KgQE=emPMasX?d)X@?~JL3!GV9ryRbeZyr{k<@YiGo#wZ@rC7V<
zT4LdE38F@ys3b<DJKW$?<nbAg<#hZ?actQFGhreCC-^ozp96`Vk^3;6TxRVotF?!4
z^X$jM^F0OJ2f*D6>{%NRc&P?%XRj@Rr$az3E%ete(6$c%*4~?rVdd4m-<xJJV7d=H
zE~k3|JI8r~C#S4H!FkK4D$<IL+$O}D<(5r+8m|Gc8XshCp15U*DfqgDDw2&JiTDr!
zaR`V`Ec2L9xUvx9?o`=4hN%4~_nFiUfN{|xCxYs{lYdSu3}XzUG;U%vR#B;P1xp<%
zGC)Pj<KV2P1F}BT{!@p^AmtAcOHhT#Nq^DAiP@S|3AcA}-h(vUmbzy{qcJxbz`AWQ
z41?B(Kwf?5xfTpU*LAq<dM&p!4JpRHmRP=X+O&v$kL|Vt$B1>);>0&9lDEbZ5H(nY
za=}cQKY?x?U*AZq-sCu#S>xUkz#h=p!g8&~4=U;dnHZj7&V>zN2v8PIV&fe`2;wFO
zK#Vd+%tF-woQS%`T~1_yWd^7~PJuZLeZyRgkP}X0ur=T^$T2rGhz>)P7|c#ZnZzRY
zWfMt95yQH4SfXU;P#T0lb<PAFsTD2fLj`m`+emsx$~}Y}zPwm4hbX@I<j^MlXa_u9
z8u`!xsi!=XL!8B)Ww5SIn@cK^R?Z?^7Q2=^gmI$tyffhIQr;YeqCPF}mr6f&H83~+
zn`=9LHU)bbtmNNgh+w~KKyL~30cCHELEN14H?PMoKsnBrL16uzDcG~THSmZX`N!WK
zx2p!~re&0=0y6YgkJs4ZWg4RlNo<Z~Zk>r^$KXzNvh^R>;f#L@piS-64Pc=z#q};Q
zP<vJdK$VR9fZ)Dw-z$GP|KD&SQu@!HkG%~z+XdRD2ZKpnXqsQ|9}Gx$rAj%Yxihcc
zu5`c16ZQ!^ErT+sEWcf}tF)p_MhAp~lc!=I<F*t_I_f*ZFe4yDu{0zB0i{>*7^C2T
zgr_JGUlE~F=6I#=Z~{J75A*J?4TblF#<lR?qwx*;VL;z^0sz#}tW%L=&=|88)ZWG_
zAPPs%0IJqSe9?)%5|7oul6cja{oX65NA>H>@WFHg&8c%7Vybr;9U$o!D#yutFJdzz
z-0GM)GNa>&JLGw^G)+rUwB=g04bDza;R(3BzJ?><to1l`UfgdGfJlma#UXNJBQ`p(
zSd0t;tH{<O6*v(iI?AGR?+8OLTru-m2dI^-^D1T+Lmm<183)QhUrbzRnI{@vOr7SM
z=soFZ+WdfuPStQi^uwZ9{(@A^Go+bjV$al5Dq=UQSRR6kC{=o{&%va?D2jz}(01rI
zW$p)njLS%A-2rM$L_-p%hA4oKqs+^wKy(NUpikg6uo!>Xny|<g(r>JDT}iP{_9Qab
z7++S$1lSmGBw&w|GM#YdQ*>I&0<dBekv=T;64Qd?w5A+F)&Jc663_hD@#uF30&`AY
z2@c9NLnG)=uBT-XHw9h0fYubKjm4~U&{YRyOJ&<T`@cH}ZMD3p-W!9!@wdi(hVeIf
zT<)Cs8NoCHCF6ACeuo~{MnQ_&mf+6XG&f8#7cx$NwRnj0_!r}`-o>Sx675R0=lAwd
z_UdEa^Zbx94tQz$-gqh>kVwne1-y#yE#}z;26t6*-+d}B_fp1%`<3z+!JOu!L&}?{
zdq$oZH_5g#)59+H4WNLm#@)$VqAuhjlm`jjkmqV3A=1gH3g-cha%&8%h>=9@tf<^D
za$emzAQR{WAS8bZP|mh#lj;ja#1L?Pc8-4N(e(qi-4=nf4xKFTs31>u<4lnrqAO^M
z_0H?J3xXcm^b(Du<FfNu4nlOJ^ryNc(+C1AVl?Fbje*7WjETzy!OHfi929SU$%;hI
zq0w59a<;y{zD1|X@Y&PY0?{ldI}v=_z<Gz(H+cEgS8&eZ;mHYBUbtD_IERc>FEhA^
zF~WIDLF!p8di!C}aZ=4DLDLC9I29oh5t^n+#sU`GG;u=pKNMrqSxcnEfFCHinTb)<
z0t6HlfQj?wZ=7QhSC>W^W<6_Tq{NyWU}9Dv1A6U8TUTDy4_N*4NnBM2?<I%%PXvU6
z7^83-1(*sKP29#fu>fZVV$ePZSgl)`V-|<%d`VWgF!v+IkygDo{Y?oBYaV8#&`idy
z*8UQGSW5Yz?G$4vwb*qI99WH#BJvI~o@#dBOLLeoC-85avf;!SpmFh+NB2sQSAv8L
z;|!ef<QggG6vWlIZOdG>I%undu4O<r1#9E`hX6rTpU(jtFs;Kf*Kb-b@^hX5KF&{3
z^xORRfM@i7gK3k-%7mxZniAY;0KicVCANf9{cnde{0lVkd-|!eRue)!BIXmYu|&U@
z004+UcfZ>%dOrgp`^tNRJi97f_Q9KgMOx~^S{0}q{90j^GVTlXrsG)Tn`NivqIV_R
z)9gS%lQG^0?2B|SU<aT$HA_t+P=>n9b-)<m5|9+j0TwP1Ba0QaL{j1kMPs6emvxi?
z;U=*9t20+AP;LPX^(&!vBF|F7FNDTL<#?$GiR~Mv$S<Ep(|GiKFJZ3~IC>VnD<TKi
zIH>PhX;o07V1Di*l9gt;&e;0VqDpAzng227+*hnbGK9ijkdd6jCXh*hIc*hAmPh1J
z=(HhO@v?FiQ<MW8gAf_r5b(|GE0i1dXk3{Sh&n$#0}<is=348(c=+G}eB*HO=pkPJ
z{Uy#;Egqk*;Yp1=v~$xO(*B51>cowR7QOeND1O?UiaCRFY@>pw#5pE{Q;XxIBEagd
z!&*UkUNj`&FB%|$aGM&+Ts4<;8wI>);*$3)@5~xi#Q6<5aF7M5%|YWm;DiiBGI6kj
zaSrm}+5A$D7iAW%j2l7pR}-X6;ltSWQD(#tV4=QZjBr7%SqlqX1D*m3%T3Gl=8~o=
z61bD812JGx!ylWU(8FEQRi$(&Grh%3?&Vj4y(~I;vR-0xn_}F)GUlw!9jAvqf^fZ8
zO#|P0M0nuWvnhMelnMHL{`2U0J|tSZ$`LJL2#WhW>4`B=n{(SrIY#~26e!LBP;qC8
zV>bs|oZc9Nw0gcJ5SeGvQvMS3&HvWGVzHESJGp$b*l;Vfu%lOajh_LCv&UmTET^v;
z_;WVjwb#T6`3eu>e+Qt=^rtrZGcMf%FaYZJ7wP)Ga`pkE9ckY;tEmQ>c0I2Evz_&t
z_H7O1?gH1f4;%vYmaeDOS$MwmS~)%m+{MgGiBbM_UHyBNbZ=mnPRiO3!c?-&N0%QL
zma{<S5;3P(tgJd@R#e&)QeJzI#V}I1ipms_q^*l}(FuZ7&P91@phTRK)vJoP$cdhn
zBL$jKtQVbk8b*rLN)K?va4jW&X~CID4mv+6>;9S5W!i~I9pxEeWiWgOp=QMkR!|bu
z9gBdm#yfZ~`h5bX75rKFCCXfZVHi^XiLW*Y_lY68Ze+#_0Z}cO$q|MypbHkfYQ8%f
zH_MU3I}aZraK!cPjcDQ#@a~f*DTw^-%@sa)c!4u71{LhYI+}`cFok&&gT^4%!y)uK
zSLB&+qXbYA^K45FLr}50>39`H(TVO}2r=oK5ev8==Tw(X=BK>}9rwl)BIka#87GWX
z712;($=r%Qpl`e$G#F92#vG#>ENHkl=Tt1oDq1#caEGXXD@H~b0zzPrr{vErS^17Q
zS8_dYkn%KS8f6Tez}SeB?f@d6Kx1G=D4aZ4Ut1Fm+QjsXRV=GUDx5)}_|ijKqiHD?
zY|89YP}VQ0T1Mj<jrpsV)oq)&qa1-S45}9y(T9L-x53*cnOOmOC5@kI?}0|Yk#bJ=
zYy(pHuarZRFHC>exp4NbT-5+!nP)$MLs5@<{yK=O<=g$8DX6Q1xf;-$wh1FpHwB#K
zpYaw!33Ls(LcsN?ZAc?&{oR7mgAriL8CH`u2rDhzfpLZ(;WYfq2{xI4iQ2Of7_HU0
z(x#nx->{s;`@HA*As?@RADUiFfti7er81TP*dckA>b4K)?YJJ@UMRC1Yh^wlMj!<J
zJ0Nib?4Iq>Y5<7q>q<N8>E6ID_xE@sV*PK9kg07{1Bi5kp+pC1!GH>52GfZvKFC}w
zVB<KMK93S*Vosf!bj!@(5b{mqD6n3Ok33G_0O{tWQ6CaF%m})I#>|B<9W0l9=P8HD
zn1>Q-3NRmC%nB?NsFHGAD0gS#iaG@y1{4#QLb-9?yTq+wmFe}vAeNraAqKTc^pp%Q
zm{q?{^{BE=nYrr#R$lbsWFO3;Yyfz8evY$~Q#^a|0$+XoEVmQD%D1>UI|s*z%d2ZN
z#CUqThELMFf{?4Oq*Fr#XMK~baOFBM#E8NXJbcpi3Ly&c4klujkzTD<=(<jH@toK8
zrbz+>0qJlK9_oaeb9_{b3<6d+G|o#LR|PK#jKDyTjhD8MVp_LO+w6!iK)B3cz_~_@
zV}h<4t$W~zsF5#1FFK<Jl010?ousCM7o0lSCKJ7qtR)aYKF`qv#iU)(d6b+4M;0)p
z0BbNVpprUAoH&A{>jBBon+q`(q}5D_gk;E3f=}nfTD|cC*2sxTv~vzk+hEm-^7$|f
z7={6xZHEvTUAMv2%}t7B3l^P0;|hit#b4oDC)WR#+_;j1#gl)M=G?mRE|%XN-<|5O
z<)>31h|#+lYV7C0EkEB42$w<K7#P;B<#)a_26fZ=)XJ&fv$;F&)&mHAIS0{5IKZ@X
z8Q<CJa|Xggboe5TPa<I;w8%z;j$2%W{{i5$$~UJ!Xveeh&-%SL1$mW+OF-^kVDXTY
z>GIBUDE$ZHc1}T_-kbKz6nEu9$7QB`gp9;#lx2KT)M+GQUFeLoO>q9p1z;NjF!b^}
z1SAWzB`#lC?mhOm`vbe;Ly=6BnHOTYAX!m|5g!1`I1>U?@?A_cWCWI#fC;cDtj~&h
zV%N6Fq7ojCA{rtM>BJghD=C~4Qeb(Ifh^@XFahTGOu0;?^+>=GlvHt?Mu1S}3>ZiX
z=^{#^<X*%AAPB)WUWWov7Ck!<M9)YX1PiRr%;M%EM4X(TV!Pd<A4GKQtOr!Qn*Evq
z&J&uZP2~`I1*QUAP2t`<Mudo%pFX(2H!rUQ$fzZ9B%pO3AAk2lG`_*J7cbC<LJm#9
z$$Eve(-W+HgHONw5>GGA@L=Vo-fC#V$o|II0S?M$cE0fcCl;F#YJ~{^5r~LrnnwCa
z0km(ti0_?42#mYUR(#BPmvr<b(uPChya2f&r2h5J!;wQj3<!`pDf)FnFhgXRuQ=&e
zi{|0d9FJX8(kH;KKy`5J4NDb&!$jwk@(gz_0r(gOvBvXW#G}f66Yi^6ty)Mhr&glk
zswLNrbUq|vPNK`mR2bV3q~6KWRBoA3ji^W1;fBm>kVa|AqC!Uo+!|2h;Tn&|dogVB
z0yMo7AB15TF!Te4Fkn0M82W$^1{e=4LV=;~A@WDLkGlRt9MU|`7PW7i7^S<44C-tb
zvLkHCjj|;$2Xa$x-58LKLC#p*I6p+xIb?)93MjrY2Y5A*H3n@{Fjf2R6lB&x-WaT+
z`pz;qtlZ<uH>bnB<R2~p1jw4i3J$Hh3`*1k41!043;Y=dFgko%l$N2pOQkuS@t@%&
z8BJ*J@_X}&RO-E)b@|>7kawTFRWNgCxwV^<b`y}eCZqgR!CljQ1^D9P-n5?Oty)#e
zV!1X>OO2%rxQK*KS$QR3<ATc;qX00DnFE(Z7?sM6fFc$3nk4(?f$9FDSgLiuOlzC)
z2|PhKWSMXwiDkv@11Axd5DP9uG{$kG0?65xiZNIOU>^OA_y`BW%2QugPQ+@Y%Gshl
z#fhsza7hu{LHxUkZWD6Hl1XtHNOVUzl|&$qBoYJ1jOJL{<Q(ZFQHxM>_UZ)Y#zFnA
z`9_`D5@6z8Vjx7^-QGgDqvC$WJ3<Ib-~sriku8gaV>R@>xb08{`4}Te{rh=}aNZ~G
z9~|)D><o`?x47y$43QCZoIZT|2;cwqd-(P5#Q#%ir3eNt&Mt6vdV;4<p5hn(>F4<1
z{0yd2CUZxqau%xt2!KS};53fT2@s4SAZp|hiv=470j;fa9RXIiDQfV-naIb$$-mz)
z1S!)(PpKGm({}<SVg!39<3cu{)H*Z_YKaSg2ty2Ds3oe-Pg}<tf&q?69c`<An?(k3
z1Pj}S5Yk+fiL0UO2mpj*XeJ{fQWem_rlwacO2;BrFGZXJcm%Ao`d1-v+H7SQGGU7`
zq&JBOjv0tSRSk3vc@OVBR;v}7W`(wGVAfMk^+rQv^f938dTcgZ0nKqpZc(D1QpaLU
zN_w;KHerfq#HsvbQ>1&w9|F$hWXCPr=j``X{5l3h8E}k6tYgYE1#$TujNY&RMk=8u
zs+|JG{Qxh&U*4P6XZp;pYa-JnfLP|6(~c>qL*?F#6Icax6<|^WE;NAPi<au477f<;
z7?1I<@EU)WRz@CoqjHiOeu4+_FF}c0R}$p~u(J=;apmSBAUibQ^j_s=t-R{30%+B6
z^c2s|fJ>2Y*TSu0bbV?1?h-(o<yk6sT4&B6%{XmF&qaqQ13L2wTk5*hzCZs+W$Y{S
zx4ix@1?+MZgv?n3gua%dL`7GrBBuWZWburpM2E-#8RbQ67LCqgiaIza?kUV#*0`et
z<CK`rophXQ9H<pvX9C15tPw;>A5J~-8?jNKsZNs9d~k%uHwZC^@dn9uKp;4ViWMOS
z6IKT04N5{*k&63#`i^n$oHrk0oWKW>5YYsB!0@dXu3O~PF9HNrL0{jGArpCG!ocDT
z);4YrQr$xN4nx59cH4nNj}IQ6<I#47yRJh-#Gn4@pWxB?DH>0BeSMuAF_-E6_uj+1
zPu~R)(eQxt<_wMnR6&hZypFMPGo=qCx`ae8G0|NiDl`xo83W7@oYa2^gX*g(UJgNZ
z3Z?9z1}!m(mbFCni^uvsR_-pCQ?v%wF|m1W)}qOhI^`E%blyowEmn)tpvE!j+d5|~
z{sxo!u!FjFApv!w11DvB*XUA^4J{N-hMhM~sLXXziYI?Fmo!EC&^2s&i)n3w0CFHF
z^T;)#*LJd6p>0+$V~e0mEU1CT5C&|w8}!3~uI~}T04V5E2#OL!Sf5$HiFn&K4odxO
z9B$i2kT~G^&@|eA@P|uTK2VZ<S$mn6Sfdg(j$Gzu+OtDI92S1hLE91tTn2bZ^}y_2
zzTPitU4GyGyI-tYD{BlO^<Duw#-MIGvnD^Qbdb@}*NC)*!wF1N)WM_0_wf$?102@q
z@dbb~l8LOOEP`F*5&tFH_+w-uy+nBJIaTh}n;_SR0Il*~1zc3$+mWwUui7^bs>{O4
ztbsb3JeWP-@oe$ETK~l?{){05wFKINaNJ5Ax4q+ZkL%Gh!#M}$|1Zh0D?bp~7}=y=
z{fo*Jm9ULG8x<e{x$&Uz?3r=UKw?$s3ecFi4a{INhqVmy1Wyg@q-CI5{MWLI`P3@V
z(7Cn)grv|O6^t>9tCb-PS*%3_^6Do|jTtzBIyD_}s_@SMD|by4sA3Ysxg;t>imU0p
z7Z4Gmgd8)SK|p|`8VQ6@#3N+^NZA-62ab1+FvOTaujvDF6z3FSNbXq7A@hnD>(dpU
zo}b|1!!!Klzxh}Ao4@-7UTrqGzP&?opr=5D06cy737Yi^U;Or;@ZR|aTIYnD=6wP+
zG2255m>>a^o}4()1tVyTUe-6nh($wB>0EKz8O#bGhESk<()B9<U_12SsN7B1fkT~K
zlS|gg-VvBYG;Lyikhq}<A%fII)9G_ZP5~>D^|dfXb$+SNo5V~tr^4n>P6>5^5fWH7
z4zd#?BW9VqfpvZnQj`ucjF9&dktKjRhan>n+NMcj?51t7TCL%nhjYSxBO>%ek8R&!
zv)y6{1McoNxaw~(3_&As2=_KJk}jfsJG);gn?&hQGTl4MWuGI<K#f8Q&K?X@<@>hB
zb8hS+Nxgp(2hll|o0`mBGs~)pSnZ7($g^_xfjFA{u7j`f_4HW{;O)9M28oAUkAc~)
zJmYWg$~OjwGQac4&OzOj@D}T-Oe4_6X!#+CTSN+}mkACn{skH|81OZ2@f$>JL2$4b
z%mko;LxX^G{xh6~e+zT8J*i0X?LFXgR=;{NvpltYv-j=+(hhxpFSwXCY}$UB{I1<A
z@*DG@CdQoAwfq+|V2g#A)HrP=Mwx-F%5}9LEA?0eaka$D`KJK2HvT)(o9oyCP(>jC
zNB}P{)PiEc_Fv9oID1m)qXCrgLb+xM#_1DqXGBceN=X!CPHUMZ%$6f{KS>0~J`qk<
zf^Hj)^GzCT6E!=pr!$7?r5Omz^@<^8cP1<`NR5le!V^R3euK3QDgzU)8^@$ZU`pm9
zMhs%~5Mol=H!fq;=m;g?iyROJ;nX=2E@T1~PF#POsLgrNn=+tje*272+E_ROR-W+A
zgL8m^upJQkfSc`B$Qw(;2teZ;F4k+TiLrh85)W5viGD$%AWE^|su=(`UxY;H!+`br
z1Vi5e<i+&`OIt`SNi&Y81SXt;2L(7Gs#9Q{BhuB4>RTFDDCRjx1aS=)K|%G*Y%P<3
z`b?2TCou&5221Y?096rdWO$NLZWi$(IdWEK$O2Qmf>Hx6nJXz|M%Gd!08YLDRu?Pa
z<m-=!M0e18tk-M!#-nLGnq~#xieqgvir8*DY`ZPC-3FV@2E!21Z94^R))oQUEJ}@F
zjhW0W^T;~R9D^t4#c6HJoSbszvwIp+7e~4O)0#3dblUH-POJ_{(qFZjjDTZpzkj2|
zqhGelvcA|<4=w{O+nL7TYYg~yfw*Z|R*PL=tHy1czq|Y1ezEO7E?;?#^mqPy3gDJN
z-I(Z&4ne68fzjaa5@i*M)C-;Mol>_mczl30BDQ#jh?`Mg00QyWAb38<dHi1hoXqPp
ztKygrj<eR@*$1|!<?hQ@{MdyHc5(;#fU>5bY~g;XBwNih$v>)BQI8}F4a#K;nFCjC
z&o0nk1K&k`W`K((c`+~hpzG0QSI=v|)Bbr_y1ytEx*5Ylsi^pf#ia$5h)5W(`J~Qb
zMX0D!iCN)TmB<#7k~ab@cs#=t#fwc+iBIVc11tneTsfEply**Jz;w!B`~W0>Zbsy6
zVViZO;22Ves;o-os;sW%5{cN#HVtSw(q9@l9y!{9e9Y=E{o_-}FmYMoA#Pv6dBb<w
zCKmRIgar#x#f9SV*ZD^D3?XhroZ|v`MmX|-1G*sL$HWR%#F7q##yhMUk01W<M>xB9
z1f~|-?X8M1&8kHD`@#7+e)!#QW4L?)y1hmdS-GN8rh|xo)3~vUUKwB#K3>=N>ZV1?
zj0%jr`C2189SaANh!emt(4N6ACqo@GEHDwbS59HlDYN$!XazQIn*5C{{TvXnYCRl#
zgh527W^7@KU(8T(v5dYLs*Lwyg=`?G18o(<8t}Bbie9LLZ-uMooWrVFVYOPJZ5p`7
z<Nf#EN92eX&!1z65!+3N+sz%WZ?2_3`vEaV(F+W~puXY=q$1*|Kv_WT=)eSBi<AYC
zY%50hBTmsc^1lcqPW0H3u*>VK+@#bICzl<?I*MGHSh=*hB~D?vksrSA?MSxzZLG2y
zL~YZbg1Zt5mYlKjpqs(r7!Vnt$#0dvmx0t2e9`FL`SYcFivVkMz3-iR-m<)TzREXw
zQr=ENxdYVA<^&;%jXKJnba+QY4G)O8E4(|5iMVHvbNo+e`6tQ8V*P_P@Uqm7+Os_b
zfARjl_YPbqv;O4y?@?ae1DC!DAoa--_igmeLNr=(-%74rMm>{q%5PUdSh}1%tKA#b
z8<qT%GzG2q84KQX{~r6>{X&WnO%^4xUW{$_dUAK8H?u_{s9dNF;w+vO(PUYAR!%qH
zWF7*nN@I!XrfQp8G#sEz7qK=q>(KG`YYrzBtB9X7<p7%Qmx(pW^yQ)kiPvymjYTX_
zAY=MfvEa&)mw2kZ>V{=etoB)B#9aunIfM1eNfILu28iJqVCNEnb53;kA~D!Yzl9(^
z&A@>7!olnZ5h-&FIlt--1?yF<aN|8rRx50`Tk)^gv{*ThUc*ZfizV@&eDr;M_~bGE
z>SuoquL4&A20+t@NS4HRi-xKS8fn6bOBjymhhB$LxQ#*e2+cAxN1K2$iZe%t6ucP+
z$c!yW!3mKENAb><fH+j)pUVgfwx}qTFI+_ck<s;o))`V4Ck6!~s&(Rvm}QO#fXO;p
za*k0_$PDt3{#><a+7?aI;_UQHM5E3j#sRMJX}m($gF{5ub=chAVY}Jj=Jh3h@bM3E
zadD1+{N*okb$O$%MkM?iO`K4ZS?hW)Yau%22m;IjV#p$SymFvpa-_{LV<KDQQP+j7
zIx$AdP%JqbF35J?GfIG$!4z|Go%A?u@1V?ipKX$FkL*kbvP?CjfK(@BE?tHuxwI1a
zWq>yYhL}B@bMKJK7z4O5;2Q(4DMzpN&bVAG<e7rSIw;!(;H;bj-Y-C%RIBXs@y|C&
z0lgk>OwQ!c-~<j==x~7^Yj}7Bgwbd<0M_^b7x6Da?B_)MUKOhWtV7yZ11JmERS>qb
z@JSh5|2<o03bLxvCV2eLNL*UEhsv{R9d{xoO*CnP-CHVSGS5Ie`f!?O@^D}Es(^0k
z8Q1HvBi)Z<xAUP6toED4E}e^G(xzehbzE!!Wa2#okr^p_^W}zOlOYn0fhp7jNEQ_?
z12zVU6{z~-a!GWnR;udY*SQ9E@@Ga0hcu41K}x_KBB|b5j37@6JhDj52#`g`ur6YP
zy09?*h4z_pJfuy8Kn(Iub*1F<!l4qd;4HX0iO!iLhk;l*MCz~1I2=UiH8R663?SFQ
z^pj*Gm3|6g&~@c7fHA~~tE)?_&dxFPJtzb`Sg)`h?$ATZU;75%dH;QU{n;1TT;J*t
zkoKX2xbci?Tp*mjpdeLTx&+a2^9@|%u<bhOV}eTimAe!(sH14a-7JuUjEytSkyE`l
zX=8{%9U2IyE*TR@=ZqZbVgepPYOJG!$QsIy6zs$pvMU|PT=iZ=qC~*zWQFy~2~O5)
zG;IrlaFe6{>-!#k2<Wz3ynOWyhTe=3B8H%0=h&>MWga@^s@`ojc=`GjKKkDG@E3pn
z=lIQU{t2Ib`6agf0OPO$EI#uLB$0*VD*y=_Pl}4RBa2uyF9RC-PW<jUkhrbUO+9eI
zm8HB_Pxk>7-(sDAj!hlGrTLq!r3ua)n%gF49DlGL=HDLJB_cJt&R}N>>dL>F3s=*l
ztAn-@gxND1{l@IQ43PG7+s44H_RhEt^>XL;a(zm`R?EA)e$zZ7dNAwr^5+79t`W@9
zV?vKB$dSbOa^ehLMs5v2+@3_(o*~~_af5UGTR1!#7hEOGmAP1*o>Sm4y<T{4`i<Sh
zxss=x$F*-2U}AO;<Au#UuFnXdmDkAk%Qh^wf1!V?H>Ut=nrAYXYV#;PtIWG}IR&q^
zJo~N}0C0JLsm|lW_qY4IJHc!!0MLa?wm6Y2+o1Wg8{BeZ$mNNHVopHHz81}uNF-f{
znFAo8MqEk=7Yiy8jq1Eq`YMKy#mLS!UIAAz#00#A%VWy%L98}OEEfR+H%hb!00#vo
z*@(aqiF05QfJB^QpvD;CJtZf#$-2+_q4>cr!`Bdlh{}*-s+x!}C;=q`+NRY)q&`8#
z(-9`>X0+a67*uZ!0sun{FvUjiTf}XLq-(_jwwtEK%U7?^4?V)r;hpm{Y=;3i+a3WC
z-+BK%eEhwS@Z10N^JF2M=Hd`RPBzUx^e}fO=3@;6jr1YFD)a-qbHXLn@eCRXBasbO
zr?|RDn5Y#h&NrR5z#{53E~m|zAw{t;;DT(RQOsrLAbV4c>W57@Y`ztXQRf|2Cu@m}
z&^BnBR>a?4eCzeyfX&?<?lv3T-Q3~w^_BMFAmB2nGh>?zHb)uBN>}y{)%DYM#~9$@
z#IVK@0vMls@=tjF^>h5_<B#$9-KY5FuRg)m%{3U5Bp78R6q|E428e)*c|wgvsJhme
zKv2EX<kCsTIv^%MpY{qiOXNbdS#a74!EUyoO@XIHU4USRJfsa*1GE%-UxrX`p8_oM
zgq1wzR%Kviu(Ag7GN3B~*%Y+pzMAXQ)t}L*)ny&IxHATYWA54XojOn(*S!whb^*Ow
zIXl50XP}qwPKf@9C{_Sa(Pztvap%55_5jKd<$-Ig@g4|I5bzkccnrdG5X4ufcq@I3
z^YE`#)kpQEXM}s94*S5{5{NkjJiJLcQ}8s(TY@hxuDSlK1Z&f>ceT5$)6V{>l|8F;
z<sMbqIR&yiz#mW<&)V48D>cAYyRMY6)E8XqpZ)26qF7+;*2L~7poI+X3L-3_lk_yE
zz$g)8ux{Mynn5>y)sL5Z8iq``GOS`ZYqyBWyc4ieI%PG*drt(Y)h9^YQkdyPax;?Z
zBAbx+URxl#T!J|X76pNMfFnQB=Zh9H3QCAcJm{P5QMpJYB84E&5KX-26c8(TReih=
zW8yMef7yV<D8LnLZY>zzH&`_u+o1;_B3M?vFZ1?K<TM<U;JOIH(D&$JoEzm*IiOvw
z@Y^pw$952l$<_n!U7X|5I$#J9fBK{EgS%VYUB5=7it#26Ww84CkpT_EpkTrXI48hj
zPL#!Zn)LswM@U)6i_1kMlLn@@Z(?V1gjKVW{@0!pDL*CsPcvrmu1Wscymx5Y7OnnU
zuTRjdS`ZP2t{0H0fN<M&2w}kG<rOyDEn*B9B4Y^RCvdbs7!tK>199ZCJCrnUDEYmE
zxuk(tAAM?op?&KZh=bUxe)IAYum9#}c<<X!@spqY7|))4jW53Z5<?J&$imaK?<&Z%
zn6YUsr$q-&W2627K?$U>Ep{tQ=Ztd>P7QK~FeFH1YKUwv6JU{G()$ds0=gs(wZl%)
znFbb<US)azt<xAXW{?ILNSfa(q2(0Z&EK1X#WA2mCExg2`MsQf4#MnS9h8;1>LS+i
z-m)lm3D8a3SoXsNC~^g`PPr3PE=mp5EphHfX}}fzw#X#^94^q{J;*Mx#s(`eR#?G;
zi}iSbkKy@IE<2sZM-`p)LZw1Wpk-&irBnlE#g7`08i9xctkrbbrrfpZ?_G7RahUD{
zaCY6Rmb;_RVb7=EEZ-@g9Wvi4mDmO7c7nYfdGqEt{orD{w<tD(M<?@<`wwI{2!QST
zWj<BVE>86+qO0|}hV!y5!$h*IKDoepVy8?51=vy2mtv@>N_D9m2yjqNP1#qRL};c+
z@12Bi%G_lYHBmNsCnhn(;a)jL06V~x#T^_>(ciuogZjm@f-Ne760k4~h~(gX8Ru5Q
zhDj6!POdqm-A==1StI}kl7rfSeh`ssu>N#}eu%;yion1Dn_m6}1oVOs;XN4!6MaSk
zu5NDdkH7jgt~MK-uFe1^tXhw?_c(1=czoL6-A8Bmm3nL*b8Sh#sT;g^NP+9ss@0JQ
zLyVg#x?yI0T5#5&4t`DN4HkY(HX#Dw8{xtl&|9w>F+k8*qs?lK^~pNNyw$lz&VfTf
z*L5Hw+-|qHz1!mDt1AqlN8fh{gNj8-xM0S~vU>!za%Cz!W*B&KW&&$iMQKq?z$bAr
z%MF4dx^2({5Q~^u)|WKW9RCI|Vi@q*r=R2b^KbC|?|mOX{mD=8yWjm5*SB|Oqmb4a
z)afq~frGX|jdJ805rKnx1G1OLf<)+8S|3DRlcJaB6nvTCll4{V85vXL?<PE~MkQ39
zm<zzm<@aIW;}4Mxh|Iy+BB4lEnJYKtn4x%<zf%KrsN@?z%U{FfUJ2NiK;M$Mw8rh*
zm%lD@&DXm?9STr4E_W{hn-Rpx*z9<{b1(5qb}IN(Kf)<~h&3KToMfG2g$5m3^eO?_
zU<1dGQ4p&9JAHEk5^D8X;%pt1Z<1#DSr5@#zvPRpT=-?#2I4Tzv-sU5kh9R9THYC8
zua#2+g!@3v!up&7$z9(ay;sXnfQOxBRY2dp0N%nJIULu$0mn`WLE_vcK&PFd3fo+O
zKSz0cut4{uo4r$qTvi4t$A*YBHYSp|jCJs$iO;Cu1caz=G=va_LbR36oJLfkA(fbc
zfD{?P^x2}iN=Y3Vq1KBKZXtT@>xckgi0YgefFnjw7Z+fSu%Ufq;x&SP6HqHv{caF_
zGaxEJ4(f!~HXyZxWd<mMod}K?3?4!<G!e^EAb`+E0Rw$t3=C{HTWtCcw_T4-h$*r}
zV8-M13J?7XW)bWip=nq6<*z@%-DZQ+W+j$mMobAH1aR!|=;0Zl<wOMuw$>sSq$eW9
zu@ygR4jk2iabR&49mT>rTZhUVXkC+y0Z6#5P21vRy~b+2!n$2yb+S&u;`_cwWDQ?6
z4A^XY^uvJ7t4nlUhhgZYei5p3#sCCVb}@k2DIf{S*>7%$f^@RO45>bjf-uLzePV$w
zVu&r))e&|eQBr0iU$CeKI7+@0VRN5_D`(#}aA)*HvKjz%xVzipAOGQ(c>44)zW0Og
z;nk~4Jp1}Nx~><1Wt>8eh1(c$n*$w_lAD*>uWAYDz+@ae*r5uE3ny(;zzFqqSK>6C
zOERbJKtbxG5XmQ_?E(2)$oK5`<ofMW24bnd3phIlZ<O+n|4!e_e+^930Ubv9=70{x
zH|KzA9}ukSwT-?#1$q@QHtL&s{wdh2gF2ZzyFp!rkn)aalb@rc_iyP-Z15Tk6R!d*
zJjDn6Z*Y!B0QES<cX5S}@d9^XbYMgbh=}MhALFuBJh`q^VCi0sm}jxn<X>F4&l4cG
zuL@<peeWC!iYnzy!PVZnO~?9xI*rnFJeIQ80NfO;SMD#BSJTbg1?YA?w|UKz>jTr9
z`>`{SqMLTMm@UOS8ORV!q^jFtI>nd6z)4J(oc3zRCoUc(RXYFzb_9bl*odOGKWG_V
zLlkk4zW3fsUI-T<-D3$HKPcBN#-Ln3EuVoHhpb2s0+HwzIXFZPVC6^}a0}54HbfT?
z>WnwA5IG8Bz`;dKR5FEpFv1|<ju{xxS?8<K0YjXy2{Wi;0an{?Ko=tJ`X1XMpc{hd
zcBwlP>1$&kiD>%7OSFLqal+|{aQ@%|rxzFa?9<P2adMW4Ke8CC4*_kv!s+P=hzQ*f
zl0#@ulZj3w#sE=W#5gF(5soXIxDo*oR!s}vwrJW$M53!TR;v~uLen%N`esJA*{HEW
zhnF|6(DyxtC=PXFjLD^nSBv4GzTa#U^y;G!uyx6DxBxK30nw@6Ug721T%(AX8Of<`
zFz^FTx}16%PyiXzVJ_--1T-<A2M4K_9C!@0OT`RI=>PP<U>s^<b7Q0b(}U5oufN9S
z^%Xw+?sxDfKmIX3`|LBkesu|k=&yoh4I8IS#LU@0qm7+~BtznIXHc2e2avgwQgG&o
zvx?;lqBj$-N^i*9L;$jyMCoG5bW>-kkOaJG$^ge|R2RRFf!dr?R|38o7jGZHL;Ct|
z?pe8e)_~L$pp|#ZvT8uDEWLz|<+G*y%g;>wI??-^5w0|`Cig7u0T2r`u%lOq7(nWb
z$uZ$u`1g2(?*b52mK7f23H}T(>DL%wR_%l>y##CMF=!j)G%^Kr6=1-%`wPHqDiSRL
zTUl0(=<ftoyXvzGxE%o2_I<Mk#un>#KzXCQ#oC-<Y}!%x>3pkvyDV^4kMeLaDR0+Y
zJoGyK<?-JG(*50?=s!o<+LOTIf&^JN$^r0fC``F<Knf$3oslZxA|&xD7YjY<r-7wt
z^GT+j{8>3V@{H0ti!2U!i8N?ClMp8Wh>XS*?M<Y{5vlsB$S(#R*<-S}%Yceol~cnm
zS=CxKjawO&I|U%=&t=E5qGy-AVRH&}Z9v0B&gL|C7+C$R0ey(*Vw9MieUI(Xql<&+
zctv#&0|CBZhGVfxU3-s}b7-3utJb6O4y|j@cyR~Vbz3y#(5yT@`oTwd{`oVkeUm`;
zV1Pu4ZDhvf%{3wzr{`z5y!j^i++!j%-lJ(mM~xidPF7g0)>yAsXjdyVZHv>>6T}ci
zPtr-awCzShRo&j*VYAud>hfAd%ON0Y1dvGT>^8(iM1qF=GA?K1nxvOz;&}!nhOLqu
z*VvpABO6j{PfsEiW*|k4D8U1%FFzYEZ_z@46g2@U>V(UpTg7@2q9aFk@KE3<j;k}l
zWvCdQNv&}69Mp;=3u!PTaGpa%!0p`~e)Gw1@c5m__|A8}gLmJ3iqAg%9Cv+(A!Mh?
zCclo6;kXb78fVhEY<)@=onvVuRFG#X8<W1%<*Xa8apc)PRg^I<pMjAoN8;lBfw*;J
zuv3D$rTkOQos}{LOyz&;qSh%_Za25B4)W@tbPTl0cXAzfi)fb&Bj)+X-@zpR66bE5
zYM>59z6yZfNl005Gj8#$0AfIkhj_yOHEi>DSOZ|J=m{JiB4DcrtQB71@6h1|PWZh^
zJu9Fffv*W*15{i(1#i<f7G>;c)41I`K+|GhE?ke^nSSdKa8`M^GjHYQp&+$<U}__U
zWMT1rrMLBmwZle{e`ousa<5X<Vx4f<`0Pk;7*b6B!w?^69tGhmHn_kME<z$G#Hgom
zLD<>l0R-ge2G1!+E;GAE0f3lANLrjzV+acnZed>H5X4YG_1~O=zo^8TbzqY#6NYf|
z6!AF?6wju&;ED(vWE@2$%mkQtE2;v&YdjpaA{u3@8)u>)1|*Rs)CU*|0F4lf)h<Nz
zp~trCu^A%z5YUH+Z5&iESD;DVWF0x2IEU6btbBuY)1vhb>$b(pHv&o(a2Kjzc>lu>
z@%+Ve3|)t_#~1j)kA8$t{`0SZsN!`19u5tW=nO*i4`04~g{!L@JbwBF5e?2S9-wU-
zG|r**5*1;HQADD#2iRe|-QsS$!S?bsVr1OiT#HlU)ITbo1CV;1@&-!C+^Rh1If(&1
ziO)NN96$o+*s#J8&P3eCBhQBn=A83@8Je)@Xo?X=c8<)9QDh5J?v~9pGSF{-iKI_0
z1X;}M!TUz7Wwj3!>@XRpF*|8aCKqgTS~-Hu$ykG&2S~*5&%b$z%d2aA^wD?mCqMlO
zKKtx5JpbmUaB<n|+~wp1BJss28|Ii_#TX^flI>v*H1YnX)iln#-bx}`%JeK*QL8Zp
zEFNwF8W{DmN~h`f-Tdmw!0H^RjejozxspR?<z+xO%|9h9Q&2Ppc$k$n1!d*ETA91R
zCPvSub<U{{M91~m`Hg+Ul(;bEy?g$iT{D0lckU|)$4z$6GyDiEJki-JYl{IiSYw5A
zZ1D{uB$n(Jud$&oaDvC;4Z(<20&vJc&Dv3dp)&ude!Qn03jksXv>XKRDs`Ks8n{{*
zyxHxgGHZF?pq#yP$o{iq9*x#8uo(W$@^bCI-I}~R>$l=t_i4`?ZfS4i$4<mf0GrDX
zAX&Ydf8iX!-N8)5MOz6l|FC3vk-?L^mc_9c5$>2Ok>bwU`$~v}xFS`6gK+CC?j|^M
z;=~FwCebM%1X@I&j6pdPF_?P{PT<rj5OdC%f)l_df|WX0bON)$CCXwqcn6hAiZ8Rk
zxa$XWK{#+-4CsadT~v!i1}rjw&DYiu8X~NHgVuShn+EH4g~k(Bp3pdl#z-`aYYbA$
zQ`_NLFb10S2|oDN2e`Pnz_#!3<A3>6blVLcK6r@plT)0mRFCcC1fPEKC4T$ar->`n
z4*@S;e1pf2pWwwem$<#TLbu(Do0mSI554Ld>OqQ~h-66(T*x5=C8&g1EK;Y@LqHH6
zv;^veyJe9`%<u+L=aWT0AQVOk3gDGXOKB34&LwG+ff(Df!2}1ddSY4WA0&E?F$R!V
zFverCq19ciaRCE;fZiy8Y7E&&A(2K;q1;~4^(?q`$m__d){u7=YuCV-&PyU}`hZ`4
z@*6z3IKy|o^C2EQe1y+F`&4`fI*m|58WkqdDD?rUY6Z%+P6sPfx**OX(tL2sn`<^U
z8E?6HosWS+vXHOrZc}$G<=?kSHBs!Ow#dM&1WeO!b9#GzJuR;WzG~dJ{JeZ;3i8T(
zW6)W)(@I#@tsCE;k;vJ-TK>^JuH-L4+`+`HLLk$`n-X$P9}9GY4Zem)15*Tgh##qw
z{5bl@v;kuaFVUqD(IU_WTYd&~Kgn*cM&F@&ou(jaSNnDXw0i=%8dtA+e+RIey?+Q`
z+W{>1-A~UBC}ZEVl%X<9XCQ3~^04PwWxdbd!MsmqW3hC-qn{5*_X2j2Iq9w0=3|;O
zwgETf1q^ZC$k}3@Y9AYiBzu4;b?tQH=M0cc6dMykl;jgaRGmBpPHe;svOSq|6XOKH
zapH=E<h#s#?I{t1)g2%OdYn-Q`XMO6jymity`fwp?;9`+U@5nZJ_KyzfNndeRq7yO
z)ex0qqERk%>ycQXgoYee#(`V4XqyJDBedS3aSo0}PtALWFo^FZ-#CdQsvyQU9_?xc
z0&sGAiuHO8*EWbCtWPh{u3D^z6P%o#;|D+bF@F9Z|0CXg^bWes2AjKETwPy+W5ny5
zTO^hXWc;qLZxFUUzW(x=xKvS#UFjpWK-EDP8yA9ziDf)n?~(Pi0=)Ma2H^_EDB;j7
z!hwldz0aY-Mh>k>CYFdhLn2WlqYg9PAsG{BU#A0xawwStz}Y$%<l%vQAb>{ni01oD
z{O=k2${DEBjYNPZr}+reU84xcl&n52;sd}1heP{}vh^`DBPcgdd~L_{E$2L97<6AE
z`vYd+_2m_AZhnOizx^$I|9juZ^XJd;^3^4rf?jhkBbAYBX&tLtrxODg?aRnElqI}?
zf<qIv>RM+m_6D<r<<VX%MoPtgkoE;2fsTn7_77=1F$PUzF5I%{b)GL1h5f*fCf7*i
z%=4FfWm)#v$fE>!_I#`>HwJTakhlcyrn-N*4LM)>&Et2|?|s9yiwN?Ld!y&Fp1~2f
z&F|6Sr5=hr9JhGH?*sg5QnLu4Lqwn5V!?~UC?0@xteVb&LmDx?y#U^3ZP^3H_Pu)u
zU_1zXE&!BNCz?Mz2;A)ea`&Al>E`q>R!c=5p4_YU2hH!*fL=|#V2SfuyMD{9?9D{6
zM%0Z)DpB>I!s-<o@3-8LC=I;OdrINH2n4a7Gq3+N>DW@|$d;BVsYAnkxdbrT>IqH<
zDbd9fFc3H!YJeH+R3{FUT$3bfq&jh|fJEGyNZ@<#MXyS9+8lz0WE#4NZV1@+JvLp3
z&Cug+7=*h6#$eG6;F3Q)55~G_(Rh#6IjkFpb<^N<)!?LEqj7}RdjWIaW7RfbNJN75
z$tf6&ll2;H>lN<`kQ;h4x&SuYE$(h_<72ca000lgNkl<Z(f2*J+bwQ48wB!r{QkG_
zWVMDPkB5&RV|{vpzx(^Y1Fr6pRVqlCH5xuEQV=79As`MxjaS5ha&mjaI?lnwl-g=r
zI~)aYxlA6CsE@MKV6za7uz4qA;0S6ELb^8fIUxZEsOt{V^+Xb{lOu#6z6ZrUA&F&l
zj9Ir10+^YjdJ-TtmT;=Oi8Pm)OwUplK)K#U*VBaORxh3T4KfE(4<;N$vnQ<+DClP`
z(zsMwp1fNuQK#!e>yZyg5zw~uA>y;ozQXC*Ykc^@w{UWDhG$>D!0l#}{7r(?+BYhv
zTaR+C>{m7JfQ~OGaB1H(g9mFjlDM0ZJ;RveYN>e=G|T^OYzyG=hk|kSH0Q`ob>lMd
zngX~IT#d!DD4xtgfjuAR$Lt!j`x!`$MWo~JP3thuKLTSgusSVgp0EBslB=4U$aXi-
zD1S`zF8p2kt&5Pk$4mDMFa|g@U;t}8z$rflVaSUo9mE0x^*Y;ioid@}lM2XL0#v9!
zuK}xlV66Jyj(2B$unXMn>i=oklV>$>#<K?pg1h2o?fw*ajqWdiue21W;BE@gif7dW
zBJKFjt~yV?v$I{5>rq_}-|^necmjFD79Ij6e^>yisFjO0DS>g6!vX1zE}RBc3^zw#
znw%4&N(7CVV8laD{SoRcSQ>0PLPRE%)p}^SDa|foK}X5|T=om5iuF+93JqTGJo-4m
z1aKiT*f~Tv+;$On+YRpe0oyQO8v=%b5rX_<lC8lkE<s|2dgdFfyvNykh12y39-N)w
z;rTfptWR;WYT!Uvovh%9v1&ZL^TPED>Xsu$M3%Ur{V-q%0h{g)udiMLkumh*eAq%q
znQylk3`tYsz;592Xmx^tfp&e0AN=@ZJbCXSe)?xW#XtVl->NtpS=zt}ek+J0G|dVx
zUtX&Ck`#<6p;q0;s9Za9v7tCATchgYF<mwDGno7)8Yoa;LF(VH@#=0wjUQmTZQ5t7
zh8c|+DX@YP7Jjbx-X^2!6Dn!tqR}9AXk3gX3#fFQ37dpBGd(!{Lzx>TTtR63TXk2X
zW9pbvokNiNh()*ZgM)$|JCraZS)ejTponw~0#V|aM75-Ku2lmR0OW9cv%x37{tX^I
zdW4TY`VL;ce2M4Zd?V`wPB?TaJezU_qv|svJJA-0)}+$qya|o;;8IVqT7u{G9%IZ;
zO;oLGvD|}3Woj3j56QPpO909UL*7|x`IbstAUf4@a{i_7%tf#zOq+tN5@42~F&_#t
zKrL%n7vGM7UoO|~rQbXObW8e6BhXX+rg_s#pt2L#(dZt7v28!c3;#=iJP0j3JX)+U
z&=nef2yzT0zo>);=WsXyV}pnW4h<Td;Uv5Z(a&?mH+O)T+Vd#@+6U?m26RjLi@ZCz
zbq4^v*_7W=*K&?m`=Neo@3Y!E-&IbmWU~Ei3Ib<kF0IkxUTxl``?WsZQRiJa?jGCQ
zy*YNJ0(&-Y7;og}XAa0Gzfw=g8TwVL*_aTufk#<1R(=SBwkvNPM(Bh3LlXi}SmtUS
zh4|PO7L(xw3LjqZqPynIR~^J12>=W%fMH-@+Yi|E18%pSxYS@^V6hYhC;=uAus&Ji
ztZ8t%Ug7-g6ptP~z{B%%T%4WZ;`9{jwnYd%8e-ht-U?)7M!(r&=(gy$J+592=m&F-
zJ4kF=?Gv-IjOz9z4C2h#D{zdFl`!izO~L_d+9q)v84;VV$7eUUxVYKk;`||=Jbi=*
zkI(Vq>TCS_|LxzQzrMvMKmP|YxQH@VOsdom>uR}r{pz*)H&hTpB9?TnQ4Wo`TXAA?
zoR>kAjgq{Asi-8>DM8RsS8U>SjtC(pKU7WA!V#ehgE}>i@M<*RK!Eqjv9SS@q-;+%
ze|6rbc)tMDqBut3ic&-A@E`<0t*c2zs2ZxwoD^F>gDj3M^Z=u}V+tIT`w(rfBVbUG
zwfT>8&Znpz5;vFiAR^!tNgRzd7lW2B-%~GpYP3<h@re*$gD+lR;_CJWj~_q6x4!ih
zuU=i^_Vy0Ua_}ND*FhR58JwI=+o6vcqKh}#e2w{SeP!$4_KqoZnNsdD$JBq_6gs_K
zGSra4+|+ns45)V3fGhb*NH<PpIrB2>Ku_|~BxnA$1RBdyd6dKc(mn3kfqnLTQT)UU
zfQ2f=t#WS>*ik`urFPL>;mh`~u%(+sLmWJaI=sSP<Ed-V#&?ybNa3D5TAZWD4TP$k
z;S^_B$EOQmZWq|#NqsBVW&Rll@DQN4=X<qpF9E*VH*4RU0h`J<cLAHla^E)pc>Px2
zVUx-^%vNK02-$mBJRtAvP(urSR$T1t)BWlGv0|0`X09$o9wu6Ba!SiUFYC!*)aYp3
z7yv$M)B#oQXU?a`<PVNDj%=~%Cf{SJBSZ|yj2spT33nt;jBqnF%$O3xfjNZ)<H*<!
z13HfAgV;mab`_Y!()Hm3oRH}FIr19LOpF;?A`mkkJa~Xr)4)+g;~T`#BfS0wuU|dK
zs}K=2DnuNDD##;oro{5lqZ=Z~n-_ml4rNr1nG$k)2htuUHB8Wcao&R?i$0n<Vs4xV
zAQr7xeZ<$7S8$(vg4Qvfymx`0{lou*Z-4(OB0s_Z_#ggfeDm2eT)%j&V)l?YSA+-$
z2_dz)y~XwQbq47Q0tlQ!TBUtJjT}OZX&%@Tbc(d1hGZjWW6dxHG1ILRcQXJF0X-|f
z*SYNUn)0*@L3JTH8y<*3iX9ZZM)g1v6B1Ekhy{)bfS7tgp4T#mqszt$;&c}fw>H@j
zq!jUI-LR;)q}?&7{-3Udz!8WNyEyw$1WdnE-sHq!%$)KB^K3=s1;M!wg24|4?z#bY
zUw(!2^D}(wTkqrb>r1?R^%`N2z6lBg9C-{J^#GI)V+4T7sc%M|CxtU*i&SIdg96kg
z(fI@{)i`lej$L+(^tP$gAuj)o#jevY?f&w(Jq7a6t0`cc0<bxtqx5EZ&+d-<i_)_a
z0FK4D<yWQvuKru*oN?xIxope7Xuj>+X8dOJ&xor2t%8)|_tM|mZ=L51;Kk|}_?mu$
zZ1P0#IDx|ock~Tj;b(Z4{xyhOtp+1t0O1_KIe-p?22bOU(eOD$$F-(she;*&-J1Ze
zQQM|~=sw_Xns;(<%2lfYzJr=E1$GrcQho1$`}fIPeLkHeHGroZ<`mqG`nI;x4**y-
z4)r1ZTaE=;If-!Y|E{Jt;@HvT+BRo17&qBZQZS#0Vn%v2ETSn_ZvNV^XdIPd6N&|4
zwuXv`$suT`G+rzUos;Ipz;KO+Z&x5u0Yg6w@{c11a_9+Q&wvMFyMl8L4O<impznIP
zAfV9o0~{DD@38g}EAnFK3RUQj5yRz60bfSU*%arIHa>G}BsyggtT9ptb&TvhiHi*l
zmj>ltB`zduA*|x>Xq+@N0*G+Vyp&Votb{1+px!ZFZSHW_Z}9A!FY)MW!s?>KdmlW+
z%fJ5;56&LpzxmVuR-#%k0QC_EvaOiVI*$kE=h)obA`eRtprqi%X`guG60v~2Y@I6v
z%9A>UCJ(S$MaGasmf6*ea$k*WrQph|!9$FL%}x-+39<>$MA{kzHjxoKjo0hE)PaGa
z2yGX&u4TwNCiRIIV&YPQbkC33I@`Ev;#)5y&K2}<!R(R+uIUdN7|$nnu|k)4%3ik?
zISg4lp}-zM(l<?q1=S~1yJOJ_)j4FL^rUoKyM*i8J8XXUd%XMZQ@sDJ_weHRbM#$6
z<bbHokkfu(V3UoMYOfjxk28n2+&i2DV$?B;g|n2R>^!G3$TOHw0*=GCyu;Gi2%-*f
zN&uDNUJ2q1c$N1pH)hwRD05nuDa4z;XV+7ZmanG94m`?Nd#~1RD@VK~*^JV*a`|Qr
zbrt=Ph?Zqp=%;%PY)J(5KW>8;{u69z3kj0Q02&b1a5$ByJJ^CLK%&A%fC2{eV6*^E
zK!|9#!PD@k+D20us?EH8*TuaVH**J2+9!_P3+^WUGP_>`d$nd1<21`NZ4c_{4*~EF
zes6v#n&nCUHj9~F17WlAp3b+We5e4#N@b>FS&r2`H@Jg<oo%W^41l^`6PFXJ6JuO9
za$<pM_a`T46Q3HWv2ac1VuXT)kWzMeh;XX=<ej>I2mu@!ZQF>iw04DdeTwzT3chL3
zwriZ8uHl;&ff;w3EjHUN`p9_w>JryiH@NEtY=$0PFR@++j_AW6t~8Y6<^XXe(lj_*
zt?-~}(Y9ViokEm}BP8;Kh&(wkwq1ntdPX*Nwgw1+HR6OS{cE%du=<n(A@oC1l;>e9
zpdgB4-N3<co#;HV_&hV90)ptm4J^86f}p^?2#T2uKuaMVBCdSEv)k8bPCY*S?t3_S
zaDv}|{yRK--eBcc`1a#(<Nx~A|B06`UJAD`0@OJ4LqI<WE4zt|i%0L^tLHD!1px`_
zMG!m)M~M^^1_;OtLG=GjtI}f9G7m5#gT4=<2El5u;UGcf9Ra<9GZNSarfi|<Df_`Q
z(8j_!0!x@Q2{JF7&%lhvdxXG9QR)d`N8r%sc&8S2OXqTA0-X41jDQ{(#8)1%#O5_U
zG!Ti7C;+&K5QBb3NTdMWC~67Hp_Xq1Kz$m5)jF0G1e)8GB#wuQffYpcLoy~(dkG})
zX53IhitQRK@(E=z>OcXXfB6+I&M)xJ<0t4hJzig4i9bRm&T?3x!o3l)nAt#{9cl{t
zZuz90JaVX9+r<nVmdTuOYvbj>QtVgvN%qd|rK&|J%o+o`F}NB(r@epkyEL+^ET2t(
z*8t`GcTGpGwB5}6@cbJ~pmlyd*55O3T^VX`4s5k^t(Pan%JkHx1hoPjQV}${F7k|O
zE)Rmg#>@T+*EGOsR1*)69svUe@xjU~T%v>H08m84fN$s}UgI^kxCOAmyZk<I_W>@4
z8zf_)RM)bN(>ftPpAzgT0Gk1_UBGUse=6^m@6W(Ot&qyIS)29%yEkLZ{9c*Pc8Blh
zIiVewKI$Bd746X7vtK`dtKCzZI0^5b_p$fxoH)<?&HNE_QSGmn%YOjxX#XZw=5xHn
z=lGZZrN92ZukU|f_kH*Gs^>e7%U(8=z5HMPlH%{%jtBIPb)2{VFT*cq^Mm>2Mj4Ql
z;on4I<HxUFamm;HNGwaJOR#!Z2sY@v<=ympYmOUW13(JF$o_xH{^YlCF4%z0zU~R-
zVB3KX0J*_^Qs04hGx}b_QNN08<6UG;@}5>qyOXV!!IAx>5|%QMW*DkhdCBnx$%q;M
cJTCp`KCJOMHvIRrbOs>sboFyt=akR{0H4es$^ZZW

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_optical_flow_thumb.png b/0.16/_images/sphx_glr_plot_optical_flow_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..b1a5c2d5695f9a8e19d0eb95a37edb7fd17a99f7
GIT binary patch
literal 94362
zcmb@tWl&sA)HNCi9vp&eaF^f=Lqc$u;O-FI9fAax;4VRe4g_~5gFC_9f(IFVuwgFG
zd*6Hie^p;qpFgL1PVYLi&+4`JS{<XVDu;tbf%WRuD;x!RY0X!!kiNfMWEf~KcifGN
z{9YlwdZi%!@w0E<34F1U)6jn@2EJfmU-jiN=&60fbMuxYlI|7ZZ|si|HfRJyL=bX&
zVr)e`|CDUWPjk7Ytv?DP2?ZGi^XNYEV|;qWNco|J5~r!%sTK_LZ4GK4H<HEM9IfH(
zJkATCEUkd&FO^sE>vCsi7YegS{6;nU-!U8Z43h|b)AsehafeW8qoX5rMfz`?E4zOD
zz`ce1KO=i9?CbycllKr*r1t-r-}cou><zN}|4es9nnEE(+5LZ~BNt$DV+{ZAA^;*I
z{O|v}f-m#`wFL;bis}CUB?OJ+{}1D~T8?oE+@S>cU%Nh-Wj*M<GCw>1@&+E(=wRHs
z`>?kT!ZC1=!tjM^oDwNcG<~jzou_#KX+KkOL}*K(sZ)RiHagYpcCA0bneOyi1tllf
z3J}+HDCB3lAve11{x|<-ysh;wc@TSGD*+04*~F<NmAbM?EVwJ+YP=7X?oQ3084-dB
zzKa77$MN)t_yn>c-FJq-5WRT8*FuJ?hIVDs1|{iIY#&_}EBDD1e>6EKa)ACQ4%}Xd
zq<Y-&f%YoxYD<8U_o-K2n?$=+%@}Gocl{Sx?Ifrj!FM@sf3GSGb{LNRTn`xh^W(>a
zP5XyD=zSgax4UFGa5lCF89EFN^{`jS(lp6Hd>}!BgX*0)(h+2o>;Smk;}$)N6b37C
z$;U!BJ{upO*6fm@Pk&I-zyHLP?K{6t8-fA<ep>`4sLr6h0giS4{1@cdCkCJQ^vr)Q
z?<nYD<xOIca=<+m{f1A{Vk&g@PKxvD0%uy0u+If?3vFm^9jZe-)bX4`Pfp?ySDWz8
zCt*Z2!ux22bb^BuK>8Q)L$vZA59keYk(AyXWGG^wF~^DM6*aQ|pQsD@Th69B+nl??
zsHVlga#GdngtGPIjRO!t6$$#|t=cs@)^N=@w4fD<{1BT1b3fbaC`_fE!o@zd%yy;O
zw9Wg|%^y|l&$pY;s~!iLgx?bqXbRy^nQh*u6b6*>(>P&s1{kq?XrKhTATDXfY1wEw
zM=gXPid^m;8`z=O^5=@(2lMgW6`|>3nXaY<b)x*p;pUiVsq8=aq{c)=LK+Ser*a_|
zwni(<Tk}&P-+IN;NRTz-<wR_!v9NQdM+mx2G95keQcyt3d39!QEwL2{7h}KBY~g<n
zXABHP3QY5r`W27%dNNUMX?c|ax_kEuE2)pD>mv*GrK+OE*4^~$4LHQ~JRSV7sBk%H
zkkZTfuI>l(@NtW&Z9Klzr(d?VI)&9>o^Y9if0m~naTuQi;s;K{$GLnPICsv^$v_kB
z$B)<Ln_J*V$tUQP6O5J`J8Anw8bcc6y)=U|&~|FR)r>sRNS@UOV|cET1j&J@VT-=M
zoC{@XCGx-|XjI1JH>Vv#TQC796WyzbuP~%C00gr*cTcVx7RH$H5#1F@YU1YG5o!A~
zO&BxEx>0r7yKw@142UuTK4_1kjM&5>*)vR5ltKyxtE{gH#S--+J2Q&(1b$8jV$>kE
z!HSnkT?f|zbP~>i(UaoE7S5m37yg<%8+*oCX&eazAtS=9kwsT$FKR-rqez@>y_H7_
zZ?t@XYjGR*#0upUS7B^~`uGxWuqh*(xy4ujxXF^<$d>R`hIs#__b6bYt6t#|YbDnA
z02VeG82bcM&Rbw3UmE#C4{EEytueY7NXN+OUKN5OVh2>(Z%kta0?w>Cid2@Eqog`F
z+05yiA`P`rj6Pj-`C__(kq_NM*|E-@bIg>37B;75S$O_DPHt?<pnUPJ5q-yhk8K}%
zgInjR6b*9IJI>B3WINF6mRXdjzUEuq`S-XiU+LjtD5R+8pQL*lhOEh7)ev^}w?G&F
zJ4N>5Fo~xUyflr_E0JW^so1K>?48%4KCUT`%vu~H&b-v^jco~<!;MNwQ02T)D!B4E
z>v{c}I$1U7O;~H%4nf&#2d+tj&lP|40@brVr+WF9Dxe!?INr99i7#I3Dya(kzM@tC
z*tU)jz;~T|U9cswXV>q_YXzOLPT_q0+m3$FjwwxF&P#XB0Rsj|pjwrG(l3p=GSfhK
z{Hs9yS6N6h4rD+wh##F*5SjwUB>5=VQ9GJkF9U6x<K9FWC(>k48+#297i-dYtku-J
z2+HI)+Ruk`9$9i3h%hV*P}ZAu`F{D5@KI$lFn-a}ztmco1F(m$^~PUO&uhZHmtSJm
z)mg3amnO1C^C(-D)pdMYQ4^i8)4O`Jf>%weY6Msk_2Qfi0&y6VX`}3fLT_muK@KQl
z3A@ZfWIy*sI)`rjyBgDW!py{2u^~EPZ8(eyvN_%gDa7e)gDmBxFCUWv23foUhI!7`
zNF#`{20!nxDg9F9fK~33kw$&U5x<y{%)AYD=vOQnCxHkom)Gp46Ce;ta6!7+xJheG
z;Z^kP(8tl^>((zp)ur*vmQH^OU#a~jU~b!EGha|{ToBX(&#gh!Juu#wOL1x+!~R{6
zq{a%3+R<j`mlz1<dlhYzmL(16OiemAqCvUZOe#kt@+~2x2|gQGi^*!`=teR%g>N%9
zVOQj>M^0IGpC&t<$!i$tezv^lOeJZe6T1iTu~(&&tN`~zUTY0~{T5;QuT_TtGO0u^
zs60pBtP&=RoSpi`PAqy@Z#%D?o%&ujtk*ofH-MKJLKY6figicvZGuq&0oD#7crKd6
z<f_(9%(E2}fpa+MU37d<GQ5W+w~@&WphAX}rD5zB000t4q2j9OsZLjvel7YDSuN@J
z>LZW?QWI9rM4~n5$b@m!MYE7S<~K7{d{I89I}ri&vp*#Yoqs5^O*@;EbC?Zuoj{qO
zGK_9AjM7$Ow=<dZV3zQLk6ZPQ-?w{39Pu5OoH`&Yk$?MzELexfQMz%z+!9<D!jidJ
zU;fS7ap2)s-dE{~7@cwYYs`iOl%KMFv6bFxkBc;N`}pcPcC*p!0$hYcoRF+^_RkXK
zzw&$qG#x+15JTRu<al*3`Y8%}vzG9Z#TFNHg2;Lq2;M8is%j0w-(OCs0FX-3DnRdM
zU<S=|D!`Q}zz&4GmUVe!{Zp9~U`8O{Kux`Pkl)~S#;AUmlE%*s7dUon5TARnLGI9O
zi+y$W-$-TJ5XE9olSj7VW-D&7VfaESi_HWi+ds8X`@;2%Ead2q$w0ioCFfQ4$3vwk
z2g`fElMb^J!3W1>rUREruxX^Z_wB(sriYVmv*CddmsTkEPw1GOAzSDnxqUC$#jh$S
zngr6fS6|pKQf8#0ZRT~kEb8x4V<O(~EQnh_ewF=V2B~l9s0yl8!Ppa%*8G!$51^5;
z^>|%|15{Dh4y$Gldy^$Vkf~qx(Jz@y%y{&fO<%u=|CcJD4w(YS&iWi_+So4&ufXEZ
zknY<6-CDB-+>HR`cZ8||JxED2Ir#k!WA}6(pJR)pn#!A;Z$zcdjC(Hx{{1S~&YuTO
zY2^oLve(O`$UU|q#~a*Uaz?h3@+xEOu!QH>`1`R7Z+KXDHrw#JWmgWFc=+sN{Kk66
z&!JLN{P|7UcWWGaRxY9b`PV#0ip@YtHpL!grxt(iF<rXc`m^;sYG|t7ye4vSk(&&T
zZ&T&G9FgvF3q_S|pwo^|t}kb%fRsS293Qtohst@tX-8cAW`X5;`BRoyi{ol}X+_LM
z(b16TR0Zd-Rya--eGVj`N#Tz|qykZzPz7CW>m+(gO`jdF1qUBW9X%fzHp|e^8rY+d
zC8<!In4u|b+OFUCk8OxcA*frTrPaspCS7+4?ESXM`FaA$PhRT!OLEWx>s$4YC0&Gg
zJlCTKff>Fn+xp4ZL-f4IiGBrE4f4HuvWzyyzL`UjHF%)|RX4m|X8TVm!3*}S&o#hy
znx7XpyWIu1O*!Nq#`fluxmEMd(?7>?VaiG%-KM8+_q$f9YWtJiDXKQm=~>wA$Zt_0
ze9V;B$tX?Jtkwn}zhRYqQIKBKMO|DZjEd1zW<(QH6jb~Rc;RvHlLfw3R2gi6v{!)!
zVvlI_kq^mPPq|L-A9X4J=6bum{JR|rVOTV81t5l99HAdlc1!gW`EOpcGWa=Y=*ynt
z>S{CENAf5=pYZeybqBx>ZA4dB#k<a*JhB*naAGy6%3)}ox6RB*0_f&rH#UIw1w?6b
zeo<KTJ-S#jjC2M>M!&Ubpk4T1m<;BuH3#|iND%cHz}i@c{FX8NN585HnTS#q=@sy*
z^Pb0-2{kN|ing5g`n$UoIQKuhv#mz@SQ*_I#&=afK^9cbHVZ6a9u1{?`L8zXUvrfT
zknShr-JZ3ZQT#XbrkKAe!&bb@@~L<93o!(*;Mtz>@4<z&NgmpIg2HvhRYQj>6{_(f
z<pisgu$4RGV~Pv4XgMw_d87XR*Ghx4T0EE12GO~|>c}cm(@u)k&pKMC6gO!o#F2wM
zMk9c<!`}|?Y?Q+mBGuDH#0Rx4X9UV-$M|9@N|5mb@r!LVyRNco;&4!7eDbJhT>TN;
zl8^Is4egynw`aKd067ksnWHTNK{Jd{$Pp&A8&0zM_zb(N2H$d=oSD8;khb14=Fs`M
z!N7IJeAkwkfzkE*^-(z{0ywD8BvF$hF!|WcJ=fLxr3V1NIW90<zIh)Hub~5si$$N=
zr+wwrQJVPMYkdsJ)QQY?SR^pcz-t+{w4a`jdoX3@15BJ6GR6Q-o1N}_OFx-;(V3^Q
zMiYLEaijNgjOOEt`LHN^V*rxbVH4!hh{z^OWg7I9P)H|ymnmT{=lRVEZqrjI^mp3r
zQgHfJyUCC!QfVdggjPzwdK#{70=->Fw@M0CShZC!^ML+0E)L~|P<47XMQm;TU){-#
ziwgIo2x!QM;Z7?*^fEz1aYMU&^^CVY4S59xZg;zU;oDzL9}`S%5tr6duWY$0{%D$K
zi<mAq3dpu6$uZ<zmEL@+NBL3LAk2ByW6)Zseclu}E%vXrJXxLuUR&p)^ll2d5Jgij
zBwnVQJA)Ls>}DwWGYoa=s7S+Of+nueK*ckNo$jxSo%u-1osg2b&UnkbvW(#B?CkP~
z)550xiUXQ_W#^WgYYYynr-ikJ(~#!1P7y`x;DKtLrVN5^*?&&rLp4q%PESs)`Ak?h
z2gO(XHmiTQ`%Abq=4{^JBnW4GiAZHaDP`iDk0{;{Bf>0@8aNO(i<d`hsd8PIUR^6~
zT`nV|{?YVKfzZ<eBj%@iQ37P=kAhUOAVh!7cYHEEsGu|W8*R@GP)&7`l|6<<18GLu
z!#*Sl0Chi_p#eV^!qLD_!okMa|5|!Q)uJIWCa6-b$rUrKM`=15ApFH)0c2-}DeZqb
zS92l_&DExm_Qsn)gD9JeNwMgF?A`kKKQ#=Y-$-@{q&_%&;}ak}6#FA<f8gdGq9E0)
z72kq;Tlmanz_e-btN7yB$wexw$C(Lgs5FIhq5d9k968x4d!?q`=GI39wrq^@g{z@4
zQM#q#^y!&f>fhMrVA2?4(?;bt+l9?K!FSUKI|3OB*r&HDJWeqdKj96}UveYpe#C@!
zGSj6#zfDg274Iv+{!0eUm)WC2(s)nA_bFm~grGbM{Z2jEnyozQHJa^^r@yfnW$rE`
zS;>)?*oP10vZa~CgMG0qEN{hD&7gyVEwr?*1{qXIvUhasm538zgh%l6K}cc($XGce
z)l2L%R{RxN86PvjTOEvu@57I=pQ>N^8u6llmX;2qjipI`-KiFQqWih>p~S#J#B*Wi
z*Q$tBwzajQ7GIxr!6Q~JJy%Ui&i*MQ5)q5R5J2J$r#9D|h#J>Mqcw+U2#NW_F^l{U
z3X~RC_hc<Y);ms-HjKpyY6pU(GQ#=y8=084?WIteqJpZM*}hpmwkh|$E<oe+mjdtg
zYKH0bf_GGn&K-@KL&dC@G6Y9jgah%9&f>%B-vjNKKIAcr@UCX!rwls8<$T4!s=*{s
z6s&4tlluAAfb#j;aC09-i(o{{L9J<Kl-<c2uv1wWk)NRg5h`0yU?=aBBmF8H@Rbn-
zlx#{`@REJ!=(9;NGnPsDdE(+!zkED>y8oRq9?jOX8LjX~JOR2HXd3cWfp2gtC}u{A
zH%Ne!iP?UkogLqmx;ZsWr}oza#kRnrpYN))-FCZyn4W8ocWa;lGGAR)4)1X9y6POk
zY>>Pdxt(pIiRkrp+vyaOh-x!MRD1XV4E{Ma8}hb#kw3&e(am&Dl_N}p)!G8cX+@b^
zPfC!y(Eo2DXRQcz$S{bFd9L+HOnI`fRiNzWn(yMr;j8Z24$6bdg?;9xpDRjScg=~Y
zKf4#owX1-XBSz%%b_T=286N~DY%u!XDOge*O(|0MVvd@EEOO``zY(Ge5@JkT=`#Qz
zen0d-O<)Pb1+j|!Cf7uVY{8e>Ccb;xd0ZD8TG~u9`uKUJ)R+r@yZglAxHzh+dNVR+
zNQh-(xVT=JUOSd&N?I@D;ntB=4jfLM!q}G;K^qK3?}FWm@8;v`m-n@?F0F&5GY<H!
z>U$Y6^Us*qCsZ0ZQBgNUFVE0cJVnI)`->z+VE1~umswk(Y%;U|CKpdk8T*es8Zk+2
zAHS4SR!)6uF(_en2hA}~+m=8y%D6;Dfi?lQw5g2B95X+itkfQ{zGc&f+V^*V=D)Z9
zpm*<aBiIPDHn)^%_QRZCCy4A74}-Bn2Nu*jZ~T9-t32G}GmnsVe_AEnW2C`4V9s5r
z$gwG{UZ&ML>l<UPrii%(Tn9gQNa6<a`<N{7F?GHdU-ct_+Uq=@{Nl%z&K3LaB3AIp
zf`Hk?V2m;8-qOpIDCWodP>p}}f&!10-~2(&j?+lEkh{^@kNFzmV5pPhulDBY(KAYm
zLbnO_1(==jYvs?t-%-%60mx~Ri2qa8^Lg~~;_U@l&qExz*MkQv>zX*Klfos0!hY<w
zv11(L^;TP7BLmf1+bz(&w?r<<4We+nzWtpsd38ZSkf;03!l*STBw~S)>i7Ff7n?`Z
z8um9JNxH+x-%|u}yhn1Sm6%9wlRFlFN*8jfn(CXrT1nh0-idQjzd|~YXq9~U{s?ov
zYLrBb+2KIc%9CLmadU@of1&J!svKR<%Ph{_N0MzwC7K#e98aFuipiMc73$x?;%l1s
zSdw9?A6VijL2W%P2^@fI4&FTsRXlT%_`Jf%E{5e#F%hM=<O?>Ja_~G9SCXzwwF8R=
zO-cREMzisu4BE*k?U_D5OvT<Xm(yw;KH;kaoo4-~e&wW&+Q7NUd09_l+38XP7Y0O3
z8Xt!W?^QmzMWfiFc4OY#6s1lT`0{RItlEHkjBSBt^>D*z-|ofN+HCf+-R=0{@sFt;
zWz+%FIm9hhB7t|&g@I5W*B^h&)IxH=4;&%k)jZvbE2lrwd&Rc5lr7X1iN61yF!0nh
z|0_h+RWT#YuyZl_H=k@FfiH{Ekb8#bt8BU)SCIq`b2ZF!zTVsOX}l}Vmg0D>zjS@7
zUziDn9xJ`t_vw%xMXlgqtsM?jquI2zwJjty^-tgsHu<jm0hLCgZ48~ZYpS}g2W!La
z=AV~XzESiOvXuUts^c@MT=d#x^?<TBm_y_3%g-5*#;adkb4vwQ)r;Qr)wU;o$OnMn
zn}Fi3xc(jG#2nn+bpuJ2PamTD8BRYed=p?gL@@==BIQW|_+Eby9{`wf^rldd9P{`r
z7u%E}kvu&gT_EBLM8Ufw2DhBt4qP0Dl5yu>kc%Vw-gHdHXe*HrVVV=Nio~SVPQ{cq
ztl>5&-v}z{nVDSb$r3l^u<*ytInNe;&2jyy;><IbW@7QT?rr<74()^<E6sY(gb(;e
zB~br7-Smkmcux|cayj9I)<+=W#EAcAi0%Um3dh9zE)T6BmEhrumFQeveW9{T{q$y`
zp4gk1%AHh8K}E;bf<7_UdD;Bb<GGnwmpD4k`$<n)tj~|u*99fvEZ0+s%S&rpeXz~H
zXWe+iGZ7^v=uektE)DhVeXEl9z>rJ1km2oZB-*Yk%iw+6kb<e63mLQ$dc}|Bv-wu2
z_NV!pTH?*O>=DDwwb&r)oFix4i`uOX+uS_#IME~WcMXGd$MLY5XpL{LSkN`o6z!K*
zLN{8*XSHX)aDk*(M#R5cp@1lutGa)RIk<gjTeTTbNLX2*KR$-kNVl}99CzQh!CHfk
z=X$j+gV?;(m0}<@CmOVv6J8GENgfG@e6R4B+v_Z9qJ*xb&CExAn)N`<|6Fw5Wdp&{
zzH<eJH(fF8)dMLiom2W=u?w27z9>GuxstYRse3RmB%zY%QX}MC-bXXNTbW9BI#EwU
z^()XIMzzlv*o{Uypt8^4N={_n87KTkk<s3=aOkRCNMxskeaWjr6eq*G+<Fr;G#l=L
z3ldaF;ipORy*@;uw3T>(zZb}{&~x+j9lw}H>fT_MB^L_FEme?hUEMxy4e5P^hWws9
z!!?CxgTod&*aQSa{g`(E6;4BOuF~OED-qV(i*H)$n7)HhXEA=LJTLCC*-q8@sp%RC
zHM3xO>K8`^i53Z(2hcWfkBTDa$0$~10F`Cs`Pu6KoO7@f#*gHL<pc*?WA)qw@fVhZ
z{_r)HqP{Wl{${Xry=IKl=<)dmLL$fC-bpk8wIH!*K#$V_CQBDyUO1Y8JkRJZ2)+D0
zL2n_V${CpSAI&|-7?ff%Q|%dG<YB)OL%W|PAd&q`Nh7A+ID|;Fj50Cj$Egm-r(Y1x
zrMS!VA6rtMT<f_*C`c;j(`JZw!Yo9JQp<KJb3I#q<i-8sGol2>2JY;z{c>srC&jKR
zGUuLwqVvrBjS}OED5>nh$8>Y$A3LUyt@V+=Gh)Ia-&;3><8)2@V;UQg;LlmA8#@m-
z2O|x{AY#hgTwgrc*+vN~Ru?Sre#~?`VJa}uqz*koHmpF^$(;t0_hGidOJOyVm#35K
z&_C3Ze-XghVR}5_;y?U0pR-SV3X9`mc0e5?LMMU|Vj~fzFg-|AOz0+cO|hX`#I!u|
zD5F*^#7y|1v;63YD~Erq9Ih4N72}^99hg^70y^ITJ><z>)BR)P(m06tZ6{zgC)tNS
zw!hD;ZQbs^hi0E6-{fmahexzHtT*DSLx{9rn!whq{W+mBde7l41?M|~(~9c{^Uulh
z$|jXIc?OCxX+g>Vj=P|z?ZYfD-O-nwA4$zc2~ZKHlf`IFz^BLMHBRFYRJW=u@d{Ym
znb@0!A8}uwAt7=l(@ru0)5jcYIL;vd)wzxv4*4pVgxeMX6S9TV^YmO($g#XW+eCaV
zglA{^9hK>}`R}1AI^g0J{yFyYyBHw#yWGWm!P3AZ4%3~7go84`McQa>!R+NlCWI=*
zgi)r6JTIHM>Kk^(RpOgDqSMp^S~1$*(L#qlV#3;s$9REaCxrymPdPhYiWpc)F-~)H
zUS5Yf`TTa%RN>JKm&OzBGmV!+YNQ!p4F@Aq%88V2QD!rG%W{(7d}{0E_c}?9xZYXc
zJl;(Y{N$rZ;N?5pVUc{Gk5Z#aVa9STRp`wGKX}!y2gWc?P(sUO%jU=?R{wEIY(E<g
zxAw@&B%5W&Dy`H>4PJcIoW(PlMW3FmM<M5ya(5sg71(J6?rh_Kv$6Fl+!(a^RLUIf
z_e`I4*7&@iTfj1Gm$D(lS2$@*!6kCbw+&FR%RinH2eZZ+9Y<^d=6(Aoj$a5TLdeH?
zv$r#aaemiNwzV--GnOTZwm8hz$0xUS!#}RC&+>U;^P$e~^hQl>In!dwO)c|eTlANr
z9AIpk+rTf&9yTlIGHdy!sd=-}!}o!F;#8*E<89ZHXc`(7{qbJ|ZYmX$R1%H+e5KWJ
z@g^mlg{iSM*(w?I%w7Rxc&SXImpR{$yfC$s(EGb%Lr^cj1)>c;Ew|Pk5H$h9DF6ib
z+PTIKP9K*s#_UJA1d|?Y@99d;(v2F7Vw}1+%U<t!9qFU`8YY)2q;9t`Gyc|Ov(CtB
zTAe$vm9&JWB5dca-n$lnEHyg69$fE6S5;srHnw*wW!Q=UPM-{M3a5YlFlpl{V+;zO
zm5x9C^PM_gXQ`^RiZj0}TtQ&6wJd%B=5D+Tj{haWby0!YJ8RTh0cs{W;2QAl--Zy?
znsyJi_P|vzLOIpQNKwPh3&X>s@~8^=DdO9_;eNNU&4at$G)m#Rohqz!#`n36wb`zi
z5&ELcwV{3yU7tJj&Vy{CD{nnm3JGGy*||uAB=Kyb3c9^-TXAVlsi!k3a;@4vXmxxT
zP+cv<&Pgn$1DO&2s{L}SsPfrZ5i>y(DV3LNe_|q#dfNGcd&0EdpjW@S@G7NR43<-=
z^riARn{C8@9%CwZ>I;LII~Wfa?`|s`#%qxLlTAP4JDCQNJ?Mo6n*jhH3HlwB6oOf)
z>E<+LK7X6{#sCS33{68AT3TC{mX>4GX5#xpW2YD0RNq7o0r)CJZz)czvSN9QDu@XG
zkw}q-w-_0&s(y}ZuYAPQkdSq#i5X?V7ZC@hQ0XW(Hn&EpYUXhmTFx)3GVodoY={M)
z+ivcR^r64%x*HwqdZ=8~h?&z@i{FNHcRt-pK60rhCAPU;iAi4R1{dgV+{O31aCIgf
z(7R1<E3Y&MOUBLBb5BnjsMubZ6xHz!@={^n&$>MG^-X_GEwLm_{d-YbxxaP65DaS#
ziJJ<1escmfwhloM+S;;EQ2~qD<(}d~o+Yn5BoT3e&tsB(#bNTa9LW|5YZ;s##_5nM
z@9KvfCs_U6NI9c?^`OC0LJ}n0010!_N=4a>t}5+QwW;8;<=Y|*CfaO6swyH)catqq
z?)z>rX9A}p1P1OVy0w`l?X77<Y2&zbS3v~9PyEM+T+doDhbl+|a8UySw;rpT{nb$l
z(ukM)go{g)T81euE*?MhX|$LT>Q!t&_tAbJzmwJ#-cH&cxHq7H(kaFNZy}@#caTUh
zenc}5M0Tt8(i%r6dO!Iu2K_BY0}F~q6x&;zeIi^RlfU@mrjlu5?{GoPC`Ac>C@o5u
z=ieS{*3<t~c3DOJnG9b0P=*aLh_q$7Q#V&0Jxk|cJuP$HwkkqfabtWD$qf<bA%7{s
zh`P$Euu~9M&*eAawlt1g(|ZAv7tH2E?9#$NJ#U#J=w<E2{Cd$P@A(j?oA=^tPuI5l
z?c0LMAs|ABjNR8_9Up5op#ToIY_hs3xh(^SFub*PkIYpcXc6YgDb#&$Mo=~!36GZg
zyN&v1%NNaXl;`Qm^ngY3o?e#aTF75BfJrPsWHIq$w#g^0FI$$q4U6qSAct2?xR}ky
zW(qE4`Wt~`J?5~yA&<P-y*WSBQA5!ZCYhuqNt<Z9E?zn)NNywdlUKj+xNutuR0AnC
zqT4mwbOHR^V_rKfN`saLj5CnUTeZVb@eo^Gc*U6SKiEi&$did2iK<x4P<Vgb7a?EJ
z8b0o*`gj+S$HjHt6Nok!z<ZBa-Zr9?&#+!=3(miAmDkp0O)p9-Yap=}(AMVTR_c=c
zI|YWTnE2g%eM!9PQUzvtfFkvPFYEMNiJoD}EM!--(~#29(NR5}0j-ZdyirS4EpcgM
zeS0N=^n8Or%FNP%Qdf+sU<ePI%!sL5XToyqgqev?kRKyjeQx2YHcR$lpUZ&UGrrGt
za&S$T0z}w_7hzVv;O2UV0_E_A-5xrfArnc9-C$P^o<wQVxNxW@C0%T!dTV*}m}V$O
zDl@F=&l{vLK8=9kJ};lagBn|G7^koZv<>zGxqin{Cq|vB9Uj@jn@>oRcax@DUFGMA
zz1nLXUU7IEPq>RF!I-nlX6^HWJtECGtow5_zkLC8uk>@=9~U0;sHk1xgC3>&QJ{4t
zviE+<K+GM#@GInokT0V2$#xgN&(45-(-;Swws3->A=*EuyV$ayE;*6BMh6FJ0y4Q9
zeU7l8U60(lwMKhmrf~Mmx{YYDtB&BOcSB*A;jT2w_=*le?+;doR!#=zFq{%Qao)pQ
zv`G>uY`uRiV@{e#iue0t-}M3)e&{y_6y>F{{ALYI!B1(z+^E4Eq|!QraNK!Zy#}h|
zsgnV)lJ<K1hg;`;mOa2y{G%dI_*LfEX-WCFg}?VhEe=G_8{gx;0k<T|j4(`c^xrF!
z2xgj}u9Y9`{P-vIrr{7bZ?QAy-h}8!rCoJ-!F=pz=9!BZS|nT;kO<od$_x&{C*W=C
zE%uDme3zXlhw1~}_{l0)s;|0o#7~CZ=Dx?iv;AEdjG!iF3V;3%m{?lbU@ltP?Cy=L
z>lzh@_Vj_0fEEH2NX46~4cH}2VvqBSz=@F~@>+ILVBqk+PG+vK>DHbuaNv2K^5pc)
zGJYEl+b9q59(Q$(osLz~b$0dj{UA!i&n*k^x$=9kWjI^w9!Q{Ee6dzA7eqwi(ehFh
zT<l&DIS1X*tgafHguu0X9vZ>@>D5!oK0-fD-_5Vwv7#nKF+2<nePw}qzU%;%d#U?q
z_D(raL6L*IXS|-Nl143|#yqnh{g;sZfI!5f<krKBnPFvtrgbBr5b!~l60N7-KPDdR
zxyP^1EVkDkKo*s;gq|d|6J+M<t*8<IU!@!J(nO5cIsY41cYZi+teS+ow7I&yKS6pw
zGxfx-kkbEIrZB#u#>S+@0LxHT+MiP5B~q`mwH}$AGNa`+vln3q4|w+#VyCRnblpLo
z$z%G{y&%&xAYC}*nt2kv=4P$OBRAocdPcy(hWD{y&N{sCEr|kAqkL1)2|(}XhJOxu
zZ85=E$^o}y=U-T{-{iXChj#XY&nT$`o(;ZTLc!R)e}c{`hfp*D>IAN=>{L*XT$YEg
z)$=%mw#6M^lmxww!tsAXCu{!916P@;{#Xdw@O)7?X-3|$KbxpwR6K?i$J8<}9mjV$
z_WQ+fjz57%C*7T?qLEcqRml<~i&Zgt*r>F%X$}Ae9AbX58I&{<{=MKkP)O*Tf`*!<
zwH0XsSM851YB^6bI~Kzu_Jy$fudt1fq{k#0&GZZnjiR)lxjwG7j73Mc{#4Y3(~{zr
zJ^@ixR9YDsk3QhO!k}Yu&khN0)6U8LeQp=<4MSHZV^qm>mVg>t9C8U!m(Pjm%FgZ~
zu>+TV&daknsSP-|jaS0)MSx=cI);E1v8|-%m-<N}@bnnmPug>>Z8YvvW~an}lwjkz
zP)>%e!P?T|b95`T+ak@z={uD=g!d9N^OX3UcZMOfTaJ)UFqTnv?cvooueF|_!krSL
z7PL>Qw35Xo$dr{3)QKfjq1KkR>n9~vC&r#qbFQ~TQAuoL1Os;ca;+P2Re*sV7WB1l
zk0gMaC;+ex$CLa(u(TYU>dZ2vW4Up6tz^{WpD#|^-B&gyyu1PK8vN8j^)tFp3%?j2
z+0Zcev$eu2|8AUukY`cg<MI?gMS`EZ>+qybCa2O%-N3yw9qru8GEbYZ{T%*?Z<%KH
zu3_p6>OcOmLb9d~F}!W=4#&hhK;N8B<#0wbf~=40Olzdzc&k*m4D)Dr*=10RtC-}k
z4_=VR==YR&wSzxg8T9~`s-h}WZJ*bF>k7;jWh{BV%hr@q*Z2~rn9G~|sW97G@0|qj
zw1<lZ3@5HhRjWR2K{WkRe%5ngI;i^91*4*+-p=nUt?ZBJ|IphT;NLE-k&6<BM@4<Z
zhBdVD3>+*iRTvx@;oH!!d2=mBZlJW0V^Y`WyL%4grHWT*OyiuWP*f6SKlJp7lu^iu
z>1k?6H0Rsj-Hm5F{>~^0Zv@)HKLOjkqMQq^o;;>JJdM<eGo~w$8P^|NjsDi`T%Q0H
z?Jorw6tG*|({`{@$#4s(>cOot8GF}{mT8LjhYMUUIZD`0-W_eL5j%uObGa=JouS2s
zSQ&MD=L5zDW~D!!7c|aV2E$C@?@bVMlz|sxl0!RDNo{^N6t?J5f$(b69ku5}H8MOj
z){^>{)>!%Mk*htFeZW;7v4y#=+;~LDiDZ~~-%Bbvlul}r7~xb*!7*eRjDS5e`9Rlt
z#ivrk<?LQGg{7^--H4qX`bvf6m6ak+s(e*h4V^UWk$u^y#^>+it<D~SzT;zc*dqbC
zo*nC6g^J7Bs3v;VrCHXR(VDTdjanV++}hd~-O!*UpBJ>2!=p|1;g`Z!U{Jw%*^p^~
z!NV69d<zD_`^S=fi#^XC6^yyA#pWvtTFuWuT+S4cjAs1>XOg4Ba`p#%h?s-E|LJ(?
ztozs6QRZKLx$zb4rI6!S9Xh1xc5j^1QvOjpSLyndhD6H}MVDdoUi{<k`{>XBs_zv4
z?lK7c<X9BpnkNS+$T(lHnYL~=T^R$^)s0YPTz{^!#*F**urV?9cc)?@gYY{(GW?mX
zQ+&0w5(Of9fbYEeojFpO!@nEbg&ONya{=QtS*qLK3^1TGV6DR=_oB`2LC_%C7R`Yg
zE61OtPa}q#o%<B}YkKU^ntUNZ(QtbdnG#?~`4<WW?LqopHYX`zm8IpQ?Jmm@m)yF*
zy*NPNT?B2{J+pmI@}6jTBcVe#BUAiI%uN55{BQ@RevSqSlduagY5}w4;>Ehhr8-ID
zar4i*)}#}QT^?9yfKGvPY~WB;wbi-#=%FgdAmlIFF}t_5g&CmB9h&m0b@MqBJfIf*
z#Biw^H5I(=5O}^|__R#>1e7Iz-gkLs4mk!t8$LCLTuSmC7RHojl2tYAUi8Vm;Qmj}
zH))!svW@waCP7DUgW<x#)Tyvv+yY&((>X~m@CbcT4)XHz%{<@9w=`U%GwoFMcl`Dh
zFblBF1t_^qYQ?`15)Qb-Kg1jSe8#tkQJNnwpmI=<tj&AceJijyfpI{8es}SS2G}+Q
zfe;l0oQax*oO(3?-YQHR#1>ls*q2wSr;m0|9HH*^E_hI`R!RRthsD~<ze;s#(igH6
z_jkY4y_%6@1dg9j2A?Y-@z+hY&LQL<u{V@*2w&VE*ufoJSt$~!YDs1NKP@*lMfR1g
z1G^vi;hw|d4<y=*v{EPnuh})$C(S-One7k~6@Q*$sbTR0a3ST}ut}2ZG37^83dMOD
z$DJG(UVF5rebsZg9|Xfc)tdB-o_4}qRTZliG-DPborU{YW$dZ@%!_7K+PlCq+7S37
zY~%{pT0=-nS8;YkD-BBoPt#mYi^MXXAB8wyHOT@XO*v2oH}*`K86mtJclGp)1aYpb
z4%&w>(&%T?PbcQ;-eM?>t_R@xKIy&t9;6NsCELzV@*pXH^<vV~cei-cA{21q>?N<&
zccCFu*ZOheNE7hyYOgNv+cRyxo4*(x-5L1JtVz`G=Htb7zbj@um(Aj)x9l;f74dUL
zWVmR6!1J$FRl~DsCgpv$;5*c3NGpPPk@m7WWJhvO7%_%Oc_GzTfp^=TA}($+19&|b
zF8BLT1PwU&p1%;WT)30;yydgBx?$BgyDj`oiooRQx<gL)AS$ge`vbiKxdf#~;n8e|
zC9&BH0DZ;8%Bo(XHjm$bNq;MSMMA9^VwnM7BmAy|gefC)7oQ~1X?K?GiT&My+F-SU
zLxNDG8u-GV=7oXn$R!QG7~4t+Am~-2FCwb7Ht!*xsF(=dNdJFKJmSE#ZA2Jxx2YaK
z)OFEbC;xCu5=Yg0U45qjnXO8C{uG(iNs{Y5a$Si}Kx}dpgJmbXMwt1C)1~UG#1ipm
z$HD4J_C~3z*yTc<w#0KwI{%#6<&xosWRa&k!r!i_$b38<iSp|Gg3VA8sKo&F-&5;O
z>bDWQYDeSao2!#XW46$j7nyZYQPjY_BH7<qIXl7^Q_RV!8W7t{Sln(O&>1!Tf`WWL
zXKQ7ggRc@eYNjcU6=h{Yb-pv^<R@^iJFA9>Sp9*fTBBM2HXNIV5ZLGZIac^s1_I&Q
z+lZdxVWNnwQ_*t5e=tFvuMVe?2FZd?bwkoau3Qk*Q<4b4jvEBD+G-J65xe!8($F|Q
zbo&<_%G5@1(&;)kZO}D$pE=|vcd^am(fB_^3ON$wHf>J?Y`loU;$L}D8UP3Biza&)
z>xPW7SC`dw_4V^RkZ!!?)pbiB+wf!IT6fPUd8*g0C&nw*0k)}UlO&y)K5SAW3W(?g
z5Uy}R=g5nNd$8=3s|Pj_GK3E889C^=mN5y~VS1q(f0*fYNepq-M~hNaKS}r8Pe^XR
z^ppxsTKkq=rE_df&Q4tYz29N-R1P<kwwKejeB0R}Dq<YdY-thlgU>H6t#@)B=|=|n
z=dQYS7tL|jSA>K})v3OMQqu<CA-M!2I=gOCTSb7PKUlE}I^9xL*>Op<?U$DhHPTVj
zn~Ma|(bfJVAL}MjlFucQ)&QCpy6yX1rhDWmbTDW2;|3P56~w#_n-g^K{&l9ORwC=S
zAif~f-7@kk$DXKZkRt*2-8Yd`Ve3B!gLqhF5<;w=hZiY|@&b&&J%=;FFjR-4LhN=M
zpTj(-X*P&asJe8q`QFtvFbliI8EXf|3G*#uY_QnTIlUUvz?;-(|I*w16GK4amh<ZR
z`n`eu>&lw)3eMA57ocp7eHmFgQ@3+e!UWZEKmKVA6UsNQjO~h7dzPHvl35*yV-)Js
z%N9otcgg*n{|qKuMY?IHScwiY#>-mPTv@f#PkEq)BTbfjii^ffziiD#k~X$1wgMFH
zuxEnBy3MTv#OPFgW1~ueb=O6uT74^cO7a=;#L%>mmi~z{p;^aKK!4QC=tp7{3hmO`
z^;+OI1^C|dNjFUf0z>?r^4-4AWF16BM{jNOz7b8~WKni|MYJBoVzUCOiVb;ILtv`W
z1ot1;^^D(6TA!`of1h^!ZI4K!4TOp_+E3aeuw97f@8V&bkN>qZ5O~;8wgA|gO7J~t
zFlq@vxBxt|6X3%uVgf(0J-l_ltpm#O(eQ}`4%@7JI!2K*15jKZK2sAXo}S_mTK;CV
zdy$f|b_I{U-$(Xoe%6&4@im$`va)j;)(rPZgl&er2za>hak1TFixf(>tA1PhYi6X7
z>(|r@%`={j@tLN9F`No*J}{a|zq7UWvD;hbl8dZ_1n$K;lkM%k-*0%jhN18?=*8Uy
z5iV}hp|`lFcqE4|&jkb>dO3tRgWVs4A1E2^h0>FC=XWGdBc}M&^I!%-2c$-@7`I>5
z*z9qGhCjRuwdT1VLCmxh^>hQX(SrK>rAc>&^b`Epy3H24p|~G0l&Bc%e`(#^4fGX4
za4aO)ezPREPfwI6x}i}L*51}LCRGO9#|zsY1#^XxTc54Plr7ekdV46=HEXBEnK2>N
zFZ`{+k2F9nO)IWquX?v7LEKS1t0GVMDTm@xb<pm!Cdu3MD+}9bEzbovZy_s(PL_YX
zmeO8?4$=6?lj3N7e3SZ3Gbn&bZL-;tRHk@2%Z_}bK!k>fAs-D}i?Hyar;sRR-o-`$
z^s;>_I0G=%<^v@ue0-ubTG{;1Odm)$BCzhf;A??1l(DU6iGht~dA8m;1%GV4-vi$Y
zli|F{i`sm!MZ9lyMr0xOLauy#p9h2~(l<_o9DdvNiMxO&3WFZS?X>x6a1OnNMI<E5
zjdTK`FGVd+4=gA#=^<?=mQp<T>3j+i5AC@UVu1tgzeLdWTv@s!&%5@wvEEgL^p-@y
zmO2Bod4g|j+pnL-98G&-_!Me}SzwN~)kEpkkJyv4OE13G5|%ln6#bvW;+TC9G6#vr
zOmFh*sINacIfygzL1SG{GfH^{RK;oAvxWEXIy-gGwQ6fm%iF&3{kw8<Y}my#{wir8
z*>7>t8^?3?ue$Cc>3Ik|Whb!p(Ci>`AOe&~ckf+rJH9n0w7i>6dI6_JhB5ci(dpw6
zX<S^~m&$H?u$eM`=(W(SW7+Dt*m(@yy@$vW!uPXTb7(l5jc0Nba?!NspFnn$L5QUa
zofqG>g^Cqc1(okeJTgg|9M%@wpUS2ik~NHRW`u_0q1F&i@Y=;!ZOYcLuMuQlNi8yr
zKVD9Xjt;A^TC<MEs<QmZEoUup1@clD)>O}iDVYLMsz(13f7X^sVNF&)2Z*-yxvKv~
zj@UXkDFzdjFparj#7A1T2z6!^7d5=Kw~(JDE(t45S&U^2g?Fpe8h0yeq`h!E^O@hY
zPY~#j^pe#HvIJFylq_Si`Zta*r4NJBQEBg{o%*FX_^PhW^9U>Ch3C7&9|WIIRyX&r
zOz((012TUksutd4M^k0!#LD5(xcdf5wWfMNA#6JrNJHg8jxS-ISSm|t>#&xIDq(Nl
zPx_EArOfUC>%f=q>Ld_=O2WScHGLvULmD`d@hJ?k5Pi;r91U%T?IN$yg<B!hr^`hV
z5s|H)Ao&pA=*C2lih?zeD&Yfa11C#FN(1Y!h13J`jklfwqPe+jovXP|<BK5$@M_`1
zUCO^2&AaJ%b5xs6DMdOkf-L%$-rNF8)au`Ua)`sM{8JSnp^m_tYSaCBUDF-Bm&)jV
z3yvpBjeUq0S-jwp*uCzg`f^-louE9t;n%L;=<^1eq2a0kcM@&5@0_bM7EI;<h=jsB
zCc&MKYhw??vc6Z#HA5{!nGIlS`Hq@ptN5nGOo`1%9fk3bv`!r@4fVK0+<CQ(U3qt_
z$nLLHtk&U8KM8-XriarV>_ebUnzle>iHFQeIW(yI{nanA{D713$HRK!nAoH^-m1hW
zZ?Y1FnFzTgH4#ycKQYWI{j2QKlYQuF<*BJmXrQ+XI(C*K$AuaOQT$^6WCX3P-E`v5
zS{n2?8Y*Q%g&ZCCS&6loP_y%VuF);IO;`5&@HY4Q@MzUPXB4v^73SB2JHNp&^$Yme
z*{_ej3aA`<^3QzMAbDFB?q=^sgNv7Re_I%o+dy&o_s!JB=oruCO&x-2Rs3;jRJJ)h
z16|x#0bd`Q6;@nBmWcJx#}`BJPVck!C-r6$%i0+v(yb`Tz^LOzo~u1w(}SzR0BXa8
z`tTp6p3}G`5b-2x()f6DSN}-9O5*(s5~u_mLY_n``L_hNA%i-xKPc4dyvcwsuPL)J
zqP~)9+g?Eq+S~R0dKGHH`L70`=67*V8F*?N5}Gif<am0La$x+FlE(aVK(90g5<n54
z_bP>XqzoXr5Fo|X4f6ey!1pdR@^>z@M+I>b0lrlN%Ukb}KeZRaDGL2hnVwH+;cPn%
zyYspx8MfmuP)hGtaJ4p(`*M`-m4i}COY5+>#Ed{8xjm{5oN>E6;%k9wLVaNKfeLg-
z1<VC%;e|Jh4DH8+abtx;h64Sg88Y$l(3J%&RazU>vFRw~*mmumZ$b_U$3k9N{KG;f
zoHid*vQ8ORZ-e&tTlsz@2t#@6eVMntq$#SFf`x>%{ULSkMRb+-U{l`9>(!rnK4UIv
z=p~z7+pKc0`20OybF1s%B9*8H@A@(x1j^ek`YleY%@fMVmsfx*3LxjF$}A#Y2pIfk
zZS*<sX6W-mjIPc#TU`zM@$%C6l2Bw778P~Di?J^7SPeXIJ1JEDBI(f5zWAA<VHskW
zEeo;Jb`d(Es;yBTB(yS~2b^WitaM}iu?_KGi0G*{?)RD8Cx58)mY^psZ?0RMZ}IsF
z_pQd_=j~qeQ&1ys1jxdXV9<^2*}d;Iov36f(tz=ehl`~9X}pkO9{69LKd0W9B0Z0A
zT3)L3>qmwZ>H79|NE{DfDz%RR7>L02E0{Sd#eVoy)vz_P&G(q;($L<W1jD5b|827)
z7U;V6uocvEGa<RZdmTY);u~5QeB<Eg0xeE^HzwY7v#{EAzmKPmWfA~?3xs{+`SIDQ
z%XDj>vcJds82&%Hod^K9e_HqyEo>5!;Or7Ec^@vhB`5hzU|Sw_0)=X3pl)zDueFWj
zi?{k+O%%Qy@-V2XvT9u1yG%cLX{z-6&ICVsiTOq0@A}Oc>;yJl)VG=R&i>X!JKx+<
z>fuflty`bnVs~q>lb4xuj2y%*{$f)qqx)0v`qKU@4l2cKc4drjNHa^uPbEOo$Ew-R
zfi&-wGM|b2kmB<T&i(2~qd5dJ*Ohz54;?1p<L;z}xk}Zn=z&6t&UU8WYh$)ycZK(^
z-eS_n5vm2|QI>-{{1hBaF+VhOrYq>CI(Pkgf0I&C71TG+GpJx&?@_+}9&7C}He(^6
zcfcB}xE8Gy&W9PkO<q#`?v-D>dx-m=14s4ZcvkhQnIC*F0D(!tDT76ZJ(yM*W9|F4
zQ+)ky3pMXeyew19^sd5afD0lE9d>p(A^D8fZIV<Md`gPBa8bVbOtcXMr4EKm!YQr%
z3X_1ob;}&IUAq^Tp~b9L8-M5~zoxJ!8JIi$COo7N5f|G=4=2dHZPCcKDM@WeFHf1*
zbB?T7eki8`wDmQUn#wtZx4zw({>~G;{if^ka%>G8lKCQ~YidiidAE3r005<|)M~h@
z7?ZQBhu?qp&UkgQO>G9IJGYMUz%O}{OW6i|<ueX^Y0o!=?Df95bQE^ozka3-fkNGm
z7mU-w1o#4BQ{Vy9$33tV%ar_;)AJ)3Ch2-GDUs6ex;M&=7(&d343UB_1xJg&U}&zA
zBr{<`yKYDjFvKb%eds0UmzQ7qfyB04E5D~N=!eRquKh)!L7us@seJjWQ6aDdNht>v
z4V}N+<$@&|w65_`;+Ilej`-W&;&ZC$&bfNm8&f~HAPvxec#mUoX=zRGvN#AkOT%q~
zN{n=+K=Gi(cpA218KNn#Ix)`De^<F&pHa)TLX_+IaTcc=t(ibQ=3RW2rg^i9i{O<)
znh_Awm;Gl*@_gXPfEp~Y$&hPlv3YxE^E{os#$f$jn(204CIvnHcQU@5@U_y^9r}gP
z1t7&VONtyu@jre=?B)ShbB_|Dt1o_$eS9Tu^YS8If|iDUQm&3J-XBp@=DwF<+B>s?
z7$!Q2Ny7Md{h}`H;)U2KbJ-S_^_0Yss95{<b&?uTqDWoXN-EfuG8Fr}U?gSOxvRNk
zmg-oEYEe&jEYOr)>dDE6wTHvkJ?HmzGm=MF;$Zk{*VXgmwc1ph`#oV24Y0^IMIJ?P
zUEotL3iR8Zx0tAmMi?ndO;hi;-+#oF)y*JLzukTXdL9q}MK;>pvO&iIWT1*J=~EVZ
zbC-ZzPBpfCCKau~qqYzz_vYi7&&X{23%~cUQaan{mdh#CbFjp>kWvqwZA6lZ*Tv*^
zJ(^*vHEtiB=g7RDD+i7hg~Wn5GwV!V%neyD#j1(lsn#?JvgFgV&t3lJ{lOX-@%`%{
zk!#tD<;1Trvj8kxx4Io}2x~`dneV`F?k)(xuueqm!~~5l_a@!3C)w@P@D;Zh0ovC2
zIVNHYaf8@}gg+XA{ZavH3bSdCc(kqyl5yH>^b=Z7`k@LQgsJ8nav#V@a2#(tjJi99
zrY}AyI_YL?-sf-bFXZDbo!y*6uEJ@9Gu`<1>kyZ9+ttY0<*RpD>Z`H~`%C(Nf{}cl
zJyWypK{}~JYB}OAv*^t;8_iF?pu6iE?urpN3qOn(iyN<gl3Yq)?peQ314-ezg5z+F
zYxm#wyRqa%5`$44DLXhABZCm%lF%gJ!f$j`)6Z?-W<x>9rcA>?Gs06G4OO$SW3i-$
zbEg7WbzB)$gh4YXqfk&rmTIOXr=L-(AeF+%IFeP>fKqHlgiJwCVXgn?)Qe6eMfyEO
zx=6Vtu{&gMU4_`ylfOUUQ^wEuK}#`x)dCyr$nv)S@3!tLzh_sM_ZV#^*Vj%obeJ~$
zh(*~^CRNcr#i)_rqDKXF<l{aDWVSc>q~dJ^7ox*xFLy5pO3WdqPXn-aPx~aeNz!RI
zTvmH8&o&?+wQ*Lfs@UQ^R6UZ=L8&7*_p8mCB*U0y-U5I;;Uh%0PRX2WL8r7E*NPBS
z7q&~>(DHZpgt$aTssX4jHdQWpr5_nizj?nP?00uaQ&ug1m8Oc3SCG%gJ?QA-bL2Uo
zb#`}0Y+0Idw|vCcmg0=t>I@z@o7CP2evCRfJLRKLG^tk$dB8z?dC22g;RVR$nI3kw
zF2$lg-!J*0VDF`nEoSvQ2EoWl%RaYY|CcwC&m&2`mz$DX9)*v9Lcg7>Yc;j|@dI%G
z<n#Xn{y+i0J_nBu_{abFAM=AZf586U9y>eR6yp&}!Du{UI2;qaC(8_-ZU=8gf02lR
z2yG;vSn!(NogPN(sIC;XVrfg>dh0E6lW^KUC7{^Z-s0Zvd;HD6`Rkk<AFmxbe~$Wo
zs?9Qfanb9n7M=JcVF?adhv+>Gmpnl{il7u;DNL+8-%c*tZm-nzvHFBoQgAIWxjuN+
zsIGt}zZE{J&uV*#mYi7XX(f<POj>waSoMS~X+397Tvw7_PEFVCVJQLWcDwv<{?7lJ
zk3M*haamH7CBX-r_uRgHm*dk@vPAQ>uY8v6jUDdYe?m}!#H46#xOC|fH*enH?YH0I
z!Gi}3$0Z-!x`mHP7JWCtS_x<yjSBX6_fWvN7^95FRTdLPP;C^wPM7hhz*Yh013^i?
zsBWIHv5|9nc80Ag(lo(V70x=yz&nep993c&^oNwTkm&caq^K(DxCUoEMOl(1VzjSH
zN0xf5b(GelE60O}2mJlN|38X%SeA2ic!V(q4;&w#v6FYG)pfP}?1RGxkMolMOVnhw
z+Qy*>V&hN>rxZbXd~gIW-e^fKF1omhIuShmvp%OMr)*vF7-L9dY(ux3ad7mA2M?bR
zlx92})9rPeV9;(SVQ0I;MmMA1&pGRlQ3T4e;INQt3nynMgrGS(I_8BJZt%f}ce#A&
zDsTPh`@H$)kNMMYeUsypBX)MXT)nbKmW~M?(j?*fwW|yUmYX-OasAp2tdEr`L6T{z
zz{7jT93Fm~hxZ<G<<ce6PM6(-9sbR~{(XM<{WrOK<tlr-J5)u1^FG>@ATcQ}h`s3X
z<5Td8EHhlbynzDF`a`y^UBaWebZG}yIxg)eBxwpk(c6&w<Gp9^@?~LEwC3TXL;le}
z{Ac{=?YH>f{_oE9D`{S^X)b&E_-tEd+O>}_k~|iTwa%QD!P4vJPCOj~TdrT9xqgAm
z`}srER&m*B6)#XsOe&~}?V}+kOy8eRSTVV1mJO_0&N}FNkv&4qe5*@j&G+>*D{Yhb
z`D5$CN*jpxjyy|95*<|sqVjp^`I~eer9j~4zV>zQ-FZN`F7?wlHhOGqY;tgT!gw&?
z^8Oavn>h!EXWYJf9~UHrhDi+FUI&FCQ7OOh3x9>LeC10N;}IcvQAq?ZsW^;ADM?#l
zt;0Fb!^g)Q9G{Uy$zog-JojRW_lm+=Mx!yVEZE$T<cvN*P@3JnefD;@ah2oJ{*DwJ
zN)2a&5@!Py;<@EQzy?s(<6WRCDn>~`RTg;f+1ij`$q>Mnj_yXvMyD$(6lF-0gpFR0
zjZPQk1IH&PlZRPFq234bJSR<4QMYLwGyTX*VoZVtQj<`W1rp04YD29Qd6qL6oRB3F
zB>2N0e4ju0lW+0x(E(@u5k*yz7(-HJIO_;NSz4;9BFi$;1e7XBGR<)4c;nNbVrOdy
zA1Y9RP6vu{%I81#HEdDx?QedQ{%}Aq-@tYiFFpSvX_nzVl$8}VS%9O%BQEWEzWU{_
z^4ZT8G2&fZa=o?u_{Z;{O+wb);_!6H!MDGIQHt?s%qy?H3IQrx@!;V@6pmi6%l1Z(
z!=n?1!-6bNdGO?bL>sDN%&^!XOLDZy>GU?}c01w~xB<QGZ4x7DFq0VJXpL^%)2(wn
ze00pCCx;A2#Z%?(nzd!NGJH?LLZUHaA6ZI()>XMIgIP7o+^H~keo`ysDHLe)LdE2O
zQUp^cKlCW&=aAO=5Q&KjpaOMIZPipe>bwpp48FatPX-osHM_sFJNR*>mJoYsTt3!N
zb<W6nb7wz}M71K72{D7NBhNcjHcAfO)9ZD)w7<`_Yu9=8_0RGChxZwb3toHmWsZ&x
zdGygo92}nU;`7hZ>!y^^keeEfub{h;gO<dK#i-!M)tlV@=tGVV5Bc<`U*k((_#8KH
z-oPXY!6Ybj%xD`VxX}fkJUQg;_iuCW!7)B)u55H0#G|#rXvN9F0l^OGbR@5Hl5Mcr
z%TXFOx;fXcY}4zc-2U)G{_x*?gCD>5p%}ouct}|nV^X}wE6Ee3QIXqMXp`ccC3w%J
z{av;;x_D<LeK?8~Loe&F)$1}Ii;51V=ytkPmB&^UJ3E_=x=R^_D?LR~;XKipU!g7;
zOqz<@vr;H+*xuSEOH;~X$msZpt(_fiefSaI_~UQMwO*trrvqA*R94dNmzAPFsOVZl
znr@I7OI3`~%5(4jV?xNpc&h`DF&<eSJv^ea1*2gff~P7=KL5sR*dWdb9vBUWU<|oQ
zFuCFV4?m<RD-MrO*xTP@Yjax|2W_}{^BQRagYg)Zq~b{z6eKx`67Q}wOSpRV3jMPa
zu3f#ta5$v6U(z28sZK2cZ~>gx9374rkB>=o%02s-YuB%`v$KOI;Ju{MkvOD^L?;++
zaNgtal;e`%9GCVtc<+PZQ@=#B$R8HHqB9Km+C5J?T-Nu~b}LA2R*kf%m-FPAc~#Tm
zJ|cqBdeW$|SFHB`S-j?mLKP^}B7rkDklKLKa{f9O)*_ed=q_iPwxyxj&vjtk_F?Wx
zT-Mv7@2h?-?{6sqan5q*?nev;r*!RxR30qK=)!C0_6(h_;jOoS#Qlf22*Bef_oBYP
z<l5Cuv>x;1a6qq{lO$tad*u?NQAxL((%bB?x7EQr&*e+oIA`hi5BQTmxy!w~@A2Ae
zFU8vH36=9GZE%&PvQ^F0i1bn&F&bMc+a*a8aE_w17^UbX3Uoy^Dll0_RUB~V;Q(Wd
zs7U%9cDH*3U+~4xzr+hK?&G}Y(c?qLRYj5*abxvTh{@=b#E3J2l-a#P=_>Lp<Jy%=
z<QT?hCC+<xwz|CX(q%TfUAn!D-Ubwf!+T4&)1}`ZQIsX!ZjLcV2y|6oD~YLro<z!L
zMq+K0!k9$rtrIAUF=bh?y?2=+3;@mJM+amkA@63AO5VGeVkTj$w*h$4R52co8JzUV
z@`_+BSv2UM9UWt>qjC-(gv;}`!g(bgc;2B@pcs#ZKzYr<$+4&d#|2r|VQaI8)?Vtj
z3`Uf;qVxC>X_Aqp37eanJb(2Hi4NSl_aG|q6iJ$rCMj8(61*c#5;l7|*Y>aAZQ$%=
zK>ut=Crx8gJ|Q$gLH}$(2nv;HoDWpiF^aYqS4tbSq8L~BRHKceC<=&1*hDMRG^N+=
zt`(g=ktEkZ0&4!{uaWrl@7eF6g_PQecABqUT=9Fl&fH|y)2g7^<Ft>@x_?b;lURld
zRck(qKu`fOMlk6>)g88Di&pW^(~78?AS*LjeY=IP$BKW-fKFOF-Yi{Beuw2Vn1V%m
z%QYY@`CPZ7<<Hr(oT2p$EFu)^FKj6Rsj8A6zx5`2dpit9N3m`;5pjx2>+zHKIXwE1
zPFIts3B~v!8(l-PVL*?tZa|s^%4*2QM&Pq=+=$3uyzY}E!8u2kCRp#d{@f<s2JnF=
z2Or{tCyAVk(wa0$#3LhwSe{n1y=B<l)&xJmX-`#JI!Pc&!~;tO2ue}85y3mQw!mXC
z>VRRX7(EfyT;4NW?WH7X%JmyLNh11jLLfDwHVQ$Zv?e%_hm}$|??`ltHiqLT1BRyo
zoaMQjd%XBs7iTMyEXBFY(N2)!oQtl$0TsY{C!RvigOZx?#u&U0R3`)c*;wK~Mq_$;
zmu|01r`G`$$g>{wc6fOAHi;ha^Iv_HRBH)d1o}luSy{9(XahUj9RiMkq~9M7hje=#
ztaTJ4OJzNkH8|%91gukduRtlPQeMJtug73~LQz&I9GzT6Z#PR>oJhH20$Up?)*88I
zzMvcrIW7l`2d9jO1%vTehyYS!#w3yz7!;i}<=W+Ks*SE>88DXpy`IGSoLKU12a_nA
z4Gcy_^b!nI)^c)kDsxjhs_rBcg7+?}3l-yWiH@Fud73a#o}#SQesP?d6v(Y#y}8$}
zMLgOnn>PBNZ9fh3C_qqVSa@!t;Amevu>?#LAwTt9Ekquz)wPM4nk9fjuoN_9YNva0
zn8w2`pm2bVpF&{fiJEWCd46}5QYf7e!n6u&3LT1;i*(bYl2%dMI<0n-0CXMdp%#8!
zp}T7xU)A}#$U3zCU(~~*XJVDSx=O-WS2jAQr0qs8XJ>0tL?|F0Scx%&;7Af9k;n=<
znUtt>9*DKZdudJxP7LAkSRI5{*Y@#=mKq^OQlW(CPNB8NS>IeseZRFlJOOE|RX{6I
zIT;UWZqV`hFh=riDHNI4O<ap)!woTtUJ!Po4V5iX%8_;qXcZY44c?=b!g-6vkmQM|
zxT+FuWUq|&5X8+E<<Z*G=}1;ktu0v+v5E4efOiVzH9kbA03<0~i21;rtz^Bvq#TXW
zX^K`-qb2X8y!QHQZ13&>nyRRH>nGnO>sYc*DiTZ(f|uBos!~+el4c1ekwC@}0y_0%
zyIqnj!BvLMUPkba%E=xDm^dUTLqOC(b&%kAJd#xLCZVztCsL<AQ;6MTJRZk%G~ld_
z8FMA6W{7Gw??aOcMQaoN{}n!1w6e0Ej^b+81vFc`Io4W{G=>$OuGG384aejjd?@Jm
zh3MA0Zz(~nDMnsAvmzTS!inMl&*AZLld@+OG@NezdkGGxHPzEhYiO=%S4Ui|P+A8m
zo%ixD;(4dPtA*$0kG0#iiG*sMlkaVEP-FF|Hjq0DSj340h>rV{bx&tKP&3~f66(Q6
z++PchK}`q;0kREGXSba6Q|l7f(kZrO#YMHl7xghg<g2X<YBJ%e7BxAZmARM@yvT&t
z=aw-PYHsId*@!TAzBSs+nij)y8%X?5)HSl-K8Tl42wq-(@1)*^lE`smbTd?9G$u(f
zfJroomZ5?djxYp~K+qbC$GVC%7*zDGGR8p6+NtBFfe<jQH7+<IHcH{5%|a_-3u{E=
zeZU1P0edJ=Lcp{(c<0f<fDTe^QEQaW@XAThBZ72|619j8QXw$c;<k(jLqe0<JxH=h
zg6P|wv)HP{TaVCGMbI$-Q;Ace)~Z4DlKaa<PA)+ygEpFeWq9&%h_wY9X$sv9%Ch90
zH-Es^{w1D!<2r|jPf)7j(%u%%OFpm3L!t042xl0J>)2SQqz(i{f%Abx8+4HGCNWgD
z5(H}uAw-Tl1e|juNh;iZ2xy~Q_W&v3i^L0gD+J30-*{tr?=Xpud*+*G*Eugy`7!9!
zdPh}B9wcofUIip0v~}VsXCv{gtOe(BL3s44@{)T^a;sosHN0_Ur5=j)Vo$5Y%MPV@
za&)j}g2m7x8q4+gK~216=3me0ai&SaVztyX(U?Oo3klMqu@)a+hw#+n%zw5+{-);<
z5uA51m`}&`)ym8EjOPi|<oje*HHr;P2*%tjWBsb`uEivFW=H2`q~#D6+`IW%PMC$+
zd(fWu0#J))(<aI7aYO5x#ZSm`P%~}j=QLpfCig>=M1sE<R{AU!RV6Qe6!|J!5fr3a
zqtGZ92uLv_oft`&7P-2hps<A?@1RiDVzj~siSr2FW1`BYu5@T(U2km^S{ac*fFu<=
zfI-wsN^5+Ocn<MiN(nwtI>}LGv}D#r39?d&!MYF;lQ;)xrE!rE2B5S<lvYjfWM(v#
zb5ud37(zg&hRR#2QiAd{Iuf)HrQiU;Gm(xA7=?EZ>pi~mF{sjzrW&Ul1W}=Zic#ra
zq#_ltW6z*pqEMt+hDkIo1oA9rV`qo6qeq-P{D3^~V2tD_(^}$1LR`CdB57IYP$A~-
z!V`kUIUS`rA6=szI?iKUR#;on?R1-z5C*w_0flpw82zoqThQ9zti!9?EgI6)MCS)d
z*HSYM3doe9a1}mOF|CUx1W{$_=&G!hCo_<ml%T-{D};$af+6e3x;&o3C~yf{8!GGQ
z<N<<s3MM*<UVIko11=`^6wWnJl%B*S+`V%jzrOT7_2Nyv&Moe69+}ry%p#d(Jumyc
zdZP6R(+FagFfAhz)4vy?+xcT*uFV|G5ZzkcHQ9~%y;Mz`Nfd}hP@8QT$D8X~G&VT!
zRZX(ySJv`nYQa1)Te^(7-;1u9-;rf|RHO9SXJRo~UEa^)na@6LZCll%<ML@xBrEn}
zr5mrc7UxC9BAl@pAeGXwym1^%DO~WVpm0_s5$`2})GIFvV}uJPrtdH&Ay_NaRDnWA
z<P=aCQIn)89#2q+#GwMdvXZzjH6m?zP)0amYb`-*wD%-BKnRg2_^7_}4YKyiffu4>
zokMBxm4~265+NP|&{6fIm3Z!W=Rj$U))cm+jyg|LA)ujlPOrIC1tdC-rD6$RO$aWk
zuM}nJa24Etc*GBX^dUtVs2q$&C0}~!2A_HU63%<Z!y%)hU}Iy0JnwP&#`6qEr}Vb=
z=xtr%@ZcWCm?r96a)c>y<<%OKh<eLfM^Tl8Ap0SR9SSG+BLojomE^o{YG)MAHK|yf
z^Nnk_4_*kgj$C2zIHXQ_{ZhN&36;fYQG1SU8SN6{ZH5X2WLy^lAvn+)WfD;0K0Yo>
zk<yfs%*|eiW@V+qB`#1oD-xDgm?XiuAQd73>x57Oq{d*KrB0nAjKrGMLqY98T1y%A
z3s@t&*o;}kxJp`RUpI?L+sJ>m4_e1p^QB?owey5UF;PiPX1yLEY5%IPi#24Tq%K+L
z>c8dluM_Q7zz{=;s6Kr#D5puv`gxO!Bk$J`9Rif6hK);2IratnItz<dG);Dv6Oh?E
zv>a9wk!L%09y_vZ#MN_I<d89el@f>xfpJ-(ypkKB0?HEeT8U~(D=~m+WAG8#_#g<Y
zcD+?v@+9GjNH$<VIAv7SdxuyqS5uW1of$k1lbBeBR7rj-0|^=(^D#-96(JVJBocWV
zR4iL-r(QdG*(Lr$5;WQv1)oyqm@>e4Tw)R<F1o<Dti+HWAV~~an#EjOiV#FaR0S(U
zGd{qjt+0b}W77fI!_-(Y?gob~19v_;rhhnQJo0>S?-3u|e=I~72;TFl>z4pUF&g2V
zqr1@~5I8<Q<o=!8y!_cOVv~%b@|2ZDs-$R1d6G!^VCzL)qn#7Ap_-(Y@o~apKvtr9
z)R}PN!Q-vC9!oVxE2&w6C3uH&QMH8J104)?ajiOqjrBtKtIEpe#KI+F+micOSugI-
z1kyAU)mM;sf#Be*8c{ln_hNU`T1VZ5Ff1NXyJ=OYj~9YyD~Agq=0Eeb^Hv;7HOz%|
zF@=tMnq=S8;aqJ3F^||+sj}w#ZGT%tgjQWUArRBO&FaY)6Pq^Dn6z9(FlN8cUOzpp
zI^&@<sIyE=6Q&92Rlo^~LlSHxH=@Ya)CPAAjh!Rj@<}g>NkT^_84}tRG?%TRy?4|5
zxC&h_+KXukCM<2YsLy40X8G6?I8O?kPbjl|yk&CFG^`{bwU?0SU7`#U34qcD9TAm6
z^34chjxVDRxXLw1**YKflJXc-G3wb7P|9MAcpe4s$kP-Zb?8PES6D0MXvM8H5milG
zT~&!1Bm}9nXQCBdr74w@DDbkhC>6-_43#QWh?0fT)R{iT3rZ3!DgvFla$T%u7NQ>C
z1VyS0K?iiA@HU7l#n~80C|QQfqQpB#YEp@5K@fFqqd9nZ%A@xW2!#b;+i1FZ%E*Zq
zlLv0yd(3M$uV5_{m1lo5r_=2)7>}{u@!@;#@ZuX^LL+t^hj<@(&m_X#Sy7=`PZ$^I
z#KhDiqEa%^#^HmfDkS)@svMFgL?Z3I5BQ+KRVX}KNnl{8#KURA28vUKbFL{srV^1}
z2{h7Com#>YBqK7h){L`+C>hG!VK5ruk^F1QL0O5nREVByl2zA7cWW?7B7u#rYKWgy
zv{Q}LCx$5rthCMnnmR%9`4FWgCTfsbl+h+Fi@5AriDJG@nEO6C+y1urMV}`=ZIy-m
zKg(;+BE;pKcD=57-g~~Cj6E&o)Lvh+BzUrqr)Y+eSk&aBZsq}}6>1wyKT0YVOla+I
zsNY%Dl1e7CSazQmF#yY8_5RHFyG}dW<4;ksxHDn4?Yc9Vo<(~f7D?W!^<^#*W($@R
zkRYmsT2<j=Dh(e~QxxNhN^7#r#0d3@BuRw8C>N{WrIFiI5y1oTGkA}RI)AMtC56&b
zi?VQzU?o5=tt?4m#K>zj-g<0sl$B+p+aXIOd7;Cjf&>zVM0DOdQ5Yq;kvduu3I@@w
zLF87;vcg%Zk)gFnRZ3}zde<JKG-;Yjsa~BBJe0nqbY+y70!E8L+B+B3Wri%v$u@+r
zSZ9S@3N`0#NVEa(apMu86hrY=YS`_h3{OkhH?6pPbi(g`=f_l4#YW!YjaLlH^x53V
zc)o%cp1*-sn!MA+dlywST56U=gjrRNEH{(fWKO)QY{PM@r~-5-t*D|@m!+wwnVd&y
zO`=p}0t`~7ST#AY#Kxf_5?6r1I*}|AXUJ1wK1`O%afGSxfoNlr-&G~)_6ZIjptP2<
zw4_PGcsM2?l8besTCK_wZ4^c)6oq3j7?CC^-b-qlJkN2?;iHFJof9i~i2*6gG75-_
z<FnD40e^woQ)roVF;5g0sk)ds7M8T1y>=NvndU&}dCS_fP|m8UqZ6VLTHj%UWahVK
z{@5(xS%i92s6B@!-dD5jCK8JDCYH*HJADWcc0fyRQru%m`j8}$h?8X#Kq#`yB4hYs
z&igzWU?t&LNqgq`|3!0A3!5+Lc%3#2?ucqVBy%b%iG7GCR|k8plt8o=1fe6+t2s*X
z%o2Az@5E-Ib)7s>iA^BU!U9L+A|-JRNs@?<zENapLY5J|sRA|z2@>NgjFMQ6%DMPE
zqq4T)Sjw`*TTzh&B`T?0r=pXuY>Z8i98HN<h*4F63+Ti|mL(8kpqzJ}%2tvE*o%H%
zXNo*cQ88vkDJZQZ=<^{Ossm^*1gUmZu8NXMjVWw}PBbQ3qJ#5fiINEaEJGQnLLf5<
z`+K{L(?FpuWd)<cQj`^U9v$G7;>vD<S6$xx@lSYicEUHl|1K}we2&k5?o(XehdjxI
zD0?ARi810q<sjhjUcAK8GzGw_=-M3Q{-WiY%Ew4=E5`o-DCe<W>|AMPuvJA_ib^R-
zWM37K=c(A0tf#0-iD}UZMhQ~4&M_EOls>xCdZDVtxL`Ogv6aVK$yepP7b4?DinG?n
zvbrJ>mDWZE2b9v$mL$)T_W^4IO7$DgvsSUz_PP*fEh;lf;eUQAlbUy4B!SGnz>5)m
z8%c*H?WYaAtNUFf2`rX2CK+z0+XUKZM793f^I5br=W*e)QcHbsvc7q>n*#C}2U71-
zCJ!+VjYHvsrqWU*N{uD9)YV3lXq}gWcLAdUI?T*r#qKY1S6&FM%NG&X^I_5Rq2>>*
zClIYU&Z0w$&bJ1CCJ96dXPBmv052$F0VzQ!c_$M>6=V>l6_v9Dudo)pPnv>KK6vqr
z(i#^mNdkG^i8^Z^?JIKEHRAFr>DU$BENk?cD9Ae!Z(?eWAB7D;#H`#%&Z}4~NGqw9
zXl;p3WvNA3F&dZR>Z--2k(*4^V`&l%y8&AYkq}jgLo3CwKVmd4d311!QVHE&PM&Le
zos7+`4Vi})53=ESjJE|!2lA-4v#z9afliu|#HxyE&v4csqsjT%zxH)r9z5XdRYhqX
zrza!s+&<*T@80F4KW+?xi7||eio5rp@ZiBAk00IVzxa)>(CMhAlyEoCFv{S9$LaVY
z2Gs;92Ja<vZB#=-kYGbC3Ip+Q@lD|?g~pbS%6pPZV-PN|v=-}Y*INZzQ&ts9E4rN=
zXB}1Pcz814<ZMV;ImU%#UEb(*8zNGa1+Mb=T8dF(tEydk363NcM}{i8{u(1SV?tyY
z#4Z<Y8y-S%LiBZ=Ta2iJjB!RSl~o8d659D7-iPfBu{A<!bL%yNq&d6%MVvmroM6mK
zF4KfzT9TMPe;$#UCN8Y_K0EJKzXf92+&USRnvmf(w10K+BseOhj?PtAu;(nLPO(Ze
z3WgKaVNe>}NTf*>Cc&KnKttN`XoO*4;b^n_vIe$Zg?xV+$bMC`MdI5W%*)92eK>y#
zE&J61SgFFYQJtWTc&T_4Vr;HeH&I1|YKp&@Tw)`%k{c>U&r}RqAZid4)wsk5FTri8
zmTG?1lO_quD0+DoBbX(bB1%jAfHtU5yOhcf<Y|IY;;rn1sIs)GjpZ6tO~j(^Un{H?
zwT@Cyxsu9xw5>$_5hO2Fu;M8-91F70i6QG`T)(m>T)dG*cCw_2>8O#YilxxfG$Bz^
zO|YBi<OI?fU7z<ddfPn?9uF85V_cB*?;SVB7{iS#TYUPp>lD`Wo$tRV`=KH2bR;pL
z)N14U)m=Jyigyz9r?tdTjI5+pNzx>$oFqQS1@SQReq!uS)1--KF-d}P66BSc$Z0D@
zmZUfYse31cqt2_;l&)2Rylv&g3rofL7@(8qB*xLrJ17EKE+&6#Eu)dbXCir&R$^|v
z7F&)<QqV|gV<*9f!N+tmI_k-l!59@K??BLjUN09<GzE1AWJ%Fts2vT;Xp>4~6#QCY
zAn^qegzsCg=d?7US}ki>7F;;}daqKWv?YaE<->HC*dhWk`8|<N=2cPi_K`LroSo0&
z<FzJa*@lLhy-=-v6bVEK2v{`7I^l}7q``r=>?p-0<M_#Fi(;q<j#^7RB|_!BSbjRR
zPfpKc+5Dz=Ak3dyc70gY?<}m#v@C(eLhh1A%NoQvv>*P63!0k2OJKDMD<YQ=ko4)+
zlB7CvsS=!}^`s`cXre@~nWiQpXfZe|z{VI46(rJG;yL6cGz7>5ktMOty4X|1ZCE^_
z0%fd;T^mQUM2z$*8jX!gM7J;al|0EvykRgJ<CFw1c4O>CjBO-%antod4E4Rdi_zeG
zg3^kjDn%OVbcMt5j`6s_x&VpP>L|xWOu_@f2huDh>vphJMNx?tmyU#_D61wTY?>!=
zeV&uUW45k!D8@yLYfht1R|t=-Dok`^)k@Lp8b0;<4U`FZ7bM}N7aLK3JSGIi_U0y|
zVno2RwYh;p4B1uX8qV8#A60acuEjcwMx+%JZCF|xvS|HwS|LRjAO!KWPqnzlN}LTS
z#Dgx&Gm<pHIUgk?1zz@{vKFnxy}9z@CLEl?BQ~1KdR$S6y3hq|WpVZMqXI70Dh(mv
zVy0&2Bf*c)r#J|>cwU$gA0#l=M<OVR8U5NekVzicR*H-{uDbnuwHiubT7p<6H7xr+
zeQld)P4_cPSmv)^Ml6=AYUVlo*gnv>(9=@;sfGa49IJ_l0uqspMIr@;O5s(sf<#6o
zK|uw+02~us6-Uatf!=2H{%eMMRtaO3nrBfTPeDLt1cOf;R`t1T?1sbn7zD&yM_E+3
zL`TmdafS8K2rYzNI8uT<0zUfHJLfs;R~QtXEF&?RBD%$<5$QW$idseotdqwn4}m0&
zb`&K#e2^Fe7X(2D!Fq)HFD}qJK`5fWGNeZA7S4N8BZljMl*A0d<E;WM(a?!W#JkIT
zDpyey;_~XPsMtafgKH;AB`(G!bW<58gg`JkzF$Eoo6XlDkf*5_?gQY&JrzY0r?9!b
zA$hNC5c^1$5)c&~Az;0wEX6kDV;qXq6bGd&Z+z-`_V+p*oty#+6Yo{f8j~eB8yHum
zWJ6V&!B}dZXcI{Dl;9%QUy2ii(lU<q0Uf>Vlu}gQ(oGXmqs6sbYr08>^N#*-M4o4W
z*iy2tA@~3uDhD>&QmpfgV<}r(N}NbhiY;YamW)P~NEqJ6fJz_LRN^Y_y?D1lz{hz8
zMKro2sw)IvBwEWltdGaUjW>jtzD6N|uhC=A#(1ao!Nqs3Mz#}AqxNTgj#lew>0<)(
zlEos=sQ?Y4U4;Cndz~dF(@?h#t+tan%+9NUh1!qlG1`3lVxrt8uG8}XK}<P;Q#`G>
zb9oVCf6hy)AZJ0j$-dTt#?GaZB#8v1J+t#9h}n6b$JsCIZzU0F?a=H)aW2fxNlkY-
zH>U;H*8|S-F|+dskuoollvdh6Bm!JYi24AXP8L0tB=Q^7L^Tit{v=IEXngsd2k%RI
z-8{NiR+QFaj6^CYiA+rUs9s347`OveF|Z=hH8+`1yWA?R@K!42S!1HdkhtYm)?$N4
zYmpR;N$7S&jpn_$rlM8!R6>l|O5?1Bh+M?e$HhEc5;^aK!x@nnvQ#qlO4bk|peCxc
zloFTIPL|`HV;mFu6}FT>IuyZ+3v+L?L*6lH3wbAxg{LG4FwZhvWhDkiY0|{d>2%};
z_KxRn?(wBBy$U*z<rx~OZw`VSb0q{qUMhJFopTi9N{DJ^Bzd7yXd{x8ZJZRmNEiXC
z$}$*@8IB5i-5f>0Bnfy+QB)``wywg9d#y52N{I<0%c`QXk_@tp!F;X~+lTdDf+Kr9
zI$0WnPr+8cQ7`!*DRU%>T^Ne8w34F5M9HzN2vIWjRrJVn(N3YDvQR~L=^CTyTKy{2
zXAJ}T<V~xM==JAfX=9#n)Ng59%9xiTCN_)362_#D`ExSHs_SXKFC#!R?PieCxw|x@
z36ZS{)VlPE-Dr_velqVa9*@QWNeJjVv03cjIYm{9tG9|a5OHjQjV;I4*0?FeHksD~
zN$@-ZF?U&5(CjHClr~IJzBS=I+O$qz>&(x&={=l*6%vS|+w0KV=;EwJX)R1Fiegkn
zNx(-h8%WZWZYM(r;p#&0Y;X5ab!|yJK?gRvxu_7LcbAGPfFu#eJ?Fe2hv?JqU67=W
ziD{ysRUmq7DUvLeG%B$SZeo&Hgi7@OSto}`5QqhweDG*v$Wy64Sm($}^1R8>6oMeP
zP!k*ii87NEApwblLA=$HB*CO&Oh!#!x<H6F92M0yGMBQF;I{5YMsKsj*<e7Dq>PJ-
z{%}ZDRcvo<Qj8`2Micj9eboQMbJzFS-pUDB44TYjGM<viZ+s9}<k}_^CBu#p_gtr3
z6#0eV8LbIpN#<w}Z>yp#8IMa2j?VC2aqa3p8{HJ^9HVlC3KArvQbUrZRO6_=G76n0
zBq|}%qP|L$me`FjaU=HL!`>EImQq=f7OTo)D~}Hys;X+d{_0Ex&Upy{RY~LGJsggu
zF1vT+n;nV7FG|n~lNiRs5?f07+bmC|9*C0wN9V*@qRy9f?t2ILsYHPF(FMq<g(!ts
zO$fw<h}3V~<Z~I~f3|b;1ch_1X%meJ;cFdPgnJett{S;E!Gt2thXsAMk13!>zyW1*
zww6E?D3LszB=#jlkMvqwwY^hfbS-_fxotI#IO|3G_I%o`czl_g+dH!ljTCYoVQO>n
z%Q^Z<*NcdkYW<$=uXTR*Ff$_KLShxByQ#l>%QG6$1H)lKuUm>bz)8ySpoB|Jb%HTc
z<t;G^r6;K?hGVI7;hm^)6dq!#kfbZ+MaQEtNh(M^NyW2@sE+d1VUn2L*Hw(I<Fr3w
zG#Yc|(jI9lT)gv+G)t-LAnT~2a+Mh5C2fcho5DIt7N~?6Nm`PqjzcOS=6Q-vV@!va
zx)?@FE-&Xj7(oPSBs{gpR+@+eq?DxRP=HRvL&zi53AKtT&*KxG#=ycPNqO;=7kCs7
z*}eQ6$K?Z*F=#Kw=b#jW!3Y;U%c`ovTB&!kzq?JMiRu^MxFv%^`I!7NW*e@G5}l=j
zNVK?67nKJuF2#-tf~3qh)==271oR~dCUIQ7bcsAmW4-tx{m~F(G`&uj+@u6N{efh_
z-O75xd1~?81B1#8d1_*mx^C2twX1IMo)J>_#U>inQ_+@@Q&mEkOKT~u1EX-2Cru@T
z0YP(Ve*+X$u9CDQArO2{K%(2rqM~dh5J`Pf79~X~YDkmzxc@M3+=N$ykEpGU_9jN*
zHX>>(g4Em_I(bu9OA6~F{^g(ZM)YM;%>3M2u#DSYbhx&Q)Y3<&<31uzXfB!eKXX7v
z;2I?-X!2HmG3hU+OR@?d4d^}&WlW&cu{1`J2Fb%zrr$+GA<Ta9^8#f9e#M@AOac;S
zuU{7y5%;iU4PoY3>vyw!_iO|t0K>tEqSqzqbz+r1h4WI9Hq8<U5^?U8AS`VlXeD+N
zD|wI7B*EHJQm6YsWk*z1Nl{rgH@hfcJgyq<IL!>5JZm#=Bw8xnWgDAlG+C0+%Q6O|
z0&k@zc&8)WcUe{90aRI%#F3>bX(S{ez-S~D^op34+&Qs5bg~SEqVk?z-oZLYuz@02
zyo0hR*y!bCos24`HmN<)tS|8+D*<{TpiFc_wvxANT$IgnvP6Os(^R2@rgVYcMwc5m
zZjg7kNO&SKG=RxciF!|EZb_0#RDD&DCWftz4r!7=6(CVmuEJH0G?mi0Drih9#%Yxp
zI(bep8dH^)va)!@Xk8VSG)qwlSX;*XAwiJ+(KzmLf^`LnNhxfFG6rK(3B)vxR3*{7
z4IUky((Ps_0;AeHi$J%Ni>j+AD905F&9EqN*0Hy}$=>!RCRLmb#uP=t?)D}GD6OME
z7-Pp094PnAQIw(v-QL_FPf|>xsVXh5*OkXBFIju5Ql!#EjOMP2uFl<zEYlQYOO~2Q
z?A8T0M=U0uSJOmcLOj~GjT(_IL*mQTP`Wuk>AgO`i~ueAoI4g4w|Oc=wh(ECiEc?L
z8n@m#2}CvR>%F~-MXCU0c)-*I-aA4JoOMB=J<%=(N(WL?MdUX#uPFv$1(JR$19`ij
za8BDbVWG6sJ~k`vtwE3$4KjPbW}A5@e@nVw?S*BcSGAJ+UzL^U;gcK_EAs^}o=XZv
zx4VV+j=h~NimHk(lVUq5%aSCKbnQB%QvX943=%P}jHa@}@mG}-y?2Zq(b~k??1pZq
zi}MzfCUk+sBoswSe=ugV*TrDy=3Pi5SEwNPz_2Lj<Q;+slV(&^$?kR!LSQf~NVA;L
zxL|x*a>#(x81ha^6)LO?cxUN98IyH$@+>7w6Go$g(mJ}mgw4DIK}p(rEvlbNk4REO
z<tm8L?al=*UA}}T@ZiA{9^N~lQ}wvHd6}XZQ3Xd~tHw?v_9TgLH?baiQI@2sA@3wn
z^_8HF;^=fl<s5k~sYy%>ejJS|#-ox>nz7MM+1%`*ji$5}&RMc7#V8{N?W&@ar6|M(
z)a!H!gc#XfF&vecM6tEiBTG_jRYb-@RFC)WpK$NN5gNti%bVQ1zQ@K!2j@IhRm4;x
z9_!%v^i1^C-3}-HF=xX<>>1XH3wbGozp@oJNLJn5jhw2isIVkyhS3Rj<QNwHnCDC3
ztuPKcC<rOzq9idQeWjT~ceV8C9Zn^XSvI;Gbh6Hx$a3O^6<QqabV~0wI<JXGO<3l2
zJ@ZJtR(dgY&9u5|(*L|vF`L6`UWGMfJYR}P+gB}UG<9eZmYG4WjTA5;Aoc$hfngJ!
z?kVt8DEee17=rRs<IN^K50OqKi6_yKfYf^(T5DLuaW7rdEJ0Xud?7NPmNKXIapiNi
z?0j2NTGU`hg|(>dltFxv<TcZucT2_+SW!?)R5fXqimRtKBqq@9bV*W!GEgT9bY29~
z(_x=Znu`?R0^@Q>k%?Qb_n@^Cn}wEC8={Jl*o4gFj7kacvz4T-*y!exg|!;uoTJmp
zs0KqqfTEQ2AX-6HO0vKwM<=qVB#HT~pp#|z5GX2#4TX3c>4MT(9v_}?baYBlI<~ep
z35cP$C>@DXY;E?)HzXfcnj{RyCFSXWZnsOflTumBXk3!?H9NhE#C9azywVtzQdE60
z04BM(%9bOFQNg{tk9hE;&z(C@c=F(wS1vur)mL}v_xl`=2OJ*`NwvY4F~jkg)3Y;B
z;+f_V-gk2|r{5oPd~%3$fleo<C@ZooW1J=wrPLyUz{%N&@zAlk+2OexdpJ83dx95-
zgifA=Qk<R@qVn~C^B9wYb0lekSB|99MRg=ur49;8=q6ieWhi|`p2Evlx467{4Xrd;
zrb#kI&vcp;F?IbAr8QgmKq>;}A|>irMNtwTO6r;{mB{wGQ~(6=YEufM4T(u(4H9_?
zBxWfzsadRZJUKkz$>9;`gyYi_jt`Cr$~S!*V<@XJl`WUPDDC$`WDPrWW>)HG^VD;1
zG}A9}n-Hkx+Par%qS5}FTsvtqKW1AhYWKLF)G<%&&gH5X5*0T&5yUxng3E*qmssv#
zIi!O)ODHl7j+{*tsEYP}NuViBkg*To0;HKkQ4!rm=jSv#|N3{E>t94f!h*e6)X(CL
zIH&E*b>|Tf-;&^F+t$t?3+Z|5d24yJP2ca<_h%&_Ap{PO`Vuj2DpZi5!T#VF>nwSm
zk(z|ja3qh0()i%%bvl$~iAoe%Vi*--cNh-F0A#5|x0jXVrRqfloummah%q-&nqjCo
zIUZ4#C6$%JLRp?+E2(_f?HD@UoZvj&PA;Wpeat;omAK&8=yvFbLX6yL%5X5EtfZy{
zNcxgawneaG#{G&UHT1GgwtMmQ3>vHkXBm%6wz^xGu3<PF;>WPv-C{HK6h(!vgh-+T
zD2aScY))qKNI)Gsc?Xqi4i1huxqrY$YS?<|3aRtleSDwtUB``=uJGLc^Wr|N72Dmw
z{>}mSADv0=EGNO8uf6y(pMC8O$*7BD(e=)g#UMQ!-K&jJBI-M0+L0(Z20^-7B~3c>
zmk5(wU`2(d4&qZW6^$?w)>e2saTw6rAZhyRXl^B0iEDS_m?YDAsZVZVb;ICmqqu7r
zHc6c$X=0ocVxyI&tV&!ZqO?~crGTj1MAaRd7?8Se=fsZn;oV2P_1;H-5|6OT(a93J
z9Z8K-ILWtmaynew6RWm?)GylfRW+hn!MRTnh((e~+f{ekelk15wv;kCw<_6T94{<5
zKR4C$xYU`pVfv>mDKJT$OC%r;N1(zp#51I##3O}~I}GSZY#4?>LZyPlaW#Dh5@k`^
z<AP-R_CRp3(H(*vPZ0hTfoR>EDS}Z?d<pWNlcL%STm=gxkSRK}N~)Xfe|D}*=DUy>
zENR<r7MKv9`h<VRNc4ar4F+Qql$TW3`DO>@D^b%_3LOl+WCItxM1`xsX3{}tJ@Ga&
zI+h4k>~(ffO44{}t?{u6U!H5SJd+geS|Y<$2;`kCPF~WU6fvT>No|3PrFEtLd6wo9
z_faDdkSO*<k}c}Q#0q*y+IJv$*VG45O1!|dR*h~o%W~mpgU41SN{byqLOJTR<x=My
zZMb{)0e1$E(H?fU_SoOO2}leAOwyO%xOAQU%^g1Y=oY7g5o0^%$QN{7hwIlrODK0^
znhoKyU8T8w=TVIOhAd5Z{l(|Ge(6dK&W!11f*=K{Y*Bl;DFRLqDiXb|Gc813Yf>Yr
z>?>O}<!Xuh?0rD_AbMx-B~BxHfQ9HHYi-52v>Y59k>@Fw_jd_iQD?tZrY@3IQCV^G
ztt#<g(n?X<O5$9U#2xvVBg<A5&P{S+jbm^nK`JU%J`#{BI(LK+>y}8Nu&OHY77A-Q
z84oC|123^eK0q<Dob?NdmGK^}1C?Fy&Mkb~0+XP)Y1LAFY*t}3FFh>s!cq&{vGBU-
zi1hX-lm8<T3Corh8YyJ1#WG@XE)kh&Qya0Bcw)-T9E1vsXMp3BiW3S-$+;Z!$ptDp
z#6%=0JcWY1cH<3-ZeFp~tvEa~B#};Jd0?yCkKRKy0a-M^ry&&QUAKgg%yRB+<h)42
znZ2g@-l90Fbv(@cp6ho7wC7UqY0E{r8LTm)FMfiS5)h*`pLy+NZd|$28b?xp=V{uc
z12;yK=eeW{caA)Z>Z4Yzj#S4=c)>-^SBe{f7~X+c-7m(YYDuv!Rf7ciduItj)KlW2
z8uKMFDOW1RX2fQYBr!FIr&Z#WSZBRXNiXx0x)`;qs3%EsK+?1LnBtuzjk%&EM#UpB
zC8nwKuEk~9D=@Kkffoc@-wYdU;|l%U&F6UT`b~;)$w#+ubL-w+(xC8TM>#5DayUnt
zB<yW;c=gI9bfO8)^0`-E;>DX+Iec;`0hk_qRpJ!{LW6*fHWKk&73iowk|dFyaZzHU
zmswqhy|PYRm4n1!jH^<lnndnrSy)EpSlqV}2MX~D^RXs6<arjIZUQFMQSDMSkPw<Q
z{SX=#;)c%-b%cOa3UhVVRYFXO5n?Qe5J?wRT`JVphgRPNypu<$qY^rq5K8AgWe8X#
zVWu6IvVSOC<(9lA^UzK%IFS_U@47jr<Iy_$dYVY6rcI3t+U@Hn3ntb3>LYrJOx2?)
zdG4BdO%TsQS!59sn?)cC+kPB~m_CRdNdg988;uK4;pnP9yNoE<;EXK>bV!vWC=0=~
zy7FvwD#~#OQo(h3*O8eb5)eJ%p67_#%(`Z!f-rM_!7`>#&PWPt?bot)VL|^3C(v>#
zYu~Fi2Cwyg_4gSesrl_}v?VQu6%vTYRh2x*T2hIaG?COHTB6OX(sMQ#lH~;+&qm(C
zJ4r(lLLkY+o9cMbXE+|QyR}J@rufp688L!NIaEN2+NCH83Ai%}d6t0>3`Ro|6B8}!
z6r-ff?6@dVNMyUUj?t*V`@q@YlswPa+ubEgBvB$LQj-vzBuTW+k!Vd(lmzSObUIRj
zG1!=9MTmzjB|nz)zCjE=dL=k(V}!FHt1?znR0@tBKH}k9M^t4c$Xf-RukdvwJ)#oS
z(G9m(DEy-V-~GcsrF2zvF&2b2t}Hzb-d33LkYD?iU*YWdnD2e(yC@~;V3d>6!d^|H
z^;Il_ZiDorf`T9^)Po8!&4eInxo$#Sfmr(~<`iq5DUmdNh-T#0gBKi(21OG)tY8vo
zs_G&@t|1yV1JJG(=OeKYo+rfE4hTUq0!aNUv4JR*q<m?6T6!UdD&9loeZ$BoBn3(u
z96D%JB9%0~Jbt8VdCd0g+cDv$Cs#~f$w_R8khM1Y_EQpuUif`hI+^y^s{2!+C7(1>
z#DsvfWt-L9_JTHxkkBl_X(PEX(<scIuO>*ktRQ&0Ko{5s@JS1IMGb-~@S&tsaxc3v
zyFuhK<fhr2v(qgUU_G1+HN#f}`G&-at<9Z>+4E09Jezg4sZi_t%Ig=+Z=vXOPG4cc
zJ!nm3uKBFgR7-2Emr()XS>r%FJb7?PhZLP?d4QFUN%I211&YdYeA1`16`kC0Wp9^G
zCuckw<3k|LGQ4$ERY@RlmUo(YB|(#A3EoEIn$mdd7*_?Qt<c8MNiztF;c!Tvr-CfK
zM+Yf5R9ZpEi8fS~r7A0oR-%4N6AqI5gdp)0S`l^j6P18Rkgf7Q8r>B>xcU1fI5Vo)
z<VmXo+lZWK@QQ4A57mpFU8iT9+<O2%pmah&LS^gg6iQM|t3XgvU)_~sPWye4B%-Q?
zASqf@tTj@VW8Qt|9p3oNr@?u~Wf|kDqPjxG6dx#*s@*0763wm%O5^Y``DF~6BuKh>
zBgDvuHpda-mPwR*Xb`7H$22*yHdiA~H;5ogl+=2Xm<ZFyn3mRK+gvYFQY3gvq|&-T
z*+ddi4Hv7FZc^)q;F_eJA;kJ5#4HKKHALgB+%r!tuhg|o)&q}XD3SjSvY+t&lg}FA
zPl!TYm!fXpmRQ=?&q^LTBKU@=)WoCptg2Z9X|u!i4PkO@3b`+8vjS;_CEuIhGn{ph
zx2>6c@F+JSzIv{03sfKv6DfX5GEjl-?f_!~gMlzY<4W`Jc!Rx-(|Mw>79nT`e#+ys
z2iB1y+U=H3WajXyb2C4}qH`^%pZB@-wVsP-w}AxuhbJ6F4=V%F_5fOkn7LJ6x()$7
zavY2f(1>(GU5yS&TC26D&sbgpRU?nkAQN3rJR%k!ECWQs&`RTM*%)D!m;6VPM@vz7
zc`^+O93&`0>8UEI{UAvh!39avQP(w*>VdI>UaWUIu^Y724UyATMAck9XW1zoBdoPj
z_=q%XuOY7j1Sbf$F8>?*)>;viq<}H?NWlnU3!s!#12o37h38~cFs>?Y-@VH({L(M*
z;;S$5;al%QtP>7`oa<B~Ix0te&Fy<3cY<W)pC@3njOQUD?3$2-#wJofXSyyW7*xz1
zIVmNp0A&(<kU&0SUm*zc*YSm}GdI_qW<AfQ|42ycK^3Cbbitz$iN^Yv6DxRvkE;oj
z6n%2NZ^dIx5{QecGZK|r<)zx3`*IKz?NJmngkkcu%pafZx1MTK|M}L{ErLNyDhV~I
zqUCbC1o6)ylUeCv>Vlc?D-NGaIO^}SoOhT^w;?c#knMub>Kmw3bFZ3?d|+?mj5l6B
z<1gOYmgif+TX*;P)b+=h7{fk4rxlFCGGx1mur`ZYb^e@lvkg}^Iqy6(`?}(sLN=?#
zbIcAUpMK?-E8{>^sMA}ACRGTn+QPz7({O!C$qS-|-3OR(x^eNfibvL<qJ2We+EXzY
zt3lMnlDVPbzUmUXb+M}uABZ}CRh=rtdNCZQ(K}0Q9<i`f{jzF<ifWst59&SD(>AY~
zy9<88sn(ov)DbE##gCLRXe~LjR5Wgb<bX0UZ3@jEMXw|8MD?S@O;$>N2MmhDDAH~h
z>)~`<a(Xi0qmMr1%U}Bn_ix?7#e{*i3aUZWChkv=v@<nYGD^n;k&?GdYJ*EV;{FjG
z1Omb#PO>=(K?!0>kJ}*kJ*ZgRDb#AycpO36H6E)Oi@Hk<5DASCPeCv~8t>$|ZkU1k
zxvE`}A-e8P;mK&FsGSR#j2Ys+t8<(Yq8;hHm$UtD)C7~d@A`LWn4$V}B5l-6Nt=MQ
zubs7z#Jy>VN0<<gHX>Yx*xRuWQ`fW+xMB*itV<+Tj$6Mg3mVi!WD!@pjIh<uX8U;k
zj8pN2m+$f3{T`1F6O4iT2OU0ovdgPi?n`tbt?Sn!AT2I@(fX#2&-OJ-JkL4yG_aC5
zEa`VCVQzh@_Im5*BmQ*H7>N$`i$3W*sJfQ%M98j<ni|y@YU_#B^(7=FTFfx3l&E!L
z-=QHKQi#db3D+V5GcmnM-EY!TeGq8sKj_4eL?dc&k`F4VSUgEhuGdP(dgzgmDcuks
z;aD49S_oE)<54ZnIEjXNt<B`6H=4?a##<;+f=wL(8`q_j<h2UH6P$|pINBW)joPNZ
zWjb!Z(T1&Vhe1(NmX1IFv+wchfAuQYZrtF(?K^0lVB_4B*3C;70A)<{P^yVpR39jj
zhT`3oGP|fo8iQK$tWjL;=``UQ8X_CN1n-zmOc<K`B@Pr(lJh~b!7D99HGqu@?pl(l
z)k+SrN7ei9quxwKBWbM+3H5`nglvVzC@<V<owD7=>WZ}^1c8`6U|l16+aMk>StB1X
zc|Ms|N7ct?GwQ0P-=S7jspesCwv-k%oH@SidyUxJ$De_CEN$09Jaa@Tv^FBHMTPls
z7LDB;uiwRK=;eJr_53mSk1nGWj4Q=E_xE^d?;%FTV8W#ceFghHb$$U6UNzr^Hl(Ln
z>=Ibjd=-IOL||s0gDD0ho^#J?1F8E`@})*_DAS<S21&&6e01^j9*hwqagD_4$!TcR
zR6=lE&D09VrCP6|kNS62zd{qjqnqC}+=wLUQxP$mMx2mCwG>p7tC~ZtACcNsO*OSR
z>Y`XubyX80fdMJgTdTr^!zUVW1Hm~$3{b0)u8QjymExofYi-w&voaH9v`8f`5}L@d
zrA9LHmSy0{<3m3D@I&_Y_qc!iHqff<7!B#Q!py~_mnynW`yeVd6$!Z~zRxvc)HULT
zn#9yay6U?1K8X6SN#GGBO%;*5NPJ=jukD^%DPbajphdz^fD4j3-Vq%VqDm(OQB}Hl
z&w?jPrm`>2yQsFPiDQT^y%H-D8ns?bdEupEqjfH-=cKlcWIIr?NY{Fxn&+h@-L#0x
zgaxV3w^dope4kfeDQKRVnpu(|S_rSf1+zq<1xtxeyUiNk&utf$x0yqFfhNb)I>>kG
zcMEyfYxAB@-F(as@9c7LCdT-OXB&(Q!$wxhJJBG?b6A9B=hwmgO9}Gqp0C>Eb6_cP
zWuf3S)nIPEOZK_#0O7PqAxrFotLuRT-^4Hg)MaPeGf`-?=r}}`5$h}Hn&l)0m&M;g
zTy&FJ)x_$2lb5U|_}7G`Hl#+^KZ9y+be(6(2S7)JR(sZ<Btatz<yz{GXve6f3m=3^
zh%^(65Fx)sB^2A#o=r#!X@aap=~#8o1%eTequ?YRy+~$Cxciz2gCKdL!8oojC>4u%
z)#@8VgFx(^({m<Y)Tm?I#>BR5+qR8&V%xTDV=}RAPOM)v!5hEf<mCHq&gJ<BPSvjJ
zuIj$pyQ+Kbwby!{-aiEVcGJ!6ni(hSGQYZ7+ayZ4#AbHlV?icy^onD{DwyG;qu1M#
z=G1@sHet&E7Lz)o(@A2z|7{5tCy^Kl^V)l~@9_DUjWsjjt}chPBnFQkx-w}+X@hUH
z{tlM~7{VnqOuvJ%akvoHMXD+={5a$~7Z6UHF2_4g5$|0hIi+zP?0?E{1o3Y)2x7wb
zT{~b`SM<V%=qie43*ZE8fb0`ZXWdtP-^Pjx%&YwxQqC~ueTYX*{1n2aESM<BVlU#y
zo;KdzuY65&j}tanBoEw#g_~C{JOY9y%hH_MBbOO^lUN{2I%H-n2rAD6MNTRvwB-nq
z5A~Nehdu1s6Y$AxYv+dfI+(M2c$kj1_s$0Yt8nlfBFXpR=GpW8k#NKHgC~?|Ex{L&
zKxER_JY{BWvCBTToq+EPO;znlmGFp$o=B=8XrqA`3rJLBtBaX0&He)hOd8^V+<mk|
z5>jk}Z{l|eI{Xxel2KWvQkNb+Lp!(d=u14mP%o;orsaHDev*2`Fat}<v0ArmI~oBg
zykT|StCV<Yhn*&Khl4Aedei7=YlE#+CZ#QO1#15DyW!quW}vuX7GfIGx|{LbC80f*
zisxJddM7W|q~-}Y0vk{C!*IiPSK{K;=YPgrFLY#vL5>1+g4|-}-0NC72gAV#Vh{@G
zM-A=cGD%#-R0PCzIxx5i`|>54)E-H38fsc}0~UsAo_vjt*c-Iyk9y2w&Gl)1-+cwA
z0eb>$KIhvql@*g0&ofi0`j0}ZP|;HzalFTGR3_fNpJmc;h{+#UK&erR_Wo7WMMo2b
zcSmwy#4*bc+cr2C^w7c_6<Qx>U^+0UU1?DjoLR9QC9U(U8Ln;8-|S)V=+lnW0sU$7
zNkvX=hF@P)w`#o5hh*!MQ*uwG$>%320mS{66;LINNv`!n`0DHi-7V|aslI`>mKOj3
zdK4DgwKWAp!olFsg~5Wyr=^e*qo#NQXU`c-O3zUwwJgnfgtktN`Jm0E#Gr(elO74y
z^@SNUeJ;G{^KJUO#lgUk1-fSf?Na^NNfPez3*YZ@1N!A?A*&w<Xt4nM&`C&cNJ!3t
z=I=ZbX&xlTyts&dJ&l8o@|m14;prLEeG|xPxBAKVF6BEehf}7?BP`M&WEO0%|1)xp
zJze_ACe*cE`<$UJzqJcC7D9bhd}d@G(X5RAMg}g}j__SK`ll+v1P=Ckf5}$gz4po7
zBUDlYg^awEXw+q}{A0LXsyd%yX4pF4EQ^o<3iZc6bCaF7UoYZQtVi^d169qLhGnjD
z$ZCI(lymjp+_otsH3S>~?trAY=_bMm9tLN4db1$xFI_=-`D9*0UDD_anO<SDycZ<#
zK4Vthz)a~Vrb0vA=neGhqqh4`wQlGr-lk*oZm*>$k0}AvaJDQwJOU&5v7X0>^cQYh
z(GJAzKisLQDiuc%yfH7M6p06H+8eW{!tmwm!Up1IkLrUL))<3+>rybgrxo%W(!;?M
zUeyl*?$KDDneAzjz;YxT;^gqU4G7Vq@}PN8#=THHz-aSS7Xqe_EHi#3R<5Gz2moqH
zG<c74Oei|3U!9foIEuWQQ9J8ju06=RS@ROkfX3p)*i8MU^kM^l@Wxq<qG1A2!OT0c
z9~0TUx~{x$`YLt3mb^^5dnXG<$w3Mx8p>TlZVNXQj!7w_iYN7lOdcY*dYlr|z`p%n
zk*^Y&yj&h}D_C9llG#m1T=N*cA~aH5+G6}i_IKr}%j2Z_R#W^aQNEjI{vxYCX>?@_
zxDi}}C!xvLBpQJV%P}a7CLi}qcQsL;s6zLzKCM#%gF|bJ73(CPa{@jJ!otUx8Z?~_
zChlEDje^~Pt78G;R*Jpoc3y}f=4w^GMcpKwLea`)jRDK;xXONct$nm2CMrr`yk}cA
zNdb&x`a)QQm>p2uQ8mFMg!Uv%Y6K8igbjZwI&8{2YKcEFA%l!&YxVI*ZnBd>_@O!c
zk%75$V*Dxkobgc9+%tUR+;C?>!n15n{?DjF&5N*fw5C#Tw!HBm99f`$fY1`Q_HeKE
z3zpFO+n|IOIWlDh_i05^W(XyTTu5XEP=U4eq5_zqxRI{{`tO{f=jMI6g=>SEZRQ40
z^6hWHB@8J)3k<k+obLGCW=MR!YJl~>Zm6Qg?iJZTjVKW%*jPmrwKK4M)%Bf2kVu*p
zVKvQIT?vj{tc!4jIs;f$415MS8N3^6F1$YO5xe>>BUfPUd164q8u?3T#hbx7Hy+{6
zEB%!r-HWshLx~akpbMc+2m6w@A~(@?>_OPW5A||clYZwdKX}Dd(sXN<^bZ|`6V&d-
z%^dNWT6x1_)R+3Dnot~OlaAW?ARZ4(3V&cZ_U1{fP3shaP21qq+N)puv(pS?({ad~
zzib+5%zM_^IdqV5?zw?YBLj8sp?+KM(t;jhk{*+!Je-qhNrhnlWQg_n^vMSA8u;+2
zKxzZZ0J{Bjjh=dlWP@1k*xyu(YG<BWz4I7({r4Nr5WK#(UHK2nhDu=3{n&;oGtSPR
z{VG}n=l#H}G5^DymRMs4{wvI5lOIZ}r<!?zl#ZM%jmQrkDJjoBu2NyvVOW8ENN!>?
z!Bz@Rx4npW6#ayIK7$eeXt9_7oX3p?GpYbrGxwUTl0VcHY_!x%+y0Szt_!JGZktUq
zh%QZVYA&=O2AWKC`GS{6vBf^717DRvF9g)vI@%pb`g9YBIwbc|sLt$Ov0gM5{jIPw
z1i|%*T4^_u`~gaM1C*bJbQ^X&&6YpyrXM|RukngtOX<9S0(Y6V`x>hX0cT}8(*#>x
zCFy{`5)KG(v|g8w;aXmn7Jl@2<?*h~P%UT{>N{%Yk``Mj6~eC@X_(Wt32wEC10~UU
z-=v&VpU925I_U%hRy0K|`m3><ZD4SBz2J(MFuIqPz(5+Q@Q9O>1%kLsY>dGq{R!2r
z;_GN+S-t34U{xRE3@d#(Wenn>+4Tf1Z2`ps2YVXZDmA6^<MOZl*A1GC^yDnLGST+}
zq)&b|Fxd_HQxvhB_<{l>&Op?pv!ENRz}t5d3pOE%l4^_Y^mAIWh9OF3J>I*xNxC@c
z8Z=xnRu^a{T0ul&yyj?EPyg9(kcs!rZv!Gh|H-K-H`#_mXmBV0Rwd8$w+g*<9sYWq
zgASqrRhAIk_OuQ?bGh<bJQ|Dw8hySW1b7CQq2FHMzn&pMqbDv9EPu-7>c1IXW~Ede
z?)4de_2+b%0$lp3t+FbH5-qdWcPB*OVvXShnZckGX1z3CnS(vwxB7xW-G$--TjCO#
zYHTv7P&)J^s}|JGqn-S*>NE<)i;oqkl-*~=Iv6JSY304YhFrQTtCEr)cC-s$`N&#3
z8@W7a*yE99`!^A1WW2eS>z0>(8FQTm<NYSWSI(zQ?Xs{x#;k}*P+K-j5e$vd#j1q&
zWtIO`xku&{>&;SZyXC=4e@Uh55U~{rt?v)zi=EvLG2mCoP3y!u?jSjmqGI8E`<`M+
zT{g)cuWfs0@Yk9k{lS4=KT1%8?PidVRO-e*?{BR$*d&eVBXb{{`*;=MuY(W%C<#?<
z@&0cEpaq?^RZhwHL#&jO+h%s2j?4Wq@GH2XO6FtO<b`ABb;@8T@ETxvl7{=YIq<&_
z%9Ml=Kg;266=IG;C};X9aeYoyc#P63p)UP1@2@?faT=PPE}yXLYbO*El2%%UdU@U7
z4jM7-d&+_R(Hadip0q`oVJ|3uo0V6()h0SvDmWWZXqix>(~KE3Z^vVRo{-fD_CtQ>
z7go1himjy~j}9b8(Q-)&4*{FkzYs~_MJ0pvCW5<xzL2s)XUh&eAyQd%&zTpjPY=^f
z1&GK7n+|0&vjWHox>)bky}G`RGP4C<x_q^fA7wH{$0th^!kN3)NnunoW3ALn;0ZS^
z8P8r5R&!kPE51!X-&*eL*IDlY0}BFPW~{L$`~*r+b2Lk$SbK~aInAa->-39-;I<V6
zH9IWH-`Eif#HEFC%Y;Xg&o8CQ?LY9i%c|j^A$?kjH)mW$xplq)J}oD@8EAm)gfnK%
zG9Ia4ak_Nb6vtdMt0kOIAqOo2d&B3rn4ABHeanWbV?0&M`SklK0*Ve@Eh`@dWP&KQ
zY#K-1=6;WvW5460u!mLzGKtCF@gz$p;t|FCh^fq?n^@?=Fz+UROd33i#)|vUyVs2=
zjKQpN&%_dH(&l#0Jg&$cKFYUmY3i=Ui}0rtD=|U)gM(y>7z43V^gsMna@Y)MGYLjH
zPC`5I_@FPxMKm?myt%H=8z*PsC#;2C-%MCoR9>NiaJKt+vGfhYZ7><MkALLR2k9V=
zjiWhLJMoEZqb#!ec9aq`v7HVJrAZR9xjuhF&*SJr6PO5-o`*rqe?7E6!+YwQqxV`P
ztSGRjs`TF;<C&$|YfeS%{zijLYnZiznmBAK$l7i=`T~h{%aN=r3Yjsl%w8_RtS;VA
zFpV!3V8G;z(j^)j{7$&5P}rfhkedn|@5b+Nf_(RHt9}8)6vVodE@Z~(X0DwQy=GQ2
zRo*c-Bn;z~R=l@8Y158Qy8i*}PO`ULjsIM^k_$b_8Q7TQ`PvEv)6PBB>HPKuG);xA
z=kkWuNWXgCgl1ugJk_YM@d=$Y@a&6^zDb!~dg$_<C8MR8HI&sm*c<T7{>mT<$VD#W
z(gUMwAeUidWJgLS#Dv%u^q9o(^F8J!$i{iV$5xdgmkOsA*7Z>-GDU_rmTBOl*}5TY
z5ETublA9pHgp()quYJ1}kd<VEbgb8o+gD@FHEq{+N-jWI4mhSdJvKVp3X%VDFLm%a
z@{R<<N<xS;@9z7BMM6?`szCsQkzMsLL}2_ERh4$SDs~A<BS+m9DUiVoczUqG3h}T`
z5A`U8nLv7sXW#+4G3xv^Vc6%JAvp8RUwc;fU3{@8FBgcZ6vI981zU33W3t_9<7~*V
z7kv15$DO|{SLN{f{M|)ch$s2!5~Iejv`G=)_SffdUENfl(xX!5LH5DC<xwbrttVIh
zjN7Q8%wqsCWUTwxAtylJ8hP<C^MVApzHQbX@Kl8h+O6|5>RtbDr>-*_T!J|KfWSEs
zr&rLQa8%s$+Q=^M5pvoh<RJ1AtR0>#Rd@jqQ1#`VkDABK-*l6)e<zC2nCF6SH<B2h
zI3kmTcIm+@H!SFQo9RB?(=dK#VhNp;nDpE03p_AprD|$rs|nL`+yO3UYBIf272l_F
zQew4xAI30^4pY(4X$zxFT(ZNPL)Y{A>DP0grWu>C^^_F4A|*nF`J)fP207kw(pz>!
zbDinr9TgO{<wtiUPg!S3(p*7V8u+}+Ms2S{ty=h823XaWWB-yR78uj%W~>}(0=Kb3
z>L8MV00#Yvqsiv!X399h#6`Wh!B}9XMO78}l@<p{|JNaja0M&E^#(gcGtZT*!ekJR
z;NSSWgS3}82J`TK`w5}03+4IB*)xyl=jY#npZrNALidC^Iq{02n=EV++?Cyfw(v@d
zV(V$r3?iMviQ&?ij}m2}2aD@D#{0Hra{^1>yXjNex~F^<@x_eX9JGOn?jMss0jQaQ
zQ43(vGd_*oMMe$cPP`zzqUPC0jnfssgi)QUD#I+Zr{Gxx3gsp}NX=c~pnM_AL1Ev#
zR}x4!+(%R3WpX{=r0*Ybp|;mBBGJ%~YlVfDiUsGM^<Y{3z?#0pl{FGk&-L3BW!y}=
zldiw3+cd6|?yvp6-C%7s@uQ#tlP-NkGUbE5?*(8JmV#s3&sY%7ix?EN;LKT$kAy7j
z3q-c&c%07`%QU;y4`-X^K)=?=Cp)m;7boW7!g1lh-+e|Ml#5<ohhbRuUODf6+KDmP
z^o7@gKE)~!2OOhZm2TnTMX>GlAcb<HjnhLpzjKGjVee^Hk$c;|rr!x;N5Z#6*VYJ7
ztBrPC=$L!{UZ&?o2%mH;`)rh<&z11sW3Bt}PrAc(bK5$>-`*H!R6hLATmBHq?=?6?
zo4ll$vzYP##1nP!v|5QLX1G|%IZf+%J9!UG8x~Ky*3Hja1>CO3BL1PfNd7yfcX)e{
ze7BDCuG91#W#$tVa2}URxJ%INL1FLuXg$`;XIntRFlVoXvM#=iMA=jx?NL9<q&hSZ
zF7Q(t*?*B?c@WqC#kt$K2R%2|Ry`sBG?E{4Dc(4f8#_}oRYZEgmaLxbUWKJ8X3n@X
zuGzmHC2>vPZdPEKcipG&PCer`R5;?p2wpyJ;J=>dTpl<<Y2Gt{gOXcxU;vp6EZKY=
z14-JNZ?mZa>JNUCHD6L##ctR<x7deLSX=EvlLaMp*BkqE;;pV|;d%CF7(RVmTBMvv
z#rJYQz34=4;hp@?KB2<R{loLdVQanUDYz+A?1OrhSlq%i_<~kBZo_slmI@}g_zk_>
zaA0i!paTO+0W(^5^6%v>EgAT^1;hy1E8(#cJKKdB5_2>`?7|)H5j5qdiSiG~H9QGS
z+!DidzCKG}=#+|G5eyZQsCVU3%OEEiUI(EXxs7&oK#VfXQrt$2s1+S54OYiVH$m!I
zF3~=0m0_9CA3SK1UTni_ifLj5#|T(uVsnk&8dk(D1m)LuDzgUKXEuoD<s}VeE)w-U
zL2L-Dkd=%0R{C!}HNr-t(aFnyqZ75GQG~tEM;xv#fl_5gv?8X)BXRp-WMtn3+~Jma
zppIjo6O2LliTHcPGQJhCnO7{Nmb;twOS~BErF2Jyc6{%wnT{<PQ%httDC<S&NMY3r
z5(B!tKp^|)e=?-io|pP&j7F?~e}mLN17>;ca38abYJy&^i%51(U3ytX>yPZYPgagW
zn*bS30{>p?3giIAf$gUGy9~C}`KKvRf8Z8@C$zO_Q<XLX8oY?(skmS^#A2%UhR_~#
zZGX$n>vv~LzZo*rb0l<6Sg8-b{~LxKd+pyv{4=7myG0;1FCu!s_jrV}(}psZdEb9E
zq5CctwQRL)1~sZE8M$)3dUDW&m66oB+aR8+Yx3gw7+Udz3E@a`O8EHKU-^h6GBt6e
zZWFxl9rBR+P*!rVc6ttC`9G);)$}z4iiT><h^ASR{fdzYDRa1Sz_(0?y4FwG<|cT(
z$JZl^%!oG)Q!Zw+04<3Jn>e!(^xnbfEBo({qeG)ss%FGw3yN(bQ4{)Sm&HaNcMC&a
z!i_#JlK?wudV8I~eRNdm&Q8IuprSkP*?<3p8#@_s0!NzLS}8-bB{y!HBiF73tH~PG
z=&0rb$>5OxD6k8=oysJz<hJBW7IRKB>*uVmzDoXgCNJQk`5ik}uL)(Q{$)LH)Xe#}
z2j+qco|J$)-ee^GtMQ6|e2ZzzN<E<1WBw@*&s*O(+=%a#C23$&TG>D<dQgZ|)OVrb
zw+rcHeSMUtg$n-$;sJeALbe#ILj)g|n3)`;{Fxp*%{u#w_kv`eZf%tTGhL5d-PNKs
z&tCQNn3UX+!;hgJZ7vqq!n4`4hvooBMGEvIq0%T9+hXqSmOh^MGzC_P@H+=FKK6RF
zK|E@SN7-e0*ahN+n`18Vou_%!<fXtS;H0iU0;0YFe=r8{e$x^WCe21xmn<pn$W3lg
zKCg0QL1G|C=GYMpzws_^LL16SlQpOt%^#nnT77QkPNo;>MiPxU6b14)jt=;JMS<4l
ztJhE}QUt@mos*pI(FiNfN0g%$Iv7>T{n!fg(&1+7(vW?gPDX?L@7*O7#u_qgy_$L=
zbDbQXBDZZby|sRGf7`;rA@2UaNs$rW90|&Pb+@^Sr`Y>-_O)KmYXNfPWXT3fXHqa@
z*<0{@v%Ec5adGGyH|{T8@{mzP2d($>cSWv5EhWpEzt~P?4*~&)?};MMX9STgn#%m3
zzZUhofseaYsFJB~+Kty1Tf3JU)O!skZ|_TWfJd+qsloClfBO9T)uQ#|-m1*`A*>J(
zRR5wA&mx340&M}C-m>s{$voaurqZBg*p<67``DkKQig|de><ZQ-?!gISsSpcs;ZQV
zOAvGv(01FJ;gp-#%81-HhVjCyGVbCx+QZK9JE3rQz{u$BMG-Piss~leuo#B=_eRAy
zm#a|x++LyxIl4NjQP<k#B%2$b0kC;N7WNu@M5h}xp_pQ+9y8mQeKKjNhH2~hRK^rB
z%lz|PawM0$Ulgihlf*bCPz#S7Yt|e;k2Vxtm&^0j%otU}-+l9%#Y;GvB4pv7J>lkI
zV36=ET#u?9;<zj2yh3Hm<VrBpN}PGDLRppCQID{0LsAH8YmW{`Ajb_19Fu-E!`ZC-
zc4<O+pnHRlTR>Zf(u>VHG0-r#S*G#m)5}iskasv5*Q`<cv#(N+x9cY)b*vK(F$`B#
zyvU}(wLu+shMaaDDi#I*2%!M-ci)v3y?~5<$;lT$H9Z<)PLprCW<TgZ-sRW6*`Da?
z+4wp1qky6SXjHimo!ajw|EuukPMGY%;Vbv8{weON@-|2wWQhCPtlxMkDBlDmJolW<
z{V`SUt^S?ff`d~NU+)fZYn^4qL*nHZL`sOp@V{jg_iq(BF`>Bd>~{KPB6wTF4Ch>*
zYa5j$dW>CG>=F*;tCqWwKs?Dc+n#b!F_(8zYNN*q;KGoI_8Rgd_mPr`I(H)#IS{u8
zKwr<ts1d5|HQxLXjWMBdokquir%~G=C(VF}<NS;lwSovsvZlojUgNMn0a2prXW^`!
z&Zx7s3b1;gur7DEcAx8^<v4+9ISBE3uOxbXHI#K$iq~S?v60iWn&R#jm*L=WcR5Q&
zCu6Zey#D7n?3l^6xA}tsL0Lt4n;6?pbB#KYVxDcFCEHnj&`(Me?HG0xh`8))Jf@p9
z?!*U<1OFMv|I|9MAtf8ljB+Psu+Ek_Aqq(3BJj`8Q($p)lzDRp5i6if<pKkW@*my}
zuPu<X9if6VFdguOE*HE+-lwbL{p}P5HoS6z^nW^i`GQrv|NgdgcK&?p812GdUH|?x
zA+(lwy&L%`!^3fENHcvI=PFzav<s5h73m2V&d+(+QUtfN82=HRd=ZOzb{$pd^T{+z
zH{&)~l#LF_8Q;CiwTuG~H?Imvtmvxpa#%@>i_LkCo9h_042SCZJrP5Vmi%1!kM9*%
z3l?U-+iQ8p>s5DoZ&9W})1q=oOdr4|f<>a5KB`pV8SSrQi%FBdQ7hW(5scd?p)sQD
zk2dQgyx)Xktb+DqtHbL{s)nkvS~3Oycuq9biwkGq#;gD}S}~_$ow*#`PN+ce7-=7J
zktem~9&~rHbIcsGEc|N`Wgr||Z~=~y&tCOzMe5+$LT@{_tjwa*OWJUA_r%kE+>={I
zbCoSLxis{m)otkK>S3V*K~mUHm4_1^XT54i6UP043Afz@EED;!Su8bJpn<Fl?)-M&
z5{qef_qo0_0yD#G`&sTOi4a+fx-$RSJD%-AfZujsWC}|_W6xLjLc7080mHTYxK+%O
zSJNfy=EK&Wjto>%x*RBBV)^M%g%9;dXQU&VV8`<Bf<PT7Lalql1&a@(Vm%*djFL*8
zG*^W@aLID`kJck`ESkfJO1U9ny#Efu?bTA`7k}*SVwa{XxCT8fbCGM*R{s<wtaKhh
zi?CQyT6Li9ghrPdF%`0~ilO6V!%-qDG??xtD9Z*6&q7D{0Aug{0C?B<R!rQZZctkB
zrm2`k$J4f$!R(2C@y<G8+pZzlk+wf8M1Lor3keG9pVnolAOnr`3X-l+bK)+Wb!2v!
z7D1eqp((c?PGH&z<dl-|=)m(XpDYm4%!92sQkv#_M|u}qde=`#RvObui009J1B^nm
zE2%o4mzG~cK7%UrQy1a0w$jH;t^R|XkMZ61AcPnKt4kCy$p0`-6>i&L>^5w)xpn|9
zqB<2pMD~5n!=_XKW|o|$?8e+kfbnAnW$Ya47E~}g>oU-(CDRVjk*QSN9y!It*2{eB
ziT8@&VVR+By*iWKlTmTTU$lEq*9L?}z2NJ-Y12?~;mZ^=KiT@TL6T4c>h@-yMbnWm
z64w^xmdDw|6ebVY7dXzcCs$Y~w-Hnmn&4iM24j=wa)T@E@l7PIs$AdOrkP>4e!!pC
zk9P)XMpssk%TU`RNW!nUT+o*g(Oa)3q}&v)jRP`h$7-$n+u7i`Ei<^ltJ6{Z+Iv>D
zca4v6b1m|lVc7?wSz>bP=K8Swu6Y_`0eq;T;~bBSS1QY++~=7!v`Yt`6q^jTIQtnW
z@0ro}1lYAxzNQ6AmTQAIVbtqgcu@%gimQb`n#J1KFiS_;z~CLQ_UOc-0z^N0-RR?s
zVuwSA&(E(b^VlRO991E31}1zsibnF`B}j(^#*exz73<__lc?i6JOAwO9(%g~)i{e&
zjd|#?i}csSfWjf&INk!0XUj7!8FnBMB$P>)rF86B33I~$`Mu2UOQP-Y$PfMMjM@q0
z{M{Q~Yu6yMg2@6l^9(8`xY^GB^9H?dsciZOmaV&qfn8X1_Z97Srylnmw7^>yTh-Ea
zd4^acgLwi6G&|4s?1=5G_l2byA2-(Ogmi)KCRxczy?S+dI(2rJ{0(JiJL+F=B<3>S
zYC@lI?QFQ2I#0)rkJ};RN_iiB21$9Fj_Zut^t&*NTu)L&>V<koA4k=pWNJ}J9b5Yv
z4A;Pl;nzwb&YSK|HmBfdDz{<qP#zOgYU46hyfkQJ=~rpK4}I`jIZ-^SQ(Cc}t|XaZ
zP54VfSzQh}(FqvqT<L5cc$%@c-0IEW$!LpEH;wr==`AyXEb(M(3}GHzY&4>6dty*^
zIQgFq&{Winj+eJcg2$TRSc;H;0*S{V?Pe{o$9Z}Y1I73SVWBFh%Kw5cF#(I!5P}$e
zd*+65q%4G?aAAU_T)ez_in>^Q$9LIxu7l6santKNeP6-HQ#nmMjTP;ElH$g6MZIk?
zBz81G-g?^YXc_UD_K++r$}<jRcYcTj0W21RJ6r+)ykRLt-2k$s&%FvI#eSVu6_on+
zim0LI&J(%8G7lpYV#3R7;|=imj_YJD0N{NeWZ>Jl&q{q~VYOp)MsTv>QwN=0pJ2Z6
zdqYvP=s(O;6+>5GXu#G$h2Mligq0Eq9!!$#&umr&&^zm}xOkFXbTYogF#IWr7lvP6
z2P#bhE<VlFosC*cHg`~Bp_!4t_i-Lx?-aJCbII{QY`Tg3r30b@i>uJih8)oCF%{bN
z%-2=lkV3ZhTJ4RC(a1|$KvI5H%N)=HQMMuGA^NHm?Wy};0)CLvpHyu-;;p^Ja=<+7
z1wK&y5>%~2o=iO|GMfHK?Rabtorcdk-P0DgPqiqk+l8`cu7>doK#X&2B%5<oc4l-P
zDo~fixsccUzL(H7g=<B1Rfob0KSer1^ot}007gdmRuim9SdUSUO^Qnq{*r5|jxZeV
zHU3b&^kTy4m-6!Z-IZdc_oH>hTk!O6hEVHl@~+?HouxSTtby80(jdVl2x#F@7%v~W
z(7!E-#kKLGNo$sv3iaK7XABvl&F8WoueDHjV!=rzclw3l%D7u=&^UsW*$Di&5!m;0
zgcRjG@fD5gKNb>g_e^5ORC!<S>qMpG2Rvo&dJZ5@yw5i|jdVL4f#$cwhEcW0`}~S-
zx+LA7?p-*7%_kga>{!iZCbw+I|Bk^(S)B4k%^N|IQ;I%r_4Bad3j*5!N+<N4R&{Kk
zQCh2OtkPqJ3o1!Am~<OF-VH}``k@iJ+&1jkEqzYm>nPZv<U&Dj)tclsdgdE?r^Tgu
zay5*{uyyL$2LL|iO3b^Ho0In+>4@}8G5hCWj1Hs$GL!9QX9+MYc;-4ef}b`42A};T
z2bzD>SZx{f-@9;`#2C~90j0>{=#qT)XIb_E%#u<2B<q|#xb)ZHl)0t7GYqSZ&ox^X
zez$(1eqfx)BT)M!HlO?nD5zxP7sV3I1!|^FWyqb&-9b3(zRC9+dhMyy^n$lQ=`wU%
zY8y_CqpC1i2sI)>elvG&t00%CS=#ZkP%*cNsHDW4OsfH~xz@}jKyLIx;1m`X!YC(0
zk_?+#ka_N3=y)uwP?&Aj1QMn4nR6a+2%-HRtG7nXO#D&(;N%mBKdOztjD(Q9h&A@S
z8y+`Wp8s8nlCfVE?t2RBWWNzu@ApyLlgw4=_xEp0U-0b_$9==}cC<{Zj5+AWS$7`v
zr&77Mb*8A%<udP1A&yq9d5pnRvd`xg9CDt&yDM8I(G;Cor9HH@g`4N4H4<&)bLI(f
z4M+`fDV+$-@yz>CmJiqlxR0)5uH$yS3;#Z^X@c*lCuL{1bjCo2RCIl_;kNT(R$|G>
z?@Gdzy8H(V)?uo`E*tK*GO^*245QfGbq%;u2iEt>*Puwpxe9M(dw<W;l+Cu^fqPA}
z>=+PcwJZJM*bS48yt{S}bU}?OHtHI{SVSQPlM%zLgWMb2tXYTJU=6xEKE)=3&rL!-
zc*1O-Bc(!Xg@al|xrEmqt~8Wa<(|a^hi4xM|BcC`=ZUe|;C(IFE|tD<N?E>Lm_-F5
zA^CP-;v*x9e(PBy3JyRVE&{E+B&o_KO~Na-;|yoal`fB~J0Bg)-7{}eh{W*O<t-$G
zspBc^7mejzC{Vc;soXUK&gVAkbSr-g4Sxe+ZqM1b04JH*?hAeEyDkNrb`?g1*?tcJ
z8pI4?es{hWhg5Az&HT=JZ)^Br85X;Dt*Sbz8La&XudeVG+8xQtyGHt{O_?N)ECr;p
z=e>{hVaVM-8`;EYO<D2YqHSzq#LN-|965b{AH?F%;y^vX9l4rmkT;Rr4n%d{gF7J<
z^{t1cAUHtAu1Dx{wNSeWtY$CafR`aD|D_HUK3$9ruCQQcC|RQ)_4Cb{4gZ{l>_HaZ
z4BVY45my%_UMb~qLkyfONB|ENw#h>HT|=2Y3^j0C3@-;wHEM?D^ht&h912&fq#4G7
zdkkW)mYO``m;V8R`SW<nNXYFa-#(qq%68*$WVqV3#gz7utI}4(IRC~kSgd!n6%N8d
z1P_sf4R5|}9M095)V|=E=PKd3o{zrH&Eru}s`U7~m4dINx3=0LLA9`5f{t70HtD)$
z`(tU8Cec%<5uShwM`LF8KZyT8kapzeK2yC8!?6|X+0Er1^>()I&P?9r8zMJ=fcScY
zepIw|B8$0}J@(6B3GE~!Ne<nbHjSSbeyDg{XMXby?PWlc!+Zme&+ccrt^RiZjh%L9
zMu@hyp|X{oV!dxjc*TgI22+-9pPKK$6L8!bET;bpg_SKpHt%#qFKQ{b`}rAQAt+&s
zlNSRH=qKNbE4C#*k~asH)wx9T+|B-0HUt1Hgkw%Vo8@FtOZ$*bpsE5?3#dnR7{~NS
zMMu_qW;o1b6veUUjAi{q((T<n_~7*w&j;8GnO-Pkmae(8ttAO~w$G&wa3904n8md=
z$jh2+JoOgs{Uc4U#YW*X(rK^Lh~5A=p~-Yg`tPx5Ao2JG*-FXeWi2=?EE-zOzFkq(
zz^mwZyxd*fyJO2R_?5$V65xT*lYep%luRg$_pMJ`eMYj89?xF%1#`R-3Fy`zT4|>$
zfITnGVZ@Y|47R>*B#ywb(CpAe&QKO~yi?y7UBs|-5e#>PehX<TU3T_LDao_<vvOjg
zQ9CWW;RtwIa79PB@We$4-ESO>raeof;B@o)8%1zEQPO4P?edzT2ZA#Ibi=T{a+pB2
zY__hjzW|xV^~4k+4KnoSCN#{@xoK<M>u-^rd45#EnpOVHoph}``?o0ji8ffqA+in+
zo<KbyPY7QATgFtA4f>b*_&x*caBXc(!~UGIWObNd?NYa%g3dHY{45@6RdVlvTS>Rv
zYlD!)&w1ZA)fEzK_J`U^A$nGjBB8yG-)XgYq$ky005?RV?1(5q6aH~dk#@t>R8oZl
zC!$;96oF&!d@wKJJUZp&dFkc*QoNS8ZE5k(!YO>@9B>D06yl#R4~Td3au&=h)B6c%
zpb~h#Lb!Qka;YASUTHFnoy_f@6XgjBl0;tgtA(2em#pV#R)wqqmjtk=XT9S`7HXw@
zEYB-k@MJ<@uxX|3@tb}w5yV7t?V{L#Ve5mfob7U3P+C<x*VAe$E??PU@>pqkqZ6Dj
z(w^Q>2Se*(H9q{1|9>`0d2CE~wGR&q4Sp*8tIVZ5K=~$;RB?F(_Lkg7zca5`NV?)j
z`7b*vdb+gGHNK4o$nyUCysbr8G$Yju_VGtU6C<O@p5;hpj_cL6KI^o~<%xr=8{CDe
z0Hygcyte-SCb0SLfA#o96_>tm&J4W$h=Rx?*X2$r;M`v?DjABcH-@%<0gOT`C<TPD
zsh4Orb)QPdLSqPejrdr<_kZFnaGF~}lh=?QZqdH_UF>s2(6&ry5-2;G3T~aHwhEJg
zH8iig1eSL^*YN(ZfHM^EC0=WMj=M73!4l>U&&HsV9J^q2z_=EZRl%TKJ=bR-PwF0)
z3GlNf+R^@IdfOVu85c}vOB1#phQy_so;rL-hPZ~U%;c&DCMe>m641P;xhQhafEe^+
zv3FUpW%(VOvN`VhqtXubuB|gN2Z#qvR;QY5*>Y&u`T=9g?9hA@8g*M$=KnThFBLIb
zv7(SWMioAmC}w(8e;N2nS0Mu7uL)FR>~<@BUiGP63P$D?;+L?bLWbB89o&{W1jkBg
z6*+<2OH6vX{42mD88`GkE=$0nxfY9>LcqNRdJ-VX3rI6_)^oN#a9n5Fe*YVW_Kg4|
z$A<L0{QAmj`9i2WzOISMZon7M8zks3J^s$L7K9RA|I_<vZ})AZlcNy$Ty^&LsQU~4
z2d5&Lz2F@`FbC?vqT0F{#OP>bHYHG(RocGMXJ+!HV@DCJdzKp^0=_8z#{ePS=3kEX
ztP`>udG4U86YE3t6Z%!ovshrtVa+WBrcs1M>~s0y_zM88-c)gXxsX`S5SqSwH}XJ4
z+&7(UQ$P?C@RBdD7N!=0F9Rm-KRie~_vW33woNX60fipN)(zjfsvn)>H1?`V<JMcC
zP<}V~7|?A9Z5#hwV`LBSmpH+9)(e}tZtVViDLBM=Vkk>e<60_2=)o*mGx4eVf!9;*
zM8*B@{QgREQ5$aP(`?crJSLnsomCm?&^UsxnbQtF*MQ-0yLnTFb7Txqo_)1E);A^b
zknerX>1!#cx=Gns28GSt+0&Zn+xZ>scaBq3#_?rz9Bw^JJ-`22v@OubU~04TdTC^~
z?~nCZW3nSMLZ&1b&BdZN9=nhYxd74TP(^98mPK8RL(<98@}I2Q|8DmeT+F}ZE^Ttw
z%QVgFL`+C-r~K_qusi>|aWv1q(x^?<;}wSW;@Lv#4ftDy)YenXyo}lnC7AJvPKc#X
zNf#zr)@@#r8naN2D+0Dj;z{sIB!Y)ZlBF>6m04t<gRFJ4BY-LR+Z1o<J5Kh{Xg1UW
zN5Dd$Y5+MuPyKB%V`u79+ChUX`^4#xh5H!Eb8deo(Ozm}h>S1Z>Z49q&t~+-i@1vg
zcP3b*X1{7~EZH9Q);{VE<MhyFz*evHrnV$Nu0)okOn)#DdZ=kQcI7{_H1_zm2C081
z^Hg&QkI^@N{-PUx?(QcIEHRR)lssrb7>GuqT_S(mo`@U`k7>+_ConK@xmgzWDU*({
zjQe+JV1pha-<$3vE@Ccfn1(<8VCGVYcrzh?X&nzI@cH^VFUQ$Bc$-jU$vo-+=TwUp
z1xt^c>PM#!+RLu1)tgXL<?$QGY*!)(b_IVqRxnTVnn|IQPG64@g>p?4T$sGiucfDB
z*lr9V6CScdLXqMU{SI8k>n9fmzPHYn=(;jLh8?+&m+sIBt~tbyAvZ0Q88L+c6j*fe
z;kW@Ybk)xsLO9!N2HdL>XMzR5)Q<{#{EH_!p*}#ZvOu{8?$nc!b-wJx?my3MIglZ$
z1<>Mc*CGJ)H`B!I=Ay6sYwwQ3hz!^h^y{B7+Ql<pP2KAyAw{T9J=-8PvGOU>^x2X-
zmm)o{aKvt&FdVv6+u+{lBQ!PP&=kjd<!SFEUykA*1PJ}Cb%;qC;a3kuW#v?Vxxh_U
zr<WG(eA<Yf?mj_q{!a%Z;^Vtf1zET~sf(qj_8*2FTFeDXXRGQB3F#Tp=1OKAj#6kc
zT#s8a<f_%3u`%ki33~P_ei!Zx6&_eyqY({Ctc9@$om>A2$CCZ&)D!2LVN&3jpLT~9
zS~h2woPrK;(%qiVWTBZDo=Z=nh%HMSU#<i`@~6)8RA5PL^?PkMvp1}-eyx6&8TvUm
z9QQqlbEqyn1ihGinN~F3d<X?`o75Yr7~YI@_vcWE%)A>n-=i)^X2jfmM^@58NdXM6
z2ggW<Mgouk(npolEd1P@9QH%5@{FVN_EVSL-X%ZdspBe&zH)^(QckY*JcaJ^FYx5x
z{qAm?WfO==Dx>B98A9kn&i^96pZ)KkJiriTzDWJw0e|s-?_<vY|NZ|=%l~5@AYvw?
z^l!f3GlAaX+;0QW{44u951RQk@ad#z4Hgkgv%1WpsxY9v5XFADs{Xq$)uAzo)c!=*
zSF}MWi{x+9caHTj-6PSR&4!n#G+vLH{vW9glf2y`J<-*(LCf#&3O`MC8D9ped)1K_
z#N{0&$?7@`3-)yG?@r#e&N@RzG&dxek&O3$b@DBszB4Q$V&VtVZ!a&K$-n!ojdeG7
z_u+vr(hpdyF)|{(#ugXKll+ITAeGD~$n9iL;h^zHF=q-0{X<0tlcxK|XA*`5kV#2Y
zefsIkOZ3;R%O_LdYgUlY$%j6e&vnO!uI@;JgNj-EbC7gd$8a`M!_#sR8$r4_gNe37
zLTr&dmDM(oK0WC~)wCV{%>TGUo3_I+YMG$oQ{(>Z4z!LCA#N9e5U&-rU!2STDUrBw
zaq5d5qTB&`l;WfT0UxgSzs^232X1jUyZ>pL?jWE?@kRlL?}i}D4V%4=sWW)@137b+
z(B55;_;&;X$NRZJ-%>ePd7i~!*>?hY1O%{o4p|2uZxFMUp8Ivp8S5(;CtevVy=jDP
z|GP}kp!t3D_krL)2!6sB-}HMH74?B_*58lIpIO`G>A|$+uuwhmf}afp$=IS0ogE9(
zl8gL-uxC)Nhh56k6<yf~kP!5p^nxPjv84wtU)NuVQBhHZzDQPj<_)*MzM8)5B<~l`
zI$0RnaXPs7YGqh^OgA52q|NMe%HI_LMRWIFU)fMMAw+iNBf1jUfy~(i6$-`nP>mVd
za$eT5KcWEB6M0f%wH|u!Ir9UmCi!crB$z?EQ1u5tBlCN^k~~t97UQFoov*>%N}9I5
z^vDq1Ew9ulUZ|cpm6W&oA04}mF8%D7$xywLo(wzPg_m8EABsM{ZPVXM-))1ROu_%d
zt#TgPyCKo1YgM#}7o&?{qh|F8V<#wsM58=QR}Qv>7ky+8mMm7A8N~R~p3Z&r2ZJKc
zc7+0$@)L40Iv65@fNAz;4i1HF%e=5HQ;UN(I41Au*>=3XA%`Py<;}-*WI~FK4-;@{
z;h1fSu$h*oQ-w)80k=h_wo6kFT$Tb`nj$2vz|Rs?L$~{9o@b$F!+-+(mfqRK;THPh
zJ0YMYaHDsPtfEi+%OJi|Z3#L-$c9i7SXwBH^x@o6;h(bVa77JX^kCVKTx_aViLjx&
zV;Vz5{!vGkD?AgIowMA+7)x%jEimVyD<e<$o%PXaykc?-`Ra#&6T0SCQRs%NIpqWA
zg0sgK9rpU_=hM`Nd|;L7-eI-VdboO1EdgpqtyC`b&jhUaRf$r!M;}0S!dUw0R7=cw
zI#%2d9zfja?sK)jh?kbYxFb^=(%5aJl7Ijw!(huELlI{(CKT)KHrYcrH_Vj|0Qhzs
z(csLxdb4x+U+N<5sh-obJqIbFBdL{(LRqWWGu(@)v`kYrh!ZIV^rBB<4R9D-aY{!y
zvLhdoP&6nsIHP6_W6*NCXpUOP_EDH-oLfgrLlTXG4@BCbW^#<?$kD^!fO)785kWTe
z?{&On7mk_b;C8?l1#4ekT?-9+2d16zZ4?G_+cICo(eu(EON`rk7W%ki==HfQyl~F)
z*(q55Y9Lu920q6ArwTGQn=~4^-TE~V1Oh8fAKv=Yq<wr2AwwE0yNrWA#Mjipx3s!$
zc1<1N(7~JmH;K`bGpByw)72T%BFLO(B4hM#sSGjSgGEde?BO(8KX4UYH>N37JpYbb
zOOLWvO(sp2C2_XDNzCYU-xg!H)y}BkT^8hP7d#qcYBa;3I`b=rS0jqrx*bmwVeTEd
z&fj+B&vP$x-H9K)cj;@(X=ovjjT1KE7!ylE7`Ht<XSnuJ5=J14=p(q;o6L7D{68^z
zAL(oI&Ypn<rB`d_sp;29J8sgL9ON)s!ED}bc7}@TF>-VK>l^hoE@y6X-T7Ya5~wQ1
zE`lhh!Z>+?Em)L#KGH8N)%hyH-gn`V!)S&L=i-dJ1*%n+AI5VigSAp6W2OxT3^jFy
z12@0>r#uo>`!bIdHv4omn(o<)qboCD(=JE6dCy$unG!6vC%gqS!7qzDGrJo}9o%}e
zma2vwRlb<jdz$~fJv935JX~(IH(o0QVMN4mdENLWYlmPJce^{FP6&E6tGEAIHOE-g
z^-}txoy?ueUOxg;`&@9k*tKh^0sZW=*kSTrl6z!HS$7}{xr2@ttt*zIbimQ~SxH+S
zQY_V;(^|U92$l1#H+g`|{mzSuP7KhBt)S8;$PV22n@Wn|CY+XPsZyjQ+rCw>qV*#d
z>VZdlkl~y_#as8bjZ7_}>%dsl<D|2n)9!}9N2jvpyy-d9GrjNESu;RcuW%@n5;wQ~
zn(?;bF+9_2i8W92h`yUgYlx98Q(9NPA0_H=<cFWYJndGNzj+T(QwP)UAV^TZ=8unC
zJX(-{snI0*#wSQ^r#5OI8Gbx2mthhtZdS*Npin{L<t=kbF%|g7`{vaaVO5B5svF;4
zDC5W}w3}!i_{;N&Z~>Ho(|#5**PcI_nibA2ssHS;YLwA$rhqLAm#=PE**O}9|Hc;<
zSo@jRZ4PyTzmBb5ch=J<0Q5J>dGrSET$dyO6W9<e^yeh#pmRs-rAE-++hh8l;)UBL
zYpun+;bLBEVL7`CwB-UEqOA3Xbe(0?ng1=D!O-4b3fL|rL$M64u+yp`xFwqNTD$v?
z9$jVo`@zD0W6JQgv^1El<6)rx!=DdX-eX*;j~N}iFDFe=>)emH*bHwwws>wb{^pQS
z9GS?wpl}d(=-K}9FzuIG5ABbwn@?rkJKcIX>7c2Qc2ETCyY5B^h_P7wZ7VmZsbJ-=
zvxQjy5PU60@o%if*FUj0(d(br*fxv%z6bt_MLKpW$O<y-nnt3{w6R=I(-jk0e{qBc
ze=ZJe^jJ-&`0rZ09aoAhI_yQDaWS&p>)i3z(0q{$Zr}H>fHqV{bCIzZ7X7+gW>s?e
zfq1U0C%7MqTE0wCKP%`00HJ-S|47eECzwrFv4{Z{Z{@0Y@3>qG-s8<EMB2|uAZw6S
z%oWXx6`Gb?+Zei6>w1Ldrh7KXmktrt-0^GSu1vx?apJ!*I0xdO+m8KX!k|Yk<)lRK
zn~SPf#Gtz@pTm=+!^zMX>UJ&rj~Ocn8qb7h44R`O$Kk81_)iBNy2#^2VT{EXbIsLh
zf0H8>_~l~lEl5%(FwnH)8SAF7=m2enRoN+eN5ndzBEOUF&z7s)q9c_veF}xY@y(7f
z*etXc6aFXaYD{~$1eh1&VqJp`nc2aA`JSQDxk0_Y+z&lJ4f_UZ`nEhLeh@Kh6(hKM
zsD=u$=l`3JekBZ~GJ8RvnP<t!?&G%`B|xuC!CM9rCzRF^G&Hm|D4^F9PgyYpD7@h|
zCHl6S*GMlg&x0q^V9yHIUlv*#JFSWmoT^y<D;)Y&Ngtt2y~K>+QbU57YQN(MUnujB
zW3MhJS`Bt}uZ<$0X@v`+BD1evoqF9Cs~b9#IhfIiB4lX#el$8a!K(y&;Q7#=^S8}U
z<&UE}$x|1xA=6$Zp#n2psxqb#k&pMFqM&=!f8Y<V@p?x0$3-7-PG=uIUwdEaO9?g@
zuDAb;LcJRXU*CT;eD>!rBzIc}1(9lCMgNgDH%TPE$eKQ4-L{=jp0(I>_|IIU#>(0L
zmFJSi0`wBA-coKYd3)Z}kW|6=q^k*>sR*U6)w0CM6<o1A^EM5@d*ns>FyR&=p^!h#
z)?b$CwiTPsHdrnmDzruUies`CnI)KO_8&1tb$|Y9ptEi=Txq6y?%gTe+g#L;;}~1d
zza(FKMR#Lp43k`x+mKq;u9@#R+x_*rIdlead5Nt2Qat}=jtxYka5h%%U9WRI{X`Bg
z7K?_3XF6=ku-9S@pzGak(qE>x^?y%bQ`ibcc&;}#HTNO57?V6@aPFVzZodge_I8R~
zi~<v;k72u`Tx}eVejvzH<7ErHmM{d9Nl=GLJG(L*n16f>2L*jt{2OZu%CKjXxZN|*
zKA7JIy842gIvWkgu=Rd?vgE;*`pErFO&fu{FEw@WgRM|B#2%bcTp~KowV-GGxNO%>
zAuGSIJgTd#j7ydml>j6H23+cC60fESRBp+?56Yd5ht!FgD1CZH0=9l*1m;;PSCasa
zh`Mbr04MA8^!6S(!c66eibi-hegsJjzn{ZQx+Ex`s%*AeK%r*o^TQvU_-!tyvjas-
ziUy)utnVS_9})3OEa`hIW;Nt=4mTv1_N0spMON@-9>RPVh79p?0+G9_8N6S+?i+ob
zb*U{$zhXB4VC2_oAU&_722YE*-xJWOG{2o8gR3bARbIh(8wf893D+1MC%}f_pc<y=
z|55V+cf-u|5YKaeEfR9#Q`OxO+duKpv%^H#iiql`q(4wsRRcIT(7}}<XvcBID6~UX
zvc)2oL)}o^5NyO;k>UQ-O4Ef&JQTPD#%=zGRl#{=KQLaxIZ<UpL@fB9o`IP_+^Pag
zow}yg^$-PzBk-oA!vt$%%Uiony|STKnyOz3ow$s8!`^7wR{4zW0*iH7CH-&wZrVz#
z9=vt|%m)y_qy*eIMwI7V@`r7pkiHsgU4QUykI50yc1rTO1QaLliG23geE0wiuX-sV
zRH2f7&Jrfeg^XJN*>Km4W=co%Bu8jg?Mq{O`%#};&flPmyS=m>r!;|4i~+PGziu@%
z;9O!p;?xgd(KblQjG5E4|J0qZnmUQ&Z<oV<%@WO1(ltvMsy>g&u!m^Cci27pNvZ*S
z-;>N@K-FKmf<iwG>s%sw0@374w3G*M0%Sz=u|p`<&*mhIB2PpB!=~bhl4WT54Q%{X
zw9=O2VPYqY>+6zn&C$<H*r>uIcPMR}@mu^NLQ-H#ja>ReWw5Y}ryP&AERT!@u6vGJ
zS%~wANqERI@jpA&z&YEgZHc{`R~xd_<H6D0<}5*Ae}}~B7Qsb({<M6DCj>DysEoq2
zuJv^kMw+;G0rygw6o$T+4>~^5l}2d1NAMydA6psnJHAC%Ouc9xd%^TcGdo{BrGdzn
zIr>gz0WdQ`Y>oyWf{fVR5ISM83M2^Sp>O2Z{~7AT0mc7|JnOtmc@}fB^)~dPoKjr<
zX<uXeu1qPeuWovgN|Ua&j-aIA6xhozg0uxkre}zx(vDuSsxIkk&;K`$-K0titiVAF
z@katRhI$$808S~vH92^kFj(BOH-%fA$*;SCP@E2gpGuo}4L7J@Oe@p(@k7O;$JGUp
zZY9_F$`Fh8$7{wdmXISf!k={i&Ow1$ua1$lsNoV8tjg6=C5ktFArNUObxd&0*@bLH
z`K=utN^E=f?X_@3O9lLEqLO1+3^+{FJ@x!zp6t|c(kv71<rpdqPZMs%=w$z_R6)W<
z4VDp^i9Ly(DZ)M`^LkMuuyN#}*npcKHrsgGc!Sgr68mwVUHUWKZzA<oEz+GPd@Vn;
z7#5GBttY&tx%M?0MmqM<X~}}57p-`$!cfRWQASkehW;Gf^`Ox5PiWi)R;(2)IZX5A
zLjNC%t}-sFw+o6&BZzbe3ew#TA_7WxcP+4VH%K=q-TXniS-O^xhQ%eNyK`AO-}`>s
zAD@<cpFPhxGjnFrcN+4Gs+1VgIA)qyOW1kbd~zo;(diEZvUwN;#^<YDT;Q$ef`SbJ
zGb%r3i%pe&O0wC!0dz|8s*@<u7dlzE(&%rk4`U>E5sqYnX79^g^Gaz-Xt&3e^~&hR
zTNmLCc;$6AjReN1#o{@NWjX!3X>O$o>Nr^E1)sBU3OfZ-f|Og$aFjw50<T>C5Xye2
zRg|rQ1LWLgTp8;-*c=5Bum0$E(`=CJ=QJuPE2rn?);D&_*EIqT%2hlII)vLr=N5`3
z@LEDfd-AD|a6ZA<AotYrCi$jC@#B|`IFrx*`Brhv53>zx#Ta1MwwYww)-^Uzj*3X|
z6@CL*>%W*Of2Q_^UX0oGW$9{Y_wAtfJhlGkl<mX2t*2Sc5mcW4q)ZViCSu%cFTebO
z4rLI9NEmZJL`sc+9)HYJ$Zt6z4X~=yJ^VJg$olgjAOTV|`l4+Cel7&PYSAlEJ2rlu
znBMfYANgtQ@@0);f<k?(F|`pK@9S=QZ2{O^S*iT2xA>&ZkaM4VrwfVydE;d}`di&L
zB^@-<=QF+A3e!9C-v|twx$hsiqS!K8^HCEm*?z;-Fu>e23cIc&<*zl-(NUyvw27SK
zm@vN`sN*2<q30EWDLGYaMuC@nO7ZceacF+qEXM1%%7Mfc9rtFj-`|G1h)dgzKh1!x
z3!P^_YV|_S@ZSSYsFq}M;NM(5-`*CRH5fXnM4C2^WV!q4L^-^ekT?w}dxmm!qm0@|
zmwhAG(;HPWjc-5+g@45)FAVWk_AWHFBHVFa_L}=E6UCuQkvc*r0prnjJ3PJ58B&pF
znET!)Kd327dYY>o&!nNl@HStm`y*2kMsGT}hPa#gLcJDAtu*u`341tvHKIOLjUn<}
zDC9E{KC{Q3l*ShcwK~h3CTMsM=8bna{nSCvZmm@@W?1%VN40VAu9PhL&94YK+=>Rf
z<_A=|Aony+n<>%Gy{uX;+~1=Yb6Ev$eVhU|bQNpw(%ns_;dVrdJoa{fTpo>BBJ<x7
zjlKT#iPI^kGU+e3TZ$-MAjhxG0G+HT{DD59j{;<^ux@hWlB%Bz9~<-b<b02Zp*mDU
zuk&ONM;a*}GK`;y0&caRnYZl*oOsQ^APP$jgPj4=kSbdT=p(VG<Ej-`p6l6OBxH)G
zvWh8sspNZID(fg|+`v+ca}J5b-8*XETT7WmeVsaaJk{@J=jze3QWdE1lTZ4n@pP$f
z?1rzMv=mi$hp#GrYhFmYu6ObNh^Q)NCM^_}45@m(#9h$%T_`u*=kRc>BE_XXm(2b(
z7HpS~sq`Wz&3$(+D3O(@)H<Wis7tr+hp(P>&r-eJ^Ej&hGmKCK1%hX*|D~YJp|E%1
zD@4W_ce4KlWF=KrvAD)wweS0(EnjSrQE3rPr>BpZ9K0xEQhm2mGGQt;rs+9b5RhBW
zK%3y1|K0J$Z1cVQKH@!5xy_e|L0{(0`sM;z?o=N2NBbL#<EplodJ5Kx+33%4Hj_US
zGBGw?zLUeTXTfcOn8faix!%o&V76}nqph14D7QZV{<r%2D^vC~V-;!qpaq7n@+uC$
zr=xG2z5njz-0gB8Vv$nRYv}5V7}(^D2jvDtZ_t`Yn{{8E;KGCse125f?o5%vFww71
zehLlFlsGsYXO#HmyyL6+hnE+p&E%xCZg)Ix$tr7BlUdF#eoQ5+uFKZ0tMw~IW0luY
zA=60zP>VKp?qZvi;0*Iqh<b=`HeaHSELPWWBMDhp7YWq&#QE^M&GrIcKD^4Db~Dnw
z)H0_VWQDOHJAz1^&R2^0YJD`q7bJ6xIp1|j#%CYg5V!MTw7c&2{vZ8MnbNNqvD{|0
zntpr#E=>ylnQwj$Z@7H#{wYOv(pe8KI(bN~o%F5xwl*vqPcBP6!<rKw=V^m)D3Je-
z?uL+3k5l!Hpmoj;RL{r@{g<mxR8%ARxUF~oGRbc)>8fiq<o2V518<%TH`eYZ-@6n1
z2BJ5yGE(OSAE>vcAxyATK@bXbLe{F_E46}ZUHyV@{q`moB+0zI{wv5k{eFdzM_=K#
zaP^MO106N+De!#G`(PHbr3*9R{gM^C5QT2HJp%@V=}QQ#AVQw20a+4v3#Z79LF1jY
z+}7_3#3^V*P1O_RbVM>Xl`e`**0k4YHewX|vnrDlCzm7{SN<ED2i`?(2r7GuwA|Zz
zxM>*$7iN*FbD2O3xMI*xY>u}T>Yi1hD^$BvPy}CcDvSyGD;!66e%J+84_K}>CYco?
zKQ27oa42~h;y6;y?^S>j8XsOu=)SUVGWFxE(i~lFEvl)tSbVOtr87^EH=d}X_+Y7R
zYD;n2euG|R)k~~m`Bbdw#Qovz{?fM#>r-T~;i1$j^)<Yj?)8!`w-}$VQ6*{=`(}bd
zA+;S%7KX9*!Jb^}s9q$E9=CuTZ2UoMOJ>t9j0jyXVwyCXl_HUZEb?+8{9NxX^QRL?
ze8<q=g~T^90U)VudqhFc3=xQd!C;XJv49u&=dh+rPMjP9X`X<c<QgZEg!QUw6G8Tq
zr;^NzA!G?uR4)8to4exoHcunF;?QIg+uNcz9JP><Qd?O=_6MZOa^x|x<nlbR%KYiz
z<t}Z02(YVo89>9CH!;i&y-M1ti_Qbk5_(zPQpX(hxlK8Du*GbVQB~D5rPH_3B8Mq?
z_bRZHi>P=W@@^J2)Isx@-~c)sqSzbT*kE=`e-L~bou+xN?><5I=8(VRHm4lTzwlqH
z&Ycg`$0Eu!Fmg<%>NfuCVvP2*ff~KNy1^gE)WTV~^rslUv<tY^o3A$1uinkl|J@%@
z4uGqctY~1WR0>#DRVTfW-`zZ|)_O-a{w|331Cdxm>V0BnrQQoW5}5jaBT=~CE#H-4
zHn#t)AlP{}oRB6Q?uDVI&Ms%>7yjkH21M+{SI`Smoba&`vP-jg{_q8@=@?XQ5h?Hx
z_jr$pr;tz<4Sk<pOce{xn%(horeZie+)k&~)iNOvsr@V*crfv-D<iVl!^S6mwH?6Q
zIYWL(scHC;CMvmcprS~sB6Cpq06jh(Mf=ZxDVXDz?HmKGDQNh%+t=TPgoJjEn^(68
z@@|CAMUD05A~Sy8pAboCe;!!v_FWXEx9|Rel&y7jUbQZFXx*~lfF2oS-4Pgk#h$+)
z%lgk+{e;8jBka~1QOOXuCRdW$b+P39rp)Qu!N06~`kpexQKLMLch0?=)|P<_DV1th
zko?kRHR#H8T6zY$Rnp7xm>l#*(QxgWlu0*$cp1c_=HxDwL{;Jc{V~hhl{qO<;b)>s
zt>XI1Q^u$7GOtNA4*#10$sO2YEfankzm<uJQV`F~`80<4oS(w~6|dT8pRc=KO&qN-
zhTCWOUy>iaJz?nZJmG7XLY;r8GEoWG!$dLrFAbvjP25iIbBQ8H7`YMU7%%Ab`g;0&
z3F&ul_R=ARv<JXS<}jw~ZfQ^kQKcaFn-mP`pNM{mD{N8Z5?l0o5DiO90Fs{LacKf^
z0h#bQ<*ZCI$Xc=Pu-o_^{H-FWV2INeL(o9s&yCI+!Yes57l^#l>@N%fSfRK294(%w
z??-N9DFpJ?GxRxBab#XhYYzLKc8yzi#Bqhwj1D0mmVDh3E`!mlx?fN$5a^bFX7jH!
zND5HBkgrEz7Hee{$<vjA`)E~jT;v_UxhVO>m+U5=yX!6pe_b#HRtrwn<;q=?8Z*JN
zulBfa;~J_JTp&uZgPFO_(3CH38~=Iz(5h^Ar!KnmkX`$>_5<u>*XU_`tNSBQN@f8>
zr0g!VF>+R7ugX#aD{_7DHG=Q0O&5zjrz&6>^4;>?tW>gNlvbPerVjjT=Jc5)0cNg3
zsfQsQyGOI0R1hms$eXkX8{L@ipL7?d>7D;%{>9g|ieKiB!8l#^PI$gMeINlw(3d9l
zl*v{)I33v5I;_;EN1Q7Qrh?>2Zw6wc@fR4&%dfbo9D0@)o6CrakPxz|VJ3H>T{oAJ
z_2JOA>e9n^^_1#3g5qM^w<N|}fe3+TZ3F*1rb@^dJ;>UgG``u7-lXU+bDwH+l=mOd
z0^PrB|Jnl2%^5j$92rvx%j`EtHZgXmixlOAC=9%lRdZfSl@!+%-)nbVs+D!vvEl1=
zbL!`pQ~P8!eurQnl_-JK@%U_5IWSZCX=(TEk9vbd%iDKD*UQa#qHUiE{=0JaA$_%S
zN9uQH=0Ox$q#D1_<K64I>q+uiqr@)x*HrTt_v)V)94N8H)mu7P<VwHVKA}MOinC6z
zW|zk3(9KFTU2-nm>hV-wM&ozY7^Gb|*mM&l>X3Kj@+I=-tCSHr8I!4NO$?H^PZ+sl
z)0=0D$LMuj!K)yP<JD(sC@H}`w(yiyim<G~{J7TINnScGXR~Q;jc><P=7J5yKR62!
zyA;K%KAMfKrEV^UA|+;b$(jSTm2o%b_+AUmI)H_!RAr0Oua?`%^D!@nvfUit7>su9
z>hr;;tvu7Zbk#IEcl(S9W_l0-F|u`avRkjbKX@Um%F!_HT^t<kkkV4u)PCk05lpZ8
z*w|Eqkd$tW7@MAffEc9S6x&1Er=^<c80Yo0Kt$DnBp!^wbJ|Fs*fD7J$N>xFAx>3`
z|7J9YMq?F!%(*Z1`V2mtGf!T1m2?^xXQAe|f0@-*%{7n+9-rSzzCVSII_WsMoJTI`
z5yxcq^}RBxBo@@)v>QI@S`B$R@jZP#!7~U7cuFvhRa&{UtON6P-k+bcm+t+s^QNle
z)Ss32W}<fW8u+^m%7Qh@HTpU<zf|wtQ{Q>_NjXJygz>F3u>YoBIXJnmtmfI!x!C=e
zCg|I>Z!V-$4DI!{-&*xl{#VW*{D^wo?d((6HmjG9nWH16#W@FsLz)D&)&V8%Wlaj`
zg$tWq{wqY!!=#7a4K9aUz8v&dHZ(adCEcW?4>&9P?hVeeUy3Bj!e}^s3Dj#DVzzB8
z1i3g>B1mIjh|>qFjAmykl&VU73o1OcX{>idKc?x5{t*(VhF`5(9*1+>DsIbhJTY@x
zcIA9L@aE~<_@0{9bVG<Cj-uLig+(sAZ@w1P<{DYmxme|{ffSN>gyVTXFE^JS0$gx5
z-30H<MeOZSY<6bFy?L(p6;lJObOZ|o_NY18eEl@94IQ!I@0m$8Q>kt29B<50hVZTI
z9N7iWX7&uVKg2gTk=N9*Qk&#h3R3<|XjT!#1rlBFW?1aLBo4xb+4W`zE@^iK0f!%e
z5kov6CtQUycyhF22hFPuJEJ012w$D|1ht<+C;YndnRopjw$pAm(w<%yu9yA{r%R`#
zf$XUl&kAYr>V0Rh^{1s%a7+3+M~Rou<<3b^`e{5ZoAZn_PSJ#zd`P2SFdDRw>fuOG
z_(8kNfR&=B3@@K1v0toH<j7<Aek9yiw?iY_XzFAS1JwWGT=MNok+Wm`^~76l{MFeP
z%)(y1mU5)brq>s54Yi-5HB5hU%BiZjnQo~?#FX-VC0X@)Gm@6}51EUe5ug}?ozt&D
zWUJCpL>0nSlDVb_Th&xOGpxDejm1_$u`8o`dq*}Y^mDIABTR<Erbn`|Y<Ur<(^*pN
zt3w5;DJ)txHJ)izcb316O2eN@)8EpDmjKU$>z8P8r13!9%-!9Aa`lHUE2Y~z)qrN1
z_rgS%<TI7MgTCRB?}`%ZH;(oa>zoDaSYcM8EzjbzvJwB<ev>*P_9y&*CitBv9T|0w
zz*dN?AY?K9!IE5GDpfS;AoGOTn-@3u?wZ_`IOz&7&i<hCa>j~i$UN8e$ue?5MUyLQ
zLMTUplo&Mxz_`=7C<%edtk5&S@nr}e7IR2X(7fHiF#O~7N?=~US&!>PSGOk=DK^p{
zoA4Bsa6YZxOo}r=Dc%y%7!S(xPVM)DWjS{BASa-IVyGWqD{1`WqZ2K>nJoL-uJAfY
zQ%)CpM!I~iEx!xdN7~MZb^i>DLdLehU;L`yufrfFqxyhzET>gI%uDge{G5jBtw8WO
zthRcbyGAa@A&Wk=Y_Zzr(*io1HQ$GKU5Cl&EKYKK*O3+~UDTM}{oWuPgCZ+!T@^`z
zsY#i95u$18Lmu|%V;rA7X%bXSt3)8A1J+Oc`}fNi(wu6g>4pVtk~9@pwmJLh>_5Jx
z36scfm%rn%U?no-5zyQnBr+jAh<`b}0KOBh3+E>iVt8jL@T)`OAD0d4i?Z_Qq&}%?
zgJqE5!WVPB<07US^0HNVJ(3sUJ5BZ(DT0Y#JrkUVi%B?r?*3)#U#`ftDfC$g?;2>h
zrwsT;{^3i*h&P>ZD@vb$r7J46AWs%4-%~ny_>7#)fIBYcCy?hQ;^PqD8N!~38dO<F
zP~|9O+V3?irMV~zr2mV%wsw69*=JhZbS}iqSewl}p*^YmFsn94q^W6o7~OWB|FYKi
zvV2!mc`fYX_wVH18NQkSzd7;2!p(?|y9MRi>H^Z8t^Od<QKnsNDlwlSV2|m(X(ukP
z3kHNO(}_QZ)vWzp2?3)%ZRDO-^!$ADeJLNiKsz_o{lBN<5-ldnL_j^r?kB^Hx>8{h
z0RN~I3SOkrtB)%A@Pe%-&Q2B14L;T)K#}?Toy;5SL!H;_vLbPYg`~gL34bJVmT1+q
zDSmBDVQ#7Y*XXtA+3|xb(X#8W9a_u?j~6JdF>YF>nRkHrcR?MA0DT2}40>J+UEg8U
zf3nqI%InJ%>G}id33CTE^c*W<e!ZYI$(p&nk=*a_$gcR|rD*Eiu2*B<5cLMax_PvP
zaqvr@+MCJZ<g$!DSxYj<eohQyhuw|$hc~?v7Q1RiWM)_a+{5}a{p+BQrgML^rlRKl
ze!{|>a}TN*w>Yecaxb0|d&4PjKUu}m&U<ZZ^!pQ|GI-q0rMKttzT@%kkWrh9)B(W~
zc)Yr^o%igW9jed)yCo&S@XjDC(m6d&xg|V-&9E}p9K^gEvHijCotRocLYcltMkP9@
z=Y;gPD}W@`>4@OXW+zP0aOWk@DvE;HEL+#s&cwA?XEBZM`)YoIqB|e$i$Ntzo%j56
zH4>xy7QHEtV1xF6kdc7AEM>zI#ttw<i`mYV;`WR0AUnY+|2GKh3|p4j1Vaz_9rKdB
zH5sUqTGTcBbyU4`V9~bw;R$irR<db{f-v{#5J9_I^{2VZONl0dPWqJLIjoZUhm2HX
zAC6xh1Fi5mr59D9KXekj)_-pa9i4tjy%&w0BBOlY=Y~~ekKQTI6v?aZ|DDg2?u*(F
z^sJVw6why_g5B{E0Y|=6)0><(M0Wa8xOh~Dp-B!oOc}Y{opY6ai?Gv}mw#nmFvS)R
zb&OgpQNQo7`Ig<kbrah_4-%!~q{KU05hvO%C^i`5%CK^LwoNKut`WKEoJKX}%Hkx)
zg<Jho;FSJ%{~t*4Pf4marg<!|l0=R#fv=C$495~Vl}zUE`@m-@>?m_nJ&EtwH8P2v
zWVB1yKJ%Y;H@q(*`@N`z=+NH+@2jE&#F9L9U-)H^48_ZOR_tS>T^X<={8?aAosC!=
z*s{uo5Qa#~j$FN<y#Ai>B|S}pEkO#LVH+cjNgx63V&zdNQ4e^=3&`7d>jj#TECEMg
z9Q|A1wcWnDmeU+KvfFiLnM@uO4BT(ZY^jcZ$ZL4U1|;uZIvJK|ov573u90w3{VQ+^
zbotvQv+sUqGuU1FouojkKkxU-zWNx>$RV%$Pxp+C9%O+&$8A*nHd9j+a(@YZ*!9Re
zL+^ig&vx!&)v{q4Kl!5JG{$h?MShWab(DX0Nv<Md?v~ou^YNcKw7PC`NT-!mfX1J+
zVs8?+Htp^+o>L*$W!xca_~yx(0Vz&fgNxxRY01kT^I5UM5aqi66tX#}>Fr-9CT&Xb
z#&>p#Ddf8DUbTz1c#^zIo(Zn1;tD}Tc97OWA)_ZSVu`yRmJZK*e!xBA_91q?cI}o#
z;)|}{`BJUe%Fb*I_4;x1ilyL)_hTdV*;3W+%}`9<+K&1}(i?Q^r1k9n86xD`rNv<;
zSinJLKC|C!J>)Ldl#`SzXX3#Oxgzc@{yKDOQ#R=AtjC{f2|5V7N-xQ4@4w8N@D1O6
zz5GJ<bt&gRIkwIv=4#hZ3P7VQDsyMZWM}AlvxX8gPu8jxY3(2ijGKrSflPA3?Im<N
z;X|hkPUA7I0ToR?dTSS?_xIHM^rTtonbYqZf6(8v6T4EqUjlJb)^;{V47<2BA?l2+
z-~85=>l`}5$($WVFkPRlN@82+!mv3yx|gp{oK9Dyh+~ANY99HaXWPabYiIwbgL4p6
z8@#YvF<rl<|FEq7T<yRM%Gv}pZR-DA9c?<$6rJTGhQ1s(had*6#WK~H@KP03NjhZ<
zXNn$b)W5ShxmOpd@6b$_c^2hvJrv(pRg$??XtRA_&Q}Te(W%QQT`T=#Vshs@IE)OB
z%Y#}|bB&$?E%d9@_rHc<PE_{M(x#@;2wGd-M%OKv&*6wkBtKb9PZ0NLF#3GGQ*CO`
zHD!nOZ=YA}wq&LJGpmYM<OF+;$>ewCiyx?~+dbvLLf<tt{Vu5=Y0*MH`riAdvu<8&
zVHAdnPU(P$v88^36kt0?r-qv&Gb|m~PyPV5UOTIbgB{BSTr1#z3tUuj_^Jl&9x1Z2
zvIfi0UO4;}Pyx<1sh@%l1#)-p#zEdsVF}#ixVNz8uJ4Ah3<ah12`ixapL#WQ5j&Y`
zFuw`{gCm+(0wUHw?Dmu7GH^7N+79|QuUdL~W>G)i$R@`y2>_wPb!o;sG0-!HR-a?C
zt7-%3NZc0{kB-FmTIu)oyrvBw2Xt85weQaLn1I+gh4}dNiQ*t-f0x5TYAFhdc$Vwg
zXx1r(iDDeIkm4UNS9q8voM8pYSKh<HetO#2J4iCmlaBuO(+*e_D7Q~<ouS4*;TVy~
za<A=uyJ@_(0O8>OxtPH6lQ$^s!?#Q}Uz<dqZ%opL`ng@+1RxU9Lht95|Nq;ju~w?7
z805RW+SN-RIKtPuB3z><0-r`jua{yX87Uq5?_HThWFTHEUFYKqgXkZ(fnWyrNzWPY
zk0O(|;p-U33r^zk_ImJ4aW}Fn>0Em(wehehjcFFf-)U|Rzidqn$A8xt(jKccDzg9m
zH};xS>5E@VFJLQ2E~)v&&yl<q1A9%lZ~K*jkqME8_m6Sx@`C&kmmPXg=C+{pCdP|^
z<$(PERE(l0LO(&SBEEr{V<RRT3yO7Cktbs!J>U=2p*|?jlz`nAP4B!LE@OOxvfmx9
zvvnb<oQxF?jCt$>?SxA)MCbsHU&6q!bsMz6p03D}T!(`u1jC&29<KuBZ9jEhJ7Wwo
zFeGD3cmAcM65YyaUM(z#_5efyFC|x2uBs*<R==OB*#(%fd%f0-3|B@FztWv?`(`*H
zFu2j^4OK#f0W^T<&&Lr?<Ih4hw;(8Dbt8Mn_<1`pL^)swDR`Kg{d(i`oqIat3<sD<
zl@>0CI9nHMm^Em1sUop^p@fSk<m>jWQ71nLe#NgW_S}({>wADc6rH_n=D6=I2o}ns
zG;aVLSFUx&N=VHG#d4kab6ihX48KxJ-1J-v<z(0LkvkvGvf<)V*F*TWka^EIF}Ii{
zvUp?;$G<Y{w!26^ekUQDu6OgH62h-GY#-rKmKd#&c*37^Tv>C?dldE6u0Ff0dBlgz
zGxbRh$QpU&LWYw;j)t4_`pqrZXX5UL&b!N=^YfiBCtimarZyXnwqxk13CY5lk|U9G
zaz7Kk_m(?B+$vd8JTm7VJFlp0GJYfEo4?;LqMpnY^Um=~i_8fkjLh|?sc?dY7Hp*~
z{dSfZN%=k}laVI;SN$LBV{qG~39n_UEAN7pFsZtgkP?^@t0MJE=Hj`fXn(Px8OJ$M
zW=e1)bOX?9U84Rp+RKm!*(HG$&mU<ne}^%26ZQWhfVaut;ff9)))9Yh&|2#74ms4W
z;m#WOc5?b`57t1<;GTBxw_c-Lr<rilO3Db{cl*J`SBQA~Mn~Vg!36PZcYPg2QxgJ#
zj@k*E^XKqnz?^|xVx?TB_;@9!MWVd>C6JJy;7(6S1gq_+iAti)_`F|uMd;M<m^{56
zgpaDelMkq^kz-q|Bfn%5aQ^{5Mf?@%6Sh{6Omw^x<Pl_|9*9m&7)sT=&$l&>>cXiC
zcrVSBqja-y`sAt-4ZYERVvFu^lE!w1IlFibU2HM_#&KTqJ0V`Y*G?VNu6tNOPqBR1
zVvyLl4!8#N)Hq(Jor`0{C9$$nR>)t4h3ZO5Rvl*#Z1D+788ErGeP9WHY;)dCMCA49
zvU%MN+Jq>QhWdVMclf%RE@z`oa8>QowRsTU=iFTd7>pEG#zMCT24lbXy|>UE@Zz;_
z42;=KqFsS{W&!jUja-jOqN|@@=&<WI{s+A&;Z#e=wukus%C3K7NrRG}F;i{bU-Q2a
zAITj#O|$0`2IP`8AO-M0F6DfN{p%C>SR@@4!62!eT@B{Jwaz02^@?7mFtWIZd)?Fz
z!4z+V#v_*KK;+F;N~CUgcew%vUpS@1aW7G4j4xVqv@7wPI#qn??%GMfCi>nlZ0{;l
zqy1VkHS!x@Y-FG9!SN?VA9+DZOq!YzB--9c4PX;GtQ7pLa0ui;qimBBzaoqu+d4XS
za6VlNl@vEov5sY~wT!A_`3nTKzT$qP-<N%y!)5`BshF#-ZElZ<TnB#tJ=RjXx;q{{
zc6HX_w3%-!8^^5FqZPPF8Nd4d8dUbb<6+&Qi?M|@9tsHgT!_!r7@MDt5M>g@#tEBa
z(Lc3+0bD$H@}38@Z+8Tqf_8I#<EN5ntCzjVp*fXTxtemS?8KsNfeE)3?~G{rT{-+I
z)d}TyDijrTz~g@W+rBk^X{UboB1gxkt}bq0xF}uD&hY8AlZ#X|<I(XFNaLE)VZ{H|
zO7m=_?b*=LWv7cQeZhG?st)zxex=hn9qyWlwI`q{TCdz}oD1kZ@^!@pRGGl>JhBVt
zA`79}1V*Tl`|A~uu#mYZ`8hW9j3|85K3m_t0!<>mYQMaMO#(Y@I|j$1xlSUc`PR9j
z(_~$I2Vzk#KT?^TGUI}PsKguej;OXcrM_Y8Pb?flQJqX~#v0<$;Fp>ISU{NI2D{f%
zZ`)2!ZuA$ERgr5<NiUb41oLYsHV;XM)xkRRENlt8Vd3EWhHku&*})%-rRBfS@u)Uz
z4+oVy4okVhMR}&(<0e-Or%kxf&eyES6qDSKhj{vHhy1z;;*XtS`}?uWRVcDaLNU^c
zs)RTZuVPh!OSxaj>-x&Dc#2F{%SkvR6or`DfH_rkLe<NvQzu!c<RAH)f6MQ0ZZKR9
z+IE2PWfPG&;kU!nak*2WfjxI*wL4x>K6O2_s;VD}U~Oumq@o0>$3OCfAqY+pA<#Vz
z?_+A$EWfDA$jNbjI76{){QNdtUtjd{0l8A7+d7cqF%Ksw=pa_?b9}ns>$)6wbmBKu
znWsHyvrwJ;keKA`!7s}+?#5r(M4V*9tp9UcD%%xsnH!2bY&U><gt5%k)O9vDi_U9*
zzVH*CdlY!6`1NXoQg$$*)=WrT+>DnB5vA-GTewNP10o>g%t=m;k<|gWdOxC}&jDS1
zwv(S*cUH}Fy0YJf>z?1=tX*3Idw$2^&Hwi4KYNX(UG({s%q&o6*@TspU%qqE^T(P3
zJI+}y4fD~mk|V4bS!YS<ZP(tTrpU0^%)Z{xldkGG?6e-&;Ki#dcF9t7lULmDGPAv0
zNRN@D>-TMrR<&Q8te0<|=iQ!E#$;ObRgeSshzM!F#6!)cd7XUK7`8nK4_rf7o1dW3
z^$rhXPR$ec;6v8QkAM6TV%#MLb*egxk|&0>B<#C|Yh_CQ{Gp~$f=%^42Y7KVKKxR<
z;Z;>Ai-o-<#LawqVBkRsi<)(n3i-H=ISH=SJRBHX0i9Nw_#J-N1GaH5=_`z;YNm>J
zMU?G}QnB#t-0^e;5<$91=37L9@j9U3LwLnBuRL%GR39U6^Paie+qJdpUtGsalA2Z#
zB_)}!`0grNCDEQQojp|P<Df|?IFwfW8Y2b3b+)JVnn!4>=k<F9z#{O;ke*>s!#(O>
zCh)dgFZZ*6KF?i9mMMbE-avltp&(`BvEB0~&iI)jc{{i#--JJ_4jLGilVd-iZyq~R
zI~%sg`Cm0lQ*-lB!_Jm<F9Jd4Zts0-PQAMcXv_Q7iIM%i?LAKlOz?Lo*~Wu#tX$|N
z<*e>XyK7osM1t4FS)#Y@ZR{YU-EqtHlNgM?L}MmaRu)gS`??G1ZghJ>jjt=`S<(|z
z{QH@6pQgj7@cUkf#uD4=wuSQN!v$|6YS?jJ!wySaZH4VKZ73_7n-Wv_$gNWXPlaF=
zpL#`x(D)Z#y38nR-0U66s*Gz8BC!4nH9EFKNn+E}M(o5ao|DDiHjQ2)-K`;`%i|9M
z!&2U6GlqR3f-qDm$b6RdPYH?WiN%tNN2f$dP-A?3xr$uJ@MorIQCTO!lHmqp>|rE|
zboDULrdY$LKC`lNA!+_KyEcrPMovn?E$M+_i}3e)Xs<%FzF30kr-Xo4`8U&ZwR3!i
z<=*d(C$E5Y7bluB?o2-LK-u{o827xKsO@oL>S=f~`~7-M41jSd>3hGK)_<{Po}jT&
zYXA5^WX+@dO29pECC0bhPEU1Que4_?;d&$bm2WEBF084daM)-bV2)3&0q1>=2zbKM
z#{8~CuRbu4&(D(~Z?oqD9{h3f@NRbIpgk<BPwdpUdxM<bJy{aBACVCCO833DhW8Q1
z_x^yZe-?mGOFYFuhf=@-fB|-|7r5Db%SIiST6}E}j`HE?^z_#*nf6u9+1Xl9IcwQy
zbmuZxu%$RlS@!mpS1=cuZY~$O_tbh}pRDNh5!*`OGp^jO-^I3=)&q*~w<@N;GY9Sa
z7!-{{t#_B&#@qU<cTx_}dtAtEu%vi?rnW{=VfA|vk<Aa^ykueR*DJgFt53LQ?*VYe
z(d&Ava*dY5Kf9AfK$rMGekjr5RHi5upJ*G;Cqt%3(S1MKywmh>_OH9;uSjY(3M}s&
z+>NqO|4qi%l}jzWL)CLG#<sel`)NIg4rMghOQ%(+dn-*udari?_w$*!RvD9}tFw8t
z$7W2m3QvqM!)*7T9NNAXJPyocM-x38GXPQC!1%HnD!op=`iWOoT`q$clqcga{;7#f
zlz}=)h`(rR6H3a({qR-_owt+B1#RgUuKnD4&PsFt=eQXBZU<XdbIt;ZqY7gnCPY-b
zWSEMxNV0^7#EOo-ts4U{6A^u!aMYwJ9wXmlzOILDV-rD2Cl{Zuy5s+r=hxT!(34}x
zt4vfZEM$N@KWPWx;{2wX&j!qg9k)A+OP#)1%7J&tC8%d!1Rn0qLCv!iCC}5q^Mj+y
z=AG+GRpb3jpZ51M;DCohYJ><BssGPM#*m}s{>dLGRUcp0JZ-&M2{`0u07R=)ROZQN
z?a<4`j{Dig*w<DNCM-#i5FXXaW_!(3q?t)NyH$tluHuzP^KDY#L)uNh#3M^fy58?M
za;E%Q&d34Hh9;l`gBKDQ1#Z4cp0u;KSH%fsouj0l06G|`RlUwl@~$_`tZR*cBQSb%
zlXk@q$>6uo0{tUveD_-b^*5wQ<i&8T9%{^0kN~7A1F*r81LjW&Sk-Rcrt!m0jqCM_
zp;eMl^J&NKvJ27C>G8&&y3Y2&kO*|YT@m0|_Iw;*tX}=Mn^BUx-4k+sdq@4mA9!Va
zUx5?8SzC41-(p1_*<kf2MpamDH^Ya0wlp_7Hz;>+MQj>9n3_DS>!egTy{4nHpSdq;
zv`j;vUT%km(>y~JyOO#Y&sxdtW_pj98vWf=m&NxC!RHjc_?v`${Hf;j-<WYte+zRk
zI47w-6HnIE8EomY2HoR2*z7IxmCSFSM`>(mE|mV@#c!HaQY5*~trd6ck22@ysGB`I
zI|I&tuQCP&RpU!bHaT!K4M@CZxN#203i<AkHH4;Q2nTn_Wi#nzDv|beQ%f7P&rZwb
zx#y!duJPkR;AmkVZ8eZS^VCOMd1*fxl5Ok2ghocgnWITs1$+S7s)x<M8}P$n;C~zF
z5;yp>!bX9|?oT5%PtqQq)oTSq1V)l67F$TGGSAz==e1MU)2CO!KG|i6`qmR>L<Gb1
z>_?Z_;ZCx$YPME*$K_S~zlHOC5~Bh=*0&!f&_x9^o&31h?;779j5zOJ-M4_M&kib9
zWkINUx4qQnVBcovlo>q)A|db>`spfgj-_^x<P5>H(FL3eN(n<T*>;J1i@4x7ed|He
zx~AY!Ck>N^O{DS7%L&nI{Hq%A5H2!!)^L)Z^PuMY)P_6uaO}G)qvJd^3u#F$QsV8G
zb*VKE2niY8)Q<C!>(cUQKmpfpE%mha+LBP>dRUoTEK6MryLd{-5$YcS&tIdZ_WwDO
zibKlK$Z&no@~vhWcJ;kum~h;K@bJ$2zfAU_JuO~}#sT{0kAM+NL)I?Js5Eam29~(Z
zR}*eMh)Q(X)~LVwsM_Pm$@^s)zt0!3{&9I_9&bAf+q8P8Qv{4CT+3}cu6`L&)a(Zb
zhu&+jzM4!@i9-L+;1cI&WlocP!zX2j{J8xk#atdrqUjVx(QrZ;ifEcaYsFm4{KpM%
zcj3P1K3<g1<oYqn65<!$4F_##z;szyPZS~<ZZhOHV5j=aQpoJ_P9_zv)%P0L_<4tA
zOnx+<Q)A}-=rm3;v?qV<!mek2TGn0)!=;i>_n)F2KO2CBcD8%oUNv_Oo0#0}$r^8M
zC-`Si1YXI<qM6(7rkK_!qS|O+%8-q?z^;vNaV2h^@G>}<iHrKIUtw9rv>|R(8!zG2
zG=i#rR}hH_S!1tUAom%f!^-$Y?{avHN)0<{3KfKW$ACr`?z!Tf1$;5a^N+@No6t~f
zXX^itZ7zNzZrq#uz)$jO`8pl83+xchU3m(PcM?bn<*5>q#nkUuh-0fYs%)W7h`Q!*
z_^rDx`8*LAv_A2QwBP^HxL}*Q5Q4gW1G9gWj&)~OpqI#s<r=3tfVI9&Oh;2KCu;0e
z{AKf28#zl+!^6yN)N)CgclT!kv4q`B8W%w+CcJSE9Q};p)_`NI_InWx&8T7S>XRcE
z7oR^S)^90w`Ma$NPTEJt)j0(x?6zGinRbx3=Ix(2R^5uovPQ)nmiOldH3+EY+EP+a
z54G}I&6it~8^S3D>T2xTMN%fmAd9g#!r0fCBL{^EGqZTo$7W=rljh5He<!OOipGpQ
zy99PBzp9`5Y9YE(ZY-je%fA<Y_GP4EJ-^64s7$C#ddZ*O*w!Faf*Nq#cWTV0c$X+D
zfD?hb!01w_%PQJ!Fuz#9Y;W?#D7d;xik6WT3o3h4P%hXS7|xGY$h)}taS9?0*(pG@
zQ&TAuQ|66jaQJDBFJQ*V0oMd$QN}pSAK~oDaRY;7KHT+#88*V`q$IMb{ov!OV2PUr
z>a+7v4(HPoUrI{Rwe=oD2me3HT5-y6g>sZs<#;IL)3bBWI!^=A0w0vSMiBGK3%5)V
zF<9BiWJOKS#kR^GXHIr^j)*7eju1K?wc1>;da7a3jXd-Mq#Jpr9QX{*&dn)^+gV!=
zJ)HCOcDo3oe%kZ(JD4&y>3~GYAg=Tsk^0X^c`k=x`hM%X5)-hMbO3co&mC{MUc;Be
zZg1BYy`N@Fo_x4ra9pd8HA`1dU5Ryn?vZL5(7qXx&nu{JL{BhMTY?5APzO8{0Zi#G
za6Z$-ubdT9?+KYkobHboMg5!-%mlY)ksP(1kj0>HOC7eYGI@~0t&8;fN}w6<Ztg<x
zb)64FLx-`uZmS&6`tlwkX(DzVme(!?R@UZA^0vc(>nCUN^hLn^BH|k`O#R4A2;7(w
zy`08oOhK)6RklEThoB0BK$xB{DxfgO=UGR@>C?p1KtGG%s_k_RPR?9O%!MxgON@Rl
zCnA-Q^0$}tzL$Zpu6zE$gNZZ#XNE1&+2qLx%XWQrzE9YP>p}l5H`7@zaqV%I%!s<F
z&8zVgp?<`73|MHXcKzWoFt>(NYOshKl2ARN*)I0EP4RP~u6wsR7NwAv)wsXF$+yzi
z<&)v47q$2LiLoR|RB!%#Mw}@Ks;n&EqCRcx)k$am-vtIFPiv1}kC1#S86fE>f9JCO
zme9K2LQ5@`xav(|?`bn)&kcsyXRU<ZOmAt@jqN0o2wCcS*F-Eh^JkcI_nED`I>d!b
zy3AsM!(7(Y198w^bv<{U0#F7D^j6=vh?E%Uca00GZ39j{1A(;4d(1$D_R~Ax`}>E^
z=Y!Mp&eO+4fE6!nXsXi(z@{Zq1Ged3Uq_+`EW%lNI^6I1o3bH>7NnrMqRp+d7ygID
za^MqcDjtsax;r}4snWX6etx}R0RBUdIE8dJ%j#;1A}^+Z*6DF{smnED-Yxo_BMn7d
zIN*6=W(q)FGL@SHm*CjJSD^s+KCp5USylDkVWnjtrDCqGDINtHYN6IcL>vq6xH?e2
z5qZ9Sj+rwKd>>ZZ-aMXn4+;F0cfZB(Z+?E;O%0Lv2zpk=r4qLgWWMP)evp3Zgd$D>
zeUlC-t8H@bEBK~r^?~F88F+Ol{!~#t1j*(IxRv*NzL!{wkhtlZ<E(a2*G%069Cu~(
zJJKwdx{%_sgM&J%*}1vrW9kAb!XU?8^1S;ok~)l}B&mT-w({~nm^iY9iPS8ayM&=*
zDDv%i;f;y^ueH$U)dvCK@?)W-roN`{I|>w;HA%fQd9d4V%Z=7L(58U-zqtFL=0q($
zHSQ1oH<!xBhrS}&YCQNaFPL)kf;bYfNx{6xNaBZ)qc~`{e8XmtrP+!SWs*D6>u%|S
zV0^e_<NJM`*~b5><~Td>SHSOM<F&(h-#@(A_YKGVlf_RHXYFJ%DndkpZfB*~ewg+;
zukg$1rczEbF&)}8ZOY^Ztp(5a7vnhf$1);1S=~6nj72KA#Eo<mq~W^ZjeOpwq{L(y
zKl+z`?V>ia;1k~FV=O|2WeQaj&);Y8TV&wkaDaS$V{Qvx6ZV<@c=uluUx-<d7TV#%
z!l;+13==eA;8^5ZRqZDp7L~Z*#%119va;U;1VFzxs2>Xh$3#v$q=d4Nz1Yn+SIyAT
z_4S|1xA^^U;4AiqEcLAD+i^x%Gj>gya{<se0A)R&K1rY7$JFmntk`68DZB+VroDc3
zAEs84-V`xaM2m6=3W~0e3m9zz$~T~*Np(Fnu9~Y5#x@P;9JaXbvTqkyy*-a+;?-h~
zSDq+kmKQ%uXc&4!9e8U_N(enqd%~J?5FhX9G4$}Psb!QePEybHx$^GLDgx%Nkky+@
z<IS184jD@^WX`h0hG)Pv9wWD@C^=#TZ)FWAJ*Ew&MvAvNtB+#h$oYx$e&ZW%U9Q*8
zM*sc>>~oIdNIb2uJOn<kteT69Y_{qvC)uD11|ajE)Tu;UCRS$bYi5X1uTRB3C!4x>
z5UY(b9xSMewD|as&9<!Fc#Qy-o2Q(NE*_q;*|~~0(<Kt4edrN>cM#)zW_fCA3!Lie
zx+WTq4zi}E#)NU9oEN=Wj-5R<j|(-<*N?tiAt<^5Tx9~5FQ`LdmRHZr<=jC6X|p$P
ze#igUg?{w{RE>y)Mz8v;y_|WFC(pymJLX?zk?wVtHhFkIr5**UR=wBZ3Ry?7k*oLU
zMbA=S2CILL?uLU#rq>N#6IMMu@s+^9Y|#uTiRn0;zts|kuMZXq2xHOu5|0~yS5xYq
zOeJo4hcy~45EQ+2PHt4eaqV_|%-qgwtG8%16=oTB4N)TpgQNYv<$^^0yf&7Fo0~JT
z6-&XUyi}$+AS`mm^Wb^6AKcNVTx%B>ghd}F{7$O+I|gQY^GR=RUKXV*vFso*Gq?=e
z?+8T-D>}}r90Aq>Xk??M>HLuGtD7s2z6vm_>6e>5iDS9|pA{l%-q6^;bwL;gaGT?H
z{QW|v8aOIgfYH*HH*DnuS7eb^Lzi15hjZ9v<py5Sfm};mcZzh6r{PE>-%*zZyf1VI
z=AO?E{T>MNOao8J^B%TJZpH(jC;)&EZ4)0^UrdTqyOFLypC8)U$<<gTD3(<K5^pNf
z(*RcIxapPxpY5Y1VSN#JEsr}DV<XWK5eYY7KsRA3&T)%Sh7LAEFMXE+u1j!agtmu8
zMFYcBQ3quSsKkV~XS9v+V*G>Q;$dTs055rVMIo_)ZT!F;Vm35y(K{%9tD6jKdAxf6
zbP{;$DJUkkK3fGM!$Hr^&JKRpby?+D0Ew;FtL^0ZJ%rQ$A)@s%(@;}275i&wj&o1d
zZmyXRKc1`H&GG6Z$x?@N%I1`Mz*!-+S_)r80u`X}^ovHnrmk;qpEy}zm+-!n9RvOU
zbq)3O;9IhZyLZe270$&+81Wr>!0hFkqYN$@ISa4$oxh8}4jvMUMZ99odl)ONJ2ei%
z-qCT<d`UFq+by<pM|T2xrhsk92Lv*tF8Uial2~>fHpHczN~+hKWlFoUreYh1dZ<_J
zY&hA`7-dc*#eY4ky}69<3J}LE@;ByEVfgU(LRHwZKg1$?kz^)(cDGk8+V}|>7!6>q
zu1nshV!!~a$Xy6@<5VS6WLm5#Rb1>y_)ge4(_$my#AAcAM{!zs_qw<icbHNJ#Ci~Y
zTlt$6tUwRQi>wF>O$W?Wxs6%70VbV0Z=yq^02rcCdTB-lvtYW(eBKU>qrc|O!nw$k
zER(mB_uV9aCipWaWuz$_)&nr2ipG-vDN5iq^yyh8)m0&NGn=}4-whTLVoq$ie=STR
zS=l#>n<bRp_^-UzVxYZOkEDLnr<-uM&@%37U*ll%^oRHiS>VN{_WbfH5_{D>jHy1c
zxX7xGaUSXt*1Y1Arrq`E-gS`zELKg3#m$~wi8!k03T9-hvdei~d%$Iqww<;|h#3dB
z+<GkP^I84N{{i4PAQuJ9fZPMl4-&k*&Jra}1w};*D~ULu7u=4Wx9qETR{~yVXH>fE
zc8++uV_-WrU>0cKv?qpSvT`^zuIBQ#xlvRB8f_HMlcn9P)hF@1Tj1Iyk0VR)1pdkn
zMJNXV#c*Z)?h<@-bo60_n>0?)&+VJ8?os8Jw>!FuhPor=R=&PtuqzA{1^$92K7A2!
z>1R8KGTXeng33?;h?6&*12DI5s)PV!_Q>xDMbO9GX5@v)A@=r2gR@Q$a59FZjOIQN
z8j(RnBma?|V;vurvuebO(ZH&!&Fj>>qB#wp8YFQ$hk@jEFE(pSQw1bz6#QBdGy7%D
z=6PoO!)eAD5j0QT=f11%{zc@!8LIYfc%^sCY)bPr+<oD1M2@q6VU%qGz0m&er)M%)
zoP+OO1Ey)~>3+JXP>?n1v_=eOA*Hy_A=%fSeT$Ld`3{5*cdM1)*l=ty@y%F?_yDlI
z@&7tAJG*BhT%Mz>pSX=yL^QRJLpiPF?8a@%pOds3Uhr4LecZ|I4OkKszRUtjjzRSm
z(KqO9n(#B?qKqYTRD|mR+?!LinJCrGzmze~jOlL}zbcpykxvodJoVA-M=)oF<iKQB
zVFKP!0^c-?=q7L?%@aX7_~op$quljlr-MHo+=!mD!wML`*b=;DU@dm7HfkNo7%t6`
z59Y?M`ThnArsPkaFz<&udk)}996&0bh(YT)Va?-(JIspbwd2Y#GPbSK|5Iy-ttvOE
z#gRjTkcc5C6<$Uf-K^zO#}lC?>gDf(R1^<-QRI<qS=HIGqhE-NYJGEOAG{D#z+q`^
zp^-)BaBwVjh46aM3zot{8*o|kz--z0qDUvXX!VM~3pp;L04t8%qI{o+002&MWPfny
zG7wQ;W>`Pgt_Afx70aH+LC-`?%?S#N(~hD6%qkrI{cMs2P&%MZwZ0-lY{-g_<PWLp
zxgNLCPmxJ;nFx}vRRRyd@r!0aNd0NYxj(?u7~dR~kZq9(INmjKn&U{TekiC7xJ3^G
zdknEg2g{aBSvgb|Y6OEn!MC=4@wVF77mSihSm-~gb&{B>h*8p70-MHPN$@j#<5~Wj
z7g$y`Z+UfRFW1>W9q?z@;c$gs?(Q&%&E`;=(ef8U_TW!muBVOA%u1IpD=e%b%0?_M
zoJPz(CFad;YXe&CoSC<9v0JX=BypAActryL>vzm2OWxgH9er^X?W|Uu9P((=;aK{W
zxXoz!p*<f!b){Z+WgsV-Y$f_7pA%j2$KkYj@`S(iNMA}YG`aAMo(pAU<a~hQ^lVn-
zlq*BaLRV6i@Tr5~ppAZl#|p_q&wy^QyegT2Qlx2@Fkt5x_gC)UJeiYyU{=2!3@`BP
zRlx4`bD}NRf93CE%5iO0*{0^j3hh~6yl~R|{#8;VyY0$G0#?l9l-f(kh`6oFMlNpJ
z5^{9jylg&Q`Zy)t*`~bt%lUA>1im3CBvx2mK~Fd0$1laye|5$0uKBUCZ-#H`mP{Z*
z03U%pzFlzP(6*tI9N)q0R8Z$WK0lv5#z^&EIsIWx(+ae@?d2k)MHpF8${eE>zTfe?
zoLM22CrjT=i*raecA(c!Tsms~$S!^_+fPu2&~D!z5ZaI7z-4%ZvexXrYqI~}(Zl<o
zENiTST$5BK<yvjT#dcRvEF-^h2fmN#ty<d-!5&GZ7@tx$9GuT=tA=~`SP07o!Jnw$
zi2QQ7tC}O8oSgHqaf)fpezPa4K|pgF4Sz+voz1gW)l8+L70kE}9Ae1C(pQ_j=Wi@;
zKm->*2ZVakDH+87+w~I%iC;(Y1I2odG(H?B{FO_Rsk!Y_ZC=}FNTInso7`Gw`>8;?
zgHdQIX}F&MZJL)6HAtn?Sb(biTb9;Ii7@KqW&-Ww)sJ3t11fV6Yo;94&qM9EuH9b@
zd*+*p#UYc@ju(Q(tXU6{E6Q);uU^>9xuvW$a8UhbdGtJmmUSwd6&v*Y`<u8@_Mu|%
zE}PizsrRx$JcFk93YNJ1Xir5&rD8$W|2fGrm13^xb7<8M#H7<s?IsX=hT}}P`rPAr
zv*X#-KL-s+l0Y&*0a`&J)pD99za!?vPkd0%@v1Imd*>-=fB$1i3ARU*$58NvBMFfM
zy+wwVZ+<C5f|FcfJ<1>mol?w+F_dENib)wu1~B9jAv1p380_ryRdsPX!yuoa;J6bC
zRmeybSsncR&FKyQ2q#r+OMpL~1unf*VHq(Utml%Y9mi#~r_j6-Imz}i3Bm)lmLT7(
zyiR4_<tcBo5QLG%)V3R{Y+)r43F4R6|Ni@h3syJl+h+5B_vwlSbg3Po!X6B|Ue0Uw
z#tiP*a9T8Pn48-99z1zWfm23Zn@&Fj1;jw`e>DBp`m5&L%G4LkX2;cmXyqMi!+rqv
z9y!77bMDD$$1AJcK03m2POZ?s0t!2;e85=}wX>I~X#<~yr)LD~==}V`&cGA9K+Vbd
ziSKpkpTQ!FR!K6&pLk9wbOM}IIRB{LkXo3Z0K^i&B2P_aT*G*TW`&7qma?e07}1h7
zzqdLa8MiB0J>;G><`l*i7)@h*$tBUD2>t0j%*a6+kF#yoGjBK2J7Y0o;+$+5)?#8s
z$XLPsY38D4Y|HMhC*5QBzSw2-bfG`q@aCpi$!)?gzb|1i70FxypFR-#(;-vE%0rA%
zaj2|($I7lpkIF#%{`j|R*J3F<M3L^MnyyB<R<OpAI(&xx<;tes(1O6mtMu9zXx_u?
z>mnMK{k^14sfPlxqK_9!oYaixqJ!2U?g4DQ>9G}aGRc;4uGn4ANj0?9^c+~HGB)qM
zf0{S+>Pul7X_6HhaCMoz6+-$%!*xVJiQ6msHhcJRRkP*n^2$=fI5=4&i;$>uX_O30
z|I`G|nM1j`QGg8zp~b*p454&5K}1!UD9h(;f#FcEciP@HwFN_7SGV|0E`(B!^cCgN
z2qzpn9%E5O`}cj4lA4-hqvS~$wN@`%-#a~5(S9Y=_PA=fJR&-;Q_<$rO*L)6t!^CQ
z;r=15Tc9Esfwxvksc8@2nA*!y;K~?h0)HI!<dmuF6#EKB**2#qj)|rj&47{Uuv+@Z
z_W8sqg-YxGq`=z)(#AsjHt#8O#Lc*K)aQCciwURZx$^yNlSf8vAdu+WBf_v9C(f-4
z>$K(-7g0g4uQu;-Plr?ph5F2bBF$B{M4zYKudbVP$CS7h>N-<+$d5R)H~`a(=_V|S
z47M){#d*{vB?M?8l93lRDhnzpF(#??Kvt%s{v6qDca4a>y*-bRU@<dGrAKm1^ev6G
z{eV}Tv$$h0P0IfOu|Q70g86)L;XuMP61{tg^c;H}dMpdk6KDk=2%H|{KE36|+aN?A
zQRxF}EEO|^axKGMcdt#1@gnEaI@5hPmeQs4+<TKYky2lOx~8)9c=x$SQ`v>IFfSLc
zJGDZ6ZRlQe5fIr3AVm(N0tQ797REJbO<}w_2PKGdWCoNCF%ZT_s&8-Si1X3VO6SzB
zCeLy*WdMiHd}bRT1ExSPKm?U#&E~#N6#2kmOocW!XeIt(*CGtY2Cc&GWy%}26@rLG
zX@iIdHS6=glVzE&T9xERVGJl)pipdY8y_jHK;flbrD}%b93|kK#hBc0ndux_DaPA5
zN=yKa0>R;INYU<u$)PZt&&ZVK%HA%@IIisF)MVtQz^GOV)=HzaVo@m%7)qmM019mF
zFgm1E&`{M|%*)E}IU@MYyD~wEU@J#iS8R=jSn+E-3XQem_iDlK<?XDcs!BvuW8TY6
z&e?p96aT#|Yn(V-oNYSPUwYG-wH|nFJws4jz2-G9>m+DW3PtI+=rwROdBAM(h~a2}
z87X!o!&yU7K$a_@V6mvtW{A>;<KqcuXUDwt<^+>v)U|I&Os|t=8AgR|1&ksq3Wh~S
zZZxhI&SrCVcDE7dL8-Qu>2yk-XDF3XRV8J)2+!8|oMn&0`8ROZK1O(vn%A1UHq`M!
z%wHd*zTSHT&a<}f%DatPRZ_8_!cv9=9&776@1AR@mq8|scHN_nk>l$a>^cBhGD)sc
zAq09;nWV*1Z8oiQlSynHZ8GUC-Dmk$AC3m1l%dGmV0aVE2*zaI!%BHOwN}0+qckQ9
z=}mBm@ovDGF;vc>lp+s{?=EB<GfH{aOy@94`G7_3nS^p6A=#Qjy!S@j16;h>$yrBk
zvc_0v!rQAeZ%VOs?LD(lWLjfU=&(ynri5B7M&}s322j<N_ZZU(Z46ez#8wLHEK9N0
zq5y}pe&OATm)w1z8i=i^77L0zC(}^c3ctD}*t)`obR#vjH^LfCZ7tSWYFCq)+yjXL
zw4k*h&d1^mauZytH42BTy-i;$MN#BFs81l%8Dc%-&2ob=p5ZzNYBWYL-~^GHVUc5H
zjNm9t?gR1^WO)(xLGh{6ZS8Y}`7|dE345y;tq~`FEeWr|h!b+-!O;Z=0vC3mbIxLv
z0%a)6n%TT!KCi;smt%2ri@J7xzn9VnntF0*tpk8LocO&#oLH*GoT(LD?E_cKxuGm8
z&SpLxUtO0})f{o&Q!u837iTZ*#=G>6to^b1g2yQOWWO(PpC<*1b2XebOh0NMs1V>r
z#e$kLn2^)!HflX<06J3N8pKK^-L;^eIv$65&0c-$CFj_$(xxqa%CdWIFPJ2Va8CDD
znzM1dHqgdwyz_Fwd{LsbL7~W<j}S*WYC`TOe5AS23MXKlKvjDWCJl7}j!Y?(_Tb^8
zeRW9w=UrZ%QVnyiL$_3nKo{>7CRV)0VqyfPL6q|Laq;Dh_TD(Qb`-^sO!;$KK&%f)
z6lEIMVkZujIr4#zQz?{ZxZb9zvEuh@R)P;ytZPL1#4HHj0|<pO28|~C?-%Ck8e3Vm
zhC_6gV{*sNxM)&-c$1Pv=Ncm#=fKGTZM>J4LZd}d)iss1DDBN~Dzrz$gSNvvpW2YL
zQfuEfqeG-Q-c{8(L1h|STc7tTxLN!5N|SQFb~ReOvx9ZsuhMIrs*&Jq?VTJ%ff$7n
zhjorj7dUa@8T)5qZ2;D+2+yWQ=f(>U4hU-$u69@(G)?7qQPgNJU_{NvSr2~NYopdV
zth1D5MP>c<W$8CGtE(F4G*wkIn=kyjQe8I-?a@mvCJmiVPjOGWN6I=$`3V)8O-y@}
zbie7g8=3Ms$IdO&y|=8MN}2zh5rU({Q8OU&1IM!TzIOQ3>bVuBnR4|F8bM#(=TV9S
zz05i(uyl`}cK6l2^cm{X-D63)6f{M(nTuh0JX6zE7lfV!L^f(5BEq7sF@_M+5u{r2
zi_o4&2vx1fvph7xGmqMN_gzQWQZ+E%+-Qm{LuutT4sC)UFH}m%jAs~L$xt8~Wppzr
zm2;@T6hk5ul=XW@I1y^4!fu4%+$=1dd**L4&rs{yXUC0<*okE}_xr2+Aam_C8u5%a
zH-76?YpqXX5R&gXaa2|1nY?&=dS2vY+VAFUOb()adU@j)caJ9b{a8AmTg=xaKA%%y
zUUgZ8++(_7a!Npp#<?2j97_Cxdx_VY{3crgR5D+y)rPwEvR<Y$A3-mMs;tlohD8Br
zT*zA{_}pbKjES=jV>HhC=WTT6*ReD{{eIX8tgbC4n88E|d7g)PQTyJB@}@=sgP?Wh
zvlwfy8Hy-08mqjso)t%3``DYZE`#>s$3`J!Vee}vKK{nwC^GFe9dc~$1p{kytbj1U
zTA$ItihqvcJh0gy;3$EgzMen66xPQO9s@K!5M6*7B}FfM{IS4eou<<{-UYx?T}NSp
z;mV@aNfRx-LXDI|lTEfyYF@e5I@C=}R8RV-E(EB;N_OGKytjs1_qHiPsz$Mi;3Ugc
zrQc2bo&ed>xXEf0`kwu!G1>?qI!J<6EY3NMnDA12S5l)xq<WARCHOK5f^pt7kY^cU
z)Y7I+wKxiuV-a+eaPeA5C4#~sR$)sUwy#uZO_O2Qm<(+cRbaxdt~?0oV4ez&%ZtSV
zqctv=Pcn21gxX8MnO0bDhzU=~_g7ifEQpt7iOxLGX=Qv|L)eqsXSXyMF=%7Zng9uq
z=Q-PZyPP~c;QK%P5&zo1@vk$lXB-{fBQtr>Y!q0SojDHgpK#^M9#{5vnUts0l}{p>
zWf>^z8GGrEl~sjvj?rMmAkRWfipGiyyEOWESHb5OTa*i22&mIq;lz54FmCM&h_gP6
zKrq2+qbQwibD&A(r5dZKl~3od!s5TzqzvK&lPN?Q>cyhXWT0rYp)6=%we#jfrG1X9
zsHQss!gKP^!!<9W(#C73*7wnaOS09lb2uE84WU}fo4Q1N&|rp7u~Jsv_1RcIx82$V
zlN}0goYH|Y^t|yGlKwl)WF4=298Kh+$NCsuU;7?g3t3kt1(PsdjxS<)8pYdeF=1~0
z?}+>#L3WzLacGD7n~e2uGUH3k?Vm=eylG#MVIPmWpm!b@?ICIYxqz0m;o)|*n69?4
z^qHW;(y>mtE-<XrleVRA?2>k=bA*(Q08-aBtY>?3hZh2aBBdPGhP}eIfQ#GRny?V*
zgS2$e0E*n3pd21ps=D^TQ(F*WkQFEcDy~0hZ^H6E{tD{^6(m<J4r9GPzgC(|`Sl9t
zycFCpAF<%PG^<UHh)+)r3M~$A(Nfed7=aWh?c*`9UZXLYA<r_jtG)S5@BpI}S?H?`
zwsRFkj>`HhIZiaY`@76eC%pUiJG}kI+r0dlPxI)NL#DGS<;*fUttb~2_uo6_(b0_E
ztu4OzxsNfrl9LaOU!-;x(V%fC;~!aGXq5O2!gW=lwfGnv6e5LxxwLOXUHi1_S?<?H
zGGlOLpu*07!Q2zI9t&7ok!cT1-l0HYj7AHMmLdUE>M9r=m3OUOlvOaLd3ddBOXaEt
zV5_<&EJNWC%G$Ch7ZigWMOX)O;DQpav_^?{P^iL+zF)U+SnIG7jyXbG2^y(4t2yf^
z%bMDGjR>H&f~#GK`w$Qz!P1LSN3x5wQL;S`0sduB7w~%8@wh;>0!r)Zi09lh)ciS&
z{(q1$VyJ$bQvcV~<^K`{bauqhyuz0HO?LS;GIfVajd`Shg%kI`L*)65?Z2ohnM&xJ
z`qwq_tN_Ck-hbtO9nGS*p!EN2z$npr&NWT-olF4eybKB(0mKPmb0rm0Pzrfw5OID%
z-T9O!A$1}_2?$;{IZ8oQTci?lT?C2UOO+ah%8Zv*YrhbmDHB9`9c<{%du54FR_Lry
zg<TJgQXx$|Ov8GHC?Pk7+TypsHGXt5lOfhJFDr~zSnIO|&li3>T#Sz@9~1*J?H7<S
zs);YKVd2---aVJHMrR?zt94i>6vmL{87jIoo4~AxLq^+MWUA(!S6}6&kAIqve&q9f
z|NDQy-8+xC{l+00!@R1Q&KKN$aER0uAAk8dZoGJ<vGF_c3&&b(L}|o1lz1uLi6E6_
zUR9w?7+)LAe99*`tV3c<rTkXJB|aSsBGkA5K0dI{BDlH=@fSgJtm}|*H$>e#hf-ei
z@?KuS{WsGn>le&RSK+L~;n%?s5Eu15MOjuXD!;y?Ymd-cc}+$_oKPJ++e}E^?!CeN
z;(A@zewzUdHI}lnWZDNAE~<*Ms>A**KGCK^`?(uZ$+*&IBF=Ktq?|aHmh1qX<3A(X
z$c)b^zxMjaJhrI~7pY_EyaSP6!K$yK)k6xl$@Sl3qP|4!zR0fmyX=|&62&PNOU*Te
zIb-!}lsaSP{?91c?z!WJ#zg8c;ERyzd|vp1HTR7_*D3+2aMIB_qIR}YZCXw(pG!?v
z-M?;I({_Cz4J6hgHHh&xdh5K&0_R^`!5~(=*ORNMD^OUh4vX7P$coHwG^KT5=FX>g
zw`K6!Qr_h=q>ZnwP+1q4rec_9J~OMU8JdDBL~&c|sBP`z4aP&10$W)`thd9fHnQ9~
z3*udIv#bdH3t*@PrL_ko@t#~xU^E!8m@lX+i!#yk%kP*-?Wn9H&x~J7P<X~$`TaIj
zrl_qK1;#sLbUmj$n{fZ#H+bO_FOj<)P9NT3oDG@J7P!h%)sAUt`QDp%xcBIUKV^Q7
z-KPc~K&*JD1>=EIoS;Qxa*dPP$8@*=)>$4TZ3(8E%1iZD{KB-hL3+33d4>?31N_>8
zD6ctMYgtrP0C4TS(3E%O_F9(@##A8^Un{SjRMi3q>m<qpfhf&%UIw#|_XrC<)i|lC
z7B$XgO`a|5e9km)q7W+Q5bI+-9FD525foXbv9&i_c|ehn#zn%8m(FWO!Q;w__1<xn
zrL3*5#|aq4>146-by#}CdS1K_OkK_3@h{oP#GeE>`Xuegt)~gHO90O+jUOfFPcY=y
z&=jCf7!Uqa=JN9l^gpDe1PZ*-QXvcwG`aexRD1)+pWAr<btFkUR?a2$rBb+b51vE4
zb@z?q2BD=Hve;f(oq9{!ZbuFN|F!oo!L}^vofr7~+&yBgwIAnsA9*vcZc3R_N-C*T
zqDtyQP#D4hqly6(V1{8mY{Td_>mIcAj1i2OVPHH4Fu@prVYFE{VAfP)&C^OCRZ$Y9
zQYxjCl&O5*y!W2>-g~VT5BHD3=N=I&R;+!_J*A#)zjvRtRy;gBJUqVl_y3NVj7%8E
zgqgs?vAI={ldYHA?F>W{F<Ho`_qkGCu@^gwpb<%QT_WwHMMpVL@WH}NGvRevdvMvr
zWJs<6R^2Ye@m3rTFD5yo(8r$MLFEI^c|;XgJG!WtJQ?VsLenCs6{!l2u4{(}9^)RI
zaB+Typ|x3rQx$siE5vB9Bd#|adyZymTXW#nA&?Tdx<WKj*DF#=)LAofBBT^^6k4RN
z%sE|b<W(K$)l4#<eE(0m_uzv4%bxA4{bW9&avtv-*KJF+-BHYdlE{9SY*tKWD%04p
z+bdl(sFo^JcyAf=#XFNU;vIylA|-?8j|nPo33o307`*pLPV6iuD`O-%Hcriin4;1*
z9WI#W(Z`4+16FiiatOCI3j<cf3tl`4rR&Yq)VGP)b%=OFbWT{JVoEuZE(bU48jJ3#
z0$mfix-!5_i?~V#+u!vhY+jl+n#nF2#+jM0G`}#KVN7z7v;>+g-$;S0tNp;hZf&PD
zkohsM+4ZJ`_|*Ob;&MwLZ@TaE9=o{jltwa<WCy8MA?nW~^50~EXU$svh?V>Skwi}p
zQ725KVnMmV$?u{1>u+VgdZYW8-P0nGeqEc5WAW4pCwQ$wJJTv|z1}G;rTji|A7=L%
zzpsJtRt!YRIR}o)3!0;=%!FZyX=ci@w<&}wCwLW2B@t?q>Mg_A<uFenRX&GKT2xd_
znjQ6#v(1L}%F)MCy2ZF<pAx<@1|kYB3O)#R9q76q3pCCVLt;!JC8e$c>y;%wwrxk(
zSxBr8g0F(*KA?a_HSP90i)6}X4ziT%oJaeJbIQ7|sOptnBPQC^<GeF7R#*ejbqP|g
zi5x0CH6D2W*)zWP$KT??!wc$F;OgZK-U+o>i0#TbUiBT9mk;>DhwrmbSC(Wbo~Rb_
z6{<krn@KKI6`QaDHHh`Tx3KG~TH%2<_5;E6IhsrmTq<TDBcd_;7Q9%X0U~s%x8rjz
zk7UfLPzQXlYLfeyaw1=at13zz^_UV}pV+q@eKa_cecM|?UuWi}uCpkp*hdg6gxuxw
z!l*N(ff&JAKtqa3yKnQFFz0#K^@9Ul$v%;o2FGg3Oe5wssVE*}ZaNVwOM$7+)oytS
zUvtRf2}g`Ym-{-EFYzYY#DQczKSyrwuV@bQW2R=11EQ<P{vaV2H*)3NpCR#pz$*mF
zVISmo(%419kqCLbk;p@ok4|GWonm53Z8|_b+<bs<_8PO-qEb2|_^12{N7uTgP3LsR
zFOE7machg?U0PcJHDkf7rFMHFp@MU2a)3&9Kx;N77^YS1&8}{>vdz)1V)o@ODRro_
zx*9D+*9mQmncUmRx~@aTGNRkom`#eN<@8NYkbn=uuCWBLI@EX(%P0m`Rc&_e9+I~(
zR@49+gEPX^_8nax@!5~yyg7x}KA_^Q-iesGtb~k4O8payN=nHf$Gcn!5oe*i;<MAg
z6=Li%)2-;WGKn_{5;i#3L2x{K`GVEihRgfs+}vy-<$6^valQ_m-G9I@{nCeQ`x_J;
zU=|E@)}C?*p3P=$j%~@c*pq1?y(%dgVb$1YZP7CI3v+PJ%ykAilZlBFnwZ${T1)z@
zy@e|GEvrzONh<pyd^S1O!K2RnVa>iFCfGL}*Ec(w&NKiYjIsB9%7wx-`<IOH`%n*w
ze3cI<=3lATft6Hd&P<APP_HTzUZVYuiY1FWGjk@%60Sh-lC=xTJ`sDyUOas?M|`Ps
z^SRrbm@^ja44+HuSRjO_B;-?lEi#}t>GMQiOG$!D#0w^x#=>~VsrhM={uL_uFG-ns
zC|I5qI^;7b6^{2w`c*pqI;#KhYn}hMZPTTlvw2)%2J=qdL`yrVz33Z^Z=tiL=Qh1W
zk%Q!5o?qWij{U7P5CLB~oC_vJd!b7f5fv&&@K*1l>w8uiLc)lw8o@&Idt+vhn4_(n
z^8|0V<d96LR%YrFZ>33-Wmu;a%_NearqyupR&&HTy4c|btEJxdR<<@c%fu$-d<<x#
z8lm-O2k$$jk4i|wb!@1dN#<?e@$zQRlV?|i;90FJR&_;PdCoVrB|0g@E^)Knqu#0$
zUYxIToeW3YCBmwr@-@$IuF&4hS(S5Cs|}lr4eD}KSydsSqHXqwD1Yf6{0p2tykOV9
zU|kDc*W^4AK{0<tN{PNV!W_J(auuM&tnQ>_W~(Y#_V_v!Q<g#kaaIu$$tE*(NNtf^
znheNMSDqLXWFoz(D`K*6X78LOhpM3sB!Q}`(LS+nI$mDydHL#wZPS_APaK=IL9d$}
zVZ`5}_w3$K6G+x|MaT@i%TOzQG=}Aznwbg<-S)NR$gI>8`(&Dk_pHTm&Pcf=fjV>4
zHyvlI6_z<KxX#R}5gDLb-bMqtO=>(zIF>gZ?RcZUIVLF6mwu5x&Ip&>cI?79J{OTZ
zF~0Gd0oAYK1;h~B(6XL04KWt~7g(qN3p@SugBnw`Az8SmQ{#M-eQ(^}L~D2(Gv^yD
z!$n`stlzoz_onjB{QN~TVQ!~0(G{+*cWlni2;T6kVl;bpB@wDXN6cD>L8g}(4eG#|
zf51CO2*SFmE$r3h@ZRh%Nl4B}HdD!_-ay7@N-3fdqAE?(AkKlt4881Vx{lQ<5S&4d
zv?;NDwIjyJ=4?&n4Vx<_WmQ*}&=}2GoqA}y8(a`J>lIyR_V#@n4UyMb(wT@Qx0$eP
zHd%47Uax5S4(~iZIYZFx6BlO}M&NwF`<gyKVpV0mxumKqx|I0rv*&#L<4<X0!mn0*
z?dQMBKlRuDI@QHFKltHyxPJaApZ)l|w9S^NmNZ8+ZNq!(&xsr0LE6~St4Z*|XCj<U
z05K&CQpnl<Rb^VCuUvL;n~y6g2+BHmLr%v0J6-DOJJTdA`7<XfCL#5aeY?-NXRxY0
z?>@dqibC7jGpmDX6)Kh$ZW-bh&TIY(rR$9`c71D~(e|0P1Xku$s9lwv<>oumWB~;w
z^f%_Hw4xcKR$r!VEo|HyKc%F7SRPBs<^LRQ{^s6dz@8KC>5excWN)HXEcJKt`z@08
z>$H;L_bmfIIbJ!uo{*z9b%ogpq9vB1#KC+eG`vsb>-6%kBl4@jUj=@V1M(%@n!$d=
z;6CT$zU}^Ad+_JI)*D>l==m&MV`&Z+hjQY2M_SlII}PV9Vl~S}@se?@{3Jv)CvhQG
zO|TQgwW{jcOa!SX^*wmd6bb9foZ`J{0}8~{lN}B??=7>rS3^uK$g18UeNR}eb0ram
z9DvQL%1|4{M&Fsg0%th|0<rHv5>3;Rg2z>X+BvA8s+<8T`hfSINB7SMbXG7z*~M;n
zKz$#Lf!2XYNA2re0numZ5zS6mrE9I`d8d|x;hZ4e%q=R)`T3eZ_vgOG+1Vwl^9|?s
zHvI78A9C;Edwl7`pX2^7{DXY&TYrz={hj~8YIwMi;atpT(6lYP)R=E1CAQ5T7aZrC
zRW3ATm{?#=;<z_<61-4{%ADW%jZVpeC48vx&RK{h22WD#L=JQ2o0enJe$qGq8G3J+
zT47@-$f;BMl}9BKdo?uNjKS4)9euR0=)Si+j=nSP$r1qpb>`E^OvJ*aBcfu3z<Og8
zeY9NW;LWHWlc5gBm=ho|;i5|l8tDYY$;fnc;`_Kw)H?Bo9Z6}6BH{A6OF!TA^R>sw
z^1iq2`=DQ){XWr7F&-mK!&`<U<d6^Zk5o8XI(idP>3M?iZT9lF@cegp{tLYRYl!?c
zz)%@Z-S^2M-a<5A*N%>PW6}N%+uJa=$u-_Gwqx=9*D;Y>oahFcxLH}fY7s19&`3+h
zr?PZh^{}Y}h?%HVmEbCTT_HiJyg5K)@q?(rS!gmFbd)g@nG9|hvZ$zFW(_qMbeAPy
z2h`rPjfq{`;V{UNYVaBKphA^lJ$ey~)=A1{Qxk%aD!~gheZ+?_fLMtL`@Uy?ZH1re
zDnL^9T}!AcM0&h<&Q}$QNFO`)tucvCd)Dh3?KP|Anw{waF_C)9d05q+&3euK%QJRe
z&$CZI<Fk)H=JMi_U;Fic5_$9<U;de2<;BbAeCrSX4v3Os1a*k3Sk)eJ25i{&hF#S)
zJ)Ji{Mb~xoDN(P?d0&NK%r+|9#>%=Dz(~u&c4!xgE)8LqXPY%nt$3Jb^JCvdw!0mh
zDzI7AR3QUSidf0tr0ko97z-pxw8X=t^u5q^iG5=bFkNqvSAC51eKgaS7*PR7*SBU$
z%L0of9~yJioEuY2rm3|sUULU2NWoGAqL~-15@d$pk+F*5;M;ps6V~hAs#C=2HB94;
zn;f6}){JI)>_We%j+-B|ILs6GJN~xpG0zrH%$560K&`QG?Uo(FGt6W9m_!A93jCWm
zeiF~mzvddZ|2`V?=RCd~bW*xsh{RetP8NQ=m6EjJmwD5E2DwK@Kg#jo4%@9x^seh@
zdW(k$-a$%KIU*@$lZ6w<YHj#WDQ1kSWQ3&AWagb!h7b{}YwpFEgJ7^0aY%3zeC5#;
zjWKDWWT6||8pIcTs54AEXdl6wc|w!sI_ts4>`YP)!BcyOM!WK^Z7qaVG61AlJ&($p
zNsJg-1#bwpSu3g56@AxHvYGe0J>t!YzY4VhFP3>wmuqksfTNVG^_YlVk5pDA@4Mgq
z0o$hGy>}lYU5~ac&wl(vHW&BUT)x9k{miehy?)7yr$6HQ`XznWb9J*PYGSpjz=5bm
zl*I_UeMemfTvb`vY)Z7z>Ty(6W!i*TSY+iLUF?auCVLxuHk&&8pOmia5jB5=ctbAg
z`beKFmZ^z7sgD_&Unr4KE^A9f^F8#vL6Im1__*IUB*TS4#e5@G<>}}N-eNwS^Mu-f
zB{f>!Pl{v^Oa)xVC$sGNzUy(xm_w&l(;er~*pHd>X!f_17ujG;1OuD=T#4v!s%b17
za_aCmVFs@~?sG7gTbYGmCZVB{<MQ~A@8vYPlHtuU^Aw_p3PFC1kOob}dhi&^nZvQp
z%qC^GdWFg-B>n@O{-1%b;8-2`a^92~9NbU&af^fdw)0Dy662HY@579CvMV{>E=$L~
zrcLMa?KBV|1ap?B7}<5bg|>=NfrVgZD3PebK3RCEWRM^gZ;Lq!ZJTg@lw`$=Ed?)`
zNhltnPm$P0e5o1YL(V!@`o1HPs4AZ`jg7f@AGkPogLz9TM_n^S{WQ@e+_QC!ieWZ&
zjn(@084_GYaK54x+zrm5wV}#-#TbAX6n$9LbZting0I)v<dw~Gy`cd2z2SDn94bqO
zM6q1J3{TSS3H6#3q1&}I*Ei62T<!P#-rxBxe*E$G`O;5+jW2!l)BMH1{EzeU$;W)_
zPkxUd{@`0Yr~*+9<0_$$l$9wu^nDuOU<hKxz}lXs=|JNE_Sj_@52uO?0ToA7WxLz6
zUe&0=)z0!sR&@nQ@!s-B)J68&J#93<g6nOWiAYX7bacHzn(THBs)7XbJA}&d@WCao
zuD6_Do`Ia@?WV!BP0xPcku(u1D{b7zgeDU-Vl)^MQAiG2HMgq^MxauJ+SgDSN&;#Q
zSu=|XvF|KuEr*cz1qwxFK$m{AH*E4DN1NZ6_`l(0J|7c-MPhn%%#r0z507`#?IY<q
zWFWW6yE^dMa@+v}K@qv4=9`G-k_%XroCWlDtXLDs$*!3HGglcR@;{N}e+B*lg8nP4
z^p_4yi85pGON?(yVCRgOh2!UUdPHFLc>l@vri<V;ZRt8XV}NhNs}}EX)^3f$icx7B
z%i`TNdqN1fougg{)^(Otqtf*jbXOeC-g_je<%noR#UnYLv7~p17}ium2~{xsDQCo6
zlB4NbnhXoCiouN#=whVpcLZm3*b~BfRZ&;Q3^40O24OVQ(Y^CCM3koMhor#4B$U8c
zfxhhr^I2M3LZa36Fw7~RAyxXQxZo|YzzMNLc!`DdDhZNgjH#HtwMy^>@XwhOx{uJd
zgm%xm@7X_nZfL%%il*;)stF;XZ9~`ftU}`bN0)r>yS-@un&>s<&|Y&iSI%WE%O)X-
z!PKup#ctox_Xg`y%za)2Upe+|laoaQXhhV)oUgYHPo7<KwqEn_!8xmYH5wC1OcLCj
zo#mS19XHz@XX`Uo)rz{V>AJ{j6*xbyjj479-k72#o}Rz^z?}G)cBU!Ov^~LDndF!v
zU2l0dm3Z)woS+rvP)BprTQn>)RqC8y)W>9zTi&9+dbJ3)+I#kGN7Gnsk*-r{O@kF@
z0@LRT#9Df@PEnH+9gIXvXM0|M$=jI7{J7h-l?mgJ<8hDviluQJ@8f7+ffv=u9Tz?0
zGXmcSd`5xrIPt92^FTLvF;>z@$Cj3BI(lLj3Y5f4RQ?8{KMV3phs0lw$2A?dAhX{~
z$Bn<1`Z~Jr1!3YyQ~fj6JQnAEvT+et>y)OVhr@jxkGnOIFeSrU3QjmX-&i)WT6m=7
zoPu60iYH`8aupoT2b_3P-wb||lp-qT6qUZWqDZSSU^_WqQ8}^b87Fx4xy)?NlJ=Ik
z78^5>Kq6G0zPI8`TGdolg_WB!XLb#w-YW5Vft9aGk_ckaSk*e1){sn^SBpEPt^!Uh
zk*&irql*~}8H*5!%1|cOK#9$Qt4=dkl@56m!TUiPi+#_2yESgAX1+<WCw0-Hwa(Ub
zyFE|7^L?&={27nlxzD2qXFR)pmb1nknw-svH=sc?&m>}JCLy0O%c@YJvEytNaGMHM
zfmD!ig%E_5m}x6UC8}ux-M(jgy(jgi&24ur-aBHK%lclPv)%3J+m?D=akg0_qO2-U
z<t<UM3KdA5i%c1omlrbHujEkO4h8o6j#iTi19hN}JzW<G=Vw$olCJGKLLG2;1Idit
zph2W$IX+fF(V*b#+SXUHFg7c{yv-FWQ8sIjb0(Ce)bq(F{te%4U7!w!(z|`Ry|-hj
ze7^to&%^I>)W_-mboP5rSf|G?_jx<-oHHO<_?|&c`2@$S;VM3PF5||+;t_k_oU$fR
zQ<La<#VevjtT=$D;vRutvdg;pcn|0Jkl{_fFK8)H9^ok_xy)FOX5=&%TW)W#2ye5V
z9+|D=aNpARLYoTHJ?Q_I@!o2}^1*R_wjp&L)v7|pByfYoNJwy0l_9J;F|&ecC5c4D
z`wTH2BPm(bma3S!sxPHyV@|e9tg97WG^eXLX9b~x5WJ)7BCZOoYtyTpcl3QnaQ3|l
zKASrt>uQx_{52C7C5a<sY!Ru%Z0#xHgTtxW@Xt1tA;l)m^-9d-7NfG=?!ZAED$Xu0
zNL_1vd8=a~7~Dv5rOrp7x!!?Vxa?|G;k_8{RWwKum!DRQ87xuY?0n7pKlw$z_3`&;
zn~vawSI@8b><6E4{|gV<tTr^Am5FWImMWMgk<_rGytg{*eQz$wo)V>G(s`(y)yNPF
zT!>vR^d&+aJlnQKoa5r+jI;AK8g2iJt1GtGTLj08S6kYBVeD4eE2c;rBUSBb`o<V|
zWqzuN)pAc7aiOB^Iut`aPSG6r>$>9dd_&S`nn>7?RM|B>ZQIf{#<&pZdxPn(E63)1
z#b&e0Nt21~u0<sy?dD|HlnevROk>JsvtqrjGMtGJ)T1h_RyA*=f(VR&4~3Z=pSMJ)
zm)qehf0~$&B<nV5_@u^>$8<cNkz~x7$NU=e`&ecq%a(UJItDzSB3M~z6kM{pH+B>G
zJhgXm!-{wKCs+}9%KwLsZ&49hqaaW6y&2Of$k<2MUD6OH<2kw~Ih?1%`<>mF9{l1o
zCzz@t^SQSRE}p4}XU)4>Vk+||F=Y<(uIHnI0*G|F-D<+pr0j0?=BEgfiIq}Um7&dM
z&0}2^w9{nK+{u#K9By!OrfAN1)X;7EE)tW4pf+uXchJ{LoAI+k&bk&Hl7RyympWCH
z)OL9?3LJPx<vn$9rYU)WJ|dc^LXPc7Lf1q#z9g(gV(e+!*32$_%J~77t)5&Wh{2(I
z=WxO0lDgJkT~$yEUF>PP*0yWHQEJ8uI7eOA1RvPmZ0Y-kDg^d2t`ZSr#I^H$`D;H7
zkIz}%oDtM<^>WW||Lz~Ny=r;%?mfQr)9<mqtj+mt$rLobF)h`e)oR6l*HEulhO}G4
zT2o@zSRt*=*@m`n+3(v7agt4PB5WHAKW_FdH~WUzN9s_q@Aj<MYc_SwYPB-t02Iwh
zd<g7z4ZHoG)oRUpU30y^V!vxal=Jg5ghaRR*zYZ8C`JQ3^uhDagLBR{9_K4wT<wV|
zvDpNaL`p(cJGx$I8;i_pL}`1^I>YjpI`vK48=Ora>7vp!oz;~0r6!Rd@?yLOuD4rO
ztBsl7LbdcJoOs<zB1s|0^Vjm&7jRl)JtkhaN%+T~OU!5KywlfNW)7zr;Om6yxk+vu
z=agu6)cihHXgXNYn5f{aBtzkZD#jXG&S>P9IpdcJ`~zI^@6z(O=;_(Wzm=K8Kc5}(
zI+){L$0$zS*Xi#|q*@lue}_F34zQrXYMGzJso5)xWl`&xFuDWHVqvJd@HuCUWuKMC
zJ1tb*4(Fc1=t4DCyQ{>rs#o-VZ*?EECW*_@IT7cx7O<{O14s$M2vipnRc-auQ#4af
z)AYP}am~f~hO_mWi?bD8gr;e!@C3<qCDbCA#Cx_kJF|J$6;&1RO3ri__WQ_o_sZZ*
za5<#dT<rU%W4~*VP!Tk7c2+}c33)L_u9_}O#IULZ^{S#|oW+>KbhFtcILEH-tTu?l
zqQE4e5%z5dsW*R#II7CgcdeOwYD<Kx>YA=KblMO+>s8Ie`}eqb?_Iw8;>VUP?uGhc
z#k*g5$alZ>8Q=NgXMFtGGk*5xf0FmV{0Mc%u#>l8iUSvf^UF>4&-A$lhqC|%XXYu?
zYME@-E4nsuwl;>hYj#|2)^x4qcU0c7%FNb#V}=Q3wW-)_RtSmxzCoR#>pocD=lW{L
z)8{w%&`{SVbiDu0LlYp9`Jr58IUuiYuIZw3SzmC|^jz=vv~A1PO+(v89^N}Q*!3KK
z?GRSgn&=F+-6fASj(yt^d*y7iW-Tka)SAiF1*1n8>`h%;*<pj*N$mDJy4ET-uIm+l
z@~t0zj%@UlU%GVK4t@0Er6Z#CrVQg0(<sbqiP*hv+ih*B-#6*&n3>M~3{#2w2h@Cn
z136a|=z#zts6Y~-7p3Jr67P|?q2?<*=3nK8|0g&6zkrW%{2K^QQ2vhy=M+Gc<uM%J
z-{F0ge@|#82m6jSxZ^W)wt1%AO|}_-8ON!U{+DKWI+i0{R+gWkPL9=st@c@9Rck7&
zLeO%%)d$k`X8R7EN42JRR+IXA+fc7+%iP{J?CRRWV!da*3Rc6R-(?J>gm)E{hxMj5
z7;%Ku_4wd8+f=Nphv3A*Z)>5dYFd;&^~4rJs94vPL4FHpN^JKXH@g-uiL=$(lI99C
zj|S;|z1<Vl2usr#wD-M>3xg?dI{LQDW;I27CB+`+0^41K4~~s<R*n|)tMqMy^MTa2
z^uZx&W;u_rYg)G3D^~T2rrksGxU^!kUb9)Pa`>-l8xF;LPYfQHEYD@Nu6Xp`Bfjuc
zU*!jX?>R{<@=Sqy@0{_Gd(8L0{V7+^TK?Ye{s|vly~khpOFu^!8+PrE>r#@pGn}ic
zsxs_JZ}rR7YNZ4x)WH#5u*#B^8~hO|qY8wT$9ZAxEifS_MV+v38oH|+>N>EgYg8;(
zNX4_;_q=@ZoR=>fe*DQRny%yF{d3-X_a4$)g-NqnoAl*y1g$N2qO$^5m0#ighVzGx
zuI)KnZ&<I^87|&@YeAsv8=MPNRc#0Z-U^wOQo~luBocdLgfT{%zGc@oyn1!b&2Eni
zo@dXWT6A6TwC$e0?@qoB$FE;8VWq_v^zen8f46p;nVcqCQ)0OEIVYI6F_GIC)aM=7
zY3*d$C!r`6zlY~3nq^#z`bwa^g<y-z%mccCj5SEccYw!S@-I{K754l`DBr{L?^Ee_
z==r}9<$dh=Of|sUBFd(}Pn+Gw?Q_jfXM5j77&+`yVPa;Rv${;H4)#p?5nbqwgB}i9
z-uUmxs>*!X8V>|~4z*uuw}<h1Pvrw$*Rd0LaB)t(S<&?!s)@~JjT5Y<gjakOsGP%x
zz<#&GC98cNyfa|Cq%`}+@R<6Z$_1;T5fi~X?%lhf-S1hiDjqz%N7FVq@9El}E=4ZR
z&$2_@v#!_dnti@=i+bvMgUPsec7fB5^Rt@mZpXumOXMD0-|X3J)&vpuyB(^M;Xgce
z2zXb~^ga8gqiGsmU2mzX3K3;puV~wzU9%;4$7Wq~d47&2b5w8BK8Gg5<^2tP3;mVj
z@uSDIH!tX#2GNACYq~yKu1Qt13Xb*JhIik8%$I)ZBUWb@T-<-ecYgRy^F38H;u7n7
zXT0;_BT{<C)$@jL{qZNf|K6AQ;KPS}di9JaFJ98c#QpO#5)AhKwvD`eaZT4J?q9C?
z@QaVw?mODvV9!agS_Rs^<LPt50J}Il!>OaLY^?RFCg8buw!xoUv9Q!x(%wqUhZIR%
zJiO0G?^XQ5SF*9lms&55%kvBK`KZP7*Q*tlEZAkHYRQhp37&FfRqA^~HwaZuWbEh*
zVlNg3UH~S$K3ei-2>8nWLT|(=w1H2aJm)*#|ClJs)y<Zh?T!$HvrUa8c=GJ|P=fb#
zE3e^HDm#>{%eGVBU#|tsCEGc<nvQ1N_j<-4g$zy3o=W8NG0N-{mb9qju^meM(__XF
zToL#atdtob1`&ifB!N1EGwH<ikse24i{l1(#P|$6YyKMWh&}%nk?(TJ|C^vc;syT(
zNxpzk${U=14&$BbWX6~0ewsx;Z^v_D&#-_(&W~~vsow<EQy#HQ)}<UhDMdZ}d&neU
zx|YkB54|1sG)AT|G7tN!NN;Jv5(|spoYjP`sFF(C?WwAY{m#mkdhemP7rJRWT)o11
zWxcMcy#XQayB?xNGI?jkm)2(+>Q$Z16A`p$Q`MFX2Id6CV#5=G{mvYzb+xiO63%gc
zw#q22R^aIB`o=UKzq7(gzCxU`>l!q{vsYL6>m7CFdGYEMZKuR&mGU0k-%zbQZ8Q^4
z<vq`yU*mk>-sL$-J6>FGX?88=n+@U|iEEm+<Klb+-T)_?wn4=~vZ79(J^75-UUL~L
zD<vCyn(daZ?+I1li(mON?|=9u?mv1=eSX2}`~h7T`SHiksp}PWu&nYbR0Lo1-WT3s
zRo(F9vn&4g-~K)RV*hinuIP4fetwB3;1hJcQ2W4U<vDwJ4^S>I*F3topzk90&sMBg
zYn=1cmF3-ZU$VJeuPVAOAvi)75X77IP+<5`a=GDL?pDk{q+$_Zf(%8gy!YUo!Jd@m
z<*dGZ-3OvtHN(CHXJiq~`x2#RNXRK!2D=rrG80`&mKfRgR$GM(W70=6sm8viZ9AU7
zyy2VQ{tll$e+fAXPf^Y{YxX<qd)vfiD!tgI7UW}Nz~M3T@5jH(i#;`3OyxuwzArJ1
z+X&h8+tM}1u^da)x^Ui-sb+rd)7n<i-W)p;kBC$#ZH91%BniB&lPG~iu<d}-Lvgqt
zv_!?PBK+TJ`HyMjn*{wW?sfkTSK<FbboU4SWvVG0`gjHz|A6eiUV~d^>*e3+F@E95
zDLJ1r;pxBQ>KKXt+^BWP{7RmG5fU<k(fRlkr=|mDIc&m;EZX)qE#p=bR#Xddl?=z-
zbUog=!6t1v!;w-;quY15<Y=3Q)p~{YJ)!pO_pK3v#)!|w*~&~|H#bJ8M5%(~`O7Pu
z3R~R~sv5<TM$F-^CS`V4#(>r};>?e-4r^QpoSmPUU!~90=C&=WiJPXSZF-)+xF&>}
zdzWkK)fufcNQAC!d9mG5HIDO(Gn0PBQ`IZ3u5M^{9re0mwW+zdxZwKgikt0?nNC8*
z(`VPLSJhy?TCZ!)f=5!K?-Q%KCdG*0xqkHujR_Zo-OY~g|H*fuYuP`4#)B_@#O33U
z*n~CjoWIXzwdVYMP3&{k#cE|4)86x=AAQFE_S?VDH@^PI{B!@pU*i{l<`?L?E<1wF
z0V(3l#6&`SOTlu^3Pq))Lv0XWSp=IAZ)c&wXKKZ(Vzm4aS8`j_@Unoejf0N`K%AqG
zk<@h-(ySJB7ZdnASCTR?WJ-pYmY=J`)Ak)2EpaeLyLQ)g-0XHFwIst=FR$qPo)7{}
zuJ4i(^j%BSb(VD6bvbbm+ODJRTDsoyN-7olK1ceH*j*c{vKzo8-=r1kSD4C`hx^Gd
z=`1;KMu25@{CwXbBY3^!JI!r4)E){6I_%?^crEpA=bt8(hu1sBWJW^QBeX;+T0Ab#
zp-5pADdX~vB;EcJLWf{s{MM%h?gyaeA7#TY(dvIr&wt8!_g|CZmwCDQ7ZJuPo#S&T
z68GVK%hA8{c|F$4=5ysv?mOMe%oQK|$RMUd%0y%_s}%rFQ$`M$_h|n*oo+e)ez;M_
z#V6ozsd#XEJrHMRfv5&+;=QM?DofIfCbgBwq<U3R)pf23p|saqD(|UcMb}%|SaqJ*
zcckmcX1(I-W=GR@s45TcozthLtRU7KSg$MkP6?sPW~YXIW7^L3&7LmyY}Pefv7CV_
zLwu(Wwmb80;7lT3g^KqdKj!Lc%kH`%wGl)(U)T7mMzA~<8YOm0zwfxbIOE+99-31*
zMd~V0Re^Ul_sm=;g7>w>O{iLOWd@uS5nevK;m6OPqBZ!+YLXNueQ*_Z?P;5yCqMdx
z_Qi9W>sLVJ!}lNJe)%ubZ1;F)@GnU$IkNA2L>xc+wR^1BEB?;^`gi%^x4*}){^h^K
z%k7rF>!DyoiCQH@HAoUoImgD?V~@!QaZ&~xsnlCp;(|czGGslH3t4sMKT=HwY12nK
zwGy#S+wkdUPuMqm-h1y|R-u9fF&fioyOzG`XxokyBYoG=G<z%d)J67ti}*{)Dp6`?
zYF!TIu#9vJ01*q(7A&dK3P(vU6qO}06~sGK9l;qBOI}FXU6xX$i-}GRVm|esr#47=
zjStV6OSPxZJw;$|yWa8lLxM0ev?T_y%vhHCm@}`r<i2#?;aKPQf4DyZ*K~XnWj9K{
z6H*y78z};+wI28uo<BbjbRDE?LFo`4QS-0h<O5XxJ<jxhq)k6h=YAE*dXArrX<4fn
zKi$y#F~gY8(YRe^bE6|?D8!kO_%kQ&n&(f)-$Qu$EXZSirG>$q9OdF^DjZa;a(a3h
za~qsHu-hksvzIJ5tA){Oq^^XfjqLV2LI?<wLpzoI_J+%Q7lv`wM|Q1oMhAq-%si=A
zu3l{EVxkU#-L~QBk6$6)18Ts3O(Lz#%#Z@C&sID*f57_>YGT*Y^%e@6G;(%!Zql?W
zRpl)QLM?n&lafR@TdfWXU>T5}uv*u|&Wb?Qb%hTho2Lv>cfDHSup&%~5yIdcRc+b!
zLO`>0ZiT2!vK8_C;QP<`TYvW({NxAk@Y7#>hpKY8P$3zpGORoz3BUGKE2|**^oQT!
zOF#b$_~2;TJ>Hq4-UUyuxi-2aR3jgJ;az^^Fa08a`1L>J(SvvRsh|HUPoF(OV>DZ8
zwkg}1(+t%iLqwCPu-!F?T9RN&iMDBJn>M4{CYq+_X1k-^HKr+eN7HoNTwM<(a!W*B
z*S75T4Jj$#-~5o^EqX0Q3$s-)CdNd{N~PSajZyQC!UPnn*Ab(cmtwvZOPtIre>fIM
z26SP`ia@f0V7W$^WiP7^fj_<!c_~@^r{9&?+Yu+<=liUU<Z#;LXxnXCNcn!65gZ7M
zlizQ~K#uz=-<I-4Mg}?6X6DBnGmyhTV~5i4J2d<rt88O0V=0ChV;@C7FiV7O{WcXZ
z2|UQmgW=Pz*n$KS{|qsFMri*-?lu25+tvS-<o+_2TwrU_9nH}=C!%G}bh2-1BC>t5
z^oTjZ9Lhp3m7WHsay)|vo!+`Ar{^yms3*2_%XN#Ht8lSfA`@?^2co=uevL#kU)4c!
zQKmq2R?E7U6;2(i;HjlTRJeDsW?h+dz20ot?;E_cFx3<-5$y8flHe=)zT?4z3-e(_
zC4@DS3nzhv5m!|OYPea5LsUT%Ud$0MPVjY2>P=#`Fj;bakr4*Pi#Nx8jyj5+J$cnc
ztSVHM)vC6nx{S58cAiif9#_ite94vXViz-}QUcBO;x*=qgjW0F5B~Uj{OH*$ny7S5
z&rg2v4wvT}R^AFWRaM~P;+)I#+E9U0qQ7Z)`C^-^G@2ISyf+^~8*w<qXQtvDU;M%s
zc=p+I{-@vgO`6p4vtRoucH2EqpT6Mw>WUsqq-?fh$f_k7nu+S>W=qrTA?H%yAa=bu
zuvK#uS4ONYvBF|RH)$HyX92pjnj^yQYHRcB#0b1dF7SjUj+P~?E))0&DNB<jY+DO}
z9vO)@4Qn%LiH*zZ+3+T3=8$1>Br)Q=I;)35%8IU-KvS6QYZAw2?Gyu9lGc~PMd#A|
z`1{c8b!3-d6c4A@F68$#OPG&f-tUqudrH22J@Ys|)?(y5i664#zbCQ<3|ntxW}{94
zMaN4dEcw&}n_*UDa=mdp=I;ok@Xw;@KgXy4ii`A&8}~1x{2HDwAQ}g~#%Z~B(cTIJ
znKBq1=;)F>e=y+Tb3u3*n1+&Jp=T#9^f~;PKJ!y+VyTnG5s&+Iso%+sybS|!&haxJ
zeaHvzJ~q2|4gvMfqt0L<R#j~T!<i&!H7m?kS%rYlQC60bZc$y{hauX66mxMYm(2$S
z>DRyjeW(#NKte+<P=oTwrU|?qZ?)P5=R7FH*jxHW$}lED=(NTAK-5IDYe1cuZzNLH
zR*^7b+EQxGmhPQ1tyxS~sY0O3^(lM~5Gb~G5l3*6v+;%B{l=g0%|HD<c+b<D8-f$g
z>p;6}xOZ{JYUPk%#B;YZlTPdtQ60};{l1xm8k6QV-&3<|kZ6dvZPytNRoAoKwd|Xg
zzxf-##j9s8`3t}B^Z2&oPyg_neDd^#ExUvf@>HbtBJ<~Eq9(CRrlP4J-e#zGCYr%e
zs~)@8QF|~AZFsTsS}<)w>64Iw$V{WljKrvU8r~0$HpgUBm5o6LO-hiQeV^|M&2{6g
zd#4Oea>-snlV?j3s2wO7L@_ByI{lJNTe(*moU+S=^l<x>Ih6C4wojkPw6DWo#W{Lf
zG?kP89%>!8+5Hcrv*wKMaNP6b4%wbyJl)_)dS3GXqvuluH&h;U2j`X*T4KaD1pNYz
zci8YNtXbi-Gw(;H!=hR5(tLGoK<<J2KLNiF>30$RcZl-eQRzQK9sgMzt8osG2>$W>
z9+6>P*gEvHw0&ysl#!8~YHy@@l<t-UigM_L^Swr4Io5n`e=klm5E0(Jf5|%!E<j2K
zwi3I9uPPGB;6T*!Cln9fo79{o!#-JPo)3PAj4{k9AttNl*S4K88z5?nbB5RCGLmiA
zb%e^3QY1JF6Yg4rtGK$kG3PSoXx?l#IpNPVhkavMT(R#!jgU2cmWG^}h@wV1@2NsS
z2xxB!W>r<ur5;k19lX6chLf^yJF5cNwFa4zw1~)bO~ba?^BaHX>s;^mh&LDE)9W37
z^2swkeEfi`eZ$JZm%j7`Zu-D){;l8P{$|Yv!vc$WtbNvcR2?eWER%|gX{It&B{?Ig
z`<sT}{_Wqxbv=Ln=f1)({_M~2@!$L{t9hO)`dM9%2`MYJGqTuK)|#Ub==?%Y7G8<4
zGC#DALKQj)S!N=%6SUOmNTaDD<(DsHtt8n&sLPi8tbG*EYRbTo(=}e+nPML3QzEG#
zl53h|NB)rAn|&n4>~zpl5@w2cb0EdSL|;Pyjx>pRAJgxrCDCzjk+^R^p5fR-(!Qi`
z%!7ER;jHsM<~ofGf0|v{rp<}UfNWicX!|Pm@&{bY|IF0vRK%HbO@k3-=;7V;kMl78
zQwVEhAXyV)yhz9uh#-Fz<O)guU)J3#_Tdjm?wbhjg8q{jn&v@Ar`ziY?(}EzBTTo;
z?7)LlPiP58TFC79*DZtTz5`IK`QYYbDI8`#&ck-A6TR;(<2T0Mfbv!r;S+N?QM3n9
zR#gQN`b_LgMUtde7%E0Xx)op1bjC2sz^ke<Wdx;fqcNyZ6IH3I6>Ul+u?3Z)6_t8*
zbwgEGeEQi_Dla^G_<(h_qS-ZUw>w<MtlIA_;jHgkT$YfpZf@xM$i>+PG|{(K?o><O
zL5y^bMU9ysq(_pV#KF&!l0}ArVNfL!l8gWbCtS4+SFf&f=~pB5eTw|x#WgSYd(Kxi
z@19@q;0qPM`J2DXpZ?%S{M@_m^1=CLB<PUK$9ke>X+29@nx%0w3E5mrQa0X_IDh}^
z-{5>5_~^@D=JN8A=g(i7xebFcPx<B5si9W$E3HOSa=8u!$q{g-O%x5os!@_bB~1gf
zo~B7hG7uLq_;n#CLun@Kxghxn;2;BWcUj92$;2uD)`8&U$xVeZU?xc|Qf={lWF&>*
zOdCha_g$Eco_L3kn-Ql&ZKB-qp=L2Bl(+Xe_A&M~KW6e`Ph|9IA9C>Lj<yvRF!ceH
z3O%K-jN_1d7>^fBlgRS^aGuNaedtFR8Hg<w;o5zhR<7}sxT<vcJYit!jvhs*c^}6^
zpd@-ONchuj14B_v%K4W-{whBHDI5I;8vaLA{ud<uqxtxy=kmtV#PXrTJ<lcj(w4^m
zy35P#fZ1=I?|<7i8}r-xK-yeO!-?6X3js?c_HE0})sC~x3U_A9CbfNKR8-O1uXIW(
zB}fe^HFOT$9ikvH)X>6!bcb|z3?Z$^Al)D#ISi$e(lDf?<Nz=K``-KBulLJ6-_F`+
zKWDGA*E##_=lMms(#rYQ*=ksFs?3p7=1$&4RQ2IMvT>tpIjzx3U64(aMcrp7A*s*L
zy%&s}-(LhYFTU9O!lJUZ4H>)eS2f3J|1I|qE*q?lU)a4re~37thcObqBgn%bo>{eJ
zi!P0`c3-u9{dRDH<>OmAvw+}3*qCw&LVEo}T4!l7*OYfGdpXwb_gFHK6OY+rrw)#<
z6RGjhN|%Z-A%{TCkYPx^u<@|x+^R;7BMC2)$~z@Nzkho1IyZN2XmYWD>hZOL7mcKB
z%%%Wo)?=>E66v1o{^W;-_4Y^gZJ&p$`GhEoT2h)~KI2jfMTV!Vdl<nBSe)1=K+aUi
z06-GtO9tvqf&3JH!Y`aK3NC{|Nw(7M+m=%$u0-Neqeq(ClHf52KUe+FeF=R_8>#J{
z(aV}E&nOR`aq2mINB~Jb`7xrCcJ|{hqYuw|Ml#6pSt)Y_Lp?tBnN}=>=bIgc(r(se
z?RRwU{R!;RIcgwl(i(2Encz^89Q~3qVwTy~&{8Kipw-m%!5v1v^J6!!(M52Jyfs?<
z?#$t#x+;*m5vwmBm?CCDN2G+-7S8^0rh6{tQsji!cW8sWr6zgT{hqeNV{v>X>Eo|g
zb=sY+-7L!Ej}{NF+d*l<)2@#$L(bN5Mmm!gCT~PO$7d;~5xC9LIanp&x=@?J4RRQo
z7FXLSiVoLvzrXNPBx}BY9rMLkdmzY}Y@g20>(88TgD_VNIXPE6l41p*Dw{L@tXe1M
zPBN?0n5t@1!)v>W3OdhKbQz}zc&asga1Qx`VKvLqZdUy#si+fXlE3#H4u`*e^BhT=
z_e~yZTNFKfEaNHq>egxWiE+Y|ogF`AD#c&as{i7n0nx*!E>PR>8+Q`zHcT7#D=vQS
zc&4#-b@pR*GmtDQxlBusUc!Ni(&}|7rOBDFPb>*$j-#fE0HUaTRmQ?)&?c_!iV~0R
zZDFT8E6|EFzo2A6QHvO1Tz)knl+K9>_9GajqiJOC|4SLh^xjve*3FP{SlHzvg=sW$
zH96^I>H&DiGcfmi#E6onDKfQc?BKHXJt@IG>^Qr$f&em8myFnTmsK4~EE2tNOpkE;
z_{2o*{TSt_<O>@Vi3r&#IXpr26L`9SJ(NcveD_L{@CDT!U5Vw3!KS6z5ysj-=s{O~
z))A#(=zGh`FhsIR>zuks0Gw-NZ9{5GM=&KLgE8b%)upJb#JMI|0ho*bxR`MF1O_V_
z0L3q*<+%6-Aw)iW)P|Vc&{k}9mQ;saq)u@Q3oz~Qq`y@NW0Ky)u!JwZF-~yqT-ht*
zT~rla`q!?BEj=r<D<?Jy!jyT=lebA~sdK`1t;7O&wb{h@CNVLaXNkqjA8N=e0{y^~
z?4uRkM<z8+uL_^J25a7tDS6zH8y@o#aO(Ssb^7Ypr;Fcm8qMh0v)VNiF<t-ARM(GH
z;S7p(w-yjepy*EQ-EyeT7(}?U;ZenevKw^Er$V=3o(@|5K^bQT-i1>1dpux_U*uqg
zI0D|<#d`eHcFmEJB|<E6j_o5k_bfFxQ<XS&{+Mlw7D4H3ia6Jh5xv5?e?Pa->9Z{g
zZ_kACRsL=ks`g3!5&QUrmf+HeRs4b%w(zDgk!#J~8@_$b_bPal_@MIyechBb^!<Pa
za;g9KsjaWup}>qBi`c<dI{l;`ydcW6BB<5srHsyOhTd{&&htI?bk}rIvgG-b{tL>V
zC!!VZjc&WEi;*dd7rVZQUK+Q3=U6KwZ7<3Zt4?tJvofZ1&>$~ej7P(tIzo|dHAsZw
zG`{uXyCC<YyDBc7zb1kkoTg#_oPr}bH`beX^k%TFkzVfWS`h-|G?h-labXB*E=fz_
zl^uuoP?N3}f+>fF)#1lURMccBcPic-W(7PYc8k$afU0}M)NW;k5{tCI1}=7pH@kYC
zUsX=md|~_dkz23EI7%W*g!Yk;m3u@0z{1NgJL=!s_ebp*DoN+YZuVrH*3o*2QmR`5
z&}frW|Iv1?Je|~K8$slrQvE~Oz%~aUm$WQL&PbYTXpl;%h)u{8$TXs>kj_l!41czh
zKsq0%Avzmwd&oYdy(O|qujB>r$`v5eVY`)@CgU_lxMexy2-?tEV!jI*bRUCs1k)kx
z+<x;s^UnZRA{03m2ZKc4X}-x(Rn*1@8pei5Kc?oAv0FRsJ~}2KZ2Y$rc@-c<(X;j^
z;*dM-BfEJ3;QTaB1;l`y8u26lLFyxoR9b~;Faq=HXp$vIp)@%Cb<f#_q%K1zyUU|K
z5zQ7B*-q=-2fu(870O=oods!bNY#ZvYYL_v#5;@h{VnMZCf|P)93dEvcn14c_EkO>
zsZ<wf;!oHx%<)CtpKn%GOacK&>XomyZZ`M2X5+g|qptSEDxI`mAY*-*Ws6e#GVqg>
zNEb(li#Y--i~gh0&#X_=IqWI=XYzGwKrNd($?b4)T>{#FX|Bk5#uAz`|L-&Bsrwb>
z!$o+{$8F74_w`QDBZh@CexrjD#>h`jHZ8^B!elz0dM~E<)<%%D7H<-^a_1y`?Y(oQ
zLw;;;6<Bo-W#C4=7q^jA(kd6e!Qo^Fjsz3h2Mu&qQErP(9jr2r^S_;pkc4R|dV3uU
zy|w<lonqSoHW}@G*Quj~o5)k1QB=23V~4C{Id7xMh4j`W*i449H`@5#`w^?$y9TCd
zzjb48<g$HQCXl!rFDi*7%~faa@Hm#ZAHO5@>??S4*(<S?&=v(9md?3LB;L(vt~>7%
zsGixb5fsxLpE~Ax%Fns3My@gF$9+$e-J>dcLy7$oC~#2z_utOg8Z<uge!#EgPv6L0
zivv+*&XO~~fL)Z~w$J+2XGbTnWW>)8>uA3oqO;q)wpNBw>;tx_6T5*v_1CoHl6jPa
z?EI@AFjTY^Va?Rl%<pZ+O@_nSDxUXkntKoCCa$>L=f!rCAGH&P=2{!L>06NecrBqN
zu7V|kh)6w{p(@ID!?pUrcB7K0S1G*B5SnLb&hYuo0Cm30=*AONV%@W<RFw9Zw&>T1
zk<<#|PKQU3cV_L}UvKHRwN(W|KSQM*b?jmssS#nzmEz*x6}@#hdIQz1FO3jbKg1|J
zF4}ZZ{QLsr?22Qh-%7w~DArVBJF(2Y_bP=2WyB>82p&gl;I4kWJ2R`hc^tbi6Kg7$
z9$;`MGa*s!r&@lIrqwDHw4%QVXg6{vZ)`4TIVsQ2DZvyqA~K>nHTAXJ^wb=II%BUD
zeYXAvwi8XKHexY3k~&29J%g4<{ld_7<Jg>zVQp*(5i8w)QrXOC`GO=@FSa(l)LHii
zc_xmt|Du8|#P4sn{6v-<&D*(QxzXON1INsF($u?e-2F#Zv2y9(@DLD@I)st&k*#5^
zce*3@n;8$JoBS(gU*M1L21kwRRW{~Jlv>RBd>1%Bo-AEMCWS#Qk_Vwn@9mhB_hk2E
zEnCGkH3RQcl-MB=K3g<aENhxb7SVMc3Bxw=xFC>dde<6zytiIs_Oo$&g98DOz0bS1
zyPOQPT0gG7H_lzs_m@y0BxIvDz?nGF@ky~x5hKb6Icb&{D#wN0?G-P`><-ao#y?Kh
z4`A7EYROj7k4YW81;~)<B|m=dTq^63hVNI;yj16pY3cggBB}KcLxpinwCyT-z*yNb
zddvdU%GR~}dOd3}%wFs@t9u!2%@y%OD=vH$5TWZpOTFr0Cl`0|2I2o>3S>-T1jJ0O
zcTY306_8;wEFLanBk^M&`HibUF_LVQD52#7ZKK{7+t$;kb*px$i8^)u#%J+$^{4+`
z6q{?!Wkk!G`{G)+#J>1>7!NS{v?6&?Vh*Vg5?#TaTOfOgN>T3$Ui2;46&cIfR?H(S
zBVg8CI#vkA6I8Q4A$hKn->>>)nAYUy4Gwaq7B(r#e(9P2v$KpBX3JNJl$e^BI+C!-
z(Lb>6Gc~iOR4FD}OuBiu^RNjm&%dTg++JwV$Zd5@njG1PA;{ato(V8lU!+jLDo$gK
z;F)PrW6Q(aj;~b|zLi<(%@}36-U@EW*qt&<mzt<0RAT7kBRZ3h3t>%{*4{SrPP=<X
zU`f2@1Yu__`%EmW$^~S2y=_kQnk}u|FT30`@)@2-Mi^SaH9n&OtnC>?JIVjKi%C@L
zOjf$JRhZ_KMn8<(Xi2umwqp^xwi<j)xlQDo-fzZUR!(O<&(Ct6#o(Z&LVP1_2jfC9
zh?nojOY`AEOtcJgNWTYU-vPY~`W&Lm*RfH=9j^$>%<A6?V(8k|aL}e}<w0u)0)dLb
zY;jZ$X(d}3$9rh4XYY(~gYoY|9Ywl7d~{S1iEf;L^aumr#Nsy*iQ?glznWvqapM}K
zXMw$`1uq;nJ31`<-HaD>Gp@1LaPo{1mOOvA?SYnJKOMUz8CMeU%1N%^Yw=C&myOyV
zj$JX~q@js&eZf*3<;)KR4KgL56At9ZxAo+A)s~q05$OT##i-}=Ickaac5Y;sGnVXS
zq%@6P5~b~l$NQUOHN+qLgLMQ;7h@h=%mlANfjXbpHg937^yYzr*+1%d(?ONe8^>d|
zi1c7Wa?MA|z`>{FC|{?l%G_bQ-Fl1O@e5v>;XvZ|?UV5^dqmliUMf4Gp}yeB1hzHP
zDrKx*!_%?wU~C~$JEWSGM)QPyI72*!6iVK~I#MvJGZ-g6hh?a<ed+@+NS_cxQJXc6
z*6#P-qn2XORv3Qls)s^8w^~$ib{rr!nwr|h;&A>V74%QKy>pei=hW{0FJn-A_Us#c
zckk})wTRnC5?r$1gSGbYk1#kJ_(vc%JSVcgh-0GKM{v{inf2t5Z36#iAT)ARVvKf*
z_$XCaYE!5vX{4lOIBAPTxGc3buVLwFcXSTd$3vksTA+h3QK@KpVG3<`4zet-J}dL+
zmsK!7#Z+#yoWai@Xppn<OV`eIGq@%r?o7tm#dyX?tjx61l7a5Zi7Tsdc%4W02%3%S
zQ4t`VWa2l%%WYb#nA3TA8F#D*Wc!kpmXUVwXNZ8^+)&bCU3TiyOGYWy9lY6>K^(+E
z3AZ`0u`*JU3n2+N4_=1k7rA)!^XRiT`pNdn1SsuuvtuIc$NGB@)ZI?iwp{jy^39L{
zH+gdIbe{-SNRc(5<uHI@Mw}dw;`V{1Z2K4U%vl!AkPEV67Is?5TnP#O(?^z}KRvo=
z(6<*|IdU`tEbr+<<Edb21rAm&)aZO*SF($fD!nfE+;Q*1EM*X!0)z2$S^Qb=*=8Ja
zvaKtrSk-i3^VerxZ~3g-ru2mb*an|^e_(}eAM^d=GQgcPWWIf^GRVP#HRb6<yRfY#
zQo`LvbsO5oFY#6`vs)K(YS2GQojsW{)+>1x8-VIuI+W-L{65=#WA`fRmrs@!4QujC
z&}M<p%GL~HaQ2uUstQgoS-ovW<(AN5OQ9qpl+3<U`urz_Jwym)row=Xl+N_B-Q<aN
zB;}guAm4NeLGmjiu$w}MGu1dJ%~a+J^Q|-KIwg0QB|6>%YiQ+1f&p(39?e&pf1;nI
z&yUrE$=s<%59u|Ogg~ZSSPHSj;VeW$89O&fji<%NNn)m1oeG1u(}-a+6UqKt2u%)O
zguG7d)&AnuawP$6G;B(cx!$t^$aX#yoOEr^jMlVjodI;8Ty4lE!9zN01airJA;_HV
zAOw=ZVJwnm?<zBpTHU+qT<$4nA^&M;NI1j83x#;v8ZEgq|B<_^iPagr;=9fU#gY!C
zE6ZyAr9Of8HdLO29+RWK6-2xtB>HbDRq7gDeKjZtU&Ae9?!scZgd%7X)rl7(l29*Y
z;zy2l8(z??Y{D{Xq46e{PlZ!}Ve!m(Or+BeOAe-^*q_|xS(OBp?s9Alqt$u6{qt3`
z36j;vz7Q5BJP7O9rkw;3Y(0m@Or+9}6#p!s^!OsYWu=iuzg4!~!MQP3!5=S52ZdR4
zFJV<v>r>#=tAP1alyQH|<x#dIFICHs{?109^501qQ`@6Sd(Lh@NUSO0yp#>buG_$@
z3ZK}9(JR~iEGUDN$Y+A`B-G64(1aA}7#P(GvN&W@2Z|&5PRcLbsqRg|2defS>>8tF
zBne{^ghM$!))Vk6NL-fQcUsLkPb<loO%cc0+PyeE6X4fR4$qVLyd-?B%viPEetgO^
zSDekC0c|&>W^dw?%K8aK4R?h#+s^;d{<F|V&~M78wK<+Icq2gw6!BaG)`5f>5)E#d
zHzc#py{7v=IaZVTzGXi)90%J%hZb$8Wg8AXwRgWmK_?7JrN2+i^L3xkCs$x(h8BOl
zQg$+y-bhvx&JDximgGBs_^uMdJMmbqw`ZF4To-3<?;N=>VLaPNMow)VrMH>i2er~g
z9re3vdAuXE8J^UQPGHdv)}{S*o+e0pdETyh30f4l7dC{Tn~4Vof!$BoYn(+H=9)26
ztrTB+S{H1mn2S_a%R9f|31EKq5i1s)xuZ95O+1$;H1(^@u2Kv3Nl=N^B<B5Z{3yo;
z0<M%7RScDI2lBMiF~V=)WZSpU8F30`A%b|a1{f}M8N8MLkv#s@hR+8sQ*E-klfqi(
z?yDdCr@IZ=2bbDW%yohdve)-gWvC*39|mqV6ZgPhhN8&>Jo;4-Sv6!f7rF7^N_{j0
zyf-heGc_&ERx`<sz4vy>Fp<kNG*mfn?O=gx%N`K-<+Tui6-zVg?STB@SL3#il0@)R
zUlajo750ed8S(F0-)*ozxCm$mO{rE}GTy2!dMS;6JK;(M!?>P8(l|G7X^j(eznT4J
zIi;V^D<DISNIy3=3^bGkHqFrog4jKV>q)ItNOJoQ>4qK0t<chFDO-bjcLti1p=}iO
z=Mb|Y?V2_TEy7*PZ-Of9I?i1pH)k!*kgB_49n>fCjdj)YLB{^Z4ml(Hl*=8TnzfR)
zFK}_uiK!CbZXWc&vof|IAucmB%vHyT)TI}%!Gi$RvtY>Gb5Afyd2P9Kb^t8NbI^So
zg~&a@z{<UpX`Xdr>OcNLKl??&WA(=c3Bu;ttxLcfFwGL?((aKb5-P0|KX@wfVTP*t
zOD?NH8uRFoa7JnjO&6lrxi|6K(Ru<@OWfWB&3B7WX!uouq<1`WzRhKH6sgA_`u6TA
zM-(W(eukzJt8thtD{d<hP|piu%Y)l9V9FQAr(QCG)+25&uh$p@{{+ghNJJXRnej#0
zxnJNE+JJ4_sQe1o5H2o)gHJx5w3T(zh_KF6VW*dE6YFw4MTVdV4asa2be<FuY05W!
zdHbvB`PQg<lwFXY+dngFNJ$>Wt+(z$0Nm32>>lq3dUOgyO%$D>_qb6}kkj(eXX8VO
z2tr)L&2Ml*7p=z7SWGJeOm$l$dwmtvF6L($*NxN*1eCVdok?EohlMV2By4Xqf1mGY
z(#)D!(M3o4|Me6fSF5Nfw;P1sUTY$(M{S>E4T;%t_g};o`>HkCW1qTz5x7_=ImgGC
zAXHU#9vrih##l?Utema|2Pe~hvIu^b9;AtI_P(c>k_K~b2uX`W1WG61*^t?nNm3_b
zFd+^EN6Z~s1(%(BBkHrUsLaEv1$E9|Z$titNWqXJCJGwBi<MhwiojBAGzz}P?eKe<
zz+9<g!KU7C@v(Qwn8hDKpMl%a5vR;>z5cj(TVeNx*?~ELl-ZPe$zvz?e5llcgOpkH
zs*wog^uX~K{8deDcbr&7byNMXgV6~aT28|Q;bA5@8P4q~>#?6pCvx{@#b!I({YTiv
zUl!H0F8sZ2E4qo<uhiYWNK~d**0klB?soJkG>~>=61$e9QsB7`2B5_0r2%nn{edh6
z<WG2y0eS&$Pg0SJD#RDes}Z?b!?Ee>n1FIex^}otk6UbAi~rV$mR<SzR})&Vb$+_o
zjejl8(t*b*$)FnoWT7l7eWK7-cq8FCxL%ox%A<D*qZqhi{5zPHwKuynD$ebTK&mwb
zHv5@~)&zK6up&Pyeu>3_K}SI2b8Il;nK&3wKFfACnK4fvoJ1V8cEyF2BlMCMnLgs>
zW`j>Gqg!w^1bLNzd(y%CP2wzvqMz&JMT^KOOvkhs(;t)id&hyF+yn<Ug~iGm&%L4m
zIm>}$1Bf){#&$zJT$*u$Lcj3hQ+fa_ClR+#_D+*yuxw~CE6MNh$~MeV>o!O5jLR_B
zN6dy>6JdzzK@p#|cBPU=qWjv@q&X-@$T$fcBAQnp{pe{}AqZUAPC=(sePp6sG~Yvc
zXG!#a4D+^sy-^<XYrx{v^Ksg>>tBCMKOYNX+`ej7fl6feQ#>^v8A9IUM$2(;3s!qs
zUt<LeWQu2x+pjy6ddvV!8MFoB*1L{(XDZRVm))8P$M;(jM@2<v4VM1d_cwO5|G?pO
zvmwIqt*46vUaZ4gbHvc_6lSn(jw<zOrRUX`zBifb!|R@ymE;TE&%S8LJOb<u%N8%s
zgLG$oqQ1f!0d`u53@TL{n{Z(BpWkxK=r2&F5Fl5PwBKmM8-8E(70~kjSP^u0ZfN1^
z;|E=6pp(b`QLLUu*O*Axz;i98oCALAX`C~BYBNafKX5h*>FsnA_L1lE-pblp-pvlT
zmTR(PJp-TK#gG+UZ46vWV71mHpST6z*?Io*J)nrl6JTYSkc#=1mv2{bo4k{_k$Ef1
zD<)1+_|!fyH;;${CG4`}Smj4rODb_pgt#>Yj8@Y0p~aJ8AHXGTALZ%Eras=jYhRvz
zOPy@Og5;ptW+_D;P4r3+yJ($lh~Jyh)%LBlPzE_sxf`sG4Q0=2seEj!%FQ!_gNy+_
z!xgw=m}0Ngk)OxcjLX-7*Yq^n0$lU`iNS!fLGZd-l`u7>2UAHIz(fOiacIW-EfnjE
z!mKjP)*+Lb(!@hJ`I(ct(VP#1tHy!USk`;%iEhVvZw77Q>5!1uo^5afi(}^F%9ZPv
zcip`mg!~Hl?nhs(S7rm3{)r#yXNY(DvX5k&0$2p+P}N@s13r;FXU;~msmA%Cj+a^6
zE;Lr&85t}pUv!xB>F1+>`4$Tx1qc&yo(bgU`bOvZNt(8@HA1pKeXc$xn*E5LKumlZ
z57YT<p;<=uW{Trc>e(0lmle5K$WbISa?M@pN8z%p8FNXtkSL>ADbdbDB!&auUJazt
zg5&$LZDa%SG40G3Xk@AJMtp=Rz)=<z3AS)^MDnYF;J;$IWP1_&nTOE>iX=$|&YXth
z_RE*dHn~aPK7{N^LuSnYoBiEcue$Sdv8C7fPjT)fV1Q;c<_Ze$gPfT7c;zhZ(HR(w
zw?b4GJyql$uK%p3|FLw_Sc|?Jb`!8sfz5fRLAAPh1V3F5&P(g&4w+$1ee@(5Z2T4k
zMWoj;3Lv+<r7}jytep|xso!-F#7O8WM7z&gVM|JOvPFkzc-P1IDr$`@an5p=M8XI$
zyU5W7EK~*+$`uSWz`FJ=MMwjUuXUZ;T$lY<UFi1rH}y}YHD8#PImSmCM)G-|a|;8f
z%GG9?4s6O2WBE))MRKcVmmA=3%y4xNWte>pW=~gpWsv;*ZNXmSepQa1#NK#~K@(05
z6+?q&cthMqqN*%LUam~>ufO4Q+v9Y$!)8-Fgmcyu?K}<Kbau-|N=knIRoBdTJ&J*o
z<{Y@W!xLdiUX96#tp6!g%5^L#dSuf`FGrUq)W8A3lyMQ3;dn{~STQpMSdrWEzlYEN
z^^p!n=OlyBDJcHJvx8sNYhAi?^qMBaf`Tjs_fo$AkfZ+NS3S<kFA7zPa$CmW&|7DZ
zpZgf8FE2B)8;d2j+QuRm0@>axEODgW_cm_<dbu_vezMqlEUz712SZRgA{0qUj7=$6
zEZw$!c{=?G?#gr*rwIQ^qUBl3%5`F+Hh#+uywiLoN%%GSQ(cLfrx6gz^IR$3PvtvT
zUEdYyuehdN=qXtO$}$m2FF>e)fIhQrly<U2o7<_)?t&^FBm7(c$=KQ;mTV0DOIcZz
zD6T=bE^6&)EK+%-|NALVMFLB%gZ;(d&Pm+2K16O_>w>3=hx{o^dV-Pz;Fhr_Sojrj
znY~*NQ*M@>Q0%h~A36#$y!?6^tJrzy!4Ji0tK96JnyPO{tgaf3b@sEWGkn&M;9Ume
zQWhrcqpB{HwRs&hY(Z*|3?Es%2NnH&*?rFZMr;;6@J4ohE4^|_ssr3LKlx)g8FZT<
zssi#mjV!v}VO@UcyNx7^vK<8}8`Vs&jb9Ars5FSFo6KC%d&1RvNhn7?wOiY2;<#Vm
z6Zz!;8ge!*Lg}#dc>*j3`aWpt#pF_O=-hrHy;W0|l2b`houaRS(5`!T5vvx>rZ_$=
z<l|xtxuHebzgrHvB;Oy8;?OOcY`k-K!(s&pJsF2PO91?5aDqeqrT=6_6)$5M$o{Jg
zEMy;ZyKh8cDaiiu_WgR8?|-9W9lR5eb0JhHWDkj^8%O6ErU}pTkRXe?QFsY9I+0m1
zC+EJZ8P2ADNo3+1tfGMe3EvGjOg26{`i|=K44FKD958fwC~#PTGxJ`$H#EL?z*-wY
z9<;?zo~)aV)}TV0Xa>(HfnV&91Z<PXjUG+G`<cZd_dGY^!;5eOC$9<|4-=C!@hfWq
z<b|E;=z?-=$taODWmtH8rbi}-d`$m|;G<KysDZLjTs5ZychbHvpeNPYBqpoRP>+!N
z`cpEG-IvmRiZ1qr90!9T`2M_4YDN5qI>tb<W!iM{%dmft{&vEz;g3I^)n8bAZK%BP
zbD_B-k9uzu(grnt$5%2h_X8g1F(j@w<R&A4nmP-=b}C6X^^f<&`;LwInMyL=6T~K$
zXR$Nz+`cGkJNsKkqX>0`qZ{1eiL%sKyk}&9)#Ux~LhPn90Iw|Plb)6}pVc^Sby_Z*
zuK)LXYbDbWT?3!KqYgsVXmr>#WFDJ~i$kRXwqa;NH1yAV?4J+Z?|g^(sOR=2zeV@@
zR`veZKep4AlqqThr_oUYJ!nb0jI~}fLag@Vl+B!6*OFq;3ZMC(-&yqYV~Z80DpoAs
zhZ3dY2lgY4O=YQCwL&W)PKy~_nc&DyWsZg63@~p{Q81ErSg!G7!2_tE6QJj58G>14
zT8u%z2PfbCjDNl!)86oP7o)ojpF7?8`{~hc)Y^5{iof;khm~pI8OR>Re+QC4vOxlF
zsJ@OD*-AJJ{>-MiHz@Lq@^g9!L~txP_desL8=?9jU&}iHvQt+sU@Yz-TI$FO89Pa!
zU1n>m5ogRS*`lF*#1lPk2;`5LsL4iOlbFXox<~KEKfZGv&325GuC>OdqRoxKZF)*i
zakstaq!SaZEz9oBrgd=4{MV~r;g6i)PfftQkaAra1(n3{3ZY+0i^7iraPut;{~}aj
zvJ{}J)%c}=z{eb#Y-!Kq>i41ATInsNa7Wh9EvLl!&Iqc{D>SBvClTepTf-Rh?u_fR
ztU%Ba$Z}ehc%Hq6cs2P<I6afnn?3$Ac}Z2C2llPd80K2r`lPrgdT3?bU;I@KCs}4>
z1p==ET0C}yN^wofG@{9#7fSmfFPz5iZnJ+L!Tl}{B(&|-qpa}d`G~oR$d^CWIW*dy
z(K1m@%iyrn3aJSa@YBtMaRWSselwpGb}vwlxVKs`pKRsySJf&WMn;flS2GAjF47LI
zt<+(4!zAop!oj8tbHA;Of{0RK&P?e&RsmK_n8MOEA=EKOU+3Kk79zZvo-1&LZl9PM
z4xQC07JQJj$DTRVT=?NkJ7jyie!06)RFg0tV0EWcbhkFEtl~3Z%dQmi_wiW-htFpR
zPF})30gv2}k^ctT=s~(4xYg;K#LmLxNLF~aKL;{ikj~#F0x(qJW>r|w5A(FrJlE*(
z{AL3Ma070-oHDNx6}akOiZi@2EtBU1aOylZ=pdHc__dC_VEDoHmISF7X`2rexdrLJ
z_O~<DIAF5FKZRGi1}rqM{ThF&sNJvi%?3H<K9}=U=OcQZwl}!$5A+uZHAOAEzJ7v}
z-0j;dd-^7%py%51u4R(;Ji0e12N&w)C88%a&fe2DQc;ZG(fNd!@s*GPxe>2Uw}XeP
z)-%igC*0PsAA|c9n~Kq9v^K{${<Z+YI1=KC(QC+{LXx#%ym&(%3Q4G%>pJ2gYzAGR
zR>-szqz$9c#P9Na`K(!c?4qF(R#e;O^8kaaobLHMo5EXcjm9TvhVNg}$cd#Mcuvep
z9=m2_L(`@&vKEHcUWj&YR_2amD7(9(tr%yzNs?ti_wFl!6Vs?qmTY+HS`!772W=mu
zeqobz>qc4g>)x;VnS?}Ct|f=X#XXg}biD{1mJA{Xtl25ArX>yIFn&`%0#faB+vF_6
zb>MAs7<k*zxLBTZPxl_7ih-}S5o6=E(C{!GS1Fc-N@M}WL#LtvgoOx!t{O)&v=_kJ
zuCF!E+Wu2d(h)?9Nak>8B;lDXemh>?WfFuK!vQF$qHvZYfY_jFek~|VuuEP?$&=ps
z_mlSSN{l?)Z?s(swZ<{r%GK77Fp^jNUlo<z6}BQ*>Ak+SFD{4;P>XeH5F~Qti@F%|
zSn)-0YJ@4;gJFF9#N^S`y?D9=)&I7;P+QeWN1$tr<P|CVv(^puM?_iA`HXbuLuh^_
zzgz|rSs?gO`ZoVCrFEwZ2&h#Njx*lse6gQdhA3-D?qg=Hn*bHGf6+(C7n%c13~e(8
z8oYCiqLe0{jiViUvDwNbvo$p&e_nyrT!aTV1bMW8HdHgJMq@KjWUgo?+k0)Qtc14`
zV8I9t<!OYo208rwBMM#W<x<bk#trBZ?Y2`|)5G}^b?8^cJtEPMVXY;=J1NQ~Uz8e9
z`J|xy$@7_6Ra>TSk1QUk-(SCmn%+;Op5+bS)7hSO1oA%sVXO_FgBxc}r<|(W!d^ML
zx49vYh4D^Ux~@zy!-kt(LZpYlPXgG5q>O89%+?RKr@C857^x!8$kt><^}njC-Nc;F
z`{JSj7lE9zLr!<^kqZ2#h{#lWVoOq4T|sQB?ra-K5Tn4PdJi5-`{?dj`n#+9_Mn!R
z4FX3QWvEWe${~groCfD{k<|Gnl5Kf#Q_02s0NBlnNvwlm<^D<t<Og*n#*SEIiq^_n
zO14s)H)GyBHa{aVbPC9rZdRwXp#StJpZff@x#4ROFY_sma!1ogWZFOPp+`}v8#NJX
z!+K{Na%1LiB)tl6f1m}n=y0z1XQ~y~99S43)1t>~6w9yreTM)d0bt3c>MxGo1)Q?J
z{|NsTuH5&#9e0GB2f%*4c<J)@<RH4JB#0%><Cm8&RJyCp%Ewb^>ick2hstrZDO10+
z4)c~O)-dP0{8)0aJPeT|pKHEDb8*0v{(z#v_RWZL3xRPx)x^4UmwT*ucJA23jgA5=
z)<znE@lBD=qKqG0*9`m(MMv6R!@R&=Y^12Y5wp8zLxa0zz!Q^DIxCHX5QdJ4S?SOb
z{!0)uF_v^3{=|FlMsuIN%weCZz{~uV4cY6<`v!VvFD;8oEo(F^zoepvv}c0#Tg8xW
z(dGwUM#0>Tvp?r%ymV`GL+2zz>Uk6u^YOwmfw6MTfcdk#8@HL^SJGi|MBJF!H;TxA
zi!lJW?`ZN?au_e;d20NHtW5Xdln6DA;yj4srXPi}#m4hYeUMN;*A+-9-B`9=NmKdu
z6PK&-WL^vZVXE$If4rG@{Draq`s3#=IKJmTzNb@~Q0CLn!=M`)w<!2xlPU1bsN)|9
zg$MckmLsCz1sx3n(qf4QBS915%dMFhMeje(%hjD_d(}kDtBr;wAPxmYw_7KmD{C~%
ztI)vi=AUc(ZnsT;?t(r*#WSSI-RemyXJwZJi*~n=H*;sn=**tJ+oPV4`iExM>lY9I
zR=_>Z%Pw+6+HTkXF7D7KE#-c<=pgYvc}orQC=K==un4|3dJ^MdPek!n$TdWT;XXHc
zUwiRE_D#4<N66J|aN=X`X2<u43)f@z?}B-Uv(F*?&+p}*2Lb<c%FrnFe;aES7_i#@
z*F{Yo7siK4VM|SqthC3T0oBF|fd7fmH1X@NaLRwTbU3^H`b#tXzhgJGIQbB7!vFkt
zw|7ndZ?V0H<o-|WuGbwlO#d0>e<Cb@`2+m_b(`B23`uCdb1L>m<&(!lU0Fv7t^f}I
EFRk{Z`2YX_

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_repurposing_annotations_001.png b/0.16/_images/sphx_glr_plot_repurposing_annotations_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..1d0b49bdcc8bab56b674120ff18b7107ca1bb885
GIT binary patch
literal 45288
zcmb@tWmKEp)-_zTK(XM^VnG8G2u^Vc5}-gJIBjunai_S4VnK^0xH~NrT8c}t;uI@h
zytu#I=XuZjob&zuek8e)eT|Hfm9^(ubI!G+wO%R{5zrAldi01$RYei@=n*E?!*v!O
z?!)tM_mZuL7jbupfxFH-Yj-a*SF1-FX70|8@7x{j%$Yo`T;1&6y%ppY<rU;%vUPWN
zc6-6c=k)(>;C<(6!#6Ce@XwdvJF6JFJ$gi9_Rj;OM5fs8(W4r;s-m2(ch>H$VU3lY
z=ThYDl<02Th<ouD(c-d!%$D-<ma@lrt@xy&{K?DM$$2Wt6pj&`I#j1XLq3M@?`wgM
z;gkf%1kwfs$$&b`Oa|L%3J5JWMhFdF-k7b>Xk1QtOV6_d@1nMI(W{mDb88Vy2yi}p
zz;UG_s^W|24)Nyxt;CY4P38R+U%$AGNNu!+qdED<nJ&gTt@ta9|G8#=_*nVrN6R<=
zEd29EX028g_uuzkC@}p0uw?dLh7k$-z47DYkhP%V8p##d>G#*-L^tbN1LxZ-52SD4
z-%+C0kKTOdo%`(VrmN?J`v=^#pL~wlHKk4`{LYp|jv^%%eRrPc`2X!cU-xsGS?Kbk
zed;NCv`2R@;2HS3|C^%-*79t;OlO&?d8Irkd<xfr8%g$M2k_jo+&N%mW`YJR#>j89
z`lVM07tY-<)kUKU0>7njn#K%?dgJt?WoHb^VpziU1hiuEjV6whIksr;g)6acqg>Kx
z<+C0HX}@i9jUKtVVswz~e6y`_A`sC_nSc+*!atDecr34i;IOC=7TYw)J!R#vb;?jL
zs?dKv2v<I2C>#(b9W!u9<O0v?rN02kD|V5ZD3Y?u$7Eztc5paaX&cEw4J;0&ndb6R
zkR$LI<fHzJncx2LVj5e(`dJ#*MDmN5TI?3R1Ws~)s17h|r`O&bntpG^20HGkYm1P{
z$HSs&SblEvHovPhsFv_Nr>ioVw{~&u8T+VaJ{3jzC)&DB)VyhubT0MsD29SZ+LA1|
ze+myQ|1z3_{1I45SFa$QnLEt2^2_-zzrBl);&T>}-M2FfHLYe|o%>(27MAn{oL`~H
zD$~_D49X2FQ&09{YJ<R;JPB@jRXqmf+rz1r3wbH>*qJR#6$!#oWT+=mWLKoov<-R}
zq@x<&MPlAg>Zn4TF`U9G_dv1hm`gbFWJo<ur$><KN-dPFeOzNq)oTR--6Hm|zXe)c
zT5w^Ivf@+haCR-x)GR5RV?ku)XrsKl`HQNImprwRXg}`M!PG3uDI*w^dYvjUiZk|&
zLA4|c&d=F&Bz84chq67|C4kjCoJ!3VU?mWR+VjT{#qnk2pL(N+L2D6aEhPUj;*7^_
z?zyUXVm+UzQGB{OR{??L-nO5{Nr<IJ48DZ`S`|ViCn~l2T;XqOR`|NBD)mUs&%Q=(
zLC2Jm=DeK!Wi-mzTb=6al&r0-`^H^WDx$5pNh^P2>kMShMV9ldqlCbM2m~>Qa)LrA
zoP^rrFGwF<vga(u&-_nRBNm8S?_K|uv_4&rTFP9QlTzg2F+3QoI>rhysWUi^=#01V
z%-nx^eqZJE+YZ#3uJ|iUm64{{PT&cAqNqh@B7sSiAWK&=)WG^CsRp%V**HydEho~}
zcXQh>QrJRQ&-<VT9#<)}2w}A{$Fp^^{R3z}H!v13!cLL}{*+K2if++PhoX_o@_D5~
zybUvUn34ps#@t!ttlWwMAc*hu&l12a_Z6f-gjS*~>UTe`f1f3hxDsl_01FIJq%hh}
zN&({G*5w}nvyH)Y`Ub`<cpb_DF^xMkqNy%YW{ZBljD7urIfjZp=ER&r_+~FMLO!rk
z(!PW15_!`fN((MzFK65MS<WtI*O+7{B3Zf+$*BuFf>ryBH6~duJHyPYh8n_UDrC{W
zOL~bwsR*N|%n&><s<DR)2fwAN^ZN7<+7l$%l5OyS0RJ~z&lgtnlQ0haG7aON1kcg^
zu~xJq9P=^Hb#7DEHJ7Y%Dee8t0FJJ}G~L1jCBRwB!zu|&<Jdx<!V8|&D%nnRlZKWc
zN`=9`4P?mDzA+|fGWszY0Ujthg&mDwYti_szEMhHXUM-@R<X}A<hYAe3$FvGi_^rl
zD^GP6oUPsuMkSM)villHC{v~HBeYRl5+lB2=tkZ*n}63-L1bT)9pa7?ailbcCh8*+
zSVSgm`xU7Tb50Y?`!l<LXGQXGI%zAJB9Dsd!5*dq4tp&eCwYoyI|Kl;Hpe4?J1mSM
zPosn@6v{?Ip@P@(ILTEVt^vi@jzCHjB?F)~4W;}>W{u1dUEV_0Hr41d!E%3fIvh_D
z>!>;!Cxfc#1bS7QsX_H!!8JrBS3^@*f(tLHf_y1~gMGu)T2|DU%o_M&vHPFI{GW&h
zoO_p8=GlkMrsf|@4W?jGfMpL8wFTKv=%v|VWGH=7ArPJ_hIk0spRx{`PWxH&FLL;n
zKqc<1^9hY`FlBME18KNO$TB`}{VTr)fdnb|q!w-H0r{fDLo~to;tESXumRgjaL;nU
zXPO=wbA*8dJ_z_3soR6b2SF+repb>mvtq|ZCBWF(m<yu302V_1c=@rM{kG`DELBhn
z16ArMjyaKlh)M+E1ezdC!qu|?sSQnZW!*&3F%+xx_d%XOZD1c7xHS>T_5@b;RoW!7
za*rpp0>l7KC?zN;N{&0cOzl!jQ5!DYLo;=oL?MJK`)kQG*8DuSo>JVVLfEzpvwS!z
zreP2#g?8+7DkrW*$LA0ImJu%_Ry`pf<WDdv1%UE+88j>?vJg@a)*khfZ8K*-O^eR`
ze@C(1Pf}N4Im@C}vhzdOSP34V3AF+5lvr{?MajpA_Sg!e$RBi?toqRjQsVerPv;t0
zAj}+c>cAqE;TW|vPjLwcwOQOym@qW2imRnb*|ngKvYp(2hg1kxfi)PrVzK~vB%K8N
zydqBhNFI;IqGl?|t*Dlg8K2qv%b+^Dg}nNf65x8sw@rzI2sE}Cq!*KwPYYj5NZ`a?
zLL-^IUk6ciWJa{VbJ_W3%Vi0eTA*{FD>DMqJvB*F0iz2f+uINNQ}Q&}ix?bmDutQF
zv>4bDSkc0q&%M*4wMUH-?X+L~^q{w`(bIub8ge9Q@}J-+@W?0#moX{KS7r>elS(yd
z4~?W+5{esr(ez$&-J7#g`VqCP=SZt|AQ_F|;l44df8`^uLZhvMSs9ZDfkohfEESn9
zRbFOfF)nyd{5y(oXa1FovQ}y@3F_V;*w|FSFj3)t8`z<(Lm$XI4x$V>Fj872(0Q8l
zc-m*K-9luM%j#))Z6h0**hfB<{yLK;p&!_VJo)E5@gpTqV}S~;6Q@z_JCPLwKu{i{
zr;Lsf({vtaDUVm6E6pCp|A-6HO-GS|;&O5fXR5d%KtG@CO3={Q6%@C(cW@cTptTY}
zN3$84wIRBYk!0z;bfHrOm4pG)tMfs2gi}Z!Y2WxJei%_zta~MgO-F0c7=Y2Q$*YHx
z$Z47qW2AvSjG<a+?*+i071D>_Wm(q3{u`{l>IUrUGH)y><&r04lAp(QPmzOM-4PaY
zg-}loJ_0A{$a2RW03*FL?TUn@z4w1fnF%;SW$Nl?tcPY@!@VeDEtH!H*Em^I>?(??
z)cdzAWMfej!vau>#Hh84Lyi?civPmN73?r4)``?*EEPo%WctOQ7F^jKfHb!GU-e=O
z)+fxI{=*lSNr_dePN*b6)v?%DV18l=*dzS-jk~-mswsjL2(qrIV<q8KTe7;Uk$>9l
z4NFeq=taaQTD2-g166xsimINAb6hO~EVRd}^tdY;Z7a16V1$aVq5Pp+NpBalOJh_N
z;hC&VdT0zMxu<4(M(Ty6#o9(IUsz6v9S99=gsfO0NkKRV0!B~39PY#?0h;OwsPDEq
z47t&tD74J^;~-BDL8LHTZd>jyX)=0NGSHKq6jGGD1CEjzPdY&o9*R&Nn_`%wfvROV
zIa$6uQ`)B|j@erICq?iFFWPW30lJ^Pltm&mv_G$fS<QqpTs*~;xkL4W#@Fx+6^1gJ
zcAOVU)}_Ovn4nD71HXpR(f@|Ug_`UABdvbfxP)n`vN1)~EvN-DUqskpdev0^se;vW
z5LZ|fuI?j%T2uxS3l+0${w^?tAVaYzX99u%P<F7jRI!AqE2)U18iOB+hl&t55GGMf
z30<_av3iXaYXd8&+z{Z<AfNzUoHleUA~h>ELRYcG5whfU5-@0MLN1}nMrDE;!$os{
zi2u;1<*SiQ5T56RA$=7!PLM1tp2DC9hl_0vT!Zr~We3vb^2;Fk9^bBm$RLvw-b;en
zFZ;`oH3+>=(h2p<khlW%C<obO6yb4GCJj(L<Tz*hZD+|C6UHd=NJmYb7)^+&{zpoW
zTlr;(TAr(HDIU<kq95bsyXRVSNsSiiT^-@rhV`$`FFcB59Qx(744FE8_rScw)db(t
zF)a~^G%N{>rLlAWwTs0t@~85`TNr+}2#-PRjaQ8#9q5^@ltN?1#Fe9-#~=uVZR27w
z3VC{r{zSiVRE#!nH%JiHO<{wS)ImZzn(13$h^FVYnqeXd3QDXib2)3|P@$xlbj4_(
zdm@e0;(4fSPo5CYqeAVnp+5O-lIS3n42!*~;L*oQDWCwxWZP-_{xpF!`Ja4P&`ng)
zj|8?i8sSqBRZj}i`;x##CUZzK7Sol0JktvB)bCJkDM3c)k{Sn$K#WA3y{rXv&!^Z2
z$pRv2=hs-6*HAA{qnEK<Xr6ED(<f1ibNJ^*t0qT9VOzE`_b@;dUILti0zMkFBARH8
zt&lOcaBx{sgHt^1WY?=L9c4m-uUz)8o7G8)xI%~Jqk^+Ku2qH~m2uiaaKTz3Sw8N)
z+y5K5Tk7u4MtoLqj^miP=91r^o+Nn;1#U=c6bzyW6Q#@QQOqY(+W{IH`ZKyLGEY$?
zN4HpK(U~0DqPJXBDqV(G>x;{t@_!f=zgn!7m=8Bv2~aGJX?MFIr4bee7CItf<^uGT
z`N|Qb0*eqXhtl$xtd7s))c}F=$(W;09aBWi?qL<q9m(WxuqEhLv$zMw(-WcE@mvD<
zRup<%Excw3+!$(iXmpk!FlPCg($f|XO?QT|=$5xZeOzaun7S(9vD4?M0VjNEAeYu;
z;%<a=wm?`rH<}wnQrIc*m0QjNrh&lsSE^I$(iWzIDaC>`yJ)n;dNh9cZ>yMopBdWU
zr`l^lzw?7dWba8O{h6WLbH3gayy~V)!KNevV}eo~f=KJXjg|`-rRg(M0JISRSN6g~
z{!eYiwf|6<{|V%h6ayD+LZaD1^&zuX9Ayh;bf)K6GUm+Os2BauZJOs0x$SAG0NYu*
zAI=~%Py10~3eM7u+$$J+3?P~|8~cFS8zPt;GH~LcIUF-v=!9%+SfF#_0X?pJIg!%Q
zo3l&?q|*hJPc&i#XYnWvz6r}$`@!Am3%AzqMbiR<tM&%P5EyW~E=$EH0h`XuC`bw*
z&7?=P44TXpnKbmNl>d3f^xK+{XZwk=OMaeSMHK~4q1uIf#Yj|L=ILB==o>#!NRnr`
z6P-YgwhkKEQ65!-aJU8%s$w|y7{w`Rp-4*)+lrhd%+(*dA)P0ux-2Qhbdv?r)bm3n
zhLy&Mx;5Q=vZ45?lQ!-c<8pDSSK!R4glAk<NhAet#!$@RzX#Yblg5;YG*x8D!0gDR
z*xd{h>3i+}Z$rq$cDQJvUYljYo4Qs~14F}=))DZ$x)4^xN5F9kT7W9F$Ncb#a+RPd
zK|_@S8n1&ovHwfA-UH8d7dg2lAdiJu{uGMB8=R~|RI%cLStu;Rm+3(+%w^0^VVuKi
zMh*hqja(#F$RsIyd6fbrAiEEsga$KXo5ytVjw0w9Y%8d^Eb#L=Hlvcv55#g1{rKk9
zDzm|gq@#1lu}p)kvW-}rDWkb62uJvil#wBFN}au12zZfnn#{83gXI0QQWe+Le(`b>
z8wLGZ9xKI+ixgSQ8bb}L+eK*WP&3E(e|l&i=mE)(1%`d>Ez%LnJeu^T)LII5ss$yH
za5C44Ec$9OvGf|j`FtrUpjCu}_gw|rvTqGzz*TB+9%n+#JLT)^`Nj9u|3UN)7)Uk+
zYyQ$WbUca6ByFs5oED3=>Jfls7O{s1Q3Z=Iw~@Mj@0L{v_l#&KubZ^RkAj9_oMiFr
z34ran{`8O~0gHk1cJsot`E+%yTV&9Ic|+MzgUW%MqDISt3pYU?w?3aOI3-ium^Y(K
z6;iulShoe}9*iVYdI}r5@L*j>Xk&(odcxx$mxR*@id&l#8k9$c272`m#mum+!^mQ5
zEepF8o$SqrSoNtiI16LxOR$`t+s@3+GUICg%~|6_(OMmfhfu^jqn~%F$JF9E$jic%
zSV#rhy($7C^GbTC;+Vx>Q+apEF8RW=qNezRq7_|-CPLNpT;-ehW8_5(en-BfBQCp^
z;;K9qy*2jL>b}Aw(fN-E-J>432Ge5@b*HT}Ej}$vNS%Dt2SY)&Vm#YAwF?nc8y7Ah
zpeR`@D=|c_RM85kEm3B)>Kh3bPGXs|gzaum0p!SU%`fsF_sJ{Oy=e%R6UPrB8>OqD
z;>Y?@id0iV12Pcj@3o8~k_>S}qw<ZL-+Ze1LI{@CD<}vufa+Qzwb$_r$e6LSxL+5G
z2%~^KNXO}GS#>KxJ*yvwG@vJ<CPCOKRoabKpG^QEXgj(rH%_oER(ZWJ@bk?V)ntpJ
zy=&M2*`Y%`X2tlL`ZjZ6e%cyV{`ndb03ISS&cw&gg;LGq?Q|2U5~P>|{zS<$Cs1}M
z&sja;vM4Zb+v;alhzX;ROVP*LT$aMy5YGblJYjQqABHxp`Ct6Hha5Sh!{Jy^{wDKw
zlQD#FT&<<7NtdoX-e{FXX4&_1S~j97T`@)jWli|Bt%C|EK{9>dO#@S-QUNN~wEvgu
z<w}4vqEaqj>!D~jhLL9cXu_9HROla8q@J|<u#UND7W(rt2KV6BXOJMVZc+QG0zra&
zS4mQSp{|};5k+a)*e&2`VPrKcHnmHB@02Ia%8()^G{tbH$*I48xrG7XAtOl75DCYb
zDjAqoAe&PG46>?df5XumWb-fugvcvnm41p`N8*7Jb%3!zNDNyM?^N#aIreK2$M6K%
zCsEc?2V*S}mM*4BJGd5WC?rJ$$4m8Y;s(=mRx*OP9ddDNS2|wcrr`zC;)Cu{%kFUz
zxS8P}^7nH*r^xR2e;3PkV@Z>bOjHho(l~V#DV7Q9`{dXLo^BO&-9%7A-n4-6V68AB
zRKIG3)LdR%w+Lmv6TWz2cOnXC9))>2fl9^#w0JOmmL(1;<(jKvX$b=;d;mB+ZI~$d
zHCAOTKo_x-2II9##UMx=C&x7Xm0>QgOWUjsrH)nQ{9~-mnGy}Qa4?RLsh6E7cttu!
zQ=VV~)no24KUW29<87{k6`-BZ_*^)pD8IhdNx3tD3wv)KwL`)Y!kJ}d%un)1y7-Y1
zWTdgm%|P-1nhit2fTdunE?qWqeb|N@#6q|K?^uguVxc71ZQz15mQF9Mf>Uz_Ia(%W
z2Oc`+f?#;p&`Ujr&5z5(iU%i+i~2Kh!=Zplv|o7R|EAGH8LZMs5Cn11)U|pppdCZ-
zaN~Af05g<CY2Z53nWUIT<pLwym{ne~pm&xO<J_1DDTO=6nJtK-f|9-fUcqb-84D!X
zZXg^E%)<#si4aAx_k<l|I0;IpKDEVFt~}mVQH*dDHd8mPGX$B_Oe<GRLQ)<`O=3%J
z6xVS{haXv91cLz{h7$!?eyKGn&I)4DixnhgNH&*RNIx|P6FDyY<=GD6E)?MUsZKb_
zS!+}|=Bca5&1!EUXM^Gc(lSam)lI~)wIucHr{wQRXoW-pd&8i4iN7_ofpatn-t{JF
zDxHL>MruP=q-*xTW+8{phAu?K=#3@baPni8*keKprU==AbHUMLhJ4tc-$qgc`uhJ(
zj*^0rV|o^4F{HpnWmXO@E>}GRsqcBJ81lKL_MzueZEgx@`q&&Gwg%onFSW=z$~k+b
zltNm4+d^kIO=pZs$IoN%aFGa)X&}_OO(B-8Co0h`LCDCEDWM=;M}<m0g5m}e*i92U
zg^A3VWhJO%E|SHIF0*$YX6a<us@r8oH_ENuNjlJ_Dil!E7U622l;@0k=%#dVcF<_}
z%VPN6DL!aREmglv$^qOJ*TN`S1Y0a~Q5a<$lrH7*iV9E6IE@tap^)wiY#N+}Z6d<7
zx+H3c=5y|UjXic%$M#tK8rzk{@QIzyfxsJ-=1K!p;l1un8M8zm4(D?gN)=K-%YkRK
zb#uUo{&g^+q5W|DK7aUs@E^XINGNW@SYJ&vmldsuEdH<rS7~Hjd{0cbN6v7oIU9Z+
zQ*v#m{h*{63r?;oenZnqJ48Q3(~3kyj)aytA#cw}vDB=qr_QM1(^P$<Fbm6tRv0;y
z_Rm1gtL6g2rZQ5VssnsxB$x(duh(B>`KI)wfk3j%(#EwtK#hNnOh?f!U~Cm6+WIK1
z9{I|MsxKppTpY)D6MRaO_?x(}fT9jDw#96W0F&1Dp5}Ftp+One6BkS+Oa*xB1lA2)
z?Y|QEl@fz!_OaN31vCS6BgcU{>({-=PhXqN`O8KL*b1AcJn0GRKM7c-C5)jC!AU7^
zD!K`!KD#o;lAMWqTN!8YLFaV`!q$e+lCXb^j}LujYp((uR)-IN439kB6^slAJrwTD
zS|p3|=8fciqmD7?g|UgXEXTAi#K}a25|plbLaGfl<h03jfrqh)r!*#iE2+Zk$XAEJ
zr^6nBAqdO|g)xB~^di#;E4rnwYl~Jh(`-|uzyK=V6vC9a22|qN{GROgh>W)KIWv+h
z?XQ)xyS$mBMp3UpI9ZG!l6ZnVyfUmxxZoAKpre%mO;#jeDG7d%I7`$m9BGZCJ5ww3
zMlr?huL$D+brjPer?ublni%0qpjb@|e4Ot*u$$7bIt#x?WB?TEThab?SZ@59EIT)b
zsVp@vN4by*KIo4w`cyO}M;;NSmsQRCoNxmsfED&#4XNx^@c-<Chy}0sNC$EiuC`<v
zaFMxg99|o`%09%cK*-WsVH-6RN~K?rB%ZQ+_N5x2V5uv5(mn&RFCX*yZr-L;YWCcM
z*RaMixBxHUyKd77I_RjN!ye+CsG{9LXIdu)poPk>Tc^0{fSA|Q6OoaH6qmqQiTf|+
zaKj#<zj<_Ul!~}4RRfrMwv=JxrOp7SQ>neMWj(=I={JtT26BHM3zi<3MrKk*D^(Kj
zZ7w!ABI$YYcy(DEd!acK!0o%y6R!N@Tc1}jC3lx)GQcjAo~lBrh8+<YmBO`NhcO}K
zAD<O#7Xj<cPaXiOv#qB@Gwy$cl_7_gQEMge;7^wGZ1<C%TYiN=TWzd{c&`6&Ra$#H
z;kcZydP}b5aV!W2*Xta^=x!3?zy=Q6*Ew#%U65IPhM+u4b!iBKObPzu#{K0$kfMJS
zLP<D)5x<>NUx&vke>_<g4_^mfVDf&eMZ|I)gl~be%QUNQV)So(vI}Mxx{_b_y$d@k
z#e1m;#kE0lq3M~)3pp(t#}xLBCaS7P6tf4uMygL==7(V=m$LPm|J9x?yKp2~2Tz{|
zF+zQcBn@G*tji|G6}*F99W>!^qQUp@s&Z)PR*~cpo7H$iTuyuj1x@7pY+k7vx8NtB
z6gCzGc6=*;whLL*g-h-MOMK@w;WuR{$u8zNGU^v~B`4h1Q1#Q+sht_;q*UgAw0{o`
za&t*klwJtKhGys!l_9Gf9l>9Rz*mQhukrMTux`uTY!&c!_YUfTRMcF8G@=IK{fRBw
z6B-zrn-Lz)tZDp39oG!TL|VjLBhttu)dgE*@O$_bpLp)mS79&_w<-fO+#309VdjL-
z>0^?C-xHZg&|of-j0^#Vuk0kKI8+hQn8mTrZ0~~@Fpg1(CA}s$`Ou`!(U=c(Jp$om
zjzy(oPdDqr!@&`}=Y@ncEaJ+R>3lyxiR3~WaioJ6YMhafsP4Q@BAIf5m@EM>W#SW1
zzM%K!J|kf#vrE;h6#20lvie_xtR{~h<5aPdRP<9YKUq5YgH<^?MLJk_;`V|OxA|P?
zztZCYNv62EESs8pc{bHx#5Bf<{@)5!*qS(>J)mV4Dn>*x6zK|uCQ6G{)GDT!!36b)
z=6t4lIl3tNxK>cbO)QE;-;;((F|GJS@m>4>3Un_fK0*QIxI<n(_(DaT3jMUv2|BZ4
zg`CtvPo++b2f~MG8~SIU2GxT&)EH>)1}?TL6~ikg+P-dmU@DR>*HwFZLttEc(kkBD
z>ZywSE)XgN4Xusv;2h*@X`)vYj?N<AIi*7Zi^8bl^su_biKjzhIr4{K01k?*VPLJJ
zgr$%$6kml9$GY-Fi~T_*J|eauU0i$8%!R>f(^tr2MXLME*9Sy)d??8<(MD+1njX_;
zW2zr!peSKDV6Pd$@eLjGA0_A?3GUmcw_Cq5-#4=2M2<&T|Dp(nh9szsuw~mwBQIyt
ze?w9&|CoQAnIN(`CQqtP-+OKT`4zxTwn7QP&Sh{JRGR^P+LH`+Yi2E#U23F}U{mIA
z3Q>3Y4jUDZpYBd);TRx`rYJ?Qbt@(%2^+%)pqm+)O_cKeLlwfhQ@`EW*Mdrg^Pdzp
z3pz%>c%5R02wB+>V=Q%od%Bh*26M`ISD8bAn^7d~K}`6Y@;=`|aR>B2ycGx7etPjz
zP}{@rSdxy~Pr_r|(wM`8NZ3M=P=&(+3bg}}HLKtaNP3Z(ePVR5rH?hz`*6=q)Pe2X
z#^^Xagwk~-{YkbxFaDMf;}4?xuh7zWbZ$P#q}~4t%>zv^qQ34+@2j=^ux73VaO+Zz
z7p63{l101ss}6S)S?wU@9YEzw7M2nOb&hsi3In)Yak0b}v12$!KfR=?;)Q-jNQ>)Q
zJ@_%=^<q+1UQLm<;b~DM`8>P)@=CG-`6RZ+P*K{$#T16vjqkUrAQdDag<`N!pMc9M
zYAoga^v!{o$sv*q)CIigv$D`*fQm75y1!tIPh4S7y)%vJ*5>Z6!}=veWx=I4c!p{~
zD{9ZjY;`hmY1<B0ia#zNARVY9vLM9);k(d;C%S>5?e}Nm{+m|CkL4sIPi_ukcaMWF
zibmEy2DhgNz|V2GH|wi=(wB~R+H(I}@3E&(<1j0>_%mDJSdkPqo>(><<+%l2Nn$Bi
z*Xvx|D&zXt=sNofE~_E7RNE+QCPM4x>Dy&4+-nsN)OLV4=ZsQ5=C%qeAZrnbqDsP~
z#)Ml%w{duOn{p={F_Zc#J14HOu*z|Vx&bzUVA+?2`c{<>HlrPGGq-l>%=<HpLTq79
z*x+}BEcARK3kEHqi5^Z=VC{`6KzIdwzMyp-$L8}+BNb9IcIB!s1R_l*29jNmdZ<9R
zV2hZjE<MW4#knNPlqx2CB|~|}XlZl`X!$AafYRp@x5%NtlF1AzT<33sHlNiT=S+qg
z?sWg!J-pn^b$gq1<kua_!n^r;_2%~PWqj34%YSwMWdh65w&e!Bsn%+OW%kGqQ9nwI
z(p=(1Mk{?Qs^rL+z01mY?fWSDrOkSXQl@qpcU^r&lzTE;rcT;9lq2;rp&%r9ogJRZ
zD}G2{R5dD2?iZ;SvsVdPE$aBd&NIB9Xx>E^J4tmQWn_hbT^K-Zm?>&w8Y9L&wUGYm
zGPrWg7$1xQ0hK6ocint8wAximKPlo9xPn%7;o9G=MLA!4^e-zQx`xvnC4${@5K6Ri
z?r$s!jc7BEs5%~pSY~-)_w%H=aaoj>#jFW1TSt8!YB~ELDao#KK|@+OszhxO)0M}o
zp&+h`%eilpM`B&JglvR@Px}amjlOaZLSmq2Ch6`~Mn6yyZSw%BbUQo~nm3aFVgUEc
zGnL8`V#QIG@+J?KQ~ZOExA6<-?%QlYiRBxC(hg6P&vj~opj+hI1@ti<BXDjiYtc0M
zt&cl48-9N$7QHXukCG0~!IH7HHM?s{YaVQu3G))C?iS<ZxIDT(U_$_j#y`JOVaaba
zX4`#e(k2JBH!$#{Wt?Xrsk585N@mF#?yF=qB?PeGJ<h<N+7Vun(~i`$B#-VO|HAfo
zv<w$c$N(E)ef_Lxs78#%Y7|(K1?bl;``+uylEfuZ?*Y4@scFIskXPxVs7W#RsvE4x
z1SoJVE;5gPj6w5rz#(E3NqiDt*@)PmW^B=ao?=ZQ)oDsNHMMjQQc&)0ajQ_6T})9e
zjAVl$cnX3%7Gsm#XdW;BeYDp7ZRgPu1lKpG8&;*DXGO?SRF_I|a|osli~V5{KfV;u
zx$if!KfJ}Y@SCq*(l4T_P3dLE>~LUQ-@VPl63qj<+T6eH*zc_5&jESzz3vc_?+b1}
z=!5YF?MNkSZ$F;281!A#cEGB&WH+LSi^$eKYRQ?<+34l=CmE0(mp{oMoz5H7{N?qb
zrK%Ls(JLE6uOZdhO4#G<yw3IDiHV7eucV~C-q_kAZW7T|N>=EPxN7H{)|>7Qd_i4k
zH=jy7alFe@6lNQ%SCsC|Y0M+#CXcQuo6(2z5*szv(pZTd*hO0+hfgqV^0_~}Y5$~F
z`sa?Yj8m@5EDlxL8&<ek(b_qISKp6FCvzg$+H=~kG~dSBK-3bf5)D6oL1G9m>9IiY
zFXRYhYv6Nhn51&7PI*Vd=UrW0)$uNY^OmGp{9pK-$KH`8UW=p(AS;ku>z|gg3%}o1
z=uyZ~q?9ETM5*?P32dW!&r7bWo_a1Bd+yk$`H9|h9j`JfX-h|B89uu&HwpOjgy)7@
zbhll(*6bWC<CPXsGla)t?Uw^@-8fm>n*XryK+Kd@BkJqLk9f4=xDb^F&(m0QYX`{u
z3rM73Pkx4i+(&c7>7I8k+N+%f!{=-FU-%D1aj`K8Bi+iPnaNyCWs`u{bynr$!bVR`
z<^`OTCg7d!FIINlgLfog#5@a`c2d^A3>0JDHMsfA{Go^*!-9N)O$$Q~wUzi|U=O-9
zZI2<7O^1D`T)vz3=-dEGV0}{+Qq83h%^v<u1wPXsrSfK$SCwl8-oEu93-vk<qLA!f
z7+<$8Ct~>y$91NCe*cBTUf%q!4$S0Z<fAdv$dwZBHkSurCaP#fTf_3p<SKrX-5Net
z3vU2WvcM#);o<7dVv^$Q#D<&VVOn*rs@-<GChlasN_zWGw2^(c6(<<^gO}F?D{or5
zUrPPnb?nbjyO7g5uH+fI2#q3YJ$T;w{u0whS*GvPTU#zavk!G9AlN-N`ilGEZ$Qt~
zTwmn0#!6^9yB=33TxHVm>P$C}vj3d94{5wVeaG#|$VF@L?D<#<|H5KKmaXRWR!MXv
z3oY~1n~EMEO+r8%1oz$wO=+a`2^sxDdxQ&t+q|FtR`}B{DR+9}Xk?hXynFOynS!0M
zqhMrC|0lYm;P|h7<<)Hiu5ATx2{WST;XKn%JZ%K^c24u`DI3ZT+q?D3DmvCo+`$!o
zNORRv2Q0kBl+U_oB04!6GjnbgCYX|%e3aE!;X)bRIhB?ZN8H+I+8>9C8=sNgUmj*w
z4%^Iy?E4mvy4Fo!`K;g;9T+@YdKh!d`BFFduI-4pA6vKZVKkxPS)JT^2|NFS6JBNh
z!LwmMvJ6$NU!NLA&Ptefx<ubDbmvDT!L;{s&TEI|RKcz}q8oqXJL?iw=n91p^>@60
z0W4=DZU39nr;gJRBfRe;1tn29Nhbc)kvPX~AO02e{P|A0!IDxY4jAXt=?dY7SaH42
z4-49kGXef13!X<(4aWz0<GAG%Ad_MzspPR8n$xFCdb`rP0zuD%e#E~VODb_tAgB>@
zJDCz_vmc|JFiNz51k&-N>lmL}snGg1TVIc3n@+n2FxhCdzCsBjEF)hSA*aJ;R=%Kj
zjEd%4c(zpJVVT?RN?Y_^2fspnLzL=Y)_9!?G%w!aqVmfeQp+jlY)!VD$<kl4ug^h4
zutBLjQ^m65ZxmTTiEgZq;}ZcXjd(S?EdnF75?{`#)mR>!{PnOMzY5{eQ<}nBIAj>#
zflx;<e{VW$tE7^xD@@PX?%hDFezmL5iwKv_^@YJ-ZGMqb$+RCw1V<7%GojcBb|Vcq
zZIPdFO9(%SZU0uH^`8};5y}4bex#j~YSP6p$>Oq2;Hld)2Arg_Aa`cz&zw^DCPU=|
zHEPCdzb%kcYwzZwpIZXzkN~D6Ah#N$pL&_k*=G`$i^*zvPQ8p}&|3=<;ZJ`Tdm|Rk
zkqZ<#s8TM3UPA1gxOY5YR6Idsx83UjS!yM{CT-ON&nekG!r^6KlBj8>Q@5J>F6Uzx
zBIh-hSv;IvoFX*=PDF-~U0v<?OyC+nRe!W6Wvr>?K?$cQ@0CC#U5Jgx99pB(R>-Wy
zsieZpr?`+VZrkLIk9!QKf<oxqjxqu{a?_1*_jq(OPudW77bmA*{AHU$mDD}9*W32~
zS@Zrg$B(Big*48JO7p_2*|@B=ulTgZ(dajYC<pR>g{TiRe(X)wHhv$MfqN}%Vf;vM
zsA{FRMH=Mk!T{iLXrz+8d#ZV-DpwU8ltgG?RpJz<NxX>_dl@sBR=m4M&o5)FT|PAZ
zE+D0Rx;jBQW<kyG*Nc`=H->l{M>-_0UIxf!s+_W^Zaq2=Yvk4YB<i=4(H<Y3PFDpF
zAkLe_?dZP-96R^1>(J8D^%(O~v|yzOoYef~Z}okBTNW|9JRXbXk=Pj1<i9KNHTLUC
z+8w<8ZhL8i@4hQ{W3qW^P3BJKF87nirpDhq)cROk4JHGl=Xr6fdt~pg^x-I5lhk)=
zo{jB~TMWK)<-Kb)OMHS*CQk+aUqtfGX+=#tj(kuxE&VX5P_9n<WBAqYjl{bq^Ye%K
zw-kzgj!a+mX}=e_rCcQ(xNM1@`-a`>CmBVwEOLn~Gf);Viq*||ognrU1i~>tD6#K!
zovnZMEI;e_K!?ZMoQ3?<2&kwKyt|Js%vJkp>(l96ubA|0xk^(31#P_aNW<OzctI+K
zu+zM&kDgiGC~oAqykM#ALP$$#i4s9SK6mEnz|Y5;IwN?dkt&u#i`MTB$cmMVE5@(P
z5oknl9}_|wn~@hP`N(Mr4F^{mC`ZEKhJ1fwFO_~ys06iZd`j@X&HlK{+1Z=Z8oWKF
zJ62~>E#@TgVUE%^6k?_xY6%$*#fa9%qoW@D9y{UpAZRj@Zu)HQHPdP+XI-Yw_eK_U
z_$zQ)@D^&O@ufOsotH3BeDweM%buYrJ$qykSYVPGK7qy$ynS~g;lFeqz{mK0yVaK1
zb2jh}=kC|XcgMN6Huw0_sfL!?R^8j^nBI_uiR6)C?@KJvbZMWIN!RXq<>IU)TB}p1
z?y7Ygm%G;ryKvZm3U9S{gU@5WrH7Ki&=axHd-}^MaYmj&1pFc%7P2O5dVAJP-}<0)
zl#u9p4wdbxNGnkHd3@F?uX6N@Sxf+)ECQ~hSUb4l;32f}56$ga2?WW$uGsnt&DCl7
zMIkjO9g|=tTO*N^m3rVJ5;VJbvwN+2Sq-K){r*SbtnRJX#Xy?O(7?TD!$LKwzqO0|
zz~n8~;wQJ{3SrAAUY8b0Ifo9t%o5?`O4<RMkYj_&=;!L}tND^T10v9rA4XuvU;bWI
z&Wvm(P8}sN<P__hz3V!k0csY)a!O{aKdPZoZ+7Ye%9}D@J}bTrQPb%3tXEe4oqKat
z?d;c%`cZg5<y3V1ffRFf4=}*0F2G8tvW?U3vYD`RnoeIn%Mf4kBW$%^=lyYPE(a=6
zU`Wk$y1-;tW+bJI<gAW?q9iN+Tb51v`QG!v5)G=*%AR!JfjJvqnOmNNti_es+*=;L
zz>7}xVx5=6Q=x_tK&kIX70M%3y{x_GU;aLm`MZ$2HgHA^q|Cj|<~t^Nzmn^6@T1vo
zdVfA*VN7g!s<_MjbA=712AokK<7Pzs=rwxp0qXQ0pzcV@E9H(S;6hu7t`eIqB1=>3
z#TDCcCghE$*?E4ZS-_?tubzSFm@Z|qui~yit&T2Y3~4&w_?(T~VetKC5ktNrO&r~a
z;g>G`<Vgx@)@`28=!eCP8+fMZ&-WLw1Ue+loeQ2ti7#6$d%q7|y}ZX}voI$N)8*mh
zb|~FNG%sFIqzpwdyGiDvQy_H-<R!X#s&6&VX!HeJL`FaN-8g=*9kG4<FqLJ~DyE<G
z(m`W3t}>?Xr>%zR^n`{EHW)}_JQbk_56?2TvhL#?uh|e-lVx2oiJa0hG#j6AeCKrh
z>v2xUdXDk=+vAOF(a)`7W+{SYO-M=I1d|~ivTh{HRZu@&?01QLQ?~u=D(}-_K%1Tl
ze02<AGymo0Z*S=i*fLzvRK$t7#4OMoiqrJi={aLVQ<(XN9Pf_O`m>A*Qo{OeDqxJC
z7m2v>^UJI0_Ti%=d}HHzgc1u9=~Jj&+4p{8msiiN0)O{iyg?Sd+Y1@1_N?Ox@@emF
ztdUwV@%*vLW}?=lYV;2s)&b8K)X&fxPZ3RvzaDncLzU`1|3!4Psj^+(tsq+}AfCrH
zck^z!lZ5=%gLUrR6Hmr6fj5~1PGQ}D5wqUvo04->q96)Z-^>?eZrcmq(Yimv;7g%i
zVf)w_M#N{t=}Qs{zQI~`*`!8EUP5T@3O!R$wI#n1vk^JKR}|78MnfLJE5*`pCF5l}
z`Kd7<TKJ`oebwUkz@60xy<#3cH>K4~MU+?0kvmj9r{U#HG#fzG(Cu=Z=<QdZ__pHn
z&wEQ|0@*dMc%#M5kl!C>t@SqfjV#<}e-LTGDJ^Mr>)1Z(%q78DN!UqQjKSKeq1oIw
zmom4Tqelon|Fg72x<CC<`qD?LANHd5kH2=$;G(E?^-l78yK{`Nz;EhmTysueBn9*o
z4*h@(#_jfTa7YTzZ+zZ-y&Qj2`&IF}`L5c3eR8^gxOQu^Vu~A?WOmAyHPRrtmEe(>
z&xKR#cpmW=o2~b+<wy+mEAPbe$Je?_E(aG3bU*G=cAgFEBbs>=4;F!i22F8N-@PYx
zuSjQe90hZ$FI)T~e=bn5#FWt<1>5AB-OP@dCkO7%(nl8>zbEWmdZT;rvxky1oHJZ4
z_f63~jYRTmOUK)X4&%X>m2TgxB1vU!j{1Vf%=9phepY+>wR?6&s!k%G;a>7|SEkr0
zp3&i-E_z0se}cuValrfQw}}v#REpk^Cp(X6JZWty$^pyo=X1ok3d}t=s1cB+H`7nG
zxA!52R*$l{NZS;giI|zAb@TUCJc)J@e7;AFW;)Sr;(7~%rM{+d8iTH6p{ux!Jie(&
zb-35D>t`mf<EN3FrJqNS?IZ0?el|UiYZPjftFziSo{Ck$%ZQ#c7k-_DTKzLQIF#l_
zHTWT&G75AzZ`ylzCp|2F)&BL7K%wH&c^8sD#bisQVO*sto$!VIE8h*s`(5Ss_|4q+
zsXJ-B^)FfnS9abxkI=(3&o9=B-v?34oU-^|<lgJNzoUL1{IPkz&gfk6V6^4?c}<{m
zxXfK=U`XIe{QK0&-%g61^&>K!uP=3EhQ~yum0#@GvI{CnJT?h9Q!r`&%lbOtl&Aqd
zV_CA(k{*&3dj5?D3Wra4{P1nFe)lfEzFrWBUpOeNDbB*of(fM<k-C?BeS0<Sy?R<b
zb}nV)KoyHq5)<#g>2>tS;kDd26Y6roLg2tc(%$oFxzFFf9&{2#uTP&z2BuVQBb#3Z
z#*z3hhko(C1?j!}qAJrL&(Ye!U(d*Pl&p9uha932#aNP%C+2kGpOlX6zR#K6W#liv
z%k>U)c;Nk{PdypGx_P-YefN@4c`4IBybF<El7HrjZyQ=khbY~?q8*&f?ds+hO1=*x
zeJ9=YPji3R73W|%3HFF+GR_wgVo~cM*Ljl9^;+Pm<KrI%Xj*Pp*9qzIEpUe&nUsDG
z*uB?bt{$72c@kSEPW)9C|Lj|>(hbe~i`@0a73SRA@6~`g;hk=7gUVP&;RJJ8?gzqw
zkzZ9LdH+l8X6nbZ-4XV5h^t9kQ>RPo_eww3U^xpn981r0Wu6hqIqsbHE5hpcdvNxj
zVAOJ9+UxxD*UJ|d6}w3cXRwh?w}+vU#p`!1rQ$XX!mOXa7Cam{8i=mkMg*=_-#FZi
zJ~S3aI8-(q*wcRt`QE3rkC^!V0a2B+HSg0O54rIjSu7o6-5fFA0Pgxn?w<u79NY)!
zl?RQbzTLa5WxR^Li3>bl>iI)|Hx+o5i&F1&<MR)R58S=>ma>_ORBQ6Aa$Qx(mH$C}
zQF?#$dNua-6&bin%>Ou{g*RvYVQdcuZ8^Zxn-=ykMK^9-o$eGM4Vvi%khOD7Y_Q}y
zw&J0#@9%E@%B<Jt-aiYNPVVQSR(rp7eNQQU_V@AStjiDc{r-UjA9rAd@{j#O>!Ha7
z+7U?)qR<+nXZ6B$UrzE#P{<VpO_?9Qx^5SZd@`Q98{R2pce!=;_pHq7eV`yNCEmgc
z<3!l^YuQTctn52TQ}g@u+l8-@@W2P7^!`DoJwJNk<#xe&e-!vh4Nr4ay_q)7S}cqL
zIA>61O*rSgJ$`ri44_vYpL~3e=s)|aiQtg@@UH|JBe@6Fxo2f9tSt6_{MCMrJ8V;j
zH_Xvz<z#7U=lVw#3{b$Pr*YDVGG(WYcLH8}^qACVF;v_3v}QdMpZob?i@WC($#@sc
zxUzJ4erNjOL;R)JJ#}Dq>$`>gss{|H3sWHZO~#Bai$Mw)|Lo#s^b0|6&@O&C6+sD9
z8p|E>e4w(0n-010>O6*OS#<l3)yaI8Z{FRWOU+&YMfNK1zHT3me09UW2wA-75!Dh8
zUNiS+W{^vaW_z;!%Sfk0nNs?q4kjqkz<IZudr9v9M?yy4)8PYEF=zJY0I!!*nbguZ
z1m&l+=xosIn{n!ra(phIswMA`+#5le4joOYj{84>*9Y}c+!a*qCn-y7j5nS4F(db9
z?_l!;)I8HU-iehquUC>qa}>oPQu7Kw?CVPHlmx#Vo3u1^>c2j9e0Om8JMz_<e&C)y
zUl-fv>>o!F?n!O{F-iQ);_OPPji=9QoM`$FYqri@(~Zc$3(dpAkvqEF``zlqfcOLB
zFC?<;5vdZ@0Xfa>7bDSoD6zL7wV2z^<+Gd~w6&Lwh?3gKO4~=rxZbR1^!YH&fK|Jx
z%FU}&hfn2cmrA<~+@6$lmG@HqH!iwKzu7sKL$Cl31{)FxN?Vf!5sB9tyfRz3H5NKm
zmc8ZEK3n`}uIIi@%uD8YJCh~{8{1fT%eU)O*#HBk6IAszg#atK$JY1pVgNF`lCgN_
zZr`GR|G_%jxg^Uq@@Jf3V9ev=ewo@VmcgfLYn?1rrCyaBg4R<$xx-eN^;oF(mF3q<
zyi##Zog5fLJZJY&;uX}MM`Y3#$0{uPuMYZOiM~%kuTm;Us%>!3WZpd+ykz9)<`|HV
zNyj6Df*=(7Rg6c!B6HqcU*`UPz>`lN#~zFRJLOq93uclIS}13yJmVKiXwa<k?f`#(
zjGCX$qVKh^>#F#R)?VL$Z+l*;yC%Mw<(ksn<{sbw?(!ztqeu+2IUVvD#ww=_3P~&b
z^xilJzgUtS6B15qD*keLS`7r%H3xnh9wzoQLg!XE57*`rtBBXOwZ+v&d+t3PmZY!y
zcG@=h<?ndp#cug)rRc-P>VV+)E<Ha)561jDV;QfpWsZ_%<i<1Y&vfkyIU!I$C8ei;
z`TYP~X+W?kBH6E4@_6Ej!9v1uX4XOcCm}RobvCSp4&ZO;<rUAJcJ6cAIGvuG17?W-
z>~5E<Qj~-6-t%e?gJP5xZ&}BP#GL+)T3wy@-o$;quQ+^{c=n1>xK$_k^OK4>^({S~
z5xr++t&UEbE=9lj#du||&EEf>dlFCC`dvgk5O3N`wui=;*sPG$kJnp(dR1S(os`m-
z#PoV0?|qi<iPt~==<S(|*QQ(h&1Ei-MF@>QFCoDMK{IWVlOSE&AI4*1lMrNw#KZ?T
zoJRKwt~n35^iN4SoB!G^(fe9C9&^u5Ryhua@zowb`*-#I>DPhR7j81$We%6|cN>A<
zWG+7k25kxd%mz~A6RUeU{5Ba|pxp4v&zzEM3=G$N<@X_U%;S~BHLj`H?@H0H1NXQ0
zXU!Saz*tY!C1uIxjLg7q(buBd0=p4<G3jK?99{1-s=s(pS1a3kesaB3n6Z!j`P1)L
z>n(h45+q^#jg_f;z3!9xcBbW9H#->@{*PLl5$>+>g_y>0A&<`CRBKks{Jk-9NF^4C
zy`an>tcoXvKa)Z8?*}M|tlQ7eMi*^v3C=|~@c((?`S{)bvGiC+2gW0fp9ya4MSb%J
zUmk?pLfW{CwcfAEOIKIZhvKUy%liv{S=m{M$6>iQ$AQ89mII5!p0$laR#~hzcmT4@
ze)o^YOBVtu59dgY%+)22){?(mPC3W|!-BvON48<|TXU%`bS`#Y@y4R>YF|~_Wkyi2
z46aJ}Z;J+|yl*Se{BA!6R?Q5}4jA4j-aR<bfv0x&gLKJF4xr+@fU>%WT;K>8Dc@0O
znm|@27BXm*^pd|!bNC=XnD+KZMu|i`eQRl84h<i(BB@-Vd#!$3R6bcx`bq$fr7Pya
zLQSV>%@ujc&HV$ze*O=`-u%@TU^BeEZRHw~oo$NLqaKMo{uB>&Av>vB^`{PX8xhEC
zQ`xAOeEUxpXqA5J6KnD8-j<3_1j{pd+n$Kz-ps_0hl5WoBM&Yva+lNI-{}Vl8u&FR
z+ESWT(e%>t?X_XA0*V*#^_!YMwb!eialc*krPQ%vI%?#5&%iB6>^rDj)!r4W*)DrH
zf0HJ&xn%<H5yvda>R?4(@!aoD?=u%EMIvCd76RoSk>`|yBKZY24ti++qHQIbQv5k^
zyKcRV^zy)>)gRq4b$g{%K;@GMbc?f*9{Rh*@cAG{M>k^DvIk}Qb5ihOz;m-pVuxJ5
zl9#Ia(U+U?K}3alo5@3!>UQn#@5fK>w<MLBq8`?L^f1NhmX(#|yg5vt6L{xSgvKWu
z5QB=akuS{7c1~PNy4>Qzg@HYaR>H3P@8VE<Ebo@S^pe~iGOphL6%A}fdT364Ih``z
zrFSH^;8daw)!3-~DC)ESdDEbYn^-gLSDVM<G3WIG%8JH{X@{mBO7|xpNDjt4rQW9#
z5!WJo+mDBx`X1!8>s^x;b+2<`=1*qT=jiC(;DuTzu~+TGW9Iz@u&4o^HXHc6<DNG0
z^SAdK7Ts@$ozE|nKYLXSVpLFo*%b%M&&>0HUyB7^^wg~tRyiQ%FpE_#h+3EA|Ivqs
z7rI^3(s?Z^9|p^zx?p;`il{+HGw@?qY5tE+F5cB`qoOC-^4l^RdS8RbFKhpi#e#p?
zkGEy+lUCLCI=;x)=iabWT^PF98@a``DQBx6BSts=_RGDaXqPE6&P@y`n6hx~E1|jU
zYK*Uun&aU9ksla2(soO`?;d#~vSMm(B3j@$BcjuOken-Pf-2+43NO}HN~SN=XAMal
zZ>>Eoo@u?JY}iZGoqAI>rDjqyQ}uo8xk+<@k)iZaNBq&vuDJQ=uNMKa5`)d&+uLbA
z`^&iEJ{xt*?jG^J*a2CB4m^IV0WeO#24Mo&23eYFKv24hkyzAp2NzP0#_6!*`1j*n
z<ybR-WaJ=D2=MCLWVwE@xe7p@DeRlS^y7iygNGB#(;i`@pBmbF3+Pr&H1#6XuG4&W
zF6gxq-+eiIp`YVTrtp1>j<d>8JT(4$<BleOId@YNpNqRY?zfc{DdMVKR0?kdOJ@b|
zH;H%ckZJL=Gh%~Nhty-74F7y@kKvzHM)nThye8<7c*pgESzeA#PDj~GN1!tj-s7Fz
zOSu2n<+ZNFzvyG{z=>TEH2&%W?V#6wZI)LSuRcxrjMU8PoZyIyNI+8cq7>+@Sv-F0
z&dZD0&I9iGNlu9+a6$ks?hf2Tde9_mEXeHr3*zcSLGWUph_Qk2(!=KO`q$>4{3|Iw
zi~6{I-4S9u3Ouh2JpTMT1VS>;?8ck=of-SfJxj?Sx`O71j^lw|dVAjc<PhtBzS<?~
zafxXPhDdHK{z<QIS{1!J;yWHTKHztvTXXK`s168fG6uL^u4j?Gc@E{aMn7*V1Az2?
zy%-An3>ic?HFkR!_5B~FzA~z<F50$2aY}J0Rw(Yl-Q8V^yB2pX?pC}6FHqdwiaQi{
zcLE6zAUH4Iz4yn<e?}NPXRkHalyyek>~L0e!Oiym4_~aIB9k>ifw5CnSSV3+joP6t
zZi7`kuasMNLm9lSw?wzBK_4`j%PtJo@|^d|bO;$`;?Tk-#2Ut{m4VFZ2Hgg32U$n9
zzeeN&bA&y4&$HHW@Ev>ED*=RN5lIx+WcKL2GVay?QnQ%}Qy1I~WoEYxExxJT?=O1&
zgJD^s*Fo>PhaJAV`TEAt4c;zt56y>d>(%JTEnI^>Z1{>1Dze^U-nIqaun*S+g{Crn
z2l3^Q?<6SSpS<4o9Ih_V^0n5F+i;q3ul0QsO0C5$v~$;j?lO`e5d^&`jbxDsZF!$_
z4HZSCIZ;qCm}fdn+pzs+izLa4vr}h@RZw?kYUf%(YHW5;qBU6Epg0&<^%2*{j~2Y2
zZy+&Pwf5uQks$ZaJ+H#bd)+0leJ;|l5Ml7deB{!p-q?<Oolv<5hu6`7JCAa^=*6t6
zD(0agEG416n%Ab+w?R;+*9`x?1)kB_pKgKsXVvfkzAia>jot6c1SKK>`(5$-<JjB3
z;azqy{VX?ESVfnqT{!`pg8ZBR_k8TH$Kyd$tdRl9ZqrX^xb}>3GgMp-R;gyh1`HqG
z$<1(@<Hxe2W{qCwSo&R4{Or2dt$T!fiog~_oAIR0^HI1a=ieNn&bT;OmUO(&!OSl!
ztS=g=sGa^d>{hSr-Kgt^n}d<~=Iq^=jrP+xeoJkIn6h;92MGkXx4LaHTI97QL&qq_
zcMlZM1E~K)MNn|V4XvDQNUVK*s*<*4x%)*Bez-b*k8i~EGd#I+yw`a(!}U6b*(JZQ
zP6d;xOmF#1GUtId1ynu43QRrLpt_TNPf(mo!XBW=^qBjbeyxLZaQ^F~`ZxBF#89z;
z?XMd)@3m;q{;1aZIfs_#?n*>*r*8}PBALYzL>}2~n4KYy+-)nYa7mCoZRVgUHZvE=
zg_eceX>yS9@NQwJ9NsK-PY5`5_O|CVc%8g49~unQ!02MaimSRCpzX;HdWn4{tl|Rj
zrY^<NrVD#T*$jDx(Mg%xKmDOV)Is1$bAG&ScQ{ZF0SUw<klc{^`T4;KKrjhuz2nk;
zDzAXR-r{6-Bj`JREVZ!osf$F1fgrjg#I|$m>}9IL%cZ-Z{TesujP(^X!D47;-Axm4
zo1r$QS};?15>Qx=bL;7Oe>W^V!TR!4lV7|?8+1hr;CAEgxfxj&684WyC64)W+8c0s
zze^Bc>irKj>3-wGD3HTYWY?js!xW|;htZKZ+-CnPVm$!32OQBme*Cy3pgwD3{>Oq1
z$#+}s3uPQFPWt$LKu|E#pKI)~rzi74Fg$M;cXW7)VWiKh`;Z{)x0qvI=ZK|tZD=*)
z+c?EkxjX5dXz6ew9qiE)f@^#F%XijUzpHmG)`<<GgG^M#A&u5`$;(6IwKY!A{qPkI
z`T)NkevIo7r%uroiOS<^57CB)nUteanKLR(MBn>Q90h?;_lf;suvafxw~6mPzd?{B
zcLi;62aci>1R2rBLUya=n6wx&G5agzdhAp#OKZ&$U9ABcdg?gb8m9t>JjxI1A7!zn
z%d8+K8KNtjdh}8deVREry?g#^&6bvyfAU3*1?8u+JX-056oZOaWL0=7!mm`Y{jhsE
znGV6Pv?KJho(c(x5LG3nwGKZ^@FEb_OS#HWKE|Fs2}iG}&<68G=-1q?&gnT7!vliu
zJ3V#FvVr=JCI#?DZD!H7i*N@B$SB196<(yYw1By}voN7*h+3P)0P4=|`5Fn?re*i9
z6sjb*zrSQ)r>D>6UIVvx66;2lx+K$W{x`_h=oM!~hQF{JCN0{-PuTg0R(0!M2QYq)
z83Y}o96Bob`Dbk1_hsD0zV?euTMG+V7m#6e@b$1g`I7r)w`Jy|#|K#GwyLidpmL3*
zLzD8|k_C~MaszX#ec?Wymj9@zL^s&6<K)SuyA4w6FjkRFV}9h~8-h(F*b)w7aor-e
z`IkXiOU^W3e0%0L^+b#g%rx|CcmL-*EijW~@3L#X?#~pq+<D|+MWuYok_ml~SL;`y
z1kt>AKiPkB&U#LKPAp?mSelDeWpC0PkQJ_ybIfiDf*K#)t`(VSn<A=yY6K{@G)Hfs
z79!z#a`rA3nRbr8p>9AEzF1vqIeCQaAic~R`d7NJQ>X{il@BGQNtx-50bpc*{<W_Q
zGNVwWq)}F=zW|#5<e0~OL18SplRh#qYT8#!lGTpJ4d<EO{O$C95}U?HGRJg1?qBHE
ztqmuG6;8QT8!*FXw#2GlNJ=kwU)FSR^h60C%rf_W{PHrSg9m~?ZI_nGs(hEj8kj(v
z*VSutw<?c!!~hYb3#F6N(jp--u<91v<!$P`y2j{+nzPQV<`y;W9Us!G7zf4F_@8(o
zQ~dKFjUDtSOa>g}FMg(mdz*ZB-W|fTmv~}nvzNQqaYdU&@g<m~der}jzCIYPJcK5|
zVhx$oD^x_rCp&sV-IMnk1bBN&<kQa;{GPVSUpat*_e)800v24mr!pdgX#_?zRLd)y
zQuGQP;8rhK7#zWT{5C|#6+m^id3PKnHFa^Q^NReFDP}j&Gw=<gR^{IhTs8<}54u$g
z;h&O#pY%*hTL5eSMLYr@b30l#eL_4p(7mWr)bqbDIcP0}C*2xIWV#g!UKLgVO3u`g
zE#Qvp+v|S+<AX#tuL`&d0e97Cn~F8`NC}<l;>19WBg$i<TnECXz;=x@4u(W2zdlyr
zY9QbcPJF>mSNu8Q&29RVi4RRJEE5NRa2HIzbL}myOg5scKXVh@>bVchOt7y%rHlCW
zIqtxHL0tz=emwsIVM$=sWBBO|<s-y*^)^m<abF%;u3;jYKNP;T;X|l6ra-BiBGt@5
zsu+rJvDpVcre3<5Ku!?FdqqGf0&=;x1t>~AS$Q&RT8D&*dF`3-fT^j9ZHO&LIaUr$
zoN^|PSUq=JqHQIm>eszKkI&;|jmb2FS#D@N5SUZnij(ipK;h%K5g55(H)+E;q@9i)
zmMSMT7m9AqvUeG_vKYsicM(im104W!FL`?T99<}hwmwe@Oc1y=w>0b4=mQz`o6Fm+
z*mKbJTM@7((>I0)<K1${R?RR@NfrsMIQqTW`{pB({IOQ(1}AH_0txpw8@zj#t8V7}
zAC`I)m-x!JTRqQhyKh$k!`4pNp?iMadqnMC-F>o@$|@s%Dp#<e`xc*<Hd-PTY0Z+e
zP4B&?rTePsfM}7M(!k8T$28WN>a8Hp>{HU}t~)Pps68w8lfy+2xluU!R;;T-#=glP
z1Pl5KtAK0XCDU@ghV!+F|8g78+`?WeDlp`3&Gvk%4&gfr6f!aq^p(GZk!tw3X?D#C
zG41t9tN`|<Hou1tWCZ`=$h$L8o~gaP;w<85^o%5iie^E*er0>Yr*S%oGmf#(jmP!r
z>hwQt*t6(RipIq$#IV#Uu~bntD|I(tc(vnCcte5F7WH+%&s->GCNs|FwX$ex@t`cY
zKm4V`<CcG|H~l88^$3M>wM^42pJVaSHBvQVo2efI$A#(Cl9)&CJvpQFM$E>AA)9*=
z1Cdv!w0B~IB-rw{si_ib(KOUCmi5S8@Ho4mKn_6cf~eeT01T#1Rp*5`0}639(}uj2
zqJ+|N*gx~QZvb!6o2@%06c0x}G_z{O`A+WEmc|}{NJ$J1$I@sCRcQ0W-?(G>Bc#Mo
zC@!MTAVLIFN(#$x@3^cER&u}%LUWq@&@1q`rTdWf2Vd%WrMj`kp?4vRgCr3<84_C_
z>zy7cso3OCf00&JSIwHa-iIBy27`pEVNYiw4_T37V4+x8?8f+-4nW0b^J$$xuIz|Y
z;9&z)xlX#-<`>KQdd+%?L+%{{k0hQxa6C=uIL<8y05?%NJ(ZpZ8d__Dau-GcDnic`
z7R{J*%vZ-Vwc}ajPyZsJO7xA~2Vcl6UM+j&0eYkKm}FnX$C62cZB(nJv)}`yJbZu@
zoZ8rU4ioW>oGYT0!_8T}XACU9mWuu2G9%Y2;>z2&nEJE#hwL2L)c?*7yf+H?P_8i>
z2g_QCdogM$<>;Goi89nGCID3-OqQtxb&QV6N*!6({!2-HuKEWC=KSv`DiHzg_v6cU
zVIwg>L18L!MJlX+pWhbFX8NM9iYG+PvYj=NDOtILxTMhvF+lm5aoriCQ#vCvg>EZe
z8yI9uLxiqloC{&yUUd3lWPl$a!BtHv!#${~FOC>j|LQ11;q|kgwf)&(yt=#?wgU#e
z_s_-UnJ5J$B!}-#@MJ7^#HE8grJJW~92eeF4WvLRJdv^cDtxOD;)g$sR^pkVu#g$5
zVo>VK;`TW1)#~Ox=KU%cxb63ftW#rDP+rcCRz3b?I~J`tpc`+~>g7Ih-U0J=)o!p8
z@IB%a(!U_hdkTba<>CvOA%M2X+e@aUWy@nQ^3oHX`p{_|`iZ<;LP&r|PlE>E<C06r
zG*%Ry$a9K8kG<7Xeg+-4@65F{PK-;O-g_nJgIToQwNqM}XXG<=^v{Guq&#axbxno@
z*wzZhgEL-_F)t6M{Ue@pKKq^gV9o7C5_#AFGHrQNo!Na|&)V4gy(LZ;nYZ{Wm4o5P
z^yeEhbYyW$X4$e8kMH^pF>9aK=M37{otB^R;Y;QAT)=@4*v9_C16SalPHz8j)GS`Z
ziG^-1Puy}Y+V{4+_o6dB-jF(98!xKKoW1vf`$(Q-aNOV+ez{C3Eh|i^E+3+$g~4Dh
zEBA%bY=TgOf}8U5i0vo#R5ICO=En{s#b1Clc)BVyIR<wpM<63BPe@79<Tyi9OKHrO
zjxo5t+`&xUa<Jqx6a9iS1)S|cGZ$lL;nj>`a)Ll<gZBNfY}F(?rz_X~w#ApLCz<I8
zc8yA17of9?GRhakQ)qP7K33C~P?+O!S%eU&nJ~t3r2{#>cl3qjdIf@b^xC7!F<KxS
z)bytI&Ec8q*mEVjV`(7S1iE~pMDN=y%&XOuY3Qv^w`c{8;QbTh*;$;ej>5x&Qgx7t
z_^W$KMa#;@$|1O`gK7qzwE0>`*GVzkER=GY+Fl=)dSdwdTqt(o7t=3$I9Yu|n=7ux
zrR_?TqSnv5a^aTNGlpNuiOe${ybizXoS)&5O<o}(Df>sIQ)sH$iC>uvPsMX!?b<jl
z5yVANrQ1C`RZv9)D)taKdAmqRj$JgG$~()}KEQ5Aqa>X;99Q+W`2JW%q>t60E21v9
z#j#7NE7$Fu=&m<0ko5w2`6i*}$!&7R`QQDg5ognRbxsZS>J>Jds#NiEc#PJ?P$BMf
zc6VMzZhJi({KFldiPLR<!(cYYQ?s&s&Tsaem=<GP-4~zp75V;{K08mfDN64vlK6j^
zxJz_>j&>4+i`JU#|1|N`QHJ)}J<*GCsr&cH1Pjs#fw9ZDfD#MdEJbB3>KI%~c$`~M
z*0R;}8O783QFE5frGt+Ai+&t;7@^JSXKGbqEp4a6UF~$U0XjopdihU-;UWE>3m~NN
zUYy#;V6L-p`T$NQejNs637QiDq#p=qfPY>>(0kx$D^VB)nK{#>ERI<rDLVa|)rI!f
z2D#DMIJL0W?oWT7rjB~1W+wH|xRP_^LPw(+m#Sww{%#aWC@U+^tX_;oii|Z?=p^B)
z{*2<QcO-Tcym`BMx9c=la|B#(8{e^=wCnzDG?Aqlku!rIYbVXkrM02Dgz*l(62L|~
zHlf<k*|z3zw6xk4Zmss8skPI`Mory0zDqsaPX+PhS{c|r<u5C>O{7&QUPiaz>FeuX
zckNxfC%An)dKU>xK>c!L>r6JkUl@VikVL)l`kMG<6VuZred}ciFspM0Id_Ack&*j?
z^=R;O*K%CD$YvfBh;>?Qrj=K6lzq7E%aaRosRrmeL+$<2b$VV9$va^*<hcC{H^O3E
zOzkrLjvfBNz&ylJxK6P4=^ceMf+^?cG@Tp*42>iz&E}ydW<DD5`ss?#^J2t?69xRI
z6g~p`2EIq^dawQx^)r^B@W?P2mxEp=u32-@tTJ4g1wG=E*wq?I4$$R!C?0P0f<wIv
zA?d8E{s?c8@ljNBCTZI^@IU^4kume7^^tn)Lkx|G=D70tX!fgvzf#;YvLl`HwZhf#
z((GH^%dA|k)ZDM`2ziCtWyCYI-&o%4HC1HTN^^}L{Oa%j%6<?@K<KMb-TG_rV-Yh9
zY_l0?K2;`S)eYnz{m;)+@=;vWp5r<Pd(%(+$6Q1DjW33%A&)_uT_l>mB(dQZj=#$~
zWk$UfIl|1Zc|UCy=k~K!zeSE}28SQ_Gl17~kr$#sXiLxElN0()K9+OvWiavd#nm|L
z)meke<r#;Ho9!$6YJ$(<d-B7)*QaSVNV*{3#_s4wh5H)muv0+J;0eWkMCei8&ZxDK
z=m}OeC`w79*k1PEO2m!bq`niWId0|jVq7C-GxN9pdc!2e{@!9n+3Sg=1j481ek-<L
zfT4WNVb?aQA>H~bba6iS`TGGPOso3hfwVNyMO}Rn-2X;?%(KOA=_=`CoPO{XzSrUK
zQ`hU$t0JR*rZ1Y566D(p&nqY>JHzPVQ_s_Q&(zk#+Uo~l8EKDrd?^j$k4`=P-R=hb
zFQ|`e8PV3iBuagt-p{hj8>>F?ellB9=*bN}YDST&!b^k_hFWpIBHF9}l%txT<$rck
zWsCxFWI0^!AC&$N@@p3Mf^;@1$YZ6NoT@*IYqo8M21>t`LqX`h=cPg_v!QZjKruvh
zI0w2opIpHwNVuq2R8m5gHY1cQt627j4=WC<oRN#80tN6zBe6{Jl8yGKI@N&eT9f7t
za#hS1xbHoq#ik%Op%NQ_{3|NGzI|2`Q6o{+_mkazoI>*a&2qbyy>&LwEh~7c*tlNM
zLn?;ky+ycN1QU<<Wxvh{m0gvr#OOgE-*>ukTv%g$W?$RSu2ZqeINX9L?Q%k0bk}6X
zQGBT^3Vo)oD#%>|M9S651*X+~rxKLFWBkZ({4&Yv+N_Jw$2jPE?tPTg@dRg2D2M1g
zJ)6CES22w4k4x`n$REUVeJ_ii4r^ZCZMmGwRr_BQS&aJ2Q5*GmA@yGNmgZX7d*$_a
z3<GoU^(R9^0p+uBVai?b>rq(Ja+klThCWT3Rg0sn@Z~y^@oqG2-rrw}LTd%HA#MqC
za~gg3$g@e4dD&B`c)%FoRKtw7%f{4ew{f++CHJyO`(-68)Nm%a%)9xYLl#$|z^xOB
z^FdZ3+iLrmMZFYiu-BA!qp@FF1-X9$`Mp~ZHaSVkC9auOplcuN6VI0DDz%kQl%2ie
z>d{y)-jEJ$V%Ww~&*?qpi`PHTT*nQgR6a4;(LF<xI^7UtX+{Svz-J}z%Mar)|I?(N
zS4<dhtABi=zs3L8As%yxdae~j$2hh0CP~DphSck7OP)Nn3r9{1t=XQ`2{ajQ@5T2I
zGO~wbHX;=erPIsnLPaUf-(o4nq^`>1WGONM7)U-!Ni)7x#6m1UCh{MVOpAdF&)rS{
z{`B0)Ii{&wtSh!9bPp{+t|D7W`*F~I!E0ffnLYd4w?%m3dc0Vx9~?LEoW;LLDQE=z
zNbCWQ%BQ%|?J@Sq0F4KL2fKwsgeU+Uar&7-LabtB_3968aT>{E>o@njaxqxRQC!Jd
z)jpt2qKZX|RU>5OMd89!N}5n_(}32)sY@N7$3I7jFaHuR=f}zS=JH<f*kjs>QnZG&
zF#xvn-b1%(_IIiS1{4Sh2?^e(L)2FK80NG#Tzws)Yc~$Yo0hBdPZz^IkpxbHlNnoq
z*%jokbov7N4l2YU^OYH;yhnub(y;Mw>p}HKlG?xBSu*Lf9zs=fjfdTmr$dcF&k3;5
z4fuaUt2h>?SXc%(j%;lUbc0gPNCih1-BFQ+so;C57UJ&Bw{Bzrh@Ms$+9<S<Fy4nJ
z){6(H!1#taiPZ*FoNDL<hYD6pwJ^4)lVFdT;0;OM;0{~6v8Q6z>~Kk^pbN#0xk^-)
z2#kPI=QjKmY|UPH4Q~Q~|7il%Wf3|n@&#30NyxZ6=3lfBt1K>LR6|R?;z6m@CTKK%
zl_PqLxREahRW1kcBJ~-NWF*awM*&jIr0GMk#bU+vM8!xk0JM_SW_02imAGS?aKpKx
zr1M8JP47H5{s#WdV#D(<K5UL7S&m)O(c+-OuXIqLW!#BBP>!m|P61UYttl=}ypp1n
z(_%KcCP3*@LSz+-%CyqtJfL^bC+oD2-SxP^-Ei8h?k}IrT61I9-tk40b*@i|uJjli
zn<0xT(Loy_Co^&9C#i;jzg-?yo7VN!BS$>+5;Vy_-{VsLQY0ef&z|;quHx}I7^GcZ
z-58`*ZKg?KTZk&pV%zK(u(2i(*z9A7pMf_bbNKxHUHG={Xn8|xVjA0X@!t(%8s)S>
z*O24fw&Cze{oa^Er0_K`w>p|#j81+(bvFFl6UiO4XMF8*1T|sSg`ZUkB`n~qpKbV&
zZui+_Nt$iO#NbP_5cWIcha2P4bbx?>PHSgN<Z=ulb9JI_g-ds-dRAhObIzVX&DBK8
zW2b+cwBcImcb$n0`$cQoN|nSvG8tqj9r{Qd)kYAXz3&=f9Y^HGn_rHg``dv(KWY3u
zqDIoec|;2Ku>OPwzqAGLioE9rJ~Ul4Ceko)P8-U+Iu#200U|AOiC?sA)430?Oahiw
z_5$rUujv~$PT>;Z4(I#dv$0=NQlx7oFwJ$pm$U1!)4x$m_BhFj(oPzu;V2GA9>|1m
zhZdJSKn=-v?cd$kQ)k=P^`%+VzT1XxJ*K1oN+55aA7_J(fonAf(E6?iQxN~?pd_vM
zSplxoaPdhMdP#*9jpg2CnvI_Azmzm}dog7w!0mQ$Ki1x!xns9m%7WdXx8R|Bp!}4o
zR!j+i&@e-dfc}6!)gZ3*x29u+I(*e6JJ}3ce_`wnA|_kj);Bq#s&JBU+^-EQnO5^~
zqu>WT9BC9~daI5ONwv(GsA5oExlH&3c#{kc#NbrU#nhz6EzFv*>SQgKgkxIh5&;${
zbP+xskT6Qyo${5ymL`RLoV4xOHb(>u57Pz}_B}grY2@lsVxBeh+>r?V&=c;N7!xkP
z3$<<;#c|EP1d%Ptuowq~tPQu^UzY|&+rqIpmua(GG*E^~=c61M)uqJND}Ap2!;fkI
zgA87Dulq~L1M=2$UkY%&WAx_vyMK#`kL$>O87BPNRxLW53!D7uyNKr_FBl_$j7z=8
zIUQsFjON#o%o=cj-J4S|Yzl9+|M`9o*0mPVaX4c#IeudQj16y|Ee_ua7(hP{fo`ca
zRUwPQeepErRw-+7r5XP_q}P5k_W;z66kZm@x$kF2ONyV4V@FiNcOk2URAJ=HV!0|7
zr$WL@eyJY;Jm%Qg%{^s0{I`dY?hjHDp=)x<D553Jh}j%k3;6>{anknszFdHxW>mB)
z$O%9=1CI|)F{hFWlhs-gyQlX*Z;I&_pw@0%&OsE>6=C^L-f-yNb_mo)pxG?ayo}Ce
zRlcOy?%Z?Bdt^59Ilu$2xA*@c;#gaSgH?i^`Pwn{GE=`aTpb2AqOVyBI<P1P3E5*L
z(PYyl@u`p*d~>@vyXDoj-PM~4?~Jlkv8oKZ9-2hrA&C|6?9|87NH1A_XItMN7K4-7
zJhp3iUJ#PHj*juBibq{MpQEalySBV%+D0#kb8-3I)tZ-Xd`oQ+YTAOy@5@Q*VkAG`
zBNJiWW(MlrtJ8ztP3_<Z5;Ly9_kI~jTZ!Hx%keS_n=jYa7o!F0+wWC2O-Kb*w(j2S
z`R0LVh4lrH6kA|?eEPM*0z=afXMc!N`-PnFZGcFh%4=6otRA`f+)ryK1FL2(@rCqm
zP*nHeiA+BjM<wUyO2Br^M;JK^uHsSGGP~U$BBd}w<+2JK@aF0;^SenVr{#c=|Dtxl
z4t;?SUgn$t+(?5eW(GXJZ8u{7@92w!lI)v&__2JWewEn#M?R<T5e4euwjnq}ANC1Q
zCLTQ`C3D2@s*NnQp8TniFe)j{QZio2d<5?s7s(met+rv>p}l_-{B~`v<Lo<RCNp9v
z?$7VT($_7V(;>5wII}_`l?#ebKa*;Cv3}CURn<11rGETA{os(GU$e5Y)^Ay0>}7*p
z0s=@Eq#a&!x`3(#fJ0o?&<~L?a6!}I#l@#1lICXKBlv^x-h_Rlq<Yu^0){WPF&E6e
zNKOo(?GWgIue?8Pmu_rxwq)cHbLmboj5tAzxSDitmjc9JgneQepm;-{c9oz_L>;1@
zRo;9J-&PMF&(R#Grg>YTlquG^#fx}z<<bwiWg$~dR4qfX)`IG-!f@4%x%>8DYxiQ)
zu>0wN6_A(b;_Dju{-CVJcn4I|Gn#i#{=5(?yb!7?r0eKgYkZ^B#W)r8_RVoD`$A}*
z+sDiMwL}JZ!dZ(e@4m@+Hw&|S<OMA`Bb}Kw;0b44n19z<Yo?5`vzl1jcc6En)V7*_
z)BmP|Wa9}j@BSq3aB}*YKj^kJFlI?$Qa|sWfkW@&MyF?}c!Oxpcc>T1h=z3iuogm&
z(vfl!O8NV-gaQ3qu#-AAa@=t4N~OfV{nX=CuF_!`ocf)GlkETLbkl8*6Dc3<Kg$-E
zCAmf5j;j~<Qqdu(Y&8+csl-T`*o??pj?^8c<h5Tf|0maqq9+wmn2vl*&6w?xM>S`V
zE-}@N&zvYZJ3qS|=*94CzO^|e5?FgYo4Wn;yYP>h)_n>F<)BeL#LrQ%X`A3>C5|dz
zbJYF&hKZuol1vDE)0nUaVbLWZgzmIG;+R)hSlS^TExGGFx3s!bddw2@`nLMQKJWfr
zMe7iEdcz#wkgl?dO5}T|*innV{Z)K;5jNm&73avUZ{m``K*XVh85T;04Fss1M&+v}
z#U=rafl0+)JUNbPi{|Os1m>NsKJ*7>&=_)QF-rPy3i~Pr2<W_}ms2+4-~<KW{SmUU
ztKPFK@*wgQ*=ST$Y1$@O1kyLq;Z7e+Ejz_pu#uc~PDeNH^cs@sffjY%)g|r%gYsp#
za0A^Ay<k;jVPG-0SR^r_SKt<Ci|ol4;?+rh)@-<Y)8aI<*yr6j2oIW0F2M5x+u^16
zSHx)Gjm}Ais@A!H=<^N3k1A6mTK#U*A~L*fgRdaS6X|uDwa?8(ob^ZPIK_TDyfXjk
zl`fpkyK8H`J8*JsM?-^T830fY{mgjuVF~)aHC1K&Mw?%K39HDS#hBC$w1};-MC{oU
z{=fGebWcRDG}>$Kx<7Tk78=#uC^80%A~<hUFHr5@Lzv*XOD6Y=g*<G&8v8ln7wP@n
zfITY@bivl#$^sze)u+YnekC2&5fXOXbCg^m=%eqb$A6goWkc`%L5_YusuK7w+w^Z@
z;6=sT9zu4L_o;GO2-I#^qk6J&RwtGv*1l!n)6?qm%F6ZbgUzT53CVS0*tmmhy<Se%
zUK`TLLvx2D@J?p_HP>wyIIW=tmNbPWIfe+gk>%6rlGfC6^G{mT_c^m!1svd7>rA=%
zo^y6}s#K%(kRp_Eu=#wjg_xrUCZfhJqH5V(iRBsJYmN6j=p_1h_av(G^<VxIIe{pT
z)&mQ5YUvo`$};ssx7nsTaD3a$$ZmJ|u@)D}q53^HLcfoSRCsury&ppj@efj8YE@>4
z*|I8~cltg|<@Nk^c8_6#w&kU%A5+RSZk6oeF<?|wzCCJb80b}}8<6AWrc?I+vL6m5
zppOx@+G(&E42A}iVY42$q-Flh*?4Ep;BIuWPSDeLG5!HW0N<)M;px>1-9g%CG@k;B
z41(QL&JAM+)rcV6y-b}QlDKnvQYQ>CrsPYY%ZYJTsLE@_Yls1(rze<)=m7YhY03vj
zb2V{<Q^R>^dR%ew+nVCUU+;#eBnk!<)r5B)(_MFhTd#}cSKHwC+S*fBr#d}<J7Osy
z0Wc&WpI;X766<AyU$%qr=Jw?j{;HM(GlL_r=kFvM&gO$nN)}*Q&|HOc$>;?emne;w
zo4a|r4f+B%kdLL2NvQ#{Z=HAT>_u@TwyDjED&^zqjytzSUSu?0=7M&4gAULc+(<n!
z=}A3RrrPx<t$TkZhnt9KKoTyx{Rw{T5!>Y?%6uwzF7&xv7=LOA3QWkwza<kCR$CUr
z?F)KVc>Pg9Pz{)(Mfa;v90GNxyI=I!T>9zSFNVKvBH<4r(kRLz)O{H$)QWJ8(dl-8
z$2&3%11|`&pE{fjxBWy2O{CN&o0m8J_Z3t>BL4aGX%|iM4%Tl<&Xp5<<#1M%I4tOG
zb5b=?Mskg4uAJ8<nrn$JEupQgbAtJRE7fblCV7qg7l}S%J_);?Z?HG|<<Bxw^w7*3
z?cVU->z)Um7w8SKip$QE-sVccSxXO<@R#BNYnQL{HB)#}dS>M@sQtzEJb%T)i$q-x
z$GB2iydS|bonA*QS|8CmB&yNeX>>=k($Ez7N|#twwZ9GpyZkYj8}~(TcxNzbv;Vm9
zgW=aq2nFwNQKk3#n(SB_>7PXi3hg;ak+muD6J+4O>ru54JObG8`qOLciC({>))z#L
zKN$5`(=*v~-ybx3y82nA=NfxY(Yv;7(xfqjk&`&l^Ztr)+PTFaZGmw_`YW06s)@Zl
z!NXwmU+|=Ug+Zr<_hYHYpZp~W-Rs#8<sZ>G$U1bK3_c(sMoN1v3;lLq7UBbeHKrAH
z^;b;I1*`>gAN~f~2l$EmJ!^}ADvsU(Bz_~~>P@ljNt(e_Ew-18?))l0F6CqYLv-I;
zQ!V_%05`=}wQwS^BEJdk{Sg@SLwg2xd?B3siO@9O+FKbqf8$n}Gj?~rWlYP6T3L7P
zCp0!l`P&TWm(QCQ*2gcC>*MxQY&m=CpC0CTTx^Du^P`)!+Tfd{+M4-o;JdV<CTV-T
zD-+OX1GV0hM@y_|t1=7cr%ak7?#=nYp5ccwkHFix+io?dmod99$4ve7W%f?PkWh-~
zG0jRKUw^FeJ*tQ=beOKR)&GGhQFt}qUEFCKHjpr6ey9-(K#S=Vt*B+@$Q-9y7uZ<(
z3`ujmiB4hqK+P#9@$D7sjz^gIw_8$l9@Ff8gUCldhBKqqYtGW?jr|<IMBbni{xy98
zLD+;6xEsde1@0W~@Y<xBBN{^~dP&mLbGbi5Eo$O8Th6=Ofb<Nl-e+sVIHn9+$2eHm
z`6-qsdA(kQonEisP0r$_AM8K8>?n=6OV`LmZqR-q^;~lWg1}j5d_6ab;H#kLsZ7>D
zmspsHs^IzsN+eb0UgREWrHf+e-oHgzT+w&&e9`dhMWmYcR%4eWb9T{+=*?(YA>A?T
zao`Jbwu9G?PrEw~lM|J0gkcPsxZ{H%U0omMK0ICK<sNOynCA)h6Fzj2m`BjzCDh1Q
z(wE7j$?s_fd-T}|ns#b`DRkB%pV1K|NyE_pE}3%UOP5AbF3JcA6~FM)2))`fnzk%n
zN?}7+J=<beP(Y%m34h~#IY@9n4)I^h2r}fCmQ*dA;qRxXiwO~&ueilA-pijFeQxQ#
z_wqTg-!l0}eziWw<Fhx}^2gzQep;Hw^;9)<F7KHz=?$tev5`IAJL}KdC>a0DM0WAh
zu3H-OPAsa1zx8p`y)r9{rb?IXt>t>8E_)1SDl9ny=v*gmX{Og(Cl)h|i+RA%yIw<X
zk?!H!{=}?FYl~;Bg$w=$?Cz(Vt!?|)V|vX^>pj{n>p!=mF*u<c8OCJLDv@}rBTa13
zkGnyZagHkk{xjjXC$)a{iyQFJ2J1!3Ymw4b<#68X{byogOC=$Mo$HggdzW909(Q>C
z>2GTOMh}7uMnr816e|p!5G!%39k69>&WW|O@TT}m=IEiW!GMTDI{&?@-ZZ?q72IOn
zhI^o%o*L2k;-Wd3$fg-QYqoiZ7<BPCEh<a14-~Ukni<cUD(QVF@|W>oBj+1-;I`(?
z&2Jzn4!fr;tz108z}Vk>rBX$5|FS)>ounF4PZY&%@@t6~`7+ZoJ^iDqum2Mm{&pw}
zJ^bckPIf}{w+9<!t-gLrLV%DgmD3<x1`vnA$qw*f2F$Je2IQq{(e*YoFmG*IXcBIA
zKj84f+YF%m>n(W6#D2`OSsK`GlU3}0Ivpz7I?-6PybSwi0=hV?H~(4G2R>b~2X=U;
z^nLS<<#TMFEK7qc$sTEto=bjc*7h+#SFfN{2G=LJjv9U$OZ|78o@qcJYobInD<`yF
z&aR!)=w9$S0SP~crme3=^jmq;I=3Il-*kkP93nS*>fkj>4^%znTbX-X7OaxvgK({X
z&J+_j-BUkmFT$_Yfm8_OdnBZO#);mLNDfXE*~}L?-gBm-<;7vA4i9Ex>^B1-g$<fb
znRT0RXU~vdk2V0yfnS?Z<8Q^V-OP_+h^dx?doqc13gd@+UNeFgUbNQ#AYX$A!Qity
zGknWvE@4bDe?>L3vNN)VStL?=YTR!wupH9sC7R5nR7#R_sH=izpsM$IMe^dC-`nMr
z>MV;{3BzS<7Zel-m%wI~qtk^7Ab&s%rpVnFQ{R}HfE&=$eap+bu?||{(~Ax!{t=@d
z-^D%kI|W>FX3(4#T1ok*oT5hS>2F8y>he7>ajSO+$84WY)%mfQ;0-Zpj7Oe&5?26>
zf9=(8OgnoIv~{tOh!bDj)hI(R6TY#sYgn6r1nHSnnBJ1M{uPM{fm#P$$*gy{MMjdZ
zoBM&VrY2R(#hK838P~V!J8ycqec1{{R1LD@T0QMPR5idagS`q~Wi<ntaQ;2UYaVn8
ztbAem7*036C36CeLgznw!6@zrj~aN9CKMj!4S@&|N#vPZ1M~|P<{^T}yqc?5R|Kfa
za9^qvV)yHFj|qQ{?`;w1T6Ut4goZHAVj;Uhmv7p_f>!D*Q~A%04a1>50K9eNoPYQ9
z^5pfqt;EEg5hib&6cUZCRH>o6JAM&*nc8~(`jQv)+;V@JE#WJb2>&ZjPp`M)i}YnJ
zicxo`hFvc|m{>nUQ<|Rkl_Kl$zZN8g%l2z_fjf5%iQVa;ieg#6?BeR^4xE+m88npN
z(J6tx(YwK?i@n)F1^?@vNG@W8yOlMkC4J*712sMPGa)*`e{gC3{j8<<mOk3~<%|y*
zT_f&he?*tXu(PjhLBqk($vg+P5<m2H3MjCLqQz#;913XOf)S<D>de)QEU!&0zULO&
zT+$X{;gdi6h$zIw_qz7_BIIZk6eIGu=ykUz^6aMiOD&g!=m1xWJcqvV)aCZfGB;qy
z4agrvltyr@IF}8++99ri&en9fn$`N9YECtkk`QBt<FY{S1zkw*wh%*wxBst3xh`$=
z<|mq*KJp>W2~7^hGFH}{*Vh56AzWm{)doD>Bl9Q&@Z$7W0P+;K;;)=cupR*W*pa=~
z{V55GF3+x1^2U0kJ`Ov!m%XASQ*Ur+(Y%=q{;Ck6LFeUuezrw6R%R){{%#r<Vkp*9
zURyK^`{v!{nm<fnX{25VKem38oMCcI<Gpn^UDlLUQ^*G~#)5WJf4!08d6XD8&dVmF
zG7P_On+;irK1|v6ahF87z`6NU#nmMP?vv}s=Z652bpbDM(vfiO5F&Ku;o;Y-!}IT(
zzlubbZqH3kTp~Lc!C&RWWU`6<{a?xUMt{iYC;lrZ$ZSGg!f^PeUP52hTvBraC`%j<
za>C5){F(XJY_MuCxo8Lsc>x9W_dN3i4U{#(Q-a&?Jw3@Z((v@B?n)7;O$D@i;a_|8
zyksmP{^DJQ3wkCq_I)^v?yFW$xU@miwLS5y)1_Lvof&6jtZ0HuW#BDE{<GY557=tc
zER62cTQ%p`z@aX`^o1q>F|m>{w6Ks4eQ_-Me@^=FkgVY(PB706Ql#W>OGD0lGEg7|
zHL*`r2Xh?5W~6$(*-Ms;#q1~BVk-=~7p-|oW+9lWy9&7vVc0~w{t32>5i4d>ux^+H
zgIiYpwJkY?<7YX&BzOub;ti%#?WGJuL<OVf>Rco;HeT~VYq40=&yRv<<iO%0A>GOk
zxQj$)cB7%X#s*J?1NIN!7B&eo!ws>LM_csr%<FXLtt!1IFTKyc##SLk)Zq(cU!6mq
zh@Jz3-6Ev;Xf<ApHY=B5ODd}^f+v)3GN6{F`PW-bN@}@P8VwgioZf=2^UKdu#!u9k
zWZzw3{g{8XO)$PGdi$yo?yYoL{JQtg*tjS2F}gFUVUijBVHV)OTk8j2r^!X<EYa=3
zmmvu5nf4ZxF4Oy8u%iNQ_d9ferm~pIGvW8Oyf_S5sP7PlG;HGvti305>T7Gd`d{$5
z-d2H*_#VDvaw#t`nu+!J2JzZgh<v0J&H~PWZs=mxX;H?Y{Ic&Z(HU3>l@az3Lb7BK
zAUHT;mhhK+bPi_M6bn_oUgGrjl#Q0E+=)JS4PM%KWw#M03yAm(*wc1>NtR;{W$iA}
z`c2hLp-cBUPkq_g4kN(=E%x74^v?ETmHy4mE(`>+KLdkJ<ED{ITnBA9#pMw4=Fh{t
zMRZ`je8%qm-$ENu4Z<s!=1$`Dr#E+-;L@6{h^h4)QBBt^e@6FZMb!(df?MLV!`fir
zT%5I#Vzm#6Y#w2vo0yQ@nXtz*qR8_KDP*v5k@lYpB37jOcI9wE6TX&BB?{hrp9gFr
z1V%1o0-lKO1D~&bRKvIa-5sPYB89~e^{|RCYYK~&joe7-0!Q!!@8217lNs#{#go1~
z!wjlmt~+<VFfZ>z>6z2ldJ=;F%CpgZ8Y{==cxPVV%^^}TJC#Ley_p>j`VahYiCX2(
zJmaT#$sc9yK4YtpbVVE6I|>GjWrrX<38TCYu+HFOFPm~=js6TlC=lV7QkKG><ZWvI
z^)?<#MPn_SM)1tJ_)X-^&`tnbuO!X{oATJ-<G!xY@ID?Rsh9}fVR!<H2puuCSfrHB
zwfi!@2D;YsT;r_$C$&g5q1Pl1W3ckcL!!CrTTqA_qCJMZ7z+`N22-PiT=;QZ{5ky*
zYpz87J8P11S6*}P&Syf#0+q%|t*I;oxWl=+yT>ip#TJYGK&d0m?)t!3>3eR>R&jTG
zQ98UJ%<(v04zJ|xF!WmSO<C%is}`fESm4*1D6oZ0p^DzSHWZE=yPQ5rxz5m{c1#O$
zn6l~CcI=tIlMK7GnAn>fj;|l2D0E3z4g0`<7U=sQI9%Gul=*vvf5mc|2zGM2FH01o
zc|kvVY;5R=PUE1(?uev(zsMVR$)ozMp#P0GV)^Mu$w5o`B0K3~J2OHh(r=)LyI&MT
zsXu>AfkP^H&|Eq3-vAT0uhH{QTK^p8k4TC<nS=ncEgF>o<Jc=k--vzNlp9$wDBpg;
zk3<Pb%y&fUZGC7DWYoUiiyJnFN9Y9v1b&cL4kz+-*^cnVF!l2t;kZU6aSLEjxl{w_
z`AE`qB^eR<X;VllO#G}f#+RJb(T7v#_?k!P9{32RfFDxAOSn?lz)nh#JqQ|oN)5!R
zs5qj%XIzm}xPZiZ7k-G21Nl)~1P~xJ+F1Pi17RC_2v@owK^zpOUYK9c<XKWRxSO(1
zSJ%OgCH-{}%yJ{hsfBKKrNUL9Me3QC1V?jiAjfnO$$5I<Bn}#~;t{OPNB{VS8Jg9H
zV$rs9isxol=SRn$bRbk;dBKyWYQ83CqAE^yrLYv=*nQ}#!B(&nNuLsrkx0vbq(OWA
zD3nU4c<kC0@Rd#PokpQvkZjK<1pUDPc*k+bwD420bC`1tMlweqA~XiG8_+^-2r-b*
zeT6EU2=^L(Odh#O{@#QU>sfT>x+RpeKNeb%3$I_m7EyC7qZXAu#o8ytGfig_8tkoc
z<bc&yh{|0n56o%$Lvnngmp@Few+~hkt6WZjY6|GuTrN3T*V`l9KAMZwMzfaB$bDXK
zPzeEl_jU^XI48$6>NUO>Fz>p*e10<uC9?v+#Ef>LrBG$`P}7(Lrd;0+5U7Y*>XL+O
zUssxScj=3e0&&k+<a;$KSZEu|*eTxYt0o)cYQZZe6`Xnjx9e+pxhVeqfQ`>UU?uPU
zat~O_{RN%lc@7$~R?+c)xgUiOBNP?*t0uzHG$|v6n$NGUlQ!=Z+?(X+=97mONpF4r
zJpYy9{v-8dr1=2^0?`$-jgHOLHhoGas!~?i*N`N{4L}T#(i4_s*E)(xfGzRgtaxwX
zobgormOsCF1oFAXj9SR1+m_#l_>l*Wzy&jIKGvIXzS2`@)1rfepoNoD4VdF^2c&zC
z_M@;WHRPyUXtrCs(^Ih@J&-+Dz%K#KdWl!y_>ysYYj<`~qZ7LG-l(Tx^r3Z{z{o%S
z3Os#`A+z4yI`tsR*4`(x*Q{3@7GAK@zQP8fA=Yl}o|nHu?4;8D^ZlJ$5Z8!^IUoL6
z%&Ur8NOj|46uu-Hlu!4lVN8x+ub=Xzuj=S-ma!%xJRJ^g6NT#?z8~Gp(Y%YjFXT?-
z2Z{30q}0Pjm7gCwXuUSNyj|egKTL0D2t8t#uv9L9m9g45%hj%Ps2ELR({CK~ICMZ?
zDouSG_8%Q0>i>_v&`dqzPo+(vS8x>r?tI%XVx07vE~PsqTf^+m)7O~*C#*0d<ATP&
zec^*Nt4JY!q65~<dX&O0Pw464buYr@Y8K}Fa$h;t^rex7NwEr+7zq<;;WwFP4@I8Q
z)61*X3nJXnv8HJmj(K)*iLQoJz{QJ~O~)Jj2!n9sP1|=xA$N4Z3AB-hOvbhC6e)jK
zaa0O+r@{DAr^{(L;!F^E-FOXMDBOQ)HF>^;`X7jRN2qNcjTMhS*|3{r_Gh`3I`bp#
zp`;;?B3y|k;=ocA{VzewIv|I6!fg)I;*|SVf_Zr1G|^3hh2@ZfJocWtIEr25zqbF#
z5-Ue_2{G<|#tDxytM~=&Eqd%M$Ng3YKjlbFa`VZ|Rxw3nG&MK<H`N9?@YXc?Uu)_B
z3kN_c;bXz_e$W#^$C<<JNMpW7p7HBxLtM<yz?#HGJo)XX4`0TWnLIIdvPDt&?R6FA
zL{WMw@O3q;s4007%(H`4Dd}u+!NBSF-*fs=H4iB7qz&|OMuiMBad%LSls=LqNa?Ak
zBe<zQ8aVOW+v!7ey2;O&#wEU~48wE&0s?CjM`*XVw|#BjxBvL6{sOfw8(n^HgU!fb
ze8qer!F;2|5fquY8Q<2hhR%NI?}WIbH{1p#fD=C0bMdgO^GQ|sz#^7c1>aig<gn`=
z!SDd=@FmT-InBpUoj1Aq-7v|NeOugJQNm1#T^=4$m4`hsqHy?kFBN`_W`N<$`rIVF
zT)(OqaqDhXbggiXA=UwI1k0UCu$OS>_-YwmdHEGu;Oyq!0JsAZU%P1Dn3WixC3^Ze
zX9`1+`=9P^)ZW;4B=VXma^HVrLKG@Ey}ar>uvhhtR^rpntpZrm{J&?@e#7n$xq((T
zjXHFWovwGU#3Oz&Xy1k0&ir;YTUiJin@EgO<ks}ApUa?@bm08UlxciHIj+{h2-<5v
z|3HXo?`l9mc{#)Su<7}_wJzFo;;@Wr@fl0^Ll_-~Kt;v~TN{~>uO^HS?1pzas_Ab#
zTOve1uApbJy187!G-uH29-p`2m)KT^dq85Z(EYlac%QTN7%-M2W6<gRLCBz=m!e;0
zCFc90M{EVH3WuXYNd|9s9mc5l()W}h3%oq!tEan|E)p)!(m5f;3tNVR=cBAC_|gRL
zZ{^6nhEt|}A*TfVrhavk4%`a(c`3mg!->|a({1^0qxtPYo&DgHWcY;+*uXOi@W>Xq
zFG2;mQaN}$y(D%|uR#n-newQ~?UapcX+QthpsNt6O%#y0KiYH+b7=u*UqmjJ<&Ny|
zH8{Q?&<a!Zo2asr=#7OZ@F1$c@!sZ?oLuM$<;fEC&xG1Y%#d>`va=W@?=W-xIgrG>
zgdNX)Bdbkk0itpJre;bN7w?;C2kZ>?xED%CMB4kyT;6Fu$t=D5^Nz)^V`{A?Adv|c
z%>7MZGZ6OO^9YSh&^ru$O1)fII~DU?`Ea3HU5eGj$IlSnna|nrR74%WK$0n1*FIVc
zm&NYiS3NkF&tZ?1HC1&`?-~zdEaQ4EozF!COzXMi5+G4=EpC$7`_7L^t}I7XCOY&=
zdiZAz15I4$g!K-gnPs*AcSK$r0XyL#9M)bdt0V<!?_4_JF%-?)&z6>KbiYgM`l1J`
zA3&J43gr`bK0b`H2;vTk7N)$BTzEs8y7p`(<&#pAZ7loQ&4M@c{MOSqD4fng=Gd<P
zIBQ>KoEk>cFv%2Y<W@6M<w|>aAYG$$To{A)gYczvCBN^49`%usYdRl3JoIBRjQ1g=
zKJXBS>ZF=||2_#C)fO&`;M$}jpft>h-Huzt9=AB>JYwDW+ZMMl%J9wFpyw9%+ffq6
zau(61AEes4cN;4|Y;}uqG84QcHK&7uj*C@^@U;M2b*7iTiZ+w?GC?B&yJM-^a4(#|
zND`@sUNWUhVju$#?*Q8DCp0uTB~pbh6QI4Q?c(ZM%lOXtzPzf+f!;A7`-!%S6jXWy
z2QK+=;IaY&ub!aPbhE$jq?Vu`yeX~5Vi|KU_lA&q6xR9KuqVkQ7^14%`0JLHQPzw1
z7C1*W=xQ{H5%ZrHafKEdG#V#=Yv#R+{{2ip&bj2&9cn?Gs>Sive~mbA!u2s5LP_}S
zXFnMnkTB-;mfDPz5NRl6LnJ!S2h2BR+!;3Icij1=xGuf+b*Ye}K<8TgK@p8Gm6vnA
z9XbU_q&yN-j&EXOGU&V{aq5I(y@FYdk2XW$c8L>cWt{dXd!8O>{e&Ybd`wT?n%+=~
zI$ho$&dy(_Cb*roG%YSkdZnW(oS>(1r4IG{8p(W5Xby(#aW=7xTgMw7nY6bYP1Y>l
zm*O6YIz;(QeoCWOv7e7E&*$?v#4Vl5HS=rtDy)6%QxK{)2>4V!&79)y)V=Vnp}t-`
za${N=J~K~vz>0PJlmFy&5zXlhj<a@8STrJ1{Bik54752Ti$OoL*JbC$GPtZk{qm>W
zZp*m6>FY;3RD?czE$xCkRY_ow$IT;6VE!O35SC-mQ+-zz)H#UcwK+tixZF}_S9ffq
z9Q{{G&-1}o)!1A_Lz;}p$amUXQ4QB0n!QCe4SiVFmC(132Ow5)_3!*M{@UiadU)Ob
zvY^f<!<}{EWbBvndSm?B^R&~_gSlH22`$WD0yXF|b%N88HThkB1XQwfO)`rV40tkU
zFU)$9mzLf9sq4s`_JM79b*-@Q4fBDD`UWBP3A-1E+V)JJhck=<!dbho*daZS;OIWb
zn$o+-9xmGPC$soiwe?BylGJ~UPw(bUVNNWho<V@cjS3!X-mc#MVK}9-Wu3YZb+UQS
z#)Av#%yIi!=Qpx}@I6S=!m8}|1kHC{L6I@`uW?61uiI77KsIYqIAeR_J|reA4}pXX
zZSE$j6Lvq`kh?v$agpGdSv5*$%}DKk3$iuVh`@jBK<75rSvv>0t#;2}<7~aI);xc9
z>b!(=&PYL`B^(a{J^wHe-<NP0Z(DL_KTs;|{-eW<-a-@KP9QKxS7w|x@eu2FO?H7$
zlAnL(CUXL0gJ(dPaG61OFRNv67st4Dv6UrJ2f-YOP!^~@r~Slosz9&x9y_05{Jn=e
zE}S%Au0YMPHS*q(3cM~NUqD0IRh<EnDOUc=UtLADW?a-BGHCZsh$T?Y<2?&J2bl-L
zURk`Ko9Y*fnz*SV>(CL`bRPkhZL9D`2*(n}aUMM5r2WxmZXMH@M#OL;f^;lgtPH(4
ztjjO}eji%}-=iRB^k)5XoZTwqJfpe6q_5;c%7woi0r!Is7|DS<dncqv3R`(wDecA(
zd5WN(X%DSNaIzRJq3WFZwxoeNqKPxVYU_=<^x5eL)(spG(!Ygm<sOwAY$W;@Wt*23
z4r^V#tR#A+CSB);Yg=EbUT01VKd&L;nx5V^H|su6vdFr!JWOK=CS`XldGpcQ_16pG
z1KgM$`pws4ET@-!=CN`M505nX=QCZ1m`&-$b@CwD^@+t*L`F&CP}>F74s8q5hu48X
z84(LbL01F#0iNT1=qlX4Uz>&w5j)L4+!QN3*ZJgr1pxe`TwI3go7oSud+v!1c!O>w
z%?L{AP~#T^YZ@CUP!h0#>St#TSBKdmyM2Sk=e3Urzx%<gL+y`?#C(|GM69&S<K%-#
z5rZT=Na4-*t96&1-P`D)|MZMbJxxyw4*L9bVtl<Sw~zL3+;ijsh)Ui<lB(XtQOHCw
zMk*s{7?-(I_o67fi~p~xuZ)UgTiV6~1a}L80KwfQxC9$q2MF#10YY$x-~$AA3GS{3
zf&>`cog~O$!7aFb%{}+7^}X|>f6T1zHM@7oQ&rEdioag4j(`cRc(n{1rixpO{Ol(S
z$GB<S=O61o&wnaO*Pqu6;c8xGUB2;QB(!Ur7y-OB>|0-b+QeYLvx^@wzL||v>iQcC
zKfmPu{K7%BED2YV_LF`|mRHT=jeo11MN}0goAVn}ay!p?86e16?!UU8BI2mT;P+ao
zOH04kXP3Xsl<4*(uPHU!)~+S#6*R1H&M)eBxce1yAzJqERc83K<nmB1V$_Pij1mHE
zaYf&ltzF7){i$QEtcD8v_+uyAfxza;^b6tx?VsL=pk?3Da>4IuX{7B)$n;J<?@odA
z-T<{7)yl*Wuf(=M;ow+37UXqE&sW!yiO-9E(g4sUHT&&$zng2^y^K|r$Aay2zsa|d
zdnX>(b&e(Y@!_jp7Z;MCXx($kE<!vZglt!;IH$B52HK+2rE|G>T`Od#X~8lV2soI4
zht*$OYLAO20goeVA)g;9xD~%Je%EAWH2$d_<`e)4@4A>9pm``-Mx)rt7@B)rnL|*L
zYICSol1evzck684rDc-1Al3g<!YA4(V5eQ1X2DID2nf#LQw1vC->GUk&MJ)Be1#7d
z`1uHJo-Q-~Ic$dies_F+*LCHy*~`5#L;+!yiK5+rm^_ID+>owr-RyjvVkL^+r7{qv
zqzIpwlw(pNPzl<dQGobv>>#K@*f_2UHq7bjzT=~r-Nz}%Qnb_l2sn#-;=U53y3X36
zO|p{PDrVxBWg6m%na)VV8)1~j7oZY{dj#xw+-$EtjG5wyffxM_o49eQ#|^zJa~I`W
zz1@CnDU-fC5T^msEoTeAsh<kragQe$Fm0wSOwn&t=dZeiVLB4<SF6`){>LytXiMs`
zWFtfB9bzLpql6*^?swP;9vg7p!ltg{;5V>BjpUS()OAdg<?WWI30{3jhuE9?yf2Q;
z{p~p79Toa)81D<*we6NKlkBM24i7m;x{c=5$}_i22`8M{qd)a@eE#KmOY+X}C((DR
zkg2c7kNqA?zKxLws_A1p04pu&5^G|>`$#Pq|ED?h8&6ow*np*182gT;`5)ZywEnE*
z^{-u59-WHSL|eYNjKL(n+uR6Mc&`bdZ0RW}Zf=vUf+7y}(EbuaDZCfp%d%+$QG8+V
z1CcH{MN>XXp5K`_FZfXBg}I9D%_cb4D56h*AYmgbFQN1>G5B|7+F4Ug>ZlDbrMj++
z|8>au!Qlm>{s%u<@X?Jq+iUkH7$BXSHSL*(9zcfxvEffKm4oDQ1Hb`78b~7lW5biL
zlv+)$=K+;SAQvD$GB`+f!{e97ycZ6vA625-3Ny!>LaefcIE=KR1GRPdvC;Dgkw?K{
zcM6i<-#C0^8J3HEm2!Q^QeD(x#Z`P5Hyp>1-?(ZQD35W*N2ot+(7;BAxy7CBy7>@X
zlH1nb1X<7r?Hn8()(d7&Q)79ZtRaF^PT_~7WzcWAJN&(>K*!vqwX6cZlbh78Zr2u*
zZf#;E)=7-2dT&Y|7M`t<;4siOQx(_vycM_B2U)Q7N&d1=gwC=6N(r2Bpr>=RAY}t6
zO%7x9PN6rQ4jzx;L^p6^Lkw;;lx{#y;GXSml5r)^C=o#5)?JAw#M_fiJzzkqznzV^
zTT3wP&_=Hmx>tMm1u&FDCLUeatP>~GBMm6+d2@3nOwXW2f1kY2b@s71si<%&w+>&!
zY@GilFLjLV?o~EY&(|H^FuDO4>NPvpb@-EhF}TnS83&I#VrpN{))A=}l3ARB#+Q*`
zewSZ3Rdu?!by@sIHQianOY1c=pI6Gp85I;(*{&tDwzQLj>4&aSX?Dg;fBj|Ntzyqd
zR*W1n)R&)~pNjj#iG(p{$cHsyLW+ZS4W$vuoCY7^)n49!I0|9XB1Is9qP{+ea!lNk
z^y-qp*A$~NhJT(IzN|A8s~L;gzD8D~TDsDZ){i9UvG_})()yc{k$#7tLyDR*1&Kkx
zZxf<}ut$$YKlrp+i=jsxQhPWs>C$RSdox6oc5)g~F+>l(;|a9^$4@nlfp#HSWEx5u
zL}8ezI~`B{K|-fqsP-z13laDYv_ku>E(Q4#Om*&E;Nh+h)@qL1*$gKNKilrTmU5GW
zJ-y6Xw!DBe_s~Vqn%nrd1A<Ig@F$@P1IHT1q;o~?x0iA#=#>U1@hdpKsC8kA8@`da
z`;_qS`WIq;xd=vLbyaR}FjBW~`_|2~kGmDlX=Q94jSV1CY%;FCn0qc}3&MczF;(GL
z`4I`C8of-I`zFnTXHdjH*8zvkENFN?Ngp30zjoiPRSy49lZex<2CQ_&VA0y#dljL0
zfFrjdH}ccksU+rsQe#d1R)Txq(>3Am)-^mW&w1~=%V5z7^>ReD8FvY(<sOQdX-&MM
zywj!f6p*0es`$&V%=#8jQ0|Moq6K92>CP)DxRKVw$P+(CB-F6)aNp-I`>nfxMjswy
zgo%(divuH1Yf28bl1=A!o2I3~%qyY`B=}tk*SzU42Gj~-le*+^On7$&9&s_(tH6rK
zQr}FMwo1R6ng<VbgTuLrL#gt7$Wu9tX-Cd*+F!fBJ8)mTSa^ii-Ici$_Tf<XaTQ3z
zB`HU7es&iBb@i!0S%2p{O|v_ro$@J?r`PS_w~J&~iPw>rA(~E~5C6i)d(FDBw}yJl
ztBF`0Pn11vqkX9%z%UkjtdFNy-(Qj`#u<LBh*&%E7r_rWFt0Q9@o9fyKWCoX)7LcX
zp=SL`w^C<DHU7Pz5;IC}^-Lym=0+5o1=k-9BC+~kAu}Q-<BgR{jc3scI=9mm;1upZ
z7lZX1Kulb@_}lRM<q?rplcaF?pV5Wz8dAb1ZENTjybZJC<Ok;Sx{#hPRM_qkIuVtc
zW6hmb^9xw&nW0pDn>ZM^`luzYrl*F6fd&g4_lpeeJ)x$rSy@xGN_JkN{}_5KmlenF
zS$7`l=DRb<Y_2%@9tW<4z3D?Yh{m--?^3q-Yq$RfnC2Z@_$ZaX%Vs@So^EVeLtz07
zHP8C~of#<XEcwQkVc~R*$wscOB!;h%#;If?P?AKK69-PSqS_V>eZtMlAC&o9R{Uqn
zHhXISriw>HPbsJYCP5=K5)zI}j&)~QS4GQ|m%8JRB((qZh|%n){fTmsA4bbk=v>88
zdCIJRo)1oHLPs6&tuqxR=O+96ypxVV_6Mgum)>1s?jsVebo9KP8MHe}0a*;eVT3X<
z!zrcfd}G5v30Ym0EG~SOjyM}-91X&U?%Z^@R<<8DREYHXHC{r3WKZz<5IMS6n?`3g
zjBoh-&N5M~#xvKD%&GFl7LmaF&?wXXn}-6StQYaHK6(}zI`PSZ`3k2vlv;0f!Nf&g
zLwOyF^76jrD7FQt4@8`cq1MW(CnIY4Jv_sA8$riv@3h>Fk|9k8gMN|~<LuYY4prR1
zw?Vn{Sv<s#P&0wd)01_XEZ4th#(p2@SG0*25cV%g>4oiu1rBJj!o?~p@ah1H(fGPj
zNcxzw&ScEBB||WLw<MRAlV^`H^nXIV)zE*mS}M@P)0z>KG>I4}(5-2mQU5apWH2u5
zgdqj6jTDfK)DuE<$LuqlytbEwz2a5#n~8gF;8Q}1d-5}BF>n+IUOPk^d{V`54;dk~
zf|nX^hfW1%7e5;4IyEUGY0zpJ3na^l$$sDRQq>@;tgfcb=i=bHaz1QjTNFY>`a?+r
zCkk(vM`rk|5&p7(2oO^q>Nt6tOry$BM?+ISSZ!ThP*Vy}W5t9pg{HtLOAIhmxN!N$
z0aS$DjuVw%%J`X>%}9kOX_M0TEC-X6W8O;rd6N%of0|t})%yHF_>o~Zmy*J;M#tad
zX-yQK?-z6Jb6_d*#_>tBL$!f)<_5nKjj%`FA>$Smq}%Ly5#26%8Hr!ABDz(D;lIk(
zb2y1W*UU~lp{qmY@FvatZJ<{4S)TK(gMjbJ70HByT!-QWc;e_#^A3I|G#F1Wh$tg&
zp@V$A?OR2VnCqEW&Q@!Ztx=kv{<r5?lZd=PBiFb_vJ92vF-N6)Is^^(`lJ^Z01sIr
zZsj>0bw^&1(S(v%(FwoPIE#dh-fGO&GtdXfhy5O5c@3hZ^OL?*I|kN@U#%|4dA%RQ
zKjigRcc}u&0gCqAe*}7@2)eEmaWmG!)e@2pIYOE`NsUkQ3$4wiEeqB~PAMb<)~<a5
zo^bSPsu!??b`fKk2h1Y<^SPKQ(Q26jVyTG&#E!epkoQ!$#rI}HcyL3LG!v8PGfsVd
zDpzUkkn%LHmzNKNh-h9m+jX|g3pgzE|HCuI%W&w(_1Hex_zHHERZvg7zh6;V3B@D6
zjjF!;4N~=d3c2z+{(aTzePLg!r<T2?T*2^$+4z-hvg&W3*#+zA33%AWyk+&cAnAut
z*F$-Uwj;ruLA+Il5mCuR_Qr_GjYx@F?h&w?Eh*1Z`|Lewk$a_bl~XyV_b(7=M)-7e
z>KYn)3*_VM-|q}X(Z?!|x^JPm34hX7Aml1Ue2aHa1;?915aKcakuyoV;BOjK6X2H!
z9JXY*&LgZDwXRpkO~JG-0WrVTf6ZoXC*73n5F-~Ic>WIP9pInNAkt!XqB=$K#>S6J
zRZK&+J3?(fUiL}A^sni-YfMu-hTM10nM9i*V=pR@#&qN+?&%PF$8LGKQqnZ=8kjJ~
zfK>T-f2Z;X4_`oMJzYa$Vz5<ng~yaRSBlF)Tt58CEpW@`c?qe$7vHRF*V{J|Hu^rU
zjMR>2o>aPE-6E5m3xiW!Hl~?*1iQi4(|Q5fdI1-ZXRF;@2nfLMg6?NN3h%@(_}_G1
zGK;i{9iZi5U0d7G(a}d+l)#Fmj0v^h*PQpJd{HrZe?#yd^ox%`vsd%uSux?L>P%bd
zh2z@jqj)^*Onl(#n;#=#W}E|xm0rKOrr96~DokHQuWf>NdKkl}cRPO%PCHcn0^K;W
zVP_I$yvbQpPDrAcw(PHCcj2Oc+{}0!X!f3eg{+>U-`*EU8l#&4dYgQIWG^XQ-~S@I
zDa^RW)X(pA5kMi!`MX42l=Mv0yStLup6Uh4V160<dNm0<ZE<;S;sj)pNk34X-@m6Q
zO0RTW-P9OyG4t?nN*&0I!?V#R$5F4gbad?NMdJ<Y&6+95`j%5$?;SO<>Yrmawf`(9
z^#Ac`FP<y&BwNAl9>?{->-a44;u0-Uc1uq26+ufRUl3c&O{rb0V6yC%F6@MKb;;_^
zCAY!G%)=1B%xP+b*;@<*4sQ61F_x^1iD#14!^^v1FHgA$vN=4qL#9(@fH;01(Bo98
z6<~U3GS5Eo%>Dbg>*2DNKf$TG`57*7W&4@@$ydzjSS!lMDfPJRms1gnls?$7U=`>+
ztFu_Ybf6NuT>{XrsKk90a%?)z&R8o!u$SF@=Giqxj(&S%p-dnq49&O?I&+WZY$N(?
zBodsnIjsQQVD)}TjVuP$_vo@=&>^+smlX5BoK21M!}g5B$^+guFGu#ff#qSA;@%_i
z2akWPV8g((h^YCWk%aUhC_HH0CAW+!Z5-q0?i$Z)-}7fqj0-2t`>rb!8b83%Mw?w9
zOP<zQd<@08q;1GmP3>7~qw<-Y+lq(p#X+HL*qs$<8ej0zGj4~LUoyi3x4o#yir##i
zim$s3o*mm9{Ty4$@RiQWx-18N>`&WR5c+C<1!bB|9)Y!}XBQ8hPT%6q!GI&I$Zz7*
z^T+aDA@MhgnM_z8gNBI}m?s@16!0jzL3<^|OY><#I2u1Bk89kNUH@~w;p%pyg2KJ8
z(I=xtzsvL;fE=63+;^HCaOcw+PVKULMz1@OQ%o>`OYNx6ync33@>NQzNIU$4$U!jx
zx_gf_Os|0cxKLzr;$)VJ2SWpB&~3|~S4!aYE>x)@DlQyq*4lbaQ<b6CG}nvr^(w(n
z-f~hYJ8b2{$5+~>S+z|wJEB(DR6)A0_Ak9k)P8dm5sgz+mL*7G*Tiy#P%6ArFTpX7
zTu{<?8UCcWjPFL4Uc@O)@RKBHDQg2wAcxwBK3`JO$a0o|te{6YgdlY7!2_I9*V#F4
zD<n<yjR}=0d1Hy0Y#)4W-nsVpa7W}dzT~=N|CHwTXdqO|mDc)wCbZOIpRo)haiPA0
z9s{W9kjw%>nJphM(vqw<CR5PMa{JdmzUd95oF^*<*xnTtmO4X}Xe_h5&l({J)NuZ_
zobRuBUVRYIk+Zu@U(SdGj*wn^-7m@$Dx>>$he!rL4YYR8X35yKy12&WcC>{dq-!1|
z(W?^W=kdk1A78ZjL3p+K{%X`IdF5)pDU_Xoe!bSp>Q83@MWC~QGqhtQ2Gc@XAbpZB
zblB(VqT=s=p6nO@sX#<KbbS!Uh}ZXHYc4DAc?&PFY>FLS&5YI6WdQI%*h|fItTao+
zDKN=)u{sAy$J65-RGv;(PgnF!wgo;?u#p%u%`H7*l73g!dr3LnO1{{zhr3SQ0OCe*
zS7FtI;K|&|hCw25wf?J^>jLWg!1_P&acB->C2W`5mIlu%(5@x|0D^}~AO%bSJDd$`
zN@p|+j8E=gu-eXu!eryl(D6k}y%%3b19=v|Z25S2hUWd7u}Wc-JO1^<CZpeDFL284
zH%=B_>C{$M1*;=J@@nB3=qIj9j#CY?R=~$tyFjYdfraab`jxz;eZv-w<*t(AY9{WA
zw5H-JMz|$Gqb|U&3hSkji*GujPSiu7Ve0Sd^10&B(Q&z*W%mAVT;VkK!z{RKD(=IE
z&tIrErS|x`QE1VRWxi!nsIVvEI}G8E2edzI`Dm$Q(^<!8wTME*w_vCRQ(ug01}-lJ
zAQd5(=KsO8&xX(ca`9odxrPWT$z!{cl!zlO`j~g#q*~9&-2bVe>eSGv@frm9PPOz(
zmSUeZ*}*fY7ua0J^u;C(a*B<&+ay$ZjMcv;`V)4#f{J>bn0UE%LZ?9B2UL*A&Gch2
zi2ov5F1x6>Go*Uq-Y)B$p*L(wkU{2YZ;}?^l)W#|n3QQ31E+*)r#>>9i)YuLuF*7C
z3c2nP8Q8d;DE_q)CLIgMhP1dyM_wV^zBLng`)Et%pJF9t5_NNbP0&&ZsP~*)+)Qe9
zy%l={$aVXN@XGBP$INb$ao2?DBrdrBc`L>A|J~49AKQ^`DJU<$nedLmW0m&O*y-4s
zcCy_EYSA4H0>tITXv)cBPA2cz-yfcG19C_sM%9%{hwJ+nry<^}nj6TP1ikvxkLg8^
zjX#f`@|jhZ7u1e!hM>M(>73Z2b{|-=TDcjpI#zkUxHcJQHB_CWJ%}d`y(Jor5a+-Z
zw~)5ST0pn~wDs3_*YFWiPM7sx!re|?8a>o0kk?0ZX+!p6mP3z5TfAONU%{hyt?)~V
zmxwY^#@uVIn{Q1Qz#&K2X({iBDl9mRI%SXLYXC}h>(l8#rRw<I9mj`;XXpN`wi1v7
zs%4(9845TQ5nCGum81Cjv2tblQT5pFGK-w$9l}E*Jrvlqwk`(;WY*^d`+Qb9>w#<w
zOej^nVaS90+Rc5v?+}=cCgf%k6|$qITO=|}C{82Ud}nPODqKk{mJy=f7^8kbk-5Lu
zOTRTJ<}8xDk)$dS!{ZiL(nBKrJ8%ot7aDNpGq>Wwz?xQ1p=X&<DDEN46+)pQ>|);<
zzdLq)+K6lO2F@N<AhW!aVGY&&`mXGFJx~*caSC@g(fIl$ehyF4SvTK`tW{Ojh_?tD
z4BX9QIEste!pJfZ^Luwi;AcB7J{z^~Ujzud&3I)zWbAeiJYHF4_B!cVpgH~x*ZAFW
zZOFNZb#$7^^m*SjcAisrw3WRt@C<#&sOWdpTqD0bdBYModmEw5`eyv(bU?;>CTR8j
zpAN+rq@(oR>AY8^2l+Vjr9W+hz0f5}!@}r`x0{k22P$<(0u7GZ6=rfTZ|aXb{!;Cd
zd`rPeI^!;Y_u#|jBM2*n2b@ZbOiXa?WPV1Y@+Ef@i4R)-(=zbWscY=$h(pk6#ZrjL
z_iZf`uIS0GbOCr+q>?w{`csDLOEPEG!E*-N0palZ47TV8+p}0?rm9khry8^+o41UJ
z?OWzjITx+Iy9D!00@O}sF{a6;cvNEkqt1N)fx7^Vd9_=6w`6wZY}u_tw*P>8afxjK
z8C}m1D-!a3Yyaa1+gcx6knrol6UG?LDn^?K-Uwc{;b4$Wsin!5?cczri{tzzYp%}@
zAGa*Q8h2^tKHtnzxZk{8k;rh?EdD+!nhM^PT2Nthpg_>tpIqkh1n*NYx5sqyXPwi_
ze7nT#cWu-KF%atN>h3}MlL}6$0GyD#47xadxbyvdOa*~JjM_f+-_@c(Lxx45r{GEl
zF_KP41Lh>xi$3LxDT6Ntw|^A@Rz!zz#pFS5o;*FU<`CoY^b5Cn2Euwa`K*nmY-Xp2
zzcxg2z<8p7Ajm?_-v7aG<h-U4KZ!cPv;xLoMJC(`)AFm%b1=8kQ(Q<`52aMz?RH&3
zOFK(k$#y2;co5I0hsdi$Z7Z%Cbd#H9F#ya^;t%qS<rBb3pjHB}{ltK!5AzJd@BX?D
zD+DbQCD)IXmtP{fkpEFC6H%0%ovUY7{=tfvwEJbK9y(K^ra|OI7LERh3WAZLOv3NH
zk5|ODqlMKzv~PzfH@=TF&Fbyr$)sh<ju)25pXV2@)!P0^Rbvf*|NFgG#=JO}LSG!=
z5gF+$Vc>_NPqIo>zn%ht!?<}S&n@c5O*&)o>I~hU@fkpv`qwu%V(|%*NJf?vbJw`B
zp=z+|-~5f4y4u(m4v+m(t`3_*3KvXiyb2ZBF9n726B|4wWoN&%vKwrGMw8wF=^QbD
z&!0wU_`?gIZ`YJlKIQ$C0WiK4>5Yzi5|Jsta-2Wa!U~+Q249L$Ma0ffwlFz}<<bxz
zNm_PqSQ|*k4-aD$Orbkh{hFsbm_AEorq_-Z_Bi%esA1Ai2i59Cx{$b#3#ej<-f1iO
zaM3ERVtr$iub>`5xVUI}XijGv;VPEi7dKEQvk5P4%o~^*R_s2Kr(W5p5)2qDh(XH|
z^NC%(A6tDh;>}e5AFHPEA486DKvn;R`x*r+7_-y6ihn**&%)8&J}swa=in-J`#lZE
zizrjMa93JFSD}pY4a(tLzL<t3=;<FD-;IT?LuCFWbD+SL-x}E<hAv#-A`N{2wG<Sx
z@jYY(bG=Z(XdPsu-?b36UNRz7kvnzG8uTAg%uxF|f2fEil;LAu_0a+w%#iiYqnU%5
z0=&jkKoT$VusJ`^5Jx4xHeG~CB)1;t;;5B@fe#b7R7Ux%+u*3|qpaqG_-(G`+GjLW
zEPLF*QCWhzDGLS2zK=`cl7Yr~D)?3(M;3s-b~{WMbI-_=QR&%^%Ln~dOSb7YP%h?!
z^Gs+bUlvzbp1@6u7^v{2;xht>RQG08jQMDsX#>d-EtsQ%drZ46dDfRmbCDy(DNHaL
z6c)GYBbkw~FB@h(eTO?!tZv%XZYl4`dQytJKq^`q)TcRhqnl@^@s2pC$ny&e`&l!O
z#>S3z0h(eGzSm#FfjA4MwE)zZleUm7*yml=WTRGZEHa=vBadB;$EPbJZM9k?nZ`Sb
zIXFN{`&E0($JJ*ebfxtl<6Fj&%l8~k)#C*s=AHh$*~2NzEBkWMuZQoHFx2>-zAg8=
zR-es3<>c9c1!G7v<vIqCGtZ7M`}%tw6Q@~8j?#pZ^`p94MS%Agz@{^_ZkQJU7fpAv
zQ)kvMTcJ|?ObZsWiV?5-^5o~}7_;uIKr-wvvXk(GFeWzB_h*>T)e#M?QNSK^qwA>q
zRYvJzk-(8pzPcu|;GA-%SW5-8|7zD|cRuIZYsuF$=^5_-&vk-v<WRN$WXvz~s=WHh
z($kf{fiGVA!%J)}g1_3XT&zDR2Fi@mp)D|9bZIE$jN`H^Koq_>Y@#JSg%MBkH5F`s
z|M7;?KexUkiGKb;wVF6i#bl<(c|3EQGA|j!N9Gj*Y`?g2dQ~lchkPD(j=eFJCOWZa
zJL9PePO+`xhM|oqTvumJ0OO0^zXZW7(93x1{Gqb<D_W5I+-YtTuPS=#DVB2UfeYEX
zLpYOLZNmJ@!1fP941cIjmCatHPu7I)vjhBZGAnTlV58C5iZ*;6UQd#->LJ~b4CPCR
zIKShQNsYm_V*mZfPM!aP21#b{^v=u=K{+C!@K8qjQ<4%$8Gf9|Q$&=;{1e4Klruz7
zHEGb2PSWD*dj}c&*yyH$ji&8$a;5t^9i4b9a<J~~g3j>WHEM{|56e@H+`#F_7XL2N
zeCb!GB`{Y?uavpb&p_n9bjh%v$Zfz}$!klb#QM2-4{8vhrX55EZF)0|_6pq9L_hao
zIASI;D`OSlCb{_cplcBpNpEt+b}%skP1uNzTmb3;PJ5aEvmF4;V2VsU;;Md3a`RLF
z3u|+gb3aE80>wq8c_f8B&U%6O-;V^{ii(r164fxime%rN@MlD8z9*iDKGvE6X?^`s
z-_W3C;^E|kTSNq$fhsRIxVm`xHRDRjHa%JP%Z;yU&fPyRane*vHobtYA(D&Ldi*sv
zH>W<(9$6g2@QqLZt*ZA9B`%fdn+YD(Z1E7?!@%t2Ed2(xCqDe2x;iRu4SfHQl&V<n
z>gF_lbe~P^{gxGH9J^r!JtgG(qumJ89?-j~`k&MR5WJr9qaY$ST4c6upGWiK+dE_f
zO?n-jxidsWcFBPaLy&FyZ-zA$Keb)4)SxWuHIY9#yQ`j})x?!a<{lnEZQ{L-4ts-E
zN6XXr9Eg*oByB=jS5I&M&f06m+*gVDeF06uR4E4+ehW7H6@NxlB{v%hrhtF|8BTbF
zCTIs!bjq>ed@w~0h4m22$Y7^w1>P6a1yJYqC+gG3gu{DjA|K6jfx0!`;y<p+p27yZ
z&rbK>f1k@1PKP`n>?(wozkHy<8d}VaCbz}T%6kYZ&uT0f5&uRcyDO$OS!`;gYs9Bg
z<QPFn=Ljjkw)*N#sM*DjpQ7axrGOy!&c1Qs)j=k@0G4LWf`yw^D4TLiWFw|op`nHO
zJCCbNPv<9bfGZ8yR~36Dv#cwcrF#2)5_LGTe|~3&;t+g4yJdiPkAF{yf4@kay@*Ii
znZ#Egzc77x*_0$1WsUOtXFy_JT`c|kuOB`MeTnv@CT>Kf!>RA5RXE%)L+ci>qez{I
z49R5SQt3+LnRVB?uiJHsSLt_6Nw*VaY*u5H(^2EYP5QH7@Nu6<ea2@X=zR5^FE`%t
z$Z2}Vv_P<aXhW=2*Mu{dxkxsAV8RT<S5Y=EMmM6Zbd#4{zX|V;Zc=3C*(yD^!V@H5
zY!h5(rVQ>kaPC<sDy%|-IF2-u-5Kzz?$}GhP~M*RBOyUZ@v*|ngvGTW|DyV5t$wf+
z=KD>2s*&W@X}$A6J1zosY>t-=g}K@%vtc{ji-`%m&>p5t1zULgi_Vkv0${Uuy;j%Z
z_t9YShAycsw_z1UfI-sWCj5$fY3huB<JiA8;}$4kNr02D6*P<fSdl~aE1>TSy8nF2
zOXqX>#J>NXO&O4r((<TNPb0O)L9&g_h=W&)f;e;eSPNH8)oxg^QX0Z1QnJ>oUsNwh
z$Xn+za;q@SQR=ZZ8VEr4{An*pw%~XQK4<@*jm1srbz}#Luq!NDu=v-o1`k_OHvsO{
zUsZ*BH}6sBDtbT~`M^^ZyPWGQD=VwpJI#3#DGNGQm7^K&++Umw?p(01p-j_4yVIAl
zisZt9LNvGwgVE6BWMW;IMdP#-HA=3@U4~tE1TIGU`R#rUZ#pOq#Gnsh9!dEA{kyKx
zu;_LFaDC;SdS;gFq}g4RG+q?~p|3J=^e|0+t#zFf4KT5|yqpXRl}IiOr6DVhiA2`-
zwbO((tI6oww2~~>2+d8eQ`fwMhX15U{<XDtL!Sme_r*{vMqdlO^baVoZ<uCcc3x)N
zF%8hYXMdd+vivopK{G6)eq+n0<jQ$`?j^7NS5B2!^y<$%Qtt-ozPsPEBTbu-a-|qY
z=)cB7)i=DkU6*xd65pF`zG{SBWoBk}T&?&3xep-WwP#A2RpF{xLdZKOwJcD?CpHx{
z<?RmXR|m_0!C3z7nAP^X5*Zb=*Wc|X_{7<W#<!V@CPnCGFnVU=Tjp;Ml?Euj%$bq?
z`JWDlY)yU)robv+5`h!%)mE3Iu~g9#6DRVFBZo*P*&{Ohsc1>VTs29uj3GA<Qj*Q{
zw-l?B2gxecxy^LJblBSwS1Jz<r#S|!xu!q7!#a4#8=9N<{0y4BY=(Yq<Jh)0iLNZe
zAFft+;Qg4JZD<8Jhq}WH|9tEYn)txcS*63&It9)Ch9Si>F6?ajwu3aBfV7*!*8$#C
zz#FPhwj+ZQUB;$#)19U-@vhzh0?jqYINpawhoQ^4`}Q0AKD|4Dt^Uv92`0-Jq8cD?
zNT106Y2kNY<z^0nhNKzS@N-4ZvS~IU^Jto~=cXd*&?hpWPRVl3gSZf;ZW8w+@sArE
z>cKfoF3vMUJLK)(%0+uc-R!e_^1|InwEDH<k<ym4NR(BJd&Ds&KqKXw+cSdB^J`gh
zj4wIi!FfrjJbl-YjN!a`sOY((!hWeSB}q@z1#kRU^giJU7CEL*Dtuz;&SKf&Mt{cT
z`@PS#DWHYw3G~rla#!pB7QQ(QPo+zKM!4?|1QwLzlz}6@oSg0wZljX+|CmLBix3pY
zM0Stu)&>~U{ZfKFCXcJ!zgVQu+0Jat2uLNa(IhuI?xG;?dH*_XA!>c()wkG#Q+efZ
zXO8VD!haU+VHx=U45yoT$g>8+e8qQ8VaTevB0#(xbzBmtnJ!o-mrXe45a>$Q@x-D#
zTn?2=ntfzLtY2$q8JXrag#QBJZ_v4p)LTM)qECl0ytv|-Bg@nhBljs`x_wnV#<?*$
zGp9lv|JNZjV-@<yZv+<JQ5zmGBdYGHMcoOe_K={7@FNyol_B#GP=jwN0^+Wej|1>t
z#V)9Y*yCipWo}pL$udTk)~N%MIHTmS4Hsn1_R--~a?#dW>|!?MtR|yQ&mjMrZZhi6
ziILLlEiiqEvu7gyIDaG~k4Zn<<${PZKtU0zw8^#t!x5qd#e#8jA}#Eor;*~&_0w$z
zIb(IpblkutL6SfSOe@LrDXYx>fu*RleB+}1U%&*oG}j3LVfdD(1(l9|mW1zo9#Nny
zy-p2&`$C!Gb3Br_*W}FI0Hh=^rhe3T!{fAd3qtjovazL$?TY{6`;DG_&BD&X(u^BX
z#gm1}T{7)o-lMCkNrSqU7Wtf;!G+OeeXVs6*Ub7A;oTZuka~d8i*38AqIw)iSvsVn
zqOSO5QeEt>!}}V&Kx9&1V{wG{-}o<n@KD8xg$&&IZpFg1tr9m$(Z>mH2|omi(<r6v
z;4wth4YcF|WM5Cv|J($}2*4c#S*0UEma<$YO-rn+uQP7_G}Q|oILFMg0o<H$dRx2N
z*MwDSRa;ysywdVIep!4Wjl~k7Kzt%2igsc@LaT^)9IS0>=aQ$HElhg1o?iCD%-Z$S
z7ylHC;i&KZqb*vZuXJfjWtkCurLpyb+=@|K>o6c?n2AhV>NM_s$1PRe`V*FjXn!iZ
zg==+x2;ymd^8SF1Q47aRWqN&<P47*DFWZu9z$HG>LYMNQpL9%!;r~hyoAxZ$JP#d+
zeoTb;)oV0Co_0Rdw_=<sh~#m~A^IuDO``%5yPc;sI-hCHvQDXXA>L3|4wUvrL<veP
z@P`j(T<B$-y}~-l&2AFWu@4zz)($|H#8OCFHf?O%ze7$GA^+vLa6M>&P|*k^neR8G
zYT-;AfmAnEI4|N=u8yI{Shgt9_DT!`oDyD#{!ckeQf$Ah^hpAG@+qPKZ4B;=0VYGC
z(ijRp8vLh^(X(%>UQh}D@N7~&x+r<9^Rl~XU)YMA9OlIcC}>IO-YsL(f#M|4Kcm(C
z=6h{Tc@m<(%ZMmC!b1`*7fn<0*hChzmvwcRWHn^TV<t*D8$^@SpP|d~CuGsJW58Fu
z=Bkdu#)SlW?i23QR$UR>A4wTvn}o5&m0i(75s40mt{QxiVI7oxvlnC<mM^w`uO;tO
zr=E|uRjOn{&iZ~WA?xR#NvEcGF0+Cq<z2STgXW~@7t!UF;KP3*^|R>#L`W?`Ejd<5
zg`XVlAP+;CKT`>T{GAr7-tu<Amc`6np(%8rWT0pPtZVG65|n1F)+ZkTdx;+qou70j
zJJVxcCkYud<udKmpY5TZUlt+n|1@7ol(mMed-SWV7g!d+w>9nxkMU=*(J<={A{m6F
zRG|Q|vSBfC`Y?nc(sd;+^vWVqo?%>gxE{+V0sK$g8{4d=reoHOQF}k;T;+DQ5s{i!
zT?~y*t}xuBI&Bz@esP2FgB0~bvjbJzGrs#TCJouzp+&BnY1o}TCx)n_3+aVSA$IBq
z+o*zbkkyl|Oog$T3Rm9JNdEHJqvPKYc!~R~)=%yww{j?M$H`7#(B(xw;bdf93L71B
zLDiVx^g&rw7p{^A$EFV~bUw@iSy{qwk4UM<nQO76BuRMSZ;~X0V5KV{7pNMZwN3a{
za|BL-w3({PSU(6aATT2R5|Q7GJ0;u7it~1B#u3se7kkORqOFsG0l^+;f!f`Fg^?L8
zUghg>>8+1xi9gs)^1Dc#S#0lAuI`y*>(kL-7x#Zl<76qGu~Sx)r+Mw}XZY?QEdB@}
zOaZYFT4ZZ3GSKAaNgSAU@UV}wUznJy{QSO?S@mK!je&jjN&uuY(M0=ZzxK{G9udOs
zalB&1lioNgmi_q?naemZZ*{ClIn_W)EVl@FhntE(jHT+Yv9;5GG$Ds7vRjy_`h0nM
z8s-k~|2(b8nEiYiV1G^i=f6LheA*q9{P*kZ`vhhA|J}5iY9zul`QID5HmBo*{_`YN
mksJEdSgij(YKXJ#KH#Hw>!0)O7hwp%M^R2ywp!XW`2PXI(5oQ;

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_repurposing_annotations_002.png b/0.16/_images/sphx_glr_plot_repurposing_annotations_002.png
new file mode 100644
index 0000000000000000000000000000000000000000..cc70fa62a278e4ad8fd140afee5c039d7c24a822
GIT binary patch
literal 208073
zcmYhibx>5_A3v-j-RTm7^wKP$Al<N%O9%^yGz-!pNQpEqz2t|6rDQ30L6DB6lx9I1
zDUk;0dic&0zvquTckY>U&z-q*;{B>veO(Q55+;%d4<3+fX{tgVJb0LYe?<}z+_!Ml
z+`{f}(mraYK8EfNKJTr)>>ub@`*^%__jw1idFE&D<qdOp6Bm9VEH3!W$;ZdT`=yA8
z>;Kn5*xk!fWLQz<&;2089-3y}4<1lj|L=NOu22Sh@bJL{Emb8W|D2s0vsA9v>6@`P
zdzGKFnxgSys|NGjz(XxbBG<Q8k(^zygxVRiG~#Dn3qO6DWY~s{eKk(e+RDw9KrR)v
zXGE&c=70V)+LzvJe>w$q32JY%TB_NfKHI)J+4tYN(`*0ue?Qz@hTPmOowT0+pB_s$
zmv4gpf89fZMdbhAcW*A=x3@jX|3B}3w_5r<^1o5<2MTV#U8vdb-TZ$Z+N|#OGHwEI
z%b)_Gx5xQS3vNK<Ey`-*|MYJE=Zx}fU%H^YTTv)D2@P$#_#QWse|PP8I}v)Rcjqj!
z_kV_J55AdDxS_o}RJiTC+snVR?A`SH{_tS+?wYXbtCVBrW)_YKV}gjXZz{&9yF}vR
z0AGXfTs(&hF%JuHUG;)Mq<=s_CqRWS_S?ScKA`j=TG{NCOMGMFixyw9VFR1ejUUbX
z67~ThAv~p;@MH)H&J^ouXZGawV3n;Ll>#Z+`gBI|07-|FMnJAFw`*G3+Qi86=5}L)
z=bTW<r<(DRIixfX#sGcE_iYwp837w?cmH_XZZ8GyPTFr@ZwzUJsUNsX4mBVBJBnbF
z!#HF+n>vC_R171ysnT<|PohBC2r-a46y;>A5B6;OVQPD^UDJNBJ?(ydzDlWm9C4Oz
zeNZ(wI5)?9EPHf)tqo4j<Ura9AFJ}=kG>0hwnyyp6f6+cX$$8mB}dUuZ`1F~Ewu&P
zGHIO3={v8KDO~q7x-XTZ>SY<CJeU1i#fI@v_IvxJC{wR7r*(ZNtmoSHbUq_b{PRc)
zKSl5FZHk5nNDN<bUDX_Rwa(+a{A%lJS^jPRGH=LjMvdTpa%1Crn;10Ci5wy3MadUF
zMd2bes>od>w3ZXgZVNA#EXRyxcm2D2v8i5>1$_?{xcezr9j_!15Aw<{rBgCir^G6&
zi6@8;liz=U!Ow<_EVVl^eQr1IdAyMy&fXqCSnA2qqlqGSkzJIA47ao|J<QxBHR^41
z@5#)>CVOUodn5erYOuj=Eyw?2<@0<7MohuV3>OLDScDoy(mk4nm6j?GFitU{Ou!S%
za?MM&C`kn5Xr7^|5rez&CYCT}RF4NSZOe)%v`{tV+Rp*zI~r?h5GEjKaayNIbez6d
z+L;13Vw`MoH9E|M8KmwtT#~~FT4`}S$b*m|O|F+c77d~7X>euhDy1##wY^O|Daz!c
zeC%dA9zn{-1BE#u^>wVjBDo)^btieN6dGWYUMa)$)swvoCc!!cwi3`MI}gB)HuJG{
zh4k=M{9<0HBoH~^iA>i<Ve+|;uKvLRNHS?hXBWV(lM@}2{tcV-E2b><UP_W}dZgWW
zI_%eKE}?lCe@Tq9QFc&NQ?i#swgWmUotB#$J`FGK%Zxf?+Lq9HQZ)f6u3aq7X5WkS
zwXDU`7%ICc)m7C2Q81JNm}m3TadZ2d$mdbn{{^Xo!c+3Um^9@(yGm`wvQsO6tPHqY
zVs%_NXXBSAoIL$^-P4Q$xb$yEk6zBrDuz8LaS@8Y-RE)7f#`#kV7)<Ep2OBv)^pG4
z=uz}nZDLCfiy4HoOs7=(;1j&yPX72JkS^yOLo6r>qAbuWs^f&J!--PhVV!{SDdMZH
zYg=1!<KQ?0k!&wKsoNE{yD(U`X?o=L`e1{g;QsqJ?XAhL9P%9V^78tnSF1Y78I0EJ
z+P+57zc59Qk*)Si#Lwrs&vc4u=UuJp3KP3ZI*kII1>h6?L0RA4)U>p=f_`%HaeECT
zAWjnKsrz=l(CcJt77%vS$Tl%7^U<(5AH{Zf=A#LT!++PxL^4e#RiKX?)7U?<6S8B_
zUMEq%(n;k(MkGCO#`XdEv4mj-*+j?cVMk(~h=|oU3*pJ^7(UQ2CpX_9RMdo|o8dV>
zJ%S_S$HW9JF#ap6{!U;)lrpkV!qm}U@cp~mwQ*97u?RXRIz+O{gSG9(*?wt=Q7=$b
zJ=u_F9qi&onGPMD{Wo3081>La?yBLds~IaVY9tUE_n)jPSO86(*&*>6_mO{c_RpQC
zJbYDR3^A`TJqKqK!toQ>j}q`9nhng^GrDB0h;BmiM8q*{EuMN(`v9yG)6=Qi#iAm0
zMtpjkcE>Y-Xz}{c(38cs))d3vFMpCP(wr$KP&;><3n;XUWx{qwkk>zkrJXATx0S(K
zOf)Y(Mp<wG<9+c)^z<g@H7jh;vVFsp-$KSq5p~1kUfhNT(6guqP5DVtLzH<r;)#9X
zCISuZ4HjaksgQk1s4L*h7}U^(d$UJ$p(79so*W;YEl6`Tf6-6B|CEob)U&APQ!m({
z;w`)zs*~ItX>hby#Drxk&(0BBWZKyNvHsUn*TCGt^0ikbx)Mh2Rk>zSt}})xgbBU>
zls+BE@DFOhKt^IqXBVj)mQrJQC0&umKK<!1Z`-w`V(_mYPJ@HG_C^k5GCXRuHV>oo
zaxRC~VZ$PGXiyCydK5rQJ}$6pImd{aATv%1oxZu_uUmCI3FSk*EpaH0$MQFYx}*hK
zZ7=3n*UcJ&Q@zp8(FUWJn~oCQ%Y~Iolks<jzr8Ioh9cEf_%1e}E?&BuI>iVMtm-hE
zVYk7@1OepZ4ylCkpGYqEcli(-cnQE@>MdN`lxb#~-`!+b?5UFu+<<i>z#+Tzcpnck
zi6}bN-92BF-aOuqUyboj@MIZ-HZR`KHbAb5q35s-s#ChJVHao&!q!GdQaK7w#}3(d
zeF*nvMLyez5R0=nTXRs<YW%1jHtU2Q|B*_!yn3Y|CwF+V{Tc$kUk`Pz_h-p!Yw`HZ
zS*|s}W3bLG@pdTzK#O5OKoAxem)kGs^j{TAtIU&dOB#t!K3+`xajaQP)dhCVs#V+S
zP+4AIkDS}Zt{y7dGS0`GvBnpc46;57w8$zG7-V!dFURz&WYglHQPyuM8MkFgbol5|
zCH@~hVo;JTXEVRV&cp0~eOqF?J~*PC@GZ$wC(#EZCzXrG5!%D$$_C`Na?KBovzcw>
z_r$lH=)sZ&1wOr|YC{LoKlS;1<+(68dC)ksfZ7259OV%}0LjG^r{4W?qFn^dU@&;H
zsgl1?hr4;873*tRhqFPIVbq7s+Csy}DA#hvfVe*88NZejQ&(bhgo-(HmQDWC2~q0*
zu*`{KBwwCt;e!XAO2%GE+Kd8S4;ELidZ<sI##^(>fz(fh3NQ{V4$=kxS<lg6*w#=m
z2n;b!#IEmReSn9PvPmVVMI>%605}Qj_~EM`z0;}u=;AurMxa~0qqbFrlgh4XkhK_A
zf)R<A>V_4*Z_QtHu~uO8!5n1~1`LrBFHG1c@;+IJM}2yp&2K9n0{95llu-=TA=Xaa
zd|wMM?hYIpq7<nb8lIcWwU0HT(}6+7Crj<UOrAA*s|ZN{c$Eh0dKnH&3+aY{kqzp_
zV?OQgtFVcA7d%c~$FDX-?yg@i?$;F9{ng|X2PT@3pd^MB$Mn5<FtZnQYJBkkvF>Hh
z;YV>$CTV)K$(qRke__R%MWI3}$N^TK+uc<7Wb_k;6)Ay&!P7@tr=rd%cF9oO7i9Gk
zk8v;2>BQy6YF|RF<Ss0Gd68U%$?6o!o97ps?c?YEPUbEEPLV%ARt(shu`d}aFjSwT
z#(|AF6K~dfVdi2F(0~j;cCz6itD?VidD&MR&sZIWaL|n*eA^$=_oreE`*KU4v&w*3
z^Yxpcgy^;NOG1fS*Z&kh>8Rrj0&;xMJHy-}o?b7PW2ncm5`nkpl%a2Pc;LP*?d``Q
zp?v{*!?871?O~rfDW&$yF<zmen$w@?`h6|UrqlXXJ0*0aHUUveaaz<;T7{Q{?+)IS
zHk-`-!ZfOj*w9Mzf~)D6FhE4T)EBJ^s@36LhnTt~?}7%bgr4GC^D}SdSkU{KN)Q<>
z*6EFv)nSMt%yjx=SFtGMKqYO&6;Ox!-CEXQ;HJ9lX>|}%AX$q$!P{7!u5ovdkSjnd
za)4n7{%=a+&Kjsli_6ZEXPb)J)JVdr^y!Dott@}zquz_rW6{Mae<{-l1d)i4THM!j
z7LU2H-a!SDNTf0OPepMw$sp0ra&9qAor!g8@}pl+_YyA=8@Aql=*1#C(aw&{W6Xxv
zbG2hxTIK7<7s_zP&9yfsPHi_w#w5&feYR;^i{wBFEEJfAEAi0hNA)X7j)))2ij(2Z
zHPpG9-_1vT1-#sk4YX-$KRi16<Q;W&lWXyKV;o$JW^ub>9R(L-3K1W4ldze$F48&X
zUmPF41{>B(2tfby{{61sdj@I`kX5)nSFQ+?TlB;h8g)XssoW?2F!Wc5T^cMh@6R0T
z{v`3@oQ@l#hp~4AL_Z((VP6MVWCff`xr>&b6qSz7jz!ZWO#I!QbW<Po6zLns>3*R{
zt}8?hh<^{GYDjp=<&4SvZJ<-SE>Aj|M7&Y}h=SSW$d1A0dquaCw|74|r{!>j9|Mqw
zZd`j=(Tk&l5GDl4YPn1f(OkpVU2@k#2oTf%*ntj5l=FAt<5^yLu>@AgP+;>ogp9&y
zcoVTY^#ZSQys`^O*<4Tuuz$${EvWc9aQ#|XK6!3}7F)hV_us0T?nIBMfAutJ`}Gk$
zlCE1c$1>SFA$Vz|V)N!E7r<%NCTjq(W6=PoG5eCrpd{OZucFkg5z13hUm1`RrD*v8
z`-Vk>@jI4hJ2EH}XCc|WhYA;8@g%W1sdExKJ69nLh54bkSuh{K)MFWB%Bx4lb#H8Q
z>6+|Yy_+mX86dHM{c<DcFXx@4i1E*wa2b&1+25*Oes5L}%wVSn%VUj)xuq4}3a_h<
zC6?zJ-QUd(pS&za@E|)x;r{zhC_l2L6c-Mpve0<1sX(vu0p1k@k#$DwFf*j8B<W%N
z87Iij`Ip&rJS>}t&jwB{uO@>wK$os7qQ?!&jEL9^>3|DFohgR)a~$f-%9>6>^Z939
z3O#_Uk?PD+YSj8Dq?!8t%FFBJZx#e-u}wLE^75fSm2yFHKfd!LltjqEQ|HGnXhll!
zP}u9W&UzD)qlfLRgd1ZBmcKcxBj^<2I*9(@M+iE{z~m%c;{GHy8)~SLxJ2XfDQyL@
zrgTZ(4omOho<EM^k_0X#rlfwtXUalqrG0YxUePzAYnl4U`qDHQiviFEZt6O7EKHc@
zsfd(RgT>)GC|OaGS8ZRb1^5OcDgzCwkbY7Wr;-KePN%*$ZYXJqsU{!9TyB5N8nC`t
zeiX7b%-AoU&c}Ubvot(nCZLj_v-gwYE5?Km|GY<o4}VccOlI!(86TkzDWa78XUDQ^
zHrBT-1;Rch`18~An>HXiJn?{H6Gg{FIk)I9y9a;cpE4`PXN<0x+%N#5qlWB%Ukn(K
zw6yP*WL~xwI-!)m6fi5?9BNcIl5U?4w_gt_eBJ*p@xJ11ng%l|)ZnrwA5sca8|DNN
zc*V+|%AD=S!<f#u^b~9`MI{l_QMH6QjQ%2&U<wJtIyco9LYk2LdgT~IIwSVswvhUe
z_ZUK15|)X6?xlC?pv0tWnz8%N51z~;(3=P|8J-1`5!+VLT`#IOLSP@&&o=kk*NyI?
zKaaE#<}4Y2-xlD>8+`O-K+?@MBtvA`_U^$n0<y`jFmilB)z;FIL&-;;gV`DCx;bHP
z6dBpXGxI9s*YQ^V;669JMhFK94s|sXR?&~|__ZD#k0=Z&)sUar_gcyF@uYy~_mxlI
z2tPsNBB<^W09(j;Snv~?Uow%55=cE51}6*$8gWxvX@)gOx=k}D+B;6hm-4U57Wk-e
zs)1d8E^90<Ee+HitR^a?oZeSlJ68#3dBK#jtfZ?_M@L6pS65fO5JONr?RJ|jeD_m4
zIf`7bQAA2)bh)i2ccuX-7Vax}1$HGUj`T14*qBcB-mAii$=#~mtLcZVRsKl+@ySVA
z&e!X^;OyGLqW>JsUw|usH8SuCQ&#4Ykr!Y-c<Pg}?)5>b4<WY~Pc|;VF7Z^5{K}(%
z;zryz3sp_(PabA<st_G?W~2m$q5OW<9_9^AB$gn~*Eg*Ukk%OPUR5|!Pn1M1;m2{0
zD48SPzvGeiTG{7lKVI!1mh8Tskd}Q%rGDdPYdY>BC8Pw}!UGD@dx7?2ORBkCmPg}y
zGDo3IH34~J(AAp<K#2A`&t5-JWM%wrT>*l}joMxI)oeL3xk1&i0QMuKbZYV6dXyJL
zHG66&a7S%E*_-aD4S4FRz`N~hu0<55#31GG!zt}*T72O4Y+dswy{*Ks6j&D+VK@Z@
z$C+~fj8jgAC^7en!g~M#A3hXex>gSp#-F}1On0CHXpD;TDKp(4<gW!m1f>C7CAom-
zKH?>rKzua6sd0Z@op4yE)wlKqPux)FHw|iK)35;q&2X*3K`~F-(^WdX)gJ$`lB8I^
z+OqzFrvaZ^<laPnhD0e`pVMY+r}>&?x#8_6suE{alFSZ$h2<zq&rF18D9CyS9GR9~
zh};d`h3`&B7-w6EV~(F;vlk2OaO+BB0!SV)Vi_}RfXy5O9lfIj%E^W-s5Mkg;4|yn
zfWXz$cTPHaH+}VC&Ol>WWr~+gTY`n~2m&(GB|PHo{_-?f;lG2`H$SCG31@g;uL|m#
z1-$Xb=ZyTpuR;==E-<X+Ei@r21~WXr+Q?fn3xb25&RzZa__#4@7TgyiK5T~GqtzsG
zd_h?&NP3SnudjnxZ};76My){#{?!{;*%7JztWx4)32u_=7B^<T4O(aE0=`WaXl%i9
zM?vdZ!KDu_fO<iztGIjNEZ23;KR6gEX@KcPTrvyzA2nAC4)K7~XTSY;0<L&Bw+MRI
zdvmBem`tz+5oJ1c?Su9m?`E-(9?OC*le5>t7tfw<vfRT7{}0W%2TLGOB-PC6M@V{K
zbN#+pUH=t-%T(JwS+S1@<~jI_zQBa&@$F^8Se_><aiA(}Wep7xV@F4gsUw`4^j{&3
zMQ2Y977V=CPuF7$^#n1-IM=eeI$vR%yt1YFA4MMjV)PYxjPZ>UhQV>=+6s|56VkES
z)d?(p26VO)GuCr44J6qIpjipwM(PxvUE{~ZsC#%K!H!dS8)01V*e9QC@|Qgu?Z_tI
zAAf)g3P3k&%KlyF3Ew0SmxoAO&C;U5%hlGTEpYI>EvDg?^=BFXS4k(705i3!l7NBN
z6y*Ewa*zeDTgLrn!5$?@*#6uiigdfJhzSt*_T&MRiyTN5LQ!-INn;VmNOUm(D+<P(
zIP(L4rqPM7K3EK}N!heJ-~{1=lRwEPun`dul{ltxV|*&}plP21n=C}t>vq3;<Urr*
zq4-f-OCD~haWzqjZQkqU$7Tar76O>T5*=d%3DSY}jC`MmK*1x_+<*Rcj%N=Q2-~IP
zBM=T5rz>R|kW4XLj%Lfxs(uaTkNBg;UQQih^Q0FDe6!SM(~4)=q@cPL?dHU}A0~EB
zd3dFs!CoEw3%bVydZCxecUL2KdX=Smc1O(!=TRmT7wB+A<^VTHPUm_{?~WGhEIiR+
z8(_Qic+9MUaDV8NF#+8-twJ5norktU-G9naqx5*6aQ|!`<=7-;zlsh=GL{D`Q%hOB
z2UrXnQo}UA8t6R?BjR!XQeLTBZ8p%Y<7#<6aVszKXP;j_x@s@ePBG^DJ-LJvB@3sr
zkfv7tn`9x)j`E~CVv5+8y1fnwb!3rse*L5Ii_wB`G;3cn3}CuPYHcP%R(2Tjl5$&x
z!`y`)Q1qDv>hfcPOxNHQJ-0^9Kf=4O`&R*@YzuC2v0|Epq43NrR{xL>dw1@t*1#sO
zP+7nxu`7AEM6LVHhHJdNng7L6c!%6<LAFhB6#{xSEgxgnOQ~?(aZh}jrmqJrtF>ic
zTx3}3-#nW;`-q8xW%D8XmE_=p+u=Fl88DuRj!RPzRy01(MBsi+5LxqLKe4f(1|Ey)
zapFEC*Yi`Wu=M)oIX1Y;eTE2;M|MI(+fD=Wnk@4#wyo}FFBDbclFMmUoifYj`cw+g
zRH!#@LCMP#3VmE8FV-GJ%b(yL`P3=W3fpG$R*C1%NKOLr4PAz&t=jAMV{<t6_wo13
z3>iL9l^k_!33oEMVFA?Nf*79lhw<m&gWo&p^SrAegv<_H4g2h}SSI{vq~4^HZv1-4
z6qQ`@1#}kdxP&CKe%dVA;;lWQ$`_w<kFa7>G|jFhk|6Y^CX(q_vt`a<3JKF+(rnBl
zwH5FsyJy&S1MXCWWm(x$TQCiGW=~82Do(Ko{s~!|$`<z!j}p^z$zmVdBfJSIn{8E#
z_Sk;)EnVeT$#8?oa@=3TwmEcJ36IqTB|KG|xIz_fpq)XC{7-^YdaS<4Bp^M09O$C|
z)1&o+(FDRr0>SaRTFAB;@`W)T7eBjpvcHnx8>jv_7qCe>6x=rxb)0SufbDjj_^llJ
z6Y(;bw1r${LGRb+=szNcmefFChN>IB0<0AFm;@veG3g)DbQF5getZAp+k@6m=47zW
zD>mO};ZVMcPnAv!`o@KEoi9hWx5I<2V8%>FM!gh}q+IWyfRK%gOU{~hdDDzE9Juit
z;CdN|(syBhtW@SjJ79IygJIQY@;Hj`ldI83WH)yio@wb@@3QMmPeLlx)=rRgIhZn?
zlF4>G&x7xAdPEO5)w<4qiF-8tn9p+yCUM^Rt|&-&5$bV|TXDYh+`3PmnYv_cWM$LN
zHLPI6N}o-6-HNiSvYyFs^HkxqSGBhmGm8bj&em#dwYg_}fhUo-_}Zn#O50AYNYl7?
zcU~YMYbxqIaRenCl$CU^BE=ZCqYbH%82BlYc<lX+mp!W^-#bv@m}P=#Kuhat6=&ZM
zks*<wfB<=X0%cLs#6pbx%75W#ou-(M0`&2N*R3-l=apo*n0}>Fw2Y_TZ%%`FNQzI<
z+@PqGTVvMMO`5CK;czjgd}B?iI)7P{@9j4KLBV+Q7qd0C$e^WJu|QKb8<r{Jn)S(A
zoNw>s!!g#>Fkm*;Gby$pj-?Ut5r+rH2%iW}rpq`w!Hggj98=TT*&K=M3GX>4d>*F7
z!pe+SY)|E^e}2TB?#w1e?}-2KYgsxlzrfTznp+dBVOy_jrKZ|bjxp^#W+#Bv3qM&(
zvo2~5l4adAa)WQ_bSp044afZjA#Z{_=qvAn%kiDEQ<xW}PVKj&qd9gSZnI#`8or%$
zRlKa?-aif_V~MOlc(KZ_N`br6Xi6oW8QL5?81PKLctQ<6!l?FCE274-8Io0Y>7Mi(
zHCGErYdF&u_8k6VF0SY;082CH7Kc3w`}Qi{@)1RoolP~GL$v`!?n1{+XVA|Nh^C&h
z`w;Y@EoSmH%%F%*t>O{>QpSr?>z<zpM$c`cAhzxwdtFZEMp&bif<CbCXg|>#jGJA)
z7)tXz5ZulMya$t)jMtjP<}@2ue#!{-XBX&A=k(MAehCN&`7}E-ofd4VD47F63mxBX
z=UTIxrN38(W;$f0^T*vCgkI)HoW5SZ>FHgV$>?Y>t}<yY{POhZ$v<x7hsy{fTDsp(
zwG{g~2f1C;As*hr{yRTeQD9m(-zSR37oq4blZCrb*hjk3rP{l}-p!A5s!t6@_B?Y4
z+#3JOq}e8n4e@%8)CW`lXj52=MhDC_xK)V>&Vj$AnUYFMf=bBote1#EqD^6Afg3Of
zE8D)28|f_;{PiS#yj-Bb8c$qf-Iaa9v>%InUyUtXdqQ)7KJt12LPGL$waz9my2U-L
zRs##eN(EQ0L*1&j2ZxR+EjF@zEj?$gU&0of<JSIEC;sPPIc-<~N?6s<$P>W6Y>+$6
zcP(J2@!fhyGS?vH7yf(oHUSq`HWvw56RpE}h6Zx)Ft5G^L1Xty?I13{QN$|<KLKQU
z!_n<2j*=HNzD{Le(%duaye~z17I+>Heg|0Fk9bBfq+;m+e_1#ethZ1`ZvdYBFd-55
zyWjti-bX~=g`@ps(E!Yq(Z%{`db{WCQxx1r@tvA1F%#UEiwQ$Shm@i-8)s+3{{xij
z3;r<AJT4X1a3c!)Yc!q)={K%*MG5hJZvIJ`O*K&gO_kzL?jGc#3Z&)94~QdH;oJ-R
zM%W-kGWMSUj%@VVI~j~bjwjlr*+7!SXfl!pN%!t}b0t8LSK5(4#LzY2Q6e)EH0+(V
zvE<VxO`;XrQMldw0CHXyU<Y7X{c{-3j_hdO>(SxKUJa%bVc4-QM!=t^<<LhTTD&Q^
z9s-2>CdAPte5+ndUpttL3&<1262H?dWniCjp$1eNn@+FGy4=et0QJvxkY`>D(mt>P
zoz)Ld55qet8vfJ90};a0uYwaFCgpcZeSXb(PYNX@K;4~eFKX_t#fJfHEf)8cV|`T~
z!mkuYMV-i|iWnnlzdcHuBU6*iX2?7hdvhMGaDyjA-*6TF8ebY8?fem``Yk%rf)o>G
zlFk--`DyRASbhCs=!p&Z`g+^StP$S@U-@bCJNfDEzSLftQT_^smZyi^K3>yhKiUnc
zyAQ@p8=?5d`U@J9b~dEe+m%umv*YucY6$fiAHd1}7vA<S=W7(iWQ!{;-B+;?+Oso4
zP@24hdAjeD#K<j5;i6AW1bq$eqdm-^gz??ri=#-EA|F4ulAe#s1^izm{rfUDc}tay
zys$90TjgQS)n`Xb5E6D3;jmJksDag-;x&G70yKvG;k^umGHXa<N!50nCS4>K#DJta
z8*9D-UY#|s5QxYkzP_7NkCB;>JzBqrt;`09w60&SV?!^j?nsAh*FOZN{CBXf#j+Lh
z)+^o0s@=26LeYG0h=)_SR+hDBs!_Qha6hXoIam*#YMbqVDAH4JOYByD<2EN8tz|eb
ztl(p<)B}A_lXEOf+>l#h3ZHm(d22IZ_9Kf>D@2hK#1p{wU3nMQ_EtkgR6mtjGP1`$
z{Y-Cum|<VT>;2Rw#0mBLx<fQUN@PZo$!122o0ykNj|KOh-oyp4%2-yRViy}M0&o5A
z&9tH~!3!xUeRsuE2|bb54U$F9D<A%2jE)6}vlE>a#yNRAUo87W>0h*rl^(MfK-0J$
z%tnJS#2a!Hcy$6DMpOC{d`3PMEscbpuVmYUsLKNJb_>vPs7C>lP*~-dsL|t3xx!y1
z+=tW$scnS<wGm|D%xlm13?>T#dwvfO+;p56(u5{#rN&D#xYBPo<o?w_NrE&VGPr4%
z_5l2-$N%wpvpk)Z@@el=epm3gvI;p`37oO((mV>-&_u3i%&o5{1Y~PA=@;##Y>ZPX
z^8E&JS3<NW@D&E>?BF((v9siVb~RDuW@K|fFCqVJQS5lOgzqk}-+0n3UfcL!eTcS6
zEo*IFb=NP>+T#)Dz6`RzM3?!S0&C0CQn^LbkFRuzUq%UV0?N=Pk(xC+(LF{VTY%hI
zpG-yomNHMMs-;!l|Gq>e9GV#Fp$zp`E?!O?k-ZbG6QV?a(q~>`B?WP(?{-cM{n&*$
z{9@g4XP&TEWmfC&T1ekBJ97J@6D^@k2kLPq3#X?vcGV$y?}~;rmw|b>&@TximFb;Y
z^1owBZUc;Mw5wbmPr{r}s*9D&ve3SBI+xQT@=TviSZyCy{!(F>LgZ>OBC?32#oy$`
zTqp^2H32wthU(i5T^yKr(c>j8W6sZd<Fm<(!rjsbNV4d0<P%I-U|xRBj<s!DLQA{c
zw>NqI-$QM`-9$W(Yu?YS`;9^V(l!t!=L{q((udI73K*-260AP0j$;?<N$Z4{jQeK`
zlR0qVdg_ILYhtNRfeE`~D}<jH=%u+Kre=*1+ax8^HxxgUAV{31>%mZCjjV!PaVtJP
z?Q0u8_+sj-PI68RJ%OZ6d?BQtf6TelDU$hFFA<mo>{7%p=C^ie`XR%uxdmD{4<E<F
z5<rHFYR6LSs32<_ZrU4Nx2}Eugj0(8Rn{tIIzJ90ZZ|v~_CkQGg2U$hj1$W@Ckcj8
zu*{j$c-+9G>#zQC;)8#yq=~`oWsgmc+giGdt#h=Kurj~j5b;r+=Faku==zR8cuLbO
zy38Kf(~tvfvo@K#-^N(me*RBY2+0+(NmNlSwdDWGY5@5^Uxy*Bv{YAmkEWKIWc-wg
zETge3iwPs*dC$DQdaO?q9m3fkIR+%)oluV>HkQzpF~X3|PI#>zG947#w!QKH-k##p
z8J}0m#BI-?T`JrA$Bk)c_;f&7P{PuqxUed=^oydYs+W#a8LYYPOufrm7dUp6Or@MZ
z`dE2T&P{W|^O|BP14w0In2^I^BTSbOYP%Et)dzca^|V)w;&#5D*yyXiIGEAva|Fd<
zR8H*;rX}biIhgeN<S;%?Rev2y%$J<fF+VHaQB?bA>7~w4a`yY|y^x<=)lX4!V1d%@
z$P@uLfL@`}K-MfBHkJ)E9VHf@WMBD6OXXKXq_rP{rk^jVlIn2Q**R>;*ADP>yV*DK
zZ7JXRVd9;pHua|LEPsQ2-@AK{M;DU`E_i)wC_d9q{Y8OqXMFrtC7+uk%<2%g;yC1+
zC;DH>+Zn{gc4z=^MbAWHl!$`=?Lp$m{Qmx=#@c3nXeXsth?EG#StfpztXnj-NYS<}
z-jrMRdnJiw2XLya13ZJrN7}QOqanHC<-7U8h}M`FYEVyW>uWOU(AzU-NcoO-Buo2_
zWRng@N_XwffpJ5BrC~B`l3(x-4t1P+J>XL0he=g36M0;tWb`t;n@MH<1$?W-q}{+9
z9GnxCqk5a^&A~O$Kd#>Y148p(PvqeY11{K+lzH_}MSpoYri(m1n|Zq}8O)ObZzzHe
zBK4ef#o=wX3tjlAGjf|=OBU{_d$F|qX;%!WPLLG+$(zBJF7ZmH0=PWd6>8@IbM^o5
z!G?t}=zZ;G4uD`}R1WM@*5>SS@|rQr+b-L|i#o3g(!r=ab1zhVo}R8D3y~FXhSJgm
z<nU@%dcQ6@B>{o=Oyi{7&rf%mvE||;+P;yqh1V~i{YUlabgSaaPQ<)CS(lETkk>(6
z_yd$gAg{D<=)S5!Su7@{8PE;r{^<BB%r{sxAv{JM-N~<_OC9<h^kM7oJ*g_wB&hFN
zp<Lwh2$fkqn9}_V5a`9yy?n5kuP`EIX0Wy9_U|i!EeYyg?Nn_lX8QD=lh%Ot0ZS|s
ztZe-FBg(m23}Y5wG)}TWo5~eMGQ6et==<7AHKq{W$-fXRj3QN_TMUS@PH`v_n&fxT
zc6n@Lf-H(xUQgUS{-a2V;r%BO%TJZycNsy~8)~P?_qF=JgH*a?3W2b%+`~X|ofnGF
z($?8S(4H)O;KBh3lv+|DoNC+Be>t}daqpax;p0+<BRC8{_jtyBQ)J{OX3?>msjzeA
zU*o9Zdt&e>egc@XT{@C~_EqcKSPy)jBnm9)vs#n$COkt+QC=ooGCPMl<mFJH4N6UV
zrKmK3PV3)2r2V}$lnMa4s7d+<gz%JUCM2It_hg8SScp6+kKf$neG_Wtv3Rs@kBDu!
z-|utRgw9&;s{%$D7;9o=CT;wn=*#NqEGRvQPB_C6aXRU&$yYBF8!VvR9l3ps4|&~$
z(~e_wT$@M9y?HhESlC5JkW|MOlc8CBWG(F4{9g!jPbQJ1ExPa5!wJ;m3a*uN=j2CI
z&9gs)i6Wqeo8E^t&>$`oUVRtkEk@gFiyBULA=4Vo??4{N@Oh`G?sRn=739$GJ54K~
z_udlF-foM|Jcso4Y_}4vy<Kx)7=CDz4zHl%z)?&Jl9WLs&MgB?D-_A`=9q}s#YZwy
zmy(&?dHwFfF(yB9)KZIk4F7N(t;^l}Q^=0{?OT?=a=%Cx`QX?yW2W}Y7`nQ<%M4}S
zxZvOnQOK$KdQL%$M1S3tS(Ok=;5466HPP90Xi*UeXA{w0jXb-Ok%$$g`6T0=RRhs(
z24OIi^foAC8xybt3vc?m5-Q%4$uzw$Ky+cBgOs&Q$w=P#zKj}v{Q2recT2sGGxzti
z&BRBN39A7QSV+28Zxkpy0T=hH0jX7Lwbgr;6c8O;wg!Lyg75ZH^r?w+rISFNn$YLZ
zEjV=w{3!|b&eIu9gJ+gVMS<|=PhU<j?O(&37N1ii-fd%#+EP!IU81>Wl`ps`joHRs
zmfNX0G{H(z_UnYNfny5Ekd~MS2!<^;)VI#u09!zOMOKOc!DrZSt>n5dh3EZ!MK6qz
znhCf=GZtqI=kdEggU?q!s#^ZdQ7LqaGkohIOR8aADc9T&O-`XqmElA}Q-eTZR(Q|q
z=W@Oz{WIs<<a|I|1tJUQb@b)01y|e~;|AD=!lty`!o--nfq$zkXnIKKKrX2U>IF+F
z@4fi_MRY!TXs2-o3U0#32sVFQcw=z)gspL>sSZk)!_UZefwd?i*AbrH(hpVhe1Beh
z8eB$aaMs!)E3PXZs5#0sAvIAcp)!3jXmxk_q)tYn_ryy{;Af)3ZCLQd6KJ@nL<H4(
z5a^K(-_$pSMUeXB$Fnn;V+nCGT9p@b8G>3gbJ<OAtY*$EgQ8K(+6+>VS0&{9Uj><m
zDJDmHHDZZM#{lY}lsQ(%bnh)^^H}}CCX(xH5{v|#+SKF&cW`A`;>IjfAT}c{J+`nQ
zjfQL8>ifBW=(kNgK9{}3*LiKS-b0)rfYAn@`6jC0SUdIVX#v|87p`*&O2zTTI|iaA
zn_LoKH~rRA^d60wW+mCs{GR=uZ1f*Iicbq2<q1Icn|YP=T+tkKcZmLQw10sEw?wcx
zRp!$Ch2lSwZebL9tsJ*wO$%9IEcH}}>CJL32D&wp-V1$8i$rt7K>^BE<SD28nSUEl
zk1C`E3{;>tS~*ia-wNZ)ZKFE#2CHkVo@@bHV<;d9%2U7A!*5H4V2anMc!hg=o4+h;
zjJ65u!Frp%!2E`Y&G}BG`4(0}ha=56e#jRqJ)!m>ee(O7N>QNpODsS+sbhR~IvVmf
z$&qzKdz0nt>c&WtwHLUlZHRF!9{6|3OqpSjAHjep&E3WDi<KG6wP7ZH!5H7|RI8i)
zAGERuK>gC9q@T~Dr`ME|;&#0*_(&EgkuX~(6Z`SnV`^f}<6gk}XY*lZLw=|>jKVXx
zgcfb0rU-e4h*6Q{!JYC9@{gt%lzl0kf=^M@aQC21sI{eBfE9<Qm%|PC{VAiPf0gK5
zN<!EM1;ct(g@AdyTB!#%!nZz8Bex~!odK-}JQ(hlCr8b87cV7dOfKealeQkPNSb>s
zUR`md=00az!{BmsJ~n_$ht5|j#GtFK-GRI<*oM^lA|rtieiWd^9w3O~XN7GoK3Af%
zn?=~|PzjI#qJ}O$K*Np1bm*(Lb&L>EV#uCl!mhjuOp;A<EVxlPyyoUM(vu}~Q!7sJ
zq@{gvvCTJmxaXb~jyX<nC0(-!EWYU(qO|f43`D9L%v%c69?KSDe(&z)`UVnBzxW?{
z9$KXSxl5DC#igZPWRmCXq9KN6{8yOJ-n+Rt=g4nrS_^J_h>A@hjFOMrXVcC(vbL8Z
z)sJ2IB@J1RJHw}#OvU2wklGDO$v=g3QHc&4`g&j$h_2e=lJ|}Lu*R$CYD^5iTO;H4
z@~MO&*kRJ^kCWlIyXKWv)r1}o76+yUuFY}`_T)pid7p2{@3BXyPF6pNe26I@Ra5%j
z#YYo5b9jX0eKF2GmHKWH3vuD>=@1?ES2#R9<%uxoLzr^ed0v~=cqE;K=ax6=+msp)
zWI)tE*9-C@s^_a;U;hD0e%)tAYIKbD*>B2jHygN=8fe1$W|N_6RfWx5b(3AIZQkGI
zlgBCgxkN9+GzLqOxVl<Qs)U0qpFrlnOIX&gpWZxUrdP|DP)Pf#oqguJy1qX2%bHjl
zc$;Kz)?~5a<}kI!qVUv?jj(?O^GNDi6NO==whiH2l6!enEby|03#CyX^IG%z0Ty5&
z>J~`p*c#VKlQmyD_hTgVA`3y2_)k$lhlqJ1$eKw5jDWps(_WSZ)zAbQQi48~A66w3
z{&dz5@TzrM6MvEk;rW?;r*5+lu}K}7zQW0&o{#^Bs8?>%Gi5HFd&W=zF76~@4lnN;
zUjLv{*su0-X_g7&)4h1n8nNUvoZR(+N8y!S(sdR<V_jUD=jqB@c`tHG0Qm8X-D^R`
zAIp^w!90A%6upkcld3SKAa3;_X5_0M(bjfL9GFqkU=iJBpH=HlAPlXz=leynl7DN@
zSH;DEd?N*hr252huR_w!A-|hEX4C1UHmN|^F*STpZU6Te#=SqV())7R7_E3ij;9WS
zkI%4}L_q~W_En%YGU@Oxn|L6eEnLoPyI&*KL&rrGcjMn~2q|0x3yGd<Ew&r=*Ckmd
z*BZJ!XnI}qno{FK=nSzn64A3-cW`^VT*gMHcTD*^v7|t!cygxIB2B2OEak~6s8TZ>
zI#$AU5j?RrK>a7Lq52=qOx?gQGf3pDH{ti-;;m|nR9UwV3}?h!U-EwECRQ|Q0S^wZ
znOz?*ILSS)x@Pu}-#>eCZ^%6PkSrCM?h(?Fh_`<~kqFp{UG_FQE&m-&L_)+8`>^&c
z_dv?tEXy%$eOyVlL4DwN)5R%D6pj}Pxc&HT4f)+Oq@zdqYn*2i{42C%ypaIYZe~6+
zqg=xOO-_!sTirO)nkln}wu#I0>bd7lH1}apu#y2W$fNZq*=h&PbDR0vH*`*c7C^I}
z2N+{$vM26DPGfr8hXDf-)6;u_^_0EQISPxV4V6tH2HWR2u<ny$-dm!ShvI`trFa`v
zUbEQ3S8hWbS=82&KG=NP3Q|O+h!rAy{%NEyNRXSlnBJ4IsS>}fLT}>UcQX=w&>T}r
z>3hZZzXU)T@ig>TmXE3@OYZ1O?vW*H=#jEo(FuNKs+OZALyPzIc{k<JOU&<S_oa_m
z`(}@(HxYMrdpVlZ_O!n||1$dN)a@sylYj`RqyDg3=D#5RQbpvsI9t&&KKR>G{Y5+m
zk6R^y_-9`mpBgR2t5?b%S4&wz6z4jhJlB;Zumzqabo~_qqqSrq96JUPAh}#~^YF?@
zD!eG$lw{KkT!{g*2JeN7R<p0sndO0b*7+C}3+=pT2Sv9>NP1a2s&sGJJ@_v}XeQ)s
zF3*$Xj_;p2u^ny1EY%opH9jdE6PPklP%@0V`UezbUP=d4JMAg4v|G7iA}Z9RX@Z=0
zijgHCRF<f@_G2*|Ed7{T_gD+A2d~+;7)at*4&I)Ols(Up<ef$uQ~t&%CO&Vr5QZ14
zxW>uFnB5Fb=&f1Au%BVOMzdKC=59Nh!XSov#(V%ve{`h#(qhvO&Uoz2uI?bCi{OOd
zN}uE_^}OTDYi%E^O%9gXkB}E3H<BwkJ2PPW&AK34Gm~E`CfuH?bd{VVR@e~Tdtc}r
z$&1VDbyBFb*Casl(0|S!oeNDlnAhkOI)ayG4pioipW(8v+i!k>Dhc-+a07aW`;)Vp
zbAHd?UbT_Zq$KtvP==Zf-x2T_Zst%muJ*GsfCug=A4Y<)=?pnBxOkIlj(G4t6Vorm
z)(TIjhl0a+#%ETv4|jV%;-)ztH@M!Hxw){sdtH8?4e}ao&v#8A5M^X9iXGDg@Qx1!
zFOAJo*>*U^pK$O|D9wBF*;i4gyxdVz){EZ}3kYp7aOO!@71v2|5USs|rCu!?tK0{I
zZdNkRfDsQ^7CcroB^peQS|Rg{Exc#z=2xu08a2WA&b^F7qS}rRi~G-8=ARI}s6`26
z9Dc$awR*u*D5I(voNnHZWqWt#oF~b@T*gsVJGRj%FK<uR$jZs(*4;h3O~iv<?!?HK
z7KHChy)7Ft?1at!pbq7qa<(G-lN=ddN2=9hxdi3rjm3Dr;p+Td^P&IYB>uzb^6qDk
zTP61dMKN@c<o$do7C6-4?r7bknbMdifpqH112#a8I}d}-L!i1W)<wVP8}oa(E(H~%
z6*OIR1X5_K;<rp&{NKSbGxt!Wr~^G)KmF`+pe{mCH_*{CKdDn%i^Q~14Z`Flw=}&f
zcNWqi9r_$^IKXRBx+s#u3i>yM9Qp92Y^trE`yOouTlr6)nM!Z|;KN0bF|s=53-Vu7
zT@gqu3_VMl1;{r~RgIC`L9pI&DrG$SPCTFEROxe49Fyd1jKGQNuH5)Dm{B853ASGz
zF7QH-qd=|wgFFj+jwupIy2{Zms=hkA^fANtfA*QEfqHG!$<lR|SgYQuworT#XC*PA
zIC;IU^6--^0Q`B0{x>)t&Wym(40kqB(<n$fY__<#bou+m&;MnblJ9pC%<<AvW^{ZK
z$Mkku3;y+6(+2mq)il}L{N-J3RkEi%n-OgKWRn(GDZR>k93<ITvmV94QJ-C<`gmOA
zBMpk82P*4pgAqMnMqg@}Qlx(SYu)Na863g+0#VQMYXN#=PDDXvIHvTmk*E2p&sGf<
zE2PJFbb2~CG&GcT;&c60Ulg#HsG-M?bBfqm!hQ74gXPJDG*jC=i~6zqT>sVT&2E@y
zy<wNL85j4{HlE(D1~-5^V{R(!*_pn^c8`&xF{Plk1XpD#;;H*n_>m$#3HQ*3p)L7h
z$J`ONIiU25KYuA~tuNyVI&8Y8rpnyT{5_(6;qwH&AV@snK5r$XaPzNE{zm8?AD6=J
zt%xt2;@LvaW356i)N<CtwRyo>UOj~#V}nV*2ix}0rT1Fxy~Vt3ca$p{^{<@0vsApP
zoT6gjW!SdyBl+o*?lqFXFH&`PY@#^f$3T9<s`m%-_uk9se+C5<fIrLhU_R~Dh`J34
zx>qf2Y3NTTBOX&YrU{9A&tyqwmfF)yuMMyNeudodYmJ-KwTKkZJt@|d79SBq2@O5^
z!}=fHef?7H1A(v3$9si!CT1?;W2vO(7bHG(Zf}vOA>QQI?TY&!8cK|B64yFJHs&E{
z>r~;5JxR6UWxalJj~je5l+E8RJzDkekyq;wC1cpE83&-4r3Z}GKHQoRwLTPO8~P!C
zwBvQhLOvX4$X1f25B+A2l1yCBwf4xpdlg)XV>1*O%_gEjKVR@RG-&RpE3#+IVVVD%
zNohdh{4@9q|0X|vu|3KDW9C3pgUa7I_kj<lQ*p_nBKuc6qlTG*YHlddPjPOfH+eMV
z6Kh1RZLWee3aoCfr6>*W#X*5bk`k;g>(b)t49rorXT<!cj}sVCLb#nZQ$~$xgGYRq
zIB3Yl#S`IUjK<rrvAhhnkdM>Y2GZYk`}N!R`8(wa?TIv4_W3FsvhZT5E7ThB_)W_q
z!Ok<nkR}ZXG|OE##nQR(Z_jOi`R|z*)W(rFot&z`QjNC*rAwR?IxfFR3SgUEJs&dL
zF;(Uddf2Ayem@+pVEdr#=Tc^1>br_@{zUmtzj8SDmd&dD>$zr7SMyGfV4wiDjka4_
zM=hteaSo|K<kM8YbH~?IFKYr%=fAiO;Y~D~<g~=t2WsCHRy~?3<{&}w<<N4^t?}3-
z<%qkQNAHV)%G$nZ5u*#?2QRO9XsPSjjnadp8+OBv#cmLSl>XWex9wK{tcw{UzAZHR
zD#7kmtlNjSls0c5Uy2D)xgAo-WXQJv_paG6z-PKixwn|rt;-+242+c>1iBE~K?)L^
z`F?vUPE&fRaT(syu~|(_ov_ObFaDhZCnzNB1}NhJy^7gZC~3@Bd2OC*!;#?S*&;IJ
z6^Fle0NX4NzP|jHfcMul?tM=m%a;km6ro%V*84CXWpTf<xNRWLJi3bO(#ms*t$v(^
znq&T6zH^5l-n6%Ase%W0v`j)Kcu{o6zkCIH0FIXwIku8vkGUDV@N6Z!rE2A$ABsqV
zlyHKxE~u+&gr36PrUiP0(PR9f07q7*-ndXk#6<mD-L(aReco%)Bhx(_MWM)GO5l{P
zujz1hZk!}5va%}Y8vtGZ?$v|yOf*vuV0b$H%B_I&rr7w`OMY29`#&~GR4<@I3qJgY
z<>T~2eRRYr4A!|R1@m%mbFh^0bmH-ot*q-oHyZyKd>Gx+^K(m^5tg5Onz<mqG9BhF
z7;A%VT~XN<-yiV3=L7w{&Qc*72cCx4qtykvRGe&o6(3Y%wG{tK=|+5udo%_n@<7E0
zjHf_bg<CJOhBqoD@uZv>vbl66IDZxb+e@TL0-x<fiZ!+sl_dREIejm=k^)Z08Wrf7
z3E0G!tzqZ3b2mk-&PVPp6jbQDmQ1@vW2IL5f;Z>p1khMkKR7iA=fr<iWSXHZ6h^ml
z+XhJ?B2B>es3-jq)6qW1>7yaECSI9ys1dxYwnbo52)ZQk4z?q5)@Q4&f#pu`+UQhh
zhg$V??*F?WBBS}#BZbI5DBfSFPMaE(^lp8l`-hlLTGiP&0?=gXPUcoJg>AqGZSeV)
z-nQMRORa>pCwg{IBgE8eHEr+Hv-b%%k?k6N?&KGzPkHOPJqt#)OkG8B5DKBd$Xdct
zdgI4e=_R&&6^{mD_*@QH?G>Kg=Y{SoiWD$MPluG|q*fe6(>X1djn9GAK6QEIaJ-C~
zy~8`Dn`>*ZP5t(0q@Bo*RbV?9nTE|MW4;#+79vOs?AwAax_sLXPNrY;l{DoM$~w6<
zI)W=KpDhR6`wk_-OcQoqvg_kDXpDSbn|?1cqCT8Fw0@tqy|?z>$KEua?JTi2M_2w$
zdL1esleiw#z7h`Qd{-ZNBBslWW3l~1eXVPvKuO^)n1=OXG;4@f)S+8ss}bDh$(a)t
z3z%RR{X#D7piHsZy_RFkif=MT=!mW+3*WEaQID;!S(F~pQ?T@e*3aO4x%X0NQ2#VM
zJ$2I`vhELUej>}g7O7wKKgdIHte6TFJy-z#EZnCo4M51rcXu@JOpf!$3dj2Ce$FW3
zfvRZ##16UqT{^aic|CiPv$fdC8mSPax<ys8RTMS<bq<a5a<_B@Q<BL|25TI@cCkub
z=Gwt<oGnNJ5a1+o@>j5~zL4@;`Sd|zwydlDZSf<u(`N%EzE#@!*B>Q0F-*#X5$G{0
zE9+|dg~E}>#>O_%fzj+8RM97sTfCwPFW??KRsH#VY2F78a>W1mH|wQFVPK)*<KWk(
z{PzN!-A&+^$xPKYOXdktWj&I+qkmi^^ZMn`7WXBz^7klQwO*A81XL8nGgMDgJnwGR
zb{&ID$f-9Ku(8k7h_c3BWm(V<HP_0T6nHktDJ_Wwg;rfpSlxLE`y1-mUF^mHoa^fs
z?<0!M-H6>B<#_4gu+h(Pz$LH<u-QBB_hY9`u76D}LE0)B!YfPfZ`$77o56K?Qa3M%
zSLyslBfi}mWZ&+D5PH+p;a-CI>vyZe9@`F5`n0~OoAnM2E%Uim`BBEk&$=%M;@)aT
ztDd0=j-^rzspd2M<b!Z#Rqqiv;8U-{kJIULyCp2YUU+uDy1pWJmn|LexAF@R_0|fy
zSRoy1@M;<^p}AN+P7A#q%9qjN?J4zm$T^(V#-l{hRQwjlPu0KfrRGAs@D6!2-G2Q;
zLT9u4le43vF!dHkAjZTt&HkUwjEq~*KA{<&fnCk#zk4UEfh?a}C~oapqEkQK_TTYa
zre8fYW=U}_@HWxNTk7i!Wa-aeDW!BW*1WZSKhd^XO@Z*G(imgD{9uoie?9<p@qZ^!
zr+$jIMt4huALo7E?QNms3sDlABoGI8z284i<Oop3hQ8Sf*~yz$H!!!Qq%LMU_U{CG
z#|f8rSx&YUo|!tFnK`t&&Rwtu?91eo@YUscIYyd3fJ^%G&b>3FEC2Sqp#S(8EtsLW
zieC8Na#w`Dh>?&>e13ku42fdC4?pTT&X{Yx@3yEoiK2UwG@F`2oB;7@+jYZ>(V#N<
zv$jwZofpHB*4BSk@ne85jD(*doWH+Te0wO!$w+;sD3c!@PU^JVe<#+eSdnMM9!r!l
zM;k(Q5g?p3FDD~S-#`U2$<<L8IB7Za8Xo1Ehx}gv^*{>0_l~^OLd<(#h0-$}Ya)k2
zQ<Z5R$R$H;CklXq$MW7kM}P8J;%!mF_hF=fv`Wewv1j*rKZ+MdcgseB5WA8tChX`i
zRgA#SjeWWlyGe?Yl>Hz6!~cZ$m0y4TN=^aQr;=O>LsV>wEGmqtP_m^}E9+i7gsozm
zCWZ#pGL19KeW#Vw8x_ehJW{G_DOAm=s?1qnz}D>r(wIoLpS(Y^Mg%x{ozD>70k0l^
zUb=}|bx8H^xqesip=xSuRE2D{REn?#(9)6T)X1qZjsx2!(~fLGl<p*ih(VT~ZR;+J
zpN#7Pc~103Mv-#DWH}^5l6BZh)$Wb4_-SNt4znqWEQ|)L$$;0Ip>r2`la$|<6<LF-
ziX18_h{4bf1DEke4vG6?kpfvNMoEOrFbsTqdu6O6mwD#t_Qd`DUAl#?k*`g4xKMRx
zaR%j~e-JM(&x8;qmOf|v;3<0+HCqvkUsifg{QC75tkukFmgoy*czb=vyGm0Av=Svz
zYotAcQRJ48btjoZWSTu&d{7Z*eoRRY66d;PT+_5VQ@pJ<U5wC`iCq-ZTI&!*#e0GK
z4oS7#xY9+*nuc%+61*Rn=b6?j&e^X0H^e=nM#WYGF-C4T$$}FpE8Q%dPf4>jMjnPf
zc`BvZ)}5Fl^E5H}36vu4S#qo{GwIVbcBDS!JxF}XIDi5_${L4i!HokoH6H7m5Vp!t
zD?N2W(GsLw&@EGswboTsKuyLFFirLu_kBEm5TA;z7G-qbRGCl<eDPmIwOF8)Mm&s%
zeqh^VB}TD^SQ6HVH>1%bO>%&u$xV{MLCF8GccUlbZxhXgOdYv4j<pN2ra)ec)(~^R
zC4tJFb6kdJS?k`HL29FzhC$Ymlc`F6wwMDU22=y<MzU?FtOWPhT7I@Fop(v{*=ub;
zC2A|=CJE-~CQw$DT}?4nLd>+9<ydNvvp`4A7*rP{l#(Plfl5^}tyQf^R0<W1mX!jf
z(po{{&ZJn!kR?2|s87wA?B%-5n!(?AEE^?N@hxeEgqsT7bTp7+_Ph(cX@pmGZouRF
zXCO$+0YD9dDC1%XR3xRkDaD>PQcWltF=h5;!&@h51ATRe>Cr;NI1*AO?TOddJ8{$8
ze2*1E1PwV?Mn6diYn6??)@Y0tqdIWz3Ip)nMBZBK2HeQLXMsmg!L>wwDEj{JZsbui
z>h?O*#X5cU2<_jX_;@X2ZAp#o9>_H^-6jS%@b-Aad&Ryttf@FXK~cQl?;WR=tgyGo
zJ4fvMI7wKd@nWJgR`cT@e&Bw;6LTc=blgAv=}!z^^N;`V11~@Pz@Pu=&+O@e86eg|
zrr_rhoMnlT?U#+wd%RMlO`O@oG*hsomO*J!ZPchpTayS0n@A<fL#aWx%+J64l^6rR
z{Pk!2G*Ig+qrb2(iIf|)iF^3{O~`-e73(G|0fuShdb?3d#W>BjuP7ryhPgCGH}qA&
z3WLUK$L;dWx-PuGt_;IKZUR;0D$yjh?*d6F^8EZPMp<tq4&G;~?eFVF)-<?Ll99jt
z`ikul{<(BM!^us#t{an+z(h$zfN`9=;Zb}F!!XHISi;%;*q+VS+99=;JieHAtP`%S
zwt};cZQXHpz?p#<cY!K1<R}wsO_gDu@YZu*7Xq?&dw!a^U9Y^q--%n~GGAz^(v0Ex
z=?f_*_U)C|<&O7*sOnnfGF(MTs5O*+*jrI1HeGwddBb}NYkYsaOKzc&iDBO&<1n7_
zH7!fNpHdBaJPX|`Q&T4E&s!lITNHJju_CpWq9q;+trg2R;nlR&6qT{lfGqR_f15~X
ztkbO9tMEqN3T0wpFpl-HbH9th*jibbDAQ6Sgosv>k)|l*AU1|^6hmhXiCA`AHS9@N
zjW=zu7}gLaK=XECoFok(iE%{KWi3;iynn4UN{je;#?Ov#udk#OFxpU4!DvfK4TYpZ
zq?$3#5c4KA+gR4UoV%P*(5j##S~rY!<kotp?--|n7#A_2dkKZL^C%QR1LYmHZK$DA
zqb#C>hVG)LZ5J+8_}zo&+IurqSl38S8!46ESc|u;wKu#s!io0v>(eB`q=!Lznl3Vp
zR^jKyQI;!WVVoyQsZ7(v<$5Fk`m-cdYZ$H*gYvA~gYh;%X&n|#yghb&vVAkM;pNL8
zP)_s1AAjP1`M>?I4E93FnagFu8pHBf2`xhF>ock#RW(CQ@|-DWFx{JT1oxs8TL>R}
z(121aT78I;S`)a<IP)8x^W0jHgf~($j{n{D<aZSxt2F2yG16g<v`0y~wgdO~2O)0#
z=F_rpN=d|rs!6($-0(di`skL!mQdAnlPEarX;ntsZCe2$P5<(jzes0nWG8tpJ<jDM
zTSZ}$Qbd)?mJ+5m=J|<|BkLYTmtJL)AX*XkNH(yAjhHs^S4=Z4Yhu^sD=lk?Q9T{F
zR)%q4+k~7H!rWq=xa#+HmlWdMvH4PEkyK0+W!N8LFfApMQ(_nfT7z}jvDS+(rV2@0
zY2*;jNS*mI3bfdLMUql2utV#?iz!R$?BVw~_9Df;D@a_;mQn@oP+s*6wDI(d9GgLn
zG3<LFrQG++<ZgDXXl*;6)@Y{n1F&E^6r~Oih7)6~cwsd6-hg;t`qvVrVr#OUaVvqr
zds;3$-tMgHicyAr*`RX<>-(LU3%MV3k}YS5Au%|)4^T=rcPh<%nTcWRNe2b1ErS<4
zewrt|8Hhpr6Vu!c$ED)jNXeXni4X+pHQj)G@L{FHP0EPsv+Sj-EroLu!)HM|hc)s{
z7;Ra%g{CB>z#4nn0E{+glcrGJQ&y>}c10Yg+*_Z?#PG3~#o)Z5<$6MU+B8xzXF=4)
z^bN$B7=>rFR&lvrB)xrKJ7**k+gciu;^8SZW0XP}5pw`&qw&@e%aIasz%z$G>VR%E
zsx(q8<W$K8u|o76O{L1Z3FU<=p-ICfDJPPtSP+z<6cMmUuqe!zE2x!qeRLjF)UaY(
zsHl`rLCkhvPCj_wA~g#KBGXQjT)tX6*QQB5#PtwJ9Ota}vXrmXj8=WMMU}PQqjy}c
zH>P<e=fYq|mhFxoy);u~_H7~O!mV?y_s2I<5g%bnnQ0zQvBo@!#X?n4C5AOJ_(5Wg
zg2dvq`T<cMIy5ae$GJ4_fYJ|+_H(Xd6|VqV|JHo{*=YTnOCewXcV53Q;h^78MV8XY
zp`Q|}!t;FP{qah{5Y}DZ43kgz&%2%rR3+*9+84eF0caS;Nt9+GA#z#aH&UwXdtw+y
zKz532r6v2x8f+6Yo9ZgLW(;dz7aO<0b($sTY~8ru-#d;g(A<}o7ji3n{rU@gSg9>Z
zw$k7zG`1*es_S%Nm`2<n8vc|7i`7O5ebp2xMpT6_*C%2s+}{@BmI*!9<xu^VCZU+u
z=_(zDHZ1GPeOdYvvV*jblvc?xP&i0qb8;*xM!<Jm-U|6j>z$*ljNW(IQPP@Zn#C9;
z#%7v2DIK8-hW`5ciq@K!mlv$HEX$*tRwcUUaQTZj14>D!^s<x!sY`j4!Z|ret+j%k
z_r-crd<ND{R;gdF7hYe#v2D@~Z~KNF9DjJaieN%Xyi0Qe*)GeS=cgC)9(j9w<9d6>
zVi@P~RDR_e*_ZW9ADD)LQY!avZ?eXG9@xVwvSKZ%%hPxv^%Vn$S@rtm8E-wyLl%Jz
zqsX&s{ZNyYp1g7Vx2_f46n+@~8!^U_fp_e&Dy3km6os4;F$G#H>|y0H&umR3;MUr{
zrB$JwA#Sp2;d*sF4n^KqY2*YpELj3f6^c~lVA=K^t!16O>05_o^K9uNS<Urw#d>+I
z%d)a=541LDB?5-rLt$0xYYd8PW^MH#`HzjlasVnmd{JlAjus)ukqlu>>$VMn>y*)_
zd>kvVrq-&v^5yUwhmj-x2oFgY@z9L}<1mR-yH*qpYg}&%B{}{u25;LcinUa+S~E;n
ztQp9;oQ7>@27ytvOs>Kw9_v1iG}=Z?M^h$uP)QOzS`$28S6VUDB!7>hAS-k=)F2_K
zx>>w6%;Sa2W#)N*qBKcK*~8AhCsL@qeS7fdfBFThEj5a8X<q}k>rKQtRlGTeQnMiq
ztVWb;@n}Z1KIPV>)KdxCn*95-RQhwF`QY+CKJSnJ{`;V|&)4r<snB{Pd&hW{fzsHQ
z4HFbONN8kL<*cknr985vrfKS(emkozD7+hqF|%z^wj?&`u8p>&Sh&o$?j>o&92lI#
z8xJ|rbng^vBPWTrFbY(kqH-ynAeMz>5_gkW);roNrpcl0z#cNrDN+v5$7<;yE%Rk!
z-L~FI!DC$j*mtCpEOI!fAOyEUQDR|VHk2|9UZQYXYdGt<FY@=b6S=ML53R)Ww38?-
z3G85dT)CD#8(O$BqqJn(^@l!$KrV%_NtbxUaae7MH8Ktpm&@go8(*WJ4Bm*5bKm#w
z`?xTULziU>3~=0p*1Epo@xy5^@56qo!am3Hz;74z+3xu{r1+8tCqc5M6lOosQt7I~
zN&HQF!k}r&QDTPDFc?}A3H)W8L5Xi6$0#cua>5S-smVsr6arhVq$nYX&N*s<6lF18
zO@(=yW!3q*lZvDS(=?6RiFy&rSZ6zz*nXpYlT>q={tnd`e3*4Vzb6MtuKU_Zt7W-b
zN>LKxYi1fI=<3Y|_;FyembX`t3QyD2xwzE*Qe$t#EK2pn@}{UsLR^PYOwPth{Lg93
z?fxF;EY^K|URp`vQ%WWQs-m_owM8bGlGFKh4u4UtGL2F{Vt*v1K%~2dk7bU)Z<K_`
zdu5$KnA#}3@9IVhT32j`yGm(Y`YdK(qc!^)Xa%O>!nQ563Rc_G2wr-lkZXgg8U2N3
z6{RS`9s%Jk58HsV4v-igx)|z|rCWIqw{3xD@PkarIcJ=eWQ)iB!Rs$?Y(8MDCDlZU
zjeB_T|KiVBr<rHZI8VI2y|eF|OxI$bp|#58@<iMVRD;nXi1B_Pv<EcBz6Dy5NFA#s
zAwr|+fhYor7syEvn4>8Yk&9SZ6tXZ<p-)-$aZi7i-hUWTe;c`_{zicC_gufL_!y<|
z){?8DG+8vHQIsfalN&OHOnxGyT_6wNUCmvVIQmgKvM4d;DWsl7B#=?bjY8&fxstuy
z$P!d2s+oP?i6LS{KS&ILTszLG6)j7IjW&wGPdx=7o}IUImYgel+@)}MOQ`|KJb$_J
zkN@x!X^W60+I?BpkCbhr7)FP0j$9MR*1wQ;ypiIvJXT7nT&F9}c?K)oiK<FgVr$8p
zdaU<TYd;JF(=hQ^7GelEZE;@wJVu$W6l^`7$DCdDhXTcPBcH+OV|#4<<OLDP>)0fF
zcIv%2=_U`-wn^+rN|K1YEK4t9s&hUCYYmr+B!!0%`b66J`t|FnB9hU)p_C{V>6u&~
zD<y>Jl!@msw05yU(-NVRns5s>QDfnHxnTxNiW%z-gB72b(u&&sJ@Y)VJyuj{=&Hzj
zB5e`@qO>Nifju<*U~xvXEE^@SgmptJBT%BzoqSb2+CP09IEPtJssDyk)k;JXrmOcv
zVK|<L*r7!GoCc9omIh*w2~$*(M}mTrpRGsbr&4gX;GCFww{2&dj+6^|jRGl(80bia
zS6&LuvS!9{;Q8rBDuGl5g3YCLG~g%_Rk*MoH&g487pOZ2*ga!S;jO`VbxP}xL#lPp
zUOiP%qT)K7?-~t-ZnCYXq`l;i_j#N~hKrS=Yz^MYex+O@DMW!(b7h{d6vX2d&kQ@&
zO_PQgGS0~=j6=D2y<DlY2&A-@{XOsRiv$w;zH+^0!k(dtA8i;MMu`AKk`D%^Ns5%n
zlqDvoy1pSfLtG=<B3oa_F-gJ5kw&4BDHmBivfg)6lxODc?M;#sdI+*}4x<YD9!Wg~
zecM)AtWd#e{~<7FO@ctRB(zb**VDR40FJCYId8?#Tl=5su2$0=+jl=-pC_KbEiO{1
z+V38Ies}R1tie>+q7)4@4A!B{fVNUKTkk>-e&rbPYmxYrZ2J<DCf=Kp;dOcK9D;nf
zlEpL&iP;aXV-<1jQdq4FtyXeO4A!HYqG|&bAyQkV^qWp;jqV31$olGW9z}_y8rRFr
z%a2dY^GMbfRTRs%^7i(|vM!Q0X7q0&XG)Q9L1Rt#`yh~(kswQ>4c>aJcI+WgV!=-y
ztgHy|ZjkMpR-im*9lG_MEw!%!XbojQu-MkS<=UmEC4cyF#J^#<%-2qci>gXu&kZRi
z5^_)(WvG2UZVEdg?5A9MUDva4Db<+g3x~_!Ie%_Vm9+w{M+bSd?#Z_5Ba&aCLDQ!p
zF%J5{b#v<>9S+>@B6%&PirKIVcb1~W<r^%c9~rC|3QNx9zTYSG*E5Y1H5KTlUFkqQ
z%85d09412E#Y~z-eWaD*cD?mORuUCNsU-*89#*W9RRPvITA!q3C3odmWl?KI*T&#S
z9@~R)9DkE|ecZR759wSoeiWg^wymTn2lBp0tU0<)a8}aEYi@kKe>=bC`}_O3-|2|{
z$+^(F6jrMS^#kV82cMD>Rn_xA(f!X$t)yH~S_)nWK{{b&I|L|PRmp{8Yp(c7GRKab
z<Gsi^(~3TSW^3)aRa5EXfUW`+_h47m6h;{d7}W-=<@v5D3)HVGF^EdDHh~uhQz)qt
z_e4WuM<)i>{uz~$N$H*IRd(}0ll@_}G0)G0JyNnvIA%7o@pcm^Q7J_X36)HbHp-wx
zj_gaL6uLaQk#ocvNwv;7L(%AF80-jxW*TQqpGb{W<QkwDoN}z&%C_zdJ=H_x-5#R{
z6pqI|aG7t^wDZfaUvbv)_IgJrNx_!n8I97Wd!D)<r%DofJw$8g<o!uV@Ts7jt`7yD
zr=8=^|MvPV_w)BltdHw=6`#?2DI!svwoS>RvsOh3F?f&Cnvf#aSRw3kB_UJCG<3t`
zA|4QNKZ{zT6irIyB=jDS2S5GvBdsT;9-Az+))?yuIpDR&J4=p1mXP%vCFjNKbAQ}<
zUGG?<@peF2D}Etkd49fOz2)`o8(+RWgD!mg_R9VK#?yQwE)jan%(|}p{PWMu*9)ri
zB3da5kaENgGpQu@O%j1`PdDNovD&gdHgNWY!dh(%-eD(sy|;NL$3iNZ!FkLuvM#G^
zPBoU+3au8bQJ@VegkEgqfUmip<FR)G&iXUcE`)%rXz*BL&nzl{unCM5;)d*KDkn^O
zm`SH;l1Z~i%gA9U1Z<3A9{ly!ZyXUy$HD5o>YQEW=}DGr38W{%rKR?RO(~{H-rwV}
z{`U1N&N%2WKrIbbEBhV<pI1U~i@2epD<~%i=dVBiO4uX!_Xmf;wiV$yy!DcO*EBUX
zk>z#{a@%%75ST9J)YFq~pK@eTm&Ks8rIp(E#x#z_a3xnFiArOfJ|c=fCrls3`-nJ_
z=}3rhZ!PmMQd`CuLk<z`ZC^uwBnW@{{FLf^+`(aFa+1%*7(=WHt)nzIF_TKdI%yV4
z%P1pB+6M=F6rw`}QTjb~_*V{9oi+CKOzG~|GtM%!fs}IhwbfIGe{cuK!4=~Ne0xlT
zpJDmq9`ep(x_NkC7BL`u)5R$n>l`s8aw?Q~7mlYiocApEg_J}=r+T_|3Oj}-DokCu
zzrPbgq$csPNl50-^?D_Rop~H-O_73O+eL(tN~8H6&(kW~5=gN_()!>vGvB_wGEX<I
z(}l@QY%a1a3#E#&RL*^A5F3SQyrPuieR*SDckzmqEZkDoY)e2Z!#r4Mj*{T<_J*#W
z)&y47MhnkZ1$nYPSFf~#bC+rKGhFr=HvgDd+usxvJ~$%%v54^V-#Z(a_P4Xr|MB&^
ziqGXXGY$hG?X)Tnud(ivnnj(vQeucy>WO=&uIjqoZoI$06S}-_7zR?>nCA&&9IZn@
zjmEM*&d!W3F&1b*R$tGx;+?=#`?gVwV4<q%?(T+$#FC8TMPfK=qABqo#SRyEKTumj
zja|=9PoXU$EuH6y=`vIMYdCiG2yyEv6#O*cN5}i?oyT_XDaoP+idnod&Pj3J(=G?j
zb;Kx3YdbkNLfmlH_tb&Hbr#8UB%BHsjlw!FOV;|%4vCPr)C}*-L!b;LE^lj`#GBND
zGyY7NuBB3{Y`JvK5>pmVV45Wrd|SvR_U^ItwBgF*@xWToI8Gna-*IU4&-yq_j>BzR
zS0n^evak#S_Z?=|+wF<R;~irRUtWF?!``+F`Pog7AtaLO&hr?_v7n4*9vr!p-*8Sp
z{^1X7n{037D5yOIRkSiiRxLQW{bQ0wBNs_HHqN4z{J!j9GF_T0g?||4K9y%t0>*qk
z(4`jUc}7Aw)5)(LoX;@<>CS7a2HU#wbdz+8(h4PImKaH~??&pcQ}>-ea*!NfLn4cm
zG8`)iN*06YagdZ!XsY0f#-h*9Uz$44FizCk`t$JHZuszCdEZW6sMJi!i5MH+en@YR
z??sadw)S-l*1EGnsU22y)^`6%hSq~d``1x)Lwl32fo+g`dW8{HR|$<)RF^o5qANv#
zzW`d5_qI}TJ&SFcJTdJ&?srPAynTD^)0)8NwH5|1#l;!RICyD(=-FpYOug0)C}c|2
z#^L(<34!#U`!rgKHOj=Hb?1-`##vI1Y%ySL<J;>uF1J~xN~>7b4KaQiXja4px$$`2
zh%qp_2{$=%Y_y(u0thru>C>qE>>4FEO>0y{E%$kV=>j-;&U$XA?23}6>NC&txtTid
zjo%cXzcWnq^Yy!mPfEhCEb9VRGhZ%f*HeFKJXVn*SL7zM)=G+VO2iORD4dl8Zdqh~
zfOMFeM#c{l^nw*aVB1zX7-Q}^V1`B}#n>gH3EiM(DM;2hf!~y7+XAgf7M#+W7`h2f
z%TlvcD$X1H<e(Wsl#s*P5+!C#Gg$AL)kw6mxiscPSVf9FUuV2`<Sg+i%}QY~g9MZ+
zrAZ~yj6A?-33wbv$D=N!mhf6jGO&_8^EoEA_XR3U<3Ot!?;YdlNhO{K!Eul&L`Bqq
zlo~n36V9qtyal#HjmO&LBWZh2`&l5gQQXGsIeKdi!#MOVr}Fah<y5`3CbIWoIMhv?
zMmO)BK&wS$$|+@*W$FHh31dX{CSIk~rynsT9}DQL6$yK;av=4^)ac%khY-&PNH++%
z5KHX*k>a8?*NZe7dyqJOqcvzP_f74hnWoOJ(fd(y0JWBfupivUbTgi{9mcD~C~@jX
zGVo{VY3+G?!#tdw{VD6+whpg;tUJg#b@h}8Cl2euJ*IBd6d27IfmU+=JdCkDn@_Uf
z{NTH?Om#VY-%#46@^a89d7qsork&FLH!&t^gJGEP&i7R#@iYpa&};%@RDX?y)FL9I
z9zIQXMuY|pYKdnjeejMsms3b{KyybbHC;-q=!L6BjD?n=q)KYyn;P6eEIX=oC1e%8
zLMu7cRp%$I(V!#+LE?NWtqM;hpNl7_+=D_jrYcH~Xx&I@!}}Yz+l<k|snyO)=Unh!
zqg17|%rICnhE~8BUcUTDt?-vW{|i=GoEu0vk#gxfraYsc7zRr$8>tr7ZQ=5CWtu03
z$I7}y_B9jAj#0A7H|{&NWCriC-lK9NA@s(PE7-cVp&UD#q)U<AQLWL`NA8|nA7TM*
z^vPR(#1DO>mVCa~k16RhEO#^wf3Mg8erNI7Q>4a9D3Qx_Wt>M+g<%?~O{(~PT~VqD
z-fRvxw~WOxB*GrK-ENdYUbsMIjT9qGyHiR!32o<vXbdrn^M9Vm*}_sl!*#wgIFG6d
za_OAFK<<S50Z$lh$*Hi1NCWmQfzCL0#Tdu3tSsAuVX};)<uZ&6#!Huy5?{Z4BSi7}
z$ckafm3lnxVtR9y_eBhL+6jTIs^<Q>kV2MK2^uPe(lVFpNHe+b3R#)_#5h{szkWqE
zjjCYv#Pu?DGPiY+f#kW#;nL8?GJ8oRxZfXVp58Q#qBt_YjfFWlvttc(YmISoJ?2La
zg3lWYwWeo3#r|iSSc^amF%w(*!1Imi(2o$9&U#{yH5r%rleC-0)TPlmXTE-Ypta)1
zAAb;-FUb0VVVs1gN(r3EYnS^<%EB{=<GoN*rfJ16%_r`?C&~7+ot$sl(3GZhg{DG7
zZ6e$6>bsI<)wLgolbdp0WZL4~mZXP+bYIu4%UV_Evoy+RoHM<!<dZ);mTn#V#|P9T
z{E8neZ*Q*%jCU0Gfg_=tCLXvL62oYFKAk|HwIb6#rZ)y6IyeL>qaTSmqAjO%K15Md
zl_EP@rD3sn??kd&+WA_IRy3m|W347ymVi=Jg7(4j9Faz&oye&WL*j^GIpdr<G}mhR
z%>u&X^&aq!)(SLe<M2wLB%=lPH`-%$MHxwGZc1Z}Kv1<R&>bmJ6{Rc2KrVrGS*hKy
zd%fN|2dbE+2~?w$$iA;=qsb|d^MIc`esDdHtq?-#!VcknXyP+dO1wyy%fL8_;WWey
zPGMT5P>4M)$yLXfAAe+;J!`PEmUyi1d>LOb*79_FqNE4+Z?D8CYePyboY7;Uue~l%
zQirACW9xb^I>!{*WL!Re`!_ObA%*4lRntA4Ky|^$N5MJHkk7yVjnYjiN%{C4g$I6T
z@hMHV78+wvMiWxvv93&2)*_=V#(A94<RtO=Yd47=euraAp!^^T?=`F|B6yK-KiP4v
zdi1+GcLSk%R+4uZ?<7?~8Pc+oa$|HJnwT9G-S?3Ssb*AFXzlt$a3z&Qj0wFf8c~EH
zRg7=`Hqvt8_4O4=OO6<4xm+%k{-zIuXWdpgL~3G<8{Q3gJ5X|hl89Lx<E_Eq2U!AD
zG^0^83*+DjyDa3oKi+AGVYQ};Hej70#GU8)M#zz{2O&$Fcx{ZcO!Gjkuq`{Wic8rU
zgBwN>n$-&0qL67O?7J9CyYpKXtZB9&=CebQCF-n^>-Ey3JG!C}LyjLZ*`x3r2gos+
zA6@=&tBYVHozuieB7kgs{Ps=O65nnwXlp19Vo2QYk_K?QJu%<po{`YU)O}K#+&)xv
zb`(gYq)JuD3hY{FN&r=(G_6C9#)vp2w|XW%SR*`U+LNRsG%@MQste!uE(uh0$dr#w
zsHzH?eBj~&DX0EE+w<24{AjI3wFtF|XvIqQ+tc+1O_BB__bm#e6w@#gqAbU=#`Ui&
zP+Jf8KHu8|_Osgd_e-+goY$x4OyEF!9ww5g-Sgy{Jcq_8oR$JTP6CTdbD=1albAma
zrpB@3sCE_IkwKSF(p#o&jcuQ!kUqjl#fPFWN@JZED2-9f(<I)tDtmQG0<A3Ch@C=V
zpk-Q-X2BYRx0>5^!h20F0-4Jy2*cnlMvG9#Y7uCR-m>LH*jC}BtD$y;VDKYKO?@F>
zq*RGF(pvm<*XuL)_XjZ+8k)EJooSM(sT>)UO5%R@xMQ5A6ood1`{Nx@T&C+eky5No
zo*6+woR!9*HJtJCca`njr9jGuSfKSuL~GqWSBEd>x6i3O$Dc(9RR)GG)K;H9v%gil
zeg3>s4X5Ah@4bE>?4|@iRjtc~!%C&HhmAe%gqYAuqUu|ZVwVc3rI0AW5)a6&HMq<d
zA<eB}8b@?DakVNgXYU=}4MMJF!DgXPTT}{~P@|~UoMKyrFEN0#mTlW<Ml)Tn<eXX7
zm6so$v2J8p@1S?)X<(YKIA>Y6mABUi`?Aw${NWEjaJ@7_+HuC5UJ~>2!k61KF-6w2
zVNrNxAVb^|DK(59u=7mKvLZX~5oJw>rYa@uERP+9Vc#-Ow<lU|TyGPDA26du7sJ2&
z@)yuDO(?6#J)<hsj2$c~6;c(5PdhOd>f`28trN$EWm|-3tpY(F3cY3BWl0o@5RTAA
zM~DTpC6soQ8rgRtvByyeVg3dKK0*;cU?}O*wrxXeBLV@fpl$47XB=j#^WEzp2h02W
z8^(>qU7D1A-?&^Z5`_|Fa@5+Pow(`8c_6h+W9VFq=5b$0oxn~hlK`zH5=M5<NOwGp
z1E%kiDXl}d%HoWL(W7W+H(+QqUD)$NQ4K#2T%Vt$!%u~M3Dhc5;gBL_tv&g|Q);D#
z$T*F%{vnH}hJHwt?oAtpiNTM6=IT8^|NIO0$3qOH&S8!XvLy)4-*Q)nifNf9J9Csm
z5<n0xXkAyDR-}|sI-t#jGlM|FDV?Nr)sh+@MM9*>wF$puj7I6h&-7ttR2C?bL8mH|
zl3C+Q3bOv2rq2atrH58#CHrKw1!Xi^2^9a?PjVb;dl1o$H8On0*m!?m`#J>?dMG3W
zHRpmI<m@R`)NyX~OqT)U6kAv*HFEvK6=M`3MB*CHR1WVQ$~xRQqwPrC#d8=#VGp9D
z^G+ssqcww{FlJ^CJM@(VDOBE;S1CwMQ;Q;oW6`BW(}=O~>n~sNqvQH?!TX8l+sL}!
z+4cu5cfK=cX%hNKQ*)sx(+d$;rO-!Vpcjn;UTc-Alpry!QCby;rl#{roaz9rI!B~>
z3PLxfw^n}B5WvSm!K3he=70_X)^Qg6s}~<LdHgt#*M@SMvId+NsqMaR<RmJqq!Po;
zV5~qCVuTcfptY9KE4Fo~q)gZ%remYdxb8mg`&AHG?L6C_AS;!=lnq@I_jTv__DpJ-
zPy%b*2sv;WN4)jiu1|dZ^;Z}jdT?m#_~{>Kv}uI2k(xkIei+bN4&TvVc)Tsph8{t#
zCGj>#)@8wJMXrjsudmGW1*aU7spJ~a*`TmE<+|F&F%C0XC#;jCZDkd$ib^HKh&STZ
zm<BOw`oVMk;fqX%v!^A4Y0`0;K_DYlb0_x|%{1<B?=;<*t|OPrOiVjlTDm%=qLQSH
zOF)=roTq+cY4#Y1ISTG=Eg`lpvY1IRv8;)hBx}u&l0Q{)A=Px=7l)L7+a$h5)OF_E
z)01=Va84z~!Z>&aZ=kEJ=4qx;`lO_V!z)FmHSciNu&xWWia!f2NzEl^YBMxs`{lN+
zT{ZPd{;W-JbOau=S{fsz70QTOu~dOY<$;s;%o^F}6q6(aSR<QN#ZJ%&X(g9VraQ38
z%5Ig~$mLkRCTrhAkfe+w$gxr}N;a2<ApGOo`vdcX2c_9#qBM!W(S7-v((+k3H*iS2
z#m{1~wn?PEvE*#2HF9KkRTR4RgyyDcrTcGEcmEqnYKYrL8zf*-AURp}UTR^Sp3cKZ
z_Rv5n60={MOo2)(RNp{aima8+Nkj^oM#xo)m2DW`m5fc&1!~P_QA_#o<9R~RG_2$(
z)<UU~80*=gi<KplvIt_7hN_CJ6j7CCEU`6IX>4V|kCySv#1>X^Eifr`Q&6pheQHgJ
zm9RyWGE^kY*I1aQ8LDhvEh*88MtP6HFiscteP_L|0$CPG=kR)>rV7?}lPsiA(8{5!
z<Mo#Z*V`3s6s2fJH)6Hn$Cnr4nh9&{(5FG$ZVDHYwQqtg{eC}|8h+lxq-hHBDK-F<
zcziw=2&o;7uE`h3xqK)VKZ^<u#p`GF+wrwTsKDv#`d2SL&re?@i_DDx+%7jB%YzgX
zIVFs;jKhc<HO52g8&92<${j<Nl#GTP#zm5#Q>u)!;QMP=M(y{V>vcZ&Z)}?+0MGMG
zMz)kvE2RemTIYCu{f0G`;pq#PX~rn=*`(4SODC!uT<3|YI#d%Zb-vtK){V#gj`tHe
zwH`ZDc-%L}$;iUHF0XAh@OUhQJ@fqh3`Jq_wA8RhrmY$a+8G(&n`BZsYp5lojbYzp
zwS)H_KgwEQJ!-~b5ZItL{^igAED(#+(%n|Ye3`k<7eI+gGRZh@jV!71gCnP%RwOp1
z<V?;xk7ecM%a72I*W$gXl5)Zz=@hET+TxruVGU<Er!$i3U@`c~lj1Rz?EMB7rQ0wJ
zjH9eQmaV07(~jea(T;5kq?EYd@3Mx#3cRzd8$bT|g;Fx#zP_RL*aZ>}=j3LNQEnXP
z2H`jqr)Z&6$#HACPcd0kNm^;Vl|4--`K=VX8<JxbQ43n(r@jlOAEa7K<ZHtM!cllD
z_g^W9M@S2FHjJV+GS>3->)ZLBfIwUGJf6QNs;`FD;t$hN(x?w#)B%~jyu6$+^<i2y
z_E=qB$T@TvQ#@xWO5BTZFIEb0UkWDc9zJ*=jpUhGCuDm|BIGCytoO2VqUvt`RA(wh
zDXquPWKtBrneOkIHBMy0>5x;4(bHIqYLd!abU`_NdXic<Bj=RCx*iB5LK^Qq*NdZ8
zCF>Sa!6-51=5nltP<<-vT-!Y0yy*d^fmnnCEwy5-CFU&qvikctjT5RWwk=3hQB2gL
zdbT8ZuW2=rGh8nhwq+BsMBmwy(=IUnaA8|TxjBww4<OBGWpQ5K`;rPxN&a1ovZra=
zH}(*Ddb-glTrQr;&s;Az-o8E99|3g3C`s;C^eLskN9vPT>$tx@uUS8S??@~-uD^xP
zj(D6SS>=FM|M&jRQCvPlcjx_nK&txRV2ySPLXZ!`&{q~qm6XaW^4V#cFi!LB{`K@7
z#dVSKvFYx1cQ`kR>^WDGDjk}0ERTiT^~z(BWn$wn$-!rA@0_C;125N3s#``susv3$
zVd8eZQfnjZk-dn~O!P1i;>H>lRQ>_&6+}&{g>TzCAx2VeY-=X%5^ZugzQ^&3n9~BS
z>Yf@b{)*+X%MD^YDON&`ERUr}o%BeLl4(dFXG(#b?db(M6lv?aa=l)=c;L>{(-V)!
zqi1S~w?&~vHK;zk5r=|EB$Hy|Rb6NSoL-7?5Hn-zt8Y{HT+~{ars)i;{M_jrhqdn>
ztK;_$71y!y`baA{6n4i=dPuPkZ28zyyRNHnMNNWaONgxNPR@l~8e~yY9lXdvMj!t^
zOq(f1v=vTJR#0e^l30FO_(y-B;8=T26W@;%s4kdrPMW6E2rGDZqx9S~SpuexRRuXS
z_<>ecpVUL|77fONb06ft>@X78;*c93NzTsMGdTh{6>$fSe!X7VwvGG!E;3{7r3pcv
zxoz8ehMr8ADIKe#9o8zWk+lQ>x@UfsQpI#^3|b4ipwtmxBjK~^2&XJaC@l)Q0W{Wp
z!r!gQ6a@&M(ducsq*XNL{LDn+PY=sX9}@l6aMt3iJ>Nqqx{1~#ug)H<_+gHObBB@~
zl#<h@(ZoIC)qqi!99MEqOxJ;NzKKPoAW0~t5wc_(h7d$J01_{yJ*_svp72h@1lzWX
zP^n9qOFE_o&Auga><72ksAi@4YtjHoGh_O+EfGye+DJ%&eGkO6aGAgKO~;;Tn7LdM
z>m&3n1yUg7U&3d$TF(3G^Az&=-~YYqZ~y$aX4CeYKmT0Z{`UJE*Re$X_gufL_#6tU
zqxeXklq@*|QSXEh@px|YGv8jnkz&L<*~fCI-e_IE>NLtWnwG7kW)xq_^&+xgSu7>^
zvFSU5wz{KqV;&`5C8dmN@^Jk0ho6}I$m8u!Z54&)c6mZ8!`ru4LfV<<iEn>-rRs{Z
znveq8{l9tpmt{$^>`W8;_DwHlc8*^nA~OLbPynh>B_u^S=>dfkz7l7g_n0rh2cU4k
z0f&4I3Z<*Nx~kd5Dj<>fh;TpWn3?Lna<JA`(_`+Q8QvnG0-o-1E@o<~TlZS)U;q05
zK+J_T3Optvf=%`4WW%~HFG_w0l4VzMI{a~YW}YYJ!$AU4N}J%QwT?!mwZyU{4u=Wn
z?8fxztTLH}@SCD^&*=NmtTm1>4f^wGV>F@m&>@Usl$Ug9&(>{$WE~hB!mwr5(p*j(
zTn2TUyXoi9_1X4Vd}}gK)ywbu%tz?dZj)Q-22<DTMZBB*z~keaEm;v65s{>SetKb9
zFEVNF#%<FQ@k_fAReJF2>b#<A?Jz?ZUKJtC-f$N>v){0KXJB2go8`9eI(OD?fv3`x
zt{d>aF;0U}bJx6WuFlD{e$;`+by;yPV4PSl$ipcDufQv)qF+mEozCdUwfgYk!!}Ia
z#}s-Nc{5nNDMMRpNm-a>-iM6^iNY*GY!IryH2)hZt+Y~@w9?B-;FfWirNx{Zx!vY{
zT9s=Yf~scg^J|S02^ZZE%l38YE#xGLVbj*N^%j$F%(tG*I9)%vl+81#vjk}p_?<Ib
z;O=^mhr_&CSZqlOmGiPtYGpo5qJ$nJ5AK17!y{9jDJ`kA*2-W+_%@|Pjz#8@O5^47
zf_FmxcsM<9JREp_Iun;9dXvtPQrYg!<-A}JdUq?5WIHQRc_{^l!v|HuR%-kZ26;}J
zpc6GWQEAZ8l6jn1rpSjM&&=b9KG3uVrZyR8QHp-AmiGgo`{Q!~<Ihz>_mxfmo%=bT
z+iU#vUQa*1D)?Mp#G$f8$=<9Hji#1G&Y`vn4u&vtI2|}YU&X<&3e=ENp_N85RhcQ0
z-3luxay>8dn$6&Cz<NWDY1;rpaCF<<vPc8j;mI-Mj3+ILycEvQ7m+KkLTYfuFyQh2
zNXwOn;gQrbFV{1zImYRLvyECVGRZx7oRLAx7#Ct%W%gy9bSk|eYR$;vjg3}C2B>Cu
zJiSq>Y~whOv=&&_3*#X386Q48@o+kEI-LON^>n4<Aq1&3vT*u0d6Hhql&r3fGPkjm
zMJlKDO3gCGy)5Sq3LCuO=RIq*hHX<8!?L_F`T*WMJ`!WxBzAh^x-Y>VG~qTW-I;5B
z;{*hblx4}sj~_%@rOGsUR=a}MDueYpxM~P~B*iSSkXC0&b<6rt<n$<;O0k7zn$FE2
zDiKJp!5J~4qZ)!v-N$tmn~!mz3!54f08vt|Dm&8@b_-!(J`B_<9!6HBe8ft^Qews+
zL&k^sz{d|CrT5?rR@IH2?=yZ~bL=(~CGXW0v2RetG);W?@L`jz>47$cF4-hdwRfK5
zbd<nS;qol3Qe*9=m!^*4_<rss-+TayH2(V#q&0>$M`B4BXE>b>41)*LnCD3=tyS;8
zOwiXX+GJFL__`8;HI)<v&XjyzDkw@q%;lsKQ1aJ?PKZK+1kcv0ST;0)z^$$~Z!D=K
zS`pjwaT<C1@W}ak;pyWO&U?IZoS)9TTwgFAmKRB=ip~nP2D3lXii{&P4=t`@=yq8}
zrm4t$Sm-8-QOFM|3dHM7P%=fOBpG|48gdlMnQ69%<}HZVCX!0fh(lnxEJE(U@%Zq-
z+s8K?j|b-Q$cGQlOhe>!I7%Bp_ncOsXwa55^Lp5`2aWqcbN}D>f#q|7YtM4qS3mpL
zx+S4(u-kp``_l21PNs$-j6yAKk_ioA#97a>3Tw+*kGFxBr!%7)@ki+`#FU|BF6$Lz
z#3fT6M6+Sp%hCt&vTmLeO(%o9HL<sjQVls4Ew#n)%Nxh@#~0>FU@Yf6xn@#KI7R|8
zB$}(rontWBi}_U+%=j=_eLx$jt(;CLj5$(DWSj=ZaZs9Vq0s5HlStw(2}n5aIL@=0
za5b5jHqcroIFZAtrSSgkJKnx~M=6=#|Ni%sT$sm6=7uz*4Zb^p8qEq3W{?pg#}EdE
z<K*)jWAN4!yzDy8I%N_$9gYi1?-f`?n0qTlEu@3Dtk*@F<LbdfZL&G_LBVu$G3^=G
zK4aQ6=<b*@#u!pboX?UC&hvp7S6*H&N*6XF=`)E!g)xR~L<v2u;>POKn!3}6ODU41
zG^s*c71<o;95KmE*Eo*&ouc%*^$Nr&PiBTJOI)y>_Nz2>Yb?2n+pP3}x`ce?e7+FE
zKyZUXbF#UMgp11G#3nQ_eMoEFH*ar}CtXFNE0ZtFvh{#^wc1+ax-57jP@)@UV{}$@
z09TR>95+#Hucgw8$mJMS2r*4aHMNMSq)hG`XpN)<1+SvEr^AV1^x^{R9EMC<<3{{x
zDrwUxz5-QRS?{%wOQGq8ZH)xOT&vE?iN_8!Ol|7sDWnKvdScQfkYgm!R9UavwJnvH
zvnb%#CLUrs3zL&19)k-kapCF1nK4G@)6C!lOOhes;2h2cNy5CL235))CeoU4UgU%>
z=Zh$|WHD#l*kRV`=RmG&?bKwHf)1jOpgqSXtgyik(oVpVi&$0ER5?FiF<72HT=?T3
ze&GCkrWHxVtG<>~>uYB%#=f?%@9Sl+JpHOC!{-CfC%qe8ul?w+x&Qta5^8<*jQ@)T
zpIX3qZH}WB%*s9y(>P09GUmbHaExK1dv%zWsHJUK?a+1NG#l>n@WQre9FGswnmC{1
z_5nBjr`D*846&t{X<1;nTB}q+R#a}SbvniFs7b7l``4%zBZ^&=DAMr7R1q0GOr$kS
zRp8~p%{gtbqj#E8BJ&2`iq&wphaN|ds08N?x^jmT(=@RxLJ~+Z^84TZRzCmrx@A{g
z{TL)$3@RM?{QN?SnR%YcxnQlQm_kW~xJ0p$aH4=c22qJF>BdSk#!xlOu2p&&$K!!-
z-o4|)hmTU}Ok(#tXb)jqBnk<ZoJCeivz@Lsz856zi}m~OzsEVt@pxbyN9}D?a;l`5
z8A8>jsj73QN={-U>@<*erq2)ty2M9GeWKe^n#^*vCXcu^O$Gw9z!9Znnio4_@4Z+n
zq=*krMU?6$=kL8|o+qx?D@$Bi*M&wQ#)VR(v2VMaNi!Ue2a#xss|di-I2<OK1G9!{
zoV2njV5+J)SV9mOu`8@wBkJK@cBfW>K>M2S8EI=`fVf<*0-@GMGeU1~)^M0+QHIFU
zlsDG0E{g`AT^3sf`Y~P@q`?cNyjC&Y!ypd3My_>jnH(dvRMzW)_ajua+G(^T&|;Tv
zX*$zljKHAdG-_3ucN>)5`jpmM#!-fTTkDWvS38iNV?&a{k_*;BEty<pekHCePai*O
zBH}R!tvpw?s?ShOV>u_rFoU<4BE2Gmqhd&lurLjt<+|`}&k{T^TrVpzu7cXAc<R3S
zey@AhpZ5wlDy4#W>R5rw)L7DDdHd!K*L9IKx1Lo{RzyShvx;KZBKM7+nB@w0%(edS
zeWlXj>wfKT_rSagQ1}16?HRRx5_tBb6u0*_3JbPv&HlOL>w-@e+G0PUp1pMO%H9Kt
zYZCXpB5DXJ6>>_nT=2%|K%Em<O{a#PwTy=&*UOdlx^kRnTCN;MneuPBfRn7Ul`1!}
z3GJ5x!3I*z)YJ&0gDM$9Z4hgs#7L<PXFavZ*R2u1^xFCbmQp{bg>h0*&sIrkMNtL$
z9UY`|E>N-z2(GKlBpnVTNntvWOsbT^x@6WR;auSD+xN8AczJo*n)K)MnNo_{3lxsW
zBLJ7nMeu#i3cC#YIC`P68cXFSWxvrQL7~o!F{l`(xjB_OaNh9na8hB-b?dPJuZ+(+
z^y9tfbUcB(nMie82&rV<jmgyHy+w6FHkFhTDGFPvXZ`}U6jBt6j_QPd>+H>lEjUlL
zZBsCp;ZUi{aFX5k&n_Eb_ZBf*VC+4eu775|kQByo6w|&c8;kcI+Z*E9%6YY_?6E{E
z!nm7ffnCJWw-AdaM%ov29;}nixmLrvUO664x^WauAS}m+neKJ97k2Ghh~4`NklBwW
z_9U%HMJgGvWpI`p8_swQw4x2yDC$5egU-`9(n5FJ6yr1LBLPmNmV)<K)b^yN4U!zZ
zD0@rQq|D9ckl8}nye^l8q`gW%3>fcdxnZ{Zbo22tfN}C(wFz|BH*kSZ?e@CT?-7EM
zb9Nit^$O?aEbwD%;y_wx;G96YDN$195>`<|P?d{yDMYFc4kuxM#!3!VKXW9j6Ydx$
zu8Xz+XsVxRFwW7e+zV41hK9A$`w~8a%FYxWG_CTzoYI!a<(M$Fv0fvshyZIV4O4|k
zkyG6uN?M^6LorSJ$x0~Eb<>~6eMR)8iew*{I-0Tn+56yh54rW%*m}U75b_gyKVNAk
zCvVWIT>NU)S#Xd6x>_pJFtIKxQ}EQ3iHl&VIcIW{O2FVqX(7sp0M<fDnJI|F>C@8_
z%XP(h%jtOFa=8$rsJl9~6L5+dTWbm3Mp=3JC0B}pp*Q&eyv2Hhv7T5eF&EGc$p?oG
zj@mNDdI**0>qoBF3xEIje@mnCo8SC~)A6MJ9+8>>7>^S*WxO|>PBSGZuGh0{_{xY%
zv2Z>wq*yr|kGy+-WL?jy0Vfapo5x3OWUE3naJ?=V5Dy!5w3Go+!(l$EwYFmz1dhsC
zY>ADPl0W#MRfE6`@7}%DSsEE2e13Z3<@uS`gjwaC-<VaM2`BKG=#L%dStzqeuVr&a
z)l6L|(Q$*dy2ngwl^7E-7MvYaY*Q_`TM_nI=}mswjQXYnUHGtTepOVh1~J>KRJ$K^
zl{I1vLMlOqKa0)ODh#~Zq|Y-9H`$vQqw2g3ShdI&?K=^{w8j!!+(;nj^I5L%Fwk05
zY+r<J#d<@C2(`&)9><Y!8lg2AaEy_hBPCz4)@%;8y{W#3-L@()1hU^{#7#D{R;Bs;
z@Nl9h7|yB1M13v%+R=s%@5Pvug*f88r;u4!=?Ug6=6}6${_yZfN`<G77mkOM=qM^A
zUaM&Ot;<RS!Wh7aduD5mFpktxSeK~Dha{Y)Np;id!yqsoBHpbKvGiZ6PUsKn+^QIu
zs9j`H&^MZaZ+=GBkg2pR!@w55xp=N?!WgT*LJjX6-cLN7-f_Jy#HAxC4V+H=E2|If
zX7%b^Ahs$MY3aR@*ZZew2*IEfUY=ichH6$-_rU3N(Eeg3w?ZjmV}n-qMtD_p-k&j|
z+F=yR>|-f?&D;l{?L5q_$JqaS(~H#C0=6XK{x$Xiudh*StPZTUf0^K85HUt?y(}6e
z3#(M#Jig&D9RRstag|ct8%y&t92J_ysIqI!w>c7PMT|2=$@ZpEO8ejfFVAO|MTQ3t
zhr=#w0B59U6P!SP&WO&{svBd+qQ#b1hW4aqR7D=&G^GtYY?YZ8<0`pE9v)}rLtsr8
zTB{t66V}4Vk57!l%s8Gz4=i||UY;;ik;{o!KupwfS}_FTDj8HQmYU(mAAY2iRXl~J
ziMU2+qR1e$=t;`>obf*J_;3>Kwk2{n9Pmzz$V5WjZIDnj>b^L5VqD2dgh{>5XvB48
zU1k1enx?IpYE*J2=2h5<D*tjmNB!9VPKGtF*DFEcw!WG5zJ#@Qvr3S87@0>AI=I+V
z1V3>u!p!T9_^!VvWxFH`rpd#-RVubaJ9X;`beg&EEV8B$RaK6a2BWI~?tjEojBOJZ
z;Zm!r=@t~4gr=TF`d}No{o7ROsr3xGwSo_xoJD24X(CaJ1SMmvR#SToJ)vq()40KZ
zy_)PWPOm_{_l&x^m?puw)=DiMcNYz^4sl6!B>}frRGm<fbKaCL`aK+m0po>Sa=l!s
zEz?@zaF`SZHatB&%Q~)t2-GS)8xem+Hzq+6I`g(Q=_RF{w)v}^%2xT5Qfb<&l*F{#
zlQ*gcAA_uOYldx2P$eKXZv771oDTO2^RB6%k2S(RtGRJ`xni7Lznq1&S1UrCvP#&w
z7M5!wtx~mG?dQxo!&WQUzOmC7M`c&WUInPXR_^mQogys(D8(etVv5Rajf_L!;rJl;
z<{C+Q4KUKnq}FW-Yk@ChkuOv=`+jTx?|snd$LE4e|N4!UX}90+FysC;MJ4{HQM~uZ
zR|TJ}Qtc99b+L5uKn#cD1DEre+M39uID_?$!Fj3?Dr9XH>kX3|@IfZ;*Ja`GaMTB|
zFdt?<e)vc!nKdq?wB9bV_E~JTG}kaqDk;-MI$WB+12uPdRERB+OJo=(a6*+HrV$LR
z>jm48o2#rYQCYUMs@(rc7JVs&aSDvX$Phe5nM}17a#~0|+ttZ>jFinWskmsfJU?HE
z=_BvHIm+hI>SizT-FM$@8ektkegurp&%knBXiX-qZ+N-WMJ*;_l7V)){M&u&*^GS6
z0#XH?PA5KoeB$}(LieukKfA|wZ%uM(<QkiH8+O`C{LeK!J2g233(am2iB<LE(qxl2
zMpHU^Z=$|>tv&W!b>k_e5C$>jvsQ)?#fM)0^Z?mmno@+>g;p=3WtMP3NFkT=6=OuY
z$vR<H0t$&WrSvz7FQTTdB!OWVHfB?4g;El6)&2l7@n37#U~AgsmS#Sxf7Vs7fmgw2
z|Jg-$sw<E8s!$sA-VYnuqgPKN+-bJY-<kwdQ3mf7mE4z7z&lYRsU?Y=OHQ0irfMa2
zetwb4(m2wZXi1gTou{f3ZNHDV%1l_WYXFi!ZNH4rM=elYxh+fAtiHBaaFM)@C>eBH
z3aC;s_Mg!!Z>^ImxX79^*5C(GR;e`;yd_r|-5?u`^R#M31I>8VOEgd{=kuZi*E3%C
z1Z%A%eOW|~og22bs3H_(E0IBH@P?`u1+6t|%NyaOW@(xW^e0N(>$-A2U$Dk=Jul>>
z$uqb1hDhj$tWHC;JQeBMo~wTBM<MR}K=%ZNer@;bX8&`&QtDstue%RoRnP5Ttmo6!
zd%c#u>$;M1W}GH2%atWYYOBO$rPac5oUl6KYiZ<KS=WTEnGgmJ;ee}-M8^7t9i&Md
zritkgSQe4l@!sD03r*~AV@e#457JY?Vf;V{hBZ~35BPDSM)9L@&N3(i&Ki}fQ25ST
zC&hkh)NHZVlh?|1U9hg<sKoVx501lh*lZfC^FZbNvM@~t7$z(M=L|7cnHDz|8$8Av
zhGS5fo{Y&Oe)3rFsTMK~KMok@S=PkM^O?(e<#;?%OXG*{KhkI%RoS3t{>wG0u9vWA
z<iT!Sb62ExqN!!Xhh@2ll#I2FmzOik^~&?}3$+QHDl~TK#~Evw4-<W6q&M-c7Zrh|
zC3$aB_XRl&qq3>`X5_Ub4_Ft(5Y7z*A84hjK&U_)K3Kf=v5K0QnLc|{I|Kz1ti*?C
z2wt1e5=i7{bPmStTthV}7@e1U*K{^TWIHjfZV8NJ6#|!zPNax1A_&>4^l_TAMD1ry
z)66&z>XEdwdUxVnDb^R#AF@V7a%;Uc;Cs*jHfG!XW|6ZP&B+`Zx9e}cQh1}%j6KFi
za<SBA2%#qudhMD_*$cZ&t)^3gSQwkb`G&($5EbyWW=gs;9}I`-M9Phv8wHDPc4K^5
z<4{FPypy3(AtqGDVZd2ST9ZJkTJd^o>C!MDm3~P@WQA~8B~*aXzFXObR4Xc;X{1kO
zjip+FXuD69&P~&-B!SLCt+J+^4K(^dt|wVX*GHo@p$%IHt}8@KMW%}#=Yq<I$w+{l
zBVZh@Spn#DIJ6@D>Dr{q6N?W+2!^hAm{a4!hl{9H<VLM_Gdt8Lfj-`0tWgcO*Rwc#
zlEJ?6k>Gdxd5of1dM?g8MYW`N)su)Gz<RaQ`?>c7gr?UE#Ae<Reo{R7x}J}Ku)?#~
zzhh3AY#FDKWm%-eu35~YrWrRlN=uZINUagZfgF?IqSi1laGVafU}!c;|A$1FJ+WAE
zwXE>`{470&Fi@KzRf`J;k?^4si&QKxFE5z2o1dw!nDLC`AdpRN7*llPK&v=IEe3B7
zV$V?22c#C3T5#SFoUnhyKi+eikJKtcrEiaKIUHu4YseUrX-25|#(Re2NY!Du(THSA
zZADDWa16{?EFYiGEZ3Em8`g|m&zYFSVKWuc`?_AQnz2jsxoAU_PQ+*#9VOBJoK&aY
z4FY$lw%3OrK7eWBF*FX_f+{r=)4F9ZU47j7zL~QzzGz7Q#29VD-Uja^5E{YVb#OF<
zk&+jzsbY7H*!-e2bg#~ybz?kr#!cCpn)_-<sQSW0OevF6w|;OK+i1NWX%b|{Y1HKD
zM!hyh8o)7CC{{@i;DVzj88|gYylhOhLY22wW<Y|>3CTyU(%e@OO}Ba5NOmn&0>aR2
zdjEzYyu#*%0T*5cqy594i@a_#kQb45IUE9m3#x>X=n4Vj<Z&K^3EUgqfDyek5&2b?
zRt96Vk6W|owN->Ehp<JxA2=RPJU?G)ROvtFOv%zeq?wK6W188h#-<{+7o`YN<SZhk
zC%uzaD;TGGb3%?7LQhx(O79SBOp&56Ue_)Yxk5)q)X5u7Yy{A`L$p1B!CDyv6&TOc
ztjrF2?`T!*t78<A;=YkP)z>o(95)!N6MBq6R5@gtNgzQZfvC|?gmmlKXxXaCh5o#f
z36?dJQo~p;iICc0x7Iy#+4Vy!9VOfM2KNMlP9o6FKxK$-Yp#?c5OdS2O0w7+oVp7x
zdr{H-d-nI);5K9UskU-oXQ@zSQ;$(h%eqVB`F!3OSGi=7y05UTE0Y~mQi3g_mQpY!
z^Rk@9t0IWbQ7ST*l48a*7<x|vnQ+f*=Iz5<p6o?bmg5x=ame%{26Lr|bbFtc7LtL{
zg(Q)?Lmd5FI<3{}fL^6QkSyM~N-7JNmz72(<;<GS3}GTCxhc!C(6R*9QVS`|0Hh1j
zjP*KvDHF*Oz^8F0E|K%HFb)#9o$(6CG*+FFczk#y)hKMCjczUwjJFPJ3|@yO6Om<&
zl$eDD<~-v#;aw1gh@x^a!BA_WscP}%e4$fgWmaRx7=ipmVWQ!KNVj8kGu)N*Js5PD
zD1@+CGl=?yz$Yonb#JxM_6xT4A-CJ)e_vb1n5|c_$CG<DzpM-L(BCMpUDiZC&ueI*
zp^`D?CQlTCz+lmctWV$gthq`0I48o1S%ae7SC2?0MOCV(!lqJOzHboDc0)g*l{c!;
zy<5eoM2b+o=WR{)LD;DD`_{zn+vB$^G^NDT)6<rbcDSxzChwgnJS$Z*3_)j!l4jgi
zVI&cy%yp$9eIn<Can^Lxbj}JZv=mBAR5C52O$e<{b#T~v$$Uj^!*E@$yDVKc>Q<{x
z%1I-nmVTbHp?fF95n3hJBKt}yqCTPSuFc54VysU0_r3~jL3ry`Ae+3^#74~5gy>XD
z6^+DUU>w9`jt(2Pg4gwE>TBW74pVI#ynY^e#-&=@DhG!`uX3G$t*SGU0d21-m1M7u
zwOs9o6{`s(*5Blfc2M_T@u06Ii=mmDVPQ&Xd#?J=?|V75l<gVcS2%kqpY4?HdPqOD
z>+@B?r_(-bs@p>EqJ>*yb{NQNX)RX989#_eN~?`ri)3NmZsucy4I=Qk$i_p1X_%PC
zNk|A`;@z8fvWYse#Et9ihVL&*w`BN|y6V1P(Eoe?Jq(1=6Q+@2@>tinE)esAwI1gO
za;~IUbwE;JkC@_?<v43qx!7(LW@l6iB}bf-&lg<alCQk13DYcVj##`h*F=2KI!7~Y
z+l*Vc7p`?P1T)+W&AQi8L!g2o41{6a)~|DpaST{H662K^6Z7G?Ra2d<CNP6|*;JLR
zH%2DA&8v<6hST35uKcPoDa~<(?K+*iNpo9ZyFM$_RHw|-C3r&U%$tH2PQvbfDI{TU
zxU>W*ZBSdg!EOCv^dX2Wj+*NAq+|aOb|L~uCFqTq_}!R5olS)fh-z=6s`9$BpK6r~
z+Gv0h2^^U<8^=+vYu(_Uo?Ui%old*o{I7QVP!CMqkgZG1bl7fL7E(_5Agn^?q~aTf
z5u-#0>kKi8^QrTW+6u-rhA{{<*hIhI`yj)kG@L_SLY4WNUN@(24feJ@Pn}^GQ`F48
zS7cI^SZjz%K)I<Vo2_3KV-&(d74y4MKWpyqiP;i_-fP+)pQ8=G`u_gwl}(ed2Nt=H
zhLk_;%k}-|)5HkY6koB>`tz`_Que*o(yCU?Hx}8gen+Yb+r?VNA1IgCLZADi-ve1Y
zx0#*$pWnlG`+knyz258l`}+0w|1IlyAB?|ze0A{Y0+Ky*G*%3_l1}DNVIrl(G>qh$
z1Xs^lstJW3?BL{qc7^H{n(4!!nq9;-VU1%N4_L&=sarEo<D?B;fob|f*A;zpE-L6*
zZW-BKklEh|Q3|g5&<kdplR%4@!m>ni$qeHlC39^&Kc9JdIdeRm2sV&Y5i4hF@WBf_
z2rSnt<2(_x%&)CrO5^#%GtPQOKeFak63AR>rVvyTM>b-0$Q-l@O~F>V!MS8nWN7l4
zLh$%-ptelWO;{LEz3Oz^pyfEpW|DHIT&^-R(t8T}ptaT*hmlf@!Vv|d46pvc?E4e@
zVY@CKdMo{fanudQx*(*r(%3IGYH)HJQ!9i@2m{UqVAX1(ZDNgrH48?6o8?g_#@&Oy
zWraqdt5SE+jRvA^RzxeUK6BO6={=5KGz))uxv(r(Fk&v~wZ}6}1IMGx>a0uLu8*ua
zsWx+py4i#D@yZtX`?~D|**!o`<AklZthKY|WQf;o71DmAE8VKziO<b-Da5pDFtUuZ
zSkA9eJeM9G9+WztxGtHJ1&WJt!P!CR>CLb#;vr-^aY7GCCS-$>-b~$Q$BaRt+uO$5
zHX@<~NV(rP5{57ki_9|Z6JDj_ck7K_?QH9DrzY_^rq@2%@ZL+ZvKtTRM3Vk{>ly|6
z+Ew^_s&RkK?Y+!hPlQ{7)X!sInF*QXHZ<+M$N8`mag}P(t35+kDA}{nZW@K3Ql5>K
z!T)_g-S-st@L8?3^B7(|-}}#a|LpI-U)#Sx@LAVYBC$qM=M*DVX3;AhswQ;D4!u-v
zvoRnt@)V(1R-K5pSaOP6xwl>-^EiQVq*a1UNgZ$Zf>YPRJWGbSu2<Dd(?KuWH^}-b
z)7^90?;E?d3WKM$LS7Y`st)gFN^atDpEH+>*db7*q1fSqVH`B8RlA2ZF^{vrj!EdC
zIf=hZG1@yRQi4y@EZJm(ToceFm@0Hqt7vI;u}!LEj?VN&CmCZ>(>jofp*5o$li<`b
zu9Rwp$f5URT@%*!YEXV}n#A0!qs`sxrf<^w#n~UMdxn+-6se?~o?{=fq)9JD0+iJU
zyJ=%phU;>wSmW_-*d7XN-4@hJEmUrk<+3(kXTPuRb*&pL&>zHJRqpkzdV=M4@6~bo
z8`NTj5#{}sb;oNIZz>xk!3w(J^$Z$lC6g3xbHh!!dt-~~(Y=v?ZSA&t3-Uc_>Ja$;
z+yO-;j1n9QwTWrp@idb1O09+CX~YLOpP%JtgA+JM*^D80oo5Q<)CfM{t%&coY-!oN
zO6<@c+IIh)+4h8o(*wq;ZMPAGV}oIlq-z|9Z4*pYu2Vy3?Opz;l!9&Amr&Z5Kx=Xj
zL{<A$iM%^KX^Nw)G-JS_NwKwLq;i!;f2V`rW~KUizh{1ZT1^>oRyGRtZNz~-fNJP7
zlTETiMkPO7uUEX0{!B8;JkwE>Tq`waAz;+k>H|Ia?1STNc$AGyu>XGF_x4CpuO|0<
z;|5#rg7`h^u?4`pYw@MytAfw@d{#SQf7`H|%nbyc4XL%^{B1H^nu3*PwERP+s?(Yn
z{m2joa?6Zv!Wg(-L=t42Cvq%YFTy4(wGjM(#gS@Ooi%CT_CdJob)l3=FO0BAtaYSV
zlnVG79_Tl#U(DWIY^{;kqK(HwZ4DPZ(>#irbpva>SYzkHbe!?CSD2{d>I<<p@l!F5
z<+^gYoHZ~T4%3ksBj?LmJZeVi6-?6%P3BWl60NwBGEYxWgb*Yc>tfhMB#~_0S*zEA
zZKL;YAcTRO#j1fObZzSzA{!${TDepqT3AD{79R$hk*~|Vf%u~Ip|<@s?2CWD@Nd<q
zRBgiY3eYU3noVXzs#3F6XZ>FEhhSaTSBv`=V0tiMQx&+W0d#-)+r93;>%7;h@b=pK
zM*qIw0=L1|+e0T0R)@j_QgI5!ivHOUsy6RssC61gmg`j;-45$@b8hw3HQZM&eb%ba
z|D=>y<4ReiB0C<BnsBX~=g%GrFQt%ECQMSvO2gSO%@c>i$aT4JT`r`!VqN3-aG*6W
zl>NBMHMEvKz$1$FnXCy@EmgmFV^sCOg_)MO;2o3r5l!=?eGCZ{BDo`h!YOQaBS_@e
zS+%aYp7_W_J<y?^o~^e|N$v$$-w0(r)KYY)RfbEgwNzI$;M9$+mr~r$uR}>id$zA3
z-cNmPHWArZeNBO7`^;A-)X2AGvls0`IpeZZ`0is8(l?i(Y=PwZOW&^j{@?vs+WJhp
zXJKEt-3QMeKw9|}YU`iz9<sZyfPTvHRl#S=Xj^@?QS_)dj-%kTRffs@AncMQrcK?z
zTL;GCRq#^Gk*rLmTNwO=eg$VCsegzwj+z_I%J8Z8mUUg|GaXH*j&n|Wvw9XOo6&Yr
zwL-h!IJX<78I|-w4G?fvoGMKdS$<mL@!_o~BV=Kt^$k}Pn2UN3C34DK&R23ucpGrq
zKXArF&DiFd##u4^#1Miwg61qi&wC!`k?SH>2FtoIgn?<A#jdzEthY+cFw&24-RfBI
zdFes)zK727SfRCx<cZmWuGj>~TnFWb7t?1AKp;u|y!&#!&k0Gzq#0GsnB1Ucj1gHD
znyi^x1k;x_<TqGq&z|XvyI0d@w+Zbtw;SHvF>rzQUeiM5p{Yu|*#GX{V`jhJ+pd4n
zp&IYyo`Lk_to1?*#?zAE_Ss5x<ZNNRu3IzQY9-joFGl5I_R!3>7P(m|v?f)@co;vK
z%jitN{&QPWWsR~dWwfA~#$`E^YoykK^8%TbC{SQC2s_gBUY;leh(bgYs|uz3daH(Y
zyK^=g<mztI>YuF}z_oU(Ao?J0L#?%~CJgK-)4I{^NiBPCrajEo_sH7@c-!EuHPqHC
zjfSy}k~26*P8sjnps%RcbPsE_TDO1q&$?%m?fZs(ZMD1JhiK?E8n-pJzjjIC-=+by
zl(MNYG(@6^=ENlM?ss3ockOi#XFKN|p<o{@KKHZEifqz8soXPQ@1eWb0g0b<eD$19
zfABZmE$?OSN0~IWRcveID2ukC><vVD*my&*fiRDxB}<7p21qKIKh8L%2s;}XgHSiE
z^{lICNgW;!U~kRg-hXn=alKx*4Wchr2{d*)=Y8hhFJ}L`WRs%x)mp<i36h3DiC|p7
zw7_wGV2Ljb-fshg<2aIIl4fs`js0|ZAgvK^1Jhxq<U)*BN-S6`$N9+VaFWGa#bgSE
zDcD-$<>i@&#}f|^4_vQT)@2brI}D}?1F0tfol!Onqe>bTrb!^L)LCmj2(+aPs9HtQ
z7%Q0<DEx>_k{c)euT^Qx)>@h_!VYEbvF!c;3M?x#VOh=LN)h*2Z=@Qt#<5<t8SWij
z@2x|9(uXMw!zeRS%2JeurrkFEqEq}mfV3vPDKK}Ng5MgiJxKMnv#+kUCbKo#YiqQ2
z>s83SSr2A9h?)~QX2xOW{PIFZlru`ol$4dBRB3${#>>1>t`Kv-H~x10@1T%z6cvQB
z6!vxt#@t}oUd8p#bvVqVma!_1n{p<-$WUWvUV87;xJ}fYjqB1!H=>4jj??jkHv{V>
zDvBiowbt01vczsfvpU$?8}ub-E|(0dByNonT7A_jg&VbADw!+_<%9fezxS@oxaj+#
z{oK|%t@%@i*1r1Pf?*da*4OULul@DkCqSLrU3WofkDlz;*;dJxPhV&McYm*bnErKh
zvt~2;^VJ|;Q9z+lQo?TP+P7!#OO;sVHtW;LDfjws{qI`q)?V3HO8W$$zu*3PeC_eo
zl}|tU-q)zDQCfwp6l!a*&NGfP=gXOt6T>7}W-k||3|EQ%TrsX-J>-;`rV(R0w(SVo
zyGX0Z#Ct0W%WlxYvntBMDJOw9mWA3X$HPIV{_}=I@Ab0wy!HLxrBrIlI#6a2vaX|5
zCl<KNO7J74h!Ei6IFYl=JAv6scba5A)>yoep}F-OWkWOK{efygrA6i!mUZRBj~|r6
z>o#g>SBn;C?dF{m7quN%!?;!jDmi+mFiB#X3f@<hQ)x`ofVEO_XjfDg88VpJC+Mq^
z8k|ZO^<g|~4Mm&XpQ||b3tJup>AeZl%TjWqDZ^{q$U!$`x!Q8*Oi{T(Q(0{#`e36m
zQt4q{uTMbdM_#Ypevf;_c8_!4pf=6$-0OKG`BC{hc<sB<`)Se>!*t-pX!W&QUwXZp
z!cw(WVyQH4&EkIa>)Zd`k9nLoCSFgn_7(g%j@y*F^+wOB>JaWtX}xJr$GZVqP)3y}
z#>y~2t5Av&h}6g|kaw1x8?6{MhD2VwzMhz`t>C;cyeMqzrq+zL9TwBM7wxN9bGz1^
zbkwQZ#)ufPF$VQj-75P=zs;J6_C#`j-&@;(Z?ZOe<%Cu>Tk_F2++Bd%UTe2k>|w2a
z(A_@^CEt3rpM$RY-^LhS_Ek{^`Ce`PS=`CoSh*)srM_mY?TG`xrV;3zgH|_br+(i3
z{rACnUrDkHH2cqdUp3tayq_9~zEWN%zgqCdh)jRZSq9*=XJVTnu8CnB=-hUJ8nnrn
zGD}_wX21uz@J)YC91ziFx~NctVJT63c$C6zbJGV2#bRZ0+qxU&y3zDz-8P>626Z2`
zd#^}l8yes+%|VsHO78+7pZ{`x5ryd@3k0L=9-TL6Ri^h_X$)bcWRZU#hXLoL;VEQ=
zCVE*aN_jb72xCwtp&<+d-+l9*Qlx({2C<=*joxi;h=S9K1c+FYk4gdVvMjd}I(S1V
zq9-?vGskJ>e37cEdnx(gwJ*~Zm>ttNV{PMF>UQmVmD8(-J*;VrR0Oml%q|^2D=K$y
z&8^Wc(l>G;_gSTUU%tNY-Si9%i!`5KB@za=kpSujN9CM_I^jLjG;LMSzSq*%g}g_9
zkorR>TXZcAeBZkowV<EJX<(8dmQsF2j4?WlDvT<Hlx)C-8wq9~(D$DyXMvv|rU$j>
z_OH%;Upe*9&~gRewmw1Cg=(!>Z%Y98^qH2FAt^*E8xL4jdU(lKlB=XutOuORH07LC
z0_Fz$usiP?ThHK(K{yaXX|`4ZR7tnB(IKa%=h`bL*T)F7LEryuFq2kJJ(%>QYRdrj
zf8>p%F~0k6n)Xq4o9afUuu2lJuY<<^p21B`I_HcjJJ@$uo%DXy{x#&L8BxLBy?0YZ
zbxglDY`sQNy}ngld&0~fq7+eLw^jZ9JokIy{xj(X+0M6<%f7Te?}N}!o%H#-AyeG|
zX>DWJ48tHgOZHaY^#zl3T6mlWymOnhd8a<R!4sTUrDarBMdpA^6KyrTI3|wDy)UKW
zJiNS|l`=dbGA4GDxsX!2Lla_;>)og9jDF$Z*1G}`8f=I}PAQd|3no`;g~vA!oUb3b
zUKXr}WnFNiJX{8K)-2-J0g!XW*+wgsw~ueBC5z=kaCj#w++he<Fa9)1=Tbn(2RUbc
z^V{Eomh_LO2T{8=0+|J^NcxSRXXzh}UV=6l4UqCD#z?~o+t5@V-@T<-qqA58RmE0w
zD~yx$vSP`2<D|mXnUNS@Q`bu=Td!wX7N%)ZsgQ-zD%Z;uT9OUNh?uCw6194E_EYbW
zUP0~ubjVj82F;RN!@2!~Y*cOGHtF5xh%nN#x>iwPzGwXOrm(*sDw5Tr|KB*D&s#N=
zqWqr4s*|bMV`E^RC%LCd5~M~K-V`a5VHm|4q|}|7uu%S8&rYqyhCoi4mzQT!N*s?T
zAwL*tLU+HMeHOaL6^sgTT7wH#dIrr9bHO>wdWp(Bw76PvL1wI|O%gBd8F}j|S>$|7
z(TWO%@7#@fC;gk<reu5aDfi2$=xV>lMy;&77;LR<OtaeL88oeJ-bHr1C1(3yqflJk
zF|_uZV?UQ0oVuG)+f@ry*Mu=@*4BfA7268=Y-;P$q+081Yg<Q6`dcZbtxDUQG4{P?
zx1zS*6;;%$wXH^no-2ql1WlN+&JnDGQa7W-zOUU6&G&t~p*_8Jo5kB7`@mDHu<xvS
zt$e+~bee2;R$Hl(q+)pWIse@8bxx)VA(VP^UhFd3)OMR%Zx9C0@_Z%6m1z{}X)XzJ
z)Be9?9esM)HnVM_y-{3p0>lE^w6@8tmr}P(vy1e38&s&VY(6C=SJfnQGN009^D$QD
zn?(bvZJUB*HC8hqS&YLVrd7GDqSfXbF|8o#7mVXfT1B|jJ1>FGi$%CNrxt9n9OeTL
z?;d!0d1h@98=NrNN?|RLV8!n~7x7K8&hYs7!1K$qNU<BSc3u`SE_2Sy24^j?7RiEq
z5dWgMGCLy`N2%ZknKLq$(j@CPR+O-pWu@AvnNY>*)P188>_lxa4I)o8PlNPWf&(L(
zR$hld`&`v9jH+&6#m6aEoDEX;w}vSd>pb3pF%1`_sckf?r3(Dl#V|YM#jf!m<I1S4
zKhbfMjjPn&vk{4sh9$<lF`<kGrCMc5zx0gGDgmT95p-;*^?0M!h_RVIWZDeWk`)dT
zZ8EG#>U1>KhK38$NM|S*TM(q)h!l$xVct?2){Qc^7FSXf_NHm#T$|AebgQtFiro&Q
zqmdQH7FMQI{YK#}3E)0>V!1+98ow85n{~~S#j6xhE8rE*K?+qm-Cc?Uf!mBaXv+Ls
zlm1KLCSzke0fDX3^TBQKYOX6u*J`&?4`C1~qpWaP6IrUN@=cwT+?o;-)S0zZRBER~
zWVb4$_rZXT61@+~O|Vc3%06pSSyj_Wt%_y6(fz>rtq0RESnV59s!mCWJxA4bYE)+E
z2A}EQR^}O-k`9c1UsksKn4FKwWi^$(vE~MU?crN1`_vw4yC>@O|IRt94AHH3(*w#L
zj^4wk1{JogRdpka1SG*C<t!B0u39<oE8d@Se4V9&LQB$LP^xN^%rgZ43g=8#ATehh
z>k>f?usi`9Zqg-PioPjjxSzc73d89TX;zpBQ1xOk#;adW#i^W(HAbd>kpVVoZ4?@D
zYtGBPlVGHIg~Y<`bLv2(E#j4vMQVLMI9@Ivh<V{~>OCd78PhOvI2>44aklK{aBtte
zB?Q6YKVCkHCSJ?bW^kpcr@O!#eZbQP4_SmWr*ULmE@E{M0!h^pTN7J^<LSt{Mv+Wu
zu*QUGD>Mn{SW7Jl?<{VZWKO9>ycrRP%_#bNrK@qnJdc`%7i!5&qi321XsWMOL<&bH
z=EKI^s<o*cjN2e7?}S<{vi7%ri$OA_Fj|@^7*pv=&CW_?1j38xnObWyNZZK_VwK%6
zPAS*4O;c^uDy)<~*x6(;)FRiyIZP`Q?U%7tK%(+&8GuT)Sz!cmkM+*cyxDqVri$vh
zu}Pcnk+m$$(}eXFeCBu@@vDlZE|pRY;4wgql@Kho7-E!jYo+0WxMx`>R@bG79^rhH
z&$X^+fn!9yQ?h>xwH4O562i#4$2VM7Bk02rXccl3Pa1K4HMqWM8ttc4qo{)>ltzSa
zX^mjLFz8Iv*_i@Ho_S-?<PWU#%=1Xfk?Zv;ntHa8iVT~#rgKNeVyl<rWLFi|1{J$)
zbbZThwbBAA=c5gDy#o8*%ziWO8T)MyQ+Tj7aLV8`Ms2Dk2{F!XnuXR{W#4u4I13G`
zk@wz)S-Weo(LkihiM&@8C4PH@rA*(ew~gA+tc*4|jaQ8wNh0U165b%g9XUqU`VC6k
zSAH85x^q6=w;(>($1+XDYnAa@H3&Hk#F7{&MZE8yKQZX^<LiV)wFz#Fk@CD_uCNV1
zb!{SJ8sj)|m@Uh)uq-QsirPtU!flPiR$qIVKo|V9+bqT~46o)eIy+^WCi+zJ{sMN0
zu9G6_{=(i}_#X8;9M@%A#xQIPdK^crF}%Dy%j-0`3CptZ_RRy%8P+xO<BuOVhSod_
z@xoxqO{G+HX2uvpOrqO0&(kLP0${L|Tsgm7WP=U^agFL08g39l66v19;eas$gHGc>
zS`s-WLXaU*84i}-NYzb$>(4oECV%>{b@r6pw}r>YN36A*sh=cmmc!ww_p{+29$1!j
z`wXv`MdhaX*5S>fbac37%53bhJGu5+xBtFxZua$ruAQ8P@g;_6H%6G%8Hx@K$Yw1W
zWvNAHK7?K>_6)L~WUslsmk=s+BhM9U%SIC~MNQzyO!Fjk-Xfx>)_R2kMIE5l%sd@1
zTB$aLRY90ywG?t|!X_O?yqDa&lQ^9B5^QRhAsVz|^Gck!&2msw8!;715w9(4J<fG8
z(c7lsjJdJL)R(6t8&U+g?S1#plO*?cg8{8`40hnUTy-j6h&Ros7_Pr<&V2)yOq&fo
zG1f>KBscyQT(p1xjq==Xss(#f!@h1_^>%BOO>_5s?;JMl`iZ-v@2f_yM^~Hfy&|+S
zYgYSHAz-;dRsHk5TDR@i^v869%Ijg{erImFeLH_0ZVa?91(NN)I)@*^_PIJNw`ck7
zt01e_uI!(=J()wRH-o`7B}e?kUQa*1D)@MXiAt*s-U<#}D^V47a~8JSpo5fy_JnfI
zEHP<QQ0b~6;EecK<dU!^Y!{%l_R9FG2cQ1%55t~zd+P`6865orgDzD28q?e_#{JC$
zq@i3J;~+z`luC(-99PPk80U$2U09PidA@t|j`Q`*<+5zJxHX;>E6EqNgt%o--g$-)
z<iS=`vQ|a3w3bF(BFlBf&?^NPhKa7!FwX~KT!n3f*<dbf#p~lR&)VNfU_^v5=1fEA
z$i~7r$+d|oQOyk_?oEF2eOb5be7Rh>UPZ`tngj~%k}m6-w(3Wkp#w&B?OKr;Eu%fR
zxF(#FdOc^E^C)VI-W##|O`_k^-v8M**n<x;`(UIBDY=ki(pfTO&Z6l9+@Q6(WsmY&
zg$`jmYq4Ien>&+EHhVS_i&SH>hK6Az1d&~l*HX)d(ep4pFwYZ>LTf?Tirsw8%6*d0
zp>t0(MNve`rVGUyxJms?GIV-fuMA-%_+D`ev0<9TZho3(f)BjBoOyZqNbL&iYUEa{
z!q!ccFG_8QpN}vr1)Aa2bJTRx_x8rBlZ?LH7=uQiEA3arxH3&6@7}$~2Tv&xV+We4
zB!$!rW(zj0Z{WI@oBlouJE4%)w!xWBJ~0~{+WSFGzh7G;$G9mybPu4Oy|>oHMqrn*
z->U@dk9$^MKki{BYgL$**S(Va-|Npy5173tv)`lkq=fzVQf-@h^qz>b|J|OnIVTW*
z7lrPhv!Cn!^KI|xEMXY1ISb^51SHGtKKEqK7>naJXLTRAzjS<E<zo${p?>5Arv_dX
zWwZHi78V(X!?qwdYIds_Eb~0$onyI*+O5}S>V6Jqe{*^e>X}1tsPzV0zX3e}?I(Dx
zwZxRv$Hsng^ZR@4B{XbP*>xYb4Y(9xX*mo=6^)w{cgdEUI3LD>bAx6Sa$d`_;(ghA
z7@}Gri!(T<tevK?lw(vs9QpT*bCz7h?Xr}j0iZ7|zXcuBn~79{Qmtdo66mHtK@H1l
z#?bJ!VXP>EmsXjk>Ggx4*Qx7U?WKvNN^G03WsDi;UT0v^sP^C-u1OE)rb#B>>)E4N
z4-|zv{T=h^HS4S2(;oPG0LMtiS<Fu6eREgj)yCE>8fxv3lRTW(d9YnJUmm7X3*L35
z;Jy|MB`<95q`GeUHI%AR1}dkydm>py_AKKlYdn|87$!n+qNbr$sQ^i0-`xYya$SV_
ztqSikCM9)D0^8NwjL+qA)qph<lMEq?04kc$h@ssEW%X+Wia_rh_afucTGM`+&U$fc
z;2ZiT*2wDBBdkf753}C$ERb_vDMkB5$_{+BcPo6h?E@VWw3f0<D)(9XK8WmRpFn2O
zoYMch5(H=>s3OKj`yu!9GWRnu{cBs*zB})J&8jN1J8K0#tT!FW{jq;mT5FrOV&9{>
zcHdUL>ap#6Ib&54Xup^3_p=R7RavH<NOlT#|G9f|0`LhDWBdHO>wqyEv5MUuWOyBn
zzjS<6@Oin)yop!Hjao=5vgmu2Q-n6$*)Oq3bufm3>*YdlGN)rSiuMJweYm@3*nUIY
z-<XYVXl@yJr=Z^Byf?K2c3*qFc=tictDe>Icw~*MYHUg7J<k(sjOv|Jc=+~PK7M@W
z#}6OrM2QYv$$}Ty@Nk$ldzaotRyE<0Gp$xmrxVjS(;7TIJu`U6<LS+2i(N{_IWG`g
zOqyNEXUaKoy`J&jb37h#*04rVUhsa@Ss93GGG}ZPA0{WZ5A$%?D%5@tI*i#DduNz+
z=gd-W%FmJzNOp94fNqdjQ&(9*Mnp=d|98JuLL}%8o3HWf{xkhyx}8hMx|N!(fy?3~
zK}vON?mDlH>JlZ}$8G_|s1iV*xbN%3d6l@gqK{_G4GwB_NUe)ddi4ov8#F2BzYy1g
z_Z`}8DxNuTK3@sJ5e7$DBI6kF*2y4o$z;91eNERUa8_^DOM(;0S;#g0x^xF$7c3Zu
zF^*c33~swX2v!Lzy#kU3a;=4eGDg|S6KP+mwt~pxShUk3)v6W6>{2o{-{9|FeJ|^R
zG18Z%Rkb$|l88m+{+sHwHGw&s0W~0!P0oAm(cDzBZ({@%x)C63F#LYg-Vexco;G@~
zI~}<vwwu)fyQ%|`{6S@?+8u-@g|Y<bLfPQFJ+rpMWg9ba7vvg3py)r-e|Nt>b+OZ&
zGbtugO|O#0J^FKB*`%D5L4SL#{eIiu_th8!(5&g?Hj0un1a_2Bf!4?Dn%%0&pL%>%
z@DWO@KCDeNo_z2k1zzPL?dqvT)s{;ZdH6xO<^t15uVPf)s!|Gr4yEl&)IGmk6ozk8
z_4{bto7nxLeqxw}OmgpUYXAK%pS~}*jkQ$Z`Q=$$9!>x3QoeV&{^N8I-L>ZzVYvm5
zFm$EKy(|*6Lh!8d3RK2X^sT08;=_jz(znHMIbX%*o11W^KsuFFBDE&%J2|D$qV#sl
zx{ATqFb1Kr8pALSSQmIXFMvEC<1~tTv4Msdd>Ng!=yElCucO|l*$08v*fzM7(nhzJ
z>;1Z_tjb<;+!l5Z8a=b?)l?6BotnGP3inw5{evcU0TM*e$E0bMBO7Yn)<z$k75GbG
z1chL%wbC0b1?yzJIPWM`_*SB4)-k-gJ-k6@bTCZ>GPT;5K9Ja_TkD8Xbi!mEf*%}n
z3qYaNB<6Ryl5!PTbQlR<Ox60D_by<)!<fQ2T74dpR&%n^4};asH**X#&U%(5>P(CT
zQ0WiJwd$WE=S*A^&Fj53X6yHK6G@t|Ae~c_Jhe{ls?eL0?2dWXlP_x<waN45Z4h<w
ztUaZwGjFxts?;ht+WkhnPeS+W!8%diaNf}hjMFHPvI2EUht0!IDLkhny<lNF(2Ne8
z>Sqwv*)1`3n;<BgFHmOyqxWD7fLR1l_b7rfX44Vu5OV)|tukiO6No;n-9PId0gzce
zVY}|N?)qMNfAo9O)uuay4YwfGUw3O=^uRBVk$kou7|TuCsJ4c;UCF$YH2wuDA3r!9
zR_a6nv}P5iOUfPckPT+=;@i??F}(LI%9f+GPnv@J1-36E_y67k#BNdc@9!a*eKphj
z2>TvSf4#kCQO}la5G?(?E=rvZB92;bn{Wu8$HzxLK0T3B#OQgkZNQQlrd%j3QgY^S
zn3?B+b-iJ?*2qK9D%7%QBUyTDAAWqsSg0h%G3Y!~!x)u5ang@)-Z6W};UF_U&mW)U
zXW=G=WDRfMJo3Yjk;gYD{9t%FKPyagrj*J&i(Xm3zQ^N{hlhvl8uyIze14{s_NqGR
z#};&IZy+}?UeEpYJ?qEbkgKY$-u`FqVTit2S}V+>ZX71BApvHCL3BgNlx<^aO_-O>
zWvr9+Y+8qSwE113rq0Z2+Mm+R&NY%U?Z$BsnsZfK?sk)JNhzx=jw8l1*2)}~WZ8|J
zBPAzXRqZvD_@HXbl5E&bJaiZaobg+tl(Sr4v4@ZfM-mnZdTw#JAU(!dBF@Tva8A_d
z`+e_KdUjn%Qy?j$6@W$rtwb6aNkV1&kmNE9o|Gf>IW-Ob#`bJlhOP}&X7{>kbThK?
z_upv^ZGLRfPTy~gwHUp&Jz3eFBTYp5elkuI#yMP;9$Igfdt;P<bh9a_t&u;oH{9<p
zTeDpQj!~9Z57Lqk-e9oJ<#!LSnOilrhvGVUYoCzq;lBN|075v~_eb~W(0!ljb%kbL
z?T`ES*dML64O$gSd&Q|8N@pgv{yZ2VBXo`YeZ~2wj<009i8!HPe`U)woHvNCy9nWV
z?<ui~L6v5?7B+c$*)-ZF*5{q0DvYxy4eXgJ9p2cNu%FzJ-Oum)LEPQw9-um-tN+^=
z!{eJb7$3N-i&Q06pp=s3L0EI*@$r#24-Z_IE9c7@<1F54^A_;l5{7^=P;16m$$0yG
zMpdim(|o{M&-3%MKoPa!nw81_dXo?trU_`o*acCAspPz5`!Yo@rN8$c7d)*B&M$D2
zRpTk~BsxAGF(#9W`p*oO>vh?#Vb_=In_hn?_E2UBeuEE6)gVKEU{wXU6q#AEGK(Yk
zM)O7^?wInRRF&URBRccxpjxkR_xG_@g>Z@jC{3HNO&MVrT4z*sXIm$>*0-JyU>hlA
zku_3<^it9dnyLHjwP$?7T$5{)6^^_O?nzHeXFICpbbMr4FD%O<J~~q%43;G&TI;a0
zP}pm&EXx(|L^rT&-ZfCQ2^|t{Rv@*Uh*_*0r1}}~e$<V!OYSsr=*(>susod}w(AH8
z8!p{~Y~M)x=I(b@lDE6fafc){wHWC^$C^HrS!E#FnXQ^~+ics7aG*w~0`rtgbDhoD
z5e3-{Uj+$;O`E+Xu(4!v6>?N9P5QV-CAmt$v>idALtpK+FH{c@`}f)>K>NV54=S>8
z^Q!l8U$O6#UP*S!4MuFA*G$|0x3j{i+Xs;SGrINT`uFHb%B_;xw>b9y?!7k9z0!M3
zDVosf$yjlN6>lxO^$PU^@l%ekhTUi)2_qLwCZe5lGRUXd@i2^B(xPmlx@{`W-cbX$
zjpVg#%6+8YXTJRc-{0)Lv*2DRa=%FSH=_r+{&$R#!{NX@&s<U>wMtYs(rO@6v8EAo
z=G`Dpr@#B-AIP<e5`by6EI79FFphyzGlL7_H&ipuDaF$oreUOofm)#+j_T^F<+?Ql
zKQabs>^7gdu2<GIGS8Ei?HOy$Hg_dbJ5e`>t_Gm0yhuZJi3`>nuGbf=H-u>*rMz7i
zArHh?P};h#+cmphuL`xPl5=l%YWCRgSMN!**6>EO;cDskUZ&mqLe~+JZDH3Uk~#Ya
z;lBUZzt`b}rn6SBmb?)nL$lgHsxr^zT%hqf5y3W4ssEn&_kKpt;N`O#dT{8^SxS*w
z`@BV=Q^jxRZ4?^xjKhQvP->#q1g7G=^m2sqoVRmot#Q3vHX64u0d)gY-9Bd><2d0Q
z#JFt9n>A9+c^_EQm0TLuc<5}$ULB}Ys@zkNQ(DR@1QsjsNmm>2z6)lmtd(Au)|wD{
znr7Oa4vxZ)^q!T&IdG<tlfr8of^vhtbhIBp6*X>~VXb0`kUG?~+azgCR8aEG4R*i5
zcL-}+S4bfzx4PZnwR-E@w*6kD?<F=9Rp*@g##o{EV9ieS6khE=J^1W5|NF{l9{~4>
z);)Y@wes5Uhd0#W){@xwkbqZv;QrsQg6Qsd<TlIa${Mt#`0dYIg{pz)GZo#RI=)(7
z2f(O?TDMm4A>g&o62?I>^0rCuZx_AZ<ZXI%?1AZcI517q=2_pr-Zz_9Swgu<d+#@<
zy|wWE=o!;}FJVvHZA}!VhhZRE7`-8sL9*LisAXar0@FA`YyAHEKZ<*6aD-7DM^lcZ
zR4648BVg1349B`&FjlI+5C%%gtc!St7@W$`Te-nB#?fP(rL;y(o%Q0jhv;%ynXFKr
zWkHv%PvpH&-E)!6Xibq}49v%wloy_!o_O>Afgw14_~Cne9SLFDu;_hx-eJ4W9$MFx
zoEzO3Q08QAP+MPo8=h?)7OCL+gJ<m>6o6De(#(G?>CuytQf@W~{d<7+5AqF8YkMVg
z(Yuq4Q5Z$0Lf1P0q$AK8LlB9hzIir1U$?!dZpIo+q)O@kc3Tw(s192r=f*q>Nkkys
zTVr1ovJKA3Ym_QP6r(e6<1i7|8#}OHyKxprUu$lztsQd8`DQ-XnhLernWzCrNFs6T
zpI)z5@rasr_&5oxvOia@L&;Itn4Q+#18q-0<n?6^PzJ(p1R{_M*EvHg`=pSKHr;4_
zOn$r8a*uU{wlz$LrS6gt;3h9+Ev3X)l;YbSaNC$`l@%Zam9_<oeFfLo*gd4Y{cQjG
z?(e$$S@{$<YIuLGo9^N3*X{|+eMR!6D$5#yAp3rL&xq^mW!rDfr&R9#9`}EG?-UJk
z)>>*0+?pt&2}bXMRoVsv?t8y~>i9}k>!Q;88ho(Sn(<*`oTqJYalJ;#9<9jOfHWq1
zz!IAFZBvO+Hp$0_hpq3@ap_ysHrqpXe{;4E<NZeGy{8QCZ}^6XA`fm_R}O~*J_N}O
zy~BC+@Ik~=4S~}<GkDL-(~~H3cel`#^wRf*+bA`GO1?Ba#tKYkfH91`Jgrz~2@GP>
zoz?ZzxQzx|lu>7b*^JU!tz?BAmPKYMgxs()3_+_0a9(C@tPzESS}N<ZVjVZ*w3<O@
zJ(@nO4U{G-lk4lciq&?xsq^-M&CW2ax#Qbb28n~w*%0+MLg7|tHN&slrte%=!4{Qm
z=iKI_VvNxH<&WohJaD~U)f!wZ)p6Z~t!@y2RcWL~+sGTOi8oTC+x!|fiaNUo(xPyh
zJVe2Y)kDs`V(7z>y@Kl*E?W<#VF=z~jX1HEyyBJaY%|gqs)*c>@wXuBoTx~*J@Zc>
zO7CS}%NS_}&vPHHj&j|HAl0uD8?3P`*M%?+jKjeB{6fhIuk_qoQkPldRjBzf;zsE?
z_KI(7q?<~!JpmFGhQ1j~MW8cjIl+c@-zvtOGp%H#H?4i7F3PKcsP&f8_QWfTY**77
z#E{SJ7-O60Who+Og_Fdvdr&p$U(nxo2Py6$xBGd)d+7Hb3fu23`+#&ytZwbHzAv=9
zpYPR6Yi*lB6N5+z<_1I7in=G!1_QR<Mitz?=lSU&f%$b0>vQ|#Co@0B;1zbGwVTXN
z?{V$#Rl9xepL%>%@JTrn#-Pwrz&fiYXNiZ?5f?n`vS3Wc+6YFZNh%49Wqx?i5--$}
zWVXYJ>02rl6AkC9A$WQHoW<-)hO8ypr>qr#GKosPsVC=>AW8-)9PY+Ajij8(CF884
zbh9Ret<s0D>RNHZk*hF%7!NE6g>xK4@}}iREQ#~^nRr<+4QXE1!u4`hsLbhm3oESD
zg2!Wh;4}_kWMjg3nqLTEWDFDL!-ds|0=Bb3dlHFZ8btB8H3@*G>qB+8K!Vjch{bow
z;G8Ee8N*8OqCVUx*en=B^9BaTx*%eW{s2l31{n;-A=Lu18M|PqrjkmMDm!<ou>_Y^
zCA;;`^$8seRpmD-#t0;`M(KBm?whdcEHeMnw2H9CQCcIVRhVR6`Yb|YE*M*haUo~<
z*`+PSECad2Fev^kjo?15ZmQ8BWLQhoxio=Ow~eWy6yVzfWUYq^O<h;51~!bKbh)&x
zw;{cuz+jzF@teULv3-X|umj%rD$r1(LTuh?{|w?9$x#dfV~SW~8A4Ym7s%3ki?OmU
zQ!-qaXJvQ_gqyQG3kG>Mr#xzcwTMKEa~Om<S6kz{EJ~S|d8d>zrSy$vp(lMJw`8~K
zw4s`2HHm6Qm)@ar8*Pv^S8|g6ZWCBdAT8Nsb8X@*YYnwIa_Q1d9*@Y_813q-FpVU%
zt*WAa(T&mRbry}lVAXQFLwa<EqY$E;swdp{a{6%(S?xF0`zHN;km@j9(;(8JK|OQq
z_HTv%n)J##gU%TFno&s5HlS7v4O?YzYvl$r?_sttosX$Xh|q+h>6}tW6s;<ENME&x
zA6O|=tF4x;hkHW?Xs;P~w}1b%AoO*F15twn!&aVOuV<FanK+Le`~g~q8ZjnmShN@`
z&Dm)V_|eLOGetIOrx}iAo73t3<ld{bfaz6|`twNpY{RCduOSWfMiV7rwH<D<L(@|#
zr$U&miPlsqnQ@+Y^X@I{x^iBwxWSM{l@T%Eyd$U{*ySZta-!sj$8Z=YN@*mjikB8r
zjyNx~6uC8Ct}8xR$sRG(ykdf*IYTT-%VFJ=N|q)mjlnunU9eR)$UZCvY`#9$c&*?9
ztyP>GaLy8}C&xs}i%fLq!rO1(v*gICL$lQ>EKs64r&bwKZ9>2HLjdh7S!c1{l9x=a
zk|a3qsP<-hmUpv3Rx`HRs#+;8<lL)TQEd>~antz`=}{>UTr%u%V}Y$vSX|OOunISM
zjIR*nx*CPPf(-;G>e(qqniUpOLn_B+t7voSb?wk!)2HzU7u^_;(>UEAk=8IqTzG44
z<aRUEGdNreI=^CRO-$u-%#vKVdh2u>aplDdJQ#eS422XkmzRqKjhwk&R~givG=Mv&
z-bY<uvU0gTGmbNdgQ&cxRK&pTC3XY_uWAroP&Ad>VE4L4RRA#B!mzS=>Up7%up2zy
z4V0Dz64s|NCK1u?8&RVwoLaC(Db7f*KoGX7jLEjUcSz7{(40|Hz^!!9X&bEd+KaW&
zGBpdu-L$%?Fx1*M%WQezQ87@hY5~8aQr{ya_s1Uky6=<RAGdzw?rZXm4XUzQRcvq<
zyFPTMbnjruPGRVZ4}I7Z^E%kv!*=XmS6+K}R*g*<TI^Hu)~lmdZ0%)f-GoIkUDxo&
z{M}bg(u4XJ==q!uN1A3fUBYA<2G(_9xm@wa64yoL^Q~q%4)4T_EQEm}3|qrdN`E6+
zvR;(Fk?l)DYmGi^W`kHnD*m+r7u*KMws|7wHXXX&2k5M+-Wz)J<_)LQiRb5M*{s%u
z+MHG(q95mtn5Zr13ni_XDl}^A9bpQr%Y|!9jN`x<W}2;7XDAfb6v@rvbO;e+nWqU;
zEqM_uf_XS_emb+pRRXUvuZ%IIC<3lN1YGX6)3;uj)#h=Gk#$+|HgK2^T<V!zvQVgv
zz#^1LPMJv6z%Lo8Lwt+KGGsdh8D=xpi0f()NhG|%^SX(7pCNWt1nJ*ZY7s(5Z=$xA
zUQKh~dUkdT2;R$vECzB}>8D;DP(3jg*}OW0r|ZQ#&pgizUYgHYgGWzLy!TYw2k36^
zz0Vr&H<~?O-+#UX4!z-f8}QUVVr^8@HvySJDTvZ@u<EGVsz8W(J@ypX{%o*=GfJ+a
z0&gm{o~gAG*ChQ`wI|O-dOuYefGOTA&&P3Om`1`dP->y9k!hZ(u8J*$sT;wd!#l=^
zEKO~qJ|4XEb~~nIL13<wFj*3ciV*^hk)*<!0auDT<?1kNc}-F&S(U>rn{StzB<RoV
zWMfK&c7uZDJ~hHlLhDrny~A9*mVVi^<vs(M_1E~^@ujcdAD;xVy8yTEY2F9D{W<O5
z&wvPc{x3MU&jr<e<=3U0_VA(wnzgSgxc&Lv@6}b;%1=4ID)>y}s0^zJwW@ZP_}%0r
zVwR?g%VUa#Ug$ZiD$zK}bh`POmd35<5|foIRtbpUqgeGi=^wd<^-cvTUvCzBYwzpz
zx@+u7mL-$iVz2Cj(3xwKlHGa3)AI+;=Vx-O*d>yqe9qzFk@<KMZMUSh16HQevu4$P
z^aKVlsG4sgrc84(S#Mp>*+dz*HgP9)&EfE18q+kgE|E2h)wHqbBt4s2Y~MgSKHcR)
zhH>C{IAV?Ac|C(>_#p&Jl%CBvjau=jeX?;hk(j_6ho8ocQMeE2tu;=k)ApXzIO2_9
zz<X99+=kS8_T`)-_<?1~n~nFL_|aeAn66J}1U@vPG?25{1sGG6L?M*uo>+z8aRxHd
zZ%HYMNUe3mbrptPu5>va>G`lFA^rXP-+QJ||JppyTY%Y9^7nH`aKp@faI7^dSw`fK
zss_}Or4np7oKx6D*Q7lBMY8zPi4$Ud(1Z56Mh>R~DP>Ypf`;+>TxBj7VVDN*ndc*K
z9tW1|MdyB`7djng#&P6wy=<HD{=UDuFWlI5Ui-PA<R9%(<SonZ@<E#08$la_A2`e>
zsXFz!?NoeWAlj|xCh(8L^`70{BD?K2R&#^Jx(AZA9%t{G@^`(P`;B=&I&gp7S2+I?
z$9<#!J^|XF!zbtWXI}Gj$<+PN_3!V0?-PplE)ac6lDHjvO1iV|vp;ixd>z#~WyZsd
zw>>IKk07oKtrTU2^+8KZt1@RHVugl2the9bgy~Z><B@F8=xjh$sh(U4w&bnMblyqn
zxi!FV(&D8QE|<$z9mw24k=}*cG}l_O-ik?<GP%y@vouYcrA3)b@+EWhp4oa9lE4l|
zhN?;}R0}x6%W}bw18?8HA?L)$AD*SRp~G!hL#{Gofl&+V+?a<0!3AopT$T&n_^TQ5
zIZ7pytN77$mDyKJGGjL7X=^OgH1X#202V>%zJSFs&j)hKl(>?M=)Bbu$=b{5((oX|
zjKgsA<yn`N)kKCNh~VcSk}ucm6>~T+i~;Av#xCr$Kww_o2j^t?a2W0}^`6!Bd8{tw
zA`d_aB1=>hM(tIzb50~-wCp!Z1+KLygG{QRlp=ABv?^@4zL9q@eGg6abJ`p7?H^Xr
zkCXrIb1uSA1l<w0wOmrdSVt~d9eLZfHbW4Pr&<g-sZgU6LWuJ+6PB~AvDSrc$BB{)
zm&+CB9P@l+xn@Kg&!~WFCPtZAs<moQLj+KrwLCsPl5^te>4|k+xm+$X_od1Zofa<^
z6T+_Sm78@pD!sKqbPew?{tj7+m4&5gwWm>DWU3BF34rUWl~r%>>&y;Pg_w<KCDtBQ
zhuH6{61jh!jA1sGW&hru4f{D5YTJnG=Y!9F9o#2L_s5qu<j;NmKGC^<pMCYvf9_S)
zbN~KdIG@kG-@Xd`-0$3@7kdK3=aR)eZN2}@wjO@kjOo|gXv?M|Z#YgPCib%VHry1~
z6=M&2k^2Xf1tYTsujg%=O33&53hzY{-daN#x(1O<n;X;dXtPbZclvR^NPR}-CORsT
zKyKSS(krMW#>lcP<ebG$pf(0;84o8O!bHAYiOZE%3lFCwe(=1UU$|T^#2WE~R4?c2
zg_t6#Wai^Zw82^-SLp|Z;5QM~X%qp`^K!;EPrPOx4i6$k>BLyds(el<VyEqgz?vc{
zCXR=L;^~H*5>FpKkmi}-2V8R;#+kS-q*@rpf!t(5yq2scdR3573QMiDGLmD#G<(~K
zo5f(cyj*nn)6r^|bm<0kJp&Ef%*bsc$h3OZpdtaF-{U^h($|Ky`#FVf0wx=vu&<10
zmNmI1wTf6}trDQ*S_P(Y5|K#+c5~V<^xlH87zf4FZ7su1$#}1xu|ZuiZ4X)AKVU|B
zM9%rmf<mg%MeuxoQ|*xILMm*@3Ja|ojC0)PfRr>)Dr;PHuvKPtrb#yHlnTxd<P<3e
zrZEuwK*@_Xg?k^(>gQ_&FKgcWfs_i%5=r?<;77yu*+BYBme+DC^0~Y9#Lfk@3UqFo
z6}9OmeOTul%@ok*v7|zdnc6ZX3mv~zgAB+@KfxGOY}ZMju-s&|xXa{kYfF+H=oO6o
zthF~4lzwmTE1#`Wzk{#t1IB%}|EDUE&mCV1K>InKeel_D{$HxRwy%BRJwFT8nb&iU
z`}f(8tK82|?$2#sFx*#_eGh=w(q(_{_^RL&LST)n_HaaZN<=qhZY6|3AGDKUyVo+}
z61XhGAn=%E@G>*eOv4$s`9fHWDyE7hcB80Mr?6b?-G%EjL6SM&ynt?5QqxUK?UYfo
zwrS$!<%N`_Ut^8M)W+lkr)h>F=3Lk73)R4zcaON>IbUCJ!842lVHO>!%T;<T;}~8A
zSFt9R>muGJLItg<(2{{HirC{Y%GCK4)||xr&Z&!~lh?UiuasIiJv^{nuUsz|x~xyM
z4%@Wnl`=U=Fb%_ivnnhqbN{NRC7Am4awV$;c?e!0Ihr^XDl^75A$OEiNJWR+YF9LF
zTX4y02c47jB4QK;31=-b8`DG)ygwYRiSJF#1RrFOH)i_2U@&+WXjqcU<9F>oarF$C
zDo}1vn+k*c+bLzZX=dDLO3{5iAJJOddQyFz_OBbGcz5f1UUmzVecoZC#%g7dl>un2
zI{g}b@QVfAq+5{wMr|3Z1d9*?H5Z~0+l%XU;V>Unx!{EqE8l+i4r{!sPo%r5KVwYO
zf#a-yw=Iy5qgZ+iqpf0mzm{^Xt#u-Ig9L-)QT&>oo}MJJ%Y|q}<_CS|Y8K{|F-EL2
zfbOf{lt__QAwATh+IsXks;WpvHyD)QX>Hh8sr%WPzGi}V6pW51+>*-9-r83X_W|Qe
z)x;h;zW=>F9JxRKRB-sx(Kqh<fYvLG{l?$Fzu)Nho4hfvug^W?*1z_rDzN?Uy&A&Z
zy?<+;QnTR}*!LCEJ!|k!9bXlE+&JKkFl<UGEbGEJ&s^3kfAe>LD--fH;m3jHveG60
zW!@!|bEHwIO<W(X(`jM{<1j;$ta?rBhL>C8aNde}mr|=;@EGNlb|vS}2~Vb!NI4Nk
zdB850Rq5LmV<k|=7+KeaM#YB_gJE(Kc#MJPrzfuQ%5}X`q!Yz_7>QXRF^|V04@7c8
z;hYXL#u|BD(=D$hQ}d}X%_Dgcb!9tw$OePkNCrLxQWYBrKX_6R&!lBtNjXX2byj6l
zRO74)JW_MQ5A!B}Z-Tv9=o-<r>hlcJzd|gwbx5(~Oma8TPwz$HJ4UIvj4_y?eGG+G
zTI-pV#hNDYl1{6)+Av-q&(V%s#n~5j7cMQO5YtMZ?HC4`brP7d>ELc61WjbB_~Fz9
zsb<D;VjKq{IjlFUg|?fTZ@~HG1wRfPkH>8wd7p*$!x&*(l@77>2fLIFt!Ygt$xUI+
zQVO++C)4H$B+0!B(@KbvFWT+jwZgIpL)2Txn>TOBYo@4_PA!#jm@vk(EUP|KsCxEo
zGYf+U8l^@}!bYO{4ap`gRirU`|Emn{cAuh>MP5gGb$%m`Yz(PgvZr7h);5)U5*WMI
zg2*RJ72;i2i<b3LYZDrO5%|=Gokp2y`n)?#F`EgY{HfbD?Uhd7{Q7mR>XFp%P2U*z
z+L5m}=1(_UxGA~sH{|}B{pa_`=b*07{oLo?>mF|F75L}QcMFKdKz*Iq-&ZQ1J3e<l
zpR2g`udydYd?|?SE47;{$e%yH4tA?zIOUXil3Kyr!0GW3;~j%@e0=#RYzIvjt-B2h
zIuD^*ayAOPG*T99-ndyZT%+<h=;E{j8MJa!JP7V)JlA2jUj6hN)>%g(1Rfq9XhwqN
zJkQz_LS^8SO5Yf~mp;*PIxr8DoZBMrTU*6A!?IjNZPyx#+DZ>Ykk4lT_361T3(Hm2
zr>pdcsF2Bo5w#0$sv2)-R4(fk>xHe=Y~|(hLd;3b;9}DGkw!Dp>v=q$Xhj%EDMwl^
z<eI2MAywIQ!f?}dE46IiEKP$?DGl7~tY^=>>VNkhnpKMNt)lF_{@uok%#1ka$vGm`
ztyo`!Xwls&)7xO~4SFkE-=b#7dELk)8_ene<H<GBiVzdlsA>;F-lbHb;R>}|U3jhb
z>(;tGT=&Ws(CMaa^RO2=liSfZblWq+Qbb_ZN$`}-0O*8)qCG8~V&bBbU6IXGB#0`d
zHJp*_XPw;VT0{-n`GC0O`pw_TSZH`+l+C%6L@7ywfuUr#eZHIAjH*}cG}4^Y!P2VA
z)6VZU9QiE!b<yE1`FCfXYLKaYQ*s*YgmT^ocbkg1_M!A<8hua2$Xu9pulp7KxNVd-
za!YqCmT`sK&u-M^?L1$rVOu`kjPHN9AAc&FzpsG)+~$A3k?(svpX<x)MMiJ-8Zs(!
ze~v$;Lc4$cF9j8N&1BstX#3#Q|L&jTUu4e5cqap6O`wx;2;`cXACCO+^pR;CSu{gG
zjwed4loIJ;l8wyaD3-v>D$JZUCY|RIiH=gU!X7<a5f2fgn@+8j6r*egrpva+jg2Hm
zR%vU34_+j4azl0J1}+L!^&pkAs1J9&t7)F(LC%flz>S{cVZeAJ8f$6AX*2$kF9d5C
zhk$9Srkpc*UC9bJc%yo+Mwo%c8=V~*G0kdE2xy`(jcqLJN;5D{6Saa>7{clx=lOi*
z+jsA`Y_OI>(`)9f(8PyfAdCY`PI}%EDUsTimD3Q-u|6?gZ#cL2PWwJE4U_hYBvaon
z!fhkoJ=~qhQA*sTXI{yhm~MNWHT#w-tK>pj6XNpOvp*w!0%NyLhvqP*V*N<7LRwkZ
zh1MW+0o!UereJHGUBzb2E<_8FquHw@x28~4mBzojgG9FHVF$}dH7BB)RYf9mPC|iJ
z=+IdiP^BqUmb0!4fuf~C7S~v31Ug-osMVmAo|mzrY9K8v?RzCc=m}26SV0<6UNQKM
z*=Y=f;P4<M9s>qPPU?7RJyrW*IVF;+UgVq@!oU#1D-A`1WZ<TOsk#<Yp>{Y-tH+uv
zB}pY$E26}%nswD^#_W`WT?<Al%YFYwZq4fruipb>6e{I)7T*7SPJJDa`6WHR)EDai
z{&Q8&=Qi*!RX+XiFTr1b>Sy)=;J)v&U!$)n<)5zE{p;^5uFnO(d&0*4oW7JC{FLMC
zf{$}JBM?Y#LVrD-9=Tl4OveMyPfz%1z@^MI4LB=h__AE7Ed!!yRmn6m<LFtVA$U(M
zY(b;dTS;8CvFUx-)Q)4y4DM6+y}hjXb|I}{Syob6@xgJq2v#a8%0*!y$K&Gz$Kz4p
zn_=YT`I)E7nX5u)hiSqE*T<Jc8fOUDYJ|?Z>Ku%<td|R>G^P+}D6G|lj!F|bp2Klw
z^b_kPl1rpFiCfOh%vdMlqq$~c5w&n?6H~3$ZN|lcv?j5v=r{j(IFM^481VDJdKRkf
zb-8X;sAq@sHa6wp+_KV>D<zjrWE31_UA6iQuY%2f0q&uW{%4*?vC~i!L13$5WxgkD
zNdOJ36Yzc$1GTTs<GS56MuN5v9$RH|?g7nfHP_UkY}=54!{M;$U-en9d+4CkiM_q$
z$DgdN&-_8N`(DM^ot|IsZ%T7Qn)KF*cxx$zX`0#BSFMUw_!i);K;9|GO)pK9K9UB%
ziuVQ(|DYIGy!ATwCf9r$1obk$uy4)}A<)$8C8k7EZNO3r1WDe4N>H^XGEB|tux>-|
zS%aqYb_(aFlC~s5q&Wos6BsVxg44=0t0z!Z64C8?x3+iIl*(%l6FKLw?R5ovUrBsv
z6W*@(-KN@q?sJgSpWQ2Wzw@Qd{7=32Csp|VymxDL|6cpRbRU53&*O8|&;9w{LwWma
zxy@YNG4nbk`%{julh-Mkm?J5zQXzziSRye;mX|BW!g9S*D4dsPe(|%P;e+RgA7t2Y
znuRcsuPcqrd_1s3p+6hj2yS4U2QZ>$5SNvb1S4)qg(0z`9$;k_sH+U`FZ6zcs68{Y
zmKYN;ojDzkOw*uE=S&Em)9Ik94TjP(YmO8ue$Z^NNk1e^1H%xAX~nL>#xlqTGtV=v
z$qdgh3_>($BD`9XPFGV%F)|+~8jZ{K!ufn*aFa3v`&ga9hk<xig$HL;4OuFek_&N-
zOyj^cXTqS^Ibuq+rYt&q^Pp5%F`#Nysqm`B6`Z5!+(pheBf0&8ifPrM#kLCqhHXyk
z)q^4%Y@cuG%G!Oy@AqV{mYq`K`T3b)7?`F}VFhG1gzmk9d^)9Z=OTm**9<sIsg)Fy
z(1wix3KqMW(*YRkME_2o4X*=}eGs!haD5ok*e3Lh9W1wpC3{f2^$u4Z3|1k{CNQNy
zPqzU)V^CQh87VMT<fKxJ0>4!yaad7n5beLj;Sf-D6_pWE$rxuhn|SGeG)lQ87xQ!w
zv$vFqDQ@)U?wncd&BU?8Z6z17&gJM`zy-G@3bl4<{q^S96moY?)?00&v|tpvEJ`#m
zCd>C~!G%5oA;AX=!Fb^<AgmW-ZWU(9MV>F~IbSYnBQZ+Fuj^x<t$z-7>N8B+*ZVq#
z+t%5CSF5-$vADhznEGq=@4gCRyCBlP-;<2{K=cU<ZA<#3I_zS+&Z?WJ&Xx7oxFt{h
zgl^C!$hGaqo^0Wr-}bk>tq)bX(f7f~dB`*S)5!+#wUtk49Ky&_3oRus=MM;r3@n8(
z22KwrPNxIc>oen=s4X(p!s+<H`TPSZWggxixSUr~dto@h`|sX^8~FLpe$Mr>(hB_3
zfBk2Hlx!s~iQp$5P7mZ%s3j3BEZ45ws{IDVBCNOATCm;<3{nia8jJ~~l9&&Vf~Utb
zxmK*HyqrIBIF1-oIKRA*(t`E!K!ibLWOB`53$Z5FxRBDq@o?bXx8D$g=X}1<41_jr
zax|q>sx1tUBJ>$^gfj8t$0vflN}!wua2}7N<VH1?W<AypjKR<lCrURsO3GYbp7GZ3
z<B#9-5C8CYIAb^-X8y<j@E^EbR=)k^FZfq~`*-~N|L(tIT^82oXNJ;v_vSs9^-85s
zfi*{_>41qH*B8f0S$u33&p8YGs?`EUvg+=|LPLW=-b5=m3i~Pqg>_vw)rq~5_&)o3
zco1(HYaP?!K-56ut<sFkmDUgu)Fw@GC!W3{ag!>oK(Il!$|`*>qjNO5WPMN?>mrrr
zJk4MnjRHmV<C@NojH4q*u|6<b{TO3Jzf6adp-Iv)jrhTdGIXp=(?R-cxnR6#>!mCX
zxy?dtg{i?*4YoAENl>~ZybHnxEEVU4K#^Xa8H49kp*0Z~ZEFEn8CoWlLhuf6<<DRP
z!@IW(;}K&AF&1}DX4LAfQivsUzFt^k<Z`{TURO+$nY4il&VVVj0KpW-cx9*+R|*tx
zMua?5%evDxel+s8v&bAxEdscgM9LMcV;&C*uV;)!sN?#+I7664*6DJ-&?*dLV6wsl
zq-Gq3nj)n|B~3W6jZ~D#AkfJz`P{+8-HO0~GHc-$V0E9{5NY|}SJmie*6y}H-c<^X
zZJ*b7LC<UVTAIMnZmQ5<g5x&ES*<D|sj%C<oK0&2_Nt^5t;w@Z-S)|fqc)MLYMsN-
zgs>sizPwLOa*n2P`(uofRP<t@suO%3g8b6)b=qz;EREw~qEYzx{3FMQiBeY%rvt~+
zjB|#!?@yesg*R^=`1#L%E_SxL^1I*to{}UZ9S<JsJkvOFS+6vk_{|^xk?;TTBhCbT
z7&yL}h=2I7kpaB3kSoPNsgX*eR%A#I)sb?7eW#pc!?9qi^l!X#1V3scTv<a^^vvdI
zW*mc5ZY7c8%9>XEXc;EYlCI<yp~@z8JRW3*1j#yEhV%6zT4*_8s>L;-NRQJ<7`+Iy
zI!|3h7A4!tVVd!ySd4eZ+LBhrVd8ofm`cPGJFVF;%_D=gTrMx<Quyn?`D;>K8OMPS
zPtQEQf5R{S>M!}9{y+Z{=jSJEfwvED_~qaH6~FoIZ}`K<?{R}?3X_sUtoA61vcCk%
z5?gSAPGwHZV(=6%!T#1K5g19>Xn_zcLl_03DVB9Tt9a#YVfN;Ce~6uv4R}paSydLC
zKvQ*>`AJV%7<MVC%#s7D)j?dLQj2h<2wJvUwYtm9Zp0c<(T<DseoD&f#iFd9Qb?)L
z$s&!qL3UJ`k8(KCH7i-=q6DIA&e4j&;$#D+XNGJUe53Jy{Qk`Ue#v|^X1BR%5}^8B
z+ni0Qm?hGl|0rUES`jpa;tt%X%Fe5<ou#!O>4Xm1;BHy{R;eNi*tIpWwk}nW2CZPM
z)t#v=SaViuY2Cnga1X8Vze$n*-TQYWgI1sD^)Iw6Y7Lb_Dy!%)8a14xa*bl|aGYmS
zT7?8Oc-B>_J{K&`%KO%aSV&h<2d@T<sKn|ARM_^jd*JSBYgy8R!d9c~do5@+_o{+;
z9gOae*3@3%j<m3Se>)X@9p&km_Gv%s{yp~anpYHJZzlq(*Jg}}{2F<NBzfv<!E8ZK
zqyl8-s+%O}72HHbVYhBNwAZV`-S;j0N$H>a<Ew(te4LpMqfj|r17BFvh125$fAhD0
zjkS&EmnWv<fiOGHmlyu=pMS;mvJ!lt7I9HL98P@u^IzzYq~%w?{SDv#>^uJQFaC;m
zZ@%IG`=9?S4-Y4%;l%mrLTaL@{Pg@pNtF;R&Nh|e7cAPEz_LL(y9Sulq1Yzs1gFE1
zc^Ft~A(e$`nE2-XTi(BWC(6}JW?jXG*jh)!P%2F05#uC)<Rp4(-UXHK7ue9saMG3~
zi%6<W#4mB>x?DLN4oZ_A2re)VBTHQ+cnlMVd7{?J={OOClUa@^E3>G`s&}$EACE`g
zym`wo&iwW_zvFOt;NSkkcYOHqBft6euW{CKdU)XH-~EiIrzgTN@ZHaU#w9O&yqq!K
zkV>W{acUJtT31=VDNx)PX;R|*U^l1L{vh<(s@~vTmPM4AUE8Xjy~|>!h`(alS+QkS
zE#TYqy%oxRD*_kxexOuj2ImJErlimDDC-8hkyo7cWD+%HaWyTX4TnB7-dNC~%xaAE
zOPr%4ExlJ}njioe(ZnmYi1LMZ%)^9hh8Sgi1aBn)YE7#0{-8F|7X5GE&HTT{QD>lb
zl0*t^dGBSmP6Kz&m7L@}a*b45a3UtnJd6Y%cyoB<aely55ex0a3F!pg7+%uXiL5y@
z9cKJ6aEXzZ%azM=A-9b6V4QB!wc(8?xB*)&-ULacROYFa!vF0Lf6#krX*g;%lxipi
zO0&qE2H=z(rT^DVV?N9R@faA!K}3I-jA@Q>oOwJx$UVx4sA#E5Zc1FE1T$;wMnt0a
zP1?SK;66EC#T>1U?&f;7zS7-m{G?vc-CJ%+z+GR3CW5c`=Xu{t-3R>7HTyr;-;zzK
zimPu`E1cFz5w~Gm+j^x0H_-6Y^KY%adTw7`i@S6BQ^(i&`M3<uQ^<@%;1_Ry#>eL$
zndT8cdD-L#agY4*;}fN1o}Zq{sc<?xP;2FSzF=G+gptcdrokWHyybX0N#>;wV54!p
zO7A4C3rkusEiw9$aUMyUDLL168e%I3kyEceP^`%3JA>fW-tzv<1EY6T=ZMDf_U#*f
z{_S_96ggk6EZ3x&tlWr_3apvacqACv08*;7oT#Z1oT02MzIm}Iuudk&HM5U7V|0`B
z#xwdsWPYj&H_~WT4CpTB%;UpbYL2v)`R==K`2LUIv#b&rnyK5*SL-+&-w?vUfBnZ_
zv95_+AYB&nd1bjSeE03Q{Q6h_MEKV6?whw51K<7ZJ5tLmzxpr4oVLYMim(ok$0OFt
z3`vR758E73Tf@16F<xvIn!r_6p&YHrX|)oQ*f=;3#@=-3_76o@2I!00DxPfJZH`DX
zya8HgXzg7^rEd~csl501`mm=Tq|dRvb}llj)C@y&TlQ`MTaBt9!{`IXh+LF624j8C
z54Vn7Fdk!}=mR>~0pl&Eh=-5Shgmih$v#nJqG#MSSBlHnUEtn7)IC`D3R8Tl1d=jT
zoD**wYYjsfn8QS{18K>mC5oI4l@gOcDb8cCTP7UOXD-VHTMLJ8zF|6@@XiyHP~gXT
zpwIN>oHxi0BSFbI>5p|NP%87T`zITJYmV9^AXQUnwjideDo`9GS(xTQtmnN*b}h?_
zZIN)8csxBYg^?OFxTeCaj<FQR7_pa&AQrlwy_KptR~eiZSFziM*p1Kn@!6ix4nDhW
zfO7x-9D8j0-9NLRz77zE%?RlW$9*9BTwm+c*VTTEu@tQedxc|+Q$szd&1@@Ns`Z<G
z=KX7ZDLCGL?av)w6@1Ru7cS|{IC{clXx8&@{@p)tSzcJy3*$8M`0fqPTgGXo)XaQ%
z5TmMPg$mc}%5_;;;}2Xe3)HF;@9^=%M}GIaKZ>MG%~0X+a1gP{QiW>j8`c}rHOq_u
zhU>D>Mqw}6T*RxRm&vtmGY^Kw=q+hk$m`1E@x=Sbw=CDl`TQaardk|<R%Ncn4hCD`
zdReJP;2Yb-MmpETHLc((lat1{l(1lQuB2P*TGWyt3FXyd#H1}rKWCmN&d+CJyi#)I
z<^04q@82`e1J7s3so<StcGbX}$2a`)um74b4E*kQe_$F$e*EyPQu&2%zWIi~{crvo
z#^AYLFZ}oa{lDkw<%u`%-ZCC1tT6=ZaKU1{W3^rERkHr``OGkeZT1FqA{`Za-4<G3
zq|z822GO(=4ZlviRo0sGI>G*$vD6#bwusG!aUw?~^IK@9Xc-1aa6(BI2uW1Kdv%r-
z+QCS3dz>aPjWw=91<noMETb1CgsPP{M&qWp+B5MC^O2kqVF))tQT1J}txEN2g&qt-
zJSnA-U8M$yIYX`@!irQ#jeUby<a1&a#egmt+Zmr7<L^O9ScgJ%frj^DQy>jtBl>CA
z!t>=BQia@TB|%vw5E=&|+yWhjhu~<d!N!GlJ>$b09*;-9|KSf@*Oeg{#?z4+D%VTh
z&fi-h{9p~HvPnSQpqQKbfHb02Z_cf>kf|nfiC3(Hw~vp~3oMEG^vIhxkF4uT2m^0V
z6Q+1tuB?}p<>i7mmaM)rJ_Jtlk#QJFrw5)te&GD_BC<+WEbBLD>yCqU?>Y*f&h~un
z_!5C(|NY+N|J)Y*NyYQ(IqLs@E$eg7IP2%SXPfTN=l;AJP@9#|W@OgP8KKs1h(d?p
z3>(1)ZlsdWeg6CR{Hd}!{rIZjb2uGv9`Qq=^23K8DP-P%^NyI|$LA+y{w$mx9*L=O
zK7ZidyLY_(=0sW|Z{ED)!;c?$c|P|s3V!=fe<a7u+qdt?Iq~M<t?G=)WPddkKYAL8
zR3ouODYI?BHp}I@fblBZ@me|A8Y_L%R;6#!DrsG0c8G!NvJhitU6W+{YHpQtAq*o}
z+0?=qu+CAgm6$Vj6h>5v1wVoDMlxF~J)tm&|BLs6Td(WNFbq^vxLy+8I|e_HTBaaM
z#E+*3ap$D*^8CzifBS3Zd1e{|OGyeTSekWw^UXK>)h~a^umAZstWheaFpRA6%;j=n
z9tVQAIPdt!U;QIa_2*o0{QT!X=f{sf^5c&`VjH3vrURfcu9|7HIl?wA(S?a)7=kd`
zWS}wY2HEEYq?!@tTmsXW#;|3$xBFA*`h8m!(*uhYoiR|!7Gp%<R89JNu&PCxp~cAk
z&PfJBMF6zon#I~cF<B~R8^|#;C|s0EB^U{E&RV8X6fS6uaTqZM7ByP4I-}ESr6S^=
z2C0CY6P8ge5pO-y;HmjcE+YTa47AkEyH&NrSe!8&=6O?m*sy#Zn(i_}oq;G~qVmi*
z=a>&8#%My*4Dq^>t`U!=rAkW+r@=B01Kv7<wH)V()A7VO4lFV9`#-A9dMgZo4}r(y
zkz67*XHs0js-i?Sc-se7GcpIds}NeFHp#53IL=@i<kn=P@rLOzFwGNdS;;x+v*LkC
zB%szsXx7bOav;Yh9#XMLUk8z5GF4vNc^JH7a325ohN3!+4Yn$|)EjL4Q;vHm>q}qj
zU*89XJ=x$($2~muxuCIszx@E`{`-5H^Zx6f?G^P_j3C!I>lmyP6^{<JNn`S}f%$XS
z=W_}SUvqq&ypER{4C_1(j}Hvvdw%%gkNn5~_#b(Eb7Cz?9}<T#j*lO{*DTBOpa19o
z!sF=?Z#^GBzR1wp;7Kv^^Iv=iEwL<D-oJZKtA^(n>F-R#%rFmJmM84sI7}14jf}ZK
zvz(tVI4^bv!!R(9GZu0FEIIS-H{X!rl^m})?}_Uwut1C~svY<IauNIQRWeRZ1P~$A
zz*;NUWx+U^?}=*?*^*quD`&_&y?n%4%j5ipz`(jjK*$0$7Y08N+yEGY6J5D=xj9$b
zV43ET5`<RViok`3Y2-M6%k$F{-+%uDKmXY;czHSV7eD_6Aq0N>;YW_gBhx%H&NDAB
zXS}sMJ$)p`$m#J&Qgwsrcp%rp;V_eP;^p}XF&=wing^BY7bCpKHxHC7v~6ema3^>d
z7{`&9mlv+r>$Xw$d70<uXN)mQo1WOT{%|r!(oo8#Y)-wWGYoPM*BB{HbOgN@f!L}#
zb%*(g(H<MkQc7mn9AX=XY36#q5Qd56dgb}UlT<mSv8F4*dnPB-`<TjL1HpP7g0&1T
z$ZSff<e0QiRYlZQNeZRt@N2FRteDbesltN^81JdhGP)p)u6Fan$)(G`7;J;V4On9~
zq1$Piu-0<DUSB=n{o48vcziq(*C>J7ipsacvR)RJ%L1veUKWni%y)0!@;86+E$>bz
zymL%pU>+yF`{o;lapdKC;h%o}&;0YRf6a%dkHm{4gl`@mSkuCiuNZ`gKohlaN+Zky
zv5T{%(fO*{>t6{w2L2Bz@!!5Yb4FDko=|0qk7_T`7{34KSI}#R9(&^5Y9|lqL%D6P
z${On%xt1EGR++KVij)l~{RzjH$`v`gtrc&9fA2m2r_=G(U}tC7RVBGtYjx9q{i657
zp#8JCyUJ@Hls;E+{khG4|2x8TG@K3x#xM}9BVAXV5r3;CMN+Bik7YL&;C(-AuZOtr
zLw&ZN_GgZ-BOFW<^Kk}5Ilurkv35ScEUane^zgvCCSIOjNHO93z;S+H9%uge`|qK_
z@o*%kLQaL?0$~8>hMW9{idec}@q?F(_ITXXt_LT|zUx&i9w<gcG=peSm0Ix5bDR&N
zaaJlr7@4M-oC??LLavp(ETYL(8^K#*m3aYY4ePZKtQ8?tKY*>Qd1XG#9Ookr$HyN(
zGED>iG!ilb8Cjz=;jA7;TCG%wOQc~KrvYaMa=emL<#;&at%v5+$x+mQmu2DQ`Gt9$
zFj#^g`S9T*$N9wDw{M9t5(W`?UDGNozmhPB+27^zLTZ&Ytt1LII7%)2{`bEVc^1XU
zr*YI~uOS4d^EpZgFm9`Q`o^%X;z}x7VX_JIASB|JA`m%_6J7A~hEd;Wqw=)N2HzA$
z3t=P}vpwn(tdOsVL9StI+l*1pLbWwW#q;p+fN9czw<-$iO`pZGc<UMbAP>F5ZWgJm
z@|tN@d}KuXtctC763sko4Gyt>7b-E^P~<Sk+)#=FRe0mn`WqoaAi}OOCvntlLen0G
zL1e4+`Zy<F>+0Zpc3?`4lEuU^r9{mQXFRRIa$QKvm4M+e27d9~d;VAd`d{&jcaO}&
zz~kw}!{Nw0P8!%nq}G{2E0LTQepoInxq=rV*eMJ$j9S4?YEe>*Qga2mtFgV>KsVB|
zh1nGTuVdi<ayl@Lf%(xh9N_8cNvf0Dc>netkB@Iyu2+)I>xdN!nAJfv(l;xW5=BaB
z3_+k$g^G>0+AACxrkN$K{OVV~BIbniUV^8x_8P*3RC$Q4*>4r%tfj-FcN%N83Sq+-
z-UI*Vk5AhMybqrJZ(+Rk`Lr*S2=;-??AuFUs9^T5|D_H8b3w^AXfb1lN^q6gd7-Ua
z#8p|GC1vf?*d)*Y%<KOz2|E4wN~*P7pfyG=7m?Y2dU@h-Jm3c}4w~bU^Yw+}bW#On
z%bU|9*SFuYysV62q~xqp7y>;V4kIzHIFmP#LuU<_Wu;bw_XB>A*^%S%#5Zr=^Sgig
zEz}DZ$9j!)n}(DUEhmQHIZP9m>qWAPVZyqBHLfJJ)6FGg5h$vff}b4Li*j>oh2!aw
z`8bhHCgzoD3-8_wBzS$fFiivF6qt?!2N%etvaSgfu3E1PGH@7V)&MKBFa%n$lx#57
zLA9)F!CPc-t7;}$A+DLUyikhtz-q1h#b5jd-h00P{zn?(n7OWJTrjv`pxVt7#{`FQ
z0tq@V)XYEp@elm+FaMJFZ{G6zKm49*9k#eFD9Es17Fm{hv&9WT#6H*c3N*%X*cM_4
zU0=`WkY6FCRljdFz}$v3yG)L?ULibH#+3#S#!9cmIq@m-PVi}?(4)AT3VfHd^hvyR
z9L9;kd9n{$*BV)cR)b7#A|pek)}1mz6UJX{Hz!@0xfvvzu7y_6#=0T+QO;Y1L|ajM
zcdpuy7syb0MPSDRLl8@hxGq>M!He(Gb-ivX2c2ke(=F>Fu#KFMO&P;-z2Z#ZGzY1~
zl6-Rrp1=F6U-EDM_HX%{zxWOluLNTl9C*`sI8J!yaMm#mj#|E<mWVNx^`C#ohmU8>
z5O4<kFwmSx2086hRnusu5mUrBkEtSRyX~?FF{aX}CSzP<nBmR4Bge;)l;yMAW>u$g
z5SHmxjVK3CNsSycJ~vn;(e>7HPKjVGBStmt8YM?cs-#kI!x87jcSr)1h?8QppJR*?
z)Y~fr&g{q@y<fA}G&9DCMMN$-C@$}PuYUb$kYFPy>^JFqRclG|%*J;69E|sqn032z
z_yoEG9>#(H?Z5lC{N;B)!?wnM{15*FfB3^6unYtz@MtT-Htp;DbH|_Rd40|CRl(=^
z!!s|>FRXC|)8$HJAq6{Ny_EsC$8T7#E6*RFIlo+pmzA0e$ML}P(+e?1m4h){+J)sb
z@$lGHhA$$0L&FaN?;I&bbd-Q4F65Z_;lq#E=2=rBu8EexS+N7iDdDU%oLdze>~TEG
zgWL=*42XCoOvjn#1Wx?fn|G8_`1P-TO&A<+-#(CQVoizTabl=})+)<#!Rl}=rjhbW
zDv_9y1TK~A$H<hhGmfkZ0>w3As+`XYHHqlu#}6;GTBTB}l~zR+Sj^mD90pR%oX-n?
z|M!2-zxg-+mhZp+o`3wu|H}72e8165561(+8@!!y#t_rW@$ks$;l%Ure@im~s!Y;f
zFE9MnU;QP&`qe*Dtz#G`yuXQvh7iOGSy^Y}I57-ilR$CaVr(zVb>75?vT~qL^@}qp
zbj$SDOrl<0bv?eEGtG#y!!Qi0gb=CLkr3G*XN6fMnsGXfUqoHp8OO`Zh0`?4u&A{R
zUdRcgDj7RQlpmBdA=P^itlDg?IujN|T|gA5o%9T?Wf%g>A^}(!LXOrfC09ltsNJ2j
zqVp7zxQ%YW8Aq_5@Ny+szpa(N!PHt=mSvN%>DNURrs4Vdh12Q8JRY%DX653#N<!BP
z4#z+I>wm@H{q<i$ypqdGvz|Zv;kW$HKYrlt!yDc_K9F<f`F!ELUWhsKaF}@a_`rYp
z-H+51X}-~lA=XH<hH;)S2Bvvp9A>W1SK8XJt@rHuA<spwiJVsn216ya3+GFuq)c!F
z)m2KWEa!!IURkdzhxv$Wp18=@XbxK{<1lg@M?%d^-s7!hKF%~6KR$h6S;cDOe7#_-
zXBbDuVH8*R-HaGYmG3ZKpN+crb=vnbHd=G7+uTcOvhMfLX#d$g0l*leN*1r-w0@ua
z`}Q8u=eF=K^@sl4u@6`VVdTC!p7>w?-QV%{bmHR=f8^~v^2fYVOU8{Otz}9{RCR^D
zHHvPp`BO+4KlS*k;Im#Ak^E4>N^4PRn33c3ApIa?8RvoD{r=a?!-PLMp?${0@p$6h
zo3~hF`S{}p43>GGDJ`+C3)Wf{dmJTe$rbMhK7M?GCfVTg%QNTa3#{;Pm@qV~@t9&5
zTp+LAPFpgUaTpkeiHc)QiB!5$K_g5fehi@59V-1_l|l9V{6rWXhvUTQ@xX_t4}AFe
z1GX97KD^<_?>|VtL7_5GtLxy1OJQvy3R#N^n5Kd)(lEt%g7*S9=19qn<iw7==xj)e
znU)Q8EL^TD%O&wQzx-R?Jig&S{ipxL?|%1Nsq74l<3LIl>kTf6oX*D&A9?@Xx4eJ<
zp5OfLH!{1^1UiT@GKPV7@89#o)6+(&EvgCtY(c0sG3axe2*_}wRG9+dgiY9W|M2eT
zfc8PBXZHOe>~b|XXwPla+|E^KQVqnII81(1F&@W>)-2vR{A3wI;Bvn3k_zK62w|im
z{fV3fM)N_Mm)**_H3?LxZL}&`7o9P>Uel&rU2>+@g7e}F)g4{E3ZqKV7R3HTtG_0v
zPiGDp(}b~S#l<u!v#kqu_RXtXTX&3IND-2Yrc|^FbU5Rg2E$UHRh<EjhX5q5muFll
zgmECanS5P{%Z0@|K0ZHjS+1PkzQG5N@e*89T&Y%23F|!Gd9qQf?PeGPqW<nJH6!ST
zR=2{akRjDvR*Y>N9wy$tKgslbN|acbh8bH7*L<Z$8R+#k5KJImBI|i2r;Muw-wcB>
zJkB$ZAuxEy7(74w#dnO;$RB_Fo`3$$KlA%P{wV#Yn1ncD3^_&y?}=UgSu0$LKc5=K
zeXftOt>aQQC)fMx;&YNX7_%`O_wZdmIs$Ne&)4+tFF{-PbnZWQNPz&S`M|+BE<b#r
zK0nh+<merzF_4YnT#_Wc4#DU^De9kAbgwGH&z$2=J-*URZe*AO^L(J84ag=i)$f1v
zJH>(x$KxzUU%_KMnA#YO$5vRb7nW-jm4j(wy(FHmPYg%9(QvyoO)Ht_mk%@=Yfd=p
zIGzqX%#SS3ks5`0R#U~{FkukN<~VMfpmUy5n#iO$i^Y-C!r^e>%{OmoMl`F|wDR=)
z#5^Au2XXcM_~9d`H#5Oz{`kXhxvp2f{q9>Fp7pvg`hhWwtnn(@G)~zh4sR{l7HXCu
z)84$UwUKgTo(Ed1#JCDWFs-5{P(*vJHJQ(7?LsY$F^v5DyPxsbzx-=nUS9a!?|#SA
z)AOdcWt`)@th7`pu`o{)-+%bPznc$y^X<1{dtrqkRBa;W#2>!@JwN~1cU<F&$mS$b
zQ=rfE1cDmj<&yO8n*(gpLC9&IxoQxUJxJz1j5V}!W6*4>*{}^1CZ$IkXGEXQ1)LLN
zg0+ruw6rW%=&?y)u30nFS}8dZg5Lt!>2MSkaUEo>Dc?M0nhfez)ecNCN^?4kk(~sP
z+r++&2^WSf_*kzjKJ8)IDw<~)X|uNi-Xh6{1|F<JC~BAo(qGV)xvFjqVs##)1b>W3
z$#hndT8KAf^K#a6Jf3J3mX`}<O&sTmLgC>s;##FF3vW(G;^o5kfBYlf3W?+}&V2j+
z8~*D*{UcA$&jjx{PDf4;5BzvZ1Zzk}0&**2BjSSu*qjv_%>~nt*`a1JRV}cAX5^YU
zXPA88aGZI3J9E8?A6C8=nv!rRO%*l_(?OKgm&+AO<up!wb2##59QoNd-|~z1-!KMI
zbaz(Zx!MZz&%fhu{_<zMTo(S{|Koq=SHJz9bTtGcP}5db0$Pl28-8ih^V>K7do|LJ
zJ$p@x{M)~U?AK`sYz#ZO4rZ%n?gP`SK9A{|ac#4r*uQr_J_(d`3dD9MuY%XDVv;{O
zC)_ylv+us8#)b3K%G-wr{_^L);5n|m{PBBQYm8xFZ54_p6o@p>?Of{JJN^F&cDvL?
z<r7w6Kn<2e=68SituVTDXfQ8l-oJSypRWQ>S;w-h<Ob^{vRpGmnD}@;1CAv(%q)vG
zcsQpQau$=X<|8fyYAn2c`-pQRgCB{DSTY~SnUV_YIbufx(LivlF)<yE)PfF}WiczW
z&`f3+Ebrew@cZxo$nkJsy+&-3tk^PQO=cK9&IPV&qwNi>&llc3yytvgc%0udn-iz;
z#K)KKH<wJ)EK82K8YH9kjq7D4XPHQ^CbQNjj)#LZsjDO#)5FNJtdwjBVUh(@8>oWz
zFMs)$T(4(-``cf0y<BJ*Vusk@IG)I*l86jWBy--(C(bEzdjB4Om^i;YL&-SPXcd0{
z{qOlV|N0+z`*Pv=<xHuoh@s{TMzo%W5D4QS>fQ~`>od*3!Oxh%(}HEKNtj=)$?)Om
zn5I$sZ$$_xDJ7ioT(4J*H(b{9mLOD9Hl^f;hY@it6-l4q1!l`NGk8N@g|${{!BeF2
zEmgoTYncv5TCH0TCDwwYF^q%ghowx3s}c$n|F5z}Mf6jgN7pFQJI7#|g5NOy>cvN8
z%`3^=HY)F>nhG`x3r$&m{amMUq7>=F)hyTF8OOtLB-KnX4ohXp1&b#GDJO=}<A;G1
zSDIBdcHP{qgO!MF3xE4pe@$IZxET41$G5yW%!FX64@ZJ?0{i(1XFcEj{4e<7!xQh{
zEwo%|DKlvn9BRY_sdkDr5C)17lgD|Gewr(V(=f0^(fkX+(sUE|-r-GSo=zMJK%-d*
z<A86C*b@1A#We`d(wq?`{G+HktmRC=@R#r2^Vi?JV=M(<FT|(J@$n7wFp!ojPv5_g
zYUbhb#B@0D%fI>=-_9fd_y5QLlkeWm{ObE>MpL;K>FNCcX#3Nc*|zk)5BsfQ&%3JL
z@yzGWQ+Knwscy0<veg_dYM^P!vLq{rtt3Ve$T0%Mk*o-T<G>J*FAk7zP6P;6U?4yc
z$FW3Fu_ap!MNy)tCPj8Pd%W{K_e}3x!yea^4{KMQ_g=OnUPLV5-ZQ;ZRqegjdY<S1
z|NXZdvsisXMCQn6?|JxWy6ts!*RCZRpAOslB}{5NLTvHIjSRMn^L4@Xu6-9invu2p
zKG#}s+h4z*NMZUBrq-^2t)zW2czfr_vW%gRX6i7>`mOR@jcb~*MU4*AJKP9$b%09;
z#;4~9ByKQdN@E?<ymLZp?NSN0HkRpmPN=Kx>-tF^pG<I&YSJ=i==&+A*-#b*$A<?y
zSNUqWV(i;^(lL?AlvNpJsZd%|m7b?h&S~2Yt#Zt0xpDUfa^~>DaO>6)){UH=olw_N
zTHE)Q8#iuIRSVvK|3jANPjQ_=%Y;0O$)-Vg`mx6mQ9M$qI4~W?-5cb*jqN=JN((m4
z8m%OKA7Q2f9dCRCASuh5**s@3DlW_dsyz0k{TN8=0!@Mno=oHf3wAoaj>b+PK8cc&
zEG>}2Op92YTUE!0EEWq+LTvje6}=`#TGBM9$cyMJ=?!&V@y6?~^W@P(-hb~sRu`8f
zN;B?GqXDH-f|8_3Kx@hA`8k`m<Hqe{7K=SL%^Ec^@`h#uK7u8Wj&E>r`6gMK#X+)8
z==v!4&GHOmE&b3l3<JhnRGN@h1xkvk1?5R(iVrT%=K}cAF$@DT2%L#0zAVcSD*9Jk
zu)B8MHat*DHB|+YaTxK=;*25H6UrsPRwD<{H5;n3rte1v@0rbJJCFFT((yi$GbR&i
zsuGH#U`y6E#-I|-V8&g0Q70NrTnr<ya>QD@YlUrz1Bzz>Wp27D*Q*suMCo`@#JR*a
zI1i$He>bpo0n?Az;7O8%a*9+K#}S8%hF3iWESfRCzSd#e7JytUg13}K#&h@YBKjWL
zkJw?PwVuUZO<tzF@ci=x2g{2U2YUy6>Wxp+tXqmaC(RSG$%~}r6S7oKp@R{PBLU+*
z#zdHGCS!81h6z6u+mTFAq(Go3BVKDbyJ8fP8ou6a7*+$m_Y_GM2?toDPVv%X!j$`@
z0;SU2zje%EmZ9xPnPuc@%KGYxkSGoh4zPWL9Y!)K*sn_t>XO4*#SgxIm!o?xaPq7F
z1k-nTmBEC`ir0li$%%jFes?=*>$!^W)w2`-j^-}dsnT|pP}nj9{M33CyU)4ae!e~<
zy572aR;B#zn1uK({yFWnX)GhQ0s$E#IEFFWJkq=%W?_L&HKT3#;PE47GDnP_)w)9!
zIiLFjKg5NxJb3(=L<rhVM^O}v4*I?&O*Mfi!jkIywfo{Gxc%$j>s;!SC7DhEQ^9%i
zG{sCamEZ&2&@-RUNfJ%lc-p=vQBiK2=6Q^Oc9ya#$;*sxH0V?ywW6Bk_}r40Gm4_-
z$<q(<9*z!g(6${zzv07&kI1tXZ5tPiM5)OL_)*BAn&gsF$%&#oPCxy#K5m?k7@j<S
z99u@ilB%5HUD(YZgpkogf|!(8f-Ebj${H1tNe(;4H>yf;ewvvWPd|*TyLRXG$%_&Z
zqEtF~N0QZX5IlH7A4o~?x?tS&q-lnfDH`>RtS}hVest!qmrHCvkWQw^QF%}bZ-xkd
z4IYg|`oOwf)2>&f_is@a1!Z2~Ohgb3qd_Z0)3ogERnvfeioI4-ODKtKF*lv2thMMY
z!D7)#Y{y6$kwZzMFxJtnn<=;~%9pFMM5x%p2{@z-=tQEenlf7?*2d&gCAWcg7BN9M
zt+fQ_Nt9yi*s#{Jn9ni8I4u_AjwiRhdAF^>>jPjZV{l)XHl4nYhQ=*XdmA#h-CVBw
zHm<Yjq?Eh(?XoNhA<E*1F-DEp(d=%*Eu*dP``sLC`!mLWD-cWxq0?=={Tv^r>v~$0
zf{(38Yc0btB7GVUV0l4Z&&aAd+E}a$5x(m@ZPRkFSFl_yDf65>%Xs*~0}`b;+}{Vu
z1p72pvzq;b1BT@l-Wyy5_(gu5wHOABkR&p}PinPz9N3l=&la9M&&l)Xx7sus7aUGR
zGJ9SXU>&Xt2rZ+kH$=OGDhg&Y<!FDOk^mi?ot}~>DT}&hZ$6_cN=zN!t5j>MvSe?u
zhx3kuqobV~G)+>>Gph<)ZkZs&mK+~#g<Ve!{fKDbW3!EIakKs0>+kEwH;wBUoa@8-
zkBbwwkL!`uA2)ICK6m>67KbBtin=XsM<P{hqvu%}0l0Xahh@`IEo$;yVV4`SqT)+m
z{t7RB{s;MMfBmoX<jE7tdX8}p*mko5TX44E+3WeceyHEi%IEmt7@cZPo;*e5lH;SB
zY}RXxaa>+5IX#W0(mc-*Dq>)ou8XouZwUmNu0dsz*<OKG8ME1((e<p(mRN6C?=L8;
zjDy7iLuXiDwG>%JT||O)o}`ftWus|o(`>M##Tids&B&4rYoga=9Qz0u3~qPo&x(w_
zy#*mS`k|Y$WHuCe6{D($fh3J7t#wtA=90yJ#Y^=|q&h(af%KAjx#sbM6P`Xf;fH?s
z3uI-^#l<C~9Z{Q>zPFgkQy_vu_-NpA#*t+Te8SbqlIF6VDBBJtH7O}BcwA@bI!o8I
z94z*D^~INImTSzWL;1kn!=vfKj6=<~kAv$l$$9&>MFz*^=_!7Afun;1o<4p=q{GEo
zSO|j4%PXpK!QI=pIXgd(i=T@F`_aY)dOPchn~<WYYjh?_QWe`uE@FBZJVl*PG0|~0
zkY_1o7_k;)0-z}B92+8`Kq!UO3Bkp<gwa|=5Yw<eDg<Sc5XlNDK`K<1ap%q*&Q4D8
zE-;(T7>0qq?}4~@1HWrAQA#0+>^+qv<T8RsSL^j|LkZi@BerWcZToBh4Li5I_nvjr
z&~zPJnQoG%)LEQwc<<@Q7{*p=6+_%q#4Lp=LVdKBX44?1nN^4m3$iQ&B`|iJs={qg
zU*H{B8<4`Gg2jcv+IKuXJ?D)VUL>?V>rI1^2ImbyjI6HK*v|05dmj*Nyk>6PxWS!!
zx7j<~Lng5@OR|*18#lNZ2V}QZa7B5yoWc&r!H_135DZ!s2%*qE!JC*i8W+we%`S@q
zyI!-|tWjFUD%5+%U`P^0CNjK$!AF&qNHlx1n&Ye@)rxtRGe10}$}?1ed0DZT)!e;(
z8+ZRcZQoJWC3#h%wPxKl+`WH~yZ7&rWI1~92=lDU@IDZj2=QS`$i23>Z{O$buUm!Y
zb>YYN69WA65<ssD5w_RJ_UrHFPCMA{Ha`9Lrf!QH;Kx?`5NDZMClKNW)b>4fHRE6V
z*`MK0{hL3}kA40N{0@3dEdT)k07*naRFhx&75>hD^S8M7@@xFakAH<9|Jgr`LGZ`^
z<e%W-qsRP}zw}r5+Sk9vX0v8Cui|?n1ZkS#yqV@f|NlQe8CcYLOg~bUHEr8+d2xvp
zlFfQenr1PMInhxfY6E3iOjHCJ7gT4W2dJzG(h=l{W59XCY*y1X4gEG##2Bu^hGx@J
z6eU$TLrTTa8>H(P+nC-dg`8aHf;7=+sc@s2NWpQ>PqUQFX87Qk&*tQL&U(FMKA)qL
z1ml>7y`Ifx6XSz}=j{BPFrD&~Tw_OrvyMz<1QS@VT9j0L;lUH0_l~A%5F*~lmzP)1
z<kot8rmPB{L@T5K?>s_Ir(EZw(QVyhjGH)RkwJAUExGg7J9N#4FSISr8iWWqV`CWE
zG(blRgj96HNH-4n;F*2vEnabgFWHgLzx#C8n>C{$_yI!0(K?W(IagOroXt54J2qBq
zM_KRi&Y_JU9YHQFQbk_d-@a9_DsmDOGgVIQGr`!8Xer2Zi5=ZUQ-}e6@joUXm!5v6
z_%V2kovs(_3`L$jyTPSoe}8{>vfgZ3l1UAAeWTe{D%+Us9Yr83u+HwN0oy<~;~dC1
zqk+IUj>L5;;_<hYNfaBzg+nV9=?6L@h@?_9Lyr&<Sz?AED&edF8S6GjK!h;)nLwy#
za$h0i*HlVjeB3~tkJnI=BnYqLeIG=eSNM1h*}mm_Z@tT+sL28>vOIDF0dH-LB$1Ni
z!y|+f4E@02Vvj^kGdL^g%)sa!mVgVf-%lcNLPBt8l^|S%g(6k738WgNMp?NNkZdW^
z>&=F?8`;k?sxre)G2TJPAKy3|&D0_6=DTK)=pto*Hb((PmQfcavwQck{lM6Eq&kw|
z$1%p#_ezjvDWmro>v;a97g-!1BNK%Wfh0|_lT9Pa-lyF(1U|m5Z#RRFdvf4ofqDKJ
z(DQnA{r%K)+mE|`ZC+P3KKq<_DENMI?T<gU5{I4WVEZ}KeBFB(hmoq9Gn>`?d;i}5
zoIn0={2Z^p{tDxA!~EzbpZ@$87zT?fYTkP95qDpDfgm)my!JZ(>0kXn@{51_Z}DFM
zB|+N0{1<;6gr5$_W~a*geh(qU?{DQZ_6F-Mbyc%ot!bMW5#X%n_~!ABL3K4;#hXM#
zMU)E`DM2L~sU?FQSZ!L$BBSja_V*7_lS0alj&&a)pLIP$ON$)?=bLltdQMf<*w$i#
z1J9&PGngUHSfq$qA2Q7#c!s`1DiKeU!gF|Z2*PuEdO|L9e2BaHa=B!0?;sYlDoTvC
zio@O{RrIE3wc14FOnZgt4P{!fx7g#S-h045xObcHUR+TW1zDD{%5uiRASV~Qa~?U-
z60$^rcTvTp)ea;XLJVc{8I!EnN3*C_Jh`~U*?}@msAdH+c-l?FIQCdGP?mM<OACRo
zOM+BftgdLPf?Kz4-~_z){<{=aML%@(0>)t=OH$6f=jLKYn`+*C>rK)ui!%%vftgY&
z+TK#{&&lhQ!Nuwj0srR175`)g&0ay0X0)AS7<y8flI97*Tbyy|L`Hf-a5x{M!|mv?
zR!|l-NutmyB~29LFrsy=5}XC+Y{cXc7wNjH0JwGQHcy^B<?`ZcSM9*F#LMea@nMLD
z*RJn&2LS-3o~e(dl#`0nFq$cvT5M_A+ZGsc!$(X>n-M>=B@!qEv${rl&u9(CSvE~W
znx(rz_m&eURh)5cPu7W!tw=My_q%qGjtC&>Jxw>nS=RI)-XlZca=GLm{gZ#fjoF+(
z^uu4m8pEb-$rDdIc5E)!oIO3~aK0bsY2K4&8J+8~-l6goWgX|sCG9vO^OP)07zqet
zqhVGn(ljGYqFGdF4L%Yvq}aBAA~1Dp8Bl5)fgV7|p}ixTLq(W2-zl!bj3a~WBk#}0
ze5p7)9&!CZsx@b)r>s{i=JOf-&~tiu#>v?!WnDp*lGHWJ)iSaHWi*)vwiI-LDCL&~
zock!Y2A);OAKwJ7S1tZyo67a5gYDP2{(0NSN5l))Hn8oXDrO^ov=#R+Qj&?uL*}-d
zm+%-DzfVz=I1jguZ}MmU?4M)*#t~jBIvcog_c=cM!<yA<MSZYOXFcOo1-fSB?!Ei`
z+kftV#DDwW{vF=`;2nx0#|OE6fp*U^*5bP#^!HbAFtlCl9i|zn(lJUqP4Q*9^J_>c
z+1uOaa<!iFQUplBY&N59R~#N5GL9o<QBqYiwps4$^@d;}*yu?Q0oKcipNV`a&&F*~
zLUVbsL^y?;oYyfZF4@ft^rY^w#sD$qSSn3fmDE+ulaoi$+i>j#%d(8V?<vY^_ma+L
za}bVovzjV5L7tc7(z6_wB#GwMty`$mOWu6$IsW0*71nz09v*R}5?0F<d0vpEIU!7H
ztUMzX(cQj1@hB;GuuP@Y6bU_I%t)F<^-!5cbK>3OBktTd<X~Ro)+>D9)3+VVAsP}l
z!-!2&I_oiYhRieE(LPqeugs&N*2t7zcx)<3g{0Pk=gN%g<{@7_dkjj_8xvcGD)#pK
zBoKt5P!c`Q$nwZK``nW#PaTGy5Io~FxOdK?)0EAoMaw{u$JvO|ioTDmxwh|#SPw;A
z*9aw8t}am$@-!m|g&6@nQV9fsqobp!y4$oI9vomsv)gR8)l9q>#57N`jN^FCyYZ~2
zew`v9rxx7x6X|xY1HZe4vi%y{In(ZI`IxAgqzOruO`DzvYth6mK=e#$Mr#;`X+||c
zM2B(Qwcd&{vhn(HnA*ru5g0erC{ecFk?Z)J1{-IMor{5hVzMZhF3wvI4))of?USY{
zC+BC7E3$cs37)2Jd2(^V`SL1OKf@3!go*SWVV{wNETkZY{Ro<-q3c>y8egX*Nk~gg
zQfRW!B1RK8OXqM-U`EgAA)S(ijCC0AK?}xyU~Q)LO-Po@B}Plk&?B734I`V?k|N8v
zar0(8NRIII<b(r%NO7>>?Ki*2yAPfqw8D5t5PDLi#ieo^iy?SM8rc52Jv?tcYuDRf
z*XKGqZnOW~cH4EO-1pm<u0M9-gYdEH>SMF0-~AAgFiQ-ajdF)6%^46X!8^e#ufBmw
zQj$Ew1jX#&h`nkLWWrwGGe0^+<O%4MzHJDj;q2jKid^%?8*lKzgZFUG?o7{HVZ`=X
z2r1cSgj^R?{9YcPX!9Wi>at`po5QqA3!;f?(=^xeM?7g#vRO3*ELpA@hZdy;QWPxe
zeMVzYQq#2^XJ_XmNro8=&V!N(N+!|hX9H=XkpcGi4$(?tM@z2L3GE+9RDza~N2L%1
ztz+t`k4m#NNlCJdgM)p_s+^t+aZ{QQD8}>@MM~F2=el!>VHg<3fve>blw=qTvvP(m
zC9A6qdwcs4qRO+9Jxx>ao;=OybV8A4lx2w^`k{pNk$`QZam{-d(=Q3J6=yBZ#>7-f
zKm?C)1M~TU&wb|8{Kyx6kUCFDrJ&3*yonLU)&)L%_=s<P_j~;I_uh=H9BV1-f=BPa
z&z%=uAkzttPEX^!qbLa&>@YCefy?s?s<PshmtNt6haZw=Io&wm@C@TfmS>bzh0GwU
zBkS($Y{fW^tX3;NG%J$inS?Y=6V`sk*!QGbV2mY6GP1O!sxyqSY?_wYY)(<s066PF
zdir6&TN7_^6@?Gh8lJoN9LER8oIX88An5!4BNfl~!rv|;TZPi~)8shaKmrko>01-0
z*tQO)3N%a}l-(R-y4j+vDau->udS7e5!oU7V%ok#+bHO$=d*Yy6EX3297o!&WwZvF
z6uX~45dB|5L=(4^(KKtl#ZNSa7{F#DOjO0Wm`W3@HFRbq2|!u$<m3rA_V%OEGlewI
zsq%`%YwDt+u4>Bp3?n_A897;<^E+>Tn@5)$jF1EwpRaS#^o$T_nZ$}Q0U;Gh8sF1?
z7zp9nsb0z`M2WtYnAy{|E!I1dw1|QlC%{JVXIYet!PB)Z);N}Z%lF=W7kv*7iyRXo
zSFcPnq}J548O~YGFE8lr2r_zVjB{vb`TFnt4(~rYVQu5<*!C7LCot}`{e%#wXxnQ<
zDYd(1uA8K{e7EbHO#GOwwD|ig)4rdWVEZxuyh=uh?=MK${{DA;{PZ=aotiBpK_I7y
z3kXP_d+r6AcBJeL<y<qifkbHbk8k3<BdrP=J1|;9)3lsFeu8OQyvz8)7rwx6f9;>K
zTwXHtT|^%_G0pa#9q<T}5Zw0{BK!*;pG@sG7*?a1^12}}O9T^8vFotYjclDKRe>_A
z*tA!1MiLA-hc|)C^GIBFHuew`nIXMEiJ12iyuuDX&IbbY){<u#n`J}aHRL*F8%I74
zCJGExBrqgO#n@gS2aDS%;|QIc)3s4#kfsUiW)l-A%NgTnsq31{<&r#$_?^vWL%>He
z!{Gr+YT9N)x7x%!5v^FQmN*wZN6tGE0&TmYs%r2vv`mm;a?G1Z2M(jah$%aFx=F7u
zvfi1=m}*F5#;wCWe)Nl<=hc^AWR@kQ5{gu_s0u`wPIWTjaQ}c8UU-R3KQac-H{bjo
ztFv=fn+;w`>Y^gJ6M~lXgJCe1Qm2GrpjociELPmRdzTL$d=Ta8&kk3yRY1TQ4?`a}
zP8+42TIiS^qnN5q5!GoyP**i>LJxsL>)89x%Zx*@hm;A%c2kdf#@@l4p=qPUcXZPr
z+aaW)EK8KqY&IK?4v%8+osv6$#&*Hjw%R_v0EB0EO>A=z+eK1Fj}qI(VcKA#m0{A@
zUBB6$y>qdW$<jEt^IH-?^a2F}qkA@c5=5S$o7!|^qH)*qjeM{%DRw%AC#jxVbx}HB
z<r%JpuJ2J=A+@9`3;fWdoWplLNuChmdVsZ#Oh`nc<2`1<I7e?rvLYwV6i-%Xyz}7)
zJUl&NlqsW^48}~eB{j|dV%4Rm1K~KDD51B{)Lzz&)6ba@L*(pk(N5EpTq_{KbkV?=
z>WoBddTTJwAyvZu;f&lt+ZmpoT`;?I6Cp#4rDx*TdC$$`qo`)fQe>hq&Z4uF(ORCK
zo-jI#aT3cjmRp#rgWbHu$A8&^T;Un9`Qwz1?PJ%DyM8@gdtCqc->cks`w&7rGpK&-
zn*F|CD{P<5G&|h>NeHCW_~3C4?mhQBn@x+nSdq^P(oEnQ4?>|bjS~>0;N;|l2k*X1
zce&)CnlZ@0XFl^8Zr!@Y)zt+_8msL{a@lUeYPb0~?{|mre^IOO`s0%c4s1YbMcZ_7
z@Q5PKGn7n_(!_XM@4>^+$9Y1q9y^X$Gm>PQb>EOCDUvk$g{&h<1R?>PN9$}V=^c8?
zR2hw-%nFPh=$e+QoYA%soe`Xys+)u?&&YI2aPidXl|YE-J2*HxqN+;1{oQYKadAPO
z$5?=>u6FRoVTc%@#k{6%H#Ao<YN4!3`reQw@i|AliJV$wW||WW%&NFJnsI=lB-JV^
zwkAA3ZtBsxXUG&GqM1;7iIh<lXq}<$8X$0VQ1jZ0_qn-Wqx^vI14WjTNicm!mJ|_^
zYAr%)7PC1&^!d+ozPzIATE6?idl=_<|Lu3U@!Wmhe*Zyqst1X+o^G^AXBh)rUR?0<
ztFLf;bi|Xh(+QMmQBzCHwt=e7abuJ<msv@fOC*L|M~ZX|j`zqAVToyy(YdIE(mLHy
zi}SpoEFH#JnoWl>9wBUuY*<uip%{C^dbL4G$!u2h<jE7x<pt~YntONeQRLNhk_|fs
z+LqXSeM8#b+_6#@QEt6$tw=e6L;XiL7a=Akp1w98QXgr<$!V@6q>PSsAsD9CpV*$p
zwMHsMKh5=oh~MA(nA$64vXg~d7kx=vVl^Vl-y;JrAyE>gK*;EkL4Z;qGfkmW`e5<W
zvwyVDYPG_S4L&5e#G`XfndTH}NmZ60C0FByrx$0OudY~)9Zq^|DDXl?K}3j2zTSDX
zhI(EjgrsS^nD*+o7$!k5@ii)n0+$F>mXh0mYdex8kFy@*QE0q1kxb%YG?vjjl01dU
z%jIR@>E$Jd^O_rrGMbsCM9Gv?MFV$TWEAs=^a?|VOcbp(Z2As|jxpRmJeyT;Z6OcO
z*Z^9`@xjNHY}=2&E`8p$Wxe11dHY`fOV`eS$z~M7^=8(;?2ApGKop~hz%x=eu@x4G
z39_@r9;0(yTrA1!no?`RV9Aw^8BpWMs#)>YTW|94z4viV%k0IMDe^c{`1Geg%{RXB
zHS$6;4E?S}$hAuF`cwP8ZanY_H=jJu@xf0$=xE|=+J>~q<B3h_n9P`@C@(SnK&lel
z7~_kxG^6W!X8D|9Gz2TqWQ=|E!)UE3s|qs)n#~$xMyjep3C(8HP!uUeQ8El8MN!A9
zYXU0*q)AGd7Z@W6+Y5sT=q!qYp&wXYEvJnphP%yXGgg}xXB}PFB4v~m_I*3ySqxo2
zaB_A@q7={FyH5x)J$1ESp=ET<Z<_Ul6#`>m=-Ws*4$;Y;Dn*b1YhtTJN`+2h3n2Pj
zA|P`ZdxoLIn1L)ynAIgkuCQZEFd5j$c+*uyqKYVeM5B`&AtGq?-tF5MJM!v_FVUEh
zcb+`P5b!Q=aBx6lESJkQetONMcLe8XHx1`!XH;c{_fPR+nspEtSS1xnk}{hYv`LFG
zkzE(Opv=pNxG!szlGsTmY(~R2G)_p#XboXIgOQRv%Tc<ftn#R!yjoG`6<S1}Mv^3?
zI^+2Gm=E6j0GxVuElwxZtq1Aj3&QnQ$5t#Ll|u7uuzcOi^YJ#v7J9lp&0hBoUGMcf
z?=jQ&-&R1^hs#-(?b?C6K{FAp^YQOzC(|uz%g)<+p)9`Z5F&E?q!eVS#s`a%@pOOV
z)-iRGP*fSqiyr4Jt9GEV4Oiic6iulzs<I-lE5;;nw!EZs2Aw60HWKT7P$(TY=Iwzg
zlDs3N)j5aI>25LiAyyIF>d<?K3~IMXB|5=43n<ERhGk^ET4B7Us4Ap)40hzQU89X>
z78KTa&X-s0)hSvET<}pjm_(dYQCB2+Hf1J7)R)dPK+^Xy$|cJR%-}IW;3msO*tWAp
zoc~-$>TG8#*L_~suaS?VcRv2OKKuKAi}1fl!4_xJAN}|13b~JdBX(vR5h7bsh{$5|
zQ{bYIGA0o#jn*kwn>F^q1F9rpk(Z=<`vh;fTwn3<=_8&#eS&v}cC!HjS}KaX;H8&d
z#F&vJjh;s-<c?#x70~!C`r)7Z+`j+g6Ky_SKTPVUICLD1AxTpN(eN0gcs2l*5dau)
zQ$|e5Fbvqy(zQKVUZPc$Zr1Y}A4Z0ur|$+pQ<f!L1Di}07ndu1Fl1SZRw3feeL!mk
zeV7zCDoSJBbYioy9j0}H7cuC}L#*EVmLy9+3fgu<mPN;Y)X0Y&(b@Z0mTDDadqWVk
zU60Z!P1{qK1pw>S1~VyiQk_KK&Qu9HAFC2Uq})1hao*yCnq~`j;@N3z=O*5piKksH
za5f4Le6Uz=sfroKIzkx8^B8??jEVCA7dY5ouxj?XY&IMo9`NRe4-iVSS*@_f;*CR3
z?S)|+ND%a9M5ZYxr)T6vfs&d{vx+DB;K|AoB_*rnnmkV;9QJBO*NtRV#jIY?_3ebi
z5p22!=PV`|P}4%)jrdK+F!^zcGLJbaiKeJiMlTqz28L17nvVU&VKkqrl&aoC$)}9t
zh#k%Db&ZFGbhltYJc(b|eeNonsWq@w!fhq9DoJ96>Y|+dpO@r#eI~Hon1CJm;JSxn
z+m;9+npdS%le9iwV^Tij-ML^1_;_#?k(Qu{n_QA;3UaJ940a@_K%OSkrki5=2;?2d
zk$W${it-lg#u%@mG%_fh9kI?JoMq4+F$h8`@lrDyi%L>vdj~K$oC-8uoOMQ}T0*LH
zjPr2Y8JI+>h>}81yuO%mK-;w#9njMOZ2{OZias=wAY)}8M4as!3*BJo`W}fO715z@
z2gAj3!|mfE$~<CTQWY!AqN)H%+xBe1zRr8HBx8T?fUxf9yu^Ajsov5!<DI1QaZ?ev
zNieb%oLooid|airoe51B!Zl^z$0}>_(Z7e^JqP;!0s?;5q2~L4(Dq`yE)|c6n;;^}
z%7xf~D9Z{xb%lnuMTo%ujT>ZnHc>1bO|v05PnM*_Brs^&j<PDIK)zV%BuR$%E{e!v
ze~+n%oDPzoY~z7Xnq-<4WJLx*(=;S$jt{V2H#ir_^NhNv@y=09gV4I3(=-k1<%TrP
zNwR`&7(gm+-@T0xk`F$3pZRQt#A8Oo!Tt=TG?$l`C@q=QHJb+NI-@KK+BQ}yLq8IN
zCP|{}I?*Y?gBt^KLfmf?1)Hwta<iuQhBQwiBQ8q_%Ar$muA@j5c)`%Lm=Q!s<J?0;
z|BUr;vD%=r66ZV68SUs9t)po=MxEe27&AmHij14rt}G@RID;Jt!IKvyc~xVaW7Rb1
z(bD&EJ|YMNH&SGZvM5nWj`fnK=S!ZRUXmmURZ%jV&7(JG!W|jwxq0^<dp8crE>0)R
zJhaUQgJ51(Y)($ewWc2p2uX({ElSRthU{v^y=u;DFTc)r-}-KBBdMvsPe4ef1XC4L
zTrpE6X~@!ut!bJT8%KPM-r-cdxSjQ+Dzf6nv8P{0X5MVCLM9oDql~@1=Q(@0rfnN8
zE*cV5@bdGoGxjNV%rIj#uFmQiAz0eJi3bK5$qXTcNrCf>WUWNZhjK*Yvl0_VWm*V~
zF?cCaAxz-PXQx+nO}Jo<!A*jND9eryIsNCIBav}_5I;X2!lIJS$KW>~<@9(F<JF~#
zyunGXEVYPPI!a9Rh)Ax;QjJaoK{>4T<T}IlfzV2VlstFuMgI67{(1g~|L$*t5A5yN
ztS>LoZYzOypaNN05Tr$yl59R>u{q$KCr?Qib2jTU(xPN<Z;#EoVYytz#l=FAmnh=F
zv#6`cQydI|DcsHvcwqFFtSAY=F!Vjb2c%QvWaL?j5SEqMV7voT#<pi147th)#!h)k
z60bC;8_QR}^#QNGaGRS4HG?&bUB}p46!BDF);T1BYF2W5aEJfZzyB9Wsu_RxAN(r6
z{OiBL##}JEIGa>xyfd-#H6EFYNyWOYROEDum%9pP9Jk_?gb>`W0@}`Zu5UWmO{v#g
zwbvsvt}DW}3e7Bw#&{LE*^z#~ZU2VJi#C1k&Z`s$JyRvJ{WnSQapog};I&s@=FY7f
z2yZA8&8#T7bL%FFQgrRW*~Jwfe(;FRW<x4-7K<C?>5PkZpf(mWjud&pY&K)PT%pq_
zwcl<+(eEY*ayzg3{?vwlC7X{>5wqU+T^!g-i57wo;&6VoT24Nf*qe3UadCOcX1&4t
zKv8J2teEB;mb3HoI3uwmXP2i)AxN^6reDX3Mmd^pL)Wk4sZheE-AuVTmeE>@G$qM2
zZj3v&LejMzWuBsvl%d_wn~oqHn|8&~%|jfPk}PtptfeS(GL^-s<RON*g(+S@rwKZl
zcvsHheFDMbgJm>b+*rJ)p3PacU3{IS#Mnp)Kzf{w#zH9tazc_scy;6yi^&}yCR3`E
z3MB$n5pgIX2%bDWrD+<>Fd~TlrW?n{Joj8m*S0)<7%w7~BsAl|Y;TW)gMD6k`DGqF
zdBU(-#;T|5Vl=3Xj_Q6eF=|08EP~~F!|^zB`}Qp^t}b|bdV<UnT=0{x#YZ!(oJazK
zX>Q_>J{oL|wagcDx~@Y8iIJ9EC(Oz@Zj2nTyq?jI8kxk57?s2rb8NiX%A%%O4m|hr
zYaAcn;mz;7N#C0|S3*$FYL-{$B#9<T(~m$Qy^AVIL1cKvVQySd95GZZcu$_^3`1-o
zCrV+)afi0qGQPHLG??&7*ZdjVHrsY)(6${~%LptDVQQ&@lpk&9ZG9aviXqO)gqoCg
zlW9;0i4=g3@@1(6LMAkQLn;+22nYfx6K);d;7|VCpXA=Hdwl-0pX0%UHwn&=XF0>N
zBTX{WTp)vwRfCd9t;mXkKl;ahj;G6p2Tz}}w>V(ASrbyr!QLLiL)Z2g;ZQ=5sCc-t
zCI;w<sV&=^4(mrC;Jm^57^k2UMX-jxZ7GW!t>U~$NJW}zx~`A;T2eFiBPv8wZI&kS
z;Gh&2n>7Qs=<Gm}N+gkFuZ4;pEuE0p35)$bvb-cM=5)sKzy6DVnR~zRi~OJe^ZyHr
zdV$gj-g?|H#K1IX88BEMDIViE?!d@fqv=Q6XV0>1q~FG4Jd;#ke}6wNRJbl+*ydDi
zrQO>Z<<RxWhyZ!E-d?LDes{|*Obh<hgPST5IaQ!aX<mHcB@PdwcQH*9vNU#NjIp$R
zOVc(0q)7}U)G`wCoAsLIdc(TuDXUpD6$^!sagO9ZW;7MkUie8LpJ?;ZNi@6leMgpO
zabYAv2&+7&X&N@o8bnQ*Pt3Qj!x&4R=eyYDsw%m-yyWEcgd)$WW(AvmiPRFE2Mis<
z2tIf$h9WO%*B#y1;eB8nqYT%DSd}QHzz140pplSBIy*A<9mWpK!U*2tMaRuMHyO<^
zl_PL<afOfxAxOI3kcb4SBwB)#AVoj~hqnVtE41=BZxD=_(IG>e+uXQ$gVPTm?z|zp
zQ?XPrFDS$fNGAy@O)zFaCux*Lnn*x4uHT*3g%s4)j4~@}+a9k3RZ+3Hb(hV0&DF&P
zZPQ^*ME9pj&b_<)+`e;%B+nR4l$hQ=I^tk9=V}u}{OWqftSo6pgJd%ISwlMU>YR6M
zHfy}U!Ht_YcyjuLG|?DuG0qVYFm9UVL>TI9KBK5o*6THAXBTKK$;+I2UNd-w%0MF7
zELXVEA-y0=QtEm}k}FiE7{;E}rp5O?$`$eSzPHC4uf4|DS)M$8g0&;Bz4|g)8eQ$1
zO$;7Pvuvkhx;|S8#0@_9=>BHfB=t16sj7+)0#{d8Q=50XnLf7gTwesPlZm(g_N>p(
z6p0{rH=_?ehTv`IJ8WAkae<I(Qr9t^iv9F^HA$MarpQbBu4S+T*+j#@d;H+|%p0HK
zkNnKP#t(k(bF42<dG79g-g)zzq=}kB=!#tkS_l&Bp{{ClTCi9g@S}h5%Y5(shdlgS
zzd&#>uV!(sGmex+K_U~{wu{N2#Dm#38pMZ4t8lKzi|zFg^M2A?ldOB(x<#s}>>LI|
z+l=&si#b*<`l6JI)nHlV3{4ZOZRc?!&<q{D>5*B%y4z6dl%a3QQo$mxC>J$Ztx!cu
zQk2a1_UW&d)b*S<KK&Vvj&G1xvp9_RfU^i6J0hfl%he?(CnrFhBere3o9kVF@8ch*
zzmsz3^SS=G-dejZVA(=twT?6K>n+=@@-fLWgnJgraJ_xEUE4p>)|%E6F;Pz4q`wqG
z;DsPbGhTb+4T`G9TE}cYCrJ}lt0h`%uC6Y*y1JyQO42mp?Cgv*O<61!41Gs88U`OA
z$tcQMtW<@Y0JmX+t47b%q)qw68_*}(e9Af_%TqReLS7`~Sp*CQX=$2<pbS-+68uO%
zwh{TF0>v!m#v4Ck8Cf<XgK5dD*h(m73F)GuZCgec+bDTekR%0t7ah=I04ppuI4lmW
zQ;NJqt5~`Yex%4s(jvzU9Ya4r77&SKJzk-N!1$K4iw_Y>QqN{s8yL($-wzlcG2=z1
zNrQ*r(Nf~9je-CzNL0X%T}=KH@y!szBSfH{)dWQhvlBX!3#^Idz7`C9%=9UXVtSVp
zN~cJral?rBHbPoW!e~c?x6I3$Jk3!EiY#YuvCsZupE|F(bK?%yTJGMtkGGaYBwSrx
z@!s3-@ye?&qY+4zaPP(~KK<%zJUKbTD$oJOHuA`nj2pMkGyzbFih*oH$J5g%%x1G`
zp3uiEBniPgMmLa`d8}~T0pmwjO~+^r^Szp?syIJCXFi{i7Me6s^nK52wPNf%iHxV+
z!C3YV_sEpMJB4s5cChqKiyZ<lzwjL11s;C*Aj-AJfvPOId1D`I4E@j(0AoEuX>^*=
zH7(ACSV2s&+djlLfp9MN!}T+G=>Gox?&P~oR)sjnxo%wj_+oT@F0h>&gh^t(t(>+e
zUgsR)TJ^J){>BX`oxWalYi#FIsZ0sZGZ=#%HxyY$QKf*Ts%jJ^Km8M5;m5!H<5YP`
zA_aOrN7$C<?>@(e58g$~6ce_-Acam8#zE7L>{t7|{`wC92|w|Newx7u-hA&JK6v;6
zPoF-SREY^$l9KP`1TS#b?dCpy>J0ecqUR{CMWmTVs<;VnNfJ_O^1Q^1f<#M%h#QeL
z5mlrGF(=7JWuQtFNt#hs1us7T0~{XJ)bo<ob&O_2NQFogNs*FN8A+i~S%TLIqZibR
zeWXqaLee%pz47t1?gN@IN$f#~G*-Qnfie<Bo{==R@AI<_<(lB&nlI=3(W|!`jB}1W
z&v#!lP2>0M`+iqx8DmJ3bm}e${NxDW+I8YXtaRe6(nZt}&jQ~j>)9m!P*hdTr#|%=
zjI}iFfb*WBh(`LWtCkP~i^UABBVuVbtJvS)XR%oDo$q{$%kxWKe&sdFdXHPry}<eJ
zd@U-7w<*2C@95ssh5b=$!tdeni8h~pSkv?d=Oe!=Ni|uXkY<i{I44Uahes7%+p)Yl
z$D1idQEPhB#nzQI<ath7L#h>0!roEN5FAc*tX3U56%_M?JTFPBgl5x|RcXwZaB&Eh
z=p_0+gr@I$<l0f@n#CeV_>?D)8-~6o&oi9txLU30I#2H#?mc%Kp`*!DB!YhINyt$;
zB@lb2iB2MX&>9eGGN*d19nngY<pseR>}ZhEu(v-a%`|-%u-@QZtVm)&Ub5r2`RPrv
zA&|t*N$hhoC3BjDFe;P}Eb0ZbYDVTPrzdB8>zm(U=sM=JIiLE}r`TWYQRNkBnsEB`
zj3<wu^4`1e^8Pz-bNlXHUV8O4s`-MK?%n6t(~N#FWI{5w9mQe~D*{1^C}j<?CE@T4
zrswqhlsr$lb?XN2eDDrMS&(F!AQH+lqpZrvdm;{fb)v}fsEAQgGK@XzCU9r(2EiK?
zaImOpwV7fjG|mQA%MEIy@C5qaFf@kH!|~BAZr!}e#l;0rpFF`i6HS#`aD3w!V`Aik
zRB@>9L*&zqgCX$~+s#{~(tyNyyPF3^+PP)5T5)i2z>OO>xVX58Jh$obamCth%@H3(
zs_Y^mwrv&x&oWlxf*@s7fPJ)*3e&)Lq7pN?@8z_p2*qd|WtPYGuh5vb<A;CfbA071
zKhEo~eVV@Mv3-wBQ>wBCn_<g}KlnpG!mnG)`|rPxQaMbDu+f3*M^sw!@~f}${0lGR
zo#6hx7x-g;^ym5Xhwt%QU;7RI?%(};(QK{aYno-TN*Tu(?;c6xG1WD`_5vYd*hTP?
zzKe%7X9WaJmeiP0VvWEVLDvm&f%TG1YqW@gc|u5p(pcxYar1zm``JIrr(Su1M-Sd)
zb@mWdRfN$ZohD23XfhA+eoeE2(Fby=**`ocuV$Q|FF|Mo3WAM`w-E7Ka-M!OpwlEO
zURSH#f#c&di|zZpd+$D`ioE`~p3x(w*<zZe6FbuFOs&b()~%{4ZhX#T&9lhw>o}k7
zb$0#XLqLfD6FMkHMaXDVks75n2Zsl=UB_q~r>Cc^mgn(Wb&mam1B4Jfee#$*&w1hb
z=VIdV<H!8QSN|ywK77oNe)-3#_xAbh=fA)=zVY>_cvU*KT(`Z3Fr`30yZC?4!h=t=
z`P4H_n&%YdG1i$FS0Ek8j(hhHFlNNMHCZ8<lVFS_FOprLmJkYKqBkMW3-<Q*u*OjD
zX_VC@<pPm<1d5<6oBj$ZHRU|x?CO+K*Ub0!&{|N`=@ct0s9eeZjhKXJh6X~QI+oll
zkD1M8bp0B_g4uk|lamv?A26<AvA>TH@Z`g%I0y48N2?ek(02-JquCP`cZs5?@IJO1
zXJyIIb!=8kr0j6PM57&%P92;_O&$-WHA#};Y}|abR(KhcF%dpiP}1YW6lEJ?k9{$#
z*`M#R$P%hF<zWAie2S6S+uNhaa;mDLEJ|+Ox{X$vH(q^}#k^)dpK;^%9hRF0ck?zk
z4-R<n=rQ+SdXe+h5}&4QhJms!aZ-@yDP!LweKf-MLr1{z(#y~D-M8N)^9HGsY51M6
z*qcXrYto{Wh=l7Tp=lb5a>gsKy~ye5DN0G|VostHmdg!$i@-Qox^6&(xWF10NRkv8
z0$5Lx=IqZ8kU?;9e#U0A!h46-0;L5ykqD9S>T9o(K6u3ACnvN`OO~cF?`G<2hY-*^
z=9t72Uw>yTX`0fu@dSA5)-8^Xj&@$5>qf(mBUrYJ&c|Kw*B6EDW^{ca+J4;~rX_wf
z*7!-(Li|}pKoMyenNE>Ns-oiG`ZxbAzWiffqAV2$`+Gco_!O*RURHQBQ00eUGUm5#
zb8#m5(I5FS&QF&FFB!sscLt=Sn%8{x^IzagKl)`tP|Rlgj9##)4D({fPk#9;{HdS+
zll=Ox|2lu`7k+`KPft)ll4hthWf0?3u^42~SU35Qk_3TZ9H8k&s;Z>#Ecfoc$bb5O
z_}}x&i?8xqzx{Ro&fodF{NsQ8k5F2prJ?T!kT9#}F)G4zJpbHt{BQol|B&Od;`G6L
z)JcI$3Vc|T<vF$sBqBith42czA_Pf2KjiSveeS*R0!~C(_uV`9czSvQ9@0E|rOX%@
zr}&I*tj6Bn9&OuB%EV{y=f@RoLWmd}9>Pa9gpY4H+j-!vTeo=p_%YVT(L>WT+`oUH
z`Fzg1@4mac1_{%~B?VgQou}-&(r&k?hwZht{cR*T1TnP>gQKoXjt&oa|AP;C>+N^>
zrN968Sgp=@<&{@)&hqZN@1m8ap4BmTYHBO?eaG*7;~SiwU9sM*sp}hj@rQnhzxQ|l
z4y)zaBsx-&ydq*JWNIeI=g;rmuk?vFpR|+|Wr`4vBG)8I8s)WliZKH+fkLGysp<Q8
z16pU8&+`bU^A6`mnzp4Z6Lc~_Vi>)}3x`Z4QpY}~&INvSC<RrSb9iGfDs8HSG#4a!
zq!=g}NyOtYAmxbH14G}Vs-7-16xwridq&@EXopK?dnucyV{~hp9(ps9&m^`p1lgnV
zj369AIHUw2V@tt0Pe9_mBuNU2yk_hySw2JSHIJS=qAWAUF;b~@l1>3{lA<apvXZ_Z
z>H8j!BP|oGvoy_`G!Y0HaK0zc6~2cA!Tx;CqOQ1me88=v{mIW1!@Jt1K`X&(dBxsh
zPOyfX$A{Rx8F{Li&1aNZO4s(3X~xa{1L`d2cyFJJqYdw!JSA*JBbh`d5shy<%h0v-
zUBlUB&Fy=~+_|$yHLFSUoX!n66_sRSs&a%7y9GC%;4E35are$$>LN#kc&ba2l%NI1
zG*P5r<H?$!@WwMM_Ne3>jOX=NUgzxOjOE1@-bO>QHI{K4Xf_+tR8tfM_wL=J8AiMh
ztk-Ldi<E%2ZOO6>Ct`%~)|ZC-C@aet!)CML#*G_1c<^9HN4WmjE;8E%WsAhQp6|1*
zs?>DQQc1kEwyGv)qL)p^vL?1y2q+m(xqa87CMA*5ks3XA19x9|p8xG%{*QR+#TOa-
zh7bmZrek$=#olaAmS_^CAb9dDgW!1l_$lA``qx>^_i5W6=RD<nHnqtDNs)s}$@2>5
zAxkoZm)tzO!NuhhZr`}c|LouWpYZBSFY{mj*Z*(6`R#8LoMSdC2^-6(1WpMY3K1k(
zs@Xr<Ct0_cF-m<`SB9&Lj-ULqKTG}PKf=7Kc;SUl@tIHm0sj1-|MRp>Lzz|JrukZk
zEng*5ih7@){OLc+gWvfU8}$@f%vrm~%=0-aNNfsP7ASIX35txW*r%8uaCGNBLFJ5A
z@Do4v)7-dokN?|$`Csx6e)*R;KEA^X_wV7{$Tz<6buQP-T_k#zWgn?fwg-&uMih<9
z&)9*{!X$Rr{<g)IrYUt@<D8?aDi({ybO?*rTvb)P^2#gx%+LG`fAz2a)yPfk`-v8y
z2sVZ-1_60l#Pz}&bec?p7dPc@DT0tVAJcwA0AuXb8iybF^ry&E&CqUm^5_HF%@Q*X
zJow;U#?f%{^a<;AG;k}WrrBr|Qmj^MhJHW`2+r`*^Y{7mr(Wlmf9ZYlEYAMC4^!0>
zAjE^6kYbk!_OE>Nxqa&%HW>P@!P;?0t<8%JA<8MmGU8K=vy@qZPNVdCFb%VLjg*?K
zR8gnyEx`?p7Q9eMm2m6!4aU(jj+Q)6BPxQC`6A}+^y3<lD2ApZNfmjLMwx6Lu_ZEh
zWXdDc$qbc^SbBQX(T^SVe9m4`(==Ul=9hsRw`brqrgQjV&FFg2mMl}4c8GHYKQK6r
zRvH@|%`o5qm82jA%hd&0k#ncM%hM+(;3YasxV*gL&YgR_{_3Z&&htCp{3hpDrz~bO
z+Gax{B;(jID@$D8u{W#8ghLFLGE49_(jl}G<V8FQTO$eHkmV^|*Km6Bm^@1u#(|=U
zZI-@`;tVqkq)Kw*@Q|Z}J?4uUcl>Q0uP&+Un$^(J50Sp?tz{fKyfqX>h7f_1)2AFA
z9iX!WM8I3k$<+nU1&ke|N{Jj10`0mbNm86~Y{mwBK+HTi&!$-;6UA)4AkT9`kc1#<
z)&mENeY#BtUec@^-gxCT=H-HKKm88pXXj`g=S^GCW;b+X#U25us+yNxdWDDShqwTv
zwYbrc<^|(uNPIlCTVs$)W2}jZnNv$+yZJ~dnayUqOrUM+<9hY7-F&W}c&`taxAUbS
zLd<oV;u@^C1QFX4F1YEq8u9l+gouC&K4O}@MP&&;^@qN~k9_G%Jb(8tKk%tfb9H%%
zRuZMsc<_*7I(-|Q8K`dG;>n{+zWL4H;h+7p-{AE0gsQB`^OAa=P}dbnqIm!P5BPh3
z|L?Qz3_tg;{|Ry->AQ}3HA7-i9P;SB2P{52=SRNuBh2eL-+B8jzVWr+=6AmJEu1mT
ztBQxK#}sAFXyXDjj+4m4tEgVH7DUSJ+xKzBoMDK{$UArM@|CZAg@5!9{~?zb7cpco
zKw6|ESwYkF3_YY-g;F^}=49ENTQ{F)b$)^!Y>e{O1;_;FQ<8MS;qiSA?!G`$&KRBN
z{QQhG%lX{re}I4c&;D6{<(Ge%uI;$EJja*;WSFGXBT16bbuqnj8_m%)&tl%=Ovitu
z`q(y>zMm0vOR4Ysp3~FQ2w<M(KI1s@^yyQ+`qi)U!V53(&O7g5hJmbzenB5R##xdq
zW4^aw^Qa}ubNV4(mr_RrmkS_NN=ztH^g~Zo&1kz0tp$k$g{50xaq;vqSC?l<C0H#D
zLq8w|93AfO4o=3HXLC;i!t~Us<nHkSqv`mGFaHQ958vg<lShohfSpDN0=ZM{ZVy$z
zFEgJzcV5A`p3P=O*Ebl`)AtR|MvRU3OwM*g(+>nGNd*jJKZSy+SRFXSI1Z#~g3eNc
zb&P|@^euHgkJMkOP%2>@EyK{_eUx#I;~3v4;U+n(qaRz6RN)w-PeMy{o&bV0RaiSP
z*jRDsTvC(;MU~KX14zTo*$f#Jr;pERLr)@0y0Ik)Nm_$(Qya>RRC!Hj1O3>*1U>b^
z)At>{2q<NdT9WCEG%5Jp=YEhs^QZm{Km5Z#%Fqw|qhI?qe*2rh$+y1y4U$YS3=%tf
z=2ec_SdI?nv??%MHAvwYyN1)JJy{m<L1mg_%|NqRv%I=sd3DCDswwg)aSyH}&oY|L
zhRu4-U_4$ain8FfS6}7k&6}wCjBh@8kImSzZZ{YSqZufYgfvkkDj`oaQVY6nL!Kw}
z<A8C7b=x7FV0E#gZ5x`6VeAX$iv?v;#g^A}dQ6iP;Wb^;5S%2<OR}^?N=uR!7~|=i
zcrA2oN4FlB<uh*Iyvuv<zQ^)%NvieaOn20anyQL$$~fG2RI`eEdxs=R3PNJ6WqENK
zasP;Ci(8iOD)H^6vfVgh{?aA$`5Y-DO&OjMhqtHN>+#3i7TQN>2wD?lK&1)eI7a0o
z(*mZ|<kx#f1PCHTI=FR|Wri6B_V@Pq!$1Ag{5${lpX1KWn<x=5eaFyl$QLu(HomV)
z$|)8;F!VhSAAZQ!zV<b~`qi)EjAgNR$k0cHW>uHWr{QvP@`THir~LX?f1Sm_A%FZ&
z{3+H?&+#^Jw12?K*s|`f=$32pvfwkH|2)q>{~SN}NB#)^yZ_`rCe2cwK7B%6RSY%`
z|E)2s*Cr}8gG0zTN70u{uCA`Imm5R?7veqm(n~M$+rRl6w3{S$2~<?{4JHsmf=X&W
ze0av<_%^2L={IL&WsXh@++Z2|K$_HesgS9rJiNidofp}^b%(VHkwq5*=TA-|MI$Zo
z&T(~dfwMi%jXRRO5F!TYO&idQFTTiny^c9Y)2t<m0%Ch<yU~R0uOWP-P3N4WZCk1;
zV!JLbE_TA5Z42${>WW|am0#g6{Dr^3*T4RCe)BhfldI*DqN;b%A5uv!E-yJcK8lKA
zXQ!VRl-3BP=?5R@UP6=C^SI#$7}|yp-hGF5c}c&y<nrW0T;D>TF%EXO;YtxZ7h0=W
z#Z0KOSzY3~mNM}O)AHuGzK$On`esevZYJ||4CB)}MG1io_L->W_ha)()e&{2Iob!Q
z1X2cuq2c`Ul#7cqnr6*t8irv*suvjRX__9XK<k{o4<G_fYol@|M1sLMMnz8Bc3hq<
zna^j;=X0EoVPwU;ikpw~G2g>mhDp+Fl#h|rA;s8;uP@7rG|LfEVFrio9a1S0C1~0W
zooOk{9FZ8jXt7v=1C@2$ePKae%z5YSmeCmY4-b$!<7(9~4i2q6S~%<w)jYA^?3gWT
z2JdK_hS{tpPYMnXZ}9K`2mecc_={gci-d77{Fy)fr}^P8{{jBWfASY`LeUQfsRFCb
zIXU2cN0tdhE-_=r&_~U;w+eiLbe6VlIX{2OTi^R0`+NJ8MZq|ZeC~6f<HZ+VBnXf3
z@#ix(8h48#XTDg(7DHKZbmN$d2k(&<QEer)B#{DdV+ACF<@9V|Q>NUwaY)mwXu1Yt
zVbdCf&{TO%qBg89R;<%F%g{<wCIwbHR?8K`dY~#wBq?1x(v7G1;7GHa^}59vL%Z&f
zQ`_Ya{n(dz{>}@0_1C^iyJ^VsgiNRCM6oxujxH|Fxp(g#LI_r?4W(1e=6l?`_Z-%E
zGL><8d5P2U{b;+My~Q5Jba2g?eqBtkjb5nhn&oo&5!KVS($QM){=T2c0@q`b1F<s7
z%ZxPBQ@CJMHeQ$D23!<)c*APFq{>S|aAaA+zy8PnD1ZLn`5&_=YrHp%Lyt3-+c%Fn
zzc}T?haUnlvLO<N9o^9KtH1WETwSfWd;dPx#_DQ+F$X+@8ED#u-V79V#jCHs%6i@M
zOTY9>+`jibw{P9!d*AvlZ@l(0nbcgIoU%MV1|Lwv$gHf8X~KX0*Z!~klVAN+{<Htn
zf5wN89@7jR)(do^@FFUpTxfA2Wz)3eg{E!SJb33jcnhn|CC(4L`_9{}mRBedP>Ci@
zO8RyLl@WsC@slgQ^X3!2@})2E%H9F}=8TKehqSjYG3`hqD|Awj7Bi0T-lAOOI2Blr
zfo`<aRl&_$x5(0xJTG|r?RUuYg1%dG^VSjj`?om1I0K?#da+or*=(rmnlFCwi@g8-
z`&?aJ?KYr~W0$rwn(NYPc-Dq<&T)8nNS0+he*Abh*V$G;>EuiM#b5kI`o8C-mtNxT
zy?Y!U-{7}?>$mvUx4yMgNoHxv-rn96Nz?9HbV3(gZ7g+JQKc!1g9VS@eTNPZiROd%
z-sZpi#b4mT2k+A@FDX*RI4Kj$GDk|uX0zG(+>9|3&MCn=i}WLZ|Nry1u*UH9-}!AG
zKK_ua)rvGpV#PDb>%?@(!+W+d-oMw!Cu6&1H>hXjbmNTpFrszF-lAYu-zE@e5zEav
ztLBu|`ij22z*$G4a*AvQ!IP*!mIR!)^nFM0k~GaI59T!OhD2+!G>iRp6Y0uLyJi>$
zybq`(fsioto-9{nSq;G<&<GijGDV04fr>XGirJ#XIYT=%48uTCX{7R`In1k!L?+DV
zDOzao9a(LV)|1UFi@gPD9aye9f{oHpqa|z8V8*zJAmYMPlu^3ryyJ^M^hG}R+0VxI
zPOzj|N<R!79`2Ln8K>uu@sniR+MeE9HXFfFwZ}{?@OESzB{E2~OtDU3okVNF-eSgU
zuf0N^=Tv3M{$h_ickZB6%!IK!B~rx3)OF3>L}T+Dq2Ps=U!)a|)Af=~-$(I=cZ6{y
zXb{elWeItnQ4~3Cr|B;{gh&xVMCZPb3avazQAo^S=*E^LiAuyOD<V9ycUW&hgGdEJ
zg9{#GP5k-B5fwE1`}<Ty$-8gAOWSk=@9DdaBux+^A<uJ0Gq65cQ&m;;_o$5Zv~ZL~
z$$TE;KBST{X;(^Ao)bj$Tx}yXt}BYRefwvd#WR5b#8%0c>vmmwecg-Uef0BK9Io{w
z`kDDv@GO4&x_4uH!Zmh8ioi=RzRVZD_yyj1`%P{h9^uW1=?C&OWnL95udWb+sI&}z
zB6?f0Jmu*4h}mL6*Y*%5gpUY#I}n7!SWPMe^?Zib37bu%P`vZK?=mz!^I6T<_h{j1
z)*GC0)MdrG-LT%QsrDBn&hXjKewM%V7ylxE=`a6Bz!+QZD!RX<gt|@{r9i5T#9F@f
z?QihCfA%6-664iBeE0$X>^Hx<Q@svdY<nVe#%_S5<WK(mpX4(?@H$tWK?%eD{vk(2
zig0)FqesbtG^wybp^FTo97b4-3#c^5A@Rl|Rl+bBUVH8JN#Pi0e>ZR4AS+V7`{r9H
zrC2Uws&UtK9334Ygy6mR-rIS8J|5d4gxr}-uL}*XSHb)H`;6nrVzJn5J|V<VL+2dx
z`JA?G(OUEFyYI5zY`EX`bX`<&?jIbmw^+p1_-eHiLR57fCED4X`mp4W{n0<hYcIdV
zVpj6t?YH<FfBmoV<iiii(*)rir;k76^7Ju2$rwXKiwOi>KQfvjwtaOpzWd<t!C~!4
zs-TbomkM$TZiu;D-bcu-QW8%z#JedemTdtje;?FtN~R<-5()x2QVE<J=yew{ATqE&
z6Wk1U>BdXe%@ynQ6(2r)z~$u`rWVZRb7Yd@gT*u(gv=?b6zeRD!y}4e!P(POo<6>e
zQOQ|GQ7EcnhIin+VX>Hll#D}9Rwh(+#&UU1niZ5~iE(a%bw+Z47Y;M2^Q<+XqQc0y
zIKUosN>-&*Re?1QYX|PWu)q^t<u{M#RI?l@JVP7MQe&j2Ukm!)5`18@Y%p0$Q54fm
zEKpYkK3KZmV+~|ki7^As&~vbNz`Jk1Num?Fu7w~0jdd_vET~nEmoQpGsxv_2EOdS(
z(UQ8V*_$t5QqR5g(u=Xh#1zjRRj3i?;sbeIaj;m>D@~ecUYqSPzj2rEzV|k(^@^uY
z9<jVUXHiuY`GRJ(Bm_%QR5V>jnikCFbC#<XA3Q>Ox_*Oo7M&E4Y+x*X*G(qi=(M&b
zV64Y`OEoJ=NEo^v*Y_wVvArc#DQ@%}>>se&tohzsZ<1sh<D);pnGqW-4<0?j2glue
zcNoTwb+h8|;0E2$A_XX^$@79dE6_US$>S5e4^(B2F#|$)v{tA|`tGcYhc@R?0+i4Z
z1Eo?neS^@FvaWWs9*F6$QYb_KC7&shgowfvsbtKm(mF1#Qw!2NJKdnuM0h$m>O^DB
z$gC_`EN0YI&Z7_Bk6tq$u){#7HD#7YM3NavlLTWU!9a+}?viNwVW7zSY5o!!aNcG3
z5a>rsRn%1V4C^d)H77}Ojt-8g>H<3$&dyG_d3+c{@cNFRBz@o0G#f6jF1UODJ~~Y}
z-aq8WfAq`zKY!uBA;~mJs$!;*5=b3AG)XE*Q_17U4>>)3$kD+OSIZUeeD8alot?3m
zE%3%Ont?R0S*|+ry5=W-`e*rf{@s6<{lhsA9z5an^a=Ox-C$OvR7FWSOEF`>iWKVw
zGF3P!u~y)Mqe!BCL!^>|l*@}Hi~SmepsZ?6Pfz*m4}6-sEJ?Mb??$|bp&Rfn@bt+k
zw{PE}uI2>y5T%pd;bR*_C`FVSZ<UMNgUB|jeRsV`!Hjtf{L6Ae2&~p?lu{IB#mVUz
z)*6nFj#+IQ-g@gTR;v|01Wry)SWWHUEoCFm^VlBqo}nN3^yk08jazq+TBCHr;$WYm
zEa`_fro5g#VH|s;3XE=uo-*(0%}CS48CPA`XdRWkZQIZ{Ep6A528$UZXt^pgg7+j!
zF_^ePMRndb+bHa!$_0GX|MRbE^9jCT=pAKQ(hvP~0~@3dXasgNs6=6g0ihCR)jo%Z
zM-)}b)3X^m5sao~upRx-BNI94^Bqo0&Q@z?`JDOuHmma$>$475L3`D5cyqyGk;Q5y
z1?LCmb%~Gxgh6Xhz_VPQW4%Wu8B!#>@;H%-w9rwZBNb+}1QW<q1wO@^gmu&5U0^gV
z2S+thDu&k3ZPxV5p7vrzLXNY7er+j3MV{7Zc}k^9u9j;|(?wR75Pa~#TReICfMJ9z
zpA(jw$fWC*oIkoEj0x*?LkJp)M2f%|QbwPUEf)A;z)1-r!&rro8iHWx2F9VMYa;nO
z)0)-AB{y%}pq|Z0$`~2$rC^kvjUO2XkE|Amyd=#^>NIdu?~`Q*Y*uG1Z`|YL<O$9i
zLNM&l_E|13Sgx8FnXCj#XH*Lfd2C1IWT=&A^aG<;c%Y~&hHhZF-cXk{?Jz=!@!@`q
z>atu#xq34;Xe3QPaQE(go_p?j-g);eR_201Hn=pJLqiBiov>^h5-rJ!oV?0-`tBpD
zG9!^Ger(ZcO`<bK6DX=ZihRc1dv|&3d*5b0%O>}D#%A4<WhHj(;teBW2w|c!f(ZyG
z*{}BKhCb3GiV|n+4iu|(<gsa~QHjPFhtLY81Y(*w1{u>}?Q|dr@)-@=+34&?3LIhQ
z{dnz_SLiky{9qV1O>BXB$5p!_QLx!Gq<J~Tji=xuR5iGOK%qpOo#a{Ac~<&9uzz@i
zqMGj#Df2YWtID!sKA(}qs2!9_=<SG1B%7|`;_NhHgPIoW9K)ug^$pqyUcUb#FFpSv
zA6%Z`2vdGm^eqJ;=(;YpGEL80Z+(Z;dWN@wq1lk>1ZzA^GvZ~*7&NM!@#?2P&7b(w
ze}cWkGMb{tfg1-0NUyOjVQ?v4MrNXv3Y|)90B-^^2nwA<^QDvo0fVyy1)~kzd;U2d
zy!Q^uf(!!X1!b19SqG9N(gjkP^2`6jukaH;_0vdFvZMe7Jw1YJ<^YjFZobcJ;)3gb
zoa<Ec?MzQ-$>Lz2gZ+JkP^c$QV{0}mSy)e1R_yKVQCB5D@}(c9>pK3%-}oC`US87K
zh9rqr6)ocipJy4?I)e8s_V(!dj>p}a^}68d>`4sQi^(s1>&@@5ygcK>$B(EM2cQE&
zMfkE-DP=XMD9cGa)}y6BN(XL0$-w#LInH~Q>os{%vToWbG%@C^DGAQDI3FhY`DC2_
z{qXsieuI{tjURY=@{}+!&Ws)D`<Av@Qr9uyD@Z?~cSZ)&fHj!WuxVB-*Ov@tn2co#
z7d&ZE5~QH-4YO>9lYuNPNM%A07SlprWh9B<?8zg7?U^lVqymq{8Xp;N{m3v{iXtPg
z^Emi*1}6e`^yGPgl!CtLST9=^^#WlN+LhtzvL{Uvy513dOHov8u3E01bOfVmFM4zc
z2;=dirbu(LtRhuQ7PA>AV+%4WkbLm`+CTmkUU~I3UV7;b+U`TjVvo(H<--r(;pF0+
zb=zX>fVE@vxHyB!GajCtqtUp&=XkM4p<<#b!D51Ay;`!qy2RO$vdF-D9zS|a*R~wr
zxWWF-Ly|nlhJX!$M<-`Ix>)k+XTLyN2&@StX-TC7v-uu&IHK=1tna_bYIVug@`Bas
zl7QsHhYeR(YorpCRY6gv%*ryN{)Hw_9c7+SWs0ut2^0LWC<=}b_L$Ett7bzm4k2O-
zRfb3%msk=Nn9mcQf8i!V4qPlBVf+YcK!$h&OQA@!=&Em<HAO0!l?i7j4=FQ6J>Mho
z7B2-tOT4qR{f6gWc@AegX8R>l7>1!m3WrW5Nha_i^4!cA!vz<6C0zsk5O2OP7>;h-
zX4P)Ey11MSjYNHsjyIuFkw~#!_(@{*BZR=aKt$3wq!1HY2bfNbQpzb-A>c;CbNBAk
zH4So7O2s%z$8vdvPNHXQw%8|~=1y_Omf(fQ+U6tp|GKW}x(=lxA-=9>Y&M&nlsrii
z7K_C$nj#S~#@pdZ^PH!rr|4<kaB*?L;o%|6<#OkZ+27me<(FRO={J_xEn7vzcAo`8
zh@FSAA9(oiL*5@Fw=dTP-o<!trBmz(qj3b4@Y6r@Gkp3ppT-5x@@mPg<73J!dqyFT
zCk2F>USlZ`J6Ui1y-?Ax%S2YliUKJEufOpIk3ReWfoHwiP?b6J*^IVp*=+jwo(Ro{
z4?ZN(DX+ZzD(}AY7X3KHfIJxxK2i?+bj08C*|x2>B#E@~gM$O4ko4^aM2t<(vYaGI
z+27kIgor4*bNdd9#e%A;Df5i;vvZne!~Wi0T!*@D$AT-%G749u!g$N=+qalk1?^_V
z*~t_7w&7@h#_^3~_74tNtyg^eJKrUdU?x>(s`CgRc7Y_#QA*SI1I~@H<1-qZGh}Ik
z4wj4M6~ky*uGi7WBPC`uu_Gj6WOop;^5NS1@OxDpe4@?g(ZdJaxOI&4mT&ydZ?arn
zGMiVVnZgA_+pWo0DMePWUbaXAO}8d560%&=v@4X1`R#s`v{!4CPASWpVcjEA#jBrs
zl|S^QukeGP`vS9S&eNw)c>lw9`OUBWQ{H~>EyljZt~F+mgx*X*LyOKN<zYqUCC&^G
z6x}eOlZ0{XxExmO?ai1~6=pCbGUs@|rdt~xd~m|?;XWr1*5Ex(IF4^-Ty$3)9UPK~
zgtqOuvv*7qa#9KsQL;BXqAX_YJ-+1VXrHI1VK5y^Cm>>K<=^~o{$G6g%Rj~8!5z}<
zg7-i8kl+5+Z*tYG&}B*@BH<b9D3pS6<oyQ^X)f1bEa_{nQ-~ZJ#_80mNwbV<KEn<j
z5<!wC>@VibXEjnQHf@J;f!=z?AXqg$=gSSxdqI(w@${Hxn5HLH2_Y!5Y(`b>F`FMS
zUmVahEAHHVju&5i8RrdGSLZzV@B=O{PZ7RnATD5TbSR%tB?<K+#S8|9o>WUryTS%T
zSx6`#(;7u=b9p;r#}*qr^Th&{^<1nj2y(+>zhv$c2vIQNjU&-1-a6DQL;0SIlSe2&
z@ZNjh<NgaTF<S(BXDF*Z@=C*c<c0fp`RYIYM+|mDkqNR~A_=snWwnZlwpuBW4(BA}
zsF2c=rH&8`S?)=aK%xaoMx2O@Pk`VBN=Mp5x{(++GLE27ND*(VNYswLH6fUs;Q=Wl
zJzE3^K~kg{-Fi(~7F?ZQFe?h$ra=(MvlK*(FDcXXBSzS*3eOm`J9Tdf+)0vfa&kgh
zmT^{4RXYRZR_(M!y<9iU_I*#5Wh6<$`FV8255vGP3_C-nbB;TAZu7<)Z}8m@-($6D
z7=w$XuP{MTr%KO;DAp*866qz)s-0#a8W%K(l)-xb#Gn2*_{pFANj94eP1|wj&K-(0
zA(78W(c4EfU_M)yw(}G>g%}dm{Rkn*(u}L~bNa#X>Z`Bw@(cHQ_pR^Z_xIS_-{<`D
z0;O!kbWO;mZ+`QeeEtVNM?IUN$An?%7{)%%Ieb8P&o)wF3!>e=w%&X8_xEwm#b7-b
zy>C~SmxKUa*T<Hg_w4WO<Gttd;)3G#ZOXFbop;{m$&)7&6Vp%>1xb=HpU=6xyks1s
zr>$8p`TB4DM%?@W;VrB48RzF0+`W5`tIKmPFV9I64MJgkK>0`_w?VMjD<_s+oPjs%
zHDll4T_3rDQu4vWr&#Y;uhukOPY@E6MiBi)k=XBdNS%)hGyausKE`wi;gHheeT%Wn
z3Hq0zb4cV!H!kTnnz3O@%Iio~jtiD?(?+tYu>|X&_ZZhR_6fL*+t1(Qulxu9F?Vm>
zN6Ca#=e%_1bw2z07tpf8bdm>;-otlm)|VSPZICKvNvOn=3Cp_aVkGuVGrAF-c)SWI
zp~%yWJS`cU6<4QMq~w?m`b&q^f^^JTFMEQPTm?s-l!Uw@AgNWwL4LqITW~Z#W>)MW
zTtd^d$Pg7jI#mqDl4cnJnC0=bd2xP<cY>jR%!7xIdHnQY<kVpq`+>3T2-cC7b8PSE
zH-__WO)4a9>*+H?GD&p7BM2y^5ot!(G+1Yt)iqKlD4oz7i^Sk%N@qL<P0%@BWick%
zgG>ZoDgv4$sSv@Fq#7ww^0Z>H=vc2WQ8Hy1JF0S@TQ^^zZ!dZ0&2O?=UgDfZ(vt{9
zkwoHzlaeAiq%4bgbBYA3V*4mj3I*ULN@bL@=({<(aX>emb8>OQd_SSg6HITx2THBc
zqM*uamX}w^0NvFkVrbbba~flK{OAEHNvLM~48t0!Qj|2*b%Hf5i+O>Tp1#=-oI!+v
zzB9xupd1@4d75GRD-x}#%bdO+(77N}fhyJXsmAtWq?03Kl!GjAeM_MvsfY`rQlLbL
zvSa7sB8;B~vwqvRiAt~FNd$O1kf$0IV%)p8j&ba#c=agZ7BW(c$0>*}P1D_?XN=jk
z?zU>HuIpH@*Bl=o?^;7;S>l|FhnPIy{rz<*`BqTjoMW|G?KZUQoVsxwDaw-P?>|qf
zl8C8dA{I&++l5m*lMo5^cCh404AqlDVy9f4VH`mw{Nzvk6j~=VO-BH-JSEeb;66%|
z&{|#d);z-?K}0RZ2=58TIc-vH+w--reVt4RUVr6f*6W6+PaZRv4Q4d4gX4YVFsg(%
z-+YrVe(?)@<tty|Z~nsHM2J9^#<}#att6g#!LG+8Y{fR$H^r@=NQj81F(!&FX0r(W
ztm~Tn{e3PjE;u_o+X-K`2Ze1I;nt6%wPs!wT%A8<wOU6CxEZ-RODKyRV-3S-DC#*w
z-_x}{Dk%`7iO)gB_kUXjTN~$8M@L7z{L+11yni1l;O+0d$@kuTlg(f-L7=vYyAxE_
zIv0gJYD*FMB)iEc6&wI~i_(tyB1PzmW>~V^Tu_iAHK@!mwgb+m2(PGT6?Ih*yrs-5
z5~XOymQ*IlB%!lC!`kwLpZQ__z5nU|g4>7BBV3A<3F8n4{&{}DSHAqSoSkp@#y8$%
z=oH-sk~E+biS=W2Dm&1LWKk{Xd&|Z7hN9BsdzvKETs&Dat}U8`EEI%6GS_p|eo24n
znacwL8)TB;ddH$Vf-W+vjt_5gINN8Q)znEvrE&<8pfpMd&M!}ByA{qGHk+tu>Drb=
zSG@h^cX945q>OF4Sy>>BXYihJG=#z82E*ujQlYqad=F(KYQ}Vqp|pgEC<6njiix5J
zH;%{>O`fF`S%#G03HWhjy=f6i;QgoPoGlxg!SYZ4+1D6lLhl74EhFI|fEXjvC(^XT
zL?4gdCS#tR`h*CK!xE`7X0tt%lw>Cct&@lhau!60VRO=Boh7egj5a<HCUcw;F@3Zw
z3c4|x;(F7Au#`oKHzVz)K{`pMa+FqtZeXtCz|}P)n{J@W3yOS(R|>OgFd=2ONV&W`
z$FwbGy{3?ww5Y(1+`KuDb4kEWuJlAFjC~*5RBgvN3}k7_W*q5<5z)6~UQ!8xSJUe?
zksT_n8P_XHDL}{eu^9}(E1WTtc`}LuAYi8oC?Zq@sS+rn5%{y8{xsei1dgU@Df4`q
znT&BhBqAxjX_{Sx!}a`_>zJPqLR65!&fFM6peTx%!<8f-QR!{v<lE{ggursS#7x;m
z+pkwu72EiCtu@cxy~qAypUbNy){L?Gu+e)GJy0Xo#ziVs2^SX^G|L9(1jFEBtV3Gx
z!+-G0{DD941FY7DJTE6Nm%^F2&~KmDGiBNJRcWg@i*!6|W2-hGqia8S0)o$c=Ci!>
z=6CpefA7C%XgA!sdxOM5KlBWv+ae2aHb%tvL(k9u{Lk~Z|IRNWgx@K{t^;M^SzH3J
zt9Ev?z0o3sB$+lG=Pg>Nm|^7V;u15A+`D%_s#K-m+u!~+4<0<&r7UmX+bqkty1F9I
z^C?3MF3wLW@`}%Y`ZcmFV}F02Cr=);UajzDLD!F*oSdK&iB5AyYmsuQkP{U<GgDO?
z8I3W4;DZl7;PUhgAp}=fOZE<L^2+O<L8mFd@teQFdecnV?L@%r1Z<T;BErNk{5@_$
zpJ?+*lp;}}GsUdVNYauUdCAH7W70yCm(h>0x*T}-(Is~DNVQL%fe?x;tzwziyI8^o
z#b_j^4ZQmD=lHQd_|p(#y5Z;|FKm0I)w3gBef9Gs>Hk2}oDxC;Au_U*3S?PIRaPXa
zLJEQDdx|_GRSMTT>bz!j1H+|7Xia7dx{X0)3C|zgrPmH6a^AUk9~m@g2%A8;C^_6)
za5UTJc(F&WQ*@B15OY8-&Q2Kmj?1fayf+kOIwg@t2&psBHX}toCxnDD9|(Rxj3X5a
zos@(k!%0IH686eDFWi3~X~2z^s!U0=z=xi;A7Vmbk&#pt%S}U;D4Z*HbR}agt6{(h
z!QT9kW_ZlSYQt$WGOi7s1*a2Kn#YT>8%TpdOBs#8Dtbo{p0X^EQgU&=iEOn*VNHY9
z<K)O67)FB)Dd+?Q=tPmGigD;zw`&}ZzUyOqD=32X1PP<{WF?r+vL4rzi;CMvw@^aU
zUNw-!MJ@>%X%pla*q`mk;Hxwx&0@|{QB~+Pi>~&IE9%8fl48!Is|J77^3u_5MDR50
zC3{tkC=5lGM`PS%%CnVWKAY2SHndGkk>!NNjW~DF89@kc=Oww$NVLLP%P@?TWx@H?
zWi<Smo@KLP&>k;g5HUC~+Xih3qS!?Ul#--MQ6!ow)ns+ekN?<@;p|A?wgl@4SzLVD
zwq>-IDnLNcw$1L;3Xo;lZn4|OZExM`d7kf(FWVRhA;j*_TU5|>^Xzq}yD=ty|H+mA
z+;h)Obc(=wy+W4-d6E&F#|$IE$C!941X79EqM9}a9|BpH;|uWZh*B|S^wrm1=8ygH
zpM#L$+`#^<K<W%<MVy0*m^!NUGbGaW%|~35JWCZLBfJlhtZkw%aIts5&-~$^h2Z&>
zU;6I}b|ld$ckUi@c6Nqyo+L>bh9O38JIkX-k4V#$+qZA=-h1!Fv|$<Fy9qe`5r0nz
zVYi7dmrJUuLMcIB<Rm&}Z4BdR;@nU~@xYtk`yNl9JjH(QbKJajlegY_i}iZNy?gg|
z(eTbWj*gCa@4fe;y&z5L`<}LGxN&&EOV8b-?K{5rop152Z+#n`B;37wmwHz7=<#Dl
z8)^C8Plj2gca^U7@sO;wMufmPww#`>xd;%DbfYCpHngMV^*7$&xffpM-S^*%ioHZf
z0gko!;CH!MAAf)Tm2Ey*nj%6xdAF-IqiYboLApS<YzW4&TCZr9j?Kzr29FZaoLS@<
zmsb~<VZe-0F_$DcLS?jlOPbeQtvZ4j$npx5r!gL#Ds&>*v;%+Oi+_+;Uikq|PM%Pg
z3Z)!Ud7K|eGXdT*_6CBd&WjlJJs7Z_Y{=0rrCBwoEJe8l)di|_#(r|Z;@%-{6cnbW
zwH=r3IeW7isg~qIQD-TpZ>Z*qzH3R#nrc5``Thyr(6U+@^nS#%+`4rGYb{+jG7b)J
zG(+E!ryA4scr&0!gY<$qDaS`gJa^|F(!lZlF-2A|FJ}Z7xLTfbwt7G}Mu)WrHf@VB
zBf(je3}i`+w@8#C2!SpN%Edm`dxB6HF9<?oJd7R)B5nxIfei!)Lp)p~ya1wM%p*vW
z6i-yLAycwqMhMV%Yc`t>;~ZD(4G%+y)<Ezq7B%x(jWPp6Ye_?iauOLdebbX<IkWnR
z^OKk|xp!3KHi5j(D0D>-ioLypI;luPLYWjuFHlG>xnOU9kK6Zd6GR}9k$Kg$9WNZ+
z<H^YxFTeOAhgnVEcf9erAL6}t-sZe_RH8u3jBdH6EGxRp24^j06{!niW7ty#Wm0i?
zc))Bn=j8H&rza<Dh7M(7^^od}Vzx&haHF2njRTj<D{7t4XpfujsvV%zO}w_P@+}0a
zrQv9Aj~5U3xO3|cFW$e)*bUe$#W2z|8<I4o?R#{Vk`@`(Ppyt=(OWK;?C<aIW&_qn
z1G2T2*=)9R)NePcEwy{w9@^skx3uLgq5HbOsp~o(J$l4Aj_mL6@BU3`6&GOdkti}9
zfw#d1f)WG)W*iYoPVmsQEic}Cf!DtL8voUQ^*1Bc+{f$d@ZgAtj~=0u6l(<v&wwQn
zpc0Aq+kw8odvRTCwp;YI*0>O7J;P|obIpTCPk8(7x4Ch+&x<d;%%A<Ue~#~d>ziDi
zpYY(}LynGTF(}eG5*^uQ<2W+kJK!Ju@;~5>H(q0JZ!fkq5m*;v7yPu)ZeKHq6-rSQ
zbX~{Or%zEzvbUJ?+N(!YRUNsEL1K($u!g#-xYU}fi%Wj_|NG0-vx>$30;MExyzvIh
z<#MMOoX_Xny?d9dt1CYI@I!9iyve8E_!RTH;#=SR1|NR-kSxzRIM@ROCr?i>Q!r)R
z4WgGUPc><p#Ooo%Hh!$2GluECcV-0Z5ISWX9SFtA=^0O+{t9>Q-Q|TBU*bF8`8L)?
z-;|OP=c4Rdgx`%C`d7C3XqljdV(bRaPR?<zB`>4v+YFxNvSAoJ%kzPpJyI>vF2(n-
zY8&hbZDYu?9700Z#o(=~o@1@&;_3>aHL4%bSx#P5w40H7mNSllJQJiz!7QI4Ltq%z
zIBUt01mBM|&6>PO$+8q-<0;NtM=+jdJ&@^)+>~gWk?Wj;`7sCeF}csFWWnO_fSdI#
zI@j^|>>+8YkU9_;DH2VZYn(I~(bFLqdqXQ5g7oBNflX4Bo_fJT#ukxr6FR<#gHU6T
zqr+|n%5shl0@XQW<9P1QedhH7EfbO?rMEo<NnR8Tp~IRH3AivRvQ&)RwG-YxNu&8o
z_yq4Q<1lh`bb}_OEL2YG6`Rq5PS9zJ4>8u-_2W*VA|;cF*v4O-;pWYo01@12dIZpS
zBU&X$spz{NmzOA^(1~X3Ev^WZ`5e;?3}*wG$~ind!j*y5YDMD)a#a&XIRD^^#qkWO
z6}I<eu0oE2Dyg}Bbc-r2>6?}`N!Y9Q(V0exBpO$}Ls~}?1kwj??H^D`$>P>c`mQ5a
znmY$a>}3Vc7~16u<6<?;7)VvhJf9P+r%EcWE-x{SVXs=S8CMLOhA@mU7>plj`hkrP
zSYy~b*r%G!7}hJAiz|lUP>WcVcps?qin^}ZG@D6c?I_9u?>w14XRoTccjG4a@7*C&
z8nnSq=D=m!PBRNhKg5XYwr$Ch1f>+4&4#}3$@6?C8tD6e3Xn@ll4N&s-Wn9QGor28
z@p|wcJR>r<Z4?D;`D9!6-4+>?OayB)j?9XRvF~FT8Ui&*f{}7c?TuZJVKkIwg)uhz
zs*()b4>)6a^ympUgy#9@Ux<t_*v*P0ayN7O=%%xs`D}A^!ZX1YlCBRlO-tVoTwGjn
zw7<{s@eO|Jr+$Y2>i_Y72N|(T@oP^fb}4qxxhM*L?brS>bv?s5k98writKcgd3&39
z3hb)Lt>9xj>*@O*V<J5x1Zdj^W8ysR?%lhbo}6*Dyox~DY3}x0zx7)?afjBLx_*Y;
z*_u$>wjnsnhYvnLDNS9)JhW~Y==%`_BuX$Ciwqu}=xJtT$;+6cJN5%@+X5al^b@Vz
zBSYj4I%fz$P!uIy*Yfnq6Yk!<%gq}%d2;fEWO@yhoP-CT@bSs;I%$p)3C0d2A!V<w
z@mS8Eo-=la^Rp%n!_q^_<d~2MLa?;!4Z|>U_x>G*s>cK6Y!>st11!%Qy0)S3TSn*5
zX-YL;(DhB6tJoShdd`<EhsB(|q{8+C^E_b~TY@(@D+%LBE>oltjQxP=quDXnb;Q40
zhn5*aN4cUcELC0d%5$$!7Yn=*_yp48CeIz;MyCQN23Fk_R`m3)<;lfUmYXZmBExt~
zD|^s_B*|EH9o}1%+(h%IO3^A{Oh79c-`o*C^ZYB^obMxoU{TIko?mi#e#uq3WOcFT
z_U*g8_0GGLb<N^n&id*Mts~aLyNDw4)}Te~X(vjM>Xb6gDDsTiqC({<$M^1_SBBqu
z^F8$Z2!p_RPrqqc93)sX&~4U~Nw&Lzk|Yk7ypKKXVeAM5X7!AsC@@3H<w=Jyo?+<0
z2FkKTj4{GDl_|n|E}ty9d;1Qb`_d0^zrI7e?D@vmzs06$`NFHObMxj+PA^XQ-n-x7
z;pLjgR}XQwkGXw(m*c(LymJ2~_UZ+*tRh3vwJkymOfU?=(EG^ivt5TZZOliJjvzc&
zSC3<B$p=IzNmGHAIjOF2`!mdVlRVGx&LhJ%rdlEqv1e}G?!<X;>xfO;(5^QG>rg^s
zr`A+e*HL*mtC=4h5VXW30uwAnSx!l*9)EB|nq?Guj`JR4Ms7a2;`HbS;b5QpckZA>
zRKUrJv6ZH2obOZ>XD3f%EI<G!Gt_dqlTFWNGx9vA>$)8j_4*LJuIrtMU>h8^9cXVi
zsO`UPD>Y*b!!VHN`3}2M6vb{n^ekFj^TKm4@(=#uucFfoM7$2Yh|v;SMPFK)rW_m`
zaOcil9=`twrSvpkvm71Yq$mnLeDHwd<C`4c+Go=k5HbE>`-nmcU@KSueuIS&Bc7d)
zuKhdzKidAR*|POI&%~auo6Tx>mwo2xJM;v_1_TLEs*+`f6r~$mRpkmh!dJekzfc#x
zw!{Cx;RxFi6^@WCOW`VuRVp<svIr6oK@uPifP>SY>@#<>y4if4FUFi}?>vG?A}lT`
ziIbT-_g>AM-x%W^Z*%7k*RNlb{ajgaar%_s{LSCuzy81f-|(^Dzm|PGwz{17LhyY5
z`+v&+@&ELHK<muE{x|;;P<wL1WwUm<^BLds5CVsXhgfU*<daXNIZKk))pb2eiEF~A
zpM1)_yZ1OaIN+m?KV-9YC<R^DF`LcypUwICIcH~Q6h*;XZ@tCu{Lb(2|M{o?jOQt`
zSRAlxTR!{jKC{^rAA;mR86)v0T2WXf$-`D?_bA2c;(|WP<P*H7X&Vd)Yb0B#oJ{!g
zSAUsTUwxJ1qaz+adBpd=`(4gXpA)0!XgMdPz}eaP@H{WmoBx`F&o5}NQ%xzWio6Rr
zRg?Nm3Wh2#u_RJ4MP*CgXv(5wKA+O`Tkc)EK{0V8?K#iqthQU)vn|2((j9f4rw=~k
z{h$1ZgX3eeQfw~HgvwgYh^q}cY0jUWaif~@_Klax{)8NReBTK47CftqGjbAYZmbe!
zVwuzvruBm0;o|&^!dc3~vR$uaDrqu@^@79Y5xY%8yX%R&z`UMQmX4srq4txfAF=K?
zJbe0$C+nxE!k`L`wvKERzE5;1pmM_dL}4w3EwIyyi*6?YY6xDOgW@Y+`a0Xyn!2bs
zUY?L&5^uYv?|8mmF*`ZrSAXLh{Pewd38xuXRw$$CyM~+-;uj`|r*)RHuBfL|rgceG
zI6{`#gJ1gkS6SS8ng8$a{UK#J<*2GzZ8lVOiL(n*h+_H}qrh&%&C|Al6a)37;`!+_
z&hDQ|0qA$kb>Q~mgb+Q|MNf_4xS9~Xr%;xz+sOiZ9_gO%m`*ERxp$Ad%e!=KPgzW<
z>XN(HZgK0{37<UrfG1}Us3rxA=>ZEfVWJ95pGcuWmo>Gu_}mjxz}kp4nV_@4kA<h2
ziVc0c+p^tl2(}`pO!PZSTi`=ah>DCRJ4L4yt`^9y_2Oz^wZY9L{BXBx2rhw6l(QM-
zbc&U@H=3PHVM-}t&e+9_X-aglr;h<!N-@hxQYEv=n5^g<iBMKiG(aYURZtfWqcT=0
z3hQW4gXyE8Yg;zk4byEyX&qDro6UyRYQ@RP$-WS5Hk&=EU}Qy&(~$RG%zB4OWF!)d
zo%YDaGREu+()erKV_c;=k7NgJEw8=&D$~h~-b)UYR&qcq$s>_`qZBzjIN-*O8~pJf
ze0LBkBKsBZxwyFCe7ocPV#VR=0HwrLKp6!(@AqV5@ZNa1jr$%cnu?gDxk?#`P?Z(+
z!4#D<?Y8Iqe1$QVqO9l|p)4CC&IRM%D>EdM4hQjfzVk<PLu5x;7Niu0`07iC17NTB
z8Fy*R#R8QR-A-Z)g7>V?FEGaP(o1)k&1QW3@yBBRI~beBAo{WK`;B|M-EK!pvcEb#
zJ>@4q`3bwWWj39mti#70rYN|6{W^~yKc?Akh5Qo)DN!j<6_%<jB{HH9LYp33)$QQM
zk$gv0m%Q`F8{EG83iWi(&6_t#F>&qWgl~TH8~ni^{63E#Kj+r1n>>E}5K_RIfoeVY
z>tz0VQ_wF|d{WFqen(AVDxN%kM5zl3T@$;+VR?kp6{fVjefKp^jt<#0E0!lSq7MAo
zkG@a3@aPcHMq|m8*5NA4vj_M2w}1FQ@rAE^nQAhjtZEt4FQo8<t!I0_p^BMTuie4;
zp2OvWlsucw1x8E$xitnNkiZ(pq?+PGX0vNi8VYA1c^*9cjLqtTvM5DSzT2T<qAnbv
zGuT*?b_rElQm<HFG+g){&sS@nZ?|;XlO58{49up(<q>_~5WFY(h;IXn=^;AnAp*Cc
z&jDvEyifR$@!ggbA~6KqwC3Q}b=J*}U;WK*aJF9Y!NX4}iy>7tM+vA?iljkj&^l2S
z1!XO8k1+;P;`ro*-EPZgpWWy0{JsB#`uHaQ?7#b0tT$V_re%F`M$QS}_dpzO9xaPt
zALxB2(?Lq)BtCCY;B00dJa4}Gim0p~ol*8J3tgkDn&onVwT9_r!tGl(nWU0hIm&D)
zlV(!SNm0|Sb`)0g%FR36zIDX&^%GL=nNJTW)r5I97sa|yq!7rVr}G;+)eD<R2eN8O
zrh_W5_NVw1pwN`1K`Bq)_Mkmgy&&a|oD|!}%QWp1)(;jtDWQ9re6$vJQc)CanhmX2
zC<;;t6xOkr&Y-ZYx^A$LhpaVOLl=Zb?6hUQTC>?~C`*|tjW!Z_F`1xI^j%N0-Qk7~
zRV(QojgbR3W!WJ-UtJJ$-rMO|>*$&WV+_mXav$I~9$;hP84t%%1GJ}64^Yzxw_P>H
z9BIq_0P%S5_fX#G?tHZX<qS7(-r(iC_xR}kXM1s{EdNF4W5?%^z=tR<*8?@2r1;?v
zen9IJFTM5#U;SF-{%4O_9vo0jN_2uq9E#ml#+%U9ullM<5z1(zD6}OdiC1r0Vd?fh
zV2tD0^QYW<=_P*j!#^Ef)9`-Bv7wVfT-OyvY56Dr<e#uy&LsFU_+bwsxi#a|Gfq5X
z{K8_fz=uG)S?|TAAp|@E%iXwfgE!uIgO5M{glEs5i37N`l(vwV@S&iN2Bclr?F-#z
zv*Azv<WD#}JYqVTu^XaO#zfTX*RSDy$LeCmbW%!KACdWdfw3j0XXixB;14LJ!a2ii
zI>V?5b!E7D<0iLn-(s;m=JfPAUDvT(&Z&xP94;3;diWW)Z{EOZ_~}o7!elxnp-DsW
zx!S1whbulQLsBfO1+UzEmAWseO~s9aTfBDnO)5L*@%?AmWH^}|K#rWJ5}HUco$$-|
zzRdIGbGE+06c+6q+ue@GPoAM-=F@jS#BV(}?%k%EP0`LWt0uUtIe+w=G8?qd99MJD
zj<oe+>e(jodN5KTOKWMnhF!a6XEtPIFl9ke)Wj4p)^O+Lmt-ND#d9tAh|Y?|bWSb{
z`WCkBhQ8}LJ-=Xc+OYK<aigj89H+&qqADt!-cW>!#r&qQye`hjMEbU;m^taHT?Lec
z(DL-+Ip6t{Kc*-Px~}K&;E1E+V@~ee;q|xQ;;k>d!=~Bt{vZBNsJg;wMQA-Cq#=SD
zg3=f*eoJLxDV&p0TWR6?_kr2+gws{abh_ZTf9spvd+kmB*?;@b`S7EUiE&N}3Ey{U
z7YBx$B*VrSY1*B@puvloY>I?Zap(@YQ%`y6)=e%hPB~c2_&Z<xWuDx*j_*XEXsltk
z*)f|;S=cF|UGwPin#p8F(*@2>F9fpecbH03&kK}^OlKAK#NxXh7u}9cr^z8;l#n5|
z?C`mVr07%6&bMs)73D;{&UXEd6yaiZPKaXFUzQbH)@)Z>%Ce#j0b@j5Nt7rFGC1o1
z@$K8Kc0&xhqAKPTt|Ai2UFLcJn6B+4Ft4uAMw67HC>+LEa!5RSx+A4X7kg}QEDsKN
zcK)245@(wW(QnXm|MVfRo%{;*w5D*yfX`>NGnizh+qMOzarE1kYJMn8yS}F=ioI?B
zZnu-Y&`6t3(zPER9#Pj4C(!pjo6U~hc1I>jLa$;U255}IXf=p`TlC=H<@<Ku)$75w
z#>a??;>>+`a?BfVzQ*Izr)Vuq$rJ_p3cV+nmUh>(D;hrj<P)~dhMa{tsWa?$=lsPF
zzt7{xk9hNq*SU4;F4KC-`L<;;u{bMk_HFBj=|nS~Ob3U9F^XJJFoBBYTG3ikZ;ch(
zDq6*SxuDsu`1WuAN8JDHQ`$bER56ekq@a-VU^4?<+pt{DMU-nCIZ07Ri<?6>N_-Rt
zh_UHCWmPd*%<x^$V!33wIAFC}5kioALmQqvc|_CfxVSi@EK2U(y~FAGDGwe$V7Xk<
z8b#CWFxub>M@ms(GbIOm-?p4Tf6i{bWxL%`6ppHxb9#EpYp=b^d^V$NS`H7FJb(V2
zq9~}Uinfm^BUVx&bSPs<G2*+#Vlm^+o!cBN=S=5wG@6gz|A2Z@b9``sHj1)z{MNUA
zi^oqMVT|F|f8!fGdG?eL12L&#@=>I}CfE8Gqgp$Iia?uq`}XVn^4%}eo$hGQTaMF&
zqIXoacrne&C6I6)CJV>naLRUZ%<g)F&r#SI+OXN~c<i3D3!b%aX)ZkJmgV$(O*NaK
ztwr@6Pfs6Eqp6Z+wn$XEV%@A!R<Y}MbiN@Hu+}i2Pw4!PuG=x4&FFeh+XU)3qw9KN
zhWTQN(~iUC5lt7VtYunG>Ai=PD5o{5s(AG5oYVyl>I1BCRL3<QPqvCc!Z^dkJmEke
z&^_syPAcZp8@NK!{cJU1efB`GXdO|NWqLfp%`K*IbQhkF*UxFD6~FVhzs<|9y+YUa
zG;K$66{ehGGK93nXe-)<5O7)#f=7X9tQNYoF<KU)QK5rEs|uBbG4#^y8(cqJ^26_6
z;8cb0M2+s-4X&_kw>uF+X0e5txB_DXhm(>_z`2SEmc!`*S!bq;lDaO?MTf0+WZe<+
zj&0MEa-s<hg=#Ujz&OKB$;23EElt}HY`{(@w6-S_*=`bN-BVm`WV%TiZRPc)oal8=
zOc95}SVN3GsfesoOC-|KvFloT)nh74pFEL_&w;M=G^r<9MMzsw;VVtBJx$Z1qhdCn
zL59xtB%6qRpv*gLkVU<+1Lj{PqD%OqKq<t;HfVG<Xzkdr#Z{I@cO;!zg%w@VvD-Bm
zEEN-W?8wsvi<2eFS$4js&k?UO(P+?`I6$JR&m0`gNL?TlvJ*HsI3T9P(`V22!~tUr
zd6;OruH)$V8pal60!rth`4ao55M}q22XU%~LF?s-z9oc~LKi?1l8F%~@{B-@S!*&G
z?G$(J-J<`4h8P2FyQR+>Th`(g*Y@bcf}))8=+R@LEZaoN9U(-Fk=@(rv-_MqeMG4p
z*Ou3WzMccA*2DxccEp&es|jT(Q-4g7Hf*h2--3LI!D~jeCppUmlu7g<Vv35_-h79}
z;WZvVJ*TP)aYM-L-vda5zQs60-?oA_XoXb>?b{g1Wg7mURbVm{+Ay2VXoKhQ_>d~f
zG#z@+;qqWVK{mS$>(vESDa4?|!v*t2&9mo^xHx@IS(Qv{d0!Msps`Be(AIl$^yFyh
zc3U|Ea-wTw@A=uMp9!><#6qX4XDpYJ_0#kld=%n?bFx$Gf;8a1Z&|HZ61Hd^3Naxy
z1F=LYq#5l!v&Di<(@;(()Uzq8^+pPhmg19r`1e$h|38Y)$l@x?l5AQ++hMih=8bDi
z3P<h}$Hy~NcI2ec+HzRUF)rali%%WOWKtK%Kvfo$RYg6U5y-5%h6m?QS#K`*=_ens
z?RQKkC5xh@%9({Npz8_EmPIioqp-@d_2*=ru$5)gTyXyEDU+(C%MI&`3p(GjZFf}d
z6lD!XS);s1WldB&Ts8z-;p0Zs`sIYa2xKycw`vx*@6z~Vw%wLpXvhX^>F8s}db6Sa
zi-OnRc!|86a<H5-n^Zh|{*3e0nxqUxRkQU8A0k>8EDo=6@7~Kid3wqhzWNStyz?cl
z-Mq{3?Rz9+=%OO(f`g+QEDlZxXYJr?AihbXnpkCpQ5vUY;>lz}O86KtR&*9QCp6O8
zg$Q5$>R0&wpZpp3KY2p%hTUdGwj1X2DP>u6^TsW5P8=L8$pnmnqSPoI==+9=Hk2XJ
zhn8)3PM=ylTiR?1szn=3*KcY1mNwVKG^J6JNZa+@aBwNwqNC|LXwImnHKuUH5QwoO
zZv&9`3j7f^B1=@Y>J)TxP4;+peJ4v#1+H@FLZgfk0bP!O!YeQAI5b_5F4`JPOo4T`
z5xtL%IBSU{&=5_ek3Bl+7oJPbS$0oB7(^*&3hi*JAaom4X=qg-*O`z!A$c$wRmx}5
zr52w8Z@v8%v-yIYWSO3`f&_&v@Q6!5%p;pisP>!9W^d^-GT!!ZE9~jg-g~y&4N4jD
ze@mIJlNfh3bkNNJ-_><JKv!wcE)%h!9;n0`vd9)f9VlEWyArJkLtJ!-^8VL#1!$CY
zD79lgpYe@ve1od0`5*s>e}VVHyh~X!wqo)aWodUCo;-fU$>B{d&d-@I59l@>rq5K>
z1cf1I#cb+`aiEQ>L{T^x0{ax26xH6^Kjn-{QZU+_1~#bU?wxyl>s#OE|Midmue|Zb
z%e?;D8+`QPhomIaoKoTss8xm$7R@lNYkc%LV~2Zfu(rtpn@%PZVY_LKagIq<5_?Zs
zmR!Gnoew|!aKDqwIZ>9yAo{f!tGRLgI!~WImDe)tv6NE#I1Zz=T%R$p+ig*UZj3C=
z$4H?c^;jXuuI}{ojLCc^RPeSXr8tPj!CEuy^)B^c>(!d(=8l)|y-Z4p^=gGJEWPg}
zhTRsd*K0Q0EsN!nZPN<OnT2)uH?s3_&Qeq*&RN#6XSdrjcNMyfMBCy+Akm?;#k!KV
zYe?FYQlRN~w7vlaT?n!$3jyN_^u&?|H|3}2k1*kar_UeJ`j*v&rkPDRC<}`D3{7C&
zCXNnk$xG3ZnRB$MW4$|Nmv?lrCFB~4M2dmg@q`=mB_T$hJbTWg^9ST}MO{vX4qHvQ
zwmhNF2B$3BP>_=*XUS(#2009>L<%jtt|9FLK6;wIBc_(gF|=jR+IOflXMSygo!6XQ
z>~NC_MVH7?qi}2+c=B{h$Th$5n}3gcufB#W9eq&HLNW!~R?H8Nan+Pw8L~2@p<5rx
zBu42g%694ga!hnx&uk)&*>~T1lNd9u1a099UwEA_f9=cs=*Rzr&}%m96{roH&5G;S
zZ_xKW_wL=}U^y4V*ltVLZz(Iwe11T;IU^$zPLo#n+|jb5vzDM5yzlVFlO~B>Y}mDD
zOp6J1IboMtIk?G0od`P8`i_a2;(es+TInKn#yKm^!RTYA#{*IAcN1D`B8ol*x<P}X
zj5Ix=@9Df$D4mU16PY3AS|_rSNi>l#<&g2?ASvl)bIv522-3`nP5%{^l~NTxC;Gl4
zC!j1arNM`W>9nBjdu(YSioRz)Tj2W$KI2;v<-PLCtK?!zXd*froGysUQ#k01r&ppO
z@iEZ(o+6Zc4Np-N`(40D`5iSc<HP&FcC+0lzm9sJt9CHs_i#?EnPL}F+M<-jXh(`_
zc<=Q9`ScX7+S}M{HalkX#Xzf8*uoM!i4ZR<i(;7CV&?SpITA}S9PY--o(Q1;KY9F^
zx8CqPdHR@YGGWuMNXj#vddfm#GRDJF2oefa78s+5@ha6?KBf`;5u(CnEP6_b-~8q`
z`G5S^|7W_kyDS<BAP?7;LcDx~IDweKI;X0zu^-?$jjNnEgd`|kA+}0ZkwRp(xxfvg
z&N20Nq;W5oOU}>F_xdbj4b$nA)oQiZ_gqcO9rtbHKF?Z5;i^4-ek6r-UB`O8-aA`3
zC!`u%7PS54?nvw5-cpLJ<y_H9i^x35b>RoGamo?b8`{3zyGCp_8!j#`n9t`teDp|M
z^apZ^`fFw?{6fV?E3r^%b0Q><sT9kj8SQ#Y;h-oCZL?y#YbYmEPIsqteFI7k1!WDZ
z&5Cj|mF3_16yqExCnt2F<!t+$`%gY0of*0mDJn<bb#y&!Qp@cdOKzWB!==PzVd#Cs
zrrly@7ON8L?HQDytHjz%=Dl*T$}O9XbdJ64$#uhewWZ^NE>{GzpjLvftE5CcnIlST
z5te0%#?DIgFJ0f!`<6VEfk?;w)Ax850x@LDV#>k633a{XY}=BIrkYKOZDhOY`QW1m
zeDJeJ+`9V`>uuo4^EJ2c9h0+R*MZU+TU8ugzkx0aLRMsL(8i%jq-;<z?hnK~QcR=R
z#YdxQnwBoXTVHsatVLno`G_ej?|kJQ^#9wi+ck8_6S@X#6wjYO=cSik;eYeL{x6u$
zCVceKNBoQb;lHP<rlc&nCs9GN0cB;s7FA;HSDgERbC8UuS7JmNb3!{qw1Hh}X>*6R
z2J0L#Ct?<8q}8o-T-}Ds7!Kz%tdm2+H{HIQ?7bH#3bpr>>Vw3b`xt5aju->RIn)3_
z8f(RJC}*s77{E=+!6IL7BuUD>RQX#*t9(rx12GQ=nUZc%A@8@=3X_WcVm**i69c7O
z70G}y3X?QGdor?H(&f;|xq^+)q>uq2r5Lft#5src9<8B|o?YMId_h?h93CF-Ndja1
z^LR*&4hGISoU`=42NYFRF_}#0y#$t7YgsOrgXh>d(J4aA$co;e2bfoBjTR!f*yV&E
zVq`65#>UvaZVUs}w4mvGHro|uae`Kfi`4}o1`1>Mx*Sv_BYG006|40HKK86G&S~2n
z+wF#gOsgp==Ck630y`2BWY?t0i5GfqnI3?gh34K%7Si*lPdPb0=IH2{Pe1vTeHH+s
zanlOfGp0li9_tJfSJLga@|*`RJ8R6~vuLnJQPl;T5QtsR#o0N>li9v#Jb3T`=Nzlm
zin1&zih|v4C-F3G%Vav)H&kOW86RWu84HYaZm-)Jp}ebxvkxCWq^c^eU%$?c8#kEE
zX7nMjYZ|tjEl;05-514?cp^}Jq$($rMM+gwtX3;BP!yJ~Q-tK{y~hv2<*N)yawHoF
zA^*(~4j69wvU1E1rkE5sZ%zSE4kAYD+FlM8uh32rO=P><5`4rIjwZEqr%y#>781FE
zO+KgZdbUl+^q{0q8IsU6K_H}hQKHIBr(3$PL-!54W-H>=T>_`sstYEwDKUGto3+q0
zQ=-%sXB9am?A&s&n4#*7?>ueXP-xAme@2-L5rXAruT^OKM#A2dI1;<UqBNwzR$pmF
z){%95Mkov1q{au&)AcEHEYG(Mo31Bo&1T!N+x8eQuG-~f%4*keVikST^g{?-St|N2
zQ`8PsXu2To$I4io&g2|X#6294hr$z+Sg?dX6I9^tYcCCEk^x;gESfF_ri&@2bnN<5
z0s-e_QaC+5=YtPE<k_=RX0w`;lbgKt=G(mg!Fxm>X&W(t>|-Y8K$8Q3M4LTp-{5q{
zSZGzpE{f9L*@`wqnsrClci6%)oy~Ej<9xlM>wA`m2l(L8Q$sbKvgun)-=U46H&HY&
zTGJ~LdJ3E(<Om-Wg19MEp>et#g1ofAEY=PFUJ}i~kjs)&BBumNq7IyP1B8~uCML(>
zh8k(4AYVB!n@DJDv3l@HLl*4C#1inBiMiqyFn0O|JGea?BQ)xiGR78S36q2*5o406
zb|3<z9O-;VAA07~>7HIa2HcJG=#lkRmf|)rnM?>l{MIfHY_p$u%Cgvdx#<C~7dL2u
zxKdVlA7ui_(#=N`2P+jJ0$?mLMUGC6Q0jfMk*TtpI3}}-?QYADfBYj}d*yWsYb1MU
z5VMAu&{{h8vMT7h7OmtxZo7uQ@6b+kM}-S}@oq}OQoAae9qHFOXHo=ICacl<<}!E3
zT7k@V+a0%V-b7`InD^fE(MKPOjZaD#CBAzT=eT|C8h38r<n;Mt-uvVODL#XRP!5CU
zM+xky0pkkVre%v}m-4>Q9336eb>04XPp8vC3zi2sR1KfeWjuS|_p-wooGnJ`yCM(=
z_!uq)!eim;x{eP&{E++i?{oe7b&Rz<c=U+vZp&;o+yA@qS{j?L@tO)Duv)Fq#?W@s
zbY+#;?sjaqTc*>R-EKPwkQIc4)fwY7zW*C2KBaLIVx<!2+cS3SQ%nwc-;$!2NNix&
zwImC(#hmSK%f-cttQ1Na*1N6T3@Nc|TC_17EEa^O$4BU+#s^J^(uL(5+3Z%h()0Kt
zvTC5TiHGMc&RMe4G@+MncaiCAL(?`iw!vymAA7c~&|ySCRWg}OIJ|X&R^n1%jG+lV
z>%PJFJr}zTL1oEM(F!}$)MYtf!omu(%8HpL0_PmunPF@Rsf1KhxnpLt8$7=Fh*c*>
znQar$7;58qa{7edI9hP+<}t2v0!5if*S7>Llc765-<;7zPoV{tF-Ui;jTm)GDyBi6
zmPHv0h2`Pt1z-QxOK2zTv(87nP9!o1hfA*Cy2&T^pRw6)F@6gvF`qAZ@2BtaU;S7A
zd%pCgFYyom;XmNp-~Mer`sgFH3Mi`xIg(99)3j{*NEaeaFr021^0p&sk8=iFmFQgJ
z1H>-T#Y9vFYYS8%UT~B70@L-x1lCo=6ls0L+JarzqjD1K_MtN-XR<AZ-H*cBf)b63
z?Ds)hYhn^pQ;+0LDXl>(LXyce23bj4W3g2+v{e#yJ@{@*7oWkIF&^QPl051`)cAD>
zeYw<1Sb?#}I15@n8<iC?C!7{b8>=j3S)i?^&z@2nair)wNu%|mC9z5kOYlq#9%~IJ
zCntMKH9*reFBppMH8_K#!D2pVz25AhoT{p@R?{@X$h#_#L@Bg$C~XLX1NS%?ja2QT
zuo$hdN{hI<u2>!|`Rdod%<NCz<?S!M!8QeyGTbhn(7*dJ4?p{aFTME%`nDT}41rJO
zK<&|{#A@g|v5m4;v)hSosj8~|KC7<ly?@=Mi#L0?EN3;K00O6KtzaMvq@fY8&azl6
z*qop8_S<jM>{b+o=(0Zi_=5q85*PeJYmVnLUcG&rFTV2@+vXOZ-Tw$3`=R)0jCHsu
zMJVTtGnV;MY>58Wm%q$UfBX~H>oqxNG1Hw)_C@6Q_!yA=Rl9EY^zEx+-?1o+RQFM@
zrM0koM=E_$6e1HCT(!OTtk-L_)~wfStdk~e)Eqfi4E~zpW)Qq5r}3KF(d}AP23Hh>
z=%tZYiPd_|Zo8wZ3eL~YS*=%t7`-Jcd5(W0J0G!3S7Z`xzhl#G2!2aSJt6h9olvM-
z-Lr0XG-XHAHP}T#RZh6LSYa2A>^-ZCHQsruvSgE2VqfoO#BHS822vKLic;W{N4p7T
zqB-B4^3t_q`mU!-o~>_bLN6n?mQlO!JKC<HbfR)kL9!>Tu}m<e&a>+`LQ^g!bj_zg
zA3PVkHM_ngs!VJmT0vD*=-qB`9TtN`XAOPVk#eBzdp>+{pL4A-YDV%kFWp`8#!Ij8
zuYd1PhlEr`?|US|MRqWM{JlS5)vWkO|HXgKmruSz*Y@J7Ku^eszGhNV)Kex3T=o>P
z!8(QVitOb=vc{pbV=|eeijs5RQ7(=+xqS~;PX@BWz|PT76eWxKTqw4-Aa^2!Tdy`a
zSMb3HAJQ}(-~8q`sVm3P(RCg@`fTXL61gZawnph1A2R5IM8*`3oICoMB(_0oOtREv
z$z)>b+E$_fl*DFhRbiAN`iOA?iFuz%*$^p-A(E4H=A_H;wY{bOph#C*y84VP=<|SM
z=fMF&XUI`NQt_x$FcgF|_*0EG{3#}4lJ94=#pblvucQ$7epq{tGc^Qjh0E`sl@W4@
zb$d#;DI|$DgeXmoGx+QYI-r%O4=r80W88wo(2Hqf@HnkV!BaX3C5r>$U_5lLIuA^T
z$!Prh-d|es122e1N3XYWdO#^$QJ@ABPirk**HPD%BnA^vN|<1p$aGTEq@MHDDeLWp
zB>38hO7dLFy5OSkc=eT6@LiA9gL_5}LWQ@6(H=PjPun#-dh&pS#WD44MmZ^mu)7g9
zwR>n$UgLNz4AW>H2v-SW3^3#>2{b*DMkLn27{g+|pm2r2v4!khx@L=2in6fO#;}}D
zn3p9}=a^59S<a@|;4qt>s&ps}#*3SL&Y7~PNFnm}J8$x}uYHZ5{`e=f!*xHxV^<?A
zlv1M8>N|{;;xa;M<L5FK-;ok;t;HG#R__bh_~*d>XN;z^DJ6Ehoj`4s-v3z&aZgt1
zTgWleH!W@3lH)MzZ#Dw&Yegcm9;}wOTM=`|A;GpP3PP6aNso+=zvA%=7N0ERESW&0
zr)SfxX}b;CB)ZU1xRPR4lAYl^Y}kbslRIo^X?(+UHskKA_c-aU@#xV5v{5W)b3*5_
zs-}%watkWTO`DQ*xw^Epv19EwoVJb-8+`Dzp~q)Q7mPWR$n>pf9xy^z?Yo5UC3<()
zMGD(M&P?kG^SLOZ+peYco-PD}(nO*db0VF<ZX-gzx~iB?gP0I@9c3Z#!K0@uIGREB
zIKAN3z1Ml?OJC>F)Bg+I*5kBT_dDxIxko9><Hrw33LZZEHedOruTYeZrfnccyzi;&
zl3)AAH#ooZ3jOMW)zgQB^;0I)=)S=>4ZiEh*)TgiW<Ec}&E^~&ESVo&r<g5iBV?WE
zLO>Ua!Wgp6+`fGm-$${bj{-HtAb9`Rzy3A8_SIkF-o1M~e0au>-~B0N>8K|qlL@qG
zgH|0XOPZ-t;wh(10_9LD%YhLyDrtN&I4o7UkOMGD$2!7Zq(F)mQ25l-`Zyf$#bA#w
zH<(Hal)bVvHITi;6{GfoLA1f-0d~;?OK{lG<v55!uH=v8%%$jZxP1hIlk3kYaUw=5
z^kA)G5Vn`TB6>_@VRdNPFXkcNUKvPzLJd~-n`XzZ+wt_#r>tfbNAn|8$hbn&Z5vb&
z@pYFXx$h~IA?IQG9y;Jr&{zxsb6RULJskq#%CclOn*j!`%zjeZZZ~+}@0nC%;S$1u
zA@l)TSjtI7=er@5*8y>$_<9M!BcOC*y*lI52On^;eNM^|Y?Am4opA$$aWPx+;m<zA
zXe$lEuoo~xF<^KLn|8;K{^Ezc_S&1=xP6<VZ~_@?%X~K57oRa>hynf+t<ONBNP}25
zCbUjNQ62X}`O?*aOyLR^i={LaPIGeYghvlMoXSk<nt4&Ms3#mvr`$ML;^}$m)(y@d
zKf*f8!D0azT()d4HiVFvPG{`8j$i$iUtzc1a(a3yW~GY&Q$ei0#op#<v@RO6nUs>?
z$*L}2*k=jMcnOJ(Q_omD%A(xcFpbtq&bhtdcx+6x)|7jHJQ2}$T{j5dWi?SaCk)E<
zTKwKJD8!%7h<O_!4aTZD({(}inPx;Aq@hzv{dE%+e!-p3ps&(SQB;m9RBU_A#cD$v
z8rJOwTNWIg91(P8wc2ogzM^dhEAb&d_`?SuF`G}AOhiZU;PiqVGAfp&P=a>kf!5u}
zfF`rru2}aCZ9K<6Yl*qXSh4s>D$@`2QavOeYNd%WL)N&mKsk*{ptM0-ht-Cd#qcyM
zg-?;7B6?C{a>Al;!w036;@aW>7?schZ`2@K3Dt1p<}uCWfVOcMb--+KgJ)+A=NB#W
zdP?d&)@oD=bgd_2`Ct5}{~3S#KmMQb_LsgwPGF5tDcdHo?LCF5xqJH+-n{)9=Z_!q
z$xkf)qM%A0E;Pg;fzU@qIXj@7%rKKVZ~f|TVCn_4!vl<!oO=@lB1%!h(5$kAE?|^H
z6(zp!hECYxT)}Vt=5KL+w&DNzkN+uu`lmnOAO6FCK+|mLlNZcC4>}yJgjTOCIR-g!
z2b&&a2b=#s;k!st3>~%BSmVe!NaLF{Im;mwlOg6v(iufUhK`Iej$)(~E1{gHlrao9
z$4~$m)b(+onU5!%5Cyc<hAXVWl%gC|&JG=N9^9v8N2C)NZHJ$s2dGNu+n4Mq*+ZB?
zsHn##P3OTL1_GWeD<^VPB$au(e#%FmCQN{^>zS$o>lJUk^ai_53bV<eVxn{vM~g#o
z2iSY74SJ0$0cPV=*!Ml#?GB}kP>_d3^%#?&wUTRh=&Z-0HJQ{L94-;xt~1IZg%|=c
zb)ZU&O{CPZ*>wE$y&v+?$M0cWL}^7Ewy3fq1W(Elr8Iy1N8jVw1Bn7R&R~s|`+L|k
z`5>}}&3eU;-u)p}Rr9s4|MFmGldx9r<GxeMSMrDia?7L1C*o`nLcj+RWya{xLvd3B
zJI5G-Ci||(T1W32GKy>0ukrl(V{+e7TElWW<=Wwrqxp<!QQ#&Me&v_G&SxL~jKgi<
z>%a6>+NNi_YWQgT5rxrY;mp%_J^%eb|J{94Gg9P7`>JQpp1t6Z;GDyU{sqT`5ndfl
zMaRMvg7`a<C82gcpUdkRc0<740JiHSHsSE_5Tz8)o;}+`fa9Yd9Vc>TyWI}ysDn6n
zDDWW#y1vB^;$m$SMd|jm{WJj4ipxglKWOnuDd2sN(h0P}7Bc#&!r;mpXA)5*Hol|r
zJ%NmyOvpyw5N9o@B-ZFEvf5pcvK(g7XL8?B>tpeQ3gQN>m61ct8Ym32dWo`0*ezO<
zjVuxiN?fg|iV~+ZDrR(&o)_b2Vj%g18r0t->&Azkq=yDvgDJpfEBI&3L-d0qBN`2m
z-N1MN86%n!>!$c<ux^g;93Ox3nECWBbv-BeNa{Q)^_0foY{_mH$Wc+3GYlo|4vNZP
zs0n>yx9zYhGM`ww7;sg^;_!%7bAj$X);Vfp$uSAzE6AeN*^+E)>}1ZSQ|OQ=OL3dd
zS{P<FL-L6%*>40d?47a{jmG;wd!Gjn9`N3K@A138`@4Mm+u!1E|Lt${o$vf1n<mq5
z&Pbsng-(wA7}2Ug8ABir>@X#}C={LuT4AhAJUxQT=ak9XkPW0^X|Ju=*XO<Ui|8m+
z5}O^Bh6yH%jyfxa(Iew8k^$p{rIm)_lSeFFW%=I2=S?1AsDWjqg~Xs>|Fc7u&pndS
zT5@FLaNs#D?#)>t1yN~oCsSC~iJT(YB#GYEnm$D|LeJIC(C40Dd$LxP*0Fhd!Shca
zqrB&Ea)@sueV>S3;ON=`6IZcI9-S}qd&V7)wRR8fjO@3ns_=cp`>-!C)leAvz8m7K
zFR$^kEKpW)aJWFJNYkt-N=r(z2T94(_d7OQjZu;Fe#?W0AJg<3vJ;AJ!qfGcNx7hH
zB_HSfY>m!ZycCTBv@{mOnWFa1fRf1Wz6FGOk2cixbRXZXl--+wj&Hj(7XvkV+{-bt
z*M@wb;S4iYnyayRl|_M%9Wex^v)Mr3mtt_RoOA8?h{dEv#em70m+suAU!7sCp)3l{
z&sUtDo(f_SVQx#_eEki6^6rm$^zhNXxQ||Ww{PEOwOaA~`E$y$-0QNe6Ssj;M>aMQ
zzzY-VNGegoK5m3s4-O7!nr0urT~*cI@qBc_7{$m}6Ku!tb2X_}U_DDt?9+Fhb(B>>
z9*Sp({eHKjwI;+!N}%lkL;r^^K5e^0YfqISX2GuK2M2WYB$t_SL?97z#PuGfrR&N`
zk&+lMxWW)ZL*c;KM8?Y^Iay4s*gelQUZG<_>>viQ>$r9MHmdHR><G#a78kNy-s7TK
ziBdUb@R8VgeBTRIvUIY{jS-dP5OSr%W<$urmP<OLhQ+>$a_H#c#@@9ZF{OP(b~k)r
zg)1qFXUu2UP}PFqCzQoC{>}gR9e(&{AF<uG*qkYpz?R;3;2}jvy9?a^<Ox50?;#hP
z9etFd+w~n?({iv}@`ZO^=NrHFC6)(=7=N1&{`|WXVav%x=(Ke`A?|v5uc#+eUVZZ&
zwlVReAN_<s{gWS%W8%%XUgwKne1q#Z4hND%=H%!aIvYrGF?OvVKBvHXz2SHN`G3cc
zfBa+a-o3}){|A4cn>VlHT*cFmpAhw$wz*)pJtqcF+iqyv1V(mB897KsHGDU?^LEIt
zLJy76xWh?C_9`hwh+ZP5m5>DTz~<8$tWlW3{yq(#wH!P{tRmvIqVwo6B~VySl5Uxl
z<QvHVa=GJ4NdoTj&?1lBG)fC;fs9g@Fiwa9yB11e5Q=Ju5uG)uMMw)e5tS#KjCLbm
zPF@3Ao>!MTyy@xsAg=7nbMw}9u3tOCc+2|viu1D#>utmG?JrWgLJT}%pT{%8F(d9j
z9w5dTq4BDIcum8fTg!aDAcP)k&Avl6#-ImyCFY1J4D;ojgTp1ypFW{&GtOlCcEkGO
zIo<Iwu>fr{Rqg1sz~gmY;#`665|5ug=0r(`O<_c!NmhHJgAyPg$ix^Vi^@5++YL{i
zKEa%?IXph*+O-=LML8768s~;!Id$cb9`{zmVXd?hLGtj|DUTDhmZf_L6veQdPoB+g
z!=xz08xqBAGT~@>z`=Y$=?a|Em|?5;^yy=o?T!Zz?sIy!=JfQ8rfDd2!QsIHcW&M0
zdw=phAhX-;u-5MV+onVG!5HW_THjbFmS9&Oc_7h@0e*r!=>aN(FofkbO(Wl@t}h8G
zu-`E?O@r?}bycxkEcV|+K667HniEnMA*&H<C#3}XbA_d<N^D{9Aq>AHBM2nhsj@YC
zXn3r(eD>a_e`P)Sg^JIU_EYNYm_$b#x7b4AR6)_#w6Q~>utsj&X<5ngHVQT^y9G~9
zk;!aISsG&MNzo6KQp?0Hxp?}7-baiQTBRPUeqBw`wZoIp+7QVY=g?XaLL9F}Qp{MZ
zu}+iAC`)Ur=}C}f@=-~;W@8j8B{6g?3`zr9lX^YuGDO$V_Yr3c3S)&e6$7>?_Y1zP
zpe(0^azd92e(N{B#e<J8_=`XNv6zCUz;1oPv@#eAwlKs7?!5LgU;MRS=flq)Va^+N
zyH55XIdXP!&iQK1yPrNL&m3QT?FO&kxka!gbURKjwlJTPGjz%j>XP!tEjT#gJAd&3
z_n%#G)+wr@=G{-9@nq+@dGmzDe9mINV0rx(^OI|=FHY&2L{b@Nq+hft@u%PWx7@t_
z694Fb`M;v?6F>aHd%XV2J%0S&_gL?CtePD@Cw%hwl-V|G(Bey`jTMauYQTcE_&<f7
zqJ+Y@J#L-Rr~xYuUY7c$Rk-qUX`Z4?nW(|qML_<c=opd`YPBq~136nzkd#1pAxK2H
z9uA&7Xd3pWno5eU+;AfuISHJTwM<m`Qs_6{)K#U)S#ItSBTidNT~inh8v2+?Dq@Uc
zvs=sIGnwG5q3bd}6kMzWC5A#f7PFc;1)*!{Hy*1Jb!Ay?qEL>LG<2CUhsF*QO}Q6I
zDy6Banv^oD^%|5RDskryIbfWjwgp;GI6PcR^6kYr7Z+!oU!1XPI#!!=Qu2ciC$MW*
zkV~9S#NI;^3i9&Cl7mUXYA2AQQta>L^X-PscFU~u=*f)Gbpsr07z?pBnykb=M_UWZ
zVU!fu^?Jv_(S+$_!F0O7Sj#A~iYb%zaO34+XO|>rC&Wxj3bG+Z*)N%5_*$ijS!nbT
zutkA&1t=}yc*w42+iWNc%XCsxXoXQpqzySUFB}tXB|6$eXal~_bbaLL#sQPbgvXB_
zu(~*->pPkBeNV~}>q^cq&au|8ST2Sv9ob9vz6Y&kzi7=sgs}T-R$mqMA?I9*a;?f9
zCSkO!Awn8f1IDmCJml!)1bgLsq7;4C(KJ%*=4MIPb<7WzRMW`-Yii1}qN*z?Jk}6X
z#1>ApYG#DF<vB1AZ=7?)6#3(S{hj}?#plE4KcRF6r!<ANOeQsDQBX~52o@E+>}XUJ
z9gJbHv=BVeMhDoT!xBtW6$(WfGVjH%Bj$w8aj#e}oa6lToWdtm5h<q*-*+T028(*^
zriKeK8|Y&s1b?|7$a3My?#W^X=tUW!RLWeQPzlJAA20ra7E+QMgnY?{Qi|Sta+Kv}
z+cZQ|Qx|g%7YqKA-}xO*ADr_3kABQ_R#DZClskNi#1v?w;_<~()>?7&`kRzZWVdQa
zeZ$%LQ|i)^mkS;|d&19t|Bv|okEZ<R|KRWNg|GiQKl|?Q<98RFU2Mt563Yp*n|Db^
z$NV>c^gZ5RD|+Ryi)$2%1t%vb#NheSr>9KHmXqT|J-g2F%dhhEhi7cOXKF2>X(+Pd
z$!DK1o1gI8|JiR-PY(FgKl=+_e(5HE=i7ga<C7B}ee!@-Y20jvElONvDJF)_@5oA~
z07H`ZE-LguI55Uyu|f$;h{<6d<B`Re%SR!4*sRf*fnqsWfcQ&lu+h?~D_fA1MF9cm
zV!~v(?vzpF7)jX^bHZZjeWvgIkb5F9hS5SfZ5mm$7KPBSW6BUTDULK|v-g=(N)!7)
z*LC#0(4do1>|%p8LJc1~abpzD9kE+)$w5=s1>S3Rtzp+zw0%z>TV}I4g|jTn2|l&t
zKG52XaS&Cci?S<3YqYil4cQLGa)g;=a%<ZjA6MkzkSL1+i$gmNMp4(&O+WwaGafyD
z2*7l@U_Lvba3#BLMd!D;LW7DVdO~P11D7f8I&4VXJec$NX=Jx;AZ7Zl!zZ}7IOn9<
zl9VCo4sA6iTPX}A*%f4&vJ!GI#-K2+z!fFIXKvlR$E{m;akilMfod|PsGKkfF)$>i
zh6p|n&B-X3%o-BIjuRn^7>kdArtdLDfh%i@vLeJNKi~Hp9V}_jo|2=dunOyt^DMN)
z4k~AvYD@N!u<GbHL3Ay;py_%(_~-*+%sInkQVIM!5FBEP)DxkO&vND<E-zy%xO4-r
z>v|uR{sP18s;(&svLJ=Y42yp=FymzZ;tD4eYUG;MN+!Pbdd0Kn&xx(Tu5MCszFqUh
zmtNx9wQIxv51A%Y8W@6Re@)AEov-MtGC6A$GHG^Q|JSM~zfkcxU!PGpja3?}4aLq;
z7KJ!UD~->QJ_da9U_hz7=i|$bF^tIyLP+~1Xi?bx&F+05gpA2aCZz}#p@r%mf$6Ng
z@R*_;LR!mSWf0zt-c~Yk$?`QR8N>G;Xdy~BV#J3jc_~&3uv!>VG0XnISb^(85ChCS
z=uFsKcqC$<Xf_?HihTURCrsQCzws;Ipn0}py*i_?j=pP<#Djz_CR}W{Jil0>Do5)x
zr>hH6-?C|1+7R%@pr=y^4e#H7$UpvPzsvvj@Bdw{-Fk`Llh5$YmZ&Y>6x@3CO+I;c
z!RfB2S2Ln1DeID`9c|K-)s*RhbW!JxC)fg2O<4DyWGh1MG009P_=Z<reTC~cZ}R@f
zpK@^XF4taunMPN<^|fE&J3sv{oe#w9K^yuQ$;wexwNUZ1bVCYDFvyKE2>rAXJYLz+
zL+sLmT^WTbO1XKGCi-E=Hj4VNz!epEh{WMD8Lft_agr_Xlc=$MPmBVWwW%Q{2x;`7
zl4&MuiBOJVT=s^25^eC3@+uZT4A-(6aD0z7nQC4S%q+>z={vFYxq5)irWMh*I0JQQ
zan^`_38B0O?+IN;QVusYn4(~7c7yMiI70XkDXWsAteDRidp6T}UyTt8<IkQh=Ir`T
zXxmwuCavVPG)=?B#RX5EJi!>v(b174>sk>H)}dgtGi=t+MK|RmsmK(@V6_yq=sPA0
z$MNBk?Rh6M1ZT(^j*e&CxOq&`pEH>$QWJ66K~%&vxML42tC$TT^pGtt-@C`*!8Q7>
zg`6phlA<WFg&X!0GX0IrM%ndc!bmvC!nBkFa->(w`6W=3uoNMq^gt(XJ9^(x7MjUq
zf>s*udl9o*VL8SW&{635&2|ew-*!yuDP7haom|Hm&9p9=)s;}bZLv>g9VcgN&CoFD
z{oW3SMtIOaT@AP!8<?x~{*l3EZL#0~0Q>(QNe7<~3k)If^x0D$KYmQxw!`i^ffxTd
z>nv4Wy}*(icSlzfbH}Wqt`jE{IcptS$!ii))?fYjg^Ev9nSfx&o$uMUJNbb|ekH94
zIf)mLGlRAHr6J?hH)!0`h{w>tBFlbYXBd<*7@g&!Gl86B(dA=HpA^byw9$Kt;izAz
z>w3S)8*$Bkz$r&+tkz2GGlqZH4GfMww4wsWh+aeyNy!VtLk;xYAx;+!YRKxDlyll#
zaO3zEIvGCv*(cn*c8Awqev5a1@MqAsxa6=Zv-N>!HIJX1^5G|+Qp`QdR&*iYL(jTv
z*fuRe8$6n9N-z~a`{)7R`{9rItuKC+kAc<cGjfJ{y5MXVc=-H+E*myoq^Bb)OCLL;
zT2oaeMvH1YhKTZ!H^2N<{_M~H6qN#geTp_2a>BUCy_fFs?tAa?>X(0s+poQjUM@I|
zj{4*_*KXeB?BW563iA6L9bV^S)lryXH=~6Ht(1{NGo~R}ZP=acm!=X&od&j#LJggi
zbk}OY>2ruem)1&Xyjkd;zW0Qr$QYC>g!vde?ZEnpVTc%yaj?>1p!N!_z-T>)c7+^a
zU3CT0Qlj9mH8KtZB*<dm7=5NHOYEc^g7`eMNzG#B@LnRwtrZG-&Y7;iU_LX{wIij5
zlsc@L;M)y0Lsb~8RU%@@5oHzHS?GmC;atgVKIh=@7;7Dq=@g|DUE2=JUMaR++tPL2
z{?J=4my&)Ph5j9UPYjXu#W~yUhTC`UFrQ6P3XG{J%Lz(lY=EphRaxU)NeW(Ia4SB0
zIVH}|&Zw1RQrD<6+09v_DGP%u=G?t=lbh8c-+t#VVG~$A-SP0_XI!k-q?`v|ynr?Z
zm905Ay2h8j@HVsAA+y;5N5_X892~G%EGVi1J$Sgqf%r2ZrUSiK4SJxohja%%2(lLx
zIKC_k`o5rVWMbSl4G$hX#D|vIv}86ZSS**s!H9MIULj_Rf!^PRz;e0d>tFw6ay`e?
zGpeeh>ss2)n(RA53S^;H4q{{>KaHQuSa?QiauR@M|7T;7VYnXmkrw;OL!m|*#qjzW
z<O}i2zcYhK+FDCpYpk(cyLNrgZq!;!+HK!c*Aqe%CkZzYSZ1>slj#(rwd{I^Yb+;m
zb1w_1%6g!?voDZ;oyRXmwbqDVl@w@VO47lRK_wE>y;^032CXld8du-&u`-W;*L(4X
z5tVn3%C)dttaJeh5%-N6j1VN3T$_JakTvQbMul*U!@o_!Y65XO(9UACc-8rE88bX)
zmn))ta2Zq%dmKM2_9lw*dL&0@Haj3tbL-|y+_?4%{q8ZV(=A0gMP&ymP}wQo?;ty3
z(sUlyo0i2SGn-H8f+t5!woceX)>76JoU=@vp*}jIR|RKXW`5%?J(kpV9NxZ%ogQ#<
zyI}S7g33sQ_hdF@QdP`mGr??w2Qu14b|G`;)z>+?d6$pR&MDm--+63|RMmuC*YVb^
zn=Fn`Soy@WXBXJDqq3e~`Sow|$AA1UIX{1dbp_GG(efHOwWyph205p*Op%5_QfY4y
zGKRKQWr>^3+3t3P;Hj$`U0I@M*!Ku!R42KhM^!gScWa!$I3Z+mpNM@=juD5(p~=>g
zbwSWVk5)?WG4GURx|N1RuxeF{sp7b>jv)}~(B{&Jb1b5TD~Kr$8TJWdEtLb`dt4z#
zoN`16bX>cNloQs<uBvpJX;m<*3rxtAN=q@0JuwG}8ErHnWwO#Nmj@glpHNR{D5dw%
zAb=ejFYi5OI%PVO!?2!-i@Q?QP?Q>)owyonZ8$zY5d+u3sovTG=L(#&L}a-?n=Prz
zn#KowAMi@z89Ia=>N*L9o0c<-u~e=?*<p{Pp(-8oNr|!r$K{fls(JY65pCDfbWV0^
zXq2_wJU(POo3K2bbNkLqESHPRgE&JP(2_KMMtN*z2gX&90xSPSKC2NTmCqEh#!8Xi
zcVq?Y^@eAsrx=q_Dsr%#GG8v~H)}TA9U&wrrD<*24qYjpJbpscw#1Y<IXdQ|X=&OP
zazYz}))n)*pbhZ_BE;3{Wo(T1&A2?{{p-E@Z>}=jF8^L64xj(?@!wu;^2f7QDaB&G
zWOTn6>DxJHQZ-CES0=BN606mQ)oQb!c*dV2)OL+AZaBv;f6j<>{Z)@&toRtII(cvs
zNFm_|dl8lO5S=SGuF+b&&xQgsz9FM{YHv*+0$N~bQcR)%PavL8qSMKYU6sYSl57l^
zEE<-O2F3CyK$DxeD1;&kiL!JUCv?X=j;vZ^1xFznyZ38y+6jF&dF-&zb!Nn4^-y@O
zCKy_@GITvuM+Y2SyG6DYH(q)bSFV_U_LLq&>=UZcpv%FlE2C^lh!8Tks=)LGrOhl0
zL(%m_W5GCVID*{){^B{RoU`@{XC_Rq-(mgq8Fqfm$(?&_^cgpbL!Lc6!%Y|B-DM23
z<zndMBRNIXAlfX+eDiPr$81kmtR8-Xb}gC2%{#Za`|8WQ`sQ0aZCWmNo(8aG$;&qj
ze&g4_$;Ti32$elmIl-us#pICS8&nc2jue8h3ymU~JV2^mihve4R7Ha_1^AxW1)@@v
zRmE%~%$c@ri7{cb;MYY_4D@Ir2iS=t`b6J$L@x(0)=^ir2=uaCPosr>jG3lu2NTA0
zDPT5=9O4xSP*GJSP8S3+W@v)OiEV@<cDvmWT4%(_RBKc@mHQy)J=}+a>2x;0abj><
z6*aSJ!rf~($eqUqkLh}R=kd`C6H#lLu9F>DS?(bvtuJA=5oTGh*T5wUZY0ByC9=Ek
zdzz-B>pjjIs;V9ek2oZYBXW<{4p%s`k_?=(s4==gVd#5L%W%(xKuA(Fb}<r~VaL*2
zLIh(Sm8-G3V^&YmCKG+&+I)hIN31tn$@Ef|s0?0NUb}afqs5%%a>?=0lIe6h7@`(v
z3tcY^#+VYMgi3?pG71Er(xpH%<w4gbG~3GwNr@cThd|DXt`}HV8AH?dI9Fn)bDp1{
zv)SziBh83Uk*abe#4UVvvEtF=Crqy0lJBK8m2=c?0Xd+x!d0@L8;gK^X5-}Z;)I}-
z5=$oJy18mHdbOC0uR)AxFP#s@$LEX9_}Ejrhlyr%w>UmNCWOFtyWJawj+RPrIem@S
z)OhY)6}yfi*q>Jj|FXv~sPjpT@0pK@uI~obtwyQgg2*Dq%cT7hbbRvTL^4*!F@r*>
zEcPcE6f!D~3ic5SOEKYW4TT~bP1X^VxC6`A%Y|v}_;1vJms^apvgr3i<<4Rgw2vfE
z5=Fo;&WfMbWffP99kDFcLP-12t{5Tu%(e@h930ao#p(Hq&==&gqBuMv7PV-kbPzVz
zWI|ajSRNd)-fT$c=a{<2Sy>jRoalVddc7fL#nd>$W{nGpXbSE>zhLeZl{H+y_d4f|
zr+d8OXCHmWgYyj=N~Wd|Gs}U_tgWVN8qj&zvpI5xuYB#7c=FMw{N8{209V%Z?TUM^
zyv93U`Z7;WpYw3zX?01b6bC0KkR6|X_%ZEz%e<IC%p6P)S?3a?6sUkx3hYR`R<wO2
zhd_~QQNfQaolXpBm(_Jted!pTdxXsx9TPXl*3ty%K@IPFk8(<~Q3Av=?Du@YK9hdD
zX2)(*v=)+gG+ujh!wQTyTFK{}sjC`m45k#u+;|O-hnQ~yB{M2J!Xz1u5lj8ynk~$w
z;4U``n*}$|U522@J)|z7Q)c37avz0!69lH|W5N$Qi(S+3<mpqUvu+S12I@(TRzlJ#
z>q?kk@_+V)m#pa9mbMYNttu;K(+LU@#+GF{On#QClCZhSWJ1k^qAS^~&si)EQ7O}`
z&WRxqvM@GPa@aiOaCStwxJg;Q%S*T3p|b+RpPYX}wSCI*!A;t;6`k6!I6mNDGGV=0
z)AfOb0avmMnV0X};r8uYl+z{C`Ba*Z)8|}VT!2<o!y;Y}1f~(m2vOkGC`ZEZ+9Zfg
z;B>@;aS;C&MIp=hAS9)#s@S$0$ePV|%e)Zrt3-$=y57?}Ls2-|Fq-4exOU?@^`xeD
zmAIUfaYkduNDi%~(HfeOtJ4fu3W*x3qwKIU$&4Dcfm&@u({xq3DL!m37M&L-9`^Tw
zbFz=Es*=fM%5*vr(v9!g?RK1<onOYi4|*_hf*7YrhCv|?*Sdr%j$bdsgnXSeLw^Hd
za(0|%tU-l9Bw-6DBGD*}HCfEMeZ?G~(A80F9Q9zt2nDv2+zmP4N8z6j9;4Ne!EVH2
zU0oVF3Jxlentb1}dl<cwAm_-oX^1IOl_jJsWP)L0vPvu_MuB0<8lz-+8M|ROHH2AN
zV=<$a2xy~pP{tO#{L1S*dA4HP8e*(b(UQsvJ)6?UmYgKxccL6`z5NBQU%yVX+fjrL
zYlQWee2-3&tQrdq+A?zvR~5u&!+g2q;?aE$7RSsN3l5J@*nWDS4?g~k^|oa?nSrVH
zj1}t~dT?w|g;0=<F}PvjZxY<OeG8>Bo$pC8;;NEu*K*qKI9<1-Vj|HFYsJMQ$oUb+
z(-|>3rqzU#<uT#tx>&Cehax_>Lr2Ya*VA?VUMLtsz!h%4$gd~$KpqH`b%Ey6cT6S@
zy)O{P7*d_6%95&{uv>5NZ9lMbpcoF*krX0!_jZ6o6zIYIV1PKr43ywKF($fBGKZ?F
zLTQ7KiS2GnN?G=1<}$cXYduhXFK>F3z-7~kI2Mn>)sgW;7BTF#ADr{@?UyNEx<O?X
zGh_Ma-FI29p0VpZ22C40MOjj2VSAmQuejK3B^0c0IZ-F_GmN-*j5~p`!AmH*t|$1o
zH?=fcV~ryC9%D+%q8KcKq~M{Ta0NwKVw~aR_y(`M@*1a4ro6ve;X}j@hfm){nss6^
zTk!ImudtZ?EB@A(ze#HqySCx_<4Cni+&H?!)14xu9ZoB%>4ZgH6SBm7^ifeqNWLc}
zX(Y6<Jb!jd=g-8!VK(D<IOy~h0@KxpSnXu?AXXMLRPPN{HTdl1jB^I(hH%4_nat*F
z+bu;|^8DFz>gz`UtX6A*!gGyvg}}DXP?Qzdt`)rY`Wtl0p~?zS6wZnkD2ls6Hewld
zl@O4x6rmT4YWEPBG-ib@_t4jPeT{o;V91=hJYUDpe4IwCwM-@x=JPp=#X|OGgB4TP
zb+m2A>FF8U?RNMX;`um!FKhqZXoS%eVSEkabNl?`Y8=F0_V|U0PhA!0LW9;6rKOsb
zvS>;P`W}@9dazdSZ-k%s#ICA!$Ijd+Lqs$?scWfJz^anHQ*Dg+5XmZu$>F$>R-;TU
zfs&3PMCpK)Mp!D!P?xe`6X-@L*UmT&4-UC?;}%sVdnRWbM2*f4tt}dbvxWF$C2<fi
zMy%me%7mD?{YOvudw=KeqE4<;RGA<B<RQ<`FX$TSyzaj7Iv>2dV{^J<Qdm0gdHbz*
zxPI$47s+81IF(t<CNL3t<YwDa>^w1RW~F1|YSc`GcDL`|CC8pW`0jW4_P4&p>u=t{
zn+iAC(#Fh87icRtX<0eyx<Xq++jS@{hojiyBuJ6#2U8B`HBB(2*_7}7`JYi8U#DIk
z6S87CDX6W&HBZpl^UhcAF?;JCUEgx=kN<+hul*XEFTQ~ql;P;Xi!DP&Lh?c%umcp9
zW5NzzVkr#NdZh&pOdY=44Qw1KWFtH_(mr#RfA5TZKINcTH%5ywChV8+3`=Rf_qI~7
zXC0aSjc*3MgupNXg_4Y<BpD@oC}bnH-S^$VT2go~qy}S+<UhrPu}0j2he>EMnaD)j
zc5HWB4rbryjhDW_%`d(topRUm?BhqAhv$SKk<}<eR&uW$93G%;fvaly-p1gA$Hzgq
zd&Rpf<}@&w#NK~4ol%t~Md3&>(6u|VD@5CpMJC|A$5=x>nFvd-DDW|Gba=w~(HZB@
zP60y<5uZKL8gw$mKC^nZW*0u<3+pX^^2a|Q)+L+mmhPvYaQOH!KluK;m=Gu<ID^PF
zoCG$mb`5^ha=17mhQxN=V6vq+Jwv-GW|+i|kB|0VXaCMVM9ZWV_t%|FIbACjK;*!5
zdPqtzolXcrhy<(khNkTXDso9m8J{vCWzdFhaE!LrQIsWpY{j`AMNyXAx^<f!hCU0-
zC;H)t$q8)@L1wT0a~;K<s22;TLJE#Em-#y5V}!MU7h*GHDlz{~F_S_i&~ejrv~AC`
zXQ#vW8jG90gyDztRsQ#K(vhP1@7vu-ahLmPa1j~a^S`dO4*a4?rrKhi#0*FBzGEV1
zoYG*^@J333%H#OEucr1&#tHi(6l26GLr!kMLI=vBaU=)YdU-?l3r4lKyl4D*@Iv*!
zx)@7Ya9`oM57yxkvduKZ*jZIARasFMB~F#pMZu)1nbcDjvjxtU#E?)~K@vJ)&Qh4B
z^@O_z*Ez@X`237pN7o>2xlk?Lu4PhB$uUt)rlh_>Y0Iq}Hz}(cpA5<wDx;{4qADvW
zOD4KtVgo)T>e8^BJLb~^Ds6BX+Rb2P?gKt#W|JE46B9DoRwU#wE-FJ=Dlmz)PuN0J
zl%kRGK9Rb}yaY?fY*up7uetyD36GyY=gw>)c18UfwcBwxbKG7`u)8(ZPSHtGYD<A*
z?&f>wDrHb94_X?K)tecd17uO!_W`Bip0+8JWQapbB*CIaqCtkxdAcqDN~RMs(|QUj
zi9Ot4r8F*5Lk#=LN*le0aub*KALF%`hdsi`ieuDs#mH{g6XW1vH%u*}Fc;X0aq+Fk
z5H~3b<SYQLmOV`2%Kc8M*)`m{So7Yy@AGum(Qa0#&~yIuoEVH~ROFPP^*v1|P|xDv
z5C&<%c<7G!`<U)ITIR?CK2z3H7PI-_Vqlq6HM8lA=sjWZ5-SUNZq{0{HyZCD6w~RP
zgXIxpjCEaA#7_1Dr7}d{5?k1vM}iLop9m_5X>C@hY_K?z?+Kj;YcNU=Hb<Jum6T2n
z(&>CjQB~kdHcd-CsW>`1;o#taqAK?+E<l_L?63n!kh2Dx@qLHy#SpdcI#Ti|0wFBX
zMlqX}<UFTsk613393CFA-mLJkqiwd>nAogk()B^Ce;D59s;cQy<oWY6Hg>}O2ahNZ
zjxm+&D4nx76jp&NEL}>&E=mq)p@(ZmX6^phcrALr|GIG>K2oK}>o?~SJ{AJa)xFTw
ziA0X&D+TTfA#42o=O0)1Jb&5V=IYP73jO_}k6)<x7}zVsMHnTil4j44$b0Xts9+p9
zKhNM8MN|7xGl@ywczB?V+}vu=izuP$#+byj&E}z7z8q)A0`&8iZ{rR~k%>W;)jA0T
zv)k@iuQucag)0SC!%*64C_GDUUcbfB;R&THF+;qtMh>9dM^cEiPtRa!dGOwcc(bG^
z5184KvNkNIGfb3>kj<{)<i;WM`JBnYl36h&7Zr=T<i^pAd8M%^e4p7hiA|fyDl(mD
z4(E=-K)dPbcW``sB>rJBaP!(+*kU8?9k2$pl?l)|NQZ^k-fQT+Vz-MV9eCrV1E#fN
z-SvWvTf=I%<GSy$Iigcy#*ULzunZA@rg{AA5h23-?2J!-_91KS@G%QTJWB?>51tU>
zes4DFdNP6w$04FXLg$>osl(#`g`0CAE3{opN{Nf}3)-&bV0l1U*kNK4Yj@||elLU)
z?=Ckadny@$$ooW2|3XmLc#w}aHEGbdNVJ38xG~A=87HNplyuRU<a)GLi}2L@!BxRf
z*A;bL55ic(Zo6f5zT&&z`7ZDM=qKousjOmJDyEf18;N61fDe(?W=lwc<-r1xOkV0O
z#vK(fz$xwixqF|OA@tfYw9HvY;VdzDLJ+2=d@gw}gxqd-1D#ppY{B*GH@SK97C-#{
zzoi^BDk1b36m>Phl{KbN!h{bkP3-BKR)pXwV2s?KUE9#8jMD}c#Rse~1yu>>+lIwW
ziz_OkD^S)E6S9A*Yy5DC?`g+IQcCkc@s)dDkdChFgf%#r5(}HM<7io9l|reS`Rp#;
zx4y-Cz2>v~pCE1#mfp8?y}>F)*L#U+uvVN92;x^)Rb-!8tyk<WHkiqj<-q~hj!%T?
z7d@#<BrW3QJ@EwgXVg{LDrafr&Cne08EEO!baXPAFq_S&>w19T5+NkE+YRgWdjGj=
zt@rTS)jiOQ#Ufw%JDGxh-kJIDdt9BiUi`B({^qYPCj5fMC-@G!L>e-Egvuyc8xe)g
z3Oy8`U@q+-GHQ>F2Oow`)a<E;+E`SUL`tp2s8jISM9z^!Ao|e;JYkJccR`VKdg0A|
zk!L=#G6s{!A%`A7DLZuL67ppRD~k{Uoo{LLmW$KpJTZnRj~+0a%$ZaZw6+wwpe$+>
z5-<GHdc`L{`U#Ko2c%B(;OSFTeTXS0%;pQOFPAK)Q&y)N9Gc^ULsU*o>xxE;70Lb0
z3f*0&s|zY4BYNAx>Y``YuW6bKLf`WBuYHk|!x^(#&C4&}8(s^%|K2B@U#!vE;(f$9
zOFf-noxGoAWf^yr+G<>(F;?MJBK1Ajj^;eN{}IM$CPhVzf%E6ju(O(aVL2{5$2u_$
z8NY>HBhd%D^@i=)nvdT9gk96K>m-ZFl_jny23<@q#E<cFSB5<3#bOXvmB12aZyPf1
zNydppY}*H)wxCH~lRShylWEL$F#|co4IQK<Cryf?aL-ZU!$>#)v0)MBTa1Zh6lu5)
zQcC!?CG;JrOD6<k9433f8hqQ*^*z=Ya_Ph}3V{lpHB?1GY3%?x20|}Jmd&<dlS-^o
ztot45u%@yzc8xI7Vvr!b7~_x~l_eOk_w>D&(63Hlqh{Bz*@zv!b&k5K$XRo-+Tyg~
zjW^z4KA-Iyz7dJgdSs*SxApt`NU4kuA_TvA^A1NxM{F*hLFlNQA?A*p8YH{kfXW!1
zD6OGsf=nU4W3xKL#()}N@ffpYZ^ev{p5O;K$Xbq$Pq^@jYBq=LI6XZj#z-~9Oe>`*
z1~^qK2H!pQnlNh&RaMaU&=2>$Lh<CubE@h((@8;0&^3YQ&!6+~(L>J8&)GFQ3aiQ4
z(Doe*EAAT3isnTY|At-DQj`UuX?Xhd8QTblH*QhaV%AF367r}NahPDz)k!BM!j+vj
z3~zfDqLbD_dmr_RMN#Zof8*IEj^!&Z&eyc9Bm}28gw72v6FH|nL_AJhvQL+DaaivO
zIVH-okmvZ~F6#4-(a`lO{Xr=}U#7|ObN=1Wd;DS|Q}MA7HLh0JVF?yP6oXGC?gwMu
z10%m)iQm(RforWvS}qDbEQGBo$uThWv8XI@vsOWuJlZ5elE05xlLjW8WP^k~n?#n(
zBcFblc$7jZSvsdUOkdhE8j+@92_8}}6mKi-RMy~g<f1uc-K{Xn4s>l1{W;?>%JLh#
z6^}1IqnXT!U1F{?o6~0)H^cOY_|r4We#K15vh(TvPxu%A%|E4HUgM&RT<koZ57cFi
zGnUd66h(y(iq(3<*~N3TP8b#WFaOK`9l!HG|0c8gfW>UiZr5S8<-33UUH;(r|B#E-
znznDT)`=RvkerUPEU8MNdFMQc&L$PpNl8^^oUKt&p+d*?`Go7bWwL(AVR^tZR=7TZ
zuZZ5DbEa|yA&C=!ttKSxh$c%wTJ-2>C=fSoV=pb1(f~6sgoufM=%t8GYb{!`tcXd1
z$<kn@qf;XIKvmb2b<L!nfXevL4bYf4CzA(uqLJ?rlVWO1jFHego!G}c<ebr_-_y5K
ziexgzT5^y^$(6-GTo^JD6#F=d(T!LejV&x#u^~}O2p68rw(Ic26r_~c0A&TcuBR<B
z2G}(Xs))=NHO;PJwwM!9lrf{cXLWwgg9i`U`N+j;!-EI+dHC=VtJMWf)39ExY1)Ra
z?U+ob9334pn@l;JAMpN9f68{d=B+p1q^?UQ6LF#tt&B8iPK#cz_d;@LyMVQlIg_Je
zad6CQufNS-{P~)uYtSlTii+e4Hl3#rJ7P?%FIv1ytWLMIji>JerxzO*bxG(utj;)J
zNKl_~gqUbUV0v_eqg%JRcKa^j`HFgafW~n8>=dO2yCfxO!+zWu4s53f*i^|L7bkD9
z?7u)6&C{o6)b$LmJDxu~=h6L#{Of=Ddpy4Xh=bVz-|Z-@?D)Ff245Gbfo2~`?Ajg1
zc`i2RtalC872LZ0GHyD@=R{#OzTFX(_z%kQEKqPF4?E&0p^JminUU#+FfRKL_d3Xt
z?mYgtb8h&$QL=C;(pO3g-F%+~bT!sM$)1xd`y3?&Nq>&+oyYrg3{d=e4>h>HU*X>=
zKr6TZeNp@k5Dr}Bu!R;TQp{NAu!9{wFf5J7e1{it$30{>FoaZ+->W1Lj5gwz<g6x|
z1gQmNyw0%^i;LW2lqkrBO#+b7c^oCv(DjPIHO4@UUKlEc#Tc2U<WorU299l0lFSM{
z_)e*zH4q3fVs#>_Kyp&ylg@b6lT)V8J!k1uyswmDY7(W1>@J?-(wbtk!8PZ!U5Cq%
zL(jWE_&!+|L|u|iO~iq%0E;pXtHsjX52lzoc_0#fVmgHoV7mon9IH*w8*jeJzxlo2
zXS>_dr9i*ilFp)NQ?y2BNIr<cpmoDVqcBc>Z8zrxpEz)q8<nBBc)-cgoFku^g&Bo$
zfTB+dn=;-<5eI8S(1z$73XMq-6&22Daf!&07|pOG*Qnuu9~a_E8LV}~^m0kVA1CE;
zw<nfJfj)F3mG;Dj3^c8$ZCi239u}ZVDf#^X)wr^vDD@D1Y^A^qKg;`Kf1U1jJG}S%
z{bY1nvWo`zZ!{Sl3lv(Bl6Z`@-U}2oQfCLV$~Y9F6nn8HX*)s&oMY4Wa;<7blOozV
z%Ce-}Y+2M(=C-8Nmi74s+weYr_U_L(J3Ht8g9mH|Z4a=gANSs~>pe~1qqCx0Z<y8-
z{`?1j%3?m}_~eK$eBo`bT|43U_=tnUCClYf3{RaSXU(LZlJJC(K}CwPK%sc)wYT`-
zgP-x_$zyz>@kyhq$YinL_T5*QOuoY}ef8Ifg=nDOKKwDq=cj~k{3=f$+~<ei|3kVQ
zBs(hvj0$+4XnW7SyDxF~mDf>4#mS8s+il0u(Ge#{M|dBIL({0TB83P;JcTi!oFYU`
z*Cq**3sRVJRI-DLk+zFGKV6a0oS$#lHG$V&euF2U-e(P*E2&v2iKm;&5ponFQx<23
z5PH^|HP6n@DQ~{QwVQXyTAWyN7V&LVA`9tyPwIO?Z{Uin2P8z~o0I5`KF=zH7e%OW
zX)!fIYombQT7%XGTqY$u#7d0cMeXkerSc$?xtwaQicrTLpPb;AP6}7qcG4gY!~|YM
z*RD=-e_hVOFGjU?lC6__!IUiqYYo<xf`R8t6}u}hor}|8)h*1nOVWXJLbvUGAS6*7
zS5r$encPGIEu}3mDw1`^6cdbtE(UbY408AVn<?<#<(oD-9JnMpAn(I)7+IVZDzbc#
zcE8{g<u2@r9{F;t^ik~T<?xF5B&?`RCZwL2ECx%V70xMa?TGz`&~0&pIbokN+9-Bi
z!d41vE4t|AhV%h#D|AjM`hgu|D64|53%H`-$!g6H-(L%ak>U^#7C63hlljRJA3XSs
zo`_E(=rmemm55DkQPH$5>s^N#F67{Qio%fmj=HSat)FrFwCBzncPYr&>>+qL=z~Ya
z0-Y0WyJNj+>AIH9cFW3GIzPZSInpIh>J_#qD2f{2d(v>&kDtw@E+_6U@X<VR{7kQU
zT#c~Hs6{BtlFeqzdcCG;n!N~aWCe{;0d-vu#FLEgJb8ZuUdcxh%Vo5-#OUR9AjNSA
zW+15{iaY~K4Y3XiWh{lUC}YS2j69}ojvAm*m0wj67K-KKfM0p@3tYc;mrSOvHBR+R
zZDe)-Gjvw6|45m(?Rom_IsN$s4<A0_$<wE-*XtJ!g?^AZjQ5pu1*LU3C9q%H2Oj?H
zeYzk1kbm`i{|2oUCnv|;zI}@uH*fIj%dc?f&K-^pPk8mUH#s;uLHkU?;`<&<!O^vw
z+<oOW-v8jI^bum)Q-#3X72FsYqc?8cgld8hk@=&COheDzyZ7)-!@dp844;#@oAfbq
zaD2pUxx~#EgzPv!zfVqy#nAzq^_C}3#7fG!0%PWARpTT@W!M?#VV5kot@mnJ(p#$1
zQ5NOkat$xtyNkMe2NyHn`{O^vc0=}2w<h#GwHcx^a>4>CMM`Vgc?L;GqOK<#9-YvK
zfYphdgE+BhL#1RdPKe|b_7lk^Q!VZP_o|5V=j~F)YyT=U^KyrN>E1Bzb8=Qvc=!9A
zQIwhoIm_kGxcVIaeLM53ao>9v|9m-RU42}=Z~r=vUkrAOVzf6}dyFibu|vJ|BGQO+
zttO=*7e@=^&~*`Q)ZiGMP+AR3OalcNHG^K!t~zqelA;(LU2ZWhVTz3Dg&Hdd)#V#0
z@eBX%n~?^?$FTz~%94Sr%|53kvDad#q2j0k-xQm8IhfRNg9jghgt4+)C<;SwKt)-6
zs|;0H(l&|0X$}r%>|#&yo^9$mym^PC>$jLJj+xAlQMN|w0#nquYC>6zdKFp18?8_#
zlbq(yfBY_0RnWE#)9I9V-g%o{i2U7u@H@Qvg|~Tj{!GkNgJ`1!_OcYt(YL~Is;UZd
zX1iYT?6Xh!>0kUgq1{mqak|Z}!IwtTE<=a58B4?}iJ+L)HOWW;Sk7kLx^)YN#7~tY
z8i68=b)*b^@Auctc%zM7)m6)-tJ;REb|m9xIT8fMB6W0h%=POx_EFNn3o7i9+#(Lp
zMwoa?Zpx8bs<ojkD~iGrQ^NPMpV+f1$yff(xKtHq1i`@d0QX8lCj+fk?IB+op^X*d
z1qyEe`Cssxzy2G1^{ua9tj3r?*Pi1yXZ+!lr=&iTlpbOs6px=g<9V~=!NW(ax4ZrO
z5xn1vZ}Z@7)(qKGRZ&6AT&%Y^tLgd<qYZvY=WQN5=E?Ij_`#2;>x%ho&T?_c-McSy
z>-IhF-g|``*KaYY#1wT}TW;LBM?F7aw>qOP9esc4n<xdppvwyD3MP{&=HiU9s+deB
zvad^n+eL_^6vUu5Wsa|16DFUr6uRX2_=JnqipNhLu~~14L$G66CY-ZOrxjIghm=x<
z)nfNErqo(v_Zp^gztuF&5S*AvF>wFULsVg?W)rGSP3Sk&Wkp3G_bm=ZZ3{|mB^e!!
zD{F4vx`TC=qAUb*D;+67Op0s_Wl@P$k(PAg(XM8^=6AcDxMK|W=XlNjya{N$Cw^XW
z+6!p+du$Q-kHts6bSwxNNH6T4#_xUk{jN+wLqYp_UTx#oeg64ge$T&w;sbJclx4Yp
zGp_pmTspN5(}IE&6(LGSjXbwF=pz7?P+ANX(HUhGQ&VE1p|0$nT5UtZ#E2>a*&sPW
zM$+~e7&^jjkfJw!SbI!Vi{nBX2nNc_5`6#W=J5>{s0J+s$S`){!(nJHX{Zupm1WPW
z)L^1I9A?9J(<wlXIIEaWOX{7Y&4K6xZ+`J@zWmi+rkpLY<(yrY@KIBgHIu238`NNd
z-uFE*1bpb|e9!%d4`r94#i4oI_88}w%x5eP7M$EVxx%!nsq1=}6e<yZdQS*q5%Tom
z17;?3_UJRt9*SU-ED(<Gdt%=ay(5`K>Kj6<$yLc>am?|JC5Ou;>a|xneES_@9&Fc>
zN9BmqO3XgfV61qB%6auB+P|mBdpR!q$H`|bOjq5}$HvF|%dYb3-+dl7xP&o;{`YxS
zkFfFv(@aK9#E=`KhR--Ffs^x?B{a-XLuj4QWbNhdD39H)bj;R{-K!iF&I$|9I)jf9
zV+-cTH_)!4NjpL|*rFz(`Q-k69-N)C>mxo&rj-;$ab>?s+x5NBv$wiKBZ-@!#gI2A
zVpQVbt`+@&La%o%rx#nwC+EET-iJ6>Gnp;<@|VBLfBKL95y3|$^9k2(-R1RnzRVx~
z{=cFqtNoID+~pbTm`<iJl7LVYWyRuP$yJtCFapi{n5d>R=F0<OhV8DSsw?VhM)HBv
z(`T$!E6TDaXK|nqM&XncYl=defRrHSJVd1^Fq*;@`#rNYmckXZO~=CrXLv91Cd-o}
zo;=!7)+L+sOiq!-d`ebuIA2m3OQ|$DM7ln)YdzO*-XiY$eb8GB4aP{avnwVP(>YZ=
zVLqw$yR~uBADgGH>-J($kj8#A`}}zf;qktCkt+QnK>)NMG5IQaU?k)KYFJJDoC#~Z
zPhMc)U1|QV(zM6l|GZzHHx&5G9={muW(~#+_@R<iz_P6OquJ#f6hH}~Kxu^2pQR!i
zTMTQLOYKD@3I%w0M%rddt8vEBb}bbKrz!%0n7bkPDWV6**34LK6)~FO4I3tvp)iCP
z7|ViWp2)Pbht-h(O<AFbLnWy+bP<SfEezApDDEOd@B@V4L0fdz=$s&dK~oroHyL9R
zvNH_BbQC2Uq$-&$4_F>u=iul%)nrCpPq2ffhCIJa7>pW6%Zm^L))>yuFX)?&YbS@8
zySI7$)t8w}Dq(KrgjSL#qXFLw<kxqXG|c9;m-*?u<l`Uz8N272zJb1L*lsnA33L-h
z=nX+xd}F|XO>+vXIath@)D?L)MIRi3ErunkG#fhSA@A$5a~rRftE2l!yT4j&uEzb2
zOV_KA!w9ii8(zSe$FILy%r4;oVKc=60yFy3cE=j|)N>vjxwV#1E;YV}@%k9t4Nx%J
zGg*-v$eL5qRx?@fcQTOCD(^K%qCYa?BW7h6X^IJbmnce&o6Jd#C)<J^&1%<j-tI_f
zimJpRhuQdk<zYXmwdg_g@cONBqLs^%VWe$=4<2m_3Re%4hoqZorD*!VX19@?n;dD*
zH@x@JCoGPy)9yC(IWk!u@@v2KE&k|xf5f);%m(R#E6Rc9TN1UVkD`GZb7<uAQhTCM
zR$6H9CUbCdoh~LSXF1<&sV580&d!-m3hvy#gR=$e^_IT(<m_?I()azKb&)eEC2&p=
zOVGv+RC+O*70)+usMzgxtX69#(~5&@$6PzV!Q;oDi7R!pqarhL1$SS&$(LVyjoa5w
zh)qYk-LY<WJbiY?wQDyC?|*<cnn_h~usmW?7F2bCaW&WqWemRE((JaZRx7sK?LZVr
z`$>Fcu#MO3_?dnl`uaQ!eHC^aC!6tH`+3Dj{;izZDPQh+uKtXxlhx0gfZ!69xLP2`
zf_N2%Gx~3&_~?E2Hg<GE)fA|0e8GCix)C0}bc7@0O$x*q2sRw-$OIu-_tKqdH|U0P
z!l=PkJV|n<Qf62Jr=bX5z8Ug-KKCZ3AsB2(EVM&XUs8GBh3-F4SuqX^LSgRt7^EZ4
zYKWSZRw5&i*nTOVGUPM>dPquYf*giuXd^n1${8v&cdlLM<i;H^CG*8XtVgt>2CR}p
z%xKv^^?fFUOo&MiP37>NXM#m(SULr^q$&%D(EG?@UWoCfP86<yN(xciXsi`#yIGVx
zd#_;X6po0i9EFa17GMmLKpF_-SVHhP1%w`_#Zsg-V%d@hs8S3MHKahydAK1ii_NIH
zxr%qg<@7Qd&t9FB>bibm+8!SQ?dBI?yBGF%BRmxX-Un$&u3V2YDarS`>bfC^nIz_3
zu_$DmcXAq`*-O!CQW=%x_j*49ONV;(HKx(2VIbWIK__CYBc#lYJ1_D6^A(%Mv)Z(@
zKH_60Yw>ZrYDzi2*2_aF?hk|jYeAI0oD~U~=F*VjnGkE|5EIrGvalcN)=E)N3%P%k
zrl@P0&WpdCm7P|Xd*;W-Wam&(rXHouK212rM2eZhmBKDf0u9O6X<|r(1Q8|orP3^q
zkEv&K%E^>)agJ7rq9|FfS5&2AHk;A5qG58bV7uM2+fC@Y8M9muMOWTxqaox#iXNju
zJImp6MqOK0tB%51s&awSnxm6L>?e+xBUM$Rm7+41Ye$EC@vXP`;%l!_=FFz;X}zcS
ziEZQg(p$IS{SU}WVG3cDN&ysVexJNRi`|C4YiXK>G1`2zn7RrPU)`x(B?5eYLis!d
zcGWxc5*Ahi>EJ5tmWN+yZ#{MO`uRNZ;OgIf{yAJNgjbuNtHtYYZ1NfLJaGw@(c#j8
zb)0S-8O109OHmkTmm_9+!A3%BNZt=Ox$KFI65**WH8Bg7Rb`8b!<(J};!rr)Kd14g
z+rO#9+pe@>EXV?@h;4?oju<ZOevC1=LX%UV_i-5YJ#D+8t|nAfJuIe&#K7T0aK;jh
z#srNWC?`2YOg6-B*uUtECXxHdlaD^;Cx7<mtnaNkIJ!=E(a`yb)=nb3t%Dr#p%=GX
zCB-lJNZ0wjPNnaAX0sWGzxs7nkDv3?M-MSpQ`Z$FPgRyI<`b%-;N)n*$w|#8_n+{5
zwPLs3qE+VcM?dAuZ@vQCQ$F~~Ur^5{w9S^rcrGpii$dXQN6wOaL<&MH4IK(_T2op_
z(&FMfyyNo!QzpuXv%e_!3jHh4!slUzapeE}@^bvz@n#;GIaeWzt6wt~DmnBe?pSs@
z!|pByoUvk7swHkic0IzF8aOQb{&HxghTjPkH~#7P%10<<WEsH-wT%m4z5i-q<yrci
zK|A`;bMwwi{NUgI8Bgy&<k6!?G<`rBp|+cGqs|Ds<QHJetFTNOT3;z{a?PWZ+&Wt1
z3X)EF$%+CY02L+=sj``i^@inqig6J79&Ihlqhl@}d?t3CIkVmFWCAJ+f-MFyuY6x)
z3w)4bS89<d6lG0Tmb!Ml^2QsOsv>4XQB^!UZMbpcm`^@D=l+BHf>5CLop;|y&d$!T
zRx_R5q9`q80d1!+2AmTqg)t@8Dt24X>FGIrmsl>QOeRyb&FC}F#l_jsoknWsc>UgO
ze(Tr2!HvZnqZG0Cltn>Ek+|EDl%|@@gk7jLO}pdSvj&YID;Y;r4$v4KNHM%{eFHDR
z!k@p!uO9B7->H259#QJ*WFss!t@OTGx-32*?A}WYoU6s=^95)8exHZouKxMw$pu$^
z<+{I#;xoQcR}07JUG@1lGAD_&l_vxc2He{Tl$B%xGRESJB_rmZD8wj{OU10<ltRl^
zS&N`d3424{TM@m+nD*W6zSER<ND5SzPE3l8cxP!N9}8NO5m+Dwgprfgu)wq!E!1G|
z6Dg^Eu$&r3+ZX_=C1yBhRFWb~jKtoHU~qV*NAo#nPoDC@yFbQe!`ag_lrE?zbIP*j
z=-`Nh8`n9QA7YHA@jb4raJC>N>B>i|{I+ZHp~uF^vMLC(Ih)On^G8n!eTUPTkG3oB
z-nvdaUb6kH;L}e&ptPFhY)U<waBJZ>{q&5uJxBSL!c3S{HO?8C{6e4$nx^w)+mV9^
zCELxKu4yEo%zLs`q&SjIGDZyoKrScK@$vI6?(z3m*+JtC@$+!yangm$LwB68J`cUg
z_s$ZfkmC?D;AxsIAw;^aXSrN3pU;PXmw+Unr&6CQKI|u75X&VsT(@!mRr`(5@Vtg4
zC~E)xMdPGMDhpJUBgU4T+<pmDPWkY&M{GA6lq)5~Et+9p1xk+xg8aH{z{dA!`1x6!
zvgNBrODdt~X5=~=i=!Rx-?A74!o&OeC*S`*-}<d@QW+}{WX^o~t6%0{fAWzK4r1JE
zTlOx_S|%=(h=WrCt*oRLYfILeh@zg)nH?N}Hq^5z2fD^LvX{Dc?K-H;>S9fd^0{}N
z$65)IJioZ$WP3<a&af5=2hKtWmcEblT~E_Qd3`BzcsOS<F9`vx%e?pDDb03^Rf^I%
zP7V+G(ih(0_Vw$uXQ!O6dS-=)fJ2O2Y#J`sEv~GG;-f^rGmw;2zP#SEdU3aTb@IJB
zZG8S9|H~S(&;R$8t8sXYP0jf4#uHHBWIlI4{k;GF{Nt+o`(O5auA&tCIFG;L@rw}-
z2Clf)>XP*}be1s%IjpsoSUwmQua|V|oS|zy&W&BF2&6Pd;0NM9UYX$#ib<@|g@K+K
zH6KGSH#{%BCnsDTXXUT12EPr*MV3iTVqkl5wjNk(K15=aL`G}u&^2nxY9^W<pO7?%
z;i5v)6}v77%`+)<HexcU6T4kQj#(TF+D@jv*y6j6NnP>A8?S*Wv94w|Ut&x_U6xF$
z0*%L+5>u4;nCZHP7$7Gx$n-<>_j<kJQ!Tdm)>^J#yUuJfp>&$QSySJ>#&S}ym^gm&
z!|$^?d&(Ezd7U?3zD_-xpwpYY|L*tMJbTEzvi$PbzQTh~Kc?GUP+*8jBE{n{&2-I<
z!Z^BSN7L*`eNW$Y^le8Ne94T?gT0SMyW)ju=jsjk!rp?>x9W0Qxk@s)3Q>%ovn+DM
zzSxY|`S|msQ|@V+M)m|TiS3V%_z+mHE;u_o$6Cjizw#vx4i~Z$BVvu)Pcd)_V2sc0
z^6(g7+AIVGIrv{Z7RDA;&V!+A8SbeVA{0V|ND8XDVsUs4A3+&M%5m7C$h4C8o22pi
zsC`nV6vh7^ZGZM_Ns{OJVIMQ|<?MHfxUuEFRc2LJR#sQ_Ha%0_Jv}p=5h+rnpa+gb
zfCK~q5(Nn&pnpV90tG!1B*g|rVnohHji#rGUZ$&Sx@*s>$~`kOGqxLdJBu%7riY)I
zyC1)PZbS@GRuy^9J-(US^4ovkU+k#>#|9_>smO*F|C0&?$%VWqM98a35#GGBRJFs_
zr~K)k{4u}w>%S`dRx3<uc=fec`R;rsehMS-GfK;@uwKmN%1Fc?TEgD*(Wt3%>%F3|
zmdR*LlI3hlscAGG&zOy2b?TT*=DhaWgkn?h=+POE9z8;-g7tcd_nPJMg5%?37JDP6
z(-f^lMO8XjSL~Ygx?nzA@Z59Pf@?s6UrtWf96x%<Y&_wS(j3g@=!1RA^_mag`vukV
zf}_QPX`bPgBnF?Ho>P>SVC;nvp3J6H#hFBSdZZ3g@Dmy(CEB~prSG`!YxZ%icaQab
zK%FzU*JoEkYOd>O-AcCy2U^$s7N6C{$~27`bz2zhHVN#x93eN<o-x$AM%!+wTHA&Y
zS{Gh2qK~wye3At7i1PtZG6oySF0Cs|9qY?G55_lkb?-gih5>d6Xq@(VWw*GzQcB7_
zXdO@wLUNQxq&uxh5;474<<L&#l@?khnB795v!W<jyE0UG97&#v4^oZSO&xhnC?sYQ
zv`SD;3X}QTqnyT764)T{L6$7b*(@toWzApw<zMjrr=Q_fLhTfj$qa=d(ME`dvrMKF
z3Tq|cKxHIJjv|o+Krdro)dkb(81F3?7Z<G8CD*TAqgb8u?QedS|K$Jlzr$}<oIm=U
zi<5i!;+*>85yjb<FFtw9YO~?*{_by+R2%-A|K=a?8{hsm?=7Z$_s{->s;()mXSJz#
zd~(6MD5!wBS0vLtVWMe<8c(RSgavyf>d0vEd)P)2+g`AJi7zshXkLk-(p?QH-lyy6
zjJ15qvgGvil!p%=Hpf@1)t1fOq)8L>ptWJMS@Y$WUt+Z3D{sDr)==9*o}*Rkzv{K9
zb%>-(BlXG)HLH?V3nG~Ce7pfMin93^60mEp$;MM^uRtf%Wku;MD%DZk5+drhrB|g?
z6Ut3shJv~&zpL0^Mav;Y{Xr=a05uXwqyeK8RaJ0&{1E4AoHeA0LK!$byvEUu8>}Cl
zK+ul8_oQitRZ>4osfgrkq)6awhFvbj@8J=@Q<gE`+vnk?z@9Ey>`hs%EX%Vs=WLkG
zGZyn1FTFJ8#*ITBKVEWja!gUIdHC=?zxc&3aK7S|SD$A-%g~|7<Y<zBnvjn&#-lv=
zI6a$<B(@r(n2jgw&ll9UZc&~*z<9;OFYohcbIwF7Mz6d~o=9NIrnIcq1xGh-Gn*|S
z_-LxSlr`&JsI02-A!p3GikJ%0U;lkez?1gJQ0o(ZALo+GOoX7FZENG(mVxfGx_#^m
zjeRp>*XMIpfBWs(u-)n!W!gF@&N;HIg?;FDFghM%*18J02w5g&dqn~hDa6@Y6h;#=
z!fPo-pp+65S?y>{>khmL*ia!PS`lA9pe7`te;xBf)YMQ;B5qU?jq`}rEkT=9kPhma
z+DhfrB+10Y6%tCN3a5ggeTdejNK}R*!PS<!tObUs62QQ^swvhb)5(I#D5vzNkbp}R
z_a8qLGlVxBpPb;W6KAc~xVj?CGx4J81bl*Xnz~Y0OM~I2@Hk(il*6PcNjl=AyPtA>
zf6nV~zQMJl14e3}fAUv<!pZ3)4yP$+ClC19Pabk|c7n+#y!zT3{Jp>Zd#umTc;WU9
zrn%vRpZ$bP!=2}z=jd=DrGZ!HoSmIgdr%{(&N(-dF9p%;8dQB3B~dZ4io_R+@~I$*
zH~tZb!b<elMDtq=9*E)65l~Y!Bt=<D;XNsgb<BeY4_U2N%^G#iHO11#YJ|r(=obR7
zZ@E}5S+AE=RRP|SqzQzAh#j8NYCiZ*)_T~mJ*iR%ec#mKn|w0wJwBKzY9;qhOoGt~
z*?3A(Tk%IE2ICc#-v(Wc_l3*(8s2*p4y~Mk;NsidI)cXI<Cu#!L=Y;As=$|yG#^vd
zHFs{`W;z-(9%neap{_Pe#)1IvN8_+o#Rr*~gj6TkDiEj;kdfvYSw7<A(IaM~ncylq
zk20wgdh}4*nv3O%Vky>+i;D}+FDgz?&N(_d<mhO^Vv)00<UD?SMw%tmb;0@h1<Q-I
zFdrp3geFm>(;PR>QAohfW?8UVmqfdTMzNSqnZEEmfBDfb(9W|uzrdcKknPRL^9+o^
z0w&M7SQm`;uajh18{KQ2xH+Rq)djnZ3J^=G^}!9z+mOGf%Q$wK;^ICMAtaid<NR-j
zL_eu@>Gsg?@jl&1)uDcNg~q38&xUHPB&0Ngf5x)4#wd(V0>f}9C9v>{$dDu-$9qWv
zt12OsVjB!#jmn9WGHigDU}=<*;&m=~*R;ZEuqr_0`w(sCwbVet2cHlDQC12-C`7BV
zHdN%JHC-8#HpS$;_d$~phFKQ1Xmxdf52_`UFiUfJNG%B2b2VkPc2HWBb&^X{I%+45
z+p~)`d(XeX^RK_oXaB=L<HoI9jHgrf4z7`>W3ZZ0l9TF$EFYut6qOmqvl)3hCLfK_
zI;E}^Rqd%N7tBqygmz=`UUII^ml9?3cmCGz@;86?*D>Q1kLRVgzD7~4sLqZ#x!9oR
zQ(k=GHKx-E-qobG;@|nZf0OG|!?npZ{<r_@|0T{@7V`<?QI64y;^K_chxd7Ka!R?n
zkg$4fDXJx86Kt?8E8mtfwgKfjYRkq*E{q8|Sanr5j{35yWSVsg0=wC4D2k%VkGQzF
z2%)w$RasJ&QXH$cmZ~f$i#6U`R?BmYNhpef*=)*Y?b+X(^NaW2<LvAVqcwM)d!EgD
zjY6@!xZv#MG;BnP46?SOi7xYiQXwBmH)zEX&A@w4Q|gyt!UCOqxJ08fjWS|3Gv102
zrZm&NF(rcU<+P$I${0QhAz2!Nu%%S+?}P}aj!@-wbU?R=;?YDMnZ40LICKJoRMn;U
zWW6U(6aMXg=YNh#b$IU#S2<SY2Bid_&3KJ3N-=qq1um)Oos{4}It@vdF`3UfF$uNu
zs3gI;ipAbs=(hIMu3}TKNk<dLX3S(_P*hm!IX+o)u`F3ECYU5+|KK{4*__R0#b&eO
z<g`HR4OyO&=aK_wYfD{LY&HRy)j5pPOeRy5g8jvU^5{DDB0)LJG&QWxH#~Z9pHxW{
z-=mWYlu0N}%A>MmVZa*62UOmYC4z8|4FZJ{)@GDPCAwKVyBz$3ban^v-eyjbF?7v(
z;r<P4Z7prlP#ctD!k{<zj`!UDyZc>W$D&ECucZlwCNz<rPtl$&+Rd28`5Kd_gi(Mt
zVvds6GSn`1vK<gF0c=sz$9s{8CHKJMD~ZY1N|Pq394l+YjV~`GOi5GQpoIi2%$QJ!
z&UY>0!9?Swz?3!#u69(V#aqa-5s6CB-e5}$#*-SU8d}vAn@x%HibUtq@5*5-i_wNu
z=VSpaHBl+5P{Kb;PqA8~Dk;=d6jCA6quAfS#`Dj;M6upT{?BH`@>Fwhd&<i%y~M#}
zk5s2rX#6N+eB&DP!vprl`|Qv6$wp%;>$zALoSk2=*_7C-z~D$zO`;TOqDeEwbTZ*<
zUwfH6pWuz)x4rV_*BIwHKm3b7<?aW+08(Cl`5KE^#=S4^^2MhgG0svhC^)*d&*IKC
z0p}zTRizoHhQrB->FhDbn_5h3uEta0U4c^ud_Xu>+YxInAf6`yHCiI)JXKXTQdV7;
zSSw-An_|Oez2WTaESMA56h*;$y>8ZIT~`ub-^nwuwgT@Y7b;1Db~DD>l8cLT{_<b_
zCFkeo>@60&_`-{%Ny=uu=Ce;e!&}E>GGRKMVv>|3$peO=f#_?Bf=yiq!ATgd4>mY;
z$xPQ`Vnyjypr#Be6#`<sM>~%~9ORREPCg&8UavxoKp;FJq**Jr#x|xeM8*@fG${xs
zKH6aSYxX2w1q2Kc0=-hywIWH!D5EIrlEwalH^2H7%5sBqsiZNgl+~&v&&Es_nn{}B
zR|T%FQDp(7kQ*#k0)tY+WIAO!U!W30%udnTaQ()SL}s2XS+CETCOOK;by8_nkhs?m
zkNNQKXLv8Te$(lMG?ftVvMhP@=$tRUJPmU<CN+ktDkwJ_Y+W-MXY9@AXl+=`q<+zK
zGGQ_v@!<#exxT+gRxCNYyJEdO$Jd8cwd1pU55c6IS;zg=l2>1No@KE@E65T<mZnn9
zpwZwJ5w;1233@kiv#&k$&+Y!**F=UGS$m5v)C*zWW1vISYP!$rGsGe7W?M_z?n4ai
zX5e1K?b$RRlL#6CG=T-$5TJFupc99@pCmQZ);3WpT4}ryqLQRhB508W(4HEL^@3y^
zt6Qddg0lk8QkA8kw<<;DLM`?*4Wpt_&XC4r#+VK4J$T85F)9;3O=9s%VJnY!it%hF
z)hAtstv%I7vXJv^LYib?J!M_uts^&M4i<;pym5;g*N#|B7hM12_Zg2TJbZ8;zg(i$
zYZC9Nikh;hB^0hMFs>wZB~#<j#ggksH^|3he7WJXk3QgWG67PoPAKz?;=>QAjbWs7
z_GWvGr&GL2DXN;)reM8VQ5FTbnnZgHp3EqY4yRPbnm_vIbAIF3zr}C;#ygnNf;>xk
z?ai;Sn2q^o>lOd%2jAy>z2a~>;_T!hpMUZpd7`*}FlTRX%3?NUI-dxtb0W|u##+>-
z#MKhcs+0mN;I1}=h1AvtkX9kOwqpi}xNht9n&`V(hf11NRk2*HxL7WmOz_nzR5k4&
z{r1rhMx#;VPZ7XT#p&r8tMy9Ikc}3fqBg8I1<U0M?-d@+a<!o-Dzr&itv5V;^q9R`
zA|aEcCP{2l4=5%}%0?73vFU6^rG-JO9oo<K-)d`%Y9+<hV#>l|vEcgAbw2&{Q@nGS
zB%#<87~et^#j)rnd3Res$b+cas}L=SY3>CYZBiVb>1+xbPA^W`+uIX!XMl*QiUQ>w
zCy!6D)^hv$KDo&;%Hf=4oEhrWU>jdY3<pohk_1}`E@N2~<f93<u3cw!cFq@{ea@q0
z!E}Di-t>@>88gYJ?9Ug>7h|p+>`@gP9zQ<j;`Ee95AGo-5@MoH(p+4^0p4mhpK|-w
zO>SI2U_P62u%D78aPqjKt{rFRE585z@ALotC;yO#pM1!RhjWtRoU`SU>3qR+FTO}N
zo?_KQ?mv9QPe1;I>#x4X@BZHJVlFPI>YB}ZjqOB(b>~<qnta39hI}mDHsr|fes<@)
zWh7Vokh=Gc-@5@n{pa@AY`kw5U18{bcbVg^YR?R>BRB8Zl~t_8Q-Kj>ud+2-$)~ff
zamXvJ@K$^SvBIBHpaMW3UZK;Z!GCf@_!}pXML~L$1bX?D*iQ|$!iC~4S|yTQZ>1It
z-k|W*l|sV2MLJKlLF<$eGfG>6)fgZ*1`<W8GmMwg!Us1GINZO+{%nu^y#o&S580dV
zktG>>^Zz$h?Md@7(|p9?ot)L<$DBPrV^f?`mn$YIjFhE1KVdIX%&56JA2ZuOU@^bW
zD4k&oPg!gBZ{1;Z{Ra2TH3gsw$8<84N{{2I_@=a=i!U!O*sPXR>m^mWLED0NzyD|4
z|NJv9&d>Q*fA+_G{%`$TeCIoVowwe46<2wVZr|a>*WcpL|LO<);<NiK67a=}i&aHJ
zjdPaMvs31?F^7ly%;r<FEEOW`z;8;{g=M1?oDB)K*5kF*ObQksf`zA9=-uFe)6>(2
z;Y0?r*=!`NyDS?&jhNgvlA}{>Ispi6;-(Ab6ciFrQdJhM4AbdMs>nu^3N$Atr<|W(
zkmpjbe!X6EbaX_LB+Tb?rqd}|mNkJ7X_gA&dWgA<nfWMz4(ggn;(2jHc3Wha-AyC{
zW1FWYT9ZPd8ILE7#}kZBa8@Gvw9=w!hB58>w7Spi5~7y*v_1i)Le)~$O==WLJxybt
zlMNV-A>xT7h}Gi5IX^q+{PZ!*Gnr(YiO<McQj<|Ag84<}o!UwujddvD668rjo=AN3
zY%*qDd6cbKpPf?r6({x)53`)3qw9ijef^NVy(u4m@-gM(jR=z3P*o1^YVtg1y1!t5
z|A528eWue1S*n@j3FoI4tv%;w=e+ywPx;Z0e$4;x5B~>Fj_)%y@WIE&OsL6{l%s3c
z7%vu_tX5cMD6HjTz2W9`%6vYjyf{ZIiQwxNZfmEus|ECF=Vh3hzCWlxmtCgN)jaz5
z8Jhd<`s)V@w0*s-MfDk4_qn@#Q$ydcYR??)=Bu_dF@!q8*?=HKYwxM+iYh?QXm7;f
zsSQacHFbP6k$G`~)?R=ovou2|8gCumDT=C++AaVlAO>T|tkeP0%19qvine0<^GXVF
z6@|mqFda`=j3<(@??S+Vmyl>p!c;F<Oj8aHu5oa1z|qkW*AAr2E?$BjGLw-gL!OSL
zqNz40uc?X>?_fM#@YYK&F~)IwF{3y+VY6Iv|KTHk_Wp;=PS1Ja#v#A?&R6-`i~mOs
zuN^SIc7)0%IG1s8QK1=g@6iQ}=M<v}XQ!tmsbP|i*qiTh{l-n^`-kX|?7LpqoSi=A
z?CgY#lM~jf3siAIU6icPPRX_6?A|^8;6MM*__IIy=llo%!T*MD{l>RBxPFKqO|YXG
zkIRD7M<<jQCn#G$z2WRkk!r_uEXBNL^C{zTs14*aiPex~lqO}fE+~U()LJNPK@rB;
z)-`2OvRW;r=FItdU@9r=^}6|Oy<VfjW;Tgv%n}xxQI;E)%OxNQwcP|gL9I^NY^0!?
z4x3}M-mqG4P(Ye&Bw$4=jP~5Paf9Fcz29T8Sg=?u*xTD<vA0K(BuqoqO!46)qD7gs
z(V`<_bYyy6kkq&*eBip#8W+tjagW5`RP@#Oh!fXXM*>q)L9+~+l!r9S@m1MDFt|*V
z`b1zmiLoUhWm%FY1k@3N&qq=ztgAtJ=8HL6>i}z0Q-|{sV@T7Cve;0rFG#a7MPM=y
z?%!kj!gFAx6on26wZ<fj#}o22BQ*w{<xH}S)!8YN@t9F!xOwf6+3Xs#(H_%$E@ce7
zM3s$lVSe7_B<ahn*Edm0k>yg3!#huw<sq;_lO_qy)tsGf_};tk@~{5#dwl%y$DE&@
zadPsQG&7tn&#}&O_x=NJOmmjynwM@KapU<H*?7&#y@%X;^q3Dm{glF4-hStuCcVUa
z-&B#^hP>$#^2@;nk!kD-G#H95>*B6mUT?$Zx$b?t+RIQIVoF0Mz@guEG0dlF&%}JX
z?-?O!td(ki-b+ZT3gzIOs~ZW`Iom?&n6$}Qj+@mvkMkPqBoIL>fr!vb<6@9#)U>?R
z55*Y?<MfF<Rp)3eT9x;rE$PIO<QcPU%s9yzO-D>7Q^w;dS(Y=K&X~>S92{H=cuQln
z(&X8w@#o+{WtuF>v9(28Bh@gSBTZ5gMkoxMO-WjbiEWh?jPjh!X!6mR$!tbGTX1;u
zHeY%34W{XvY>G8M|L9X5p54WpDRrGPpWi_1go{ndY2jFwmh;tylr`t;nhgo_(qqz0
zHh)=idV0b6=>^NPC7acfq*`(P+I2R^54bpeOubp6>ka1*Gyc*4=O1wI{yl#EJAZ@8
z;Q@y)yv&y$f5Lk8fQyqAi%E*A#b=Ug(NK!A<n;WUJkO-0hSN;aDJsdy<_oNYs;D_V
zJ!Ne@XPbg`U16PNv)Qn`Sh89!*=)pjlOzd;hlkX4&3e7&;^IOo)CLop_*0}*gmaR|
zRM%ybuhZR-N=;G9u->eR5In{hW{W+>(-}#Uu)n|Glvc>|oXKRy%dfo7csyn_8Z{=U
zE&_Iw5U2%}yUTpyv98^|*&ohJple_-kpV@9Bt)PEvEANJEpdp%X*?b`>)sd<sC+MZ
zHhvCCqkU6~bFBodwIMOx^V?^nd5Tvp+{>d!$1L`yq&n1vvYz>34?LWoonmHicDAA_
zOFsVa1KzxO1bL1N)(u}f42E%@Gs+SSiaZx4^x)o?ymIpfI+iLgHekkJDx4R@3>_k1
ztE$Ahn)xheGR`FMq*zgv1!rd$oSmMqyjZf{tl6wr+`D&=yLa#M`0-;t`Q&5P>lO3q
zoGi;YU)G$j&%jk=<1wqMB=exhQ<fI)esPb-C#RfVEIBVre*L$9n{R&en`BwWdbMum
zB+qk-qS!%rpR5gZnM`-ihR*LZn;|^5ZtF9MU40l_%RNRpw8py6-lcI})t-s@IOlMc
zpabODi1Bz5BG(d(5@g%DmgF_>BnL)*>)54z*+i1b#!tnut5cM(aa9b(4VYcE_<oGa
zB{2?<aS2#Wsxz`=gbwDj$~#=GP&#36G2+IJTim>Ho5T4&hx0wg;|ZhDh%7phlQiUU
zM2BsVQqhvN9<(_V`YK4<DnukHY%L+eMjKAgFR1d2xltUuiba+%%X75Ry!7Hr+<xIj
z7T1o*@-eHkV?O!xQ-1V=AM^8%KEtXJ*RH>aHlL9%Zm?1*<z!Adov<n@P9L8z8J+X!
zV$E!E!gw}kGL@3vn{~mesySN=q0j7w=U;poQ?L2ikItE98tRhu=`rvA@Pbe8-sRno
zKIY9|{WWZoQJIvB((>?Zi8gaGB^525b0lbri#4Mv#jFdo(PT-+g|;X&BKMlIwtRX2
z0e91kvrWN8S>UzCAr60ATh7nUWz(+LjhP~v;L1|Uqk|-jikT}$M9Lc2Y>G}zdWaD|
z-h)XJMn+6y)5(NM_&1x+Wv;Vqi?J9j_<}AbSB6cKrD=oH7M;aysD98?ber(H<y%#%
z<KWkH!Qgz1Gzz}wi0<t>5ocX9jSKPTTAL7+C40fzs__yHZSKBJzx%rgiLMa<wKiLX
ztksaf<0#95e4JBN1<K@{oIK|6a1UEsk~GCT&v-OJX~X$?O`c7-_r)WMvf|$T2fS8W
z^eBXIql9T16M~TyqdY@7aCOD~FF)tiuYa9`y*cZLC;aTEKV~u6XPVFOCWBN;fuP{*
z>`W@RZkCi~!Fs*n^z<>Ge}0z-4<2xO@|fjv$tDD(WLefkBBn{gXn#)v6zU>q+6m>R
zq^v8nQrx<EhZ4gFpMS}{FTTKQC~L><7hd3R{m=d$qsavCrIM%(zP7R~o3ZX|FkR-;
zZCy>NZ(bWB=zT3|$UoDc&t00ykfzjq)-F?8|NY{5yBOEgv}a;IWm%#VA;{6T?i|#W
zLjW7afsY4Y>tMPV<FperfvO7m9s<3gbb=-!CQE9v5O%9cjc7D!sOTAEamS+xi^+mi
z<!CPvZ$>91CSx?7a(Hl^y}d&g^L-8vuQ3@<F;0;wsl?fUP}UL|DjCQEp`#k+BYv!A
z47Gq|YL!w|6?F)SErJ=yInU$AXZZ0Ll_(Nx&@SWr^o;3v!sGJ`zBoQ18%?===Xs8<
z9Wa?6Fj*Y(##i^b`P{3Fr`ND{#P{F*fJf&ioSiRuXe$^`sf&urDO{SP$1@V|DXe34
zcEV<}W_fYW@uOodP9KxGg6TqY`^A?yJ$}gg@dI*`;_Hfw(+jf6m>>MrU-4jBGP!<}
zS|yAY`<%Y_9?m;5DxSM_h(#)E)-|Y6B0!^+CG(E)NC<gRRBWo6$M+xbWtwAAEUN-<
z6t%M?so5$X*p;TDuSZZOibnFUZ4FA1Bq?=}b|r2kFq^<cvMgmjUvPBe7USu>kwT+J
z6RV@fjirO%k&u`ik$0O76}1x8ka+oBP6~gk@ZB{D$88J8BLwzVi<_44Da^<_)tK3o
z+LmCYzEIW>F>3{(Tq9xN&3)J=23t=J@7DT+BJ*@u3xu*FqTxueM4CwePn@sGWE4uV
zr?{%bdnws48jTrE=alP;`;X7KC@VZLnoL<0CD~{!^QASJF_h~yCyyV4_fodLs7S{n
z)~gi?$7D3(^UprvKmX7ED~e@Fy>a+7M`amlCT7shX2tpWS(p{qKrB^tDHaMO=O)V}
z-^SV!?`t+2OPc1AFrFoNj|4Uujmc8VvZQp5ho|S9FPCC!G%4%Nn#x&z{q;9^_06{g
z+|^5<lhL{}KbL(TagO@GM-93^M?z-WT2@<Ya|dCM$A@bD>|!+id&F~xnAlJs`@hF$
z|I&Rv#v~-EM9Em|IX_=A$|tBqHk@@5TkpL`yNu1Uz}97m5*m|^W|W02VC5w`Ef#T0
zv=EHM*C-FE5>ts$38V3t$#}wKJm&D=kn7iOFrVzPx7cH`*dx!!Xq{kdF}q2acBpJ<
zGPF||FF8ap0;RRj(<Z?ZOdeA7n?`a8lCAZg!d9WKy+}S;DrT9oDA^Pxc3o1hU~^&c
zRl#ghkme(f?|)A14B!2$AG28O@%R4i{{{ci|M~w4NlsZSAm#YslD*q6^P~4a;=_{@
zt}SM)(Ih%$d9mVrwdTwF_c&dyn9OGEE%wMpqYz<OV-kfbE;ubtnJi{J_sXl>EiQQB
z+8&vLdk-E^xSE5512*dw7x(UCl8lEJ7nIJksw(c@zssv%d4+d=<tx1VlON!V44Wl_
z+^rNfn$y*q!_ky1A7Pc|VpDLoTv0p6sw${mjl(mUj+xD7f=?IxI#pG1e0(e>y})=P
zoK9WWxWHg$vkaqCl5xg-e!zG<BF{5M`Gm=I%yc?ovDn8eEj4|D`An*KrY-VD#~I(6
zy@m<KrKS`yiP(D|z%prq3gQ-3oyaxieUB;a+Klbh?Ps{VQsnt4)Uc6ts<n|s=boA6
zDzi&GF3jJMB(0R1Bqo@=-PZG6fa56&CqCBT=MwEO4?*ObgTw1Q_tLAJ+)r7bElH;t
zCQ13#-}q~ct{t%nNx0rS5~FzZ_%Zjs{E{TkSeF~>vZ8PmYvnjQKVf$BHsf)|W_>}b
zGgJx%5}D>nf;rSP8fBD~<a|}NluEE6veMT@WAb!X%4a+0P$r?SD={bA5Fx7#buA3j
z<QdvS9ZI|;+OR=$vMzY%+rP?h|K8u^q9~b!inxM$l91<lGrpJSSQ85pcITq&v%I|A
zwxOo_TF%g#>idV{b9SwT?tQOnSM80iFL8)L{!*Avnx>>#g2DueO$E@+6p*A-u~>*-
zCRYY$eDKX^u$tLuDo%VG%WE6bEG07|7PAGD$%OfI#{OcT$#lvn8<C|M^U17{21ogX
z#ALxN;?dS%NJw;ah$aGsFCq@!f<@p|(lmyaZUa#IkUg$~<SPs+zPWGAB|6h2DhYuU
z8e>wFQM~-pD|~Tr%BI{<Ys-bLNEBES(pr&E7EJRg(*un)34ipxAMumVk4dI;?mYJ*
zZ@l&uvgwqUUweg3lJn^2@AA>{L+UiaS|b`|kesTzqN;0-PaaF1o~mGZe#Yt9DdqAM
zdwP$@4;5d1?k4W}=g3@1W-DH}b(1eo&pCPUkVIvio^3cdIAk_oP&tRL!D_>@DA~Vx
zi<jyOXDb|zvaU$;j5NzxZ8j_xdjg}Bzsu=z$?@qK+8D~Z6iL)TmTKxsd{d7fpEQd!
zP1CI!JyB!H^AV%bgz;p`VzJMBK5L5L2}DzG;{=$aY$PC<I-qI9oFt_rNjXN5iQ=;D
z_?LTb`!flL6bi2-QgX-%+}~(Q1%D(@wtq+SULSC>OCaX+dGmXkrqs1ZS>N=#zv26Y
zAWG3f@MusXnz}6Ey2zY@O#^3zi5k-uGpy?xLe5Z{WaRmn@qC|Oy#I4%qZxN@9WnmN
zPnawYSk)GnN<^TmEongOc>MSwcD=%_H)KhQEj-r_793tXVDIRVgM)=YWO=YfMOAxh
zrKzh}7*h6PY6K6msw-++5exSvDxf%M1x916qINFODl1IO*Hl)_v=~rElj>YFd%$T!
zI-T<3bI<Yb{o1$q#y8*L#;u!@GT>TmD+Y|X&TrioY0A(VxXLW{w8VVKH*^)zAL?Is
zja+s9(0B10w!+M=`&`wYG2kGe4%ZH?3`vsXU4Up0a1V(QbDF9&c{V`>d@QX^$cr$7
zY^}ufF&R(Uo9%NrKV&iAV=<qzzu03j-zQBnicNvDmNZEv`UgRqwv}q(@13LC)Ka}u
zJ6zJ$Y^(s6npP5)+=#Xq*EagBL1}O?tyQT`Ib<*xNu2Y6L3t@P@vU!ti~stc{$uQB
z1*xK3U7($3Q#vNoJscbM4ia8}>+8&KJjaEDcYpC7?|<@rvUJQFuYHBz`Rl*KwQEyu
zKX-#SRmJjs$LGh#7?YA@Ig}O0kI#7U_!wnUrqd~jQQ&HfhuM6NnvS@)IpwoQU-H7O
zYdrVDi)`+F#@Hq7@9j~ec=-84uHSkd?HzfRlOODXfs4(Gd^+Xc@iC7tmSnR9`6wgF
zGfG<veqCzF%QZ?H)~6R#!30>;HRWnWZ7XV9qO`}klA@4A-J&R(=&(G`8I49w?#yg9
zYr>rKe1b7K#$*EB6chs4Y+~+m)`p6yQav*1Re<c8A+}0cH=Ad8V-0P1uN2NHqDB(6
z;MR1a;{OrIsO{|ZodiLJqIJp;%Pw?HZ(3=dfBt#$EN8Rcgo1klx3jIWNb57hJ_xI<
z{mvR!64Kp0Kk9E%Rn?%JW*JI8CrKhm23BfAIp^6wy2;I#UgwP$U*-1oTTK7xpYzb9
z5X*~ejdvD}M|<)88CUc8(L?ey1*LiEt8enk8?Teh<|J8)Nfgy;6QUwBbYmU`2n<sb
zsuKrOtTuRO!PbH@tw2hUATUBIAg4GlDFI;)(-7&WlSI-tyh3Feh4p;<H-C%2{crqj
zlu@7+)A0yjlq6d5`0?Xr1w|q}%piu$ab3&H?kKY^gBn6S=!V$0&+}XJ{Ujewmx1*A
zx6Aa_4M6CEILAIJrAVY|<`dg9F`ufG(#%>J5|fge6tA_IsXRo$%|z!crh6==d(7r@
z#v{oXA5SJy-`pf*`6xJur;O4u+8VUiA!!S6U^kvgI){W1iQ|IwCfVe!7H5#wig7+}
zRNAU4@lMo!lwizN)dnBXthEuJl`Q1eNaB^CFgT^9=$4OznaE~TLX6PSZZEk+)_dN3
z>rGyM?_Iw5^dr`1C*=8vswg?zl-Q(ZQxp{Bypd12{mL5@Cg<jBZ*lbQ&-mdFf5e~v
ztMBtq{^=hwp3M06Z~ZlX^>==YZ+!i$B=5b?mtUTc7)3f6W3*wqDFiuKYqBIkk&-Bd
zR-U?o$=)8->Xgs#Kj5{uUS?0%%#>$!e$Ms7L&lFzQQmX&`gJsp+B>dYzsB9W=hR+^
zk1~lkR?AANbZUcjin^#Mox|6bi=q%@=;e~rJQsVut0m_P%Vx8}amsYMU^bg^{pdQQ
zTmUB1EM+tje6Z<k%6L2>O;WVZuud}9t*xcbj*32;RFXO~#V8-46C*(l9Y=iE4&MG9
zO<K)Xy0oc=G!GS`EgQgspshIEmNwO#FIKZAWEvt<=u4;_1O(rj@$gaWhAhi@;rW*s
zk0+d-o`IKA8hz}9L5A5QAkpWRBx#V+gO;XDBeje9$bq+1wnkZn@ijV4rF6A1Br0KD
zt<Y(P2VVcmTfFj>H<^#7sM=$al%lpIqX~tpn(*iOV!>oKW3^ec80UCkxmx4!tjm%#
zO*uU~<?+cem|&$yq>dNFz6q3+^P(|pt?_{wrCACZqy}*v<{y+X7-y)0_7hV(WRqK*
z6D?J15vH|CNb?a@RbjFbI?Yhpqm5?0*^mS4b;*10z1JE|2TWLfAI*@XzVFxR|JJ=n
zzldD3-Ub-n(C<$ZP;hA<gfVN5hyQA&h&bpGM#nc+g?8^fO?x(<4<n4v&=@k6Gcglf
z4QY}w&c;k8bH@3YQ9kAN^_LswV@wEH)=HdoMD*f-0=+>>7CRd>s#F%Awc-QNT7reN
zXiIUk8brF1&~8s<D~vYi#NeG10_JM04N{+1DD4CpI|R2WMF@UqH9O~}pk3thwz6b-
zh>p`pf@@q<UNoe1KH|6j=I`+1@6Y-9kAKAK{2Z5LjK^~dol&KV@!?Hg{rWd~Tv#U8
zXRJ@pI9r$KY{bEh8$5XQnEQ_&@CX0Zf5Fe+f0zH&-~add+Rd9Ryyp1y0;OO+)~t=8
zvW`trlcxzrB~m%DF7Z``4wV@no*Z+sxy`ME18^lqYgP{)adhi8<T+VBA{&peBe-$<
z2KVpZMZ1dS`2}@ZvA$T6WGSnY6HK0i5$(0MHO^X!)h0xvt+`k&#V=+p&RVn)Gwpaf
zVgK48i}`}R{k`C<4U)Gpf)NN}((s<zS|nFX<F!Gj5sIb_fkp|$y|NA;Z0oJ5OQJ1X
zOs(!lY?)EJ@3y};j|k_5yM&{%XiS&8AV8KLP%0n?dmxr9h#V=hjVANo;jy3^oaz`j
zC;zl(|6sva-h7?Uzqm^hd=!~(rOvoM`;v5D18m<dmNO58P_@L+hO*an$kFoFGfFeO
zbJ(h)EDEf#Xrm}xDJIS&lfawWk!US-Z4!g^p7m<Qfptt~Gm@LPcyR9llkt>#wIok+
z&NemKY=%X1e0;_ikB=#<n$>25Pc>Fq>dKKSF_|emStjK@lom6q4FX_UTWnnkBEA#Q
z*nFe|s<&vA*7*=r1g>@%uhFT@uvDKF6c(%}OH!(Jfvz1oQP`>=O$1l3C<;uHNQGZ(
z1=+fnWZGSC?Pn~m8LRGga=seEKdfzB6(|rtw}mEkX<ZC4zV3Z4X*<q0{!0}FErKHH
z!y0wLj|GkOOYT3MPt<SE#C&cZyu_{BH@UWdjf1@d_GbI+&-O`mD&fmo79-wZ>x4AR
zNlb!|{+CAe4}(#M&Z0<6hDuc<HCh`=4dX*8X11lxz$lax-%JdC^+C!^vW$Er8(wOA
zbRg&H_;ggH9Nl}<6uVPOfFxu7hrC9c<mbXb7g(Nq^(9_><puuw@B9ru_}P2h`|>`<
zjJa{^Iezl)hdlq%t9*3-5l1h*&P7pERF*UwvA1`~M~^;ax;JOJykIiU`QD%Y2|xPY
z_xQ#;-{iM`?|0e1alq;6DHp3X*AEt)oiC|O%Boo5mDu4YSwc3yhN;fjQ1RvC`+V{6
zK1cg6Guu1FOwOqvpFsxKkB+!@>jkWWFYc8*czjA}Eh99iC#T&1;vP3{T;ufojI!7;
zPEu-XIbW_Q)&(|3?RdCz=Q*;MUnWTm#z<&u7BrhQ%P^@36j_!A(_3pw(K?h(Kp>N%
z1f`4?0?{>V^v!1~7WEs>L)dODA=oAHb+x0oPQ{PYkxIkyxPFu@y2fPX<4I9{C$6hb
zHaa#eGC5~kQkKuED!6guI)C@?{u})HpZ`nBvLw$_eAO)>KAd|Ik40fBGNLw5DG(>u
z!lJkachNDPOaK}LaflIB4sA5G6|?|CNfGaT<72J8Lpe*9N|Ns5lT%JU`jl+3$8Y`i
z-{24b!G9~kK<S7_k5763gOB;MfB9W*JpTfDHYOiWSZ+4B1j==RtE+%Nmy)JZdc;>U
zKPGCB0X3yuFDW(!S!URjC0Q<g@K$_fN`mImXc8?rkII8GA_#fssmh8hOGuO=(V;LT
zpuJ%E&1N%x>$iTB?8o0H%d$ob>o+0h?VKtV=T<fAzdPqq8*SItHuLDtX}9e%>t5ym
z89H~%1n6bHwQK$aCuuIqb;50YXIH6h+rrbdXJS6T|DAu6qoeE0CUdeRmq{cM;7_l`
zv0qu7lMu}|XQJIGT`a10P^IoiTV~r{8u%?-A*C8-BLu9SeCJ$KkgTzNBFTbUZQEv}
z6w=t)n++;e3S*T_xv$;g=<zm@_>J-r>3E6(_GSz2+<t+&bR^k?%4$y59+#x-Uq2#E
zGwP}$F`DUQj62xli#+Ap!6C-O+35*oSx{FMfBxNn$;0!<9NoIfXgtDNDMfJii!V4o
zU$I=SxpDI*ufFy=CP_It-tgt8pHQq$*(}d-^^suUWf^N_P)W+g`i#3j{t@fH`Z1)!
z^a~65c#q=ZlvJnO{o)>b`+L;2W0a3clSCrUMq@^!DQTK9old#^<$dz6evRc<zuGY9
z{$jS>xz^Fw5ymkPLZuaAI|-(sUewPvleqjj)ShJnc3YQ__cwgsjPj&=NJoZ_w*u2p
z+iV0%=#^JqX&{M&fW^Fnr)zOd$9xxgwh-%B=rNoLsOG$`&nPD2cIU$ckYVptV_F2H
z8gU@<Wmz&y63Vh9(eUzXuW@weIUb*!Qm$4cF6Yt73IE-H_uum$|6l$S_7{7Mvy7rF
zSuHns7t4Q(2`J`kWeDQ7^KFilQ;K{vqOJ<kG-b0{VUk`baJ0Z8DB_7p!akQmj7Ayp
z{Wy;{CV<I`b{)skd(Y9)5x|Zpx3=v%wylwN8P#shCZ4xz?d>wb_4iQ!9z)L%OS8t9
zF2VxfJvzDUe~Mz4YfR9;1hd<n>#s7K%}A7F`Mb(V<w~c}UVJ}_6cc?XT#oi=;N&|H
zB}(bwm#LbfV4KY*Y`B5J4#cX<Fq(k~=Z!`qCX-2%r7cV<GL2RWHi>R&L&N!X0~-_>
zQ`f;49p_D?5(-~~grGZY?J+7vWtw&EIX*q*bhF{LH(ufT&07?;V}Cqjof_7M`^?58
z4DkNXe?eKS(P)<8yZ`|J07*naR9Y7Hy?giYIDYYqpYqdpf53b`XExj8;NXD6#U3N&
znHkW<1*e~Xg0+?}@89D`Kl%Z6vEtgnjO4A?NpylX3A4R@D(g92uK4W!33u-wqef$N
zp0hWbFy6b4v!0BakMDlT>u<cy(al@jxN(FTW$a%&B*}6jfV5JI#e5D22aK;@Z-k4O
zI4UqCp#Ykas+N+k-a6}*)}rM^SpOzct=%kr$+EkihN8}f{vK-G&3P50dzv<6!@uet
z(Uj4*#mDnUqY-soH(;hpNwTISHST7P{b$D-NJ(NEQ=MpaR?N`h`NpJD0>G#yx-DXt
z&1SQ$`!*4CfUCRx4q&g&h2-AoQ;|MiyLF4_UwxHFKlna+l;f*{$~hiCKE>8G*AEYQ
z>y0<~qd)#*%Jqgk8)54hk*d(9BGnm5nxdVjc5RIr=PbUinM@{ZHY>@C>I7OyBDx`#
zQmN}2lbRs(h%n%g03vJamL~Z2i&Ao34o#$NG`$VAK9h;93$N3i|Gv5Ef66-Cb&ny&
z({I5r-MoAEJq8iEm4wt^l>|N1o-NecPsrSqED;PYjF#w<+6RD1C91G?cva#E4t^CJ
zsm`@#uke!Ic_}^X*+S>-Zk+B5HIt=iij76)5C^9VlhQ__bh9s7eb_gBQR0YHZ8Vej
zp~`UQxn1>GBm$A@ZRMcWC?(irrIRR{ho`4}a`#JK{>qy?|MF|pK$aR34JXTU%JoX3
zw?d9ZUDq5xK1L~c?)m3<d~!@>SInjp>T1KMAKv5sXHqolV1J*3#Q}IYU7s<XOu0GB
zx%HjzaDH(@wZ0(L8lw#3@suym?vsosq?0M5nI&IVxHL!QBleE&FxH+&X-;)<f-*V#
z2S>d4;vKFX9b&SCBu%jhbMRr@O_EU6H5V64s;UCouuF+>lmtet6SUs;xyX3x=69_H
z1CRkJ?JDy^-_Cti8@k`FXLUDfpE(TmXULh`WoBFbppBv&>R%-OWm&Sfx5ww7e@-m+
z771bhnM3`F?<L8N?Y>e<iLj{R*e5vUo9e-F{}^K!kH=eM+*&XCJ~cvwoAgQQXW6ii
z9P`<fH{SU=AAI>a_VkP!^Es#I7tHn+oSvML<|E#G?REa*Uw)U0g1W34^KOzz4XCm#
zNm3a>F@lJ;MlgZ@%GciFd*A!6<nll#$0-pXvr5qzY8&f@$-S(VQ1V144PB9SERav$
zOGz0CXp&MIPjam9nw##twCkj;OncQ@-xV+r@4IVV_bom{*EaVIE)pM|;vte*#e34O
z6S*j?t8mY*#ptT`Yyk&ep}oRY9+X2P!3JI-C0v5`?Nv>Ukp*ErDkQ7<Ca2+Y)Q(Oi
zW4JpALviGd`Auzyo5t(9aPwOe72$Gxdwv`+32cx>ZgePu=C)z#mD=_RNS3&580KbO
zQYS0%Uj*~k>S9ghHQwaB{mpOU2z6mf2@FWJLVL?}I-%NZxPElV#l;y?_8}!3jhNaQ
zW#O34#u%lU<{8zd<ng12eD>%*Unm96qrFs-)JekP#vP81t}{sz4(6gMRki2(&D-25
zJ$Fx5yn5|6uYP^P+A8wd0k6LDG9xU@NB8;oXFuZO_kKb;nsDRh9VXLU9KfN}_-a!&
zCd;y{Sgi!eQ#QCEO(JxX0iEL9Hj{~!Oe67VnYsjzpodKc{T8*NUCd`_^Y`y{m6SHL
zaj#-JyUraXQd<)5<$lJ!V2t7Q*I(z}y?b<);}e-|_dFMZZ=m%T8Ix{ByD4R;oE*>6
zthp`{_pa}#c}IVlaf8Ps!gQ>&Bq|bH$LYldcV2#lzyBZpNBr#FpK<@;BfN9`%GbYv
zNm8bh3BUfWU*o_1um2mWtfF$BRavxs@lv)0@1<y41JMO|Pi-sSe*0~H@Pof<_Gg#*
zdT*uDYL=3uiqcj=BMJx-MwqSjHc0c)94`WCH03*IgIlk?_f>7z`h}i5p)+UCBHVPZ
zd0L;)uKPBY0M*)uqDjyr!Cw+TBPbF|Hf|8@RbWKFgi5B`dlEcSGFeItmt`f@Mm<5A
z5XYr;RahV@pbB&|&0CaTSy;BQpLUb(;%8h0&1Cm9ogpq>*R>!gB#Gn$Aq+!@>+wE7
ztaz!k3E&-dWkVRVNZ~dRT<=_v_C<TEf+N&P)Gc*oscTE^pmH^pEjd0rXEd4dmAAfz
z^I}V{t%Y&}RY|Hm(>&wo=o)xOQLHhE!5g?ZU$R`U(23#J{!N^#STC34sbQRqF>b*+
z)zn3Sw>1WZ_Y!GVZ!Wn1dBMSA#-|^D%&RZGz}~?=c{XCU*yq*P-z1;!v$%E(GoEp}
zlFa#es`%<_ukfAU_%{E~|NH-qi{+Zyc^*7E#w3Q-W=%F8Q`8dm73VO?bF$I6nZJf1
zXsHupbkfv%Y6+Mm`nI<2$bdXCB5YeS9M(em^Zz8xr0d`7{?<(t>@$R2rjwy#eF+m<
z4SkCUd9fDs&3;`UXIy*lyz>q}_`wf2J$Vc$Vg=z{gr(c3*3TpFSXEVk9uiZDbE>rx
zpql_1pkX@E>=yS`OLS?h1VV&@KoKNn*=8MZ$~2|6HQ8j!SKoez*Q%1bs!5Zq3GSKA
zX1w^~i>!Y7Gi+@`a;-#`#k{a4!P7@AtZsb0ufP5}dwY94eE5JYPddJ3(ZrqetT!9U
z5mFwfJXxljyreWuv2}$BT3=C=p-xjfwuwnv%G??6zBbyO@2=UZi*nG;Rcq!Sn#cY<
z;u`6<ApvNJF+E9m@xh4J_AY!ztxzOKP^v@A<8yxrJ|CrOycbYNTZ`>DF$p%9s3NqE
z2fPZl{8(N%w7%~oZu%rKmxxVVQ0?<oxK2z-@qS&6C~mebm~$7gvBTtq0#w!pi9yb4
zN-lVmaCjtaQ&jD0t8Z0Rp|xhc-cVL0=VxaeomEs@-4=ygibHXyK!HGUcP&nEcWH5V
zE!N^%+}$0DI}{7Wi@Q4%0tDB4{`-(I@{qSAXRp2HoZqan9c{1THzqs;eIivPj<HlI
z;mo>DlN^Z?6pEO_8%ea8^c67SC#!2o<IJ915?nlhv3Bn!DvRz-WUtJgaF+kk(MURZ
zzYgY}o1&92M}f%mxEm7KjbzN~(*nZJP=6Ta-G#FM2=wIz3iLYn-3NQF-xCxLzCH3%
z;#o=DcF*LlSaouBgjhnK(F+yJ>eN-{QZ%eTB%CDfk#QY@uyu;w^wyittD-__!aqvi
zd9hkyOVb9;zgb~9Zw(GpQ1d&2T94a3kFP&R>lx%aqg&HRz!%Hgd4PLUYmS6UA5rY5
zjWSAru+zoOiem$C8eMLG(AE@$ZUl(pH-3W(7f!XdSzbMcAAyXZb*&I2oCDsz6ZD(8
zO4<ZXHDY9R#e(}H(aYu>tfUh3dGF47khOGzSv|%GPlbaJxvAgiZoyoycRo^KoUdNT
zJn>oAmm0$kT60zM+z*cF=hW3)wAekEsI|gIGP&A;s^8?<$}1Q&q*FXqa2Uy6?er*$
z3j|-ZtEVWp@_UpXHXi%Da_jzalwLnB2#bI=hqLUVqj!FX<lW`mYtne4<-(+WZ<Xw9
zN2r~>Sc0_>SMnmFjKA<vO`T<o7o^M4<L9ZnB;=KN7VqB$PNfq<zXBk+Qj-03)sg7v
z@vzna(lMI}l%4R5cr}|*k$)zX9pfcCzwkbe-95F}0BJ5ATYeFd{fR+-QNP=vJ6m2l
z2AQv0H?FLg4=1%-U_L<+z$srBz-HoZR97RHq>6V!LugqF(sOR@k<r!EC|GfUO^`W=
zc!oh=_lyBM4?kk8D~TmS3u-YNIMPLpYL!kZg_M@)2GcQ55%R4;ns^+E!fyaZ;X6=$
zU~J;*{F9wNaq2@YXXG47rBI>J&bHZLzsa!Y>(k`jRig48WT9K!B)AS#7(MHwL8ZyO
z1BsmY#}~Rn%NFP+DLdv4px2mY?kF;BTNE=>>*^2}_I^X36}{&U;`n#Kw@{oEO4_!{
zT!}GSqZvz+^(g4x;bFL-=h^vEX)cegm-cH|4)FmkN5Ns(RK?s4e21gY3&~q*$sYF&
zQY`7FJs1iw&T|z!Q(LgXFO!XNf}`B3lOXhi6^klM8KBWJ-;ny6A4r!IE%8;-MXuN$
zNuS~92fc*0_Qy$mZ<b2!A~os42?LRndd&hyL#GdCzc{~r!*46z(ll@8jRDVzYe?zd
z;a<({2WhYsd4L<4NrQw2=NR5nv>WrbdzIy;rV|Ww-Elpmb@`6KiiB}t@kNTN#xXor
ztkXZV$&2BVz)Z6x3cnp~Pwvt-A0whfrQ6+6q)5A_9Ym}&{om%=r7I_mT?)zAt_2rE
z%J^-Wmwo8J;43&}Uh9qd_QsbB=~$LCs6rVJ%1mSlFMno^hhBf{w`gz}`E02RxxeQJ
z2DxP>cK08{(ql6y_g2BM@2-mllEw6R<SpeY?lhWWxNqUBD{%-bdUCRJ*wU4jlDMHS
z{|p=1S6eSW=!|-~{m`O6up9#H76U~%UR&@NeC*q7hBc{lO$0{-$~+{lw^2R@9oKLZ
z)*|<JEQve7Ls^=KoP8}*b&fj>M(<fO+4lOt_%vlTyEw$I_fhtZf>V-Am^X5E%5ON`
z@8dYpH~#os%6n6^pc-pRrI_`?j}<3|RGm@QZYwisx$Tk<*1R(;!&eLuH{jd#c4<Qk
z0pQZRL7Rx;ZrGs1l<}dtbn4Dz^d0so0hMi2h9nN-TPVJfF^|49r|Rrei-G{<9jKF-
z>nn_%-e9cKh0M*CIAY1@#bv#42Jb`|lU#T=3h7ow+SJeX9_za%;Dh9*RRm89=?DzG
zZrY5RaqP$NcK2cW-@O}QxXTFoth>N1y@2DDhhI}u7fw)?^p@Gn+1ey6dZ_xNf()a7
zB!8@Gh_X#8_fLBY*9~7@Gl!G)oeJUl%ASrfx2a$fUO*_gB#hw_(<W8R4p4TOM66Yp
zu-8mv{Y0D8+)$TL)Rbc4E^QF4n-X7A?e+l|ZE(Jd<iOud>0;d(i@Npe(?M4Mw}m95
zpUX-q=F$BH;Dn+&Y)2kar3FSjxyY>Q1)UR;Uvi|bK^1aTb2g*2cFk5Lhx2($c052a
z6hL_=8L#ocFAvpkaY)#A73uJay%oJnGxmGT?7W6l{+o17DW&mfqSE};-MYE`4_zC@
zXqIj=U14aI#QbT=_=)jliQ)y#|KY4-am1Nz*%U8<3pN&KX1Z@JXV1G^PB~f3ro*M!
zA;(r_0pm}+$l}U9pZb*R<aEm8_<CIHyfcrsn}{${1fnKR6q*`yodocILH9|#Mt<pY
zmsH3oD2KldG(cv~bMA_=U*kP^Y&uKEb#TQ(!c7SZS5#b-N<qDv8Oi}OLh_^>oNkIl
z%G6(PY6GUZ4NWk<`aG;z@M3M*wHff-d0Kx~p}2hik0h$1;!k}vp|Ni~VDCx*q0Gy1
z>*C`|N<g~qgjq7-JC3joM}zGiPY*J%bauC<SFP7jGga$<!k}-XEZdBn0hZu#g#(<y
zZtj}_RCQZ`ogUt&4j3YRCu(0GYGIf!z)lXO#3*=MGU9bA<TE@OKIaTOl+>wd@BYBS
z)ya#<O0Cim`>Ff+rb_&eL(k0|$y8jZ{~Z6b&#6-A8i9X1x-5oW=6b>3S>)7s#DVA6
z(JvyOK3UE-cN~&w8cKvafoB>~ni`AoxlZDU4K>XRizQvPNz>;oRAd@=3g}2yJZCLT
z+;qK^i^Oealc*vC(G~1NzAovFsgZ=C26Ga8Pl&N1P1MW!aFSmm$tfrCJ8&xDk@_zO
zMyKP4e0cTgek9DY?CoSA;V^NUsX-9{ncR2=%y3WG#@>6HKQS<gis9ZHXEB3u`>%lo
z-T}Zh6Wza?V)ulu?et@c<T03arexCGD)YP@>D%^;nf?j-F%0^ohz~AM;Y)HMV&0?v
zrRL!+V3?N}y;lL>J=rBw8V&I|G?O$J2F1mG<c=d9PtEmJXcDsINq4@y3${^D*ek^9
zwH)w`D2^yu6?4l}%*R`XZ;15V?A<CdQcKL)4`OTbnZf{DY1^Muo|ZeR9dB43Hq@HC
z1$0z_>1z(BJ#v$mm;XvUM`!e?MLGsNt~>$Wse`N#o0?~le>R$$7Zbq#+3~r$XD|Rz
zxkdYO&}8Q;uJ6<G*7O!iSlx&(dBnwujG@kN1F9*NE0n(rSd_d;oT1smKp5GvQ_L_|
zuuz|7Dz31psDM}a+YR5SNvxP4S6wRb7q#vuT}6d7V?SeUCn{aN1}fn;GGDB~Z-+UM
zwXN2bkrnt9Wy;g|0O)FyG)~6IAM99eu$eW~DVi{>>3oe_a{Y6AStNqdWJvPzw@D%+
zzyR%3@Cj$ro##X`2$T(q!1ErdM<isP<ghbLjFqWcLv&=0nD^3FUr>`MDwFb=l0ILE
z6wlKe`EnTvb~D%IrhP{d=UZ4y+uY7XFE`A5F%e%xpEwNK>_J#*PR1H#inmE#_enW8
zP*eBkO30w-g<mMjWQhIEt&=qAq*bsRt!9;|QfjS!I_tF*4S)hh$6TDd%^Es4n-(@U
ztweWrccVKzdp6e(D1ZUO@%(0?XM#()JQ-rf{}_c`dSEF802q*<Jww2hpiqIg@wB|q
z+T1hL|A?t8EOKylm(~Eot*H1tm*TFfo-E&ZsTiVd5??3h!l#k}@{vr51;bD<mHqpa
z*I$Y3IffUe5EDskY@$PjrpgL7=L9|PY@xnEvHmZdA^}j`^#{npFNoSQYO<`eC<P9l
z`LMMw@{bTel-j=xj^4Sm9rY4OXi!Ti6O#NMmL}683?9*_xX%B_p*Udr9K1)M`gvVu
zOcb@;vSaTTPEgRzz$)v8-Jd^6dS=!FVDFCg!kQDrzP+XeBj5@qU>O_Bn!UlQ((Gpd
zW9+%)nG}7c+PMgBuYtS3J!Gg0rdNS$OixRyZ;o+FPGd@Im%OtseA<>UQvgY+_hly^
z-92?)<xjbM3#Jz*q(8W5G8f6_Bva-WO%~Q$z_#v1n5hApg;d(hS7zA+9M9L|%!wai
z5@!MT{1FSYO;R+d=6FW#X=N>5`iU0&CyK*vJa9;p07-Y|KDvE1ILRt~h|its+L7~n
zH|_Fk^+qUs*w?G$G!P?PR#pCDtUr!z^hYxPGnF45hR<%4L{c~U3m<AS+-dHGOopu%
z+Dp-B+%<h@_c=*EnYE#-7^YTDvIJ5<>Wyki7N0J>0xKD^V0eSlMTo^^m8kXt@hWu{
zWol`lejo8QEblO($rka7>SP1l&P7jcN-=f0M}c}FOCgf4LA1MBn`8^_VBJ}<<9M42
z-~tx2eAqd(z&)WG5enuSk;lo#kSwe)sn=r&sjM1I(QN3Q29+`}R#mp%-j~ub(7UC^
zv^vASvv#)*VFp}EOlTi;z8qHyc|Ot-jhE*7LLPRMy}f+La-WZKoo->Q;QFkfAlssd
ztq{Q$HeN^IEE~3IDx%}n&;;ueWMzFo%a`a=H2obs!XaH4A<1XXPdm(}3-N0I!j)rE
zZ$K0bX7i2&<0eQZl8i_3{=B=<cf4pKG};buJ3qtp@6$K-!xiQz(iQ(|i@v5<AYgOX
z49TMsgBVN?bln!bbtg(#t?n{5o{3m+4<O|r{VX5Xp|)Rhe)#e%(Dg4(TmaGO0iHG`
z?}pCcuS!eA=M5u@!U2!;vG!;83HIDc@5j@Y``vy_QZzA7$E%HeZWyt2A@=8;QZCvD
zL;FIML53PR;?7F2+S{=#1=`+`%|(`<v{1#Jo(P^r!vSS(iqv%;EdD}nAg1y>NFEEc
zS`*K)2|;N{G>J-3zCJk{{#hQa#~z)_=v2CctZZyj1U%3u3L!CmE*sz`c_IcE1kIlo
zJ=kZkWlpsD?~n*g-Himf>)4&gfL(l&t~{;Dm8Yd2-=az|lOJaa?zo>e8OF46PNYIx
z?bUWJUyhnbX~j0MUAXiddu$1B=;ztLsR2JI*V{A8=pob2+bXR}$Wl#{!wA*nfh(l#
zO(erXF|zJ?k1S&6tWs+0%Wn<w9$Q-=B9t=ZCNMb9?xvZa0AF1cmvl?Xq(P~N%ChKE
ztVh{+@wzTym`q0#eg5>61p=o{SZZi(?i7*{p{1)8<*QYu-MM|PF`QX!OHMLN!7?q!
z70sR~BNs*v&fnkamk|;XHtk+KXcJ@wgoeAj#15gC(Vp$++}EsH|7*od-~ad+z{YkY
zHsC?s|7nN9)#aPC?p%UdhPj#b2x$CIVz6%#C;3725L#pCK~4u*c#f??+`NEqc7i1}
zB#}OBA+{CvgtpoV`w%rKvU3#e%Rvq1oED-UM#mWU0_mN|dQVrF_(3V9t<N@kCV?2p
zUdYkOPxU-=mJUT7Rca!e5lo7u;-#x`Our484ZOdv{>k1&J#~AA+f()s>VqcA?$hh|
zdjpm#_j{<-?2;njI*yLdbL_sSeRT@EX{qD~i2OUT=}M(V*30mi19iAqIuXc~WJ$^J
zM`r1}3qbj{T{wD-P|fJ~c2lh9fQm<@3c==ngrV(DY0jOkElOz5>kC0pGf&=kLHUew
z6?`!TD+nltqHWRcPXjTqwo!Z;5`B_jZ^M+VlFE(OCO3`<1%~2<8bec&+$brC9u6}t
z|5V0q$abguN+Nu5zNdeNN3uoJiAYCq`q{xunthmW!P=wY+0YF8#v%~r0IVoLehlab
zW=g%(1%@0J)ES}%1FbAF(fy~ZDii`NI+uPgF2uYFM({iwOA2F3=KFPtl;W`iwDo(4
z;J;JWG6{QfXhI(I51nXq>l_tFhRF8c^8^Y%`w`8|V-Sw(O`3G&@4x0X8*RQcV9VK7
zF;CHcNByQe2ELF}Uy`#~Ih6ePGBrD~mGlt{q$`~gGX>#N3lJe;Cplmua0RkE))|Qw
z#0P{x@&$|~-KGoE1}JN=CAe%Nk=Jc`|I5VO^ZNs1+2rZ(^5zPGw?79M>j#1j%0AMN
zeBIu*mH$(pgdCZ1j3Q;EHjkg|iKjVUIP{6>=7!xmR6V0TCGUMP{=JUNXB>nigcF$|
z17Gi0*7bng>&w7)rl}ox8yTLFGBm+I9}y!8oKAYLuRw6^GqD4|mI{3A2;VMVj~Qmx
z#aFd-%LFaWk$V0KUR+S9v5D`c1&agv1G$lHMGxeu?eHC0w-MFc%Tve3o~7ukH!i}m
zNoQMi_o50X9S5iw59si*Pbga!_Gf-)GoFgg){2kg5$l(9Hp2yMiUd<3O7vnB*y@s`
z=UC*CYLUyDK<GN|`I+Cb^Fp8ny1Ku9;0acXfkiIMi)7XIT7Lq=%B}XCh5jp0C0EFn
zzW<V=(Q8~z!`_st&vzxFZ_43hcW`%V_1dmCpE%KDgE@E~dGJp%R&@&WtYK)q%*jr1
z!eF3!akrZWYv3H=U@>D4BO@m_sovZE7Y!TwB=qeH8A`KY*<h5o$A54fEg*0gv{72c
z;Fh7Lg%36<8wqR&oDlJRbX|iTM14n27$Dhc9O}{!S`+=)Q)C*ab=?pF(l@BWn#|mR
z^O-<V?FWs!BKN#PkoS8OA)Q(@9w~f|Do(tNisrsw2?kVNDuA(Ur3K2C^D}MI5=i?d
zaG46_Y=iHvW;&jp4xsGW6KTprz}b?qzZkQ#6DzvE{_Im^O(*&zT3NTs{8DvsgqO>}
z+BuK5{HT1Fywk)HDgaK1kCBdEuvMCzkcIk*WyA*86_i!s!irKi;~HS<k>jYw$-5XB
z_2@qS6E7x8b@aou<#%sKB8hO;Y<TTn$ucmb53G6qXyWLClqBB1@w2x`Exgjr*?3v|
zbD@88O-Zla-O|&voVYyo=_2~{*c83_Vx3n-uwlvVq+9tyD_WYm^EN9h3%CygP!%lX
z^zhX!zaSQz_<}@`J@UV)srWn`Gn6%%<^j93Ja?L{{JP_wO0^1<gKp;?Y2<xtb=LO3
z3hO>g0b(l=s&0k%S(qYTvD!xd@r5S~fOIZE5pYNDor8uKjYPo|LLLTAIA<O7R?aX3
zt`O2Y*fkYJN{OZPpK6+H3fkLE|Kh)Chqj_U+FOx{j{ABbgw#;LtXJYRzudO1z?R?h
zEfvSOFAV(OOK@^6uoz`)qpv<^j>}HF&pX)lXnZ;KNWrE>Z&_-IM_E6{KjhM8H)abq
z36DElw)$riahoGFsk*N2z*EJ-lCergsDQZcQw2vFmz%7*t4m;F^&z?1saQcy6)#B>
zeEOl%y?5|l>hmgq$Ko|Bth0Kn`JN#UDiF&|R+%ihg0C4;O?_T&gUB;3Jx1tLMA`A$
zc945;^OBQBi{9HpQ~W_mY2yALY!Y~}67BIvXaTJfzYB+A+ge*U&(=EBjJt2dI5g>?
zb?)GlqVKi^ag8Y|x6Aw$?J<ki1hv{p*AmBi1U4zL#Z}d6N$$uqQXwQV1W=)Juq=)Q
z&j^tvaXGtAExUK95)LF0r8v1y&)vzNN3ooF=<y@va$Qa+Xy66$NMNpm^r}KA(lFDm
z)MH+dyf*nMR#uXF2Zep#e+6X)XW72MXAaTbUmtduOE|t%=ct<!@`t5I%8TW(Vw(sB
zPfJr%zfIutPL-pdM7>k8_IM6#y`iTkpxpRUIgsFi&nRcU2p=}4$Nbf6!sL(Jzk~I4
za;&~)#jMokT6aT)YMO~3cU~vo`g5O`axY?9dL~|cN49$3uInGSCk;58%O*yHMkQLC
zk&8!kY)<}M#<T7H%BfBssVS_gz9HGD6>V`Q&jCFd6nhKN;|YjR`-@Z);Qj;*@Yck}
zw;SgVk2D!;EEBA7(+)J)Y~+!O5$-7_>YHT1qRtQiJFp<lAgs%IvaqC3s_Utz;9F<Z
z>DB<Y^wLtof2WfdawGc$n&Vyqg&sNXx?n}j1E-Vsg;TQs+BGB8O?>c&hg27~vDWNR
z6F~TP_Z0<ns9K|ed-Y$s=*8$EGwDK;-bqR-B35s;?E0E-%&dR1N9@v#yH0~>Mx}wj
zl=OW#XgGsy@xp4%w=OSB!C7`Q5C+)j>9lSA9~yF9Utiu;-3keaAX)-olkT*WP7SEr
z#8#)q8%?8i8b!SI1<BS+tD6h4CzlDN#c$Ro0&nC)rq!Gq^>h<7INB4r`V&8@T96mU
ztF7ww<!4vZp~`2NYeY~gqDKY!*$5edz3p=wm7Z8CJVAVZaO*UqYxB30S1WOf_ttht
z&$hMo0iG*yFp-yaZeA(ARPv5P``@GJmRLrwk+)DAxQiK3LPU=dE4{=Yciu8_`lgV0
zBI=WZNxHt*p}Ob|(l@G!ADv`Pq;;7Hgl<<|;WCqH$9>8nb!)fK5a?HXK1<T89m8c6
z&#XdYFP==43b;5fl2U>nN>O{1`}>Xi)>E4zcbo70pAso9cmoau#|Rrjf7nuMDx6f{
z9I`TJM7v{XYo&ooD;R*9T7Z%;!mNy|v6k_vD-{_udEWkyGqkdCdDLQXtqflRPUChl
z?q=gv7!m&JaTe3;u(>;e@9g5m$20etDPs{ZC&OSgx+(QW04z{S_6`s1)j-cD*zu){
zp-gUc1pc?``!e`qblUN}ORK~{cdNvF_dsowV*Jqk<WdhOkM6!IUXDe0h{ik}o|IvN
zS-BbW3TlW67YS#2bUDMQjV3Uwl%JrY#LNL|%gs$~K35Sqa+-dmD0--{<e_Aq0j~)y
zq=>}xo9%(NQGGjtSiZD1y|4iuzNXPI;_Ub7=uiB{{&y&^hZL{4XTo0JrwV8^i_=K_
zG)JW|&{p_Yut0kVuVY~M%85`%8av7lZ<=W`zQP!*Y3L`As=@q?>H`m3BD<?};yfN@
zRC+#uu1u;QsQqEJReH}xNIFm?&|{H_t4BX3Hlct-r>uiV`?aMq?GLb1IJoA$sd0Tc
zXfa(ER`y?+cnsImn??5EY;gcSz3FmdiZA`%lY<y(7e-Xq?9(7)5^t{S>JRdr3ZAYe
z44&61Ewh!+parI`eOOzP@)CK-SCbv<^>Wf>au}>{`olj<?U`%^S}L=Y;yZl<I{Pns
zh~SI}l$bNvE^K!<E#M0C)yMx?9~x(Fy~7Ig`fJkHqDdL8>wO6XZRJ%*WaBixWXVzb
zQ+(nY(q+VMW3cJz!uIaV_?F?J$ui0Eqr9gkjO+0eiubc={E}x#js4b%%d7#Z!?XOf
zZ&gS;oMo@Blwnj!JS~dmQMp{$6mjzPA?RJBWFq}|N%P%65ZGCcs)`9U+2RmH{XPmb
z&gqs>w~D8Tr)Q(qE6yl!A6ReY`b~kM7ilSc%4)<+pf@eJAB&)@Q5t{|ks+<Lq5sTd
zDFO^f5pAEHy`XFY)*$|2b1cN&<2b9xH#plgqm(RGoorJ!hH)jtKqe;cfO=wr1&%9j
zQ>P7Uzcd*M3eT<;K`c{mzMH20*0U1Du{iyT<?W1Kk%ZfKIrapN1(2tJ_atN?&kEVX
ziwvdlJfqL%lI!3GZE`E7g!Z9VC;~9|WLBmt-(50|{)@0rO)qD?e{T<|RB-6;>F0T8
z1Iv71mb=NZC(1H>?y}J+i15~rN_zAC@fONGyc{2Y6@;`5wSUg8kzAeou*~L!*=?u}
z?1MgrN>Zq5V=Wp*Ws5C-3xzp+3Oa30Ib?*GE>(`YFX`%2f{f+;*yvSSW6<Lch-;L*
zq0hIn3BiND<T#qb_5yLs953a6Rt?6znD+Xc=A&i!-P>rVP^@i!wM)~-Xmh#Sw);T?
zdLl2-Q9k=A7I*l!=d~!CuqaQrdt4t}-dZVjKV+Q2GL|j0;d|Xy8t#3yXf#deBLl}v
z@|YMqgX$YWV5OYc!QQj?n&+{6r56H(xzQsk5ks~?>KLVJ3Oop$1q|AJwPK}KiR!xI
zy6-~3z=Tn^*+L(p&PEcuw6S3+k{fT4zI#Bo)aaeUOZ!IkG2|`(NppB{Q()2wBdNlB
z6Nk7#vN8Rse}w0`RODpRaPKv6<Q;}&KY*oOjUMV*cmVm)Gyw>;1J2-j1Npxp_=2s{
zX|a9sYxz1vC=`vMw3Su5SPVE~8!Xaf=jC=!D^{H@tpY6TAL+}#)Jq-{OQ|PF<m2Pj
z(3Vpwqc;WWMf^R|_*k98EUN6uZe;V0WK45eVFP~ZxjG;>e{wN266@uJJAGy``%iSE
zwZsH<Qs=E|V~?G30D%cs?aIwicn=g{jBZ(q`rP73%KJ*E63HUFOk0sB>@<weX1HJq
zl&Hyxnz9R4(_AkjK-#PEjn;1I4(W^GW@8F5Dc?~k%DkuHV7PDuk4W_U!cAnsSU|qg
z()4~Tm2eRT4S0j0!)Ag1hDe}^Dg;vbiSXHt(UqtX#}hu}!rumfDEb5=Yxm-qcmJ-U
zt<k?&zgK7_*R^5EX1?iE@N$=jRRH?h3E1z5yrWYzm#N6TC+dEK%k`e?l=z3#;Lv6l
zE!bt+Y!=UiG<GO@h_i<5o00dPd;Jvpfe<+3OJj<-rF4B?gV0cW#;_Md8_f=O6Xt%y
z1rYKdcv%-&qWi=i$~6AxQsL63QS4JYy@7?o*cT)xobkQEW=34Xf6L7d$L`3Tviepd
zGP<&F-`^yKsUUvXs90XnLd0Yz>J2fK3;FY5>F>@(n(^~|%iU<tlNcLrB*Apruu%R$
zkU=fH3=)Ovk1nJ=fdM+%QihvKLu8RGC|v(EL^g(CuQ@z4ZB<A__S}d*Yc_k(N9L57
z^q|g?M)dxE@(xlP5XD6iQDqy=c49=j%oRU1rMsw|O(7aq)Is^^di!Ld^bN<X#c=Sd
zrn|&k<Wf%LcawnygJIt)_a{S#>S32Msp_H?ifV3FV4J6#qKYTwYOh@x(OrD@IPjFy
z?6k_m*)lKJUgkht;hNw@Jeg?Q0bgyHqp#Q$AHnY3^gi4=r1VsNx%=JwRZYp^aJ?iK
zky%o;vM;o$ruC_Y%ocXVcqeli2RQ*h0SO;*<2ec%iDDE*zea>4w!;agI(w5GmliA&
zOMcl;@e-c7P|HbnKvX<6lAjDOJLt!Ig|Rkicv{i4AhSSZqs(D1Tp8C8MSNz{!Gz+B
zkt22T4OsC!ABNxi{;VDx`Y1bHp9e`~PpsQL@4RN*$gAf9+vsMX9E<5xCm!Jy<9X+(
z;~_BM<t8H7f--32`1L&$r{t1;A{?7-iH+g1g;%TD#a$9Nr1dh=iC?KuE4uaZ?QohD
zD`MAbsKZxxEDbc*W^{pqDY8&$-gOSmbBBd3-1EMiZR{NM+ySH|E04q&ha|TVq}67&
zY(Amop+k~b8!UuiZy)?>9}G6JMGQWD@~;L~8YZC0g7_ShHGI|#t7t(@u)HLkgbTSG
z^!|KCxvX>0OGZ~tjCJPIY<AO%VHSf*Ss8Qi%FL(F5~MAxZSR-TADMB=4Wq$1Rfu#J
zqj6LD88$wW`;>h#hh?y~FJI7FV8QI}U!SZMRb*WpI0x!#*SAOqf;=<|;!%`Mn-dOz
zR;*UBoP%vVXdW$kOjxwkI50<p6IAe|3JtwBQuNN(6W{2CP`N4vw;s6>1md)p&{CWU
zV_lLV2kdrF#iDZk(vf&!E7p)D;Eq9TT(bM6dA_u)5-BmN<iwEFgYipkUN<Ybp2|AG
z*(gU%@W1+cuaipSJ=X49!Wc&qninN)LAQj!YcF$zy#chU1!eyy)9&ku6*JzSDu`9J
zL6G7=v8SWPNJGiij`fa%tNygdXp2<q#{}-Y+bXiB%jxSiOS-uQe?5{_Kdsf&sYxVl
znF5dhl3||R*Y>nCV6WCQEc-dO{C#tcQ9dg5h$yCij6CLVr}iTul*e5QI7izC=gI>Z
zR&z<)H!(XxvW^o4zZ8=tR89JV!etFAu)8&WizTGDWrfrxLCjXBK2IaGiE&L0qHyS$
zHc>`I39^-#s9*lqulFn9w}ob4SbO8$o97)BI(xX8YHTZlI(Z*SoVoc3h$Y;Ef!Xi)
zuiT^ptJ=gx;>8%x?{Vre!?(e2Y~3+xo08_INwUNjND}wePN(@Cy3eV=0^<CGOt>?}
zs3BVgA1eE;JOYBw0x@UXT#J%}^)vlO?^w0*PaQRg5xHBzj|sZkj2ak?sqD7U1KFUZ
z*|3lA=RVX8nD0=6_#ackABd(tK@aWLvNl4^eZF!iwj{0o>A=Xzvq?+9JQ+h8@+4{U
zE4&}kUf+epkOC5|UC39@{GjHjOn!gpt$$bn-96#!w(({cW?wd-;W@cZs3xayEtnO8
z8?-SoYVzI1;i)tunX5}gMaMY>ch;qPT01!kwCGKZG_?Z#8>6)YA#`9O*6iXw)%HlU
zq@cITG{sF#_<tBv6D-v0ce5s;YX^{VQb0F8WU`$WOZveC{w;vB$39WeNtGL|s>HRd
zsOse6F(S_@99FHJ_R;El@8jsb(Hh;PLtY5`8vC&xy&i?W5aLRA!stR{IO!>m)iV<x
z`z>S07nRV<O|RG%f(^PHWrC|dmk)S&&CY0HE38s5@MSdUv-7}D9#T;pwBR3n`Z%X`
zs4;5Jfmib;?>$j@R>_ez=rs*kTGlq+tn7!Re6`znb-^w><7@`OJ&;z9-(Seirh8Id
zKYS6Lf9v=8_<al;II!l}A}BPBXA;Y&(*b1Rdw%Z~;(DooOQbR|y0ZQU-h}25V_i>y
zZt;_oH2cZwM*`%ppec>~WNA}2<acS47M(8-+D=~dj31DbdAiIS<ol>`fVwfGl#*se
zc^fAsPq5#`gNWGSPAj)~0;)+^i<lEO?Z?4cy_A`>m1T$Nif*cZSL5j7c78jtw!ygW
z-d!a2vM~;k$xT7?HB%1ZwJ`E1N31s@FLyi4Wm)->G6x29rRQ)`syuCiir>Y<_`O(5
z4nfFzK=u0Q!$(c|wUmz@5#PCOHrd2Z2y#|1w9h)bwLsS5DW!`s5=&`2)GvA`Ld|rt
z)8GGL!`T^}UN97wk$vOZIy<=adX8DpA*4wQ78(W46<fL}AYF;G`ykm!50w&Zh~Y+<
z&&2nxfGg7YW%bwRVB%KRkxDHfS7m4>j1AG#n%_{-G-};r-pIgFuSIwktBN|uYCYMj
zuvB7cG5Pm$z>B8Ez^5k5O0#3uuuzZUX@{Zq4~?}jdsZWC+@=LZ!-})dn2M3NMu;s#
zcI_Ks3M>*skGLfIqHmSyn*<x@9w8qVuLM6tVyu^`EocW1d3=q`@QfWRt+|-z{asb3
z;S7t6At;eh`JSLMg@s~ePmFW;fx&u90mn4>4I>$RA=@XT1ZBsycA5~|rd3`=^$YxK
zo(Y!IUGQmr3P+Yzi@e&&tFT+ul3m#s2u?O92a&UyXP-e6rYX_0B{RF@{Kj|3y$S)r
zXXHGhA3S5Am6MM7cL$)=hyV1wNx-fo_yhSHTc0d_#8039l$d`F9Y7ZqWXHLmXvcSy
zQ{tkz!i0-Fv4p;GimuALylIXXQ)n1{*79?_DEF&9pHG-mC4ri3I!-Y<3Zh%!)EDfM
zqt1SPWnCn_?+|a<MrqRi$qk>_o@Y=0++}YCDmYhFK*rc3v^f}7r%yUeDyZkqhz8*+
z3<8O?hP(!NsBPV6c&Dk9uZ}&Iv)`#2|JLwuIY^xN$hIEO^m+;X%+#u=ljb^zFxAa*
z9{eCOzJ8jC{z$ShvyPO*y_wgz=%$~bDaYG1p51V{Er9a(Qf&H~89P>m$k_!Xv8>8p
zj)o_=zLu1#>7xs0h&V+Soj*`6LQTR!=vI5;@GwfE9_x${xYks|gX18(YwJQUPMRRn
zK30=#^Vvs26A5Ke%GLB%@4|Vo_o7?5Rj*07;-P)-ic2rQtXj6DqLe|NXH;`07hp6@
zl;Ci&|IGm@3d^71)`_^D_<JA}tl<=~Q3MeLkgT<H_x&+dCmUGT>8#!;?RLUjkig?<
z=W((MiSwf96-ojfI*77z?s&Kg9t^{xQ;q%GdY4ET6bK#(aG9X%U85#-UpUh3ss#@~
zsbfU+J=i~tPTFa~3ADc`N2gO%Zy&zR`O3WS@rLe;s=^{n*HM@btMqH}lp+#0h}a4B
z=V;aC+wfvlQHK$do=u{+hD5zn_rCSZ*LF|jJ{|QGG03AZ)~7I$f{o&)%0ObpDBv9K
zdxe?dq<debgd(rn-q5bKD32)U+_hrM$;EN_@BF%N({*R74k&^-5VNS`e|r{gC~n{G
zUOI^WYGzETP>^Qus*sqJ0Y)%lwZZMQh+#4Aigx$8pC@9p5bm#8B3eyXE9gkgl`q_P
z;&Qwh%El%{R2Y7W6}8pwyxt2Dq>AW%e;z2uD~2r()7CnXLh>8UceotzcdV4Gpes0)
z^Jax{aYvx{gC`jUR8$cUHLpdc`xlLiU#8M4qtmq7qn8kjgD-PvlZL!ie!8fi53Y6i
zo@j4i6tjYG%6lTixaL?2TWxTS=JCu5-42c!LS2%J)<k9K@mEcItv?it`+jm6*|J@*
zs=F|f$e5^UT7XIz>l_)8rc37bN=XwanyIxHQ7PgQL`k_i1h*Ww+*Pr5-vx@^1&MBo
zK4HF6?*l=A($e=k&MDH!Q)1y<`InBXL-wK?oblRn&8b9cmIfxW0;wnlg5=CIH(ql3
zU$Z5MgPc*w%c^%b_rOXCG-~22{J>3JAaOXUz@Wjv-s?cSXOQCcBKN%a5z3ifRK=o2
zL`_?vsF^@K&NP`mM(YLDt=0NKP9B#IAJeCqGf-qW2`gql3$$-WFS|y=K*@e)@024w
z4uC*cObKqjyPAxgg2d2F)x1hsV*HF{p)cKNnKc_>`CYI7zsQ^x|H)uk6&GbBr4s(&
zK@i4j$E?zf<Pm;PSC`0jAnMJ*id)740=KmNELwq{XikP_@>)Dd>niiK<0Iuw!7W>%
zbuQ>fsnhrB+Y$!bgcnaECf?B=%?m_S-iozfx)GK^bQfm%U6dGFN_5G6oq5u*`X0Y6
z?0O-;yQFkjRr4MVdb}+hO}p?KD_M+#=Y-sRFe0zzyLo0OofdDIbfo!X7+fP?6gFn)
z2bskR|7Ay_B3DRif)~*_rrmK5stQ2%C^ZyiPc1>y#m!woFpGX3hGi|eMP^4w?7eBf
z{oz58Y`&D&FUY(8O}gqzuyJeW?rEE_XY=k3fMXHQd!n~Fq-Ra%H-d?jZJ7Zo3e#v`
z2gHxhPAV>FYw=B8G|o3hkW_<y7R<h{q6xBHmOYEc>6P$1x`BysKRn^w_EWbig>3`T
zU~g~t?<-K~#@+Y^pm&QF(G6h6H1OG~Lc%RVsM+h|8QxEBv`eLcD8qa66D6wz$b}sA
zD^+Y5Q5rj5+T}}X2R$ZNq#8TKFIlQLd{8FB!%5j4Yq^65kh@ZjZlN=Tv(5XB3poQi
zp*OcRHwt5xJo;?eN||Bk!m<b!1&xl4c+eFUp{>7bRfcskaCBYv50SSk(`*LcLza50
zOg6Tmw7+);X2OW%bAyIp9_d^UleYd*lb;+u4)c@5fIR7OqL{{)>}>L)#u2PZ7+XBD
zBet@WCG+WboVp<p12V?tlpCsY9Il$0^iOb7I8j^Hiey<8x(w5%b?d)))75JjIGD}*
zlRcmQeS!vtaVY||)^=Pr=ad~gSVQfsZKIEi^3i53S4<~`CqYM(Ls*4J8!b$uB?jL}
z{@p={vaa7MsQs8vawjXOD0-+?-)stQ5JV`{2zNwR=Ya&{y<7WI@-{}4EfjC)nmhn1
zK0#N^#}uhzvD3-FKG(hDGjsy4XBpX4(+^sCsur1v+Q}`kR%ZZ6vz+9y61m9S<p~q*
z+MB|_w+lct{JD@($Xv|=o`=d|O7OWg=YMgAzr6ozXe)80eUCMP4z#kY1~}y|f1CCV
zZo(g#OR9{#uu@2|!NH5TNE!^4jOrN?ApEN}53&eU>)Y*lK6Ql2yl!K*JzjhwDm<Sn
z0fTr^eGR{kzK1Pk@fvwY^t>$i2hASR#uQe)<5J<2^k;ly9gZ>5-paIzaWCEOW-+=U
zUrFp3;~KD<e3zL&+PCujrK{gNWWnYnm=5bS(EnJZw1Yfs*&}!lKf}2cc+RR;^p-6?
zREvI@TDR^Rr2lkm#0l=TCgj!o9RQQ7;`iI<ob|nux4SmgS6pop^H}pC{Q8fx=nUjW
zD9f}j9ar>$99ossHjgqBw^Ah0+M4w6(+c&gsYiQ>ieg!-DhX6@(k%g<!9C2^bud{k
z5OujgoWO){;TsZVejS}0WGQ@ZDWU)Q-EWuYW$vt|*<>W#tcq^3QI@Z#a&FGl!%p|T
z2wj!eG+o_a*tle#2^wJ3KODx&jQV0L<sNM8W12_|TW*e-<bl5_mty))$at%adXj-9
zX$<J)gd^?G#{$|JLPV3E$E8g_#94EN1iFgWfL9S<xi>pg!IJK0G6M)9G_{?VazW5}
zveB}^KTyOI{l*{T7Be6po}Fj-N--|pmV5H)+&|yDkifQw6O~ugA`60(-?|8V2NiB2
z=E(PKj_dQP&}7wo6XK#}%+^gc#P^uzeMSoKVCNPw)wYd+7ghtKtew9fu7aa>WI*A=
z?L4U3sXBYaNA$cH7M+)?iVze!^9!P^l-o=Aq%b04{?Cmawk?86Yuw+GJ*JGI${T@L
z6g6{GAGY<E^YHYqieEeZH#!0^PY3gFf9Cci3FX$)foJSacy$w3)Rd~~aQS<*z_@l;
z-0`7rzpCglBsOykvShY3`_GU&)`osL+Av*s<2ku>FT}E9(?XzRnJ`fCu>@%$V!O@%
zv|V<z>Em;x7M`X#dp>{1d#19$2yuEEi;zXnVz_sg;}4OIVg-!_H*+4UVsyft!Vo#I
zRizSMIq&8d4suRXMU~|XNM&skEp;9PPf-bX$6y=OV#!nAOhjVr6^vdcNT6-4*zwa-
zUXe@9#wm87sl=T`!660kjO0$HPnH5g5&H#v@D)+XhYj~GqXwhe;zl2~h->aM7Sp4e
z>_2YPAh_i-hWJ&EFuF~?gZsn6SO6!DwG$z^!A~F8Y`{WlF>4cNKbs8(A=9lFRyJY#
zn#kGX7LnmD9QnL%Ht&)>rh~wo{y5^yni>m(0OYf<)0~GqK3Z)*5oLkECt8L$@rdpU
zK`S#bR-|LH^p;$0KBbU!AuXC?#6&TL#K`}``={tDURT@iWYVPRS?PSm5i^eM2a}QN
zApXgX&QP&N?OK3(fIyC|*XPpRIM0H6Iy#>(!hB@Jd!_$M{5&a5bjp$OnIePD7LJ~W
z45sgx?FL=(^9V4+Xal{J{)ao%G~|wIu^kGvAD4I{=egY8*9q>6h%#Dd;5!P*!NBlc
zn8m00#n&n^R$MDQ?9I%(P7~vn%Q^e-wQk&1<4XzICvG`pgV_g%^-dI{F)=s0fwjC)
zhk8Bjhe5az(?)UP7>w?g`c*C>ZK@?%(T{zlGb|Io-<Y-fh2`;Uop2*6ilnR8%CGyw
zw26v|S%m_^YW;S!%SI#h^l@n7qzxl#D365;W}gki7aR_6{5#}=NCnn&PxxwrZdETS
z!uTvoXXtk>!Us*(XGc{9tX#EJAce+z|NccMCgSptis?Hx+_1{9NcW7#;Sq7yEzif$
z*ODG`G!p2)pSc89w>;+p@lKxHl@9U=!B9V_&j$)ohnp%kwW;<(NJ({n5wUdXxQ~0f
zWnmLE&6ap>UXo<N5k2!}o%f}ghh9$)-(_RFb=P|W2v{5Y1HQl?LW*tPtwX(Tkm{4L
zPNzp&*iqWBS@7&dJrgetujBgWXu#E^)AHq9ZD(jbjdex>(fBD)pQxj7ae{C#M6(<;
z!g+a-cJJ0FG^Y@=ZRHr+9CGK8pz)Nin&(!5&DF8PA^E$j$F3%}8d?~A;HIygM}g|G
z<UqE?5~_cj7OVb)0bQu2$3Iu`WSKn7@;~D6z~CpG1U1UquCJ|A(ItZ=m4jR$52HC*
z6Np;^=ew_0Mq?VSqc%pj>*Je$S*{jqsC3psumVm*3gtgOyc2Ew`=E%88pI!~U2yOH
zxv;$YbC!W<yHZB+U6dxL(v$(NBEMn79|D*)dHK}xR^ZY&KWUL>*!;?KqNl@pnpKMH
z`Y6LJGqv(`uxdiS3c%}3Z0UJ%pC@F56h=Gqh`;qt0b3&=l*=`*3yWA-z4K;OxU=4`
z5G?2@s_UNhNY<L28}t1_RgetO4FNu{^XW1-K@{;67ZDPx4GYPjP1`mw2kq?aFmEy9
zgo>~qp^4XF{?@`v$7r>D_J%&VY;!vsR8;J4Zx@sw;BHhyuYOLZ!kdmhKi-nQl{q6V
zWTx&5M=)piZVt)%2*tWxPmGAd+*!zus`<R{=c>iIUoRqArpuhFP^$j&@*m|TNc1k?
z0%#aXJm7zX(J$QYk4d9Y@1z4)Cxny0c?Agd$T3N?4rk`%{f*$pm!#q^pyZv2YElIV
z@yXb4I-~=TwFvkR9+B;ElBLK5+dAT`U(l><v)93&Z6D#UA0C3Abi*J1^7~ac`#2O~
zG8mE0FB8DTn9htaH#!*0+(Qr-3AWo8zdzr|Pr{C*(IZ^vxm67>6HrUZM#K^rzA9D7
z!<&BUfX}9tW9s(3MQ9Ihy(a}1kYpsaZnYqpwUk@3`5IJ)w!LT1l~?MW2xxHxMbG(6
z_P&rO4u{X#gj7mBmQ_^5$31SXvX`retR}UV4|>z>r*ciIHsk!ARRTrbiaf1x*ebI@
z%e6W#<^J_Uh!VK=rh-h3TJPZql=p0GaCkJubTt)9h`VcKXU%h0<F4GS>ApqXSU2WY
zD&TH0wFjJ;C^##Ez<q8U09>3cCt7xfE01~Qcb417zucMbVybvqBK~8vcbHi+RJq~^
zIty2dU%SwdcjK|8#yRC`?cuF<(42cyH{lH_fBuBwbnW#$4sbZ20*ym37z~0=6p~8$
zar*A1FutmzCZM@VvT<;^sNTlb1~))QF~N)v`qw`VtNHPxB)vytH_J{;vVI{gOKL=l
zktROC7(Ox!9lG4lG~RnX?|F^ApTc~N%=Le*B{D}r4D-N;XzxN2NR>}P73ws^3z#l1
z*#RdTX0H&%bzfkKX`T_jv%>3`P)t9e2T8>4t{BMJWG=sn=IFNqnQLkM@rhiWUi$t}
z11Au=g28uKGnMa5Y|{bx#Tb6Yos)fgME~Y52Q}1Bh0E{JQ0n%3))Hzf<+sL$+>VSh
z&F3SDcmENS_7mw8a8TY#p10AN%S;?*txPLye@Ew+-pLwdk5;w2I6f3o7h{Z^g*=>E
zm7NK9sC|jfv+i5>o3tRCoBdfbg@$Ua2mgvUR$|42dn#s2F>NeaFe|rJjhSCTe!!T;
zs;bAZqjT_UJpxyyQU$XNx4a1cT8uHHk;EH`3^g?DxYGGw75+9-zDOKlVo741Y&1kL
zDso49T0?9OBdcjQ5!-SwGI!%SEeeL#coSX2>~$7HC{_-+uMt$WcdmcPtUbxnw$X22
z#@Co$=M?DAq>dqITBFjpU(jwDv|=?y2fl(OJBXWr{RD2P+4mxL>*@srLc;(0)xyHt
z;C#(Mn_YdjOFSc1Kf0XjFi~eh;x0b^uElYqf?~O@;aO)4>j0TU-<e*4%^*%A1W7a~
zzzyHq-7T6jd5A70#U@(tW-7&sUtLix)_u(DwX)IYNLSz!)-O!@G>%hMnPHQ^)|SwA
zib7@3tkDRYnjgP>Po3yIT5kYh)lkp=ub+sM;8BJN;(F$Hg?0=JvpR6v!_h2NDKWE_
z6A!Icr}Z9z`X6|<-mjvWB_c?jvRk9-sV4tdd=;;}9zc<3!Ka-_8q3bT<1D^si{ZFd
z!a*-b<CUE+tE!bqPpuQ#j58YQ5Yo)zdJ1?bx>-LPnlBNf4wBa+HeCgN5woE-)8s9S
z!dk=*Dfq?n?lSN1F>+evfSG0W=F%z7(}P<}<@0L-S1y6dFZtDuEY+Xx@kq-PH^uex
z8=jp8b{(LTe3Uc)1!Rxaznj;!@j?ZkTYc=2Ir`H^w-y-GCVpKMGwdK9P&t*IXn5My
z;=)Z8U<_^0L@IINH(p5;{z4u~d0yeLz`s5}yYo&LCBy-f-VlS%Dysf#2k2Tlf03au
zu^lx_fB|nVY<uqQlR|W^8>MPnSKFT*2n$zCs-9sT!7kbzMcF))R<CChUsKG1=c<?{
zX1tN%{;c9=6$1(ppQz{oaCBdPThmo1>uRRfgiO-O^7Q{}-;biCT8;RFew%dSGhs9%
zef(qA#$_I;N9-FMluBHgafic<`6`O(I9l*|V;*yauE-#n3QU8<KtQ=Vr)RAna@)p0
zb$lsw6H$rwG2J?&VUhF0TIV1k#cLo)QC!0sKak$0QU165FhL=WF&K`Yt2{n1+5J>j
z0peR1JE$%g%MrS)1r%VqryTi?Or7d`nU-~l^CK`kl($<`X1ASm1>8QeaO&pjUR%qg
zZV5$%x45~kw1dp@yP6dZ9i^JiNY3;@@U#5i%Ww8ww1YQ#lmH+0yZ(26nx7srZM0=N
zf;P;}o}C8s4{jAjR-#o<%H3CR1qRqM0=h*r?Sv05g}?bl!530z%LRTx<0pS48vk_h
z00K8pa6-6_ombmMm{|A;aQ<Zkz5O`Tf>wRx8W%}I^UGD}Ms-sQ=n8D}gy!&Vp;#pU
z4BhW?EI47oZe|fXVehL^E*$k;?#wGSwY%U~jRY2GU-@A7Nf%}BSyQk!TP=3Ufl!}O
ztu9L{Q|c(KTiavSnFVRW)8qJ*Igq>Pzw^X9XGBrhXe8xsXlQ_iUbq!5gwDHm8l5#U
z7a5|yQ5BK=T2DiC8fo$&E7u~#10mL)vM9;~M?Zks3rZR@z4RZe_oAnm_zvzMuZCIE
zTx^a`vp^G$=(ybuBKHZ!+`ilv4Uu++cm`-&8>Aocj-J+G^%hE83hWA!P~d@B2XT<V
z8D-rHKg`+nRT9_7#g8^ZBgDNt6+GGMDsvb)j+;>?Burn3ar%%RH;yD;J1Hqkbt`+K
z<w`mPmXu#e)tDby*@~L4i?tUgA`XG0$tIPi_Ea{TW$So@+J?EK#B)J<HNj~6*+8Dp
zB*CPUhv1fBG@+tGcBb)utiFmeaNQ?*qCgYA6rWc7F8JY+B*wM)UzmzU?W2;RyHJKl
z=WRWayT+C<&<#eE{I+8FSX5Z}{lMr}ww1Eq<C#a=vlEeZ3#%csl`M6;+~=cO_OE8`
z5gLq<KymFjuvlG+Zcxh?8v4CD=$mELe!PG2=e}dgFN~)~i;OaVsI_?=TMW`nx#Ve8
z)ox1gc(vL4OdCM=f+#HXTkLK1i88V*Wreb+FzQ;cmCQbF;avOe$t|<NdRqQU-TKdO
z|GHOd2x4O|vrKmQ+MR;{G5UvczzcQcvZ~-@NyVRIk|gW_O!Ns*l4gWEw@wroPfFDV
zATljhvEDt6vJxqmXk;eC^Lb-}r3G?;Qg@h_`EQUcI5sT>Ye1wn5Hsax1}VgAQC5`h
zcOTn#Cr|sEA$K~pD~2DR5w;rcGv5r1n18hlC$ISY+DO9l)<O7_uSX&CIBQ{ob%3hO
z_EaiPnBBwk_*^VjB)Zif>t#Qq*ZZ}TM?bsR_r{!DKRB?p<teqzVLWKaPBl%#p4ib%
zT5vyRh-+DX9KC8SsD&hhLS5y*hq|tPjyYt*=p`~Z)9#n-oK>Kdb-qYr4{c8pHv-Ci
z`n+wO1xm|%z2ytk|IJr>rjcb{tmnYGmtPv2x4R)r{)D2Ygj0D(>B~GvSH*V@bWYk|
zBGt04PF;*^^j7TCb4Q_#gGB8%cBQ#%K9J|Ru7!(>i<B=Z=DPi$(6mXFjZD9Jl;sGH
zdhl|135&$Z&z%{$Lx~|4w9x5?U>)SVS@Mi%SHN_EFLF10%Pyy?r>v=4g^&sl?x;DR
z^>TB1FH83bpo1xV&)??y-LS`!ubR3)An!5jCTwO=ktx#XY#s85uyJtp29PkJP9TZk
z*dgc#`yI#{n~3=BPaDpxIj?a`Uo2`W2Iow8d0LA(+TYxsXo(6Ag*Luca^-)zW^cCq
z=|LL!?NIRBV}S8?g&}i3HUAKTC<q&>ciurX&6DcDt9E;DNbh?>j_$b4g1y%5|5;CJ
zUjR;2nj(lD(Gh>pVkSih8H)~*2HC}?xrbbpxYPMCZ`<I3{O%U=-s&%GHe)|x6Shaq
z@#=GCHQ=<yne$g(XS6#h&~n-ts%zqVXA~q6NNz0~O`}SQtECcHx3o=I5TVMS;T}`I
z&7X5=M#*b(r}nnDl~_UlWRu|%)75y^ZcAHZ{9`h{<%3>;#Mpeto7Y)E4;1{*+`QHk
zq<TGT?*D);&?h*nQ9CG=+1ljkQe=tB;9kbx73@ImaqEDb?f(H+L8!j%l~VMd+V*Ap
z$W>z+_gIoh$os83&r?}PW(-ATsjT=W(4p3j3TPEbz>xEvqTKNC(IXz8p78OfpK|Yu
z&ncIu93IS=jmJz!Gf2mrT&z(^4m!arDQana7>`E#34Y`7zwlC*_QTd_4lrczTL`B9
zdLN>BLyYK3;Q}Neq7$tvk-@|BI3&PFn&Ad!vMcxuF&yvxF9+4S-`Va$?Y1rf*u`+V
z_j($Jwo9@ak}SJsK;h!s4%2?0yZvd1TjzfLbEA`Ydts=pbG`SN#IRbe060EAMkz_^
zD~f8%5gakXE;BXI+F*1-44%gO5SmWoe{l}*q7ikMm=Y=z7?f;KttIzJYhfrwS(1fv
z=vkKH>zd{A0;T3meBNMuZ8jV7ETOJy&|RmwlpPpi;-R}VCC58s-yietJoE*TF4$(5
zSqL*%z_zdwHLN~k>7zdn2_Hl2qrXA>f=RdEJBhVc0bxRzn)4oIE+ZRsg|GP8jcALo
zH>1(0$;at3^s6<?NHkiIP>9K-n#$Qr#pz<XfbM*B$8r~e?6*Fx7#hF$J)T?FrTAs6
zBS}+ci+!r)8SAp9LQ^@<L@5zKydzdXjbm`JT=L<E@AJjIFFC)s;Cy+3c8-U~C#*&r
zW>fH|Yetg^WFu+}DpWNVa2=_lAdK2K1YLV*KwP4h5Z@ur10JuhX10CWI7Cb1oc6W;
z?wG3}Lp0ZVd48l}ga9xBBOacUA;DQGnl;;>tN7WcgI(e6N$r`<Zo3As|64yhx-V6B
zU!dD|Npbz>?0UiO=4kIP!`UZt-D`&W(d|#aMFu336`@bIovFp9U^<)f(MKQg!yo*x
zfnnKf3XDz~#NgP6-Tn_Du+i@#>ITA;v;!mBi|U@&kzxp%j#eu8d|Hh#KEt^-{53k_
z$D@={@IPtO_EGm3a(~R?cNKR9Pa~CI9^cFNy8OI2K11tdh_Mazwfnrw0#ip*xb)m!
zfAB3__)rTW)D6?QLZa;^r?wUc!)mo`m_uX;{r$3~iEZt<I2I1?qj`$XGZeP2NdmM`
zA2;r*))Ky_o8Rs^eWo3Mx7LQT=3@SQ>E&1X@E7k=ZkCj_!#a;P1|5VdyeH2SBD~kh
z$zwkH<P#n|e24>TiBzE4tXQuKj!z0|>-qXS-(WJ^qjHdEDa2@<rb8}BQStAlqoVnE
zzqO;{j&;sA+F)!~X);&QrT(05{qEaN?HYBZR1jjXj-qS&qEv$g*X>8&4BS;R;yN5E
z@iEk%xsqx3dwgg=GHpm(=zs4Mls@P0Gpiwjva65IZ6gTxPv5nfx?0aJV$<)J4~w#U
zyLc=g<tU}t-{0r;*I(oO;+)Sv|AKp8JlKllcFr~0&yc1leG`yHv=3m1uFa$IfVBww
z={kC&1hd`u=zNc%z}?LsHI2<?!)B9VtBRu7Fq=)Ns%oc7y^A3Cna|L5yMFIV*ZrWD
z_`M5u+g&5O+Eq-bf1SM3HdA_1yK3_{!5;F%mS3*F-ntTQqrG^ym0TMAh+TB;q5H}A
zuAm`WLe$n}-wM)2EDt_Z8e&Lr*wSQrFCT@O{&_=dWyl<xm;{|dq7_BydG(F2@cCz-
zuqxJ}{D>mUM;I?bCqBthqbVN2V5@D#a<yc$Dal6@bdq4}3hTiDN+*~k<@uLhWi*+i
zw8A<cO4SF=A(U%qH+W;yZeLQVtzW|`3<AdGx!6^IssHzrg!U~u<~zQ-9qTk&O8WEH
z*GPT*z4t7*9ECa5*JxIJQhR3R)8~4ZOkA+Fak`!Bn;3Sf?A-zBZ^Ye1t{sT(JsQbv
z$NO}zYyaLI%ih&)n(uh}+J*;Q?HU;SZ+`QeY>I-_YR&1%IqT(S%OBAnkJhHDsjUs!
z_Ywl?5rhMA8XHt%utfxH9g1RVo9l$T_DwudW3^J)h3nUkxN+?e=Oi*qtIKmNd~UZ>
zx@CmzwS6>)U2RC4>CfBH@v9{D{$|?6u)6cy?_zhZwhwSA&||32L;IlLhd3wER~0w(
zwQJY-<dcsZ&3=2W>>jfyjCA`W<*~QRV8rn4(P*?)KBe2=tKP3muo`K6halSj?eZM^
zCb!IcxP9k2UVh~@KKw<6SCZ$JBncxEv_xglY1DiyS(XBjW>ON|n1qe9IHmB~P+G@y
zJfgIoAOGwZy!F;sNRS*lDR3DCHy=|JI%clUTwbChgq^5k&8^Ffr9bz5VZ3|p(C=H%
zWUE7$&h1|_w10N>uWu2!^z5O%c?Es$w*N_?*0FO#(#<YabBNQ%-**l2(Biy`Xhkzc
zml(PMg0a)g<()`Ay*}?6185!o;f7lQ2i|)Ydwab4>Z|<x=fAiVA=3}CxWqVXfil9i
z$lnbCR|=&}u*`+`^Brxkzc^f%@wbdQY7tqMh5!M@{{9}<kB*vP2%}@k<=vWNe>{ga
z=MbUnk7w6!v#Slw&#nzJ^nKU)-F~;YX3x+DPnU341!6}Z(8goy3up1SG);N??Qihz
zyFYEvwBxmdFi~4;y=6WTX2&T{mc<~90M(O-k7qm{Hx*NNjnUv3@0zcMF%L80;rq%u
z^xk7k%5*y7t*`wG55BmIU7u5x^8R_A;Zay`ATgwAhSi3>y*(C-1*aElYG*^iJclxx
z$~nff8P;ol{@(i>pPzB-&Mj`<IKsOKitN$Qn3;7<q7CM4t;7pto3$rh)_~XB@3U(?
z?)q(rmJVqwa=g9Awz$#N)7l)rA(zEgA5|#J&=VeFKl(|{mz091+%M0BkKKJ(S3=+Q
zZ;MDiW4lTt+SQl-rtNBF(G=&bCDz{Y9Tn315Dg~Y`%C?}9AMBM2f<D2_b1MLw4Lwm
z?Qw8$(1iD<X^KK^g<*I5Dk?wdL<_-p4vR-=Lz?7B@|%fUq}}%eV-uas(dHW^%cw0$
z__aYBMVbll=EH{%`T2W4<Kg3Dlre0|3Z>2VedAqvjAv^ccFgTia^%yFd0(o(+%5|Y
zyYABnD7~o5Zb^ox5k!=VxLD%U*Aaj>z%9kZzxb(!xzwfS#r2S8IluaA-{R=TP14bb
zBo(6TlnMbU3hR9XClv(>8ed>btVAH|nAC7sTQ{24cswT0bLzV4jB(WP9TAf)bh5>-
z>S)4v-~8QP-ygeSS`x?7<EX5qsx24mg8BX-Njj#gpe#MEhScPY(h+$+#yJ%bD=OyG
z881BlJb1@uwZZ6wG#QbmW0X$uPLpLLP?~a6keQ6Ctis@l5En_L(n`Ln@I6>E<uo7u
zb$;utm4^0o+yVV#U176X!9DFL{&smSclL|cO+R8^VsyCmE&F$J6Huy|Nlz<8yAI>&
zYraFuMxXXSO?zg|r@MIi>VBU~cO|me!Cefbd;k79-S1t3&~N?vEwL%UsLtZ|Av{>O
z4&@jyAezHAIeR*3xw9zZJ9)ui%d(8JC^<Yl<U8N_4&VRFAF?`sAdyTREzN~6U7(4i
z5z~ke2d|}4skH(ys&o>Zoeq@20k*Em(+*i{`tC#EfZ&nE@7{UP8e>w7NvNulBuPlp
zgu|l)zV)qd@!~5lQ#(jUW4tkFA5j$~@~&@>zKU7&nNy#bceU2|cXzBWU)M^V^4qrX
z9sh2h+xm#Q&+Qj`YuE1fd+l0#&HX(c=F*T{oV6l3d%y-EXW*W3pL8{uE=tC=>o@qV
z-~Jo?m;dE|L6VOIDkk`Bob%{l{;i!5lEk<Ef>=*01}r#NQ<Uq#aKJl7U5P1pu~-P=
zbC!mL=g3q80du~kVzK|;fvdrV_2(SQy9Q4X6iRpT(1L$O#oyvMDg+PDQIs3fB*iBN
zuT!epF*~@y@%=|wYbXowl}2YN%4F0QtX1Ush)w1B{qOt^%kwpV`sd#zQ3**h;o{<q
z#bO_0a%PhW55Bm^^}Rg~_x5qV!YJLOThzGL|D!;(A&bXTs~|Ayu>Rz)js`%xpVhW+
zy!%^!e{|P(Jhz)l(50nA3=1!hj=@D%MeRs!ZH^`bq7jr;r`m;lNz-&|oeyc)eT&W0
zwtp?3&k(bS-*;`$Ee#`_KV<gmw;{q6@7MqBstCA#ANqakN+4Z;)Oh_+ICyja-o2Gl
zO|*#C0-<A4$-e*a!w<2+T+($s?yThlB55iKlT==X`XYKOXu$_bHxj!cWuVXOy{If<
zv1*(Z(@xAkx_Rp+fAPKV@~v<GDp{6cUClyGh1k)Wdt66z8<yM{ob$euxx2Rd(pG;S
zyPw_ni=RVt-4$ep{_aLxZ8M+C$PL}UyU&S?!}ZS%<>Q-mM8_mDuwjSRJjgov&ENV>
z{^39Rhn$_A38Hf_!KrBP@xHl#)T*N<*Ui)Dj!~MX64vYlg!Az5kUW=A`#jGAq)+Up
z10~vi)c8YZg^4%MXiah6$1s55ckF5eG4f9QB~cppRMujx<E>x$I_tAjE*{<I<b1_^
zGUjN$psXC{s}1|tQcxPD9eI}X*MIxB`HS!UCEh_@)nwTiYXN0hEtjNe%4@H_hIdXX
z4BJ+-^Nr8PH<66iyPzd(B{tA<nwAN4-?P8gyX~rV-u>QXN?m$8B=EK4DRUL)sDJ%X
z+qM4sm3RBidw(g}dsq9{j&_S3+r{bngVo=>eTEf(?`uxoXLYabA}j1-P`g^6c=Ycd
zpWmIJOG$!T@7k)XO0~faVk(hvYFAa9o}TijfBL5s#pTKA=2G;Z2@vb=L;}za!Bv=6
zV^D%#-5-l6{VTeXm1}C*M@dvE!(=kyt6zPSH{X1d&p*G*y?gh#b?X+3y*X)`vc5QP
zG&VYnyT1`4qC$VjhM31LCeZ)A&)kOQr2G5RZ1VWM->=9xyMVGo>tyKN`uFJ~t;@5v
zO*7s554~gj?wsTK=br~e8eXkd<avH63Mpz{D7Uq?;(hv72xAOsKEip;rdV-saKLOf
zqpr)Xes#w~a$?l>^E&Sog(09d&4&Bzn&a-fJL>_3wE^nKd%V)@9UgM+<_*61{8K);
zf1fWOe2H~G<k7~mf9pA>d)JwH14&9z6uj`l3oI53zPxuIlbBEzM6~|pa>>z+Bi?@d
zZMwSZV3SQmW5??Bd58|iaqj-z)n<m~cW5mS2|3-d8d|&kdFZB|Y>n|iUfM2tdKGQ$
zzGuIF*RJ-jZFcK3lYR$xNv?fqw|j2?y6%SQ9`8>=m)PubrVT}Hbk$h6ob39vQCZ<3
zm)<3A`XouP-m%$i_~esMuvJS8hQ?#a7t&}$S|m#!n20e6XhA_%N{Ly`x(-x+(?QvA
z9Vf1J4x<wiWdKi-h*{>Xue`y(_wW9@>@5}?KYGYN`N#i=zw>wgHi>x=pXYezP+b$}
zz<Brn_1~jAR=b#bcigw5&xW;-?nWAl!j$ueGqd~cNgIAw$a$X$$A0ZHuUy3>yWhKO
zakJU*`@jGD{L_E>Pq%Q`hQ_VSWV)JmS2B*l3(8ou5lubIvS4!QrvcDL^!ffc3Dk}6
z?Hi>OMr%Ae_)CHY(McZeo*Tze(A<RyN$CqPi6PB%Ci@5c;*(GK@oI@HHe4((SX#~e
z;E2L$e)sqOCb>>XQn<LdU^bgl*R`1Aiju@=k|e`>m`o-t_Vy&A6N1^zZG|l>Xz}2B
zd%aKdcFk#QO=Z^|v~<^8cIoF3?TxL!NBT_dX&C>m-*@fRuJBV;zl2chZW|i7erI<v
zid_RgwBftBTk$1_`mw7o{re3umZ5vabGn=K|4-YWG+UBnX=2!?s%CblH~JXi5s{IR
zV^vod3f0|JShO^XD6k>ylE#AdA4rN2d+;~(y8i+f5JH5TUepSNU?5opnpJ`(x`wRm
z$c&8HXZQEsoy=5~i=#*F=rOhD?$HvO>GAHhYo?}p^yrz-QTcp1;px{1va$0K2~E@B
z`uZ9n1oVB65XLeT^9hco=+hV@HZRY^AO>3?XoxY7sbD{mBc8@%qMdCJgC`~y0&L^(
z!J`lG!J|jGxw*za`Un32|Mmavzd{TifBL8Yd;IuEKf?C9pWPr@Qq!=V(^j|-jW4x(
z<h^qh1^BwQQS%WPvM)JqjhCZFc)NfliRRqgig^pBTw^?%X-i?O?|b~yfBH}H*=L{O
zFaPo{a}^2=4<?&>>}-CMa}6QT*+s^~8jJOM4X&;r#fuO*=pfR0QxN$-Ab|!V1%scB
zfzGaq!K71;>t;~_g~YV!ETaHoR*iGG-tXb9#nY=B9Jbdu9uF9XfFU-xdG-pw`ur>W
z;c<X#JleLyVY`Fxdn^|V1jEOl8w78182eXWeuYKX;_Unc<?@mLNlI8z{8z7;oJVbK
zrNAlISZjmar+!xA<=j5erd26a&nX4`cMf;u=h}O36Cmf>cVa%bac#Mw^k8bzu#~E5
zO>gGg{6x1AEj?~t6de)ApCvFSo4P`Lir<*fScAUrF$@D*13Z2Dlq9;rW9WT4MRV!{
z)9(SuIY{UDSiq3|d?X;{&X7<bJqMznm?GW+ybowQDlOkO4#UvH0N8BS*za~&E*3aD
z-QfT7U;dYP`SJx;t0i`+Tu_Fmh@R6@SviWjR;zKQ_I!<5<oPJu_uMljjBl6rviY<k
zaPVhY?nmFF&5-N|D#k4|j&WQ9VzJZ~HB45k75?N;{sdoq@dchge~zoGtBj{(;7o*T
zyUc9Z@53c?<hrq@VT?hy=y3P$-IP>p6l0?RmppIn#}RY$fpDzv`-FFvG<2o;@;GEd
zEBSnqc{1`53jr_&UDIH(ImP<o4%*cQuV25$qU{i{fcL=HPhKLx;q3AbUcGpZ-EN1k
zzy2E5S@<E~^z;NxyM%?qVnG_-mtTH?yZ0`!Ia$KGbarfvKnf}_pdclfbf#HFK-5h#
z3a!=Yx!3Se(<{BlZnMm2oMF6>_tpg6H!&bdBjt5nF6?dEch-Dr#6ki`4zQM;t|vK1
zaQHXfrv~m^f43osT=<$PNZzBR6mH`tF5eXa9FNBwEiw!Pw%ctsX*_!WKK8pEzJ2@{
ztJMnL`+Tgngj-g%0diV9hNl=`h-jNOJ!61Wazr;{C|uh$jw0sLDW^j}q|Et%rgb<y
zJwe+xaL(Xx*yG1P{!yBfhcW43GbWZ|-ZgxHu%YX9(AwvmIY3^UdzZW}H(Bi-Hap5b
z^?SxOn8HzxRc}g}Ff5tb93z{{&zZwR?>$zl6%L03KK$@Q{OO<mDgNxw{tTaf`f1j(
zLkJ|;Bn@P-SY*G@;c&oyzt2Zz91e&4J|BU_VvW^mg$EBFh<OPD2=d_(a}y>w9=nJX
zoP=Gch&;q;jXk*zG|vcFiy?a0w#^!wF$S;}O}D_upZo;>?$ckxECXD-g0%~{VDQ&}
z_bK*+$BS350a#qUdWEKO@I#N|(PMpb3hNvk9AX5z)c@Jp8Oo(7=}SPQbO%pq5PYtZ
zj=-%kjoKRPXWpbWYr-ytBux)BL)Y)rOlOY9>U;FPdL8B8H?{9n+f5T+y`gGMK<EAF
z0nQEhP1=<1W1@w;VJ>=sZfmu9j{l}(Va+Zig3q}<SI=ML|M-9SKjB~h>wk@}zxgJQ
zL;HG)@w3!r$ZTOV0i|mpahqZAx0V+xy2MCaR*MC0wp;Xv1D0J26JyHt1`vGKa9|D9
z(_$f`f}ADXli330TI%(zA$E;<+_oO_`5Io{*0hkp&(s)Gt?^i8;-{8O-*!_1Wo<W!
z56(GUUteRrUc-BjfAo+35mu`e{^h^?m)PxgS-Sy{PjFo<7O>W0yWL{D-NG1y^_rMW
z*LA6s(jJF?fNK_LyEaEXZZ@klrv#TV7MH4Jh>GP%^B>QQ3Q1E)1yv)P?dHu%S$m5S
z);Lhq8YNMO6rvt|M2r@j(+fO(@fycI;Lg1@PR{RP=mS<)*Vr8ccKaUdMT7132FLv#
zjdO68G`yxu@N68cb2uChI6XOq4?RWMC4R{aC?NT?(y6==Z%;RBLh>B5sy#P{H?@6o
zn+8;)xizN4&*@+Zouf3jH^0}z<m8KFegyxf35veX?M<6ea%uhE+E{A8)l5Me0cybr
zwP&>EGi5;RXh|jy37Z(PUat{jglk%Ia-N-Gy<P*TpLN46Bzw6r0U_me!0^0C5_OIw
z*&#+~A5;kwVGjo2`vEpmIr>G{;{EsE!)CpP_eZQ&CorImPa0OBa_uugAR6}Kjsj*{
zla%x1{-h~ZU$4DKZ=jmvpT{EayKNryy?)P}kfkHb<oNkMDfHCL7<w)ePEv>qA>icX
zB!|B?O_QZ?=N#Im#iD7@_dPL}!vSsE<^<BkVgctIx~@alb=d9?hycF*`fL2zpZyvB
z@DKkGzx%ttgJBqJ`cbsx2>T*PhxePA=17eXQ6C!4HJTAkN|=l>A93J}LF+n%0r>W-
zC+ONW7RwHs^(9uDOYDvZ{LX*)5Ak=O{tAbOcktUk`w3pXdX*DstqtfFE5KN^4dpQL
zA%ue;sIq(T#ChVVxl=wI<1jLevzCV`Et(_xdX4qk&j~8EJ*3-hw0I8YwZ9U_wO}y!
z?ssk9+2=DSRo0~18sp*XZ_;LNlV0_28Ys8<l4L8x1gB+P-ZYQ&V=)c@DZ(e25fY&o
zh5_Az4gh=b-~sx+$H~bF_WON7T&t3d5eyX~lFAvhNvrbC0=M7=+tR6DN`7oy13wJ7
zxH!lE_CNhk@xi0_vFsK&?)Qi>B%)>!49!y>$~MPH#?6)d1h;G|IbRKD%{*#Ey4J^B
z|I}?g%)a)lCM*nd{Ve~+<I?7lH|0mv#;XM+_RBcuu;1_DoI~IDc>44y`o2fkb@}<O
z>tKMw6t`F`uv)FKTCLDDO=d`|)e09E7g+TJu5b2OtrqxK|LR}iCqMZ~&e_O$Xx6}l
z={m(2A*0{YEb<WtF_Kv?R`zp^I9p>7j71+kY!bdKC~|bS-C@7o;mNnpaOdnUuD3n<
zBk=Oc3%IVu?q-X}U;X#^@rU>E!;e0|AssY$9B3@d<pRec;PUb^3sZ-~0Wm~k(9QuS
ziu}D))taTcB;RQ(PK}_~aFoq;8mDC<sk}ayzgWXp8U3h*;+haYN6%{4b&yJpIo&p1
z-M;f-a&Kxf4+R)Ef0m$b(Pm_O7eL*nS=ETuoW@goUaO$xi&!Gm%+Vr*RB_mve1w9F
zR9@$|fBUy_etr(`Jr;{aH54|F4^YTN8_^`mjLj{^P?;UERZ2MV{z+5tA;2^iT}x8)
zpZv+6;N)bTq~kUROziesTwh<~a5(0*t!=C+3`ZZqoFvHC<mP>oNkgmsbN$MDX;T6<
z=2IhpMljmiCY#fIYD~1I&2XK)nnYUH!ph(jW;XA=_a5%uyN8>bo9x$d&SBAZaK_-~
z`a1g`Pft&Aetr&sLXf4SID~*j*I@`f9z1xE7#9;pE}ddvY??<vQNfrAD~~z|xKqGD
z%*<zEEnv7K;i#cx&S3~}peo!p05B1+p1;5^fAM$NUfrNuS+uUh$>tnQ+rlndTs?mR
z2ZNWduJHKr6I`S)>KH%_c)iBjwdfWLG>yZvXHW3({yj8pgTCJ*L<7qj53f`3WCjEm
z=bfI6TNAqFnium1Mu%cV_MO#)@%p|J8aQ=4OZ?{r!})z!qw^Xsx!<?t3A%mf^L%b2
z5^|%|qyzbU?H#h8+I9ZzHi=a(j<pWfxG^#d5izF4imXo0_jMaGum-`1n%@nO)aNN*
zpzB&V7+3&@!xrb4r#L@5$L3^>tJl}5a<$9RLVW6|0m$YJDpL}ohw+p%(JeZX%0fUi
z2F5xB1FC+YT)}B}=~GHpjkJD;;|^`R#?T+}@y9=awFcdSoUr@-4%_WDd^%{XANtJT
z#=f$pQNV)X!}m!cVq|^_(m;-IBC`$kZHwpmJ>v<fQ^rxQk0#zavP_PLpDBeVjmI^O
zs4BrH0gMlR<9kC0=!d?@2Coq<ox9X^4cXMA$L{6|-h14=Jjahe`5}Jw>8J3)17Uzc
z#IkF!SS--B4$T!E$+2E8&^B~p?!7ygI6ph3bEXc56rJYLCjsMRy+P|-_H)pYjI=%o
z>1?Se=8u%j$m0*C)0;I3&Ozf7H+jG0wTz%LMg9oq2Q<dO1o9b0AMx_)HU9P&zrb>{
zgmo6K?GWPuEi6nhxI9_oH=llvHg2%*dtBVRhiEOl8PIeV)($v1J;CXEgPWTh91aIu
zT%4otk4ZQVKuo9p#sR!vnuPbcOj4lJJ}n<ssTgVuVRVrd3Sk^ACGv8R^u&N8yqQkL
zw66^BFV5LZrp&QJ+F<$Yu1NJDt$XuXYTQEbv<H*<)UvrfBu$X-ixD26<6c<n|2~dX
zD8e{Vk<JC+S;NuXSUZBXx+a;LW>k#GlH+OyG$jTGh-{|f?K7GQ$TU#mhU7JK5^pgF
zAaC44dXEuZ)1V&)zz6t#fIs#yHlSSqN58{rv%t4c9;59#9FKjPJ1x9*CMAAaz}oQy
z(?C&2Ven|0216>p<GKz3hDwB_aM%(lLpB-EB?H%x&HB(EaMGM$f83&NTfBJu9NX;;
zhM~u9zfHuGYH<w%)e@uX3ZRV<wS!+0bd&bm7$h-;D}}I*-)&w8xleK>?#x8$QXH?t
z)I>HPeI%X+5Q;I^a<TZkWNaF@*$kq?Z4pbKkM38Gk=_$99QUvm82SU+rh$n8yW<v{
zg~QSTwr$ZyGOfiJ5#X^|b=ckngfO6O0B1eA*5Tx24d)u{c3ZrD9Wi*1{jP_D!R6&0
ztd}e7w_CWz(Ry+XfauBj9|6gv5o3{LDmBhRNVSPrhfxfWnmT`O;sECdSmy}mTnkuW
z@B?<k9?MM!+jbaykEU(0T1ITQSGc}@jmCJ~Y;W+{S6|^D{)6Ac!;e0~uRr@0_Wc&7
z@mMT?Mdz^J8l0?G_|Yeyz#7nrs3suBAyo<qaF+Iz@hN~J?I9bHe6cJvF@s2u3WZdY
z_lo{384Qfc``j3le~Um0g5b}w5Gbtyg|H{4!<G^xA6-ZrGQX>W_eFpVJk8&Zf`LJz
zN%SDCtq)09rg4N25Qb6Ex~+Yu+HN&9^GzlXeN8hB$(-b7*JeBYuco@I5s)+(C3<5*
znD`pW6(KQzy81acSMGzy!_PL(<%-$s%?c-{8&ICjfTpE_o^0ZYku?W$?#E~mA|?Li
zXgCn#&<vWYQa6acmqAI@@k<HIT5_J=++5?&|NMW#`|m%*&CLz0HE7#WVxj}Q0tF$w
z0Vd=3`7rF7N$O3o)%V22_%u%Hxz?Y%-QTF^QV14rq50@{6>~;Kh*9v-<2QyFgo_j&
zZveb}{t};l`YE11c?Kj6v2hK$wnYS;-nrZF-~-if^D$tz-=l3iG=~Eg3zMW`N52m@
z_~esM@O!`adwB2RLtI^5;o{;PAw<N`Qz(CC;0Az<fJ#1x3?s_*;REovvDx-ktr0*B
z0~#CQ2kJY-0sG?)uV25$`PmuTZh<@Z9|8bgyu3o!G`N0!fbS3Jk4GRneE#|8_^sdm
zIez-{pX2507kKsh6+-Yh9QUvrI)eL$pZpN-zyA=8wZwF?j|Z92um-{5<9dzqJb2C0
zv+0L4De6g<T9Rv=1sa~eWK@7)0;RCbOa(dpAu%za=u?rvG^CnOHKu4+Q_VFfg>wsw
zR7u@4&3yBo0=}c>GY)d<3#n-?JjvSMxeY5dBpofp%!vV~Lt2@!F!P9!%wH@8@ZT7D
z%p>DTZ<&mzX77~qux?r^hfOV<>+|s8L;Pp|^gqK#Kl}(y(_p*Z<MiZ&B5{(>h2}QE
z`v89o=zR~@G{CV(+XO6^^d3J9c>{Ytplw}R4}&lh=~$E65lL&?ZFlfphkx-e{ssPv
z|Kk6QhYugXxt7j(@kaq~sgaDEO$#NmY35QA)TvqQHil4Ra<yo(+P(6gxpmaAlK0hc
zvNqQ_&8lRO)0`%~w<;qCV)9)qmw5T&1^&N(@fY~(zxpfew>xx;1)R0{K)|b;8#3Km
zi<6TR0Kj&;!+yU<*L8UQ{CSRaV~ujTT;QMmvp>Om?>&UI7N=(y==-CH8kI@9D&H&y
z^eNOjMk=$?c|R#MSrZs*%*PpE-}l%bj_A52p1pX2|Mjo_3jg!J{!0o*@E(U_kG5Oj
z>goo@G<f;)6(V9fzT^z6)e<jWJjWmZ@gKuEga7`2`0w!i`E%S{-QecM9v^=25&qyG
z{R6DmEA01MESKp#tH|1eO@7s^V`m<eKZnu`#afIp7Bzg=WxOO_%;B!qwDrBWJqx5W
ziP$7(n1Ligf`3Oc-<b$_HtMB$Rr1k_eU$uvdOzOQ-kJGmLZS&Sf0m9~ZZ!=_-d}6A
z&0Hg5QbW>`cKJLnkbDPUXXbQ^+92LPno!1CR1!wAl!o^nmzS6LC;#N1z&0W=O)$z4
z!9?CeAc<lK146Lqx)rR|aA6I`!a9dAqxDvi9LaV73-2SIK6!#yuU_Hd!-p6KUpkr%
zlGmy<QB9=!cOd#ap4*<QJ<re8`kU*srfJE3^u4wIYj`N1;f<#^>on(w@>W>et>bDr
zcOnlS-VazUmw5d6F@F8ouW@yAjWBq)6h%dWD73c2;IUkh|L$-&P}JA)nD-lpUsJs?
zhttzDTwI)Exm>~+gZ+L_!cH0|{bdQ)!I;s!9wXWN2aay@6lupn32#GWGmHT2haT3E
zaJ1d+@QYvkE&k#!{sOOFUE$>H6#emt?aej9^)+6;e1#A#7Rx2hPS0?Deune&6D(It
zyn6Kt@4fc`zy0@q8`n2i_?y4^E0}1|buE7I(MNdiy@y~6$$mGWaVF==MMF}(i4ifu
z%qB~U_jKr7ltSG|5~@`&<i8r-tihxv80mZEJ9PP;5JEQR0Wsqr3E&Ce3eC%cqC4jO
z^%G&!*6ViO*1or@)-~z%yAl^YQMq_E;xH|~S07JpoFg$W$0ViRDS?QSB8l?8VHj{c
z9?^9jPEJp--QB<#%BVg)IiVy#;}GLmL74n0E?H6x8cW3)DNM61jz8w17-PoL=`--H
z&(~nNT;cWWSLnJ1*H<^_x(>t8gU*1VsJ&tE06dyT9@B`?&RroWF-d^>`W*akYc>3M
z+y3T+pf{P#Y8y}foeQ{`yC(70`yri`Q)fPMY{bk0n#Lf6h^NnA;1_@Q@9_24-vTkB
z?>+o*z|i+tbd*%uG%eOAXRy}d=H?o^{V_9xWA9Vmi^pzHc-J-yoS!f8{P}a7pPyr~
zSm65RH65qnhcveF7<L+FDU1%@V-O5jT1#{^=NzoK`C&jC9bz;%_5;58_A#D6e~GK>
z8|?Q7xRj^#@ZLTA)^Gh5uC8v-G!33TdxkI^5d8wnu7e2?pMUllj{N~Y|LM<gd47)7
za*5N;1@2wm#nr19*gA*hVnKe5RK>ws3u|Ih77Uyr3r0Nl2qEBjJQ9<MW2vr~2s$DK
zDCnl~N5Y`y!!Cu=(b6M-uicZG4&MVHs>sc`u>pWJA@6O(hzJ8~I20T}I#z?De66!+
zT$2Tmw`uR1d@vzkIskzut0%)inyYZ2@8jod1ruu_+_Fhcg6zu_S(7r>3fn1j|8h)Y
zZVLCO=TjTEHUmoOR|a5gHXG~?JFHe6-hcmnde+iCW6%PR5HQxFbuFxIGx6wJCnV8;
zVd#?tgIUE*j!&*x07YXh7E2tD2fTiLjqP@qj@0J$wP;%^MPCGs7?TlD!f=FTJ*P&G
ztM<1BU%A<8cqZp5pVjwiJl31NCW+SYTx&;j=A_!a9HheEWdfhxW2{wgcRif5c=_@b
zUcP#bV?Q8<fbGqe7(@v0LqO9wtTr0}K-0EZZ%%ML4tRa_nxY~@fDaKN1rJ!~0C+ro
z`V?1JR|p|sx7%U4Tw=T3W;;F$a--05+dc*^KK~6K{l&kgYk5Ct&AHkI)f-G@r>t}M
z4?q76PG7yC;Gv<%@%k0)^C#Gxp5Wt;e~7zx?!r2UUw!*I_PZUG33~c1&dyJ<Ua#Q&
zfaP)tAj0o<h)<uR{pt%K_#AeP80i4Kq{Z-|rGSAoDVcY_P0@-eUq@)X(Tw!M8oUwr
zWt$UZ_&MIUvaC)GPiJW>kHL)oAX>uO{SkD=Dboy-LbaH{7_o|Kzr4Y6#O>XfkCOKE
zCgf`pKy!p-Zc(&7RGQUl8&h(5Jq~U(7AnXIjHoZQhGLpxEOHL~E<IQ3tel_zoGZ;H
zpUGmefH4LieDFc;V=VZXBCV{2v7<^%l2}7wjTjSRtu;6tk8sYQaSjKez184c&VfyX
z<j+|w4F2LT{sRBefAoh~tycLSlQ)$zO##H%QEr}6L!H+~q+p@dcFkb4xq{{}My1WY
z<4r<?gw1V?OiP=xk2y)chQF*`%Y)ct-#ST9z9#?}4n0=OHGcc={SGcK?_j^*V>leK
z-EMI>9DoqfG!0gpHO?;1u})0r%P+sgpa1!v0|0tY(Qi%DU>FA25a2y<d44DFB?_&t
zpU*OmOa4a5p^WRVfABG0J$s6?r%%%}NYb>zOEJIvk1pWQA7OpK)$7;jzWf3o9QN4m
zcJLvfaV??&hF34}<f||7%sQN%pW$M)!g9I9Cblq-zsCCJH7+hL&~^=QeNBR31Po%<
zaLf&joyIc~S7TD-njzseCT97o|IHtd>nq0Zlcq=2K+0p-Z`1PR8uR@c7xkG?rC)n`
zW>oEOjloG=EABI81@tjbj%b4D0VQ4d9+OO^l+Qq5K*-Y)fjT#?ErOQF83`q$lQp3D
z+FUD%1`wkMRKNjCUF9`X?W0DN@_c}FB=^+Er9Vd#sWzIi04z>VPjPv92QQw#M4Jj6
z5~26REKQ_Jz|N-p2vKY_!$D7;wvJw>%)n%_aI}E47zUr}c^&ZCXP@Kg(`S^7XiCSa
zT$@_yqIYQ}F;-KkY}z`bu;7R*-8WD_*QbV)bIe1h2I$}B+FX?6C_q^(c3W(=Hwiyu
zs+AUU{*oLMC2EG8b2E64Zqebv2agbaIy(D!#Ly49KrIZJgu?NNrg7*_PVo5j6fa)9
z$WdsH`|f)zJBz>f-~m2bE)l-{76=};;pAl`yaJHp^-KSs@dG6-Cx7Pl;Y0l5fA=Sa
z<gVdRgFh#w_Wh7^hPJycUcGopb?CNR98*~wV=emQF-7b(7*Z97wr%mEv1r>i<wted
zC)&0RY3nh-S$ek<+B;IdjZG52HI!1|IVFSTO`pt)X>RPtCGDbKar(BsB897(pv})~
zMm8s46+*os3k7Q>k);4+HVebcnwSo^dQ|ZAbM5<W?cck$o939zfo~g()F5ioIDgmL
zjBJ{)%InkjN&B==8x1p@+Pvz>Q=Jbe$ttJ=#%h{aK!6_x1Ozlqm(2ij>@1C02ln|1
zX|c=;$;c=YC!KRJaf&cnV>8AO^MS)+L8!1^Z*Vy5)9H~m^*h3h31ybNigCX^;>mH<
zuIcM@(UaEFTIt_4>?N^#lXf#_&XCVCBY)Fe<Zp6}uRs1Fe*FLVZ-D>yze(=^SS^?{
zTd+o8>{vCLLhe)0LP`MiJ{@fyBl@9FW;@cT|M_l58cjcB#%isFvla^sIRE-HEdKBR
z6`XaM>9}UZYg2qE=exWA0Jdo|^Lew@I==>k_oO+v28a=Dx4^~uC5)w%gxzjO%+tE$
z-*PaG!|}*IG{RfXcjB@{ZQEvH*%{=7SF6Xv@D7HK(oR>cwWFq=1ct&4h7pI1EosOc
zEtBw=6QXMO)U@om#zOX7HG+wzWVtk0swZjW8cDi24bpmBI`ExqyK!SxPfEc$M@LKQ
zda}GRb<+_-w~ga=z)v^xoQdvD&+AQ+@2R{;O5bwhS?A!a!^OoVtfBKnNb8_dHp4I^
zKb6gri0n_oGc(ZS&mvA+A}GJR7Lp#5DA%?fj>ki;yd85N2FPZ!Q9I0YeZv~e33{3s
zy{XlR*Z979&1vphO{=XD#u~vWn7$g%oOG}E$d~`|AK}aY_>U$`PMT8Wm|~=C`T$}E
zecvPc0ECERidggg0H3NAurvHP^nfuq9*;1_;O6EAecxlZ+vO1Gwr$b19hQqFE>177
zUaxZTQR{%kqAPI?vI;~An`Q={x2<(;AG)Na_>fL7H3rAyArsK9>(DjrXcj^MEYO`$
zp<a&S3`vXObBt(QNk9u8#zZtCe-2>(ED>27<v5A)FjFxxz(n%-Oa+l~ZWF@(mc)5Z
z=*@zx63!WKs=|0NkNKbjaTnM!ME%I$ScsP$8<^XY3oGqApXZZ@af=GA@6oNs$=_sx
zk$q|))}+yNcj|lmOwH!3dZ~TP*(~SA!S?AI^QlSHgYOXyogZ(k!D6w%kOB?H2qCJ4
zWh^nP00YDTYXN6lG>x1;Z8l>@iA|1YPEs0S=3rpZG!)?$0uw>jc+|V%NPsom-<Fxw
z)_jgY%}LASz4Mw^P2+jfJLJ0QN*&y41f|BLYVEdhj2hiEXoa5%feMw{h6rZ2=rA06
zSVPArELSVaJuwEy{)ld|z;3rIDqc5DlY>`yzKv~Qf+gQ!IziYql#|5waL7Fq=h|{h
zp=%DtwYkc9T5I895lw)P0TxX3N!VR<T?&UL936Z*6pyQTB-3)+TJ*io{?QOpWSU92
zd59oia0;nT9v==gDE2xC_c^bJHB`>yi;?VB9O`bA5A<#JhNfq8I3*dn6!Q5zC8#}b
zkf%vMC1IvUw*VkI-s>gqk#Csk0{6^*!Ff7UYwyT>jQBggAcQCoZpsj9QXmgf0#T~3
z)K{6ilSg69gu~M~rsclny~e~O(E~(KnS)D02yy-Gp<)b*MPW6Unuyd20nS-O6LEcg
zjpbsEdk-Gq`O8<hzTN`C1HmBp0T?180<80C+y;&FIEH}60Zr3_axNUkGjr&)$QWYI
zcxGF!3CL;J%rt}dF#ZTX3~-G@2p;=AMK@WCY+6Gc=TQ)7gULF~5&JozL^DT5dc9%E
z7owR}DP_K?%_YmtNtyE6ZQ3$_*CywxmMlJN+d9KTzYhSMv2bj1V*x?rcp+&8F7;uZ
zMU&3vY1=k4AFg!WbzLT$#*ocF4guX_k*iSyCK2q><U);eMgnSkg#{EXmCO0d{V!oK
z!Y`m}419u1<0yEA{TCtl<QoNAXAw+<@gA<Bpq#!x!W%<5P|m?3CO;0DlfZ;Df+D$M
zOr|tH`d2NdFO21LvXud>Vexh3{a~$0zLb$3@SHgSD$hFt<XZ(;gW!6GQw@q301U{y
ziI}Sp7~azy7!-Ni(oq!b^BD!^k%1OMGC(;^wp<%tLk_x#i5}QU`;g!YAkO<jWqW7l
z!v#5s7*4a^8PJkzN;Q$c%S|S8R%*zqTTRQVNzcr9#(rq=tYrd5n`tB{nDqN5f|fU=
zVP?i-Da{mV<NyM~5YaAL3Xe9x`JD?4F<^)R(L_S{Xp`h*5WHZh)}udefi*c$<B$%&
z;<;EO3Z+`uFa}6EW8g3z0v8Q1!~re>#u@B(TP&9y&d<-VIayQPw_Zk*@w!imcTEEp
zP^!SJzGhOWnMz&dy6g4Q%t{kKjbD>d-m=!35!8Oyebuzy+xnIL&n0h%5OTC*v?7;7
z^&tf~HjPVjxITx=Cdy$Oix4P;*t$`&9{c4d0{t@{g<}$9bR+v+<oeISgvVFlK_Ek8
zBCXlYlzM=($i5m<I*8#SQhhO(_MUC%8VxVPgg}Hi05J>8qc4ZGM!ue<%aDkDUXzf<
zA?L_kCS-y_@Q}<StqteD<@*y%VZhbkdB2sFMSy55OfnDpI2LZL9oA<I3N0ryeVZ7s
zcqR)pCL)!L%yVPSXB5~?s+?d+h)qmLpHw^7zH^0xImR&FJ59I*<~O+3X5K{fxjq!%
zTYJuA1gcB2xBWIZ_Ns(zXlu+#_F7e@nr&ku-o3l`#;V7LO;~h1f<;6hJWQl2&el-$
zeF$8AfQYb78^CAjS<Z(l@v?IjCLxU@DcO3GVt2b8jEVT8Kl&qV*6X4s7fEcKq*jJm
z-o93)7Wp>XNv+k^N=H?`t-AlFe&^cT^e>r4EeDQ2Pqn?$DX*ESHnC{mkRGRI#CnWu
zhSFn@%zwTa@h6VY7znLEGOM}}p<z?5gMo=&+$*{<QtKUwzL6REW@MDE67gsf(Ufq{
zxG;0<ON1-t8!<8m9^*I#x!&V<){0n@k_?U_a;5e^(#1S49#0A8QJa~-F#|qpJ(Ipr
z;*+%o#?g@$p)k4HbdN#Pj2d;5v`U14zV9crf3Cf=&qp$expE3I5z5-ek`tDbk|wt|
z5d_wjWM8^ZJ)W8r`8Ld_re#g;rypmTmYh=#S4V&~4t@wYJv+nUaDcUJI|gVoQAu{5
zYi77{^bL9M$sS%v+Qnp8CjzVd(O6iJ<kh$a1~&T<j>jW@`qQ6cz24*;B0c|FnD?CK
zL7Q`)_{udO-!V@B^*Yu<so#c~Xn|mkP}lmp?LBfc^S!reHgnpT4#1h$5bEQ2o5W~6
z?-<2#2bD3EYX~GR07+=a&)Q5^zE{sl6UFgf7BI)YX?*fsOswlQ{^k<Wt)0@4x1l|`
z|McRT)?337-Vdz(jC6sn$ua296XD8`6-8RO=AtA4(uhnDf&%YL*7oS2a5{4-8mi>&
zESzyUcdJkk%$U<<t)=K%Yjdflw`uRheAsTBCp#l~rRf%0Qf)z)n92px$*ImwD=5qN
z*UT()gic<Qe{X9wO{YX0sZy7%B+?`E;T#N(nrU1M=UR~XlcXt6=k%Bu&<CFrP}`QA
zrA&nMd^tRJoH8L(M6xx5RB#(eT9Iq9SajI$clo4CV9fEMIatVz!x|Pd!a5P3*GjdZ
zF*2UIK%@JfuETWAbko1REwSP8O0D8H)18LJTw=D~KUy2npCumMrip2$J~u`^7pvB4
z{(1KK>4>+I`SAXzt=*iESPC{{fASrA>a-vHEKFp-TC<!AOw+N_2VSeB*~WQXoqkp{
zXtW;^6F@Aq7)@(fa1;YYG^6F`jV%Nql=U`ajIgjN0L2wKfKBV*8VXF2^qc?6NIMhB
zA89NA=g=fS;+xucK1}XSO{(>Bv(zLh$uzWzHz#?@fAaa-80R+7ZHQVUX0wbe;Zg0q
zQ!UFPgdT0z!5T-UZH6Ap<qG@#0Wlh|?c5jx0|)1jDisqGf?ycjFV%drNfjTX(y~og
zj3*>@ge4f0j#L<ih||+kESJl4i~?ys-qSiZjhk%xG}1T1bx!)PF{D_r8B6?-YgE%>
z-$a!47;n43CXwH^2V|DH7Cvf!_1NajXtj51Vl97v8!d1mS^Gfe#LO^-vOklZ?V`u@
zT@orl6JEW4CxjIvGtt-P;9-o(!5&(msTKIt@7I03O=AcCcz@l(0&uWXES_X;Lv#0l
z@i1l{SLC=u9GMm4&22)FHGpgQSeDU$Byo=NrJ9`V9%F>>2Mlpc{EeI&mO|VOfFVX$
zgM1iYRBJic-ii4P!$4*p;l!_JfKL-d=hf=^f-?}gB5v(>-83T2ZE#5(^mDa$-G=lr
z5%P(=Utxs&y=^<R?E(P?9u~*LF^z>CpH!gF!~kau+NRCS+*(^4PaGqPbAUo5XR5Ju
zl*FiC3x-9RnE<F*smaGd(7E>O^%`B<0VaS?Ff_&#VaNQ>T1$BtZ9B8ZWn$_aQx~h4
z>Dg_}T#uv1RCKGodrs=rOis626_47xxKHiR<j-@_ak9<r8A-I|I5qLrjWtz0XT+E@
z{ojU>6Ml`=j|tb9;Tl03N^MU{&U{_(+wnK*cU)sB#1c{0#9hx_mwlqX>|`8#zYeb-
z;bHPv=m!oOD92iXlz>^93jpam9a+&lt>i`mjz!bSh&U1feT*bPSgMF(orQrx*E%e^
zwxHd_;K_8)w4JM-%aS_$cSy&$G)<F))d9Pdv^?}j^!*Xb<r0g<vN~?v-ii76fl7Cb
zq04nmgMc=z0$tlgS{l(BQtg>JA|~(GMB_FM>0P8B{chGOVhl4#J|zg02S8Cl&N=M1
zdkQ1wC?NpBM+&>n(I!J;Ks|hjXxb*3?<|I)Pnp~h8Y42pVF#_41ql(}qwU%hm}78$
zeu1C;>}NPRImyh7Djm=~y=Ot7Zt|%m5)*>YADO{0)G(e~gR1jYuEorC-Q@2fgu;h7
zr|sM(*y#7_KD4=7<Gt=<ZV%L$xc;ps^y#^eXGaN~t@vkd)u5uOl3FIzBZC-)R3>)P
zRIAN8BaFtrMg~N_8Xh}q<@^k%2k5c0;3jDoC!p%N4XI$L#6xRsI@B(Vfj4&ijm8_Q
z%&bbQgh0l=Cm*JQHuoikov!QBdLhAM6l~hYq3gOaPb*iO@cFDKpbRi*nkJp;RQjcP
zo%sI4;XqX-QlD1%cp7IjeK;Hr6|9)H?__qHBFR*OaqjJ<NbXy|hbJxDZJSp2BR6Ah
z47I8XH9wRdm)_V01~HTcA!`JMo3#%8Fd&)+&KW#=_8fjl<!mDI!FMsH1L?d$1QSRq
zdtsafh5@c=(f2)0Pfu|;9KoA@@B<Km#-!3A5gFM6A`XWGqA^&n*Ki#L8eCjl;2;0v
ze~in^OANz+w&Um?9)HX*lJk><WC|+_K5~0*9m>Y}4)Y8~j-f_y<$ZJOP}3M{>rLx5
zf3As~s~k(M>9GU1`!s8F83|ThlbU<>w%^8nP@I6P1;&yf12I}eP*K`+2RJ{43A^lv
zj~(gzJ}0{J{24#Wqv&lepA%9fRe$t!){q5;=wU3U0z_EXKxWzVds!=&bD^;y`!sYU
zFRvqO1;!YxmZzXS&}88(g2u(Lky?|~E=%nRh(WDaezsgM>4!NUj~RAtLvwALHs@=}
zBw}l=uo?k$?oz}s95an$2E}uXF~C}?eK+(Thr<!a!y)f))_lxjf#qU>MN3T3IfoE^
zGRwy?a`z4G-I$M3*=kzAT(QO5HUUsi{?mk9_Elqea`ETxmHlbtcvG7jM=gZ$wr67`
zKZ<P$p0BU2K?ie<k`Tqu8<Ms}A|H*F!bf>NYzieLq0c^~cV;Pc%mv{8dQayJoh%nw
zBKpA(evm&?^#`Eb5F9Ex!c!KL)o>su@KgY~4c;}4qo(1^t!qtd(AtC6Qs;yQ-3K#p
z$pqwYwLJl#Fdu8FTHdF}qM57A&zTdtYH*xu1p@<1)6?Lfsiv2EkXHflUS_R;aJ}z)
z?Du<Yx7%FmhkXl72&Lvs6&O-PW+IaOjMPl5ZSuP##xu2KHRe7QAq8|fuyL+Z6JR3C
z`A{rG(YVKXM$UDbfq5B`Ys8oxkZ1rWgj!=@tOHk*;D5|?bwqGPMB8=~X2)}r0wMd0
zLm0%m0*B)s$HOrlibqw}o0gcGfkWdQ+;Ul%>LlK0lU?7#xDn=WBTxa~A=J9YAg14M
zPsro@v=&h#?zcUE8zHMb`!)>bwl+J)8L3;ovm|C?#R9;<4<61nI39aYg#y4v^2=Bg
zFvI~|>9`;~W;7@8yJ!O$L~9XZn2aNY0HiusF-AI4*)Gs^U9RUfq-qCb(nu1XKw9UT
z5)QG1Pf0XUY;s^4XfT1N(6%%zYgH`xxf;>bOi^oXw`sw$pXnIO&ysl6et%Qb?~zV&
zIbV5QGr!x$IrmNpgE^l=nPkepO$*$PM4Z-~3Fq-hjOTDTV7J@lqM*Q30Av`_$*Kkh
zqn6h+O*Rp+)-=95V6kj7ftA0JA7X(J=m0n-k|B)Rjod@rv}q7=WOT%=4PdMg#!8dh
zlwphYD1J}XvmF3Ta~)aeSS%Kq5fj6j?m1oqsWHVE;X|LaLh_>zzR&(0hOKK-fmTa}
zP^@#XBKRPr<l7kOz_=JA)!?%hz3;P`?t9UE7!Y$>hXkfEQ{qvRR(WBigKM%-Y9ZwF
zwWc@EDCv}aGJ@);yf^h*W0=NH8R=XYdI)49K#b{hP=Jua=VaC31D!mm-WLKL-D~V<
za$`o)bshTtm`olP)&fY0sa(s;3*sHEL5P%G>b=ML`B^q0#2C=D4l-n(GuC4qk428c
zVOd)@dKa4-Voq9f)|EoDdS*`Jt@+jF)=AzciQAn2CEqXpD{(WAEk!xy^LPaF)#EaT
z{36y`c3q^5_9-*reIxrC$2JOedjCze!u)NDgY=pbcJfa}aMG}(zX^sWW*^Kj4Cy=@
z!eu(PyU+N+zO~hQog<=*v2f19`;h7irb_7{qH8-ClMcS~0}h9tPHgoNz#+{Y3D*s-
zuLtzQ5iv%zZIjazZf>@4&e9PL={GWA(ditWWmHsc8-*zer9l{mt^t&Wp-Z|&Knal=
zq=qi(PU(;iX^@T~lm_W;0qGQENQdS-@3$5|SgiRqhx0sl?0xMvzJB%D2(!UFgU=?M
z6u>h8l#Kcyf$$eHB389V+yb1f+;rmG2Jixgx6@5+C~KOrW&MzT1h7lfP&<4rG5}Gx
zFiOVn2Q!4CD~~24h++7Z2Z1ShTM^8F;xTMNucpqB$JyCA5e10ku3Ow%rQ6ufvxKBo
z#fE$$L&#hBqwoG|{9RK!FV7$<VqxkWfh2k#h2+uG?%pjf3iE1Ldie+nL%`ERgBs%4
zh|uy3TfgMB>L-8J&ctD2VGvJ{ykM+?I$jRrBwrskrm5^Ot8$j4%YBh@>FXY5a7YOs
zh)_4tdrIV}V5}=Tan$1g!As#5Y#CEq9nwG`3tC?x&GS9uE1G}t1j{><wq5pXm-30A
z>p1*52L@Y7Cxiq9Bj6035mWTcA<Dn~ww;oKLHL1Zpl3<riOJb_%#xl#J3$#`e8_!l
zxRes1+vl&^T<9l|n715c`BwJaX|_>`!y6!$irD=6`=KEEYvVv6cZ-#HYa>k6^?xb?
z>{*HLaLDk!2N4fMje2973{J$#<#yqMDHWe7sq>=bWCCcML};c;1;XM6VCL1Frea-C
zr)w>t&!hN!J-^al>qEyea((mENA=$)cH~b<XZ@2eD)^y(!m&7Gidg9~jNX~vj=6)a
zZFx#`8D{xGI1vr;%1__+gDwW7^qxmWp2&UGjxodOOX#=ZyTHuJn`M>8k9?yqC(2tj
zxoV1LMJ<-O>IqtuP58*T+SOBP=+Un)@~p3M0#@X1n(k&F-E`>W>KE(Sc~1e1;AQ0H
z7ZBD)A=MwYv)N0OW(U~!@pB8-VwTkAN$!XS<fcOvyHz0_of<xSfg#JjQuQhpUu?5j
zHqYf@D^I}(1H_J8RH<9DsFB0W#MyyDJR6Qg%9y7t<p&RLYj<{8G|`OMB5)^<@7?P@
zs?$$N&@pqE9d#(&L0e~4`#e|V-jbddw~~|YQGwvT?F)B?zZST;y7!^%*o5Ovwe$IK
zE~)FO0l*k`OSBG_<$l5cLY**J@cH@HiBoSFL~SjKF-=hA5md|!D#}i(vw|p=Wy%y`
z^Fo!T#Am28GC+`~GynV@Fn<mxy^ja$dS~Gj-JFE?NS{a<r&GaM+z*r#AB`KX#bQ1>
z?pt1`FZUdxzi6)bme)Y4M@$i<mE?@|yp&esBc9lG3mg;1UdIbo<!%b^aTMS*Wf%CZ
ze(HNvDCn5BB)RT4IL(taf5j{QFn|oUH?@B@E^rz&D%`L2@e>>)97co_0sNDP34~cp
zOiP*wV4viI6c%Fr1dNkZ>_8bNg8M$-@VoF$CHm^fi;52j(K4F~KQqi*oagd2vf&^P
zVz?JrV$IRB3N}D=)dKo%1n(-|ntKxbW#}adGF(hD`-N@~>jsyW_y|4{epp}(4J>_~
zx7jcu)jL!A+S$V+90<7CxY)+JcShso*QG))qZTL{yj=ocrTFX^;!uB4Sf|+dG2h{u
zbdZ%kfjc|REot>G+3>S(UR(kAt$lL91a&J$U*<vh8rxnsY(~AJ34P}ZB-)5;ICI<*
z-()k@5Jk`xLGO>dKA|%FdAs2iXzvfFjpVCezZwkl`lT-X_#Hi8^(o3yPv7S7@g|N}
zpn_hoSH4rT(-iu3wBA0UxwMo?iKVK`zX0Z?eV@2)%xG$ECLb`9X%E$CXv)3m5Vrp;
zJIKo%vZ3VrH4^f)1C-OQVFoGj<;Y}z9zBP|=ZhJQc>Q7w$&<`rkTk~oM6YTW6+yF2
zy2(z9sqM;V;i4dx#9SnYuaR-`ftd~}vS1p{>8WK?vg4A0iE%}*)=P50qP4~U2DJ)v
zDwM`l5!?HABvk&Eu_ZC;XF8#2$|dcPIj^Gq)Kwc2s7?X#c?^0lZ@6a!@gx^GKZg;n
zUo|*sHIs#)px_E70(sY~3KA%UC)Pwb^+B=ims#k^#yaiaS1!PTAE0BEjfSlhLfy<Y
z6x(tO?F9#>>~PU)jN5zW=AKvA)R<NMBm(N*m3ALWxO3U65rUw-1G*&IOtb^)Ogj^U
zGU;MQ&U4#JJs*m+-P=29kRT}+J@C8-pJ!mM(41D-_lR0nP)4vGf$=S<5kf+22`~Im
z#~(x%x<o(IpffF%peyv@Q)#;RBkUaT3g4LX8`KxqO9;+_NSWxE<0Mt8`qO=noN)64
z+j9dHduGivmempz$ZMm1^61F3acy~Qma^*JJ8*SvqkiIIu!_fwAQ-s>VwriJt;j$m
zxbvD<lKcf6Xiig%hU!jX3z|2Tl?aBg#_A7hEuQ}#S+95_uhgl_y=L(7{&4zXGS_v)
zCN6+J+2^UPSeXyEGBH;c*IK+%?5dN+ffpFHhJducv4IlfF{Usgb1{3r+~?i>z}HC4
z)X^LkA`c{nae`RIM>7&!xSwS-VFVL7jVT9IvhKDT@(jf%Q%Tsk0h;uY9sb)S&lP@s
z<zvQUJ;nOD^bn&oVhrOp&mgObTy!)|O-(1@noAe^yovUmM}->h$$*lW!5-R1CnpKM
zz4s4(kNB5^LFkaovg6m4(goixih5#!Q_@1oPHCNjF&r3f=n!Wa_}y`=vazq4QkJ{R
z{=Dk)EkB2t+<*{QTPXo4=FSe#nG(4+d=L{}B?8@P$=}z8>Me)5z9K}+_+ZYY>nq`a
zi*9uYurE&JDi;ph_(a959q1gB;sh%3A{#n2ZZme(4#prGBj;a6J>$CgkLArM<iK*>
z++=TJ_-t|tUUSIQb_TIHo)spxXd1ov{7_lWW}ylCXd$3vpn)N(9VE58m6uFkn)wZ_
z+C%D?7LWe8Gy9Wdq_qQmosBXn+h5qnT{0#c>FzY>;vh|Wt}}#awWM`gJn8i-o5#C2
zKEX_AenNUb2w`w@hjysGQM%4I<m5pGr2wcBQPpou`~u`>zFR03&I=N7bC2_5ZfUv5
z=p|SZL}Iw^{`+FtwL%ns`QK6G@dF54_~c6Z0!Nbf!A4B9^slN!BAn;R=juXUN2H&n
z5NC`Z4>}H1!!#H73sxF3PvFM^@sjA*nB6b+d7@(qQa;q@>j0afGev9H(y%G>;@jSC
zdjvcyg%}*9jHCAi&HpR2Kl(O1A6-P~7tPd^{Pg4?R>jx7wZBqa{r%(qYs$j}naTD9
znR6Nb;=>Bo6c>kE)za0{6Lb7C1&Z=!uHGc?+TN$}&?9N}0njchsZ+K<BzELf{6EJL
zuu@<eSP2UaGg)(lW-smtp1&T-awqSdZxKxM)s#0kbqLaQN;^71(6)e)KcB96Wux-F
zKl>83ENDm&5Swf`5|UrgJYOP~)^qR|NLD!QpG3#gR&E^c>Z=cbPrha3_IFTux>ZE#
zZ;piju6@4+%|a^&f4p+sP%AbIJr?>%pabv<xOwL6kqQc-3SHY&vHS6dLMD_dY}i~R
zNSkjv1{O^I^H4p2|1$(|^YGNI{dZIWwgjOD8=dsozH9pUM1&I8Xg{cdqTE@2x(La-
zE`y=goQYL|4he#wzsOSbcsFn@>LOw}*8cg<aY@$&A5nMcjB(juaCXOQYu?;0tfo`K
z>7g8b#EZ{-cD)`1T2B&{TbjHU9g4A-=Vy$}cbW$&o@XWOX4|7JuP>rJuU36LFZSPf
z7`m*vlByHrddD%}S>tDvD8-N%SjOa1UF&9Bt7a1Kqef%P$4}DH*&0};E!Aqqk#WkQ
zkP<_0Z4svA1JC*wYJTA_2&{Z*3#mislYKv&>7*Aiehd~N<m={|z`9eQ&kf%ewp;O3
z7Zear9<3maKs>&Me6Wr6Bh0<v47=C4cw25TrZWDjYXLmj`Yy;<TNy(qMrIUd*3bEb
zV0?pWfIQhl5Z}*OYEK)Hpi7c#QN*BCLHA??tI1O3)z{7M+4S;zOM>N`gkjm@N}{gv
zS0kh_Z{#o>Ei4djl$Ix|G$abxFQ_HF&ahtRO1xTb^1M1Rt5(Mt<bRLXnKL5G)S0%R
z{Iu~-)25(`gJv2>DKT`U@^{_tWy!yBOW|LFGSJ#>zf3;$;$L2hMaU`@2WW$wy}GNs
zs=&~<P!lgH6Q%aYKlh2S1L}<gT31v4Q4EEN!9f8@LQyGr`8+9^b2SvokW|jYXtpdA
z=>)>aSU5H^-aI<0lu=deT8{L7AXHy};><;X>bD!QN+uCdWV$GX<=!UA+i4a~Ic2=(
zp&(7?Sk%>>1Dnq$zK;F!IXr4kN;&v7UT|vi-=4CeP>zn*kslP~KAJJ+AxH=KL^X(K
zOssS~{gRFw9OqJ{^sfE;sjjoa?deKMUu=UcP7Ul0lP3C#s)DFG`IHA@3zR2|o+ZmR
zc~vJ@K4^oICt_;j{y2KEsKoA%ph)Gw=Y28g*lSx(66UQ&U81CS6wkaYyu6Hwwdl4>
z&BmW=ve*?vy8G_uD)g?-MtDx=^i1*r(f#AYEnosigBvVSX542y7t?<hYuR70!C}VF
zW&wh&cf<!`ddv^KV<whwDFMsHCdW;^&<UPEX7e|C7gW%^j>YRxt8l*i<wN2sHJ(yF
zPNqb7SzII#k<UZh*nO4AkY!`@yxpr_{&|v{1MNqKkb>M8sPz&>gqd=&M4<R5?$h<|
zkh@Cx7NZdrhHu%)@`_vv(bNqb<qUlc4aq9Jf#?N{4WIF!B-B=B*umj73tBxAdN<f;
z0-KE*_{3)T^Jr+xDNA{wVH6$dS|FWVlVagoECTfpKPr2~VwS+}$~ozgZZxvfoMh?*
z>V(Nvc)D%xim{A)UfXKa^O1b?FTG8Z7ns2o`^QU7e$KS*o(ZEVo@X7?Z?5X>P}=0=
z4cDHyj^1(QQ}lE;Cce%2QT^Z9yT1-i136x_>en1o%9yRDZUInqDgL-Oh=@qXj%)0I
zOTB<Fw%56)4ZqggE$U1kpGh-6c}u~YJ$-hS!|&UUEvI3H*sVkr0E>2HVXd|OE8C^8
zyARD^fh2p)Gxk<{uSa{*<xgfZh}wQ(INHpoL!aT;o)X?7rf9Cr@X@DIMMa|tvnB%J
zKB_7oS<qO6L2z~xc14D(B8|T6u{M29;*DBs%B%aeN1UtMfJfxs1KJWz^iL){5}l#Q
z1xup}?b>u}b__RfjdDy#`W=GsP*`Q#waPcIYatyAN40hkOjV=i&r`zc7pe7BG4(+V
zPBuk(QV_Ra%@m0_iRgAQiTY;MDjDn~+m<*$b<S*a&z$IqndP&<o7VH$&ihdXq`AP2
zfg8q5pT0W~(<1v&;Znnrr>;Igmd+PpD?*rjSw-S{Gn!`+X7?vRRbLbZ&y^ay-t{_R
zOu)9zB4lHuwi7x`Ve50fjO^}ese2K06u%)*T&OcbK!yU&hr64F&i<`|_|vl;H6vr=
z%|gI8VT<^7^mF?47p|A%JKhN{asBB`-`@5b&b;HY<2#?1x^1>u8#;nyC`ks&FoQn(
zY|AS;h`mLwe&lq`QsD+5Hrm>Bc{O1!duwluJMYi+Dd>3hx%8};A+!*s+L0w)+*)qh
zPxk$R1XTW9FjNrvoaI4EPd`{NJhDKeBq#!C3U(@}1*2IX)`=d`r0=#PK;e9Cjfx2%
zl~_K7zmO1hy$Xd7P#T?xYwvKYf&%OrL?;2|#cuM}T#^8pujL)&sdOZ1K@8|q<O~@<
zj5C9Zgh;MKki8+9!MqqRCb#167rPgU3<B6eIe*cZhObZ8aJ#TD8M+oBs8R%$Mx$UQ
z0<8>LD0<t*13<aHwMI(nuuEZ)@q|QfrY{KfL>E5ys)Cr3;g?!?-+elYRxg37boR5G
ztN(h-Xxd<~wc75Z&t0yKCZi;t@bzcUrQl`_wvs>F#?LEFs-M??o7Xni?y>>^JLt*7
zEplrLKp?cV3^pd8pT+QrD%li}_vz~=Iuu&*#UV~v+7tchu3L5S@#6-z+{|A56@M!-
zDKnQ`jN}*h|JF1lWy74d=Q4;If}hdWt1!&oc%{8dW)b9j*-!rFC^Yj?Wg_23jHX>l
zkm7JmY{8g$Pblnax;(Iuw%NUWkRB~7`?qeKb!CNR%+KpzcsCD`4=p&@FW9ZQ)+@UY
z4)B2M`eZ0RAUxjxL{6e5fZlla<JMq$?}ipJyGo>B0l&E|jw=Q3*%J66;31na>F1~I
zXuyP$=6z_`9NO%5#6+^cDQSzDA=n2pXZ+>-XgJNhN!O)mZ}WhrI78mm5q}_IQIVm0
zZp18?Oob{lQmo_V<@;iviH&z^JRAHD?{F%E$h60ZUC{Hp5{o)Zo2*5wl<I1-X|i=!
z+5B4!V^1Sy=GH2e#o<e8x$I;T*LtiK>9LQ>d+A(1D6*0XfC~Em{?H8c-V=X|yE@#G
zSyEfW2tK2#?p{&3;u@f=8GyxGP2_FN0HW*v^y&=_t`eS3^jTvnLG*TF&;FC#?zF09
zj);6Hv@%S<4!rj_k>82k%BI68L-mY}jZrUPdn)b98T4(L$;N!O;D5mkl|2zX`;0Rf
zl0r3Z6jxX~Mjngc-0oyWSN}wXyN#pB5q(=u#y?hDR+AJte@};friEbN75*o{WvB|t
zjqKJz*II7kjZDej#W$zdUgIO74DplJvh-Z;4t(lzBV&aUj16-qFpm_W!?BykmcHG%
z-H(57*DSo~*nHxbw+eSgj#K%cY~Wnwtlj!acp@=5)6^~oK4`ZqZKlT`jjs*JHG*<k
zzA9?v@WnNLHiuz~wuYElYK*@@CuQUUL1P(1T^dO7e;tTF72eV>ntbz=u2@k5TO!3T
z9;_cEP;#WUDHr>dM9SI4>cwS}=CXyko+e!Eb?a+uDONrlE!{V!e+9ygZW$6^XF*%c
z<D8}Z0-+H}ruAMww<W;yOEAs~R_mS=O!_O71YsSrfL#x(NG2LL(lz+1pE1Yhqq<kl
zMg^XYhN*?2eAI7l_!HtQ@;iOrW_oWdq_c-N+m{o(MTeSh#sxrcRwZgcS(&db)sN5g
zqBb<KkJfvFq<0dwUoP^;KcEZfSZRG+7PkuH{R5+;T>-D>)@k5x52CDb`OSdLjlt`N
zHKm?cnMecIk6>u=-psW19e&{E{ldZmS=%?nCwv5iqgyn3%rXS9y#NFzvpW54bKhx|
zk|+txTTh8ifOY~%vK4Ua=1WSn8Oe$hjroYpg-T<V$hGyW91I&07X~J_{B&8rw;5lB
zsZ&9<b*xy;uXwu<T;^E+ZmExSk_;o|MeP1a8v;UpfSf~&enb?p{aB}8r24(mxut)@
zqM>`|nnP0zE%xUx^Dw6lGz^lSpEu-0^nQ_RH@`b0aERzxzA3=ysGccZ>=Q__s<VzN
zo|9dpk-j-Y4?j1=SBwx5pr!B;IGvE(dCk0nX+h<S2j_B+zv>%h`=Zr7cy~d=v=1^j
z=9zy7s)0|NbCJdu+d%4-dlX_|B%>rH9`$vt0V*wO8)0aqWGiPa>Y;cbL`&bkTjO#z
z3!z!{CNfD>+vFo!uMzQQAtFjtKl?kOD#oA2FF_5KOi!fe&s7gcTBPqSj18yh11?%$
zh?5$hNJ-HhOd<eWM0B(OCk5?Ns$6Yik{(BMM1QL0&0jT0jT!5LAzgK|wA~baB)Z?}
zPfB2)2Z*+>)<cL+w&?G1iRh(55khdYtV`@{GD`E&UHxh;+4pJxQfJjcg#0hwC?R$j
zvdN)rrzST!#mfrs!mf-p`9D;pwz%#QrUIUni3wNFY6#$CLF@NcVZ;K~IsZ9RfM@P`
z`R6%cVz1-7s*xveH&C?TX>EZ(pqosIvKV^*<N^kK12apoQi8}L(^(mAe$hl>bU_J8
zDD0QJz|r%fM>t|4e?0d|LIwbz654;}H9*M?*h>5#Q4g4aivWmf3y9d-0MzwSm);@3
zFV%>BFeVu$4L*3GtBJ76mA5*`KDFhOSCsf(+DR|kg3y6coHr{Tfp|xM?Z|X3?Seit
zlxgl&vZ%R>q-$rJW!v{Z#_j$t1cDvb5-&mtQg)5JTuN3g`apv$Oc}8>KM1+1ZO9~Y
z%OS#fSCzMB*<83U@3mqILwg5(S@Exp4L*m@NB_OjaZL<wH709CsZ<bmBoXt|mf0w~
z#P*-YVibi(t-!XHCd7XrYDG7LXN{D|D3oTnD4qdt!UljqcuOxzLDD%YA}FDl4<&z~
zz824_1@tt82aS_|J)(hv4IH7H?~gG2&(Y-hc#oQDvX}BHg})a$Bxe3e6b>SZq0R6P
zo%AeMG**6x6AshB-9RLvE0Q9MDL&5Srq=ExNjcgA(iULuzzj^AK5bWr(-=)}aMhe|
zyou;#r2GLZ4?Zg9V=j&>o8AZ}C5}nRteqfX;GatAtDu<bk6jgDwcY_KHW(%mv7u}+
zl##`u)A#;3TZW!QZNAr}5nH+057_!{T)sqD^mPW@Byr+LX<SQ9-3KMTtKOXNLnd)T
z&4|{h-Ew+uHzuLjIZ(F?4;E^1j2&XYt!k8=w{>I@@_~qxLahr!E&%JrR!wg)i=f-p
zhp14Ou}Y5Dq1n|RFEu7FAGQq-vy%*|2lT3g!Z>;~S(!6Tq#@p{b(r`Tc8W2@shqd1
zTqy=*F4sREDi#9s%!14Fo*4anlt0g<yZISYsyPUUm!jBulc-`Xjz_gF7Be=c9APJK
zG!igX(E|gLAsh8;<ubhd({&{@?@d}Tpzq!158W-xv}<=Sw_jj@xz#auJS{{n9Gc~D
zb(!ru*S7xLJkOQ%3ho!Rd-u|4wKXJO>IPu#OKHE+S0R%dw80xM$`-9(B(ppCdEpQ0
zGU#JjQNb?LF|ftRvCG_$xnO&e@_)(E`1m-$R(P5v9GI>9oLO`ac=rEIt$<2$Wo5;}
zlKEWG^J_0J`SX-wuEt63K6!L<W%adE?VT$x!B4Y-(Dn?^!r255CY9(X^WcyN&A~(M
zgJ{-*x~o4~H7Csh=L;f7sFMz*Lwtwonp*4kEA1i!4Qy*@y(C%V>z*r{HID%Scrg5T
zIj{DE=4^wAB@N;0w_IY)=)Qpj+XA*Ra?9@4Kb{-|YxuLf;6K)kfZjB6=hvz0Nvg)}
zvt10%Y*(j#*#_JylHE_B<iF+K%xYD7H+qe!BNOw2Bugnf;(J0#M!lj@@8gqA-=u0M
z=Iy1&&WgbO)VMU_U)dV#?6U5DpmLnoMBiAeAd6e_wFp?DynVW3G~Ajt8bQ^n0{L*S
zDQW4iD4>g!6MB9Ktt6cWnPG96+vmoMFWL#UdJAo#-`X!;e5@L4jDH{7l!6a3_+g%h
zx*UVXMKEhFP*={s{O?7(yQ3kZmmn^A7b3KCi-l~srgiQp_yM&TZ=CO;4q;5<(W;J(
z2~KI@we7v^;`8dX{I|BOQi89}EqUBNp&ouO^5eHd{rVAkGC1z8ouXc{)cIr}6YJgH
ze4(KfKl%Z~d>`ERok0AOIDP_2E9AGps4|lsnt*A5$hxEd>8Go_hrQ0e<~1ozWoQ?D
zygsMcinw&|uOYD6{)TTzzdiA*XYg#fX8rZ?C$vd(C`n)$5<dxiu8E-MP;Xtp9DF#r
zgdN@Vv-g$=@VuZ~ai&I7z;d=pAcoXsRRf`uKa<dhV(j*eq?D^I&AHH3Dj8XZ;*eAF
z%5fUiO5{j0b0)YUm7GeNSo*7{+DTI+$vLFl9QuP~w8aN+`9U24!)3$YeQ#)!oc>rL
zw!bZuF~|y7>e`bQtih*DW=;p&F;lSA{&5uFr;l12*3bK>BlDNU)HZ^T^U|6S6C)r@
z7#Hn}v{7S?8H)`=v5DeTfg&AkSj^AwBeobxT55&^A_wgQz+7-_Oyl7&yH|<cKl1H@
z*$Jl9^>(s)muR<c&(ETDbZQ%<m^k)R<sb~c5<&jqdzfF2nI4@q-@Dt?^Surh%())o
zyojQ(3Sq>j>c9Q&KER9O;NYPAQ5}MiQg}PpBadMDHn%Hkqg4^+wx*w-nlWf}?R*{6
zMqgGDvE}opB>zu8AB3zNyp?#qfi=DtZ(`nVymg!-E<zL6GSldJeJAQ0v!DI5z%{bF
z5YCyZ5uHH1CWZ+^r^Mz2M|jkA?cg9cq#ySl@jGwkYEJYXch|biVjSxFyW~|pm$U8G
z!4cg)_-F;t6S+R~Da)ugJ%Q2-EQR<uF|4*l2W(_c)RXBpaZ#bMp6oauM!St4s%%O!
zwQLlJlA0lw-yI<oe8UMV>9@Gu&x|Z>=6lmgP^`U)u$ex2lLa3+(&t>WLxUpraY^0y
znN*r#f_Tjwg)mRGD`iXOh0q#mf!%O|q2_&~_r$8x??hU{n0{@0UMyiUGp5-0g_U)r
zIVIiZ#oxs7nK%5hAfxV-<HygOSZuC*gd8#spMqih_nTcWq@0kuyAXPl0OYD9pu6cp
z*ROeJodtOk9cuSxqxBcoODirT&S%h>6?Tlo>exmeMHeg;-z1FWCXoYZYBwn<TMS?8
zF3r~b#);(i^$~z@@{9lNkQOxX>`6+-SXc@pu}5_sONnq;YLVx3EW3|(@}1k-e5$;q
zrcS}bVF$~v$L$-tGEW<C!034B*x7Azw_~#T=LWm9l=5ot@mhKyBm!BuMn8#cj=U$~
z$&>#1cnkcs2ie@R&q)IS@Jvn7pEX7kYAl=i)pwgC`f38MpSAg(V(hWO?TC~ve$d|W
zeti$PxoH1W9qo;O5VVgj%rE0!7HP}~)Wn1|_LZ4^B}sdN29Zf(A5;1F^AqnudghL;
zNbwWuUnUuZ4C?Pmyoz^shy9gP#SXlmzSnG3&i+B}3;wiil@**g@&MP)5C2P3xxHw+
z0F~Gmv*8Di8zv_?S?=nP6}U;STsTu`i%~Sl;*!kQ{|c_*j{`B#)@Yn;-W!ZJD6$)z
zm)m&WWn>VFU(4ZIX9Fw;*3^-mKJvGPWcK1cx%YC6p+_T@=G+G~?;+LI)%BhGKcgvu
zFAAs$td}X+*x2-I=0$#2PKk(#O!uWfO)hC@Cv60I6%vWy%ME9I{?)7sdtFY#)XBYw
z7SH!JVvhr=6Gh}AeU5Y?vej4FExM0){hO0MkzD%PG-m9LtxFK7)^1{UTu72nqi?5g
zTX8N39mqJmhO{Imce>sD87mUDs2Y2qgVmL}#_oMWjeod14LBR{^z#cL2Ck8ieL&hZ
zj~h+FR`G=UweVHI{ijE+`Dlv6dGiC#cUc60rP*$4(k-;H8eS(Q-aU5-6{>hbxPU$K
z6QkefYQIjO+sVZve8lHd-Ad&U+W}{#&AMnjE7qdwbS{eo)IRd@$%J<0OMm9EGQxOm
zYL)wjU0B(RhS?-X7+~PAk=A%+#JqDo$Ty5q2G|)P<oS}Uib)~^xNxvOrHn#VX{H=O
z=O1}*Uqv{loni`g&cNFdk%hzJ4c(rkF?$`Ie{BixHZin_$iWND7*kBxn1KZivS=Tf
z6~3nuojwb!d!vkG9o_AJ8^RnPanCs}3go9}=T-uJhs&I&%}2A%_o{cSLDe<{0O8RQ
z*h~QkvJW3VnoI+<HK1%x9eE;j2+b&s2z6pCK&S+MOUQ{-{)kSdZ^mTGT##?EDn)o5
z>9Qb*+g%db>@8n0KVAHl`BdTAn<OIT{B!#yalSlC9%H~Uz2o}ALW98z8+-3=_VfH)
z5nFD|ii~fkqQ&<-7I#H?d&0?B(so_z8c+z$!VliBcQNU5RT?xGeCDl&-8fw>5VGSE
z8>b^8P@EDcCBgCU_Q18@roTn(mI$p%twhAd<$f3u&xro&5+a-eRZ^;5!Q%TxjZjMR
z@W~cJ`vtRhgRP@Yj3#?t{bw^!Uh8$q=W2E^-C`rvtNly%V*!Xb6dG&@*BRdO4A9t2
zUE}bsrCa6R;K-YTWil)C?KX}KIWDmQ?$Nrp*xxjaexU-<{M%Vple$ku9G8oJO~rTx
zDU#x*_NWczFkK36P5H&;sgar5|7B&~T66z$0CpaZ^Gu0qaMgr=?@vMWp;P%mGOP%y
zSEPHbT8(zun=Wk=ZOCtnSkJTLN*&@VtD-}3wk|I(p&mbIg|Y~MipwR=BJL}7I879X
z&0G6>W+wBc0M5G-Q{fkra4tJBKPUz+1fy>Hb#`8INtCvJ^5JM>*4|Bf_FgZ3gb72k
zTyd2y4~-+K#;n`g4OM&O7Nj7+Wzz+=uQ;t{5Yr%peeM6XNtEx^=mn#m$Rz6ZtN~iT
z;e}j$iB0WlwrQFs2EN?uFd6&4?}sNS|J%K*3&#g+wsEtWMv?c9PJ4G46=UGPE<qlx
zGBfTi9j%JZ#+#?jYdsuu-`^+fx;mo>zboLtxc;GGFsUySvmr-txd@aFjP_z#qCk1F
z1|exBmxk~pl+rQLR^|w_Rb%+e)tmDtJFV&UmWCSE?76SYWSV1S@w1$xuOa%ip_K@k
zp>@zWB6b+{=P3@@^61mL4+4u0)>`AR;H%12=8@-be5|9AV8I)YRfMn*N=*_%B1n*t
zgYV(5%3rz%ETmXtfF-X)<J6b!gn+cM0ZRewKl{4z%jtqzH!|BI+lyDi^SaH;Tpf33
zYHj}aV86Bh59vN3jV)J-+zIscBl<-_tcx-e&t(KjFlEh!Z2Y!7D-ZzMkPi#LySH@P
zt}wmA!SJLEufTg;;`Ol}y^-*LN@^NHC6VSl>t_-&Rjp1lgTrKr$r!^8+_q>N6YhO2
zA8yMonQJE0#4lqFFT&*SWND)8Z?cflH0}e@iUTJuZ2`HTif|FVr+|eUxfOk^XH7Qq
zssF1xVMTSXd0i;{eR@eZKx0$isx_)agH2YY?%&e=F6h0vAO{XDNg864_P(<V{3#(B
zF$XgdI<+dCK$Sj*Op@zZK8xt4kDff{Uc-OL_lrg0ai_#&%nj+UA4-V_ehZpO+|i`a
z$=R~=PQ8l+;gFrZf}>?=DqC9Iy~Hh~S|i=MY9dxF$XFvDJYr<s&b$~rPsI1YXcFPl
zD7?pWl{fbdXBtg+c41i-Axf<sd9Yo=4wzW7sf)Xmly%xx*zk+Thd*eZ(5sO+?{>h;
z)b(0x3a^xwrbJ9$z~qX(*&{4Z=%N9(*AO0X3%auI6PIPjr7V7|z#I0>BRnp{8}WSv
z>a_E{iBllo<3IiZ5M15|WP6%%werN|n}&W|=qhfbmD=r4tZ4Bf;{(7%A?X^aBq^5N
z3$l@sbdggf!y9;y#ah+_+Uj0Nv!WphVOmzBnbGgZKFa(%CHA*6<w@wSUO4V}??@Rq
z7;T^HUxX%t&<ZJZSKq3_35{UU!d)$lwua9Z^s_QqL%*s$?vI}ejuvfBn<q8?<>V40
zNx-f4hfkLr0FVX2RXJ(Xc6~E<Z=BLk^j#*BO4j)5@JT@#D}q#P0OEIK9;J6n`|#Vb
zLk3{vd7aP^Rx7O7J9~Qt?Flb<-p{2i8vTc^y9E-w0*ID-14ZQf%w+aw7+qtc_?UWh
zNR)YhL1iLw%>q#+Uk|lMzA3yz=Zi2rM`Y5+pN3Gd#Y}*@sM3aO(1fU3%X^%Hf9w?X
zlMJxSaOu32PimdX^o)t~q@&DwYu<zNT25srg2k3fpogKp$!t+Xj})IaCLekS-Y{QT
z0QU&G@KBJbxgcXuY9&J@B9`7s*D+JRE_azNDG&N3n1QJ2HtY3!-1k{eQX_G>&0nHF
zmA@J-i@}S;-A19Wb8b9|otwT1OaFrf!ST7UrrD5P(}`{TS<w6Q2EStG>F)x$g{G+M
zq{3BE{%7!@ke&+-OFMPP%0fx?622EKhmtn$c$G_u$hbqhw|1f_J#UVvf>Brdy+qF5
z-kBrj1b##z<YQUuf2ziw&$iBgFh+#Bdc`CPg7M_~WnlO{EjdBf<&AH9-fT)ko@HDc
zB%cUpT1-<@P1pxNM9Oi-1Q1y^XYTTr63nWzsDHh39b@^5?EJfQRaadeSvu`nZ1C!V
zd!SEUbzE(2LIM9;*b1#?fcm<tC2?*?YctT7rT&#XtXx+SYh6wz@YN5aZmxwzN0YBX
z1C02f8)h*pr+TBVR_t!jQIq)H9FKnxP&uhDoBMg;xp8?gGI$Vp^rE@06w*t0+|m+*
zIXT~s9)pLuOmie<i4_y%Eq)qar}wOAluxOGY`9#Qp6h}If{B^-7wydr8(;0)feT`X
zBzg`BFv-49usEw#XPW(WwUE%UJfXmu=TOWNuAjUUq#L*ST|Wn^7euPBf6I+^juA{k
zrHS(PajKe)Y`3$Z3h)c(;gV9)H)(?T;IM%=F;UKg!3D-47^@_O7j(>n<Y@&<F^n6`
z*q(Xc!a*UuX7MGA;Vs5IuE0Ro0-Y%sc}|mAGJ?;8m3Kcch#e@`Dd6f+U}+ef12^(6
z{kh>Ov)~R)_O9+4*4D2ZtN?rpfM*0QaJ9F8df97P$^<nq(<8fL#<g19(0)@-OR;lu
zpk;rVN@tb)@{R(nbvmk_=WsZ<us^-emLSUZ^J#5G@d_#iHX9IdM`!!PrU+yB$nd_o
zA6PAvU%Dc*1rP_f7d|_v?>6Rt*i3EqG)CXWWcz0kK8>v77c#E(WmYi(9lmltYC}-`
z_)S+aM*7=4iqR7|#`QxST4k)koT+LOe*f?*h-;<92M+akNP-6}8~U%Wy`*p~KoWo^
zHFhRhEp)wZE!i!yj~MD(1bH#w7(3}no(&&g6}@a*Rh=P1ZF%ZgE23%OANr$<YI0P{
z7%lY7tax66JXfbls-YwukmSqV(2C`BvCaU8j#1x64-O=>MB;f`K!t++(5InW8{>}_
zrUXJOxZEy;Fe-WiqKt)}%$&w1UL6aicKpBg%M=xJpD~`!AO87n6C{VDV8g|3Qc4ur
zGit~6{GTAie9EK6+b;}=KB9@Im+o)>@d>~%M2%VMTvfSQFim`kpW>b$OP9UMvZc9%
z?)ucf!;gCrqaeNNLj#opW5TGX{bw`hm&<;2+)F(YS~i0d7Xm9)tf1HodF9v78ZXM&
zgqr#G4DCm~)nW8Ij@t07rZ*MS)VgGDa4s7B{Ysv2vhu}j)4l8vdTf0(jhTWEoW$mR
z&LGnzvc>8UVTuwPw7pvWH@Q6<PH0Wbx@WMU=#ghM@jh-JBK`v-ZMOYPz@QB`wiDCo
zdDqL|J65Fi@g+vXb6AElvYB}4TIfS>CSEG5L#(5io$iDR95ZQA&8n`AI}4)aepkN6
zv!Uzb;1Dn<s{Zx48j~^12u$I9r(HWc!EeNJZoY{5Tsk8dj)#-B#v%9RW8IRox_i*q
zrubMsYj8yMp-UzXy={q2YIcQ%1YaCF+D-fX($Fh=8jepAbj|=hNRWU|+CK&mYyo6i
z;PwG9AHWQZJc!Yov+!%d#)yc-hhA-w)Z?Hx6=_n^^Naf8!YWxKNOeo$2kO|7j-%Fh
z)m4*&D=1<o21)|<-dC4?uHZNJ>t}`<!PWLov@2T49gXB)Z<EeBgzw8_$5rHiifGLD
zRk$X`M0dP5tc;Nv9G9YcNe}DG5l>re7wP`@N0cyRAOt1dW!~D=3~aWxk7<)ME4#fL
z7zN%xePzO(CW8gsqo+U=*+xZ8M?VnF1Yi;+i8*GMGAYF2Q{2O(g7{={q%UYrB`9;9
zVlwos^)Zz{N;(kLRYSR^4iU-rV1u_k0ZYPYDZe(rR<^EFeusMw&tg-rv|;>+1lWC>
zrt$37BrFfsRd0d6R>OR{oqmbj5?3l12>sDy3GHWJV`kQR_6oZ$a(~N5ge4P^1Z)I;
zng^BCiJIvOP_%RNDFSTim58Hw<VVZy^UZJ5^9dBI^zn=Igsn>>r=ER4CE$OSoTv3i
z{<Dl-CM4}^;sKkB=B%pkg})IJoI`6ibQs!3C#d&qoDACKnH_bN7Dm93Q<`$qxXt!u
z>Ey5M_VS8O6O!END=U!$w5kZA!nY2`QWR><hfge=NnS9GsnpzL*sRKk5{)URztSJ^
zVz_J_A=X73^3^5fh7UP*KdhR8YL&I;?{SDu?v^K}kdC$Dge|M|P9py<CVUpF!|bWE
zb92m5o6mjMFf{-myR*pA6c>BI1^^xxUckw=(=@G8Sm1W5Zw(~E3}4TG{u-vvUCctI
zEyFj*qM<?65XvJ!Pg@|lwmaqqw%T7)R^FRWnkTyUUt00!pU;e!Ys#GE0+_wm3s?oq
zH`7Ucy(Z7BT%G26vK};gR7s8PgC_faSY#JBvE0s&vf(V54xe1M4WPm~|Bbls%)8S9
zO&uV=GAD+~bKH9AM<?Q^+-@#~_<mRuFZZJ~=(V*r@ri%%>=OobTV|j#cAMoSB33@w
zd@74CYck!z7W;|n)hM1Asl(m%yB>NN)Bq74&=-S`Q6i#;i}Uk`V*e%3b}%Q#m{iM2
zf2<&3>frv{g0MmqA{5;OYl{^jYw)Te?NID&V}0+AYQEUciY-iDjp{6BV?dOp<|Tw%
zY=P%Y%I?@W=MpRa2?_JL1Y<Jgc6di!ng=x`lYMpPwGXkZ`>G$1&^YTsaR>mgUn^yu
z9OwMlkLP1;s#>?h0WB3hth3e5LV6{LAW{ZmRW<E)P}$z9KE`*;r2_(kdmpdZNRn<K
zn@AUj>~pBn{-t!zWBx#O&BEw_>uunIez+EVQ*xnbmkAM$=g`k@uGynPw%ywLv0H_l
zh}Mk%cL0<t&QD4voA<|(RKq+*6vN1+DWnd74jb3h8F4EZtEfa<fDjZyww{CMSs~?7
zUnOWJNfvIixT*m$t#d%v&=58X`5Q2Oh=P?MFHP{$QcD!^nn`dNO^3I*)s|!8&sGz6
zMTL5)80})`@mlj@R78U^HSw0-PEXZ~1+Tu}4xz67eS9kPH)qTtlVsF+3GAAHCCBqw
z%J(z1o>~HF^pRRe{6JyxlZ>4z;q{gDSdQCF)sLGs-%F6md6-Gw@cwHhI=$~THemkV
zCt@%6x9A^mc}ye*oI9UnQUE|rG4WS!gx!YOCdTXW072wb8d&es1e`~kY<BnJ0LwB~
zpesFI{ZaI--JW^4bHGhx9&KA|#w+|i1E|PXXtk8}_vtkUeq)!`Q^79JJGX=Aa+24P
z%3Qyj+wwZnP|Cj9y<6aray(Ao4+8WW7hYr&r6lSDA40MyFfNg#)rj8ii)>F$S4%gq
zzB6|8nP1{bUx#5jZs=NuGY3*%d|DwkBbT4X|J%vVB%|ZThOIK@&2)(4Rmh%@ootA5
zQ$+XN#Sqs@)735iJvrMMaT|q(z4`P(>7(~eE}=CRU!Lg7lqbEcff`qD40_stL{BR7
z#1+NyI9SY-yT%>6V^B#KuC@Lo>liD$$mDhAvvWh%fXgW*{9csqhD_5E_ACk>)!3Bb
zo;*F7CHV4Qic4jPpgLpFmpaNaw$^q?dHgLI;cqOJA^GH*=YA!zM-hMozE(<9VMauL
z*%Ymoh_3Ty(^bM3DOp+BZE`<mvUR@Z8`m&HX4m|lCLX|qwdT(9;ycBSJ^{oD3Y03*
zb#_88sp{vk`8jp=+0=?0!?#n==h7c>-v>v|`WH<yMv;>^F$I!7oPD^F^rq3?`~us!
zOz-v5Xarh)f17V}9&QgW@m~G?{e3J}L9K$HKiaI+tT5xX#$;s%#|X>B0o6NZb%uUH
zf~Qk(d4B~o(3t&^_{ND%>zX3E?Nc2(_?WuMkF*Ni5t<kCgvOozvA~G*@V&<ey%?^o
zI6e_V|N8Kx>Y-~Ndftm>GJC4puUuTk6Qe!sRa0WZ{CQNbKJ=87xDS8D8}e0FV?9iQ
zVrJjdVGyns`K`=1VigMa8tFb6SY5C?^=9tVr?U{uq*g4KiBpJo%w|@xzL6|0Vj7ox
z{a8hE{DN~jKx86<{}w&@XmIs5+ky%mU&_?a<#X0wsE1#?!`z1LkDn6(|8(^CoV@)B
z4{7K+P{jRWZ5`kL>x!#uRQP?a&se1%?_96C0gTp?N37Y`{emW&by$0!n%_xGx_dyl
z#R&^Z@m;Y<SiB&P$_#Z_eVlw$R4!pNEuQg^ARGftplL~DH#9L5asmme!0rUE&a#{?
z-R@P7IFH)A^d`(ynVr`Vf4CO*`ZqEdgJDhsah(ixr+HcRg$8HM#80?iZWS2j2xVzd
zkh4gFqFnl{gI^37W;M`M7%`WKuKri=05{>iLj;f1EkVcK<}(oNbhS6)YW-g6zcGb>
z=YUIVs4I6~D45`BbFiI7?hC{eF-FVr2YW6Az0S=ci{JL`FCJDgjulKWHmT)BnCvjT
zC$Vzfp6knezIokX=D~7ffMebSP2x9u?e^0h=_*E>xN3A3%FxacXIngVwWc#+Nuo@v
zR<phJSK1i$A3Wkrtwnfo3LX+!Do8T2X^CXv*<tQnxt+eT)$r2?93p3b4-AoKGa~ri
zTmLP4-QC0Xcm`K2tN#1S=_I{FOKeWIhjbXk8pklCuev3{O1S&CLAKU!!;Ri|CZvn1
zDY&92lFkJ&ZBwt+7G08?EDd^Ns&(+vmLo+vYZl)JAE@+HM0tz_I4^@I?jcl!sM4r}
zJ_Pvanr?{v!gpK$+Of*cKX!lkVdwk?l~|VXC!ICF(uUTzrsK8hjKp!1J>X#cha5y5
zeA2n^pN=2qov+P+--%26h|mCVaF|5EKm7R48*J>Z-U!rxVk}t0uP#ss5M*wbOHCB(
zMy}7Y&SA5ZQdanluManLokX^Eew<2BL)>hI#+Yk@8iTDGqtzui8~!Sq13}KG^*mwO
zx~xqT=~6rY{Ec@r0-3$1vyAqAj#&Nm*QRcG<X|46(GV66HZ_W<5b9`kJ``Q(&n3ap
zCF~UtP;J^HmgN^8>OlWY)v!(~9vK)#G9F`<U%L&($E=R?X_+^3t?xgs9Z0sCa6)!V
zV#(oK0y}p4hi-ZXGyvLq{a(WVjGXZT(D<SgGp9yBi5oyqqx|%LzNOHjeSB?y++cb<
zcuzEKKq!7ND&-)@A&x}!JQGjdQVi>IdgTp@CuLBA)FQRA=!Eg_<MssRV$4ts9~Q?G
z<^Z8ekI$&#Pil(`*(YbntR)(`Ichyr)d0&f%}g~3UM^AAYEg0?D(7foNf-4pPG7Ak
zI+SpJ+)QJ9nkT+2ZEIVG7&p5Fnutt|3C_^Pb)y)wMRXqwsiYR&c1#=^3wZ4ocFnZd
z@MN~(oEwozraC3T7-j@ny`+8q;dkEc@Bn$C`aBd@>r>2as1r#5DeCk#v}9*SF{nxH
zi4b}V`=2ti6>h;M5^i7G{sP6)Z0g&EOro(@nPabbBe$$5oG{Is*k#%{KJPEzy(iso
zv345u(+;qoDJept$%k~4RYfzdF9s0pU8Cu|KUua*$_y<obl>iud~uqRq0a1AGOlrr
zujWNFWErhc=rb_Z$Exp;6|s8Zx4*H4*40(C|MDAzg0vvy9eGpayN!nM%KYA@7%0G|
z?oR&q@6Vj8gxAe;|9|Is>*M#X-x6-e(matND5uBv2;S;fNKA?fRyw~6bgBExnv-67
z|G@JGols(@`#;YKZ<(vB9Z$qNFMoL6|9J|e+;}zhjd?`)gG{x5k+yb(HV!&%SdzgI
z;f<`{e=^_X;=Wn>h4t!L-<v;HuGp;k3bi)L^D9iDxX3+q%bBO>MH?8fC`zHU>6TLU
z_&?4zt|5YZANTkb{~8K${T$y=PW9hPAFViz2>3y;Hv(nEsDxgE9aPKn0@8xep&FcL
z7;J$m8&f6hHIb2wMN76MwWEOvs^8`33-IznP<@r`PQn}HIA>@53~hM+GHW{K``+#S
zwg@a@fE4!>vga3w8pP<=kDSGD;KX@uE_0#l$YsdzC1Bk1kdxK5**>ecBn?QQxs|L$
zv~+y$+3UO}FmCsaIzJZx6nT`Nj<aD1hYdeqBW?uSxPLWS5c<eiQ0PiOW2w^kj3m}w
z9uvQtu^*QxT6vlES>yBvutay<d^dw45&gcKJ`CUds}ECx4W0@@&~X!P<45WmAH=K|
z1Zv5YE2@`7yAC$!dM5SuxD<u|t3A=$kn{yM(%H{77En6?y?ClN9BRoNkX+S2y4|_l
zSOFGDSb)EAZ`9}CAQ-ahnsAY1tC~)e*H^Vy2HNI&)?LM*iQ>kvH{8+atx|7sLyN`@
z7a8_QoFtB&C5}(V7EIQ6M$-0|C_;#?F7!HkqA7WtA0PaDFX!|Cc>ur&WgX`kuU-4F
zaXGSQ4gThTzYbtoPjIB=55X&~ZP7ppj779!WKw7kzR0&~XJ8iSPT7mPqm^;7%p|X+
zj5;yB8!bUuPumgnCL|v=uEqU5{a76m+TH<)f>v2&NcY_HuV>`j-fc<==2zQ%ziD2^
z;<?U3(E%_#f;d*ss^SsIn7K<wl8sG8(D6k2DK7aC{tnV0gf~3Wg>=M#IbJ1JICZ0Y
zNs3=$0Qz!`VDOS9R>(1e^ODT}uZNrJnYRXO{!&x1#duq%5=7uwsXrR+DT9VjBKDig
z8V3q&xqwktPxaxUgIl^@-u0Q+(m#O&rYK^J|2#&%cc)7FA|WMpi}W4^jvGADTh6qQ
zewI)}hRK$)8y|V3PdzE_4PRFePF`OVX}pCbG>V$Fa>R)JfXsC~A7TGFn3<6w{Ghzz
z?)Ne}wuRz(>S>PJ(>GFcpqRwFN19_g=22aas-{o>t_Ry9K0)OMeGqP?!rokE`4Cat
z&Um*zm6;r0O;jl|8B38Aoy(d373LsKe!xO#e!2<MqMN3;8at!?0&cR)n>fGtu^X%(
zf9WVFh>klW=>25J^^ll!0k`D#GaI!?XJ`MY1w9^TclX%Kq~YeZb{`@+`YCrc;1<-=
zdoSU2heb*GDak-9AtGZ;tTbx-(2@=(T}$=qkAPi~q7IqhZfHE1P#tOYs)*dd_I&QJ
z@(K05UF`MXUQr1NY&5@{RN9fOvat&@=7GAS+LfoM34)-Z%L;vUK&jl5U?F9GOoroy
z|COw6gn_JGid49TNtSEf9Jnj)?j&wKL~Le@_C8KVg?I85dU5q_FQl8TP!aO#L;4$j
zJDl5;bY71+o{R~vt)FfUIF0Wx6Qh~;^%9%*rel!$4tD(7G^$V@eP`|+1KHU^pVF@#
zr-T&+x8sJzM=nIK6(e9#IBm!36Wbh>_1}YMKSK#sGO3w~!=mCo_kQnDWmphBu*KV5
zlPg2e0}3$6*OS136MdC{U7@!&6}0GsQ+kHC8$-|U``={7C0ab;{zNAybCG>S@vQ6Q
zJad1r(-k>K`pp>4imPo=7ks*N<@)h(yH>&6eDyrv0Zrr;p>lq4qptmOcPoX)%*LH>
z2r!q~YDca1u08zasr^-1_nWTvSGHN}5wRKHRirgRIPvM~{p}(>pxC42snvHRR+q=e
zEdTul@l8eVta-Un8xQVR^AGcn4|H@#-xi)P-C5o#TA~f}N1S2~{^W!^m5=2V>#epW
z>5oD+Lbq4fkz!BDyS8Z&bq5{jNddC>qB4Dbi6WM)`~(Xn_r}j4Q!x6dtclP>DE@f7
zsHkYSWrGUJ+4`!Cb;0y)mlPyV8T_e9OkvwECF2_-di)q1K*ca({U3MV{P=Ipym@)!
z=m<ZaPO>KihloA%rMK8+GZ3}RWyf?p4{nUErlG{;uCsBGf7$KCGTD9WYe`5VVEpV^
zyRdB#FSxT&%dv_p4m%BH8$k8=?nn((K=hgJkAMAZl8-Fs%Z;B15vh^YIpKWscp-L%
zPj?~lt@E9S@S$&FanAzjRhK>T$Jm>t<2xZiaG4Q}qJYu>X1VgPbla){wodG6Jh8?|
z<s|H1Mpe)g%p4T35O-eu&2YsG#NizbQDv5NJN&?5%0L3F^YB~LF>`2O??xmb79_@`
zllBVNtEnMj&Z#-u+=1(Ft^4)OT|WO+`9W?k0e5j$Y60w6_Q(CwE6!^OTJgJP!^Fp!
z)Ig)i5vJuzZMZ7U7P$g>dx+xe&Y=dZGx3n}asxxLtZU;8X%?zvb)(9^DtN*apJQ@X
z)Aj}+I$0RRS|&&^^WLF4F=aFmYd7b)B2%<&HQtbj^G)SEU*hJg{>#Aj$^z2H>B$)!
z%r1mm=0SyC%?aS`*~oJ22y!JvA=d^9mExTqAMR=DSDFYoF|PD+;`UvthZ^Cs(!k~{
z=FVGQt{jJP3jZm4Jj_1)X`D+ZO(4su=KDy&*i;`i<g=)}JD=_oyQi7NFg(89u7dQ@
zNX8qB>-^iw)*J_?+wtdw7TGA8pmKyLUITWm=q_gM_S>{fD0-yI+i5>|xfb_|y?;0<
z#l`zI)Gw;h&O>hlMyxAhlCurtAn^&Sx%3OWym5$4w!#qr2+jp^KvsVE&;xt@tTF7{
zd%AmPnKcx^`)qk!^|Nzq^1RzNaLE?t4Hp_58sf;6@QU+69dB)KlLG(}jhRg+2{iN3
zOoQQT*O#PNPmXriITGJLj9(Rwwca?|-(RNNuI&l3_|c2s@1*ah=N#*7DcBz!y-o_E
zRJU$r3LN)Jf+bigWQc1g|Bs`y42z;|!!Rw~FD{65Dz(5;QUcN|NO$)F5{h)EEV)Q5
z2uOo;BV9{(hja=^OXD|w$H8xY%+Aa+cbw-H&*~_Pn`Q?C4YZx7jx=U^f4vX~tX*VH
zViDm{zB;jS)PsPcbAP{>uPRN6^4TiyUHoK|vlUWB;Yl`^o&l_HlWf)ZoXG#-z;qKh
zKmQ$-KmZj0C;%=9V1WbDMG#JVdjrBFHjVRwPey3e)NVm2g0Xw;QuFlQ&MDzrS99!V
z@M&5qpZP!)nRdjGp~YjAf>tUS!i)uW#{RZ6pFrog8FzH4aUanL7km<8350}{_?h1I
zK=!u=dM-I>NksE1?8wlRv!Xv_XhVo;aG1WAgJUBy@ro*~GRH)ObJY~d75oc{xkrz`
zlQKCk?*ir8yDxq5n>$2QOfH9-<VdB$SK(+_XAklAUD;JsegS!~m2PUY*1eM>T~iGm
z$S!Ya;D|R-M5J24&8L5%m^SX=0Q9tWbN-&3!p+VoIy#yFSj^Z22AqQf?r3I?nvQml
zrTI48oaosc-~2mHJ-JdzRDKd>A*C4uzJp(KZVx~jq5WwTGPUlR6-pSTY0}Uiv&Zod
zBMX!0l%cZZ27jK~%`0s%=bkl^jA#ialsVuC4C}8ffe>ALVE{zy4S%3H<hHV378S;e
z+6Bj*pX~jhDl9Anu05qSl}r#U-z!Gd-n}c(if4GaNvf%k`f1w2`vyZ`vHJe(_>mY`
zk=Nr%xxN_-h^?s+a1=@>3`nc|-5LP_r}xRnd@g^N$M$MXe~O=JS!T_6av8rRHn66k
z<1dnd=itD8J(o>=dFR#copl=B<JT~U(PuJ3-v`<vtu@T8tU4D0=pzQ+RZn?pFunZI
zZ&MapnxN_e9>kj-aVf4jB$Y=75^8>tMxUpONrFkYQ)A&94v`5nM$?m#bxuP#IA8CL
z6!nz~Q?0%vGZtDPNypdwDKB`UAB>we;1Wejk5j7qa=H1_ljehGv>t}FW!(!m)HJ(*
zR(sI{9TPYg-=9upJz`auwD%nFnlgq)A`k+4b8DV^=0r+D0C)1E7eg|S0!lP4#7+21
z5~Rch3EsU)1Ctk^R9W6ls}|JLEeAOLgjLSCBy0SzVyTn7(ECu1w;ACt-=;Y#yNhGL
z|D~cehOKa&&F_@~-}vUF`-IE)b?;wR{F(g*5sdpnsqYBP&!erfk&OJRz^m1}6)=GJ
zYZtMe7nPAbdA_&03;JI3;e~BvZ3-L=pn-jDPb_+z8@ipFd%w?BoNk?30?9}~g7OAX
zvi~EG4o27i_-4rXE4-r;V=JcrK+Z=**iq}j3R6fR<P)_Z<l;&PXUP-LK{{2te?f!8
zdR)0f_cJO^pYtPK2rbq=eo+ppql)>MJ7aTAcjjvsh(rTavzR`mG)7C;n#D6{92H90
z(`O7E<&$}pHr}WLT7;`*99M!N^vz8DBY%>B4qhr-(#50j+?VKtQIo>^EHK$5NEzGi
z>GRu7^gjQNoP4YCS1BnXrBt3bTx+9034q^*9*iSmJ5m1}NT7Q!`S-usbnLfpbsF7|
z$<u2HfIT_TKrN}|a9tsq6(^sH&3`>*WnV)RlFlC1*gZU<W{|^Ln|IE$;3zF!jby*`
z-A{~z4dC|0+fj(pnwRh5E=IO@%yCLv_R<P7^=BjcGO}MneIFo}qdkdpa$T%>@n&tN
zX{{{pHd_Z>%p-*9Sz8xM*Owa=@^myJKH|-Z&hv4(Df-|)uKOTANN1p*`~$v(fqLd&
z8k(Bq4a?>ckj#TD7_gD{idFN!f_vCA&h>WXoz|BZ#~Nm~h;5MleVNv<!mFXz7Sr~t
zH9w4ez%H*nes^SsiJz`Vr1RBdW*Bae+WV=tDKANJ(Wd<!5g`))w%&|q%*S65zFvrq
zBrHXYh~uHZiTL`CFD42-`UlwRp8u4R2oX1l(rV#2^HI%71k(m7z|+6xu8>4>?mL2F
z0l(w?QGH}LdJJZ{Nqa|4;pzsUWde!QN%{r?LPH~yj1A9uj1#?&L^&V%l4CXYl3yy0
zAo3~n#6YpjYWc}IB<quWQ+;R=q-Si&!I%vOW)UkaXMPP}|DRiICHCb=#o1a-&d|N4
zm01V!z2$oca)0<#_EVxyWljfPqe?>!VdX=jR99L%%y%>ANrHjUNDTt4K%gHsVuMiL
zG|PbsAWFc#%71j&K2_DzNTNW3D2?7&X~}3(@$H5S%ztqe-%~37Eh8QO3<#_liiS6A
zRuN8heZN^{2~J(_{SZ|5jKG$VR9-mnOH+t^2PqqvR{)2NV}oZs7ZVC_XeQJ=iErJd
zR_<h)F(bi=$H-GBT7^{->>BDm#Y^_}{pa0R<{p+r?D9Bt-;==W+>Aqtc`6tZ&18u~
zF?TsiDIQ}o4%noAT$xYe^Z993^=3NKe=L;gbQv+#$azEKS%acnr>nNy4^?;kEm>Bz
z1Ub7!Vj0)b)*_gHl)ss)6LI|XFS`X;dUrX$<qY3M2Ww~>Bl3EA+r|D(3WU-r&N>+v
zW|mleS&GY?IQIfc-@o7_S;^I|+z~xDZKck0hlIutH9TG99PCF8AfsW(Y%0n8&ozdz
zaZ=}vTc}cS?$YfzGOCRy-NRSKhBKYGKGhP}yR?o}e<3S0J)=Eice(l6q_y7OBF#*m
zfy@QJohNVS0%*#|fc2LoMMXJ-@$<q$v8kwyzRiW~X!46eRkyusXV*b-)W_F+Jr(V@
zKn(Isg;`ne-En&yaLZ(5WHQKgc;q$>`NXvyN2`ClF`>smG;XzzqK&LV`3olsqV+48
zDVos<!A$EJI5#K>?=P~PuCEDzMkT)_;2G>e7p14txPU0vvZI{f1KK^X#@=Id<=X-V
zk_8~3=OnK#PlR0PM;y!TJ0a<fYb?hqMnVBD0!tK8CoU&z<GrDrhl`T5%QeI&H;2EK
zwPj!o`*{DV|B(QHAsHT;zRYr;&Zbk>U2@&#TluA~X&I-TgFU&+qB%b&+ZFfUkC7?<
zh{HqYD&h#tdH>u=*z4gmy1PD+y4hEA0ctx^jtHwNgUz*P3}QiAf<9?hI?P4hhnwWk
z5tsWEh@}1>`K7bn+Go`0Y{D6W*u!sx?ZY)J<IjHdAPfL68qK@1nU$9#Pj5dx|CqP>
z5}WIR*e{^zSo!YG=_H}u9zh)7!L&cr)uDhE<dO6rI?Ht2h=M8b;y(0fRxO=9fKWwa
zw_JHX+(OVspKzIT>}FGQQFJ0*I$aQLkw<*lP&T1sOaSNvN}zRM^A^~$NJ)8G{JE1%
zx5AX&Cb4#f6et{!^{+uVdb@Vc3BF<>g)SRd;6+WDy}VNjRkY6z5J^A;%@(JH7gpY<
zSfEOeiL*T!Kdew;=)`+wd*p+Np>A6#yl=Iq5rjU00hfn6+?+6aH!#Nay<#t#OKc8W
zb8q!6V7UZOFLr|111VnH7Q?Yd7pVXhywMmg85ph+wBGnfaYqQIN=#`cJ)IN#c_bvk
zL%`MgI<9UBpl4N|i4g`Wk+~*o{ftkD%p?M5f25M#I*omnO6L25kzzvP@0m(z%Fv;#
zIooa`XYk7vv0V-gUX+n3X8|4-{W!M<>p?;1wyX5j2E@7Cgb5fSk}|(drvhB;fTugN
zt?TiUF5TzT=^g$5&+@LPSphMYXj~bQt@H~#%&n-A1edrr34HdL`@a?sp6A>ZCcEw-
zjr7?j7;+S!yh6c&Q8C6Fok@I9h+vy!19)5f^1<!i;CxawX6_xi+=y9T_o`^b=BZxl
zO9wfvSOxi@Dp3^y?N+JlRxVPjEF&fp{)4Yl?zvUea-mGOQJ+UO<l9|4_%1*=pABFJ
zH0Qv94N#1W^#C7#<>Kv~NG@RS>+}+WXQI+Iuyl_Jo*j!zNN^H=YKbe5j#_t_?*>H!
zZY8r%%elt&o+l^hG;?cww3W`fdk<u7^l>oP*{(F)bcerbdz|_b8o$*oqo}J%pJT2)
z#P?yyejKM!Ko<cpyCAdQ$k4eUzcvn+CuET}-f}S)-F%2yPo}x{J%R4ZdoDHDcB0zZ
zD_As^VfF!)UWFRsgivpZ4XFULE6aY46IluRxVyB}tSy3@H~{~aF!>HO+);o_;>`2L
zGd?OxQ7VBS8450*Hsjh{_+IaZhDv5<HPFpG4&n<JWx|phE3~ZMLHtUfM$|DQHjbVh
z$a5>(O9c`eez?BB3ZR$zYx<ZX*r-ACqTJOqm`;9O-MY&|v5@t*$E*84<|I-VA1+@1
zij@^>{tdqWdn<ghvKesQ{y2BNmrHWmeRtLQg+kS$K5A1dpRtUnt%{l)<caU!+PS__
z9dEL$^;i|X@3)*M`Z~Sz;-RMgChb$3%9MM&Wf53f2`$ju)1|Dc;<bZR2gte*FE%vl
zv9OQZyc@1+I(4;eU<}`MK75HvW!c18@j3LcjHx+4Kc|)9#txqp)C+$%hE3jo3L|>U
zzbk_Vrd;<>^?DEZI67-=#{iBBC?q3*EN^T+WS1;1U-U!<$8PHC*xx8lZl)j)N-ktq
zttvb*_NJ1@?1TY5J}kC?a5R!`;mLP%o%HYZEokQnNk3erN`IpHT4AS6rSAOpXue;#
z3oHQN7;s-JY?7b>`zXWZ#g50cM{L5rL=)!{T@jWDo9+ULj-mb}+b`kwnG<u@K1ahJ
zqv96|5%KQj_G_JD9gSL{?^_g@#U-Ae4oTgrb-a;dJ^Llc_Zw5TU$L+L{o@n$(8YgZ
z7gK<4ygmcMXf}E^FT!!Pmuf>#!t*A36KQ|o-~^mR1O#UNy>A%j)~Q7896A<5w5Zjw
z>+6ma`5ZQSBUb;LdvPEdw%T}8ZC@R%d#_LNeW{lACL<HUW>e?y<>vq|V$g8l;nyf(
zs~nCo{Zj~0dhow7M*hCe5-L>eRCOS+Hf1DG^j13+*CVjh{9Y&Y<iC>}A&v8?9bJ3@
z9_qv8A4<@7UCII;U0=~B05r1|1$0|aF7`FWJx|c+1MXcjE5IL=?Qw%fhxqfG)b4Zy
zmtmGOumrzuM^@5G%XxvronKbJL{*?#IOmG@lLst)oMnpMZ@sxRG=GbIF4)#I+yxe{
zI_(>lwC6M%KB@43byZc74wJ@pG~TyMk$#cDk6scSSa;PAWfLwI3W7i5w)&EK++0Ss
zA=07OFU{hu>qiy?(GoIIYDFbslif7CCLtJdhX5-glI<1jc%?Nu>+z!SLQ;<Qjv44V
zuFrQz#H$)ZHPSKf*MJsFl}VZVd_!#g_cz`BAHU2&x=YCXTTfXIS6O0)>Wb{4w>~ka
zLor2nCF@<wv@BKA&-4V?C7wzsK=TA$IRvlu<eQ<nksEnxL|LLrS_CR?0d^Y+ci}-R
z&(E4jWl(B1V!oMLQ-T!c>J?wb51|5uO+xInC-c-#{@gwP*PH)Wrvld7>jgFc;8&l1
zuepfl9Ee@FJt0g>#)yM~%3ju8*%hz{w2xd_z|pym)aE&fpIPF2R-JSx{a6erUC+ZN
zii)^hlP*X|JUzX(R$^{rgZ9-izux+BB5-fw|A{;P`)r%0*Vm=Iu@B7Hq=}w!hv-V$
zP78OW?iWS_z7FIW&Z>M{Y;);JRH3zWh(z<<yc3I9Kuhsdm!j@eA0PqvrW9X3a#f%Q
z)^JSw>4eGRF5nO_*anh-EzE^_if$e_<-Rhz2z~!Ao_u-D&qlh=%V8tiJlmXum99zu
zLI8rwUCOwqP$AogAYtx=Cr%(7DQ4k}jLl`wZMvd$3W_)`N3E6OyD2gqvBc=VGy*e<
z*SwnsP?<6?{Ydsgk6>SEnx{VOd_M<hzC`!t;4(CE<I1!d-km!@r7G6<{!f_I?2err
zrI>b!rVs>s|D)WMPoS!wqnvRWQ`RSBQrY&9H+I)rxdvauU{rDpl=mN5|0_mUDr;gR
zwMXUl4IVHY+wN#Ia>}hjI@S0|jVG1Lqr<11V8u(Xo>FbIJrQ_K%Et?=L06Aw4ETr3
zLDAKJved^Y7v!)?KbC6$@yed>@h!+C!t-5-sH-3JJ$v7aL4F=}(wlSOs*FdAX&(Xq
zaF$kJ|5?mc5(i6<p>+tCXwV>f-K7J_oO)FJFZc=lIOtN+EV#&Ea3b*o(N2R>IE~cS
z-PZjj==#Cn&Jz)zy_@SEXLa$DmSWO9F}<>2`@q(E^Dd7h!ZkGCT~xA7!2wNN=SX2Z
zv8D}mog0FCL0594XB7~hIR4r7+$YtHCq?D|yh>)y!4IYX&>nQhz7R5>w-HJRwel>J
zJkbWa4gUm3rhb>$s*%?}S0A1O4ilR`(cQ8h#ewma@ist^{H39U0F(U5_OaV=-Q}<#
zgC6Z&JOS}g)q&i*@CHh;6&6K%(0Pq}W63FePRn8OR&;<RuFn#6)Fjb;HR#*7wD?LO
zYf&Yix8I;dFtDS8CkR3gERe<1dV!zw7NKil^7ENe+2Z?lEVA9O6b$6uo-1O>Yd#k~
zvu4ctiHSxQrI)JlIl~f(0_8o|W8G2>^*R@kyS5`O3siyI(WWgSM{s7<sXi(0h(zPU
z4za2naEIljSA_d2J*U9nnF7A3wYtCr<X<_-T&GQUk-|QnRo>NHsGcS*Ei_YX$YdjF
zn)I65_)!TTN<*@3wB;F-#wL6wa=A!t>E<Ndrc3@ri4xW(kXcB(YL%7t&b#d9(^7q}
zO%2dED7X!__&X%3c&!9Fp%oP0zU>#S*;jDr77sX<0I041w>>Wp=0sjebOWYF^Ey2y
zLb?cP|Id=zLj9cR<`)gb_)oozZohiB_*P6SvOb2AYOs{41gQLASl6DSgkc1)G$YrB
z{0g8CXJP`h`poeGn-Z2zUbL7M=>@vTS*4NR03dYdE2v@b<HyG4zcP>>P7=vQ3+uBZ
zck0I3fr{0^G@~#o!3?M$5`L2kTi;G(rfd)L<mKn!=sDQI4E_eyQ|FbSsC?S@F;O%>
z&`}Gg9Gmf<n|A~@D7a$2B_84EmrTKB%VlGv+oJSc14E78E9q@r!wN8*HM%xeA(j&^
zh^upm)6$NP`_8Ynx@kQhQz&A5Go^j*@FzL@wO`8sEA#G^@=lRq%-UYl(3SH1f#@$V
zH6|#YI?9i{kNG)lWcP@RiU6dk^Ev*2TV=-%NIeLli^vFYy4PsD-bsIda<iLd2@GF8
z`R{&OD-DVX0>!P`$`4|(p%j(&IBhv6hdzHCr%FL)v$P+WC2sgzqKQ-+alD33sXjO!
zJwWtke$T*-<e`II6_Zz(*+q4IGEeAb`i{RGPaP+3-kJ!~8@_g<i{ueYc?-!@3?f&&
z7yios%GG03A~#@)Am3?$)~ipFo=81Li|yfksinfO{Cf)+Hyqk4Gs1ywFQD;Xay{X<
z8Mt!n)Ts>~U52rfg(~Z-NKG1$SfR;GqffeyCGnfrZ{CPIkaMb+oI1u%RBm4ld>UG{
zGOfFuXyh|}dN|_#MTB2isEN3Gu)cl>rZ?lFDnRU%SOyl3sX*S0l2t6Wyg#oZb&aj0
z5@kB{6sKB@oqh}35gO29b#%Z&JfnDX^EJ!7*l3pMVr6r#{SgABQpN(_4RP-aTp$a8
z(eneU16PI~LCM*tQ5rop;h$feF&6{@(oF{HX50l)dE`kp7%%sqA&zH*-#TUO-NE1r
zlTyZK-^0HfpzVqtT%FF>y%$8KLMzcKA0%^rIz)5iuMuMLb(#eZK<%TCFWA#|DH(1m
zlBZ#l0~uDT!%sR0DN56>S)NW8utu?f<p=fn$tGgInAVvpN~#Py8L*rS+0Th~6skKO
za~>=i5F8+l1Tkp8TXFard-Ogy?0zEpwhhr~uhZg*B9vku5!CShGCAv11_VX1(L$wb
z{}e^Fhf4*tac4_Q@%jnzpD9rAv_@?kIj~84C6iRN54%peBpgqQm{{tZJwW7pPb_oR
z9|A$XCg;F=c=@@cMWF0;lPM?1-rf{V?7+v#U}Pwn)R^DO^76lt-Sw~UKoey2x$oHk
zJD~bkVs$O@PybI)=GF{yFN>?R_k6wkT#S|&J5$kpnYK+)hL%~|>F_Q-_ou@_PMFw3
z+<Nx`3$=R;weoVjk2(Pwa+Hrpxdl0uxK#g@&o;(#<cjR5;E4uvvY@A|Kbfi@-{Lo%
zz%FtpgqL>A{4_`bIykM7YK};Q6s7&VaPOfdeDekSEifGGh>v@*^M+N^JW_Vhgf0r@
zgoOZYTwIlKK4G*rcp`Q;eR595v(3XQ2qgij)2bP(l-41)N)sH2SMH6G`dda0!-m~P
zD#LYmt2R7|qN8M%oJ|L=)3+&Dikokt24(im8s2j~xonmFuCSoE2Pj}>?NlOXvTxhC
zqA(GP49eRE>>CIK!ndckG|_C>6IpbqyV+#b($nF`NU3DB?#CF#GX@C_{AbGpc235!
zrm9W>-H05VY%EKvkeQ}5h&ue83w$7n96bd>i+J&w#U@ekr3rzJn71*Ij8G0En)Pix
z5R(&V4`3l=8D^o_tV<Bdp>Cxm`O6mF^2<;ECGGIAf+6(|7FXHKAmrXeri_ev#_Kbg
z<oIi6$EQh=k(3yYlTlyArRinFY`eR}*z|}|tUZzMk~WnHtcfmV|EiIR+KZSqJJa?(
zwe;YKHF%xM;0gWJkEUimtavQI^95Rn-qHDb(t{o<CXk6DR#A%}El!F9TV5D}2nc8M
zCXpZbu}BEA?H66|JN82GbB_s8R-Sfvt@TlYU8i+4{8e&Bh!)7y%rj+^l*u%<Eyzw8
zwbIegGl|5GI5V)erwX1%drDvah|>>HuH&>W7ONHov0}I|d9a&vV$%R~F017Z8+FFC
z9Ba#^(G76uB(JSEd8&<iQLFP2N{HyU*o)`|mLN}j=B2QD^3qazlHDg|PC(&pgQ;-F
zAYK+vl5RAq3zt!gb}1kO8h=z@-1{%UbY~QTpXgCkEtyz0>1A7MdRb|Dv6#jyU38<(
zihyMMqG&@36Ge(E4z1KK*s|j6v7*+!SkFJfwO(*lgfL2Ref1NpEA3H?n6{q_q{O%o
zn@fFCmXfuWMP;`23XYC9;2#L15jp56$F^1docE@n9sK;S1^l%j*f2ic{^jL=5(KGa
zfv>M6eI<El<`uQ|pV@*#<DW0Nb)i65l|LbA>u_j(_p`;+DCsGg>SXhdx-7tZ>PkI<
z*?3!HPGOAIp_G^KdsoaI3|4?SzAMficcDa2diIkh-IH4MDuR!}id|@?b7QxdC&_wv
zSUZ!9Ij2z8)(Ky4jqtq^S6T-;>o?X=M;CGLYY<t^YfFmFXfjDOFp^A3!}!$@Au%gO
z<H)oY!iuG9Xw1msjdB@z=5iKWsIRfY`gEjcOr$ccrJm9h+?TyubM}oT9=SA|V4#S#
zPMU7*0FE?qkS%u)2!U63MTdh9=#;XuC|}Q@Ea7``Nqh5^yu%oyLSYD9<sTWly_MEX
zeZ#V&Z*Gkg8hf{n^8`Nj!<C~l>%WOv4RaY=P6(+0o%oO3F%o3T{_R(jwhXBMa2V(n
zaTS9Qvr#CEk<|B=lyrcH%JbOO-94n<zS8K!NrZ%nWI*tPbDWQ@EwI{H$do0ln}%j!
zInLB|j1!7wxEqNs0wbTwbvRy8@I=wWCL$s`Se@gyrfY?UGoXuOjPkY6l&$0V=(1Qi
zZWOhKS$rK8+4g#{#>+V$nGLE<A%UktPn2V{H_CkF1wO8*L<aeyB)fV7$LYy#7J6K^
z9tZup3cXXUwH)BJnL-+w=2WGnuL)N%&+RipC@A<bI!TMUJUjE7k`eJ6B^sP)1TWUV
z+Nfz;mkp)!V`1tCN2tB4wGg13l>22kn=*9#U2**EJ-jQ_X7Bj~(9?(tq+ojc;^|Bx
z<{7%taw#rW{9+hwicovJ4L{Z2=maw8Ep1{gHiD&g-*@ZPhwlr$mwsszIv7NPsw`Cj
z6q*y8-;Jjx+ie`ZWmVYg1t|tHNW3bn8p(y5e)U9#ScbPJy-36R;}>Txf&sSpKXU@K
z1!?2@Hbm>w5y?~AlBw$_pF~aR3bli1`Mrz74QG=I$bSpVBcs7uRjDXBxWerR008he
zDUS|q|AVPw!Eri>|5^>7;BXh8af>qKr$=<*sxMI}x@{plz-5h8Ru$Of$XSTQhHBCY
zxKml0msS>lqEX7n2lFzxd|-?fgh~fUB(;TvOp!uya^I>=(%z-vh$6}zZn`KqSVpzQ
z3SqkWov^N=3FeT-!|Ch^CRXwuh86p%$a1+x&UN}^-2|x|z!V9~r9UpNRMn@ClY%0d
z;`-*r)+a*?*uo;zGmZ2G2o!r#-YPdmN;m045|EZ+i6GbS{HcTQ4IXC5n`wpHWu0we
z9INS_K2KWCW$hZo6z{nXv~97iA6Nh2;bY}7xJ&SxUNKEivFWe#T1B<6ge3fNGR}WJ
zmTS|6wpa8|?PiAt6ln~U!}EZrBbXgA2*foeo?Tb`p|5l5FA%>`quSeQkLX5cb9w|u
z5>ez&Z<AcuHk>CX4y~bJ8yPh131HgE8A@YTmap?iLEw7vcK_p)XLg#lD*V=k!Kvy%
z%=h6Ea7}Ie<LTcD!%B6A@|4h^OF1fpRrRZ^Z+lXQCx$(@=!9`b^%(QshOuy7j34Lf
zl<X&%tRl@g`?PakFTQ`#myPX#r`z<YHa`A1D(H-WjD2h_m1IhHHt_1suly8+d_46W
z{&wXGdQe=A8)L5{4E-#q1%pJD-p3@FaC%jLH6|xIx+r^iiHhLt%wmqM^RcFf$fe72
zEI7+L(0!NVqf5!0!b*du@f9r1qDhlYmaxu?=4}i0)m{L84&{C0ZXe>e<aBPdZl!Jl
z{Oh}S^R9#p0z^DddBf<GtRU}_L^`zDJ^c?i;}=t}K7^k83+r51TMSkXM8z_FdtW2^
z8W;WUp)FEgoG$K-I+j4^d33Z9e%9SKhv4reQ2@9aFYw9JkEj#1F;f+fypBmIqIn9F
z#3vGN&Ag?d@VjZYY5~T|^Ue(b+`PNH+xTgh+*m+|DE?#?$#}(j<!@e1sQH{-fIaBb
zy)M<ZwRplK7bn-RhivL^1+Q-PZ`3A5{DsrwAybJyuVZE72CCU2_mTb5m{CLbpZ*Dw
ze&GY>eILNxulO9d!UY_E*Ud70_7i0B_0_z4t5lGPt0-L)@|LIGBNX`B*0NMmw7E5=
zN!3W)Zeuxd(5QE=gXXb1EbM^d4^>80m|NsI9-MD<|7hO%y)E%hXIrhn8(j9S+ZyVB
z0W2~hebt;K0F{Rntgap9d+m;k4yefHkFP3|OJrQatkIfO$)3NGw90w;Qs5kAlZE@<
z!^TEDb82czl|x@7(S7HN_vulds^97r=Tqxav(dJS`uxGpdkNio7asaG*1j2!=N9nB
zA3uWY{@FH~7|ND4^*3v*U03U!+@*oiZ`6PS_Vx!;l#N^io~V1*=R<b<9}IPMyJK0K
z2Nmt8{e4dVlMs3!P<!G^gjSS#vQs2w$c_5Is)gf*+l*8r9sv$_5Tpv`PwQ^%@QY+*
z4N0B|JcB=Lio2xSEvV}fUDUMXx*zk&qVT?K&g0A5&p9;wJyxBN)rV`7Kc_9=(d0ZT
zy4)T>Pw#(iP8AAbKbQPAN7*%=zW53zN&e41V7}$~t|R@4F*7X^Hw<KsX*}0gH#_V+
zDPG80N@KPgc}_bq(_au}a#SN{BI?h1baMBhEykB!iaDRgkN$62+OL)A+9b#mgy@D;
zpP*+fKOWSXRy50CEhPEZg(t{gwetEoKc=Y;_KI@@9|};*X4}&knVJGo(}iiGRlJxC
zQ&ax!S@18c_V&pL$QByqDzL8NY=uVN*20({5R&=r6Hm$zVMN>UL_K$DU6>V(MUO{w
zKB-CPNg7$TJZkQ1U8JjTiaQUz6E`-sy#c??FM=sp_EoOY%NW};cTyVqD_2!p-Bv(I
z2HeO0G-(6A3nCx79I4TIz6r+6rlPe+tLbf9MMd@(8tkqGX5X$i4UkK3O0os$LD7=<
ziuH{J#q$Dt$6!ddX9C>&(FneHnFz|qEuVeqmNVoZ$bQpFE4*2wXi^o4QQkJD4#Qoe
zar;%n$=;)0R7p<i=6=@WiC$Wlr0-;r%&^r}O&#Pk!|!k09v##NRr%>oVEC+X`GXg_
zEma$eqv^e<nlk{I;<;2OLIN?~BGHu$4>PiY$)rmr3=9+dS+h1_XFse&-1){6b%}o8
zH_tl53=-N3RE7cWrXU~d$$kehqZuGCauHbr+6gz~0+y=vhQ}Vw;&eiNRcY`MqSC07
z<FiH=x^_7He8qAPajZc4ZSyn|xVnPS(Jbl*5s8233CPkWiSVF%J&;s}O?9Yz=rvbL
zN(x1g?X-yq%_x&6P2a$PlFj&VWU6b-wJ2+*!S%#-&efOcL~Cm_law%yYigN}r>4F^
z|1v24X##qCv4Q{k^|V>|37NWJTSEc$RC_||pYCgCI)cDME_xi2ID4lq<Ynlesm2r*
z)emWmPLe5v@(#R&%wDB+1rlzV275kJCbK)nATwM|Qxj<A%*S7znT|FH_c07e{FG)}
z%A#{1sOSpR@}$%Zs~)R9nT}<gfQ5t%m$hRX%UbbUec@TNAXz%oj+VA{u?B_B?{jF@
zd2QH@s`3M*+2!cDH)3cN!Zf#1k)SOwjeOpMD&4*Pa=a@kI5`^gF8$X|-fN;Kt*#Iw
zmQ9-|LXa@lozU-<)^xlZk7jH>od8L?xP_E14H4_1IL6^apNXii<U3(o9KI>T8E9s8
z2G-xCDJcICfO9SN6lMDS-hgppQs85RPhds(ALl*-hgq0cQ7p`+dWVMo`o~J)8J#F`
zOv?SzPzvuUz)lE4LrA3VNo{RyT}A=;HKi*XltmZ8HtKCcezj#X=F^kxuVn8(eV~$6
zh>;~EFmhD5xDtOCJSRV@mcH-sFoKuhW^!RieaP7NV)w;WmXTDOn!w|-+MCJ%O`pD!
zS?iC%q|6+OVrtK%TdiW2JvH%~a4N~>tL37TYaM!A--^Zmy_mO>Y9PB8Wz$%ybsw`2
z9syDan)<8unqWJFn?-4Tc`Uf9&@kTsr;uT=?Lr}&Vj@F-OvRkiAI*>~wzEZ}BJ*TC
zfX*QR+KZ(+-^GasQehvrYS{`QZeAaPS=3{m^)c0{rAc!LpLqXNHM1-mK_&trf^#h8
zK>Fr1XlZ?aK7>uc&StRA@3s5O=+>3+=tZsWF{PNyifs-awAL%n+Y8Qr3K6u9KKg%o
za6`1oJa>HHyyf=LgJP1XT(bUw;v`k%AK6<XaF>C_q9m}yr!n~{6tdX~^Zp9qnLvKw
zGefe4fq$~Z4(x8(E~JFlIBKL8AE7JF4PL6Wzv<8+ZkQT=+ZN(#Sb)}lGx%vZ(5$|d
z;{5cRcwRpZCNs_;hu<?so=jT{Y>n&>-(<{?ioIojh)2u>HOJ*@Q7k)xFFDit!DqpW
zJ0*271_6+P9JZfm)?qdy!}+Zy$*L)pvrs1Ak@tDj%~Vt@aF!NW{SFeVbd_27K@LG`
z8877L|F)YP^t{83`P3b0-+#H`BU%kbo;tY4ABf%Nd3*i`@5q9ne<ph)7UWwfIH3p{
zI8|({Nw(p}R`}P`&2Sv2C!wzA4_3J&*?YJg_vbzwwwLfFx#B*Uz64xc|CdSP{7WGD
z+5G9B*IX%8l)_=s=IbGdk4+bJ(pB2W=Zp+jm3lXhMLHG5$5%dJP+G^PZJ26cd*R!t
zB!K9F6nnukc6_$zQbalrcvM#G?i1zOb$&`u@h*`3M65C67c0O197f*s_d|@N{bX5Z
zWcO?g7+c&KcC2SoxO_6Z$Yj96po3oc*TFs^5AlPh7|;GFQ(U~)NN}?_Mu9?+yLY%%
zWkqu-QhZbt`o=WHJf{Kt8#&Dwl|ZK(gT7Za3@%0{_dVFlyrAx;MN~K?8_?IC?lsxn
z#~&k_Bn+mePNstK-wU?s@_*ta+=J};KihRPH?~BTbdgJ*welllb{XgN1-zL5gs>>z
z-vLAg@XqRF%?bQYscMx);}2moq^P`D9Bi^aV(qkV{<$HX<p-?Onyhz*JZTL0>mOcM
z`g;Va!L6VitM>FJgTgzoSX~9qlVi^XPn*4-+Tn>hKa&uTU6fphq%^YGrxoCJ=*7sk
zg-{1s3Ar6@8=SqeUc(T`LebR9)FdY3x2k<m4)0Opmfktt5lihCxh?qPJlqE_al}!7
z@fDrHwy_HtEb^Z*ZVwNh&ewVLd_1|50`i5MW)^bfyfSJaTDY(Dy7k8@4dcwDoD;Jh
zp7P>(UrE=cfZYdEa^lHhMsowp?8N@Lop01%GHY1I?~VTqwa-LyH4H3U3{RGtL@(Z7
zHr7#&&Mk-zYNUQR{Plu}kb-ys#}2bvr|bAj8+we&{qF)s_JFBIN6Bb#&T)kAg{bYC
z&vBL7;%A5l#^1$7OvPQ%i8caB2BwXR3rgVR1q8DvnK=)O^{+P@7iB3YzkdtWW^^R0
zxjgYFlpcpw><<2AS%D3Rpd#lgOgD#k+vQfTt|#LBlO9~$==p=5LlFFfb>W7zWE}Q^
zw9gv65JF^Y1=LsM+h<9A-}oN%d7hP6@P<`<=3{p)K@sY|+&UnvpyBBQMgIbYjq~iV
zo98m~1x5175!R0pGLd#-o0(y1zlhB40P~)0IapSJ{X~DS1kkgmouP$ZbPMUxpiQDM
z{+p%j*bb-`v9=-1?)nO;iK$mn-j~IY7T|!IU82gt^fB4wH+6z6QDtnD?rg%A@&#P^
zQOG>r-4S2{Q|kQb&3`tKn%Lwu!NeJwD`N!VrTmB-2FGlGjN*(br)Ui&``pLa2^dIu
zhbyhz*F*3N^nvm)U_3q3JbT0|SBb0h$CV3l;J1w!ElqBv0#w&spg-<O!e{Zxmfnm>
zjnt|f9#SG1wae}z4i+#YCME@U>>XVjJwm_Xltx;5Nr;-tEr!*RrTj$fBqLwH2t~>w
z4ukyxPLv+d=>U7y+Qo%nH}sYYRH&y{MCR&rJ1a@r+J5O-C?ZYf?<Ylt%Hy%~+oSu<
z%S0?I26PBh!$?WP>0F@rn>%j=><4hHI*s^<S{tb8GK8|jB%@NuotJqDYTO<QRCyVQ
zwd^2w19~-ax%G7DCFa!1i=%?t_?g&_W%*j^AlGY2{Agu{{s9#hx$K@Y24c6wbWN$T
zRX>g(UFH{%DC5U%9|&WKQWC_1<AK;4fo?725_5O(X(Y`+P*o5;^+4BwdoqIIxI&5P
z9$wo!c)7;%hV9yo3&}Z?r~11BM?fpGBSW9NB6rbOgK1humYY+*OY1$AjSw<b%sMgt
zIG%<?X%HwSxyxpeR!Pf+AE|1`7OFCby=4@oCzFj{q}u<)0IL^W1AEfI`><y9^~RSv
zDjXsU+7W;9_tuPtQCCK9-%uPIAOQ9`PJhX&>=&S_{>K{qAeOpcC#f>1c&XuKKltT<
zeNOwfWJ|!^Q#hgqP646F6O=fWU_L}6q!W5>OV7AM&E(5~dP7(h_L1#BCLZjQ;y;kK
z03NvMl+16nG0pCBt1kXCCr4zE2#jKg1t-tE#4n6d-F%{xMD+b2fDC=~Y3Sb4GFRpP
znk2`%bGLxkKa<Z+Kq_C8URwjhD=7GKVjGp+oBDFqdX?OMS16559oau5)PR27Y$M{{
z&mm0zcjjs~aZ-@+i)B+=8xIS}eiKqqn+q%8>iZs5(W0xI;@i$8|1#~pQyXfLR*F;D
z3VO1(>u^wrO?T5W{lO|eh_;5Zq&8d&fwpZsV#Qj8CQo7ZJ33z_S$qoA#NwMNjA3A@
z5oT^ywFgMwaV)v-JAxH~thwyLQ*^L*rF^f;cw=XbA9n-ZYwsfw@=H%~o0c`wKfq^a
zy;GGz*;+<7P#}sW$nm)U*}>A{B(-PbwU3_--Ihpv+b?lg^B>zpm{pFTbnIv5R8i&e
z59m$ERw$AeP45Z)o(LY(2Jwe!-H=G-z%Mz*;{IvU-1gbay56Hg+P|HbaHg$elBv-&
znwfj}Z7Aa}C+23w^j_@{44B2^`M##&EqL?`=mAg-OGz0+vkoO0gHU`FXONAj9EVg8
zIE-FNAY%zP?%uHG?`$MOsKK?JrfD0rn<oubg;WpSxNM%7rh)XCv)u{5L{)qI`1R=*
zm^)?2(Qx(111P9L;po7$<)4_5iz-Yf8!y;5A{h>uDkHNJm267x6)TavY^9O(t%yZ6
z5&c<iV8MadFtrPqZ=Z?OxhkgapAPu5s@8Pa@X^U`gpsCz)ZyrK9p>!{Q-ar>eNxg(
z^#F1JERL@YVi7Xi|G_se@!D_t)%E?yK$Ft^nU9FnK2UrATkr>_jxPUq(}Zlh>9?fM
zaj+NZx|G@o?G}_VlD9#!5P}4QUc`NLlEhXYl`0*`HC<6m2J9#dhDpvYC8lhDG&xzN
zqS@*uSHW+}LFs+PBW>{Z#7Y&@jEr3lTyi&0smoV$W?}~3o+NM7KWu`y?T6$0HtmuT
zEQ^aY)Uaxj;&5B{qo1Sa#q$pi>B$Ku6c(*y0`EQt!j$E6q6}ithWud~=)zDmm>g6X
z8o0G9iqZctayhn{^yCEnl12MrWf2ksf)rPxYHz+8F-m8)n;^n3L3;d%<JK6L8OLRB
zo@J>!d}|l!n5I)&5IyTe%7eXl0+zC~^YeA))lT<uPaVtpTmPuIMXfGBJH)rlr<<yC
z9wI(2!S&nm<{=|CVWc?f)XxWe_vf)J`O<1K()|lf4rqwaKHA5&w6&pxSA9bc4h}Mw
zhwyTZxVr*Vh136hi2Ud@_Nz*fUMrz9t~e^zHMmZ+soerz>FWEmT|r>Z^^x}S&5b(w
z3~`8>weREu3WJSGVYL5&1Dcswte6q+&u4ER?<rp>&^PeIL8{h7s_8d*d6*PU|Ho5L
z`cVmm+0U+Rqpi;ky@W>;9L*lf*;Ag_7CFA%(tJ%Mzst$T7K+X{I%y759-Nz$hPovo
zQ}4@mbjP9ZMI_WR5r>8)SMHBw55@w#D}-Y8{=Zw@G+Xw+`a-WbckRnY7Gqg{xN+SZ
z3nbILp4;SZ(*>k}@ZJYJ_Qrq}${ztYCILf5yl7}#<Vtd{wVFh0Z^DGd(vv>CHV|cZ
zT1nU}!~WcVHFt}UlcsYhd~gzT9XB9=5#HvR13L46xpyfl8KuHrrQ+qTezHul<o>Kd
zla|Q`p6Lt7U6$28@^j!xm%KX^@cn0_2N3*!;E$JA8uYzUOReiRwWRNT*m8~ASbqoO
ztbsh9KbIWWL5`Bh*{(t7D5;DXNRxD&($!AdWfOBcGPX(|4xzF4$U&sv^iv<ul$;8C
zhKl@03dv3I3899qUNxywj8z<nCCFl<k)hCdws}YJD#(xZ_!yySf`ZZnO2o!XMP}8B
zuT#pVEB_Ui)l3yhi=OJe1**HR1akzVuILh2I}}Ukd0ULI<oQR3lgZo&8JsmRHrQBq
z3@G7;3M!$<wX@}GFD2it2MvP>0CO~s)l?>T)0`W|LAsp0Vzr!bkB6!piQ97eo%*@C
z^T9uwE<2omv5b?Jyi@AZkYvWj3sl!fT6s!XPZ68j?75lnOXukr%f{;eo#I0r3>{BD
zOT-Hc<^8*T<AM$QJr^aDrwCi06!{9)E1PR<X&LU7^BSGZ>Zu&}?A%0n(eg*gmm?lu
zo`zzhm30CQZD9K5d%ItFeSba%nA|?cNj-vqlpS3f(~7y=jxVT=J_SYZ96bK+2(^)I
zAC|TOgvrOAoaEW<MFUtAPY~@P`4fRyEVOEq{BJcb{Hh{rXbhdAH@-#sf2t`(O2hup
zFxu)>)el|qWdI?S265=9BpGb`w!9!E+Co?|iWM~>nEzj)>JnO`UPRu@8F5|&x!2SW
z*S*{H#pW=BmjiOdsu=%ucwRVo6NkTGBIN2!2XBqxqPYb?!ZC)^hcCjX66+6TA7>O9
zS<#?+FoSTm#zgA9sv*}|WU@Hy7%|t#d-JK>{(2S^X_HMOiL`@V|7tpWZvq}e3VsRD
z9CNTRO=ni=)Ucdt5~-HrR<$crzvA942;tvyua+86Pjhi_H^r5N-B0LUe0#!VHrWn{
zlYlL2t{>Qelr%(iuf`%)yDhiC??}~fZK`M8BpcG?SVt@48;p3|iCUaW^Z*kB#Xsr;
z$eXz_@$<dmBU}hF<wO6XH()v0q*pzX7I|Y3BY(O}_*c>Yxh)67XMAP^L5G`+C0yTy
zAN5TQ14#TXOy~iD<pPr#TgPblo403540-dYIQqumPZ-Rt+gAYt)|QyxsD7yQb+=70
zDTY~BXpo*IVrd+GBHoMd>{0u*no!KVoCt;aMp1(D%Mq;iXnM9mj^qz?IX{^Su#6kT
z6GTu^LvVNMqhOH23{}1jQ@$#6JpThvUk+<Slj|%C=}pFc;q!nxf0k-knM~X}|G8%-
z4)4yWN8%l>2Eq2wB9HcEc%t3G=?rkrr_Xw`5^gPL{E}s=rF+R-^IJBe31+m;#%kr2
zHBEUzdn2&Q-e+KmJoAF83lgT`(Gf>5O@3Z`9`xb<ReJnl{Ro*wbT0<*7wKrhWS@d9
z`inqX+2}t#{zqYpYBqO5<1R&h^vn=<VgRcHXy$*@_lqgD_13=b2$EBvgKXYpWYNC*
zv@g`%eRSKhmX(<QFt4(su~$x?{i`y1Lb^#|lZ4<S@eAP)P9rOVmMn`<u3icYa&^ld
zQl=X(Pi$W~L3E*}=Yp{_C&{jFb{dGf0|90dXo~@Y3b=wII01#`E0VCR85Y>zR!FL`
zca=m9`yL4HAbrA+L={K#bsXar0X=|;xEDdx^!#UOq@mqwO8o32iee$+ZWawn^KP`>
zRof2+<bZkeeB!vG73NIX(5WOWfm+%r+I<(I9WyR)T6D>J{!%al_Tcip(~1B0puvg6
z@HLD1#N$_Cu-!u&l7YchnLPqn3AUM7+Tqz9*)&J7L)*OJzT5=5uG1&Y<=w!h(ZIk!
zUOaO=Ilk<mM#GPtbuD!;I3!?l2diW;GhQ<J`}dc=cSjBX8YTeH1K2xH_a>Nudvyo%
zE{|TNJWI&yEJJW&y1=X{^Bzt`zU@a&+|ou)FaMgcD>Z%><-TJ#jgiS_#$ZT$nU>Qd
z1IOh{y8&+0>p)U*Ra`!ur|i#V|KuwzfHY|fs4a(glq5C2Qe<&fkFvbBG6<6^XMH%2
z>}0~B$+ZD#Wamku*~$coCwi#WT8AWAA#dTU3YluXjQR-N7)EWX0i1oKlI4S`P(s5u
zMDaq!AGgUG7mquIEFO=b!wM9IG*jzi$Vxu)v74F;3>Eo~GSowU9@lR1YJQ)wWlP=1
z_+h#3&H92gMJLQT+z1>0#I$}}mhM!9uK*c7<5Lb_ijahKU1e(sj&ZYEtf4SB5o}#n
zfX=JJXb73yiS-<>qB&gfD-gQP^;G374$Z2(;3yGG5N)&*lmI6RnbhN)9i!<{-)LER
zMI&$EgKz>`0x9kHtp7j5?oa!!PvZhI)WTca+D0GTlo+EhQK_^p46Ff1HRc;!eFGvn
zb9HRB!(i>7gxn5nsidC!)0tal7{Ttwm_iJX6d3l@l3@C1Zx%ZzH3uRq;9-YsGLOsR
za>~pEg$&}-!J}0P;w5YTeg@?YZ1aP713~+*-YD|742TzAl@4PMeZaKxU>?f7@qf!d
zgVn<JZ3jEL)-}VmGMvigM0Qb%2vDML?wOUpsT&B37g8`QxmDU%FqyMK(vX!xppmQ=
zb7txxZ(?4b%-7X+rN902>Djkw_vGWm9cBM;bB@)*v}#}ZKbLH%p{tR!3^>@zP<O0)
zi=%yRHjmAYPhjFeP*|SRTzP)*R7EN`w9IG>iN<^^U<Jcb{mk;2tn<#%pPE@3eU}Sq
z6ZG-ys5Dun2bWMbt6Q7MHIt>qt8q38b2fhAExg0-lR$947M85$tVt;z>GHgAGvJxJ
z7@)W|?}+oiU<ND(eR@1pk=c7i-JRWHn6}TZ*Vbt;F{4xeG(yf4@m3;Of5u~tg3fe7
z-2=Zk=WxRBdtwt)hzMI4cN@oilJGdw1~XtmTbP#64uy`c{A{%374*D4qx2W15n47V
zn&>x6)h!uw{5|n22ayYXe*=8xW|!riOQp=h^EmYDeHRYB5n9`<U+({A)eaOKUY7eQ
z*Rv!Fgh-OapnUBtIJ_v1DVFjRRY%t*?Jk}ZLB$m{b3|zQW0Z7ghcV4-UyX6&<?DTt
zkWpW0{53Z|+?!;oVi49bD}n_L?*A)zLv6hhPt16M+yA?h0qv{8BtGs7lgrX6^%4=T
zQ~bB(VDX-oxhxD*uPik~m=m^Zp}mIG2e`qb(ew{8GL|9_bwQ(AZtV7nR-s|bvrT^P
zIex$SjJj16pT6PM>=5~emv3pO*GN{DweRV5t9GZ%=cQBC5UDGJx;{kT^UWA+!K;HZ
zoV(bpbBswtaqNii|J{~rI%OHrl6K-UOg5WTo`@0|J~a`h(~ah7#a_^m105Tgx8xUj
z(pVVarV7lDzsb(!8Y5*O1BH{J2Os^xAImgeKXWs5R{sxICw(SkCYco;$=|=ur!ze0
z|C9~$UAS>(@f9jziRAL}S3kEkI1}(^ZUC?M`6Ca{yFc8(s#+%ID*6-)C<4<$+6ERJ
zeOWT7lQEZ5Zv~1lF#i!7hIQxU8R2jS-WlV11|<(nI!Ff4B@&7Q?;?iaI#3ARuaQpI
zn6zyes~kmM;+EC}+K2Bv0p*@VV2pR9tl3w|vxBi3@EIdbo~}}uz)KuPEX967II*AP
zhCTe=qBZ!<1<UquHP(NgqARwz8z6*@W<<u@mHme|uep&)VU=j{43+$<#*dO#n&k%$
znSDOX71p>lgXfgj#&tg(NpZNNv#l5D?JF74gv@NAd^J`CoecpmUPWQ6s*rk4eZ6bg
zH9&rhu3AFvpw$Ls?yghoQAka<MI@8@A5v2(o2=c`PS3VH{Jl9&{vHyNHhka8+9?C1
zhE!j8{uFl-506XMUc8kubSgW&o<~Teyf-YT%uLc~qICx003Y8#`qB$iXJIKz%k%uV
zznX@l(2_X|PVwm?*0rf=o4=PysIZvzPEf)nv>}hc66NXRgAc$zufGqtdj`a4#A|e|
zSB_PY)vX;mV4%*c{u9l5U43l$`R%2!LwKLhuts7ti0m6Vr}hJNS;KMLgsR~~z{}ax
ziL&ERbt88ht1~@ASnuX|tNIj>dJh`EGzM$QrXKfC44v`Uq%41WDNn6Fs!{z+v*y_5
z%EG@E*M00*PTr2oo=zqDZ&j!IwF3N&x+KnDZS!~U|KpR<zq=|-xX~YrZzVk`F6lVM
zTcL70jEH_+9oBVRSG$xi>YgPYp?u5BMYUNlv!iLQq$t70*K)G6+p;}g@~ltw+l&x_
zS-R~Id9AEF&b=y3;(kYuT#WnH(tK#s5%2vNU!i@MO^y8|YutC#m~UG@R@BY@(TBeW
za}_FoG~n)k3S)){vb<CS@sfe-`?y5l<|D~E)8-|KVrgtM@*){5n)^<-DY9Oe<==I5
z#J%!(g86TrW;<5S7wbLlHdDa|9c(#cAkgIcfMu1UEvXZP7KG2B<WQHFgl|dM3nDE$
zW_MQ>U}tp+R#eg(EEj6eRJpKls&=!wN3}l<HoyeStDZIb%&)8qy3vL7R(wxXZ`Ef#
zZ&dg~{DMWcKGZ^&6DF=}q7wVQp3+e5C4^VCiS4VF)17F&IF~7AyUO)-n9-z|z8Of?
z&>|x)$hOo3tqx^=G9Bpmy)2pGG<b4npSQ!pX5feyFDpMxob}M~7|gtTb3LC>XGsHh
zau^--BBbYLT=HHKP&^oe)p6guHFL4Fho7V*QT!fz&-a&W)1UL}bSm>0V}59J^Q)es
z2&}6+iIw(%xJ&PtZgbsRs~=`NU7Pzqv<G~Dw6=2pmlrc}`U3@WajhyQE?Vq#Dkx^U
zp9(bxDC&mp?rzKi`>r@C-#DPvUVK7}1MgPHShxR>84QXZm*!7NP0BKpVV&<(pyXaC
zNWI}bEQ%TVy7^Iju3BD65G4caxeE-7HC20m>BEbM#$fs;Fnq&i%<KNnuWPW^yOl>b
zf}Z%vhvKZT$}$zIHSqZtkVSF-c#e_@L0g{sE(Ka~7&bv1jTDE>>R8xSUR+cCjvs!T
z!S(f9aSE&GO)xCD6PDS=`{%Tw+O;t&x!&1{*IMXC^635JPh{Nj9PHsSja8&qH6`?Z
zQsmxNB{*65^W5!GD=Yfy?a0yDCvF(xuL+a$-vv#<w|}XxrpP}KmcCmOZ&5$P56kK<
zIR^5ZUkV$*s~Q)0cz6aM)eqEuKI6jjrTyZH*7Hlu@s-h#D-GuJPcNQOiT>Wd>Jo_s
z{qeb9?OQzo9$#SSbN_H0&<F6(z=?gm>?o9P^fnTm@wu+MS<8Ie%;`%@0`s?GiuU_!
zmC5)RvESvcFEa{?lTNSjV@I0v)h~*|NDPBbZW36Kd{hoGtLLSxCf%F-)@#3i6E`$8
zpamYyu<E4O;itD`8c%N}+8yWC!ySB|pZ_lk9QETN=Eo@klIklfJRavZq`2L9ZcCO9
z_vgoQ3VwHOx0bZ!Nm!8BGykQ*d4%%9oyxfK4ALW)&2xUn2U}Zh`OlUJj>pKFxfj4s
z)e5*jCU%b_FZh@6<;TloH2km$ZqNA7tSxV@bNj8>jUNlUPdw9V;lY@g|IQ1AR-Dbh
z(_-kx*tm9>&7}^Q%aNa(`v*{Jasenc**2REo<DzH#=O~V&~+Ui9v*OUu|67$S6pUV
z)gQx`#*C!7o->|jR~y$#T(8#!ALBS;x7*>vFTY@Se}{km^B;Kg<~zOHWgOA>dLbSJ
z7}JDdIA~uXpvwnJ=Q1@$Um4%#?GV8fFc+&d#vd)?3NYcMVT-7VNe(Y91wcgIbK^MQ
zDbtDxdLg+A^SLZL)amzFVXP2DN_=64>y=agx%Fu@*7rT`@9**H(<cnWfSa2eJbU&m
zS6Luh7nEwzF{;&L$tTPgOlWesDbE99>`TU>7e0Q!gb?QJOqz$s595OA@0(OO;03V`
zy{6Cc9WR`g+G=IX6(q>);}wxAhujlm3mHtLzJ8hy^tTwM7X;j<*MNSnONjg9KAEAn
zGV1I-?*igYVDGou^@7W5YKC8*oWRd4eeTE7nyY8D+!wdy+w=6{1)>-DK%K$2B&=HQ
zMy8*ZKxRLCN(m`nL%rE-G<`Z(DrhMyo6QEpus^~*&yD900>CoZ-|Xt~Z)I3`zFgPM
z==c8q9v?q`1R!|(!}oZ3bA!G!H&i~8SH|NKKSG?#(EugRlcqXb(8gcTAc8cuGYaRi
zc3{$Bdgkf}dM-*3B$N_h>_TQG#RLu03cE=yImFhZ9Q+Ic9W<eV_mNk+IGd0&M6Deg
z&zCP>iq*Q`?{WC;H%;%RF?Qx=vwm`{%nM>(-<Fc9F4cs=@h6|rX`rE|R1!68j+|-h
zfx{r*q{4w8C^Ld3%fy3cW-PHiVQ^bs+Md#Kd2!hK;Cpnq_Ec+(e)D@xix=E$;Q)?^
z9+=ZxTNZ`y*Z1#*UCVM~e0a^vaA9+|_c?9ZGx4VH8}$ir?$6_XX>9hKo#OZ3Em@^L
zmgR*ATc5TD9AEMQ<`(YxZ(CO`dtxpp?yD;Q0^s7}0xw>?!1eVt34*#Fnf>O=+ge(s
zAEPqlw)MuJcln%($b6+WKWAh9`0*pAX~NsLZ*jHV;A*o*nkEdxKD&9+Tz@fJC0yDk
z2!JW6#UW~ebWv-jO=*HmX)c$91mt3v>Oo5R3{;+<;oeTEgix~Am}l_5E=q5jFy=TD
z)1R;L$4bbQGX(7-JiGAD(^N_c5`e=v=xv|wFlfx5ah!$e*yzW`|M~Oha*x(DP5Auz
zGZF+{-=kv-DXK+Ts)46D`M~;Tmn-iRre{@()b(c1vlVlskOfUV8OPyB5d-0ue|x%q
z3ViiBADf{4USg8`4lhPaTW)E<o&j>7t<A@6wWL3Qit%{szzaP8o!5X_ka%Axf5sc%
z`(ZtJ3&7^*D;c+#bIV%rKdo3<2%-3(fv0ILPtnr-_pQ%7es~_95qsfazgc<l;N2g*
zPF?`5sOF&W*E1CxySK)>uEUFb2IlhevTX8Jd}*FT{h6opdOSdkkFUASeVPP%lWDy+
zmS+9yufJly-($PoVzpXf8b{pjcZlIw5u98Gp0*be5Rb+>HSq|9n3DqRpd&2tAY?o+
zP7-3*Bgk<n9m74-bBrBAq6|9VXdRCf$Z6hBR<rJKIN;*q;>f#DesH|+On6zAqWAvr
zSYnL0zP>KWAAR4e5R!y`m1A0l2SLto%fdx8_S90I7+4?`1*z|_I*|@v3oL<jbMaf_
zb9VmC-t$7vKK~|HDi|yY0{hG(#xItaw*I0b&XVIGU`wd+#61J`t{B!n*#+l;1&p=R
zdd6#HOIoOB5<Kn}c!AxTr=P#?(-XaISi=u%jV&B{XbEQqZ_TyF=y9>+&U>u?r9OB)
z{hF{_*wY`6#jmFc=MX{>B*1LR7^B&4w|M^iIi6izA_P<$l@G0Y0b{AqUa+>|I3B@&
zvlTwq2}_kAF}KqF{XIkk-+lKTQqntbF4nzXH4tYvqT#fy@XQCr{kqt`F8Gy%5Oh3x
zP}Gtu>XiO@UkQ-rL!;cc*Ss_0fB=jsHTR6C2@%0MXKR_@I2;aGuh%7(W-v_?#-v^Y
zA)dv2ixo~^jYXuc9XH>}GnI=R_rn2Q->ZPJ@#g9ZRvkn{fQIm4<=vDqrUmU+6$L)e
z<d0Jn`Y#8SYh0Ux=5boDTf#7YgYr5CMwH&1=Rus_L^JS);Lp4t_9@JjnTKbp243*A
z7^WW|TT9G*T3RZA;>%?XVX6HVc){Z}4f}3sjJ_SeMy>w5@O;Yp`RA7y;}R^bX=`3Q
zPLFT@oY#thWoyy@u<sgB#$%Uy`Q!uLoB`L@H+c2x6)rC?k*4}`G`5cYJpIOReSeIL
zhBofTKhtTx8oKKs0Cu|_?(XhzcXx-)W`nnH-(ng^Y_?lmU0vbu@I@Jd2p|ba)9l8|
zLLg=)uY-sX3B?^JqT0HZosqz6Ac!6GlFyLi7t>sRzwD*lJp>r05qc%{Y;{272%lp8
z+ZbK!qwBk)gO4Eu^s$4agu`%vh?HzZz9*>A4~ao<Rahw>w!qC&U}e+j!GxSypjO#p
zI2>wv&>QP0=2Kff);|jr>-X5pqc|@$>PkPzu2i1W1cB{?+0rycYNo$j^N#V9QqxLq
z!B@N8T9$}UQDp|fp8@TSWty6`=`~h<p0!rrj<s(K;5FSPM$Vh|)W-M1mI=$h$8jtl
zlkmW}S`s`h^VZ{($K<UwA8_Y&Y2ny3!c*h*O2$%&q$TTYscnAGTP8fd)_Ck$`<&gM
zDJ8_NvF<5BBqt0=t{NO-DI_@zI{;d1OF6B!Z|=1h&tG7@Sz#D<c@u+vbORH|hLu@i
zI3<MGs+;Lhg^QtIncqmyYZnnf-}PEfKP4auwwn!JzI=&`)fz)kwj2Sh`xPV&(4q?g
z5(1_%3n7u5p{8>Qh-H!@qKwUslIE(*VvH#St?!5lC}&EufXc2i%YIWKCYecg<;GAg
zDWmNRaxy^(vl~nVDQLw4zu!9PO7>O&%B`hc)VS2cZU;#T+sjMDzQZt%b47s=bY4LV
zB+s;C^e$w*<?}QdPgY>nF3mGuWczBf*;KK}F->#jfgIlffSxaK6HTn(3mSQ^us4JM
zHeB>j8kQ``5cE7)3|dcbKGZvEHH8pLg@q`BC;~)N1~iplAP_1_TFU)A0$*qOkfi0{
zD}!x(Z^`(em}g*|!>?UQ^W7E$G6w47ET^<%uSK)4sHZGdfZhN086zyHd|#{qwS=pm
zn-?x#d*{zvmI@16{y(n^kCpM)8WX?Q0xsF&1qb_wy%1(v;5BIp?ImleC6HP|UD5=p
z;w(VytwJ==<?UatKb8q;(t2tsZ`!WwaCv!!?RJap_5yt$k){#IMICvpU8kFyOT2h|
zw#Mw|Gb;~uiJ6(f)aO}ZjH>Y`!MHyteNPFOn+;a|3hS=Nus>v<zB09O+G89Cq>?0H
zMvzL3E=I&ID|SHzj$jrFCJA$W&LVFGqL;0Oh!oUar}8QwgwBc@WX;~zQWgJ%kUukJ
zK&`Jyf|k`ZI*SRw2t?$#&xDkP&4i?uPYgMAIoBM7CL1IGvFo%zs7RI{x$3xRuc5dI
zZ66AN`D~S00J>)L-cD&ME-OoN;>rhHA-W5-8{CHvD(g@$6soK+J}Th(-OSfQURSC9
zU-n$KjyTg!q9$jgN!MCS{769)07%r`0|4W=KBDawe3kG?Qfr?rKQ?Rryg66`d+>-d
zXleG`_fi0vQyg2C1Gn#!3QP-WY1}Qh7c*LKW^+F+M(&x7#|;x9&!J_#@NZ_?-tuS(
zJ8va<>$AlSd5m83S_1$6l5q3*_H@u{%a5a#ey%CAvoD)9#QD&tDo%yI(o}8c&1QqE
zt1DbzUl%6A>*nXt5+?kA@A|d5_%^-U&YxSe&zYHpuEXyB4qrZh!PV6jR{g5XYaB+T
zDOF{5l8pJ?y5cpU*9<6ASIaNaC1wThwnn66%x#O_b{Rs^uKoIWrDUAcB8^s?EDO0|
zv}{BDxa+#CZBHufv*jAYOxw(VhGD=sj_3kqPlh>GW;_6x_mKWy6%$=Aw&suN(4J%3
z@ktE!yoc6zM?wWal$vkx`~GL%n=077vh3dtIsh$Ogq-%><(qW#a=-*N00|K$e7?yw
z+J+F%G`*G?LO2ReElrl6@&7NIs3F`~!?WMaWW5I8I_2!ONA{Pq5NTz}nTF5A$ppS+
z9e69mkC}0N3ipg7|2_x9{apOB-rDeM;fa_(=QZ%c$B%6Z3|~BsQ_vECEy2ZOGv=L=
zX8lCx7GNCJ8&%y9G1K$K#RZ-{dxq_HTb}3SjaC5HYp*=kmSFXCZJ&9C!YAk$!!(*d
z91i&O+eeK1eK}_&0LE#=VK`uWu~q#%YrV0HNK-(X#%j#0Q5*OQ@l5+3m|2QQQ}9^Z
zeve1h5@sa~^E`dEgAh=zLGZZq<K?}~``hrlM2oG92wQ8Xf6tbqU8SJyt9velm<zJb
zf-J`QqB-UdJ~1c{0upkzpr~8ZtWBi}$?h^0*jBB(@DH(Oq)O_5xj)5XI_5@JE7dNs
z6EQcV(IB!L4yD?}WAM8Y4w%8TGGbbSqNU~e=NOn7cxW+PwuD-OY^jZ>^y`JmlIF_e
zVWwyAFfsW)yntZvW^~z%)3JuSB-lAv?inrPqs7RrUC%r{w5(b`4&N3F20#BLTrFvN
zr=Tb??sJ|$GjV%oy<VfsZWB8&%Jk53p?bPp!YTLf!GF&5(`I(cMECnW_WL~^9v<-d
z_7>aq2A5Y?7{(DYYMGoVCC#qN!piWx3O)L6-p4HPtPlPvF%?%CqNOAO8Xe%O)Ap=a
zSghS~5&*McsD-;~@G~xu1y3D_2kI+1Y`k69A>z2Iwul@Z#ARH2K48!A{+VfGq|aHY
zoRb20zM%2&8n6H@<Auh8JT3uMn3Cv)b`|~Pw2Oe_c^K^%JWc9?jKOYA4LP2s4Ojw-
zH8atFzKQTLzb0=y-mHdzrOjdQFnCR|rPa5f#WZ;#Ss=0yXjuXLneYFU0BmVZo@jg<
z_MUAHPS$t|BdUqD#v%8?7NKvy<&N}%r6qU_yoKnMk5l7sea=FLap>2R$JG*K{D0FR
zSwL$a7Nfw?`#cZ({Q31YE-vO9C4JvaV-_1crC_zbV`wP>kG*3;<Zw9P;o$-A-o3-&
zaKNjVFLAZKtZ-=u82ON4?HbRebWk(imQ}|%@PTcf&yv!@xaXK{QI^VFjjq(q69X`g
zQ*n{mc^RZki<x2eu`BMha(`I<=VOqJTdO(^pa>$Rqxyzbo0As|^W2ZG99ZHp>ZYst
zsI|%2Y-N9B0c4-qdbJX*T0GgWBkw!FVoCuaOe)m74lxBF7_Rb9_9C$L^i3)pEHRNI
zsVTL~i+^)mq)e2a_uqLTVf}gLZtYlGTz9KA4sM^3?isybJikbsQ_q@}Z`a!6Qb9c5
zQ<jCFgF(y1Wn<)bS-Wh((z^2dJO}&SZ$uUnt>-<o*1|T<FbwFru2f)dZIUHg0geSl
zajgmpbM*?*YPBlY6j=Q=W7gId9|2g}*r$Z;l2x!Y2hY&Ghrwc1B7*&Xj}IR{;M1p1
z==&bq%S#{x><<SZs78Dk293P{^DVN-=Ww7UdCb~irGWRAXR=F*mZ#}S5_m`0_qJb-
zT%<7qV!>`L9;$c;(l{Yy0h&U9L|{UIDf>Les=7qTL3C0AJ^wJF#%rO>eEy90lW#-k
zW`@8lKvQ)_XL9s1yKqnO(rT}}@1u1pQdI;mLeJXp^q2=+?dcRs24ajM3mE}&;zvpo
z(v%Q_V4Ow(5v!a&o^Z^=TJV*^r@i4lf-I=Fq_MTM4KGMq?=aJ5i@_5QGesk0F9?_+
zv!?4AsNYETnU53sc3M`DjoaGuS1_EiHasvB^Bdl~KkV;T5Z*Hq9tUr*7GrrVxqxxr
zS}T8LgSFXpdV?lVyIPp`>@%)D(9+8N^ZHp^DDLug0@v5qc>eskmj993wPet$55`%M
zeoSApp1@oijOoV5?|4mzVZblH{DODy-jxd45W!&_)$No7W6tpEU{#!rQ%{gV&;?-K
z_azo#{dlXCnWM)c(@qr&3YnQ@0nk!vj+8(FAPGSd{Z{u;-$krf7e#mhK<wI?f@SSp
zY;th5%J@6pSGpEGLYUoT97nppzekL*SYUj~-#pJcF+r`dIo_f&LsRH|+^NMI%eaGD
zCE{=xV=VKv4nmC84bmLTuB|dHa1;=$_^4olWjN}ef^nR%+Z{CbZ`GrV{S!scR|=nL
zYBTYc*3O%+WtMZG)}9BJ<3H@8vbKGn+#drkiq}3p!}1zDZ#>V`eGu7irdF2PsnQcX
zKhLPFZ^j#M4%XIMg)DhoHdZe{nQ<^<^-RIzih+f%J=>ZO6F<KPC}ZDlw-rv8){E)E
z&!aVW+e`ecjrHBTcX;;f8D6}2fzO{m>p6^ZK+4vI)uQvsD?BGlspWMvYtGNp<IK|~
zYu}7lUbppnjr;q1+}_^er=NbpdcDS*H*a*l0J^@%DpzP&op50W05R0D4<55A0X#3i
zSC}U3o$@SDVh(-{xgtQtXqBT@>umJVWSXIEotbs$_g0s!YVSTa@bQ-r`if?3FWUV^
zUPx>&E-_!3t=6S+=1J2h=KT>9v=q{Dv0Lx*Ea1*GP197)O~nw4ODq6I5VaIYUO!C;
z(ep|%6!c@hf?P1<8*D`a9v*fONqF(<B`&Y7^!UxUi20lXUx2Q4UWS=C2ReaeR|N-u
zjR?4HZ((p-OWK)z$7U{Xdas2xwK6PQMfTeZr^kRbe`cnq+*^LkJO=x_^*IY4KA&?+
z<MQ+IzdciE&D#rho)a?;#sSZ*CHzha9exkM5v^F-t@U$#eU0^cUFiAC=g+0eu?cni
zza1U~^!kMLdW}`SVt@sYKXJ|s)4O09N1jR8cdUq5F-*Ih?uQ?KK-YB`h5?tCmqlpW
zp5s`6$4`)ih&h(v1&)8-?=SnCaXbqj$tPp@AY=RBb5Tr8_~c~`k=%6PScAHR%Hdrh
z7ZfyZlbuG{l>{J!QlXla7cKzm744L!ijM5Zl#xoADxrOp=e<HPuYH|m>3LC0_(;YF
z3m*XLf^<0b{OpvnQ|6Qe>ch0O>%7PNSS;w>euumJ2i&}Tf%RH%{GH1>Jr2L?J|Fwr
zCe7NQO~f{acTaft1qbvz0eVESmTM(!{C5US+`w^03P0zWKKIYy8F@=HWoY@rmn;FG
zF6L;Ko<S>^f26fIfxZM@n6coo?+yICPBD1iAg#^III#b>*4YP5{j+SXwUmMu9Tgnq
z;j>a>0_4XZf5e+NZ%XNt5CUFYKP!0Unp~&cPrMJjrEdS{HE7XLOG9P++q!-E@&&*B
z_8Wfv^;f)p{ThGz)1R=}Y%q>v(a70SvUhzfN3v_nR@d)`g1?66!^6^^EZ<Wfi-Ir&
zM6!JN^CEWKzT);{>?)cvAyLweqaK)z-F&6Ht*f2$viWM;0V4I@x93f`+xTatcAhhj
z(UhyoVuxB2Askmg_c0&<h+RY<R~0?9`Z7FwuZ<T-Ug^%~j$x8NARu;n_G--Y+aC@{
zBG_!UdPsMg7I|9>{_Db`ege$oeaxe!!E*qY+xJ?r_kS}&eSiF3%Q`tS?{UnKMT+8d
zBrun>Y>$(rjrkaG`^Er_K^krtx2z#m>w#HoG@r2ff@9W3)6^~*<ky24J>!w#*iCV4
zIkEt5ZHOhN%g_0?E8JG-or9?&=2{A9`PDBkFY*2N-ybpD^?JVS&BoLBeX)Mnoy?lN
ze}@xbT0+u;S)<&Br@3jG@bk|<<HLs!=(-NCUcJI*vq6k{UH32y2wjBL-~!bBV)jDy
zlcZE85X^gymAIDT(6WYG3P{%PF*Ovw76UCSD93Bew8Xy|$4o<hFWGp;VF1Q4<76K1
zFbqZct35h%4L6?WZnrD39_C4hQL9JPmY`ZhDFHFEhan)PF~2{Rd3xoQahHnKH?Nt9
z%vCn>*OD~NJqEol!34OqW`CFWliH|CpIUY&#(=Ji7{&uc)Oy@(E(%VE!vTj$V_ds@
znc`#c)#Y{kqFSjE0rZ+#wRzEpUj2{R2N`0_Wi*Nt5fITLbfhWy4ao%NH!u?dr#mib
z>Yge1ezYwBu=bI)5j#I%E<R%(HF{d@x0tbMI7O=n_LuovQ5Kv{axc)liEdyU+}6Xs
zkpQIJuMr!d_ph{YUXOX%rDy$nq2!em?`HAwG@lsa6$3wpkP#^XfJn)tGRFrpUJbt~
zr8$s_h%sp?ne}Rg=hxR*uhtNm5Mo5%1;l)x3a_C77oYUP($iMU%3;CR!YBXm`;4po
zeveO|KH<xkFZlDH|BRcP8vwN{F(z7z5Xxn1S|TR^F&7kyqHZ@)8f*zjPkT%AU}bY&
zuVs&dU++R=+4Av%jr*+W&-hX(dvls<IxIpw|AYWQ>O=}a5Ms!u`eSZO>-KP=Q&t37
zF`nrn3mv^J9YO2TU`+?`Lj+>z&><p8hZy<byjk2Ww_DA7i)%6jKmw*Im30icZkv9e
zvO)$(PNqpk;HY4fNxusi5JIQ54x@6C*mbz*yGnqC2nYhY*ypN$|HBFgHtC0lhiV}O
zfL#TZlAfjwmR~>VMNx`GujN|1L};NdPh$_5nH=L>!kjg6-DIqtAoppT)imv66Vo7A
zy$CTz-C@4BFh3E2`gXrr_^3%*pMIXlSZ8zrm@E4(;JM%+6GlKp2(^Z3c=gOoM4&NC
zCtS%2LNO4su*h?0GWW(9c_S<_q1L-8<!q&f#lTv2aS0_W?|yu@imjC@JyOz!=UL5S
zm+2)+M3>iVng$#Wdt6>_AxUt5_eEO_z-rZ@iyeS?Bs6UtK8eNSZAnq^>&WwBx15hp
zF^=!<?(oYmAByk+fWu)Z{Z@W=>P!9kNv}>Yms1FkGy#$j1!$$Bn<8dSb=`4-P)lGI
z%PB8RRp;6XmHF?CGnuJ)Zb!4-?6>lax$KX(po{y=_(-6{CkVQzrFkY3J`mJu2J%J^
zkVy(Xn84A}KCz6cToWE6`c+>PS4ieX(7J&IZwUw(6*m)Lf=|Cs<$V_5Gf{E8Cevqj
z4{~ytX6oh6U^E42f`evhP6>$6G<&T<J1H+O*?Ve1kK<Um46Vv~N~0D#jU9FmyCd3Y
zL6FCu0PydB{~Le(>t7#x><ni(!x_$S1^~dn{`D{X<u89({QgP8C#8h{{O3Q_wZLaM
z!x_$ShBJ5&(W}CL{PD+=$xlxbK4&<?8P0HqGyE^%bWZLW&Txh^oZ$@L0M5ea3}-mQ
z8P4#B!&&&8;S6Uu!x{c?I18UMoZ$>-IKv+fXW?^(Go0ZJXZXY6EPT#zhBKVu41YMB
kh0htzaE3FS;SY!Z2SZI^Ae5Of)Bpeg07*qoM6N<$f-r!uz5oCK

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_repurposing_annotations_thumb.png b/0.16/_images/sphx_glr_plot_repurposing_annotations_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..fbed6047b3923c8efd6722ef9ae77fb4a6b5238a
GIT binary patch
literal 105726
zcmeFY_cxqh)HWW`qZ@)CTJ!{i4@S=rL_{Y{lptE7_tAR`647N0F?x+?iC#vD&M?Bn
z=w-r08NI%q=llID-u3=)uXV3e_PN(t`#xu1*S>awzOE+Cy$AR1+_^*ZTua^X&K;sp
zw+|}?*=>uLMFI2<(VaWb)t?#%z;><{K0lhvET#(Ceq}N`vcM<ly}NK>tkz3Pao69%
zW0cgrOVp#ga?i^nd639C@!#M1Mk&6Yg;8VQtI=P5#)QFla^yVpEdT~~UVdhUM|?dN
zFK2VZ?2>TC8RQQGmLkX>p58EBcOH}f?|Z<~jmka+H=W{M7ct!>{X?R)+X1&dTs?f@
z|J$axrxrl+>HnJ<O7Zc|!R@%)4!UIq)&JibkK*0I|7{~yiPpRW{ht*Amg4sQpV9w|
z_q6+}_hyCbx;fyaGv83dhkJ_XsXC?dQY0dyUvV6!E2S(S7;egLm2M@^V;?Iba<A(w
zc%!Kk;QdBB!F{nKbu(vsJHt;DNq=^TlTx`l%)u($;8d;vuF{ula!Pse-m0w=?s@5w
zV+xUUr%PuSS&OIJVFVV{p9?jxajDnlzd)0f`WWs~)%Z_d{X{$(#~}lt0bKfz4)>%c
zdT`0{tK);A^@D^Z$wuG6sLfIq^lbm)^%?sVGnZgV1O45lnnPLyl@sr&4qnz+fhPb)
zh^A*X)cPs)QqN+Rh9x8a<`A64%%x~YAp$oPnRE=Sv=AHbDmdytYtx5G1`6KO5$j=H
z;(Nb)+@{1m5y+FbbcV|@yjSSajvWlKt%|~u6M}6Q{oj=%S)T*xy-WDQdmjIwKv1@u
z#`*|-x=^pCmFBjY(0ul2CFI$f#%_;dh`_0i_-@iyq}*JXO7t?X8n+z9)A({;{-qqY
zud2qe{sCP3Q@_~Y%K7@%34pbLLYF*4Dg>P|GYj`P5yL<Si~dLNRA)#)SbN9O%NzgN
zo73~7PIUMA`2OPM`1SJ5r;^i7YM$_UDP%-Eqn)}zAl^cZWka}-NW&vNL~vF3HGiJi
z^!HVmV2PG)YNJKYCY#AJ9-jv_54bvFHzeO&N1_M-m74>>(5s`};E*frtg(aiU&HFz
zcdtypswE}R@9eg_vRjgcT@%KuFf{s+{yxGa`DG@ob`qlztV%rUFz<j}Gk00tO~8+C
zyoel#2%Ppx`{zCmm7hwBF&%VvfR6{CZD!PVwntJG$)!!{>?L>8aHbl{gW&Wr%5LFH
z?SlRB6ak%S0&ypTZkXgV*y$l5#x+cEGXcUbS#adGS><nKW%c_cgH_2HK3BXF``J_d
zNFthn4`a38%~vQ_aF{VX+#@;e67=<?I~0?1uJe;QI3Ag8Hk*M}AV{x9@s+4&M4WUz
z_fQp2#<u3|KA=!Cda$JUm%Q=639Wec;%bsCt$hH|*L3!!1RL;-6iKXxezyXDNd$pr
zh?<AHX6ZX(^*Io{2(p$eeew3kQ1F<O-Zc962`60xPtjTeXyT$Ela|C*Cx8Wd;di0Z
z56!(9_<I<J5Q$!5d#3`UgJ<($gW67XhLKjXkpa@Ad-orMeWp#H0~t!Ay&rC-=Z9Xa
z!x<Z8rr^-C!;wur@!&JeA}&Og!$%}&ai3B->|EV~R~aPF$;I~q@&hJbg&FGDEn>>a
z)L~Dazh}u_{1h(9J7oHx6cvMV20NG02JY%O#}?`kbaHv?yE@2>Nm)2lZq5WZf^96R
z5PtKl4F-EgZy<Mzr9CoAv?%~0w3AQcBb;+t)ymcOuso`Vv>X9kd)PEX2M80`4stgk
zL76NdPH-iX?}t`AEf-IJLLJS!-H~y}w?2&Uj(=r7)1VC$cn==3h-;5JDWKiFixoWf
zCH6g~cba13Edz*+vtX?j_r_2I+}uLMzC)#YRC(H%3?PX(90owx#F-Gw#g(EoS5c)@
zrNdH~mipR>3JEQG)pon2#1HUjz9k01cq4e|xz3a1lUa5iN*+?yaG^1fdN-6#xq!k|
z`q)&T*_O5_omKKNXfK7}WrGowPd62%4S8Sl@}>`g45Ql^KImk`-E3d{WH4y+j%LyV
z@u+Em7(P2LxFeyqWsJ6}7hR;9PZU5Hwim)gL)O>lEFMD_7a0Up&gP~geBAN+x;czz
zeQ=M)GoyOJYzzPmwUyVCu8v;YzpF<8C<nO7Qr$<vCMtFszhxw|Dz0hi<e<qKX>R>i
zi_#+sLIu<S>-?b4-e9w69<BpHxQc`i4p8|Cp&M6M{S&(Qn4%mxg(=&t!%tw36nYNo
z@riC}#|=<@HM>Zn_e0+TS{V5zK5v|$o;u{Ct@A=vO~d55SmED&C3*{PAI*5UqJ94s
zHBM1UW%312c`4QuCNjo9rpKZ>*PO>gyxY4^M-C%Qe+SqiY5X5eMhU-ERI&~T3F!r~
z+AVqCk=^@6IE0q$C}4xmTYoPIM@;e*kLc5v9@m4Gvs7-PsmA}*WVEPN7Ki~NHDQt(
z{75Y())fYleBS&p=my4J7Gg>HoD#_EOD{de92CH_@~7Yf@05`J0O&<Wy|>U^)x^{#
z!KOwtL#=@z+cr!0Rf2f_KGw6nL+bnuTo6WhOz2@HW?nxJ<$w0^BnABd$gkl(B8-qX
z+sL7he1=8uIcpd|-UnYF=Aa)C($Cg!XOEd@>hO%iRj!Wo{-Bc~>~n&~(^f68fUvNE
zppU)Fd7EBPjaXjdKz6Mmt+){8qG$6PTipO7hs9bFF5&lrXE9f3Gk11n^&tAYrQSWx
zOMVNU0aFZ?ouOCBQ2A87o&#w<-al@#xjO;BikK3BOMkkyx%*Y$)952X>W}H08Grxj
z8c-bp>*!Y)9!Gjdw3`NUDOy`;DFf#VvUIDXSUD@$dC2;=)Op7gv~lW!Fu3kkxRv{(
ze70u+SB4%!+mNmHRkcF@AWCDBI4_6;#H2+gcyAit@0FY5P+$_AlJu!d@cZi@HKL8a
z-~$Wzm_>6e6?kOt@)Eg@>uDB$g<Xs`RBiAG&C!CcV4B6h(X|PXX^%vdj*F+)6~0)<
zDW*|0ybQQ#LX3$lpUftIdV0mJ(FefIm)mpqeyunZkW&N(Zc*-X=yo}~xNV<XD{5b!
z6wIRYt>LT)!tpT;3wN<P#pK;Wt_Jk&o;t!M+hrdGa4a#Ez?_N4h7;!JJ&8bp!wuMY
z9!aH_>4raG4%g>&olEVd={?w*;OjHW;5Gae=PPz)xpNthm92JElLo~xtsWt%C7zm<
z^Tivll7HX@<_xU`CcPeVO^p8vV_C?j-k$AQc5rWN!uExNFsb2~*2$m0DgLQYxN{=`
zlhrBJ7lZQVj@7D8cD;#BqbyAo4gx0$#n8t{w6)J-`PB4H-Uf7imDdsmzI?bDJsH>D
zJSGy!3Lr@KhPSW;ZZ}|mt<3rKik98$)=R<)1+53e2aQT*BYb$ybMPWsuIcl?k7EZS
zNc!o)6ocQ!Jt4Y>B?DKa7aa##<0=5PXH#A83u2I;#ml0aeeInbm7gGCY_r;lDcMT-
zVdpu{5NMcaTL^PS<h#u*Yxvx-!uCb-1YWVs=pc9nC9Yy1WMTU@C1Quqz0)<l;#At`
zgM^i=h2bA`0?$fF$&B`7O7GWqoC);2W14L`0h<hL0WX!XT(*0D<*OvA5%G=pyvAU<
zf+fy{JECKtO5^^5*?!RYW6<{w>ztAVt1lUWt-ZeF;qem?1Uhe+9}{CNYBY9d&H3qN
zmjc~NfdYJNTx-|Q;81rg^W_Mt2XnzL0*A^2z;f4rxZ`xVzjf3iofu?K2hW=%M*)-k
z*yczoZHfqbPLQ5lipyMt_d}<V7Y<2@dMTxwI`I+X-}ms_>{n<u%ZQQTVaGtO!|Q%l
zIQl_FA{~4wG9$~9h=<{K^Ia<n;EE-8X^ty~t^MmLZM1gRi=rB0y=Z|XERQ`b@U5H%
zo5er|mV7E5YVIV$J@iR~yTEPYeW|RCynTtYt+*(bbb>@)zmfZsz<5)Q3nnpb#EG_c
z=;By|<*8!y1ZCPDfe?dzU}t>2KD@ZqPS{kr$&fe2x+#DiLsKrlbcU`hCJA0+Zcr+`
zFT2dSFH6HNHK+Orn_(A@pSjYvhaigvwMY*e@w_J4F;or#r&F#V!qLq;71*~*`GDTw
zoHeNi{oO~q%bU#Y=0D!PFmi3-c$jeU1cdMBAZs34o)<ipLs$p@;S-_h``&TXv=aO_
z#}RhV-MVvby_Ln5dIzK7oB@Rk7fZy3%N#HTgoG)zP)4NhT%J-C$R*1~44>&dA!Ptn
z*jKU`9aL>6(=(!&@N*ZcoH3wLE#{_<l{SfSs&;CK>&S3YR+lSgFs<>2ZK4jeq%kpG
zWsL3g4uPZG*HL~PmYG1WH!Zn;Y;T{g%hdye|FD1ugf;s)?=CRDZuCx)H&$$?!uGPF
z(TB?Pk&^Z=;)7Tq!dfh`ch#LDWXK{^Rqjru3(W6v4~2?OkA7A}&W(TVak?LjAZ*dU
z!wrTz=Tc7A;a}OBIoXG!5Uw|1ij?gmjauSrhT7DtCZm%Sxo50DvVWBhS}TVXy0H7Y
zT&}8w*l6Dd|7VWaS@$ILKGfq3y0RF0jm+x2>JPhAxw(h^H|^SbuzMrE3k&<XH=M8X
z)phecuy~M$OW>1^bBx$f=`dBmR&@lJg25v*`$s(B#cxQhpa0U%<#Ig9$E2rwsnDE+
zNrZ?CCfOAum#52BpZuLAdQdTcALz9py4VrIcTVXsNPqFduEA*ao9Xl2X~=Wo?==e5
z2}e7~$jpA#+`AXm`VgCYfD2!>JP-N2I{{b6L#$B(_ic>6e>WyM#`lN5K07=RjPyrJ
z*JX)Pe;Hkq=6ks>y!4-8MWeO0a8ixaH_c^^$1li6+uA#);w-h*6sFjriI+I>@x#}p
zCa3t_SGN&0d$+y*m!LzeCwAH6esR=N2=WYxZ<&lexD>8`{7bA|uFjP-93%(<psPLQ
z+x448@#rc?oMN#oWV?vTxwvjY=1$v*<{kt1u*ij6rX=V5;pq{cnNz!(G`0Uzgtl#t
z-r6sKz_W($)-k415%DbJTTws}0{`n@|Gu{RXY!}9d`!v#%HQtCDrUP=zsU0pBsI=5
z1uZC-$iI8@FYJ2v^yCJw0!*7w{@vLwxoPbrd)gGrV?VH-6+<{cZZG<g4Sg0+XK!x&
z`Zr#g0Ss2B_e76@_aBxh*12c2)S6~iUZ|1M-Pz;_@v)_TTna!j2^W8J{bE9sMb++a
z9PQff8Q4K@yLRBSeAQg5#aSj-X$lcM?@E`v8uZ(Jo?Bt~rSwIq?{K`YYPsm<rj#<Y
zX<Pdsry%0xD~}-pL8=*(>}%<4Nq;0q2P8G-uq4A=C{U04gnplU>6IRWcr~nV=aEzr
zkgvaACbwY^emn8E>nF8gf)Vd#^t;vVX7|i3QoXc9>VaV0HJ^yT0YM8tJ38y#iJ8Ml
zV&1s&>>4z#=`e%kfUyVHrr8ebsBuS_cTu)}1omEcucD({b7bP^EE+>2p1edHFkZbA
zPXQ#;sFjh@k=dLqE298Hy~ZAY;OiS~4s%OyG!Jr=Hdd&X;m9EfGkjl3E6oH~6R|xk
zB`S@8h>p?ejsp@0`Vvp`2<z%vlZ<O7znhq$+n*k&+#Ds*5v%+sO%{5-P1UY>eascM
z`dM$MS=4vo?DC8S8~VB&pDaxpdLGA;d5B!R-j@wd{IMlu|BOS92Az0lnotowu}^DD
zUBo0DWVd(^_(+{Qv-(>Ja-Gp7fLSSM&Y-jm5G$>1WSDe|3n9%r61wyVN2@XJ2<M@l
zi+)Kx*oKs~o6I)_>#?NyL1D%al88xikhsU&OW?-pp6Yhg7Sam2g{#=Ph#h}eN?R*q
zM9**>qBf2AD<`g3M&Y^x8*_YT^riFBd%+n<&a<3SKm)kew2+jA&C=QK+ZrWT)y_Z^
z75tOy1V<qP-ILDg>?vK2e3puJ^|;Rg;jEFTkE5dH<MTPbo4~et*OlEOWSKBNjuiHK
z{mP<&s-*EL7xA+nmbG#&@M#F?=031M^?lS5hkTV|uHoB7(dMgu>1Vo|$TLZXLXWG)
zI}!5=t_we5f?hCa7_&!$q5Ran;kiIqrn8V$23mexe0&WwVXsQR;mxb7zfC))JbrfO
zLj2oEV4Q0Bv+z;uhp6^fGR%Q4m#sJ5o$Kp9%fpM;fcM*%gs@Y(%fl=w0zcz-!F&Kd
z-x`#!M{V-1l4N%uIL~kD`Y$~)l@CVU<Xh@;|1CIvVErKOuNBcZ_l$H=f+FxG)tG|1
zXyDsC4IPiQbWnc&r?1o19F8uwV^Y+baK>9F*UPu=0)@6;_XiJ=a1rsuXdV38uziqS
z0>Csuit%?vXf_outmMQU(T1Qr7PO%gV)?V(5T%JiR<%i!E8nILMGh07+7x_iUFexm
zkbMrzkQKg1H(?l;Bl7^^&)4-s>wR8LkFa2WSp(nP0dT3%QTb)OkM?_5iFzr6K=-Uo
zwJk@D8du8XQ_?nX)#e;q&(rjRXMW{>yt?OOwVj!>I7CvW9;ii6iWbw^#+Om&Qj7(`
znrT1yYkJ}DPCB-oSS-3vo61>A4hwsBJf1e(at$X`UEHf7oS^{KB*!&Ea7-Ydu8g(U
zSY#4l0=qv(IKWZ1P1(~6*#CSxZDPTw5!d!;Z%9$^RQLcTo?qrG_=tW{sJOm0Ql<N5
zE$m#SuvWBoF?1DwJa*i2FnhDC7hh?iobU90KyyIxNjijqjhu}ab)jldsP|(j0lT5t
zyd7J%l1G|N-_EZ-J}l)bWG4sKYFjV`ACBlrVTD2bijmU8v<*;=N)qy{VI9s4Qw88t
zk0|<>3TAum7Nfq3iU~#YyK3BwI6H|=t@~z+ZUw184wPcrW{xtM7>T@u^?h}PYUa=K
z#_S@Ie+x`(ezE+CPj8YcT5?+?-)NLZyYlM+Jcp8ipU^m4{LhwuMZ#%bYk{qwy$aAY
z49^<s9<#jl8FEwdj}3R1Ud&M;{bDqj!cOJhDbsdB!mIn#n5|HPOc)|#Hmn-FcGcYD
zV=%3-B0TlQehx_GNrr=m;5CPNxTEOMzgVLpde4ybnv3L(^NPKYq?{wPiyRieS5ZqX
z80{p5K#MZzi|cGAF*>#JxC!b2tQ1ozxdgPIlx;7O0j;%u#a`X|pW?U$T)ZtbNRwnt
zj|xj^ZS7vPBbPtk77AU*85L~$AI=d7n|jkXd*dUzx9kic;Pe2`ZqM~tTyXyH)|VxX
zuJWl9h38#TEi^A*Z*ynjG9eyvhKbQz3C}_KH)c<&Ku;Rv&2({2oIYS(SjpUP5A3Qj
zo9n<4n~A14`>-C%6lgNgE`Uc*8~o|_7EK%14t~fPviF>lQy^K3MKMBIp71v35N|?H
zG}9Ffr<j8;deCwy4f~WdaT=#8d(?b2YjyJB-(<1dYK&|m^IY-ysqgpoaL9^ke0&`3
z;F8?LoN3=(j%dnj&^equJ3dclRVw~R%@DZ_({<FI!(aW2L_OuAaafn+2eTGvTLT))
zO?bst3|a!n65nm6OCl5+NavZc{krn}JxpxW8g$XDIbO*3Z#pUEnYcJAXyfXJAnfts
zPT$sQdRddv<UG)>Rm+v5?Zn>hcsUi;j%66%`3S^+J{(g)pB|jcCeP}o7DfLVSOjBz
zT8!olZZ_ZX0b6zr^x&I3f3`{FGuAF9Im?bFh$+l&i=r-boOMt4#?I>P=sG>Gw-d%Y
z75A^hu0ARbAcso@trizgFBUs)&zRv-sda&^Y>x^jE@{~+&aH0``667|<?i>FP0z;z
z>5aIddOC0|CYYag$mY1pO^lplztiDYMN4=*&r%A{dAh(*Q>6c60%hvh)<mU5#e}9>
z-)nX9wvT1tF#v+HnCYFX&bT0AoL>(^{zi&eR=)j}wo>eCBZnOE^!CYC`e5*aYCrct
zk~QPDD`F=`nAA&q8q20O=-k8L+{jnKRi6$K#fIgZ6x=^jh?*A#!YIZFXM_(lb2Ybt
z@6)`HNndWOJ+bePW%TEdlQH<A7G|&otwyR@?JNASQ+28#*ko=t?`!IEv|gv8)MbH!
zJtY;7*{Ab^Z3V2T9c#v%f?aR&WCCzCFXNzKr}&nebbhXbN)p*A=2>xmX}~#xQrGG?
zt;CG6YI}i`zSZpd{L<M}|Hz5ECs{S`!mybkFVg+G?}r;Z@Dao_ptS5Bf$`QDa_e&z
zN~)bU&`>#~W2tuu2i<52T@7sY`Wd{Tc!O**1t%gs3eyv3W#1VcFQEMT=3XJsf{Au!
zLwt^_m)ZnRyPP2uJS**po%K`rjp(z)8`|iy^g6y==owDET%lSvSj$4!G1B{R$_e5r
zMmRt2Ka&Kgm896>>3!1|AfGxK?cnr#R-R&?48#BZu&Sj_M-d{Ub*a(syt)~p94%#o
z2VH`m`gq9adyM?4A)SUy$jgg$!5kG+T^BsReqN0})nX?$am*vZryE|+{Crm(LcQPK
zj4ZKYf2ly*-8v#U`Z!<Npb6Y6quBr1mik<oo_x}=#2(!LX2B1;XpOU^v>`HJ>IRVy
zu(?;KdON)KZ(?r%{!kyst?Qu3cq9jOsE7P8bF(y)5wG}gMd{^@L>jI#+8rj33X|8<
z&BhW~r4a1S$oJpaM22uaWbzkQAAiR4^{Z#}AI%SXzk%Ejd~Lw|mb%D1P9v7X=aQ5y
zYCgFiQ)xGHmttI8N?B)sx<B$d^u3~z(msyx%dGUn=Irf_tQOsQeZ=*<c?`2n8`|@A
z+Hk^Dc+^OSbw+3iORBpmh47umaFjQvoS^6B#nE_rhNjB8I@@h_vB0sc7)46uZqo;t
z3*GG;tP;W}QV)rE_AMg=q7W(3()%KE*?@+cJj57Ax5kmPf$U&udTryt%f)u>Y+mN8
z#<nI59G6<HmLw}$%PFBW8TQlQ^kX24O9Ce{5VgtEH=6erY2ba}lQ9_In>WwaV81{{
z+GpA0a*k17o#Dwd%jv@Qinm~6t~r_CywYY-q&>m%&m=-!oy<hOf>m^g91b7bk^TCr
zvexciz$L?Ioy@!>iYkd$E7CwUt#np{CIcl5s2?Ez`Y~^bw~T?oINA7(iH1aCfuF7P
zP`;C+*1F%5X_0^|Rs?ZW`eqayRK-tr`ljJmHjII5=nf5f_t9rCZ#a$QW{gBRXNhE)
zLV)ZtMZ>%&EgF+NTiYMzKW3&eIl834n+NeXOo`@1U$I-=4XS)Ao4Tp>+Wiq1^o%g>
zQorhcb4XJQ$3`DrT|6tQ*pf4o(|wt?cC`=VK#0~{X3@~!?CN8Nd~&lL#&K)iRCezh
z=5ZjyxT&Uo=ue5n(Z;5Bi9D7sY<}ZRj8h8Hs;_L~M`ugy`^m?TLY~QLDrY-4?v8ku
zxV&nMgZ@IY&Ye`6q+U~~QU^XcobpgF5|)%lxrBrX(lYgLMMDe!p))Ir3D*wlDXe(6
zAA;|AqCe&=S=LW#@cj%YL@@D!$d7$jVd8$v?lJVA8!H4A?Al*PzJ9NCr~zOY2<7za
zxO%F`$w7%$RnN`>w7Vr$B;7N7YFUt_fduLSgo8Xr8{BQc-4L9nL<?#!g9yu-t{kp~
zFkSY0l0@50uC730pDWNdeAqbI`tamg?@!%$8CjtV?o;)9l$)Gy?TkUYa_a!2F}Y2l
zAktCf4{qXZAt%SzwZU_pMEB$}?FHicSvgZRwlmh#p&=o53!w=AuPObwV1e<TrkXZI
zrF}2x#`Yprt+iW#B7(BH`D)@&Xo^#wLoF$yBwJ=;<5R?Bj1Z0OL3;b%5$4rK+I~hp
zTu*)1$=9R3hC4+`x{XRvrBnaK&ERpG283QDnS(fJ;kk0EMNxWx)BX1dYI<?G+A2(j
zUcT5!e*W!6vRECnAz56FOru1=^T%Aabt@GnH8+F}n^n!JKsi?2Eh8Sp#jCJ{@BLdz
z`&agRSUz5vaE@N>5mps31Wd}v>rc&Y68d*h9n9gt_HK9Zn7({7hGv!nG!UiaZOlvz
z-W%~Q#FG}v`lV9`Yu%UII1_Ru=6T|Bp)C$xAotE?t@c=OLkb~52Zs&Z)lw}-52EyB
z;_;jri4JjuyJizwlTLl@vc~LVl*$$Rjjl2j?+5_qu*+rz=li)7K{$FIM^S?;`>!*R
zGJTawox%c;_cFT_IWWnfx~QA8QkIDxCQja=+1griQ8$g>=CgO@uJN5$?sEC!=|`NY
zL*r*i*Z<%x9}~{ubFbK=Q^d-mmL40#lPf_?_TMe2uTPLB@eO)OoL#?}=sjExZ0GG@
zltNOV8qAn26x`iq=f${FP&nkLr*rbV7~Myc?MvztzbVrkYm-qvU^4XmZE#mb=zwah
z{?JmEb!g$t?T4`AaBt0B<Lq;?4SRwHnwSym;t5Q=&Hlwz8LZKhF*|}J!#!#qnjYV%
zLX(L0Qeg|YE*_E>)C1o-p!X;=CY<}!8?s94N9!3MJ3B0jU|Ilu(aDW670uj5cR`wM
z@(?Ciphhi;BPSg|_LdLK_D~bPg(!JvJD%NjyX^90pz={Fy_`sdEKh*@s{<Y>3guM}
z<XV|4e}B-*#>by;o29l$H)BV=+ZvaWY~6^;J|@dOEoZL|T3LJi{EH-zB+ylh<r80B
z-{<A*q+(qAreEty`|80V+grO|sje7vcXl^;e5g&5KX%a-dvwdsB@oVXTH06I)vj`U
z7v<LDy>kqtwoT{IL#(#8uA6%U<F$56Ce_J;!6TLS*@G4au)xA2j;Ei6y^U?m?+;td
zd@ZyX(`%SYEt&4{{*?f=`fatCWHa3!f^!u072vYz>Uh-@$X9&ZQ9uE`7tQZg{mN}S
z`MYrUxW(cJ%A%FK>O&oM9iermuPCU?Z%JOSw3*OCp2viQ`Fp%krKN&h^{xty_f=n-
zoVDI5tVp+QpK~7i<5rnrZ~qC53>q|glb9oradOa)TJH-C3(#OLilQBm9PJM*PY$jV
zSvl|#Q0B*Wv`=YjeylnhdqAWGL<J*UV#=itU^%5RmXCuL{e}g^^r)%>0*&4QZQfF(
zhKQ4JF+4!+oL%DjAiULlA~3ilmZ~sHW91k-VJCp_qUMTfHlJmru$`H~Zu4A2*s6iW
zs_%*8OCDQ^I=66SooWlq+K!Y3peuo{Z4grq%MzC5{2SaZOy#A{na=yD&MUmEm+r{i
zh7lon1Lw8k-g(Wr$%`1h#H+OP&Y%<mW^*D*2b>cJ?gYqAE<as;hA;joCZYaG?eb8w
zXDbCejSNxC0Y570vmnW(anH|(v(jknR&5UvhNaZ%J`6r9qO^Q4T&km!@Qc)V$qysc
z?g1BjtD+`4QkuXJ{kr|&_&Cr)H>Y`)UHD1Brst!MYrnp7h_yn1#%uF~m|<K3<nPmb
z+FNEDJ1Yy`dpr!NNWrwsKV#>(=#b*dYCh?2+i6d`{WHc}s*y|voNJ%rf8{7YBEcSu
zHaG1bTqcAIG8R+<1t~|IOVEjlOM-yrt3!h0JoZON=U3Ik<<^Rz+z~UL@g4h8x8xSt
z5&}~3iPoa4hLNmo<Y|MxY)Ynaqr#{;m18cTSXwMA#+lZo^~!uRV^a@MLBlqLW%Om`
zXKz$E`p5L&Xo8%uvVHKr&%H)p?74!sZU8w#I6%*pbM66DSxUo^Yxm7(^FSl?`_g~O
z9Lywz`9Ip-cHY~_lY7Z~w%)%s54DF&S-m7b#t{VjDT2?AlR4nfEBSznK(#ORIY#;d
zTnSj<sEm%dy=GoQnhAdMKzRukN&#Jk1{Quy*(JiKwrmf4E1K<Z!E3RUC>k6)u=v$h
zt!q%HjOXXiVg0e&U0e-46LE8qhv9#CaqA;i?A0}jG0_%O!|Up<8SfV;YYK#fUk0Ba
zCU~_4;M;lMy|a8f@nyB~Pgy&AXE3F->7@(4zL6QQ>s-2h5?)Y|L<F3F!H`yZq|Jqo
zy60q!4A-DQA)SU?7d?0ierFv8w-kI>#atwx1>59GVmekWBZYK8hO>FTuNfRejLUoP
z>H)V|vIHp-0~^1814G|OujQ<`T)LII6q~Vk=}otw;-)I@XjC#WApbiSD7xPzm}@Gg
zC-o+i(!P@+Q!4FI;Cm+{4dPZ?s&4z5-8?i&F|ds<|9-(&N{ENCIEghIVwt?n`c-9D
ze}AB;&e?j|agD{8V3Eu_3@Fc4V=8XyVrd=aSkuh@egSmjvvPH$-EQJ`F?6=xqZf8A
zb+fM$+-(B<5^tpi<gsKhC?UM9I7e^wn#o=l-1l}$GiF|3*NstmK)21Lr^P<Oci4pw
zS&A}`-U+f_yoZGr>KOxHq@7^c6TKjkS@z5LqrBzxe;ghDV0=-4!=lom4(vX*S+>nj
zh|SkS+4A>Eo}sAG@u2|}diFMI#cid|YvDN$&jN$(ieKg7@>V-=oQ>6|lR@BPz<}=}
z1s2uFCp1|c)y+3>Cxm0Z+Np&%PWW$}M2jTs(`qC@Z-E?R6%XPHZ0OR<SN3dYe#Z{W
zq~Zw}a$O>ck6H)Sb%dzJ3OA012rn_8TRj(B8lPjNbg$AA*2ZPXX~yae?j=nMxjH5S
zmdYC^;&_V@EBbaE`j~z!v!Ec^!$OICfdamJt~(>+;z`>Dt~9Z~sF00nVV{2=_poyE
za|=s8c{rK)Xn?+$T;YV5vpizERrt>=o(7k|s*5h1%CpaPD!vo7E?rk^N_zh3)aNY5
zy7O{~N%gbGMav|_zM4HNOVBFx9B^|OcG1t2KUglM?)mu%&j+~TEpg*qoZX>Km(#_d
z*pte^tvT}(nSlO^&)6U%phhn6dCt0-O7V_X4w=x1y45QrYcfQB*txJ-V<ZKn6P+!D
ziLJ~aFq+kVp=sr8t69u0Rb>z63QSlDZs`6)KW_?s;@%(rS6dfQ%uG_L1Hq+F4jRdS
z-(0Pq)^P7-6%JYsR%WwvrU3CDNY}aE-sZcSq5~U`PS8_0>E-80R(eiTQ1z=RkSF~9
z`MVkL0h@sl@dt|CMXb(Xnd!zlhg_aibKO@)3e;Od3Q(x=_nRuFl3U0b@oR3p+A=jt
zvYHyK6_6j#6CQp-8`rowp_*Nlk?x?E`;u?!*$>4$7dPp{Ez)uFtRQ*|@E76JY7XB)
zk7+WjJp!V^PpYYv3T>k5&ezou{#GNKX@A%85DQd<Q{o=(-hEj2m#mw-FU}^HX}_V-
zV6`hbIwg@EPQdltYE>#<!Kl*PR}<Pg7d?8T=HHAjPfq@<76rY`G`Rm|`G-@nv}SoV
zx^<b3MJb&B(Y*e=uV5@GO^<J<=TF$)0}2wjnBMWE<^TxMQbB9{MoVy%=m^jaMW3+O
z(@i%`YTKKSO`Dnxv*+iJyIk~pLG<e!(+maT-aaX69{n8=?pGP#-1u1#Q1P!=LSMmA
zm@&;!%aO@OJOlfnFwhC96A-MN?@8;{eDqBQ7#Q+i^S_Xvtm}%o&@U&cC(W|cl(=*)
z<+_gQAInY<ZgpLP&Vt`D-DsS77i<hL^;N6y2_Pw+cEGACaJ$;@J@;r%_DD+k?9(y0
zf$OJ_^0tX_9fPb8jWq_a;)d+ZAEK%B1f#+aE0LQ%%C}J6L4d}Xg^yV-$WfEUyKrx#
z(UASn!J9iJ(~M&1v!kJ;$d%%FK@|F4DGI!sY&y^_t+Qp5w=#oJBjQ*0sQF@ADu3Jz
zw(b5HZK^5vOgZG7az+=2S#EW_ciQak?$>0zAGh=q9s4&_F41L>RgNW@GO+A$V6sD!
zPnKUJnstZNtr!SihUqm}O%?<w{t~p)p6*8b9i#`e1oPV0lRT#^^3~^W(Sk8vSW!-X
zrG{8T1L3FZn=Jk4pEa^xDQlpE&~~<vx<s51hZb>Poq-OsFua9EDnRF9+{(22i*7-B
zn}%4u2xu;)*yqiz(AER|k*J=Y-XS(&5;#6QaA`+<re7AQqsj&@Kx=F#Pe%&ZV3wev
zcYUX)vg3t%E-<e?4BDyQ#na-^qCP<V55i_qvR{pNWx{}*I@hddVb=P`g0Wza)a>!W
z<sjcMwr#7N()z~aL=SoUU{gCnfSu-{@~35kywsXUyR5w~ukW5AxbPuw8i>^Bn#e6e
zA#yJ%SvcC6z}cx{(_gt$Tj&h^ih4tB5Owa%n~JYp2R>Mu>n32?d2}-q78JuSPQnD&
zl1}?&Pjm;r;n~N?$2M=1&MvhXP4~RYLWv^*U&;@aw`imajBkMY<jQ2FW1#It`Hu$6
z*K&hGAdMz8>$WZoP}(r@UQh$21@*m^jF^rxFL{9&3Xg_oF&a9wa|%;7DOBB{CSN{B
zsee8dmuKSQ!7Ua)I<Wesk4q`@IWC%d!wq7#MuJ?VpAq}ub({-d0kt<hqUV-9`v<8W
z)>0NpopZ<f$URYC;Rv`mjtQ~O*(A--)n1PJk7e}@M#dQ5cGl}-V8X!sLe;LIO2L4*
z6!&(gT@2Q%?yi8D;gD##=Xvb3$m+Xf_ja%VuMp3t;m&Ksfi-`z?IT}TV@jG}Cd*wK
zul5Uw%Hxy!_>TN^YkwT|c4p}}G_+%)aegiVj(t!qoD!Z$v&V=2PME50lqUnu{rNsa
z*25z>Xdwm|=3o#`GJGzE^6S1(txdU40{iJUg3CEMVaB~vev5fye&<0cLH$WaIv@2o
z9y4|{+!^lQ(5T8z5#_p=vyp>Z`Dy7t)M#`<B<aDpcRvTG#@fCdqOud&qlw`>&GrRt
z?v0rr{VdXgGcICBojr5zgA?1Z588b;6j3i^ML$p7@isr|MurvN*M1+sl=~?QfH6o3
zVxbUD70^?r49S+LYVP0OmetLY0%_{&A({FRLKnXL;yGu~?($Oih4iYsVrIp>Jd?+v
zZ`a{k>R(pcN|SX{d#v2#$lGOMCmRxT^KqFyPW_tjZ6#ECc{N8(yPyWUOwd+jmP6D$
zPlM-w-HYE;W&0AQM2^s1pqA{ZhbC<*lDXLvG}`4Oo=<5Z?z;mYM0?T&j;DUKTT=3T
zvMod>E0Z-+L)aVq_y=<4?>WnK=Mx@sxqg$4o%s7EcP#uYyN?xTf@|m0{Yi=UGx!Kh
z(SCwL_XNpOB|_K{vg$p+g2zZ|adE2Gtt;kLmTZc`0oo?$N&IK_WcshI5vVSEs76px
zgM-@BmkowOCR#Sj=QFH!>mdaTtJA9{lvqh?giR}-zEH)Te{X_*eq_svw+s(M-dp=V
z;t+poy2s-I88deN)8x-&p9r{kr&DJ5GOd^uP=<X;(|VSJ^EicWTs3{yk!qydu;&@m
z9?28q<l=BA&+}K#(*mt{>d2=`*ISDgn`Y|e5TFh_#6-uhd60vw8B-#_4a6pV?;vF3
zQ7=hNBMYQ58WHH<U537JlNtq96+Y}m4$`t?t=%0$nA#9-=-H3TE1IPliB#~~&d^4J
zJa?ZeZEDMZf1D_6Ap$J)4?ya$lM&_mY6liO_{&Ad{EF~<FLHk@)L6U#GIyp$oiCK(
z#a_xude6?u#+^RZVPkstA6YiFeeV5!&(3VR%sJRnn{#(A>yb5yApKrd?JEr$61oew
zB31)CuLKzlSJ+rW?e1dywK|uBsY2i2QApe}#V;PO-*Y<|9QUo@ELf}?b@UpsRzU;!
zJ#A@19aXhzJ1s1`OyQw-F0g+YZZnPx?f+NR64c|T;&}WLTxEyI8hW#Lc?ReW>IlBL
zr)!>>fK73M3Uz;p1PVC0_K^=r2OU?a(~!`dE?2iRNHt0i`K}Yh^^&UVMp96U`v+$|
zO)SSD{G&M<eWNmuS~~F5%;2#f<zH!zo2FiDmS*MezbI&mU7^Hs$iD!_xF>1;oM*3=
zc~3{u3L`RB+J<C%^Tr<95Y@TI5S)Pn5ECqR@du10a~SY<M4^;spyMBx$P7K-FbZ(d
z!KcjKJMX1T>IK)Qn`eivOdvR@>uQknW$mHH+z9{a{{la#v&?Plawcb1ar!_UIP#W&
zna?B~%qig2f9MUYrlP4G8zv?x?fOXVeYP#dUKIYx1-bhv+__?HQAl~38@I&=g|qRQ
z1!*J7|E=oy5aX7RQBvC(^CPdf37di)p_V^0XluOW1(icumZbOAXT=X|<hTOP&H%n`
z-l57;M^rbso9o2;q)#5n@&z{ZsJl4*h1FI6JFa{;E?yEIKbrFoecw;3n)6gsjREDg
z#$T_fQimSGT+Ebx|NE3XMx8EdNddLvImXxkOKTKu>-{g=sN{b4na*x%h--Sw0WM4G
zrO4UNPYc~Lz%j%*L{lHa`eV-MIPd*WhSlhNlAaW@^}ry(G^nyQ>m<xS?}HpmKv?Tt
z3Gp<L57)Wi=JGmfU5%L0t|<MTl|L#$Sg?oZ@t>x=;WMP1%PZ}<04S7qBmVm;<)5Z^
zVhq1f4W9CO$xY&siG>S5*piX074Kh{G2oe#mCp?0BL;hAv&fUL9o{Jrkzp8xm<Q%w
zb^KiQz@T)FoX4VkvGj)Ua@(lttSi6o^1-1FM(6z_d99(Rq%uY$>4G&gA|;Sf+M@qn
zFy;{s=P5}pwMvd3_O)%shWi)2VUF<WtK<tnl~T`v@~u3)Ht-?#E{Sv*rKNy;m7AXJ
zX8*tEV`vdQaXpyI=(Okh|8Om}o!HX@afL%&pOhVzW1c2TFd0#$P`}*o$i$yMNUc5E
z$z7tgF&EzfkpwgPwpTs}ZM&j>Iai{kyFE7I-^&cKs$hieq`gv#nDiCTE|aQgo)F`b
z=W=Mgt#}DW1;a!9dL&#wI#_}ai{AdrFdA{e+#@-EsK5KXl)9U(_iej**bNFmKp|V)
zPnU#>KW9-I@VrFRddop}U4npFa{?rdN&wt%%Fn37efty}mY38T!~bokb^T8)Y@icj
zuKi#09}q_EL#nj>kd?Eo-zs+QA^mVX&!~WOqFK=6weXL&;*QlnmS`V#mWMytuCwxd
zk~e#lf$}Q(5KchSk3<RP9zYD5W|ku}7JaR|HG|62Nd0y>+PAtp82cqM-d8g#o|63H
zmuezxIOU=etkS<zNmjjLCyri&Ui;Yo$a9vT{9w6f^zD(IwKdy9>S~gAYLl#+dxoLN
zZ1aqZ#`o`TF(_0F=IC9`@N1iNqTb1A+HVTqIV4*BHYHwoCOzF~b7uVaWT2JdO*{iA
z@L!lT)9gf1nGv>Gw2>~@hkm5o{Mp>BZLjur^Y624Mq&9wX(@3)b7$w>NEUvPPw9ku
zXV>%JKVTz7O}osy#2~Ul!V+igdft-5-<oLWrVo$hAlp^jZl<f_4tT2|si<}vmU7;a
zVRap%Z}$+5y6`kg^HR#qddna4&w*?NQMthS<sT1!L7qI^1)z!?0$HnkW!Dp=pIPkD
zbC;+mC0s$A@oq|Mv}!8+_NesuUE~raAAS7*Y31)1l}j*lj{wd12s8<shS(2F+1QGl
z|B0Y^uz#e>;+DoJh4tk2wcMvenyEQ4agw7BXS~l0jt(7I9`kZ#m=8LrmnV+?x4-5D
zsc#a+QYza%86Y5iyy)d|9I5YS{NLENyH?45`S{XCBTpFDiOXo76>T)NbIu%^mV(L?
z&hg1D{_{<UCkA~xU)oyiLix!DxPm?6++NZh=dopY#s9avk>OWs{3YUXK0IDHyBuSc
zLXqcL|GSrKBkX(DJ7Q+V<`ee)^~k{YVgFS#mG5ANtZ(6`N-?0hDNQT^xAVg_Qhorx
zK<0UyZvq8ioBh!p--~h#(?glt#x?ywzdKF6<m(mL@-HDW$4ydtQUqrQcXnuk4gq;|
zKGeZttf-fzx5Tl#TdQLvK=OqP>7+e$<cDbP|NfX<?Y$Y5{qEpikzyT8(YmuEBCrHz
z-dt%fBfnQT9io4$HvJ+Y0x80tuy5iby}={2lMadKe}h=sfg{ts2cxO8U$%?=Z~3r#
znAqdev`Nwn?)8x-*ZU-Clv2YoDOn~70UNM6ciGHMM-Ox<4bM0CD9=-eUpB+}wqRxr
zMF$d1+=EELet^AHKHQL9{}&-)T@n?7FyyAhF%5$17DXC`gh!U{4a1yfAg-YCwy=ey
z@G`2WO7qL-uA9eJW@9NTSC>7$VpYXa`4qWNt4ZyG1GXySYJUU8y~yd<*mD<sd3dt%
z3H~LzHwe;H)bNKR#49PibRHO7;8E^T3{$1?utS1DWq<IODUy}J$qj9cgv@y7*_pW3
zs|uFP&FFKl$7+D4FZ6k<ID0Y^w^~`<QmPLl>7zV-khjWE#1C{g6v_iry5&^eW&0(9
z7-Z9LDXp+jFxwxuTD0VID!u&MqLAe8IYZA(%8S!bl<MZQ%-?eCRG9v*aAK;zJqAD_
zd!c)HjJ7VDiH&=I2Rnpv#T5@`!K^M5^E`D`m`{J6HmU76P$2ol@%&$Q$X&W1d|d9I
zhPG;1_2nkF;A=(i4#iuUI%4nX$B&-8T>juEci^Ud*AF>&Zu2!>B~9A8EXEuc${ul!
zrb-aUrdc(QIZiweKy2MvLu2jOs=V6ZHGk|F3Y8<5zcn_f<LMR6LGjDG)vwL$+pYNE
z#dJFd+2_;~aZ!0i0~W;;)9Bh8FTS_q1n;#?J-xP@;lzM&9UmgL8NZNI#J9~F|4cPK
zU#`r>liA;(`gKRm8S$voWBk?9)v?<9JD?}jY;7GKE6B%jn>b1Xz;;MrAn-#9(Ns*<
zrr><zflpxni-MVMMM=z}%-<rHDW{sr*rPwU0!2~b*8-lFC%vJo9;4IUR&GU19`9mZ
z!;eNzA_sJinymCa<szS5*+tn70Yn@qnnp}rwPa&S?^UCFjkK?4!pn~Q%xq=ZHRy=;
zWh4&a8okRpt#vvZNKo-E<|@(-!!Eo@_iX3vJN>GQsrU*`-fZ^E)rFbBZS_yKh=Ahd
z!TPYK_2LdUuVnh<{MQGxik9~gM@_amGBj%ew$}b&+JRy_xV;a8(u<m9Mi?jvk*%=z
z`RAeYaH}l((-yCeu%~i<f^fEGYJJw<+d8f9t`b%5)tp)a=fO8ko<WUh3ydq48H_Af
zQi!wD25fgsAxzlC_43WE!~SDGpM0{1eSJSrB8K*;-8++KAr9*QC#b$cyDNlx_C_lp
zC`7PWYatoP3V!p;vBT2Zd^pxeQ}=`OV++Te`{|T|Xw7k9;3(Crkkzw;gce_~L8Qs=
z8c`*mME0G9d0i7SOlUn(5Z&eZiD)S*!PNBywrNqnYU&pVK+Aull+cEWI27xLRNaev
zWvH72(rbY1(Qd$Qv*Vv^FMmIPvi!py@$eF}<=LD6KIx&4d#99H-uE%Opjj86OgKZP
z3m6SGcc;64TJ%`^b(fYzY?F%n@Lf`0yNUMP$nzj+qqqxI*UbNRFUA&cu21OJgWr#S
z#|vmMv4RtGI8ZjomeiCZDhVBL;HZO)IVoP_c&kuOv5>e;utXi#^bAFQLAbEe>ON-Y
z>4BQdS&JnZu-WSeu0xf-_+Gax_ZR$&5jL|uJn-4ry}8t#bN%2XC(oOrHkvZ9`b*gJ
zwYy^8*ofzHaOetto7PU^)5n_Ln@VrL&dz_CXD2~Z9K{21OZ2z;(a1xux1|n#8U&aw
z)kwtl=zXPPAycvA{)Q21paW5_!J_Dnc4gjFGNw(7cSc>MsknrHCAb&*^gsHjQJOYF
z`i!YyN3c9lgZMwf4-`&kH!oz7f=5!&j-hcrLAh{@sfgq6>e`e+?tdjP;XX4x<yg86
zP0lEfZ+!p6tIVJLel~Y2Q|*8RjeZzaX0kWG|9RNSd#AB5utAr$<XKXxbZPI3{nDPx
z7gw^K8AbaI{v16sIclQkT=l6~szx*&*H_lNuiuZL?k?+7&ALeyvhEiSveLVjztTyt
zRu1cdE+1josr|oUs;FO)SstDKnG|_BiLkWIh;K?cj*w(+WVlle#wQ;%MFmd5Mr61K
ztdhsfioP%S`H%!lP_ZEB?Mzv1kB>ob4~$Bh*tfUGwr6D{AL7+p>5>Y#&%3KyPONux
z^De09E>%bYTq`&3yj?e0Wz8B?oJ?NXctf^?Qe2Fsi2FrG38xCbTv;xmjddSM+A#mX
zc~%jF8G|`muogM=PW-ESCSTlp(4+8Lj%8=zK;6wf=2nIa)z<mt)ZCaWwuZ;!5&Qvg
zmhDA4{x}!i`Vus<Qjp`qpJOOp3ghski1liK`T$<s<DY4flv<L9aV1tfYDxPn7#Htn
z|5~;L%KJ0AY|5{WsGe2L;NuHdxF9&C$&1aN`oDMk5)KY<UW6vNN#^I41Y(MwL2%mk
zwW7*rt;TzQCBOV`objYi8AW{jn#3EYSKtu{hn@w0JsENh8hBpdy0BpewVl;<lrWBW
zw$-NP6u@PEC9f-;KQR><=`{`b&^UQx)p>xjRz^JXq9EZs<c*G=!)MVlQYb%cP2%Xb
zsc#@E?6Lo8W;>JaZjGIDS6F`2Q%)#L`&fT?;e8zBx7wZ@d~<X)wq~}sNndEcImZfm
zH{D<rL)e*=e?z?;@o>6^5bVH#&o`7Yj%kvyh}cParmi}fBEPk;91_b@bSF9Q>@9ki
zwm$Bm&@cyE$g>&mAn`TDI;&bPwT&eg!9ezR^)bR6SGGy-;!B6$z^hdrAc$Y~|9S}C
zmU-L(6Hii8(ANf;kvWVTd{`O`$e(Kx<r6ZJ)++;&`Bs|lCLDQ#AZm!5Yo2F>2ZruU
zeHeUfnMAHc=`Xa8WD2(W`))D=W4rL!oQ+G|M}4?J<Vl0CX544P4MoQa_djm0hMK&U
zzdS7d^T1gf`Kx|GW>|gt3ugq0I`MXOHfX_Ryc(ZjN@X<kCQw-P#cMlh)V$WNOZ8Vv
zPwdwP!uj^l`g+17*@Y>^7$ElG_}owbZj6Bf&=GKIqY>t>)YFlxs%;FtXnLS@A(5Xh
z#?^Gs2(HY^b&_>{9kv(ZK^y#5>m&*Hk%Fzii_I?kX_1+S9*;+5@)jqqU{$xfBdGK`
z8)#A_ilF(faY>Qss_D`P#17J60mV{1s%m^1n$-%4CN15RK#?i`w_SKNC||JS`WQnF
zLRO8Ea^7Z1iUZ^pY&T-={UrlENaZaKPjZoxsq1qDW>0R(F{T*~b<3njp(~2sW{Pb4
zs(r5Z&6`4&^+I(2+K?DOrrfCp6rSy`xkwoXk;^x~SDPV;d0{o;G(mTePS<*-k(gj8
zf?`q@;oDV$=1u@VP4R$A<N$gl?3gCzU8O8kOgqzGnRw7L(mSY?2LZ4wYIr(<WbHW$
z;aiLU(CviAqzt4YK=dSzBqTUuW&+V;jKgjR#iXO#Y)139)RK`#(;+ly6foQPeZdr_
z{)a)Q)ET@^|J2U*C4)_?Yo-g8%j}oue6cHUvrEYBO4ng3*#(^{)+bUYL9cTc1s60s
zKWw!teS$Fx-5#ll2iM0MdqZdGlXqNwj7Un=pZ)BxWWYt+H$^Q8zJC=x*n)R)<0xK@
z%%bi6K7YpO+QrV;4yMk*1a;f&&<`N-kbv2qv%9FbGD1KSx8|80Zm`2xCjZZZxH}te
zO1nv8S%Nk8^+zpO7Y#%0raAH7U8nHLT;fe#9YFOPwuBD@%6Eo;5BzX{=kvI9<PA53
z?uLtC_kh=_+F>T|u3RPkzsS6=_Z8I#>aHdy-#vYFCsOZUZ*)y#?061+Rg6jTvf8E4
zllGGatWfKynQiuj)h(8_9+fuWtOh&q$c?j-a^a`&YP8x-4E!TnqZ?E}lGlCE-?9o4
z#qN9RS!KohN7h!|Ia@mIr3`!W3{eA6Kpd8?++K{xK;Hn1tx1}Mo?WU_$Vlx+8Wb`w
ziMf3DI@j%P>jl}xZS4kq*UaIxd)wf5^Wj>5?}0MpKV_24VX$aMta*PltedMfK<n{v
zg6l2yF>_o%Y$Nyi%P*A9b`+*%IZ|Sj1tVF?>>9<ytU~TZTjkTHr*$H25QzEMlr;LN
z*?0ZS{w~99S_R!xAz<4I%ctNeWWbQfRn3q8il-_yQVt%f!jo{wBM_roAOA88Nar9T
z6B&&he2f|b-7&Pg(U}(@f^yoF`t1J!fk1x0iqeckp*HYst=lSRX|`^*TOwgbe*@>t
zGN#tuswX|{Z*a6)N$I;OXRDPdCDi%jXE5sY2iwS*O-rqfc|O?!xZ69mZJbWCdB}~(
zM7D-Z7eaEh(Tx$=?0LvBm>#O&ggIwh#c<fKpf+zN`VV^&e8rK_;i<-z!fPhQ08u_%
za$7UXG-^z$p@ZLk{eq(Wo4@=*?e;mG`0v&~-hH%we;*v~xRX;vzgH*s^Clu_AAlz!
zg3s}Td9k)aX_cG<PfsU$Tj`}SxtX3Ck}5HP@1V`x8uu0mArT2^SJWk={~f|LQ_3ZC
zcDz;SwqHA?$Pgv^ncx@JF62|@^fTa1X(_`~y72Z^=v5HyET_o2-e~*4siXCRPtN|m
zrl>Yv<r~i1ehBZ(cP9iZ;q3C%;{~=o08_r85*b>MR!Kn!&V6LMW3=O8n^|M6#T+ZW
zx#He4Ziho^#|@*;-nr<)Fi}|YY+Ix5o#?0I3pEPueRZ$gEfk8W-^($#L5xPdoYTk4
zou>p?Z`H?!Q%~JGm*-FP*69@jm@v<I+yFO(6pqQ6a;zct#Nd(9*ve)b%A&O(s&?EE
zTa6lKL~R&<KWL>^d@^dZQ$0kJogoL?&6y7u>al$bp7qw*H?e2l$1wmmagrRF>s|0E
zYI7SCR9efSb@Sr+y_hh@i1PuGgZ9T7qT_5qnW9wbNZr`W&hyKOID6{8;<|A13oa|<
zP8bl!F}oX%Hj*jrheHSmfgEO=Q4DL5YM+ojGNiz4b}v)Tylxw(>6sy&CWUy4oO8eh
z!Nq_yc)u0)Yq6xsi7^_tid6GK>ar8F?JDoBa+A&W(G!m90Gu<%Q>&eJv_)|oisww6
zCL@7JXN<y>tko8eMQ$HDg9xn^dauNED5ity$`s5@bFV12X2E-h^p2$CoJGfWZHHTK
zJOY$`mVO9%@3Y6Mc5X{G)|6LInrxAoU9z6+Occ9qw)R`Q*&ogMiS^d#!`vwDt6rrW
zQ!WK_y4Inz(mBlwF*x?p*lR(9iH=7X8@hKT#6S)K=lwAk*6)j+rkUsi4ojL^Z78;8
zc7$W(tw#%4l`mgz2PHmO1;U*TuOAis-2_v;&7sE_xM+8IZ{6;VBfxxk`bA@7MeDUk
zwrUyix~+tqsNMYchqK;vEtTO0U%vcC3Yq9k*fLEg_U&fNUz|m-HtNy#5%zumINUJ@
zmg$1HowD7)cS>!z;SsS7cca_Du(lm1!Q6O@Ef@0YSxzV0-@aQ}Kj#@ps8i;=m}Ps)
zC=C(okBU=7O#vRFt;JNe{UiW6TcA_1nGDXOVmpdM@Py!4*A3@9iXdv%;yze)V!|6e
z)b2JcGbdRk=EyuHKn_dlO7NK!r=z;DXXc$*HQdX{X_`!9a+HkwzOyU~*7LQ1b2?2%
zerT0a3X@NkO))HBI-8f&Zv9y>I|5Y9a>S+{$QUfM+2`-2AgWCBVwd^R7q>gmf==}I
z@;H<fgYn0YfHsca@_jukhk`g~v%<HQlwAUM+iwxIzbhr2$_Hov{`a}xIVaz>1{`)c
zzc0T<?3v3Sa3f)g5!6iR`WQ4BIPY1vLUw^6iXSj`6L8+Pl{wk)cZxF_9NX58f$Tj%
z4^$JaH1@Is2gN)FPRlcUePigL)s<ETt9kIF-Y%RN?E4s1*ls(cIb8R|n3zuSW8Ub+
zk<)|@Yft5DtBmctui$N;xP3?F_)Lhz5D#)rC4@jc+?;LG=6CW&jGmAqP(R!n%0WrJ
z_vObhyYuhKMZ&x9s{~TYBo|Ny+h(ol6eo~z42ti&zV#!Uz9(z8iaVI*hJ3h0SW7AA
z#B!QU$Fi?987#|;OOc!-(R+%s1g7`)5i9{zZFc0GvsN6~I_hr0!-qhevL#m*hYiO)
z5S+(h?Ia&<Hh8Cpv^MbGf;@oeK;8GdjN{Llw#~M6irN4?$3Xd=Q~8H|eTs)i?LfPJ
zRFA*k@A9!u;pnJD;kzBP9DU5>Ote+@ic2PFK|Lu3_C9FMtip531OoH?{X-k8s8Gw!
zzI~tI-*;K>YoYZ@$P;_rX`^#q&ZHEHY3BXi*eX(?P#JYFrO6`Vkwn~?9Syr^a?T9Y
z%Iq$bR*1>g)Qa8@m5G@~Yj7}iO!QPJwJ_y;Xe_Kucf@0UQjdjB5$^+{mLb``l8ziZ
zI(log-%0&dZQI~ouh5ze14D@Prp(ifMrGSJGf-1Un)1iOD+?Cjy2EAr{f|A+nEE_s
z&-wYpx^-CX0>1ZxMyo72QaaSpKAxKnWm+aq(KFhh?1enpVc7P~>>^S!+V7$24!C2W
zcDUd~hjX@GQ$#3-rT12^R#EXi5C-qbKH`r%-CSVZQ4SyNshDV`()Ihy<6bJ`Z2x(!
z_2Xc+)(&r;-|t2_+5y8>k^a%pdpEs@8rux@Oldlj1Iw~U7m~Lf5qsTj-+9Q^_DaFl
zJx71c(#(Qqz1iNykAOP{ad>X8D_xY3C*E&YOKc|3G#^RFR&j7x-wY6#^MrT{cD)Z`
z2+Z?DFM<y?k5L!s&8XT#hP9AO`Di!#U_k<D|7u=UYsIu186Oj^baNP1r8NPqpaF9p
z4o4PjTPBrzQeu^=IUt1ZEttcF0c_@DgK<anp5@p=x^Fwus<oEkWAW7KbfWGRueO8e
zUICe}nz0s7mkCjaUw2A%#3|xh$8juc%aPX2e9~4^H(YRBE*DgdK~}cPa+&~T-FN&^
zVMrf{ijjZx9(?TIwq;hvdaKoFx4ky%*#8wX@qW&{_c$>o;yo|~9}pZqCL|78ZS?V-
z`hI6(fwAcB$=khEj&mO~mXF0?$F??yBer6Wk#by0%d!xnaiUQ{WiSQXi09NZ#f3mX
zn(dL8qi4#A+ggty<{$*yJ#69`{Vs#=Lca}gffN>!kIa)YpC(dFZ0m|NGwTb%_A-Z<
z&0`1Udb6zT{q2=y`hH;UJ$9RAbdt~D`mt|X>0<4Xu4=v4qnCFNKIZ6ZW`-#yYPY_9
z?{bvnW?5jD*zdi3^mu`7O{V^iz3{=MR$CVwV>U`|7xuDRV4N}mM@k+aBK>$)_b=aj
zPe|WqBKLk}NP&<8Tia=BEuIh)>Y`=y!+7g`U{)tb-Xg<z->sR*OtFe2#>CKO^uBlu
z&Ad1FRIz~iLFL0}(dJk6ebb|DY!hoY#qXKpG)-u?to!$r!FwikZ%=V}4CxU-?t5E8
z2#ka6_WR#Q?R~wM)%VN99c{2b8fvRVm#qyE6M3E(qDRGM&nCNsVta9x6o@V|n$T3Z
zJf8^O*hA&W=tD4e+wVnQ<}BUlWf0PAUSm@rX2)_~jH<g8`ql{{lH5es#xiB{JezjI
zn-A8uZ7kDbtfAKN(PoD!O}K3%#=y{yIyn}++4!jq`51nugN8p%6Fyj0ZbC!ThB$0)
z(e92H>W)m_#2X@N!ZtN46!rJtm+$s9VpX~KfhnJivb$Hd>rT$`aJ6>FpZAR4-AMp3
zMS|M1+xPwBOe=~{p7-r)?P_P|ao`z8OE@?)ZFP5pQavm|hESWCN{Wty5$EWmLOd#i
z)(Osh@I-8O>)df#RT-^+oV78=hl%A7F;m9loP`j`DIZKUOXlv}M5FgA!#jM85InWj
zW7FsNmX!D2lE<T+bx+>B0>85g`Z&MGI~Q$GqQ<MnRzTVh5*X6)hq+)DTQKjuwRLiu
zX{E8QI~wap`8Gt34PoY{+iE>TDo2LcIzB9n81Vt>ekD!Drs%cfI-m~MkE|-%%OLLl
z&;pms6DrE}dIN#;`Gj+UTFvY&PXR?(P7AknwTC_wA8h68zO5ev-sxaFscWcOYb*GO
z$k9_)o0V7u)?@yD*e84b)sQ!)^pTBu?~i9?l~V`_?;^Dqyi3SvhicV!JTN>y&32jJ
zIoOAfl^~>)iP5rv(L3Tg+ggu7T4sy{-r<JFOQ1W)sKyezS4}?LX?h&PMx!r#4xu4{
zcf^#8FFpoN^t;dqWBT0x+x|NW;9bP$kHM+6y(}B+rE>J=4jT-2PqK`L*asT05}NTb
zce-DQDJ3i4-S@wHbkh{vU+Q}n^+(izfDs6sGn!}X%`!b7kz)_Hh|)%fVdk7McnE<U
zC+h@+mDcOgwlX%4^T*}q!}qN4$PiT>P6POiFzC)U)kznm4cs`4KRpBj3^QiDyCW=%
z_1Z@luGcF$CQj+X5c6|$&VvYP%AmG`XNZ2ZKUF%d)?>ehwPc+WoD*Bg>odmD0}hs#
z_qKi^MrZAC<uE1mW;>*|)|%sI3_hUFl0z2)XgH(@iik}kMwesOC)Bz^NW^Tql)H72
zJI;5E(U!|i6TLUYK^>L8n}MWr5y2`0>46}Djzd~tbR+T%F)fTGH%F%L9U83VJBAsg
z`e1uCr1uYZU5<}>4<zI8psK%T7m5&UxAe!w4^Ya^sO^};+TY=iL<{eaJ8(cWd!nG<
z&&r)*9gei_p167MkDxR@&e%r)@%H`4yT{Dv{rD0rOK+_qDoj3^(OD21+=!$0iiE?s
zus6!Sk&^ZPMFyqX{2>IfHcq!~T`?u3>XC6(1fh#9OY)AxoUQVFegTvw9Tz-4Sjl<0
zn2qv0&&aS<mgR7RowBu@Vz3^g^N!?g1*Ki!c8O0pBGNF^L2GA;m^mEMm{OwaI9grS
zvozkeWs1>TMjS&pUIH^~9J|d^R6+0zfjlQfZFhHV1sUa|cNvnBW=Ac`Xu$@I@kDC{
zL~vdRPU&N}LG-BLB3@0*1lEhKMX5#5;qk#i_uvAh+a|%gn$&xa%Uy{nB3e-yq&y+r
zL{!ohWgLTQ+xY0cDZJys%DBHk!=^zTI{@*~S_(0n37{Mord3YX4!PT<sPQyS;OL-z
zsLA7pgXu?EErND4RJ@nh_WR$1EP8wf9L;u{@<$6JL#*=ctsQLvMYNf{|D4c%RNgU~
zE=OUsV*0CK)j%E9TAwJRiS6wP?if^?$SdbWRNKqbMj`uXVpng}Z558?Oz`sb#9lUB
zu*_7Cc2J%Yy;i1ara6l;Lv7F<yjiMy7pSc><#4=PdJNQISu!%rHQNH)y#fX~F2$70
zdR@lA`=duM<6|B2k%n8+6N0TM=tA%N(O&Y@W&*vV&wvBAvL1FNN=QP?o)}>-igXj)
zIu~fQ5Iuxsvw<PibW*(#f<LP1M#_QUGQC4BHotQyP7<AaRqil~vf_jLs5X3ZwsXkY
zTJ6!4A>WM)(~(>XW?yr^xPRY5Ybv;4n*f6kHV?eFc#kfKb5Oe(+6I5OR05o19J-Ui
z-TucavB1AJO6<KKE5qe`_v0{!v;cwyuVc<N)L3-U$9P0W<0W?yqo<}!P}}8j$|qv*
z2ldWS>i$8YlmQ(j1k+=T-YuJQwiHQ^%EaMJ*&Q89qt-%*fjOPn)*EH-T%Mkf&0H06
zYR;nA#XP!k#OuCWH8IclkWpv!MAZb&yk<i5)UuI+JM86matf^Ll{{sFi){W_n$nb@
z#&~dEaKZF1;ZTah(Rcs2j6Ho=rB!MJJ%l_crFO(cT(HbcotkxhT8_m{8_o--^K9!d
zW2U|}P_u;|U5E}CN=*Jx|KImn+n$gvjw-A5Gpm4_w!^9M_uJ^;l_?uNSfs;2ZFEP5
zSBJF+gyW??7$kZODOJt<5RPo)`2W_LJAN<NdSB<Qq8RrU(!KSiijVR8wm7ka6JxZ&
zt6Gru$1<pUxqTmu--A^+0>THiP>jxLLqZGc?*8XpVEh4FSn*J3yP&quTrITE^GvUm
zppIoe8<j`>N5!(2?Sm4Hcv>CD!FKe9Ikl!^P>an6R54n9fW5o}g44vD7QTM{isHGv
zJW+dNc-u7+4(D924i15;UH+c)(*+e1zqzAjs<>mvx6nspnd8TBwYNg=jpej3O%v<7
z)4PHwfk=)CrQyQBsT`z|JG0Q<33tpNMD5*d?YKX@QGHwQs1G<FQEhmqX7uQ6gW$gJ
zq#RgI_PHXCwRPfQsdFb<c)eRebwY5Q&I>UfJIRN22U@LWmMR{^n1ni%GAcfskXRWX
zt*U9ts52wfE*!~}JMLCTI95b=G3}iX@q0_we)xlc0xks8MNQN9`G~c2KDrAIN6MzP
zvwQse9(oO)Cf&LNM@8=x+m?FFQSK`xQb-KO$7sO176CPu3?W7n9y{#25hYGbvZ&*L
z;LHux8I>&gL>1d6W;29)#{Rwk`1_uHJ?N$(gu{aSd)C;xsc@f8Pn5P(>xK)M7k~HM
zZWZTjxk}DQZ}Qlok;BA(u(DE$^wEyU8AvIRQ(&|X{@B5x1##vjlT!K^K;H+ky)~Mc
zt|i4}Ewf=6VDEQUR5I4p5sgzyA1x)zCIyCtcKi9Q&+mOCrU=LH1sAZDkO0>HOL;an
zX1Bndr-1Z9OdM-m57m6_D3%q@^Md8rfi^s7LZp2R<;@1fETqRU-OZGfT@JzceZ4hK
zmxUqS=(fjyj~q7W!C5xy?9w(C+dx|Xpv;D7%o^`3YfSN|IL0BY?8Zj9&s|gp>VC&D
zKUgZ!JIH1QbI-i~^PcRzJ!Zja{Hu1MsuHpV_q%G{yQ7bM)Lnlhq!g&b6x%u!#atm%
zvc+Mev=1H2o%%k0*E;yUW&Qns+gTVN)f&KFc9KuzY3BX*hNxBJ^SrRFje593?^`wH
z_%W<?*o9MT!)qdE%SvSoTeNmCl1k|i0{eC&L^EQ|(?kfK+im@*5bwl?-VV2hL(k*|
z_RZQ_5I#&u?`@QO58nB^hZHazP7aHWZtECAjQE&v{_d<`=lEW^q!h{7A8xqzj-|=t
z{V+1ju*9Xcj`OD8HhYpt@bS>R9D-qvHN9;h+KJwX9lbY34EPW~K7SM23zFw-+AF=A
zV_GEi&Y{sA_ktt3g9+C9!A3lijWMj9?%bjCir?P{Y>&#FFd}N=RE|6M_ifx515VtB
z7%&90e)7l9bbqhzZDHm69QKZo6XURpy0^m3`bLONh8fi1KDO80mlF*J(>!r|znUh(
zn~L&2w4CSpJM-mtLX+0MGcfKC)*XGn;gmWW>L7*4X`1=++gC)vd9xX>t(k96tJG5P
zM53RMjLTHgwHgQAc_B>`&Ik4~$V(!5P-iTYJMHg$J|7p5Da1oaA1t8W&0^U(f3Tne
z_Z6x4-f~LDjsXr;ylo2{T~P0lW7S_s{?I%$JQgIXc1Ez?<uuz%+v@mG|DMl_34Q3+
zy7H0Jyl`7@<ZOG_*V_gJ#!zDR*5h@ynX%O!R48>bn+fA-VZKs<5NFo1(xtO)J4z>K
zV}WS57u$&$-QnH$Sxf7OqpG*o)SbC;tS5Eq4nYw%15x{z{UKD9gK^Z4pfS($$NN!h
z#f88$-QAHbP}IXm>GwgUSycVVa6<^kAa&3=4kdi_n*)x1KEuIK`XiueN4>RW2S<{2
zRAy635QMkaS6lApzANi<&WA_a_u1P0oRv~O-pTuOal_h(Wi-0jzLt5KU?{KGD;&%~
z==czsrit6_#^}mCUx<F9?1PZ>Ag5S%pK`>fNK{Yj!hFj3I2c&(zctHxPp8wty4oqF
zlH!EwLBH*Sa}i@pnb9HKckHweCZ`ROo!J#(bCA((4fGhtED!tsy+1T6T~V=RNHIm6
zzccs1sI<z{(?qLswEPBBo^fjP6jNjiy=^#Wx{)}W<}0U}R_&RMVNA~OLFqrp2q_U$
zq^<9ECvF4hEQwP&hSDM+4uZG+A$MxGn%Qhf;p6hYGxnsL&{o7c5h}+y6Yy$t6^2-F
zJ1E2F^ZDc6>%Eh5VxH%ZWSE0!@xiRZs>(jbvA6!H{EtGwhVpi{#-Sy#WHJ!JdaSKC
zn}K);IoKkn!xPEqew{eYmK7e^<^6KK<73O7-1l6J@8vh}{P`1aWkYggw86{s#oQ|o
z{&Sv!ZM;hk?+4LMxN*E}A>s%OKd21jr85qipHdHoLX6ScLou?xwGX-f{Xq8K4z9FW
zn6%y*P0Suy6ca!?L=FuC->ow4<(@zGB`anQ*{f+4M8}bdA9}B$ADa$)KQ@OsJz71s
z?>vNrIp_+l7RD%a74~AZPVePV+In+A&6EAB&QsckbHeF#Mn-47Zinr=lDsWrx(9qc
z^c6ALr5)U1lBfC+3~bLh?DcmC**m*Pjy9xnXgO5B+d=;hcNU2r{ZbbK{n(#;_aExL
zGgJx33$0iH`n~e}egMpS$Kmw5iDK_Y?%+sDq?_vbJ`1p0<Bt<Oz>;H}X{5p-pv^hr
zIu;1zJ1ZvUaI|af;=MEQ;QN5{gFJ9J6Q^W*@8@Y^e!B3wUePJib#Qul;@h{c<`Jcx
zWsby*C15H%EoauX(D!bR1`ev)9ujNenyzSwP+9|TtCUYq6SeJZ@89Y4cg=u^lCxR;
zk6~^RPQ49u2Lzvv703?h?Z|}nBXM&_D~;pVUJg^#n5@m}Y>&s?`i9|=(Xp*_<D=j2
zh!&_^x0ab7^x6<<hq<DyRgiY94j$I3n<o=!dN)x@A;rk$k}2CxE2j4nXY(78f8F-c
zfo(L5-yNdGB;p{@7zEVZybrdS52imG=s_~DKjV*%$?v_W?m@sEcWv~6fhqX!&;R~e
zx>~>mOYU?WNno;W)4c$2M-ui?_4EClxg)W(XYD^I#;QWRm7DqGL|e}C{PKkNf!nto
z9fIQMhiO^sm8bKCQYu647n=={V~otp%z9f<rysh5QS3!u=83dS<WnTeiP?MT%{c1G
zn*pKN&@W9^DM&Yq5IwTekOJbuv6Ezg6Sl43Tt!AOBS<go0o<|b)x>}1dv(wFQizBB
zi952FM%4VBY2dx>N4Bew7{gJ%7sfEH$?=U~YIEZ#*~10rzL%1!?)Wty?Y!QM0i|P*
zvqKISTIbC45w+lK9B*qOj&A)=8ydThmef2?W{pyJVoKyA!@tiB<miis7-7^8mejGO
z>0@YS6${5gr=#EL95F;{DU4c=p`9fV;W!lis5b9G_Ff492i@@g6m4iaocsQ4<oke5
zkD1B+lKc=Wje%49w}ST{14R4TAWsvWii+HItcE{U{9r39@o;<y!SVe0lbMf}<Fkv2
z!9-vpGtU#H)DQAN>&DoNF%V<;Xcwv2_j~uDdU}2)o+sv0!1cjpO59#w5tNhybfJx%
zInDSu@Dgd=F6n#n<D;E3hbhW^fa~>&cZtzqXvO)+wmaIWMz%Ptd`h)#q$+Y$^bPOK
zdS-MX#AJzxSexA9s=a$DS<BfNT}EGZ$3#7NW9SK3vT)ons8vd17~P(u8nSoG?hjN*
z*)e)Cn-!-<sqGEKAFUnF_}!WSRcjcJHW=M1iPkG^bn=vsmc9+PmBVQPYd>{qwl?}O
znbTqOg8N|Hx%)Xf+DHd^=}yDN*}W3ncQNf<bnBcm?=JoRvX9=6b7sxkT1;~T$LiTW
zj*3lw*B1Uh|F|b7>KsysnW1`bi@c86%rs5(+K#{l_X8MHin!pIPnq5-fq~P3b2rq(
za(<$#S*N$rsI6EwrM7#zwT3bX&YM6m9oxU6(G%^H!Rd15>-8J;#rAINP~N}15)Y=v
zIF_pHLrjqAAuOIbWTYAeSRAyv;atW!%vOO&Yl9#W-<2g#?0cg$J5VaZ>2$KJYBZd<
zk3o_RW<BD~f!E41XMOvFTIihOe7n!BEP#Y!=7!Um(FFCDv5B*C+R4%C3jVkl%;tHt
zqcuaua}2mu@hadQ(=?Og#P+@!kKP9}u1lHM_5ENdDX(9@pwfxiTyg8@c4)fLT(_RO
znsvHpr)kB>F+6LH96Z5CS|2n~MzoALc;FcD24{3l@q_3c?YN*-ziTb-S@7=#O}}SP
z!M!CdA2#dG`Hu?CVVULqz>4yp9b)~cXgqdV9?fejG{!)@`SRTVJ)2ei>#FFF0H17m
z)A@2@U2mkIG+Du0%XFSjyuI##RhB-Sj%v!$hokp_55mxa6SKq_Vi}?LmRRJ|1OdKo
z@91UbGEZ!`jqP?LIdfC+5i{}}gK0{1H%RA&NfOcvLmOygl!E#hr=E0dyOiOH1G0Lu
zX0F%D@SavyQjFwjVjNl^KkPaS#}Yc90%bHKg?O+FPL9hLf>UtiV@B51Y)R&6VU9Cp
ztG3oa@0TOEey5$<GnNkE=}l}st{m0#FwUS0ii?5MX)^WxtsHD&Tb!l}eK#Z5)+(>B
zSG+e33u+GAtv1F%LC-On-fha%zB8oTy&+<2Bc_BSu=h>}MhYm7<i>~2!l_cWofIO0
zdlhBMb#?xuDzKmPdkgd4p8uFF9u>)*N)Nv?M#SO84w%zJr*<b$|L&aK$w~|h@Giu|
z!|<qHj&r6VkepR8@w{A^W?Q;)E0x>Z8&l4xI6}yzIOAN!9}APlktuiW<dAUVpnM)J
zhkM_R;0V!wum^;}_4N(iSAPCCe}Uf31=mAJi9AJ0+mPtEH8W3(>RCc&IW2s9do{j$
zt4y=4)m(2Ir_+KDfq9x)$Gg!PyPzs8(}~{9HaU75ey1GCA$>UUh7h=1615ByPaa>?
zoJ?0Sx~;do>o9O09i1VaJ}SK(!}Dl+$<gmjqWjRwi?w}PuOIUu8)`+{!GaKTfNoaO
zB4)hShbh#XYm8xA7-N9*>CC#`aC&5Bec0fSgOZI9`OaYCQ1GiL!3Cx~(`!4_#Fi<>
z6wuy|wvqLn{gKt6d<c*9|H0g!2kVxlXMNyrO?&vJGoN$kA@k-?nKhSGs?ty@VM`RF
zBoIb`EU?^w4Iu{|4!6UNahMJ{+NRy@aD;IOgY9T=6KZIXgrTv7C?u6sr6H?ISy`3y
z%{%5D&isvgSX2M8_V?YB3AEeYL6AElBkwu)4EKCzuf5j$zR&Y}VkE(H<Emmfdy4tz
zT~7w;aW%YKmi(@hOP~4R)$ic!4K9UT_4~bZxHAjq?Dc-U`v>J(8w@Z;Lrey;#dwnD
zW>J!5B}!#TrMrr|L$Zk2amRETBqLVyMr$nrhzj!DtWJrGYECtGs5)nG{ye8mO~4bh
zA}b1{7PO(J6<r@(Du|Y{>|-03s0A@8`N~e~U(FYkMTzhVp}@6GGC}$PMlk3NQpFs_
zid<cmC4^M@FZ=qLE>f$^F6G#1F6KpQ@d_!(GP#1HRxYix);4J^gpd*uWxA(QOSC(b
zbtoX`U2^p2Ng)=4M=C*GTk;|?zCujcbVwU>j1WSS^E(S|(@^9^T4(clnw26_w?QUG
z6rCzGc1U$8C2f<a)xigHlc!9NCgE9p=%NM$1lv)kJMqFYD~GNXwo<bZlGMGs$KSJI
zyzhE4IA3~d5~6D90sCjx#rxLTvNAbqux#6QCD>RV5Yx9f!_g}9tW(8EvXv|qf>$KE
zW72o~kf#6EwwNqW8s}h9M&P?=z_ORNJfSKnu(o9|7%-d7FflR^hOBWcY{kab8k5-s
zU1(bGV9CT$0-+?fwz$OvLqTuY$3~0F4M9YtNDq?J!y`7<Hq($lGuW7nO5S^1NIP-t
zPRni6f@ZZaV~n&-jgTqaO3Ei?CL*LcgR^9rS!JF_JW7*dEL}(bG?J(VjcXvX+9YC#
zgm#Hefvi(4rCA4fZ|STNY1W`5QiAs_na<Nq9<%iMcx+ptinwy!S9MKuIzbvu?4l@*
z$<oZKE8}AX8B*KC3zftIL39eRbdpJ~TZl@gpR*eyEPL$TeEU5&x8UqeyS%TS+}z8L
z<$a&Gk_eyq+|}D++3b4nwPak0_++qtw){SeGg;zzK4@3g<&-F;(Or6C2!h}h-hnYG
z6*7c~kBM)lv?02Hk7+RMZG+-TZ`v=h6TzpDA_1*!$dPQw95<PfDTQ1cP}>@n>m&;<
z3_?q!QHU65s)l-DDT_YQC-a;sa$0A}jA3#(#y6IY;T8bi=mdmZO8<SDnFbrvfy!Ch
zpu0*i%`ihq?H;KTDFIludm*jaLZvv~Wp$Ee8lyC!O%hun6;Xv1Tr{;!Lz;m>nkB@P
z8Ly(C_7<YTx`;MlTQH(?q{lR;NhVbwm^AzJE#3#b4j3)bCSx+5A&lz0P7-9ozPQ$|
zsJ(cBiYf_b&?r=g#HX9>qzW_wNZD;5=^2rpPb(btleN>Wv$JdO?7DjY_5O9Xyv`~Y
z{}<O;eBuFF{&{5<J~N|C-y@)9mW*??vqUY~Sl?kjpA#2}Ith*}Pe@~B(gJ8~gHmQC
zmrNBvstT6;e`C_X7$VNM_+S~02CVmn9Gx67YZsJjeH@;m+n|g#)QcLOWtcLZcA|#h
z0{#Aw)A=bVN!2v04~7_}Xc~)MG^n5vM&n6|V$QbEx!GeUEmlI|T91u>wX238Qw?oA
zd0|#B96<Cg5raKcm8B>RQY7%7R7)Lp8gM6rV4AnA<aQlnXo)_NQevAnJsf1#sZ$^X
zg>w)@@-&pa0d3WyRF>p>Ar0${k+@J}qUgS#qOmPT>4Y+fj()#~Z7i+r`lLZ(RF*L7
zL87I`ww|^ltSX}sGRYp>s?(K=rDI)o=B+2d!sR8;nDySjA)LL={@qy^+fq8XoSn#S
zzTw+8X~_OL8_@D5Kl?j+4-Bq5=?cL0`9RG<gwY9X=q<&dM}N@c@ZdPLOGO&)8k1zW
zS(yYKt#4Oe3oRuCMOl`qnr)LoGuK%f;>t+fEYL=>IanuCibwm82nI|qBZx>Qp~#X%
z)ukH_T}r(4M3uxeMOM%*TAUA9>lhA(hzPc-ai&QMqCClsQ%tQRT9GOl6}$Fns@j8J
z;`S5T$j9U=FMIiu3KI{)r@W5Zqf{g>G=zji5+Z<DqJdI(;jD-tiFLXe!16v(g2q+J
zBcycd3WOBR;9N{9L{kuAdL6YiT~vbW&?62XYJ|_b3P)mFM`<#Qk)WX8?=y81p!0bk
z$#jMf$v7ywHR@bQZ>wbVRZ3FXCtSC5@aYtf@ky&{c}P1er#-uYh%-O$*)|kBSrg0G
zXzABmt~Vbe-r0Zd*ImB7tUOljE%}6*wc>2)hz=@6t-25?`+b_|nA@7X%sQ=@rOcC@
zt7($7Hxg-Gi&h2#s7|xKJOu*X<UV%onug)9&(6*ci+P2Wf&SV6FH>J_F-h6wu6Aj-
zD*{fm^hX7ab<~Rr)oOI6u|#}K>M5gj3QdZnoRSP49UejS7(t45?aKa?5mF@Fi4QED
z(vW(bS^nffv}(u5G}!cBVLC62P=Y8FO<ke78BdCm6swzf`HM^PU;(I=KPFXav7_>)
zSWIHa>yH*;B+>d#HYka)NoBimm^>-RoR{QTiFJ!)G*aoLr?qA}o#LHmeK1T)5et%t
zYul9H9@6kI1`-ii24eKIZN2*aiB1&~16};(dv7iw(r|A1Hl2L;>`i>OG=BfJe4Urq
z*}>jYF|y)Xow;exe!a8)qvh9JsYzn|Ky4n7`rl1!kw%l{nrT~Oqod#N<6P>yBSez;
zdQU(k(T+<V2kZQ*`Y^`O(sr*&=YAHE5FDGE8?24iI6OQ;l^I!{6go{)p@K~tQrmW>
zHA&Es-mu54UNBqC=nZ@5+~Av}`}f{ql)^R*3PDzuoVtqB=>#pnNJ*KON%hqEGx97Y
zg+}35Ct;;DS*B8GR(HT!ic7rc9(ExCTcw0dYuaFG8jBVRL?DKw3gpba=e<XCLy6@Y
zQ!25l2uYD-bb%CtHn;?&6bdc6Xa}3<##%U}3N$7VwV-k>xhjcSnLv$7P@|Aq;H_hN
zdV)Z(Ia;TU;5(yrY<(I&+mw-#IvdG!Yh43Ep`^rh4`r=VGHLRDk>($(gOmJcHiP$H
zXE%}eUr90h<YUe%td@tcROmeE3cP>ap4~i_ZRgdd7D3y_feOUnsGA8wdp7q92ECj?
zzoc5!G*yM}8HCQ!rpIF1aQtYDj4Jgo+ZyK@5HhtxBvg&XdC%6?8uR&tjg2ANMCQ!|
z)dR1R_BzDWL-pPhon<g6F-1;eTPCwP(F<a(up**$f$|_UEUE=hUAn-fy=}(hIai;0
ziq~I%jlIE$Su+PM5F)XuV)SHrlG(->x_)QCdq<w-2$6KeS!P!1r_?4$WxkIA>l;L^
zP+H(cOr<HyhbNr@Pjrw$fl(QdURNR0JV>deObSjSMAr``{hI^=+Cb~tL^;+vrJ1P$
z>jPe<8-IPUPLb^*#18YxH9`>Bu)y=HRLvakEvT4Il%&MQ7#uM;GA$9o;o1~u=YvnC
zBrM7#=8noF8ifd+#)Zy*Av$D9SKbP_Qe>?v0U^8L-}_{}%L+k=)rP%PYArDet4$#$
zgJFcu?A3)1;kqVXMii2OBYKC7dR5}9PR1KT!gGJnF4x*5VsI4!kIn=;m-}q24Ui0}
zstQ?qinSgp>jQ?#;gs>wg3=U79r100bpk~433yx6R!LDb$`!s@Fj_0IuA&JQsvki+
zjF9w&VbJSQH4S+$!{i2;85Yd~JD=fY8uW*iyj&1npsMD)_~LUM?eDX(c@7~B)4R7h
zH(X=d%+Xm!2&qkFi~`{SM-tVgcv}Eta$;n4NLfm{y$`EaTCk4BrUfn(xGu%ACADIc
z`YN@Dy=@aBL@6Ta^qz1@>7*YT4?(BUGXhq?_15ySxhIAO-J#;^PQ|IU0hy2{Mk$mu
zWI|v}wxS!<O_SmdyvO+@x*5q7#~Xd5ah@+Ka5^K{25%k0Br`8AwI&4-E0TFwMxgV*
zq^y`HicKZdO6j$1b0u|FnoGUET`Yu189s<aVi3u4BDziV?8nm0jJjhVo({+=JrRY+
z5QV@8o0MRPB*KyD;3#4z=??*o^b7_C!eq2=#_ok(%AUc?Wa3j&FdPoaO_9p$te|x@
zYny$>2MfHDNEOif$%8m{KlG?SpfEYQKVoCF&TKZPsT)*Zp>v7#jyx+s3fk7vMu(2+
z^s;nNs!U;tC?o0jO6q0-1hf)tY;2-|C=JSJwl>z;Twmki$vzK`_o?bGPBKfXtr(Zc
zb7*~P0Yy50NYyf%J;_O4hFt{#s3p->(zYQPb4->V3>{9uHV&3Eg)W@Qb!{yrQ|v1~
zP%1_2*gnB06K4>C&?J<1h!IQLytGNFYTjCO9_jaT3h~{%d-G4zwgtA`fS9;Fm#+Rd
z=<@F|ZA|ytwrHivbxt%6?|~?B4t$72V-Y%4iDg-!GRb_}G8&HP$qXD!7Y&p7JZ*qc
zJ(2q@Z@n`o79V3mN&j<NS&4nM?7BvTk}K86*-|{2t$5Oo5=j~zMe^mOKsvn|#-H7+
z1Zl{geCY5&8>MS4=nV(>s6kowE^c9RLE{#98&FEHv%Qndc3M%*EdyWD)|PfrQ5b{N
zV01ck(JA6&gw2gDB96VST{4+*IyuBSM={8;SYjl3G6>jU3EiOpms+&0bx4ys)>)y6
zK~OgfluRj~+dDf1=g4~{B1$go>@r#(^2WWJOzR3?cX>Zu<ZSYU7@FFxelOl7H27pP
zSydoPsWjUONifhA8iWqqYO(@UXGjgnWbEorN*X{b(G7G{8(jxb9%8Tz@{+j?Y2N8Q
zwG@2+tmZ#E{9jX=9{n$W$noqwZ+_?2Chr6RE@BBtoeZQX==J+uoUCSju#WLEwbW*F
zs=DeD6+3@V_w_uaQ((8b_?SFcN#OG2dhZ&e^OhuFWBRQ;X-zHLW9gtKJA*jsS>)5b
zk)DkqZS2c=N~ii0f=;m&(D_JVMNO7T_RjB;m5Mf|9(wB{T4yY(hNF{11d@K4oWCa5
zpd*$R)1e4Nv>0X3LL#MOtx(9|SYO*<F=?318l02(CZwKw?Qjlin-T@}61mYhwAAD}
zMW`wPO+8~{Z9rK_ni$yH-X$;lynE+e4)zZi^ha#(Z1MV=ud%VaNttH|pCpoPTjP8>
z0n6@%MJxxwk;P(;Hr>W|rj4acI`L{{kV0Z~hFE%E+T<rGRgSY3LmGf9l8PiZhM1>x
zwxk~NNR=L#+Iyz+Dzz!v)OWSsr<?&HR;?JN1X88Q)1uIX`cu4h`_E#VO_cN~)uwPE
z8#&qkF)m$umB#%xd7e{MZ92(lfi?=!reR`?lBhCN-{PEQv_7IY=+WzExV3<vMVcuT
zCTDLuXKk2sxPM4%>7?7sa|^2;?URiq^^4E;X*nYqgeP;5rRgc%>+e-urF->^p$Ux_
zX*eoF()>pxfoAnJBXn}j^sxD$UAlyp1-((u`sN0xKpi|)>mdq+&N-gWna?VQgFeBg
zSybN3s1`GH_qy7pxm5@>f;@vkCa_L1Do30iPH7ea=LBsN2ufhvkWxU4fY9lI5u67v
z$ch9~w9XMjo7UOJ233^~2v@IPXFQp6a#}II`<U<g>}ME{kGOUJHlwv6t!uF+RThhC
zL6)UR0n=ed5DByHgQH!vDP=S)+m1;^C5=G{l@3i>A%(_w2ZqqO@lB`vlmdw9`H^db
z(S{Iw>U_k|5wJ-uEYniA4MBK>jl`G^xPN1F$k#JX#<C@w&fz*UXlV!whpR2Q2=q+G
zpzL8X!^vb!HXJf(TV~BXDN{X2UGfLVGd?w+6NRR2BmJN;1;{*6$_K*%!?j@o&dv?Z
ztbx{1<OSQ?116I(<I_asNLH^jL_gbFS_*$eOwhyzR_ibwnx3qu)#rFe=r)`8%o5ME
zej!PjTM}EWkQvcOd`Qf)NSc8z+g{5}$cKn-ni*SrTjWK?Y(8gkG(%+yZ4&x?KC5tz
zM+gI*dC+=C;~eFBNv;dpSwlT@<jSC=q|ix*7oDf78>*_o#gr3}P#OA3vRp_*K2+#n
zf=M+Ltw3r8!7=O&aIK@7&Da<YIkA?jm#;v8yw}4zOIugCTv7RkZ7t~)ImBe(L1djH
zHlVVUj~ru?%LeZ#$|9AAT^};6<a7yZpJG@OKPLoG0&<-t(`y^Z(G*3F2uU`ZG~G)B
zrb?KQD2ZKLtfzDGVUeWG)vUv_^l)8;v=k$>ZNvIT%0iL8;KALWplvo8l!81fF)Fcx
znzms&H@Nx=!{VFR7)TxykGBbvA3FE0$CDL`tWY$QIa)~UV$Ni-$N0ey<7-D#Uqhnk
z^=_d|pVGWUaDSIJ0*Isv?#kt5k5qN$M17`bn_!3G$z2Xl^zYs0zke9MyjGiw6sRcJ
z9FFKqLsPZPZJ=^XpU#t$)ha9GgLp~D#>N^dQ|#{_Vq;5DCf9da^qI~kL?1CGRUNs>
zIIU)&HG{1_r74*n&vA{Tm*>csJWpDBnyRL5Tk57w`PUIlW;!Wdx+hhx5v4^)iEF!g
zLXZ$N(FHP<ljk|MX^GCV-XGAmj*AyA@ZkO<Oj%GreuQ%^(|Sf5TDHm!rbeMvhh)j}
z&Ro0<0ZaImwyiPU^CWdp#H#XFGR-Tb)`@o|bi$S(5DIdgLmQAlp>pypC;CLg&Xk5u
zwk~3aPj{Z!nT6N#oMF*V82o02>W=ll(9ijwHii@-vZxkl4ZXfbputJ1dcy9`7N36M
z1{b%t>F?|!%aV6~`)hpj?tM<gh|^l(WT0(4G8^(j-SX-AoSMKhzTu}&$7m4~Rw+ib
zo)LM}xzani`Po4_0_&cTi~m1e7C4lW9~lif?L;Hdxz(qhJ&36@hqL#4wch{!OlWpD
zHlcAetz*_&Y<O}QJnQjVt~0Fy*hND%nd5?^zcwT<3L018Dvxk7p~?e{MtwqbNJ%Qo
zG{{_{TGQp1Cj(8UG(u+hAgOHwS)?452=9q5u(mnG$z;#>m80CqsBFbz(%@{0uWg+N
z0p1Fxb%h84Zvvr-l-ZDtjZMDs$~S1+hN7nvfzlYzaCmgU=4iyp$q7YS5<-f^R4VI;
z!XA^Q9%DKUrP3|XmDRe-OOAo8$T}5=fGASXRY?8UT%|Nd6-XOR?A+BZVFk1@sYHz_
z&pDAY)8J8~5IQ3t80xmd>Bz(_R=we-X%aK5Npk{)!L^z}-_R@4rnH=~NCA1Kn2k>;
z*3v_?EHjKWxOqtZXyI_w$U-pOO0BPqjCR_fg`=8&Cx;Jyh)g>~B1AGCe*|j<{r)uu
z{eO&e6HuwNCBDztB-0SlJGYV@hOSSF6f9$LSDVW}=T!9mD^<0EQfhwSQ{P54KIW@0
zzf2<)K`hbI|7<TfNw+^>RfvmJ(~72TKnX^pbxdZMEoMo{XTjEvTxaz9CC)pfN<DCq
zNn|E5nW42FXC0Xo7^%pzK26(FdymOHy`4VO(+R`19%i6%(&K8EpnRj8#k@)iET3HI
zjrVvjh~D8;%BPr9;jH8OQ`f2Mnnwo@fXJYi;ar2xHO;(bKAVy&$)v3?CgFj+@50xb
zMM4sFIUVUr?It0Xa^{XbXAH}j+u&Ip+UQ~c@;oP38WBKBLzDqR#I?Swf;GNf`WVts
zS{Z?k5+M~uZirb%vFPS8-easqIghMrl#v*n*js+lkjudD`JQj%Tt7!SPu@(CZo&1f
z0mp|q`}ZF+=m{JV%7Qksk|sy#(7xfickCO<t;~Q+7?T7(b*Knsmb1)VnJ?xn>K2{%
zu&!k(WL_dclRziEGXNiUj&<)5VaeeOt2KC5Y_fbG?^|E*S2Zr*zPh=?_kZVSc=@X@
zad-cqo1?_eWAo1{z`89!(bO$PQR16OGqc1Nq)H~Qn5gUoCpl`y<im6$OBbmZ6<82a
zA#^vG@g)5^ipI6*JY#)t11B7@u9+_uG>xJTb8=HsniAna1xdf?v#<^Gd5yI-CNCjI
zs<uW%P_o5kkxzX5c@7VbxO4X|{k0JwsI29}xu?kP-y(QNo*M>(KC?wd2$4KbsQ0>Q
zi85uC17teg+`*?!#=B(RL&lW-u9IuNsp_OL&&y=|5|UtD>M7HiEJHF`X{o!@UrUyk
z_~5ZkgUK{n3RDP*_UQo`$&^7VC}pH+>NNNdp_A~!a9A=uya4e$MyLLx4<Vi2I`d{`
zazb0<+J>ML?6EK!Tc`CaLPQGBg)6&AC0HD`WKm=5maN!ddiqHU(^6!l0i#GbZEFKf
zy94<KQjGE4rYk}uoH^}B?^9dq$uTJ1tWu)XlQhAVl-DP?+*yr#S~O2I>gl8^$d#fm
zz$nGe+KA)BL!`=(Q4_mllI6#LP%qbdobhC)ps8!BNkeEoIs}^eEIGGpXo90^7W4)=
zS*dB~71n}i(~wjeiR}DM+T=JSjq?mQMigshYV&Bxd_JSh47IE1n?Z7Ow>7!S5QSnf
zZJEz&0u5!^qpq4&u$Z-xQ8D1krE9$Owcld@=#buEnB=Sx_AXxFWO{;iEgPF_%pNvW
zbq!Hcltsd?!~}9&t~0HZ$0*TpEiTwpjcAdgBc#T+j;5(mLQ@twnaNh5K<P*=9X!@1
z4~kKm;2cVu?$DQFqfOS~{X@XIfb0UQj7<L)BMrg&ZpPc`B$Xl8Ijv~f8ui&76-d`o
z*A>Ge=h4GQm@;Q~a~=2AO{6J^T9avoBz0T>5FTmL{rhtpJ^uH*Lzq=$ra%jgZUfiX
zH+lNPMMgzNUL>{IYq#Fz)}vz%=W`Y@5|l`X2%cbq5P~WMes(<J^DeD5=aV{5sw5SL
zCsOy7pyW)ec)8A!Y^N*Xg@6G4%#c~jxzUjQlOsY~ry#tj2+2OYI-o5tU12FRMdj-b
zuMYFc1T8%MUWv*y^`vG#jSP1B3^#LpsM#3wDXK^{4b=07qE})oPl$qOGHlyWyBYmn
zNpLN}OK3BSVnk*%u30cpJ!D%@c!O#b<N2J>fD##PT%fdNI2bXVERfP*OwK?TY^?S8
zt>64InUchyh$?4XTXe5Se{IO*XrE~{Wqo~v*&-nZbgrq}1#>%ReQOg_7@V^S6W_fU
z(rAQEPSU1n85Ct2`U9DQU>(x8c(2e!j?|h(U3UXviR*6cB_TFp9Z~>SYg=-e6K%xT
zEoD}u#pcs#ULxt~0<8tUcF0W7*oMYhntG0mf?<ChEkH?y&XayzONDioh$f2y7ac7i
zy=P5zryiAP!yBV@tak|o-8zf{UkJP!A~{D<YGiZ?E79F_{b8TktVTB#QN@%EEoDM2
zoVhWVQt9Qe4p>zkXQj<gZtisR3eHM>1wkr=Nb{oL1R@$1(erq6O16l6@-v^O{?<3R
zdGj`*L%kq6&r3{MFi%F1ADpY!4s9~pHifT=5SSdDP-KZ<8w89OGl&Jlts(tQgS8b|
ze?;2~CWlqZ2v-KTun01u%4nK7Hdbg|;u}j08W9Ty!yF%G^mRcI3W_j9g@ReNU_P5+
zy~eAI+RhmcM&x;kYXg}|{-nJtSJ*!~WdGohAWLd`kU1VLDy}|z6?gkhj++_%Y#rSn
zFdL6?!BXailku3K1)H0rWF~GZ>e#G?!Lm?<ZX-!(kJMJqgv7|0m~CERoue#DTyV^6
zT4yTC#M_c7xe)=1l%!hQ6gHK~3~wu<bqx9gMAR$kK0wbDl%1MS<%+7QsOkn`JWpTV
z<3qdW=?BI3-Y#wJxww0Yovl55Y-z%6auumWpbhB&P{<TL0Vzaoy;s7c8t*L1Su!mU
z1Bt8~K6>>E&#tdyLd#~EGn<}p?aD<ya%qo${N<PU;x~_wx&$G}q)0rk?ywkR0+1#J
zRif|4&Of!>9G0@_rAQ%kX3n$rgxJARmFng!A?2o@xQfHmDO#`b-Jkyg|L9kKneli=
zkP?JThqXX5=ZU03)%~)h$V#gDf+EYA&*o|8QWAq8xC&8328AK-E9TRxOB{<t>r>xf
zC}xu>S{QOw;9X18R2Xf@vyAb0OcVwsbMhkLsYmMrlv6z1KVUGpgqDGJF=IAOO0R+v
zwMf2)GUkXNsb>|tPw#O00F3=OdwIZFNnSi)Jbf3MnyvLs=HpXNPN!VCxJ5rNm~?HS
z5IlpmKE3sfyvXogvRK&EO7Wpn78sg^PjQt#Ar_nu^s<7jtu1nGm`y9@vw5Q1>MjeQ
zLpTT_uq&N=plX{=PoHuzlvZSUMr|8>9alsB(Av}{6Dee<bu~&!8XK5QTk^1vQYS>V
z6Kjm*`c*1@y$}_>(FW6)hM2x>X>s(j673z)r5noHmfF?m!jMaY^DVh~$j16TTzvx>
zYg7j;OnoZNr3;c!!F1Z?ym$v8u2dZGB*OsSJH&%0go5R3wG5kDE{Z3W!%{`Hl#@t?
zYn%LU|NTG5C!W5+&;Omj&+q$VKgyqd>I(neU-|3&wZHaP(n(4_A@!zmUw+_PSfgT)
z4jBSs6wK!fA_5x~P4HOPkdIWt2t`jd3kfl91HmS$bMTSLbiyDTBsr*U(g}Hegmaep
ze1_5eB=|6fwb6DGt4tl^2m5^A-VT#R<l+7-Md?ZjQKRlplF3XcA~oK6KDO(r4)3z7
zGC<-M3u@=!<}E(5F+#S2#mv&byUz=GpMyGJeL&g;Rqe?~IfEFvGt4<GJg)83KrUdE
zrfFkJ?sSRB(6%)~X$qaOySGJCwN#6m@p#NK5+cU*tno{sOZPhuZA+p}cMngIq9>Hm
z%<Bc2$*@Whx`Tza4qfyRSxy^Tq|{WEBesFbtf6>nozq!G2sQ1z=B<}s=jyeq>|ETU
z9QGNkud{G9-US-hpvyABO1o?MXD=d@3gab>YuFkNxpwXxSGTsYvnkguTtFL5IV_1r
z@v%>SoL_nUPHNq|#3YQ}z%edojY_RrTT5B*Sry|GkxDlcM0CX9_a3Tb2bmNiar0jM
z$jABffBrw@^yXb|z4Z=%^w0ij{@C#$|Jh&sDZccpzeJYFWI5<QKeey~+RmC}ypM{y
zYAA~yAXqdFjUz=m=E(^yM4+9uRO6Nq(sJ*W16n^rkb1IOh=g2Fl5-dKsO_AboXIo=
z%oie~sT_|VAJa@L`g%aG7;yX6J+v?sqM%fL$~>oMQY$6I<cjTOk;UN}ufO}J2;myh
zH-sP=4PIqq^Z!n6J-HAN6;<Q$wPjfJI6Xb>R0o1`P!NP=wx}2kdSpGFw9d|=gy!UA
z!K`u&2LQrAJ-?f{wn4}=InQ&$#`*?EX--e3=|=5t<hE_GxGr8&tPF!^PXjSVT4%vo
zg2~WEv#<?$o?}$U;c*U=rO+%?j_``w);KSb;Vr!Xd-(XLL5Bo9OD>@F_xd5VzX^GP
z77Cf=7*ilns|DyoAkQ+KaQFz7ZSaCb9&oWk)F+B@>WxA`l5M<=HDjW9Qc)mTq2iYg
z*Rz|5Jky(9ZW1dzJLxcb7M-xVVv-sA0M<pm_xrw|Kk=vjUFy?0MY&CHcZ+;&hadm9
z{v<EG^qai&)vv7lC4vq=A3tc@2Q;&Wa!}yg)Or{W*QlEo7Xmg2gpn8#ktPrWOphuA
z3g<jU(IZzmm2c9bRSFfrwl)2JpS`_3j!yOy$)T<>CgE?!;{}uPn8mc=Q-f{(^I!WK
z<@?!V|Dx-`{=fL|UcGhZjpKj8=g)y-e)KcvIo!%9G6UYDk$4}OpH3m>l*IsBFOV`R
zsQUdLn;V;)#xc{$>^;5uWfuEc>VUQOiIS-Me{EZckwMwR2A?uKv?R(WCOIV~Dl-YA
zZ*5{$3Al6rU4FG{`P?TyL73Il#}iCU*(=AhDUT+nOxx6&H!)I}5@qrvB7rWLDXEsE
zRy5IJLcoQ<Vm?R2NZT~v1Ml2?hs~{Znn#ww-WHFJ4hbsFT18CjD|Sm`#i}}E*4C;c
z0iW@O{IkyVCnVq}>ukB<L|lvrmGLM4ouA};f9DVI_{|4gfAKM{7;@#tE`R3F{(HRg
z%5Sr%PnSmdF5)3Q19UsBP(o1l`V>XMyjsv&iwK&ksfhhlVz-TBHlC--BN*D&lNX9b
zGsm|PeCl-@l{DOC(PwRKoyof=SZ6yla7yvA^FY79fesl%Qzi;*NYva?=V+k`E^&5(
zAUUyPMOB~Y_Tm4IvUrg)2PuZc7@5sK%{#ZBV_{!q@7iBxI+=qo96Wf$jc0CfIvF!<
zDs+}qOSY+*&8Cb`PEblv6v+)f9IP`K^@!9&8dR3b#}Ff1Tbl`aA6T*W<(U~o?8K@u
zf{MxbwB-5=naIRG2J$?oaSl&JX9h>4t`}V%2B{6uFDbGDUnQwvlrd@W?aS1Q1{D+O
zS8Ius2B{u1?ENHjUSqz{coiu#Llly#ZfM&Um8%q7CF^v9V$dp`Dtz$7@D9#?4@$aj
zev=0OQ3{Use~ze631Zk8nG%;M%d$?<7T>dxoW0K8&}ZK#WJ2bL5LSqI0NJImC1Rdp
zd-o#8<BDO?=lTmDp;nIHe&rS9yyA18`y5~Y?cX5cR|$!U==(2z_Wi56npuTWk}^-e
zr|E2-Fg6anwe<Uj+~idAhPFxZ1i^v#N%dCE7X&MjiDrgT>EW`yy_KqrwgH)DUy(>v
zRYVDEYum^$q1P+WA@Ek7@hh7fD3|c+gxDee6K11sYBDYP$g|hk-Q1v;WoTs(QgiRY
zLtcO9Can}Kf@fN#+2qg|hU+69A0DJU3#0kos^t^YmZQTX9@T<@(HJFAQZO2AGF&fd
z8$h(^EZuKPz~0_&mqQfaQ_(InE?DyFmYEsxOfqW65<{dp?7XQ`)T>m~-chwl_U)Xb
zZW=ZRCD@qOtBt5!p-slj&N;28#5}UGy~$>=j_&1r<DED7rI)|K!b{q!qh+f^3}|iI
zG|I9hH>-0xvRh|XSQbs-&D(d;=XNpPA-xA9$Ok>{ynCN7{^~Ci#4@-|g7;({jCPp`
z6Qx|Om1PU~Y<p@ItFUY%cX7}D%qAmZnxi5l0Y#P%D0=Ih&I7~ka}2ikI9QzWjj#MV
zBdz%4CqKb2{mL&S8-kQ5sX8Me=zclEwU*QBl&qIr;L=A@G_j(cwuCBy$Pg)o(6-4(
zflY8lK?0HGSx&zwXkuVzXPfbO%+b*iMZctVP~-_^y{H?8YwJi4bGzW^_>i*9QleNC
z1m_Y8K&h3M{A9Ap!^8iW(cl>}Wf7>A3TQaUK=iqL=VL_k{bc5KW{beWHZ+R`{j$f&
ze2P#>x|qqFVIe@I3}f#d9<VN;q+orkPt&xt^ArJ#SJZVyJ*~NN>FUZ!y9~=p?#?H@
zNZ<JpWCwg(2JJv+ii!c-I*RT@WNmVMmYKmv$6~R-<eC@*S)QD=dFgQT7Z6R3^Yb*g
z?_x-AT>Wo2Zhn)+bbuq#K2@%$>IUUf)v!XTh?M=Ve46hBZSc{Po4-dh{{lj8qGUiB
zlNJawh>x+M1mW<ZL3)rn!<a4;yK|Jk2Lp5FI?JnCI$@W?JRw9<uf=Zsp28(k9=lMK
zgAq61xzG0ch@G`HZr;1a*S`8&Jhyj&kA3uG=|IrszAWeON{CLm{=vBlQ&7)p7PFdS
z(8FXoZDr9RAgo|^T$357YmYXHUfyTXra`o=+Z6gmN-OM_J!JG~U~^-Ww{O42d_E^D
z6O6Dw7;t=aOi-H5^$n~I%;qyx47^<S`HOp7lx^B+gHIeZNb$6XM+ZD}?L7bH7e31i
z*RQcXS|`s3DAVWY!9#xS7k`0YedATCUZ1xgJVsc{wiG=5q3gWzX3Nyp_)(Al>Zsxq
z)0Up;vEI)y-s7X|9=4XeG?abKV!l8iD0E3bAM)VA11|1eq{z$nRj8ex>I}lhcakn;
z62oqJKSDZ{w;|w~mZHdML!xFwV$-FTU+8d4k)lv!xg?j8-R*6H_B=j4p<+&3T5>nw
zt%vuRhL$#F1SGiBTCD1N^?kKAD4qVC$&8!=Z!JL@+UU9UXrGH4L)Ocj^|b*T=k|yy
z^61fos<pVF2taTi6_=*WC#L0;zL}mS-g`JRXM{8I_vPCw+5D3>mgu@MLZo$O@`5N7
zufOpQmo~S_GR-UB{048m@-m<Qj?ePKi!bm`zW9%l+_;-_3PIv~HGs%Uq$yI4MrFxV
z#%Qg_V9-O#9NR{Q!!@+ZXj@OepOY5?AuNMofi@PS1xgrP6A3oZI>%`>Lk|m7Dbd<v
zM5L($&P!(Vn%YKYjpgKYO4DhNTj%i+yzsQX0b`gq3yvq>hY!yoWlL*qce1urRgy-U
zOrvz<_}~H=5*Z`45Zr$FnC;D7N?l-tp{ddd&sk5I4-vsoRgOh#X=~5ybk1~gLT_DA
zj0`7}ip|Zd*tS4mPztglBP&d2YVzb-lNn9FEGhGx%p^aI(Hf%^!8r)s$vOlS3L~?m
zUW*EaP9?6dQ>KVaPW!Sjl$l~+B!0S%;s;PFAsb@sd>%32*}>!^2E!hMGDn4o5t>|=
z_@Hn>P}zpMt+0z0;{=kH);sFCqZvn>dkrCf2HV}hu9M^nk>j%emL~jN>cxLVY<~zH
z&moD-=M~AdsylQAtDAVm%#u&=&}VP(?hr`~qz9waq`sIUq6tAV=nZ-Hsq0`H2K|zw
z(^KBP^Dd*o22-0J9#ZOV<4y)cIOE3s09{ThRM960&3xYC+CZ7-tZfX*@&bt=1i@&u
zk%pQ+FdXF=<54QmA7n_~CbWB0xH?r2lf|3{j~NyS?J2Sd)?=;2wvoxKq6sirq*<H@
z0a>>wYsq3nyBgyyMo8NE9M#O(DK)ZgxcSCgy!x%zxO4j!U;oW7V;5tdyLuV5s1V}`
z>rqgIK;5)V4^DV$`yA^i`VK;~jDjx6O`qr$vqeMWJk`uHJ(|$kDQmkKUIeNdHaD(O
zEd*JXBUGU1<;es~K#NGO(~PKB6cl-mQ3~CyGby?R(U@om(Fcr3#z4G4Mvd%Rur@fN
ziU=7fa!sB|21N!nTwK4*cYXT1(I#S)q;6+Sn-gLd=?yngrpNmFAjznsz{m_O3zp}m
z%2w1(m40>!J~Y&Apq@CInd4;9^75_6%mOT$mIn`x8BZ&of9|9Fxu5(GdG5InlbeE`
z8IY?Uwsl?1gha|juvn^~mg{U;`Jb&m-oMW1q93HtM3Jid5H-OGu3Wst?$$cWd5SXU
z?*04BPG<B+Tb#`26nU9CE6Q}8m8AhIpv!)V&=Oxc27?}x$%ML%i2i`ZqDrhNDL6dd
zr|4@OH8mcu12#D7dV!BnlmlkACX<@lR@mtS^uTcC;(5j=Q+!Pt28Mu^5n@9K4X#<R
zb^R*)j}Ea-ineThAU7FOfYgT24OO+G3zg|kCQ1`JEV9lFy`n_o$zni7a8fdFTCzN2
zYipC!`Gl&ifXK%BI*V#fT{jHY2FN@xIhdk?Vtr#B*LtSoieYb)t@T|VzVj_U{cRtn
zsV9{E9zqD_vw3=G#xBq&%LpM*Rn^KZyez+CjO2OVIZDH7UXV@!l1$}^Nk&8=kufP!
zdRj3#{VYoSd-RHrW1B+~6{5p|D8aP(GZeksjEY~Ul~Bk$QBkeMxt7+?$+RFV3Z|zO
zA}ZQe(KLb9wwP#X++`x`fM317xzQ=BB^#iNJx(Uq2;mPy{w8~S|0M?}Uqz6yOge-2
z4$#j64k1+{-nmuUp@>gTQ>lVWEvz&k4V=Gli4VQ-0#YlgYQcCmW^FWLI2iEw;X@Wq
zGT7K*f4@#4X#ZcBu>MtDMZbrSj=W=}O{b^$vcNe_T{o0PNmbX>lNwccYEcnYfmIG0
z9JcZ(X%LxWQZEqR;ao*Cy~pPH9$B%$;lm?*n;2zUM7#{dP!U2)U9~G$K_CrfT?m-G
zKxbLvA%#e8GS>Pd4iCrlih|j!rPo{M<oE<-GOk{{$kP`u@ygx%WJKy1@mevSO?mFd
z4GvEaaq>z2gff!xd_o>zb9)mP8fJ%c`b9~<H)1xMQ7<glFFZq2d3xo5?VTNN-<mKS
zj+joTG>fVWWfQA_xaoAdYMZKV-qyAWFu9Ddj-5H!dH=rFkA#?_4P;Q5=*hF34}Ibz
z{Me8DD7Y!V`0yrCrW943=Ujd2Sw8e#zk|cS{?~bU_?Y$nfZ#<JIoozrZcsXsWoe_T
z>xQPz@Rd)F^$=)dppwui$I*Pw#;{;_XOpdS7y0%VKf|dU^5}2<H2thVM$fpOaC~x{
zhUZ->>2k>|H;%I=+Osgkq=FPH!OK$gB0}QYrM&aVM?dyS&YeF`-PRnR9^<@YYjcyW
ztu6K+KPDe;qKW}7x}>WirWaR8G^Roql8u3)C>5$q4}sa_h-_pKB4MNVb_NVLOAbye
zl<*Al0mnyEN>efzlqvh$0wG}L3T;|yJH^RJ)PeDQ##VoiOgWsb=?!`mnL?EX!%@lM
zaWXV2B~eP_5eTjEB+PZD1qg#y9fDbF@+>2$kP_aEVbIH%&YmKR+w^<K#A1kWmiv$H
zv$j6uWOl%4qolvpPo_7a(0NJ3voRbX+CVjPj8AJKg8C$I?fN<H-MNkTH5V?NV}Cyo
z#SutDL?WcdTZfOH*=)w%-X3k+QrC5wJuG7=<CCo7C<IERHlSbqq9r0|v;-L(%15?0
zzK5#)1JvsK5TZa!$FM(O?EDHC*z1>AzfL{=t88ukG{fv~Q&+bLC}y(?=PW{`WVG=~
z!@1q-eC|6x&lms6ukp_0ZFIOyv-mxz_`__hJxz48&Y-2bh~A`TA$G}(v8Af7bNBXt
z!eH&MAk;4q!czGWS5D*QWsIRL%Oo_gmc4W5Iv!HmEVa^4hJZ0b46Bpk#Y>m)G4jP<
z{6(a2tJiaVV}rl{_x}NZ@DKecpZN55^8fn#e>=I7yQt6j0rZ1Nvmna_d7dPY*+4*S
znNKH_8wJjtP?n0dwGp}>d3<<Cf6ym)n#s6g)Em<8_i%OU5G-abA{u<0Q-v8tu}kf1
z+D5Wo_K|Uc6p}K_SRV~o(^GgfBNA}FMVNx<l59F^g?9~krjQ}$+W9NU;5dAA%(Krt
z%L~tam}*jS@@St=T)e`!J^v!_+<ibJfzc3*#I_Abrw8;0IWs$3Jw&5)j5d0NMNOVd
zMtT5E#5IwVgNmjJ2x}OY1782uO)l>3vAes?qy0yW=M6&Y)Gm@H%>lfnu4}fows`dD
zQAccO-*afG>pFRkv`HB&K0!m50jVNtbTrs`%f`+czvFX1z~?^q88(Mo?5uBstC)QB
z)BO5ZUcxzxDL44Ur@o7=%THtaoBZmRe~DlH)h{!hRjCDDRVyx=wJ>i}1NGeARdNw&
z!Wbh3K3aaq5B%$V|95;FcV7Jp&Cw}k*&{1TL~q3Ug{x%6Z)00eG=aLU5F(-Do90Q_
z-TP~7Aq07zQ<f#u=`_t@@|^j6p4e*gi4QAwT@WHjky`9e-*}qackl7Ful_oDrm35T
z;b1^jRebp?zr~OL@juRmD^Ky@-rEUw>ODau_@lo3)OpIDrmkl=Ur`PRINvhdOlqK_
z%+jm?Pvhro7DH?s5O_9pI*Ip24MkxfTIRDkrBDp>A$6s2p~Z>7#(9lw7vPdX(Z@*H
zTf-nZIz1-uDe6X%nVfI`)MvPJ_Z}znLu@l8@AdHE;~1UL>v@(Dd@{)>B^mSwsaNbG
zMZU{Z*Pi8@4_?A%l7L`5KV|R28dt9DGLJUNU!`0j5eB^>%0@QVH<)KF)jWXJtZnVE
z(cj?qn_s6_melQnYuBH`3BxPj_$GDRcJ5e33@My`GMO+K3{sADS9zQrwhJji3T&`l
zHIm+^MMA$g+tTZO9KoMpJpDncS`otnrQ5Wr)lE{onJkGCd_YMEq2OTu-(q{~huPcy
zD|q+w*k(e7q0HCt!7)Cb@}K`7eu_WxgTI@<@#6oJKlhWr#p3h_iE)Si`hdZppqezG
z0tA`d>REx#3S^Az?(A}Kcz{u`sJA#h`VUy&cEB$Xqg>^loSlXAdOZe%0Zr53oa193
z`xsw+=_L*hk5Jk$7>sD!hPrN2>nx_2(5AIq+}q;l;XQu-AN&lhZ<&n8D6MGfhU4QC
ze*BO85x(%Rf1Y3c;y<9S=ENw7>7V_9yRJWd5t&7*YC?!DgY^N{TGn?oSz##45k6RA
zv@{{Ib8!uCJx1h+tfVLf<zPXcX{;S1W^)F;5z_T(y{EM^)^;V84qz+Lo;Cz}dO&Zm
zNiP_r*hfeZ8uree<NDPby!O^>)NY29mY!CWd7leA+g#b(!dZ*08y531&)&FBu5Zxm
z_qnpO%d=Ona=V&wYjT3f6}I(E<|mvxzl91p`Ekc76B2JNgMN=;UJ{&TcXx;JctL0l
zqVZh4dX@RfjAk)K$ZZht;)g$id*xN$8IP%3i%t_<61#3eQ4|aY1E$mIds;cm%|t0h
zkP++C@YzY3_)?_A!C2&c?{|I|f9&7*H+bRs8;nmLGoQ}MB<vp@bL-{<2K{x)odl7c
zHI_^HfTu59<HaBU3I5>k{UQF=PyaN(`O-`1j`syn%{6q%reFA-zlYzSW&GUV{F`W<
z;dGzh_d`F%IULP{d(7KDT8+?baP9d|LT`f~`~!c8*WSFzJ2zjWstvd9-6hZSRnC$)
z(-Mv`;=J#O80qy0A+WKr!S(Cc`O=rZgtci9KUf<wnWU=CBXL1?%GAh(oejS6(pUJ0
z|KMkcF|e2~kWw<4OxRf8;^`|F_}=gQG=KYVztt5$vXfCp^oO4PUiyQQs-93c6ON7_
za(wcbsy(Dy)M@CN3A|4csC8`_jz(n0&>F{lK4&raoqO9-3^R<$sU~xR^7J=KblEZ*
zYW%$AXn(;l@3E*WDsjTB3Y5`M<OQ4C8~odU{NLfnfAmlA=B>B+Pyf<?#Qi(3Q3QoD
z=ZM~8+w{=)!7-iIWJXgICC&w$gF#<&<JoJxcKR}h^BGwT<SH^-oFaAL^x%ZLO);QF
zQ6%c@e2x+lRcd6eaehuSsrl$fKgK(6zJm;bjrAc_wZM+ndHR{>aZd29S6*R!e2P{Y
z5t9;RKA*2V56d8?rJ-(FX6F3@Pg0~2BtBY%2q@X2#E+n|{}VCx(n-pteyVdRa@TpE
zlqCZ2NrJ0QMs9LMbex<Fna%$K7ta3>o9lmp@$@S+&SHb7C^M$hF~9!XuW)>vQ_p@3
zqDIgrPm3w&jgp%l6v`M(o~0I|NCU#)eUchtX{|zrf0vz|H+cB)ZYSP4+kbs>Qd+vu
zUw!peE?&CCjc1?b-G`5uFBY7;aDkJP6P%AkwZ@eT=ec+DE#7(KRbKw`I(Bi2lY%1G
z-KOjq<chbx`YYVL{RTSJNFs?|7gDbG2kd(OxzA$?NmY%R&W<=be#G(VV;X--HER&s
zB)op~)KyD}g0;0ZlEu_gw{zkmA<hw&;Z`4Fq#jr3TrpfTWCi$5@QWI8RHMp1bv<Kj
zJ$;*kx^$+w@%%IVAOHMM@v#>^#?SuGKg(bKYyTz9+_Pyk8?n#q^f3ow!LUEz&h1;=
zz4Zn+o_d<ASFUk-dP1foAN}Zu@dpoi{p*iu7LHtaYFA->MO95`yN7641WZv9W}fkQ
zhL937NI_=PlNmiR;M~SJzViB47?e4^Ue0VjXSlx2M?d~4oQ>Rm=N+0wh1MB9q}j^p
z=_$kEFwFzI+1Ju%6hc60MLEc6x<eksRhvml&yWAukMfiM=8qw*XL@?f!NERRW_aVR
zSNYaAUqj|47dAJE!t?Iqhy3h6_&I*0Sfiii9Nl}&jpv``FaCx9gkS#3m-vtWi~j=m
z-~a$299f=xAo~aVsLPkgREqOGtpxigf$#qQ&(U8WFxwxK<y-7tyGZ}SMHT_H>9ICi
z=keo5Y;2~e;20y{{_Wq+OE0}dvsk=meJw>hS(ec>4N55{lL-$VJmC4~pJy;y<F(gc
zCo2lh4~Lwdp5S9ff4IZeXvmwdyv*0X@+-`z6QZp_DKaCw$Zg1D#V`Hb-{SSJ|2nyf
zSTE6ei4&IKELw(5)(7X=T#I<$u(vtJhnl0=yWD&5Hq-erCWB@%1|3-2Di|M($uyKb
z&7hyBobgi7x*0+xlBUix+IB%}=b$|6+iS%2K!3|I7_M=0)G|5roKzLQEr~J#2|xXr
zPhpAdo;y#Om(;aI#8b95zroSr4MJ$~UAd}t=b7k$dOk>=e(IQyeEc$twcpPxuYZ%1
zgGUIv!1_cB_AzCZO9Zp=jNl`gDVaBxq6x@M(vBO}`kORu&HmA2THA7TdcwU2_fSUD
z?>o+)yFgu8zV*t>;JXYDWm1NEWxM=wowBWcOh^K4XuYFxmZDb>r66?jJQqAhL9Tum
zi-kaPl4Q0aoo<$DjAZaj8+tMkroOQADHB0Thizxvz4sx!|I4`e6AadW8z(btaQr|1
z&VSGSyPxKZzp##Np9bMk%CNb;jST^*Gm5+*$!MSC6D+aZ5y-Tm6%B%x`TPnSTR(tS
ze+}!-u(5<#wVZa(?J=EBX<LgFlH-#x7Y9QIg8`GtgxPFHo@JCK=hL76{rui9{7(Mf
z|M)-f%5VNEuYTjVu+;>Ym9(v6I2sV5r8=2%aQ{v2-+Ue4)M*IcNh;H9<%4zQdXKG*
zZQ@!>)6TejZcc3{?4RD{_PsZeCP1iYsu}rU03MF_XAJr!#z^wf0Br<15464_3PlWp
zyq^w#GI&fD*t$5Ptt&<s3WkF%-n}zsHgUM)St7kU%jHXZJoDVsI3$-ZUrC0zre*G0
zp1*#Xs$Ebm5}spyZHSSQ?X5LzsA;jtp5fBQQ^=>DrX6kY_S>(~&L@O+Mq?e;*_43+
zIBOXnwe*#Npb0*byELb<vxaj|-QeAa_nFsobf$Rs(F4{tMp*CZ_cyt5<2mZu@$Q{>
zP$HmIfe@)JRn2N-SSo`e-gWpN5pXdk8Ejc3-$i`Fgg$?6muGHV<C|aqDn>w&XDcai
z+3$1Xxo7b<5Pi$h$ssRpZzF`}_}~$7w8mmSXFi=Xd;Eycede?L@bCHE_?Q3B02F;F
zI}_4tuYHRz-F%BY?{PY5`RwPuz-Pbbd$>QV*!$3@xbo2wD2SP$a+Y465mh8FOQcX-
zzIcgme(RewF0iq_&GyzVhqEUhoRqREa81KtIH2tHXqwbMuiHda-`v?H&vWX!#<}Fo
zH~lSs<d6MHp1FL1pZfFv0dKweEnGDtI!Du{?Dqbkq^f2#vt#bR{Y?(uy^RbWB~=1G
z_!KLyU3JLx)N^!v$m!$=>lPp^LJ6dcm>4<PKcH@EgwDy=Hu#>8eJ9w+!{ggb>V2BY
zF`@yf1+!*O*&EWhnzCoW1?mYL95zhGGe+wf&3H<4637+<o*rIeFdFdg{%z)y1OC?E
z`tSJO&;Jm2?mpz^JNKDaf;MJs4E_cuC!fTJD?sQH&^*z5vLbmFS{Hcw*>~w}oaa+(
zPxHbvALr=sA@|?C#m#qK<@n?g!$BkphX@iyWIkC?^!jP|B?8(;qKsU<y3PIl*T^?4
z$||g1kZI3@d#|&xeU9P!B}QB8T)1|ghxc9wpGrdK8%Bd6(F4lhq+l_gCX;2RX&OgX
zWQn~WJxkk!5|J1#(l#GL#6?nAIa2<7C-)U1d2M1$3n_N-vwF$%3eehQwvy5j{5E<1
zd8GOVN@Y0fc=Y%I|MZ{yBOcuS!$b_kmd&kgiaY~iGB0%!zq1eFNS7*SvQ@jlIY+gq
zIo$sNwEjE9n81K5qm$MQheHsswY{C@MJ}?wzQI%1u5#<0o7}y18!IIJejlwgO*Q9d
ze&+9TIyq(DRDA7Q-$dygEcDm6Qk%=R<eA{@JGVI+&k~v;LX0*gTE-9B_5s!80b{k`
z_MKa}P*c?tOeXQ6rKdAyizAN5hr}pAW!$**AwGTWI`jF2N8>w8tH*@8#?_uEBMZAA
zFMHG>8Dg7xgPSYH59YM(oVE2H)&4Q@1cq+Jjjf9)`4UP-UVrUt{Pf@c8(h8mQGWSL
zzs$+FqHQFe$Xo9o!k(dT)@bSlCkGEvcwAj`<;Hb<Uhvwx57@l%5w>@(@~>?_#b-YH
z9X#HDm%I1g=AGNGp$z2N0<h$nq|gxSmcgi`wJl}zgw|24^%#r<H*dYgU?;`iwswZ1
z;?AvaaPi8s$ZUgim(Me+_L-eN0-xgp5Iw!#0NVu2pvPoBC)O>-7=%&;;W4>Rd^XYL
zXLM_AcYB-A)SOJF*k-|SZH@C6E|cp58!f~BFc}xymZvthu+}mh4jB#ycpqqj<DGZj
z1|evwhUcGpmchk~#GCh0JG*nlA3c7=Y_v{OMTT4F_}~1QKZQ1u^}z;GXS5=Dxx5SL
zm<+H%!QQ1yjMg@gL9jmPb9}Pkjc>ip?$&wUx_AFf3oDuWilU$_OLV5$*xaP99a)yM
zwY9~yE0@`S_=vlA?=l*V5D^Bw9&f$zI&XaAC3?diuitzN5gbK-4KH(sYnxQn3=tE`
z{nfYMruE>XAc(ZFif)dgZ@=>f*SNyL(H-8t`wG2vh4U?ogGiAXwDQn6?6@J5Ie+HI
z{v3bs2Y(c$48Qd2|CFElOaCeNZ#_cij^Ua?=LO?~ia4-jN)hJJ?hERp!1zegtQ#(F
zZS(Z{1)jS86a%%vTa!0An1b+vuf6nTPLC@d9gdl~Y3gCtk%~K%;_oo^|2{;I6f$KS
z=#(L1T?8>>;a=yPH+wXqhxP#xOSU$z6Uo_~jX6EO%i-hO2o>2_6U0G@HiA}1WQZ84
z*pfqn-lrTc$c-fPC0Wc6lVbvblju<<Lkdl9S~k}Tgmc)s!t^z{PEObm13{H!l^{Bc
ziXe0Vx4^f7o>D}SHY{jx?elmxkXV)v^Q5Mb#7KuENb?0_bmz*2rL#1~)asHl0x1Sp
z@@aI;@jf7sIOnPA$jRw7LU8nY1;b%7WNNM7`|y2_cZ1GpFKKQ1_ocyki}%+l%a0KF
zGBWCwDnu#8Xf#4eNwuhm5i(P-HcSijTi^O7Rkc{@_hazfe)Vhg${yae^hfJFdi0Qf
zp~$e5Wy#U;Ax3LH`l)Z{aQ^_+TSGK4_3K@dN`LS`CUF1mEv~2w+`0EPX6+&Q2&%^|
z_Czok<OElv>d54D#<SNx$xr>6zr+jAevH~y{MEntSNQM$Z~s^3vx=+?j5c#Z6f{(b
zwU(YLuqT4*AW|K{bRl7{$NA9~FI>3FrOl^VJDwrG{M8O>xZofE&wrnt?N@pI?oFbJ
zRO1E}G<Qx8=-D1?sz;e?h8vrdMzX%O&3Iljp0wOMm~vPgAf|$V#(0NT8=OD)99j)H
zv2#ZEe+8H^DoQSnGP2+ixxgYYk(7E^7#8jho7)g7MJWa-Cs;gwKnRYL!}|<3JehHn
znWxw&aaHofWLlG{2Cq}jiS`QXA%>R1BnP-?=Q!`_YYj1w$#ie84f-^7&Ah6J1TvST
zvb|o97$ZedaD04>l#;czHJo#u7elN}#E%|7W_^7fDJ6TmyIi{QEaB$86@mw#u4|@Z
zj#~u&&>#JMoIiJwURlsDOT6<bG_UKP2n5!H&IEPikkWAH?tQd2Xsx+-?;d~PNB#h>
zyz@GYQ_z*@Xb!!yq}S`=ykjyR6N4l-Ia?c>Jb3p3ckbMwZQE`x8L6iy+<Nt^WJ*$G
z8BQ1$RYMRpy|o@nX=c+|`nmRYxO?X=`DmNT$qBY&iY+nm+8`;7Vy&;(T357ULBMf%
zRI|ChfmSd%sfb>2;oNmT|J^^x=}e+B!G}NoeOx&Iv)sOU2Q5QlV<}0j14iW33r}tf
z_To0@FFnm;Jz?TbIiGLx%-&@-dp)YjDev5R6>nO~LbI4pc;#y^k!M5tbIWG7#pQG7
zdG_Kp7Ska={VOvXo1tVxtsbbJAT<5KUEaL&6@K%!Mw$ZICb^o@kv!LoCnsoSu`#2m
zXBh1n3^E$q;>Hdg3?|D6##7`OdxJf^4{Y=|$bE*^ik)-Ygea+8K<5SL&)?u!_Q*vb
zI!p9j;FeJIhC?3gAE1gMy`mrr!T$J|qgdf&3f)=uGqrLY9DfhvX_srzxwX^mKy<QQ
zR#iv3L|4J2Vc6MRjD#;S7^M+T(At)u3_*A%lOEPy0X&1@Fb(Ep3X550oiDGXY)Of7
zF~VRt<mm8#EYr|!baf>_{4id8DTQF2!I(UJ_<)NSFS4<|$?@?Cr_&R@^6r<}+S)*C
z&C^fcz!<}1Iz<V%w7tf3Hs|r<N2whW1g&-S`?_O*!E83;%fJ2`STDJ7@gh6t2i$x2
zK7_Vww|($bSea0inP6+(z|4bmOzMiG<CtcK^UjAf6MW{AKfsksALVEo$a;Og^SghT
zzwtl*EJ`*g;Rr2A5wWeOR3-BhOVReYJh(=`zr($U4<UN4j?Qy^=REyPGM_%eg=A99
zGRebt?{M&7f<Q2uMXqjK;o8<^uI*i+t>(;|hb)?kwO)>tIfvs3K`E+iz`esM_wRj`
zMi(fdkv1YkATKl_v>dl{>L!UOrd7kN2V4D&YBEE%5*Gxz%yF$}eP<mP0-4C!(Cc6$
z!?gkHn;R@@3uIs<dskm%c5+A|BerfZTH{;GXgI_n*xx^(zjZFr5uD}G<cNvt;-r<L
z60q3<%y{PdRbFl?V%y+=A_OLrF<E9X#;hV0%ChXzNzH1e8Dper8fLQ@CnqOdx^#(h
zdy9UrM*w^qR>+Udt!>I^5JK|V&we*bX{5^VK47$5RYVYpLSY@;zWadZU-&4$dHz><
z=k>2M91b}=oKjU4M@L8W`~CMa=SarPdw1^g)Kk}SE-;@}+<oU3j~_fjD!6dr!pcT)
za&khy$T+{z=iP^oI2liAeWZ;7>r)bJ(<Bbx{rmTM=jKhc>2u-Y1*|kYKHMjU^h^p~
zpojf+`o)mDcjmOmfzy+Q$w|vl?BEX|WR6S>(Y+j#4|x0bJLJPP%HEK|^titJG!Hkg
z0dwm4F=8H|jbtYU(hCY<xOV9ppLpt1ge#gRR#-J>5stZe_&W2p!kLtzY)nSbfwqCH
zm3;ilC+LZcLEdL?Z5y$Zv+;7xy9arvdWxu!LhPlI4A%q~uU%v<WJsMMUB`=wjzu*i
zD>4@2hfEJGb?s<{Bh-P*=PvN#`R6%#G-Wv4;NJc%9=>&(>sO!V#b-XuwXKVI+fvJx
zI#jf?8BRoGX2{flOTB_-Uab(*T1eWeX1G3LXiDt7PN<f_fMHQ`&IHQQV0B{>J@dx!
zPha{xQglIAO9zmEwzXuLN&K<ZMv`U|-Y44T@<v>4AOKoB>iP<q*(EoRX{AbiQ*|CG
zHW757N9Pg(<BU7}$*s-m{ZdMX!%<333s1gp)6U5A@5b{iF@6gnvehhPu~;x2kC{zo
zG<8dif;=m@cmF<!H1HgaMmTHPxp<KW4<E2tR77u?o}Q3(2L-7Vi^YOPRZ&$HgQDO=
zH(q3ObAvbEe1o+?kMX1>h7Vr*3Se!xfhJ?RKf_gU@OVnQFpSg|iw71=MrL}fuL%})
z!&|rBM&u=BIY3&&bLXGs080on>XUKGr?-;qxIjh0K=*jzh3C1t`wZ1}i&r(Z7&C4T
zxI4auwEK{nlN4ErmloeR&JT9E{>%$xUa`5h#jwAQAFXj`|Ae}lv44ER@kwBJbDf=Y
z+wAS_vU}ojwIpU4T8M<!(F$uD7K<6UUS|x2#iXTDl35-3@a2!NBX$@cEf}ti_{Pnz
zv6$67cjW~>{M-v{6(ed}QyWWNPw=M2%1D;y$hAK0Xh^%Li6K(v1+HmOLSo9CHC;lR
z=9?-r40<KLuNjO+=|NkTJoWW&!kq&OCFy0F@p#O_dt{aoTaUGt-Q8W9reQc7u2@@3
zZ&#}P6K`xjpL2S8%I4-Kncqn8L+A{`Y5uddxs5uRAyvk~!7<lbhqD12B1#%$r(92K
zNzuAUK(V`fo-cgi_ws*v`M2oxmV|}K%P+so(>HDqUit0SeI=Dj^Ps9?|M6qm#?iJO
ztqs*;!J?THJAuz&Fd)x!dcBfcw{9n-%VLJD=43KqW{9v@R4f(?Y}>N6was%M`WRI;
zXEvKK>KDvr)8qpIblG30$OjDd7GnH}tJyUe6<pi6NIBAs?ICa8e+z`h9-i`xKld{{
z{o;pM8*UJ%fh(fq+{b<wF;uv=O1aokqjZlD1J(Q#5dypYbz-ZL@&d>1F|WS)I$wYH
zo7i5UZ${`SnV;5_W<VsPP<@`d`Wz}aiXvwouko9&e}cnFAFTs1BuSrA8l@Fdd79?i
zdHdB(HZOmP(fVzItneZ6pqi#)QCIW=eD8-o#iNJsA_<g5&gI?ne8-hfBCloCjprjz
zUFV>>$FRS~`L*+uL9(DB=$3__(3WHB(6VR}>9Z&@8Uj)nTHm0hLiGhlZVYQZ1{-S~
z*GsUr)o18^a;tg>-lCLe|M4~Ei#?R`#Lx}Kok!?|ichAz(D_Qz&GY2uOIJGehcoYs
z<k^6&wH*!@`>BN$EcN_FtkobIX7gEcLMu(OeY{r#yHuSoNubVJjLERJdh$UrFke7c
zejiG|h7@|0MX<iH4ly#HO>s8x;)@^Q=FN97TBm$_A(&04sV(s65oIwzD8uf~I!(2}
z*@k|<WWJaa>J;I-v$Ml@eCp#|yZ$W3Xv$3T#+$DqL(Qye2vOpshWY=R{Of<YS5Xe=
z<!j`AjW9PnJ9w6&7;<@g4>O3|-oMGC$pIL}n_vAJzqY-{?!}7?^AU^3Q@;E1i^#pF
zXq#yQXt_wHN&*@o0>?*>I6v59r#HYWgUmACI)0UJy>W}zC%3tpX`~dijiZ^h^kqpC
zGB$_1T)*-xZPhSbU+2nmH+bXbtC(q-GH&pcnPP2oll9?{(b^h6@o)cb{?0Fai8~M8
zC3=ewExwH`syS_2Q#Cbo3dCo&E+Kcf7!^4mIsY6lY(Ga`1!`+~_R4w2_K<Vy7uoKu
zqg+V1oUCH%$5>f0YZutc($9NP2%JhDs)-+CdKw#G<~(L+okHmpXr>f9+k^CAt*a#O
zZ7iLN7cB*uNr$P)=?OZ^iFKNDOuKt=DOz#P<D6x&SO8#*rmhw&7K@c+Sc)em#SXdx
zz~lXWrggyhoPYYoFS2{#GFPr$A>;xndfl4RD-2HncK7y>F)|pgF&M7n-AVdK3&H8>
zDb1}wo)^$m>ChKa$73)YV4IfHlQFy7d#rD4^6tA25U@hHjK^b?k{o3@Q7ZDnu$Slf
zwnhpl%bermV}!{tCgak%>wM35eK&*ACPoYXNi$_WGu&zB1Yrn1piK^V`0(WycvfEI
zTycRS49MwWnGsCQC(b{Q6Osq}M?8N0A(MH7?rG$B#)eaz@9i>e9o99NOwu$nTzyQ;
z40@wry}wNz8b)Q7!jNW`z49F2ezm}vz+%=gC^eb(j29Jm>M8W$VfBy)hdWgEhj_66
zVea0W5;;x4@6NC$rAQ9d2#_;7`6RdA+GB0)CKs-JlbyZSSuZ@Y98vT(*ftqEx#8*S
zLl6~BU6TQ)&D;18Ihu{JB9IRY7G|HD%|j$2Wp2bs6C0*#%G_4ep+?xmhH1JMPhF+>
zP9X%Yab&VcirbDCrHn~vdfO0<plK{x1RhU6NhAZR#I9LlK|)06&P^L5!G)B~E@YRH
z43V(HGso^fI;W}D1#24@*`Ixt5Oa(fF`Yk;b1gzd>T1q>mbg|*Nqna>*D9&omSys?
z5<v(C!x2YEb=TfXo0azt#Wk{A(?k$;pfnl9Xu!r`L}BJE77Om*yM<P8`Qk<1eCrLA
zR9z)vIXyn4(sS}WXCYv7bBnfZ@IYIQF`36TfnsBq$*f|0bi!zT$kz5In@8JZWyYKD
z+(|n1s`;0)5&Zvp2_?wM_`Vmv2YDPgKfH`?a+FX+Q}L0#t4x{&wF}&Ryw6)-e~WkS
zKcHt69G-A7-=wGeRBcO?4y7aWc8V1svy9Q@O-#|l#+H7+q_Ii-aed<kTU(b|ucyTL
zE<1y5K78(ZPL9SrZY{O(+?qaSw0(i!_v3#QlNEgP{%xepP)Ll?^oxul&yY&9m`6VM
z{73QGfdBfx`m3~+M-`431tXJFuJ^e<G+cW00Pfu5lb^iC*2N9%d_iVr>{qX{Xe`Im
zQ*=K=4>gDK9ggaB9B*BVL@;H>oH=v9p!E)AHP(8pORdqWs_2zHthET2kRq`IpQg4`
zim`5j1tA&F9M82A0Gx<~Oe3^L2ZgecV3JDfbUa2%$)c*beCaA7NPK{%wMZ#2TH>1-
zzFpv4n?xPN6;Z+2hGq8pqel<1!vWFG`L$p9M|||f57XB@<|BnZo-!J3(CZZ_4N3~M
z5_lqwZP~kcnGb*b)BMI)ejX(?Ay`~1m`{>LfF|LdR7gg{-ufCtVbD_JT+4%Z@3ObM
zOVw2DA3RQY7oe>d93SkXm7!Ob6h*=I)(*})YG-kwMu%j}$ch~f506n8)TqbaxpOS4
z0jIMW$FmAAJi5$F@+?O;mfl7|FY^dDW8r2jnlZIq&}fIy1y`TH!1$fF_|30<g?T&U
z`t}ANI(LqB6L7|IIy+!8AERU-F9dC{96lH`D%N=V{8a|@aey5<Hb#!Dg{rNnHWFZA
zG}u5)OCHviN&OVR_S=7$@!~VM&_hWmdxn{n9fxJP&x^$0Bbk<To=$sr?p)>0o%3vO
ze=pzreGQ-d%-3n%5uu%MGJTgSyo=5kRC-FYupDF~Jm5x=);pXB+J=6achEM7ZOUq%
zHIBNq2qjafix9*Ne2h5Z@gh|UK8q~kJeiC-tcTW`JW8SnL=h<l1%9TO%|A;5dn9OF
z>;`0tCxbRR3{D77pm30QQmFJ=bV0G*h8k&F!@G}m&{;%y&&JjUZuSfZ-Xc}H>22Gl
zFs#sFRZK?cD6L`{&9QtClFXb+CNN{d8bkz+V4o(J8`yA)F`8ahBtvCu>Gygp7K<c3
z&LnH25oM9nG;Q+H7@d;mq`=yiqoW1e+uQ8zokMD!2A(k@Me0Ex>>n^Hhb*cY)5VnW
zWI{P8P?^Np_Fu{-@c++M76qBn2qRhB$;nhjEE;A_L+z$4e1+En)9-VA^8#P~);GBM
z)+4m^yl~+h!|euBuHn^;N8?A>wnocHUP?}9GiLiW=QqxAZTAwYF<21@yOzBjD6OMB
zNC+QODt52!(ME~1b57>)?1!G?kNm_>&>QusYscpKYaHHr4IvdWq|h?&VP3av?<OwN
z^Upuc*0~LS`Ja3lqig=)ANUbI^=%*Lt6%>OT<$PsYGIx5kRx+UEynCuie5gT)gHSZ
z@B*5qr3fi)GQ_0piA}^u!DP|mV*n)xu_fx{Zw)cv@KmNoWP;WQB5oDm>b$2g7L!MU
zNSPNsQNTM(8m%HzB}N;pv&<KB_PW;962B8;<a9h?&>yaF=&g11@(j?)jftYJTGpam
zRqm!#h&HWHAAC2ngu0sW?)|%5yZV@L;#gnbq~Bkssv1UXB})(kLab;UAN$xR`0|&2
zo-EI)7jvR*kWM34#z{#bHCdJ=f^(+n^?HoQWAZ#B&ojz?!QtUPf<*9+ouEKTkuWlK
z&CbqFs=lfP^V_#jq-W23HfR6wKAFlGPmURn$JA{_Z`kLgnn7sMxlHiCU>wJBpQGs=
zrn4!njMTQGYAv%xP1RTiWuN2mg4!i>OC#sh{V{j#0W)7yXO2cU)J=sZ#mM$9XS8<A
zecqv-w#?d!h({}d6cMEaP3EX;$w_sd<Kt&}{jJ~4!Q)RL;wG0a8?quMu8o%XP63c^
zmV`}D09ToC>oJlS8A4fta~vJ!{K`MQPF?l+<ahidoDiIfmZSQF=uU{T!JEK-J;C|F
zXk(p4s1Zt1^fGL$aM2NBz(=3*mz5-zfQ;xY1DWzXWQgRl1kyuZBhb7zM_2&Sbq}~8
z&@p1{b>`FOyUI$eO5peKY~m8nEZ>v#w*(d@Ld`OaDODySCgs${D^Jnp7g)H!$s@zj
zVV)?*-4$XWct=Eb`%tPmme_<9N9PRZDv^j?+RP#tB`C9Pq%FvDFfxH8mCo4M*x>Pl
zhlwJP=VV!y43OHeSfq^ZrF*z(8@%)6dCuwSX>uKxCElj~b^_z4OpN>Y?x8cqteR2R
z71lRM=`lLe^1*3e0c0wH0T-s?;q(Fbng!#<jA$g&#f;iIX7h@GAlEsMXVcEx6ln7~
z<NPs`>NeBZ&<s6QyTCh(iIIH0pjZ=3;yyR+obh~;R7ga$j6^NiRVhMt+)lXPobb4=
zIhj}7zk7@8&wP?P3T)%(4K%S-R6(#_4p6lt27z;m)4Aiu?s;^ka7J+M+67LkF|wKS
z)vtY>&wlo^JoEI2`0aPTiSHh$j~2(=t0kzK7$xoW08<oXWln8dRG|o2M(|BXcJ?IT
zlEJ4eD}+jwJ3^$nXM{raR*FO=(jjt5i7mQn9B<P*0~9Kw-!Hr0hD0IBw8r@s7jxFu
z)>gz?tu^cG>-7750P-xS&9jcOEeRNW2Lg50uG&@t*txVxQCWEFVUonh^ehlUQ&m$Y
z(-UN=7>?FEKw-zT^59%R=oDX}wPI~FVs~c;qGK_iPzuHKmoIW&2EuHLF%pBuNJ+ow
zQRW2)2L}{IK~WTm!`HSbrD>XabuTKTna>xiHTU@OW9qu0YAR%>AQM1l3{Bl|`}S>=
zf&HTcbf#$PDyee;Es<neapLE^wf_d=*#W^os{)J05`rKvb1E^T)~AGihM$6Skwsi^
zb9|Gca7=AQBRoypq7c|NVr0Quf52!kVp2~ynoOu%lO(RrQDzx}NE2ru49(H+;`TRw
zlB35BN_wQ4QPorO-e%`6mGt@pf~^t35q)}~tIUw~`Usiw0FpdTqyaGvO*7`}U%kf3
z$!Gc87rum}Pi1q0C<!qc^9qyli3J8N3W5tPLQ46Fp)>O3S#r`Eo6JiBQarT)9}zJP
zxt({o2wDhAksQWi$-e?@Ch?*Zng&B|4tVxs|B$!7@jG#iN!viDh)Oiu^ebI%7RwD|
z71-1bM^`k?^o2>#0Z;2&HZP7Cguwl`o}#UD1XWt6BH`qlwx-wbujsiewonIei!r9W
zV|dR7;yVLlZX{)Xp52XaLvH>C-j>Wy7F@k}h3i)@bN|jAl+r6k*mO2sg-|V%9g}J-
znVyz6w{wYsHSG0htxYO1A9(cO0l`I-kyLGkj~<;#6sdaa_xsF0K%0mW5@Qrc^8@ZZ
ze2rN<MiiQP-D2ASNc2EcH*<QXB=i(+0a8Yen?ssui%>07*U;i=+XgK(O&u8)1BSgm
z{r-T*lLJl{Gg{w56h!aH%YwxuJwWExa{l>`@b#B|k;9`Sl$Ct-<uCJK{|CAL+zpyg
zarK#N*rRWUqdVw%NxP`o*gQvm;Ubq__!w<W$BMmk=kWnt+w#iG-{S4J?{<>p3@H=P
z(8hw9lgMaLD2iT&(i(|G`i2-`;sfza@R>0NB_cZWppriYnX2k#>pDU@*pU=&8Hx~+
zykCm6He01oGC&YSCGje?TTf<3%w|VO*P?x(Ulc14DkS_)S(fyBJ-mx7>KV=@RFUZQ
z$od5?I1rxbYcl0ACat~oy*|Ts`4AS}8WSKQufOpvo_X#CE^l6^w^Vp#2B4GRTLc2F
z3{ok|vgG3Vi$obXK6%VmujIp5uCOsVMF_#g3ws<kf_CBf<i|e2!@GA@%sn9lRaJom
zlWDxyE62K4hRl?!O+_iS;tBThJeew$qMpq;**{9=K^a)o3#3l%SS8cYMoU5(<Yo#{
zYK|v!S|6}TCW{5pLk~^S3rzeJ;sDeraadGfs%pm3BXW&LQ|3Li1K*Q47x4m4784o~
z5Tl%4C=kfh_K68`k;MWb20VE6bIc|hA=By0=w8Z1Ze0qGmy+Sy8ud(oYaj$dgskkN
zvJ&eY%A}i9#3yDODHW5+5w~u>$j6UgMXEYQ@sbfS1fiIDPo@keD5A&-BH(?ys&ce3
zh@c2QIm~^C2qp2I+}22qk_u~W`Z`_kO%xiP$&{IqrPh$L0wFv>TYL;Os^P*%{s)E|
zzr~mS;ZGvNkl>R;{K?PI<sfhtW55dCAhT4NXsvM05@Rx31`gTS`z5Sc(8kDgdXA=w
zDX6M@xH)HOnuem+Lw8luN-7+m)F7SZB+Y0<`u8ZUz^7B1l8R@a{v_+yYTkJ35%c2(
z{i4TLf9tDVeoDq<GQnDl$x^`BvYHAZBut4HPvCJ~)#ciRTIm&Kk|;aJd^$%ejgD%S
z0_U6qA<1+}<vv(IED=2+Sh7Jutv!d;9D-(UE%RA}R*Jq-=(6R+E{I|sD2XCaD@Q{T
z9E98;L`FaBqXOts5`CnNo@#bND<fuofc6n7lfkkjLdbzbgzWL)v1L55_@=?to=<(^
z+vpc-Ji33(_;kwEi`TgNg&p4b(m!O_tT7wU5al*EKKrjx={~2^Ia~RFo+=PFVf0N_
z@bJMtN5=<z>T};g{D*%x{oFyKC=sd2#LP!BZBS7Vl<343ElDA!P$B~_2m~amLnH}&
zNVAPxci9-uQR*xahmlC7Q!9d$!{c2*dr1(MXdJDM_%5g8?RQ_}Tk(Xe`5taMXY1(=
ztaBKx84L!i{GJdaqKku_&E`O4XLpASm-cwPe;<1=rK!gp9^B#N^ni81ii$R*_FX#E
zspLxvfj8fJm4l;4oV#*~yeQK8Ig_Ka%=;i3lvGHmFj>ZEZH<1v$A#@}p1F8|>|3v3
zjN;iFPjmX#T}&{Hk50IG^CmGSD)`Y+@`~t8ty<^l`*<1`C~}p80^1g~%$M)_XuYz;
zInQFTps5>_GMHSVjmG*;Dlasd&d~R!uW=F!6fOPj5$DQl%wtOvEL(f&n@Snb+9PG4
zK9)Q>Y7tVhv9ZnY{3cfV)VB|yq9C>wCj~BgbV#91Jq<_=t?BYUB0;2J_6!j14eH4=
z)Xg+;jYt&9AQV;A;+@0$K(CjPYm;Iyg+M5SLvlKqA%!5%vlRJh3{6`hh>TB9+1%Pf
ztAe-Q&RM^7z`U8@>=bDfQDih0v`>;?tu=xa>e4ThfhhGx9YSiN%xSGdMTv`Dppr=8
zS~13?o)Ur()9`)ii-434SjaLx45TIE>6JZ<1SQ`^>zI1L&Y?|~p4Z*xA-YXBbPv{L
zRpgx~IFA>;d(5Y_xyA>Io}*_RZK2qIRB?Lz9O=B<ZA2tD>ywnlCy|-U)>VvWWMAD#
z07pT%zFkU@`n6I55z`quZ{NPmCGjrzZtp{s93MR9@aO=oHNAchsnW?PkT%|azqdlP
z2n5ch&9W$pgrx{+QE6>BIXPjmm@^m*xc>ChXp>R)3QVRuqhkufl0tFx;35ChttZ%r
zf7REm(^uJV6e8AaT|9@VOBT`6kD*hoIgE)&>Ci%PaMRKprVNI2=XPLF;bnLF4qd8o
zYq44qoke#sa|0tFXmAD>Jwc|Z+@5Ig5uSeP8Qy&T7DtDNl%gTv=xuCs@1*7K-2;w~
z4)~RqUg!J2`#C6w96i2|)H!Pxukc%UkNL(s-{i`LJ+58ZBW9bZUY}<1h||Z9_~Orh
zk)QZef0C_@b!^wBo!1j4vx;08T+p;3l81m+GO^V{K*dN|<VgS$0v?dc;Dp9mi!z$l
zT4IEGd;%$59x9y|xXuGa1m_*mMRM6A2uqL-n;RBgW#C0*W7MNPI-#CTD2f~@Bzazv
z=j951+Sr7LAB+YJ*ZLgnKjQJ@$27J<Wl+_pJbqZy_!%GZUH(ZXIk=y=SdlJHRtTa}
zNU3@J_<)bgjH2w3<%S^A;JghYQLRI_#swe+gTat#HD*xe>~3waF(@Ghg7-YW`+$cJ
z54pT^p1ZeiQa2UD;gEj6ze<1g1j?*Wuh&aKa!Rpu2pDb1ixe^8oFmV2ymy?OoYFK6
zJ3Bimccqj5N-bF?uVz`EHtAO$zC@90bP)9O5{#zuEkSvV=`y{80I%p}C9R!e+m;v%
zi|GP6j>Igk@-4JZy~@U=8zD%EB%nM9XOS^=>VtE55mNeGG#uT%L~PE{%a3X51jBQ|
z;}8gKQrinSna)`&GxEH^wH9SErge)$VYH_4U_(ThJe^^pLo12WhA0%XW{Xl4<fFAz
z!YYB3hOCpYDwVR|1%jY+g70#Neaf=-!4pJ^v#hM6X_{ox3n^b=*^&|6^f|aL>a{Do
zwbC>$mF`8J6M`aG$=q4usqgxtNnI96I`8E#y(<UOm1gDLNhu&z3Y-x{8bODbmPD6f
z=R3%?oAmpJ7eDbbhqnjpKYE&!+n1tXvpmD7&X>|T&m(k*kSE_%bd!7`<q8(L?BaoA
zG25iaHZMHChfoF`TFzZKN9+KTCNs%lPedAx(FP#}MyF~^2%X|QyG=Vu9VG_`hnTVe
zBInPa=i<eS<az2{#|YEul-X<sNSw9Q?t|q;|Hl2_AkQRO(3E9?l~DN>slgaW1gsNe
zS%FZJx|!j<LP*8*bcT)=VV*<-1f_^l(%3|E_5?H_TAzCIF%pBr`9M&S(9Hv)i+uVc
zpW^XLud^7>v7zDq<b)3mcM#j@LHzLen7{fn|Cm4i*%y$*V_K!zHv_)(=0ntQo6~HI
zS01()-AC@6qdB^Zs^(m~at)WSF}U<Jz3p89bWxJ!ic%;-(BwYNJ(ZFWQ;RD4jNlbM
zkRoXTTRVhS#GtT?wnMZBTu2Y*C3)9qjdy7dx15ndM4OC7JHu#$78%w_qE81Q+c-*N
z81w~MnHXvt8(Z`T1AIr(4k3c>2);5hpU()vv$?T}kAkW>rS%QbYiKO76K)6sqfu(#
z6`7(`0pTnt#no%q+1cLV;)N>=27}ZRmS6+vx#iQrZ0UQFS&FqdJUpaVln@<Krg^16
zuszu3Q{Q%-jEKhb)H6?~hr7wpCLPGU0F$MPX4#=whQOWW)}{a8{{8zDg8^DAZrr%R
zXf&eVA21jUXxoO#WWr)mp;g9WK1Vm!p-M^XJhOS7!hW<O7f+m*OX8y-u$Wh=e922<
zOgk4cwM&E$6h+=WKm);AFkM<$>o8g(MS~JavL=)v0)<@XarF^|%rRL(Q&oV(`<MWH
zI_=abh0Y6Xh{+`?I>&iPFQ|}u1^g>|eHODBu??6kLuk$1If&oJrAyDVzyAhO<mi}A
zGuHT3AdS(6rfKMxBcw>Nn@X2y;fhFG`LtteO(7&sXe=HZ9Lj05$y52NL1w<2T`WsU
zg;F{h-b9yKEG5Bda#M4<b&G>XpG=!g*AGs}lF(&`2$WV`&ot$iYt=n;g#-~(E6sb9
z2-J1PVtScf^$p@$$&KIfTfF+^OH7YHN(c$CmlZjKUXLu#kR3K<g)muNsUBHjP5$4^
z{a3IpS(@L6{g%nh?an^_+PbE0)h*Xu)l=<zT#o^UFq{EFiXaIQQZRy|Q%EmjDEOck
zJt#sU27?TRL{U`eK_V3bLJ&v_W&m(7n8EbSbWeL*-CeHhR$YJmIooCCT1yY#TA8``
zIR%75WX#QqTj!i@(yaB_|L44Cu~<lEQ4Bt6>)AUVGb?+1`WN4za$sB~LZTDDk<pmq
z!K9q~G&hByJC*kBwXa=$u1SxcPN&h0)=Vcd`@y?Fk@rB!%;w>L4@%(wpQ8*8rC?S~
zIIgGY%%ZKKXDww}wK1pQag}B^o0DZ*7_0FvP*pXGR3K`xHm58r2o6_yvOFWWn&3g3
z8l!8phMqAvqgb1#{$xFNFTKF##tmjq5Ab#1{^MgFADmEXSmYVQ%U3CkX0I-I{_dO1
z>Oi@Encm4BdboiZZK6~lXvG^}`gMNi-S;U*IoF=Q#jL;1H=k7e^OtV`ur}CXXRS|B
zOCm~H*Hm>)ZZkqGm>mws)U{_(FJfC8nM%5t2ygaGPG4&otQ8#2d2*eh@<NQJUX96P
zAX}&Iv;|7_8?U-9EQ3PUpr2cg>lt2W1Z}u<<tn?EE|FQwWGYbMsMlk(vCd?2#^J#}
znTD%ZuQ42~QPz&h{0f=rGoP1CCQ~M7Q#8;weTG{StbgwEHlN<z<oM}6wdwKsH(%#-
zpZyfuJ6rVoJr-q+(+Q$;fRC^l<b?qRd;15x`~C;qy?L3*OqwJHq$~5~D_6Pl!po?#
z!~}=8ra>%Ha9){qoOI$pB{{2&G1BGpn>TJZ&cJ*=m$+(BoSmLYf<w^siay3zYFA^6
zeu1%CW@)=vQ{V$AgHk%y2bd<n<efq(OK>tBa4;BBn`v`1A=DTHz06{*!YWO_=+nz`
z5vtyx&=@RAi4}3W*kJF;o7BO9wRmk%K?-JS=g4gqGZiGbqk+;vUi2wziGzp$rAc+|
z_51YtL#E}Nyy&qgD=P0fc&yp9x7oR6(0-1l!h1NI2w@YXSjZTQ(E(Riut6i>dPR@L
zWPu`9$fg22=c!{nqFgLw?sKnTQP<M93e`YQC{ch3m#k|^2pF3MAAURV@V=J-eelc|
zi)IgG^JxUUz)dvmN3>GtEGI8|)K!i19=yYb&+wU7;RMjLWA68N>176M2Gl-a@|-La
zqVe+uHk0)ZKIRsx&=8^0X}+>VG=1<=+Y{%&S8k2b=9k#mviOjb;|bcJn#{4d1o0e7
zi_PeL%%u-)0ADG^+S(dhTU!)G!SV4iPoF-eD0~MN1Z!g7vDi_yT$C_;wp2)-S#+kU
zyay$9Gev|3Q@f0$K!OLEuVQeuN9&CKK;u-|!i%9M%QY$p+-j9!*dJrHK?j2l3WY(d
zj2gvIouGk}dC8&5P&-#RdFL%Qwzt^Yxx(S`0UO)v^m{$(MPPmBDrYFp@;;;84c`6W
z37P4Ohoq$2n8`DK=~sW9umASnp+8rwUE1Y`KY5E{a*b*xg{IG6xl4BAGFi2ts^M%}
zN}se?tDzJaErDMX1)Q~35(H**tPvwh8wrk=acEdw*OF|3LT8rB)iHJvhm$E;*FnFJ
zQkpU|Tz}_}fp@sPeVKce=d>(QS<d69`^;+3^=sG2dIcsE7<)c1adlvGeG6+Kvr_ao
zULSJH-r?4*7x~$bzRx_H<4Q4ze(vs<QD>*XA2TX)zW9k(s7@xF)S5dtpJQWloqHeL
zV|;SL_RcOux#nGsQh~az@V<vJVyG=DsmUm-nu%X9nJw@kkY^c}uU==du}zN0;t6rm
zC#o%R5LyXwRu2syL)$q!%Q7}LHrU+U1mNk@ryL$0((f;i$TcQlZJNAj@y=m~nPo8S
zu?SwW9jY2#ho(1D@;;mw?c=>P?8|K~J*ZqDXc6qhl(m42c+HGK1?cBJF7IAp*dH;<
zH@LiWmGN*BgT*NL=}))$%K!TsKGdKM%34BQF&GY|ZonF}GE}wXROeQP)VSbd_f3Ia
zv{5PF6o6i?nJPG&O&MrSPQ~dH!^>Nr;-yz!psE#h@Klv!Q6_*U$jpln2)?9T%o=NZ
zzL-;FIr)xhgg-!@=O~i_in^|1#)hY=YG$*l2}GwXDiec;d7jbl^*K2`;pvmdROKPn
z!ZbPI+L+kBoksHdI8`jq@@O3dVH%|QB1+Ng^_fkk1TPlqd+%-W@V#Ts6foImax|BI
zUDl#p0Htx>qjijBkS-aaEEfcu4jyZBGHWo=lk&cnSXKzpQ_}Q@Tl5Bwytl#7D4Z5}
zFA=QwqI9G-#>cft!BHO}1Zn9qnar3@A2ffvn0Y4-X}nj{R^clJbyfd&BJ}^eJVyP3
zY?w3ei`VEEHCDw)R8*GEIf>&06W2wc2IoSAR7BwlY#FCgD)C5lfbnRZ?Tu~v`H1zw
zHdl79(#r-YUC_J#6!V|{V}ggH$&B%ptCU&B?&Yg|`i(cph6Q)8?NDv5v42vsy<V_g
zz|(slu-+@Uee(vp-e8NI{xGKw4pUg>Cx+)P4S4&9M||yj-{a=Zn_M|KB0G~H?&n|n
z94~+R4g6xxqIA?%o84DewIndsQq?U(ljc-a&7fBZjb>Rh206=eh!ZU9x|XS;-Z7ua
ztfItclG3k{OhoBc8wP#W^0xt~rkYuy(E*&+Z0~MU42IOf(=T#qnKL<~=#98}^ER0^
zxN<JWB!Qyequ1|q=k_bS|HB_JpU-g7_}qVZ21in`i1Tdpdi07Bk54N=vv+jN&h8e2
z!7yf>iFfWpU~)QRIz6RV1g>7$;L@dS);G5(218CKGjd%Ml+=hAr5LV_F>7nA_j(js
zk|k0hW#R{obhSvch_Wn?>*t#`E#3!?j!rl{JZ@ScNyb|n4^4;GE-Dj&v8}FZnZu}p
zG|6%Si=wP6P>Q;gc!V|vt+P0oU16H^PX)@wNm+nODW&4g5a9GwdLbE|VZCB~w8^Ed
zD-^a*KkMU;b1rT4IXZYjW*#VU)>RPfc0P6FR^}fp>JsvdLBB`vfwC%@FG{h#jm)f&
z2h|uAqIK&CbxC1CRT<y-C&%d7TYU6*&e3#6(aSMbJP$xx6rdm&f){J2t{ulG$K;tM
z&rQPw)>wwaA<9_hWyPX1%C;&?%Gs<@9C^h5wUy?Ey<U&;c+B439{>0s{XTAbOjQj~
z`b~VDEPgG_HJ``23&DRGNV=gsZp>r~9_L~c#Vlh!tIz~Om@zrs=WyY0XL~H7QcMa^
z8-t0ltE9bDjIpGcizwF6SdS(aM*<BDh0<7a!fg6Y?tO5D>o*F>3{FYFGbv_3<}xc`
zZi!`#8$vG2N}4w!3R7#{SXE;E$}H!lhVx2Af)8l(Y-L~pP*zYV@RIOgQE2ZdJ!tJI
zeZs#D8jICd`dul-VzI!Mpm1baYgLNZEdf=n;K}1h>^<3+8V4-9+gI2aZIR^zb`IwJ
zd<aYr_NhL)$JJiWSX**{#f&HKzl}Y4%4ToC$>D-~r%(BnU;H`lJbxYM0{7m2n}fp>
z2BSWMVL?`8D61)Q&E`;{@SGh!X0&CvGtL;+()#0_AH2i&9-iRJioCGG-!VC&s_Hly
zw4K;lmL(@Ar;J8@lomMFdqE=x!vVF2MOiYBy{VMeET&UjC6(e@8%ej1vZ$_W#^W&z
zh72A{aQj=|e2c@+zRF;&&)L}lCbxK{`SIItb8>RR-rgS8ST5~eVSR1P%^S~=={1z{
zfam!1j8Ma^n=dfvuW>fr!+VjFXUB)s2S>8yWyy3hrI)V}f@bT|HS+Z>#L#qPUIcM~
z=pl$FvVU;M<x68mqcyCtXr)<{b1ECKsU9YHKK$?jLHFogy-bOth9((?l-}<lGnS%P
zG}A$6vpL5{CoJX?w{hMRe9Jcp21^;9jX$Ge6;`a}Hb!GT4z!0wNns2sTJfe9i%%(q
zcajzmoFmhU%+4@Y;Z;rP6qEsQ1RU0i)X8!g_LPdxyz^9EGD#kvP1&11CIo@QM!>b3
zns?rQ4&`f%t8m`2|MZE>dO$^7TrBtX@fxF1AFmYUq~ghwC-~rLitqvvY7u87502cI
zvDLU@|3EXT9Q*T9dQz#Hp6*jSDGaJ>hp`r;70M_wlY_1(`XcQ1_x1(uLfl%N^%<?@
z^sJ@V7u<C+nP99T&rMVCl;xRNl6gjNAT<|L9|%Q{-REvoPEWC%@${%9)SAV7!Rch$
zG;#~^xvVyvo}7{wIeC`Fs?&lj%cEBi$nzny=`ocz^ag9Z_R43N2TN|}oSoj}dq1oR
zUS@yv1_P8W8d>Lkz$CX92uh<ccxuYZVU09#G+H-xAo4v3{t>rdyv6e`-lleQ7T!^p
zC9xA^qy;DNS|s?%WWwR$A*pq}F`2-mUV+J_<FN8^R!KYYu$ev=A=U)zS(`C@)}kQB
z6GDY9Yx0!k=RGFk${J~LWvp0?MW#^tG{Vb<(q-sUjJ!ZimKo|=rc7!Zv*Cimxe^}&
zl@Cn!C*0eAAFnN!jw}9E?<vA5gGXOwao`x&Cw%zsTl}a0;s2Ux7*K^iSw6x$sLC0a
zHvWD7++X-*PL3=7;1B<hzx~^PpP&M{%`r-&L!j3$@bwAXCh*+!9#^NHyuQ{n^gJmY
zhl>hZPSCoCGJT1Kl$97DR-=s)!!)zh-Z31HKq>b24@ljo`@NE4uqH)zdCp)sq%0-4
zZnVOuqE;oT<vvJ@m%h<BZ74%PMQ?t%yF)P=(es{O4gcg1zQWhP{}w;{*#}f*iSv&2
z^$kW@&OiR0-{JamSGjfjIbL}24tH)p&y_1T=nvP}*fG5D${YO7@BST0+R%!ZUw8w3
z=?wV0%ze$IT+piw{q5(t`^Kl4)H#Eq&ukuh=7k|PR|F?bF6Vl|+36YY-#cP)Hldz5
z#=Rbk(}O6u1#z%8<K<W1V7@*ggo?TI_^K4-qKj4p$>^-=iqq3mo;-Pi%oDQM%t*{k
z+nAOg8VA6q_7*{UX(>RPw#&A)7HcxBE;4LpFs9#F@je8!%Az&{=a2xnQDj&9to1TN
z6<C-G)2}4L-_$6`11=VlN1PWutSMsPCXi=fW6gx5k2#dKOr~pS6_5sr0-NOhoT4`b
zWn!~G!R3w79G^`wMw1WvZ0+ncgNy_wku0R?*(r-^PVf%Ai?tz+y@ShKzVsZ!-2#Ot
zHvy%s2*3`O-oPzd$7*d+Qj^K7gA1A%&&WxGn9_U;7rW^aUCV^=Cb&x}3Hljr_|;$j
z3uHF1|LBPS;NP!VEHu;ElzCM*0e6+siM@Clv&DkcY13#!mS+@spTTg0f9_Wgc>M5y
zfB28D((Cnk;f0r}bitrs@$GMoBSZoii?KP%80xB)VkTvTsZF%}Rcf1JTX|?~>;wu%
zDU6aVm!0h?ckf<hZF3VvNj6$%*q1th#bUwX;UQHe)>=0MME;JQ1!8y7;8XEfz=yh7
z^X|;`be_gqicrwAb{K0YVy0V4$p#Nf_R_lmL1QzCYh8POo$9#8`5HfT*usz*8<7IB
z6utLMr!$=Q^m+q~&ImzabRf$mo5yHFg+fhZvI35^+@Z{Z@klc*J)ip0n`~Tp5mRh1
z-rU6~LtT{|xxnB5%D1?2<2qmd%|FlR#$`hA3<g82$#9k9@abc|^7sB5++u;7o)Kmz
zuoySZ6mL9t9s7yb$*LKJwYVVk)b5drN<wHLz3!Z6Hj_fPgwqcO11SZFbx}==b(LzP
zd?N$2HiU?lpcJ9s=Y<a*0soMH@n8Cv`P$Bidtdnq)6(;u-~R*Ne<<FGj-w6ErZXzz
z*!$s6_``308<b--9J0B&!Oa^l@X0qm$FKhKm$~`!8>p{-g?V)kl;+m0JDAwSGx$JN
zIZmn>S6<p?xW0)hEsq~O;N<v}t*vc#cE=2c8O~|)tYFX|N;+)`MQ_0RrEPle_h6#g
z$TQY#CPl`^vbD3rV%*0X=#97NWzv~<I-PQKbkxvDTH6$-PHQFMI77=yGUTqyV<AYH
zMcRkt;TU5CVlOgHQAB7(8$zsWP}*Uv0<?GvXo7ZR1&oFTzV<8#RFvegvt$a4hG6h`
z2`syScT(V$S&diV>Jq3iMrtg)t2x+vP8KG*?Mgu}FIZa}VQh}E87a@ZD@cr1tdG~4
ziDO*_qJ2$0pF~`*mhR}frmipX+Dn!&7>F<CWd?P!ic`iy$}UQTEsGiwa?{|2@>7F~
z!>wY)v?Jhjr|?hfSV^7MH8lOK@BytIyH`C|u1$D)FlJHK-5goc0{9?SLh>a;fZBVi
z*@BbPieG$9@o)Y+2Ymf&8UNsSrB2;sIl<)EOmNT;1R+k)RSJrWm*;S`ju~{wz^kC5
zcPgVpO=bnQM6os+G3fP}FG{XmQ*7+)Fy7u|bFELvhLq);ql0}ZR*aBUm#}?j-NgU_
zkfKegLGc6}s|sXQnMjR<RNy~jF?5Q;(kpUQA>Oq24(BSAPHEfX9V-RF)a1ElywRts
zD;8>w3l*7(Fe5q|4h5C3VwRSpP7p(`@C2%Ij@Fit<+vyqG=a=l^t59*)btd5>9e2V
zh1WjA#?B4a$Lkms1S(pTvF5__%I%xnz44r+<Q5sd+_G3y>_2|O{##$?Cr^LM>B%vZ
zlOE+s4_AutfBD51xbfy^Fx5=rWwad>mW_wiJq-;4B0T1B9$cN0RvfO5z0(LXm`zU9
zq`fPjg22FFK%Nf(u*HDQD>r%94tRVt<^JP6Ocp)x1RKZXoS-cs9|D21T5<a5h@U<9
z7C-)ApU?f`uW;qL+gyM7H5Om}ww&8$=vV;kJyZ@Vld*N}Ic8;rt5xhZ=Q%t)<m7mZ
zot<q$fO4S-gA7H^>FJEwykh(MWt4`UwKe{QU;Q$>qcy6942Lqg<=S&s85TXf=`lGw
z;^_1g=NzNas9DoC!6~fpm}=tM^0#Z|#;@+j>bXD$X4xVy$a70oR!sq%wOYoHIu~(e
zO<j319cR5|HQELW45rQqp+*~lsU(kE5wu2SIkgMqxg`XJR}vg77ZtUu>G!0?^x0&S
zlf$bR6|uV@_;xv)QTO`z$`Nc8JH$rhJBWr?2qaQDNrTHXOYjb(mCTY*axTGG6@_(t
z@L|q-?`rP8Ua_bps|zLRkxibxZ|7<xt46Tcj8s^cK2z5<Rb8@}&pACkr6_V4<?1{-
z72!#xo4`UUS=I@fi<_!6z4ZYvzZUrF*B7F|n~aLYAn~u1jukB|xzB)7hN3q`_bpd$
z&Ka+l)Xt!-46{K6YUfeffKudn5odwOeJHJ>K&r?xL!JwiM}m%m!e$v|Jx3u0V1r&?
z{7WC$*}hLUxWZ(1MmamiF$CrCu3S>eO=W#Y0d$F5DGOG`wRQyjIf70WQI|oCF)=bS
z`u|<Z|Mz(0MUJ&GLo8;%WSNnDtbJ_jE%0HcG?+lJ6($SxLMADcs+N@0%E`JYKwFei
zxFAUpAt-{<RJCI^D=``@8MPCdF=+ByF`3O#p~js}dH>trWwY31a{mCgdztZQfH#&<
zYurLSKjRF0Po7XOrVR4Htt%VM51+8p%eb|!d6;>EuQ@n8Vme(HJPuIK5nLT7Vnzry
zW;1y&O0ABasuMkk$6*^ot&NXVEL_RiWWi)Ik%@rDHgh!q0!rp}>r^Tm<FSp`2Oh=N
zwc9*6oN@2*0j3y`;PC_%`=F?TbQum1yFwU@_2k;HsB7*$e8ko3ud#dMIcfT+K-(N;
zr4zI<88#n~_tv<4>o&z`oxHTnCZ`ld!P(iA5AJ=)<*Q@X)<o%KMpKrKtCvTh$NcW^
zO<3>u`GwcN$c-zPF(JwXtqEFjbaKe4Q`D-53X7HpB44FMlVt_cP1Cp0C|7`_T6dKP
zu=IQSX^dq)n_(+o;k;@zk8{#OKxDaCrW(*DN4X#q?i5t6BGjOR!D-1b5G*=b4u&kZ
zRAtS4QDTkE+^$QHu|4pP*+Npo<|iYjXPc4(m4dja7*IPwhA3$nqo^PtcyCJAi%-V(
z_6GQxa(c#a&|{Dr=Dh-IE#+)VK>AdLS&1(_%BzSwOCUKwV-=-;RaM6xQzG;hi<0S7
zrh`3x{FudJ!E83e#5C9sKm3qaUU`K#UVj4<WGzF|cp%D5c|z(G5}_Fl_@FUaj~8C5
z#fzfts+-c2+rV}@Um}KfDfagEnJ+Y#FK29SWjx)>namu$jes&%+ABmwtc;cp5RyeN
z87tmXmZb!5t!WgQEVpEAsUYzm>XPZ?j6r`u2!~wSb?B^zucuAiLm-E?Wl5K)fhN;T
z2)s2y(m$25m7rOV^xZ7WmRxIVEjF{yn*NZS`4A+r1gPptnt%ELXcluR!qsSW?J=2b
zbX8a26*?$_iA^W1rmjSpo6Lf?)K$%FQIcl`^MwS}jU7m~&D^sn5e?!%SBiR86P)n<
zcfR{A9_}AwvmwL0M=!Tj+TdM|tE3;9&7i71S9dn}pZ%MEg%7^}Rf_tMt1sN(7e8?m
zef&vs1M~fd%xaqX89D@lbKpEC(@TQ9?XJ)nos7`PY&PTUY{J8b4>>+QW;UBis&8hQ
z&1O7&_>kZDjo)BBU&ksFOSd6%jjVa>Pz~V1f`BNspfvf~Ce99-drcjXqUWH*kWzpS
z;Oj_%2GztM0~*Eg;Q>GV!T0&%=U?T{3wJ1bJqV|0YXylw5<Zm9z+{B1zy(dU@NBGa
zaQ)gP_V<sufB!!3yz?_Y{h1e8UpHv280$W+hNn+Y*ys<rzP-)kd-wR!<e0lJy_if%
zDYV4X2oq?ONBTXjXR5VIMw-ZJi4lj?KvC%i;puuBsf3};(2^KPo@ZF_;5}BStfGXk
z8dAR_362`6ybJ;5V$Db`%j;||H=z_>NwK@DW#Eaco%CSV3au<v2!smi8CYc*4l)i8
zK0&qUqr;*}P3-jsEMhvRHL(I(OZx*E^%`9&044ojMx!CFu2{^J%=svCw$|4q*gGh&
zRxrY{tXV8(Xxqb(3)~cBjAcrNmB5o&J(J0V<KtsaPEVLlr_84_PR~x+d%DMDGNI^6
z0Q<uaKV)rfjR*HX<oeYsF{qulJ#P?!hnAv0tx|3_=vaMj2=s@!w0n;XA)DLP8<JfX
zBBa$8p+N{PrU@>nXA>^%9P`#&H+cI;j!*yc89q~}7_R|QAkv~my4=cyLzxcgoMXPI
zur4Z`5P%w^4SAlCTgzfTrSzUW57f*V?g|8Hb&mI<c*Qe{#z?d><(XO{O@B*)lVrPu
zm^6@Pju_MCrKg{*=b>X}h$6G*S<@de2B#%4Lj{P{_6{i&RaRP*pfsf`$>RR_3UG>`
zt<14NvfqLdr9>?)KBNv!YOGLkA04<kWW9o(%Rxb1*38b%I6Iqg<?3y!;Td!1sH+KM
zTcA~q-57Iu>k2;P93D?`Wr;Nor2@b7<uCBk&)wnd=tJK9#=EQ!GCp<ZCbu?+^m0vI
z&RN^tVEg^Yz*{I?VXR9oGv3z`eiOLdISHsQ7Gf-(o}O}YauTQhopF48%)$Nvi*mu{
z#s>Qbhb+ocx-sAT0DbcY#uO5hiwa5j%_6iE5rhbf1E}E1twIlT=33}Xj8<XIl=Px?
zG<>45YK_JyOI67ry3c<0=Oh+m3i|yageTbGsZ*zGtyu)m)@TR@_Kx=$=R0_(IXs$^
z<ps~(xW&ooF+cvvQ#Ln-s1T^j8JW&l9}ao()-CqGSMc6<-(}0(;e)3~yi&&YK1$(K
zfybd#9Cf|ITdQffD#FxS2f)V!m=NWQkMYjLX9*cdN+`C34*_d6+8S(;Wzx>QmI!<Z
zk{VZQP3=O=fB_#IPCHyE!6~fHs9h~MDTZFJAA`Jxs*?0wt09yMt)#*{qhOG2(6c=T
zqdr%+uk+g7JN)kS6MVJy6h^M20$8I7UePOZAqHg#)}WNZW*Iibo;S`zs3{jSPEJqQ
z7>^mPZ*Z{pl(MR+s|C3=<Pmz(O5vU5JKrDhnO|H`JB14Z5LJ~-9-B-ioSd9+d~(8M
zI_3EInA3<YDj<*ZFy@OH0gv-FAHDy8ci(-FFMQ$iOiri#>}U7b+S;Z!7}D$YMOzz-
zM^X*bKxC!F+LobYl0tpu+CGC(pZTmW!j~jyS1fag!F3&pLYO4a`zTehm`|~KPH?w5
zeB#NhW1%$LJ04?06MUanC0Yf^3NaSvYRb+`jk-#6p-tl%9vvPrpPZqcqsTMn3r8{9
z=cU)5r!E(0ZSZKclJ?vkJ<%Lp7kL$@RWYU|7FJmLPB-QeOJN!~P--On`HTH2A4Mh#
z*GYe8glSFenH&uQVyhfoOD6%O{%KLoSyUcnbAf}stW{M?z+9p6WgyS`m{5=x@~>z5
zjJ8JHzIB(ERS)}r|Kq5H9rM{S8<#FqRtx&;V{YBJ#XzqU*0-4q25fCyVzj=)=5)%*
z!9JmyGCz30H^2J(ymEV&FMRpS%+AjE!$10c`fkGF^a(HAyv&vBm#E5V?2#rA7NMf5
zrNv6^WpwP>*%@bNXPlm%a&&Ui1b(N}DJce76a}u9>780@X7f2e_`weujn;VS?p+={
zc!)6>YvXl#{ekpPi#^q(ULuY{Ahk^wi##)mOa(4o+r=B{qOGA#zz6}OiS0fFRmrj(
zYjbqS==E}X<387(yM}Xcd^BM*K5s#d)3MmjSbXicva^dT7kv2eZMH{Kw#VD_2R-iJ
z|A>RbeH4MmPxjeZ-{9trYiz6yaMhgeeEU26jlc2NsqVjn)r#T97T@~Gd%XJaAusfM
z(JL|pr0aHq|GUSkvDJm>A@PkU(XbMiY|{L-?i3=Xe`4fWUpeYZ+5-r*kwMTbswrh%
zVN8ZbI&R7kB*8Gt*l9NyOW|^CmE*l3)Ee#jY>zfsAFk0WhODn|vAMa!&h`$Mwl7iG
z9{p^<uo%+MGyaSFBl1Es>J`*gN$mr}(Hh>%{9uB`D?_Ph0>Z{vY|I<4=QY#QQ|4!<
zET&Vowzlxva(Z@#3XXD7VizUWnwV2<IXexU>>u&qq-HXmvj6lcCnqOlS;mtmPngYS
zWVyv;ma|hK_LPr4FCBT7ljS|uH;43leQw;i!<T;k7rA`tGNZMTG%w3?eDvb7Slum0
z)XGt8h*d70SMZu5m%d)}h6sH7`Np6SF{C0Eo(v7GHP@dz!nuG}GNdW^nx~KN6Y3tp
z^!engbB6sK=T!7QH8y6hC?kZQiX9U)G@s8{EEY^B6N&q{3S2FjA1dSIwB}2n8nC@>
zn9b%4V{7;XOLbk~t{2i;JLfe_la_6k8DW_@+W1szTVb?5cdkF}qmG58u1=7E!4sHQ
z6TFu-C&;?gu}g746Y4;%mDEtV98>1FTH&44dR$)LV%Qtf@2|17wZq2xHkWoUv$eiS
zrgMhHh<<;K{G*59VfXSSe)(5^k=s|_VSMctue|y)j}MPHnHPArILJ5npsG-7TkPC<
znX7q^2M<2t>HT}$`_Xs!;UE1jKmO_;@^AgS|2EfOdWj$IO_{#`QwB`=*#{qRWp~VI
zXM<n5|A;MMK3(wi@gApV6HZT#dHVRNj60}n9zT8zfXNKTXbuhzr3*9A?+uyHrS;EX
zG-kXurr+=LTfcRKwecD|J3EYqLyDq@$#YzUR8t&Nj>KnD7@b<=C<Ij*4o1kt9xnlY
z-)3<^^!h_5!!H?nI?(G$FDN$F2_ELt3HZSNlSd3klEKl-bJmI;YEe=+$Cb4amp3=C
zy@J88NB@QAI6fIOolZDDK4NnGlt20C9lrgoZ}F2Kzs1SvF^7kb7?u-0clS2$J~`wk
z4<3^D`vmXem<XDV!O$lB$o;D&K3k1{v`9efsipV3M>pZ6tB8y;Sm!;9vLrYyvvyU6
zt308$s0wChrF3r0a<s{CF6LNc$#ll%V8qs7o6&HC+zuFxwz%=!ZLV!!V^~OPp|oib
zYd|^a#bUL^RgT)%T-rIItVUE;aJpV@qjfD=HA*WMWksoFa7^Xpejrd*(nEhR7*Nfp
zoE;y^+=#rO@_|r020caXrqU7`jc35aClC3fKl&z5CKVXP^lZv}K4)!hjlI1+CetZ-
zp3@%;C>IOrIOJ$&=Mrlh+l<F!wzsz_ib4waY#zOeAXv7Nd-V|O7r-S;IkYjH)U{V4
zaDlq^*j$n4GrZ4eYvQ66ADxh9kCnV9!d5=8v3US&lI1=6L&M3$v441qt22}eEGBaZ
z<EDjz3vh6>FA79UYiH>)APE8~*ojK>NutnK3Kt9~XCJV(@r3`;zwtZ*J+*FhOQ5YQ
zoglYvHct1wE)jRrb<>fs-0pXi)oM&TpZZxk>bfG#6-E^pokK9xu0qurzN(om<`Orw
zIZF3f)DDXj7543jjg2if`dbVJV~T9Z=Efy%-Mqu4wH;}mXY!_zDPE(zwEoa&>IKwu
z1zK}*Hs$ba!gDuoGTymDt}Nx;;e+OEQS;;bAMhkEc>mD>_s=S}FJ0#3_=x-WAF+Gm
z7B9c_BJ(F7@qhW>|Hr)gYkwX$+Ty(@Cu|P_YBT4Z2m7d_Q(itf1K@{m{gk&(r<CQ4
z0Gu41NaeiN?CtFlLLe`4ie8`jd;vjbxm><_o&G@D*sO1?lVw>02^06-W-;3-v=d1s
z<2?qNz3^iQQfHAIQy++(y@zOHs1`N4LvJZDL;ent-ZhzZ5?{mNe9e5ZAb1##*731M
z>G*WQXea=xR*HU}v-jj7uj#-a{m$?3!8;GhM(b#6*xP%|Y(C}a@PLE;r_5(ld=PJE
zFdU(*CKyef^?2{_j0YdR#n1inFY(&i5ckcmO9D~o4tmo|{IHtk_;InPZg4$)w`<s~
z@~EUb%lj4`@ovD?fYDSXsCge<d&b!o%4n3bto7E|S>I)MdzbNOgN?OycD8p}A8(Uq
z1zO2aAW)cK(Mp1&-bvS2r4`;tPb?Qav#KQ1j_2<fX0w2yirsM?N;`_dfP5e=md&E#
zC<f+4z%8T>0ckyw8AHF{1Lr8K3a1P~DW(hfbHDy$?%w(ubEhO5$>vy7@%H-@CW|R+
zn<M7)ipfNji1(iDtu01t8*Fc1Vmw}xNqSjMZ!kb-d2FqawHiJK`cshG%SuaHrIy+n
zqQ$I~2)QPOSGKo_6^z#$n_EwK@??`_RV*0=A$lJy3tMR)9a~^1OIED&`lsF^Gf(+*
zfAJ0~(`3rA_wWI;`54Fr_VR)0<V<2JD)rkEC@9UJ2c-EOQmZ4GAcjS)dFj;y{wM#_
zr`&!0F4GfBzn`yUf2bCV?sx~?Y?1TXAuJL7DlxCdQ5wrrsRpWemJXw(4y0O$_dw9N
zGJ}~SSDHbdL(o{0u`%3YXKR;BTe}SVBepiS*xkO&pg$HvPD!seP}mT=8;7<I1}o!M
zQ8G?o<|+zZgHoKFo$<{d{e&ySb+~uH#?@Qg_~fU!e(iZKZQsOw<NF-n`+%}G+<W+t
zqeY3;F!zr45082Ng%=oY^_ZMZxcAeyG22%;Dobi#^NYXs3ruGxRK0?1uLM5ggGW#J
z5QVR%tZ!~`a(W_36-JZiIcsZU*2bIc?Cdfe4#_hqiY*2Mh<&h(F%e$Tt&mJHL6D3>
zL_r2GD2XP~vcQ~PbvzIy@@##5QyQ}B8ltD&ttAq{t1u>$Lg-3Tsa>e4VykzQW@q;r
zc4HfL?>^V>yhNCZ@|>3o9_~M3BhR^XWt08=2YmI9ejk_jF_|R<M^!DT>Jp2?DyY0?
zv5+wjS-&q;^2TydD?a}lf1cm^tN#k&fA_y>AimNXluvxsNoVf}j4q!+D=$`+L^`^B
zmNF*O@5%VnN~4UymXn&I=drp^KVQS@f?_b_(%N+{?ObLw8Z#IS84cH1A8ygNeR_FM
zh!{ofr0TUvQ7t__)GX=>5sJy;l*e<Slpt)x5tZ`%>Ywp^?T<g;5C3qDMJ4f)wc#cz
z&pAFhrP@Ct@AaAvmI^7R^G*V`!AiDQFE23O;mSBMFtd0{Y8+Q?L8bdR9XL5XV(;N0
zi)l$+FE~8jXMJmjYuB%{v9ZC%#yXdFcPV-sIHz!~mR2rG%f{;zZ_rT*sHJC3nL<jN
zSU9CH(Slb>GHSZn4y}-sem?8t4PLx^z)#-tZPrzjebq8fI+R+jr8qk~L#bVy*Zjr5
ze2M?TzijCD3#PSWcXL3mUr<)EKgL@8ypkk`r70mvaV!^l38BS+ozghxdF2zY^QAAp
zMm={NA3tC`9#a%WN7#2gi<TbJ(Y1)W@sQ=UVOeQ--w2gJn%^F~v1ay}Yo@rE$$Xcd
z9Z=W-*7O*zZ*X<<2Ai8(j7Dqp`$NWS>x_#H3frT|3K@=u5(vx(w2zQQ8KK3xjENtq
zrYPXhc(2LwoUE`6Mgz=o0SZd*d3ZA6<fP{M%^iOESAUcD4vzV+|DXRAi?aov`h}n8
zwcFSD=+Ox$dcovmpWJ4w4Tn5A$~kcpUVrH>$4{OxId+UUHyLlObFz1z+7_%|yTxKK
zAYc0t0FMt3c&HSrnsRb>%GUNJc6WDKUmvrzwZ-=KF2)S0YS&o4)>yH=8;@6^Z8AJm
zLq83DDbw^~OF>($6v{7K<4q7f#uRVeyv5e$7Ehi$YIt*%-cqyFG-6;;iomZ-U^bN|
zvi-F&mv6rS-~BG$<c+a6o6dOr<S}7$la1|dy!Pn66pMQAs9XueX@+$rhH#ca9jK}W
zK`V08L*+Tj7W~>@_~*Ix;$50<3@Hdog4<Su=j}7648`<&cU{tJtA-gMNll&gOTE}^
zX34F^zH;a1xqasb*LJS4y|%+JTO-p2{bF4pt!ULc?~#dNv6)%$)bS<)(nlpBEXHK9
z#w`klSV?V_j@6F}8#L6CI=6F0@xT3#2mJK=haBvC{>#7qB0qgEC({O-<&vRse2h6e
zV=x+Gl>kw7<z!Yeu<#DAELoAGVkS{Qg4FN5x5FE+OsH*%QwD<J_U#v0TU(QKv?8ZJ
zlrfgZSgbMVEXScB%VY#+>V~VNZG<dvlI$!IpN@-607LRTI^m1dwmDi+Vxi0Gg%HTC
z=9j;Goxl74N$^c-^Oz`G2&T0l(u?xh;H9Fp@}A*n8y^%i=Lo?unR{OR<R*7+9PyL)
zuW)*L#?>2&hmS<`2+phHjirfkGSdX5kfLAJeRM+Y(Pd4j>ZWt<a(0!D#2woukYYCH
zv$47!O1CbcJy**hnTo@4l4lo)UKG#V@y1JE=FaoC*<HWP#$b~{HpHre!C(Vxq%o&7
zEDb1?iI?s?b#Sc>q2+%8DR#5Q#OysC;myQnCK^{7;}xM&7_XqJc=?mB^RNEf|0dN*
z&HhP+8}Bfr$HwjzzWx56@V)>1KjX!_ce(q<8w_+{e}2ZSbZqWiVxj_1_V>x(c!6AL
z^2NSDpf=;|Y{tRUeFpO><=Pt4iA3;Tc=0adS6`(-I%M~UeR^>kUzW*?9;Y-aGtwT#
zM1`SL?5}q|i>Un<g&?sTT*Ro9R?Y7%pVhQkMGogmu3o#$CqDTK5B}jjGTp|~y7Xpa
zTm=+$RZ%L1$<t6KhsWcTr`MAf?sag|%^X-?8<S-N9_}6U2Y>K2Y+f+e6{b9;$b0nq
zIc06AYtM8x$5jgk1}QiW2JMy9v=|KzR9U0T3VedBTdPuTS(Mze2Y)^W(q?TgD~r{2
z4k1WvG?r#ahi8dHr*GJQ@i+d*T)lb?RV%24*tCaWr5C?eVtFeqeTGt6YidU!ch_oq
zV+`Z*m^{yErrSlxMJtW>sbDA03D=6?G<mkoXTP+=_kWmi@4-5~UIs|p;=0IjuBNVP
zj*bo_D?mx_dtVvxM(PSP?4gaNa-Qkr7;Q7MV!*7dx&8cIuHCvxHXIRxjAt}jQ&o~7
z60nhY+m_NQ#ss>Hyy_mxYc$XQap`mE2n`;&EG)Yg1ZZ}5kFi-F!dgQl-On}_eRKU*
zaE{m3B2F2-qeyUCY38#z<IO(*>R<UO|Ka~(lfazMd|LCv9~jC~+Cu1b|3Q)G6Z(UP
z93I@k=;V2Hth5e6raBTh-$LO$?3B9as&ntB@2AhM2HI1TIIWS48Bx=+wz0*+vv&OD
zKl5*}v%Q0xd#Dst5{wNhdFoQ<B-Q?k*=){aGHKUDi3yR`RN$XA>BXiyqYMD0By&hw
zq_}AO$odY3V?Of>zs}Fz`H;cPv$45ORae}7@izO99<Z2C`N0pr#}9w{9R|GtMcyOV
zIcK9a?mv3S<|tz{8Z(*9_~Cos;JvT^fNR&UGu)kUa5Co?zWmFKZd~JqcixAuevMaO
zeT}Q1`V{qKg7XlRWJ9HRl`$6LAgG3C5)aW0=q`J$tHLjIey3w}B+rzNGD<Y1LttZL
zBYM8)lQ>M%4O`-m(j7a)AtPdk(aTkOnI}@Z`SY*5%&qI!`QdlI$A=$&#HU_)jo<#;
zzm4;rl9JwFKwcF1uwXnMGd=TEbw!>PGGMKN>mmvg6>EBamtvC_KvL;BQ+s=*)KcuQ
zyN+o+R(UXlfDR(wMgD8KCZLR!_0bDWR%t2li#TCG++}HB(n2*MKv5Kp1(9&VpjzB4
zVc$mU*a<c=z4NiQK{Y|y5ReuORiKL5CjYnp@)EThQWW6p8fzq#x2mfs)CA|O>k{v3
zsu26)WI4eL9y%Bf>94JWw$d^_+h=R{7K7o4ne)`%v8bdCf=tQlgO=1>ZQ@WCr9g*9
zSa%oKbOh8zn*5@Sg^P&(qR#@<E^ztEfSn!9lgCj|t{h@?hnDdg=uCo;i77%z@fmsF
zd(W&4{Mv85!gH^jLO5f*w$6Y3Uyaz?KZ`MWadA8_8Weo`Gkt#NA4u_IGVp`y-c9O1
zwakRese-M5@Uh*<^7E`J$>r}Xy)Vw43>oWwUrS36tL(GZN)fxzm{Q|Qhfy*t$b}l`
zz{I$dG_?~&F&d4en*r^t72jC{9oxtl9lg^Ci3T4CT0ZN8K%!M0@QXQ0!E9dh_J?~Y
zJK%-aZllUMxi7hXX^T5A+~V-$kYbpl{FJk!BMv|M2sfRgoM*6kiSgzQyi@F4yTz~m
z`QPN%H?A^V+oU%?;@<ba#_r|olqMtPK1}8drt^wiOFMR@i>4<Wh_|40%NJb(Y}vrN
zn9nZqPFF4M)!&=Ust~wy=ML6de6=`d^re8hwPH|aHWv>!NIh4t*K3RG{Q3BU?GN~^
zzxLPo!q5MVot<60gX_DO`0())NmB7Lnji$YdE+MY*#<v-`z=WqHew_v!^oIS_Oxr5
zk@oyNB$e*J%V!sP^IdxCey;hQwnno#TdcED%3?JoiW27+l+MKzw}5xzwuK;7zOrCa
zjMv>{3AVNy)?_ev2TH}(4MHeoSy3a}bT*r_SS&a_J>~fDgkUp1xPOIjee-3CoLHId
z(OOXy(r#lmpR<_HW9w!uLHec(Z(uf?VRM-`Y~yq~<vqtoS2#F2V{NMjlhg0#?GUCI
z2t!%P#!uc<Lv$+|x#J41ZcK84lQvIPn~51N(*Escw&(Gs=knzqZ+^PsxBo6^9a!E{
z`KeOOA`mmM6evxvH;4o(WESFDCWs8Bs8zw;*8&>P>9J#TbI#K}dEQi)AlI^d;*(Q;
z?{}vJKa^|fjwPoQ#>`k*l7=8xQvl=BE_rR~T&th&zP~CMl0Z$)SL?ksfW|X+lqlwu
z^+fJ9fD2OjLs>>X^Yr^Y@;o~yOr7&f9!R5{z09FllqJRp@0L}KSJDQ4u~;xIOWeEf
z@`Cp~+1uy%bk3dYH|Z(xb%m~HT-qG-=9{l`>ACBC>zm(X@52wMCUe%;$EZODzM|L9
zn4BK*jc<RK&wT02{QS@VD#J_HF;;W(;k*3w>;Hts!ciB!#_H+!hm^x1);XluIPOj2
zv&g&M=Uq*dNJo0+E}x~>yD_tic+59J@m7J9RmB@`yutSN7Ec~NkoCOa`x3+#Ifyn*
z0zjTiv28GIrl9w}fv{WvQ;fLv!ixYLA0CqxS?p>n6Ge+m;NIbA$ki*`{PgX&8spk#
z8Nn#>JjW{W+JxU*K3dI4TvZ~gymt{yx3o^(jIAV-QUX+z9RgqpTxO3>W~^;28kZL>
z8&l1s7_8s#ljqV7M2fa#0F(DJWNA8`Qq@&6IWNz1lv4Pz#Mc$m*$k~@(9(QSF{>--
zqcWd!bacdgKBufouHU-F55D^XCx?!W^@y>@>PKY^<*elR<OEz;f_t>F4Ibx1U_M(g
zFgbN8{kc>KvF%1ie{DpWTrydIQV=%i*!N!f*o{~gD0oGh$hW%K>G$rYDYcxrkY4*Z
zNUME)S&(Y2xp8w3xZZucInrR5))q<EnUQ4^l-k7Fc>kyvXr*!0r6ozg5y(d~25V<H
zw~00_At^M$8$R*Lb+&gN@bvM^jZ2(h2CV}xzj};Tmxx_a005l1Nkl<Zn|LMoa`%~6
z300o$eEW*_z4MmPMKG2uNvl}f(Vay1=-Ki?PR?fZ1{GxuwXfr-TB!`r^PFC<hf>5W
z#?bUwla9Mn=8HLHRZ&${Q!qye_;Nv2mMqGJOzbO5ri+Cj2ujBP?j0Yq_H+;L1EaMO
z$H&JU+DGi<H*n>gdVa>*cubLJeB)bR=l6c^D-1#op@N|44+a#uLVHD?XMFVV370;6
zOg6gCqI$r1En{)~luRk|-hgbhAx%0F-Zj=re2O5WBS6rIh#o}S<~%HQ0iUhX$8zj(
zxdCeTSY^M2;tz&HwzjsT0$J`AmFERluU_To=on+AP6wm$ow$NAslZvz*9P1LO4M5B
znj2$ydVIkC$q{L$kN1H^SvKCxn{U3!@BE{GAe}Ml8fT@||6;LVeS4Sf?d@|9qm*i9
zcdQ;_w6^@X^qmWkkLG#OKA`Zb6Hktic#E1D?CH@7`-6S4aDu0H()&sIjx`uJx2hdg
zU9(usNzKs8vZAg^=Ce80W~`0ZBtGIQ6ryOdET^t3w6!d%ie7&}ZVZLZn3QMW4ez}5
z6SmezSo>uT4=rnJeF@Ba&*bz>TEkmQ?W;JeJ+`}$RY0N9#!%Lt>7s-<_(Us3KeN33
z{Ucuf#3klhqk>AE6`~bwq$f-WDHBc>IjF|k?h@!l+WsP~9+9PXzhhTprF|oaL%3M6
zcprHEjVrMu@_8XV9*;RWIf))$A}Wp2XPE2}uuTXN`tV5zr_B$jL>NJ)N%ndL#CuBZ
zT%?iQ8Jn9KmoGo%@q<^I0XPZ>9yYc-Z+v=P8gY8x+-cX^yGV(o^@O%n$@0fqNKWoI
zoD&Xd?Lv^&3XR98YG88VHUG4Z<D+8{@RiKEan50lA=HWB<cBm449ph`>FTVNH0!Kt
z7V|k(RWTS0D2jrrEHO%AgIQ*qRB@@#3G{jc`gwuz5Ne0gij%{q%#T0fJMX{6y|;Jx
z+-H7{(fS7ahx`2fzyIHI@aPfG?QV1M;4vpB2h?s3)Oas5YI^yI{@NM$_y1dZU->5F
zmaI_R+*;@L=U(9GqsMGs+G<-<p>RRc*;A*41hr|Unr5S`#@4Fw(H&v@PvbM82@`qF
zM**tn05FYp?Y(C>81QqS`V>F;(GQkr)qBs*&Mv?F%fG_k`dfbsmFAX2ZzT9O#z61{
zT7GKjR${$6hzB$rj>v2li)jtkn#K?dL391uEe3W#Z~<*38OA8^I9_=9MRqUm;_DLC
zd4A_SDyAQ&V-==K@<)uvq#)b<NxvubsLN;aUNRPG6rWPk+|xeZdr0Y~92^{QaBxUf
zl~SQu*G&-580o)LInSc1+1=S?XJ<z&cog$ex?^TpMz5dApqJ^Cv$GQ!%wjWYAIOo8
zy$b30s+5s{Q9fjE|9~r3zR7R?rJTR}x4+2V{+vO-N9}43504@<NpX}h3Z-mE5z9cD
zyjMtNs+F$20&T%!;(6z%0TWUcbc_q=mRM6q-jpJzI<vXcrlhW+-Ho~^mpZ{$DXlWS
zA5lG2@J;i|i`-7xOvQ_LVS78U_oT&%)AP<|vt>d-Jh05>6nP)#>PXCqK$At3hA*`<
zGcuD=6n&r_zZ+EqRcTqw`+VW^Ip6zs00Tk%z6rrL&*Z)0%JrJH4atbp!4o=32Nx-h
z?&~4qZ=qT1mdG!{th<DrVmfK<n(t|eEj(N510Fs8h@;~@9zJ}?+1Ui|9qVgjW~Zkt
z7UhzHUsMj~1GjJ8itw$}5lyF3r=*m}7?{sz%x6<3lQXVgzuwTG%?z$Aqv0pj`FTD-
zYr}Lp<-z@*ar5$ZzWbx^@sqcI%!_yK@F(B+CV%I*{{f52F&YjS<vH71W3p@mn@K^Q
zQGK?zuW<L(Pw?FB7sv-Abe^G&VGt_rfBie`-G9ub&wZk$=h#w2N4T+p99ia<1`b~-
zdP~Pe_<ifewy>^uzCkBYb%#ED$54sii{E<j?p=&AOLJtj)|{P9_|cDkB=31gmWy}j
zeV{0$b(6$%ePiSK7-U|)Z;^THYhyOnH+cMHUxamCH_c3E^95HfU**p2JG}qiyAZ&6
z$>`EH<NA%~q*HEPHP4Z*9WL~^R{5-jct0*qB&S)f%j)u32yL%CAEd*Dee&cX^Z6O~
z?%m^Taz<t|PVz(Uzxx5lM@O;6d-OoOOp^QT=RU{AcpyU37-~Gq2CA^24ke|Vb8vFP
z;lU9)>(d*K3Cc3B1JgxC>CVWqf}kv$JC_;s*Vx+GVYohEyk79uTle|B-yKs|G99ce
zO9BB`xn|Q<14AX@V^~xrXJ-@9>ONxm3S4PqP?86f<cptk<348b6G5Ap#=GifC!#wK
zy(A?9A*3c}QQNEKbF|-S4~hL4@?0{+R{fOOY_@#O_`lRtGTnO!4cpS~gxkHY_cgn_
zagKQ7&4AJ{nG}5E>u0?5(#<6+Q7kycrORhzc}}Qhidj?byAnWepGz-!aOobLTNMi3
zwGr!a)yhg9NEZjzT&%eMw2wzm9-@_MrrxcM*Ld{Nhdlb=K4)nFQGABd)v*!ii1qcc
zSl?PEfCl3!{eqf;Sv6tr;F#HT#@4PQ&ocrrT{vd*lFCiUvL3}?%qEuLJ(n(B=FQK1
zj{0oIpM3qBxT@m)pS{Nqe)vO<j}LLqaeB0ejl82YWU-H%wS9_@9`KX*-^FA-Di^4#
z8VK~XVxz42tAFKRLb-;-BPso}jfvFari;;EWI%LZYw2aVh9Mb0?e&*^_@unj=X<?E
zGDKBNUjqM4_{mRx((sIPb>r!e$7@X`yksV|!uleC?z|_<GDf4JjG-{8K1z7@;ll@<
zo*whujcdH~_K&4qla+Ww2!Y{nB+3FhYbM2wdWByuf&jZ~2Is5OyX(CA9Z7C%JMu#G
z2m$->;gi@pdB)k<6l*Q-J^2}jj}IE?swpzj0Ti5^PWa%X2aR`9E3w2!qY;Nk(ip9D
zfx&na+Z&P()|v@~8=JdiHp7^VEGw`^($%~UXcKtnXFuhKKYEOFU*z=ctl4rELZpij
zEtuxlD3l#_Obt$pF8H97`;Ru)aau_;!wW2()i_3a?(@PdEPW>Z-d*_CZEVv6Q`_!{
zEV3#PR}b$&>odmVhkO)10n}Z-ICn27Icu-iV{L651Nh75tn0cH*l&fm&EW3c3b36F
zZ&7$!D}Lv99`UdL4-LkMm7jvwN(UZ27&D($3`{8@eRc6#rlYHH)8|%STdjT-FQx5M
z)WkaC`Rh3SPR?P@G0wAaJbLhiqR5#|=dm@p<EL-^6jcZ6SVWh+NDKsHdAzsJ@!6!A
zaT1J*U1u})_70d%=U_618-0{5=&!BI8Y#t&mbNI?=9r?-=;<CPng22vjrqI(@E`N`
zd+$-s=lu95Zy}vn9acy0ILLTd?*ru`;EZEl1x}_jo*wR#<vE}mqowTStPR!}Zmm;y
z3^M6#>rg@HrYQ$h%a(BW=S8uQ)z7E*6OG7cL!8za`L6@hdgq#CpAft41}BXji=vPM
zx!N}htb0Al#7bFY8S~O{<;s;fdDJcQB%z^4A438r@?4f>^alf}Z_zq>+!Dup`t%9^
z?eF{pMxz1VR~Q^+xxfTVW^(59nPgSc!o(@5qT$$;HC#>j=_=tQ+qx~@yD7h2K1*xS
z<+Fe$c!xbYI>Xh9lao1<Ny%uKk!2&inM+eWEx4c224$55ri(SEB@BimvOH%z9J0B!
zNx$D0ms?A#0Ar;pGqvo{D%O1j`B!^I?Y#{0a}KLCN5>}|pTt0X2&Dc_$&L;xxt^*`
zC2auEt`JX4EZX1na~4{VO6AWJsY(i`Bk*(wj_Dq&pF0;IkDJ=YH>4_E`d<lCCFkMe
zEHkAPMV@i%mgPr3>{O%+ZrEgr(RIgz+;g61I4AW5tEZQYI@`FxB5*<+m;=k)AVKo`
zA6_QU?lBnbGo4<c=^(5E$A^aLWX^cYqJk%=<!rN64<sF3OC}NVs#kK++WgZj+4edY
z$CdsxkAwX)iX!Lqc)?;(QRHjbd_)<%%ruX|S8WVRLx0ex^cL+cqrnJcE$f>bjK^d8
zy*^nkZQ!+$daFQ7uSTVsEASXx9azLziwhp-;ql`qRF%UKCul~cVF+mD2uc!I@EsnA
z^(3yYff}X2YJ&?UQ|~!(me9{AbxTEQM6_Co5w3crtFNWMKMvxJAAnWsTPvDbK`eVQ
z>82V18_&JHz0Kz4CXXIHS~`6yRmgMcyXTyXjb&@r)?`dXRafbx@n`$&iqu9v{nPYg
z!J$t>-2x5D^WjJL_{7~iIA3E!YK|!PdT?+c0^gc8gEXD@f;COg>0^hG)s*OTw1zb2
zzWZ@UGSWd|uU@^$rAwCyNI#)|zmGMR$-JWI_2?CSj4|YSfeS)3Iu_U{rLfVpNwbI3
z!-+*4J%G5Jl3A5hL3}^Pw}O`W=y&hF%<G>V^R+*8Xq)`c&_FZ|h`Ri=&4v+ZW+bbq
zegGaG>{0q%#JuZnYDdVc?m{LnV%d^j#cLZ&I&`ll8?OirCTd~^gTOpBuB*h7dhFZ_
zN|6^iufMMO?Z21iDYT1_XBo9~O%j3dM>`K%DHoq1xyp)*Ah^U#-aDYRMp9!)aLnf;
zP|tJevgVz499ApRNQ<XW6yN^V5r5&&jqzn`v8-OB>(R(M%F{Skcejw#&r8uV$Wjt>
zqhQkcTYC(IA$;wgwHP|rZ$8iF#s=O8tTl{ABR=~DPgw_sgJJB5P)Lj_)NyiOr|3@W
zmdCnrCLqKfPd=`R6zp}pW~G*dTJ0Ra{N-Qet6%#rhll%dPEH^=hYJBhkg9N{oAU(k
zaaD=!6`+NVRR~lH(9&?_a603|y{B9Yeu*a=UV|v`TT)*&Fji@2)l$|m`=y&|o~BhQ
zBtGf`1n+1DNQHp6(t9et(6qwwA+Wo>!?kPIc<|uC63zDeJ+`)Xc>44y-Yc97*i4My
z;c!Ii%@sPuwOVV5!8Xs+O%w@~<pS-~cnrv_iUol(r|9KZUnTDqR8@&p7ULDUHhlkk
z-{)6;=}TO_zT2TH)fkbYXk<NF8I`iB!wR3J8ZqA}IhJ^-l?y3uvdU-9Nrs?({j;BC
zXL~!kCoQq~I2leW89zyywSc2Zef1s};?%KZ@kfhNMFL!QheE6pZWlD&n?~kxj7Do*
zx?%y}F}iho^GT4Z=5Cv}lvboNd-<MZQh{b`_ZqnhGIcBwY{z=(3aKQR+RZ<o#@n@&
zwbp1S)>iO!)CHlj@`Lxd7|`opL(vt2t=XoOjOh)|^76|Adc7HCIV3eP)Jn6lvCiK9
zepIgU96?%K$Rx?;Zs{J)^Lm+P7J?A3HmQe@?l{Tv`Le9I_wZ-r{ea@t2ye(H1WL;K
zguog>c~VRAt{bgXRHPw!Wo_YG*Ar=!a#Vy?$rWlJIujhbE_o{;((95X^6}8%pVsmE
zr$5d5cuZMV(Qr=ih=&lBf-<O}aCqoIj4^Z<m8Q^D(<3T%is-^UdZp@EBXCYUhxdUi
zm#;7$Z*X#SDEy!eqy!5mO4uliQZkQTA@e*!t;zC&+99Jnl@vl6XBls9apn4RSP!Pf
z31TXJ6ms%37b=a5Ag%7}3WbW0Sfat8e1wMuA`7luF1CwCt-?3iOA5I^6@_SjFyN(^
zUglfh`qt9*UwrWnue|aafBmoj4f3pqQUT{Hw9V=Fhth4^w*|YzCv4q*jYQ=jWsPQ;
zp{FxkZkQ{FQw|H5tl;ou#^b{ybkT!_Be)<Ln?6vMC1qU_bep{17+XzEjJvSzYBh$`
zC}pKF3XGDBuPW(^mG6eI%x6l0u>#-O+1ZTwU=dTebYt<g5R0_0FiI1&2*aip0D-4W
z%#KQyLib2aq9akS7K(`mUHSsryajwg1aUXbwexVQ`=X*HFlgbO<`bTK?{DGigILGZ
zYSe!gUfzv`wAT7^Hb)2n8%*>-e8efE7t&?y4pDoF8k1&$gdkPa>Cjp;ozB?Vv5dz?
ziu@W?A#2*ed@(UV_^9=aR@*Z))_ioc<=MrxHuemq1`q{OyS57N=bX*o;pI=f!jHa}
z+&~4&Q1s5Ye%%x592M&L`4&T#-_v!PEpF~-l@_d(U~ltkgyn*6zqo82wR@f5Fm27V
zRtCasGN;ITl$A6S73HcTo#<E_K%7qxuA>~4QfMPQnASD%SyD#n=oYayo=SSNWA%5d
zl$BD=`Lxz31KwBEPO`m}GSMNGcvD{VNd}%6i=ceKEh>~-ZVQs=to1^d4;2m1GzBm{
zbC>4;%9vGVL+U!4cudTl)RB0VCek{1Ynu!r7v~6BCtmt=I^o8R8;yKQ{Bv}4479NX
zYYf_2s(8+10Co*o#j@;KqE#j9Tj?NmX|@2bF=)Ki>>nP1DY$g?2KRn;PbR76mR>Js
zI2te>jS01Z6iG_F;Zqmki($akqltUOTA>gT52MwR@wm!oi9Q;S*chxf5**UnbuOV$
zT3kZsVvcce5Yt|lh>gZrY8E=>_otgkP|!sIFr`%Vv)b#YR@F&4B%vvOO47c|n<m$4
z|45dYiZ0u1QcF5y<GHL_K<DxIWXQIb^fDfsg3@`OM+i&Wb}J(;UY_S<akH$AO(<h=
znB@Drcz(6SKuG|kRA8l-y%q00v)Pom@13=5e^2@77ZJ1fA$n|)K~;uQ-v@OU@5d+=
znh~I~B#K^djbHt<oBX~1y2Mph+Sj<io!gEVZfsLkUId&@DTd2|$P@?bR_AvKbT$6r
zLff(}C=pj`<vCV~yK5mmYlpPV$(ucC)<lU1;#`QeSdO4Vll9Vtf4Y1I=l6Ei@@zh<
z+ZirmuZibxt;L$GF-DSyLjao_s;Z3596bh^1EjQ~sw$MWvE9GMDueNg`N=6y9zNv7
zH(zV4WJTL&NfehmXN0Axf8A&6dQep*efjb{mznN5K~<tadPO0SXD5rRY^wWHyQOw~
zLhvp9I9Erga9J?}JbLty_uhLSnIJ0brHu66%i~_2Lnk`x-zcNF5&)Cwgvu{)Q|TVw
z%lfF`DSZvSjH}nK^Mx<{JRiRIK1!9iU@0=io1b`@wO*gvE!RgmC+F|RG`k9THCEe&
zgwy)ycyFFIuG}@2yM|tR%@~75VMXY-1ex-sRS=CP73d<+RkTi4Q!zuQ74zzGQIoDN
z(T5P0+K2~WV`Dukju!49X?m3c)X|J?gk0LbRShhGzRYT>s>W+DYB^IUSwRVIYZpwe
z-73UoVb`60mU=-cZRi($Hn-Lp4F(uAv)Po%bi&!$gt9C-Z=ol})UK6QioBok?z?bu
zY-zHlI)rfEqEn4BN{V7tI75poYa|dZgQF_3{H-E4((A`)8EagYC7FU-Hv`wM20r>g
zNr6*<m+x41#x*nVaLS-NMNO@MlNlvlLQX$dg<kjG8X8%FIJ)O)V6Tqx;+=oyVjvP<
zO7kqF2-kTJIH_UEGo%16@|amET8)hnSG~4XIA|5ZXn1-h1v<+zMiIw1I*ZC6?uk}X
zSPPWZ0_VNN6$4F0c|tn^(k-@!<?`kZ8{>5rRoT!$s;F00wZvzw2h~1fSFUv5v(`5B
z?7f!;f7UV{kJ;E5GwAhkb<J!#<@D@~MJ&Qh&${~j-8Gc>ZqC!EPnXurIY&_pTgyH@
ztI}vgBPT<U(GrOjQ;)6gwUf8!oMOJ1;oOYP;TqSsMqInJLtbS3^k?_Ex3`bgo=?8^
zGGD*4#pLJ+ZQ<6|kk7t$hg?_8bS5#a(B4Cvsd!#tbbqFMO7|IsTZA@cSC`)JtkrV;
zmB5A0Ic!~ds<Mu-Rn-1kI@5NoL+Ov@y3#rp_O$}4yZN&4%Xql8?cn<^u?Go)$n`^r
zCHt)aPaaGoGj&YG?G*5|zbq>et1Z3J8bw}7aUBtE>nMX%ud!+ci$G@OHI|+?3A|pf
z&uBE_^5q?AOQ8c*S#o@E$fHM(Bwt?9>86>A=2R3at)`dF6nJp&sb@Af7@ZWUJlAYC
z@7!0yLlu`UEf@}^r<vC0E&Vn<Ho|QI=JSeAzP`(^|8wv1Cx7$?J-g1Jmtm}?tQ2db
zQ$GEP4TgEacx{B$GWZJ>__>!KVZso!Og3zyRp%7+3ar?9)~;6~@5O*|kTx_vgoty~
z1U|cY>FM`$-LL}PJqt%&IjXXbN<G5Du|+;1c?fAeJUVqfjq9XUYh6RPr_p)LI~g`<
zeOB9CY9-A9x3{;VAr_T`QfMP1fmKjMfSjtFqcoBKyC-V}*~@od;*FPH<?{Bfcnt9q
zEgeV?ywv8SYdCe!v-};J&x(;?7z_t&Y^<}hvlXlF9o2l!)2B~4I5=qJPLdg21GK>&
z^m2Ar2!V$WA1+NIR7x?Q&w1;ux8$CXY&DO;g`{Zo5>LgOZ`WD3O(R3_RHf%P{_JnC
z_vk)XMgu;5_jz8vbDjRs^7S9A@zw{2T;1AWb69Zu`YwBJPJcM!>Zo96U{PT~P;t&q
zysKva_;^0kTF34k-MFM`)`1}2z6|4vkKsKig*CQ;2A7Jaz4T_YHq&Ie#W}GMluE;r
z3{5NK6cbS@&h`*5K&PmKYF*AoyQ?-UL=jVXjX|rfU`fuMk64W0saYm%NJ9Kio=UJ{
z6)mICA&~B-B}*Yq{qqW@<$`%tOHZEE;=Dtn(z4lK;6-&It?s~i71L#<rS@PjU~O$J
zwlWV)CKLAd_BcH~r79P~pE1?a2WgQkYi*kKRoX_V28gp&yoRTHdl1%X;XJi`Z*-KB
z4{UAM^anL{z1(mvJx6OvbYURFrQNIi2mh-9|HuF60l)t{V<^P6)ynY6Pu}IlTaQ?H
zvDTD!Xyq|RvA)}<B!}9FE3I10xV2_NqoBJk@v8g3w1<&NS{y&Nh)^06t>IMnlb)^J
ztH877f)s@z%QRIbDVw?(N+uILwN_2arHXE~@J<Q>BRF-3&$KZl<G0%!RL1(I^-$7=
zBeIH|)1Y;r=qr@*ah!p87cy~DhUt2v2*!f8_%hb{1Zlvjq?ZRK#+~+Nr+nj`x48Aj
zi`>nwHus~N`${psWlL6$bg$jID;3v8H~cjm4jGL`WLZXCR~#Q7v%i19baF-=jYG6<
zU?d&kSFJT6&W<78flmTaDHc`5<HwJejI4Bz2};hajfP4<#hMwEZw!{u;m1~>r+TZ<
z?D3~R{dulleU9Jzul^r|#Sy03pa>g$>eWy3<v;rhgIorQZEo#y^5h{sQ?j|ejv0-q
zf<@Jd7FcN`pJdc>bH%RraM987nZl4@L8+&L<WaRAe)^7&|E$qQyqK`8NtfMZIaCPE
zke7@8bvM8Va9vzeDVcY`vMUX=ZHtvcDNk80;)BI(t;kg3WVP;R1LB0x**LX?VtFUS
zV=5N~X6$O(J;YLIF~KC{Rts^un^F`7{eGWbFDC@Y(a|9w&FfB2PbteX-skxqSso-E
zQFUOM73(2{KwVe7|Na)zUDh{*lY-zE?Nh@fE7n|6j;(-Q9wRHk(KWvK#m9W*_e#9)
zHBeY;=fA&y#G+i6sefK`<*MWMi#d)gYW09{ZBld1L$x%tBJjHwb@z4RS(K7CGtPD0
za@APp-E*{b=y}$n=*cik$t3ADSzkJycDOKIchv%Hg5{nDGHEljXcP!UVQJ#b5L#ZH
z&u7U*!$({{ogb|**2MPZ9$$O#W$XhLBz{J!KMFj0@`SVDguR1(UQDz=%TqM1eOg7)
z7K|?0fX(-ek!G%gL7zOgEEe;opzieal*we$_JZTQF_LH`WlOc{DkS|xAaHhi%HhFb
zbKO<gS4E{9g$w~Ny{Z}>>M&&I@J%l)dDA(M3yz)LOI*HojrskbGMmoG^FAT#V`Dev
zM^Bz&iylE6yf$2V?k4$gOdX8WM#gKY<}q46>w5C(d6w7u0=cdt3KCj`DH;B04ObB+
zaT$>qh`7W09CJfUlwG($U!X*~H<K=8^m5>H*@EuGM|^7?Ct=opN4$lgTH!%lG;vJ2
zgwg#aG#0=2lDQFr)Kdg?K7%Ly-gT|JE^W8bm{QV8WHOnMEQ@r%%T3>W_paKtirTvi
zwS*8_moolcRTYPaW%JC*O6)%DIdW($j&`k9t@A|mOHg8@f!9Bov9tY{qr=-+8zOcr
zO8rm1@eMw@Z}H&@3eEbK!}g>txdO3Dy(p2S@`0wY-tzU^W2vqoy{DF$qPu?0-^r-x
zTHi5P{fy}knKHKoX)Ifbs|HIqzV>l^mV`og$WZh7&b5~zzLwYON_wAl`;4W2b1lT?
z396RqoE_sNIH55yxB`p%l-y>NRTVQeKV}`&a_nz4y;SZ&@C|S>?&0|OToOyyvhJb-
z-Sxt^mU8k+<O?+GrIh0I?3A;~q*?z}&)vo$W!|;&D+KAg6|dMnVE4h}uif@6Mhkp+
z{pM}H``%kjN=L6ODcuzBYYvYN`P+Z*@A89he~Yz3(<{dG)_15wG@2yTP10NKlt7X?
zZ?3hv?yGyy(n?I2@ynkLANP5s$g&KZTdX!MR<}y9s~7KX;wtfXKkK~*rQ@*s^G8>R
zNE%vuv(7mlKY4<*F%PZPot~&;t~DM=+Bntd%<k0)!Lj!=<M6<+u~p;!vc=ndri(0)
z?tOzI1W9>r&)=?5!t;d~SyC?Pr%x7;QEmERge|&mz4wOWqqML@L+89AZr-PBjW1)(
z7g-+Nds0dfoMXIhdG(bi{F6VpMZXbjioJt9zVrR>QkI{gtSY?M+`Kcx4lJ%#WD<gt
zwQxR;#f@vK1zS}tTS(otPlt{{J&|<syhdx3&YI_Gm5gM)Ev0;3Yz+RXA6*{n?#Jri
zs|x0#&ny`@MlVr9cb}WJYKlVBhv4k&jI+t48PwJa4f(v$IdzZ#xj;`zxd~K42r>I7
z#92Qj$0x^BWhnx?OYc6!N_0uuxCmnD@}R))%LZJRZq7q8>KxC85TxyT5+FvO%P=DH
zX<3$(W!c0EnnaEji+z>HlhSu9-cOhQl0xeK9lT@b(iObzGp#(Avkb4mTE$O)@)Lgi
z_S@Wl`~>4`);As5a0{m__<)Ig2B5suwxoI%ufv7+o9?xnV4<}ZV=pOJB`Nd?!s_x_
z^PDTON&@J*@|}%au5aiPVtTKef|_1We|KMRWI*S<DT^X~N77*H<~sxl1b^-en`~^x
zMu1&Up)pFu9Z#;;1$KC&OeBpkaBxs?a3C?5uC<e%bvdoJeg1TBtreFF3_pfh>AHN&
z*H|Jdae#5b$4cQ<tK3MY%$mSpOG2Ux({=SGfz8byFI}_yFEMst`E{j|bpv@m;LrS-
zAz4-fY3!(2gVWhu3W(x49z5(ZDQ!%{7GiGVVo{i<J?a*_ty)N{7HDGb?$M>AuE0yd
z!IrLS>Z*jvvlRnLJn?aR(|v}<Lx{p`mCsW8IIsj2rDZ$+o%H$W5jrWJ`_h-b#CSYj
zQf{edT?jIp#Ar)V^w3H=IK(hWlhu^0Yppez&6*)g=}3Ga_vV-HotD=U>abi#wyb0?
z@?O(9R%t4*T#@A>6b>v8MAMpnzfYFgre<W>IPPQy#&-1SJB?fn%Q29pdtBi&f)rI=
zxq6-FUwV~A?O7YGv9Z2JUKkeTTvFDvj7eFsAmhpLR1}DcA%Kvw;9Bgjo6z)eYtrSv
zb6(1_*6XZ~YUrp-qjl}5t2(79FAKue<06+R3EfqONDmXd$JG_CuB5GicX-zlPfHY?
zg+`Rj^=s>_Zw$nm?Od;8BhiIux{H%OAsZ<QNNLS*XefpnWwMX;xDr8iEybibmOYb{
zJ{Ma#BRF>YetVwf@1{cvaiqUk_er{f(T6~m8~)s10ApJBwAt&%8(-d+%U5zFZMMey
zxc1HyyhC{>T^fC2==SR3(Rk(+yz+@2SFax9oyNwIo7-F4y!ia{k{+q#**^GLpq#bP
z1{x{&GXfI^P1B*kdACGcX^qm}cGp)MEyaLFMk;%S#KDX-p+jrQ#ESvk^t1V=EzcZ=
z6^~U9J{^glR{5-}3|9F}l&-_o72eel9Ntw;n~5dQqS@<~d)|EWO|D$N3?X`Xh+<F*
z*}o8j{3;_5#~-9DzR;aSYAE_r>{bi}yVqtpev|m)<9N*4<sw?SZQtRw<-*5hK1*?w
zWnpPl<g;d-Bfqn3fLwp>Io|xtXVEqTZ3s$BrWL9|D^g9C^NxPM&*tWaC}v;decc*7
zF7jAhbKl`HNdi*x9i4IG`Ini`6}@c8c)W%8@Y0KS*&L5iWY}VjvI@NQ)=w}>H{Nwf
z#&qPbkoJr7vihR>wZ>Sm24xIEtCm-_lDZ)yJVs}Ym-$Q#OF;}i2Ifc>&#H%#ZsH<B
zXe}8<x(BUj6iBi*R-p{vFqM|PbJrmTx~c*aOciKrvm5g0D5Uf_#`L)L!dnc-hs?{?
z3R+GjY@fNSy}Qreu0cc`V}^zI?m|XS_u3Vf5aSrmH~&;D>|0vn5U5J{)SEe1uRXwf
zyA&fyOwtfXoL#@Yy&IWMvP85B7!^qMt2+2d;L`F0VoGlSkEe8wY@qq%8;0quVme<i
zUz8Y~@hiXhi{x2`ig3oTUogmXaDmEu%HXJ|sX~oY0jE4pwUxrq8Yrt)P1m~d-s8Nd
z4xY+8>foupr}FZ*3qDrhw@M_-GAWF+7Mo`bpE3O*O8?>jXIF8oDu`7+(-A}O@|hwu
zfpbwLtw+7e*U>RVDaB+mVKHAwFS-;QCxYQx*4TMX8u#owJf#$|0vZ$sr4=u}e22+=
zM(KR4aY_6HiO!Os@9y<QURmNZQ9R;Lbc+zXf>M6AYn2`}@mYvGree%-^}8X!=f3bo
zw9TZAdX}SYn(MAHMrIE=SJ5kShJzvN>+7O8@Q@l$66IZ2k)b-zFL9$fIObKw_SG9?
zYg>Hp$M5s)AHK`}(Uj{~uJM^S-o(`kYlnE}xpU(dCMbdnl+9W}g==^$O(9v$fL!$)
zy7cC}#|2MKpj_fJN9{c>8ueWs%d)Id_|{sqUkPAe<W6@9d-e6MOWpm~byd3zk$zr$
ztypkVJ7%+4!+5Rrly;@#IkZ*%E8I#X7R~E#^!SBe`Z0g>l|P69zHm;Eb+OM?fp(D+
zk$<fKNWU+?@0v>fk2}xm(e><{3#_eK{=#n_6MRVsE!5PkYlBD1Lk|W+@w&P-IKJa8
z#&y%tifLey;HfK5T?c};eDcj1$CD$T931fW`)3?a6t2EPRc)hmKxvrGG7b;Gm`(=Y
z(%LUSU)Q6%$Y5E;u)FK!oRcx19k2Hycqc(N&lrLgy}oen`MU6MdObbQMG)&Mzg1PW
z^tn}QbJc4BNsLfhbMM|gKK$^*r66yL-KF2PR&o47EWSOrzX25-&t1LBzxJ>E6&~LI
zi1*%pJGq{6?Pz2Pl(x(d={y(lT)PhG_lspzmj*tT&pK{<_uR>#efi~=dF{2=mI@HN
z1L0E5j`N<pD1^^Qyx6XbTgmcF<7Si5?3}|@p4xeO<2Clm1^=hN_22Lx{$KwK{?S*y
zjxrg&Ufx*ud6v;D3hEeFl#JH)S=xPQ=_d*NRsLF~i)Md1ilzH+b)8q=civ;QUdG1K
z4FDJ3K=<EjV5WP6iI7wLAf3Pad)K|zM#4kRdCn$ts%o7q8z61TBUBR<s1Rp^wC*UK
zjGxZUB9$sLe)G4E`1}`NWpDpF-qlTxbC<}Z;h%4WT=)I1>)plSmpzzN`Q6>DR=y_!
zTK#T!Ggk$!Twn2<zqP~HzgF>`Z&_&-q9p&mOGeHGik@SAy~Ncrw;tuup)Hzh7%E6$
zySt}#EqI~QibV~(*N+))^q8ED`0##@Z+$msV|+-tuxK67%5Zd~dHh6i^=e!74xz!4
zyZd||QVC1X*<G7e3o5<-Pg6SGXLFU~Svyo`J<_vv`RpQZc$F?z*Pt6a3Lz}fZK750
zg<hlqVf5UiusAw8k`WA9kE5fb=Jlp$U9%@G>_Nv~ad8@+3l({8*;pU*8-M24sfsu!
zBRxk)Sf?O*@)o*#*M*h4{Lx((-_FWy?l<w7QnAg-1vDloLA#!HU8AFi=A2_N81PsB
z>R;vX@R0Z3dyfzT!(NZ!U`Sb(GWV#iIXOAu#k(&xF%$0{-geUc<2|M8lx2xoH5lW*
zm(DTTy+%3OWw5oz`;U*QgW>RCAFZGmjOg|H9336uy_bofo!)@n`ExXqWbj3vE-drY
zsxrQa#=0_n)!6aQ#YwImD4_eV`&t*O>bi>Qz4K{y#P?gqS#_^g@zC^|wQ|kTXw(wz
z`Pf-|tiBfwVdoS#Zam?YS012(Wp`^+1h4AANu9;&V)9AjrhD%O+0%2L54<m1PRU~G
zYV`Ep$9rMzYhqzK#$(Ipzc51d=1M(4105RRn0Md%J(=_1T88PKuOTGwmi`VQP%Dj7
z8s#1H$&kA0p{)f|@JD}?ae8(Hl&Clk7>rb^HWo^~v}PAsF5SOZt>dmT=&pbF8M@z3
zpR1~>`RD2y=N->lAtAILn2kkgUEWErt@2ryH<ut9mK9LfxJiM}uA*wH-sKA@CA@q6
z`gPuT<BbcAmt>xijH-@6PS<pGjmC5H+BJ0Wtd07tuZ@~ZbwA&Iu5=Bd1K$YpZl+J7
zhpzFp>REKXj&9OHlbDmf)1fmNsj$p<i^YPMUw)ZS{oKz1GMFvT3(7@__kqE1$Z$BM
z*X!}%!2{lV?>)Ir;#t~rcdy%hCIA-#PDw{{T)@=^W!71{bcbok`RTjwGdVj$D@|6&
zh;F5&d13e2L5%8;OJ+(+a!ICi_G8t1OXpj?@AUoT?N(JqUDwpLyl){+$6C=q$)f2p
zcb6CvVXnU3YBH&I4H1J90oSJHb#va^igY<gzvv6I7;&9zS4V<u0i1MItu(Gntj+I%
z;47|PeT`RM4irVr{=oq!C#OpVj!E!$J%=Rt6JFb#v1>(i4j~Xna*Wbfv{ux0SX!6v
z`Bu}SFS5e5Qq1R`>4Z4ot8Rov5=LUJX%=P8_rLc8{`{Z&Vsm}(ykOj21`N`m+vqs&
zArOlL9o{(#>-qhEG-fs}C<fv(f9E@z&wX~EEQb(2fkJcT>H~H!IqF)1^eMerHJP}f
zNy|;!C!GvfUW*j}!8h<q2%*{TE={)Ii>=U`?<Wt3^WNw)aHOF=T|+7T>}LLSe{a9L
z{Jk!%qzQ~6!n~~#ZsQ+ap@W3OG1^EILIp3r_yT|Ml|Mky@fssklIlXTJcrZ*h&DaF
zu4+cT9)Inx{p-AR_XQ@?2~Q?x=W3i%{6ZchS%lp}%8sz_3dzn|$mGUlBTf{L#A2Fz
z@2pW$+A@(a!q!eE1?o<{nbwN=e8$FjjjhcM0Qz~s+1VM_uU;iDa%Qs`z5alQkDqey
zgL}O2!X32o(sw&4h}t!>R1zF&8Fvxlx+%rHt|(MsGMV$vkKbY3-y|<aJUuwzt6%>{
z3|_<Tr5&u20!rs35lMcwaZ>S25}$SW$+tzGEqWIdAL45&3ahkJU74KLMW#7Qm+dBD
z2$h^DEYM(J=~%^2yC>`_6*xZ=ycL!Pf~A3W8eTsS*93f1#A3BZ1#q=SWmX8)CA^=d
z6xL)N(Gf1)fV<Lxd-n_<e0a>())A*?Q)z#JiTB%82D)LwbadNN%bKJ{g{@`Wg=$un
zZMT8o>Ut*PPCr}gaixAk*PI?Xe)NMLd0wMbvd)qKk#x(tmP+DRUwValu>hrUE|8Fm
ziy*U*X0yaS$$W)Ufu{$SZ+&ZxqGzyLL8#e#s=5F48n^C#m3MyhNrt1R{F#6557^xN
zJie~MS6FSN*`AZ2en`(3^UbBxt7-^a#ni(P6+lB1-Zd6iy1&p_vu00x*$_e&6`(6?
z>T<yu&(@Kuq2Xevca;v(?}@3G3t5wrka<0F&K5RBt(>pJvmpe6cal`$0>e?x#`Xx4
zS?WqMC$n5*ZHB_5oJSjjwmDuKz(x5M)3y~<u4GY7*+1Op^z4LbGfj_}MB6Q>O$xI0
zq<oy`kpkiA*}6)wYh<-N)lIcMZ>Y5j#^F>uI@x<q?HoFKHX;gYrt?z@9k{$PmXuv%
zAe7`8Y_1Qvw6noS_aF1z)oWb4a+P{M2cTA}DJoo)B_YO998ksk3rULmfDI*1Tkh@c
zapX%jZft?}?C+iN?xRQi+?$`kum3S>s3`1+>vwMBGmTS$kXx`Zd((%QofQ*;TuS$E
zQ}YR(PU1eClT(dh7T*tq#zU1;c~qz%D6EnYLhUN*g`-}~u~l6)?n$~(a%)ztrL_2s
zWl3u>th(3E)dXLQfJis1MNo9F1E95SHoUF^y#oHNfA1mhyrcL#|4qsZkLj~`v{%wH
zBU=0yL_i{O!IQ^FeC?|T{QNJRP`lI(bs1OgK1X`CRpr?V`(+$)2@~v`vwNOZe6oA3
zu3O&?w#D7Un~D!V?D6iqW%O{gOpikppbRLg@m_P~$|j%x+-F!Uq~(P3HPRbLH3GV8
z$x93C&WiaO&z!&cH^;pHenGDv6_6M2#|6#5@Go!jpZzEA^QoWv6TbMxyDVlU8i!H7
znMWdP0ca_b)P99v)Ad3}F+dyF>8wk#uEpxr@plC(dMaKjf_-c&p%4pmbdDbWbPpeE
zo)Vr8%R;W}DI}rSHP{-Dbjj;iINx?EfSB#3J>BZ;E}TP%^Dvy_#TQ@T^7YFc92{b_
z!Kr``p829gInUa#N6@u+PZ2_+6Wd5#a?j2t{Gb1`|D1p4-~G2y8eEaL$&e9RG07O{
z@WU$qK-a*!sH^jNqqe(tt83boV@_*n1__i(G4URg3`XzCtl_&q{2u@4kN=pEWfDCt
z7li(R2S-OdIys@&@8g}~H-GalarOB-oVkh+9HoLP-qU<O$4w;x=-MCsA^j&$(7qE`
z52AcQb7Oy>PaGXn3<lJdV>+Euyj$?x2S4O5KD<wL?`N#9ZSvfMpJJ^RM%0>(AN`1+
z4X%!II~wUJ`>PAfuNo+;@sU(=;Un*~jV2S~sZpw(lM<2#6XAMWSIf9{_vEWbSMYTS
zA>~Og1)P<NF0Taf0x?ZcvBVJauZ}k?a}h#t{QNIDKKbSc{P7>&=D~f9)}AJ_gAm(F
z)4Cj8LD*ris`BhT?V<Gq6WWGq7ZKy?;&&~@R#BecoaJk_3MPFoz2C(fyTSr!+Prr$
z;JU6sEqM3+2|ipW&&8dvJ&V#gC&!i$YLxO+m128)k6v%aVj=FcSbj1zCt0yw<+ucE
zd5>0bd^+PB-#DYL#>^-viX3Bdiry4!zRu3(F8|*D`~f>#o4Cp`IXRY8VI45a5GQuw
zqtFS2$f%qn-eqN`m%Ojl^`rY;l~N}Xy$L+371%7Jh|t8dbO;{Z#-djFtV@So!I$<X
ztzBK$OGbxSWlawXnHaj#Yf92>Til!gWm&PdHfFpzCZJ>ndwCki*ov~KT}`fIVPG6u
z)@2v16~Sw?3f#Z{h`RRD8?G}LEg1l^r(JyDJUygmkB_qSne;5(wOW-|t>?eIb|3-Y
z<R!fH|N0wT|Hr=@*GX1IDY*L4M||PnP!zY;*et`jn)zbUsnaq%|F`}+Tfg@Y#YiUL
zm16hpw+Tv9S0y2M7Na$aqa&_<?T<T~+*$v+Vn7$ngPaIgKEBV3xxkBIHpM->*FGcQ
zvyA!1CbRjR#e5#iBz!vs@&Z0<_pI&ed=4Ffrp|{(Hig&#Hx_LNk5-yI&&V>1)uzpc
z=t3&#!mAcTm&q@}bkpXCIF?h+u`*(^6DUmDuKQqOWoz=xFc_ppf-2(SG_aNU?P*z8
zkK~3a-ulsizx0<+;<Z-F@ky5Lc`K(IqkwKtq4SVaR~Rpi6YZR{yP@5RTkm~i%_ZwI
zeWrAl7hWp)U;HosgmS)MHknX+M_FtB@DHx@{qNmD;4DjjP;q*Cg0V$ZbP02od+fIP
zPS@>Psg-k>%<@}*`5pH6-z5YMMll+V8L#))+~_mt_1L<TfnUe!Xideo^wLc@86&ba
zJ~C<7JMA86og1MVLK_32WrpOffLPf|C+L#p^RMjfvvR0T6>~aP`L0V-i5?O?HRtKj
zBM=RU7A8r*ry^Flb}G#4dc5t=t*tG_Yh%t%r+{=8&{^J$rgqNDfH|d_b|8(y*M>q_
z4iArb`edKm&)>wI>kg5w+bW<{xUTE9%WH|nTi#nW_AbWkm*rO~P^^@~6*<*t#Lj!~
zvir{4OXmu}^Nm7|*#AT<bMG8!;I{$HkG|iWpZL1BmihV^4V>M6p1=1W{U^=Xh;D{r
zT??Nb93Jt}M;|hs&Oj+_ma#ToXEYkd&cg%gbgTkaOQ&mvK>VQbTC*s6__8DzBL<_=
z7t(&$`&hc~6$4l+1+{BBoi{wTr1VXr+>O>)E;v?&<3%pfvI00iu}^L+bciBpv1O~X
zJ=bzZiSsq<8=BYOD0%zGQYqPNS8}g98@LK#DMb?$Tr3n)YS}Jm_xj7<`)3QE?sKPO
znfRAo>U7R-R?Vu#dC@&N@5xj_f8g2Ps&Q3KIUkDUQ7yQ9<wJhy%O6ozHNB$8^(#AQ
z-6zlU7_^lQkBz97?mq?LyU$`g^af*oVRsi-mm-XfY6!{)-2_Lqsu~4iFeDOBU0*Bj
z8)4oxJXViYTIw1m4Lt?gxihqM;Ay?Wb~&Ft<srxUK=^8*->T)fYIJlz)6!MqGnRyJ
z`o2;r<y*B(k`!5lW7pTmj7Gy2Hq)uwLNbQLAk<nBlwP7$r7W4vQ7W)lR4A5>F*>|W
zXZO0#bkUIvwbgZKuX{fEr2E}2Os<qdd(VSk`8A$?{)@4}YRrW6v7L~XUU05*xUyog
zn6s!Vs=8t}oiUkC2qCaGSmV<6F2m7~EH_bED>1f|q8^XQjG?}C8Hh$2QHhf#G8(q7
z-(=$xpP;TPp>3n^D)xi$f*be{@KgkCqVlr%DA1LWY{@ihUg_m_LEW|Jda&uWuFP)v
zEC$Crauubd4NM@}p-YF=QKN;Bis4c#^VUjD7wuZdU4f8pth=Gc=#J}3^ZrE-nB2&g
z5Ttsvs^%1Z4=MxdCW9b7T{F8|rT=vVx?B)|wY7>A#aF1M`)ECf^Usmamo6!<waQ?X
z!Mhgys%O&Oyl!xG)zWU)BnG01x=$Od$?5loj8(_d{A6cc>mUqFX!lU1HETV+^sK9}
zfYAXTDpVkj;%#dylv3bZs8PpxD<BStG73dImt84|OeHZO864^PqF1Rxljhyrm(Ao_
zK^Q0~nSR(ELrRdU`Jg2@;yt96^YSpy9)zWID$qOMTWQtHhjvdd^76azRGYrrdcdvs
zoi3heD}6tHJ_^#hu36vO#H)a-YwD^dC`GMn3Twd_i85-LwV{1TAxOM;_)uXCSgY|N
zD#hl~!k1x*23HkB%V#Y-)p~3fc^T>4t9-o5XWg2Xy5FacEeg{5H7UUnVh_=EE`U7o
zS+vGQ87OoRaCo6;`7F&jaRE!q!%ZSd^lEenSSo^-?hv_6!EM(FZjr<B+^Duwg>sPQ
zbkvSOOaKTWw)v9SiX>NvHzhIWbdTa?NGlZGh)gdRlc?&7ycRx-x8_x2kccv@TMtCi
zZb6*1grTowy>t}EMb^Wzl_Xw=Fo13ddew3V8h5&_=8h?-b>R6IPf%v7dBB9#$qEO$
zoT5;)Y?88z6tBEGL+hHVii@}++*j|v{d@Umy4UV^FIt~f0`JB<mOZrd>$WtQD_U~i
zIk7^_GSnfb5d7cAw7q1M$~6>e@QpO7T80po*CC+1mtjFs7^hYz5@hp|k<}<l6KSR+
zxKWrO#z=@z*75)`=`Gamk82pk#Z*CpianOv?>R|fYH(;DJlbeXrek%kLTf`&<n;3#
zt9j;HnaCmc*&c01->Q;Gzpr{EY0tX#Q{C6P=QsyTtLT|48bi@UL!dt#;+4W?7FElz
zE^vgfK$DB-5zlR-U`rn7d@|#?=dN?<@-EI*1m7`L&+!=N_@K*Y>Gz8atnRw3@>zP`
z?piHfN69>*)V4_j_&85I-lh-Xe8Fc@<UxzIAAB4Lry4%PgVM=pNgmjFh1Q%e#7wqi
z{VSy4H$Ja5U8R||JeCx5FcLTq4uWgo(d3OLqg-JB#7pv|&IjrDl|J4zy1l^BN~;Lt
zfyopYQ5ssyxP)F&(6fg5e1^?zhOri?m#m!>2kr`)C6}=C*{=0;Q3?Gr0j*fsD-6&O
zdyrmQ{ef<TVvtpdv%HhFnNCc6Xw}ioilPTEzvA#AF<ZjTm$6bfpP`VdX4CVpDuyn>
zUX;I|K6?>V)@)i>hQ7MjOXuyb`zoH<T-!^UtX{gWs8Sp2v$aH8ORpQahXg-3*JZ{y
zo>=~f_o`{TtO{|4&`!c@zuWnK5Je^m!lCqXhFb8Rpk2enA*9@S$V_*9XzYEJ#WdvT
z-T2BetERZx@vJpdw8l1ZjpahRWk{yWQ;DC_XS+1fJ;%qvgN*@0R{*P}qD1e(nhcxe
z(s4J&v!vHi)_6@J2yG=_$JI4>Ly;F~U9q*b$#6Kt`?{IuzIq=jA_eEY?KbJ8o21h<
zz|zq*ur7Mmi;TkVxjruKesvE%mXCd_KrS)@iNWRs+pY4Myp|9TNCH23k&WjP0?F_e
zK8wvR#T99}OeGJmE7Jl5L+Xc@QvQ?i8K^;-R#A`)gCIPmKu6;+&oi_!%{2h3#ez`D
z^V+JesH(Eb5pG@1WfyN1zhABY=n}xHaBJP{<<0J<xu)-Qak)IVRArpQoP@mW(S?@D
z>JftG`t_R2mus}LXq&9y)$d;PT4#Sc&zXd6S1EKEyZc_3=`U)J0yIIsRi;@L80j+?
z35zCX6$MV41=i+*6S|pms>zy5`_sLCYRTebUmv;T^8IwP$>M5@bsbpx%raaPy$2Pc
zxs|l%5CWN2*i1SVXPJ$~l7_6AA0DV`nXO&NG{&+lDe@v_DEX$J(X(_kmGVm!(y{Yx
zhgQ;lq&?|oT|v86-EIb}8EmWYaubA)7H8r;tz(6|&8bR<QWk<o5sN|N^^G!U6WfWa
zMvUQ{Tx&2G@cQd-#G=-KH>PQsZj8ZLi}sELucZ5upOAb~Xj6k%<wf`C%8HBlDRgLk
z)ks}kqgCbfadNHU584ET3$q17SlZWa%6}^}62Hlw$Y|yi%j+frT)Yk@jz3NfRNJoL
zX_HmsIfc-s+@||f3Njm;avB|booO;FU3LW#kWL!L7^<?2Gk_%7r>ZKv)l^j(A*>+5
zopw>(g$uO$Fx|~t6@K0Cb(x?mNH6kWT6Yw*)tsKr@O}%Cv#MDrp}P)Yg}88W0loJG
zZ@6_kV`BrHPl_=t5$)=|uI8nyCWWod&6m%fZ;scs8k_Sfnr4RGpmp;d=OeUm9<Er$
z#k-2+oTa4&Ryda()(iqlo(D{)6ON7!an3g_GysgYvGb@*91|Qh1sq$!maL@&IkwiD
z6Du|aw^cVeLOcg+40Y{g?u1@81e8+rdpUWYH{Cv!QUpA8T@pf|T$CuSFt$&o4K~;G
zV#W}Fv6}f}&a-wz2YDs_)e4Ptw8}&jealZRU9TD$U4@ePc1elInUh?R&o;_kVYOp2
znWB*4QQlV)+|>qUEO>`DHl_o6Y~0hjuBqyZt@Uj_{pmLw?^K|g&ci2RNo$QSD~07}
zT7`UDdYAJ!?~qa4%U*w42OY=Hc3%$?C_AT&R*{8n!q2iSkR#2kU-neOIX-hvrYW8r
zAERT(h(-qLtnmV^wQ@e)PCd0Yg%RYMt0fW&)1igsB&6h_6C``i8cUkpo}STKL$Ao2
z?z&n_L2vCV>bk^vG0d_oBU6UV#L0t153tuGFEWCUGtpO#cu4_VwywLDb=QhW3%Dv6
zyMM1@wB60V$W`qUO<g5yF<mDb)R7~V>L`=sT1KMv9`4*Z!?_CQHLK~wiKs(Jm|vQ}
zrXytLNi(gaT|#^PcnMOxK(w`Ll!9sq(%>T2Pl(+*n@MRjnGI>SLOPm|e@8io1lFoH
zE2$x*&R#9Uh=O;{=Oy#mg2`k`Syq%)9f$ZVn)V1<Nyd%VEy2rD)MDBB*qU8y7K?HP
za}VTspUh+^t;lkN(TXg$WSK=PgE0!5=M+UjURbiMI}SiMfk0PRIPcKPa(HyW>FEi(
zyE_B|*2=7j;2qZFNDxpmzI^b6+Bd_Ao~1);)$Td8&#afd&2+T1*~Z=4@3nk(Q8G(c
zNP6!jyU6**14(djh>YZ&Z>o@$(dc0DvD>PuYcg%IMVfEZj?i@>FrUpSc5|{kqpGF7
zd0n?&mh<wrb}bKt5HMC+q^BURQmHz<l`q{j>3R&-L@y&6Upi))r55tR$#tb0LRuV2
zF6r`a2tm?3REzn?56d+Z>Y>tFhKA2nhyRocprpCtY&K^$TQHl=sH&PpRpaL+C>mKN
zd=|*Etf2v=1=cmj(lyG<vTD4lv<|&|Aacs6=%v#bT_K2fkrx6z6@|rE)ug^_Qb?ed
z@xyTr`ol+$(N?iOUdMY+9u-UQj?88e`ikkz-c#37lO%S1$DM5n*kf%wF0@_PC{!BP
zuub`EG-SuMKF=7-7;txCmSQ1Zx=(Jbuf_fK5s(q9I8j0H%N|3Ba90S5USILTi&Ya$
z*Lpc?Sw+iVGB13ij3ihlVPF+EdpU63ph8jc$^B_18<mou0DLV0GT%`w&V**rQhQF-
z5LSrzsJ2={U#Z0aQ7Nf_>L^3M;`uG769=lYWImfSolXU=1oB>=LP`CF%`DzYK+)IB
zLyhEj6Cuv%iPgTDNdX;PI(j&GkE;S+1^T^$JQD_3TkEs7Hja$5Y-J}{K%QTU*Y@`I
zc=YHYSFc`c)+SkQAq1+}sX0MoAq1>Sd;F|5Q<OBz)%XQ0*9|4b(kg*fEc`r-nY_Y_
z;9R9tr(sV#zwY$h3ShNjIPBqr!zqW>0TVK*Y<2-(meexUOn4!<#?lQzarf>W#^VtQ
z_SV|crtT(1X$ejO^GYd6L5+Ma2^q=z=t3<m8evR?M1!>NFiMfStp_Lhe%h->$f2Od
z#Z1N)HBcB7CK_y#K5LW&$P*fnl!lHnr;%q#UYt`PVi+#wC9~O#`Fu`YSD+02;TkE`
z93>f7CUDNk5n8a)Wg^6^FXzbfyyF>lKJOvmseB2!As=SVScQ#^H3ox0l;<M&Rl{e=
z`?d7?edhByAAR%@i^YQL*RDZOSlyx$J_Hu^qEpkd?3LN%z65<f@YRJUy4ADGM6AvQ
z2h(S|_+vMgvZ`#l1k<&m0fxh2Z0sf#)x@Hl(5QtrwKHvIlwqu`Id0yrQOYjCN2`mV
zv~GwD6)l+%BwJy{#qHj6_kLBZz)m#_II)NY1&a4=B;zG08jM)5WS62lZJ-PY!Nul-
zstt~>KpWj;hKsyK2v}=bUtec^eLb$T#x+VoGMYjNQpp+w!i?-ig8U?&A#j$=Jn$?Q
z^X8h?TCyz1R~}<E#whYas#OPrK7&D@q9~%Eqr-gddM4=Q=;(-}qa%8~9?w62=fY8)
zAGa6t`5ae2d-Fh)xw!Lm@-;8Q<h!p6G<rS=pN>S|s~w{g{~42VF_$6HM6YY35rlxY
z1{EahPHUMFuC=rQk>rGA?1`~ZE*6Z}#}q};bUL8p0Vk!Ylme|A{)HG%ga*p&8bB8*
z!}fk7pP^!d#5B7jrU@I3C&dUNZAC;P@tIEi?j==K09!$%z9pGNYY?x-2Gg-D58A(r
zqM+aJx8Kng=cPwmN(Cm+7==FghU-H_7Mv50Aa&DrUSfRddRdlXwWY3I^ga|tFC#B9
z2K_$6;jl?vj+7^fC>PMog9i^Jg?oLSq9_{Al6Zf~W7bh_5#&5oRk2tsa6We9?PB)r
z1(qvhTUd!D!lV;iy(mzg&YAvQ^qE!HTU<+V8M^^th?aZkFqu-KV47U`=tcz1CtqK%
zwN>K1CP8u+F<2U#8A4lN6*?$WBHq>K>OP0SV7_B{MNcZl5Td7s0;^06A_wuRqQ#*I
zag-<#l8H5-LGrq_kybBBpm(7v5isVQfRX22>7k?)cyP*NbW}>w?NK@vo+Ucaa(!bW
zvm^^YFdnb9%%sKpiiIGA;7}ThBBLmBy!T9})6V^cb|hw8tFkP)fB!zy>6Gi&uTvC7
zOJ|`Aq00Vuy^-{6;@zlcE{)e{2~awxR(PzX#V!wioRJ0R`D~SUSBqG~Qf2sx(H1{1
zkdMZ@S`x!Upe)NJUeEKKSv@Bqsah#|y*{tK`dX7P(a4OIXOY+1er0lB(S_(7-6Wds
zk?2Nb(eiU^6FpD_w(_AuDUFSl?*XY>Q9;G>2azw4QX$<_Oaw{Wlj5zb9^&dfhTx;}
zY!!1_YjkLynb4ZO+N4xy8o47po%kz@Kl|s5?iA}(DiDzC_Im|cmNk3U8kaPI@$~QH
z<b?b8?_;gy`t|D#9d(m{x(0ueGii^z6A3LW_k?d3U^%VO#LlhhVu?^y6BAe_go_kN
z`p!j}MCYu@NGQ#H#cPQ7v3$AK*^j!zTY7`1eEHYz6T;<qo_20WTCCQq3NbxVW_KHF
zV&RaqzP>1{>LLP9KYa)>rl4X0d}sohjl%NnecPn8qN@qf>j*R{gDSdKP3DE_3Zmuy
zKkY1#W<}0@P6d_l%hsDyG%c^2b@^u`Np!Q*VkHvBYSzc=42QkQu=WC_+t&Yt5cu%J
z4|(_9cUfOw=jP3uO?FxvjBlT<yWY$5SOgA|EUafsgQS`8%X}7%q7Ge&uu^D!-k|H!
zSeIA23aD%JEw8zF>Cx{adGX$b2-(Ixk6jD8*-VYlOme|Hbe=&_+`M^{-Ak96c&2kw
zm>4=rtgdT}lfpM~#vK~*ib4lfSTzvR^R?bUR6t3<Cl7i_CB*OPltmcVC=ij1Hf=Q-
zO(D?qk&;y9fI@>-GS;<2@2hfP*`sD<FF<+7Vr(ZVUR0Z?)a5gx_p|IREQbYrj6)Iv
zS#H_b*rX^j(Nz(?T~e+Z5(HXR6>q)u76%6hT)ldgOP4M!Jy-X;O$@6uw<jr)CQFcj
z?Z!BkbFjPl$>%NaF2kk(SXf!?)uVfk)p~?<{munYYT4xvAwfhP$f+S1)xcO;=J|_%
z=>y*U^e$D^&iat+w&|q}Y!G54Y9vDCRnt`L;|RL@Jjwl3NI+7BfQ^;h-q&<JH}Uwg
zrl2V?o!sN{S}URz6!X4Y*Vv#^Me0h%ig+i<3rH!2%hq80V*I`JEX9gbO&5Xq#LfE2
z&tx?lP`a^TQHrW8>Gg6hU%o`18+=GXL&Yjt2s9$k`R=>#@{^zZm=|7nk?rkmx=>7A
z*Ep@0tiJRdU7^)YOI|D%Or9|eD^=1%m(E3aoTsVflvD98mxJa1G(KAmtcMQFl8mIZ
z|2lqVFc^psR~lDU<Y^X9#OhPVp0x(;1HE34=dL`*Z~mo!f$?}vltfH}T{Rxl-ud7g
zKJ&o~w5wVeXf^9FeZMtWm*Fo%ob?eLGM}8r=u|30Ye@!Bbdd)*$7i7x=#Dl4Kxx~I
zq`l}!-cV8it3}2kLg9_liQbM2U4_~1Q~TI$tsN})<MUqTb8K#|v$eGuy^vaz1m`{3
zs&W{ssmh9Pe)F45CKFzI=_Q84VY63B9y+~Tn3`zsqCs%hS|+CxEXY{KEVHbwt<mrI
zn}^A=EJoaAl;hdi8OeQ=3nlHj8~|(u*Lhbo*+7yD?MXS&O>dgkf5gRNGZh6;BP2T(
z6eu!lD72PAF}2~>etnOB<zHLJI|(9nt%QW-cL@l9x^Aj1RRE&(;$nN3<bfqZ5uz6Y
zaxTqAXssF-mNTQe?x6^c&{*N!wMhc8iP<FR41vgki^SNmFk6A)o3)n|K^ZN&YPqKM
zZpbx67hW5*Y8zV09=EQ1Qi39AB=JCqv?6#<uUD|SxkgpZnNCXr(ZE2**hx!>^VyWc
z!vhYF4p|$m@rBQSjsVnE)p(nUqr6CkhB?|HTJ7^E!q2iy(jcET3@b>k{$S8&G#ZlU
z1;q79f_Jf4aCCIUVzFpw(-`5GZW2dA2!s~TWAsfdAkjq-enYiJgo~Mb9vMkn6q!Ik
z2(6Xg(<=&0XMV>kw|4MlpV?%}=FVmQ>~H={+<xID>RK@Rx)wh3u5M_?`x@seR1l?w
zbJ848ci>J^oPZeTk$7j4Lv78G>r{aUe1x-;an&_4#IScL6>E@Cw3&{}V}MmOW|fKx
zjTFoe30?sb4@ayb1i+JJHp-Rru(osE(F|gsNfuj@C2f6_JW`DFb|5Mfhw=q4<8hT%
ztgQ_h4tq={r_^<YQZ&$)b21s!d*RF3^pvN2Pnpl=T)upnS6_LF*?d7&Rp*K_6Hg_h
z!CKqIz@UYUjWHBOAsJ#jyO+6j`xc|&h|y@&xF?d5HskpCn8{?qbUIye)o27!2uVwN
z+ExvsTv9Pgx<O(ZS+^iBSr1)I-^H%Evw0<(5JJrFZnIWuSEH4<!bXW@o@bUXe!gaq
z7c5GT#Pp;#A%R9(xA6ke=uT=1tfb<R@L-j)KN=~3bn{(~Jtzf6=~&}XcQAV0SWVrZ
zkR~kZ=6+;-LKGTpR?3nB0*N4^mMp#2I&rl1DZcr=do&PMr=^B&_N`m9aSj-4?j!BB
z_w5{+OP6;U4Ej`+5NEP#yjyOL=$y>re)#Y{`%j;6<;oRaedT3({T@|W5u|f?Ozn(H
zUoK37L%NV!mu}M98DpO@01K{Mxx#ZdZqV=d>Gylh8kc3sbUNkm@Q~Bf(`G%pN1ErI
zzLPwyv=6WxBS=C>=v&l+@KAeBqeP%eqddpFoS{`9ILG$Z29vW>ZeG94{f|DvdC&T&
zU~`a>4~CfS4Q}6kjXO7Qh#-&mBVJNyQ)g*~j*mu05qq62)vYM;*xUM|HukXO0i&=H
z&u_-HHo|w=z;UiM8dL01H$1iMrF9Vr1u+&V=d0_gkzX<avx_Q(X5whtgLKIAC1bIL
z;fxlD^8Aq?tX2ldHN@~L%aXxh!0zrY#t7Zib=6U>E!`-kSS)7jAMEqs{)fya6JCAw
zRbIOLBHnw7EJKC1lZ1@k0N3HsRphB_q@|z6XzcA<w|VZy4VfkF9EXR89334onM_zL
z7AzKvr4wBQQzc6}Y3=s2k_x<xZJuxK(RG{BHLNi_e!R!u{OdzpDH$Y1mS7!$d~$2Z
zvrGsjwqV%aQrx}0z&jPMz2a7e<sh?UAsMXh^q^{ptHqLA>@N|MTlN%COXn6#+c$-I
z=^15n>LnrFU5jqKrTgA$0a+LRYhzL8A%Jwglp)h*Xtc!85_>nXAYI=_FQjEo@%{u}
z+Su5TOgW{P&lgR>UkYX?Mo2@;j*pLdy0^!0FyQv5Kh5RKm&vjWT)=7xeCp`M`Z#;Q
zoBwZnZ`LH&k)?@!zT6v2W=3uR5+p&8Mb=bx$y6FirJg70NhZzc-$=hee?-4wOpkh!
znPhsA$+}xJW;La*DzZUv13?f#?h$c)(ZliMe%v!yqsf;_nfWmz0c1wp>&w|!4Tqw1
zuvR1yo%cD8<AZczJ^%DmoIQDxY3ls^9OvifxVgE(X0y3dalV_*?rOPg@ykFb8Ns48
zN4uC5@FKg!;P=1(clbB|=KsdDS>tHAz-nf((j&GPukhq(#C$%-haY_edvk>$Xe^dX
zeDcx9m|2Sr%3VvJef7FzF^_i0OuUo~C`2A*oN!SFt@)0vwW+;0eT%I&J|k->=uAm4
z3f{TICu+VWak#BJlei?6l)?jiI=nhwAo-aTrsUZ*$tBE*1*&>>=_pWKU<}4+xQX!M
z$&)9T&*ums<hkTZsVdQgy+zYB;qBYExV*T)hi6am!P7GwAD<wmca6rvq9`NmDVVf|
zk7aM{mYk-k;%$7t({_Trz1`sbyLWi=<_)f{uCUpV2C0;)a{c+Bvp=&6pzF*mTI7}&
zz^*9i?=1oQx?0k?ExvlW#<OP!`0N*3++14x!{6D|h)e;R);EkJHn;2Sho8+Xl%*jr
z?S;tAOU+)Z8euX*k$5hj6ML5~6s1%qQr>4N{z*bFknb~i;hG8c;GY)JWdNHmlb%_8
zr&1HA43gUF*fnyPK3|FkxtNXJTJ4<yHIX>yFrN)LK03l|HUmHdvAlQL>e_5J>1Rab
z(=_4w`Wo-wzsJeR2|juD5mtM9;4Y-(meh#e!KX|Q6#yS5co=x$#%bdr(wcp!u)ev$
z>({UG`t@sEUteRp-R5dvsXzh947M;lBxZD%Ik(>{I0R&6F5xaCX%yDvyYmZNTwdZ|
z{mXxefAwGgH@JHH8h`w^e}^Nxgc&U6%O%d9eTc>05?}t?-@~sR4p%EYe)1SD2EYtL
z)3236dRJPv(IQS+*2{5`g2SbaOB)m6Y&^MlrBuphB&CMv3)wvc6bqwO9_8;?A(21l
z)+>@XNdCd!lX2vd2}uIZX#TeaJa;^=dw<C|uiSeGRK~DeE-{Xj$++2UGJbGoBU}72
zQX2ck#RWE-4L<zvBRqS0hWR)or7M*hn26vgwOFT?K+XktZQxM^DzA&*Ei+gFN`JOr
ze)*3$zj%^dRJF8bK~ig^a<Uy?M4d%JpWD@ewkE2=B+(#YgfavE`G5Kh|Ji@>2J;d4
z`p--JKfnJJemh|{jxd9zw8$8-O%JekfE@>@nS~o7^b}~0FC&CfcgCi&CdLON88XEM
zYRgC>iWDFkDrcS)2B(w%pY}v4a2ndwraBXiBn)|htd-SntGw!bex<o#(eUnSNFG{S
zC{P@Mta}6RNuvqTgSz;Z%n>&2ou^slr>Cb_t@aaf&>qHdq|c|6MoCm!>+A-7_Usv+
zJb8jaE1IRA<te#7oQjAlfO2*M0zABrh%nh-EYlh#!{K*XYaW~n>rY?&3Hy5o6`iq`
z*D4P{J2v^Q^~|t)x*@<W5dbt0DV;Sy$ifQ3{=or$_q)G^HUnlxWBvXOZeP8?*=)qK
z=O5$X^ceH~J$&>1cjS(m&++l6pI~`(h-+&xML>B^(<g=U!%7locgcA_ycd=Jtzz9I
zkvl%HQkt2oqM|>k+>mfG6ype-x#?*}i7pvLt@aGa)Y<ZGDHvkZNCu6db;-Ai=3nlr
z#A~b_fS8K0p%5cgb}Owi4`l@!D8-V_In3t^JbHAB<zk)`2*LqU*kW>P7N*Yc|LLcn
zaC>`;XV0EtwOUbu<mBiLX(D4#5ngkzivX`YxifuA&=EAoXUE<PR*JRO)P!@h*-~&u
zAQfLQFG~Kbb^p-8$GoYu6;MDIIiqKn3CYN=vfGOhi!2On45C-Kx!U67=n5Zv2z>Lx
z!WvDkE3E<N^XOD%6xJJsi^~yDjyCDpiDp2yfXXR<!gN);gWiY%&bxeIY;onzvgw&I
zgzWd_0JJtLWq`zLj!De#sL!I+xN@9FxHj0TajvvMr+2c@6KjX82SDC~B%*AgDdbp@
zbD@~EyyxKHyvKY#$H~blmdhn57R2Vah%+feqd-$2PAP@=@89ExAAZ2$YK60>Pt)A@
z04M{WX1^c>IVVM9LZ!kIoP&nRg<(2rolyu728v_32Ny(Nuh*E*mH=w^l%xp>QfVuS
zL)|5r-72Le3V<8{W;YfPgnKhaigN^)ivBcUje++QZr2kQ^97!N`WejopYZhf7-wf6
zzzr5ZonK&ncnBciIykJ|1T!w-U81X4Vr5z=Wo3^6N)+;C1wc_+uy<3omQ=!5NsmB{
z=gjztxu&trIdwhlx8XT_mi&H>&k#P=3JF;NKGROpKY1n99hJBfeO6k?P!w+>D?Z(c
zjwy~pJV&Q_E9CwKR;zuSo}9uCHqjW>vC%V0X*0IChP6^~-r>#bAMx(pJACxfM>se*
zz%*?Et!c1zN~2a#aM8mlYEsLINWlk0Kf!2fTFSGQBDpClMNMleXdDMir2wt@B@cX;
zAZbmy!n~L;Xrr8}%Yj&;^AhBx3!{36q=e=RaFGsKMT&2Xi-=!*euEePv_w=9CTRs1
zn>7W}j6qZp=NAgEzg^&yr#ILH1*Dej)Mg|#CKDzzB{u~ciX`?EV?^*C2&5qzAe+jR
zMyv`kWJ2+YkbUlZ)1GB<*%joLmIWiP>?kXJ{qoP^`bbD1%=CiTJddTg4{1G9LSp1Q
z<sc#T^a|hr-Z&08Jvqg}!2vNXPYOpwNe+mywxrB#M2rzHU%tfqx9@Ovc82+U4#(!U
zqSWb3>?#$5g~U8Kl{H2T?0re?>XiblHHZn>i^)%k1C$l?uwozzr5BsDHi{C^Xw?nQ
z6agd73>=e_bf#fUp+&-dlfSD#wM8ab%|4{|-XXQJQ3m7c7*`iRLc!x?xxhCUm-xer
zuV9Xk5yOO^Zm)51b&b<u1?3{q7ze(!h7Y3k4^q&YU68SHyM=)09TX^$Lo<$oQdm*Z
zSp@M9iWhiaXe3(@>67fP$x5PdJ4;uz6e!ZAEa|hXw{|^C_Z8tQKbI^hV<JM><g)-0
z@AMufWj7K%IUois_m((5I>J0vvj;z=ca%2pN|D={GRLSH<7TtLmtTAdqcuK$_AGbM
z4IyBg((YL)BQHZl@a(=<g+H2vNealLA+!aWk^$!^yUhpg#*4ebtJ)n^G$wfg!0x&V
zMpmIbB<X!u+scREyw?~_)35>+k{KKE*=Lvd-}D{;lB!k&h>bNEti^U4@Q?pE!*BoN
zEtHw<Bv4thuarVmq57HJgZw`2w_bZ|`nFb<JU3|`^Vzi)nas^@-y`osuAl#ujH#S6
zHFTs5nG$Q#2?4!teZzIzgT-Qw{rx@6X0z<l<aNp}q2xNUss1J6{rNeruC5S5z}eHM
zn9pWKxFl<ZEzYPorO{-UX0f8EaLllF%KOiRPzzscHJr0}XoZ9TvG?L0DR)?F*+sjR
z79;!U`QOaU*y><lG?sYB8Ex5wPRbGoPhHa1Ih}Q}w}<yo(90z@3K*9QTwY$nE|)kt
zUSV~3fFFK%h1sxzwR4!EWNv1BMsWk@R8?RnowZ7j+#mnG&{@s&+n5Iby%osv-!ir+
z|IQZwu6y~p_sIKTi>CapxRUE`$;B-cBc-J;7IPdN?88`_JwgDeWr#Ret^()a_Pe)l
zvEFPjpU-i)TIB*<rbn&;le~$QjytX@P&R2XwTRASjFCQ1ZFP*cu(D1-GA-%D<^{Ef
zE<e{Y%l3fT*W0$QZa*tE7mwm64FUM<vs*0pECqZbHGvz30h6efwASGD>m_F69Fvbp
zYSZUpWUaRjz{>d)ay@jwwNPbYD?g)>8Qu2&mwVhLjHTr#;|8r;tF`##`7yMkhKn*#
zvPa}ISpe{UopY4NnTTjMo8j>A5aU>aq;duYBR>0bbN61Q@c#WdzJ2);tJNW%o}J}H
z1m1H!z$)E&Vqh|mnYHT3E(vx47MB<AwP=+F*VnfXE(?o`$^h(C4)|PUTcMWD<aZd_
z>^xEg?+&PvPNgW~Z$3qusG%VJ?t>3L!n^NZ;>W8SJVM0s<Pb;4N4Q*5m)XZp9$_(C
z!KN%opQcYT4otDQlE4~ypm((U1(01)e@S^Q=TNNioFd!^ho3tm{;Z_aR#^jf3k!aZ
zq{I6A_2f~uhG|rIEwI=VqBYG@S*=zurnvA+)))QBbjE^%dhG@L`05Az^wS%B@WF>z
z9j;`4LRRY273AlpOF>BuFYiT+xd&D*Mn=g%Z3X1T<rOS&x1D=?kn*s3as0D3h$k6L
z(UP*02b^{&kpK#J5T54rk}1pXjMLK_Jbmi$_16QnL;&*ivlys2N-MxxVlHKp7Olk?
z>i!u}YxE(Oirm^_xjP(+g3YGd8}SCx>gC8%DU4dpMV^zql-44V#MHVz+RUn!nGi#w
z7SDX<#r$p)9Aef~IrF2;Frar>V^c9yAs%Mj##*?|7GHn$65DCQlSfaed1DCJ3U^*v
zM??-BDdcw3ZYo)0{9dJXDshXfKxcen#AZ#;nh`N5TwPu5M(V#7w8<Od?`>ebwVGR{
zt;Q?l&(_Z%Ccw!fGAN;%XE6O~MFDL+3^;xI5w_Uk27yKeFXy;gZ?M^JaeaM_)xl}L
z3u{e<qq6N0YvM6dQiOtJyp)l|7D`<ijYMa)vK3=VVBw!h8f9pOR7-RGJ*kENr5LcZ
z0xLSp!pVROstfZjg@U|xfHKdlwVBSO#idkH{+RxNx`^-M-Fv+H?t2(x@ch}wSS*%s
zTL+i2yCNuYf|F$A<(t$YgKFqXrmRZz7Zg2%d8AUsHC+@n4Wlh$P_W#iM<TX7K>m9@
z>8^IG$O~atn5+rmiy(xqr^RJq6q9gCCmoadMBk5y0kAf}4xC;~^LfW{hTHWTi4Jnk
zaz@x}(t9|U4yHj8tu<v)i9NHq%qDHita(53EO<YB;2fN5p>_AQ=U;0JWj!)$i?vPp
z^Shd;Noy!~CWDyLYKbvooQ*hKtuUX@D*>idY7(h{=*oWN_usvJi$DM4A2A;W{QB3w
z&YgF?PmLtm+)$(q)0&EJp4aHDuW&|ddanyUOAaCoir4UC%lRqH7fYO-o!vS2zb1H3
znJ$%8u!F(z-x;JvE!|&ClaiV7Gs(JaQqsM!U_iQMBDOL@IQ>8P09cF5n=34i4{`kD
z5kC9;bG&}@e_?fWgzKwo{PfdLI68cqwCp0>DdWrENmS9A#uccF#sH))1x!n9B@~Z`
z6ds(tSll_l-PY*x%v+^PS~$C`dp`445zBX>kX@-h;Uv?#+%N5q3aKSe2q9u`ZyyH-
z`#FW1lq5>;ks$%H1WE%#0C@4@1-^ax5+6PN5YL}Kht`Vbt!M=unMVOJ4(ZQUm>1fY
zu5_iSpp#qL^Id{lAr%{da7Gy{mV2<=Vy8vsUFMrS<mynx#TWBSYLet{CylZMp7=jU
zwX&qWXXdZry+PE}VN>ad+uK{*-rhi|Lu{ul73pb3*$U}E)G%N%Ut&IJTyHspiL+OV
zCBjTdDMeM8S@>yi4!Lvmoi?S7+mI+#hD^2RR{I?9He-~5xY})V*L}E)09rF2igNAg
zdLxuIaD)Zq*G!1zDQ$)jBQcTfcEWP8!0KQh!(ei2iVz}<H83NMI;DY9$?XsohNK~V
z|NZy))1Ur?<D*0T^0VjA#$&U+%E7)_YD(#&2m9Y03_N#87H$d3daLoRWS_Z8xEgJt
zwW05Oa&Jr~;`-+L!MU(FS@YjVXRUDIduN6shI@$qUmWWy_Y8=*94BE!YF#qL2_cZ8
z<)fiAEgcgR1KzxOh4uCl*PCm6^TTUknBn-*DIPsK#C#kupYP$Lvk!54xWf76HE6Ug
zF?(v8K0$HEZ0`W?pu7ik%t3Wl43(<yy3hqNU;vQyN%ZefDYVjBXwOv&2-z>KjaRnx
z3DlNOexBBVp<!A`R-5wyD#dEk3h4PsR4Y)dE_mSJ-~jvkODM%zSBaj60oodX6?#PI
z09d7A#}VE+{OON>#P_eh$7j!<;lmG}VDdMBOV$Y@7HQqmRW&Jv2%hDUT>XLiq4`j|
z_h^`vhx0BA5Ur>s5({$b6&A5|5!W}ju#*d^5JfgYlUgmnzGz-oYdR91c<^Mt;8UM}
zffU9dMel1(KgUH75kMO>+E6Bk_bHt{A{vj^uSdLnyN6+*IlkNNHuZ5*P=>4#?=8;G
zUL%5}XUMOLWgjUOQfGwYm_PuxW$~Uhp4vhPN%L(HmH(}jx}y<E*OB~e2q81acHW94
zfHok=z`tWn#wV+d?x~odtaDQACx06NRc;ef@28aOvd$PHpee<VGNj0Ypd4`Olk<l^
z`~ly5^9?@#{Fiuo_K0wSHLeu62e+ta<so>u7~nN$f3OQ6B>z3Y*<&132gp^TfHbSZ
zAcnO6G_PB0Md^(Q*n41Uyj&qHoBB!Zde`0GNlXbqP)RdQ3X=~3N|nyV^3Tj170=Be
z@X<r1erds_TA&ya9I+bbXr%4OAAJIa0e|>+f6PUXXJ==4^ymYO!xC55*N7=Ais+@J
z*=8Q0=syK=|ML1uQpa7!T@nWB3@OPwTBR%RO$40fS#TVIX-&RYroGEpuY3;2`N)Hj
zTEPUF>6YSkV7H`H6v!>SO4hyB^&J5~RH=8C=qrfin!UWd#NYq@-y`^dU;X-5SRE|j
zJax0MmOEi2UK}H!QgTf2q(o{SpIcIYp^9Q(TwJ6G3<;m2(a^>wei>6&4HBYuu(-XQ
z!v^mXfro7Nmi8a5tEx8fv`!Pn(gIv8j5(;n;9`u~wIqp!|4y-Kd7%o<NBEGYjs*{^
z9j<Q<5aO7Iz*_*6<~xN6H#au`ZUKy#FTTRf`WjnjE5<8F+A_4lkdbIt1$U+K^O1{S
zm3*drkMy0&R9C49wIDT7KI@W)U67ogO^hOGNC}ZNC_le259uz6SglrAt!Nk!JzEr?
zg#c|+kyHpU%0O9_=S~0g(@*%PfBGjl=kVL#{uU=Er_hRGE6J&j&6Eg2K@6xN1tl`-
zDg{JTJ;Yd>ZmryOAh0IQO3;7-_cY5X7ce_I5d5{l2VX2eu=bj|O2?SBcQkaC1qdHs
zj9EFOcpmP!O794zz-i&6d2Tmbgp}#Cb@c8DWUZ>?4LDkz;Ox;;ES3udN3P*_@2>FY
zFaA%=#sxn7;A8yavtMF=|A1nM0w=w5+2avHYCR)d!NkN$>k^6I(u*PKP0~s0w&g$h
zdHf#9{N)+7-)UW@{BGG<1L>TR+)Zh7^!()WDa{(ox~(-hJUqm~!9j{?kYd`#Hof;y
zQA6nxb8ya)%j(;2ad2>ePd@nsi^YQIEG@yu)PNL7y(uGW{a`v|MLC2}x!TxRXz_;k
zuG0Q<MZZs0RiyFP)+ksQDhLaaTpK<FgvbG-P%#y9j@9m%7$bBbvnZQ<f%5#lDnTY*
zSZgw-n*cZY&rBndnW$ofiUCGXaPAa?9bgU3+ctW@<RbR>FYsUfSN}7%+iNTqdswaJ
z_~{*`7)pjM5vp>9h-?mK%DdtUyZ+4=BmcgO!Sj3N=cQ)V*5F9(h^o99A#MJ>d~b}z
zK9gOV5vk1_@!n%TpJBCPtD~6dtq(Y7baAGOK<amI-{R!t1gEE`IWXsA1b1qo>=QPd
zi!~RKDvEX42cM5T1G%?0?Unquq!83=jj&q5sq)!}0!n*?I3Yx8q#1lb)ZA59#ufp@
zqMhe+NSRDX>7w9sQgthr*7^t`AQ*!&^|YHp#HIJpj@B^+67Cui2o6>&JpT9t3;^~{
zk1$`%@zH7@-+uQU{Cp3a;TE@(hrPYU{>?R}*+{NLrSeoi2c%3aYH_pKY+#(rR={5h
zR{w|4*{&E6e^>G+79OqZSJEb@$1>CnV1>|%5h{?dSIom2I^%uq?Jcod(RkUCD1$ru
z%yb44_SzU+TwLJAix=2#Hu&U|PjGN>Py{R@$Sa+ch$N8N@{37_(^}K4p|p=S;YqGR
z;!mB~#x6v9s7~=8s{n%t8$5THl<B-No1|<J&>Ky40OU`jzW%HoAtAj-M9`ES#cP%p
zNd+qT*9Fp(*-fU?pdyrZ_~Oq8m?j0pEu15>G<b{O{`M#QxBuNfygy8i83P{y9cmvb
z|6OW2vMVqDK3iAmdos0BYALNi;=l8SwFo8OE3FABOl3wE3l^!JNp0-TzOt*4`rJg^
z&4phsmpDB=#W0k4_c}2$o>Sfu;HWK(Hn_gI!8c!jjTj?-^PAt~^WZaK<R&yoR)!g1
zBWrj_ZN$m1QCi}Zy|fGjC_z$rZWPlBnJFp7Emx>8?V-)I4c;LJpXn?{j$!O@FUFyw
z1pt5#F#H+DgBalzd)n_ZyG+mNcR?GRzI=(_{p)`X&2F4j8$^>CNV`0H@)#=e$W#Ms
zEKV-YarWjtv>hpd;rtTni!Wg9kTU2>jgB`4r+@eZ0AM>&&o3kfw#zldbl5Uh*~UcD
zkEGAKB}i?FwC`!@O#Yq!<Fk`%WypV5sv1el7NLCR(a{kO4-cVKnO9S+qMFxv@1db!
z4bA!a@y9oK_39N)4v%nja$NBwfZ7e^Zu1m9hP4JJ4Ff}DT2cA#TbgE&=evu}I8lX#
zbp*TTC{&R}ZYd~BDP2g>_yE!4$s(ZCa%62gk{07qAjt)vtBT1Y0RU<y=*3{2ym;E?
z9rqEx!vzm(9hS=%xV(4{pW2qsXA7J@n&WT(_9cR!LBk><!st3paF<EUh$D%u^_xn|
zM4H5MZ4#u`)IO`#0OT`y4Xp(rg_~S!Yv#w8sz$*ngw%(Jdkn49e2J5j<5Up1g>$Uw
z*E0$3SZ7+pY2f1G0xw^Fjn%;cPEJm+xw%0Ebz(Kf0PH4Eb>@gt8XFf<i-csZ2M;gi
zI7ltB&Gh3xxrmYGeb6|~^?HrjYz|`vVc{qkJ3hEpCQ8XbCRYLL)UOF-QG_ywC8T7J
z%m|GJ0^nnyl;%ji$%0A(S(Wpegl}Fcg<uTaG$AVB@rxID{Nlxb@cN$s_-FUncW`yj
zCdoMY&ZXur=}cOqlCC5zw+dVPbNNh3ODufaKzU1NyOhruC|-&fVFpT**T&%H<~qkZ
ztR1k~+ybb3-m#)4YrnMC#Is_=_pe^z-P^Z#{P-~ziy3aOuPBf`Sio2Skunld;yu<5
z#9KW|+z}DZhumGa-TR#wOf1$KH_>I!hjWfng*mnxQygSA!ur5Hwn~;%j1elirL3(r
z)f}wh1L4EhF;DT+L6grjHYO7k0!#{QM|K5+o91y!90ghfbRggB6ap^ZZ?WC}9iDys
zzhkrAQhVrehNn*tF<*R)X|seiv@tf}JnF!-v>pKL{I~o$BfWGRwH83@N0ox0HNji$
zs(pt1H(z2)JgtS2e-D7?s%WFhC8&VSc8$eij^pFw6l~|#C$#|L`@GxECg7(ZU*rAx
zIX?OHIhM;MA!4YlClv!MTdKwo;zo*zIHRFbTmNaA;9}}7koF>lMr#rBdGUF$Z+N?<
zOtI_h+v+^FNw7S$EG+UYDp1{(P#RKu91<1)jZ)RQr3WAs9nF?Xa(M&|lLiVA6yJRi
zetD%e{OKwFm;e5MfJ#j>f#j+8q(!^Q!8wPw@805zFaC^+^Y{7v#uz+*{v5ylo8LgI
zVi8A(s&j?vPnCu^T|a*g8|!$2v<iUYn&scMG}hADF1nH5{W(j6K`My+UVh##I;({k
zDQJ|=OZyGM!%q&!$459iIugQ*i)G1N;AO}mV!Xy_TI2gy-@%7~U;gTs#CNmRnQ$c~
z?+lDTNRt^Od2GDZ&}lzB8(s3Zc4Ppbg|zNiIg2soZra9JTwGjL=QO#1^UE7p&Jvaf
z7(#%BP8k*oT3KT8J|qS$Odz_sY~p0i6Nwnavz69}FqNMT0JP=In$p?>F_KA^4#~MA
zoSdBCpa17;(!7$ZWw0Y=!xC1T3NfWw%S0u8ru_5P9VL-m{_f`l0YAGX3K?8)g^WZ@
zd5Mf%{9LKEac?YXs<-itC{mO+n+>$qI6gj3rZn$EA`L3SX5f8$@A2Zr3#@Oh@SETK
zM;M18w^@k3SXPoDwm#_)0%<Zz72Ie9wiHBCU~=~Yu4mz&%P|A(8Cz>Hj>Ch?!UALr
zFvVYEgoa6hNs0@Fd<J*h+;HKSv`|TA*KQ%DEmBwWe2;1z@%mr<3vd88BwxEqOv(qM
zZ#Ql6WWC0rb4lY&#hUXuz8*$cK@(b&>q@pK;Y^*(h@FFJtZjx?>5??T_#{E%D&N1%
z#@a<^Qqz`Ncs)O*b}zRAq|jLzIK_(0G))-CF<A_Yybk$2;zc!`G2hTy<Lc@PU;O!x
zI5|4R=fC`{SZp??29dR5DlHnxGmRbdK8InbT*Pd>%V#leFkQBHA+3=1=R{|=xUeLn
zG}=Hz!y1(gg_z<TsybYG5KP!bf~%4Xo-f@XB4r6dAsS9s1AHji0r?VKVUX5ZESGz5
zYM-QROk+8fLUbO3uGduug6gb`cF*#1q`AvKYg23`g2?YHr81+FYmkJ&guve^W~jY`
zwETC@P~x-a=Sra>XJ*G3v0N^3baaH}k~+Xj?SuPZwL*@O!x)40dX1Mazd~>h|Ln70
zV$cTL^#<E%OLO0M2w~~2%S0I?w(E7iD}I-}tM=acKiT)Ey-z+n*|xsiS9ow1h;DE&
zD>YH4coh;|P@tUm-)h@H-TK^U_QFdm+rlPvivR{=08KG59|C3O1*%<8Ap#-7#0Uex
zDur=AN6=KL>z!u}6UzIn{Q{t(KoM(*F~W+!=e-N7vSsp;hO&#OQ6!|b%^-VA@pBlY
z%a_-|${44-%g-^5Ndd#(cWKaG;f-~7j>E%493351Gcy>s>Dj)bXBqP`#^BAHH~9Yh
z?{Rc^h>t&d20xKd+D=<=%@p6Q#D_c2j1g|~)t;pFwM#%}B}D5=xa?8_CC+6&MML07
zny~4a1xQRmu@IA~N<LfJ*GMvB>0W?R)WSGIY$mltP6<K)`Fx>L`l+hhqQpQ`3Hj#$
zxsnD15;hSSQX3N}TOBn2DWx$VxI~alghDHphbs|JB7Td|QlLme=1Y__hWYc7f$oZN
zWDQT%LR+p={+n8ZP;BVn-~a~)2bDWW?zJ7>)e5ikiwnH`>MNX`o#FKO2ywf`?bUTA
zAgv6dDs5z1LYKr#**z*V5&peAuR?4!lWDcgvKPXilJ;5$B%N@HJrqz%!6rOpS~E&R
zN13i1$@15wp<S!=@R{T}A~DrjNW6%mf}76=I@6g{m;hwOkMYnZx3c#Wj0SSY-zGkh
zUni72c~3tV+mL6%I4YU6{CWG`7H=d@@z6C%XE~}Y)*)%4J>&YjHRN?BSJKJJ2^Nb*
zRtT$Rl1c;MD*fWNlDy)ZZ@$I*_wVrOFMffAHJH}hROqWPsAQRHl|utc(TE7Hv#K&E
z>8f*@eH-7EXDRt$%gb8fbSG8<q=I5#@{TMJ1wcg#!e;kRLTt{e(Mprat84cIYfk*9
zQc4unlovbip^yUZky1`g+T;`{7^u_y0<AR#2jg8|q>3Io`b2;Up_B%2w-tSxiYFQJ
zF2*bwVH@C<-<KL8e^#!s{e8>Ci54lyZ%pcB>RjsGRR~LJ60HWt2$U~D0l`Nsm(*WJ
z7EQ^0Y`0rjTLO3tAq05maC38mx9{HK$&<%eF6P*7*03RAXZlZp(ikV_DIGOZ#k$m(
z(ne_C(79bkbq+wK97u~uo@-`$30XN)%qVJ)5fSqkT*l{z24wT4l|>f5nz|r_kXwx?
zZRjl1M+t}0*EmBiP=*cnuA<)+SBF*_F(f?Ir4E1!$dLPvQSM}k2bAYTAb6PQp|ycZ
z@yVblhVz%`Owv_LYrG%1UL$aRmjaUa)ObBw*QFE)`OKLHLuvfULIyy^gm<kC0LEcJ
zOoCr)O|93VQ=r`Dx#UU#9GfEHUOW;EoQ>tZqjAR9*Le2Phd}UfTMr#lnsAJOrnXKY
zMucgiuG8FWj}=t0X72+P4CZ8w+Djnao_tpPC)1uz0*WU`#>nFnRDPdUlmx@g_U+6J
z7(k|0P%0B?O2~vBb@rvINan6eW;07Zq6rF$>li3oBo#=>3pQFo8I^smN&(6!j4+6i
zJ7rP@rPT%@7OBo=REP#T%^43VlZQ18sa;6~Z4tY6wUrK?vMHamWEj~vu-*z1X4DeV
zrQnbQYm`#hZq^tELg?)@Q3_sa%aH|x)>MfPh=QK<>_BtMtsOE_&So>rX8V}U#;l!k
zEe2;p<uxS*-ExYsQdn;`IDhjRRs+wUoDwtN+#qO;U`(nIPeqGblPgkD(t=O^a-C)(
z6u)!?lt!wN&#?@?mtCed6Li){0jS^r4mc(TYBUQa&K4sLel!*)+J}~eg;{bdxd4Ef
zp`t_&7(iQUYO0{9-Xo+SZ7hCpq#~}6>NcYDsR3pkTT@0sTed<1kQmPh$`}gXBe;;_
z7X|-|qD09qW*AE#o_WI#@hlY_>AsP|f$x-EG19`2mVl%)`56EeFKD%ANoPd!G^A=6
z25dGolVcbL=v04Xc=SM^VoY9d78D@>F~ANc2?AoQ%f%iRi&<VH#iR;O7h%RRDpdu|
zZ?CU#`SwSwRx6x7IfV~fL=OZ_3c4N0#UBEV!}Uthlt7(?k(;2jLG&^A;$jaG;H$G{
zfgu&$*1{_a9Vq?Xdmw7!^^ubwVzz2n`CSYa(>TB)xj40=%!wc_lr+1fHHHX-G=#?O
zp6Q#JYwC-`vzT}Jui3)6qge~)pM!lB!<J3f?gqbi^{**iWouVJYk};_6p<PcBk_8@
z27sImBz=hOd$^xJ`||1gx7cn8VT1R$Uav8ZGiYsadwUD5b)LUrtj+EyX1=r8h!`o;
zVi<<hB4GwjujV=ByVjo}mZHS-^K;zX++cNhi2dcf5`t}UA8WH#o7|SEFc7Muw!&rT
z1><ks96<3&^Sk7|WFe0-`z6L$F{X=)3pnSfw#8alV`&oG1A`dIN-|m_YL(BfX!4~x
z+fW8Q+$57tO1(}9DfS~6gS<0&euV)SLB#ZqSTZL8k?S7;>V&IPiamXc8I7bBX$iD+
z#&;#H@s{QoByWh3l16TCDZ`HGY#hg(_#9)*G&u|dLI_x|H*nJ?r+-h=gw1Ay`JB>G
zH=7N{afC4j+wB(1y*;Y();iPGU=5}%xr`Qz1@`y%^LT?8i)*~~W{8CW#?Z)#X~Naj
z6}H<6XHP%CY#b`SC|zrrS8AR6*hv={jd)hDNkL#bYb`!$?K6J@3P0qt<9pQ#d7h8l
zR|o-b-@e6ixy0Vy9*ngV3yKjI2AL@<Z2*+~d})FyMOFj#MXJ;4WDuP{D-l|16|GCB
z)jat*65(28kpf2|YP)H9NK!Hb#iC#YyQBHWl5HTL>AlCz%?<YU_VP5a^Ye2o77NVh
zbD;sz{@6!4O;dj7Y&?QeMF7m_b69IJO&)uDd-=kb%Vn<qChbZ1ZxQn>fpHwu9&1ZT
zA~gR!2l~=}-n@B(%gakVe*75oaexaxR}=%fnh+R(oLA1VE^BQ<PAcw;u}-CI-2-ec
za|R3lPG-pgrBPsv5hj&O?b;Wg-*&qt7hs5(w$no!NSIST6)RYcn?WXD_z)01#U3Sn
z@cql7L6S~nCZwdVmLVWn6VlBD6k28Lk(=+ecvNZW38c~olzA8YJKo>VW&3a(j`4fF
zUL(esX^I&zD^tv4_&o;)2iX&#l)_>@huK?zlTBDr(b{H(kBTG>_V@QwHMeipLC-!s
zJj`7iXrCd!t5rH>j3K1wZ8jUcdi4rQDLi^~O8v%C|FL9qA|yJCFyOlH7$RJB`FskG
z146E+Gk5NY&zTbdWC9Hf62^(_Er=0DMJiHlaKn3_`Q60sG#jy<94u7@keM32M^r^C
zxs!bakn{m6q^$@=FbrPgl*X7!B>cJ0%7sf8QTusD!uEN5C~`g02iR&kQc_E;S0ba-
zcovHV7Cdq?nZ?Y&SW{wcj4zLYnE)cWp!k_YB-AmO1xPOFF$VMbJnv<iCRk%|aIl{<
ztQe8GA*MtS$z<f7w%aYf`|djo!+^(+A5){h?G~k|u)YI+kMo{do1_`dCEr}Z+7TQy
ztb1G4F%5P{whT<(t9(B*yw)5CN&pxj*NjK)!+t2B0D+t2-&RdQZr0|ZFpRdsdx@L*
znSzN*7mhq1dA5?4+Ouy#S}?oll3k}Q4e!cg;@|TcC0@zQC5d-OM@Ilk8u%~_xVgE>
zaXe=BNarf+9=MFaxw00v>bF}`#ukf3l>rH%2y+VTY&OSgwU1#asjM8klzZmqOA8_d
zauJ@te~%wtzsB+LF_z0EuCH%kRc&!`pr3I-wq9X6V+)It1jZ~tcD0e`wE~4cq-&8s
z&;PFYr7mgNEsy8#_>gL0Vu0lip-iYi>hrP#aWxc-;&G%!E0?}cCVm+ZZJDr?Z2Wim
zeW{(aOiVIJmZnTVQlo1R3Mefx<Ymjl-c^IqK2QFgHG{pqy*w{mUZ8x3L>P8pl+Tmq
zKA$i0df5~Y0x4rm4G<6Zp;FMUr0*8@4xd+RJuyOEU0&kl%a>TKRyaOBE@{F>BQ+{c
z?WaAUTGL!1sM2jTCZ9Z@QtDyd7DAo}uY-Tr22dqINK3<La$(i+uoQ%D@M5$B1*9KX
z3}k8rYVs=sHEtM_`n;uJXzPlSXZ4qG47iKu{CxS{R)DlZrL}C@cwXx|lh%e*&ZHG2
z8It_0<Tr9HTI(!CTB~bUdlbd^8R~gAF-DBz4D61RF^dp6U7q3^j~+e3Vm{_Iu;nZ*
z!b)HX<%ie!!>d<#`}QrKJb8lUa*0jqN5-855Mv%?P}57Y617eV-nG}o`wal1{Wg}=
zTAO^{QhBIc%S>l6!kW~ceupB`;@2>Yu;U1eG_tiw9Z>d+SVXPQ)p^MB;9AYGMIx!m
zN(ATcw^Foi{UH&xeXc}0W}LFfh@lmDS=uXwEQguE#KE<i6R$;HV2emu@M>MXQ>lc2
z8F>sQKamrXv(!?>cys`_zc@WT!EBu4>iQakZHw=a*B~!j`bXKEe*5+<o<4n=&3?WI
z$z0^KB=XY!C^OF(3NBJ7TmXTJ6MgVRB#N{dt&0L7@1#Y104cM@i~A0kTtv|1i#A%r
z3>Ge=o>zhhl%kBjZSW7R6&5IofpH!%ln#|sGHb?N+*iAr58jOn)(RPaPP)Yyq|lL8
zgQOt_#))=30pqm9kJha$l|!i*7_H%%(TV-YXG`svaYK$zmduvl-4+>Ae=lOP+Y}R1
zP?*9d(PPXs6C(i6F028M9vvh4i0kWX(1-)6C=va(nmt=&%zG{`FR@;)@#&|ZW@T_T
zn^gkMdyg@lqZCf^9CT`$=o}Bq19+BCa1K#ZcMv+4!UK2Bc9-YO_gV}1klbk@wTai5
z+?4tm)*ZpO!Iv~%@M4up8GMw3rTOr@M^q{eBB=voErLoZ$d@V2>{f%4NVdxaYY~^9
z*&@8uY}m||T5OA?^1|eIrO8dKvi1$P+K!BwaGXQFg9GO>VCF+apb{AP5TPTvhC-;j
z_i{n!ILyd*Y7D$17gF7to)C+xM%(@I+TXo<hwJNWig(QCdA-u?mPjHsRf&jNQ(R`V
zreR+(P=T7&)PCP;gP;^#hzQUKj=Z|kBJ$p2yWQT&$Wf$x)mpcvJT4k-o9;43a#<z!
zT~<c4!p(XO;9-rljmcBNxD-O5Ssdis%9bYWD`uA+aM%4L?}l-z)iQy)pp%~?&sEZ3
zOCP(Q$xK8F1x}0QpGhGm*8q_4kod;WWKPa&ZBx*JQuoqYK>R(qcK$u}<f{eL6dV+{
z%XfsoMj5D}usS%z;o%{GNVP;!WjZ65QMglQBKd*V6jMDvKZg$iADo?4!c%%YSOD`K
zNH<beHY2dz76q5<n}(DLrL@j;h6sd|KAc_UKn@?sWEEPwR0fkUO~!(@ru5JXNl-;$
zvBe1G9j-TPSYvb!jsS__16&)M`q~)s{iT5;t;WiKmnO56s;x=YUOXd}L~bS?S!g6R
zk5*ffrgR7ax3{-gEEd_H*?u-Nbe+<aao6{rXKMh}K6yRTT_lkx6GpRsxy~sR(|rmQ
z6juhM#V0Lb7%dL>4zOCS@_NV702si%toU73QIS9tYb~y?ukrHbOBiEte0)qAL<-<@
z>_)DQiI~q^n)+F|c$#!Km=Stl!I08SQwA@<t{Ngf4(iGKUa!~CT8khjpP8o7s4+Dd
zMA4SF(*&aw4C=Nk#u(^a6uFOw1lE)+JRn)CvCMU#bWe5{a7~n|a87Fe(n^$C{4N}9
zt!8P(0JTEaYV%Tf@ca0?>-8Gz^%|?ys<MDs5L7Or5ULDI$p>1c5~v42Nl!()NBM7=
ziY(8%@?z+i1&E>zVzHRvaDSEC--Hk_Dhr?p3Vz04>YerDk3Zt|>(|)d-^amxj?HG1
zGa@-uAlI(7&OP0Bl@K6Os<u)zb#*X>&bXK_O`xTjKarZFO791s{mspd=<2|3D3i{(
zv}&RK6Th0A!;tu+RL)I`%^IcYge2q4h$y}k0qEp&lNzcl|Ci4I>Z}s^9f>ThCd9vQ
zH5z6xKwTN$n%+zZT5Ft~oMfbGt(=PBc>e+!r3Tghor#CdQs*4Tajb||S~SuGm&m~c
z##)t9fHsK;wZZaWiIbCK42CBbI&nVSakHMGA!7?H6fQ0<@DKm+4_GV~c>er((axO1
z?d@$f=xZ2;jPU$SnZ+Spaf8+{Hl+qeDk_TkdmJz(q@?bprGTr|BH1^*x3@$@fYM1j
zRNN_;%xR@apc^v+2pEK=B85H&Js%Qqx8RDIffO{$0zsuBzcgh|x*7QMZDFVs4$_56
z@4PE4q*Wp_-&k-+fg~Bcd>-Q&r?xULXycB5$*a(ME?S`=1sR`#)T||~^SfFFlYduL
z+MB`$hJh5SgZ=$n4^wqZ)`Zpk9*Jvw)=X#Le)}zc|NGzPp-=q&&1RF8H2&FaHp{I~
z_#QbcQSM7Cg>j&E`G_Sh9fMC<eB{aS*>mBleao|tG2-Ut1_uW#SZhE9e3Vh96m=U=
ziqBVJuo1?PqGlDfjS`myV+@9Igyo^{t)`)s%0+`pQH7%A(W$=fwZF?WmR5;G-1c+w
zPpSE|F05V1*lIKKJJM8^rm$SQ)NXbW7{@`Rb|<Y5X=%xQNJ41$+ae=delbRP=P?We
zj*gBpVQ#j!)bB{8{$c=Pq(Gk}bZG{2!0+|z*ZA_wFL8EuhNGjSI!`~{4`-t=@SU-#
z-rgy%e^Atq%sHAm7?c7`gbNPN2TVR7Dnq7!a8Rl)00WS}yWMUvo6Sj3>5w&`tRRE{
z#KcssrM4Uq@S&VtzBj}>KKq9N-nXqTRIbN}o{Gk31cHYd20*Dw(6&~3dmmCTNd*b0
z;|wj+mCxbtGe`j?zxQ+5SA4(iI<v57=!~E18}EbUCz3wf?8;VH%JoYxgN(azdwK2^
zH=obpCI_gLcFgV}8rzg(0+RNn5ZP=t`0A^#@aD}M{LSC|2J`u>@}6)iZ*{h+#1m@%
z0Th>#g3$(pCbxA6REwt#d4e*X1t3C`1rq91Y00bE(upy^c}n?@vBt^ZhZIjqaby_8
z=yUsc-akMt*~x9mg(c0ud}-+tsBs=rY^0ub&i=L5eFUKDuheErNaigOxkX<2oK~Pn
zU#UcP{$0gXnt!(p9kEzCeDKDYs@IZyhs?mJJ};Pgi*W7Q>2sv*ghY@B`%9dhoMISk
z9<W7z<C4fw&aY0<lz#%ici(-7ufF;Uzxc(cxdl7@I_BpBlGxMQJt#<*0^n5d!TpdD
zQwV^EOXrr73!va8hp70RDahl!M+}a%o?Kf1%A!gg4fd7`AO>tUYov@yD5YucbR_?-
zX7f6k^jw%I<2C>&!0zwH&;HtA>|HuD#w2>u`5XamIdV7MTx!>H-z~1Sv?Uc5{u%$t
zG$1WxX=z458?8`it)Mo(A>V0?$)i^R(tWv=QD_uPekLn0%n+qekk4;p2y})H)()9x
z5d#jeM&tPSsJMllhk{6V2if-2>ZQaj`TcV5&N=+?kAK8=yTz}5{j1z)QKvzA95bOc
z7FUr@HB~fykgN;msg;r9`%w`wP`p&ucF~LvhoDed1%8^KRDdzIyhm$jL!Baq!KP5f
z7QqABlsGT7Cn484@k69yacb5X)2YbVw;cv5+~s5E1!e{XHgKq2iPk0;+W@teAQ>b}
z)l3LLbqVcz_-Cxmw_3<9WRXIswSJ^FzRO=MnLhjZIpd6cibYG{{V=pH1Gxq%G^8Ne
zrQK;oK`S5V{lhTe^z;-52YXqt5RpZAmzXU-PXoq?W_rSm9;}6P4qt!$HI!2LPk#Gb
zXiZGsSdKH0rc)RM`9foi)Ix(9QSi{#?Opdbo9+J(8dTLDL8WNyy=SeUVhi=OL2FhB
zMQYcq*s*6RQN*adimF{|2dzzOYf-a9jQZqx{*3SW-alQ}Iq#qDbFTY+U+>p)xTWMZ
zwrT8YTQO}saq$Xp7pa4{(~II|niAhS8ecAjcGE&YbNg7~fFhnISDm?e>S`kqC+CYh
zWc@~ZXCvZZ8#d2=h^Rm;gy9<V?_U1V2@&3@%Zz4iUeM`zdsX_aD`p?acO#m%-<RU*
zBWtQJf3zfkDGbWu7|3t!wC1sBrkQN2K=O|W{znje?0AvD)=tOAzdh<ENy(&9K3uEG
zbpENPSF`FI2pAsII6o?iEzLRF><4yU{<c<Ax!dpN3jo`=U>V1(iL`PKp8?TwJ(&)m
zFRDT&AC2?5V?FV*Q{f3bc%w2*CTL0t3C}v_AXMbE_&pXr7I!Qe$mb#4Pq0=G$n|`x
z$K$I~IP^Lk&bXmnV<;QVU|(pc>ffpJoo09|aYnh?>~Dd*XGKKQ;2_Rb{%EEuYx9?f
znlropSOSh0Cq6qhpbc?oEbYU&d@YgdABkEDIgl8Gg?8P<x*At!N9NVdo3SyqSAOXV
z3v(e1&_1Z0Z6)P~Ye_lE4WzEoS1EHAAN~Dd+WM}wJ2aQGJfqidVPn-;MS8njsnvLp
zR(IYag3aLX%=>#l$6Y{(9-nN=ViZABw3T6hg$h>-w+sjf3Aw3C$8xU-^B9Sjf+W5o
zRDS4XDaWT9P&`9c34UZ1@vyEC3kTn##_OzQL%wA_wBmSDeq>6Mv`6PA=N*>^_eOO9
zj+Lw1EALMr*we8%eU$G0oq<?$C-TczdH-}mMV_D~+pEs3DmLcU<cxDlua9gpjR2<|
z+!c0zJJ@0SH#hMNv6Sn`W8!urh1b!(xWb~ide9lmNC&?IcFvuf-%p5NMHyoRjVG2n
z=)#uFW5Xxs-$-bf`ajWu&gxftZBTmw>fVAwI_BN~mGuzOsR1U(-@XO?Rz1-OdTa}V
z`LkQpR==%fn|!(1=NZ{~%(p?nKe;2>qAN1wgx5qen&tjLr?zA6iPoLYGqsfvEY-{8
zHHa=G{BC|P$TMLDZoPWYF0d@lPC+RR&_}hp-oVR&HA1nGkrcFjDdbVi<w_IOMSQEn
zAwhH6t4Z%JMsJ5*Ft%i~2mC$ym?OzH@6)o-jlFX}`C|^2I-d30=SHlIQi@EEkkY{O
zf)8rfLa(G~>kRq?qNR<QtG~u9Rln4FE4o@$`>$f>@RZuD+fJ-d$ZK1BXszNP^W8`T
zzn(-vEa>2!6~lU@`!=0dX=WSLkV$a!i$4%b{Mw%5RoT|kOTsx98*HX5&c9eKX5*Y~
zvUk&c1Z?d+_KKS~tatNq)1A2QMpMjJ?>8TQ+W41S@8_5GJ?5;Wd3rr#kA#I;US9x(
zu*I({IphiilYDe@lGz9vp_KS?)!G-m?jYDCc>|Xi{||VOSTcuI_&+p}&C2%u&yy=$
zxr&>kIqp5kI1@KPuD+wC3=}}5mqqNFjUw%_{dPi(x@*wgse4tjkD=7zMe6S2#wO{|
z6_pfQ{YAbiu?$y7FcTSQV>_^?oxvh^4j$<i_uBb4>nBYQU;PgrQK=V?3^DJs6&ww~
zZy=#Wi9o-GMae6*fwT(kA5gG;{R2x*9W3;hNs!)9>)NHLt_1(zeUjiWYXqns6s&C<
zx{Ue!fkux5i9@$(-8>Jc_HdZmORsn%|I?09yJC+j7WB%0`aDtkkg@d_*`lkpTptLa
z_4AeXm;^R2kBh{VbpTP4Amwg43O$rGr{U^w34Sv`&xHG@nUqk+OWG#AX})_Rw|S#u
zW@v6`+6vYujGw+Lxi98zTiS0BJ%aOOl`Dj#Hz^5M?`GMb5JHp2H@2<)_1h&l+TE*p
zIUHBFhZ>RJEkzuR%Bl|?q6R=a$-ll$7MG^l#cgKF3wzO2E>VRfeHnU~;NoKH)zkuD
zSy|2);l?(E-x%UErKvZmk+PE0dJPai>)CAbJ6Hui-T+)=q`jKi^&QGFj0e|R1O+@8
z^fJW%IrvgGkfP%fyAhT0-kiq1+J_(OK$~+d(NOf`M{1HjL`Y+sD-+T3eu<5D^Wao_
z@t}DjRp(7sA*5(HY^14mW@puZkfnbHp0-8g#;qiY$+ii=0Hi=ei#*8&-}@xQarGM#
znEu-=R-x#Jv+hAnv#7R-8(G)#Q_@|U^nsRsK*23|O)*tIHFhd8p7_m%{J4;-|M|)v
zb3eUU-wrC^1j-+*2w!qn@cM-5Pa6$5*m;n?SIJ5)->j;pW>h0eH~3M_nrAlY!-7-k
zTU8c8eO&lS)XTy8JRd<i`sx%0<10v<WEvep(F>(Dm=p8)w>62CswgX3D1@MeEA1hy
zch52lXxNOz|Bk4B)y<}DK(^pvrM!=J32nQ_^A&0Pl^Fx;_ei~D{N#CwE+oN=zlL3O
zy!-sRG+bA~w8e%bhoFwF2vKz~f*u7d08W}pPJ!{O9FE6W$n9IfVn>Wg*PGSGpW~hs
z-MRlETZQqq8jDg-Eao=uI=dgq-JbCZTEumGI7K64E^nqNF{Cfu5gYhfNwkw>dUw#k
z3q<obqMCQCa`qZjAVn|?lF?k^>8xx?n<x;YvloWjf~2Nhb&kO2oMl>G6zaWKkrIki
z<a7Gk?$7LiW;I3){{VmFFac}y{{0)O0>$OD3?h1${V1B~h6{^ijqVtPgj;*o-Ex`B
z8(Zz@dy<D>#xU8<&<H;=Ta#qmI9NJM2`QG=Q>tBsaz)aZWyzPP;mMHh9d}k=3#IE>
z-^hgfe^;M35RI~2SK9gj1O$R3Cta7|!z!z9oU7%rK!%}YLq{q(5=$C_N&2F-OU+8b
zBDN9mHHO?h;_Yx;Nn}>Uh`PCWvFA&94MUmbnc*#$k-6a-K$Fu26T-0gvLl%8)6Gxu
zESNNfK-(X=o1rD$W;Tm?EHuC({7Qw!0@{<lOi(aq><NaezDXgqKHYX0{aO|+O~26X
z$-TBwkKpd+yrJk(S5TUP&IfsM)^oSDW)ma%?rHv%xCB$H_eKG3pFU;LH#B^1NVs;w
zwoO*-I_gCmi|E-QGdH*Prwd{}-IJlj2x^)8r<MzkGm(98ny<rGW@^hhqsu2<3o{ib
zhR$LH6AZnxO0DY1Am#`QbGKjc^AvI|KFtIvaT!2!)Aly{h}K2ra2pb_CqkLAV09k+
z5-!U4fQC#saPiL&!Jy~l*;X)?1z98#62&0*>QC$!C~D;}C6{fgu>^PZ)X!-5lc0te
z>e)}?$;lt7Lds{HG%bpXL+%}8sD>H74{-aZRj-v<3Ko*CMd^eOH8TZ(qhPMd8AA?L
zdNtPAL4URvF$$FFl0c_{I)H2SP`J0#MT(WFeSiR73NB*L>xC0@G9g07Dx5Hrs&_Kz
z3R&b6cy;#f)*1NRJ}7N4G}JE0_PIY@lhyt(soINdM+xeu2=SMhr5++|%AdW{H-8f3
z+!AlryU^nf*MzCLd%t6~y4bU2<2dM`&d(7LNsBf$nsNk*k%(sKuokmviO=UAL<BK#
z{*Rn-t5R3DLua=;XeGa|i3EOm>{Rzh)&sJ4QcvG!zz^eU8+uFc!&ZOu=6jdCl~<7`
z$RT)l95RiV1DP{s=2X_w8TD;zg`Z;`N_`|=xh?J!o+l%Rc_cqj7h77PJ5COVL_Qg$
z19SkzEX(VFYH0jjk(?ffTU6t{A_6d1VRIMhX7)vy>v6R3K4h~1@x-b)dvFBjVLWSP
z;Tkn7aczTOh^uTWN9bg}Fmp>tE958%G@8*-@d2y{NB2cW-r7Zq7tXky2i{B`wLIW{
zM4~KugS^lIn6akU=6wi9ecOo56JjvLTf`ALpijG@Vn7>r#tHnO-*|mcWVx49L}D$q
zQo0lt{dWIhN%MeQDu1_2|5h+q3ndo&hc%n?<8Z5POO`*d!K`(eHj<v@XH2pjDUV1q
z^>?{Z_mc(nRuDCnI?*IlBCB=xV9I!L3RBWzYgCUuDYRM(M`)!_&<GG85!<cmi=Ru>
ztXs88H|@8~$gFfV0yd=UiE91lbaYlqb6@-1724N@PnE$vU7oI{sKm-j(O{B~l+s=G
zXnL@OUaK6JQ%)ktrO#x<?1|#dco`-}L!0};!`ieFK`2)sC+WwJ$eM;v!Hndn+&z5Q
zkd16#4&P}=`^Ena`@M~YneBDKr$*@DFM+nr8@NcNuMFgy08R6(Y%+Z-fWXswUkli}
zGiH4x{eGQC>MOP2CP^Lc#+1<qG`EPdFrnkSQ*ztB`Z_2_-u43iR2ZM<IUSM1HKclW
z_YLa{n-~ZWKK}k&67Gp3$U?Wn*=pk~toikC#A+-u?Tuthx}sO+ImS}yU2F7Xy{wao
zc8$O3G9J+381J+((drbC;;aLxDp>~R^`D#Ww*}MZR%Mj?`hwku(C;HT_*;0dtM@kI
z5EQ(vssa63q_J`CBqc&d5M@vKMz5vL@+DWr{o%1+^3n4ERcruKnn7Jd-4pgXt!=1%
zv7Z#}By{gu%W&>Bj@tBqX&QQ>It+)-PkZYGXu<#E8hA*+Pi*)~cQ$zh<{V@+qjXL$
zI)4S_zZrsgxhv%D5SRX5j@wdmq6m(S^D85gUH_twP4)b$B5hBDAu8!=(S#R{QttUM
zyLbN>^ziUqqhh_IheErB`swz^Mi7)kx{~f>FL0_WqM_y;asD=e;GSl&V8OI;hgj<G
zqsJWsIlFLd^Tb5mFu4ey4!xp>_p|4Tq~3!>*)v{a?jaCYMw|^X8_=)61fdC<Qw=GD
z3!ydB<xWY&$3g718s~?cHw@g6X~?jro{jTW@nah&uH%^IJrYaGHTUpzh#5dLJ3C9{
zsL~>7Q^j<gc2lgNY4diqkGlri{SOX_+qg2~w(RT8Pv;fm>rcImWCR?MoefFS>{YNH
zWJYvBVxhv!;4ck#{6G%67i&^ajcae13SrgjpO-}77ZS17l+sFGcT7zLDK!&0B1$Ua
z+)!**h;%EZKqdNek|Uj+s)yIJ_JNpW6FBEVAKD)FigIU7(KU7FmKT4@3{_aC^XI9v
zZ2}C4IdWr%X8+7)6}@d2qcWXspKT?Xpk!Scn!P4Oh&ZZYeehmAkWP7oe9))4r6tC-
zy+^?HNM<Ihj7|V$I+Hu3@pAmtvwsy@(e~XWZz|FhF;`^(BuvROGbs|8*-)wJBcKLG
ze8SQfe|$3_I1ru_+;33f9@xjSBoxonwm2sa3r6%H^Om0QHlN3Zi^S1A)SBh=D-7`{
z^GNdGggU1AHk@|+$$9mWg)f88-sjn~8$wb2=(sVz8(i9Q&|GiB2#(>ph%)v}%V9HH
za@1`PdG~~z%)pz#4WG8PvJC8dCi3rxgE9JVDn^wi5ZiAEIdYz?@=XeXThC@jGhW2)
zDOSmvJg`W8D%V!vw^W);2566(dKa+FNV>2*6Z^wvsl=a>$>BZTZTB^-SlYmle;}3w
zGzTB`v!9Zb8>EB@8gNkGs4XfHL~bNRX8Ao76(?U3NG^~i-%pK9%uwb_2e;<OGoJJG
zkQ)O-Q_*(;_ib#kb5oq=27L&q@W_N&{7VK~w2smA+is~Or*T0(;~(AViN~UR5`2#g
zuDD&EAwB9kJM%qS7TX!WbctXhmiv|_BRF)e#RiQM!vt8b!(6^l(L(QLBy>VMk9ww-
z8SXNEf(f`q-L@f)W(lv$3j-{}WND%MCR(R8_CoDskrQYB54WpI4vkK%%8?#C3EGa-
zatBfEOi~?`4qrYs^^>4Y^?);$z7#!?kL^dp#}`HRhRCR)1a`^)MsfT8UB$mOr2FM9
zE*Udk9D@ZYk&Ik!-?3F>oLYtEQ)hR0mF1F)6B&8&B-VeG2mT?{FgP*tIO)YVO9K5t
zLia49PxW!B3?~H&6eicit5PaU9Y;pKfW!tBEG$^iMn5u@o*Z?lhSkf)bcDp0oiWtU
zn?!afcU@?N-rBnYkvsRbY9bm;=9S>@C|Rm=Ld&EV7|FW*IwTfiN&)N-xYn+KH{DWi
z0d8K~28^*AX>9o-^ScUe>#DcCpuQ-(5dmHXaM&fXk{|CnJB{eGGLRd0u3v6yG)GV8
z2Im&C(W1ljJGDcn9Mspg&hG9hK}r2=IwLq+C9L^@LU1qVOx9%-1IcMnV}-<A>-$t@
zdbBfLDHhOC_Gg@}@#o*T%Iw%wE=_2e!Ue3(?dG<CN*IWzt(P1y)}zP1!YD;D|CIf(
zAuG7eAQz@hA+Z{5R2>!lR=7{{K<39oDsLbeNT1pjdF`G*su|c6Im7!<K+}tJLXIv+
zFPj`;{XQ%#;@joe^z`(mE%tP`XY)MaN*Ht6rLw)kv)~(BU#VGL9`*CUyQwZW8p%3B
zJBUdu{L8hTIf9#2cx>6bx$LuLqw*{?7vGMae+>^83i3BL{HxhKo_5)%?$_ptOyu!4
zn0V+##@1A}iB3|~=#1ddyjyFNe|K1NdzP(J7bWEKX#CPN?2LJwK=2L>j3b=|%n(I1
z4O2z-X8;Dktu|>&9ZuECo~&0?kJ;tDOxvHN;Ds#dZTXH~9h*mm3%7n&)-L-qaT-&2
zW4WXj?h-vJ>sxC4h<*}upxglh(273i?2*<bpP<v!myj}Sl}Z9x<U!Q$U7D<KZEYFy
z^ZUloajW=zJG^2AKk5zsZ*zJD^dXyi4JB>%fwtj3-zG@w=UVJM{}xWHKAHgP|5w+_
g{{LnFmtV-0qz$!g7qc(113-FUx+Xd`+IBDg2h0PydjJ3c

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_scripted_tensor_transforms_001.png b/0.16/_images/sphx_glr_plot_scripted_tensor_transforms_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..8d14925bfe50c8ba0f5e7f5d1734f3ce4c4f10b3
GIT binary patch
literal 171473
zcmYg%Wl&r}vvvp&+}#5Ng1dWg7GHdU1h>WA2`*V65InF6?#|+_39bvlU9y3NV8K7$
zTVLJxR`rn|GkvD!RQEjH{dByJmI@v=HTH`aFYwe<6@f2ay!3f~&&PcApZkcK;`2k?
zTglj4&&|%;&(hQOg{GyqyR)0Ov!m5(Ut3QvM>khN9#I}auGbFU-tJx!yu2>|JAlW{
z)1G%yPT}xb36{I6iPwu41eX7~P|9RW9bcflc%i2FPTxQG&x2ZW^-ab;)&smJ=<8b6
zucOX3O+_1$bQ^;d6md4E;nvR<Xw3S=xRgx%JE*6Oo2aiT1<{mWrxO;E68~Zo{M33q
ze2JJ4AIa&uUb#1Yg;?{wFDVdM9mtbMYkt#na=paAva<f*1`mXvOR0mviJAWK|CN7F
zM|@8kp{TrP#Q#U;c38@T{_iMBLSD5s_0GZnm5p$JwcqczS^ifZ@dwp`*Z)_E-|P3O
z?f?Cfe|!(>S?YY(e|4ze)5T>Lo%~nF`)#CLE5qMRr|RI#1tVH~48`52Iph(`BfcUy
zaqeg^^FwGnzif@Kzlj^~ne|R-m8ido7w_55&K@Kv{z+mdwx)$4Z9e3A?)`70p4(sR
ztLpbp+V>}ZJv*5__j}2=(9Fv#;;@x}EB}9HOLC^*3Ckk_(&O0m@V)57UF^l_#uLMX
z{r!XNlO8bosNr-X?8Zpyehc+!ChRZ6UqslyQZD|pH9v-bCo9SKkJlUgdG^8F!2v-E
zVYHt^MjE*E(ZJG78Kq-XJDUq0+U^-1hrN{YhviC~Io$|0pCKThRr$i{)+02H7`c8Q
z@o*+LDhPkkoH{yAP2p5O{ZiNnU{XikiMMO^Lwpuw7b4-c^77#%Mk;9>Mx3cy9bJQP
zEcbjo?cEFR$ES}NiRrBkJp+f{RQ8Ne^xOw|n~yARJeqdjU5U<%Sl+#wnCU<r*zDz+
z&rt=lX2y9X&2n5bkayqN+2si;NAaL3aut(zdeP#mREf^depCi<eh7P@%be?9v)l}Z
zwwkfVglM;%&$D(0U5np&s4&i5%O7%7W8*Qx-=35PO%g;4VVQ4iLH3TCW|6qxp0+a{
z#HM5M-$F?_#f{AB;%tK5DLJbgSr!L#Sv?)IoG?Rr#Uo)ct;(TK&wJ<gX!Nvl06Vwd
zzLkDzd3t!t$zQyDtxSkE*Q;8X1_1k1dZP2ogYo!dmgNTJD__5IVuknj+*G>#b2ll+
zd>VQ>dFqX=?N%DxF0#az9RDhLJCz*#IA+wO?@kWkPvc6<`L}({8G3V-v@RsEa_VC%
z9#Z6&lp{DCaJv&uO41jUi|oa;qMfCN**VqV-bpJ-SxQ-SpqOcsMZVrO>+`SjdHpK?
z9k&kCus$G{eN6rn&ThIHHF(^D>uw;$-Cq+^+QJd@`&2$m9-OGW5X%+wijtX<zX^VE
zIT&F~h(gUYT@jib_qt%}W+RVhskxT<l_H__=U4L@)#)IC>E`PfuctQHF^ki*!<MI~
zk|JJIFC0rZ9ZNf;gDM}}_+!ST&cEFgJlTg;US2&2U(5buc)Bi|)TyUUu06a%-3Sey
zYGLTMeQyN(*#xYA8NzUQ+gr1ut8UAkBCG@e?5?V|$-Dgd(6hCFF1uB311%98x0jD^
z&8=&mF>RGxJ>5UfoX^RcddoA{-(dH7*n?J_*46)d-z#|O%sXr<mf%CUKK_mzg%rJl
zJ?^?m)dEakFJHfGcy^G#cywouL|a>13OXsJh2zDQ2IL#w%<U5&IdR`?oLX<7Q7$6i
zZUBPw(=`$Wxs|Fg*QOhX$J<#SI&)+)vjoT5bvTT{lrhkU(7}~g`q=67-Y#^*bUUs#
zxO9@7Xha(5NLmrxrdMf-<=Htn#aK~Eso3Y<`9<l=9NKf)1Pn3)yxFdz#O>i`*a(61
zf9_K|bLgD?$NBcMzWy6qc`OIxH=e8+{cCIB_3nq<8YDmb_2ifOI08rqP5X4@Ldcv?
zI5o-B*wf9^_S-y<r!P-OVM*6e<w&@9bxVTkkd6lsYu4SSh~;8|hD^ax*)Pt+LbAVZ
z<TA+l(0vkd;*<#b3jMb2fak=$n`WGu?$}<(gM$#z^UT+m)=R9XEDX|D3{wW*(hpJQ
zHi})dhm^s7r_|>W!2@#7+dm`@QK}DEK$I%Fc@?%M*3b9MwGU{uIMziJf0tqy2#g)2
z9C_PyYXqP3ojrQIGu!kR^IS!@GBW`V#1&SNXqS>$agVIY@nkznc+du2btHMLdP&jx
zxAv4GBr+WLg2m5o>@j3|sDAQ$=GcK;Kkn`9_20Ff-j((<Xr=gQ&9hd)>NDiI6OCyo
zM=A%PAshAk!=%j0xa6%|6xgr8+{L6-FfqjE<MqGm*1Lf%Sn2?asD~OA&o|!E@Bpn+
zl4qmad%73>A1`aZwZ(Tu>4v5J@6=<@JI5nr<m@El;juUH>BnB@bikAt+WrVmUtm%!
z>MxoK{<upqBHxa1*1ApO9e`{040=`LID5TJo(JW6XXr%bqFWhv<G8&;OwEdAUDezm
zqC;}{0h)F1Yi2VmzO6Ck`jk`OVcDK5_TePopa@HKI7T^F@G=+c;UmciDYJHFbS}<A
z&45o5Q1F)s$j!Yle+2`-Lvp<<WJC<#*i`ir%5?qb%ED#D|McT$tD&r@Yr1M5?{2F+
zTUyj&8%VByYuMD02c!R3Lrv4G^`P6D0f|Ttl(T}bKd~nzAi~~^pJNAexQ7RhE9uNB
z_(ZHb6c^DWM8&}F&4p=E^XV%my=M-sxdM*m{?_&Qw+a@1lGD*2$xvb2i38*)dgXZV
z-0zq$5k5X#X(lb%FYA^N3+|}C#-Q9shvWOK*3hY%#y=RE5NbKFanv&Q1XgwSfuh3T
zv=eG0OMul2X-Ji~qD7e0-?fJiL`fKrKf|sz_9}bsIUi5IZg@Ta@y?q$*=IAC{?O^U
z3XhH*$vzjTFB!kP7VP=g*faI*;Z{9H!up(Q4!ipuPOZnJ9^hTo0D&ZEY@U=qA9cq8
zW|i1SnV$hM{v=9{j3ShZR&FFmNcMg-&T)=)de3OT9XKo=>5<45ze?PEUX(`Fk@tSw
zXQ)}Q4_oGsN2s=oyvs+2{x|&@Ss{Qy8}+&zA+*T2LhL5P&)nf1GWLmAxl7V-0~?C%
z7kijrG0SsDg9Kb8FKRQ#uk^kj1+_l@f|XA<zj^o6A@w%$)<=^Ik@3gmyLr#w7H3=Y
zo+&Ym$<2XY*DxEXAecu8!B=CYG?gqPcKhxcM^1S9Cj88OOlnGMU3L~!K%<;gyPoP$
z)!yb&t->US!Y;-@=_~oh-<ho4vPg)nDlO+I7TV$_v$Ov%Ax#jDy)JHcKy@7yL8ejZ
zPZU8G;RAyFQ?8C^{Ya^ji4Dhu%l*t2HlQ~6)+I!ZNAvPq$fJ1e$%z9jb&={oTA#5+
zo+J_ny7HB#^IA$q1_)qUl5@{6#1Xd*v@b8&lAH>a_4I1A{;XVSLPCK{RhR$PTe^?h
zw{JGSPvl+I&}}%Yd}2xlbG03itBRe8Oaj?QBj2)fbxeV~vv~hl8>#}JYB|T<fv!yp
zifyeO(<5<}spBjlg#>gk1E;Zh$9{>i$GT^$2ULR|<BD^R{~iZb!ESSLNqNJXd>=&+
z)rg>WS5pO;j(T_o5w$W$MntJ4ilfU)c>xlqaOnGa9K+e#M+8QfbsnqxVn_sLM9{kA
zWWadFu8FmWX@APOaQXao`H<;Q*46!!k`Pfn{BZ`!6v65;X_+@-T-pJo)r>Nv>663{
zi52YMjDAlHWVW5a?SR%ddUq8fB?0AWtqmY|VP^4q1_`SeOFsondN=Iub7NGvSca6B
z-;yqo=d9Px!ZDpfU1zuSM}3e<TgR_2Ui<%UV|EVxuLGwxqmnam4je5_@|Cw91<vdH
zN4$2=(%6e<2E9NeN)Yn5@VRT$8UM+lx<NFaO~^YZ+f@dC3C<V5X0#3_N^4B>k1^1h
zVT-QP^Kc-J6>S!vBC(YfD*Mo5`9C<1n#^B9+goAe$b0K(C)-F=5>y`S@4xxgO-8`n
z@Ow%IZh8H%VgeOH3K$KdhLIa)66sN*;^q%5kG4-%A>=FY65J0}Aei_VaArC&B&PZ1
zy+yfZ9h?bnmCE=piABjNryTwL#zW59$q&E3WOD*vwiC(;55}dY25;o^NN`VuKKS>r
z4hGQ5zZWFIaRl)J10I}qPcFkfWQjj9mt@=CW+1yC@fSsZkm$cVZ7RxK6{z)JBysKy
zQb_@DM~fLv>TMs%q?SMgUxAqSq9|1eElmvhVkQHfWjDFz%kIQW6sxFY1Mwg5aVg!X
zMv{bC;og0|=={=?zMxB!NOt^&K9MD)WX9(@D@r#mOLuS7i$2Q=#-W#_Rl={y9Mg$$
zs;MEu2~D6nLlQL3L-h66{+6pN42?~X%Rf9AO%;k~(2Cww1oE3{=OO-z47;c$z#DSj
zk@f-y>x2$t=<F_n|NXwULDWU=jP?BUdMtdpkYy1XG5_$Z<8AQqcj4~)ZL)0)8L5bs
zH39F<48f4PL0emMJFbZ7x<1C?nX)+8eAUmCYdUdz05}eTnU|kY5N9=6<Wpjq@<It0
zjt&VIX6f|W^~oQ?;ntFpeqSd9N|>guf7OY|I!Grql0P`R`B<Jv$~&Kd`S1b~w`8xc
zC+c;VW&<m6FrTovAk2J4wpM8*gu9g5b{pI(>m`Y%^aM?5x;-Kkcx6xa&`|*1kj;ef
zH#L*-eE)e9vuXY%ahbX)p>NuusomVrMg4GeYCxTsD<Z#ncw|ocz;Vw}#w9+vFF}b`
z?BK9-B;gP`7T;m??d<mnPKNDpS`Sm^=!yRrh~^o;lsCJCQ0w4kbR&)wRLqzUMhJ|F
zY$})X8ZSae*1mqQ>PWHeuG4zEeW6J4GWP3YXS+CXZd@XduYu*45CqdInk!*2fvA&2
z!j0<L#r(<2D8kaLOe;~lvg{c#841jGV0l+92L7>#woG}o6L+4JN({5<6#gTjjij3g
z2=#L%E^)v?&ZlT@;%Kps(YYk-6hIP4BV}(>g?J~KCR26s@A|ULyai#P(sxoP9(`)~
zD+mrHf>J(NFlKD8|7a_I?JVYDiN8BEF*D(_xHbh`n0ZAoW!C$N$@FExH}&f1+3(n-
zL~x&>Xdm@$&T0{WmTW+mIS~@<VE|s(op41KQ&p6_2rNmzg2XZUBCMRT(pIqFHRdl>
z&rql3z+R^w=Q=wF=@W_DOQSu|BRPvnohH(NGGzKIp*`IP_BQ6bRP`i8b>xVba16pv
z`6*v@F7mAumy%KR1WVokILv^1b#rKS@Cb{a?5SS)!(?E*Nn<qkmmQ4LpyCqDni(Gy
z{+L{jbt)jJY(dA<k`g0IdC_MD`91#N#=9S+o*19QJh_(Y#d<KDx`~KQw-pW!O6W|E
zccZt5FF9Dn`Jf=Nc6!g;<si|ZZ1@kBT4O|+@E6mePnJJZA}p%Vql`JWm5@<pn3E*U
z8+~qUh+y$A>iyJ=qS81+w+u}sq58msh(ZzRtO!5sl$K%J9Ak&vDH3s(t`J8`)I1N>
zIVV2uc*mq&v-0Y#+&<I%>lni*V*wTeGk5<*UDF)_ZnQ8f+8{nMv+aiWV%h{nW>hgW
zzOl>{b8O|Ww|^?NIBX#0zAr?GW-!TDX6FhbI)Qq9CecfJ<=*eUjEptuSS#eXi))W(
z;=ph_1Id<>YmL7!mOjHZE8b{63CQ&>EEUkV?eLNn!*>VIT2!Z4=XcfDd~8SV@dGl)
zoOx>ID9hdbHHtx>{l|x(*?@i@RLc&VVgm@7l@=G+M#^^;H>gSwH$-L8&lsGY8BNL@
za#%MEWKT<N+Pf<|UY}ZDa}TPlh9Ac=3n|%G4N0hLp<9#6T-S^&MF@SWmj^F5r(jP~
zI>o@7j_uHgL;vxsBWe?mg8THN?>Q3<edwqjm;vBr8KJA?OTGCxymhyPcS^Vj_W|mw
z&YQo;IfnP={((i3f6n@L4oJ#H@JCSB0`1t{gD-QYiPgAv*68|7i>jk6li3BkaON$e
zNQK}1MX^RTj?Mar^U+6Wqff&6ZxvR6OO$Joe)zKSBGr+KMAW|v)fU7Lf!9{XnSwQ3
zW+y!ZsVTH-;wZ}^e-DdWH_$zi0sa#8p_gQr$Kb2l4U0s^S}IAQK_R}Fm_lLoOZ1gq
z^0iOL%l#!TY}qv!pqWoFgF$0J^>nj+Crb=mLi5Eh9vtgtk*Z&C>@-7^KC#H2hybGt
zWDxhl0sQ3Z_M1*8xMT59snHyCyU&&XC?y|yAWb=z#`J8Bwk9Tvsyr5h9Doxk1<YlU
znrJf|Fi$QoN$m$5(OAjA9lptP_t5QufgpNyme}vAY>9aUon1i#9xc{woo{O?vT!C%
zaC2TDM^aW6d){0yv)2sz<QE%UdkcEk6ncfvViUhCG0>!_J@)Q337OKwO-ta9XzZUx
zPr5?^lTV%}#Pt#6u_|3NsFiG0+qP`%w1}7~9kxR!CE21IYQ!uKU*5f&$pi@qjpatC
zUf>E3TMasSYDd}`XQE-UD8F&AdJkj~#T>RCP)6@{M(}bv8F32Rd-BBLw5}1YB=4q`
zX>K$&4SNO_EI@X8rvnlkn{rgy5rMVo!pcrsX}S5ro@UvZ==?E28g`0`cf2lJ0b#1(
z*Oyf0kL9|efeP#Z13n|_$eGf0&>Y%#F_3yYZw~ZVJH_!8y>PyuGU-EkPVf*Z%w5oc
z9WVussy-#{oFjMVPE0cbCd5_kIzu<Ug>IC}*SJ<MgZr=Vg=$Fgt=V9yqh26vHlGrl
z3=13{Id;4Qy37F9;tGo?*Io>j@@U)c&%pkmk+as)JZ3$84!G-|vq^a;%BTi?38d=8
zbTTu44J|=XXUpqE4_Zg1PxF+_9`BcunYFb^xF@H%6e!CAwg{5#LxY}vSXWh<u(hS4
zR9Rx84oq)#7G{h#@azUG$V=NpqWi>o*!7MLt3;K=?z))|<Q)y3RzZMh4_Y?N14lck
z-SB$-MzUFkq8SrH=;{0AMPaH0#4?Ax6UW#C5(|aB*%#jCFR~p1sk?E1@#1=E?H<RU
zYh_h)nT@m8HSGfxV)0y1_>kKt5!onjA8o;%O5@RPp@H%tOPz6u$~j;1!WhX8;|PRJ
zreRa=*`zhNLs>J96o1CkGPAH&@n4v~hHG%8xFNrt-i6z6^M5E4im54V$<X7;g&I2a
zAAFEIaQ&6YOrXsW@b6#vB#p1JGh133Td_wVR~%)`8t3q%FkEib-V?wz7ttb((2oPZ
zY6AI9)_1h+T&~doHM8LUfx$&pL}y7H(X{r?)ne4J@Oo(X`KVhR6ogkTk&9;>={lX&
zt6xP%`wEG=j!Wx{(SKNHSf)h4R(^O8Ngu5K-qA!fcu`zQr+J_%ZB$4JVjN>}nVMda
z`^B!V<VpA0>wKC!8oxdwTQ)@heU0CK{F#V3K`Xd6M<1<BHyA@?8Zr`{^l+A8*<x>Q
zpW$OSfwzIa&q54}Y<;a6M>*==i(I!XwmL^I&tQ!pa@WsUNVLl<jsL4Wh!Nq_-e#_n
zMCAOPx-wTSO|*@N;s=gr+ol^Q;wa^E01$Sa(d_Y3Fgt^SCzE7gT35YKLj5qJpnFV%
zJNjYk$BI!)@cBg0$wtozNJUbvMN)1~eJnx8bjV+uSqD6)uFfGZUsSlz%%ZXUNJ`Uu
zP+`)zmg+n))Kt9M+}tW?w=?Bn^>z`h9(%%!M6A&RT{ffyq;nY74E{iQlXl@)Xvja~
z>oSp%7+Dn74gZEwPSY-3uB@7v3Zu)=ds7mzxpb_I-(iXy9gx@5TAmJ!zKc~Ud&-f0
zxcuhjb(b?H<<(}^YR1!q$F3?a54hi+c+S4v@T##0e%sJfFZtf7P6#=E&RxOYYzH&d
z3T6Q8>DGfjS7ff$%o+ZvizJ4ff=45{VA~NVcuF6<H)c7Mzx&1f=9>9<ERj_Efh04!
z!K};>^A5Gpj|NY!_c=tXCU@DT8sNL~2VYu|G0qU0vcdRyIp0SCImAWxGgLLdG?0Ed
z*#J2TGi!aZ(BtF`y!!eb(JxjCZX|W8+y$0X_MgNtTu~zG@O<xRuI^B^GviY!Jf*zE
zjWU#`1~xBN=Up2=gG<Qv1ZkM75dB9NQ6BXCF~YV=_%3S>s5EpRV;Yn8Lu2IS&v`uo
zX{kg_%y!h2V1CV?=+$O7bu><fHVSTOzOy+W{i&62N@4G2?5ow>(b*|Y%S?3*8-0_~
zsI5e!T3w^Ru^t%dKT>ZPnLl>Wn)1{CG7f8l;Z2YNDpu*Q|8(>&iA$@aU)|A_u)<qY
zfd%JDYdm!;=n$c0_|Na;Nh2w#L6aZCY)phOekEpjxL2tUib#Jd;MYg68;E^Kd#PVh
z$AwNMwGr&6tr^|Q<6n9~N5OS2AiYs}P)Z`Mk|4*0{a(XM5i;P1x-He`mu9AVfHR9d
zb~NajEi$UsZ>->KQKeADOo7Sxd<w#grgWp`y!#}zG3i`P%8O?K!$Et)5$-Y{j5E?V
zLyg<73RBy~Sgd%s>IFWK9|qm>IUs|e{6fP#eCFgOW22pR5u{pkqksAc-f#WM;;*r!
z`*G-4&G$`ERAe)vFHjvRb!VwKs8gf^iYZvoV5dUhl~AwuJaHr7<YFfnz7HQ^^pSPr
zZ}%Oo@{DGwO%Bpi1-lSGv1CG}&QJaO^asus!uhWLTm`whxw451YLd)34vSh|&{Z32
zWv1s#cxvpLxxaFV$vr<WNOsI;%#!gmq0n&D7$ceUW*w1tPr8fE_W2n_c)G3!*gHEy
znG+vg3x8;3=EV_L%231XDirQZV5E$th$T~~)bV087-)Aefcs>uqq*D_<M2>QfcfK^
znQv3p@=H~9L%o`tn%0VC=7LKK-W+rq!i9mr^?#xGMX8=PCIb2jNmTrJJFnFXz|fr0
zWD`)BOOdgObFdb@;y64|XSXE0*)}mmIS!ig?HCbfaWChEVjWSSIJ~JNqunEEW%fON
zJeA?-REJ{LyWocT%D;%l)x$=^SLY=w#28!_P&^t%>5|vLl1hnV=+NfMP5bGZxh?#Q
zGRO!NKi~OzGI{V$g$?I?Z~S&GX!u5vQrWZmBBVE4`MkfK%iDZPNE@<)&#6;z(s90{
z5;MSEyWV9P*zO+S|0XJ2iJFx{euL9UrSm=hu%)@SV0erL6_l4UfYrrKO2qmbI@^0(
z`4CrRobyle6_)edxcJAe3`b=?L*CI*vh0;qSV^sY&9}wNmqY`N779(Hi}=proInrT
zSz*4d(pphDbdc@}{h}{_v5vY@+3ITd-qjV}+{0o~+Pm*&GrJ&vDd|~Nf8<YsYmLk=
zOsL0mNJH@6<ylfbL;L|of;?-EzsuB0&~s!K*v5B2rbzT)7cJx8uBE$Nb4}SR(WhT&
z6@v#8C^(32p5ZLF0VeEbk)4sBl3yuIxl1|ywTa&6x|69xN)>Lved@CI*~Im%2mD8-
z19ElI)SKbNz0U!aot4Hs_IW^wAoLAhITb+OsQ@0qP6np?Mc&{x&!h9^i{_y49wEU)
z93l|HKR%}cBVR-k)ygM=^wIeLB++h!{M?ZOgR_POCeoS-<v-?)n0yi>=P<Ue>tluk
ze`jY29vo@(^b7^6!wgNczsEUi8P5#DDI>3CrOP{56&H(i<z_^<lyjN>u+@?xn!dC9
zDB$7%xtT^od^EpYEjLHi7y>5K($x6B0E**vK~b2q%^?nS+qnfN!x8?9*em36ipUmZ
zi$%{965IgRTuG2sU{7RGnmm`+OxfIePPBGkr_rKx8FjKr7=LA}$RwUiD`O{KBm?a8
zC8^d+nbE7V3q*atz_2$iiHAq!BTO<d$pJNKOoPHR1gK*<WdSUEiRV-#K%8|@O#TSy
zFi;zbs+~*`e0b=aQCb_8%pXH)A3^Z?)q(VbtG1)&m(%4|DVaJg`{D{AvcFV_rNF~L
z;^u#=B7W~otChdw(7`2gz2>p`l>ushVv<;tCo=_$`_A@bQnv{}da@kl9X+y*Rp>N!
z&MU3da8DD%PGMa0Ui7Jfqw!AG+i4G}KgPf}pXwN-=s%SQ`Q|fwqjU&XiF>wH#gXRv
z7@vtz@STkI&7A8HoGK^a%mvuGC#939gIDR1hwVwzK2=~#Xqo~=X*#XZUut%K&X;;5
znRA14JCNGxfU%nE&WC%v7Poa>>R|t;lw8hrarf*^>GOuuj*X|JP|Vfip!H6Ph>zDO
z{GS1=2XQ#~2i(Io8e>7ZqVqX*&BRya%;Byr10TpG4w{Nt(gUEF+R%C$LL136;gm!~
zdem=Nu`XvS_Q41lKOclyhBBB?zHa%12WPz^RhU9^v&uMEILpO4v*c?>Nk297kg}PZ
zhs@@X?o47c7l}~yc%nhHsY`Z@CA$Fyuws{WdHsE;#KvGW{RTl;)Npyj?@>IG>e-g4
zcaWK$k&G9Z#s_T8*3ER?j(7Y^x~5-P+$hTPchhH;>8<Q2XiIGWoF>#sy1smgV_8J5
z;;N=3Yp3K~ie*!+BWR_c^4T+3<`i;Xv6RQtl%bZ1kug_DKMzZt>z|8t`^j$QY~3EJ
ziTmq!v7@=lA)Poydkr&pV`_6FN3jN`1?@1#da%bpCzl4If09Onv#CiZp)XKVi#jHU
zJa@%%fZpaiU=8ZDLfROXJ8!<IQAx)9O-z{jZ~WU9mnFoy!C6ASOQ(#kz=mLoX-y{X
zsoG2xZAFQG^YOW>u|1i1YeD?S^$6L#0W3njm}Pygd1_$_HiMM!YyZJ-SptY0_J!|Z
znz~ph>s>c55h~D03Z7Du8g>G<g^E=@&ih-651w8@rEKdrXlx!`yVVT>^gBU0m~A#7
z4vQ!oj!245<e_N0l6#V&z-$gw0p9q&l1_%D-NjlgBj-o4a~!nUI<loH>xUX|NB%jj
zFRH@gj7f*|EdUmsLsRW41UYUbwLYb3H*i{;*4$e*Da=IJf{r9(=aPv`TcW-j7#x)V
zhJY@8)()vhijpt-&<%jD2j6J87JL#;$4P9Z1*7Dye{(-fdo3NeNy#{48rUU9`y@6k
z3*x><M~7P_7A*Ef`lY1V|CJET)5eW8UFF+uPl2a#ngLKY_>@gujp(km6PC-0b(~6U
z$WrRg5CBZEUo{Rtoukmm3=Kr0B$<D`W?>XRIsnmBJKrBu#<%GJSV}rQ-d<Lb;=+vw
zKZu8{H2lgb9;W0j`lWLQYB}#^OW+NIEYDQ1>I2#2xYO6Ed}C?eZ>&wDe`+K&(>7=2
zIly^O&0n21d)P$?DJ?3<>$L0@W)p9C&$@3i0|R~EAM`XaNNc1aDs;MFk>EK83^WeM
z;XvoaEsJKO>D|6FNV+Bk%{J&HoU|EosGG-wnf5d|QU)m9^Xz+8Jl=1Sl1v<inB!7L
z9c@8<B=oi`)_A}MHp9NIZmUZ63u{)C82h%v#@b*!q_nyHMaYYg4ZQ?w5V@9iG8(e2
zqr)0%54gAV`<)E<-Tin%9|@SpIN(#_GuQbyjZ=f;U08~s>kEvCUTjh64hptN;)M5i
zEVg*t_2viDkZd|SMNLj0OAdP!k-ww0ury(2O41?T#w#YBOeQ(=7^0KlOSfE|@u7<#
z+!^iEiqi*-bImD2)E1ac0^X@Z#?^PlE;{KYs{FkDL?I7l$DgyrcpF`V%Z0*0#HHto
z753|)zD3AnNJmOU>bSz`Qu}ACzE=E=fN@_G7_bi0^3Q4CoL?=UU9$c_VB54)zZmAw
z*3fp2r^Iq2i^`nBY~_SfO*BhQ)I$BBDwl}??2gDjO^EBT#OwX3mJ&;o@<CvcglEOl
z_vw(ycowC7A6srtUw-e9A_8b?6pL*}D;sTR1Isy?y(pxZeN+c9Z1ba;_~09K9<Kv+
zXF`kD1I3<hl6^y)^R)qn<bmU)YwUFC6=#9KUr2FLjC#7ABm@RB&xs&dK4tqwT>Z8K
zD*oYgeS6p_sxB7P7r^o+C{};(+W2LvwD4Pv%K4_xfPN6;LDU!Jg}baKEsf+2d5~pZ
zYw&&2rc`Ll$RB+V=it~sDn-6=KE}vi=i_xAi@cHB(Lixs)BK{U+GDA^9l8?Ph;yNX
z!`KvZqwP+eoJ~n>SxC`d1P-bdK^8GPLNsj;A{+e5_aFV==6qC|m#{Ua+o=+ZDCF6;
zj-o^4oWs{tZN)ErqrQyJu}^PfXcOXG!rb?c;!eBN&nw_AS=i3C);0?KkhTa#S@HdV
z;h;Wh^}6tN3~r0(sz`V#a=lT5q*!^-&1^Dp3?{ic?I*nvawnWQqB7c^>NIRR#=$`z
z5aP9avY(W!tvwh^P<19xfjjAXMWER<G+;<5bn{-)RF=k`&RW)rD`9j<>H3b3<C%G)
zkIP<KO(}pD1*GXhvW0>#smOBnQh#_T7j0Gbtt1k3oD~(h2&p}M{O*{aO_<|_8=`&K
zelRs@fjJ3unyZP&WVdhMB(&!**$$5NdqrI|TV|4?s7LCf&CdRTT*Zt>)qo6kN!KUD
zQ_n9gomxuLhrJyo^zAc<$kD&$Om+yU<cQ5*^T%o=bP@CY8j8Cw^BuTjEUu16aKV>0
z7H`Nc8zh(*krD&b{APMf6-C_S!dp6Fr;hxBYjfb>YPth^hA&khv!CZ{^6E{t`K}i^
zvmb%%ZHH@_v#gKPaG&zgmPYtaRS-`v<Ogf>ujv82@E})7x0mtS5s}rONgDI?ILH-f
z$@_u^>{~=}L!OrbPVG_#CVV#Xom3bb|HNwlt8ddU%KSS*qAWetkRHkK0|7vlKh>BU
z42G7oY<C{)pOAxX3eSYwH`}C?04`>%FbT;&fxz+BNnP;Xz#Rb%MW?fU>JWyuhLh48
z+dU&UH)BthmGF(=J4GmMrcyu_mD2BwvG^vQuNPQFhE*+5H3Ko0jWFP8C4BtoqKV8e
zdYWM6e4Zz(-j1`qA)yLi^n&p@KoXOdPT9VVgwH?wmi@W$$5#6uB!49blO&RIgvdJj
z>!M0TtXogl+Y(hkx}ymp@9OHwvqReR7n@GCZBul?aVc8&!2NU0@%F<a843^}R4&F_
z1p3Uei-}-w)WSN#&Z}H{*+!Z{8xOWIXAP_cOb5Id$rkf3#n6Dn_lZk@1AH14I<1Mv
z7}mpzdVUGwqt<?)V^@DOnKUYYh6Ng=%x9o&ZqbkB2ukeR!hF|e4?xq6@cg+P!-vmB
zc%Njsq_;e1i*4*U?opCH_}&-Nfoi@eSW80bP@DoDG9KeQQLScr=kj3~=t$-1I4W44
z$c3lZz>!D$l^ENFs8qQkyE!FO30K%4!`2o=0dHs+PA6yF>)obJg*|O{4aKo8)W50X
zJtRXl@NPUP!0=$lxxp|bd+I(+4(e}3ZITdD{vrg6`^6JnxWH=2%ljC2zC{Sen#Hs!
zLeKS>HzZ2ns8CEZ+Ce`+?WZ5G|6OT=i~b2I)px^fsrw}QmW(~au$q-ZS^RZ6q}(V}
zorRaeMZ7rr{vze(esGDsg#ejCc({s5*&j3c@uTn(p@XJ);w3=xbBvFCQUmN9ZAaRQ
zZp{_LvbHmyxIS`=*8X?kJI~i82@EPuW39kmu>{ID-{KjxmR4l*O<XY2az7hFZ(OFE
zIo78EH#<f6t=LpGhbsolua(m`<&{TSeSf8TTQ^m7k<&)7_8QT=`t>=&(0_jmW|+?0
zeA8qMb}V@?6Q>(%0Ddi2wQiA}PFA^{=C9^t*a+KvT#+SIXY~%>#8^|Il-QWL={-7v
z#>T4bk5O^PxKi?&j~s9HPS}4FjDqfYZa?n$?T#c<>ha&DgdSpb5B#Hlr`q;;E<unU
zA!0Lx3I9!B=Cq9^{El0QgI!tv(1WLLKf)_%^W_9)Q$z~nd@<w8P<v8fn}uheGg?sN
z<*FyVe?}r!IHvjRlr)_u?WG!*A+A+kfrF8OO{KRfaPh_4h{jj-5%m#*P#NQhx$V*2
zm?HTEB1a2dq1Ix>Vj%kVavgjoS#mbD?0is4dnuP_r;gGo>Gr&N`@|%lyF$opr$UIH
zAD2<@Q!%pXdxPP;1iQUc`R+^b=n@F){mrNmw3MjNq!MdE8}~B5^YMBZJtJXW<0nGU
z)BICG*>3wW#@dJUgx!7#-d(ew=E+TX25vboI4D3hEZMAb8$11bRdhJaqvesDEihd`
zpljEA(?J;d>Q!g&?{y(%UfLb@V#ez2ywiQ<n{BV$pT$liY&DQc_qN=&Tzcq{K-@q;
zvKL+Em^^z5H3h)M`=PhJYP=LD^IXYgBm(<w?gA6juqtPEJ>tem{7gu*2}C8SqRv^_
zl3U#;xm8C~Zd<r-s4(d+^Ub(Ko?U(ssmiYBWDCGipJh@7G4tSPz8>)*`Q$>bFj@IK
zQ~oe8fzrzG#p{pcBIdE@TiFq_mnhcJX-o%WMTn>02z%O6p9TCR{y)tfr)~)h>GLL0
zU(KfwD|DIH_On3ajA%-dk*}y-$gG`cVP)X>e3qKemQmOh)kA$LLl7uh+zvWP0vCRd
z7dv66@G#jF@sUz+F7=F2r#QuQfR#;88J-}RF{|&wLGH{~lpEk5yzL1+3bDyHbQ;&e
z5x38+7Xlx(%l*yR3+`)pOa9M2wEs%F#nfuwrI%LnKGae(FH}{H?U#EiRefIm_mZrL
z)c6jqyK{DX&!6okrUS+(<!OfB5|gXPR@X1NT#KYJ*?BI^N7gENTRQ?2oYLnm;vD5U
zbB-92YfsmH@P#V9H<B-f)5PU7NN@zY2Ov*}Uql0`7uf!OR^RiG*vhs4@Y39O7uY(!
zTwyu~P@{58%HSl-7wYt@z|8=-01V%I<&(X0ig+z#YyYLvkng<VKCAe0QF)QTd*D>X
z48PH2$;|C6rrpF?1Wnf_=ueKJL&58>vYEQ_zLiZSilKWk*#|vtkbw^Rdf&za9roMO
zW`#oX%MV|O7tC60S;!aTTsQ`5J(z~mY2WUjlPI_swK^2=#<+&*O!028Cy#=#4S3aR
z00NBOsIUsA;O^X%G>;NXbhtsf`9C?=XJi5axyY#`aSog%d50?IG8T4ghhUmvEe^MG
z=W`T|1@-$ErM=ir)9Lc4_0Dz?a&0R)4{oOwAxdi=S$JD*DN>UT%AG$Zby-|*HH#)#
znOd%E1C0vy{4nO08JwR${I2MJ@aES=OnLT2aGY6j3{q1CkDTq3ty7l-SehFFlZ2wo
zI8&J>eJ|F^tm=;Xg{CT^Dh+0S23TQ+x72QU23=>2dD}^RGnu52T86T->zUatt!?)w
ziJ3c?XeJeVO)1woY32Q7Ng?)UR@Fnc87lI^N(%CT=@M3X29xwsZzqP0eLJ8v6jtM0
zW9<PNASG3$YN=;ZogW=gP<-r)p1PR4VxeFpDk+9BbF`i7_W;{h4iN4zs1xRGUV2bL
z;u7xjTvy^Y_F}$B?7W>2Q8MThHp14?7(&lSDA&<N+Q46WbXm4P$xRxN)YVwGtngRs
zuJjQL$9s>Oj206%LN)EIQzg8}X!MmBy!!B7nQb;yS#X`S7Sg;5tS!g~TjhPbc_nMv
zwHM{36<|Cys_(;C=BVV9@9e7p{DiIE+!bW-wZ#w?(v`n+3`i94RcN(s4DS4!D$HK(
z!kJd{au}4Kp{=@>9#IdA1MAUHDUw_oF1StHd_0q;ls1yDTs=W`b5)LTiCW)^@VxXa
zUky7aBOuUTs&Lr?3BWNI%12BqmDriYt@)oRKXW^%b9)8*T?I~$Yc(Sn>Z1M-<BWh3
zM3f!s9Ch}h>Vb6e?_dA?p&sj7TyY-1tK@70;vnl@p%Rf;TnlG+rP+65(fhREMh$)e
zC0Nd=_!qr5Lu8x@xu)Q6BwMVn?EBh2>VNWaQYWeN|DK`Ys@W05R5b=*&!$~4TP8q-
z*^ou~Cle{LfgddGSD&HE$q7Df5;I)9n%aq_RB<$(l7yg?<XZdM49aSt`fK=v&%IUp
zqwcyPAiIe)LKwGjg!<FaHbf7rvwSx-zPO?VC&ERJXS|KioM$kX$XZRk{}!SevYs~A
z<o<LP-{<E~#O0o2Uotknbm<+^i$21lEEDYiUEwf^`A%W#>@s|ydLU5qMW@dn<a)Rs
z#bGWd6SR9{D>sLRdWFw-$0tl1Pv2=4`xkZ}`^MXr{)YDs6Wh7AvL|OH)+X2<<}K<9
z6b*ism6SA$XzigY8+7#|bu`La?3P_aLXpK=x!pC6PLQLjo53#*=|G$b{VEEp1gX3)
z1#Zf0+>OI!vBrj|b0F+$p<GI<iCtlY61R@&`s+{^m@10d(2^$eW|P2E`3%IM-?5{D
zLG<<kUeIf%hqLGQae2tBkA?X&7~QeS4s%^ESMNth1vidELk6$_R#CkQ4w|AIG=ErF
z&bSxS)c6ZXvX7%EQU<+eLqvCu7uhn6Wak##UKgjQ(ON}WE`d#3X$a@le=b$yr86rD
z@e7*AQeOT6JSXL--0;R&uRIFO%CZb^EsHje%ZCxsrWXQ31j}&scrzIHY8h81j-|bi
zlhT|Tffaf@!F;ntJI$$NUEyC)ioI1=Q-7(pb<mZ8<4d?RW^Pxrr4DCe;x|?%vx$)!
zwsqk=R85@=E8SWi>+wdsh$cL4o1HClgXWc%C$4+ee?>2A6MZySyH1{S52vT@Q`F4P
zacR?R;>7LD$nJl_$=cUac~Dh$-7!1#`LyPD+<$IS68_rNx{!joeZd>c)Ac=fsXitS
zmx^P*iOOAMllk#cVA>`WxpIdRFzgiNFL>|9vQsYwn&B?0rJ=*Ov;jQ&01ldluA9Pm
zs5%Tgj`^Zct4S@IlV=_@x4Zli2f$x$@TG#%2(YmGRde1Zu^1s0;*M7MGpeE>eX3F^
z?ba;8XWoUuiN2jN!lY};lf6hP1u)dnTi0RPrZKn|!Jcl3kZIyvaxs|6Hl35?RQ7Zn
z%%wVwD{Jf2(q}hKX*pW~SeYAXC0`F-#E5<i8Fs~u(oFyN@RL=R1qg8nFJ4cCyRw+O
zG^Mg@lB_hdc+W1);~Y|rztyMZVp3!0FX-twp$JfJ&s(sWrY4(QsQx#-(eYWd(H-1G
zw&J8zAvdF_j-DOk9N3AhTT|`n8MsMl<(kI6oAjjQ>hd#2G?~U&XxO%o4^p-)x)~#X
zPO<UjYTMi4__e4EKhT#)`(*P?@}DR>2%2Y>vozAEcybptrJ7osldp#cj39g3+(RY6
zDt&zcIoYg*F=OQ)g{&M$5iK!!1kW5O!Rq)@(tt3v;2<82w!^533bQ`(lEkkZO@?QN
z=x-X>JM|LM#<GoKp|U|<3hddTenktzNCswtHuuSdHzyFrYEE|B@nJe2N@-pI%YlX&
z#dbS!?iTS5EAide9TgUz7B^R{j+FBNTXQCL$e$||(9YReJWvA8E}+ek`|ppt#VVbu
z@pc2Dh>&KktvkIvZVA<uf;R3eW{~Ed{$WV$CEa9Tr(YGj)2yC=7l)kfV&cjYpXEA_
zB#Ss_a+Tn#PdWe<EusUo1MfL5kG+PP`>O#A85-Oku~knoZ(4jh^F`)~xe5#USgb`3
z0A{F8;Q+B3I~XzY5~#*fublqolZ(*2-)Mk(wPCeF!qtM8>%|&MWZ&jzr|bij^{{_*
zC5k5u>y2i1busmeXJ&eBFVHEt45C&!;v<X=@KbXBu2b)t^|Hwpd_EJaB8}NR6I$!6
zK5ES(Ef;o)jVVm4nEENgLB^}K)>ircu$;}&D8b@>_)CR_IWwh;?Z1<Q&v!Ivd)y&F
z4OjM$W~N-G9aEITtZQ*A!&wKvNec(J!$#=Pe=YloyI_GMa|O*Yr&7-+K0=r0UHj*s
z`x0EFbxR4-Zs_Al0>o1?;!|gbV`o3{p@!(NG#Gk^+f@Y$Qb`LXp8u}ji!cm=yC~W*
zmX<*+bjcRhO~@wSRLdt`$N2@$xTxBc8%(q-;W@1NzlJi@6;EH&^A+vIH)q>XqwLUC
zzv)WD$@*LuG8BUXei=Z*L^<vibZJ6+Hs13K+LNXE;G<brtHV%NJ*CZUN~SS=A-Hg0
zD2pXbUH0Kusp8+!w#ehg4Gqt*q&j78sw$j+gju|AOKz%;*=t{wSF@enUR}|JJuy6v
z*m(Rn(mg^M1}5GV@!*Ab`1M$T$ejSYq*!pG>NS1HbTrh+Ei@h`&Hau2NWvG~A*9it
zGtn2u{705Xh~`sUpb)u|18K-w8iXcOONmqS;iZx>JbjE-t^Bo>==DPcOd_N}M0era
z*)+DfyA4>iO%8IgQ)Zr4lDi7}dwyUnQfS9dv$15tg<k%OJ;sc9E~l<<N5S89%GXfI
zOq>GvQ@>MdzCJ09^U|BO<OeOYcMv>F4aV3VIbHtR0%~$#+i2&MRuW(L_?&2n$3|0}
z2UJ!7#r#!jIN2wQj^gKX(leE#2@mDJ`(Y4vu&B$1Or!%p<1Vu-0h^lh;fXE`!A`Nw
zoHz4vtW#4>bJ`W>vaDIIn1~EacsnducW~v$KFad^m{4X}s9W2|AB_j*D1$!U9nzLd
zmcD*bHF{xETwtD@`y$1hw<RN^T_jby6%mMz<IWn1n=;a?wWP5*RFAx)&cX$a^5wI?
z=Fc&t;zG!C+t9^DTq3%xz7xc3Uv1?sFClt8<SOP>uz#2~XCgnEJVAM{_fJYLSoy;*
z$M;8FJ9aJ1nWGJMJO=+JM&5RJ*w9nN!L0J+)kme1{~dGj7H^_K^swc-sdZbF529Ow
zMh<F|g3pFBaV@80??gYG*FLlPR*Ly~<4Z3gSfDDZdKz9+B$9MtxyUOpro`4zW}D!T
z@_kv+q_N9Z9+anOtjb!pYUTq+b(%9YU0bvxoL4qz;&_eDmy(Y@3UG)T$zaSAe?B)j
zeVGtQW^jq#WFUSez2>f?HwgL(+YV-&@qZF)+(7yV(NYbk$oN~0j(ZgqjZQx|2|y>E
ze73t@x99i-j}Fs`pL9<CL(rD^3C-0l9rFoQcz#`ooDx6oY5@}J5n`7JOCXPeOM$<0
ze$-ni<sXAY-nx}0{VCqVlb|RW>%VjRhIxgu@)F<j0@VY0Gw_!N?!zh;i&W`E9?4I&
zLmgoJ5Y&gg@%EauR~_97IV^5JZK~6o3^}OmFH<Mj-qz%+)%st)ex?d9qLI;N%nC8;
zMgfo+q&O@HC7{D#(L%f<UbCTkA`NoV-<uXbHI3hx0WYt6chl+=6|hHvGW7+$h{~lF
z)DZyBPV7NX1aSUlfjb3hkuQnK8C~wBmly$tKrW~maChS(5kSH`v1K;@nf#}>qrnAL
z_D(tVFKZAYvV2`)@FLf3W0%78SrJdq{QV+dHeu^35d67TCNo{+=t)lDX~xLHv6(j%
z_3`?hY0@xo##dD>-bp&Yqq$1gjkor`_HxIsXfFPwZqdV<e`~OAf1EE#s7G(Nv}WEm
z+B^TEW$BE3_51M>>(tMS1)1TT>Zb|c5oZcJmnqRn!72TUf@reK*;{Y^-({7c=Iwmp
zW-U%&Pv+o<MdiDTE+a(DH|ATy(;c}Cc@0kO>y3tCG78*x5+Tw8cWX(qcVEKd1j7j!
zy7w;p2{cDuXTw$5pF7kR)}D#_xs%#e-_au*F9tqVNvK(db>_o{FaAN#cQ6222~7!)
zu3$5FD<6rtV1Y$AcfpTLm?h15<-O@A=-8rRo^t`CFi*{VuJOThiMCio*vK{7wUsqG
zoz%X<&A$*W7&3Gxx2^$kOz*7Ns;DD6NPJh-7cG|_u~HImyk_jAmfkSg;tdTwJnR~}
zd>1mQh*O=g*&j!j)MOZ!Ng`Z{2jI;154125EGPGEsr6v|T#yPkt^TSRKkH>RYqIVn
zuG?}lbluK5yam}z(7I+1u#+Cht!5QG9~`+${u2weUfdBaJoqa}N<8^|q{!eveqZ}n
z1lC~vRYPMv4$9Rp)Q>7TH?Fm8){S?r?yLOp_JyW2+88VJMgDG8Qv~R=+EuA$C{~Un
z@*;AwoZiiS2X8wN@ctW4SHrL5U12LQYiYKTp9=m?ujPSHMoo*^<lEVHfjn(*_}SSG
zR(ZS!r8}#2wIwZvZ%4|?ijjBM&NFk5L3vxCaTfqf+NqS}NlqTLEk;P94Y+QQMC?7p
zs%RzKUs=Dt;8^Vplkjj&0dhVN{*X&otU7&^I%^#lvlipUvX%g93|2MEIbBq}%z@8e
z!<-Mk7^`D)V6)RCz^SEbPS+krKiuzx{rL8Dvl07SY(4OCdnfb0G3*iJapCT5sCsx+
zKvY2}IcMFVlH-f7m?|$KeK&Dsox2{&ig@;-{9c(>28K{ACDSQPt2?!io0ViYy}^&A
z!H=cOY}zExf@2|RJ3M;*RRnNMe(zrc<H5GHovuu@@Ix^^qIA9>qyw;y9s>4OuxQzs
z!RsQvG-dII8cK4}iFQZ}vjE@#p?&#^=o#@&LENr^4cv=6C~xRxM#op?a)Xj`9aDJ-
zQ)HIBLfm1r-SI0r(@%o*ZZE5aGNj6L&Du2|*MGnW8v{Gzbr&CM5R_NUwq)IRw*-Uj
zd{;qxD#kcARhRz?6dyMwuI?mnLw6fOZF0WNt1e-kOSzpRB{Xn;Q~1*zv1s177>@j!
z2llabZUkwQ<nJm^MvRbvZE}cbA-DUs(L3tsju}lh8p69kf2q83(qZ%0xP}2?WV6&h
zsY8?2-*_*2LZbMH<$K#-l-<`fw<pRi4~rWpDVp90t!@4eH_RQ#NQW@*U16R5;!tia
zpvPK%tBy&eP9kknyFVhVVrBNZv-dCNxpa*I#7pScj$bWd{qN5om}+WN)t8u>tgu3K
z$ST)zFU?gPE3ml5XO=Jhbe{Nic&nEMs_iP^**RP^g$ag8A|!uk5j7pjU`%(y%uj}4
zxkR<KUn9_)d$|C8nUeaL`80XmBB1rVc6txP6#Orree!EaPX6Kd%e@OwT5M)pai~Lj
z;n(slr<eLq41!j4{un+UB9x;Hv2DS}9Mjf>J0cWMh_nPw{;!%%D8WztcU0T^Cm9}L
zn`eJjwFJdSCJ0+CtD8F$kJ}`>trFvtub)^xJZxQaook$$0-0j$xcV@P`fkp`KM5Hk
z9qgtK>1qOS$;#8&Yq*d-mU;F?zx~C))xH8EMcgCNzojQy4-ViWUrDWNs~rqPo0s{t
zMBeX|-gHg|DeJ$G1B<yiG~UMVQBuXw?|9RdCpMFd_OV~%)=yWa$`q^Symf+1H}gjo
zASJy~m*knZ6^Yc<6-1Cip80$7@!7&PKQ=vFdgo(AmuFOqHy_GgO~VMi`da?}5&E2|
zI+^n59xJ*2E`Gkun=ViN_Q<J2>ubH?Hx-R|TNBFuIG9#6s<KLo3G*&-zJyZ^7ujh>
z2xPufbB0LQ06Aq*xKq`=vD{GLKvThODrpF@Xlb1K2;ExDsC1SufXiSFLRb2~Nsz-G
zP(JjGXx4l!KQXwsfPB21r#`%r9wzYaKoz;jdi?0GE$GDMMkuVb4jENXm@i@$BM7lr
zrCIN${)HEwk8oc<_m&ZqmbO$n?4Wmo*C+s<eT{1}s$YvJZ4yR|<)HC^ce@TSe}7!4
zw&In`sI;j_P9!Ipb!P&F=k|qnAI3;440T@|_=Q}rBM<wJMD2BqpBBk#;)q+Gtqk{Q
z@^yv}4I{kyEYt_wK@EmZiEa2uBzGESRFi!&K9Q}t-$JN7eto)-JM_b?4V38KL3UA5
z`7w)LOFsVndaZE|SpPiRw5B#G0b>8no_*n3wl`Nc+~cA7kcCbu!E)>a{~R*yv?@6$
zx!mrVO+I;m?quCVLj=CL>Im8=NHi1JnV$2rZ`ZPK@!{bsRej&+npN~B*noiA3W&@M
zWO3S$`z*ftO5Q%b$$^?(blCMfQkp5z0u_FZ{l!jNY+uYem*pa3Eq`ZBHXycW(K}?S
zPI9`UfUOxTz>(54r6AigQIp~S0p&m%zp7QJ)ES4Nqb6Z4nMNhIg%Gw;IZ|m9Qcl#G
zRDskA-aE$QD~9eBPal6rnRGfpE}3eerpEsEjxX-M<Z}ARx8HoF9G}ftca4$D`MjZE
zjd@;hV#ulB4b+Xedi5A;sgNn;s;Z+@Gt0WF+33(=nrsHDRo`RQq|G~>qnfgkmG2IB
z#2R^cc%qif{IoE|mALkdqoW%<)?4OvWl5Q?v)mqj`j{|dgy2+JS2wgr+SWETZX*I$
zYBP#bwDvsVpjMTLTTuycGezlRTSE6d57T?pE>3M0jMesYsdXzT3UQezB`Shav-0l3
zj?#MQZfWAEHt2|=v{ILemM%y+)45gKbxuVQB05*JZB`#@XQL*%O)4=(#OjD3ZEI)c
zr#c5Es#QwYG4zA-*g}U`k8vJk^9gGi=Y^Om!C6ex&)@f>a;8eABvpkQRJD6qX53IH
zX~Fr9FzmDyTW4fSw3e~MExvn0^8p>;AkP)-wT-GBqq4P(uBgbT)=@19TKS_Stifif
zpaf>MCeT@ja&7w>9Z}exyRfoxDrK9M`QwJ;{8vTak3gg5wDHmu*;*HE=^5kL?+)Zz
z`TF<2;WRy|hg7Sys0Xh@*zXUtCS0ZqSq!-kwBT7&!q$p6mf!+4C(fsfP9G7WbAsCt
zx{6MlX$@jj?sC_8O;TvvcsCUkUHNoZWDaAVFQ2aOL3CjIysR`6L4<M8GaEp>W&ida
zVdx=6YE}f?Zo(JaG>PcEQ@(Ri0Y<LvI;~0zp$#FrCfhpmy=`cde&*JodxZ=oQ<lb3
z8!b1MZ8mADm9l)K+6%@Sf(2_gW7iFl@5Vi5+$$)sERYL>cZ5C=f)-~Y4R0MSG^`O)
zGNhQcQBFp^^hDc`&Z;k{BH&r+4^_|AFpl_6*L|%OYdl@(e)4*AUTvCc+s^N+=*O5%
z;PuIa@%tn6R8f9S<hGGt*S71k*JD+r>AA3KG^sQxnndPIreQ+IaCk+(yM<wo>pFbD
zL#$(26S3-Zp(JXWNo&;ckk09OT-Cy;^Om7^gqFzbL^OrGMo#B5^E_e1(03hAkB^FU
zwMvaMyUuZQbL4n)R8?t7EKd`4*}OhmVI23QxUybW;;LdLj8W6!c_uisq5N4Bvb0dQ
zq%Nk+ZX9sVaGoZtHFVC>_dU~jp^9+6%uI{Y0&+830-vA9B06{V=CfN1T*qhbh9mp^
z$aFgK?W%lGDVbK`;j8z2`<K7tSO54wpk_-fm78~WydGcii^r8;zy83#{M~Q)%WuCS
zt}8ACN{X~n$vJHTF@vd%YClah`5V{Nk6D1hDhH=VZ7Vh-*kGxmGtt)CNU?BL+n?u!
z^)j<o9UpZ`7f!X}jBvae8EymXvan7I))-1zuPhsuG$U;$tqoHh-a3{`WO|q|Kq-mS
z`GQLp7kWasxuK(6gG6O6>TI5C!39sh3v|7s?>$|(fhriIn_u1X9i|x_#=h~=)@eq>
zF!Up1f1^oWt=gikP|~_35~A}@ozSFg_5e`Pgtk^u&p1C~%${6Fq`2pHUCxo*8a^j*
z0o&~m=bwqYb~9xzxvH&&D9JX8Cx(^Y_uSpQr2@IE)S6jmb-Aw6*dJdLY{$R&SO1(f
zMBcr-!x+OdZ!;5}4&<JenX)QboT!QB@%0h$(2B5JF3=2dR*Qs|AubKyIp!F-ED32a
z-~SBlSHm{zG(A_b%DY!*bum{t{Wi07LwS_*)u3c;W489(3cQk=&6W@vqcaLkQx|Y;
z^KFCIOMOPsbf%>a`L=!sw%c0k9%q7DK%^CEja(xX!HU|iU@c-A)^}>uDwecez`+?>
zRl5}NkW+kavs{N(t3MaYlky&GJ%|v-z1l_?N2#0IaI>8eJgpYiX~mJX?JNSWLwrXU
zy3OjR;qX)w$!Vo`I#-q$8TLm?3)EDo^Ak1Bv@yeYi@1&^hFsao2-#Xk@SNST*_Y6^
zAs{+Af@^zeLnU<XtixD?GlDI4YcZLPGO!6Lw&x-uD!|w-vC=kj{?8E8G*Me+x?Gez
zFP|>TQW9@I|B~QG?mv9X>HI)02_3VhwiziU_BXeD@hAU?`w!pn@ZlSBD>Mnu%I#DN
z&3kr-BXOG4YNIxAI-j}#!5F+(ey?+m>2i5SoV+oLY6^s*$9F+}ysipXluK*8MT}=0
zc1q-4qpCYgg@D>CbdF&+GThuU-n=5EL|In#2&<}Ur3oPfLN^j?rK~H19cdfNn_@(2
z+o&FDFI6^`rnR5^BmK75iWt1jq>3VO4N|U<^5zs^i3KikA?BIgZbx@i%0SzAh%R(^
zYltN=PZ#oXW(bZj9-#De{iBNTih9+FZNvsCsj|8>eH~3qbV7-hZPXnNXQ6kF(sdkf
zUonnH^%^Thtt>kCjMn^=FdJP&syv@P+abr**!gPb)X-<Em-{C!suo3CQ6+lzi2<vb
zIL#>qr8bl&o-N*v$ZgL!><IfC!gzxX1J!*(A9bXhw&hDMq-9p59c}xADmi0}tV^P_
zh;N45H+$Z_eT5Nqv2flq4kN8rE|&{A!^7hv)(CHI2KM`a_ePy0w%=0_EuNdwRpL6c
zoG+UQ$q@RE-tJhYg=wC+yFE}F9QJ#rwUBG$64jW!8N+Fsnb%dPMTABlI_7C2r9u&5
zT2|&cQM0;Nw5lGEh+v%_zSlGBO84lyj@$hm<IeN+@d3nct!`n;i4+U}_MiO=9v&|Y
zw?~HEk$?0L|6}go&%Al#`24e9@QZhE`4|84U-55#|9ifgpEP0S6TKprirtcQ`GMlY
zAC=U3r~HWOEEPv}N)!}p@VyEfVmvX=k*D*S=9HA%dyhDS>nx|sL>B^gw+BFb`?Y4Q
zv7}Ayv*wk~IV?^`QI*1Bv?|+9ZS!%_O2oU4u?y;~-ma*Ywyml_7QyyCgH-yUye1V~
zM3tx;6&b`;+ZHh`?1nwV5Ln~N>2jfS149V3X315JjH|cLWO$uD=rs{Qm8P{-<!+1U
zAaJM3-jp-ru;<n7XFR-pk8OoDEjrDCjLeUm*9*tP9ml(c%j0`h#)u)f;aM%#bv<1d
zl&^r!2<ZC*)(4ijAX0cbpCL!YTK?WY_^0f;J^$+8{sz+`w>O_Fim)cuwbCC4dT%MV
zQet6^m1R-ir_1@wxZCsY?HAme@5sOVnsqJsP|2wR%5_VrQKMFFvHkR<?&_h`YSA^a
zA*G0JM2ta<*G;$iZd|JwwZzbt-du%T?DPA4ZcVf+UB7KZwADmee?U{-TTPU7^@}p9
zXm%YUgSLUIj_5I0AqAOW6XGM5ifbOB(_&i`brz?W<dsYh*J0k*q2)#Ii%sozRjE0r
z$y_rQ*Yz~xo)a9kv1nK+IBV2KrSsIhf~h(g0iIja8KKz@?*jOOqu`oe&|N=}OJzEr
z@Z%lUJ8C)8+JfYJ>J*W%quHL)EY9j>edTap(FDw9UaPOHN!=A?yMD&o|BCk3b)?m^
z@IpC86`namSI!<=0)MSt%g->D4&XAKVVh6-{c&^{!`+)NdHwbmTuvwIcG*~KAe(KD
zrt&>+UVVlSy%q`1VQWQPK)O!3O4iVYUePV1YT?jrhAtx7YWE$(?ndu-Yv^~Qss+sk
zS7V-A(4~}To{tKpww9|AT<|n3%Q~^F%O|dmC`x_U?YO;tLvcMzOgub3lIIJ1Zzx{v
zT}qe149an>S1TNYb&h5%si8bwl>b<p0jb*lx9W3et<pLSDi*8FV2h=MN)r!83wN<j
zQTa7eTVh@o#2R)(PqvQG^|)=UV!6ybK0dI{GtLU*(BX$Y-dV=e6Yl*-ot}XjztSM4
zq(Td0$wgK1t!cMKw&$W~Yvs7xak%>e*A3dXjVr6~7;7cQxS0U|w0M>^Y@9k_qZTxt
z-_Q3w<*o+6S}bYNvsF^h=&lYEsiahJ4qQ<SFzI*L-3YG3`HrqX;M_pPvPwZ}!*1)t
za8)mF!j)xVnI^1kywFxRQ~XHH>Weyf&+TzUXgobVQfpyeSDjm1^*$`Q5Oc+cp0U%}
zt)17Af~{uyZl|y9JJvKktAp2dVH|rZjq~NAW3`&_-TQk&aA2Ia_%~5V%9*FrnTMx|
zR4uvcJ-x1p-FU+^X~!$2#2gF834I7uE6mfX&%N)J2C(tETN6^z!tHvG9QHd#3&(Nb
z;X%)dCc-7Hd~=%k>~z5Y<~^T({wx0UFa8N>DGZO9u)kxtyJf%Mv&XUz0srl9`Q3LP
zh--S@=YzLuxW3UJ9)G}^`cX;UyU&g+^TOGIvzBNLONKH>mY9h#W2+Lx8$!ach@<O0
zcQ+%u(a}9x_J@ISw?Usxoy4m&PRpdTO`W56Jt9_N`3*WQH6sEcbokJ#$U+=cVO_S7
zPpxhSwT_$z^h#_LiE54+6Vu~NP7A}U!^W|wJY6Dgvn%OBV2zQ-$0uxSjJpH<aDzCD
zl!igauaqkS7zP}5YoRr*23w+0s8TVmfmCA5JUqU~n0LCNgQp8SJUgauCp9tl9sA=e
zUL9Yv%oAxoV^BC5Q+0-6D}-U-7r*!g!E~%ioybd#7y_^EK4%;TYKHmo5wqL#4}SgM
zU<^HXZ$HP?%zyB!|2C)f%wM*D%U}KFZ;{#39|k75(5kj&ODrrg^5Nm0es|#2%^OXw
zdWUh)531~S-eJ9`S-p_^(9!Swk6Bv(6|o(1)Gc9BrLnSyuGNg&XmVz=b=b@^MU6yD
zPNZgOsA5eycRZ7eolzAr@T@Frt?7)*nl?^V!CS8!7ZLTA+qfjuqJCtpZ4kZL1SXoG
zn4mKpp=`s$HG$Ki>|5?*lyIBULi0rl<yBqgP2J|3iq(k|*3m?Lm#C_)Oig{$thdDJ
z1VC_}_40(Ng}B+4ToYH>P~_G;#&jTVYiIOWr#@)jJ4#$x&)?AxN98k?LOwkp?n3Ez
z(A{9%UcbMZXQ9cp$`tJLa~9FI?6WN5d%Qc@`s!C39rea4w3aC7w2A9v1Tk(CASp#d
zHog(<r<+}as-KOtjb~+_71XvH=!QL~%bEL!52PINU7#QLtT}N$omkhE)*26|CoU;7
zpU#L_y5Lp(vRU+4Dsf#XtG4FVBzRC%JtzG3IkUCum#E79A@ual5xggKJ<wQ|S(S6>
z`V*}B+lh7$sFfI3mh*{aRU=on@ep0t@va}KSl&OKdH?<!K0G|)ts$76TrDx_M1p0V
z$W28R>#~xH66$lVT)m8*EugOm6WdyIio%N1naj#A7J@kSS2Ug?j?mwzx-dp!d!SCL
zZoFJBSSse#EyLjkKkix2CrUjLmkYV*2&j#lfmn;#6qGS;if+`<Ipso36=N$YYa3I&
z4vV&$t=T;CGW*@0aQBwlinbG@P{;Osex&@vW^w-eqHQYYTGW#3nz+fvRjsu?7n-d#
zeePM?f`U~cguS+JRXMJETRT?OTH1KT9A9IGBXlF#s;v{blCyRrgp9R~-1O}7&fzw(
zNnTfSj5uTP#$Yz2PylMiP#L?xtI<Q&sYFs7UEeYE9aI>Gk(vu1Ki<=KfuVPhE;upl
z#*twh^lYw;>2l`j>4{njcQ;2`(~d?;N$pRp<#_jsuYUVGnh5>4qYFK0T{xXDoX%(M
zIF-zP9B9T<5YFc_!CU6@nYhk?(yD0CcLR~6jv2B72G==!L!A{G&&AADcoI`&nJ@f<
zU;cvIao~JDv(6LK8o?R9J+J)XVa3t$hyUll#(2Z+-5aQ2I^otG=!TAL8l^K}D|Kyr
z^YH^+-|18%4C^vutk8vlumA8juc;rE)QK4N`6&%2nKc>Khlykx^JS&YD-tuG-R!wJ
z?l?7=1;(Lc-1k^h>B8n+vnH0fQgY+GW_lx9(W1PT-VWr_7&^ncZA@Avu9=c5$D5Jf
z_ljJkN@<ZbXRu&oB)EW#>NU44lbZ16N~xLoaiUC-Yyz>%oJwYjg}Y<VI1IE_IiJp~
z4>S95PjDk$H_{CqwW)v1x=v6Nuit#h&FwqR=LarN-|Bdlq81s!?<h@GMsZn)%M5vD
z3@08QFC>ItfA&jiODt)o#sqHQxWDD*c*ob@eWhb~#y!J+Pj?uxw(;<APpz3T^myCx
zc$w(>PDkBrJcF^@F?L5<C=ACp+}+$_5sdBd{gGe(>7Nt4=iRGca(?<hasS<Sx=>eb
zn}iUEIbxb%cY@zpqFJ#${QB>`#_S5O@Af!nh^t_OL98dOJ)JXz{-;}-n2paPRsF24
zYRf7XT76dc87o*3s%ZP?6A#g1tWMpio6@qPzHqH=_6|)4V<T#o+qjA-GPUXu=Fq9a
zs-iwyw5m?$IVn-Nu_;y!p%Y4)!TN1X)T}D9_6fchBb&*#+suy@!P75qt>luGoW3<T
zombAB?|O1=G*P8-!)l_qE_B|LS|d#-`u$F;=$f!bDFALuj7$@9su(Ziq%8@hjtJIw
zRO=}<LuX+*bGbay?+#ek)9PliGCg5RMna^9BVvb*Pp4^{TFU72Et`tCZE7+Av+?Eh
z0&BFI$!Q~pqnBMN71_w~t!*Rswh01k|A=+a67^@Wmw=7i=Xge!q*@YcBXc=D5ZA~u
zoe`kh@40#PMiEjIxSXH3%oiRX-Yb&Qg#9>RZBOt4*NtEe>vG|Ae#9&@LS-EG41JH3
z#?bHC?+&z*n72`d>Y*l>plHG{?yx?P)<wM#Tfus-`(F$hO%8I>wtHL`Qq<PjwTCOl
zF!+InVVYI+FrQDVmfplRn)EwrZCb>(tf*JfNj>v6#lURnn&mpG$v|#u61#n%Vhwf^
zTu5%zs%<evXa(yuLAFkfcCXyI&O4>hq?Ht-njxb}SBa}8vsN%p=kVsFwi9e30)uS)
zRh51i<1jXBHwQfflyjyDp$aDymGRCIywd`AGX}RCj}tx>FFz&LWubaavTDs6RdnOf
z)%rZW_i7`g6CG4tFKphCW_#w_@0c4{A2{5-CEUEG4SSO7nU%?c$!Y<lg3Saq=|rIQ
z6AEkeMfLScCant=XwFeLYKQ<_Cs9yY=QCm}!%fdP1VY!5V|<>7lhZ;U1Q{KM!rj|<
zc&k&XyzejwPxlX;pDq|@8ONS=T}aEqy3Y9E8IL#Yhk<Xu`3CD8H@63@vCPX#i4g=!
z)fR2v=>(yMAyw^0l~jl^vzEHiwX_yUF+x>9qH~UE)qN1Uj+iUXT1*#MqwX#5Jvk+o
zY2siVxy<Z+;2-_^Px<G+`Aa$v!3O^F{)F*{{ki9x|KtCa{p~G(_K*L9@$MZGgxzsZ
zZI!krrjJidOX1)<j<0T@6xQ>Zlk?2WME?WZ{XZ(Ho6ZZD(?W`gC0Euaq$M(yLS7QC
z6hchgzI%o59OtJq;w|3799QOLqNYkLmA>z24W@M^#=^U|cibM2Os6w3C72c^L5m@_
z1Vt?$Xa+Byq0`JhN9}v(mPpHr>pk*pikzuboo$#3bt$B2=JC5nOlbtM_(P8j4ks4V
z9IZ5BOVnJDYIU?$$(%1Ia!cCw9QLY0s)?aHau{C`W8#~2(IMH^0YWK4DX_-Ghi^~J
zr!!yt;*M9xM#+V~-|_a%7c|l8K2Hm2BYrm;*7xjRz2nupFS&d5ilHC)`1A$;{QvQP
zXMTJlSw*h)`#TPY1N;4fb<M<-@jk2Tedw{F$BDK}q&ar`JKo&BX6OPnX1@6B3%>c?
zZwX$Q<`o?5N5|pSfLVmyjmHivOL<Rn6JP$xk;6Mn*Xz8zC8|12Y9+4~s!-EUzhq6*
zL)2Ln*0c?=DXppV<~9$qO2fDS&nd&EKosTO8L1mbX#2t2^S$~viCA)0#7z?)DAi%S
z2Y)?C3&u1`i5rh-0d!WTAUK2an$iW*N=pfHVagE~6QSz}-sypZPT;sEbL!;ZFt($`
zN=obI%x|BqB#iUeuA^;;VZ*2rQjC&!b7H<+==zRw$xNq5{oYnS8D!SNc9P?aCC3$S
zENLlN=g379&FhFX5uw}NDykG$ObGh=RBX)<&(suQU9tTU)9t9Din1%=N1vU3c2|En
zK`URK0wP;$W}7GZy;c*54o;ViXut93w)+=a*^o4CUHu#poW&Szy<36M@1RuT`GIx5
zsAWcph-d|!W7H`x{m3$1v@^Pu2dy<qT#!!pU*|h6X<=R#N{obBu@sie6QwFG<MpdI
z7!jt^$}(RFL6fzTA?Jj(dnNu~TiM%8UXivXqFVfvtRfg%#iC~9t%|Y~Xo{YR)#*RW
zM9Vtsd5sB6(-CL=z~Oj@Ymrv7CIUdMnVO@z`}=?oy>fy}RzbyP9=%mDSmz0$qoIU)
zX$_Tjv+Y@Z-_D@6C0PwwJMWR&NLldKF!np7==7cSe4?zYCUi}3M(F&ewzL)>tlIx5
zxeV7l-mAzdcoj;er1l;{`!tcpLhzm(3u!vT;|GdB-*xQw2fS77>CdIMUY7-0z-$8V
z>$D0j=B_qN0^7V~>jSQ|?PFILh7c@b9l08k7`kp?Jlx=SH_TO-=15#0nAd~}4zW$g
z9#%(A3*H*M(`gSmWx8Hf^d)DOS<QmYwuY#6qf6<y%~lmT%(XD~hHeOWYZ&^Dd5)}?
z6D0)3u~+_Pu(+E&RvKwOlVjps3J<55w62VO&+)jY>jRg|#JWt3eMi?jTnHTRZaGgA
zwO0CJU`lFZlXGJ115b}nc;^^58jH7r_kq*-!eLY%@nu%t>4(QhEtU;DohE7%e6y6M
z7B4Pzv{G=^G3@rlQaN8{Ks4Elkq?jed~^Si*LQdP(|_{c;{WxV-|%XG<a~Z)*zL*X
z#P2?QWR4489F7<piA%(Zp&DWDI|w~-7<u#VHS;nvbe_+@`~~F1(}(xGf4nCydrJLb
z-H0ER)IHpvnJx<@H`Z8~a%QQGDP{7S7{nm0l4It4Sy`hhV(Oasc$t{y35%ic)Xmyj
z{rf7A+Bi-PrgdRSD^{%f9hF3`jkH#R1jb=x-}f8eNF$fb({v$Jk8ui_f~tp7&T8Ii
z;Au)cJk1Q&kZNNWMncykF;g>?th|_D)zrAL$@d>WvMe39uii0QRiLf&%(oxEBDTn7
zQe|k@?dZAzV+@zelP>I~@c3b3ew-M0C-%1=ndXSw9oQd_q?8C@M_V#$TzI_ys7gB*
z_z(ZlUvPVSgR_=h-|_bDj^BUt2mbQk{0qiT=f=H${YHK4mW5?qvBpxYWgK_>@{2#=
z!~DekeBt%($Z;6S^F%J0IW`{7Ph2Lgj6w)ZaVD8eb&1`<(Vi7wt3FQ;!T0n-kGO{O
zmOFu1GV2;ymP|>F^ZgGSKK$>CS~kAMM#{Zb)KZnVD7DgZBF2Si9ljeduG7{7n(#?0
z+q&9h=~X4oS0!F+g;I62n57~mlH*L<ywR>Ba#Vy7iOakh&=yiDWTTf>Nr|$Y@nu5t
zgegUpYgSc)G3i05MzDnt?B=<rXk<er^|dytjTDKbv|@dSv7TlFRXk$CrhKy$toq?u
z9ru+>M&=1CLdlWXDxKBwQG$}Po!x{GB#g7<rQt$Pq2PQ^YdQi)Kvj8m8<7_;#N|R5
zZ}gMM&|1PtfjUv<3{+ZoR5frbYI(3-PS4XLMBVOnLQlntBHJSN8KKjBP*J+JaVL?@
z_s+Pj^*vLIQZ0-%YS{ZTgfoJ5T39*fxO?@6hk4?>oLFnZR((%GBgL7o|LQlqdh>4M
zKi20KciqVMQq{&q-R~>YG^wNhhDe&$lx#j<`1O~6%3*iI>2yy@3$-fQ+DeCUD(p$y
zY*Ml4&)*W^D>cA-ZS`I!TY%a@4Yk7?MXj2#_|R=6aj??py=Cy8$*Ab%=5WJjpZ$`3
z@2SfZEvfQUsiv7$EB!b!>~5c_3bhnUzM?x-XV$`om^-9q)+OQEX5rJI8jBzI_|TJ<
zg|e-^z8@7mE}FFNcRPlmC#@54naGzX$Z0d0_jv0Vx}M<Gq9i!0o`GgVx=~Stvkof>
zBZf`Nfi;^S*@nUyE3{fz9=~JBmC}v$U;c`I+;bRvot*MB#m!;It`Aseu+HF|&iyoE
zHZjm92-++)Fq<K}2w*YZfwfEtmg=z0QV<sFc|57b7~Axn+Op1$%?*oEPeg>A>-Iaf
zXLLQqgh<;A{Z;uGbApof94?Ji5}gm)e(gQM3qBaeUB~%+La@Akb<2Lg+lpVP+nnLk
z;{%|tb52QxIH6NCH+W6bhMwEQj?3kQb&kW$5ooOQ%x)a{=KhgsTF5bSdpvGR*Tk;x
zsOO18-y<T#xboTSS6HdUbs{Y@WtrFy{WhJ%P<-HVTDY9fVDz&{>C+gnd6_6pg-9_b
zR6!!#pP%@b-~NIB&Z}2^alGNX;Tv}2z*k0?*2tHiea3u#;M?<w)0+4^6%Ow|@at9>
zJMD-p>kO`Odv|0x-!R<n`SMqP0&(F(@7W(m9v&thr-vVQhW(fY`2O2_T52k>Ek%_Y
zVsN=I7@=>KedlQ<@ieKSp2M<EiRE%(nHL?ik{lNt$K$}*_ju!R0x2q@CZ^%L%?-U3
zywQs&FOilit;uGvDcgaerobs@WNpORJ`?G^^MKACGzi`**D7N-i<1CipD&kX(HX2c
zUCk;JYrN2p2lmGs^>!j{Eu+fA$FG^D50usz{En_aGVXR5JS7><=ZRgn=haQ&{hYBP
zOy@_M3A}mzj<>Jh@NhZlMO%e+o_KnEAeO?LcVF=O=9aGSnBv02GI4Mn-g?f{6E}x<
z+FHvqre!`pKJauo^Z94LL<E*v`24e9V5e~#o}I}#v0k3|>g&J4SkG^N_m`x3;_mHR
zj{7~Uth~83goDEcA!G;DQewphgLme6CaNYpf_0X@(}Bi@8G6Xm&w-p|9hFmxA~<Xu
z6gF;~(*r!Ni(c|_Rgjr&tWJWQvDIRVt|#ktUT!OuvZ))zIBYAFc;Rw>V4lvjynZ5_
zi>SOk7jiCmr$(t&1mE{q^ui6^;bHUaOVqrm6@XD2g_0tn>j}Y;b72^J6-!vJZ6NCr
zn>QPut<BP!dSdyG+|=5mWECJe>nS-=V#Zmpt&x@)Ni!naeltZ|iF(l%>V|}+k_Zw&
z5>!iWWmD2!TPucEEE3E{%-71drikhHR57$Hlno6kb*3&T**feMNUf?aR?(__QYCN1
z?o!o|yXL(4e(|ZLouf`*L2%nRp9at5=4GBqNx6UC2Sx0EhA;xX_jF-ox8D<no+(Zo
zyu<VZm)eLWlUt=#uoULU4{EVfE4}yB=13doR*dEK+t1kV4t!WwQccg)0mG-n?B?bb
zyZsGITzNb_Q1YrXNn6vdfpuHkO~)v@rirH_n*f{o7)7(M6ayoI4IXQS8r9zrYiVK?
z%@Z|{AG$!-1-!FJh2!Bsa1M&9+52&)e7BNVbK&4RT)(4Wne##`X-h0OL)ubUQ${c>
zF%x{ni6GXg;<YH*-0k-G@kp&OPZw%V^uwq|$(wRE1U<iN$*iZDw49+NoH2yqK-Z7>
z;OV=dyv<ahW<{3d%1KjHUDdT_n<(K_Y-E*pja>O`+a#VGDRCk;!#wLi`nG`)bwjd#
zYW(W+FWBz~oZWo#oYg5Y*M!a(j97}XN;lXLS)HWRSex(^h2@gSr-^A^nPSqJxhU<#
zZtdBtlKN_1oU<xduj|M=M?Z8JV~C5MUEW$`o9?1>TFchj|9{i=A5E5IS(YyL9jfFq
zGjsQd$VgV1)zz;XgO&#f1f+pE{xZfGv^)_2?*Y|aJgbsL2619$mnb_FgMG}RA@Zav
zQzBbSY7oiX{F+|Xeb!#<TN$Il>&o#ku<s0~{eY?)Q25d3@96u1z9DTw*yM9Fnw*n#
z4_jp0GRuFa>pX|Uju#}$HZNS)8Kq%7?1?Ec&l@$t?R<l(@Xle8cTrA>&UMr*O8Kty
z@>z`?yIoJ*CW<i-Jtg+cG2{G5jFF+&pc8BH%(2pSj<jyXRM_%D-wzTJ%2_NCEy<#^
z;SbNB`1IiepP~65zIe+Y9&dd1?k#`+>tFMu-IsKo#aPYr`I+UzNA`dCj{kiqXliSs
zsPscmNqdgRBlm|rZ-4X!ahkA3G3-Wg6Zzu*sC(*98tSI&gi%7@t_HG3WO3^`XhpB!
z^Fz;W*YUgSM$UyfWujK(ZEG(?jkX5sJp02Px+-!EjDz^&N*2ntx3;Nm0{IQGuzs2_
zR<l2j&0{3%gp{NeK~>Gz^_VKsddPufrQ6SHVb7G9@fr@D_@4ypTG7T3a%5gta>`uK
zSLSIIe@zTL><0Gx9Yt4CS)~88q((@W6z0R}M^r6rs@@*}hci3+u4CwLq?(zxjV=YW
z)(l;THwJGEW8YJX=5l>zH}2W*j#y*3%oDCgYR-K7^og{s{PRElXVMzz`#rU4p3aZ#
zhdoX!_Fa$B7Uz1rj$kyyZcpzVYcqDv=SP13`+s2=_ZSLT&A<K2f5%~W;NyROj~gn(
z;HgSjf+dL(luIO+rVJoSCY2#oi^__cB{nZc_CE*VU&{)KX0k6Y)lmiGd#aZ3gSHNm
zJaAM>EddxIZK#~^YQSs7mNF$9NhQCA8!VKKxXrA$M_GmCfFuKGp#ot6#vGc7UwQ=H
zFyj0m>*11crBZ{mI~Zq}mX&o|>34ga>t9!krHBd^Lckc;O0tE$WxMXx)8vexJz}Xi
z^^*9M7fZ<nRRo=rvoIcw0#&5jb+;R!yr3oIl&q8zRoF>dIG0S^G8S;dzL|$h!|}9-
z|2heH6e;eG(-~a~ei#H`ozp8&xu%6011bfo+fk~frtn{xzZIZSN`f_|%FA6T)eanq
z*bL0F!c!>M3Bm<MicyT+FP>HJ1%qOQ>GkKhBoVr2zu%*@W?inBxNzq!<9?(!hTFEX
zrHFN&f+FQaj0vNy4Cq;Wgeb+ZKj4OuEe59LMy;)Pv#Q2N(>QB!Msq!X;<`*+Z=dL_
ztQ(!SEs3aKhq{~-&iGd>k8I~;&0Uft`7~gh6}G0)Xax}<iBBshfm&Oo=)GgzBAv7J
zou{Ni+!nUWBN*^T;kzBf@kB0_h#`hZDV5tik=84^2GCa4x>$-*<gk%zA*95XGp6%s
zC9rrXGE{X(LP|{2%sS6>)?#Cn8pM=%>7>mmuq^@d201}3cx~zTN5<1zI;S8D_#40*
zjdP;@ru-5V5^RZ44SiL`!ca)4)&L~FI)!Uy_*^nIMQ-OaPoF-~DcBDk8VMNu(Cdr8
z_)GSOowVU-CHIDv76@KAWk$SmR0#`;f6g*38=pQr@%7`4^RzL~tFT)Ogpvj2WhC%m
zR2KG``JQ<PGT5mK-bpQ<HAec?*R}Zvfm~kuj#cy6Q5Cyk;O*UxyU|eN!sy}cn=dFi
zNFZUp5o?sgf;0HeGL9n(&3wCYeSYTr_yiTk-GFt5oMfX>au$wg`;L#}p4+^#t&z*q
z8AWA3j%?Enqm)$c<-~p*S(cgm(-CTwKxE(HP2$bN0b<0(gwH)$dA69jsD#dnhr`I@
zEwiqXX`YA}s&-N>rc_I=8&V6Q^n;{hQ&oKR=>tDM?s$8*$7fA>{K&nt{O(u(%rE}U
zU!vWB?<M$#()8Bhw3b{!mR|00?CA&3>FryNrxU5`aNf`j1MZNyyWag#_tc+8?Cdl-
zYm`yAoo6>#`l9&48@4%+L*g`!C@qym(;UcJ(K*YZGbkl;vFQhP$35LJkhUab@zfsT
zP5SRg&lERERbK5;%$&}94(-oalbB1lnAo<BP8pnbP!%dBSd;kUYQ=ktYH7&he&DZu
z^oHHrdv3L|rHH(2Bn`7Jndvq$&nu_<1LNT>es`qkN{kC7$qqkk0jab%D6Dw1;=;#|
z?-}}yq2JN%Msf_~6vX`Apv{-dGv9vuHSd4?Gp6|_4-x0ELq|VI{h76z>w4p0cbxh?
zE+u~RPydBK{N3MDi|66=Q>JZVdVbHQHhSOF^_u<R#GALDqt_XS;xz0z?e=_lev&k&
z690%ZhU^S$nn^hl;>K-S(0vu=l?6>Eg^Y6<?J6X2-iXPxibtm8g3(sY>bj6)A^nl%
z!~g9|wUd8oYq=yQZzaQ?DzPbdxh0Ycq>WmYtX@l^)Qr+e>g*E4Z6RVry-?zj%iF{@
zT?osB0*rQ4<xpM*DXo++0)v+J2_$&XX^rbVMmy57l9wmS^%E)1)G9u<R4dCoqsCTa
z_Tt&gS$YkrSiGzltEg|aXgNI_Ih|7>73ql0<psN-no>|uH3`wZnaZ^_%=3h`!bB@M
z;ax9=aRsr+;H!)bxV0+ECMM%j5+!7e^=NMdj%<XFrK%+B;(VsY1>1R1^pbGmw1S#f
zYG}3y(>EBhtk7g{EJ|I<l4@2pVU*@2-Cj^_IFe+maTcQy)g($MqH`W?Jykc3T~7ah
zHAz3@+V4jCVQ8MTNLeOaO;~5>?I`}RuBU404ZThiF%;>_G}2mPlw#}$`p%<_;ra0c
zL2ItHQk`S49`7=?7O4U>Qttif{WmOWqmn3E==aWhdOuK;gvCONtZ|ccqLO0|x&?fS
zq=spgdB&KA!2xBR1Se{4idt=m6tD)o)%4!69|op%V_9x2PFAeecjRp)#}Vy3<1nz)
z$}-)kVZx|_=`0x`1lOwRdr8HUiv&1~#aJglFBPl@+mX=B*Bj^Ob6cBQ@j$m=lQxE$
zrPSRzgVm)aX$q;OPrFV+cqNCXu2-Bfblzf}h9CpKR<*4hn@SZUDHe=X=pr^<qwA}$
zHU&u|=N#EWkdVwKt8=FnCH?Si#NB6~b2#i9%TKqIw|f1Kt7f))>62d9jgOCKzWMGW
z-#%aYc%7N2jcHm4Nub50KrESD3t1<M74%NdS=4UC=s%8wtS8rvoHH>5l4^<SoH0oE
za7kG#QYBGhYPE)1+Euh-+crjL>70H=kbHRhNbfAilf1{qog`dy$xOF}$HynGPfvu9
zajqxEz;5hu&dLFlLciOiwPNUdrfH^DIb3(%GxVNqo^Vz?m8!za-_B}jsHwGZ*zefy
zJNnMzl;Y$08EYzc$DWWBg@g`;vSn&gSXFuZu%qbAr`y(oG>RMoMQfz&+|l)e4B-`w
zvV1e$_|^FX|K1jU@#CNI)4%u;r{441_aE7&nTPwgT$T-^GKYr~wSyEk^0r~B96~}1
zDBeHZ;m3j6IjmFk-M}sumZv{%T>F!TI@|SR+15xigEP2Tsb~U;+m`q+ub?!ycJrHd
ziZ;$j7{NQ~kd2wgr>hKpIdOmJu}abP9d~aY=!c$sc_Q3aQY_538+i-3p%=4xgIhZ9
z*zHH=+rstvMoJm$<fM~hk}g(bP-Nm3nU~059q&H-jBY<LUuQxLXf4%SpqZA1^L1uk
zHx9>p9`4^tf+hq~*jScXN~2R{*zHkDalJm%k00oV9_KrTamU@=J@ah%`2G*9VWI0S
zhr^zsAGw{cy#MC6JU#voXjNI3iS>46iyPKjx^du(FFr#nc$%&(+sqnPo}NE2U2Ysc
z`ziOw&*}5Z;q)M<?;ODx-hA;Be)R53;>Yj!czwj=fgka};5x>`z~~K!{fVdxmu*JX
z%;oZtr%xX!F;HVi-aw_wmZc?R7&@HSq7LMWc0!9+8cbF6#!`)(Y{M2BH28;IFR80S
z5)-jDvT71OhM-K4CIVWLQmY~_4kf-A2?i{9G9!wqYGOk=l(-SLjWC~y>nsm1YsDa|
z<Uv{~BH_1HP|gWeD+R=0UJG73()APh`g_XtJt{^4N%tMrx>v|=X`C!cD!S&)i?S+K
zqJDVq=yyA&^Lt84biKv20j4T>U^^?SvsRK`)p?_qL`WM7si&){sS{efd#{9YtBG3#
z3U1iphCMMRO3c{Kz2y4Ht;yoWhp4LjZ&+ibN-1nDd2Hn{W(>KCN3NDYTF;msl<s6m
z)(TY{`QGT)^}VdZv+&xq!TXo`CG~2!&#AR}w4}3G?4<Q0BM<aHmbU({UHf4q$BZhX
zHX9``eX`&l@RnidNoX#&8&c?2$qM53;ydlRd-ImDA2?s0nWihzS!(yvvLdOv-a2Yn
zQRD`deHru{_*E&q9~k<96dF>dDPJ)MbW39j62(bE@4tf8YOPy(wor{}eX^pAT8-?Q
zi>z^}g}&=JDU0(PUW+1Rb?f8}E9>>l{^2u*u1A$jT5mW5!psvMXGYTxdw~R(1i7HC
zq3gTWe38gn%<Hb(u`Da|EGsr+wa`(tX7HZT7?ycry-m^%XS5s;8lTX1J<4dpwqQt<
zxMEdq#1@6OPAbE!9PXU9c-QKwwkWIYqOes!;UMaeZtKaM6D3AWC*cFv8YDuDa_Idt
zSnCdl9s9%BLO?J5uC&3sZAG+Zi-qU&m9M`3j(_;4-|@|dGuH%5im3cw@f?jOrX+{N
z#$eT1669Sc9<wEE)Jkhe5q_Xj>Q&LqDT_b1DB)Y?*411KwMyFcbQF){beoAP@aD_6
z9QHe=+r)Icu^T(yKHM>QQA|@xZ0ptzj#aj2t=B!Ijl)R>I;eDgk9P-r--~j1yD={t
z({1Mb^vs*PJC<pp*2I3-qpRle;}gTs%kxiF!WOX3wBU(m?0cRbKeDADVUcRMELUQR
zLP0>nMT$yFEBnDym0^yBB}kw~62lSSH^!)zm34`QYt8)i?|#qGb^P*+$kFwDc{*~w
zT)2dd^YzTbn>Xxt1E<g4Qk@lVYO_FSC9KQ7>lucDtOdS2>K(gbU>6d%y!=u3)Ssr*
zg-T~Fp=Rbw!dr({h7TbTHdz7XTnH@zdKffGDU|al=fr?)4a*vMzAWSlxyXRn4~ouM
zjQ3zQ)hK-DB@I{uX<1mOnZsdV+>h*!2T>2&X(n#*wIV7-G3}RH=sJs48mBE=TG&$J
zIPSoA1mtw7wLx2ja-up++s64S5Yki=x9Ln?D$Z$0(mnU07>s>SQ3V?eyWKsfySEIx
zRaPoil%C~wW122hE0D#Ur4uns7uI#+v_D`JT%W%sMa48t9N)fS-}Rv2+sAJ?&w*`Q
zWyKi_%XMbjR@~V0lQ%!*-Mi0N<4VqzyW@#B_b1kx_|Jdy{~_i~uZ?hZrYp8qf~x$*
z&;N>l`&a)Pl7>&e`3)rmoKlof$u?7UkzUAHsJ(!O1s3hlrcw}pK_10$Ja!(mBE`)0
zvi@mLxBrzZrNp{k8`sS;bYp86sEv8n_;SE2B(43k(h6lgB?np8){Oxz4IB-0nYNiQ
zPsDAZX8A&wA}9(aokXvN+922j-O8-vx==#Erh;}IrbJW~fOIVZT`OrT4Eq=Ch<H3x
z5$|WMNNQIVa@rbAQIm?nbpv)7sfY(kJE0&NEy{<|nvypZh7tpDo9LVt3`O(2C~ZhZ
z>=CuT822+Fh`OC~MB5JIQu8heuv5P#6hWncvX&fGdl(l|ZOPf3DPhAHsnfEq!&*y<
zK|tFSQZSg_fbLr!R1_jrGE`APgsEv|71j=FrHV4E>I(`*hGNqcj8a-tOqMnRYr94v
z{NYzkQ4WFaup;Y3MKPJBB}3&wo+(kl=uRsrg2zY&j49mTKcGtGB1A<A=4kmGQ_jL^
z^IkkPS<<$uC0c7sJ8EU<`@NXh8|zQ0msD|v*6-?!mA=@PSbj+vzt|*D8m%Q|ZJk4F
zsZ=UCVQaw{@riXt(H%yPr#rl^*u2U8PysD3VcRSM&PiAd3YY71Ywa1>@9%N8BS#5J
zdF$HYUb=8A4!1vu?V%~^yRoB`ftZt^Pn=`FKY@|Kz9#wCos+7(xUJa1;r91f=Skbl
zx`|?_ouL{n4HsJDjl;S`twyK<T_yB#KT$#3o_^;safVP(N)CxeX>u-Dr#QZO%l_@>
zIOo{bl|(`HgVeD7&`TR~^P4K<yDcbr8E#WiOt+Qa{NV$?`t=|9?&BvOpC{%mvxT*-
zRKM>&mZ4CT+S<(guY+Bde@#}}ign$@YiC{a4VQ+pkpU*eM9C3{ZqF(iG;P}mZ0m|C
ziFfY~yuBaV<|gsxe$Tsy2TG0zjKB3%D-i{#k!{^bF>yGIuj-^$8t0m?&^olWgl*$;
zxp2K)xZJL21M4zDEsR5lN-LL7XRI}JP6VVF6yDo*&nt$Z69!;TjDy%YK0aRMd!yw$
zplYJl!nCa5IuwRpD|9Er?=1+^Fs4AtnZE3BotN!NEtsr{O7S=EA36Bn^UJrN@o#_m
zOMddxpYr#=`3+yc|BjkBOcuztO`2F&N$}T7$`URbjz_e!WGxLq(jzUFv8w+#_UoTC
z)ZLyZoU;_GSaZSZjj@sObBNsa7H1T<kfq_kJE}8M>C$&(kXn~D@ZsZy6cg52y0J$a
z&2cXi>c`8KI~r(r5P#7$Eo|$?{?OC!I{L9^3yF0L%+tiv)0t&iIPLeOoS_zuhn>i>
zso^k`qSJ=EaffqWP%P3vCCsemj|m~5t5DKEJ$+<KvfA1oc6jF)JCAdN03a>&t|OHL
zhr=5V$9MR!QR2#NdgSu-p6Pnw@^q!&?I<NN&u3zsSwiA`dBkbm6tcu}UHJB^Ut>qh
z!|8ywhV%8po81G)ZpYVmZ_rwE`}Bbi@4w;Amp`T(M}G37&jH1!%Z0-@F!r9)c;NH9
zFZk`ZzvcH|{f@A$eEjAc4!a%7cl_eZA2T@5cfa^|{PW-aXV&KnF&6Y%C~3o32q{uj
zC8mV;@?zeWAgGuWKr42+7r8$tytU1z^TWG1l~PSzkbkbVV2l;WaPy`KQ%RGBB_U<-
z^`%P6Q(MQRQwoPBZ<Q@XwvY(hLRfCil%GWDz{tS!(!E**ew*6>p|rvEJ6u1C!QVPO
zP?KWIsr|W%p%Zn@8%<Ua!h-916r!%xoMm9Dl34Y<d}(VoQU>3TU>sF@QEg1aeJDf$
zsgbg+s1k_Fgr!H-!n!W_u9swMYCgo45l<<QQ+hp_OIKoRK5?$q|1~s*)}*lF@B;re
zFDo1gFNj#0QCgs<%d{~xgL_E{WeT<|)M$k(hl3)6vu?^zE%~)^pp-24Q<aB%Eg9Lo
ze-D(FFlT8ATBxO&#%m_0AAX1cS-&}3aas%V$h0JJkaiWS6m1hTx@Pe}sf^VIrzNP+
zb-ry00#+-=p=Z^aP0^Yk<USHIy?h>8DT<QOUXro0@>$m_)o7enIAbM1fZXGzd9bW*
z{iQEofKp#8#Y#<Lx^J9B?;Os%SEV6s8!>0RYpoPUDwxL3QG-K;S%$&RW1J%<u{>Ds
z*q=^7C6*%I(c=UC?nDmK=k1+i-0j5rqSR~Gve9ynEVm0PZ*<lWJCDxd1@4Ct>nth8
zR}Na|J=QyNHYg*A6s%`W8|!q1lI3%!$SMg@_PuwwEl`TsGqh4N#3~tfP!{V4jLJ}!
z_HwUit1!(vwA<g|tdmVmj_i6bb%6PY`-Yckl7pH;jy*~&4y9_=ta$%4^PAtl=byiN
z&*S+<Od?o>hJH~sLlR-bSc8hvqVvM|w9d8rwmmZ&ht68Lmt>&G+-^5g5|(9WMF@$t
zP;+68i`X@&&7)^2Sy9!(+lPTKKReP}jq$x8Wi%}7f~tkCGm^+|_L%4AGw0_kAq0kD
zWLq~v+~}Rfpqk&<qOIn3IWu3c+%8w1A0M&K;hjd4INcpt)(L8Eo$s2?%c0Uag?F}X
zmn;UjK0i@XK~+r(iNRY`&X``)8_#LCBYnE@9152$u<tsKV~^=9`(5Gd4_Bs;QBHu+
zX}zJ6rTV=Ls!D<KJk2Zr>HP;DmzDo(-}BR7{G87}`;3p@{+{b~;x`}O^X|B(lqd(~
zlqpft;AvX8``Me;C{m~~lea*Lg2D3cze7;se7n(khZ!9qD?%#dRd`54sl0uAq*}v=
zr<vOpF-CK+J>F<BK^9S8a>%3*xy}nEgL<Ag-!?w~{0`kYPNjk}4BbEu^00}i!46TZ
zAyMa*S}NNXn6C?$=c}y6YBSf;3d9==Nn1JpY*E&|VcS@n$F4S0uEmgJriR?`3NTFz
zSj%a+!x>B0dB&ai1a8aB7K8Beilp=g-?4_3+kEA6{zTkvq!37<VD_4>vs9x<i$jIP
z<9EL&PEUOL^c|_nie<e$@$tjAJUqOS^<AlW<C-Tzle2Vb`e9(VKXN!8={id);0Mpc
zQM?FASG=)ww#Qk|>2ShzCyXEY@{6C~onuLnl;J0z|BP?KZ<uZiu2@tq5)4ooRArh1
zrFt}m<+jPNpjx$#R&?GlQCJaR=x6Wtf6A@(zkYN2j(!{^xKLB;hzz`xmTPTf<<N$&
z*K}o(q3*@QR*HBmpwOw>)}t?8EEyb2P7?4a;>}S>Sc8p!B57-5sAQBG(8ghH#W^tR
z4P!D|7m6`##$jBC>-yG7TB#wD!v;pG=Fpa;I*fHvTNfj~8>HLNSZWoxu(6(!3)Vmi
zK_2)qQIjyBhRz~T?yV`5F9z>6D8zuaPH+Yiw|u44FL}mwBgKHOFTJfYz)I?}6;x}3
zp<<QzK||bdMoTy))>k9`wytPp@xBwq&P1HjV0#%bvfvR?t<)4Kwe{hu+LE=K3?|w%
z2(F+TYD+Il^`5MRjrBM!KHZWN<<GG02Y9oy^qs>ROV>PMF*<JT->*fy!=!}Miq2X<
zn4Lo}7^G4Pmk(bzSwk^&j#@Qq!&I0CpN=_Vs(21xst+~o{Xiv!mCN%-`mX2H!cH%T
zCPN!qoWtsuB(0>Eb4eK7>oc%6Sgh14O2Y+8d~Tc8q@0PpAbW1pBFs8_LhFhfdUm6D
z31d!j2-2EbdTiG*jC;mYLdS%*9XSQo^+H&$l<eq-0qX{gl~tJ1T6#poM2R<Q6!p_P
zhqju&A92o$hj6<=TF_NY@MKvPcYQB@(G)4`MB0{DP&8G)4l$xoDhU|0^!mC5wyiNt
z^TK>PbKDtB1Dv25$Ipo8m|Dr(%&MeaVe|&8J5*a!{?IGjK2a9w*sV&Vj3(o_t(9*-
zUihcqz30<qVF{6#@+)k3=seXnFmnDLK8?~CbmQ>FmOz&wPAMgSCnKwQV+=WGLRhIq
zwk|OSOwQC0WiwPVPKm$~W5O9LREVs2_wLB&4+HyNa57nGj5CzD5kqV!6v09*^ZeSY
zthK>+9k=Vuwk{YAb67E|a=O2RQn@^Rk}$>76EP%8*toQ{Zg&__BywIQF`s3nKHuh7
zML%o-)mg@FWWHXB>n3V?t9BEQVC4QVQnTZEPWZfW^qQ0uy@8!BbULx$durPAyT^&O
z1?FNb5{?Scb)qURC6U!Yy#;<1p2++?|0h-XH=q58FF*U7`nGa<`-acH{F0(HRO!(z
zF_XhaiL&-y=S807F~YWzw?g0dbnd@{QfB~NLASEolB$>#Q{v5j<hVZ&YT}kMA8!jG
zCGNIJKSstwzziLsRh&30C(LQxgz=>+eQ){2Pd?*r7_n&aX2ry|iMnU3CD+O{ZET@5
zW47%1Lrmlt+3iP46*Vc;&`!=h#u(OhV_72~pD)}$Kd}L!3e&El<T1EK=4oMGHV&tO
zaW_gTrS~{*B~fJ+p(LI!S32L*+Y#MN;cMJ@Jb%ZxU;m4w!b*}sK*>6}A3BNxjt4?W
zeEjyeJQ<62JC28U<X9j_oHpnzY@xCSzWVMDTyJNtj~{VXGal}^yL-!zzx*+8PRBOb
z6i&lP-x*RWtjo%GpWahy<%^&GlrK*Y{MEnz4}_e#9}nbOc>nawAHM#Y{dmXWaL@H}
zWgG@_mEc5-0+uXPbZwFMQn_8uSf?>o5thidCTc1eWvIpcSsr*Y#J8%js@lp(R(Dk)
zFLz_n)?wX9p^_RTReBi&(;@B5!dj|XK?>9&Qid^_6ceGTXC)_$wRq=+on|cBS!yn%
z93*UDI<&SJYj8E7(v5WcL|M<$<Jotz(zHej(Urz{FV#{nYFCAvnqQ47KVHgMp9M7B
zLJ}3#)*h9dR*cbvbw<~WP8&&;Rud%tH2FN_VQrNpV~w$d81Ysh$6{rZ!ERe__}xJs
z{z_A;d<H3GNtl<?l9CdL(Z)bZgqSf(m{{KSSl3Cqb_?Vpe94q1=F5Bfq`)4Kh)Do2
zLC?NH*Fp+2s%VklN|eu`)P{->rApO;wk?3ud`>Sf1zE#N5-;WQXB+DF-f-LxbPerL
zLMChmYZQK6DW$N4_VXeJXRUk2!+<f4%hNO0<;FT+FxKK!kFlD;8mb$}#uC?+6c$Vt
zoP^ct)%%m%U?Bitg=$`&W-@3iD)BYGysRay+<1U(^(rZC`JKM>nd%nO0$$f)G?Y!L
znvyEnXHpS=n)9M``L0K4Nlw?C#9zG4#CAwg+LB3xu*ko@-Y{*=H?PlhLq~6hrj!)Y
zI*~TPl2DTb4qB^3Erl31bjgq+PC?Z`j>Ijsu!XF|DG8Nx8){X{;0Rrz#tg{kq8r{P
zgdl#gBGr_+ByQ(N&R%is6`m@5MH+b8YPPVE*BRgKWze&x0nydVujB_^+wDrUmL+Mf
zp>m!hmnHM@>B{-GvTkB$%B7(63tFgBviQcf5Qr(!mqOR~<Q%D0D&S<0Hi8Dqg{)eQ
zrgoAvkFf<bM72!NLMee711U*JN^nCj3_Tdfj-UVZGwyp!Zz`SdQO;6R;d;3ew}sA1
zRj2P9m-921%Z0<?K-YVWGCV$AxlS|QT5^`etM5Fj!Z%<2fm}8j;H_o7P1Ky}oW-ch
zI!zdr@ZJepMQN6Kr8GZ!ObM+NyWNguTBs=#=1o-Qwzsgx&^t>l^50is<ICfSxkY+C
z5OU_r!@&J%<Z)imr;aT}K3!KzR-~$05`88ojqg0R8=!Wib>T5P;?srrxBrDdeDz!2
zy?evmu;c#STXdFw?ou=%2y3w$N3LrmtQ)$ODwK7hZYvtYmJ{iZ4EX;^L!G=g6gpLW
z5-yiEiY;Ugy{Gez>yijsqnzcsMb>;p&oigE=jT6q%f7dKTC$`x@!}Dw3an<daJTE}
zot5s@ZDpBeo}aH=FAK+^C!|81W{j0o92BYkNSTxiYfMtPqqUe}OH&#&Pzn!+fgis;
zaT<C)uCi7EsfXE0Wx7pVZ!_0*<?ihs=LH^%0pCe=iMEDqUb)?7%-#_~#B~R9Sb2Q<
zn%nbxRBBBZu~M@HI)?q8ielLU>$-BiUZF_B?f%V(HKMx2nDN@MEelh)@$usy*p`KD
zx{;;{r3}luvc*jXl_Xw9X+@9{?kyx9pC9@4Z-32CfBp;Jo$henQ5DSVO5b-3&arR7
z6QvApJkTHRc{m<;I)7wfPg)Z5WuYHB_QxYu4b*ZYuaXYaR*~5x|2t}Y=cL_W`Qg2Y
zF$Q9gVMV{Zn3Q@|Un_`LtQ2agC?o2en0jAY30ils(rBx(O5t>*hD1?vKZCpsBq=Kx
z1X+_MN?55|3y9Q;wjC)O?tF(cmN;+JAU+>m3W+2`W?O9-V=2wUq$~0gY7gmHGOCrN
zd*{hD$^e$4c)Gk6)h-oMSgAQtGUTwx$QmQLR&3u%&uvUH^t?cDt6*MC5k8{UhGJfZ
z1F13F*7jYhl~@bTXj0rLB@0)oSrY7v>QfrGDko}V7Uo*e+RF>nng*6NROK)x6Q*x5
z-3<1G_D888vjy`~-IfB<7@^2Y)8tpG;+2afKeLivR3(Z;$)WMipu|7V`n7YKv3Crk
zlz3+^X;EpJh+Iv5^&OThxTL-x=->T>vPG8bEZ-*zrYdq%AjV;2f4t|K5|?#rYwy&0
zab@`EdwI5qz$c!tTxvU)h)P-2OB+N}k(<wnR>@^v{8e)6O8`tsFi5<Mkeg@GT0?QJ
z`2oRoJ@-nmB-ScamEeYic!E4@D3n^|o-BcCEV)8RL6X*K<Ap^i{j)GlH)0IJVk1Lb
zDPbXnDCntJs8*iYH2#}XGFV}?+{d*DKFB!<(^!Yq6;fn-U1@7v-(j336R@_e{+h?k
zv=wFNoE%oS$nAP&>^*z86UCxrNzZC&qzI+rtl@OJXXqV<#Ap@!&XVmP`Ih}}T(_W@
zZkhL&g-^H0d5(l!Tib;plq?lqS(NQs<WQzi#8k*dc#kO+Tmk1g#KWmk6<KeFfbTqm
z9wf}tP%+d>Z9sA*CK7Vi)GCJ-Y9S{{%V#B95a%rW{f=#&_;}km{OBDi7pB`ye5NU}
z8wUpO$tm*j!zXGj{NyJ;X~}E->gD_F^DnqOKeKHMN*A7=&Rm`^)U=V}f}*m_E4S+;
z+}K)KjpF{aldVaSDnV@wVUr&2%lXWkH*c}l2<Ps0!zm#J?E0>OjZ+H;Iclxkt~bV^
zLmN-YvshZJWp80LV2^t`Ul=;ecelh>A1^#!W@4#?R`0g15nHk_Tlbo*Jl}mhbGbb6
zheylj-+j%SVdTp<cl=-OkEGdCjRcu;l1)oW5wmSzEpb`JlC^HAswgRd{qK-|Ed^4k
zs93Ptu<I<Y>v#%L{4c6<URJP%{<x!rNZK~Smgqu8S;Np<j8oK*ab7cIhj;R@8Jwlo
z!m_Qx=UNkC-k8rfV$9T?LOF}`;w#y<O*mCfvl~3MdYlyq=+Je9sz}>PHIQ0j@y-2_
zzx<1zFpdM~k58!3l8QziC?d;edZXD5(x+M!f;1R<y1}EhWnEXSmJ@ubkvw0RZ_lKf
zsX4RXAD9=zv}VYf(RXx>2laHmGGDJ;ZdVT70JU(wJmHPz`EtQp!};>W;r<QFbY;3e
zv)pd-%xSX!hfg2*>;LjUV@E@_(nj!Uo>=CY_01b{Oq?H|*!5Cuol@c{Y<&FoTgKxF
z=PZ5icsL!ndw7eQ*OqjIcl#4-iac12%a-}PkvH&aptBuQy0A?f#%i)DgrEp(VRv#2
zhf3!aHU02oY$^HGA5-CVHKL4d|Gp?*wNy&jP`3GqN^W=@Q44ePWJ&!_p|iTJTQt@6
zxSXJa*w7mFCCq@VbYBKUV{B7D1R5E_Mu{7)I`nX&Y|p47Da~ADMXt4K{5ViXk;B|p
zi;#eB+>=sj>^M!#Nmij+z6)AwQFfE8GeSz#90hUWJHdKr2Ts<m##$L5l|og8_hJ&(
zP1%yqQK7ZRcbXUj+p^GOs7ek5s$^;wW}LRZHBiVi<)vy&Nc6F-hckr?%A&NE&e@zW
zT3*62%}}neR=VMi)gMWrgS?{B`r=uaK>&@mg1^Y65ZASNDdj#Wt*l;EQ5ABCRI)tu
z{tQ>wd3@(yJx4~H7TiekyVP<I#+-;LqLpQMIAFbFnI=*cg(1~~t)dF(syK`X=mxIo
zB0j**;Zu>KdZjQ{<DF+1`exLZfJiPc{ltY*vm^q&kOJgGXomlnL(fa<RlT4wN>jDP
z$ET_!i(k4utu|C=**lB2J=XQF3{X*QH*$)sdLnEKdCAf{s<l+fX(L^&#$vUhsQOwZ
zmeHd6g_)4WH|11Cw^eA|A}JS)8`&S<P*Ic^L035dDWJ*<RJ)O3w?`uiG~VL7PQnVc
z5@W&-J+2>QwWb5sC^{#rbfp5?3f5v&vT{y2y|%?@YcUF{ip|$(=zL|3fx1ofPH->I
zBzjXh3_UgfaCP`^zP@q|l_gX{lA-M-`3!BCPo+Wm#U_;7b3t28??!wnvR<vth5{_}
z+*pa4u5Av!@0k{>#cmuVWEd0kw8--{2V4vG<Smn8q8NqK$jaVjwrxdWxgQ+bDRPMn
zeNWaAJM@CZxm|hu_zCYlT_-DB!PR8E?-)jy=Z$q+nWlwPXGv&M*a*uNt24{8u`U}R
zDz>v0tt!KApw>#<HoD%kPOJPrw-r##%R+BGx94Yuv1eJuE0}UbW!-#vl^iqM7TFDh
zfSkk3e()_I<G8=y$-IF@>q0+v6tz)v<uL4cS|U%^t27YhN?bQC`I&WD*_}>QrMaw`
z=V|4i)0xpK{==8=_>TqB8qr1|&v}iqC5(ZvOe`@9Jwj{p8qj2XfBd5p{(sU?x24Qh
z<>caw#wJTO!m!zQ9Ybf?sOY|Duts{;9?yhnW?KW#&o|c95Y|8~^1wUnI_?f5%N9|(
zGA$FvNNUMciJsq{Crp9Ee&l#KV7+Hv7S=7$cO6nmC+LZ=1_^N})1-JSD=A~})-v=x
z{V-6qBzpSJF>S)u(N;^njMn(x^X7gp0S_zxtet+9t7RFj8t68&-X=c8L{^Gn967#y
zpc{6iEuw5wm~tVkqJn6nu@<~8q_7c}NQWenoO8q+NpZt!i&2uE?E4<4WW5&FjqiT*
zEB>j^c<XrR_uQ^$e*4Yu`SAX0O3wWH*Z;)M2p3BNbpl#m*M%)cjMnUW&u5=~jt_-G
z<via|&NHt8yW4ZRe}~5K{?mKn8ga&PeZH~Hk#Q$-xiX3vGt;G_GN@6IH9z#KIRW3B
zLtl)(>Lr4nv#ckz0aavJDoxo@7R~qmu@{4|s+Hat_Rg>r$O@`4lwwhxe*MpSCxcvV
zYF93WoGacND4FZ|5$8L6KY;4d2J&{HEKkxQXiTgBYHKm0$y=6n%eIJ*vx>5aYA6cV
z3AnaIq%TEQle%Qut5SfcwZ6pAFs*CVS&MP9asnBCVoJ?RBPQjX5)~RhOb{a0I)S@Z
zNnqx<H6C5Vx0vrk6{)mTD#%LNg0T{0GAN1$(;{*?%d%DF9_Q{zX(NS?EYBHx_e%;+
zlp_o!HAS@)=_|H|f}{d-MQY)IoW=BKt-R<<$$~qnf&}?<TrbaxCIch1ytc+`2iKG$
zF-E8`obK`CfiMf}u+)rq9#dtIc3Sc9_A~Au-to9f?VYol&J4Ju(OH0`5_FJz+qPk{
zR=Ediks!uyccLG5GURE4Q4(NL&8zm3d^Os9k20y5nhVNE-z8cGmY3wXnBO%RhxI+a
z8!*l{ihy`HbCj2@R+^F_QE<MOI<mGMkwz3tO|<N$u+rkTp>$;5N%w2kFABR77M*r<
zyA#fL<QSml7Jh+(Y+2mc<GN7>xELs5L#u-7_Bc1t?FuRcVSpm_f+`2Bg3ijp+iC;6
z485v><3$xS+JZ9T5muIlCT@Y1!utFz>+@F}hMzXI*wEXT8m}LAefzj!{UE`gDr%E4
zrq!ProHGqE(u)EL#GGE+T1rc-%I{JRw51B5I>snUfANG6h%u6qplbTQmj)&TF=AWO
z1ND1VSn35;vucQk+M0pRYC5ZV^Kg&RhM2M-&JH8D>xFOLzZVpY@6c9byk(0*qj1iV
zmcaFL;k)lXv8@|MX$tV+<0rOtWj}VTTjFwAAQw9C*zI}%y56Fc_~>2|)F~Be)fBXV
ztUsMG#^H?EHDdGkZ_AC(?oQa<$n*8ex~+`Ej@p!7>l}JpIqo}@sc2RnPH!3aC%%2W
z^8OE>*lEST|M3IA`sNcMLUM*{a;%NekWwTR8Ni4AfxhoZNwKQL;})3Bj_L-if^?h6
zHB)pYR{>~+X-2sWT44>%N<>3c{c-o(pET6X+XhP04;@u2u4`mjBYUlQdpw}6=a%GU
z(s@TwkXi{hxo|$;IG@Guv>O~sOAqYBY2>yh2|zRtPt4GJL8i>}#u9^g=-Lf<f7sD`
zLr96+ZQ(L6^qt4nMA<UhNZGtb`W#;dx^Wm#*0Ck=hFBv}cTr{xxl}?b3{9ES)}oxI
z>pa$2Y6B!2{h~@F0+r|IM~s>9!$3C<jN<`HVL6_84{bGAfx<ZS-0yq38o7LYW(@_U
zJ*(^4mX#QVP+w{xmW@(`be?M_RC$163{Gjb>y6+3%inW4+)+a2o8SL4AC!<Z|K(S|
z;<!6vP6z&X<NwHh?6Joqby=C_g?WpF7#NJj_gWsRYh>OoTs}QftKyeG|0VlzM+ghc
zywLfcbxGs|x{9*YDZzb|0Ex_&D>?jdZdi#MDM<T)Qtf3>qyp%LzxL80YMqDLf-fj}
z0ca!4wj#KIn&iu3wC2!RRLIO><E7+Y>gSxS0dtmmCiSv<O>Lc?NHJr5Co3)M(7p$)
z81BC$8%<m;0^N<$GutQ>a_^{0i5s~>teUtk*nSi>5GCt@HIZ|{l|oHHfXWJT5w(z)
z!fz|SubMO3KtTpP?VJokwUDXO`B`MWnNkvbik2d7V=dZwveu|v06EcGB|Wt@N+Oji
z9x}mdNZ+mWM9Q$Mv?PX8X)xTP+2SUOqjKb2P!!_zYih3K?(-(XR%+GFhSC0hu0t7*
zvkFr)+jL=D7f@ob)`-unl>Ex(Qa?OwEmwmW=eE`bs3lUXc=u8kHM3GVz5Rmm?labT
zB2HIISaIH>ER<}~wQx8dI5idOC8Q<bZ%uC$DruZHq7H8>R#glVpb-p;tnr6o$1sf4
zTuDhdV}7@jgN1i;&}oWtTg)~ustn6jL1_&EN>y~>_0XcK40l?Cw&De)6lx0|#7*i2
zQ|n<i+F^W8td(h7d7gx?>5P$vkeuabQd+`RF|gHw>m`kD#HfYpcNo(FN;r`*NLRBl
zSPcnHwI1g?`7F{(4J*1v>LMMuw3RGMhVQbi=px~Z+KTEs=a@{ys@l*&6=O7g=OIdu
zbj(FsDYT)cD5-12iyPC-a`}#Ye?;&1*v7Q0<%g$V*HAGfp;SR3>$_Zdg}3)z2R-ee
zBno**QMObq6qTZVMUm86sTt&ZCBq#`yqIlUFBp@~zoKolmPByeq_Qu^Knd&X;7F|)
zXYjt`i+2yaz1!oQY_d?Yl~P*s{QS(iEKmwt2)uc?=dkNZX%%%$DJasJp0+^e9nM;6
z&dj$9AttQrS!3lk1-1aawRGN#^@NJ|;&0aqoUxprt^y`Tku!vtB{{q%#?diti4qGs
zS31|xbv?_x5@I2y$`)n#tyL3(gieNG09ELn`0iAdj`6qmM}GO!pK!Ze=)6Jw;fa_6
zMOD-$c9E{{QKhnND|0ER<3V_K2F%d$=ItAX-lI!^l<>}yR3*+UAqEPd`%VVe5YSc=
zmkpNn)rRt~U4OEt?&YCgN@a-w2aKIm9bK1|$LA}TCDJV$%4%w3QE8*-JCC!Pm@>v&
z(iYfaWQu`=AW-(M!&c2;EKbWnrrV!qti}(Hao6LNW{rVqTRBfNWAAzQa3XJ+pf^!w
ztoSHiA9@tfkSZjl<mHz|+O}2^Yv%d7Fs*@ozsGus@oS^;+M$agg+$jmyfLT>^L1uX
z!m`=#nu@k<+-?`fVL$>@Nd^-wzj$k4zw219p6`~8n4s$&>$WgWH;i|})hUT>o5^`Y
zo654yJUu;8O60Wb=)C}zFW>#1fBc(DQHJNI5A3@ET@*UZJf5HEAMQ{_Gg!<1c)(eQ
z1)iR-eDmoe|NNVO<|l7IXIUmbeD@8eBuWYlUB_W}B88PLW#(n!?cKeg19B#&N-P;a
znr7+-HesUW^@l(36p(9`|6K|xXL3r?^=GA7BF8`qfe_ZF%q1BJlt&oE8m+|_^y0&#
zRywT&QkjDwWvXgP&(@jg1XWRMk#&%kfghwssAPG`bye7Q#*dKiaA6WJL``TUOi;B{
zTv@G9!;Y}5#1Mt{UotuuN=#^1P}bphN8)ys?@e3f)%G&d#xx^5XzL{eqGjD%OH=+z
zrYeo=26Blq_-Ne{UlmIGT(#7EDd#9f4Cfgcl5$&1g%G|En;`XO&Nz&bK~1`ATLqtz
z!gy;1IZ>0;dsPG0c+#5jN->?k#qKJ>?x=Ra>sI`(D!!#8OA*yIMM?dI5b)lM{Xl7y
zYOqx;HF4Jec^V~3mQ~*Ckb`QfpmeV00#!T4!$^O)BW;oGc457osalHUbx|l&Iov(a
z4?9Xu+@2n}&a<?DC{W(gJHvi=kI{<xc4J#7;<n-y?Rg;w4BHR%{Qx-=!%B{kNXDtk
zVYh1#SMf8+%e*$g>-TBSk_)v)foNl!o-6*qf()*uwRWi5w6<m?$uSaJosrfC<E6@O
z+g7G^VTl_lZiH3(dcEuET`wV<AcgBXCv+8ayBvre)i^Se&^9IE)l2Y1tP9pk?OLss
z5*DcdtI{3=kk*SFM70;NbqSOb<vwqr5og8TBT3-cJZQPCvgG&rf|jw)3uOAm$BS6{
zVir}{Xc&jC8TM(=+8=gJ*Ryo|Ht?@9hKj*8d_`xQ6~H-)6W?U7W&NmIdUab@E|*IS
zl9dL<9qIdydD)sTRiMpA8({~w!$N5Xv$#oMF@->lo7DT|`g*7}T5)%G<hUQ%br#=Q
z)^)=69hb{Z2E^8ylv3dH&);E;Vw$cPqd1H`VO?356|E|V!$6@jO%o{x{2)EpF%+)1
z8&8)Rfc@@BOdC#X4u>7X;OTl#NtwsTXP!Pji{`8<%4qh7BbVEZ(wSJ`)AJ3dEc;He
zKlMDF&v-wutO=zp^BRaLld^0##$K>f&U^OzJ=QyP=lJ5spY!v6$2PCbapsHDz=z9~
z4;M@CDrr#6p%6+_^;2cpBJ1_a@pxj~3Do>J^cbU1##7ZyP4P9o?>b9ZBg!^k_q@^#
zJx)vbN8slF`t>Ibb=o_Dz>ZRyyXB2iVL$YOz1XCGE-e9F3Yg}RD1|L1s#5Gud+FdT
znRQ+<sc@Y(v^G+ryCrI_oQ4i>4Jk(9-5AYo^mJY`g_ZY@Pkem3;*BDd%)`Thx1ZlL
zpBDnKZ3)wPFkl=~MHV(n9a!Sl%(a=26Z00>YeUV2+fBR(%4&>vGMG7s#S)fCL8^dK
zmc(t{5>6pOh*7GTQd~HnKN7+q=5?gzDb|9iQmIvyV$F&3yfyZnBh<_qBFD}%cuUEN
zWx26#3!Sr+6j<jOqYK_Ev{O<|q!aV)18YodYec1u6f(=YNHD-u=4IjA82I%+{w=@!
z@@Ir?Wtlsc%Z=Of6CbO@$iM#GuQ}{`?(PrVzj;F>;m47m{_Ge0^FRNCRP`9mvIasf
zbiJXA1((2NIW2D?GcSMK4fx->iZUoGG45J3sx<JfwrJnAPQ=PQT`04x;%$SlN>yU(
z8&xt8p$)2aW~ObE`Y`7ySwa>WMQjhiS}QbgufTkz3@#;Eh2~lyY=mXPqp4j_wma}A
zv|XtA));Pz+!A#muT%<$HxKv_Kw0U2EtMDp$F9STI~g2QZi!ZDO)%|6R%&fkV!~b&
zZlCy{w6@feL}5@`(z^(}RrAt;Da$34U=LE>C@G4%RHff_U1n@M)xXG9FXienMzoTj
z83b7Dylaag1x4%xSt+`%r&I;nVXTD$B^k!ivTctz>sh0wI^pl8EDu;0WKCRO)ys~Z
zktC++1t(KVB8FA`Xd>bN;QI|^Wf)?Ts-cF(5Y?(svY?i_VUNZUZ#U+7Cag<a%}Q{j
zs!D&j<MhSPNbAJ<{Ky)D1a=C9l323l^M^a`PWPm^5$8#U)gmF3^5SFkqArva*}^7j
zjur)~mc+VFbl!^byJU)jq6ExcRkJE6xo>E{Go_07?b@$Q*DCET%Z<2-w^o3*t&Jo`
za*`F5*3v^7LSUX3X;&z<tpP=0E-K-IBvaRArRFFdrb+@2%4C$vDCN<*Z@fK)wW46^
z_MNr92W<%JO3FdFfTo4skb_N0;rsiYimu=L(a`b?@B8l62OD!C=1fkR(hAzcwh=-i
z7T3Bt<-e|s!h691<xLJZ&WeqP+LHJ`@VebDfTXu83fm1>-(!sBc0IG)E_8j58%F84
z2GI3&&{GP$w;T>Tu`i@7_L}vjH@79-qrj(MFeb~gV2t=MRcW2Uing`NuS%{p!a(dg
z&*ggN(JOZPgi<|KS5^f>^K7e<==%=uc7hCBC#;5tyA#{Gl0)Dyj;wm+<Kq)0Z}h!o
zKX$}TP-nO6g=Gs=tw||!y<O>dmgDh2-}+&v+r)Ie5!Q%x4r?K-6|Fkfbz^rJ32R_W
znb3M#LrN%ZI8PJ#y3%(Y&Kl0=3q^@aci8XArO@||an~{II{N*<;o&XQyrOcYI{4zt
zFZkx;GovZI9XnL6DCJ0M;KOC*nl|yn6~RNLZDS1qRpk4qv=yd<F{E`RZkqtf6&UN7
z<_YgS+8Ndu=+NjQ2i-sB`jdt_-#MW@n?|UYm>^^Cv7KXjzLA@p`w~@_he2go0^fdo
zW>X5B!nb1F&RII=2q7^qE8jgoVLHR<Zs6#9XfR-<D(7{@pP!j?VV(n3ReUc{NZT9i
zU>Iu;Ay89684XU<v1MD?Qa~92a;7YviWq~SIkaNiBDMA08KdOysw6I-qbNloM*;7K
zNRA1w#Cs8=Bx#j4(y5s?<~d0myP*Y2sW=6__c-UJFg!%O_xR3>3BJmj^@X^<E-SZr
zrFwzDDkU$Ttt5&1t15{@mE|G3Nir|23(K<5_Z`nqANl&5-;vkI^zC;%dCzz6zh+(+
zK0SWqcsSAZJx>=&BJF#}Zhyzme)0?UhXd2~N}OkgaR8;5mRTO;sZcgWo{Ln3)rzVD
z_nrR?=~v|kYFkZ}iaeyXLHi!lm^~_clG4;xQDn74wH4}%i8?iayjB&AtQ&2q*iv!L
zLqd`StyYSd#Yd-_XXeGL6r+^R+vbss+sX`BKT0R7-(!yQ@2PMmqr?b1J>pd%XAN0P
ztY15{aiC=_W`-Tr_EdT*5$k$cg|vrj#(r<9Fs3ERTB7wO9jXm63p~|Y>EP5_lWQf{
z+=3StWi7hABpVemZ1m0%)=gSaN@<6e@39##pMx=mT$1ptj3LK_Y1o68t5!+n`3hCF
z;Q@v59^)(|M^T2DH@q?AD8t)Jn)HQuuaF^AwLv{aY3is&`X~!ZLLy4a&;J}(DZKXU
zmQ*P!zj`=zQ!V!Up6e2suL~t^f)_E4l!e1tYNFpC$VIU(8|$)&4@z5U56A^)E%*C7
z#(v-yHgpB2Evhw8l-AR!m8RPGit6dTG&6MG<GpxVw`CDeo)X_*RjU2|+8S1gs$F0H
zuNJ80bv^qMwh%i*k^8>1n!E}r#rC`x3{Wn)J=e>N_AO09QGRcwR;Fb}7kU1ZGtq!e
zg5{8-SVpVFI4Zinr&kqIa_eU`Xy-7t7bRJ6*R3~Mwe|sVm}N3LC2DCn7cEUNFW(7e
zMRjrBq87OmwNkD9L_B3-3!E=EIuAQrq>ETpR5e&zVM{FYOjuV@RkWZ_UIIiv<kAh<
zl!F9@T1=U&H{#`)b-fafcl4(>n0^OnYS}0h8G2j7U(h!CwbOQ8w^y%SR+5$#sCRj_
z4XG-$0+j3l=4pZ`YwesXomJ?XS*8_dcVI0kS7IuhFB4z>=nc+Fh|uVck^_g+o^=sk
zoXX;3UT+g6Y*=Gh(#*0fcyD>=P7QJ%xyG9uH;uw|j!=_`KU$cdF;!fxEa!=Jmis-1
z_&s5O!tHit7)Ok?gt!t`S@DjYX6HIST{do0#5qe05tkA%%jf5vq;<7XBF@z2tF(?7
zGuQK(aew0Y=ESljw)l}ZV~4pbY{hX|Hwtk0?7+8Ait`evR<Y#FJY7+yVvVSPI%dK$
zi47%{wl+r;Y-<VIitjp%?+BMG_zs<g*6_z%f6`EwVt%c&(1=_)ILmI=Q<UP<^(IBw
z)?#}H2~-k(oT`OoT1Xjo_dD!ClBI7S_MG+uUw`wFVnAudVb|ljhDOMVlnUd%CmGHA
zryJwo`RufB5X;1pAS^+uZte!=>y3P^I3rW~76Q-b8zmJ^<G^X`Nz)8ew1IvQ4o_S}
zmUl+e_nwp!^Rj?a#9CivXsBeec<rnqg~&Xu4BLk9Iyv2Fs1&wsWt~^NbsX*vxK1FO
zwt4)}N|HM@QFCr4Y|Hif%&;HOB&?~VB14igir8=%Rm(r4Dmg>U1?vrm<BoGlY$ej+
zm_B~X|MZ{!@2u0tx4-*}ckd27pZ~xzo%!nhH+=r1pY!7%|CAs9^k@9t|NDO-=D_E_
z_$9yitAERKy)jKU<`AgbvKx2uFwz=ttpI@osH{jei|;JwKTA+TIVVP6V=zs%l73es
zhU{9g#&g==y}khD#mlR?zoIC{gVL-q63%C85oAoQ1y}QH33+VrVJfofE4BG8WX-4)
z()*bt6_~Q9&#2b7sy+B4>eP|rf-;Ie1=7>+sMB{ir7%vD(?-sT(Fwm!jYrh@hU)K_
zw!-rK0b>i}Fan|y3LmdMjFIX+t+gO~lzL6Q%0oz$5vx5hX6ejKFTTNsjA-{=PJ&E`
z>k6d^3rs7vb))OL#<?@Es;Aa6bd{214V@AxfbMlOEQ=acYh~RQjO##aLd-3pX;IFz
z8}@ALMlqJlO$HWfp(sT{VXc2vSpi~9tjmmRRaYPnL2W(7deZj8)7EkcvofUurC?h4
zLo1859o}ZPB(@N16xBImk^!vJ`gW~ni;3yed)Dg{I<BOuNJjo{&RV{B`ws61a{T@v
zDLN`pL^)F^vI;D>h|yx2H^$LfL*IG2uBYz?v{nM7)<9LS@b#C0OZr3QqpVqwB+As+
zM0BZw$)M6gHT6}6s<leuGRu8ZN+m1vYPh$p(W3VjRXr~$=A5Ko*fcd=K38GdRod#0
zjM9S1a9a4M2C1Pl&Qe=Yh?2$Yr%d}kD6~<ySM@rglBC(S5wAAwAZ>D?>pHrw!#Yur
zbp>Y(s=;V)n^expxlyAruPT=)TZL_!Sk{@GqP(YGhT$K2x5a=}h0CY!h+z}8uJ76H
zkF3`VVZL&E{>0svk-IN`LPk^K21c{3GuBv~bFY577mN&ipN8)G{#EJExe~Sw?>)wd
z@G{@7#5Q!Ln8-;ES5%m%iJCWHhf<19pPo2+#TQ@R%Mj!`hGEBaozc3=`*>csK0mYT
zI(avSRk(#zdf&4@9>jA@Wtk?lQVhERbG?ZGvj%R{#<DIFz)F?p>%yCP#pw3;tC<{R
z;7?m*nHHSw82gbmL_)}{%Z9d&{bA2~yD?1*6~np)yw&Xcj;`<7Lf|&v*!2Ua!yUKl
z#D3qCVj`}ARHjy?CV=%%A0C;XAK7(|GJ0axs3JHgV-yeW$Td}d^Zp}C*RgE!EKSwW
zJ0T9#h9WC5N<CmjIGrlddrwVLCrYV=81UAk`QxU#KW(TB5z`5m&9;QnY3O-(cZar4
zh6-!gQesO9Wi4nWfa1naDlJB?v_)I!dMT{7+HtoV`24ds_|6eh;nOtpe7R9;p&L5J
zu|s*=UJgnCry-+(P8&Jd_nt^)OA(a}U<q61{Je1I9mm~{!FSN;*12S?)j~9H95nBx
zVx*SB<$A;S7S}tJ_ITe*2&5!-`<}r%=G#QPq}Q_cq4V_n5vwid$7i0eSHAf2bH4cU
zGjx+zC5a|~R?3-mS;$+)7{xetlpL9-Ge#?h!L_xZoMf~&40>RH*vZccjpJk)`=0%A
z$2{NAMiMk7M}GIKf8yh}pP1*Fu2*O^bG=>p^{@Yd@$P}K>-g;LJCs(8$C01?<S#h*
zfp2eLv4w@*&{344Kb{!&2V(w~N(8OwtC4d0)_tj(ssD#wFFZ1%<w2ZtY;2<1)=LPM
z!F57oZxGTFH}WRxZcz@mJJRotOv}pj={td-npUu&6skzVDkG|vv4Sp%X+<mA&mgPL
z(i9DA45dh|meLw)z*;Rx0Xv{=hc-gM*V+>M9X8L9H;4<SWP&!>?uZ%h(SAotiRbgT
zaD8St^wcsCLquEYq6ATUC2%1t)Y=qtrOa!XL2F?jRh@)wr5#mQY8F41(FW~0O0JE|
zoJm1I-6<vdzQ<bo`jU8gflDIxC5fwm0!miLlD;$zJtFm8IR^PT8}dS!C^8t36`sVr
z6}0j)geV$5L<Ym06YDL!CNYf$H;iqiC@mk>O2{LJAHMDf+R7bb5?HlwB>sZ@9*s7r
zlCY}M4;>~s%CfTNNY#ej;hvBqPfwrNLPFO-rwvuxw*6F`c491lJU>yVOOpmGx%uIw
z_zpS5<Vq?<h7r@fs-?a64CDA(iY_gAvTCjMg`D52*ed7-H&!p}$<o&LwepfirD0G&
zhL+q8GquQI7n4{?avRz;urrl+B))xo1m{WpK&(mrOs!k8R{Y1>O33GBRd2Nt1W<`M
z-(jjpl@}aOhFsAl$tENf!m_bH?b^Gfu>-9^=}I(}BprcruNo`EE7FEwjS`!HZo$1;
zP)b&fTMOTH-V2T?LyZ+xnhiiz*02%6CiQj7wKVh}DV_W`u4P&1RDqm`(^VQGR7S<O
zgmr-wxn3_^o47C>?xcU)Sc=gUCyge?II@y>jA{32EusLWm?-V{nzN|r%d!x{iubbV
zs<JC2Z6Y3cYf)92mv*DYJIj(1rnOOYP9WH)<B8f7Ie_c+$~-UZ_j^pKINk0=Rm30W
zJ-72klDCZ=+d0d&trSY*Tte8yvwT}4PtOxS{p_A`6*54rh0e-5r^HAt5JJM}!sR+~
zzTH?uz&b}&qN-M{Q7x?<V_?_GY9B3~uGTsFalo~i;^E;QbZP$H!Z7qyV_4>iu&x}&
zHjm`OvKDrCdzQGNo#8^l*2LTWK&_f(vFr{bkF)f2t6Ffzk;*E`-&$HQr!sGg5DT<e
zS%$7_K^JHVn1A*9(>QgtwBANhF!s*z+1-hL=vhmVe$@TQ^6|_vEhuZLTkFv@a&|OY
z9{g(v=v-N1#I6SGWXkTHM;ps+4%{9mzWH#Wl)`QtxIZ1yy%Pph-*I<3;l~@7>lHur
z;!k>E`WUGO(^ioqge`&dvaV^TabpdoN+04lblh4;VP_rV*vZRLC%Um?-1TVWMwV!Q
zrtcib!;TOG3aP)*T65U%*pE9><K5IsOZvz|#aK(uv2`IDsZEQK7$U>yINgnOy#pf+
z38_YMmD7{bpuMKwdG7AV*N6Dfc~XV_X~(eZ@%;;yAhE6i>k4U&jo(($CKB_+<Hv9L
z)vx}){N<Nl@bS|hP&IPjdoI@xe7GF=?)g1Gdiy0sD|XMH_&5LdKk$!#_y+yA|0MNL
zjk4Z#p5?Z^^2E}gfm5fokyVV6#9qm=LNMdLsVR{ZRw)5*-MzIGVu4Z(H3zmh<EoOC
zt#=HElhFE0k@~i@K}x9BBicDHH~_>{%nQxkv{Yw<U23DrOGGPaH*l`UIbpWd{8Cd_
zD9WPtpW(*`OlioFBn&O6C}${D6XQnQW>T0*>%_3{$tAJHm0mjuUuf~QQ3@#rRM)@$
zd_ve0Jb^KtR6lJSN@a3OK;|Uh-w;-MB}%2ThLV&7C7i`+Pb%^-%PC7b))?t~EV&uz
zd0oHiw)Tu6ptV7(j$A|?kp_?a`f^o93v62(rpq$}E519CttV^o7ixv9mCVo(EF08{
zQy%B+>+cxytJ*`YxbkNj>VVAmL!D9@r5&lQAC1y@4cTbW7IdL*V$bly9zX60w=>H!
zv92q<v-G3KI!C9h_#Ua8uUFJEW7Zi(#X2i6QY-6br6kxTmbkWZY}gWO4bI9Tl;)Mw
z>LuwaAsF3sy4-5GY>^r*-6G7jRHQghio=y^f1d^-*GhfQfNaBoZX8F!3dxJc-7cK}
z@o)Km{4f7N-}k)x%fI0G(=Sm@Izr_V-$<#D8bi@*&FC$M-mx1TL*HSM189v=hAv&H
zS`p@nzy5#yEuS9G{Kx<JKl1KJpHZvu<g5`j%^C?bq}l>STK<koSNcxV`GyU$vewsB
zV%)@&X5~OsN<o>vbw?*4HHM&}Zh<t<=#+3qKF89YbwB93y{C{lnZVuAQLASwk&q@j
z1rMLSL47*&@$$^auYbq!evyIIcvR<!R=~T?%X_nwDxls{UVX5#mTkkD?_Wg#ISEKJ
zLfTexkR~JX3|5v|>eue?jtoQ3c3W7MmCI$KmLqTO?pYS0B6#Ommzl2fbiSkHOxRXx
zFADS64OnaW^!P~L0(W;OtQGpi<>{%Z@EURg#Mg!@%=3iNhU+TL3b)&p-Oy9h#^@D?
zLGWfNCX&)zriHnsMng#Soo`zdC1|l)u-=j?)R^e3k;*~ew;??<j6ESI2JeYEGwye+
z^9(j)tm5t42d2(3j84|~Az=F*k7vacGs;`S^#&zk3VgQf={&4k;d=f=jDgN}^nK5g
zGutG|{+tsbY*?d7YhqhxcJJQdT}KV^b=&f<U4J5cSTF)8P_;04$LY8Ot+{Rs$$;zR
z6qS}$UI?|~J4e@9ash2BhQxA>!Vls|3Yl7zAUsmUD6mBlmyPpvA?HH3^Njlu)k&fI
z;kf7Bo452s&pfYu@%AlaH!#mPYRq)K!+47`j<9Sj+s14hmnqQi3&uNK>M>@amMU@l
zS_vuBTSK?&F`Z*nm3==l_9K-ly?RBFf*56;kyD^hkeUa{$$BD}BFdY0l$@C6iI5Xj
zH&~=MU=6mkf@!Pi2g~8U=XAfv7)QvFR6GQE!Du6=QA}a08(PViP+NoT4R7AuG4vhp
zzkQF_inwjUpXwdTh`(uBS4u66gQedY((=f6@BfCczWRGqW9*#Y|DNA06UG`od-x1h
z;jrK1wW2c`XGDFqy`>vG`@_Ixo?GF2#rzo%6Qu-Vkby;O)#CR?URbqIV-S9ivQ15h
zY?~Nlwd?5pp1~!ydE)%t%I<JPhd|f_5UWv`m-dOAF%|4f#9JGbDTEN^&u%_*V=ctc
zwt|w#*2bZ24@O@4ncPhALZ?>>!H^h*vIfhDMj*#EM{3>(%T2sdwj*g-rMrgrNv%RB
zHUvq^VoPw1$(M7cbN<ygldGg%#qBKr*;?h$MX3zUc(09>E>bW3&BpjfFO&aGYi&#8
zA;^e^Ku}6(_FD*azL$oAA|9)C5%f!~84xAWI!`R35NV~+en4ACxO^f_3;w}k+@AB2
zuoQBtR9D;ZVn|W?epCCiDT!JoY?Y%DZ_1wm1Pf40hOAl;MVc0J)spCL3aM6_&TC2t
zgql&cN<b=SuGeR7%T20}OsnxS8rya3jAv7dbq#Fmf=?U1w$*bI&f2jb#CKRCy4HqK
z5`+HG_jJB{W$D#i$W<!bC@;Rs*78wuTkQr=8C9(Ozt&V^sQSMK(aQZUaq;HWeANrZ
z<M*-(f5G#ph6m}_3!koMK7VtEHJ-2k?pIJMZ~x8Dnm<rC%U%P!mrUmjLqD+ZJ%_Pp
zKgfDi0VQc^FDMzRVwxv@^Dn>U{fCb{%^Sb{%~!np(Ysa&rdpkzv?ow1Di?AU>x<C^
z<22TcoTdX$*M+c2LrBRo_|zf+lWm!~TyAt)VaenupQ%!cq3c-JLYQYXNM)q?9&Y|a
zuK(>%k2qE6j2w36d1cd@rD)!NeCBl8Q*&ZoZrsj~RAsn3Jy5G~AuA#ppulO3(sCfs
z`uk;FO2SkW%SFOkN8d@fF>V_vB^iL5Dqq_`NEIgS-Q69i7tz9SJRW)b@RsxW%rY<J
z9QiQM+@DS?>ndK|oN?CTI!EUuv46YXAZPabJ)oFw6WcnILwH4*shT<NI<%5)h<8rb
zl{rb<Ml8(BGo#n+`W=_s%rdW}ECI5VO0&dBx2-mcWnFNt!+VQ0ij<ng07X?sR--ZE
zoRMz!`f_l$n7+sNJ&zwhG3*U*-n}6Nsl-#Nay*P|L7|-DwppI06<^@gdj@N`U9WWB
za=#lmuaV_;=6Zd`?B9@L!Z`U(YGdTE(88hA!hRSC+twOCq&o1Ax&E~DYcp1tQc$I`
z?<KWasa&>=s0&*TjeS(9DWRNYKX{J&fh}d0uo1VF6rt-qB=O9x>q4q+<x?^x6~Zd$
zhq3eg#m|4tyEkvRtTV%L<Z!xU824;z;?42I=MQhO&I;3|f*+-iQY+1}1*uE%o~P@L
z<HH_jJ$;v3%Y#VkVUuc^yZb$dy91qf9EYB};~ln_msgHSOpMyFtP7>=UZJEpS5hj3
zn6W(!-N@bDJ%_skyZwk(2rQM1cb2~Qvb&!G)*Fo17!Rg1I4>19ra+;iDoK|?C1p7Q
z#F*%u#&(9HGk&z3?hm3GR>is{24m?)&)u6nVZQO<8_-tcyu(?;?l5rJcX*YsD!&#)
zlUBC%%v2K3(+3`(zvj2U`a4c0^WE=%-3<7JdD}ScJMnFynU_tz{6Bm%!FD}khu!Oo
zC%r&~r&;_#CCR{49I9%<CMyL=P6o6;;jAW=KzN?%;)XSf90Dn4OgA)iK&94z&543W
z*`5*uC2!)fv>mB}a~3M3DrqsJ6wY_DI<gjJ6}rd@M7&D54LjeP%wL8<YHO|t3ps9V
z+bYUq48kR>g|G!#k5*ZG=BlJCwU(wxiZWc#6y+$YV6-B~C}wcy$+4nzL8Hm>W$;lL
z<B9E5Y^{S@WJM~UhCFE6y0D}Sy7_gAV1bOage~A**TNDF{ZSk1j`W(|6?I1zw$68y
zoanqK8H?Ix!t>Xd;enb*Zc%=(UFT6&Qnm7tiTctu?qc6C;;C~^qyEUa|G##X6e-P<
zi83f{T8oExErmHH)mtfR0RU_1l3cfiHExo$Yi4}soS=+q<}ePB#Osq%LPa^imZqXQ
z1gfgElLLoRjNZStMRd+H45O$mwMuIFRfWy1<w3^soH98q=qd;mgQk#QeRNv4-}TF2
z*BVgd@7)gWN&;hA=B9S<3e)W6&~o&J-xx(V^z4R_RN?%G@92Ba@#jC5!v*RU3sbAY
z4;^FYrM<(uSJ<w#nra}cOiqer%RGMd760XL|Al@S=u72%enwZ}O)8}3N*gCA_*x+s
zX`HB%opD1)EwF5f_n*!zn_y*9No3^Et)t*sF6T3yQuJO+w{r6(TCH$Kv&=Ikzm%(s
z()~lP|J(oh-?2_J%2?Jd@%1<F>AYfp_rPEOgCeSqoCdW1oZmm+czXPnPFw6CTbiQ1
zv@K}Wz}FI<2%FdzC<$^AAhkBo8km%>#1Pn)S?(iV4uVLqC1$L#cr8H}r4`;edS^Kt
zMy4&0wtzK`+cdE|9?8Hu&%ArMqqkQ4@Npx}t870?q*I#tHp%^33)_0bXt8~yZRO3e
zCnwLg>v+E2c)Y<>3%j8g`^*|z$Sd)9(QM1cd5TOe6j5{FZr^dR2CXaO=w-vGWC(V~
zV|+oOrPaY&vazI8>7Ao<R(y+FAa^~dhX-l`q5$5793YmH&UmV;^r|Noxlaz`j%u0@
zPJ2#o?kPD?Yv#^tVw`w<`i|r7UhrzGbRXxqV4cNxgK&O3&*+@6JX4g-(I0uc{YgWe
zs`a&V&??Pc-_u!%+McI{^KE5WB1(bvnp&EQWyR2~jqJB`kX4sA@c4X{<WxvJ1d+ZC
zu*?(Bk2h-0{OIi){_00x(jE46M$;b-42L5*!|igBc(&FI!$9Bn6s_sH4%ZLOfT*xa
z@#BXBcY{Ux3&k{?OJdssR}}NSvL8C$eRg8E@5xkl<G}v7Lv;drB^QiRy#LgE0;8i3
zqLgVVn8jMdaev^j+e6Jfyt(Ji-95YUKunojMOiuR_q=;^;-W5?&P&p0ThUtMdM81V
z=JD$Ko>FGkWhKW<bq=d7&WPNtou+6BekcT^v8IYs6=elBIXKI%^HQCo#QPYxg5Eh|
zP^7i6ZHXmBLCNTadA{=NfBK(E>q?!@{NZ2zj_LUk=M-VfT$jLp@GR>l4~eZARMihZ
z#V#SEwXFHPGgzb|O*}RmA+AzftvgDpZIBdh({kQ;diua_7}^Q862jDcLk4XOMJqym
zQ3VQ`A}Agm>5W2n9!SkLAO-U&MRMGv8cA!M6-0{szf7_f@r}~bfr3a81w|Xx!auPc
zZdOzdq;00AjT~lDL!9KCv97^cb4II|o(b`Ig|_lE))PyjVd-V<`{GgK#d}wT`PG2E
zaxe3`F4wKYQ)%&DiC;yAK5ay)s?AJZ@=MRC5Y%N0C_`vJjBHz=(_I@lq*hGK|C*fS
z=keZS_V<)pNnv6Qf&JajK=&Bk;mCv-2r0_yq_l1hqcP6Q&#g2m%Q`URKM&HRHqV){
z7PQ5fjt~}6J4<b>FL~Yf&Vw>+xsp~n%+_KW!;Dwsy|WT9(#}fi+BCIn%Opw*_sD8L
zOcU7|hOx(M=$-t|(Uz|3*zNXs>%>2q3pKaiWBK99smSn@BQ+(e%A(vgm9rLM(B+)l
zq}IAsO93Ul!KS`OEE;Y^x9(v5+B<5RFU{73?|LpD&J^YG{XnXfzyIsM<NxsaJG#4r
zc+p}+)r@mJW8Vu1*7rEuV~xXFK_#IT21nAGPv3sa@BZbtgjDFYC6=|dMI`Z*DKHp#
zNe#CIq0yqyIA>bpi)CImcDtU-ZDq+xJikh6E~P%N;Z5R3=lk}#h|ft>Y#Hk2d81U-
z4im!<zTbZOm%kvcf$MEz7<TlBJ<h;xcVdm2^KHY_%)@@ikAL(T|I=UpJ<lJ%WqkMo
zZ#<Zc_5;qzpdUkm+B~gELJuWnQVpz|z|D8IX+BjB1Em(6(ywdT5H@PnbiJeRh1aQ!
zr7Boc;+qejs5&ut$F6hih8?}{Sk{Hp>4bBhWto{SXO?+}m>9+Z*E`mjxLuyalc*HC
z!IPr&TEBa9!fL}j&1~B<)>_7$k?_|tbG}^2wNg~&d0Dxvfq7b4)`-!TR4Z1)lAzqA
z86p)$*dj(HcH>SC@v1WPJ(ou7Shgsn0VV!)XCS1^=nS2;JYUYNIq>D%6YFh;lsO)c
z7^`4fSYs42Mu@~zTa%3D`F!TMA8|$wWYGo8X!-W5-|(}~|4lm>!r%wCutC*eEyH1t
zDFst2T32+b;vM`Ws}TRBq0Tsg4sMG`mAj!wDVUdq$MeKA1#%ESTR*zy!+@$Y);Wf*
z7c|K<Q;J0Ym$kIgc14&L=If2pd2Z($pFW*QSxki2+YR44?m8z)lv0_NjpxT_^0uIg
zRDoD0zBPrS>pfN}YU@`#9`?L>bC2GQq!cKt7)M`v07Hyom>s1OYz`X>kvp&HdnYwr
zTC*-QMJ0xP&%4jwqMhdbcTX~4?K%#}k=<dCMvKNe8itY14)mSC21mcgVYt6P66Y0V
z4HTGH#b7!{*O9d}7+CGtQBkFmVqv;%qBI(Th^m$l3RNJQozrx^mG!E(q>@>~CM^ZF
zH8C`NK`n(&JLYL)x(Z_AB|t*l*uu)PoXK0DQ?ShoC1nnMN2-x<UC2p{)Jj#1s_4>4
z=0EWA{SMprSXCR=prBsXMmZ7VhILXm<($EILsqgT%Sx*3d_OXb2b9v1$cqwEu}YJR
z7=i0c*h2S|)PfAg<E<lS#g>Z<lItS(O_IQdQlijU<7MEI=%Nf=B~#N&ayDa(Kt$zK
zCY`Z4%33ssmFfHe6&6Zb#rz*5S;-4?d-;2|<Y}!k2-GZV-lC!mPt@;U=sCWcmqAj&
z8CkiiBHg9Jt<tjAHO=qUJXf-kYsULl-<3;}`zMR~W}SWQm3>*+=9I*nqFTB#OLg80
zvZN?!qcGh_HJ%azHLZl{0@wGL-Io|YN+R&~iT6)WobK<~4?DuTp^$@=)mjF*DxsV|
zL)cnADQ&Sv@IgYUZ(Y97)Iu%3VrA$IpfmfS4NRRQB~f*^4A*5sDd?Q1_Z?dn&@G5s
zuAGCCUcHIZ`3kCJI%C=O9;;=2>HCp!JhV<)4Gnce7GE4zi%+gGKvlzoC?x}V%F@(O
zTC!D`Znkmtw44VdsUnsEQ(AC^*ZW+zpG7zIEf-nw4a3N7{>1a`8E-A);LzIf;n!dB
zSO3SKqA<ifVYOixM*42Vx&h~880q=}Z)HVSQp6a~y3RcQ`d|3fS6>sW<9S(7Mspem
z;`N5UdyvFG(s2wz0x&OYGpn)QNi&D>?8iOFap04?vSuUdv$6P2K6~4Fv=%#saTcc~
zKt;_cU2t1so;N-{oq4{Bs*?T;N?m_Aa(b|g^MWy!!^4Sbx`ERC1@PhHnOGuWp18Yz
zV3}tA)8G6v^K`-Qk95vseJ3#IoTbhy%Sv9ldm}M!uPJtGHQPLsN^ZWlg&ZXfUvhzz
z+Ou0TO)JNK$A0XXLnefcm=g2Oa=UqUZ$#+WjUyoj4u=CdMQ*o=r;i^XC29%0yT7L&
z2c9oy;=JH{FIAxpD6frSx0B|Nk|jW~-wlMkptg;$PIOjtI_`)ev26=;QBqG=6#X#1
z(in11Od*p^#cr7;-srtyaFU+3Mh^J<aloKBPYYp9TrO8`#{>In=}SR7&$!<amznu`
zVRzc2JHz^ZW-x*i(!FJRzTtgGTxL4&czix%vc{;&ySI1buwu0+`iIfuy<wZ5nXZrQ
z4{vc@j~NZVWU2yp+@lnP>y0!`5`0sF=bFOuN8M9@+E8bN-$Si9Bg~~~i9D|}m-Cfv
zizuD(oxyY#?JP!D{9rlV?RdLCa$086?;@A;#;PO3vBP(k6vfw8DQr1&TLSZ%Fet_V
z>p&F0p%dzNiX4X>UK!zv0o$}NdIKf0PG`2;jlLARTG@Mxb(TP3iG^E8B&~UK+@pek
zV4t5a<O=)!fco}~PMQ7bz?*k(;rWSF<pKE8y{VePJjB5L@xX3(U|LoVrvu)5?jG)W
zc>BQNC?Lbs0uJMD$1wCPaUn-VXF4!RK%$K$W;;i@RMst0tHNnZKlb8N%ZYWGv1sBH
z*?hz(BkQqTh$T^t;1c?-V}BU9e>mW^#tojse&pIp%5$k$uW@eR{>_2izC&wav(eTd
zO0^PuEy#d1%<@8@;yNqAkuh=Jgb|pnq8Kqkry{AnA9{JC>HC426We?xG^xJd?HD?X
z>m`=2733-i1@ArMZilfQH5NiixZy-Mp1937e*dfgCzr?fj9tgw@m@~27^=5u*P(Mk
zr9i2Xs>&;=Ki4WPA60%AwaWVSrDCa+46#URuf6<>)>2JYX{**KO$m*>P6J~u%f#*Z
z3BnC$6#j5R8%vCvtYfP}HG^}Eu_eT0$X8lUaY{G9a>~L1D_PnXa%qEGZa9ESZYrZN
z`K0=Z)&zmllq%~(Z44<(n}wjeq;cW}?Nb}rHs#!WkD6`U@SSf52uUF3BL2FZ3y^TN
zvaTzB+|%EG&Mif{T%iPNiCm@yZw*mfKL7bIIo`cNQ#gNkPu}LXA&Mw8-7vDXdci-(
zrQIm}z}7S%h?_RdSxeT6k_s59H?jndyOGXG3k;>A#}jL{43|gZCJ%c<mG!sG!+7VV
z*xVbek`$jxiQXuNuBZ2&e%Rys0a<xkoOiG5XWcL>uS`9JZ<!l&>}6#OklPYF7tmFS
zqMb$Y0R2)jB7>rm4TRO-KkH@h%q!ZaRJk{_R;19<he_%c*Dc{Y$9{L@>6=HMfA<xK
zpZpl*6UG|6>u}biv;>4eYOz8PACa^oM1J$@-}C$5{*EmvcKaPcTSzOr-N3v|nEi=q
z4M3vmhLSM~?_SU^8oLHMpYWa7N{m)0M4if|VwA=>@m{$as5Le}sDz|sAl8&BYdTdI
zK79CwKm6e>>pW5Z3_ecn9ojoOfLiDVhw+AaS$KGJU|J>~-W<89LJcclzB%#5XLtPi
z+b7&OQu_h#JFJzYea(U%EA3!c^7r2jtu^bql5Wxf;hbR{MuJ*p^PmLZ@-h(iy?D%G
z*icxkR*d@{yS`%_<j*`dw3xAu^Z7#O9Av3+R0iI>J25!Nr^hEgJwI}{m-nR6mfQ8p
zJg*#&BVE@sU2mvTS=Y#ITDje><dC@AbsTm*>gmc{Gfs6l(`vn97T;@5cvHzr5mIIf
zSz!Ohu*;sDz~~iY6+7Qy6zumSYf4;~m0T6S`EW&9$J@U0aNOZK&wM@ORHYjS>9AjB
ziVkF>IE*KhZ7fp~cZZSNZ388}y;^}DtURxZ!bZn#?0CLB^Xc(-jQhLxUC{W>qnyJG
zg9OaXMlKD#lQVG(?cn~eUw<0XBxlkRD5)?W9No|nYvtR^m1&V`F6RuocVf!c8sA%n
zy~hlW=qgV4^uFU(Hi7(VIVI64+})nx;e^%(KHaEFv)_3h9!}DIIPTev2edJ4DWa1Y
zLOVa8s$#m_xSY?@qM#KAE3CV1*$6SSEi2P>W1sH8D5)y53f((Q@8|}{&~^O9&ws{`
zKl>6>4f8zH8|esK;!JXur{`zR=PQTZj&&8X`mXOeosJyu?%0n<I^Pqb$fczUxY>11
zAegmKDTEY=Aqk?wiNCF%R$=($N>x(Ou9YEI2r;rP8zmItnxQDHleF!Y0<jc~GVS4M
z8O9z{D*Mx({c%sf@32m${;un(CE~1R7&^w?fa@(GXF`%HG^Gs#9s6;FBroX{6B&)_
zJ$vssoko0bF;+Sqx0ner|5+GLalNvs^m>|i&P&?X@5IoJCWeh^yOH7wlvY>cu~yPT
zTUpuC+SZYU>MTo)gl)wb&$x4v!m0(WDxI^Wq)BlT^LS2DzMKoXr9MIav$`s&Da!%l
z1jnEh)K+DrT}Rdinqo&0aIT~(1;Im2ln|uiOewUn(!E;cr5A#nWYufeZNV!G(!I?V
zHaTFW@4a@_C_z~?C~wIXXB{a^=tF5kQIgCynAcv&loGb7zbGjV7ay5v%A`ENWXQ9v
z`%=9ou%%L{Rr<AXNu`BOl#$?+k~Jf7CM^}$_Y4o8vym_rjJIfekcX4ejCY^0JKRBy
zZ1Y4g8tpvVTB>#!*OQlLw(0zX?>B&=9Z7(1ZQUsN7KM^SlOM?$O3fIPQO00O7In3$
z9LI;ZWZN_JJ@fR8K{$rdz^Fx?FjaGZIB*zxS=Yu$-WI&7?0Ok|oFDMr0otLVqKT<e
zOlkeMl1h!avTi|!PX*R`su9>Q6>83GQLGUrBeiU`Vr->qQTGdqq7(&9=PlmqHq^+y
zpR*uCVy=Xe1sHw1QY&mhOz~Tge#*<dG7bZ|RDS&r|G>YyyCdd;R^kh-Nqz=vF6{bV
zQjayGaz>TJmNdWp+kfOg|MokA9vF-h6Md<ix6H5p@QHhK;_Js}T0hEb{hHPqXPgWb
z8pd(pa5|9E%7^D0(>#&3=5uu(qoqA#*#ejAL<)s+%GlEZqZA<~mUWeIlG1$h%{OEW
zyTgfEf4Fhj_aoj(C_%NZR;x65S<yA~=CmhnndN#V&QghZe|O}oPgiP|L0Cde;+lLJ
z;9i;xYRy=q+cQGKds<6<rb0;#pnRi=DN3UVa(`=W+17=z|B@w15AHVK*t$w*@7N6;
zIVYBF<9b^p{FO7ybR#SihhgCIJn`N2#u7Ik-oByt9bvt(EHhdgd~ex8<m+!fP*P^U
z8(G%KhmTM6y=Qm7XY4HNGT|)@Z|;aG@!|0ii=>`&%-GH|^c~Z>pb%L58Ui^)a)#_R
z&T0<(0sa4@?N5GfOVji|?Dw^++1=@G7hOhXRu#!&iKI!<5G=!hVGktO@W2{*WI%KO
zWA@N9Hv?OOEy$KcTBKQ}tn4B(;@-ITo@O_z`&tj)b#BtIjLJl$HW0{2WaK?3_Fj9f
z_j}&w`8`T8O(R|31A;B;w>ut&#BYB%;$QFRLgX?Gs1WGdmh(JG(poEqVIXVRUTyJh
z14#hVs%EaXJrBnV-Wl4~5$6e~#mbR_DEj@za~|FihOg1hMp~0{!5Jg>@%1xQMLeP<
z2z31q-G2Si=~rE3aJ$>>IBYg(4fB#osYqPhNvcy+I_a@ng5ok~LY>$&j_%OX9!IWw
z&&C+WBoEu{Gz$fzHS_8V@?IEiKOCRf9Cmb@${Z%1h6~fUaGC>~wr38RaS6=1puMB0
ziZ7LM92g#t^j34-w{(ri)WRHQY7R{E%#ssxNnB1dFW%g7xVgf&JxM99HV3}AdC3?9
zr)i|DYua3(F_t(lJU$!=bD-b!bZt-HY(N=mjl_^XoeXo4#F#QdLQgre%nRc*<9vtn
z2Gba}`yJcuh9yO6l@3I;uEkQ4RkBqIZ^S!z*zZ|NWLyHy+D~TaIV^Z3fs4Lt={wK1
zZ`pM{jc=HiKq!f(?`YbFrW3ADRuyZ76PBZ_RU2pV&hYO3L<kA(#0Rw9wY06FZ5$L)
zhEm~AqMRtLP-<aWB5iwMvpG--g!v@iwi1y@enlxseL16Kg^*-$HcF$A5QX-RFJAwI
z*H?GsWfHTlkwz13Gzw9*s<qg)dMH<oDzYw1DHH1IB`%V<OhHb$wTMEgYFT?#HO?5E
z^6UE;p(J@|6ftwBWnp;yf%m`tB`VMC_ghidQd+$(f~TM&5A`D8?-YXgVRBj@2Jp<N
zD=Er>pcZt_I45;#%2@0=q$mTbh_6VzX{#4dNeDz*GYn1J3p$0hEd-6R1_9g(K7%Zn
zl!{d4Nh5*WoSs=(&;6-27nD(?IHT(d5LQ*15q8&<;fWG2Y_=~MgJQazIi8Q~4p;c5
z2kp_@j<}*doNXlK+zzzf{kLC4f8^`&cwy7H)d#pHOHnBEoEE4oYZs_eWmVB-xT;k=
zjKefitoSOukv}U_L?J5)oszs~A(7NfQU6w0!6;HGj7y-X0p_)2y<`bat!i0ads&}p
z^p|l06o<{0!_|SN?I>%-*&IhsmzjB8=P1M4{we_^8PYXbDRPC@JDh$-sTc~1kPFk2
zn3qUc2b{xv&rIdxMLb2)ooqN>CUON!A>L1%et1uAw!~$Y!=F;?N?TH^H780H)|&Gk
zn*+n|zU7+_M_#^Y$OyO08Mwd9j7xxrM}Ga=_wv1{nVi$V#c%|(Vqdz=j_r1fVj-MQ
zEK7hQfMYp;3AgPsPCT7t;Om=)yc3&8SOO_a<4IMDaa#E9yYFdsGs^fMeZ6IEX?h9K
z#c86VaY_>Z!?Lj1d4AHo!qmby-@X;wL@I23!#Gd0QM#*D6*S2>UPvLV$!D=;B!HC@
z>zw-pD=sOm_w4-1Gpe;*`bInAJmYs8c8%pUoUvMwOGO!rDv*{;iJ8khirSkq#x#^F
z_uTXGNQ?>REY~+Tc%wL-9#L7oQ>kPgo@P$R6Q8|&$!6;bF>?QKr0+eqS68yxr~**G
z+3zsM^YP)46s3vDX-{V@oi!_CF|+G?!cusj6PL??)s`VfrkH4*W141~#`AC)G1}2}
z8>-9PPxpM76~lKI{!&|1DO_z^#yn9=Vi*c-*DxL@>C{dE16;;|T!2z=RswxO9x!<C
zvC7aIPc(^Ypiz#dfv4eDT=9mEmnhrFaGnb}201Vrxvp&qq_8mjfh^8HZvARC#zAWg
zhiymOIF_)m<lh(83%y}2Kd-V1nC47Mf!)E<_lEx3bG>Z&Y`dXVhTpt9b2(3Z@JE&`
zE6Mi{XNmTzjJBF(nRq%ql56H+8ks}naTpj@gZ_udM}{SEf4&gc6l<>VcpP~=PF!zW
zuC8|Utw)zkT^5$pk?}kbRAqP!oF32QS2q$eSVoL0{N+u<zCLliyQXV9PRk_FP^0L&
z4sSg<W*(kSY&Ki2+C4Yjo~VR_qpPLlN?s~yE|e%fm}*2)Jw2V7rU||Aj3F?EgzFql
z-=d<SWT`>1%Fwk9n|(*uS)3D!yLFDXX($OcTG4tB1>z$0YB-N?8@jfWxVsjQjYgA#
z@YlM&#WoJ(G|nnwi44ogGzTgQ-#9T=TbRQlv2jx{UZGos?;MSHge1Oztm>aas3Vn1
zN{*B&F<qE~sBv+cpccG$v~4ed<tqLcr{&aXl+=BtqOitEeNw3?<8h59Vp$mF#IEhy
zl#s>DZ4E}d)#z?WRS+YSQnb}GEo)s+PK5I#H7yu(;JJU&C`Cc~HKjsI6si=ebiro9
zft>C?^7!_*JbnCuzA>n)9mYt4s~|(IvzC|=F)Wg#lpUFz#AC!5!!nPej1}?zl^oZh
zL->7m?Tpm+_v<_7+@}h^k_*ncb@fywp}HE$m6oOlV>Gdd_f@LNWN0fZv$3w&3x!$x
zXtS8jKgIB62rflVn;v=are&f=hjTrlp3rqh=Yf=4FfF<+R0XE#@vb9;d9AEcVi72j
ztQ5A}{P+tF`1LP;jk7D0O(}6muIP-a3@AfNm6VpXa`5;6ifX;+>%);Hi@!}Eu7xEA
zLMkjFQs+QYqLSp*(^#{t$Gg@uIk_$8{E<=>%MwUM)Uf&toGxYM(~0>#t|6B5)g9UO
zY;DD82~J#w$o<pAa5|A!)mSO{eZ@+EsHs+JAe|A<S<V^TG~yXbnd4=YHK~&8+UG}p
zS7+WGM_QxFMpj=bXNGa1U-4EC@7{9t)nAZ`W|>ycR>>&qQA(3y6c%Y`FeX#qf6L?j
z6CkQ=&V_x~5>w!KS!l0BZCRoWa;2<3LfQSu8n?<iRxCm@C5^DiA<$VsD*<v>QBGsb
zl%iHhv-uQ?(@N1eC$<Y;n3lDJcbQQ3$KT}i`h_%Fk(BsltEn9B?r5e+2qQJkG~VEi
z;jrD(9B2Bj;r_!D`u-!@^~ARO6t)qcHBzf+<on8My^wxbu~(@2|9$u_EBM(ME$LZX
zxw*bVYuIf%F4MwgnaL^h{^KJ{h;lzN5|zH%ZAejk%B(z3wZg`iIWn9BF)re7(-L4P
zrSSFFKfwVd1=136#&Wp1g)08soE42VG)>F-GBZsxFK=$xZ8l=b+O#agKta)lmSv7q
zWV6(DJ+&&H9*>kw%eU{3w5_MA3n68;yDhhe14bJfZMdX`k0J8Q?=x5KnnT|RZNgbf
z38?BwDe?ID#O>=FYOc)FLR=#It83CYVT}ahVv6)_M+uQRFS4p#{c&<@oLQESSf%OQ
zOH6PO7EBI=VWg%+SVm6a%;V+6-~RB!A9PLqaY&PuP1bnNzHjl~GK7hw3$9yvM7dC7
zUcda7wr_Bqrx?Y!M9iXSy<+=vL)&d>5||GI<NXEg4b8U27{xe5Vy+Zxv4@`auw^KP
zWg3`6M4^~tq#`O(XDmY=IUY~MC9`v$S}K=uAugG#yR9(wROR9H#M9Fw=Z6bVPZww`
z$NNEQJ8B{f15qjd?6AjBIp+pHO>Dg9hcGgRSt!)4$2dphTlU?S&b9Qmr8NzXO46Rj
zIFadVB`t}tL^dYTqWBOZ=kqANvP<NA9J!okN|qCJ)3~+!6F^iBrwz`E3fj6xl3zuk
zio&gRSTRJV^Fj`VP?OZQ86(N4T!}FeV&*(ZTG_X9db3^_fWtU*9OYEM*>31J9WiJ8
z*5f<JTm!gDb%m@GAxqOiER|{%)`^<%V_z{=c86=4wG!uie&8}Zp^~67#JjVTp>2)y
zk4chHS{WYB?|FDQQF7$==D^MM3!3%@>paF<Tx}$&)bFsyl9q`QCh1QM^XeVbpVSEn
z9H`$zSaV<rc{O(nlS$Zg;%}?#6F=p+DhLMay_n*SXLAK%85l1Q<S-D|nlH4YqNP;6
ztYzP&!75E;O)??^TZ#)M%lbjmZ1THXE6omRk$(?sozT%)doh(pT9@SS3u;ZwMHCjd
z`d+A&l!EXV?fP&di!aYvba4V@u4Qc{0BsE3iW;De5d(fLpA7c08vA`xS5d%u+-64_
z2I6=`YsF^MW9<&t_88w$v>;`2D!A3tlT#3OnEMleV1Zx!_P5f@tgK)UvaH*maUL19
z65xfiDZ_uk`$j17S&%@gs+B(_YtXO+QCYJnu)b|5*`c){H)2Rsr{z2P%nU1qm<!`L
zu}qWr9=)hTs)(P_T93_wm{D3W#l&e2s5WBsYEPJEQmIsJSaKvMg|SGQJ4ltDaaPv)
z+EJ8ZNr@1ODAwx%Iah%)%i&!^TKI4pnU>5KcLyF$hIbDaS$9=|{#vKGfB%73fA$Z-
zN-DAh=?d0f0xxMTncuaRTb;=9$Z-zq{YasdW!q@JKOOOH!)2KGdc{gr^<0By0BKW@
z#JL2fQqDl8q>R%B>x42;wW4T6jD?gk3WYNk>m@O(jT6j~Uj3`uP+X?<jwOn(R_h<X
zji^8=GUz^gc9k<Y-_mb9!});^mlM=P+ZZ$nRWjRt!^>OESQ5)LVXZVuh^1o{R%#I|
zj`wRiQ%C~OJUUjt=<_|M<U*t6x|3_#$#1ACecRx*B1{vfN5%Ew!2QDs-#EIyk*d%t
z%`XBF_o6r{QTOJM7*7{`V^FPQnHRio=(~>d>4;Z~Ql#E1hR8zVuoY~}G*76!)-?7l
zpS^y;)zuZtIHHvV&1~yg^2+`!nZ9w%F<@NJa9O0=6~p`EL>DR{%RRk3ocQA9%wK%<
z6E<zjB^9PQ@i--pV_<(CFjld(2B%?~MlP2@hSzBkR8%TrlR>R`BkACEO^?xrOeCa$
zE}FLQB$#I`jW@)2WSS!>9?_<!^?P#3jMIsb7EYHZ#yIoa5AXTi@%#r}Q-9P@XIqc9
zhHcm5SJUp8qx7H}O-+@ODrrunB~gm-Z<>wAchUhkj*)q)Sf^<9MCTo;)0oa;Ix9^G
zNv#R3l{3_}sIH@G%_&S6W$3J5Q%;6$*Ko7n(m2QCabVF4hrY*X%RI}?bk~cQFHyP7
zf>1b(6XO(#bKvqcuvM1cJ6IN$Qfc43WN}+UHjG&IRk0&*JU!7ho_^D<M)1OZv%|ZF
zWuBRb3&u-nyh@4IY8q#0nR(GiHfqMI1+(w?e<;heOq@?M03~F3?}9uu)jAv1T8T^I
z>2cz8nz?m7DO4blmdv;;IB#V|xXc_+XQoR)xkAa}c}_t-pLqyO!@$S;d&ZSDWQ_%7
zn5M+8YI@&MOTv0hF_kflV5+Rxtqd(;3OxNVaIt~C>A6gSSSzNre-clfaL*RfBJ}rK
zpe7~z&nkYGd70^2MXfTdq*SPO?Qpa^Zf*`3BAu%cDuI!<^=Q`+!%SEvOt<}1Gxi)f
zC^=)TLupG&GUzFzQKjOIp?Xga^O`Udc&4$In#HC78f#ZRSI(rmg4(K<d#N(mwr6+!
zk`hNwAHE}&LhBthORbaE5P}SCIhRlA(hvgPSz?N7YeQk4CV^BI`DfZjRFD`W-b)#D
zs0H74g0xtb^tu+f3s5-XtdmmvoJmRG+NDU}rnP=m+vG5%wL$q7rPQYiHyK7MvaK}h
z3VqFJYc0LeR(yOpNvDCa5Ef{<9VI5TD(qXuI7dzoPZS<#_E%iLdLv1@Bz?UF@p|R)
zi87sj{PhM*OT<`VT-GWWo18@X)LMbD^7&;^WUG=v$JvVUmb|8ft=7^=lO<G@R~1A)
zS3}!vXtsNtYj81AyCl|uBK1{zz1GMDn~NCzt;Jd=TVO!CPUT?PKKFZC@k_FXWpYaE
zwJX(dI!001yeF3+pS?Etc7tmgslUr<wQ5P#qqP#!sH<nKtihNxCN_-~m9SK<Hyz^~
z7-GUTh8Me@eWQ8z;Ujh#Dcc){X%s}0^H}dzrLod#&6hVV!*BnVo5PmIA9>L?TpzY*
z?V$#YwuDmX+YQ!O`My<AzvX!$3W*HkVsntRv0Ax}4V|`7#78!l!Wff4V6~=g8yYX~
zXRT^|4`fx0);KGzK=UHLEn|QDK|-xVPp-1oG;39dR+^M%w36UcoM*b$bG6^F>o44g
znZxxByY@i#JF#k{w1U-F#jh4=)>(g!wX0ti8Tyn~<d{Cu5I`Dfs1<9)Zj(x6yXgr@
zp-aLkMPohZ^NG!-r)e9y*0b#!&ZiS4MK)cFR*I)3P*=-FiW%!PuU_Bbj3O?H%jv??
z;}czP@y4)(6)DrSbe%>!!{vOTZ99C^VXWqMx8w2gNSG(Kn~t0#V;Dbu9*4t@#g2Go
z&>DtVNK=p&3N2mLF^M|uj3<}E=``_pI$^z~Z#<VNpp|9Hn(rP)_MOFBOVhx^<AwM4
zk6a%P5>QNu@gnMFj6w%!z2P!0;!jrM>zvL5067M%bu5>G+O@<mVJK8)mdnidi)Wlg
z)f?vtWi_$Lz5AIX`ggtlIHgW0yfIwwI_W>Hm0^kUAdNCS6b*SwQs}Of_{ow0HJhd(
z&yjfvv^&eiBt{}-3(eGEt!BK;ESJbKNH3c6mZfCQ(?qKjkMo5@W$!l(!^m{HaR2ZT
zuf$i?IZspJvzr_Co2{(6VwUc}QrS2M(Xy1vC1ql*BE5$~v$0%lH#ntOLclaV(d-Gs
zL<s>@&zr*?Z#LgDEm2S_=M(3rf$RMhV~qUmyKl*=a(8u2&XIY%;Em?mS8nV`DxPx!
zod#}Mxb0hh`0kNXDqZKvsgOcqo)%0jQbT1lIah|$#Jdk?J|1TdSB1FB*W);I9!IvD
zjuI1Lo;jUHrc=VTYOR4%OiN^(6Fi<dJw0&$@VGWM2o7PHBl9Bd9!?vwQ|u0VHeE;C
zc39(3Rf*zwUN~JAKD<4%?K{SPWSJtd7B;;U$^Y1wcNQHd=4rw?&#Sv@?hZTVWn70a
zE1kWHPi5cQXtWZT<Zi#CXiJWPvW)8uv@lK=!V<U~A1G-dq(IYc1icUzPNySbp1FG^
zsjRM6HVT~c#86mDWy+Cxp3tUXTqEGu`b_6vl?)WQ$W1e^57-idU_44DO#^wE(6+($
zdrD2{s;H|15pt4nf>xw;WlNDEBOxGWqb+qM^(zBtP1AOSBz=XYM&`>I-*?z<Bb8@L
zf)DkRAF0rF8>vM@t@TcoTqJp0mGpp0O_hUGKVdagmDD9xR@g?#DYYtvwvJrN?=c*!
z5nk6yPVtjJEXIVXnVJ;FwWMXjEKg{9LY{bd_y`a07{-B3)8M-e%6DuwTWr&FKQDa$
z_U(_p-T<j9-0VMbxil?d8OdQ51y-9?T~X@+02v~Z1S*VmG+JX&c;7OIAS!RIpX%(|
zZb!G>(eL)6u&o?FU6Uv#&+Fx!S<(W^Vtpt69BZUARSRIeROvKr+jf0W>ot{KR~8wc
zhH7K15C)X2+cgGb97>6TV4YuibqcREl|>FXC9f<w8SskMl$5#5Gk5zAYUMaBbOv7R
zI;@79>m8rJyrQOs@%+ROzxgHY*F9y)_}0;FwrDR(xzU=hZnlKWiQhghyzmCyZh3XJ
z=gTi%^6_aVhRSxo;r-KzyO%Gi*5g`9tJf;4?oTOCqXh+{1;b^Hu-3ZP(mO>;Gbv2y
z^85CIW(AR#l9{FnR|{5EVTGZ|rBYV-V3xQgPW||x1Z%&wHV)OuAy5@9K<ToO;(&FA
z{eH(h-=k~B8BJUQ4>4ls=sWpbQ59P(IRtV_SZ6*}%Q@#($hFuXBv4ie&j(Vi*OdO+
z1)LK~TduDTSStYTXAraWFn2g-h;gRxcY=aa>XWh=Qj%*wOjxDy+VJ|-E&E+ZjETpG
zBirqkzH4Y(O_*o&Y;o;|@iI{e#FXgz4c3X}#XCn#3n|U?eYbji8^*JEUP~&}nrRzL
zQ#D<G;G6dk^sN^wg@UB(8ql;96fr`Hg<+gf8m#i{y~i5G{b^*hYtQ`sBUgtF-@p6F
zI0tTT4fDKkK3;^yR~6^ek$xxDjn;~<SRse~WuB>}u-$FRArNxpV~tqrnFbkTb4|QI
z9T}Gh+F+c6B&aiQJ%`O7=HdLKhPoIN-Wm37v*Jf!m=<CXz#&zJgOVkgrPeQ(7^SuF
zpA0xt>6(VdNg@wd@f$_cG%VA?!v_gL?AF2Y>2zVboY`oIC2~%glRhzxnRFR=dY_n1
zN0j#b@Gw$y;my^agVTgr8V+(UWa@g_YRFY{SrXa`JzOd1*4moJdX{;SzD>27NMoWj
zg*jUO`A^<3E|HHPA9#3r;^E^Hcl!hLvheolL4d|3pljhW9J%W>FZ#+*dd4}Dv%$L^
zCM~?!G{hw{ukg&GG&L*cDdH^{D<`)YXQm-=d>R>-M9M;~PFXV!6T@YsYdmFD28Ll_
z9v5P%q>xD=G0c%61nM|)KAt!}4wSe8OH*MS7C9}2NGgSH({Xq6f}5*5<{|Um{eaGj
z<+Ko6&v2TU$3SBo$MYnEuG1K2P|c4w5gTnVuA{@QW!s*)E~t{Jt2sS}fH#_AMH)Y!
zPk8UyZ1&Qz+V9x(J;_;?SYVk5b7Wp5-P(3iSu;<wtVgY5v%hAuD{OZMl$Hmdp-_}Y
zH-^SK%rw$G_9Wfnx{aKGOG2+^VgWRQq(cSto5n>3vSsA>;Vtv|k$Jk%w$kmYtzk(O
z@1#MZR)tX(sdkX`bK~ra5fNajwi=v;YN%RCkD)fi5K!9DZCd8xLP!NC%NcS?__iY#
z3pG=gnUu1uO#ZF9DOIhkS(W&{wAQOy<jFNiqs6-VN<|Wx))}0Wl&rB%T4+|aPP&w3
z*njS(R7y$0*h(7DIS0#yFg?=WUUAJGn|6nHh72phWuc_NX}-wA$9Oh}H$VD%1K91a
z==XbadiF#L-Y2hNh5|SZ5mD=^lG6JSB2x@9_!zNB^nHhGTeNe;IHQnzT-)t2zM=10
z;Sbg-Xpp4H+F(Gg9i352M(a%1S%Hiz@hVy8*LCQMr&+yl+E^(GFJ-j~h=QH73_V&K
zN!M5D=dD&0IA!!Fh_POw#M&fs4CF9N$R}o0g}&>UE(1@)%<J1d+fB#))0vtAudnue
z`QnD%wqd{VEMsO#@btU4>`#%Gdr$M`OZ?_Df#w=bqbr?reE)Ev-}J;{+3q4$DV|Of
zZP(InHuw~Ie>$U`=d@I;Z!lD`Gr%VwU@Un}_gb393C4wlwhGsJbP(Hu)&}pJPpuaz
zE}YL#oX-=f@3=i|u}%VDC1p97RVjckS&%_L_Ej~~zZ^l;wY$1TsEHgW(lC+7kxlEk
zy}jk}IAayGP0M^fVe}(P8(g<r!$=8DrSV=A-*sp%HA}c`_0dXNlTQbTe!D{h^Dr_^
zGv2_9+dWq|JN&i-FDuxm644mq@l1A>H($PCv)M3C7rK6nRgon`9><yc^C)aKSNZ&Q
z&(*$X8fS8;?DiX8zr16f#3NTTDD4Q#$S{x8nCKhBW-ES9d9y0>oY-%7kYNf5LuS)?
zPR9#I8D8IBlk0*hl^CFLhQ()WD)eo`>Aa9Ij3Ht*T<<!rw;O7zBx^Yi7qm9aG4U=1
zcB^7OwuZNl4}5=|*;|dyk;kVC;}`@{q#;ZbZQ}$Ao(f@E2vfxL4KXDNYHTYdFJLUc
z{_auUGXvOxMK>%{B;`aYGsZZYrla3>bV@PC;Sai|{<xro)->AD81WjVTm-QZ^QW#v
zSwF2cLt0K8aY>}5k{XS6j$N<Cq?$C1=_swkd&l)=!!pgh|L#PJnd+>pw=M(Jh_c|j
z23sM_3ngn>Z@B6lt<(JQbm90kl2hQ_?|wr(zvtWU-chR<kPC&UY2Xq9NsI@4@nX+#
z9{KGLgD{Rd&r&nzII+ZpwUU5a&W|{!$+l(FZuuYogP-$%{(t|Pzxl;CEXSGQ^+-^P
zaT@q||B+M@j>_G}^5-`P7S%E@l^mp@pgK=X1y`j2J|(0oM_VYOa2D+#S8`0mkchK<
zDN8QIoEWEt>Kx0su*}P+o6kH4F6UWzMJh?x<dTR>#sWIQbXs_Pe<4~!S|TN9F2{+I
zDshO!rLfr<ylFAYVbH>Tn=+?MV82V`D4vX%Gf(G{5HpRH)Ti!${Axm)8-&PR!orl3
zq&T}BwaR-G=Sg7L44jS+{PLIo4Yd@${NgKKUF}Hog#keZn5L7=UkvCR=$)bKTD0pW
zl(Bl+yza;;5MsiZDktM<VLCoS(_nYEnCj@aw`}_@-FCNfq7pSLjFMo2GDcJ{Er^j6
z15=!s<};5E?>K*aEA>YEovghIqzu++LW%<VeKyt_MF<PVDvA>5Yb`5*Z2g`UrO;ZG
z6j2WiwQAb!j+C<0x0xagwh(cRrz(p@QBx)+ff*|+W^ri|Q+U@>tEhHPS+o<DR|=D0
zj8=83tWaL96*Y@l-C8NIm;cF-mj%|Ql=M81XhXpetNi}D+o9)yaXez(3wFC(au^6n
zcv99X=6NQh#0t>Ebw8f&52D6e=Kx9Fl~N=Hs$?Cm0oQmL@@u6O1gLC`%-*@kgBl5#
zw2c?Opw+8Sw!u0t5As}4juqaF4DrTE;u*wu)8<uugT`63uEM4>k}!N$pToR}N3B+J
za?~tczgBB$YTy|=14>m&NvJBpsZ^wwm9iS>mHIRUO8r|Qg+<g!))tbQc{+_eoM-N?
zH-wb=_+jDg!-dx`uX%H~Bb7u{iXmjq!^lr|nt#x5=*$BT^EH?5AP5_+_%H?Dym`Ur
zKRrO4Swi8r4;S8kJTVR<D9e%q`@SWuQ2?!^yVExf-dW1J0#i|fB@%crE9&PtIE4&B
z-eauT3eHQG1bs=A3Ui2Z&#7BUJ)3n1N_uTvdHys16l_Gf=CPJ#S|o9;G`gy_`_vGe
zA;yKq8NU4d4O7zm?T25ko!^2gnT}5wV`$vUBF%}IB06hgimU%NNf2sX<0-9an?|^C
z#*mgkT%;P&w~p;@hqlt_u<tq<f~}G7y-qalhQoHl@$pf7rYoLoUINF{h3&TE_05*+
zT}NwWgW;NnIF9t24ad_7qcxk&hSTZD@pPot$~cVltu#6*BX}w6q(P+LZprh^v@BdM
zGc^UQid^4b(Qi7GR{ZesBU&q#7->7loHI3BZm#yk7&%Q*LL5rIbMw4#J`I@Nj>q#!
zJiCoh0|GFHgffcLWx!bYV$<@whnX=XcAdkPf^xz<e0n-Fq(mr+7#F5#WHTqmP&i%!
z!3T0qBr<ba2v3E^_iS3n)m6v%G~k@YO8lHDOz2rScYo-*|Bo8#ZtfZy<R(?9#285-
ztp)26Cx1?N0Bh1F6Q;<8fmjMnE8T(CY8q?leGhHGnucz>A;c(5u__mx{5MfRLJ^ir
z<1DF)@1t=o?WU#i213FdJXcM}o0kVprz1Z+o~WwQwk=pVF9V4J4ostYb9dmO7QRK%
z_6>H^ke$MA9Ag;S_c!vz&VkGKzs7&|hU`afx1N9Zt5^JI|HZ!{r@+JUfu+LjZpThr
zmg9-n*DXIeXo6~(gC>WCa5<uz4&BQ|6GM>nU{dI%;tHe?rBK?fjTcFx<Ep5YPOZqP
zFpdlDRv1pT6r9$?B{E!QE|;0Gny&MlNt2|I_Pd_Lwqtx4`SAY4?&XfKOvDfw#}RAc
z>G8tH4^L=snT9~<5^+hQ{KZ5{nR(8Huol<nN{$&->N?=B7>ghIIz7I})QHhye0Keg
zZg)jv9OJyuv>mtCcMRt<^TY4B+V1)DpZ=WTa>mq5$eM1mTci6K@4P5O)<A7!ZG-yM
zo7s55!Gy4oQ$STiN{aI+Yf@4oDnnBh)^zOeKBv1rpsf)OKrIyQ(AHv<_~w*Slxo3P
zGGK}#kAX0a%+n;*P%*MQToInrhAXDUIk{;Us(_4jAy!3EUivpv04;;#G7Z#JXuPN7
zOw%;P5V5Tzqp{jjR*0pwhB!&HhHivLZ@pdd0f`VswCY6tP*9bkX**(ytP8G{LC5Mt
zv@%>(K^2IhUkvNOicG1Wls;7@Oj6c^plMo4)#y@W?b&QG+L6u!a~ipXObrWd*RU)T
zU9Hp{FiPPYPbdYe|1jG6-}TBlu}rhzX^@0%Uf1Vp9a4<}T`eIMk^)nqX*|{#35rNn
zI3*`(&oHv)Su=^^&&!$|#Dl6(pY8+7+O>ea(sUhd-wEiq7Q8i}%34~lKCK<Yyim!|
zwt|&GQqx_tLp7Ezc`={Yg433y#RnKdSVJMJ2QA5(*sNO^>l{WqQR`9^Ag^Xsd=gik
z=NI37;H%GGbF<r!C-Hwe@5nK8eth6zihTQc;vaqal7IB`*X*45gC2kTE1o_ELfbP{
z$FB3d+BwGLX|$#Hmb=eiQ2Ha^fBQSayx^4Mr*B>mL&7|VeO5oHHdYjMm#IOcQ(oV0
z@eykjDOctYXbghUL42-bD8wp|ZDTFgSmE>4x`uoJqtV9VY{9uih>5t&Kl*x8WerrV
zB8WCt!Z&f@GEYq7NC?XcdR}<=@R6DVx7Rz~4vC^Q)>*oyCzni#vSnG8S<=ha;BcS9
z5ejfk`)P28-$TR4^Mx=>@}6bIG+pSrH{88=$?-H&i--cPcO14YxBH&=KYWkz4JjA8
zt$5!`o!Rs~yH3+uu)5N3TCDXP&m*Z63I*4AI^S?UojF}DG+oC$PTbwyQgUXqX(;O|
z6GZu6!h)`qWr-|v<n;8wU;h2CdG+Ro;c{dcXRK{`csLSM!dgY&Xb$~>S}WcfI#)?W
zzK1yr@}=Kwd0G~7Em))3`rm`bkMqLQICGvA+eR_XXWVgNj*6>R(P%~I4Et@Nlt`?Q
zmE}AKPM48>BLMoR^T>y&s3nmx>>GtM8R7}8_gwGpu$P&(F<f2k*!B%i$0O(C888xV
z`2${m+)(#ohaOk;K`EA$8DpXpsn?Mk_49xyyZL6@Nd47mL|f3=@Hi*>t)o*Oofk2K
z>%^=J=krOxgbHDw8AGuKtYHodHOpSV7{i<wbSYd=1f?0g#@{yVHXF?Ko_U%GSxl#;
z6tIeB>+nudRAO^>R8yHkqA`ZUzQ;6%Y%->Sr{&CEiy2wv!s&Ov#j>!u(bV=8KmX!$
z{`p`0jEB<+*E?cPyxcVGoaT$GhR?df911aMVxGCY|Bf`C*u41(wM3TlM^2-t9m||J
zp9fOP43~j<o@mw;rmhvMHLvy?UR`e_v0D_XfU}BH6W$oSwuB`y94E$M5zu5Y<T){2
zW@;*I`i{;zzW?q6-+%v+pKd+lY2-YNoX-<lYaY%c?|*ot?OVq4#MB1z0x_&SLNt%(
ziE%BB&ZV$~OfChILS-Rg|Jc_OCY-*)7(f}6Doo=AO~#uBofT=W7&6B7D3#f3i}$Up
zk@hdR{_4+!q&^&(rx^oiq}3r8DR#~=5$2JUgRJL`Cr&d@<A||_wh^_>+`gfC@j0sL
zXsV+%ka#vFvv#{WU=3TW>#k=tDlUOEj0_*{IljN=av6v*vpZ}teqs)Rezy}vA&b}3
z_>QnFk|NbkIz>&TDvkAK?Qs=0m(}_cZ>rXUYdtX+s@7}2DAXd~S7)VJz!}RF=aqRU
ziJBB6$ruz`+5(nD+cxrErGj>AdR4D}&m#S>&Usngt(-ndfyMQ{AgBxL9JL7grIfYB
z1O+8i=L?FC>N;Gvp-q{h49fTDo%BLhB`fQ3xU85FjqCQ5^oJ$G{;t<N&S)iSr?VDk
zrCA`RB;LiGXuKc^q_qLGv#a_c|4vQgWXQ8x)}<?)xs<}PM6|A0CmVWIMJ1YNsjE{)
z6ULETGkbNrrlr-V4aoDl_c@?7k5aAZyrOFxVFa$}MWZ#=YBsecgoq;Jw3BMQX(q-b
zwP=l}D&e)|C<#gJywn_`2+v;8_1Z>(Zw+bF^2=Yp#k+=mzu|7bMVE|Dnd9TY-+uFs
zzkP~W=lSx@OWqHKZ*%42Jn_rN1>G(vYY9c+yN=&BTl#Isp>tTH*lxCTU4wb`6YS{`
zJwGyCj!aQ;eI;&2wZ6AG=e3hpxSp)<4=Zv;o<}k~8>iNe><T4Qav+8%th-t;R?GJd
zMJ-B(rLrzt|1BjmO+Wr1QCBolDR~8$Cu&aXmZD&d!JuUjDw0$^9gi49$sJRMm?&Bf
z1vv$&J+q>K!pgX;MR=MOrg>NoEQZGWPw(Fng5ZM?m9T3(Uw`$b9Oxd7>~>oozKblE
zg=8!5e)k=IzoXk}@ql4irVGwh-n_aaU5+R+P1kb!@+I%yePEm>?q0m0ZDpul=9!|P
z+wGW#5o4_2nX+aaCn$+c+sgC1BC0H!@f;W~_w2VDZtnKv5}D@2+jmDk9tWPD&TKXf
z&T4M2_I!MQ56a-IVb@uzEb8Q*7&4iPsf7l(-tVYnQpsQh)09#qM$Kc$44np3NKx^$
zNX~$oX^s;q1gx?6rX%FSykxqXXe<O}as7^6-*Fyi8t>?wrBRV;9j7JIb}h98hRd1V
zrj=E5fq6-={C6<aDPsiGssvwhSr(RDK^vTw{y^45TCFnBZC`8&=b3q&SuPWbhFS~L
zIdZuV?DnnnAePGH&Ky6C<XMtnjni1A=(VHwj!-hg6wo0NW957riLNrUMLSDr45dbb
zPQ0HNe)aJp+_yC?ro3fywUv};2y8bEippgS?7Nmt?<hn{G2|RM1M}q`=kCx>s%`GS
z`<^eq`p4+QXSADZ{^S4Xe?}~hkP11Ri1US16iq(y_>2FX%L1oz=3oC;zhKi@?mqug
z9`b1<g}~D}u-`Qttmehm^LUzAE+^7-W@)6)uCWc9#-pw_yn4CkS8qR3RK~ZOo2yPd
zS3#=Y$MMX=!-;tgTx}giTS5pt9nV;8&{h#c;<xXgczhao`8e_L?#R>e#N}~hj+rGS
zmLV{nW`@(m)GdTDaypMJK`yMY6k-tFCY4I0;2TS&&=`aJ6DW1A?Wv}r8pRS9Op2s1
zk)}b~F8VExA>w^U<DWY*HNKIDf;XAv@q2KN5)*13s9r$JA&yL!GgUchtneLLHJ`da
zjkb8#V``!698*fP*LTu^Ixoz&V6w(H9e77BnY2WVk_Uo99>Tc@K(dxfObe&ucf9-l
zS6t5GjkHeMB4U&SK&qCQAugFh!gRu=1HB%e1P39l6>>AFO4fELMJxroVi&q*gL8@!
zC1F}}UE3jqrmw6gt%sy#oYt^|!Mcu$u&wm!4U|sN6@Or4O;wdaX)7%$>bbvEDx{=O
z)`$UIOVv`&K{}3&5%}&W^~Ye8W}XI&FO;~fy`2U-Unt5`yu)>#uy{}wzq!UZM_4Ap
z>4E#l4?p^P1F2z4DWaWQ*@f!^Z%*VmQyY)f(xEA9<<+-V#73c1MeDVx06|}XG-`la
zTL|>(Zj_c2v$D}r5%fk1fmD(xG{&q+^11?g*P{6<gJ#Kzv?d0fZ!nFgX<FL0!x&M2
zo|BHk5tLN2HJ%t|2{aj5<Lh<Z^&DK+T469)*PuKYFJ0d$1YB#$eaFvV-coAjX5+YD
z0@D=b_s<LW<HB*S+`W9wF>1cK%v_@8VGay2Q>O=ft&kSlrsc6H+Wig7)fJ7-JZgpQ
zw!HZKbE@yK@xnaLY!9A8t9ggC9938$>7umB8KF>fURjR@t1X6z&eAj!N`V}Siy%Tu
zTA1gs4!Y)(-&2y{5-=!b#S$3@On(xkF2#wQrFzmniy|p74ri$d6t@K4w-{TPLtq$3
zj;D2nSfo{^CfR1K<el}ix2wP+&~nxgOo~#m=e)GMNYY*2zg%TQP)p*~ix+I$j(MKh
z9S&UIT=UjiAhQe;83nFkm_Bgx;st4m*iLbC-7{TgmT}_x^)-j<Yo5**hG}GfIIO|1
z%rZ}CV7uM2JUx-Jgg{iSoK7clj=Z_M#hFz>?OK|y<9z=}jFHW@XWtoy@e$v2%v0s<
z`!jDJ9_hP|5G$X(dC4@K*l)Ku-?6;^h}Y{di}b}0d8YFXsb)HB*|r_`<G_-nY(Z(k
zK{cIXo(*HFlx3l*hLC_pW}28mRYNR_gn;3#mKsoP4V~3EZP@ocZQIfvHZ+Gf<f5_R
zf{7zxo*0J{`(4lXzkADWyIt=U$hPqhc>PgB-7sc0x)JtLDICuOVGcs7XZ3~{IlY%$
zY4#nPn=Mvr=BJS`EaVu4&R;7N6EOtZji=_qF;~XNK#2vbB#qM<!<$Wq^PXv%iOYym
zq8^OXh|`TIc#9!~LN2hx#CPLI-oHPxZyQ=`=vrBWw2f3vl?r|1X@O;t6O3u3@K@JL
zSt3K4IjbKq4pNR3t+>9q#<d%i?I^WSbqhB?qp6@D{u+8uE-)VxZvUG3yYKkzuYbe&
z!vpPScc^ZM>G!leaHL`59Y1|@%V#e(w5{c@zj;rZCi;0~vlvYA*qYJBQVAUPp55Ma
zS{7#Fii#-|BZckbyzs;0N0hF-ezjq!ig6y$DbQG1fu#a-Slb8+3{L|ee|Y3_nV6OU
z3NFLUbQzh3k>TlyhYufld>jZ-PBo2IM5TpMrl9Q{*LzRnG%v3<oF2zN=F<9aeLdZO
zq_G0B_Du_+N?p-7vdj~k%}C$(m<$geFF0#xyrXf37z4}sfy@0{LW=ae1FqSS!^H5o
z;H^VhgLVR5j`NI75v>KJ+a0di_gmuS#M8Uq;hTo^_yE4aZVq4#$|`~BCQVgU+C0|g
z0A(DeNUd1Om1UV3&rjTc{K)w-FfEI~JFP_-&t+POIj^g*%4uA<?za?WnZrb`nZ|i?
zU5C7)2_>zbEseH*HNtzWafBSGSq401WR2*YBg_fkGysx(UE2*x&6py6uG%^=@M}X|
zy-5INSymvgMr(()GQ2zw8qXA7rNBE!<@erNV+^GlN)=Y=>P<(RXCSsD6?4fLXNgg&
z*XrfWaDGcQ_nh9zpwl-Uo2!?YX2Vz&>Ga6M$HyOiy#b%F4@U6`=qv**pX?>9JkAs{
z1%<MbPOns~SrINJf9g_I&aFUasB5hg>KQX3OuT|dr9_$+jCNweFGBsVMb@lJ8NBfr
zt5K;4+GSl4TW8p88hqPfU4zwn4ZRfk&OV35vb0z$<IpCNN`Vko0j^?|MTSpNMyjf7
zBZ;B!J2aJ`6;IzjaqTq!gTHr2ls}MZY;a!G!t)eJ)ndH^-|_ZR7#AalF9)UwjbG&m
zH4}z2)3PuRM{0P@VbkDMqK=Qy!u3~Qvuh@PZm+og?2e_g{K^=?nXJQ#Gw#y>D=^}M
z)rPid=)A_{Kv^;&OoDm|3ngVimaIX7TI$LcRKgs5ZWMWDRB9xFY%I%K1NmcKIZSA^
zS(BIxWjd3x;H>Jb#pmlCVOmfWu5S-~l@+J@$cHeab43>^pfAXp`FX8sRvVN?*ouPX
z(JQyoN-6!imQ5j|sCc8PC31at;Okd6oQD%`d&A-8hHvd}IPCWf5AT^{V4EA3^BJ?>
zF^9l*-?6{BA*2XQ?5_8`e*KzhoLQEMTC@z!%C1(dj<)fPgD_&H7@#nm&m49;n%2^K
zIk+j6={iTLQp>pAb{Lg$I+K^cIA=aQPCU(lm#<&3b&hTCX}#g;<G_Bm<uV5ehDdJ@
zN?E2^3|>~56|bXEh$nLlnXArFO5kd<;l>({W9I2HQT>`iUsJ)N#3vdr-@RJJ3Zpf2
zjYnCel99q3CX{J8?A~zIzTj`ez@2e?es{-*4~CRxo{nc4*ML#<ZHxDsKh(kbM-6o~
z7}~AJ8$&6X(`Axk?#6+FT4gQajHWh*uJ`zAz&qN`bD0OqTu@eSjLJz5U>K8d&*sE(
z36L_ju_&YIy&<N;DF`Un<;tct3}YcCNdSI050o*GFeqIaYGgW&ggNt*yDJWT%gyx$
zZCCzQ7OtDIh8h!LlKw;E4OUAudOA;BPG|0P$4|a^%_+9L`svSj@p6OlEk#+vGE+^5
zG7a_d8|-wXT1UZv+hA`#r<n>*|KbNRHERV;hqsWAPb^up-3mMD)vG;YEwFQh^9*UG
zD;+9UN(v0MGR2wQm1n#2jq7LH00961Nkl<Z{Pz13$7N(LGs=h@JkBQ`jwjx}{eS{q
z96I`#v1#P>ZqG}<<>PrF6~)$D_FcoqSxj7r^TciM`RP?h*I4>ZN2@YS1Lw!e>HeO)
zNO+)D_-xm(be6yQ5HM6;9UA`Lmj}Ae(OAd->+#30!!8dW==zqfY5DN@$PXW%@Lf-*
zXBuazCGzy}1H0`8WgO>ak)cFuoRVQ8g_);^d)|EYHM+|@zW<(a9N6uyu<e#PR($7Z
zwWd`{)DL6m=9#Pv!|9RJ!#iGo@dZW~R5fBA&lMdP=HVmNdhB+GY4>ETnE7k{<R!~F
zQo=&GoLEL-Bvz$Jz>+c{!+BV+&Jkl~UM9}N%;rW^>WpF;CfFLRH5g?{HGcB#Eh%AH
z)dl!;)20^UGUJ+_5*M+(DUETS6eBSPQTvoIo~+g^F{0L$aRzCJSl2kld(t8;Bi0E{
zrWWCv)G80>k}Fnw8O+jJJ5;1hU0X+ti#&K#UHMrWW4#P@Mh++_O)bKH>sm)P4%O^g
z!i<g=Y7C5*kC^ERRWsFFUVi@NkG|eu<%Wp?ZcU0-NuwdZ|1;=XO0HP{5-7V;Vr+ND
zV6+l^#HyCHtwDIN>l&?QO0j}M(TTj)vxN}FM`T3n&M|%BkJYvj1Gg0hmsS?N{yw%u
zItxwX*1=QQXB5`91klEjt0d>qV7$Xt@l;jhtf1Bq&hy}km4``P*K@7+?E8inuMgCB
zA9=IU=)*0)`R0*dynW#B|MY80jf_hr>IS#zX}3F3Zk;A8`J6Y6=cd(M8$(~=>9X)J
z7E-Cq!+TC?&t32Ni)&3^AGvdHFx?HCeaBB;d(`HZw}a-J_YdOBGzRA#-ic*JkP@Pt
zw@pjy6)b@?M!`O1sp6|yyue<Acj5;Xl@;bC(D+8G-l~Ly3<{GpMi~J>|HxM@Ss0IK
zYE`7FrDm%L&dO;^P9xJWP_^am#SK+A{LQ;F^E7gqGp60J>3ZqzT~p;+>rZP{g`{AW
zwVx~@ZQDv)P7)94vWOqjR5_UJcU#U+Pc$3H*I)gd%M{i_h#@Y5x~VBK4;Qw5k1B=3
z{z|T~Wx+VZ=U;p#wTr;A1iHTC>iWPuE!2`ZTwO_uco=a24ngt0P0P3}cvac$wpeRu
zJ7F)*(|~s!IYhJ-_9Lodtywveie*mB!^o?fD{gNOkTTOWFijcfMaU6vqKxR)S*CG<
zBA;o=N$MaQIbf%hv0D6tsunhVBPbw~*tU*o?aWQdNoXvpPxp`KrWaAvA(euuqJR(c
zLac?$<&l5((|^kL0`JC|yEnJII&?HvlG-0XyyJ2n(7JvKM9I(myI+6QP)8_OG_7~U
zkQiqva4svlp;BRmuUcy{D3+6SV{Z06DOHxsj9wpja#J#&k|@PimbBh1wQyzT82RSI
zJs*ZZh=FZuNqEjnB9%&rmE)9&3XD@UR!##C4<nc3$ct-B(`@joFKSvA&eOu<(|}i%
z@iZ}B7H)SN+OEM^L(CF9nCHyD{Kc<%bA7|%<_*1S(Pac>h}ad5>(F7QJbh0pNm8N`
z7(?5Zo0o6+tA9^%|IIs`Gc<k2`0x%4yr*zf6{{<wRYYTX@%atmGGltbUVw#I<aC@g
zG_9lamRc$<Crnz<RU~U0GyO2|nNhrWvEg;=sGXy8o;Qb%^AvbG&uC+KbG@N=mCtW`
z_O|lVo1UK@zT{@p(0E5Km8BG#8ZcAlm5F?44ZqD5O5wV-4Dpi}nLBIv`F=y+HH4he
z>BryjN~x3-(cV$Rgi0eHzkko0mp9ZTL&W8LqHR2`%J87AMWL9Ng;Zq?)8D+L+3q=w
z1GP%3@B8n6M=tW9e(~atrDld&>H8k-J&Yfjrv=E=IPmcP9ZkQb-4{%^!&z9K?s@mk
zZ+ZFE*Qmx*jjYQ`VfDb26^FAho*ywKu`J?c3ONhx)i+Er^K>4#ySZf=Cq90>$LK~>
z$rW&!O65FE^j(LwD+&V{oTwFJH5vJNu8IewRzZ%~MpUGd3dUH013NE8(QEQjYw@pY
z=_mViQ_@PU4}2vGUs@r$TCZ$AQSwp_)Vz9+s{ORip(bhy<TXL6v=;NZR`N5oLpMFn
zwv>`6Iigg>l!VF=?+uM@Nw($1%`2A6g*;tooF~PFWf|yS-_l&&{^;usSUrLj)|B~k
z5X3rzso<T#T7}gLQ!-0TlBR50!I`YV6;*{jS5%c1js@ckrkIrvD68z43Plx45(sX|
z5<ar#xjjQThf0Z>8FRss)_WuZh1Ou=1r?sCU5B#mry)I8;e{$KgMwPWA93a3=>l3;
zoNG`<eFD)MC7~AMEV>$&5E;gaQVNGdM{5*QiWuE+eD@KbGbPRZ-~Z-s_y=Ep#$!=j
zbi?-c4%bUnlGXCQwM#(P#7;+EdCQIIc(bv5e}vXpmI6P7ic^`NU0XB@Zg`J={pZ{t
zGydI)!!>NyMv96;uj-jumDZ|+rKnd}Unv!0B!|E<&P+L@bU_z^7H``I=Ny+M$bE+j
zO6(KLdcmF)U`0=<T0eo|fAlq1GTK`3hFmt(WrnITzHmt=LI@NJ##>^6oFYyuX`0EE
zxGaprK)2b*@P>#cDW^~22mum5_Xq1W)m>QTiDg-k#uup$vrbraAw;gPZ`o`%y#M$>
z+j^dxVZ5`%5U5&_#*xj7myDNbts{dspS@wzcf?`f>FJ0unw#5eu>e@Z-PH?9$%G}y
z@Y;B8@9t1h62M9;w!1ycJh9vC&{T}lgtbfdGM=$o<4w!M`NTXY4mW$gy1SwEp5y%^
z(|AFlxVgFI;o(tIz3Y8*94_qo4(B_*`{5m_7F^@F%z-gV@xi`zScI1fnXC<d+mox|
zG>qbZT~;rm7r(R7QjO-kmG7rDgq#U;l=hM=3{OSFzTL1-@R!%G>AH@~+k0rBrpoQj
z4bFN>PQ;LD+GcGs`nz9$)KFKJjH$xC8B%6R8Q-@UXQ)ZK3FnxIF;TT*JdMOz>U5gE
z;qY?L#TwGIu<bqFw#95^{YB0^emt!Vor*FrrO2;8JYkJxDizy0$Qq&|W<f5DbH((A
zsx4`b=vK_U!;qQg#Qk~V=4!)IGc7`t{`hd@`wvfCwVIFb&V-ohds%D9&A?#1Bcb^I
z>CCUb`+<LQc!ACrQXCmC8<bn<+K%Ejm@raZPf;66n$dYC#|d&kDfr@x*L2+h-8eQI
zLkSCi8!e9pKBUA>!Gb&-ZZ|D`*K*smG+L1hc%@gvYyoGvQebkWkC{6ZjS`c#HJX>Z
z17H55j#>mx8`sJ)ofh&ev~lY-SG$gvd(YLb#ni~nzDHRZiruqP5*Ab#aLQm!rO|88
zVl6DCV72D1cQ~uL@s=(t`c%;&qy7Z&VcTt~rSksWTb9e2M#0^oXIW<2wxQI@X160#
z(Ira+-wRjj6rV^-U>FzbxKOmgTX|qzU0>0*9pf~Kzw`8jvzBw1I9-ltrSaY}hCt59
zN%p&YoPyg|uel4Ee!rtkXS&W~Tq}OE63NH+;5)KunPWti!ZHuc!<n>9G~V;-&1*h<
zyk`yzm&=H<GTcekk>+$d<Gp--DJO=@K-0EzL-U0>M65Bi-eC;FM2Z1tWpxswq&SW7
zq?Fdbw?G0VO&qOnP)bYF0Rpi#O4RibA}Or{mbTcYS^Zdo?l5b4cCDF|g`#eZ3_m$0
zKpH@-m31mbh8ATUwrQmlTpP4?NLJKfYzsBZT31)XGQwJ#P0^GH_vc5#G-665^*d6n
zC^F0AiOcY(NxvFnq@Jo)@%1?`4=VDib}Cu1mxNP^Y82W#eAn|_8UmV<M``=e(iN$y
zqH1{$pILCvReH~%GHrySuZ+g7NQLM1pjzS1luFgI>Q<{S%qkgx%5+a^w&>2SNlM5W
zw3Wki&XNe#*1|eO)||<CO(n07gJ-OX;4I|l<|=5VS}HfT^5MffzWeq)Wg2PQhQE0?
zu#`f`m3NmT?=K6b-Qv5BQZ<|1o<@m}GKPf`7v?nc!<2X;F)WdgGC#St{DU`pPEqlj
zQ|9TiaMc>(hwpfIQB2((zj-+G-}_n1Qr5PI^#Jm0<CP`}3H}%a#2iw@u07gX8EQ?R
zhB9L$6w_!8Mspe_8ID%vt5DaWI)55^)gSnP{kvY`aaEtz>vAtK%;N=Yi|KZ#nCP_Q
zIL@4gk>Pw{)A!un-tzG6dx}cLWm!L82?h!ma#d_|T5S}<!OJN_jl>$5rirqmh3a}O
zrjRg7VVvXDXP=Woz<WzB5$8SDd**3ed#yFgH1PiI?_^*p*Yx`hN-3V6p7`OL-=U4-
zXFvN|z||#jcXx+3mho~SE(=yGuCK1BVIho}91Hz^N85HRvxI3H?@$`boS2t|rtvh+
zG7kd}#|y7szozXxH+Kh?X<-QqeJ_Ez6vb1x-E3Hv3zy4<{jTBWuoFuNl~e<VuIIzk
zk*UDz!=6KLNjY=1-9gQ~dpx0arE{A0uvvk}ndA8)+mx6oRnfK`&I)>I2n$)XZpIWD
z=9!oZZ(iMTeSO2tn;o~_a=e@<SBiJX@1<>H5;${BnQ<Cv8b{y#cK{!@>t8WM#%UH+
z&0CtzlgAn74QWkz`Nok`6{55eW29{yrZJ@A=x?@MUGGT7;<|=q4urQS@zZI8bC#G2
z=cfr{3~k@i_8V+tSGxC%N|iV#mP?}D7-G&OBYlf8B#aU!kJScc4Y`UD`EofjjsrRw
z9v?=WHng2*Nr7P+vDQ&)LAioag<pOD4gdHjH_XdO)9v{FyYG4W#W(zufA&vNwj*Ai
zP*pLH3nrdv>Oh&#q`06^IF;F6ZLn#goImi*oCu~x#YmiIQg(DP(I^?Hc8z0WJ<bRP
z{5j3o<VsisPf}B683Qg=Hr}lT$_m}L?0SjG&y%djLzdKPjG2@w+G&ibbd5t9!vgpf
zvKV6_lthUEV#H}pR*DcJmno8qoS62F<?Gv?Kfi53!S$}iAUFQ9#I*wI$GvLJJf24$
zfB1nHR~xRc4%|PUxW2oEQfYj{I8U6Wc}>HPWHMeU!ZI=~num`kB84R-T2&!O8aFTv
zBd%>wtw$S=-L!PIAx(jA{^r+gI?ML@z~=gfSQPW{#JD8B{V)Cn{lETic=hT9-L~V4
zpZ`7Lc*Zm>Q9HurM7Vs5JKPbfk*?zLOc;)grzd)EaJr#z#V|#_{oPwakeg*yisNPC
z@e;Yd0<<QkN(dE*Xd_;%m^0dXvi6u7(1`b{m=<!C8n^QMUdJ2~x@k!Ys2QsjVVQ)5
zR10B=P!+muK0WYjtujD);i?&>q;y_MGl5dV^|Go`vV7iJi_%p}k;L92YZm9EaNZk3
zoJX`WB!qovyd~#|jg$0i)<UjeZ9^%6oCngJnF}mgP$Je8&QC`|2-pU^-w;kuKl*wD
zMAfq5uW?O_vC<(Om-$obQK<?gpeZ!YqZ;ubh7d$eO97js1j?v_4~d0Lls|WEeyD)c
z>Qkc9Su3o+T2V%b@>)S!Q;P*iF;Zg$qh-BYB8@e)jYF4Ai8IBgHSGu#wRR0_d3fan
zh!?azM?*`J&dyKkL`7Wz*l2~Wk_64A;;mv+XFi;tcp4UBQC#m5|JB<gzdJ>CeZy&t
z*rs87b4$}U#2{fI*H}`nc&(^fQ?=n?3{Vu8sQCVeCp3mXdm-%_SH0shC&Dmtd)Tnt
z{~O|0|D2_0zQ4d^B;lJ&S^bLjQ+rEH3H97JtmQqeN@2R5uEKID#3*b}sKQY57Pr<@
zruE-l25y=Dp3}5SYn*d9`{M!Y6-`wFeoiHVGV<&Aj-oT|K~YfTX+noWjAzC{pzkRJ
z8t;fmi-R>re(!a-efGkpQm9Hk+voZ}t)!zkgzyQ~Q)|Ukh0=u(7v6mS8LnwL9naF*
zqZQl3j>iuUTRdSJDY3F?8z_Y|PYfrGZ!M1>A9?@wEsZmL_WCtwx&GQsk5MoTXPzD(
zi1Wgx>nK^iPn5#p`Ub!4na%^H7TUg}>sl_;fVP%R--|aN7{?K#6j!?))>cS?r~60Z
z66m^)X_|;BVx40cMpBIIb~_IHmd0Dor;)3E%XaJOwdb<w7?k6%={a<YkB?`jabnk6
zuKI@7c`B9t;UJsDIbey5A<=rrFiub+`^}EN^X&76(=;=uOe`=gi*!FXt=!*VZ17FX
z&CQl22J*Ls50Bq6N8uJ4pz{q@E1ItPgRZGRt|m;&Ww<O0V_3*Wk=3e7TT9ha>xHUx
zU4wFlX%q-<iWAC+kF8KyLg6w6vP$?UX5QtppmL?_h50m=%reW0s20%ei0eFdC8Bqm
z4py#P@m4r<&-EPAcTMLlo3<gA%EK@**1+TGLWxp|HJ$=q8T!WI%<maYIYnG8w5IUu
zx9|At|K_iGp)>DO!~ga_{WbsK>(BX%pS@<geMOm%jK?4Num9!0CQJkW?mzl7wvB->
zuk!Q-7vF>Hc{^K<PZxGAs8DF9KshI_ljgeb(I`1x7Db^z6s03)shG-hWM>W6onyQ#
zj1Omg62rN*?WYx-c*_iJ*V8Bk)>E5_Z^wH=%Ivm|P4B5n6h6JC>vDy#L_7s=6s8Ko
zWJ;MKti|re(0k3R!-jg;tk7QJ=anM9n38||Zeo9ZEB>adE5^$Sm7#AtEQ)s1lc;RW
zhNfwl5bjqvjYJgQH*_1vwzs6JFinHg29}9j3qfmA&4lxretW<rXxk0hbbS3+f5p3R
ze$CTq<o5HQ@T%W&eb_NAiT3t0?jINa(|`3Vt~Z72SFac!A9+zThnJtDYem;WJbvUi
zzx|f+c#o>GicIr`c^*+HIt5MN@w;3Z&yiRaxdfih6D1eC6V6=YrQJkZCl~|UtOL|D
z1x|u5+B%$br0l4OZw@7^&Xg0xC^c2akW#>CCCtGXMS;_XP%1iRw2`Klx_WNX>c^v2
zl*(u&p0Jb_G(y;~wUDA1vCH}p6<da;R7v%vB*BS-bRvVIF7oq2DH!Wykf<4ufz-Gb
zn=5(>#A)H_;U23Cjqm8XjvNE2OvFEdX&r>WS!%&qfwC49DFvt!G;yk+l?=1iJ2F<*
z<MTjCqNvqRmgXWp!vd*jinb_g;kjz87CG#GS{cgISQI?z+!d5YMvFa@6cRCIQJk!y
zafZ%!bgp2O{N(3uUBoFN?`f&TB`}SncuixZ=8RPuO_nOX6hJAAZ74;0gHz6*n1neO
zY7&oD;|=@WmaIL$xSu%AiI5VtHuBl@TYT3O7Fqwb*5EAcoad@hT+Wb#;jr0|O6E6z
z{S8l3;``Ig_m2ac#_{s{T1@mf?jMhI&hX*=cU=GE&p76FqtbZF63|svhiYxnXxGGR
zvxbDEv7kXKOQSSqo*6F}px_%BiYP^rmRd2#LM;-2QR`<?R(pXUL)@x<|F~D3PN>d<
z5&rA*wO@d`A~19agiugMlGp8~#hkOG{GBict+yB>Z7-!3l#vuZMJnMcM({aN{NtFq
z@<fq;j^CqU4xQ(2x8uv7f6WvYN#Vwb^#<KIs@8-MSf&}@3F@fKiQ~h)^p$GQWxNPh
zX}ckZLWzmC+YqOPhvShj&1j|AY&xuSq!0w%rC_|AX*OH5wzPeZwvw!BQj5x!>^;s}
zjt@^vOQi7*tt;o#g(gI%;ly?;iPY2Sh;Ld_On_#$ZMix0jN{BO!P8}ADTSZj-jb)0
zpMUw1VG-rB@s9l_GEEcc9j~vp%*#U8cNmjt`<B{Uw5pt@g_ts}?zunCEFshPhJ9~o
zjOKA#Sen4HB1^{cL|zvBdBG~niA;m#y1SzB8`9GUj4>>6;jnGEyWOn;l)wA+M_rYT
z{yo8bj*+^aWJ1aqqgisndnXXwnAYiDygn%vh>7aOBz^zkNd`7+K2<*DalxQyotLVP
zMNYm7MTnX5aHiXI^sQxc*y0*57>TN=N|CdyH0EI;g~W?Z%THb(*!PZMocTDOnBxTF
z%nN6EdYlMj#5J1EYj)ORqSVafC~2VHYv?`SkC`uj_66IqlG864jz?-$xc)#LKJfnA
zU-5tXPyS23|9IhF{L6pCfBYZ)`&h(;diwa5GM~Bq`~N<d@88kQ3124EoN#mDn}>nR
z<G^*hp{h#AzvocN=~KKNwN^Uk`KMpK;ER2SQZSxn&Evesc}cXEmD%LHC<xj*ly7iR
z9;!7XJ~XR&9H!N4AgQ+{O8?~RjVLtFJ%V8gtlzD*-w#4sS8rMn4QW*r!-~K7u`j>b
zk>kh`dP0=;jBdBZTS16qRcWlFvyQ&;EVW{t#WkMykN3E2X^f;aZQs$h2g2n-OabS6
z_M4XZ@eySut(`&uq!!9H9WP$KA|XlIu5D>u;^y<${M8qKpVMg|z5gAFNC+d-c&5C6
z%kX&5`SG6Xm#_HdZ~vOh{X48t+`fFp;d)O6o*s|9{oPw09?x8s!s{<v?q0p%-4Blp
zmp}*$F-EE~Jf1F?w!?Jd`B{n#QQAvqXGN2%ru7}Ug7cjOfU0=Hs;UC7u7S4iSe6-$
zG@EEG={jc{)Vd0lm7lC%oo^{=p_E81c4dBLa$004C{b8iPOdq%CPQ;btWQGmrL7*i
zm?CBctk#@i#hw_GC^n&WhgQhXtCfPrn+Bs7;xaH_2E3ELN8>!F%US%nb!43Xup8dr
z^{TaEtrOL(WI{*`mov3wy1vEhicV2fAf+kt3zIG<0m#y!oR$RiI#p?9aMpow7$>R2
zTBLVV_-aOg?azGAD$Nfl@kPmkx}d6f@M0)brzBNppwShp#0RI0guR3WxUM4CpOaH!
z8K*VeqSmCSd{0Y~gP>Ak*f-DX#hh1<nS=-$r!aBAsls;C^24R@*H04<vtTz~U*FL0
z518E*A5SwW7q(qPV>LAg_S*(xW@kD&)nS$5n}-SS8_@8J_b2Xp$KQW_;NQ7>K{*`I
zD1Pzl-}39XANkkA-|#>A>}%T2Q<M^Ta+Z6f1~~{Q&2F=yv63iVf-pOuId*FGJ?129
zS#316k;AY>6s}xk=q^>Es`@n8)zxSE+^PCwUnNfDFjKuj8UN|ev`Ved%|y(ZVVszz
z32P;k7*k-p49w$9P@2|@ifF8cs>oHU6m=0hQ_jNXRJDHcW<A>gtd%jKt{9lUGyIc(
z^cPISk+=q-%CgY=hDWXFu68^<-mlN3LJEN)7F6SDs^(XJ`zx$>Y_}U?Te-h~q-~`F
zEG&U#o@l+NZ96u7&t)2*R*q+}sNCG$;k#ZAJYH00qqLAVyk}z!!}%=m;=ZTU%s32$
zd109*HhssY-|+44zQ<TA&q1R}>CFCW%YM^{=Q?NhS3Ao2EWAQ0-KL}2NVoC4EFAWG
z#%Yqqp_JDFQpqYlM=H|2+$+W9GNQGTa90Wp<AszQ-Zki2Xslt=I8sQArw2|C@99%w
zYhI(b4a7)&8tIg0nip!RTpe~OCEL->_P+z9Nvkc+I<%38|2)oA6lQNR8&8-6DJHD5
z#1NTJaw-%hKxFc`1nJL<iSv@M-ce!}aN-!G3vewEj!91bZuRt(Tu4hKXs}Dfw5_DU
zYGs}S^YMZznYb*N7<uz@&re?Ns1=@07jmAtvXz(n9<L4m>J<2<RE$zW8DAenxm0@Z
zaJwzruEA)<U;N}R*uMCZcDeAs`VR^(UcO@Y<v*cX@xG*K<mohHROQW^8|LG^+;Fs}
zCSjXxOu<$dE+dPxD-g8saGCk-+eZ%Dp4K~ZE`*rY@^v{O$CTw^USYTGXc|L^vhtGR
zV2R^XWw_GDP?RC-Dvb;=vCM&-3r*WoPz>Wlm?rVUBri%p$?Tm3SH_rFmVhZ0?+ssH
z_x$F5<T1+0KjpHjH<}m<%B&uXQVHe9uV26a{yV%=+}_->-|tCb!EZL?B%U%yWf=#K
z<Ar{^#~6#>v|tQ3FYZ`U!YR#Vju_i9n;!q_1vO=^J4amxx@Jq62b}ZFA>kU&yv)3@
zmd*Ye+wTCy<^DZOeBw$4{L?!QRpUN?fi4;M;RgsKx^Z-kVQC!Qw&S1uvwy_+_=aEo
z`qvz8Zusiwe~xLloG%k^-oE3TZ@yzHil6_*8*UCg^Lgg+!xJBVcuOu|S=J(JgEtCq
zEu~cY%?8$vPo*@z-{GC6rifMYnLJnZS#2dfJq&1Tq*WqER4x(<N*Qg{%ICA|seBPU
zi`Ez`jj<)?wK_^XE#6tF@~g7$4Jp&Mzh4iiH8@pkUDZITeJiYqSLXzji>NClRqDKI
z6t&>21}wD-cWhY(+NMP}4cHCiGJwyV9!@xCn8F_h=KQ-~>eFUmo<?F2u9;C4ttymE
zj0=hc%YyD4O}~R_pu~tS3#!-^=d!NBl~C%Hv8d<1TO|kLk}^fh%1n}p`V;bm=l`e~
zucI71u(Fo5ijo85NR4Nzo1xiCYPHCLP(43<R&0%T4r8s9r?37z5%lUh(7|&rsDP2>
zKBB3_A~>XRDLljlx4WU7HNSoL$V0GTG;QDW=EXH%e)WQPqvhc=qm`y}8cX3PcY9uS
z230z082R?Yng9Czz~gygE`{yQE8;Zr%_Z`6^;BiJe({o8;b+~B|L1rA3FnmfKl~s6
zFZ|<w{Fi+3)34+ZP>a-p<$|h}sujfwgn615hVdemSYG^*N*SDXf+fjG2DsHbN0pjG
zsv@OAN&+pmRtsaXBL2v=73Igiau}#-Cl!Wkcu-VaSy`tPg;j@iuMg8KpyVV<>+@!Z
zDv4(2J&o4PQ4ZsEMVXW$b$(h27AlqkNchQUQDs4(@vmN8qgV*jiI@t$?}_t_N{N@R
zUy)Vi>HSBDfm$l#FrccTwO7O?V!dbIZ|J(7VVub|p}v1hW5v_H-EGOaFijImYw_?d
z!sNSq{f6E36{hj{)DS}@rN|Nm#S|li9BKQWrfDc8Gh8mz3f;<!YrBRR1J*hY*IUl#
zGkx3O4BTC9@YZ0B;SvIeeZ#hRGF6o2l2`1G<#;--Jk*{T6H|&P16$M6Z9AIQGtH6V
zawe>`dX0CalJVZH_pOBx3+G{G3c?5Uz9HwrzxwSj_}O3nd#G>jVdHUES1iL!qb=Sy
z%*)I&&+PlXq_6+LYwC|1>a2L(N|n=JU7~nMtig8{?+tayEaO620~5*!c+Yx=Dq_OS
zF+rA7a9WaZ?ox&<09&sW!|Kbjt;aWBs-99NObcOI7{`Tu$lP7+DJ1edp<|#mu(6ek
zH;hYWNd-q{OXRlqygY2UI&2B4@XKF60UBowo353#iBT&fuh6ag^RDgr;<KNlnggm-
z{`{*u`p^Fgu0A79Cu**I{TKg`|IPpUz;yn|%bjNb@~8OS9W^a%KKpaR`(M)7!t2*}
z{HyQZJwNie+xNUU%v`3#<8h#GB<PW{m@Bnbl)OT0i&WWAWF?}s7}kx^cqbBiUTbIO
zoCsyY<&5tXP3L}J%VgwXu0o)BUKM3oy>{B*jV7c_Tzw1P8Jt!edW%sNYhc%VHeIvc
zcnVg7^Oj6u`4jl*0B#OD^0JUaqH8-E>x8&%_DrWI;y5x6M>dBYrQJZ|F{xu2CR*DH
z!!Je}BFeU8?Fh{Q+eun8#EHiDSmWgjyWL`%mfi)jJJN18H2WKtn7Q>G@w7wN0GD@+
z$0u69S(^~fxPC*nnuO;1#SMzYZriXiSNQhtv48O;s@>6DzodI<*}V9Io1gv#?&_A-
zbf_>8Hs5o*Yk0HCU>odi&(pB*!`mOQQPKGZV;!5e$5^u_xeDI2kb+nn;<BnkG91Jh
zIUkQ$>liK<IgpJ5y>CSwUsan@A;mxn6S`(nm32i;V!p=MRRNZVXkL|gl+-W9Tqx6s
zs#V<irF_E6gp#qU%8+2qx&kVorO(m%MwFxVnx_^@)>s4MlCci<hdocH%w<Zv{_+Jb
z_JsM&ZhI?hq^Ca(r*1hvQI?UW=_UPXq!~lP8#5_KbX+iI9iZckX%9GSs3l@`Mj>l%
zt&}9NtCXfc18t=jx1cK$L@=WI)G7x|ZJt|56nV9T86{pWQ^n&d_l1Bd0fl%Si?USo
zy0*=b3MH*+Vk7IMrfpEAN|o7abkFO+u9RR;L~#`FpLJF`kde>1gh2Od&xePefBkUJ
zH<O|mM+J7<4gcV?D|Y*qZ|*O$#R-vG9n?&3pldCc)5LE-ocYgw@dL-x%$I-lPx<`j
z4S)OB|B9<OUvYf@9sl*)2fqI4$kp8)D9xM04gbzR{42iy>-XGUT{8|B!ZhI;ORkbu
zwQDM~0&@y1p^{U@_)b#1Mi3aPLf`i6c6-7FimK}&N};sETZea!F)pjhUFH8DEhQJR
zmMA{eVg8s`okz+%P>m<+OhwQ*DT}&Rv-rT5B`{74m-EQ`A3kz?I+NmzRghHyWGIhG
zf=apM-+KXb5&Og{5Y%UUP?c1)SS-Yb(X<Vl?Vj)7zT@R#$7bK~?%j7Z+M%r=dwko_
zv<=0#Eb~Cf5pP>m$uz1`(nOvO$N9qJd1SZmXw7Si)2Irk)0wc$++1DDZ>SZ{d#-P;
zIow`Jm@6b)(@<-|fDj<&g;ERq>np4_%$I?25(RU&-;tJuuuK@E2`SNZ9aG4h#~EtD
zI!UC)7zKf24c=Lrrr~rM*QP9~5^FYF_J=+1-+v(I%6_}Sl0NmXPSdpFV=TMAA?D1q
z1Qf#OG)4jwF+`?0(t5|!>C6x_F@v!UfBpS8{Ez;ff6mL@72SSA{rg|iJRUi|JE4u>
zRID`&mkXPX`-2Y=f7DRttd`R%t8A;~G?T}LxmJAFV2nbgh>8i%3h#_n%6Y4x0*DE6
zK^ZZ3I;F{Es#f&YVl3pjpjXPar4owtcPbP$ib^y+T{xCX<DgX;=V9wKDuZ&GVa{B}
z#8q#&ZX2$5Et|HbF%4mvS%P?f8fT^VbA>3ET2VP+tHxHvoBb`D-5m)_Fdnmijk^7s
zR28+#>P}h97hk+)dOu49e*1##Hqczt+<5xK9d3BbpI>bl?jISSo={5hXK(Jf+4gt=
zdI>v6%Zbf8S#wHQLQYmj8>wU~8J!CXK5>bp$wPr=O|@kRQ9ABIV7!b{6kiG{7F@1^
zM@flGj4Y{8jmBz+%0ig$o3$i7Ct@zN-m}s21yM@;C)Vq=gxIo##KviUez)Pj{N*2W
zb^ZUo8kOj7udv+?*KSyr1?>#kIV_r|(;2h9rhWZ0nr(;oEz3ADk3UdIXzS3X!NttS
zZ+?w!x6mK4-3DU~hy4bMqZ)@pfVqBs$PFK7KEA)_i^D7I^#O|}7e$$mq~*-!aKkvv
zjMKv5_A^e$C%X8aW`B)QJtSG<)u!iW{~B}sIZ3sYW=By5XRo-q%3?n#kY~-_e#Ve8
z*RL(_K73?+I&*h(&7t{%Fim34uN9vxRXK<=&RP^1MIy$9oC+mO#CaswN?3=xGzLPR
z*<EjCt?D{Lm}$B#A%o5rq(*5W)gT_VDPc4$Ib)lau5Ag^h*Fu7BAb58FpdId4x%Uy
z<AqJzLQW{9u})Tv&dX}acRi&VR4kw(##@H-iKgv7B@C%@&uAM(PBRpXvs?DJd(1o&
zFHhXRe^1qlrBq(Ld`b7G2_q7wfwT<N80q#mIHxeCpg|5(Mat26-ae#-aC%3n2BRyw
zt}PnYg958yN$b6@f>ja*QMIDImx?jxey`H4>uP6L+eC?|CCYnJ3$bMC+9X1mvAPIe
z#djz#u<E)Fk~xYg9;Hx5PWskZinHkY-1uQ-l{b$Vt!aBF%22Iq%~X+9bSmhS=$x$e
zpQeTTB{N1bfNNv8?LGTW6yulz<1%5jWe$;v#N$abk0ZZ%Kk#qfJ#kKs>zkJT{ogPy
zmDfN2bGC;&-lf2IPw)5_zx*9tw`G5Qg>stz;lKCa=VAHJxnu|466Qd+X(?@st(C@m
z8t+#BpZMQw)2vBgVK@?J3Siov&9x@_mM~vHW$L<8^{%q-8+qO+MMcS4+ga)5u50bq
zAFQ4H|9_QbCXQ#oqS``L=J`y-664yrJ)xDsTF0`4HQB1^J4aA!)uB-<usW|@?6Qh3
zS)j_s%4dy{!454zWl0K0VvrCIjpJ9p{vG{KZ@HW&QhuP{@96rTWe%heIc)Z5^e0uh
ztSX^5iX2A{haFla=9mfdz+zz-M&@~98fUf}$D3C#7>0p)nc41lT)%qBX1AlR)`7+w
zY7SgZXTms9QUo;Hs~xv@cf=4-6^wo0?$s*}+YQHudyEzJ**A{UyL$<fImhvI;%8rf
z&M=&%Uej0(+lF}@IG#t8_D~}?hb^@f?w_7$+m^4s`jWTbeUGNF-F2jtI9&#mwkQmT
zo#W;00p}f$$1@-A2ebljET&|%F}OzIb){bQygv?{r$|W)Z{L2y|NdY8U-@7DPygq%
z$0NJ=nz>U<DsVm@*>^o%)3IqS&S?4E{(#pXHPm%p3VK756-n<bQ;EXtS;7jmG&D|;
z5mnDNmRP~Ar<zg2P@t?yFk7YDRO*E^SDmHz5><CpT#n?q5JOu108;y+bfwHQPa)tN
z!{>*FtKP6N!iXD@hf&G}l*gIEwrzzWl``XL;NiRoN<fY{=w+7LCKO&7Hp;MVI$mGh
zt?Q9YH6E_MV#$W_>4c3JY!TzPQ;O}IpHoan!jN;K?XSdh)?T5@fN>2!`PonS=HD!Y
z5ZQDcZe{D$T&QIQz3Sg%57w<PT8W=c8I4MEdJ+79CMH?CQtQeQTFWJKAx#U*Y2<jB
zL}5XOzbO@*vt(t6HLaKuR>@vY>!c(=-cndH?6xg!6=i3b7Obw)DxeD)S@T?V4S)6d
zf&bkfb8-E*z5Jfto1X|d+-P!|QO2<M8{UQsLHBGsPY46);~Aw4ZQHTiZ~6G)9VRT4
z7`Plyl#~eJ0;S?~rbNY}g`}@tL)$it<B^3z=N-GY!QWI$8tGy~bM*@MWk)!DB;0?;
z*1=)7A(V=3divHArvY4p@|J8ANgIsaU^g!*-7U&^l6B(Ku?m$Ij80fx$X*jmL;vQd
zFr9eS>`22SyGB+?v+fu#1BQf26VwE`GS4%r1opjyG;_S1h{HhNG-%`LT+4ZkeEtvr
zlBV0SaV>bmbbjRH_umpiB&SN6CsK)o%bA)Z^D>hcDV8rlx7`4RI1LyD-WZ1V$m7!k
zo85+KoaovXYYoAS<Pg{&_N1n)EhdGSB5h3gwig~7u<=`Bj?}4;YnFxyg@l7*q?Aa=
z5#<_kIJ47`b>YQpc|SE3ef$<bzyGmU8vxtQM%p*L!~3<rc>Mw@4Hkp7&AR%Q&;W6k
zOoUMs(+VnMO`J;A*2)PiNdWC@@BsMaqkC5JK7}J{6;)hol2R<ONbp2!6q%Z%sB}4i
z%^2HKji^+zUYFEvSpldym$hL6l#&DDvoT(p6UyopTem9KkSdrWAm844s3&$tHY|PD
zuxp(RkGatKhR#7unfv33A3i>zaX6zmp9hSGU&V!QKRoeqNI2*C$=wy(n;WrxJf4Zu
z0)3#CMAI~!=Y@x1;US&5t}S1Gb<6+aKl~5*zy0-Z7|xGU!KTF1RPyt^aP`0G-%t6^
z%4e-r`ihl-rloN$8|$RsGX)?6lG483JD?I`q9FG;YZR42EJ-jue}bVd&I@I_pp^xs
z$)=$q#0HEL6;5lkwSwDnhTZjn&y2^oGyNs;-ErY@nozmW^&3zYt2Np<sx=g&NFkA8
zMk!IyRhAYR34jD3p_E~mMh<<4cb@zEd+t=>>h6Xx3{rf*-?2>Mg)a&eqEM$=dH-<F
z?XJNl!*+MYJWqIQF}iX-pSl0|MBjG!7gwC71#1*Lzv1@fEw*jI8e$R4OpSpbzWojJ
zIAM*$ILGa)7i<n!<S;=gT*isr?trt3m;&3Y12HT-Jw5W_<B?$wG@F)FNZjpr>^jfe
zr-)Ss>;6Bq{nw9WS(fIB{kA;n3RA8^WJILStZFT)+2n8l64V4A2=H&@%YZKg2|kj<
z%rsf->aOBsb#_%oW@Lu&aCehe9kmr7)-j75puo=P$T~m(4hwg)YxnN6&)RE!-|urw
zF>$jyu<K`<*3eo3P+Ko(reQd8u{~$IUNQDP`~8u&ZCP*D6bjnFFis>UY^<=A2eqRg
z22?53>lI@^&@?rd7Y%Kt**2cHyOEpyz%WMs-CzC{4#WTWU;S&2pZ=0JH-E$R?hQsZ
zL7LjKz1YxpHB<U~$bN0Cp|V0u_9b!Ict$6uo3?Q@jiad~9%-5rl{c)`4bEF?=Rg(q
z$ANd(2l{>%VW(1}tILIw6PN3Xt}@*87G<F8DpFDGvy3Y$t8qq?m12srQ8g$wR`aa(
zY@H^%#Bh`ml2TcKp@t{HBn6I#fn7gS(6nCDYE7Fp4hcebO4AsFSC;AKno_Mvra@O5
z_Os@6`vuc9^4_Z}{2XBzF-=9%H5p9^GqtPHB}vh%LE8#Sr0p8ef+WZ>kY@o5iquA9
zKpv(s{hm!fFBzH0seRF`X{!YVk)^H9;*v93i%V$|onuIeagIxzk>LK0N#fS#7>Px3
zi~_n%rJ&KYMiE1%S{`a4!DiL6UAL@Ef+C~Wh`heq+GSFwsc_$BY5kp7_wxJrtB+Bq
zcf9@dujqF-)OE}F>W9=<FL<`ua{tw@nQpJC9|ibvrdhApUYv9N?v^kH+Vw_)%B?|f
z*HmpK`JN=<LY0i3C+s|;av;xwu2@~2q0&gW{gTEjNlPg`-YayD=&GVSe}Pht%HMGx
z10jrrhl%W3-1d^7D(2!)Sy93XXC2Oa`FxE*s|uTXY!S_5R2JnbR_hJE(Zumc)Ec8A
zoyu4xt$nN_HXD3wiBr#UJQC&+I|a6zGdAbvOv*B0SU+D;*@pFMP05*+wY1)G$ceXi
zJ5seG(J=HQUwrx%c^nBT;GAJqdAzT<zJG%v(=8(mh1^$^$Z+bJW1^@)Yb?S<hkl-M
z3Yw-RgupnC#8_yGl2LIf;w(#<k`w(QED&FV%7xQ7pmo72g~|h`Q_tc0nrhS0u2xK^
zff52Ao~<|>kKcK{0kmBUO7eA`F@jey2B+opt&%|{(FlemhN~!z)|tn+pdu=JvMAAU
z7K*Vl?o=|Wj!_CJo9)wBC8x9m%O^2mq?vg<5vCqp64uF~G%tKQo#SF+5z#g&l9)+q
z!dA_~E0dxwgh-qQ2oofca)d0PV@hGwbyUvdE9nqwD^Pf?Bw~t?sCB`R+3g4BIr4H{
z^VPOy7(_34*;tOl$nh}q{&~mi-HgsN&o>U{iTk@fr%`mo@2_kA%a3Y&mHEv><Y)`Y
zY0BY-wKH67FZtO|zR&jRic=`OzCUm(Z}{Nyj6eJN_qe`a^X2s&d6@=jWY1HNoulP?
z+&>?U4Y7RVFfM>G9_?Bw^pX?{+Q>wtt{pK&>12f<#rV>3MXBY%``s>;1hSpRBWR21
zoQ#8G5D2x^7^~6F;i{U=`33bfQPhrH3Rz9;XT>fA<`7ZCfb%uZRg2M1q6Z(NJs<H{
zqFvQ2?x!a%A6gMYVt43ywrv@wc`-#5jB_~WIUP?j@u_RZegI?H4<n~xz~+kYU9GS=
z;*H{RRZ({hhvP`?D(QC3iQzcVb}i3etfey<0*C&D)(}hKaDPwV_tZ^8UA3HDUa+}5
zBd5&JNBZLtYd!7u97E!;yQ4eba&vvh;S_j#v*+^i9Az9~ntA!j`{Xo<Bq9r{ESH2@
z@~Nje@cjH7tu;;467!Lp-GLay)!)w|A(_IF)w-3KGOg+RfzvoL-5=1#VNAtlCFvFY
z(BrM-$e+ec+f>xvaK34|+a39<-@U~EzxkWLWW4*5_g*|hRfUpgERw_Ryy5)vT$)YO
z-$SB=)dIQnmEp%PH;@%$h<N9?JZoudDOz&L#86N<VZFn78409P2u;OLFE*5ro?KX!
zlg!M7BvEJ$pS(IFps-HH74tA*Q<1=56j_M|OgqcE_N*#L>m0Q+48z1U-!tDHpad#!
zsI101*{H{1=H?U`bD^s&7j4C=s#v)OZw!heW5q>?P|%|P(T?lu2mX)$r~iuC7@FGi
zU;fK~!jE6AIKFwqi;sSQHytS_TvMaTlrWKu!Nmh*+)=b9PZMzp<S3Crk2d_gY_M4s
z88a`;s+<L$s28nYS^iT>%UGz$$^LQUdOYc;`I{`mToQAR3{zknC%SdVA;aglM;WDA
zEl^eGIm`*A64q2;mVz<EFhw$&t~U6b@JeL5c^OkoOQ^P25)A!auTlc#_O~2&*TRNc
zU2%$;6i2*on8OH&YxPhvUAw}j!2QiND8+iS<)AD@8`kH~$T?D<Ur{;B;qDzCO^&lT
zd>5!}3M0pHqF#46t!bN<yW>0F{`MDC+bg=&hPi4v3^UH1qs$tn1ATv@7)R5#FwX=;
zvoVb$vA@H3M^c`797We-oaBTKCrBeDj-+wMw=0s7crzU)$);@_E3F8}11e67Gg8x9
z$ArSSE2g1ivK_gaxxRTzeO00BhE?5>=0Lc;;TQ(os%DxTWt=H7^6c3K=Il9DcfoYr
zv8vbX?(doN%=4=Ye(@Loj??a*ah}+89kiaDCf4hgF(z^XoJBbY%YTuKts0bdgaoM{
zMJAv<<1k4sblU(3+CYo610@jiIVS-rlO_I$f(n_dXYb+GFBpdtV+?%p^_O^Oxfy0+
z`R<1ZK;C!j&|36#ISFZBOUyAELrU`^u7Bh?Ed_;^v1LgC<^io2N29hVQ>0ih`pE<(
zCX)Enw#zXQgNWlrG%G>#C4ZB@>U^~To&#QsX`;4|-Ma})hjD(%Augz*NJWa~I8Wlz
z&IRi{P@ZC?s!INiwFYfOzo(RFu2M?45@-yxji;Xn_NR%nuHuu6me)JOlrr-)@$it*
zreZz~TsDTEKJWP9GtbRk4Nc9z`P;X=dFa{L!p;32WqQVFxoT^kZ4H&y^i5{nHlze^
zuJ6dkvCEd*>7Kj$1Do~@AH2Naryst@{W$UNen-mROg0`FjWQ))I(Iog70(4>)}a)}
zRrsoqlysaZk(9E$cS=k?(;TS1rM6Do+e+TkKgN~ji8vYJGz*5yTC%QDM%?W=Ckm1T
zqN|#f5u*!ArZHI^xt}d>PCMo(VB~LH)CkaciN;DXJQa31WsG6bc#2U`Yhe*;zvgB)
zK>CtDyZlp5hXdZXT)lY4yLZ=o{pD8}?`2H^870ME5yMpFJ#*hvY0G8jFy3>xJ>p#@
z_fAk$z9NLg&Gikn*U(jz6j3>o6ojFtU3E0wnv07|s-|I@#N9fdPNbN*eDMmb;dt|o
zq9CNm7!$X<neO}<KlsTHc>U#9oS&c3whg%-IKR5&@Ng86yY)0}$0-b?lxe#TZ;I^M
zQ>1M=_J^Kn4vfQ$QH8VZ3eX%L4y@Ob;9<4Jw-wj7JDh{Iaa2xA<G@&&rlyp_asMEB
z(nTg-pS=G}7>r%b#mb=bku(<^5P`a?xVSpQ)gJFWyVLjwt*LKS)V1EyIz#O(&(`l@
zv?0!cuBqtSW&sQuk`*M0Q-zWPVHn6M(Dx(#IMdW#Ebfb=%4jf10b{HYRLX*DP+CFB
z=*OEBh*3@-)@ngmm}LwQ<>WcdnWmNkPf@5UyNpB}SsQLo2flbXVvYPhl`#mLQ3D38
zq}{F;BZ57Dg<d_wD9`==4S)In{VM{B))-#+%!lXyl2a@k#z0$1C)27-(bD2pb5GoV
zNjTl&x=U`}z7@?%-C#oiMk*!g5v3GVNlzJdswnAxJ#Ok&iEa*YXlZRkyjw)Ps}u-h
zS7LyS5r<?fnZ{}?g)mQevw{q79!{irW~CKZt>xm(v$7hm3XK+lwzrCz0xIKE!VD+0
zm6wnNQYI*qu#nc3<eGlh>-NiEGp=fyrlHzwQS(F|M#|wtiZJZ&FuKs5UkF+vCQ>d8
z$0O4`arNRQo33G+lfZ9%MO)VpM)Y*zw7aHTZE@WiW1SQWv&abM@x;4dAE~-Esj>{G
zksPm?b~l81kfe(8n$6V<G&5x$vFm53x&_};Vq{Jec|4$FCJj4WyJnh3im&AR#=!XS
zP7(y_4((fpc@Xpkg189dNV16;i!FxQHN-GeR~|!P%7KEVS#Qy{rk^KXo^4sRHE|Ba
z(}~t*LL*p$Vk*M3Eq+t6Es?G=6lFO)+^}+vpZ)BI__o0}j@0)&+~4t=-+jqESnm3f
zDP%gY$-5q13SCpP-%ZrDj1JpoIf+jr);p@U6+J{LbZtwVd&%Amf-A~PAa+$rhGOL$
zVeADf(^SH@at(nKgloLjtPqXZ#db~B8jE2+_J4#G2mR@Q^$isY7bfY3q$uvgELvM)
zAg2kdkcUtzVz6@R3G1SBTCg&aq>Si)5FL|qb}=X{sEL%9Vga0!h_`7RQ95G{R8=LW
zm<2_lwL)o&DOOyZuHc*@8HIK=zFCRrI|ngdlq3f>wPaE%={^>#B@AB49`mT(Eja)M
zRWzjtJ8)|Y&sT=cx?$=M9Qv8nX2Zn>-koj<X=0ocnVQNf{^VuHwsU;%`3gto`I%?N
z^Yz^huQH9(40B)%17mSCE6s=JTT)Rxl$KHy4~G*?(=q1AsTj1cxx2kb<;WLbea*V5
zSf8y}Z_hZ*<tgnUmnD;TdB4hH94QNC##o0^g@l2aNq9mLpnXo0XeX6KYsC~9=Risu
zY;CEWV;bkD4&ZmaVvO>9RI)ZEWuaPwtAt*Vq|ibs>b7GXhb28{1u@ap0dFQ$l8KN(
z&?JE%u;nF2jim@x%ln}LQf6KXJ<()Gv@w#zlmgGT7sM2qr-829F!VjY{KYTOC9&D8
zI8oSMob#|dV5dM*i7~<Ten6+d<>dvN$})zDN*6*YggLU_Z0X0Du4}~zuz!Fkl!<lQ
zuzR>iDR}w(Ilhv}yPN`X92gD<R4HtqKgX|D#A#&ra8K<WMOS?B`i6(o$Upm6f5!QX
zEnj@`6?N6HKb-jJ<qJa091jCd8B!@!b;EpfT%7xgaU3zmFpfQS)e%!>nr6DLVH^fx
zOr#8})mp&ERw6o8hIO~%G>o*S5?Ff{v!T(3##>hFE$?pcISwPf={T=iqB3-?$JCCg
zB-E^EtBS6+7^hgZ4Ru}7c5B|=-2Opp>f5<rz2bb+P*)bOCHu5wkI~=^Pa9u|naUaJ
zuA`JlnkRBntX3Uew_@r~G~WJRS1HTpqYU0VvDPo0sv^e?X^$^D3GWvMRQU#-00rwk
zAtXY9${8-!HBDVn)ipT;GMT2T*={=i@ef|`$%_fC6=$oKuC1Sdu96uFx@44c)aUP`
zyGvBcoL^n?@%zvCkAL$iv2FO|llSST8Q-1pm;e4h@Z%qUkF#|HXBTK~DdUcL_nLJ7
z6>Zy+uuM|{BU~`GKv#{Gn@zY>24f{q=#k{DjS*JVvJvD-;H`2TTXG0XK^K9NYb6EO
zBQ(}3Mdu7#r_s8w@`lUC(WyeQif!e1vGR0frj8kJG|Yi9DqLfzt>WX$m9TuAM(4sb
z3&`?Om#CM~#}p#@_cSg4xz}*o;Z)(}^H+>9W0b=tMLZoCP7l~pFxJvGE!v4mVj6l*
zV-&5$`Vw0r;q-=Z8nI2sKqgKDhtq(m&(ZBAAx_L;l#xQ$psh!x#D2HKOfxIzsa9tM
z3WuAwJiL8FRj-)ld)~eNlCz5o+SQ76H4yp}-YuLk4b^Imo(5Ezu@0Q6n1&<F@|k2?
z$Yd&0$f;*Ko~V4qwpKXb(a!<XZJ4a(wBMtgBkgw_`-yJ7!s&#~k(@2{_5$rZW#}1C
z2R7R+bDYR=!Yur_QX=#14aV1;*_L7GvFnD_)icWRmK=_h(?}^DW$gK<Kl?F1{OCiH
z(HxI^V&9{!B({{8+21{|dwAes7kIHf$98Le_t~#G4iimRaj{j*Qy_(zm(L5QljJdu
z<4iw{G;PcAaAdt|Fgek5p00N6@9x<?e~zMKY7JlCT+_4@XBSsEZ^_fZ`SzS@y+->6
zzBJ!?y@3=1A|;JX<A~N~;r@x1N{0y@C7+VtW0i%N$SAA<Us=h3T+k>nWnu^<C17Z4
z?c#)ebRiZN@DnKqYHD&8r?FN=uti1oJhh0W3re_zg{3Hyse&>Vtu$%j2r7{2$<&i#
z6bG#mcWx?a(fVb{O;$?;T+UM9D2s6hTVP!&UOd|{oO%w26QRf$D;D7->LRZ(mCV^%
zg2}teaX1W|uPQd1mVH0-@yiX?I6>ruNX!FenCOh6_J&~yG?ixz38yV{%G@0XlJU3%
zr{jr~BD;sgtC!CR>ByCnvE*Txm_vX?8z=>K`6lX1E387yd`e55vrH6})+pmZ8}#x_
zvc}L<mBc?~p$3RSOh4tFf7dI8OjH4-6)E>9-B9^86oGb&A;(cv!FkKWewLNuyg1;k
zCyW#GILPZviMq2?zNTo6(Q-Is(QT%XVCfLY5EgBimacRW;QTaAIAyWM@%2|<@zbCE
z0Bbd0eDOOdk(bXe>HD5_TQi?+DSSnY2^VL^dE#$Azvi+vyu7-gavpTx{H$X)D5j$M
z?DK1;5LvGVcK3>lmEqZTh0&VpX`raW)r)70{fRKoIPb|Zu)Dj#+J?)QFR+^x##g-l
z^{2$4r#*k4FJIsDx4-_1kH7x|KK{v1nED+UOH7%&hduxJPkzYpbmH>rifP$@a2n$t
zp5inqMOdC%p>@U^xH?}^Q0z&h6i6v@8VYq)VGUU833JA1(aj!@N33;Awmlq=J)^PW
zEN(ldd4@QmeZy*1(|M0A&@>JGkO=d{=FHJFo-|DuYl$h!6zC7QzEx58-qjhKRfqRR
zHiZx*4=_ezOqi-7&I8lw0h~b%6{cBHFcL+i6iwHnmk~tUv=kC)4x$s%in5F&1?REM
z{FfHGdKA2}Sp1hUR7?r2&BCn_F59#7PDJ2J(^ZaK$wq3e;pMXnE-%*dU|Nbsvskv}
zHWEz?T1jk@Yskt_$h51L|K0!Qzu<e%TAIqzy2P-%<FikHLqGKV$<Mz>dv=L+9vx4p
zX-A1CtaJG84EypYocT{U{q<ka-+nE*jEg8T=S+#>%(2cOr)8F0%>`9ZmXoLv0i9Cf
zsC(>|xdp=zLL}uP`Xp7TyvFDjZ7j30v_xLHME!DwQJGESSyf9WWReU=jO0~nt!bTN
zRm<mLv?hfpZa-^DMdDI)lJ>hs@y9SDuDT7I%V*5x8#mr|yP>RC@;ax9ew@Ho%wvyL
ziEIGqKo`H@`THN@w^ul8uqCoOzglt%-I9apsNETp-LQ8Xs&dbCyca}4gc4`GbzGfa
zNlt0b%+pBSHgsnf^iz;r#O)SmHBKwMcBmW~r=F5MQ%F=KOjDDsCdEwk2(_IDOo_}P
z;g&ck=Op(vdIPZ+O=jhJaj`+Uiv6Jnio`j!HO{WcAu&z|e96?VLgg9NPc)T9xt5%w
zSnHb&MHzI+WM$FDv+wWmD5_?KEsE0biAzCVVlNMX>BRY}rgM&An0WE*CC*u9CF^l_
zcaM7S17R^u0l%&Zs_@~DYr-VrZGG{KZZV;ZwSjI;qdlMh;&1uvx3~Ch;_m*2_s>??
zlIS|ehwr^)P8HW*-tnW=du;5MXCJ)a`Kwp>s-|ee{{Dgcn>U!e=i+&dU#<TLD-P<q
zLRS?9gVu_=uJE;JAXAuxRW@r#Gmx=PG{mLINKYG&wO*WtTBDSuDvG&iIV>!P9hv^A
zC)tD)b-<F=p|!BIN|8|ZwCHOEb+U9^lmaDV>kt#MBx#DKOv({*l<$!vVHy~Rfs_Ox
zld^z!MZYK9HDhECAur%+3b|O$Ys<5Bg*ApS2u7l4O)Q|aqH>NkiDw(nx~(t<ah9VA
z!-=k`BxVi8s&;5IF-#JprnTn9<r!zI8eJe}#eSHXLcussP><QQmT5S0e}9Xyp3V6s
zmp}R$AAj!y&byjT+t5z~fBUz;6x;-o`C5wXnT4fOS~8`G!J=e2NUH@*jCEw=DLJCi
zG_InoHFJV_7THS95>-@y{72B#jpHc#W2J~=#0E)?aT|ltj-;$~c8f+8gHx7VW{S=t
zXG)Qn0^S<Nm??808JPwcS0j2kNKr<{##Y#r$U)rCj}4I&g_W3MqG~kj?FFBG{s#Zy
zIVlKMWjGvoc5#k1idHpj)-8vJ9Zg-)R5ex$Ec;;?d46_*cLjNj8P=8I{@C;OVbR}e
z%fq3^+Xkg74u>NK(XHCrLovMj@-@D0NU1Q-6DlVzU$)q)qODu5-@M`TU;l<zuU_zQ
z==tok&w2jr1wZ@KpL2e;=Ie)RR!zg+>}Y)>Io3wH<i4&kt|A==8t+LVayp&Z)aHrv
zsVQ`A&BOjA<cGTAc-V0|4RlS%M8eiq&}+uh4=2Wy$y1<pa_I53Cog*T!_ZTC$EvCX
zV`K~lweV^o<r%caS&KJ?i?bC~?eJEkoDn0&A8>uEqVD<C1zp!H+7c+fVi*R}qZL0#
zS|{Q@MMi}gWoN3&k|#qPM`B+vi(_h>=BGFt_1G#dERI#_WH=|`)Q@6UZS~V=R3dBy
zi(<6K6rrxS-l7{XZZeeU;f$8kVaXWn@YM>}{9d?rUP91GSo8Q=3b|+qgE*~#s%iP~
zlOM2NuZX7uRt1Lr*K}*o&;G#=sn!>0T@l9~HQvDS9qF{gxB}fXbiL-ipMJ>M$3NpY
z|MmZk)0e*_hC)n$77d{`hGK->_gL_$QiNTU3k8Go3)_lC_?k0T8N7cKUMsvR0?kAU
zLn#6cQ!1#~b5_CTqLuNNGc?{UGylSz3(8oE0;e^#TE-1ZVXYyKGtMgE#+Aa!X+jcw
zgRzEr{_e3&tL-_>=6vaBNH;?((Qr&*MgiTrqg}0;+7<D5N2d$bdP8Z&VQDeMsujU{
zRZ~)8JRX^*!ZFUwx9_NpjJZ~>qFZ;AFf%WS3(9CxoKYI4IZ|(5^6|w7lsM3?I#%av
zNP*M4JNEYvSnqMZBBv3TA|?h*)ev;yu)CpcS_ntRA>itjL`<c}pl~^?81K<pbGtt<
zghabq(I57lhKb~AR_7Nu?@7~)QyQZ(#nz;vF#Ulz70?-DJ&kWE3Wj;2C`Z-QC|gtF
z5vx7LSdND~rrWP+pS{oKYC|cBVVuyeVjxm#&1!XqHjbO!Juxcu0&J}-Pq$g)e1&ft
zg4R?&YMGBahyi6QzIgqXs}DcHWzFvPHKu`QFW%#aH}@DS-d(=~l{h`zadEMxsXd!>
zL3MOZ$JC$r)n|XrPs5*Z_3C{})zX}uqmAOQzv1r9Tj}Kg5iaKq)_aPw7%PWBv*0kC
z(G)Eci4qm1M69bp$uTD7LeU1LJla)YMWbMslC2<ekVnLZ`F%`55gn~n71m0kgwjxo
zBbR{`)00+KICu&zpdB$y6-CHNM#DK#mINA>@molld6sNZr?qs|f{ebkwcqG@bB2Yd
zmlC8okZocfBmE(gW-$yvVj4$`a-7$WuE{(<tGL*%xH@0DF8hI+OpFt8E*xXwi<^Ph
zH%HmSNnD+;d49RYD#PoCk>6aO*pHFMIySmsYl}w_i-3QJenh*%^Y35LZO_R{bBu7@
z&(v+p_kQqme*HIpO`1m}a7zmA$0<U{Gln3C-J+MY4$+e-qbbIcq6QLH8&<VtT4knj
z6g{lg7>Ntvk0Glw$`GR<Qk3xw$0I}fk}l4aswMjs#x)d{xH=NP_9!Hqb`FwjTGtg3
zBPIZ&ktW|_z*%&Y&N;#)#0Tr0bV_4fI-6>F21^F)+qdu7RvG7Oyz^YY{TfSQ)izk8
zX`6=Ocw{`CFviffEvKWH_mncEqUguKW|gR{XEz_QB~e>NU3VP%nQ=<IyE_r4fo<!l
zOySwtnlO#L{^j5B^7$+DF>vUmnE%tC|BUmiE3CDg?(X>XuYXC?v?LV2`^|3|r-_d~
z{sFH(c%PI4VH!9;+i-h*L(Yj|n29M-w=Jt~BZYdr#ux`#rha7U(1w&S*6`xhdxWpv
z^2L|0Icpov&dxaPMM4sDq*P8~GL>f4HtdFh%2!KWBVE77JBCsA`SEzdDNWs0)K;?T
zr*URFNJL*!nip4RLJlzs=QPvMOB%)c4?DPi%j{R7ke8)vqLiZYo&YEnC?yDDz^Wxc
zJ}`wK^ZF1-K{gw0tmsLG5vP>^je!_5S|L!x-#Y~TKDbl(I0j=Zr6|%;JfxI}3sg`q
zIx?-rx~#P!p-9G{mMUCj7PzRK$V*Y^BpZ!iax_tjlrnKXE#tJM2sD;%d(Pan%*T7I
zf)|$+-T6z5UvvM(r>NnE;mv1E54Yqn@S}h54@7{RMpT|yZO{18|L}j#m;Y}Vu79^|
zxa!H4pH(3)t#YFwry}HeSjNsv&SsW~rcyH2igBWpg01AwkM1|)48ACN?Pd9T%CaeF
zffsux?*|H|IZ)8lZ9`qxgwug}$=56eOrybAadHC2dFGT^D|vh!`tN><&5KB|<b=*b
zc8?*WwFuLvSt75-X`~pBbu}dgR2Xsdft&?2*;O^h37GYGIx+VrDy>+p)-obD65&+0
z^)gmdgc5K{cwlwgP`fq580a<^7*{dfeTC17#(SLW@Hny-XJU>CYifMu8Rk(iFYA{5
z?!Y{ssH`DtftUK`Oa${WE@QSr(iUwh>a&JYBKL<QAtyHHXC!5rPCaD`U=&TalJ`g%
zQYvUa6Q&c2h_j72G;^l*l^_ywB!^x`ThmNZK~fr48}Kbdia2Fa-jHTZ2|YO#bY0<G
z#q&$~oK?xhp~h&%oFOSCF;I1j?<}f0<5&Ow-_q}o%vk2LH5Z@!0Ci?b_wQ)Nz{Sfe
zk~RGNPd)-=84d?ZiR3u5a*h&aIWZf>+3Lr{FfiSG&7lw6T;CI8<ov}YRvEVI<~y%9
zkUh><HQH9<`cz6>=pqO~Mw|wP!)PZa2xX)<ZRB*GDT~pflJ1=nZEP+}c-Nz(LhD7`
zzobc^p4^+-EDxCi)=GN7V`q=WC9k!?V8~gKiYAcJVT2MF7_m$>Bm-DRh9zfG%qW$Y
z98;dKNKZO3r56{wLnAQv+8gSoAq*qut>NZ2kcwi{*4S*ZIT4q<=V6>M&hz@sE&u4F
zEn!ZCT=?bNp5I(g3<=JvilPd;S@HIF_dJ|tZVnUJrJJq`yD>uSNm&Z6)6mloJ>l5%
zy^`1;j+~~MFh%;~o)~A&w=22@uFfM#uE@lrXoblUYot3zF6f+bDlKl<f;A3bcf$Y6
z!hq9l;kdl-AUK~v@!xaV{^wpL=Vd~LjP^396gb8}eSSsqk>qCZHF262gPf7)0gAih
zk$xNm9;`IhSZPS)@&wuS&SPNNV>HExZu6Tp|9j6@t(hjtTaT{bs*2m|JBCpXv6Z(N
zZN&jQPK@Kk<<%vHd%`s1tS6<3!)fNt{fWw1+R9Ve!t-rIRoC3~k)A@yisJ~eD8{+4
zX(6jUT4gq8&-nW0$nEV7F!QsY|A_ZKe4jB)m|k)7`VEb>M4kEi&0DN-ym<M7@BjGw
z_|}ud%x2w^ha*bCFwWdO?6^2varJynw_0;NoY1#>`Y~Wtp{Xm{MswU9Xxk1<!M8P6
zFJJKH=9Y07*q&`@>ISPN_U`r9@2K0F$|UNlruB-PBHlYfSsDWhQkK{;XRPSHr%1^O
z(@N()E%A>-7#YKiaT}(2W<?<vVUWi3ht0pfRZ&-RASWf_VDs36F4%>oQI^qM)^c*S
z3(Qv;jaHt@FWI5a5sS2X&C>aS#TcNat7M#X+Q@0Kp6lh|_00yQmNsfkNd%&R{k_7(
zXaP2AV}$XeCAY1ph4sdNKf|vS`LlHnXPumQP)uc}kCLA_pY~7G{;I05`T}d7VU5E&
z%l#L>=9ho*-*Ef-H6LGAeDu95UVQj7+86JMP_KV0g+uv>_UwWW{_KCo{a^pL<l~JX
zIf|G)w2^}CaqRL)eMc!u$)pq)&CWNQv(ZZ0=<>0MogK;Qh2`M36oDGql)^kOn>tHp
zLJAW%1wu@;s}Aq16o~W09O4paE#vo+<mW4|@yq95N+$gg!mfwi9^Z7BlBkR!C3!gL
z+GDB~ox#@)SjYbLZ#nG`y!YxA&Q(nPiNoy;{Wx-Q^@5^}6oom<3CMf8vooryBKAGw
zIHTr3h!bg^c(}P)M&pL<YQwChgQirb<e7PY!+3kdsG08UTskjKF&}3#8q;=^dWCCd
z@^m2Pgl$?18dDN+oWRyk>#syv=T+qiITLfC_7!E8!^-WKUr?{NtS>GEMC>Kzs2?FI
zOS2T4`TT;sE0{15&?Kd?SgfgKoc0JJ?OLi%(VqsC(d6bra#)E_lIZU!jAO#LEv0H1
zOQEZ443RLN=)-_D9;}y+=F}5&k7lN-Tl{Ln_3eS_bcEl14%)E2xTJK?QB6eWz;t}z
z{>v};@WYRZK+%S3eMakRPScFakuaTTS1VA3?)-`u+Hm>of<q`=y#E0%&OChi9ZA>A
zl~F>@uWMP?%U5+S(kUrL_7zj(w1NdE6dyThM!I0Sl!av{r~fBWy)ICey%V*JQZ>p7
ziC*$I7rm@f<OQE%m)JHCUS`Td8OU0+#i<ywc1s#g5zu83hi<{5wEX)R6JZ_@=W9Y)
z>C6BqZ9o|*qO`*s%i6)St)XcwMk{J>>71f*hIX~WSwmeseLtbJVVozrwr1={-n-~n
zcNP2HiQD7Er>~EEc^~LQ=2zE8+NQy1%N&Kqp9^?0PD2ivGLLv~Xe%%2L&(TEroh9)
zj!%F2A9?rn=iL1EH{_CdetAX6k?(!*icfz0BY_M*vOdeBqc=f{gpi2^z)(8_x}b7l
zvR33Td4}->In!u`uAFrKlqRNvQRb-=_Fb+JA|8ROO4CA-hME2KJ!aRlef2$*&x~rI
z*ovlU&{}io2Ojn(PU9r5>l_GErl2WWQR@y{d65ADah?Usf^hpD8wk?8fJY~E%<?Rc
zC6lsZcQ|3mcvomzfjiGrB8Hig5_QvHo#dwGIdL41@_o=;??#?)TF$p?jDjj_j7=P;
zk=kUGZP`ze!;l1}g(j4OGm6i?{w*rQrt^INlaKl6qYs#-k(FO_{pJm)!vg?s-o9hi
zb$IW&dj5h^W>Odl(}>X!rUB<HxA!xrVdmw_hRyjF?>(n}U<#4rF!Ox7qFc2zt}ylk
zN-Lhddckq%DM(c1@%kNi_d8;Mv(;MioTrKX@kr-utkGPZZ5YQS=9G|_=YX$jDrZUA
z;fy3$9Osdk3jHwCxQflTWp_BpBxRA9`Kn<U1GXr9fujE&#;>98sgc4-<t4%@OP-l_
z4w+CYs!Hh9PX&_}mYQ0`zQ#GR^rwjYZ$mCD!NL-~QPnL<8_^Jnm5CGr&Usms$U@Xw
zg6OT$Pbmlbsrv?I;W|ktOBT7DwB?2rYC2CAaMA32!|f?RRc5A#H=I&tIPA$OENB&|
z>jq6Bhk=JLK4m)X`TqC6$E&L6lTSWib$&s&y~O%jis5<0@_^a_O4E2l{nLNM{O(h9
zDNBCqVo6^{D353pd9$3j>Ov6x%VIWpyd;uH!=tUDNP(A&FpEmgu*5Mb)Kjr2kxo(?
zKi=3WMR^~9yxKekkV~$eB1|LmB;6F}?4r$)_pbRPj9<t3gnj)PZneQ)K0`T&O%bh~
zxQm_W)6#Gvjw8CQ@fR1|-R%g&K**Z86!xbF`8TGPykJw|iY6ZSST=-tf*hFp30KvO
zxsbJ@tt-5h;!pcV#`mWO(lkhcKf~+KzodKEV|_)@nxYG7=uz`Pii3<-OC*y}S)qzB
znI4fDB}J4Gy{15D9hJ9CV-N#Mx237qY}A@CJ}?YN!Zedp!8C?dx1xX8F`d4ms%qS3
z3$`PfR_>)382SUV(p0`7q=a&g`s`dX8rPaIjkw0+N<!s8Of$|HO1%N)82gb9N1P{S
zBL}ODN9%@D-&1uOXDenr)>cfX$f@5`ug`e?u_a6+7>DzY)6kQqf${E!k`v9UWwly!
zIP6fsaOyek9vHXh)X&~$wY_3^xIqo~%>5qFyuCX}hw5;HesC1)MWz2oxJ2_NrY%=_
zjM9WSkz-h78nW*ht)#$5GOgvld%Q;<yI$!D2_m<YhN6}cszf`1#O^%`_r<WV{4PZ%
z8;|mo$D*z*_j^&AkU?dUm}u8vd<9A}OG651RUuJ`4eJ%mVP>8O8M`494z0yFVSSCM
zI$}}iqIkKf=?LT)DM?8e%Ne{ioUPg=%1FbkNY(<v_)-c=j&zk}7$cv)KJbe#9{7*9
zC-!4Ow=1g6hJKK)tk+8xsWwkj6jynSal|-soMu8uoQ6@RJbC2Zm!ENHTW-JplKt&9
zb=z<{?3qiZJ3EtoO&$Q)MW0AX<eZq}L`ji?rmAb4Hsqr4&Jc5A8hYY*Ko3WO;~R@x
z#&XFT8f!^)LyF&h{+e@Po@Z3f%fno>irx7+X0;`j3^|~dtmnKmId;1fZ?A7L&P!1h
zqqyBuW|}9alvwL$PtzQwlnCxakS--WeePw^b3cylm0rjS#&I}JynTDk#~)tt!ykUk
z!~HG85b)Lt_*^Nr+btm_v{91GWG%xm;+?0e8mu)q4}NK;yfRxrvyYB>krXNC*!2?+
z4<p0ai@UuxT)cP1v-34KufGx*%6?DZ9~s9JyZwpH>Kuw<8cq!3fsNS+)SXN}oKR${
zN+P!M(q+#jlZs4aj>AaEinyRywwsE>{y?#Y^Ye4282R<5pVQSfm!EvhfB4%^xjUZd
zr^NHi6=$0@#adJp{d&<rrKHnuG}m`GSYz;}aJD`p&XEEPb0lR$iiNREXrowfI))Hv
z+nQ;fDaufHYr;74aD7M8e^|Npt%^Edw<v2CIHd?UT^5TkD8bEU=`zjJh;t5CdDJ4T
z&QYM3)>xFas8X<caVlxiujHj2?R>Qq9U}Hc8{#|*gm2LRNKW3e(UpWTl9i^kS-N#%
zxJWU|2}KDcQ(ODw1~N;5ShDC(w9%9pmm*3^ACy9;o@syLcYpm$`u&OME#ov%tsB~P
ziz2{y<mE+2eET_@ZO3}Ep>A885yuq9fT>7Dqox7%@H@#`oQ`-cF({^yxp|K9Nyi7v
zh(rr7uN2udAB)cAZ%E<1bjV5yPy8#CfK78sP?BVGJ{E~(ff48QBoLJw-Yi|g$FYN6
z#tl+5=A1}T!p>ugl6|?1Z&FOSe)u*^>+ig-p1<UL`wY``k}nw}^KhaR=~!B=39%5v
zglRguvoox3XxhRwXLM^>H5E=P(Ja;#K`E^9jA0_oM^cu&&0LB+fTM_rn|4hGd|h+4
zUE!LRjA1yQ7)}SqR8YP}Im-|gPJ{KHX4Nv8o@pK_7{a1snofg6Y1xX}NOxK*2vOj%
zBuVW!9VeWX@yZlus;0&Hitb{ArzTA!Va}i&Nk`^+qH7yofB7Zbvva(Ef%7#egDR0Y
z_L$R-6lP{1YeU<tQDJ7B26SEu{G6DEBdUn*G&?7TlC>q4MO0rh!#rT-k)jpV^D8Rt
zP*DU%x|&f&p<`gTyGIw$x$yGpijYJnW}Rh@meMp7qZsGFs_jT6;jAO;iEI^zz9-zi
zmEX7D;mKHEFYDqtU;Oq9RMj#SP30WbX8WDj8;Fz5`HDq01&d}i=f|<992(UkiAXsS
z^Y4xNlt!1bxI1SVedZ#7Vk>&1QWCjDk+S5B)()felA}CgwL%pMq)#D<RAL$1rj(e&
zj9u_0DJe?Ki=0HJ5IH5{EC{JsdP<S*UEOy0<Qe+RFn$B&Ez5mhlwnK|=D_z}TvAt-
zwN<1!k;%-{Knf$PuBEygs2fRHNKSLonv`J}BhxTYV!&={hAA`6k<afB-0lW+y}~)m
zbl7vayOR-=wP@ecuGd&)m~z2n$YH>nvIL=vPPk};a*qCZB!`*v7w_@ny^lEFe2sU8
zfAnYnidR=>q<Kc=gtvyKl8Cw(GX&|5mXsy>%Ss`ck~9;%wbFHs2^}*!MU;x@oJrbH
ze8MTqrm7YU)^}WQ-l0o`Bwf3lKvj|@Y6^sD!e~d`dWMh)<A}DN6fzV;jx$3)i1yPy
z>MoP$f`=1xh*;}r>sm(tkD0JqiTkyP-mc_=!cdk`yit&{;%|TVhKsG|>iHGdw|B%4
zI9p3{!yIO=p1qK$&}H1;bPY|>45yK~7#<EiFRof@UlEdK)hV(Lq_ASXiPX*!m*Ps5
z%+!y>IndUQ&ALKq%`g7?Gpvbx^3ex0ttUqK`psTo!pQ<f_NLQ``P4&GGavU1haJ|{
zoDL^W!_04AU-R*Y&(Snu@Jx}L`y;~`37tfb&N1=q`E&YZ8h5tY^7EhlgunQ={~p_Q
zyn6nE;rIo^H1YN}@#133+FObZoW@b!!%}cMaB;rkaJVNc$U&avT_-VQ$|&x4BPhug
zPKet(rGW9C?b(^IPpwCl#Njw_`1|10UA}ssRa2vt#<>dPEJ_<G3X-%?WrpvVrx9Zv
zO}Bp1f|OiH%Xmv^vGzOfAEgR%lKs6)E#smTC9cR>^E9Idp}wOu#!9I3BQ8KNAo@ug
zl2cr?bW4ld*yTjNSiJ!)I$YF(2*^wE@~Bwu6p#XafxrAW{~o*UxcbTa*pk`pdp4Ub
zKm70oD>b6+bF8m$RV8}9Fp}d0D4buBwH9|>48+p|gd<^pPnZM7x2!H-V*QFzB$h_r
zn#ZoETZ*yF@8eu#-fZN=pCUOfMVN?6#9pm4kfb0)J&l)2k<mlWGL|iiwnuBdjKmFP
zd7ouOqbW+XT6q!qvoN`G&cqPO_VjI**57%xn{&EPeu!<?WE3XNgu?^V{Vnrw;IUw@
zyu<m1^?Hp>2~#=D)g{Fwro(~ha1gA7QTVFGH7(utj1Wec2I<}^N6O+x^VYCgZyEa|
z<Nl6=Nn)TdHC9<xtw+&Ov_@-jKjj4TIMJU5xyg42^t!NWJVTm9cUmgONn(dOQ!(!L
zq5)jib18+P@1^htoO4XQr>a}5uSNWfAz6#58uYm1)SqaonrBy+B3!?}!8a}U`;$B@
zRe^ba+Q^l)GLDbZT`k5CVvtTxx5a8jnFdNssPRZPp1N*09p(M5iYJGN)(SrdP?G9m
zs~Tf#%u;|jBSlHIZU`k2O5|+4Wf~@|Dy+^g3D!`}fzH<yl?l^GyIP~Q=zvm5#Oa78
zW9ycF2<+b7P;Jg>R%<rfbF{NmR?*+@zw>$n**CP3j#+u)l@+wk@*pUgQe;FaU3I?M
zXJ8@rCzQ^ZB1s1!MKVS*p_hYZ&Jr<Z7U;E6g*i@yQH*B=MM~057w5KKP!xe_97xvS
zeI@9NtjH+?Bf0OfQy@$;r(t9akyJF5tEj6AqYB1IS3NCGPEgBOQ(oK4i#40pQgY;Q
zI5GA|Y^g9SM-Gu;95E#NhX>A|T~XDB`8IIuC&&}q)=?WnNtvk>GKG&`Y@u55?i3lv
zfz!=9(mZ3F!+MV~ik!rJQ8x|KJfXGb;V_7MGelyT0Y%+3#A#$W9En3u?JZ&KDapY+
z(E4XYEi#q5GOX%~IV2AAh*nTl4&yXa5N=w|qBpBMIcN!gu%J?+M448QGKvT>p`1q>
zk2Bxh;~zpmt4xWJX`Bcl<Fle_H}r=dwf4BW=G6C$<Ahe4{y1Q)6@c%&1E6ji#wf*U
zE|6j*%`@JKYdOV8UJf5ACPDC|DC7B@$wkmwS{a#)rOfdpi7w~oEAH<1Tu%r7<Yzy_
zYeS&$-sLmU7L_u_)to(hhIuGUCU3_ujfA3!OAo<$yP&x=`$M4e1s4@1$zd-{6B+rl
zuPWy3`PFZ4xq0^$Kl|}Z#;IqDC%*jZo_E)GJb!UURVTtQVvXYd`iAGvE*TE@?C;)@
zLgBCv>~{xNtB!;w6~!>lI4eId#DsT_x~{R_Q?J&5V(r9a8G=j`|G_`{Q-1XyKSe9}
z;N>%>5E$l(!*Ne3iB;F23V5&azF`bKq?z}gKchd6IM+$4PT%9aqjHYR^NL{}ah2ub
z{0vHFCUSjy$Mef`+O}bydLZMyXS)BxQggmtqJ*_r??sDdg{-Z#CM)5h84AvMd{yJC
z=1KIwc*mvKvIZ4J=&v%UBAv1v12LCHSH@y3D#ST5j#ANDYnP7OqXsOCQ29~Zi)@tn
zY2(Zbzb#RcL}h^lts5f?EJa~x37_gQN@ZzJo8RjMVo1KFzB=Pa|BHXkdmnwo$7dPB
ziJ2B%Z8<Nulrdra8K!DvRB9_y4p?gmF%ypuXum@HLL6r1``5(bKuHn<G!C~Y1?#I9
z*ybE%8ZsKS#Jj*UD$Wv5BqN@si?@6pT6whfqGw!0)gWUVMR_Xra*Cqw(VCJavaBrK
zxcmqu<?(xegOlOWxI%tH(4?gkC_k%^T>Wj9*57><i1+t2rY6S3>E<nE*rP1OY2-2F
z+`EQmeMX5BZ@&1PQZ(=X@JGZj5|u~;)}5m$#XR*SuQ1zl{HkHP8PU3+oTX%qDT)$j
z!aOn__sqvVuXlHN@7SLPUc7vdcD*8Jg>n_Jd=@1!3<JCS2f~Y2R9COSdrF9yP>3Zk
zr9|y(beg2=QM42e%X({?hV^<S9iBzJoMIyMN7OV4X<b2ec0tmz*1Z$S0wv>&ra#@W
z+r8s7j+|dTqZo~?YPol-3gauR^^$b3e7@d`erygAWer9-!r_7A{T<r3azL?`f}(C#
z__GUQf5J=?ikYYj21`s5DK+*3#(3g%BKI@8a<tZ=#zb2;^kpPtSgp>ayFUq{fi%-p
z4ZDXOa|+;Us&0c;g%S(CYrrU0+cTWD=%UC5h_e)?e~c?<L6;~aPDf*KUc~32k4&RL
zy^Gchg<qiHi`zl&4K3U+v;3VHBQYnGl|yJsOC%uR4H8)t2Kg@qa@K^9m}8QZ2d%M7
zj%-dr_Olq7PCL0@qoA2`nwZlq*4Bu8A}@zjh%s=OCrXZxMW>lkBwLNQ9k$w_jpM~x
z&8w>wb5<N~??}f7NHb$=IGzf6F~3b?PhD5k-jU)gLiGCwroplEhV!natqjvN^X`z?
zkC`zh=4oUc1kUZ9qit*It|1_RS#88jay&f9LA<C1++1YD7=xgulsG?0U=hw<-JbFB
z2QRrg>lP?<psp-UQ*rmO=XyM%Qsivg;;dm92l{bX#*GjTM^uiOX@-y%Gest`6z@fW
zN|YcOzK>Il?{XPqF)A_7v*_E%%;U_dzoFWmvof0Lc%&a@ls34!5z~y$%wb}lXX?6U
zB9R^rBBfD^5GKk=g6F;S<ShHYR+_wU3Ne<P#FdExRr1p3P^i2}b<pgNGoOF{72kjV
zimsM|+Sd)%d5X3~m1#Fy#xYWbmZq*57k%<!irnl*+RoFpVsH%m5v?N^XO7uq`jjz7
zF_nnU9#sk<BwpVfIqpw%z9tsS=U=>`Xu}s@UxTtdoC;sP`yFSMV_SQ^_wknbbl`s4
z(^QV*?!epE*KAiSbhqMHU%W;+N6dxRdHQ2dS2wJ?npIN?L85M{npW;%Yj}9LC&hr(
zh9CaqM|}3TzoKh|4QTzD=$&sKP+4()zQyRwI7QCSH=@6-4JKrqu>`BBS1sc_GtD!#
zGo+&Fx|Y?d6-j{Ae0_b-&BKllU!0TUz#I}<S=N6aoVu9uqrO)j;vtgCz?{SyU9tpW
z4_ZG7_~#gwF{{RVFS@xLVR7PwVF6*r$TSB~ZV_q9iDR52Qz)#~D;nzv^CYMbz`Bf5
z;c1MhHBS&;39???qe>!D@+0I^D+&tlJu%3*z#1bA8@=QuB8G{l@vN(0ea5SgYtCQw
zPzovDVpF2J{1De{(4oVuMu>w5ldYATDU8IDnEL}^{~%*uSK;cKG@axWs0?w~Q8g=0
zy9b=FX{}%2hSDX{df^H!h6%J;^f?9iCIO@5ByeN>Xh8YBZk`rBVOgSs)biy=K9$_v
z`bj)pQUqhfC7Pwn_oVqN<tc!@EVu>f3Tl#HeVe8Acdxf;Anf1L=g2q(e)}K(JBGWr
ztlxV@<z@Vzb0HQet|7-Lo203jk|Iq|b?5l)1+{a8VPFVvao*y*qgywm{SEUxQj|vd
zj`Qs~$NiqE-_xzP)Yh}VyW{nnd-nSsZL{KGf22LTV$)qvH4SCxi6IdCk+Z93RNa>K
z#Vg6@Gy}QJc%Yvqobzm#k;`tJ$*#c|L*I)aw`n?xub5My?|UjMU4@VnaT*D8kg;k(
z4o^-VNU;cZLAwfNJay~HXqt9K<!Y>V%t2gjx)cgY+~!)zMjwOZ&t|YygVUC-SqtJO
zXSBCu45sQRVL*9@>Ka0r$YVgjxfyFKnr$Nou^5RZa6k22u3uo!Hp>A}L7zG9PLwdQ
zZE6al=Oahf4(54cF5oV%utuZi8S5(aiz~c$OuHS^@kq038Pmk-?2p5#qagANQx*+e
z#!#pXI2*7&F{Mmi#w;O9=T&6CkB(NeNJolfBd0`&g);W|N7E7l2NzfREbjNX=wuh!
z$uv#`1$E`YikLno;X|q-`rN$i2W6QqXeBZZrN#Acb&<|Z&P>xl2r}N)%Hh_R`0XXW
zUXfxV^?TwlvD+bzww&-*xM<@za&_?>LrsX0)pjPwK!_v0a`dMY&MEGW1KJpF4g=?H
z&Glj67jJrwQEJ|llc1xlWtv7_Y#smGKl>g<xBR!ie8X1{NBSU>l~ja(T2c^Gowjly
z)za0oM&td`k!x4{=!ZY%ryswf@(!yN^E_e}&Gx=T@;pdkL%Wy~#L%L$7|O~VQ8A!H
zKq*1b6r~rZsOYgo-YiAlA0zv9o@POSDVeO~T)>EKb2vP(-|sPXi&+b^Gv<Udn*099
zG);1#j*0#LKu$@z`yt?rmDCwkuq@+$XXUw^%u}-^#YkSJK{?I{{{h_MuFg4ed$;4q
zuioeT-~WVR=*b~)zC8zHDO$L1Er#oNcR25f{Xk_Mn^nv0;l#Rjyn1y`F^XvpIB$9W
z{G5lwM9~sI>Dv{rzdq4+)#A#RgK*b5Y=L*zC;H<KwMb`NrTF^hHIz(weomY-?cI)f
zKJos0&j={qy}c#OBhR0|<hO6`$Qi7aNW5v9CAW3xv6Un>QA8WNIy+}~f6sQiVXYPO
z=_KyyIq=C3KjzJsUr>3Nr-^k}(^M}o#;{t|v|Yvh-5%>LZPQR<h7g#hKrXOZcbvI1
zPD9Te5>40gu-~)0f8g?BL+K-j<G|Oq2Ue?=);p@ob2=Qb<_~lDe><Z@6riniqa!*O
z(xOchi@l68V_e1*S)5{e5qTD+h$yUYNLm<k3MfjlBbD}OEjId?B-^$21)l{9TGh3j
zLUefI?obv;<}ylr<hf<QxJnjlE~H{bu;`qOi(`<{iZj&K3tMJ!e>yLEBI|3>-=V-d
zreG-biY80zdK_CS6=~MbvA&^{4$6d@Pvp=`<V*^rdEj*Y74zW^>pfaqbPS*i&es?e
z({YDN0e|_Dc6C9Hf$?xhodc%HsCYs(S7_fY;_0#giRI7w5lN7u6eyU5CnW`{T+58;
z5o4gW6ov?>Q!NjavUK(oz+)#+DM~SvY(><ckpOCA$zej7{M4l@IZIdKJ6_Z7o^*OZ
zANSa%p}Sg9x63-p8Rs0PD8iI@xVx94vpwVN{DP{g8HR(5ZN>!EijW=V>?7*SmzZu%
zDl_K%W75rQ(r~~p9hDl;PN9;*uGXy9TbhgKT*gVXq}Eb(E4pq&PLWkzGb#v9K^)nC
z^(p<k-_h?6)OE%A2j53+-lMB$)bS0fKT%4i>Q>Nh$!P2lu{kr1J4zg&7{cMete|c@
zR$Do-IWIAJF)Vq>MvN`mh~rolE-#;9ohQbC(;C}$INy-Q0i6RX1tL*&V^IRMqjF#@
zQ-WI8=*r>9#4s`s1A3pxT9dbd+Ic`kzFJkNM(z_~V;1VV76)GUK28_XaHO#YooD)K
z#-zfka!hk1s>Cqv1sR0I3L34jl_zDzbUa|zTYTF=il}*HI^7fc6KNh%5u7TbOZp>R
zj|UUAupbkJG>_;KaRhv=Q8Nsa6uTiVqbsej+K6sA7OYo1Ht>`(IEU~EV|aqThB+}S
zaITQ&k(>$&fq91z1lQy=rI;tjxiN~E#3iehaK+Guyy#wwk_n12OIIWY=4m9PjB7jU
z<|Tf02GvURxiOMsWE@XiR+_7e7Vj*(haF*(q=NIc=jD4Z@Kr?^2W-<(Zvw~LJIIAo
zf1+y)c`T$#QItFb-W^WdpC+bxW}Ie_urjRHY|gjrZf-fAGMjeIKl*2X!pqf`fA_lw
z{_@wq<uHz<6q%=y6lRf?7>Dy7Z#7pJYi=Hngn3}K>DZi~^H2ZzPidT!BA84`N~8=q
zq2q{CP|Ra&mcl7X*~yEHq@;{0l0~kRK^I|=iR%_|3dbV2loC+tyF2_-OsFLrU5sc2
zWgU6Z{B1Vp*scR#GmZl>!EQgWyF19U$~nqBGfm^NN6Yt(L5vf|SeXKOE7-3X7YTq6
zD#ZMgo(yg*T}i#L8B57DO~q#25$2Km<B^vyUNX*s95So33o2W28h12p%MeaXF;Y8A
ziIzUWX-@RhOqU!b8=BUUVy3l;o{5j1tr>G7ZEM1u=t(4}Ng-1(OmpC2e_)s;UU0ps
z*qsg-8nl8h-`%rrJX4tX_`P##ZvoHO@9uf|{0!$6Uw!?K?Tc4fA|>*4*t)*I$5t)A
z@-&r4Q}EW&b{$iUthQU;9(%$lj`4Qg@zD=H;r{v^^Dtty#W_t{y38m=(>Ov*R8569
zmMM(vK9`_@m;&BuuC6XP9D2U|>MbP|aw^>2Jus)t@icM0|C(_a`RL^tl{Xlb(PjFB
z*3`FJm6i8Pdp4s~K{>%+n#I*p;K^w!#eZ24A&d5I$q7|j;cO)$aTH}SHTYT}!YH$J
zg1$ik80&G)EpM0zXYFEm$V;}Q+^!2nTp5&FxM5kGt=_tYSEiQcp8S2YID=3!;;`0>
z{%DEKDN5+iwKildIY&~QsOtvrTeNdnBV*+<_LMl0#v}1~&vbeqjR(@y<JuM0RVXWD
zYul`V!hAYFinyx6wp*&rB{>Dc;g)$CsGOiL;^9k7dr7e?C|X+N%Sbk68J%h?%(aq(
zbk(#t?i86ECEE{eme?HWC`jk-QRuyN^-6itIF?-G>t(?$JQng~`5C08Y|P&TXRGh7
zH*u|GUdCMHw4@Z(dWCbMFDoTW!1FMmQpQy+-TEBuJiFU><S;W$z3372NXAp0Z!l+P
z*mi|giLUFYx)pa{eI^>a3|2$sJidN(N-55r7gU7;N+b*?!u@MBhU~noEng88_|{;Z
zKz`?(7;H^db@)<oyn9VLoLF}qahkDuARB|ynPHfSd8Y3Vcxxngauoc+`PquP?Xa#Q
z#YlhZ8Rv<rY0=g*Pm%T688KV1Ew1vwD8`2b`E(?VN4a;kK}APQ0b@1U)@0j|OiP?5
z=9qCM5vL$7(7Gj=Bp}jugVh>LB!no36l+LnSqo`m+TWq(9%~(yH*(lhilAHc`U2*O
zGM=zRs@hU?h4C#)7j)<;X=aKA#vtY;D}_s#Cm4I;G!f@XMy|&b#ECHcVIt!H)TQks
z=Ta?MsYnMgMXXy`l|eL1Dl4o~ge+KLtsobL(O~VO$1UKrXn0M^62qk=-2ko0ITGf$
zFvUEbZx};nRN{)QD@R*<>dK>)l0CBsEkFu4Bi+IzOtqK_CC=zn@O2CI8ozo*-JavR
z4cb{$Q4FKV3%pVS7!EUaXE7QPXy?Mi!;!`&HtRLkXmPrS^z_ijSPsVlZ!O+gj`v4S
zQ^GleE{Rh=GYqpNG?a{Xj+jLb^1YWI^66<{o(BHruRi73_MD&pi+{?;KehbBU;T#v
z>wo?K;8(9-W00<u^`5qCd2i+U)9+pJgXbOpm%sRoFYhCDW%(!n@F!GCBqYu?gjx0^
zV=XDnRK{@8b%c~CQFM+fMO>8PIYiO^8DCQwjW(4ST4r%W=b{!zv0{#q5(>(H_fB0H
zXGSH52&Kfl5N5g8a#kD<2VBnRI!m`#8*+?zYw5ga)wJB6mKd|eUG0oR8<|w45SWKi
zid5(1;E|&kFsu`!Pk8+K(wUax<Z=3=Y3i2yeb29d^EoeHJY$ZDG>=?8e??yC3)&er
z7aQ*0eSx=zc5PX&D&9RD=yT#`*K@XRY1)P<2sOc4i_-(^wdZa(QlD=b$C1$q!Yl>L
zFvGit6M+aJV2vV%v@|iI=rChKl}zp6p-&8-y(7$ps<W)t4cE7O<~dT4c(}VmpKW<~
z`;HJJxBH&D(^Or%m^mY1p6I%c-Tt0ai8$-|^6M|zo~<Z3a<=Yx{_-V<haJ;tz<Vo=
zDJ{*d)ppBazvFn=bH3S7JA?P0&DjQR#e*<RL29C$B@0`D(vegQT~qOJz30o<x9si@
zyu4a*wQZTEzXzj4?Hx@m#s;lrk(4Y32yF%In-<}`=!XSmqL*Bo$GN@$hz<Ucm&4LI
zG7e)6#yFnXam&d;87uk_qp24AH~E`T<!{E2336WEWb=&(J&OR-IuF{C;w&?CYoD+K
z-(Xm@wD_~Q1+_8K_BR%d0rCP2dF)au5%CUt%KdMN51$kJJt@rOl%%WJtT4tcW(+|P
zP%>%i;W0qlIk1+9!nhXG1aiM4AX+%<oY=MNGxX{?+0^LYcd;rV)jtI|=O~3(DU=wc
z(33YC7B|@=s^Re}Pe>FV{d~*kBPZID<m8KzF;!8I48i4lC_~W}Wxu;M%yo`cyQSY9
zP{3RgbIHsZeC1$ab@|GZ5E!SgJ5K$cI2=)?LYbPRz$i^qS@dy7#^Nuoq)QRR)oH65
zm4$am7;vg!2s{F3OID;HdQG)t1o!<3XDoAn#CK~h&M)P4#!Rt}zCV!;JBb8)xWSYF
zDX||X##zL~d77CN%qd}%VLFYJX=E~mF&4(uqWyEy`b-!`VTRL=q!dOQVgZX1_EJ?t
zo|((YXj}aGM`Z1o-+e(m-4n-ym^&I_EaemlITFU1VlvS|9*@NS2Au|S2v}E<wWi)|
zu$8B>61*L?1Wu!ky!Uyc6ybM;RM5&%lp)L!HTM+lpsgjA%vXdwOZVJ2q@tudP9e=@
zIb0PH=tpq|H%)_45W_58v9_gBV%o`L0sY76)H!Eqnpz|hx}cT(PF4kM?U<&(97YPl
z6!gY&JZ01pZ&bO8s;*E<GYzv$5R8^IuUur`StKEfOt!MFSL-bweDWh!XBX`6cl_#a
z{+64&d%-~%VS`SKrYuJ3T;=p=P)mf6g&3d|MQhyl8Ejt=-HNh|=t4h;j?a0<(-G!C
z(NMxfYc&o*epKE{SL=Mm#wVJ(VmkIr<AfzsNHps;_qX>rZ5gJCX&PDA4sR{fG;=)l
zq=a<NY{uA@Y#dX_Ow+`(y2gI^5no-u<u`W|SAY2{;*k0AKmRZIXaD+NK#Kg||NPI<
zTCsb$=h>_G_}QO+#Iyc$e7j=R*8K1PC;x=E(O|k2&R5K%B>#9TygqAXggnJa+caEm
z499UG2g#i+b07~rQ{NLsp^x?st?h!o5b#Va(1*;J=O<Z<tdH+r)Ey58bgJ;)5o3OO
zMh-(y$`F*ME(Nb0&E|}v46$gA{e&S=IY$VQs;U+(y5%%Yq!gZ#mP%UCI4si>h42(Z
zczRCj@_WxSga+W89R5$^$jw8+7Ejl<XzTdlPkt==Jq(z_de!pf?OS1-dQ0srDJQzA
z;qI{GFw7WTQ8hIvsEkA(ZnrBinim(Vr_U?JnRysE9*(@bKXIxw)~;x*<}?n(oT;6s
z@eN^|+2?^_oViz!)jesNc(%RZMQb@7dv0zHY*#Iwz{A}gVH|n$o6pJZid+<$f-xe;
z(poc(M?&Zc!;v_^c#Nbeay$%ZJo~&8^Uh*Ci7{cE5o4N`XJy^C496q8huvc4^JuL&
zJKxfefniGY$CKC%N}+383W38gv*{|VyTW=;iZeI6BZtG0rmFs+HTCU~CeBe;^^(J>
zF}|Yk7~v&eOqGPxEr|yfZAD0%Qk1O1ML=w=m8x|a!R7_6Ku#po6EqRDbeptV3X`P>
zECTvOn<WAUlva?^(nVZy@3cTTOO8^AT8A+X=NI(H(vh+#iS|(v4W+de6T;G}P9>Al
z(!l~y3g;S%f;3Nr@dV=zoeQkr$E`j<_xF_XUI5F!0<DQDJmo5eX@U^YCE-?UOuHc|
z8Hp$>482HUKJ8FNz`MTNV8etO_n7SmP@l_dUc|$qC5%!$$C<~jn<_;LNn>T~vW!_3
zk8Hce)Us$^o=g$2xGtB%?U9xBSS+Tz;9$z)&=u`Z4u4py`2XKE4<obo#4w63GKEFw
z(erQ^X_^*eELKZ~UNQ1_y!W{EmSGH_pj)lTMSz)ee<B|CO#K5njGzt1du$kB9#E<<
zo$kr1V3yH-&=ys1(5@2o;^Go(yr=P+({V?L6LB0xz^`Qde70#hzWIW$zxV=QcQosc
z?RJAcyOa%88+N;Uy!W7C#z;pbM}~1?NQUg!B-5Z*&rr2tzPl#QBIup_BbkV?6)9z=
zVWO=XTuk(D-%$1kjP>;UTXN``ha(YC!-%U-=#n4_v^1t0l(S67J5qm-EkXEMuEF_=
z!(BtWxnO(oLe`P1@ZAQo91KjJC7P)~)-}dClvPagiIOK&j-=x~Y8*+Wpv!WY&=yk)
zgn@A!2>p(#Y4E<`Y`tRXKcSM0zP+ym&7*|rrL@I0TWa-(ZFv8wtFA0(n|3i7WELMh
zpa?ND#|#LJx3-Q=*Wrz2jEQKB6n3?I9;rlB$*kJ;$-#I!8~`~KmlP>kvuZnDe(*g$
z`O!~UZ?+uwd(u2o+d?9vv?C=kn%UY?RcqXGI4zi^$go81xE~dLF^m|dQy{B*8PO#X
zYpVid2&8eKgh^5cD0H=E)p{ClS=AnAHR-HjfBTN>>mA!wOWSzzJW_XShFP?G$J2=t
z1DnRv_=0zi!+y`_V~;9@>(fXnl2dA(=h21z>DOQJKl<sPV2$N84*Yj-?zrs(AOFKY
z<NWN5?|=L~l!C77`0a0gjqO^_y`~A4&Ds+ZR8_^cb{q=~!|()%)*9-nr9Wl*!+~+P
zBhG?ds;inY_rxrEvT>Y5n^qLwXi5ykoJiBeJPZtD;5cXc5Q*ATdAne>K-up;jWBxA
z>1DNyLk*_nMKkCmqqOq0+YP2^7>0={%raus9k-{Q2>FdBCSlDvt0~GbjU(9?ss)Ve
zjKF<!kS=e|SyGgg6}drL*og{KN!VoyHqMFrL!`Eet2ywa7ta}ET*wju=M>X4F&qbK
z?=dB^Ia}lFiqrfW4vKL~>`w#Q8QfCvIcuou2JI|O*I*2gBjf4FVK-BvV>3&QF-(C_
z)f{YQ&RHfQ$`G|C<jgS_`Z02x<az5&Puq%4@Zs)6)vXCb&+aafW8u(84##^om1EmF
zj{AG&l|`KjF$SzJ>>u`2?S^-69yr|XFui4W8~FK8f5<Q#IUSG8{Y2NbQlA8Q_T-YW
zRYjvj4%Q!z6hvRU-yPAe=K9^OjKfo;YHF@;Z`p1;F3vWnV-IRX+d{Y9a2i?;5BI#g
zz5j#O)VFiLHfyTY87VD-UM;M-$}bygDOe|6A*B+qXui^nMdO=J+Q18&4yEM(KXr}@
zs-&g!=f$bHI9AJ&2kU%AN`V{~L9n!n7T%MU5c&+(I`I-bCLGLJ0-FI+STF@khl>ox
zGEXOI*&9c;9_{=hUJZnl$p~-G+DZ%ph$&$nPoy}BtF_)hE~N1QaiX&MiN%M~_>#qO
z8z+I#mW=i_uIbQ~U@}siA&lraqKrdZC;E*vFHEI~(TRBf8Kyf?>NB)<@Tg0I-_yuJ
zK`<_`c*{##M%+@+>5@q~VT^s^)rq!f>Gr8Y4omVuN{gU0%WE@6bbMveuqY{nbNud|
zlW-d4J;{N79wqu_4vfP{1UB7=oL#)YxJq24MSg#!EVy$*xkqI~$bzHLrQlX8N-P`>
z_l)~H>ZZn7&*An4Vy0cK*xx^pa-fV8d5*M~FX%pc4_8&PiB%3g4lwR{c=tev3Af&o
zTtzlXPJV+Fe`j68e1Ah{GmSIkVsPiLuu7xTj7@>@_LgJ{rE1W1M}74Y7Xs;S$L-x)
zY7Oa35!Njw&0s2Yv!+=-U+=wW(h3aIKp0Pqrz4@49Mfr>q>E!K99i6=xA(*#(N4x%
zd63Pc;DdlE2I4%C#(9Yyg46DS7)I)PMb(|5wB+fk801h>qyRG35a$zCSJ<i|nR5}W
zCjr^2Qc%u|BmH!bE`=FLDO1%gRkvm;YRUawumVO7WNLwvqm*<6jksXbAHj(T&^Sd`
z*~QT<$t|FWQzXQUQJ&{#HQtJQzG+&j$}>(g#aOzFOS<!OD2l!xuqZCBuCUH=*zXwo
z{)rDKBK?fl!e6_+xnUd!v{F2O|2@twwiF7+)=P1#aZZwJ?4tWs#xPDZ*LNp=`NcKw
zPCJg{L?~I3V2rR5OU`UpYf8z?{QxnN=b4fsFP?3AwKcrBTH}bU8_R0dGEaeV|3J>7
zt5QX}fiV_N4=21aq{XncX==PxoNa0jRb;K<-LVI)7#=1LLm=c6T6vnL<>qu^INtLw
zKK_tD|Mb`R`jWSUq!FZfWP7&aXYEgEw_7frT`|rRUhOyyU($F*V{5Ma%yBqzHwcSu
z-87_mqCXvpK5}@t=i%)&{r*HKSq|3TVzh`ILs~%V>M={V5XONqM#f=g90$fZGUq}>
z7@);^FdF47s{ZZ=iK?!MVIs_8mJ=AXIH+Tmu4Yr$#4r<<Jv)Rz+cl_=3E>^9wx&-q
zwNOg%RdXMfLwaExP9iH<kW~6nN1f6)FGOjK)53ez!mZOrV~S-CnXEk#i?17TFwcRi
zTTwL~^VGAu-w~H}c6Pqz*|SUfX%cxy6rI>tZ?B~oZG^Lzk#*Is*R0Prlq3b6DMcK{
zQS5z{*Ti5jrJ_GLcE^c674A<b+U<&Jg7K8Z$Z0h(7w&ftcoVqO1!o<49C5|7dpK}7
z1P*;-KOFhsd`(?zj(2z5*TO>0CGh_HSKNI4j%Tkbrc+OvCz`e*#=vKve#ZOnzl2<v
z=Rk}*x^+j>wND1mx^1XeEz|KtlMB-*Bl4W#?%gfN!--rn>-C1J^3-+3@pPbFC+f<w
z+eHq$iGH8RYQ|~Jy8Xk<hu^NKt2>PKi|$Qs<`AaEhOU<rN5D4n%PFlzYvL;u8si&*
zO=^uYPP$wnNRhV)>vM`yI91J4hfSDj@^$ShY@C;tawbjy`alK0qrei|8mo<r7?Ttu
zC1>)2`_PLkvn;wfwdliyNhjryc^X3&N(yM_Wz#|}TpxiKD<v*k$z7FhkZ}Tc4AVdc
z+#?GrMyz*~BG6@6!m^|2ovf>H?V4;WwDB?yoDagfN+XXMk+$AoYAK3Ko|&f;&RWvL
z7nFJj?Q@E*q);wPcTOR2<B|wV?ru(#d`(-hHA*4&_Z-PiK+!qNdxa{L7(}nJjL>C#
zF8UKGsuP3>j6)f*<foLG#>2N+T7T!2Ov4xx(-b)E_RJxnv?a_lShon9qs03RGjU21
z-D)+;6$s*zrX(3f7)O`}Ow42ww({s)xc=%5yN4Z{^_pt!;Or&3t|-SN@%9E^H{^L_
z&Vpnxr4=m|iqn)xQ=r*;RLW=-31iP`90X`;9cS--DEdxinC6MPT}l3;Ybf0XY1dO$
z6~5~*A!F)>QfK-xu$^XFHB$C(iTND6y2RHo9dG5cj)5=_*k(n$-ZGB^W$J0RXT&V2
zBq&3gN8+%P)m@~sot2>&$(xN~B2FW*EM)tVm<LJDu-3_8D$Mk|9nQDp#YADWI0$o`
z$fXERFDKG;;?zF~+|<c|&ia*TMduy!f{F3IVn)`L^%WRPjG&TG^kY$iN0}#55#5kM
zAlfP>iZ+CKV0_qp=k*5g!ShR2>xQaoFvc&)qls}iGR=Xi>R7K=lvD(m>}xhj#&W(}
zvA%pocXo-^<_WXny{9DM@Z~3jO+lqXjx#X^Zr)w<#php;$Xs5WbM^cM-DZW=;HnkI
zRf68&5uy&D`0DKg|DV754Zr-&YmVb6pX)L{EGc81r6?ur(8YWa`x7bzG>Nuxs1iBf
ztYi$ks_EJaYb}S}k(@Gh<*6!9w^|eDnJ+&7f)WC4ZAqtsF`7PRHtUY2(o9~_)*5q^
zPVlA{BggG=CXuL}1gQV{?>^`M@a&TR>7V`yzkK}~dwGVdMEGx&qPx1lZ`Q2Zikq*$
z=C40{&3>PG`Mr)+Tf=tE>)*ZsRai9*XPXTQ&F=P=Qs8*E6U5M*P}a~|ht-;t6Z13^
zQx=UKViE{hfYI|5=?5|O#3Yy+oD<Yf%4qLswimed<~y%9tpvmsV`iQgNsza4FgN0S
zPjle(a7VR1W0*2^)nKcd-SB|cTDoIeeKY!N8nkun_9r=@FQ9k<Jxd(jI7vE?G6sWi
z7Ogens$S9|N@3A1i$nF;7sg@Y;qH#M>qH7w*WBJ*b2=Q@9|p=eaQ6N8XxoOnhaJ;A
zVYTBlOoXx)mvxY&mCa_0>snUl8;o(xDJ>JajO_<8h&jV{+frLY|M0+dPW0nMh35Ii
z2G=;k;4r~al;sc>Tu4*%aOeqh;QVY&&I+PozuS>H5R#{^3){+2SC0GP#Jjh*7z+Jq
zPaJ2oHhl5B*OHE54Eu*Wo<G0h_Vxx<VAV-3y0wOJoT!@y@4Td#cu(C((VtSLZ9DEC
z_Kfq$es^4S>`Q*|B8h6-j_aEpRo!rLae+2Hhr>WWOV|4kP0jgMMV)bO8HIXb*pw`;
zQ74R-loHx{jIXiQ2{k<x3@cI$GTzLA9HNLijY4M`D?DlEl#B=ETCCPfdV;Acv@){k
z`wC+`Sm~HyJ=S|E=2D`d3Cqtk&QZc7V@jncIWEPW7e`&rkmAw-kZeWr%ry7dW`imP
zlLI9#9V`tL>Bi0d9tmU*#CfKma8*s259qN+%@ZlcWnS;G-eFC}JRAf|QLiYzBWq92
z8Z}R-=|so@Rnig$E{;;_)lyU%j1{4~^$pAiQVP(%AnV^dJ*wq1STH0h%;b`=$_g^!
zc!zV6ha1C$34xq~bO)Sl3<c5Yg*a2<wCFJ^%A>uyD01u(u5o(mNb*OZt?v&<b`Lk?
zd4d>lRYkWsC+mXFiIumgc}A5QSD%q=AeSr!>>MfMgwdkcvPPpzCQm1Z!wJ`}XzLZ%
zUtja?^=mM&Ba&62dv=9uT2Pu+7rb}qoSDa-(Dxj}NL>j)a{K(5Oyrdz&l9c)sv!i)
z$t5MKW=*PFXgB1TnJ~<~So;$i-EC-}zrb~C;==>+c!#SUmuF|JoS}6NZ#5;3n0TTT
zMLO+q#!z_?sT*r(np(iXN*vo^8Yyui7uY@AQesc#D_q^+U5Br0LI@mo_e|r!QzR9#
zo5w{yA=uS|CNmBvn&Cj@Ejf<NM`2pcVZykYlr!TrG4)4Eni)=ejDqduGrCJnDwcjK
z<l@O=!l^`<M>*8FN{W*B=;#(xl`;io3v(J#QzXy55I(d4V;K9MdG5daf&<=r{tWL7
zb+=jy)JzI9)3}m2yk<?kxg@2+cznPbjqy(IW9xBEOVcz@KBHWOuQm?7Fwhn#^P|o+
z#|f<z+p{g&TKcgEl>AP8#E_}lfvV{!D5hZ|<U$|dKYo78FFt+E{&-@ZXTm&7f)54#
zn9aQy%2G<~_j_#4q`pVzfYl1ydOm#C^88{Woz{6^8YlL<9sOw}D~(E$_k23;x&8VL
zyW2a?H!aQ;DyJYt+S<@omSMj`XXw1*2Ujikr-@@o{OGEqax<Ubo;cm#Qq>&~CG&s&
z#ozG1{)a#3fA!D)F?|vPK$>P^KT)4);_(Q>#H!oyZ~pBsIKRBaJo_=*tA@?yC7(U~
zIlhTltuRV4_5<Tyw6gs;(GL?TXId`~M{O-BW#(Zd25A&vtgP{vm}YV5VhjbnK<x`v
zRZ+Da^=i%KhaYnO`~~LsJ6->oSEk^rikN0XOr(^FMUqN<rD*B~R6!dfR0%Zgs-quf
zQpmJzLmft()67Xz*DY<=QPjx8{y+#3lVneJRYPU%;`9}1fH7Kfy%*zaTC%(!CsB{C
z>p4c|Y&iA<acY_LK)2oS?%ivS`<`LnGtYrdQ`4?mrg?c*D2>&My4lc=h5O^o^Qxw8
z8mhJxEwj?Zk}%$)h=e&~ykqMGIzP`7Doh+xpdi`Z=Uv10Y{NA6w3jRP-mo7tWf$n$
zhQ@n_81TMg3W?+Cf!2>$2XoGZFw<5wjzF3-^E7kVAE}%t$H?bjyy5EVlHa|%rfE7p
zxH`vS=#K|Jc<&{zUw@6Wj@|BvuN^O5KEruK2s86o=(J(dLDrWw0xO;ZAr`cAY*td2
z+}=O1KMlOQx#n!UC6~<cbO0szm!_+z)|cGh+%f-wEY5FL)a4K<l@PTvO2(Z^OP7X-
zP|y?O?snOPiezrq6)8u;+~ceOP~+H>mrX7&yfrf9lAuiFSxMF%nX+i`NST<#d|<7k
z#HEvE9Jv&sFXtq(04<qzF;1fI(M!?-pj?Hqo*ZXV5htvE#E+yXEoy6JJTaYsgo_E6
zEBXZ-8pl2T{k!GwYeEp+V7<D)SdTjOl<~kg9H{E{X|!pTM!R|$6)K8$B&4mbQ-C;;
zhod;|O2Hb7v6^Wb1VT)bag6g4`H~XGStujyzogVxEOLL8_Htv92}>l5wc?%(Q8b1E
zW|q-IiIZp|T}@IB<yN3XdlylpX%d}L9MNW$0=h_7Q@M%~#LbM-5>xgauiyRV*TfXK
zJlk@9wm})gZoj7hHVml#h+AE-ef1HhsZk}PLqt_ItUF?gVrnp&yc7?4JTZg^Y|fMz
zc(}Xea6Itry_ZyND@DRENd8}#Shp3ipV0k;bC#92SZg39VvgvlCTmTc26C7gP6=%-
zV@UFO6~(*T8$x==_WTT06t@q1=4oQJTG6d)+RD;+i?3^Tr7#?K9QTRUro%gnby|`D
z@<`lY6SBs{DDiCdVqBO;4)@oDDG{cDv>Yysw@lN3Z)!*qttP<SZ@SW26|9DSj_ACY
zVwAzH&Lv`P=uu?mVZg>jn0lt8CdXMac87tY6`PAIT-7p6lOQ?5%>M2M9TR1qWt4LM
zg6iy2#uWX55PR}?!qhD)i>5UrdBuWY$tWukTI1;`9ro#jCep4}C>(9$A^%}u&i~Yv
zbHca^>nzS1QV?gK^J~!v`xd*}3d=4@LV>S5w(^8|VtUvU_t(;`o+ffijKfGtfy!wa
zU&TyJ;wGmAoU3KCS5P@ii8H6Wcf`=6QB+OCW^=}rEuXzT^4pt{o88Fn*wZgL#3?4G
zDS~m><@;uxP(}-K#b_BP4?U^xCDyKToS(I<8&4{k5&{M|?2oi<h4-FR6!S1}`!+E3
zJC3^p(>O8NjO!}G9O(BCth*JbhXZjKa89$TEOSo$<f5h36EB~wd36qrcYOAC&#B)L
z;)u3}|LV{Gf*<_tFZq{0{vqH0lb=XLVvG!TB6`;$vbnnApZxQG#{cX8`QPxXzkbg9
z|I7aYzy0*rl&q+njLlMr#9rL+Qi=(|pmLVVdm$7U;ae&)<Ipd>Q!kjA7$7T5yT&WS
ze0#&Rzej1y)$<pey?Vvvi&wmS?>*|;ak#$z?h8(rg-8I(8m2ii9DA%WG}%(@NaNQS
zE1h7c45xm;S;uz0;qDMvya#~dcswzM8SQGEm#Icdg%BqC{>Z$T+{E4e`0o}z-qVCf
zEscu&q$|(Ga2g}We&l%CbGH5e(e@ubwq@CxB{r7a>IhR;xJP(Al?ot17Fb}94R(qk
z_($0yz+XT%*dtM36<Ar7`SPg<5BDq0?5M4lWMgi>nIMvt5%A0dM8kn_am~&-d#yRY
zZ;bbp82IqVpV3$-U=+N0`$W^Wk{wlw{ccB|6LC#k=g7y`iN5#jj(hReS|`t9ke|OY
z__m|gLYziIF0{kIdRoxh(zl-2XNJR$!_g7jj-!YB-t*}+5!MN99Ns(jZqIFA7$><j
z-sXjHNp!o8U2EBO9g52JHZq6AZ4SiCMB^P=!R5mjKE7VDZYC}ZJMXbhGtM*KJAVA>
z16|wm^67=^H1op`@7N73%QRDR=H2^uOzVtqwu9xglB&X4p=xO0cs%mQk6#EGUT-sp
z-b;i<S&XVIw;AOV&M371FW`QS^My;e(6$}MdXiFPWvI6cMpZCgK#}8#X?jJw7C#&)
zd7<Qj)=2h6ZTOR%zS$!xm1fvWaDQuvWu-`K*lrOmDMWIRgPY337FLx|aw-K}BlwQE
z$aH4IUex7EPI1Gc^a4!O4h*PonVG<{j<mj!f8Q_VkD8FIU*te$DIN2CV>*3-QgPiN
z>s{B$VJR%woG4%Zo_zg^?FLFt@<B7gnUbblkLdfPC0I))&R0kic>^}9t&$Fz?(jei
z0Zo>M)Fxrpbwx+LL3agUsrt86<xynOJ#7UBtt{k7DH*Fp6I78k@h=4SS|QC>t}kC`
z_mA{<kEmK%&o7LpXWXV#abC~}H3f2*DJgE$6Pe(K@z-m@!ZOnj9m8&5*zbukGEEoe
z)0Mo=JX_71zuM6^o|Hz?<%MuMWA}IXcb=}h7yT6lL&j)FyoY_Y)aAy@m!BEhiu&#?
zZ-4g#uIb1zOI|4lOi4U{{le+>#M9#&VX~2-82VjLN{RJ0;hPps1?MQ<6Kg>`k2Vh1
zwk)~u^79Au5=3^rEGVm4<_o=ED2?a#xzNocX*{E-g0k>}RnWjR&u9wDb>dCdnq`@!
zz*7<b@;C{mAO>Pp*rp@INZ6E1RAD=&n1Gq}mnP;Wv8M<tE21dIp}jz9Z`YB|TZ~p1
zE0eU_cqL3DB?ifmmckl-l82@jA@z8@GF>kaBInmv&X+6ucUSiBeuH%l=*n%JN%KUL
z63S>?Q>D<Pq2`6<bYn%)?(cBT9f?3%7A{{ta$FX=-9XuLcmFwF%RJHco?5dMj>wd6
zUTOKl&0&I^Xq?^{y@9+Gfd?-$p(Ii&#BFj<%wmeD7VRCj%G6UEO|9atbVfL!Rslwn
za+cM%vy`Iv`wtU;{~UR~MOmX3XxetmL&_zs*=z$TC)R19A9mcXSNgsa<3|XT7)a|v
zrqZ>R#v1vqw2n0d!V(zz4r3K1&)jY|@;vi0jl^|?7&+{Fn$DvXtRZn5S1#9;-YE>3
z$9>1V%uqGGH~jTe$K#uUW47#u9s923fB5}prZsSW{!BdH^Zy#JxZNH9yJ;bYK$s_q
zI%p-Af@S8v`9J<Q{Pth{D|%xHx+BdaZDWW@P!f`>6*CG7T;n+$dYaa4K<N5yqHU}q
zEGuhR>4zP=-GFX8e80mV-w;xvygZXa#x|Bi+wuP4o}q6saVCtfOrJmg>i63(I4Kk;
zX9;=cvP7JzXjd?)(4lCXj+hJAX#{O(eZ!(6-g%5E-}-wYtW+wyu4mWx#5D@UIt0`*
zi<ZrKDJv;0Fn%RD?O#fBDaCE0CD9|9I0Mzl^V=vIZ~5}^3!i`b!u`{uJP!=q-yI1d
zQVI;io>ahT$Jei~O!Le*M~rneZ39S_JLN3-Q0rxAPz1fDyOy1A&^e=gOXnMI<H)dU
zxI1>VttP9&TC9MwRpP!k47&lPG$ABRO{nZB+VgcBQAYFjZeZ6c4!c%X?NeZlh4ZwK
zQliO1fC$%#jKY-4ZC+`uqh|R1j~{sR^p<rAOmjdR%kO{xfq9&H|LzUDUBg;tmhnc4
zk!1}M;Zcg~bz&HL?(SOFd6xI3f>;XkII-&lJ7ct_Z#%}Ex!y*M*6jNK0`TF><;ug`
z_ssK%^HNNR;UaKDYbhy6#te<I4r@g$9ILcs+ODUTh$wgrB7f235;3X>+1oaxBByfa
zy&M)O)VK(LNlREY#7NBnqnE8!>$eA2p_WKq165U&b{O9Ql`x%2cHAb67SkOBoLeP)
zU465yUK6TVl<}nUONPbMld$uRFkgvbre+ayOJLbzo1V_!(JU`SBBgDyjVI+u4wa=A
zN(%V6-KaZR>8Kn?)0MK`<j<=GAX=*;=T%A|%Z%@ZQKV8LCn4;onlX8kBx{SY38M_f
z9Yy<<Gx&xa;<mc~1z21#wxOy(+{zIt%-i(P(30Ttfn`18tVe~JhLyU`K%nGCnvN-m
zGj{V}I^Upuiw?j3A^ZLN_c(7Dx{eqFDF&Lx;@{kJJvr7?=#P(3AYHx^*998OGM=a(
zR))ubwhcMU@02*9tfOlhUSB?Qy<TXVo~Of+##@xrIIZZrmbU4!wNT<B_yesNmx;b@
z(M}XnsaCP6%#)xX`horao?J4qi!!M*hDRq4KxZ6>`+F&^C@qtN9FPKw6iyENj?^0B
zx{`CjZR?Wj6j07{ynDp8o_u-1BOugi2{`L$8YiojlCVzHU-Pn}Qlx7e(pp&Og=L!Y
zPKpOr*=>UB&;&VHG>&Fy<zQ+H!|@=8*;>g)bGx3gMiJMUl;qEuVwS^QPB`z$IWvv3
ztZ%#{guv^Ek6gY!)BX5+ntq3ETkanoa7~Za8BKvWlVT#Lh1>HB(<Lzc^*^Wi;jd8D
zW7acqUbtKqmg|i-@1C&UGB4p*zu!Ren1!-6t)<8nQ;_0J-SSoGigz7t(?QM5)0x65
zt0rfuRb!msQKVIFwh72D))j;TV2!M~bH!Ur*Y~o14lB#5vDWiEX8wmC10P>r2y2kc
zWUaV%z#2`z-=l2*4MaYTBg;HfOOjTkb*Q?zKv(&h5hGpixj%F~?BsWKyN*yIzHw3-
zC>dQdL+@B>!})aM)62~Jy<_Mt-dip&XL2bR-!M-rFQ=KO&QSA6$%Q!tv@+Zu2Ob^|
zJaq?tyJoa5EX&AkZFyZ1<9K1%?YTSb1<AB7<P=cWp{l@&)9uRjAO4ZY_wQ-Do`3yM
zf6viqhQ@If+7vlQT;pkrNa}qfO;v9#F(yJ-8Fo9GeqeWZ$9SEo7b(NcYrr%Ow(Ur9
zWjtLdDf8~l8xDs9`>w&}Ogg>r=~<Yduh(Di8%_z771F0NRy2N%__M3Fq#{-&vjkqo
z6UtZ)_Yaf`(=^jI4KdyX_eI$(8;-thSyus!>#dMj;wsF(rWK@75%;h{P*tTAQcB-2
zVN@AH$k9__o0chQzFZ6GA3yNZPoL=fJt-E>*B9Qsd8F?;V%Vqx*09^}No&D(EmMpX
z<B3?V<HDe-FaeSOMi~ffD_id|jmJ1oT^DXK5Nl?*+tK$2_WM0KPAn8u7TeLA#{*AK
z_c$wSP^B$DeTeLbf%9BgLgI(tzTsgo0ur~H(`jXz14|No)#Lq<niKOZQ~2Z1(pp1F
zh10ZRn})Sme*e=a9*#%mWn?3}^7+dNRW;$Rqw0cEGKmXIBo(O2psnY0xiRc^biUy_
zj%cN5oJZ?Q#S&BH_4<l4(iDF8^z?0A{-?bDtV7-L{*k8fnBzS)ENE?TjWiEy&M0R=
zJBpI<Z3#E(d=pMnPQ<v#8nlRxqtq<4bfc+SGp@7b?ocG9DXSvJECABYpPIMyrrN^(
zwMtmN<V2cpXeCR^lw{&kb0kh-yJ~FPVq?iAvb=tT?N*?ESw(Je<ys<EInZpx?ai-Q
z1bmEAgt$t{042DEm~nTGIA1Aw#d*IKK!isZ)|uoR(#HR?!x8d!b6%!ztIMq`zBP>H
zi+>%jvMw#sG}X~;)3Hji4&Mzjy~|K<AHl8U<`I=w>Jnu-sw_GG`@>yMkr)@uHi@V?
zf$y-!$?qgfNW7Lv;~W-GE<%Kp@NJ@|RXnrX!EjwOItEhs^)T3!6W!x?RHYcdeh{3(
z?ttCxXtTx}hwBE?EwG$k$VSub_rzsl{q(~8`WZ@PofmRkF-FlI?iu!bbk}hF?zdRu
z@m^v3R#qG+Y)-`M%(zArj&|6yWcjjdYY{>A#+5V55vXWyG1iL!+-h3Wl1gUXK8)|5
zo@g6~QQK`=$>c}3M`{+ljUjA&!EH+88*mM&WNdqf-XHN#-{EzY0CFv2Z>WltRzeJv
zdBrpy*YEKC0pA~3<{N~OuwLZg-S4Q@;`)JF<bm&uM=vXNF_@<3;qfiXIkdG{{hsx9
z;nPolBrglYuE!e3x~&#n(@HaW2`uYO*9+bz<|K#GBJfy&j$3M2h*S)j8d{otkI~{`
zUgIdh)MX*9Gs*(i;ay9du5f)JFC)%sbkW?-SG2dJUk*#Z@})h5Fi~s3`&I(<>nz+a
z-^etqA_0djfWEF*oN1(hq&6>g1?#0PFH>|!&XKUro3B|epg9TWv(!rC8jN);DPpYU
z=hMo+`TdzsU!S>6Gcn8M_I7%v-5pr3XLj#?L;vuWq6HaZttA(s<$GhOwXiG;jWvX2
zCM*F*CM*-@^Ob4Z^YqZ8jlnv_aye6%LSr@F8s0u0xmwMqk6)QX;$>7E$G|wP<YnFt
zjs|TM+8X8zFV~5ov6PtUTZ8jP7@HcJ-to{V{;TirxU8A;oca4NC+2X))l6OjaoMK7
zDWTk!0jrMx_P_tv{NMi1{}-G!{LNqe!2kY#{Tpgtxs5A#cXte3gKE7TWHi)V<TH^)
zf7rN&uJM2(-rqqM4OPzaOsLj^H7s$Z*YLw%|0QqUKFM$aMX8zdxKi34e|Oyexxeo9
za+>KnxA8?QIZNv_7sb3TtR<41A)L=FF?09$gu?Lpd|{p!iZOCgTUL}cv~7cL<PvF(
zw54;B_>HEDx-UeWb|?WsQMRdYSXWX`8?r?fvs#OpD5=D4nfU2T;PaO=4~L%CXxc&Z
z^72a4b^Oimeut@*P@q5T#f<TKW(`50+r}`i3;+1j7b=yu@l+@Hm>hyk$C1MjT1p#6
zTh?4q*06u`fNdL`l^h;(#^p>+1&?B9J<AeMHE|d$##*d?;`x?&xjpm!+dY5vw%f{D
zigk%B%SuLK568n3$Du{n%-7SE*J(u;DdK7EdATjb>x|VJl_SPaSS?U)=N!KAyuO|=
z29EbT=4GYp3zzG}%V`o%y-v8M<$9Y)-qSQKPj~Vh`SHUiK7Bdij6(rU>m&nDVHi4=
zdH$2$Q-9W>Zn~aOX~wwAmc^0>k5Z0lx?(8!cA(Ti4z~?Iu(64XC~|5kBFl9aY=hQv
zqt!-0d%v_+jqh;XKv4_TILa!3R+(@>TyJP)C^g}<qM~K}tBeGwwIwTqQe~UQ$(MeP
zi4<>C)HcaV0$TKq95_KF&b3nN3^k#&`-bfpjACas*)@RvCbG{t6W4|DcBS9H$1cy*
z`2{vI{d9h2Sr&%JceLGJ${!_ugDqm@P@s^XlN=g0TDST4p2->oAd0r6ZLQ|pUe><O
zlhcB;4SD=18AZ33pQ#4z2l9L)t~1`ZP$DS?Db=hiSS!sc>xs7)%DO-;q>`xC3rfXr
zrI;!i2&2GQLzqW-pczLkkp4va_5a+f>3e?oi@zq%6VuD*?O<xK3i`%E)1Vxj=aICm
zDDQAvo-&le`STafuP>yqVz!BqKE1GOd-lVTVlDQmp?`S9`Ub5uu4cmd#5&C=<nP&g
zkIzZkX-X4qk8V5aDvYMKvDn5#O2l=<nO-=KDWkXb!kQ9!nb0+JzFhI$p2NK+rW;Dh
zO_MZj?@1%3lBQ=hj?_5Zox^u~Ow$wRD=}^F<(lY*0bdO3bR$}68m`Nkp|Ldm5qCHc
z&M!>kh|$muJG3<$gm%OD=t}7u8mr}y9#?Xl>AQhp7;sI?t!Y`Ni5wGi$*gPG&_P=A
ztg6uVp0@Yw_Ivh+o{}rxH}aXMu<r-7GFac>+eYN<RE+CcmxYQEQ;jw#1MBI`<>OBb
z)^q#%6YKSrp>O!jZ{BdeU06dPjyKfiX#TY?XD!z9EsrbhX~8MgDxY;}w$3c;RW7JS
za<*1i*z&O{3Oh_2hbp4&QmTIAgF&U_Uy5KwGI~JUiSYY>_{S@M`_n6~uKdNmaLy~I
zDKJmZxOirH{)(@fpZ@I^o__e}sD6j*JAB)dQo$N0W}OWLAH%{rPADolMuy{#c?nXM
z(1vARpbGX!pz(^h$}~g10|j*a)3X46O$i(tNXtTul`*V1XGli=7cn@mXuFoCZ6(U0
z6t!l)d)l${n)8zQ<Lg9CElwLuyrQdkbW6&Fyb`BEyWdk&<c~l86JJ04k>k@7Zy$S<
zGu+?rxU7ZGr;+a;dphT6tTgp=E?6f}-kjpM4TS>QzCk@b(hohVZtqcJC`ANXRE2L_
zbki~~Gt+G%0<X)0QJNuijMME`zu!KdZuFC8zZ=j-cx}3I%*(>Mtdb$09E*bfaEGy;
zc?ry`Y^~;PwYsh=yY`4RhB-xXnp$|cKX9=lr`I#fJV}h8X-O$yjo%uWWkY#sw04q1
z-V8&vh=*M((QSsTxviOMGk2|}JfE<-^5*F+t#?fGD1VNA$D9JUaokqnj{Wh#<$PmG
z!jbLz9%mFWB#dRVH(6oV8UYc9X(p`;zHPDI<2!+s$92IpEhZ+E4UjS^tenp$X{GiD
zoHHon`26~b`@QAO{hq-UqR}jiWZdR8P;23M9N4uDB}x2X$(6Y(%DSSI!8Q##B*|<m
z1<G)_O;}a%u3?!M9`1HfV47D<<D_Jyu9OPnIAdJP?yzT?M!K#;RXAU+^4(}OUE{fq
zfi*`OZ)uwbr^S8jbcgvbv3^}*Ml-{Brs<C~{hoQckX1stUKBy=jHA-^2T(><RmP)@
zVVN%9<kMAEjBTW;m=dO}*k%w*MT)F3lE#Tt5_t_YZI7i0wOQ$J&Crqsgj!0*Xk@ab
zEjg^zBJH`7Bx9lVc5~Jmiez52Z@1g6tdCL>HBtd?r>`{5(ey{u)|~auvkNzZlSyIT
z4s<!^Z_?~`cZa$EYibU%3eFMNb^`mYqVh6LCq{%)Buk);2&+>-nIf#96wq6Pty0KU
zTD>CvPTPM^BmO{>whyMI8jICEas0qy1Iq4Dx?mAT7D}VMql6^J0F)qHv|_!U$VQMB
z%6K3HUb0+y!89E@7v|~8>E#Q%-5#3~)AbqK_PFi=V->C&{+zegfAp2YO1`|Z&NFe9
z_F#%LArwfNxXhf;m~N25Os%LCP&pFji4Q-0<n`s1rt!imG7&W|EN$SAKYhk@9ltdi
z?>Y_#ONcYq@rsTM%6fd$Q&PqnaV)OOjheytJCt{jGrkydiIh;VK+F-PD>+AQudjrf
z$y$KfsTQ18wBFOSEjdPBzJ4L3!rk#m?+wO6b&ix0%QUgt78)&?m1SZ&zp_pj;yR*Y
z!kCOQ3XP$09o}|?=|YJ!m&?TdZpOWRFBz@P53G}bb;*UeEaWiLSViA;^uz8OZfKe&
zdaLQJ!L<#){q1ilr7+DC<9J2&J<d6lHo~j3j&|75w;hL}MHTT^x~|76m|tI|fvwC|
z*03_2C>g&Y6>_euYmf_;?`Zcjb)UcdNV<NN449F1Rlh&bpeP|R{fW2O|J2J^BSr`#
z+{e6(pbOr1xVD#RqZU13j1nT(#*%A+9K?o$lFOt#F)3R%Px4bqK3fP=5v?)Kkk&v-
zGs~*^!-q4UpU<=^(>IF$>fN5bZRvLhzJ5J(xlT-L;NSi6mB0V_EC2Z6pXeTcD{W<$
zx5HDRhKy@GzOh_ig#ky1cm?~uV@Va1<>lpsQkBkItkz7^%v>|XKuLjh9(i+UP~YG4
z^YfMSI5Un{1`Fr&gwdM2oug?*?tVD#sYwo#ZRasY)9m(4w}r#Q9VM+aPLW<G4zr`~
zJzgsqzcNiXKw*?H+?<S(H{^rnzy4qRmcRbXANc;xY#Kzz<FVy5kGxzLo;r<TQ#NUf
zbAHR474jNb=c}x|Lqsb@+jn?nXr0G9188#0jLXdF^um%P!@Smt?s_ut=@0)v{zqA7
z|Jv7*D!W}L4csc29Bmc8X#j(+1?$|l!O=|fOh}b!ozcn>LnWplC?YDlibBg-hpLgz
zdCuqSW_41e5LlNPZw$_u?O81AL+32N3@T7;mjana*6pwxOQdN%70t3HOrzQFI}ZDv
z*VhyMzTxqBM@p41pFVLLZ}_eujth<VG=0yK0|^;YlvFUT#k&sU95H8{wP>vf%ZScV
zurpP1>xB(ZR%At;V-oz!Hl2NXd_-@?pJj>MZYyu@kNoDHm*L7hqm5=i3=D^c*YPL5
ze|p2?L&toZ$jK5)WmzNZy3koi-*&jhkV?ySoN3#Rby+s5g!D_aHq6t+Vc(Fe=Iir?
zVa+J**$o}qI6_R~dN;!Aty{DD^?Vk*ls0(pP&z}-wAS(d>4B6Ix65TKwfv`Ff1W1n
za8H?DL3t?vC=JcP?)XM3xaTjdw-cXVUg-N9FQ*rF!-4y!xAJA!A83YNV7670<|{Ep
za;@a`hV?7W?m&%;oW@h7Z9QH|9<G{(u#DeUrAEt%SvI4YQlZ8m;_<wNm8+bBz15gz
zha`6jr3OsflhC+kz#z;hYb>aZD_6xG2*Q5KXu`4z2(WEPi{xxd;hSn}nQnx`irc+M
z`<@y`>eqjVt>A}(-#@_c#B@E;I3enz#NmiiXj8}~qKzY_AcYO%$Z-{}&{iywH-;E0
zkkXK>b8^BPi*gRXyF=&1Z5`#{R@Ii1a*z^A%gRA(3n`G->o?wBLW&hyZ4l@Rr5a+C
zTbt5`!`+jBe`}?`drvHdx=j_+^6L>3)9J*|fBUa#hJoX*gB-yc+TEUL1JD2XN6vr$
zBmLo?ihwSMzQbx5FK1$m6s5SWD{2i=@W_tgZqLK_N4mBnj){N#H~)tHVb9jm;n24<
z`v<IR(MBG;#W<{wr13_$JYyS&sp6NcB@;`gDw(#rrsa0Na=l%_7<99hORPhsfUz3y
z8>S@^l0r8<#WYA7wNhEhapCpl#Ht%iw@0rtx@4AVB-JRbSJkisviFo~WTNb(ZLZ=<
zx?VV)Um29RkvqTPE>yv|hOn&kT?bCdf!Q^3Xg5wSU9QD9Ms|%w(H(YJYuTp*<E3M|
z-B^}|istd|3ELexKE9=IJY~5OryIjzkLx;8N`xh_=8ErnoNYiMpZ_?MS6OA$n$fl3
zZ9~eQ{{EhnGt=#a9tKK<wLr-d!)RPX9Xg6>fA#wfKx+xrOR1_QAPhfYnc3}P2&js4
zek&nqVvNEKR0UP)xA1u>@*7QQ#ceT#N|nQ|wv^4AgI26l;1B=!nXj)CKCd{HV7l?o
z{^kFQw~r5?6u*m=by>NMBaeUo6L0?R54eB#6aVA$R~{bUQ>!M0NL&`eI^%<9nGHG0
zq$;fo&Ozhgx&*$yp19w?mv(H4e0_ap-&mG)kv4vbbWMX(ifJ4f8iOg8uo#}NBQeWE
z<m)XEYUOF)^Y*c)Z5saY@tLnP+}$};E!ftuj91#eXADvb@w#%q^EjnYDs%e!Gw<L1
z4c+dD?i<4MGb*plw-<i<<oTf;czSzBEScNoMmr4jL&KYg%JX?ab$9F@j~HjjDSY#O
z+o7XVigj67PA`;r!zc-(TWxUGW4vb`M?QRg=6oHQmWh%Q)>>-Q(e3ZpKO6`#b9w$E
z%E@2(n$`tn4pQJ~8s<5mz2vGc<tCqzoH0c)&nt7Rgj5*Enb!A&P+3Bh$(E`#-jkC|
zcNA5kAWg$EEyPlYIT5Fcti+n+m6NT4G4j3sOaAUZ$k%z_Y;+1iy8tZ9LfbmpuA}K2
z#&v?P3;*KruTe_#>BC1Zm-Ch+%Jf}_X<A+{KT=9%oL8>X%+NT{qV`;-nQj<x-m@+v
zd0nV65>l4jt?z+IxZRjvFX&o83wFjh$HU{3Fyq34)0)fcnZ|ndha;6jSXY_UIzx1j
zke2M@(>)J&4U>1QYvMXCoGv5I8s0rU&}xPCa^QWv+<3iBcmw1pK44X8ohBE@JO^Uc
z%*z7S($5QT-n`}QTg8{>7b)>n#W>A0&f%Qjq9+w=9JaUY_YHAgXuM(Ab@YA1rFVS&
z^71FWr~a%%T^e6$+h#kqYfdkpS>i$;g)daMyhv?#Bx|{Q`i{%#OtVJr-hMA72W463
zGfj7tRa!d`myxgM3vJ_g@I9q!Ky+6he*8ew$jG_m#BsmFw*9w2c1js*W;B^*IO5xZ
z<#r~fL=G$Ma3tg;n2)%kmD$#v2G<|4+EKFjB#pL|s!*;W$Cb9*Z2{XzRfTqUfQ>!o
zm7;e};EpMwos}<;`G&DpCUe$7|A0BXr^X3g5?Lvx^B30hGxe~SfVvS>iP8dbR@xJj
zgeyx<n4BeVm($jk*{&r@kyAh$#X3(|=P*_#43tL5>C+kCyd@zku(VO+ixY@41j&nJ
zQPb$wOCvW$midM@7NZneiA92Kg+fl$6sT2gZDMGDS<U9SO*($vt8E+Rd7;0%r)_#l
zSSVG~y90G^Sz_U~{GPx6!#~KMT|;9uyP+e;z;&E3UYb@-(}|92n+(}h=#=KP%yJV{
zmb=GChTRTh6g4NA%q2nA;5?~HWH{eWEGK~o)9ui<#To+@v=<1oQi|rVLpw`c0y$)S
z(^E?o4`+tUxH2ZWRldw;OpdfF@%GqYN+!mD-tRF(2esn7!5wyhOoviQ7_G=*MfslH
z;{$$wK>J<{5w{bkpMKAJy~z7oTYO{jtxV39uXtnSKuV$Sb`)hKo6|ISwzXYM(rixR
z8CAe;x2J0wc}T7i>oiIyRGJBCVmW={_VShabx$|!QC1qIP2ZrjB9*{4r~+dosBR3}
zSlVGPf378v(4e7dd$jJ+x)PU(7*`Am&{zyk{bv9p0&UyTbsn_Cx|W((>^4QI@E@oH
zE2?VZEL<~XG}`;EHMoMRq?CvuV4TIb!`8%BC|zZ=U<_6naxFNedAXiZso-kl;4AO%
zJ9^i!zk8&8`yIqclQZsmrliPkzq@0)TzUNX0v|t-V<4wS7<coC($Z#jS@KCGB`LBQ
z=&Z(+gxk2B*YO595yyq~mgsxg0Hl~$mq6Xt?cY7@nHPt*3TGAV2J=1LR(^h+nU+M~
zHj)u+TmIpB#54`N!xil`Ye=BQh1>fL<uf>(62JCs`~z`)L)*SZV^Aei=82D|uYA3p
za81LPmmA-I|BlCZPjYb14R^bNwR-M<_)DI?e@~7Jx33>b>w<Bu6mwj|JdKRj&{>Po
zTkNNTe|nGFA6Y+r;{53==V=x#X4_$`rMiZ;-?6{D=kej5_57a8^RGvm<Xl)o;IQkk
z&a?MYaIzN0+d}>Y^sX&SC>RRY+sL{`wDv4foWRCOX4M*nD+*<WsIblx)(R3V1Ziq&
zCk;zL;t$$TO#YVTRqB_?m%<oJDT2fi%gPotZ~`RnhaJXho?p-W_Fw!p-g{nOUKz&)
zyM32*ZNp)I<PSf8q-{Opv@k7!R0K!XbS=v~3&Sz2%-1s|tr%69r;$=C?I29MWeI4l
zX&QM(*ObJ&stndRl+hTksUg$0qE&l+d8X9u&t<Gk+EvDRhxe9g8hQI>AeBtpb*x$8
z`<}=9f#1A&BF-{oatk=+7{?jiI2zZY?G`oChLquQU0Abh)w+}b&Fv~~+KS{4A0HlA
zSK(3iZHw~_a}=ko)zEhhcW>_LtUS~D-qW{F{HdwoKkHD(@Q7(W^Vi>F+dYcP&>hK_
zXYAb@Ldo>~o_9aI!&uGrd=bk-Tyd`9_Vq`)wxeoKxV?f>_;$xUP4WYFmSsF+o1M5s
zjiGyZi&C1qUFGYCf8sXX@Kw<^ax+)Tpl!ptjI8HZ+I~+AiDepT4WzJ$6BaGbRi#;%
zQE~$%QL7@xUz&3p=$DaF65clA^(=+fJLcPqwArl30BucFYsgU=qcIEYw&`}*rX@2|
z;v$Eh?m!7M>+OZ<dKPl|vQis`shUKQptJE*3IYt1)~y1lWpn9TbSeMdy;>oLgw{#`
z&&JYtX+U<vftn-r{CigS9bsKaMe-BTh)i7@d8jCD$x2|pv94Qzq`@>TX*}VaB`za%
zU9cNhYcs`B@{CfRv<eahZCTb|zYepFXJ{LYRm@?ftPzVv8HM)^k8j`6`j+G4LB4!)
z;&!`1tz6E5<C}XPpPty?-Qm7XRx6G3tn<uxdPUX9!+yuUZyDN_#yOT{BCQi~U4^4m
zq0F;PL1NsnJc^JaDz13zBsP#`RhvRYS;uhyB>XR~IO}lTp4t}FbYq@IK7RPXZB1z3
zkt@Vyg1mBFjtpHd-c=H}uV2{>d*1!-FVNm|K3`B(p$(8CB}R<*IA>{MqU0O3R;<!A
zzNhFwOaWIDrdEvIp`Anls1@I~KnJeDX~;1!UoH|5wxXiaRY{y6MvMmK<nqvWJ*^AO
z%S4z~mXgrh@51<oak?;#H~etGckQ==fOD3r^@hY*P})e4*fum>Pf3Zk^DGHUSkOrV
z@ALSIHj1Y8DC6md9i)ld<<}#00I`o~ndS?Ju423grMKTZ-_Sj|MoO+Iqp;RvjIhu$
z!Uc1-A<ltB#`gkM|3xiWYK5H1Rbc7sbtPUW!fhf>Gu|q8&hc>UFf|BHr)?<96{RY}
z1}pyZ;S0ykvuixl=|ze!F@UPy4x=e&>SmKrfVYO@(DM2D#Pjn7YUOww<o{DDA!NdO
zrEg>l;;m%2?)C#(6}o1JvyShd9w5$qK3%B=)>Qa(y0L^$WL5d@%_G_x{`()FdABn>
z-FGZYU|pp^qpHH{`i(*AoW+{Jc>P<*H){1<u3!1^@{xb{<qJ*MQk9_{_HyYOue7@Z
zN<r5-jyRgGquU>tZ%T?KIisyyG;>LmIMei<lpm@h=7@1E!~6Gi`vcWkrppV*`zNdu
zaIW?q?HiomvVqkS)Pek;Ayc>O2lm5`Vc224WYLy+CR`>8QlLQ#B_gImHI<MgTRw-C
z%MzHQFd)4#^xmP3Wf4sh$~kfc=N2hFc~8^z__n7S^KCMhV%qFYY|oXgSg4fXPL#oH
zsG+85C{-qSw{hWonR(pz)C#B9GdYRjXV^8kw&CrYH@v>SQfuY@;STFps?rptu--A=
zZfFw4SeEe$IigEqMYeBU+hVjlw@Zl9xYuCXhBj^O=$KgJN?tSj!$63MoaM5L0;3(y
zJDSc@tw!65@-6*-N1-w=vqYDwOd%WRxj*dL^(}LO%d&DEZ@haraDN!M%`4hitf|aO
zE_~w>h%w@<p&L3uuOUUWd0DvKCfc^=^>m`~j!HyFL#l;LrD}zB8fO%J>u9tkhM8$L
z+#l|E_x{bF^q%^&4s~Du@b{GaJLXV%{fGYv+jLlCXjNpsYW)75^YzTKOp@7AnlQ;e
z|MK#QxURIm!5GCdUhz%GbpA>R3;pgu*Y`5bh=EiJx66s$-6O^~kOEI{-m!*6QW{j@
z^7V^&*^H;{2TCoxeErPvcu&_2q;+JyUC>RB8y@7LsUmLoAV`WR!uzU>e6dtW%S~F0
zO7I44-{A`w--330`tg6HZwH(o(Av<Hz^Y*hGjW+P2)e?6FvK<%7$uU{h57o*>BB!V
zw1%)Q(#1zY)Typy6@;BvGqoy|Hk2x^NMpU+8cP;%a+U`uRa$r`h1Z&tf&_(|gnRf6
z<-^}$%1APgB&BGb6Nji$gj}&r%RJAl%R=u(d)9S5noPWWL8q0Ht>~aCl(^um0cV7d
z)!mD2BuD1)jH2N5uWuu^ha*EL)4vi{aafwl<$9uxD@GgkeM9rzTS7{tl({<|$T@R3
z92j;x4!b@3`v+{>vfd_Q$?Oh0w1#z>n66jm^NHu5f6wjZGu{3mB^}pMY0v^KWSnoP
zQfQpT?ssA@F%F|Fslsv_(KWF%4(|tA@5v!@yS}1R#dj^b5eRk7nYskxGP7PMv>#DT
zLr|c)o=?Fr7{%LuhpOU-RW<SQ`7`^wBc^QNXurpL*^VkD8Iy6ja{cs~JkA`R9;pht
zhdcCi#m13VS5#Oa7|JpO71y=czLUIbl!I!{iCVxEjVeX-a518Pc{nO9pG&O?%Ob1L
zMndt{d7^Werk6~Ru~g%s@@=x1#tGv*DM(Jxc0Gr#$9g9#60LZCxo{l=ELRxM)F>>;
zrg1#Jd&6$;h}W}d<Nh<bUt^p}X)76!v9?89A@cv?*Oe=xqF7;m`KASB1e^}?=hlYe
z1T0$>C@am_stwzdQ&vv%L<xa(UGcWknM_hEfAPEbym@<1>n$Z!mg{Gl{SyU^X<BqH
ztaYVt49B6tJ4b2TO@Sr)Hs!VWJ!>V#h|z{p0{!4Q90wX?>gn7)&S`Sdj7ju~-#;C2
zR%0q?lGqNygmYfL7rTa<E8{6}*tP7uk;`)}n;uPbz1-OC_vi}$=BG3NECilUXC96R
zuIp_Jf<w)6z)dM}c}c9}jB=m(e4Zr-s^p)&t}|=SxVB^HTMk2uX*ISn1f|JgCC)d-
z^9#%6M7-S?h8@1!VVomZ(S;S=5mJ=j=L%57yt3TbdCl+s<}b->q@;yBFVvdpAKuXI
zkK~jXuV-Gv^e0bF{^wqDFUu@zq)<_2-OLhF;G0vXZ5^c+uG2(aR%t<ML#lz7%ffA0
z$y(Dm%VFqg3gp{_@h$TrY*WRCn=<mquOMz$Z8k8!kP`$;B=2LT6e*=`I;T)o%4=o&
zK3GDe7U<iC!{LFdH1irMx!@bgG9LEweLRjghN0(l9f_&%{5o=Z*wZ^p?=?9I4^}HN
zxa1g6-r~F?l`3pcr6@V$+aBHbln`KBk%zEiokLqgh>2J;%4_z+0j(XcFDG2n;u_Bq
zBGVlC{P8E8cl3Qv2-1+AmX&2)xjXLIbv>8sm3dl-RSO?)*RwA2d_B+7*fmD}`(;80
zedDD3g~DmI<^Scs%Xe}aR<$s#iPLSORK;~#SmuS{%{@cc;;km<NYiW=#hhr{KP_VN
zXC3PN?m#G&Zufw048w5Xa(=~kJBGJ^!S(V=GLEvY?2dQ%;mGd$f5!6s6HZmyoycp)
zd1Tl>u-r}*Z8+S&A<h@7GWg+v_3{}|93S5@jVA`{u-$>QPH0y8rXx&eay7i1ZxrL{
zi>7g!-R{Vm3%hP7dM>9iO7Zgfk94~O!*BrWQKn^{z6h@;O_VB*Q{NpZIigBt8gE!#
zaLSO=itZb_-92nIeq$`8nN}4tnWZFIQH4n!`r6PA2U1uF(?X4bm!EzwyZEdKsbF#e
z715icPa)n~trWiLHlCkIp3xd%Hygnx02)n+i##xYX>V<8p|nmIXXzimBg`waZ>Txp
zcSlk#qCuLk%$E}w%i(Zj9WOL(%RI}2LGpqg9VT)NRO6&vf&nA>qVe{Mu?^07Y<B=-
zsp;3R!<zjaN)@Kp7j#Z21)(O)GLxvXDx^pzX6Pi#m*oTGeJd-2ifG=Hl9n}EbXv7A
zPx5=(9UtWIRwYNGhZc2whm9*NucY}3X`y5YRd{@+>1o>mg=A5(s-$^kT~^Hf_bBHm
z=dYN)L934D@POUjk!wY*6HOmC#@Z`NW$>O$Hq6RVl+gH#_Oxx!eg_#%UMI#2DBI$@
zjuIjzC%l#_O{f|2$o2HX^!&m+-7rJXaQBwhdg^)=<yt9(P<i?Km8S8$dH)W=f{CL5
zYIBi`jW&dJCafztCfxRhC^@5z!2`Z)v0abRk~u-?ZCzt9c1PMS9)1u6M4WG=97v0B
z9<?qsjgzo*t#~UMt*|bPr)RS6u+|c5M3GRwhrVYuJ;rq8ysebl|2*qgy)AWfTE#SA
z#YU4-kU;b=WfRv5z+8jUN+viPv{V7g+70@wzO9w3k_ISM3RdVs*X>AkqN<sF-;n1U
z-9yLszxyq_yMx3W%0fJUm0XeaNKaOjUc1)u{_&2+e&RA+QB4sseH5iqNeO2izG+aU
zFs97o<B`XAZ|MH*Kf-!r@bd6h&J%&Z{Qi-<T`Nci?-;K)VvY#%QrL?|X}&(ca=#zA
zf86u=>nn#Ppj74gG_s~b13bT;SW{tHXa4Tv3zW#)-OAuJ_kAExxr9V6P?XvV4p$x?
zpQyReb%v5N&KqfoqvQ}YP1Jq6an>qoqlL|vGgqrvudg^GVC}drRM)UOK42apEECt4
zXF?Q9jH(LnEKTF-`j)<NynO!1v_>iV;b^;_<NX8HD3(Kq*ASLp&*E%*%bGG@U$3-{
z=9mhv=bP|WwP7h0LL@Gc>m+(hwfzi2HN0N8_=A^nRnxWjs(8I!Ng1e_p|wn%r`U#C
zDq3r~XecGVR|U3HgjMoy&N-nu#Q5z%ms3I+%d#wJrRe*9!}n+oy8*4?7Bb6eV%N18
z73n%p2!VB-I2`tzE;IAA^89+Cvx=CNDBG5RwZdj4XK97kDl1jrW4lfo+{=t@8m#ql
z5sCs`UcyS%ihjFnhZyO)j@3C(hItNHFUXPGZQ^vk@p_Hiy?x95+k1>MtjogfdPAAa
z-S)dV9`|T9K$bFtF$JTfAac2lgsNC#U}(js>%61HM97((#R4==5$C1UnM&p5%U6E;
z?umxPc@DhX7Osofqgre6-Z8W-L*I#!Cajc{<dUuQ|Nl4AfAISA)~|7*?+%pZmHzMm
zWcJ59n)~nMq-=~74QfEUh8!a?o>(r=)W(DDxt_k#IE&K;N+y-U&^Y3v$zf)>y--S`
zW+=wcwmX*P2Fs1S&YaJ$<nswt3LzG-hQ^Dc$C#eexX=#=*6GALzM!lqM{o_<49vxH
z{qQ3>iu-W4+mpjeu@3E8K0kkC8870GYkJA!<(2dG%ER5r<J}YMGU59jHOS$j(~3aB
z8^b(a$jeCk=KD>%A`hD!7M3y#1_DH_rW7d#v=W+piYrz@jtkaWoOPJCLn$ZRy6u;0
ztfL^9j*=tBHDo2MJZ;1_P^#Q~?K)#!i!os9O2J@kOEn5@6-HOA)hx@*dihM-AGaw&
zAVhKJ+NQ&-8D$(rTZ*>K6<oEVHZpP;ttHYuycG%h>GWs4wf<wT-7uiyN(g~QE82b_
zQgP1GHJ-3a5E>&arHZ2M28=cmVm6T1mGO3AUMKMl7r15_$&(AM@u<Td>s!`!LAL|l
z-COkGQSc_8CZ^{JQlh$!wOHzs(DjOoLDs&$#VU>K_f)^fxE|#^fmak3qYZ7pqxgoD
zE5^4dH}aeBo_O>A4sR3{%gZhD$FC=_hPMxQ?79{eC-QhgD^1(&L~*x&LNz^V`pWtF
z6SsN9dnaquQrN$J!|>*bGzP}k8{_3?-o1GcW?u}rro$XNeBNWxQubLF$}hKF=gHa<
zQzjaZ-8Iw@C?J!XDhG#L0>x^yZ74;oENNK;<)Ou?>9zw+N&>UB2C~;UH%JjhL)$u>
zX|P=<ncNsD^TeB>qxMH)TDdagy~p=QN_QZ;mf%~+5gz~{LEgS^q-p)t?>BiVQdq@m
z<0RHmicAq|$&@TnhvHjN5mrS`l06b@f%!Jc5Kl)D@NZs4rfxi1*>49CtrT^g*!K;}
zb)*W;Yr5TlLJ^i3m2VhZiQ_AFxW^brTt)$kSD0>B8fz$PMwwMU6I9~5P;!D&a9z(h
z2_*Z?!;$0tJ-e>qcKOQJpMU1??j8H%z?-`r4-ZFjjtqyr6nCas9GY3|CXM&hEKb|w
z{gHq8r_cQU!<mO;hc1;8#XVb7W-XbnYnWr?({<uBC;sj8%wP@P2u`Icnf%=y58T}y
zI2`XO&Y+evUDuLRAxtYx;~926z8f&sGR-r77;uBf*p5TC__m?MRW98a=F?|hPB(_{
z|B7~ZkMjZ%c1<gw=px|lu5GYJTHHo!thd`=ci5&E79#~?9J}3t<}W*NzrLK)4?7y)
zv95uf3)38#R)HbUsgjgsO@?`0$yqT4i3&|?W=@%LSrI~x;x<kwBcSERONryI6Bb%b
zm3&LYR4590?ifSPVlS{-P&~iz>3&(iIw!+}5CRoJ<TPC;F_WBV4?VW&(7xf*%Z<i4
z*2{#gf#2=k5#mhSdR|}8tN|(ODKOgb{PN1d!|s0AxVGY9&m~Jfux;_fAWq(S#1vUm
zhH0jzC@${}t5#4F=h!F_x>U%8>+QxkuROhdhw}}`r#s%J!gb2nra?JN44VZhIELLo
zH|!Xv8>L7DB&CG+4sMFuyb@C(ZA#1-6RXuY?Y6Uz=5}4ER`WVXZlUnk)e=%>*dO?O
zI`e!w^K_8!zuS7|vIc6(w8uNxN*1~%3KdfeIcKidD>40PT*N=?P}f*5hpdt*VPRPp
zoK+M);*U>+ai-tx$m)*E`4gRW><;(n-GMk>@ckW5AuE@ZFvAhGu8gnGID_Qf<{PdZ
zsCgw#XL|1#4);u{vMeL_yB%v4;(5EfqxGI~x{-3G-$^s6Zfv;z_?8f60foAj%jpF_
z9Pz%Vte>g5z^d>@qpc?8g{R{kQ{z!e<E&v_D^1(+-Ts~u1JiV6x}Nc2K`BE>h28$Z
zW0ToqNTr~y)c9)!ik4vfvNB&@X}X@YB$RE4HB+LvZgUD~G?|KV9%C$Fnb6*EZoG`H
zh+ut#m1-l0DwPYSwW3y0(qu9kbjdPZD+S*)wBgKccFfDfaCks9J=5jF`FbN7!<*xi
z@U76;W`M9Tk5@i__#<u8^5)Gua!Q*m#8Ps>+JPEIO4Y;^SWlmbF_Gi1ulvXGg3}_f
zcGl3e9lk=XnV2)FWKag{W%8C%rftM=S*rruVyc(0v@vwPA+CY2u4pSO_Z*X~A+?e9
zsPi~oAWi~}bv<^tBZmp&9BtEY>x#&<%(z-nuAql|T))Ha?+9poh=4`wMh*!fP<@Xv
z9*iX}Gpr+x(<mIuYMzcgyLPMfJ2@;C*I+B;@j?xY99UzAcA#6&Ftj8Uw+E1{4UMrf
zg&iz@7l@_c`v#K?%V>$kpw`ULA8=X|OCrRGH4QX-xzQUZ*|K$&DRLwELbN3;GdWD?
zny9&Mpi@iHk&?u+A`@i=MpK;}ddZQpj94mu=uxhxtbrUVMJ>dFpc@2qg6kZZgf0nl
zcaQQNRXbL#DJjCbpu$Lqzr+}R{Y!5vL~S%|nYl`l%$#pXaiYXX4&#=2Z8rDwh%pA&
z>?G%_k#AR}$)ym(N=*yOw$yFPUP>n4Mwmw0<B{h6jyHexH<TC{pFhe}y=`ck9!(%E
zBg%JjSq@=4P$u4g_eA?|z7Wqd{y%JFmV&AkXRRm<*Fftu4||cRACCu4zyEt~U!OTV
zJu&PC+|UXvz6#8^Z9L;`qHQeu?hd0XO=B2uGk3?1^Xr9iDSY?#4Ms&&&RoY8XAChH
zhQ@PU6IvC1`?%w<R4(&M&4pA%Rn<6)-=;%OxBSK9zog$Oo<ILBA3uKN?(Tr!R&jTS
zp5Ok}4>ZGpbzK;ziL4v+;6)Xu7D@uA6ew8Nh3R_X_T%r#apCU%fHsoD@=XI3N{n2e
zKcHPpe|#Xvm69S_Ro3xJTqe?Tf_#KwPuuNiyN>(&<F9_dsjYyZwm8FW746kEWo~Q4
zIK{kVrg^0r&l029Bx2z*uf&ZKQA%L-eA9om-eR@lICM1LGtP;nEL?6gg(7N51scgr
z%0x~+d*6%_$l*?Dv7czAWo5gS!jx9hx;2_{jX2wKx-F0ve)ruy`~4p44AZobQej;)
z%Mv6i)Hu3fN59)k0dFfdk`lRO`k}`f&vYBfQHT?H5x}~!2CW5GmP;0uXsN_;CQxLR
zOyzRB5liKVzxoSYD_Tdb<g?vmI9*<uL!?vzpXV%Qsd=6`op1Es62m4NK;ADYE1KT$
z?%f-5$&B;Lyy?DbsT-<Da~@Z!F~q#ke7$0H!S@60u;(&O0)y6#ScV46I!>f@CFfpj
zJo8FQnfv{Ywr%O#meci5%W(c#--IErT&^z|;5J`y)`8h`nPv_<u<f4Z^qKv`cMNZS
zM=goCUYXXF;c!H`mgAcrsO!jh`oi`6iNpPOgz-$B&YZ7Tnr7hm=6jhq*&XwIBh@Q3
zJ-D8l3;W{}<LQMj&o9)PI1D`xZ@;5`{EqSY6S-REk})N*Z(CFn!T7#T=w$}$Dc&*9
z3yX620{L>LX*|_vn%x}=!+QS8=Rf`feb;gK^qqW(*TT?uDC-45q6ay&O*hy-@b<U=
zg7)bRDhIUfP}X2v%Xqu-kN?wuON}f2VNa8lJVdQ#tqHIq^DYVF<xr=zKsc!h+xFC)
zgwn4K+DNrrTSqD*&UFGY)LKw|RJo;ZUOlu28|iijnI^=AdAg#l=l<ah3F!7m;yghP
zXzfW<a1DNUN7r{0WiXAyv^~aprs=jxunT&F$!cw}#vxUbU-t^jOw;!q4hPuC{9Y?z
zuB8M<_(0A#TiGL_H(gRnk@b@4nhx!nZA~Ncbt7OQfh6aM_MZEvH)JaFctgjD-bo17
zwvN2tlcL3_A|6TWnS&#)hLkg=9dL&yRKLgc!&d1m)Hpy*T+Ux9&y@j<(uU>y%Jt=i
z_4JHy4E_BhrU#V4YfW!G$~vO6ByFg5PYx5eWyLHjZI~I3kF@u1>5lg>w4|Cz2r@&(
z2y1}O;~(zPMdAS##nAO!$BE^7#cE4Yl4)9|ndUH{njP7-SnDB+@t`Du?k?*}Q7h|w
zL+3zis+4r{g0`73Pvmu!vQXPoQYC~+4H=&*mP`q=tbR&GrAVBoZJnqvIRh2DP4{As
z)VN@g5O^wy+cdJQK^T)cV3qLD?C#e;U;(VD7%LCPs=;rJO@UJzid7`D3b0jc^fyJ9
z-@xh8lsDRn9*l~0qFajVN+~N<u{EI$F`TJuV!2(h!(J>XIg9_%Ih@lNU4{8+J+%~+
zZK0{q&R{R3oN1jU&KHVtl9N)9Q@~lt4zKe{$-*@3x=z@QUg8tW^~~-0nf38VdmOlp
zBj5k#4UKP@u2+u7BiSjMwqd=EK&DnHx1Fvl|NLM6!1?9Ed>fgN+~nQ9<2grkl!Nlq
zFyO7>p|4=?xZY;2OT;Kc(;6;I;13^9-0i>TfAd%WoGDIB&wr%v22#lM`%b1*O5>d2
z{^_2A=6bnun<8OZF-GDFZRaprqqN5F@95q|LR?TOl9!2MjKn2YB)?fQs#bEEnZj9c
zRw<#ZWN1+qtuuN2%9<|Jr9-0#w_jh*(N?}gV#z4oaEX!g9C#iBQ;4)pAtWUzDVG_S
zgf^NrWtNo5SrOJG@bD^_p#9LHRKZzzb9aDJ*!3&@syN?7;if&*?R#`7GEEk~n?h^7
zu@0pWR;e<<M{Ta*j6u`&lJVBo30WXk)+Nz*lB+GHavKBJ%T)m7F_BB*u;20UaL>F1
zF4vLf;IIfIHZQ9T3`$|WT*x`nwvMz!C`i##)lK6klT@lA#mMb;q4geTHOstkJRI<w
z&->HYpHNm)OQh7o*RMA+mZ$IDQ>)r=C*YcvHDr`Bq*7T{!9Fe1N=kw2?Ivu*ha+8M
zIiIgQziP&5mHc~&5}hf+;iDRUKArf@<2{#gVp&&~H3AJ`J+mKri4OQ-v$Mc{-?Q(%
zeBXv3&GjNG)<5Yx?9X`v`nKV4><J;#wnvQXQ0<=edgA%#zvsU5^li)S^#fhIqiRhm
ziZuqzq9B}E<{N!K2>-?nthY1PYK)cJT3lyNpMNj6rrjO7W~S3eY&(E51ZAjUrIf(X
zTHGK-ir44Q^vg^(enX}VtS=uJjpSHzSkPLtEQw5}aZY5-v86OUYnaih5OO4kulRPr
zXwCQE|CU-a-F{Ds38P@0XKIElb2VaI(ft8^_m+Lrqg@N%J^)f%|MG`_;KPp}@U3I5
z8rz9))NS%!>!j7WAxTlrZ(6Xp<wXP>iGt$f)<l(2x^|5`0L(A0NogwDrh}YB2^MB-
zzeml1;vWd(Nt~VY%(5)_{T(4?`p5Uo^Taq`DcTd2rs*15r?AdSb>3L!@hXQ#1*ur{
zvS6H}8nDJtbNsWv;QoJK=9gN%71Rs})m|%uvcg%?n<rOqKSz{6G;>ZY<HDQ*)><@%
zv;;JwtkTvJ*T{0a;r9oe)08!!Q^0CXQ&$?T(6+E>P_Dt5iYkyoq~<D!9&IVckTep4
zk4bKHRoHKN9m(^=*Xxy-6j>QAug~OFN;>N*)7rj(sR>;orZMQ;Qs#ismNd`!TySki
z-#Hp@(H=T2Osh~dIRqL7u{lucNVr{DuFsf$hwdG|6>Nlc7V8~1B?0YU+HKyq=(fXj
z1FqTOeIs=LrXx%v*UN?T=bxD_ujI7w{de!thdm*uZ-*|uHPVby;%rQEh(H@^&E%2<
z0aRc<UnNIXE3WHsz7^izvItKh7s?_B5sW6LD4Dcen7%%fQ|9h&kF$mpSHgb=Ege7(
zqx?>_K_M51S_>(pO@J;F0+kgxCbUUd-BMHowt)@UT5Cls0ihdfP)ct|7=hj@Af=Vt
z%Z2dzN*`KmKVaK|5+lB8Xm$sT&eS|n!a}MUr#;DPupW2!4t1JH`3s7QQIZ{1T7PS4
z<`_sZQ#KS$?;F1V;d|mT)9-tVR>XPY^5rXk&>G)1{KXIN$uaYK{vyU04OJ;Y%~(rL
zks&6YZzHX@gxd{W6KhyeO7Zq_54o}*dVUyMrdYNzOXg|s==VLhrQ_qvjq@e&aNIF;
zk!g{DeA7F8_l~=t?)dQOM;`BczJGfpR6|(fH#Vekj@|yq>Gl=n3~wJE=?3}UR>I%`
zsg_U^!~o**5IaaM2}oF{@vfz+Do*d3aZZcB)OQ%;DJ6)Rh(ZmMWDNf^usGL{F$!XZ
zFJGQ{xdyIt;&q8&3`HsCScxH1SIKA@Ynj)`nhIf?BpD-l!%_=vV+lcXI1C*21J}!i
zwzcefuvR9=8o_1FbH+O-9%>+O(Sfv;C9E-mR6$WUzjsa<H^>lSURKn(a@_4mOC+U0
zOv29l`1P4}i8yBnArP{{dB=XYm)Vzf%xi$+u+~tm!#0wA4%0$h<aaT>ipsNXJK^W$
zj51&k2bv8zv&<tqccjM1GR>qV5XKeMmGk9F)rRB4k;B6tP&hN7T#FxeXl+>6g`5hx
zNH)$n#nb%(Zxz-$tW!**A;iiO5-BYB;lQD5NU<lZkq{EzIIK(9nDEXMD_rK8hq&iD
z3N(DYT}eY`XdTDC<=8ejujPW@wd{8tkN10)WnsGBzL{a%e~Cj~(UyP_x%urCn=&j5
zRx5^0;$2H(|0D;YFTejgmhr~J;{%GyniIA}#`(nM^-6!d11fXLs2qqfVA><qIWDgs
zS;iY})1r)Fxt>{bC0DSQ7g7*ik-c3xzWog;XFmP@Z*i_+Sp&H$D3Qb69qT-iOOdrj
zPLvohR$;7XxxJ8cWu6!IcMp7dxzL(I*GelUmyAL2;m3bq|M0}?%U6``>HD5WE8Ol*
z3Kh=M_j~zw)C=vydn%e*64Uva4?q8b>k{a!BaZ=|FFd|`&wN`L`VOoiD@3Nfsf3*M
zKtWk6g6pk1ZM;BFOO`L{#`UNg#aFv!DN2=mfw2zL4nTpL@v1V#32LOVo>~lJhH$&l
zcP(+5m|vfHJ$>Q6(d@P1diuh$f24Lj^K>Oe@x7W(@=o3;3Wn(%aK;eA%**TP&v|S8
zM_-MzbZtukViF)`h>?`!0d9=JIVbCaZB3AqQ1o5nDd!98x(cPcczNg<jYYI?DF*rA
z%p(>F^y`u-&JxFjsgl_<R?{>+))-<~W!>#W#GZm!RB~DcYn>7qP-UguUYNi9Ot@aq
z%?{V@Y4-Op+@t1KxII(Cg6nspjR_NZoMq6LBBff4X~@$Axze^4TQh0Ak#1+g?Sk*_
z==VEJfpuDmZop^>2*>$MSXONC*uEjvh*1jfJ!ZpdC~Jt?pxYMh#5t&QrDnie+WoyS
z7wgK)r$6xW;V1f`g$=xHjNMj@5}(jUE*PaiHnLta%5Gq6c_^p~DG4gXS)t`uMCz?d
zA*GoZ=557TP}RtUN^87@eczH&!c@4tzRL8e`Sk`9fO$GY_E_st8fsH;$Vw8nNJXvU
zNv?&OS8-m~N-1p9x$Oe7VTV+$813*)kJ6JsaZ@F&6JcIJ6`HoE76|i1oMqJ=mzio3
zhDeQ(v|L0xHoSwX2{8)9^*S++*R6OW56&0@rO03B0&Z(#zkNCor&mT@IouuCzj@;G
z-=8@@zfhH7cYoxc{^>`S+nL50?v8u3)*Row5xv}4S;oTt{tl%SFP}eSD739$eSFP4
z>>55lUl|5ZwT8$2z}K&@<d9Glv@SSP85#{cnQ-P1F;qT3f8g@RpSl05zr<v<A(#~V
z{Xlm(aypHC{qPy*9JcFOV`aQfw93+$gjWk{%Y1_1o0Nj8Gumcyo@LdDMj1KWu>qhp
zk}<AIA|8_M)>KuXCTg9bNGVFRk-t92ImSv@CbDVx`E_Aj3+J`)IxbRJC^8uv$AwZQ
zY+pi=_(?8Oc5wgxPa9IfH!W(s;=QBa$=Y#P3;Q*3931DWSVE*IXd5Y;)Xi;;Qo`c=
zrSZC55=xfKm@*n`4d_ZNg}h|CM$AyDf_4HkKHVlRx5UGKAeSW5<aW>fn<p-(3vJsF
zmz7uy)QWX2C0B|~#F{owv?isD@eN%kv4e2C5Z8e3dl=+9?0PyQWx^`W*=gnU<wf#;
zZHER9#{*s8VZ6sS4UI2+{P>w^895y8iD9KwSeC-e^9wm=TJNz+)N&~%G%)m@uJfo`
z3G+-z1#b=B8sZvgY|EkTWP6m6p;pP1nsIGQoaEn=bLP<O*)=j*e7rv}_$n8#%@OX6
zW-KsGE2Sv*$Niu5p8B&6bukyN&tK?=J+A3UA&}ET$%Sq=ppD@+&CI926Iu8=bAPvI
zURFwkYz^W1jH^MY`_qk*E9*MZHlD8Qal;+1A8790^ZDQXcboD`CN)MY#?y($TVjmF
zb%hw2#w!48%y<_WhE~uEWMW*ImkDJYtJ2h*3G>K2FXULbzrQ2bg!LU?KK@KiE3s+O
zI)hf&#&f;gI9(?0Vy4EGd3|MWJIbaotJZQk&m13q<l*U&```RCif>u3uS_9x`TR#(
zZ|HYBE^`3e@zc+r8SbBm^Gem0{ccY~p_YJ(5tH>6zLqZ>^-F%tIZ{}0UJM0V0cw*V
z=R%4rs%B~GS&#BEP1V{!nlb7|36UHUm54Hyx4mN-Pn<to>AC}_*H?^poaR8O8igm$
zD__2R#=4faam>pVJ&p8_zhO?5@#{zCb!OKKNb)-U`bk{{LKL=KNP>0IT4SxO2DBEX
zO-f1D&RQnEF^KZ4Z5yo7vbGK@ZrI`a9_<>G((H#vn!C3w=V#_wCPXzyE@5S8J;Q#7
zX(Y#$izr?4W-kG8>y#8pA)uEVuItcMVIrtFbN=*4&L4l~beZ`6U%h8|{~hh&frnIx
zub)_rCM_poiiF#lI9&<Tjk)V-cSmB=p`Bw58I2=a$9#R}^5KufHDQ_^55M^xuIr&~
zsaR|&q#UW<aQFQiVyUEgWSwrLWtO(EbySp~0kp<7juZna%&a$qZNQ^2w#6uM6x$r=
zoZ{WvN4lZMJ4Y=UYaB6ULI`xN<jV^BTQhF^F>P(}j+A6A?>10&%7t}Z(Mg=2F|MSr
zkYb=1P205sHm_2U(OOoa%QQ3JZcqyCF#O4N|Nqph+da@VUi5vcP)jDo2~(S`2%*qQ
zQ&hnjvDKt-qiVlRlO?>30_!@NGR75yMq7i<3S~g6g0ph*?D~%8a3IIb?d2=lcy`A-
z!8?S7IA1WjP}Y%}KcI>xtCldX<Wh+#^RRDtxFt?$MHyMYZUYu+XQxPtE3c<3AO84(
zzx-z%-M(kH+w%#<5*8lw6}7<Y2-nvWk530u&iu_k|7*I#jyT@9yqxHVp5vQGwAQqZ
zV;Qgf^z%oim>Ha7SysHWe0Vu?=zE&haXjvrE*Fed#1I+U25Sv}{CJ`Bmc6&s9Qg6;
zng8p5`+w*EbpL<g)BFi<grS$KtN`D9_XcX^dOg$T%)DgQX=Zn5Azx70Z3o9K@LwQp
z8L?V2HHS^5Wu<&p3k5^T8m-Ot_h~{Hg=d;tOtZsVFUrTVg$0z}$OOOcHD}F~G+$mW
ze0-UhvtmhwF=TY8#GuGkCUUj@qV$s7a47|?<zT9mA;u`q*eHjczVGP!p120ypfJ|4
zESY6_VOiy{Ov#Y7T(-6YT&?0I&cCeuN)e(!BNIL6EUs;Y$>x5+`gk55dYo}A)5P_9
zV!vxyL!$ApKOX4%JFK<z`vFy?aU028mW8S^#E8*V@@>mZm`AkHv`+9}F^D_9=FH2-
z51d}UFmxSyb984i#6<dqWtR(98Lqd9k3W5+-yOJr_rSyV_o%AKF>`vp37$)73WgX1
z%e>%QPtIjyglqb)C)vmv6W8mFq3@}xvd#<EDBe6B@K&RgB^iybl@JRlLy48fJDjrI
z{0Pa@d52R~xS%PrObfm7C?%Vq(pa=o3_~x^CSd#%*ZqIq`qg$Qm073hTU)5@Izq~<
zA+w}_^_sSAP-SHz^W}77KeWWlS5&`e*bUsSQWQ8meaCuvMPDJM#CW}tZ!0A)s9dSF
z5at`pbfd9y8&PE?hJ@>PRBf2XiFIE1%@5zxri7{z+^!{4bxYV_LZw7%$i!8$ZubuZ
zySpdMaA3VYQ*vcqR~(Mx@qls`?^~wp6;$SM?6Jd<xUM)W!AS4+yewb2zk8zXSFUx(
zkDpIu^9K$oGM-=gxBt_>#^^$$!2>lGmNhX=E7QDEGPD?~v1FC;wnZb^qIHe9#zR`E
zCF6Xzt#KkME@(VxS>q_Rt!}G?qiZRcEMwbJD$Y03;!}$8^)uITmQ@W^@ECqTqxkyu
z1G!XM1GX2f+IYD!fBwvU?=fVSaiO_?!}$3lT76^*iSz3#$yiL=((Uh9%lXfFYyC%G
zmvN?6QMKhw#pb<Cfj~mMzc7`wF`Mc}@)c<-k~kn!MPT1{c;DhS-)%GO8JdQ8I51wX
z#I&;9&ZHDss>W7Jqa4~w0U?(S_*{g(uC>K_jZL7&Gr49`O<)zRwG6#SefOU3e&ul3
zVZuT^eIhP1;c}+rM9!6PyAt9;(H4DpOWNHL+5<^xjLLXXLz1cFbz#O)w8q51>B|Sa
ztL&eiFsdVjGt)eQR`~WngQj&Bl@rl>a!wdy$y@0Lq>WZmAg?R*1>3Y}B6Uubv~v0K
zfiFM(9uos^zW-h#0JSoXlknO~!5WSF<xp4U#@YC0`z*?rqXg6i|6#Ozf!3VptfHhS
znVS>=CFFPKaK1(C4Mi*^#?tH@7$Xp4s~HYQminhP!~Rn*>pZUAqm-f6xLuHfsMBf{
zg`e#qRmsj7!*sh*Rg{Rx#srgQtW4H(43wI1YKPKxn@m<|t2@Kv`*#$riR(<Qg~#81
zFO!V%gxL;f%jJv?6J?oDIM}_Bf=b)+_J<$%UuwrWe&EYb7qoU;naPpkLfK^YwPwa8
z^2ZO)eE-|;vF(mE1e9<1`uxh{&hT{X*tIRsV`g21=QZChr0GmtChnfzu)jMpolk<C
za+Wn`mfEpDdX5c*+l3{EjqTN=Dx5AiuIHKE;Cb+#X`H#vnXYwE5>4Ci{X<9B7+ya8
z9ske&$Bm*C`(fbyZ@&YLJk*m+&aan=)9Z{fp2j)u4jo39?Lem~WwU3X#ipV(N?WXx
zGJsZ!9EIv{v=t9=EwW-%$ds-Mj8{CJdCT&~O{<0Y!hhxKk1r$Ryz+dRdA+TyMT!sN
z5Xz7vRniC>NkPHp75?SV`HKOf6e*v4d3mL4EO&Ru&9SU$yy4il9C}ZQGR>qy3}tI{
z8@v<bNu^2@DoNQysVclQE4nx5WT>*P3s!@3iv6zP{;owS$Lahmh861waiQzl4bE-2
zoX_l!2VzXvsyUsnJaz-eUr;U6L@f~&BNou5GCsdTt!N_v@Ejs#6~}jsQAinzZ8K#F
z^CF*D;Cx${*Ual>;pb0hzF!wyr)hT`yW^gh(@i2d))EVJUBkou5&cUPN>yTxG;Oo3
zlwV{EQWt8@IKX>LqcmX&R4muHFf9vOJ0VsW!);#itz_Qsj(Z&P-7$xid0BZrU(i3i
z=YHo|r<rJFN<FQSwrR1}Fin5jy8q8Q)Xgs+>09|=jHjzSKwU%Wj%1@ST625(2w41K
z!0sQZG4kch7tXhpfAO!rr@edNHm<~Z!K@=DT<E;!=dY)&R=o1v55Hr&jTmPbx|Z>B
zVVr^hPo2d(!CBZ+c{m)n*_qd`U+5c8>n)VZErLHDn4UjkY~}KNrrq!9hdtVgex_E#
za(N{#3x&%5{*lf*jL|Houe9AjfB!%VD=`<+{w?kMM|`@W!wqE|-Qk|@?hSFg@>o_<
z>dEVs^;B`b<zN2I_e3+W|NgIe`|&HPZSm`g=YRTJvNtSiW$>P^?`f^V)j*9iIR=(#
z!mKNnA`bws$n%J)l^g=AUQtz;c1kN0`5SE1P~-z`J)w!1;IJC3_E@XfJ-laGC-M?O
zH@I$KzTL3Ku{#Wum{=|^DA&<9mc|@utR~JkjPJ>{5?#w5KfmI=!|KYrw{O5TEaSwq
z{QA|z6f%c{6>w;Xq>@q9V7;frL`g~3sg0-couDR?{5<Ayk}tXK;LtP;C1uv@mC{DZ
zd!|6w4=CGmcz7a)MVer3i%}Zu8_;IE6<5&Gj?z@g*OWwK4XSHNF_WeXxdy6HsJ6p)
zf=}?k?erCLLFvjG7s<8g4FuSB*lxgdJ50AnwFBL7L{-6Eq<W#Ig`5Hzjo#l$6MCJ=
zCG+{ik6cf$wEaP(p4zbM4)}sVaka*JVXnm((16`0z|@MW83U}-b-O$`#`(s4y)r(3
z;^oUn8s~X@`;P8-k5;fuBg-<OoyA!Z!B>j>vy7}%op%_e$uSVaO0K_ngOOYe&^Ikc
z8$w-Jf`p`VsVKyr5<(!ZE5=B+Qab@cTj%j@OHP4hnYgXLo*4$PEF;bvtnFkqw{iKr
z5&H_tfN4Q5lp2}Wfbx67d_~0|uIMV$gkQLaX&q5YxRA!Q+w`(RS!suk?(Rrf0?RaV
zcz9rUcO>4<gxd{Ol4!+J5d+OFz*1<wYbj>HcOACvxPQ2#v!BWHM79maH7G4yy%?gX
z`?O+NBVS&w#AW9B`NX?-k61EOi2V36^8Hib-EqfdjQsrR%s2;p9_gIs{>=k9Mn3=e
z2kzdyC8L<9nScD_EC2Gp{@?OX|K@+h>BQZBhyHrPodmBE(!$VsO3K6(2&v+&!>THo
zFr^u93#o$Dir+llVYM8@-aho~_5&}k7pCb#+xNJ}a2+T9>Bke_{pNdq_g9bfyPmdh
zWRglHQ9ygi-<C=)NjRWdIEr!mE~!OPQb1ezoy4?aZHqM>)oD_Nw62t2CRn8krlwXx
z{Po{)U&qMpwh%ImYb2H}Z&}4o6Jx|!xgZF$S`MDILfM#!zmyu9rbQdehc93F=YRP<
z#u`El-{KKk!LIf6jU$%Cv@FyJ#bmU#G`<mTYb~NZL=KA9$uS}4gr>4AGbv}9#&Ulc
zc>A>H@z~Q1EibR5pi!)4Nr6hC+wC#N$-!FHjUVecUq{fETn+Rsu5&cKC9D%mjnr6}
zr-ihv+j}_U+D2kW8<0P&0aB4mlF>A6%Ty~VM6S1q%Vi`NC5J(!`TF(5{jR0|>mS%Z
z4&2}EF(|sGrL%@Li4N0R0si(~`|W4%oZ)nN;KQe9=1De`q=0uCZyjqWEbGF!Rxaa$
zG6kg_OD&`+&?JEyKkQoet>@u5Fm#?TB{NO4Ok=a7q{MU^(aw@n=9Wg@zk4Uei$CG@
zXC3O+xH9yPm<unjCkd(^9A>#Ov<>qbX&&CwbUWhhOn>*t8X~^m@!S5M5+>H$8P<tz
zKk)kL4|3yh9ev+XYGUXNtT%Lxw0_6+Hj+Z5@eboVuJcN~3aNZeg)~o)GwXDtbw{#p
zsa2Xkt*Q9CgVe>lN9NbB7$@b#cGwXk9PjS9KCjr}j{U<E^YsL_C(enetnis)V4WuB
z)8Ep(`3u+&G{ZfbOlvBQw;bO5OO`LcN5_eB+%w%SDDSYlzhHRyYgBvSH&1_qseycZ
z;jr6N&L2tZOk5XixKgJ?NjK)(Ssq^9k-PW5!Ia474?l7K_{`<YC)Vk@J=hGkM68ju
zPt!QO(G2??D7E<xBWjtjO^eeSrz_4L$zj4O#nDI1e8Si_qFZY_#@82WDfIgtw>8tg
zc~5@+iFKZEofsEtiYzH}JRT(6KRn>9B`R=EbHDHZoVV70^c6#72@xs+Gd5f9#W``<
z)*_U1zfrfhoS1NZF2r@AY5n#A4dgkIW5VY|GxUUUAx{&w-=VFc=7gz9>=;O9F>mfy
ztz{CZjH6;n8B|_Tm@RKrw!@z$RYTNrGet+NRiqTTO($7rY>l#cS@3O#JG{lV!zKt<
zDBA$i-qXG%H!C%ccvVo|k#+}`5GYFF8jI18N@XnxlQedln8g&(s-jG#AP)wk<wCQD
zRs63AaI|DG5rk>tdcDeOsTNX*^v4If-JaTZn=L^=yIvctRyf}X&nmB|swm2#Y9g<*
zOukeFRmeGPH6Vl8ZmQmjzE5vK|F&;QwGxwHK;~)WcDu@2OhH>jh#i(iY#`%RysY-u
z%hdr=T!~?(ltPZH_)pgb=SH-380TS`B~wR5=}5ObQg0W=(+h3uMGo)ecCBl|cq_sD
zG$ZvY$tsl^xw}6Cm7D`@V`;jcx=!TV2~{P#Rpx~_T~Q^H*OfeeWsVj5<_C=3%je%}
z))Z*0l{TkVlC7*2W5vmBwEV0;Jzu!Io@l#{hj&k6>$$(@$BzrwIq+L5gDrf%X8!Qu
zOqgfBzxO;}Ci36@fxr5z-}3P8J%9YWf8cz%@#fuo-hcOo^G`ju^NH`?Ju>X?`G=oR
z0#Ub?{lh)C-~WuSg>_s>UT;%%Q90F87}tfa@hB?KuUGcRf&H$58hLp4hM#}>%-81=
z)@k}Jt}`xj;g!auKwdM55)vpil5#+?WgUy~EUB8R6h$R+O{f%P3sN&V2l?5dHoHN>
z_#W+ckZ<I?2z&5Pe8~RKy*{5V9CkYln)3X*O$sfl$S_)|g8F7NsoxZJicML!P3X)9
z?f&|FA}uprE57JbMYrjij+z(t!@zMru;v?DwT#okx-OVxiC$tU#v1W`Yk^WPDPT(7
zChv+61KxS=?~c5C_sHFGN3DTnGOS}DM@>Q8)elbt-g|~&$K`ZGD@92e?L5v%yWcpE
ztqNZ%r!UXMbw+K*qv5cpv>m6@i5x2-N4&AJ4o4yzN=p<XrpS1`vd$BYf&Tu$u4(vs
zx}s8%?@MEto=3jgk7B;Cu+A&Dair;6y8Vvb&SSJ?p5*YTY7r_$jePfXM3=%e1?G9>
ze9456h@qfuL)D7wGBRh$_gSY|N?{I(Hi~6JVE8aN#jY_NfA>Ahyy9vW?yGaOjimtZ
z-#;)-3+pmV_Wn<J{aJ@Pqv7@RitTo|Za`T_cAnw*K!}kLC%pB{=dZZ7<NV=gveJ}g
zB0Xr*awA?Zbo&E`yCd^Waryj_>>_cM^1}Ww02<o~Jss;PDAsx8(C@HT$x2cQ1X1Ti
zZ!D_YV+^?AfKf_rrX_7tgUZf;Iu5x0DDw9>aro^I#PNjfcIa+TSZ2oC6}@wklRLf8
z`bOA4?ZD%GPrGX+uv`Oaov0c*ueg19#Tvt+8@i^$7di3gT#2_=)Vk2LEo#0{Yr*$>
zT)L9>d+IVu%gF4AOM}vm;drE?aT;1bupAB)zo%4>k2C+D|M!1K3W2-M(K$uaT8{Tt
z6e()KD8=ozVEi_rvlgo=!_cE!Ppl2s^DDO$s9~YC2j<Hs)@7z`2khNj&bJFiYt;3b
z92eTAqiD_Y`VnOeZ{PohwjWqtKN3z~IWIG7D!h67j?(V_{BOAb-}f7A<%C4u=;k>l
ze378K)e`89AyAF9qEeQejIjpqcA%=XoxP*$de$XMi`R;-spd+WFY?f{?%(GamDvto
z0QK8yS8M_viYQ^Mv)Bq=OQvPI-G~WV*WtSss$*FedIOZSHIoas@g`h7*U}CLjF%f>
z3=wJ|hk#Lt8+92WE*P!EA-m-Uo86B6VUV_M%G9dSMON@JL~2Z!I+Bu1xwYM}Gh230
z@<fo5<R6=1hYFFr<!icrM?JO}Ese(z0{>sy{-oE|?9R``p4GnXZccO0y*F+|Fd~vk
zDSLHQE~x}qC>i#sfEs(^ui&rZkw1WkHM9q1z=mx=vRtK<rOZ?|OftxbVBGdJyM4RW
zda!ncJSa#cg4zH90wlp3=kD{qYyJPf-}89CP*-KG=f{aG6u1P<#W9M)Sb<F*k!Oeo
zWeh??5Q>;Y!4*YETIP}g6$(N`=RhSjDnm>jKhG?~z;qhEFCDZ?hg61XoEY{8gh*uy
z8-Gjd*Q!OTwp;?dBW0+X;yTSj5kkQGu{=XAu;KIW5$_zsFpy%Psv;&PAmROt)e58`
z1<!OkFiq#u5Koy#tV!O1F9+~;)sdYeOiMWgrvOqQv?fgpekr-4VSIxLioCu<E5+ke
zK}UsoEHA6gw)m|n1tM$81jhToI0ydb=il)6{_uC0#(}D?dHL!E`{RMPyNTcbEc3<7
zo**WEcJKM-DRO_<Ba-Kje*6++B(Hz@CA+5wa@M^3(G|oAqv3F#*xXz*&5rXN`SHhB
z<gl>5x#IdX^ZJ|DsGO*gG?nFg)36^VR2EdGWpbX`1**!Hi<VHF_6MpibNB3;Uw!+U
zZ@;}SSzp~^fW<k;jvO;tl?JjX_I5(>WojyAd0zzw$bvMwxU<W;x~yMAbO@=DrnFuS
zGg)YqS&?%>%DJ@0r9jFb&PH@DavqBnMVpG~JduLLAqkSs<d6_te8iL{GKkXf1)%s_
zQxxp?3te4Wl*@FYu3K^zC?QL{$d+|=S8KeR%k+;#O5x%VFTmXtBUNQNc?VHottQ7n
z7SOh3ifyc<Y7NKp8JRMhjV%!c??FVQirijr==v4oGIJWwG{&M^g-{A@6v@N={=}-z
zBo;EEK%qrMiVJg877uvUHCS5}y=O|_<Fu;wf~<0-;Ii7>poC<<JCc*8wW>^P3FMR*
zLm)lvv9)4#bAxSrW<R5hf^czpi_GC%kYCO@5`yzMl9OO%D_T>-rhwDmzk6Wu5hD{q
zg3qa3O0&d^@O^Yg<V3F}T7bB4Rbv2|9i1)f=ad7pPi!_FLMEa+v+8=BbKgIw|C7(K
z-{cE8%$}DYe?k_erPZvq45t%M@86J>Wqtb*-Z{>vk#%dRj9{1+n!8sh)ncV2B}dgX
zh?H2ZH;@I>bS9aaFq}D$3st)zE*_mSI%hUbjnj(Bd6a3;s_0F8@F;DmuAXD6HRCXG
zoCd0NM#i~hpK>P0NZWN>zx*7zeui`-Zuh3hu_@V~{q8;6+vjvwH~49!QDATGm`-O*
zzrtShU2!}y(O@n(2!Y_~IH4pQpWc!atX4f$(~*Tl>4j?kfQSuDfgC)>HY7K*ZY#X5
zIm_a6RmoFb-_hJWM<)*?Tv7y~xxIN#j-IO@Jwr;xtE&yeTqX*`@x(NYm{rUDy9X*!
zV^kuJ1HbzASHu{&?lhaO;_j0#uxrQl-8D_j=*lqNzv1!216NloT5DOBg+`RTOB9NJ
zvt_PJBXB++L0fWD6mh%5fvc-6+h?CLzW)aQ;SKM<{T08BWmwfUNNbq<P{RL!$PwlV
zrDXZ#@t%2JFj^HSrm3)%VSRN)rT{jpwx(aNky7Kvf#3_^QVLO4@g+r4MD!V!l*;A7
zmdV-TBR*!ND-SDGYh2325HFx-QT%^Cv$*2=_480lFEOEILKsbsg<!vgqW98D65~?f
zrc1$c%)_xf%u*@6_+>)ogvx@t>x%wN8~R=sa{~&AqOuh^OTs+hrZe7Iq7UTrzA)8{
zEr<-O$x@<~yc}8;QfD**)vjqTQzUnhyGtN@N17)@Tu4L2jS}rWJ}m^Lk(b=tephbO
zS65r4oFR@RuQA4A+nUO1LKKKBnCFRM|A<HtYYW#Zr^w_MLb^N@&m+_RQ~=uAF!_w6
zX1l#3)s8V*WORf$GyR%*lmFDmjVErdUT{1<ULe($LUK^pszsT6ncmuRc=nzc%7k`y
zg*Pd*A$m_pf-s*58tSeqnz?CVI-DQ}Bm%S`+raSfNVn=q=OZ~Lq*nOZfy}sZBzyNg
za$+7PvT4A1a0|<ln5P0y)mm_#&J6n<-FjP!9w|XANERU_PsforPiH=P+Ht$xvc9_J
zGzFeLzvbzEhfnZq+wyr<JWV5~{mA>>#HN<~^yhDodBKH<4}n!xvAtfQ<B9X}%yC?p
z7soJ+ECFoYa&vuGHX|P;mSN)QY0r0!(@JaZu2;<G;&N{5>aqdo38GA(kNXq*Z{BnD
z{0?GbnFDVh&fMN8p1r6LB!VwmLvqP@3Q2i4C__}`aF>E7h8d}=qF=jMc=V;vlTz7H
z<_jjO)|QlZCAdC55_2gI2-709Af_L_e*NG7^S{G~<IGnNBVWl4Z}u}j&P*;6ghs@S
z6uGD)mHgiHAcgo35j(_1b4c;1H#NgNQQ5*YJWUhMIbsGM99F@m?}#ZePRBC&GrAo5
zgTu}0;(EQ9QAD|9S!1Z1hD65N;&y%b@JQcSp5NY}lxDYk<obHcG@tn7lNWSd$2bf;
z?slY@=&x@O+R&}m%swJb#pEWQ_9JGcs8_4vavk%<ZC)n2ecylYe$BB=Dx0oFNrAQs
ztrcNj2(H*fB$>9+)JCycH8>wR9M2>qS{sIG<njFjF~aJK3nHtm>)kYgPcTm${fWhS
zq&CdH_}Hta!CuzM?KGlo&1oKxiK3reBJ&cFMqM)P<w8ml)Rm!YYjpMm?{UkFxlG_T
zn>8{g-oJf=&VsJD7%T8u74~ZU^;C$z(V;H*(jr7^&@HN7W0NG=f?YU2yyYZC0dVei
zW!+~q!kwAZcq#g8e1zS@6Q`&5D5Y69hP#_9Mi~nOW;_BCw`8imVRQGK@pM3CiSv-d
zf=H1#4)~lQdYs+j%$nR@k;es<99eQ{u$I80(UMjsmb4(6hH)B^rY1?jYI}pSH8MC<
z+Y+=!+loqLOxF`bL~6tF=@B6olF)Q*ho8qwBS`V;C;tE|W;WLw9^Smi`-r+6K%*ZZ
zMU-hsu~4yH%J|cvWIyVzgrb!}j0dFAsH#Ehsx;i3ATBd?x5Y0Lo0~1q>W`QfN17&7
zRne|m{5bLLH?O(+=ri(k;FB*#T!!Q)IwLVkv3QSL9Ouc?<}%&9=`u+gY^~9iXE_}a
zO5p{haiZT|<3nOvKCsLS=cjLJYR&#UL5i4WLrj^~X2WrqX_{ZJe*GW&%)`j2B+F7{
z&i#5#-8K+%VUq=iUtC$w+KR4SQTJ<<ED!eN7L<`#Em0(rFOv@MOFOEdCS<{G=z{Z5
zS|KTj3O5G=0;B<Dk$J)|6N5VzX?a*sv8=z-A{eKM7#-SJ+&q`&R4(i&Kh20(GFHQI
zW|>D!U8CBL-~&P=lABqEBi_yB|KGds51%S$q}51c$rqof$bu9LpjuwY^d%}Th_c3O
zL1u(F7VZ|swa6tS0w9e>D@|1XGNmkyIPaFSUP=iO0*#U&H2OPu!X+H%lE9dY^~5iP
zVL)^S+gohi7awuXEY2f@$NN|uq{kiO`E+4ljnuZR-HQ8DlZ0d%28Pp-tPTBU%lht)
ztCyd$y1HXt0;Xy4!wG-hb3Xlc4s`(YbRaF|pdH2$=M#0e#h4>?U7W8e6kT5o6EP%|
zs_9qPEYpnlg5hwa*=(_<VfXL`bDoQl0}w(JeIy0JkS2myaK{t1l*Fz@X^A`T2-5(R
z?|AeV#@{8^H4g)%d!)}rUpJ2ztV(I@W;dbIiX_TQY*`kfh^SP)2M<qYDq}HKgYDP!
zx3Jzlv5XVnK8}3!@eQk0#pf?FNlP9?5y{{09Opm(`eHT^Jl`}_Rub;6na-yYn3Rh1
zX`pQ_Ra0~Q(I>>5xK0_P1cwjrNulV&AW>H}_iyh})`HPI?FZV%@~ekEZyxqsbq0|#
zfBLWghOY0p-=A5QvabK=#Rj9{eB76;sI3uDG+!}L4v!KpTsm{X_RPz%9PY5VxUft|
zVwh2;EyWb}h@8rEKLyCk1*dLM+TxahW%~69|NpViKl_u9v05=jxId2E@6Y`EZ%+Jj
zm-*|jUb8<9OevAdxf3BS2e?bF?mLIUcdO$R6G}FOlt@Bx97Z0VcH|6Jn*!nQTdemu
zH?ubi7bG#47J7=|`!Yljg5a0ZJXRJd;CS4Fl=N*&^b0xE)b@%DynFke)v71>iM!i1
zFP?3gP9xJYqlBbtYGO{@USAVZnXvn1<}ge&qT+VFLZ}i4QPv=XAUR)jbMwr!EZ;Nr
zVsK2;z<S+NwRPD<Ny(f8q=;KQ<9VdA<#4WYroG;<cuxSXuAeh4o?)DL7{2A|=9<&}
zOjTE;oN>Wpo0`qlnk*%!)4&n};$m%@-BS3Eu^h5}2;kMF{jX5kmY7fokTP0Hv?|-q
z+9*~=VJ>)>sxpu-S@;V~f44ti%oeQ-5m$Wfzb5<jH#*ey*F_^Wr%a9u?|$*O)J;R^
zSFE0Y#Psl*Dg};nBpZY58>|Qjsi?JtdE_(?2p>>QL$|$Rah~)3iA`g;{p1S{rxWAu
z1Ffxa6b;GY+b@~0MB89;L`08VZwpjXl<=tad)(nY&1yrO&qS@Tb$4mZM1;1CySFTd
zcO+4=F;XdlTga)jY;zXarbSAPN{QpsJ*&-@@!`I}2AduQY@3o*3C?jk9a;4&`t=6<
zz~S*NaVZEDQpqy8c!A!U!rV)jF9)W=`$Sv@f}4wT6Ghcum5}y0pp-9@!nP*JjPxVH
zR0Iskv^3f>rUcp$og-1;i80LRW<?Vtz0u^@plwaFxgiY?<uDr~tG*@70&Od_g5@;g
zq{ggo@Z*VRKl&Y-x*`nk*}Z;U1ki2A`q{^vAKr3v+q2spAbm}o&)lqQ+B!2W({J=}
z`=9%`apd82M5>|+&r8TA005mPNkl<Z(`0h+Bp(XKL}|SA7gd!kx5}Ilxnx1IpBN{{
zc-T=_H7O?2h4k;eL#s0F5<=i*5w>e%P+E}^%!?-?sk(JhhG|0#3w{`I%Zy9`7Zi?6
zh~=QHY)#X3rCF*p!3R<nmr!jW%q6!LqsKH2C_@4ysSwrU<wc?t%2X&(K(a<=Y+W8+
zIbZZW#-JqSpFvVqN<ou^#S2TUEVA7qoWL|qfkWp^7OFIDl`Io7GB`?xEIM*@glQs0
zKx*;@8Do?t#Io8>RfFnoX@o+i1m5%N#VzaR6;0QZbD4&w%WB9k1yho9BF0eoW41y`
zjhhyRaUum@TC&DqnwF|-X}m{BP2DuCH*40b4r?S*S*jP$nTL+!z2Nv;a18_KdW|*}
zq(I%Y1n)UMeM1ThQq_c}rS7(9+mMAMrIIOBN+YyD2wPTzrX^>IkOrZz7@fqO2O2A|
zro(yA%0dz><3Lv%nstjZl6k)??da?no*r0ck21wUDV4;#fb*G@M{K*WxQt3=Wv7Hd
zD^2!+<UOjYh)N^0E<kXR%b__--n_eKeRYjAmU`8pl;GvvHM?<Oc7nQYxLak~x?;G}
zjHeUdzCZBo<Br)EqW@QKo*+k_efz}AmshxH;&A_-w{PDugaV#bQs8~$>FLC(Z)iS#
zi5o_ArAiUSd)lr-*ELaS!iPOW47l0x?|=F&cbkgZD4q@z-#(nEtYx$A`0S$%zw^Z_
zhOlrvK2q5lA2Z(1R9VAxKxRtOgffNGwj)MgR+H@p?-!QoKnM;fCV&_tj0cJus<>dk
zV~Z(ZJ|D`ZF8{~BOTW%1w<}CzsI6x6{E8n7_%HrTSnQggefh}$`(OSI|LSjE^Jbix
z-Benj1s!|=O^eHAv<T#L46NFQd0KFN!|8P7{{D%6y<)vy6~DDi<zP`+ZZ~V1wqtiX
z<6OqMrTnaNv6sXY5T@`smw7@6iN3HwwK2p9!!WTw?n`DDfmX2TTgKB7x0Hi)${8O#
zRZ}xB#qczoPvtqTHK)mwWd*VnK5~Y6c7$Onu2-OGn*vGB&^8rH<zlUx=Tc-@99GD3
zhF69Ch=5_9n1_jV*Hc%9X4BJdHzn2;GV>fc><>gAIGiVzc_B$bw_Ou6bn6w*ZmwAO
zD;}Qqgy0EFWOq2RKMhn`^89AY!+GZ6IO1X|dCgpKGfHy#`_jmq$iX3$Bnd^hU}5~?
z%f*9C)zsX-KT;hB1TA$f$Vm|0e}<w2(lCr?6h#8RxxQxe?2dVwIDYt+>}HIRT(>2(
zTFHot9%*{gJeHP-Ev>i|XDTb%Y;IV(8sD@m*|5wDWXX~Q(=_6(LD`Pe<Y{7Hwdq*A
zXF8pVpHG4lg4)z{ogj;{%N~Zg2#W`YY}e#iX!J5igenatQ?*n^kh7(~?&)5BfuBZ}
zr?*T~V10FsQ5t*mjOn-|WT@({RLIGsm|68JvKxq_Lh1_H^o(;PM32@QBQuj*zy)es
z5n^fTB=6A1P}d#StVv3<oR6e=B#{fwB20*;MWj>|WlE8$$<86QW?5#egsxwMGC;=d
zAK6}QXu6u11p4+lMhN^e5<<ec5v5x^&|N>nXu&)V_;Cahw!fxM<z^(bX1agPj35qs
zy6rVGdW6VKrw^>x6-!h^B{?5X^nJsAcci+yMp^Nn_%i*cKV8?9>b(@$s_>`cGEwOH
zWm1%L(T2(3@w-Q+;Y3|k7a(yWCYZ;Wad1VW*3@OKYIHF`sM2;0ArPV`T!igHN`fzw
zrRZmjtx49Pq@dOiCP#1!lP?*UC<Fq95{i%_6^1yM@p{gope<6#Q6#A_b&c&-g`pRY
zB=4EWp}0CPFK{IkTAQMDG75zxF6Dt>stQ|K{4^ECjh1Aw5SEh1^Jzpipo~V_HC0`g
zl`Dn!rRoMLAO}zU{}V;~5D_t9ltzk7K#+|h#Nt6EC1ixUe@3-^2FWwO|Ay!WnoUQ)
z=}=Y6vN)6|IEfqsN?OnhGI{X6z@){5R2nM{qxZyI4zS8rn69O1JLt<~Rw@a}6An*E
zH>0&hW`R$E%1EmAhZl1IL@0n$EhXdmjMA2?yU#f6k0dv7+7F!F1I=d5rf)Gypp~Sq
zYjRB3swbn+vO*+BbOEhvf@snGOE3meRWv!Vl8H_VINo#q=I7vM=7VQ=dSu)kaPx`N
z{)qRmeWnWnVVsGvToR)2m-v8W-PdRdbye2ulXK+Y(3h=))I~ceFB!Xs)5wSYz)u9s
zZe~0lscXx#mpA;yzyDhvc4t0*al_Si#gA@p$YJLG^&2`1FE*O{#}iW$ynQ_I<*y#u
z??;B=jIC<^_HVxA-NV2}OLh-Wgyl?3aR2tm&>J>=X~?hi3ft7E%A(H`A!Ob^JY0a*
ziMyK~b(k6Fz<xjQbX-Wna@*DX{>L{w+cY#yP1oL#3E#KWjcFiG$Q-a$i}PnlGumtr
zvRn+?euKOS*d;w7c#N&cDCXe^04PD%zCzHhuF7@=$ht<Gg!2PZM2t2}=iLwfzEv-O
zhq`acVIuhvH|`)TthD1#K40-k`+NMs@4evv@^8N6FTQ-oIFv>yg3FeH3rh`B4pg@2
zyxzTgPuJD#_6N@AfvPFy1?Lv3%0kR^b&X$Z^y1kcPo$U;x=aLfDwnUa%FYNRBoOL?
z|F~qllx(T0D`HH@TsV{crJ<^<DLJv2aEqhu`_f>)a1-4!p)YI9vW89ct->z@^YKU=
zN5Wht)pgfWS2bCHUlu}0<#3x5adF>cr`+OD5-O|dR@VhwKYNDbiR2>MXzI2?noQUC
zY?~FrbWG=o<8IIUZ{Kp-9SE~0EfMO(xZ9(xAqh$Hj%f-wFNuM&8Vw6Z7o&&IikOQI
zjRZ!QG8nl`=&X_j1EVBHYo=*nnR~2OWiuibP1jO)9ryQ71h;T~wIRg7JiA}(JM|kK
z>VEnc|6hLm=_~rH4em6s{p6O@{d?SSA|=l>EeHhd_KJ7kydex<^U3EgsJn)^3@jo+
zjF3Fa+7en{26BoVo^~a#B?>iLA!w|nX)9ue>#LURCdO&LC_oPAx?}tNW9DJN&jZ0l
zo}TVGj|(6F=u^C#85UT*cvZ;iPu~#y0(nN{29RVSsj7-No;iK{bDloDCufhXEVk;I
zr-^>k(X6gY1G4I=+J?jXuW9-YKc5Ls@%1mhVYS_G{qj?!luWxvvglz6=yr?F2TWSX
zwj+CwsWjs-ko|Ju$QiU$EGgh!ah|TOKPnR$d1-5@iex&JUck*LTNhvLg~ueMqF!H<
zrUOmaF^5cC%0aePsbmnFE84m#nY8l}Av5iE%{)#B0qdI^PLB^bzhLU7K#R%LH?N3c
zEJ~I0327usSG4_#y6rhXzNg=I<WNDF*qwejr*5^~u-a~!rx7;}xMd>8h}D`zY0i;L
z@ZN`1a&_`MDRnHyhmcDq&J|u#Z!NO2*vrAsXp4~=;R=poS>|$}GR5qnawbPdoISHw
z#Z-}Eff^fik%+6Zf<z%MC5E^-NTRUqzFR9QjjTJmZbQ{IKx8@Y%KDY^FkReI9;B&Q
zD+>5iVuk&lHx|R?hMp2R`eGZ2vK&&QM=LNCAE-55ao9=;@uE!<Wep&5VKDhnRB}Qa
z41(xqq;P~~xv;Syrhp6O&l$pkuzQ-6NO2~OJ4E)J_A|lFnAI)XHdK`<;`aPqcCSJN
z$U4(>-375xAj5XGBKdMt6;(yj6;UW`(}MHmMt(kW9!_Xusp|Fu-t`!*xqANm2Y=t7
zgzk-i>@!AJOyh*<@6eTEfB!8h!iNt>{QG-8e)*i$x~H;=EHZ7gMbv8$nrJg1X_~5J
z=X}P+TwviwV@0M)6U;~ac*HF;`(HdToy+sEZ7M<tMa7o{ewwjeO%{T~aUkVPz1A$l
ziBxZ}RmG~Su~yOaE8c#1Vmu!)ZAVH0V>C(>w=`Mc9qdjshx>cRdFI{QN1kuH;)6x<
zVK?&P_L`fQxAa$A!ZZ+<iMFda-{14|9FSGbb)EU<{!Ge=x9|7V*7ERSN8i<K`<lP{
z%b#Oq24m=|8j)J!G+mmFm!Hw9XE;uL_~t{oq~*+cn2NC7C&tAgL*ldRp4wVo-M0MX
zlP%9beO}ab#-NQPh6SYznN0QpsU?}25Y8y6&`Pr`M^a3*-4#-mU_aKDqMofwu`WiG
zZLoEBVFi|)q_8zY3M$heb<N^__{H_~w4=MZrRvwrE+L|T<paC-d$N#ZtN6nozu<R&
z@&*6rfALfPhyT}~@id(<R$`Du5<-;cixh&Yvg{8Ba_)G1ICFo0B!qGZ_1@F8Ev>E?
zjsx568X?aZ8F?pRHX4x&Og)EUb;wGfDItE8L~2<Mbb@k`NyWA>9356bZ4E|2-&P1!
zFd#8ynx<ymuUK^}j;AC0-43ZVix0#sn6l*QFmqj12wfoGCBN2E)fG||hq029#@fR0
z6A5E2wyEe=C3Ck71IZT#t1N2_x^2VeSqmYNa*3SeSOx>8GN`7e-Spgi^o+YtUhwq(
zo_D``&+&K$smr8S6i#YoE1vd8hB-1Ug3-rP<V&8xeYXsU5JQPxNre@H#!C81v#At~
z5_DCeGW5*~MIv}lZCi+my0&aLYutEdo;(j>hth_(AD(`#@6>N}sQbM?{u6qY@i}um
zPn`eur_9q#+xPUhTdr?k;ulBLHoVXcfBvuj75?X6v1w|8kPrfg!-=|?370_Ys_hEq
zp;jmvhyhI9P@>5Zkt4>m)J@NvU^wl`IZ<ym_}OuKy1!&qBwi$%bx*Y}CtGVPtd;~f
zu*@T2zoY9KQqD}{h>!}Y4PlPxw!%#ZY^^{y1n)Q<kF2j>Fpoz<@Ko)Jy1&M>9rfaH
zhX*<(kei-nwJw3`sT2t|>kU4^Fpk*v3O75fpP+_inVHT{OdhW8KBZeVOuG+gYiTzd
zOx<C+EpC}`!xPdLsHxK%!ZIP}C!~}#n_E=XlaPg9<OV7wsQM11q3T+6-XTqghzg-b
z=6NC+Ll}n(?<b&Bq^S%p!ZHn*Zbh3DS{8@w5OvXE9d;}+QFkkh(k$K+<3M8!bzP(C
z3KI=(KeBFqc>5Ktu@@`EM8Z#FX}jiJlsHL(w4f}CT<ZTh5<}6xrOO)Dhsfd`RZ}Ah
zi-efVfn}K&ywat`YBbIZNEer^5|?p*#JPp&Jw}U?-MOGiq_U-wuQhS@Br-|K3sy%W
zvt)7$(FbZFsHzszG-#t3PJ5>FiQr2^id^XT&1OYw%%%BdFq<t>*W~ep7)SEr@F6lw
zfw<J*rBq0j(WWZc9$~1ADFSQnNLpP2*O}ys{)`+6^IT~1+7RFZ%$4NJ+B8!z3S3C;
zF-FGIi8!8dZHH|tl$Mz03b)KG!%U-Q#1dimSa?!ObJ1^<KSRzaA!*vuWSyNOnTi}!
zQ4ht$JV%6EN&{@1@Y9H?G)pBh#^AzC2(esbHqU5R*FXIG2JCkSXlv3iVNA_5I!qJT
zuD7h$*XXL_>OA2t>Mdm(QuG9^7)^zeO<{sri4c-=%4lopRvm2yZOXwLH^Tmj@%7K~
zhdp6g3VU*xaI?n`XA}{m6iQxz>Nz8n#38tUI`d37+$c-TsZ1u?7RaTcTw9LE6Y5Z7
zn-(Qi`E!OqPUQ-596jTB=IQM{(ipC;KSHL+%a5M(c)!EtLh2XFqHLsDuc*3)a6Itp
z#g^5o<wY;(G-RQ9*q^vRI-1sUv#xl2I`i#g!J7D#xan&)U4y-_&>nUNK7Mt}?eiO4
zNbJrt+fBJ(zI{4!z3Rw8pn>hG<+iU_cQsenJ?rb9{$`7@mMj%H2XxgSWKE8syE|k&
zqr^m~+mb_*5*70$!k|#qni5r-ilbN6Ov4kpZZNh7B8h|#Go)CQl|q3i*+u(XX!!s1
zFaMUGy}9Qf{G&gnzq&(iKO#&grVo#t_h*D|n9w}`o!{ZV{a^kGA0F=cH-Gjkl!z?K
zBa`~ROaPe_B0+$2jyG@KeD4Ss8HV#%rdvXww4}9$^EvYTX3NuIB<@DiWm}d~!YE5j
zk*rDvHwI72MJ3j^ElpkH=9&8s@2RckcGXjLO<}(&iBy8DGs{qrT6JA>J{{Q|_blEc
zV!ZJ7Ah-pwsR=-DYt*V^3?79*iqcH3s|H<_4T2CEMS<)~ElIg#^t4$KgGC`wMw3Nh
z^<}ppbkW8sV~|2$2pbC0LJS@u7453QR<BrHU-S6z$o^^1cpiwUWH_t3=KAIt^ZlOb
zH1c#f5xmFN4K9KUrCd^;aRsDfw4$vol@`=i(>4u@_hd3n-(jo)^aW}y7{?JM6_u??
z-g7)nKq*B1I-lR@P*+zq!Wc{pym)1hl&Y>n^YHF1yVGlg)O_}n-yx5C{{G+peVXg%
z9Cz=_%h}enDI#OSZ0}0TWto_#1y#3%OTFHwL}fIfnP!KK0_6&dLX(Ng;-f~0fNgtp
zvn7uw{L_xNUwy-i&pv1U{9{h<zQ!d``2z?rbx~`b_j^)R2+<~kj{%b+IYpM?%<KZU
zFF(d^y3&TK8qk`Q0{gdLlHFV;G2U@ljtC#<KKTM|4AL*S<Bt038I4xN<QTvDOWNBP
zgsNx$@C`{RglQO#NBS%X$|9N-IXaZmgz3zD+7pAv&jTsHC%AGranlh$odCt^<^{JO
z|CsjX1-Q8o)J19Q$>0ggQ*j`=5n`ZfI#fsyJ=T^9C^-|Ck$E`MZEgrt;c?Yg71(MJ
zNbk_C!8OlW#tE~&#<m+|4m`a68fhfW^-E;Vn5IPOi?IF&eo`d4QeQVpmlbs?TqUUt
zn2Ib5zc9KYfF>39<>D5Mt_rpz#UigB&uAgA)}V~Om=q!?LEE*+x}&bDvho(<KM>@F
zL|#C$#Nrdwrj!oC1(%?O1X(g+wyqHf;ye=EObP*`D@@%ZOhw9)c`R8Gq$)t-X+#0*
z&4#+`$crPyOt-zJyLyhT8p5e!NFOe0IEfoahT%lpHZ)a5&XIDzNtomi^9f``<%AX%
zZ8Rdw?{!I}NDK>5ID9z;G8wHjiLxdPvE&y;&JYvsJdzgA<UL6#Hdi-P{SB#IA>9bm
z2_X&R>A>mn15${>xXXzY9Yz_XQdmV{x|yO1bJ3H8C|T1a%9@&pG@4}z5CsZ@G!>{q
z-JfTNkP6Xwl>P0%hoiSR-=OP`=K2{`jF?)X>mE^WK^f$A#7Dv40_V{a1^nvhOMp9%
zSYxQPq^}j7RcHb}4OEKqEDJMgIU&yX2m)>zd3yJr<Kaxog7axY>r7Rt0zw6JfcdnL
zlqPEf*79^7@b{1Ox6c`dvZhjitt<$|a2g2HNZqbbQs6Go*n)Quygw}b-~aS$?pBu1
zfAo?{8*I0xc{;&y<n`0Q%a1PNdXX_rM{+aW%@x~M9}&hgB6>baiP|b&-nRVw&5@68
zTTa8mw0K_KU6E7d>2M|{!B-C_Dhaic%=1z-RYI~mPGn^{&mJu`oo#q{8d<d!x9g7U
z-ty}Cnts({FSrwFEkYK<Oil`I8_4o||E{tt1Qtso_%p#xNZpa*LQ2KffE1;D?iPYt
zAeDo23NwO?(F&mr$YNu$c8!~lNXq2zhkkx>@A>7=z9ebMKm9NM1l&YS31cc+<5@!H
zJPf=<@$u(({I~!7zv9E|6R-CN6p<__**YQ00qLSJ^gbXct$aZlV$9RT&CNAF#PTzg
znZ9r6+m?@?Uo$O%Ie4_FSbSifC$v=<Q<cou1?^<4MJvfNjR=t#2S;nxSfkNO&^8q!
zCsHVOpLw3SYFp-I=HcNHVM`IsnBu55wV<yXmM{~sWs6Xki_mmB6JjZDG)+yr?NACX
z|FC5)mLhI(1m~#g3Pe?=wb~XMf~g7eOq@oPl2q1`T*i-cDViusj6oZ;ZKykgZ8S}1
z*laouPe%@i1Iz4+-x)<y_grlob}q6zjf5b`%904goUle?rKB|)Bf&2-+h#@ASd<im
z@ZAAT5PYDj3`(hjm&xhU;?9i2Q~>fRVXgVKzEi)^w^A#E&XH-Eirpav?CJ_F4E5=V
zIGvG7b9M8Qw7q8ct1o$c{VNoa;3IJ!u~rjMTz6YkbPPFCTSZ72yV{iLfb-~nO`S6#
z35N4P)P`jl$*M@Tqjz*|M~;c<d}ROhfJlk$c2nA3=L0Dxtf@$DqFryw*i|W#sS#Q+
zE;FqZEVCm_f$6j>bbPH*VZoRiMi+=lmY1SR_`@U4XMXkNmuzlssIPA@%ZMKa!ey5)
zOhp_{49mzI67*}vL_rSpSM*mmEKeVJaVzP<4(-b7z~~}dpDs+cc{=0A9aXbN8O3q;
z;ga(zYq#|Ho~G}~%bDaIs_7WdCzkV`n~y#v$qMNw6oIN+6C)rM-p#mqD4xBGaD1~_
zW7`_NS>va%_;hn7Cy$YW=oW0%(ryG=1*Y)<qG!`cd=B{1=VmzE0}40)@S;STMp9kl
zmZg*kQX(SoMG8yyiKM|~iO7Lr7zj?%uY2@GQIkTEnrmAM3EnLT5~xf_2_*$q8Ayq+
z%ms0v6fu`uU2y^?w5|}fDE?VdG7?CI5Jm5laxTqKZIG!r0q5hM96VLqvcA5-uJ4e#
z6jc0hL?GzaYlhRltU04c5^<*k!|{x5Hnep|B@`w_Dr?EB4!`Qkgz$I<C8<n75E!NC
zD-9~*#)&YGNG&n8MN-!AWXLfRQptQur3fM7{6vnv0EvYt09!IjLJ|e%Lkh&`K}l58
zA=Wp9?hd_qj%wE!HxklY+_)po#pn?c$Ymy_qBSx`lVW0+MwV%$>Duyh&Ivaa23SH6
zk*KC68$&Q9mnNGUVJt~s+G0TvgJ+o*3_rYy2+-brURuevW)Tt7^mrjjS{GhjWl>o$
zPBRau9dGaV+&>&RhfK_n1uToh7|ZRtqwfq-B%DvYx?b_=b;C^!!p-FKf#LC<^V8GC
z@&SwUEPlZ!$>aiO=kYOPtRpH#yXx6=4V&xIG9PD0IG#BVf%7mF^M=f<I?JIlOw)iK
z&$NAq(WTuSd^x;3FL)XfZB<&}F=zJsBU!+DyW!v>&PVE$(RGcG3U@lvZf_w6{P~PD
zn$2c~6q$_T@BO&N1qagbv1w>4#kcRC=(>h=+cMoB2r)1QPg`l84kP*P6RXu0sfzMy
znr9G_rnWr0+HiN(k^F?zn)SAb+3Ttz`C^7CO0PunC#+eKE~_~=?rAsA(em>9Vf)Jj
zx+bT98=fvr?}9f`N~29zG?gin(^B%VLZgKx<v>ajrES65T(B%Z<P-hOhj;IJ_p22T
zAHTph6{g;xR|y{^{xC4lj%f*~s^P~U-|$cV(U17Q{_CG|JVC$_$rm<XF3%>w`$$fi
z5IpO?;e0yN^(#O!Pa~_o<9s}_Uaz^{u1Fbn$0L)I?9a<3V<=gzI+oxOlJYJE5CuxE
zO#!m^jYUeRtl{Q*jS?B}95KT2T;3zzxzb*>2Im9MZf_aJ38@XHu8}F131f!3X_>-+
zP-Q);FW~Vk62UECbeRHP&?-tukfv;Ew8eb){1ysn8f+y9LgF&W3@K*fyx`(@6o7!X
z3TZUnEhJ-6LQsiJd)x8iZo@pk<UCC5cLVniJEmn~e;80XVT|FtKp?V&Tylp((O5%Q
z8&YueO+yZ@kQFFJ3MnN*eCIbW1{&`J=V4-59JN;TmjV+R62G2n_%}M#efZ|*Y&RVd
zg<0RSxxQgq9K+KQtu0@C{t4}BjTMgZ^uVe!_@r3oLPTHAN7g_3jKw7^p1Q55>5(F%
z`YU$tz9GyD{q-|U-Sh4*{+j0I3WuOVVwwh#GCnTlR9jUH<HYW1k58FbAAd|GGRODd
z;wy_ZHTw1jTgL+9Jsmh-lq1hxeT<BS#kI^cyZwQtX~<IWbpMD?idX#&?s&jnKS#G4
zggc{}f#LBhZeHHeU)@j}$^QI=cMGe&D=yZQNZOFA8eK^qAMUAajgXSQv&dyY#D%`8
z%a?Q-$T2d7NO<~Cg6q0?Gqtt^H-9gWo~IezZOF?+aE{}{J06#baAD~<7Z?u*K79Kv
zH=q8v$inlCQIZq`=fg<-;!`GfV0!mUw%1!yj7Sk_`W1u;oo4cMVjfQPU5C(`kQ2*!
z25Z4on4EzCTHz;8Ek;t9X|@egG^F^$8BR^xmo}uJ96VA4rO^hWa|NIqBO!aVHW*{d
zFJ}A?)qkZFwrOasMN2~-Oa3RtM2wN9@95WSlr)4;8m!TIq%mmIkdt5>4;XD}x>cDT
zhQRlbs2E`xQBopJ0cGbH$Z;V!S1v2IL3b;vrY*LJl+hv)<BXdIh!II)Yc11QvQ#-E
zjAR}U<mHGkr6zBi23xOCs=>AsS_=GhBKid>Jhg4G#xUo|VhYwkC|%S@OGL?%(*wvz
zl&MK+BDjGRJjk+s5mJzH(JE=F$wm``N4GVgu>BSF_67Cox~Q>I$*Y8vaMOU+iq&>|
zL9&1p-zl~fq=<;cTa2zy+F<2^A05*;5T!uW?d9=QBdsL*g^)GEbm*(Q0us${n4TVx
zDWdXkXHwVQzNE6{HtnZVS=rV7#Xp?UF_lYYOuT<Q@$K6OrfFdbiNooP5}9!*;_ZjJ
zL25<YR75o2zSgL}`w^RudQ9~AVMK;VT`StY$Lh?}!-?AqqivpNR<hzS1<tcWSi{B`
zwl_D-v*)}#;iBO9bS!0pall$lS6k)~Y3i!v_@w-vzeX+@teioqk~@;RVwe_&^Mv!B
zyJxp#x#x5mSz`H4`gtU|krV@Uzs3zqQ8qRWrmo2`6a-L=bgPzVV3re2ZL#&tN6&AW
zN5}fSFwCBuJjcl)7f0Q7v<-~Yf|4b>6f)dB+w$se%gyx)QsU}*L(_NVFrjV1rxZCf
z-RhRG>@GPbO(J3R4Q+o@n2=_5K};-{T<A;;Gayhk#4s1IxGGR-r4)h{<V4DW=$G=}
zwGwG+yl^FI^8-J#_ngO>WpLd8>NU?ly~D0H=zfci85;$gizOzH4kZ`<@qhUluOANl
ztH1gcDHL92CYOAnG6iZ&Dy1AzTDvJ0*X!wY#?57#tWCw$^(|eakdIHS`-(bQViqhR
zP>EtHiZQb4I+~{Ccpiu`BC|v*OH7`58tLja{pEsoIPA*gEM&sv&{I@fqWoP*SqKma
zaw?#7Hw;vj<T#8>71V80R^LULflviNY)qMGNhv^;m3mYPYi)693sE8{mxFG~0vAf8
z166QPejZCnECtd<BN=B;mI7(3V)lu#wAoe38lwzdD`@K#AHTZdi{E+4FTVbO`pa+m
zn{OW2Po5=4d`bu*PzZ7?*rHY6q9_&#qvZD{5-BAyMuK<5SX#`>;@R(yB_G~c?q1wd
zS;;g^4C8+Wq=^!Ox@kEK6XNuQZ)@V>Ff!8LtdSB<_xI#!z@0~ArKp+)qM^R+s5Vzz
zee@a1RAqFX6SmtH&CxPaSC+P4@$Q?i8K!`2Hw2g2ZdL^|@EwVM@?@!r-lGKcjX_FD
zUn$<_ghL@@#o@yf)BV8j{^Sel+R|(`Y;O%8-o68&aA8Jf$T871E&a1+9N+zl{^pu$
zbH(Aqdz2RJ{`yZj?e<6#{nZt!X}Nj%g4!DL@xW@e;WW?XAg3z`rKa3eHM4WPxVu5+
zgfuNE3x1|DmRz-jYKy&kM%=wag-~vTIZ?GOy6rf>`v%oiNFizaatQ6Ow{*=*?q7e)
zG7ZerP|zqb(6lYhs-;qr`1A(S71h;-c|79Ak!f;N{soL@y!Sl3e~+mfOuJzjj$|Q;
z%Y^fe5CZ4@Bhh>6rYVd)5+TKc9tZ(jcT`=?;dp@LS?(vg>t{t={sTX4-%@E^$_Ubc
z)Wx1KoDnHex2A~Gl_X1p=$i6imV&xzh%rD&2(7VQhZK^L$h?#rqE-ei4T{1r6v7fy
zsiMaWB1TfW<hHsMtJM}akLBkSO9fqA9Ijake9Gl^s0*}t5*k(4=>D24E0SNxu29h9
zd?E}7ru`F=gch>s$&^K2%0$7<JidF4CbDW9nyZhf*LRry8ihhiMXDRZu*Z)_v`)lP
z6KstXnyS*oOMxc%;>0y&cY)0E@;*$c91Fp{Hq1-Fhg4KWLXlENX+y0wLR+*?<n{`^
zx<OT4(H0FyT;3DhNEn_-u`nWylxT5TAyyV;ZNbWTL8@w`ZBVPLVh&IPRS*QBao!<q
zO;(ze1+xz%A24-A)o#e9A-NGD%PKd<nGk;X_YEK&UZc0S1#M&`VLBqxh^SU%X$hgA
z7Isf3j^~kS8p{hTCPeULBGE0y?VAFXu}sTMR2E6g<Nio`(IHc$T6fgdEn%#2i$ms2
zjO8UXEd~EFpB*j)tSN!;m<0RNnR)gMA+ww(c2i_@k#|o=?ryKRS$900M>c(f&m}8r
zP05`GDagxmv9J{8mXCn0d2_#K+t(Lcg5+>s*d0e2Ss{}nxrKVWEkz<*lS2SmA~P{~
zRvDHh7NmoOSFf%(pU&LgZrFAWM|t9-xaQ}tABkjq%18v$<d_%7X<F!O%{T84c<=e~
zC(p1(l7eTu-ITT`Wu33<2B9?wT_OWj%`%h-iqeLZDYE+k&W8hay(;LLx+^g2oN>#7
zu?^Z(r4*1O)^@~YL~C2V$5|j$jS>pKj3hTxt)C%9g%raN{=SVqvNF)M9nv;PyN3D}
zr6<I3zz+*uW3Xf#3rQK)wdL>q-V6TX=f7f_zjKx9q70Lg0EO-#O1?`7iTAFo;GH98
z!P9Zz?&g}x8e&M0B<sHAG%a+EWu7e|!!XUPS3Nh|HPbj!8(9ip2x4;7MsaghkRQ%3
zJnjqRd6`C@-P|H_v4i-_&pf5F#$T3l>933}z;+?=9&&3gpmtXdWfYBET{YOgSf*sm
zq<JiB+>}vTm%~?KJwc`vBU3_ZP^w1Tir_pVEaaHJ*Sz_0Vp%4%E(IH*B1+A~rA&rP
zJfs|&jVZAmVI-Ya{K*$BzxV7@{`!}<{HwpY=g;2k_+~$mgCHr%=sfGbLK}gOfpy;!
zQYKL@TH`bqXgreRalqG!X&TA7L}1#srD|%bN|r67`OnaPoqQxp!;5Rpa5(bx`fKXG
z!=iBKfiO>urxS;Vd#bi)-9KYIkM!Furs=>`1>Q#@EK><THkp))OEad-G7Jc%IG%Q_
z+B>Rs2PBM@%u(V+P3|k0k7U1)Z3D((wPF|sj1sI`ixv^{{2IF{b^P;jffUMUJ}0Kb
z6N`&f+grRp;Fpnjwp2!7tmOTdKO^Ks*IEuAKJf1SBUN3q85g#<&xl#lHkPig(fx+?
zYD->5AYv6|qLV!5r$>~vSSb+6@p#;`?rIE)-NOUaE5^qMVpv$M*EHP<)kvn(p|JF-
zwuF^0WAx3{Rq?V`n$;Jdv&<96-4jV!*3Vwi%J@ChadE!1!E!-X$f~9>E#tdipa`ti
zThJO)w<Q~SJm7OC2!paUZB?;<IuLb3y?utt9+Y94MuJwDZc7L=DmvEdRcZVH2#Z?%
zdi($X*e4-Np#O5<skNbQTc(tlV=URFx+0QsSzt_cDGX?urn#&%%WW}9QAX>4sp_&u
zMdCwd7)Gktz@<1Ng(L*WJkJ2w%K^j-P1X&viiJbu&Lk2-E3~#3=Fvh52_G`5>dLec
z$#B@=MMkKiZOhAq2o9wwjI+vMnhsla#1J^|J}@3nMOf_@b`P)(mY3Hr5V1_Zax4==
zS@xIs5E!S0Mp~5B2x(AKlG8<VCgt}$KP3uXKBrJ7Fm7TFfvilqnIbSoBbB}^nKBuT
zFp&B!W_<%nu^gT_zx{@|jHKvG)7mPg6d9byYJIt3gGeB9arc_K!L~hBzb?5>@4(oi
zR#OFFJ9)?GW@1X%%A&0zxfwanBsY=WTn@y#Ei3fj!pA3t(VFe`HCiijF2yFFFT>d^
z7`$h<AJ{+c2#aG?88(}ir`?hL{*3M$oDZ~hg|!vqwBX&s+OGKQZi7*<*{q<t=J?AU
z<6$UVvBl$K!e<Bwl&!I?rj-U46WzK;nu_BXd9#~2xWE!3hdJ;xI8KA3>lT`(=F`S-
zyMghS56r0mS(T7vDa)3_+Tw>*8r#%NQSt4=$Y$Mf9tKY5@^V>1CWU|>M{H9S*!P7u
zNG=T0u(&ei1=cs)@}C`M0=ej_e8R_wwHBWRx7Ta-#}m;fcBhH^^8=p+!x$37G~i=-
zF{cz+mW5SU;oZ!7)l#=TQrSz!(_i8Mby2SMFG!)ZU87s@eqtVv#KjTaK;K;vqq`jV
zVnKK0d@=vna?!JO$!UqQ))lH`P)V5_J%NZ@&ZG$P*IbAF$3D-lu1L}G?AeB9)zYl4
zsk#;O(XlL!M4)cg%;Sg_caRNnNxZsR^U2)}PveYHg&UVec?KvU%V8~+-}4Y7##n?>
z4CjGy9J#r<p{`5jc$~(<N3)jGxS$a9O~ZMfsf^<GdP`SVY`S8h$Wn4XpD{{uy<O3_
z4LJo)r?Jo_QsCLMYhsLOYpE_9wUiQ7Rbwxvg=O(br8%F^tX3<Olm)j!E~SiEa`aM`
z3oNM&=VDk$IiW;GnUZxUls3N<#n<Z*2xUs+JC}l(QHmHq2ni_Wc|qaJ1wEwiwTUUE
zlH;}(hbw-1A;`*K^pY85fvyXo`GZ%A&#tfe$M4tt_17Ez>cazX_ajTHkV(OiiALjN
zMkJ3GrO7LD`56Gj3(UK%8*)xGm%p3c(~i}qqpEAxSHB*%={GvmEz7`k-Z3w85hm-J
z@#%?aDXq91J<nb|r`z7({laiOVwJ=;J-X@e%L0C87>=Y2S_;nh1JbnAs|~{(a517y
z&DFE#5F*3j2u(+b6R+RChki?MG`)>TTeE-vhO6zCtLr<aVIVbqxp_7#dZVb?hWUJk
zl$n=>*)JGVvE4MR@9y~NU;P556V|psLYszubwzrq`0(x>r_%{91V8zs->2^S;%N1r
zFppGMTSCsTjD?Z3j07cU`qIjh+R|*clow>pyVq~|_TfaxiS4SUwE~6a*^B3dn9%}E
z-&0jJb=|Um{}wB93GXY3Y>P}ixuqzxQlPb#_U<|JJfX6~jU)5^3Dd1gLz9f3C#r5k
zQyEUDk~4Fz<ZQIc1=d_OsMUt^>4ftGO=S?K;_Bls5Uc0lM*QI|ISaD3BvpZ}3Bw~k
zdaBh8w(fvZKKRa>`a?b;XVSb}8haU;6Qqc38e;IobYW8oK}sdKue7GBDoj-q-AoD&
zCCV+F;IY~uF7oSKP(s8kSwcigFjf|nLW;O$K^udqDo_?L48oQ}UzSBhlY+x73n^wo
zE)Rs@JjztGjb#~5gwqbKKq`T4TZ}H)gaX?wlNq5);lnf?(k&QUVJeT0lEp>B;*h&X
z`VZeiC^z=BOoU}bX@hRp1xKJ;g6TjPf1^l&bcq>&xM<DFLsDByiV%WkNs^EiMpZ~#
z{<lhAe7BdCLOBSR+*nGE5E8@wf!*t0lIH=GDuCmDRer`XlBJ}wZ7H+F@*@4t>q!oT
z=~V0=!J%X(TZOT0fj=LQNFms6*Cnu?Bf<>`Q_L?uI;57gZHLtL5B|OZZ0=rRRvlSY
zxLBqj!nT-{aLdBtBg5o)zdxXjWW8=ni_d#jjV1bq<MBlE-Kt1y`c1{lRmVry9hDMX
zty-$IXL$F>d^j*1M`DC56w21bX-1fetP&E<s_R%>AnS_5?D^Hxz%L&LvQ#WCGA)5&
z2`nxUVrJd9T&)`p=ZW?Gk!3&Qrs;d%ER^=Pq<q(x5E<uyOo<P>6ZHuan%if$j7tCv
zLMfKhnH1&%lTFZUSD328PiM}LJ3`2e=ZWp}8)~D_P0zge^5^JU=6R-R41HVCS;d=o
zPuR-hmc;#OU|Jk9MgGpKXWU(PeDd;&zA=TU(9~$F$)wV()S8$QT1uocW%XFKWF=5i
zK{}#sS3>6Xf-*HZ&Sbw}yDL%(#28Rg7e~031-d;4+&mPnADLwy5EQ(TG%dC+myi$#
zviRYdx+Da)HXNpb&AeipCABhyR45Ql-?C&F4hNnsGs8474FfVR++J_c-+mzFqJqm8
zcyf{&B}xIzSc5i}y6ZWe2lh{UVvO8g3$!sT&f!C#>pD*7kyTsV>&gUt3^Z-Srf+$5
z_pCHjmEt^2T=flY+Y}2@O2i1KQ_0M4wp-e!VVVZI$}mh5<2cgPbwO#(Gvic{T!;Oh
zd2vPOXe;JA<GiMmf+Pe=+wv|I<#OtmnXt^I%}ps*<q#nim`IEfp)}TBF6pxPnhAm6
zX7Z&dq)pBB?cK#K9STnA;%F`FZ15L=zR~)!4T{8ZE*lS{kT!!q;d5D~{@q*6-`Q^Y
z=joQO9~S=Xn;l>626pqzDk|p1v!n%?GBOvn+2Ur5mW5|mRrGy}wHhf)#OOGbSj@9$
z&&c*aL;JPrR>=OE)9{Y*@t&$%Q>`{sx7P&cVLGCffSlRizbD5;)i<2?k5pMO4LhtV
zdt_xArt=XY!KeZvjKdxgBF}#Rf5E3WACWT%Kj2SKeEnDdhE*-7tBh8erYjSW_2!!4
z_=LJ>Q5Id3l%dm+d7O#E3DY#>=rG1m)iohF9^QUSSWYy|h<1%_*G#(yV)V@C1D(-)
z@!6-Bv2dWN>$0Y-3~rv8jwgoWf%f_5h|>q=;}gO*H2t=$-$KINyu^ezRI3eHSM<jN
zcYfrQD~?%n)pyi*!aQSEH^k}0ayoE)`!!*5*mi}9GyW9uv*Y?|4T~eAaLWv#C|zt_
zVcQ;r#PpkD#IOy^@c~jq6KSkw94F4_6Yltks#<ilB_~hQHRUBzuee%0BL#<4iWEG4
zIFOI;aMPLOCPFGXkQh(6-8<rVKy@2LT@E%e2HbdN9DjIGVxA@}i5MKtc|<a#Ty$+x
zYP>IgU}G()LONHfKp@@Xn1^#|m1>PNWnC<#E_Mhd@u5)hjn+l6b*by;R7xLEiWT!@
zs4Fn0x~vcqNiIlT7i@^K1<dO_-YtZjsg*$j^Wh0U9BJwbv+i(yA*M`lp)|#+3hzBt
zt*MlzHijh@W}ei#z(8$6X2~)QJiYlPc{!uC#V8G-I31(Z#S~x~qA=fEHHb^eugHrE
z?ZVv3%9IO2)EHeOs}<1|a(fb>MZqML8*_2o=9F-=!-Wr}QLQw>k2EF|TtZd`-82Pl
z(-iGpOcBvEM5$TE8JSDrLMTzz491e-a!WUtDR+6eu2&76EI)@B%DP($g-S(wF0@AL
zh8TTmA^$c$LJ|_3&I1{NkrHbpWoc3X(#O-8d7Mh4{NkXV#=ztq)9A|2S!?R5rdxGX
zO<nw<hdno+DC+aX?(3g2J?+R!(A{kC9ulGixrdo~i6kG0(sNoOC!aZ-Mjp-!KYO!h
zKl|@j+bKcJWt6!1!0ZE8*IU}A;`670ho^6dE}=|Wt;QHpT4Sxf;2<*FZI2Ryr{js|
z*A?5VD}++aeqjueDrdqn5|;&+2Ewv%`tTN`6p74vcjA1W*_}sz<YwCSmN`VSuFw*+
zR<v!4kWV=8u^OIT_2iQSoj?|UaU&v%g3nPgAaj{8hk3!BPRQs`3T)dVqr)BU5vz<*
zhUh2Ue7-c+1=3U)ZHP*vY)A0AL|{Y#=xSARWhwd#X!r7B_Y=eMaCuf#WH%v|L^a!T
zfC}YO&2M2kyjy0v>pLFLksGf#FOGaZU~P+SSNQqNSKmBv7-p_N`G{YB`4!9l1FNeW
zqAZ~E^c_nwMx@jwr=<jS)s){YZ5W1$^ZCp?FSJcdRTqcrFiy0!MeB0FU-v6Q&X~Fg
z?Vmk=#@*d*ah7g7GH}BdD6yq+95^3Gg7;i)I-0hn?_0)kAmw74kW!RYqP2{}nR%Qq
ztIY~Gm5ox>G)QAuBDk2@G!=fHi3kX>SY$-LII|VTHe_4jLn;$f2{EM7l-06)XHxvW
zjDQvxQ4ua$Nnr~HN`!!N3t@I8%bs)jJ&(oht}eN^@BZbSE5$plsOm`F^(Y_6i{l&@
zvJ$kl<_|y9{M{da%wNBD{OiB`z=y*lZYiGfuBotEAf%$IELMq<p%VoIW2`P)4<*XR
zDV3~U@t*#tKflqTt`(WQe?V!8?pDakA}U+{4@J4P${cunyl3_D3zqW%yI%9|t1r3#
ztDkcH{DzM{{+M*}52g!XSf)T)CZs7$q$~ssdqmEpWrStKC{2HLMcp>|@l5ib*gnI)
zcu|7sQgR+g#x!F^#!r#tGA_(mt1j3HO<IPMsW2UnC(n9)gQ;7RcQkE_HXHKg#ILo*
zXoFCt#b;B*kLBgjUEP9GbX|qdYoedfb%UQrvRzX@{~W3ns1`RnRAW(Y;-l*oqFEKv
zezU^wJ|N~3&3X+wky5~}H>7PvKvHGPI8K<>aGqxB>z7~}#CRrLCN&~^j8M4gR6@eq
zAY#DCglaqd0Dhj3$#c_dV)v3ENv3H;5U8Hrp_(mHTWr;UFPXE9AOuHTF4Hchn1&O>
zeowWz!fL3mZkUrSv~5aGWeO45ZGMxl*8l8hnP;TVq#RKiq*9lBkRc#gav@3S9MQT)
zC`%N&G*g$QNa-o|53Mw&t3c=i9gP{0W^(p;zm&X=RE3omBHlStyeMcwK-61Qo4(gv
zdAA_kT(nHul1*{`RVh=YNDL*Hn4%-3#A%x7>Wb)RVvL2;q!mdDREkJ9vW!QB2n1j9
zBqB!qcxIU$B6(s?9G(KvE%bfMs@;$-O%kJI$(D$+x+n_7@{$s%D8OXOAoa!n8cKNn
z@-h>mU<0D}$W;Ell*N*fF0R>_BEgL;hXHA7s?~;0L-+g!H&~L;sJbI?VdCXNeNU;Z
zSLf3n9UP+82q`eCLai>1bmvJVl;omiy1=AKh%|#MnH;4x)*qRD@l+}y@i7tHZvmkL
zh#W84wbF~Tjm1<IT55VJIPOmfAy~B)e(xBDqBpy~+G0XY786lO`nKlos^@q(bNcXx
z+eYvYfAWlv`-$P}U-JGhf5GB1S3kN%2$)Z2PUi`1Mbd^K6~}qu&Drziha>xOVK)U7
zg2(fO^9ifVf#-W+xF{He`=>p|eohRQXPcf?t$5s@q0aQH4O&TD@CXv#J6^tc!Od08
z<HN`j65VQD$N|#evLtH@VIajo_JPy!cv07tX*y&yiLfk8<HDPtzvjn(P^0S(Q%OE~
zbxZProFdojp11FwI37-<BzZhf?3YB8p5914zU_Fn?U5vs4-m`Yv#KiuiSg+H;b*9A
zxjgi@<lwnjnlarq!cXMrNI8;3Mwt5Y&s#EDaW2zsTdlsA*au%SUQ&pn?NimoL!2;e
zSCoHlB!vAXwpAg13+vZcKmHw3lI-5y^RS=LAKq~%Z&}-(!!Yx3|HRLK`5N0@^YAe8
z{{4<xDIVUR`ROlT<GrVDYI4r_sW`=#X~Ei#^=4B_5mm+RuxFlTtf{D~svw)ZC&xh3
zY<ReTD$_`<Y5SIOalCx>lHdD-KO{y+XH9{bTf=cU6BfthXKJhIpWl@{Uzm&af#T5a
zs}+b+T2o5my=R&xV(@4oaBg9ori<;uFfJY)BW|-Iq>Q#z`QNz(9~>$rLRcURvamE=
zxe(?U@xh~&E)Cs_*&^pi6qM(BF2HuFH93`)v@{x7S>$5TmmaKJT$#=*S)k#f<Pgaa
zUE!5VB@lIuYy`^WOI*aFr-Vw7{DN+Ks;=W7{!zn^KfC5%{?$GI_RoLLJSUa_wgPJm
zT~ih6g(^J3loHOBKYMK})_sFA1tSyXe}?w!`?v3T`TQBD^TPh@X&TR~RgaragwY{h
z-V#+snjTo*ea-r_A7lD0&t;LFPs6}z|3ufXaKk|G0e3o4buII9W}ar2)0xx!Khai}
z^I^|)I`X40zF@UpA*zaZZ{M;^GeH=tD@DaURrV-ZargX|<KYMyrl%*UTB5e>_D4vi
zwr-^)CopwY+>LHVS%X#}Y>N-+a+58$JSinEIi_KR#?aIa%i<B<QC9{#gu>^!zA2nL
zWpGjxO=;tdyAMo@qyOj&7F`pgr}Z-po;Y82_d1iFzNNo@PDloq6w7{(YAb}1G?gty
z2sEO*qVBhd`ph@~<Np)k&b<2g6?lhK8liMiR{5FiJt0n|;TSw&o(O&^8LVk0&kKuH
z<oqeRYVga<v<zq^k$yptky0`4KcJK##6V*e+s}W5pBE%hHw_Jf6g^30?0QQ)pUJ<4
z2VP1Mge{IzsgTB^RQWq;y27hk=ApR$WJW20ONINDLqKMUlEszkT|j7otQ6P=AuE(R
z<A(>*GO@&nwlxWumq73l?+c2@`2l@8pz5w1Q04(YoLEkKq{wWpuCY~pso-mpD~zWQ
z1Kqa6be8jZU~x-%XoOg_I!WWar>hJhdQOKYyqiIaG7TQioOVxmA4(?f!dOzuAW(Q$
z$o#shsBBdrzX>7{ltolk$*VCHWQ(6mV_0gEu1i}u2f{L<Wm$Cy2|}pik|m+EKtzB@
z%Aw?_GRWYl(@fWQnC%_GUlDv_ae@7A$2eUwvu%qOkthUV9+=00=!@vRsx`vT_$3xA
z#-)9hq%14!RD#dbI5C{h7sSjCAq`F`k~TC|gAwJR{oDAgwl}n0T@-YbhG>`<=4B5#
z@bK;(`-gW#@44AD+-zGy%%Bu?g5Uee=RCi=rZ&)8$$Y%$_@U#vHGFz!Sib%XhPQ9a
z1VIN>4CIihtzcG}VH^;3$7<8jR+8U++3?SP?}oOmd3S%}=kLz^`8N$edw=BNSeRzz
zQ(j8*TqbqHIB_1!3+VY(&;D4xf0bV_S{7@9F;upwqh?pkBrk5)Xswy&h4s}nA!Xh@
zp7?AfP?bUCOh;*fYnkc#hPcf1ouRXe>ve<kiQ~KXtZtv7a*0mN^TOl;n#?C3-SMUK
zJnRR~OJs26lCiZ7R)G<4weHLBN-bz?0Vki%6UW!@c>a8azFA|cav=68lXE}{S@Nku
zlk!ZA8Dlgkjcr<@8_NX8jV$8<?-q~+G#+C)^o`S=VI0}se2Nl+I6r|bu>w`KOs8Lw
zQPl04)A96!zi)r=M}NrqIPm)GZ&=cnX&iWZ<2ctHuV3Hu<xk)8;qlCKyWzvTJ?mGW
z@$dfa->`fCf&0gM>dNqw&%WSiUwuW_baYL_K4wBJ!1b%`mg8w)f0|JQB7!XvtBU$>
z)wYCCa&>*zqm9DTaJAm@#my~E(_l<XqYZ6sndb>PWsH^BW{tJB5GkgG*-M(bA%sBN
z_biKJoTh?El7e|$2oa2}kV5hFxFczesvDG6Jnc>dojF`r+%~pc-u;Xsp%A#>NiKp^
zSW`ANnN;|*)*^Gc2xKT_4q26FnUYAO5J1Wk*=2Ool0rb4h^+(jav*Do$dX!@K~-{P
zDwr-n^ytcS8=*>+A!NcB(Tzs;mRuW>PZt=xMTSJyf$N(M|LcGDoON6CFaO7%5mWh&
z3#kYxvFb~^JBCEe8Y>LS0=8@^lr`w8BFA4Z()1f0>aKqB_o;i!@vFb()yt1cZf-cE
zVq%^L9-kC#-%(XX*Yx=AHQ9f~wBNCuMuufX=fK>bF;z>PCsfmOy5A9UM%E3(xiHc`
zq{QMQRb@FG_B2h6k);9Et=7EXo#5#mo0~0G3EsTDXPAG*vzO0FE|98*u5WPX6XP@%
z%D>dyf4JxRYR#&zsk;@ydrnh1pjeYYYP2rF)0hgYOy<InlFFc)o~xT4x6BNOC){vm
zb#(*Uln0$i`1we>{{>kA^MS6vVbeEM?v`o)j`K8QHh1V|1>(f<_`v@4S6ttoxcca0
zmXK&R9U>Pzgph^o?uQf0;U1L})BPJ(mBjSdEc1+3nx=0N%1|j)m}HeLVc-~9<`L%>
zOkH8L<Yu$s`1pVj5T+w@E-I)(5;q7p5|)w4MT95-t{gMx$G7C`TWr^3jls<$X&LC+
z4y{|F^hC=gi}S-iwW-Qj{(?2hLS1C!k(^R_pomP2p6CKDOw3DhvKp<3QkQkGbD$))
zZZTOBM8RpuoG?m}mEt?$PYi{XrK*~s3<$wI9LT3fk|{1jKb{JrXFQ{&#M-hd5IGav
z%yK@J8-7z`>k30n1jgf;d6^LiL>2{fnm`~i#!^=mIhFCh^Bx2=b#-Z)C2DI~^*yWA
zsytNuR4gU}B9Mir(G{ktFl`UECQFNs0p=s|^nmvZQi9MGrfMM07l3ELFUzIvE4Ubn
z(U+oy5CSDK-Q}OjYDH7o3t!D*nucHl%lyFU=@Gxo7;CUvkjwrJTPdn~i|tluT@I;{
zg!l7>6Bsz3PNYjqF1f(GEG%;|Xh>DsxJhW#W=q#!v&4)XPUQ64nbhSJnf8xp9r42s
zEmPrrxr|Hj?(45ev&8C}n9Az2X=^_I<Q2EK*F3-7u<kly3ixGWw6L*?jR`E@{5kLc
z>Ptj&+<f#5-8SX$m@?N&@RD258eO+VAV0f8z_%7vHGKYK%YQ8u&W}8ej{ncUd&__S
zFaDaZ-aa8xMk;|40b=CMheuAsMBmgrzv>ysz*p~|IG+v}U9@InaOB*v+a0-HHCQdV
zUf1|#!CZ<IJ_L3y@!55Q;LLJ57nY+jSaq4KDoIR|e%)fEpj~w|x7RGm@bK<|-NO^j
z)tYVJ^0*(l-<=2{u_PEnc~KcHXtba;qF6{IAQbQ3J#lv%Ifuahe$S>eH0z%A^Jm!2
z4MD<kX`)M=nfLF>$`ITcscYtOUkXF9tSPOnibAR?tjEQTkQK%hSv)}7-;sqWj5=wH
z)#1F5aY(Zz$3*g;H2?7SYw!o=>BwqTvtDiZ#m|1h=bwGT?)6*tPe(pHj?Bs6b%km>
zoKlP-G5JKd-WE@^b68alu%CSL3T-Xp`9$B;RNC-i90}1Qr6whbsSVl+j51{aGENw6
z==+ZS>C9?##pdRkrf!PYxKhmH$oVug4ri=Vxaiq#w$yc9R;W^;f<Yi@+LqI)ym!Bw
z(w$EyhG8VeOb8MmW{&5P{^}ZSD)#$5hy59|?O1%oHXULa@S`WYnIO4Lqrn(ctRPSd
zYnME92I58MrPQThq19zlSBjX@)+nio*%xi1DBGTVxzw84;HL#8N{hIxTg&fgme5uW
zbyJgap+K-KxaersJ=)fwia|$MNy-sIL<CPOEdS&m{ytUH^QV9IQ;z4$vm41cIi!HL
zZIP9s-mKZ)Tp^6cPYWneb<H@8zt(r^H#*c^zkJ2;@Rl6m;q@C{{?XqfW`TcrL$`g&
z>F_|65=m<wp7y+X^FVDC&#yLo^rM%Yhk>qdA$Sh&?onFcb_b-=RF$DYF=`l&Gq!7~
z8iQ4m!)}Mt&v1tW=cfm*uC96Y;u)u>N9xTjHhb>cTkhY#McI~Vd}7ts<(Jjgm^NML
z_yVJkRCTE-kLM%GHoSiSKwDMZY}W{psM`*iBTd^7d|(_-2ogC3=HW>23)<H7T@Se>
zWPwT<VOmlgsT;*IoykViUfnWI6Z>y}%HkdM=9X2z0o@dwfS<8-i@kYH(31V`k*aPl
z5LmEvTMPg>k=)GFw_mbT-w=GD-&|8w70YnOGz}>WbWSuvlrc4+MJ~c*s}Qp0>G6p~
zWOIE}6j`pU1Bb&IRaZ1ki`9xS&j=+;c10WJ(}`?r^lC*CP-%e{rE#^yKu!TA6Ivv$
ztSYeDANVPtUs2kkF%n`R3y)H&ycFgc?>);hF`g&<;&6+@R)+0nMHHS81KvBtr9{Cp
zGS4F!=qig*0%<L^lErM`1AcLMC=Yb0%Z<t{jK>FLPAua<atl5LJYc6&;Si=+%ofYU
zxXj3TqV87+t&uU|7l#m%uIuoNXE>cpspNXY_4PGX-IW`5@Caq-ude9VSB$5D>3G0e
zOVe2Dx*XP#74zcCDpyFP(x^Bgj4r&crbAR+xiKoi;zqJEC|jXTi#D1_M5aj0nV22k
zhms%5`2rU;C?ksoEk?9Z#O#qqqjXJ33W320iHIH%0<~4Fx(#*HV5<tPi?mxCgX*ua
zo7?jHblNkYpBR^goCMP}aylMKDWSB&yTCFpC|Pc*!WbkHYidl}(KZe9>BP9-lfwLi
zzi$9qC6YTcK7EZ_4)o0xw%cO+2Dt<_+ZDUv%rehZbwl45bk5`Z_Y998aEEJF>lL=?
zN~16dwq}9-d%pbBKjRm__`uywOVTI0RlvJQ-8IzP9w`OhM;1SG+)bb@rmeWWd&zP>
zaypE})h#bxeoXVrzu+Hyw&KS>{@?I_{O|uIUw-=*WJYVrVSnHkzYwfeJ)6GgcDqJo
z!R|D29411FD5J=-GzFJwX8*Y3M=x$^D#J88R{e^@VbAOqL=w0t(Fy9hDDG^nQ6kZF
z6~=5JEJ$OB%LGDF_ZzC~J3jflU-04eJ?~z>!?_tH1&7nX+x^JdXIzLFCAev8o^_Vi
zXeuL-Bur&FK1II%>K!R&R<-5!#U1_iO+f`UHQE|t-Qnhe7#8%!lv8!r#UtxZC?si_
zN;^Jfa+px2BNNcNL8OE+^#yIBFN{WTbH;nYIGkwO4YsP8gXdT8ANlyMq2By(->?zz
z*(V?IqmMr2XFvTZpM3m+^ZCs7>KRSDF0Sp6S#8&}eaFLx_pG}fah3Qm?YX<&65Yh#
z{hg1wySwI({_v0Z=>RuD$iB~h&ii+7`RwCQijO)to`wlG2lnR?L}Jy~th-V;8<&N)
zsoAVMX1{QAy`icqlt`@G8n?{2abO(J)Ro1^jIJ%jKyV8&McTF@PsInj%nQyrlv1P=
z8OM>+Y2bVw=(>*GVPKjas<Ldap7G)R1LyOOx?T|?gapdgNCA483Bi$*N0gI($u;L#
z{;sm5P~l3hREWZ4EEmHFJ|Pt-rO=fnMnPo06eB3WW+~}bJs}3BabOxJMCAL6J{LfL
z)!6dv4uPCX*(mvdxNOKw-&0qmu$5v!q==+2A+yZ1wd5cFSHH(<v*KU;i+{^BIie8E
zKGCmQOxsZRJ-Ta=>lOXg4I&Aq@knhAqF?`d&#>R*3;5H2`)~O8^Ur8M`HaV(|BMte
zE3Mf(M|X8iJR83J^Plnu|LBkDZ(i`(_BoXXEfxDWZ|HBXu{klHPi(F?tX_P?>9EIY
z!{Pn|`}=!Vn>AgxMXc7O>5N|{Ok0CexM?UbL1p3iND2$*rw??S4dY=?8b{tgKCylN
z0#UD+<_S@)Ng?y%<x9@T6Z`k?SVCsKYDmGC(4)*HuQ82iV_?2K?3RW0>Xv?cg)x$T
zy=FR(q!<fY!B~)nBor|Vk_$MW=&d5f$O5#D!8R3co^kU;wT9`qV>tc-w@k<wSZ%IY
ztvA$lg)$Y>`N;k50ONuF=9a3mg_X9tqFQCfX~BArSO#(|2OD85%l^RKwy><!MJ)gL
z;Vnw2!YGQF{reB-^_oq0gNPOJ;SEo3-!r;EyY2~op{iSu3LPRfnKTY$KO@OB)}S?{
z`3$x~=ECIj&Y`Ube?Ulu6+e9a>b=J;;DTpamclqP2HP}b6sPl<aU5CZiOKn*vdNi-
z2D>ciN*9g>3GX9uI3pKFWsA;6sDMyyX#z^gyi82f$Z#G@EMD4zR0$s2Dx{x~0u+HP
zGTIn&ip=u}DN>o@v`kX6qzp@7bURWAq*yq`?YgJwI<zjiIgOyd98glCm?98ZS&$-1
zmJ){+rRf`E#9i`ahC)>rvcOoC)}-aKl2Zy{3|UA-TnKJ1Tu5uswn8gVIRckM<VB-b
zUi1M;rfX}uw!I(#Aoy}?HCRFxEUG3~4U<p=H<70aL1MdJmnY(7@{v<2d_xpc)@0q0
zqhnqYDL_^C=(?x!4pVoyWdh0zU+5XB4NbqIs%uofcqm08k9%qYu`x{JZ)Z|B-v5H8
zv1GqcH#bCSP-canPPo$!5htE)G-2-{&bT4a))JdNdGH+HO&Ix{@wnsq>Wa1hK>X#O
z^6<@jcK2tdCG(eGpZMFiBftB_Evvd>oea~oXBsBX=Y=s!l5RNpgv*KldUHe5cHIB+
zk^kHO<FEMN{&)WrV>;$x$JM7j|GWS0pYnhBKm89J#sMi{agL{_l1bjISF}d)$*Wr?
z=lJ=rKCmC=0v?vKK$ho;92&Ok62Wlu!ufb0AgC^CMJEJTS8HYhbyLxAyP|Ufq>>1;
zDTn;w0Dd9P4!8S&ZfZVywc**bmg8aM!~Kb;A@Z{iXO6DCIXAVXGm=$PSYZGuCXV|P
zIhRsHx9Parc4XA|&fgpZD$G<$Fy&12W66GAiXg@`ke8xgthPWPMOUIYZbq7lIG;(0
zvdIzO0bxB#Rn$#KRn?5=Q(@zo7VBqzw^^}Tt(d1D4p?t1%guJf;c(>krpC=DthFru
z%yzTI=x1c9F}6bEX)R2{BSIwpSO3jF;rD*`$GrLGD`r3O$AA1s#NfGIYkvRpzenG8
z{QT!%@%g8B{PES6Z@&JPtSYn?)F!j41wyxUtzy-+rC~1C)Rn@yiK?v$(+M{%Xpw1Z
z!#tkRU5iqR<6*~odsRwlV`055)66goB{!!uyYa+uo*2d%Q`MY?k^TNiOmNk$`1-4_
zd3@S&d%NX)9%(8=%!0{9R#l6xXObI<i$^3%BVVhs$w*LD-DQHGNEeM+flG&Ct1$LI
zDAZymT?$oFYa*H89OLPTA4;Jv_)s#)As|FK)}>J71yyB`MX^I|>p}$3-+{2pW~k~`
zn9KV*rwB0=9i>)<aTf%C@<%^m)vfq9|L(8Y?Jw`krp2}$(il`#A+^D_EjebCp9(is
z{%0sk{KcRDIos<iW;dg%hF|{mUm%AAtG1?F_n0tJRg#CdZ+Z5~k6EuQ%ZG27vgYmU
zcYNNhaIGO}OPWRwUw>JYGV5!)uEmXK(mbJThtw7IR&v+{y0$8ZFC}SegK!hmyl^@U
z9PamQKK_{9{SiNpoX=+_;N!aGJO_}Hd5+xwPydcpZ+QOkC)|GeW5$Qq_z;<nN49;3
zpBI*4pt&x$&JYsLMRYYl@(hQ8<71IM>x)V%Eei-(LZ)tdqMsR#J7!UuZ0oBV=GhUv
zV;asRC2{kF8_$>KucU5j1e8fcE{F^t!P}l%$<p+!n!+0sjxZh(ej>(<NQo$GnrAPO
z=LcT>|JwVrU(2)fJQI7?_#5|d=9qG-fh<;$5=~K(8#d4e8ff^!*Zx24n}z|w7q;87
z<(5b_ha#)U%E~D+BF?a<-@GPYtX*YT-Dp)RsYZi19}qVQL?GkD+H0@x``-6?KfT2k
zHA{>vV}}nDUYZ=AHBD684eOgrs`ZA;Rn5g_!-GmpRmruHz<KSZOUmsvbi1svl7V)+
z&137}AS}eB0ZB?Sr!%-YXP!zySSH5t`%mhk$a#Dz1)@+Skr6>LWHy{KhN7xS3C1}y
zgsin_tuw&Y=fnjiKnaZ$8PFNdh>mF-VVTi&gDy&B%9xQD@(R{3$L#)$5hXHyq;6Lf
zWsOkh*fJNh@kUCNwJc~vv!$%JXln`Qb*vvp+Oj|tB~j+Mr@E>#(7Gz}7okwdqQX>d
z&TU;BNn|IW6B=C>s4^#$j9r%}Tv8&-25U31J;apF2}vP+1mCmlKQNqj$mkGSlZ4Jt
z&zOjA$=S9_l4PRN8b43q7i7w;C}Ark0+}LK38bn3flnD35JZ_DZO-Fd#s*|Yl?EXY
zDzl>GS;HE;UCxG#xy!XhLAAZgC;_E0Q%?*FHbU6~(o(f8GQ;(dc#?M{KcgkY+OmY-
zjkf+DD|dQAIZL8In<|@N&TF5lU2#{f*z_G2#U4|Y1f_{FJO0O0&&BN>L$^oxnQARj
z!z06SVC=Hv@9O3X*-bol6F>iA&)ufs#puym^6=F2z8~<aL^g{1-qHKW-Mb?{`O(KL
zG2!Qh{xov)`XkJT14e6p@3Tw(<R`EASO4~JiNNeVQY4&noQ@r-u6gz1j=STDZpe{#
zwyCk&;C!I146CN%@$rd5OUk-tnFe%GBT{0XXC`kkjYYI2Wm})Mt)PkmNC>46szA3_
zgnrL*yk|Nd&YHFS<!7Yi(`(DC%Qt*{Rr25bySMz@XGFxRG)Mu?2l`>6>#{N9_G-(`
z^#&p|+cnk-reR@!zo(iU<?PWZ(bz4MP(;9o1(_<Go3g{(C;ZY=R85AFCvuA+YfQ67
z*%}d+GdNrm{X_~gNuA@b4B4nO?@>b1)Geey^xwacFedPD|0P-%Y}PHlP#m5PTwGt|
z^c-1Yn-wU*w7ciU>r1NDKjY=?OU$Z7%oAx|lk^2$cVM1o=6U4(hqn|Tf6C_if|nnC
zgqx0BZ$9Pj&1cl*isR!4(lYYuqmS9`cYr54$F?!dej>U=^a9A!ri)F>u{&~j8d#>D
z%j;{(s>l=uAvv8+ltqDaK4-ZrMf4tPEl<lH=L5%Mhf*@HHe+C#MvkYBx~zG6e4?n<
z%!}j0!z0~`4Xb8_&^Bv*gGc3HYeEdDlqicT4|{yb<^`F->d8k060HpgGP_S|l*)03
z`K=~MDI2Uri1=Z~G3N#vK{`L6QjF)OG*A0N3VEm_<++*3rmBp-Dp9t`F=nR7L!T5#
z-eHkQtteZANAc5t_WRtte#Ngo|B_$);&W0~uoF|t(D7-+y9tpbF$VmyF#nE3iSP8N
z+iV*C{%`&|M-5q>*N;Vk(l9)}XPRd2KK+FL@rnNJ-%-AJ#fOI{x_M@Exn_8JORcY%
zf@A;wiHFA{tIIX7wdH!-BCjuTLk|j~?^s>FU~~O~&^=<Ur6b{^r!g>ef$NVyB`q_f
zZK+@1;O7DJ^fk87DB-!fzC_j=zIgj3NhH>nSJc%R2)n&STqNAE$2o`d9%BvT{xQ2B
zMRrQg{Rtl;{V*a@;;O7jXZWBXJI}O;%-sXV7>qR}a=`oS9r%C`k#cj5kcs{MTY~eL
z(qeQ;dvQzXjx1fr6!X-}C__<djA*l#F9!THWXG&BNSVKMhjGB|p4iMIwzNDRI&7J8
zhmbr{S(@!OL!TFi6AG<iJ|4OM_0Q4TP_3@G`@#R3r?38warc4f1G>FN*cRvBqhtZ(
zhtx84*rLGAGvkym5M>O@JRt>irz6_tF2((h_Ur$tvPFTdDy9@L>k3o0EK^UMXH+6*
zK$cZr8J88~IOU8-V-P~nk3CC>`DdsFF)WC9j+l}BmJ%*9<$&M=0zp}pgqSJfzv=wF
zj|i!<$W<$p$uThLJn~mcpJTAd$qOQD?Q~V<bdRD$mH81+G!5&vCC)Q$=nzT}Qpy>#
zQi8UW^_qCjuC3~pqPakn*;y9mk(3H%{24Bt>G^I+nW1<%<g?iyh|baN?}=f-C_%7g
z{{Kyy;B#iCJcEfL=9Ti|nEMk0o}zK6@(gO7Iz)<8RgE$QVOf~`NRVZ|#3e`>QH3Jv
zhN4|#%ND7NoOe4Manqh<+##gOmrwMhIpB`@XP`tx)fGZEgeXYoU~5FyKxW>b0j)^U
zp5e&?Wxqcc5y*z0kn?iqafU(=)5JUv_{CAJwj_~=ZlJolV(}^KaNIz3xq{G9S;H_*
z;Ci~J2l`;>ogfqivRPy6z{@*8K-Vw)`oj(>6cdW0S3I3YY$|A8d_)p8U-k;|e&+2d
za&>#jqARd1vM70Y+EJ8-k8U?8qgnhya*jghc)0f;9;rUAncYHFR$O0faRFYwdChiJ
zvv{9D)Z<9-iDcv%BA)R55swC~aT`M-Vyikgcith+uFD+bbe>N505XHLyY2yh=1k_#
zh|I3O!t&D#&5!=KFZs}4^Ji~+{^qM4yHigo6yvb4ZYy5gTtW(Tr-8fIuSiOOHFSr4
zo(S!G>Q&Er?9mF=MTxT&qF4ctQuwXuCx)51Ur<F$vH6r3XCR{M4J1iiQf4m>C&F?<
zAW+2x^Kv5PPZP_0%6dP)AaupNeE*gIyKmn^GQ4@SCB~W3Hd$^^6||cbbJtL}TR!X_
zXg~T0V+yXW98n~eab)g0Vsvb)nsId4y5vVc{1{^l2*sOE-!Kger@80jk3Yqh6>VE0
z<%Sd*9u5y&ZZDXIBZuLMvRt8~K}ttq5;5e@bD^O!S+lwLJh7k0fx2z!`jJW-o}M0w
zF%W#<=K3~6-L=O1z&MRmP0c({G>u_vF4*mR8dIWCIFXr{``wWbV!{6b7?uCKQU;e&
zcGL?&ijv?1Zk}1foY|jJkrSgcr9rAJogqe))+n9b%!K@Ra^pmp9IhWTJX?aF9Yk@~
z5ucsYLS*W=5Sb+#6GmxtS%S$pA6Zn`c7-VmPy(S9$rOm9&VvFq!P4i5x@3_yCu6+&
zWXt7E%d1x}_`ARViVshFkQ$!^%c-Xwrktde&fib*KSNB|AN=%(+`oTfJoTJ*JBk;Y
zyVtKN!^rmKE%T5C&&(6wzWu<%{f^Nsym)a#p#<04EpBn_r*r-vlJ3wmR8K4h=E?Et
zM=vSLiqS_F5pvSQGSggbIPRVZ7^ErLTv*2OgbRYZk3Qz*AN&a-CEow-|A#OB>fiAA
z?bpNv+lwnMw-tg4V@taJ$l=faA6#EwVTyufcTYU)ky4EKkTVFQ0BON5bEZhg99Yut
zpRk2SHf`4HS&IufG~P;ut}Jez@;?HCp*wQ%>JzH%3&#BeMOiZ~PY4I5-m+eA@ug)B
zg46xCkRsb!GS7RoDRN$^%XkgH%vn?~1xY5xWoGe_xxS`&^#j7wH$2^cfp<&JLRAv0
zvNe3@CgzmDR!C7}$}QvEQw$>)DX_V^Ci#&lOT_JG5CV&js7j-CV)k>^w9Scxz}9Vc
zQA&d-@)yB8JB(XU^@@S-&v5GYkI1klMvp8lRlUMmo0Bhm&g}b*$zZ+CUD~qv--2@M
zdPPbxU+yvyqt72mt+84{aCriHjtrXTOg%SRXCB)5ahPJBLMfdmrxA%REQPj+7)eu)
zN&&4ERvV(y_;F<JCZd$YGrmL@B~{(huGiF6O@G)C#y;n~=FI90`VAgLo|=Xjvwlq)
zC=4zH+!PQ-VRUoOtCSQb|Gc9y%=5_c!@wZP8HO_DsaeV{Kkp|*$}4A~B;h>0#5<gG
z8C~K7B08jXM3cFFZXEH`h+7uwb`2<Sf#61*)EJQ{n-(JsNmuB)L7p$Eq%BfwWLR)#
zr?#>-b0No`;dB6Ps5e`x)f!clI0r+2!n-A}u*ooBDfGE(hh-+`GOnxdeck}3X{4+w
zkcQyFCyQJ%ze?GX;5_x}0&VL2;1n~9cX;n9OvyO*RI5vxwq?J21h4Qfe~eI9TpNv8
zwcLe-_mQ$GIX<4)fByIAc0+w}!{@*F1(&7a<?B}{U63^Nv*W{IK;PW4USIL6cTb42
z<d<JQaOfP>s^ZUn@s_HrV9rTNb(<Zp@w|%6k#L%;+beE9{+Jhc*R+M^cs!gZ?GCN-
zS}VkWTjs2dN{V`W$@_iJ?`>;D3fZ9@LAOQz0+sPRX-LST$dm0B%$TJiDP}SI)Mwpm
z+^*1<1=_+#b-`!9_lhrej{noY`b(notkx}csS!$1RaMTPwl;&U5fm5Kh-E^T1$NyM
zk@&Gg`G5@$GYz=f!}cY?R>YWjZ%P-~q9rUnLK%e4*otAihm>`EN|%^wjUS)zVP=_n
zOtWELdWvGr>G502mBl+x@ZZ0YxZfT5{U86BZ{B^5NF$Z1u&T+RTxZ!m-se;XCFu4$
zuC8Cu)HNZ+{Jmh{_C+QyRP8#;RC<Zg<}7Qd8T*;l#hS9Ih(5C4Jt1VIu50?S$H$0}
zl6f8($DY;Z2B}~i2iEHqO|#;(-?88A&|0(JZs@uriVPAzo_dyfK@d^rjNzM`E9PlL
zDNU_4w;#PBhCuK&?RHJqIl7^zsS89BG-^vX<tf-a4pds;!-!BRPo8ywE-QlbBp*o8
zGp7+@3W}o4kExK&63Fa$g?z6WU1W`}^Z9|U^%;faLBy<~jFA`uehDmqRuXNq^d$>E
z1cZnq;8UWi3(BfSX@fd%la{f^iilBV9vqN!gJW0-p(DkJ3<}Z=-hx(?TJiD6EzRl^
z{^pm0!y#moRLY#)^`@jOO1uDTe>cGTcY4$bQLukF@akqmX%z8zU|E&Ce)EExw&Y*^
zzy2*LN`#2Ke>n2l>kaE7ljKdk;<WGSr-_JUbA3Uz%`m?JOwz!-qe`BYkyt7Ci@*Oh
zUMQ+Ub9{IrdY{LfwnDB7(&H1|!;Ye^VHr_UaCP^Rr*|*#bDv?nh2grX>HCSUKXN)A
z!8_WvU>OJIB+=TkS+5xT5m_{pNnm6m8HLgtin7cxF-qn5F0Cnwk|-3;2V#uW>uc}}
zOP|+VLE0?3*A2o{Sx6jaRB+U*H8H~B=>yaLffs-9Cv0x+5L3_i_$8(_n5HFgX60yw
zNCJ_179T-r%CdngP1W3?nhQ>EzaaDvs2I@3P&F+nfv(#e%QOYT;u&I&ehTd!G2Rn=
zM|b}f-R=>gE#u;N`<H)(QHqN~!!i*Gl+79+6Mh<zB4r(e%K+UZ4YR6ng@Ne_&zTqZ
zJ!>8ZP9djAm?lx!3Z(=}s~i@+j7X``rbH=0S=!8z^PXi6lubifw;(e_*Sis!BtqoG
zh_E=Mk}OLO>kiK0od*NDuCjJ7oZXKuJFH?#C~K)#8+1{g-A|6V3}^``4YoiC=)1h?
zl10UOv&A-RLWrb525Zjwu`bmA&#e`g2{)c{!db`{h%%aKIl?hB_(bw3l~94qeUgGj
z=58*=h@Yq20Hl05*|NY2OA-?AJ<Bu^&KeJ8%-M+xZaz;Qq|7eql&6|0Wo*RkBjYeK
zj3dqB5Vj^DNHJlpWV5|ILqavdWp>ou?=vQ-yh7>*+2o8==N!&0#CajOGtg709Q`9D
zWmzGGV;UkNd8`&h&za$rzl5Ck_~5aoK$_xvpErQA*+5Kq=kOt-{lelVrtT5bT%c7X
zq<~C`s%}s!Gxf%9kI@Z9*&vdqswyNkYI{d(D~g*}gz1>W&9$Z|YTP_>eRo6r@f%FJ
zhN9xffAWU}mw98$*fAXMaX}(%jjcEAo_gMW^_E}%`djWtNx8bDi=H_se(?H^i_4an
zpiqjx`}Nn1vt-p+1c}YMWm8mK-P|Gx^rt;drKy?*KYM&kJU#AMrje)JfniwC^@{G{
z34DtzZwZHvuuPaDJ9MRxnf$L5;vBW5gdxVv0W(d7-)Dfe0NPc1mOjj^Uff~M5`>?8
zR<Zs+-tm9=zy2B1I3eqn=Jt-;kKbevv6g2yd7W`h<B>3)80UpBdE9WKUajG@$7qE)
zoe;wwdHEVwY!JFeB+opZh;x@G)4Jr)e}j+_rD}?@g_LkZhj;Fb^%3Y|O^O1o$;hJM
z30@H6_ty=tnhjC}3K^LDnbSC6%ZZmCzhN0pG^IkzNN^*i$^1v7B!|aGhW<!hTI#ak
zY4-$Dayag>+9Htz=ZR^=EfdDT7hn86t9ng$Jn`|zuW-vmiXI3g0z!fnlFhnB^b6&7
zMbotOXUBA54a3wk&oliv@zL!|PWye%nYEhNS9k392duUPzc5V$x=h^OU1GGsYDu@>
zF--&W<XAT~>vfG^99kEo&=Wkg8bZHkv<otLa5<|OMCJ&d^K9p7#G=SaIhq&~^D<{G
zG6J_OXIX$^S=<>uuQH%sW<!N46n;^^b<8d)6Ck9NxFvwdS>j3?LhuM{sG5qT^Ex%A
zh(0F}sRUY>e5bn^Vk8l>GdmWD6j4baJbCyPG)ZKK`}L;fPyhIz@b=x2x9>i1|M<k%
zPiT3{dOn-+C%;R5r*Gb0|LO}?g`u2JtkyNjE7IJtDkafJUc9_U_9wo+-|>St7x)O2
zYYJO%8s^_R<sQGi2W5EkPyPvyUwwfsCBx*2OUkeGcFn0fadW-lxFn>l=({6x*F$^7
z2P(GJ1|<(D;rZsT{|&$XcmFpOaB+1>_x?TIJhQ&svRSXNg=QLO1cIh+XqpP=JW?jA
z)rJK4L=F-cilS~1enuH}Mt4}2u_r`N)vSmyF}cj@@gb9_qiqp|M1_DC1Iy%aDG;WS
zZ-4O@6m3oN0VyJz%NGdX*nh?DVNd+KpL6@-Kyov&Xc<L?ju735(iUA*=XkIRsS`>_
zM2Z*#OY)4nZ`uFi-=oW#a<xK~79RyxCP<FxBJH}NSR7*ZIjvwi62gov3dVlqus>35
zFWB#o>^|&i+nT$NUt&s&@-tdnobw1<A#~0U6%nK@v$%cqxY;qQ8f{w2%iHgAwEjn{
zYFeyTXPw(Zf7(-(IbteB$2|1Eo!I;)kUMzCvN(c|luea6YBuwgrXg$c@>DEq)esP4
zIupW6T+AYUYc1_+MfCC9#l_4aoabyzP#TiJ%`?O4M3_6~Wn@YUY>6&vOufqMQ2}<f
zrLYBVS&%6rW1elqm=GeKzeFN#T8Mtk5kxY(st|(UQ`VKpoIMysBu)b|I+RdMeV12Z
zGC>R=Vtyn7AW1@zk|YI>OEV!zSn_44r6L8G+=MootO$*{Iduz@KQb+exHyE;IJa=@
z2BLNpWzEI)C7Y`&>g^^^y^|u0Cr(dK*&XL2)S6^#Fg2EhoClWEW7ZO;#Ns15%_uvP
zq`~{&tXu<1<fxt~GNAJ}lP9+<aDGrM?jJ@|mqZS!9r_c}HfPL>$D>#`ZJyc#VP3FR
z!?b_ImKAreJ|QHDpN6v_ddi6&(K7Z&=4D2tNNQS^;lPkSkYb>&*B~TA_eAU^X?(&3
zkspjHVyg=6EsLA6GI6oelpkL(^fNDB-4GERyCa|e@V`O^hdevz>#E^rzxM`h3)-@v
zDlO5^j0+HRV}>6HBxacyhlOQ!5Cglzk=>zV8ngK$q=Y6h21#43@L?f}jEE3OlqAOo
zlGo*&Dg49!z;x<J&J$wJ>NHj(v_=&Lc5^{}dxy3KB07Hj+H&>gBmUi={ep+b1EQ?a
zZA)Bc!aNcC4r0uiyl95rae6!uPA9Y$G}mtkVNWy#qAHOoq4(dS^*wfV!-N6TW(3DD
zVA~74ceI<!tdE^1mf(>o;9^3YCrx4KnZ_frEb`QNcDSUO{{fENi>nLbG_o!m`b9D<
z9gA~xrz1muVzb(?xRJ6h7>0=uJjc@>ts;eroOTa<{)-20U%p~YQ<l_}C4JX}ID>tq
zqd%U&&$RUwO|gMEvn&}=6rE$}dTy?6SSH8uZ~~!;%ZzPTj3Tfs1KJ4s{z#e>r*7uO
z_L7^c3vAsG13ccp!z#gM-Ll&~(sd_VBQUn$`sNb6=jv*O^NGLut6ya%;kg-Irh&U{
zjqmonXiCyN@$mHnwJ|Kmo)A32IWPuWlsOtvfwtxOF31_KMyd0?8bAo7(!_W+PDqI|
z8qf$^U{@=oa0q19Bai2#5FsI@%oFt<TW5?@2m!llC{_)ou;<C6!>rbrrbb(Zwu+=I
zLduMaO7Xllo(X==Mhqd4+Tdb_T`QxBqQ~fpk6&#m3&G*A=bNv-;qqe3YF$H9B7Vog
z@OOIDNf9Y(OCLRYPTbvWxPSkiFTePP<FZg*-!Kd#pMCNgQy98a?h+37@3_0WrK&2v
z{roLS7QFgj{TICX>CX^C6ZUWME>N~DfBCB~Dc>Kdw_A#;p<Jz~Z(i}?mw(OZ;qIeP
zkyXohcx2dpV6$EG>#x3|JN784n7rrje)SbGMP8@CX4B%LC5jwVQ<fF&dd)JA>>oaG
z+&zJixFR#pj8WL4{O#-AhsZo03BhNRL~{xAh#U6Al$hovXUo=ml(EeF%+xEbh5o>j
zCc1H8ci0nR<mzgRIXs}WX6{Z5UB}bAZz$42Q5M*;#z#oX<k9We6A~yXuqH<}NhOE@
zm>1mp&)HmTx%|D)IQ0{!<B6yx{cvK63!C)@R6bxbCW<&SJoY%f!`lKGHQUNy$`ZZ0
z;q}EU>h?X-SbUNMG<BtL%fd2t2%$)3l`|F#gG-ti5=FbAxnA+~{u@>=Kf|<F-}Qd`
zA67g+V2#bwB(k<4EV&EHIyOm&+12EH22Td(2}`!XYpwDqT?xuMCj%&%ySl1uA$qh(
zXr(BXB_IhQQW!&7mLwsW`;jQ}6fdQS8%Hz(qZIwadk!DI#ae@tg3(W~cnV=Dni{l0
z#)x0CcwH<LT4tzoiavM!XH6#<bWx+4JB&3T$u3<IOIVl_B&|`=p~Henu!uREIzqm*
zs;Z<kigXr!&y&x@xfqEeLGTElvinu(Jh4QwIM3o7g%hacAvop)-pAh>I8;%Ql)+XC
zVY3FN)P{DmqS<aKiaO^p21igDM`Q?rghI`y9I=E%8bdM#$rMCokvd0S2_nldo~MQj
z%9=a{P%6{?g+vRR!Oc=&>N?{Flwz6xVcf6Sie=pApTCL-SyI}PMI=tAJ*(B4;jpKy
z*9@m8QkWQz1NG_xAryp&GMS&Hv_jhwRcZ<Z%Xp$*ZIQcsy4@aOHNAU82o1VGl{He=
zblp9t!vn+V!205fQddM%aCr9)V-#^ZvR-X5g(jf6x&4%J8gXN0{1xVsvR+g7*-e;Z
zhq5WGvI4ZhTFc=nr=1kR5SEGHX6mwGRTVV$3g;t_`yL?_{r-TPdXh;9on!IFzGq#p
z@(N3t{ADOrj@nvgf_F^)Nc0hH^41}QNV%<Y^C=S&mmSg3ft$JBHvCuryPxr^uXcR?
z<rBx<iFK1**-}}&bNS2N2hasR4t#ib&+VHRB$<Z`ph?P*B>8*^P{WRTwPrc5)nmEE
zN6E4TreVyDo+_Cau%^J4D^Ml#GSVLonFt^XNF(L?hW_0>(?38mos?MvZHg7m_JaD-
z6a5HytTmKn!4f>P4-{3+_Uek$Vb9d{1lKW7PgIqpDkR6D$Ji<-jKqksl4Us&+(c~!
zcW+)Z^qI*TLnOMy{^3Aj>pZb5$~?6Vo_-kEKRr>LQ;ve0xV%_1Ew?-!j~G*7w4y2s
zn)QkgA0FsVdrDhkjY0ySefBBVY95|;JiY%wRTelYc=houKl{T^`S5;6-<|MjX1(4}
z%fRZgq}7RTcVvHfzzRX7ZJwN^g<@?nMV0S@BoNAQJe_iMoXVzx5HqU86eY?O__Og%
zo>O9GA3znTs%0GZI1*9h9A05G##o|{D3dsyju|*CMJ7TtC9=#?fG~P=b`larhM_wN
z3P|1&LN;>=V#b~zkY^Zv@F@fJ(Il6(rqOi>8@aw%@q^DkVt+W%cLU?p^CSgz^`Bw=
z`uP5lxhlBa)DR{nS5t4deEG2Bu<L0bk6f=xtghH@w^T*R-~H8J^Vvst9QH><T#zDh
zb#uY%AO4W&7KXzCcbZsjDy}ZKd^t^g`SzZVi-Pgg6ZIn>-rrMMjb3j#J$zu9e$8@t
zqDd})xeHirS6sY!Lw`83ZcLs`8NuQGTMBWDt~U5(VVOqW9rkAubxm@8h6XE*R{0A(
zdWTzl<}?j6t965q9x)8q#nV_rVGYwVGc5~q66`;Gfg~}Wj<j`6p%kb66REO5%o>Q0
z0juVsVDX;K)g>~KLwpOvvdnlP(WaofevJr;<^FS&sW39(+<ew42%4%y*A<mjP&6o2
zFm+#G;)2B?>jHcA1LAmKI_@}~j!e_U*ykW+V*_Q`GVGo(lSG=5!-sG2DbdZ5%c};Z
z1j8`k;)Jz1$zd8Nl4?+!YxFX*^iRZ5AZu{_gfN!q1H;q(ce>pE$0{r%(d2chOo`|c
zi%3YJ5JFRw4IzOK5g!663T#mje4fH;`y1WW3?X8W*<pmpD`{I~XKa}8KI`iykqF77
zB^V=6x**Ih>k$i!j2aOKqMMof5#f5)>l)jv@O6Vsie+*rqe=K|MhKoX9|=l82xz6U
zgEoQ|iaB|LE->u|ySc?~?<kc(&I58hvGhk&ba<~o2r8v9GGTR%s9HRVA~@PY(O7{<
z9+G969b@u2$|*#Yv8XdcFr`S6I)jsaz{e%K$8?^iE1gB~s!SxQk=Fj!Qm>Rn8V$jb
zmYK9nr0GPAj*!kw!D+;Yj7pKg5vK!gJhFs<G-X~lhXrXA%B*t4SxVXcB{in70C;zn
zTo{dyi4-E4m7F_qA!N>IHQ#>~1~4B#AVuP0T~cpu5EzE;z^17Y+Mt!CC@Pj^qNr@f
zv>1a>Cg)0$5K>b%YeEVH=Q*8DNG)lq7C(#>#fr3s>hh9#?69^X#zeioWFC90u|!=H
z-2_tNrw(H+Z4>ZIzIVNwP$D491}PI^K5%;afNj=jp_%$aejG>(Xxx%x#S%ICO2oX9
zJM22%ynM;3Z8#l{9J`V0s})sY@xh~If&gpj>E3>cOAezo^~Dy6W?4L1M1qLeI#c?E
zGMM#-Vza_c3vLO7c}6!C#nlGe)+m!vPozK`4`=j<&#{nF^XYBHy8VcM{onm9<4Mz$
znzG6fS5iuX6d1M2*~3jqRhNv9_nBM}Bi-W%gcNLEyhdw_pH38hM(G3B>n*-{!P|G=
zuz&x=`o}j&0tiECRmM{+3lYQAcW7fN+bdFVDAl5M#ddSS{QmKKpSPlH(57Y>2bP$)
zxqU-bSoR;jqOP|Tt2OiPJ?pB#RI998%N-dZDzvs-Y%1=quUXBJCFXP$Q{?;2`<bG=
zqSO)@94dkLjud8GoVnW82w{<;WFC7cBE1VFnXr{%jtc_GIQ3MuVck|p31zWnv)-VL
z<-@n{D9VZ-e)=iXG*LD!hr<zNG-X**6$Rrkvt75Ov><ij#jV9nH4hK(_`#c(Y+t@+
zJRYb;!1o8nX=0fJPvgSHRmJV~1qwk3g8i;1xQGjx@u;oMpGA>KVaXu+#j(A+CTfv0
zUA;%6m_tHUEoECnvIHlX#|76fS??(o+Gg<dJk1aVamkHub+w|}tnz3PfwmTDa?`C+
z%<gGrkiijtWA4%lf9hF=iE6V(G*@U-;krE$2qEV0A*FN9EdrGmK7F;ps-N-mzxk54
z@7~g%2ITMfT>DOsx>ujRWZHL}P9qns;kZ9#u3S}+d|*==8Y39b!NG@Pht`TQBpy!_
zpImIw+Vb^RUvc~O-*fTeWzHEDn(LPzaeVlOYSnV56{p>fd3FS$F<VV_x8ma;e1vzN
z&;Rnz5q?HD1%=Vv-Ck2%+;I8w6CS_%f?7CKq4R58OQgzZi5LUAP-vCcnMF~sy1b>^
zJ&=T82@dbG!)5WFadP+=>8BY8$dHIBqKu;1Y;e=S-1WF+CTT_8wwSVDnMX|3kXDx&
z!!aJS^R_H0@+^b0-C%A0yDMYq`aK~fN?S3^Gv(MbA0Cj)Old7?@hEF4QY0;XMxCTY
zX<LpIoV<YKY1@{fK4;6$Qx<lwFH!A=dF*LCn4-iGAuNQ-5RAo?HFB7c2sW!4<0rHb
z9FGUevY=j7glWPh&--8flGWuCtGdD{iBJ~rW_%b4Zo!Qs$Nla*UvU3nC1sEzPuW8B
z_~7tMq9~f*a@l<FgkhlX^4hel%B*>`hPtWHTHz+24Gq!ZL$=^YrHO!50!TQ&5JE<p
zC=x;@lmz?Rc$u6sQJV^{6|vjl#uK%La<eAcnn9O{sF;oe(>$ThnhRp2s#|PxL0Pw1
z6RIt0o-jp)jDcDkbh*mEjHM6}WJK$TEkJmf4<p?vi-&C~357+OmV`woM_np(9Em;)
zhP}`FGo(Z+g)s)DBtlqBjKrAEBH#?tOwpgSYz_9zm~$~ZVw3X-tuhol24ob(=)o^c
z$2~HIoIx2A-Uqz**-%oi5iUcZqx0ZK;xdqwMoEQHDn~iVEL7LFK-nVy8S@Mzw92fq
z<N}eL#Nm8i>1kum=#lTQ;RHB#NAB+KD9akZ%*5m=tD1T0vb#<~k~x4ogupO%)b)ya
zJW<yT`-dZe1je3MKMTePcDo0rgQhfwa<xV^EkbIR@tDte*|5wL$vfJ1#c5nvUtZ>a
zgIVLI0j(@$-LSrHN!}6Tgh-xpwI%uml@`|97fj27)S9x;_~F3N?I~)-vKS^GNO+tJ
zTwbnu_4*BU)iU-yjm_JPX<o2Kuv#}*8EMy-q~XN=@I==SeEjAm!@(hCpmqTza^zTb
z1{vF`r6>(c5X|u${kC{gX(_iW6nW(&Y?JXieqy|Tq<--cD1{paY*})3-S98}`5*9q
z|8M`Cw_lz3gWvxc5h6uh5f?`oJC@^#wrzR;)mxY+ZhrVFk6*t<#z48*<jMQ8WJFV;
zGN^VMvGSgeUcREx7Of1Y_g`bmmLwGZ><SOdLJ$(64O4&0+YV(3&ZDb>)n)s=&zq2%
zx-HMho`#~@pahId<l^dr0N5hC{FSjNRZzDDZL>ki61N;tcs!cT)t1FAoDO>~*BhkH
zMwsi{J9H9s`?shPqRa9DtrAyP4FOGe^5~+;({vw5Nl+Fow%l?&?UANL=mJ|RK6~+o
zy4hqNob#+VEyw*4f<}sL+In$!$NTpW)K>-bR8eS!l8N<ZLtQmIy}xH&8vgW;f5zqZ
zlG75I21VJlOd_#+c*NE#j)(Wee&SNuvvCaE66lU6R;vapax|px7NU1(Es06cY_?ca
zG8}r;8M)$pPSv=5-7rrB(r0(6DQl+VlqD|++>+BrRACW|V~(CuOKedfQvN&VdoWWl
zLdt^%Z7QPFBn(m}l!CMjOoL-NE>znRFKwn#Y&PHm$A<^x<XBzZV45{amRaBE7C!yq
zOX_ur{`Zo<`isA1{2kra-{~gYefyZRFz1DJdx5nTx^B7IY*@EF($7q%j^k-KgLws$
z6V$rl@=o#L;ecry-hcVX&;R_-_~eHVRO^<w4)nu`Wm$+JU`;_?*WBK0`F0n0^H2W^
zYLhS$rg7wxAO8^HCd4w~7RUDPhRvHFuuK!)%`{inBq7+`+|eD5EGc6kq!i5Fqm>~<
z7?+V^NhpyRx+4fhd;Nki5116Wy}RLf%m7SN6~wS){gL;$vlzW@8>CtjTtF8E&DDx@
z+#!{q7Mgi-v@c##Z7#7=FzxSiaPN}$>$)rnLJ^fAxEcIH*B$BazojTNWmO@`e>Ndy
zW63b!gvkNv0#dOeYRkoo7YJc+KH;KZIXx2H%xOPTZnoUM{+RA~!c8MGC1g>eZ9(0v
z@FAe4{Gl)+lExE-EcuNpYn~T;NO+;SyuHJ!j08!hL@1S`#Z<yqD~ft^4*LFnRd22+
zjYPzSG)I)oCIumKNOTf{5VGze#T+B0l)~7C(iW7)X52!`j{rB1S+pu8Nl3<df)FUn
zf}$+(-V;KgC@YGxLK{Qy3-dJKeCEI@GMKvSI{Yv~Wss#JM!`5ubjN|mZ{B7%VPR0l
zP?j~T^%c$K9qr{Ux~P#NJ7l}vTgLu`h=DYZEd4-=3%axvI$^|ukPA{rnlvzv6G3Ub
zS)p~6ANM}HQxOVbTSx=Tbi(<7Qi{T83S)A9tj?M|ld`oxXQn0!lY_3M%3MTYv+*Dd
zBl9wm+=7x4X);$%Xqh8-Oa)RfE|Fy!h|8QYKSi0*BTD5IV3~#VLP?fo!YvEVdln?t
z8VYNPXHl{(3(yMC2$hW!(L3Cn1=A_zIIkE2N$1sU{0DF%0_d}&`QhC+)a@EmuNnFy
zwx}4650rJDUSD2qI2|&=r)^uLg86jJ@Kh;T<{YOag`{rQlyys-dXmhNfn}Mog~5sm
z;)vB3#E{cGmT6$<vL0$W9Z^akWTI}jbp44+3sTvVe1-4saekrM-sTB{G)NWMZmu#}
z-`2zkZg`-mu4&5}sVtAj18(Z5S1aPO;N3u3T3)o<oTs`>%zokSW=nf<#nRV^t1HsO
zBfTl8S6fcI2MSZ5t!11Cj_)5pNaS%Og^A@f6M|wM99mi$T`=_%Mr+LG7SY@xq-VMx
zIQfdQz6RGL3W;+8TPto}-SNNwKm6DH#lQdSJjD|V=K^k=5YvRxhOz5-`}Toqd&MW8
z{V{J32|xAN)fGYJ=%*+W(|$)<z2xHN4RP3Eea|O9{4p`X&D~25rzgJs{BP->KCrpC
zqg-teDhIl!kTE*bI1q*-M(TeMPF>m72$2}anHMi#;i6z(ChE%@RJ&pxI+U%L-GYoz
zNW<px79S$*JagJPz!Oq712n6a$HylGlHe1Ump6>VgjNBU67A*=DN3f}j>r4I$&G=~
zBqa6fl9)VqA73&J9nM2k71M4<y}sbs50qtzE(&Z}aXKETOG|$|vftlRXoa<gX*_bV
z+0fRnK?^Q7E#uJVNXO_o?RRv?6VoYjadW}V-8I9z2VTB@$^EzQxO?>mx4vW^PHZq7
z4hOz|*b$e&-RqAql_AZMXbMzSqYW(6L|s&8#E2nj#nXOI7_%LG@{T12E><<e>3~ZC
zqd+4FXo8fu2q7et%y)^j3J07cDTpbuOf%7EQ&C|pvQS7<p{fno70MWp66I#%+>^ZG
z)CH>M5_9niRo2AdKom$JSzk%U(~%Is>H<-(fP_tvzU#Psam#;s`HIzM%g=xQ%irw|
z`z{CY)Q>DaQYcLdfe-IK(6$Y(Ti8E5Vzgwvs%WHSaRJ*@IqKz`uMr7W*VpLU^6AGn
zRGSsDXjwHi^LXUTUww{IlJR(=DlJV@p*Zl-&;B`e+aTP)&>!gzPx&z<H9{G*PHb;p
zV5H!<yU%M?W6?!P7zUPMz>AdgE|sCI8|LEytu#ehG4(r?F=(an-ZLF{XeC&t;jA5T
z#0W2M-muIg0s%fdqij)7v@P}ef@v5S#|h;F$K4~rXkxjhXfG)1ib6nIW|nzkUKFa#
z>>NKkX78~gzx=16qcoC4BE-bvJor6&%4;$~4gntz`#i2q32Qtbef%-?ddu|m$ont8
zz}GjheU0D0$4?W>I56yYND^gXa1KH`3(#Z0ja_yPDTNde3d<Y<wk%m9xR~8O>rIQb
zhJGBfRzXUn*352U@&enml%>X(-+ywFO#<W<Tgot1mn7CiM9jaObEM>{n+jW&)OCYW
z8l!X;yE=b<=ov(^za5<mnYDP+d_pdsL=KcM${MNjs&ANPmSLjnPI(0xe4apg4<aXg
z1fw}&2@5<OkL=!mVA$=j))0vl)?|mDP{?^fEekSb#v$W?Kc1Kl52UmZQ^L<PY3xu^
zQd@(HF6XjBp3<Qys!MdUMidnyB}ACeNuZ?xZSc#2k69$F&SGRuqR<*!l_Zf+A>iCX
zH;=>+(Mln8K{6%LHn_4P5%XIyI%1b|SaAyhMNuk@X;8W%2uWH7{M><;NI~xQUCvh&
z<Q2Nsnve|1`3zM~=LokLQ0KhQ(K}*FG@C8TRD>lXVU{r`9eD3B)}l@3p4r0WX{`ST
z5I6ysS9hp5QPwMztpSA_mz*DJDwbtnwQBHg#z;roYzbkeC|7t}66cwVt6QR<S;h|M
zN9xs0o~)%rm^)C_Z#6z{iO4jtlqZ&Pp{N^_u{6nKs=bVvYn4o%lo&^F6GBRMkM~qs
zAbo-~v$@K7+$Q+!J{8$eAZ>$;CO3~UU`t6|SmtP1t+)6&H>>Sti+3|&?lHEdF_x;m
zAdDT8kWAyi^~)PBZ*Ngr^JE1}2n=5Et6zS_-+%r!Kl=0y{&-|L?Wwe8vuQc?E`MGw
zHE9-HuD6JKLsI6q=#*nOaS_SdQ)%fRp3wb7`{FgGDEO!U@}KkP|IdHJ>=vxD2wNcB
z#IiVchmK;sAza+?=Rg0F|My?Kg%J7uLGi;Ef$Q51ZPD=bH1PGWf5rdizq}+YGp;{?
zi&PgkNEDmZCA+G@*9~@cNr)b~EL7#1xj!O`5@dF+yWl{6f5)}biv97Cd77xICxkXQ
zH&Jc2=qgWdqss|F<1#UK85})L?#$8@C~1ii1c`6H{xa)pmCX8KZE4GvvFli`tE|tP
zPlPyvC^+r&dRfXf$|#6H3`956v=_*7Lr8aYhaF8_<<ujQh5YZo{f0L$Uos9|9&l9c
zd9q!C2+YesQ*21VQ5TA7o@ttf`RR$<H=i<hiT&;^$J2qjvKR!n*OxSP&4OUPU8Ad(
zX1(I#a?Q{!P}SL#FilK;;c<6ly=_@-uCUrL3?rs082TgAIAVm%P<o?q(}J=U@85ol
zQe>F<vT)k%&?Ib8F?D0kq-|@0OUxeD>lI2BEVDxb>a4rAg(M&{i?dpjRNfLPm2lmG
z@zfz><kT<NtA<IJ5E5dV5W$m-CL%Fa%Q7$M(*c=0&PR-?v8Dkub*uQ(fB7c}^Pge;
zs;$A45<f1iOUv3SKL4w~<K>$xrUb$kEIzVyJ<Vp#>h3j9Z@*=lcU)X<DK{IE5;X0=
zhi~7Krj9p%^e?Di{E*G)u_|$KIU?qPx>=#OH%w75zyF3+yCMjUOF3C!oMxKB5ZuD?
z{w?0kJl@|EVx)cX5>=KtW{AY&`)}E9*Q6=phYq0&y8R9z4@}EUvu>%10;?r{8KG<_
z>ovRG9;xzpT%?4IIWNvAnIWknQCdY^7fdc<qGZ1RlFhnguq&RPj+wZ<Y$#v6LDw5h
zObA=kY_6&NiQ(xTUp$VKm7%=(h*}96F>yNXnSDYR6-uP@3eaP;#gq+kUNA~x$})p7
zk2@H<d^n87_xCLGi~v^a4W=+is|haWBf2FE|7~G$)4<XlDC!ne0j9{&Il(iJQ+6~G
zb53PlaN6&4HmcIp?PXqzc27)mcBB@5msf4yv%38wx~(B(Ce%F52)U4&6-Ee%K5HGd
z#tKbgGqXr*eTGarQb<q~8B-F^5M8n(Nfw&4WYovv^9o5@OHtQ|Bys)1v?PMiB=1R!
z2L#HN*sB#$S{7vp%gplh!0>Qn?mBE^smhwF-B4X`Y1$1w2=3o~#qQmg)I|YE#$jat
zbk8)8C@U$-0<9GxL>?a>S;m1Bd`_Ow`N6GAgDPxBco<6x2`OfbB!f9q$lZR_IlDG4
zDDP3tDw~h2K}1jN4+s*a(ad2Y`8gXoY>iVz)^B)+NICLt^p3gf(M3(D1<pm3mjr8x
zLJ&<&^yH+ExDc0x;3rh@Xr)M6XT(CtX&$<$P`bd4T^0!^fnOFv@I;`jE9^OR-T6ck
zAsZ)@MGApPfh0mUQA7#<Kt_qGuCXa$+dI5lP{xv)iK42Qhe&W8LKGBb&3rmh)R%ZS
zF-|kml<1Hn2mLbVbb=)^FLPdxnhfQAxc@+DpgRUsb%{$d>#DkmczVJVHAKbIcNwD)
zQs$*8P1CNK`yKOXhmA8vw-kkCnHQp;iPMp?StFFjFAGvCuHO8Zlmxb{axU><M~$K=
zEGalvn=9(}0`Gb>24l9w80q^xB4k9%<;#~0rvp9)%E~a$J(sU;8IK2qvi$MC_)~7~
zUgT-w{)x~ZS(O%HB-drb{kwbKA4YEPTGmCw93mzL&<*nD6Hf1c&eMl`s?9aFUJ*jc
z7hinIAAP*REe?n0<&S>IumAGTDK9Rl?TT4NTuAsyGMzfUdKmbt|L$LN97me6K|M|!
zo#3A~Tg>H#!}tXcA0D~?`(JYL@n;CT;&lIkad+VAC;t>76c^Wb7?}t{Qr0aZ<)*%O
zGxIQT_u@6t9dJo<8jj!ly!EG^zUz=e@bKYVRu?O@HfU>Dd}6=bQ!o({Ov^&s7Cb)W
ziKsON?ZpjcSuyq#!`vgZp*97}GNV;?PutS4OyqOra|1sPCz|yQ-Q#;i6ijJA+M2R#
zczpPlQVYsL(76G>Oh_UrIKKSVSG@lCV>a6xt}iZ_#{nfImlv1R>n+YLG)+wiiS?$Y
zC=E~d@31<@4C+u(Hx=s_uc?YF`loA@kQ7M9{ee;mj=M*a%MElljJ$sRf+Q20iyNA&
zD@cjQ5BGG#z`CjEM@Ms6vU47k=HZ)fIX*rhgy7hBTx_>AMMc@HxZfT4_KU9(<hw&A
zSeB6&w^vNFqn{jeh&(+!(X<Vlc7-@^)8=_WBanr`lm$U)v@TFZOAwaAl!P!d96n%{
z2^D9K$BDyf<T?qYDsam{DHEe}h;cyZDmRb)fe*j>8``T&bkQK{HBsjTqTnP#8h-NA
zpZrc|>N`E^KKkSp*Xx?oZckn3Ro4`=WAysv9Zg-KoWmv9+`Z=Rr~eF7{vGe%zDG|B
z{rv-?Y%o&t^5rdF7Rai>BT3VU4-<ljRT_JF$L7rsAVsF3!@C)4G-cheOuwN%`>cTo
z9$^Zu-+V@YI<kAXr`kqzxxy_I)@mvv5W2)(yv#)Dc_yZan+1y(w3`OzJZ%s<XJ#6s
zX*U~6o9*#=#uFI3f$7XEliG4T92ie$*Iuz^QHu9peZ}eVk@d|ro9#75TS$WGbRxzF
zPG9^D%Fjpw!b{lx2&D~1DLjh6iKY;2H)|Au?s%k7KyyJTYy9p#<I@B6#SP8%3r-=y
zafiTVu9wj{GqoRJPJ~6Vxh*lOWbE@}LCc(&;Dw;vY%#XPC&}@!Cj_5YbxL8*s1S;p
zP_7AH;Ku{a#TBaBGVMMv4Fg6S%A#NyCVcq*k2<3iS}Xc#q(2=AAyC$tdnl#+Z8o0}
z*&Go4f_L+8Q6cj<AcfAHL22<U#Oa8N9uWfbII>JL*or6wX_}~-M0HM-2*D#!)TY2i
zLDV`2(qFxRa)VD$EF<Ccz&K9$ro|XAl|>abWxJ+mDvF{cxWIHeaX23N=5%5iCt`@q
ziz7u(QCY69uP9Z8P%?)_TT6^Na;7XZ`z{Fx%S?151r^3f9F90mEaW&Fzs!VrfUqD_
zKp07M3vTQwR0GLpwpd|ktAfH+41LG4c*1xhhI0tDpHMMiM9gR3W`1JXtPr{&#6+5B
zlr6IXMcPbRj~*v-l$=f(6Ci*PJ;4X$d4fRR&f>;_#m~%rmtEG&f|3ebPnm<30HKgJ
zljY~`fFC-FqQ;sMq4U)7AEMGSVy&HLJX`Jm$L~@#O4S}UiUiRhMvEFzqIQYdYEgUE
zxT#sA_Ks+7tBS@5QE7`>rAn-p7<X#6R@7*eAS$+heoub?*Z=wdysii5!FA4ga=or|
zo$GwQ@3%e2c=r8=N*oC9*FnojPda7nhCjRktgR+~&LY9?;PngkewLI{3WKG=+CJSC
z;Fv{SR8!g;5CK&-#EySV#=cRiuA-#+!^qM1=8)AhR;Zg*4=R#$Kgf?*)PP*s%;EL-
zpEP^B-4U|ekPpyubvNj7$+$Ujln0@K@lsZ@`Rd#+&`uL9$jB_BYCMb0lT2D(8)b(W
zuOKJfX3!g7@9XZeld5Z5FibQWe*~|;m9}m)q7*$~adv<abcB>Bunt{p_4kJJp#wDN
z12lV4kp1I8&WE_sWgpM&ytvaAM8~se5;L^Kr{zZMVT-eJsAw7G{GrqQ+-|F8(9@^*
zA@OH=a5eB8*Mq5&eU*?8o^t_4dfPJ4NW2bkKBmzt`orsoTFe4trUd6$O(k*9Pp#q*
zMUQ~yUgUqurT-wb^<Y_cU}8L29`94>%T_60`~C7BgtU*)kmGt4thzw!xDej*)>wj@
zUjKH0?IrMZBYv0a1T{VZ`S<npap?Kb0X==c8=4K0YlZV=Wt-5B^`OkWt>JvlO`89R
z^^QKB1i{h#2yUZd;^QA}KJ}FDMV~P+LsA2uKFD<s=)Gl*m;JoXVRB6Y7ce~^lM$MZ
zi~hNUpDJ`8|Ju;jVQ`~cH7tq#Z8J7mEUWwiab1?o0OYlO`1kEefBTy;;ed&o4&F&N
zy(nT7Kesi@Bj>rFKPY;Vu0Qwp^S*z)LHhSj5^e+HK?rs@SSbiZD(!5Yr7M1l`MK+g
zmuYzzyBA%yb5<AlE8MUe!Xq&1P;l@QDW~==!~_ONB@Vv({8Ke-q8uj4@jUG(S^@O^
zmI`@R{tgKC_2g^e#jkH32S2D6(F$-VOu4W|mtq=+e8!YEc;~{$vb7xqvOe}kL=9w?
zon1itf7U3}MEUIKJkB_0-<k@#{aVrPWKjm{E_5l3Ga8FV&p0)<PfOlIgt{T>g`34=
z`j<A|9v;=ToRILte*_O^MsbIvQ?X6{*b~gTWM=Y*mDXr?Em3bX((Z)Jn}riWBCb(L
z&*$qVpX4Ri?Y59oY4l|lNr|XD3vbZ<>LN6Fdt`T?+9e~ceASLEzi<53jVyWf&C&)!
zVK25;`AzdiPym|3EbL)<*KIgZ+J<M=j>Q^6Uu#xT(hU;RS-*20RKojG$P)-8A%AP5
zt0HE$@$if2#C`&byX&AXZXi@$Sy8G$z%L6WaxdnVu%P}nPqNuCMy%3Z%+Qh{p#saX
z8a4FmTZl+b((;uTBGomJi3}g(L9{e)F{ErP0MS!XSv8YyhLO2P;{<08J<@q1w8knw
z2)CuCH3Pe4_|+1h3|7v?=L?<q*vlLHRQo)CS01k}B2@ickg><Esi4K?X<`-<-{KrQ
zb-Wp8Nn$_cP*T!BSdlt(fHEMXba+akYuE@=Y8j?vs^Izo(`#VAL*<9iq2cLpaH^>5
z13`?C3wR|N_ywG6M;x54Q;C&8qFlPmH{>Vw^GUl*QPGKVzV3Tq3}YgvrA~LdhrDmb
zCO=vJUtU1j%hXESFuC@>Oh1>8IY>D+cN8QO+Kpee1rvPD;cE0BSHL_YFKVJpm|-O$
zU^DIWt1w7*tJ39RYxZ7`!M{jXS?~Uf;2~;R*JxsUbR#`M#>weiE*znZ`p_i{god#J
ztTLCy4|F-RRygAiu-W1L!RMw3BL=C4aS3b!pmCny_7`&QUF=<^+@BL*Wz2)?K$EGW
z{&|9W8R6QHYDh#>(lXaWThs&_s^1V%CeS_*jdmVW<GZI@a(8tle8}JgQL0gUXEF^U
zVbEj2xao_Jh=4T-*wo2e(+Wcrj<r|F%GX2^@D_}+cQS;nSF;}bu_e79W}K#P0So)>
z_QC~OB@(_RMY@V<W@Hv50dm>CV(<9SL*uRVP#=R*@9zbAgm9UeyR_j`H$MkUvX4UH
z5u=`8LYc;9SC2Ul2uAm@WXf4slT&1Qe2a=W5reHI6$_rBaQ8LWn7(IGJ^VNETiWI?
zHDX4hujcJERn#cn@6XA6N?7M_`nTM)Ng+4bb3I3OlTG5Lx$PKDYC-4Kw5Sa3c<}AK
zDBX9z?G;D8G2iv<OIh4bui;ew+-d1%9}^sScC>?3+=@P1Z$#RkpQtq17H;^dOP?(g
z22Nby!BJ$NDS5n0s%q&m=6N&j@FX*E=BegB^t{EtnXYEThzJ~zr@B^2WPa$V`o$sc
zT=kSwp1a-|)xh?Z7JtKk80acilXTlZ#Xjoq!3%LK&Nh6uME6D2tJaUz#eNwY+D^9a
z7%6J^*<Px{6k6L}Oy3Gc`#7vF`n<d(!(d|6AB+~iiye~h_x;fn&nP5{nTzhP)tAkF
zFc<V8=F3XarR{@*q>#%6^m@oxX)v-){GDyS8m~ukGmx4`o#x~ehfFqfMwyo|Uld!D
z(fwOXt32Dl-VfD`@-Uxp+NRTE7`|wQiHJdiVj@{}^mEk&9)?*klP82JUL7IydKKZ<
z90cFwmzW>d<+=%VetNvmjwjOckS@QKl!8N|2U0mX(*%QL_c6%hmIbTI&l?sxEBIxx
z7pkW}LON&4MIsa<rbQ#D<9<n?ZJk};PEv%(!RlOe6a3lUQAQtAXsuEmpF*6r)mN~O
zB`8FDqV3DB&O6w1ZvbJ1Nn9+<KNccYMf&wi7sWv4yT#UV%}1em`g^jX&bfwKJ8+i2
z_izow{naFY`=dc%Y>T0zOXNpQuBvhn7-MZ1DyB)`B^rgFCpE~oK##45Gc67%=P%Dy
zIBB%|eg~^(@uw>f=eKS<b+)>)Bm3pw?IqP3nsn3E^7haJsp1X$UJuv#a2TtJApe@W
z*gp6)y#Ys`^EVv)KV3a0g3fiV(ekD}%`0PnG)5tSD{~jadfNk$0Bkh?SXAV7`>kA_
zRSkOhRR*2pu3sXp7er5CV$!<}_IxkvG-NT#P>>rBZ$Z!8?7|h5thi|6ECSau2*7|a
z@R07f(>8fo{OV~*Vl`Y<;PTur-Mvy`YKs}b*KgX%$#RN`Mxhvkw#Q}Og2)-0nF8wr
zqQo=MTaT|9I8BJMB`%PannWp`E{4G$gkBWTs`mA<ar(ec%(}H0)1%6_tYwY5iHDRY
z$@KO5h!nk8B%$y45mZNOCL7E&uD$oQufz^6{RS=R!}8_P4%Uwz!KV{mQ$A6dXyTK~
zqeev%m}abUF>DZg1|8s%V5$Ncb6J@k?m$;K<Jd}*@Cf}8mU+sr9MPes@_%6Uw<7yo
zLt?RZWdC9fmps5v?d#-S;pk&fShVR(K(4`aI76l@PI$F3QonXsLe9)su8w_tQ?o})
zk}a_r*v!x3_#Ufl-YKrxkCOqGMMlkDM?aoO(6Sz<=jjwxsD&=yOniQfHC|(;t85VW
zkz*2z6%|<1;%awp$UfsT<dw{0m5p{7?qS~ewB9O!+13)P<!Ey>*a}^A)Vr3PmvEJz
zf!o70jjJWgjM}r#&@&m~GjdV|+e2RE&%4IuG7NVEF-TtXb?4{y0{vauYK6*`G+r-h
z6l4T>EH;LSd8o8|lMVB&5g|z54}|x3B)rV4scz@TcI-RDlD~s8D`q}@Ei85E*GT2W
z{H{LU2G;#4hHN1vL&jIP#A;Rh$ihHVz0hH5@88eRPL5-5zyU(F#cUEBM%_QOU%&Id
zQS<%#aIl}k{bvXhNVws3deu&iomoXi^m!@4(HI_3SimJN`BiZk!&r@#kj`HU4TGAJ
z!|+;ry$~);^2CH6%%ldulHAREBO$jnlD<6fI%r+Q@QiZ|Hb_?rCE>Fz)QML4z?kz~
zx0cx(O(4O0(Vl?gfO1(Ar;d}t<&TPVPwdcb!%cY}#kg%Lh4vhK)5W1cvBMb)D84wa
z>||T&>}ZN2o4v#w=%5h0pXZ`1xM7*x-Nd9wG090~u59q;LIThfkX&D%xq(UP#E|9@
zo>9>`A2pHDfCNm9zsh+ue<DF0lgi8+JwJv@rRaW~vBKG#VPY;#O3OC;RptygVq_oc
zv1m}Vz$gL`iv!vFwG(N902sHd$5qhJRMPZ{=TPLMARC8wjx9zPsHzQ}?Ozg3J*oqQ
zlX$_)pJTO@XofgkAlzz))OY#1(x_D9C}h7`L5Rk<Le9wiF5s^iP?W+-VVrd+-sEuj
z{EhN!pR+QTo;=J7sB{b*H?f+a^B}2vTcY}L7M)RqmlkJf1INxpXpDZAkg3gd-k&}o
zYqu)>HqN`P-y4TzeCL$^TsHA0*U5{|Zy76D+5k80;Qs*K3R(;rsk|QD9cK5X#+8+_
zQyZzBD<pEpqO>rmW`i%)ZIm-^PpCjZ;MoqXixfrVPFy6BFWI6xPS$zkfEuFVs&^Y}
zvBS2!&jHk~fz3Z0vkvo}(S~gfAwNXd4K3ifbQ8!S%G}Z+uhQMg6N~;Dj9AMqeKX18
z;q(3SY`|)rb60ZT(utV3*2~SyzmJ6Tb^YwsFOc_(Ipg>aPBOz!_E!0N?0VZie>AIF
z|45{SR>xut9dBz2tpU`3-&1u6+7-)N>JvX@#X1o=5%1bOsEK80yMjH+!s}qOx#x8s
zmR9=sh+|x@mdSto_}6mBxjNlwXYzGnLj=xq)|J)bMb9g=z@24~zM-+YxX>Chpamau
z$~R?S0FbC5vq9xW5Ezw3P8Nmmrs|$$7M&Ld11^#&NiXjvp#V=n8Oo`}FEe<h26eO2
zcS1=`n%t*ZAYYRSOMivhY!AWc{w360Vyr9qlgIqCzI93B$G2C9c#J_ub8aa&?rFEz
z6!IRHN2QLZ2r?MVp{pL#4lX#NSn7+8o_mOCJZ01Y1DIYuNvu^AOhI<D+WZSZs+U{|
zt*8(&{8!bDG&ns8P?i}08mH}wc>61-*~!u8Y5WpbEyutoFILNiM=GLMN{(_oZKcp&
z=@9VqvtNaqq!&`PLPaX?w?|YISk?AmAwBB}mQ3jS^Yf9B@Y;ow(%@n`JO|}|MMLb#
z4y%AywFZXO#+E}8ZN>p&3_Vp|#pR<k434!G?p7~fcx{sLG+D~^WtL4L=6R&Bu~PXU
zLbLv+u~Cwu><hij;_<{n!^*Ovs^Opo7qmDte?j&Y>hJUcDLFuH@gY9?N08(MKl~e5
z2SMwbjhB-(*K1CR6heCaB`x_epiii>4Hk5#<W+xhpZtTsHU1_$L2mdBRGN+Sz>H8G
z9G<(H^>!Tt{?-*YyYhyFg>ec!MQ3o^l-kL;y`S`Kt>XqrzHnwS&ecOuJUuBzukfbX
zvJd@VX&`!y?u)o=u@JeIWQz3~aelceg$?`)1MgR+OSR&LmBe>M0#WEh-s>>WvNU3r
z)h&!5>JpZ%m=Dj*Vp;}GJ@1}5w9pQrY&Bj09mW@k-4usZU$+|T{3ydh(Bk|iEfM-y
zm^S<Z|Lq2h&D-tQv9nv<&DE3Z+au#;IlAuTuu7CH`KD={UN5V}#raX#11nqrd@kwR
z)f~>6sG*(jfy$R^ow%~@;HJPV`N94C9P`Ysq3o$w+DqyDzC|k8x<DWlFXqwgU6HHT
z2$ryD{4$6X<O4;Xs7m(JUly%Zh7lGvK;=lZFH$)kr(SX|DN^^J{Gi)NS6!p*+!Q_2
zrubzdGrz#LcZcO;FgBo@x8^H+A|2Xse9G=~i%&r9TI>l?ToPSz|DWNx3%0oRFp81Y
zu3<q7*wG>lGOV*v_6py}cswU%5b~T;tRQsc$R@ppUk{a3Ze%<LjY$M6n^IORV$D2E
zn`-axm!{$+^uIWMp5{AUR@3bk;U3rTbWoy+vTNgs+93C=L0aeXy<=@Xz#DidKK`XN
z7W<{GL9W5U*%f<pxd}oOt@hOru`pZi?F-9%pVE$t=!hH8Vo$@QikN~pB!G7X-dDLi
zVo)~55=1xmxhogv?7LRqX<AP5ZDd}~Xs8aqARyJ?Kqqx|MWdTit}<Y7Pz#pmH~U!?
z75;oWCU@U*VLeNESO`98kF`KdDdUZr9trWB@1X2nGyk$8=Sc<0!Rv(;B=SCo&fP0>
zg-e}!?=)ZCueTp;uUvozpKo0=oBKQ1k!Q4B!|iFcO6%{xa>lJEfq^>5g$@3o4UPd~
zdc)bBfc-AQnvEKe!Q1Yp<HGCAVwatF5l+vJSm7NN*R-8KK`jNdLA=-pjY#ubx1Xvs
zI!~J<Uiq1wJo?K7JNP|UUs=m6z3X(v-iCjS-^CCXrDsg|!O2pwEzwYu;e0}uK6XYK
zjN^=q$3-uh)JMdfFNNO@lsvs6v%B%OYn6`rJQ`u9wUutvz5PTYyi{BGMbj#4Sv#_*
zKSxn7je2dhoe?(<bE0wQ8XM5C`TxwVS;!W;<-wso_Qt(9+}cqY7R8MrrrW=DTH*U1
zUb9Bsf3)!7(`aPknhNdFLT4$3PBkxUjGdzH)Av|QH|UT0Z;PAxGxu<IEeiL)?a29&
vHuV2=JLCFBt=RhSHS`nlJO4jjPU9|syQ(1WLy(%k=!dbvJs3{!al-!qV(-x#

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png b/0.16/_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..1c28df4c140c7cdfb7eda00d799bd3b3a2dfdcc0
GIT binary patch
literal 77760
zcmeFYXFFW)_dfhlqJ`)+f*ExXU680_v{8o9iHP1p^cE##^g0Zpgpe7cM@^K3L>r<D
zqIaW3?+kz6`wiT0;J1#wkK;P_gZ*IdYpwHK>pZVGeLW2-3O0)W{`-$gOA`YB?>~an
zyPJiK<nD?0^N&IQ5&ZX`7DUA;FmDIno8mr|Q9vHnqA>TbBItvRZ`w~X^%8i2K7si=
zh_M6Sk`Ou!sq2d%k!Ilfs19ANPtDdecF1Ias|$cB)QI1qpMBW9dAjZ4(>yova^Vb{
z`nUCq7BJ$7U>7}8=j6YqfZJ&}vuHc91mgOC{eNN2e9?At-|T;JzQC{jpZWh*fein*
z_`i?ve>;H<du!Pi8}+}q#n<F1GyN|(y6FD@7g0A3;rl5!BTCB)_=7f8klJ821HFY1
zf~FG{E3PuA38DeFW_hnsg5T0C<8Rj~<(!D$zRl~CAqRw=N_$^CRl2gcC|u*e4_LUp
z4!Gt2enkx!w)lkN!)QWjv19<+qU*%GqbtrVLq6`o@Y7w5*rSvIw9yN<!R!E0buIGc
zx!;0qG8o^mw#bkhVKi#*_6k45xaPE$7ruL@LWW$p`mliGyy*x(7ZM#+y2ah*-R`#G
zU)M}r@9Z6Iuli)amjLn3%6?E9*U6+SWIVjSNS6eYq0-hW*N=~w53Wy#!mqB^j@klm
z2^`ObIM-W5Gw+Aru-}81-L6_5T8C0E&)r^^C~Y%;e-^eBuy!VT=+w6I4x8Ql?>zik
z=9bFO>bCDSp8s~WV88n2q)l+rqVGr{eXvciU=JT~W_UUETQ;0zZ#OCcrW+K!Iu!bS
zR%Y*q%-Zpd+g?@A(0mrxhuasIO>cDaZXaiagnw8FSH!vH_wKzeIy)<-2|uX?2NhiJ
z9EIJ^hwTcsUNwO*NB{PBn_z1}bV^rpdtueGN<z!HU4E~yd2hXtvu6uyoVP3C*Fx1@
zZRgClk>RI9H>}+o1&^7RaWA%q=4_PCjx$bH@kiHnu$?*lWr0$)QutSVZOh)=9#i<*
zv)+#=<@ka(u0!FeKIUi37tOzq@WiQMyDzYEANJyo3UG&VWk*hzyRMz6IGptETy@Zy
zrY1lWC;FqsKYY#V`WkoOvn{JUCUrd)ehJ%b-mPw(vu0Di*{8lu{oc0rtnI?;$MuqF
z-iw6;>KoD9x2a1<Lv5=^qCcX49&8>7nVEiDtZ7`Ag4qLM0S~vD33=4bC$^JW9C_mL
zOR>tH_+z=a?DvmWd?E444DC(y76RbIemt@)#ka+c!@K(pY^FYun*a3bjjvNv@L}G<
z#lN-zje^h^nb)^zS68Anj|JGupy>z1xSEXRz`Tnk>sc}C2X2&b=)^XapJjT)*g^aD
z>v_0aSI3R=*YeGYM=igF%BJ;Y2tf>XvqzrIPJiD)R-P|zrBh`wKdksdOo~l3BIo%g
zyEK}p8DCZ^rheGL8Pdy%Z&A8-$XZkgzhPEd+zYK;YP(dr34NaUnJy}>R9G0US^FO%
zB;2t7=h0sHA(c!3>CI~Rx#?`%4~LWEoPTF8_D2$ORu$@eI??Azufsd*^bA*CJ_Q<r
zUEg`QLIwpaBlKgBz{OjP{tc>slPiSf?j!2SK<zLZb?-d#5~8VRjs^PAu@?#$*C(3h
z>2+K<O}-Dzb%&nUJ4+MxNNwdeXB@q&sKJ0IZ;69mINNcNP=hj}j^|>s!WsqAe|G~h
z`cIU=nQFVOo9At}O1IXX$J<1M1=qctM_vC?ngVa~{+)upUMW~FwA$x8*$?`bM^}0D
z%B|Rb+#~HCscHIF6m;|0qlDv9!I9&$@#-i3p{ZSw&9gK49hfHQ{P+_r=yKK0W8sHG
zTX={6_o9&;buGb9wB+E(3c=)b+{w(>))1eLITyZ?M5D-ra&1NYF}=~?H!q&(Y}wO#
zW8>YEFlBBojp!gA?!T{0Uzti1!mzBR(%UDQ?p7nv2wDoO1~g4<7!Z-DA(28P$A79i
zB(u;8Hl`{rlil5ulb0Vh$uXYPpNjWllcVD13_o-isjkApX%LtfdmAPX5hxwiA<o;S
zg6nCwg9}{2(fFkGVN2Ut3-O*Nh%45awp0zW#;OJUv~M2SxZ#-a&!5HeXOQai$OfK3
zSooG4$7#PEjEok*1J75r)>}mxSlCqF8_cx>r9$m=qKqg2(qMs#QM2rzM!FqWHA|4I
z5DlCThwPY<X$|w~F_j@|G_TvFtV==vhj$4NW0J6dREbVhm{?uxs1nK-Jg=7BIiplX
zb1GP~JB1zl7jqJ+z6B8zMxMA1G*_{PLdf?$94U~LEbhY=cJ=BM5p;9~bh;REI(YPq
z%))iX+sov;bS!Hwa+9gQ%PWbQf#eg%-!m-=8Sa1Uh2Y-LcSXo59X<1#8|y8#`9?S5
z6S*|LnY}(gP9M??L>bJs1!*69;t%-IvR*kP4`QSXLbn50d+4jJ@rYGI)o`xvIIFp|
zdi5?HE7n)W>1;zK%tPP8!c@ke*P3g4Uq_EW)z9Xyn27JysBrz9o5Mm6bwth9?aHg`
zk@1$k(w+I^%bfMMv9PIaEs*7Jm1}c(w4;?*Xbo>%zw2DBtBGdmrjvHvlTN3!JY}KY
z%QBm}DlNU#PY^7ET2h?+iEEUU!$=ymsNtnQy>;tT{B%sGS;t!Q+NOUO13S>cTgTp8
zPD|6={|gTPPFdpqKmyl(`ySr(_8!O2fH2(I`3AUl@Xh6zvhrAuw5W*Xw5MtgjBN5_
zgNjQ+eI^Zevum$-QN4TtkfCqlC7b<tp}T571)MA2;{YF_!!=G(Glk@ql^K)HIdMp+
zdWPOro<7=T8JsCP(#@fTr+Wjt_QOc6^pH)&IRX4Ur8ICbbT42^oOk{o-NJ|GzZXOa
z#MaFG&2W1Ia`dY9-(TELyAix}@A41$xx1Cd!=nQJIyFKJ!`#a?*3gOLVOR(|WsQj;
zy&NO|KJUp-Y*Jk+sJf=NbGHrteKD~S@qgc3XlcFQTL%*ZEm{2bzj~Da+={M*n~#*u
zJrzz{2i}SwA#LuqMHL6L?2D&nEU&;69a~<gki`}|a=htx>~b6*E(kVp?<%h9758Q*
ziKKh3ICWq<Y2)247sJeE^;Jr}wvowAu(k@T2%N>r59UQ@>zi+lXFAELCeHqQ^JM$6
zX#R8C!$cqMG4{{pIbJ(8ZwAIscuglKW%s<`b}h;Ob_pwqH9CTbF3U$O>gN6h9<H+C
z93yiL1V3r95K6#O>EJGDV%V}S|A#ttBG%e8MWluLqiOJBcM=o#m@`QL1iat55hs3~
z(H|*Jf?<F|(IOGX5ATzO{8vsaDq{Kvg35JHTC}a4st~FlL3r7VN=igjP1qZcn8yw0
z1W{5pygX#ZFytC{tu_8+>5|u_p(`J(wKOjl-yfKZ%c9OMU<Xo%{!-Hti#}Zjm6g$m
zi(mYgp>n^SC06N*GHK+`gl$rnk+e7jl0yTh2|>He+{*X1SCesp$$+eTMCuQh<qy{f
zsLrnP+3DgF4I*iPJ$C4VM{*j1m11!5H!!B`i6`Htd;$YMzIfQWe>|pOb(GgU=PCa*
zghn~(NwlYLYWtg*0`?JpJ-IY~4G4{)I?uPD7Tj{tn(Gx|;WLl<*7f?Kzs5-qxxFZz
zXKI2E8SG!O?PWh-UUG!+{VcZhx!B=0vO2ijt`=L-ahC?(1Mh!xZVGJv<RWZrB?*)K
z`OG^*3rJ(QyflAw+#7wO=A-x6UIHo1J2=4NQq#CI+)FNV4`l+$GN?p1Qh_PkP64!J
z52)YlF3gLOS~K+-!Hpbn+y?MW-lM#X_swqoUH7Xnm>++M;w<<lfFG|r%iJ_iBb$Sx
zlQmElpLEqu>FH>i`L-!AQUcmJq+D;Su;yYX{Pk3^Ho_X)uJ^ngg9BA_u6)=G3cY_h
zziWtYnXJZwYC;d$3ty3|*8=YwGM%H|(y?g8>1f4xD%xP)(*oZa*#ndX5rPD0;^?#o
zBXGR%8uxFjf<gZKL3*SpDJ}I9vb$U&aXA0@!-Al}Jx;_sAY`DB(^GL|{PA`bFYnNf
z!aJV|6QY8UWA|d0Wzn{qycibpzL${omV#?O!J)j}Hozr0K-#^<@#~vkVyBuiMF?eW
zZ3kCkBs=t}#2Uhc1R?N{1@zZfdNAmLUGdz%{Sxk;CG1+)JqxW`8mSBwgtD6BOv9gv
z#nu&B3P<Ka{tj<WUQY7T3JFEFb<{{7Fbd*KK0<5vGXGKVch!Jrqkj!erEQv*JdY@X
znz&t5Vv#hGD^e?t8hXgFP35}jMOCgoL>aeuw|A1_BF}lU4j+PUzi(4J0^TXCshFBB
zbXHFnTY&|CFx*6LaF59&)0DUVGpDnKuqKFSvTz^Ci@}hnJ}2&>Go!=<r_HazSTL=A
zAhWo|$R8%&dsdi{4Bn{rRRwqWXP(%3z`Y8S$5#gso?2HSTRNV57SMWHH088eNFSnC
zlARPzDKL)g(9w%rSv*gWdGllXtp5d9VxgO*pyhnW=2E_x91ZznUUNZ|d+9b?DJ8%u
zr2jJ%Y1pkRY4r(f{{SDTw|$GFUCkn)dAky}y8U{W3~pmWRbrRImyYh#rOmZb((6<Q
z82_3`>oY5MQRj1>iT5)c87E=pLnNFBasAeeC9)31+vQAuG5>1zNLIde>D;sD5B2#J
zpmzE62QGKs>*6?VE?p?6*e1NKhdQ~{!s{t-&Xc9}x5(%XC`ThFd&B4VVB7rU;}Uuh
zi=IYIr64*r(Qarwcwo(=Cs2Nwfd<&1?_5Y@o(5j`^|eQS)<0H-89%Jgly92%_NpPd
z=MudrAx?~7>U{6n@fJmBCj|()oWFxzmss!$?%W0z`_{`hFFKBe9-d+6KFEbsmJ_)T
z=t~oluHDivv-#dqY`Nz-kECUvdsI2ZBg1WK`e)ne^RTir5{P^b<3rL2U}%8!+#lWO
z;@IWGo{B%(>+N`3&lLW2<b{HoYZ6@?1O5Vn!UC3y=^N+1Ojl&X6bwT@rctPZ%VM9B
z3jNs%ErVV^Z-xqlhH27+FtzPmgMRFBT<PA*<W$fsqxm4(v?6pnWRRDNfAuQIYLFZ*
z?9$Hi``3}mf$B^5q|8VZ-N*njbqh1hrD1t}mlwdB=QBI|V|^?|sblYqvV^{<ES7j;
zTN)-(;OTS42;z6-Ol5NgZ0^0rg1)#~Ot>dBGPx5^)MK4vJ@|rCP=p1*x&T>Rsy!Qb
zjboAv+pj=dQ%nCO5P`IF+8N6y^8hLfZN#6>y#1e&dFq!P*bn-jH+sW}>E?qXD>dEu
zD0Gw1mCyQJwZqwHbV-f17Yf9n-Y3h3yBr)FSM)Nk=WQz-zL)7w7ZFS8{@#&~>BtX$
zlT^!SE(lHx+#4RXO}v0@J!p%}8w#H!Vs+^2O{~xYrP=|b6E)+|+)wX^kkJ5tl3uUA
z&*S$7-jBi-y$Y@u@v|0={O|42P3YsI>n$B9t(KaO*aOK9*~5dA<n08iXidH3y!IZ=
z%HQkWTPawE(%kr%7@~L`OP8X8;pSqA$QN4cBk8pA>?G@~-efGuQlX?Lgbobix;!AQ
zPHz}n)W4_bCyDsDrKRKpUv%CMFWgkXvM}nG**ncOh73e!3?xT`IBvy7(2vQE41E1d
zV<n3u6-8$q?$;saXi-{fLIqHNvYm;2tNdlALQao+raU=ePCBXdOT(&TK|Aks<4-f~
zw9-GC!{4Sx7S7PqFVpS6Zt7bg53vU~SWnrdm}~~gEBssExRUM6gz`tsL-}9!_`y0x
zP-W=#dBRZ8&2?HzzBM-r{4fed1a|f1W+{G%MS3_4U#8j69A7CYJ}q9-%nuMCj!evD
zUwg%?HR#hS2Ndi&Ntd3;sZC%cK&bF&xeM~}rf@u83}o@LE!q@J6r=p!L8>t(MaHV4
z{=832<9TGq(y`66%DmEP#sPl~GIBeHX=mjk-KhG;RjVOh)l#aSS1ggrGw~>TV`%4w
z*C*gz38rdn5sVrL*64d28(U_wx`7Sj`w-B&N0GmBlRUg2IsYSwvlCr_g-WC-G`i4C
z=vP0%1cpDQd#4(gSo+N9D-nil9&oAJtNT!c4|o_$NmLt3(gzgS6#5S>r~<ghvxWi;
zsSW1=#Zi0o7Roo-EXu2N@q>&z-qv>w!M@%fFOgYCKan4|m)l6mZ&QZVuJuMzeZ$67
za*n&OH@2B*%7%uW#;n%hllYE*md8zGVKwGQJH3+gt)X3FJ2j2JVqzN~ldJ3#Xs*6@
z$-SfC#)LLxMvZ)AQFf46@54W7N85TVyuARrD^d$Lvj(b+iM1P(G_PcPkg`x>rGB^A
zAbpySyVTLDoH@+yp1MnYB#q5`1ueH)UoJ)qGIXVZ-#fbOjt+$&`7d)7rAdYE54#IX
z3fN=D-kwS44lp@LcmIXoCr=);8Cvh<H!7zBsd+yTF#cee+;%e(_|>Fq(aR$eg(R#{
zoa9gA!2lDr#6Pa;aFFU$aPeHrOAbnkK-Ts3SfQ+2Wqc3crQzOee&E6hGjJ8ap5mtM
zppi7n+~=ob+EpSG0zgnOpHliPR+e&gP`@<3xwg|~b;$;%L(3~!5~DCJP?hhmmg+-{
zllLL({5)J+)DREt^`R16j&ONkmT@t_ob}vVE6iJFG$xTc|HAiW;I}AL-{oh)#0E{n
zCqF@%1=a4oWcH4U;(^QuB5>A@ur`eXB3yW%C>58S!{tglKWCe=F8O2PIoE;R-b_&7
z`V_sz(0Coyq^O9Qw2&S(JtSg0r>xXDCw_y_=lzJuqN=@OaCA&|AnUhjXK=jy>?86)
z$q#(RVF+P;_{V$lGww})tD27DX7CYJZr?&``EW$F@%D~68R`@9i}V>auUd4q-hGBM
zB=87Y)=lk_(R%PucF>kvOl)iOol%)P+UmUuV1B>*_(&FRqYTcZ@gXFDW^{F5y-j$4
z;enGUhYsD<`IhXkFD#|BO@E#i2LThzm;!0ZL8&F3_>PGFYLn~MPc#3_tQA1_li)0d
z6ouf@+9Ok3{R29T^v`>VV-RAwsI=v{{Iar<WmdT5Fo~VJR=8S}&I5q<2d+-d!+drL
zo835{p};Gus&HkksCwz<vXMS6l*(r~vjH@th=&&NyQ6rEZKhIu#w|5c_QQL<H}->n
z8&x;h08&5qw8Fqu)?8>10ndj7Mlh;*$UKSg;(LxL5%{6K-Xf{DU=$23UohH7nClIu
zxVQI58Hn@-h26l72Smf3Zp6d|&v*wNAWC3LOQ)y#rgvQ<&&MfWqwbG>zYq4?VIJYc
z;_&K;kjov(iL6Qa*3g0DQ#UrEasMvII?%<ri@ICFL@}BlgP_s$ie^6m7ok*2W2M<R
zCDq-UtJK`I``&~sd>T5#z>ukuP<u2smffrH&p4{|Ed+|R$O?WtPUp^3Y~tHkB#}17
zTigrq5wVSO`IV43IPClIQfjGNMqm#Rn2MrDpad{pE7*ZKNPM=R>6kPlRK=fMNrQyB
zCiX0wpt4ViGepfBSmIgrjyEmW3y#78WcM-tnVJZf?jBN|@sG7UYY;MIp2Tv6o;GM{
zE-~Oi*Rv(HUvFTEyaUDd-4xoHjPR`H)PB#Vs`J5wM}UqU?xRh_A-7@p+dMNFH@FEq
zjN64n6SSiGQuc$Uc<r2(zG?Iir4gg(4l&OyQhzJmJoRLgnoTx)!D?v|m=Yh5_dZHs
z+vxox5!u;cZp6p3hbVDpH?8S3gzzSq7@P}N&dT<Z_j(#wFX-18#xkgK!Th=^EQmsz
zOhk0z`OeeJENUG?tpdu9PCTAq&+%2=THW3cW!K)nu)w|2T#|QcJX(}UP~f55h(vPo
zfmm8y`#`=h9e~8cWA}ZsWQM-ZN^I_<@+vjWpSdlh=rn9X@|>F5<oL#Rj?+RFEj0C$
zV~bUKEq@fyt-LILDy35@>z61Y?h$@7;-v3aI<EElX_0SNu;1O-EfXSVVg2IAqIW=#
zRmeLRsb&K-19_1^oCNL>ESbF|e+p*yiKQ5c--Mh-a`D<dSVMF1n%jfpR|SyjT(1OO
zDi{<69MU8OEDgpV2U^2raI;D{BUsTjhrK!?M6b?_5c9FXmaB{IeUf|XG*?4d+kffw
zUb|XjMCTD={|@l9G8}Opk&9TlmpM#^U&#t-!8F*<8wSv#N;R(78|>^^?BeguPo+Q2
zXeeBS!RA|(;yULH?E|q}&#TFZHunwfeWJvK5EH!oFrd&|;61$>fp4fMni%aDxphGG
z)ixkGZ&tm9UTKITw(gQI3n-6Lx@t=AtL8>@%08X=_n?xpWDLpjdsnffeS8ci$B`mG
z8`iO-AfZ;q5*QpbfRZR#A$X^ZPONz}cS+vUBD1>Ki=$G3#(A-oqS`;FxfS48dAaWj
z%9KXRM+S@B#`VCQw;X`qS{54=Gxa<&Z)LSZG0G;4ihS4dG=>nY_wvS8d{v{Fkhnz#
zRrH{*pYG|6u&k%-UGHrFSy21y_to?Dhw24?tF+e(XX)A8DLWQ7adFLQfz+UTV9$-b
zzXf`U9_#5aC5*?@`X@>=bGH@{1M4o&`B@dcd@&6=!Xi;W6#{aDDxX9&%Wu4Z_9DYw
z&bAAvsne=*1*TAQAnlq;z`!2m%+!o~hITFYeH#FQJrfl9?ym4-E=FkPB-01c%|ChX
zi{oZOZR=czjeqYDkwr+;jj>c1BlN3EpQPan`PbTS8030uK^y!qQ**vOxl``fF<9m^
z$&%Cqb5FKMiaf2kSaD#K;7`#&#ilazO4r(oi_OrXJUTx;h!VXj*Hew;2zwhAJplmu
zV4?kW)KK>wdcesVCz&Zf;e2<dj4Y#n^u)xis1BVJuUynQ*}EUObQ=|~ibVaYPa`{L
zaW2a#Hs?mHO6C7ep&w%i^fw38F8Z>N`Lt{PEUR%w%67%V37O?(Vian}LT+UWyhEpk
z>mQB0`z%z?=xol7ewjm(DE;xS%}cZ8Dt}ukEfD1+8E__pFz257n6~cwZH52lZYyx4
zf(?wNeb~es2zbeWHfX*m1e{`z(BPger2t@~g156%b|`U-SUl|=FgvY;T5&m&JMYR7
zTGM3zvpGY>GU963J<~|j@DnC1AR-l<-*B611QO7OS}TtL;^61OC1!MaemL3)h5MZe
zXxJc8-aO5z5-<Df*s(Z<v8Fb_C%@9Vjk0@%Tztc9al)IDm|k(ZciaQqP?ba0LI!7D
zFx}7;4~VL4hGy64P#%eu{i=aZY;2T21_8^-=d6%A$j(00V2N<*Wkuan=Fx>78$T0!
z1OHEQ_aV?dO!LKT!GDLLg$fu51t(Mgf|9I-q4cAl1II^1=5Na=p^zW=sn;XBRa2a&
z&KU7BG<(40&|l7!Z|tQN1l6Uk9sXyiMWUYr%8ve$7UmV>h&pF5=$$g78d9Id9UFVL
zi9!`}2J_*x3>ITq#^~&TQRJWQl^c_y)U2Qe=&zv6Ax#(G!9VJSaGoKgosxu!w7a{R
zr7A|+XWxFDs}jS>7r7vb=$F4|JBKWb9ur&qz-^#c3sAESn>+%N%VikWGwF;);MvJo
z(@ZK_lB?W@_O?5!)Z%M2zkzZ|y)vgWlMLia9SO`o*TwHSh0AacH@^~q=%AnH=j9P9
z>JTfbLRm1Wqg-2$$4@Edc886VXySv|4}7T8Rr!a?B;Pj+4rK@sOmwko@!?`Cg<Hyn
z(h3YL#9NfV(w}?1^yYA9&-LrcTDV3*p<8~xZ-(i?AkV$mrxiIlRio^DJ=6;iLP9+u
zJQl9D$6Q>P*e4El(kTy++*y_Dx2M!UemlMyXuDm~zS+<SxL)0j=pA5j&__)#&K7W~
zYDNU*eYU{gU%3#4*%<gc*yI_=)wv1^K9=Knvu?9qSFWvNC#OpfwikKwsN)X(5IQek
zMWwP?f||$UX$XyLA*{3$U%pQR2L-Dd-OXkwbV2?cDGKIV&Bk3Coc{tZD@&xxqk{-R
z3iHn(WH|009nJ}9Y>%mq<hfF@6$e$qi%>Vzgs<*z=7kW|jlIuXOqEiDFlWH$X{y{4
z>6P=hG_<i0esL>pPYwuAj1DkuS-Fqg4T>CSfjo#5W0w1jich7J6a2-%T9R$U?rchu
z%itnhJ0W&OT~CO0o_T5N1`!a%9M1b83yY1|;#bZ4a=E<*(Og3b&WhSJ^wUOMeg+9$
zkW+G*4>T(Z-hZ<mSu}JvX7S*(3i$`<xrV#B>OoYwPw9LhXfu=rYFgc!yA#30IxdBw
zdCit4yv6j`r<Bw|tXpFDwf<4*s@=Pu4T`x}G~Maz`LZ1PEh*UXM(1rG5b@y)8{R=(
z>&@1ye$TM#)Px)X#&LFV+G<O2!>A8DXMR4k@8V7?A*`w{_%dU4czVh^CiO81T}rv_
z9|Jz@qP{`lJ03x`1S@MAfl=7pT+`204u+E(3`mN*jw6bbw1UaZPD^eDhkSK>b;gJI
z<!m_`+=Q-#i{S73&85=x*Ks4-Hyo!MRgIZcAT-Xe%y+h$?fjE3zVY|GHDPmqY#hC)
zU_tJi2x0ad>-S}U=^(x>ZT=9f8LLNM%Xdkj8)>FTf3oJowHWiJ{>dt(cvf&SoDe4u
zQuoCS5kSO;_}5Xgg=U=8i7RBx%$lW2viT1&pDS(rk$IghiTqI1z&+51E3{FNA_9_^
zD}#Zr)#>mmY&n@m2I&b@4*MrmJS`h1F|S1GTIz8E((HK7vxC@NP>Z!S@+~jM8mLMN
zd)|<OB(7NZ92<px*EvnTYMk-8EVX}7Hl9-<6&hZ|i0KX3fc1}cdN{?I7iIipQSFZt
zrt=~D#_&Fb2Z7+BjpVjBaQ(LSq|1mgqY^7FF(mog!%?SUT_*hYYoK?-8Z`80TX@~t
zP+M?U7(2GCf)x~aEuUW0^9)vH+O|J+o+0s~l7|G_2yS4iw+(0Wsw-(O3&gKy&IExi
z2ejXGIF8r4TKro4TcKSexzWGmORe&O+q=^!IyC&35o5~<NhPDjcz+MkiWEKBpAYNH
zz4R|ad{wNfv6@Bn-|{r4Ry1OyrTyG#!0C}cFS2>$CmJ;~VtZHK5i!83^6OrNLUR`A
z=CHyfU!q@!Ugz$-ejzCrGr<%<NsQ}54)H#*LYr4&<!EtnHf$<Ty7;Vx*0n8Hs?Ay0
zqMy01gJV87>?>@eqrGOYDahyJrtr4AZf5q1EHZDUZs7DC2OWk7`c}KtHv^b(b3I#0
znK!P@`^Dbe-wx#SVo}n(lumgt!Q+r)FSsaIA6%FtWPip`C<xbC^@4G{7CR;DwJD>(
z=*S)7Vs!}Mz2x-2Z-)RM)o52tjH9Y7(e5;Tsos3VuuMY6M*?bf-+Atl_7cT4?Zp97
za~O-wk+VqOt9z~|lp_iNs1+NVkDTgip0=mJqX_hSuBbB=fP+4~(A^4UiA(=uj`K5l
zrKOKcpC&E)f<}D{iuAFD?knW}leS>{alTBN7tJl6L4qxr6pdv)r9M0V$HwR;ANkL9
zAmr)(={l1e1JoE()36!OBhCM(>Be!3qMYG~rB@En(7C}j=lj=%q;y~0rkrTpwQK!C
zBbJ5ihiF2DhyuHhNXxE)W{3_?_U}?X&Eu&;W$idHu%bIC<{@7)x%v!^#PQXnv!zc>
z*$~`N^}FkP!~Zdbap}3>q~K5D@UbV-g!?{Cr|LpztQjwD?<X~IX>@Rd+UGRIOHW1<
z^@ma7(Ws$F3cxTlLy9GmMb&^m(~b-jSN9%>(lr&V-N@7aKxa8t6;RJSgq$OB*M`l=
zX3_hzyon{?mKu?Z%pGBTH7T88aK|j0ijm*i!rKSvThwdjDx906QNRux`QH<37t-oF
z#u?K`E_bS0StU#Bioa62gie?`7TD!)_%e^3x+i~xvBA#sL8;EvLqME(%~s>{=g-Ab
zB!9js?mbFZ!qb|XGHcr1sqH6kLY}aNot21Q+~N;7k3N}-vUcb^zyM@05c$c7u#vh0
zAwsf>X9-@Bvqof}dM%I$TXw>L9{N8y77r5*I$50ER6UcYl63R{_AX!U(=Yl>u?~_O
z-?t^JFnI)p5~lIc>Dfc<rry&TwFXFK01~i&hwP=wW<1$QThHXb{{Y)}NCoqd10<mt
z>-F!!n}SSS4!`(QjD#AqIDOJyZ}gCYb3E<+VTw!t%0&VE^%cX@!wpm~AaI=uQ`3zy
zo4gcC{+{#{#g(a}y?v#dl-ox$yZ5NC6z;g6N9a$lS;>Fg9kMGGD_vtL*&0JB;fE9;
z<6d6q`)D&G>vHNq^|vVcF)6Vmd{aB77h33>G~6tl-*gP?Wbk@Hzy?Zy<G%ad+w<Ry
zThy@K$E1J*$iBa00Pw*#Xyu9!Jw{BAbIuy>x&L|7lEtDhpR~%373Z=gEH`Kdzo&3{
zSS!WD093Q3Oy!o#oXe)E#ro<dl}h_QKtdnL$;MR4q&kis2$2yr%_HrTOqypW_Fi#9
z$OY0U4?mprH8DjJBeCLXAt9VneQ}joW;-%m(RGZlIWLR736^+OXUx4R<dQt#wKexW
zDR*t9L9#+KE5$Y6r@--uX_MH#qRt?qkG`euNGx=$-m+fW+3hnLQVKsFQr4j@9UMr9
zzTC=@W!4~&Fmr1P^X}67GfT~V6+_x+A~BF(-{<W1-YI32!baF5rx#_i0_k>+8f9~n
zW>c+@Qe-2fD0M?slGm;HhP>?^nTQS^v&rC{-D`Stl-8$N#$RBY8@;vtdVi2Ms`IMd
zCC4jbbTS?K7#82}G3Rw~Ec1Cgr*|N5ZxB>Uy;2=3xH$X$ae^~%R#+*G9kQ954EHvz
zSC+FTsk8B+Z=K3kPO}BTNH^_mf1q`+AzA)rAY$=zVuX?ZCdKEZ$mgTDx+4L1GMnxg
z@6o9W!3xk>i6)I*6Q`Z8GjB~s*Cf@~g;s}vn&MH6mRJ5GysVhI3Ry&BX`F{Wo6~b0
zOpp<Y<A;dR>s;h{>cQZM5tV4;v)!5KAbBr}WYHhZ4rip{KNl{mZ(*;hti?DMr}!6^
zFLootFJ9kD-)^Verll8rP6CsumNQ!5m+6<PD<UesQgzHhCemhB68D_$=Hb&8XlZuz
zYVUZCRk_y)car1A8<IU*4_kM)r*q;Lh-c*~V60K!sPW@KmIp+?1agcw2{tm&O0wN6
z?Y`m4R#oqlUix~59x&{&$K)z7xA51OYolmamk=q*{4hn{671iFrFTM9{=aCh`|<xC
zsm~+Pi<|rl>$^9+B}>?@U6A{DZTL5phribk32`Z8NaU1Q%v+<#f9q*(Cwa0wVJmeD
zxKAl<U7YwBrR6<*<*m@l@R1V!?DqAwxUx}~<1f-nvHg+erNfKl?JLG|<0uqKJ=!OF
zap0;LZGD@n-$@M5Xyfh!{Rh1iRH^wFt7c3G-z?F$_%J551`rz1kE7wYeIVdY17`#^
z_&}7>Q_buQM$(*A;%RN5QHOK7X#1%-asoNIU-DURxfuIU=^9LFJd|X9BFm(Z<5dP^
zuSzm+7q3bg+Z4-3mA>RWAR@c8twZXuVBqo+<J1S50qNcfZRD{qtY0}~!Tm<h{GR?L
z5pgyugxnhrL<=6ibfKLF3zlY51DWH{X@tB~KG8f3ujj*5+uEF3<_xO8wuW9s-}o;6
z77f?b%Wf0_lNkpfmfB}#NGllr%(0RCT50!-JxduuDW(pPfQVPjGYm~L+37(Mx%`(M
zg?}W_zl@@tVEKoqOW2u7(aj!Y@C>bOhZJK104I562wO<8$Mk}M5U@*9CRYR4LiMqv
z9GD@KHOiw}Ojo-#^c^ZKDCh6qUQ_U8#c|YtsFsbov=CfRz|!Vb5(7UrGPS)Y@qV0U
zv=QPzHWe!wDdsG-FJ77UJuKmce%+#h68nQI>#TdwGGhw!992-QJC0~jqh1QLw<O$S
zJPqH=2L{06#-q_t+?zwBe9EfX7R<t}Lj2qXI4Eh7@iGk#S&wa?_`ogaXg@c6SIQVA
zNJXAdC6)fIlyaY;vszqQ`k}eq0Q^SlAT{7RaPZg8HDMxi>YiR`3U!3LX&qC%SALwx
za0)jdQ+*C9Gz}HIOmq=8(5(GqPY2ec#K*y&Saz2Tx#)Je$R8LN{0;SiorvB_XK{vh
zmb6_3{C%OcBy%&dxBRYSC_i^BqPUhahPy78B}^;f&s`?%?7ex;xdD^xu^aL_+~_^{
zUjA4!A|(RF{d)HeWU$@}B+$e2w@Rk7S3FN$pRIuCC%-*hs#xv5nm0z$&^WrZwlgmd
zw1~qV3`DHtlA`I2Ot*^~d+9jkUFe7l<T$7+WGr*WA1vwB&)vfZq6_%S$^f`dsnga=
z*Ob^a^+#CEA170<ZMH~$T(BXz?)M(O#`6_~T$7g@_i*BS+tx*IR;is^qMb~KTyioQ
zfLitwxk5WlVerX^ieU$jc_Njr--UO`+`(*wW=q{pbMh{HrITvFVRi5N-QG(<EjovY
zeZ$q=6QbqajtSxHpRTG^4;MpLA4Eufhy`)gtLf{pL81G+QKb8W7u;R8^hMrN7j9#;
zhCB`_4@rK>XU;z`7Jn9#A6=9YK>zdVY=TaW!t$wiMt$dori*~pm~m45<6lo!D>O%e
zh`XN+b^H~0H^ND*(iBP;*c7m;ZR|4}y<h<87^6$)0rhI@lw$;~+~Zlp&r=cV=s-n}
z0tX_vG9j-dC@*1CSr7B&e2_H3D5S=|I4@C=0r1755F=2#;Xh#d+Ip1ptdk{MUH9wv
zmvjD<J={lsxFuimSdEK0$et*4C+#IEDP8Tn5g@o~kzv8m7;8N<X<q$)AW0b2P`pzV
zDtF+5EUo7O#Ys>d?Pwk;^R_!U-^p6P`9I62xAI4~%TsAuF_kTnLKKD45s(>gLYR#*
z`Y>C-u0}omorYFdnM*2bPO~Z>%Y(C37qmqDv@@GqZzP;s{z}!{($0#f8=}ejm{^@L
zCNfAi?e&ksx!u1WPXtaTQ;qK5Y31;bjcOHKQOj1Y0uIgy9<_)+OopuTIp!=ouG&wi
z_?cOEPwmOOomVgFD<H^iAy9S+vSu$KQ7$tHO6g#T7V!JfSP9^j%C4&Q^&c3sis4ND
zR*D-hf5z}-#+UcsCf_Ruv)2LrfRAV(LChF3Kx@L!5DDk|^v#tD@6xHu9bwxY%)(x0
z)VZ2U{%RbVVUtIaSTkF2+Ir`u3ixrPfW9Puz4u%C;^3zJ1_rzC4qLwn2%~nV%57Zh
zI&!*BS~wSZ0kZgXv0h%_9q^`y`cx)dL9~K_zWPUVVcp{nG4W_J#bH7T5<NVk>T!$=
zsa_AaE(WOuw@er&r2Y8@M#tgj(dzT5heF$O(wRq#{4|u>g9knO5sg5Xx^i4UE(zl*
zcWVwFpdnn5`ZQFQJ$2H5q})RSpdU=S-slNW9F1?BubqD(cFK+SQt!@L`r{K2-m$y+
zHSp$|dzWLE%W=LntT&yx;BaQ)>e1F|-n$pp-aDKd&1@KnfPiY+zArQKYw1Hjw7<Tf
zx8r)Cu+Vx<ha(i%s4Dorxawi9VBl<7hyY?ni3~TT!H%ymT99rV$ETG2oacF~t6Cm>
zSDvQ`LDxJ61s^5bZ7Q&(cODY(v4})UuGh9XaFD;}AYx@Hc>{N*5rX6MCM@lvHM5MP
z;6_BSumjaY23G-gm1=z<yPqd9q`buNW`0cKrfXW2vW2%u8kQZH250Hcl8R<hdyoXj
zd|@L}L$M=A3XX1z|G)<Cfeot@800jTL+;A~%e28F{K7#m47}42?9dL|UR&CC{J|Q7
zAGdASH$q>spOGmG`3!1>W4i<{uI7Tfg6mdfhw@I@3Z*}3e)D>Hr$)l-pJ3udpK452
zZsa7lj1h^exdLc#$w$Sz-@aco3lp0I>Vx+>(4A8?S@7`EvNB>QfqLcVN<MO3FwGpq
zNxHnTTdwD1T-s`kOVxlVuF&bb$b=-`3+Agfd}lr3i%ZGYJP7zbpC0wgW1{~^C-2dO
zMVU|44>!x=j>kt!9NfGoUA*PS3FGlBjui?I30%#an=v93iVoUC$yhSpeSCg?HEm<}
zU$#oW((+JEL#Muw>uc7wqujO#8nrHdW4RNGQY|>QF}aDz&w=+rM`BBQk}--)zc^?x
zFGZx~9723oX<W9O1(u%Iyf8pnl=#rgF{Cv9&R8pQ`)20SY%(@M^<kX%9e@}V7|i-U
znj7>pxJCV8=xz<g`{HAtq!Q!?JqSXah60*^CNfmWw2aLKGXhidgY?f|d5IrhN{J6)
z2IVf^GSkB|ak`+WA*9lq6X}JPu)ec`7wK4eN`;wnr`BM6+D&=Z^-V_{GhJ(e+3RON
zPBt~9Oea}hwrrmFPQE_BR^qq~Cg8#KA4t6nU@RJY6}b#Hn*bj!7cGgN(f8OC!BKzI
z3`gpP{AYAW6r3U;%&(b=DLMj`^d}jOt>_74o~6O+{nR+<Kjo&vZ86FSY@K_Dqu~-L
z|IT?pqcL>gSu|Z=5%5gsvDZzk550$fg=tp(zswjs!DH90^~kD6cZigDgk2XtE3Ypu
zwajV8xLG_HBbN%nAG+O~FWd|{3#)&foo($F)OCNj<)JoN8FVKu+<4x-_;;4|&HG}H
zJ1{Cwcztwdj;;QjQzQXk`b`V~#6b=tQWLUmi$8-D^U{<CoNfUKurXU9V_y`IoQLZd
zB0df(l?U!Dzz3rcVjLTs^9)f#O=JZ=wmBN!Br|Li|C$4ZyEv{9v68Cc?bB@L(s_nA
zoh(pT%o9y_Ek!2LI^oX<;97v|H04OTx4_)LqgbOzx#BdE1afyCkcOj-4F-s=`6H&2
zq(ffFbwwK>BGp|q#6$g6a}PVd`<<;)ipBzQX9duQFAlN=)(U*M0x#p@Xn^XQqlQO?
z?n6!{<;mp<Zr&6uO8W=?Y9-P=L44#JCFFbT8IK>77%>*R1j#BaUh80l)d!7?-(lkl
zZjNIp$MSW*7L~c7pURX6YnU=FsrGdV^1JCeNcEvU@_TppK#pm-@3~vmxVm<`I4b_j
z*0mp#_MLfLY{`B0COL-aoZjNTK9f<Y+#q07E2`>Q5%#5pp3ma>my#MLACqIG<WV3Y
zE!L|TF3mioC?c&L0AAO0*Y$cnNE_<0O{SHqkLJ+|OceA>U3EMS2$-FnxB2!mU-6Y&
zhR<^P!ujJLw_k4(Mi)z@4`-hX4Fz2V<$Rl=4mrBaAF^KXX;TuvYG<&eeZ!KFmO9jp
zhHAlMrTmgc;_br0-9;7*;^2J<=eJq+z*;x6+|Ih~M`@bL97uYrnPK0@9)~)3+S*?x
zbZ!Yxbv3nGw#1AFmSiP3G}zo&lLV7e=v?hS*MrOZ9j(S#wCMTS8uQjWb63a<gn)5l
z@=HXqQiXmHig#6lf`A|vKE?1O#@nv1N_KM-*P`K<PG}w%8B?|YJ~9G{;rmhzYs<Dh
zC;SEf4f|5fY%^aUWQAW$-sd8Vs1ky^J6ovH69;?Vxl8d3?~Ah$%FdP*&$if@>8m|b
zV%evp)}o%vH{FrTh)vn0%ZL-ls1b+isVSzt!OJVuK?!*_uQ}s3VlVtU5@%`}62yXa
zLiP{fiB`mpEKP@0T%vC@!vnd#DbLkoV%PZLgY~(L=E2binls~nGS=>l-}?iT){-ei
zE557Cy@vUUZ@_bA-7dD_XYtix=UE}gn`c2cJ2KhJ#6Taflf=6pxb$z1YDKCDgZ{_X
zJkI7o1sG}#!ZWV8psVgd<)TLnWYp|mK798_?bf4Gor@L6njQXqMIhETPK*k%s1MJW
zNH+2?;pL^F!Ci35$+;q3ik1aj5owe$3S$S!8|=ld30AQm8BXeXgJJOzLGp~xl4dy5
zcYlb$4cy29FMT^>;4*;z_s?ln!K{|TO-M6`*jh9Vf}wM<{lv3MDR6JZKslbLa2UGm
zXuZW=_6sHnCji}%w7a>)oj|545ae$KAqMlL&1Au_6$#W#gn)$PHuOsdaM_t|@|u(w
zIFZHJ@e>b3-0)f%LF4OZm{avjlbbtZ+_}M4C}8&A83yo#DQYk{V8HeJ+UZj9=elXe
z$|n`m(xXHDWBTyA+UMKj!3jD91>B!<OHpVEr)}E1nF!Ra;t84|I|GbHn56PJXaB8u
zs!syq7UU;;Fj?UI)zp6PeR6=22^~g#+PL)7y%h22;(w_;WhW#cp%-mIy0ILE@fzd)
zEi32y`hCM-lwZ4<ck;L1Pw<**7r}(a%#BA$k0)4vl7j!RQ0hNtv=vtWFfoB9F0m2t
z7vUVW&y|=dzh9NF)X@^FAG!Po7^k&AkZXB!Jfwsh3OlBf!6O2G{LR7E_V(6F`It<k
zLHMFH=kgvgf?rbpI6v%K{=n<w!dD@r<{*uCS^NYO)eLjY$JsW`o7az!5syhpK3KVW
zA!x^t(NOOmlbM#A`|y;Sl#<I~^Uof2sLP@@aw^Eg-IA@D`i*{evp9PRy?EEMEMUpi
z7!cj=72kh^(9iBOS(9Q0hM$j{hP2M?2eh5#-W@DCJ>;#c{1$wJa-F1Irw!Wu?lP0<
zKNrCum1OgEx3m&Jb&V-lG0j*0wQ<nklt*On@S5gtXJ>CvqI88tf(ILVq#v+9KphT^
z&HlF*ANU>f;_Iu7kCuLkSM+x-6P#$(w_xvxl(-+9<l-L(<>rbTKMO;{A<_QdyYH*g
z(K+ylkR;=ky_09IzOt*r3|SLVM#j1`<L5lz!VcXLaKuGtZWgut>_W}F6%8I6_|)_1
zY<frk!6~my<5J7Q71h?Xr|iE{cOos%+tcXlXVGm}GPfdDlWE?La}BLcK?4Q&MHsEP
zS8H40i^>bq@I6kkL89<urCaS$6wBEINIUK=oK@+5cMU^ji|)WZx=P-WF6x~(BF}Du
zwzNPt+nG9iKV%*=>{Vy$M^paI`<v6h+A&42m?=iBSqPOZcq&=~0vIQaQYp&9g`e<@
zo}pz6|KolP&urE{+!a0{2EjHi^SG_u$!TkixO{`GM$+dHduPts2}=cnr-DFj@8m!w
zBJx7YQDwS|?*<=_?<ghGp-6Qwa>o(wi`%ube7gB)jR8G@8QFdQMU*I!7(`6$^bqyD
z;@N6bsW^$X)Dalp@N|Tcwp=?2%)>I|;|USTH(sB`a2ghsS_--k!#~z{BI7$vT|6;k
z=Fm8X>}7}ny8d#f!$`hL(A7@ANJ?1{1f|d<OPs)Ha1lcU^}dvs*?#~2afD9tOp{k1
z`U!0x5JV0<*jj@XZaK?LB3#-nv<)Ynt?X-{(BhlU;&RQVpMT@W9W9&dOKPNc)ASqF
z8r_QTOb~I6f}8T?QUgs>dIVMunq<Cp#c539AENHdjrj?Q6v*=j_WX?*mAV8lgOw@c
zg8U)4zI9bHk8W(i?FM7`;XBbl{AvD*7+s!Fyt`E=Jv1x$KX7XSfE%)@J4P1mC46Q$
zcsDV46GP`9i=xTd(8yP?_GxhI1sPO6)B%0Ziz1Q49~hwm%b7y`2dcJeOIzC7a3|%H
zcZtx*Wa(B!SbopX;eGd8ljz2VRiCUtE8}l<($~i2vWXE{CA4;e`R;D@r>zP)nEY(+
zqD`7niJ|@5i)A5Olk2STm9vZafVKf<>#(0ZJPm?dQw)SQzV&xxzxS*RTJB5x;o6S~
zi^p?R*gVK4P~#_t<+}&@u-#o3e+Ahp$t)yOBxCS#c(28ifB9h#D*P?;`99wPkuB?B
z5l_cXEWjtA1-G;tG3Rd;Z*U<$3C_IimC_^T9RfW}Dp2q`U40+;t!<?YzR_q?_m8E7
zoRfRn!@Sae2S(i*bi`PKw_Kn1@~f=LfM}g>;=(PYmIOHF!v<h&m*l3CFq-UT709OJ
zi`V|ot*wToq@kI?D(-9yYY@8}?R+Kt`)Q;4ljue(@x0(sN}rrKiBz<Pc1NJ#eVu)R
zQi%y?|J<CV{RoQy%MpLe;?kZNIJYvYNV4eI46Q#323u$*LwV}A6K)il$=ze=0sR6_
z36J7Lta|(W>jxM!`Zvb3**RVdm(@C3(u08)$UG^Ld!JpGJ%7mt<X7rDvai-zr)f3?
z;uHDdp8chP$6T_O4XS)d?}L_{=t5ygqghK9U%4{ngV07L&<R2Ov5~&~M7<i;?n2mA
zHx~IG`du}fqEOl;pY~$*XO)pb*_v`6Yo$>AXrNMC0rMe&8xw3YdMLj_9K5`q9kUV<
z!v6j;4oYred0iox_5Q(G=O)C`GED-L#cMw+|H)@6O1m(AJT~`RTWHsvjl0ak0A&0`
zbK=gppEqaS8_&+2ooLdf0mU*}WZooCePHRK53Vb+kgjN91n@_X5&0R`S-6{7ha<EX
zt%l?kKdweM&)B#}{r9|$Uyqsv=sspJ{z!uV+846+RufZQY*Fn65X<8jhNt?e%9gyV
z7d#wu@%~1$FD{Xn?7O|pktDon@7+zUU_dd!ynSXHV;wF|Ja+RqU?F7Ci?ZL`^Jww>
zXV&YB&D^_#2(fwduoPjFe_L~Q##^tU-%Twu6af(hIB!lI;*?fK*BM?Tu40Q?{!}i#
z5ZqYk_RzNP`~Z@tz|%GN&)sESH*zXw^cg?>cGXW^`T&_~x%v4qr5!C$&8Sngguyz~
zMJazt?09^)fFAXmSurFX3((Z<x7sIJs^lvb0Hgez$@6M9I9;n8Wn#zM0(oOAyRG=I
z{2jMvN1|~ON7gs%XBSs(_)tDxmzKs@7e^So6MRPEOu>cSCa?6B`E3|&TPrm&SHv-S
zIvv#B)y?eAsdz4X*HU?VCy%?Vj=}P|9Ex0|#)Q3o(wD^jZsd*X#4lmz@DXy3(Y@*O
zv!9G>hBx>%#?&DFi6!~XrLl@1G0Lm$VoCjqFB=@>PiGa_o0?x3ewFsKGp}^hkG#T1
z7`HfDx+Rx9d(n0id39y_?Z~Cnyn)P^vh&CJi1nB6A$?^t5>#*Yk@&gCuV26LZv770
zOln?RUM6&ds#G^J#*D=cg<r~v=7;wOC|;nHuIrQ>zD`M8Wg9BhwuI<WuzH@a^xy1u
z?%F1}Pwr3?iDUq)^oM<;!`J~bob~n?7R}zay+%FUJvpbCQeilw(}2$E+$$<A_2RS~
zIyjAC=a3;ei;akTOfz~+=fjlABjCXKO|tAG1ND4eJP&*u?j*Oji+k=Wq;e_oNLS0=
zFcFac5Zg<SL(1xsl@FbeAMBQfa{h;r;_1`sOd_o?<Hnx52w&_Rk%OnqMqjX^8s69S
ziO?%+J`az{c~siWM{eGKvK~v>wJn&BPd=Q=KAFRsA=+EWaqD%V*QC!MEEY9{9HV_1
zs7uH{NgT=h1PD-Nl#j=|p5?WeiU#>y#ph9Gd~Rx;9W{Cs(_-0LAV!f`Rhq*F_?cIJ
zBd1Q&BwHdFdirB|Rta9Ql|hVQ$S}gljj@^eb)pQ)x}81?CXIhY1K3m0vXUo1(4L5w
zo(f^IT<JYg`FAH*tR_FNlHv2kQUVL~i@AFc)Cz(cf*HMX`Yzh;(rE}&jJ4Vfg`HI0
zsKW!T#66{>AdaRWJe6CJS|t$ZBv{VUp<_B;Z`DyY!E>t@e6yWT`EyC?u%r4c<iK%p
zUCKvx@OUioe)4;0I7~0w<q$!W+J}0<f=ne;-}}vC#~U|zFu2Tj-<&TXnds*b#w4>{
zD(81XGSqNZwAGHo$<>N*Yss^nMYqL5awz{N@v_)bQHfIi`yRXA=wo`ra3E27@1Kl2
zvuy4o!*Jov&hhxM!k#lba!2UPO1>_HPsD3?i?kS!mg1aWViOrt*uMJF;VE{5yW>j_
zg2F}W40^C|+taqFac6hkd%ko2u<b-4KkU?n)i0DO@0*Q)bk!KS_%rXhyzd=1Y%k3I
z6R|Ir8y(LnmXLxX)f`RY$J6f@cf?&3;x6j6Q6#03%rhjF4CTdt5qZJA`<|;y14Lj!
z9V_G~Cl>99I`^`p4t)x11Jp%(L2KZz0HrGg=lpE?^ZyK@b^ad!#y~m0$z06Zq*a*4
z(ym(G4WJVwrsF*?L$OZZ?yi{j1C1?3$eY_ecMo?|NREdGoSvB7%&FfaC&|sLFA=fK
zw*J9i{0IK+|N1XkuZuQ61kW_~xM8Gi8urHn$Nd9ZE1q8Nc=7S4s3fSWj??}vM%B1O
z!p0h#EaQGAWzWF{<~ee4T{HAE?T4*bq~J;6Y|$Z#UO&b}%0O^P`Lns)IhAp@N-1h3
zut3l?Rpp9*R*GfEk%}FIlp3WyW4~t-VC)u5hYVwRL}f9B#3$VGzSz}NHKuOSU4;)b
znzEymsdz61Ka;#Gep$LgOI3P1Vm@1Nox9<oxPc;Y(_9oZIg*1r=iN(3a!&Ew^HIFb
z2m!4%!OzH`&{~uHLy_oizC)4~!XHa<UC5Fm(=~n?;C<bw>_EdjzCZGR5CWQ}qN;2W
z0%UM*K&gZa#guM!wL=(PB&Gs#j#N@}y}RP|*(SScYIO1#qe}7Ima)Mh7Uh+$Ds0`o
z-~9<uhKCtrFtRLw0#I6)Q&~=EsaS71cGpj-n$o4}+{8Sbn1-Hy^ySp5BoSY%X>3Ju
z2^SL&r+YeIQ7g#KF;6qP+0k`7;yfXv0<BP@!89w3$|xnMnl0qSJbS9HMa6KI^;(qD
z)OEuaaKRJ22c=36&{?@T4>Pq0JlU;T+l&bEuXtAdGNgb%e>E^%^c+u9IVp>BKZM{(
z$RgknP&e&4eOWe5^E45Hr>a{*DD^p|6sxWw#Kb%mhXgG$!FghyiXzTPoO2)yk$}}L
zU8^8^jt>tQW2tM4(TW%{DJD++Kuu&g^oTg(rh&1au(rZjg*gwLl2R6<&S_?zJ$~{S
z#kppe*i<bcIC6{_W9gdAvsDp6Tx=?YZ9orbZCF<gU)~)#91i$-=HcyIx@};t8~*5J
z&%gcrD@I}X{lE2h@iF43nRM=_iQ`y;#x%WelKe&diWC-%<A`b-qJY^&K7MhHForQD
zblV_iM?4MO+&rKp7%izg%f+@Ur;WZxCxJ?W{xGmvHFQmbF`AQr`~3+QJoFi2IYKtN
z%(+-smZS|rD~v3i<vADMo84}Wtt*<gLP}No2&5)>&*A=Sq`Kjw&%VRcC(k(C+;Nh3
zq~NJ5&6CSDcQ0?4MvpZOfnr`)*9Ivl>K!HU<H%3`?2q{mm)F$wikGi{!f-rtcXQ8g
ze*gPiy?BN(nqeH7oa1VD!T9jN&C8eE-rRHb<SF0#jfStk{+#hNG4&2x2^4}jC&nAc
zG<up3Ie-8$xRQGgfs|7jj`PmQDQ80RXWNpx7)ho`3XVo9E-S^u<nc<8bX7WlV+kNJ
z1TvH(JJ9W}knKn4JQ8oeIQLJSBXrKV;lwzO=w^j6_G}|x>KP)sVkmh&Z75SAO$}tE
zmV_7(F%nWj30W5GoRCQ3IUN@f<Pgix5xJzEQfWR@N}(u?jkpOR%1mn?W`t@W8^Y-u
zg!kn3d70Rmij)*>K-UNz7!UVkA*h=cNvSUtqrDGvCIb*kzi*Btm6KnH5g#I(>lbL#
zaJYGcl!<lMqJ`k;(-%lv;ZC<ym0@jH*i_PeVIHV;TTXmg&@>Hgy}}2N)w=W^8Hq?4
zC1gpd#z@MsG-As0DKD3abGEWnZBv4-h`6Ch^My1}wG7^KdpwlIHF&Jj%rP<dM@;g?
zM@uNE)<{#6-Hgf}qXkJARMnPR!h9glo?LIRO^Z+gyo2aUlVoMFTA%A!kyTq0(Zno3
z8geXyu;P5;`1Z#<d$Ps0-C4r-0oOsm{lRn9dvf&Uz7nwNTGs2*)R=NA-TJ^h6`S^)
z6DdbR2-sY#+Ea|QU3a#|iHIDFH`X-by~nEJ!tnHB!_(c0-FAgi29TWkBh%QkZX4Q0
zBPh@5IP?tt39U-N8v7nn=5Uy3jHFYg-qE%dMoIQ}2b_<@q&N-}2v1uHh>2AtsC8g}
z>`Px)2ImsB)!16GTNzZ{a0>A0vumEL?{UGALe1nOQe^yaK#-}LiZNBx7guN{@qPK3
zK03S$=(VIwk^?tRrJ48_eEz@k1d>HD&~q`8LQ$<eoCa>++@Y(A6g`rRHkz($@WC_A
z6P5RLRfEZbT1vbs5=(0&O>3A!<Y5>%O(V~rT_8P-hXG*>Rc&ymfh00xNFXK7mG0rY
zhaMLbb=$J;R@Ab>jSgdMNtN9n&_>d>ioQS4t!ku9B$-Ge(=-O18@@OWRCUMEoz4qz
zU>HU+p5v)!b|-e5nj9QI`tyH{Fco2*al_2p*KgS$ZjtKSSW}uIKmW-$eDTGPczE-g
z&wuczNR`=awxrO)%P&h_I%H<=+0+$HmgHVA-g|Z*{rWubxp9^V@29p&q0q)MjXm*n
zz-WoG4W)-lqDprjLMWQ(7)f|#K$uc{kh-{aDnp1Fax7+S&SR<$xq5*Z4y5@Ga!DQK
z9I#zS6cVY;d0c3w+U^K$K*V>Bx0I8Eln7;u%}Vm6c~63}ikv(uLk?vmYiTMEkW0ZX
z#IyB@pLl1ES4oXJ7ygqFMCXecp0Q*hQMJY76>hk}`#`L&Fv^gL7!`<9U({xqGL6sL
zb7#6h%c__(exMjDgvc}(hgDK8>Y^m6*H<WA)8F0ThZB|2SZld>_7N(>&D&SZvjY;V
z&5G;QHHZ6K=CLO#iEhd`g1V~l!;us`vZ_eC4N{mgwjh=JSx|Pj`5Ygm6ev-wZ2WjS
zzqg_$a??nR9%F0f7}=i=^wWTwN~id4dqwmUe(aIK5n^2iJIR(DJTgaGTP(_D-I8K1
zMwi55TR!eVR0tH=I}jPAHEq)%MX?L_-a{@uJ>Ew|QznQaKr2D=GnEl^m8G9fA9gK-
z<ZyW4?aQAMV|YK8TuVu}TA_?VNmZoLTH&UNp&tm|p=9aRc-QrAjQQDrhd}fWKTU*j
zMCOb#6?JWCD$S}=1V1qhnR%MY(bKi%`3u3*T1A{4MhV(RA<zWb;HJQvhkH(g$80Rt
zz;@f9jpVpLaX6gHDYLRvl={i5{YXWot7>i@j${NrIL0}kq-IrH)~lA)W=*xdU~_fF
z+)t$B`N>U%l^!1j_ftHlw0bHfsg-2Y*2VrL=kuwfWP2iIlUh8<IEy*{bAE-47i3+I
zSJ8X46f|AM*T<gd1J-DQ^LQV4dv_w3vV*thv7r4h^7{Uc-L^$>cCHkOuOIH{M^Eq$
zsRf&>6($LU4`7N3;AuRut_?m;_%M@1arkxPj8Xz+i-gfnktut$opEk1-L$ea*#+10
zaQ}v(KfRxoU0I8tCwk{-w>#!xMnoXYoW>KwP(HwzLU9c?f{f$1KasPbKOS+>^Xlan
z{M*^_^^g97uV4NIYa5i5^uv+cx3Bn}zxg-$?)QF+!~GqvUVh1R+LuPGkZ6vYT4a3m
z91bJTUwqh#B}`prlgog%HPTdN5sx0{9ZFcNE;E=@l7zESn^G2`5=~W+R7QlzbWW?3
zBBnfdatng@^oM&)-65+T=IXnMaaD$L9bc1UL~F}-SEIBcTZ?YjpgTl7r^OI(wuC^)
zGCy0K6<j%j0Hq9JKIfx_LMmN$;eIZuvcy?ARpz@WRSYBj*|#g@MAej&O>h&K3SC){
z8r7_6Ooj9jmuK8u+!9hMyL4qL))!Z#=*wv^gH*+<>qCPFfD5VU?sG!s2)T^8?ONDB
zJP_Sf>Ht1qH!FngIKKUwX&MQXh4^y4V^!CT_xDA;mWq+5Rti7%WSLoYEp1&DkGWi?
z?M0E#eb&4m0i04qi6UYUDdVP|c^c1iK}1QpP^|2obHq?=+<hz>zC(XP1V{FPQIxcz
zkPd4MtM!IfB&^67W0AUI8hWx2#RXdlgbet3E`@ca&T70&-8NWjIQAnFff9<~CLRta
zR`m)iB}yx-wT%66ZrUr3+xY`7hyaqCn2+~NLr>LhF{Yxwdrh;uVD;pqvtT14&&o2T
z6(I_yxtOLZt=|t!nx~njt`I2^=DGY6619Rb4LA$;4|^^;#giu+l#xgzbypMSiSc*@
zE9r-HHjRYMx@Fxp^!uK}{gEud=gjCc359ck<Dth*Q*liYMR~VgHB9$=q>|K)=I-d2
z-Aq#%ND&_rLMN0_XkCJikVOErx>zCZ9{BFA;;Ir1KJz>XUOtSx$vtzwFC!vSCQjvk
z@Utsp6ND!CGSA4*4x@j$Gx9GFAQ_VPocalwGf%fGx~}E;_P+QqSyQypE^s>bBqh1J
z-f(fTWo<P>Ke5}EdEv)XUpi8oisNZux9%Vq?r$DS5pCDvzrID28Hbs}e&la__AT1V
z5>vp|iX2Pt$f~W0A#nfjz-k?7+7=-*Rcpvf@bK_}ohmNY9d&IuolXcP_~^5bNIBu=
z385tO*`7ly&BOjJ%1DGRlgW}2sJfhnQU;`8yIC>M1Dqm%`5*o*e{k0`_5=I<=e&M<
z<a^)$+kE!fcliVLKk(h}{|?WdKIi`a2DGN#Y!Olxn-+iet4ay)Kg>QO1Hu@jwb)!1
z*NiH@Vz#5QT}kDofOj)lS<L#HW8X7B+*1jG7MaFMe9CB9y5VywmU7k_VoZ2H6<MuF
zBxMNekIAcN=p3Q>5>77>p(pzp*>vd5bF$rX`{orYooKs~bC*&RV=9*3LY!4F#e1q0
zQza;UuIrR>0VyTIdGJooC2+jcoxdxlW=i2Uc#l+|ba7Ahemc+jDAw{qTADgSD7H9d
zryh{f;Js&d^SNoNH0<fBV~(jfVSP{$U6jJTFD|IoNR+Oq%!HjCO;t4WQ*hX}!>7#G
zUwwge172#9kUZU8K!(HZZJEh|D4$IX%)^NwC7Vr4TUG2fYpS{}GfU2=j&QagS3;t+
zD9*boo`VT+4%{&oEBg=v&U<|JMOv&SDmmVUNN8$eC@CjltFup-CT9TztP#|vMx}t!
zs-%jDxcJVIK$WSMB&3UErO$G5fwi8hwp4Y6r9w%;<O1ouyVc3!&w`3$UzB4xZQ#8)
z{nuZLI8TJbJ%^WH5lM9GHNHP_`%3d7N1AR+$O(+0u4|Gu=M0&AKU+mgf%m@D=R_vX
zr4cmxGPSRmZBDF2$MH0f(uUd^nx?~zGxIbv_9slH*q@7psVy=kDr3ko;wy#F<^3Oz
z1JmU3F7jfS5jmGBo^fUxN`_$wGs8HSd(>+7{mkq$=uBrU$@QgaQpjR18Un(3w6W|T
z9?0{+N4thrY37(o61KHsZ8Eb91R+q_7i)SS5vgoGe0GSe$PSDwzGA=ZTKq~cEJ7s2
zfXVRU>Jm(atcsrfbT}b%u~Jq_p_O3r9v2fKOSY>u+trHW$?>Bv-w;#g$&(A3tD>M<
zH7k6Kw9Sg-Baw*JmON#~!6St&ss~k^AgikQBXwPcGzzq!?J8<(fK1a=sHWon{vOC|
zH*2b{0VK)@>ZWG$GiZS^8b6oeJ=ae!FjkO6M5u&N2~{iBT}!<&r6^fjo;<nauzw)>
zz-fQv{>?pATOrlR&FvTPqs+SM=&T~S6JNdloQtb#o_+QmR<`C3{>>jVJ4aQQ@sWO>
z&uO6#c}j|RlR}ct>JZa(B~9bg+4><d_|gaz{Y-0JnNfc3i->+kXQ*|#<g+UdtU{LQ
zhKN$Mj5(rBMO{_oP>%LO7}V3>MqYiVjGWC0BxN4+;f|qypw*Gt8KlT)inc!Z8KZ4+
zk)<qZ2+cgtq#TO*VN9enm!1e&rtqorjz>^BcIEqeLnMeIwVtQaajTR*_m<41o?<eH
z$k^W^Wkp>*!DWpxx-`3a2f>%w;VF_*NnzpK*_`sj$O4=X#2AZJJsK?(wU)%zGKN4(
znyOo~zq?_7d|=yl*t%w=EY1hM{OU{e)FX|?)^+JAA)T#-OW%eu25YMG0z;8F$09UI
ze!|F-zY<cAr7uBO$l{HriamQsxwyY(iraD)w9|<<|Lza@{LP--cfQ5tw?8R%F1a)Z
z`VeTWVb>V8tzp%4r3c45P;hg5$G`et{vE&lJHNsAzW*%*(ALGQ&;^J>lqsK;q-!ku
zzC6DvdW=?-E*XiG6=YXTKa(R41GQ1O%6!;$Kv2yOTno#SClynO^l`-OI$qvB@bwS=
zBTrs@3#tZj9`a<WhV%Fsv?=e|`%vWKqMUG@^UTADN@q)uVdnXh4NX-s^aBb>TUYEK
z_5|<HQa~<B$Ewk^l_>_6AyU_tm?M53k(7qdy0-Mg%w7sYD04S9m7q0(X&6aK(7TzC
zGgeZDaMgxLA|i1~P_-RZrKxI-X{z_5t-YUFtva4RzasjX&1$CinO6@3QbaZex~68#
ziT-dyjtOfm^VE|mDY)!rM2fT*7l;^ssj=m+sDWh2Ma{D|hRaRI$<It8fXKvtMp?tf
z)fyy(<d~0v)LN#PQEgR<xwB_~f23WteCyMXNVQ^jeMw~-h?y7HPx$(qud#JS-L@R|
zJ^e6pbLjc_az)#=%=1IZp9;bDa>IJlv2KeGQOZRj;oVHtSY)ZW5_3jN!{z0Uap>t!
zM|}3!#?Y>7vP(z}>#LgWt}E&gW<m@=z-LL!k;+=SRm1huOMdv5Ul4rc@=42Xni=PS
z){5=6CHY(Kj;}~Du)q0=$shUTTffWHKhWO{#f4UYvJztr<HL!R;xAIh{bvXvaN~jD
zbS!g9QbPGiOhs~-onjnL%rQ`RYm7)7zxkZ5?MU8{gv1yNDPm$INkdYZ=n`3Kl+uLk
zA!STeVQh=C6<L-5MLN$((6-o@q)40x+_+~PdSWi4iKQ!}Z4t$eU7V%9AknG<6bX;e
znS8Em@h<Ha6H*(Le3w#6sGQ#i*z;^focHH>9*N*1-WP{rYpb%D=Zvj81QK_6ja+41
z-Z7kdY&9TcaR!c3AadzdFvb*>*9VhH?|X41E|x`IN=ajM8Ac?_N$U3Q4Z}2IjX@ht
z-PCAPb9eWE>krr*NN!@Qbs2LCG}@xIX5FodDQFC(FUH67Af(c#Ai{{y5|uK^806WO
z{@v$QniET(k35@SimXsV(8Lk&TyEAJzI;h-4ZClDN(#ALOL9?jZ93L%LsM5^1;Gg(
zZV&wXKlyWR4@X|Udd2s@Uu2-_{2Ihr9VnGR@2)ru18?q6W!IZZF?To~dHdFK(N)M4
zu}V@Y$=naDRvkjb4|1xGDf55++n@6He)~K0!_33sh!>hK-<;?w&5z#RbNl8SHdoJq
z;#4bbK<gS2V%bE*^nT<)Ra=~!@NPcqQ!;*XrKVEXNFiyPhHkZDf9#ot0VzStQcqAw
zg7@s_6DgE5={yYhxo4hcDw+A#lMU-i@pgYAr_4h?;>MoXX>NVw;W#3Mq^TsUMq?^P
z%n4f=j42XcvLq`-WhHfONiox`H-rG$c{W!YrtySP2BRhMG*cVHuCY`(VPvNErHAHt
zdcdlRe4g{u)HTk{G`iFXa~7rl>aQc?6#z=0gk)1!v{v!zc;e=vC&m(GwH8<-Fh=ok
zD)oue>BQbAo?mTAZp-m*&vAd^d%yb~zI}C#SDAKu!Rxp8_;F;@b==ksrm9$NR=jz6
z!>*~g+|*=Q@{-d$F^_?!GCX;{LtDxAV$J4qht-<<yF26&wpS~vSEXyZX=<*ouQ=X6
zFb^GFy=L+wo3`Vl=T{u}r()2j3{_*m!A5Kthl&0)QCW*swmkS+Vk%v%4gg&#R*gMB
zq>7XU6}ITQ=TLmP_KqL_@ZWI%^%vZ{euGS9n9IXqptYKD@Qf$-!Aq9&85h$z#~tva
zXL2W+ZgZX?4X*F$`#tN`8e<IJjYuKLvbY$N6DlWkPP9gmq9tSj(U*~&F<?!FK1(lC
zO868ZI&$`<Rv{989`VzF9}djN8~i+=R4!v}WnCnO$g)eGXSAuvA|g|KFKZQN^<0+4
zk0{HGM3&Q#QWzy6#P`z#b50}#QmC_YZaNPI(xomC9OQ%-8L17`=1G~GaQ7v8C6M(M
zGR)i`_L!<dW>JD=y&^<`Py9lwhF?m~Xk$sayq*+Fce~Y!wQQJHiI4}>9MPgFK3+z%
zZW<0r6u+f+4#G+jrV$Yvw(E{sE3~QEY_>E_i-^8hjOSd6*TI)GVh#jK`a;G?5(0!O
zO?WXD-zHVM){QQ;q$I!yNv!~7xgREswtV%MUvT-+C+Moi&jVfCux{71^_r?`sI7&p
z@B+U2!4LWA%Qs|gk+Gyel2Qm+kp&oCy5p<HQZ*}vVdjf3zhRgiNhrFuLyiO9&qORr
zmJWNZX7>97E<jaxd`LbqrNnRl-f!`D{{H`g(`n*(JQO$ZAAQ62tCxKG(KY|mAN_#+
zonc+CP^Kv(9zszS`4I5a2szWPR!EW0eIFi0CPZIMBr8Q_z{|uqIez-`4Ie$bVB6Lx
zqsri~6uG~>CHtAK>xh2F4JQPdrYiUJN6#)e9eZN*jJ{>)Po=>(XTH8WGL4Q91=eWh
zcy=8yCEdGT)g>LM4TnR|zMoi~ZC4NiV=et@PrYfGhMu-<i<9;|5`7W4Xpu28(-=jq
z6;kxvzkS7vXWyZzI(+n~RZE={VlD#!(>#~W?l1WKf5izzAhbjZsH=)$_I&;Nmea!t
zZ3MQfFjms+EFWF$(60q=UhfH&Lko?}nb|pRAA06F6bo-R^W<`ePJ*wGC$e|MxhKvu
z6>za`$P!$Fs_R(qHtgpCQ`e;Iu$AG-i>GBLB#H@N(^R~@b*$GNPo6&E=IsM*x8~VL
zPia;)sLXn`VZB<T21#W#>$brSW9i_IV3non8d8%u+&wTIW>S|iQjtO;hLXp;yj*d7
zvZih=Zm$p`ptNAKYEW5mJNxslsmx?{?nH`?91|B8YbF%J2%;=D;?c(sE8J!<)>73S
z!||RJ9Jbkjp>)y8gsv=CPo5Tw5T&TsPr1AMhL@lJdscNv+iuWRhcq2>+Cz%SoQuKZ
z*{K#1T8c6)JjV0M142B{6px-1OBy317>;`$UVTYpBy=soIS?ADOc~%Kfs{yoe(&ll
zlK<ITKD+3qbT(EbBeh0rTe`AyCgrk;NTN87N>Qd7at<omNI;)$(M6H)hCu2n^wmd5
zH<H4En6Ge1?DmS?<u#|<H%M{r{xym)I8yr103uMPD{~v>8Ij8PQ&W}M?ILEbq%5}P
zX-ld%+#DXr(P5<|xqw!R`s#|#C`5>u6liV5X1$|oT9h_aRsH@H@ows6IQIf+DF8#r
zWzL3>85trdO+=AXk@*@`+W(7xmMRUEXQzdn6|ZhjbWKHLH8(&0ijRN)chI$<X;w&8
zk<sL=nCIeU_WJ8r{P}<U0a0p_D$mpYcw$gB`B+Zb)@mwisOCTy6hcUhl7zF2xLH|j
zZHTE@@n~BW*I_+#e}BuL|KP_owJFWAzxLV4*lxwev!~RXkzKgpa6Izu6qx-)7$&~|
z>HnXy|LV0Z%d<PN-?YnZcigq%GczhHtBS>>5xcrUP$JNf9s~gmz3II_g7lq)7W5(k
z0wh39b~ioQrE<wi7e1~zZrkOi)x+Eo)gZI7ZY?14;YP;2_nfoWG~++UZ#?1Oe11zc
z-B9N%V#-7p@opx%k>Cp&r#~I(Pe+U?jJ6btwKw>QWGqIDLbg8FCD)aqX=}#G5q)4D
zj-(K{TrKFw(hJ`<bukB|Lb9G_$25DUX-4C@UN<O_7zW9y2c6Z~B`BD^=f0mg`pCMm
zR8`qT){6o)y*~|n`RJ&tn)hE6o|O`%Snsqw6f3qdNHx<nhNgGu#<FTkHFR-5X72Ak
z<LS)@)Llz$40_QKr-^C0LuE<yg=F!Yp5Jccbpg?2E>}xjhPT@zDJJZyK{plF8dOu`
zE6HNra=1IuXa&}yFBa%chkw1Lsw$ecCQFU3T84gN7-poj9FIMH-&1SFip+fK3zNmW
zf<!igs0Cia<<*)W{_uzF4hL3MO_de>@x<eHhmwNB{)E<=CpXtzuCIwA5`3g-G`2E0
zFY(bajx$+G7Mm7r&G-EQ#%R2cgfWnPqOB^r^@5}%)|le0uPZLEE@1M62yN418%1Xp
zJaiV9BDN|ma;7dX(lB~-Q*mPqp5pqLTx9a`ot!2J$bKM<j~LUS+x1y;$v8I?eUFqG
zdk)=&n3<<i8C-X(g1ivr)N6DZU^_pd$rw@CP&ox8iRfm0EbWguMx@r%W<l_!YziMy
zLg2@N)BW4>K?Hn^#4M1yNH;zObP~kq&(cZRgai>GHCmO%wV2NdS1G<Pze?XHL(BXl
zr!vSCqAYx52NauV&X7`(kG^XejKSgi9kyH2UOy+Cj_ArTdW#YXr3}?#Q}Q&n!d5NA
z=}_ha?-)fNJxs0;aC1s!R?{>PGBP=8DN)K0Q^sy?uyuzU4$LV~C&4r-Hj5?ewjp|l
zj*hCbWHce$vwLhr$f+dvW27IaV!2%{kzL__oel@?4~4301n0a31R+sc)0nb=yO=nQ
zo_-F5l+gXa;L45mG|glw_~!F3Xs@miwr1!ZTeUCU<)W038Krpq^4I*wuimnL@`98i
zx5o+j<t=Z=i1YJVLKCG2#2PNHu8?`=aknSA8EGn%)||$`!@~naFc&L^aVDfhf9&};
z|N7s8Zix9E<{&wvs+wlGBm|F0i9FA2Ry7~J|BPS%;)@c3)&e(`tV1zg%uM4*oQLAT
zQu2G8N%CiDD<qOH^aqlsRuZ2e2Y2pa(ey*lu^-SvGMz@GpLlibvFCn}X&i_tb2yEN
zl&C7h{x~3kXIE=Nm@!&%u$no{rR7m7q=e%*aU4C5r;%|k0phN<?9K<P^=idecSrv0
z)e+Oy7+pU9^YKWEfpIr-adW{`Jce|Y;gln7U7@l>q6pcOIB?j1#cJ^qtt~lbax7do
z;b)SY&+p{lGKr3YuCYA5Sd(PN=U8qfQW6vRP(nOe4lLJ-7aw2p%+!4L?SZ$qd#Dw+
zk0%hBAQN}{EhCA?Q_q+8_gvI9Z|)y?vpcf1nz5hQKOA^@y=K)_xYL2-?LGI8J05S3
ze0+05BC@U&KU!Pp7Q7xE5$Nig<P*EyflohbdD306pH2m-J|)~3A!X)yV*9WoN$@(e
zxvp8P7dRKts{GK~&eB@LqG?#JmIV`mVjLz$=V-f@p{lTEDq4F?++4NPU6E=I;|Xmv
zPcN^rjbWHZ7M;Pg5|b;Ghfp9K@BG9ph*>kuBWXP1h7(3B>bgT1i_FFL8iHe-0yaO;
zv`apE|I>39L87Yb!q-s}iAH76rBRCzJ;US}4v&cBQMNfZ_7w?48H*o_S3gVxcdx$T
z?wc>TyjmelQ%+4v5~9aM0~tcj<sd*oM&uYs(V>(^FV<&C%pydYv!pYTpui0|W91#W
zU<u1SB~vEYnX}{SugZf2XL%=;SwRp2Wm=-_(dkIaOS~I#Zo+mOW;c_y1_8EdzVD9z
zo7b$&pZ&$pczAdy>3@+>N)ts%KNmn?L8m|_eEXIdGg=C!GrlKHj%jKMs#N#ZmBv+;
zyN4aKFBuc1b@^IjJO?%f+E#q{(GR#V3;gW3dmMPYdvmr#XL1N+1hT5I#*}ok(&(yY
zf9UBa$5KXuyW{cXSXP>cer65>t4i|l>NU&fAF<u;i*rpIf(t06xTytqzy39YPplDG
zsdzYytS=_+w-fz16@(1>99FA2IRSBiIZ$Z{=bXea&djdV3+J4X87f;bY0qxIBgl!(
z`sUqUj&q;8AcM%PuCA%Zk?10CyM|gTI%~P#J#cz_Bs3i{xPtf*<=_(Y86lHXAw0w)
zx5_gE&nV5Ztw>X1mm;U*$T#2IA~jS^g_M~Pit{fzkJOrn`z@7%e0CvrZA(rLr4(12
zB|;=-6qTu%M#qQGpHXXxC2<%gLQ2Fa7=7e0Ow9AlJjZeyw}$tgJfW!!Up=1CU%-VG
zG_6M3g6ugw?pdx{`qPM38h!q|AO#`;Ibth=tu?cI%{-Qyk4TX`opAGsv47x;x3Abu
z=Z*G1d46jSMAVw)qG8ci+>S@EAiN-sf$`|DZNvI$#TX-LiYzvUR2kx%%-By{xXjSc
zSfj94Yi6T4o%V<%2|2T<D=wOrFGE1v3g<KP6nOG-LllX-`+MHJe&FumK!4w(ZmyUf
z9{KS4ijfZGB{t0r^T5-m*XS%+wJWSJm<XNFSk5HSIPp{t9HhsMo;e2cH1p)bqoo3?
zfb!w-E+Nt7?>Qx8ZE()Be>gGpuAD4g!i|Cb{=jy3;*$?*v^7j~U~-;KXEBX|TA&*Z
zIx)_UsgKms5EYoJ;K$$Pi3y}_2-BYJ<6X&2D9JiTtTlydEDVEl91aJ%aR;j_y44k0
z8+`ESrX$ZI<9H;@bJ-N;vCIR`Rk`DMB85QRG~W-F#TnEnWU;H}MzXwl!D;uF5PPJt
z-`9u+H<Rn;%s-06SP(2m3370y0T2r7N(iK_fOA^>eCvQ*@(V(q`#tm-@lfV$w&}{P
ziHtn|J0VL*5h!z%4AISiM0ZR4JkpttA!_D+#MCQ{wnbk~j_ms4fO<#W{nuZ8%NSKL
zZ1}NoxnyZtw7T4u+NOhC5Ijnm;>2;DoJ!@ZQU;|hByrZ`%baJJ9moB#&~vqhOmW7w
zElC=V&S59VGz?5W5~ab#gw~=Asg<R%ixR?}2U3L28iY1LSvdRCNX~-yu2u~519O&a
z77cE{qy0g{6e2ojY+V%zwzX`Iuh<e<ua|UngP#Tl7qHr3ZDtCYoOF4gfP}#L86%)=
z7HDQ%Dr<s7uv#vun-+w^D22(IP*==;EVkRf#S>yEp9f>;>KdaY_piRCv4Ul7*en-h
znqo+C^Eq!)4qfsLQ!Ihq!jTKUbbe^gK}S>Mx|M)$=b5n|IaHSAs>Nu98wc9TGNpi*
zk<)3SYZPr)GxcNHVEoKDj)*L<R*{mRuIplPu5D>l4uyl|lBKblN=Z<H<P$DM-t5P6
z<4-+8C@$7(YBRAP1K&ROJc$*T8_VtOBW_O2!_06RX_}VGD3r)pZ4o))hpAAxLqg=4
zVfYF^-!eJF=sm+c())>DKi)Ej@EZ+3zx4(pq~v<Jz#7Bx(Gh$kg`&+xpU#>zRCUGN
zk4Tfavol;6YOmQpoEQ%yi;D$^;4!-$$GI=k)$Yh+m@(-Jn-$mHf=X$oxg>?9Hp~fj
zbKzzY6W6N+I8W9!LvD$EVAC!6aJ3+M$K%5RB{ZMC{g$e!SuZ-G8+d<Ru`&riYcAx-
zg|YnE?Sa*0gBv}=ene^mB6Iipz}K%2y!Z4e!*OH`iFVyG*M^6?1E<HCFlYKPaQ}GZ
z^|!CsZjZ<;na02{Mk;ODTy9veHw9;%VA@T*zB@9Hh2<4;#<cdGUl5}taX!%)$>RA-
zhVeuelIq+Hs6s$WP2DcIx_Jh{;hkp~P7J3b(N8!(BZVp6aw%E0OE#uLDve75wKL4&
zzVN0*E})@YPGu=2-0X?dT!N9u9aX!)4WVfCQsR?GpqS>uEQ@|dC<zp9moXNRG|F0n
zn{i4|8H+e~ILF|N<}VdeXi^H_BPWb8rP4N~vd5fz+{w9QU}Qn$ixkiLZXl+@psJgW
zEP7-*a`3Qwc!P5zZM!5j3$zp*_lL8?>)kf=rdzRCUl!Zr+?PQm0yigw$mAr5F;UeG
z+LU?5X&gy_(S~+;$@==b6byNP-dv08ar97kg321KF81btpzAs+YcWa{Uf5Ya)W#G8
zMBS3l^Amf9O-dn|-HZ^4+FEi4=~zO%E=FRAe0aHFxoUZO+474o-tgwvpHX#7x@XU+
zx&;fXS&P7zuf8HTYd+{Y7Rv>PLE{bw>g5JA9qEKR9~?@)2r1FVQt8a9)nwlj`W}R$
zZI^gIb2yz)siv0Vdt`r%85Hk&wmNq<D|FRBD!0%u?Wv7q(KRI%9s*+yEcAEy_DBjI
zeJ1lNr3k^F(KZE3Q&m+V)TYF8*|1zMNQV(4Gttc`ky)+Ua<bRObh002lApO*c07B!
z!8wP~%Tlxz9MTvPCtUQ@wjz;HMp5O6NRp^vI)RpgtIHOPBHbOBCx?$^V*^2JG-(cu
z!$_wUIVC>3JrcZU928%@e&G4l1&@z=`a@3+k!IPV5d`O`lwcaiax0FG*TaA@1-N*B
zoH$JmWi8e=EZP>Uhu;`E_^lg=%CM{~QV9BfB+h}HB`M^BsmKuKNJ@rg0d=LY($HO0
zw7TW-)t0fJ$QK>E+k3oDNUIsVAdCU@%<Z=WUwwPT%qO;5F-;DkBx49n5v0*P-L!1`
z$YR;zNX+w$whnhZ5~~$|^uv$&7r*?HKl!s?@X^Z`JPgHQb-iBk-db|CY}pTz5DY6-
z(Mp&G&&Ab(90l{}QKoc59QOmm=olu)@ifBZs2a;S&g^$R{c&RIXNJ)eo#*k*j{R=H
zjgG_PiLsxVr@*q^aIw6mRV~voFnh;zni=O*I<A$VX{wSzc!#G9L=7=Jq&C>a1tBK9
zbNIgJ)wiEBjUyjCe?iqQ$wGoML^mOGpj|9VcrU|Zb%7KG!Y6ZP?hoj0!(wqovTK@d
zfjD=KAWfNB*Z?QVFb}->@-I2uenZ<f2&HjRpfU(y(B>Rg6oNqcb**(lAy|usVwp76
z8U0=o(aJWU3@H|DO1aRK+=vi@jKWj~H=Za%>{9iq&O?8tiw#p`L5gz;BdS8;O+uq=
zO_)a(tLLm&*MvBdlgAzInNmVGi=zL&!^u|&LAP8{H*J~m<QkD8nvBMje9k$qky65H
zLDN`FRWnWl)>c%_0&8@!Od?34+(Kl^)J-lp6Qel;+VG+9%|tHcS~=ws@`AHmR#NfS
zTEh9{FNG`&F_8(=fRA8|;S>V<eqz1q`0T5<ytrKPgO}Ij<hVZ$Z05h@r$2m#6C3*G
z1=2{OANa}d{*XAm;MHzmx9`!$(zOj!R!C!LtBT4P{OlOqT#y`6A##ByYEjdsM4(}u
zBhESQA0MgI1<iU1K9ui+EcEGjee!o`ot(<B_uPJ|bL3*Z;MvVJ&F7EA!^C*lg0|@M
zpA$pC2|-F_<I|#iPn)*ID$QY-5JGTuwWixF$--g<r0GDbZ>WuCbTf6^aC^VQ<;43>
zFL|<QQBvZZpj}jGA?fFt$vOIAq_&ExE?J~3B*s)6cO!1<X}g-Mu3?x*#(t(b8M@YB
zl;-a7z!)N%%3w+C4?VkKL~2bxN)G+R<Q+x?GCe$wG%}-<$EOp$i`<{aGBfTSjR!>L
zG<&qES#LHtmk^Xz#w=XfpZ}U=>aVx)N_g_@YzK|8kSV|O<g3CAN@V)&KnB`%AxhSA
zf#{&CJ8oY;lsJhL#2lGYJ_p$Z*+9K)m~x_e(s1!)!56>!hS^7Ktr?h*jbMGzu&{#Z
ze#WjFLIZsqxX{<Q?Jc|h4c8y6_<Qd?<=_9$Kj(PbvRp3dg6CN}aI>!22hDKz2HkaJ
z>zT#@H#t)8aYJGBWCYm<I%BDnVxBy~6+gVrg5$PlzZ;l`z;tk=ka@aUasTy}hoR@R
z-?QJBlfu_u-SPg1A8@ty%qPb@dG<qL*Tj%<BM7U|t$fErmf*C%#~tqpDPa~Hv@Q#h
zY1i}*uX%d;gv0)Uh@@UWDb5lywXwwLNJwHVU_B8s<1{dk$zdj`3T-<U-IA`UkV+Je
zl4;KJkVfaa*ny7&yW6)syuD+&T4K8n-Lx2~u~kJpQ({vp^D;z6YE@WSDV8?FGwMR-
zTvFvClv{_f<ODG}=2$G&WP!2u`BW=Q;#^9kvS%c_M5{ug&19sK$Xw7T<-bwI@>+LD
zp_$zbx<Z=;aXR6~9$VMtcJ!_%UcgOfIYJf}SqL6;wtEW%K^2#d6dA2#fgEa868Ug0
zWttg7EG#98)K%3rZCCO;MnjB=kTUZ)5!?)`B?!rKv7qZJrnx~RPg0p__T~3i8h^%;
zYilv>qELthpcJhVyn5KPy13@q#fq*nxMAd1-@NAJBY*Ux7oaqs|NNIc)O(z+xmYcE
zwr*HAE$ws5$=u@}Z&|mAVRi^1&&`ijsoI@QfmFswNSR@DEUFS<n}g#x7Uz>xikK<X
zTv;s1S}Tsnk^GLXIC1WmkxC<#Bzz}P<wSIjI1jA58Z%62Ey+p{y(4&Eq|jV$VnP%Y
zy+ELq;&?hT^gXSSR912G^ciFSZK<}92^B_m5BI350&A&kMSHnoy)ax{ZP@L5Y*S;K
zn&=%<ECz$ss?0w^3}{)b>w?VjIFN(m(@#I<bUg6p-tq8wKxx6vx?!3ki>9Iq86yi!
zadss@+v-xf_U-=2tL;Q%1)uF43dzz+7PTpd4`tb()9oo`x`jZg((BYEPbDf$+tQiL
z)#ZxacF+DW{OyB;LI}D=O=XH6cA5iP71oLmkt_wPrz_^;#PE0|I>Yhd#A!8QE5%+s
zvVHrAQG(hkR!u`7vKwZ+&-BNE-Bm@ZHFB-l8^Jdxk2*UdHa9EuWSGZ+Zy%0q7B!}Z
zgl9kQSl7?UF|qyTE5b+rfKPt#A^(^E_7kR%Nz)^SM85rxy!qxWzxeFWsVc+s-}^)S
zaA0(S_pTOP7?}5u%*RLcV!_Hb{P6h&n~Rn#66;mLqevpJZ?`z_XcrZ`?Vc1Ol{Wn9
z^H&^m=GA9!dHc-+hskmKX2-+Z9XG3<{kF%sz%Y2GIe;v6q?MJ_LcQa1Rcx#YNg(0-
z1AaJRY=bhT=CoNXFs2N<l_(sq=p)8zVn3A@LLu>~NJ$0mQ?Lhebxm%T2wh=KO;);e
zaGZzKAr)KlJoUW!>KEL8`<Br;Y?GKzBi%+2QXol-p~wM}a?mK54}n&elnZ;xn!03Y
zL?Xw6Hjzr7Lt6u=ve4%5I9jQ+&SedpyBySaIPhYtK?+Sg&s=m>pVPDDYYy{NsLr;g
z@*XEr0b~vtVLL(=oOX{m=iW)e%23MAyu(&CF_d=A+DN*#4C1Afg&LeAwyRMx7hPXS
zLMUnZ6a!i2vpv@{91esS=oU-JMb42@kz!=(djt{N+)y<Q*6QL6v>Im$GRC>m5uxi6
z7W~dtWKB^K%vmzYCqrH>Kt>j;n&dNg4_p4?n=RedQ$9N>?%vM4IUR}V5tRa8OT#bL
zm)u;gsjcGT*-N^MYtq|a)9J*)6=^%iLPih(sRZ<_snMwb3}f=e={F6G(}Y%~ayW!c
z@R5AZwYYgYQ=tE*M-cp2I`4%f&J)vkB&U21Q7VREz|AvCNXBslSs<|T+<%@^B1I3t
z6eCSjh`inh5CUx^?>~P=Wi6|V3l@!K+8()oeM{{EtLtk-4qU7&rrpGHy}@S)DY58U
zzH<zk%HpRPL}VC7E*33SrD+!n=6*ygSz1;7$g8(Ue*XDeDyw<<^n!UfAqluSQY{RK
zM3^Iy%%*AR&qM!GJ|;Q3#Li7rS#Y!(ah%Xdj488>rq<}nuw1TKEf+MKD^$J2#*uV-
zWP92pa-^;*s@fKp<8OHW+6^Qnh^}bt$J4;+H1M6;VvI($73-T7gA_b`^+1?slE}DW
zrtU0ZiVTkfOCjmB=Ax>297fDIGESaxnz=t6$*#tc`0Vu^Q}6hlm)9&7RjJFIM!MP(
z=0H{gYZPIexjQ^^tv0Bp;^{~4V;7f16q^^HmU__rXZV!axdZoq`I=Y1`iu{L|MxI;
zO&lUu>jgjjWJRAG-@Mt;_Io;&zzvuxaB<V{I1i*MVJpqx4*dG7FQ}~Gy%!yR*m2!R
zhOrFYt&r^BY?%%{@2?xy8r=TC!>ez(eSOd90;^gha^jRCKYD-7dry|UdA)n5Q<e~t
z-NQZik9YK=qn`pM4OpY-ng*d1DJG1Os0_DX{u0|Q(Dj1jVN2aC$flz*3LOKpny4?H
z&>tLgf1ooKsVzb({Ao)zHKr~@Sg9?-m!$vk@s8U!uQ`rmLHx%8=DYNXadcdF8-go%
z?xwCv!=H#Kp^*BVX;1=EMj4IBkfcIsQ@TH(++Jcvng(?~4I(I(TO?JzIG^YvIh79Y
z?=~7Oi!J!OW<jmX;5f??5>z>&2@L&?>6A%fr0EKiDuyDRy`!tTERZ3Pbzw}!RC-f#
zg5*npF9|Tppaf*^k=D>OEg}5XuwH3{v4z7VMJX%_vnv^buE92?+zBDEDzaI%R80d}
z;C#gSNQjg_-&AnMk{5YTLQG_xL6rrpGKy?6Kl;H2)wKM}zkI~D3l6&hss*`T^X&*p
z7_5brj&z|Uy!+c7x5EW5R~4Hs^7IFva&!3+&z@iL&wf4dW_Lj8LcYypoPgR|mbD_E
z9H-NX*^MM0kc5Jsks22Q@$6Jmx>)9A&hKn-kA5cSNbn9n4U}#aLG%G3G9P{NKEHTy
z`1^+<GPd@7qCNw2aw%T}uvLXI1~<>=TU=(nu4&JPxMsCv(RTEAuW>_<s4BR=V!2r}
z3_VYtKc#IO9=8w0h$3W3@sknzW3k$zSU<Z#2}r4!E{5n?u4;5HPSl6To+nRlSXKtD
z0$9yB2aqM<@8*OQf=X+$pzyO)nKiZDn(ZlYn%%iarx-u9)=0n_LuCsY!MmBmN%G_O
z{*Y}JJiC5MUM=|S^PlnN?LF7m7bq$I_7WC=awQCP|9Hg5fKca`J|?Ub_`^)4HH(V{
z)1k*`O>JwevBY^K`+%w}V+{Q4t2b;%$3?3dd}8*QH~YCrS(RdccOo1oE<U=T=?tfS
z;F~v(Oot=eZO`-T3ruC0=b3NcKC+Ns@ppgx2Q*?J3e6BT*+s<ZHQJY}qH5N>|Kpz!
zSkhizlKg>JDRHurWv8i{iWillwgM9!Lvkns7Yobj)gJ8zTGPOE;$;q8z1(nf*|DlE
zF$#X|B0qX|L2V^ZY~r8%<RzDFRZgEeQXTHNk%e^mbWx$ngdy?=R||gh;*x*<cKdEG
z38k;YjVBJb_uOuGtQHMzW0{5ntL2I$CC-<VvGak8x+5MQIBp+#{puA$Tb5mmAyKt0
zdiRFI>BuzAERK7wJ~A{pvHj+2me1ZRFg&SnF_S`I?hp9Wz69B%0x39*1=0Tauw@t=
zo13Qu7YM<DG_-XMGL}q;0FeX4!b3s|lqjLLB=z^9MX4)Dxu8pO0#SHq)>e=c+7w=t
zNavC*DL@sSJ%v~>Cqf~NE>cyyDZsJh5oxBbB%xX$bca9f(LxZD<ND>x^1{Dk8eJ$!
z2oaeQ^1H>^mb9{no+!X}3o29CaWTxKl+jk9lsU7#itOY=z&npn#UdY^FLcz9P_Zl&
zT1iak(SDC?JCrWWCMZR0L5T0WU9{3@QzRs*6vi4vN*srg#ipav8QG7J6L*h~97JiL
zlbdVm%^C}k0^Kz5M9n<Y4cj)OgrLufWs~`lIUxOktzOd|M=lo?yP+KXs-`ZWM<vQk
zAqVn2Go2>p9MCzkX=}9B46`o=ikz^@lwO%LI^$g*k#h;g8jG+Ch?0B;XJ!G$C@QN7
z^UNH|yhO^9ISSDi9$JnVZA!*zo*DZ7j2g<Ndn0Dn>kYf(>+ip=FgxVwNS;08X`)@W
zT&@;8zI}_e6&IIlgw#~VaM*6?mMc1|!OyhI1#TP(lOTmij1eUvMaA8|=f!*1tTr2N
zzkNdp32iILLy?qoWV38}`|yBJhWma-E5WL<bd_aJk!4jg`h<^(VVDVLh-sx{F;L{d
zaB{qT+)`Id{+pj}_|Viu5jo@&%Wj2HntAfrs_+JX)AL(sycrcU#Kdu&P)&u^3Xw}K
zZS;}d-JYt_)QgIkGMkH*#nl>FTi$&Brkpgj;+y*eX?vo(T4Q9z>;m(U80w5^5>5)n
zA>u~Q6eCH(JPv&Q`Yq$`MBh8w4=r<Wr7xY#&%XSU-+ytzEE3<`^*sONcj)Yj93|r{
zsMDS>>{%}?Klp<mfggE0A9zfOhjHMEYk0C;vbwq8!j==A&6!F_mOjDgxbT^$!hlh{
z|LMoX5a@@25F>B<iIm~`V!>sru$81+tT6(TbEtVldxs36+)}jU@4vrc4uNNtrJ4d_
z3hxM01feaj-@HY*f$L?BtgA9}$ce#6bgdvp<~eZpaNsz37MhZ_&IyqNF$TOyy!q^N
z1c`dRX1DA4`txt7e)cP_Zmzg_{sQ82q8ss#ZwR(!bb-_Ej%okM)#VbYB(H8C80WyE
zTkyCan9pvNzMl|eTpdsmw5<{GL<oVx&M!=+=sl`xh?MRHB@2UzEUB7?7$fu0qiv1V
zZJBpy1H^OSFP&wtL~?;d87@nuAse!?-xFnpGR3JA5^)~z^Pb-I2wxUaWG0<XC5``%
zgAip5rm1M!1=@IGn1Of>5hik7p_-<Um;H>)krV=Bn7(g#Ohi;F6G~GVQr2i&6<1a+
zsFN_25SPl4R%g=qi0`iH1V!#Knlu%wX*|mqN)$+DK9i)g5a&wb$WjIlZ(sBIH@D=f
z;@9^Fs2#dlvberPD#Nld#A#%lBfEZP>m6@)M}GQz#ov2T^Ll{oc8?nVoL8$K@z7Xg
zd+wmgB}nSLKU0&7yGh#$Q`gLcBjrp}8&qYP{QevUEq&uL7THQ`^UfwLge1mH2t}IA
z3H17g#wUh{Esu|PXar4FGtUt>j-@l_EQ98l3inS0hG8I{H*_V7yD53klNZlXQkVB;
z(IAY*jRRT1{_bswD%fXy{`r@v9Qo*@_o?a%tqh0VmM~8&+p<8*w#-Gm^R%lqF3bf;
z+f>}&9;oVuCr>UI#u*_>Hp;o;oRq+3wLl?go0_h&oF>QBdW8`Yt2K3!WF+nNij3gy
zamR6rOmkq>HMFf|H_kjBM{c(xRej0o#Z#`IJ!Kk4@|QE=+kg^fy>F~8NcZ2cSpC);
zh?X!-6TNpNCCR#I+l`dyrozX_X+I)PMPnXZgwq6LVAxI6QemxOJB}qltSum^lwz@3
zkVDa^Ns8zqo7Qr%Xt_TP-0ugb<B{Y2NOFl~WAM{Rh>=<)-s~Uv`7eJ)qZ|J3|ME-z
z+y9ecbM=H@{rpe(XaD2B=O6y$CBO4x$#~k4$AOc&;PLUmRrH959@{PW@cIUkAkW2t
zkwSoJ;zd()=xX}A1EWxA)9@X>vu(Q4%BUvN;eooY&{i^sM9iL}^VCEZQV`tCahMQ7
zvaB_~|K7#9HL#Rq1s{3WDM+_mQ8f*R)1FP$uv{$2Mq#8u7)!t3mqfFLr%zXix}%nc
zaXgiUG&r(UG*w5s(mdS0<)XWwTdo;L$JQ_S<?UN!{fO6}ahX46+}?8<dj9f1{tT5p
zl~shPCyzbhIN+7xh~ju0*j!wkGcXZbS&~rafwv+CQFvYGQd9b!zO2%@oz@d5;Volo
zLKfsKNSTs%u^LH6ph<o@qa-L~T0zmubB0_(P*N4iOT|cv5o_yW2|i~eL~xw8Tc)w6
zwpHO6g>Vj*z1z<tQwG%98iwIiS^%{`=MiHyj*K6U=bKzakjT*!vOpS%mWq(d=4rGh
zd?z=FVmMJEkX#`wYg@5sG%Ho1NTuR7ivn1Tk+C0`$Fay1Kn@<)`$FWEqAZHJ<T7lU
zS(wbXvf@u)pLiStFINqVt0#D&5Gi3*X0@_-vqHy~`@_JOpS^_;xmnh{c=im}4?Nu6
za^E-D-}@<0x<X3Ov!t|_Mj<H!YNZx5DY4z}Nhwg(345MlO>-hOW%G$4qRw#4clGh+
zxrgAs$KD8S(XEDNAc+&k6q~pZ38B90G|{B<9K>Qt(Gk5bv#C&KH#sMWf%l(XarN|?
zVdzmgVHYh?N~UQ-*A?Ub#GCwz-MK07y=ONBADPFA``sHPg0`&*-r=T^rt8qUqFpV@
z26hxImP=0iJs~ERMsaxD;d}|0-al+17sJSVPgkWPcbHkP7F=~Tt7S*bC0&0!p14{q
zO3iSquvLZDP#;Hj+mRR}wKY7mmYg!Du_d~ZdA}oHZIE5fb$5xvupb9jo#kTPqSbHy
zS^O55qM*?Rq+pn5=D}euZD~Ifl4+dJ>>ye4A#-3n44mpR!1Bk5Yh#OxB!a0db3Z}K
z=&L2m%?c4C)8NR*o-|B+`eMV)%JS8>w-_Dx`<o^I;q^$LGRpuKLFv9qQN_}QaIx%Y
zE6wJ^-{rDh@=s&O)%!n2>5A>!H~iUOe!<f&^6}*-xN*Ym?$K8lr0ofn6Rvms`m1|>
zcyj^3<UBcN!r8<AgQpifS(H+l!~Te|j+io1Y3i<}(ze*!ba8zQ!@%G}skh9Ly0r`;
zf)8A^hR&3dB!4dPaWNrPNtjmewqDWC6Q*g{JbOWF3{n@wM|3AzwL&ghd@zi`qjH9J
ziCn*+$%3`?<l_T)M;3~Ot!bA_y2~f{am3g??>+s9@Oz)ad}2K98Mm+LPbZAEJXu$4
zw>>XE`aMLo<nv#B&Y<u3@Q3eFYr(6}zb1r83KLR7-Bh3qx@nL`qtps36Do(oU`mnb
zX2RHGj4Fr=LDA`(s+>lnFLG5*MDOt_;+f7Bt(1&`l4s<6I^#|XC0CwpolzGAd`u-B
zuB1e&ij+=>6j)r`5OSj5?}{rg^UjMFxzz0rrz2zEQ)y5llbj<Pfvy(Bc}8ZBu;+yP
zneMD?UAR<JY5I)k0i`UOUKzYI6Qf|ll)rnPQE5Pmw)9iv!URj_?1Gz0I8PXTb{1t6
zrIu8j`%i_`Oq0j7OYY*4&yUb9HvHWmeL!!P{Oa|QWu>|43?Hs5e6RS`>qoXu(X2Ln
zGxYrG(esl(cuM<W!@v47&A<PPulOCk;?p00g4Po6Jjr`1YpIOjcsvnvLKuTq79})O
zDAa2U7^&I&nGitL1py?^jI?)q1aUf%{eY575asK4(J}i-Ka9-IGfo3;oX8aNKvh-v
z7z*d<Y>QS>miJt1j8e?Aqpd6-ym!fXd_Yx8@;q~Kb;at*HN)<aX&CT6(;^5Vv)XL9
zd)Ojnq*8^>&^8SrB!*!^S<CY38fi4BM06fq8FH>!3Bfea#Ce7g=<15ed2H1XLSWfd
z+-w%Sdb_3RsuI)Eil(X%Df6(M3MOfGG_|HM7BeY0B$2K)_>>suf{S;vqtcpjc6|Ni
zuX(RtBCpnn-lI`CACS~Y_}dTB1<)9SQj$1(Vn5=HqHZhP9B@<Vvk>P#4tE;MP;n}t
zIN>7AS}}z(9IdP&ILCDAxt(Tm$aIZHO#_|ESY0|Qh=Gr+<oWdtO<VJ`zuaPVD%44>
z(IlFtVg2MOAN}wT3FE*&`Y)f-z5l!P``7&V_kP0v`Ld#|HTCn~p-~!+uMvLY_VpvK
zKk>WoT`^9cyN6vFGzwXI6Jj8yT(HhnRXW(ULgYk>FnY(FGF^sooXXej0$uc&l$d;^
zHM$H$$#h1t=`@X%w2dv_10T><5mKf$f_IF8&}iuD7QMMaD?t`;c=a_xLbtxc8cTS1
zVAU3Fem@81Va6`5kRo6=8e&O6w>dHI-qNnG5KlV%-5b*3jxbMnS9D*WJ>BvWTURW9
z^f8X##a@4iS>3Sy`0w(&DPe2{hqruqwPC-#C+dp67sb7yMZq#ifz=f<JE9AOaU`U`
zI87v<Ftx^<yDzj-c<+eO6XFCRmAQtoW#Aa+0*@4>yeLPH$^~vGq(X`E^NLb|PoM+R
zJ4~#}IbxJ1h#6}&&FT_#OLh}>c}272Y_WYeP4Tq7W8YYs#R{W<R6<>Ka#%p<5hF5M
zM6}dMX(%&Lfsw`eW28cv5^N)9v{Wc#2tlDk(cPsI<UAwOgfx<@YEnAi+Pp(4Nz>H0
zu|}Q?wT5vb=R|GFJgzbd5y82L6oNOg=FfMDNmyQ7toi--HvGF+Q<0s5!sJK{fj|Cw
z$De)q$kXdfZl1r#7~#)vzv3saANl>?{gj{ngMY{`r$1+&9pf-!D~&Lkx~@UM5F*01
z)TZ1ZRmxmkJt6EKirE24TU%^D6JkPIgY<zg&ZRo$9p8;Ko$$vk&J^3W3z>O#xM^k@
z9rt&S^oIjhXB>dY_z=JEOOIy@w#X8p%I6^XvbmUc!TrOI-Q7KQRSq?a^^)~w!?vpF
z`y;h7m>g(?XV~BK=IxBkFAzCVSBmwrqn$f87aOi_uF$5;z@}ql9ttb#^2s$hB*tli
z7)l?9NX#MN5ghx0n{_GEv$o`wj^oVi602oHRo66aP4u2cXUQavr-9W{a&xhuYa9=|
ziIAW&hH3Ov3j93cT;^YY_80u*$3NlfYQr?17+mRP5#-+rwfrqMUg#=;5s69)8Yz(d
zh?Zqx%$346rQAsZk1;a#2_w!yu}X8%YHoW+MI_H7VJ<~JlFZ}lJN&}($yLpDrMT+_
z{p`s|ZdM(err~g!cs$M+B{3;teZhn;5j<-zFjC`fM_4>#d+4##j1qA3@&lr*@v@>>
zUSrxNjl1R3A7{S!+0XgWi)+SlB8Esz31dqlyjcI_c~I-m<vN-WOJkZf<;6eVKk~R6
zNI~O+#$`zaj=p%$HHO96!#}#jldCn0wqlwcN`Mxnfa~{OZunOJ=AHe2`Q)5vRyRET
z!H*gC_qcgR*K0yMpk-tlM~qLHrsi}gnEdH*BpvtU%a@3@fx5y}6)^?0%v7~Sq?twc
zl+b*_^ahq`BuEX?&~G2gEJH|SdyQGWB(;|~Qz7+|mVi!<(5+D~e;3ibLQY$rFKdp6
z6XXHy1ZgUbfkrq^hb=NCgsG9T;_{<USS{Bm5qbOd7aUJB+s7mQ{(<Ru#Lp8(YjkB1
zIUr-8wi+Kx#=#hk(F&_2?Q(IhZ-us7k@1+y5~qL(0i`twPZR;EYP3+qd3VmQDC*-I
zOe&P&cX?{7F$%O+kjt%4Q^ImWc;dJ#Gl;qdxkMO?mW6><ZZG-FVJaw=G6*OE%q&aC
zNI;TvK==unW|Xc#SzPdBS7uHzm5oLzQN}|#;)6d|x0cyYqTs=s>`}py1%-zsB$u70
z?+3QK6OzPm>iPK>C;r)fU<&Y)k6)05WS=HFsX6U5uXa7{+aotGUh?1k*Z*t&{Oi}u
z!K14RohVq2^Nn808l{Sanqr{6ddf;0!gv5Y%hnQeDd!SGpp{^Z;hf=lhewEVDB0hW
z8i=+<CV8GC38c}~UB}C3SIpwTFCQI;@kAp)Dq9YSNt0qkic}WNkO;nDcTW9@!*0R-
z-2)#imI$4x7cJJ<GFL+ub>R5Gvx{dOJs6RBynn>afhW%{$OyXiip6F{l7bLChr72N
zwvW_x!_CVVR7SFYc;w>h8Yv=v8hN<C<L&*9DN3@|bY0D9d%#-D?c<h*!_3vHqpmcQ
zpJ|(#>^#oTg>j{2sXh?0yzle`pQw~#*&0@h3)YS1>-!Tw`_*6a-~IFdC;v456As6>
z9LGIfWw^PyV6krh+V|qOZXm4@EG_)blLhauYZhJ2#kxTY@GjEShNh{AIdeFUO#O_=
zV2uV6U1NA0m-urhwjUj}QXs%(SS>0(xainanqMWwZSS#0(ln(<`SI13ZSQ$;wPd3V
zH^QQ&pq2_fJKWfFm^A<VfB0tzDfy#6_=u%Du(^3gk{TmKNmQrOvs@WjbmQPjF_Dr$
zONtCaE<n1o6C|Gz-Q+TaB$vdqj~O2V{X8>AFs<O5!(6Z(Ipd6C2#F_^Vl6Vm6uDZr
zT(2s~{v4_*C!Wf_BMU2oU0(x=<NYn;G|{~1Xl|ad|N7_5$1RI)iK;DGC_IMM%@a(!
zq*^WrQ;!S+5ga1UjPuCiVntUa##eudUSH6*6>f&*llMqdkDsT)v2uabF7U3919Wg0
zTOo7CxKeMMtA=*5#41k^nU$>(ZYXr!YC()0^Kpw6mTFN`U0#!f#Yly<k`~1_CuUn=
z)=zmnI3D^V*0y~6>J7s*v1%&XN>Lk87Vg^6NL9j$Nx&RXs?_3opRi5M>?X8;;0nM_
z39uTB$z;_sOfw;zIPJG|qO@tgt8h|R6?M~;ir1v5l|W|+KH)+kB`cXp`Mek*k;<Iw
zZ($k({w$eFsqphmObO>Eg6mOaY+WN$Lb-%i8Ik?j%@t8fkdt8Yfq6I(rU8T?#!?8>
z)HSxUrFvQ_Qi!Z{`Hbg~_|el9Sqf^IdE6doCH!ck_~g>C?<05n6E9lH?=C%WX3e~;
zAUe#4FS)vT%E3jx{`@sBt~Px7;*v3C?)#|#S1GpX@7PkDyIfKRl)|=~GP0U>NSUcC
zML+aR<M>zK|8p|`Z+wIxJBL4Q@j;+giX==)Pg_N`Sn}lMOHAD$wdG;nvwa+x)l6L%
zIZJCz<XjaUBT~vUC#!5?yME&KVb7E2Bg>}4`NaNkK-!8(&;6k%g%SI~HJw&$79Dpx
zM`bn9M+Q7TB&Pnze2m~{PNx&=%S*1GJVjNS)7>3XYOY?qWZ2$fE6vY;{Vl3lv(^&t
zX6i~YS%VJ=sVxalvuLr#FuTCrZp%12qH}E4lJ(LcGaUMXaW3eox{9RiQRlht&8lM2
z7=H08bNB7f`S<_Rp1QN>3{6vW_4Jauss36?@wfJYEUe_c%Z`t(e*h`7Y#X$-IAJih
zLJ^pc2c~%@tQR!4pd(a)Zla7V!&F2evgY%wGiS6e1NNK+^Xz#zIQnp+stw9mlCo^)
zz|&2``xh;>wir>Q3=oLr4d(I#w!i#>|J(odzoW96_gDW%o~<O4I<Z+Tu#F{*ThjJ(
zGM?@2zBKTeiaHnUIVqwQkR@8%VpT3pM+hV_7R2{=Jrsg--+>aOaiY>|-X2D_+a2#O
zDy}a&7DRMbNG&j0u!(^-pONP0{stE$ql-|#%R3@?`1aTAZm+2~SA@Q2-rf^}r>QG+
zwI+Jc?(Hqzdc$(Lz`N2Dprd2lKhiEY$f{xvGm*q$`^eNcM3GVb5jURb2akPoM^h=P
z#Rk(X2sxAd3DIv+^93PgWZMy^o)Bhay&wofl@lUmvZ<h{%l9&Rv`FO2&@}JUUpzxa
zN2L@g&O~=&+JoBMaP|B>Zj`_U!LvX7UCbZ-l;~&v#sB!Pc=PZ`njP6n0^nv3&Qo<O
z_J<Pa>z0dRa<H1?{sB}8@@ZK%>M}BoTwFgXL-xv0Hyt7d>TX?5^?!}d`=39Vp+7K<
zn{)Of5OO5N8Qh4<iJTHqf;J#hAo+pdN>7KCx+K<vBF2JhS5n|zU=9&uB-U1E)?7qs
zOQkBTO?V~I%@QpObW$cqjsbGUSX->Xq7Y<r4u}9rR#*{u@3O*=fqo7=z1Z;0VP=>d
zG1dI|NyE#lmKW=aq-ty>*dHd|Y!4iq;72c>QJaSSX<~aEn2$4qsQLEo1C@xpxLmQm
zzCczj-gyw9M2Sm><Pid_HADnKC{kDuT~CT5g2*&aOs?i)vE(F&ck2S79KXDCxWk@#
z3@8kuUJ;TY1Gej!L*OE3<}pyKW2xmm+nj~e1?nlv|3Ah^bgo!<CB^Ez9Xa$9^VqZ8
zT=M0weob8&E}vWxQ{vz}b4q;ucE|fqF1e@`gN7j$$EuF-aC?jPr*eayN3Nbd<?`wZ
zAv3$%w+IBA7cV$IZt*_wxbJy51^!q6)gSTIuYS(<_Lht58-%p%wg;N5U_YakVH`$+
z_taH|bAc#J_e%JV1!qcQ;ouy#EvTK~1J-DEW0~K6aD72*6m4h7DYIVGY?ckn)#Ce*
z?{9kk+6_bscz(U&*^>=Q2*T_TDU*c6jRP_`Y}bKMSYxp|qlrZ-@E)yY0l^{8Vx~eW
zRX*V7MKNX2Mx%`3lb6>#y<Ty#UZAa|t}FiGAAG`3K6#EhqmWC&wKUPuU6Sh!*H54D
zzxyYDm&4;djgI{K^H*G6?78{r9~M;k@ms>-mgdFpP`!8`z58QAf50k*Iuj;KM{|)Q
zGUYm*^7$n=J3EY4r4UB|C|+zDqz)`}<`32eS1)<Ks;P@QPo|tKKfLbBe-l%IP$HP;
zK!~MX?p+^A#x@;MNET8Nr;#uv&^6i^d_2)!KBc>Qf{rupxI?9Y_cN}#BAW~3@CZOv
zHHBlAMHxIR$)R`W*}}Opb{?6=o+K63Vu7>*qf5nX8n>tvQCgy9APol^Jd0(6YL`45
z2WC+ts0bk-{7kiIS=22udt{CX(Gn7bteNJi91f(Q5i>3gFnSD`fBg6UpdesUAPB+K
zPxOZa^EA_4uXw!OF&$6L%>^+9k__Zz`PDDJ;4pazGn-oS?8$~40#091uWOd8C1!I?
zI31~j!&p;D!|&*;wx)Ei<dkUJwpfd&A{QtLm~$bLu;{8T==vm?g)FkEmL=FFze|lr
zq&ok%#G=7Rj})cM_q(p{lnY<X&84$COg(O%irb6x0+@>})mV+6JVL?KmEm$(@!9Qb
zrj!XNYHe6Iiq)dx>Gcwk5>iPj3-4bqi2?j9xTz$Ujo~myCZ9;EWVu|Pczk?BXv@vV
zKjP|x4_L33eEs?x4!dosCXTUWF=R%{T!OQytP|N*<Y}VTmd$c??rkq}aS@Z=?eVi`
zK8;M-VMR-I-5{mKnwnq*hXm4;*1i}V**jb>uyu<z`i#L*$YdbR<ufZNblhC}Yu?`N
zn1+%5ut$r;i|03lt2L6$I85yN17F<k`Q-T}i`sCp$lSjjc(^~Ij3&APF&2(i)zlcf
z!;PMC9MMwp(T{&j23~*h1$9&L+2?Qg_|s2$_TCF#eeoGFM%u-yIM~tT(}*<(yz|r*
z2r^w`NU>&Cf_a`8M#E}RVWnVXrXNS#Y*0$$rio)NcJEpVs!C&}BKlllm$l$>(_o}P
z)`eRAx3cj{DS38t&8k}<$V^HTrU{h;3lRw8z#NJe9fuyNBvS&dN?Tga;yjRrs<vpQ
z&sczS@p~whn=&VA1ut(lrQlwcP*fIz)q2gv)lG3n#01e(+d^PXL}CooUB^HD-~LmM
z4{veDdlsvPm!JHQFgy0IK4X6KOV*nedi9JCfBaLL=Re@<fARldI&KTMN0kB|QUVFy
zCz2FYttJC02DB8V);5-|YbYCFT{}_^Ts-MevZJdlS}W!dK^bam5TS66a)>Nq;UV>t
zBP4-Pk+<*gf@Ctw%@wv=;f7O@W9qsr5_N+x6*@UK-4apPBrmD!1|byNhqn+W#={-P
zNLGs_mWcNOZ8VaAPz@ja-akTyfve|N5CV5!zapeq2Jb1Ln}+D85(acL(w5L&a5F}i
zTHk*5=yV{C6Ir*|wj-yAB9fGa{y<6-x^B?61|`r{R~Ef##-~6WPPFSAND(&=WD{8}
zHs$x9Je97vy4;|v8rwD4zthp*-SY78h-^2k7Awa68=k-Tn878c5b)!Xt}(PCQ&%;o
zX~LcM3{^uvc>Hw2Dp~r*-*M4unu^6@flvx#3`$tC%p?hth}H#DACV<hU$sb~kQCIs
zR0=IMF_q7r6uMw#lqfXWa3Y3)P_m@<g=Y2<FC2AnB~T^=K1RlArm9UTLJG0`9x0yf
z&5Gy)CJUk~?TIlIcGdfvhUbgK&E*QEHN)vd4wAS1#Ak1h9ELzwTUN`CSKFS?A15-B
z&B7G=feSo7KC*f7A&X}(Iq8o5-lN(}*dLi(fm)WfmV}bNP;*lN89`gJsi*`z-B`AV
z1Ady%>&Lqvu5wCvSBL=P{1vsI5vv<iy)JG(EkGIu@5p4@x+d`*&#&~E%QLb^=`s(=
zxwK>|p*fy<_H{wlY;WJPSuSatn&EV!X=+xh6{l&$=fr-Pxb~hZN0wGFMayG9^VQqL
zG&*iph9(5|Ltva{lurD^fB5&262AS_ubAeU$L*eWvEV0v=l2oGbGd4{Q<CFpqFr=c
zT-^}Ae2o;Zky4<gBx|^NdV`lc`Zoh@(~#%DkRwtTT)q^=#NmTSE5Xpug*t3Cbyeie
z`+fwa`1t)BDyz=@CFSt`o1Wj^NErS17E7H|8TO@sFp?Z)xe7{gj>RitE?5AgHBv}~
zENB3wN;R%HgZZS=7@>%=oanXDD6L9MohfK=B{f3na*aQ?g+Zx>)JhRd<}~&A@r0C;
z#pMlYxuVh<KMef4fB7#lr#F23{x#R@8*n@1{&Svv^h0X(zvuAFe}7It7tYjoox$PU
zJDyV@r%cN6j2t-YZ<N91;!H&)m2W(`paPUG^`A&ZQ(MN<0p~;61igopSuF%gmZn9y
z`<s_6fnYrK4Er4lN~%<Agf=80>F?j7a-^vmhBsd^&5mxhDL;=ebG2O3ELP0p3_>tX
zGyDCPs%ls*78qMLyZ+&hUQVp9pU|v2`s0XhS}+Q-E4}ccgp^d(mI0>oXj1}vvhFx+
zZy62;szs)sjwKW*HCi|1Zbj@9A}f5%)TSc38Lbt%Z3>HObzPGDDUzfixQH<=m8qx}
zfqwfy2$3i<c|20pl2IqNcW=?nz~X8{y?H`iop52qHj6@?%aNvT@hNjLAIT}8bYO5N
z7MlwiQ38tZx@-xhXch}f2}yDC<P63(rTZj{vN?!S2Bwt(1TsNB>-*}=ViL$QG(Hz$
zNg=7MB`Q@c$~lq|xTHxGmz$J|lqEt{NU8B|B9qT_;&QWzluhZvDz28Ay0%=jhE=67
zb;YDh#juZ&G$mHG;$c7XZ@%2~C*KZ4WvQzg_vVp(KVvFGTQ^L{k;-c3;2CF!*lqda
zfA{Zr_1P~-^UQM9@#Ek50gLsj1d@dW5z)+rg(ftW0*pbZCC2qsDPx+NK)I#-b+*yp
z@Ny7*X&%iX5vQ4l<B8^3KrK8$8f;Z@*bdz9Pq^SqXN@d%0aj~_RCqU+pNkadj&TJd
zmrk3MiMi&hZ*IBR)LdR}aMR2@PHZmLc%L|)jvV`;gqkl`Y}PfpHpDkQv$yP~jEjlU
z&lnY%$1Rtuiof^wKE+gqx8HulY4<=s3`i6oeex1xA^U*Uf)FBye&YQPUs6?;Uwrk3
zVM?r5OB(4g)YwJKVvh9NJ*QJoqbrj0xN&6DEx1{CxB&a1N1zMs*(f5=4-;M0V6A0N
zi6jJBD5hy9%ER~FK>z9U*KQy=XY$!&{+$+^l`dLq7T6|~j)rsLjS!;LdQwV$lrF<^
zTeoLlH)Z&z)%o;P{!yxQ1}I&sH*-qGjbSR}8OkPQX)P0Hw@N9d6NJMnUVZf~yX}@J
z74@QFjsxO!%abN>^Xw(hKKcR8^0Kty9gEg@aj{_a!%x6@gw(|gey)f`=|X)C1%r@t
zK7W{s%s?&!!W2E6H;$Ys<k6hJ&l2SG|Ar7rIU}XOO=Gd~mZ1w`af|%ZZ{F!85h5W|
zMs#a*Ja1SwF^xTbo@t*xVGM%9d|n7o2V}iK>jpngD5=T8lS270A{A}haPj0B2+g>^
zWf&*MdFFVUFinRv8b2S&?z`&QRDSlRDjRPM#OM&Rge;?~>8@WwwIt6+WL6}KOXl$K
z$oj>HsEe1x=!yGVeApvRg<Y<(l}78*bSd14m?E~jAS+8qlH?<b096Mjmx3lDyOU!I
zf@dFoR2CiY3DMK{N2FF{H$%>h{lJtJ%j@T*J0#9fwWw;Q5)o_l_h80%d6YJ2ZPD5g
z!dy^4@(g$+;e8}h?8WI!9R^4!gfgU5W@<`Gl&OfppTmY_m@9-hlVt;F#hl7}W=snr
zGYmcDx71v0t{KMz<5c|pIVFTyFlEEjw&mGEB85ZbNLy>JR}I=I4pGt1nHYgIqpjsn
ze*MV*_s_oNNR3%6nci-ZfUX-_V^L_55X{qza{;tw`*@3+Pt2zi_>uK`fvt;tSsPRO
zOQOfmBLabKJ3vqyL%%=4euoX7N)+eVLK|)tOOklEO(aun&7}rdV(PZcmV!qbOA3+g
z<B4H(XP4E?>;?iLtR<NWb(SNuDDcmc2`N@zDJ7>d@Uzdp<-hy~pR(OPB9b74z-qnW
zc$(<i8XqIaxoqysrKM^l9}7vuG4DM?@O*VUF!e{CY%CwYyyW)PTSN@R7`eT>rLiqB
zgP%_@_sso)oC7{2_We|94oK{dLt)|>%c`rGv*7ybis9tAy1C-ek9_;E<Jr}U^?Flg
zy;)*fO%M@3xl(#(EW<prKh3xhXxpmbSL&LW3!wAi{uR?OmG$>Od48*eb^YRfu3tPy
zn}TSLDG_4Cn&LyQ>lS$)R{FUhEUKz_&&x?pp7sBT_j7SJo!jG-G6<~-jzO2aNzs%`
z_zOfS%8{ak$G)qa4DtKTYLRC454ZgJzx^|w|L_?LSFv0y`QXWtY6Eq*Dkz0{Kvo?Y
z1;d{7@Fnr^Kn#iY;we>kQTj6|ClG}CD-x*4pbSzJY2my>oJp(Uj6o>iw-`c!{t0ou
zftAHVoaMWm&IHyJ$SE<)2L9~Zzxf48iHj#MdH#3)5k3jR;gM-~%j|nZPF&pF(Ab(J
z5;rfNGw!#{(?}v>>kh3odG1lx5<^0X#Co~Jop#KR8Ci8`A*dD|LK;HKxZ_xCrUGnT
zuJw}-1b?o;oy+i62*GeDnHXy<Ro&8BOH@meZE&YuLE<YA$1VPJAPI@`uIS6AU`n!&
z#i^qVqF#cVh+~h|HPbwOKQy*lQfUL)6du)hB8DDoi>0@=5{E)Pdx1n^&w8-b<&ZKT
zAJ8H*MMa(+%Zn?T#ey{TaF#XR<<Uw(3Pc~zxQR@Rk(3K2JDtrQF(t%VevK&;(HLzD
ze<{V%hL;3xJe5pBES3|k6)^?Iu}3rw8BLr$>NMc<j%l1ZO%5R~jkV>}oO8LY2usR}
zi$%?5SuylGLY}Ck#XASS5VMVuJia}U{K#czI1mXrvtBJY9V3S~uZZ)A?v`Yw@jfuR
ziD@o;wB-HwAj@I5r)w&Fl)QNPjHlPv1V6FtYMl3+4w1I02re>?<+i4c!o-B|0o+td
zDJjKJoiUu=JxfT+nR%LtQKGQOHloN(r#&tf%XEt65L!1iKJ)}PF`tger8>W7MS+Ok
zp@sZjF;OHQ_dV8hyt%z++0-nXhUIDvy2367E}mS2QtbCfj+5t^P%M@;n*i%Ja5pRZ
z!Q)~?C`FbPfBYBUaJ`mDS!my<VdCaugHDO2tr<3N8T&oE$0I3b4wGkcWz8AK8EY-l
z7*;n|rG*p0^A}G@De>&;ibYl9qGKE;n#!QHL`t|`cMMWNEX(Vnsmpq7Bv%(}wx<(w
zNI<4P9hfFZ2=O<bt$sU9(N;xc&yZuppZ(eCj37Am2gG-hRsIgu5Xx!Wd16+?6!9T4
zx=7bGW!CVWV??U+4Wk69QcB3P?Ges5AFXlIh(e%bxmGG|&`L7f1s6Z~l%GC-$;;MZ
zl|(OJun6~vI3Z05U8OKH+`UHd7G1T(VMq3^6z&9%eqpFq*Px3jpp??&;vN$qq%KmM
z)F95A;`wHhb37v`isYBm8IO3rF$p0HAnyE07eWwYk+e3^AmzV#hZjT;=ZWz6mVR)&
z`NdygT7!~B7HpOqhKE~n60C2Yk{6k7eSsf(kg$64l<DqkTnto;HPU(9uqQdsFMsw6
zW;gTfz4vG~PiPhk{4{df?Gf5?@!~^dikRuZ=o3u^HqH1X%RpBMR;v}>JA4R)dBFFF
zLZV$PQ6eFfAdop89*H^8ELSwk70Q%)0Btl`$kLV;$^ccH26s9ja;9!-v~4j;Aly{A
zQ(2OuK$;d+6&t73Ew*V%ss!S+6c`dhXKaW}K2lvg!J2c=NLwSEN94pP^V!$`E;~p8
z-~~4#R0)Y14P$ipq=@CBTCyHep0OC<oyR3dmZb+qX-O?I!3{(|;Uo0JT+9TOEuXFA
zaf8R5on1K|m|Q?eizE}m#KUQi$mQ3}@)FssFgbG7YN9KqgIYS=>WZuk>g*1Lz;3_i
zahJJkD~y0YeKYW;4-CVEpGR6P_^*EdJr1+tmv0aFd4v>+zEA~<yps^YGY^jCYR&I|
z`Vk+$cSUd$)+oMy{hFcQ^W<txa*=tMkWwH9kD8{^G#St47)jZf(r;d{e$!LVdc|lE
zQ(`(CkP-4QVHX>;GI%%R++5zj5Wk;qCO4x}Jol&+89hml4<3O)es?QLaR1no!VOJT
zmDb00L6Vksb4h*Nv+{vyn)v4a#CtEVsI<gd#q*n*K6|v0?D9--aOfxeIHQ#1U;l?+
zbDU<f(0ujk$orQaUDvW*l*;7&<0DrWYxW_aw82diDMZ$*79|Dk#T821(z19u^78$c
z7$q2XkB}1c6j>}9td)e6XzPZ(+A$0x-Fk&5Gftk#2dc`@hKg@)AJ}X<tc*w!TB*O?
z5WQRr%cemLkpAk`ny0ZeY2}R8rVItO0$~Ut7skFa2%*b~T_$QHsk;sr%1uNXT^hvB
z8%jtSAxqRArKoImrgUaht+Cc(jKNeEZ5x!)SXprsEbHh=(-GOU*u^Ep8yFu5<Bl-y
z7#_YM%>&w0RNV?&)uhRzs+#WlDQ-R?doW!@RvWOZGt(#W-4|Z;QT<mR>YPh&g(%H<
z2w6VKLY{LYf8`^|F_0)<pU4rN5>g6GcY3F|tw7-VJ@;S#f+-8Kj)W)>#*#wfal1$A
zj?HR~cO${iAR0t_O?UG?cC#j_n(@`=Fb!0u#xynS`Z?|Tntq;;T9x`yA+dFXYew8K
zu>Ioar07|H@MD_A6S(~qIyu_aB_2aK9+<{G(K&{G#9D(f1!dxP+v04KWmY5ViYJ#h
zC6m)Fk+vpG{aOAiWZs}^91?3Sm6U|BCrys1OY%LN23d6=Rgv*TMrnb*SYk<Jtx*Z8
z;PCT|R|<i`SVa^?9<s|dGDor-7!O;FR>%~Qq6m-g_GCnkq!dW=NNr|pE7<jk*@KS}
zt4h_X7lABFf?mq=30JE^-4%g2kB}50%F<C^8MHLzww_fX%_>bimujULOBTi$Vv@9*
zr<itKj0y?d>BMC<Vr6Ex-Q(wpmoGQ8izTywX0yQ!1B8IqlH(k3r-4I1^Tln?$rb7C
z^6Hx7;{zXGF8K$4|EK(mKY7jn@W+43>+R#Y=UuZ|EcoQvns08m%)#?_fBL)p;Js(8
z7B%x!<Ww6yjg+J)Fu+n4iI0?MjYicKlPh?WBs8-NBp0B0x1WS}o)mLwf{c{;Mhw_9
zkM1-N^!p=qRpWhsmJZ8Jkq{twRBgyfAeBaGb5;t<=22QPB}NOZH5`t6E-r5f8En&H
z+LBN=5-wK@_Rg{GXBNwrrma~=$$M*9Hkt=HGxUNOB$HJ<9*^w%zI;BDLuMF<iI@W<
z$2cdtWzFMVk7`<oGedvi?RE>=QdOm?P}em{W=^Mp*-gCo;067vVcZ`vMj>QIE5$fY
ztXB(2fngpoR<T+x*zQgoUf*-biP3w$`1XN*@O=Dy!?Kp=hRFZ?-T3V&le$z?imE~g
zb)H*L^jz^*4PY@*ie!{HpV|xkQ<(}SB)JSpP_lGwm&HPSA2JfT)H)&xGe{t@bzNY5
zsle5QkZ1Ok0whu^QdU$!Bm7L(2HP&lDKa1Ki4VWTA0G+R2_+S#Ss=ARRW&iX;?8KV
zQH#s-DQ`y#6Gq>oyBkFJ1XMwYBLqSkNCD3BQ}Q!Ho3l)%zYp5v6wjMNu`i2U_)O$P
z7Nt;1NKu$P(!67iFjFmS>ZT!_dg^6|)D4-)Jnu;{vABGJ51#o{q!3iZ$OvIV`4(F(
zXf{u9`+Eozoh+w4p$t`Jiv@H#F^7q)EUV>)*sX{t_J<wgJY%w894E$UppM{`!u!B<
zJmCF_F=`rh!INiCQ893Oc*7hfs#deutk2d>i|(G{rV)SpEk4YsW`iOT$pm9@F)`i0
zLbx8GG^Xunn<d6tg3bt~Nzs$Th#Py#xHuHkN43E$EF#`gRUOjQ$Q($s#HE>OJTX@l
zT4a(N$_xxe@YCPQ`D1HK(^d!)Mmn^ua3K+bqqUac1z8GIT`&{g6<oJeh7dwI!Hdi^
z1uoVbKK<xZ?(ZM@<zM~+Z4_3Qju3J@50nKdovT}=CN8c?-6cv@<^7NncRV26jBq1$
zlaV4(8%@(%YNI&q_lW2*O~bH1;=M<Sj1ierKj7v_tpuX28C>MW#WiEq@W+4t4KFU9
z@_+cR|2KT_^eO+Z|Ng&coJO9#_ne>H2tK@7^7ZL`9;kTo?1tb*>S~Q7;pc&iMMv@w
zW=|Yv@^M$F6m3V<RODFZDuc^RMACO<T<`D*AsCK5!)YKTaI+&2i7pbfrP|ab3#4IL
zSBNaWN9^ZNzQ19f&q2d-z!R#pI?7y3HbQ_<g*Nfomv5*PtgoM7tI{h{H-?xZU0btS
zHhlf|frtIb$M3&Hq=40ltCe6Q6dzu7xT$A29{Kw10YpNIgbN;9SHwAR8b-c;dt`rf
zJXtUKppu+U0~bc}>Z`BW-EG<R6RV}As%)7psZ2i(9JdD!hXe9vN!`|jX+{F3(kN3h
zd8gBnoFa>^rCYX4J~E9nIYh{1ExTT|#F+Tv^*yajOmq5;_1k~Fge3%Bw`8$gllV@?
zDOday0+KR_4Sp{AnUxq_I-JYKKuLKTQ#K1B&ek|{?wuCp)LwF{XN_9(9eYJ;ixL7M
ziX<fD*`0L``;j4qRAdaL5y}v~!;c4K%!uxN1QoS;C?{7@`Xhpy5n-lYERbypkQvHN
z$PdTj5^*!i4XD*iL{%z8(|PmYJZu-L{C)Bn8P54uk>dX%03{1;xVXBCrbx`>v-_rO
zEbsK15NAh@Gs*X)5IIecREq_h&2>518d<dUxFDRh_24JAZ@ysv?XS^VkYgkb1O0sD
zIGrFS^wo3h`l6tsr9y^@dEBFn!D@?cDq;_}fB8SNy1b!YZwU2@aopp#4@6xW0w*sy
z_72%eXjZtx4pA>j)3F>pa&aJy_ivGh?YYwRgbxn-6U0CaiIBd+&jV@P5!?jQpzDsy
zXODE78_epe;6CP|6bT6nM&q(0r=Hwzi#tga`)Kwf(==eTD6NgoK^Re0jjEbr;T8s^
zZ9&1m;}B%kRa{>z3Eq*DBF2dj0_(10(RSnv$8pA9T+l6+oDL_trlD$EwvP|QSZ+^^
z6ufxveLng0Q(inj(pg#h&nc2Y(^S~HK`6s81-`f+cykQQQ9?hXq@&iB93o*jk=@Lb
zmEq>1<z`jWJ-^`o^(*>*WZl*nYnf(;bH(!e@slM+!slOmzVap1my$#(#kN24kAL(j
zpS^j_|L5QS1t0(6KjNo<@DrY0Ub5XE$Z}%szocy~dC?%9Vjd^lG@zBl$IRi)18#IQ
zRYmW^_nqLxfHjICIL70V!)fN=6WE&OJV$+37?u#==1^|sNJNUo(4ZB%t%>`AuB|ag
zhZLnBM9PfIqTF6;k1F13lrAhgtra4j!yuw;RHO5}-4A>cOVi^pO=v6WcYAV5Twbkd
zx`x+}dmax1yZwRXqD5)N<#NU9a)lC!VZY_>O_`Z>wPG3-`_srAB~xXYo#XED#Qm-(
z`$&ufAt{{6eDh`G{&Ap_l1_LIyFGCpscJz;g8Mfo#<2uiZ?0Ax_6M|5bh5)LiD?^5
zZ8)U^LOQ0&bNg`MbQ)+DEk1ab%Lbn^<D5WBPWv6+|7MH(Z>_Yo-G-`LmAOYQQTUuP
zvMQ&moHNQ-DAOR$Y?+kfd2S%dF`;Z-<_YOLXNe^RSFizqs;cihjiG?dgeX11e1Aj9
zWXjMwriijNNvML?$QGk5F$TialX51i4n%`?CuBU9nTfH)=nGQAw#aTnQkHny68%_C
zv9`z~3CP2a><2`11=VG_A)Or=Ip?#*IU-bD3YMZj-zLO)qb-|~q->liOVA4ZRPgFL
z<Hr8Hnf}es?4a)l9`EnTxhyh^u4BDy@gXx$4(}Z~1Z0Y3RwOB#s-GG66Zdc4(zPvS
zy(Y9ZdHcvZOJ*04BGYI%9S#U2b?XU!rhj}xiiyi7PY~6DXi8YnS2ez$P+f;r7PGp*
zg>SLr8^WDLE^bh%LWZGOyQ``M#R#~mptHl=lfz7=++ssaxbcK{Q?VPLyBTaXkW)qo
zMW;0?I>LA=d>#SAX@DH4+7@G4{BcM06SXzOOi9ISivmPbkzyppRM6--5VL1G9m=iX
z9j=5zl-$yXPp?2~{4~&?_88mJwae0os01j>^4WW=E}!6hX-_03=BXsDMGPeGczk^1
z&;ImJxwyLIlOO*M#u#j~EFnWF@mcVfpTFk+_lsZf?cI(!fi@Of6=MJ<5&NDz4Rp2U
z{hJGJHZ_}tErU}b%jx)H!}k6GKa6yh!h~WquT?-yE&VicKS`#3CPmHe)N^;(@<%`V
z0T0tuYH>y8hrjzH1cE>Ri$CSnxA2cXx*>*&Uw?MXqU#Vb^X99s*gx)gc>91dhVEij
zI$m5L<kAyAP9x*!*dGTbAHbCU4B4$&bsg?-e5ZHAXpQ%g!^xd9%Mz);jU#cpB_&;2
z9L1ilwe07Srftx^V!P{c<A9q+s^tnF%1WfPMoWR<eE7=}VqkDE&Yu0<9o<!C7zf_I
zeuJB5p8oI!$J0QLnd9i#?R#o#sk?^d^#zO73O|h`?^vyt7;DJUbG+N~#kYw*N|F&|
zH*gpQy$_s*fxbVnT6ScgxWC`xqT<QALyOE;Uwngzk=3$7n3{PGSfy|S+<EtX06NAJ
zY&PdQ<QOw!@7ZlnoQ^%58GtF5Zg!;#P_mx}!sJQW{>Jb9ZyllwDG}O2j2NX*N*9V{
z%1C1hc(5+^Tq#7cACk{@T+VuJ`L#?btj}j6tg<EbD^zKt%PA89Q8lPkHW#!hWJ>Rm
zsa*LX5{3z-D|86N*bvCf`&<0f<AWpR!qZB}2i$m|>y~8*TUo}#uAmO|g6IN_JA4>R
z!<e10DWFuj4P`eW;~VfZ(X0XaJ-Q+WS9IyQEM&z7=*sQOcBpy*N)X2bq<~cR{I$#o
z>5HY-&+qhtNGbX9+i&>!Qqxrxwyx1ilZF%h?tr;?PRbJ@NusFHB4a8=9QXMBBlTj%
ztJilN_FGVrE_9$Z)oM}x4#!Z6x~!sEuZUSN^+!YsoW_BGB1nTWk#yWKy}hNaHO=ON
z?)e9t-o7TB9=W-`pc0uXjO5{-Fiwyol`c|`NclYQRAfBkctXoW)vjsk1^sc)=`a(0
zEZG6uqExX`j{SkD-=gsJ<4BC2c74ShB2$0l_;|;1vE=&WPYGs0ND4WfAbQNo;$uEr
z!~;1;hOvinK&yahT5>4#(|37<5InuS;OWg}=`Zm!F<xWZ3skda+}~p>gRYk(0Xog3
z;F$UyVH&8cAx3ew?nc_CCd9zppLiJeR9dpSxaKgy=eIqdy*}`ZuU<3s11Ux7rY+eK
zUF3%2cpx7Ss1$f{b<K;LC5@3x<3LUU7ZNEt?q7e)+iza6>?}q_e4NqQ;!XpXo#uxZ
zrHtv5Cu{!g7hAU52maMB6In|BiywSQt0aDy$mjC4oHajx9r!Q)@h7YhTx}yp2&Pld
z@$D_6^UTh(Y8z}-kwU?rXQ9YSF}ayocMv3Q-m+P5c>2*NeE2)RLyCca@vq*&V3GkT
z(3QdMCx-KYesN*Yq69XLF=St8){C}A9gavTscQoqm?n=p9mz_S7RPgm)AyT);Cz6Q
zj6Sg4A93S=9eZBgJu;qVR?8OK)P%#4wyAMh@cMD$;v%zIuc+IWEEUp}BCT%OQ7;<e
z*fTzCSu}>{7i%PuXBB+$_Q=C&qOyizn%Eys9QFfQ3D%36N;dR^<954;lxen_J_NKj
zymzr8=LV-_v1rRo)iyX+nhDKjg)o}KICB^#7K<fPYhK+x@UY*pXd9GJ<QTEmFb~6D
zd(i!@jfDOFfni^?L6<3R3BXS$gp}CD2BbcBIoNYeXIcDh)q*T=H&Xd0c}EKK_qN?E
zG*KBy37taex^{(-D|LmWA}foCi7@ww<WbfZNi0m%IbeO@cDF6{q*a5EWn<&?fP44?
zsV$Q+sOd!AED%*kRHod5`YqXyMSjWw6B{xTDF?Kzii1}v)I2~Nz^saSfZ`cX!NYkt
z4#n*iV`Mt@RI7+KhW_ymnG;RdmYj$yjdV#${BYpiDhV^w>4*}NcC}>a2aXS0?qlSm
z^;6biVt%|u*oNlW2Po5(9tvZqT;TBb8>DbN|L6m{iz}3nR5G!<-*N0G)~gkDQ=z41
z9>#*rkRWV@%#zAhoE{#DyRWcSQ*N0UR96#sZxOyn<UnIuh=J4Tgl!A5W|&JrOU)i3
z1a9`Ix+4ZhRo4Z_F%F~<AqJEX4AWdPMT<3Q9LUb0gkn4$81|32u`dMLySGSV=uZQ~
zZpQ<e?RL-R{ZFWx4nL2mFkmW!HZ{@DoF4bc)fKwEWS$*&Uwy&F)w%>&-|;p`2InUb
z<wli7q^cXFl87|ZwN<fd&wIi+GKIkG9MLCo3h1^5%J&B(Mo7BalH9~Jp6HY2XRluK
z+1<qUREm<?w!_wS!FEjJxpFnpb~W=j;*Up`wc&DAqm*QJ4kHCtmyq1;?H&F0o<$`w
zo0{X%@$h)y;UTf91X^S|E&1ttYr2~!JiET)|NSSQad+JE&;Q~t*!Det=MR1#oM+hY
zsfFMtpL|I7(Fgd}(0}`LnyMnFj2lOW$zdzYa@Ccfu#~im7VjO?IJ4ZW(W?!nyQB}_
zAmxFlS6BS-;}2L?irZJO{@U=`fA*Y5j#$m^(Ib*THGx)?8|dU_D%;T16-J+nl#WN1
zn+w{uX16^soDL{!(Y8VheeTil#8ikE<RYo1tQcJ)q{z2le$C@?!qgR4R~uqXv~9=x
zA3W!quio-_awMS;Qj|JlX%M=!bjHqOwZ_GOF^Zd;6_-~l;yhC0ndh17b&ZmWyDjX7
z$is2wX5H}qlLb;pwuhdF(^$+pgCOKgYaRDOus!zYie7LraC`U2#~(hYThtg?@GlX?
z{xoB%26L{dK23$GrL;yONwcG?D@^!X_JP28?x5?kXr@SrzRUq+DNYLgfh<6ta}Oz>
zV-i9@aAnARo^zz&kvV`WDdZ3xGS6gXkm4M~(-INpf)Y>zY3@s~EQNDLXDW+=Y094q
zRAC^@1LJr?PCe38EH0j+#{sPcva<O8z;Pb%?TBud<lu<YiRh-HmFG-IkvI$>6e35I
zGbmXqMorSF_=;RPvan<#2~}n_2vQyjFfS!A6%~=R{hFu^^|?wmj;Vy3C_UC$>+kD3
z|HkzyXEv9YTs(itxV;A=%bRN^2c}xm-@Yb^MA)Bj#|MPaxT%yi)$M{L6`QBe=q{F+
zx~0w;8D?U;DAv(=E(mU^2r&?HpzS)EZb98;d`>*oH|R47APa@5+EUj#PjroC5)q*-
zAsZ%L(LVi<G#)5zF~vbdaA;d&ny%QMN0^R#5@4+?p~8AWI-IDxj^)J#>gtmDwCC{n
z$kZQ6(J_uCFgNuB+8SIaE<r5>yW7{y#~t-z&GqxAv|U~5dHVyyVaIryx%}W$s&2)6
zxMw(xBsXCyb7mmDgYGUM%fvA5v8E!pnNeh_s=?Tn%5+TA@q06gks4#l4PHu(_hTVH
zOGOL`qa@2^hZ{AY50U@y%Ljh>>VdI85yl>^HOrf)RO=18t4PTa22Tv7muEU03FFAJ
z0wd3IMb1R;$SQNS={W87eDgT+;dR65*faKH2>~8@hB2a3MCQ!<?{9ea{E~JtGY?0;
z*^d0`?m(@85b@(k6atLl?#r+E*RS4U8_Q2W=n5*MZfLAxy;&k<F_m~%65khBS3H0D
zjP1iC${18rGy92~^^)KJ_<fqi5_jD5W*m5!9PjdW2m!~Fr*{d2M%kL#IgXP<9F8Ox
zSzcaoI+Yofn;ar1nx>+zE!CN9k%K2WhptO40J+%Gop*@yEK>?eCh~ea^FRIRH?#`c
z#<H;$7mEc6$>QR&(6-x}@i6fEv1fg;qN*CwJd#r}(%7a&Da|Bc_4JCT-3lc#!}gKn
z123KzBSB>qSL=#jy*jd7HhlDaMGBs~-N-n3E>|7vw&HPrWDfEFPuYL<+_s)+df4^G
z5$Bv=TzMyMvdFF~c2EbkWm}LXx#1%N?H}M1|0w?f9~p)X8!(^~$yTe?K^>}8fmKCv
zbEg%*IMYaP^uareg4I>zstNFbAPMfZ<`{Fl<9Y7;y6CLsepv7+qjF-?)`Z}xjN$ul
z-*f$JbJ~LirX_;X?1vd46sx9X)Ak$=j)7tzU3E3R)qFJ$f8}=YH^z9E>BzjhD@$ro
zmSQ3YtP*7>K^b$oc`QUxt3g?eE)SR-5?Sc-V<{9$mLyin5pimq=M>3Wmwg0)U&<Rn
z%4IQlYOhLd$YfH=<asIoA41Rw*jPhsI=l~f?+fs(eS#V95NRO}kIX(|Uv(w%<vp2<
zZhBIf$-$S423XvT2o9#AVXgc1$@h~H{w>a2A(TZJi*p{OCE1^ds1j_|K?sQ83Ulbx
zP7Z#;%`>*Lc(*{aDmVB?y?|t@rs4YO3x0L;7RWSJjnRhLkA!ic?iw<g*T4EDckkbq
z<V91_tyY+-<?6G~XwR?8!*c&f7)Es6a#nXNZi2ceRUK70(injRq90MJVi^WPj)<HX
z4qK#Y33XY-Rw|#~94ubfxTdA8dr*dX3B;Ib)*S@T5+Z3C5yr5(e$L@efQz)N9<3D1
zFwk~2Pp&RbyED&ue@3beDwO?~Un15P>S(=cS#8c}>#B@ZtVKnSmIhnZXsb}#5Msio
z40QwEGk*UoW<Mgr$kP{B9HxP3oc?Ahtg3X9oS(_TffghTG0q&^krX3J8MLsZbV8d8
z@$0=t5~m(Q&LvW=B(HA|{J}5oY2tx@e6?_>J=?n-_s5CXKl=`S{VCnmlal;9g;K_G
zU>*i+N(GSVBg?WNlBdc=(W*#PCS#1`mv6Vcedp0}#<@UMsZ*Lxa^5JKt^pxg^%ei%
zhgV$R4s1-KH8WwJODQ;bgcRK0-tnjZ{$KIazx`wC=Et-#^I^a6d}V3-ij+m!*DVuS
zB!tLZzj(^|#fHX|7JrJFlm$U4j^oU392i1mf7pLi5=#idhwZ@KZbB(Tw`yR7!#ol8
zdsen4gv|ap;1(E$BSIAXlJkzuy1^|Aj}Ke?G~()ps%}p-;Cy2El~7Pl1@`!G@C<Y2
zdS%%7kuN@dMT{O(uLwCZEi-jnGrGk6=n8s0M56Raqfn|Y1n2R9(wfT`FA&PG3`b;D
zA?-{$48)XZy9T7>$>iuyeWTp~i}PGutXOvy5(pXYw|lHHIFVRp4~P;GmWsBnNFuY{
zO-Oq{Ny#uyv|V}dFTQx-Yirr>ClHy-^A&xqs3_Q-zv{>8H!jfwf_a{(*XQ_@F{?Ae
zam(G~UHR`w#o6UGdVPITtY+Liu#6*8mD6u<GufAgcoY(?BNUVwNm*3Rr=@1V`7&>;
zY=Nwa6v;XxZG#X+Zl9GZS@|qV@`}vH8hq*)WFcuT^MG5w!o2v8Nc*?gB&d~y_7WGD
zGT*G4Q|`YIPNSPjnK90YQ)51pk@E;7Y+WQWD(o0thmM?zkOFK4G-Njt{ZveiswwSP
zK>^Q&p*X%HL{VN35&7eSCPFek+;jiU&#A2}jsMzm|L%_0U%y6KNz+&=1-Bn=xV_(_
zjHY&xplbS7bNAtn`8Z-t;ShD}bEMEz#vq-6Wg#v{jEsa(`Yg8Tn8%Ub;{&R$P}U-q
zp|TYZyCZYRZ2GRC?U#wWHxJ}?g9&fA8;>Lc*4Egn!swd%?3xf3ro)clzbEC4wxtjk
zL%>Q9M$<lhj<hv?nW>CK*qYCO_$6+c3x<B0kyT53`JA)Wnmp}j+6HAS9&g^CD9<%9
z1>(ut7fDFdU@lj<Wg)}`V>HCbs=pxi8L7TLZR>vf$J&~<KZ7h-#vM`{jIB|W!-kZ)
z1Uhor0ftz14&g*KfP^~j5rYf7|M0-~-#yZXBfs;q<v;xgzYi#OyCc8)<}Lr9fBDD!
zi*G)lH&<u@^YMr=1&QO`f+f&YYlg8r7*bfuTTt|5Khs*pXHV9A_s(&5x8raaP(rZj
zG@o3b(Vw>*4^t@`s)AA3tUFeRnNKe}wD63(Tdtq}fKDdlagP-v|HVK2f~U{Una6>?
zZ+V;(PFb!lFAC2rW$LuS`Gw={2aFJ`&d#Z;4us&%o7ddlKa_`3-BDjZCqCS<zxe>F
z{iyF^bOCf>Udfz`cx#+MsB+T2dtewBQnD<;BP3Kx;6<TqSJt4FVxA|`@lf85q$r^W
zC<oKvO27wsDl~%+neKYU<R+Z+1Q&UK^Ny4goAsL6!xR<{<4kQ7+FB4AV=aE#6P!cn
zx`b1P1K~K5LP99T^#Bb)^1hQNWK!V`rpT9{JS%0$7`Uo>c8lQr@*FK9`@_uf!y{+w
zp6gXlyLi5y9>}5uQ&I%)J<dzE#~o2>KL7L+nl>^oo-8s_Nu2li5LvBz8e2LouU=fR
zy}Ki);jjF={|ziGkfuV~ie*_Cr;)}7WRBG93#3-;@9&sHqRoOF0%U=*4Tt@nxa?SG
ziO?FiOr#i@4twg_P&Ivt><V9Eb#=pZI27(yRTsTw_QWt_n!cFVa)bm)B$Too#u23*
zN(eNGUJF8y2x0Jk!PGUX-H_T#<kKIM{mkyeuaLu*ZnY^CTCG7;gk0=yh1r!)Z(L=p
zwoJu)5uJzC3KbS)obd@#E)7j(Ou=xZM5YL-N`EqzQD0Q$!k{eUj5dhi@jiUi4I;#8
zDi|D1zaqQHdVP(w9gl}UW4qt8-xVkL>>Ssxp3`qONFg!C&{$jC&02Hy<Qc{q+;||z
z3F!;EBp!CmaiUqBQ+4aYGcy&!1e%MB0zIsIM2ZZP=l$KDMQGmM?dWynqOb7Ik>pI%
z_jDH<{0Z|BQ=)2HHdjx<R7~4@?!Wyd!Fj5tqHbG=5vesP*XXJNk+D+#y0xD|q|%1#
z>nFHr!Ud1S5$8R_{W}gfnW|kQssi8?svz*gvXF9B{xe{S5|bdyOmqX8L}es8`H!v;
z3Iui4&{Q=sWt3Q#C^x4Na4ftrT{mc>Nq!>C3yWLmHWwu>EF!4VNDn@+zuh6`BR{@Y
z{P?pAHlO|wl{0mBi>)oUH+THucRyn~4kx|BOpL0K8<quWGrfWFII!CviEbhuc2ovb
z4#*TRQt*T47pxnDpJ%rFBd>4wjLSXGt^#vXq&Hjsaa+cMDM4oy`n*F+iJQMi#48Af
z_jj-P>h^m=rFk%xXSGED*B5L0r=Rik2cL2O-B-lZ0W^ky)rz!uvQTu-K4X!Z@otOi
zBYka=ZAW+UgcmQKb6!{c@vlCD6L<PPifVR^RGP=f1Bc0xg=Th%C5}v!BjA}rVfeL`
zVaa7jTN!gQL+WB0M1T~f8J|uEUMYaWw3Cw4628GjIF1WaDsFFX$su!jc?CkS8)v2|
zGy8IFFVl!}3v=+~IOCR?cC|rJ{yCJAXsfBH$yuO{WH}aU?r=Duw4!M%lCU&wO-O;d
zv6w2;^c`QmyhcTs?{;+Bvc!mOI)Yo6ePp^nAWXS7=c(|uhH*wK0ktViw!C1B<kgES
z&YGI4l6?00OX_cizj9jg8y+Bu<9^S+(nL3*g&>j0bq`^|NL!Ll%gmAkRoxOpC<{|<
z=(=mFs>8b?dhgcf)YdS~GciSuH@CPb&{c)fg03sGy&xPYgCZj970!Fo2^%j?uaH83
z5)q{;j{7@g%Gj!*lM(M6B6_T;ks=e5!$;4!y<_|STUwJ*wn4{4^u82_VnV12K`eks
zDT!_-3q=T?7z+bSDoNX~koz5TZpf+>0aG-@6F^V79oCkcDJ>)9NE8aK4OvQ(Qb=Wq
ze);I7ZO)nQ{G6uk7>@fAK|O4#>xSR`$(Jl{!8y-7&%~H`^7I+&i%aHdrfSRb`0!9P
zvD<gwVpr!VRbhlY#b%2S!!|3VY48c~6Q=HHx=T`gBqFIdSCA9O<M-_EZU{o)wB|uq
zJW%nnZwjPPbeO6}S%I`II%kv-j1TYFzkNs5w$%NaFde8_kh5VKCv0VC`o8R_mWdp5
z*)cf>ZXqULc2Y``LfMTeVQDs(q~J>oTbc6RO%u`2EN)_b*<q`iaU8k5dn6wo3CBlL
z@;rI@9GMf#<WAqmZ-0yw?wKR`^43~aO$k=yGAlRIkWa-|0VE|6D72~2rusEVG{sON
zq}wAljy!1+Km5TfHtUY%u%lXSP}b5lmM3Q|XI;a0^N7flUQbH+5Qr(WS~Xm(8y+9`
zgm^*9;Bw}yX{eOIO(UTyR=a-H;f5m@>k26a-@bq3m#^<QKPzuAbq*K;oiS*w2|3|n
z;{Av3*&X*BonQ=s$8kpchKtKfS_^e0iC&SsW86LBhmpGJ&`K8(+{Gocb2#U)t)*{U
ze)^C80n=eeT4tKfIsN%1!|svo`_~`$K+l=lmKM1S0gWt$+7m|paGVOmEoB~uf!#22
zb#cb^*_!)dVhIteBUM$CmBv;@$)L3)YefqGlYZEHf1+eqq?EXj7^i@J+OgdqAr*^X
z-}UScJL<Y(cbM58XPzz&l^se#8b?S0-L@nkEi<aB3o?B;;FqP0&4nOq!Rma)H2G4E
zAoH-BsM?0AD#4{`aa^5ms4L4fOqfdWqff7Rb9dmlEEuCoOj^K7+w!BIiw3}_j1r0a
z{lYkV_QS}tixr!`#+=9v^{OwW$-m<5{5Rv=bHZwaZE7G>t*;p8iDjA4O-FsUChTqy
zS+EQ{Y}K)NN4qK<exWrYX-qBz%q%i~8Ck|6d|!N|Zd;nJr{S~@nP)tTc6EU}48>|E
z3*OnsNLb1kMv@taEltytW5j4(N@E0$ZX_%N-D-nUl4*C(@$sH{9I&ev?_9xe`!XxG
zRaL-?q*GH|Ax)X>r(6)(NJR`~aas30F+i+)biG1^iFx-(iUDP-f`-neK`-4xmV$U1
zYXs-WF;SmgA-2blo4aS|HW!4L$vz+|L-I4>#O!LCj<%Ob6h4%<x_HvLDgdQft<R}V
z!@SI>?u`2UIqB|O{QWz;F!&@`S%EPEp=1F+njY1yk>h~vR@CR$EX%^`Y|rof_)~I@
z9D?HQ!O{0E-MYrOBZP^tw3ziZVLGrJ3ND1~abaP;8$mU!J413KNk|UEfmLp)bw>Ib
zgds_Vb2HKT0#g)*oFY|iF}5Mag48#*LNqP9T0siTVT6>an+j_Ljn$Am@wlg!f~C@E
zI@Ict<?NEA6iOaJnU74xm4|wC6T(=cp9sm9Vy5}aLZU>9baH@{hX_by@(5#*LZ9$B
z3SBAG)taiQc=GZkaXzp-yeTetB}r~zWfi^ExcyiNz`hWfo4Ub=$ko+3&z?Qu@a_#^
z9<W`*MR&<rS7Hp#FNHB?Enx}Nm0|Ig^G!>NkzsU1w=l)b|M$;#42$QyHmvKCz^s~v
z^=8dF1bUsQjH0P4nzk-dGUqW$Q7emcBZr%}$m}@YAF0$QJpJSomct|L9;m8{rZRMW
zhm}wH;6}2wtj^B4cyi6v<r&Mo@b7;1^~c;A)NRG=6Yk*vLb5xKj6smcg~uT<hs@$L
zw}**&2?Xb9+6L<ugvd1K9h1vUKL6T;XtYMw4StzXO5vO%r4oek-cuP1La<t`h*I(9
ze#8=aq6CW@3lU@X><%NlL(MWR^p!+}Knwv@8LHKW7y=|;ND#j4`Jxc4)~5)#6dZR)
zOl4`ij%iuwsd#pNPD+{GcE=N2R0x~Crfq9tPK<|{aT+;Jj@n9$v4jwLJUU|DW2!Y7
z$>VY0`FYPUM}Bp)MI(51T{=@SMPkSt4*L=?`J13jfMTN};itJg3}Qe@Xx8V%6wzUV
z9LTaoD8UjvS(}n@@^e|NYl}<{U9HMYTPuRnTs?V;6q#yOWI;f2A*6syk#RU6<~=z&
zOxr<LSf!9nOZJ{=o@m;R90DP_f?ZaY%T+~I6}GCVsv0i@+iBvseI%=jBr9eYkU1d9
z7$pk;F-EenWoLmzsFF%jQlpJVD}@vpKaa?!r4lo{;|@|H2gkh3RCRk=x=E&a;c<B6
zY_-8u4YDeWIW3oxR&zt~gM7?6OF(iZ6RytA(N#sxsa%N47P_sF1vg_gBnUoH+Y-x_
zs<_>&)dsyj$JRBC6}WbVYx;8Hs@v1a@d4dkQ(b(DSY4L<yKc*O7eh(gO#{6M*wu;)
z-Eh6}C{vM|o_G+X;}$U=G5wml)+|$z97$EtcRfi|3@%~}un0yUPdk~+VVtP@b7<D3
zVZY2s?~zL5mVvw+(5bN3+%gffKaH$Pb66yYCQ@Il@pC9Vw|N2Ki(c4DLbD>tOBQLF
zV<b%ns^mVl_*_CAWP~WO+z41x5gCcgo}>*@RRr(RwnhqrDaSs%KR{5$Nu4rz|43Wa
zTz&dEIe5nH165TblYTmuqM|Pf=)TeT_XDEs$SIUS#Nu$v#N~O<)%6vBJ`QZ(ec*Tg
z{!dvOL0v1fRJd`bvl62OLq0Oik%(opTH$Apm5Om%2sv=Qt}(OaI63w^M_X46+4J(%
zEB^M6evjAR{{h|=?%A`a=Y09QKO|<sG&`bg$R}Kp)D_dTWB2A;axB-cwFZ$g%l@vA
z6;eP^-lCS{9fNYLHWyq!IY;S_!?~NvVvNQi*bkofhscs4QpQKa5;)F*$I-FOk>fa#
zgrIL4S}Un-#pork?+!!~{3T2^<zJ`rgb=8c3{Rev6jDiaT{B0;arAUH^F`M&FOI`;
zVw{R>P3R)Zi@`AuM+6Dm_AJ8@LLj-Kj*zmbArzAFa74+X>qSY<&d*7}s#f?sF&uV;
zRIaJYTDJSciEUPj*h_*Z*H>)Z!ghcBb)os;xNtjooT%BXdqS4Hx!aMlpsy?9><W2%
zSs(@ac1>l>>7e^Jd;<|0gwBu!vTix-w{*G)S9H}e%_A{os+_5tj_I%~=&0<NcK0Cn
zRNV?|Eq?w$k_M_4;md1myI!M}BqVSlBHV-}A(yEb)ubR~$>b|=2TWC!2yyV>MpV}W
znJmh3D<H^HarN2f$Yx!>WK;A0+i&>!AO9hj&z_>{HI>yErKx05c&fT7O?y$^GSC7g
z4I(R)&RA2j^(1#<VS%y+s~fzV@QVi|wq2vEnsL|@rJ-uqn7YQ;ngICVA)a)e38a=E
z&F;&=(Ka<cRe&hkQlU%MTmL0P433A9jHIoLds!PpN|Et6aQX6Ok#o%>n`l^SN!*;F
zRYpj~tP~+wV%rnC3$$s<=}5x;%{yiviBfU6yC)BitWh+bq!*#g$m2{9f(eO9Gj_gb
zb$(60ykuTHK{dqjz~U#&sz=w7x_^Q^9+`)!2)(kxbm!RhC0Z!*IIw^JhWod_Acu)(
zPcCr|bb=HYj4U^|4~aYvsN^69rtN{463R9tPU(e|@bicgf|O^5-GODE=o`x-AjkM|
zF<vsH2_Za!T;QHsU}S|CMJ8pdnoRX;nu1aar5m(X%+r9RoM!j8kA${DwJpR<)Adv~
zLmWtMg8Blvc}lXkh@6SbOfof)h|x353nB-uF3+gyVl;ew_l7sW+7O;z^4Y5^a`0@{
zJ#}@4cMHLLhP~kOyyfBE)6|CADEh{5o<O9?j~mO91aEE+7;E|V?h&08KmFn->^}U8
zeVllDy+KMr)6_JZ3*O&s5mNHx@{+332$49yJR>Yef*&C&atH__)u{-ZgQu@0%632^
zPxnk=$8>l1mzdq(`UxpB$Hd(*^Y(G#=CGiQCWOr5Ba4qrzU=OTkHiGlXj<EIzG=uI
zaJFf9_jn+s5=tmR8d1_$xlBL=f~@{0y|7f8*_F4EX<TTkn#cV>I`w`saD8<NvLwv9
zl`6yzp_sOh#IPWxz}ou6O|vws4G6(<^8C1QMk>S2_iw1zZDEInz<PCt)CRw};sw!~
zwzDWHnP-Jmux@L*YQ-A?caO)y6BL4&1ULHwNlG4e2T}}Ntvi17<ebV1&Nm$|UR)Es
zBl(gLR7(C8kB{F1Q;b5Agd%279uCFKha}6E`LJbvyrJq=><>o{4<AS&7FkYoXr<8Y
z8N)Pi+&y6GvK;q|C;O2Y6Cnof#{J0yaQec8z=epLC)|8QNrN%E<o=zf(t_1yMNV*Y
z`vD~~AqncbA|dcuapnU3<xA3qW!UZb@aAhChK2d|K&(dAUBi0SU}P!?zYOB!IDu5k
zh^9gcurebEzn)2|szo)cVuf2~a!g2>S%wMqc!RGxwvV?|l%nB&tP8d!M`W`?D9wC4
zq9wfk{`-%LD&|}+cx^Px<VsiL<VBI9#OtJ#NNdm%j)xt>YjRyt8+*FoGxu-5E418{
zuwg`0HC0=qwZSh7m4O^*mid4Q6KTqbdB^^{pEEx0*}V86-PLpIx+UCyhtLWU9Lx9!
zO;56kIM2li6DJ<O`z6EUBbTo}M`#M(UuvQ|k{)L;8J7#UODkE%Jf`doL<HTntlJ(x
zZrT6QJ8C15;u3@;go3nDrf6ZMIpy?2Dkn!I(pZoZnIg-yXIUl?r5tLMVAEN~9B{*k
zZmW-cc$`SpZYC@aq(cx9$rcApIuV@H3__Q;>vWQ#A<?R$syrzzgr!8g`*qK7{|H%7
z*9~!*VcC)9J?hdBg5dCL^=?LKN%BQtmVrtuvO6MWWc}otZ~o*@x%=j8#@&|v-3`y5
zp7Z?Wb84%}`z=piy+ldDbl4KeY*sbP>{x>2vTHeP_cRvPl_kao@8PPoyngc)cMl&(
zDN-5DC!ahc3eEoRmJkB_$BF(zvZ|({8W1N}P8A5E1Zm3Unvkjn1W`o1n;_*9Mu`)_
zCt@7<7%rIrj!WY0e&MSRJ8qAj$8n~sn-biaJ&|(2h$$jfNhXDyX{=^*wkCPc^C@um
z*zoS|NXof5J3@iL#t={c+wvx_jK)-!Kw_FbU7cw94(C1Je|Y4)FW&L%r`Nw8vo0Z$
zG03_m`Gv5|q!ci9jhjaNG@<H>s%;2vL5a+K9C`R~i;|ikBw8!h=WEWg;%@s$f3_xN
z*ll+RGJRYA%xztBwh4#~$vgU{Vwz@BQeX^=%XrFLd`OT(WYbh^+KQ)_JzcHJ*jG!o
zi^uzTN)G;qpWgseltdqK<M3;lk57^~d*bu}IpI@cetbZk_&?g#SYw$Uk4z~e{cVv!
z#Tn-vGAEobmM|9_$M<jO`%?i)Bz)EcHy5u&h=9g+8-x<LVL&8@t$VU4eGQit(l(?p
z62?-_LTL8C_zRwm1M=oI-+%oJIwM%EFIi${zdIterCC>G13Aqs&J%KBBiX7c50c~y
zX*d>1l#g>E>sBCaKnV!*k;lgp`!!X?G#=R9zC~)ws4Vjw5XW08t&z4SPIp8k`~8l7
zbB6ODH*XgLhIwI{CcIB*Q!db)bGfijlCH2g)G6~VfT~@U!jf(1pT1(;ZyAR@huu99
z=v9T#vYg&ev;ZX&az2pnzb9(2DWYYf-}JO?MU@w%$hfLy+TSt;k7&+`tEXTrLJc4y
z54T6QH;>r+k(@KDC(lTJ!TE*$a#Mc%TyajMM2Sw4@xz|!@dm9eF$b3YJ%{gq$&2St
zOL}Zs@JreAbn6v~;u9%irifGxqOMNfl?W-|=c&xFg`jF0qy%d#s*OfhJH{>8rh$(P
z@d}{s&Zt+{r05X%g6#o>5=bEkZX%(O+LpI+sff!|N~*>pWI*YK^~Hv|>o{z;oIiO=
zb_3pz#A%0cM`V!r<{I5~{O%7w<>&Wb^M0PFjXf>$XQrI_)z@$M^7B_zP0bWD{`f$*
zw!D8^&>olPYvL5?*K3s07-RVQtBh^~9gf*;Sz@HiktGDAQ%DstQlJfdalPi<!;b&<
zzyE*G)QT^D^eLrUQ!~vow_kmStb6+Q0-48xt%-r0Pj*hBPC!pbB0)}tIhQjb94XMk
z7VoCwkjo$WT?oPZgJVcLzP&v#xQt7YX&Dg+(qEDza!FAYYiMSB7-*|C>-Cy>9C&fL
z=EKd7_cxEkC~#zqsZgr4=)=i4T35E<|E0irM|Hm9$&(FD)A087dwi0FnAx1KkwTyw
zi*u3fG;+Pxm>5ZJA-b99J%{6gj378ixWAz>7J&pK@v|dDI37nHwnye)zUIe2{etV4
z&xk&dgU9HKrm4`T91@i=R87PF-3=Lu3Zaw{&sOXQUxa+>RPN5u))lrfSS5MfAMk$R
zix(Hnha*DCl1>xFzW7&teq#zN3_H@=^03>{tTyO&MF^1`Gp25!>KJwhx^7)0C(BHV
zrCL88cGPY_8H+XsKO8_xA`&eV)@V}jEXN(|^$MRP%2p5~a|qP7!migC+v42F>iiO0
zy}(Tq!U!7M5QY&}PuV=H(K%q0z*a5(a7S=Q;^KJm`RC}XFZn@&y#K)Q%`b4vLSrOW
zKwK7<1hr}KZf1Mf;SP@oGDa4Yo>5R4eS+Mz=xLy`h8%SnPt6lm4hY>4^?VXbC2Z3n
z=TI>A+OQucWOSUbS5$qE()S;6i<F<NVRnR*bSnd<s_@fDT~{<+PmY=KuxHrs5G1O)
zVmTg(-lOW7%FWbGkJgfKnv=#Pz=y&$GYxSbu(~XO!>}cYfaq4VXBXt-$nNzoQA*-d
zfexzmC0)BhH!E~|hC<-w9btS#H$C?HIk=gJ?LFf(V%wf(wZWvqKoi{xW2$nK3rc9g
zEezWS+&mI}V1IXmwiVYeUZJ(&xPL?{aWY>*J{2`<qYEzFpD-vCV`d78QDmC7C8WeW
zPfX*$*~J-E+m|1ioH*{c#27wmqL_i%4a8W+LBY=$laabC;Zq7|)8fVh+IY0K$Rvny
zB;{241-inRh`pAG46Cy>T~{$b+!EbLng-&qg@Z%A6f|AW_0uaV@hkGOAlr@{9LbeZ
z-`fvcwl^Pm`s9LFpT6Yg-5Y-QGuV%jpZ&!f{Na)7t95CEzk0#_;|{xd!mD;eKHO1v
z8-DrKw<sZ~+lFPFFuIh1XBP=EuxU$w;k-ZR$@Ljy6zpyuKqS6-JMiUCE?I9@NL6@o
ziBx{DQ#!o`OTZB00&yV`kwT+Xi}Qh;OY}B{nPvI7HxMb{`-g$2s^c&%EWTjJW6tHD
z1B#C)Q_vN;i0W=1wmiL9A(bQxiP46Obw^(t_T$VnEYL#LTePmxN)b+x-V|XtIxa8Q
zeERZ|%d?u};PDY+_Lycv-Ly>8LUr2XL^Qf?u`V+VflNZA#JE4=yf1rJDe=xDQ$mE&
z50WwBrWxlvlaGA(`aNf>73;nw`-HWbCr{RdQ2aw$3Tjgh3?FW`9F8L*2b44%=g8fD
zEcybWIQoRmk*iJ3le4abeHK^T^3J1A{TOR)+3o%npWn#BT7qW^iD)XkG=<%xEaC*p
z4dYC^I>R|nb~9&ZYmSe%rPyROUE6>zV}#&ONt#mxQJ`6!UE#7ITZ7F3ku$bw%4x7#
zVPyFcn2wU9D^d)2UHzKHrM01AA}@(ZKpROQ5K<~mZy+}-q{y5bNUIeWuRbBVf#e+-
ziOW5vs<His=p5`1jKj>=zx)bAq>_oYuc<17HM-=rrC{A!Hdil+<^nelxWhuS4St@m
z-MWadNT5qcB+Mgc*H2N&5rzZ9!yUEOA3bxILJ~q^9!K2bs8>a3<efvxk(eRogwh3>
zVY?QqtKwr9xx{tnV?nkng;e@Pvpuz>l_lsheyJ;kTRg|(fl4<hBd9QBQI?D0IDx6?
z&YvK4Y1Yq&hk|4ub|^nlTZ^w-^lC-7YRS2bmqG{}mo3>ZB&Dgl9+6WiO~r&zrGK(a
z16o_oo_|6Lq1<G$^o?@TWG&gx9&0sf*^-T-GMb=eL7}^uu%7@$B_cvA4oAoF{*lN1
zj(J&FUwDSY$h05OD*ykZ_<zgyAt;L1dDx@1g)DHhBSeSMrjU&H2c*cX`ZcBs$n4PC
zFduFh$2}TJr_KrULfiMGsh~?@Jkg40&+_n&ya>4doWrm%FB4%-WX_90&uDVYyuaJ=
z{hN1u@#-0$eD(?V4_n5?^Wzt1{EM%S{HwosVDt42|L~^{HxKL{4}AH1zsvoHfk_$8
zo}8h+d(V?+=kzY|%Rl{!!*QmqEs2Eo5`O}os>(1PkLYg0JWqW8<_=-j1y_?6QZgq9
zp(drJwCz&>A<H#)l4Yg1ls|**iQ@pG1|l)fr(gSz`6i%Ig2pI}G7QUv%8FQGL;xpb
zKPMUhI2XBpD8yOsX0njP3~gO=zA7k`X_{G<nW}DRn+h!r`yn7xiKJU=S#Mf|gmH2#
zb0GUj(<!blFPOZ;J4e^|Ai*6M_J@&89?3Bvl|nR%CPOVUITkOHB2!sSa0_)~I9pY;
zmBJXsVe$|&Z@&7Lr=Pu`>Muxc#x@l)Cyv_#Q;JN^v$mG*#ZwOB#Mj@x#m9)xiR0o4
z<U$5m9M`=?iwrSx)^^xcb6V!-62LH)^c$FDzu^re6SKm$Ys|_LV`Lmhuoc5R(MpBR
z<;Rqmk+$M6j7%<2R~9fh9}!`pvWBb63yzB?M}ZL$KToK7g{&@^LZDu55y>OTw3}<X
z>rZjV`;#oC#5Zl-A{oH03kg^omc=o<KvfCm#iLWfo>Yx3l+84tj%CIwl))N7pvIXR
zV+=JHWf!6<a@A5>!_%uZUwrmydFl@XrnWrZZ!uNP?(UAmJfXBfj8Ls5qFRxJKqy7O
zS{00iP*h!y)aUGP-Y`xMcl@6I;t8Acr-XOkFdmN|_j^S_=TGy*5<Fc~le{Me$FeL~
zQ_){Np<XvArO{ThIX@>y7>~zdj}!9r;3#&$BqRw*6b9)R!ZI>@M=cUE8~ifiyl0t)
zl6zN$q&rVD(>PGA6n>tFe#hbA10i@8H$x1xbxXgwpy@i4kj1X2YnC)HE(?B1q{AJy
z?U2@DwL%JkFg5i`)1GN!@EpepLc|1z&NC?}2n)_iq=3hp8!k30WG4z&Oytw()v>?*
zfE!0DQ=ye4OUbkZ_P2LrtugI})r%iexy0_~4dd;bGV=PE2YN09>Kp>gI5Op#rk7Y-
zp=3qQMQntuncPC&Ef^ah9?SkBCdTa}_>2=CF;CRbpOTd%j7NeSSmwE;4X1^W9uTr)
z*dItS;id!bc)%D#Or`AhxSx3a{*f;~d&%XiPq}{gf!#dwd)GZbUL^naE6@G*$iMtI
ze}PbmzOpoxWE>pXHsm}}wLPEw=yw>$ndCgven;C_o&w`>CJ4zA6NlMx^Wl!e@Qxt}
zzJ2q^=b!hiTg_qjfYB9FNlaB2L{M-gt(nUBNGZ*-9Eic8>kURlgv{jS0U;@vqSM>U
zZ~OR|_&@!pKjiQJ=o5bScF&)`-SKDd2EKj&z;s#wODRj@E%9rnUJMQwJcnuKaW@cr
z*@s@AuNbF=!!Qz4Vj72&SPF7Fr6x0g;dmIhyH~W&pW>E@vop(QudayRb37dByPkHv
zW_E%7e#g7}BkN}kwomwJfTd6pd?;_eLO{FfX#0jRPDD2|4<qZ&vT1AP+2J#IAvoMV
z5=MuvE1W!aR8EK0WnMsNv^7|-k;-r!XBMCFQGzN;X04#LhQ78XKtk!%SS`7|y+ese
z*Eb~v`qw-}{`w`ll$g53`#_i$OxvMW8~pB$w0LISvTkcMfs{(VT1Z6{1)O;O>NB>F
z59OpOGRNV-f}t^rj3OC}pAX3CP<$S4qHY?pR4i^Gz56-BR3Ifn3asg{eMx;R!yc&w
z)38S>!@AOCJmD7B>lHaBV$3BzM=s6ulZV7f^=rY?G?hhc3=v}-rtQfDyqD;{p_ZBc
z>>RBn))<uSxO)1G90Niq%;}AUED=E=C8Q`=TtFC0Q&$X+H)tv7yOw31+27wY?~klZ
zb(*{He(Oyl1i0bI?!ygr-(j^VZ&0~NX`+lo=L!?YPfW*&Y5NF5fhrL^YYfs05;MkV
zd`K)IQ}r7vt?=W4>98Y;#3;ZyPrE)tX^Z#8v=>jDG4CCfRCvGOrvbOj1x+p#rncoo
zB#YBUt3tWWIS_MZNrqTg=;X0&Db;0d5XPX9-~`c)gn1w@ftZR0l?+L0?9<Ol3G}k(
zcfNSWdUHm~Whv^HvQvqn+-Rw4*dIsYJfSN~wO-R*Y(N>Vo?LOU>B+*dYAx@=a)Ri6
z#N=I+lpb+GOUZ7(<FG3}5Zhl+SEjh!s|KkmqD!Qh$+2ZSls9GFze0tKnmks`MKpAG
z%W(6K{lh&{8M4T@C9q68OtV2ok5&>R1bLazO^eiqm;zZyZXXV~1ogV(>iV4V;hw=~
ze)xRD)2lT%@80ugU%ln_Vc?zHvELo}?8PPL7iZiouJGnk<hVa__4ER9-ePP~?;fio
zhbgk`4m=zNw%Y^a<oK{pJij{QZ~fp2UwrYB`|WKpe&&Fgwn&l51a#G&I!X$+WR`h{
zRGQ!&P8zDJMyN7L@XH=ImBsLH`{b0l{`3?6(f{J_@Y7{raSQuY^RNHSH~c^U#n1WU
z*B?kBo!a9dl|YF^rOScy?qSc{cegxw_5?&^-B_NUt%zCj{{Fx`jYUO}6G>#|SRNux
z*D)!<G<o*N136{ZZG{#I=ORi7Ol40hf<qaTI!Lm@Dn|&Jad#m4fbCl(0((+h_<2V7
zfJzx-3Xpa_PE<;9e%{ePxg-k9Fhw3d-12z$fRB;&^(FoJirQN49uFLM10NptRJsr;
z?sr2`OOw$mW3@tyNF_5lm!J?iA(bKt$%mUqo@G#~)Exe*$mn1D07)rvfBS*!b%oLu
zDkqlh9l<TMrsDqg7Fxs8r`M#gU<4owz%M1s<KrX2dt?@9K-dO-$`a4VJ+^6)+7g$A
z-NSwH5a*0Mdx0Ra9PTi7rtV4(Sv57&cw`xmXj`LAg;W;Ptk7CA4|`+|JiL2DcXov|
z0_O`(LdHZ0BOzykn=raA5n3MzF;K17Xk(BFLbXPy1rY+uS}YKArd@fOx?u_u6&IYF
zFkMe8B}C6SL_&x(XJ`0_I|PYow?j%pU1h>NW7aF|5(&Yhs~ULxs8d!>na#xo&p!K<
zX*jYR_hkf-6`*iFaNIsp*Om|h^AwndJ>6zSf7YCOEEbU=E)(N86r{ZO%#RbXvPdb=
zBC<FKN*9(?hItrBArrF)AyLhm&1Fqap7G%W!G}^Z%3zeFYdR`r2-AeBiiF0ysf-C?
zM8-lQt?Mf)1%iaJwv@G|u`E`FLP(94WmfFHV{sngBAX_W!ayAF2%=>QYjoX^5~Qrq
zvY<@5oSCMXj3g@!)}s0~o7I{)&-lYVyKg3{rX$73deaw;@<$d4muVnOEpc`<%{ANY
zo`?63+z&TAe|F7g)3WY+Y=4DpE#!n%iZ&UN(palm#|gWA!{O@>eE+jQE2qxo$ndzQ
zUG<1cv)zsbs~<gGSF`SG)^lK(22#I9TZ2)MV&wgH<l*5SaTwUFJK{X?{%+v!{`3Vu
zxIE|Ky5ZHi<?FBCb9<cm=G_BdeKpg*y`yU^b*1@l{;PK=X}H+5R912I^pfUm!~Ok{
z-C<&M1&OjflqK-1b;IBP;Wa<{><R1hb0&YpxdPFYcrt}}os&l?LEWA~N(FzfE0*zq
zsXK&sNK=tiad+4CLgvFqd=r^P)NuU+m`Awz71cQKkAMFuKm76;|F3`fHUIB__J`yg
z3$jDx!uk?$^WlzVaty=3lc!)Si^-)DkTTpq9$8Ys;Ycaq6p#=~5}d=ylF%9sBaM=D
zwnOVeShm)%S#6lck!dU=<|!$LIhTjbG^3=(G$)J-MM12!py?V^-{Pkuot6kwY^!nc
zXlt>JroDQC>dtZV!p-Y<y#4wOv-3>H5oI+wcn<p``^hsc3!}^UOd<1Dn$8;5wPsUk
zTB~R(jYxr<BxJ?e*_sb;-tyu7JuwD`lV|0B{`qSkAW}-6K6`>~I>zmeVZR4!Sdymi
zI;w7Am`3JhA!W%j9uQT70JM;#Fd|e*$0$m{#h^Iu9ucazgmu+0&4DEZ&dx7M(UlXV
zpQuv84<pDTgwfG4k0VJMl(J-L$RZ%LB@01X9Li0s*5@VF6&$*%IgSHLNP?ePmXY1@
z$l24+upyz8r0RMSg8lZMoM6cke|SI}!_})Fq1zR5InwoOgwoV9AhShlNTOT}iA1F(
zN);4*yDoEm=Oba6Swg11ctLOrl?=4Hr&5aj<Ksu|Kr*0f%liBbkt5@7i;t10EZOvk
zoFRF<OH50^InTJvnAsAi8G>R=1GcKLl|7ju%ltfgho1+UrlDy%l&MRc(g#qItVKyO
zEgowdy3Gb5VR!eycsLM4r0zSKrY56ku3lh-B;CHl4-?0*M~X~sYjj=F8impzPp)r?
zfO#gQP;S(}+}mi2@C!1A6O(LaKHTCCkF2ksU^dU#EhE$Jh`+flSQBGTWmiFEDx}l}
z@m+VgkT?znr04bn#t1?;BM*-x1l^NQQKsT!7PsWg_Tk8(S)){ktr}i@@sg*qVK)bY
z(6}VHK@?-B{>z)x(A70s38bHpx34*V{|g@8zvuDxk$Ie{ZDDvh=ZFY)E;0`T&z>$w
zZOe$kDvr~_uzRFi7f@l0iNh@Tqn~}l#ir#S{=L72Kyd%zj-)iwRBW!VkvXvHD(1t1
zS7(OD!@|verql5D=8<t(5JJ#5HI<UQc-Zm94_@*(IFiuphJ|@@G}iFy`8hA2Y<TtQ
z1!qq#%Nv|j*t#KN(8440Ku(Sri)L2au9Pq-v{_ph)JgCY8HLsjGDBUj5JLUduR=-5
zzxyA5&h_&r{NsQ0_n_La`{p&r{err9%K!17{;%0S4*bE-enr*+sZ6OTgus4(#4Vn>
zvJ8iTzHPC!LW+b^Yqp1p58DIY1wsOA%t>tll`3BK``cTj?>T?*gw3YID9IcHRnss{
zBlr6qAttg^2&Gt(pz2!0s>4Zvl7eR4Q8gNw)2TBgsG6F*F2Gl5YP_3C^N2DUDav)4
z!$93yzW7^TaQ*5TAKrf8{=+TH;%V9*Z4<-t!0mnpRTE+ck*TDl(UMhVIP2<P*8%Fv
zV2wGk_XNgR9=8YF5{U6{=mQa=6b1!?_WT@OwLCuV*uH(u>imi)PoFaEA2|%eDgLSH
zFQ1d$SS)v0lT*UfEn+z^9*<=PuN5&#l&KhI&$PIL+end^GQtfsM&rFA<b-gJX>m-`
zLJA9A+i`aB6qh8Xt8x1~ZeM>-LKJ^=9H^U$OeVXDFgq5zMhcDX&ggT-!~r2JN@?OU
zGaPsL@ev^n+q+w)q3kB2Tj=@?DQCvxo^E}Hv6fnv6j}~}*^P))3T2zkB}ysA!$OK4
zqagM*8`XkRL<ytI1z(>WoM@lF?IQ%_Wx*c~gn1$4(uv6=a<WJP<8+`~op@xiq&l2i
z7!EtM)n)XdAf0H(2~}vr%3v#jkp`m`N=4!_m$yc(@y;`CA89rlHqSpr*`DRNr8~9M
z=ff6DWZrG*FJ7{0S6G#i>6|DuJZ=f|0R(iFMi5x0JCv=^RHfJyLb)Mw#4Sg}=>cL?
zNhpOm65UJ!e7z!E{Q!IZ96LM^?!M)?eLN+fEJ`Yjl1#~!lbn={#{uUeNmt0a#;Ibb
zq+Q{Y#~*gc!?7UwKZ*<>u;*W*o_${4?y3gcbmh$|B!2$Qdw%uZ4G*Iy#9~yuT6grF
zK}*SxUu^iajpT=S9Pe*g4#jYH7-se(Og<19QGLt#*@pQz^WlDv6OLt`>1sm^6%U7j
zrtiz&A*D3;C*SP&2ftJB_PVa1ZrC1Y&dxkS7CnKjYjz^@nSiIW<MSo(IC@OivE3c;
zK63MTM8sH#y3&BGa3PfY$in9@F1c76UViqR_4%5NAf*Dk&@y90B<D!oU6cGw%q1!w
zmw^yG&IgpP%7`m2C{;$);zZLG<!JF+KT-<b-4Fb;|Nam8<g(*@(^6kNVVG{%9VUKo
zeZfEd-~Jum-QV->VOJvh0x}WjJTXK>jvV$!thNYMSbm><dcnK|q>S8b2Yf6&HKAll
z5RwZ(R)E!#^G%P;f#ZH+7zR2inHPr^imGWr8<dbtITSiU-;|xN5LA_>Zfb-Ct<<R-
zmN893LLf~;Wg1-0$P8hci9Vq#L9&|E^t49v`6uVRe169MF!69d5RiP>k9;`H9G8qs
z8LKs`%F-Hv7EtNGghWafSK{%@lG*K#$PAlJPY#LdIRBMz=YRbOD@MkL54c8AZ>~tv
zP*)Yd^TSV(Vj+zajfFXtk;D1(SG4^KBR5D1$H!a7<B{#-p6CYp^%`AwsNiru5HcJd
z?&#NRTv*t>dyN;Ci)Sy&NNO5MQjt(B<B@6cNRikL3ugOBjsZ7&M4Zq~#pd#oJPr)Q
zNM$uq3RGoL8Sn?1rbjE#ZAaEMnan&-z>Kb2!ZMZ%SShOYIa<KOhwpg*%}C0E6ccUJ
zarx{ea?>&I?}>UvTn@|+Z|F8>G+hrU!m!1UdyJHzD^k^foDm3&Qq1E6rq)y^H2ZJ+
z<P2@oprpWsL^c&VCWrx%Cr}lpYf;)@tR=^U>3T$r%*Q=BLsPe`R%@J(Ec1f46(m7M
zus(Z2n2rQLBVs^=xs>pL#Vw3cuv%SFcPpf{2ydCU9+`^r())N?IPSRr=1<wIHq_5P
z!}J$4-96jyeoh!4sMiW>D~xJL@-ICZ8Kufz%vwR1cO`A-7LWodG<DMubH<o8_4Oy%
zW=(VxZXR*-$%T?mZEstOo5?FgbAfHaOo4gYBPvajEsM+;RbkFwL0s_1@x(9s*c2AD
ztDe>QIc}a1N|X^@h)l!Gx36#b{_T5)abccD#&O`at*EPtY--~8oL4`&M5V}jT~q4|
zlmJ^*9LI^lMGh{o?j=>zA|2fBMsAK<_S3}G#X026-~>u*l*}g^XlX~gthl*(<ofaq
z+q7(t3*+vQ<Y!FNB67ymHEU}yhXJ#AawD;^cxP13r1XK??T&}zM7tZQ8p&atxF2VJ
z@3SYoe15@Z)6=cbX!`Z(?L$*H9bw)Vg|iT(6cDPSu9^amT#iW7Q<+ebl$_F0ITE5H
zhd|w}QF`~$X#K@6#M!aG-*I;J1aoOn4+ASzxOBhs<puw%zxO%+*Z=wtaj}?BC5pv+
ztX70jXxYbMU>1^>&!4gCTh@KeldBajCWd(-3BhK)Vw!zXHi%4Z6sxW!#=yIG?~y{H
ztReWqx6-yLFvgsjrimd5LN>H@4Q@thh3(sdh|dwE#?);&%xHZoS4)Ix(A{OhM5U2D
zINULU6sYQeND*x{C`G3Y>*ozGpT6Sr7cKw%-@oOb|KgteA)zyb80eag&Xz}X%$b<;
zuYs(=dFFZH)r%)QdwR|?j+l3E|AsVPz@<!U6yyCnvaTQ{#$_VLh;3SoHC#M-hOB$4
z%92AMBFVxKDMg#x`#buxO-Wva#2gKVMD`0^-_lr3G>W?IXzB(jBFkKOMj~Zmvbc~i
zB4Vv3Ao%9ze}<oq>>hWlFD|%z`iy#>3u4>U1#zKDR8Yi7Yc1RTc#`~NawSk?y0*pE
z4f8yq>mH*8&t8<JmDQ4EIC4B3s8;8+rYzr!?>!W_;<N{Mr0!Orif`SgR7Pr5&nPOo
z#t>6v$r@!Fq#LQab@^v&D4V^H_y}zPiLy0y)3Rud^b2)kaY(eOsOq-7T?9`Mf_WS<
zO@kDQ5Mde}(pq$FO3Ei?+Vuv}tPAvSNQCGR#^STYO$#1_>CbUX;PKse7-PZDq~(a#
zl4^BL-In<F{oOtH<3QV=^V#_;R!v5xgp!#hIEG<D=oV$`V$7p>R|J9(V@bJqPeLG0
zWl61y^Fo-Gw0VJCU2u5w9m9uj$<a|;#TX-2Yx+&O&a^hDbxpN8C;Az?Uehayi?ADy
z2rW@Yk-aCm87UIh)E~{lA{X4pX1zjRp5vS&q~eWu|NfrG$1Safi>r>?JC90+x+(%R
zso=7&>HC`cdc*Clz|R4x4Z<XJBdC)up6kBn>)Qk0Jx=Uq&)dhDF=T${2cPon>Wul9
zceoTWN)xpv5KvmPVEF!Sq^b?myfDlLH;e^K6H8oPSxX#`NGVufo+IW&9Yxt0dB^!#
z(J0(K9{Bm|2i9j1w=7gf($*y<Rar&Tc9^OmQ%?5Uv?X4?>~U@;_=wSly4{eKE%%_T
zFttT#Q%ry{kbFX*NZAo%LS*?@!u$S%uOh>$@A>q_CC%oHdh?9&@k1##tD<CxUh?CY
z7hJA8?v6)_2rMA2`E^G{YQ@9jj<zn8+i6~~)=(SES=SOnq&wU2^zwobBeU~pCAd0Y
zp~$>`^MS50oNZQAb&YciS`{^}TPEVNAVk60Kqdx{t_<2%C~XQtDMe&;`a2?*gOD;9
zt4q01XR>q%p_!)x$vXssn2*0cKp{9y5(L_4K5s40|G_o?__Ln>@tZY&@@~uR-s4>+
z8^!D#B4(yxMk`qyF{_UA^CGC*?+=`vos~ECzX=OV07(cQ_Xp<v9@B4dqBJS5o<8HS
z-y@a<9}|oR+)z$H*)Q0(XFBXrw&N%N@lO#c;&wOOz5W%ZUD5U}A(XUD(uQnWtf|pb
zmWFqT#0)AsDj|pll<G--q3`PA`SmTIe)%Jst}UqQknz?sA3hW`dEKF!3p&#yv}V5j
z3vNfxYSnUSJk7;ZM2=<LHajdH5d>{jlXJo?jz$~Wt1FT$cnzg1q8rLvgKpVA?10E>
zv&LwHY&Te;5F(d_wW{#PN9K89i5@u~5INGEKP{)4<Kd&0wn7xI^E8h|!>h^z!+TFk
z0;S4H)Yys;1Isj)wrNT<byxEFqHx7zN@y*KArz69P(<C58-s`ub9$8xOJJU6Ow%##
z_uRh!j%L-Onug>fT9niJX4BAiEBrK|tBG+=Z1;PbU;Yu{>M1In(i`nL-gbD^klaG{
zODPYU8dWuA@1g`GQ9LS2m5VdQa^0rHGCYz~V0w7N^x<1n(_qXcx^Br@fh|e5%1W}I
zSaxN^P*obMdX{A&n~1Ru<bZMm1c$N)tv_y@SiTQIwZ6omagK8B=EQLtxqaB8WWhub
z0$NBaQ_-~@b=T8b$;&p;jPKch|257zvND*pMX3d5PsFhw9pBzg{K@M_9!5`2FfRdJ
zX{x&7$6q|>{dVHdzq>~@E#!<b7NN_C#bwFcyFKUU7bLApDs8_f%p*A`l$OlLfe-KQ
z5KYbXtCvVy(^LlC8n*ibPjW}}kq=|!{y4KS@S|tvJil0Bj6$YFTt-r?2-6YVT#)2J
zk%Xb?Y9tCy@7dZ^B@iGqDFu?7n8yP#Eaj9A=z3Ksu;h=6RFg?s5lqGH;JNexVH)c7
zhF^XEfw;Qicz;WXp}fiN4<te<6iP{5Yjjny+aKBO4)mX_2q_{(#@dR_YE@FB5?;J|
z$@9xIlq$F-C?S(&n%MLmt5uKl6M3A<k2#6`{T-@mI8GkhHY}}WiGo@wkQG{6vUj9d
zl4@D#!aPf<gjsa)lf)EAp#)i~)dsR)cz7Tzo)ALO7|XJDaJdv*jn#BrhsyA>&3y9v
z7yOrh`<kEMJO0I=-E#M^Wg0w<HKhSBplK>rtCps*_<3SDOtf|T*M<{+BTSKLWSKe+
zOX6%(2I$?YD|7W`MZ4bc`d7aoi^#K=FG(>YVnj?M%QT@Xg^UYfI3knh^74|LBg;G^
z*UykK5W<4<5feS5PZ%pp*)2sD?{P7+KD$6<LCh5?WzNnoPa^}56#~_?Xxq>=H9D?H
zIpcMYSwF*l_zn=9_cc#1S6~{#G*bH`Rb_DrRNYc3jhiMG6wUSL`1>1nx34){wFtkI
zH-NC5ur89Y99TD!qtgtB124Y(38ufoNKiuHmXZDKTcTg^(+C-CWuR(^!L#_poCF_7
zQxwe8L<p`VbCoEa10ZK#aOS3Ba*jAp1n)3ajcmFCsq=|AZ}DzIi;57+19BM$vb3!G
zE0Qb!UWSFL=@7Ey!h<`I{J`ud!jOpj1B-LCtDe3#Oos=gh(tf*#)0Wr9+J0jUz3-C
zRkuQDO<mVziRgT3YsXxc<hk(Gq|)RVuvSBqAWFy~`0}6%hmqm#J<V#vle3yweM-n>
zDVbtGP)cFr-Fp@hX_}UHeMY1dDUDWOBvNMPX-DvZ>^&|;5R!48%cA!qLk-pzS4I{B
zX>>^;KJFnpS}pKNV49Y??djJYmwm<d?uHk=;K%(VcmMv6c)Z!Mes;leKXN!OP<4n(
z5$3=m1=~5WJ31Z?a~Wj-D8b`lpl)hjUak50*KcqP7pH~*055w<L_t)uq-|S#3TQ2O
za@~_$LRS`BS&~xBalzz(8z+LFX|zJcfR3K~*E?#hSU-Ep)ylHWj&)P>_VK{vBe%nX
z(V8D!p0TP7VV>#R739SIw_lU?H+0Wmk=M^iS|D|esuL-Blxa@1?Sywnw64f9;#?s%
zEK4c$Ntw{5A;d#@6a84oXv)0$ogZN~Pk4R1r+@!Fy{Y-d_qY7{&)?zf8HdAxuV3GB
z`}oLuwc;=yOL0#s+P=rf%>BbHm8lB&)jMii5mTb;JCe0rfBFeepFLq+8?-7oqGg`h
zbQLKCv@UoM8w~MuNIM=6MCTD9;}bMy!~K5Xc-C>=RV455(Sdix&e@n>$L^&pKg}U}
z6qMp&U6PSW3S`%z=b0Ff<Q$7*Cc`op)Q@wnOn5qhZVYKIZjY`v{Kqe@`N@x8@n8Sj
zZ}>O={#PWSXzH5HszwVzOc@_NRol`xEvBjd+K2gXWMRq8GeVB&enU!;c-XN|9j#S_
z!<KoO!FdkH-3fH3Ng)-(U54S?*L?f)pAkaf)yo(37Z;4jf#VYCd}3`f5Bm|@RD@V&
zc;j(j;%3Xj=4?&VbWGEX(WMnBWGMy(?{VWmU7cf^4nG}nX`xy5gvGJ=f$;D>Zg)d{
z`2tmUgzTy6o}A$R%{Oe$uh88Ft24xmu@y^<Xt&_TBhfpSS<$pj!F{BN6lKX3<`Jo&
z*O74;aKpBkzj8(;k6VgyEJhFBu{ygbBBW)c(vbQUkK2zu^=6FH)ZH4JB1lsxo~s_`
zCW80)xr_{qGWZynW=FLGqq{Oo6_BDMI6<Hw^_|qX-5u$sBm0H%;R6KE#pM-BKpyu*
zH`1+IRCmqe-2?M5Lk2f5P*>y_nT8|tT>2Lx6!FcfThTTx(NB;xN`mqdS@o!HLrQ{t
zxWSKGKv6dqnR3B%gg_RV;LF1}lc|hky=v(iN!vE4iwnl%k#Bza3;aB@zBq%_Ayq~;
zJ*sJlabcN9y!XUdn(p(oFr5778HKvIrs>zD@v-p0KH?*UC}|bHBhEW~hS4Y9|LQey
zUf8r1`mE>4(<glK*$b{OHUu{?-=5RYAK3oEe@7Y)Jb872GKTh(bFNc{x`PPTJjXOG
zb5MLZjQqPlf5ZR%pZ}8AA0E&WhU37mzJEtmNxpb?!OvemAaVc+ewN(bKl1W&!>Y9i
zrSM7cFeRRAU6O}YN7ran3bd8v#kHflJSPaj-OUYIE3C{UA$gbs(-P>7DF3QUd3S&J
zz}9&h5qWxDrV2BO95XTcqF6|goJOSWisF|{2n#7?w6>&NjCoQO1w@E~ge{1jH1RRQ
zwVX3oS8Fb>dXA3=zW>#0vatN@m$!_IzzM^ip0B?DimEDa6txwcpI>r&cgOi=&F$?2
z<Kz*lqN+4nYnI7VSxc0PSI?hwett<^DUzRY^Mp~bd3uS_hUqvEPqg#hZVw2?abOr`
zKychWvc7)CFitGv#PL~!zUUaIBO)xMm@u|32$mEv+7<|9&ZS<UHB#w<V?l$Tklv$=
zL3cI6OWe2+7hkUHoN1c|tu=lsZ~Dz<MNSf73L@oV)$_mq&;KJ<{f0mI<F5!w;Ik;&
z>bhh7<btMe5h-J={%c3=zu^HQ$aPQFft&=@<ujCh<gj~$9GOGLJ?_|aHCN3wIs`)U
zWUb3`H_U}IqiU=P4D(FhJ`h}B)pbZBc6EgqJmYbpU3DNNrWZ7I&E4zo$)fy)KI}^r
zceC!X){?R!i;|ZXLV$P0h=?F1MZ}=e!hE==?px;VJCfGe%~OWOk)uO&XDk`QJQSm$
zHE5$@8oB@WFAzD?Z#IOOS(eDTDkoV#4u}}ZZsumY=j{1QTpalBkN+)6YF6tNGDB5c
zl-5L-kVd1k$Do+QjEssp8O}`lxPw+MKUyLu{4|mBgsEx<7damH91cgOVWMqn)@Lhx
z41`qB1!@|pt!7o%n102y6eqdQf|MW!&-Czy=oXf#fDiZkdupv&h6B6f5j7XwgPRuW
zwjpPjk0Y$tG;P7QNRxQ>=_}ToBjavM(^#}F{G%n5jtpt0QUblXL<vnu3!aFnd#o|U
z><KC0lOuRfT^qWlK}boZ4eRJgrpBwTP}04ps|>2Isq3b6SuUPXKl_5bj7+!R5z&mn
zbJ&#x&NwV+Q`2;PNo+MWDhKA>mgN0Mj!hsf4@|f1oz+(yP1Nq+BxrDg1{mDkoeAzi
zf&>rl1A`3&86dbbXmAVe?gY2s4#C|5gFD}Rcjw~#2j}!vud3C3(W`28?Y*D<dzy;;
zhg{v1v{-V&3doBKEWDn6|498Uo?F4T56@}?J~urfjIgW^3b*PezO6#)(_?ntBH#tR
zFhnLgv1s4E6Tvy2=z0o*YAX7j-|s(VzdF5AoKXTiJw12Is*p^49$@Q_7xcY5o{%ot
zkG(vq4o+wH!<YC-iriceeSha|_-7#O&w7y!_+JRszv;`>mvE=MW>I9poxLZXR>jX6
zKS0b2<^xl#v3wC4h``wPj-_e+(K^oT8DnGh2CUk~ljL7EI2J!|86h%gx<g~bxmwX`
z=<NKh{}hDr%*_nNq!7doOQ6bZ8KKn6-W$ahex8O9=s5(MV)E2)!gbQ5;CR1z_Ck`)
zQ|3OgvSKhk;q7;xFcCitzep%@5FnFi_Z}@hyua67cWD#n;mEokabBBK8U3?+FYNlK
zo>9J8C1&tBnS|Wkanuxba^PFgMl}$oxg&(-Kp7cy2o}sKORgtFY&_qlf&J+s&27A<
zD0(LbBx>YDJE|0|Of6Vy1Z|Md@((NY;RMa_CS?sVHdAmH;JQ&&c5XK4Dut~!%0c9)
z4x@lz=j*giS?4bFR_z7Pdf)n~oAL=+dF0AS%@zFF)pu-nRQC=nXQe9?3{4<NhQwq)
zZ+7k9zTTXH5uHS!byxRJ$+q~_ZurG0W&B=zlhfLa2Y;bXfhCQXVw*nqsdZ~>PlmW;
zguhPWObPy<-gqw;a)yZ;zd;^=CwvztPIYjzI;WAc`8zz~<q75mQ8Uq{CqA>@+dznH
z8XdnThm-IQ=m$Z9K2-LpzGD`0n8C06W=!5Uba!KrY@2f|npWCt)4#(@O)rr6*LAf*
zfqAPCy$g}xErlHv_Or`}*qZ$XNv7ys*>&WV+QqDdoLJ+&Z)lMTMprB-V-Hiwi0L~D
zN(G?ek^&uowfBYx%G`_$#Bpdr2q;cOD_>%!q)tRHAEHN>8a^uZ8IKZK!YgeqZHHo%
z4_}JPGn4>9H?t$l>25(5jMZ{fYoyq7(e~hKV~?)jD&tqx?*3^#Gq-D2?)@<zuN&_;
zjgAA?eUuBk8T@IMy+OVC08IU*EqE6ZhTXj>qhmp>z1&gHYu04_%k@nLtca0YJ(BM*
z5o8AAkG#CIEa0Jk&aY>u3&PxNru9OVs_ASm7*%%{?-p&1(`ft!VzI#7;i8=hJW6*N
zd0>*ZFzbB8UfxfAcw0c-LaSD(^Y!1bK20JUh7Z9b&bqV56kg}6>%T6qXwX9Yy68W@
zmmx-CjE3(!W6xO>gz~F(YpgGI*Zk0%$c5Q--tlw{zvh~!{nrEeCzF>lXg@e@@eBu1
zvc221=DdHVl%QpZwe^nDEGHXv7L&S2UCwogocP)5R3aid(`aMx9eqTgz{{fDYtR`I
zTOTs255DPj70Il5eewS1L7RlH61jZ?QnsX<$Tq1H2KHVNBH}-8^kt@!vMhSx!1`b9
zEYythN<{IpOOy@vN=2OUPhpz;M(afCNzP&&%hfBYUC-V^tK>AI?JJj8!)ZUl6=vX9
zhbZ;*%>k6J6>4`K2r|?7xQ>I<8lo!?p?{hwi}7nPb-nRM60C0d9x_5oi@$7)XH-*&
z4ML$DVg}cMye@CD$g$iP>iYUT**MZIJrqh(5o!rjsvrgDtW;v`!P{Ml&k^Zu>@Sfi
zsfCQH-UaFw-ONc$Xau|eF$bulF)N~5C^444yR)@gK6T#K1bk{q?D`VL=Po-=zsgiE
zqf6Kd;y&iz)7oGVA_!lee{y;mhV~ExA7H?jglq((6Q;$e?eY`p&$hxt`m|@mRJ|tr
zq0A(3Rt)MAK0E0!?<{ojg2)W2;hFQ<(}!alLvLfe;7>d9RbL_OMk_E@)ynZWFhesk
zkX&uFGi2{RrTxh}+x3$8_1)#a5IJQQyh&9fF=9k}x_BCzQLh*xUBTEFo*_Dcp;}jM
zd^_v@MMe^oGrdtqQz(OmQC5`ACAP+O2Ar%MZ*Kk`kdCN=C08~m7=5voO+yX5WQq?n
zHS02HzT0x*wwSo-b<``@#<F84DfGvMovyNL&zuj(LbC@Ky823pmW-<Z`h7WxG?jcK
zf#k!IdURkdg^AfoRRP>%2<OH(FI+Y>J3?t%&(q#3Z_NSeV~3g_HWbi&9ATl=l*;gY
zRZOPOFd<?wSc@rd_$5d~s31L;B=N)TOYV+NNE|-P550=`dChF2YCWgaCY%jMGbIvg
zjY8@|nK?JDeYG%Q-gWQx3ma-YVLWMz$A8|*p|?SnQeg_^D|u_;)Q5P1Cm%+6c*jwb
z>2NhaVB^#~hE}ut0pNPMNjV1B2mi=tD#P_^Pe9IsT~c<Df5PD};;9kWOQa+7Dl>r)
z?D+Q`dfMXf#3xwmfeCblF}s>sO27TxSLd}{$1%H^6jY^gy0ZV&@YBH1!T6ZFr%*$v
zPR;nopl`1?6s{vkzQJhmYQ}z2m3Dp}$CGNt)sbnt&CZa^0MpbdoewUG3Y}eE8+E!9
z!YdAUqQFzH`->%X6JSb1gTI#;&zXF)P_KzreGjggOv%hjTU;>yR0S#EmaS=J>zew~
zI|ZS(@49<aMSN!seTaEXr??q$_L?$JPu@Erk3#%b3PRsduW1~IIezr<{anvpvbvFi
zKr=3T&|j4EnUcm&z4=*9KH{tCUSQeX77lh}1Wq`$@gfOAV#z}THgMdGr04j<*d!NJ
zKA&IpBhD=+l$49*_64&fOFvF*A6!FB2O_YZ?Nmqb_O0tDF-K{I$Lr@J*xsI&nv<f|
ztDxawY%5#+<>QHx*Li+#&!41Kz6?96(<S~3#O4$E@XI2N=pP6W$Mb`&nA(=F8}%<A
zbzPzG3Y81fHw~lDFU!jp2GrCXSzM}pxXN68*~{1ZJ%<vbq*tZ+`tJ1MC4@4uxc7Y6
zu1JDZnZ2|2eiv<OUhs<y;C-Ew*Q4nmmq=^xCxeB)8It=dpU+S3CQ)K=Kb;~57uB5v
z;)%PZ?3awtHq<@K%`F#gt|jUkAbIg4V5WEHtrGeDa3pI8-Q8}epq4grcY?Mw6x@CF
z;2kw|h2nXW;RJ!m2wpz5?t5l5xB1FLx!jCmCqtBM8T2%CTj(^FW84d`kX_*HJ8{K=
zYQJM~A-kyhiCscJ#g6M-2W&U@PRS5{ZV<&$JmtQFKqdvH4?TG3SI_UkH!#!BClJA}
zcs488M_J1^f8g=LWNr{_(BR9eAQENs>OVcuP5alfJaC0TkR`_fAyVEEi4jBJPdzeR
z%22oL5Zb{|b}e?DB=Cyo-@K&UO003xLa^LQL|xf_-wu2x3<e`7r@q_z0LCC4K03kt
zg*g@aR_RN`zbAaW50{|}VdPw`H1q-K@8%w&uI~w@SY!~A!k@7p!$rjihP6Y=3H->Z
zQ|b~4ehS5U;)Qe9M$t+u@u{k|>2G#eli(mzIAVGY2W@;(ifSkzNi{MPBVRi8SoD*g
zCmgfuUy9_Ml-ex(3BC>cVlU$zoJeR&iUC8k4H(Ho8*}a2F7|;*f2E?R*d%0*8aWGx
zVx>&ccbJihdG0Lf3jGh*Mm!j;TY59g-5Ydv9Mr2ggxg+_n0cr5El~U(L@ie1G#Li{
zlYhd`{iS<wvRp6s#BW#PW%)d3E=*oHUnS-h5G(b*uU{`xht;pT^V4XUfn1_7Zwy9L
z6Qtc2G72MD8NS2;KEg;)Ypdz>*06&bhel}jNo_Hptik@=$;=kxz{jAjx_|dix6s(v
z8<Uqngj;PL!e4u<6qnLrf#z97=x<#I@SA1c={p@9Ef7P7xO8fA2J$5wK#e8AZex^m
z%xV>wjVtwo3+>Q(^f%9aY%3?#%q*7NckI|(Ok?L?*@T%=s5Gro!KpF03`6m$yv(R>
zjeWmV1R6&JY(=J<_^0hOS#ZA5^E7H|sn!&Y53)0(gjf~|5Xzx*aOm(wBaT4;Kx2Ky
z=EKnknJX~>L!oVM1*=(5FINpL>mKeHUxP(k!r@Xu@Gfc0#xkr^1y6?h6O6}Ft$g&I
zydcqxR_YQr$CvKcyNeD>YTRhqCc+V$dTxLDLgXhcJ2YV`v$sH)?{8MsGD7ItEpYdq
zcVfct1?!d)fSY2irT^Lxf7EsJ2H$+Ec^$qpOj@QIdH$}d3mRRHnPSb}C}B-PF$$*A
z8%6Y6AF93LY15Og`}O0~03VMGN)a@jy(k9*K-ViGDpw=HC5gY=i+!Po-c-GiErZ5F
zS5PIMyOsOqNR?2t#JuDz`3JYNrq3t7=Eq-$@J!<EHu$=K#?JDBBl0SOtQLp>+H}iv
z5qx_?7R4>*{Zdch*9rX*7Sri;ZYL{8`yAECUsxKT@;=LUdAeX|BgXR=hM5>@0lu3K
zZm9S_zmR1fFEw@kc^N92BY-*2Xl>lA;%++4EW4xbiDh~u;YIA4&#a?aboYERy-@O@
zKSB5o1qZ!FOdo;sWtbLhu9T4>whqx1!5s8<i#NhU`?I9)y7@|sWrp)*AD`$&JJ%WI
z(z@TLM#4yvVu6LlG31A!U-5pDFy%li2_Dt=w)(TSbg63Q0=WbXKZr2iq!xvj)|TtK
zzriTX<6%gNlXUx7#WybNf^CC`H?>_9oWm02sL0G~-CbDW1z%!4niCdA8c1(C*$;PM
zFA)x%<3D(33y?$&#32U&z7pZkmFrkkpmlmG1Wgq&xIZ)#hT+ngj{NwX9yr5n$^lc`
zIiT`teTh0CSK5=xA1f(1s5f0&{UD_{YPpLm#YGN)gCrzL*tr2Mi)dL!z;|CR6V@N2
zxNwBd6!r3c_2`fBoDB2uOz;kZBl)mE8PxJ=%e5!OH;`M*Ugb(%ZBi2k^+p=yXmyk`
z3Ph<6lscvP%}&|9v9Pr{r?q*<>*i_ZwHMmM+1a`M+$%MW5-&E+K7<4s7V~>%S2GbW
zI>OX2a64aI^>$drUv>rT|0bXv^0#9pGI{wcCD3ds_du&e@6w1FDt>4#1pZxGQk-bz
zdc#c3A=IQ!s;Y!A&)KyAFb+I>x2w4wE7#8>?Z3a{{!?yeQ!b-4;g*3$ZJwB;=R~sq
zbN2vG^@mP)A^*vle7R1$&vnM&DfU*BPT+_g3&Kry3c{|dHZiI1?D=<QmT;miZJ)pW
zes=-wtXebw@%O_Pc?FxIO-L^uyK6jLQF65kojUPcohZyBSieGj+u!J|o~;E@GnftI
zwz9BAq{uF<r+}AerQ$b{h{Bb;%#G|(68;n?Rm2k}H~h&=^{$)u<+|&hcVIg}+g?rE
z|MmRrRnVb?XH674LE1oZPT=Xjsx!YSgqRos;xS#4FpCOV%5T^jF4J{b?_#i&g-%`m
zJ4L-Zko=pglTu!l^`S6e5VG!~!+NAt=XFCr^D@-0X9W8+Xg+)$`ddgfwY4|%PE7rv
z4$^d6kT|Jh)*5_SNi8%1f_xMD7a&7VFHA3W6_i+$DBv~IU}$WFC^wOyTGSO91F1pl
z`9W-&giVuk$x6!YhrfxPGaTnF97#ec9-p!I@wI93@_R`FA_H-G9P7-otS+{=^=sB+
z3+5;7smQ|!sra`4-N>{&oo18x9OmJ|FT=PARolAze}kDW&AyNxRo0)tbJY!yXWgag
z$<TP2zqhMs=}fz0`JoTZn~{LqL`f)xfM@7st{0l`)j-Re=0U&m8VL*SG_(%^Lsqih
z&!mnE<^%FiHzD9F%#vcAvN2>{!Uly5ZS<cMkG(@+!#MvM{(=NbNjo1GCT8_gsZyR%
z{vx`FX(EM#ov_}RRB6Q=_7D1GFkr2Us97>oMe))58Ggq9j7^`A&PvkLf@to-H)y4j
zeEo5^wsok4c`*%+ZkXMnH})loiD_PX&LwOi+L(PC+}g2~b_3ghCQmJb*)bGT0&XFP
zIypsEN3y&)=Oe~nTAtRLa;1f)#thgqGV2H~t7$E(3&T4_AsBl3L)zt~vImYROWFO&
zt}qUc_cXb1=VR7FY{b8ls2uq~P7T#uo>HUS6_hi4DaAC#eXTNpB>{Lb884yY&LBIx
zX4&H9Jhm&38+Lr`TAPSwK+_8ns>KZ1oH9+1xK7wF=2B1aoX-ZI@gr<>n23Y;3fR!N
z5oTIV%%Q!lB{s)3Tx)*EDQ_A<C4&2tOaj)r4nUt+M@&<R=}Lj((q2xoQjdOZgPLVA
zbHikq9lrWPh>}c5Z;U0t#_dxR>tDl4PGa%--zO9jLRp3;=R3o^^JN%AKRZiovkbSs
zkfOgu2(KRIYg>xLL1UHthQ{Lb9Php(fUV7zz$gUg&-$u$>jgaU7~rv6-N206w^6bZ
zQ$Xhj?xX+K%0Jf1h~Dd>G=y%h@Dtw+9*>}|m7cxdsIOG!RJDtv=VuQ3=C>$?uQeed
z8Mejxp2J*6T#D4Vv-n=^1+I2a`Er_#a0$CojtMO(Cw#HaUO)d24}l2%i#?lE*mdH0
z|A%HTV0AkN5g3N6=;{`eEU96bAo3-9M@JdoSHUE5rKLUk!6mvGzW0wAP9iZY-f3c{
zdDfRlX?sV)lz!L58JP{Kg`Dq~yhK`9-(wptz`#P^(z3|K5g^|fjFEX<a#I#Thn3Vo
z3|swF)I6Sgum0;uei1L1D-uUrj>_}P|3+0>eFyK(vkGg&DuH87jP0CXgJ)UBvn6OB
zhVYR=i_|V#hm6_VL?(L?jD(<%j3SxUaDO$tB9?;rV0yk&xN3BXXjp&)J4hn#a!~9O
zPb2H{A3jgb&z5~Q@3V(EF$VPKt@q34?X|{(aL+IO<KAZTm^;t9yW{gRm$UzSu*~{B
z&%`W*Eq|18_u2q4nSjwP_zrgc)azU}=Jg=9N0%y42vJeM4DLe<I+xp0oBKg?#fDBT
z8Hh=M_UX5a?y;5C;Do#5h^5spK<(h^I7i5}kA#zJk&!m!x~i4uGQ&YQpX^;qEaeQS
zQXw*<|0rxb{WbNv;+C7CsKpZ@gpHEI<Cypcm9ryn0qF%(v_&SHa?_TnDM~9@5ip>x
z7;Zf&WYH(;$-#)>4x?V>4Fq~04bT%^-RwvN3_^TQ^5awDw2&hb%SJ<1cM0>|57-!)
z?Gqbv*p1rE7qM&z3Yio%r8UOTlm?}|l={(8#<wb@gM-=?)>|1xsB;o~JSd8l6F;MM
zZ=ysB)Cfvv5z~F|LymPPnpH?HVj5Vap4g#d!977SV&70#8%yQG7GTU#3HhvKh!$k3
zW|R?UP0u8&JyE@yd|Kxnb%>dnRK4Qj;u~}9U7BtWLJFqzuFU38RPBB_$*5=uviQ96
zl}E4^dY&j_pLl!-$?6Q9D~+;ujS&9sk@=k9L{-6*kTxX(>T(U_u=0T)>S8``^6Y{J
z55(60`u@2Xgbx2`yA|TFW1%1xG<HWH71`(0&M<ua^3b_GobPQY?yJxs#GsfU#8GHY
zN?Wn!q57GM6=Pbg$he#|iP{K|A3YG{`Bu*YxkwNg#tRD}N=VvS&Iy#L7n*N29ds~?
z=l5Pd6_0R163<uQ+0*BSxy+q(d%<Y-m#+<kj=S}4((4v$b6J>c`59EJ<$2xS8r4~!
zc=$YRwjH<(->^`daDXofS3+YwYtIhdJ{yl8gtA*EZ#4u?<&<S`t1f03@m#&jbS<C4
zl>SlkrpUY>Z9S$SeBdX6<A3>JNBMJFz2w~K_uAze+QBZ#H63P6IJ5Fa#YbVdDD@);
zjk*O4jCw}BxXniFcTMt)iF2=8(6z^^*J}N*LB9)ahhdnGs2r4;xo)5AhZ^$ASqC(x
zOfu@)-6Ln%tlL_4u5KJvj9uWxvP<!7oL|@cvirYB*RlJq$0yb>=)_x|09Jdi*)O-h
zgml-oH0Lnk*Yd@RFAa(xeBD}tz!~*P<fmty-F*HWurbqaPrCZJD4h|W_LJtH+LSYv
z5zStcFEqIcSn5nvr|e&m)SAhaQ1~uPvj%?@M4X@89ncE!$6D3NGqarG=je~!gURjw
zA1P!D551$O(euf;gf8+1a41Z}>e@0w`(0G6Ll-l-S#1Oi@G&h144}5r)Z@(;pr~if
z#DOTl@m65&cfdOaV|G%-T1-MQeb#mepV8<dcBsoF;SUVAv6#Z8{B)E7|E!8bSk!=b
z)>_x#Qm@lcU80XLV>)Mb8^vk9Hyy+7(XctkI3VZ4Z&83#6O&-v$6%Z|3;x3sqSE3i
zRq0YqF2dDj)CAaO$wkJJUGR<gs^8!wg0flq)wGN@QG?6#<OXK7vZ*B}E-JNPFS3H?
zK}<qvePQD?#Zc+0u83{i&z)1#ef;&++I^eBR&(a`cCk}Fcc|7gdeJ#G;98)8puwua
zP*aCjK`G;ON~C*fk87ULUP_XRQz@b}&z@8A=T~n<1X}_}FYuN{NB^2AVW&A4*~$Dq
zwzj85LV#a+`YW+Rqf9V-vdX&u;^u84bu+-y%oeZi51LT5_zyBO$r(JWx{sLy5({Hi
z$}?Dh#i1%$bc>foY)AOOGOuaT_a<Cr3sHpyx`d~GRC;q@{GJ4?h4sh(k|yqtaJq`v
zFjQ)+x3L~w`k_x_MiOn8VL-3_L}>h2i<jf@I`n}MdacG1p|aifI4cPJ=I-)gMUZYc
zN+!5jVvr=!Y3X#WaxOsR>R19ODq0UMW}07EGI2ZySsaLDh%gS1d=-#jJw%=8T;;CX
zD^OH|rI2KD#bkLvf+6TDk?e&m3YTogTl(eb;ndA23k~4YFDvm#l!JjIfq6{ESH^0A
zJ^4(>!e_7fE6pC##&mR*QO^-kdvY)sF}hw0T?LCkHgx>Had&y_xR@;;VI3*q9aWdC
zr)@e7L%pTsJ`qgCE%V!9*xhnbo{T6}&|BbrSQ`v3P{e2DUXU+#!CKyaR9-H_B2Evu
zkheEfpB#`1w{7?a^^+6eqw9aG{YMYLexIKGO!o3z_44~JL>~;S#zvBO*$_B01QyO`
zofhUAQflmwz8t}hDW6Go9|P3RfZc9z9u5Ao8r$Hhv_GGGefl15LKBMw4_3rw)8Z0&
zPu4ayP6UW+irit<En<VH(Wke6F4F2Hx<LXb5Cbk!`7&7UlBrg>B7SliGJ0CKN4=`F
zCPCEN4CGA&Vdjs>M~GgQ-NR<8#{yS9SxC|C_Ht~#f!62;&6Tap?K_+3kL+!Q)@(#~
zW!43X{oj(s{5&N|P@;|x9YZh0@7^R?L^U!FyGot!$SML^(liEnw)~a8l#d^VhJ5!6
zbWfhbPUZWx4q5{(Gl4q>xG-y!3N@Ja(fY9MMUY60GK`nqYt%*@LMTN3{tK+Jt@&X-
z^ghv@Jc<6fv)@a>EkcR)!u~9mh;gyA$oHJy@*$C|$>~dZOtFrAZQeIm_u(I+*;7ss
zM55_3d3#&?#P&@7<}q1plqvG7pmn`?c!+=hDWtyfmy^e0rTpPy!ANmDpN9Hws+eWh
z%JenXJdpk_z`>2~6BFyj68;%s@5b+)JD=$5ca7`ucNFN<$^D<c%q#mOo8Ic4a8#+U
zlr;y6RiTaKM8Z6}1_cd>j?JGYDO)Q$RONew0J>VP8ITQf24<$R<4>QMCs{1%;&f32
z-LxX@nSPNzpk&C~%YV{fbK;Jnm2R@c#M17dVofRDK=(kjIF=KG{nvMR2Xd*A6pm>P
z+ayKhVGEwbku1x(zp^Md<>i~mm@o0=lr@v#3+E)tRy&r89qZ29^Clmp$jM`TM|iWu
zMYc7O9|jFzW!^q78-bz_z_~Rryu$eQF*%`4SE%Z4W6!;m$1)w|LOlmn&5DBZC)q*y
zR#`M+<e$V^?bBY=E%XTiCp~fnu~+tAY${5jNzB`fVmxuh*{G)ZydLd4brDlD&6n9@
z_WDnKT=*vp^~p9ACMq+38&YXoJPNBXJ#0(1>kO-tI`yA_b37=dUd|h_>$qHK%11gM
zUBHs{n^bPkU3I}0@zpAgs=RJ?0DAg7K#_>(^#_1qs19lI+az17<MgO!Jio=6*vTfa
z4@;FxOl79}1Tj;^@6O&ykZtH2bIpGe-v<&NWAC=sXk&=@p)U!so=>|A^=-;Z_!nZP
zzV!vVX^~~(wgz$Rv7`pdS^MM_3&pi`1<3gkmVqSO5m#30(@pI64vPC`9Xvq0R;!(-
zO$yL+zX`rWIZ+&O&EO;OfH+JlATZyavDsJJjBv1M;hXDc_-1wj<?7bIvsVn0Y@Dk-
z^2oiampKz&6OQI)Xz0Dqc<Aeu)0Sn)USK_oZ5jnj)T}I4cubi{g|%>1i|0e?*c^j#
zs9ZvEJkP?;vFsDbO(K~_FhWn8@<EX)utDgIq%CI8W%OhlI=m$h9a*CElQkkLF8}6z
zPe36TAr|c@Js8h*0^S$*`q>YAJ4Ptzw4S5iGdIIKt-_lcRH2P)hrhtXsx;kRuw*oS
z>R~Y_R=LQS6BTaHU)~$fU1kk7H4wKx5hWTGZl+j6wOUx)T^2q2(0PY>-PUBAmXGcF
zTk=;3og&i(n+Gdz%|Qlsz$E3k@k|tr>Jo$k;dP|Gt)M?K<@J6$Y}JVblEz&&ROQ>Z
zl}Y&}YmJ8U{S#<3!9)7Hi;NB-T~X6^eK#o7m<^n726oG2<K-Jm7UL(-`^;~$nOS82
z(Qtj>n&eUWCSTUwTl0+B%JoQP<|d%h(KbCqWKN4wp5U51$zoRCUVxaaQElQI`A`&f
z6~-TjXbzUfQ`|v97`f(;Ufd|zBBFc>ue3KPQ6Vh3yyh8d!0%~egwm$W`>YP4lP1Qn
zJ+M#z^{J^^S0>+>qMO*~h^4n@DARd;kdz%cJU$^D8zx=W>H6Oy%6hs;*22Q#$P^0^
zTqr%?gOyzQ0FIg@;?gS{j%s=qn4Pio(+NmXa`v$lw@3&Zb!DBZk_Fa>jsY9_WJ|4U
z_MSz`Wl9JMp?cwfh_cRHqOH<Cc~;QOb*P(e>~Mr$`QZ*BF|p{l0kLQT19JA7SI;T!
zm{0Od)5IbK<<dC#^B0|kh4#mUSag<X4Am9(prE?C(8bQ(?B`?X_@ukocjxSaO87W_
zNY=fJ<Ro+P*Q2#R)&zBgioP*ZuV=BHIB$xN(`JqrmW(CxoybCbr3D1EEXz+VFL<&n
zK4MkUT2qZ;4AA^@61^fAn8ao;OFygqYGi&DfERow>I3PY^axV=5E1oVAeK=vcUvA&
z3Lo3EJvobEtk9&Qop7pN%`i6#J!>paQTEWKs3U{dgjz+)$7|Eu+K~(fq@Nj?L-e?>
zu=F%8r(%J7H=<3<WolG5wRZhaJ1D#5d&Nuv@#W)R3Hb#Rg4!SeXOEWshvBHaH1nf0
zHcl>-S}pU7h$*^avjHE`+Vzvvs*i!C!)hS>a>k#TF8fdu7c;x$!9lQDOiY6Lnwx`<
zlmAIth?>M>)Wn$t9lAQX5v^~1aHpj&+43vgp!4qe%y|vIyBgz$;D2ol;Y&BII&Wv*
z-nm_}E{_Q}uJrfY__DDY6*B9^^I}t|&g9?_g1nDw#^WuSy*_ekm(YnOKF)idaj|fT
z;6&|Q(gp@LE^V#|+;H4anb+nG7_Bj4iY~7L9Qo!PV#G0JSP6XFdYPE&f)G&TV!6+1
zu?mr-V;q%3Yqv2HeJ;+w0L+o(YKEh^ICc1eJW-8<(X=A0hs#1u9hL9D6H^UObq{3s
z<QI{Nu~Z02bad|>1&)DU<#Ax92lq~0_wQn>8u@%<H6rM^ZGH_$WYC3Qdm=UtOM&9&
z$uv+`Z!92*zCdwcvH|H#u|9&``Z8_n;mgHbiiMX;Gm*m*@F8vUNkYjMw)A$PvDr$Q
z>HXj0)Cpn&FS&JoPhd`BxHe;ZzCGCqS58vTL%;@Gi4Ni*Wvo402kRtT1bJK;jlz$<
z5FOJqsB$U3w~c8^-aKL<^2X#yu3jMhL3#Kb#zI`6hrG7k76M+>{FRpXhkKbT28_uR
z#C;Qt-p8v=IqJ-)H*g^K<JLkp`aN{wfDI=*qXTaJ0EvnA?d^KR>Dri&Ha85zQ13Ve
z&r4QoSsLDzs<H{2*EXZ{2$PL^U^<Gm0ptT?wd36-7{`<X9fdP6@@L*UnaiSLzkBlN
zqE8>w&>vh|TQhYF<gQ_hxs)d7qL5cHHfuZGtQ{e-k3$YVBU3RRY!Ei5vXQpJcUXwi
z<sZkcVe5o;5>eAm#dLf4xV{-TGINZ)ji~B;U`r0t2w?k?S|27WinHxdBMz2Mn-dD_
zCv1=lHpwUaCU-KK>mI-jcMt8Akm!2a9t>$netpV*5!Fp+Z#&f{Zz4A*z2}>#0o3N@
z9DhG5l|mi_xq_SPn|mI=N=mazrzP*Wt2RZl+gL{un)`RInB(y8$vMHt`#UM9o!w5~
zEkC2IoNc9)N6df|%5$7JY7m$jVXrLSE3COKlKc-Iy2N)==F>qmYD<Qh2}!*5O4YtU
zrr^`1Iw8*>dA;$L*F&SfleKNF1qxVR$paFsEXMtZ=~+d$l<^<O95lc_wKR$_A!f7A
zC=AC6`Jxc(*z9SC#_|PzuLinwPSQ+ky7RexliHUt;=ew(Mj_){juXTtKF<S5R4hId
z!2>2wTf|~McL)8MJ(t&4#fyTQlg>xy(!13bem)u==9uP~|6MC}fBJ>p#noflk^KEx
zw^TE9FtEvJlJM;2Nw{b9yIg3OIHfsSmC?GNaG7xXMHP3u&od`rD`};p{qL1eKF;gq
zQ%Gp{+(OowK|1_NZGu>FwbMN^`{k+#*qqt*c<OY2`uhBAzjB|s@XHvqaM{%#Y<6@#
zYN6EOvx<AGi&n)(pvhEWzVx8mcz@3(q+aDq)Y$<(n)AJ^pdeYlJ>POY_W=gu8ZQ;}
zZUZDRoIE3lt8~TJv!ojqbeoR*y|3OGuOz%VYDMs9ZV;<%iSm5Z<ONpM+1We#?0ZMh
z@kHXI?Ag-RS+_C^ak?B??QFwZ_XC==oI=FbQ=B}b1fdi@>+dKURXZBJyDlBpB|tdY
z0lK$S20$mYzs=yx?B^3}Cm-<%--}_!r-y~t{c|nOF8eI#$sAdh9{dP0$-08yW8Ws$
z-uh_llRRf1do9MdRxe%!z}P%L+aFwlyqq~3b_zKd`+6N2R2g^t&F8$;Z8s6C34;Jk
z$cw6cbOA4UNfJJZTaM;s5qIhj=Oo)gor^cML-+R+(9g-1RZfSu$YHorXmIJZ%JlgE
zy<x|vl0u_qi2n~y3XMA)2{9R_h`;7{)b_uVflqJ$k72q`LAX*XG;e5vaLw@1ajAoF
y6BPW>sDn~v-pZg+_tE`d-t_+;`G4$--M_+dkngbHZyf%IxD;hS%T!8z3;ZA0#&J6U

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_e2e_001.png b/0.16/_images/sphx_glr_plot_transforms_e2e_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..bf1b534dafd5dac3e5be7c635172672f65f9fb54
GIT binary patch
literal 278328
zcmeEt1ydYByEPsxu(&3$xVt;c;snT|!QI`R;I;&J3j}vt+?`+{IEw{M&;Y^h<JNoc
z{SjZ?Q#CbJQ#0Mu)6@N&=k)0qbyfMdnB<rU2ncT#6~GV#1f<Uh2#8zgXs=iJXV-RL
zPr@Fux*i(NRvzA_Zk7liOg&s2oIM<D&8WRB-P~=Rop?C}Ie6KqZ9F_&+(kG!9sl<R
z4re!O&S@!`->-#WxF~#bM?k<e{jUSDT(Zm-0TBT~5iG6glY9IGuXEH--+%L@cNPd0
zKIk~^p5folCBP#~TB@w(v+ay+e4E~&wcOapDMp=`2vF!k+@{*1+Cl^x0gy1l0X)<w
z%UZ;x-_zMN_E<7lCu*y!t0$rUxzX<MhulDY`zhhk-`#@m^^VLZVb>B#cn#ljzpnRA
z68Y!B_Osog68C%d#+@;hEbsmYb#Dg>u24Px2agAFstBSY|A7&&2h#HFe{h~%NYMA?
zzp;d!hv4<!k91l6MydbyAMEhUFn<2ulF396kAMFMK2tw1OyvKAQTxdMtK$DY)a?KL
zQNMrxe>pDx#XXzWq$Nn)M^8*BgLgB0Km6B}&`Vg(GE*RV@N~1u7+x^5apG?vFF*Ri
zxubP0=!1%MJ1b9hepMIzMDld`U$e~1+TUMMR2Q(iz>|afoSson50A$e<e;Rdw?-id
zMn*<Re|ztZAL~}~62yF56qzFuf*vtmeg)rBowvR$Ztld5&SYL?gMh9hLPk81?qK<5
zO*29>K+oNg^Lfo?K#p(r-HznF1l7|nZ_pma<}T-pdvKER(^Z{&P*ODBHWvaCIt`*$
zFDVf~8f+su!0v^Ij!%sBJfBSRl9Tate0qi|tA3LOzhU|sp@&tXX;*Wt-^=x&b3%?Z
z0V^ZK0Nz=HA>`5AwFPmHn#0l=6j7CF5X{PsfEaWjN+cf_cFfjUcM$9x(Dz)r*wGUe
z9Lxjd3<#=8(>v>td(3L4LP6;yjtNCOegH%Rmp$2oSDuRk(SpQ}gj!x&?4388*3Q;l
zt@GY;w!cZ=Y<;Lb$%oI_!3ruk&InZxpy($wAce0X02B-m`hxKr;PMq}h?GdXe)`OU
z9`iyzJ&bMBzUKO_a=vonvQ!&<gmvR7LgF@_n$0{nbjc~P_l@dB=6%rp!|o#na8iM+
z+;igUJaSON9Y&HbrlBOhL$UdI`KZzRkl}k4=G?fFXG(HrKe_qhalGmivq%yYEb_aO
zXn0A*untCOEK{<4g9(d`pBnG#xylm2T0C&~V@H6w<uW{HyyUz*H5xyO*#uv1-km)O
zl4@8G)5q03Lx>cS!=)-PQ6Q8tJ7Z!qYm~L^?d_(LRKoAA-6|Y^CS~|X5mFD2G0;Z>
zNHyq&6cjKp@Gz3TezkKy&*as}s?hrZ!I1h$uYrLn8v%F|9v!K$;t-v~>#?j3&eW_B
z9t@z{&-##uVJQbgNK<G1Xp$q~aJp_yL7g_h7RrQ;h$LYx`=e-AFUviok{S`b45etv
zc{|u<nu50K^$qOCtj?>)BCMk{KCg$N7sK4R?m!l)01XXCKvE0`g5|>LKnCfFp<sl{
z)%(nYNzK&C>uNWx=2aXHpn9NMM;5~cH!aLfjIvGmy|Ai$J{9kh`+g=bzuK;EQ7Y)F
z+ff0zzhJ>iVTcJpGsuN+P8X>Vo8*6>&(R_}z=a<<M2c@R^Fg_EGK<;z3q9{a*4H#m
zEC|cOXmP69K`ik$S8|Iem<A*@0S%2W0F86m?pcZOP$f+21-@Gr=Eg?O$jDfel$59A
z*;8SUu$6FgU;9>~ZeOZdf+WpC6T*!$Jv!ti+PLl{Bnod_fkK&iMnb3)LwVF!hWbd6
zF@Vu$QTUJ8IOW;}l0@B}#48>7tn&cV<wQKCirt+g+h7@Cm3h53iT9bM$aCj?&n~+{
zGnye%mMXM6UT^T9e6{TISI%yZ4`TEd_6=%3e?{Fl5ZaH?W93E}Nfm;F1>i$gl*{q@
zAO(~H5196T?zr9$U#saX+AQj%scYy&+yitc(v_0zXJjUg4B<#7H5wnoKpSnb0yKp5
zjahU!snl`6?ymB{XOxB}&EsR}>~v>N4#jiX%izm6XLD~qE1>WC$8%ipo%f29;I7!^
zI3up_^}op1xg&Z9$Tj4jD)x=i=Ab4K9i<Xn1kMq&-OX#=v>=uHGCDBFsMg*QG!PX0
zK`H6-(%mdP5np;6WTfo|NSA4U_u#F#_J_Y<W6%)a=jIIb_eLAu2WLc&Why7=(?>b{
z_X8RH4+l{tJn`{~mb+O0;=~Yg-{|VBuH`nZI(B%G4M4IbC56zGg0X)^Jit&ZCN|Uc
zNN9eZ0=u9HW^qzS%!tC>R=p%|k4Q&8gA)NzaD++m&Gdtu;fp<e$o;@_esX9=!cyF>
z+JZnc;FztJ@uocy?U<Q;=xq8@?5T@%2je5o1nF`{exmdX2tC>1oluCum^J(uo-tG5
z^)7xHW=*6HHtrhrGFg6@N)>T^I#$RpaBaJz{59r9AVtxhy%Il*bAd!6`s>dnZ9(9L
z`TL$5&kf(3s<qDG0--NVt%}PcwTDru0~i&efrhkSV8n?nMu+1tZ9Db}gK47Iwu=S!
zMhM+^%%|83(|<0TIXRGh$VtREjUM_cCqz$ES&o99Bt`v}Hs>4)AP|fa|HP}2j{yTD
z^U(PVlqXkGFL)*YMuTD&yHb~eqeC*$Amo!{kP>1C&V^W$$W$_Mfm&oGC%S4fDe2`r
z<9gvv3x?#1p~~IF@$$fyP(t8*s0}whegw<+fG!{!uZcN%%uZh>J|g1={+K%hA${rL
zh|1S_yBjXLNNzj|yl6ZiPi3GJ6uE@>!eh&2O+%Ssq)#aO7x}jnT3hsD0NK#kCN&5y
z?4VjpSG1_4ngvOP6pDP352ywem#?zmCWVookJBeB6(5d8N?mu~t79opgMsA~i-S_w
z1AuQQ4LKGwtPH^d16*_ho$Q_KWUgyc=wK~DJ02}sSHK7$5ua2;s+p!lDu0&Ti9jkf
zQ*MBZ8VGVU(9g7PxS~_~`8k@$)^hPMQnN%GXd`>`#v>X#=N&g%ER!PI2zTNEtro59
zFmx*0@e7hI!OE$CD#Fb2NeebJHzqbP5v-_}i#<S{M0ni(CPo3Bp`cvESBF**{m?&f
zJc}+gG90SU2d>2Qb+>r`rqp#NKr9(zu}W7u)hG1j7?(gpqDYvaZ|PeNJ>D2SK@?a-
zRpoM$Zz{F0h=`9Boe;l-kg@U_n_yteOjap`w8$NsK0;~{gcBsk5>C2Y+~=j@Ff!&P
ziXSyQ^o6>ZRFS(7SeTi~^#z$5N5Ka4fewWNIW#W?pWxOv#f&bZEjMr6caW33=mLwN
zFLJ`)8e3-fd`BFYN@7=Z^G>2b(oYrJ+9q!bpI>xXSMqAn%226fcan?leV6OFV9&Xh
z#Pjb_F~5J&RY#+gI6Bb=JlM2p9S#97fxU_f)uTcdM&zG3`JE_d9Ev;Bn|x~`kG&Sy
zM%Npy!B$we*Z%hm;cRby8`50TB&A-*S0x|aIDTuBRR`TBKR7P6uvltUu?C+g8&U-%
z8C0Ys!_T@a-CNHB>7_qN<M(nHQmxfk%fY{X!q~vFs9~0>jI;B(5wU0IG!kqf+km7P
zQO9|SvwVo(Tck|+0Y#Qx(0RFMmLP74K_i^A9U@G;hML?VF^<3>GE#P*^6eeyl01#F
z8cfatZ+>chW>H5iQNNhHx;hZ^y})}IB(V`Mg2Wp<T&0^8gPLFai_)Ywc;u$%#qoXD
zodCRvm*XdL+qcI2%3ALOD}TAEdfU1<=A!QxttEtC*=X@-qX`KMF>9F`yS#`D<E3Ik
z^Ci9pBdLe5d1dbFsekO0(9%rJL>n=+Xj(~(V}@Pkj<>Hq0;G=|!$7OEI#r+U?JVR_
zHIbM}T_F`rO43>GrMIg?3wiC!dKFsCp9)!E7JP(eN_6-#q>rjYX``_dw0x){W^WiZ
zm9#>cenT0?q!miI99nY=3_mZgW*_(h8JR5jwQ0;|4t>Nr@c01c)a56_JN1^G;<Rt|
zFZh~t#xB85BcHXCH^1HLIk1!qpq>pK{n#%+0Lqms5teAuhA@Kxg(81)FjI@b;iSkK
z`Hlt^(aVLIn#@Reg(C=Bcmr<eaVC}VD2X<7OE<BuD#u>>uLE(=N+x<-LHASqd(SH)
zb(-3Jd+CZ6Eg?la%<O39!IlUu4LsP-p@h`Xc%6{yugp>v;2eAvvm1)?-NA6pMCu{B
z;{I!*cW8+r?b<k((B!qga_lg!FCkM5+}KF~$73iumg31%PFeMdOMe}l*sJ~<z`rH!
z3fk6<Rgn9}i4B{%p9`{|1w;jr8Yo21W5QG{3T7_FIy;z(`-jq}_OU@o!pkB|5y=yv
zT#AfxSP^%}s+IBBJ9A<net8i%Msu7FWnS|l8_U$3#S<J8WCaMXB(ehzeO^?`XN2Q^
zyCb1r-oTetfoV&{6=We|v=!EP)jpebIrp;?NG)VRSOf)^0xrfGA8#3{xYZ(<XazzX
zr3G~2KPf1El;y7Q&&GKT3|=k2s?NH{>z+oINL#;Ovr+dGmyD0)QuZ*u38}T%o*?X<
zD|q4C`)yfwDJt%Z7ahyKn9?g6xOvs>RPdpyq%!=6ftj(Y>KO}Dycot2EHc%Ix?`-;
zD4&HcvICk`S$jIN_gTafaa43h-+A-LxMvQgRWvj36lrAC)HSEp*$Qi+&im8f#xGA9
z%%9!7O)?Hp0qJ3mca*vY9gfH3Np@Ik?`tHR_9~W9YwY06zuKRguJ7UrS9=@VCXWNy
z<M7Ps<_do0HK5a!rxfd3z{C$CY-&%{5I*XY;yW~z@AXYx$<)?L6fIhZbKUX3iCs0V
z(~9Bdq04N<PXXoM{Y$~U`^|WEJ}r7yFUm{*t)^W7Ek2iX$2;n8&G${aT+$nx%Sm~|
zN&2~-&=&P@&s_MU)Y`I5f%nDTM2f>qu8zD`1SYB{(msd+QI;(7ZO+seVEp8s35zB-
zf732h+mO)9cE<JN(aGHiDamGNxDDOJpi(38K4jA48mKFSe4IG8LiqNh>!3VoWXR<@
z*`O6xiF)%_VNSUjv$0?PHvFh`BcCAvqF8Nkglp5@#RVe#B&+ktnqt3llAhDNmr1*4
z<1$xqkes}MOmMBYYi8y2y+(yW#&;p7F<%)DUcJ<0Eb^%ccC)aW{(OrX^|f#Kt78A^
zT2@-#F?2kTq7doi&;j}<8`+8h!Yw!IomA2>h;oWHS{mS!5DE-z(~k(ULyF$?N+zUl
zrl+TcL?z96q5F%MO=?vTrPBFEgR<z8WC=^x$Z#ZP`b*+rsr-1Zz0$ue>VQf-B1zmZ
zCT4VU$M?2GcIq@Sfbfqr%&3us3g%{{$kM4&sL2!*E~T{XR(7aF{N6WK7EOx9bm&BQ
z^u;y>b)H&}n&^&Qboe`;OFt`tt2S+>ddh08aoP&$4h!E3+zbt)4(I!-V9@bk$E3bd
zIp#RN0bO$?iyg>1)Vsv?37~<Z9R&3EeXT^^j}^L1UaxM+S;o!|H0%8)WCc1OXh1_M
z(+=S&#Qdx!EU8NB=SO<$Tc4TxXW!!2Ot$kZDXDM)$SmICU1$tlPe2xV3s01tesp6R
zWg$#k4jhkux8im4byX#&iQWiu*}@h6wj3V+QJtYQU^o}oq2-m>hDzSRH@EvKtY2RO
zT$xgWJ{9=eO~OU_1A-2+kMNy8nPO)g3C3E{?6TGFA_gPHZ^O0e5n&(+r|%i@*^S%J
zyKU14%NxQS$D)=;M}K2&SOv4`_d|%on+x`mx1zi0qCGML#DcE%J}U4;6^Xi%>IP@-
z4%Fg_t1hWNwbZ0e8{$ht*dGq*=9iK2A-{O`(uNIRZ5Xlpw>9!+0riunRr-QtcZy!O
zQB)_2<@-&qymJ9ymwh?g(2_w1q^f-auaXL_2$bYzy|x~^2tUs`5tl)q$DTXSKB|&G
zzQL8Q1x^N$u2|^@?vzxL>1sm4GgEVpSc5KRqq<)DmwYcz>Jmpz-0BM?q85m1_gA}k
zB~=fYr2vwlRTbz!K6)4W&nVO+U^eYqys_bhwR~L1{O1&z+C+3s_OEj2?)=H(Unn@H
zIWfScJCh!hD()@IC5<D(>Dd+4C;-RvdNVp8S|z$WI~|2$TvW9uuv40CqqYV$4g==O
zR#H^*6(y7yvRc1l$b&ry%C$ci`jA~%{MFP{GPbU)NI^4}5I=?xKj-92Tf3BDI~T4D
zXo$!FkEuf}I=1uu0d&NI#azW8*VxT92*VUb+-FS#)D$rvO$C+3lKx<d*ISEZ!;P#o
zJPx$-;fRC@SPmKVew<|1H3Zj_vijG&o8$_QE)B_6A;R>jcCG>Afw;~zG@4~~W4g3K
zurT4FF)#0ty>7wR3>4RzTgUEP*z|HqI9FTa(oVLD;{|Q9e#Pm7tOLuPY4#7xu!dXT
z2i2s?+Gx{ADS#P2O*yxZ&?oBBzD%?hc#<nG&ybD9B!Am}VO0^KzzHackSr=>iYqyC
zDDgG6n*dx}o#}+LwASiB{-D;x3w5x8)Rs><+ev36;vL@*@oHZ<YygPh2ale_Bv?sO
z&=>$OH~J0(57fW5w)QnIx0_BSbvOojNDn9ygRuRJSakNXNll^Z^`xEafDgjZP%t{)
zSD#})I$2!zv6e4%+vlwjxz?3d1;#O%MWexWxqZ`xpfnpGqSnVO#c@L2!iqp*@}JkY
zxz<He3hy^{kzrk{(>LEyxBv>wu3kN_8pQ4A&F7QN-Pu8F1LI(99^{d2M5R3DD+O|7
zVKq%H4i5Q^fTP^Y%XbvGT`D^{$6g^j{h^QpU!KhXu@Aq6J_zaJX86gUbolE&RS%v0
z+g#Wy=e&E!c+=IZu&s(yLvZ1Kb7Z;XdB}2ws~0_3Y4PO2SMvK>yyi4I-NZ%5vZ?8C
zSkddg<Cr+r=c}^@-`}avaw@9ZL`I}XUbuV1J6#oB#uvYj{e*zU&21S3T9;)B)8m18
zfDG!R0R0hZL&T2DFN_pTIie`4_-O`~7>uu3B3TP}?CiF`rpG72lZ#cuYRoTy+;DBY
zW<r15m!Vwie95O!>lq398MV*$qV@b=!J>Z~UPjW?UA+}zW7U0cS4IPFf8qMR9QA(R
zKMv)?=1e@b#Mz+PE;IM@Uq`{Ht#2Xi<@Z{^F8Zw5ZK+Ro#~O_S=YkEZc68>#>$;CZ
zgNSamR=PEVagcyuFZjEhsdO=&FQZvIzs5W=GY_`i=}a()C8C;>>a9@TBs1a+5m*XY
zYYeh5n!jU<Q~;G&Y5bauhmi}*E?n=#DSKHT?27U3eP<ikMLaq*r+nPCTQZ-igsic$
zJGGbbaY4W3O)bQjMJtpOlMta)<VUAXnCYh3jf;i6g`4g^9w)D#A_0?>xnWZ`Bao9%
z92!Q$<ExH?i|Iw(Eo^k<CuBYH(VU|i$fCFYigcR#y>QohS7p?XC6m1LvThv2=E@oo
zP~C-LDe-d!!YeS*ugAyShnuseb*Ga)os!C_u%fFWpN6~!`e0b5ajb8r8T8mo^pyj0
z%az)0k73)*h(#xged=3;rZii9jxgC(w7}@Z_=wKU^CiZWlvI!welo@&+l<0QhS4%o
z*@6Ku_*h%oF){`qU}k_|t1ibMqZSN8a+0Ff?VIhT91;XO2ba1eCoGM5S)-D%_N>kh
zBFHjex&ij+ja?GMZwIe)Yx=m}p%3zFA*CItq*R;<%c;dC(~satHA`i!OyaF$&s%Im
zl#+nim=eWX7&%_){5Qb>+J}A&ALfW$tjdU@1t?#GOSma%fb$R({j*)ENvE-vn%8(1
zDHb<&91EU{J*}=!Bz<mWj49&389hWo#i2yf(WKVRz_`ulyv^-DDXjStfg@8>?sPgs
z=3i5lq7xx|+is&2xYKuA_GcG0#(@dW0XxVXXL)#LvfGh^K-a*)!KtYsFaPnq;XtpP
z6Lut<kjhrv?9M2}`VJ7>-pG%NGr1QYu42T(+uSNjA7U)WqeEiZB%!u5ZWN@R*RkX>
zk&_bsMe^ixr+;tR(kIw$q;UrFZhG?kK=;+IdiSMKS(HP~f@ZHxog+7)+^La7Ehkex
zXM5CDmdlY|&#}?OyWziOI~>2nfal?|mbyegH8O_tdViM6X;MSDWjUk2Ix)L@2gM}q
z4!hVM=4qS3rj&-)<XcBjq*V9^7GgC#xepw&;+ND`j3rAC-vSBM4D4FFRg8KIV&mwf
z=;fBr>TY*;3jY3{Ez8~X4|~X&V!sIX<E+~Dg)?>3{LuT+Bd*{sUXh1GO;nUnka_P}
zsgOe<i&4m7w5fr#4i49lImig9e9usL7$po~YgY=bKy1&B6<)fk`mt*+T@N6>(ruz9
z#x4j&^(-SLgXl$joU=V#mz$|T|3EX@@*VPdE0=b=%2KoRVyl=2gq&7RIT8WmdRbLG
zh#J!tIu9~A{=^<_bQB-ir;}-lGY$3#yzOU}lXhoH{6ZC?Z4lxChuJ=f$#{#dO*_Zi
z@8KsnKp!vvLKXQ7*BOsBjU(u9KbtQ(>^q6apYc_(tLL7uLeU7QXNRM!x6Z%S^2N0K
z$rg?3>TQGK2Qq0I$<PtqRCj%8&^jm3HPuq*B2*fyb-0o#_+BI5bhm%Taf@g&UxI?H
zed)dmyTEDVS_HXix$lQ^4GqdUAucDu(l7TY>g>^(aF`>lIdo{Zkj}A29q;W};;}0v
z9z`zhh9Yu^ORfdpkCI_hpku9HFwD?;Cf&wJlX*`*c|x88zu66P;67|r<lJCPY8eps
zTHFyHrrl8+UD&w~)ARC?Ef--Qc{*_)Po|A@J-{HKotK(!nyM1!o#8bhB8QLz(Psx)
zEl3kP=+WAiPZEh=$Q)*j2Ypiy3&l+67NK0cTT0xhkX7B=7E&`hY};S&V^4lLa%e@c
z5b;4|%jpNjfIz&R=*7dXc3;fE$ymG(hCCrJP>ijstG6=+B-c^mFeVZb{h%ko-tj^C
z0(Z}=T($S&(spr{ke1F6=YXfcp7W`jyRD~IpPThJYr#8Ef9KpJNbx*2p;_qQdG-vI
z5pS{kM$pl_>E+rm={Q#X8EzYjU|RH6B0(8ZCZpPxyk~#Eu?XrQpAC8<P8|eei#RuX
z{9cti|CNKD$#;+2;@y<3^PCI%$4V8%gIhTXpT2QcI^UY@oj$O!q*}V5>0ll3@V;~G
z8gZQvTD8y)Crwdyuv%D#wUgs5PF-E2*P1us|K>E^_Es3JY1pQ(T+jH0BkefTwgIGT
zl?+<vQyKm=3V}t){07~gb@XK15VhXwDNs_cnhl*YebyYpHMQu*(af^nw+bG<v`!E{
zE&u{M@0X35tTtiFhArB?;iS^B7k_7EUPT$&ji4K34o*%0hVocH594EZ=vt6bFuCe`
zc#S6jL?@>sn0J!JdQ1;qF38FyWs=OfcxdWxRu3-vxlQVLs;z}{ImGt~GBkDGH|5$p
zH0)~c*!jSPLLeum+Qf{-Ehl-*J|>xACXFCXDj@3B>5n}ajdUqi$d81_K9l8cN-FPV
zeIo>SjJBoIv0MaI3}{x;i{9#?SC2#$De&XtX^Q_4^Qo>xU748!?`PwGla=5o9%h&(
z#{r0iQJ7J#oSeFPoqdQr`BK@~#(Nn+O@%x8Ab$CaoLyVaoV3-!%i3Xty>m(mZ1^Xa
zIC*Y=L*xXm>RQCdteP(N{QZVfTEQr(a1mWbmanLUYeeBa%_Sw<#N2wr5WTfk7VZfb
zN%)D^#NC}R5PTs<J3Io;sD5@(Bx#HUX$Fs^8#I!)?AX+<Y>_pV5MGGEJ<DbDT6W_I
zL(_;qYQ#EzK|>;@%~EdQ#^{fU8*Wj2)tysG!#_kT@OeAeG_E;#g)MCiV&C=eEDhDY
z(eifIwwtBrX5ipZapiaA_aI@XZ!;~Hs>v@dnkOX*!z{=+#5K@jLwwDCqjY|dMk1~|
zWJSJP>gQdsw-iyO#-rzQ@G3T?9yX!Z3ssK<f7KhO&@yX&_&&UNnOntaN@8H}8Iy)>
zeq!d|H8k7Gg;iRnT^0cj<)^XX*oI)ui4kmr@vn5}w(ihkV|{&!&B%7}Uyd=hZhCJB
zf-ZKfDQ~haCqjf&2(32!9@sa0cHTv<n?GB>T$ZUWhEEAyZau%XK9DqPA|sWri%1dU
z7@Mfv&24Y*`Q?(i;`GAH!FkH<RE_i1xq>lAd>OsTp{-2+#jQT~*+8uThp*&nmm<5^
z<tWW@P$Rfb+rA<R*QnMk&tMXwT>GIiZH)hEdt#&f9fKq*;;RGXXeVQn+0%E*O}jO?
zB>rFCxs6Ky_+Nw6gu1PlJ9!XCdI<kaN?M^JvaN!0{9y3a67rexfPUw`p6!>u6q$=~
zKDO9t42({3;v?NWiL;FLoBRCM9c-P5vs3DB8-_KSInKZ{Pn<jnW7QyHCvnk-+@~w!
ztxL7%2dpH}q*-qji=d7NT%{5-+`(xa@KuhgE=z=J<&5h5_%KT44lB~4j<-p)-G9Yg
z_caEqN9d*%C(hgT?4d{P@Jc-2%0I5wJ)vFYwJsk|*QAcbBrg$?4+}nAt7JFS1J(p?
z9w*lOW8cN%e`8kG>%vy>zitfpv|ZhSX%nSDJAC9`*D`XXW=t!I1>(~AM$D>=H)K{^
zN&IClDPWoRV1Uok2}hn)F(h8P<_Kz!X^4(@?z5~wEgQ#PL}n6%@a_PEsKJx)jaR38
zq-j(%*SiD-7&+FYyi%;eT5wd;rMpztkOI8Wih&3RmYl|q+-pm`U7_ctfj%HaH_~jn
zv-hoUZM2ZkE>>Byo(bI)i0x>s#EdtZ<e%Z}a@LRh^O`cvq#@;UDAk(pb#d@|GK#i*
zIH3yNwrS77fBNn;cb!3cYFCV#iuz!jK#ROAQI@Dhzk%Bk@z!d$t#Te(K`mFw9{qhk
zouG6IP~l0J@jVrOs##8`L2f%_JS76LREfE4;x&s-bpV4d!}qiJy=pp;$kqVOBKbZL
zK;~gDnH<OTj6zPhACJrwS!HM90`>{F+(;(rklkqvdXcs{=S+m61AEwTWu-va*fved
zQZAM%x_Ny^AaNX;R`s(@@>uo$dy?TD=1=I;X3(nD_w!${9OZI-nO1oW2Af^*Z9ml&
zV)|W_*39GCWi7?sA{bqdM<+#S`h(2*2jYo2Tj?im)uMhx!p&j?X_-;Xlq7-@!6Bej
zZAlnMXoz90?qgwYG|97y>f1!iT4CF2yO)}2H7H6QhO^&c^0`HWZ$7kG5b_62=So-q
zvQc*8ku!J%<u5$=0vP<@928}(_T_l`*M;=n#YTFSb=IFuAxZdO;}=0pBo$sMxwcOj
zk!uWmt*Ii8Df?L2g!gkR3!D^;bUOvhjMakw=B{EhY}2~}#Db=kC4L38)((cBbUIs|
zR<vtBub=gmE&UY?p5NwP|FMzVCH9ap;Qfpci5|SC|F`Z%q^rT!HH^LI>~tz%{Pra!
zkdD(xj&__gaC}Ax*t=^<GZ&+0DGI(4>RNF5v=BFZ!VyxQK5oxp#)zkz*WkPA9{ALr
z7G$btz*Z>t0}m02H-eu{dfuK$Tlpw9sC+T|?s2v(;Q2T2#=}-%i}e9Y456torK+>$
z87b$75JM;(uB9%o>C??bL^BUPE20>EXgaCY4Bb3JA<}SA3n|_y<eA=qbX$6jS+7xF
z!`MAP>r!QWA4y`GVAQw8F96IrZq4B=i^o}QUprm;_*VDh?Lu3*7BkFe(oni7tpEea
z!;!T-i)6241yC3qyMA<3nxZU4E}>jjBM@xois4d}lA*J^Jf~QRGdumbpcIYVc@BNZ
zAfby<$gA*`tC+l;1k`Bf-T!)@ey~F~VZWDm;+1RIH9UMrfeknH-M4`a^3SB+{cZ?t
zWWb4#icPnvhdYqc5JJUlp*DC!incB~<sVUK5oGmO2y<xntF0;qec{C^C1Pnp{dMe$
zW~9rnVl-mIL6S0TP%INMS<J|N%>f<VGC!-bNEisCUP!LdGjP_!@i1-KHPm5X-QIQf
z$N(~802ku;GxVbI6{MF-`gy{U8O`z7-gAHI?yh7%3VIt~-|6()r{QBt9VpZtJbIN}
zGg%@J{F>-8;CeQzS#2{m4%4?V-IY?zVlE`&Bjv<lG<)uCVtIWwBbSd>c~X8tp(Yc4
zEM$%tahR1z%+!NY3Z)1v>F?GtH+34^T)9y+5)Q_Xr5Pd8so{*iKFk?_fUtjorGBMe
zy!RjITj|(x2o~_Q?2JapjeGUjv3#c^g!n=aA0qE@AS~o6PqZR2M$i}jp$R3vKcLud
za4Uggk0$i?)A=FhmiPPfakYG2wSk>c|HFg=bzzA6_Ti_!^_vg1FRjltAus1IHlqy3
z%Qz&t1$j+y+0%6@-=lTjpr_WJ{<^38R-c`IdVDNF!5QDjTMXlXXXsH?R_n$i=H~r<
zo$j!GtI1IjT{TKExAzrP6vL-}->+P6O&RNlGq;mV<#M$}b5dc-Pl~ClLZX_Q*dW>}
zd!1PA{Q-mAaV>6*Aw35v7hg(m<oqIL2``_qdkIHmiI^Hu*|^}Q3j4E_YQUAE(Rmv$
z3~Ki=VXq=0Hz8V6Vk3B9?x?Iwuxrq%X|HL}I^gE~<_?Y<oRCl_xij9wg-vU1Eeh`{
z8vX{|Q6Ol`hBIV<#;>1RUr3)5!Y-yRIE+`iCY%i)$$Ky2ozMFpj2UN=2wQcZwBpQ}
zH-v#pDia+|@kM^yN}f#`;<ZEgOxrcEU+|Z_Um`H(SFTzx;Xt7{zK?(YLEL8r**|Zp
zVy;&<`|zMHA&Hup#Bw;Ocgw!_eDmtyCx>><c>8pMXzRy}3v~6W3QWWay0>lkqK58I
z#~7nI$cmF`k;Gtvo)T*E?wql8X;WcRc3YHuG2a2kf`Eb6pp)3R2WKCi=p~^32#H?f
zw5`K|Ed+0EI$1Ivuj>kX$UwD_Y;5_lPLExqL)%>+lC+BR)Mzyvb5TZ*^ECNG34L79
zgmxYBX|1`PaNo?;cX6D~AJ0(KMg~iaWM+tt3p-tFrCeA4pi=I6{zZy)vfVAvb8zJ}
z)^#m$`*Sgvv^D(ljY00Z^|AvK{wp&<GMUc$<FlK|N-s-_ra|mpYJ%#7TK7VNY`*5!
zAincFOb<W#B1V;oGgD|4+u5WXGjXY*RW)&#C+RS)E1z;1r##JhMi9j-+GTIF4Yki=
z<88Chkdgt1-Jh{*tZdWgQMN;9%qr#ofkwA^)l~Nv{sh>$Gy64y(b>&D5SuMFvJZkF
zZ1z41NmD|%jG3-5YpeV?`q#8g3p*SLfhCuPorw1v*4s6B?+7J>Xkg%I*={{?UG7Xw
zhx3Ak-A21BSLJDStl2tWg(RTt5h($*Z?18`F^hV3o2G_0HseBCtfI`axaFs08%W1c
z*xnvM8$3MVU7U%eRjy2y5;1?<tr-F;OpJFCTq;Llu|Mu<2IIVK(fOvWnFpgmz3nMr
zt%2H6<Ns7xGfeL4p)tcC1Zo1R+5k2Ss9o70_EeEiUQX=M7NJfilr;daazaEc<>qlq
ztLz$oHI0)GQ+27wb_YBKj@_+4DD)Y#Y(;oJ$X|YrNN=;UUsD2+-PN4j-h>{pTRk@i
zRc7EDVp}N3_A9?Uk`eSdzl^J8mMH*c!gJ(tM#Y8IROD?i_kuVB?iFe9g_ZRoD*HET
zg0Gxq`f7@d<Y3dRseM$8>UKNsbuk+AA?Mv?*N%gGg0-4AbTBM}uOgFQ`y9&Rn6g3E
z#K(LeKIXgjN{>oAl}tVibQ@BDxk}8L9e4a1i?&eymZ6Tsfh$Q(e~?$4;u|Jw<RUyb
zN3CqIDuBAQb#!@lonF@2WyjYk9TN=b?1?G~i09FrICzzQjn4FLNj~mMZsYdc%9IA|
z7KpyQ>&XcoI9MvU^zl0+R<odNd6l%Ly+lQqa|!R%S50>ar0fx==99-u{@(TcwY|eA
z2^f32mwzmwdJ%qJd(%5?{FoM;AJ>HC7Gmm-mdsSC;%-%&U#!lCgQSI3JiHnQ)aD{#
z{5<glvwj#TKdh-<(-`<^<~hHP-abuu25D$(S3O-?OZl}J(yZ%hb}BPL8D|rpzU6A)
z(|q>$xszXVRjfIlR}%|ICxveEJ>gM3tzOpA(7Dgl)H5@Tl9Ucm*?#KK)DqAva1)@t
z`KoBbrp7SWw7-lZ7l=Qmj;i%+O?mf%$XoVR9r{0lE>G6Tdf|~%wv_XBOfZ_6Rtp*G
z%*3eCi3-MrTYo*1z$3b!P_79MSxd4}9eL`=*|Ab4Li^1HfhX;OFe|9dk;`^4`?!*}
zkG8MfN@`)+U`2Cw5xYPJKx}R)fB5pq9{fI(#kuEPfY}kKJ87U2lPa)A=`2>eM2LP=
zS)_XTxb;`~L&et!!f?}+wpP)D8h>ZEAg^Ea2ACihn)TZFVOEW}(SVj?J%<9gQ;WmM
ztXARic-%1N#$uL;+Sl(tpr1|b72QlZzI%xA%n<nIcV8FGwxm|?>8ksq!C*4~p4Qi@
z@}D-NqaI>&SlT*3j?)~9;T@5nO!zFmM<m_X{nb3{%x3ld?UiPVpF;Q=#VU*6KiCz<
zG9NI@9_7=70O=t)N1EQISdi@Ovj$5iglIp+Lp#~FBJegB4Pvq`RtMq96@J%0P5t2b
zoen9pkr<-)4E(tL9mM_Mlv0ma7A;=b@1N=1BmMTIs{8j&Vp>mPZ?$vbVnhAtTV{PR
z1eq;3IY5pNMd#wtmS6Hm^x^mQEN=#dlD$YAI5^aXaHYp>;=^B^jub5$U${pcb~Sh_
z$zs*9c0Xl-YZl8#uk|Uv(y!{0l*%nTObeo!mITa4#}mWDi0P+qy7k8Al6KCe(8tY0
z=-HeCj&8$8xo-;=?XD<n&tUlH2%1D57;&!1WbN;A@3FShFt!*A#LNVHu=)}OD9aG%
zACMVfx+H&PON>@}@h$iyz4F_7I=27J+1J+IsElUiJS9;Dy=e5@tmA$}A~`rPI6eK;
zj;E@Hq)Q*Fl+qE#X5LnB$I8Wg+VXewkJ9jWY!yY@{=piP6Ns}N49HH}Rk`%8tHz_6
z&W1juENkqj?9TsA@f;*~oVQeGA@ww&{e0r*!D$z2)!mHzGrCHaioXa_?z#WxpK$QW
zo8Y{c+qvGEYI%2=v`OQ@3qht5t}lyz2M$e^yiV!qANbxDG@{s6vBoY6@@8uq)rj`-
zeNYL)k`A}OevHrQo#?$8eZlBG@9rIzY<(~cx_sK)*6F>Y2sgDSGphATGpO-Q!~76n
zL#*eoO|0YUy{3({Z1FVb`sy|_-k%*mJ7tOZpyl*V?lieipC>inVDMstJ$-z0M)3N|
zw_{*CBC5(~&DO=7Uam+ku3$pmE-UlPjPUz>d<z=C5J+b#2*a@_i$~prz4zBk6D8P;
zWsuvt=XM(}fl88LX3orDrN*~qJBIVtgF~>$Tj}|fW~2=ui|3jVU;Pe}9=5J275o?6
z>oOsQizV&Ma;U8PG7~@LGveiMb!x<KMkLGh0aci}K-1kQb^DtB)RI&T$AMDbYY+|5
zE$^w1S%NW5>3mWqA%RdC;uCZ=xu)8FvDM${Fp@Ooq>P^A+tT*OHoI-*?A0s!YN#ZS
zilAt=Hc*m|E22+{R9JOvTh4Rs5jCeo+t)*%JycJJbjzmeerVP9DzB~CoSbu_=cV)M
zTHwR`mRQt2x4$zhMqiScOi{ay3LWMdjWW^tit|9xB4fh)MGmKX9*P!XcJdd36lXx7
z5vF#KR{M**%-&P6rRZcqF5kh4UuK2KpM-SMghz(^zTYjVWvBn?%~&E6jAudI*Nx`$
zf?4x{W1vS=x>%vnd@FNT|98$P%XjV86|3*JrfAv11oa+-@zO!_#4^?_keUS%r)UbA
z(9T-+EZ{skvmcsQqqSyrRI}NTR%M6&qCp-%31djQX^pBuT<1A^`j0NYLlzRdl4u89
zY)uRWe9*slOH?Z8crm)cx0v|r-G{D!YPdtD4EcxsJLw%-w}xlJQoOEy3J%trdV5Ld
zT_5AhA5Bmq!@`H{%X~}a=-<S})%0umRU4WodTmU6$6YmQq?>L>(5y-tHTg7da-Q#-
z&u(j%owaRYr~%O{P;|0@%YZ^T#<_D7H%axKiLMSNw(qda_AK}BhVOeNabMTPPo15j
z8?8GPQ*+~8A`~j*Q?Y+m4w1xe_|%*uj|ng6>p+l3yT9_g3R&b!ieH4C1&;*Z-}K0!
z$JBM*put@S{f&J|%CT*^KEVb32oh@kq_>|A3#f0VqArTKZ)xZMk~LH))VmMxOe=af
z)^GW*w5a*QcB)8NOJGZu{&%gC6)q&*Pu)9>X;+|7$Wu`VCS{mDWqyM2<Njx3S4Wy<
z{V<Dkd3%v1QtDAo+2d4Ha8&R0PH+FhbF45GMnB}xwaT2A+*DE$?iXyzYX9-#_)U$W
zF|<^@MY%>R>(T<R&8|#6Q31OK8orQ-iL&pz55J2FxLJvMyuUiTWGx8H3;wm<JM_l*
ziQ(X8liDX(1?VRajZN2L(;FR#nFS%8CQ3>&5o&<dCYEzM3@Q3;>GhmCgO)fcp0>1J
z)XK7I!DGHAv|dy1>djoe+F81v-;v(Qa4pqQDP89qp%qmVd({w(?zmq+j8e<zq_o6;
zg)CK%0Dh1&*@e>z3MS8du%j&l0ruiJd$#_}F|Mtxy^1N!<LN2<^&2-6jT`VbwqW0j
zCiw0TO8AKko$kvL%loN#W?k~d1BRai;(8;dy3tovCf-5P<nosslcjv}=iLajm2q%9
zw#-gVPO31&GEnOG{m1tiHyajGCiy}t5$4RTp^qo4p1vbfkj%@W%cB_E{AlU6fHQlC
zR_mZGN~X)9^Lp!G_o-caF8u91J_B}$p=Oe8J{c2=_`$_?=!sKUv}Yb*?PNhH%~+1v
zvt3~4L82`ENzO0?#zRS;xhm9oo^R04MbJ!kp`?JJ+^*hiwa?R_tpC$wolRZipBc7X
zdK?L-5fo8g!_pL}+$`QgBwVZQ&DDveW*<?c>l9h#*58HQztG2xnur)P{=NISg^-3v
zZ66Z{RKXQ)akXP+krEEXkO^l(9lAe+a6#z$$B0M^U6|E|8!JYCkCQLoBfBrjJIj_i
zKIZzPedNO|Bs^+m_2G7_I*GRUo|*N=RD-}b@*86<82X5+o6JRV_nOOx0k%IY!Pu-#
zh-FhE5V4s*K^^*zwwULY#=KmURI19BD)GXBh%sN&vaR9%O0tZctzfAz42(u$TByF}
z5W1DtQIh{pRJOOY($mbAW+_-vdP!om{ItWXEHs(f!ZYQmCCkmxWPb;@%-OZScBi(~
zI%A;#={1qV`=~>N^_Qc%dsfNCPce4VHHUZ_hVR7Xx@JL;6f>Jh^x<xV2uQ%!dHmgh
zMiPCDO6oeJgMVolCLCm$AHOJ@G&!Q7V({9hTcT#EWbdP;xtp!sp(P5fG3m@O<V(UY
zR~XgqW^W`9?lcf9ROV=>y=i-|<8d=LA1bvzNeXLfaS4-nR#D=zp^0X0X=2uAio7Ib
zu%NUs=>xh@&su(BNC4+~eArIK)<pV@G!edK!2T{vx^Hl<^dY9S!8h0FgK{4`t%iDh
z5?gra;v$Z6-Y<t31)5itP%|w(ls)!;3m!6l`el5;C`hGI^19pNut6h(<RJX3@}#M~
zRE=I%nbKSBG)bI|#`0mbJWX$B3cF5{rq7EwZz=_%K=lG2SN!>&6`v$FgDzx(&h~L{
zm2rckjQtOmZr5*~ZKu657IPM`<lcDOXVz6H9<1HpT&Hx$XCx)VcbJ2shpv2u&qY6&
z2*)EDY{~?h8xb8Xr1jz#bzX8cFLijXqHlt5;8eE^oAVzL?WR;lS)H8Tr(C{OM|Y)*
znB5V@ZHd7R{XGKSbpaRAG>I@9H$9;ctc2<A{~3%ba0?7{bet-JJzRc6?`k{x<I|(^
zDKJa_8WjdN?2+Rd;1TN)J67yD4$MX*WK-pQ#7Ez<a(9Z=9ZwA~nD5>1(D#dIQL)xh
zBhlfLs=?ktb?&}=pED)Aw;!t>AF!Lt{%!Wm7WmmHbZ6d%xkL}tM_p{H3+<)`Wm#P~
zaA>lp&o`LGzc}^-rNgs%Nvuf>ayP9NkbN;$bj{PXcUmff*}y*^{q49Nj4I0sd9SN+
z^jm_Og<^0^M{L|@V!ge|RitA#Zag|yc%0<0b_?R8he%`W=r#{WD@F@_eW}%uI$%+;
zYkHTlDRJBDdX+?Ti))9fhZjoy=sM*Qgh=6IZ7U7y4h{*Pt-aB(xH!9p05~^y@elwx
z=ha-?#`I}n{WLU5h<9Vski$n8%%!D;Ui?F$DZT>y2}+ssH#A(~-_ut@?@fV)HS;>x
zDEoXXP-t$f7CCw?xPnofkanqj0gf%<>^Ab<W6=(STDp3gKK`37{vv>Y+)yg%*zs3U
zRV%S}qM7VZDNJp7-tF^)%6WrK%Zm%9eZ=UTcRWr>iA;7&h-tnXO(MxtLJ6i5*T>0*
zwUlx6cD60xx1JeE+G6=@c7A$!eJJns=;MB71I`#|oXFyq6jWySwb+KSLIh<zj0h)f
zC3xnuK4L-M>|j;S@0S$8I-54&HPdXXn0d%<g4O8YuHRl`oF@p^m#^g4BsPaWl78e>
zfk6?bK91ju3l3+KC)DCCQ5h1_Gu5Dp3cL6>&cDNO{u)LntqY7~+QG~QlEKrARKe3b
zqiRBwI0hTNd^9p~oeiIWOyhBZ+P2(O(}0W=MUxezV3F*J;;WFFd9CrHjgQ`ogD0{i
zg?N2?-u0>GHg)Z?<c4HoTD9bC)c)Pqa?NbNiF?&_dR2{<1j6~~_QfvqtT*?Z18&dG
zaV4Ik){g6<H`XX3*1ZOWu(ey&4vXZ7S{ZU0_2^!0OzAhyQJM~9&eG9ICEwe}6YN47
zm?3VzT*PhzSNbE-ff$-4yIC^{3Af(3FNlA}gNK}hE?RHv0-x&sQk*>tx9pU()7O9h
z<C$fQ#cK8m#)!FLm|LmJ>DkT+MBLxkEjaT_sCmC^Xb;M$I=^$dPzgGGbce)Tgo<_!
zd#zF$r>SsdC@}Mm?z|NgR0{zqS(oEP=K`f;Y)xwA=@_P4os%G2D{~1<mmxxQnfv+;
zpY!#RnM$T#{nwc}Pj4^9n%)U)RE{r2V=fRB8A+tK_V$WiTT!=Bv^^zE9Dg7hKDe{}
zJ&a{;{Y@JEDxLL-BqBLoXz+SXb;<W3d{Jw3j*<k4z0YgS<)x;sUbt62QGrx$e}`Y|
z1fBImpx2&b1(~+~kM{PABDxK~op&4@o5eJf!YS^*d~(HP=29?0U&mG*L$x(Y^$#Gd
z+-MttJ8w69j|?FSr>8LJ=98&fMeV9hXvuB_h36RLfL(!_Zsz+x|3WkFCQGBSd=uFw
zb6tIJJ#P#+Ma{9C+46sRVU9YsVoA*`70l|9>-{DTm^WO9Z_=ri-mYP^pQp7!%g#(;
zKE^u^L!)VXa+5zgju*H^fN}lFwEHUrI<fM=tO1c%<5@CM!?k~$S)>8Df=4}UKun)9
zQx{2z)o)|oD6|c$jV!LL5Z8OZ;A~3ksBv^TU@AzC*{7m7*V+P;QbR5!6RUan`#pL3
zR;2-ZDfm+#IlJ7f-DpRnfxfa1v0DmCxKibJz@++ElW$q5rcNk3p8|+42_!BjBCEd>
zdPVlWb?6*>+hq0_#<NH3o1a-mp=lkNx$HQt2!Dr9W4gxW<n7>XojMreC&aUaks_{p
zS~#!7p_I5y$$&rjnNH_O??LeF-c&BZ_N}8nMo-iNB&LDw<XJ^FFZ0WrNK<50FM4zt
zN<|aQ(laXU<O(8y;03$SI7z{T*SZ~0yoUOAf~dNHtWE_%tERvZ8ySeLTm9#Y0nTnf
z^Axg8$#za>acU?U?H9d>oT@CU=PmR168EW8!MBM|p3-en_Hx@v)@sf^|Na7(mX^8<
zH$jnL8~*;G;inX2fiZG|quiMv0uiFIGZX4L^uc}zg~|l>{+%QkCPF~WYu-9UPtfr|
z-qfGd&CS(54|?pHyUxI~n+_D=)}XYWEn5O`#?s8(&`rTh;>$YKQ^D=;U$|MN-<%PW
zQrRXPb|MsV^apXWy6B|8rvFhU#EGIc%{D}(T%%#`S=n7!KWr#a(ukQ#Yg&O);>b(p
zRENpBv>J(iMgbns20tFd176@%7Zj3ro*SKypM&?pf|Kfc&fhEO@xUju@BpzSXwPKP
z&*L{o?&l3DrOpxh-?zzM7Wcoed+TkkB}eE%v&nGr2i(5)94DmzS<%eYp!wF8(??_s
z8|)^m%JLFgb6Gad`dLU3>60+EAMz<}j=-&bKmN_x#gy@rA<jeJP0wZP-M?>CMMf?i
zyJ=myK{nAYo`J(xSMK}1W+b5t@>NIO%F0$YBOfUyM2pXy*tTqQzji0|E)oQ<UI>`0
zX<K=|iBD`leS-XaxhyIVO+i&gxSZ0-xo=?MNKYwN&pznmF+4Lga0?1D1>#-eW%2rQ
zOt$M{YiPKv-ZbA#=eyd_YjAL$T`p2Mr~HiENyX6oo+^BtFc}s0CIv2`k()VRN`2qO
zUIGZcxYl?63#;oHJ$10=f3uD*`u;jDma~?NeHbo2J9;gvjDRs~@NLnqhB=+h-*AV%
z$WH63m_CFLmFp8kzmWxFkV_ZWTnem4$eBQeJ8Oo#P+Ou!{@3QY<aeVQIGwI+P8gpC
zMR|*LA%0HaK&!he$#1dzO0O00RVmN9s`F0_LQV4+7hTI>pBFDr@$W$MDR+Q=W(=`X
z1UV8vS~Wu1*99yreg72MKTdWxb{4BwqW}YGSOGt1Zr}7A1&3?9E4)ewBmZD!e$S?O
zRiV9VAio?y<SiKmGb_?oR-kU{G!lQ^D>Y@nx<Gq!WZ>YOyzSrP|3``8U}a~}u#n{H
z2aCC`E2;Er+roU*9Lwpywd>ww1|Uq8qthbE(1g2=BW*yDQ!*B950nWwvt6~^fc6@}
zDR`ZBC9Y-;+#yG8ow>co?4f`#e9-xi(#F#XnELC!LsN720MdqJwU9&_+hz5WdVM!S
zZQNBoVH3RQ-C{Cn_I!jpyIXpIlrl@Ey|jnG5kn@7jt;5DuJekDT@#F%MH}%>vns>W
zo%b^AED#hOaa0u}_Qozt9Tu{_$%IQKe(>kgzLY2BQ4`S07N&jq$3K~M!9YjZBX+IF
zC_Tc{jkh%wS2kV3|2eWPo4_N`Rd-WTAtfQ_$O$&DR~F~_Q4nc~xRWGY2>974k`(bT
zD+Z8*qGQYaB(6n(=8%}S=kD9#d2qzhlY7wFc<_j|n&jT${ddSbtgTTut2D;7&9rm<
z7B*x#NG6~^!8YIVINi8z@Lg^H-UjjMMBRBu;=tC{QiJt0l6eA+Q&v2)&8IZAqT?{_
zr+t<XrPXu&qxtjJAIW{^;A`(&!Js|k9pjgUUQjfJ37rDCg!v7o9S4n}^+3&=+UisO
zK2h;{y;XF^{``qtTW*MgJ9i#_bGk)ivNP}L#6^}gEB|7O<gCE?S_(^$cx~lRv$9-u
zJAcL|fsd@ecn@O!v<dQj#dIBA2{Yacy9tWR86*gL+>t~<leir@yHI->4o=IF`6U!M
z{Vl4%H5Zr9inh>flC_QA&Z**QD*xz+LdhwIINan-nBPgYK-p2Uk6U9%V{+-W2o(=_
zGP8pH8sIJ8`OBo%eTPOm)eLXhN?7@C1CM8_9olS_bCOH$&9yHK(&(vU^IauTigoMI
z>#-;sGySbdanib1-27yK7!MUM<<0zSB)t+Qu>~T3_<$&x8OMCndGe)o!zb?5IOj5n
zz7qX1i!Z6<zRtfFJqIQd4XC=w>0ZveJQY22X%RR7PBum6#-K)U>A@w~*y4nOr!{fQ
zVR!YLO=Sqfq*1pTvJiH0L@s%FGq)cyjJ{YBj_#A!1@wdbKL8Iw@V?pLs)*RZ=m)80
zJLsE7zm@oiD%4mO1rEGeFbk)%IoFJAY4Z%Pq~Pn!|0W2`er-!gOY;DgO3_TojrliC
z$-EYWCuhXqRT`XlMJIs>b|2N^xl#yds2*48s&T?btX8{7)#cjEMlj+LQ%k_mB?dO$
zQi|+nLyAu9jMNO?aG{h!4jCjaZrD}}PnxT_`mPg9#az^HZ7MZxv{Z7np5jsqDppWl
zL$#WSo)neR8oOQcOrEAC#$Ib#6s-^wX|%ALybyfLN}i7sIv*(>+t<3D)cHV}Mpx%7
z&@718Ym5nr1LD;>PBV-#PH3&?>V(!z9tP47NJC=4&N-x&&$(^pn=#?;=NQS^5>6#~
z9CcC3G%<%Gs@D2ZQUC^s7IUj)#QGUx3#WK_vlE^H9z&TIh0&5nUTueaWdcYjUW~Rv
zu){r!5OXA!#O+QC=C;i;A#$WJ6TF9CEy~l!Gr-Dz!(@{~g<J+6etyO0PoJ_sUNi1I
zj93#vql<sl(#}uXTrI)!Cx_<}YjdKk*$1E87L~wlckA*pU5TB)xos-(ZUVvj6`XLc
zD$**kh<HYm#ATyeEZF4$J$bP<OHA9vE=*T}d080VT*qD3wN8At|Cqo3@b7bdbH(9c
z3vja;7}Bt8*R>S(H`f5<T(Hc=4dZ5HNCQb)*#})WxRlszHp{hJoj4`WnQ69+D9ip1
zi<?QX5)fK-$sFGA_e`ZM*0bP6aqGNnYE>&S4jjsJ{+%EHzc}AT{>AtHOAP5AB8glw
zRV{O_<+*v~LqCK(c1eX=^J*N>f^E|>mi7pI;Wu}s&q+jF1qut{Y1Vq11MWOt=VQ%S
zB8*tv&GNa~#!f-R0^;S?vu|}P?Ft03jkvCD{InKoo|zAO>O2t?#_c(O;h*_e_?Q2y
z|81T;{Skld-~Qk7!|(nv?|${q@%}e{mqR}Cz3=>8%eIagiPI7+En*6IkVFcK>dZ7v
z91hb$&_nW})K(6O#lrTZ|76{q$ePPnBL{BO_~C4sltu!oi0j#y@<R1zhgo@1p|+UQ
zvcz$js`hFzG62Wx8xLie%@`0t@%YygNg*+WYzD4aJX072mB6^&kt88i*k4a5NAmG#
z3dTH9r6PkBf~Jue-(Q+LJVhga=jWHiVPKk$RA#2*fjk}UfSHa+QKm9es<rsCfsM>!
zWv?U(DGC?64R2rGXFF`!=VIL+=83xBbKGB3bhdhL($q6AHAF&VAXm#?9EQ=AKQD1o
z)OtBDWf)G&vB*N=n-v^e*`yE{)5sVHH@s{v2e6FUd7_lr*34$Jbed@Wv4dDIVzb-K
zO<$d4MgOeA8p}v7NEODA&|yQ(g{jVVn7Vpsgwhi*>TE<^eV+NbV28iQ3Cpm=kCmDk
zi0y)m5C?M|n`&AROf7D8_WS^&-AIcK%f+%}5z8cvF)@tek_b|7shw8kB8vBy4td?H
zFeOl$A*sr7F0Kj&chR@E(Irg4YM7o>&Hz{9Tp{&8DpjFmhA>!+x!S!bl`4ghni6aa
zT(vUK2cQx|^ti>@ih0M;D%Pl?C6d+RZ>PhNe7K?>EqO6Un^z$QR5Ljr2(ziXgTTRe
z{v<+5LM}E>V~nT-i~SU(sP}4CrIg5AGGmEs+1QwdMD-Yc3k5K~lRR;$d5R=>%%}@6
zF?i`WweB9d!XVKq0EFnmk%>4<(wAa2TRog1O7*Y@JHt&JG9yY@_lRE<uTB7)Qh}KS
zRqcUVs}k<;C%1xd#X!qywVwnY=}DU^T^76W`R9-M;`66mUEOeTxnUd<nokZ_#WvA)
zMPlPi3s&}@w<CFlF8{QjTCv@Fa(ww6$nBqm&=j}Z%H3(8{}Laz{P()awD)Y;e?RTB
zYy`a+h`J<<)jbOY#9gP2P-wOyzLvIs;78AX!2jWU|5rZ#;uH4!10^fovYs&|c4s@A
zY;IMxM2EKN_`_MQ+f+pl_AZ5Kn#`I{X}NYIm>V|3VC5j@88j0&(Gxy0G1<brxw)~b
z_?wZO%}s#U3Dd<QvQZF$&1Ph-Gyk{m|GONIGyl@p{w2<Kkr)R=#zj^aA2pFS&(`Ph
z)5iSN9oe`{+o#vq3-NigTI%gOD|I34?dQc^IaP#|*8<|L4xaUc^=v?0mE%*lsktwp
z6+5mTJZy!DtrcI+Gg_7HxZ}U@&-^9ce*bH{@s)G_!k_;gKK=2Jxjetm{mTao^MUj8
zbB19g4}+C~*cUfcTxIt9mx3})M{cgKndfX}C%ndHh*l<oY^5hovwwdY?#9kGC9uV@
zu0LN@4Pa52PvbF{ua;=Qs-(RNuCJ~cH`WD2v=Y+lQnH$G-e^81%ABd=hGE<?Y&Oif
zQs%<paG)F}s#cE2y|qj9o}_CiM_81iwvROx?&9p6aTv*y)ySK3;jrIxbM>4$&!#+y
zm737d;)reQ8)87Cm1()WxZvLT1!u_&v}vBKWX1l5>FPPh{S}qUuo>8ngBKhQo*4;3
zw0$K61}~Ou^_VPaM~n3yZiUK2Xb0r_tb(i_k|0(XdD%B>Gj1*BQBE>-H3Zu)f|XPe
z!Kw~~fXHmbBSc&dFprFTW$cu&1EF}xLNl-iPo~MPfXXrl+rMpY!Rp#(`TG#o@DM)+
z$;Eox&N-v$q;GcHd)k|;t9`gxrxp@b+IXT?TQH_2kuZ9~ot-1={65`-Qm}e~a}oDG
z$f6RqWQZ7Aq2vnKS{7G!LlDH>!zubRRwGGPLBK*n%p>H4q(BX~do`>7t`;jIjhfo1
zDA^U%>alOCWffACCA>td2?mG5o_ahY&1<A_C0Y`wIWu#F5XjSU2|=<mSC{YY%5GPD
zsA}qd(aKze;$%@rh)JoTP-{Y~yM{wJAzyk+?TsmV*qP7eVhJ&5fSehN5=T?dC5j6I
zES|5$c6Kygp#no1mT-bwbLkhvD-*z`j#%($Bt~l^qq<I7d*P~>Ei$Zni}mJ4ZU5_a
zcuo&Jk-+|D=JU^=a(%U@mNzUmS&gu?$$7fpFGLH>ZL7W+HS2WM1D8X(5uB>k`z4&b
z-^&iOlY?hLJMiTsM6(gUm%QICh3`)p?e-;JQ!!}krE1>~F;YWinvTrVOeuxq@xV`>
z{g{9AC;tyVdj1Lf{hmB$iigb{4ux880OjKHjLm2@ry4OA-)!0hZ_i~Zg%abk!Ix6l
z-&m4B%O2irH>6~p2&@D{Wu7L_60}WZ&a)?;*yg_5ZBRyz$D<SE!85BSj>kPAM&@}k
z6+(o6`=fu4XHTE=uYBXb!n^N&%@2dYh>N&_M$4piLqu)fzHI1EHwUx&b?KkmNF}U>
z#Fz1m(;APUOQ{}<ubkVf!0TUGCyTTwC%SYOY04Tj*Osc0O!Y%kDNB4!D}-KhVZPpb
zB7u;P#mZ2q#qR|#mlT}n*N~Ubvtd($7l;i4HCOigYo_^Vgv{f<YBRuEz}|_{b2r0U
z+qk(x(mv4CpgV@($@raS!>*wf;nw%H*5z;7ig{HWrvo8w&D)GxcZT45jR@mr%doL>
zEVWdt&$r!joHJK9H)uI>*x!(|89-B>AV4Bfs&}_&!QS}g`ym`cFoiK%BE~$=9F7NW
z_IvU)F&&Snt7Aj5;=~qb<nc)!3$@*DIlDOL!Nq+e0lwz)aiZ+6n6IyyZ?35G0f|Bt
z>x(*sjjKcAY6fh^L>NYfXo(}WJ)i0w4r*A4U2A=}#5J~O9n6hs<V!q5*0WBld6L?m
zE_lTRTJ~;QN@7^TnI)HKY2iJ=T)ZBlLI_^q*yc}*84p2-$>N61yJvew$dM3@Jmj3o
zwKyrV0_s*3piNt^`XH^)vyq9MvzK3~Am-}xiM*=gP9_CBWWZzUn$aVk35y3}xXaz@
z6NW_<hc(I1lp!lql%hSa?1m>!f`WfeNKVj$h09oIh0Pr6TBP2mTKxWeZCMB$6eFiK
z=jFXk;IrqyxQbPwX25G9iee27jPzI*x1GjA8VJ#fpldO=Xsvdh9Va7yNt+?BZukTu
z(YMjOgcsTV6kNfCRl#fqm!NqyR4Wz}7=!OU0wE@bAgGrh(<*-RYZmtsu7WH0g2xwT
zLKu)>*Q+X&V&QVJR6=pry?>6W{ww#f`L*tkp?#hUMosa375owuK(Gt8?^>CYaFY3I
zukdLsup|ViF%T;mbQ%w|T4yV35A1IW4?lg(vuAtexsYP<V#dwov;MDrsQqp!>=^7I
z)2b(b#qxbSV5+D0>xp%>zUA#B<yQPy`~F1fJNcz$E;flnS8^EuUh-LY4^I18@R<A7
zlhnwA_Y`2B3&-OP)3hh&%JZw|{J%f_5BSH|f5Nlr8LbRVp39<m3RYlwnr2JLsG03{
zvnZIg6pAY=61=D>mbt5#n^^I*`eKDJty6|g&f{_7`Lipoudk`4GLD1CdabcG<2Wp&
zCxo~JF~?|eg*P{QRFz@a5Yyl!$CS!xI<UXl^Y1+S_xb%Nf5iXbTmLP-^{w9qi6;y|
zxLc+LSjga)VGaD{gaoqw&53F(r{w1jVQDc~C+~D?s<t=HB`+?7wKy~<>tL~8z}=)(
zty;d8%A9BBJXtc0K&<eOfB*M+^DDp2_0>mw?>m12B~vDg6`ZC$hr=ET7O$42m6>R3
z(A~jUS)}TUZ=TI1KLp#bOLl`!*MyEG+3U_cJ~f!`DuC_#rsm&zR;OdBt3`jy06SS%
zjcmLs>=3KwMpY<e+Z#-RVjQ<vVSk&eyWJKcaGaldf9AC;#O#Vm6mN5wv$YlzcL}L1
z#kU*jiZL?f+16Ug?2mhn)4^O~^X!L^t-S#+r4l2OMmFPy;o=^<d-u72?>@sYa5x;;
zA1CTObGW(YcymSF?+K;2K_}MLqP4XXYh7rnP$Z@egy^x3u5#4s9tXQOqLD@dc1VS`
z*EA#6|I43Gsm1Ec2}^kud;TXbF16tJoHKaXh=>t9{FUnGNQxFhVwDV7iuQ^ttF$a@
z)?}5K5>id^iqWz-&Ba>$%L-SAcv``sxq?kuGgqE~AMU}8gzBH8b&W#<GoGbbGSA`}
z^RS)bYH0}T-dkyBrF!y+ca}Jb^K5f}ae)WBhTs{$ErBG)=tSINEA96d_ZOqrG8_U?
z!s0-yrI3raqD&A<CB$H%LTc`6iyXFCydbN^YO&u!sG#Ea=|%06L?jIi<Ic!Zs8&?H
zPKa2UpJsA5!ljnwM2qVWey<`zs^*C*F*1jEqVM>=0cI#PL*El!Y?9b>oTf|+lTFoV
zaj^D2f>vt{**dtG5iKoj+Kdt3FKRWBE2e=k4CGQd&XcXBki2}&X=yR_5@2^Wp_A3)
zm;cJ!+b=}{R5f+Y0)f4{HO6LF*Y4`#X$-gInG$AFOs*`o-nnh>zooxXa^Ukv&v^3q
zIagOVzR_B20jvk)>50C&R@<TB@7tu5Wg}=3Mk8b^Icm7;h63MMM3zIZ{a(XLHkM7M
z9Zu~g9W84}B)BrANIm&vk!8!YKA}E2)HfivlCQ6xaoF#vSWCB`Jo$vb`Ptv&A0GY)
zMcIr4xvbPT#z4+%oQFt6t1_f<Q8(=LH&wTh+Icp23u5k*m=a?eyx6u_k<?<Y={Xx&
zI2;Zf4@ahX;&8BDn$b7uFoq>oVw#S=m`#yNF>yRjJbV6}7$cV#_pFcVXo}!Urmb$$
z_n-ZMzy6Q@I{(f454e2eeYV>j^PEX(SYuGyCK&`-%8cM@gWP@SPJsNdY8#1L(d{r>
ze&xjFl>D`4thFN1l`l@(f2q~063=i@F;|KD^R<85da@6x&H@9GTD(uGAKI-flWHc<
zdw%fUKjG%)|IYRCA)kHn6Kb9~-aO~o<Ij11xaR8W8Hl=T$<(9vS$5zlK_Nw<7G*vj
ztkxTFett<va6CkwKbsIgm~j`Zy1LWm{NmC*SqIJay%09H@Pl`j0oHYOhZLTc9}1p`
zr1_*b=Sz&JUj$Wkv9+tZ>q{W2GH$H6dD;Ya+bt<tiK8H9$e7x0eS_8S(JXz-0L15U
z^7x_V_O%|u_N))b11;QQnmh&*wc-j2qIMAFGE=k?)5vBJcDpmS=a-yc++#cLXmPk&
zz3tlJNIuS%rS68u`WWmGu1-8a*lu>DVT0AXwWfkn)t<Naty7QP6*s{B!7gLj++x|g
zro<SH$au(vsCNjElXjkq!e*T)T7_yOW~Y0tt_oJL@*S-;-0kzE9ooxo;AA@y<47%7
zOVgr+FndPrs>-ZJM=L_!VneYp2x9Tl<478YMd_<G&<cY$muz(P>9kUUYCPG0ZC*=^
z%e^kfcAl-^wvWAym(t3FG(^GFFi*N@uH=}4C-SwjGp=uW4EEsNG@3#;<C(!Y;dP@S
z#o}`)*fR-joJ^6d-fl>OSD|ltbpink+eyQQ7)MlvF$fw8MXgrfZns10%zR^^P1Y;k
zLSLGZ88k7ZEyFk>aj;0uT$t0|6EZC1rX?MgV(#kV^-Eh@z1F=4DkWEPG&MSTSeqbT
za?6N$UG}A>f?GDdR%MnWX*<}y>Sb?gRgTBWk|>w)wr8A%0ko0vHRPwYI}E4oR`KpM
zuYg}p44}3E-$;@`O;FQdwSI0DVr{N78{CSwKnitHJVf-C8o0uWJU)AR;PXe1O%bXJ
zzD}(4haaeIW}T2oiz{jBiCYfMYPxe+<L_eszyADBi)XUWvDY{GS=8Lux1Gr^`@8CD
zi!6sg`%I@8wz(~CNvkK0*BjbjUs3Xb!~UA5S5Nrv)ergi5B~|XWYU;i<+DJ2+vr*x
zj>Wdrs_f3T9QHR1gEd5F7#W7dW@ACe^E@wZ+v<l@>layaUV_1EX*Hyb5YO|30B7fC
zT*$eXnm8~|6O03)1Sg|5A3`w2Am@pL9-WjXu5b3ln7F)mVKvW0mVU$IxUs7GW8m5R
zgdhFjJN(-29wEYzwhOtoF>O=egz%k^a`EM_ji8)9sJ17cs29up(c4e?)|_%nRhBn_
zb#A(AxZ-OdVm&GCRp#2XS+J;JRqe1!FqO>Va07Yvnq5AZv{2@2Zmz%J(PtmBpFd}s
zo>B7^SI-{u*~dTN+5U<z9(}fK<cpDH=Z^J64HowkvmFE34S*OUXXod{6us+(LQxCK
z_qE>CmFn#Z%!pZ)mlkt$YdmS47+wr;e%V@GTsSS3>SX==GU%T=A+_4$t=O|L&y!X4
zFEeQv*qxm-41=|sQmcVi=9%ozWpNX=IcE}r#WU7S&T4UWA`F`iW7<$^=IHTE(=;*r
ze6OAarnUw89%vnY?C_m|N=kw4IItT`(a&nN;O0DWoTg>X4dO9_+Uhr03|r2XG1`6|
z$B{Uk5jST@p3yu}#9X#iuen&8YMea@D=Z3K^2!8Y=1<YWTxMc5f)Q$fw6?!$T|TUM
zG>F<^ttc9WX_^3=qfG^?>I$?Qu8nXnle*2V_6YPuSMf4P!!V#JkV_$kqlLJ5ScOlB
zbst{fh3%aPrg30*{=l-pqa_tuc66W=>*L;%^I9fxsk#_P5@k8ULbP%yC74^a79(ND
zoXPXFxI5QVg5?WOP-_|MNnHJN4^0{@Y%8U(l(`Ag8Wyl9zhGe=mY5e?&21zmY7FEe
zmeA6|MciNtcDUHuuR;kX8i`b&3V}FSAK#cpV+Ott1c4!qMw-t}Ft{d=_tze)sg!I<
zd({<7Q{gwns7#6(|K3t?n&)M2Si(l4kci8^H%~KQ#8VOGnqdY_!IG#ja;{C%b{=7>
zpt+Ex8aXdg2_=&d;xJNTv@&HyDP%YvIKvWtN(fYeF^&u|T1hpvB#`D>Zr+4jd)g~X
zfWHI<(BZ-LmN`{DV<wqI4~wPm5*I!rurd}Eq!L?SL0$gjHgw?n>V}68A9HhkV4kfX
zVauM?MLp4F!Q0&P_^Niesb?@Loc8g&^I_ag5qOzP_LLr+x~N+7dgGv0qUH7<-u{6d
zfKFij(6pDW$s~$;%P328*E})L2j=4q$Nde@o_x+HpM1=B_dnu4+5QP7sVM_Z{`W+T
zaTwTc2ad-~O2THdAqZStoO5x$Wxu~+o(rWMJlNi@rPRuh5{e~WCRg$7x%;9Gfv`1;
z{&+lkODfpzHf%RrYOUmw+3mJWlc|Fx7i8R!;$W`kWuBN-8$z(!U}@&*(`OuyNA5qk
z=e;p)e2#~M7n4^0Q~fq?{iAz4_{s;Iou9eeyP~>w@2Fl__e-H(I7VCi(aG~_1nFcm
zX+Kw1gJSve_md{uzVzeV@~;&kLGYNPg#@<vzc$B&G>oJ%FxSHK=U2=}bN#MIKt|+1
zo}O}Z{W(*8Le1CIa?PWMKjJ&zzToNMiieLr1%V+B#L#Lg8d<fn3Mc=H#bH%d#?6N9
z**PgiZVoqAsv%njcZ_MhjutPZ)jfNr3TbGY=)#2ncly0#tu1j*y1IW)+xVT<oos98
zXEEA-7A!@2#hy{cjDwsr*Vi|^@#ZDF-MMAGA1z@0`udvd>uV0j1G$u?URt}vcE9Rn
zm=s&5_wT>W<>h^9RrZG)bh3C$%Z9JF#=-BjF+kf>bEZ-#^M(`#E_P=mcgfDK8nwEf
za8j87%P?FLQ81ThE)#LsvN?ai*}XT&wPn1|7T;Hl+=O6<WK*k4ZCQ)H2ifJ(;;^kN
zP-YI6q#_|%PwSRgM=NXMO32c2rG@F>#p!L{*OSjXt)u$F{(d?~SdE1c<FcpC+19_t
z;Gs-w_N-+KxA@aFxy63pVx>*}vv|R>7?bv)y4-`h!fOgQ&Z^dHiYa=RidE21lu|4l
zAeU^l@me*2X4Hp(6{TGFr$uRxfnllJYUfS>&zN32Pl%HoFK%RUsvb+-_Os~5dXzww
zjFz?90?Jy8by<eh=R1^Y>doK=QFW1~=&JnL9JCs>n-R)3b0}VuR+U^8E!hp#jfVxS
z9K+58Hw=TdU9<8~<`%Biv%O<Pvdy1Tz2QN1l~5`qzn{?d6s6RR;3a-MX-QQ%<rLZ~
zD5`riYPL3nIWryhlwnUkKO?J>w$>P99E`A<I_Trci}U)Iv9tY>vVdC_kGp&u89dp5
zmr128g_)bn)ZzeIT*RqDbMi4N2#I-C9zOh>r%#@8IK0C+jH~>9qQKa~s&2WjWUneq
zXM+~FY=7_)68KX__;hmtZmAD1xPt+fnu53g{^SR)49GfGO+9I!<wIMl7l17jhr>1d
z=TG_k(WiX)$q)G3<sb2Act!yR8ExZt%R86M;drn<g(+Ga8&$?Ja&d7%ih<qkob7h&
zLF=YAJbChD3D$083?P@pX55g5mMLgf^E6Gg_%Bq=E!GOVhJY8lVjQfXbR$=_&g^zO
zrfDM2nHJ-i=ZRXC?QTQ#jJkQ68P0Y#E-vk($`B*-H1oaheVg~*`vcy2=PT@Xmxu&P
zwvGCBfbExPv(36|`|cftMqYd_ThMl!-0Pv!n$Ner-9ET|oxJmjC{$apH7g0!<$dkF
zW3Y`PCUe89D#zo&JFc`jouO7rxn`Q4k$FxnMgSi_{)nkWo=*p!KYt2pg+*;stSia2
z{hgG;YM_)#N|D`e$7Zuxl++4Lc{X)n5HAwjqEwyOoeYHi+%{?}T)ZkEZR}q5b(iF|
zf0xbw?sIxYw3>hIx>~@$NVF_`@V=py<1}$N9J#)_W`Ec_A(y2HyrS+Z_2=&eo~^yx
z!1=``Z@u*%!*<Kf{u;^C9g@-7o|(^s<~mnXXGK^-6G9kFvGMgk7b_BeJRaA~TCFTT
zhQ-(fBrbcFq=D`EJudFQgQN|5bV8alYum@#9`5pdT4uXewRp8=$X1`zPHJ;0?h*8|
z5z=C)W8*_Kg{~=Gmds&$L2d3&R;4RQG$mTwIncr<Ud#+{9;23NIZYGOZ1J>qkG8H)
z*I;$kr^U@#$rGFa3J@tCA17{5*|E5WebDaLe(S0@%X%47wC+675B}z&EY(O-T~*-Q
z_tVNx`MBMIsgjo29n|9VLR0H$#;DzIE|!=hA+2W_LUcEC`@599-CL?l0c!QSkhXsn
zTQ~UFiN##1TBfvwNU0MQ$kpNu5uvKJo76@`Lxh+tOS%@z68B7RQ<Zf|zG?L)+nlj;
zxm6ml{i2GKNNhc7n`_P|>M~!}eZ-6s;}RJqdD$b2ue7o>QL7bVr&7v{<{1r<lFbO4
zjz{uzWQvKYWOh`}!zJTTnNuajVsZ4rO6e_0Vki6EF9WZx0J^&hCW0iZ>7$Vl2G0P#
z?E-ES(xOLBkA{^n8KG+`q5qu(hLAYU!sm~k^7-e_IP8yHoEK^|S5noM@X=gyzG3T$
zdUCR0bjciUF@P`J+?U?>atKw>JJrHph*i=?0q(p#cVFVyC*rQ-X3e#Y6I85m_B<as
z9<F)v_#vNs^dUcf_#^(o=6ifPJh74#8Yy|Emb`A1)dO{XSKV!`v`b8Zi;FWO516}^
z-Odk_Rat3-G{!_qky0y%!-4tvM2L~?b_=j<0LiV_C7HxY@{%Psw9G0a9^37fVH`ZR
z#bT#wUE{#0wkeuLZX3xsj&?uu>>G|yMR|1Zn&11qzr*|Qf51Qc&%QwjTPxpD?YWz=
zD}Khw`VgA&c;^|lL!uo<VM#Dql|ETW-AO!;E4D#amGtCgx3G3j@Z!}~&838*UcakN
zQ;|qa0}?HE4=he|c~*_sY9%rg^gz&=pq8++-#?}3$jw}k6W7*Klv~WTMvSIZ)wPY2
zA`HWZ-R{g1N8J4%V_aMMabhHKVi>ibpA7P;qIXK1OEp(#Ir&}}U9oN2>!B39b>Z@J
z{pAqA7}gBq>gwN;)fy7x#@wri!wu8XYSxufObwBhSX8goh<HYyZ|p@ax#`U}-{HNl
zeuMMNw>g%KWQ!A8dU1zfuA;imr3C`gvS4Fkb8(-G2X8@$T<`Y`7_m7Vt;<J@7UofF
zv{<IHzMrZk<^^wKc6RoF`wzZC8qdg8SGQhmC0e{IY*jClXwNa{?9bX;s0Du>ZQY=>
z)@&MEY#S$Yr#VNldV8M4)$F}hfxlgypebM>ge4Za^(4P*NUZJGY;DzAnR8*9>(XYf
z`u-%el1pCX-s}1)pvm3JV&to+Qp*fgt>>;5L|dsR6HRJ6J5nT+=y9E%fH7D+SWG8z
ze{N8jtF=(HEVU4uaZ>79xEc?sTApc}N70f&Xh~Ztiz|E)PO^G3gIZEh67PoM)f8Mr
z3E~M$Eo)sP*<(yI%^M+TNUpZhp6EjCtS)TEdy80{_&fnZ7`8~-kV`WbGPxM(3jr&6
zlmZ%rtkuY*SeK<-ipQW^-{(^!ppA(UytcpSa(yD=VJ#J6S~Bw2!VrFjsF3TLp&L^q
z)w(?YmbtEFrj|sh$~0RG)Z#I)`9_&-!MQ7ZAVq5_*s1|6#^<s&)_Zsryt)FYfpMI;
zyf^U1{WIp{!1JeG;Iw77flUyXU6(n45=Tonx0XARzy`C2W~`_upaUY2tOp2z=g*bT
zKHYPDleu3GgkkXX>|t>+oJK=5(Ls<k6W9A38VO`|*S71m$9Zd0S#J4Mf$-M~Cx?v}
z1YQDN?XT_(++7Frnqeft+#SK=?$EfzQYOvF(F$VTJo}taKKudy@cAF{_jccAt{EvN
z(`T>8C=#gdlE;b2W*EJ-js>?`X^R;VxW0bQy?ghNIGSZ06Cn=dTu8~Tp;bMXTA6hw
z4I`GoG121LL;{DyfjrNyZkc5s7q>CXMjBIM^x~$+<Ahc<w{!>`4ma*bF*mMwvPhZj
z{wzjBI37Jd(QS<SpiqAFh=2H>{Vjg$w|<j*58gvHTIQ7oD4ANbTko*g;OdiQA$;M4
zxC*WFOiS07wh>X0V70JjlL(-7Akpf*Ypp<Qn@OROK*hSCG%B;y-D{-Z3?Lz)L?97{
z4Mn`N|7>oZxfE`0ZkXoD=4*|FI1+S*=_!%QJV&A==6pq*H-tGMWdbQkv=|&>w3t1t
zST6!qGBM)WYFpBH&M<DVI5T04ks$<%&dfPb*_hg^9_QC$p48)~bS=x#l#`~$H#Jp*
zQiBx&*I?w_t4^CDQ=w!a4p@CU>lT5#wX6W|z^fv8Cj?kk%B2%U^wLZ2mMVFsl$l&?
zE~X(tspNSwITcT~Xb)d4>FTYwzQWf&_y(5`-e#`J;&q)IwQB!u4OuO{=E2Z~P_{BH
z!|sfGZ@$gNy$4iP_QwO0+96wNwy*}8`z}%?u7=C99ogo!+nsa&{#%UWc8Tw5Ez;_V
zOWKS>UC4^5*iIhAx<Iv3Q8t%7ZtBE<YX4^<VDQUmBt}=IGqS8b+neKRbn9AL#b3Vu
z?+LkV`}vZ!+YbC`rW7UT$~4cG?9@tOg=n1}JjOXiPim0W73_mYElQq`q)=@QsqX_?
zJ;Qcb$WK!ir`akS48!21v&^sq`%J+~GHI#H`<XquhRsx6Rd<nK-LZV%H)XWd&9iaQ
zYCB)3-gTuB%u*?2JDX!5B=xLncX?tJ7*q=tA>xLTm@->xg&b_X1b=8Au4QGzY#vlq
zq?(rpCu!3>E%CLQGij{kBIHsme4%)}WeCVP5Qb#kH6)mFs1|cOP1b-QS63fh1X8SL
zd9JMmT>G~r;kXb`*SXTxe5<eC43EWjaM31U=S0hXF7Ec#Y`utQ0*AS9^ePHf13Cq2
z*qgB(lhv`MFor->ti_!R59$+HwWTxu856*>dzo+k>RbHEuYAC6`!)|hKJdfoA;)7z
zLR}cMSq#Bhe_33K_wQY>+g$Mc>4A9)SQn6h^}4f0)Alej=KcsA4>OM*KIZAuE8cv2
z%doM-@vd9;rLdERTYmbNyk22R61koAwOXF1IQ`k(n6Q^$#*05+ihBC*<~e&kLX(`e
zDp%Lf$j2L=JbuW>AODCy*nfw=8~&In8~L&Z+jdCK9+YdvnRlkn4hhw*W#1*s2n!){
zv)^-ewj<}l^XJbwKR;(2M@msr5>nEotjp=x?RHxf$n)$9qHWsKG;usy@cJ;AGTd6`
zG?KO5ZfRwM5WECUwXC_|HKLS7NiafG8HR+2bx*jtx%MpI$TS}@_h{hh{h2q0iEsbO
zAM^YF;6LSG`j_AHBFwAXrKzjN*X(dUc^apW;$%{^xDK4yLMzxz@G?iP0M=R-Lb|BQ
zb)6B-dT&Hvof{QItx<b*qeT?2=V0-8!3$moJcebUs&YIYIUXjnK}5*466;Kg7H4KP
z=uAyJOxKh=P|8Gz$`Uh#E_1P!8W6iZ6eXpBd-onNj$1-VXkBYQTFox2@i+{dg_NkC
zDmM0pr%4`*8spEUqEt1dBm`^kS8G!|S^&A)1vcZ&l^>~q7a$Jt)jNTV!-&K{RV#iS
zW26SC-UUL7GTrP^Of9z>iP_p|1uGR187x6VJ-*UiXek8VeDiJId+%#pTs%ODl$t4d
zVm{t5w}Q4<jF!f@5JL6VY<^x5C6LA~=a+AAwmV-^I&+>l=Gg=HTMaQ>jS7$kBX4Dm
zRjU-Xy9+MwKVaOBo_SlLLO$FuPZQH|T9e)UK(~F(NrZr<Vv1)q+iQZXo}gPMv~9ZJ
zae!+W3|Ksx?{9K)Y&G@CRVO=yEs^Ck5oldrjX2hmk2fPlv8?AjXXdF|9o1YH1<(#6
z_acZFGWM=7cDA+iq>}SQo|PQcN}RZBLBztzVzNYnQbF8Mw^%=jabOseT_*}&j;EE|
z2`8aarc|xZbx9&=35W`^#PiBZ1Q#vey$Y`6Dp(Aum-NX&AVy*pN~{dYT%Jm`8hj}c
zVp=kYabnams+VvSiIBFCMyogKLIEXP&DK&8s>Qfdm%V<Tj+9&}HB!`wVxFzj)41C)
zp6&ekgX(NCksfDWvl~0n$vc*e)|9=u7Upb9;SySa8C=yDm2eW1-p18y3WE{9C??Q^
zVBK3{tb{CHF3p%^KJLl2FcCRsWzMdCM@53Q%bSh|_QySOpNR3CvotWPP=i<_1#oYT
z`g!mo5BYiV$^`JA|1bYKzx_Mk<ox`c=TC)aPmWZb7)D{rCl|MnsNjc2WxE}D_p5L7
z=KZ($?2|9}$&VlVYPHuColpwnXahV=0jf!=pMUX$r%$e!rU$Ht^_LIE-N1ZK+yP(u
zH;d(XqWGO0Vy8cEYS!szoxc8MxTPv$fuc6?n>zs_LY-%?R6bMkk&iz5A%FDYKj!bJ
zKjGv0h#4>Z>x((h>!eO`V6)jQwT49S`g(zJNaUQ^?zYtDRvT~_26E0ke}3gn*~jH@
zZZSJCdf!fO=B>4Q-9IZCpbF>b=d`BeEqgdcZ_Mp((IqPjxAeiwc+5<b|IIL1*^639
zF_FEET+1>x7n5*#?~>iwj{VIwwH79XYndq5%J2V!f57kjQ-7H^-+rIch+gm@=eoAa
zYekn^OSkqc*Fr*G3$7UgO{bF<SF+Bv_IwwzLnDi)MDOL_uak<UbfLOR*$&$nTQ>kZ
z6bCP&T|EQM#1OGUtzN1?QX&ikg`*Xd4FN9;OG`wknn9`;##LgpIG*4kUJ@hcmzSKK
zT{tPKp1@#ni2|1VGfkv4E-n6=^?e#5(Fo(+dtX8_mUYtVhpH)iOF{1S=cntb#o)F$
zpI5~&SaRCnWsAn`mMluCrZnVv;xHd6HCu+X7ITTGxRfg~R}&zSnr$5ggnO6wdFSn~
zaDINDT7;U991eS~o<HIFvnNdZ8&l#PF$P8%td&(M%U&ZuN(1NT=bY`ffLa~3q8#R#
zNh<^k$rw^v#w(;`uKo<nRt`qhVrkPbF3+fxqZL!mnVak99FI5TJbT~UxV&G31=HL5
z7cwDRLFnZ1T5Zkfowas46_#G&>Ibj5i0b5^#h$i9+TvAj2g<kiUFVXxF%y<h6m9W&
z{<jL$nk^Z?$-r7?kCT#4{$A2*-?lSj4gA*{mh4$0rjZ!%8hnvTD?JlE3phCOP~TH5
zaSC0_>@4?my075hxRKqubr6k=TiwK9Mt@+5_iL)ER&zh!+Us++`1ue?L$&_HwIb2_
zZii&ae(HV0o9f6(F=JH-Awf!naf1w7QXKp|o0w~*)ETK|DQw-0$v`D)r8=1|ie=Vo
zp~gtqoDqhdookvQw@mP|D9hnwelDRUdKwqH)=+H9T8kT9;#aXVp}sJWR=2+?gr<B-
zSz}QVH!SK*nUq4J7K^Ves%W+HG?Fti&!n3@K_Z(JDZ>`hYQ&feU2J}-86kh>1n{r?
zYk!G%-o3};FRu9X<4^ePv(K2O*{zqgn5|vDse9J+JfoVqxZLvo*WTpfV&r(7_~OwK
zi;<q$odq7ean9~+!#pWZzOX)|wPc<>x#IKBpKx>aHV+<L;6mX~#jTi+m#BZY11n!t
z_}Y8?G!l3k$Mmz3KN}r$o2Z!Owrp}9BN+qS9IyDncmIeVe)fI-Kk@f@n7^>#eBVgh
zq%XxLYl>FFWk{(}WKxQppP!*xt-OL-P;nXt>)z2~HEMMQuPlMIBH}^KpqiGLjJAM=
zp-rc@d6~ix!-jDjupn;B023{%5~jLj@x`z<!rtz7Ow*CgX2X7e@Qu{6ORWHE^%_>z
z{An|8sHL*s-%v{C(N`nCxxeI-kAK3qzx_x23vYeJ7in+;XuV!)ZSIWr+)fW78`ou%
zRkQwXcdTgS&f~mJ?H*Wy-<z9hnJ>B?elN@oY4L_ENmK22W^wzotEP?&X<)ZIV;l#I
zy9|M<!dxRG1KYFvRFr+bww{?18Mc=wfjAG;l9}g;QcY=S28e{#F(D8_;qvkx-}sep
za{vB)RIN;i6#@^W!9M$DzhCO6t%UF8B=@vvbX!9<Zk~<kYO39PT<d!J8aZ7?cjIqW
z)!k^Xh7@gnB^emETjDr!oDUplg}IocU$YZvJ5YkFe-(EPX~24=x2h<C?RLu>_ugQ$
zJ-0fJp)wtBxO)1O$B#ef`SmljWCo8xDuTb4zb{_r)DN4@xMdthP?)Eqb^Fl5z7&pH
zAq0z8^zIHJC2M!7N=O4DQ2h|Kj9v>RDTf>8siGy5=Yy3TYVq1M0@M!emB2R@&I-wA
ziV7MN5(b*u-jrOMA1xK)#r?Lqzwvl$i?Q+w1){!p1b1`Z_ST&4$)`U{P7HD=RL?XP
ztd-#{H*u|2h`U(KW6s55-$Dy%(xnBTY4k#DS@T$}!6>G{wIbhbe1nQ*K+9;_yRwp)
zZ0eY1|FU9wrss*$`cj|9*9KKm3Qp!g+?5Mu6%}N4kq0*{o62eNpDip3tGS1{gf&??
zoEmIj51U~zFwNEL9izNERaF-N1Be?ftrQFh)$0bg4lpfsrNt={Ak<<ZHEPOio-+~?
zn~QtIvvWw6oG3-8WkNCcYwH(%ni%%d#0gQ9vdleOPfeG$GublLVMqZ@t8wXXaFU?W
zVk4Wo-D5IkHqxyYK`E-#DxT?V6+%J{gp#eM<1~|}H32EWN%paOJk-jD-N`Nex$(-o
z<2Szf9x+xv`sj1M_uUV9^zb<~Z%p3RM35GPBUr^Mi<bdmn)g(lxc|n;TW`NXE_?p)
z4}Qe;^^x=Q4PXDp+r0m^ci3zOKK<kgH`kBY?=xBh&!1a4nCH*;91oe@X5d!I!S<<l
z7C*4C{ZqVfIQ_Iq^Ch_JDt?)p`1I#5D1dnIycgFFcssCKxw(GE55NB>{LQC-kAFOT
zpNWYv4rtA8VYE&hu3$+-k}-!;vb$6PRmO2-o(pq!*XcAdPu^>>m}-+;9W>=Gu8Z3@
z59`&NEiNQk0clk{dv~=;n}wuxC3&rmS&O4^?}%N0_6%iDE+|z!Bg~q2?{AKNh*~m8
z&L-zyrj4r&0aa5+>L|=x$X8eV@gM#Hzw_sRmj`dWYs5=}wH%t;*sLz#^*O;x(A;Ri
zl(lf$Qy5u#KekQw%jV9lIk6CxyK}-_Gq;rO#s;m=r-+p(5O)j3nAmJKjN^`!Mt3VE
zV!Ytq{jc&HzxKQ2n)%)j13&rn$l3WN@4feHlqvGzhmU#k<l3t5*XjY!?RrcBj42RD
z;ogI{_~tji$;G|<7LN(IAKDAcCL<so<8tagxw{rl-M-M`z2rm{JQ-hCf=+*S%AHTg
zy1mEU*xg?gF(%SxB#k5EW<$;`Abs|-MeA!Dp)JL0iAn1=fF1a*(o|K>cW0z9G94#0
zD^(|Mo;~H+)2BSUdS)?UwNOK{Si-p09$QC2oQxRts8yHJN~XP4ia_y$Evolw7GaFR
zg8FmoauF<4D_C6PoHNg^uZUub(tMmK<$#t<is_aSy{K^5e6_gURh^{T_0*El*r+{=
z2f&Plc<Vizg5yus_KMJAx_xa_>nPE(XIJB?sljXZq`IXh*v}HJziy!vk2NKD@>(R;
zrf&P*q-z36OEQVA;Izla3AqVzi@PoegxJVmaFW=pHzy9RKsH4?N}yt0n8?M>OdreB
zq{C(II{^z=VwkuPAhpHZR*%Q@9n9DbV2L1cF<$L&jEVYKI}vQ>3o#HjiCO}AHgaA=
zpwRBmV(Qxd9ZF?P8_U?$?8#i-0@ULTBXx~)6C*r9qq|})ajNZ+<K~Rb#eHP6wevC0
zriRhB2m$qH?4GHV64SI2U!MrJzEeVVHnP=vnyXm2j#}BIAeA`<Qc%V*y`cW#pS>o-
z6@prgP}}Ft+kzxdO7eQBBGx~-R+5WTA((Mu`=b|EXN@<v7<5@Hn7jgBnE;N%$kV4!
z_|XqO=EEO8;&8Ji$d(u?DIPo%xNTJ}rCYs<cV?Py7`Mth?_5Cm3Ple*e|E+DAAFT>
ze)D~HyA98u-|&+kf66?0V6+OyW9H$fpY#0L4aZ|<w`mI8?R0He57+hZeYuPFHnU&<
z`)(%TU5jGbJntMhU-rIt4=3@>*?KXa1UsYDJX7+XKm5@j@OM7{J^oSsE=1e(s}reK
z*9-L?#xYq50c&b#S(&q0Hx?gZ2ZRxaz;v9LX6q}8no6izNkc*~SAaFH-<XvY!qRa9
zH3coX`qwh&nv10t%gnNg?HRm&Naj+=^UN?>Qp!Bdgdo(pGEFx5CwCuD(@f5$zE$S}
zvuAe=W26*i95x&eGcgS8_eY+jYd)DPm(RC6`s8DN@}nQ{#)CH)(m+<2OEGKQNt<db
z&e~iUEb1Sq$M3Cy>bg)}t2c=>6%qT7_Kd5gR8TA7#27h=20S54%j`LstNwQWv&H=@
ztp~4oY}!C_x4a$5i69#;FW%zacfL+VdHVbbpFR47dlzr>m3O~R&57sFzsIwukFYo^
zX_xLQuDekrGLDJ&-v5C2KlnNq7nf|dTS}b?-juy{8c?-Z#Cgtax3+P$EW9sUOQ-*B
zhwP#So&I)Jgj#mAzvo>-`7*=r7exqx&1S=}*&-o$d#a`s`&FtLN7czi@i}ib13hC@
zSK_cqiE*<fYhk{AZUvgl#It9Q*<as~=gEnO#WN{Ns20Y7dX}#0qKLJ?`+1%SwNgaL
z6^bY&Metg3Rh8pB8yOjrDWuT~3=7d}Yk~#ZGo#rO1xm552txKXV{v6M#>MkctH+vQ
z8N@+ty|%3Iw&r6UETJfRa+k|ZErx6j&9U+#!DE_T^QF0B5!S?%HV5QnUk;vtBu>m}
z<$T1}Vb!HOi>Y`*z^WjOV`46a90Ji}=+?~i!VnYV**W9c*+TwoQ?M9!yTZ7T)f56T
zq;-F3`$Vu9G6GZ;U(?l<9kpbVZ1I=U?pZx^H=G!W{#^=zj6ycm3|GP=hNXp_`n(J@
z7rOem`I&>QKXJl}hwQW><EjCTVjX--HAUK!+j59R#5>=ZabEL82{(kSRLS63x}_Fs
zp3MasFeOaX_MxV1>e}EzVq!=m+sk{%u%$|(#dOzFt^1BP;*WPK5f6WAYOPlo@emA4
zl+>jXg*t~&wRk}zy_Fn<ITyxc2Hg<jYD|E8?N$LG1PIkgZT3z%P^prQFe_ecIv6o7
z)t-M!q+zt{iqXS#@Psz2eqmu3R{ki&S2J*5nE)QA!n0?4zWbdIdHnba(Q{%L$@Oq@
zKpJZeAtBWcix_OfI~=b~1-!rG{QR7Y^9^&JdGoCYTwY#qeKqmXN1yY_Ctq+lSXqQJ
z+r)qP@G*}cKeM3si>AEXX2mac@rnhDpQ;b1Zr%18xJ}aTzRzha)2S=#PgQ<i_TONc
zj(6$}Q7hm1;=BBxKmB)jR<D-QDz(haWzAy7U1dQmxhAE!bjujTdX82Rt4Y-2SroJe
z3-H$JwahXW4`)hH43@Of)IfC$#VlXj^wNMWv=kY}5hFBCR=ibQt#)nE-Dwi680q0~
z;BYuHj8-=2u)nc>w$rrCos<T0u2^fcN-hwBxou;z`<;#lPYzjQO>-_hytqNXci`iX
zKID7f{v&?<*MF0=Ij7*+Xr<EBg=T%X=hHUpwz+FqqC#msM_;aPnet$T$yXOtYl*hZ
z5p8qf<a248iY_E<t(E6x28P5ijKpCeYqr*3!LrAOG!o;$qL8U_*x&H^qld&4dG_?N
z*HyF(aIKXxD>00OFc?oDpSb*!$FdFVb~}FMo8RIqU;PT_7w5<@P;y0!?RuuB#1jHI
zFN)%eW3;ZJmGsqHKWju<yd|2srduLRTdynmIDK|4cI{=K{qtgVCAG35g<3Op&P!aD
zh!Gksx|YK!mUUZMYmXtqX54yQQsg)v$>qRtnwj#E{o%$-`BXoV*ZTvOwL3cDNoi!W
zIYX-@Zp_85%QCqIje<5;p@)2wYVNM1DuamCF>@m}vMLMHG*dKN?ni4}AFOqm(n`8C
zm%PV8`<h5Gk$exG=4ojiSX@<$!MZjCC!D5kwYAnl0W5fbvl*73wfHoDFt|hX^kJEL
zBrAzX$yCB5!81_3fOM|tY(19MyD*6uX&iDP&Dj$O!b&hPC3kav#`gYwhK*&;hiWcc
ztzxb@Rf`?;MQ%M~fzZ4MP5{*nIwuM(q2rX;`AAsIbGuJ3P+FU}KrFVpk@3})-c;f>
zHuBWXE^g%6MS@gT!d=BwXTi>gra;z8nXRN)oh>mdwN{`3D-^Cu_Aw~AGUXdc8I|Oo
z11o_fSW=95EM^6wsUI$K*iD2Ot?#&(NhxYY=VzUfJX2?LnI^yKkj$u2EvDxBnKiqY
zNP?-JMJxm&Mawg2wQFltva+;XY({BQd)pD;45l`Bl}Z%Qz*2v>R7-F%_1deWn81i?
zrG<9JVPF_XhSBz1i6=$3X+qUnd$R3IuYy-5fVEbRhlzQziD1g|jOr2PNDGK<L6Sit
zLZH-{%>ct-lXD!;xPSkQ%gb|Wo!IU+>~Cg%{P`8%|L!Bc_-rC)O9Ze55_$Uc8IK-)
z!Sm-=Jh*?)gK1y%1HVkwgEedM^fjjk@!h|_tJ=X@`sGhk8t$qSr`MvY=Gw1H6)PVi
z)#@63I(^2!`IEoKv+|r6gdwHH{n}hj&C2BD`m98Q>r){hb+xReN!N|wc;V6!^(yEs
z9;~q0>=;rai6!pDw5nvPrVNObMWt$FW*HAvj}x>_Fz3QBj%-FN;C*@TKHKfg&Hl#7
zy{jFnYPF%-GZIq+YRz~$6vLPZ0k*q=5QXD$kE*rCODS-iDu**G!ScyRAMw5KeV07X
z$WT3PKb#P;b;GnprHcX>C|F6CbrG%wqV;65oX(Bg>Tpv_@4Ef(&XHx?mpe*_Z|W&p
z$()o-?W01jv$;Fk?_#tAf{Jk9+0#$?!$15zLJT~;{uJsB&z^qD4}bW_%m?9%FTOxE
zFr+OhDy3Yb)!n`!nhP)l?!WN{U;oB8xOe}7Z9s1L6lrCb@Xb-2*dK1G@-K_WI!)L(
zRpb5Cd8y%b`gf<bC71eaCnWpDmR&zTTD)iRG8EH1F?nC>db0k^wb(w}Hg{cWj8#Gy
zI6J$~c6(vHaErC_n&z2lIx^?U4n|qm7uch$G*aa6UtqI6<K7!@GmJZKZVt@HEAJLE
z8_5eohy#R^L*J91RExD_oF=L&+n5-KKyVc?M)7Rqytr8x!&%hJ;pk%jD6Wt@Q4T3B
z1<DnrmLskBan*vh1FWgWn2S}^-cy1Q+@MV30IF^j+V5g3^1K-FE2*(ol17R|l_(=|
zu%wG1;+-L8M9UhJTM9+iBL1yEy#xVIgaRxvZnx$9tv4AjF8q)$AWF?F*|R~60F*^x
z^z|Xp)FBm%_YR5R{b+0U9?0P|e8=3WR!6S*s<7C!pmM9qu7cMA^n&hfYf`nC=SFb-
z|CWtQgd~>CSA(Dgt2bL)0t`?p5ie&0R)^3cxnK@JK!aksV07YYs%lZJsP=3sml+$v
zX1iq^tp9k3&0w|bmMT_QIz$UgATW@SXew%*CyIh-M&|=%nxJIj5Urh{*9IQyKpqBC
z3f^nm=2}p@?<gyo6Jx(VnBGov*#w#k*18-un$bqm#=m(v><mdUk_+a+#>ip}LQ)|q
zzQ#O>Z?oa-?1J6yjA7hZn1~NX&6zn&q%aa>SU5)rVrRfB;FSsBb~AAQ{w06mcm5pD
z9w#1ss+9SN#O!4SR?-us1!S5Z{c9h*%Qt@I1J2LS2{Mx6hKn~ZDD}W`y5Z(}&u5=L
z<)aUuaJaDtr4>&j&+Kn*`1F&9JbwI?ciw)Jv$OL*m58l}^Ge`ebid(J%!`EY?!T6K
zxfTb#n}N2xU2~DVaL_p_M*hUR1gHRo|I1JQDxW<2)LjrK8B2aR#$e?!Oft4)k9G~4
zjdcQOS-SH)dH0B53J_-5$7Jz!X=v)rfvH+8CXrxn+F@|Z5YISW-OH`*pQ<U1R>oy@
z6X#rLq-nDmEvR~$*>1Pwd1iDg`s(_*U#GQ?Gs0IrBRGS)ODNibQ)?k6E5HkzwN6z~
zQn0cC8Y+ixM&5pO#^Wy@vcG;#+ME$LBSpQiN;@=89{xgPQ8R*NO=M72@0zia#C3i_
zv(oC)E^0kknhSX8w(*jAf|%tTS~PNF*J4fDV@kG}Ew0|C$OvgjOPwv-zz$6FGjnOw
zN-0;QA#*%D=F^WWc}^S;2Urx7_MTxe*zg*aZ@&2!Z@=>{=VupoC`U+|&@>WczmTpL
z12auW&p3biHgjqSow|o>UHT%>+$!q<zY=wAnR+LL?^c0dIN6?makSDlMJH;h9H+yg
za4J@^=K!QYTfg-AwQFeCopYw5+&_DRx8C{+NFpDP<m=~_)t)DEnOyb5-Y>9ZQ?@;`
z#b~LT%kJ{Q8@&D1uXC6(S)WmLvJy0^c3o<&Hq_LX5H0Zl1aYORW#W&!bGCzJ;M-?r
zN{MdJvk1zrm!rXh`2AT}{F%72XQZxm3|f1Ho69Q1mK5cOqEGYM_7lv@F>W_D&%&CG
z*`BG9DOpM{wX16}tKuCv)B?(|QbML~idu5eRM!1?-PeQ1j75gQ{++XXT}EZx?l`;u
zfZgRi>w{Y?(^qq)79+-%<0GIJ640`tiOFK*L`X3Z(}tJ^RNy#IkgYtLgg`=;I)Dny
z`@+&Zq5AzsuR?I5h*zDxjESv}Rui!8bK>NyEsLeVv8vV8Q}ygpbH`T2J8hT%pk6h>
z5<HCTlv<JMX-Pg;Ei2dvPOg=?T8EFOVy0w0uf;t{t$Kv0Lh*P$KZ9GQdck6iE1nIU
z3tDH&bfC_Yk;*WTHd|5>3%!v@oMtZv<@=kqvOboaW8o`c-Nqo-)}*4XXTM*+A5|>I
z+SNI<2|<>m<2KAHwAec#RH7(JEEsMOQ`|S(EoWz!jGJ>p9KE)rEjU{@-gPGlnxZTE
zZdB;2?QgG40EZO#%2(dx&;Nyq>#H07(eHo6CqKSI%8ppyBFJPDLp=pV&E;@;KJvA%
zzr%Z9dy~x$j+2!g+3n7WL70y>?DrFo9(}>nC)Wrg=uD8=7Japa`S7zV9)EGe;W#sH
zk8n~CuvI6AQ|q~|?mKDLNLbyXjhy*GZmt{qzL5ZNrAq{~cr1(`oIzhy16$@^YsqA-
z%bMBY<brB_|CR?DsAQfUo-xly#$>7#?wT#dCf!`JEkN-MS+#fuB=~>Z0&P-ph_*4;
z>cvw%_Mrr`G-Fr~RC5PgG1pu&buJ7kahwiT24)xtVnvZlnHJY%kkwsMsxl5EbDmKx
z%+thnyXAP?Q%h#E-Euq}(5h_54Q0-3h7Dyd><@c;53gs)`pEnN*k%~VxFn=hwF25y
zWwRMM?hnXi;Cvc+<Bk6>asSeD*_K_2VPmef_TJ}wk9gci-prd%l9eQrSrm)qvQ&~?
zl1dh;VWS|c1gRtlkOTuV<c{r*8~z1uxaW>xz&$r?z-|}@3^xSWDod8(2UIE*t5{@_
zOfs2|n~xj!5%GQJ?7i2T(~U9b+WUO*-App$8r+$2BjUvQ&e@Ok7<0@q$Gpbjm?6T}
zQ|czV*gemP55$4pp6%rKi;o5KakXQc{Bz=c52XSju-@8cZAbsPBxvr;ALqf(@{Ry>
zV@48SKtn|Zght6)1XX}0fP&0Z6Gagkx=KorH9;4^P@#B*ECq(Me^~=ikusRa5oHx{
zrkd&I_!NiZJCNxbd0-l}sGwHYKqcBnGyrQ;%xlBi6jLhwBBQgy9{omoJlF~g=FS;`
z9`K|cS!<nt-R4;eFE$4Wwy}3_a;(b=6G2sHWY(D;ajij;!O|lcfZl-tNK$ZpeT{d&
z@I_qTKEtwb9xx5$LT^xyOB402(1BeIw(%1dB*&Xuy!XYgV7PjM#a^LIBl1$Zr)KOz
zlBgp}xpa0mOy$iQQqC+Mnns#ko54ow-CaU!8EMAuIR|RC!gx4I!oitoKzf0Y2(uPC
zO>Sqyi~|`%g4Ea~HNcYuSj_}~+obYm{($Y<F~tEaDPLfkJ>^2NeQVs@^|DRaUJo8t
zFey8b&4QsMBpYxzUSqtz!f?Do$py^>87!7r*GlUxFKE!5U~YN30Yet)NMUh883r<1
zFi4F`Ub3PB29R_)7}4?yTWXiq(1MAr$Iu!02oMO`gWZ(fz3M^ZZqAgvtKN3jQk5Ns
z6zctO6P8;095nMnyd8k%*b6bAYvdZ)V{?zcb!~vPsIefp5<XK&5)lR<TZMa;gq(m@
zso7k-zwddTp>>97L$U@D@1dRxiubqIrLu{^c!yH%A>L>pn9vBUAzM%pduR|lO%({u
zyT5M>@WjaoMH1ay3Vci^bSS~9W&w%_#ykLdLK!CZR6kyMpuEW{TC1oA9}gr~TH!OG
z?liw~hU1L^;H$ecuCEVx{?;vi_1C|Q)&%qUd-%x*6~F;Gjs8xi+%ou3OD;$`(WG=e
zVP0Ng949Qx0l5HWNH|;-y!Flv-hTH6-}{qKFeE+zZLSxxf>)m|_~^we+~1#Yb33pz
zXDlj0Z@2KC12$Ht7mMsRl6bKDgOegQciXN-0{riE8NlXE_u6Op>pNuMqXXM)Mg#;l
zA_`c;JfFF6me8ni!3uT^U^Hc>Nd8(|E2fdJvEe$cTzIY7zmJQyl)-O70#G3W424z_
z%Q~avjKM7;*3~@(qtbJ!xW2xoUCX-S{_YN{hG7^ni1(_^n}3$+<}*0tLYtY^FpU$_
zX{xlbP)6uvBiC($DmS%~3kyK17e*5AVxDJT{FEliexF}FpK<@<j8}J`;N{&X_`>a5
z5SY91KdhvR@eq8+>(4mev-b<!4(YF5rh*-qC)q^nj}=7p1w_1ih}&s^YzKs%eq6Vj
zpJQ{R2uAO+rROgCWSbxfEAyG!1`Lh!gt0G~BZQclzo!HgWL6jF0~GA|p}N582wM$m
z4b5twBTHSex{X0>7-S(ICcb^#dk_}gK^`UU`mNplyZK9<dGI-vD6WP)_CbHXV`&vy
z#fvN`T4xsq`m3>vQ=}Gw4ZRDHfbn>Zx8M0Po;-b*_gPqZZhkF-gmF0fMpDt5Kyy@M
zZs#w|E20bouAe=}FkS&oF(kp_8Zb#2mdVzHIj?X*MJ@&6Z6TAV1cxM;q6o`_^w6%E
zftssy**}}ru2E3n3}GIS$195MNr`xhQy{9yLvfoFvhyXe`-oceXB4%uo-&R#v5|{W
zYaN&%=A$#Ek<#oez6Wzjm_|0Y&WJW@wKg=D(L~YOHcr$*dlXh>Ort;!8ROv^GF@S9
zOpB^koR%5q(;01<p>;v=K&3T9Nm0bXHMpee>G(oR6rbyS%o(bStKhu2M!@%r3sFSS
z{2mnt`au?VX4s#gKs_jPrp~pD2+f*oy1<kRXsgdaJ9X1}GO{8xl%EFf6=Vc?Uqwjh
zBnX(RWuRcbeyzGPjcW>9DF<q;0wSpEifI_onqf}BkZ4bY<l{4CSajp64JIp=y5PL9
zV6gTA5@E_WIm4X@n#)k2L&MM(l+yue7?4Uqvenn1gn~pJp+xFj029>OuvP{PO%-hg
z?w1AE%4P{;F8&#<FG4D|F|9$Fdl8*NTtjKdLq-~|a6H_iTwO!*z!97bW-Hbe7$jqX
zK!%JGU1X4Zo7Dj{h<?-KjRD~A{8xX7r%$gjP6Kj2;N5rM#aF+kIN$#mFFrg2Qu}+k
z_+)5Hc=d9|55D&k++5$_c$jb;6OPAgG`+`=2TTXU?UMoD_@%dT_j1Ace8ML`xq~Si
zeY0_M!mC%W@X?D;aDR6~Yfsz`VFOqIE}}f1$Llw4uS?n%JM^D_>}MYXdXF9P{Xegn
z^TSt?B<fy-(LgDju(eemU~HKt0IJq*YDynCulTJM%fg<(Oc|3Qk_fg$=^N-2ew^np
z&|fJUtwRzH)3Iw(N*Q{p@BZ!{)8T;Y>l>WUXDstVt(|yJ%Up0g9C4mcZ2X=WO_~`_
z_b05iBBhMFuE+@(#*91*4u^^D<=_HauXw7pVHgTZ5nNx7ZZysIXWjSB4C_k$O&JQ#
z&w!tP_#yt}PyPh&fB7$=1u%0qZ~r*<te<s+vHy2x{&wb|+w;5}XYsNbh@{k0oz9wU
z0C78?2r&BVejX`61&L2oP$y!5gFZ5L>|MJp^j@aMjnNl;j_KTA3+8}(&OWM-w2`NT
zo!M#HE?ATvl!44Z+=gHsfUU?XDuVks{|p`NLtXA?9l@~fQ|<9HpNA2&nz#RQO&`1k
zSJzSby^72nJ4Ga>)G*@a_8G2jpScg3qN%Y1fl^@3`zp?^6ckl<251f14AR#xaTE3t
zk_+Z#g(bsU7c9;WH8ZaU=krdvK;3k>J2E)l$%&RH09h(!WZGMug`+^|j10^QNN1J^
z@N&tWH7j|9Dyv|%@i}B?pNlZvTf9<K;C_Sd85`pZ@QT}7S(K2SnbfT~Eo2IOeVc>9
zM2ST8q`98WdaJ&$$ywcCKLUvTID+tAAEpD+74JbR6INuXu2|}dd7km=?hf~-6V_!x
z5yNpTl-)QxYU;%+t)aFsnkLL+FJDg`d+{AqQG3v|b-ZbDo+(&RADdum(JaUfCo-G`
zHEdP$!N?{r<dI-n!-~e4x^BnA(sm>>B;A0rNrnn7O&FJEuVf|&46SrCgJkc4tO?98
z@I2Rb#r^#~auOWI0WGo6CFcQo7*IyGk`y!OvU1G&d|Bo*mL>GJf`W`R<corx@c&CW
zW0(#Y#v{hV5#>Bz7^q`M0GJ7S?{Rmppw5@qxDE}q@tH<&7X?42h79U%eO2m(L&3PW
z#+GzLE+~g1?0AiQI07mAwJ6QxE3iVTyB{vG!oH4x-Cck7+W7O1HwJ+Jm;dws6VIRD
z;`v)oaC38uRtwhJVA7CsgF8(+n&RUU8F4<p!uP-TQwTDyuMT)}Gvb|hXsEiZR8Ag_
z1KxS}DZcq@g46wsfAn8`4<ElcdGySv@q9kx#fy*d>B~DTOT{#d=qZ3*b?!p}`d)|k
z5d6hryEg=Ws1w`+IRPI1tcS<Lf4c;M8(S2=hu}QzCh|O=IN?G(D~X*4b{#4b;FXGP
zt6Bu3Dzb^(2trgHU?zFjhHk--X_vx<R@~ktJ3Y-swd4XxhWon{ro)88;mF9kHm@L5
zXlAM@<$`%W_nwadirC7H)q)8P>zYu844f2Y$ZnEL*4Ug)m<|)Jua1Bi&gYXelwPGQ
zsWVAAB@E+$kDko9`XJ*U|D)f*Kl59E9fzAkXGO{IT41Xky9<J?W}PXHLtALF{O{N*
zQnz0}gMhji2fOhg9TNK3vK~!}Y-6T5FInN%7tV@g-`Hc(2_d*d<(S2Sp8*U$S2Dy}
zKbieJG2O;28Ku+%g`oC=RX5#quBA|laao;<-Q0RSc<MSO>{BhD9qH~j?fpF8YjWT8
zFmsQ%02^i=*#mG8WI(TAjle;2#yC!xj@MXf18~M#758^{xVwLaw$dYv-5CUB7$K=}
zqi8FzI#_A-;#|xO=Xt?LpMDJF1Z@@f_h;0FI@z2u!Br+GYmyL@V}Kh$RH}nos@hl~
z>HX1}iiQV**}s?I%W7=o0+G_!1TzPbiL`uEGD=Nae-<St43fzjIXhCyh~mX6LGrf7
zG`Xfq(_x^y8bG3W%lDLYIX7eOMR;Y}ptYf`3)W?J_Q<<ja1dRNGht9P4lK0F!vQ5{
z>e!YFZ57M;gqJU0;*(deaGn=H1XIZ<YA8)HxIIoY0{yCjqS=G;6#_wRt4oeN6{-#A
z^UO+E*UfS*)d%w<vIha|c-Jg47I7291}6eS+n7I^hx-jm3?SLER|wYp^UN0{HJUxE
zXA~N|l^uX48D;X~1&~)`+sT=Icw61ITjQasIGs-r7}_+VO#{;L02@b0nUKaKThO%?
zR@sq-{dy}}^Eu(y78W-UxQ&bTt(1D=4bY2w+S23Cix>SSm^BY%Z5z+V>g;4x#nfiq
zz&d+?9059cQ~%;57wfe-)hp^1$I}sVyk>Wll+fw|tuvssv(dC)JGblE<CwM9US0o-
z9&Zc)zw<{g@q_7%ls~{Ujkvyg0&N+mb9Mixh!)|I9z~3T38(u7-~H~7@bu{o-v8pe
zIG+^S?osQ4^L)ZI9RVBg^yw3P{TpAw-A4t#fBqBp85W?V0rP5j@!}JF{Ng3nr2*Hk
zvugncKZ_B39I)9IRqZyGhb&&&{rMq-`0yO!J^QwC58ee}V?*fi6GO{T>xy|<oPA4f
zxxxx5Q>Idq2(&uIUfDuzSr#0o2}w$4h0N4-NOHl-bjHb|jZQ)69jtjZoNpRAXIO%p
zmIBMXBIm8>0SY$b#)>i&<iS(Nb!8!ss_Q@+!XWxI-?`7x)~?Dtk{$-KSHmzcZM-aA
zyTIZHR@@PkHdRd108@4-S#AyWR`ADv@+bJoPk)RrAD<%Sf>uM17Md~idrpuOt&v{e
zWO}$QF;28=0%P27B3&Ox7XWd4U!cG!>ez;`Z0`Yjy+N}L+=Vbdf(VfeNP&VnC8lE>
z3^g~Hm5h`I_el!@b04fbMqvYZflkF2H`^^reLgsQd~<Vy!*uYXoYcW^$SHymyttpc
z&rrJre~(Esw$pvEf#b5e_{|UN^Ok{_ZtIzNU}a1Jh0Q~D4WoEv`#j&FE(z!JjJvy6
zm`|@DN^2y^hFmhH;}NACpr%+`MLM4_-{1Q+x{Zo&r1z&gtX@Z?ZNYkGr8=#UI5@I+
zQARFk4xCuQm|<W~S!-lDWLaUg0@{!~kSX4`mD~`>L2GiC2Spi17HT-ss;xoLy6(EE
zGV_oa01j@vTynygS=}jMCF3wo1k&1IQ9RZv*n2m;I?0kUlPRG15?#ODpO!{<9~Pdh
zwPIOj7Va!8<dH53+SX=tHXYp~k-FOv7J#-Xv@Mv=_c-0(VOc0U5q~yy0V)c6u47y8
zyPm8yEXq`L9<qa8uD{wCkTPqtzzjW@X15)mXUX;3!XPkt;Fydp+K4D!N9y{qXr;$D
z2KWmwxn&HfvhuyJ2gHI5_VH>aP}i5MB?cEVVmRDjoDL46IEIo+4@Tp=yI(E<ZIn2f
zR=`G#L&4;%9n;m>Tf_+gTdS~Ey;F$-;s)H_<`ghGSa^0xM7(WXB5Ud2&sKR2)Y)@%
zQ*uQhPMaj~`N7AYqvg=NI=b`Kfm(N0I{J;{=aPwRBt4XDegbF(b-YFyN2oQ-TCo~6
z`6_@}juqF}nM<Exf>WO(Z*aUZ0Bp-Oststi;O^6kSHnHn>o@V^Lu&`>k8&Q7ZDjO=
zgmqPX^x_^bKT*u5j88s#g&+LjhxpNteu|r`XL$RacQKTLR(aE|uWzs{pMaiR6+pqO
zPw(;3M=x<cRkWwTG;X<aGrMR969XeA<713nXj>jKgAX@4AOfxJ^~ZXf?mu_?{7c3o
zlu=s>OVO+$r-WLW>S42P>C6?YkhpB+%%qB?jA}+kpjFKCidq$iiH*C9>-WN!YELPv
zHZFQjUk3m(HP>aNZt~1FOw=8<$_2i)hGjkz1Ugtq30@J}a3~Z8SFiF-DPvt&fM90c
zW{Ih<6qQimZ);W5yAy_zQR~*cdtGO2NX!9L5)5hZN-h5USyp_!e2VGxQ~dCUKfo8h
z_%#%iz5#TQtJW9M9#Q-Av;AMZa!ec?sT;(`SlOOA`fD9fUI4(D9i9pe53`8sEf#WT
z3Owm6DRsbW4&GaBIG-!dvteEnnvEDFA<2RBan~<#1N#nCeC)=O7IsPkM$WaxIc(z=
z;(A5DZ!wmyuCAb}D0$p{SK5v54U~0uahLk*hQbf#(|+E>(fh?dGSBvm!<!v()6kjQ
z0O^!fNYo=SEz_A`_tTqCz&gXK;&eW<!TmaWp(ypWIcH4M0psBc0>RQ)xtJg*OWyQq
z%Hp2$=^o2cVW6H(8>J4VIQyFdEU`mN$_2@-gG9)tQQ+UOp6_AHg4`N6*hI&M<o=5+
z6v+N8Xl8vFCLFG=$Pk)m8?-^J_OWALKCFfw_(XO_U>H(H873Ov7BYaXqFbk^lA$rL
zD;Ub)EVT4B#?0+k{8@LuL~oUaRQ>+co)uAb?Pf}S-3>i1Gbx558ABq7Ye;C7Ym90|
zT^7t|R)V)$2~>oIXGU?L>sgZv|6Mj=isI{a(*~>51g9SKYrn_U`#4j+L+zsgER6B$
zTmd==jKDDqSx2YsI@sFAv0rhF-$W3YGPTJGO8-3N=XMX$ovg#m*geKO$>d=~9wtc2
z$db^Yy;7O$Z@2>l78rPsz^sC{idu#8pqw#`gReCW)_jlCMq)!lpNwrRITIIzN%OXr
zlA^&J9I*NC!Ny@1aND+?Nxz?*OHI7eUxJN`U|$eCjuWHYPt2T=6x60&Qp1iw(wX_}
z;MG?8@WNbfWt->brfc1@96`>&=g6BJZ{8>q1AvLba#_(d0}{Oew-niaz!8ABhqo%0
zbH({y@$%CJuU?(-5B`fkz#skL4>65TaC`d%x3>peT^(`vvEh6^v&9V}jU_mp7kvEj
zE3B)ct`*~udq32lm5z<yKZ{*>_-~JGG_w2io+<pS&)dH@UgPIP5^p=Q>uZ|Z>4=NG
zsTJB<t3^~3Rlu`r7+O=zv-gj5AUt_BUM?Aj!_oWwR^*Z}uQLSbI=ZMfR;@_Lw3P5D
zNF*G|GnQrD7_iz<m&L2-xY6XCX*ZDt<2X@N?pneug7Z9616bMVWtkfehm6^^AFWxB
z`nx5ES17BJaTuqaSlvRD-`MG7S)tnSllK+3|MU)@e){RgK(>7_=jo*WY+B66JqdU5
z{QH6-QNexrT;n-+Ak#LRlE(l_EJy)1NleJ)_xs|{3T!T3>v_dbfBXV}^he*vpZwAH
z@xce5QgnPs7!s=vsn3B&$qPgV#`Us&?~Z!xbGNgdG2We#?J4U>$@zjjO%tm~@wg>Y
zRBk>Du;QLRc0^kFv+mntb9LVc{!Na!{@_8gY>Pb|H7H)EeZv52)!)xxhJ9SwqOGnA
zv@Q^EY^eZZ1qm$B7^ew>1f4U6Vz5zBB;kxRROj6@qPHG50N;p|T^Cf0eSR~mf<=Kb
z4;ZH_hzM4z-7jukSNhL2+dK`;&~yP-ATJribOa<rQDH_o#_{TitE+21i$<1YSr;sJ
za;v88zMHZz%fwl(LOY{UMuNW_2JU2P8WQ08ILiPd49-UuF~w#|2Lhx^R-MIvp5ANN
z()P|AwZ;HcKm%stdi9`=o_L{;Q&T``_oJ=CRx*^<Dv|(W$!Ka2TVVz?b5B355Ti@N
zG#ybgP_jU+IUr;}m&^;hc&8A>e>~3$)wTQjN%nra1|YGguMbdP#HcRJ2F`T}3m|_R
z_SZ^_neYc>J3(V1!s=_6_}T{sia{VEw28jP<&-fLkQsMI+yrj7QmKa&>dzesyFtFo
z6r5p(DM~8HsQ`&-<k|>^>q?(YH3HPZi!hQXB#{`&nR|fvJ|q)h-qavk2qpuP2^3y5
zC4y{1ur45&PX-ZMtCeIhw62we%O2D{82K5->kKd%1PHiKmXx81i*|Cmo7Ng^nb{eo
zh0m<A>O2X9BT4Z9c0}5aH1c5KO^!D=g2s$y8@m>1FfQ=xwy*<;X6};U6>glU_oq9&
zeEAAL`N>c4$;Y4I$3OlA^J&0xpYX|xmyq-mjH5twz`6i}?5I7V<N@n)!iyIl<JHT1
z;ET`5hO`YC|2#nU*lQ1Q<7cLlc7X4(RMcbNyZ`ro4zhuVJ%}iSyX6Zym7bEI)wMh7
zL&zHCoLwiOn?69?EnEmN%P_)<dTBiusQddn2ngs@-OV0k4yI{<%>h~h;BwQS=QBhI
zE}C~QNGWtF4YgA5vMj9btS>8Y!zj7d+BJoBWxB)6oYgQi-Ku?KzfXe$854?tEljQ9
zEq&O+3fUuiZM`D{V5mi*1MummpJ1)?j*SX?pAh(3pIoNq_hT1Hd(^&-Y#?+=ef$7!
z$3vvhoE3l<OQ;!AXQ_7prh~25#}3o6{^#iPsfISM`00;7#DDqs{~`X<|NDP}55E5g
zSe6g*>{{`)w{9?e;R&WOVP00ex_d?anXUwnC@$J>ObOcd9l;lvpU?i3sQFW<k_~Px
z?)8=e>=nJQ%>)002SL!~{IEws_s=@^GK>DGJ_ijT!5vdyDu&b-U3HKP-XEjC+W-Ny
zVWnXi8JreO9sB!Aq%olchIQfV0kl(Mb^kCDnC&+JA0}K=H^~S)G_2fgbLp802wQ}e
zZa7TVzO8O3-x|c!!K?&l0<9}f_xG#-wSzNg!1dJ?j>l^Z!$7?f4D>K&x>BuSz_0_~
z&TO>=&D15sQ>;TOop~`Yu1N{d3Lhw!Yi49$;<JN2rHm!k{B1w#3{i{+iN49sg+(UB
z$E0L)4W)RXqtu|SD|;ckPOIdMX(-U8LfyMHiJ&C1w&w9*E(PN>VJzIt#kdwRh$H%p
zLX=?tV*ldbsXL3-U?mT%(ssRC0Jvc8&HBao__*9VnC_*sLryIIP?#4wyuLTZu4G<l
z63kenIb(l`;M~(nz8_d;(T1Ga5)DAY;1$`;pGm?{MvP@blI)u6in><RWk#K8^sH+o
zd6Bt4`BWiG?wvexE*QrF<2blgkGcV~GvQH~v6G0HaW&5%;X-?zjFy0$H^vlT@!Psq
z<jlQG5ZAHiSvOG4pw&=~*Ci0{0m}>-XEe(&qcu+QIeMHX4AaE(?a$*<ine{%n;mZq
z042{}2^my>?vq)uBCIqJ=-3~|q77;*L>f+~S9tmI6MXd1h!-zD#;2dYg6R#cjsCd}
zr_&jO05Gpk4V_0RsA_oe;YaxJ!x#9<SKr1sZu#<`2gV*_A<VC5&mev#GZDW(0Dq5Z
zlrDa5N9U5Qc);?2n!oJrwUS3(DdoOE#Pt;+^H*!@RlkdyUzUO090&&Ktu^M6=iO*0
zAR<_nGb9mM#gylQkuXgougf-jhUOrj-^cbkIeV)&6Lwhh&sdfPxeQpB1*f~a&IqQI
zdJp4XkkS;j8dO(YT^$)EMwHGQ=U^xnH|y=k*nC`jD=vd!ndc5z+fi}<p5en6KgIq1
zJ+5yi(D>rPxbTWM)#iF4aW>A!lfqr^#RD@57Z#7~wS&2y_6-qq+v$OF3#h+iwLF@4
zHj($+Jo4Yz$%HT=2VT8=h2Q_Z-^YLaAOBzRU;cx?k9B#4B*6U>U}`UMbvPrJ1+AU&
z;^U8TclwyUBo%;V2xgBzl2>l-Ju&^947;)IF9jHl^9DDub)9iOpD6`-1OQ6kJ1*zU
ze$Gn&+}DrC=2JYo&p-VOud1f}q3t4W!DcU&^N1l$NW+9w7wG8@Qa#ON4baNX-&pM_
zoh_!&;C#A6%GAOQ;{js)-j+DV;c3gb&#^J&J|8v~Mb%r>DRoUbZ9RlJzt~^9diyMx
zBAac`8s;fFLDd<xEvR+ISOy#pH+XXM6xY``h(TZtXl-?A2djt)>B2nTIA^rMw74H%
zfHfnl0_z|wyws$p-7a;p8_<WR(Kis-H%Y@Tr^3qkAt#7Y2i=)@zv2M#Dbf5HJIkWj
z<h3n;u1IJYa>h_t@JOQsuHnTQpHfB~4hQdabj6^7x$5qRxZUs2IY(vvu1%vFhH>!9
z?h(0+Fm0&1qPm_Dq8->oRlZ*iVq{1Pt0y;B9*W!ffPk%W%(zO&C%Lnzd*fjU%o#M^
ztI}EHWZt!+t~CEw>XB1Y3<a>}mR7>%3#%_A!%%QMUg2;!a!$1dYk;l`beYkX8K@P>
z*p5yCvUFfEOw5Ip5{6Q6m=2i65hc^0J7=~X9fkma99tw)NCMEG#E@G?3ue+C8<JVH
zNT)Vc8$iP!j?I*OKg7j2O(|gz>N{oKWIF47f}LJ=d!}ZMnShir9!HGR5#wRPR0@hy
z4G(e1Mi~5ASFy*BHwS<KIJ<45h>?69NiggeOhH0M0;>^2P`%WOAO7eC&U3@vs}s&A
zMx=(->ER5u0p_3E%z8c@pm6j4{)`tNeu7soPq?|h^4_wcrFsb70vK-p7W>Jzuk29v
z1D5Y1O1k*i%b)4*v8}3T&lq;TuWz*R{Q}4~!G>XA8l|y(KIhV_>LQx0bp^nPex7H)
zw+SV08^$=4&HG6*^1$-*@o>QX{T&&_!pJGkCFhJm1|&E0jg7TdrY6R5x?oGSk&X|m
z0gc{^#uPwZ7l@2ZZ#2QWuB?2lZq;CdwHgZ4`~Q;S6NK>FFpVR|fgXM$!14NsRx9Qi
zpeJ4|{(`!O#*r!UpW;(|_j`YgPd@$x)8QF%@h2^^dzPuQ9gIeq`nCXiK6YaL?wf}s
zZo0k@6Bq{j=Zu+!orvw+=Dk0wfK|F7@ChIu9Wk&w$3*BbFAH9L^fCU?KmNz~{eSv<
zIL~L~T(H!NkMDpFhKj?@ORUI%&ba&Z9!(ATkYLpSsN3)ea#B!-Zv7hh-36LPs1yKh
z>tGXEm#mOfIZu@a)$_XGbbp6wI)Z+a+h|W>?r)g4MJ(pV#rz65?XUTK46CE4PJH;Y
z`_KHWbn&0}C|>CcH7`6RYs8Hju%^gU!OhhZU>IQMdo*3KtTPsCO!+E{F7m}X8DQAZ
z>VlUae++Ggt0%X(I^J*%Y^!%!@S`(-uNR#KdzXy5s&!p(948zmcAQWTlGc?TYQew=
zDI)6MEJk(&8-}`4Tt8gj;@OjDc=F^Ka!yzqEt)g{1X|a2RimMG8eC4|7EUZyqxN>K
z6*&#)M4@edrL-m4vwqU=RUF;w1DmFMYq_}H-{{MoQ}!J%jC;u<G0W-pA!R_$#e2X4
z(1dEB6$)IIlM{xLP}YQ!GO9FhfF>v+n1%t@#~VC(@(fR&KEWUYYYx`d)OBo;Lx8Y2
z3=RT0Ze8}11mod|!{N#U4rV#ToGJBe0+g00X1J)_=cDx&YzC-CK|zote3PvWwKa}q
zr~~q7@#zH;;<7Ci4C9DA9#BTQZnTxg=vvu^FaZQONg%l($qMsyu6p4@67FBqbj0Dn
zmZN43t*#tbbwQg?Xy+5!x&X>9F5>$?)AVE?6POkrN+ApDdq$+FL#~vIxA$x4V~Wp$
zKr4*YxxYq=n@=i-vx3IT?R%4C<ZLWX*##*9zxRP{Yl^m>VW&@M?UP1S5lAT*k5>fD
zWk4B6jAg{=seU>hU6M*ErEOjK1&=pR0k6q#WN5%DkOqRk<^vMEFButTBSZ_tSXZQ|
zSgYatKX`$UKVES@FIX0!q}8L_3^746I;KZ2VE`_)<mW8shEG1Z$N9AKvW!t2wJ4i~
zOLSo{i$V_%ly~ewhqZdyD(K-aASy6-a4|;I{_sJ&H~!tHez^#3i@69F^P!A*@Ax)8
z7=Az?QWEAfbw6ACH7@F<P!loFXX;jrQBJEYbQw$WimVX*vKf2!c0+lXur3S6b#c7|
z4Y$n;G6?j%8q&BSgodKG!jrA)t@?rhWI$<ElgfzZ8&8(4^359i`_^hbv#?e}URJVP
z4Jak_XJnX?S6Lzn?(a@8Rqr;U9vJXkhEk{jH13@D&nrIogCF7t-~T?o^!_h1vP}Un
zjG`=%RxlN43%idU7?FpM*?9!F3+O(Tey>6tZc$-op3d#unfrVxL9UpxN2QM|c1{TE
z3;)SvSXJ@re2?${@CW$LAAbjF0LDYc=|o0vRp8?@@WYSqaerPh*aB%YuBHsB8Er90
zVyd1RJ9<La+R&OIXQoF?6B-JdB?ua%8Ip}?$Y^qhhMA0$r|q;>%%>BETtLsfEp^f5
zShxgi7wnCAWurxs6->eQf~8Hozwd<fn55(L0D#gNz~)+1_n(v01>ZN<*SNjC#d)nL
z+0#EcVJsQ8X0+ORP{Oej=ykgnwJv!1=|`x#0NPMehIwi@(s40|wMYNE8@dBNru6kZ
z91gfT9ME#csa7m?!MrZr(7HRQL`NHJ-8j?%uE0=I!R^gcy!HHT9FJF+mlNtrFuANV
z0ie&ZU}RcrzVRlvWD#`3a${GA)vaBk3N`e4+|&c?0$b323HVDXNTA+V5)FK{ZStS~
zzFY{Jns+g2+F32;K3O<33?t7|sq13kN#IvX#?T6V9GfDi6)C$UXQh}pOTuwF;`a6l
zuCJe991qCWU`f$X!MtKvDuD%O5Z9<lzkV}nv8VBf!}T>%p2!HgH)0$S6!$enJqV9`
z6l(`Km(=ak5Wavu2t`P&SZy#-4>`k!a4S+&$&jI-Oh=Sq0^n?3Q?zxV{>~z0osi9S
z)oQ3^YbTeIK~lju3^*K)EKW)SZPg2PD%Rx;T~5%tdiShY4}CN_0V2XISIGr~lOwTS
zmn5j7JrK<~?SP+nu-+WVIg9HJ6{sae&sM}Z^jtZE1_4Z059(6iBO=Q4U4{FBgKE?T
zw%oae&WIHm`EWp)NT!Ha?hnI=Dbp@V(L8H_Z4Nw=1^M$HZ_WUA0XP@zrVX&%OUnbL
zhRBy`;H@1^Fk|QQ8S6sUXH3(UIuN^D<vRz$?@PHL=K=G);?=7Y&hxSXBI87-Pt+Do
z0vm<;7~{8H+pcW=A*k!ubjc3JXFvA7{lO}$jm8Zq5a%MCvH0!VZ{c^x-^2Mdv+_2l
zzR<E-EK;7WNB07zU<c5zFK24!#$oK5sLjHH+Bv3g6WbF>Xhwu7R0xJ~^5*F5EnH)w
zTI<&!J_f>3p4C_@uCA_eI^X*yuQ+>mlTsMXuIpyW&|88juq-R4iT7=u72_xv$I>I@
z)f=w^^iqRIz^y4H5t?WB=348@#ZeS5p539=#iJ^}?|r7)E`r`F#$&;Vo~PV*cM%!3
z)agDA*^57R6ZWCbG(hwAS#n{W5F5V;5UtdjD%NGj{rx>Y{O}_@dGZ8*`P;vVAN}};
z_~3&N@bV>n=FVrsM<2h!y#QC)@boHS9Q;0es;RTOYjfjT>XJpQGYoZa^epkLww78Q
zkm=S7_u(87-_%1f^O#;X{{FB~WM8e`0ppId-ODK+)A+v8A!I^LodrTd@?N7k3#Q`%
zH`hm$k|33x6vhOMDWQtMa)u2f*2WaDn8R8jQ^xCS4)PUNS1flgVZ+!12l4b*N$zkU
zAt+Z!_g_p1pP*C<o;-Vo+uNrYhXFIlN>^)W@aN;>x+Dh4$vhNwW}U6LfFTdKeexX7
zo__)3cmP~}vn~sk^BL=Mf?1`8IO;~#fTloF?;DPu7~1zVjy}2EWOu8u8N986>*>tr
z#kR)wb=5(NKswvk*CfwG00cu)nTfEhpbdzBM_=bdTj)R213&BRUnvDcGnBk^K%V^E
z%P`{V<_5>>D-VEx3y=<i*04tkFx+azDqAn$>2P4m*ZY>MW`fAJz>MCm)GWKs&hJU2
z*87~Xglu5eO{gcVjTOo%Z9%AHAX|pHd3SRXfl_@1Gb;i)98iuokbwrq0Ni77MP26p
zT++x4mh^E(5ME{V^QC-ioJMa^$Y<W_$}!(o)a4$w%n+>z#|se>BgyS$vinDl<A|{g
zeShOUqPz&L!Hfae&Ikc&8w>11jae$!Rd$vSj#s>f0U|JAAEuNB6cHE}-<P9eILIXr
zdJ?mpNfNYHn5>YrpeCSVKwVapXeSt;+vng=5~@GTeXzV|dVkU5%>kgJ@i-aP2FU97
zAU3i7ZvI!Iq9*m%XjWXE;;tx3k#`lDJZ$&N2YE9j8PM8-S1(Vvdv#CWyTgQFfnu`(
z_Guw!!km<Q*q^kY*CU%0uVL4E2)hO2k@p6pbommm!=l90$OJK<BN`d<;4DzEQV|hz
zbOlA9Qo>pntP3?b>$;KwS}R(m9~Kv#f{3KI(=<kdt`{nK6UqXD!?E~VvRC(U$P^d1
zH)F~|j8${}*?5FqSK2R3hog5hp{2#?eDaMf<8(S<UD($G=Iz}KFb9FLnBU(u)K!s7
z@(mLnDO1>-bL!g2c+NSokVV9>tgJ}88_xLgPd~uc0xc|00+dsfE50FO7b}zbkq*M+
z2!?}#YJ%~L(G(!a1RIFi{yo@cHDl+8WI=uS-m%q)n;Unt$fn`?`WnCWSN<wmU2u1Q
zhmSt`5a0dock$gn{tiBU@nf{*g!}UhO~5o7rWzW~*zC!6M#W#GFS^#MMGL3REc#;M
z5rJR>dHtSvXNTA33UINIJ!Jnv-xXtNd+h;R7*Fg0Nc`MEVE<m%^|()2&VuXX5oL1S
zUJ_^j2-v5bV5ML*gO!H6GuaQdV7lqmC%)K^dGKD^z`D%P%I-eV3z(bUWnoLL6)L&~
z;R2Wlu5NDe{H?b!9d6h@%v+v?{tO}nM4q*hFby_8A5W#KDh}fTH#avZWx%|yNU7nx
zE;z3<yZfv&vKJ7<|JW3rV6Rr~&H~-(!U!T`eUA)6JhM(3@RjWu8?|N+HBb@1^)ra;
z+XeASopca*k=K?cG|xk;GihA@NI*n!OcEB;v;?G7kuv)n&uJk<5J4^pr6gn+U=>ys
zO;?y!TBjHb@y2mPF2%P6&zP8@nnH8NP(}>nfwo>?<vFX39neHEQIG|akSt*sQ!gf2
zmKD|6<EWZWs+iXmd6_Yk%mSz22y-FOQUJR4bb1uVNf}Y5D~!_>khVVaBp$r`lYwbg
z!0p0XFG`6g6_`D{#EI*;7qvvORGnd70SzDXIrgXNfHF?VdB9Kxnpit?M>{lTTvlEB
z7&aeYp#co(LiMyV1U<DJT~e|zposp<EMU#xYM&s9=dNIy*>)Hr_9pd;?u3w6@Vb$<
zH7tpA<MU-cqpk~D9Z_?}V#(bFxWlM%4D@I77YzV&$~~e}2N6vbB5N;L0JcfbG}qSt
zC$?QRbzyq~>K^F%FTsP?yhuY>jngdQ<)^Rk>eVYZ(R^sbL9i9oihS*|8TdmMa1R<c
z=Ir$}a2%fn0v~%u`}fAZ;d@C%Z$B$QilP^F<IA+2w_#(RlCW{TDRN5Agz<KjU39c(
zVj=9HL9j%PL{7V5A@LsptpoTo5`J}>A+mLUFdv|n)e9Kth@Z>g;EW$TO-HP?;_CVq
z=kp17clWqIoiNWUmUZoI+_af50D@B3gVU<H_QQHL<pu!wo=8QAa}pwgX)0J-!za(q
z_};4@cr|9ZeCj)<j>2Yal)sM&b4h{ppES3N*p=mr@IJ&?aVFfe$$dkAbUbK8=DwlB
zH_PM!gMV!)6i<Kkt6#%=@4d&$x58-phada^zyEu`hu{52{}6xhd%uexe)qdL%?s{M
zhOro?Q7{eykuWa341Z!wwrWtRSX**q_YJ_s*a)US^rf-sZoe+)ia6c#7zo_v#3h>=
zpgx{KX8;ZTwTFx0=fwbqMaWPx#pIHaO2O6j5y$DkO<N7t7Fb=7>WZutQxcS6KowxE
z0LBSJ%Bb4V%+QP`#UgHD;oVchRM3o8Nntr5CA+s-JeytjmYm$n3>XhbT;1Lx4=kcH
zHSf*5`kdILAZA-7W|HLWWPmS<=cjQ*5~ir<nQPDee8T<xE8O3I0&Od<UBd-Q9^2ZE
z4GwnAHA}Xa;CPTFo9;Bp7D(;`@%}fTqt+BEt*1d_LkyN?KMpA&fEN+?IO=V3`11xp
z)h&Fqb<J54U?|yNSFDwKu##8^G7bZ*S!WuxHPq$I4k#sczy@s?OU95h$}lkH=NFYr
zMYDv8gkp*;48%2<+qOd21#LMqHQLxCJ$ikMU^9C*fi{7Z23sn1$4+`QQ>?Aw?1A+-
zFmSe%J;lg1ls$MG&mDl|5i(u_<-l|D^>%Sj#JUiWmXwfm3n1c9-C8LIpo%e)v$+=_
zC1j=xQ@$}4VLk9E(+C?!42LU}VZt;{NQq;JXrgrhC=ocpP(<7pGinE1vgSWW>q3s9
z{+)>+FjPTI`8oiUAVatxxNrh+=Ln=&r|79#RM8T(<t}Sy7h!7<6acK?pYKn&8(w(<
z2mmj#q1gwC52$iWia_lRj?W<loV_Pqb*9KSJ|o4xy-$!*>EHt{Xzk#_A4v2iY+8Mx
zN`QN;^&X|ZD9{E(K_;W2uuRzf@nSRa9(=nf5+J)o@-0%-*4cW*_A8f6La(&iu?-JX
z#E)${5o7I-t<Pg@aBwi<M$l-O=b2I5O25C_Dl&5SQryg#*_lICvCI{K>+1{{J;EBt
z@nREgwP9|H_tE7hRJ_YZaUparxp(%6l<{_X4j^p@inn+j2H#xVGKJ}xyxrTBx#(%b
zIF2|Rk2v0+a5|mw>eU_Y?(fmOA~@&do_BTkj9jcM6qt)MDg=zyQ`$7rrtL;915z%y
zZwsD0d+OKWo>2kBJvc}(sC!4*L!gf+MzFqdgdsG59-UpLbv=#iDg1{)#lDI!Wp*&B
z){q6{{&J>X()iuY&0)uZ?e_rS_VxyEz4aWWWL)3eU>XlFHGKNXOMLv|BYZM{h`9o*
zCe$XF=Z0wraL;kXD#Ex9^d7iXbJtciRg8H94?Vau8r%k(xo48&u`y6dec{^AuZxuW
z?v{U6pk=$@=i%pkZs}kDA`u2r<H*XqaXR2I9vGQ7MO|mFggpawMs_V~8V9r_m<7mE
zkkg8#Rjk@5L>~lYV3CFy*dN%Mn~{5(9UHJ#iOp><ts#S$C1D&VjMD*<GS=2$wqkB%
z+~>N`fEZgpY6v|5`%~(S+LrDVVRxaroS-Ya5xqL!;k4XA*BMC+^LmEa#C684P>{@9
zUk%$cgPTxm0sF32G!+kAI5$#`RAk#g4z@Y-Fx}tTV}xMd$6W*?eGLPxlb}!$5)B@V
zC}Q4{EQlrn@tH^&mJKas<fPrBHPsb4Z3T7^<5-vz7F}T_EJj!WmBh)M#BL+e)uZK#
zQ4JNoU$xD4#k8T-nV_<sA+{nX1;kUQ;*yc<CA><eD3^k4)(g`h5x99VNAi2**%Q+S
zRNZ7aFu)6-(l|oKD^wKf1kte873b3#^RmEdbzS2CGv9-qOl&QTt-YTx_)JZWsbjxy
z8Gz{ubFFRnk5t#<IkB9|fMGnKj8`aQLGeE5G>hiOPE%{J2nA&AOXhWQJir0qWndA@
z)izqu8Oz=g%B1Io3_vpJoj}aIL``k`WQje)*KTZjTjC_@b+xf$&H4TgcO|o^<S<cx
zssaW3aOWtH3TKEnI6eme7#qLz4H5873!65mYk*P$G!ka}h204BJUxN1(&!A-K3W6;
zVFFYaq_X0jcc0*^UwaSFo<G4b42T6|56oiY4evL35NSFLMF7&+clIpA9`OCFv~PdS
zcrW(v;dSg&ZRl)XM6R62jJD_VjQKqKfT>;YBYtn1Z`$ixw*}N1%Fg6-7+3|EAUL1T
zw24UtrDPnhu1KXD3jpp<CxQ(SNKMm(!|}?s7aLm?hO9lJj{n&<afB3>JG-X~A%9Ax
zdjXbV#NnX5GVE{|U0}|t=VfVFR~9E2xKWlONF)89);jm-&CG$hKg+=d#<^tFrg(DB
z_`=oOknm1qM{k7pUzYeu@xWlO72PpZz(th1XR+cr?pZAY*8OW601PXe81wPj9!vtF
zFzgv?re^15!TsGmUfsQNAjsaTSI2AI+}`5m$rD`N+@efJG)uUj71uLR4+$Q4!o=GO
z)v%%P`>0?+h$^ZkXZAzn|8UG(XXN&CE?UFI`zaW2m-p&3=FY>^Z*2Z?5A3t<{pWm$
zYrn{_QuC8brVPPN+R;|%GQ-vxNfjj#=q3|XIG|0e>P$7UXRP}H4#NnDV67FXinaNs
zuj=|YMM|Y>s$#W8VoNSZ3Ydn0mDf2#*NXG~Jxm01Yq(z*oLj>R19GD08HHa++JJTC
z+$+UB0TYHSz1S*JQ)Yr?U2$5^Slfb<02xZpP^iMKrwB?Ca%c3+jJheMl}1rTZC=o3
z%oK<SSvnEG+Xh;!BVb{gK%hxs(kNXBkGlYQl00w?ASlkF(6rfH-$t90EpUUy+OJ-%
zzU@K59QA(2g};}aY4n>33|dpn=QHkY1#Ci6TH)lO^r~g`{_Id=J2(K<&$GCHV?$U_
zHMZzm&#*cJ7Cv}vf?&aRW)cHTK)C+Sub%}-@E|<xs`7mg6lT&SSx*DA>OI3UWje?s
zEKx#KUE*?oI$=JYX;&tKVThDApHtI?{e#VVRdb)$zL%}lpJ65#g;$TqCS$feFIB@@
z8xE2%4ih_&c<`V6S<;GbzXp-~$%Sk-gPL7dLV-y6iwMEGKkuH1^I_54;%2RsR<%aG
zFVnI>YP|Qb7Hz8g{>x%0#o9<xSZk0>CU0#OE9jPy@`7bx+rnv@u;h#^3rY^HY2YJ3
zu#@Hdg2(5=05S;KT9|EM$FQGU9Dr^M;B+ck5A1-0I>J_DgdKn@pxm1&9LEiYlpEf9
z`wHLu)vw}LfBoxt=iRs1?ZVT$DQ%I@rvFQQLD_FI{hR@x1Y303Qwb0L{22Iq<SxDT
z_gJj<8)LZo2UwWO0Ye#k<Rnm_MWcJySnNc<EcoEWSJ7@lId}^w7LkyNySf@NP6w1+
z*g`71A*+(bX|C(Izk7wdyO%hf?s0u}gTwI(w0RMzZs!WGV(qCrptC&e$GN&TuJuaT
z)*7{CTG{7#o)JBFr*UMut(K0lQWjWcgw-m|XQL%g$xKlOy&qs<U038(kcA9dU01w?
zr}*OY_xZZBQ7YDHVWy|{#T#3g5YOuJxpTpaSGbVSHsSKIOJ5X#0I3@Q3}CYlj3T0h
zDh%Fo%bA_KyL-HP`3k4g8S{C@-Q68N{`g~z<A}SvSNP=PPw?vQCC<y4>4qejD{x*6
zOBJ-5v1)}CMfLGx26%xeQet|>%%k{q?V|G@{{GA(bXzQX2nO)aazqhG+=Bpsoxy&@
z=>2u+#peSEi-W(zbfA_^>8H0gH$FFosC)f^8dmYj%UJVr$v|#2cU>#ODK8bC^IBO{
z)p+96O}W<71nvVV4023-qvYof&P<M_z=S}3KHuX!pV5+FHDETy3WH{LzmQ-=64Sm(
z70H;c9o<wndi<J*gFkMtYoo3$MV3d=S+)+SBw-jQlr+L>=*iSKMsX(rvX7||NQJFe
zJhv1k*S?7wBHogoX7-Qs=kNa}F{ds&(U<`in)L^BIvAQ}+orA9&sVp#N$}5QW$-wT
zFa<;wtc{vt)yDq4EA@NBP`rSIIzp{0+F7uq0g=LmSsz>LVlZiFR-tu))!BRCh9dzp
zDeNF(1i(r9`jT?Us~Ety>}+N;FFui7HEeRNF85szEy-YsLefCJpA;Cl--OlJd^%y7
z7w$)%VFGqU0DOJ+bS@;S>0Nh+Eod-_03cj266wAG_TEc(T^6Ku!L@*z^dyj7|K0v;
z|7<0y$^}_CpTpZz)QgsaIdCvnI^dVSZZ)G88bUCtE#@yv!(e3B6@Ka|z-2<p4F19#
zU%g0*gbsi-VCx72XNHplUq$mdtKC9qlg_=#@j3VahDXz-mW}`Dkpuu-JewGJ3C&`m
z-+mc3tBhd72*iBg!yjrkJb&vJzxL~2#czE38+iXK@8J6C$f(CV5||}Ox?~NwPF%7K
z!NPSm4(?x;E*Z=>xl<bujE!yk+?0Yb@ieJ<TG!OO0u-#vrS(dDUh1Xrbk|?&ip>Mj
zFdYtEqZOYu3?p*p0u<G;0D80rVcNKqTLQO6D*kl3Cks+auk`NRsTYKp57JsIH8bmq
zc|N;_jBf48Bi2+hfYA#?3g#0{gwN*_Bv}WvWhfYijCs+Rtx%(N42xVmUG4Xp5vX@i
zFft9{?R9(e1n<24F3j9^plv-ut^0p*uyXS6C#HW#q<1BR<OZ?U7Yl#5ua6o3YjS&x
z{<3U3K_4CdIRQu;Xw)2?M0jn_o<8^9yUlG%5|(Ab`FzIx`GmXEE8L&i%|XiGU>sN%
zgAN&H>~(3*obXy&mpxL%-%u-C>MgDp?X1az!+<{5dJzbu-=~M^^v|3vBCi30@OkUn
zU2M$X>l#A)9Of*KoCVj1BaX)-v?=CwhSt@CO94!2RiV@$CE=D3OsXX6^E5?@sQ}5_
zr6sOa>Xs~71j)j-LRcpU_&BW{Fa$Wtso-!p;4n={U{!6MXUq!?`mH3?l+Y4WgmKP6
zc8LheP%x$hTUWH^Rhk}1!CW)v8!Q8PYK;4425ydJm(4H}lsve|pBgVti=F=3SnyP9
zMa`Ml4)o#>UTW&|D0GSsvoH?>fPJ=U&mhOqUynJq`-~_DYvOD@cBy9XB@DO*Fc;$A
zsB;i+Pbo&$%!fzs<OFeW)zDT2PQ1=M9Z&{#Wul^~Az8xGfVHi?3YNOB+=ayEvZ#jC
zi;5(%TL_Bznnu4wQ-!6?K5Q5Um-_(fKo!3fPxCX#q#tPp5_PYs3`k`nI8k))7pD6?
zE8yLiHiCrgr1@TN0I~tCRIKv?b?s|^Djmr)72V+8m&qkyh=u@)W+V;5W0tEktz%4G
z*Vp$}&)yjIe5DlEx))&eR5M5_5nTCw_8=f7g7)Th4y~H+Eygufg=uLuGzg0!A=v0c
zcqS?NmOxz7v;u8KXkjG{ok(L#yAA6EdDQ@}4Jq}i^^JD;#g5Mb0A>!J5Rdlss<R?J
zinDHj%kBe(KZ^uKb<}LsjllXs5z|t@szPfGhKgK(ciy?hZ+!cg@aw<v4Sea#Z{g-<
z>cV3H$T=!(H@afKA^fZqE+C}DU9q`UsBOQt)b|D;yJh&ps^5!`mWwDQQhKrBF+vrv
zH|P<=z;<`k0Wx%DlqA$#g2f>vm#yz^=*vJL$K8^XfyFt&WU=kqd2Z~*Yt}uzsJ+u(
zw8N8a&JI!26d*QkjyaPVrpL|G(xV%OG8!tqJze~$8qH+q(~P@Uukgvsk8wWVLruMa
zCHKf%Kr4>ukx7Qi*lSbO*}<@GQ8bX<Qe_q2H}ECgT;IZ)Vr^@`Uh`IF(jT;++kTR%
zu>pH9bxz4!7nSyhdQs$o7=ga8iUh+U{ogKS13<_i#LWG>v~RlOkZ~MG3}u8!f`Ue_
zOI>h({|W}h)2Fw1{`48HuCAz8$_cB{W3e{Bi^ZU+p=812rrxccG0$7E2{mShx-MAP
zg_UjSx2H$@aowRa%Xt%Ad#(j=i;)=q{mx3vDZVW%Na8++ZY`dF1VLmMAIxJ1f7p-B
zH#kblIE*8PlyP^SX=c8=0Jd;53@NtwwtI%IRfcyI0V@^rFv3#8vQ{YE*f)ZSejkEK
zXZ3DVbpVtp4lnGjeS7l+<519+74v%b0L5Jb8oL&tGelcJtw!-iLh2R%ZB5W+y}SYD
z&sdwsMFxm&_duzsYaX;G$ta0BV)mdmv`P?LT~n%E>nxNXw92kTF8-1Za2&V_jeZk$
z_F|6!!2aA3nsE{gVSN*A<W#vfVK+X4xi>h&k^obBSH`ok=GsjF<1nCrZWcKgl)6Gu
zG2{TQUI361&E2#1F)pTP%L0=VreQ?Mxu?UWjHqBi3Ohl$f1d+VUZMa}xa*prIiN4&
zh{JSX;mEqeY}cmE*F?AlP0o1-04Y208{q^CiGpFIv}CO-WS+Z?4ULtzy_p5Fo=Wz6
zKMYYZ&wycB79=&SUJ!x^Hj;xsc+rVpmw5q81CYlFX*^;$9H81TFDDE+5iD9mTNjvC
z>ZT)Aifw<*$(d#mx7Ct9-T-fDmkR_d&GNG#H$Qi!wmykKtNFf@Ixs|xRUdzl*1<3<
zfjN<wmRD-cqtGQ4CucHp9+1k|p9vf!tCC60VEh;Z=}p(SHwS>00XqS9CKA^HW(i0u
z($M<WAkL;)6C_M1=^C}oSnUL$LDA5VP<*m_fvPjdLgVO0k$f%-zWn}k{N``{GJg42
zzJjlO^*vl)T>$`AR74s4x*Dcwe0atenEAEEV|Q@DhysX>y%QvmE|>m!!EEv^Aw<Sj
z@$?X>`!2ZOI01jZB&5V%jipS;c|@(OKJ$Rpv4)|$g_t6hgjTnTU;&Uy;XRN@yK^Ci
zQGk1Qe~;67Ms14I`Q)ixikuIJLw8o6CR$YZE)H!ju15CFTO3F+t!qAD7jIW2!Ucf9
zbVeBlilBW*XOx&ArC>T9aRd5?)f$U*LK#Mg%t#Q|Gsb;zIYA08via|E&$aW*K*q0s
z>01~wJ=R)XJNu`OCIe*kjlh5~`Rp#N{_o;IMI;!f%Q3YtP9U&+eH;qJ8-KgLQMlP!
zq)N>7cd;~@U@8NyrU~Mb3X=&94MQ7{hlD%`n$Gz6qxbRl^XC{!!Y~e)?*&T}%nhib
zKyFCc<CP4E`t=0!WPqAL%&@L!)H)O32(>IxVb~h0uR~y+k~hB3&JG~b!3{++?Q*tZ
z-^VqC9Ti0KCVfcafeTy%$^{=VFUDzf>F?|lnI#oPdN}WXf#Wc`E~{2(rA#9w@%2u<
z1Au1?%vT5>2P~p7Y_WQxDf>`R8`DkQCPVq$lXyj`VAq<@3yaK+j<w-XM%>)o;Much
zFjdTYPv);Rw8m}<NCosSE9~SV;b|rw8&e3M^N2AewB&V^Zf?8J>d2nhupRecdq)L0
zBgtOD$wm080yGvv4TXN9)wJ7%HH#`y2DDKCC;?Ow&3<z@7{FdzueUXPTbadAJoInD
zrgYt{?@$q}MH{uK)QsRq9C+pIZ7xOpToR6lf^jq)w8E02Kv7gNWbarJat5WgF@Qb9
z5dB}pK}jA4ST2w?`?)qq25>fD)h?eQBxiOc>`8JNFddH=h6!3La$@E{z;&nygJ!sd
zDZn55D!ZN#<}xAiVnA9P1z`}KscNm%4Exx8Xt@J$4H(7|Lzy5Yp|+Ll-?&2Hnqv_R
zL&lH=*_k?ZJvRI*O$1U#496$P(-q2a0JV1N)x!(mSsb#FiDSszMZzMR%syn|BuGlp
z(LJit$^MxLhLWM<2(yK2&Tbw=oq&j2{zzugm#&n7#Ce1Rf+bk?y-y<B#LQJr;KEi4
z#;a=_Z=T@l`r2)i1R<58XREnR)Vru5yDiNxe!MXN91n)3Gaw7N(UM&<Rn+SCD}diH
z>D?<@8<4ahAv<@d{wp?0>R<M&OlvbFt@z@Xp5xo!{w99y*S>)-e(4?D-d;l_qt=FK
z6@}Pv)FTq~02m%0`fIO8kn#72oj|a=j)&Tk{gG0n9pXoPXM6Yq`vuHbog~RM9W+Uf
zC~b}A+VSI_0S=a=3>j*QWoCP(@DMDekbMKZIK<pq!tn8nkMYT;pJG`naw%x4xWB(g
z*vMR69q{zYGsq}7pJ%kDI2@*aCB_}!7fJ1rN~_fuh>CULMsl7P1VKvQ!rD^?A&{P?
zi4m6Zx(7e+HJXM;C8WmU&wH%RJpD_+K90QKz4%6eGEm1h<q6;X<~LDt!TEFsWV1;y
zzc2gJdKH0y{o;ao>W}O*qOW~Qat{C>{>)>6NW4z%iGcky6S5>)hj<#a8Sv^nV_okl
z*dHehQ|X(Y2)wcQk=nJPKr#j2wKdGkilrKw*zS2+;QI?;X3k=*%hJ<7-R|bKi@&_4
z4vGW*Bo5lgH}8o3x6OCYqVChWBFrBsHvJ^9rJsM+^ZD~0B~jl~n?f=$fEGk>23^c8
zSWJM#G<@m|dx5r!wpJt~xGuf-nQwBfp{;xmd~6q&K&7Nus;`SeAl%G=N+5lEeS@4b
zs<QZtKxJbGBr%<7VlMND)RCf@K@{vllSBCLbI0~tEhjbn&wM2DTwI@N)a4}zQXWvo
zBgdk5)5!3k!rVw5WVS^zO@VnM&?g$mw2J`rMsP+ib@lcm>vp8rAFj{A){{{ZUyoDH
zy#-r#e^U`}Czn7^#LNx9c`7yK0%@D|OvxEiGMel0vH(R4NfpqQZPgsGl9^;AV*va6
z4ui{cq}u?Q!X+}iXE}mHGt`;s7fK*1kVd3|mEXfKxvYo8jEI5RHg<k0DY@A<=UMVX
z9kx1Suu=xHPt*zKUQA+2)AQDVVMu@t04fU`YK2qSOw$SjxLgK|(~;NZX>ctK%`L%-
zv$H8>NMVp=+K?=>53n{sGx9jFMd{U%T0<{n5l_iimt=?-wcrRj%_f6_XCcD7G0DPC
zJKgCcb%r<%30NyM8(PWG!8Obt@RI8($Ypf>JRq$iX@=+mTf+{_C1_66WJ!Q>EH!2E
zl}N&PJmC2BElk(XP!1D@Vdzc?^oOkOV>loW8K%C5_L=zxYug+BG=Q&t{V610;|D)b
zeDK4MAkrWyb4r_2p1b!G`DP^b!KK-wQ3GA_nHq*|(L6n(Dx}PK_X|()t>5@%{Q9?k
z1z&jYIgZCeUjRzU@Pf{b)zRju004c_kk=NF3(Jlk!CZ=nA4?fO6q@f@pWTbEz2~*V
z(-vKPC2+moh?Fww8tEUtxnTnqu}Plid1F%id!}jZ)o1Z5Oeg1=6<lw<^%mZG>m7XW
z2j7PoLEc9%KEl=26>e{CvCL;L5MT;n7{_i1?I(fgb!*Br>ar|Y*qoOY$FBRE=LN4`
zeS-Vb3GfJ!QCr6Gc*OPbh;b+gU&59$lA5;{N}UC|i1wY95Z5_|$)<q|ek@|82>!;~
ze+}=y`=zdRq?z`nEqi!`PDOTX?PmhNSmbwO;AhDU_W#C|C+9&HrqlKeAtO}IsdZiP
z$*Vhj^64x5@P{A3%<$#+zl<-u`!=Ry>Fi0gN4h#5admaX`T82m{TcP{gn3cSONA<R
zVPXFgx)OEgi15!8u*hbfXSSn~*r+eB|6yfrXt!b=mM#(57@7Sz-ru+V-^1X0|9VvX
zJvIm98GJ4vP&}|vmj`6@bXtPYrXc_bva_Rpz6NCDds>jko+39jm}~p`H5h!|jOuUc
z0`{&$byjJdMwBu#lVsbyine7Y0!s;+HmOAbHZ!u9*)>SC?~w=l34@OWGEa!?gP)%`
zm^ZNwVjys4eE~BW+n9)=r?k_@ZJSs`9b|QPt2BSc|CD0R_Za%&AhiC$`@+6`1}Dxx
z<%9xf-+Ya1Ei9Iz%Mg&&ph?hT4RsbqJ=2p*fq4;KDG5UoBxTQ6X)6na0ElF==K#~I
z<H#0u7es&%q~rleA)`oMP!JuMLa*(gbGX9s>Kdhtu*TG}d7)%V8Og!ekTa{cV|@~I
zO(g*2JZ>rB#9~1W-EC{$4|(f}JeCX?D}P(#_%N4?mZ4yr4sHv?@9TWLGksBrkxM~x
zQ);seYnkg0Rt!UelmTfRkf#YIB#hQ|VbM}su`Xu-?9|dsA!XO&8F9Tnw3Xh)Ln+9q
z_RcBYf+&$8)TF-Okj2XaK7Zd+AWYTgac}^bVXjYS+BfyND<&iolDK7=TWA?ul}<NL
zaJ+ej;c$&S(JHSgy&sbr1YB<qV~G*mGgLoQYW8P1-WUM>*Z=yjqE>eQ`0)oXu<E+=
z>&v@ki-(A(aZCV9Zk3SGi~?G2HrY2*1MLzML>itv&G_c8y^p{2t6#x;UwR8yS5x<Z
zi=qcgrK0TI+zk13Aomae?mgM|mA{wge)$=XRT(}eNdK(kH5Me&UB1Y16CZGOyvFs-
zE#~<iLxuyf#`F-#&YaIPud_$UJf`X3*OEKX0DzV=4#%S#+FqlKr1oFEZ20ukmpC2{
zxV^r?^XJd-(Tf+jzrVvcvX~_FeF4yeH8Z6SZCz<PyLi>;_4PHo8HnK3-5m_Td8rUF
zTwPz|*|TSO`SKNBe)<x3FJI!>lUuJ~b~Ys?Xl?G{B)}RGogoTR>sm1m2jr5_DzMhY
zb%S0_C9Kr?wXc2?hw%u2Y2@&kdI$g?lLT;M`Hbhu+1No`{Or#xMtE&Zl7JBr<N!F%
zM5rVbkfr_T<BxGSH~iio{xSZ^KlwdOWxzMT`Iqp^-}oBd`@-9}y_uk3Z`3Rq!!TeR
zM@-X*aT?LiXROUIS3|RCBXwzr-tmZrjhI_p%rn+?WorIm^l95Gm+Z`Tt>X7kkTwn3
zYqupp;`TYcLV5=R8F*Yit3L}v`{zA|fow`hbXBWoZ4!!Ph|)SJXEHslRTsw-6lQV#
zni!fn<5xX|b1Rwb%wRL6)Wtl_1W)-GV3p&uq>MCVXFe;mdMmMI>FoV96sROL2Z1fd
zmwBC{hI}NU_y!$CTg?l;a<FNj5TDrz*EA*%T3mxE5U#sxrC*(QXrK@1mXb}f^W{S)
z`=T*YXTQbJ(~40r7(Wx_fK$`EAB2M%fMWY-F~>9<Q6V@#Z3{Xg9SV@;tDbGKoE@xl
zUumkYBSSYjj#PB^wRgxF)4(y&lEGF$8bQB$QH$h4=C2gO!PX$P0*K<U+-<8wCQth|
zRP*(9UQqqHOot=Jt7{C?5lJ#70$vPZ-ij{`8D+>U-s{1DlMFsbQck@%BCLPNMk=(e
zsOti4E0Q&o1XwVHt&s&ysTUQNVelfut;L%*?bbIEOsKJEaM-i3(p@P4C<;p%Ln1ho
zG9cvv4PJp;7p!i(q}u#m6~JPD;xmf1BF2#<`T!DmCgjyc0D>+7+LI-b!spMDF=sGn
zgu)V)$O%$bH1#4F&?H)7$oY4OlNec8i9D1-AJ8-a;*B#rBUS6_?H_@$IO_?(tf95k
zb^dR1yfFa$m2bX}pZxffp6XSrGy;_jv#pJa0EqTU+Q9du0%;T=aTD@I-2V=%h|U-{
zw<CV_*S~~s|K(rCSHJcio;<lC7;q2}TTg(&%_3zJkH-=F(SC!-o|gGp))7CivT)C4
zTmV4Z^kV%S9s-24`POyTDJT5u@mKIKfAMeN|Mc=xeDI?gW)0b+#?|N*)tX1xX0T=`
zL+J=G#H7(CC(@h4FkpN(c{LY%G2h(WfR6vGQ~sGYDy86fJbHR{+!kVRGQ>s@>EE>y
zAf%iy9VgVLI9^@h$<yaJoz9SyaDO`C`sNlxDY&`5!8lI%@I%At?iD_L`4UfVZ@N%7
zr-TaXDoRdRB)~v_>=!J9<pf&*1-mG;x<U;&9<Fhiu8`dlr?alR=kk!j@>k*S+*rBY
zB4OB-LF|6scJJRG9jwH^%Ko(gNd#u%tXa~;{*$%g<Cib-o$vh+|JCpQQ@r@_BRqZj
z6o2?f-@*OMmw5U0_wl7KzJue{fVOh1WN+_h!eSv4K~=+Af!YMct>og=1$)Mbi<>H0
zxbx{m7O*v@ly_s8f7Xj2Y8%_=6~NY<p-HJ%Q}0>$h}a(k(HFs)TCb9Q`20Rk25=k_
zrXis@N@+$Py&<`7OU(P}a$~Diu)<KQrw*O53`vV)5ZZbu?XF4^SWajf+Anq1gl)hq
zN{b~yQb8UIOdIC=I|vd}Mc&&v=Y%wjXa-adyjEu`)$dVg6nmkBX}8u%iK)KY=p~9m
zZ~)u+y869&n4+(1#p!hNO_mlkgSY$|g7I-c#|<sgqx_Zp2F1K5O9Zd1-XaXTHd%Ks
zh-&lw`h2mLOBnAr&`Xv@Thv*HZ=e@jzyY`x7E&Z2D}gRqb!TvzqBccaXKMaNR&R0-
za_U0}l#rO>9Y^ndn-WvGFlbw`TI+ixh5=<9I&c|B%5Itossbwau+}gP6UO5W@-QKJ
z#w}$7R)w|&ZJl}UIU$KVLF7bBqRzhZ4^h<7@11%e(-xXg*AuL=m?6QRhwy$t5(Y^G
zn7LpYrf%#Vg>lwav~`Bo8Lcfo5`*Xn&2R+aZdWr5n5L;e7X!7uXccWep)O}wZJs(7
zA3qN6y}YUK6TW&!59WAADS7))f3{{@+soF%i&HxxWd^tSSzSv{&?(M>>waG_?m^<?
zKzuCclrSW311SK!k%ei4Ri0~YE84pD!l*1;&uMLnCM&pn{ssF0j#I%;fASH2^urf0
z%}B{m+kzzCmMX$#i8Pl}s#Y8Tvb~tZFH6n<Dj*di!0qh;zxu0R#kc?RFX1a+{Sux&
zy`_bQ#Ac9tDj}FfrisGm`vGJ30Ej)l=|2tze;)V)F13R^tIMB%G))#W(l-choEfDY
z@vnU0U&L~r@qc*rf5iR$$1pY2Rk5zIsPRXsh2unO6+NmAu*-<r?I*xBm%0=FeHaGZ
zTwi+?Z$pA$o)?J!TPgGovwjbE3$26Rh@O*2f@K(xvtbx$?0j`~#OcEmbtV7~)8rAk
z;W$lzHsrLiIi(agaApgfVaQ0SqP2@t=qb7>4NzqpJgw*zd}_e+@(h3Nt$(Job!EtC
zZSynS|JpyB@TtTO__EN3G`}_NLh%6mfNx3<*yqc^^!m68_B?9f#X+h`LUSvQB!Z_;
zZt<;O|0dR@;xJ8^Pcwe};~(RbPhQ}|58uVtzxHL^USBgJ_m)NH^O+2R0ZXGF;Jhlz
z5PfQ-1HmmL$?J*0VV)Pv^FqB)_Urd_(q3S{2Wk(mt(jN+BfJA}5qUqHD|?wk-20x!
z-2L3c@BX}w<4|yzGPK0=&8|rUraW{z9}NH&k-8ep$dH65scA4*>S;6GGA2THY)FNE
znUO|@caa&$WU~kjXv>28SD!j)YG}wXtB^vbDf<Me5Z-TRV|H7AzmIHX$PIq*wL+Vs
zHEd}}4g~6ccbNv;v1))}IiDek1p`?UG%*MrJa84aj>;+ILF%b=vR?@sAh}@vVq9==
z^x%5)<=l^-Eg+*5<PsFB1G=pmmW2=rsRM+~3d+zzms}(&3IIuHZ9!cdmh+6G9Wjm>
z)~Krl&{_DDSW!E=hK;oe1~G-T1**HCCyEFL76Rm)A;ST!85UKnLhF$s4UjT&I5h&>
zwJxaZ8Ff9Q)!D&xMjj^W9Ni7zg5`2eVlYY!JP0Cdo1txiDO=ew#moCHiC%nkDPXmB
z8Adt|q}-nkU<9P~j8+$jHQN0+GpMa1R3Z{{I)#C}KXdV7D6O!%uq%xhGa?lT_-_z*
z7s*?|sKF9@p@$#>7u%iVc7%q#)z&+h(5lH$6%bi5&-bYF723q{o#OmD`4hk3Cfj6L
zBQUzKxH5KnVM;%P4OtfMh2!93QoLXVe$LEtw)UW6|Ic~6F#!DdqnG%jKl}miUNt0{
zz?Lj$G#H9kS^^G2Qi65BN}wl_9Yq@$R~KYB+D(S5tAww9?HzpU+rNabfBh?X`t*rY
z#<>H){YJq}dI!FuOnbjL#A5j1c(`eNR*Ls${rjVbpV$6~wnm$tNs&WXE$FEhHABf0
z{^tAt68^s*{sa8(cYg=VeC{oE%v_WmwshpOVR;|bZQ2W_U~=E!71Tr|Z(7gDx}h!$
zTQ(UG<&8O~KCJmlpau2Gx%ugdN4&?wu`e##8g6fIF-;TB=QFCai*;RLjexVa>ygxL
zB$|t^8_aD5(2H>;aB~0vY<eE8oUl-))^Dz!;C6b7=oDbW&*_ZPZXEEk#TP|BkcUHt
zk)fw2-6-}UNW5gnHhBOyukD%jvB!&HM4!~zaQZ$a6eLS{@4ct^$~V4&ahg!qidqfd
z|Nal~yVmfXKl%f#>j}5Fw<vjpm|z-591ar>hY?8<+EOvUyrU`fsyMU_z=?xE4P=(k
z#~J|Cx?)*orYDB2ayi;(Rc}cMY#Q3wsN>jc`{KUngSoV40pp$pOW!>``};0gzt8))
z8ZwSKp<1vktkx6&k~4<Hl%Fc43cB;rD~%23vn0$bZBXW5>#)<K!9Y)0K{bO6aP6<N
zK)C?OP@@L4nL_J~r7gYvnTfC?giufq2W`bzWgZ<fVz(_uePXK>t0{^#h$_yi7yyjp
zfbjJsOT%IplTBX*Klez<M=#-eULcbJrMTFiI>OO=k{3zcFOg*-)(a~nGe|9ExUBr$
zu~so%G;D>iWA1#N0|-{kf$4s1<(zvuGLBp_@3C%Ps{;%*8Hy<m28dK(ZM_fiSPEJ_
zV_jyPv{GYtJR;=;*a{JaCZLS0rp;b;o7@Cnn?b$ECp*>IbjCba6rE7Y07(TR%>((2
zG8}r)WMWWVAuzPMVmaSqoo7HpSE^X+J@QOFX3mDe3eOKAIX6f*yCU${l_u0^E5sVI
zw?B*as7Vdk7-*39_UGbYGBQKrWCChMY8Ba~UKQqn65S}E%6`^~ZV_C7Hh5zGIa*if
zy1F%>KXX-Q#{KykC`W)Bk2m@P^0fp@;sg^28WAoAzOFP@D9_7^d97&8P}P}11u_hE
zzQ_4=NA?&akQJ#_*Qk5(l)&41hhC4;i!6c)gIPjdXAGwWDks=rXlv*dnH2!QN?!t~
z9<;D8aKalBfNu@}zyGH{!jFFV3EDb2-McU!5BlHKR=1>q0h@ub=j0{Q#Rh;?!O;(R
z7zFQs<z0O1H+~6U|N47)_UsvkVMGYxxt1|S9EOkPgK}^7sj8jg_x0g1{P%N@$2N4}
z5;#6&F`^jZ=Y5|zaP15rxR^Jm`k%;v|NgiBd-z|;{~Evly?=_kS1-2-sqS>{z9_t>
z8{HkiIJvMlIdes^s*g)={l+c_mDUxVne<J{BjcWu!-i`&@5O>7wNIv|>c*C=D$FUv
zw4hvHF%9e2zAOt`tw{OM8IRGcf&qFNwwXRzNx*J`0e%gj@LqStB4k67(B1#~aD{*A
z>;D3BqNiYVz1V}l4TMPt?MbM=`>gc+KHcBbW4ke6yKxdfvjLUe>yQ3H`nX7-cIYr&
z;rUx%z_aI1a5xOK2bvqMrxBmL_z>Uw-k&%-678U9=kxyi@8kOR8p~Yq!JqyBfBJ`i
zgr+N&wV_rZ6(I7duY)KF9uRA-IGxW}mz9m#JuMdutUaj0y$Hqv9PjX;^v}LLcwhH_
zw*KesthRgo|1a<dcu_?bWKwI$*^8H4_mv<(7<9XhlCTXn)<z@rXfN1w2H<00r#nrO
zpu#lZAOo@}7E_d?rvoLEX>t?Y8m-shh5$*B3g=lTsK8NFlAu@uav=+s=@_8PLiSo4
z3tX%rcTFEU^SUv)JZ}|ea2=6F-&KKelZv2>?l}FKr`@xp)ex;z5&#iy$eoo6z}J6?
z$Dt4Bp&)qVZGGOy8xL7!5Euij^o{!zz;V|%xHW-k25fM<&x|nivEHSH_9`{gj;JnY
ztl2XrDT;&=N($dRuXZ)tX+523G|wU^uW4;(v{X@sfo>0oImMu?w(5F!*I#>DwAG3n
zeTRj6MAHS!!oYVJCnRsc5c*z1aI(O<u4pS6J8g?Ik_3p<Bl2FVDpXwmhup_>^F0qj
zR@_?iP9`)uR}t@V?qfz9sy0|ny%JhPdQ~}}54z4g+99U>INALVuA`Ua0lXMcz3rdI
zx}P>LK;!eT)eCHVUW&6OQFK8sS<Zo0SIp;=Gu9PT0R}OY6xK%8HHGlLKOA~26snaT
zg6p!PEN7(Y8pytP8(>LjedOI^c)hU?Z_ibvhho3b@x}n~2S4~JUcQ_G41g^#jS9?$
zB>Vn<@Kh})E2pIU$RPt;(u)g20^WP~Eqv?Oeg(huOJBzG=eHO~Z=DlEXNTsxkJAoR
z0DA@y{X8zgnDyWP?1bJUo6!S??;+#&@VM}u@(o*To8(bhl{3>pLn`>{%{TDxeCyxF
z|Kk3C2{Yhyf9Jy5wi_fH>#{HWh^RLf_c%wkJn{v|^_sLcD9-HpA`iffl=x*DDYRQ#
z$}i`<^Q%ISHnUfo4kfcniu)<O)y4rLWH;h@hJ6jYx^Ro>Ca`Vm1kpECXCcL%)T7Og
zK=3cz{MYcezW6U=7^r)coVrLjp2uTMQfxf&qyyGTN{Bv^yN7;}&htUi&r{8|#b0DU
z@-D{G-nGyvoBo;OIAJ<Wm_~wxJS2z$x7SyA=bg9kl`nq<hvS5+tLx76USD71>)-eW
z&dUk&X~lqopM3uZm|w1#7n(ws+>^azVF2IECA_X-=KV028MUo0g!gj}E1kHe&d9$$
zANs%9>({;q$e|zZdbo#`+JSKTSNtMj-OoRgZ>A}eJx>`JM_Owjf+Uw6R1q)m;Iq!n
zef<(!7cR*+E=1v!-*;!uibwm31StvA%~MQQN3=z8eszadeJ&{a+6tco%+EED)M%7o
z6m}gk6kt3cjUyH}42R!m+oHf8`1J8Ed8>r<^`jxIFyLb)Dnp%l^k0(1ZC1AD6wh8&
zVL6la+5K-Hl%*6!Z4C1Jr6PbLJlM$>&psHGSo?r&y)`p7Qri&({XQ7iF|xN_6`B+E
zWtp1xu+R$dZYID%P{siUW+>Qd%xz;}EEdX4hkFUmHs*Jv0`1CN|L!TisuiuSt`Ft4
zG&6SOaZP4xWb4{`hU(098>!zc7Q`r8i{iYcKht;~#n?aIbVXgy>}=zyU(s!EAP{1Q
zk=h!XsGH1logy>p^VjH+=?&`$;6+^GvuR>u?-g*|if4dWVB^m^;i9_-3zv3`Z-2fD
zAd`g3ft0*qfW{uRNmHV#9I+Sp^@WTnNrKr{r5|fkZ7Y`L%&O_6z~~ykvJZ5^?Q~nS
zlK;w*paL|lFeNxHc|aa1i9(_twxranu1f++_H{@ANkDtUI*>O8fX|*k$1wgW#<5{J
zD?~Gr9$1N#+76*guYgglfYa7VMu-#`&b?YJX~WIc5x?~H_wki4e-UrL{S3pD0k|7?
zN)SX3Tvido=m_pPOh_`MjBe^1tRS#~1`2{V)(sa-o$P*X$3vFEc5jQ#zau99-DBJf
zw&LMo@CN9zeP>4IwTXj>B>3jjZ{crz?O()yp8s>a_{ooPfA^9TI19|0R~Usr*9W-y
zIhZp;ozdDaj<j=#OAkv2@w8WMtGA-tv3~|%sA^w4O|?g7P(=bdl}+6oqqD5${<j7x
zh2Cf>Ar*?-V^QV_iZ2`)C9$t&6HcoT5W^>ISQfx<r2+UA{R;lw-~79H`}z9>dc)#2
z3DOs~Jt*GBh`*H4^-cbNjva}pj-OzCfs+0?d*IW-w!hXVpAkU$A3;&^CX$DP`BElK
z(-j~Ib2TW~iDlJ>G!(r1?%Vi{Z+#O@8*Xo|QIes#Ez@zh#+W87GUN7mja){|4On!-
zstp7Cr2tluU`!3F_}<+JBE$e$(ArBZ%S)(E*ygoBvcJ2vzEM>F^FtOUd1^Tpd5#}_
zeSQV7?7muC0T4BHTQtnq18wa8{ANdX8wl|~4<*ygxok$}Ft*I&er5oQuYU$0bCH)|
z0=ASJwM)$nuBpilJ}n=vF+F*XoClbmFbxg0ESS>?Q7I_khQRy4CiSX~;53sv9#dkK
z=yU}u1(?rdJ8avidRG;TT9a6(`0vRJIT#T5obifMaUJ7W+)8B!7C7F4&|`NE-Y#^a
zebXg?i-nc43J*AYzE*Y>!k6<k#zw5SY4<xi;6wizk`?Y`&=l(0#U9M240s0;k#LxW
z1Kfh?=$|Vr{wO(^Qm##=T$6{);txxS?IyiS*IIZyl8s}>3;-a2YNlvQ<#UtVr9rh-
z`Zt1XRDiX(pR`-riXWqcjHzNipS;K~_1K}MOmlUuP+d7*oBzG3x4ZLJgH;^}6G&ea
z21Vdz4br=E!21~UFU2)&TB}#3)4wleY9)h27)l{qn}8%PkSYpPR#Yp{nb<AUAdsAp
zhqC$J0yMkVv~@7?e%DF1iszCN*SQqzpLf`@&_)|-^Yv{}%qb(Mf*}KA89CN<1Ggj?
zoB<REemOboS^@LE;lh5=!*sxOyhhH#pYedMBtc0mIx9KxUPS;SZ-xNhybAa)y^oJy
ze1iYa|Lb><%Zi+6Qaj`fHgMUPj3~iIb<GrvokWm4t<aEQ7!7!Syu}yay20B|uP}|=
z^toBIK4@cyHwQOS=N2AF5hd<ZNZW!DS75$=@6Iz86MaB2JaTl8Gk<3+K#|Vv-?xXB
zpTSgY$?@&y<NXosjxyoj`1*erhr@ushyMg07r{E8&?-A_gurxgKc|=+@N={N*&bO3
z_|v$ZzF>Q5IXlD0t_*yh7w7K`!rp+_%-p<`?47xoJ(xfMK&D8Y%u%H2SyIo;#%2Fz
z3(7%usUe?~DX0D!+8UZ{1p(J~z`y<5|2lr<m%d5wLJ^1zFszW&SP&SDM}JnYAKrer
zFJ_Q_SGR@d!DjUE_ZVB|_pR$V!CtC;eE3t^GlWodMrUn``Qw*l!(!6`@buYJBrz=W
z3=zS)&fTEfROpJ|48h=51Li7ND?RFBBN9is9Y0BQ{Af^A)OyA;-(y{7<UI1;$Cmzp
z=?Qb~NEz<a<vZSQf2aLfN3h88tSWtOg{Bv~I~0#EB7FXPx*T)A;E{!`(gYN}my+EM
z#H|?oR3z3T15#?#=a{+=U$7Y#Pn2ig^^De;D|j2O>G~SubOjSZnpduv&Ugf1%bw<9
z*9?9SSw(?6&Z0~iWt@OAV#Su0Y`&(516Xs0y0=`EORc5=-j%~!fC&LPSsq|03|yjv
zOE^PBfM(XHp|VY$5GB6sgR7Tg-qhJQzQXg|fZs!#uKh5IB9$#LN58(lE*rdGt+`g+
zS-}22A;9rAX3`Dn2ew=jFLn@iWDu_kPl-RTjXi70F8O-s^S)uf2RV}nGVIJoL?&QY
zNs9eGwYyIV1bit+(QSl4yTT5Bp_*a!mY<Tu3w$yQH%*a(2@cI&i$V=U)95eEYe<>a
zED~)!V-E0}5tZIa3}O2uzHn0&%Ck(#tz*W6@2Alrg@i!(2(qAxV18L<)TKeaz@pFn
zlD+tF>;N!Hg<6G4a}USdJ^ms%O(}Kjrd)Cdyn`19saCY+*`UOJ#mh3YvU^B4PKCPQ
z1SE@UV2=IG3z`0^ngzo!K%rRORwftTgIoq24+VoaHZWb0Ofe+ESQuo5qXz(3{R{2{
z^0jY#0j*X1_{SgMcYo(QNSU1|>bm-*gI_wqlCfdnwt0edbwkF0iU}3hVz|Z`$9lq~
zD`Z`nB24CHlz^L~`w<%gF49hWZBVqa`7Eu_LzW_@&>jvv_HzNuKKpZi@uP!5k-pI+
z2(Ay;_&2}tui>A(_+8X>#xfL~?p{K3!r(n6%P`<D9WU-dZ01cBP_Px71DA*_LkGyk
z(2!(TqHYms`Ta5Vt|gI@@~FQ>t0gZIaDi{6T!WEiWKMsj(AsHRsk!r%Sb%DNcPCr=
zEe{P#+;cT{P-(4UsssMcJO3*FTfhA`F&&R+zUn4;RiI=@R^PNY@cLQLBC6l^&pSZf
zV^92&^>SF}pAj&qAG`1En_pN-80{UxAoTR=4)>=M)@4O52{+eQxISKCC<Rwn{Cz&3
zaa!pO*Ss=2@L9CiGQd+p>Y$s~>09Au)DZSe9yEl}_qwi-H1x9$R%V~7<#_Ps(H#`V
zSb3fC+ir(l{Ce$P+NSy4zkeTy_~Uuwqa?QLf&h64bJ(rIn>IIsX!y>woI&T>Evizm
zWzJq11Z!g#n%2nVm?~^lNEQrN2TWJjti}yS4Gq>-3iMq|-ArMXsiBfsA)Ev#BB0lw
zU>pXFhY1aZT0h^Ulk<=%C8Uy2v_QmJ3!x?~B|$35{Z^@i9ETC5B&d&+1fV#$ZLUG4
zJ}mT&;+q*$<4u854k!bwSViF0FaSh7IEWx*0~S4S-Ocy??+GD+=_RFrw2LeagB*fo
zQ&*UoOF$6q{AeKpbJku2@Kkh38ObuikEhmCVo`;pEp-h5=5`|5s)H#-f-!qhJ(ykr
zYnNkO-Hs;;SS*+-<`%KV!qV+QkO)@Wx)5-}paN7gWFdp8rhun2wJBTksk2I5hSgjO
z(A+U#^GLPExhoL%tR0US#%tG<@-xYLu3kYMCguhRJ3l20=~MIIOpqP?jpO82KNal#
zEU?V$XQN@oUK#)c3rLKQE9HbV6r^Eb>nG4pL=uZZOqAf<^>xLIjo4^_K8(oHvtC&e
zMo|bVnikJ&G;cKFI?zUjS8{S7%z2qoj91T?`T9GS0cI678>G?w!n6VFf-$q*>5!>8
zCU^$2QARSBf^jGqOU6(<o7w(?0pPP|H~7XceFeYuSH6uGKlvDcdj1is)LI3d#3SNm
zL=MS)+Y=a)`yZuLkXyqM4Np?Tbp8bMPri@&#e1-)x5$Shl353V{sB9h0vE=z`*dq;
z3_t6LXkdFU(<hhg#OnvYa$^@iZyi#Pc4MOvGOGy({M}#vJNVz({~jNH`V+KPF^vVS
zu1M<&K(>a`Kwnh3Ir91_XoZJQ4#(BY@Da~#({biq5r`S9+d9`G6hU_;M7Dd%)Y1l$
z`Wc1((Zc|Nxqi^?1{^%@F+x-&vS%onF-p<6sr6QA3jB5ZHvaa1^S^^9&z_>zhB8cE
zSuIGQ=0#HW=-%1Up8-hroD+6`?$2iT-u?4?sBydGR385>08CGJHj99%RaTN#pJO6i
zm}tIT(Y)y3d_H5I&sgTgdr3x$*{HE8Oed{Pv8>bqnk8<SX5oNAwn+jo^SP=jJH0Fm
zhVjtvMX=#}Ah!P<2%YBZ_v!U}pxF9oUmkr<`$qi`ygy|7-}FenQ81$S%KE*P1~x`2
zB69(tZY12<r<4$_JK24P9et`xC<N4bl`>*DUL%bIsws3`P?s~7<xJ2B7E>iL)Q~)+
z!~#0Ai+MVf5jWSj$m0>`YQRqDPBYxh!YE59(Z|$5UDwBU#!WDc1;aQ1P$Uy-+mf)F
zl4KNHqUBw&*^a>%$q=x@cRCyaq+Si0L4Um-+`)RW!2a1k3jPu^3dyTd$r!Sr1&Hqt
zkYx|zIJYQ+%V8V@#(t0?T{B2)odj{yV@U<2(E4O=qhe;va<m3*ZgA|G6oJCV0kV>*
zL_-Hz(gCEG4M?Jv%>%ZmUM*#CD;Ytv=z;5}URMW^=Bp}!xw%adi*_P4?uCHKHtjK^
zZfE~M_HT6{C=8s-aKLoD!8p-WzqX1h6`~4puTgH#MqOX9wWReS!k{AO;%zJs$f@v|
zs36K)JHL{=Rv-%o-@CLb*-2F-H`k^OP4S?E;Vd(aF|KLWJ(L0g8EskDPHtp4dsl)T
zq{Y_(1EO0}@^o2{GfCLJqqBXMdH%gZu_3zVbWj|@f+}pC(atZC+U#DLw3T5Y4_VqG
zWZ1_cV;Tz6)d1GvFBky+^e3Mnr)#|R{9U~N{@3xr_dmhQbwSBTPK3&KU>sB&&<IT&
zn9|L@{U4w>LUAh%hxs0-AAf+mAO8sJ+i&A|yh3si2yjsJND%<K{%<?CSQw(ZXD={+
zxH<IbD;NYbXBq9$_r;-6HTM8;UK|#Y=Z>X(<nvzJuyzYvuuNt~e<V}<)i3-unpFI6
ze*b?<zzv1qK*Qei*`uX=@x6wq>`O3<WMdd(dMM>uAa2~D`?EgidrCHXaS0GmP+Tlt
zYsI>*D8*nH(9<v2F*q>vuIm;B5<pL`_SCXEdjvKRSL?c3gF6O@>pd5yP<|zS8GrX*
z{kQOyuYLna1<?#d)L8<HRM}=faS6`;Y)32t{rLd>9=-NF_n-eDShyHBK3;+W5C`WJ
zO*4IPeNDaNFcgf_z|9i_YGMz~>Oi<}re+jvdrDS?+2FYvYGuV^bZdzafzKzQhXW42
z_Y7c*xwV^L+vQsSZVsP)I{$URw!ar)qw)|uZ|t9K_sO>R#hj1#<~MAXesMkl^wSQ6
z09#u!-#8JvQfFzx05+~imBm!Gt*9O#w^otzh-o}vy1vCQ9bpEn^BH<RVV&<W&!^7d
z<wV&`HU%<4nl*C}z`nrOH_!0o`P(=iudza)#+0PB>g#~}Y??q+keea)>B7@RgjO31
zX9zE^EZ`;u7Og;>Z88Tmtdi>De{PZm)8PtN*S9ET^o}z8ck@(P>UWkbSfBs~8LzL~
z!3u4gzCQ<Zz~qy%FH`X@**>HH?z%dzWkBR=vx&Qo2*ho(qM5-^26n3m0H_YQ>gpw%
z8DgoE5NyB5_fmr<Xw}@Z#61Ux0i_Hy02eoJZ>=jhVr)mCMt~V}$(eX;!MZe1QwG|r
z<wPI^6Noiv3z?BQTg7`oEiZv<Gi8C@Fd<Jzl<~k6IDlk=>|S`{dkl+N8oL`|Ti<%?
zxs;L0fMJ}F%jkDK+9Y}w!79=?A<4J{#09w+%BZN^6d0rxn%<MT@zq!x(CA^9e9sYy
zqKXtSyQlmWhMYD)UJ3=^W`=RhXiEG-oom{Sz2w1-B3M9k3JHA(GX<Z09z`E&UCx+a
zeT4bdhZqi5K)LlXsSr3|Pun?De!@Mg$mQDk=RMvS0RE?c?;ql=r|;qAr)Qk*4X_D0
zT>-Ay$-6S;B*j(x8P;4AqYZ;K6kTyGz^x=)OTxh_PA@*fhkyD#y!G~bxVpNf%X&UQ
z@~&N6MmOP2CS6>--w5J(XwbYrE;bfmGqA>IGYagkaQX9(9zGU-79$}(0E6v|*B*k#
zl+0Vk(SiNfpZ^Aa{he>&AGP0s)&<j8u&#^ub4;6epDJkF6j~%8qP^rwR8c`vWd#_U
zN;=Dg?(3u;1(^@;enpN0ODbrn(AGK_TIU668p#Ii3ULVDtbayqj@$KGXCy=awe_J5
zZDW+%17?CU7W@*vi2vxX{RjB9Z~i8R=}2+)z=&^O;b#dK&wQV9+#l-f-$QV7F)jiG
z2FMfXtlRD(<MAQT2a&U)slPXc#w#Uik4k}*0n@?fSIX3>7*H&`X#0v4sq2En;ectH
z@bcwLGJ)^_8w7zW>;_URf2ARo#JCO@c}u(!HVD;y|0=7}Qub#jo$c!L%?C&rYFiAw
z{e9leACX9>wq17$uP*nZl0JrZRsNlE-re`lg$3k;a3%r{%sqgJO|w7A%XOg7HHq-R
z2r_gl1mK1_V7hvWtD7e%;{iw+tv1NA0_#f9=AL)aSwekd=m0IzOE62qG+p88^LKFb
z>^Y<iWP-JA0UfOOFwPKe;7K4}NRYiaAp$Z1+)K^RIc>T+2siVdVG#4wE!ih=CcC7J
z>2SpHc!g310u=H0q~!bzoS8{(D4b~O>wxSr_HX;O`JH;`;VgD@;3mEf`}l$S_y*`k
zAyCvl2Q32k7+*cL97S3z5=l9-Ga;uj_T()YQo$4>+U5H0AlZ2&m*DW?2{1=?bAi{9
zs^Pp=)K;;q3z{k4lj(~}AG^jL)XJY?W(Cf|DFO>8X|SBxzd8$@JNh%&)*{K-P43-?
z)H2d=K%TCUhp7`Jv2HU!tbNWiW@1rHFP{;3^*s9%r$XHuq6noCu!+(e)+(WtNe1%Z
zL1coYj694OA!w!geqVZk&G}h@y1E4o8Nlp;MwS3m(gP|9q4#c_0icH_GE_5cr1h1q
z)tS6_?xeApli;Kef6Gnr7^=>YQu#`&4a@0MoId#p(s;m-fpWzHJ9r?QoGIrl*|e<P
z9QqeL-WUM>XaBd~!SU*kA(rv+i%&40DQHcSnSbw31TG$xzSwJ9fwp3>1qZBnmISwX
z!~{WVfSwf}e(y(k=iMLT>iPz0C@5orSnivW2s|z95tYXfaqL;S%MIpoV|WbYJ&aiP
zAg_NeHrC@YR*|o9af;6jcH-v&!T=W74YVxD;|lV9H^Vdh-EaIG_@BS}F6QN)-cw01
zjgxm02=KQtA?C#c@JPg3qk$~IO?RE?HMXv(+IlKkU5~^F)YB9f;3<IR2IiUo^Nija
zD7i^<fWyA1Tf7oXllBVkXxkR)T}jgWqDqR0+fZ`Gd)M#bfAm-WBYgWeejCGZ1GQ}t
zq%hZZ-<2S;S;2G$@Bsi!4?&?Z=-Bs;3{4nzfTZolP#n4&o52SXUj!(P6qbW9XO)tt
zEv9Ki8YT+Kqd#(?1~Y;EoDYWs(@k~3=`{5Y3}&|3Sb!{_8n9N-$v+ieupFwGZ-OE_
zroo@BnYT)!_Ae-v$L3j|HxUQ{uP<A#v4?yv><RrkyBKpX{=9!(JWw$|$2UDB#T<M%
zkCm*DZ|;1C)CrlgGlO|$vl_=}(=e4~IxZNe!}Sx~Jbwqn=+ADgoImpkwk$xa9I-iL
zNCIgqs=B)Yz%(0J**T6AZf~FB>gt9-D=a5WclJyUs+_GU@G8hA?D?7maBsI*hEypv
zFeCG9l3?ooo;h`V%gu?5Ah9?(nQ@$OJX~QIMjtEu7uAa9n_6;!FbwRll6w`XkDBP{
zfgUz-?7)}G-RNR_Ft>{`hODw<hskjJfOJ_*->=k@r+tp^#+|URfkmSp)MhVafq1YC
zP?%=@dzw~+{=s0qm;yi#n2@9cBjgNq!|2u;=5@hb7tG6orB;+-#yA|@{a{6H(Vk5E
z+M0v~8U*LO7p*Y}X-S~;R`EH?AUX$=DO(W?Fw&TRJR+450`Y*-*ZFmU*45L!*#joG
zE6IXp3Dyh{?gP0nT^lCHC3g)j00961Nkl<Z<`u|R7npYo5ir=$#h>VKgyhjdvUtTj
zfhuelY^uRf9ROBH(+<JdaxV3jeq$+^b9GtJZdAEOy=jB!7L4bdFlGh|EdKE#NmbOw
zs$Pqm?-RmW3|r~rG@=wrK`0Yg=9d^wA7dE{hWr%C=NLCRexAjep#U9L{(=GEtCxcF
z{Q}hm^ZCq+6ocu4O>gECRwBY<6^%8b@}o37u?4p>;V213CkQf1N|;|R`0*cqA2&~)
zVz|A<;pW;Wvy0SCJ^M=rFnpu-eMNKL`fGmH@sLIE6Yih!St;Y6_d5`e{TDkJN$oU7
z&ipm;=Gl_){OTS2H{Sc_@%Mi4_W&r}<NfDo!_h@o0DK}uWUB^61sd4_Z4K)@qei!f
zGW7EhU&v!q;IsA(A`^Pos(q8snf5NBNjn_iCgD5A3>#CcW)R)L7;H-gM4FLI(gvIu
zA2O<b|D7-6Km05I9=`o6e-))%qZ&8Ilv$aWpvb!o!ijhUtM#)0xdDE!;=DXD>R2s6
zY=Dx-zzGoPE&ucG-}XLb*x3#Q)5MBjH~03)9Nz5S(Hc7o^fAt#t#$1hKoje6Xb9t*
z0c&OdT2;_qp##Kzcl<oG11mFnHnP|&@6L4tYdt-ZQct({Yuw-43&w2s{IM~{uQ7kN
z6(l_t;O!a6H$5&|O8Otv(`nvHhK~PU(MPI<znZSjyzwPfru)`L!(1f6;p!IG&)>z>
z?K7we=KDLe^+c9%KBF!Rv{m;LWMZhKf@IWcCZ%13NWo#6FdeTkO-Dcl>r&BV>t9NJ
zcEeUE;hUR~12ebODwziT{r=b0^;R>W#nZn73<?$NY>34(6lG;B8pPbcayoi@N~Wus
zl5ViJuv<#bz&K&dWTMy!$U$g;F%Ph$Kbru-;tb9<jzp4MbNHNKMHLykKIR$&YH{wy
z0{HR0$=}~oJ(k^`!<{BrdBPweZL6d}AeCvm20?1#dd-TuF3zA=7UaRTrHR_oGLB^5
z%;&2yV3_NQ`(<`1M@5?(tA<6Otr55zfW%Bp84AInsdsgXj0#Xqu?ka&QQ&0ofX|8-
zATZr6C3)3%hB?cotvUNn2Fog0wy@<Azu^IRs9q5=c@Am<7>gIDz&-I;EJN}l=K|9T
zTkg5$4F;4*@Lre+G59cp)fLpEH)LaBk9lBXh%uks2F9P22pZUp3MQUagoEb@$X$<D
zGMYu-XhqYCwN{Md$a`bvT2#QSBH|2y^R@XJk;FU`P@!$`HLiLQO?B%Y?wLb)|9P;o
zmkYhg@x}m9Yk}l}1FR`hRv@iVU477O;DhJP8_B@SFl50E1kVh(PJ%Jo!y#i?D-Jp1
zqaXbQKlsicL!MltTpw^O&U$!LYrt<zuml#bT`~}{dA(*I0Q60(zu%+lhsQ&<0dYyY
zzduY*z%D=k&vop;U-JCFyMF<afn*RJ@Xx*b*YNjW{C(UlpSm7_y@0cWuZU8DeWP)w
zdBo@jr`pi!in`2L&L?k_p1U?Lft0!SYe^V~0Z78<&k!84%=_Ddrqm?TsV8}KpF86g
zlM=Cs$Jo{#Xtzz%C9<LvaTdyuOTri5{v!S-f8&3Mf9@;)EIsblMqLgm(8uqsImn5z
z6M!~dJ|l7VL4^Sa7iQfiUyQk6Lq7xj#exz?3e`3)$7#18vH*ma6W(^;#aqKoZmWQe
zv<yL1{5kr|+71AkG3&Cf4a>433qfVnU-UU^Zl>xRl9{;<aDk-J7caeN>9!xY+S755
zuUHTDeZidXLEXN>_b~==KgS=k_iu8<{ZjzZlQiy)B%W)q!Et%pn&YgiZuZ_xd$O`|
z7!SC4_7<kA8vuns4vMzUsOK5W=^j<tR)k)Q8P!?v9GkBhaw(XuZg4ytFick%3OA!#
zE9O>lS{9t=8Fgh~ZgF2E(sjYUC8RX5M>n55v{sy3?Z8R3qOMFom7F^R10f3*!3)3*
zJd0pD9Jp4%dX+0#>J?h4TTKE$8wL@KW%Q3s-Aw>aD1`C+?g{ySZL<u-_IsqV0qok0
z#kGs)wj(eMa2Y`&{JAn567NFAU{xW?wrCQS?LKc2MQV-<o3l5qHE-WY07w$Vb;&6q
z4Fm0}zzXjOewRuU=XIxNS6iW8e8X00h6Jv!sT4A7fCu;C=AeMG61fUcQzfC27@Ty5
z+sOqbjo8zx9Lp|60zZpIC7YHmJTAAT{BjRVZkfYJTAjpcfHCd+Xh5zF*6x8;VdK%O
zcZnf_9RQO+Y`+%252|Dn6RZb1D&iR$AE#ag90gN-%)t5@9i!s4ExQN%+uFc71NK_Z
z68?(fmG8S?-T=67rtjh*XGdfexc(_S4wcM`;PAbK2c*tQzG-;-<^a$uGDE9j0aL>h
zX}4IeO@T!AZ)hu~cE**=c%BVU<iNCX%B~#|v{fONAhY7*?|y*mXSaBId&D$ml&R2P
zD`z&_R?xG~b`^|3FU=1Cq~=cWyNx7%k3ni4nQtim!DhSrxp)M@X7zuwh&odDkV7yQ
zDP4e64!#h+$GKp8x;@mSRf?JGZx9X5Grn;65`OEgzl#6tNB<9~RvgCMqw@1SBiiVw
zY$>wv3`2{Qd_$|O?yRk0JWT9NV6IomK6&89FTM#2T#E%>XR8fy0c+)A-&C=jXBy33
zUtiq!(2jNF(T!=hg&W{zK8+)i2<COg>Ad1_HR1a93I5J+{cZg7-}vijDp;HK)L|Hd
zN8|YbH0tJ@QS5@<Jr9R^8jefMchdXzpz1WvF&2@KIP~RXJIvgI#$Qi}icxPf<_o5#
zMz%49ZVniHC32XG4jK_#*A@vyYBwT&HA84W0kU*KUP5WFzE(wToXY}S)Kf*A%}s=9
zKQX2X+j@c}*Lf+t(vaVgQ<zyhIRb(o_18$q*u`z~ZL@!O09x^7MK0bOpBo!@us2@$
zC-6^uvm-hVm<jhOY!!+DFFPBFOMq_Jv)D17sqq>i7_V-ShXaF|WrlhHrfbD=KA}~&
z(*oHF^Z93%VDOD3515WuxP9v#h*ue^DXh*|RdHTdygHw-IuopFFi})bVImjgL@SrU
zZC=J<f(Qe|6^63r4)VqAFEs8#V&NDRUegsGfyOG&l93(MKn*6%{qfi%c1yP^*5!;P
zXB?&xz|ff*u6B<Hau(~8?8N_}IlCB5Lhx?MSx{$~6}^BX#!CRlm?!}WEb@8Ed0=hr
zo^QJ|^5B-mbQ)GL!NfZq%wUo_L~(7rAp7%SA=kPtSk@KuvY=|~bJ|S3>y;N^HIA+1
z+BoC_Meg90dm?|XR@#b$?_bLq;~3`G1T<0g4iopEWZp<25LIMt0N4|m9V=MqG$iay
zx;NmHkbs=%8|Wki+YoxH)(bumETV%bLlo@+t5#?|(>BSa4e-{n1i{sxPwRGZ`(_XO
zJvV3i1?J3lYf?RX;bW&6{fxu>TD5Iih_Gb}@7XBs@$t^O5i;Cr;;w4k54gu`U&~6#
z9vsGV^;B~rxiXC-wThs(a!&iP2t+~FFM7N&0OVriJt-x5#pM7rgQ-*V{+{X;!MUAr
zv;|L5@mwls2EeWcB0=mnY7wd;+W>tz<HaBT5uV>3aeZ}#;p#w|JY{Gyh?_LReWMuE
zZ7IeudW=ZCFBm(9U>|jC<lgq$W1x{6=0+nkJ;A3Gx{8!iv{~K0V+Zac%IMU2yx?uv
z;0u2X01)2~ZoG`tCj7U*_OIc4r=Q>tUi>aj=M_WAKEae)4l}4R<zX5dB)BLyMXeR5
z^9k!(ar5LBgEzd_#)-ZtSt5fWmjS~tFscL_h}XJeo@boSXKs!;bpXPK2uvH^-{0YQ
zykZG{%IFciFKk_3WwaM={>)NJn5GFi3D$MRdtZ15hpTJ+=6k=1fBjp38)Y0(=hfRc
zX$KwXMx3#zqYL@BsP(}!_C-TP0M*^&iZp_Y-a9<<$gyt(kHuyJ)ztqjOYWbUr1;sf
z;#$SHQHjAo9gg+E;SO$rJRNyQ@d(F@Z-KG!^G?F7_2&sCvTXp?=H7KmQFv4WTHG`7
z%}NE<62y{6+Y8JxEt*Wwb#2<%^2xSzT|~Q)ipjzN5*Fh<t{?qbJpA*1qfJ21%rWf(
zH{;D4+yAB^GGe}SjMG?qbM|k+Ogr-h0HV$qb~f@-@DE8SWrU=HAx~(kuAypJPczo}
zgtacL*cDo;K%xqk#@|f=Yl4&q4AT)$p1*^9a|^5$wl1ja3|r4wYQs`1EUIk-s3)MV
z4QPUt6k{sLC1K13hjBs~4iHh)g*r#JuJUmnj1$1F3E7Jr;G2I=-3F!%L$_uL@Y9UF
zVc~vpl8Ubb9vHNWbvdDqBL-i)1l({#{t;sW0E0QN?pHS%X7%qjo;Fr;E6z$DEOiDc
z?N9>tGdLU}d_Gu%)hN2C(7K|@mW|p5D$k_3EsJ?UTmQLWw1rwdlMI2A6`-9{*gLJB
zF7E51>!PFM1Or6^wP5L*SbtAzing}iKe)kAjYUHO!?Jpxb@R$#RY;<q&n!EG2>_Xe
z6s<w5dLa@EQF7jSWXk1!nIt0>20m)cMlcwyu*P<C;+8GZ*Is<@>jXn{P~TRP19<>k
zvLa3#cxj=pS<;_kxAzi;n0t^_QicdMx8`2B&2{ApxXo5|-LAP`F6XqailMBoPv2#<
z5Nvtt_n>DtARFT@aE99pEZf#RLqN7P=Zpe80Dt_ZpJ#9P^Nu%X0Bwb8f*MmtVu{tb
z$~?LXg_@=fcK#HDEg09AxRC{K4T2}BAfaIiL9AJibOnNF6=Q0+`|xA@;Gcd6)6G*n
z9S0mzLB7tgB4`q+<I9NuVd^z*@&${74b*LamTh;rc>VJCo;JJyg<Q0^hyLJa9(>mx
z&~?xOvZ>wlQsniY<rPq0krIjy_*-B3TWB_8UGA~WXQ<71rLO>}Hy=k&TZwH|wptrb
zrxR+exOwsf(=<67#+@!FYWYqxBd{{87~06q$;jf(^Mbjznc6yg<gm{02&h&X6w5TZ
zDYI{q4jh_RkF01!8j~_-GQ_0}I9}f(4<ml%`WyHk{N;ZiZ(qNIwS#)FIN?zc_WH3;
z`*XwCz@TkB$Hn_|=7ta*Dgb>rT!N?Te<#k_0l>T!7iIU1V(6ntmOKTvrKTZFPZR#}
z8)A{{4Ce03dOA25iAY}uD~OFNX$Dkv(Z23JWdHN=ihs7nGiuH61NvCM)aC8x+9s{I
zDC{un-eC_lf<N!sJuDE}zyA3g;^wd1fT$}Kb|DB5vJ;_mkq^<IbP_5IZBr3UC8LGx
zd|qKTqpmBKd3N1qMHFbH&`fy%zyvZS3CM!0r_XVC@*GyEdDE&uC7alqQJ)1LBgVFN
z(UD@U6<Hc;YLIlmbYP)I9tMJxb#YeFkR^Ly;#Ix^4=NI1(S@S9+gCy<6Rk{qUe|@`
zKI&)dw$4J`&^Wu@4)v_dGNY8TvAo7jp5v-@paBIP4tje~pU3|649VME##rNbFm56n
z>M-N=aU&!Xx-`XuErmk0!s?2Ow2eQJotGc?qsXq*TFK(6b`L+qb0Gr&=90Nb)pf<P
zxP~;o+HZw`K%}WUljc`!8B1%(b;V#FD!OeG)BoyVOVA8hQ2qXRhRGbfHM&4V=NfT+
zVDZizNC_EC?`L1Xf(6W8Z4C$Hl34VSy`3FYkxfyX`dHdA_7YuaNQn6D282E810yt3
zpjH^#HaASS^)jCSZe6hWd-U)g3(8tB3>j(W_j5b69wacHb@oog^?RH{i}K#pTA{U}
z)rPgX52vTv6<`qO>!O;v9iXS73&*FroB-?U<XPYg_C3x$Kz6Nx&%@91O8v7QZwvqp
zyB1-5fymVWQ$-B~t}DjnV_fMyuJsPjbH(!km`t!JuzK{8ysA(n!Nfhw!1C=<1s{Fy
zM>stB4$AcvhU1Y@MZSXMjF!X&r5jVjO*gc7ZWQd%=|kWK@ZtA9d-^(ncnt(Tc>U4M
zD5?i-6UXo0cfmUPq8J-K1+&>TB2LRc02Rr7{8S9T`sADVwQv8AF%H0bdWl!(pWuJ`
z|NfuBtO7m&rA1UmR%2aPoKI)K3^z}n;OgcYK{uFh%mzaq2RwcLoX-Y^bzy3v)zGwf
ztDT{A&;VgPR#HaFsf)nZy0Xt=iG`C_71konBjmVTwh<)wqZda)$s_)qU;8)lm3O{@
z#TUg=axWHnsNsAJ5NrSh{haqH{9u+YAz%PGLs&{c(rawf>w3=+4*TEo_S_}K2Ddxc
z08btz=A3rpF2;xSg|15ncGudOkGLMy4OH(zpvB@BAZGWovFP|OLo7b-52g)j(GzsL
z#}74$ab7{Xn`<K;2Z6ZwtjCYnuXhhT1MI%(5t`7Z1koj>&HK;6A%vPZp7xd&60h1P
zC=s{w$tY>U+SrMO75594bwTy6EeY%k%r<z~s<7i;dQdbB6Ao9`$iu`Gsbqrrs;CPq
zh1Jt{mJ;W(Uxz9^Seq(LGsZID>TreQ;R*`H>iV=utp^a@>PLLi23xGoZmCkdU$jRI
zrVXnzC}Cy6f+lhAP_AQ9=q9c!Ugz0?2;4snK38L-c80tc>TF}?V)V!+op`toY-3ar
zYwtxSAU@1?t3Zsem>;pvB>^l!RRA}}2hfYyBF)F|N!z>)7XZO!;adP+4X;iAe)r>S
zEez9nQ$n7z)Rn1RCp?yAA-iPC^r$yv(8ORMHFNh2-PRUz+#a|`Fv_1TrC^Zk)+A)}
zMX*~dty^wM$;f5wMM80pWB3qUp}!XTIA9}105DZcAucOxbp!74o%?yom#u$i9}|ff
z#VshDwREXizn^|fea#EjPDL1Cne{-NduYWnSIpB2LSyZ|iyh?iN?C-NGpLH7rnWT}
z^sMuOx~{!@jG3EbcQ(=YHpg3k;Q&x%1AnZ<Lht%OI@*1_tm0Z<;<?@7iJbAg2%ePE
zJI??>^!-l0uvdsDu7F}dsT?sMe&_qRdG-{C>nkL8|F&GvT==;SAm7lV!V+6GVSv1c
zBJpi;h|NVV{_P=X3$VG}k4yi-hz9#GeT2!2!O<gp5jQ~fLF%RWee1u#70$4MRdW!-
z+7j-n;xOI7+8q~dPf@rlU<Ww@3dPG%aC>`;VI28-7>p%AQt6v%clUQKqIR%q9<ezi
zqWk;IE+RS8n=lEd$htBul^F(+YFDoe3<iZa-oY3*VHhxt2mFVB_20*Df9;>e(v-}K
z0O501Qh5zOyaoWpHJYB5x_GXD>y`M%zKrbI!QJP{!=DAZhV<eKckjvjvF8jM+*lW3
zkD!dcMdDX{PgepmVh`<}T#i@UcFXOxomEY2U$rp}ysGx5P#ra5_p|Y`|DBAwi@##=
z^>f&#KrioCoZ}_q5<yyTx%cNd9y0J<d-S;!fo$jOSjX8Gip!ytDp1=<AR#_^w5^P#
zoE1otaF_;EWT+@=wQlnPFxXI|uD7?3OG0)4OoLIEv;a9{JRFeAfL1w%>%4LksTH+S
z%cL;(yyQ4;AvA{o%9dBh@qp{A8ypWurkD*_bsIZkzA@*#nekKq!`6>hf$V{wDpI1b
zU$|3uMqgB6UNsxNb92rZ^3c5=Q}i27EEHp<D`+he>#Zn3Rw6suS-su3dAMHpRKDti
zHFSHg_(PW6>v%ABH%6WJM}1QC4jvY%zWwKW`u1UwQtSu*U9D|_OQfvv;J-kuh2loy
z0Fc;!UC6|@>Ry1oV1}QiKA#1OrirQZMm4#+)^LsT;Ebo~=HpqAN<tno2FVx_kW78w
zj?N=4W$I(t3tN&~iK%w?3GByVph5pV)K8&{=RE<U+J?Qg&5Lq8v$fxwCHFn7uLm6r
z`dIS4xs>8^B)`A0{zw4X&8nsV&}y!8=Yv>l#rZsg_BIKyf<pna2RgCW_Q$7j3f_&&
zHIw@eOI!MQ%e26}9c$bwpr@FB(c^Pq0G%c_;YMYEm|{qbZc{x2FbrsTV)u9^XFSgh
zHv&xReIZjq>cUFp4Y$DbhSeGd6ckGk6`VhPg&+RjAL4L4qKqSkX+p|0$W5|QwFa7|
zHf$R~Y+^B3T5G+16Mzm3c}Rb4KWRUU?Rzb?8=qxcis-fD0fW=GTQ4SHbW%Jz&YSUL
z#}b>S(1)wHB@)zjf~Abc+6`(597M23V}j+3p*4)-2sNO33%_7ugON0)?vE*54x8-y
zgf<kIPOf|6AYJRa(dd4kOqrfiVUb2kf_2vkAXvYqu2tz1Ct(~%48s8s!PD!v@b7->
z-@?E0mA?svU>Gyj@V^pIr<&nmaMYh0!Z%WR-*Q5BGt}2eC&SLGYd<bxENqeMYr9If
zL)eaz_M+h3mnBfoiA}d85Tq=y@MD7mK95#Os3;=U1c+N3q%J|=`#EY_kF?OS9Sa5o
z=cZVyLM<a%04Xmx*a5Oy*+rtYhBOqg%C$L3NQNT3_xAmO_xJy?sP$rQn04kWK4;J5
zT{3&G2XOoEdvwnIxsNTq9p7pGT2y}uHc%1Ij`O1+XSR7uJ_o6dHH6?jHfWMZUhygc
z(6<dS+Ftcyj69$W2kMp96Pi}c_jg$D@1V;9)7o`R2I=j-T2yrV27H`GJh{2W)pYIc
zxWsFx^dcxH#@Yf<@#^I0(`vN>-lG(1SeG;HYqA$3sCOU%vjI6pcLVmLO_}#AfMCg7
zUqWY2AQ*CzeNd7*DBI7e*O<PGAnw7x`Iy>EdiLPYBJ%OivE7b-5frL!6;Hb+2nN^g
z?EoPCSY5T&1EpZ-w>3v0C{j>cL-lpWZQ*>{doZ6i4?lj7u+^%sQ7H+Ss;_8(rxh8<
zH&aA!&!{+!LsenfKrd#3oRTt%fMp@ZqP4K{j<qW13}kBmdL^WVG-C5a6laxtkO~iP
zT@G;Zyd3;RdN=~2*e{tS3)$2b;2N142K)dkmvccW1>?x{bWRyn8jJlj04Q%XTZgjP
zB)c0204&bZR=0v#Yh_!_>_4{)JfcK>TPOS>8W8PEmwCoK&sgU3|Bt<Y`?YP$&ik-m
z>%EUL=eq3cxt~*2tSX9CltfaN6v?p^JF(-~aN;De0mp!0_$hfw93&5kVIV<(`~e9P
z7)J7vhZwR08%b;_hGE3AO;I#25=Aj8k}6U~vFcQv+rF>soO6tR$wO<ckFnO?r^u>(
z0xEz9xohvU*1U{MzkKa$U;7&Kyj6P`#pX(f-C8=^A2kEm%nwI=^TDElIw5t>kS`Hn
zl*+wWIBA(pD<n(fdEJ)yMK-5#S2`lJjJ_ZYO&FVSc>0{jzxVeUjz;2kWD^qU?j1r(
z9XNT_FMX3<>yj3Wh})^)uEATSfY;33nzaZk8sA^n-=jC^m5arV*K}aw3)%wcF!l#a
zfrjp4Z<FMu{mod-puIsI2<WGWz7S|HQ2`QTRY0aycr%b0EGnvKWm$?2`jWZdM^dbe
zY3R|tMzEQjXSJ)T%ZBCmp;|;g&of8IM_M&19&NVH(0U;a4NlsYf8m3FhM)OUe~zg%
zw&9dN@$gSjDj<PeG{uF~?gh3YZEoEipMBV(R9DYvdCw|Vw0AAwPsAh6<qbZ^Zyf&E
zEhy1-dP#AvU+!3Ub#gi<tUcR}SKprV{(M*G)5nx;ip{iFee+p;mo^+y4PdElam4p8
zGp57Gr~c>`)6(zF8icLB<rXv9uXz=@uLCN-znz)69bB#E;O);}zvhbofVF$i>0DPU
z5j(fmdO=n2#_xToXqr#`>;_U*hTE$~qX;DgHDHc`oEt$Z5=~RRx<}1!N?&*4fP_S;
zn0B`@=ZR^5O+D-gbw+5^VreT>cUHU|pY4vvn<F0FyT|>z_egP|mZlnE3sQVN7J*PT
zU3hb2L#tbZRVi0Rj7>yYsFDeih*5gbDmWl9pcfSE1?&8g(m;ws2Y`ZC(pmsx^UCkF
z12j*^23e2&Wu4TvonZ;IovB-mmD|n_OHrDBsn4Bd28I5N`ul85uY(p%*XCvZwYIA0
zHBDt4ps8a*UkAmDnr>cukiZaq|1{ls(sj8y7`D$37Le#z6!W-SlBcynGJifX<LHX$
z*!Fw7s*pwKonUm2lm2<V#T_W{@5cXD9M&?@tXdb{g99pmTyCPdO@xT{(KWw5DgC~d
zQkLg!bIq6AAfgPPEQNl}#)Mi@0kvA|xX6$OhNNIHC)G0t^t!z;GX`ySH<P>PCsGje
z^IX_k!$Bb;WcuFGSO^JQr`1@x_3<(1N|B|-=CWR|$)mP_{{C4%{_zDcfI)@Mwn=HU
zQK<Dm<iJtA1Yi>y_lC-*SRYm*O_HTkR#Cf0oMH=?2!%=}Qbe04#~{>c;_Ty3*q$D9
zeD@x4JW`tZ^cWe^ifzyZ)vJ?o&b<hvPx@8LcD*@uP!|^co*xUFWZU)n_p-o+W;@yE
zrG3|b_l(O5Y}kotWD~7j4jPf)Tm-S}wT-vVL>j+y_PZ>?QC6R)F-3@hjcoeFs<ray
z)6dvl@3?#C4x6K6_3g6=DJ9i}4CXPl+&}fVQVpM$-q&1J<7AW7J9(rbaddQ~#+e80
zB2cPVKZ#yl?=jri%7*{SyMKnCeDJf(MYVqwXu+7?;Hd=*vg!&T-##2mts>`_0M#oF
z2YW9AeP{n(^)8P(ePL9cYsYEp;Q%z#fIA@7OYu!gU(`X^AL^uUo+rLp))fG=-&HV8
zS|xh*VBhp?e6<z0vw!=T3H_eP>iiac&4Sg`YND+^-|tmlu*Ei?)!giz5^g+`CH3=4
zc?L_6_d18RgAGQsDxXkl?F_e6JLkUX&^LFHWNXKun@Fh4xhSU)1F2+cgjyrR;H~Un
z%6k1e92$79?<BOK6~t8nhr=GJl{{DGd?3$zTAmn0HFcU6)wFU&4S+Rk*f`#7xqtT#
zC&#By0|zR#&J2^K2=mliLE2PAf1^x<RBQ7U^u8#lbOx{@!GIful9l;XeXtE?S=gLp
z%uNcBm=eRV^mSG17BOkjQY0X&!j4)SB09H<GfQ^w0-9=5P_Yz;sa{+^!$C$3wU1-5
zNUh5lXnl?LeR#c@`)0i`+{^hX@g>SwRrm4J^r5cDjaCY!sSHEyVGMxV7#o5R<ES4t
zWcIO=^8ganE0U$C&r8Y(4)!#Vm)6uxqH4Q4&6q{vu+e9;UK_+ZKZ&hJwRT1~Mf1CB
zVg!XEoYo+*1iXT3+{CPuG+-)PNepUr)q<v<?KuU@Fg1i2hW;#qE$~_&P|R*6;J%Cv
zXqwZ;JD@mF3L*A&A!4a+DkTiSUJLR<Q%$M_s;G_F;Py*44r0{5j|6rH(^A%~g+$Vb
zQLGwTh!U-*c3M1SO>8X&qTON&5vEc#NGvmRwc3hO6iBZ^4q4sXKh!?(Lyj+i0n`Ve
zV&|-bPbx?zrOKVCy(N=0jzWVfR1!4|cB8?4wft}tuf+yMQ<JTdMJtSBt87AG+U<Gr
z{U5Nscb{;4%CJ45j6+Y6!9vN0RYZ`xG3K&q^wh6!B8U#m*1B!Qt2UT+>;CtM1AkD`
zKljGJ^@BQy(nU!6f`n@ag9)$2*-CekTOw*lG(B3V4a$VX%HKQxj&A&QsV4OeDp>>b
zXj+6I!qN7Kdw1_~eR0WQw<pFyBRc}Sd|*fkAuxvtSkiozhHHIt&K#zLTEi4A5ZUdn
zIlsK(<ao;mU;2RKlVbpeLHpa1p+>bHXv@k9z+3SFCnxt286jv&U!qd@Wk1pI=+8B3
zg2+w4?$6D^%YwoJ#M&3ASIwusnDvkAPk2d}_zrn%Bzqdx0hsu+x;2h$KQ!%&+!#wf
zc9!pbdECA7@W{PcAwc)7THf429;zpr=f*HBiB(;k>^oPiptH3(V-<j@1JxyE=5xRc
zX7s#ES>MYgEbdYP51LjWa5>jyUfl+Nw^M%W_q@(==iX<W91R8r+{&Z&dnznxAw5;G
zjjvKlWu6O1b7nuy9CBtVmAQ(xh3f@R7}L_owKC6>3e0f^CM!T1M$|nhv!7??`odDM
zYK`O_6$aB`A?o8cOJv_*l16PpijmW8=I#j`4?DKuxiXS@CX^XUp$%Supq_N9J=Ds4
zt^N#sp6Q0>)uhUpxE|8W<3)rtM8<Js90rCI7*b+L>O0uI+0vkFRyUQdEB`znSFN{*
z3oQ9iuaCEfuk&mfjB0dkK80rL*w<iDl+ovGAMdTtSbT}_z)2tYvbVIh?oVM0jXl>=
zG>xqW>NC?k;~jngwNVpg;uXwT#-uiCorrbj{?Q3fE`?lY4s+$%{*u!>k(PC06|^t}
zV`*Ja@Bel{>!VaqFS<0QZqC0x9JD1%m!_21q>=47G6eOH3`U80ccesE?YZmV(X^m>
zrZu=c%se>+#<(F!V9E#HJT41Q5v)C-KigJU>sV1%yR<%UT<Sm&m4u12k4pgBXeMva
zY}Ir)usylt=*~5Fc6;VrwW1dd#GFm`EEjl|tFS-J?550I3UPq#X2+d7fz#v4(HMv{
zS}r{L?5K|opDVRFgV`)FSLV0ntkw5D@G$UgSfP(y+ynUa0^Ha40c?WQpn9?@Dtv4i
zX?u=j$Ae(a#Zd!hwLXYc0!g$}NNi1YM`+aHT?(AVX=v(U28L*Q57Qj%&#rm&y^k2~
z-eq%q#ITLTWM$_yP>WkB#BMpF>4b_5H?O6WWR~gyZh*qoE0^mJey-9fmA)adY=DYs
zWcdRN2(kZk))TO*S7R<C5Y%fi%93{WM_+1L7!@%s5!EKYrddkOc;Qnsy;e1Vp=n|j
zwn_-xy>|~#J;H;B4?%?O_K552YsRBvc9&XMqMOdX>n(E8mO~+kRqQ6}zOvh0bA7$z
z?wu3fdi!lo@7&Q;MQv=3j!e&{i?OpRevewK{Pge>+`adZT!lhnNR>S8h)wUA`!nfc
zrT4V649iD#z+f_NTJy?Xu}SCv^7bfSHjw3c-IN5ZuvtG;RJ>+`7XY|U-;0f(+rnq*
zMJVd0**!H`iPY_bWx=G&SvOrMi;Z_@U|$7PRR`3_O)X)>3I@b}=bxu3F%_8CD)zaR
zK&GuUDvhd@%~HXb8CgA#{<^lB4zl`hojf-H@XLUHd8W5h(yvNG|G39L{!jn!xqJ7u
zKjyzY{>$UXI==M?{`M#O?;Y^Bz6X2VR(a<X{wMzkeChws#%F2pZ~cGZ7k_X0*(Sli
z`sd-#y!S_UUSITabvg3C{n!6V_Ej?h+JL~;vR(`l)R0{iT$XvZc7e04qZ&{-HzcBk
zTmN7HKp@sYa)D%o<0Ckh%yxgy1FS7rgV=&Lg|>Ym#>POVw90@miyGg~0X%|<BIKrp
z1681s>@xyFqE3bL$4?l)`}a9M8Mt@4Aq^v;>hH^}_60G7ZdBVU>Siv!rN+MDARfgk
z``7yYU3((69y$8f=FAfaq76HH1X=}<NEM~VmGak5^({&(m$fN-t_4qx_y!tcU_t%*
z&LxsVH)RV?o46lV0tTxRJW$F^3c~_w%}YpuNub*|#l*?IyM*f<yTgugvq8qhCY}<;
z4VxG_Iyy!qaQ^(6rfkN+e9ZO?LuA_Rp%(Vjj&YN?d+#mo+`Y@N*>qFuFsN&N%`<8v
zVU>bmDRFPD@!-K*Y_`Wp9LX~bEs?@VE82!>KpA3gL`$0PScCVD(K_=Ap)FPo8mzDM
zfc?p@%sIt@9Ud*LaM!IsVI`PfpM;1OJuO8gOQh*Qw5|YcKvNddRv*%>RF?C)d44ul
zjXib?nbunWj^(?ps;~rd4a~JJfrJfUiB>_+Okq!L6H!ipM5{s#+Q%}Ns;Q%<W0*E0
z&3x!?RqJ^~E!{+U0c>k6mYXSP!Ak3MNCCTBcPv_&^*vwT;~)2!renVU-9LE)R90g=
zR!;=gYlR*lEP%Ng8}1!$mehW6cEnDznu&B?^==hWMnj9R)aIzR!Zd4pFlR-uz%xec
z8IA|A+QFgV40Et_wKGGY&eF;`Qi~S1*oEosmLAk%8&7vz@V==&5!Ufi{}B5OD^yVZ
zUohrWi$7dWLU3-%-{1OYw$&I}e!rf19*iw)t)EK|Cj7qit67ba&lxW5rPs-7Es>Yr
z*z8<;U~Vj8t=4hLt9;vX0ZmW4vy5(G^79{^^B;frDG{@DnF~|Nv<4po|Jp~wzw*r!
z9&X;y{VV8qsupcIw|46Ud+h5B?kc#ouO}GG8oO1B73PSK0b}=##%j0kdw*tjnU-n!
z#ZO-FSAOpq^Q``{KCk~X?(yK8?^xPXP#=)K!}h-y=FGmKt?pmH-X(zacWPF}o=!D(
zJ7=ajbC_qddXgLW%Eyb(<zjl&;%4vtK1PNy8LJ)n*`NJHzd9(vqG|H1d%9_egAqL~
zfTTz(jcLDAKftV2^IrL`AW-|+M_bokSK;#dDd1W`4oDQnP2=r1-{RfVhop-;oPW0v
z(qVaRsYvw+vpfj!Di^vYP#?YlDw`BNB_g6N;+i(Io^}VGeE(za-n&P<{|0e9X1qH<
zYSu|%wK$?cBNt^xl%}qTH>t+{h;6c+F*4Ips->nrhbR(QO0<@R%ohXfCACWT)522^
zwrIzyi*^uAzqVQF;TWQBuBDo_j0tWH&4_t}HCxzA5jAa20|(oXyiv6{z_pENNe}C1
zgb>+mH{`>CX+N<ac5F^gb%Ps5yfEeX_>?#()vlX~0}o+$I1ppx&9~mxt|Yctdbb9(
zoLS5)`=SU&^Xp`M|4UyYCQbdu5Y2_2K7nhu6j(7!tQYJ>r*~5q*r51B@K-jv2}(mZ
zw^dO@^UX-))@Qu}?DeO8^Ixs6zCZjBZr8ZpIP=ea&+@k~=!=HwRYz^j{KKppNK|c(
z%|2K!?$v<8`g40~dCdSWK)P=()n~*42tN1PrFG`<=Fb;4Z~@*onZR|b_>0g0UR}Py
zum8%wtc$*}sX5QoT>9JzBE%FKhooIWBxvvDte#$Zp1D5EOp~U7bE}lvn2L2A5VPaS
zg~NX4YPV-DnW+>~OpMzj#-mfxFhDKr_B+lm&$+z5WXdyTHu+5q9BoF9w?`Uiq(lpe
z31Myq0z=e(v#pY%kQ&HghB6~<Ce@5&=p9d@>h~lhox>8?j2kv7A!yM^XpR^??W)MJ
zuY_WH<L$sOjBJL1&1PUbZrF@lwwohOMW#^$Xpvrs;%P{*prOxW{Vt1#SK}45YIvd)
zEruCP>f-B95Y?Rtta`V?%}Lxk>vE$DPY<AtEC>O?HD+!T1G(wiq@6o3n~%9$eRy#k
zt#-X<sg2q1qd=zd>xbXq)9Y^`(>3LA;POy-cD<uDl^*<NYy9#rCjP?i>)d(cZNll6
z+^Sl&3?m-ddxk@9Edw#eu3H1TZCMJ!5Cda~Y=+1<4s13Xjy7AyG?;ygj)Bm?4Xo9}
z&JwatFqp2mHeomK`L{m%D?HC5`|CZ=o}V+-$`~V$erL<@oBtT^zx5sx2c%lRW6J6|
zdEiL*$J8RHK&@V&5lri9!A^)e&rDw_t?1|G%+>W3&(AJ-c6P~r(kkA<;vx~z`|Jgd
zhr^!T^)*v&Oj(PyH0?ZadgquscTTu>yybuWzxhvj@Zj2lg?2Oh*w;TZmKR}2iQJS0
ztOfu>H4kKG37ZwoI&S)-W&pL?bI<m4!&kocCg1$#JG^@@F+G03hw(t#Ujdg^x)0Q%
z0HpO<1*|q^oAhE}LTuxb)~I3{fK`MtpuL0lXFDE$_hZuC2OJ&W;dpyQje!~~C6ulu
zifF-sr^F;AFN3m;X;szhn+@hS#1frJHy+iq?|<K@prl71{Z+TpS!?ll!Pv^A`p8yA
zD|tggicsguTxTx!J3iYz;@R|!!{NYwzh|B%4%5WNe8I0zf4c*{%lVw#_6UWz(USju
zs~hH!@1Si}sf0S3V;jZQ#U;~}88=5Ns4WI0YlCbZA+oxd*OJ*^@1ViS@rk+M7eaDx
zCh3D;Q<PbmjA|8T1S-myoTgLWefU1PKy9<7i>;$TtDPmh0sfjzkXry_VVRa|e3>DV
zMGts`URYk&UjFERAPyhD;sJrIs~YRgOTTgG1(jNv=E5e53Z=z|uZzG9l+#wiXl*-1
zL3?yM0Q1Sh0{Rx`oh6=!1-Siz*C(qY5iyM*H-Y3AVf_5^(aeOmbx<z8p*o`~Dmy4G
zATi<j{3h!BUs}HLd5Fp+#-ME9p%l|lRWR!guX3*;P@1;VtgTQh?B*T0&baY1tIGah
zuTMdUX@D5}rkH>tff-!`TUK3PIqXSP1_M};(%B6`tyn@)iAWkElA`u@1!%<rWMwvD
zh!DhDIhu=rF@{4L)Y2&>vz&2Le!5rO0{oWtX&Auo-avgF<%UgBPyZWOZ$*37R_ouZ
zBtX>u)dfI1x^#`W@kYwl-PDQe*H#D#@Xjo{@HBBIT6%CD%u!QM%|A=uqeV^4s=aOR
zY0hs-HxZts@O$&$(@ce3PYBvc#_wIunX9trH=q7C-@KT3_ou&xY!cBdUt$<QOxj`h
zh{ht_$;FB<w9>T&?YfadViSZ-3XDT!GbFY{WRrv;2`MH*GNa<E&nv`0%i1TrwSqJ&
zexn(yocPW2-{r~SIW^Dhc6(i98qBrwe7fS3Cr^0)t#^IqSpaT9qXBX)4pUIC%es#d
zZB5v!;7gtk4-H&B(2Y^;Y+8zpb!0nj825V)d4{r>b$eSnwHWiy(r4xcVGSs`k{TTL
z2R7rtE=8&pBUS6{(QKUzG`v@*h;~=m4kNW>4n>Pp%l-g4Yvw0ta2cbA<HlgNcGFAt
z$2h(KBk0EIY2YWn@issEXTQ#uf8rs>DRXt&n6A&MPhTMNNk_?Ewk`I*%{MFmNgtdB
zj#w7kS}Qepb1vIRav?ThAaMEVbDo@h$nohN#?6NC@E&<PXxY02l8B|Dt#@$~6E2&O
z*gb;$Nwb_ARE_ki`wC%;j_2YW=q-X%F_umwsQKxNk?_=9@0}R5Vvi~`iR$^48`GS5
zeDMPwoju|gU;F}B`wI?-J&&g+oRza)Ve8eM{(q%v&tI)Ls+^vTgs{~v1_%uW;xq_V
zcjjh`kT~G61zRJoqKrq<);ndIm}YJH7ElkrlGVpAms!cRTr~aNAf}~9WDrr?856;0
z>;6$|K}79)O4X(KPrmijeCh6&L82K%t3chWq5W`h_cF*^kJXpKUU=1O1=}!u1HApd
zHTSgq>^6q^VGP&*PmUgI+|nn$;@n<!)T)Nir3y9Jh*;tmi(*$SpU=ytlfoA~eeLQg
z_*Clc!*C4rhdQt26a78*>9RZ<2Zw9VPG2)#`)JshYpZS!%ZV@FXMSE;$FWyrE{x>j
zG3qIQ0csFJwF+J9k=omw*=usnW@`nx6sCF4oK*<cE2Tw@%~|Z9G%cP;Nz~>=gxs{)
zqcjCcd7fw`TW{v20u}VTS}lYUAyh2{2qV2kWxo%tcV>vz3%D7;@#~G2{?%fFVOa}N
zRuK%q_W;Z8oqN6M%~tT-H$L%!*zdpg-OZZHwy~nMC=H2fRa2>EoV+|kS>N#0xLiML
zcT2#S(1kjzEO{*ZJ#^4OcQtU<##t;`%>#qmD|S(=msSjF)29Z|S{b?-2VzLfb6|fs
z@bP=ksgHh-yN~X3=e;+{QO~0`@0f8b{Oy4n)+I#2@PH6fQuc2gNULYojk;qZneB{V
zflfmKdyEJ=wrut3Yte3h!EavtUFMwGUGLc+Cf%EpuKT$a{!aQ1e`fk7;b`dRyCzMP
zjkcb}X6q8GRZIun^Q!1QbF!$`qQxpf4Xx9lpV4w9)IgpaSJ!)Xhe?|a<dqS3qj$RU
zwY1Oq)p{`30s(fr$Pkun#9^A*Pa4?aE)d-VQ40x&G_V;&&D(Qf&YI$P%eG-i#H2fS
z=xa_T4S&=OU@W}<!CU<CKlv4Y>W_bk)BBNpxMtp;^59D!5br$(eofmGiKZxmrJN-A
ze73T2yV+J()hCZfgi7;P%}I!wP6q=%8;RsY;l=kq;rPxe+tXu?Mo8NS6pv0DYNw*%
zu+qmOh_%IWhR*-o5T1MLItrnRbuaM>uGKEtH>6-13}d{4^bJ5P)tpl5*P)H!1Jg8d
z*dMsMy5h-`r~Jy(U*^}(f1UH`%m7eXrP~@#?HBXm^D4C3YPS#w(%A10Y{tlt40yK1
zUcplXE*Eht8P}+c!@y?Tl51r%Zb;)uE0w8M^#+?~+MLZ3R@)%uT!<l%Ytd&LoQl`T
z(I-yH-b=RjQvTYMey<H4Jbarl98p{&*bW+9w`J46$sYX>M6m)4{;#EK&=~~(osS3w
zqOCdXw=v<YW8lW;`4IAP&{NuN*#N+7E7sMx_P?#Ug@~n3URt(S7t`PP67uY0cVS1D
zF|wrl>CtyX8K+#JWJ^g#6B)>jZ|tB(mjCQUJbqGskNO;HzRRuQ^yj6`SvD;ZWW`cr
zpj5<cBl`V(*=XH^tJgnFYS7xvxu@SkXYoy_tvyc74BPM3im{6Jw*>W2t;S3Q(Yfj%
zETsWy+LF!vOy{B=nt9a$z|-L&1ohKffuq_g<e71MOdL1l=AAnLPu(l(Cri^JK5?f2
zHK%PTOBvElfGt66{gqC&__K+@^v-A>-hi7sr+$vZFer;;EiTm?Zjt<S@aDm7T{cZ+
ztaNSd(uUf4OD$<;3~xro718Qx5oPO|(BjSUQg5++{pqjPmtI6A4J*gCh12M5Jn-|i
zLO1cnie>@K>(AHyBAPvkj;pdVPfh`HP!_#ph$;5ZNhuO@qK)wQ-7|jY$?xzdQ)K(*
zJ=&<PH&;{hCg^k^qyN-&rv%27G|lTe%@`H%S$B_hdsFNh1utR@1WAN|+1@l-jMNPO
z(bJ#j<HM)yb_Xu6_Xsc~eLY#*pPf$^Ow%>n?WW()#ZAHGSZ?-gSt!dehSo6uAsdoL
zG&MmBDolbjB+XzAAQCw$mE-Fpc87_1J}kr&O^X|?`dzD!E%+3jE7itcr7%y8{XQ~U
z00asiYZKPpK3&ov;%i+qqxiry7xw#|IssJ0Ov5mc2HTga=`0Bh<Lg1V*9L%h?~i=*
zr+<QPeDf!G>+SpGuw$A)HXH7|^)}<(4?wWu4k5K-gIztK^bVL9If&8n^oS+O%DB@R
z2uU>awNhIYpfwUOUGI4DgO515e@fd%PLBpsI5jg(1%Mhn1Y^8JL0~W~hv@}+gzgJM
zs7hyBYDYpdze*LWF1wl-&`cJ^tIK%{9_hNlWmt@+MW}P(`r?WooPD2ffA($u_VZuo
z(f$!v`D(FRu_B3TwhVcm$?iUGDb#LH0k}bDkjl{}a(p~6&uYEk3u~={NX1*btq<uT
ztj?(@LDRce+Mrx)6M(iVYS9fk#-P=?qS0iCv9nW+g(-0cP<;S(l9r}c7{@2~_+xK=
zgIvQ3fYCRIo7JAH`~P_$u%rBbSsjin5%>a_SpWW2pM~qxY!Q@Y<6TGftu3Dm<|Wpe
z1tR{nt#=4nTD7@;=GB+e@3_CFEq7q;=I=nz&1H4*at5k((Rf8!z_~PrsQp<TnD$Q=
zeZ~r4D=nA@S1a5!&81hF_o5G$kKqrxuC{0hmuJ5Kz+fF!R?Qar_xxef$v@K3=<Is6
zb5WqP=)ojLOa0otEVZvo>uXXioMk(LP*J8Z1|ih|ArgbOv>alvw2@wS$%R~IN||ZM
z5?t1**$wZU5!~uTnVn#)+?XQNM(V&?I*_9YHRs+nsrY!caa^mVi$fq!6FDU|DVlcK
z##$9)_aj40y?1Y=C;&{U7k~`IhRtT9|7|m|&pET!>A<8eOv|g%{mt%eU;lCjGr|s5
zfv)k#vR<qMgU<f7hC(1%0ytwtUwSQl9!ZNPrCF~%%fH0{U}tqL<s3pa%N7SXz9!k6
zHt$QJ5v>T4U?$-eO|w^Jz0|8UQd8yQ)2D2{^E>?U_|v3^cWE27yI49;czJ~PGRDAl
z&{l6rLAshBiy3lD*9!W&srQjyAR+{sRH|6&T?EkmHn+%kF2BcjFFs=4AGo@@rskpy
zV@lmuv((DR^(mjVr~K6LCOHGE`wwW!e`UWz)O#O%ya)?OMN`3Qu~n5jN|7{<eeZ4(
z80O5d(R-*;3TvTQE4Agm`czwNW6&;IKChb<_T;%SA10<W^zl2-nVf?a!&M8gM1vDF
zC2X}%8cQ^_Qk%^t$U8eF(_2bxeSI8;*UNCcHURwKOK<SCuYbT>Z$4y9i8=4J^;jI(
zoZew`@&JH14zL}dWGw~})Wi{bl}q=yv5nBJF9OmX==~uS$7Lpp)I*zEUp#uu<KN#f
z+}W}{*>IEwf+T8FO`Bm(dqC1{RV*y0BOcHN)>$AN!vqyRiy*fa3|+_|NW_gcy|zS*
zZA1bw47i^D`Q;0K@Ztyj+T*{;x1N8C=Z9zXmQOYabIu$N2d?)!ro(~RC~0SPZbX#*
zU)?7jnhtjN=}RdpAPt`K34Nhkszn8YdB`>WJY_6Y=!>BJ&RbmoU^A9iqApxGWxrxn
zTkF?Xa_PXN;g0!*A#V5=|HMDfUwr$&Mj<*Spi#Y5lPXgwUB{=VyaHu^n4|fqT6Bc#
zsy+YTKh#GxU(8{tPFsDpe_}IDS#Wq|3vpX%+RQU6$P$eFsE11qIxzB%7P6${n=uT_
z+|m?qDNvOTK-*Y(0OkNy0}ZKOExc-RWyA4?l)~9QW6VVyxH?GH=O(gjvbE}>x<<hZ
zx8guI++f(2XW>4DYwg|Z90or9`^D^;Y^_LXB!p;1BJ%<o`k=ZMRP-ULnyxVoTK%3g
zo7%`tn5(udOfhP|;HH_A7%dm*0K?dk*q^lmKL6cjtD^lF0+EUo^DPt&cnhJkLA6zr
zZ7jWQ*Z*#{lBbC|2ssAACRrtKLZUL6Lkx_A0z0WH&UZF8SQ>a7R0|lRw;lxAZFzx?
zxn4-HdbT(IhOW_UMN_{j-)QQkXAI=+_pXbDmf|x_ulBoZ&uN_^?)`MDdERy~>FdUF
zp3b&aBRCYCdwt<zbnf5hKv0d08-Hj2EnL5bsJ>D}4Wpg0RDd5%r4%{bpZVSS2efZj
ze)>=SX|_klG$0T=bKDtK4`={HyNPU4A_ZmsQcS%|hOQYA?-rs#^h$1`H?!NBE&iP_
z&xPMO|GOOO!2nU|KD0_inDc=JS3j5@@eTQ=t^@eGMgNZOjVsbKT~)-eV(TQ1h@vXb
zIEJXbjMF$0f~S~;P0nnOkJwyX662)Ls)^~EoyoV^YsFQ`@QMZ=D)VZ#h~_-60#T35
z#mvkdOLUf?ua`cD_q#p&>mBnns}>l*P0)J)Dh%n57y$m%H$UJ@Uw((vlT$V4oe!k2
zrG-M1N;)|~5RUKO;pENx91aJ3hmbzYO>8NPS*1qV#>h2Uu_cz`WbslHr2Cd73FLXt
z#iK_&`|c6j;|=4uB}k+W8){SseVT7-AkmayUfL`4A_@!hJYLdYhk#U1C9G3SzPU*U
zv>J&teXVVPg6bWv=N-TK^tbqhAN&G;@A>a8K*%=gQVWN}j?1encGuS&=E*?e&EpS-
zJFQR9G$iIc5#q)uX)`x%+9nBnueIeix*d!DzE#~&;}A_G?m8!Zd2OoW$TjPP^XjG0
zKiG6Sm3f*dIUBv+Eajxn<)8V=KgNIOYyUJM9g$u4-PFvYHt#LDGGgv*k*5s*5TL2)
zGqqv0Yzhwkd@=vS;4dFb9oYD0voL?lcPyO%{BQuM=du8W8yM;}k2O&0#V>0#)sz9O
zrBMQCY`Ox*wVAHVH_zKAw%QEb2-bh&V_!6g(4Q@g-J(_Ww6FnwZ#`u{Tax0ssW<Tr
zWLXGr6j*&hH_xl%Z!>?*Jvl={B-!;&8Z_Gd>Mgan2~<6W7Nt>*NeiN5I|Jn~lhBPf
z1tEzr7a^BGOS(Rm)=V#pRrl*WaoNQ<jvO5w6NZ88;2RZ>hbUSFpCYvci?0fqLX_C=
zX^hd>TnFK5<~vV&V%#tc8-^4$?bixwX;yVJi-J^RY=WjPhhgXj@G&f4*2^yxbb9f}
z5`c+S?ykW!%>kMDQ+kitw#qP>gut(-1HRS#^Kn>PHCeaq2_fh#SOquB+3MYp)-?49
zIv$$Bt!3`YjlUy-VoehqyjTA$yC@-%!a@MhgE{TQR9odRWn+dc)o%d>!0uk*_dfiL
z```bBpL_ouCIo344tdXPmC<5jvQag)o3zq*97knvV>GR1B8EZ#SPDOdL`o|Sy8)Ai
zX1ogKyIZxV_b)&G75?toZ!=F5msi)S#Y@_>p;XO&D412!<hKsr<)3VS5$A(!1Q?L@
z<`doneOVcrpfW8~JGXE<y5^;%*BR45iX#opB&5TEaZGH-#E^u;LC3GkKFU&U+>F*R
z$-ZBI9D)*CSZaEy6zGAEv}hw-*R<;UtxG_HVhd>GT*%YJG*_m%k_)7vb>OQ*R8;8r
z8prE1fM0osH{QC>cDrF;4@etmMN~5v0<%@=9NjtPgP;03-#<I2on5frJbWt&tQKsW
zyImV-!^<P3{zu$fY*`RCF)>~3c=F*7*q)wfIyxoVoqM$HNDG>NT@NgeaD%0Qy?5Xm
zrT2xqVJY26EjJet-RG794++}MWI_BoNdMm-{U`kVCqK{C^%d9q9kuA9rDnZ_tLsai
zzj)4Wf359;?89FL^ga*#$?CVYHJZRUj!cI=C1)6-wq-yN5+P`Wzw}<zr6^d73K6-H
zo<w|6b(YW<$*KcBtn^W~=oVvj%A8H;>Q0Ju9pHO+-{HUgC;nSN)PAB8aZK9KRJ?oJ
z^?R@y1D$=lG0I+j7&hxs_|4zhb*`xZe}=GHoP4a@WU*kC4z0dOt?!ecFYEH_UJdPb
z4DItC@GFn?yjs$zuxxzZWoKbND(lfws%9N)TP!VVb8Tkp;8m<>$Hzf6os9c{bvDrF
zwQtDKB@|d>zkXJZE!Q1D)(lMlw{M<xnVWu{>lSjae`qD5d#W|@dx>q#DsN~m@;0#R
zfYE_qkmbF8wZRGzbS=|w%ci>;toL^#YP-V_$R(=Syj3Jba&2k@vvNo9do+Z|X0uVP
zv6(#<HwIsUp^$`{MP24WQ*oF<d{{bu==oY;D?%%kRvI$seI8;WNi?&1>8afinx=n;
zkzqVyGaPBhkLc`nyE$Jv%39_~UnD%(TLo;**uD<@eDhe{>xLJ&w7zNgvDF715bU}f
zu%WX1K9^Q;6KJ7Ss0w`CkH*LPjqz-2iAn&>m(~g=mH=mEOkWKY`W%&I4+=4w?GGS<
zQs-XK;<rvDaPf@;-}&fw`R3Wzx%ci@xSSiwtZ5oE5~5{MbUjs2KJ?zzDko5n#t@RS
zY9SKiKpK*^?+dcnnY9X12`$os%8W{@JUg86cb@(>El=!rJFX80;;^y(Km(bodaBMm
zUQAc~{`L3y()eY2{Cch;Lar_W(Snb4(GtOe=C;f)?^~<|EDl;DVpZouQf>S3=9r`H
zmhHG>x1VS=EMBt88L3Q6RL5)nNcz1c5aMF;yxyC%sY5i-9lBkc4PxChQZU^rc8!Ir
z-Jbn4b3J8Fa$zWX{<a0FOT*_Hh5g}<*9L&^fB6kIcScGyp)ZgGN`xE>$#R47<cJ5~
z_$t>gp7HEkzfId`BnI|nqJ<4&Mx@eO*Zt|t2q2M4WfH9<8)Q{S?9rcZ4l+Ou3>5N<
zYo31RyBr-Iaddi@FmB*z3q(t0%#=-KGidbMY=df2)q<vi`tf<yo_<FS#v0Z27Ha?r
z+qeX(G(@VtJ_$)SOuNEgfAZJ(`A>d<vx{@~`@JnJ3ZkwL*F1mzjPr{Nt)^WcKN1k|
zxA+39?~fs|9go@XcZgI%($udvDi0D=yc)cU&`f-7v(+LG)8hn<*c$3pRGUYubl}%)
zVtaL;vwb;d(^pi}G-hp?#VP;l&-@Sh>VrSV{-A(KaJz&`h+>t@OK-=1t$vZl4>v4I
zb8D2ooIBHC_h)5{)suY9SznLkTuiL5bE$$<kIa=Hj>c4#=4lllC%#Fofyy#@*KDB3
zO4yDA9+v0m;Y4S@Zf2o$CsGMQ6(QHE?el`NF&e?gzNflCJp{ehSyWatw?=Bdk?WI~
zi$<aPHKKh&MRi`e&S*+gEiQzveZT{TX1{mVxwfSvjc<g_d>uj0$uWVO#rlS=YqN^)
zlaI@W*`C=5_|eew=NzI{htA&F_;%*4wMMh{R??al^LsqH{(f;rcS!;1CZumL2*fZX
zHiisJX8^%KKEE~_RP9e%s#?J`i4+36`M_b)HdfdiPNJQ7qOpQ1t$>oVRWvR})=F3@
zum}uDwBj+}EG6`!t3c}o9zj$9BRbmH3{iO>@qoHfi;jadZWzWR25b2y%T?)j1!Ifs
z9%)TQ=SyrWSOt7uEK{3;0MY#;;FaqO+uE9RuS=gkUU0VD0DD9Yvj@|$<tvK8Hy;ib
z_<(E3YirsS%Yr|}bAbie>KAvZlY!7?W7Y@%QuJgg#SRV)z=US{`M_LC?=37WT0Je)
zsMIZd^0f>8?&II(&)<Is8Mll_r%1SFZjF|+N_7H+I4}$vECg}T7m+X!;$Q$US(`rX
z4S(AV+&%TH1t%Eb`PSL*^2yo9q*j=6#@!~gs8KX<5+MkT!$6F}#LTDHkNJu1SJgnB
z*6&lxmP&Q?Rn>J!OlR-f`(Ob~hY~S5%?4gXg5Iwb5<^UkabSpn6g0aKTU(?ZR@mhQ
zUt%s1X&BVRxm8<-mlSQEM^ytOniv1MepT0~Vr%U%6owqF5agPx>uZj$t~oj$8KR!I
zW(yiAyxh9)M?GG@2;{+CBuY2&mDL1Xn<d9_;N+b*c<XCl;_&IG%pX6Y87i-J5sR-Q
zdp|^MlKEo55)cO3(yUrQ3HC@Zmg)N03m$*?5$VBwPLGd>+YJmGyz>Or9Me(nIXyhu
zRjAOb(mE>`mK(ojq5L1_uhiT~&0bS-<t$(F*FO1q&M(f{?+&^Nq5$m0#W_!(K4F@s
z*LtwGU{Rl-+J8{39L90vFzpu>$jUt>2ApvXCSLT-$}7N@O2|T$LN$}mh;Hh|BJqkg
z%XY>=Y#}a13qSmhm0Hz{s#U&m`jdS9!B3KNwPKldF@as3jgK48MB0tL-a)JWF7(%|
z0T?WxagkKD<?nt6{YMsLERf%TwJkvQyY-Xv)Ze1>Y5gm1KVxUMyimiFVx4VzWdsmB
zkVGwq))lAW#`$`XayvDuDM8q5wb6VjnbMRAtW}92Vfj4lS@i3%g>5aC_fK{frGsoi
z7f;(THBxG~?!`^vecv?OGW%<5T?=}9ysx!gw_=1J?+`&!!Zk;sXzN7<Esc`(e4AM!
zt-=UpT7&6g&3&Lh4T+Ej8g5;ppf`rVW>iMLTBWWA$3WAn#F_>|+N#6B{u&OEbWrVK
zvZ}#xP~D$ywpDeV&X(PHO2QCFHLxz36a(Y11b3KZBU;_HW}c$z`J#|wP($h9hQhIf
zHLY+QNaH|ALpON_I@{ORdRc-gN>elPDtJ%m&YpJQ>d$)VL)*1ztFhi&8Tx;-Fxt0+
zNU?d=YD2<$uKE+D2bc8Vx~j&o2PO8a{`%GZZa3$(gwX4U1?kI^1Yr)Di3w2i|22bN
zEp1Avt2L}LifQ(S0dl4W_|EQoeB<M<a`M4fVK}13L`#E$iB`-LPt)>4+E`(RgFFR<
zI0Gn&;EbTpdykn}0cCTSIG?ZhmB+t9%`+94N+S*f$%-&tZ>fZW_te&B`!}zDho5=)
zbJ!5n;KxQwD;a`6bKSp!1;Q~jnpxm1ZX$vd9s3$62B7pHT0}LmQPf*7Mq$pTiN%6=
zOO$BfCI%uQCUvgW;1`gp<HdtjAJ+kG>?Munrl?Ids;y||En4wSXoWoQxxT#Q>gtls
zCK5&1&PDq#hsKZ+`~A{r<HtE(8vv%LT9GJGk?HLX#ux1n81I~N_k(wN>nFd;XP0{p
zm)9hX*~oZ$L0>MbKkK74w>i1F;rF9ft3FevaNf^6`Q%g5gZsq0_c%E^CY)&Aze-`=
zY}HuUD|*)7DWOy9zRNDmq`q=(B-|HvOWnCXZ7!8@9QowMWB!B3f0-v|PdMx+94?ev
z`0Vi`o<Dz1eQ6W@(GQQRVOd-k?U^!ao6Uw&3)j0n+s%d+F`q<dYAVip(Y*RmIuKZj
zEdrQ*gWod;Lav>0A%1IWCs9f<fTR9QS|+a(_Vms@j*d@(K#Yl+HANgl=tUE<GPC}}
z-G?83c=fF>&i>{#P2WFr{cqcF8m;4-RVH@@92daX$CFZ8%O|bUT{i$s2Y}|YweS!x
zA0oY=>WAsyB&=IR>83Jn5)U3eB<GpS7cY#Jt&|doPV@M+G!<<JKW95H*W$I++!S9h
zSKq&Gq2?QPe-<l?l^ew!A_Oy)Rc1je*l(`^w?Xw6q5%|vw9&wCRa{_ajHez$pKJQP
z(aWf7C6OQoa&5ixvxyMWfC7$#F$j?%LTr%~wMeR3g``R_Dw{#z=ni2#A-6yqHpFqH
zZf0_xpk`tW3doQS;#zImKGD)pG*pX)`n^dTp%ymd$abSy5>4|~O3s8<Rof>p#K@Qu
zL(<M3LmX5%A4k$=LyXCub%iim0R#l^I$^+n&31<nY@%pTY~|_7lzE1sVePt9H`!=}
z5|B`>AkaX;4dZ^XS)+^2vL@>KC7~_G-Y$0uEbkG2(!Q$dYl8=XL7!!5YER|b*FHC#
z!q(MZ35(FoM#lk@7kgU}E^s!V^J)2v-~o{_D!2ED5Q6&n8DqIv59$4QdVIkz?|z+s
z`0+8vZ+`%y73=ye2gFjNPQqvbpT4&mJ0&J#@O3Na+3uy%GAeI!7Yae@j)}nk?Zf|=
zADn$esD-PkGS7uHj=ifD_FNYRRg8IttGZ(^nV2I}DVoVBjU<T{syTV+6LF9iM2n=l
zKGDHn(CT|{iJ4+yGgd-O)LJ-9lQOE(NKvbwTgfUx@?uPdtVS3a0~xigr>^0dhMKrr
zA`CD_xeeV4vh{#kyOVgCvgWunOBW=wpJ%Ri=Nw<0u^Dv{-i(34t}zaY!yh#O4AFgo
zg7&U+2G9T4>so`B67k+$?tkqo?4F<V{5#*HTpgfk)st&ST3rcjjmS|6i>ESQh5QUK
z`a;~AR@RZ(UG8}LgOAugxX*Ze%yCN0AyBdI6>VXo{BywdQDXnE34~@ZSUPH0Bw#6#
z8_bG%C5<t%-%UK<zToeE^jpmHjDx?NGoL;Blo#hORufxTnP$A^VQf-c^m<qdVY}T@
zYGHpkFbsn+xWNcZ!Mzh#cNaRFw#2Q$sAzvSOGF>YdFsDY%a`2mRV|fio-O5~3(wK<
z2_gxJL%)JxtWlG=N|idiTaSefxo!X|m2UV`efiLop+(=$F5r`T`G$3oh`Z=_9g<Dz
zbtSc*Qjd(=t@CoA!_DvTG-E#x|BhB~UheI~S;p3SRphG<X9xWhYikVS$o&To$kW8(
z>XN*-lwoa}ib6qOu@qz}g=kxAb1%vDc`n~(jEJ&;9vt{}$g(2d1lR55=jeL5m!JFX
z+pqxWFXj*k+TBCHd1Xsvn~T(&=OgRu>+4}+7!o!6290iFwQ7285Q3$}1InajQF|RX
z)h$JRHIGm35VxnyEf7=Wuo;-AnZy2)l4ot>H4I2=s^>E+l7tyFqsHsC=w>nw+MOc>
zVY3-YLBX3;T~o_EQ}Sd5CXF!$HYu?g2dyxUi73(h%C^K|G{9PDM7+C+TEWB>JH4*r
z_jTXKX7u!;XB~8G)w*!-NU&7k!l-zqh!+Mr{_;^%MVZApYr#h}=~8H7K+4w~6s)yn
z?`et;0Vq<tznhIbAP~H|zpb8q1DBsvn}5MRySJHZ&8qMlF*IlzK(wl@Kj*o09h~CM
zmosIxYR4>>A?bw&elLE)@r&Q&AG&jwwA})_)Tu;9GZ?uv=D9FQA+(_BdnwE@GY$pX
z7Md_+GuW1<thVnf&Fo*SI{MrD&-nJE?@*_SnhWQ<gWVzNbHm*@e63ZFJ$>F+`#t~p
zvtQ#MX@8pQX;xG6)|6~&x(78>Rzz#2kJEOcDagta$=EniNl6Nc&1PU2_Qa%(BJy-#
zcQ~-0W)4$UKsaa0tX-_C&2`Tfin!GlwBX}XyHwTUdT1Rm(=-!Wf}{mIzI1ud?_9EF
z!L<K(jOr9JO_>*GFG%ACL^$mBjGM^TbdoJau6Da0Z7I8Tyfy&T?hUpsuLgXFQJvBg
zh>37=%JKW}^Y-O|`SQTU_dZ54{WJ8fIv_-v0Xm&Ldfe1O-9A*WqCQ-~PD*O#^3hXX
zeE%cv-M!29XiM8}sS?Q3ft+W%$zGLnGm_VZ#<s4$x!A@n?)B%@P>b_AM)6-??|2%Y
z^KbvgU*h_D*Eh*rGLN2o#`Cl1-CE$s1%DlY1_L1$0jj@ZV;saHlR|3|36Yo*$Hyn^
z&00o#6b2P>BE;Z~rw@2vxOLMPf@vPae7d|vj^20eov3=Hb-TA1BXib%syXLAu|Rm|
z^c~%7^{R+*x&Qh&i;acletFtdS;;jBT!S=wgzGUQ&ISmcLawIE@n`L0tpls(wTk-h
zAT9@x<$H7>t}u!*0S*wG!~fQQ?Lkeub&9>H#sbCu`L8${o{IEOUIC##4P_;QU&cAW
zxEYn5iHX_{mI4-Pi?{)KvyJ#h-``v8<uX5)Yw`Q6sd9%=Y7$)0VuF{h$FIZZoCAU%
z27+%LUl;=zqNRFW-??n;L7D-DHSpBe1fT`2T#Ru8X&_R}Ba;q_(=Z?mlrj^{$Fmgm
z3x!4urH?ZS!gjM|dwk5)bTkeza0nBD0&VQUbhQmSg~$d$ia;sm`B*4cfjFeZcC%3c
zIF4+RF@VOl<q&9AG#01~AutTlTDqyGPhAg2L=vDsuPl1QkaW!DQoD{Vgo-6ldjQ{z
zWpZw*lw#H)c}Yt*XMLcVku<{1PMjW8RA-DEwa&Bxf_lPHA(YAxgk<gv5(3eyf%}@d
zA|RsvqBh92AyJKgwJ@QXUm^DFSSm=Vy>O;hGhVN*a@K-~QVY2VbF1vD_Jkb_eE8>h
zAs4{_t4%VUlX_F)YN2o$%I}8X=No4~&5684E;BV(1DA!gNyH}9I+3TsAq{45rZ!Jw
zij2K-+fBY3?qwEgHC9=LfIA8N;`!g^`s$KE;Ci08p0oBQ7U}!MdL0d-a-g|rEkG`n
zDHo<vd3LsAH%$mt3oM*ysMb3dCtP%`(XKP87t*PeNm=wXXcv{ucEhmU5K|z}6Z@+x
zE-rUm?hfpyg97eipuMU;EzwfGPWl8ZZY&yTO*wNo>^bb+v7>Q)y=ROOk&z%~%<c24
zRHi9gU|qZZ*>T7D`6bA}G*!m!5ot^e5_R7h68ru1T1(sOGk_+zbdydI>Bf)^v?>Q3
zi@7vf8X4~0<K)X<;^Em9*Do%~FD{89glcI+UqG8rbQ90A+H7u!-q(jN(kVp(jYME1
z(dNq4XOB6*cbD<>7~Xo5xZSERp%*Ub+ZV<nxL!q6&({~01&~B>cXS7dSX$f6az#y5
zf9bRTg6HSYdpe-ecyacE=P%aG-zzW3z7r!HEJn70*Zs9*0Ybh=h(u$LZdQ3ZI1Lg=
zY2f%+BZ}+YPK!gffiP-iRuHdPv($hrx}**>40Iq$ofiWM-WxXOS#@PaMYN@q?qi6U
zw-v%BZuv*w{fkH(s6|ae+se19R!m&%>79i!^?g1Bueg;SA#3{S)vVD{S!ywgEwp9L
zXt7kkI+*I_+!k%uq6=lPVgY|<(gT)dY_cB5{?zUJgNSM<mn+lnY}Vby7P$7xak2`$
z?Ao+iI(AE<T8p)DO3S?xC^aKNI66LNb9BV+>Pi8iEuP47FPd*~&7=0^y4I;@FO=#3
zcBaQJ&li7tK|kkyPyEI}*OYbk)KX0}ZrsP0Du=(2Dqvl^YTvwFH>`KMb#UG+23BgL
z36|!98U#|Jg=ox{%~dICB8{f;Jng8@ZApt3GVaGIA(%F)5@KW+BCQUZIjKhor8B@G
zRMks~d3ai(&y@30rEA|BnpXU#U=|)C5Y4A9C9`vhNGqLTk+Ar&HSua&sJZl_u$Jcz
zo@!P9Q14mnZPD7ANpgnIhQgdnXJlwP{)1pG$|7x0^WtE})E)!&F<P})BA1!Nq@6;*
zMD)f?M-d`|SaF-P5k7nt-JhOVMZw-w^!aVmplLb5z%7K3Xl*H8sD0dOx0qZib8cpX
zqU`!y3q=C431{Jw`LO3G>T7awum~;(P%SKd&1q<F-xgs=UdX4<;i6pfgYsSe*#0i1
zX^XOEU@IOXaU6&tau8t%2Zxn*&$VO6#;bk}==M?un*;(#82GjDZN9zx4kb^NR=C>F
z%4}96X9goo+9DlM@<N!4OCNOam`mpA^K*6w?Jw*`+Cl^Ch3z#6H~}FjF@xE(glONP
z*kss@93LNXe0)M00;Nn$`)l@>JNEmT$pVg=1rJ$cAFnrqT?K%JDbMT=6T97k{eCjW
z8vXa6T~%7rLat^_5T-KgUbyo0Y|Y6fVV(--XBSLUA#RUh+>kIJK8(!MtwPZs@px?j
z*f*SI@9A{C2=p#NO^78hSD_9YhBqE^|EphNcXq)?-}*M`V9ZD>2xbq^ng_|Yk#@?V
z1CEyu&(#|kt$`$sS_;$I8P7lZh<Nveqs_oJM!E@Z(<vTIl-t03dcpz@S*o&p5{f9B
zMp!)0bf85<Dp%J#{>F>H!7u&&U($zVo9J%0<H?i9x*+~Rci2wo_*Bc(g9PSzVz=Al
zy1u3Akj{juFP2qz+G1g3R0n}BoZf7=Ah5gMv)k_s<fzcpb$^7?tvlA;8G72&uisKu
z^E@-p6Z1Uv3F_&?2I_3?0Yq!WG_u{^A*L<c4NTJk?>8zztjI*4fu{icnJ;6}m;nbX
z2(>PtxsMVDb2xx-mQXjg)-;vkpxfWs0nQrOh%;*o_-W9)7u2S}!SB<;(Djq)2vAw2
z1>iXd>KMn_wRYnZUOjGD&dIeewN|Fti;N7s2xMbsPw(8}^v)eFUOczLgrM!f`j}dn
z(FMe|l|9hvzUXiIi96@HzIy*GFL-eC=HBJOKeLq1u=n?w$JqM3{P6dG5jHTT6fQ0=
zbqy0sIn`NxDI2tw&5mOAOhoOZe27*W<V+kRrB$Y!Z6h<XxT<AFvBES7zA<VbT*~5=
z8Je)4Cg$r)kieV^c`g*|J|Yf)Xx%(o5M%Vvl+!%ZO16}4AWt*11fmSghXVyoITLgP
zEb|PdnQp8RiL{35rU1zbYDRX1R4DF`+Ny2R`qX2L#1wBl7Y7es74AhrtyN;wE)mnD
zM)-44;5MYdb{yG^BclYyI4JWNyl5?S@K>7lG(OCk!!$Eb2j+PuHKB!Ih3XnGCfio3
zDYClfV-ws|PS?ypK{%8dLSPdULx?KScMF&%cneH0&qgaMQ1GXkwwzT~k!{_~0^2z7
zx7x2#^Gq7G;<qA9S%IS#KUbDM(4hyaUY+h4ADY>gNa54Vk9og*mEq(L)1ffW6BGsX
z#Gv{+<5`0hGn9t)B}QpWWBF>$GznI$Cm`AzI3K;>_cq@rANF7=&fPSrotH7I-Wjr5
zt2ke0b?-3hRT{OZ^x*mPGY;9v6Dwx)MOYgX6w?;#v+|U!2i4je8$#4d=3#1F9pLCX
z6H}npnVR?H!+|>Go^A#Y!p%J*tb^2MAW!G#;V^NSW*e)E5p=V4w)?00eC`apE$8BL
z8NEM7t}-3eS2THXrDzA4*ffRwI>&1RK#$t{1@zz8#dqO8gbWlRCt-VXOnT!UcfRsI
zcOO4x_u*%NR=}yZo|!88XE}3b8~f)GWs00xSh^69R)`JiG;{s*DbGIng!^yYXE;95
z0aj*e&BQpaHhW{#<mLnK080F!8F;h}d<{}V2>hMT{tkcR@>kjI_B|q~wesZ2W9B^3
zsl8X-+tBeTDJ32}c)$>ovT5d_SL=H7>*%g;(BP49=#$Kk5G}=Iji<+PLoJp4e&Xut
zilgljtrlu5y#ly({Y_m4fJgU5C#JTio0)8(b!|g!)vDP{AioCiBBV5MbaDqFG0g|Y
z&4zi}=>}$ORuiF0MQm|eG-qOC#$aC%TD=*qIjzv}bn$B3ILKN}LVq$^akl`HMT_RQ
zz><Ad)!GaY>0WO37R|i#9RBoK-(fWFFNm>x`ir(yxi@DcKX>@wIGUZWr)z7iTwh;v
zetyn8EvW+mO40K>+HSdj|31$iJ!XG?Nv@ivknZ)@mS^eX-tAm!?F(D2RTokFZXcfO
zGxu@pg&%bVB5(iB!Ru1l*=TL4fW8erePjG0j*=_SpI;Ja4C)g?*YLG!q=TQk1slHh
zgs2?~5SnDtoSEmwbU5HWj>P~&E!D<bT@_ueG)|%^vPvPgE9&7u7HuHka?$j$pr*vl
zSellZTVrl819Pbg@D5Y2AZ;~Mvu>a{&*Wz7?>g|&6GNz-&8$TQ@spUTF+$o+44Vz5
zH1qjv<awqTTbY8kv?4?ejA*8TR~@!zeSE$KD<;W#=5UypEw~@jz@6h$at#m%;zoL~
zt4ucrf+M+AuII*XUpVXw(?LO?wpdW@*t*wuim^*t92kPhgGGA4UwRfqmR4y&7FQ{g
zdjzW>7b|8dwK12*G-n0xcJGU&=7q#%yk<Y{6r|VG=iOv%(b{VhS!74PH_@5}6R6?0
z2(-ZW_=G=^pApKD&YLN70AeXfmcF;PR*5tPf2bx+Nh)2#eV&J?8NVTogxH89d{{o>
zGM^*mK&yrQq0pKrI8LUeZDO9V8ff%RGtHYk(8-2rnwj&=e$viJ3cQ@SQ83h<L;^R~
z25hg0)x0VNLesXVr8M?ahFqa#+C0IWX}RE90<q$yDABSZO#y5xwtOo_m`l;QH7l^c
zk<)9%GBDO*X36CvBBYc^&9h9EYW|BvliVQeb77M+DNnR9Qi-&ACfC=i6?lCHaFy=y
zJ=Tqo8(QKFXcSVlRJ$VN5Qum0aOZ<}_~6AQpT4-}>e&<hEMEw41l~WQZzjzb-Stk~
zxM>L!giwvXH9$9CUGe18PiSwy$@cC&#_1lVX49q_a|o!R?~(%6Ake%AeS;S0jj30t
zK~T_eu{+~$p8YD1FCO(t-hk`tYp$*@?UY~Mq+CPPmRrZk=?O7LrfKSWI&aB!v#PSM
zJ2wtZ$LM0UfOXxZwA+V95}S=`A@{pI7nc`|)&^)uBPn@LNqs6}o=0wUtL$E8nkMFX
zW-{<sYVKnr1g-AHHfg`VH5V@Tm;7J<=D*Fq^bh^Z+`n_eJZo`EEm<u?yt)n-)$4s$
z%LW5qZK>4mZNhNY)|YFHaO?loBeYN|Md+YuHN-o(cE-=2)eTn47SET!N24#$^w)87
ze78z1An|63x7`9xFaP^hO!)Gg{9b?-9qspf)jHjJTdkGtcFSAuyu<S+Px$QW%2+vw
zaRH$>f-%=Lt^ivH-5%j~(A&?<?)fryTxa#Np718S=NoeCpA`Df{XNd&IuL!`qm<0~
zi|6VVP>V8uBAt<}SQ;E(v&<ha1hBSQ&R9ztsikq44z!$=eX@$sQkw>9xTTg}Q>jIP
zMQNI54Ip7w&5>0`HrtRl!_a*$MPSYobFHc`voX$`3HwQdF;sF8hD5FfnN;sv4B(fX
zdu2JO2jGw*ImHF=7$Dc6R7f#tKX3D>oAX3N{nS!Qs;M%T#sP=7^TNlF1Ij^lT(vdk
zZXYF#<Azd&L#fmdk@2V(CB@J*Rd3@oPnq3ZxIScd*9Z3dJ@aAT*G>iU4#u6~H$822
zot%AoYq}1F)NRZf&Fo8}f^@Ik_OUo5V$gLIv{%P8%^VIhhpc)~F(#WRL`k$zm`^7f
zg;pbnS+#JvsI*3!7WTMyyfwNzh9IgBRIeKiFp{!#`#SUY<h%S_eutP6IYup?$*l*1
zx~6LoQm`VY=tU*c$rbhGRB#)Db`g=|#%JS)$UG5iWiE}msCkR8_pJwWFvPT4E0qn!
z7)Y`EI}C%p*Y%aQ?|aZr>soX%nNq7_QH!VzlgpUQqDZV!L2cSQx|NBRXC!AMb8Noi
z(iF$gdLF_Obe6I(_uADgm@&8Q5k{T_lf$VjZrMNd)WvMB5>+b0P^~aTKetwuE=Vpk
znXSq>+VlGHkJo1ayO!LdM*O<%0X7%HRwk<UWwgjL9i1R=y^VZ*=IZHlKG~g9u6Lv;
zhU9g#51Mdb7S)ebfg-K<FHj-2^FKnNQlyZiAfXa1QVT88a%6gP$^N5HncjTJ^zbfo
zK0s~6s+zePHOg&OBOllI^(tQn03y8zz}c~Y`rZYE|M2hs6+SJWaCLpHRY6sKg06N~
z8Vwe&inMzv+TEhm(sfcIaB_0QyYIZi)%6u~KJ-S>4P{OISOdoXeC%^;X-u<5wAzZM
z;by5J25M8EqL31!0j%9_$Mt?stD1@#$B~rO|H&tmi>zC#<XV`giD{lyv~A|n{vAv8
zpNVMgOmnSN5$4wTkG}gK^Wwz?|I*L?i+ttF@B82)B;6>yIr?J!?!eGANS%$=YQMGB
ziUTm`R@UJ~uk3SB0$tN2B7}8;_N}9}W(}BpTAQb&V<ps(S&NIx>cR8}e={XkyyCTW
zt!opnw9~wR^e1iY32*ot3YdJsR50EAIiS5-Pfw1yd*_ski%Xio>FFuQ+hgwDxyRkR
zce!}}jLWlU<lPlBrt6~hDmN->4%~rs1*wE!DfqBr9`uRpxNG(!Rluvpv#p*}(f-2y
zk3TS$=jiqk$Pz&L#$A@inO5cc`n`bXoO$~6i5V#8#XnAz`PF}0Fov#gDY_Tls${9U
z+D@!MO0A$mRM1}u?g*jRWZK@=w1}gYVxUq-TCucowIT}%Nr9Q^qiQY8Wu|2F?W}sg
za#js)GmUbUMrxz3C(!^GjgpIck{0dk(?lg9E^SD$QdVs1G@0&+)^%=k_H2b-R7_@#
z)4Qi&&%dty*!x<WhNPJa2l%G<DotqdKuaTGD#&4GY(lKwxyV|0mC8KN><^jUoH^u5
zu23X&ow5hV5(J?dV`J=VDW+%4U|@)tv^Y(K*wo8Tny%-qR_))K)k{-CB;`OF8bek8
zw$_BhRM}5z^-*$Bn-$PDWrGnQRp-><a9}?b_Oqscm2uPiG0&>4bDJD72F;5gV~Bm;
zd5_*F^*L>tNt=;54U7^Y8fe9c1u1w&FT_Z+z`_A~F-ZnZh;d*X5~P{m+Wr=`?g?cg
z(zq^#*_l$?FMQ7nqK3;DaLqO*=P5Ies!?vGa=h6vqyd?#0Z0R>I3D%S3|3@eYv+n0
z5@X~mYHCiOv#pP3jP%jH*Uip@IZ&W-h=8&4t~CbtsrBHu2ZuU~Rj-%QA{A>2SfmCm
z`09j`&h65)sA~>^7#EOFQ%OUfGeSg?+}Qtroa6NYpgjO@$07zSdJ4hUb(*FMLSu-+
z5R#fGp59^e&fDDosjqT!_Ke;4zE7P!Hf&%EtV^za(XT*}WseN@zzl?g=nf^VAT1*V
zE!k&8t|xXMea6|__Zi-~&z-ziF>Ex}GB#?c1oK|%GEm>NI}5a2yPHF6MPA}NpZz_4
z<-=d*!Ta~g^W5*xoM$et&dFt7u^3o%JeZHyQsK5e+VZupeU)+CaDIM9so9K_btBRM
zXrZWX4+mdg5xBg(<}l5qlP$FsD3NucfCb(<Dx@Ldm=FTH-HyX_;OhEndG4#oKN{$&
z)ixIUzF^%2Voco<M^GT-Hgv7?0Exqt88?kzIsa9j|I#!5sh|2M`3pb&=XvnpE_vQ-
zun~mTH+g4l^~1ylq(^5S$=zn-d_$&B(Ixe|#v)pbR1IYMc$b%0rA3drLUm$5-l$Ip
zl));K^_e?+wJu&TP=x|=l~Cx$Q*BX;_M8jLBG<3bFQktxU5HXhY=#YY@7&?|=!ma;
z`6rmBncelCRvX)s6W)FQeZKOQuOQMGQ{e3RV?O%uLvqtf?%XTY=#`ZYZ?xdSW~PDF
zB^bPknyY!sx%u&8EZ&yFf%PAIehxI0O$rd&%8$>0dXWBEu8)1)Bj>_rPo65=wp5&^
zqKt6>%OaG%CM?XbEYGK@4ovZErIp!O4vZm4tSxaoJrqLF6qKc&Yu^l72ZK#-kg^Za
zQ^%D#tDwHMYI^=guD;pp{7fm?n%WkuZC;-tL@hq3O_}ND)wGS$iZUWtASl7Ki~1qe
z(qyHojSNkhFI^LsVWDHbE)~EcgcQ}+&sw5|xNO8BMba>8U{xErRt|H95E)kXMy;3}
zW%lkT%HGA8v@oREUQ!FSiO~!Ntr&F%+Im+TBh(D^NJza11_6l)EoUhZ3WTVD(13bq
z${tnIG!r8sC1NSG*eqbwYnKS0jh`{a16%ZawK}*^DvX2L^=L#KX)SlQ)7LefCz`g6
zJ{AL%x$vw#=Sh0PJMk`?lOtLUDo4s0s%an197hBsnRlZEq&BJr|6UL%0x49^H_y3>
z&xxf{%0vs1+%y%v^bB@DWBZO3rWJceB0@~j0t)@v(aDi&(c6I*7Scr}Knk21O}m2x
zcjch(0}eW|IqStZr4&Spb=Ilo=V#BdfwbJgVza^?f>01zQQt;ozDs9b#mJ4=I|rfQ
zISneK5~0|=3mz%AH5Z`4dlofVA5sEz{nf%+%{TZC_PQG9*9L&EOn2RQfxk{4ghzIc
zkV6P?_mq2IdXKlCe3`2kFUZd>y6ZinYocaG7QeB+yR5#t-^3Lp2-1{#tq$aLGVabV
zdHT_(9DV6s@_Rel9otTW3WJ-K13S~=1MDRRsA1KyTD2!lIKR5!|MabYlgoO^o%vL?
zaa94JQvZh=AR>go_GrU2%{EyeM&an_h%bNT%iO<zpC5emk#3YmH~S*&XZLxBLz32n
z%ZqDXoLz8qbVRL!NTn1Z#U+XlL7!;|kzo*oBSeIGn0s1xnkLF2_l;ZZJ~$}U1zyG8
zF{PfK19}Sm#^Nt(mGk)V3+~-}g9i^D@>}t@`S5o?<nfb7{Nq3OU+2qTdS6q)ArOM5
zm0$Jb*8@U_#{vkseV=U8($|Wwzl(#5ZG^UZ=>LiIqMVywzl;?9=#}=3MYH3!NWM_q
zJb(LswR*RQ*X$b1l-w7o+v9Nk0sT|G0$Pn;w?{{$LD+1zy!F;Q93P)hixx-`A}4q6
z^6tCuF^-8h9zNje{EXezCC{HdA=kz{H>*~ok3nk1J0|o_DQ!uHddhg6BCe%aYqHXd
zlRWLWUQ2GLf7f%yKlj!W(2p-F{j1knxVYXa2$4vTNDTq3A-#xh@;R3t5Ej#qb*VsW
z6mcz}Z9`QV&=j=&<&bn!D<~7vQ>-SL5K*78Y^lkTi?NN#G_Q>s3OY5Mg)9Bu>W!m6
zD>3jWW@IcOs()XMNU0{AH#tw!bxFI6EDU~E3HrFTkfR4uw&9m~?g682gt1qJ)0%PB
z766ElQUtsMMd)h^rtxxZpjPZxd=csrkfs?Ibsx|_f;I^t3X<i<I;1l!XnWUSUWp}Z
z_9q&UijkPqI61}v34t_NMnr7aF;LVRMBk(>7)~iF$xs`EjcaX(wt$srnUWH;%3Cd|
zyl1kQRHF6&9)_gpakD7$g{%3ww5D<d(I*4KlxO~a{4Vc>pJYf|@?N#0RXiBgaT`s(
zATbc4chv|`3UjRFm}$VH;bW%LOAebWwP4GIP0~KOB4t@Og~b+bH5k`{ngac1fu{b6
zmO^#b-F5k`*-0Cz(AR0Pid`!%^fsimX)qUzr7NuvrP1|#2EJmjqM0U8JYyrhqF<WH
zDXbXBWiToO2nz0vy$7tVUQnRAAUz|tq>o!2D}j`>@TsQOt$(s-luGS7SHI5JIlce@
z_==5n{rmpACAwR`H--R58-};;^Tt=d#O2dxJlXv|Tpb8a)44a&uKqdwf!+LW>HoL1
zu+p5omo4a`RrlH^ZvX5BZFkU0H8i>*+S1n6=L*Pb!GjJEVoYWPD;(xM|Jir{D!=#X
z@A2@>hgMF$EWny3vB3@ygyWNAro%*@oe>D!y?cju-+hO--h7jfKmM5gZfBcHxN#0Y
z413?WE24G^Kls6?lv4TN{r5SJgHDW66|4nIfs3}osZ|(8T?ArEjGGNx+qe(Yfjny}
zI?s#0-!P~q#I-w`a`qFg{n@O?M{AI$nZthKv*|OgcRSvG>n-F+_`iSl?{oGG=ll~t
z_fPPRpZW%I91srl9J%r`FO9ms7}$ogfWX^#MwyWHW=)SCVJS3`z6lGTXT##pq;)L~
zv=tb!KA!84&(C9x=z{Ai%zw(*l%Z(+*KWP{^ABf{RCqs*BOx@7kB@lcjW>As;0<GL
z6|APsmIn_WvfXZg#-IF|pX1SIpYYq?+HrAt$(&V2P?{Gj6vV~W4P$V8BXacHIuO*T
zyLRwbN>T01=K#M~-B&-nAVe*k7VW1;_G^Iox<^C(+LFK!bTf=Yf)=S}4jm0bK&57{
z{`Fv>byHept*S-znpPLl)M$u<GE${cS~eY#LZcKSRf2+@Y-xY1FqL?dU2&F9r6kMd
zs=<rsJg;7i6||GbFlwLPanv+zRVx=Ekm3NX5eMtYQx*Zfq+rEl5UN!jOQ1A}L!t(V
zxe%hVjL}$4U(-EBv~IhJ^`gZ@^Ae+Wt4K*Vc1+3<(EgJyXE8lxm!;IDi-`u{D#aiS
z$RNN0iB+v%2t)_BQ7C0bP{D>E20{|Sw9vsckSThAtAUL))5}(tR9_T=IUYm;Ez-e^
z1R+KZt{vFLmKc+2pM8%Tqj1?4*^~C}R@<!<qosz`TtC;Q%w%iUrY|s9(ajJeb16Kp
zk2p(@IhC(5q)KUtI%}U^Eu7Sj6fGJk6X7g9=RCfk)ymWOl#6s>dl(#T5~G0ZW?IdR
zo6)q2I?hrIj47E!$Jjs9&Wi;M?lapEa+x_A5-|o%4T}QGKr<t4<tVGv8v)MEVnJ+e
zutF+v$&#%fMYN*`2APA{IEbXy2zA9STCh+>tcXJcj9iwnrh9j^`7Rym#1zdhI9hTv
zn<iL#twRv3EuOx~JFGc&s8)#QJ-9;<ofo|!@{f1CJ^=iK*4JAMAXp?7g89G+b%@Nz
zBd71b&6h5IlEbreF7}_QQKk1!SusXdZ21xyr~j=3-wDDcPPMcuYKsu7P%aMao;+td
zJEzUF_O<oGgtnw`^fDCSz4V<;*Pz*&MgO%=exASl-G85}-6e0Qw^iQ(c>U)*O<hA2
z0&KTO1QBwckpSbExOeX!@4WjCZ#;a++4(sy&d!uIF@4E2X*$E#(9eS+!V;NkdQCV?
zneTu9Q!X#A`O25x=iwU<^m?nMt%B0usyEWXHEwhPZe9Ue3*&a%s|FAIy;`=2R_BT9
zAMAfxYo=jSQ0@H}RWB3_P-?HpRx4-cXB-X(-g)O8?wsD?uO9w$KK#{>_-9}Iclk3v
z_h&gd>B3XP7=or9m$8O{|E~P=$^VMsDISB%E1s|38cj!8dTn7}Zvzz%P<(@2eUH+2
zRBIJ<yGr#oEOuyGty(R&!p5{d8b`I&D_3v30i@^)>tdm@oI67IdDLv8n4lRT@$ike
z`I(>n(_CNg`1ZGdi>WjYvw2pD+Al~GYYC@dd40WJ6ypG|SGhSzGdA#5fX_kmx;5Ml
zkPN!;YIDv0`(A-;K9gSv3)mcd23>SRB!+}Yq*O81Z!pWE0=S;7^>dCf(t>s{C<dUU
zDf1UX(*P_cWhg>g4A9l1L_01)uO<imm3y4IB~n0ByrtCMZ@b~(FZA`ojo^o2WZWE)
zMl;17Mm^VBkWjREpel=%qFrlI8<0|qG0Ya|g{GCo08K%k7OtqdIF|YGtU&LNih~-j
z|1@h611%{<#$j{QKBhLRjo%ydaK$*U$IK!l5L#POm@x#xpt`X!C1O%@atTe-y|Whd
zU`%Klw9jrx#!QAK9Z4{KS;7Jm3lmr8pdi)W*8>EbFI5BNI3!|<)B}1?#oME)Erix~
zEIMeRF{FsLzZJx^qC<=w0B*)eEkZ4US~Byz=SlfKciKCI;Rvd>TC1&aHau4bE5Ub<
zf8PRPp-3i+Y9TAlq&&jWCNV_Fhe_>?&}*z&_mGsD^_Mm|LmZg0wFC9>>`bf(lNBC~
z!^rW`k#?#X2TO-ms)QRwHo;iIS3Ph4S+t(RE*a^%ePfT^CpASOB^YQ(3rNdWVGSmb
zKuV@m_p&ufL<5LgNNT{|QKK@_ts6tHE!2E?2<ZNLV?`IcM>aP~W=KQ-T}p!j<JUO8
z03Sg2AVN{`$`sfsZde9*7>2%3D?6mBi?$g_cTTzY{yV(=wJ)<jzhL+Lf`Dp{0<bo0
zb}i7YQ_(?5=oN?|2AY-TI|z&+5;HItE<bz1bauhCzb4sArxBksfcgMa3=8SlHyn*d
zKiYrH51xI<FFgKheD~>hd2#j}L4EG}y{IarF14s;F=)ZZ(b1N(vol(2Y_<arA3osC
zx8CBx{Re~?IeT$No@Wp-rft30{XyW*U$%gSQZ#aW_WX?N>nk2UxX;Py2}egq93LNR
zBk-6s?X+4%2T|LDXwShX;7~_~aYLRbt!DEt^6#~cLRJh~ZC)|16L6j@yWOFmccZaC
z?D**8k9hdt4c>g?O}>?Wlc(Q)&Wp1P{^Fnev)q4pVpY(^#*?#hI?1GOE9-?|!AJTW
zE&Xc@sYm7g-Zq_(!GaE7TtXD>4Cb0Riz@p#?_as3FuNYiez{nEEcZ%$aEpr1bA_B6
zxeD9kBZhIS*Ge(}IjcMqhp~&y!k?QnkI{OtYIg~`h`UuNxq$?R%}8rPF|V<4d(8V^
z{t3Que#Z0jGd}t3Blc5fw?8nKOkz+yqJf<`&l+tTsB`1*o)W$V>RjLF_rzJk^*vgH
zI88||3%mvv^lM!MK!4r3DCD&cWBSK&B*cM~w4*^V(VT>&PqAb*KF^bNY4Euody$&%
z<IZ{nM3%L%(N@=D_t4s^c_nd(sWT0}k)@PqP1F~xR!vv=250B50i6ZF5=p~`VHio{
zmT|Koja#kuG^>^ton6tjWbkxZ3>q+&(C38~tN>yXm0YceO3RQWXsR@stAOh@RT5+X
z5PDFE1)rhW^(IYoM=L_BMbo*#Qhm})liHR!?%tG(HdKpV2({}fn*<FanqHfxrZsg)
zfS5*`pL*~8{2c%aVpb-4zeNI}1`CEtUweK3s=<6_o6gx7LuZG()s(>0)TU|9nYkz^
zmeqRFni}r=Dwl$2rl5kvz%V3+A(En|kE1lw5NItK+gSPV@I6l3yQKIQzq|Ps`*uyO
znTzpE0}c`04_7i7X&R73W3R@PI6fYk4-=QWiFvMU2xJ182$D%r&m{)o^z?*jnrSWc
zp3aD-Nh_{T)rJtmkdPQT%$XPmWt3ef=w7W=0kMR&?p6<9FMO@-71Ge<N<zeI7J_K8
z+>jt4ialDO=82cva0()kMpnbRpc1gwBA(K=yc6?0BQ+6HMZ49RrnY)U%oj2JYgq$5
zt6=+O(e7cUP}O;JBc=YMA74N#xb=Nm7IOW2d0lHhiKM^M!GdlOf{X+4-YIuJc$?$L
zkGVb^D2GBKFi-PMcBVO+e-#g?krGuSIn^#yoJ>?vXh3phKEGnVxS-CHzDHj`Xx8S1
zReG?42mQ|La?StiyMLL#^@G0&8+|BK-m%?|<atA_c}btdNNu@q^5Zx#49PaQ!Z0Qt
zzVU!J-guKcckYmefs6AC+a$DMb<Wi`SWjlZ`f%{)^!oB_lu>f}d4HJs=#xjb5C_tr
zCHH!6>NcK+#AdT$v)Nb`ur_@cZ~tV*nlUEAJavs)pPZiJv#}ypvQt`Ro+qw%JI*dH
z*&XK9S?V?9JoDtqV{)E(`>nV6X#4?x<&*z_XMgPz{+oaPzrhC|yhjSrfQB0Dc0kdr
zD!(m}mQLCQ0S0>({!Cn_VslT~N)#M<xw{X8YAH1J9ILj8)PC(ZBhd9+SOLW4AFJ1k
zb%Y3=jT?q7CntB<Y_?VriLrpov+cpqYEU@S*tIha)5U0I;XKcb!v=}!4_9j?rNs8=
zgzaJC==hZ5lO13F^4Gb%zGAMGtKG!e<t1+(4veF+MYIJd^m}OF?Pl867wnsw!1X;V
zD@JgAA75hs{PVp2dIbR2K=O-V03`%An=QkrY?-EN^;v1QUIR^OLbVFoAW*|n1+Pqn
z0`6uEffv&dd`{PeF|~V@=S)t85DZKUR(0AV4u&)^jvE?*Fwe9wYs$&M=*|15o98fY
z8HcTc*>PmM-7;)8c5W(}@$>M-JG4rz6ExAFIxOQ(KVCpw8w|ze6)_Rrt|w|Wq-x5r
zN>%Gp19anI>`z-HK;05Vgc3?;(z?!gnNEJG{o0!aOx<Ee%o4<3E*d-%tOy|pwM1Gd
z7G&8bKMoATsKi3`y@X|-P{AJwd3k)7N{g+N7q`cF5KX7=!_4=G@4-QT9#f(ftnHqX
zET!rAiIxWT+L9CmMEeW}f$f+GLCDk8t8Ry3V9J#zyx><(ei_<^&_p$TQV7IljaQVh
z+)KS#h`^Wv+aV!MHHe2S(1aQzm-|`Uqczi2igukk91f&3P@9m;WPY$9(t2hr7u83k
zaiB_MU!a5tF(FhfAYBE5)l>ZzEUvYJz8MDmd(5xVKzD4xMBmzCRJHGsLDaXPv_z#+
ztc@wARu$16)S+(=vG29&O}Ol5`~6-kM#h08Vx3~lDrz-u9U}qeZD>BfRmT^tY0J#!
zbM4hxtsmw1LV5w$)>!&r+&WeppKo~mr2PV_wjgRDQMV&U@4UeqKlx>@uCF=&!Dnm=
z#MOqUi^I(nn;(MSP}ddcre~3+Pu5noB~z`;m)Bf8dCKAXnxm452%#uYkg)iTIVe~k
zzxn)I{PhoifoD&j@TD((psmZ=mgmo&l7`5BzvJ%RP2T{dsc>^UZipdpdV0dsr%JcK
z@$dm}z4;a=r>BhLhQndv*|X;zU}tL`6Jz3L>Gv&`3j=@tf8y|9(PDY(#xJK-ZEQXt
zD*MB{M6Fg0;r|_Nu^q>eaon)oj2s^yv)yh~;4IPXdunG}-ITC4p)^qois``R^);85
zR~+W4>ulG<#cy|Qh3C(oF-;R6y#E0o#^2}PJ^LO{e(^E?{6GB9@s%%s&{-2#$Sgil
z>rXTW(j!2BCbn@ce=o+b%UE*(ut$x{&!sIzFb+0fm4;Y<p1-~q@w5d{N;8|IR+w`k
zr6ca%d%)?P`wZi#v5nu`SAX)2u@S7?+J94<YF~PawG_x`u!Kh9!#HyH-hB?!K~uH4
z@Ka4YDSY<nXS_H&<KpUq?U6lqTe*bLTR8c-G@JK6XT4gnt~S0k7<^UgcirBtU#)#+
zDK}HVu%wU6dSJcA5kw1Tw%a2%+anr6Ew=sy7x)jf*2srOOIq+%Rx7)*m4Y}A*cx=b
zB<iDS0I=1i(l+P9G|z|?^WcS6mipARqy}rvEGvl0R!(^$MAxu-rLk5Pj+-rM*y8EX
z5P~r-mfH07+JGpv(u%PjRf9;?{;jToMvMcwg2bwtvZeq!MePmj?a7=n4`@M0iVC`r
zs8z$I7(>)ioEBWzy5+{YAqHck=wr)Jw7y1`aky-#8mxujOj7B)Z!NW_)niPmy|Pb0
zptlyRKIS2sl};an4O(z(u-XA%t&LhlpF0{P)uLT21PDpzo(yAOw>71z*Q~l~)e6qJ
zT7|EIzT^c%D2UV`Os{zyMrvHN%cVB<)6CGI?e>JEfJ@9`N(<1`j95XGRn{6?2Ou=Y
zVPH(cVLB}PMr%}3mkMb|i%K>IE{53sA%{WR=XsZy)*5@$$dAKF2!Y?*eZ;-_OH>&R
zm?tGw@E};Pvz1lgSKHzf+<fP=bbnM9H%22c48YMKqQ=(2UKd4>8Yop;lGb8Gfz8Ee
zMPW;?9kC!Ilf6S*Yg}LN^f}LoAw@QW_EnbP4kyx|RW)O4-xF#lrLe9(26XM`f8;hW
z)h3f0p1yo6%tT`i;`$E=#!O5=4nnwh!o#2Vfa!YAbaqL7c1dh2TM6H6U-5N)tMhfB
zHe?6{vuFSnk%BH7EoaW2J?HZLoTKT$v9o{v09e2REEC+n#+zb?0K*V@_Vj}1XV1BN
zdXKp;z)MBqFbo_YA5)uxxz;LgzVVQ^-+5b^zmyo#z&vH6%2gCTO*87}_bNtX0fCBB
z)*8I^M9^iov<fXO9W9!d$$J{n@1NZG{9GEjOxEV+vTyV$>H@UcY&bgJa(uL995i*3
z%?x^)W-fO-u6Gl8wlq<nntq-)!EFDXz~$8?-~Yjfy!*~O93P$VpIrZ^On-gipZjzF
zeSYd2U&m%9La~LRSz0Apl)h*&{5kg4a6TRk2zJn9U!JSnNC9KPM4+mFVn0t!!2q4j
z`n<y-k-b<WS5&xM(FlJS22StX<IdfCjN2m+AvaCmxPaQ(z~+~6`&(A{cijrl&C9E8
z@(LE+BsFq$a>_%ym|7cCnfYTs^-YFx!|v*eXV0Ir-EJ6CVhG9h-PTj$zM$8tCd3+)
zV0s<to@B2A%ePa(KxbNgF1Drx2ep~zcl9}50P-qr;g54@>Tkoa*^)LJK-1){nnt-P
zOCyRPEZD03x$0+FnzU&lP$W=@<i5d62NjDxY1zPXtvz+yh3?Jtq|vKn10`Bfh(^q6
zJA$rd4gh0}Y{rq02As|7>DV9;uxF~`bCp8qoD%Pd&;WvIdUbp?bmI*bLMTM*0#&Kj
zV>mFx0T*lNQmO?KaTpCmiP;HZ1s$P-KD`EgW3+P?*WKAEmAd!|hSiu@0{JDlkghq@
zK)*DQS{O)utX8Y8ZpJ88RF5k*?)-Z+SZ#eyvKpuAEm&z)D04NSt7E7Xpj0jNNl80B
zN$zCDTr0c%YzEqmlr)1fXDu8lB4%9Q*lb2ZP^0c~NaRwKiC&FCzfnFG$~;l$nIMg$
zQMIQzLl_6}Hm7!ts$0~%<rUk9xiV+H$76&TNA+Q>2~s4L!fu*K!F?Aslb5WTy3veX
zHEp#$s~7_%SEi6TG!5v7Ae1oRY^$G>CrVp2Gje;rxw%V)%eUgvu_Y=Qqg_EXD2oA_
z=)Qy%Hqo^3a>)Yedf;O_i%?szJxBG>`kJYay~F;XjUdXv>CuMmn0lbr&}Y{h85DqK
zEU(Xl2r+91zUJ0jSZl!_bv3(rd?7Y4j^*@!^l>-?b@Tj|>ak_FGBK;OAu*ATPe^YB
z-v07|i;o}i?Bbf1XY~R!)Gt=-(F}muZSQPEF~6Q7ki>K^4MK|qG~JL0NU7wrOXl-)
z>V5}BD~_r}n>d>0(UglbNDYUYN5@B;oSdjv-hRi$g|^vht#a7!>Dr#A%+z*!WI$nJ
zn)VzWjXZq#po?9D3JLlh^D5hEni<D|>990^c5vjK5uDL-9gJ7Nu5d>GyH&ccT&J#G
zV`(3<FgXjj#8Ww93wU#2SlJzBUYuQa>yaLv*=F5)RARjveFWH7tnk^bF{VG<?RR|q
z$;Z6=&U>63pYYf6f5G*y?D*%}e~&-)&7bXyj70SwlNQur-vxRrRJ_7+D*))5lK;pA
z_<gJ{Od26*O|w;hNof7vfLE!muE&CpW|}eG?`_P}?V+Iwhq)2Qfjg)7IK6w1VcaPA
zr!E#yVL2yk1Hz+S2lRj`Y?T?Xs9sEasX#+$3DdhOlRsFR?&RnMdDyIR`K(cAgU>$u
zl*3`-;_{O1cFXZ*s|#6c2oYN7OtWsb)vDl@lzG?S^yjdEQ=iRiE#PubmyOl$o&tLx
z!wp)&`(QU-p9K^|)7>GGl2(IeDF)u0&1tN9fZ14DYm%ftpJ+NT%OL_@6l80K7i8Vo
z;1nP>w4r^oNiHJYPDcIOEOi&`zQ^T0I&+u?O=Dt}iOm?0S}UoaOEZIgv&Ax2LeQcj
zkkxvk=MrLs+EnXQlm$}ztwH@*5iNKT1*u|y%5_db%=0fSqjzD+)o(38PaD#j`uSD0
zR&mg@WL&(+Mhl&aXfSS2)foc4c&uonX(2=s>)aqg{SNI_3L4cWtTUgotZBXbhyjs4
zf<zWMOg9|XRC%rHL93pRf=IDxt)7Et*L&_9ZvkN(60=$K#9%f`2!~nw@+yc^H-;Fs
zqPbQpmRZf2AY$zPL=3`%qr_1(ASOa+Gjj&UV_Jv``b;0M*(2Q#=UgbQF^-AjLD(b!
z1VQ`0wW`5_?phoBd1NXY2-bp*-`1!(lg(s2CDkUDQgk0QNr1!%MZE&`P6oU~SM+;k
zmPE~hoGvpdMQy4&0pfEb_O(Y@HLXl8=4t5GF4eO>78rXclMtYZkV`{KBNt)4;<`@K
zk5UWstQaH@$@Z7fGaKS0Om8LX0eVBMxKVYSSf<A8uWCh0P=Ea63$cOGTf6e|30B3m
zLd1whzO7}%gtpFJ=q#$XA!_;m$(=3Xt(kYf`em-3K4Jgp8ABOq#bm|yaB8!S7IQwg
z4KZ6fB!z}hNUcz6Vvr3%D$)vLOVq;*FRnQ}dQ3TgPJ8fxJe;U}w?YWb+`zHPiA8sz
z`~Q_vnGO?2M_cA;CIvN51|8@zMzRfH_XG+taX9Qay>rYk4z#AlDFW<v*OXk?ZU(L{
zFWB$*Ho$ctcYR;~<5fsL3{?|Sn-3BnSiYgE#@cm6x|ztbsIE|0j~elfWR1}5=U|B~
z{k&Y80_o2iGq-=U-bh}0?dsb^*iRFmeEKQxz4t!HM@Rf}{TqDv>p$Rs+WyD<xu5w{
z5N6U4kOHj?RB3o$Pka%RWlZWO+jKnQB6M9aTL}8&5?Jwks}0pR4x#U&_Is}`lW;TL
zS9?lgF|X0~UXj*Jix;*WA3xya?tR9i6QZ?{qJm%^YJLu$5({dVpc}b~-9^W*pQCOl
zW|2_!0~>|8WJHphGiPl}6%jURLplkxhjsBEtO03(Cr=-9etty=k^101!<g*4BdvLH
zjZO2=EW$US?M92<wh;wS5)?zN3cCCKe3hneosP#?s{R`9Lp)Es_8|)csp*-l<D^%6
zH|8=cfT#;lTAgVB@4+^CT?f?`rPc<)ea&>^aTd@W6N3HBs8$7NW`ClrmbQvYgIdXm
zKuUpDQ_mW>MlBgotj#MZ^)8mITEp%+hsQ0-NHp7gLUcw%*B1{yLrgTO#8OCuR{fTu
zpBG335r!dlkk(o;K|K0zVy{TI2j4y6l!>ggnjs}3(%DXtpq)3Y+l|f%t0$I)y;{~X
z`(4z)p{{}uL47f+s8-9Xu(7pI;|pzh=X0>)Ri>?bzUnwEU9YM3I{}OlZ=#0N>c8h{
zdX0j=6Mjb-?3%UnLs3>fG$92|{nB(S%~^HjO_VGck{Mx}G)14ORsv^}XG|65teLwp
z1&%fYCr5)dKB#QQh>#g1a6Rv7!?6|=^~bWb@<Rd=g&{<?Nf^Y<+r2nI7{-yk+0%Hj
zU$H8E4?=oNPa2y+B>~#;MOy%h?dzcHm&%a_)AB{UV~&Jkju{1FM5^{$_UF4E2c8-7
zT7uF9DV2H7S`1>9_+kZMUVH&MFQ-yzQ#Ii4fe`&XecwfbRc@OpdT|Mi$y>xSHC9xr
zg@-27@;yETLS5DyFLGUofFJMpf;xd<CwAjE3Y_HT3BJ6^x1I`cn~@<UQW`kAdzUvq
zc#jvKea89u1#zbnHdxL-q=6N8vf5VFPsb~NS~JavK<{N(ooY_(uXmh1e#*n=FF1L3
zVi=BrW%00HO(0QqJc<VOfyGD~26E8>z1?iN`{l3j^yxDWb76OVZOq+dte8>PG1A6?
zyZ7!gjDu>af?A?zzu(Hy(GmOoo}7yse%4AFB5tPLH%4x7Q^t_GE1B&Xx9@suYIlI&
zlUKY5I;g+()cYNG_<#H7E8_5rJM1&3`M@Wie8dOue~Haz<fHi`{*S-@Z}7hy|7ZN^
zKmI3eGto)Y?IMC}I+Y=EKQu8$1!GZT)D#P1=~$o8&M>Xu>y5M}PR%cDsn|cdPc2LJ
znpb)0OLU<)%o9lD<m3)_?%d_*<Wy~S#M=`spCjT?;-VMnFYt8-*Q&$?OgMXjg9wf4
z^E^{Wy>G47Zv8N%fxCC_EMOPusqLrF9&>SVMU0Wt(_=iM>smYq4!44e4rHK@<8?q2
zLZHY^w(oXI_@(=A-*H<<_(kae{j<GWK`oGT_P9X{oTPOx#C4Yw4_1S*78<>)?MSv_
z34(5_+AU^D*{V(|_HnbEkLt#xi3ZQ5P=XL^Tmhdzp;(YCH~w_?4nO~f7>I0E4CDn4
zt^>tPXlrfOp(dCyeNk}J^m)c)!GAE8v^AnY3>agx1fmX(w88QEe(V3}Gt4~|ouVM6
zQA7cpq@=+MIv<jO0ZnN`i-D8^xthUlsT5`nrkaA;B*{L1sXX^F9PB!6%BXC`z>0%H
zA&^U60Si6XXmd}!@XXY?DfsWah-md_ZLNRrEE5ehWxg4ia^>>+!13`B(9DQiTk5$z
z*c#LXGX|$r@-%lwu(ifvo;l1~t-nc;yT=>u9c_##(X5TzqlB8Nb5WT^8mMBG#}d1S
zcawx6Dkz&ugQV?K6&Blg2*Nfs4q1!a=6TkDyjpd<0h&_O`*O6xk~|yG4#`+&9S@<e
zVTQMY80e42@&%+Nq!emy1V=fh30(>=wD(Q3IVvdITbgQbX|)GP+Vw@>1A$Pq+Bb)6
z_BpMOS$yxYL*Hu>EL#(+Z?`=tvHs$zAuSSWBSwg|5zUrnZnJ?u&B%~vNXz&9IL8-Y
z10y{WzwxsWdYZ?NAEwFaK7Bd@nQ7D%8SdTXjjw*07v~o|{?4aFM5CP276HD0f;J}Y
zG`mh{w46UiHEV8Fqesl(F$mM)z}e%cTt0cm8`syQlY70&O=)zsF+hTuODgkMt+mYk
z`w#f|;~y{`4wPCLL(uj_`#n#dJm%eZKj7-}y0d|5X;3&GkJ)aH7>1EMr*}*++j#Nf
zjA0m<XVrPc7};z_OF1p~5*8+~>-<*W!U2fDOZUB}&=%s)K5zA!8_(x9p7Za*U{CwJ
z7k**KTK_mr6Q6wgG4H+iKAUmFC*@QA55M+r@~@5mbAIM0|2PuWw#MVu-cQj2m7cBh
z3E1zGU0rLPiEzesx!!i`-}KFDEy^`wW#MH-5Q#P}Ti4gL22(DyCY+v}a`*0Cj*pMc
z_X=ZiUp>*+x7#;%uYz-(;}ZF5z|t%=y9SNE0Tl~mQg;EW2JlXMW$*I3%=yI`FJ7Ec
zYhg2P)LhkE>uYCj+H#-lf^Up1V~TDw&$lfOUda%z#(cMzxy2NI5r;?JWzGbNsz0oi
z)&wz76m8>lt0wOayv!|6pEZ?VnA=H%&R!~OssTqd)+m`^z3WhGTRmG(BQ+a?#yX)4
zp?8iD7ND*7MuQ;rle6NHA~eylk&9W=Gy_ICQ?rWXi505ayX1Lho({yMJ$t<sR~kZ}
zQx3vC*wFcAY<x4e-dO2g$c1jgWnRV_G)SSFA5#OdYOmkg^z()xG7bxCDiT%F)2x+V
zjZ96uY)GqGoh$<-D3Hxrfrzhzf>@=11yoH`L#?C2f*I>jBnnK5=^{f&s>ifYytG*Z
zs~$`$NfHP>P%5QRQZ>KGLIQ>%adEvzVq`lGw5$a(TEt?7F1C1ikfGyFePiP`P;+LQ
zGnf0BTpNQ3N1K7;?Z_}_CBHQ7L{e%cOJu)0Fjv(ohGC=rd+7BwQNZ77pr*hqT3|Lz
z5fRt?HuI>|dl%EdU~E{a+Cj*5dnpCRaf6`gWbd+}G6L1VAu7S}In{gv>xTs&sDEUe
zA(E6ZqN!brczoQtrp}(Ejn6=>!d%n>JjznFtLD{428c3piu0DvB_eC7ik_a14P^oP
zT)(xp_`acpjxnp^Spd~K*Bp8q$2?~wW~BX*`v4l|s<%K1`p|DPfG@oljl^12y++h=
z#blb6a7Ww3y>oTV^`i^+&n}J4Qvew31ez}{=2b=W0SXP%b5L;4v=|~{+74>v#iK{O
z`1ljv{rU?|9=t(J3i5oBm0pIftG1>*yiJk<lK0<xm(M=^M0E@yGR+e&UOeZ$_dd`R
zSi>SmkNkA;-i#Yd^9eQSJ>2bfT<>-qA0Kmkd`ygyi_1%PyK76wEgQ?S`GFx09pHvQ
zH1&6LL%hKXdU)$2bp1MizW&0taA3<MRN#GkgO*#L^F<vF#`cFjj~;!-d+)tRijfZw
zKj44&EB^-ntL?wePrm<k6LA+^fICyFX`=-J+SzqbvVgU)-URwZwAH<A_SUA1@{Q}a
z86Xa-{U)0DIIi1SsXfop@d<bC-skkr9fpm0{*~H4XMH%psJb^P+tbSOUH+A~Z+vSj
zuvdEVNetR=^hT$esA<U{x;Wpxci$Fnz5i~ue17qQ^Ybf?j|zh%1vL(UOo$y<`1LmU
zyB*9ZYY;m~b>Q-O0C4@fH3$TH>iV^grkhl?=sSi{yF}H3b{Pu2q7>V>OVjGoS~OiO
z1U2`qg(0czKrK#IdOy22t+MimGyPa=D{w^b5$r9-T-2}qw*#>@iY&lOwS|zg0@M&H
z4egO0<48=VX_aDHJk@FES$mdGhe?@|6ba4zFKTAq@0kvJV>u$@X4HyOS+m=AT|RY0
zsG^dLTAQ+B{u-dwqI$4Hreyth&J)8ZgfMnMOH<K&wR-E2T+pwM1q#9<Na_R6^@gR)
zy1CP6Mg8YmR-O1#$*uu+AZZFPMoab0IxocvM6AFtkZiBfXPVN`0Z?nRj)%T>SQbcB
zr|e$3O{{mX6<5}#b7#(l^NTAuJ!TvRYSUizV~m6$bWL9`9;%wktgUf4Of(U$4w?O|
zV`3Z<$D0ie;V>&(9fKBH97^SSnwe@dR=OZnKck}Do{p73DUf5b<*hNqX7@q^*BA|$
zCuOy0Aj3$IK%Qo5%M6>1vbm+GeGQA*xvp(u`-^H17j5l&)P#OteJ=|!Fv!r?xmK54
zl@14!M%WtOEbCB;O4_8Msn_VJQI>D0Y8#`{tkMnb^;s{BtiLVBr7PA-b$W}{zZn4c
z>v5Tfy=$H`fr&I6L0h*({n3vvAO&m>Miv%mooipO*=sTPjZO%wDBc?+Yb8!Vj<$>s
z?sNCucX;sbCwzQyxtJ-lfDE5J4p;<li{vRSotPAt@B(O>I0e{UT=C+gPq}>ZlzZ>J
zPuy<Rrw46KQ!N6}KM<s;rw^6SKKY3Avoqd*_Z>EyEn4EPhM`wiSL}Az3`5dx7Y)X7
zL#Y$PF!nx&)uQ?s6Q`$lc=YHI+s&4v?NQ$V@7_6OI_$~T2e;OxMBEuESWf{Pg0lYU
z>aZ8p=2gJ!bnfcjcCfg5zcXNM{b8#wercS%0*w5K$HKxn0B!7c*F1jwi1*%mA6nyk
z`|tCA{;U5N{`G(Mf597X+|%e<EG6SD{yN}YT~#v>ySny13GMzGpR>9L3jh&bz8}jS
zcSa6cTbPk`8=qkqxpU_}rzfY3<G2DmdP5i8NB=lrSqRtX<rS#@`w(tQF-p$lIj_Ky
zrZ93Y<~^t>KZ)*?ZwqQj+`W6B5Tb3&Q~!*Mi*qjKOSZ$vFsean*A3c$=s>S8jEmL{
zZdNOXfUPSp1Hbk2|6zaDuYH}P0rNC>Yl|@jYEaYV<b6Kv=!|@Yc`meEj0v&SdDE)I
zA&IqtS{>pd`<4FB*t{TE|Iy4~HXkucFC6ecTHR%X#KwT?Umaem1AyKaWnrXNE!L2V
zl!lxCw%W)gt98??)tiB4xm%!A4%b)YJRvE;UfaEel#Gd4$beRr_4FbS1*D$VH_)<r
zzMlF$91hk!Nqzs)FzEUoO@nP#Fw*aV%`fpP|3(t+t*XTZ7%)^eXNZ9o8<J<Mh!^D!
zTBBsW%$SzVTi5=eDex8u6!k5%XX*W$Lk!j%R;hF~fmW-VFz7W)9F*zt*(;)HzF66Y
z$ef~9V;kt1=E5`=hL|j!>=sF@XudUS)^0S_>{j;EOm2a_`N+l?*ltGcYLX&lnkf|y
z6{Kp)cb6->tSlHb5_ItD`+-QHhCq>72@(-<EZQzH==jU3jbJX7WT0H4+PRPnptnkG
znOcfEE7VG!wJ>Hh-EA^;oRbYbuyFRwqy$UB$8z){NuOqZ5v%bAuEA7C4_c)aUGu91
zohP(j_3y#079pw3K|!Er0G5nKumI7k)^F~xTHxXtkiHf+XfEPv%1m|unKM-;YMBXt
z_`aa8JYE|BdZZ<NnWtaqZ6NT&G(4bHXI`mRYo*u#Pba6GzWD|Z-hZ3Nk3M7n{M^zo
zLh9zKX!Q;5z-<wt8D_R-BJ*Nt7FOcl1{@|{eEbQQpFLv#wF@@)?^=3g0Rh_jM*V^!
z8Wm0ZYo0xRq^6P9@O++U6=U0m@$~6q-g)PJ4u=^{#bC^#Rvy*bn5Kzf-2_UhOw+{S
zFfj~^{vpQ5_Gru1<t5(X!2`;ED_01nr_tQ1;U;$JgqD+2OGEAcL~ecP%kRJao96fP
zl_2mY5cq=*5v{(Qrlq9*HXvxNa(#WxvuDqE<KY|BTKSE`Z}6A?^M8l`(LeLAaCCHJ
zj_@O;&MWZOzhwb`t8@U+!IsW==__Azz}*)c4}8|^il#!V(H=|e==Lir2nH?wIX$_{
z>79Fw+hdTmPSC&4798t6+r%Uj%>>9;(TP3GSIsroi#F&Q9#0q7TBtQMPZQHLv)NQa
zlN;AV6UJf7$*J)0joDa4O$#3m6IYiPOuIeDCr3c$Filoe<M+zai3T`dV(V5+VDB&?
zH$mWS@ON7q_y=View?GVrW(oCXf+Z;?a}-CoNLpxqHBzDF4TE8jgbk<hiLw9mc9#O
zn}$7Tx%uqei9nx`Zjvdu2cUEYP@j;c&8mjSzLwrt$I$d!XhKLg!)RHx-K;hDpLb9V
z#!NLztJJLHCFcpLg)nHhhB3q&%&P~X%@{2a3wmi#-OJ3u>Pni>8ZBp9DGG|4>F*jK
zK}G&SSyFFn*NbLcek59m7`i0Mi}`{9ZS^uN2Dn?bPB!NM?936a4F*CmCbl+;1T~AM
zE!Ss$uLd9Jm8&YOC5B{ub|Vc<nKp|a&*Bwl@fl-{<fh(>+vAb#IB~r{Fii(GsrQ0-
zX+DSnei6(s4(8IBY0Op}mSQBtsMaJ+ixP5G%Nf_sPo<Ijx|qp_tPFQu=CR&W>w039
zvLyy1Tg0++`Y1C(i3$?aFcOD>5E7-T!E&%?I%i{jr$QVeL0~g@RjrvtH$Mkip3e$A
z0bSnEGX&X-@Qgv^_L||k(PrI^v;|x(X1X182vpI^-%>OUTeEiFQ628Q+|%{CBv^p5
zbbZkY5vav7K-K@aJV$K1QShf8p!T~~kvuaU_VmthKj!iJ4B$$q)1VJR7hB$5j$gI;
z>xSkjC#$%WfHfW?4vC|Ccewk(Tip5dA=j4|lvx*{Xn=~kqJ=xyv6yvLk%To=Zq4Ee
z(R1ArIXrv8^N&B_!P6%kzx^h07@P{ibNW&<DHKMjjBTPEc1*h~j!sUvy1d}%_?YYK
zE5^-6qaYEUJbB8y@4VlcK4)e`gpxJw7GuRz4%_XPyLaw#c7ARez-siqFb)H|7_86d
zP2hJ6*4+4)5Bi&sd%-u(e!a~9t^d8!alPtzeX`}|$JqqlZns(q{VFyHO~sv`pD~Uj
zckkXK=gMDy^oxAsm%hqB{ZIT;HlYeb8nhbQ*pKx|x6T0S?^+8z;pQf@Ywh3K$oR&c
z5;}8*84gw7u+b`9-OOu0g_s8J-h06D$*ES=)@4J`uW-QO;8d2hg-*KV%WeGMjF5Hw
z<eYojR=pV~Eox|P5mC+jIhz4)==DLhHEPBx>j&=Ly-%y!Z?=@I>84V-c=3$O%L~Ra
zbgiIQkam6K4`J<kdZ)D;eLHWbO4c0~)_=ch{Ji#|h`9p@W!`mBb_Pzti>8C}Jd;Z`
zuRvqE&~>L4D=re%`1vgKO?3@=`?b=vn8~d$#GDDl$D1-2-ld_rWt9OxF};=pn7FEh
z^g=!TFKad*TDLaA@*s${vDEy2??{kB&@w!KhNV)IGKv6rdq}X>a2@9^0V>LNG?W!>
zv_5A+=Qp*m*=`xf4a2Y@4P#%IQ%a26Et}2IQ{(*}S~MzN6tWBs+(M<5(lvxJYL$IK
zHHW2GL4aiK4J-<Rr06phX?8!uGHz{NdWVx>UZk2vZb(&@G^EkkR(*!^G=c4(x&;PW
zY1*A92I@F4q^JzARi#gd+5X+grD!mq=*(maRkh{Tzd6LfFzWp)HB+^=hwDQ|qV}_%
za%S%7<Z2_MsTR-!xvC&1+4r<E0f9Jdm{TH|rIGp&YER%?a<7I>&<j8mSfW+P5-3gW
zmBwLY2ymPt2T{_dG_NxD)_(S!>r#9sBE1@ZZNx3my&5$T&_E?d&yrLE1<H(6cMoI<
z&T0dSj`32dv)<Dfpv}bsluW6R%hCpLorUqtjRymmhoF-RORHBMvr+>gRHFIz`Zd;8
zX?doGe*iRqt|RIxpRl^%)%N&#3$j+DdJ4lleh5f(BrQl9*gQPt!I$3Q`s|$f@wtkY
z3+7*EMIFx8=$W8{ca=^P^U*O7=9RD`aM({g`SerXK0hOu*(SI}k!C3qT4j_<DmyN(
zpOE+0#8%i{o^xSs)7;_R@4^1C=f#T`oSfd}{OrQ2i99N-)|*k3DH0vn+wGBl7LDuc
zYc4M@tkO`+|BI#TMBn!L*vmk}ZCZTw>sOI=tKY7Vg+}bD$5#N*#^<wuKk6|}6PxXZ
z<KttludiSFz1pv7kHoXH7i_mj3~Au$@Qi=&JO2;fd-Gj>_Gf>V?RLXFPnHf?r)%8`
zC#<f)m{7ZC{m%6ZTrDiCTeM8>285o*3>6!jYD`^?>4;Xr6;on+bj)IhQ2W>UhPkjX
zo;qwV-E%yJigfFTexICK>5+2lV&(oh4f8KkjbJb{RNW?%w!fl56;d3yd-ootWb!;Q
z&-!<r_dI+4gq($GQUlalDn&{MfHy#Ezoj?-TUH>}f4^q^)<qt#0)el3-Rm3`ET|2b
zdVJ9$8<@IrK23eSa#1toAS|6MY{S>*tZDs{tNBX$wN%^SmTM6UUVGOV(Lg|o3(R%W
zl{O)9t79cZ#9|Fy^Dxp+Z1XV?Cg>X4Kv0$lZwV$D3Bv3Z#x@>H6C`M#-qIRa%25rn
zor$J{qFSpJ<XMm(2feQa)n|pEA+S{ryLX+4jYkQL+kx?DV`*7spkf-;IU;JJ(APV=
zk(Ol<s`&+aKon~wM+@joD_%=QB!O6$oKT?_YSw)~f-;gw)2>m*u(k@dsMVQ8YQ`=?
zGjsYFtl~97NQpE=Vj2`^m6=d8shE4kqzwDXc0+5z7$ZZ{aTbHH8Py}u?NfTz!meEl
zW4!py{1uIXlMtD+0@ryiOu6FC14_{<+W-qgkEOkvgL2bZiWMhWX9_L6ti(1#8W42P
zY*l3mDVV-4K&TUObIdRdXb)z!M2ayokeFlOXge?t!U$Yv1y-$zVsKx(>={SPlDWU0
zXC-S0)^itXE9L}JO}i_QG>fPmRT9!Pb0ZO2(Xr?Aw9sg!(B?_UOLG?ueeOk;abC)L
zQ`YAu64XAYX^Oa%RchA(X0zV&CXsXShzPMLgXo!~M(zE^f868s0bq3is5h7uxDwor
z=~c%9CK`(fT2*yopP_{pQJ}?v;pBve@4v(0`30XmzqCnfgWscH-@I45V{N|8+oxOl
z42yh214?5Efft`W=IYrC4p)~P-@R{!t${!^-9;wIfnWXpFY*8UyZ=7Z)j22Ik*8-@
zeD>KR=F-?4ZF&Ce8N)F2N%YC5pYX?j`e7G>)3mBJrp57J#o2*z9H@0eYmJkW6W(~E
z@!^Lba&d8?5obyaL#4HeTw2E~eqY7cbBj2dOTu1t7+#juf3NRfyZvHrZp^n?#g{(j
zH4hO{3!QD?=;(;uZub&cP>H9e&7VJi#=GynN3DfV<Pm@Q7ym=vdFNf;efRASd|n!p
z9p4!n<Nqs887wgC24l6l3(MGBB2irsm+xyDb~=9OMIbRmZFJtsqQ|rkH>)?_Q}pws
zpH1If^*)&V!4{2R>4ux4>6`4}^54oZ?e{yT`9N(&7ec!ZjGfXnioS111NZLTSM6YJ
z3i#$d`)SYR<#VQKTB6&YI$E`%vp-$-!2;OUPxw^~;18FR{J}r}kq&{FVs9?*Oiz%d
zy;Q%h71+_Q4t+xf%pJi;nXX8l0<NuDD=H7JUY<qEGt+l#r7@lTTLun@1Y$D&xH=<f
zb|f9ME+M*pT_6NC6NcE?U!nG7amxz;Zsy~rzh7Ff(H3t~51914jkdHPC4@u^${c2A
zXF{-Gsr9ZIo-T(<O4>_ZD~=;EDu{A*m4efz`J<-H*gPBiUA;pLx*6vsDACK0gao0+
zE}2jnL$MTL?DK9h;IeaW3;182pFctS9@}?<^)-%Zuz)Tin+;)(P=x*UwVBK}4*MPV
zPe(!$rb*X`AwaEKTyZ#L(h!kB3z8Jj8sKgVE2dyyfA2oGx#<4u=F5U=O``$UVp1bK
z_>75aRnpq3PqoCQYR8OhOrr;y4cmbHuh!7oq)5vXaZuBFL3MsZtE40xjRTwQz?dRK
z)#4H`w$M`W9vItXAcV#4%Ey`WV>NjD2jky@t~qf?3a*0?14z~Mztowz&V4>Ro7jCG
zU5a3ZDT)*HUKMNa*Ow1t;(Z-dd0SPlKeW==&vyZ}t~r%HN7tQCs#!ltc)j)Q^#R~q
zln$3ltp_cUm!?e$B|z1sPajsWNDRFw--xPM^r;AbF!Wo16b5~B8j0I`Z0{U+@a{XD
zfBYHy&z^upqy=IVXo)Hn%;;HA$~Y645hPTGTF6wU5Y<o`VW_&n)drVW*X$oXp`KsS
z-f1KqmLMC76oiT2c=T)hJHPdpC@&tt`3vrk16Lt(F&Caaf5BHyzeEawv>6ySBPCb%
z`yG#;JmSIqH#k2(*JyXvVi>(STFF*gWeD1LQ~_Zk4T-OO<*WSu@Be+SuCEahhS4Gk
z(Wcw0bsu+Z$5Sp2l5ZWlXRHR@`ZXQ+ybR;kn~-<$cm?kMLXOtd?qI*)b9{8f@zIvu
zZh!MUu*g*^*Sl-Z&(FAb?;ZyVzgqt`Kk--pGycW@!N0&}3|`V>8qf;W{M9@%qb<14
zZ$vPEJO_a2M(%*GRplKtNLa<<M$L*ZA_6T6b3V`_jN4>4uF<MiTh?ZsRi*-SW+0F}
zO&0=jh_s}OK+B;A4!*!6+NU<>N}fx9roDg$dW7of;AEhrm6==`bH1RpV;~|jKoMIy
ztu<lulxAeJJ>l*>J?DAO<T5kY!dzzNS&gmdTnLi%{5;r`*o(GorL*;O0T6!PuI&lt
zlj%SIqCEcophJwA==adm!6*<cbJpreV{dgsR1;ebGR-u(*|l1FEN8u|IV%9td*h>L
z6@M65()H0St!z{1>?akxsK^4p-W5d5Vx=kl*VY>-l~S{{nDaDKT=7ZEBXQtcstMhF
zLzR{84iICc#Dj&VW23c7GapkngR<_vhKTwfiUCjwMb~z#*pDG-s*cJS#M)`4MT2Gq
z0-ixpA4$EAVmd<y<h{#+73MgQ^E7Y}wd(SF>smJfO~spEquv|0VKG^TUw5sv;$2&$
z-$x$<#e#*@EUjwn^0(!;iK5q+(v}nxb>4BbNsQZ(!`RsGW@;%+8TN-_4GCa7DwwP#
z>vPtM8(L}0T5tONQWYdO!3q&AfI-*NTs`&cy@rF0x7v-Er`ak{ZQoX))2PL?)-f?x
zGbAsnlT%Z0W!%O9(y~e^hJltVl*%|J#xW6wNRY<v_>!-E@O8fT_=>aLp<{&}JO&-N
z!9c9`LJp1#?G)QWbO@vnN#@}@4oT02YW)BMk~OtGxup{Z+-h^2*VjUtz1KCst2$p$
z_B2{eM74mVR9ko5Y#Q?zwm{p<Z~Xe)Pt-rF<YHOIKVk+j&l9Z{LRu=jmS)-mguw|=
zKo%X)?W5h)z5spU2MBF|VPH5q;qJpXxck;a<`-wQz1vdMt{q(TT{^X$w2rnJFUu-~
zX4;viH=`X6Ts(Qqe0ENpARI+X0%0PR1Ap)N-{Rl>_P@tGUDEQz{_={JHrzQm=JM>4
z=`eA5b;&r6%(-y1-E!DZw9<I`<OvV%KVV3zhiI*lr`gg5AjzUmfpIf3O}R%Ir>Cd<
z#7}&c?|kPw?Djjo$F@ktU?#Q22cri?tw*+>d-x7~W3m{dU<~Q{{p-~4ZF1#hmEr3j
zf-c@E1`dZkC&$NZw;K+J10MYOBh_fFa(4cLqoZR|7`Tpme(C&I`TDPYjeqnn{8wo$
z>os+KnKL!F<O-|*?FiS<u8*ERl01+hP=ntX=?kLE4+O}yFz2dm#fHs|EA<HT>T1W+
z$4|*ud+uzHIQsuG_h-+RWZ8Kq_S=KIpWzNMM`UK@P*os6g5W4l0-9{KT9V1yOx=3x
z7d^H906pqIk?BolZL&6_tYswiBGb@q5Gj!jut5?e2-IAam19hIJd?Zcy_X)=+V1B@
zWfck)Nn?B2t%$sF?zzXe?_sTPec!j*j;3j7mJ5Pg=)qR4h<r`3zA9KZWK!<-C7yUB
zMT5ZDS{Vu<=Yj3^hSll>=Ns)9k)5GjGoJ=@yn_!cmn$AT(6JoSo|Fdqevc1xd#B>F
z36_pB_oJ(pJL>0p6Xm;uIv;I{cTvpwy^g=Wj?1}9IM;HJ^V(W$bTt?-bku_Cxs9SX
zG*0`U`VbYxH^puirE%2YX>$Mf?kMn5zM)7V22YGF-Z`SZHlW=!szr}G@{IJ};hU)4
z2Z~q{4##NW-SWDYqs0)HjGM-}*!)Z>QF1arNkx8qaE5Fe5`Bb(mXu9JouM>DA0|{8
z3rfAOo`uSst;Bal_Q0ICi4+<yM8_1&)cI5`hqMQ{687u9kV{d5z6l{haLOA)pdz<A
z8upRYQ*J`k@9c&1F6Ru@ue-<O0cYjBX_KnAE4ogyS=~5?s(PK7cL$*h4Y7@cq`~>3
z4UjWA6+(zCmY$&>Nr#@>{eiY!u(no(HMuKg<_%*VrbOIiXXCv$`w~k+CQhRznuabr
zrzPhmc)Z!J#YUCQ3$NWUN(lDz^-R#7zus$i4H7h7gTAhDGyw%^QcrBuyxTcP2~h*;
z03iw?s#V$-@df|khyMYCzhHQA#V{NU7|@{9hd?y*Yw!KU1FdsATkE`1y(B?<QwMG9
z7-K_lf+JHTAt~YVs8j@7^F_+!!H3PYTC!EVOU=$5qAf`yGcsxO$375aO8CJVLik4a
zvlyxb;LdS;d>=FKSAi?a{~$`h)_w~01j^zFE~;AAM{=o$<1gNbj|rSGZ%Q>N^xor}
zNC}Q+z2d<;&)B`XWc%eMt!MIj<7Xhz+G?oG)Ij9;_@+kdn!RQ=7Z>z58{DBFr$Tl;
z$c`Uh{9FG1cm6B-?FG>Zxm8}&c6VTT@0|DEdBPX3UUPGM%fkovIc#-t?#u*Q2I1Aq
zmppv*h|OlBFnU(API0IufhZZcjV;>VYc!=`<o^8!eE8u<{P0IVU>tfVLfdKs@FHqe
zF=}V-BgATwvi_XQN|eGpNvA0CXn2lZe|&iSS+76qfn0x^9*%+i;lSD1nR3;Rr1s7^
z>&&1n(XOwrc>4GmIVZjwzso=VSO0`R_pOh3=h<U(&(|&vQjYI=o_oi)?F`kEJD=;Q
zlsrYD+!>nsQ`0OHJ;zGc@@d=Da=iobrUr8ihk=)0KIiJ?YwoR2IA5Q#SS>ldcg{(0
z#IP`~%6#x0n+~hrSe^p5$p`BQK?=UMHp^4lHRsI!aNyOeS2WFvzRx(<;9X0%=%{uS
zI_LbIpt9>0oS&cb`0#{&zoS3w3C>h>Ig>+1@;uhZ!P)Uzum64Q?RR|D{h%K7$3G9;
z`I)aR_(RF_xfn`sJw}V@XK+F7PpJD<jreMi;<WjEW38*4WaBJp3CM{|>Pm0ODgfRc
zp&1&0H?fwvRQ#a*Z#Bp&MUhrbh?&p?Ts4i>U@|8wKuc9c&Ne(XX7QyUX8c=<_I1v=
zXbZ2Dv^TJg0r(@7V^&8OQELqKP%KC9leJQm)mXW5wxE`Z9mk@I;0!@ZBd#PHqdM==
zI!;wqV#d2c|9k30Q`fNVhu)jmpo&l?9%V(!#gNomT2_vtw}7Z3x;1HZHlB5kdV_Bj
zajLJG`JXvu4y9*IiAA@diGkpl0?lXv$(TmECL%@X_dUKX_z>|P4T2rq?6x>3v<vkI
zBm@o8MUlmn3gQ*jY`RX7Y!fon{VGQD@)H%QDB>s9<;Bqihc@7YP%=Vh(M7_b<2u%@
z(2(je3k6ls=2Y;}(*(y!bSxXsP=v80J-bbP?m}P`$j;-uCOaa9oBo=Qmp{d^R5aMS
z;!F+!pxn}^qO1_2nX=b?C2Im(Jhe@olUj+eJp=(MNJ&UeW_+$Rm|{MWS*mZU&L;~j
zaF~#^Z4rSe;3qS2Rq5w!$v6V8(qHCe+ItT{-&gmQFs;eS9!o#y1gMF=;4OBxuOR^R
zIpa$~gKN!%ypQVB=fZqQ+VzVz<vWO<pZn-mlT-GNQ3M|Yr;i`==z|-E-9WxR(27_v
zTJFX>Vu$rqZ&D7l7nx8XU~(>R9Q3zWY%gAO*lbxnI%7Xx@LzuKzvTJVj~Q>TaA{CA
za11QEmXr^io~{V%bHu~VZpY>2CA-~$MVB~Puh{MOh!n1_FM06bK3&(+_tqCN7u=X}
zJ`rLhjibu_y9K*(B;~9MuO{;3$zz6L;Kx7y5tUrrG#%at+9t8x?k4lrI@G`HL0AKX
z<8k}bg2y}KTo>|J{JNSM{z{iY=Zm(X?+13f9ry0t<Lc__P6vq^Fo)pT?zS8bd$Z#y
z{L}K^@$3KM*Z9x=<Nt)Vu~L_<rU^l)%hCBe^LFCpF%QQSB06XGIUZ5F`XlskHXEoZ
zhaR$?5E^2f=@CUL!suvY!`b?TakB%_9-prFbi+U%6HP8iaGE@fBT}r_u6?}9K~Z(?
zcsG%9e(FXu1zK_x0Pgo}wmYEV<n(}dEirU-%LSz*7K;V<?%mS^V|uo_PTO|D>B%|g
z=MT8PzQjq^{;)@gpZ@vPf!$XHg>RyOHAs9j_<J*0{Mv&+3vNnvgEUgR(g&)!vIhdV
z+IJc}TE+_AT02z>66bI(+8B-aXr+JAJ<PiX=K?-NGg<afYALj_szD8yH4yL^6~Yi4
zJ)C9)3M+3a+N`vOQV&G@w1yR-(l@Mj7OtqsPjaz}NiBt0xfzd_2)T0O=G~*xT+|go
zRl{Re6mJUjy~DXe2#`gLE9vpkjChN~l>@QA#*GU~Y1C6MJ3Db5u6pnR*yk^1#pDc4
zbUsW$fHzwcy?aGi$CPWhr-_E>IJ;oId~0x8g)}Iad)hm~EKc#k<7!(xr}QUnbnVPW
zqy`iq(yG0S=&1P+V03sNG_Ywsq(T}JqiEZ?7#yn@xa|jCZ#D?fHH{)&hUkx3%Za#|
z6D&5LUES|pt-g7Xf`X$O+(fV4#!61O=m}D2nm{nR%d!a!#$Q#9itejTaE#&^iuQ(=
zBKRf{gLz<Pog<|`sEWGrxY!^B@;ER$twy*a^iAgZ{uTG#hsu2vIm*uG600*HLzt-t
zIHxp%oUMU}m5VWy)~V$R(dKu`NEUKYI!j8Kek_cos0ECbioN7J$C;-PVv9qA!0lMI
z5YWWHIO^YP!c{r0g=|TapMtvT4?6{aMNvk{_V=1}$lLF4Z;t>TcKC6ilt$Gnb%7{|
z4`DV01%75zuh%h>^Hm!;*rfE3W1u-*bN}7<*j#ORz5SFtX2Q|oJRQBRA_`*tEA`);
ztLjgb$m}6VrV(M>+_1fT$)UetzxVuCKlwlN<JW%-!ycbU9kh~(&avn^Zf<v!{=n5{
z!`1bM{aC;Uniy%D$o+fw*zNX!DkERLe#Mi=Pe^H0^`EF`SZp=h%~BZp9`8L}*K#=Y
z<k&EdiLUK<=bd-SIrGJrUw{;vM$Ldius(}Nh|e^JzhIuY`h2EmP~)76-oE*2j-tz-
z|Gs}}{P;^;6(JK*JBE|fHMh4La!#C{p0e3&tjYPz0m?byo#*EEh9^&+kW*sY?fC!v
z@SpJ4|J+~U!w=sjc&Dl`!i?0F%Efu(HtwK%mHRa#CzS^&&dQtiI^6UlN8_bGAN<@C
zb-GuSqquB4o;`WW$znz72hF-eAe!k@<%{)0&tb3jI9;da(dt;BYQ0=$bWg`pgUs4t
zV&V@n+cxE#*=%=&@S3Js5~CxH!lCaO#seoOCoC3=saNdrdiFlhbt{&O6>YmfazAI-
zbwkE|<sNVx#Jw539slXxd^kUM-n`|+`-*Gg;)H7JVPtd6_%~<}ktO0wqV+Xs8mT^X
zB4Yk%VrY+0-GPc|@6WnyVvK4RVuu!<nd{^fO(;3xN}A@s24-a%Z>eSBij)|_R8W<&
z-pg@<HWSh(0({BFs+m(86*SEpLl*-+1`Utv?}iB#)s7rYWjc+U%TzwKr>130Cn(=k
z6}Xc<4Ip_>>TmIj4&Mf{Q<c5<<{29T!AgnRbJu=0%*NW)wx*sZ!KoKu%C>)`L`ma}
zxD+&CX&No#+p`2ala2}ixFf$e`*Pk-s8yw12x+QFr7TE4Qe46ZAqtE+ldLCv4$eAc
zC|X>`YyueV=n=glwfn<?u50jk`e86HKP@l1pp6NRfq~F8x{qPT>&6SL^T-FM{CW$V
zD53`6_69KqB&#z<aE2lZILSm0t#fqZ7*bK0gy{R%>Ui`HBtz=e7#cN>9*S_-^(=#m
zTw)gfaQDaj$IFjs8wF0B@eM3~g?e98qFaM0b8n#rz4MVYJ;RbogW$>tMH3uv^Ks(!
zik8%8cZ*}YC9I2?r|lxC6!Q(Nu;@(f07X2#sjj8yFLTb>yq{yByR?>abkxq4HrgWb
zSHIpm030q}!Ra}Exw4ivVLBKUcC}FA=w^<~QA%zMlR75GIK8&Ewh2L%v)UwG#UY_<
zSwDNk)BS<{)rR5u6@hG-uYsg)M*U`0o<7_L7$?N%`1^%nzvcGzmwb8sL;mYO`B(h(
z<_EZ8hjR(%YKE(-rLJvQb^$L5H}?23L4xJ!l7|l;Q$%Q+hSSqEyX~G_3fEVcMDMx(
z-~pS>R#BUzrQjIiB)K03?G}<trXPCCBvn23?45Vn?+@$`dvYGB!H=&c4cpN5K>hj8
zPmZ6Bvq+|n$yffIt$>Qy)NdcX{wO(n>r3jd1ICoN-!16-fz4*i!-o$T#}?=rM-6nm
z*HUnd4g39$aU6(IxoH2k`40d0Kl*?2U%d0bq3gn&;ZOH)mxFqI&-G=+)jO-a4r}oT
zRQ{<$`CGM|RS(^1tX&P#%5;yC%XGZ9ZA&;iW6^b1>ekKQ$3VB}@ZK||k=xsAUcP)q
z(>6SQ^qBQ}HI)i|(gULFx~XLAeIVz`=MqB?Em$xQwEE?pHxuoSk`2wU4i%G)MwmUQ
zD)*`DmNd<R7(4cdo%U8uS^cVH+CA;Qk9cCGOdDUl=i}GBnIOLT?_bplPU<25HC^6&
z+NPaQ*7_{93~YhNRO*Z%1xd;s3egioG*nJGa?TFRfQjgA4HvUYiA{r#jiJo*=O{&R
zQ9GK08hA_2h?Re_&nHFZUbKd4PnIDzXM-8V=Q2N=x-M!c>YVmYs@%kiCTi)|C<QTC
zFlQo&tVl}j`Z4u|_j3Y;*@!qxkgDPx(bi;AtaFO-t5gq?DdP_5Hx%EIaD)ghR)pBf
z$<_bQK^1*qieg|YkLnGh=EXLTr}>Oo_*mdIQIRCT9RqC(PIW@gbM;7pe=KAwQgGQD
zA6@ql#PPX&%7HDuXbG|(Mi9sTFfjB&w@`E`jfsp}=Jd(L0YMYlejMnUNGd{~igsv0
z{Tgjx^pj4X$xX<zxGJJ4Dv+}veaMOE)C{={I)5AIaK&>N6Uk7_C_)<?F|_KTCqnUo
zJ~(z`F$*K1A4dy1Rn$|4o_+wg5F|w7p{j*an?_b$WD^hUq-PvPeHLB}Z8YTC;)w~&
zsB1;$wwKu4G?)r7DuAh(eID^ey@e-X)3K&Nl=d@~_b233_rVyd7{zHJV2Vu3R8gwV
z9QCtRQ~cTzlBue2ytmC3THbrewb_NOk+-<sIsm-+<flA7zo#fg+o>D8S2uF)tvA26
z@|Awk_2$MteNRNy-vyHBVrW-O?!EU8k6&N%e7mRYdRwet`Nr5eXArj;OV;*hnv&9a
z=*!59S3lwRU;RV2-GR^_h`!*11_1g#igC7GN4IEc9Gtb0eGK$t=CIkb{KgU-jA`WF
z`I#E$j+w*Z!1EW+iA`j+T+is>VIX)t<fVZ2#2hHOaL7Fl&vvuLImhYgDIa|BA^ZJ~
z*B7t(;`2{gEEkFZmZ~8-%1nRZprcvHSG@MkZyBmST~plmO8|nu%B8?$O%~KN`}FjL
zSFc{P-R?L)Kj-rDk`N<@{b52m(l`>F=i=gmci(wWtx8<tpM3F8`JaCFclmQ4en^c;
z=Fe;#4cjE@_towK#AJLTb3fdg4bFoyBZ}Y+QLOJ<81ukpbIWeGrIdlE?+U}%b8-24
zQkFJe%XBU%id>t$H%&8jm>GvXmzNiO?|a{&>pEiaESDXNr3yE4)>3%WM3&1XtJR8>
z22viU%`gO=o9eNq{F&u)i4T#pvolUlPgyKZaYTI7@;X4<ws(T8`3qubSuRglE>;|N
zH>9NW1?`Z)wEo`=+`j6+k6-um0=_r@?U+0GRUE=!brM!3aJ?#e6JjK`Ez}aDQ*^C}
zmf*7~hI7$C(20(F6(2}RsvIneqWBPSu_3fAp>6Q7qlgC4t|nYl@})?xt45%nxl~>M
zJsStw_POG%D~)J*Sjx;LRZlMs09<t^I9eUv2V7(5pjlx!>SqpCVpi3<<V+fqavS|D
z4!h$xAaeA()pA-@)U~}<wtnwIj9T&xA>c~Jje8{QDNci1Zvn4Zk7e`az)?jtx^d0i
zKI2GZW=BUsT7BkIFz4#r(m=vT>l2Ii>Mn1L6EFqJV)s;&tRhFh-&rthEzH`Wf*u)r
z^3aoVeSU$i@wk*}e89QDb{IHh$H5XuFM<n>eLpaaNlV#&Myi8(cpk3@UjugY!WDF%
z%{fg8w?}&uXHgo}kTa4qF$Bh}E#xYdCQD%;(>0O%Crcce!<cECNERyv<JDS5gvJHs
zg5nwatpY#`HNzi7=*L88wU=`e$BUZ{Ib&Q?eO8$a;->B^*nBFM;Z(#mn|C8NXG0B!
zvm$VnPpYdx%l%E;lC6yGvrL~`q!N1II0>++NOUT4ItEs1f*XwIUqmL(y<n9mKvntk
z?>XM*=74iX>AyPI`Q@*-4gjBj|2sUo|9}-9?%gLeEl?<B1XqlVtCzD9*;if1nf;p?
zfAx53a1#nxLclE+EbpK3@V$4~zq;V&(=W-X5D7o?{u)HO$r0X*xs+FU39Kxb_~|!C
zKK+d^xsTA52q>Y}l}Z{_1L-WvM{Ju&8yt;`jH`|@9oUIuyV>ycou_QJTL6~J1$#M6
z2+;HA&w2kFAJVoR{m@fHIPCZ8NFO64h^^ks?d>gBSJ(8112JkTB8CWf&d<)dxOmNO
zxBZ!M{N=8hS8xpFeRUPh-ThkyS-<L69qskM>NN+PnmKK^Tb@39Lf0w6w^}V(tylDi
z1Jbn&!$^$Yf?%QV4~(Oh;3NdT<T?NK_x=I@`w#wiiqw_kAPz%D?goHg1!Zdfg&m}I
zAs@r0&f>y(gAfO4?0NCx3tqf<!RhISzxbC6SC^Ok^Z)+$=(?7s=}Z8m0a@F&v~9zB
zeZu_*_vzY|X41F9#l;0LUc8`dJDxs$%A<!5S*_N@Hq!U~)bXKd8sqHQp;m<Al!tUU
z9O#+_QPi_Cgh111uo*&Ny;>21634TZ)N1Lo@-OG0Oo5#ywlqyk8nu_JnoBpj!Tu?M
z+%f6-tAfn>{Wk->uR;L7&Z`P5z*MAEMAv-BLewAvNATH1GWxUC*RLcsqE&HtLCESm
zCT1aFs0g)$Xk0Sg<I{vN6)B8a@RFb;q8Ku3LCYMFI71mS;wsn8rgJsJ6-CYTXEb<p
zML%b*s$5r9;9@FGBjJY{T-nc6)$W9>3sdN4=!shB2%$KN@$G7YQw$iCoRN}|tkyBX
zsabE-Ay!dFPLN0NnUoWu90{Qe1W+2AA%T9j2@$*3lE*v8crm0>4|x6VOpOs#bf%&f
zFV$YBSg^>^KB38wo*9}f78J$5Gc5&@WCFtJYRS0l2>U`RNs%fGNQR`ri#Q+%V=f$0
zVVfMu#01_66R%9PyHF`=)YHUxWF4eWBp2<q+(s?E)|_=hwn5FnE%236$XTYb8<J3*
z+ES%qVBaU4E1WG`e9DYDv+5#@J>!^&!B1-J+OxVJwLN0n1WMW=z~};DxyFTnPb#`f
zMOdaK&b7o)7|X2YHGFv*S3&oH>Is>pm=U;^>fO;{rAdIkwu)~MOYTCz(Gu`9B9zfV
zHd+E^xg`Y=ZM%1eTj{;EwySf?_TA$EyzV8;`&^}y)PEPL35*7}`Z&Z8OOPa_^6$9>
z{ONbU!}|0d?=>CkZh>1akP^%uKxXd8apd}y*F0j!!QU(b(aklQ2hfl+**g~NB@f?y
z2AeHEzPY8p+};g1Z2cCdf`*Df?6>)&MN5G1emL;SZ#CR|e8x%C(t#`*D?uxHRkKFO
zPB*W`VuADrIv-d!k$dYUp{UHhw@wx*k1SVeWULZ=h4P<2f5Ee7&xkRS%$se$-?LaO
zCa<~BG_-BY)zvjOw>O-fo~z2Mh_(09lctnR+csA6Oux88$X8s|;)NR0)V=R3pVb@h
zZ9nSX{QlQ=4(sO{$C2G`&*|wI7Z(@Y-rn-)(IXD~1B*om$V44TSv{t1Z*Mq1zi*qH
z=LgS!#1}80^WM{^Q}F7J5&@W>F<?r|V+h|*BcSJt9iDeifq6mKah-C!GID)$$<6I^
zM1-sBD}L|y|CznNr)?Hg)7Hge!TRKcU;DLR<2V1pZ?ZN8^u^UBFJHc(ZB?PW-EP_K
z_ne()wF$9}=*@MfnWzRdDJ3?W4gFytL@f^=`U5E!Zf|!8fxgc)acRiOaj<4xLu4)~
zi}l{#ZEkq+@)cKC*Q{5amVirb!8lKdH~(Hs%g5il8ywcx%hBug;Gd19>tS8J-@b+k
zX!l$BZQf~V)!Hjq;G=@KVuWq)oYGOeM{+$}EbA9T3ntGnHH25rv~vBXvZ>y#DmE?n
zMj*=wlC;Z=sP}2`!9*OHtiO(23ZrOwuNsAWuUtEYdq64@Zk}vTltxhJtG@p{hibfo
zOh62XCB_;RQvnKb8D}1wO4M^?LB%9$&_{IYGlx@$PRYiTMAg!*_cVT_iH;Cvu6Lp6
z{d#98maW?YL5>7gcKAmh7)`yNOP=;ksR6VUCI078qsK{osbNzP3pi?$!Mx(@>h>`r
zI1TKwGygz`q@>Tj*!|ZypeT<pNW#fz!IrAbgLkB2eP^9Fn@8>GJ7x_8rRX!LNNV(9
zDhC(Q)}c+?+NyDi#+zs?*Y#0-8=Z3QO7UhdVt$F{0a(wPiefqE7*pXeB$B8qxr>g*
z1w}?nMd4IT61;wfoYhWd7zc6;EMj2QHS}p@9C}>0QssAW3`Hw{f1!N9ijHj<2+}Ce
z-GWp6T)GZYyBFyI*F8YD1(^(+T~Kir-kzNK8&WNooc7+0&e0-dgd{4ilLC@d9qlS=
z>}<_ala%Qcosm9~$*CeZZ~VeE=_KlFs_<hD3@r|^WG#jU%{w%Se4Fd71HjE^pYw6|
zFX_%!G-n;nI@r4goLdmRPCkz(x(JeS)qlxVnB2&ewt?#6n9WZ`oSJV2C`M)Qg1Wf-
zhSmKC$VY{judg}met;VaZd9^%LP(7xJ4fT7DIV_vV@?wpTYRQ8!VkWA;5&al@#yIz
zPR~}j$!5U#065)TbD4<#f)GQ)X4oRpA(B|Pp2NPU^pVTgmppv*Kp|%jXXhsj{Xn)h
zKD+If*B38%@aO@hI0}wDj0^&?ZIDt3F<M}!jl)wOI2`uq0#$^;JI{K(rfDKw+wto4
zrA^yk=CW<<2x+U|c5co@=c{hD{A_^dFjef&{Lrs{2{SjsDH{6p=`&(fRq!wjoUGRz
z_6L@WC7aE5o;wuwhaKnl?paw2el&i<AARy&o<4p`?c;ik`V~PBOdAjF@u-WBqjJ`C
z0(d)kro_bt06Rg%zT!Bjw1Y7=3*LF>1Mc5{LfgK^`~MRU?my)3{LSB)(8}%pz|GAq
zo6U~>en85O!(IhLV?T0zeaUB^e!}(D1*az~y0#+?1O2|Iq=IV#o``2wlU44C9srq`
z1J1)R^juxP=IZ*2am*b0o|Fn9bUb<d4o@CGqg$*rK&WA_R9;P*H`r<+f@8ni^5Vtk
zAeq&w;hm?CsSVFH<Gm|TI{umwY%HK;`qBVH4;KAWgJb<uzZB7(%R!|amG$0s8XlpP
zN;FqueT)$wv>jS03874!W{08qbw>EQ?I|BA7`fiZh`5NL?V?1A`If0|khWK|q}6+T
zsi@N&hDgcGrORUIM<Elnou6|PM-WQ_0FGOUX@a8Yx%a`|Zcb#T<Q(14qzIWzXqpKj
zt=2154UGB8&HDH<2OcGn&}WGAI1>YSry_$$p!4Y79Y&?%n9|o`eU7`{=yR$moZVZU
zPqnmGgq#v-7^gkOTXjlOnmV0SPPMDbOA97FnQZNOoUf%-RsU8~5p0dPQYPN4exR8j
zEJY|;J3D0y^lHURD^(g82Mg+oICgy^75lxSl_`&Q;Bbyfq%s;ux9b-6`LO50gwl#}
zEn`&ik`qCSo+o2%59d`_sAr`oHd>1Jp<#dM8B<o#gQ3WWRA?6sLrS!%20n$RX>r-9
zO_^~qi@-39jE&M<7EMsvQ-7dYED>vynzGP)As0tzBEnG@AvB?)h$TPTqE7{RwPl@~
z0(I{-Ky9Nc#>Htha603hiJGw1eO^5!hm>LHwIrXFJ6-Suw8~l6MiC*Dy8;d^voj%)
z!Ok<~M9#^8HYZ@}+dQo^OQxo3Xn;_{9S4kG^?K_7(Dgm9KK=>q`3b8BXT;@#mOu#$
zAgaM3hUQuJb>~CZ$hj`s$*#rEBvp*e=|qio0kS%+``8gy3%ZAJ{?Ui*E?#r^^f^oK
zwS?e=6qQ(=i^CPjDbd7;lcMcjeBwJF9r)2-@;rX>h}Ei7w4-(>DaJw5dI;3gf`~md
zXP#Ng?_ua!H=di}0I_4Y-E(rfX17;^=;d<3?bV*(A_d|4<`Nf$^V0{4YNSF+nhl1g
zp=m}!44j;<5R#gJrV;N0>-7nXMaS9M8Qbj*UE45vZ3>=_CeqJ+e$xoc-R~G=edp(;
zB3=edGjlEG*SzuPTVL}b<%8zh<CqwRk@NF&UcY|L?d>g(A3b7oyJ59jvsxDVelVy1
z#Fz(C8mx?|rIPRe;79z<Z~kpoi`6^`QqjJe&AE9DY&^W!p`wlndKE!jx$~LtD7fl(
zouh3R+<y>RES8*}JjFTBy?gigTYvL!S>N2O0*}px-LA*ENbr&M$(ob(DTjT}ix)3=
z`SLlALf1x`ChBHoX0eh5raqJY&cT>dVoW2_Hn=R@+}!ZT-~D60|NZZ2cJHGe#&}|A
zIXyk+-uVOVRqVAaSwxZQuIujlXn~anRmI*mp51QGI3}Enh@^>ssGsL<S@-TWK}&bO
zwB?$DGdT+-+reF&epNdKJ8a!o&_RBc>$rSV5!I=ERt-1{+5u(gNfx|SWF=b5ujmBr
z23^G<%D3y9hU7fss0dR@8Iiqu4B7!_vxo_Esz`%7YC~q@<gDQ9tSk+f8dVXF`X|@{
zhzcH532F*==Oz?zLcU9~GH!i-Op5ZF)K?@|%VmPMlJ~UN2&k2!dI@GjIf9SGPQ`<W
zqlupApm9Ru)JLuwXzNI*$C?;nzmoQyn~R-wZdD#-4G42eq`uc6t=f!Ll(|Nhf(2q#
zuwl=|nFs-pWcEKkGIhcbkObam@Lu_I$&ej9E|`$Y(7&8AyWzkeHGiTzj)-#Xd@{AD
zw**stc86l7|Mh8$nhyt4WGCxD5*rnb<O#^AdubuIjiJdJ@T+gHCMzLCdfUGGVIVk1
z+e8Cs^la(J#5kHaq@huSz_w46oH=O&>xETCoHM$T8c(YyW}G~5`&<Z47*b)7!dMRY
zZlPxWQF+JNd5U)$payN>hEs7)T}{==po%|LoTYnu>z#UM25)|gIwvuuKE;B@;!tsw
z&H+!vheC)+sS>4-hcfR6;V7xn$;Q-baoQ7g9Q5qes$R}wfvK*}YTHyy;89E2Dx|9W
zxa~i7)HtJNyv_CYHZVSrb_YKH-VeF==ow9S%Kfl{mBTlN1lflrctrfv0?NaYEW4sA
zbD80|bX7}09xYFdDKX}(lOhJfYQg!_$MEeBd4AZ_zr11TBTXr|RKRD7bBxXtm!5vu
zAx^k?ApDEJo(Lb@=lw^g_~>keh3fB78<AJHekCH$SqM0~MaS6pv~5SqDrFBM<e_K3
z*>ZMv%K5zqinNqOl0>vDx@|j37Ug4g3swE(YUQB_VCR;!F>-!-#y0G=l}v1C+lJ+G
zN!xYo54H7CnaW4MC=R;=PTlzzsWA^QmoBG&6`5XtOzPH`&sf6U_xx2an?E+swS4aA
z4?XwppL2bErRKCb(RD33C+^+5$IF*5r|i6xLO%>F+LgA5@}BQ~`8{5|c*RE_d^Ce7
z)1qe%c9E%tQ{9}I#>UNm|0QqMc}H}S7=vbxXr{SdpHS6=sU%KLPf4k$qBb<z9j1xw
zb~jvHTyS-DN!zx>r~zQpMzVEjs69_X%s!?rbFsrW6D4X^nlo+F@zF=$)-E+oM@l;P
zOM&yVdwlT0M?8A`jILV(j+90-u<guv)x6~}4pk|P!@%oTU$SU)(=1u-JjxrYwV%h=
zyK8|kBX4)Oex;&;M`o!Lf+#=pKELWE%wBIb3lG+PK)F>|r-{VS_Z<2?IjKQ@@PW`q
zV$}W4n-Ij?d{U%I$IVxS6Q?DvobU{WXzC$H-4{chat_4W<5e1h6d>TemO|BcQfUh{
z_^W2Q^_s_3KU0xnfuWB7ox>HK^YyjPX*UCm@=&BqJ$D_>+HzvHNAr8m2IY1yq0l&0
z7Gov$;8a~1Ex@cnor#yV#2Wl0{yQ$cDe5I=hjc6ovb8Oc`hi*kuG~gU1an*oVH#^I
zDd$<Wu1bld^CLTifOEPBtGO_f5%%<aEU)xdc5T|iC_P@P0H)ZODKe_#RTvQVxiE}^
zxLVG4EV@?9rZx}Sw#5Y%Ur1q6|05hBej51Zlxjm3TdN6&j!E<$?@=RYuU%x)WP5No
zfm6TQ!kG0l^~r=l=4aUjMfz$_+r?r*DR3AQZSZV|%q}HH52fko&QIvNmi_*~VLXt6
zC=!ZxM6tup0(Ea?T{jn>L_w$62BO!#!rrN;p?Y^#Pr<pQuYpIfdo#LBsOIY$<ckPH
z%D9}hWnI*u&KdFam<wL_cd%*+P}#V}kae--ttKvYJVoyE1}8<iRdebZdYkL51Hhc2
zAyBTiJpZ@fW4$<|ZB{Jac^6KFk_WvZ4RiHinX3?Z3$i9Vj_GD+$n)42CmNYGB~=y@
z137w9<LMrp^5~o2zzuso*?kw;Wqipjyp~a1$fO3cPtYYURwIA-+YO5k9<n}N<AW02
z-Sqq_l&h0LWZp23?J(4@uVWkro!H%iKJDp(XWRF<X2I?4j{A@AbAJCmDHjgAPZ@`S
zljR9#_s<ar{b9g4Pt$gcLvLlE!hX9`)6L})*F;X%Yx;gL?<>bR4s5p@v&$*;eQ&)t
zfAO<BJ{s$5t%5Y82{MIt^AAn`POW!N@ze3I{8N7YFMZX;+_nviMaR%5K)uCQs}<Yr
zhV6FC>FFt#msfOM$7;1^f7q)KLWH63S#~Sie}x~%kNNzw&-w5hADSJ+e86gAqRLDW
zIZ`po6c|?JxNaK1d<3XyoccEneXqfV6jB<Dd*Up3vrM^BC9t67vYVSLUcP+J{;+4!
zol;7o6!XwZBjY$Qrtyu#Zr(t|!7vPzoLMXvv|Y!82aoy2H&(p&?l)-L1<rX@O=n@f
zJmDMff5^M<e8BqT4Ck~j^EfI3r~#nzS#njMi}IL0`{XCw+`Ph}Wu)j0K$p9L-*EtV
zM>HdD3hZDuMy-Lv<WqNie@E~2b&5f>{OYEF+D~S!38j*GN*Wo5fpIXBx@a;HL#q}h
zb~IonxD__k$D|v%51z(1<Wd-hgVNBdv9Q;M_MU7Ri3XCjjiwzJPy;2Wwp;;+kB$DD
zy-&6KaSl_O>b#!zyIhpN7s9*;*Agms$6eod+6pW<NAKT>9@vO(|CMWH=%89$O!PpT
z1k(e=+0O;z_&F1=sOZH{L6n=;FjJsj_pxHF(KIL*>w=*HYJsU7n^Ky1vqraLiWHA7
zXXil?3>6iO5UTbmUeTKBD{8kg0Uer5%;Bu_nMJhwuSVNwP*a)3ISczSu}hggX?4nh
zb&ZH7Wboh>U5l|{Fa^F*IQ2ZOrBA8JpqXNOM-yY+{Po#8N7n>ALK~z0KARsgD)&8=
ziZGVxesh5_!+y~3#YXSEY4km^sfdGFzU=lp#sY&lwub`}J#o+qShRT>UA3nIXXf5G
zhXg;3R}Ho^Htxj{95ii1>jRC~Dw23eX;h1<x{nx;<K5BTZfNklF7#nkq+HMzdLb90
zFGvgs%HNyBTh5vJJBG&Un)SZ=Av&Y#I775O<7hL^AoGcvZ@}V&aC=X+p!rp=w+;YD
zftU&_8ph|ZdG(zi5Ms*%$t>S{7uOy1ZfbgVth7H~Thkh=hw0CXMzoZo#;DrPOL;Dd
z!Waa<Ub24g38bF=X2<LA{Fp^n&%Z*!`9g}&v~alR`N7{l<^0`qViOUvD#E?;q(l^L
z)C^70m7LOK1#t&q5Pjfn+c6yWtk)}wILdBdltjPZL({Q3S<$UKPR~wRpRTyRyyo=m
z1n)eDVc^jB?E3=`A3UJ#I)<SK+qnAu4v)|*mKqQk0@C*fKKt|)!!WQ|v~0Hr`a`d_
z5!~Inj{o}?hb#49#x`MevDJSPnRa_A<&Ev}>%S_>>Ah|`F=XTMWKK^{+3$B&68F=y
zIzK;W7zX29DQX--Kp3=i<~(11{skhVX4uXnHbLtVs<&IUlXYlxerk4hgnqse5dXZ(
z8M>THW*85E+AeIj8_Vw0M=Iw8;>ljkp}q4Q4m)1Ie96_tYl72*PWy=lvpSf+l>a#!
z&(<<R8b_|LuPG&S|GswgShg$HixVC_c+6_Knu5QQg>JFn-su_V=jWWB-pAqCH9LlJ
zR0Pr#!DAXNDHFEa4N^22>Y9$|YZ=l|m^zHy4gTgan%DpF^(J3%e(jtrR*K27=;iCV
zD&FRBE=+SmleCeXO%yXGN-E&A46B=*9@6A&$U`!NV<pZ@PV|Qz5?Xw-#K#sdLLpN|
zwE|JUuMEYj=s~c4yBO#8&3RWq)Zo50kw<icn}SJ00G%(>hM+-qCg9c4(z&UOsV*9d
z>S;CPxcp?QU^}_ykE49PK%sHMm@-x3p@jB8_BOs|&Tdhei7gVjB!VBTq?&Ou%lIaR
zWkoRUtf^(+<26tfu2s3MM%I-&P&ukaJ@Imx5KR~S8;P(+a}!#FKq)4mu>cxHn5qb*
za&SvUtPIs9C_1JcDVY2~U5G>!*-Tl;XjLiW?DhIy&WA~P8e?F&?C>#=3OL^&1B_z*
zn*AKuG%--A+HKpgY&_99a?1D^YJ9F8gR**5_DL0}c#nvdb!+ckagLI7ZaZKo!fwp0
znnwLGvC5hAFsS-EG!5$z8B1n&*pqT012H(p{hrUi7&t#Wq3s%eB;V&F_vcvnr3rjY
zK2^#D9w{iJ&*B&{m+TRvJ5=x@XCWnZK`6#nZGtAVm{Ul7!j)rh9ja~K*L}>Ar&5rj
zz2K(`lFb(H<}NZ57ZfekYar>4k{HxdDWi#E1<OnblC%CUH+{Na4tV%^*INgGr3gVB
zU4bqY_FsI-7mY`jfk#~+t`q_EPGRJnb#X|^%H43mvY5&_Gv11?jhzGIsLCT3JjwVe
zB5Atyf^Zl|wPRV{;{lGtrf0a_v;Fupnp|*_&`=Th#yQ{r+YQ}=&Nhmqn?|+)#;h*N
ze1K@H8EduVbmNF2PQ=bqgr;p7`vb!;uv&D=*XR@b{f^u&=(jrt^J2Sq|6Bvg=o!a>
z!~Q@@BhEQqTc?n=RaCJCJNw<9f@8T_87c%r-!qIoo7-Eiu9VcQ2mZ04qz%+pMSxPX
zL;HehVnaPw-k4-ZsNcWK;IAeCAAIlu%f*6=iwi}d(x_uSj;z)Ty0$ezLZ)dNhGF3T
zy$9TMx1`k1rTgfoJ4rF~lNUcCAFf%QtQo{nTx)HUs+tikH59b~(tAD*{N}8%F890S
z-}$OOPqVOTMh*3u83N>>^rL;>YcR3UIe2yXnlC>8gw4%0i)E|uzm>0s;lTc|W4GV3
zTCH@wa5umTL3yDsU%uq#<^~_W#r=Et$YW-|J#g3^I9Z(-YLis8mlM0~E#t5!jZ1t8
z4C9{NZpUV$LAAD!+p*st*zNZ0c3U<#*Elk#>yDF0_pEfEaCH;>is0_egs-Z8kA)Xs
zh4$4k{i;5|cf8%+;<}pz$wcHI2B^I~_3&2XVrP3u6KBVNSIO&X#Fs>H8S%<s%Hu%E
zLL1eGvEYcVkX*tKBO)5mJIS~*QkCdMls`4kMXOdY8~e^m#pCgn`7m-bvS9Xk6UtW*
z6+zX#gj%Z7eyw)`bwfo$a7eCwcni*Dg0nuyVyy>7d4Aqh)B=jmRDqCW0$G)@X&}2E
zA2QN}DM3@PN<TxfGG{nK0}(Oq9+SzmLwl~5*f4~FXdZQ%Fw10}ROeV7yW>(E9PN$b
zrT`+CnR*qh)xi6B4kk$^RI2Z-0p81-o=q5N(O%H1g~<$KVn~I3!dk8w(kZI))y4@L
z+%3C?7<4|y*pNq=68MVBdar%R>z?R6v@y^%!8{lBJnq`o)X14bDfAh(Lt#isg=9(S
z(_mazN7F5E4KNm@@)T`)Wed1!7oE^V{kspi=vkedupg7!--L!#GAZpK3qyb4!GrtA
z!c9&d%AYJ|c_TKT>N@qtBC}xED?+KX1YIxVFmmV*^oN61OPmQ~y!8k^ZY1FX)uJi-
zNE;(#)Um>YR6z@>boWv%f99)}yel_Risd*KJZ4LZoy8cwsD04Xm8vGo&P`4P6CK3g
z-ADhd*V}Ud#p^-o91YO#w!HZ94;dQ>ji)_50kqWHI#0GNq!f+zXCBJTxkBcQeg@9n
zw1KIQA4t-6IrZ`(;Jc3aF!J!%62pE^+HDzLzF^@p>qjg6UwOj6`1O&oQ?5m*1Wo%b
z+Cs)#vuLKzihUk29)}Y%yOufIaL&<m9b>;|(JeSxbjTU(FK*b~Ub2ZW2)y(Dd#sl$
zt}iahX;f$TG~iH0v2%{g%S)a-d5ps`h@wOq1n$5o66T#+ovc?YoOd)$<kc%xbjF}r
zp=Q1_=Vc0AW}x-xssV}pm#e%anQ+;u?HTj?{ksSP>n8m8@k8$4zsGjFQ~pOTq?}oX
zhS5;J<!ZrpyCJ1S(=_b&Jux<%oSblboAG9=Q9ZAU2)nT7rx!oulYjmD{KlXE4OS=j
z8KaOwNA~8ah9?_m>yAg0UDedMX7lpKXP<%`Ic9~}{Bin)y=x_v>%Si%7b5HrY6$-2
zm!I?U<qJv~alTb!XS9vr;V@8&@ZkPKE0NaUGb3RYtsaJf-EPZnyJfvv8G4tvxxQw-
zURz7S!PJP*4_h`jmv$?{%NH;B<kL@$YB1XS?YO<&u-$FgZ8z+9Tb?|A$Y1)+zkq*u
zt|KSd);Obb)xM#=_dK>TmGJ7<?C5{|{kyNZ`}4xrg#hNPbs4pO5`u3OnZrtGt{(d8
zpC+O<9u8GnXXl6>Bn_055E(V2FNNrg#x4m@QgXNp6e2Mwxn69}cwfaMSb0d%Y;PG{
zDrSA*<~3XK?27ye91+#${aB9Bsd*rw*4VQ76;9ZL^A?C62i?<2tYxoKbiYQ$bkW+2
zN%1%dB>NE`&1Iov3K>__<IqXMI}LJc>qK#mTsxao<j&8ipPSk>D%C<tvazeT>}g`q
zAV)+w3Fes>>ONmf>~>z&!Ki1{@xknIa}QusgqD((e-^`(tW=*(oFIg0Xe9|>xG%_F
z@J%F&qnDAf6#DGgjY6N$c4>x;*8awV(xMh^!y>Bj`-o$+PYl*;m|DI!^wT-}%udK@
z#5G>k$lj@uu?eU8F_}OmaocAOLlyQoT+^~VIi(B-HrLlWkJ}cJMmdF|&#6z^gSYYn
z8y|3HWmH6k5R2uKlQFaH6Z<^Soj;*npCQ|8c9*Xv>PH$!q}BD7&6GP!VUoj0A&7Dn
zovC-7*Y(g31Idu{aZHSZjSpJA!<z`izVNkeoPb1PXjn8Yv`GbqGPj2-7&2Z_cDR$^
z2JeZ%(L}YaN~R7k#hWXG2DC8-1-$Z%$WxKbx{h_h8({8@U-+wEZ?6J&%E=SwED|r|
z-N+X|_#W*dvVQgfy6p>%!00?#s?=Exo$(e}AWS_}-OrYQHR82)D*`STaz9!r*}QXn
zLunh@$B(%8tq(a|UGjQ&!O7`@FaO>>e(-@3<aJ%EZ3E~58I3!Ui*oN~<(nO-V*Pf9
zk%Dge6#+@6Zi;Qo(C-<B9udcC(eUv6g#GJl_BS{9<(kWv7c5WI6L7cPvRtlNExXC1
zDMsZ+TwGr9?8#Hwre$+`OU_F0-`?I*a%Q<&Dxo_DPESucJw0W;K4tvD_b4UN#D>G6
zM`WbryJ6ZC|G65l+|4$o@vpCcvs7mr;hz@#9RZEM?2MY$&iUCX@4WL4+wF$UX3KK9
zVCYAd%M~K31`Z*xJ~`pA-zfsvw%R?V6i(JB91c6gNaz!n)PBzw-E&_4;9qgRzvP`?
z`%O-se1k(QjIm|V6BBpz$=@8o8I1}YqXyIXz4<e4UdnF{lC$}!#1Ip3)E=e7z|GAy
zFTVVO+v_W0;}Mzr{BAcpwp&Huo<4oXgYn_SQ4=^4YRtiEN{Pck(dz!tb9-~ma?uil
zv!4l^pP$h-4Q=D;hYk69WV6}uz3+XGKl(R+#KwYdJ>)!n-_wr=N=~FPasPaQ^Hnhi
zBJR$3-Wkm|kUM=nGvDus`*-{^H30nC@81=Nyv-%0koUXk&>5SCwplWJoPuxL$xK#*
z29H_W>CXctWyk}h5$Dvy&IK*+stJ7|4?UsLL)jsdSD6$&j6Bu!S5W|GmLYXsc`Fs^
zURt@;vlY$p`kE3FnUn6SKn#{Z9tVxpWVjOTy`fme%BCTx51}H{S;xUXj~<T2%GG*s
zRTX_j87J7JWHoxGA|2KKLMBlJn1u)0jicHcLYXZ5s!#~AGqVPE_KZRdgwaz%#J6fp
z9h~ySD$-U>s+lZ@tTsBKrgz7!<gA@hEIIau3Tppf^(d@dO-*VXl*C2*IbG*zqNfb1
z@*Rt?A2Yjb%+90dR9!OECl|V)HY!aUk)#zi@10iIMm0nB=9!pMhM);b(=<wR3K8$s
zs$>|nC$3t&2#2Kn!+soC#+LgJpOBVIGLAUx>9;$*zow;KtaSf9^xW)vVubsC#Ujrc
zi--ogs>m+H5II?`*c^H`hmpM7Qh?40V^nJx?;OMa0Aa23))Jno>aLbGAx5GPHb-<F
zRFp6k)l+9Howr@56C<b~FE~{Q>ptzEKp~spt3^UWy38^qa`5zNFtpuS%ei8~HFIBI
zR0I;eb!72DErjaot3s5D2s=0Rk9I*I+nG_O)vEI?D0483|DFRt<vl3lXvibQsp@Ng
zdC6zr{~jkFs++vzjEuvCrqs>6^(rr)GWer<<=p^KCIoPTQgfjUqjIXFH+gh`5b<j`
zefKGD`%T!r;=3Q+@ci9TH(_U72FGM->Akh}aoU^Mhd|Ebv{@DtWVB9wmQqTJNTm#}
zDyz`HH<A5eha(U|!`ZsyNjc@k^&Sa{e%}+K8m{J?*luoVnuf(<LDQ*&dqo>xy?V`)
z$B%jN;31cnuaTTcX=J<EkV|H@Uh9JlfngX~tyaAE?z?>U*{6Web=teMmw}q}PeVG5
z=lndU2?q1QUH?t$Yv#%L@8S{>mdgcCo<3o_-LT&us14+AZf-bP-($DkQG>9yQIAB>
z=ZU~Lj;z;fy0&9X{cK(elX&3u;)eeh|BCIGuSwUpoPOgq_dog#miHfN)A5`T6m2MX
zZq@`3t|HVm5IhS0+%a-+JXZ5~-%%^p@}(VE^*Ko{^oO2TFP`)2<qO>~W24zHx>;Y}
zY`D3;0X$b%mppm$ltrgOvRL3W41=kJNA`yU*Vi|E@x>SPeb4LXueiLt;PT=%pMLTQ
zr4-(M?-_sfulyC(s|6?P1<oagF>!nQnv2WlY_=EVoN1cKVi9SYhSfT<+Y6;chQpx2
zpA@^nI+1M5z9JGn4g%{|@Ma))%<r3D`-UjtD-glAxzyH11KUy(sfXMih@m623u4ov
zHYPASEqT^;Rre}aef>;GQP6;=qK6(ej!t8bJ6z*(r^i=PQ(%W0>np05mA@9E?!VN;
zMm@>A2+8<wuG%LVvUjxfCsm@8xeXm?4jyA;LEurTH3Y==r5YDif7IYr6|Dk~5osD+
zQYCS;5}J5#f|EiPM=Fj&RPG>(J~@>4=UlZW5h}*$eTQozE-Vn&0@2E3%Cjp#w6VzM
zv8VT{_u%8KrnRD*T9Y}aHZH;8P(UC!aKRIU0v3uf2uaeEojAY;@S!nx3TF-sCM5By
z@D)^K5kyyfF51qt^2Q=+pxg%yB5T{gu?SfQh3cxPf*?0}*LeqB@GRPBXuVQ~mThD^
zlp_SV%=SMRmB~aaMLa`Fde2T1#@+6~As6<e2BCG|>Y~u}R}9<Nq^w?%Au3w!QM)5?
z%5NKo1FkrBL!ya2%hals)P$DW`U2Xn<<b2I><$B)-G=A}d=puno@t4_6b}16a;m(<
zk`;+mQ!jnSA%@8k%Q;Wun0oEjy%2%edZt!~_N;2JYBg#v+HSBU?Gz@u-h7M%I+~Tx
z#ReIB(m1Gh`w=AXrgK3rMF6$cpR|5TP}D}v=BA&K^bi9+X@E@7JwQa}*1Dz;S5Mlv
zyWTzkbT(-%3j}Z_(-q<H>XOT!`~(p4FfeX!DBXe*)il)ELJfXK{pz0W{ay4=oTC^)
zGdNZENJ)&TM|>g1mZ+Ypc<*U}#WK)7Uh(^Xv*r0&gP@sdN{J8~nxLkVdMPRbjwZTk
zQrtlh1X51K5LJYdvju)C3m4I*<xSI)hs2`QO?=FW`==`e$Cp<dLKDcT<->1&#7{o{
zF?k#q$AO#^tJMiHwrV`-A&-SGUq0u_!-qV2_>fnxUjr(+-)$6uS*=e<<7k~P;OVnx
z^h3{!7cYr1&~~kHAMPrHEs0`M88IVKH4v<4wH)!W{;acZ3h9p{&H7$naVg2%v#}sI
zjU#Q_ad~;o+3A@k52BE{BDZ>AfwnNKSzp_BbX`Xg4c1di6JfXp65o?cUilrrclezD
z?7J_yef0(Rzxf;7d*^)?%N4!%jBbH*kx;ap+>^;ZBOZbTLM>4mS=hjXqx-k<wbIng
z&74B(DKOBInJEa-ZUr~DSA6m1=Um@j6B`ds7*bCzg}%>p-HLnnTT)J3UcKgaa|uq%
z9{n(~+wa*Q4vJt2^oO3y%NuTPH~jRI&l&rXz8`pbagFysCV0<Z9e<O*^jm*PMK>`{
zpEY>LYT5DZ*;7-*Hq$!EIngazhM^~QsuI@uR3_`E`h+d%wfR`H^)jhZry#J*!C(zy
zbnuwQ@VNcku^r6MR1ki}OGHTHNXg0e3&D?`;NsEq)!*s8LxP*D+*0q$n$asPG^2EA
zDNivSJrah)2F8Pu_>Bl2qEkLw0cUWRj;SJI-cUmCy!IN-R+h4|XOUtOCpY>3;ffgv
zV-i#iE;v>Rd#BuJ3qp%%d3oa2IqUx$0x_t=VF>~fa3~t3vtCR81>N(rhy_-~i1h+F
zQDlTX;?RyC!Fl6f8{*Y0kfuXIqum+|Ro1p^VqS5|shyGr`(O2vs<w*GP(dxLDXk|Q
znTUtrCQm{rhK>!CD&Pr`HZ*2BY;j>t<ZAUujN>KD!P1z(#>JEIXqh^tG+VI<q@pFm
z+5SlHCZ{s(+rfLHn%bGCp+2|jCwdqYQX5x`p|h4GJFnHPaTsw?=~=l5!&ul4nL{cJ
zDKn0l<#I)6T1xI2(@0};b<Bdqz}eY7@-Py+1*JdWx{O%sv>}b`21mQ^S=sp<n>e+U
zq<{;sS}u6<_%Z(CQkAx`V|jAQetV<6-OYzIddOa9bM%wTrCMUOapG}0XRQ#A_6K>R
zBH)5~N7kP~pP?&>GMLe{3y2qJT6_pJp<xk6+}P0fdzDKS^@ptfOme*KIH#)Vwz0%3
z&$c|S0uoF}H)zn9IZ9?H)c7bV7XAmo0YoAE?1W-Gi;#p?pbUxO<qHj*4ts{1Yk1UB
z3<{5;lJQVCfEsX2T!o`e+&OYkZdUNlpxQzrSt-B+nc`F&vd9NUuK0uH5Bb>N>JgKn
z6lZ=uI@xv8w)RR+mA<YBv2Bdcff2Yx5yv7Fr{&^d7^X(|H4BlVkoID+VjKo7y_JkF
zg~z8$@;Gw2yJl!Trw`6~=e=+6*-t+P$@m!A?{;|C(=L`2)FhQGJb&>ePaZzx*|TT7
zdi9Fken-lg!){MWnbpY}=N<iCg$qxfzQds(xVpSBpST8x^2NjuMpxB$%#%#TBgJ|g
zw?ykHS&#2w7&sh;JAvde3idNs4Ipjey6GFgcuw_wuR;ayXkx==v*q;kjKyNfZnGzb
zK-2gsz-Zb=iQHyjTZ#&Y+OB07dtzu*b){)E;?yo+E3hb;FSy|U8vc8J{pP3qjX(UH
ziyyzsgAcyV$u~b_v09NvAr`d{aIMq!SspclEkHo|t;d~3EIHCZ#qd7S%(LWzw_u|7
zr*tBy^2j(l2;<muaruhZSFcEnG-1Jh=sEOT4#PkUE1o@lhjAQueenf9`0;lUH}de|
zGxqzQtE(%<(QILy2j{0YQimZS4Af|KOqy9nUpVy29rc(Wo7b{d;~P$wr}XFQPsdTB
zGz<f3X&Pu^!G%WE@XpoSwoPn4Fkwy%@{t`$4g6GLhm{M>`eHf|^$mVe|KG0z1$0hZ
zYSr+sZdE>b^VF*c>~WGKZ~VU^SyQRXp+TTC!I`Z=kAqT>YyqbN9SDKqK_n2J_C(it
zJnxSoY7=@DTj+RFgJP#Gvw|Tt^>;Z#Jmy6?D|9P*37U|L3t&}}JIBt^*azpVHJiuz
zKnbQI6;snY2j0xe!CKU57Yj94Pv#q#6}3XeBf*=XCpO?CI4zSY^+*$YCwl*LTfn=A
z{Uj=ylG*v*N}f5n-sZLUYA~JCUBAg>J)JA~p)LPv2M<4uNh$)?F3}k(C3r$?5Xm43
z7hoT5coCn|Fej!{&-wb<sS0}gjJ|f4DP|VUl&Bq2W=w@4X(_pBT97JmfpN@uC+v0u
z>qU!?Q9X5&W8aS)#>`l>PqCDY500TE;$n%TV<=kA&R&z3&N*^Q#I~hfEL9ve_LCKg
zbI^~O?bM&#<FbD47$a@xX$@_DaQ`08N3L!*_{EZ880gxLGA6F_1rHBT2n)3)(k>o`
zpkbA*H`luy(UdIUnx655)c&kcOq}`(I*c0|qmH=<Z5ojf)z{d8q)}-LWQByQ1Hk-l
zG%=dPgZ{2@RNi$o5JnI)YS+EMs=-PNF)>xi8CTo{VpL@GZLYTt0OvMBZra!pEbDVH
zD5QSaZn=H=n&om$IBUt08O6~$IWmBnugaa6Jg6|(-C+kSOf8DcwAK}9%Se|7e!Te+
z|M2>E>BkKb4bq%LeXW=ej*2uo=ZK+U9C|G&kKlc4NT8{&EW0c@Pv*MgFwis&v55@*
zFm0o;iOLH|i6&^DRY@aHADlD1y5iOA3u4>w;PGQ#cMG<=EqP1?97BI#Oo`>`T7@3t
zNXmuJKL3IT4+ow+d&=d-rJ717?Tvdl^sG<TES8SLeowbp@b0_svfu3(2UWs$T}RWJ
zG4eq_YxO;vd>E;(YiZjS?}HLlL%@6WfWxT~^E~%{_K=;<j-z02Dv^R&5LobecSa$o
zDz88EELSUX5xQ={+37jE-9fDh5ZbmyL`$5(dxgtQlv96S+jY3@h7h8flA?<Wl+-@>
zz<|&=;kn!K86WdS-t%`}ZphcSy!`BAp8UpN<NO=nWE%=&(@=b%8HFZAMS7fil2xVQ
z1o%xzm0?78kgOS<5-c+<Dk`YRmUE8HcFU_*FWGLlgy<=v{d-?szU20H%LniOI$gKK
zd(Z3FuleDRzRMr};U9AE{2}}OK;QSg^Ugbb``h2<?Cgx?a!J>Ac<;#ht7Gvg!_2ek
zaz*Vldv6rwE^1(0vrtvNHsnUpaZR-J{@(QLI}Xh5n)KH1RW$Ht%ei;P?kfVkJOB1I
z1b`s~*6S60uRYFjx<55dBsR*yt*ZN(%A(w$iu5@{vz)W;0;uyr5K%*4^jT=uZ^%O~
zTJp2Mi?~TxQV~*bL4BEW4^5_HbSea~h*j0KnphNfL^7`jq|68z_8GKHHLoQx#KEJe
zU=<iR=O)u=M<G@K)N;Kn<~L_*$1LbQ*C5kOqj4I%1_YNAP$<4od;a?1@jj9*!J5!t
zXBmPNT=75WkvmH$j!<T0@9X`d)S$X)RnZY>SvO}5<nCN`E!Xm(S(H^ywK!)^9&SqV
zV&myT%jkN>+~Z5npTr*^PCG#Ps6vtuJZ<ze$uU;r+<GSloqIDOcZMPjqjrZn^aGon
zQhvPiR#m9wXG3v|5WTh|-R}pI6ow&l7!%oSg-R((uW=q9qs@Is@oGaOUb*~fOj>FU
zNfB95!G=EnIUx$pn~}3Awkr~ug%C}Ni{t*ib6kwvY!7&#Ya5Dpe97m0RDO+&$45hE
zMNQbfE%hSu)xCc#Nzw#zR<}=Z@4Ot;{z$4I)Xj*l8IDIEDLx`0kVQz&At~W7^lu6@
z^}W@eELaP@VCRqR3l_Q7)dOZ5W9vD?gksil6of1~H>c$DSG?Zd21euv?$iKDVR^;m
zS}QjkZnwPp$)`MBpR?+;tyJ-zL}C!jh>za4Zln_mC}>~Bshn`^P33UK;j^&lw|tgA
z;~!lAW0LGmd8ZZ%Aw+_k%eqd~GDV6^t%Ss@sH0>D*7(|4nWY%&SP!hG(SQ+APo|nl
zl%jS7P18_H=JN8I^RpA8_oR|}e15`yykd8G$%A|MSS?oE+-$U`CRPy3Bm3K1PS4L7
z>;Nv}cztn6KMcJ4-g|V51utGa=TNkz(e>3eLqBkK4ipVS-+lHDAAkH43?^mKEm*Er
z?6x~@Z*NpV=ajpl5PD<1C(UQe%`N=KQOoCdr|aDSVlFA`zi63dChMZpqfVDQoAq%p
zA}Aqw97kekdHUorrOarp_i8K(ZsIb55}c=b<Fz!U@9ao9-T7=;l;g4!p=$z-Wn+Kr
zU+`tz@IOyme*KGwe0F_J8gFPm`iL={QnI5fEnU`L(_?g$qa<c_qSrZA|6T)<Ipm#x
zMuelm&~YEr$i>BLE-x-ACDDcjhr@xJn;V`#f6jhC@QwEh{o%mX^%Xz*(NFl^_rA~V
z?G_|hpV7cy|LcE^Z+`QebX`Z+wZta=sio0M1E>dl70^gAHEzo4kETO=Cz!3*WY_Qp
zLiZ+DuL?Yl1GiaVV`!E4vyvI6|NhF)ef0is>y%$V!*aPWqjf|4y~DX+)*l*#N$tpC
zhlNsA9L}k!xR-3mUo~_Fj0>$9zjBb07hIO<V0T9-@&ujMa+P3cm7xgEX)GkQMOiTW
z2+-cIPRm&q4AsP11qvC-+5GqlO>79JrbKG_X-*2XgM{u6#jDwOa9XYw$%;IWgZ54}
zQHKvv%ao{fh6-5%<ZL!Uiu`FG_G}19p1NGvQ`TUQf|E?jo?;$>wNup3LxWS!lK`<U
zOjolrEA!&+qJ%jY+gFrh>|HZizNjg<?lt}humH?Iuv#8DybIH34isYRxp5n=`~~ig
z<&mP1%O)@+p-+i~o5VP3Su*pCqX~6#R^(3+J`<GKZEBg^=9;O-gLe#;6z|8Rv?Gj}
zCQ4`MqGugFtIk{ZqLx&Iu@v@0VlyW5rE@C&I&3vyP9qMnIV3EaK-)wO;$8KaR0Ma-
z8HNEth#p)J?wze^V&L+6V}cg7@d|BJpF9yrc5{`>YHQ-SY6vwJb9!iq0CPK9?Xi7C
z6|tbTCOiU@!oQfoizdAR@1yc?OJkG*0ns&dyt`@_o;b~ke)J^okWhPV3#sHusSPG1
zb2824n($P$KhBv`$-n0SPy<;vE4b{^&*d$XgK>Z0)sH@*J-LVPmiV(%nv*4XEvW`~
z_YKsHHqVXJ9e64DYP;fe191ga%D7C(BR}1J!av&oQx0iQRDrezb`|xjAJ<SAmE~K&
zVO>2+)`uv`fol}rw+U1%x-W)`#n><my>X?2_M6n5LN|#Valvyq99XX#oD^1V<mvr2
zFE?8ldlp?s2n{JGLs`%bA!lx`u34?u_|OoV7Uy7pIPk+C{*Wh6p78Fw?{Rr?!R6Hz
zrDV379pjK#pPaB-ujtyAcb+}t>iSA2jrZhSh)u(Cv9j`naej<k+jboQeSaWl%~qtk
zy-QteOqMY9`Di&z1vFK3p<iVpf9t|iLh*4n&iqyYv~A1T**WKD=d9N!l%kEpo2F&k
z-ztYBs!BG-1`!n@jcKH0EyFv@`s=1Jj@HMN8DaA=>LHWM<n83ev2}a?QTP*nOYZTv
zhL33er10)(!zz40Z#<77crrpV1)^t00NoMVr#EK@?GcLLj<UqMVd+@tm~XdRUcP+6
zcDGUPkYsLduDHIr;^z9Awp-CO9fw2DtCz31y}e}|ll2tVa(Bv!+uK{~-Z4&PusLA+
z>NTxCTh)yB2#K?esG9BCxj6@ON5PkJUDSZLmMW$|@+kQG%7CyGJ3D6!jl28HS8slA
zib8UJdlj(GlLoBxrwND!GCpXFEIBgim2728Ee{8$G!VpVD>y8e@`9rwWfk;b+)sz6
zc%=fU{fY4&-E`jJa+&yj8V|W?KKU@0i18kLRGN)5+byS`Mfb70E%r3o)-oMrauA!B
z7K9oqoKs@V37jK%EismqDX!2DnKbNG4VtZ_Tleh(zA^8=YCBVRXAxZ1E;xk(C8M@_
ziVTtyC5@)gP~l;cghYYtp=m4`QSd>#j8xi(CH6(nmXdl%374~$BSp+`TWL2%GLj3U
zQ;LK$uDbWC_KrSGO$lN&YZK?D;6b4skB{1(OiCt8CXJdD`l1NbkHe?zORt=E5yr6~
z<_1wjmAf%QAuy(73569NMChVZN}uh^Ics@-y=)k=qaV#OrD$1k)kShfy&bKb4KpIB
zk6GtN^nn&(**KPs%4|~ZNn<97*Rvu<ifEr)ip|L^%Egv~bA>jlz_D9&#6~%jUUVMk
zq6*!DcHwG6wF-)?mK{wSIqbE$fls>LbM+3gq{Vv`fCdw}9Bl?uP*{`iN;gp8Ame?Q
z>}cw1>;6+IE)^x$G1mE{J>xYA#5XATr--}r1t&*1q*;DB95FcBwlxl4u?Qtk_GTLQ
z=>wk8cfH$SL6-NP28~z3SMT7z^!4^7U=5L(EK{u1FsUXj0e893zq;c2_kTpYT66EC
zh+j0gR=ZkMSN4iHlwx9ypdV2s+DAx|5m*C`Pzp%T$HPm0H~$`k>}j1DapE<z6@^VJ
z>A7;nYMHr|Ldv7@V)VZy6*5UDEi)P^wow;R1?tcr^uI9^=Ac1jDUPOXSgty9QqQ4&
z7)bqrvy-)QUdEoTiL^~*xmqJ7)9*GDk=Ki`yWJom;$oz+FXtRT{rF?<-Mi0|$4^)+
z7QB4<io^cEVZW#Ed-l66=l9N8uhw*ng^`hk-Tq+03L~{Pk@I`^SuK|g!^mM6xW2x|
z`(_@yT&&gDNNwMAoMa13GW@1#=7l&N#xwF$?F-aEcyGk&`MsJLI6FP%(Zfe9mn#rw
zifm`@_t15%zE=(61xgnB!vIdjAR;oU#Skc|OejV!N$n4OpsBnm!!|X@$%rKWrGL(M
zn`{2l?SbXLX?gnRpnLiex1D2HDvncfp%Jf&%%i}{S7^VYX*TafGRFZ-6&y{mp>w2h
z<mTp@tLsa083|2fyWMiTxn;N80*(g{9&vtl4md8at~l&_JH!kmGEDW$FTUhE-}w&x
zq389*Yi@6Er)THhKw;}u`}e9a>h80v!S&Q;Y7&9uDF`g3Abv)XYXI@4qW2g9yknAG
zlY*mQ?#-)?;W0XRj2wP_T|hL-t#k?X-ZE<vLlwN!E&vpp-&Vp2-Vwbbf=&u$1Sw{5
zB3kA3J}P=x@&xdKn*|Efz1R7vb6AsW53WoolnWYsIOmAQb0P#YB3IA2VyKqdR~?}t
z_AEjjcbGa;l%mo_4d%r}aN0Vp_)J2}jYVvKumrwj6*YNh{6J?YzVqa%y_+-S#Y$o}
zf$RB;DLHe_xYPrV;sfFm5GfM0ax!(hfMR3s(B=c$i?`V49@Thz9Q8aY){;)<Kvc?X
zFf7GBpP_z@S&gVAnNuNPJp+x)R-cu(*Az71a?Vqds__q8^Aq<|oU^KcU}IZRRv`w@
z!uBkOKbjwk{59r&kgOD&i;7e5P6J$oCVE1#XHbd;ji{whit1<CHNudUKUjmoCJ4ti
zbK<-S@M<t+2FX?}Na$UM00961Nkl<ZicPEKU!RFCqFq%StL1{OX*4Kwb+0HYauF?)
z`yzx;$RX3lpn-{KtiN7#EV`D`_uPyxc*whonjLk!I;L_RCz;bTovwM$o%X;tR(s7W
zkN2Uv3)FK=#3`rRN1PV|9xv)Q>70P=n^5hObo@0zo+%iTbu0!qSesiRns~?^pP4lZ
ztZVbDHtk>GdV2)0Dyxub<C0q2b8Zgc#1RgO+s|I`V*L}E<$|*|Xf{~kS>{c>SozQU
zFmZvZaIUIET}}kav;zNT{2u?7AJe#j#$@;2p!Z4M4VtCc`&SO2fi_B_txexicsP&6
zJgEGfA!A2E&c+35IUEjZ<I<=qQq7$c#K_6XDPv0P_XAfqJH}ji)Ha+hJ9fFpNuHX-
zKYaL*-J)f8bE}>~DbobcP%>%A<TTKO@wfsPuV1mbz2(`nr+o0i`+WZS7i>0LQp(zu
zVZY<_^n{bMQ#w;IouA+1_V$+Rt1A|Z1@C|GA?N2BNGNKQxw*aJ#q$?jUS3Q{-7qBk
z`y4nZ&|4kv!!*{;)O)%J%MlW08>yiuH6R#<qXkohF(t-vL}YI$>uBZr!fLssX~KkF
zdOPI2^XzteMF1=)7>1GcYR!Isu%~Qg%rxR%ft1wB!X1?gjuMSR;UEM5FnpJPdG(s#
z{*w#-#`Zs8^_#!Rg(M0s>(FTr&tvw9$mHy9OMYH5&a5<!ww_wDuBh8?zvbnt7i_m1
zJwQrkyS-(*-I!W@#XIl3!~OgB`Qq~nuCHz;YbEDZicyS+_4?w1Klp<`;Ny=!=JxiM
z7cX9DY26g;+JxlphVoUyiYf{y#=E3GYcuK)(X?J_kY@qk<liSpb~n>7-7VQp|9zB=
zzxn#(*VX{Ae(yMlbk0qi`P*DlAg2MbGEF_~)dr{45-3Jfabi}c^>Z?n;6#+imWw;u
zdrXWbD5NgNPcl?YegE9Vldk+9S4`^bMtyhZ@Y0x&$Wsc02+k>YtKh7J7tE4I-|HCp
zt7vcShV$lpD7o7DXxqCqjwWcS2KgY83_YvpTbVjDOnP7+EKt(*FFB*Qr&m9^QaOE&
zHb57Gb_H<)GJ*s|^uAn(I2ZA;v0kwr@3c!s<rxbYHOV@Yh#*Cq2xM~x5Rs{?#x&n?
z7C`7jtY=fG?bqtL;<N;htL1v-jr$|?sG3CA&Ml6S@44?#T&9Zw=Y%}Y-iPCu5m5fP
zs9&JRo6&cDmCkqV`(6lMllfd=x9=G;q#|^!rwfkO3#ZGLMeE3^usdY3hjy)0nu|?O
zD!S)2RcKiX<Cy53!w0l~ZY~yJWA~~UZWA1?ctWGzeH2Hx=qB3LX#00IqhseAo72j>
zD@jq;Tnc_nc+l=9tu{x9%fJu2KW3R1obJyiG`#j#e>17B!F-MM>hqe@sk*@D*US&!
zO%Y#}NY%<CfF}|yXq|fMWewDY@s6xP>YnH-)@@D?qGOw_vKE~hMu#v5vYI^9icQ@s
zW%`J36i)xl*V`LG*T0AoI%WE<E{tqPU+|t~XvqD@)hC~`Iy<FXFNm#zm?bp2d6sIx
zYzE6_%~JVC72?-JTOjR8e&jphCs0Pb*H%oE<CWOf;+350tTc;d!S$o+BDK1vQyxSm
zmZotex@KPVAy|n*5v8_k+3vQAe6_7^jP~%7iA21Q^y7hH)J<sEA2?|OZ7JN`Zg_aQ
z=Ee1v+uOj&$ti8@*ljjsKQavcWGY()4C6QuV?(#-5P?rW`;60*Gaf#C$jSPY*B2KY
z`aLBVuCK4zANHJ_oN{t{N(g}m_aE@^;X{_I6|2=+$GE7!od~Q}C%pgulBZ9e^77?N
zUcY`x-w(v-IXyik#>jquVCV-%>%b9$pPrqfb{6(Hv&XRD&oLvtUAIv9d(5-CNrW_J
zQ%V(<OK-xMk)~#n2A-GV5YbIA<znUdCC)j9p=Yz%u;^NHDqv`x7)iNpy>hcOHREW?
zsQ;7mM<m_C$8y1^+keIvpWO1-;uY_I>#vd9Q|+NCN({cUHtOb&@j6{4ufK7=*2KnX
zAU&p$tE)?1zI?%$w8i1Y^(B{Am!y<v+m>z-xPSiv%jFulRAi#*h)qk`Yc~1t;X{_o
zj?1eHzWnkz&!4}Tf}`_$=RAA*l<juQ7tdcY4g)Y7Rjb-D9R+A&!J?IPMfp;}HH0{q
zCTcaH@|AL)%WyTIJHFoJ`c1)}$j|J_UllCY-&+H+8UW@aWb17&k<4Mc(SuLGH;r1Y
zSoap^?Vzs|38~@;bZ;>PfKsM_E`y08l<THtQrkam{&|m?!wOYSDbpsaL6R7+RnT(3
zv(j|I;j*FHzFL*22t{nJ#So7W4N;FjQFO25%m|a=v+loTdN!5Ams4hBAjfJvu0f0|
z*4oeqE1haURaLSTJ=Km-il}%ia%m#j;z=|?Xd<e7cFLO)r@gzgQwoP~I~q%FH9;3b
zbZWBfs&AwsFx~?}f41(YGfy;3^CtxDxvbHap}mbarSEa-rk8<!OlwFSkWrI(?bwu(
zat{|y7`rX6-3t~`1wqa^Lf0@3kjHElE`2Ud2$ZNL_NsQRbM4R%G+t<%NQ_ZiVu~WY
z&O?V&4qo(}ELs+gNBf_{vazaC>)7r`8hc%`t{}k#MY$9Xy>j%V2J+q+#UW0C@3QL{
zh7l2`0GI?T#cO$UOj!}bs1yTjfi}-M=RK(isUIz%)ifw31s>XNLDLrgP5cA?#}9s+
zd)sqvcfAQb{hduUAMG<ggYZp%v7}1$9xF&x*I<nartiDTyVIvE9$MFuW2B79Vx5c>
z?NT$pf2Eh%`Yn}%s`ot_s@{aiqO|x%qn9Q&y65CKt0qVH@|Fqi*I)xvKY_ub)M#3d
zJ@GU5r}dt8EbOl~y!!YPy0azCvSryV<`z#y50It_(;*Z>oHn3JP7Th{cnCSs%PmQI
zd>(O)GZM2{fF={-Ql!{s?IsDh*84M+E$f{lG*Ksz2_#C&jLsQ(e$M>F8EoXV^x2QY
z)OA5ObAza}1}cZWmO9&}VG%o|A6T^wPfpj|48m>Ou(`d(`M~{q4{-zzn++6U$U4z$
z#wjH;44G0gUAJJlTr!M3KmPc~+&{n1!-o&pY&YEA-jH&lKMaiHCA;08v-5KvJdCW?
zYn<~8{XnSMudCikdf=>9Cw$`@3(n8axwv>uDVerwHS-)s<xd!oVt?2(^!;@3YJc9D
zG+qM(XB$XW0ZtylOjC&>h5i1(VzFc#wSRJmp0?GEt)k7PiV|{Ww^QV^>lOeEhaLwU
z4!s_V#vK_`BIU%wwTPKeX>e*Kvn-Q3Q29{7#J&abBmeK+zv7?&$&dKYU%%kDzV+Lz
z9)HNbX&8=IrpS?b^bvwpm96%_`mcga0l{v!<@M`VYJ=lc46@m5*zLB25a_y&^LzJs
z_Usw=?%m_t-(FBk;pwvnJb&>n00Tk%zCZcNPv{Ss_uhM-v$HkNUwp~;zyAYnZ*KuT
zeBXQTJ^s>f{T44?yyW8QmT?>^RAn31T+-EmDlrTro6VNK?-kv5p0;h2UXYZqUk%cO
zC>>(E-Ll(lS#+J2rs}S7wE2FO_+tW$k5Iuk%fWB{{Y_=?F*^7*mnIB7DTyVuAmRwI
z(Xu3?1`Fth4vL~+eMQCQvlk80ydB;tk3?64*7>e$#iJBKf+8($Qv10IWa~X!z@fp6
z?qfx=9>OK#ir^Y8RjF_zgW0h}^HnXSer~-|9m7)UyH6ZwDOM3_^z)RVS8Jro@k7gw
z!Bwh<a^$opvY!x}stBGM+q*C+p&eR=cD{g9<K@~L9_Pqbj`blBn-144Xqtuc?et9%
zG3%P5AcTf37<X$%h-;OiBJ#&1`>DN`?N=H{>iH&y?iJ^7rC2#Um>5S@q}AwKiqboZ
zoi9yr{A>3;92t*V4Ednt)ijQ31f8_J9-CGZDRDRpAjJ?1v2)gW-M_}9bR}{jy4i%<
z0gF~U`ULNg(m_bLAhd?A4UNz?YIKy0Q|O(Rpeu4(_i3qUR4E9F5FCppBB{5bt4OZ~
z{NtD``PH&~QALt@K^oUi2X1PO24X&liRjgkyC24h`~8pO@A5l~-{GEn%*}Q`M<Fs(
zN@@V->e6!aK0NV<Cr&i#<3}yin$fp6i*`feDw-?Bl0Ku<4%rbx1Fk5i?}+QI!MdM4
zG>_1{3UY}Q(*Vp#ZLD6LZmv2W-{YIX;IDFhEk1xy@ff`-BC2IIgJcCIhdR$#gsT@X
zi61Xno}bX3p5RP)AjKJ<rc8u&z+ChVE)EU}LP&*$7k=OUh;iK0IlZMa%~bta?{$L{
zvF7!adxCBN-c+g@vf5y$Odbb|V^!f)CeN?RSF{#MMTHaF&4y*SP|vVoG<C7|ugw|t
z7xaNPL>g-pf7tI?ELOO#VGzgR)q(5FOGF%PXbG_)<Pk3;zLY7b@ZPant+_QbXn#<9
zl$0{BFJ80RY*?*VJb3t!?PkMnza!_uez#}n2X1d}dGzEFj~+dy?K-nMP_WzkV5&Qz
z`d;N+ST42aaL$RL9}sb@Rx8%)6BdgF&cQg0><|0N$8W#ilhTMtveK{~I4%nW-6*pm
zO*P|_B1kUmc028J?6jjv)5JTASdd9<GLDJOW{WB|mlP@V7L00!sfMT}XNFz_!>((I
zK5Isr%`v_dD;Jrl1GO8Fpc$W{%WySZ@_+pFAMiWdPxwzi{7-oJ?cbs_Yl@G$<qA;J
zbU+t_gKAj}WZLMa&9r248M(Z^<mIau?E4)}6S=v$=KA`|vdF;c$tmCZ*01rw`yX=e
z{0Wa9L{3iD+-^SP^6CX2|MX*?fBA~la>eP{31{bL91aJ5{No>y#*uEZ;F}+P$Oj+1
z4^9uDy1~?~W@=JyR!e<<;C8d&`udXHeus!mo3ZzvQM5#VTsrRi1Fv4aV!7yOV$12t
ziHWXe1f{m;vMuqbY{Q++?H2`rUxg~ZSss3?OQ4L|j8{FiyGjUwoCYd+Qv~htDVc<*
z0=*`f2)?LcY&OH=ToO37nbE*S`CpoVSx{Fp$qAwH1WQ1@`N?_FfW=qNkoQwtOu>`#
z2sXApc8COq!qjT28agY7Em+TRnJlivf}M$5l})W|Dq)IfvYd7I7pc6*fDc;r3DFyM
zU7ZE|^n7aB4h^=PB}YeshMALVB||(}w8G=O+L5Ht5DjJ1185ej)snhHRkyilujL}O
z%ZO4(N-joY2@~B1DU2exVCRs*v(90G&ygat&Pl0AjnY|MafHxm1q6uRlT+e3uXyfX
z;-f0aOUWwKNNW6^Y@c^Uk*vyBb7mt{7e?@NLhY?<31UH0j6flJ4fMN41#@H4Bq7G2
zB}b1bP_rf)7QBnAC~<(|5Ot_#YO6MCBH`QC6e;T08j7$#3^<2z;Z4DBNw0EM9YZe4
z%>>khTdd?>ruF5W)47i$HoBJB_xRu$x`E&A|66{0`S*CZc*=IyGa?`XA=$Y+75Mb?
zxnqis{!wf`)_`q-t|rvl=t16uD>!h0L{XupGmt4bC?27~G0r7u=A;8_gII+TJ~*;d
z5C>Nk|3{QH6BmZyRq0-Af3>e7bGMsc^?G{|2!-U7<eCv9ms&>*982RxX?A4@6Uq^J
z_1PEPyLZO&-WiMa(gH7N+7+RAyi@o2loKTn#F8mKb69A@@NU0nn_lv%`w~|MN*5?%
zg@jsn&)Lc!y4mC*sf;{E#NhmzS?WTx(k_BT-IP4w3zT6bv`({q@9%H}q8UAo!>HtI
zz=x>5O9-)XoS(O5#9YW_pbHB^hz$EZ?E;=G8aV0r;(E*W@&eB85n@9OYg}&GZf+=L
zoF-G>A1o_Hxh$oqN1kN*Vb5VW5TlAq&hI~9zu(jE4=TRcZTRdb8!lcv=ltOVPS5Yr
zb_?}t!V{d5orj$0`yIn@&`n4zO!kowBC+YLZB$jnwXD`_M1-eLpRwI;xw^il?+==N
zN-;C(gYt;XjIL=LC8(!FBCE12+b$?-T$=N!;uwYNro&?#2Ci?eto*y8TXgh8&wl7h
zN5KVB7={BwzvpDVq;-)~SBNE3b3T`uO8bN`1td-!UTS)vnS7R!-@E+-{`kA!;eT8H
zbAJ7s{}IFD9%Ecl$TaMUxmN_pEyzw);JT3(#2M-$0$B~r_uCz>E-rX+c|q~Qw%>4h
zbHQ%EC5D!>^*Qf7{g7|H|Lfd4dBA$HW=Mqx5AL%%S#olAPSc&RTz$%JyJgX?ShWWp
zKX}Ni7hjNb<niN2eE9ym+&@3%^~(iKFd+kGxq_ik1PaPo+Ydcgx0hVrUU3+E-K<BQ
zZ=(!4ug0Vtp;SnzkjBERm#<lNE6z^tv0AJ(*g#48fRf`)4VTe!XYDC|3=1DCe(z?$
zcZ0$=g(2yPjQuv3w*z2yVo`)sMLjiWQIx0(L!{=R6-_H6t{C6f430IUvtz14c@@2~
z@}%=leE><g;0TRcV$@!}cIQOY6(QANBo`2mxQvs84?=Q61G+1_<HJS+vWnDC+(4-x
zO^v3#!S7Ov@w)}@YnKyMis~URN2N0h0CJfsD94DcJvc?Wj4Nv0>(I_6CWf)+Do#s?
zDzNZWrmYk)6eUgfRr?m_qQ=YQOi5Z|w38kmqn-TT1S<M|lUln5eufklyH}~7%fwFg
zwJIPp8y>O-zCnd5V;T9({}hRdMPqIWMG@jMMJxI|=j6qL39C5ibJZX@<xKQ}aZHr#
zSS~u+)pGj$HGog~uG-f*M~G&o?3H`A=o)Z=ajZ&NORkO*s$xoNMTIrsZ<;_8gH>XT
zf3I$i+J>+w0;W@2jDIh7HtH=0>p?!*3t2)_a#m`>2+<ldl$05y2NC{%?w{~G-QVT@
z>M6Tn)CG#a6VQPM%Q}Q_R2wRyTa!;4Z;B>6e%kNrdhwI^t$Md2dJlyv1~b&S?j0dQ
zw*L|);3>L57h6i21O19RhbpXz(VoYQ4%T&hw0eH!>uW&(>)=hERd(|g(%08i_*m6i
zc^DY3Z+ZU7XRP0OjJtP2JiCYUjT%gP%K}BGzblw6g4ceJ2&W<N-}q0FlxV!F-ZW9G
zpTkJxFnRNc6kG`^!;fBlvVu(fK0yy>ocib`^Em<sNysh{nmAWt?4iV<>fAJrl%|==
z*148OL^wTJA);L_(l{~<18v(frX8u@(=8m&?wtU_>s^ma2Q_%LK<oa)C%k&~+zuW=
z$~X<McZ!l0#Qe(C-)Mgr*&PnV*sxx$I6XaK90n`-35VT*e*Xp6uP#}xR;<rXS)Q!5
z=Voy*7W%`UoJL{@lL9b!SS%Wq>Nn9$Jq?i&9p^MEm#Ax7mWu^l*U`4kkpg&Uf#AsD
z&~trt#ddpxFN#VnmJ2}L?7f>?SUHEGiGiFG*H_nUH(L<La<L+fSp_RuC!7XxTE^@5
zw*)dLtCq%PTH|jOGm)j1Z_r0x;rgG!+6@UPW#D$q{2#vm_xUfH6@U4o|Bw+)E{ahi
ziU@_I2iFwbnC;B;cDlLlwmV+FddZkaK&<q;!+Xcc>V)^-d!Jwbwcp^~ciyG#7Nj&<
zYba-Zk4H{V&p0_*v$?%t=zDw!ESC$`>otA9<K*Oo^?GfJMmyBsT+>G-hMY6~Ft9ru
zI2?MLf2xS=4?X*$h;J?593ymnKX7$*#b&eR;R7wz<L~lyDk4}9c?WmwdZy3xMxe*h
zzu!^*y&)WV`%7%jc}+ZyL?_k1u(}|`SSbn>h0-n%#S+lzu0?Ebb4oQa3njIXsFozg
z|3gcN+2XX?7OisLq?jSMjh$L-2wuE-jRm8;sA07Y`O&y4(g^w)lmi#$f3(l3v<+SV
z-mF5Lmq{h72PPIw8M^KCKvS_t0hdjwTlvDwYs)!B=@s+S4m&okHQ*AoYe%MN1)-Wr
z>j7l6g*>UEDV5hMq%<<v{8cWiOyU(+6G|`?Ezcboiqe~OUodnUM=(l}dbb{-*d{a~
zgE%$bPAP%F#%=j5ey-<L)8I`oND*?%3`VP|Cs<XD>fb9uW=I;dKV}oXdbkEf`fA|k
zEr}gQ;}1%KE)tvh=+8y{XT?JkoS|s7av|y)Y4fb2p*a^~RMasz$2g3(*VRYqCZ`O2
zyn-{TO_sSmUR6@h{iSWK`6^HoTuMkA39Yto)xBSEF_66Q`|{8ETix$)FFaw`AI$Az
zPBwfE&^d}nYT{Vq1N+P>9;Fukl-l5Rto2#gQ?8YVT<7`G;`SO>Oy8*U)8eAwJcLMT
zwIfr`a2T|^R;{+hVCc1*JEBdA0M3L}GKs@})$8jC0Bg2uhi!Eo$B9i{T^QLmGvz@<
zQW?2?@tW74eolA)jCHrbuL9W@vezltpa#z_c}j6KzF|=kp*R*^*rq+r1XbHZ#|E!1
z%1XL)dYHLVKX1SXp*rl$YP;IiK@eZPlJqX}I8Knh^VKBU1~aN>(U{T{ta#@jM$$Mi
zj3bL~!D6xCfVNc=)bp-C97u-^i`ANU9-OjXEx8>EyKyA8t+wXLLeq4UA}@<rreJ&)
zr=`xZWY(uA^n((QYd_twKhQNTtJRW|lN0)^N;pZGet%#%9Js!^q+Kmo-#cZoS`##J
z)CuWC4PC92u6-=?q-dJa7**dM4hQ<ar)e5it2J%YO=1w=M4H%ga-tlSO)X`6EuF<9
zQ3=x=vhr!xl0YeitBXr+u5T$xXxjygcEMpcSm5XFv1>2a!)}AiBVFTZU4mT5R9g>w
z!iN6VK%{<NgK}y;!8R$~xbUzUZutk_{YQKd-{JHde}SEEDL4iqkWDm{>^#+y$8n6}
z&_nq4_J-~D7K(O4xxKz6k0Z-&$-B?q;aeYllXu^F#%i@tW>YD2ZKP=;hkjrf0!}g|
zkCZ$zWJQ0L%LVVf^A5}9g2zuDbAEP4*S3~vyZI2cTRrZ=0jP6<cw$i7h+!C@Rf2g%
zz>b9@U>j=i?Du<aZm!wwcEl!{m)_l=^GyV=me%g>F~67(SpWQagy~=Pa?aDXtulI@
z*De4uBMLPb4zW>ARgImBXiK+}G<b6!pOacdWGe+34L~_<*7H^%n|ZVz0KshKoQga`
z#QA1Y-HVe+bmN>SI^|%woXL2KZ@@R!W=ao|B1{L}oai`9;B;KIBSDSom=e7D`iXz&
z&3n+#R$ba?@z_d5=Tt>Gpo;x$#z_sb(Dq&&|DGCbYtT?|6kj=nSdU~wlFdg@+#NU#
zwb>bwaU_o;sT6J1W@uK7SO6*$@>&DH`aZQiW4(%aj>%6_O{gsa31I6%Pnafnej0v&
zkJ`t)mT$G)oB{$l&r}@!+y&>V7q0F_y7v|}fVQ@zZB!23Vd!aszhe&Fcooi#eNPh{
zU8@#a)y4%>wcPlqVigPUiXoHUd==G|Lou<|Ov1&`Jw<5^fRSR!poNicPS9!)qru-S
z7OR1F6^-RI(6sHe|AyG&VoSSTPbS>KtC@9-kw4;J^LNhw9`|?mxV(C)LYip|kM<EN
zk~;@}B66p9uQPPBav!OpE+sH%QfC~%ijdZLK!GNY56v|F6i=F!y%LQ7-UhNbQpPx?
z8UT)?I#g(R_h<tP-W2mxeERrbf2Hf~RltsE5{Xm`1IsY2KQP##f(Z<gmYaNW<U``s
z$DeU>|CD>Dr^IEa{J)^(i&7lEDCsgewGQZtr$P9xe8y|Jp$i@_I$%Pv47?(9fmCV~
zZyTM6+7xK&I|Ylyhv_h^r0sfdHOMbH!#Ln*z)Yl_l|@|jNwQKzRk{gkj5(%}ei&%m
zPSvla&<_J$*J<mx;XuFJ(ydlJST!tK;q~o-%j*mJ;^>xZPS>Y2v0=B{DuiAX!mp*1
z3bXI_J)khV_H+eA7={CB=+y}N<b>sNp(2;08>16p*zOp%TYQLg%O%}%Mc1_oWjh^H
z91MN0Js_I{(NuWexyj#8Rl`BKb}^XeqML&R5z?3#$AK}8tX3=9=%<bywUh?O>PNlD
zF^#-_^_uIeYjP1*s}oL7&OoXqg#%I3-IRI`yA5gRSu~DS7wI>*6esMwBdk|y)*GZI
zc)E5g`r+>W_wj3liip6Eu7AKk`=kE{f3H~)AHGi?BgJV(?j+kYFr@StRg}uB8QE?(
zjKe@0NA|ld!=a~*9giP9<=4LTZQlRJH#j{#F*G6(UBm|%$HLg}*>0}6ynMy$*DrbX
z>N(?nBt}ouME>Gm{7v3_?>*M*HTUn|S3y$Se048OL$?r_Xao-)JYuyvq3?TMzkW59
z@pH+HX;6dMR6ASvsbrW^nruVrK2gO4b)CM6+SSDBILNDL$MM(C3j&Y-eCx|Yj9%v!
z;H~Gb$SiaS=BF8(s8ol#&q#!l)pn>P@Y;V_yz;d?nIhK1+5$B1oVL5gSz=`(b?>x4
zsjEUNdv1n~)&97(v0Mq(RRV8q)Q%EXsY#CANAO5yO5mlwFHBVx?K!KWDJS45;Emg5
zTx#vHYDkb2$YLV6yU$AbilQY#kqI!+q)WTQ`^hh@Oem$Ge8JiVOap#!*4?34PgYez
z&sIy>cy+}IW7ed&)cf_?sX<HCWkRirsWYoZ6ZLu6_a=KA(S%N$&0F9vh3sa#A4J&2
zTVBLZSVT{>3PLW0vF}w}*ES?e^ubLDaIx_T-dPn!e@}F%zh2PS=Az2ku5D@Ch!j&{
zhrpN=xM(}`M@<^2*8B24ki-x^3++qE)?fO#MCzvdaZUw|RUnf~Ar}*Jh>ARFBB3I>
zik>lb)zN9DPbpafnUbRF!!TMx6$o9&Vzr{_It`u;glVEFja}gn<iF!Tdip=(+45aJ
z{q%EE8m-zSL?6|^x6VPaswHy}XfdB^q!84XQxh3Ykj%I_iz4br)MJe;f_i~QZ${7+
zd#+o(uDM`AwD)S$GqmccAZU3%kBQ{09DbaH+a)!yMiKD0xxOY5h^d^W?A;Dc$E2)*
z3S^thd<I_vK6>_-w_JSuDUTjKBA%{rZENtp`UE*~q~aLNRJHX`O3y!&A58vB#~nKA
zLe)X`lOG#myo|bW<e9sY9ry^r$I0`rlw@U{I6cLjM{-Kj1au()T1se|$ZEC1M^lzz
zC<rPbh_RUrRDB4lp3KULYnuiMswzx<Pi$cwBJpfV>picoH|%aMXjZ2z7Ykx+=*NNG
zeowL<jnp8?DY~fR!IbIf{Tp{9CCb$`LkxtjrD-EEcAEJPNeRel<f<PjS2wh6%WA!%
z=~~*hqitK77^a{%=Yy5GbaTko@nGnCZc1Uf9B7(OH<Ef7Oqstf#U=(y@@gV_l=&A6
zyg~@X5E+Jni;D|3+bywaI9;F6HZ9&t3_2r~^P_yjoi^;2LgQOn7a4Y2LI7Dr`yKju
zYn!AR_j7cJU<ZSKt}<;JDUFnz@Hl?(#lPmS-1~K&o!ujaQxd8_j3RbOVS%J&mQdk8
z(m2xZ_w@T6!+y_j=<&{R@AN*u_RVke(FY%L|NNW?Xxl7BUQ(a9xVYqt=g;}_`R9E3
z;tM|c_$U12$3LEw(9fPd<)aVZ=Y#h@;Qsylgb;Z3>NOswc84<gkicZ-Y-?t<T5FK+
z717H%^Xk=0hGAfT*f9(}MKll<(S)ao4ej|EZPRjkcBY+XO40HTcagd`|2zVIannIM
zx7qqdDBzoaf7{EU6`epW>sz)jiUNn=?N397gP#c9-YXrzJC9Fxu!<<LzF2@U9j)q{
zI&JR0x)%u5L|AP_blqu{!BafAYC`U6G18mjQh9vdYxj+Ma8Kjpn96V^phy}RGRYB$
zRu40M1|j-s{)*K5o_T#vJUJ(Ozgh>v0t}gH7R>9dmeIutuBdMwsHYzyddA?$PLPu3
z$_{m|DjHW?sRcWPN)6PCkctK-+S*M^?CQ~HueHxuirKEzGfiYNc-Fnb&>(a#z&f!A
z)oae1%2g<%@;Lv*{js_m97X~{*L65E@fLwJ4#uI=GX>Frs&?kk4kV7jY+Bmrh+aEq
zh!ol;uw0lgrLEl<!qkAE@y-&v+7i=}FHxG!kc7jK=|>HwYw$FgWQ%h9Tl4OXA()8L
zGnz%xVH_D#4We}}M9&;FR?(%1s+aL|kj=bqt6m@%q3s%cY*b~92IkH{h|!*r2uK+D
z-OcavcTfK(Jbn5epMU-tIcH+f-tu|kC)nJt0k5l7OZyC2%l1+-BIA6{n_Za0Dd1Gq
z<<n4FTSWX(X;#-{u{rNbA(-6|Ko=WQr>)>p$q0osCXRFRneQj_dlubGDuv-!y}lL^
zh?r8c-mG&vD9#^QGfW6hMGJM4%E{qN;`Z}bTs?ix>cImRs}(*5TnuL9=@dx}x>+=1
z=3=~JaO$z=%x(nZnz&MfL2LbjqI89#C6o%?Vi|I-K0*Sn7#~piJ|fz`vKpHzl&yz#
zV;fdQDx7mPU5jHTQA5pq)Kjji;1K2Zb&G{|!$_GJTS%jkK1X5$>(~-bTDo|`)$NYg
z`whEspj|9kFP1nL=!c%e&}$z{Ltr(Vn5T?}Sh^{C?};L;L^GZ7a3Bv}RnARIXj{$X
ztW^@F(C>TtVFa&zgS)O{xm>C#bF^~p^o)-r^j0eB4}1FIV9LN~9#T5aBH9|M6h)0(
z_I1W6I#!Y?9pUQgitTm_&T)EvP7_-_*lNkp##YKm-*4!L!MJ*v=p1cqIBXC2Tn{SE
zsH{or6i_I+`>xWjri`@(p7x<Gq(NI=1@HN{;WK{w(?8<T`ybLi`H;bPbz7k#85lZp
zeDFIF#<6F&+pxL0X1m$Y?|atWns0viO}_QvH+lHrA*-sEcNzo^{mA9z4L|zPkND)%
z&$zn2<nrc%i&tNAbN!0#&6e%9C#QkG`j`JI(MJ}o+T=v9;~c{*)T#*n-IBf#e55tf
zdf)3owc8!o@3&lEU$WV57{)~3t9kqV`}cYOy>IaN;X_u-73XJXrWBt8xI1g4mPe+*
zz){=g)t@=Gr8y4zzAAv6zyI|Ff=aQdru8kA&m>L^`L-ZB7sNW#SPD2*Np$-&17uM&
zOD#O?HMP8pvqa9}f-@1CDJB&QLKEo-UJqfsw{9QafCfu?EhrW7y63uiOf-lzZeKlI
z3lo3%m{(ba7M|MaLf_N5`h3JpX^WP;d^4?WH(4KnbHe0aYu~HEpCTh-yeoTO=jLQY
z(>mSP#|&}A=MkivlC?Kg-?frZt#-1Zk#~AeTfmHokenCFsIT+Z(MUxlD*UURhiU-f
z97u2z(4Z~uj@q3`;;Omed3g?5ji0+pg|Ng+XNn@MGIg#h7P!~oM4b{Er<9x+z0u8d
zpKY5)lkb!!pFRQh{Xj0lx@$2L3j`G~<Wfk2{TUcVIeH3!l}Sn)f+sdnKYJT3094nB
zei*fj$XFPXn(ky9m#DIFrm1VWHO@^mlRCaN(5{UKyqayAy^g*|L8ov^iEKik+G*pk
z-LnrP|LySK@V|Wbf6e*fF*ld5G*NIf1;M7CCU9;prB<*(OXCSj!a41$?7TixXFef{
z!4-llN68F`abHysc9d!1@!nLv%v``W1~QOKq8|sy%2ka~tgR$C5%WE^WX3wjlnPhA
z#r5@+fUPY`-RM*#YeAs<*$vn^ZFy2kB8w0U94>b}|KTUB@7<$Yu4sM0EuzBG&Ldv%
zWID;TL(ji<KZU5UxZoAO6DBFY11mpUKJ8Faw7EM6n`G((D7dVOq~P4deQ_Z&X0>^+
z61}x(({e-}N8RL(OECgCmDx|u`HI<Su;?}OFQSX8X<IeP97gTe7#eaKCf;G=Jc}4;
z)(e(Rpxq^2-E7$19B8@)i{+ZGJ7t+xY<61?{b3fq9F4i4_IA__5Fj`@GyDt%N)ntT
zlHF&F5#P29<EY+wBIE*NKaj@6ez)h!X*;5(X=vNlzGhec>99a-E*SDEg(6lmvOpXu
zco)GRw{6Nsj2)%`V6j}Aa{K5Yu}#Z~kcWZ3-;uJ~z@$-O(8a1V<kt?Rd<4?Vj7m70
zFtyBh_fU~}_W)|w3pMP_jK1*y{qpyC@|~x=cY4Nh^_ZS&e_>viRZ99sc<!B}q>=5-
zHMh4ntd^&||K9t2`<uVUlSj{3c1sIb6D}(MXB-dQ-du8V@tXdy!%=9XXVFF$O=Roz
zpu4)d=K1q4xxTt)xm;<vDXDb}?x>xbeWIK)hr@wk80h-})%SZ2hdq~<S6p9ff8p(R
z$Hm13&tH7WX0y{iu&MCgd++j%_rAe{2M@V_@17#YA?lr$V(Q<!<y{?nH;u`0u=wWR
zC#2~pP^n1a-Mz-~bzD{L8GOJ66?s*$Q$3)kLnM_cDHQ8qAx279DY1ErS)ezGa17D2
zUucM*31)&J6C!GajG|Rqa*dXFyLnC6_yu&oEutcV8r;`=^1db~VhUIdMXLTW8$||~
zg&$MkgawNBGxocc?-Zu}(%C+%T~usuw(_H|2co@?cOG9n)>}n?kD0w|FsKK9DhZbc
zVos>&!`nV>^TD~h^TaAaN_^C@%~?@BDav_sQzs9Z0$bb8kLDXwH;BqR6pmHR$H8#a
zQf+j69zUh{k#%%*T{8o#wHz1%J|tq)XFCi76NlBj9o4wIfQLn+QplWCP2RMDloE|u
z_>^3TUCX{7*luSBhMe5Q*{s$#PJ}Uy><$BmAzSjG2GlMnuQA3*G_&t98OKk0a@K(H
zFcwl4iv#4&+$u$tyxusTf+4Urpq0WjrZlxHoz`&`a#$dqb5ZX=u-{Xrs!u6}gY@in
zJO07&_xWEu`(M#+7mUM>Aq^HAQ~;`2Wyw{jfhk!NOG1javU6saq@JFJ02p|pPfM(N
zQf=Cfsz$Ds;XN^Ed=YRKXLxe306%BEQ4_rVzKY;@Q~@1>qnNLy27qth>-=l!1fmI`
zGm~aap_XvuJJGQ%p$epEA3%hN$ECvd)itkv^eO9;6XP~C5TU0~TxMB>bxs`eE#G&~
z>AZ@~B}9e$r}(@gjxwV+%&0~^!0X{%>g#lMQ-Q1y0wF}kaZti(JSw3iD|4nqXxb^e
zsfM1>dnjs&;e25t**nd)Rg9t=Q*92N$3e|aL(umwUf<s->2<wSFTB%5;MMh(i|wA>
zFc20?Vz=OQwI+H`KlW;!V;Qy_jekLJe*}gi#BrqafzElNIHG8VDQP5+JxQQz8x<XR
z^>7@LDd+^GWX7Vf^<jUQev$WH`xZCt<Q3<mmBP#*v~JL))IcvyWbw+6sbpOpcW)?x
zKBqY-&`q$6BW38xLyvl)X-TpOEE?r$^?gq?ouT)^46rTOqw)^z%so~@)=l3wuu5o#
z8L1Eh5}$Ly^B?~ko_y=ivs~R{ESi}I7Z^E4DrI_qJwSBh+}z%9*zNe}{crLczy9ZW
z_u0Fw7c0wrRa8*MOrLaf-)(Owc_esf+sLwXoS&UCJblE;$&!+V;5(uZY;JG4S*~c>
z$maHze!s(^y_5T)pXUCdKk)kXYc4J>EFf-hzU6S(@#>}a6IR<F9izkHz|GAD^Qz_f
z^A}uQT~bP!5R<CD6j%9E^Yx}(iz{<CpNbS#?&vZ9@69suaR-p8#Pimdn-)<Bs;KoL
zP6)7~8Z(+JQUwb7j1_3LfVXmv#eoMY#+|D{qzgx=GgPn98AaAV3hFN?KA2CLmDcK5
z)Q!0?&sD(={h50Sm?~QjRFA6}kiOjd2xDkT9S%+OkNn-dRYOFqbgTSrs0gDXl(n>c
zTp1FB0qe)~-kXoC$5o+))1I-@ekNl4mP{#$;2S`b_xwikFd@xW*{L?M$8*F>wI>-T
zFVN}VLM^Y~&D-i&&%81n^NADh1EGOG4*!Pd;R|hvhO=_4r|nt|%IrO(*GgBWDvDiw
z6tpoCeBjWJbX{Y%SUT5R-%x22*?4tr+cKoYcE3l&v0ij^jRt&bG_9{o0yq1fKI`w(
zU`$l7<SSgL9XiG#QDwf#%IQV444X?uD-{uQ-jh<HiPaa?XkzBai&f4l;#c{`wR)!K
za#1^=qGr(JFd$81)-igTDgCAb3OXjiJBBp!e>?nN_#ZWYgY|ATB_wJS#Pkr$^mmIN
zyu<6krhz}lfgj4;$+niPisQ{WM65KVzJJqFVbLwL^{MR>!BBBqMc_Q4ZHb~{MdeG{
z^Ed5+Bn7v?x-(6v<<H%;eyQtg=mm^<!`WiTRM2jcPt=#$fU3YEmtq7=-Dn!uLSK0P
zlTTT!7c^bVN%U}ff`kaJu*f6bZqM`K5?9nwA18L8n3{;A)s0Nq?$I}GZ`6$5TSV?d
zFn>SG9&Lbq4EUmo)9Aesk*zc+SWncXC2QMcwVhg=OIImthF2e^ZL(Ep5@V!oBV!pz
zV^7<5grGegtH{hbJxG>OXhWonr<|-7yt+MbbJ#NU2V%EiwdhzZIu66YVd#zPQtnKc
z>A7UJEX`zkuT}tE8EKuP^-kL_;mCsql&)!rF|t@LabDX<jVW;$28J=sF6TA|iX!z>
zfTV|k%@Ug*21y)c*j2fE1fkoZpmqU;N|aZxN6pr98j!KK{?JCQPO}?)({v2uXbsX)
z{!CL9)OA+W`C`APWa|%lgm_HBAN5IUxn|<1!I$TWo3i2Rvmf*H$%k~^J%)tdvt{+i
zp}8I)*_yj+%fUyF9`VhOzR9D94|U_;A1GyH9D7pk8FOMtz1sY2Hv}iFRvkG<+Sb!`
zkp~aX8HNMiwKTEg<m5iXptdX^+}vKX-EIITg=oFA!=dNnk3Z)8<cxRUeTQyw!eVhs
z&LfAzFnPdf7n+sz!QL@iAh+A?`Q)dc@ZLM`p%yjm)cZL}o;K@ap5ZZKqg-m`3f6Mn
zS9$-5(H=CO{@LAd7>BRewEnD@RIgj@xf?VYRBoF3dRER7Hb?6GEmGBdny5GwTboRq
zgD3b(m(ak;J1cum^F)dI_5EwlXlDVY<f3Y9s`G&Ef%Tx%!_-d!S}g-t#Ht2!*k|_t
zfA;>YTbAUy^TdAJ%-lWV4ENs5#7K}RkSq?Ru{1vPQm>^q=|$f_AFr2|9`vNgR!gNS
zX*E?Xsm5lJV6jLHOdvCncRc4rgu4w}4||(=+yua)kmy3Sl+Fca=DlZ#@Gv*q`@jD~
z3~n`IsEbM%@1^f?RtQ!5UoRL%F?F{Y=hvFYvlV?*ca7ZkL!Mr!2P+VV=UNCzPtm|Q
zWmU~DQVHyu=DwfX?-DOIn_d^u(Tk2Sp4qNYu2qn{IVZgCAOZUUEB)WEx@p1fOR({J
zb2}d5gsSnrzn}h$AL$R6v20x%D?^;pK)y^xZiQAC9{_`2m!eHsa<MZ=96>`Hhr@yd
zn@CA8RCAi9;ldClvP_VY7t~_+s#e1~E_q|kl}ojCS(`PxSt~7Dja~gSNink3YJvs7
zMLjiY_)#j(?X&ezsyDU>o7#T7OvddW2GL$H_ED`B+WT6$<V-qWXsxnkvrExZs5z7K
zW>Kjz+JwQse|>!5ZL4Uy@SmptB^BNiq1baGdlTu%dQxwOya=Q<8%5V1Cb1W@g7Zv)
zNg^7|gEMQjtZQE=X<k4q@{ekP4s=6rZA{)kX^M7sp-$H_QP)~3r9Y{o)^Y^$%T0j4
zd;!=WLTF4<S!m1+qC}b`hJi{aPt)AN9#kt;%raLc5;aoZo%#HSA9Hwqi#)SYl`g_T
zmAmy7?csC&ar@Ww$RR}yv<qGc>XmCjhmrjgAEO$z8*Cf5?1q!gDsF-p52D%VH-<p2
zrj+d)WlRYzuG%RX#CTHRi56=75vc-dycnEA)H`NXlYJ*g%Cs7XD(A{_I3Ui`Flo8c
zTL0d_jU%K>j7)Epx0lTM^vITHmctEq$2;aYQETCRIZ@n@x+xvxd=J=iGkkKhVP9-8
z1z`!E5>Xn|71foqJ|ZEKmYKux$jt}OnC8Vi;Vx%$v+p^w{(c7%T&)Y%zR7FnL%{pC
zL4CvWL$&SPJf)(onAK>lXx@n0P_2|!Saad5uwf7Rv*Qhu2-~`H41rl1bHL1T@k3E{
zd_KOjQM1zrQnK-p?gAEDP<ZhoC>~;tH4y*e@R~n<_ZhtXg!zL+3kQlMBalbC`K^zH
z{r@CH?w{TBz2E&Fhr_}LA3mqm%&S+Q(Q0GMD{HpmFPA*bzt=5Ov{1EDS~h#2G;^C0
zw|DnUX(pwa7!q4qdG~0D>$`Vvd3-w48iY8}3~L{!-KRf$#g9Jz2_Zy2{7|VSkfz8y
z&)nZVBWK%qpFMlQv-=OYoHNSf5MsT4^NNpu`ZvTW^4#;It!c;>Fu$(koLSd(F9hR`
zviHqReW8v+c&w10|JUmiS_HN&|6RATUwZX(U)2N^=AW3X5Yc2}v8&f5u4Z+j<~ZcX
zh|rC4g-8sx_KhZNMWxWp|9mtXK9UUXBPmuCYDb*pnpxDBbK4Or7h?2cPo0aZSLkF@
zL;nV4Ck#wv5Ze2ZW;Kd6@Ma$FcU;v9ECnywOzGP{%#M`7xF6CC$x-KP<Gy)Kl-x2w
z3+AI>JYUsrfIS&!Qn<QD&RMjNpN=YazPY83n0O|HU4_~kKlL&2H8(XlI@<4L|KV0h
z1xHT}B`wbT!$l>3D*uwd;YY*(%Mw^paPeRlk699o1W@-Rj3E-En)O)_<`hY;@CBHs
zMBcX1)Pk_C+fWQo9z|Mf<>q)K=SmA|qhS$pEjE=xhUH8z?sGAbTIX4+W-7K1ALf}>
zmD0>YC`zEIO_{_!U^^qNwtbp0?j1%SN+}GpXvOJ3qdDk;&0Z_my^ijyRJP1oHY@zj
zXIjY~!RJCvx3wC%zS{V9`4;t~j7o5IHEvgAsWN>}?-)Sm0S=2K#r$>E95=d?iK-`N
zB@<VL0q7Wbu+_G&!w?B6k<=QD)+$=F#f;J^YC;QQItNv>E9wc<pwtwJ)z(K4_rcV5
zwIBX!*Owmwj|ZX=r7%l#afB6sjeSPlfzdFUS?9=k@fJmp3J;&X<flLQ5#hx>9~}=&
z5}CBJoZj-U&VR)Z(@W+)?ide#zZrd#^G%_*C5nwhIuYMLU~Nimu8InRHlg8TVbmga
zQX;96w~eZe$$jHuObFdlpwffF!N$Yp03Xn5#iZJ-#_RjtpweLLK$;GusgSqKRGae+
zO@yJ~{2zcy43#M+Ufe9GR$>$$tMcxWd3=1wG-VElnVaL0c}`Y5Y#VEbn6dMz`Rx&m
zBUlxRy#=BwYuN^8A$qZlDqCJTtrs4i9$b+XxWBu%Sh!_zs|?7sux=Z9+i1nQ$LbG|
z`sc>ja20}Y1pWYv8VYH#rQRKDK}$wUu>%IkwQ??nbG117F7H1sGj}&PJiUI+Aw&|5
zh@pqQm{I3i*z1vjTBw>i>re)d19mSm{5Tt{yCWojCa?MU^cl}S`y1|`eGN&Ach`MA
zUO%~ZK&5Ho!;e1V@BhIcP~AW7<*T1jHZPW{DaD$9K3XmI>5?~2=M(44nVgG-u8WT|
zqA*RCQ9Y&E(p%C*Ys%B<#OZvtLSPiz_QTD_!Ng9N^O?)}Oo}rhEX*k}#Uo`-uH3c)
z=%bIm#@lyq$*w@ZySu}@4c`LIknLa_@p@TVSL5sa?DL=T^7EHGJv|LVyNy8fy<2ph
zB@DJ{_71#%krft<g0Xyo6ZorMsykQD3ItRCS_R^16cSJ$Nvaoph!q-rL^sfH32y2P
zJ{qyYLBJxVjHY0<bQ_0Y=cOSg*h%<k_G&!|{Qy^|eDpinP>GnmmNmDn!I+)+>*)KV
z(Io6pZha30Lq@tJrQd&VEPKu|9gpl|vET>WRRM^<b~{+h)w!UIPyyBPWT|)~79x^n
zQktw`Ku6O7SYy>>|31a<cOu1!nr45_eH_XAqM<%D>aTT-DEm}P>~Ib#5z<76lNW@M
zDuL1pe>wdbpG_Y#OZ0$(uI`j!T@q9D&q|PJ9<J3sT4FS@olQOJu;vl`{B*<aoGUS!
z_0W_iz^D>E<9MFp;G6emF&9M}l+w7g#-%EkZ9hMA+vgMF1#VF{C--yH(ok$X-)bA@
zu!%OH3xKNn{R+VwI#<D~Qi~(_x>GtjH?UK2Pzxd$YawS0buZ++`nhh=eyxPDMAw{|
zO5%8XLl@LkSBl4&oQrGgGKM(zZfy$E#`9O3{VqeReIVETJx$gi8K2K?SOcasnQc>_
z0JcWWnOZkO*=Xy9x~(LMgs{ghiZ-HlR55sAY{hmGLw{2UOpdlTaAN<je*N0Afcxef
z+@w)LpeSBQ?UxZ|127o)_VCr}uxFs;%&VV#%Jlq>o7)@i;>>}HoX-5?@FVMCy|CL&
zZ2KsWQM|pk+4$ybO<WsA&rp4WZyI)mXa!0iO?Bt}=Fw%NoAtkIY7buWaP4aWfeB3t
zT6E-a29E%WK|K4>zIHh*lx*buX_`h?YwrVay9uXgBPchAiL8xKY_Yyu8(TiH<ufsv
z(s()CFfT`rd9%YeZzCq{>La|4+I0&8J0N<Y)!pykAD08I@%GJY-o1UpvK*M_nd9vZ
zX`Y!<GRJtVm^Kw;|E!uTtg9oo!AA^U7*^|++P0ykfYy<%s7+aGWAlHmV7RaZL@UQA
z+%6NB#|JaV^>tCzI^yHqEh3g9974CBu{B{EXfNi?_;VyLT#YCC{p%`8Rdtinz)$3n
z-+K9myYE$)JE7m-o7#8n4~CG^%!_9qu;y>^`pxG&K0KIW8AAX%@xS#A2dK61@a`?2
z{p@EJxwox0>c$F`X-drV%yKyJ{MmEv@1C<Pc9hgwX~phm5{TYe3WhX1yFK#k_Q;e1
zC9hmAk0_ZbMl=3JnPTAn?v{CqeEsVmQA%aY*++?yd78Mnxgmw*0v5)|XPRfix^lU!
zeDdk1eER9n==6$?Qopx6-z)a)-VYB$z4{t)go*27{mWc_uzBI8u9OWS4ep*57^dhg
z+Ki~}n|TX{df14si$Qj($$0r{`{SU<sO$mt{#E^Ow4%(Az9HUm3W9y5^O8D3_l0{0
zJ5p$kmD<|$0#Jf_v+7Nk7p~f>=W1Kw)DdrW(T4hK>PptJ$MXK?*~k7m_DV(KUO3n*
zkG5CaiR(yYgpjQ0O3C*B(ACObL{z8g1WZ)cUpqpGiFq=LjTsLY%Vw^H+V(8#FLDTh
zaR%6uFp*%KmSIm)RmfWSe*9B@8h*?%roP~<AV+ue<hTnFMhb@59umZZ4b;Ze$(tgJ
zPquMvX;!dBw~UF=(>nwt2xT+V=28tD5TLj^c`J=v)JMOSOR-#D_2NvTaic?;#zNV$
zqopFg3yB$JcO<2M_QELef~$1JN7+G5p&SCEF_B;odKaVhv(~5|4QeIijn$h7t(BZt
z=l%J*?$4!H$emvDJ^A}QpFgrfYAbesyf8M<#RN*h-9>s6W6zmqde=g%Wq>e_lbael
znAK90zMciGZk3a!UM8@u8}$N`FGyKoTM4z;6B8m5tbiA%Gbun!5bW<$Fd^(A+1CUK
z#8N4t4Nm5-di@#-z`e-oM`C|?IzOO)l{FY)*csS@u?IM%h*Tx0u%yK0TzL04pYZ19
zmT$}l4$H*DyH{Lp^YD!t6p!)HngZ#IoBnv9eW3>T!%}}J_m<aRr8qPeWh_bwVIXoj
z7kj<E@C*?f^%(ch$kCPNC|)P9C(j?o$0B`96MR!>sClO?iz#hOu@Shx4?!U9sF}Yx
zlE5JvFCpH}gsJiNlF7AEYNnhT=gZ1rnpx(F<<9&DPkCj_WsGZw4V><uZ1LOHL5iTR
zptTod4+CbzZ8ik1%x;zHCc&4>%IWEepj{DVK4zvgj_<|DjiN}+hp^-oTJ=;~J5X|i
zT-C%Us$^_zQT#O_N)qAzu+X;6Vvd@1`wf`q$o>5-AqGlWjciQc*iG%&*GTokw6)&B
zGS;@2e2!G^2#@6=NB&v*oIhSq<fk)o?4yzft(&`!Ck9k##k%ZuU3vZTC9hw;qUORf
zFJ5d!h{+C3e;qk*oSvR|_3|YzU%g`8Ha0ibR{tF-CGPI-_?_SW9gfQ_$HS2jB6FH-
zWF$(IMhbgm<(vY~?{9hW;yHK68{!l|3uROD%=8LWf)$JNbn6X!@KK&U8zBUyX{I-8
zX2YT0Fx3$sXDJU#Z5HMJ#e$(<Uo2x}kW7B@_3J4AKwO-a+>gw8p^77ciJ&{$)c2cG
z3L5st*vyj+A;E1<3?*}W8-L%;M|0|X>7WQ)6>%L6YChhLBF<07xv}lN*DoDi)BfLa
z6ii)!G*UbK&&~hT@?v}AQ>zt#xKdFi411KZf41t++x~gIw<q_X<*Lu_T3|-jPK@{B
zN4z-})RIwxzjNnD8He89pH|~!n<r%>1SPr^nboU?0>;_BTh)w*bH1=;Mavh=$a^v3
zMc_~c+jpZ*;8%$>ZLB5pXVahZllVhs$5mr+CAO<rJIW;zM|z6VcH5Z{m{Oo+3lxYk
zxCy!4w^r<PP0@IEwN^7@_VYIm3Wb>9p(-NDBIar}p$3&)Do<Nxt<{v(-e3~?<iMH<
zZ7#lYo?KHio30&I_98Fz^U;gHZdR^-))jY25C?fggj|Z*COL;t5F`q!nGpr)O?9c6
zpuv*36k1hMnwaLrg_p%gw`L{w!Qr0&S^7^{mZPPiq(s)9>>)##IQjT?@LkoBzhX_!
zXd>14e+q<@h$)g%aA60iR&bSZF4aV0{=jK5vb~goc*EW4OLm^xeU#?sx+-mQlr%au
zMqmz+Fj<P)8Y^olp7H#vUthie>`w%}c(^L~+WQaGwKk>~fzbVIJOZo-n76|E`CDF1
zKVW-yNBZy?|7H9sr;x356anYjp)$B!s)j1{hqqtdug?lZtgs8+7QxEV=3Uj`N)N)Y
z{|ra&a@mN@kj~!0cImvfZsci;Nn|fJ`$JJ{cR-J$7_2}og%~5tvQSGUZ{~9tQ#Uv^
zG^*7)7Oeo%lw<eS+cruvG%N<#s@-30aK4<lT!c9Vi)TCBa6H%pp0~`J-80XTO;mTj
z;P|xYaKsRDNvkfG4=l62-Vh^NE2opCMFcSkywK3xo#LS-OHde!U}Y}~I*KY*6trsY
z{TqSoMiJITSmi3%*m??ZGtcI!l#3UxjQ|{H;o1Gn6sjqmi-`onyyvQRw4@_G{$a8!
zf9<gCB;vk_&9jMQ`sVp*d*CnDkNNJ?BjJNgljt6e&V77O3M=Gw<>}#xm!JKNAN}x$
zy!`AlS}9CZVp$HP>A*A{Z0wX0BFeTEF6Wg%bcD%1w~{y8^t}PgIrGs+UmK&pzG;UL
zj8ZS^g`lnXDMoH@Zg_To$KB1*P!?=4*n!-~^olwUCrGf-R0u=WYXyVFBbHLQTrNC%
zuJGHpZ+ZRtH7{Sj<n6-)r5`F+_k=I}_w{b_ce$?cwO{!%;5{I5BE&?TY%~{wKN-Gf
z*l|^TT&iOd0A4`RyqF4O<gQfP?>g;7jsIq%2e;%HXsDsv4TWab9AJk}E!LRDjyC$M
z@mKF4=04^ZRmXajro}%`Ls42YnFRt&hI(o1Xrd*T1p27d3l4CeuJM}t0p9uz95>VW
z)j@1Dmh(1>gw`6<G@D8pjCbs04x>H>E1<<1iAY4%6tG&d+2oXhmO{=Iq#|J(D%(~~
z*_oWO5>zaaLnDyJsnkVB@1GwIYSdo5xe30SHOW>hf5yM!&GHGi>44RHUEmR}d}MJ;
z72`@aM^w?qJSAETJf7B}o=+w^rxt&|V#@5=9O-OIb$PTZY<V4g${;ebshjS~+*aEa
z58Qm`Kx|JL{7412*r}T-qx)f4*nyGvh8S1(DVOZceuCB``WQ;}S~`$oIr5zkSc;tu
zAx2l$+WVZuEO25<%!dO!`@qK9+e+D1vu4xAG%q~+=xh9b`C~pheuHv8+4))<QO!p$
z1mm|){d1809n{)f*JNC~K(>RF=%Z#Ob<=l8@r$2JxEei3BnJ2M#BA2AfeTivG~>Po
z6;Bhfh6=T_TtvXVjpLN8>32*#MInaCY9&WRzs&V(Q2`Izf$RTfhm`x(3yG$l`={z3
zh)eyu!Ks22VG%fg`WgE6HS3-7C(k}%<n@YI>rLI*8As!KHQKvo=w_Nb8GA2QhmmC6
zxq%&(>&m9qI|1AsJp}jb5idA>6s=<J@sp8E^S1jkb(9Yu4>l*N#u#ZX$q#P-5<}p4
zIB+_hEG2{)LO)Cdh}6>9N*={*NWpTGr+FX}Zyr~oC`FW_jV9(|zm>|GGvRWv@!-60
zxL>%#5T$js+~U&Q#>YDuN5}WB-qrkFrFS`BD0v&5V45akvUn^}VZznndPNTRmC;ps
z=1OlJYQqkTrdQ|$43*UN!6LvCU>0Es0jXwn5PWA$0=I|2{V}mbMd~_Y`??iOKd^VO
z*gN#vcKZ$=In`VtbR+>6jo8miGbC-q06*as<?%I#dU9S{8nyzq@!LA2RkrQIyLWH+
z^e3P2$;Ut8)1Q9Kc6lO5U|9~tv=GvPG#{Cl*(d@bvTYSw<b!7)v6YRd^Mk$qcH(kb
z?NHt_r^hEwPmkQ*+|YuO*Ns-npf6aATYpJ@;Q2ZorHwzk78|C+L#>bTnjeUzX5Zv3
z<Mez!b2^{7tQX!rop?N*czk^1a_LI-FP3(~&liAS*m%a*{%zN<2?GqqOB^}pxbjkr
zr&nR*aytx2-|=j1d(m|L`h_cn@GdZFbUKU6E;{#6b)OoD*-gnnz`WpU?yVr=+$<R!
z!OlZO?3o?qu_Jq7pljmKTO;C4tPyRk2`ifwae$1FBz4q8hG3yv$^?1uU7O)_ZR3D%
zwK}CF4HOVSLxs?l;?0%$g`k0$?E7=A7O_`N#F0xe9_&`wiW#byirJc@<cTT48<rG7
zr)f-%ls-;=9|;_zb$i$y4Fj#pshs$J_&5AC{WVLOm{M1C+kJ0!_cW~j@77K=SD>2D
zo9I>iqQ`0qP`2WuLa~L$&ROip?3RoFUR!g?vvNB!*mKLZHA0HCZ0jV3fV%gfbefaz
z&)OO#o5}tD={YcRH6ph)8Fs%!fs|(7Pj&1mU`7A78Rf?K*Tx^LwNrrb0<`)0+>v%`
zv=O+RPRxfRtti{s_<`2WG!jaFc;aSWIHZ|N+$frDbRU8>%2<J!Qe<`_ZX3@|?T8RU
zAf`YNFTA^uGrH2fDSfC@N})AnIoR)Uv9Rr8!R+z+XJK4J+yBM+hxYKQq4z3CGSsoD
zLQ+TBcTGbKfhk7jIQ^0x>+g2`8aaTBLsaC78|IgeC2!l{9@Z@`Mrc%Vco-Wiwzl#0
z7ti?4N6&eAIHR%GBEGS<E+CL0Mxi-BOaxqEWaiKBA133b+Y7gaFKyqV#eY{{=&~bV
zF(oH;R(C6h!6WqD)zopj%BpY9(L1qtHB#vhfw(%bRm)&J-W<7HE;b^cCRDMaaLpqa
z-OvYv?bB?<)*OYe-Q94UBCj7VJYFhMgc2CtTT?|^waBscV$of0<a{_BhA-b`+sMvI
zy!KQZceCxsb%#@{XF)bq@Zfc=R)kF=gSc+$M|(ZJu-j19S}D~$s=AotnyqLi5H?~m
z;goIcp*9lUVPPIp0p&Oc?hc8Y$;h<Q{S6$I0IEP$zqJAoN20kpEIYr;c~sho&wZ&m
z_ew1xAxIb=cOu3h_}ANy`RU`2`GfLZ;v`g#qhpt(W1Wdqw)M=z!yDedeZ{-CZ#kcy
zDBFov3g=TqHL+DenMl*byc}4T8&W!gBwM=@nbVPZUddabhT`#lnK!Rq@snxc>EVeO
zBInbE&p!Q(ckdo~I&b9K(<aOw;<KN9&a>NRJl;K0DP*mbTzx$>vOfbQdzN%<<Z9!S
zzGk+~nu(k<dCQbqjZl7d$oA*z{p;6Lxcypwj<BQceP6(vk|u+2Uwx8cO_TqxHE7kj
zJ#MWM9qFq$dZuH2ca#;vKx+F4*o0I5uC$LOLnkj7LJ@W}MVpB}DwUwUm~`Z(PeS~+
z8uDiXso=;)w}I;oQPm37Mv1Len(tu}>iB*eN$+9_Ay@(A-k*D+YwM(BQ*r{tW;7B>
z&ZG9RdQ*n=h=`)4kjjQ^D~`>e#ZabgTUm24ge+I*p&ODPLZo>^9%H2SqS-tD!MFC;
z*t92$C<8_HsepZ~_PPAk^5?vsKISeRnLP{I3Kv0}t$DnO?NlYygdQ;_L#Rr&4A`!a
zttLP#H0K0*++YgStzyTUA;ev^+Rvb#m8*`tr8Es7rg<?Hq;Z@RjX<u-CD&2>_Vdh;
zHM5)9(LEXTg^nm2<t4Z;q4NXlZlTo$IHeS_bN8m0$hizBj^>R(h;ipjx=C``e2q93
zA=<b+KDbcCbF`Z^tvMGCCGmCrI$8>opPjk3QP?SsWl0>DndF~ot>$l=wF{7-CwHV2
z+=x0+s<JsUtqoeR26$Zys9ErYIbu-@cF%j`Cw}k6IkJ774G>ioskR@~+#3k*C&i-Z
z`m+l?u;OoPlK=hJuV(}dZXqC7qm6#a)tBy@oQ{8XE`fbG%@<k!{%?N$Lk_orf0TY`
zq(|^h(1as?cZTzOt6nnrJ}Q2Y@6<9sbbKe@yB0?V{b17mXV-5*<Z6+d2VTU{!mInC
z4Pv_DKp@w*!56Z>M_+6b`k2O4VXZZbT$}feTEL7VYjp{~-+nPjI@e0d8`GR1Ms8AM
z`QVnfPZ!=iX4b022wQ9BvF6WnQzg52;Cwz2Q}SHnTaGtJvl~%m+cM|N**2zX5qqjQ
z3K4chPZ=IpW-DSH5RUo=d>kx$$Jp0H1xlmaAAVsCbzPqoSJ7j#Xh1H^0!xTYg8RrR
zH;0+y60PvlV*Io#UaE09EVX3Mn+^Tr_Dyfk*4$G8WQ^jwMNJ41QtzS-nOJM(KY#iE
z<nNvS5ElDvRqarWZF5vpeIu(($-Dql5JatTh`~m*)m70=E44(5RANx3m`KyneZV#g
zL6?bTn%Tl?h)Ayd?6a3VynEbzCQIYt@rj4EvbC<)w`cOQ6@L8jPdPuGnNnga8yRb8
zYO(caO#$@mR7KwxVCgOb)BY)r0_MHP;g?*$t|IWuUgBhRd$>DNp)OA7Vj-&?s*44Z
zMO6CwZ`DZnEr3+)pW=I<?`Pr)O0TBucSz?If-M#qV_`aajIGH)*mGb3no`sw&m@lY
zfq_k4qk?#2B|XW>#cvq*4pl_mQ?ymvdxMQY)%PR&dHXq__#rCId9}Mrvlk~_fDuHH
zD%89V(TCP-2~DNB-=okll)YVqrYGjA)R3Sm+A4Y5O^%g4>8UkLap~TnmQ>W7lHmop
zzou*9hqh+pzSzMES0TrV7KFc;{xzS*A9I+ZWuphT0V%dlQn0hJ)NE&y7|(WECPL^w
zk(LBSZ#spYcPC1k=JDr&4uXgm#l=1POkE#Cpw(<VJ|F3qQi-u=w%d64xXkAG&_J^_
z&RaIeiEfeBoHu=au7@B<G+|9I3TrK<WcIacqKe=}YVonTpM}o#%U=Ae7fn4Gri&+~
z7v8>}WbBjuTsI{3csz1?It>nEQTW>1?{NIuffyS}tPo$OXnLHW)Vgt)12=~_0wmUK
z_F?9}VEb+o@dOVcsu$eV&huKT?Z@MBwKcP@nWfh+?ok?>r%%;pNOg0QX@AajXBmxs
zBw*vUrVTBHCXV(;JGYzQjntRh*?tXKz<Sjwic7#dBKPxKc&BzUj&e{@n&%;^Z;FJ#
z@$QD-{_St`)A_{r=Qr+1zmMZ+J-77g)wDgF*<3eVN#w{|znXXB>TQL6Pw?&?Gj{H+
zuXnV4(`?O0$6f@*7;Izlzqb!&=SYc|O+@V*uIg2Bx&IEuy1uTU)Ygpu2cw9cQgrc(
znTbx*%)HE`6lkqjhb(Fc%yuC};rTJLB;oBTb1qI0ZryR-IVx&Dv$jfEE8Dv9^mJyL
zCXUAghodQbZ|`o%rLe6Vr5d^vaaFw|09MFaKzPwBPbEy(7hMq+#tQJyi^h6|{Pn*Z
z@%GYfx6DsY`v@1f5xAWLON1oKk|GBm?FiWonBD)(J9s28$3*9ch8Rd4QtzjvzldJ2
zcQSqm!j|nJ9iuxxsil7n$~*=BMg0*k&!6zkZ+wl<-kvzz+(Xlm%ZZPb4u_d<eB&ej
z!5{w-H;2Gq|GNV}{^4Ko@a~<VFJi^v6c*xi!{Knt?agy;Za-k2j>Ku=e0d@~+In7G
zk^cDj;6-5Je7<lxoqcTvwp=M1@RjVZ$KKD=W#jV6XSim6$SwNw^?sA*fB4HD{QgH*
ze||ki`i0lOuetpXR3AFvHx5PPZg!yy!vs7;d%-bA(iE7I&{Cjm8>N|gQ>%HO3enLH
zpAx7&7u^eO5RtD6Xa}kc9qFUYXqnnEMzb2Jc{CMOE3L`!ysNb!vMX=jb3hmknb|Fw
z)r6SuU^Vo=mQ1U82&(8*233!1w5PU@S4D8-aaRZ%GT4h}+=tK=`ZZHGL8BD_+N=Q7
zx-T{bB-pq)*OT#!<3!Lfc!b*RmYNk`!A-k+V`mYF8KF0^)^WDxNj2KdvzxE%{+UEV
zOoT9TnJPb+{*s@jA2B6iio!HSdQ=}lXt_BZ!W&O`uSp&z{f^uUux%STmr;PGn7jxv
zoVR((RG~Rvt}&-%1$b*@jIuVRiK^9+ETPtH;u9heS+fhU#QwcjTL-1?++F)Dqpu5B
z^!hn(3<4SJv-|1x$w`R26GnGE=uJgO8e{C<j_M}y2Iz#~{68#Z1A38F=@S_dyBF)a
zSrJVm#)(B{{@#mk^U;fE9Og)3p>AgL*1VuD6>g3TcQ*%DRJ$juShC2rZid<~iQ~;g
z3XvYtSc*-!*3zjMkumv1%<88%iVhTo){xSmRP%|{>pjw&4KGw%U+9f1s~SIl)Z8d)
zua~|Ur}QgaU%CJ^FDe`FrLlSWi`B}m`L&IuN%IY%|GAJ_U<t`)?F6bcG;G}6E<C@V
z_~-Je6>2?OcWhq139+Mq_~F}aVo-*JdlXzS%%eLh(Z?)Sym-g&KTjF|iXej*6&w+a
zzLAe{rf))BR3J$p#6eUZa#F!JM*E=mj>}4;9ZBsYv&P-CJKt;-G2euppM*U`tu;b2
zTNCgxb1vM?kx3di1U3~;rLpE><(2fYMIfp=Uqs37*T&m-q-El8SXkzTo7)>3*K03|
ztLNd`gR2VDp{qM>!xt2~A@zU{?MAfLJ0B%<Cwwo$MJ=kSKRj!J+>GzB0QV7I923tD
zGlwY|@6Y|4)E~q#-n~MYlPf!;j}ZDOy^njZN0c2M?p=DoHovCENVGA1LZTrtT9tzF
z1Vey-`|0=j&WGRT=}jS}-=@?=n#9+XZ4y%oym;}PIsGor?rvG;18?44`0Ul$m1TVl
z1j__*HlM5e=l3kj(bjxwG`S_tN~sA#rs_&qZBw?y0=)1rWpXDZTY=kg*UtOfYvF$V
ztzGBe_IEpwyN}!cTW~2D-&Go|WRSMoPx*+;)|nSQ(V7WSh|-{_DLxgq&TwR=bB?5=
zTh_I9>zC^oO|!=FHp-m@%)`v>X)uLx53}DokG3gQ8c873=&Jr5X#sC<1*%wJy^C9&
zZ|6MR(x9vay;Au4xUYYXMgcIiX^O7y@B1QZQJ21#Rp%E6Q3yM-uXO_|rQj&~&iz#}
zGkj4O&xAnTl+1>R*=aBe3`6aLAFlB#Be!*tSt-ur^t<02Ve%<Y6Dv}+B`c=VftR=M
zIK>MerytNl<`@IVW%8`uJ~|9g8rkhbXjf{|rl<ziZ8O?VbhH@QR*TX!(TMG%)3n>W
zlwu0iX_<LApPbWfg=7dFVG~P`qPkhL2|h};BEF%Va$_yk?R$J&?p$8M3oRY9xq)oA
zTA8N69HY6ZhCuNKruit-#<I<6F5>5<j`MQM##@h9mPKes_B!%vYWD7+(M_X=X>Gr}
zsGHJcXNCx0U%$!szWp7Z-%QMDww->S{qw0dYA?d^uyCB`;USx2Ahd>-mAtK}R+cyu
zY9b~BNi6o!CO<V><0E3q>W*xRudmc&a1}z%NOt6}=S{P#`dfpZCKUa7YSNp&Mk|Vn
ziR+As;FJcr=F$IZ*Ox~CrII9Yh!ax`gznDXF3^f+-eKLID3NLfp!PyAFtxx^CzOcB
z0I?9`%5Qz^L-KUuPvTF^1FlQM`<PaDM5dQ;-jvhPh>pt9IUORHS*4HeWJh;GC(ZUw
zMXqWTzd{U1whuOjz`AAB{5SWd>4jgE=$&rxqR&MJz7V>%SpR0@kQut!ABO0PH3{Nm
zTI*;f_Iu!=>cLJPX#!OU+6HM~RhTCs=gcwM!(1hjLga*UF4f%MyHk2E27xL%gcx<(
zcw9Fgc^qC<DI_D;#z2}UHzT#Tba1aNJJ-~C6gJ!q!Q!Gi30hQ03b(Be!{gfO7%x`m
za9Cq+4@v^=XW?tNk!MR{79n8Vsyap(>YKP7a^hsp$>Wuxqq7Yi4{}$@4RChA4?3EF
z-m$k{WTMoTi5mTDOcC3#PW4aT{5k*gpZ+}#hZjt7qv=A^-DfM*M$pDIC0;y#!MbMR
z^o)1s$j=^ND?$v76f-GS(z<c3g+d?;91phutlNdl`b4c(sDuf^B*Yd_naEqW95Lmo
zR}O>p{l)4|*^$4$Wf$OT_5bZ(*8lbIQvCf#QXlRO0QOmV!)Y8bD@+U}bk_ql&Rw7(
zp+aoqPz}<Hz>0e8)ObD|h1am}S%xymI4tnJDFx#M^#ZKaN<`QNIUPAMue&~3nE<nZ
z(ScC-Q}5FYHxCyt|DASFTd{&q+xQOkNrw6XrfQ?%TnlZeETfAsfW|Q7HNIpg{rk;r
zokC##w(2o^`0qh>o1Na!YxL2%7r|YrA3{eoyH&ydV+?hvtSy5!Lep`WiYsUdfn_=J
z_V$VMG4gJC&8z9(QDdVt8(UKiIn=t@_q#W$bUKMxB1oDNC2yX9;Q|L?7+D*2sTj|@
zHS(4T>DsCcM-gFJ4kmQjvNagNnzmeQ|LSy<Tvk$y%qh{T*(+^Ld0H=QHIO^%L`Sj9
z7~7A_kq-k}>>0m1EZiQ<apG{e;qjb#d%CcB6rPqc=vd;@kv?~89(Z8bj7_^b1tCZr
zsI(!&9=oc2DA9)Nf*s-A^+WBM2$96h|1kVR{@}a6&Edu{yTPBt3fo*Ftz?#YVxA2-
zZCWi>wiV~jdWzMR-I+i72tL7*u52-71L|h_s_!)u_-m(I8Nt=MZ2K5jjbj^n_Z7qV
ze#;A%QOgXVDeO=0uX=sy0x*a&Czv8es2rPvp>RKcoV@=h0w$$y%?>QFBH^%1{Jr1)
z4*&M>z^8JCYPq?ixVu7Qyf5}mrr%`${a+{&n;V`A@k+w(#Uu8>hVg;x_}jkYh%3r!
ztt3bOO3qAaqSP|113KZmN7|2ZS91cWL+hws-H{H3@9^;$)2J9);{@Wanl)ZWMTzv-
z_!tf57io?t8mgF5@2CQ4TDZ-Pcjq&YtMa%tHs{_Ihvcow^&}24gaP{37<zanGdLQu
zE3Q*aW{DCEWfKu<3%)+=5E<`5cLI4S8@0G)g0}aTOGjK{Gbzyr$HX_D9k`zYlaKmK
zX_VRv-)n?pA032XQC?|sE{m_D7<Xj0Ka{-?ynY?s?A&|=?cwglV}G3{EU^w5Rlfi9
zDgW%Jf64EB=f6kUE+99y4tzOxn+6eKtC@FCCqDW7bN=S1pYXI=B$L!i$X>IZg?CRU
zUcGtXaQKvYK6uAmX<CeX6)Iakv92q%s0%lm+eA3mRFx`yYx?T?#h2`bME{d3^jftI
zQJbKMdQ2e|X_=%e5|o&f(i*ubO$8-*Q)kU=@uI9%9nFqWKd!tse7tFeri^upnhw(H
zMt4P^`^<I3x5BnDyg2cF1VAgAL4uk$T2hn(t(pmW&KAORwZ9_44kzc<+CgXkEEUj3
z)5skC*$83Kg*uf+f)2jpz~-IbCEjH9LsG?QJpEwoh^Rl;2qQlg`u^=aKyBl>t}Rn)
zCbn#*)=dc>m9|Bww-Z0Rd&Bu=<zarZsZ~d79O3EWp6dIe@u1B>*~Y!zoJ;48chswo
zNbPH<d19o-#K?BJSSr?(sM+qlMJOusWX9L)YGeK>2Bsw$6{nSPJ~bc9YqjL1=nfdA
zL2bsrtG#ey7i(VASb~T*>}EmZwGLaZJik5g;Rp9jX<=C|+#P4$JY0BO3v2aBj~e3E
z6O{UA(D&NDzWbRhQTBbdpOwm>R`tt*1bR_uf7i4@h<x+%JACiE-{t0J=6Fbh>zQIV
ze}*WP>gLLUiUw>_*)meC$r%<%zD|dt+y8$Qm}5XA<Nb8Lzi;O6z3xD#FNxcK#ewcO
zdM37=hgF)z`nl++gH%H*=HD08r^H`!+y2r8;1nuzQl_vQ^(w(pqiZ7nFTNBzWDCkN
zMMLdc=J~T*mL>81`VtMzIb2tJup=?!lj=Wgl-)AWM;ZI4?1#H^dCXK-4Fz!%H2WBJ
zZ2Du-`u~FaoOK?9I=Yo4U7?dYqLsdQ543eZaaWS|1&=9#t)WnVIQ@VDR~Fl!hiF-%
z=1kwyNSpxe#j2$wL}E~CsN}p2a&e^cY>C{=w>+#H%W3138|R|rs@}aRA^S)dP-!;0
z6)Tb#G5E+f6Y=+Iwg+2ltlMRbMTB74hhFreePLfIeMOygZVR*XF7U6ZC`FZnz}Igl
zK6-x3F$Q9*XsMK9MC+=}M|OQAp~J{KIq#%H_Bf)tTkPy`f%b-`A2{LaA&1>O)IYn<
z1ykuLg`)}zH5dN$;bXp={|MbyWG-m4oZ8-K1PK(aJY6n){K-%GH-GVcK6~?qDtqEj
ze}&o<T`Q04Bl7N%AzT3neFN7qT5oo)?vKY<alJ|U_dTiq|9oB10mM6gIt{?YGd*WC
z3AAFiCV@)OjI@kk3eKn?t|YVs4;dUw^*vMBxx)Rh?8u2?l#UuAAx^~T&rRG9mHwK6
z;W*HdHxxpWoe!;QpMA4uS1BZol$c1Uk>Wj#N1N>lqK3G%@rukwsFrMk4R8vhE0tq(
zzOSQX5}1O!BxrSkhmNi*T<^6$Q&;-5CJp3qG<6~1(o`#~d{1ubh?Mge`=6C+W8X#~
z%ttgP@=f5g<0qVsnK#p$PBsRD<ua>1K)uinnDAnr0*7Tb9%cx`+&lp4)?X$X($Rd+
z^X&QWHahB-C#9H~@HE?=RdS&>sL|hd&4pSjDTM)2h$CG#xv^rwv`0O+winBN4f`JA
z1=$!%_5-%x+g^Y>-@S=Ys&Kv(;<S*Kh5IFOyktUEmTBVcx^Zz{9$2g@9hJpIFI@%R
zFA*Pk_h+oL!?lg3+yIGQ7@8<CI!Y!YNJ{+Px4y^k{r<NNdH1xj<Q&;SdGYC%qMqs{
zP%~OKgi2Ib?G9i`>Bi*lLWu~ZP7%|6Z_u%Z-@j6bXmmbpGmjI~GIaEC+z4wVL-6w)
z&_;2w4{<U*gG8%uf-ud$VH=njAw?r=N>lHFgL}C3k>B5bbxv3d4XuUS<HSGsgWu=f
zZRJnXTj!5C6ulD%tSEY~0@nBb9_cnC``6oYzf$h?%~rx*xcN`$Wd8lKfA7`+o&Tpz
z2luU%vXfo)>iflcg-yqU*Iy=h!O#&whumweG#{_Vo?ha+>-F+fho8+?&>ozAsEhR1
zt-wT#!7Vvfre)@kW~S+ud0u$=?uo-}hPUTSHuF&*CuMD9U#zN-lp*4%n0mYV{fgbX
zLhJ+Ty;B3DJ9QMn-~aBJWFnCujLbuGkuQ{{Yzl`YeCyf5*KZG25O3x;SF)+Mw2%F<
z#Vuks04P^>I$nSxmR=&=|Iapu_8!+n#+s3?p!7{ATp<l1gx&UFZ&M7>RV5<)i|r@;
zC;2TmWkplj5x}e8R;%*r?OT5MlaKkUAN`2zD8qFD_=V-@pGg~^a`y@wX^Y~;-_KoF
zKj5qDmtNy=#GhZEzV;2@C?r;FH}P~Dt%Rsf^=L#5XlTsAL>Rf)sIjRz7j#}`>0>a&
zEslJR`N*!LNg*hyk8Q;gl={goArf(`r(g#xu3EIe#~N*K+=6YC_B*P2nuZdRkJ$!?
z&CQ^l;(}P)P`${}S}i*}ns=Jz*#@&a35Kvo>E5bVcm_wJ%?hg$h~xY1javieWh%pF
z?7ha{kwm=6q>ISxHLBRurIbPwAuUJRZQyY}^ZDU(9_L3x-*j`a$n5XCTXcB=rT#N9
zo?Z%Op59sp4Q#ayWoro0geGDKe=Wrlb7sph7V)ueF~qFz;W7nFNU6IwW(d*HTTjxF
zKKc`=MaO74r8tV)?lIc|fl>>(s`CTw`@3;>7vkV|5T_NTG!a7}W6lxwYa3-PoFAW9
z4hN7#qY$Uai^Gw_a^T^z@^E_MvRQ*`bgTLvI0C&%#+?p!!}H*jCqY=Iz;T&bx5CqT
z^`^qIj;FMYru>_Suk%0s+JB$GhSq?VN}R(Gru3;&4`DD)p<Bq*g4AL;yh%vSOui>i
zTWPL9_gn5~l)P79gN^`$tI?x<eKfB{X$_U0^=o5!ZFQ_^JEI(Q>X0_}{8iL^14W1$
zkO(nO%+vg(_O>t20gRD2McU>Wp#k^#a^>I`&X#L0xgcW1MhtzcwQ`sut(+<C!v7`z
zl=HNi<%_(3SasOn9!~Z{UG?XP{J}dyP@kamBG9_$YxG2$i!>C}?F;sj(Ba-7NQjne
zC^i--)v|kg@poMW^w(q&aRqIz3>tzNe;?yJx<8Z!h*#&6n3=R9aql>+9t`wMMLY0q
z)Y3sMvnvVaKZeWhYhlZkNg8oZ+#e657<v8fky19EJ-?+k<+5eg%Y~^~XV#Ti_`RyF
z7k{SSwIK0ET4ttsGPBlg<B~1tJc_xQAEt$bIJeW(Ur}HV7EeYD<Z9-`(YO2iCGpXV
zJ3crjbXzG~q2_K&Vp)=5G7eJ=!FW=@JTHKyVMO0dRa^v;66>~&Vtt^@*Z+=OH5U67
zLPvuwx~N;@$YeZ@-gTSdY~;=Q#6LU#fdAF{UE;JNsg5-jB#>Jrm&VIiZ}_Vp{E*LI
zz1i(>u2AA%R<HdcQJC(J%++<Z*?(31{oO7f`>XcnzpHfjXV>d06Ta1Mzfy^!5ZZW7
z&HUAire-LD<$&7ZU>V(g0<iO*p)1Y-|GO9h_1c>fhG8{ohiP!RxwzUi#bheY>P<3O
z6XeI6h!qF5R;Z0ujE|d(Qb}I@dc>D@nu&Anf*+zOkep|mCVLj6n;RpBnA%wo{WhAY
zZTQCd*tjEhT{YYrneJz2jb$+bS2UoYit$^kAI_l)WfIn#L_G>0-~Ny~3#WA6KVPl>
z04Y65qfaV23fp~MwN*<LnJiH$I$9HhAd~%!;sTtUExL4>%_gQ>zXS=R5e5Sl4B^YW
zVmO+oU@e8JjW8M7lVYSLC0FB;tG~YUWgUF*7~?1+&0N2WHTpEb5vUMG0dtLLi3FxJ
zlcKO~mJ}l5MGJ6w_rzNQX_<h)vkyO{<i>IkZf**<%goz%4?JBitnMW^7AZSW*3Zi3
z#b6Y3Pe>x%-5$7|6E}y%<0WvptTc>X6C7<7QI_Sv|K#?6z(JMubRkX)s`dSK?u|@q
z&6<%;*-@KTMF~tXaG28Y=$(T#ET!19*@{K>1qWE9sj1Yx-U-HK+QA(^kW;b<H=5G2
zyFX||L+xPe`zT;D@tuKa(kII1BoG3%t;DI3mWdd?TnYE(RlsTD;)=|u%}q%iS>qR+
zNaNB0;g5e-`~E1(yR`Bz53iY`ypN^{W5XEtY-|SF3w1hKvZomIBCB&=x^H2BwZnN(
zjXR;<Nq5pIj{4vobxP^IF0CU%DJ4oS)|ES|i26f%{XzSm_xIkBt7|l&NBJ1(cVn7o
zqL|3Qe2+Q`>W7GafiO;AT+T=Fnd>x7RIvxBl|m9BCAeJ@hmW3{G4RuCBqr`}Z+Uk2
zjIwUDoQW*s=~Y$MypfBNo3b_gESBSeZ+_=Hq<Ln`8!unIq&>VN9cHF95hZY(W}e^N
zQqE_dUcV-_N<xrkk$9KYOgW1}3h?3mfp5Hc#ssX74{k}J)&V!`fNyROgk*?ejJ7$~
z9-9}fP|L;mGbttK>ogA{-&dvtw*%@O>MvNm_IbxKCIJQky9G1K4KdM!-lIFd|D(_U
zYyQ&@{~^z`Fys~9z|>k<&lg_5e#0l9e8%}wyr`33T&L|I<uCqouKxbL{z-qy7k+j9
zqN|I3g18Fr`WX}3uZDO`tt7GNyk>=?AKWTtZmi8?5Y@PPoqrnD_5tyNXWvueK>lqU
zDrP-0CE`38il~q<@6ApnG4DN40%=kchz(Q$disS#r+Wy56p7l*5=9lQmN#0f<qa?M
zOq|VgQi0PYLldgtugkc<Q;N*<OiYp4^S+}uWX%PDcp=jFt-k*SgDvfzc|kJXm&P%H
zG42$ONSf0@PD-lHVzLeg-W^vy3qRx4@{XtJ4C3*5qB}*Xf6ik2SM?lo5gASg{r91E
zs-+Z3QQYOgdB^?m*PVxH1$Q@cwx5ww*vH(sNJslb$qM%3NNTH=<Gw8OK*wSP-3%C5
zw{3j>;$TFIlc%zHN{{Z7nFgcq1NY@~5T<n+fY)mCG!b1KV+E>t1YSyED--eQLXgOZ
z-~PItt#u>JGf9;xC1#0C>4CTBC)V7J%NdQa>~J(;o(VMOU>06ChlN85%qcR3#)o$Y
zp3W<$wGLx#P;PE-`S61e`1Z|5<b3AzRG4mso8v<80#1T~6e4ye1V5V%A>BYM8~A29
zpiNkVi4c<`>E>DK`M}kDWTjR^^oqr8Ru?Nag(e^#5CH}hXb9HeXhe!NCwA@!OgyDd
zg=#>gMy25Zix<ef-smH2qk>)WgMWqVOBaC4i;p<JdCgs%Y!T2Sy>_S<a1iB+=y`Po
zgy4Q|S~It|iI0~@Q|E-)vv9rGiJ@;{#&7Ziwr`dSE+8;ln*Jb!YiP9>k<L>H(Ql}4
zVqN)WPWt<%>2G7$#DoyY+cvmh!d|SRc(C<82DZPq_iSolA6<;Gh6@awbAfu{kc6#R
zq}Dty!<(skS5rJsQXKURVNWNaD}+e|@ez$%?TC8{=A5Y|GtUz-Chl(+Vz9^_)-#H5
zm?zTBoqdk(+Z7sUt>!Jat!6Y`;l&3p`0&LGWSYpWA|bLI4jgU|EXO0N@aE-f(lYbn
z{+6&^Sl_)h6{HIdqZnX0O@WVI-1EWR4efH_az5F(SKU-P1eST`a6AyAc&dQ$L*{8B
zW7)zs4)%Dy*^)Hu^RqOI$vyMry#?x<l$U}1v+W!t@dl%>5Anb2YBp1`dGQbeF$kCS
z%*U6X@uK~%aVPOZ4&$1v=arYAzvSi1m$t__C-nU)ua~95FQ-`ilE44=6^mb8KYs~h
z(~ZGKlzo(G`;H>A^8%v;CJ}@}Oo3@x>|m&k-2K<8t&P5Rrj$sI+K8$>H+b<W!N##6
znH^G^ILwx8L5MbOsWwjS;)=cKw}Gf?FQ%$FHQ?i9-|u?69tV^u<XTWivZiU~`TaBQ
zo_)Z4b4Lu3QZo+^C*Hn!!<+WbjGGA_Ll!)tr<34i6ezW$Ypq+dR7bXDcq3}7qhZg*
zGGeRR$TdYnA)71xO|NE$X$yf*Z{DJ}6F*q~lo|tB>fkPR{+V$(JAFpR1jVv>+tstZ
z0|TgeIi{G1!6qxEDQQZB+Zdy9^|p0${+b|KiNS15Itto+TwF@A^b>JlwKYd^4e2_s
zCqgu?cMyvp-8SQLHy0wR^B;4`#`p9Q@|cWaXNq8vpxwqQBzEXSI~r%FxFMj8yQ2jG
zoWwi|`?>S9RiZTF>CD5c*DS|_oo`W?#q!Xb;EMLfgI?H59&4i)Gb!0#eD~o;gzb?<
zw?P*WzbO(EtrouZ(KF6fxwuc_yexd@_P6-m_+6g28)_|-47Fre+=u;shk&L?2ouAX
z()a5$O~%_TAR0+AvlUMsxk5{|HYhFtsuit;x?QNXk8`ci>{HFYbnMLcV%tFn%m3Za
z{f1CfsP14<N~RG|_XcdLt}ySrf;A8!jvSz0>H5+I;5WbfJwD$u<$PiicM->Zd#+BG
z{(|j#|I`UUXv!P|(kdJS|MJD#fg{?)!);ld<f>}_ZcT){lt&{bOEns)Hbba_*dVu$
zUt){{Ht&^n4Et-_k9>9%s26d)(2UU#CMOy9L%b7dYpsM922ZRPzyoG)dx2~<sTZ8;
z<E8GbFD5jZMa(30y^yAvI8UDS*ilAYg$Jc&BWU+7gR6nGcS82RW(a|IvD;dS9(JE%
zgjvjPWm?!;<#IVADYDE9F-}B+C2mZC=_0HRLT=pN-Eg=$5|#r~ZKzvTTrMXb^NH!O
zkP&K(q~n3Z{VnbCMAXJsZAKBA@a!=2!3WQoqi}h8<nr`nDlG|w6j@><rpV;tp;9w3
z1{QxPR3XM>#B2|!FST+AT`{X>{E7~)6?9I=bz!#S&3zoyKhLgumhKtk8>y;}gGBMj
zq!f877yjh&FZtc)|6LA;!}vX&OY-#e#OFWz84vHChyg6J^BThI&rd(8bs)0uA<TsL
zzx<2W-v-Wqb-j0qFh-K|JX;g$pRu8Aj%W#P=xm;qkw{DeGWq(cG_4dzu9|`bvyzDk
z4s)ay<2pA->pJyAg4qg%7@3xZ6v3L307cnqb=9X^t7vf1gzW*XM`5W633?o#@&B$G
z5rHKNHFOb*i5#B2c)>^C_%_c!`Z~+;mJo!zZJZuXd_FBq3Q;SUY$z6j7ZElpZ>5l1
z8)%}v$L7=x9edJ9G3l>|-65bg<5X(s95*X8iYljR<I@*E<8rG!OlPY2Sh0Cp(Djta
zMPjW##~}`nIqkeG`!KH_o&7v(UYv&LC~OR_XtvM4wTb~ifhcbGLUZ%(V19|~hJu3|
zZFfEhLYi!iolj>k{A?|hYLpTCUE)?lc6Kx%Oo^V&Emut|il=e(;>&zdL+HhWojJ`p
z{AR(@Y&wLHB(S83o8`#k`H3kd*4lX3HqsP%^V#RbG;#a<1-H`yNs#7=gJOP&Ted<*
z7jvj+&9dgJnI_*J7Vd72ARDn1QuH&cPgsHw!wf3CI4nF5FZkp7dmQFlj&jFQj%@2n
zBltS0Ltr(<1QCJ?TXmkH+6ceaN>mv+-f5aZlA&_q4MwS>0Hpu3Hfvtoj;B>q_19LA
z7!gea-4%=)P_HI?UA<iWv#OXSd2UK>kXxm+fQmV2gp>@4#iDY-M&+?j6~Dywr3=7s
z{lOp8E-Q~e{vmC<48vk$f?XGZ{_j^^m8i>_{lN}F5v~06ZsoJvM3Tv3XM7V3aaV8l
z?xx4gT;uprOc7`BE-cGDWazOMie6D0|HY6+ldkQ--3|#}y$5lbyb*Ir=F7Bg>w9y5
z5w~3^g)q-v6kd%rV4!thyj0zSCA#9PHS>~+5m^@2^+MiO^KD9TY!Kbg5OGTp_TSar
z`a9{jQh-MEL!>oB8jVY57B$w*+SoMGI8GBugsl`b7b=2I_B-Y<F}s2FwiV)(h*Kgh
zGg=zQ5ILXDTux`>5}hujX|njS5J`svEecCoY-6g8IZiyky@S?xdi#!bU6~ilM@}L6
zn5e(EO0AiIFwY6Cjdjc9yty5VJ<O#RT5ZF7t6S(8s=K?`_vKWbcVz#)Bfw)dgdN=j
zSATkeSZj4YP8w@7Sq>k6cKRFs=J6B$@PqG;&!|>I4c@+e$LF8F;<Dz!6Z-k<7dKM-
zfBqeIg|Cpmzw;&70Wqb-@p#xL5Y@@@!94$LYcMMmJ4hND#78@pzO$8z$C_H<7u~BY
z7~SKLlK(tAX9#;U)4-9nFdAw@G77a6wvq|z4KaI4h>BU2)Mj62idg5J1@~_7y|ikB
zDKSsNjIgAco9Fj@@bz!-wQv0vFFyQ;X*nPwl$<#p9;qdhw+k&_c=vdw84;2$$Pnkw
zl~O2P1gdkflAC55=h?NwBcswLFw<f&XMt#k;uM&sL}B8gt^8p5cYJ>H3Y`K5(s*&%
z+*Ds|Pwi8me*aqE>wBT;Xe532>;)r?!dj~f>)d=ldg6ug!G)9(DF%pvFnclYW5Ts%
zYFh~+#*IwLyiRMgdyuRMp5}?{0+QebTh7_jKn(Rv(Z>Bk7kd<@Xu0=AVAIG}%9v`E
zV&ahomU_XC4yI(Zf}SGNBiYW|%4u7<zq=z|R@QALh2Y;ak?Y2*rzein#6fz)Yg4vl
zL{w6W91jOOr>Dr2CR)i{E~`;*P^K8UStgz?<|!GcXr|xl&3D`P=Q}y_znT6QJk||u
z3%My5Hsi|Y;*^L=Yj(zU0i}o#i#-FSnAntUL-K}hc{J-HN=k_t0}CuJ&oDY~lQ2VR
z%RAqEf@GA9=)QxyICWoF0<BWEjkc|5&Ximz&Eo5}rmUrro7<uh#yT?#B^j<1!G)oS
z8LEGo>q{4a&;H<#5#UGbh0~9JWFrGlf2g7;#lO;TqQsH28^lKv7{8|0GnyXx?AtSv
zj!g69!U`L~v|!`nR;)uiJwEX8?j5&xHym$|p2FaVv|3j0G|y)9k*`s*Hf&QAMUsyc
z0O_~HyX9WD_Fk|T9?i$Aaf(b#W`j_RZ={7~p4f6GY8?)JUF|xi3QY-PT+@*`hmgl8
z#JY|t8gC7q*|wFMH(G(15;0BY>Ex;<AsSKjiezk?TpOKTrJ>aXAt+4KOwAX24Qe9-
z6&uTMm$RXHF(MjBvlV!oA3B$ikokbb$P_lt>l3Y6WK@b1s*(9{1M!W+>V`b17o6v9
zV{33siDR&0cS(_QIdM5%h(Y-9gXc{1#Fh(}^}=D9NipJk2+MS!<U))Sxf*G?YDKek
z%TW@i^My2-|6fDR&{aEE^ZH|V&ZIx=;yzz=JkY-f!r+GW;R)K~`~p*mhSFk-GKp{q
ziN{y3_{;mh=8yE7{yT+M1Ld;u^!6>Ezj{Mf_Vkjg2IOkC$8MnW#mms9udZL;6@oCO
zz%sc-LsP04dfZVM@ejJWA#B5F3T4|c1gE$#D>6kPEMD}eS!f)WnK@0&$tuLHR8#^*
zlokUT1Olc)4mP43zNXSZ9Q^~MMcf`q%M6&H>U=tLp^?=Mc!L$8$q)RXb^wc-b;Xh*
zcZVZ~n>&uT_bkU7mgT^_EQXrR3!=*HvwNOho};=kFYbevtgs9rp(@nkInvFL_&KTx
zMN%Y9-pB^CT~U9w2+@Xh!BoKkp%Mx6#JR%1z5I~BX#b9v`nmA{bM>NvEB682M{$ns
zkEl4m`t!c;_eNvt1)%z$ndl4V6iLyFKRZu^<KaLuSQY^v^C@{VfFiqsQ~Y!5=`lgf
zD7teFdoFJCT;J`gN!4KXV|{ut`Q)p5wlAn>?^+WwB@;6?Ray{_e(RQDMpfww<38o;
z2I6?L|8O~ToEHu$v27Rb@9%l{@W|ojfYw0G77mbWL*|(~Rgw!`YBL_=a+oP)W4XN}
z#YkRPVhn6sWh*Q5a-fxs?Q-Eplv<T(2_(7-w~=!p%`*aAPA9glQDQ@48fxlR%rQaR
z;s~=#DT47*OH*>$$m>QqpDF9gb~#z}o<YGz*^x+$$?lO@q2Fn3G#y|<scq1BYzpKk
zVduE|@+(`WZlG1|cV)9J%cXEB%2pfMQ*?@|jeev1(#8ZSj^U}IXb?0+6WRVkfZ*?b
zedz-5{@cIBix}8me$J0xy`sE*<KvNHs<WOAIuJnJC&Bnb3FJ~qbKv0?-u?b9_xH~L
zGaxo|TUTe{3$<7d;I?L#V;P5Xf9%ielZ{Iz<0hKTL4VlQEfySJ5Bs6o0lzK+_##Tz
z9M(0Jr-BC3WW0~g=a{FNTnd6k@WeE~cWhr(Wn4>TOpWY?W{iYRa@J0uHA49$7z$Vm
zEoW-89K?{k`*#G-Uf=#Jt2)m_x^lTg<vy|w9xc~52{(alt#Q7bxxc%&LcH38eREjQ
zOJm#2P*ZE=FeO@yl=I0xi*z8wiD`;fcs2FTTxg}RT`pWsC!XEk6TEN^z<NF-+Iar#
z4iRBpSDxNHk<!HR<~VY0YrPP}9`KxtAscSe+(#SJv~WJJ<N~*MN89wYjKgB6sAa5x
ztF_R_zhC%sUumZLzEXtxni*<w?}QPt;yDB!&L{r#>4*HU%l~+MHmb^cx$yAr9Z#pH
z;br)XA8+B87k^(}|4m$il7cb?1}C{jMuXN<XCy#Epn2rph)1;jClHvUar(MzfGSKe
za+qhMSA;@EiQ7h1C8^MYK#ZiL@m-gpKD9$-UJ^BDq$zVu+&_E9&HX*W&Ca)V<MHsw
zdOjMz$e(+0#i(F=p4Lhb%aT2$U~yT~Od!xoB|8r+M*F!S%!h@$XU~YzxV^u{?5AA6
zERm|-%vNjsf(Uc6`xt{!BVyX8J(fEgX`V8q+H3+s;8a)s^5Lialh^;0&-jeBZeyfA
z+9VMZ<6LvB1~1n>i<qJt=Hq-!PwsFY+};TG#$T9IU{20$jDZ-`#UKf7)y2_vzNHlW
z`zBho{h?G-VNcUEPMcN=#ZjCTOjuamuB2p}0;mhk)F(SZ#6${CMMyE&@6*N{jVCH<
zV@V0d8;r5BW~gc@AidV@-ld(IWITGaty18Utzf!cX5#HFXBVejww2@Y=)C3&=bV|>
z4RNP7bwn-3h^ELqFC33Y=T;i>EwLauvK)@2Al%>H&`$5{JgdSodoxu%qrO&(R<_H^
zWv#S0QFR@HOKEsaTcPBQysm89Vhy}np<Hq%UoNaqXSUOsZMzsfNbKAUF`>bCeC^;t
zp!HFDGb^25cXct1xJMvjHZjqkCs1-;SueH*w{<hu2zR*Hs?nY-jt`rxVdqLnGm?@g
zt6*o01Y}Ccl)&`>U*`JqK7e=kyqG3_`;Yzsk3aj2&;D|yK0Sdp5|%+}ms}MiK4z1>
zYYu5bX`KH4J=2GGw(y(oHJBCXfFUe-T`hrQPR!Fh#1pkPE*HyGoDa+3$90s=dpdg^
zen{^^kACfl%P%})`xslX4<s0pmbYw$pD&bYo`+$y^zZ8&ve8*G3Wkmj_EDrkzdEuj
zjprB)@dMc0BhL#Taph@9ItsbHbFfiMYrepU;D?r_kElz`jey!`pFP#1RTG<ZFTA{N
z2r!38m?I%gT-FQua;7xp^|}%dH>BeoVLpJ{vea0;7`IL8{O~{m&W}&%wvjI;$Dg7V
z^Xuho+>{u&ySrsM9<83j4RK2uPWgST+kEt61mHy8D(BNm43Wd(Xot5Wwq2z0#ZP|x
zw{~u)7kAR>2R?oYZZz7tBJ6Qi;0?n7lJzQ<s93J7|9RESnlC(^A9;Lw`X#=7UtPb6
zON8P1mZA`Yt%nq%Jx>a%-S^FkK&r)S?RF;sRxB*b!ZIcE>~u?b34wVrgJ9b#7g7>0
zSPT`LmYL=5j``+}`FJ4ux-Yr1ZkuzOta!VBcF*zlmL3mwxtzJb+z-D@LzV2n3gXQS
z*2JL}wirFPbSsU^dE?ESH*D@kkfxp_V*ASR?v80r!voDYW{DVQLoa=XJvfai+Gws@
zazvY8h#227#DJ<PB%6=o9@mxs;>Z7(|JUdLf^{=qZ?_NWh?cnO^XHoCae%jdMD9&)
zbHPn>Q|dAPHMkISeK4_^Xd$4?Q_p`@wske$S};y<Ot$y5sy5QAHlh?C_jlVIP#=Gr
z(RE6brF2v~<5I9F$(q|}GA(z$G9`;#t+lwljEPfPDQFbq10u!^b?>=(O02ni%oa2A
z9($F0pmrd|t<-8$9@jHdggGU`lql28d0W{o7mo8{!WU0z=_1WOz3Nt3%d%ME8_W|j
zgg}@U4rM0H3DL}QxZ!+y=f$%}p{9Vuz(gQ7N8-fL%+q<Zgd~9_%`DSI3ehcFEEr&0
zS36T{W4)|g&QGk5kF1X;E@u-zmRbi*BTgntA$c<(CZZI7Mn>9B|DV@2rUBYr0{Tg!
zz`CxSE{5vnY`{Ztb$xN=byG18qj){yMQ7`ug*d_(1F|g0JVTmjU-F#z(gmO_H>8^z
zKKlEA$Zx-S&E<6B)enAv)JklXZjy05yb^KcZ|{S|QXce}D1Ufo3##r9V(+T#SvF2Z
z-Zs|DW+ZDhg-#2N)9FN-CJwhZ{`^`8P7J%VzFdh^+WRkTe005X8h_BM$o=mXA0>FF
z6+@tw$}~+(Aq-UD>TZ4E?sf49Jr92T-VVu!AgD)n^>FDJBXRO#GK-I&O_ttK#Sjw`
zyGq96?fTf=QP_!zoQ<CnBcl^kbY$t$3-usu+sZsI^qfm|a<+Lb29_W!l34S`)5AO7
zy?(=VcTYGRZJ~=W&x`w<2@^_wcwl{eVo`IqpMwX<2g^IP=pS*^u8ofcHDABl8VSaI
zZnYAV8Ln!BoEzuUg>}u`-QSx<LR-lm1RrD?1)%oH(3l6?|5Wyd0RdGR-Jyg|I<U)3
zpB))=<gZhj#1GpT0^8+6E!(~hXymqWzMNRM)!wJbFW~@wb^V9DVhFsrJupoZhba=H
zHDEC%Un_><)v{R;p~f-F0k#s{yf+dgGR2vj<;d)jlciNpHAtj+c0=F5G(}4Bknxz1
z7+G#^INrZtx_L%A9POaO#=xaiDs{LYEXM=OeDLSAvaJ{6D!T%=cJ4Ibi`u3w9(5*O
z{FJ=$bUO3!bmH-JB2I}A?Y&PaaC3XZ?fna;d9mMT(1)XTGgF#rq5ERmSh!ogbo6{2
zu+q`^ZVGG-?vFqJ3IErB_h0gV|LjkAS1zN;Yt`T;t1UY&uLN}iV9X_<sb#9CDM8(l
zz3N%ywx7AVweF)kbqKz5&r8mvh(}0Tkyc8!OmSDY2H&f+H5Z>1Yp!EYFzGyFd)?D)
zM!qoyw$-Qtt(dx31hly|jM(_s^`GNxnWp5Ublb!7X1wJzId4*dW`IPgR*)@o7tOi-
zj&q%BwUK={Berczc*q;mCVVvA^5L^*%v0j+;}Z{;3z8yI5-B?15)DP}*}dJOY?>$M
zINJLPA&?X#2WkwQpPo3}-7_B-0(CS8c&(6<SjtRQ9-dB|)<SMp2;SY^ay%S3Of$=z
zyg0TA)8%v`Z<{IbFPY2JJI)VpS<f5WrVeZvKT?57JY`FF-`!e0a<z0HNSl~tSMvw8
zv$cr5?kcVtwnfBcTRE?G|68%2wUtb+*4%ZJ+Rll9`19_%6X#Vnx3r1G=|DOjY)!e@
z_OE(<c_ZjF(dLQx;zOSQz2D=*&pzYrt5>Ys#eAi%=Di|fj`i9HX#O)$2+I9O&p3a>
zUFVx|ug1tW0?oU@Qf!W&Gl#>?;9=x-167yRTNj~y6o>gI^-a$R`&~fdW4zEi#{PQ4
zRROR=X=WL-pX%d?4#9FJ&u5dnr^M7DP%kb!l&=F3y~ZTRFCO>HY+TgG6g{i4b4O+u
z8f;$3_07BYAKf!+AB`IarNhR~|F~8pk5}Uzj2btgcRw>$fYw~l=)Pyoa(jb+J~t^M
z%Y=l=>GEXVX_}~Nb_vOgM`;F^QwmcG9OlF_o2qV^=i#H+T4T#5=t(IJIzaU?1O7F!
z3o_LHZrcj$CG+@rA;idGIrunezn>xQn{ijp`5UMI-}PGUpJU%3>7S>+hSnM_wt*Vr
z#qQvc0#k}S)(da-G=8RvQd;HW2V8$mUtPcXYfg!8e)s_mB_=E6Vv0=j(Fzzb+oswI
zq86W4b0+7>yeD%ZghWaU^E`7nE=K5Y%DQDkapsx%aASUVWwT}{1SBMgk!d+_yn9AE
z+;O<MV_I(B6FAi_;7~hUr<8Vow_2Fe#Bn)}XS<I?JB%mb3IQX~_cdL%Obd~<6ds?>
zoSvQxF$==sFrx_5e8X})BFRR#K`a(D#+i^7rkKpG!qGk3V>(|k_@u!|?Lm6uANb+x
zPx!z6;J@Tge)?bW(|4bdOQQ^Ko?Y6@l&f>5ITtIq8d;jq(;&bEd{!Lp!nl4vg7zoX
z_rrh{E{HXZDKO9G{!q%slqRFC2=r76byL1R-nai}`%}+FAL0bJ^eWkqsGQxFN{kn1
z+|y{5T&nIrYgxWlTp7nrB#_sO2}DvPmo^&oG46N3K@g^3S@ETsC#CwiHVU`Cc2uD#
zZ_g{oX<@mnM1(sRLzyr{%%e6<X0_A#Z{pvN4B;KYK643ySR1t^R2y&JJn&&Uurc%Y
z^-I3?!3z$P42K0#8{3B%$te@dL<<(1w@ed<<HC)Pl8q}btm|e%4@X_=rEq?F;`Fpp
zwoJ*M3=(WaY*Yv_;YBnhuu_bNxn)Y;s4k#s8usasJCg(z=xeP{jc|WO6-sqWquSW2
z{e2%3tMLXv0!adby$2I)1)__FrsJ(m0B-LI(?r>>%*=m<>&q8_5@_>6ySeA~w|<*%
z{Lz1hx36CElgD?|nui=_3_F}-A%dHxDnesQ%HRB9MpJ-hW{X`>Oz_27O+{0-jmzas
zOp(K~j4ty0v?5^_iG=7?{NRFgQ^&gVU?hzE$l&VRK10x}9K(L#_pN7)mU|JXJ8JiI
zx|mwfmA^ynEPGMcEeS>q=-ylJU5-}TLl7Jdb#9oWbs<F3G=rEb&*F<JYT=^b=`8AK
zR4-e^m4+!Ms+LjMUYDY+?h^Qr%z!%krU!<)#+Xb@;hk`GBfm*t>Fx?OlS^gO#OsF#
zntw*IR;F!Zt_4YxNoBNAO9StWdr)@Y^lfYlG<Cz-G(HvUnoQ(UigLLWQ)#uv@#a99
z0<9HF$*!Jq8<#68Lca&iNk#7!f56r48UHHcMxEM*u&CAwVm1dO9Y$;fmO|o9J@H9>
zyWeZI1fAmRzO^qj1z%miu`8y?haY@k>PbhAVwySL+z^tXaH>0UR?6l)BrE1E9x@T9
z1Jkmw+}x1jL~F{D3uRjY^EInY(K?%ga-Iosf;2JB2Sfel1Bc@s%VJg`D5LSnj*tg$
zw3`|)S!)3;h7w7)EV*v-JO#s3C5+2v1yP#ecymjw3KEUx;bSv#(tIwJkPEfhSUmZZ
zL7SOVD-8*=Jzcf=8aH%1NVUQ**z3@cxD@64pZqod*<bwc`02xEJ~C?_IcLu;y$K)c
z{V-Jd_NMHAeZ)EZS$o4+_sLS{?x{OPTsr}X1XqYQ4s+som=JFSrs-(0mrnf%L@Or5
zipA{varf}Ed+pI+eqID&SCo2&u1$SxlB8Na!xpAFx$sAZnY0;f3q3s~r9J901R<$H
z6gGgSY`Km-Z;p`=l*@Jn0!XL4gn?RGf|c;{bRyTr*Y5B6;MsF-Zf-c8PL?sR&1gH`
zIN<9?)G*65PvhR>_qu;BfxzSGOndc;<#^!T1=3UDJNNe__t&gM?r)x>Y2s(=Bj1tl
zFiRo^<J2zG%skl{qN;2;Bc4)2Ghg7m6;4lQE|&{+Yo@ezp`J}w?RBPULSh`|;~mXa
z=Zbd2?|p4`t2guPGmrEA%(7`yH;HGrm62}IW@kshp53ZSD3y?Gj895Tw>QN5=PdVk
z9B=QOHyOyMFHRNzzq`Kt2pB9?Vm=@*KH`Jl{XKr?;SEo3Uh?_}e~pG6qU?_9@E#^b
zkjBk@B>&bewY(jXMHW?LButyJRQKLlDY<YsSj1OyRa7mO5m>g(I++y5ShJ^vG>;7c
z#6Ua0vm(%ttq=p6CPMJ<aZ8vHm~Pe4UhnDt=FLmWNk543^F%nGF9b1OPcH=Rd-j6Z
ztIUBrT^E@BKRQAg#8ic5djjYck1{6D%{61{+6t741|1V_GtcrUDslS=N9KkVf-4Hg
z?ujo{M+hW@ffxp{OyQn^TubYgv;Ml>w@523Z=|p=-5jZ_5RM0)9v`Udh1q-f){GAk
z)r$FA8!hb3ewyZ95AUSd{z_ZVc{BxFDVc3k@@BAmN`af3g_xk!Ol{`fIJ#24fJk@v
z{uh;C^7DVJTp@s!L6+BMqdik`g^`7tLZkpXf<WxkZwlp?3O~NOep6Qnft%xjk#}1o
zk`jl*<lJQ$8P{l~)<oVmO7ipsRa3Ff3+Zsne7GaT$<jM=Ca))2zF0#OjAI*<p?)#W
zG>K?1p+pGDv(v4mwv-IxYUR!o2E>1FErs=Rrk0HqBU4H~T2kLfdJ$;FnYQc&qb2Ad
zMr}DB4<`J|N4Ap9V#^ySBi`o`i5b0-lMtDgGjTaOx6$YlwKeLx8DCmV(M*U`5`X^o
zhy0Vj`KSDwH{a*9t!}?^HK7n6?K<s7I;hZT9QLNhq1+QUI`Y_yVtda%R-;>W7#Lw_
zpN^b&1m-z%SbCh>elB#=YTKObdvo2!=(k(!SWKklH`ij;WD>+|z1GbqTb3h?^8#ap
zbu}?zcN1u>AYw&EkGw>?(j0GEEvcmB!p$<7NqBCoxq41}BL(9SE{pB|r>zpS`{EW_
zvv7tG#M40nYm2<StTb(W4dvP0JxBM#eOfOZj>oI>#spBbz1Y+0!nM{+B!W0k8*cCJ
zc=`G*hx5jVAAOBE1zJ2J(Xy}8w2%%5PUn@7W^RtRNVEb_bRXec!E7qrpdQ~=jIW-H
zQnJ0bqNcp<HZ#*SnWILGhR(<6XN!%AYc^EB3A?~0L?7Fl;A^~ZN8~#_uh}MoM3Ckv
zcx%=KiJ0II?3|w_;xscw19fCd+<fqY+Yeqa-M--V_LgN%%ps6e*|fW?{A$;iE&vmD
zgH{BY7v_(?&NqJl4|wy*$2`1#P33d@Lfs3Ags`FRU#A2L`t8UMp1q}P7pCA2QjS<v
zx8$H1aX)YNP*{g#sShO^iWMc$wmjt8<H7eur#|`<%uErrNFXg0jorI2GA6-t#S3IF
z{<^_tH_aSljHVoq2To5<HbTjH*r#0A6uWofz~A?o;ukJ)>8@1qj;6X`Br?Vy5{#QP
z3n4pF)+(*oy_XO@r?ex1cu^w0;g!+Z7~1Kh6?%89#yd1PmYtty_sm9M)zAdruxn{(
zE~pk{isn&<upAa*5Z2as9wP1Of&1HqWl8RH7QJH^=P~xpzu!O^9ji9Cm#AeFfVmV(
z0o8<{+}<8I-pnYOT36>b3eiU=up_$HAA$P1y)F)aLH#%WAV{$Ip4texxOGJvhQG8%
zn#^KAT;$Zhy=g!RgJk^G^&ipYMT&@T_yJ<F%xa0rKK`nf?Md{n#e@>l5dwjjn2vYE
z<<7rPN$W<SlCiP>G))%AmJ-u^K*BU~Kv8dSE*Ww4W{)zgu8{2LPb)o`-gtakH#Au@
zVnw6zd~4eiF^s#FL5nxWi5RVbps6cy8TK)#jci+9$#tW(?1@B{azqdlt{iR$UlFH&
zl*@@rQPwJ)a^r1X`A2X52|s-Ln8)?x%3kfa-#-WO!qvuSJ)H$G@rp2tLV%IN)X_ix
zv)%7g9MO3nk?MOp7`OR&SUAomOp6}z)vEIg-QLCi%$zBpPbVVT-72o*>*jej9@u#W
zbWCmNC|s~VtHsd5o~qT0NGnVl;$Cq*jG)t6Y?E?hD)szz8pjlnYHHItz%nHhKP3Ad
zYpz4EWO0QVZC#XO_c({h!@8jl59Dp*u*}T!%+2vgn$kd8wHY8l2*!P@29Rj;^Q$(e
z3)nh49u8=E#nZz(mO1d@i?1^;NAm_2EDfY&p5DG8gl}<kyctcmsx{qJ4MA<Sp&r|2
z_tr+>R&w@aEO%&76X~V6a|Z_xwV&(09%?lcck%kc)}M`yRm2R{{j=)g5WuEe%VM8T
zNDvC>#ca^jQgDJq(ljw2j~s6A2`Mto3%7UoJp1S)Zl6Er=Jt-mVKI(*t!Umj)%8nl
zZC_pm?C-P1Knx2h3WpCr<lDdh2fY34GxEcm@wG}r)qS31FEFMkoIg^w?LrJ5b}ljs
zMD+qgvvL1aD~E$YzWC^_8*xr4IrQG3K$GfSnmX^yd0oLh{oKHn);xFB#0o+C2I7wT
zeN*v&?>s<#Z<8{_a#?1!b>nn8b8~xSmMcZBqjvi}*$WEYe-QnL4(6a3I_C|Ce`81N
z{9cAY3KLPF6w8wwoxR$Wx%H89M*+Q%6QXC8HhRwRcuQ9ZVb31yqb4FFM_q#Kb_+h@
z4kFO7Lu$L23b7VWrzfJ1w9}Nhy}KnPe@#CAwBIY^bFzol^HYOhf}lpu+bCXZtseeu
zpUv@j;O_1g2}*I~!uCi{L+KU>(u=ztZTMxEecfIbs*ipNLfB>R_Nny7r`hY*<~%8X
zpJNT^^=C0w(=WNcsu2G}UfK*f7z~OS&2!WJGVM7k?;*z$ECV(~^RVkefS4xB26oO{
ztPzcY7UL*Bq73ICKe8H9H}dADmqBRfM>_wnD+_Iu>vj{Hail|2g4hT%PKh#2s6@A&
z=n+=E2~`9`{}f?LGcr|=uWK|cX{ou8M4OFfix+~mZf3~t#PtvZIiG2{ay%YMli~e2
z7oJXymya7ic{=lNPJhV{>R<D^eLgZ(?U}H3-V45N{J-<U9l=*upn8#ImR3TK#O!-z
z({7?0N5j`qUG3rz7mxHt^RUc>V5s4eOaND1;Yzos=_B85vq(R;1*>jIJ<YSnS2iRD
zw$1iHoGyZ|uUcJ<QJUNH$a`l@t;Vfv+6=LZk!{O1HY}EAC2{vOwPxM+y(vbHlOQy<
zV5;ClvW6zPZPpYc=Pk2k0~l-{H&wWZpwW`5)+*G~nRUBxJVy>Oaab0fCF80AbYTO!
zqst)nzP&LqBWv?J{P2ShdHe1ij}Ipv-#+n4&a{tS@a&@xm=B9ls~#SyIr~28lqf&j
zN71g1=H_C4x6RRB-1!1V6A?l%{$iSjIe6@KhrfrEjPKihH?7$AajpAtf_+WrSSrEz
zfWBdb=|D_3wyT#4>+0%t<7{dJiGjoM$g>YVV7b3X=9$}@8}9F(bN~DWclXaY+&b-}
zR@$~vTE0p!`_-;5T>utU0+kTltWpH-@0q{zUGDzyHR~sDKt8n_)izpC)(|*rq$DBa
zLTnp<diOJgTZ7o66)9GQfw<3BD}`3A&^vneYx5$gwqhN(D~Gy*Yp8(L3q7xwRIQk>
zVytro?9mUk(q%zi1fUKNi3hfKGOdb5@&MGJpoO4~X-Zr+ShvdI#v_ruP}70-$>0U_
z7xr_UGe-%v8~9o3x)&GrcTKCWHUCLd@&dO)F6O_bHr_yNbJuWXU~CL#>l!s6T8L^z
zPWwWU+cOr&VTivFoBBX=!mf*b0``(a6p+UA<IKC4pR=A<hrt`qkBiCOi9}TjjW+Jj
zZnWEiQpJkl;J>fd?AecgU{r-_uk$bq_xB6a91S@xmFgqbt~{jQHvg>OLuCC;zgKQu
zjqnE#RK<%<9o}v|TCVQ{DmLb(neZeuh`u>hg-d`IDpM#(=T3^eXDRc3yYSWZAL>$7
z@_MoL5iQ%-wud^<y@n;y1TO{@8~?{J0gkkLWvAetc#b0BNl0SRa@q<skXHN<tw>p6
zJ=4lY5VO}G;q$fU3#*G*mAqxQhcb>`3PPI`^WjKJi=i`H@y3L4;90>DQ1xQd5gTfy
zwrs)lh&9s9Y+{6txNK0&6gZ@rB_={MXN2Zt{7ako<Za<EUq15Z=RfDK+V?r>(-5$9
z4z6~;P|W%RZ&*W&`?J~+KIh?;VQB2n9($43eP4kQg=PisZrjs5;%n>+A-Gp69Fpxj
z>II7e$^Bio^+KzaX_`r9Tno4uXP!pPTW`c`u&{tuE$zenOLcFyfz&zBwOZ0gE~qv(
zx4mjjS<hy7rrpC-luj24^k!b&CJUrCmMI{)x{Zx}dP$VKB@u%<Ad}3TxY!g)RLR-U
zwdA1#YgH~=<DI}10?(F}XD?pRvQ1ai->>b{)eFBaVpJ%F;3|48m3dkCoj>{`e){8|
z@bKytr?v3;>nCDbh{5uG%jII=kqe|$4Z#lLhWkP6{K3xuoXf}@4#5o7I{;%sQDK_m
z$Y@T%8gK-+wh7Fw(xyb71NqX>x{)oP(~8EJ>}U8it=lU_SN9@9y#D}rMwLL_E+pPi
zAJ1$}?Qm=jjWAC$&yF`Nx3?T_?@aZ7cgNk`Eyv?*lSEPSb|I`QT%Kr`r!Td)eR&ax
z0<QWJ8kI(wCb+-j?mOS*{^m2_zw%fu>qf2l6+#Q77<l@IpdZdmQ*u8jta%Zvb8uL<
zBYV?i2_d<Z0Xx@f>Zj&b_Sc4<L9P&+Q8DjC-yMRp4bKW6=LZf+ybC?L#8wz6VDN2F
zv4cBe56@*&MCw4(I%iJ@gtz}*btw9a_21Pu`iMF3o;#sj#lAWGuOdjAjDW3Tn{!jt
z^GZDdbjbb3E$B#P@Nu>CZTi;;?itn#O(*)=J+p$Wxj~Yl0~JY;!{I=cgrr0PUfkR}
zDrlYhCh(XWjlco-^y+A*c67`sR<ak#dlBZad+lfjOjG3MX5p|{(0Y$tvckXHP+Y&4
zUexa&O&52_)$Q(IJ5XHhh@p=E7W!f8Ki84a+8PukCJ3sW9^cycgB7+!rjU$R^-Gq*
zSJ!XgQrmlMABZIF=)YTtIL&0NNA~@~s1j)a_GYtU2dj-AGeOk3WqS^*wo1t>PzKs2
zUbHm5R(R@&G_(<681azCrG*`eVF(9QEy==+Wc&Wq3qXslv;EJb!c-|*tswOAVn<C(
zk#9f0_#|PDfn|18qZUrv#7|DjPu@K6XK(+4Ka+pODV(tKw~q@2QG4x8)X|j=;B`io
zjAv&5czi6}g<U?<lv<sCDG;JlNPfYdue%D^F-P2UjDYd)mSr9QLWpssav0S{hwW20
zRu(UUO3qv^8@1G7`%|3P7lN3|x|yw3N^wwY#(Nz56JtFY;%jJcN@5MKSm?l%%o=9P
zdAu%sT}J;|%WV9@6wQVv*T%znBha`%&OBYR-Ope-%l3QmJZB|MlNI8gB62F`MX13j
z0^PTk-bbs&n)bt3wAaW!jq^q*O^NS*@4I~Fk*6O#d(PqZVB)x94aA&aGOKl(T7s1{
zSC0#{J~4E}w_bg&Y1%EttZxV`hd7$JARSTm0xufZH>E@hU1w0Kxspq^^^qns#Eza@
z91XzIw2zPHn>#ezQ@nmCPr(eWeN5d{ZKT^8lL*I@xH&9LDKQI#R*_<Bh*~J8C(8Lu
zK3!PfJ^re@8t`i<0DEy{1(&r_tu-V@KK%OEm_PVkTbw9a%u=DLkXvP|8Ij5-zt`Z!
z!C=%V12O4j-&Wn{%JWs{c_vNfH`n3(=Hptc@OS6*_d2VAT3z$;^u5i%l)o~Fy$!c2
z7`)#v9Lr;L;+sBVlJBN$xe?+ovvTWI)&BE3+W7vD-}x$x;_>Qhu8T(5A1t#PNa$=S
z0=1aum6k&2yoF&kVbHhs-{+f#RQK-c*^%@E&L8exc(rb)V+DBi@y4{wgn2>Z%;miC
z{Qk(=OcDL;*#{IgtA=g8@H%JSZ5z*1Vu`{O{BFAOr`4JE{$t$HRO#h#M*sz;*_`L+
z*=%xJtM=e_tB0_Ako9}qJJW0N$B_B=|9=0@@1IrQow<rqvV~~Ai+z0(P)A4w_|M<|
zbN;t#yq;Jgv5>+YDa^o^>+HU|e%)6^Lb9ntGa*@W1$uKYKkyNd(*e|rwcvY-jleP?
z`!nMO#*_r&Jh+3l*KKIA-A52atyxmH7O{QO)U%z+(4X5Fl9>WGi>NdNkdX~+EDkHS
zYHi-@*}>Sys3DucB#4@6bqIuD1%T-y)Q}hl33F>`D@1KfDKbR|dL;2?OZ@chg&#k?
z<j?iL=4X7&rkn95RmU`;7tdV)0gmEfDGuEXU!C7Q_HM5R`avI2pVag{A?(GZ|2}b)
zu@~{$_Li#`VaH`=cAq^f@bL5yW!p?xP)aqv-;{v<dD&G;@mNSJKD1d%g>4l=sf}E1
z&ri;ERxdF79v8&yOw95sm~Uhkx`gORY`4Rbk?G#;U(^dxFf=p9L_Z2+^d`6}F$mA@
zp3zz%1wqf77u~_)#>6?!)pCzPM<ErX@zqi?&b8EL8OdHSwbq=oY6p~QSZtgx7&kAZ
z#4!cpJoBCJ{ub+H<>q)~TFg`N{LT_Jj5O%ajzlDhr`Pnxs28l_Un{muM2uS-t<dg`
z%`_WD!{Q1ZsjNUVF0RC6;*fdrxWWL^=|&d%FinY4CL~N~3|8a@Gra}~bD}n}vF*08
ztryPgN-o(vJEsMi5<wfGDM=t|Bea5E&Rn*YEe1j>)Ur|63w7Hlx$^G(8%6-Z9~iN5
zonnxE)yCoej>GhIK!}Hhx=e&znW)S;BUI|#c>MjuvpFKsMuoj==<vPzqEpO()a2=g
zHudxh@Ufm3Tiq*eSC2t27<eB;!^*cE!9g3^%mXNB^i9Z%q5cc3JsU+r|M`emn6m~d
zx?3LnvU}HGLr0%}!Mz=vG!eIca86ABow|=P?agE33c=myC)*5Nsn<$16<O-9K^QzZ
zFQ5s6#AYL$=E^GRK3d^DF-!;^d*vg|G}(rjrkNTd=d7GB8{ypp#KiI0bM9|%_}QB`
ztThvs16?ioR5G<HcPdN{P}<1d?nC$V(qtrF@iBI79qJdRDX}awX`WEreZ7{k`S%XJ
ze~_*9`+xn}?)T<By!O3fzTbrFPqJIrWEAtHv@suU%4<TQD6r)dpomsd+n8ixneQz+
z&_DP0KL4+--`Evn1W$Mn!Hsy0(`=W0Mzz_fP9zX8WK7iv?_C(xY8azS!bFNQAtbjW
zG5(UN2;Fd<D@ALJ_B*8)n?i_<*rLClUZhw-(0uY@T<tDZbR<0JC|5*`bJne8f<%ih
zbAQx1MpBw^Ii3{<wX5$v(k%*evuRM(z^iTM$FEm@c>0_lwZG(V<lhd<sqSYw);KZ*
zd&4&AYFZQBhHCKbN^w3QMiWVPUrHB3_(!rEPWPwV?_0Yn{QBO}CoKZ=Y!j&|b|keD
z#5i~2g-mNk^_iy03I||H!QB7~LGZ#5H*v<&PB8R#UCp9unUez#=5v~&C!sWDU01fv
z&}VICKSMQ_1W+#7Tp6ZlIn}vjM55dFbRIY~E&GH=oZII7#d)5Y4@XdCy;P1%;!+B^
z1fm!$QQQ$NOwqgpOF@$Dwc3<9d913~emx3u-;+@_n=XGn);#&yQ!AP?<;k+y=fi<z
zo@v@F#p&t9)0;Qk%#jc#{=EGa{~-M#H}Td!4-pjel?^FCi)f3)vd5&w7@3!3^c-)5
z(iDdpx>hS7OD;rdEX%>p+YpKK%ye8>Zja<GgKYkcTk%?JV_jD}c4I|DG~1hgwMt#j
z<jczX_{inyj3^urM{cHt!*U=j6Sv138)dg<PAEm7WHXKS^d5XV(11(X_<a43p#TIT
z;`Rc~Ob}xr&Ijga9{{l2-*fY=54il~XT+@#>qea6$KNeX&+grq%Db?Rs5n`aTA((M
zK{HcPD>M!DYpo1ezk8^S4v`%_yLP_s@kd}yf<-#H@145RNA36c&OFAgR+~Cf)W8)G
zbe>9cB(N$}m4PW~N2&U&xI+H``Z)26e|}vI*jToWqGS}F_5;|XjBJb|Vy{0q=Ruqt
z+aJV`%)jcrM2mjEVjyZ7D%<zp!~0m6fe)tO?b)J+EF^MWx#YrJGJ^8<-D@~KpfOUm
z%(N^hhW^$Nxm-@fl9}d6bSeNr2-V27ee644>v-MPJ*Rmh%|;C<tx{?k`NJYVkNyop
zxX54#?Dv~56n)pDGH))pA|-p{AtE$s6fC?wj<2_M9TP}1>jvXoo=CBo*H?^ub^S+l
z34~;yXVqd>4o!>~)v6&eSYcAiW{5}C(V&Eon5G#pn-r;r&Zo2>lE!}5)VX0TQ|e|T
z-<Zs}-Pd_H#%<y%P>i!@X6$Z7HFA0lRT4*{&EBKus(1C__5KK*D_x6^&iXzXBJ&iH
zXhIHi7_j%6i#`7-MskHyhR@!t{P^L-|6~2<eA0f%+Meiw4VxO2U3nWUAtH&T8kM|_
zeZJc<^v1Log(*hV{o6VR^1ARdKfU0^dK>q_-l8LD^uJ^O9a!>9&Kp~RNoElB_(q>N
zxKX<)n=J}X1m-CXu3w6YLLnDd1yAO}P-|l?ZWDv;Ii*x$g2OTseN0(g*wd;RNUK8i
zeJaM$8IzlFy6t^Lwqi~UeGe|B^CI1t8gtFC{AJ^_R&7wN5k4FboY##ca7f+p!D869
zVid2aGCUbe;R<N+H78|{#`LL7LmjDe>IBZ?_QE!3<L%oEr`Hcmha<~8qiWVQ>*dVd
z?7o&YqYD4@;h*zA;vaHzbK|@{L-$$`_tZ0&2laK3QtU^x0Yw3)FhqJZVl9Q*DxoEd
zCrpt*LbOryjoI~-2H|Yu{F2<^A^P~gnwPDEGI*V^o=!YHojE_9xSUQz;P%-)H^&3F
zha-pOK%6J0Y36u5TGQaoO{s+~XEF_O@T*FoP&sXtbNLP1z>EX5<7jS6KngHLS~8{X
z^6Z`ufB*OR>DxzIzToacY##;EaiUXEdaycR>MRdV7V3awdjS|zGQ_RgCfuR^sDQ)E
zGE8{A(;bD2j|9Ak3u2>Yb*mBM=-5{Vuux`Ym*?$wV4!z0{vSqg6(feXj_x`7)W$ty
zY`*rB-Y=_u!OeCRvLlhY^E3AMq+{XjNT+RnF(rZoD&j`il~ydeEIE>=;OJg2mZH1Q
zTL)|wCS5^&br;n*LGNwYDbYgYvToKjQVA+UND?J*oT8(1g2u>(GRKM2`b5;qVeSQ`
z4SPd~H1Q%`qSsUAg(i-8ra28G&)TXRG#dw|A0XoQ)5i0<-#y!}24W!Yx7gP={SfiZ
zxGSc+AD4CJ{hl;hyK<OkXNpE3iXVVlNa2FYN|hA}``CYsy}m*Le{+`zGFg$8%L)j!
zW>jqq93&D#0&Uc0q9nAM>ND6gqq-2&gsP#@Nf%N`sALq|=)sVqQqHt`!5Y<QZcm+_
zAhwEnqar??Aeu$WFfsP$*@{!QQW6Nwqa(E!UrxDU-^;BYt<LliP>Yjmp)pPNGqswa
z(feXzR8%9Ur_5iy+4%m$Pxy=culZDeV)T>#fO=zPlN&=zX~z53Dzv;=VL462`G7<}
zqPh@7#v30=LDfyTdogE7r(A8OCXO-tnqU*YA<D3Qw6~G9WP(_XX7hrrq1+r7P($Ea
zvjS8C#%%*=hKh9&Nl_bdrexlWirJ-^sWjHm5^N;w5nt8mF?GuW;p?_6o#zgQo+*&C
zS<s{?q#$g$S=^hLD9MM)3}TTsU#oGT)T);D5@I6F$@b~mIHu?YniZ8P2BsivRajer
zU?{j~_4Q>)Y^&MUO<%A0M9I)0Bu#YdozW18l}&HYXP&l9Je`4^HyuOdCYg;&#W<=z
z<`aI82abAVT4qzjiY2NLBP_k>?}9&Exozlm9kh*IC?aBUep8y5QW!})q0Q83VL2S2
z7NRI?RPq+E?LN37U1<^^bT46huyWog8&C^N2psS4cyW8+`TZ@+%`MY3TO$^e`w$zd
zT1w%vZDgRu$aHfLArMhoxe!|!Nb0Y8{aO&fehpt96mJRLDIey$TfXr}f1ma31Ap_+
z{+yU*PPdtte>ig|f$C8#W-M&Jblrc-YzO?97E^+4IeVuIxm1>}zKngt^@G?Kng7Rr
z%InppCxQkWk4SSBUpI-Yl{$VeFh)N1^WHm_y*LPBoQ@#7C(<uEzQ4%#`&)~@@m_cz
zmHI;QH^R?#FAxE%VnY}<B^m@R*<o+vvm}|!PC<+YVW^gkBYH@^$=!FvalDU?7Kn$j
zi;Z-)T-er)T#bVlk|PUcBLIgqamkhJ7Bdp<Gr4F(*9vjr*dk$wZ+bqlZS>8HJ2P$#
zEIphYy^2<9<Mj;to4x?vujMhN-~QgXT0ZvIIy&n9XX?na__)JH%C^}D>D(Um*X%@f
z3FegstmiY2?_QHz;PLW~%XazOkz(FQlE1qCo4xFijbuMiItMW4MpKW5^ARp;HY%-d
z)K(FRCgA9c&DXJsHq0bnwP$KKYR%;BOl!qBS5j>p*dJ*x@J5uIc45}u=<Y?J@d1V4
zh?qB{UqG<?{~zmIujn%kZ<Or!#Wax|NottiXdu%h^7_2;CvW~O|Lo15^0|I)Xicj`
zb-}|h#}!lim(2`zYeiyUnr18HN=CN}vo_}A?LKu7<7CHZN%H%1DEricZoT8JmcNhW
z_9T|i+cUTw=~pX+OR>)mNuaQ<D@!sfm|82#Jkv!lL=#=)Z8Kp_2O{j%IM2+ZEY)Z?
zL1e_!sY1yXMnHnos^lOn?`1Sw(FA7nt*T6J+;8F^+xzA;qs_8u6&p!Le0&*`C%35O
zqaz-nIvYPR)G-aKwX$ATTFImsP_+aQg6++7RE{K0QCPDXU0XvEjAEdBfz^v@zsdd_
znQ3@OXqyBg_F17ySyvw4z6BU!sFWhlo<FxjdN~jsV7k;ZfBN)&^l9S7ix+(TgJ+~@
zC$fG13{jPI*A8#kbIwS~j<`2sG$b;JaeC*&fs~RLmDXIPDYyrpGNr__Ol<4w1$q}=
z1)3w(8sb3LU)&rxAhg3m3W2+2;raa?H+MHohXrD=G?>zc%~4j1pgZN277{7Vq;#Mi
z7CWQPg2eN#P52eBFOL9z;a{415?~M{MZc*Kmxa6Ue4A(A`<%naFS&gBl6N;5T@r8J
zorp1TvmDS;Im~82ssS2xs6|6e)={TOIbST9!nqtH7qkCv;xB%kc-b~~JMkH?AyV2Z
zQVUH@sn_d-HV{PUSn`kvs3C_PIqIH4CSuU>K)~K*kMUxEOs;FYD|Ar5xM5%Y^SUe;
z%Bt&PxBq^C1`Q_*$6&mpWCq%`3v9#-L?_O61TaR0*+{b8e~S+JtM`}vZHNbe`%S-G
zE*5gGmdX*UxzCqYDLHdoj>HCGo>@zwteL!?DO=_s%2buvQA1QwC%+S9yzb@&cA6%p
zdGXCdch4?%WW-*UUgah8-iEs)p#3iP&-v$9{y={5=TQ26^Hp{=64qe%%l10O{0(i5
zu50G?=kIumGcO<BaJrn|EAYN5_I|UMh?u(1M{DM1s{T*gbHoYv1L5+$$cj5qEmqX^
z^a@q9W~vx^-J9Ins(Y7p!{dT+{(7@HRvRPxThvBY{UigR7mXc3?tD9j2&xOB3~6eP
z6nQhJ`vjyn_Dw5gGql1+z&5HbRru?-Z~3R6|2hBf^<S~I3jt+Lfz*^z>^Tf#Rx7om
zL)~Jg6>7CHVCUlI+{h0Ps9L&`p`YD9RuO582(CV{=<|qXE&}`CLC)FUci(3nq3gw;
z_&rF`ES##RVDyQ=JR87Z9)r;wDbz;2W=kR_9u*ljEwwAhy@6<*UgYp?p9JMn*c!0R
z#Af%_))2nG?zE<yt#HYuxLL6dG>_opb+Jf39ec9v>nWM>a!M{}h`}c}7GYPaWm(Tt
zaz9d~R+ta7S;9m^0|SlYl&ryxt`u&aV;vAbXIj&-25W6ZRkup=k*tCVDvX1-<<-E4
z)(9?E32A1UXGh7c4mcbS#FTgw-*9?dk;j!7A9;Sekou$x<Gsch#`>?m{v$Ra0&!=E
zHv9y^$Mv?5-1ON@(C20H`<IyJ#JtT8Oxx!hda!_>(a{@`Z+`HMd^A)qrO07kINsb4
z=NSz`)k>>a!j1=8WKk~L#$(P9l*7D`+*3@|ieN()f5S%5*I<wpK=@{5FE;iPHAd3&
z=RE)I@9?$nz2f7?3qStDC#HjiN*{0TxI8`a=H+WrfEO?BSf;=<Ey!g4e<?;&Qz4XA
zSf&G|6jw8uoV#7Wu&WAb=W97?)0M3LIzw`WX7x9t_BHBocIzHf7`RE7*ZW2-18p<J
zF9wXi)^-H&y0{y!vR4pSe}6r)?tj<EKjJ7rFBipM?)U)uzY14}!T$Qx-#`#E#+<O<
zRf`qU)P@k*e?@Ss2s;>RZG_P1N^AS^_9k|<vK4FV<-+-NLcnt(EfHmz4z%@bA>;_h
zyL;|G{2E&>oF3lt;~zh;)=Jh&IKd);n-mDaBCvYxTuXKymXYj_$D`8&4E-y$y5AQ*
znf4;yZ~Ao!{hp%N|M5kHcHHW#C3gKmT2a|MUmM>VTIAv9&S?^f2$4%p{Ij3JAKky;
zTi^U1|MG`F<n6<WZ7c61G}jHm&%L6ruKy5M2zzW*|7><Kk&eS-@KZfww13_qr9ma=
z1sI~fJ~m3th^HUPAV=AOS!-s9B;>l~Z}hzK=4hSxI_#-AR=g?><&A>miz1_5s15-}
z3h#|byZLk3R#OCv4gQ+KU%h+D|Nhhef={<sph}9SJe}r5sEsOaq#Yxr3hP$PyHgs+
zLqw~o`*Um5hN&V$=CWqC%ZX`T2C703n%ZmZ6QJs&y?y_&;@Gyj&Y!#z?+hP6hD}Ow
z^lqNbV7azx-g}36*xjrb<NDQB-2l2<_E_<pVsyTq4$;Tdi&t&KM7S3hF$kpS2GB}v
z)yBBR8ihb(O2PSN1;HZBYM+wTHqb4^3fbTZEH)bNwm)ffwV|IkUI@n!UCrtR%i%!I
z+W;|2EyOfY&vo!YLlE;ljM01#t5Qp*mco=KFD!e*;n~x^_O&)kC6#n_W*P<JvaW17
zlct$8&6HX=-dGCD90h3PQtdNrjT|x`r@!HM&PSduiNh3-1k%h*z4r?i_@Sd94I#OD
zK6(Y@ufO&rD?_%Mc4}Bb{ke@P*~uZDUSZGB6fM1H?%Ck2k(`72;^s)5Ec;ucjr&zF
zfoQ9gys@q;m-A{xxJ0T1a#O0B`njFOlGgG>Ei32KTb|y%=KS=V7J>AeBsQC;8vsj0
zG>(Ey)RIDOUVO;&?|z3LJ$*{~#sfEZcZ6(2$5I=I!x5oTi}JXhIW7~5kt{V$$h4ri
zE4m14Del1|j{N=NBXVqX;2my1z5SSX|2T_BOAhCXXWzQwuba2BqdbO8d5VYz%2ud(
zBS<0z$gX4QFKa)<-xsCqT)@2|?C)eppssimboJc60e}^UVlzp4btt{#RzWn{>utFI
zlz4!69-Dn%)KxGJoU49Aw63q#Mv69eQu;@()=7`<wpr+Q*_fko?dAX(<?-nWXh@vc
z){W(GAkB&8_Ky3zTO<Yk`Uiha+g8rC@pfGa@xauS(9G*83Y0BVb44X^b9>9O+z>E+
zN2z%nTKgzKt`2Q$Qo6}B?9GAgN=~3hQ1@~ii7;a7FT@Ia8;|WDvW#Z|T2-1fTBz1(
zOYj2)wkrJK4gBGUGw;gGn}-XpU%e)m9(dnZfv*R>ev#LI|K38U7wcdAyI;;F{9>W|
z%gW53FNptc-}m4COF|$LAqFH)z6aRqQ5UKia^t+*V2Z`)RuUn4!{k;S-ITZ+eQW2E
z#)OFHkamj_imjWzHT1oN&M8$tWSjq~8A|t}P>G|+v#;+8=H}~E$BxtEGfjXY?pxTS
zIF(vgv_&M%AQK-yea8Rs&HtIV?F~X3=>NRTXx+FwPHd+QNr8Eeh(vNRdm43<<8|F^
ze~O8x^NJb48mBJj4FsmNIKm#r8g6R*u;50}C<AeqUZ~?n@@^cgwWl1Ja@6d2WbC?A
zkE~NmZkgwqloIQ@j>0ZR<59O}MUaF*Ou-9fx4~*kOvxiEt4E<(v97fm$IN{LbFP$T
z1!!rFs!;OA6a%3NxmI#%WABG9WN^`;Dp7)A5)u*Ld-}S?r%^G=Zt>-LesM~aoKa~U
zmV=>-P6>+1nis?nur*hrxRsmOxj3bXrb4ZSb-NIQd-0BD8yhEMzKPMBpHiwBFRK9t
z#iwK`CFT@~G;VHhI4*(X9EiatPeB7()%Q{5ucyD_(C)dbzsorrYJCWbIk)*~U41X~
z=TaQ06R;6?M^H68(9_IYF$97nr=$6_$*E-_MUr@misH?o7lHvGjl`|vVWEadX;#2)
z+l5juGznB1S`68}<jsoq#3&RcRJQXI=cflA-n}+tw^d5MQ1XR*K7Xlw?MoMc9ZLc1
z9v6^mUTuLmY*U0l&_tMKmajc0{qffc-+DmOgw)0%D9dqX6SKuA+X}TY-_BGqbH(X+
zAlE|OJcwL`R<g~}rK^HFa&|oe=!M_-e`5qP;+x(>Fzun~g{>4&tzKC5;BR|ZU4q_J
zDQn@~>vyEoNV9opsWnIgsS8&sUT@pJQ22e}u=fr9`f!!rzKxe5-qCph+@PnZ>~qhn
zg27SE9#bKXZkS?B$Ndujwrd~6eMB_wDt${Gxwj&$TV~4}QIs@UAvqrsr{zdHUCj4N
z1<mH+^ybY=zV+L`i!2l0oW8@e7caP+o_O=wXKe4@@b<FuY&ILAG$*z#lP{(Sn{Vz&
zxA(M|Xl-LHnXMLo;Jx7M@3sG*qbY;yBpqq%8!MRgjDH4dW8)CS4XN#Go8QuoxF~p<
ziv5#@jfUD@wA(?}Fg_qu0w+zpeLC~@)m!sN`{EPU*W9@8{VR4eZ?78q3mVGgflR?}
z)zF^>_4Uq<VE^JmPkzatUtd?9&i{{HV;l<LO7a*d>*kwM^GeGXdb=jRKS>CrX+~Tm
z0y>EJ09vz+=z?*jL-%IVzP_x;3u6D~3yeL}G+ZfJ_m;7?IvT!cOe_Dmn~UsRXy;yj
zow{<}AK18CMseCOrE$~7Wvl$;vp4)7KL0=Q$>nov@RAwy=Dbl$<93du&^SK7L$td3
z)Qrxzt#G-lq%;xJ8P&!-FGPuK)yBnBy5(uhJe^m6rqo3tft36JAG@B6d;i{>kY4oo
z7<-_FUKIAE2r=Haw(3+P3n^%=aap&)DKsIS?a9;Re8|?!9^^`G95DKeVD1w3$QvrQ
z%w`Q$TeITVtzDuBYu?BnYnp4bg0NTts94s05M@~=q&22_^7m#G9`kfHK3eruj^Z{k
z;@@w@b>(n40ydUUhuPKPZ4@rH{|8eaixtPEWa?(oa($j2gQ?8d%ZZ!At&2CD>sJi<
z#hZ#K43yMt+<dKT<N5s&6<B5qdN?eLy|-FeTlKZ$E)HtE)-*Rhu0Q1K>qlt2r7V#!
z2jT=xIzJYFwoIUbixtHI4)qAW{y!7on5|63(8k@|x{Cp=pq|`%E{X8}Veij^ZCkUu
zPVDz}bFQ`a-e;e4&pG#IrZQ7XNhOew1S+M38j2KBfp8(HQi+hEK<33!;i&LqMY)V4
zT(-kd^<Y!=;Hanw+ra}5rl0}^jZp&vg9IWm5~x>7sqCAZw>iz;d#yF+{F>p1F}}Ih
z$rK25(@KPx88=V2_u6aC=KIF@kAIWmNQh3)#~Hxm1WIYldEscbKGU38nobqHHGIWs
zWf171SZ%^u6%PA7)A5KpK#BSX##1Bt>ped2B9N13T_2I!;jqAD&*k+suZ1JK?anr3
zEgqbAxD?29Mr!5qV#_#0TGhG!xY^QL<yb4b-44($Fs4LTH1$nVH}JD}aP~&rueFca
z#dE<CLH+zn_l(k;LW==Yz1oI4&kM(C;oiLwIBmc^i`dyYq89<S8C#uqDq3-;!tJzM
zpWfYBLAm+|3NL)FJ~Cd51-Je!U7_X9?s_$nH8i5v7iWl9_XCr&FOahWb0uSAWa4>J
zZf0;R?`gB!;x4YZJ5Gcc5eXDvyWP^FI%fOe7{-y{1DCrCF2)TX`{0o>9l1Ro@!Ji}
z!F-rlmcnp(k4InfB{Uz&)53JPW1lk;9Kk928NAxlc>5f|5Y1+MnKgB!pGez8`bP>)
zprW~|-iqDNDuDZjj~HRQxfO)%T+!2cda+EBR+;W@xOw)J!~S@7arNK5v-mx`9h$VY
z1YGOMe!XlA)>V$W0E%5py-C)dKy=R%cdlIi=R6YYyZ$vFo__-m$GN!5*>lmS1umG)
z3YAviB++&D=y`%4biA}uJi0|mSK90MbY}x#MUV8}JXQG}z0*iW0h_aQUsNp!Y%TVg
z*1F%B2b%Huj4#$O6KJoP_i9_!*FcSD)O+6HTx5SNy!rfw_doqPUSFQ-SPej`d5l&{
zX}F|GJj0+ul-|UI5C~}?8n-KLMn#9510R?swTlu_^=&K9I4%=|-=P!TPWiF>*B%t1
zueH7FGCp4a*R=q&o4nhF^=BvaqFT@KycE=aaM0*9r&;s0)wO_M@?!bss=7V-{@JLq
zuRlQ%pb$JU=-9O_ntoA>8YTD9W(Wb1N;b~5`rKM3y3=v_k}E@0Gijt5Md1`d8DnG-
z9gjC)DU~q|^k>|86{M7`<IOlmT2|3zGQnS6^12+h+E%~a5In6G^?vjLOxTv^Bdrv?
zH(^u?iflTEOGy(gj0cIWOVS!m)HCsNtLDiab#+b=I<d97+Vm;Jp!vNvfw~!<atL?q
z@|H%Rxj=J{kOqcvSQA#Xjqw^~X#1)|?R^MZdjt_;Yliq`q*N6u<wd;@(S1LL$Pkhi
z)jfF!R`k_5S>s8M#{>I==eR6P^UPtMn2wWLlL>2q>|%hBAPUVU2(iI-S6Swn<8h|u
ztVYp;iB#HKQ-OS53&8&1;Cct)Uf3_hoa`AQ`<+UdrB*G-+sPtBM6(LVG0~xGMCbE*
z!7AgB$n#|5l#|-ZH*Xu_iyPHv7k%CC;QY__m8?6j(~A>qF{*Ui0)x1n5;E}tsX872
zPT7m}5m8Tq=s(AhCf7%%_FDSYr%~hjHqP3UH76Hc$Uf_Xo!)(4$j{LtTMD(c({lzo
zTIvO>eNRUPUMMW&Y``x1I?mq1TK79x7VTm|Vf)~9B<f>iv)gg!17R4cML8kCdoC_6
zbm0zh9XI1WSN9)q|E0&g@rjS(M7WzKQVuMK8RtCX{Rf15_nC7e?(Xx_%kSX2W#)NL
zJx*#XQ7di9YZq77O<aJg=z@JH*X=RRMyUOv?$^%SIqZ3&y{OiO5^4phrw+HHh>AKm
zPH3Z@vsTG>&$)f_g!yn>3%!2DyJd{W*+>s7T<tyE_38eBdP{`0`EdPbyHi)X`gR4U
z*~_vzHT>#D;EVI;-*U%^d)M?H>CHq)R-B9am7@1ft(y3t;6~>No{GqGFHPmMg%yc%
zJ@2;33AlIetO3K28vUm((&%LO6WztEM_R)Lv8e}0=MH+EUU(m@i0%LGR<*7}mBa+S
z_m-*h`WsLA|L#A)`|Hn|-H86I;2bR|LRFfMZ_{{19D<GG?EQHyw2}|T-*wa)DYdbS
z%HeG?QcEWI!1d*h<@p_XnO9edKK69ZnR3#KzRt-r^X+C<Q0uup>!-UYru8Y1t$`gl
z2y60#Uh8p4YD*HMjS4lE4bhr^yHR@5aXPMrK}Y%|H#1+3CyTMQQPZNii!X96T9o*(
zCJ6QW1WX(REJbilh`|$^TPZz0D0=C}q<qU(SaM}b;0Jw;%`hUh&`KsGb#|yFuhSk)
zP}23vHH|8)i)OATD_Hv4gE(Rev|3n}Bb)8zYJ-xZqNy$hTZb3*{j>e9U$>=Zdq&jW
z1{YL#BNhSc)PJ!je<@E=&Bw=-_<;N(588c7i@0IKaJ6N;*r~Uo3Afa`sox{cUbC%p
zi{6;w9JwsyX{H{JcoDK$l{p8|1%{9qhfxbcYvzIrgcOkwv|h32@U@#;p5Ig*^;eCN
zE4e5t?se_(c%(pW4u;AY28PX!w7Fo|><BS2j2q^|-p25XUYGg>_5nOU`huq?#hpP_
zivBg{X#{?zewc%wh)%QfT8y}&=Ab%;>`I@sdIB~Cw=A|;6eOrw=Qxh$DRqW!oq=pG
zA{M72PameXzULP|$7w9+ty2nOu}?i)QyNkoc(449GpmNa*@ti{EP!<r=Byj;@1+-Q
z>)Tyrs;f2+fa=?d0QHe(FFyJp<h0SBset*_H+J193=Vg;l#4!?-iXu6gFENd*!Q(&
z#f%bSK|9W6QPr=}Ol-pF2q`k$yTT8N+ryp|H-KZFCl1TP+um`lpVWI(>1kJf|Is5p
z{jrZ40_&LOg(!}T%S$9AvI~6h6Q5!_WbW-Q+3q%Mb_1K;1G_(T2@MDo?dD2ZkWwIL
zT-8Om<ZOiK-n}bndu_5z<{zBG{{5UIwuv-B4aL0*T>6;37yHhcSK`Sk20-e<^6WM4
zp1jFC*S^;S`fG47b}vM)Kc_k!APb5*48urFiL1*CE-x=>(pY9yiS<peTZhzAm=`UK
zvCmSQ<;rqG24A>OUU+_fYaPxx<%WqixK%1Rua+nw8NaOJ4G~a2Y&bQ0(x_!w=?H3y
z+XWb>;!=&IE9yw4O^3%JF05lr5X3cH_mH%Lezk=`@4KE&D|9=Iz<I2rMva{U>z(<V
zc5amGhU<U}5f=l;edfuVH+<ygll+AIEMQz<v9-pbT&?JFLo(&75YmobIE(moJ?mV>
zUTn2mX$V6KloD11G<wf4Bo0%i)@-Wa@Iv(0Dz+1)*Enm2ntA8-&(~GJ>$z&8z=A)i
zMU@W*YMoF?VYGxQEv9QWAj0(qOUI*2&dl?S^MNrbcu~3-!Z?+C6U8|z9&=W~hx!aQ
zY7<flAc4gU`|;fm)O-wWG+gZVs;$wytb0yy%1`t*ozg8Qa5pVD65GK<2*$~i)@UYB
z$+duY3}MvkD4K2)V<6Q|V`1Iwo%6JHI#HSPo?+PP{RV3ny8;9&7GX`1?X^oO_S&A3
zGo@7Wl2v(YLY@%%9#hqo!+|s9^%Rf1xqHHw_%9`0T(W!cfUE0!43@~D_DTopGSl*G
zMQ$RPucZr-5C(l_q_N~eh<gg2&?-@c;1+_@dojgvZQ7(+Zt&&;x7|S+wdn}z*LYim
zXUCa5&p6yk%qnVcc1o(Ty#xbHd}P?}*j!w&ySU)$@`B}XfU-bSQ}=l}SsML%j<?PM
zG(@9W7mDM=yTWN#28u?AFo()d%Lnl>s<PMMRGe4th4a*$Nkc@uW$}7b^*ET0$F->H
zQfn{-MCCqbMVI^BH~hlCxjuzm$=IhV&TdXM_E9poa}-&0EF&ON0CTIC&NWftNQhw_
z`vx0Vbj)de2j`$8XJaPYG<p#za$2yO@djiq_c)zrZR&L_w1R`D*t7oJ>U(!Do*qj(
zk&Z6tPg@t#Rr%Brq+Zwy8r5c=O}&Wf*G7d8_Su~yrh&BCP^EHx|0Nw^;Mm{Y>OyRm
z7tVQIbfvPrykdKK$=#c8=&5dval3^WIV>|<aWn$+Jo9wg^YqCbm4=UjanOv_lp^C0
ziP5py3=Bgejw5NSe5~L+2^|~js+uWfrj(2<3w4>9n{gUjBWc-;*WJVl@jk9_He#~^
zz2Beyw9(pD&a(-Q9Ju?;Czx)YTA`}E(O{P-I32YNX<#!9TwPsp{n7(odFMNL_q*TC
zE3dr5qsNbV+beJ5o$q=l_wL^Z^FX}2ea?QrXPPDs`#m=|H{9Lb@%;HSo<G0gI8D6q
z>`gxOkq`6X4}X;B&!1C@bdmdt2x{YXQjPvrIvj5GU(*x=8~O(og?Qs!xuAHhClI0*
z`1*YGBCVbjz}7VM)=pZZ<z{0huYxw`lnfpkV4Opzo0;x>VhEmot!B?^4!4hLn(utn
z^o}stisx69TY8h}ohe2Q&9L`WL>q))>swv+OmgG{n~&hD!xuBKcKW(w@}y`sCgMnA
zI#WX1re<rI^H!*ovaUH|0*aIZ7i^$4rfH$JMhXG%GChK8Me`7~q&XY?idJV@K>O3W
zx+YUt(IL_;qHKax3uC-sDJptVZlxCf!KrW|7eoB45UVB(JZ80SN<&h#tTu~>)6bf7
z<#?DF1|4(O(pYjPStOg@A3A>Z-YW{Ys6Qeij4^4khOm$cAyJwzFN;mUJPkE$&J=oP
ztrPVQtQ4lDayy%vcUnlMLe{KwGv_Yq1c4RO%>;uuMeUZN*QZbTyz^>}>%fc0Q7Pp>
z3>)ROI=h?9QpIF*+N4OOw#rm8$7MFOv5{gTm&&vp7_xE!hcT@a6j?zJ-IqO{sbqeI
z5AhY3U&Z$7n%(`ExVV2!8j>~=%Yt`sT#8yf+QuKeMwZ55LrfzsYEhT6ic1%1L|3Q*
z#$ltV|1f|bbepJ!B8{-UfcuXLy9*>HeGY@tIl^vBcs5g9Z|YPa6hh}T8uIKky=1(+
zV!YU~-HxO|MJsFuE;os~%#`DimM8K&b8~z5me#h<9|C9(RY(6i0<{)^=prEkcXA+k
z4-)X`m@thS9D;1sUn)2)YDB6@uN!4Pl9!nwhP8HCt8h9})!&a5J6)CMPVeIUcPo-+
zvw_dv+PYG%o2@E$O?1`=3f~9a7}Rq}Nu-^}p(AVyYVTbqXFh;uZITFQ9(rx{jiQSY
zL_2f+3VrI@iXeD}wjGE!l%RL!(!I(28cEiL8wMWgwV@eL#vuCs1)NaD7YhJ)LMoju
zhB+@fiVyCE5sGw+i{RN_TyU5U+#QeF@y&&pqHd6~1{7<dR6QY$Jb3t+yC+YyDD=7z
zxp@X3Da*p(BZ-I?r$(RNgLsO0tR2a?peFI}`O<gmh^h&1?C*H`^x3M++>9gRkd(VL
zZWy*#440l=bckfyvS)hwgf^+ynb_>KbElf<`dKZr*r5f0Dx^Dy2l}1%V;_U}zI-4*
zdyRTLfQ_mKrwX&pX3Hxtzs%Qu?bq=2U;p)d)o=ew-u3Qx@zSG*>@F|WPs&uLiZq;6
zZhhY69W7@~9iXQRbpDNSyT9f6&2xVFgTKs={pgSJ6F>Qr{L%-0iT(bLWl<AZ^_5dp
zp=YD^l!AZL%fZt$!3vkPz?V+)fS$bIbrJ9QR%?!v4SYx1&pI!A0J~CDeZcfN@i7oX
zu<z@<H!B!Ltt)yL>6|uOr7mjhUMr(>ggAh+Ce<l9T~>5Yo6Kh8XhnW?blN*b_55ob
zZVtTp#_Rm>@%?;C-c;UO6N_Q94V}OcHAA<SqGtFWYi2#pMQcv={_EMh%|_~J7;r@k
zl9<$zC1(W*I_Emi3qF_#(4X0`oS}cczdx;2C)Db^0Mzql>tSo8G+Gg?%^FqD$+G12
zb((aZ<#|)SdR>6PvZ&`{=i}w94e@rjvp!p!h;dMUnyg{~M@U*cmYhjZbB3KXjlQdm
z+^X^#t?<-+#4|6M5{xd=&8AUgGTM0GQ8_MZ{_mX-H^~&<Xwhzhrkods&47s7_b881
z*LwY0S=Ge-_reKvE^X?|H-S-m6)Y*mCP!j^m&d$t%!@Y0)*R^it%!AP8lTA0aNY7P
zL{sDft`f}S@V4CY3Hb!?9KM7!jKpyyrigDAFyVrYd%J8%*8d^OMRb<kEv*t<V;CYu
zwhHJ}!4Z()X;se|Vtl}>`^5X#Y%VVGDKX_@=^&oVtNTpbOmq<!A~9(lkz&->_4=&4
z5U4H^$1R)9hLi%{t9M}V9*K^c0wp-=?3tPvuk+V`e4aUgRw&qw;BdyJ0#J3*Dp40!
zGtQhdH5E3)uof$z)xLL;CKai5^#b%x$H2+VpY`YJP1pgPew+h)`gOk+(H;6&%pvB$
z-t;0lZ~PT+vi<(z=8_*&`rV|W>Bx$tbP{UkoOD#~TsQ&X*9EE>_rn2E7wqP>5N#@{
z7lJ9YSowuc)EBC?hPY;<Pi6Ed!MkG@ec|H_hSYPYrm|<}u@}#0i)O!`D?(|s+ENNj
z*70vI{(8WB!;lXg7gtw2efE^w{hnbQiQaK_ea|c}wBW@|lEu2FmmWRh(;xmI&TBWm
z8#f#e3(Hctyt`%a;95~HM~Cr2gPj+BzpYi?_VU|EVc_Yr8*p&ByT@_A@ch{g0GL2$
zzYfoxOlJskH9k#X*bZ#Af%m-oZG>@Ry1C=RxQ%NeZD?6*-G#@{=@mWI*VQX>iaZQX
zczC(t;ngK$dd}m=ukh7h`?Y-C*MB`<{|)cu>weeo;`-r3O=Icf|JAJ9(gk3(GU~JU
zRti3NRS_<gm~<@nLc`M<o68FxJ${*Yzw3*6-@o^5<ed4?hd#vn-~WF8(Lefee)1>(
zDIfpDr_KroyI)Tt<MoXF8+nKz-3u(}v+PlizE&nJZkldx-Yat5Et0$yfO5XBfi?|J
zZyH&VKxgYqw-!l+n6wyliZnXs@G%%EzS_pE;r6aUi-}e$t?Jt4*Y4NRGt2dHt_I_G
zQ9vIfSG=!Kl{Vm=)A+c<k=M&pJ|k~ZOSXawLwJ-Umd?hwDe3yxH~SvdDOT)ZXx#dr
z0~Z2g94KXBJ{}dxbSGo|Athu{l+(ww7OuLan)2RthX?ZlT*vFD!V??YtyqCHL)p84
z!V|J1*TS+mV%%~(&YGWlW-Kggteam2AJq(?omcUerddRSR~r<Q=%|si!oWKIZk1{U
zo%fz`GcpW=KEqB&*)9fZJ{fuk-}8riuje=}g%C9cvx(rRg?pD5L?4L3v$RHT!n722
z$#|zhQDLb%7S>$cx;i9C10TD=zuA4cei?PK(@C7Zf8~77c7JBP*D-E0Yc5UES(r*@
zKTqHTzIl@ElYQJ9W8#I}-VQ=wVjF9U`u!WD<3r_xd~N=ng!ndH?}nt;!_>^BR4c9>
zLDbkAm$eA2{Z|z;Nl|6Dk7gX*biFUqAe<Ut7e{0>lE#5?GuU(M*Y|w!(gX4mpa>x)
z2--OId)xKg>)^phd{FLVzaO1Bqlh(Zy>2+4ApCld&wBv~_Q6gJWk7EhkK<{3Za0lL
z<(MvnWuDkC6T8g@C+a%i38G>Pj4Fy0^~+14N3D57R(j{BFZ-))NAfw`CaYk6^}}{^
zp?5l3l=f7BF8gmZq<OmulST~<Vi2>cvrY=>6CwZX9nb|DC*;r3B;bfhOVsLU&9PJs
z=FSzCs_7c%q6UXB#6S!Y?}QX#NZ=%!Id#LEAC-R32@UY3f|F{l=;n~#M>wZq+nvFu
zdd{0va#3DcN4uPjlk}nWE6IGb+49QU-_G%FrnbiYhgVb^2ShP3kn0GpUf1zr$K}NZ
z)6Fv^=;Hd)^#ymi@~Mx1l&MALxe>xZbxw_Te4x6*sqhAI!qewZ+3vRFW#;DD6CPaO
z<L>5`T0|QZ*Gcp``kJ%jI4?ZES$O>50e)#5j|<z~#<;C$wyHQK7mF^&3J<xp;Btok
zA;!0}n^8(;<163sZSUi|zVEm5&A;yt@I_zpC0t%y;!`kWQO7}zxT1$W!bx!b0aLU|
z(DNVMYAm|BxTNzo6)ErE6Y#B4Br|Syj0iOszV3JZyL|1}{x1IbfAS~z=tn-vPyWPD
z^7nr9$M^^T;2-jt&%Aopq_A4y{F_<;Hqi`V-T1ATvFAowrn{bb*L^0fdBthYD=#8W
z3i|hEPjkJ+8it;|t%YAF+q)MwUFX$@FNuy!)zi(<e8gL9RBi<qgB4zSy}MyOpuVjL
z-5#kw(6v_bvLKez>l)PK!Y|%@j31nSg4`AyYFUDLZU*PL*d$W$wn6E6>wIi!Rqbv<
zvLEB@yvhp`n4Vd@zMIyF(OWtP1Ol6FLV`w`HRJEClNA%d=<(vl=xoxm7Bzil)Tbjs
zd3GTNLjQZwYg&p57M<^D8t#-%<%1$yKu;jT!8j`Qq}6WYvaHBlEp=TNL)7ng-mGoB
zV;mwO>z<KP)Izo^ibEiYP92(exCTouYw|}E9YNR%VScO&yRCBNgT2QlFc%f@Y=#IC
zvT=ixO}ScYlw8-nyep)gp~W#qq~&#Qb54aCE(TMt3zf{W2vx%Bm)et?aw~M3p%A^=
zh6T;SZKVS|ks-x(y4F#(X6M^k2t+Y~-|OvlUaz0w9n(80%f)K#L#s$uDo-l~@dLv7
z9%a)R<C3<TX(i*ZA`Ooiogq4Ozbf`DBr|<)l{U{v$#zTXIk|XnpAatTBXguGSYrTy
z$DY9$7&bd%7!lv>{Fl{RQkJUgZugdyQ>x6b_xS%y0k|RzoRBxr{qSd(574mg(;710
z;ghC?pmSOww^nsQ@!GxCQp^WQAFLG5JaBXcrSmzQaJDd<zNBxS>!`qXt}diF6|857
zmwjEc1yL7xeYkW#?+cGG43TM?4FR$^ntqW!dxv!Wr441Qi{5BWvynvt!br}Z!%;aL
zi;on;^n4ttK}o<`C~3DHGq__$q>=&*N!SjaF*#Cz7{EIvID7Bb2io;>==*E&;yP;o
z{^DW8_*&K6Rg~Dh7Jk+R1snhSpj?N`t9xum?SwDyZ8;qFtA&LipYvmU2;6(=0mr)=
zYSEnNIC!>0;PU=MCO7a)zw|MV^PcKp*liKl$tnGM3rn53-0j%xMs}AQ-1A$M>!^#D
zbZ<a=cpK=X%xkRS&CQ+%W2BgWSQB;JugyVqYKq#OC7S5>8nS~AG&GDp6DSswc<<pO
zzUklpPPXs<O5($JQq#z1KK%(kv6#nYaE#jx+uas|D$RyCGN!@!L%Jw8^YH5GPe2Lt
zt)Ef1CZ%+<T&xL5rmjT5r*5|Exqt5=*I)C1ulS1J&Uby+ckzoK_yB+BM}CCA{x|;L
zeB?tP;&43bvufArbNZ`)0}eq{d=PXzr@T)U1&9<wyc0e|Vu)uETxa~VvxTt7kKx=p
zVhujfqCQ8#L_N5mVxrd2$Se!3M{Px%8*#;+D`$#V=WsC)%~ne+K6Awi`g^0>(6kf!
zX!}sL3}2~=1RnPXer*0(Cbt-QU8wVd<U+uahDeN%hOi0`ydiO&no`L|cu)FXoX5F_
zv&`APHg@rqw_>Z-x_ZZzB~x5fueGRSYE|6VRV2~T5$Bx=S(<&ep3HNuZ`LQLi?Pz6
zh*MiRZ+h<gdC}|a*HI?k(QR6ajZ(eS?BHS0IJjlW3@H)9i1$!(#)m-33UDZ0CYb8h
z6x_C^hIQm`@v8Rwj);2inU+HJ4hd$;Zw++kTsCoMsYsB#r=en#<5Ec;9GQ{G@i>!9
zWk{M6Tymip7p@m{ooCP5K<H@KyQDylaZo!#)ia>yJX^$DutX8Jt~;t9P`j97=-V(1
z>vd^OQ&UzWY}fY(b<NWPNzpy$;U7PLKVS3m*DxJ!X?1~GIo#bc-8`pGlTmAS44W;^
zD<?CgNV5j08p1CIV&I|FOjW*>uI0_sa;BAvqd?8nvQWIo7kkhQc(~r}xJVIcI)y^2
z8pr2MtJ$Uj;5Uigy-SASf_|DyE##b8riuB`pca;XoxM9zetw<DTQ2}jLw()_DEFgV
zvz)$EyEQdbj^5*|fD3&<3FycoSn=35E6kYJnPoudT{-M5iF8VwAsfB8)I~u3hWaQ3
z?6IdSUhS_qsrLGVqy?q3h1arSbmU?d1HA~;heHv!eG^LaL`#~9wkhjtS#jl7_b#2>
zH9ZTp3PPf_k);mY9x8K%>Nk`qw4@8w5C>j<`7wuM=C#kf2EfZNKj!)K=iI-(Vm?kh
ze|F2<gehn4=FAv`2bUw;2;R?xWI`GY$Fy8zHGkKIyV*m2QZ4Io4gk`ima-NdE0596
ztJArRE~E{b*HKW234K;PV@T`6h9MQM>A4p35xjW#i1w*ZDmvU8(Rr4tirtF`_t<PN
zNKrTJcfR{wG}n0Y<SDz|POSu**$WLxKeyLn{bF}T2%cH8T@<xI2th|X!3i;Hp*>F%
zad$zZYQjfrwBo46f#`<4o-^N3KTq|63C}SLiwL<z()Jp-<JCjv@BY*W3BT|W=2}=#
zGh7WSw{sN&+s#Obfvc-~Jh=Zdj~>3l<Hs-a@X<qFe)(m#n+-7ySfj2esCX;({EA{5
zz^cuLG?N?~x~toVajwLC)*NZr@bHx_-*A1+H@x?~{INgw$N1ZS>xcRNzw-V3?9coR
zcl!f^Tb0Rb%I?=ON&kuVum94^%fB9nB7Gqyl)?={MVhu^;7^EmRHNcrv#xaxGk@26
zq1JPuoHyGILEYxTC05%EXI4H!&*jz8z9I=u1s+jR$!ZOWHeE_?*EEj^=%y>K*)r+-
zr^2~r>!6m1!3Uz(5h7t=sWU&?evy4to^AgOF{l}@w8oG;DXHirM2n3?i?$}sMpZ&v
z7W{CE!cr7f@3_`2s#-xxMT#m<MFx&@R&g87c#qyt?i0t)_dR00WZes~7knoGAOsa&
zNL7mwL>N+1<Sj-`cJKiwiy|&LgZFGNcIe(8h}Nv;(wG+Y#Pc3DgNn^+J()fa!?2E3
zoezfIs2Q_&I?n9d80h&;F@mT%A5kyCTvX(dl1D`qj-}{2Rcj^1pqahmEPL58roa%R
z+EUR?j?tL0xoW1bIF`I{SPD_0SLUUi(f4IVDIFHQhZ2!mYvkI9Ns(Th=kA~e-#GoA
zA_~|v6M{JB4do7M6b(AcZOV&mw*S^D_|Ql$tXp?4^y%kH$_f|Mreo`tGxc!G;)HTc
z)aA(W*&EFJI~7@s7i@+NF%A&D)hOyU9McHnhE^0nX`%>jojtx~PSZM@e$+w+mStwx
z?C?@7TiI4T=eWGMU|Ce%OvB>r)Pq(^qqc$on{i8UL7gkcLCcs@C}m+81BcjD+;@!B
z97vlj65d(~_jy48KZhH3Rw<kctPZ;y#-pGfRw3v^Z49;A7cww3(Z8@r*^|7E2Wzc*
zO>FU_f3LkAT|52h#%ZTTVf`A{U)xYZXB#9&a<%TUJKOkP1lre2Yc#4w-Kf)jiXL1z
z6{++>uy2L-ImAH=NQ;yjXl=*c(Q(Xy;zvqwWD1w}FWFsgxPSkek}I{r<<*uaPXpWS
zhL;{aAP$kuHZtr2yX!5}A#?Y952aA%i6?i3ZHip(E(vmA3<V^s!73gk8hll8p$kdQ
zxnHs(8_sNWnzVHx)&{))`{_FLj_=e?JDd*;gI_ztTq|AzJ>RmAr?p6RY%ebfNtIAt
zuI~^^$!tSV#aAgruLRqB53YFSmB&`p=psK&6Vp5);<>#&aD9DEiiyk13#2K5+dHog
z{h|%WFa(yOr1!&d0e??NYFMUdGan}u4Xy~{Q5Cxrax2{KGvxrJErf)N1I4V1=Gyqg
z^XJU-9*01P)Ot!T@LnUShNSA{yPG|q`ShE3_bV1x6}i5?=Fy|ayyG43;N_QJ=KlR_
zt{*&LcX6e{nU+sJfv`3mfGSO2c!E!(OK$=5-bRq)xPe;u@-O>x-u<q3@<027e~=&k
z;UD6E`pbWrpZuwxWm+;#l-Jlg!v5mZ_M7l<I5Z;6`(TJ*TTu&*j#Q7g-~%T!WkG7D
z(oNxQ9NZeMSw?H{^wF<$4yRF1v_KS^sqFQg=u+vhp4U4Qn)K`0v0iJW`)CW4meA1`
zqdDN5HL?2adY^EELxwHX%8%du96z^wgj&0;h@($78jY9&o6SfZ2IWe5XU(}?c`IOD
z)8@RHiLyhRLOmCCZQ6Tr>%F5D<w0HUcI2v4jc%DytExGb8+r19?ad?ZbbqVooH43R
zFoh_uu6xqO81*@D8jaai`#L@phP1;WEOQ3u>^iGJ#0RH^UovxEL+46SH;9}U6KVw(
zvt{a@vHj3}MmL7;6<Bgn4tlMG0K46&>txQD{UECWbTE#j2urE-Mg<^eeI}dt%*&$Z
zEOejHg7aFG;DlWaFc(S_#%P6h(_&0&QH0K^0($X^^c{1h1&bC$uft)<L?1}e)WxkD
zPte)Ks4JEAx&-enaG|VXr*+@0mF8qcO1lWh*>&|^Lj+Pvl&YxfJk5Gf)=FKJkLZK?
z6^7`jQfbRfS#&xQ#w|k}%udV1O07~#W}c1~tEiri-ET3aq^*?q%4Hu0H8@WeHD{C1
zi^~h#UWGl&qINK?YElX*;+qMy;(#}cn6w!!u%IDzU>PGZKxnGA4sl>!7F@}1X=Qus
z1>g#*8LQ$98d|;L?8uTi$gx|atwo@T@pTaleH>tar_~lcZmlmZB6efg__6Z?qLHRC
zSv5}9hkt#a#-})c`zv?Ep~#`3IJ#RZx21QLJ@c~jDUh;$-Vg#QMh&$dlJO)?udj%v
z77A)x;GlV=MQYo!WY6tUIJQ7`37-ZS9FHEn#Qt#0cDG}m6&5}mX1vn}9N?oL{TRbA
z;M^s+#CA8b*$7wnu5jXb{WGsJ?f2Z|%zS|B?Io@5*`!8nN=o+48ON#XqwIo;v%SQh
z5JBmC)oHZS$1Oc_Y(2-Hv;P~UZ~r;(XvIj=!BpP<RCo%S5HfCe<inlr7gE;_J^G;I
zPFKmT@y45Pa(R70h@O;`XCqdy-8{eH!GlLYU`UD98n3+an76<EZ5r7#&Ak5l>o^DZ
zuCEC`@Uf45iYA_U5yTBfLbsgg)z7M|2(frib}&)MC2K>GA`!iBYORDtH|%Lyc=gRU
zXl2IJR-{Q65zlTn*!aNFT42eLBE&H;jsx56md$2Mt(A{{^rL+2V;^H22huo@hRBzH
z`M<+UFTKR|{re1Q#5r~9*XQ+DA0hPS3unqo=QZshM)lyzGiexk=eys{pZt^m3%>PR
zzm*^O!5`u;{*Ql|5B$<E+vvQV3IV%{zZnlL(2^BxI<ghZ{`E0gey<iGHS4u*W;)xs
zR3c}_#Ww1540QEmAN%#mL#rA5i5KjgrY&^)jXn<Ub|$(enrEsR4WF!s!bx#UFCY!^
zb%9R2@MyoPmK2`X_pkBucORlPGq|=UpvfswBgC+h<om|mTwmMt`Bd>4)_}<R8SUii
zrfvmkDJl}ERur~YnI{!y4Kb37IZc=tr#mI|HK>ay9IP94_b8RKbt{)*-gg2)e@^h8
zal2I`VuUW@8&bp}%=5G=P={eK9-ud2q9RjW{I)FFzy~3w1OiK447fV^N_s2$R=-xK
zBEoL2kaK}nxp(iL7HSrQ+L{VB1l0~~MvD@3j_CcV`RYZr_bf%l5Ya@M($wlF&y}UB
z)mp6?=PRwj;C26v!4rc~eMHe(Ef{qzTXJKTMMW>-NWtNq#iM%9?&6AEo)d?GanSX4
znHR-H&is453l#IWQx^v{Y*sHwaK``B&#6dqv+La73kaJuYH?`yMlD(Sn`zLb9fXo+
zYSGUjDKW-@AP#ZK8gyN`V$p(Ms^!X;LhYmb7}dru4P0Jckv3aGO2%XK^pJ-%48%oq
z!b@H>%GNnTPdqV;Csi#E^e9sAy`|$=J<=N9Mbfxob8+A}&D<UiZ)t7&ycB?a;@VLz
zXUJYxwA!C|qdnzU^2e-Gum?YgzK!5l!G({`y3uajX@uN95?(7!(0a=F9km#295hxX
z1|@FxLlm;@kuKeomDXRbFN(d7w&Kx)M{hRFS-%#At-VtZE!MAA+4eLo3fT{7MJ)|!
z)^UeaqzckhXwX{XFemPoLiT}{JTAg3?|Q^%UVnqlZeZLz<muB}ZlB*N)ZgQt(5%73
zK`k7fy!ng}wUfAaeZ@;JT~kWq!OJf*9gjSH^BMD8c)l0zZ8yl0*+j25gHz6pIORZ`
zl%)cEcCpZl@fSHxeWPE8GVTohTfgu0YcnFbevU$nY2~t<UJbx=d3BHJ*_%4{65T9X
zUI;XXAt8Q7s&IR==To12l`ned+wB#EvSgeSHk*y66r@C{h5dd{ijnL454gM>I8K3=
z9)0=BK{_6ertXF%tN6yJM3q@l2dq&D9&wPRY0jhf6bHqrC?&dxxWF_Q8jy6p%H7=^
zzx07$Cb`CLGq9A(u-WkF;bT%7X%!yaf54Z1>6h~*@A)#`_V%}Pb#;%+iyh;*QLjVW
zTsvBuOXlY0DSzun{w6>5(?3CL!WVtfyZ9Zy<7;@&d%lc^4<D)Vsbw0Ub#nR`8`p-8
z((6vGi*X2eH!2s(!DhGP-}#EK;Nhdky!RWvk-zX4{sKSngFnda{-_%vXRq;_un6pn
zv5ut*HD}6-LPICpw`DO@us;uW|B5;tgtN(nB3T)ln%k?=-x@Bbo{QcrWlsK}_PjL*
zQdcV)?NhbDY-T-$Kq(8(6)OTg-bba|h@y|D=1$h3Fjkzsetd)f@zX!V^LnEVne&F^
zXcC4_UA%6BX;3>0yR^=V7-<!UIui)G6-X^sfd<ypD0(jYDJsvSh>+Mgu#~LDk`;Td
zszGuWH7OG7*RMafZJ}m#UpkGxPby}+zRq#6+o|{4Qq<V_aG;e&aanC*HXEuGVoH>(
z;)m^KP~n3$|Kn!lc${b=jN`bDXO+_z5h?oh=>FF|Q-^p$=Y5EHZ+0@ZvD*%Y^j4PK
z*dJ$>oYmILkiY3Tfzx{?VXtLe5HSO1kEVEppw|(n_e@6LUv8q@bEh2r?2OKnE8Z1a
z6Em{+M)5!kL#Ng(n{m)R&wI<mwXw5V#k5k%PSZhRv@tpDWFzFA#)7u8UYOX^QA(kS
zkj4`V?_e#qnkY}tImi1R{vL*Lv`L$a9({~>r}P-#DtSIIA14meLL0YS_=pP|#t}pF
zl^!Fm>HTS_sdJ8K_m$pzLQ11W&PDU##rCuQ83xi&7&aqi9?@J9uPWs<ATE+pR}br?
zp@V1QwTbIdxN+E6LdcF`IwH^C+S>N{83DJ?E&zD(LF|a0boV)MLDe)NMatsLKA`Fe
z+YCy6uR6A@#q#gW6jP)TjB}D4s#+CO%$TEjSHu#65sgvhl)?Vz9#K6C$jU#uWyiWT
zy53Fejo+#m>ZleRt$AJ8(6NI?&!{>}VR@%N+v(gN8zoC~wCZtg;IP2+sZwJk`^MvU
zJY=}ol1t{D?|2(e-+0EeCpRn>@Qp2+y$kl{IpzpP;x5ed%)CtO4|fdX$mQh~xe3KZ
zT=LAz%*`y^8+Wv2&&43PP^c~AV+0q}nx^w}*3Ml%=Q{V_cLaDH*}GG>uS%;EnrfIa
z%QEAF+TE05u%Z=h^hL-v!QI7`8i(ru77$#klzCyZ8NuDz#_V|W%^O;CJbZM`xJiWI
zsI9QwjVw!HzdsUFU>F9*&8Qouw~>{P<UA8&RDM(Ngs4i&X<8VD4P`pA)CwU`#WT0c
z(nPr@sO>{@4wsaRRjOm2j%FxZ8It;c<!RyBr{Ba&;Z148-N@s&y}~=c=w1B2Z~G3u
z>T7-%U;3qA&Sra|wn!++UiUMxecPU^Gr`fl>q|D<fzQ15DUOGUkAM7Qy#M__&dV>q
zov-=oujMPh>Z|zjFZ&91+bi>Q(Hmamv}rdRok(ps9|_Tl5idlGlPY<^`M{&ckNNtG
zOTO&OzLekldw(zg_y5QLlb`zOpP`A64J|ow0)KNI-TSbVre2phD|ZW_w2VUwu8#bj
zEF8QcZbB`~T5yPX$g0q8?xY-dK7thr%?aYu>WSuEI9(@cXGKgefP#&VgP~%YXIf~A
zCWHh&o^kNH*r%iSJ=sF|`0<lGS)S-R>b#9dwFpc;G7KBKv9%&qi}Jht8Omw4Q^BjM
zwkJ*mZ|1>gM{AW<iXnO(t@Fx5mA2-<b}sUo;cM@E6@QqmRY&Eqb>5*$Uqx4KA39s)
zH{-yNBGm*4LrP5MbEM7&p1fp@1`PopJ<<4%(>${obu*uH)+Q-x3Ph>BsP*^&t(<t8
z-Ew6#zE^M>m+6CNnN)Q?44R2s>q2dgT$F~=3-0Yk3#K8d>Gxrpta0eLu3M)nwN-|w
zJmo6Ni}!|7#R#=}b2BhrxPSy}Ol%Jnv<UBBzg;1#{)y-qd>rs;tD=)K<Erk-i_s=p
zt!qGl_nv6iNJsOH&1O7nP@Jlx>0+Bo@+x!!Y_#jwF}{dKwTU$HQgE%R*RT&br`ANR
zW|qSp`)5zNxtqwthJ}DQrP+8_iEY5ShUB7Q0M@KTGwSYc4xQ83YNh50l{G}zjGXV2
zy@u%io~;<#zXDsjH~HqNH9%3&2=A08;yjzpj^-e3Drp!=`^YkVp%FmNMgqX8Y|@9g
zsSHv#6XU30Um>-vK5xn;Q<$^XMiPTdt)L@x6bW_AHqM5aP(ei0Ak*nWU311XX!`S<
zTOay!!3t=B(YbtTZ_z)B6^ebVU<;Di!&94fb(;Bj8Y>6$^;za@pGz0j6DFWzCB%c@
zaF{)}vm?id9~1ZPT@uC%au)V?Gp{~-gZXY!AI0D)?eqZZ`ft>3AMMxC&b5Q4JQe4J
zo0~!!23~pleY`lzapHJ$$38Dy4S~Dq376Z7r>W6s7;)5Izgtnz(W3Pv{qsScQ{{JB
zaL*2J=u(*{eSo)DEGCWb1)%cLbX?)^3>%9Gl!a0)o4Ho<@xZtlEwWC<Kdn06c;lH0
zV;)@c$}5iu$urG+^~efX$FKajW(EGL_H=<#gxlK#Iagii5RUsjk1j8H_UsLgOVOgQ
zHfARjAPedhCFTdXC^31l=3{+S?B;BxGTl}-URBa@mw0gXkU#mK{3*WS8^1}(#X+-2
zaBB9tLMkiB`r^6me>WYEJbChl-4|*q{P9nGh>w5#L;To3{JVVd7r%%1e*HJ|HDCR;
zJb3WHcsVB|%7+s&bpqXW(Z}gnCNk&<v^93S4c<TCKl~$qg!jDXJ^aPL^q2T6-~U&+
zy}Q$~w;>|G`3}*MQ#D1RQB8uWC^aI|8WN{Grx*tu!Sb*ZB_Cl$pM+NIS@ZoTcuRFK
zDNU<unnZ{LEsO*=sA7<S^C!ek0O#Au=-oUiHNLDUFRxXcPY?z=AFx^R+Q6Qr2Boqb
zZaYzwFsSlZoRG$iM&9)Xx)+*rq2`$o2YPIs*Ekfr7o|ght5~4lGp(thhK`zG=vYyT
z&vv^(Tx5}nsaAb9gYO;Cih^3bwjQT3w_mw~eIJt6*B2dQ?=}O=G84xgAtl6VdQhG+
zRdlM<%$&Kls$)j)ndgaN9N2C*R&cj<y>nKahZu}ktP0dtRYe?w=G_M4$k~^#_eTi2
z;@||Sjgp0V$)qu{8AdQ$H1$Q)b1A0}k2JM`@`fk{uZ3E+a~PwpXJgVLbE!fuC!RW=
z)vWbBv~x7WYW1b(y4H%|xd>Za3+$zGG3<!G;b<JEgN|GQqBW~x=(;V6!S-|)g>f7R
zAsRxcXk)9z%&EZ#y~i}_QhAEl{=b-;h!Yj7L?6IAJe8JbYSTE-nkSZAxqI`8p0s-8
zIBwW%Huz}gqSGC`w{#tUI>}XezGXfzAJxR&#|^bz89fa0GBZzwyv&M<3pmkWi?+;o
zvquuIX6Wp$7-XYsOpJ&`Vn`6f>Q|{Eq{791^Ojb%&yxt`9CrWQf78uM*w?!b>R2bH
zM9APAAthSyEc3)N&qR+Jj+#klH(?gzh6rN$irQV5q6<hIB9cv&TFUC@<T_`>E8k2^
zw%>VZx@4Ivy^W}~9&9Jsy1_0!sKT<Vn$-MN6-{6fXdzf+lefjz(l^8wrs`-dl4W3D
zVDgb;tGxZu+nI797vc5SpRi0bQi?7Nu8=GatP_cAsmj3tmBBmhCUgNC2nilPzUI;6
z2jEnN)<@E1zTnv~QjdlC@SGuBahRTRvGveEd^;J;uA^YRu6Ek`RX)4mFMQdGPBhWc
z#@S!vnl5ruGsk7&a(iLdT-M^yttbV6BaK^P8ff!OX?p&tg=IRjdvK2sJ#AK>t$tt5
zQ{l;*&zYu~cYM*?h;iiZ?uN_D3lK%6yb+t5Knw$Sw-fvQf#Y%F=6Szw)%f@JW+u*_
z=eN3eETYQ5si8)zqWfsSi2~iGN1K4N<zqdfm^R5d9zJ-5|LBkWQGUl)el5WbASNu3
zRqJG^(!baOz<JL+&D`GJDylII<_oCz{hLqry!qtQeBc*<j<>)4i}=QG{M~%xH+~Zj
zA3oOS&1`CP!3%5aW>23*iYdn`#!i@rn$%wCcmB?=<>gmi;XPmS9{&6P;eX(@*Iu*e
zzu#PkbF6d#Y2cy<TeLfMMAs)B*?JTI#5k%wj1`1K@|2P-{;W?#I&UokQ42I5fQD3|
zwhU6BG`uJ$&3lc-3Vm=Z(5kG@W1mcf=<vl+t63b0@>E?jl)LjkO|T_aC@gv5<HuL2
zC9i~g`o<4u1)QyqRSlyF$q*%0vcAJ<0qV366YpQVcbp22G+0qiV_HLsqD9(e6GfGh
zRo!kIp(a|M0S>FJ4V?etO*yPZnl<az;CE}6S`!zWk$JJtEoMPtYrBgPsg+^7Rn*6t
z#mzYAc-;Kih9QwsA{*M7VzegQYp$+&Gxk_O=NxUW1k2uC$4=JZIbg|ZbrOQ(Vsil^
z<i%`*Ox!h>#T*#idVfVT=k9B04-V+pvfGrYfinzI#dwGoBu(VB|2b@JcJ^9kG%m2T
zOs<|<g;%bl#smT>L{fk@Xj3yylkQVEOVv`(+f`WT{c26-Lr`Vz;7!pjlwvV?qRQmt
z0{s?MR}Pq_nH-80&>c!>gr*az<^q1WBwcyNaDl5(oN#yZgqja*w;Oi59ci?vHmAP2
zDJ6VxN>%HNu?VeXN}egpL=%UP6Ri~kg&KFypE4aAb)KzAw|7*F?b$QVtKg0tIOaWN
z8W@HRY1|Np4U$HxY*fLHxe3%laBr=de(MFGA!jS9_F4bDX6yn#zW<3{Eh6Tp<(qb3
z=ja&S;1pAE)xsJ>V_#AY`SV5~b|sUGqBLPp_`O<gqn3!^VqCky&TW&^PX&5GXiKYk
z;Rz-&$CN!@O{|M8N<NsurdI_s9YqrD>Ra#TtwVK9(JAK!_Oq}riM@C#l{cO~;ql9l
z*&pv%${j=@CPiVc?_KepFMAiY=&1bJv*(mj*o-@FZ}!~Y?!kFB7aMk0JFc#_M4#|p
z3#nypy#D&@?C%cv;JCWHV7t4<2T|1&Ldy&DyfBUml~=J80&k4G&q`7`9|4~#Kjlam
z^_p%6)Lq*}Fhg2J1*B%mbRdc@`u&{hc*|bf#3vzHR|qtxS~sbUx-5(-5Tj>0Op09g
z!qW#@^W5G{d`e#DZEt@}j3cGMI3`kTl%k#ca$I=h_2=B&+;P}X2oRh((!0!%gq9~>
zd$v~zzxT}Ocw;J}#h=}%I;c0iR8DtVZ-D%|VQW#58@%IfZ|6__ssAP4^B;dNW84zL
zu)a!Hf1VK8f8mn}0#Bda@bu|(E-toQ-Ma)j$-dC}f6H><=I%{?`B#39fB55npWpMX
z@8cW4@tb-0(#x!4LVrSBY>THhrS57qq8oT?LNa1}2$6TZ?QMMD_x%Z8diWCm*Z=mv
z<0Bvbh(*i&<~f{mM0<9<*L${`w~Fy3HI2nvF{hbaAr8cl?4EDnwfL^i?05Pz;G8u_
z3FnfX)P}@>uK{YdYpu-=r~+LX6<tNFI9><-y9!^8iq-hJH6u+(HA^cqjt*(%m<f)f
zKk`e<r;TG4=y7Y_MqN0eN;dTIv?}`;*5d2r7l(28^d9Qt>~*bhCQ=bC6kPzxrrPyt
zXHt!Ktz%7RD7tQdxRcj_fO@K}yNQneI$)q#!*xYXdvS%WPn*rwXg(+JzxB@mRlO1O
zJnO#)&oB(6m{{hSWl>)7U=4bVQFC|=#r58^-E7p9xG9PilcJ<88d~Z*2%(>&D}s;n
z#4rr(hQwyOWtnER6+u<QIcw4CrJ=&*%61%S0(Z--LKGa6MZ1kjg95tXE}IQm2*G%4
ziqtDl&-a$k>~8eBD4JT;u|QPxk>bb@6UjSL+|X*FP0FDxxiF+bQD0FM*027BN+a-o
z)bTLrgfKSq*fbELHxh{bTt(7tn>Tf?p7)r7@mH-W>W!0BT64sEjcq9`rRe*w3r;k5
zI^b|s5z6Koxy`hiDa8UToQGP0rgmwy;X5C^G|I8AoYr!93OV9iCDLfF;ew%(V?c;j
z01M7H9X&65N}HigNZlYg;g5mh+`1*Unqk?$rM2zzV+1X-5|UrslZ4JMIY$W9{IssT
z5IU;YPbk>9ps5JNDPjk#lJaU1a>;V?`e?VIlJXdXzJ3*bUFUE?*6vvOfHt~z9p2WX
zEA6bk)Q7PNRm}Z8sKdUq&m`y^-%Hbb!eH%&uCpo>XlwB!)!}L+3mh{{5-8Tm%=3}g
zUjGE+E^zO`S-`xDI?k;}9Ll{JH;v1*q1MLZw_P)CJy%!vm};Rq&v7YuR7H2V+q2tl
zdFkOb!=PQ%vuDrw^s7&h8W_Y=a9pRCSe7^NVIlcSDZ*d|uT84?!pZ5g*fo8jfbZAa
z7Gv?EUEY~TST77cAhoc6{*2432gL1mJ>Ta5SATsaohOQ8<*PUv%Y5X@JH~N9#>!?G
z$Wk~QW)+8sa`J9&_RQ1Fb~o_w(MzPL@lD6$fszZ)pWiUg1qf*To-XvkHB$6QUYHJh
zWYKFdRpAIs-Diu_N3`acZusDxD(!lM-#L5@n3}M?*zs-O`3Lxh-~BC=MS#eB$oN62
z6misuq0<@v7aye*KKaQ{@~KaKn%!>0!}$T@7zqJ{f~Hz{V#vghaW3%lzwpz1{No?z
z$AA3C`Ht`SPQLEzzLD#D_rZI5jNs`<tI_PLn73la1uYb$Rx`m3{E_eaE*?C1z<>SU
z{5Sl}&;0xgjp4uUhayXLU6Z60#Dy1%3thOaKwd$!b1_K6iMQfVbY$!l<L&}0=iS;2
zYZRi_2Ecx9H8pM(K`2lmh{rkO2})R}sNU8zq|t)MRYO_K8&AC8iAG_GC+=u;YARhC
zrRbY1O$!_Kb!=zj+*1SX-HT{kKX1^HHrpRWeG^p#a%v`fcDLa0Nxk5L4^%TFE?Hwh
zQw-Fl7&_PYmyRG=QEun!6xpUtVy7zg$<KOSt$-dk8(OIZ;~R$<t?@Lg8i11Vix$Qn
zweGsTyQ3@%DJq{>_XAH1TKrYx`b9(Pn&^I%3^`n86%?hIw6H3TZZ8B@v^vK)j;pYv
zlww?SQ<>X2l&lT(W-}0jC)Z}YP32)OMLh;LL!gv`Lsh$-R^**9jDx+dUoB*+^((<U
zV$fK*(wc%CQps8U9=Dr;FeJh-C@4}3Ef>mOQN>a$&ei;IW9UZEx^~u5af5QMJ3q12
zssf21h<IG;5188I^mQkA-IuSfuDHIuZ+?$DLf2a(s<_eiV(>_*)TSP>tt_;X*J-2P
zr}})>rbg^77aEp3ZIlL&#@za79+w`RacM#DawN!rqv_8N0>R_sfb$UxhO;KGLd#}{
z23{(@Env1mb#2D1=_&KCYfbCNTQ30XLMzspN?W5@T5FV631&^8I!HTVPJP0pWt^xX
zrVAeDg*ZfYFe=)mipjsLs25PTmhd)GQ?8c6wBBh~A>Ne00wM|*yVlG|+R|}@a&4sc
zO6D{#=x{gIg^Cyy-c$er&iEAFUkf8w>xCBwA=v2CssEA_livg462<Shz4J^hQJvs?
zW*CLdZc8bJyBqb3s>XltE-DY?xbX7J_Yqf^mm|SPHrq?27F-aH^DTG#g=K+@iygb|
zj=_%{+k{jHRTxrWGeicTc=fd>l;&89-iDj#iGC!Pg&57!MB8{<nEHj+n|l@ZtQ-1i
zQ5QR*BHCd#ABf#D1?9PV5j6y#X3Fgi!}h`!AHg&~N*K59FF40$GqBt^yY>!h#pRj2
z9J$<WIo>T?+`CVyGqnmYKYoZq7qi>DTMqj@$9>_|B{OWc+}-UB8Q^5iV>zBU7)Q_L
zCa~-$=KY>h7K(!`m8puESv$RjEd9f7rS({+I_N(~ZyYW&`Ay@<H+;i;`POgwW_%1R
zwQx8d3DM)TC)dI_4h;5Ru15~jp8ee&^E@-nnxDEq92AYnnVaWN`S1rn#FN*b;a<Ap
zwPhhjH5%RSMpDwas@Waz{-~X{PqdrL&;9%@AN%-6`1?QlcX;3L{e8Umz2D5m<uy`4
zEn&K{uPG{K{ZjL8HJkNT<OlBv+b!SmzTeAcGw|pB+y9Oq|M7opsCxIKTW_g<4G*EU
zOpixYj-;I3_d5Dkdn?^6yqz00i#EW^e!q@B?epq*E90YnpYyT5n0{?Y_J!Ft9&B``
zs7r4ux>%~m(HW78Ac70}GjL)8G{Hx27B<~7L*p&;aVF0Tc;ye*rki#x8GJbNVPoae
ziP_BrA6Dc~o8EfX^BGctA*NP@*(-n{J3fsB&_a*SjWaQaA&N~J!G*JGo7Fx;nm_B^
zZM|`wv#6(a6^7U}XLoUdkCA1b^&DF|gfl<FX4llC9J6^kGO!^#$2`w?U>KBp9lbWp
zd6`uRR_i(%7ef<$P_$IXv0lX%wx1{uP8YvggxrfJr&BEzj})^Tt2CoC^+c%Mc3g{V
zo#qT|$AOxQja_y9A*iaoxxjW%TN=%vws^Io0*cavC@N}nbl*WwtW8C5B9x}u*6xeC
zMqh?Vbe>^Rp0IO5snrA{NdqCG`ss#Zrpu**6}!;{5^B>ENNJ$ds!krg;76N<V5v8|
z-G<F}qeY}UA=KuUxgx|4u{GE-ti4c!Q(ytQ(WU}i9ncmr-`j%uJ+GE<J+~VN?aZGq
zh*H6P=nsNdN?MHgFc{sy8Sv5lDYbC*P78Nhql!i^(v(i-orCrT_5l<rw5GmPU7^*B
z1d-G4ixjgcVD)_JE5338^&yB!$FH?iJeY&H681}}>Qrv#oon7OYTilJlhj96-g}M6
zX^K$#W?i~Ni+k_&<gOdPy<w*<^jZWuvzl?nGcXx{R|B0Cm_bSN^Z;|#E`POx5l1O7
z6*v@u7!j$ATi9Mi_J=)r)@~&PPfD=c>_~Cs?)E?}nR&@PefAnJJ-j9k(fCP??RI1s
zBM%-vV40L}vCIcvfBj81!;bA{OD&bnxS<oc-|>#O@%kIjIo!;M3+(5GdqZMg8k>zN
zt(x)^`li(dExrGL0n9ly{kms~wvWw<$eyp+tmYORQU%|HHfPIJ6kM?Hb>}}oJEt%V
zT9h}T1xFBpR-n#YU2J)yJs}MdSu)$*1{aLy5+nD*CDVLhd$A)81FygNgb)&sAHU4O
z>|l<^iDg-6G{$k@{`JUmcf;}ahMF_tVL>=l?OMHciYrRMKgTeq39H0KDzRBP;2mG~
z<zL1(e$zKmYvHxmK5cxK%rGXTRh~V6hIjD#>#y_Sk9?R9ee{F8@#Hn0J%7&Ou;+L@
zvMkH`^X~5Mc>44yhr^MJiyg;FW1jTblVW1K+p^njiCZxv@JbAY7Av!xndc)<p1j5{
z|MD;JQ$O|3_=A7&5A(%e{H4S&peZ|XwM)Q|h#o>OrQk(#g@bW*x0``)`KE8=-v8mh
z;?Mk<Kg*B)=s(a#pw#XbwH6CKX{U>|el>paZ`2`DsmmVc0<JZxl+~U}diIG`Vd8~O
z<f=78ZCV()u#Ou$GAOnQHM0UyQAfqOqy?nehn)QF0uWB<x`{<jVuy}7sFj5e-FO-Z
z%7GHpJ5E)!-9NBV^F*!_lHoHqZ>(sU78Pe`-iia8IH>=7M~Bb%05OXfwD41vutQp*
zp^E1o*Vvn*%tO(d00961Nkl<ZzD_n2mFhe@oC?00IX?Ih*Jd4e>Kl49;iMO}UF_G9
zNJB<JdEmiuvE8uQjQafO$svZof`Jc<a<{z5>VMcoXfm_k9|^%T4vBeQ*ltH!Z7j=T
z5owA{r6G~CKA+V_gPWvip^N%iyon-O!)#L2>ZfxBOVProi6)HHc7_C6(KV#NkUYy$
zRHUeIzfqSQ8x!z!FT?40)P}U<A5Ql|Exw?ogtJj_3PH2y&389=M?z8IWK*@W%+zI}
zIfvg2H0Ov-2qyb3S#y;^a31A7c2qG6)1*jZ$yp20S_}!bTyw8<D?((+lbLM~>$a`7
zIDr%aG@Z-tR$HffR+0T%)PKXNfq(P5S1C8w+oZxwo9&s^XIJkjPgln4v`(LuM)8J}
z133M?US6frIDd-s3l`NVG@T9#mJi-}y@puUYWcyP=h{Ehum5<<1z<H#rv9Lvlkc57
z5tV;w`Y^9%bgf(s9lPStcw9{d6kSDDr4eHw1V=6lwPei>G(L*=UTsyps->LwYc6E)
z!FXg!sBR+WM|A2=dNM*E5wAs!HTvDtM-OLYQ%BDp)P-6*9~{PAaGLO72v!Vsp*zyi
zJdijp6_>UYDsk`(o5)lqHWyo7dUU18-7um=UR@X-Ji2Co_ms^xam~Gzc$~}3IBxY_
zdSTc^He1hj@!Z`VI7|a$8YpdHNRero*^DDMH@Dn>=_STZV3{3t7E1LrHzJ~bSmP$q
z$<>|A-j83c@LOku{ocpAULOrj)W7W>fyx`JNNbe5D7UI_`t%3jw15M|VMDIEPzC3R
zHtL%1Zn(I5%$L05G1XUO@Qj;FoP#EXc~L}AoG@&ayAz{klHT4uU+o=smmAt$=3=}g
zcsSfX<M8x}a&DYssg0v>D(0;CIY0WTQ5C;+^U~Fdx@lGWp4Nm<e&#cL;Fmv$;Mh+y
z&u^dcu}?&vK6}cmuYQ_e`1zmXQ=k4M&z?PFyW0|nux`?wr_?!4ITxnss3<`SRzL(=
zH;)zJaG1E;@0G)}bzELuu-opaK9h4LgviXyIF3Ad`Z_=IcYc7M`PrZ5+rRzy^9R54
z5AopsBP&1@xv#b9LhS@EFt=<-pz*j`BS>UOBfsM-zmosmpZ?SQ*+2VV^LPJ0KdL#e
z=i(4&1ab3P`@$E1jd|LijVzV09+MX5jfUIU$f}8sgKYh4O?PykG_9Z;7whYqaWt|P
zQ_3&Y=2txH-%Bq5ZKu=cqBBfSYc5!E=X8{-2!%&{Bt$`)7GPuzysu@}jjm3#nrX%H
z)8(TSx_t?xm~34?6`Pzrx4k&k_o;}V*m$<5a%g_L?h_szZ_^E`tzz`_QQPT!h@5h*
zdqW_GU<Qv`yO=-W`qWv0c5+Iv>()U7D{?@e#mmdBod<2=rCR#OsG_N5o>t<2$tvau
zPNTg-2<ko|YTCSR^j5fa;%@h`>*mwLI1mR#v#_G!6cecD#EWMfM^<z2R!<7t5GX}Q
z^CF_k<z;3}i4Z(X)^nXwT>13Qo5;dBO365}Jy2})>K(BsZn)0BJQ<Gbzw6#yswO7o
zVq88eW<-R$<DSu}@1GkJk{hlzg6oz+h3!V0NI)q%DDvr?XF5(SOEy#OplbQ3MWCjF
ztjG$rm`S?r<?q>j5s$}vL}xXL0T%j`)H+PP0sGO7?i(Sf6sgLYV9E-=wQ+EusC2(i
z6##JVw5#cpd0MX&jPH$z@jF{HpIs}#oYU8pOp`^w3`I$r^!=|Goy-b)fyNgY0c@>N
zs5XkQ(=46);w+8BK1@@1xR!z+hP4*gQ`E_nm9!E=o7!)Pht^}(JT(`5G#?=|-xVta
zvBIE3#;cgX)Hkg)D=^)9QBV27r_Nl98NYUm-5ao*wyGj+Ax13%Pj#X*v*~OKuwc-$
zTB{(5(h|8jHrH3AA#nBZnzTuT0a|lRhl~@yM%p~Tc}{D>ZntF!4d+HKFYdAAiIftx
zHh%u+e}V0G!<T*em)ZEIkw(vh@g7(AuBeOf^vNx`s{GyIxxBh&yBoQ<&;|ba>qm$q
zOUaDsg2VKtAyAFn8bgX}EvAQ@#wwq4oc?lc4!l1Vb>dz?MEYhTr4WX(zxrDAtw5j?
z(soP5Qwv0Aqe!XTK6#xoWj5O@!nh-BuU6qn6QR@{xl}H9BiJbC(Ss{)Zf?1}+%V4z
zA?oIGeSN`nJn;PKbCPY2;xt2Z66N)sK&#J|5_J>?<^4E}I-rYtvxOtd>1$tq@`RuG
z>3@b#kxzc=<9z1TPw|OQeuA6l&&g%MIoNDQQc`sGcAxb=w(;G0yi}>m`&^bxjEQ+J
zh9G2p9&GcgwKC5KWZ$@XeqguTaPRtxi`|x47RzESIHzRy*IsLU|M&mD__?3^DgN*u
z{%+p;@BVHZ>jYI_x2mEC7ib$HA14}Z9oIOs*%^j`ulee)=D+;k{1<p1_`84a@0)CT
zy<yb;$*%nuviO7E{P&a!RnM+*W==U>wG^6b^yC!p#&yjAYjxyQ{5c1%rwV!X3RLmi
zVk2C$Kr#vqf)zq1`;@LAw!3W|T`4O<#TvosJ)%|<9w%y9<9fl-7WK9>UEI7J$;)g#
zoREg3_nFsq)T0|ut;HrGZgp6Y)+nMQf9O0Vdxj0nZzhjdah8cXZ0l^jh!uUFng}^7
z*L$946;p`uhSzo-MHYDx1$1Y>??*?$)<V>KHk*-BGRvYy*|lUHQ+rEr$e9qLP8(W9
zM1ASRgldjPGXbLhd0bVsh{k`Vlva;AA0p#U6DgX#C%ic=pz3Nt@KIB(YAFPBw1_Fu
zEV4~wDwSux%ri0Q=hQKySU`dCmht4OR1@zSF+|33<Zw7@6CERJ*o{qAdQy#Y0y|nK
z&Qm&<(K#&|4Nc58<#EpO?0Cn+t7}q7)M0>)$7PLU3$|A(r_-;<TCgH|u=K3M{-C~#
z&bW_sA*LQVH=9hEiMTYp6CRECxbPtDu-+HWiHWwlEsQ=S4rp?k0%8%4PKwqdrsUN+
zOz%|%g^b^4_iJq)q=Kub(f8SXccMFVZzdI&Hc^CMLR=dKM7i0ts`q#=cER3{8K3)}
zN&eQ>#?QkDxOc7{@^Pn4yQ4Q99kLJav94*JX-$d2&Vd({FzT$ssS0B!)GQbyK(^LM
zKB>uW=K*MX{kjl|@dwv&F8<URnlrh*VZ)*N8}{q_|7^d&-ff>!0UE7kQWTo^;Hts$
zS<#o^h(^9;{Vy@k4&sLG^%dbFF_)SBvpa?{F{TU3)VO)}4DW=^HZi8aJXfAQxy6ZQ
zIX-^*B|=oE`Tc&+em`+}d5v>PQ%Ztg3)~o0z4iE&`yBQ&pZfHt*zYH31EtOEE=I;-
z$8vzYZ_Kr@NgI~p$XPRur74+Q)xam7-|0R$|GlAmuK!s_3ToS<q-rs^P*G76eTYz(
zg}B{WhoyX?_3Yb*lEw}AsEAh8>l}jygCCx}#+&menZ@JGlWZJDwig#{3{|}z62rK$
zg}u?<_A*sqDV3#EZf>6A5#GFcifaXGAx6)#R3>R;E9vdi{c|5d@BM1HdLDW7Ifv8D
z?n58`D9@fhCm;9JJQJg17zU*kNTW)|PkJ75(Trecqvwu}w6m_=`^dPt<S;#_6iu&b
zS<@M?MM|2wj+9)P=RJqR%x*Vwb?<_4Ow4m54v|t9O3lp6k)Qp!pX8ID`~=_e?Z3z$
z_yd24hYueUye=g@EnwVkaIlV;5P_miToWB}ef977YX0>9_P^vfP5k&j`bUb+8e(+r
z75IfN0G;#1VM|D=Wc1E>vE~bE)d~%b(W=?4i?lODyxHejks&*RnKBP%k)+d*MVW85
z$gd=SEVJ1`ZCO>b&Y9JWBJ6$h7E#NT3IsoDBjg2|neO_IhH27)cOb!$qBl#Q2HsSm
zF|k_4h;q1EUKHW!LKumL9G!*$Is%oxf33xxj)Qemm0+8z?GgPRR;!#|%(k<lO?kl0
zxpjKf3r7VR<TNr5-uRMY8@?%_TU$G{QX99o2X^B~UKU*NB#VcX)?+DcQ*ZrV?7U0W
z$M(!v-#jHnN)fYXag4*T8a@xBHgF-O^;`xY$hlA!MS^Rs#FW@<x|gf0C}VJrVn1I~
ztgIpNfqB;Fwi(Fc%}zxd!_(Ms(}gAi+ufGq@rVbKiH}MtYOjDcuiK3_<MWcqmd)Hz
zTOWdYTtvT8qr`b`?{0bh>V`-6Uc%K%91|@#a?>$>iqSj-?MBg-i4eML1MDt#x*teo
zUS<Y2n1z|%4`s>BxggFnZbqP4v6k^}F_m?*Yoz;M_cnEK5-6@Wr+$A&;iVNMFF;lI
zSM#mZr@+u=L8%5k`p{*4^PGzvJF@BQKAvri-@rw@6lx84AMr8aF-tP$vvtqe?P5u3
z-mlYu(@**J9iNv15ZAk_7YVsqNS{^02+}lJK>L!0P2O?l_Xnt3d<ggy@y)uBst;}V
zQvlgOCcpC`ya<O{r`sX_TAM2K4EA=~Vx11B3%irPICrGaI>{BqeBsm5-Oq6WimZGO
zZ=-9K!&kEbaHbU2F4en$h^I=V`p73f^D5&#%@S?ZG3^^pCLTY$kGmL|j}u8EwZWA4
zlv$H1Zf+*VaYkqyk2mD9aDDF)w>L9C_cOo3xY@AVZMeMJa(T7Y2hJ7{MQRE!z4S64
z|LE&1^BrIG&Q}<Q$Y!@?I_z=cC=wBpDXj-W2;@>2hqPYz?#*PUX8rY~3O)T^MLDtx
zB|7@&Y{6IWr_Kk}MXpID;H`K!<N$Cfse%eKi;a++CaZcX%~hNf+M+&*lfd!KjzH<1
z7_BQ0DH6vGakC?Cb_|;>n;3Zd<SEPHj(Zndq&cQJbF4z<H2VFuj(-`?yW|0TPMQ>6
zfBg-HfZC2!sPm-HMC%@KcJC5!<QLSm-n#K-*QS)p-TuwBky&jNUbtW?!tcMfT9~Jz
z4aL=l%gY_L3N`6wQ;Us9+N=Dvzy4SF@P|Ll_k7R)oZt4_z6x-wZHRN8VY^lAyFo5$
zI2`PqH-T{+`O2^SO8$%g{6FXa^nd;{{L_E-Q>QWCTkLE<pNATd-&0$XT-VWVF?*I^
z=!vrhTBPb2$CS6;2Q{g+Nr&3DL@EnSvLT6{5OuSyb<wD^7Qm%5#II_jYZHzx0>Swe
zg>#ld<Pcge2o*05AsND8js#~tt=gLgWXXgfQVk`GA+G$hK0<A+BBic5?7=CWM-UtL
z+QdgtN`e^wDFnJ2Q=7fAB7b;IeduE?T{2GliT9cVzH<fV3sVt}S<i>Z849h}iE#QE
zU7!N2YYOX{rG?`%&s=VH%<}{cY=#Rh1W-f&(9ul{#YM-|l!eU@Nx@UvY1CM9(fz;&
zVj7GSi&8BHTkGxmc;{H=iTQZ6ra~Au8#M#A^{zJ2qA_?%Ekr+1OJ>QLc{XoB5eS|k
zMaxZAQ9`b5E!<LyIH%2-iU}HndCHcYDOF508`s*UHRVBhbJduaLa9O=!%D;2jEN95
zFL+)mv!Udldi7PD!2RoM+CFnT-K+(fbB6S#fb))}WCri;Ub1IqP#$b8nNo_5-hI`P
zuo;1mUeB+dOGWx>nW)R&L^r4F-^8!TvO|k`y*6qc*CFI{Z35mATvP-dziJM8BV&)x
zio|&r`W=00g!&;Mty1ztEt-Rjk6Ob7_3KTdlmY2YW2-NqiOy;k^MA5+BfO=x?JXC8
zOD?3KH_zu(qfRNqRp5h96zTI+Toc;7N9t^tn%E8Rlpi8)<u?U0ZPxf0Z_zL3S%;^e
z)o1Gp*(%x>UFVY7;Gm;=eLY^k`JZ>3(5^m$^fnp}Awtvct{BP%N{(I)q8)3;<WM8d
zS{>Cp#KUnuQgdZ+JMwZQ`G~vsfG4lM&c!a0HlB<77mR7(xYtqB^XEqn`&Tvc%{8@Y
z37%F1wK$eJb9b{4{J?a0PAQq=@yKqsV@QcHZ2-qGjMQ?(Nnn3_V47ztRXJVah;c_L
zPuGP&{rtL%em^I&f}8!E_VXmIQJryDzz_`&)b&B>C${tX$VI)Z#Qdj3P%e?x?+);q
zbG%q1--QR!d1CNLbhxVbRO_NdL(&y3XtXlX7GZI4z*BvshD7md<uD)jTx>_8ckD}H
zZmRY*9Ou`#Xa41nZ+{Qm|9JnCb|0k*wSp_Knm+M@6`wvwa^gfehZSGFFz-E6_v<=+
zo%`&!Za=~~a5{xjwDY*<uwS^m+%b-UaWf#JE}qUQPw%IG>L+>awb%LX@BX8F%QycX
z!sY_+qao@rY<D;pk=r|#X+lC}d$DC41`fv~DMa4;_3!0>{h$43{O|v(|05s&#K-y9
z?a&ynsD%D;T>THd8MG^5$7~vE#-U^L{#+;mYL-P#5aOD}Dy^FDpIBl@pp}_<IwCki
z+Tqh^Z$$5D9p!m7=yfWl=r_j6f6rT?p<`u`wMf(o=~Q4<bR^|t<VktT&+I>f3(k<0
z&imCOw00qat;54+qbMM#U*0OBSI!ts(Zt#oRs}5Bd#p&wkd#Ky*A?fTIVz~ZejG-;
z6;e2dP+6+dF)9Vj{$tHxc8<@^W7a)Ee{(YY(j%8jZHg@Q^Xm*{$)zAxz;-*9<Fb%E
z3?AY*P^B?MPe}ST4o8amEY_l;hiF{ARx2&5D!3Gl>vUc}=P-<<Q5)=<i&C9JAdVxg
z7LwEVwa<lVTC|B8BGX}Fw;4Nxp{^J984WK46y;M-%Sc|1Om)IJ<*Yj^Y*UOYZKW2C
zKGfz@Bk;uX*E^rBHuVw=-Z7@gykwpoW{|=guYa23-7_|uk=)G7QWQXmDQKaKPUBX@
z-?>vWV4#u@QE4O%sxv-wtyB?G8W_f&Ok;^#QYh6K2NRnzRvn>AZ;x~{5VTfsVhEpe
z)_gd<Cpn35JWjm;>j{Mo=4<JkqIl~$e_<rXs<Lmbn4ObyJRO)|wew*u`p_mh*lTs*
zsTa2DO?jHO{(6tMTmY73AsDp%;>L0styY2ee-daFFKWk9roCDuxK1v%idLnew5H<^
z=L5~;=P@N(EhZa3i@iA?*{)rm{hd3#7o9P;^<S-KT%q8O{<+h4cR26`q^mkHpCZ4K
zPdjl`_i*J*#Nbv0OfWRCi8+O+nQKacwltQh(yC{(^(^xp_pSm%f?AHyHeeiwDmpHy
z)zO+3ezhvEs<=$69U^Mn-t4)5eZke`mi^s<*Is>taU6O1<%g6-$hi;*1P0ygk9#`K
zR;rNPpbLNq&LD&H;%^=Cwtk+@i#pa3h8O~fJ&?eQUspT1+0&M#Ulbam`4c9Q?`bMH
zh%xmqquPUv=tw4by|KM*bc%ifoC+opO=}P!Bb4N6DRJ}chT{|%Q)J1Rk{1q3Wx@RX
z)|=z=gHR>25?BRetRMe{!ze}TJ=*F2=OVCbi}m_v+sJI+<kSGY_}boHP3#=^CaG3v
zEwjvfw!6f=YX@S24XdaxE&|VQp7NLe(qH6_*WTnizw<xf-t{$Alv{*%>o}OEhTY4&
z5PV=rs$_J|^S<}JkN?$w`k(TD`E!4cXV0EdKTqfP^L9vEIa|#IHCRuoW<=DJuWuBs
zS?;b2cD+PB#tZ=;(qIVyZp}1pXB|JT7fk|88blwX*-rI|LZ`o2iboSO%Qcm1S*Urz
z$)b}Gqt!SU%zI3yBfY`WO!Gpn>ei6PkvJ%Y!r^UG(sfTqTtXa%^=mW{B&%Pm(`)K*
zI^J~}fygRo!5YO3-mJO27?OClsd<MBk>pHqEJE<=+jcCFJ3q^4B67Y4d2hlEZTJ+>
z>9$?6I#G!8RB_zx4-Dx_NBll8ZbpuWBlB@qbkz2ZDu({j`QTbKZU?m>IV_7Nl+d8*
zHEJSE#|g(s>`{o$DNj6jd{pyjEfSS8+p-o3<G4W_)V!z{r>K`-u8s!}t~CuK78Qd~
z)go#srXCK2pn?sI$J-HhA0m0lhGeTJuV!1TNF&;P>76$_AT4Bv6jx<<zu$9S=sf6@
z0+-tnsa1{ngV5@%2%RlN!;lnRjZu+3RA^MCts`;gRBBP1uq+GGjE>+vLe;UhKe4Yv
zJhMLXUY}3b(`wA~w6Af7zT&-6j~#t_*vI|rT4`U~Rk!+ewnlff5__&}L+o7N(-dQc
z7uW9xr>-F}83)(u9%*WT{aF)8*AWa1Q1QK$`2rV!)dImgq4&g>(>E<^l^}>z9hVql
z<Lo3?D*$o9yx*K%lVA!*Q=;`zFE%nm%zw%g`odu_Z&yp%8#%ovQy9A!-j|))b*8S}
z4d*9#EgbrIDri2jbpeXF)K$#T1pprXJo>(EWbK2YS&kS3rI$JmHrtWSb|AO}WT26f
zvvH4h(9Pg$)(znd;-dRy>1g%p>WbP5FF$^TXYo0kaYG6l4%0%(N;EBtuuNIGWSIOJ
zhj)qGP6dHjXA;ezPtN{ZPqRH8T?H=oK0{I!7ro$?-jZuqoKiCTpKhCcy0KTnw|%ce
z7puPLm)a;*{kNQ5d*{1~TZ_ORGH-Xm`aspD9dU^RZyqLm4EU7TPe)v91P9f+WH~*9
zAm_h5`^$fY!+qB8b&TZ9hM-rP{dZ^Qz!p*`-GZgRLjS7MecYd=_2-<86i$afz5DJ*
zW$L(A9CK}yGE++B@-nl#P-UX`nsJ-L$dESt=#Tz=p1k>#Kl~s5N4)JFZ&x2suOp>#
zGvc`>9}cR;tBp7$yi<?8%Zm%X`@6rJPrdpofBw(^dF~Ex&Efv@c?gZ#vaLl*O^}{e
zfEG30901f6pc+Ct;6kFcV(Wia1L?FQx<Zi!B!ZE_Yt6Q?MDPPr7k%cX*jkvZNUW>S
z%{8YaPVrW-+XiI$q|#<$+A5OL)G*rn0cf^viBcw9&=HnxWIp&cq<+a+1s{%8rSC%^
zjf3SO>)KUIH&%w$?KgssrTclqHuZH$ZbEYl1Q4^sXlpa#tVmGzhLqF{KZFK7iDYd;
z+qxm2p2NQJ+KO)6@RjC61M$prCXRajTC2pEC`;z%`Ex>WkP-!90H%3av)^+ArkNBI
zDX0i4IGqx-D#WOwi4=4Uk>{BfG$p6DM$Vbt<s~r<bnbf3x^>Qx#t|PPWtqvjXi+-N
z?2i+h-38mLEAE~>!+HHV+wA~ADeC>$L{v~QEsWmc)Y{2bD8{JiB~g{ay*Z5b=ap{d
z%upFc>-^eOx3zeivyP>Gh>Rhs{Y%vBdi^>TL~8K`Dd~8<RW0UP)kLsX)p|>LlU=OV
znCGlx^$-X#fdK~5xbBD|$1x^S*qHT_?VGL#K6Hl>47F1`mKTs>+b{aFjr~h+B6PrL
zC}mS!v2l2m!)MJ<Da+Y9D~Kpc7~`PGq6iwbY1)LUR5)y(?G2B<r4u45R!4Q3%@^DT
zCe}j45nTicwq4n%S59mxCymf_yyv}+2bAw&{z_8FHIwngam1Sso~!C6uIrAR#=;Hd
zYV?LNM&&jLt1+~5`cU^JVU=;40`*kP38#Hhf8(9+Gz?z670^XXs&$pB;PcJYotpoq
z%r#SKvKG4?`Exje`rp}Tx`~qd@9q|&8D6#~q+w)qTZUcarN>v)Jd@{2u4+f3KTMQM
zSsPBQOOncboO$!jr(9oO5o6@~`g-M)3^m|5(BwdFg^G<1yb{EzhOAXFbU}Gry;|<q
z;*@Q8I#<mr-_whYT)iixQ44d?QBJj|)=Q)1LW;>Yu0>Hu>7;WTKN(TnTbCbFqR)kM
z5sl@GO-FNhLbHzFIRb^?goYW8T60i|IOAC6rs~-c9P@m{VM@i|b!=}NiQY(`{S^HA
zx-j{i-y5kNz_gD2-&`;J=nIk}H~*raF9_Iu`Y$@#+3TE_kRr137;}M=7s?`(?2vmA
z=O!Sq-CmH(5$75|{|i6O?cq89!5{iAzWghG8@=uE1kz?p^pWFqi-6@42R556IcHve
z{Fp!Wec#8g{PHjJLqGh(IFF%Vy11XCo4=7qlgjaUOG+C;NJ>v}JuA5&QV<{2TEkUp
z&5AC&!HirZ!H-A);y{yVjZDP(x;9L?XNeCR;<%%lQn2J%N$4r8D6RLHC!Ve~;v7g>
zH+d(O)hEzdiisWGYYf=h1Uru>(3&E=(AP@0qAu0UhpTx>CPS=xWLV2;)1^T3*?lA@
zP(be~TX&na*IrB<WU6OtVl^I45T|O^&^K(W-dZiHi1ki0kK>6yrVCU%xk6yPJvrUm
zJz5luUmGHa{hlFR;hf`e+}l_)Q9@xzK~1r%+K!YW?2kGMzJI0YtTPX~xjJeSZtsrZ
z962{~gl*C!lPqe$Jd6XaRkgqJR#YwvK19-HgHI7@aNO&AX-&AjRcgTXgZsLsn(%R+
zGY{@vA|i~V`u5dYttj@V^WIS{RM9#l%ljr6uT&Ai;Cm#T7CjF2>-1vGP3P@rlvR`v
zyf!j;M@pK|<wS)}r3%hBh9TnCz1BwS^P)xC*(A@0PT5i~!`c?+Y*Bl7jeARja=24e
zRle7_H=G)&_Zgw~5)N6*olbvnD~;x4BUB9~cWZN}1y^q*v<McXEHsPt!zoWNMy(CH
zVjT>DCE|P}$QGq#^gLr#{Du(CPDbzD6F$|!lrs>^{!e`hzrY3H=poq+P{r$*!0Gno
z5prvM;zoH*t`vN$CkeBzW$Fs$6>4s_U^x+jM5Gi-nL&gx42FCKbH~;b+@#Q)a`Ia2
zMTrHpJImsgGeFkd@K1<gUwk+rI~{HG+FhPH%aek#m9ip`-g$Gux12<?Co#&Kj!>Nn
ztn_U=q&kw1q#+<h3B;`luf2H-Xk(AR<53%Nu5xv6!z&LS^5&=Cz>Sx@`sP!P$4rq-
zm4I`qEb1LzYmV1nd&;sD9z3{O{jy4}Jb7|MDTRj*FCmBquP8kq@1f?ZNCy~kTF;&I
zbJ?Nlo)N^_R+}Y{ULP+Wagh*5HGwuHu6h*$xu)?tdfpq3Vz1x3cRSxs|J64&4R>Y(
zgg_I`rEMtk;JqiI{0(t}7cIsMp5T&hJPw4&oIS@X%%XQeSu&Tqkx~lBDN~xJn5>7>
zQA91DzagcZ{dYRr=X}lQ{`qrD$Y1@df34$lWi7wjXhE9iaHuTHo{|;8O>yAv?iQ%H
z%X_48;M1>ug8%6+|3$v-+rFLO{ky-Jaop+R-Won7VjWiHGIl>j9|&XQi{JGwzVA=`
zNq*@AzraU7_Ayd$CZJd^_iy|mqK%1H&Q5bMEk{xs6xC}LX&EVvWu9qeAtc983L#FM
zAPS&n7H(-dBeiJLW|NAXXF?i?VM|DfDvo@3j?7twIOe_9{pHlo!dnW$K$C!X8N^X?
zHUG(AzH}LB6W)#BbUZES0=y=RILBwo4QWUWDV^nUD-x%p-brgKok8h%tz_C6V$>%o
z(i%1Q&uEkk?=(5e&?BipRRiX*aVYsOSo7AcXL^ECZ$5Qw9D-Prs)($Q-KMHjLUh`q
z0=#}t%c-HVuMIJAoEF@mKd-e)E0xWt>Qar8Y&LrHM08YFYbE+<sZVMO96d}qGi5_U
zjguQg)abOi7-vqWM$=(hr)QdwA#8VPO|iJWy<wSVwwD(i4ikCFJb3Vc?RKO5LZ>Fm
zS~Q#?r$LLNGH0FGSRpcm&`GH}eX36FuMEM|sJ5hM$xPmzQ?+yhBzL+%U5nDC78R&D
zRUV&;^lrr6i+3@VxlV7o7jH<B={U0qkx~oC<D@*{aZno^-$*e~im)sV9~y(Dk@SM!
zImhS|?|bxhG|Z}rPNgtiK;OiiO&k$QD<=m9sVkcACE4C0M&H%_TdlW@w&B!QP>jnf
zH3N~;YNdzrf(QFMn;eL@A~4&ev=`Ev&n<4yWTjTrO1++qu-9liuhf9IaJ+RDa5b+f
z!3<xOi=xPv@-W{23E0mHUFvMy_<_=mLajse>Zskv#vNi-7+U#b;+%OoiB;$-6_-V!
z>+No<UQ-Uefq@R*^lRcQzqG3**DreEgBc#OE+%$aZ4px+A5*^7ig@Ehh-P)JFC&Ux
z=(r-R1o~DR1jmvK+lwt_nV=Sgf>cLLTWTwK$qZxV_V$j;i^sH5*gtul%ZHx9H(q-C
zBeKh6gs1Z}8VlyFVCq`9d49w3cx1cXGEI~EHZ9tLUf+MfvnRJ4_Qep5Z2Xvv4`N-J
zjTc23-nvL?EKRfX^Wr#DA4(DB3`q#aKhhKH`XVi6UR;Wvzo48-*FKBi>0}-M_7Sjh
zL3$x?LXc)7Cm*y)py9|gA`pmpL>v+njjVa*FweMQ!|`ySw8n#%u8A@5{P~R*jqSx!
z$anSCZ?U7REkU`X&u?zE@#@dCG(+4qBlXFXC;ZS4{Se2)#J7CQzsJ4n`(V#}8V8nX
zS}7f!d@kbHY<IlxeedIse$Ri*|Lwp3?^%|t4aw(>t$$+=vHN++vntv{P(pWJXjKuM
zS}N0V0x9?mc`3v=D3{HvT}y}qjak+1t<n=5a7Oy}4O$~aV|iL}HXkQy)?8~9-05b4
z7n~?{#`O+UoJMO75ovC<ooPj*H+{$mK`n~>NjOq#<)7SsK$99=w^PvxOkOe}Mz!y1
zS_Js+sn^;WUacIb+ALev2b+x4mDlO)eD<hCS|iW17Gxpme>S~o7Ot+NM;*ae`#$A0
zYSFlscA|Sr=mHl@ZR$<C{<bSMd*PjFUWgYv9FFODRPmm7gfQUjYe0GEWyz;CuECNk
zLu@u0E<^`IP{G@50+Yj(DbX`Toz4*9u%DIVCu$>8Tj8Y#4@iT?rp<>V)A2|eN9K89
znrFt%Xt8yHyv(GO$jd^>LMhEg<w-pRYh})L73qYaBUl9oRB`NUV_Ftk)6r{Q)cdZj
zYrdX`6a#bjM6@Z8Rw4Q|rD%bo8OY|>TdE<jiqu~2RD9@zqqd5Zfb*W+#f3Qq6t>$f
ztqFIxds<b*5D_zr_T-}AKq-w9)uM{^+6zfEfhR^?<GY~Zyv{(>l|cEHd0mGbJ(0u4
z<p7xVLPsa9m^iNu`bfAVs9KY@)27%e;`;j7aGH9InHD3>oPvEIq=XBDB8oo3GC@_2
zWv$)zN2~$*)#yC`YR6mW04}qR8)~({Xoq+u^Gao%QTJj^$DyYY124?Bbrh=eSVwJT
z^=b>Lhopnf@*7^fnnN~GWsRsKx{d=*6@Tj}pPV*su?}S|GWAWK{c;+|tuX8g53+W-
zeGJ;XSQ3bNdxgHhXnE8+bdGgOqGRg)-I14Heu;5YIZiX1P2%$MlH*~aP1;q*Fd(fm
zj6j`nE#oBf@TE(9^gMp}7>NN_Do>wkWYso}#2C1^+!3O_kKM(Vc~Vp1H(q}O?}eM2
zCmi;IBdPVnVBM_4>=62md^WzZh2h)=W<9^_$u!kch_n!hDd}riao3B%RZ*CXKY`i+
zokmIhXWD&aKhKAFh7zI(Nok4%$B1*9HxThG;wdf>a5xDp4BQ+J<oU>csnlGl&hhZ!
zBZe{Y^yw4kDdQ|g@cdYP#D0q%0%onGe=bG1y*-fgVxDIK=OV3nN^Weox~Tu)5B(s5
z<9+Y@HpXFNc0X<{B%0ZE<eV9XEklB<t1G_ud%u^z_xJxE|KunC8QpxnbAo>34~KJV
zRZ?xOaTQV;pf$6o$ck(^2tJwtvv&Cwm_n<t%o#ECrLPW(NO_z~w5HZ1wKQs;b<9?p
znH5)Rj*Oc@6B&GjkO)4i00Yr+lvKPIHX+g^*y!A>Qe>fJ^DAwFiyFtK7G36l)|x77
zkrk!%=-AbrH+-VsqpCKG;9MK#*6rlh;eZOAjElSSnMDf(fn{ItX~3rqga|$u+UM5I
zJD19mdjY0vDy|DMY)kQ{Cbjd>WqoE}Y~NbzTe$vyS+el}gBB5D1!7h7ZyXcSgi!Sy
zHW8+!5`AVIL^)~R;j3_5G)uRMqg37GOKpsUqqNL0j2w;&r6_`S|Na%@W}}VuT$ztY
zh9Pl09&yfFeB+L2RDj?eEp%%d;W*D+>^2M=wVRpenP$aV93m;kHKR6l^|Fof^E4TH
zr^s5Xg~{yLLR0_0(6{XM?AiWm=SBrDR&@8#b1#(6CUC9l>$RrExr>3zt4r=~Z`IT~
z24LWLoT;TU3<*TP$E;yjj$Ett>{$|8&ES)<n%(;U>x8lDnxl4E8L1iPBHnV)b?-QF
z906;5*PrmhM6nl%&Yv1B&VKfKp5*k^CXKYAfU|LPfH2^~78gcE0m}qF8y%}ss-l%1
zZwT=zsN&aqyyXILdh<=>-UHk=;a#BF99u+Hi4u&|FA9^w3Dr{`U>7Z5sNt!a?;mbi
z7KULsopAd;Z=PeluxzEO50-bU1OECt9Rlvf+zJU>VR|-SvU<me%VtSp91`IqE?8|}
z#L-HnxwiU0jWL>6q2MeFu$#cfppCv0Pf8n>W#Q`HmU)>8A@a)G9uuNxI!ru$^M?K1
zk?m$nj7=AX-3IZE%{KA;<~djH9zx^gmtSJL+i`VuPgPw*WEc}U>j+&fc`Db}*W4YR
z@Y-iy<LMhu*^aL;jxXUuVH^UjWFqSH9(-7(=gvVl!CnoA`?=GL40ip4cSIKmA(6d$
zI_d(^A1Z82pk0%Ooja<WV(0W)v3i!BnjKk-MFSIU)MCn45i4Tt+K2-eXf9I4^Tx9~
zre#5#FgL+R$D@~DCJh5`yzvJ6{Xy@Iv(fi&^<ypbTTgn>^SvyEH{W=sNXXKd=fd^9
zW(8YiyWR0O{`y}-TI1Wl<2#5kDZL{`sA`L#&zq;zY{@#Bjr_;|_#fw&e))rnHu8nL
z5Z(vI-KFNCMo+8i0jQ{_38IV~p3*>!k!1O#8XT{v0yw7Wj!rH<Z8i{-CJMDyDf5xJ
z9B|CE)|m4`E*bB1R9=d5=Ul6}s$v|Z(Y*3Br4@pAgh9mx9gL7xC^b_`rj(gwSqLE#
zQoyyU)RjErT_gBJSJ1CsuK>;wgDN<|06b}@5tZ3L^}?(xm3=^wujUnDYC<1dt^hz2
zie#dXx;B_nxj*YU7p5iSW2DX#(iC*)K7C;r*PL0YXBozQy<RuVQ&G~-zoJEOb9W>~
z7{&{oPV}ZjuvpTd*S}<fseG5JQ>_>yr_>3kq9QXNB6CsRpie<jIfR@2je6+~kxR2i
z%6TEh#57Mxtt`3FxsNfLmm(VD2bv=tY)uyQXN5C9WbJ#FSD(O&Ff9w?ID&(Do*B|W
z1dh{$*7K1R!<x7vIHLD##z70bCd_$Jp@nyhgI+K1y{X8p=nQf7j}1YKLCo?-$HT>x
z>!N1;+YS5u0q;~eQcLC7geB{Ew2IO}VzTjQulThh1XN!DeXo#p({Md$!)P2vPB-MQ
zgAAB5Ub(kTREgd1W#q(d?e`Fz#@^1X@Ln(*2eWBPIIr|0q#?fGTv>mv4<jLN4gK@l
z$W^_Un(OH`+6=`IP2=D{tLXe!Ki+Zyc=zc~QeS$+5C>J4MMt-EY0{XE3vV0@DXZ1S
z=cj5|uhY7f>x-`BWM?yZugM^-u6&se{wP^gf3J0(ZroMW+*v!8F7aM}g7Q52lW$*h
zs_D%xz^e3KG<L2)HHEMSIr|mOR8nU_#4&8fbwOVjQBeYMaH@vB*zI`qj%%9C+}=Lr
z`oRTvw+kQn@F%&tx8dQ#E57V2zKCTmJbCjCKJ}?rd3gVli^mVi;)uHq_kxCG3;JH(
z{<g=27z{ntuHOeG-X4z=4_>;5NMm=orG8p@9VvNy%tTl4-jU~pi@2p7Z}s2odaRIl
zZPpcc<J*2>o#k(O>m;=Z_SVfeWbGD3bW1E*)qEiYvz)+)x5h2B`tn5R)n8-iZtDua
zOY7N{4yX*#c&Vu35fQR@p6?g#jz>Zaniev6UV8k9Gz>g@_ME%hyEV40)-Pn)@vraE
zg&6@yEy7d_PoCUpJX5Ji!xY0;s3CgdZ~nmFK&X7%`@S7Nj5t$=s|AJYR(LqD`$qV#
z@A@u&;0J$@zxB8O?h6RQ7wT{>u-)9F;hE<>txbp&Ek1lyBo2saAjFM6H_qcj!nvsF
z9nxrJ)-2U%Bk+=mL2yCuXB7ryJi0+NV5ynpBjN%jJ6d$Qsn$kVGNguhM_H7ESc6}S
zKrz8xDMiPHB~z9o)8S4V>EsDcm|9`Uldf@RO5Ky6pDH!ekd(Y{qLnoSKu+toHrbX^
zW6^4ADWZ+2lM^pXnkco$Qb#(>6hfeh`oIc8Udk%w=r%68+&8Ema$Z%w-lj2q3Ge^E
zCYr#CPN~P5XPOs~#)Ioiybm0wBi<?GRBKypRC+<FYFQO8^hr^S#+!^zeNV+R6^%*t
zQ6t%CI(h_QNTy^3?rv|@&MRnhu`IJHdBZ>}MN?g(H}urj{WG8>=M$d|r{`>5GOf84
z4Gl38?DxZ{qvqS2JH~OOw8q`tky0B%VoXudscQbO_I#!xQCk;5VqSP|ZQQNJw{fM0
z4><3MA?o^&bi`Q}<-o-uAyuu1Qi>dplc9%j*dLYCJtT7!&|TcSlUFdJNbDN3NN-#e
zr0Aa=@mjRTK`mg4!&O5raVqSzDQ2mY30#!pxr*|ftoM!HPfBy}UagwcGN~fDt#wSd
zaSCw&7t{>DHcDJ5%cAKnVZ`|nslu`x@f1Q3aGs2=6GIAw81U{)%<_ds0H1y2L)_kf
ziTlGyym~+uZ{4tv^THdqGw=TdXI1(Mg6{~NohI!#t<vt0YAb!bq~8}@k0bL+@OKX9
zPQ`+J)|k;QlfIr6Cs^^}?Ivke#R7URyH}m^4jb(RLD~Bh>EjX|m8>16Ek>mle6ffm
z>mGvlx`<No%$k#k<9N4_M&Y=h7&n((UhKHNyW{DTXB-cE9=|fN+l{>Kiyk5EF`xX@
zC#cKJyWaDq_!RIlXt5asF4i>(0CAe54CJi&qTB5yufO(&{`bb491b(r*AH|-@tNH=
zu}l+QJV7*(K#AGtd9uQ+RgJ_tskX2JT(6lAYSbK@dSk6wqAw_hwL4P|hIZG|+UhCR
z|6Ll0H&bY^WEJHE_0Au9@z-eH;k+U&&N(CuNS$b{acshq{gD`hWzaYtJ$k@)x8?cG
zo~KWrK_7SZx4AY3|MoaRy?Q?S*hi!=9W$-n0T$h5<o9Y8Kdm)>;0OLD-h1BnEx(uT
z?t+>#yNgTip5ItE3q2L4Hpe@@=pB5|_k0ik<R|_qPo6zH<K_Iu9<64kn>TQA%aC>q
zoBNc!Fdy#lb<&2YHOA4cs9&RLhIuXcI6_SL3USylw~B9qmjEsjgPH=DTC8d88K+Sl
z5D4U@QQVC8iS4)nQ9f(HvCM_)bWB~AO3gFDLy190_U2IF99nFZid09-I^qo_P+jHc
z>t}dvxiw$2a`I6tl|0WVw{C;9N;zhGc0-@L+W5HAsXe+m>zYw>JK1ve>5F|$E6sS7
zm~bTH2aoH|v<S5-aX<RNyvzu)8WSTRwL;L5apx%*PQ4a^t*Nb(zJ<=c)UT_xaj2Ee
zsNSl#cXw<y8yi(N#?7cAx-2-al$OvcbIX(_EV;t2e{`?Ke5nk<k<BAAOI?lV#VM~g
zMYTD5<INk=;J6<GLSvbg5|pCH4MA1U(s04Esnw9;3FjiybVqH%WCb9CPNZrh7i(h8
z7(9f)Zo5Tne;bB@i^~i4w|j<|*o?|ETe23Y<B(J^*Gwqsz_rSlA~`F_(kUQaF`kP~
z%|g_SaH)-@6mVP85Ls+LQUo$<m_$luhCOgROia^690tZFOw)|aj$zPfy{=I99n^7#
zIIk2lp|yoRb(B_-=vE~;Uaf1Q1Bq(TO|7(&70_rE$&(6|qQiNkDKxELRC(=n|HZAs
ztY&qCL-6qonRBWXc5Lxsz=fy^VrfWGyD`^1G-!FE%{R1|@Ql_AdA#f1iQcFV$^&-p
zUpANOS3llz0r=(zf0_H2_ZaV8lVZe$dw@!?bD7UnDqZ0JSxb0tE6)Fl4!pWG`J<2I
zI*hj>PY&xoZ4}(POE#<Va4?lsM;jfA944O(j@B-rk0aMNbcR5I6?v9U(d>QHR@Zmi
zzophnm5EfF5^b$J?kd;d7hIjGIWWd+Aan2XC0>7H&$!Jze0Yz$+XJ_^2k!1>#!X_H
zkKDU=pLc)Bm$1LRW3$^L4oX{~nFp73jA9Bi?MMSDZMeI;<Mr2{@}UoXm}#0xX~X5!
zHQvD{ZHRGTnkOipF$_jLJt)f1N9COOTD5i)*3;?ozN4Z+J^3uwrWb2fPK8}Zf#-Q%
z5njw7r`1X@MDd*a*Sos@L;Lu~b~%Oaf!MpU5X~aPQz|^Wy-^g>IdZK$eEBgK7dsBe
ziRaIrX|v&VAvwLne|voHp*MdvrD-Rb=cV%O*>gk7#k{-<*Vor9i=&i<zxr4I3X&_|
z@$G+r5Ch9}WV_pu7xm4mwGx9@Z!)R;`@jG9^H;wA`}yI&{UcwnPup5!KI{oYV-C+5
z#tVXnJRd3Z0q@j0BhQHtM}$DyRH9b23{>J~i)&uv7@en@d8`jnz25Sod`dYroVBVo
z)G+J_@yK-C<K4_KT&Pc|2zgq-3d*zz2%crmP$nFe)^>;}p*(n1@>WyL<CBWn02PgO
zq{!PkTBtcwmW;GUOv)jv=33qn{+>VE5Zjzg;OC5EXAQmR$+UvRtxdFm2?hir#7#$N
zjPv9JMVeYsnTtS7i7J-OW0Rj!3eow~<fWaa2J7T!rJ$TPo%8qA#IfHW7*k+~32Dv5
zFG($=q9Z0Tn-4|1>NF#0K6i|rpXD^Q!V4*?iF_%Z;2TvJMVy1Bh>l)s;odN^Ec#me
z!$fO>3nR<%$lyJW%4V~nEHgb-Wtt|WHFnz#lV#(UR!JdJ%8?;Oq>7Goedl89{#U9B
z`Gz6E)g{OMo{QZ^uZ8P~aVDjxlQw}6jh7kX>RT!*0j5*AVhq#<qY0KIHYCF|PYh{C
z@J`PkkYZ$-XO7dtTnhK@T`>-cWnMV$Rd5wTQZ8&$L@cGCak6&Zm4>8#mic%jk3ms)
z98^tAa?KhJME4fwl^R4>)Shkz<pOT;A}Rv%o>nuhE{a~0aF)Z~r&iEtby3c2Yl!Q%
zGWu*)5nNsD`PXS^E3>_~O075Ky5L$>|3c+kAWeOS(=g&<LL?()wrQfL*5g}R+um{k
zc=O4R^NHOBo0nhV-ezPo4#ZJCla?}5n|2b8&Ye3)=FShd{=f1*Y<%m?i^`zo)l<rG
zI!T5ecC)n5q0yE1aR!z4o5udo)odclN9&NWoE3kr|J-%N;qj{-OkZG5pGZWsE3ZNt
z2D&+<6TwmOzK~pFZWSr2l#O8^`kCY1#4<l;cQNwt;eEE-k<I3U-Q|e$5C}x)nU6=r
zgDe^+6gLCWK}g!^ElXy9cgK?_Pq@3gV_DQ>b{rGCiyb~BV%jjKz~Oe{ZhurSu-OrZ
z4S9Y>@JCX-IGsCD1)eIXWW{c=Ya4<}{53-};>Bs`o7V-g*k-T_gG897$@nOz>(zxh
z&Y=r)Yi0>yC~_a`*!PHa0}43Ub+FCOwTigF!SXXV!^qt{asA+$%gd3R7oI<R#&n!?
zWY>M?=mr1Tb^o^qpWUsAI1FT=i@0uYZ;c~bC`C#2_wU~)rpVo0<*)tKzshFZ@h#u_
ztvaf+5luD!u`Zx$wej}1zn$;>-tXlf{o{Y42<;c*6FW0?#xYL^;_(?RFDia(jfx^)
zd9M20m&TB<Gz+#AhTPax4V71NjK-FgQZ)L_DR;})Mp+gdj%-G_o6U~h_JY#6&dwQ6
z%j2YRxI3tgmN;D8uo<?LbmVZ@lb=3kTnZ_TfQkrwNJa@Ufn8O8vWjuM6@)@@x)H5;
zyiPY<YNMeMWfH_jyy|7w$|)jjZJ)#vPOJ!ApQm+N>@DF){nn&FI~-I<=rta#Sut5^
zw+=$rqBSZD)=82Y#A-`!jd!zTlGdjOoQlHp<{lB|rLc)`|K25?`lG67x2^{vMSO`!
zRZEu`m7a8*3*!)p>D1>A7WHE$Dl~yC9Wga7buV&@@a*{!=lyD`yxdOIR=L`Yj6>A>
zLn>*Aiqy?BDJsP#ACBu3ritEP03n#~X{~H_J4B2+<tvgE2~8=%Zp$)Fgh2t96x9wZ
zMa}I+)B#`^%?N(bA`x-KK}Yww6diFkr%6-6n)_T>vZ|6dgW0I8^sY!Qg}d8{k{d&c
zwhRmNvS?26YCBZbzi===&QhpeKf}fkOos!$IMT3D^J<$qX~bnPM7q*d=pq<0?M~Dr
zuZ5rfohqk`XpV1_QyPxayRy*|3OP@-Y{j`hnKbtt?~*5tlFN=(G&9;cNAlh%7uD>N
zA{DiS)5*wSrtCt_Gc_N<1T^#E<}IylZ(Rj^n0fWXA7*&;h%dRkV7S=fhD4cEntp#1
z_?CA$e&h{jH0To!DC?iCQ?tM60Dd!mgBTfK$GUDs|5gQRN2sJ5Gxq`zSbw!Qe72Xb
z-Jtz!=he`<iJ1Ms|Bt<Y|Fvw(?)$LMn2)vA-shaEy03n*yZLH~fJD-?V^VS?o0Q@p
zwi4uL#Rw1}0s`fSApefzzyaa_iDLw`;79?CNVa2%A|*rAD`K;o4>j4h+4p^`>YTmz
zTJxD7#+YlL>TW_}-`gz*(gNz%t*UeO*?T?a9OFB_4@>rV<y{UMg6uzY7b!Gs1fhRl
zn?30E9-Zq51u2Oro!H^|had6QkG#Qrn23|7<e8p@`{yf-=iEJAXg%?!yh)6KnpO3r
zX%_->By`BN@ciXVUOv3!<^2QqFCSEK*fjI^$=y@qd<+F9oY%y5I^$(R@YD>8p9nN^
zzGt4mc`dA4Q~$tWUbz>(U8msADaR#5Z$y4)7CGG>W<>?AWn+h0C~4znna7#cLfphc
z<MqWB1LS(OhKg{%dS*|fe;!ssPJIV)oRjiVw!AVO7H*$Dp|!&Oi+j$eixrjQK4gF8
z??SBqr9HlT{5rfdd+WP`u=md8QmFO8#=vclX*%8<S>~CCcF%wQU;HvppFQQL|HRMG
z0Vkf^l3I&&rBehS$))g1zw}G|oBzhY!C(Kc{%hTof5@L)oM*ndBTP4RamwM_E(8&X
z6Q!!fOHP@xY4O#hbC4&bEi2DwyWqTIS#Ah^QdFa<9g{P}Gx$iAjC71|W;QB7oMsM(
zBc*g&cciRb!j>!N^U7^lv>10YVZO!B8|(Rj?fJ$$8#?1I#7QGNrE9)+O^Gg;y$67f
zm`YJJNi5mJdClAHup>zwbo5%aNz)oil~}!L@u8GCeMU>N*K%X@s&*i~!-%Epz2QX1
zu;PQR8C`uzOB$9PLfQF#X6e;TX4Csl_0jaC&af}i=mw?xfZ+vP8$G?3mFuU-)XFqz
zu~U0fZl{CI-p80I1@l+URf7oBJ5EJ-N?T|Y*9c1_<%GvmYBis?Mu>h-yHGo;f)9?g
zB}IuvIIY<dk1}^R2SkL2lk(JVZf<z~@{DtVV3uP;VZU0^hc~`DXbMNlnQ5MO;G<KS
z7cH97wt}78X$lNGFEQ|Bh+#@rLD67Jd+&(Tft(B5rZIQ5G^Oyg#$`*&wL8qjXk5vz
zV|gdaJALuu#5_-k6V{w;%&DTnDG0}9Vp%5Q94WO>vI&r!<9~4b6@2rQs_rDUWH3dy
zuMxLhbWhS7PMaZ#RHyri8aQWaX*eHsvQ+g(N$Et|&V;xyAvp9pwVEh3QA=XGJW$ID
zU6Fe|ubo;Gy+^vUCdw4Obiw&)N6DQ8#NG?7+6X#=RB}F3(wSVgeczMb$oXF|0@#T3
z^UAwl{~E8o{u)njZxD${kN7?_-`?@rTTpdW$J+wa?V%K!DuEvX3**5W3KuP^w67eB
zu*b!p)&;Udj16|q-zzWL`)&!d3q1I4rKb~B7y~q?9F1P<&L7ZtGm90f9b8vqV<&iY
z(QD$_vT)l(3qS8^E+GykpsX85cjVMM-Ur@$@q**cftFXcwc*?WCyso{v>W*7!;g9P
z+B24A#(U5C{J`bq8R*0ic<0^s_~tjiNiCgey2Wu|iO&dbp;frN6gDSx3L##{-Cz9T
zk1)*#?w_Ce@SP7>e5G#BfwLBjPOnp~MaRFTC>6pYSez&iDFjD|8u#NQSTcb|*>$n&
z-I`hs3;5QNTyUij50g3nIJ-_2dZBA{g0=*Gm<HR3wrL_vmB#Qr6t09^X%(8M%VbtF
zogzXK;j?c(!#Uw}pLn=msl_ZQ#=U2^lbV$45c*#}<CpJse8+<#*CwiWf$s4VD7BKd
z#se#(^J+%q?=<3E_~y62!GHGu_)mHAzq;eczW8JGy%R2q=~p2V@OVD+=4bf}|K7jH
zum0+<ab9)7`(g3{wVCrAj!%g5k@dxUX38uq%S>%r(4Dy_Urwa;p4t+g3NDhfW_{0x
z15=#nP9xc-5OfS>Rvx<9dGm6c5z*wDh87d^a>S*Iw{c`G74d<??X4nIws-h&AWXNg
zR(f}AHBd{UmNQ-|hh<`!XBA=9s@d#|@OR7G`)BszC@IlvWs1sEL}X81=vbrHiVC@^
zdc8T3ef(O>HIr5lbAf=?<f_I**VEQPd{CsS>zJ-v5i>rkqD2m0&A*k=q>1sujdxlK
zK14%LjOT4=3hpZW*ilHZky1ClYUknMfoY!5DPm(<78?<F{roMmrH$x4!8uYbrj9Md
z=<qJ?7G)lYUX&Nu9MK6mj}Nbk*aW0%V|20oO$HU9(3%#s!B1RPwFKI7<?eP?ibBdv
z^Gqp8BRn^yR9v=AmG=gUv|4b}jPup(XB?-8lc}fG8#V+_TGe|_>68Y{IEQa$NbXdG
zgoq-Lb~>9Cj-{w?Vw0*(T~j{y>AX_2x=qYez{zlM&~zIds`U1r)*x8is`Ek!LJYz@
zMZ!EGMQp-g7HxRMYqoB!im>%w!Ou2HDw@e{+|^)a=Yyx(xyG%!xaey&E#S8GL6zUZ
zP{am#J5$nyS~p5ntDvz@wSB)rk-s(?25Xk+_&7sXUq6u!U3s*nB{VuttRO6uZDYN>
zpaavSlr37_ZMnYR;}aKvAjA&o<LA8dTffEa)7Nlup&p+SU1XXk{{64OU;FQb^B?*e
zYLdKLtBlbe+IflHXs=A4WUvIU#hf9zTvflaKm7D=3yq<L)|c}5WgkO7Xy@-mAQ!fr
zHILXJKUHCMT2cFjK5o?g1s%0%r#?-=RB%zN`skS}Y9QM>FFtzC7ryY87M{G|`G+TJ
z6?%*G>NwwLUVr02behrl&ENlhVvM~0`s>WI6?mujkbcW?IS||ddQh8$K5QboIk*Q}
z)%d12-g?4vj4V^+#RrMWH6~ZNq?atgF->Mo(L2SGFRCRt3~*H!0?mr_!S9R0SS)qK
zG)BAK7Kw&yjhr?cPq-_>+S{(29!G1PCMN88L@alHG!A{ge*3kzUkXTc^WdsHd-FBs
zdFFCH^YZ?IEv0=F<QQwN(A}%Z^S?U2`}j^1(mik4&S5RpEOpe(cPQwN?m+Mpzw<l4
z#sB?3`%n4b{H4Fl-R)Cd%tS|ft~(?;-2htU&;8P$<0pRn|IBax*4IfX8w&qJx;DL0
z(#GX<#`yzCq}4**vRXelRoX9y8_MOL`LNJRC7mDe^9{W#k4JiEJ)gOdn2xu&5b-_|
zMaM}4Inh}3$P_0fh@08=G9N(HnqrAF#RY1}2u+iBIx5z15W+&7)zH_?PuX&!UOuAt
zjT#QH+~7qxhC8<POqEV40jeo`ePCJ~)%a$J+eeA5HyE7hT2!dg#uP^Xz7|DNdhZMa
zxpBW}>{c-chcVIVM#5JwxmFu~(kxuH0MR51FQz8OOrWh1t!6!=6KGxQ%^h2hLE(|{
zqyvnECu{*wpT~8}2p;bP$rWxa`&w;Q93Ic<AsIbMk;{N*YnfCk(j6{%D9}Ym;A5<L
zvuIv#pB3G#MXg@6U{G;Gt-96*qs}zErij$eWz$ir6D@?A<)C|~w4}5pTCJ=PXY$D7
z>L#+WqPNtn9JJm5N2`shWF2x(7GtUu4C%6SxB+|&)SSs#0|SCr5zstG$|K>|nDoUs
za)n`;wc!!eJi8PXlhoEZtw{^?ePY&_yhgHxpvdMtO-ysd2NfDd5wcG@a`jHpx#b4c
zd3Hja5i!dtQL$R<-H<;)S|c|VVGLktB;NR-hRm&)2c&3bJTyvG@lC6lR+5gTae6<8
zU}84*bJkm?Y(w>~H=7T^c!|ol^wz-Srp2W=+2inXIvM3e?<eI-7o=Bv7QX-ElP7_I
z=C0<(`p)~j^R=(x=Y{2qUqo*22)?tM+KYPH$e8BorM?zo&RGG-o(3|)n(KFvcb&+i
zF7R5#X~WVr!?vpus#)>3TMl4~I9o8rP;ws|+yA$2KZj0A8dv7c>qk1Y+BJaP#=(9>
z6#*Sbs*UrOxH(1}C2L$zZ;jw9<a_3L$NAxb%Zr&RnYZ40jcwg{xKGd{t$0o^SGM(i
z9`2v>xi7s16yA8_GaQy1)=S~-x8LP*>f9dQqU265!T1gXW!2A6>ToLW+`aaM*FW==
z!*t}u2QOLgUvh|@vb{i9IUFYo-9|UNTK6J)9J=wTKb4NKf+?rj$2qlKo#<Pum?0P<
zos>4Bw;WB|&{fS_uh*kniT`M)-w|RFRsB}&ni;(VvYT>uL-0~g9ByxT^7JWN+PIuo
z&g*4YGTZvkFfSeXzjge)e|7w!$77Bt7=k|H%~Iivj*w!EYM$)iFbk8P_y_;!AMjWH
z$}jU5|AYS}z6*Zx+&?_<+SAw2=q)8%fB8qh%)j+-{b_#l>t8d@Zr}IxAN1k8=P=Kt
z^`85j@HB!`MCWvVP#cd1xPXr{x6eLHPMPhpLa8bsXqEL;Sl4^j0yob#!n_b%;Bc7n
z4%?JFdAr6t)zaxr$DPhcrZ}VCmYy&Lq<eZz(2G5%h2gf~#nGF?g&9BHvV}><k<-jN
z&xjX(HUE9Kd8PJBD>KoWUT5Bb1W#EnCfLxGjlq$oUsc6p^xM^&u9j*!(gIY?YxRm~
zN-v{c+?xe$)^v;<LO{G%UYXPJsCtF!G^Taqm0P1JcFzrK5+A)a++u5yt@DQJ55Ax7
zF(wQc=esGg=AA~s-7R<ao;51Wi`TqfQG;n8JRt=7n%J`PzDw&!sk>E;M#Xu=JDj(H
zxLN&V8m*}pV5ypBlrx%t>;nTuYsJJ!xoB||BEdV>wUdj!FPcUX_7+K{l4@ZJ9*3c^
zLC87LG49|rbGe);DN|ixTN_WG-cpNkcWb-Gwpr2@T?;A|{<M89&RNBp+q$yzD+Shd
zV_7DR=}Vc`_4@X%rrYPsiWvPM<)now4&Q`n_5>$H@5~H(SkI^lvh&VR&uoRFH~yn?
zDm5iW$IFgJ%ZUI?^MdnHv#nD`qygCq{D?SxzapI~iabxtjW!s*?<}NgUhq!$$B_gg
z&Z0uAO|($vK}{M78K(n29u$2P^~CJPPjpVGPA84t@^nj0lyo+P-79UQm_P_#3Agxu
zj!%vNc8BksrGh)1`RI3l2cqLMp>y*iKLRBa9Q@Lc3IEsMaO`;)V*x;6Z<_TC*4+&x
z#s{xZ?xSK+Z%7*p4!FEq7g?MAs?=7grS2n%om0kj)8|p4>6}-SX|d{i<Yl6~q?|{_
z;D{c3#As_e4(ncvk!FR*mNTta<3Fl*osvUU;$O-Go*Q1g_YrS=@iRCleC9LH$mK)U
zi^1A`qSe(l!_MI_G0&0vmoItyoA2`c`31*BEea#IR@LsH1L)1Di=C?G?(S}Q<Bcb@
z)_D2ifsfw%7*{K_6Y26H$4ULEKo{j~A?ON|V-MA}_zNNK4<CE)r^sDx%8^ksOXc^7
zwL&cy%^cO+fMK9{^|HbidAoMD`3P+<0LHKJwS%mU{roP%mJ==no;<nJSgS4Za5_=U
z@6UmavQc6-eJ}q0)#&%1eO!}#^v@dAbrn~YTsfUqMDEdN>4xYdEf)^+E&t8m`fvE+
zkA9h7{C9qlYI&=9OFAxXN+Foy#Gn83f1Y3dD}RL#KKu~);S_%Wf@3}$kd!FfhFCE>
z#{;D_(t0Ab3x1jiVa88)bUYuv{|@KVJ>F$2kc87Yb54!-Uz}NPZ<xGrTn-%ONVH<2
zwMwm*-DaodqBUL9tl4QflGc@63iCWsHbs&|JaO{;&iQRV&T0)(s!;FW=R>#g)$Mly
zcA5mh+BX|Mccx`xA}Z%r1WKY;v$&{gr(x7SJ^!^*^Dz21K>*!c70foIUu{BmE3Ocr
z6iOZWwRWDZpwq8+CiswD_z^-NSM?2z!IQ0_^+vcKT;W=at>f{P^+0V*bYWDwiA^W0
ziWFPtHFw;+XG;Os5a%s+ZPR??rkuCM1Z>Xf`zp0@URUni0Zfe1fe<}0PJ8BVEmnlu
zltD!rdaqUQNTpGGC$~z79#==~rUDbWRJPP9)nr8;z^PqOX{4eekefLoXkq9Da!MR;
zW^zi<#e6$eQ7XXc{>0(17?RiMJ)5PM-w|9tMWkw8atLkD>fY8ww)-K3fcK749Hmy{
zo(eT(-KXpp(+Ef%k#~anK<p?0mOyF0Z|68oOj9IIW}0s4R3m95gkULPP+OstWM(0<
z3o$f}XQS7IIH6W&;~sEM(?fbKnyh0eLTj1U3OQYn1}+?#rUhKHhC<Ux)DvH+QIv!a
zfoWbqqVCryog(@O-Vx^mhvQSiG*fe;Zs+~;igV1<QTI9N)Si*5D60>iWxCaCzn)2#
z2ZXCv=l6Sj;sVe)&rD|#K~m=V`|tB@X~@ljUn0lXW}Ltm?yTdo2i`M0_SiS`$A=oo
z29;YUB}~*(sb*HJ9)4Cxv}@7ViY`iOK~XKO!bVeLX7u>y4Mfp-EBds!^giHX+^dUe
zisUV;MBcl-TlHRa1Z-o`-gO+@%!sp=My;7CM!*?=uBj5(3#Ed{%q15-diObRe$LaS
z@#bfq^5I7h+`n9vCs7*WBJaKP0neU2<^6X*=38&SN7@?F<Ib@VOl%^$A=K8bV*<qS
z`s=s6`PMUz$A$ZsXWsvV_h8E$1TLo+1QIvLnGitSc>j;gta-Kgi_wrj?~TWxh5E=~
zEVXF9a~~(i7?_4sOd;q<*RJ`9=Nm|Dw_<UiJ#LKt)mjkm1d(eoIR0IcLm}15!+POx
zcZav!%!h{uwk=sRU;-9&bE<|~4CL@%R`l;Ww*T%D_^QO0y;vXCANzH)6Q{9Gkha1D
z7ebhbbHKX?o<4cXd41s0FZ`AN;+Of#SALSufBy3<aVBj^6`-xEiK_Sf+|T_SKmHRx
z&ifyHXrss<&eiF3E~V(kZ|HRCI_5gwJi$9*y{wdVCBzx;W_-9|IV4_wd{5cVx=yst
z!$aa>Z9Hrl`S=CP?0I^3$J5&z<Ix5V%Z!NdU|w`ig!lUc-YvA6i6!HM@ONJRW8QBs
z)PkioeDXx^_%OfZ(l$$NXw(g^vryd}-H_UeaitB>al`hh>AKZ&+M6<QT8ygCW2M>}
zJ2-d)8M)pAHf?DtG|&zA>d-*%_F8TCg+pxPZk>`6E=FoGG)#0f=$tBtb&b<?NSvqV
zY)eevy+p5~LGvIT>xY2DKjPY5{XT2eT-{n!-7L-%XGhM7oDwNzN-0cBG+#UQa1>iV
z^i*Ech<uE=81OMLY{r~d-&5x+lLSKVc$ug<vtoJ2T{QKAM)g9#(-2a%sLl?TH8F)D
z)DUVeG&GDLc$mz@I9q;kNgJhPrZ|!FMJYn4sAidGQcloUqx0ygsa36Toby&ttJp8M
z4IcuvW=g4Ki|D(W_K)UJSl6Wc`Rv#XP#Ao>==HvJh70LE2|=1RJeHyC3>hAtw7Lg~
zA#Md#_Z9<6hUy!7&eVLe>!EyDESb9GGbL-YHTj9?1C>TAiBc}qwCeNh9buY5%-~#{
z(jU54=K}Gj()ZvZF&-7967X>%R**0e=NrOw(CgZ4bS+wVcJa8VIrbIsmPFEPC3;6#
zZV(q(#SFDu*`MR@_xQvGptQcj+VqB$%K5{OdH1(}m&5HHPu_Y9fTwNZhR8W>l&r9C
z2#)BrU<fiQZyR3(Pe}!FLQaX`9KBWwW5nck|GrwaNGUmMu4@^Fn1&4Qh~MZ0#|G8+
z9lrKL>(`T6d_i(H^_$}`PekX;w@!A`OTGBkEoC);?amo;q&KkC0wFTg!P<ORg7?f*
zCtn^|riuI0%E#}0$ZMZ@#?5i2v`UxG<*e}Wyc{_{6h3(O!bcxI$H_#?h7#JXFyMDD
zN~m{U(M|7#Cr@s8{k11Z>%9Evo{!#pPFoXqF|fY)0M|BdZzqn6jfzU4mTp8_EqKRh
zC79()2>Jl1SmatVjZLGr4$fPrI}!r4V5U{2>9nzE4@Y;sE{(o(9CjZ-cx3p=UI>nH
z{m%U}ionbBneH9S&4Jn~DJ8bc23GueLo<~hs!>*L^a-Ez;rRdK!#P8K+Ld}S3cbe#
z;IK|m)LgwBN4_d(bd%N>o%OQu(%mynI#QdaiDg-+CG-1le}iBC&wrV}^uPW~Eb#y+
zS`lp16yUR;`7FQi^S{8a|N5_ME-^oxYgB9He1BGy!q$~u3N>#`^Niqe1aem4N4=aV
zlOx1Pm=@@j;KdSg3g`5KVC&jSV_qD$Zb5?M9iH?1m-n<#N-*aLPWXCxkM079qdOh_
z>>J!z*NWh~MK_8TjuP3M<?&oG3L!(igl>MQ#nKfDEhXGeY3cgxoHyHetPwQ6w%R{=
z{oY#WbxdoNW9BWbskf#<pyN6977N#v^9Lj3UgO-xSlBs-ct<Jfxf+7U)rK_LIbzZ}
zx#T^EmLfRcYz%8n3OWNGyTF(%Jieb51#Cu{(IPNJMUu9ZG`&LS%0yE_5AQ@p;sT|1
zQfVy1969*i#)I*>i#AKqBK7L<gG+Yq1QrwAl$?o&NXm_^T45NJL+C}D>y!%4HCqey
zvrQo&O?jZjY=s0!d4n#jIa6Eb=AeexIVTRw%(||K2svj$@M?k$=G|xLms2zT){4<~
zI*1AZozo((B;^VYBludc`MT3Qky6^xV&{a(Ys?v_z;Br+rsxUL;iHK*f`to+pk97T
zEmGQ#BH1B$MWpMPBnd7qI<EI#sRU9eb)%taKF%Kq(}MU(sSl-6i}AzVL<rN~bc%R2
z2$!awff6*oS3Ih3(&uz^LgIul&4lTI_^9LmbRk_HXz2lh8bkw>jwS@f8Sgx`t~5X4
ze8h{!#D)2W0|DtO^7$mkCocdo$-T3}!+RuGp1=JL?=4Tbc~f26znG7_-WE=M<>QZE
z&>I|Y7H)25gTK?R-VGtLtsA{5Pt1v<`AV-^?Dd@#Ty><Ei*}<sPeqHA5id66??a?9
zIzkwL;5FuL{NAgYM~=?b2TzP1aWg3;=o&|+3SU+4<zi#!uKYiML-dGw*(sS;?+r2X
zt(VGSf^&M#(LW_WB;MOjy#CqG@Z|1>o8v9te&-!d4~g}h`S61iP6DmWn!AkONu#A^
z<s*rjTZ`|^OW^j&foD&jFi#V$G}0yW?%VI+TjM4O>-~F3Kk@X*#O+NW1fh3z^k3IA
zt=H?g70la7Z7aOj4Crz6&K}XTtF1+LU#9`PbtBed=XfM}sISzxCfDe{X77hSVu`FE
zl>r>s=HVP81}c@pdD~c?JfV9>N{Q3MgO07NU^JH-{TV3tZ>WC%Wi9~6v#@h+-L(+k
z$<4j*MPh4B{l5CROnpaF2R)#-&U#5a|M0nHRmVW1GEFn|&foskzr|1gi9g9N{@cHV
z_nviKIr^hMKlUE}%%Ax){3rjH|AhPdhaDCEK_9hLUcPwFvpao;)^%gMKjE4p^E6E~
zxNI3&JU%Fq%XQ(&-Rp4sln?{cG*g@B`OWY0{^bK|_k7E}&-3X$=J>V>t^_kU)`G<)
ztjLOpsL}Hn*(n`kAXrY3A}g}gd()YJ-w<F7YR!I?8OG{r*;J?{tCv|)(_76h_i78$
zmFiMk(Soh(n7>J*<ZDV$D;E9cyqWi2fx*GC)%Rh<9#O8SPE??ms{BUha6|E2wPB7i
zk}5c-qsHDE>!um%iXz&yqE$-CIGZNye8IkdavobejRINkDHZss7o)H&6D4QX^F=rJ
zu0C{oG#pyUwW;bF)>Nz!g0s=5WzG&tQAhcJJ<uph*)hu-5gqyU0AUKm;0P|ziV%FK
zuL<eK5hGCDH7~f<M#^f~Jp~<MchtMDc2t>b&um>cQmGc}*qOXG4yCF;<oSHTd(Y{z
zl5695m@L=Vl(5ZiBoVWx>9+Q|{d}ZSDXk$@3pxcqydkUhUc{8$A~=U4fzBgV1bQ>p
z_J;m>?^Qu=Xt6WT;F-Yd;YObtAFP?_xDb(PHru0(-t}C0iRwHsJtGEH-G_Jt%fN2k
zL-&MuWS*a@s=GG2^Z5B_F_<1)(Hf|CTnxmxsKrys*BV0vKh4DXxF^BXTBt=4<ytB+
zF1mLj%*Px2v}o<oG)mKZ%@MAxXyY(1NO$CR-Fv^E;}aKv9sw^@r{UKPB_HQg;{D(K
zCWp_jx}mp9ER8rtq7PhBV!dqmpjHavhw8Gkm5rPVB`0bw+}++%)~wxPvjT7+ewH-}
z3MIQKT+N8ZUAs+q)CHP-#aNYWv(wisyG2U3zZ0`2Xtm<J+e5ocEvR(^wB{*Ixf@%n
zqzpkkhY)e&dq?+YXqWO&8ZpA5S5hi0^AlcPDj$67Lk`a>@es-D%3H6$fe5_u`ZG@F
zRVD3J#|+KV-b3hwpvC5LIMRCMOF#NKYAeLxS<f5yAHU>se<rlf-Q>8OKZN=*PwoQ8
zgYq_7ZFCVzE+*$6qZ=5h9buR~+d1B2{j4MFrsN&A0J~|oZA8tCEV1*PbLPBm)YgbM
zH~amiBZWaa9}Bf^T0;da*42;k(y;$IuPYKfVVTLTux=}t%eIS<yf~a~1l<&Xx&f>G
z)4$Be<E+bp^gSL4598+X|N1?4T}+NYzP6f$(y;SswUbifd^zJM9eX<Qtm}op_E-NZ
zfBeUPiZ6cYixg|_u#LqB&rkpKpX7^Q{32id>eqA=_+eClqQ>ekU%X(RXVPV*T~>UD
zmI^*bYHs9nV(akq$r~)kgR0k~@KHPA<HUb?`iES4=K0+N_cs@m?;MCcrHfEU+#xo?
zwEMaDu2FB88v<cJx6Yy{73fgY+VPC^s*UTId?>Q?NEy2$g6tL?QX8%}YE#>%V$B?k
zNrE<=6-9M(%IKu4!x(w`>m<dEw~X3dbwl=sEsMAgO*9X6s2_WWuJ>Fi<?3W0y2kiH
zDM79UY+%pkRoj#XmODP|rF3H8b}GvF{;QgO=LPL!YOx0A{{E!rII8F0As$~Dk4AlU
zwk=t~)%Pip7JZH0hsbdV;l$Ld!9<0{3aCPyg1Rt>u20pjgO|p%sLyJxU3tvS(Z#dX
zqET;iRPm*E<Xra@ja)Rjr`2xx%BnQ3Rn4*y$973fQJ>{f8;Ed!f1>2V;V|QzBJ(ju
zZKOr1Hq{EQI7LRf3x-fvqjR;|cY&Bv2NX=0qE~dib}RgJq>t?xVis#-va|2y1KDnk
zyRh(Mw_ha0JHm7%Z!1m;5)Sw<>t0n>YI;eLRfSw(#?MbsF5kG9nr3>(8-1RfM{~S8
z;PCN?k2iQ6umYOiQAvn*ct5Kr=yqX<u|iz%{s=y5A!-v&Cr!m)t!ndEZITiKK~Pn-
zA?CFvootE&F@MRfqv}s`d~yVEbXQ_Z)2<wdKxF;+#QVRu?Thp;p5NmCcK8;@n<FV{
z7g<Upm&%$dHEXfeYEz{ZaC<z^ip9aTPB!=K5#8pzXFkkocIkuJGPr%=1oj)T*J_UJ
zqi}$s)CpGD#26Kpu9cDs&Ti~|M2R97Zi<?Ak}9}HZIFs5HP1O$Y8RHEsU5Up54<Xq
z>7uAvGgt9mnPS8D!s%4FS>9spp3{dfslMYpJh{1HS!Pa`m%RDbQ?{*;Q>9j2Xx{(e
zeSYLezJS8#f%AFe+wZ*32k*V$cwBh@`E#g+mN%jk4$-l_e4m;==E<Gouz+`Z?RpnV
zsjTZ}MIVN?$*aYm+vVN+n0<%=27=U@jXkjF9xD__;n!+qT`xSGF6uIG;|9jaO7Ai)
zm`&b4c!JmEdMJ2h-|TX!q*_^SZYYK@oKI(s*FxO|9~YwHWk>#o^7n@|V*Kypu@}>(
zPTLLHAN{)%9nU)B`BiI(YQ3sC2RGGOQJ`F*&gtQd1jl@sn3rOkq42x^^mq6h|Mg$z
zFaE{<Kz%TC-jij7&YN$%$xr>%Pw|_-`88XVfAEKLN+(iMFR`9Fy@-+17vkZD`6kl*
zhltDcsq^1$zsdXMB`@mA@8$PxBXL?3hE8Bow8A`48!*gJ1?Zxd5Y^H{R77$`x;ta6
zI{Z>~4N?~-MO(!D*PI(xSwjSeJ%4pK#_hf0dPS9%R(#TF1xhaaXtf#VPn;%^bgrTe
zZ3b-}*w3DBR8Ew#Q*9*I?LBH8=L{{h*K1u--&{$h5GGYj*3$OIxmWd48~cy<s)8NJ
z>0Wd-RfLK~nJL0o_x0%&dXVhzn2ux1jr&XHvK6*1aaax}1nf*P+M1+mV2JwuoH<!^
zgj5=mvc|DFcXg<+^QOhF?N_~1jN%fPNvA8pt2NHBLPNMUq;Bvgw@t^isi^JN&0!{Z
ztd-E`t(br>#=sObcCdCmZ_#TpovQ*OtqXZwS*B_CF%%KjE!q07*W1Z346jxwdupwg
z{-Hd*Jq4tjV}?z6oM@DvrhBMhF?fdyVp9cOFGOHOfnLWwx^9naIfO<X#}x0FkI$fc
zno*y+1X|OgN#Zj?JmBO^D-X1q@NNQW)V$J)CUxKsgmARNOU<sGbKoIhsHHe0cv{g>
zc}{07PTPg*@^yR+%+o^c9=gTS30mZa2yp^spwqoV;Lc6^xZQ^Xv`pU4&@(P9_@LbA
zyq&4#4>#BQ=Q%!k0q7f{dn5~ufQY9^C)KW~*~6+Ze{;-r;RzG<W}&3S>2%@pu<~%)
zNEK2tza4?w+ljlo1Hmb`qNGepXPgmGy>s|^QH4?n<`sy|blDe&QH(%WewRfbu@3;Q
zH<)&aq8&icVz87#E=qD$|Eo?(1@9bloLD8Z<wh!y5YDxcI(Si`M<FOru&V`3vrb}g
zcchO_UXV#xsGQe#3I51!Jd%1~+u*&o-{bDtYs@z)CUM^J+M73Y9P74nxJ*2GcI3kk
zKjwVi*e(^R!uk1$`moVbVhONJp7s2K`_qTaZspnC#NiNe=D|FW(b|f2`^tYA8;zR6
zUW2~{<6&4K*m|Xssi-xC;54Snx~n#LN2SqA=iz+e;r>KPmDzjdFh(XqtDWZi&dG62
z#apdwWYxeKWXyKF(h;OL)>5c3;^&2)EB%sKw@eiXUD3bXI~6su_BdKEYAeR$4eW<d
z2HKi7#&crhl}`Ub|KwNR!Z^HVCxB}#k|J<vP0Z2T#tERJf!cIKsC$E?qNFiGZm7|3
zj1eSIijYqg7c+-L!tub(?G0&5%wgtl{K|jH&;Oa9=f}VN6-DtYP#}hxWm@>R|LmXT
z|M^${Dy`Qa5&-B3{y@)#Ry<*OLOAH>ONW`Sh7SqT!r$M1hxgM5_?HvKT1~_uBUJ#@
z^p)MVZ>OM)k@-GKN3&&X)AdTk@6V<4x=wnhyuE0)CyFxZBm_}DErb6i1EFoBbkt{Z
zynZY6v_Y?CsicVb7!P;by!ZGl?{>wf+H_f5ml$26n?OsyK$|zQ99*<m@zKcdv>hp{
ztr`hmpG%BqU0Ww@NvA({URu-Uccgv@5PYDbCdbM%*6ZZBqNzJgrC$rk-gO%WEwW23
zT)OA7X6ERbqbFsJz<m1j3Bjv(V~A=zycfw%i{nx&r;7<6Mz%6G+Luw}Ul*Hdh;Ka1
zyA?|74e=h=F)JFi;wq&m2it>N0xoyfw6e^B<D|uJF%|GIYM-ac(li{!3zvKqNQw(=
zrBZX&eW-Uf^ZCQBb#Ow;nOPzpjeHYvoX@Lr^4!%OLr1;B94F|)d0jcLRVOZfG?Uf1
zdJb<EE+cIuT3&IrXJI@i&W(wl-odTu9_qZaqQ6n{s$BTG>67c!d$*MeUeP#l5o*A-
z0^;yaNAc3o#>45o(P*@m@zT{A#(}g<TTf&s&?_~sl<iDSD}tx^sz%f{Ztp<OnUv22
zCwLJm8cIM{FF%F8y;|kac2SD>g(==>Q`Vs7Gfo1*Pnxb$%PaJ&@8|gBHZV@de$F;6
zMFF8$mlKBZ<CeKO3qMhB`A1>n$&(xAX=2Njwbb3}K)U02h&;JHAW}3&r3uqCF->=>
zKK1&xhjOfoQ`;4@0?;CX{Q^9Ok_V54p+GiN>S`4*P(khP!W8i_P>MR<XX6@<g*pTe
ze&T$eIp@UIm0jZLAWe-M!DjwpPZ;Su+VJhaoz4_hu(7Tuw#$V$-4Ww~R5l(y`hb_u
zb@80$iH~`%YOUZo_yaHAe_*{&Tuv(`H9`|6>C9B-gOJx3oId`TUN77(!tG&UGI{>h
zVg)5rx1ybCGc>`tDIr8H0Ef?>Z8BIv%h29*U{61FI^yVk7iZKhbAI{2<@7*Gl_h$P
zha(|G%>f-9`aoa1mP3O`+=?L;eTYGHRTXsO+Dd`xK%vu0Wjn95Vq7v)h*jx?Nlkk-
zeoh!dogezKpELWzA9lz;`%Au`V{kmZzoNV`Mtd$g!9{}T^@qXl(?%-iL@8=^TXwsp
zs{#0<Qb$omKhav@a!%a8JORz@gfs*5o%g@ZU;BUkHU4-1+yA{Gy+X?s#1VbqXMg%n
za(8#j#~;7=bWg$`c5ZxRx_g6{>kIDt1s@~d>L2p;^&9jR9^4t(JWU!kcwFm-c-n?#
zr10^)ds~-AA<~uPrd+t9R1ieSwNb1;bxu{dr4(A~`)ITG&eS?n@CLqBTEy1Ak+9&H
zW5Cg^!O>z-lO9C<+;X;!#;$9t7Hn_VjnPP}>b?#>w!s{=b*k7X&8~JXZNE)}SLJ?f
zO)a5XXL9=W;$-KQwQ5Cg>#BYXI(}=-W;b0?`w_Dhouin5OMq$eq@@0=nj}*3-J#YU
zHX^tCP|bJu?Lq&|AX2erUBppw<N{SH^Aw2QD=Orjs*k&J)pdW-T-}XKvxjPUVrpSz
zlo|Vx;`MWiHBsW62}ZD#DM#;R-z=K>u{!5=RK2&v%@T;gk~P3a*lNQSV_=>FBUwc`
zbahAegICd_l)iIyM}(asuZ6fwx;NBH&uW(w!Wd6#I*o8u{bwtdHOJZEq=D;r=Q%8s
zf&r~5@^|G4?9)5v9j-Z5P>(&+idBOM_jRV03R9()6RlkAUbS&7g0tAU?fenjxiTH@
zkZ>TzJN=o`iE%>hH$5R7@O~m^5U=%s7LtX!-P7}V2PtYP<do@E)r_jDMlEHmP%UXw
z_W<wd&O^`Cve|VE;LK1PrO*xTyn?=Zr?{Y|)JmaIy@d0TI9=&d-_P;MZD2-cWoVm@
zd<-$dX*UXNmkX!oFY#aK;Jc1bLS(L;xfJ~A!bx3~i5})j^9!9ljAL_VUp%jhB5^im
zwAE(FJN5yhb+po)w%)Wca~WTu-(lS$GD60iA%d#t^Dz*c3KiOLYmb3rJn(RNN!hmR
zDV0acXdC^gg+YKWW^6rLLH(9tBpn3&;)n&#+e_B-OTu#Gc(|jsiBvo3#lw|bWdz%k
zK}pd$ZYNbxwYrhl=iIN)p`DpR<#6jbEHmB(?aZ4>^1CTGTdO<4xmjkR76r47;>^3N
z50qh4FFGdgjh%Nh{G9ZJDso4M)5DqPAHBd)S(eC?n<K~RpnTFe>B3u+DT9?<u45d#
z2do>m7dB<KcHg;HigesEvDM0!GwZh5`_V<KbdAyS-YYer*@oGF;CSCFqifKH#TyFf
z2J-m>eh=UIe!N%L{%M+sF<QEdqJ7Rt?7dxiO(Gh#<ZMh_Yf`mfSAHK{g(HvogV*8^
zwk`4S0P_+#pI2^f4%F6|;>_RqJHN&+{@GvR=YIBIC#8)!F2<E{eDRB4;tOB+0?%K(
zp#Koant$~hXZ~^h$9&K};@jzcTva7vcXr=bjk#3v`eaS1ZSsS+WNXmir;f>lf^r3|
zr)5XF#xQeOxd@t4piNM{Do;fyStvn~FbrkWNGbI!Yqc>?fh9(~>qPHN{XQbpY!qs1
zR82|qi*qq&0rp~ExqidU-jR!rQwM6JqB${-KK6~hb{nG^r)~%%^jvgnKD^zJZvT0$
zHM()?G`j4HI1bd@J%X=;cW4NiuC1+AdeLHK*pMk2D!bR<pqexj>Wzt!+b0o?>L5&B
zRk=mSbH{k3ceq@XzT$@>9*3_(kYjOsXjZ!BDUa7iks`E!?p+Z~1kG12%B5BmU9Z8i
zEPBou7F~^Hp0sG}jcUl86Hkc3mK#r=-0Il7={d|vN3|gYf>)766wR&|=jc>JEJt0e
z9Cr0ZY}A(NK%6F~II*QG+NC32yXK`TLg(C`n4&1HdLovb$y-uUW7P@LaNTgCBiDU0
zXYWI~<qgN+2=1k{Spzwyc#BwL+HIfhy;EAEm5nF|;$-PqqI9g@11=s2lV>_S(`gV|
zaEhp9PueoQS7HFiro~^r?vLHBEzT3eyn{+HPCD|%fF)7E3%w(Lq^D%O^TaS&a+Ru(
zTgyAp<-BN^Ln)e&;{z^CpJ;9SbVMK`G*N`7N#Ac)b(PnxY+E;8eEghWy!;sdWcrBI
ztKIY+dc{$3y%+>tXf-p1sGJvZ3cmveHK8Zcx`%!9?)yT!AEWp)j*%Z7Z@WS+JAZ9(
zocb8P84A_LVxzA!P1F9`-s!?JPXT{Ux-em0Q&{?i*`L%_)$hrBMJyG=)-HUEf(3e4
zdz8cEi9tA4So4YVM;`+R=H-UtG}Ezhi;Z800HU^y?coD1c}2(wg(bpr@XT}6NGGv{
zzuEag<K9wAEX%U<hq|5TVfHNhtkSlP-hoU*?Hs63TBVvROc%gakVeS`f_Astx@k1q
z?ab|yBTJkVC1iB3rW_XN<~!!cO5ZBIYR<4=+{D3I>YcSzm?jF4a$(&PwT;o23VLv!
z5GPX3`?(*e%&P_9r&|t=J5$5i0UE&v?v69rQ0B4dU1JM>u!nOxcA4jiX_|>qN#-#I
zT5U}8Oz?p;r+qQswuJYqSz&1L6JlU9-?%;Yt6nR#p&G8uygZ$wl*Z}df^)XH^-j)-
z`}~5x@i+e8{PCas3Wxay=R8wTj@i5{eB~=&;j3T$O@8p@I{fu-BtFV-(`ZPTl^<s#
zuWo*L*7hU8S?Axc`;#5=%Bx;^#ko}0EmNuz_Ol^tW8>d5b*rlJTWwdOJz8Lwd@U|g
zE`;C}g=v6uT((T~8&7W+q<77w^`j^^9;`L*trTj``(y&aNb6V5tP@9n#I+XTk)NNK
z`qr#!45=)(mbTt;(wSl)MlGhR(FL~Es6)Nux^c`LkH*c_xPqhL9!;IT*HOf&m!LN3
zy&0<Ly=@H+=i0u;>kD67E^jCpHXZ5$&R;W-*KOmq!xKPQFPH1*2IhI8w#I5cqd8@2
z&`3ik>bK{-$9cW3%2PB$YxBKrtr&=6ygYl4VMpW*05Qh|#L_4p(TO_Hni+44nP}(2
zG)=5&#e2`qalza7yXIN9%w<i)pc9*p+S3FR-K3l;Rncu9RJ7njM2d1XqnVhCnP|7F
zP95W%PLmlUJ0S+I#^=syP=Z%aSsepA8;|Q$ssYr7lZH4GNZA`#YLX}TX)nO1c!Tqz
z>x?M6EDeswr}UDUmODkxdf(StN91_-gc#Joc}*{<z2PL{UC<iEJ8ISEqBjr|YcvB;
z)C*E)UL*7Y6sZ%5n${u?=b)D5isUie7MgMgyU?YfHLk~rHr-|TCAz46r`^||<oFas
zAlGGWS9(e}W8Fc_g;E+XUOd1yOqmBVr(BdMoh9m~5CUxL8R;|AG$BDf!$x=3Obz^)
zXEchRZfK-15ZJ3Xoc;-LkCCKq<G04J0URU8$6kUk0>~c~gZk>^vJ#A6mpAP=w8QH=
zf~8Xz?Xb0wab)A{Xi=pVQ%7_&sTJq+0y!s4K{!m2QYu^ekn@WKqINg~aZ?3u!~2G~
z#&Pz<;LNO9?Ht5zPXP+WxHC#*Ev>N}4$5zNG{VzjAR6Ds7N34qPPdjcN~#IH2&E|k
zCk;BD(khcQdN2C$)J`p#W%3+vW<qotc~c=baEBvP(}j^o*P`~O0B&#lcxqg$q2fw4
zlUj#qR)f}DNI9E3ydD+Jv=pa{gtX3BW`EFtk^sl!4Nspw#km#XBVvra_Uv^^Eo|$?
znl`qa!R&E<&_#|I3Ew$2dkrBl&kOTBYv!kO%+rDwMdf|)V5s~wP4wQ$sq97HKypUd
z{6KO_sn$V9dsc&b?44H*#{-wkg=JAon3Pxk-rxJX{Lb(GkNKHD`Lo&$PpXzY9v1%i
zAODIH@aqpA0_+GrMBES!*yV87Mdncv0>afVahTP2*rUkCSBb49h&Cm;)E(X1ZFkx=
z8E2%Tw64vR7)Ps#i6EpXn&3p0rJ`$tTDEiwjcrRrH`!#S8-iw{I(y!{b_DyFb$CK{
z2Q{m$$jJ3pQLEplG6Us_-Yfb@C)eSo;LNwvsc0{lO_$}=Ruhb&r$nhe-6%a)*wxm^
zhBA(8;G9<hSJSoEn>9#lhCYR;JT=C^rSGHd>*cuG<+Of1|6@$5X2Tk(c6(6Ej-1m@
zh0p?L0uk2JnQ4xc(hPMpkxLy(DV>@VV;s^9<@4qV$Urw0Cc+!nZ5PFfp2JoRSSZFl
zw9%<JZ5(3o#?iI4eC$|G1x(8{>3vZvC1s)yJFQ^%^c>U#yN2HKX`Q%h;Zv0JnRBu0
z*3`;ItPQg5Em~wKX3DMcp$ClwohEGtKnr9iSLE9V<;U4IvBE*GopU%J@F5bTdWx2M
z!lTW_NcZX;d|2>dA%q*^d`tA6T9U@%#e?z#y{FX`A8$CsBbdKqE6P9h!KQ=K0SA2~
z_avrTw0Q4b0hw{J#3OD}G|>PQ?<39)<Mm1($s;-K=yY%sA_0sW>7(UDiz(e}qnH04
z_yBfO{#MvWg8PWjPD2Bp-=C?c%+J2|8h<Z*V6UL6pXePudh>nL2km5tln;)1SqO2$
zd)<VjcU-$#6but!9lhv-GPr!-^BBl6x+)vxc>gEu%-1~wS9XP)JX*w|WOFocaJvL@
z$)wuw_?`FGOzqoxrPM-b!NRQL&Xquf-Umk2A4zXJUIuz1%(1h~p4ukm#`IB`nvC91
zQSZBr@vht>o5|48IHHagrkj~b9+!pPj(x!DPEB4%aC>dt)Z5C1%%xCDRsTGb(AEyc
zxDd5hZc}IJ<}KKV7g^!<_Mjo@Re2OORW7M<xO>7)zxp-qwitW;-OXpIKT3m8v~nwz
z9%c$!+~s5fo^h{&`ZC2h&h!-l{=@H<Pvua)P2fj=>__>^Pkx2g&xrf4c=OFS`I(>j
zS>AvDJwE#ABTknyr^}gj-DpY4%vVM9r}IAcR2SR0rfFuLXO?AQo)?y7CIlS`&(n+(
zM~Q`HUYMqd5MJ=nM;|do9eLi~-m-2hn=MGgCZ-u>V%&@4-`JN&O~+X;n~53R9w7hT
z`|t2~{?6a#kNxByW0?**hH{?6;mDW1^kt@LqU6h`c3%F_^F>#MtPCD2dLPQTM>RuM
zYdJ-qaKrXyG?|@dHaN;>wRO?fuidCw|CU;89N8WffupblM)s}Py`kfAJ+G|^sb(9a
zg}rf>)>%_lg=h=Nq=<ry_kh_|IkBAKus5$_V_oMRI%8G4FL9zGh%z$b486ms7)3|H
z-g~ue>X?;>UH>uO3ZA-U1`THvKeaYRMh(F)O%cx4HNQ95=sOQ}aMLN2FziFF<K$gM
z+8=TM9z9mOVYC%NgU^OF*0okOS9Z>jZGEqUA{1ied_EI{=K5B_ykVP;R|i2Fqp;Kl
zGF05kC&Ttr&~q}1!8U-8J~)LYZZVI%V#-_}y-vb_H6<%5W#1!m&N`ioQTOelM$wJN
zVTttK*;3ipZ!MlQ-&nV$>weN(MO-irV4!66K5VVA8G@Nj%{oi8_kC?<YOTZ=Sq=wA
z!ppVMQPXBbOyKDi!S!B8vbKgeucP2U_14cq7iwOqCE??NWxi45tgE>>y<0)vj5gLt
z>BP34S(Y1WQ$?}R@P1J~darv?7X)fcjMOh7;9aCS^^XXCqEX2?T>&RSU#BW3P{R%M
z9uPz+VG_w%fu9g<O)yJ}Zs@!BL6d9RMzvzG<&Euh|9!7(@TnAldT1|Wqf<eZpFlU2
zN<(;YN?cBv&*Xy=-56%Rf{QxtZLJ{UaZdRLIcK(1*tU&nURahR(>(2DQSn|!8csD_
zy?-ZvO0GyAqwX7U&~<d6{~dRBr;942N;V%nxpG5Z>2?Qt2uu>Vn?3jEjBDBjjA)>p
zyHeF#FPf)T@9JHpHJ9=@#@lGJzwK*Li$dtm1p@4UJ2nEr3aowMhTSZ-L#-RVUlSbE
zgmThueBaQnMWqiB=L4-fN^Rt<yoZ$Y&Oxaym@-%P(LtfB)r#}98W7az6m$Mp6K2hM
zUemUd+qWv*`naptc=Y*uQ^0&&XMLa7b5vU+w}!-tTsvhew4@IB6{Rcm4&7-ggWfk|
zas1xq&+<L~?)y0gl6rf0$CtnKW#(nEjR}7GXMTqN@c;N9@f*MKRsPn0`#1U3U;7ol
z{@cIJ`|rQMkJPRk?5B5Vj8@>V9CiE~XO{WM;c(!1JaRl9*-~PeXAbj$%lXW*99Y+t
zr%ztvu;>DNI-N*c!TX{rObbx&y_4<v8TXX8F-q^=#>aY-=TtTEhDPrbuEP{(e)ZRW
zg@6Bl@)!7#&wq)WH~e%Y_{f*O^koi*ne*ufUjX)Asp&!0yj%;2ZqKVjxm7Ce=n#W0
z-agPx@jt{cqsi!29H*k@y?b&_RZ7PYI*4&D$4xG<KO2IMSpB~C>0_>{VO>OeEwtwL
zBBOIfR)P-%*RT9!uVZ|Tm@+<J8&7}`rl^zvyJj8be5(pvaKV#v*$d)Y^=m{hM9#*x
zCfLfkQR>DN!!GX9LR0<j^!)UF&CW$dG0BRmXvpLgBjvL0&A$`%m+Rv8YlsXhDpC6m
z!PXNS@|Y@E|7?RfqX}us2?Q6IqPi3e?ql$tQY(Q;lWT$}XH_9@Dbu`Vu{%{@Yr)%%
zZ#jHv-F}^0Ye-kbP{;AgXdDP?3~_H)<IH|P7AwxxW~s2IN}MKcZ#5C7SUQ$CRSz%I
zjCUGorb#jCo13#*^|UsS(Cf8UWU=GAfeq0acO15qnWu>BnpTo(r8Rb;hxd+XdrYG;
z44emY-s*K?(Cwm#w`^VvJZLaRsQN{wV#e{s<7a7grIfQ-r!-1A>$z*5j`H^^4#^&W
zX{;r4|NH|=RV8v?PxMxa$0xYSv*vrK37Rt`5b|ZE))fgeA?gOJ(?`FXYHP!F6}Bjm
zx8QsPHzCqV^@R#U<VFP}&g10(q9|WYC)UeLrZ6)vZ{orY-WNnB;|oSkThgZD6C9t4
z51>v(Iu$k29VdmZHX8sq&p9_feEz`Q_LMj~HY^RoLX%bHrd_o66CpUJRw-@M!YLJ2
zl)n&WXEikX3~((99{>8@Z~8k>2t(L*gviFek4=Tg*Ju$}_o9JEUA}jPLkP6s$)@gZ
z2FGhUMupNT#+QOIPSYr#Q4GW5B5aJDuIE(v&SMa?_m9VJ_$nB;aUclL(XPh1NSG9j
z93$Os-EJ9{52z<#ZH1icZY<iGdPeo`k!}eEUB~%OI&%<aADHF?$EYsm1A)}^kIq&y
z^g?c#W)@^VM2^R!!rR>h16-daaRWhA6s7g6*q}D$8#NcmrO>lR`j~|pPN4RVc+dHA
z!8ap`|8w8%-`_D1h!6sw`~2s*eR5|AhaS#*KJ)p{@y1(k@u&XupW@r!{$KF7{^sA}
zKm3pW5B$S__zyoZ&*M9f$LQY`^-~1!csz1*bHm|q*nfo(czXAY`w(|3z{u(zInmQJ
zalAR;g5z{P(Q4iQJ;rFlkn0Gm^=@OaN~x90Wg|}dy0sNTh_qJt_B-F^AO53%z-K@6
zd8ET+RvVxH+~>Hxz2(CXo`0(6=6fBjceZuK`)Wv*S~~@Ae81?aQ?UEBUhmtc%GW{q
z9^9th)w6F<8&p)L_jxlCeGGoz40Mz>$m4=kjXk@1Y_g-P-qgWE)!q@R!IXm9c`Q?8
zo&q9fwIMq41+O+0N$p{fF&;MbsWy8q2f0@Z)l#np!+===DZONP<tYV2&qoN})UCpZ
z2pfKNwO^m>T9u<*nyz`dwOtV`R>+RB*xBU52aD(I&}-j$-M!VF`lC&<+rf&xgS7R&
z8SO&7?1s%uR|ULt)ULjw!8=OXnCEEgsoPQSM=!~?ZA3p4$1`teTDM}s8A_`PU*%5L
zS{_kgnx##N3I41Bu(diDebz#Zcse0Q;;b|m=dahLR!x;iCF@AhYY|pzLGl1Abo$ni
zy|@zR)bzYHa@M`1v<)nhEriH=z96FUb%-OEtcbBVqD_xlwLIz8sHM}Iv#34wlofQ1
z9G<SDX=!~3(R7C%=HJ7CqIFF<=@Ri_L7ajO>*WN}h|?{-cGk;>y3dsxwu|Soo>}kD
zgm~cIY}l0o6KSn-USB|~I)*j#?zEoiy%G?bc#UOr9v3EhK9llBm}chrW?$pGIB<bD
zA5_g-6Sd7q=r})Ff?1>XssPDI3CSBmXUdoBq)LXjVq{)!So6Q=0+8$ItnamoR>r+U
z+8<I!>B9RTzT|&+|2b}N0-Gy{Gtw;(q{HCNHRW}M7>PK8J!O~OYg)y4cCFy6*UZQ3
z82DNo?e6Y-HSy@T!T&H`!w9Ml=3cK~H#$r)3xrOs21T2{Oq9kFR9V_PG#t6K9kl|D
zD5}N$q0j|8m|>=lJ<49Bj<t4-EB4osE82H`f3_j?$BXo^K+v|S)Qt0@usiCsuifF*
zsI!c*h1pQ3iKnew4RknQ%52mY!ig4sK{_GAbO_AeIEW!3s9LJmt_y$Zq)n^7W(DkF
zUYPH0Io{lIZ{rZg=k1TKo#^`*{)5i=leKl+6j`h0cy{v|wEEe6*_6OvOI1!+yCTCs
zutHF@*t)s7;j^Fp>^}DL0(a8@XO#mqd&iIc=#TT(n{V;|`J4YOU;T}*a({aGLA)iV
z7&#seEX%^OEUzGc%d+5{V_6ovM>2PJPl3+yctc8=)9J+N^uV$lsI9ZE=RGASbyWk8
z0&o<ALq*)GF19IYeyelpE0;EKz7a$a3Di>fm0$T4e(B%&9}y-oqgmnEYp?P8>#y^B
z-+22|KR4fft{g5dx;}IW&J(8Dcv6ZY`?z<)CfaD$wYdHH9L1X2?^&VKhFPyZhoi|G
zey45<D$Ww!s9Wt8L8<3!pm}3`Q?yY3zm&TF92a$Cjb2!y7J7`LQk;&mx{4CSCKave
zNO;faZfd#lxLNFiW%mN21qPypXE#@eW{sad!ZtP*cclk)%sNa+RZ+k?3P5ptp_ffm
zFqHb$J|FhF!F5f8@}AnDrb2ln?od|;Gq`Ncn)>Ti-iJcFWsj{bz3Xyk$YrtI+bvas
z$Akm2g9FvpQr%0;%EWn~+8%W^w02&FC9Rmaz{cGoWAD;?h1Si-))e5dQ*c6@nCE#%
zfIA&mbX1;mx2C?btF9a55l*1ED9(FzMsS*R)lfuE{cFd*=UH>se`_F@jTn`i7os)Y
zy@RQ^L+~`C0BHPKrR8A|9Sm93eOn`2ugY)G&+G<Wpf~7lFZy;w+-cORNMK%GXPR!!
z>|c@4QZ6*(YSy%Zcyh|*bYk01Om4zUBdzzUj-QWKOJu#|A<`<;niLI=I?Yly44`XG
z;(fsNiH2jIkN9a}xR*@JfmQ<DdvdF^n)R9VIss9ht*8fkAI=+<DMniFT-Fn_jQgvr
zQoMJ(i$uSCqP6YQDF8)`|7XPz3Keg_JRb5UNSk@_{KUiki9dcf^SAFVh}p4-!8}Un
zZRS6x!h#STF)ZwGFzAhC1>EpAdPTl(i_8cthid`2ubBIIk-dWLO>fBYG&w&;zv%nw
z&`k_6vQGOW_u$?t!D((|ZP1!#1baFmIB1QU3ngSrF3}Nyd!({!hNkVsG=@G(N5nq2
z!Qo;f(UDoK{z}~l>gFlbHJ{QsVP0l@(AQ6S(@mu6c%>H2bS%ZB=~nCw%yAT)>Q<xY
zNJI#(5(4-sFh@^}4j%+>Z?yLczEi3AE_CZ!s-c-NI1cm7&CLycIU+Go){909IzhT8
zh{xMzs>Z?%saE{D22IE=;HE(C1#(d}qZRV%4cUls*6Z2KX?@(k{aPr#S_yvFU;L2|
zFODfho;`WW(`QeOYz}ub{JTG4XV&gX*91J|yz=~`5AA-IE7$Q8R;fEmEW#8b2OIAm
z4@aIndBWjvAX;H~dwa+5yPT$poGRXH5q3BP+w`uA-<;L&>M$=z7fP=9#ZU+<L~5<v
z+}u!$@^Zv=yl4}#t}Ecd&Dv0fNX;8x{n|g_n{WRfKk?&VA?Jjj9MA6Vc=Po)Uj1zH
z>2f9a*L9BRa7#B5zr;y>$gEHp!1kWx5u%Ef?6`_JbgWiOC6}@%Ahq4kQua|>?+q`C
zjKtaFU17_z7tG#!<|%43)UVeS`+jMT5KM*Jwb_W?vCNZ>;SC{-(d))rtC~|=;Zjne
znkZ$Auzm2j0Nq3f-l+$lCQ>B3S9IJc2yvPS=s36(9Wi^a{&lUJIK>I2YjKyVj>Ii`
z81>=PHAxILZbtiPO;wT|Q<z32#4MBg(dEzxHBbgg-Y%3{UX6=$ErQ`uBkjz^!HFrB
z)u&AnICVm(Hd*kK7QC%tD7U#MbZxNaX`<vxi<SdVH(t4NyLwdvX7wc0dnD|lnjyr~
z?B8Ns>0w_pY+;;u<8jN5KyRByD8@+(LwYx>7~R)e?>jelpt`lHZOd|)aRNDewv;W|
zN6c2pc(8OXIV0XfGxXLbR;?wX_j_$1(sphyBFf*kb3Nz_!7DXDoDipFzuvmJ+1_p@
z>H}cXo4?u~*YV+ib4N;E=`FLwKyOcJwQxB<R~xhGNL-$PJJGua0CXqDQPxn2R#$3<
z+7q1x=YrCQDEo7+H9*j;VH*#NkA!Zf)TKOX{IH_mKoxVIa1>fElv+)&6Ljqe6Wci{
zpy3yMIMBPJl$BbP%bW75bxQkW>)j_ef(C=^%`muY!BDA`uF5k2>4NW(bV)p%PJHna
z=(lZOv_}GkblE+C_96&p;sk}IhY!^518DE#anUEn4;jLc@r7cYXy5-l{%LSctWd=5
z*_AsC@2;x8W*wSB%2wbCc2hMbFKk_l;?_r%AJnVJ$hWl=yf{@9iFyM%@0h}3qY+(%
zujirf$~@_g-E!3`SnsdiV-4t4UoXxRlP$irH7-fBmQyaep%gVvZB6W2I_Q2!aj}JK
z*i;U*&pStq!W=w4Dv#4U3o7rb0&WPU!8C}wrP7n-I@6SovP>E_9-&6HvZ$?s4}0+$
zL>H7MP|X@&6EjpY-<wzQNAT1&gCzk}J3Z1Big&>fs1b}l7MknL{9X6<_xy+NaeS{*
z(TQzPA@cfbud6#g*Y#_Rv#*7jq7CnS`wyrk5&S1FK)=@!#Hm`c3o!(4jyD_+iu{c+
zu=k#2S(svEO9^_?+-`Av@WBUc+o~qB^8|EKI<T#3@e{mS$T;u!drccK$N6;GQS9C;
zA=t>Ug7Y1}c_Tw-W_RG>bkE=a2Y;U*{n4L*n#oC3f^WU`=8n{KYes$`1)!MEmuP`B
z{P$#lemmXd(K=(OuE1WvH`_eakhqbH+7n^)0O!nH*^0BF!1l%;4#A`Gc*&*&vTr_4
zGvIU$cFvwx@P-NoFHBLlvEDj9dZrlETBjE^Nw0;P60KxfF|&2d%hS$n?}qp_wV<e_
zBGEkkL}LLxXkmyJRpJ~Wx}5{4MS!L-q*6$wkxN(Jn4ThSKFx-zi~?kzM2zW<PL75R
zlyhGBfF-LiBXlHLfjm&CZev#&qtw2Cu3=&gTpM~u;l$3v^){~USQ<<&v|0?k>N_n(
zQ$Y0dl-dZv?H+exK5%7AbnJbJDw@YVg#%*Lk#;S*ci`#9=c7Z-MG?nRiStC9l{Y8i
zZ8WI`@hIfgDlHR(UjM5?U*CH%%dHqAcu(_V?0=;o*e0W@=r(xZja)L_D>Y&^k3r|K
zJuMI?-5;{MJ~v~mjbM#|7RYWdxSffv`hLB}-z)EZ++$Un$)LiFTA(*IS<Y?6g}@}Y
zbQi$|=H-UT5B|M2D}t^$+xa69O?Zh>aADSarzK0`(L8MD6;-b_D=$*Yz9zIXRNk7M
zyj>de0%2YV?m$TwO1_xolOR@#g*YQFs?qj*U^zU)xryEzxoilH(#>BnMq2q7TmZJ-
zG`h!@oQ`&B1x1}oHS)Saxuq7KzkH5AO@ybFI+^jMwX(Z7!LlcZFhRqz%_d<VK`5%F
z78YhF;0-dC>#u<>?uWd}mjkr;bV`UFf(qrSwB5dC7lAyoX;9k*MVngF0CwjD9~@$S
zT|?=mCY8GQ*IG31I4Bno$9_YLnAeh@XwhJQeR`?&s!SVldU&kP(uZlcS8$%zRK0sS
zXSPjEkyEbJ;%Lo%$B9rhM$NEAy_aN6KUA?jqI-s{KU(f0*7ae;U|Lsun3Qy+<s&_3
zbNN?}kn<itPY?s$d0dQIV7HD0v0`_qA8paL8PGsCdV}26=WBGf-Ev1e`&!P(we@=;
zZ>FWMJVW2>f%^U)GF{N!UP!(1#v7(`-_P{-_$Eb0e)o5Ohm?}tYqIlrK8-_QH+&vy
z-BCEj7)=#CAp(cP(Ny4`DNe>|)6sBijdfiKW?wN+GthY&Jnz2q4lza@!_;A{K6pF2
zXiqV)ZP!j&MD3z-Dy)~r-R*(aI_u@aumAer<A3%){hxy<-v{sb+~+=L#QDCz{txWv
z+T0<|Mru0)d7nOvoL;>*+y48YmUN?g)OH;a4`jfZC`=0zvoCR?(*mP0)PY{vv!G%@
zJ-5}GJr%39T_mSSp>9;ciSfo%JsiCa_i9ly>Y~LTBKWBJxwTy5_G)RPL$-~2tb<C`
zk*3cdPP0<Sdv(s6vQd%G-t?N)UfFVI%jzPq=4wT}v*);8pI<}&dcRJK#@{2duYW#z
zE0#M`Yoy>b7R?YdXMUmGgm<HfzqbG5TJx$%p=<FO?AmwHSg+C==XGNW9w&xG`M|Wy
z<fH}sKpl5FQm{mlrl0~%*H7<*joGu_6PC%?JEq=cM+vLB8w^x3T12NfeOA3|oG&Y4
zT-|1tV>wq+${d$v_Xw<3sDv=$J$0|`ee7F;^lq_iw$FF<dsXDkQ&QWDy6C;eo;hKe
zCxp6t&+1;{_HpsDTqg;wIz$?^Ra#YL^+3dj9g83qO;|0<;ZfeR=!C78Ld~1<yIUot
zmEb1AH0zYo1w?{!^Lj=6f(tXwLEcW(R1rTDmplA)G~QyR7o}AR*uI{XHqrDcT&CN-
zP-}ZoL8s<TQQ)4LsRSP>HK~1^JJP*}d?syY`tcd!A}$=656_Tzpf{U5xJZ{m7e_ok
zVRD|WelmK{CoTZ1EpnxMDq^UgK5)gjbnvLg6DT?J+6Rr-Zz|t<t*HvtUwxWJ$1%Qf
z2qJ2&<}PmDR6)l}8o+&(`;Q0v`aIicW{iR#z5nmpf}QfohMJXb)(wvqhw)uX%9P!j
zrqxCf9b5Y@M5mpfqceCV&U<2r<f0Ej&V@L6q{}Y3f4q$ie<&{ktughH{bz)8#zOY!
z6B(LVT2~%S?wl@(ZOf!ojXx)-E>``*cHKETb`jA>-O&k7N8#RSmg6Y?y!WQ&wfC;E
zA~sUaNa>U+)bj<Zau!NY2Knnk5~JCJh{Eb&o>Z_gY;eYg=Nv9Lo%Zeobp7XQs(BwL
zBZ_O+Ix=gmys(bF)6Hmg2!UQb<KKDw^A{DbTsOORS^x%~JblW|?Tr?+kD=%9@gnQx
z!rO1Z&C8cBNoL#dX`+CF`hbl>?|8gnGM?TvO~yrY+}+)>Ec2cU;%$U<KA+7BCMW`u
z(w;jy(g<`D4xBF+mU-4qqG(=f$r+Juf-j*KMMpY~R+V56p1T`WHA)k{^{sF5^8N+S
zUVDR5Dszmy`Nms#A80LqYUkz;J!du+91WH>WX2<7jCr*nayIe-cK#e)i-sYN&?cps
zXhZMwW(_TPTCL=wt4_Bfy%Z}Zt2TZ<=!Q%m2`zr7yo~s<eO<Jor`F0LL=Kb3Qz+X?
z%d1Aw?Z(1F_vREJO5O)eez=Z~BVyNCizW4Z9o)d$2lS3L;U1jYLzP?*7g$qfO@*x}
z?WLINR)tyD2@gCf_8$FCXSayaV%I5WJ2Vzdty>$lCU$IZA89;lzia~KOgJZq*=L!r
zo}r)WW2lKa-jKSqt>DbGzIDr{SCjzbKpek6#Hj0x_mrlVC>lhdt5O^1+)zes^tHk5
zt+6E)0(e6W#|qFMfgs*_YR;U`=e_WpkCUFW+9@UB_T@u4-_F((^;i^5ghF#~TPqez
z2J}Yo!ZNG>EaLE1;Ma1!cXb~yt0!m7H(?hz#c-Y6VTE-Fx|Zm_52btCpMfD3@!lIx
zd)x~;)d4Mh>D>}EX1qYn7t-a0p}?M6vnp_%x63ZpY<W#SESc^C({xbaDX4l=`~epa
z;2fo9f>%e3Cf=TZ_We)!vU#UU=r}zYXr%l>Cs5W0w)1o1bfBb78<ypknlts`BT72s
z<Ah9iI?nciS~In-w7P-m4MLo7KC*3RTKi<~5TCdJEU069wUbgs(Wcr}z4aJ*bD|H=
z;pV`a@7eOkx(#H;{Ihm=|BCz}qvKZP(vEKV;kosl=$>4E>?k509k1-Y-?^{$D^f8~
zMn(3nT!TG2%|;Zx+Y+@S0it|A9Bf7TaOTISQvAp~d0->wtg7ZX?S<=Z$MVYiGOmjK
ztVYujfptaQs_SY*=`ClrEs>Ji=%}6H5N}*J1=$NeLms6mZ0{7l4-SHJRsecKCf#0O
zYm|-}Qn#up0x2hQ%ebOGy=7hLt!bvRfb(!XPV{1EsR+TT)q<a9YB$>*+j5<|A}@dz
z%eZ~K-q6K)j**ftyLT9m_aNtXHSN_H6}^W>H;$3hU&$ByJ_|1y<6lKdpS||X#te@e
ziN}}s%0+tq{5jwLgKzWEM;|J}l|Q`#XB3~_Yu4|wEW|i*yty$Y@I(la!$Fn7p!dS<
z?Jc!dwr%5hJhHAE>w00Bgr80)&ZjdXS|Hxu-f}vhIUEj@s(zkBIjjYs7I*Wk3gKpp
zdMSg?(|PgY%#*_{^v;JLe!w66!MAw*^*1yc?y&Im*|Qyq|A8Jhk<kDBjxygGR3~mv
zP=QAyMUh7$icTq+Y|Tl^NsGK|T}czSo4j7V#2#@#2gSuu{x&8Cdy^oFz{Egp%AY%*
zE}DSggeOPOO$_+9QPz#Nt<-JPJk^fr#<Zz!y{l^1NA^nB=ryf19N}7o8Go$Ha6Qny
zq2wV%ajZ#`4?+xVYa-{$R;uMk*K3hy7FN2h^4$%ZJ=Ejhx1=M@<6+sGI_E?MXuE1w
z3wA{%+dh#28=pV&fEq~Z7&i`tJO)SbI>qt9v#pti79g#4e3ZT5F1aWw>Q(8UOQnx7
zV^_7WJT7FnF{?IkL5Ys^(rFYLYPlkYFe~=pNO^c|b}s9MrB-}^FipgHqOPlc7AsPx
zDO#!v`?PJ0Ue|5YJ-q2XFLZ(-(ajmPs||2O-)m{jz5zA&mT}Is;7vJmcXJ>*G`YlS
zebDXeN0x6BRg&w3#8Pcqp=O&tnH|jz#<Z~C2S5AlkMSdix3rk68k6S5&Y^R73Hu)I
ztkD#+-f1mSTcVUr_i`#O&iHsxq`M?qP4pgiUaAj~Wx3H&K2XZW<x<$vi#^9VhJzNU
z_RRKPX(f?U(tT8#74d~}^XYh9_urE5DW@~@a-`P|%~9%wS}MzY18u0B2O#wQ9iJQl
z1Vx-$C-<f~wXKnQQ_hPWTGhjHnDJAjWxam2s&S%r1VCed;D*Vgc4DH0Q~P;)sJ5+p
zUB>1n`_}-BvGV?K_3PrSUC@;~G<>-BU%7n8!rynp>79o#7Gtzn&E}8iL=X?f>B1kU
zgNANf_vXFVj(S_Q6ZG0)bx^U#W0Xj)C?+z*9^FPULT!*!BX5Ou&6<Ewt6B+Y3AHa8
zx?d{77o=K*nw}c(<=Vv=$=h4EG#1w3I6oj=<?Cq&ol~Kttcpe|uAo1=s_jgeJe%u?
zS5L82lPbcTSi*-8rkRsL`r0UWb6p?T54;OB995i(k6L4zBh7iWGN@*aAogg<xJDK~
zmh*o8aw*gff+tQhuf6^n@4fd9YNMfr=<Z4Q&v^)gm*jWh&9i5(?Hl^_==*n!0<iVY
z2OoUEJMX^3<+AEE{Pe&-I-TGn^E@*d@_slhmi4R&AR&akm>C7$V@sEHU6pTY@7a6x
z^t-*g<>CG%H#aw|>&o5T6E5q8ciw%6VMd%%(uldHsRHY&op2pomJY4LwpOAGl&KS)
zBc;T9@4cr=*sdeb+uOTW3cz8*@q;+TMz*`DFkGL3-VGghJI4*!n+#ASaHP6yDevR8
z7%{QM;6oQfuj@{n7ZIe1+HCc1BQY(u9I#B0=pD&Mg0)r3rI3q`30<!ogX76@;`R^;
zuG7m#&uTPYtQZq<xabj4O3z5ss5`x7Boz&Auoe})>oeBcm6nle<&ui3MMYTC#wDqz
z-O|i*1hsC`Yc>ilMNfrCh5dIt09Rm9jeyrJ;oP*RUX1auP8!tgy6@4Wy3Xo-J194U
zOV;*6W4OtVXvP>l%QCSnQ70DNnt3z#rB#-2Ftsn&&mlVQ_QUZZsx3+_nL4}&ZIm&P
z^WC@`ZFV~_La%i`Wm7WqimXO~a9%wK^Eu&NAWXsBC2V459Px-IJ&kcMD6Qw_=E&)E
zQJR&KFez?u8%>366Bg&3o}t!h7SA_%M{rJKQN2U5n1R2pwF;c*`z*VUVxafB?~OFs
zX%o&Xs@$Eh7Zc*ZE5h1(MY<rZQ?^wRc?37z;2o5la8y0NG?blRX%s{PrFNP`!Zd5T
zOe+YLCLQlOt!B!qRx(v)oCN$7@uH*Y+I5d@d8M^NNB4HYku9@W_kv*e+{83%++Qu@
z>oYM#<Zw&vM_QO6ujI=K3E)_%4ZNSA7qw%ni7<KM;A4J*<CAj$?G~y)L81R@e}0_G
z(T!&@FEh0l4U`uByHpCHROac*N7GR1zHi#s#dK6#`_gJZ;aZ#wp~n6-hSXUJqu(>e
z3thYNT9q3ykP2NqA~8fGXuE4j#^@w%q1aV?PDdj#Drc*CC`CwFM~?9@Q<838!(S?d
zsA{ZIap8&%?ba{fIkpx}SLx!&wX?37%O$a%jc3q+J`g%Th9^4am5%pD%J!WoO*=vJ
z7SiI+?eFasvh6s__V;xt8+)hqqKP1>(sNNps8>&};36#sdQSLh!Ns85kr=5d&|{=o
zK_@;y2$XfxB42b2;A~uA@GK&lLz=oSN`ow2!KD9c7116QT+KG3l8dGtSOcI%Lr28(
z<uCmxpZol0IG^she|Smn{1J=4eb%xcPSpP8=Jw{*FMH)HuiV$x8t;7j+kE)pN4pKf
zr+El+wOxu@=#8T3a5x|$%(ISt$M1s|2ug%kdzZF*0*-%w?X}lQ+sa2De?*9p`-cZ!
z-rw&cg7I^TZDtx7_Xw@u3jioZf3@1Fy|J!mzV$7&d=XKB#9ME^#o?e4azF566n<j%
zN78qi$d1q%vNx2Q<H;S#4Wp3p&Rc;UXqNgh5_JZz*?SwwTV!Fb70K<9y~rpq6%}#f
znWkWgI+<pofhUXSW?nc<p1b2jhz@F^tZBCtsl8jIWM!I^2G9i+f>h-`HKVI^E4s$-
zt#?A472%?)p>J>C0=X8ptf&jd73|LQuw{aD=4mFS!ru9h(X80L!1Woqu0Mv5bzRR=
z(zW&WX0eiSh!s4&sS<anYUw6&s2WLGO$a68a6y~tQm+1ogBcjTXPzczdj_<Lm#dvm
z@5IRyyvEz5E!mZ~35gQ~Lu*^7q>Yj?PE>>@A?#~?ZPglew`bd8NUVpQ1MfXVAMxVY
zETean%{dh~U6{M(jF(cN7EP2e@4k^TFxK$iD$!}&=tz6oa@L{@c0{;!jrtrC#AZHx
zyUn<$@sStd_ONIpJuKA>@zXUUSkt3P8DlJ4(ncwjXa%(1GbkS))J$1)YN&U}xNn?#
zK>7*7L~R)wc<Gd^O{SB`wykvOgs|A_9q{o;swY}W%wFiuDbKidI)z?4sjb8iI7~CG
zL)uOz+EOv6^No`2-gS+fFO0nI-m^|!rU_(GRMIts!W3tM6Z&*uT5i+}IjzjxQrbpo
z8?6;eejxaRj;>3hhd@|nK54)EDIkE_l~#-xjE;Z1bzwYO+@TR;&<&(yT|8}^Fh-#c
zllY6^aOU-ASpRT#e^gkCEwH=A%cCFp_)lunH~3*@!;mw(!Krmg)1;nC-f4ie`e_x?
zw(W)DFjiI|&I&9yP)YQ`YT8x0pbFSx*|ot`YQy9?_(09tVY@a^0PRK@pUF^0B`d2B
z)w*TwPZ!STjaKz%s#TRB*2(H8Z#eZ`B?ONT`dVJlxH`G=)JEaApR=9F&5nFtA9@RR
zZ;jFvZRoY(MLCfz3X^!Ei>mC+1wvz*7jlE-HaI?EesZU>_HE<l_Ks@E26nv~f&^oL
z7`gzO25>2b+`)Bl%R=c&RByHKW2k+kV&9*OO7I7g6PAT~*>KMD`Okfh!?Nr)4Iy~r
zoc_@dF_hVR&y%|+mU__{Ey&|zFK*ES>+QGS=K04T@5;STs|MUr6C<Vrmb*EUB6NdM
z^`h*k;Q0M?I&rz^*!S-4j*mb7_)1|2fy?E*Q#l?U9zfvf)2BRt@tm_2X;)S2xCY~$
z>F?c8U%C?3FPDvF3UudG8C^3j%&c2tS=7UDw`ppBlw$pla^kcgyTW<hI{HUN{G(=L
zG!A0RR&6R|s#Zkhl*mb+`G(!oU3rk+MXm!2%jnJKf7iNt5Y8&z30_BeAvh~eI=#(G
z!&ntPj$SxSfqC}$;Gk7nQsl3dW{A3{d!<Iul%we|^u~PlaAN*?-Te7Py*3fV>8Pz&
zT?@nsQ95_Ca1(>6jRUpLY^CvV**F*D0*Xgulv*97cJxQj$Q32Gy719u5VPXa`zlvO
zQ`9!koCL%;XE;jMR~g&LzOM|=#&(_Zj6y}%U@gGZ&ZcshXBD`WLU4{w<*>|nuSj(+
zg(>RTTx?R+YGuut;EXpNj5BGrE#o1Nrux<9&!{;zjT!rlm^h`BVxL73D`(Y!1}bD^
zh56u^;zaFc*4-;~N6-0@x>sZSdRN3Wgsc0*`Eup%Av*3?&{EG`?^+Q1a1|g8u}&>z
z_xx*30hW1I#E*7WV(V#l)SM})*ceoiJ>*LJ67RwHecyfDKnL^W9IS6~Dv0T|Snt(I
zb)~h&;r2CRn92QK*PswsZf{9vbGdL%QM#V>zHSBT$|YrTPB=elgXg+^{!VQhtyW@~
z6i|_&FmHVL!MCVAaC`d($D1S02hyn{{X%aSE|(R*+}ZS_C{W@Td<&$M5e#s0o@IVY
zS}&B6nGZ(=OFqHzsT6?fvDT?w4LpaO;8mnf|J<6TN|dbfeizj3-3*DNJ#?)p$4!Q}
zmdNh=H@ZB1N!iiC-X5(EXaUi!Ln}oW-eFEV5RPeCUJ;NC_`Q}&-3ph>MY$1X0kDsI
zAB|3-BYjvKyt)yHbtb(xwryix78{)y+EF!ovzBc30EPtF;&knRHCJ*ijq@dOzGQOI
zj!55AL!{zdBgp7Hb#w4uJ;9t4S}%`Al*_C9ORggv>}$jshG@v@7?1CEIIU5d7Si51
zCKu^mDH^Vu;#O>w5-L+XkhaFN&wi1+H{Kvj5pv??yYJG1+P-wVcDk?(46{Gl3xon>
zyXU~+r<qG>J8C@C=wloNhAQvgje0+76VUNNIG<Nuym&!MNzs>i;?gd=S@b{qc=hXg
z@5C5+`t<2e&i|g@HJS(!UOa!!x4!i)Uc7v1F>0Sq^*6>nidIaD-dMq;sCC@`|M6J$
zF*RbK<F~iB%IzzaoHH+8ys(8{qtWs%T!Dv&2TrFG9Yy?d&MY^J5vQ;A5+iM42g<tU
zY;Tg8Ox^$Pd+&0-T)4TNc7piLjUuZ*C<_>UZAYA<72|C{^F}4n{~ySebTa|2Wq&@L
zZOAR_gdwM@F^c_)R>gU`7JPO53@sj9@GLPBL(nngJQ-5&=&hOx5>+GCgv^1q9I08y
zP%&r$D7{d(3w6EFvRO=-4aI2C5i8KmH&jIF8OjS5T-B7IK8j%P5OiEWEs-b3z~mjS
zH$)=hE$KpZwAV_d)&se8&Z+RQHC~)n)?8R~BiB(1w*9&SujH?fzD`7oekbgOsmQQp
z8IgKT)y`IQ<(h1L)nd5T!DF^+#;#c~!Bg)FTI7@wI?FsG&XX=@rYTTc<@QFE@+sTM
z-I}am2GG8Big_Vs4@=PZ(`v!$I;36shP7yxtlNcymKfDL^$~q#i1ZH2q)rHRObpC+
zs5Z9qg%FEo^_qWUYuSojtO>En5g<-oIfRjdrSvW>(rudR9>ulU#c`s-jv&ndkfx@|
zg3(8Ww66MRFX*IPY|>>S&s<n9E7H{7=CDK)H>q;kc@=$Xk2nfIeb&-;p*0;F&j_to
zGuKvyddZob9}u@7eNuIO+OEx1IN)x)jg5-}6=`Lw33WLZ6;Vmnn8Svmux1EfTEhoT
z11fn_b+Fy7wIvz>7Z<{GKy$l2WlPjvaOikmjcu#uAkTNyEz!$LsVfo=Oo!K)<{PYQ
zA#EqNmoGoj+V+VHz}}oHxG?l@e{^*t_TvE-RmQfWhNCiytZIbgrt|b@VcTvW<${%5
zTvOP!AlSJ$raaS;tlJe*wH8}&^UkSRmW4RQUDPlTh;FaZawg}(wr*^fZI|*-^K9nI
z*TTsD2y`8p*cccaYk6DvK&(3&yb)Cc3Yr?AkwUs5xIUD0deaRe_m&Eml(}Rf7wwcm
z6?(ySbHRrYgemCf@-~uoHi0R%KxsM(2op}GUHLXP&&O&hec-FA@@zpFwiUa%tu2z?
zc_tqT&K%|GJ9-(sM{1?zN{f;Cc*orvZ{j8mBzK3v-RC~f!;2Txt~@cU$Q&ImW9+7P
z6HS~Jfixx`lx)3jxM7TK_oru|amu?)hFm(d<IhQ5*>lc(^*8?s(9Q4bD)9M!4?z*-
z+q=8*3GAVH{D<o%*L&mLci-im_rA@;`6a0)6}ZUvj{=VG-9<0xSWGib(GAc1C)K;m
zU(pZ!5Q<)J`UXZq$mMcj+qT^Trj*LMZIsqnmWA{A%>Bayy?1VJZ%H;ziiZOu{!7zz
zu;TC40<Kl9I)beSwN>7G|2@{riJRjc!AEY72cA87`sg|Qw94i0K6H7}rexSs3_;jH
zrW(ZxGUfF3qD@hCLfUjJo3~8PmF^<W(68V}+?d{*IjiC9GzSjzVhE&Tj#>cD)3ght
z*ad025q27ywQN&~u3`vBqotL$o+)eE3r7)$pMu7e8R{#}2C%lKUWIrg0@I#CcNO9d
zrFyGMw%0Y#t8(Ag6lZIe^Ds?tb2IZ95qiJlk}K!x`QYORK7Rhd!&YA5kc+KRvabUJ
z5$so`F3Ra$*G*AK8##uc>reHb9w&<B6gM;c?@nl9h^IX#wN)$Lb)5-8d22Nk*3~98
zs4!$2lY%~o{OssoDT-+6<fc)ogXrnKSVP*?vT8)g!H)I~?^$Q-u?R>AHoC7y82}M&
z%%c@%;x&DwwoWcu4EL(`DuP(C;>|a3fvzUjrRw@xN;5#FsQ{!4xoL5Z6}Q@?*q$N7
zcTz<oW6Chp?Ss$QMJYPoMTVV^1!z%9CT$zt;_Aet#6AQhctVUibK6r|xPmC%l;d4N
zhqj$mg)4&inVJ)q`xlgQ;pWMms&4z4vR)_+W@aQ##OSp#rQ)jc2`8_1F)2Zao_V>W
zQVqyb{=Rb_CxI9y^P3Hn-0i*<p1t-NT0h`@q%~E0pDrKM$~{kR-Xi!37qqd+yD#LX
zk(%AnBr?aD5Da+)O1hBKN;>}wE&!_;1}9*nSst)<3?A79x(*1v6B<n7V2V^Jlv;?Q
zhRZ1@mCko1Qi^km0FLIAs~|vYLn{bI=40)o(kM;QwwyHfY+erB-rf?YxGz*lH$>5j
zzor&5c~hm%x^A>4%!h@W+Z*B(X|<TI;qazI7jZYrK=#bW0Tp(;!N-I{7kSHgq}yVv
zJg45Oi6N>gpjs!F&LuT2TO(B=cjZkv8bLZ<8k2`93NgCfY_xYwXcCPmteb6|7e?-4
zGc?0l6p$jC!$wC9XG`ApLa)|py+^k?T#NDGC*{95Q>vN6y7vKzQH2!S#(Z<fi_<;N
zKJx|qG+8W8VTzs#8g5-H+vS26EjE3)TI2MlCsX~mVC$VKDlDqK(ew7ocp#7a4)waq
zX=U_fLMO<0?N0BR)8$pQZ7H9GV4>*AVL31_^Q%bY4(+~*7Wdxx=G(u|2Oq!BIXzHX
z*^P0pmCY!3`s2U+XCBUMOuQF0Of@?nrDf<w-CbwPiRI=%cb)D!xhiUPb5ND=2&|7W
zGS3sIlNNcJf}lvT3xQ;@W-&&FY3)`U^E7k1oI(7~AFAfFspr8^F+ty#_oSL}wR8V)
z&&!why!Gbis5x<Z{(?iCm@K=qi}Hql5ZTE!XG5?Gt!DFD6KYXMex*osc5#OU-ke|K
zt?4yt^!$VOzs8o*jtVzXrDL^%Z&+O+LYyMg6bUvyjnlB=k-afio56a$f5A{lHQVh$
zN$pP5E~e!R^>QX(E|j95$<gqW*8<5Y-A0_)b9UuTcPl`g*CY_!b&V78jQGxKHZ?AU
z{npmC4V+VS-_1I4={s_#wi|PB97EvsCvS59+70hLKl9Fu6YG}8MX>wA?td9YVIzi#
zoRTJ=RJA3FlM0-2O4<yI?3uJ=9kr?0Kmfl_T@+O`XNSG8HKeSWOTcgpK%+wIJ{=PC
zuXUF4p#^3(Bnr@!Co7P8F}tc-X;}-xV4P1?mMfj7mSR%~y)F(-7xK$Y$5KgT2qzt3
z3i#-W&a=poj*!-kv?bD(%!gIeUYbqALh#g<O^BkV+nVJIQ}CQ~RnuaO7cEx6jlxt=
zp;XuDK`xmXwE)*FW2FYAZG+yK=dc%utu?kaUnf&Rh;vkQE(BcEBGBUy@iZPMBMkr|
za6BAnHFLTDnDu;5bCKzAM{66@q@2y#5O<;EdwR{#Jb8UUvSaHixC`;1Ngn2nS!)5;
zS(YOZ7`_`dbxJS95b-`DqUF?KUMPM+=?#u&Pd`sCj+8R3UJ&ni`t0-6{KV{aB9mXZ
zJe=50D{h8#{+M_?aQoU%Q1eDQJ+ShC^Bei}f<PnsKsi4BL~Glplmx;R4MLxNk7(L}
zZBpa!oT%l;AKV4reX3)WI)c!f&e%0uvRSCNa=PehmfTz4CHv!hbFFMSv87DQbsuv*
zd3wikSXAb3i^g~_bao`GRei|vW+UR92|lpg9yuJ3_-G5-6o{=Ev6cM+fJX?8yt*xS
zt`CadtOLSk?+mMr*m}=M@P_(9DxEb|E?c8CN4GQ#P2y;H*O@#lOJtdX7Js5|KncmZ
zSgB82MjW{2_J?tZMRfF}8>h38l!6NyThr;g<egFciTW^&5ti}?Rl%t|3XczA@BT|k
z#Hz4+%9$DEn-H947aKaTURHA2APfZ=1N1eKkH@PYr;Q_Eo)<%T^;(MlwF;5?Knq@}
zqU>dVFu2i<PKa=Od%JsFUZ0M~@51hqt_Z+4zVQt{`0xYH=QE`~y05<L_@lpP5%X8H
z;#BR2;_lHmly@VibX`}5C?v$dc{}d~OesaJU(RQMos0MJ3*|1}-rf>p<Z`*}h`^qX
z0J{<F7<<AMg|?H(wkB;5CXaKSKlt{ydFS1C`H3I@V>ofV{r1~@^x=mr^UUddf=93Q
zX&xh<iry)*kz0bYffssBZ0m_1x%mupI3TX@=bruRJiGloF$r%TU#G_dfBM<a@sEH1
zpYZDsU*&n**s7uywdix_ZR4KIoLj(<L<ieu$hB_PI*rf<&ik+n7ND!gQb*|^-n6te
z)3=p$xscb)mKE8Wq9s?fIyAVN0%A>EN8`$d)m9f15RK-0{M-g<D~db?MV7Tl&?E|a
zwM^I!zT;ZJEeEypYL%K+Trad-K{|&e@P*IZad&s*?f0JZ@?oP=Y??G0c)d0wgLg}b
zI7M>Gl+uXPn8-MycnXC{98z@Bv^{zhhbi<cYnx4Rbn>CRIf-Fc-|ilG=8?K{5k*J8
z-psF4+Oz>y`75=n)lrbX^UpzaRBjydW{qwWv&kBMx8AaKO~fKzQN}QNrWR<X0(WLK
z9lS&0ggDQ9Sg2VkP3f{Jx6D3gGo`L{sI{0akL)V#Avzk1QEOe@A<Xm7J7IE8?Y!)B
z3{Yh}*V5I3rdH+N53b$Ft1d|k#yAIj3_3pdL4k>``@WA3j6<mcQeDfQOivN#NI6l`
z1_JZ#QyL!U!21J-!xJvs1#yA2Ui8_S4ovY5*DJXsi`fG=A8_L7QrOZ;jE;s!#8X?M
zBS?2Rr&+!}1cDF55Gl1mn27F*g2s8K;hFuU5SiuqmQ6UH@A0ycE`^sb&N%NmM28Ox
z-bK#$D_eyZFQD}kuOHwLpok;-NLY@aXl?s+3P2kp4YzSYJFXy3IRSP!r&9~rXZjAR
z8K9cFH08{}xJ2V&*LWlu99{L6E1gmrsbo_@DOvyK5V^b6K>OX3iz`|MsJf?CN>(mn
zErpT`)liFNIdHf+syfNYrQJ5o{Xw<G!cM@W8Ps)T_t?x@3m9d0_t9lGRKZ|l;j(SC
z5~&Dl?p$&wS4ZuhW+UVvS-tsuCkD;Yo#MXk1Nva9;!77ECGzWg)}pScRgF{g*nJc3
zD6wHIjCNf|2RFvs4lDkUYe8ZE)oe%O20ilb^9VI3j>m(Iw^3Bg(B2RuhnrhozWWDS
zU{|$kQLilH^^GQ2Po^OgI-c25(y^}jYkfCPCVXef_k-e$4eUgKo7>y{@4oY^#-Rn0
z@X^N~^9SGl79W4~T*ri0q57wEjAq3epE1TaD~`cMELFLEP8@en?o`0i`i|yhqe%=7
z-tFxz@4WL4)2yh=wr#r+cg|PK6Yo7QUc4~j!<EAEy{<#+wt+ckisOln<K@eHO078Y
zeCbPH2B+hD?=)!rv10vGJ9?+LOzRMR(2YE;{F}><@%pnTtd}!?B7Tw2z4iqT#{-vC
zS&pw0p9R*-g;p|Oim&nPji2ME{1^F^_kNSVUw@ZPv<RL|Xya5JLoSICo3^c)QZm6i
zmN_c&P`zdFYWhT~>PP3CM|wqCB}&2dqGuyzw$q8rd1c!wTW-Xl8(K&COd)jRhYi|;
zjtSMmMDHC3v{Lu`#)YWdv!vn{O*JaJP+^*kh~Lq;#@;D8Q_>2TmBZZ~({eyupsk&B
zJ{jdE(8TfV=?yVXeEl1Lz_xWI-H-cNu4|53HaXUd`WX)XT&-1UKA4a|6fHW;6I-r|
zB$*0Y<GrrydZ+K4^5E44qL#|WM(|qf#2EJU7BL!+^2yYrF6Cry5nAN54qMLL7&_^?
zC%cMt@L@7kt6%Me6s6E&)SjOZ1HCG;wm0v+BQ#_RyYNjA#1CWe4&sPQ#K%Bh7PeJA
zXce(!ZwS3t|Gq6{QZ*q1YBi?ob=%Lo12u6L%r4Am5yqPwXtjIvoO|aH-&Hl<t3BHR
zA0r`pVpM~9=e_aIjc#?d2)65~b}Ft5f9`X?pt<om>qO4aNC-^R!iyIls|Zq@s%)F!
zrL(3&2s3n8*9#INw>QsJnazw4sQrR-RVfYPt!~)#Gb~T2U5Mzp%PA3jU|tT!qaWtw
zY6LCrRt1qYv94#<lyz-vj+YNFxw{M8+}$xR2L)=v3^Xq5%4NOq{>S&E(s|=2Kg;3S
z6E%Yt!Clr*w6uM44xo|{b_l&=q+x_8@(KbtpyFYz;hfr(G&6=&-zjxNm)7^ur3P<n
z=5DT?)x;3l3ONrPr@+lJb2!dSOWbe#t0fHTG1Rmht!DXBMy?_nk2TGcdZBr5BjN$|
z>6R_Vcjz5e96c)Zbo|JEr#@Cs@f}KKCquxlUmHRsZH3DvQ6;dJPHsXs2YqKFo5^+N
zD1@k1E<S+w4ei(5Zt3EjGi1?V?O^^@cBoPEqtUL&t_0TW$xZ?vBOw`<38D*>e$6!(
zRj=_Vh94XB>VlNlmF3wjewiON@_If*JSi&*>rgf7dO34@_e7UvGiFs4+y1hMsHMjF
zbSAiKC+^V^@O6+iuAiZ?qnmgIk^6LxF(&aLupE~C47~ak*PlGj@$Y}*8@&JCdwl%y
z3rZgRIIf@f(f58o$72L=h&zTM@!Su<uzA^Nz47$P6OE(mohP?XcGb7qxpYeoi99@<
zsI_uBoya+Jb92M_eCEYRFF2h}dxP=VH!y@?0vzQVK1S|_!c`pPqDXgbjR3s!?z=oZ
zoVb7a!22J3$S?f-FVKN)+xYO~k9hI&<xX(_w2q&CJ&-<nM(&P(_s4#T=~CHV{2sT<
zLhl}bH}mrTBR+ib0iT_Jis_)n==HKuwvF?{iHFk*nwweN#J?7QlFz5llV->NSH8}-
z-Mf^YX;pb<-ptW!t3>Y<5ebGOx2A@-UA+f2DWKv*)Qz?lCMo!`(#lGwlP{TcTFFT#
z3$=9}Kbk$(6g`J!B1W@#D#{I1TA=J>N*VXe5K^HwFTJ56gS8lHR6&H}jc_|nAj(N=
zr4mq&*mpj7U^yp#<a2N6Bw%>Bp+#oumFWgvd+nAloL}c#-~Pz#gY;)V&g~ssy^)>V
zdNWR4Wtt+bVZ}*RMRF7_TUJlI(Tvwd7qvzjd~h+LjoML}`$H4LGV55`U#(iYp=ZIG
zrSp2)dEw6KNLX5DjzKMVx`}_Hj*x4sSM+cS`--J|iXwE)8XwG7MvO+_I<%ViJmL{Y
z*liD$vH2PbKOZK-6!wvDPT2q#wIvHph|ROrRRthUsTF$$v+)w0(<zPf-W2T4P(Qaf
z-Xk=iKWh4NE>~2vcS2BPaEND`+0ZEJsbGTcza8mK8y7K!b<;hkmdxZ8U28O4Z{(Cz
z+~%Bm33fYQ4qUakHj<&N3oagL=|nCcG4qsGH(D*!s_qpY%{y*V2vfn=Lc5%)y}~@v
z+KhBX)0-P^8eL7M+e{Nrt%+%xSQZs?Y?l+IzKfp%PoCVd9B(N#<L77eR_Ng+-B0A^
zSe6@JfAbC>C#K~{&k2`5+5Yv33qaL?nH`L6s9v0oe>++sqZdh}fiz0*MsB}axYXVl
zA=&$Aw)dSQQ%Yysv~s_+j&UqG&W^j|!jom<5WEt4TcZ@^=?wdl)+)JV(w5YArCiNc
z=Xq8oq;+P)=S~YZ_sZry{Qrhm;LaVvwq`2`2k{lIV`BT0Rew%9GeagN1l%-}ww2Th
zsS3EDP&pOv8j}|$C(M(>2f?|H3)by9*$Xq!La4TC5VbDA?$P^*POWLurX0U{UX*{b
zm!#b`0f%M@Ye6%9hU|$1ZZG~A5BaF8MD`-O_h!6`%CszWI+t~`tVsQt>$d5r*uj=I
zPNx%h+s4%z#|oIH_7lCccgrPbe2nzkkft4}v&LfgqwMU~96Oil(?*NNeS(fT-+%vo
zh4qieU195ua6gXTJGB*l@AtmJ2k*bn^XD(g*?6ws^X+|q?|F>0#*eP0QcB@+SvlU^
z?v^gmHo+nC5D^|89=Mz@`yyUz<>BE05n<c5D=N;_$lZItCx~1wm%Z4lhSt9-B)Wd4
zp)f7gkm}j)>DKtQzw>ty3H-)C`5IsQ`q%mSpZ^8^@?ZYTIOq8KZ~r#`@&D;R=HcP=
zsb1H=^shbPgWvrNthF=8BYAy+?=x*v4k~z-!&4n|2OV9}J0+c{>B16)IyoLLXKI4m
zqvy5PKFj^NbN;=`mv6tsKaKyCbJ%z;_ed+mIpC@ay8vjFoDwxS95XIPMNqPy%iuk(
zHG*sSnyK5FmN#0d<YcC_qW)vSt0!Hmo!mO9su}v+0`sI}E%mrVS9MzphiynTpH0K7
zz&Z1w(_$(cr?a0g%!ipUO}i~qO6vZwZJE*v>-oUEEV?m=h-;metNC?;kAb(I-SP7N
z#Ea8v@9EX&R>WzsV`2yuY-B}=aMWJ##`BvdRhAx>$eI$aIdZO4>8j$SU)Py2327r+
zS<@>7{e1|gTDzu3_14IFv+wK7F`C-j5uz5lwRgk^q}H8hI6@B;c{D?6n<V)V_U3E%
zI2=#gX!u2iEEwgbS#p&#nvwHD?;dR8<LMmcz`<hPaw()#*|vlm9DXOX?)H(eK9*v9
z%AjA@yZ)}9_<BtkwoI}=>m@5!a_q6HGgJQIG>2V5zkj~2)3Kqx6hl~gt5B5N=_INs
zvK3C-OM;JtX(8te=ag8MBiojkyl0+fqz7`g#vpG9`aDPoYK`bSt!>n<jc*7ONJd&F
zt-1$#Kan0@l20dEN-7|6o?a8@hm~bH+JsC`NiEf!I3|L#{obn>?s)8c_A{SRfl%2<
zNku227o1ck58-C!csy`iZkRk&r~J-}BI=*y_{0U^74_2Jhos&kgf93;U#Dnar8ajR
zIogE{GK#f<2x_cZZLpTkc`clh#mk}i(jhvY-5hv&b6|;rli|)M`vW^h$SG}XX`|*s
zZ))y&JRWpYHn@2dZzCMrJHPul?P#EVjr7r(+5P!wY;Ya@Ub#>E!##?{-gkn3>sqA5
z!$Qc7%Q>0LdS(iaDRd^Ud_OOKkCjt3oJAUSi^wx4`r$988+HK16nta++4k-W*zIS!
z6@J~`oAVxqA@N>t4b<;Sl7>N?HZI`fpJo3ey|b<t4(7WxO)3zfY4NYwvXzuG>*b_E
z9<vk~=X7-2+L6MDcvG}<aLJ5JjGB@+n%UT}zXl>)F{dBQv89dQ`mNvMn{U6(?d>gh
zcXu3*H~KIi56sW5_tWKc;&*=gxB2Ebf1mZD8{MZ>!;HdjoWp7xxpU6^DOL5iuIiH&
zLf~*XGO~XG&X+T<J$sE(3NK!~AZO+5eEj?c_xJbqy($K|TrT?}UrN!Du%R@gvD;De
z??4A1|K092hBktAEBwY+f0N54^YyR&7VEn4?tAa?E5G))dF#!$c=r11Jh^+Kwlkk%
zT|aw}(kL%KM$R8`P6<jOrOeH1&xn&J_9y0dmT*0i*Nw8Bp%pEP(#GT+#S71#9H?~G
zhX=m=%=6~6&+w<*XJCqat$dSTzkHP~BpzDQ35SDqOO%|si8pGk)Ea4c#dW}IQ|PJm
zlA!76s-&`uO9na=#VbNtHHK{~+AO95TWU-}1ujF?>byW#WU#hQsoff1T_d`5oEH=H
zG|l6#jV_&SOQ=m#KtzkloD#KHdWLk_bd)$pWSVXD&cvq02FB<6!e^hby<It5!`j*a
z6=YwtolPQC5v^v)DQCJ-147gZ54|x(Pw#=+je~BfJ*8H>S58+p@76Agc=fIh0m~E#
z!RuaT{*3j}sB<iM%A2;LOw{p#-Vmoq`w~<I?jG$6;;!o}Hm&H`^)+f#w`e!ll?T^5
z`}9pI69pNKm7Wl4?cmC_FwqljO@YJBfoTfd-Y#rg=EeO*g>#ooYdX=`)3$8N<-Hp(
zqw}7wo&Ok1vx{s{;H4_9K;tIO)}~a=oA+8!4~4T-F?IHEgSdho(2$ntMI-NIT8O1_
zI=>*ez!Yvdmk(ICm$Y74j!(=|YsI-)=}hQ|-Z>%qiEiN(hv`W0H-vB?Zy&2ZAYVvz
z1ApZ1<_%g8w0vSdF1V-9@beuV5BUYXWvGeP%=cA<!{HWTp?A%~-qyr&d*J`a-hVY~
zmZkTd*l)RQUvpfV$};U64KxYRfdDxijwlUBBaM70LL7x+WI~}^ehf+9V05jR2q{7m
zbtNgrP?{h)<PZcM2!KSRtE#(9nqyzH?Q*(UYriL}0dN?|9)bvBcVt&q<vDrwyUkk9
z^MC#Z0WQoe!bVdRX-`xJKX=S-M9YXnQk01fQIrN7lki8$N0IIC_jvaaFzMm&xzix?
zFSL{x#9Kcg-8knQQ5cr+^pn?>5T_R?1n?0?>yrp#_6#;KrE-T_3R)#tm4;1WsI&l;
z@`y#0N!6OWA)gl<>ntKbS(NAn?fXf;vs|v6Uc$vSJr{q8x0gjA8D3vDHklQOOM#+P
z^7i>1=wF7?mtPCY6Y~OHHzl+4#3^-4MhOZn&@vl3>%>8Nq%EhSFPXy1Q+FKm*jr^H
zZR#LT<rvB8Lt=__hkOo-Pep4U+U9Ok3YpS7ayKeZh2P`n$+hw6PRHzfm+m?cb9ShZ
zpb~#iDU~NWG8dx>9rDwJ%SNRx8-_<X{mdv%l=sspG3HJyk=Nr1<x9ylk0<xjANc;>
zM~n#t6l2=7Qeh-dBhZJ7WLfmtkAK8}`XB#a`A7foA8`HVO-dv9s`U4rj}Ss~vl5T|
zKkpshDXg<O-c3;y#R=i_&NGA_Yb|9_o(erkBiF9$5yTTp_^^8*acI-ndr1v><SM0P
zf)fG#t-?NCuXyqx3>kGmM63&VKPJ>R9QpLqPwB^zX`C6Rk^SMo%a<>C|AP-t#pF-(
z2;R{juM?^^LX-uvTw}B&M2XZDrmP5dN->3DCg4y;5rjgUnzbnqT5u88INNd9zvjhN
z&HLxjtZMeXXK<FkaPe*aZ1_5p5B#nDuki!-1*0DjQ6lPwYPBLn#XR+hEC$d;5?V-M
z^7uR3nFs|U#Xg4Ibg4XXQBmNOOyUYJa3U$?Cl~M$YNO!9!%j9lc24Wv<b6i;1X4w$
zO6JS63s|3%MMA{eNIg4Looq6q2ncPEYKZ|;kjI#)T>J?~bSWBC3h?=})TL&-E=~m?
zu<$ufE8)^;k}Y)vOZrSoQYngrz`NO@q&$tDjm`$w8KExAG_FpJnXLJY2`MVeRQy?&
z_?6CDiacF8i68QFBb3P2D+}!;q+&{l$Do7-NsTe_%d_Yv2_HQ1v=(O{Px00{5l=;;
zltoq@=ZxVHlj*lh@4Cz&hFE<1czoBE_il+8EVbb2dC6I$xZhi@_Y;SC#@ZBVmzwX8
zi^-7hqd56mChn!d2MECtaz1_X7EGR+D*vp7k`jDCltoG=a&tByPe4R)9z^zalsUpK
zqfXOw;Q#*F|C?XD{F8jQxkNA%<AhNaWwik%a4}G*3WQ?p_moBQ<}8|X@DuZNL<f+i
z#3c>_Vh*SvK$qC=F1=%gMn#RaJ7l2{MUBgVhem)gXkFriLg<oZfAr2m=+eFw(}+KK
zn~Gqiu^}Os-dpV4qf~lUtn-vbNnO=oD(rNR8x^WvBSplynGpVfIe^PJG%nS@OhuX&
z5+mfv2M`hhXx)qu=8*_aJUz|}HOo5BWIg@tIZlB&q(Vm$DFoD7&=!ieQW%*!4PshM
z9$yYJxjl8QS{and*_ujaGs{Pg;8MIkc3_KQB^9_KEM8lwvyL(OUBwWIQW0c0p=S%B
zdx`ftefnIGCKKFv8XqK#9z+b3MzX34Vpc~<B1&mYVUo>(Du^;!!6dGQAcz#a?qfg+
zGQT3DnEDZeRa%~Eid3AwJw_2rs7~&BQ+JrW%2ev2#S)Uvsl$U@!-+>O;0eAb$JXV$
z0$F{Wuf2$1hY<m&vY;@D3zEh+X(VD@$~BJp=L#v3HBf%8j4{mC<}pw>jW<*MyA-If
zcvdZCEBrLetKa>I#O$0urLI<_rXWS`DIR26j6eD06aK^h@c+d9{VltP9l<{;+}^@a
zf8WOmB?$o)!)aV=t)(ohl%1Rj;W4Ijp7KVka&9L9`mU$RqwK1zD9fb$ou(=Mx2j@x
z$!L9_jjWeK&-*lPUg#8GMRXrO-y;_=-D43`GZ-RB8QE+%eEU1!;UE2@e?nE(NCC66
z^nK3{fB3`H47}6T`khBo#T&}9P6oB@Dkll0hrK@@DD8}!W=s)i&(1KWCHRTH+tIAf
zQN@VIQplFWuA?Y4ZkU;Nx71C=+2ti$B(t4Zl`F<>;J<tRXV}Z1<5v&=l+X1w*Ws8b
z5=BGMDp5QZb+9o&AbOXqYOH^ZL=0&%5th-TQq03dL{OJ$<U1#29l9uRE;3t}N5ukS
z8Z)OH*fhDAeU1Q#P{`!(HrY%$F$FH9?A0k~Dj!nxm5}+}7c^B#Ri@}PWeT(@6OS!K
zA_1v1-g$~j;A1L?wsp;o?uqX44kz?4Q9g3@mwY~TTDzT1eus6Pa(WlUZi#5a$87rS
zi9%461%=jWB$!dP#jAA@eS{Ekc7|O1EFOePsVqiJVh1T6P5o0$p3kvo<h>yyf&#=c
zg@O}`x$Gwh!D5q?>c2qh)gK#i5f?NxD|->i_<1Qj#bT(Rz>6fXcq~?<cq=_hq)Cnu
zf+rirwyrqzhWl>fIM2*3?J+q-{gfh;0wKZ)RefByk}zg5wszJLlsI`2MiPxAWE52u
z1uk<Fm0tGIfb~-nN{RIMDU(D;AK&~MpWXd4o|b2PSU=@YZNJAx*AOkr6~y2Ot)(t1
zgs&KfJ*H4pRm<$gRBZc@M#2i*+%x$-QpuD7y?K@#B~-!eCZsG;MMKqWiP8|`%-BCb
z%o*!QboNV)F)(&-2%#iKM^!z=m^LBmQ&K}0rZ|ag(zMbc;*5}n7#vgIAuGYybreNK
zv@_n##QDJC{>yh-+kQ$2pvW72BJpYgS_rfdnGU)f>ZHs%9YSQZCZljJ<vR|ur=KHp
zfFP2@wI<SNNn2^^LSYI?B;b5<q8Bko$0gw4Gy*kQWw&r8maka~m4q%WomBQ8WU)wC
z>>6w?0&@YFg%gsj?9F6ZrBw2Of>VbE32}4Irv)M`1ngx*n^dxhwA0l_6Dq?vC0?5_
z1x^V1Y34Y0xO`rTPo^<SQW{C2C1s%~l|)L7T0Re)6ujhaGL>!F;5j73EOA<v4I^6v
zECn&C*jo6J@hvj8%*AqAYtmT6$C&t&G7)x_kk}9@wITWWBr>qIGr?L)0cBO5#s@`d
zU>r}%I7AvTWw%0NE-YnEmsrr_G`?Qw1#dljKl$7EJO1;3ZAS<x>s1Pg2w>-w4LKC)
z=a1va4}bVWj)&nV^b3y*=<oMT^RzpUMAP44?aVL;L@ls38K%md=R1xgTI&;{ycjX}
zeTsnd-ZPFP(=;*9BYoeab&7YKrs*_p4k1t!DY0Y7whM(RSd@TEQw5Lj-ST~JfBw_h
zKEIz=msd!sSg+P>w;N&zw5yige$SihH{9Ia{l2c#KkpC%3N7jSfq9<r`FU^K6;dkh
z?rssmvsrKPv7qaEBpuET?Du!{#|M-Evpys05g!A04?C3i*ls``4{X~D_V@Rcr6Cq2
zWnnlT_iVJ`zkT)#h^^w+jz8vC`ycSi?h=AkmUP|mr0^$(6tU?%NJ$Wn*V(1Wi;PMN
zP#UQ<AuBd*KY&s+>kWrPkFyg#`mCsbMDk-yqg3nXybgJ0*Q1JLH`1<JimKu6;Yd-I
zOd&AN9v>7+LO)n0=d(x;toN7@F=7#(CHp8J!Ot^AC=>DCIkZlQ|LojXnoL-9KN;ba
z5~=kmb`Kt7I@6RRtsBDoyrv~uf|8QL7@D$71y^od$Rpn+KGFNcyVBa^G?3(J=zOx6
zSuC-978pvUm@}6dKpKtro<fUL^L!fnLIN2w-*0(89#gK85aqN_Er#UT3_Q=qky25V
z)m!^k-jVXuX<2CGf})C;GSNwUg_3#NMcT_R8^w9kF#Cr6;JEK+2IrE=z7GWQet}FB
zcb{^O7eQiS%!%Lcy~o->U6lm6h{=j1Hp)?Nh_ny+*`D(0LsktdDRJpTq<tHK6C+>D
z2fpmS<kxn;#rM}=<IiuukFIN)YK8EDsoz1+DG5mioE;&~ggB>SI!K1@h$Ld_6|+~g
z%~QOZ@bf^ku9$}-S~oZu=<JcAXbI5~6%-iUG^S>Pl;NvX0(PQvDCO~XK<NVSCt|d?
z@tF2FRiuVUYOECu{XJ#f;@!wRJ8H8~rUE|9C@pcLBhK%ZgZ|D1U_dZMg^dy^A_b9F
zdTJTbVTp`}7!@Km$(${Sw~)bv8b$%$M<xfO6O2wWg=N&>DG;<u(ilN)pfC|BBZ+00
zMie3&#GT4cL5`ylCobP9V(o-HEMeV?kYbT<rx(gQPjD`wb{Y9&vAdB8(TO6XI(f5<
zkX65rX#vjplcG|}B_KYZWd29U5!+fsszPI<XO@EdVPdonWop{Z8JA~g)J>b>qYwmd
znTLVlxaT-@s9|PR6s$^(2mzx&h)5LjMBWx;|Gi~#5KCG~LMTP<I@8D~ou!tEL`+s?
z$p|WviZLme?Zh-?cXX{$A~~Zw@0qniH7&(zlg9KRS;@>X@gkZ+BSSg|=K{_M5IXrP
zLGEzh^03SN2PIN2Fr=~j67(JybTd9)W67gfXDMO-WgL<0^+-VEcmn}sPBhke_Q&yb
zZ_@G8JG&DfGjc-Wf=&zSIQCRhVC{(ab@u*ChO4xaqKK4bNpvGh!ZeS}^Td9?XB@{Q
ztg;i%yQHe~o>D6cEpaYI?ag*3A+i26qI(?mhSTP^@BmL6`&%`yh;Vsvg%FC@uix;&
z2k%oA1;6}ff01u~>s$C3*zI@xpZ}x(h+q5ll<0DzSG?o(vvy{8|AuMo2@#5VgS9jJ
zeMFf8Aq=xk_poUy=CQ-iN3@9qKeOux>e3MOieNerlFO&>(;Xhrrsn+VbF{7?J}}vt
zFndaspJxJn-y_g`ul|s4*U$LD@pFFY@A=XAnqGG4-Ug2gfshkdkQ%QfqR=?kBa8qc
z(M`)}XH>0FWl4WHCWnABHMhEL5JOUhN<#8yD~mLas%pBvM=L|XG1`fyu2TW36v!00
zRaTBwYchK5m|dhBdg`WPj)HwZ(zGou24)v2W3orlB8g~(n;E(h7b2#v*zXTXU1=(o
zJoHEZ$mQ8iYcSa!#b#@-5JJjHUcNrYOdD}YeYz-!y~CK2x~@=Cpq2Sb+m)KyBxF!B
zCFU$?IVrYOLgo)Hisayvpn+^5BXt&Qkh$z4Q;af7D0u`zlEqdM1cb1xeencXEEK1u
z5nsls=~WM>HCiNiF70QhdU@Fdmb9&ygqd0FBZVaBY=091(Fbfyg=}dePb<yYw&Zp<
zvpe?by5!Ulsg{B^ug`hk6^o`uBzv00)+_nX`po|h*lEf@jU;N4{aXkuw1%Z|NaJB~
zQl~?9q)?GQ9Qc*_H~0q+Kjf^w;4eJ;)BK6mcd^!@ixMRbN~N**91BWq@Lr%)L9mH0
zE_6#_Y6`PrvJP!3q$u#gBB&^vD~h7T+sHId6uLx)j>%2LFcI9qVr8d_EmgB+8t<_#
z5bcnFAUER1J!QMWhM8_SP?dGcvky6)W$2O;IYhMC;Dbk%fmOTCwsr5ew*3?ez(sf=
zkusu%PYD%-e3Ktr_!t!-q_17F4VU?68ka>Tn-+yB#8Ya61dJ3I1xiGsw@6XsLLm78
z`H;x$0zr`R@gX9Sae1rY%1Z{1O4diN-hzlFyM>fb99)WbT0D(})F)pvp_1p5P)nw6
z?x;uz;IiQ+0&wawm(-2rNDW!$?PLx|QD}q<?E4YXw0!Lw-{R@>52#ixN}C0;JdM$b
zY5Zfyf$4bU_SH-74tv(6p%lxF_7Jl<tXRU$-DxxSC$UFX)0xFTkgWdA{JMP2mPkJk
z%ywqzd-`cYX_?aw(!DNME6UA=!8)RRL<@zKtTr1KX|5-eft5i@ofqx2k)IL(@`zt$
zk&jZzI6GLI7rd|f%PsdI@n}wy;D0H|-|u#uMxQ*omBg5=IsWAwX%S88-vQY+Kq9wK
za~kEU6b6g6maZEZ#u=5Bf%80Lt;IP<Q4|13IVJ;lA=n*{Df`(uy1q{cN?drUkK?VM
z6!!e&^?CbPQ05Ol_>ld>1IOc$7w^5tFa6>#^7DV}kEOA%6g+$OjQ{yx`S<zX{~!Mk
z><)*YO7REjoWtyiG3AF71E#DHQZr67rYQ4TqS@~cv~7hjTV^+~J#Pt^gaGP-c^=TF
zKo@0F`3plo&**t13c=aM1rNtD2{l3>O7JnVU9XvDkDVP~*H8J=AN*;qyL<lr-9P11
z`vtet9aEf>C5Xtrp@m6d@kK>x408xnn=@SBlXV?z)fNw!s!Cj@s-&oEbW<@N4@d)1
zDYj?lC|&S=`vK#A&q}wv+THNs#WS8gd&<r2E!&GL2$5mva5e12h*jiqxuc;zz4NQh
znuq&6qYaoTnfGE0)>?*PWPj{2Ma?i+tdEE&ICdS=oYH=Dwhmex0w6W}DL-5@O(_7|
zKqSBJZt*%Lqhtb3lmu;x6DUyE73<9wNO5LrK`5<F<8vZqStO1uktHtCdyg^2={W&T
zE*488V{J;0S{j{cwv<vo`rl^L{mA3QDJk&ciKia(#7M}bLS7~#g0~z67N8^Y$kCGl
zsTILP^ggGZh`cX8(gTRu@LC9jUiOGcRH<=LK3TAUM9368@S-xTHWhEKcl16RB$Lgn
zb5f4TRdgz1(~UApqjFI$bplRK993nK$~QL^i6@>ut6U7R1!>>Sz>SQHUty+-07-8!
zaI^oCzt-LGx7PoNzx4DM`QG}Rrd(s^0g=5Cy|;vDsj3yqR3M;gFF+`U@rLMogltm&
zuP%{B1B&1*eRl)G<D$Sjfi4P^(DcV4-$O~cxndqAqNrKb6|;{78<0psNU14hy(YxK
zJa<S@vDrMw+KJoy*Hoouebx|_!4EScI-Hwv!6VFD)70O`@l%L`OU1|AqhCdat>xQF
zOF;-J57D}a6iE;uG6~e@7`MrJ<{(Hk4|)1~3z-VcggGaU!qWLb>N-M99)kEZ0@NmR
z*2F^OeT2P}LFwDY0{O*x?6LwejZJhSF~40xij44)mBNd0=i9axJ|sJXB|=SSo0;_U
z;^WH^WAf4xnV6fEQ*~Q$)igZ)+SfUI_5xvQmR4%%f)UG@FBO$V(_*TU_Tr5G{(<|K
zUnJ#iq0k{v2oTXi1RK&QWkF+oUbK=IRy>W3)55dxDl__d@^=DfXU1XR*pGN3qAajN
zVob?ud%-X{e)QWP<B9y-#S;(-^^-Cck;@TWE`pWJ=ARPllAGTt`;^21Oc}JvKEeCN
zJ2D}HKRMn_CF<$s{lQg#zsK>r)I9&S9o~E9ne2b&sj8BBo=%EVmkVj<JX@vF+F++S
zZHO_Pc!9~q-<`xGE;mbcU32J;ClQJB?yV8@=@aJ-;jKdbCoaO<1!9OX8O}<{?cFU^
zS#h>K<NM$LW8B@{@%iVU^WE?L4AV66(MKQgvp@H<{GI>uSNLh&Xzv`6NPBrj7%Wkg
zY|frCO^z6Q4#yp5m(M7x7Ojp8xdK6?+ufkbHD%E-`Hs08Se<S0G9}ndgQZ=c;l?A@
zdya$Su)m`|zd$R)I3LlbLYW5F-!qLPZL{Kd+_SCE_{$&u1^T-;d@+2^Z*a?R9X=uI
zh^gv?$W@NuXQWhALGo{&{c+@z3R6@k>Pv75u@@ron$_1Tro(|~XXbw5&<)6<=Dm-;
z#xJzr<;|y`aD8{p7wVdy`_6Y+tr~92TR!~gBM$cu++4pzYz%LT1Cw>kF0i+jf8c(=
zEEV-di&m2E;fRYa<wy?$cD4+|#PKjP%^s;E)_Qc^a`XBwMQMdVQ#YhN!{-RI)Uf)H
zIC=9tV_nK4UvU1E7?SK;thFf%UL_HhHio7s^M6YY7D}bHC-L(Hby25lmBn%*qj!sY
zLdp+M&$IQ85EA<3Lr7kh*^kb9kF%axfiII-(&uPU0jF{8vcH5pvQDVk+jJuV**Sw#
zb4=nDlKk^x@L9;1CL}_opSM`~<Zn2=j|;N6jGSXuVyBo$nMd?#^NHRgwB~G`VnAQM
zeqf$mo@_nRbC!KBB9h~ThzP0D$UTck(o{|+5GChg3z0nxg9jB);u@*+BUdrx1hEK9
zj<Y)Ljk%c=5dFlPhadBQIbQS4``_Taz2e_|@=Ltfyg)jOBw)uc8HXLRxj-w48%I=O
znEE4ip@=f#{6t||ly32UVD3le;{!Gv)4)cp(QS<jpmlK)xM))mLZCYgNHY?3gHjO3
z9-|9%wW27ujQyTzI<j7$A<C9YNzN}WDC-JTq78U&Vd~OYy>3~FXYaJO{Zx`bhyjI2
zs$Y>e_6*TJy5S|RVDvECkj0;iM;}-lm~6mBa0n8WHhs52lFUlN$+u7G;)G<M)PoUH
zueKPLdbBa8pT9uR<gRgnp-&$pE;cUdi<5b0QXD0jd;Y)p_<}~6A?*dN;&DTZIhJfm
z;!yeFep{({Qmke~D}iP)dq^o_&I@%RV8n9%`Fu$)79w?sG8e@OFc*~DGu~@fJY2tF
zfAfY)N<>H*!ajQ-8k5jO=Unnll#k*S=NvwJV3AF_<$~O%;LubA%$8#}GT8`P;H~HC
za*L^2K7aL^o4b36g4L=`e>Z_)nv?&cOhkV|q(2;KnwGoUTSD%fwMhZkX)zI}dsCL>
z>3mOP{d_-@Rl<V8a=OlmL-?=vv0NXibha%?D6leV)H2Nz!_ZTf75n{;Vd$yr2AjL=
zREQ<D@PY;|2>NU-<2cZD9m6oNG(;+^Ez|K;+n2=}<|hllN1K@RJ?r%fsRYlTKWE*p
z_!EEfPoTBp_V$)fKKX<XzV;C&Ba*-H^S{8a{{4S|a~A*8n4}AdDqD(jjntm8+cWHM
zDXKN0FcekAG^fw2g{K>Oj8TmH2ZpYry=XZelkh{ElAGJxl%=e63Z{1j!CNl2Yh+oW
zE}l~sk~g<^blok_-un<YM?%QWs1k&E#3ImDOVh6SUi&kAlM?mfFCj#vtXh;Vao+Rr
za1Vg-14W2LT~f3wWMPOJlv8L^vb(>g)Dm4ZgnGi;nR%M1nk~9&kwr4I{qX0%$J5~(
z{A>UzJ<iQseeeP@FQllzyMg*Jv))`GiwYMtzdU`Dv#X~JgJYft{`Q-H%zkaTy?#yG
zRP4rqPr|E2jgW>Y719_cp9C7p7_@gOM$66=MzLyJ0)oTgKxqtRQ6%$m7CS3}GGCn*
z#E^*hA*SfVs;ZKiakkg;&M}VJo<qvaUri%QmB)}Fq>N=dCwkDS__OYmA?|$^MTAt4
z2OqL$rb@z=ll6=vdUxWvBC_gNF7{&rvM?{dLiuaNBM?9?e3VIz+7iP?wy{wvt$|DA
zqkI$!F58OICVT&7Pe*GJGDVr8)m!@~B%{a(c-BpYdjAP;ZtgiAhh$a_%+8^uM4IG8
zq01_%X>;)|bG7DUL>q-N<w;~H1VoXnUPKIp=n)|hyn~oH=}Inc1v7$dHXb4|xJN-v
zTzdWpA@IrVZ}G|D$NbveKjY`m{usY_^?kl}{!LW-1Q{)LQDBD=7dwVIfm?2}L^-F*
zLI@0z(T@1xK#VDUWEe-<wT6&Hb4|NVAXD({ufIg=7D;3nEbVqf6i1G>r>@SZ>L<+3
z5@p5t#WRYs!ucNSXG~e(V@#&jszmF8d3FRbb8&u-x9^fy|I-)&$AE~E1d2wCNdJuR
zXu$#b-vuHntdF<=<Q~oWfVBbVA|XnAOrt{xD4<jlBSDIYDKuIaEKcfz#Zz!`M^C<Q
z;nbC7!te63@D>`B;j^V+JXH%~;R1?GrAVmL@|wkvkU_+wf#-sru@rx~V_AeDQa(cU
zZ+qu0XQMJ4xOmh(b~%ymc%a;zq00J%e8zl@#qv3_qfN}2$smYQ5|m>7<UMp*ar5Pm
zGta|kRhwan+FE?u7Sn7<-JtWSP?RbmtV*af_RYMN(K`0M<uE3j5g!H5o?haSeEHdr
z>H8_IgfV7svK0DV7@ZcI;7~#`^c^q1_=0E8pR(Sp`0~py@jhe~?c2pbdZ~-D%xdfG
zRhMhhuzcNe{e1r3$7Nu-V*lD6i`XIrPu~wj1kGE)?bccP{zz3dr+28z#nqTCZKf%i
z-AXAL$0P<Bhv8)Qx}?h_)L>CEKJsow-t4|=lk_`}<<Hx;VH^kk+%NwU&!0W##f#^}
z82R$cFSx$BKJl)!F}!&H1;!LPbM&Wx17=h~I-eAR{ekQ2*Q~ae$f9Al+p}s`%>IB1
zmdSf`AvxaP5~5%uG*L_XVPxI5NTrD}vb(#()D>l!EMn%N1MhiqamDfYK(pRZR*GRb
z;G@RW4VUYuL>q9f&#%4D4JecmlrcCrV~=}8k?bRyy26D3{eUPnx=P6<-h0ZbLd$~l
z%PRzl?|htQDkBNOvs$mInidyQ<FMP`aj|KUI0~)kx{ji%n8zuJ{*<DqYP^MYjO_03
zX;ukg+?FlI3A8bYqUL|}OaBg^{pd&B^;akfR+qR&F}4H8Zlv9;*{fUr&h4+G?3&}T
z<Jsj0OvOZJJL;xlvtBb<%epZbt&<^ypDD}YWI7$gqqpG_eO6UPvT~6s^NCaBUGP33
zz9C1X<x%N8&q>@Dmr?El7KGE9>9U%B5q?bbObDbLUYE?!wb7h}7<r9W+9aR7;7{)d
zWNu-8N0u~>Q^6P0J}0t;))If0ggPYuzvT#ThZc}|bL1*3eMvGZPGEx&kCYpclQ^U{
z$%1TohnESKNIsEGZFuk56Rzbw{V)=UD4`M;T<dIKCK1}C`=4Vzm3kECcxMS&fEi;z
z35hT!ZM9ND@EO@nBBCggP$+nxn!lJ7+rhhJWtOj7`Wfl_V>XSRrh&iL|4shM%|GIs
z+i&papZ_A24P5jZr3_lH&{E;z9piLFmn(!!nb%HOhW^O${tJ{=Ts(Ons)S6=(}Xc4
zTA9=0k<em;G%I2lnT7$vkboDT4Bkzpu#?5w2_YlWADE7}9QO|tRYg&?bp41>sQ_$D
zvX`0t^iE6LI~RZ{QY=C1r36FFFl;i%T;gLw6nG>uN@fQkq)j22_@;=r5K`_18J!cv
zlzS_nP$a`ut+BIbocf$a+B}XzGoe>yE?o@C(n3Rik@AZXVUf@;4(kgNck)}xep-*{
zTsosA#o#nX&R_fX=y$;(7ldOeK)-q{DBAL`nL!~R$;|?gAutXD{b5hlwn>PwbV-nS
z6>&khavBKf;rAI0j1tQAmWvnfbN%Cw5n58KRLIHYg_knJ{@L2aFL6pKa6ZJWwkBq&
zie#7_58c4Nw@g9eh@79FQMMaiy?)6!42Y0~K|a7V4BWkWL$%r@{#bN`;MgCIy!ql4
z*3EqR<x2vA{o%-L?c*5XbVX8uTvgSH%Ys;n_;f8)e(|-GOny$6H-QtUBJy9gG4Q|4
zBLuJ^DeB5H@n=(_$4HEkrm6Er8Hv$j?aVxlNy#U~Dc`p!Qf}wrc%bXL)Fe!Ss;W|6
zuum2~OB`0nVyM*PJjT)eZoI!`Y`l>DQ&P&0xw*dKzyI(3d#u-M_76K=e);Nzth8+l
z0ydisZQC#mX;k9g5e5_)v00*}it2)Ra7<HV)2#5m<Lu&+hc|DCF}<%^R%|YwAcSIb
ze#z_|t*>y-bM@>Q0zsz=RG~0sjqMX%WSX;}1~|WZjuM`6>M-Ss5G90};4G$U8Tuny
zIzsRmtr>?sp<hDi0}uB%Oio~wV|0$DsF3S5(aqSoPsZ%~3GZjLvJA%u+PcO$fv(mm
zU)fpu+t(b2k?qqD84eHJA74`z1-mJsU4?A{sI<X*Pt{cH?t7Ze1-f2i`vc?Haj`n5
zC`*RvNU0UhkF3`Qd_%B~$w>;`GU&*Q_nst9-?8ArRQ&k(ijThjb$;giKf{j?KjM?u
zpK|r=DIyB~#>f99_hqDQRw>oy><K|fj{Ol+)(~UzQ#FQ2K?p8q*rsvhY%N0T<Yk+E
z0j&!-J7;^HyrRc6!d>iKth1*^hx}eF<4ijz5lAEzH;ani&K9j9M3aOy#()fO6>N_t
z&@soJEydsId4$JCN-m?{be2mo2pP%C9Q5dX`r5EC_2LQq5HT72=bH3{!Y$iVL=c5c
z{6qOT_Fe=x2$2Omg7;oL<^Ew$KaA<xoYcykfQ=J+86RUMlRHQ%{!Wv@#mn*JzL9L*
zM7FgdM*N)3(shB+CXJF8Zk5dH@2mhPr$mVuiCChu^ls*#3=jO1yWiwleaZKm5Ba5w
zpQn{8+RYgw6V{GQUB@^LG<AzG0g1unVz#Q!C<@1M=+k~ZO%%DIPM@tO1UMY-kg5PL
z5N^U04O(lWJ)+GPVQ0GjGkkT1h(n49^Z~65LpHacr<tnRAVlQu{+en016Bbek|I+P
zsiZ{bc$kz<D+zBmB2f~8WU`)5)+wXbxx@+dKC*Zi$tX~nu&V_qk&GLY`+9PB9|nt5
z4pV5FrUK&W#wP$s5=1U$G66M;M@1`f@yJ@zBXV^h7N|X@5NH+!R_dGrU`&CMDJyu%
zGh9gSOA#nCm+jFs`6n;iqpAxsLX}_Q1?m?eqKPn1bPxAbtCqDX$??FC=S^}+JY^q~
zP%<eJ57P%hkdkt}<^075+<fs_Qrm@)xVCwWx)_G0dzA>vQfDDeR@g3zv={})e&qf*
zGCF}nqRWD_r_XqG^M=VfmguQP-Wsgqi%&mg7$??eXPBZO2G5Uv^dqXW=EaNm8Rwb1
z`}@qHObG)ixgw8tNyTdN^-2qdF(%OlmZ%|-{Jew^FwKvX`^QK2t<k{0o<}PB#pyX*
zxI5lE`o7PFzszIev=A4@WR-AwMjjp>nCI!V={T1xFvfX0p&VN4Z2RzbqRX;r{Z8)R
za^!uKNI)A!U02L@=6t*6gAZQN^+#@RZ`dFA{Q9r|I^X;L&++p=|MN6$N-$Y`-<EOQ
zyBtay>a(Y~c|wIA)J(n7m`Y-d<@~Z`m`6-g<7fE#haYhJaL?yI{*({jf1hKwCjw<%
z)072no|vXN+gDW}HPMvRt2KH!qDw{9Gz6eKj+m-p94t~xy8CPV(6N2*BdWSWxQVel
zFbhFH&OCeeoCvt!SSf=%PLzd0b_2pIw9t&h$TZEgZ9`o*jQcy@y!@PT>|wK|tk2l(
zZ;-lRf9NQyIvLjb88`Rb9&b=O^04bUzkEhCsrf!U^b|@GLW_?UWg1FdGaT;GMak{M
z16wqvlzjR5$0)Pn>}*4AB)hu@YH2usdd1kam}*5g_fQMYK6uLazyHUHN-*^u>rU{e
z&c2E5BEvB8KY9Ldayag3bc?*G(YoMQum2(8O0eFXqm<$|?|#5d*kz;P1z}3fV^XUo
zv@ZzIQXsTq8XaSBG)<e-s=2@mAux_ZYM_eZD`wBu+T^#H_Zle`E?b!_o`4}nhJIiy
zB&)j4>!d!3S{AfC)i4t0eD;;iQ;qx%Nu}SKYlz%6<+U<5Rx$I>LySb9i@y{7h+GE-
z$X08Q6Q%6jsvZ$Mu^?$#U`h0k)DDrRM5Qr2tDeyJBm3Qv*=Eb044znPQvsQUG9ud)
zEo*Uz5ki{oE!}$bAt_^>OZ!MJK3QBkyu-w_zatk~mdtUV>AEPTiFz5agUah38Ljn1
zZl-(w^XY-VyZe|wz4=-Go5i1FynkT4zhRy|rl_!LVsiJWqM@wT1SHmia}z~TVf&t{
zESSfh5IkB43R&Wu<l^i($8jRq9vPrCE%Q7ujSo1daNQ0YCX_Ig2Ab8D*?P1onViQ>
z6Uo>e-0bMO146!A3HR<kfHCIEJ&%B8uCTN5=qa&6h~yXOBmFq@S)B^NaLQ85kaJ8K
z!U;M}encYk3YC&MTFT=y2g3)Sq|v$SiGfq9#$wue+UP^>v@(QEM*d(4(9ZEjJ`m$#
z`1~je5mHiCb%L97heCY2h$Fv;k80o(<n;DFCmw$z9?O!)mpQ+PNzfo@w4xh(`iDEp
zs-bM!G&;97p)>Q0Qktr&(b_x~kTRiO0g+%psaTy|GV~+IhdatjozTV@9{t&pqkPO<
zz!Y+x-08)MJLqDh8y&ri%z4!R?D-3h(}-7+a=k(W!g+?>0q-qZfN+*qKmIXa{`fPh
zs^rB7FW5FU*LU|kfAJox^%`X~{m|pweIiZ!gmNJAcp#*>GHnvspsFf5pX@L~UJFfG
ze8p348K1<{TKTvT{_A=q1TJ}@wQWODl!UOvG9@&pEGt&)ma3?TA!cqLO!EveQdM>G
zw{(uCX^=pFJy6#*wUmt0$S@3P-1b%@^v^5)-abMAV-oUGl_eLKmo!boCm(-8VGO_g
z%YTj!-v5B#{LSCw!w)}Xm_}~yZYc`$U-@_c3tuNdmMdm2u>Br8If|m8s@LpyJDhh+
z<HUM(0U}Ku&d)FS^wUpxeSOWU9awL-tk=(YxOv0w_L{10kka5LORxiWI>Oj7b$jAC
z@X05?P2H?``s69jTk1kHdPkuQs@))>!_Jn{WREoOcyjfOh@c;n!m(*uW~XS+HpJ-Z
zjss}T<@p(=teK8`x|<uO{>b*~3YAA{gWnPS5ur+&VucSb8JAzbq^%o7`2s1yxoh^j
z9f*LIFb@ZA@9(L$YpS}%4g=F<*>2A;b<3w;e2L|b(ije-<HM^BclWO`5~`|1l?A4N
zv2&EArk^}T-O{#c%qx^6j3ZT3&=i^w1y>hmjJ9XhRMhnfU9~(s?D_Lg{sdQ7=U5lf
zTJlS)?{Is(Cx%E<);!#P!QZyO$*d$6KuGSx0c+CU;GIpthSto}oKlla8Y?gFO<MaO
zo8}l2S8!>1jL|7yceW?ha?FUfgUs7bYt}M@5^CXC!(z#`h$R<6#iH~jk4nj4L*ArV
z&^^LiVFmHgV^c&P_orol&Ia6(#K#jlQH^B&X9yk<1S046E`l&(jw3`M&?IXcXj?;3
zY&jhJG`bWSWN@CK6R$kw6hH3Moc8df5m;<((&tY5R}!2myt8;~Pn45oQXulQh($HM
zOzOmfjGr!VI<hsIkP$`NV<s{3H+KJk-}t9LL>*^-S%01HUVe+R-m*G-%Ilk#OjA!)
zD4M!OYe9e5Q#O~BRn0Vv6h$hiV~9NLk5tte^BgFvEmCQ0Oo=(xM}h~b6+_=4>Mc>4
z<gYoW8f2bEP!+n`aNNJ<aQ}vSyP?^v>H0t5EMRR6B9H|j34IA6Rq39$_t-QxbOFXG
zAxTT|mqoh?{bND5L>8086oIDli{m2B<zREIGsM1R9Hy5?3UTruTKv?O)QuQ@g07ct
zM`W9rrI^d6sf+asc?_kCf~u}DWl0nwtDl!omX1?a;#*L6`h&N==q;oV79{YIe;D$j
zl3w7XgEj&xDVVw=w{Kok)D7OdQ}O3~N|~tZhV^<&yKd79hG5B>41&ZD6q@skXMB18
zz&u-wlBaRF5M&RcCAW1s$MktZl&6<8S@w8r$Rp;uq*_(X!Lw=3DU87&iPKEqk6<lY
zrVK+|D(L6V@#$~<fG8ziO7@36=cb}CCFf^n+}zwe@?Ap6O5kjLBLz)Ur}N9cib~7m
z<3!4F4$eQm==q8VKrE3b|C(ZesqrGegYIMjqqL$hB~?|Y#1=Tk!>!gU3T@bI)(pd#
zqp;xm`kD|To6RPT1s@W6+Ydt;0Xv`hhKa5q@^$~J;q&i(o{%g*rfK4EI8uJM;BY+f
z*Z;HsjOYWadQH=|)OF2vyX8OrPyZ9@rePeX<h>@vJ6r4s0a|z@NzrOuWU|TpxM^yv
z50up!Ydym_;=JYT@(JJj+3#b=fv!6;%`>mQ_zX>C=sU`yLKbOEbocVdct5dTZHQ7)
z*Cki_DP^<5Mv0D!V6z%jYRb|eq{2l@3}A`^?*+0*mRRdD@eGG~qFgsfQ?a`I2;Fvg
zp|D=kR4omWm#<%9=Mj1Ngct?ZTdJbQS%pFp=7b2h>ou$O7H>VPdQIOQxj4Jv{{EV7
zzo$DKaO1$xb(neP{6g^T{0cW(_WL6>3MD0n-93ZW6jj4+_lCpnH6<D)YR;=G>~O$O
z9g;0FSdRAxHtP*_S#oo8jhlPUw=F^>A=%`5j)w!QW(5Jfb;%4n%v|5RrZ6?f{R1yP
zc#mh*mf1Q=VGye2m&W&a@A7MSQ8CVec`ex41M@U-H{LSFkzbEr(lb#CL-c~#O+chD
z8t+rW$S@2DNC_e`*_62GPc|q<OO%#~s8Z6E4@7TKBI2`0=G!zPk&7N8`0yx%6H*XE
z20x@?Ne7W{t&!<@%>vAr;^h_~0}{0-O~}+~^;-m#NuetVBr6=jGK%$xkgRwXTaTDH
zSLt^#+r5O4{tl7G$wmpbn-zsA7)Hx%Q?79IAt7{9<#hqH&bi5O@@rfs3PPk2v^Hr9
z6@ADlUny9jC<?SO$)3f!6bVW~aZyY0>Ju=MMV>+@f5s(pPY8ir_>M6krDiwp@Y3_w
z+^_MqyI1_#?Jw}{VvUrN<Nl5r3!A{<&@m5tyo<EWHt~;Vd-B)S)s}j7&OFaZX;9^c
zakhyks3TLqr~C2?3bW<uv#$}OV6p=;SlsL}Wx@8zH)xv;FF#g%cJn3WS|N3X*2^sL
z_jkN=0a(5FoaxPLOtcgdL=;#d7^Gqp3YY7h2v}f_lEI4{7n6&{5D7W@Dw4z=Qp9X2
zlU2o9;(^KfY}yzR!X=|iGy#E9i&$gnmPCG0;>jgF`W!Z$xefw|G7^K&*}h4=OZG8T
zIXBo;CDItY$mqo~zI_Y*I{|*nucgRCW+*xrA>t9`lYorrWpdFIg2QIhQJ+y%k03~e
zn@8>s2cj~lg_L_TN0!)mpB6xGS#8!xW%8yW(qbMXQAkWvQ*Y0Bc=H99n^i(_MI>eu
z<)j8n#dHb`lDH6%cogZHh2U_xlMMn^bwN>9G=-rklbTBbQ`ZxO!pB9MlK7c~Y_lVh
zHINg~9Xrmh(nTtYf~KknAzS-o4?<$hrruCj^=TxcwZdpc@6%#zox_E%j{g?hSOKR_
z@n5&{Z+RD4s1P!VvW~|NA0lm&nvkL>6QZYJp+)GTM3*Jj2PSLT@Av6BjUkB)oa5nm
zq#Gy3OaxzO5ns*y{TI9HDx)Tw&E{kjyI!q1=Iq}uzkJ2Fzx6G?^X>2Qpa1{<2A_SN
zNZC)GKE+N+otJ*pyN!XZo!Pzq5p7%ZF!yOwbu&WDlubpcD%Ptj=4qm74c>ZWgi>o%
zS)lcnNHPP~TJqw>`#2ZbAC7F+Tcjyyw`-Ith@+$H2cpt!w&zS^D$aLzcgVJ;0LE#=
z$eQ)~jG`zRrz6%+*Z|(z#OsWnqAc-(6dNTJ>&-b$-EiziA`w%pXtz%g$9s?^tFxy>
zKVhBY^2Ik&XBQyqJ!co^=w=N>4*Pq&u*m4Bs)nhPY%bQUstP3nhiRf8CREWN3WuLl
zyyNrd&rnj(9rhGOgPR^u)s}8|<j~*HtQ-7bDU`w5fwnC%Ml%jQAqKQ?I5)Ao?>HWh
zoLyWnTgUnK4Cg%k@kqN7NZ|<4@^Jsav#V#UYE7^MeK$~4^-0P5;=K>>9;~0(uFu)D
zTaNvX<Fw~YNj`u0f?w8O!@0=!zVS_jG5ko~aX0Tc9Cl37^8@z^J13fnDU%AB7@;ni
ztta*qM(bqoocD~{Nbt!qV;=h?hRb$gET+;zoC-Tmg^@qeEta)#*>n~j`$CIB=3HTs
z8@cS*fOxb-Tl_eaC@J{`sw@mlZ)r$AyLtTlxVZtm9lIHbK7vR<L|9H)2&B-ojiD?H
z9`;?Pmn7q7V-|l#kc8yncZyIAIb|!a+ajAvqfjIpt0kpMXOU&}4m*uWy{sgL#B&eM
zr(AfSlUcM*1+GZ@fVT-0!YE`=AU#3{kSfQ`2|kNQzBv3HJ~;jcADIg_H;N~wMLNOd
z^Y>U)Yhvg*9($rNlv<rYhO%m~bHG_oT`8t$C^pX+A8wiZKyVQe9hdKa2NxpFP88J&
z>_AnmXxa;;E_t|rNsREp`|o2TbhpzZ7wq?WymJBg{Aa(zD|;l~-zEepNL-Z6Q81IU
zBuNjy18fus3oeN4-vco$rBK?HBx(GqvNB6&QmAtg@0ZRbsezJRNiv8;N{rH}6N~{B
zBMUV`3X$E`6Nxo(F;e!1cOGX`+|bhXNU2aoK~a_HvH-0JB8ys1g<pJpAC?MGE~DJ#
z{~zW0>FW~wl0+qB5lT`*56;n#BjfCtLk?EYV*(`wrYIOAL|sr;4Mu9px}qqnB#xP8
zoU`<YBT@+3)#_2G0w*s&5UkHHxc%~Ts>!k`RkC=I*?3c?hhB)7uCdP_HOu>nNwJmE
zEz;jtRu!xK^NONKUR5y?%Zj2b8TSXge2j2=L_9gg^6mEro@Fv{*Y&L0hN`YP8M?eh
z?gf&S!`lfhBD>4Sn7KH=bK_sG#XG(?f9Q_|6cZ6lS)#LYO~{mI+89ky7AZDKrN#61
z{x0+WQld@b`^^l~$l-Wkx7*Qm-6;xi;WT~K^cntz%kF%$%Ch0O-ywu#n>YKmZGp%%
z&&<v;&okpR@$denf0ym{jGH&N+}_*)nXB{;hXCfW$J-GvB;8@pd+$A?U9E_splr9e
z5K-Dt%bLS{pxfP{b@Ed*Rm1x10wD{!{XGWH_03B_5NuCTSJ=pjlG0C(W?d(XrC~sq
zn$32D(*oajXerQzVR8Y7M;B`<wL(O}G!6&?ufP0^*+mZh$n&ceo6D!TpgHXBI2@1Y
zs>T;JrrokC3z~~(n7YB+j(Hjw$DYml4DSMsSQ9h<t7*3s)do9u%)^fJ3(b|Rs8{D`
zrMdq6W6swN%9Pll!>regyL+m(rD|8~cRSYYDjA1MO|^XwRg9F=9;p;#*U_v?Lhww^
zB4o+t?2L!|Yf2;74|nukGKseHh<5?6Od2hE%VB?mbrWHp_}YuBl-yBDMDWPy>2?ns
z9IV<@tho@Gk4NsWzrdt+mf$9?F1H-|9s91wNJSwv+p}}tZ$9L=|LJei4>eWW^2fv1
z7~~9%<nIi>0fPVyH*!zgrqAJhvLqu$3a!vNLpWN;<P!Lx#L`%UG6G$s1S=I&bMKsE
z3BgE6U*;GpeR`)B<!%~jD$s@y97|GDnhNEqQ_RNCq*sI}Gh&vCs)VAE_cTRJ-<Zd%
zIfI(T%Ql`0(k1<35u!w95E7*d)@{W+O{X!scQ)DUWU+|SlHhVOkW6uYF-BzMB$AO*
zLGan4O=nYRnMUQ_fpbaKT_o^h?1yZ49g^p)l4zyVh&j!fQKm==<e1Od2ZYO9XQ_~;
zI7KLaGQZ)sy3cU^z%Nwa;4e0RhRb@z_WT9M*FRvrS<!AU+26iC-4{3Y6h>jEo<ex0
z$x@UxemLT+<^1Y7e(W*rCF8>zl!(m35oc$Ft`Y^I?_io)uUq=Vj@_F#oIm}5QNA11
z<(&(_r{DMpx(DulYo<RQ(InRof=|WPX{}G=*AOG~>~U5QO0xKagextOjNmcqOOlfb
zk^<;7c3nE37?|f7a1^mXS4lO6Pu46VBSwiYXCFhUbCKYDvcm{DV^C?GqJB()E(??~
zNUbvRm(DMv<>^Fa%*dY1<6L<vvU1Ny5Q$$wOlY4L$pPFGnS7ud1`hqip`V$o$AyGH
zW(N}_ky1r!B9@4f5v>hZS65s<d6q^5fSpE$zN70p`o2SHMP0XPe85{4D7iQ+n-#m`
zo<^@wN<MDZ3%W^qp-&t~x#Uj5G$)V1<*3REV+w|GN>&1SoDZ<tY}no1Cv)$ZiMf#d
z3iD?ag8p~}@3GFZ-#u`7d4VxHyYDZB1R0r3*~D}3`8>&<h9fcLdmDf6G2_1y$lsnn
zkP}%{?NTP|htiakW%^YV1=BQhfB%5@fpMI;zrW}9<{qWeGgsF&ec$utmtS%`9+~Hv
z{eI8={r!nP@HWTqUxe%TyB&+dQYp#)aNy$Ng4Jqu8cP%5?(UXPKK=yfJs*GkF@No^
z{WboB|KNYgU->KlKL7Ne{WFH)-E0yd30c~8%kAwoPoBMCoM(!1g)tR#5D)_6uw$NP
zoV6*RyDA|@>~usE5_Kdhw9!Zuo3k@!J0LYEEqS=TCAyih@0eUj7AOz*^nJ%lo>R3o
z8qcwVzMm-9sWCKFflw-CIwMhqVeC4pwnfY%y4s+O=Hp-g4Tk-Qk%6=G6;ED#n1m0j
z3TtIHyB8df2ZsH1N)0gvqQXdn(G~ms9wJmlP1|e;-k}v#Wy}7sV{^7)b#X@9uG2Nr
z8X;Ev`2Lz&3JO_J=@MrLns&?n;26B8JiFq(i*IwheMMEbDUW!6M^h9O%?52GMOknd
z9&pZ6Rhnl{FDa@scHKZx82Wys*|dacQEE+HC<+l7hZ#Fh2;bqRBZaC_bqU@wbO&_Z
zAdR3@l6`kiQC1A&fwGc3x!h8fB~86ywOKJwGhKhpvu974hZbFwG{=EM7}&HM+P+{O
zEoWD2UdwA%+m?QH<Ue`!>vU5`xvmIWBAi8pWG)_@PkwnyQ`O}OHdx}oVvH!gq+}#P
zp2&NlR1za)QuxAYZ$hZV-4kiV8hqk)`z&}6EMA;szCNf@e+B)M>QTIyu#Jo?W+7fA
zBD~LjehQ%tLTG}wkKZTNiE~Zx6h?D?wr0QY>4%gaq?Ja9NZU3<ACrZf4=AHg0yRNO
zt&#xF6YQM88zoU8dpAxwW;GYEN(#n)z)coYm=x<80#cgg6J@(IfkdLKI;T{DjL97%
z1V?Z#Ic3C%)P~X&xj=UOV?N{c-tu$$j(=nP8IB+SD(`*xC%N6<W9Q^WNsPE*AkL1v
zH*YXc-eWr4(p$;>?lqf>=Q!(leZQxAa-`p1LvXaKCye8SQ7XmIA(_2I$VgCvYTZ(l
z8gdHF?|W@~r|$-ei!JZ{%y$XXz?<LtZTuV(TpJ=q^r-N5DU><~>r)XJ!eYqGTZLW1
zh$Imul1fz)h*4xtT8<J+B^MC!!Qp)b8K{a9sh4cgG@gn!@#U0K7^4#$Y#m-CQ(`F<
z#uON1PT9f66shZ#Ny+>c5`Xeg5+aR{Q)eVo$rE!zNgDR%mnEdr&J|AxFj~iM7}@nb
z`*Eh9Ll(;l64N%jDaJs+rN7+;RAng33Y(+6mhoD5IB-1dsjHIhcFXRtV;BaEDN)*-
zil^kA7!gX*Y_@!PdxH~DisT0<vT7bU5r>zeOJ<bQI?rSsez~_X1)~cgv)g;Nzd;~{
zp{#2{ggK|3h`8KmGG7Knh=keUrjb{#-=MT+wOTRR`BAVDAC=z3Ma3%87;Ltfd=7||
zuN(Ni$Grba6o8-dSONb_Jic0_!QpUVnn#MFq^{DlHI4(HeDZ1H()pye6GGCqZL%(i
zF=YYgSh$n&T_iQ|Pa=Q6-y{8fX`^YHnzASeAu^5=vvm~K(f3_KGziHASP1^qpZe4M
zum4y78~*OE{0e{jSN@B4ItL#-H#axz4||HHWwY6^Uac9&141c=F5rAh_Gp?FP2F<n
z?uo%udW%+p>zALi-kdWI2ZBppiy;Q2DN)K$RW%}}0{n0s*sM0(-#^f{HFcqh(Lx9u
zk4N0>(fb4YuH)j#GuGP;l8Bux({RKn!;_1r4AY+5+Z$fJxnZ>~_~v)Mi4un6{sFXN
zoJS^WQA#mQLss~MAb=0d<3!9}Zg%u+Hs|0cKKbq6X1!ifS0$n-vE4mie)=Q)^QUYt
zFL?9jbE>MQ(28TX=kn?)``s-9kv6p=c^aOV5+N&uYFTZ!Jlx!13PVsO+7z^Pi!4$G
zYrDFnDq92vk^z)vwY|WsB15<5vR=_{)?8m-^YqzsG?H%j5?zAF66cBYZB4t{;)7zp
zpNPTH4VFVUvfZpHs}0UuqBIC)XxcSTpMS_|eV({;`I(m?@GuS3Rf!auX|!l9*{)j-
z@j#<Bzp(u#^`_zO<&XK=`Y$s#6E|m;&-OP+?fLESIs0*k$WyY>c}kghc|L-jld#Gd
zow#>VoJPRQXwevxX-lbSS~!4Z-Q=%Lad716*Mbh7s7W#<jR*lor3PS0)L4#~Coe+k
z{9So$91}OtPL`rlL{$D+Y2UEZgj5k-l<9l(eq7ZBrZDK8<mPr}nr8}Q2qC2PTZ$7;
zaiLdv?-GC!vl81ngi51q5KPm|G^FRMt}4(5FAY&TkSX3S07`3+D*ayMYvBVfc(h2P
z<Rt72pfp0Mw4cRnG3Go%C+i-y-Y^Uucf&n@^NYXB-}HZr%Nc(0jpomt{|u#!r<9f$
zLlVu&<frIkKuAry-X<QO^0cch(OSGq#^d8SvOU`pY)X}Jc0!d3p%m+jD>fGwOv8ZA
zHfO)z<K1mwnu@cJKEzKG-Em<5#Ve#wF+3(m#6c>8JT@y-HEtRa@vBw-k{cpr8a*nR
zM!;V^mK#sz7)YW>D%3^#P8JD>ocl0aOC*JWi_DQ&;@QfoO1XtPUy58*M45iQ^#Q(m
z7J_^pnS)1C$)1M4B4ju`A)ztk+&>@LAA9chM-Icx-~%3!RZ1c+0FEeJei?G%rNPC3
z$}jpf^^{eU#>dX_u-h?oN4&FaHtU3rS;uT=Dy6@&Aum=O?RLxT1C#YsQedPwp+hnF
ze2pH7!0FNoChHk(veC(FQ4$SAsO#o5q7|fZ_xa@|LL1yPCxR{MzDg-Er6w!{XC-<4
z`R5$RffpZqfG!N<Jg2LhRb+3gcT@2hFh&zX`m6wNX8`}v@w*=(+c2bPxJOgxB0Kdb
z!Ie#@d;T-;9mnJGE7nPGKUcro$oE|jr4;i#VT`7#5}n}T{+@@22O$0YqAYWKA#`2G
zU;pcWoqy{u{x|rmfAz2O@yDO=Z~u4xTkmuZ!FfQk-kh^qZHU%$cYBMQ4}|E@<px!(
zsVYNJ6a+t^6ewLFykQ#l*vaxR&nP9hzuRHTipwXLc_K113?oWO=E?Hj^B2SjPo7+{
z+uyT)ymNaV002I9Nkl<ZxTP|N7@(<2s1?Z`N3-AG5u(K?jkAfT3yPb&TM(Y2tZCXn
zp+V@h5ti#U+L%-<D@ETO5xQV@0WAeZRa2ECfr+9zqg|h4U7###4);eYqY-{)>}ICE
zLz|kqD!IG8<?7jUx<f~t9il2xF}223We9PkYFnCijntZB*U=P)_1QVy&{1r*=%U2C
z30=0SIAuhY4cqf8=4oaadgei*lwq74^{Qc>%ZzdfE-#-lTaRbL3&HCA8QWO0yZJHG
z_!=QboRgSxMFc*0euj6Jahy@*nlE1ch-Q7xVVJ1uin?8+P08Snlv>jrcldciN|?uv
zwq5g$uf1S&6VINU(;p9sGfSX**m2nHz>U24`uluq`<!o^1JmKiFT`)MecE#S@+H4o
ze#+-}uUU;Pui^u$F2JPq6qPJkLh#Ejlf;HfVVwgXPVZibX^-<R?RRNUO=i=PR0N8Q
zDoT}zH9@3eD3X!FJo#Dt^OE<UNwlCm0u~`Di=B%^G_q*G7*6q~2z0STC}z7Fm7-FE
z_ec@AxVk`_l81*KBF0qwsca7u1HvzQfK(7BGU{V$mdxowAq4D{4Zn$$x}?~&sJ20>
zf=TQN!6(kVDY9?kqvxRzOC(^x&Qlr}$3RdLRg|enb`B{PC^Hg+Ll-5j&Ne(T5@W^Q
zd;Y)nw^+?<zRh_;ExqUZ^{1?wOIB6EJPinCSZ~iMeMQw?a)0~V__${1k34z$f@6P+
zpul_2*mp>&k!3+YJg{D0At<n6CgK<d_fBiuyC;DdjciL+AAZ2IV~_KZ-H%_QqeI24
zFd`F0afxFPOTP0D1)u&K7FSCI{ws)(%*s_s%qR~TP+NlK9{rGn$coPsYL%gFvP#HW
zl#&!hldNK5WVSYilQ&J0&y&@+`R8b*PX2|<kA4(=aN_SJNL=K^;<r(%g#M9@bP0yw
zCHrCG?%1&(XS&HU2Z7QRbx~4ilVE2oqwOeL&E?fO`$LC7F?2K2IAf=o$ptoNXJ}J!
zcY90Iw%B>bImbL(4u=EEXkrN1*;3bUqj{N|2k0_+kj>T+%WQa>U%<!BJr(FG6OFx(
zOwMD&TO(ys=jg0O=fsGRc_D(7+R?094r5w4&sGJSwnk}<&>EMGeviY*e!u4%-~JYs
zY~r1BiDL-aP&BLCzB1HK&p_Ar`Tpmp<aaa=e{}rh@t8`GNanwnZ{AV}F3BKofB!%4
z-+#Zya-WN$psFe^FE6>cxIhTO@o;3j-Ee(<!#vN_b)Dk9iV}d&KKqQ{{_WppwOV~8
z3i9`TNJ&-K$s_S_peZVD?{27yfRO=HfvGeP57)>TI3D+yvSPcv!i7jT^t5G-X;%ml
z*q#YK`}_+ouFfd6pr2;!I50%d+1ZweN9ziuCEM*9d|-e706tmvtheX%c4pVzaDIMH
zT^5NKRjTZOU^r_UP@ql4li+cO2j+gHT30;ueR}tlrZgo4hqE*1n+-BV4*MQE--6zv
z^a`yC=5b*5N7}kVi%Wb6G<89DJfe#d7bWNt>jYPqPq};jis&3gS<sCgF3xPv&N0O%
z`I^l=``wOtibzw^4@Y#>P;D;h@7~~JQe}4I5hn!FXr?KdD+^_aF>#xB4|m+}4qRPa
zCb6I?sfv=}aAduy8SR9f4@3cWhHiJn$AV&Yj?_getfHr`3aYaUY#cf6?kRNg5>-m@
z=FKanq2uZ01!I3i<<YxRf})h{9uBmdE#6JEwPqYAin?Z+23~yQJ9PIWyWJh->XLO^
zQ3l78>I_2U;=^zA{_p{>{cFxIwtQ~yaE0Nudf*@5{xf6%H(4kZN}DX)0Ejs2@}8M(
zV0=VM&DallLX}-D<|zdTNSJK`B#<(zi8WH{Bz|1Li}>gZDwF|bvJtdEgtVs;1I~q<
zHkFE1kp+7T6-ug9j7pgRkci0fcWF|yJ=>tPrtdm@2!uJHwMw3S@aWC~nNZ~Oxh>JF
z3lA}+RJjlkMS*Tw5DFWT=rn{#p*2yc<o~w_C*>oROw06l#Oy8T?8K^GC11f1AtZjN
zj3kk$OrzNl(<mG%P^Lgg$*=K}Z?zvJfJ8)O=(+i#=gHIOcvY}HzrfCcL$_zWy`n5D
zqA*A?asB#7=*mzS#XKhvraK;~*DYb5QoLk8Qm-mp421Xx>;o8t#41I3dBuA_n_^Uc
z>o_vKzD5Tg^9u?1XbDlGFsyF`mtVrGZ)AIeZ2T;;0(kM96bpY2v5f9dsME5+3n9=-
zr`c=-V!%4bG>$muD2oCu6+$MX&S{#GnX}dll$R1s7RYu%_wo-|B>D;66DO1-S;;J;
z-wYim2WEhrgd!LR$Khz%4I_K&7y{Ir3!a@{B9RCmkOAg?#QKpa0`oL*f4@&D1kMwp
zq9|AN!vQ5DRo&3k$?n4C?yqedymNG2m!f#I!IWhl?c~pr)nAKaJu(@Ma=LU0@|T%+
z8Qv;vAr*GBOJn>f9(jItn%F(;D9VzU3$;{mru@e9^E0~p2kKICwyvpkvKtuZG|C&N
ziGFZA|L_COHyggZyW!Y%tlAYyDkO_=CqCZ)TzD=y%=4Uimg2Fn`MpNKe{}q-eVlA7
zLb7~VwJW~!o$uhC<A)#rka@N&8PQtnWGggJtX6Bbn=RMZ*BlN<#_`=bfRNsEqcsG}
zJS9}3&;_<X;3|vs5L{$9>=Di(i}cP-*5$F2z*H?#1g6RH?CCT5{(x=_!45pRSTnkr
zsxch*_ssLi`S~SPRTF%qZXCn$fQ+!;?`Y32`P$dMiP0KwM`mj&ROII7h9^(nr_=?r
zpYhX-(vorP>4zRsXk_$gsc5PeWlFkkhcW`A1l{g|IQLBZ0a@2ntH9XbbN%KeN<dju
z2wk#1e?klvH;))o;_O7-wj758-cGdZ5<z5kBT~Ss*&<{ELNmLW-NTL;BHiHuQzQ;z
zS)DPt!0us(@;zzl7LYn{bA62|O3HSFDK5~m;Nku*+2%A2QVPbQr!7iMp_p|<7YzlP
z<MEyd6xBI%w*v*o?t#lEFW5cYq2tW?lc&tK!`lfFJ<-j4_T^`2DLCJpaU?R2BkNW2
zICLSfTD8Oo+l!j2D!ILWiIfrVX1uRx+m`bWzQObNE5`k2C{-~GJxVL)p<@~+qW5(D
zz}3|`?YiaX<RzOYPYK@h-McFcfg7=7SfyCB9}Zvg<KZP^mwZnrJ0(SJPNf*FW$b%S
zdljjv*R@Wxm<Y3X;Qh(7Fy{4BWMP+5$v02PBnXTm1sVi5gSSXh?A>f#Y9w;Xk0gmN
zMKZQS3XPpd#%Yf(3Zzc?&QU6aNR~LOb(`&hb}0?yt<)^3HFlbEktgyVm5{g?(pYs#
zIMJm+DFbDNN8pB$uAA5&EzWz&BGQx*qa@KJ#|W{ghvOoOO9jSkJ%?_f-d><pn#{$N
z`hc6KoK6)I=hl=&k&0p=KuW`%I8wHo)4nu}183(S&<~#Gyd)wSrk>5ZVea=BxdpA6
zhY=Yj!lc0D2w1IIwGCgs{*tCD>5oV1vZkrd80Lw)>(}qJuDyE{ob}jJP;bvzueK-z
zH($Ku%iTS}x>ErdiG*nJ!6g5mD0%+}kvAWVnMb#H7)6#MUCMm3)1QDy`65X+n$fsi
zsL8kDa-4O{^OO`|N~DE52JGwzKA@FOT$7WMHz|%%-fc!ia-s={Q}%5-KXF0;5!rZ}
z0MR2v;{1gWi9#`2*dGTD-OMmYM(-Istj;g^&iB5@)srX8{lN9hR~&bH20zpFJ(HiY
z!L#dAgiMh=qzaSbZiac{cswQ^!D`L<*%>he?(c518gk%pJhItrY1<Xas8i)SE(tqM
z9iI><bSGsu=5by&=}eJ$F&DVte8jqdU-*jo|Hr<kX<MdoN~)|Fao)1{D>kb(sU~el
zjE>+1vkMHfW3--m@@&@?>$+kdN3>GBxxd4h0-aTBsxT*&*}_Xq>Ns&)G)@KITchGX
zI{xq-Z;zF`uH*C1?-|CC|LOnuf5=B4eZ;rE^-b!k;^R+#d|Jd~jJ$a9g7fop-n@B}
z#9jXaeuMv2htUOBFTS2U>c)GVg{M!RP-w?kXo}4RMb%=o!Ob12XqiXPet$qp#iqTa
z8++co{xP+VlvRt|tPnylPJ6m;&*jAh$bv#O6z40P?U5o`Ud`hKDix7-o_O-&Yj|a-
ztBPsp@zyg96QL?l+S481VBM5r(9lf%19m*J+FlT)X6o<hc5hf;e!y^eNOYaDVvdnw
zwZ++i<INpSW%2$9qNZ7IY3r7o*KgQtE5`oFYPH3z&RJbN;bMkqJW>e7VHlCjNK?^X
zJZ0G3VsuFqCAw^Ib5GN(Ip3T!PSZ(nHw_~Q$HV<=&f1E$SyQ(ge2ghMWwLDA23^zy
zr6}5#Lak})n!|pF7BLseN!YhL9%<?pQ*IE#piIC~u+WHh`<|jK5mACFakFF7v>3pc
zbBem*?8$Ss7w>a+F|yme!rk8?k!;T{Xx3}ye$O}!NFiyfgt8Tz_XsW(pjrruwq%~q
zsEs5>M^uI?KyQvHEm2+1WF1e=F3?(16ao<h-~8rB#29(f-2<8sH9xodI-hl4@a6Sq
zT$cy_@xy2A<{tGzvstg0rwKPXd~}EaS|%>JP-*=aLgIr10mpu3y~!Sf-X)ieJe7$=
zgih0(#r6!T)AT6Qr(y`F>}078aS?Q8Ck`9afA9A_AquLpNo8rsoOlw7Td!A))5O^K
zC&8vvI>*ETO6Bo$LjR(7IO|RcQ9>zHgeW9lDQ4$6?vLya1Cvcg$!@SIRYj%rvAPK8
zG6M)iq|rzc&BabL`@_ijFrZ~~0!;|XrYFP*N)yMPn5Sb>E28(Q_*aIhh{+>a2z1$?
z+bx^zQ&gep_V;KhF}me=cSGG=VAd<5_Y8*})%lv)d+NHz+Zh)FZM&i{is9&R(J|VA
zX?AQjm+!Q;y}J>#kW^Jmwcb(y>tW>m?|cjQaLdEzU*Mu=5kfrXf+Q2jx(|H#dBLY&
zOKLwYK`D`h0y*+zfqNFC;>S7sVwn6G8KZJUjdK?79705lGU!Zrcllw^x;R<EES$0>
z=zCE|I`2-tb@>cf{`D(kVL|3)Jmwy|DCh>q-EN?t0#lS6<{qgF&Mz-`{@x2pWpLw2
zh@O4-z{{I2>BsRTUU(dK!{pMa$vU4Paw&PZe?UmX#pM<2^#&ne_i#_wcii9I5k(q_
zJ&sGsXq-PGfZlnQXt%|DG}l$>S8BRg)J!o%taHrHpUk@hk)|kEH4P$04*Lh@?7_Dn
zfX-URVPbW5M)#(}2hU&~vy04L(2t(Yx?+2_MF8_Op^avHw&mrkm%MoLoXu*LOr%X#
z!XdK7f=e%&d~4ntPhf6t68;DU{KIy<JpwMPDiNutiQRtBhaY~3F`Dk!@%fi86Ms%C
zf=vbE<>e(+RpFe&x_5KR2QY=^;_{OFYlpRaWDGQCSJ=5j%VeRGjCu>2)j38<oC{Rz
z75Iqo4rv6_cwpG?aMn|-t}yML<x^W-xW#cRF2!8~rC2CX0u)WKQXGmEFO(t;TC7kY
zNPuER8{7iLi@Q4%r??a-?oPOQ=l&7*%RMuhBs0l}bLO1ApS_>8ek-(NG)?D&vw2I-
zoM~0d%f8pxXNyp%Z=an&Q6z_P#3qR!f@dcA!<NHlMPOF3`pkl$tjcbE&JiBo?Ah{V
zyu(iF)_*wgAGE-9fg0ldL+8{jh}mq26d>tI{@jsD*zXgRL_myN?S5>`!C!txR$if6
zSd+;M`{Z-5>a0gZua4nVC<k7wUa${Q_o@*s$?wJA2hQ}lTM;*3s`~w&ZNG<hZ-SVl
znDCKRmY_wkkThf|+vmI~f>bdNSde5ida3aq&h>IgvRx8hIx86Y<(JsR6L;4|i@?!M
z735{`$TZcc#rqIF_l<V<{rBpsH8Rdb^}mB+hDUV`^iSn3Fe7X;4vW&nBCUEBm)7q4
zooKo$`d+Fplu-<hm(`^>_M|7vDf9=!niyx&`xAzq9F|+?=yP+jxXA2PixpIE-`_JC
zCjv|Gq`6~N*w^+|)ne(wNPluq+W36BVd*3sjUs7U@I}aGoD2+UISi+{WCp@mn{U}!
zXUknXbf_a^04EC>g{5j%;Zr@4<XC;y;8~J>j~}5vSlJ$=Bp8~X<{=uNPx>>-;`l%X
zk;ML5%vhp$y$-LLKI_K&J)3}}eX>dsO!~Y@+qch|g2ANRw2au{oq6jTR8<yE0DpmV
z#CFD-zztX0!`hS4CdjyWpA>H01jDYYBfBor3CF=Y;)I6RuU2xf2mBH0jN|axvJ<G3
zg7C0LbgUPmxU%Hm=V=$y!Q)o#IzbeJl=@SUk8Daja%W^$a;rkfJh$uP2QqpyP5)ZE
z43)eWXw(}<B(l39;%%`W&TH`UV(a?;esGo3H{}RM(u{KAggU1&gKoAH_M3}N)(RrD
zmLH&&r>2H5qRwc!VmY)Ad_jYCI&_}|2(=QiH*+4YlY1`Z%ygCfJX-54rX^4J89##(
zL|d7IeBcCVzOFAS66WMmS>&P6@25~Xjz8IDL#)6%nngYD9-jvUHN^B$!)Ti2q8XAw
ztMza-?s@Px-JVUw?$2``RCP_5j3m0`#Lflp4V)xQUX+M)s1#&xC&UJ}XG^=bwP_U-
zx~{W7H`~Sf%=_&*2Z&IX3~}g<E>gc&O)-(;o45E41SCVsH*s|b`ds(9{iMV8GvC_>
zd*$IIol8x;mo(H;w~8dkNp*54#ZMlS7rcl2Y#shpH=v~*rZ}vUKi0vBd`78lADLtB
z#w1-se$BUI4_=}+b&{agS^LTT6pvjz{o*Mn@$9S%-EB-sp%#Vw=3HBd&dVb{BtN5J
zp_R)Gv_~forHy$FJ1{zh{m>*h`rMMU7lYn;w}#s_nTcs&2j0i}XQCM|;UjiE<@bZ-
z=zi%<U)yD@l$@NF!qdA$+;;?VK6J1BZ1S_;{#bh%8ize8S<TP*CbD2@oY}>7rgFmG
z-5=c6ROm#aqeS#E<wxP~uqtE`d=quuE@?EQRYzb(Q9y6YU|<{V5piW=Ah3HuU%OzP
zI2b+A?l~z+iY?C-Gd1U-SJ0_45b*e=`aAO3JnR)4M^tbTG2ciR3ZdS@qXEvT9(bwt
z1sZ4Nuj=*q1wh3=1r&c6he|T>MTK8+22_-$F>q5J{T|_rC``K76VRPlYA$F(1s5Ml
z1zuuW?f#PnLr>srGQaUJn;0WI-;h@>A)-8ndU~W33L$8HMj^2^8P1CZK{GEjq2klx
zl3T5D?w%kzzWvQx=-kq})jC;u8rq3dJ0`cjJMT-riaJC;IP<15sw}~0?I@JEW15j6
z)}B4Gio<)zuDS!#^Kc<h@UMT8*24bR^65V^lsST5H4o@CR=ZQh)*tNY-*a)m+)YSN
z(lM$eNA=JRB7w}io@}ZP+ShTuC2X|J;{(FF5C@IQfciGyucEkWbr}}s>-X(FlLA{(
z<3QY%n^7UxyREdSk3(aDN7ALWv{_FOeNYs-+q&2)pVGgl2@VgEHwo~#r4D&TgLlM?
z<fCh%_@EnN#<s4|?_O-P!}L`PD_o<65Mz+-SBH|MyT1aa801Y#BBS6LiVj0Nk;yq0
zn2?YBkOQG8a|lkDI%w2D)Ou=YC&YyV)E9x-PrgFQ2_V^V%JSb-F}^|v;LAL~ianJ>
z*2{e|8HlESaWgBQ9V6t=(+C2}BM#&@h%Wc@+#eG*&#rB#Bnp9~L7h|L3cO9fcD6G_
zxhu@T#ncm^UF`5<hpYG&nJPhypAPQIUS6?L6vfo&0aW1|+-kF8FQt#-h_}p(%<ukR
z?DfuvII>o^q-A!AaW4tT<fl8=zR%T`hxj~u4s}Ay4j+BJ7*ne9gt^hKYFT~Uo_XB8
zSXrwDFJg(hSHZse&nD`{UvmBH;K%Rs3eY>b8Y5M@q5D`(u8OJiRAm&0C!GwO<hO&4
zHjO0*<Z^u0awnxo*pAbQgb*R0C#vOIq{c}X>tv^ruqRssQl6^56!8(3R~gvc;7eLf
z&{%kK3vdlgsh5kvF%#uXQOGiLF%}mO6&Vnhm2CC)>AP4GoRRXglWH}~<4RAb#^nuF
zP*19NMQ3Xx=1Sk9OJTXYkq=+;Y=-!G!|9R|6M{D*Qby-DzN$BrnSqB<Fe`1jgt(D=
z=~N$D%oK%p!#kPqECM}BN6zElVyw~zTA61z>FEhMHP5Kk1mGgY?K8MdU2!_<elgqg
zrH(`#z0z2e#Z>t1Y!0L7{gpH!ivZP1Iag1&I4B2hQ2V}}fzF#9EY7gzH-GDMQ`B~D
zc+2)#lfFLgvIckWX<BFHsWJ7%30bt{IjThtW{~^+oZPV+h<+XL?H@~<&zUn5{g+i{
z-8Noy60jDqMBKJGwYMKQN?o9`F3IABLwmBU;4YaS**kKBo6A@<sIJzM<2;rUssPv_
zWiC16&5(J((EnKZf=yq(x-2z}RPG9+s8PPY{dPTg<K#<HaO3!*SzC_tET69M7?r})
z(m#Sr?;Z9c%^@0b_4Beu&Fbc%fF;Z_3;$8i>35MtM?AL<ozi_ayuIjk-42>5HSbi6
z{C;7l=#EcB{|^@P;N711q4`WnE;m2RtlVer0KIQeY5Y;RAnOl0nwX~V0W-fti+K>@
zuCFtrIVd8eGrD){q&Z!>b4!uLd(9Ii?FY~J@sZH^)mlnnToEas?whA*C{j&Hq`zEl
zx;U$O3meALVT|+RH{vsTZ7YSRB-M<@KYnnAt{z^7={(uW2FGipw=lC4jMiz1<m2TQ
z8`8d~M>}7C*!zupgB5;Q+xU8pF1afzy`kZ@MEr|^*_=S<bTei0=be1&S<qvx`7b5R
z68$F}r>=pTf%^_)hEAi;D|_Oa_VB}U!GJELHgOkr!5hCl`?jw$xVA0Xa+evIh8V!~
z$U(k<GEqLjamNYm(iqfQ1?5tXdRnU?8+5jx^0Uh)>P4guPtPIYKh7*gu5e(!NtIU;
zpn@-huf9ocVg^?Skf&vQA{$@mYV1mht2oO`5kBKhdrCJirXe%p7z$8tP6rSvPS-@7
zQkfzS32H=aZaH$#hG*<qCyBOoDy3qvOB(ZlX-nDQ@Ra==4Ww3D!QVNue$efOY~Gtl
zjE*PG(bwcgzi1xZI1+<%bOzeI=nJjM`9h#`7@savE-PhU5y&2yyH&{8C*?UZcuFvj
z61_+^oRfrt_8la&$oZ>Yn(zX(h%_coN6G^eF6`7(dN(zn;mFh5keg=6b>N|)rL(4H
z=0S}AjT2WC6%+6CUl9d|dsBS-l`XsS=>!G6T?Z)mk#hZOa21JW{Pv#G^09}pi`Tac
zm0HLj{~2)2O)G(Tj-*y?_-c_eH-SqgYZH~6u}cx3rI^amldl>v!%)>w6Rvvx?cJ|o
zOjEn2O*apC>gCgf<keuT1kkq-O$)Km{!Jz{xpz(2Ed$i}b`3%>`*)f~U!Sr?kv|Cf
z{h=#vZ-KGY%R7Kqf&284iwHoi_AE}e7^16Fs5ypP0Emt=+@oPvDgf9vsukBaH|CV3
zQz8HTkb6j=lRCT#rz;)|#&sALv-r*~5!D_V?iJ&cMj&gIS0G3<NU4#%!vlu4W;-Ff
z?I{OZdAwm>F_yKG;#37vC9~AtV@MHdh0UV8zgXeA^88;sf30`L$||VD-M~e?z6om$
z+~)I5U~Oh}^@U&&#feK<H1r1Q-9K1LhCo8~%nbHD(bA^V=KJ$WR=0qF-UMDPt+A#N
zo+CBY|IE%!BR#9b<&VAPZD$SL<n3eOI}QdP(k{Nk77Xj9;ttXqEfLa-&GyxeO5yPm
z<#N`6fg?SC@w?gFC`%22`q2PGjmdV9W`v}<c;W$lub#UZ*lfBu6p+oYmiRky@1y9<
z4_^j$H=6|(>-g9-9n<D(yt@SkR<zxRP(f!JH&kOY;Dh}>kiyA^fmw|q=h>Tr!Htqc
zuJg?Snzs%JS)W+m<ltEs^8%-1x}7_gTqcP+i7)MqsTFd^>pfr)oLL__x{6a0W$)0r
zj(zJC#q4Wk%oB&pBEiJQprrk#+~B<#2MKj!_ibtpvN&Dyw1FyqXo%>i9=|~WCwhl<
z0xcg0^c|8)UrinG0Mbin1VwBWuE;=(ekzkSKz&>ht3t$AX5j*LZP!;#j#3NG|2(}2
zFqJL84OYk<9={l`vJLd^vCS?1!{Fi@*_cK0VR2t{#gyHqy$Ldnu<bz>@bi~u@|B?5
z`vv?|u}$g@WcG%%@X7G2nCv}C((l`>0kDf3sFJp@gXZLlh?h)$amrTQLb~*yX<Vd{
zqPArkmdus=Qu|E;uiiBNt5^N)bF8Vc*t8{r$)}Nz2eT;r`-8i{PM2G))WN!LJr!%k
zS0OmSYbKI3kiew5a150K#GU*+p%5CkaTs)kFx8<B2EeC0-ntYzSncz2450R1nNm21
zpTBvEt7*2rmHJR=?;fnVa9q=>kVvYVc;0E*^JNTaeMIGIX|FXKtHcf%qNaU6NO{Dp
zZ%Ii;F(%P=f+<v9?iP^8r=dp1u1V_$r$$TW;-$Rv%EI4H%p=gSP2(-&ERue5ZA?A*
zyeFyE>7?Z;i->08az2*ZZyKeJE)3S1prr$nWfwj}fzr%**2AJboAYHoESrd{_<)r9
z#?LBr_Lq0~F-z)JW|S$Zg)I4Y_fF%=UA|*&qX|Rm%cgCg3x6FH3S)ie;|~9A-J&^I
zz2nL-Jde&7VPX4}sF9^dM8(04>95d09k@Hoh%#TW7W%T2z-P0TqZ}?$?53v{EJYKh
z(4UZn`OfpHRy#FKBcJX>Z+31e(VtqLux<yVcu+_p!G(cqEre*Djzg3$F&;ug&}rXU
z1o`dpeBE!e<E?X3jK2%LOrri!h!QQg#p9}C%?tNF>_60KK_KZ80v)%ozjQ>WC0m91
z&J^uLz-+tgM_|M_uBJmxl%F8<$N8}xIATnas-uC1B7*!c@0AM7Kr0GQ{cw(UE`8yY
ziSta1qPGpMf@cxR@g(8s2AK*!=e)B)VtzM_5S-=sB<EP;?8}wdB#@^p`)uA+PZZ<j
z9rG<+ab8yV8$gl@o=>U3$|<LY-j|yB;tW+dWQ>HJ(rM7kK&rRD3A~&GkG<GuT?+`T
zCuOIqy0S+NXRcb-ZOlFnxL~<Tx-*e6e@om|E40uVoO&PUgvfxNKeWr;!vY_FjB!)a
zHOY+3b8DBS^b$&t=;=oxYm0K;9Uh5L_8x{cw%Am+S}{0BoW(ibvGdGmbcI935*QuQ
z;~@ilS*67Fab<FBiENU6<5z@#FJlVo6}m$jaf$(qT~1();x}e}?`7;4NHFE6o!g}&
zk=zi)Y;o9vP?3SjE9=Cq{Zpxu5nMHImc#N?yr@EDhOW=nd`Lkw!1~WL_g*Oyam!h~
zGw)>RZSraVa~T4(aj}=&Et#v=(KAU_wENv5t!S6QigLhbtw=*rKiL~JOr_mEwX$LF
z+VLi{1y)T3>k3*`SvO^KfBrhZ7hjbi?GrRY$|!6pyjVzhgg#&0@sGp4$Vgg|`k;aE
zg%{JI#rgpRPQw0@<?yh6&YTVT)XI_T0fbgDd`CKby46byso4{kCm<!e@=nIVymXC)
z%i2aa#m$35nCUKk3o7lseETj>0e>C8PH^|ZWYDxKZHP_NmeU;$1<leSjpZRAvfCQ!
ztnfs>A}ES?WpyAb*88D{7eX=bTwb*B1c=8CJn$rGY{!cwWD(R2D1AnZR&!iRTn$@m
z!4Xx(!fI9K<*B@|Ux=uxwZO}cwh@;l|K3OFRh?va+$uAyD{e=$!GwR&DdH7{Qd$w=
zK$+LTuC3gJ$xwh-H@Gh#17P;f^LzAJV!maYv@mfD1QN`>C%Jquci{U+e`In<EX-|d
zZfp(~bYzYKJw_cpNSkKV4OeeeAJVXvjJ<iQOC7e<C5SdR8f9Mx`rlfR>k&oMH3hA%
z-h&7^3Z%qCHFb~g|8cu4FXrNyX6Rq@9k(hG$jiM%c2d>kQR5v6E&;8-wkxA?AWH0X
z;HWoe0AC}djUbI5PBvXwYVX^S=LzqClZ7bOHu@Ws1;O<aNw)b~c)y#<ZGXi=(BT`j
zlT$F|m7Keh;U`bGQQ9jifg~p4W}2y<??oJ;-}VnzDq8$9*=WpGS|<No>hfVEoDSTi
z&Xb;>wSz8jBq-DrTVCR0B8B^PvI|~OSm9|Wzx$wk_Rjm(`nvvC3OaZ^^GybDM&mv`
zkx-)+2C-?inOyzBiKAuweW%K*3bB-(%leFw<88_26_mdUVnu16Kv5Z8NRVLziw`Id
zfjkNQ6+%uZZ~1X`F4~k#41e1VqA|bfq1yBZJ<Hv_2e^3_JOh@OTe$fKMnRuPfdV9)
z{{%k(5+Opq#PijVa3@$i5udaPX;WOSpmF&qKrLDi4XLyJwOk=3*Ofw@{MYPLQY^bN
zSp*$@EVK12bXo6xs*i`UrSD4t3N3psh=s0W(O&QY5(C7p=7hrygk^6ZZUqha6wF)J
zw9{!C*Xr30tc{~xE21jyubIcS-&M_KMIqRLZ~qOCZ0g~trTUeH$I(sFfnJjNY?dcN
z((QSKw?kTiL{Bp!(RmkGkFHE2N`ne+ZmnAb(^7J`*Se&z5WMEkH$>9LN{}(w;EwX-
z3mjHg&raVyS_{8kmxtB5<r~upX`we&Tm)TAGAt3#)!nn%U*<5U^dcM>(9iGLnYgcS
z-vO=MrrLDkf`9=nqo;%reGHjK?%Mi4=$5?5-`|zb-u@YRDxW3&b{|x|>Bf5Ho00i`
zfyR@C?v{l{y)<fSgeYI#4|4fj$04$+JC`6p#W5Q2Aw&;8`Rc#yjJvM59H;jNG>yjj
zTXk~i_6hWaMTbA{JNj?ZB-AQm#F#EJ7I%Lnjxk5aGIMP5SbJ?^*9C3O$W=Kf*nld*
ztaxD?>?b8SXzfQ}n_?su1#Sm!O&MJuXSIy$e?`<ZY-&9?;IJ4C#{qlJ3KiEVO>Mua
zC7y)y8JxrL83@x4xI2ss)<|@{)Q@HE9vBsBBHpeu&k9|Y3T3hnk#xUu<s9V|>TYzL
zKpn{R9YT%~3Qy1PWYT|THL(_@iEC=JS?5u^`aXV1fBC{tt4Jc5AWN<}8}!JqQ2nIw
zP6#j*9FU>iYgyXvc}1;1@ON4w#SB)2J3|V0wlPt=B&#DxooOnQ#^7f6R04$rZy))8
zEnA`DBz{m@dYV$ztK7kad$EPZdKxe{<qpHQ+n#qC0~pvR+V<dYbU&7%j!t63R<<6`
zZe4Vflt=NOx_7<$q(X&br<R-AlFP`f5BS#@l#X7_(GYdmV84*u->2KcPg7gp=ITGu
z#^m(S*KvqwFp^JXx@3L;1p%jekFx`U!!h%5^c}99Zzxt;Z(qFq(0kHe*=$|y$d?X9
zEly^F4Y)mP&YOY)GQ5u-?C+1)!1sSpu;C?{8PBP)+WU%zt?E0&DNRKJLjV1{IJdNK
z1)FQAR}Lyj9V;cJpK3%&rV<4r^ug-f2_2qWDtI3verr#2_dQS?;idr(Nkwu*{WURE
zW)lyjkj3N*m2VBvWo0%xh(m;6Us8vGFd#>I5gpdnhQA>q{^c(bpkpC%yhcXi3)g4L
zt83;MFD)zzLjGD$8@ibD$vC&2V^0-KO<n9a<tC#Rp!~a&5<Pc8Tcbom?cCKaRtDLd
zeoBqn3-UO=Dh{;TQ}?}l_k7PNSF{c7UwcdUNVT(IYYFHMWnBlgZDiCFfOvOLB3^tX
zXL>_G9CE7E;3LC_s2Yj^I)X{~4|^5Xl;C%0$8(D{JeZHzc8NUSvPEX2N$yOn#-qor
z&YlP{X5O}bxrWE8z%g{1(6gfRjYroZNPYlIP;+!evnO*VOo3Yhx^1uE?p;`vo~CaC
zb@S-ma!QCa;ua&6<+kzOBUq-WPW1Qkf)OLCr|~$e$9W3ws`$c`?H;)mWXZ8hR3>X}
zUo-Uw$Ipw7aV|f}i&gE&$JICE@Zg}nTTSRR85rKC=7riH$><aYYYlD=JD**@C?2<Y
zy3Vy06UF|KYdKqyZCI=ksRLgDZfgv<M~6ZNdwTW<&cMp6ZOA@AD38i$o%n)ghyCN@
z<)IU}inHzLTH<^4md^C_l1;TLsXa{?gSX9GV;ZA8ZP)|Wi+%fJ^&8xKJWhEm;t#7&
zo_v32!Vq8WcJ%Z6Fv#b-Bt>g@DMxV+LzO*{^T$e|ygTOp;BeVK=w>MV`tecj5#ZAH
z(ByY80!HWUjuGgt-&7Yi-;KteeJ77g+ye26|4mx;y&46mk6EDpT2!nYnBTfd?G^34
zXe0fR3;Jnvga`~4<VQN<{?;+H*PH&1T%Nc97YIZ5FZatb-=&k}mpi0~IltT#%%dZC
z2$I|f_>Fj0a}0~QA&kpkbWbKeO(32;2}|ttV{48D9uUFT*4G3QZGTuu<4rguS?%04
zCs<PT($E>Mw_saqZF31;9a?O1QnrR5dwhm=P9^(tis%c2mPW)q%qCutSd#90O8dnz
z_Osq`rdQ)%cm6eZc@Ar+lwK|Ev2v*W;Ph|K3*U<F>~}Q%Ta$^J?d#612=)Aq5Xgda
z5mA)pxQEP!pYj?bQKpw-%$-2*n+lmGEn~xT7~0<Yiiti}0^UcGJZ4^D)^s04p=kEA
zXfUnIGL~0_DTa8m54asPzU^sj`Fh_I-IN>=0L@U|2082lhT_xH5pFvtwgZP1_P@;0
zzS`O%$-oPWr>9&Ac3r4l+*&?9KbAAQsCZ{O$)-nsvADhMql&i1t>26*9Np?$!V{&n
z*Tvmw@xmm2*K1?*P6ku9$|)RsBZ})&Vxh7^o*F)_F0*7t78G0IXA!ishRwIP=`{yw
zboS0==5HFh1L|`51kf8*eW~etv`k)x)OalZOgFI$tjC3uu{ZDh(gY|pNiB6#iqmNx
zgfl+7bRq>KCWm=y67#beDWQY@Twd1g@y7*OKbdFHvq?r2iKgblcMZj|KSzZa9TQ}n
zE6rEp&wJDMO1AZEn+2YeICQ4Q`^<@5&MppvXA0U2`Wxp~|B&=3_Xv|xhyF#)AD!Ht
zSo}VCP9MhThdCA|;JNW?n<MaiEtbG?o&V5hEt{VhdH`W(jc)XcOV5UVEA*eMdPk>n
zJ0w7qUYFs&G;tt3vloW^)UBD$The`xV$kJMzV6ZA6hauiC~sCd6;Ftm8c7=}Gz9j*
z%V*#ntAu(U%GFDv7vP+_kS-b?9KT<71zBc4&$268b63q-EKKUq+=Io7gTIXOg&<MK
zo!IlG7_^Tn%v!ej0JH@x_4fDdoo&67zRzKm3OG$X^(+zL$q5C>gT8}MC;elOwt1z`
zW)?kfM;8+6<hf>HT;V<+CcX)=DI2f$_ptZLx=P|Hgfu==oq;l`pe4y_5zlc709)ko
zQFbR!Czy);f?OClpoa%yG4PFkIu7ix?T7X#S<%$R>&h|%Nei7B!4=eum!#anD>ZcG
zqP5}=qZLAk<sXNRUl|h&2>4x|577q_1#<UZ>*jxQHgfnVTfbL2W?i7sd83p;GCPZ=
z?&^G}*{i-^&+LtJ>)#IO1a7UTq)10uwAHU$EmeGEWb(n$i!8KCOgtPN`|;Gm%^O+u
zGJ2%iXQz()F^C}lY_7^m^1Qy*<^|y%GgNI@q^`DuVYha$B!a!|H|{n;@4Bw6R$3%T
zU9y)+S&mV_a{-=ix<f(bkjD2#@*I8fjq4{5e@SU_H2k2=2pl}DslT})e~+3uV{ap@
z8#_BNIAlewvAX)-gj_BA#%A7c%9#6KU}|N`X77DYYg~P>xQ7dCfdU>*mTuajqI_!(
zA|-hn58E>zZysM>MEm^%gSoUy2C7>*C&#yY>*>N04IlSBd#LT=UFaHPBEqU+F4{mr
z4XQi)%y^wVG&%>zb4-@oeGHJ!kKgFS2A9$Q#qqMMHM8B@iYRZKLJ?Ln<5YXaH<;Fl
z6y|ZeSeGi9W-yuAZ^8~0JMY7`jA@WbE}vWW4VB#9J7+qDRW?3@V&7A<jiyYN9u~_9
zvplb?%>$@erSKWz>boIKPP<b9!%j!yr4nQW(C+i`?}CIwQwKX<vs`jNB1b=UqzK4-
zY*?2ZX6kF78Y7lq?((*2IdG`8Z%8&dH(C&dMRZQP_XS<N6dtpd{X+r19Fy%$<hIoo
zW|D3`p`L+0+!I=YK1oueHym|7*LywEq}eHUi;ywTkf5mq|JYyqZdGn}d~m7sJ$d$M
z!mmbTp@6J-LCNENzh$!D$a!!hHG9WwNlL6Z)+m*`Wd6-g@^yn_4Zlp!qSbnkC6LSH
z+x>8c3MowHp?*7BV&z@N=S#z}l&@=B1=hX+nMmXgsYL0*ZCSkM{#AyYH;G9iFn@c_
z!|!%Y>-3rLHl(Y+W<`t?OnS$BfR6lz#)@nXh;+Fnd*wt@Zh+X8kjR0KI=KpTzT-wm
z`u%82;7F$bz1+=4;p5k{YB8cCDHibMoy-QX+ED=cr_xy5J7Me?vwJALxk8u`IiX9-
zJ*xDlHfxS9Oypa<JpxLRbJZX-gyj1NgH$5D{T|_9ry_wdlq=E0S)$yIRt&9{Y{y&0
z-=%KPJDu1@-S5<!Pv|xXbhAoL8rii!j=AX$6|z3!FshiluujyPv|jyi8E3v)giAC-
zujzTXWtkDjx%PpYlVyNiG0!G}#p0#8Ehoi!vi<V%)I!7taN>4xMTou&h&$LguzhI#
z9Hb#>UTuHh<f0EUA_4t+y3PX6c3^cYVk{HNJlox~ce<zV-E`b(m78ECbz)uKcJ&C=
z44bYkkZBuhZR(%{R~w75T!y3^mla6)9_tv^TF7IB2uGvuK}D~3FIgfiM_uY?)?SZr
zmDK)Ca`F3dxRn+QLu_1-yr5SzB#*WXk8rzmdK{cIhZC=~GFvg91sx||-F;+flQ*Ag
zkg<*zCM4QU-Hr=j`iCd^TCC%<3X|bd`_xs#imuvlAx&2&!|tF?GHyv@PBBr&NLMmd
zLwxZOc#A++)HIgLP|-#-Z2m2g;YJkE(UVrxjER%Phlz}xJI<rlD^gaOn{W}$Mbf$3
zraC{tr^$h-l=ei#EKz;S29J%VMOM)xC0T|1t=$?cR~p$Hr^;YspJn9AuzmxYr1#qA
z-^<O@VuyVn)W5<xY9H6uj&$9oL^`d9ANyvN>ZE2T&W4gOIs8JE-=PAk1Hy$7^y>39
zX!lg1ftVM`XDJUfmQ|5ULu@1CYrRJpe^Tj5ed6Oi(>CbUJ=FFhPq$53r|c~P+l|SH
zCM=%bg)&Jl>3+eK#Q8P%W8UJ_Lcq+@*5QqCrtb@CtU#!<uWLb~_uWC=&7oJ@kgwcp
zzhe{({kp3up(&~535h^n<^f$7)b}s36^y_8J>IPpmN}v}wNQx`&5yE{0XG;sLo#H`
z*A%}cO||8NN4|1~4)uu_NLsh#mKiEN8}*&JA|r;y<nV>_{7w{nsN`foqsPOuw_P9Z
zsr^PPTL(54!(yr&!;N;K6=m--RFKoJA~{>%^DX$kJF|(G!i%)adPbz7tzF9uz(ZQY
zgw)e6q{2Xgwg_wo1fgYjh8<y_RAz&C>oAzNHYi(;GEBy{p`dj3{6SVi(s{JP)$!Qb
zHNWrf=^oSF1C^R#5OfYm6~nv%(k-jSnDW#@%Uv$QUT|fcnacPm1p1HISp(7{4@%0v
zK!@F|7)a?1C(6bo-I?<rOBTBl`xjY@dxNSzii}n_3hOC%WSqKXk*h%B4wS*VhCadL
zPTutcvk}>Q6}CZ8ZV>28iII$VCODQVFV(WY+F~1!IZrHxu3AWZWS!;@RIU3Qt&Kh&
zZIpyb4cCa&c@U1YkMD~(6wK9Td^~dVKcjZC@{c2{Jo0)>A948BE(u1TxreS;4F1d&
z2NZPRDWye+eK6ltvxfl?C1JK>>DVq-H#D5p5cypz=wG*u5-=>Cu@7;uku;^c#oqI!
z^3YGH$dK_9j3s6Ejk#lF>G|@r3~gcY*}F(RT5<B(zQ#PM#k``fn9<Sy2Yh*vQhv7q
zL5S?v&wF*hHxkEfrF(tQr+|7Tq7sQ*_h#zYbaT0!8hn;VZ_N`>5n0;x*g->R@wT~~
zz|xKhJ*REnXXo?r!ymtp<L!P@)}9UV`I6{bX2axL5#wm4Q%7PK<9!zba5r5H{850U
zSIlCJ^;_Z&aA8T=BJ2ta3_RnRMsD_&-&FImk}Iy>SZ+PG)@MJ<{>a;F4XcjHX&8K>
zU1rZn=67ag*R*sxvn%pw^+4hp(@P8b%qbDxYAo|E@Bt1aafdxYB6+3|tt(WM#UoRa
z+O)W>rF4Z6p1!>oQ33^0K|?<a-}TP-J2tV0OolJ4ODy6GkH(?AH@)Y?F2EB&OG@vJ
zvqI72p$XQ22Q$<`V5`l}%U(q=9iAzNgaZ#PZa0C*$h}keeKHy6hM#mHuuXh>`o~zL
zV*P^!UdEoA^I&GhcE2$Yl(ylJr`dPF6?NX=<@4dRhT;R&C{F2*BP{vmB}z@HX0Fro
z2t(_a2hZ{*XC^5WW3Ld6Qk!CN==3*>yV020Hkw0)@s;l;-#14FZ)!%QTqWL(_|;zc
zNc`)}xW!HzbGD8V1B~QR7bDwxljQm^l~b`z5nGsf-5~s#a$-qr51-UoRuV6heG5}6
zb&G4CrB>Ay|Bt>p8l_thQOrH`yaJW$UO4+|VR|w=l4pItk0_-FENQHr(rcA+-q<(H
z+l%_#T^?%6uDn5=fYUU+{@k*Pprp#j;zTyP22T=AqHmmrmx?9#s<*oqJ)Yb|hqOH&
zlkiy_Ufhj!F6|UfJ{`Jhy^=vz*)=U!I2(qzZB{H`MiiV<lR~Hf`MPIUpuX$f-FQ~i
zG0fC-(Vh<9D&uTEHF@~U$Ze!y;A30s`N~;lb-8HlYV^_lAZxG7&*b#k<Ji6Qwm9F1
z>)B5wv27=R+qUuB(gWq-$0xFZ57h$K8<|_>Z60K()xDcNO~UdrOa!JJCTfA!SwO>N
z_h4_8$kR}6`(_&Kv|vbI!1=Q+YvHJ(-gNWr40`dQk51X^_pHu;F4J`wMmr`ufU6jq
z&C1-1|7RC|a+JD+8oC|CJia8kaSYT<`+O-OL<fuUVG<soG?j;kbvdtY@mL-pwor(z
zWN&WJ>xx^T+fDx4#YMd0Ch&E`%42rjRohKVU~=2h3unfWm{;h~l-|c9&p>^OP4C;I
zWD!*4AyvJ=hZ}x7S(Ai2{%Bt+i~FW~{@RCthvfha7`D5`-^`I@*M(J3$njoFZu!m`
z>oXRa&)uI}>~?qy0Kb7B&~7ODFpBU}i4VI4ol;$Yr(?*mx`4}?MDqui|7Ff@p}yjS
z{7>OatxZj_<)`%+s5{mN^s3P4rwOvrW*5KUA7lYIhwLyKn~qYiT_8aZtyJ|B)@IHg
zQ%7YdWf&H2OrOKtHc>54{MI0DIeh%39_T)qm(OClY6I993odY6$ovKE+8pbW{L3@T
zL+ekiA1$sd`%C)j+zkA?^wjgzfg#f4d-l;z)8eyb<5&Uy!A3g<?6Yy*Sr2o7M#KAI
zpG?O7zHfCqW<!q9%q3L3aEwo$eoN=vmvqjD5h|yKoO$!j2-2BVe_Vh>s1}ce7D6?o
zIQN`81!UHNh;i=r$>SG@cZ08aOAK^0G-EN|#Vk!f*)3r24j1@fu&0T*jkmMP5ZO8x
zJHho2{6Z~4?VWkYo|bs5uSx9V^xn)sL$_TRJvft37PEe|VhGcfzyha$)m-Rfd`o|!
z`3}}mlQRm3DiqeTCOf7Pf{X<K;S2l>v2LHy9<-(jc#a;~qIc9TlQZM(AfxLe!$9J9
zGDcwI7p`eV?$RZ*Qeb$TC4-hg=yNR#2JGeZ$@7wCbvu%M7iZVjEGLu594iDbk@ZeY
z!#IY;r1Ozwr}r#%10U2ywX}nlW^|<&7p(nr=SsqhHSb3?Ip?G+TlXvd@|3V3AC|Ly
zQ?<w{9=Pw88=GZH=PR`^&tIe9j&)+UVKM)YiuwlbUE-mFJ9iP(*e{gfFr=f@q0kx5
zx2i#Vx1sp5h2L6np4B6QCOUZ)2&w2qTQun=L14GVKI8K)^5xIq1biS9DIIa^N6duS
z*A!B9h{y9|c)>H(NenIiPRX-@xijtIZI((z!F$Pb>9j2t%Vr+Dp~GWBysqr8Zp&-?
z^bZ+yUwdvnob#CNBR`XH-J+|bf4XTCR0)=9;|FBKc4LJw#>m@@#*^#nG&j`xuiYWh
ze*yv9_^f{I=`O7|Nr!q$o2-*}M{Q$_F3l(Tt>9TnikrgK64RAcZmf6kxLDa+v=k3|
z_0LRcvrbhlV#&F0qc8T%XIx1Bzh8d56rGjuj=<pKKp&4l|C|2*{gem5hzt)8e+I|O
pLcjHQ^8NqU|GyglS2x4(BlcFS5(B5PFaQI6s4Bl!DpPnD{68(HqhJ64

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_e2e_002.png b/0.16/_images/sphx_glr_plot_transforms_e2e_002.png
new file mode 100644
index 0000000000000000000000000000000000000000..beacafd21c257532a8620ef2c0d35819c99e31dd
GIT binary patch
literal 205620
zcmb??bx<79v+fcI9tiFdSONrx;O_1Om!KQmEy!ZQVUfi>Sa5fDm*8%Ty9Bq#eYf8I
zz4!mC>Zz%kI;W?*rmOpW-QR?%C`o@nB}N4R03T#!B-8+acYXi>+y*k@+Y`>eE8A~3
zUKdGi7j=6J7k6VPbAY0;i-V25i;blTrJK2vv!%Tq2QxP_2P5SV7Z(R-J{A_+|5?Cn
z?*w8Q7ZKloQ-b0kqvH$!U>g6|1@}j|&=LR#0LV&+YIvj{KkJn;$<}9v_)t@^@9A5g
z8O(lL9izU~|06D!tuU1*p-yERjzB+BDY^heAWTHK8)HQ$VcnqZAY!Gw>#C@yIF^>b
zL_=BdWwi-}Duh#c(q*DeY<}`w5FofY@N#!uy;-v^<K>|4Kkp%oO>r)-W0^H*SuhaH
zZMo649&eSfkdk_WQJ6l7kB&Ch%D4PSvXT>DB;7Zrm{0yALd$M&eo1bY|0BZ2C+`3(
z|JMrN?xptt!=-$K|0B;qb!dRr|EQIJM6o6Sc>G62(hbfWGvNF`FJV5REj0ja{6~es
zlCpLSp!5GRuJZqjAv@DT*~OP-_C67`%9)f@SvQ;i?Te5q2S=u$SO3|gPX`3*TWhm6
z;_YdfbOT%Vh3+NJTW0YAp5_|b`zBtFCw!Me+WxnDk2y>89WSoFyCC0PBNuL^CG++-
zzaKB`g~Odkj%t<3079OZ2!5|SugBm~JMOoA9sj*g;>08UPWZ*>W!l{CLDi~)!{~5X
zIOL>51e+!&l*ojmj@;wsAJO{Dsr&uj`jZg%1OvOp@M$>vF_u4&72o44P>%g@lW0Qt
z!4HxtB%T%<EST5JgMQo}3w*EkS=NNqO~Cijp#A(8OI1rV)lQ_0m|AzO$rPOlE*nsE
zy0;3CI0cI8NtaAdn4MrIg>BB=Ih#x`Zc*O<mBEUVDw(Jl&*k@lL;>4&u&%k(ALJS$
z#Sfghv*Ah{b!%`5>DXGZ&TBi$u!|>$fs<7@4oWm}IVeJp$^zzmWDn=#W~%W@EKxFv
z!Yn1;jTV2VK(j>$1(AFXG?MraQSU_2m$B>=Kjh#TiGBU@?|+Ev>J}zg-{XiPI0e}F
z(Fgx^=~uwIItz|twe@~_3Wj6}nWx%esq1n8<bpq;6h|!)J7*M?{naS?W{HOp5>_nD
zYCf8BB<)2myre4O;Va;foT+vulzqU20xuOqrbEt#4jD4v{z;3Td%h_xEp&H>*b}4Q
ziC`@@Ej1KY;<)`_CB10Ai*g$MVazLd;Edg{qwn+rr~DSO@PS&^`$sZ8n~V%13UyeQ
z*DG#88u$If^W=%Nr=4fUO%Z(OY5$$`YX&&$NE*TlC#foej@~l^2qf&K3h2l|BwjDE
zk08>-whAsbv3PN^U8UI_sJdINf95?Q#ooh-eH;j#cpR8A+I~?c^;(plaw?rWls}VS
z3epQlj;r5u`?A<#?3)WvFg(OxBJWX1GW9UxUF6<rWUg5le~sX^tf>H>;W$2F2wB=@
zMikvJGtUQwuBSH|J@a1M)^c*KQ}V_>oC*6qTKfo9Gm<~w4_RQJD#Cs~``VsLDo8lq
ze{XM0rY|7wxV@NXnXb5QT_34qV(6r8DYvaiinTck)#Rw9FMct~SI!&>>^whvN)=UL
zy>LgX+sV9b7o3so`WqM;1B{ImUa0k^V7lkQ&g$^7$p&)u-1#4AmtWYVdCh}w>^iK#
z+5<YT+oz`(cgkR(L1~+E%j%sNKsy907*3Y5z|0>iWftJ%-dXxxlkTy%aydTOUxPgk
zH6SjYN>)Zq6wk>1Eusq}p<?vn0(<CcgEZ<l{5M+j`hsgU%V*~?7sr4}smkE&Q0-QN
zuqyNzBw!LD$6aF$szoS0?E7&!8cd~;ift^4kHP8%!xV49e=JndERAOzr}#tPpom;p
zL2$}<qVPm*CEU8;SXEUZ2``#r6i|$Lv1=uUPTW}L7Zj(bGIu++3)@mNrgzqEfE5E;
zO`Q-Ko^4W#fAyJCaxjiA@mW$XEZx|LHfoNzsjiW$gC;FyX%CB|Qp%EOc2|VXfAi|+
zmz3x0%p9{+E~~QIg0J63iFpZbU?M4K6PQT7&>X;T+(>=0$lW-ie!=-ihVonL78vK5
zkB3?JM>aX7P+Vw$m9d&}?Pw!~=aJ)<RmWaASM-3Ck`VbI3Qm_s3f}2wa~A&oPmZqN
zk_jtUqXkYEAzJ7#ydIu7oT~OeklcblnsWI4>6VXVgHQAwj`ju5bmL4F(?YI*g=N%g
z@5LDNfPp=%o-y@hzfMICl+K0u;`X!ZAAJN>R3tChi|}Z#Q-xzMHl01y#kbD(UDNeO
zpn=nQGph<xfYZ%#&tr0y5^SPY=lrX$vUL`zjD0jC3DDUsE6->_r{d}8nx`-gfv$cG
zpZ*~!PI>=b;+A&$TFJ|_eU<G46~n>{5O59$sMKE;A6U?DfnefS(%SWuuiKU{niS$V
zkR~^+GvkoTWSo-1o{^CJeb)2c)2a^oAtadC{i3>t1q)ID`4LsMbfc~C=g$idrG}{Q
zCSKOwJzhs%U{poYbPbFSEwFd|K54R9be+!P33OFy@@$ykzn^?rcC5d79B_6zRje}}
zOA%Hfjv4bG{Hpa9s-stt+W~nJa+b!g4`$1{xwCo&%X_34NE&7KBI7Aji)})n2o<@U
zNO)c}UfEXPW*gHle-PWnMX-UwQgSa@$@c;UY={qDvwl?2EkCt9wt3cO1sVEy^-|Jr
zT^38CmIZ8?vt2gUld0>}>N0k%tBN=}#8z^s!J`^~EmW|zquF5DHJ0K(<kvrRDg4FX
zn_0j|%CN(Ic&sPpx#K*vYIKKHad(~Bq6PYRI*)xch_d_kAyFOvAE=*QjfT2z>SHC*
ztW*h}{JX>y>s%(p-|$8<c^V&A?l@>p9XB7WHp&i|PBS|ORvaeNpA5NdCoSSo23q>{
zrJTO#pCM(gD}9fNi&5{63lve6#M;%RH#Z|h?DrnCd0~7%NR_O)BIUZ#G)hw}Gd$Av
z)1$wmR=K2&4M6yH^zu7}AD<1y*)SlT%)(!X@w@HF=|VR-_V*nc^o<*yip67e7aM|j
zf-3A=CD8K8T$p<X4?=xpsx2+l1M30wAxlM4MC;?>)ziULt8MX5x5ed0@B?E#`VWxc
z^SP57vtI0dgk6utrjWt54+U&?<Ng;DKo~(}pMs_HruO<Mc1O4S`+`8$4!&Iq4Q!1^
z#M>!p%++<*v8nxKvw8#*O#%VSc`}LXd09aEgOqwRIDZ)dKkLBlD&%o`ag3t0rTR%*
z<Oao~MOX#j;Q_wh>#<{I|Ng>y)pdc$8cQ@)HO+}fi4?<KVLG;29*Bi0g(<Jr>u&%)
z*CkP0)nmg2H3S<Eu%eGVLe&bO-K>$3ut6168s5$lMJ&gUKjP9gG;^`CnI6IU9!3hY
zdyH`O^yD2JlCq|D3SsQXa%d`0*>^N3M5fh2G@@mFPiF(Ngq|tD53*viFNvJCltP8O
zYOK!Gbos2=5`$_n#~eSbqkzPxv&n|D`16OnqJW}h!A&;R2l?qJ4vs5pRF>}M4raqd
zUja56ol>}qjG|a4r$jDvORGK^GtSg6wM5^NQ=NE5zgj2r@z`2bl?LV5_O^}u@DqVs
zYxj+S5ws76P^_778JGQ@!*{DXz{*@T-i3Xl%>|;ch*N>|?3BmN*$`xi;v@4EjL+E;
zIEp6P)=&_{<0kYht;r0RNbKz~0BhFW^!cC|5sj@KxKzE+wApC`t-aA6x6Q-y<qzjA
z4R#rf-4#DFRBN5BK=6?d1*M&oZQt1pFuX3W5jWn49u53GX<P>gta>xbwX)hJ$P4?Y
z8yYvv%qvr5mvGJ3xqgaBhpXA%oBf`{mw}(d(uM3I;^D?JV|XvN4zFFYh`-_9&yy*(
zIsVVKya<wZ_g3NFGiBVKJ;Aq4$ha2RdPRUeCf%LxHk!Sw<bfby{e~dGkUJgU?~7xG
zWe(GL$cW`SY(y*S=rKb-dK=8lnAyi;RQ3yJ`2-2zk9h+A+3}3=)X3pSg^eB&*PG&=
z4-Xx76q;I+sCu@FWIT?nq`-kl_gK)){wX+4V?9Od@%>IGMlY2Z6fEC>K0Ir>^Y`11
z)%?N<HV`Q~*bn8KpznlC`#!;-EFt@LDaN=$(s{BDj+l5^6$z4@rjwNnqP&<=xD~u*
zo{br9?UaoPowVB{QU3ev>qoV{eitpD4_MUsL9#uipLRdJm9ZVZ9JAuqp4roIwlTA!
z$R;WxNSOj2qdGQaSk5tFIJ+6m4yk(S2Atry8s{kNbt|*dSal5DGT1~$P3`i2C501z
zQk?agekI2$3U`Cf{1d2S&66uQmn}hRH~|W(Rv<kQNh@BjLGV9Ko`ro_Ju$*!bNYfF
z2#D(W%CS8hvl(%Qj((3dkwsz27l@ZSc4{)fnv!xzmIsHmDSw7LgFRgeOc`o=7xw8a
zW^2QPFv?M@$_1FvZ6{o_Og0qAn_If8kj*&|AuAI|!pckJT;3{Bc{t^}aI*{e-k?p9
z>NSgxxZjo7d*I-Qw2)4&SKaSyOK?TNZIosRmOagu;L83@VOlq6x|~Gj5`i+hLba9|
zg12GI)N{Q80=PNmu#knLf#L4-L?#$mJgf9YWrkD?ZcorknM8bH1!xvqQA@U%eSSj+
zoGgqh@A^*zU(3=eA!G}FL*L8t;EVbss}@79=iEoW#UTx5aeI!XUGh;jLN4lzFp203
zPp7BPwzoxYqw-zTJ{0(#B$`x?n9%h$Rsk>~C|-;|-svhDw!R8V=?HZ>f+h5nh~zP<
zM~RJps?G9@Cs<xeYB-^qe(uZy_NU)XgP+o#Dqd*8NYkMdQN8HO`wPi3v<*W8&W8HK
zg}0Ruh8E$@L{r=K277^DuhZ)xrhyG^!{9qwoQ|`hFxO^s?YY~MfzYxm6K?A{$A9Nu
z&T`0K17GLuQNdRw@5jP~@6@GY`?Fq?er0(cwc_|ZjT}6+JZU!w^44V7)c(Zdk~R?H
zHDx$m4>%cVA}$sL=GR0Yow20wj}OI;5z~#}S|&G-c+&`D>nX3YbN;KMt_Jp&&d1Jq
z7Ib)ZeQYcdX4*5b@42~qUL)yvVd%)}xOshPF!VM^x)E>SMg$~@5=9QnC6x$UuA(m?
ztynwMpWAefilcuQ-s45qU(Y%DeI8D;8TpM5iCg|%P+3I%;k~Gm-r8p$7szPa^cpx;
zS>(s@kzG8wusayB3F*ZU-ftyK<O#rl><uUEd=oi-_KK3M==8F))eSaC++C{?Ofy>~
zMt9;XvU$8L)}tbu3~MP?>U_De!PWr_ehPK|*s8RR2*N$1wZ5fLMcQQWst5dzYVk7W
z4m>!HKf_*Wa=}nKUB{e~CI=*7))DwM&fBrP^C!`&`DOgibONv|VTNC-l8vg1Gu@TE
zE5%GCE2gwapn0V*Oy$sKKkB>%eg)sb1Rn3`2Ld_xiz1J2yxdyI#K$q1`~1)?zrWtu
zX0|>vKR$QH1Oa1!UQ}`Zdq7%7a|x$jBQ&BIK(Y8c+kO_`y`(#SiYMZL^ZgU|`!#NG
zs_^S680}!5hLVRvL(a(!J%4Wat1>CXU15N_j9fK?6xRwwxQw{X!*7S^>}uiAjQ6tm
zG;v?oa>h2F!hU?uQn9}I6Upd#;1yL^_ysy}mmiR~Rhom$3wkrooG>PxI<|na5v@7w
zFM93$y-8|k{O!|XnX}YC$FPf%(67Pi^z?pSuPkdLnTm{I+Tcjd1kcQ-3=Gm}q}@gq
z*W!rS_EBZ*kK@F3<1?D7JshFmuqc!IlOv5aB)(|pYDzNY9v6M7wOAoLo8*gUkL-zV
zpy(L8wu7Wx9=n;OGBlwdyUnBXD-&Xe(<0>ir@Kc!G6WlcA3P<W8`R~xS{PwBQx+Od
z5xyRHLDbTs=(un{qHSyA1q`N&F2SvDRm_xeHY|epm2VwlLl{TziuSUBj1XG^LcNV^
zza2LWZ6*F%-x`3xwAXUSKf)0>vy%zT{nM`w1V%_P{;#LB!o83WcW-E(^m{+lGrEHS
ze-bkhK>?fm5$E<d2d^WlC%{b5BcsB#hTM*&Z-K4;q4IFj13s1lVSa>G@B7@DkwSOk
z<xm|X(Y$dakk?#BsqEf)`6a~C4O_lezoz1j@n#O~X7y%FG~R~qXx%;O%o@IlAk*?{
z8Brjs@#lnyen;NC9p-UIrq0}_2UAP{2l{&~l!4r-WL8B|%n!D#81v{o%xj)q#Dw8;
z;@_HD<evh;9(~eh%A?MrgWJ`n?$4ppka9sH@vkPoI4s@A_>CUgS_OSlR?Ahsn$-2K
zVps(fpvO@=^S}>@LuWr;kB+nToNPuL-93Q&Lx4bYJKD|D8={WAyCC=eEJU5F;Tv(P
zr0sHr2w>8E=jKgf&t{@!vuS1~$Ao+KhuRFnEm`cuIh%W!=zY}s{g?H-q|CVTii;s*
zO0pXPPyM?(9r{a-(=UqfRPQAQsi4MY4eM+e)J3>X%%SKp-pzk@nddZr6`#hPQPxhN
zwAq(aZs@cu5l6|@_m?A!55ZkNdEWXYx^wcf;y+!%1RR~+kE*h+Vq4Pm3VPsdnMN}u
zW~H#HX~k7`@1vK~;OjW8%mP4I?&}3%&y#fG1-PPS1OSI^9<&fk3%=;lD=d}ZYS<7=
zQ27SJ8gW)AH}2JcM4K6AC?)p>zQ6x6ZZvUs9rNX`CP4)%<E9kTR)+%n9VrPQpPuJ^
zU$h#x-T~HKyaEB0@QoPkCKl@f9R#f4mzGjL$E#c3bbi6^ZsjcUGn_^dqlJJBb+3)c
z>S5E?KTO&Rjvtec@j`c>{xM7^eDx^EL_P<MI704^HavLeP+ts>g|z=rHGGZ-#dyAR
znD}DAeN^SbqTgq9xhlSxhUr!a&lx7^>lGP8Fy+9Vw}U<IM}ZmGfkQ-wAMjzDv1Mhd
z8uHQZH?X}2QO1t;f(QL}EaXbwyH8M?@K^LrJ>|>4l@VHo4^^UaZm7-U`Ueb^4`v&d
z474uodzyzkpFZ9-pKP~*e}H5Ae%-(?=?becB?TP28Qbuba17iDl77~Rs$Py&>z64%
zqn-RQs#!EGVLd!{MRJOPlN{sR0g<!i$YzE!p`@?r_}LrZu(mY>lZLYW)fF!pn4!5~
zyowy@N6yUq-j!H_Wz&&ON1;67nYvEfcN4NqA`xk{_WTaK837K9yET)|wAB~bWeTgn
zY!dl9%8D_=^}XQF_~^91rYj1qF0d+Z!je@n%-;J}u>D@WL$W!*u|a(7&n%7v<A<-K
zdv(#BK&Ji-3hhcDMmG(U&L~sl!CcUi%Pi9NaYoqvEJsuZoW=)wBEYo*%y=-5v_Gjl
zPMzLgJG`lPj~fg0Go<TmKtoE6O@?W$ELgF*cUm4hPZW)p%sM77oBqC@k3&aNg6aZ<
z$?o8Mo<}MhriNOK7r}GB$IVR&;V>o2xQ;^qXAz?)gR4d?QVo)n{t;OO%5!h7mFDpO
zo63A^1V&aK%5U@a%6;pC{OkrU!23>RDq)t|Y~DLQbbxH$1DdR*uYD#KD}|^_jdWTU
z<VypZ(zYZOD**({N*T<R%_zyug`%_uAxz~>*WxR}dtq+tH!f)<n$C(Pz6NpVdU3}T
zcD`p6yRJ&?Y25J(*l5@?Kkl8wuFf;-Jh-08^T*oueQuBDy!P}~Ixik003o263`kZC
zsSdN5f8+R6U&9_-3}(hA!pCa=cDOtO)&=2_+h!+B+jRD#za1Fq6lT)go%?3q8Ll>0
zi}$MI@o}}V-ks+I(~b7~T0<$uVm+e=rzaZrB=Y5&_g?a(yKT`dl|Oxy>czT4$~bgh
z1&Ig>*-|i98~XGr;hKfZZMo5*s~G#rN`GOod}D85n_@4H^ohQ)H2ANs<jkL|pdu#$
zg|D2|xK2d$`5U}6;a@ZY<8XKh+_b#N1>3iV;sx_@3zm`#!h<r?nU_>Nac;k8uqAnP
z4Tue=oL$*|7L38(j<YOtba(D)a5I9X%^PiE`fjdw_QJ)q09O0{>h^KCllei}vlWNT
z&FH9YQ&;oDIozf~fMRKz)<#}g6{j8D=fHR;8R`@&l1cXgEIikFI}kTq;kMXPUZC((
zTuaG(<m47Ch)|}K4Ms+HQ)(C$`8{g;qwNRlmsYiS8U+hu97P#AEBXSHEb5B>*c@PG
ze-(P|C*wK6uN>d-Cca*Pyjs(WoW-pp+#7Pb4yjOeK4QzVn-fx=Y<h^;x}_u?r)<{!
zMD*YgFA2d$H_zL~UV;eKJCVy+b`IL~{2q(cBq_Snj)xjMA*I}KDtc8f<hLE%5Lv)g
zlqui(jMdi2KH`w=drNxn6JB<{GsU6?vLI%rDSGyO9{HPOv+exxPa);$w-bG(JeFnQ
zAN9xZpwO#-9Sz5SEhzwbGG;ad+qiO1R?UovBN7C!NAb6wCcT37XOc&+5iM=x*7sV8
zo0V$}ujTMfS6=tUSIACF^U9`5a$9y!?@8N#bGtnpQF)A|rdMNoFzmP<#?>QT>^`%d
z2i9MYx9_V&=RV$>dA=f0741KTURJVsU~bq}g9&ymIO&>2Z(u9IaYP}<54J&x25Bjn
z(irYY`^Lz~^qofR0>OAEvg^oV^zd;j8NBEzRFwEmr1YMpdPEBb2F|W+dl%Puyabo}
zLq)$i|J|fcQH@Vrbx~>bG9t{AeM!M>&+j;at>I2fF^9dKSz20lv@)N4d>_id2?LKT
zKhaH3{fs%6q?lyveh^XRx=QpjSP41_&?R@Mn}2lk6mj0ny%pYVum>mHOe2K{^+svd
zYQ{(mqi}4d2o`?cS-SGeNRTt2u9P5kKw7X~Pz5~+kV<q-Sv^GzG<kV{U%@#N(F-JW
zL^vonlPOb`0V?XX^v_GbR~r@I8WN-jg@4{{rcRP8pR8rPT4yDc{o=~!#Ixd<O!{XR
z%5`MdQ_8+8F+@ogvnkb?dqn%S%`_8KDVK<DE|Irkq7kMhAXU9B_67d2gu$&v@H4!@
zy)w<#?;gdql&BJ5mso0qAN3!RWsg71Cu8*!h=8S=!ZwdKG%0!hs`E)U!x?fXe>?X?
z=Bp|LRa;{L-+lauC@f0W<g)e!*WhzWS@PS*Cx@1XG3&XkQ#L=ABcalSuMhPG(x`&)
zM|VuhaiP8C`41B6G@1+(JS&xjWWjBNJCBolEMTF+EsCRpdcSBq9^uYdySUg>XNCI4
zM0~&3xJ<^!%&NXZE&Gqy$bwb-UrWyW=6?pnCkbyQ(U_UM7Z#2r=-VRPJ3cQdfBjVL
z1ZTZRsk{;5d5u4W<I6+VMRC1*)Sf*hlAsAuaGz*w^Laz#%@KBk&`(S`nB`_$)eL04
z&=Nu1-=oWeJb3{8@zACX=bT8d@(Ob2wsj7SKnf!~l4=fJrbwL+i-~jT5^`ImaqkX4
zQBUmdd9#8cYEq@-KWuN@N1!dsd#%F0qi?bD1$_9h?)pI#3xulPJ#+nPL98t)->(}S
z7n`FpK*m@1owMy{?{&F~>xBK<6Z`apAlms}*awSDID_D}k%L|UZdkG)_R?P`q+Hrg
zma?N?dflOP^oy4%kEAOf)){Lb_;{m^K3I8}#>tR!)aS<b)xD}+dM@aQe`3j=jIx68
z&>x48Dc^)I^cokR1}pP35-31Mrz80dv1#1>=+-dRWyHQ9!0#{l)@cGs@*|JkH`%2>
z4n=xS*tLo|`V6H@J2}Dwll<9_2hEW@(k)polDg}bi{Kbl$L5FAv9VvQj#36_FNTtC
z3YJLT`_VF2Mj!UP`}jT~M18^b9^_T3m3h3~T1fHBF!NtmgRz9r7-i0By&it+^%7$S
zb6!@hEy3%Kj9eoXckTR0^o3u^?Te0*9_q@E>f)t})2W2~uUFYeLZY7d5GI<Zrze{)
zEdqcSLnoA@`{UYIs!uEj8Mp`QNPeb)5f3QMk`J9tTES7SfVNY8GdOAZjf`67&U){s
zcjpa^N@+cMfWa{@ny&U%o|j8enVZW9kMEC6_1*_-M-g?$ZTc{=LxIWp_nn!po29lX
zy<ROD?&l$r$--2&eB_o5+%tUQC8HsFLJ#NSLJxOD>Ky&ggyrf;2IC0fps8XIyu2K`
zmKL<jdSZ8*&E1esah`yKW?~82MU?d?IH$}+k@}!JMW6^z7ijY5FLHkg+!oDFdr615
z`ERIyF%nRLUH*MdzreSbzepY36favH<1Ggx7o$^ZvY&T?ppoAoeKqMZ?38>awn$#)
z<xz~46ljvD9rvrZcQ5hl`9b_mV+opj$*<o3Tw<*l`q?Nvx|iw1P(RB7nVjEOBA2oN
zRQNgfE)oST6ixK%kQtURIt=+HAk~%OtJ53D$$CpVPK0ygcHxacNGRwG`^rJCn+KoV
z6iRBR5&4Ex^YZlZ`?=W~3EjEeCv?U4OwBx6)(yW|)_p4CMIME3F-Ed7W4qDvT5JB9
zmirllNjgi+?FZqIIY8F;Ox+??OIale=kTb{@kRn~XlNi6nZ7XeQ(e+HeA3N9Ew>ix
z=`m>H9mWJ;)wCufI|;eMV^Il**|7B?{bR5XoHFP*+(W-BweWhVVkf>n-W_a>_yRa-
z#mt!WI%&Vo#??;FAgF?rV9XpmE1Q%LRs``=Gd2T+h^O8;V5boHu@=Yvd+hkAMdC2I
zeqZxqtHUDL5k&^<3-|<A4Fu%4ex3vdS#pyIBTS8-_R)IYeYRWBuJpd8=jrIQLs^Fx
z**igac0Ct2YkItL5%+X1qqSKma=exuxOiX`>RQ_)EQN&b=+3p@oS4IZSH7Mg-;d9*
zO1{pLpF86vXmmw+O|HijzVmyq=^m;_$*{YBxfR|BETo{CD`D5rh^JN0n+#X`Q>JFE
zw-yNHXwi;brQH*IDx<f+laGS?!{_Mk9ymE!utJ70!^e;OwMWC!wV;kSC<IP6`KNdC
zwgkDm_lWO^J@{}FmyH5>zf>Wl&^yM$ca#6s=;gTOmE82wqYcPI@_Bcb*Xk|z>t?Xc
z(uvv{b}Ko?+Ca~|&#^izz?e+4Iuabsziq(&^^9w!!yJ#HEPE`}iXyFXzpR158*b=B
zhJX8E_pRLnap?HL$<lyg$N^wZ_G|sOxRlqZDmj_OZg1vqnll-t<#6<Q$8sX_-;y@e
z)rQf<9UZv%Le)4k*T;USyTS14D@KP{6`^L4;#7F^o@E34rDF<LWF`p;C`A-9_!rsy
z5OJScipKV(tGQXd70XR8<-!>1h=sq#BEa~M{Wj?E#%3-w?<WXvn-)byBuJ=-{Qo6C
z-NrQ*CV5$A`94PKcaoeVtt$KkME*)JpA!ttVD&w?G)*&g8TtE8zchlX!i-?eY?)n5
z_=m#Oa~nRw`OxDP;Oe2+_qwcY9E59(dJfupf1>xz$VJy+rRg+0j^|8%4mNhxG@vpp
z4k@qk0g<?!)8d!EH}cTvHO=J;9^ksqZRt8&jhOc~G{K0;KSC6$91Gm2$a*s0(>X~V
zlrd}!ygpM*h(cQJsOSo{4V~qgBSJ%FA+yhP>5;`AL>u@9Xs=Nn95I}V7q`^`OZR6X
zb)OkN?(?s2+~<AbR^wITpz!hTseuSn$#$_W9e0MR#<o)9<<t#d+(h@Bd{`oOZ<tts
zKw?_mZAD#0QwmP!Y-Lj*(%U!m+!UAsOwt9QUp=d_%V{Ksj)nk@X{p&ioyN?9adPei
zbTl;O9p^McarU`aWc#hG!pR*;C@>FiDI#g#$LEUk>8pzK`WIi8FH~~l9$c(aB{)|u
zow|||MJF=0I9F9_g>u8)RTxjZn<WCu6wHh%#Vz_RY&F0?mg0U(*-A*oD>=A85~^?-
z__4Q4<9A3@Ok8ec{Z@ES4He(3J5~g4e<u(W1;Ql|y^}HRSBbMGLtN0)WW7?*PD`+o
z!n@i|GXB@TfJ!KcKeM_-A9<dREv_`Plq|oS%x+2)Q^F8z6Z!#t(f@Z#$|eu)?tzFc
z9^1q_ayNcDe)RdX<r6ocdgifoc}2PBzkL+X0LZD36ma;3W&(lhLdeYVOcl#LKOWV6
zT2~0t&L!_OCfOyLGlqKsd!v}7+1tm!;o{$}xQTJ|#|kg9^P=4yB|g!98tZO8h8vP^
zja`8utSNXcZnZTb(T=9s5DshAm3g@EJ>EADECghp#u*8>uUl%4W`EM69Hgo}>>6us
zyi9sXDxj5aN_uW{G|&7umfZq6>+-V;z5?*97)q~KamX(h(~R6;u2l5(Q6H`MUM*Iu
zRS?PvA#yXqJl#A2z0T8nzXte^N@M|J&PkSjolO0DoxAV@p5Iy^(#l;M5sxF)Fs|LR
z%qqujjq+y>&pyxk@>eFF1lOl@%zfpbC&E^??Oy$>(|Ws2j(<2NFov(vP|Ct14c68F
zT8$z@dy~WGt1_ZzU_q8JheuLXW9Vq5`C1>WywqRLyt!rb9*M*0&MyObpheHl&Yo9c
zc4VUN!=LT}ukn<0eGKa&X#pV<at2+<(X^Y@P~E?oNiJuBl}l1s(vJQ4K;|=d47Gd_
zKbA7;&=$>9sDeSx?=m%zWNFp%S!3=6=QCgG8&3V<dU=u?l*z&ret@o96nDS0dDP19
z^S2FQh(p_)aIgnM_E+Y=>|LEUlkojf{WGyGvpwfL-qq|~#~3^GZ=?D*5{mRRu@yJa
zY7pbE^d3Jh9&&~OSS)NanvX(-qklEixIqs&YUX`Yl9hJmw~{g@ni+KC@@&`cQr@)G
zIHdWAmP5U3FS9STR^JVEnusFlsha@sI;fJZYxc^NO(i(fHu8>vL4iC24&gMQR1|l2
zY`CI4jc0*Rzf2odnZL9{=%!kQ%f`{|UuGY|?6Xkd8v~-g)7Gz1u2a5vBR^%7v<W&q
z{o*?8G!eYvN}J^3w47+Q7XHfNvDQOCcG`u$F#iLQ7KLIerK3@QG2OMO{{7J}H^3~D
z1OC9^JFI;Ldbpr>0%?92VhkeAtlF~%?9*S_COGOeb*-5*`7n;%azUhWP&6yN2>Jc%
zz-c1?rb>ARAzu`g=GPP#i!$YCY(FpfFFzx1Z&)MBz8{V0RhoefEBJ+erpIQyOUG~!
z?6pLD*YFG`#9hT%JiYKwe`<PHF_z)=_x$JP-3mF!i_)A<|NM_?x$ibOzE2A;HAq4B
zZl1Qg<IVmK77Ig{n|o+aaJ76KV>r(0(RGdj<D<6SYHs!U+noKVVZ<ejCcXYXm7Y>p
z63b(_3}4YQnl!TWXcVNacRcilt`igA7ooaJ?S;VvOzX;N@?p<*WN*i+B3<O?rG*{|
zG2b9ulj$&M-)hu!EdAL%rSMM6gP!;OO@wchYxk&5WjdMnj6==mzYeA(VWT|w?bDf;
z`a)FlgrzRx?mji6ly%|bNK=*K`R8jVNZ$Yc32KO=($Eg`upgJl2$<~Zg6o<_!-1_f
zx*4!Gc?mXbfiO`rgor>PI7p7XL*05+4FjEh>_~zZk=!`GTw02q(C)!Bq^Zd(I+Zm#
z^Ou43wKE8$q&|UKKU$ig3qrEaR~a6hAui(t(`XdA{+=D`D$#DuC7)A&o5B5q&02j;
z8`=T<-mcBf3ktnh`9ecGrJ_Q@t3VmYv$gT@=?ogAM-V}i@;1}^YHoG?sGi0#*e(dX
z>|WT-16q>W*)JOT%-Js~8+w@n&L^HyUg_-xc}D~c9Ce?Zz;myk90@DmyJIT8u<-O0
zI&LYK4ec7)>*?f8ILUHsB(=YPZ7_m#aQ|3D?2()BL#t%2ejyO_X$^VZpIEMTs5t4<
za&kJsygB@K^76s!+$L++Ss3&coULxt;JX+vBzPzYxwL}Ih;tY2ycSZh8-WLB)VSO4
z=KtLPLkkjI-Z)U-0l##3P#fGwrmE_hW+qsc*M1})=T#q;p*O8$42s3NFhkBabNaDC
zb8s9n5=-fcrAQ#a%#xyYUWs_MGMGs3*c#`s7@ZzV=Wg-Nne|0hAR0|Q?ybM?C6diq
zI~H3e060yh*^MY5uW(C7YWIEiS<75ApA4u&)gcTPlMAzuOCw2+i$_y!Nz92*1@rp^
zsw^Q1C_$8j@`7aCRBhAAN<v6nyn6+-?9MVxYjW^_B(TQ4Ln|sw>N(;NOOsT6F$+Rz
zHLTLH+PNQ!DZ4R0yMC2skR5sUkl+5G;aHC4otS~M`0UiET*rC1>fbYlBmlmFrnTe-
zmMrsYM0sMk^Yi%0uco2{)`)p&<dl0w2F=g&!V7BoRQIUIR0;Ee!Nv$UyTzm!<d|~o
z!D%T_+$n!{^vEWFTv-iHBX3B*QDv=aUA`BoMEYYc5s@)BK2!oHo1>`g*W*&3iGa?e
z>1~KwW0U7N!5cp^>c1vr6b9fBX~^2lUzQWahK^^>JMdHRtFPAPu8@K2!xha5YtQS=
zS@$*I^>4?O<;~?Ar}AoDlQ+&GBSXdxw;G++rq0^_XHkVC9eP)9;kK)dqJbN5N8*Pc
zfb!7Q0|__Bq<D0*`0DV3oM>x%i3ikog7Os6D0;gE%Ko^@dBTe1fJFdMf~TQH3fj_X
zuj9;CJqahV^gb!Oy>s&FPH1AiP(iHdv1W>6K*yoM3|r+<>+g0hY1Qh00eEQgM)-d%
z&!@C?(a4ql-KC!TLb;L<)^gGLZB>-rJ%3kXoYgG1*k3blUYQzXI!J^E5?~lx{kyq@
z`z{Ecceo>8L*Im!R)b{P<-EBWml35*^Uyn_K8&ir+-#uyaB<$`jSOgkEwW6zy#)1i
zTkTq1qp??m*7V?(@GR8F{6U!z9a>CLqw?;I^vqP{j8>0ebfU|5&_e8;I|P;$2}8E4
zMd^F0Gz%<I8yZ*}!t3Pl>4toS=NS53Skgsh!#>fZ63wm%+IEgz6`tp#`FTvA7=?cP
zm;bz9-*GXq`Ff>y?i~QZ$Qoyx-B_L=V|b}}Z3c4z@7A9igfn19OZh&TX-QjhdH9<s
zyH$-eN#9?8TS*4FA(ICt)~JNXAkEt_G<jSm61$cI^$GcQzJDBBANi!E=2|6eEB6fq
zPUUJY&>Lo6V_@3$?Q=!&@xNVX@`G4iA1k9VX+9B#^TKQDv08f-{2T~@WL;j&dZ8m3
zPmU7ovQ##Z*(*XSZJ?IOzmRG6M%!^;EFUo3uMG5IezJYPmn7xJop#wLrwd1C4AvX%
z4nCGorzbeN+&B^J+$xMQ7;F&xh%;kk2UqX!e&BWH9@F3in!5$UlawTwUIP=VfEB)<
zSR3SuqVxrv=OuoNC4jhum%LIMbU|@+G$1*g!|p?C=!raCW;zFYiX$muxVXpA&|eKa
zwO-Dp{qN<t^HDCIX@8VgaC^l<m%D7roCgujodh@%zH7lx^4(SZxVm85{<I|>Q1$PL
z$Vx+$&EfA6BXk?UCcQ6~#!#p9BTX3Aa7<g%<_Ngse!Y2Z3dz{qX%aiz7kE7DRktYw
zyO*`RG}zvr!Q~6~GX>bIn<?~}#ZPe&@6S-UZOyH?9lFBD*N*75`(fZbG4V95xi07B
zL{e8bYAV4x(8rYSy$}@UwT*ILoqjMfp<ZwKs6|>0H?A8~k?w`H^!T^5@Y|hXtdK!S
z3sS&q?cLnR?jBqy6ci0revHpDydgVD3(X!(CbqrV-`rKN{mhS2Y?uDjU1C!C37|X;
z+XLmC&|X=5H^GC~*PP-PC<>|UvqQ92vgkB#Y(~*2{O}i(YK4B}+L)Gw>iEPJCfIz;
z7TdoL?ukO9<Ikhzv}~nSZru^&7oJD5>^5<!oq>aof?kx!=Lv?>G!c#mc8%*^-O>nr
zMV2xNO`1Aq;;VEQ?(Umh+Z(MetCqJx3M*-+nn^EX2fOWw83VBE(~Ls%RZJYuM$GDd
zhyJ}xBpet^y{h##fCPNN@@jwQXirZQY%KD*Z}d-H0B+X0iQ$PdC2FLnCHM8OjAYE)
z7P9dhqNQSZ1c6-P9SJOB1$q-C5+;*pfjK3y^o5QyM_2lL`Oji9-goGAB+_cw<>lq~
zclgRuGQM{3Q<bJhZlr~#`yp?vmbqa85=fT7?;Vm|w-KfWUtb)RMm|I8nv3yGruN3H
z<ilm})wQ0%?TaN-D(4%77*Dahm|?;&VrGr~9Ix^Iu;Qeg5u(^po@jKK2t83$Y8!90
zg)=ks6Vl^Ob_s444j$cw(g#VpA{8w5t<%M*&MUVz$6v@470gFl&Iu`TH4z+z+q%(<
z37I~x4m-4t1Wt%@myDZL=%;R6;_>qsZVmT-ItD=<Z00i5@7>oa3W%#V4a>(lDSS>R
z@?(B(r4aI(53A@`_v;UTF7(Y}w|d`&W2D9m95H;rAgpFhRFEB7-Hwadm69$Sc;qWi
zx{+EULGn}1dWIXFJD!w0cZ$zJPA`W8p-E$Mt=ZXZrn1h#%2CKI9e?Fv0K6O?+~5R-
zC(ydq<q=IG;AKEC5?4U2;Yo7_qtR_-WG$exma!Z;#-OSl1PZ?nTPl0SEvXi_5&!LG
zio;V+RMtJ_z5R?Sy#4&voH&?<@Aw7|Ia}gZiKIr_t&M;UVUH)M8NwNv>f95k%mjij
z2&pSv7pZdBv*eZAi|W+&Qam$Nz6p6BhA`;dudc3}bgZ)_`oDH1y@;2)dCd?{UkqM&
zYzw#CgX7e8S-|(g8yDA{l5s7XGj#Ni5{u2`i{V>8MO~0SSpG$ctaxs){XJSBn{{MX
z^V|dHcpD-C30sVm&u!s3w()NBy{c*>X2Pe{*SKi_s2Eh(rs)`co9JycCj(XX5|!=o
zbEdId$A!S7L1c9m6Y-t|32D9`d3ibYoW2t8t=oLY{uc1l$hIb)X+E!*HtV^u;Y9Et
zzgI1sw~{__)*u~j_gfUqv3sdkV3l<OhpOAR1AfE~suRi>^1a23Ky57{U?5=(h{%pQ
z_mRl5>~|S$6-vR&DdrFkt<z0M8nL*4f$|W)-qvPtmt+6|x2rPg5Rr{kA;^qa0crBB
z$8LQ{YNH-W{0;V55cDPC3_@|)(ln{I6}Zgro{)s4_8S^&Ay$1nPvAFP8sNDD(FpG`
zR~K6F89d8ci(Zg;XC&Ebc2w?4V8ASHh&~Nv=Y=m~Moa>BG{pR<f#`LteH@s7{2V`V
zWSArA0KZg6mbR`H$o|$^v>1w~s|`Srj7$_m&}X2>3toAD?^wfx9BY6tA*JR-YPm%B
zc1Zbz%{#H8C?xzS6y``sIN<NN!rQ}p{{b7$I;vn;;FnaMOzD8&S3eq2l%p`YN|K-w
zM~65{{bh5yZ8g>aqPAM@wU@q`claRl2R@(frAh-ilQ?1Hvw*(`nowK(835C*cIr5f
zOZI14BPJg{St^0&K!O@v>{sEo;G|gM<qm?#d4^xLLPQcE)2*v@M!D+yCJjnFB!mxr
z4!81wgHu-@w-<xXT3?0EgrAe&NdG%oD^J+#Dee7bf}lhc6atjyfb^Yw#em>SOQrkm
zHCNXMp_Gy@G<XW$+B%qAXVa@+?|;X)QMNLvVfWu2c@0L@q=9$tz~-cga^p{n&)}@r
zr-Rpw+g<y>ZljDB9JiL(j;EbhzoTb78c;FOKu@n7q4N7`{)Ho$NgJ$UY&%1sVZIZG
zr8`8AhpKY8(Z=0dQKiuiA4Yw19eF!0IYQqUb@E76W<u*XuRyK$@b7u~lwlHfbmrsk
z`+<St7Vx_!feD-<F>N`o`~hNK@qXXdulS$BCdJ05S7ff5;4RTBLq{>{nuZ)r+0?31
zjM_L_YK4K6o~uGq$OJEWG&0t$Bpf;(5>70Wm{PY2d(Q@AltoUA;4TYIxW(>b-Mg_B
z8aiCHO9`AThn0^*7Tf9fjlLP;UQ)wEwh6}nwh^#$IOlJUOe@j*%*}vaCPNn@?afYe
zmGiCjji;#UBMR!f1!TEDC8H!Tvsx4=_hRFamIQKb3+AY`O$nu@HXLJZ`Qk=n;ug-_
zNilI;xw6?xNNSma3grQ-(05OWS93&d>}CpRZT~#W@M<HBLbxSJWDK`B-|HMApq1wr
z9`gB-F(z$AT1OX?Td<Bzer;;>AxhariBdSKLT-t}H&u0zZNqg<458ZIa&EG6?Mz^J
z^a*Ac8J@ux;iW0>k7YlV@sv0mUHS6c7m_+Y>YQJP>T!*;2z(!hs1ZKTlITko=J*k$
z>uged+O5RP!ACP3!vZ+3`C(5EXtz;3ihn@Su8!9+q!QZ7bCtz8@nFdlGUQv&V#daF
zF};6AK<GZ`UV($P<!0q{AhE*KY!BYw=0)t6?}hcvj9zbZMO5_dc%X;43L@?5db)WH
zoNRWkd>OFZvfw)(Y3S%|D72r_XsRn5y(e@BvfQdv>(vTwkUAzQk9Tol|5U^FscyV=
z1UHX#*=LBz2)cYOg>IDx=@)IH&N--iYW^N!@_xRb*Khd(%ESaGVE$_Xpo?#|OuJX+
zUTR^yvb8u47(EodV%ZD-Ona^r*NdX9TI>Dwi@lrI4b`bWXj6RRy0xu3GAlhSS~&EI
ztMkMKWcbpJAntC`#rP^PlW=Y45{aMY#+zB@^&q6F$J<&5{il-!t107e^9mamtpI;6
zQJCn?w0{xef8vEd*x=33wX2t7aJ@Tde7))dyc;L=A6UJPNu1$0je=erTGPACYNYH^
zUkNbrrPK5+Go#28(R<Y%tH@&5#VHpTlpD}mbe%tN)cl$Jq?$vu_cxvf5dtZ(Ef`Pe
zbIbq4LugYL%oqmBmy;up;KJ=MVNnZin;<}m`xZ`?Lq;9Q%VLON8)iz9!pKH5`K1;n
zEwVrM-a!s!%I++sd78{th$nk!Gn#Sy$9TxzF&gO<>6MX5rXczFA6jcs$PmZKKW#CS
zm#8K37Xhb>w~4z)6S6exH;o{wO`TX<CkRLM7+Ub%*=*lGd1l`g=<_1|c23NN*Pk+W
zWMpE$R#tQ(6FWEAR0)Yd1Xh<%R!KWWnTkyP%ml<Lx?w&zu>BzptAJESazT)+PWQXS
zCQkjXIBapEfH6JEMjCCA^Xn3?!(pDu)aDB#<^z;0$;acfOL*&bP4C+vqJ=olG)a@1
z`iFKO<E2tQwA`C;JfA3YzOsskYFA&KqwxoXHilcwNrPCzA2VWWhSKgWcHz8?1Bewh
z%f|9l;%>JV65rUBqo8O5a~{5put13L?#*trQPx|IoIShD@$Jl56$RbCTk)4rby~3{
z5ozAuI!(c@l5%WoFJp4TKFUQio^as3`NB9th9fGMlbpj#{l<8?4r}Cc?=Ul3+0#rp
z9Gx#*&yW1|i2!4On#LRmYJVG$xRpH3!oc0RM*~m!>yKwwKUTbswa|ZI)Fy#UOFT}7
z1QYWq|H5Y<Kk2mQ$qFMb!~JaTDuk1>d3<N9EpY6za81ajbJn%B#dr2IGQ{imY9UF+
zQ!A1YfkVsMbo(qy{c@{y?}A!Q$l3F*{>bXlZ3+hQj#&#q+E3x~vA!g0+pioGbSGz^
z@oiB$@xeyQ(L=U|AEAIf^$=5R(Mu7ciXS}~?-ALsKiD#NM`*G<%QPm%vqEu$3{2>L
z<v<<$-5Rp8!;(wb1Ag;KGwFg3<Q$0zS#KU=jT>n3kiZz&B8<GDw#<54&S6^fjVY+{
z$FWpqs&c+!^1~Oqf%#tzh;BxwM|sfY(<uN{nft1UGAPqTTC+GpafJ$p&4IkPaaOTU
zF@iUE)93m@{(nt7`Y8R1k!R+9%+|PJB?ZN%caeQztH5BcXMA5XETB~Y>)q5&DG0VS
z&BP2%0sTa3kqxTzXf?;+s#ZBX&R+=+dD*60T-0q;9~1CxCSs~^mX1oc?GTI|${=&F
z1}@56dVQVW{RH$#7R3kI(0D~2ejQ`h$e=z5l^{Y2N8^zCAw~EchD?8c^(A;LEp<2x
z=jKJa?Z2=QJFDxNvkt+ammOW$FGh`?zR;|fZe}(uk~xLY9$Wm-YmGuii3x}0YN80g
zy3Lz4BGt=DG3w=`xUs@p4pb$K0%9G5twce59o2$ATvHpqzkov#<!OQ4;Vh``17SU1
zot@oBra=?;vpwsu!23sUvoj7yeTmqx#x%8Xy_u!!dmwaMy%F^iyP`@zCWU|mV!D&Q
zDCOh^`0in0kObaeM#F-x!2L!}^xk5m|7&5KO1n<rzs*U%SMH<9&|;jo5fM2zthKB~
z1c84m<P=xie@$Pul3oSN2c)5uXfEZ+#68*~u1!jq?1m}1bHu#*%q%w8IjMT**nk2K
zWm^u5#LYIW^V5C^M`0S~qRneSy}1@aBhHxuh<dI_kO$R2|K>E9-n1<UCu#_%YtLhD
zsGib>v-C7y@EQ#{VIvbd6)&ngIamFCGO6`!Z(!x6SzGpU<BAV!^o6umGvQz%Ia<9p
zcgI6|+GIbq{!bM{*KzH4#0<_ubj=<P(+J?;FDOzHUX_&fEQ49?BGHz)z%K`fao|&H
zzQ<*Pi<QS!x|0)1JpQQGI6N=rcRU3_Q4^VCw_kCh%S^?1Gx3e7W8(`AD7Lc@@f^!7
z)(LH{!Fec@4QJ|<xD2VF1(JN4`N?um7K@(EOjPaSjyDq8HdkQ@yV<2T_g0ujRn=wM
ziy4NSSy55qtr}^c(DzX~+W10Bl?hE!gZf#I*8SeRrt|y<#IcN|;W(|Jdt|sI)aiHL
zVWr?zkZ!H5W}i>#aSUD}L^3Bt3ReASi9AjkH-~<VkG`pb>gfP*cjsm!1C^Ic0a1vl
zUhq*!CsU+yH9?Fe*Nw&oyH;v9cY2+Rvk1JVbCl>o`iN@4*ZcIXYcP{!+q!lII!(k@
zN`}zqq$~PSMe6Zu=iaqf<c0|VV3<Lxe%kZFA7wSH*g<lk04lTI&f*^XyhOkkI}MW^
z2si>|@|LiLTsv7T$9L><-i#3hJBhbUT2C>rMp+x4MIq?g7N`m_Rww5!xYtNQlWFj@
zUhT(t$T3z(F}%<J4m?ertnc<5_#Eori>#Yg!``WVq>DYF{e;}2X`%hGqT`>SqX@rx
z<E9f|vfhgFr{|A8ddbPa(ERWCu*SL8*CC2GY7g4H8@@Li3V~E7oe?1Q_wL*uRG}s-
zGmP*_jr+r^bz>gf=#Vhs+c_|#;TGbFJ4BTTwNM=so7;;q5m{(M^5sKJlEX*y5GEs?
zgBb6@e@+e<lP=k;5^{crF4udOk>TBPRSipRBN6%I&U(+a+IQp%W0GSZUTkseU`k4U
zSa%w3Wa{}jrkPa3>^3qo&irC*VaHK#m|G_8b*HQJC9-vw4Lq~F0)^R-7RXsk5ecNo
zcC&IaMt+8<AK}@mjrU$gq_9n3R5~%2Pv%HXkt_e0+1(`UoAY60TpB<AnZk|ayDG9a
zP853QXu3A9nS#M#Ne8uw?2E@A2JFW+t>x5BX~5oo!x5hq(^6K&`3RA<9%F5yfy}}P
zIh#m4-6b<JrwjBx3-wOzmlBObW1U!2Uow>j^jtEB=J|Wsx8asiTDY)xg}zd6R>i_p
zR<6_D+xCC;O7!U#^-vmza=tIao5t6vhHr6HW)3*O&5A;YMYEe_QR&XeL7Lb~SOYoi
z+@%j^8oLt}%&A6}iKe%Q2rSUl=dDari6`vrQlgnD4G`VotX=V)38jIwv~S<2j*Ytl
zZS@^g6da;9INW&piE763_c(cFW*|CYuyN64HQ^nidYKAHfn*FSXs8I3G*ME|ujj3f
z**ixyK!pS2pKv(9@juS?Of_(tCZTxObhd6FlO$3lM*&ykMOffb)BQ?laDBag*VLDd
zkWiGOoeb?&v`=|4iyh6o{Y*BmYIo1sG~uK{P#oN{eERmx2Skbg60+wm&>?vK?<Ux&
z{gL0>;}&i`qhly<U~({y=E^#X=unNVtGnv`eun6Y!Nqg_0ODddM_HBLN=JQT<ab1G
zJ@YXs*CSqh44eBl_&auVZF+6>RC-<-C$|VzP}Z?7G$9@ulkWXn6}EdCT`~F^l(wu_
z-Y*E=90A9&7=gpmbGS@Nohmb%iIM&Fz5W*fqCj20#@u#Xc5dqYp{~!lZ>UQw-KjCK
zy({+~b!-#n!kIy^oNH^WyvR+u9J7P*c<0Q}q$KOwaV&8_eyX^_%~+IVCx3gd&J3>e
z_cfjMX?b;QVG6fj{jj$ELw8_%%q__OJhY~8cbquR&W7#MnviAIu-Z$j^?BAk1vw|0
ztuecYX#yD;MX+k<RLPb|*-MrY*|sHRia@sHERV1-HKc?@wtd5E`C7!m^=;IQoQ5d1
zCd~7ZAvycz(lG|sINZ$}@niuS)@J6}*(i&$aXytb%-l<3I@69_$FHEWBcbeuf#dN&
zs|Tc{$b^>wu+H42>|mxfql{OSGE(P;4vD+n9q+Wae4bHA)#t4eDhsv0Up<s0;;xDK
zXzA=(Yc6s;WM@f3I?M3i$>C)YkB!SYG46J>xiTGRL=v_6_qBQLVb!<>ODD!RuzFgS
zSVLp3Gt)FP&B}4Ah!mfb!+`IM-oi`7+M}T_nDFmYT0J^S@=|m$4|y$TMY1mB39R#q
z$Ash3FLPy{ClGV->v5n(h(ce3n%0mJI}2sGv^Um1(-^=71j+-Nc`uu+?=3i-{BEva
zxq!h=Q}mby>F3^}l^666SioQBS6<Z~iVa)#?TcYrKVQFG=#=H9Apl#BZYa3xau`s9
zyE^mi*)#r&5B?e--+jzfE8o*^GTR(r%TKmw)@>-AZmEL=QDi6G*lNNdQFy#hUgmLs
zzDb{W?Bv>06boPSP`e9JK$C+s-6Vf3`$cfMsz#Qo@03IBumf__P!tuAzJH27XK0g`
zu=?uVZ<7y+s*Sg<-{Rl?;=jumc5gdqjXtM%AZiXSWdmXsS8HdmmWOul$v=1>Ue?Y5
zqMRLDWCOBK0O1M-mk});K%N>&r->8&cX!pDyaNw(d37$!_QR83djR};7+V;Od;7@t
zWN!|UoxW0XA4~rjH&tin%!m+;LN3n6cqui(BH&enWx;f1fn(ga){tzGnvtM2#3sfu
zvS%ebLzQ97kZ_l<Hm`KCm}(Nbcgi9+r0NM+;0A>@Me~j_K4#eCF2d3%xgr=hca7;-
zo!vumNxI53SEjjP;)w(<8Kw~GRJzEo7@n|lz~-4;MoLb!oc;RoI7n^FKy(-O_z1xm
z_CAtxPRx^g6(kjQ!y|Tsc>LaBrZrg--!ek8NPI9T;u+QMI>#AGbyspquB(_Ztj$#Q
z^yM^UL>1A7rLw$9LmANh4#XMXT@uQ1qN-9-j%SFMvXdOOk;U1&?!ht^=1iMg=$`o<
zay7-v;SFU+9tJ8d2p$$YViqxlt8MIazcN?ZyvqR=AprcmWH}pPwYfq?vW>;($tuY^
z<>K4sNjZjKHHUtzYfyfE-rgVW_L(PvbhK|9ztPLszt~{iPWn<}>nM$Fd=Zol3wVDa
zI|<T1ImN?$kcT#Y&9*&a+OR51Lddp7Tw3eu=^?%f{$k(FGtX{r`CA|SH9o%mnBP78
z7L(4-qM0qRIEE~lds`;U1z}e`EBe;v>G#_pW5*<JO}4P;`vq+^ygEMGf*%WOyDt<y
zVZ_YiPWtaF$Rgxm*7!RYOZCZhJ$C45W$$n6K+<BO4>V=sY+u!x-+cBJuI3$+%{<vX
z=Fh(S=XjDH2T=9Nsk^R{Zb0*LF{cl+&t*Xx#>j}{ybLF94|5WJ@W2MkIs3TlfN&k5
z5A7!#1lWIfZAuBN==8l<PH}dfNz=pEu(geP5c~o~`^Y2WSi247-*liVr<e2tx07&+
z%_YRy9P7#}AH|PNGxIU6>cYcG+yyHsg7#xPC;WYDTH}3Zc~{-V>+U)YEkK&fo3yHn
z2y&8<VVBtrMiGU2WSWl5;!IN#FAwH2QHGGCl$#rgp3w>A>KzW(g_N%u9u1%~xy)Yr
zZEnoXnAPLzl6ZJXs{#HXvVPU;{H9FGf`lfa%UxLr4*M~zO;I9k4M#C;NY0GA5!HqT
zvt)T>_ts-f64UMT$Z%~u-#w?`i6fbrl;$v0*Vtw424u{ks-Z2=<CG9`N$wIY;(f3%
zh%?U0*G8)|IrTZNC)x~*`%F@(95J~em5gLhjwzA6mx|vHHJB<Pw6C7YyS$ST&)w(Y
zaLY?}0qCwRNNUtrL%`>!Qv5t7LMkO<FeT>hzCLW?rN!fW%|JB3C!tRn%wmgmC*yUX
zI}u>&vEd#c?%Cl*P6h8T?aj3}4geQ-p>?6!ZQ8(VfWHUY2?ODLb*un#ww-_8S&P?M
zpdaqrt3lufhd$Wfv;j%Z>UlEDvW*CqK{TThWq9V9!#pv~jlcc$Z}Rl+8NYe+8%*`+
z{S2MIH8H_2HU+?57|3An*2N$eZ0R*G(^3s!S)<Js%hG{^gk*uN?&A#`0R34TeK&+P
z{!evL!S{Z!eBNq1V-kBNnFQ-65qlIx?IZZz%o%~!D$>*wd14V%Gpts)Grpq#g#9=&
z&A0ryFZ>1Gz5Ws@&fVR?_Rtf201WoUrT^Y>mu&uiVcWQEHXog|&gikk6rQ6dIoWUP
zN|rw0PmbR$Chwh`TwWLg+koo_bpB4{RBZcmGsa|P0#5|dAG{$y<mCPU0NfsV`(Aw>
zT5IOobSCr!08dWH%R=ppy;r)J=gDKL=pym{E~QeiR8Ywwq18Z?R)wZ+bk3HjyM$zC
zkCL<yH}-MZ^#W6N_hp?cxm8+ii+5oh3h(~hAK+_`e}jA+xw;Z$o=K^I3><Z)l401v
z@fq3<s5a&%Xxh;;(n$kS#xV9PLdEJ~6PIk8xrwD@9?ej3Jr&bR8Ah5mZjaB%X=H!>
z2+4)R1S$hV87ZT3S1Qt+&2(3?mm`xDl!M|`F}eYa-xs4qnL9yD9f&6Inc3gnlnbT!
zbI=w5eQ~cx@|cvEBMnQ&te*HHMJOpb<7@s51mDNBaXd_@HpWqxbC`HsUQ=^`UKPnk
zkxbKu)rn@}@dZ63C?&G6B$$x18&u(JQFXVwS9Ecgdotg@SqNN<pg}%}xV~gE2gKy$
z-YoOEx>ipLQcKKprB-KdbGXef?ajA04*)mT#X0!v-@mdcu0zG@I_?YohU)VleXZ=&
z#3ttH#ckno?)B}p0P2QGIX_jOs6^I<u`?Aa>w2mieYdI2+&;hKZ-4lob2}fXwej22
zZ!&A6;a1Yh%DH9bF(O1fi2KF8OcW3<&ox=Y&($-2t)JgE2_38f)9xNntQt+W!MPpK
z<ptPmD`42-9&$IU+O)bhePd_xB6etrQH;#w1@!go<Q|I{rIQfDS_i8ZF7v1igJ((~
z=b2f3QT(UvpMV+v&hX#m?eqmdXPjr5v9=&-0btt^vM~^(-<tq|54nL)7{6Bo!1ctk
z{M*6Z%R&HGAb9e07FtrU=;KD_r9X;v`uCd9635xb1j^}meK%O|(f0MDk3m_t`e7s<
zy4>gH(XjpXha?w&B%3yWkHqU5Zf!kh0>KwdVX6L-vI7PX$}k`)v%h`|Wk)!NL+S&x
z8WWY4g)&y=S*R6uIY^?Ye+w=)y=4E2VVWKI<){Q?AbH@%?)c8#H>tIfbf$=s%fPV;
zb2BvWebCOgv<k?8412Pa@MsvRT#-^itRl>A#FibY$CP)sqx)Y;IWg`A>~O?vMq~hB
zni{vaAE7Gbyz}ZJb0y^ga&ZA=>C!Ki>3CqCj);i6u9JtNm}z(q6cIO|OY-^CBEz<|
zK03p*5W>-^tmD2LYHE#=!lj#(mes+&8P8kh*)d~y;z>#!Tt(uBwFs%1{n~@&+s0I9
zDKsW`)i<csGsP>Fq|RDeYaSOInysn3-ce7o>0+Vk@sxQn#xtpuo-geQiBo09XyP(A
zr4&lSPc8$w6;E&shA+ijucjeK<5f@6ri#g>z1jAe2thAdl-yrDUYiV1TQVG%C13$~
zD;Vk@uM6QC7`-qavC&D~R$pPy*^7;(`HIV2U0ll{Wm&2G{GBn`EP~=m&1z7aa(lSr
zZ+`!;^ZfRPuiXA46pwqWF<E<8WVvrj&+uG<mnH>WRp8fa+4L(|uh!biObH=2x^hm^
z-3}Xf`*zapqM)a8C@jBqfMr>*+LlgqReX)Cgbt2!4sENB@lsEWF}Dz{13u3bt4l?Z
z4m6`aW?vvwJ^Qh_y8Lh7{w9A~|13}JEmT_+kMAXuI`~`1(aG?XGc!Rpai!-;J$<s@
z96?UMI{9=NJJJDPz(0eN5`YVcx054u`IhB;OZ9}4t<Jz)7&)7<=y9I1ykaNf%pG;T
z2((IHkk?md!F?N)_``U2&Vk<xH!@CRg@24&ZFO<8_OYqKtoPyC|7)i4zGWmi8!u6Y
zY`1$%)56qh)C&)z#njhmQy2s>q1MKPQldQ28buKAbvJ&NIWp!H1Chde>AM`}8}x98
z>VaXGX*F?o$W(J+W?FrAE{~vGA$XEjwz>ctAHyDoA|@e8qH*03pSQ>Y1wbv?<Ivh2
zR<5bFVy%(N6__y{X7Vt4=*TpqI-`&F+@*=z`Ia2jJxVEzyB#Tcd>Yco&GioZd(-ER
z)|6?k&KimZ3+S1yLrF`el$2b-p`=b=Byvd@4Prif-h>Zsm4haB$rT_x@5b5O{*Lv;
zu2S-16t|{J?6znArkJ1m<n8<Ub4sC%K*$AgGM6tgjgd&B<}rs!qGohXh>LQOk}|i*
zfR>EPfJupULRWb)F>>iuXT0i(HFxE^x&%we!lk|W_QnC=Qm6N2?HqKSwXPd(#sYRa
zD3k5}8**u>wy}wwTH=(pr;G!vZm`q7S}<uVB4PnZ+rBF1b$t6_qpy7}jdX_4@B>9@
z3Qhb!a^_~5_#5B<>-_2mzsPZ(78Z<TOiL`DccKFoccZS0?{aQBpb?Ls2?lV@)Y{-?
z2v>%`=*v9~P;8w^D0#Ve9gHl`NY-q_1&H=IkHF7f=BQ@`E6Ea8f{6%g+55a6qc*R3
zsd1f)Zl5;lJbPd}EE91r3*{072_k!LZpvH-sy2S**)Q^^^1scsJwj4uEF~26Te!x0
z5(_8yOTu#9YEBxfd)iDQ%{IVv{}}*rAsLsp0e|tQbhfX5l){8rjHv$Lge@HHpJ<FX
zm#DzqF>H?~F<xV`@9RQ8XYxog#TM7kw!5|wFyHoP1)FkeKx~-5{_i=H`Z{f4I0cUz
zk}b4e-^b+@G5MsY_>pZI5-&Y|*Y9<NFeHsRBC=CMXcut?yz{XsCGt3;RU(0^yGDqZ
zw#+zWcB5B>=%E=RT@$WixO)?Wu?zvQ`%T9IVG?Ss+}(Xd$}`jSjFJr1k!R0uINCKx
zcA;u>p=23%Xu4uZnd9w8&~9mU#_BDa1`K7$9wSD$SV_XLxJ*3`u-AZ2*%Mq$EAw>g
z<2+@oO-y$mkcPLAJW%IbbUI)dcgH=qyXGut&x=SznM~YODyIVXT`U>##ErSR#$rN|
z;-k|OFT$534#Xs_QdkRNXfBL=m>N&#fx9ZMS(paYj9H<oM(7Ih-25C#61rhUhtb{h
zZd`l3v?O5|O1%HF7!Mr`cJPiW)jY1TTB50;8u8Z%wFxO1W4M>x10)Q?NJE)5bDV{$
z*%vl()=(pJCS~u(xEt8*LN;brYD?s@#;0F|pg)`T<^dqFAa2_U2@yJcuwl&3w^^Bw
zhi3JYrJyrM-87c7eKu@{EOXapK{_9Iwl0KgiKCST={i{_eU1KhK$swaig+2T?2L_J
zRPjZ6GUIQ3|NqCYe*DW!+OPmLHGoMysbXO#-R<#t<i5zpg@rYjBr1Ar1DAtRN--fg
zFsmU3okLXC7i&{RF&b+eVTw$&zA|?@zat6*G$nWOHV^D}Y~+hlPulPpCR-#<w6#d^
z@X54k3(Z12Ai7Ts+7m<s7jbQr5WSV;#oDbk7axtuZmyNNR=)E5H~1q@{sd!s<S#(n
zu6x*Q8IHCvCU7bfBP-fw5nd*ea+c3G4C^cCt7U=dF?}1vy_M-%&6zVW_L83<Cz37Q
zNHI3d&PHAdv6a}|+@I}n&)-G|Vh>+qe;3v&ArG{3V{yYyo?Q&5Uu?eG7=NFK=JuVe
z)i)yc4~wi=kM?u9CEK_~TSI#J3Gw%b)<ZjjK}`Nm?}m~6)yQrKB^y29Mq?0bX9~sL
zoSHq3u!mkW?S`pPa^ju&F-c|wmpPdx<z`Bhu~4T&%v)`#Z7WEK;Vbp{3{_*;@4;pu
zn5QV!=29}U8Go|OF)0Z{8hoy)&R+IJHr0wu&oD+v12!LNbt09KT7^a??;g``-;bk$
zHP5PU8gmKm#c=Rvaos&xZZ-NW*fWpQGmJ(~5-#-O_evpYY#aBmnKD%{b8+QdMuf)P
zt9aC?#L?z3l-B~6Ca$L7Avq56J<HbPoO740Aq;f$8on*G`#fC7<#v2DIg_VIXfbBb
z{H=}I8i`fx-($%km4Y;(k*E#-EV9BO`X`B!3PVn4?xpB9apV{9_nEcNBnjly(|QW>
z+;!Q(*otb-T~%?g<xhi$7m<f{Th9HWPQKzIgoRbxuKr7~xd>?`W!w_(j8AOR!4DbF
z@Xdmv)VVU5@^{|<FZkv6e<9=u5*eX(R}Ic^sdcN4$inmNsm@sSN!`1&f&(@4U|sQ=
zlAZ~;0Z?baWGUyil}5o9iGeNwu-qF2OXA;K=Nx(l@Tw@Ueev0xhgjTtJdgjLNQ(u#
z#^i0|43oHEw3zc%s25Ww2PACff$7zYluORkI&+*RhH>N{J^fYw*!7>}OX-W$>c!0y
zJ~;_1h<32Lvab@Bd(Up$@><KUo4Y9{%c9V)^#^X<ry4@&az1Wl^I5-V#!1F~iUpvL
z<qj$nt5IRH6!W-`sy3h5+22ch0ps6+rjL&|M9;~3`p?TU;?gG}>5=l11mP@u%m+ik
zs|Phteu%o*y)2g-dCvIwAZb`@)N150GmeF;tI@NJX%t)^P?DEs%$l%v^sLrMKFgA5
zA@1&(vQUcnF;qxxCg-T>eNaeyT2tof$Zo%LZO<}#7ON@RBGJRVgq&7kfBgg#mril}
zC0M0WAQ-jqbA$CVZOS+fsLc+RETMDg72?u2Hs3lRDkV^gmm|Av2Og^%yR9~~9T%g&
zHIH=z57(&{hLaI<Z&c<8Fgw>qI#j@))i`=xUo)SRv#t9|l8r$vh-73K8M$T$BxVe#
zb!M&}1DFHYTU3YY1K+ICW5By${^GJnmN!r};;(~%>BhLnGCJeiNLWg!c-FDSyr39a
zES_D$L>+HiVHh(O^{5rrIo%3LntQ;4JsfgD6^_&7Alp`nu1kCK?ac!~?Ep|tK;8)u
z<aEvOHRTM*oGmKrK<0GE&G#q2-&-xucYOJRbT4=x0t9vl(81{n31Sn^$QlQaTTA|F
z1E#t0{O*ol{P^$kOCSDi&t{ES32-;3vt&IFrf1^zn7KrcUUNMt??W$N)L$e6=6QDQ
zrD5<mJ6qNJ*VwzXxHp9WnK?%SNf25A)Y$CL=5q}<NEb`piblKMUdUIZ<n@39{7Xze
zBrlduOrkAf{B-lQ&1X>Pz!yPscE+!I5p}JRY1du*;>Vh%iMzWy{$cwS4u=DO>YYEs
zn170-UR5NgjfNpCwUL9eSPhS@-*#i<EZURS<EGBzPk*@PVB5ST;@Q!*?kNYboV*jg
zyZuMsdl^H^zAP>!@AJkge`SfL7pHZ}(w|d9;I$I1w%vaKM8}YM5%&DCZfa|sY-hoo
zkW+uheUbqpRG@iwW#31lG@CYy1=(Y@+F|mH;xT(nR!Iyw3?bYoOBuYL?T##uusXY>
zOK7~MNMHa5i)HX2{<%%aXk1@E;`Z6c+}_M+N#SZP?8cF%e%|SLWNsCkZ<uc0_i}XF
zb^MfJf4#VQ&8lnRWlwZdt0Wu{)+y?Vi}!ms4jiXOmdr4W%=4Y^Gv5YFc10<f7L_h$
zRlaB6<&ix?Rhg_h%ZkUynXU6|Q}tvP)n$6Xx~tTUUH5X=#b_5v?Dq~VyYZg=#*rwb
z<keCj(ZoY$#C8boJ#r!ZMx7?qpyfm=1D3?Io>5YOx1pp(oj}!)>~Fk;&L6O-Af%K+
z1_+Y6``G7SK}nt{WMMWYT(S(IHDRg_=5%V|E-s{Dprpcanmpl1GQ&8MXxMZ|tCgFB
zF%)6H_l)vVii?vk>&NHtMg9?NZ{7vGxCS<s`IhZ_jn+ZRd0Q5!GjMkT{^Vi4@cuIO
zf<<VH<Ti}n3B4D8cj(s_qs|5z&KA1*x-mE9?l^ICIP&+O{$2jTvtQW6Vc6mlcL1gC
zW>{kQKsK6r%S*gj98x51^n(bl)6^L}j|tgGzAQ{z#IuFYSB|oA%X!5?Tno2Ik2mVV
z%00fs%fNA}ABZ!aCvk%*ofrfun1$}XUoc0&eJ@=UW$JniQ)gz?>sik8Og&E6St%jU
zqICDg@puF?{;B>t3E@x6pW)r%U3brDI?*{4WRcuBH829$_G|1+W(R-^@LR`5XGUK%
z)LvUmniq(vb84t8fX>c7+rEr!z~HHGn+2Vp_VT5DvsY}fC;@C=$xlSpNyp-+Abblb
z`#bk}p#aSy{X<rRXR)6xDnlfaHJ6*RhP4{8W;tX#poSS`QAgWa1u0R^EFnShc=CuR
zPt_(2Wn>(74B0jL>I`*0IJ*`VJ49H*LWUwdV}|66)#2`z>2RRz3o=~u=<yRIXNqO=
zP^j~S9q-Wj7SxkJa><l3G7ck>1|*kY<lCbCE=gPkK)YN^rh7rODR~&!?{}^(Xo99=
zXaOG4+54NuYlmUxVSeNhO2Lj5YpXt~Yuov;Thl0;XG=7W&yThkx_-RIJXZvd1@tUk
zpMykF=r^|2jFm3i6X8Fas8W1Rg2i}JRTrer+2;atxt35jjd;6E4_*(FG5)i-o|VlM
z@&4tMXw72*vxLaAWvZo+_UfJ{iNwXkq={DDb?vczg`*0y+F~nu{9`63VKSl4E*o?h
zycV^J$EXg&uz0yHUf|EBeFiarK6z~yoNWZq>pticX5%I8KKP4IH@htXu<K5*re->-
z>&HnOphsMXtzX;e37MKAB~t~SA8vX7!;ks(+h68a4u8+%FU&=a=SYX|ZUVI^QeMiU
z)#qR<wDeL!TMKr;wWff`hFxE9$qB2CX%5+djrg;#^NrR*F+K{*x0)ggH{#ji<*ePR
zwH^bc%Y^0=cOj)=x!1ilT|twXVG984Y{GgU7I$_42Cqc2x`ZWXh$<d!uC%!_Pe<xJ
zJNSD5Oqpu9GCZNA)y8i;{f8vrUw`~>@ul%yG#3ZHc_X4)+*0py5X)=|OFY{G|2G<S
z%jgSb9-#xEPoX_9W?wsd?rfU{v^@k0KgHJ1(?dm1Y^xW>4Za3S+*n+v@Ac<n6<G{A
z@x?tCH;7U&bh)_Hh_-k~T8pHG?Am+g(7<{EmNo>h0h}dM(%|LJQUFbBG_B0@(chC&
z$pU4!gUG&bEu@x|h2dH(U4akNN8H}s@X-e!xg#+drW21Iy~FkOo@uUL+uPij$|$wn
zdJJDkQ`I^#9iEfgk(@`;_!t?k$XP&JFpdND@C>Zdbx{G_O+27^$B;qwIi{JaH4?dI
zXtyLADZ@y~g4GGFPf6^_snFCg2@N|~uGvetwIyE7I>L(cLQk^j(lMSuHw+^=mF1qP
zbuUjCwyhhqidf><42_z@j|P{|$v_@5V#1728>kd=G72s9{PJSZN!Je)q04d!9v9hS
z-5__@a2Vo|&?czId1@~<&<O*hnxIXXL&$kI!fsy>2A)2z+|5c(9(SEnWz3Oem@<$s
zRi;^y!x70T#IiY=OE2xsx6h~&$hzn)3n;8F4`*|=oDG64mg1E8ICmwVe7^a8Ava#I
zK)!IT=nIvsF1`5ptZQ=C&vg+s@ocZDR&MWZ`Nj8tfsa0Z%CAoU)$)=muES^eL{qgb
zFJlSt*WI*%qI{A|iIOvO_1GCFz9RV~ZK^u7wg|@d`>>W{3x-NvKsEGF0Z(RF;tjJb
zAh7SZND`>CZ$m4t*A;Ruoe?A4%-cA{exFYC0s)KiP}_TKVYT+-x{D<iqgv5cnWs6-
z_{{?i7mW#MZ7X08VcS|8|M=-Yq|S{m@4n3M-~Bva+<l2M7Qf#Jh>^?~EPGj|e#iPF
z@BirfT-I2~1v`5uKCSQ;Jsry({M{Shw*B>WA>B5+cm=z!x$8^MbCLmUr$*FkwYZ_S
z0g#<J?+@f9DJDOb&GqOCJyy2gKhG$2aN*KP@h-&sIh27?250e&rCb>(BFa#N-LT_o
z_lPl%<U~~WFmeJbDdK2n=EEJ?W@6w4s~DNb)CkLfrU})|)GBjpjAQ1V$A#~HI5AvZ
z@%Zs04D+zN1Z}Fc`N%v!2W{?#E{Vgj^3E3?vm4;%X6EkhhO#?SuHPXIdveL)5imMG
z3h_=voMN3S`RbayyCPRb>kS^#=&ti*2W<Y>bq>qOP!ivp-{a?}cbth!5kDDouSvz`
z&a8G&)nAV_uGZtV9i$b!GRU|a8OAZ{7lWF3onHkLUDOKPo8lE$jD$+2sxUX0nlg&(
zGFmgTWK;^~^)*nsDEKl$Na6k`jNYy%C^(QcEaqNYbYuHI67*(|LzhT0!35Rf2`<S$
z%LRs<nQLNdc=eFhywVGmqd=3WGhB+7!UN{2=;Utv*_7jPLPVJBxdg$by@~dj0KhG1
zI{|R9SoXF0gavqwIsApJ!4~{|N{hjX>+a;&i2ge#DmvvZp5<#P@!<y_^UZI6o4@(|
zuZ38#@2j<vBSk+;ft`?+QceM+VzF$kxz&hI@G=bKoT#<MX}gZw3ii4%vh*VI0Zn^?
z24=1Y74893&w6W7^gX4-Fcx-u*Tm`zj)+S$c$QnlwqzGnwQy7Po8qFV*;z?jb^oMm
z@GVxVf3M-j(a^eDC2TwsuFehB8t#@p-E}6b|6%=6&{2bDJFl@(et(Rw-2P*J^B}x8
zewRP?_)qd>`#lU}@%~jmw`#$Xtpe53;T?oZqX+13T&GLOwaKzjybMmg&Uk(32(JzN
zI^x>6-d8qr1+G3O=)pfHY~3gA+j*b*<5_-R7MRz2S4~)4^yi2Fv{}m^h32ui9sIen
zLJ}|ORkG{It*}=Eyd-kjk@Mim0zko-&By|IDC~#A)$WR`or9m8jj?!OewK`xzpvU|
zg~1vo@v>!RkhLL4mte`!Pc!3q=P`h3xZ-<HXUKcrdG||{T&PVc$t6ZwotUTRwCOn|
z!K15z*4&+VoC;GDEbmahkJbY=J*AqG(xYXbF<4lC5wF)<#N&b$2da6zq8&$4o!4qN
z76~DGK=lgAS4;!U4YH_Lu8{2BmYm5Zw7EvMvgb(8O<kTvjIj($B8RhoP_j^R%)7$k
zmIpVS>s(#k0Dlh>!XU=J2(_fJXA?~lvl%6I_S}V}p&-c(fy6*gp3^c8V*vijk`*qt
z5DB?J&TR<`vXD)ZrD(nPrHAq4l$_iZ!AR8VN!BU?)|jf#fOk`+s&CU>&Q%-7Ss@E0
z2}r>*Sn{=Eo+`J;#_e>ZsWRqr0f3)T`)n9MWr@?f7X!6fEMFPGe%4me%KO0ap@sBC
zM8;}UpB|$hMWAQTZ>+0z!RnNH&WRSF%Co)9gPT+0W^R1@d++hf-~anedG_qQ5Vlq`
z%7TJ78NMB0Y?vK03;jFAMhn-l$4~To*yAX=yRm1@p2UX*Zt4NIYZ<pLna~!XREn1*
zs&n<Exz0w2xQi+KjoIzi`lMUet;@M+Bz5Rgr?fMevRrF`KW)_7mRQJycnoE@u^=iH
zl>u#yxmKp*k==gJ;qDHisCd672^Y>Po8tY=Idl!J_fz?2$KOCP{;2*5zVzt#Fz$CA
zBczJiOv6Q{OHK@Ba3+F2f!eaTYz372vs0|2m$NbX(7@bg&#%9)9&@|Dn>+vn&yU(c
zejLm;pRZQS**TVRw03oY*73JKcnuF;*2=~&6!~(MaTzav@tyQv8NWb}Ja|Js2Th_D
zt)ZLDc@M--IlI1Qe}^R@#o4Ca)nif_2m+px@%LfK3|V-*-*dI!F=inT77`_i-Qd^<
zyvm1X?*sICB$0AKP039V3q7|e;g~Sykd%5nddSLK?|hN*>WVsh88VO8Z5-~NQRf39
zFph=ms}VIATjMx2nhqqHdGcsvo*MJv8KoXM);sce?cS570RSW8Uxc|%WD;fcI|vC<
z0rBit44-$*^#VP%uW>-R$?(LAgSR2m3~Fnu7N@!Cp2X!}j>iKc=Cyj$4pJf0Wo=lJ
z-YhIB5nJcjsLGV^a33*a%!%s}YRlZsDYO79$JWRg1&<{+ZIBv?hTtJRrHqsULmp`2
z(lNc<Wv{T2qN-9&g3KM{d3Z~Y%WgVF44=n$=aSKs5L1t16rrgxH=w~$4V6aKgh-~O
z1kJ}$k7;Zc0{!CSchT|`p{j>TVMdb+`1`EdX955<7L^{y0xWJaAX4{bT^wFAZ=bk-
z<99g&H}~7J%&iM#P}$vseCje@&(PVAl{i@)OljHJQ{+?19vrL&C80E?+OTAN?}HC{
z|Kq3pX88y1T1a6jMroyfkG9#*i<?opEJ`F`RFC!OS*;qSpD<3OV~jPWUIUQ0%*iSm
z>u%%D1VWbt=<cTOh8BDh?CZ6Qg9eC_empN%wIrbQn-<V(Vf$D?R}1Ygkdqj{oZa0j
zk>%L5E%&AiU#FC)(@d+CB#GT_$MLoczot0eDxtGw9{Yoqh52-rPR;o3`76|$#_!di
z<)`iEcz5?@B)Jw=3wec{T@!1ULSGA6SSi;v^ebKZj4p!O8!s`ZK<|`6I{|ZB5cZ<>
z8W`-{O{~kY<8xi0-RU81?a#t`$3NTY`juVSFw=3clXIMJy6A#!48kQAiKq_Ix9Q%5
zWrNR$oP>$EOo=swf?7pXNQ^F#A_=eS>+xU1nAl%mv)ey*-O;#lIk~GYAzY0EZ(Z-$
zj{~H%WC!QOrd(i805i<m!VYbLdY6kTQ5gaerfK4XtEY^+J?PArzWgP|{S&Gx){baB
z(q;`{+uzX~GAnpRjn)_fm^%ADl8a$&qRt1HCuF*%3|B}>uApFTH7Hf7^NilzP_9b!
zIY*+IE)|qSGO3L2;x*;S%-63z;13;sj>o&=_kEhd9IW}Yoe(Ls0Gx4CXb~TsSd-f{
zM4<c2tF^idbO~DkixQtzTX;>xV=jxw&DCbEbW<`l*gQePaCXTYta)5|xU7wY>KyUp
zA+o=CS-igV5|6xjw=7DtHS_wuB3=eiijRXOvm+S4xp@^C3#O8A?I!`xZO$G)Z4IQd
zhLrf6BPo-|fzpJlNFLmcBg44+RNtXXd$@f@F5olL4^HWXm)QIZ$lEZ7=ZxR07=!hE
z{qNUgE8_T_6!EpC_viEsCF0w1fV<i%lT{kZG&Mf>=pz)Nlp!W_??tbpJ5OHuqXR;4
zcW+AK7Cy!_PaKa&Og+P`lwcX>nbaa%&AOGjsaO#c##&#|U&4i*>yt$COyX^Mv~xH8
z5<4hFfUqv->P5U|eNZ75tHBVgZh?hyv;a&ZG;MBe=H$sN;uA~L>feW(d!A>KBwCv}
zPDfg8JbHA^TwOjwa*9PWod5$jbbUcwt|KKcSRNUKe&;h^=c|0}=6CqH;~(S?UHx(1
z9lykG7#Y*xLaqYG)@X+VxfI5nDLJhMLf{<N)N|k`!eUk1smR_6uH<B>p0IQ$zrUcp
zk^$^}J3agIJkIIx?Vml~_N05S=+8qvFnFmd(<(l_8FwA*vAI7R#&ZSSn_hkL&izO>
z2VNPGybzr#k!2YIzD+$e#FCIxVHkHIJ;fsaoy45gN-&m0Nhu1Zc$Si24*t54NF<Kf
z8qeB|CjZqF#FPM4X5^&du1=J3$A13>QW=@1nOcw3>M>r!STGV-b`4gk(=BF=yJM7#
zjQF;kuhF!}@(zX>Za;u9a+xRhQe=N8B_*19Je{OM$s<*v&Xb3xqyfa|Q+ga=9!PoT
z0`!pt^WEKd`FVbp+RP`Q1O?YaY@{W*E|tV?=Zu<a1#$37wJ^SY{5BD)LaIiR*^@GI
z@wq|3<!l;Lt;{XaT8=y(2Vp&Ir8Hs24cZd%-71NxHfpOrxwJ%LM3Xa?s!FXhgAItd
z>$#VfoagGHLl*Jk$=N#<@wmri<mPji$4>g(rp6M3?3#XkelxE>Eg)2KGKS=F)ESRe
zPog1HRD%t4!(kjl#e(DjaQKYU6Mn+mXMUHv1#r)8+pYN6=A7v}PwH6qgN40%`B_gl
zc*Qz-e}AV5H>}_3fy8|BNhzTM^EA_N7w`ukJ>{d1KH{7DyR_uB51VeXs-lPM<my+S
zLtC%b$}~+J4@cTGBasM^I>3PiJXvS?0)4EkQQY@<k!w`R5P@MB+({Bve6Iwuh|NZv
zUl&AP^sggBxJ$~cw`64qIzUe;Z=RPfu4X|NeZxHFg#}#3f_X`*?pm+awZ2kLOc5}=
zzrNz?>WYs){uom7iU-J2xnyyT;cl*!)26C`Tg1JLR0q$!4_}>si>X!qmGVb;r+kU4
z-JYxc6+78EBPd3p2J?ptN{cK7H<s!kMm!G(e|-`@8Mqw~zhI#Hcdr<@uljle{<>kb
z7)a-qT5nE|`ViVMm#5v_g02^L!fZLz3!B8dU~$5ezwaG>GiSs_^}5KP9|V}_>U;n$
z1GV0ZS{u`RbYrL)xeTrjz&fr4cL9$g>{4b(LJmX7)6{$IkM*33U|g`{{d|V&wYjEe
zCZ`sm9WFNiz5P?lqrJ1_O_}eG)On)SgBzPtKn7BRaU3wa<|trVT?e!Tpu&^eNR^aE
z?_(;EN2FEqFp|f|{$Az5wE^clL?)j%%rkgvox=dlNE%R`Nohn<CJk3U$F=HgX-OP7
z@>TmC9!-CMvU3oZdPoW4GC7tgMcCn?AagyqsQ+9cX&ClC0w-;>W)UkmkaKeoDm}(f
z7WcC23o1ud4zr=G9tMr=kro~rkIkE#OJcQJnH3HgS~3M<?U+%85@AZIA1}o6iE)xr
zqKq!hgE$+PM9><EBGnTVn;9OW*Gs{rcu!q-Rm{Tm>qt@X@^85qVlCL$jeL<rjA6XL
zDs1A%FYQgXH!lRuikEs!QMWF8>z1W2M%|6GVfJ1`@bv}jHP^&X#QeReb=U0Xm<Ay+
z60v+{sJL+F;pvfYeCOMI`0+>l!|VSNe@BA-%Pw2sb#NoG0+xVkBwpa-#C&&%ef$OO
z(gZ!j*?JM!)2^<r0$372wywUMa-2KGIoBAClvd$sZOSxH8>ULag3fN)rImr~i_StQ
zg#xgdc_>awSkeUquj4`JwS$R1cruE+ylZt~RuKu&Y+=kJ`~9AA938;!Ms~x%JkO~5
z1fy2NT|I<kM(7*%SWDv8Ev`^lREtIUhW>NrL*?h~5A*ijJ6vB)Twh&twcE2>EY&&O
zR*6j74Ek-@L1gJ|e5RcdJqJ#w2FiV=@CR>p@7N~>RCi0=@5K6fFuM=FQD!fEUAM=0
z(N)NHBUq}EtQ&0m%KH8%SzPb79O-kiimU^aE-wPf$FsP5^LRLLxVu9r3_E9>ScK5y
zT^D6ciL0Sd2A7P|?6HKldG78a?C-FjOU|bkS+zs`%S0abeE;!Neye<wV{Pa(Gf(dF
z4Sj<J&^N6v*%U@)8TOber9{SUn-KG){#s|OKKK5aAZdq?k+P>;zk`S;CCF~au$yUj
zA7ky#vy#o(Ky4MRQy97#QYPg+DUZ&8wuX$wD~y=&o&ERu{rdB~)yy-ab1#eMjF)Be
zd>dIqRy+j8-|^hpKZR+UJXX6zg4|LaC?UmN_yu7<Dw8T!p_ylM4Es!0C{-CuDJk5;
zgL_KUAd9<y{W-eSm)1D|pzHl5frcCMDY+hGPaIt1xx1U#!<VWmttF<a5eK<SlQoTM
z9^LrY9-r;wQ=5^puVp2N5V_iF%msUB5o*+0r8bv_NmV&c6&T*wD|Kn!K3fKGF<SeA
zxPrWM)~~nM&J29%WWO?iKg{}h(i2-h_4Uw}_!F9IKUGyee)g1aedpVJbUUFtjTfe~
z8~6Lu+0G8S>RcW8G$ri@hH+egoSGrM)K8y`yK6O$wrGg;g+o@+O<-a9gmoeyt>VWE
z6PUg4E#zrjv~=ZJHn3rnz!JAAz+kO4`nhhe5v1Hgx<R$#G6N3!=DDI8b(F)Ejk%!e
z<42D?qb*|ma&cf>a`t#dA?l+BV9jp%*4miUy!3O)w2r=N$kME1O87Qkqt?ckr=Q`;
z{0{FPzs%!Dk9cx@#ntF+V9Cj49;6rOi;)+VsswSD)#kY`EZi~<UJUkL2AtMyUS$p0
z$w2P0k<RRG$4<vun~qxWL}x_Y%Kq8O=Pzn6d!su538YU)c}r&X1}HC?>n6ScWrE=p
z?A3ST$E#TgeMGI_B(AraOF?aBZZp#~lZ#_>lZchoKK@~s6T6&9MJS^uFqK?z`MK;h
zA0%RfH?<zOda7Nc6BH9Alfw-=^4+VC`3Li_ad&g@syy1T=F%}ab$OJCkxm6qo^#F9
zl)dJxq=KaZOlbA!#vdt&0oOWJaksAwJ5ni>@fw?(&tb!eq<3h?E6iNNMq(b24xnR<
zlu{^Vpw^kD6=X*l2272~Cce4*7LO1A5~9vZ%^JFg=3>xG0$q&Jlrm|IGI^;WxQ?G|
zq&oO(c;%5~Lhf^@2t_j2yOCLEp3NYdLfAU)Pby@c{XR+$WAUf5+l}EJNoaK))L9!X
z47)N6t}<b^@#@$j_D{@#JubafH@fF)+)mA}J2(R&IoV)6o*JLCqvpHo3OYkQHNV=<
z%xCj#X?KAc4e=O4_4_$Z6UW(W8#_3?0Khlh{>8X}qq()zp~|MrR{uW!y5}xF`~6i@
z{SPm`ySnuH@z$RCZz&1Vvr3b{MDKt2A>aAl_jrDD!`<u_bB?-`L_);^yoj|qyW7GR
zP$XV9DVN1<i4g_7J3y8t31rE94WO69r8`fP#|{>^{H^3Aqjve%JF`^v=$Uc!KBiuF
zzPpw{lP=rz?Q+h+P&zc;W*Vz1tvaZ6w?hs?0H$NGoFX{ptuBI{v-exv-8BpYIVHx?
zi>l|8n5T(x9AiR?%(~nI;L}pcL#>OO`u$BB;?Dg^N(YNR*>5Mi@0D-zUHv*=(x2rI
z9{(t2U-EKQ*Nv;)h}nRoQ2t+Bmf<Gp$JcX!w7P=NU8@hVfUmA@@rrk0`}+pTEy3Aa
zmvjGT3sALlYiI9NEH(>6_`(j@tGc)Xj-Y~72!Y~0Qac#zPD=~K(+{v3n4gEXBv=Jg
z=Q^8-oC~>R*QUe4c`n7l$EFlvpSwzus2>|f-{Kk`Xruty)#a5UF-u5`<dcIz4R%mh
zMmuM|v-^Nwo_>v+!>zwKG_=huNHRwzXv8ZfGb9aBe>M!U92G56BFjLVXKI@<iLz)h
zFL{i#k;Xml5_NZMohNi|Za3tNDCA+SQN9=z<~n;4K(>$j);uOI=S&`Vd_%v@&&n5h
zJ6)q{%(DZlZFo(K`Aoyr=~63^ASVTFGeged?GdJGo{eiB7iZ0wsf(^)NszN3)t{Rr
zL0S*7iMqX=5snu$;qzKf5-elFERn)s?wW~NQWSRc1dR^hL^5R0>NWLZ1aqxS&8QWW
z0VI=3+`=dpAhTb4n;TMnJ+ib`4N;4FP3mRhMr{n4Ff=?7rL}qS?({VUUO7cv+Nab$
zBLKMC<hGfc%`4USxjYD^u=uwSt1k$$J>K%`V%5B;{m_F+AcE&kSPoC|kCdl)A$zXs
z&E0{of8!f`{Ttt)YU6hIG#0Lis|XFfB;rcw;u>1L*n5ORyJZICB2o6O>o)cE(I=cG
z2GG{BY~AwADPy^yw3Eer)#6(<>JVA0j93VS#&$WawJ!TYw6c{wc$)x@5tkB^YOe$A
zeb?BxyJH=Y?f3im-p`vgL(!1Bi1dtLFH0%Zd5#*LQHp3BNGWkR93TmK7#2{R10ci7
zP%gESO=+#inms&T7ULud-{%{?xE}uqwSIwVI`a6@71#ScY9luI!^z2cz$5|@6LfO)
z4ePfBe|ECT86bNVsA(_%+R_;S3-+%wd~5CQ^Q+%BxqpvrvnIWwy=<G6^INy}Wjf+-
zJKwU*7c8vK#v=Uul!2qq14a_pp6ilMB7~Q~1<=PbvcG!FZrCHiusJ{oH)Vi7A)yVL
zxSP&LerNNu=T2E&)dAxCJ?_vGwtRu#UtRJ2#~<)p{4>TXlwm|$<8U`~cdUzcVe92z
z@aa<3YxFLRl(x|2bisE3NkcH2KF{RhnYt38Cu!JYy*7GsvFzjB<XU-vNeVY*&oY*%
z;$Z=B9q>q~WY=k?jT&ASVb+;%PT%7C?q9}GT6MLDPTU!8AkVD<-HT!o!V|J9e#;#s
znxZP0WQK8vsjCX8HA*kzDb8kVb$5A=`yJLO?ufRfItSsg=<{|eCC+PDFl{|4C#OPM
z1e?}6@P0p5R4D=_J6k)HL=msvp=n)bMIl$0rIF;-l9E6Qpx&A*LP!`OIV(AfA1`9e
ztv04<UI>Je3nt0k-=7!j?MJ-5c>uT!u#Mz|bh$L2)IA5S$NzOTb774Ke84&UDwctD
zq1hNqym0w>(W(^KuCrr3_DMvLaP9Rh){4>uW>Y?V_KdH8_dC4z>?!-}$9(hMUx)7U
zfEE0S--FJY3GOPL>%3IQ00(s%i4$o9+AJn9L{RZqkDegXrDRe{?Du<Wt)U^<7QpS-
z5eq|i>25);tl7g`Hn6YK8tlZv=2^rvhGtm}cY>KS5G5CK83Uv`K~U@LQU|RrPfO_8
zxx)e+bIw@KOtVYhlwn{fg}cKY$HS2)Z#|)F-1!2T7e>?<fPU|Zm=w<hRPja0M3xeI
zZoBqHapUSSUohX}8`ReL>FMWrJU`}(^~(-)A1zA(IVo8ZLyk%euGFy@0^l(_%Qy|A
z0ND^0oPoesBAZu0u=B5#6#&A0@VCK>>t?}iK45g6fWZ4ZE+5MSKYbqXztd*prKr91
zjOq<YTp!cvwVi|QALG{9sw7cZGj5K*R}=Hm%RBC`uDIG?GYorYShU4{zNqKi8GHva
zO4Ra|00l8OOU!frd>dwom}1;mij3nYZ@oqRy^Y_V|2bnT6x(B(X|<ANMqA}LMF@)|
zcKiL3(2%ykBa*_*8aWNF+bMZ*XG@eVOX`(;da{U_XPs(u_jn>|cH8W6XGt*4%*W=y
zggrVNs4aEQA>!D~Lh8xGPGs;38{gW0#LpZjhFLr_mq?yTECTp_j?Mu%9dtGy&td@{
z8)(AF3?Xn&=mECYn1LY-Fw70=rp~q{9K3fmQwOfL2>JK9Jc)&TmOo|H1kxak_auG(
zI(;)zP9EQAfQ6uUUw=zR^5B(AED9}|kE50oc%?ySabvOc2tyCWDGA1GUMj8_V@Y(!
zOq*NRnFl21Ul5$xzO*;r-aG(I$xFh(!fHX_s14hqxW!~hEcg;Hc}##J5j(Xm#0#TC
zU%^>ijps=wi}?F1)`Y%pbhm7GO)p}~sk=lh7N>ARr9?GFa_0H78@~GOZ}Z;s8|3N{
zIu7xYg&=YZ9V&1!PlT&f2;(&Luv`q;kTp&tCyz%7*G<Z+JGKjamd=(1*mbc{7iTIV
z8eLllWxdw0&mc=8NB{05#^r&<zB3WQ?|PB*(A|@svFNcjse{It+@c1OPbMOaL&0>d
zV_d8EL2_QC1w8R1F${&03#Al})6u1I#3%AvXKJmO_;KqzI{+Mpr4&-PKs%GQJma1T
z96h(Nw3mp=1vvUb^g;R-ALMWGC4L`r{5O0tlZ;3qNg-nlCA*H6ISZ0A&^%$t6I2>C
zO56=YM(f$F)@^2{|NSbN)Fou=MdO>Z_n;rpSml4TziY9VExbYOzd68JA5W@_uqFe!
z3saU;n#4Wb{OpEeKYDPeaV+1k8wB76*m`Q|JH>na`2c{vXnQPaP(pYt7R;?O*G4J>
z``tDB;Tn^IY7JhmKS5`qp~T!P9*Y*7l4`J27$l<zYsdy|Ve)zs$CK+T?DtOm^7P9b
z<~z(b?+{Q*VkpAl5b=b`UDzEYcztoN2I7HrGuD(U6UiERctjd@?1qunDje_7HZLYw
z)k;a3in}+B7)L2HA8*LJE5B;VjQeX3fe}GaYznqfA{3<CV~T6q6_{(oTH{s^{7(8^
z{^0yGbREHNWG>>{$TAn2`Fhu^u`99|`4)oOlICC?U6IP7+EOzwH`&y@tY<eC%}BVH
zC(-4qT7(F#Gl3iW5&+AuZO{bPOINZObG~Nzi<5fXx>>y2Jt@xT-XTVjA`C!Np*3M@
z8r4<=%Pt?&SBa{IWhDVa^l2=@I2a|Tu!MaKx}K^hru1=WmR~-JUD~JC-Z%i1m>Bxp
z+_hP|m2Y^w6iGrcy@QHyV{R6wllCA}@q<nHTNVN~;O}e@u8c)Ys^QYGV8V9e$kXGI
zZ++*xeE<FTIaWnV=B7MDN+Rh>x488PIRM`Ac#JrX6iFI8B=v&hYlhXrZbiVZh4rbb
z9@HF?E-f+un>9?29td1AsW>B5ta!$4FkZk4`Zjl^OTa*|eX0xV*ON&~_T&o<CNu@R
zrQu>&pD8E(?_fg}Qt0<}*VZ_$&w{uf&s-d0pT^;E;5Z!t*zd0vxtykraU2$n#guZy
z4OLQNBXg4$cR9A|Aug9MYj{F`{v8xqWH4j!o_rGo{?Pm<(2pGmzx&qPJi5N7sd967
zU@V2*kb^~aho&zAE|j_;neE9=Mjk60_bJSctNWDGUzQ)X`O{86JMmGUy>qs9bd!1&
zP$ZH~Ht*G<E)qZWtY2L+!js&ECPEI%WBaH+*Ub!ThGip3LDGOop=o#lWT507yZs}^
z@tTIjBATI9Hy{BeF_{Ceg!tU75sspOCYsIer4V;HuEtLf_;;h`=G|xf(#OBb&GC68
zCipwvK?GsQyFsR;T-<<@63<PM2uE1Qlg}^hP+hyP>|F9n3Z*1WN`#&$W|exp@wm2>
zKuUZLG-W#6vVZgz^XwK?5{bC!gbaK4K7?R%uQ5({pC+G^jRD2@_U?Q9)buV-=C^3w
zr;9Bav9KD$Y{-Xou4tPKStF-aD-`JxSzf|!s>U=|%zSNd&3iX`5}BM72@gdV++X9p
z3q#5jNoWF*0N+7qY_t6rL(dC9Lv5twV`P!Al2t~m_g)2#W{4<7@P6fjS=Y7=qobr%
z4klZ;#-UkV%x+UiSr{cFWuO5zD{Z5%eQ9sDy?F+3vp^(pkx9-b?Enb=VJynAO<$Z(
zCyIXP2LRtEwEDMMu>J2-@V7}0SWqm%UIqd%P?RBMo;`iW*T3-%zW@IFE(<awzV_r7
zo$ZO8{q>F5oa~ux(>!4_OD3^fvpr7D>pu4KZh*6@23@Yi-)DbKQ_8Dum35Kjid9=G
zcl6Im>kHq(31Ll&gcx%QOIOWfZrYmA(WO~ZXl`xeH9MH?pXZd^4P7hqJclM#fU%sH
z2~FD?>)2zvB((cF*qx`DX`X0JBQ{YYxx}-HuXa~{r`5caU9FVd0bwtR)UQ((W~;l|
zJCk){1jHEZ0Jz7m3c~mJCWi5c_1{Ep9U#`)cyx8;m0sqFs)<MAfRGkR35z>!TG6SI
z7h1e8CjHbMJiU5Mzs@Ug8aKNEn43jn0e|)a<}bEAcPXtE2F{M(U!+g)j(Qb^9&6Df
z0qx}b&pQB=J{dxIM6zdv<~_}fG3_a3pyWM*kMUC6=*dzThXLnwGNfIE*%-A2xHct<
zvy#+Mad&JNH%{Sh^|7A#aK7R1e(Ucsap&=R<_6h*92v*5WW3K&vQkya1-d~gT~s+_
zrM%itBx(e}({{w>TO?&#^6On=)^HZhjC%W#$0-kwJT9;dFkVxSHyrM6DZ@akUd1El
z#5~R9RImgk*u`Yk!5{-Ed16_Yk5X;aZ2aTL|CB#|_#<3TPyF8X8sq?zKy1IAJ&ZL$
zqHhi1ZmJZ)D>rEiH)$`AX0GfI4_HW;s7uVK>*a+scer=mur})nI8l-_#z@LJ!h{Sd
zL9V2PX<KwwEd{gJfgOZ=jIT8oRH#c;X~a8*G+6vk(O8Q->pBaI#Ei$7dWLh-glQ#&
z&H^m^I4UuoQUWZyo}pxB18F_{=F&ck_8I5^K0G~d=CqFc?tLBycyK}5E-kYC?sS(k
zBIV(QN6tX$BF5Y&8#ave<`9A`Y(^3|9FKhe``_oi_rAyB?tl%&1&m?iK3qZEFbp9G
z+TFfN?MV?aF-jy7pt{7I0E^z9h-ZGKR6J8{nvu~1YCE0pZr|z?)5bNot}5&CY^m&#
z!gUt{vC@U2GrMr-E}6kT<8Ry`79eho4+MC0eH{rLgP-5r)jLfSt@?zrNYBtg3iN>G
z8po}zE^g`)?(zOPW(=h;j$>2|@x^eSCI^7@=M`n_79GH@=&zpH+qwv@oWwzHS>~mj
zR4h#2lW$^T{Gs+Ic>X+dxVz&EPu}94FTBle7?x5|eL>evJR3M@T-dR5@YH|X#=QNY
zO;(`S-LsvzJAJoK!QV+c1AiT}u3g{0g`6JvB^Pi4FoSF{dPXa_sRe06|9$R3pvC)!
zXO@;!$ay4}-SP}cg{C=TwHo6n3?);tvXDJ_#AI9m*IbW_5U^nk89Xj6SSAy|b>skA
zo88sd61Pt$o)5Pq$x)KhnfR;y!1dL@)o#BWyCWn|{u0)=$g^4{=7MFL7&{gq2=&Sl
zX3AVATu8lmVpT3!ok$F5%anmg{<`uwzc6O!$^E;%M!VNChFn}=x++WPQD;m&p#jj$
zG2QX&>7VjP`9nOJpI~MT2`ndOPkY^jAF-4KZ9*i#`Eg556oG7pn1)ej>s%+oSTbW%
zW;LKWb0%Ic4lH8E%Muy`EG0f$lyU2zoem<C#9ZmSR9)Fix^ZveX^Oynmwj<}v844K
z^8_rzGU4I#=a{6Bl29bkRPf-yH59?3UpZw=%#B@G5|(@o*oNU;+M8~lK^Cw%!>7Si
zoh%CBV#ATYn-sQKdMBXH7QSuy+rRB>!#b^WX_@nN>wH*swy*zw+OJOaZvd(-to_M?
zBa%oZ@zMM5^S$@J$L;N%tI{JJ?5TqaLUS#yx#Y|+4u~YTlr>`BnxD5bV5#^cN*3om
z>8C|@S6U`5GuIC6ae+A!GqwpH{lDcXeQ^mUZhL<$CN_;)yt$M`_<0o$T_t03j&g6s
zkH40mV{yl51Tbs=FZ$x{=e7XGo@JdPG+;hX5tq1Hh|Y#Y3`_Gwl+tT$s=KMjGEhtP
zQf(SZNX26a!7cVA0zIQRw8@qUQiK#$URb6xzyBTPI5ttftKZ<g^ex`8FY#ZS{!Qjq
zX{LPX-FJDs-!sjw<s~_o<`fxv^jM*_=JENMeVu9Xwb;Lhj=6bdwuaX0Fn=BVy`D(*
zLs`Zz+%$|2*MJPaof}B8Nivosk<V`vO4*UiKpw6r<tp634#rv)nk1S)Ri$cW9E>_C
zkH>~a>}<+|Ck8A+<w~26G^_p`2V>0YLc<Xrq1J*4)5ZNdq|DFM-@`xVLn!`i`kXS&
zE~BD)V4j<U$sWd08)P5z&MLV9C*qpbn;K)wKyHly1r<p!7LOBi@G2k~sY)&rX?KOD
zL>~RUZrYYGkWw-_w@8@FK&ICS7s(4)l(cFNY6nKOk<!4lQ+}O)!XJJ9ukyBh!83Ow
z(X2I(?{2NTdt<&&;s$BRZw;j|h$mTSYaFe|dE@g;5%xo7t`7dJ`3jdr{Xc>sq%*VY
zDCtr`Rih*ufVf!ntm+_C+^uXzB%A<5NK%XRHS38S9uq0zUKZ_hH=b;WxF^dANeAqh
zc-D0+p9w!->dSlb^`)fD7^O_z_1_?p@{)K~^`*V}_U0MDSIzd_1-oHB%q&D+mxK6^
zvqAAtOwx<u0AJRR&iJjr6B860Aj)Dr`LBkTweSGhxi#K<|9#$n|05LhSUfR)=Zn7r
zDoem_j+lwg{#i)7<j7K8&aXas<jz1zPqv7XOBUcSuYe~9!=>8R$(Am&aGzG+f=PGr
z`Wm~^oo+otND1Izca?5XHiy_Sc5w+C7je(3?LpM3vu~<MbI_+<B+qOa7gi(2F#yRT
zLlF04J(tJ%xa>_@7Ob9KJ&Z$?8?xo|?R6S%rKg<9<}S0QE!@4Hz+q;wcx}4b`w~`x
zsh8P%U%tn0<X_~!UjLiO$1dSBzx{;A*L&AAi;4plnrhu8yg<ut)yN9!T$=6WQ0G<C
zrmVMjLo%-S_NCx&>+<3U0e`&*FYEv5ainL6^qVfHV_*ZnKesLC!Z=<t?jDiK2vSf>
zw8@xe(3-rmMZ}RgbeL&Ls6qr*%7T<zi;`faAhpqE&p23~XNppZB4i4!CTHGSgqLLa
zWAFTHJk1~Ry~9V$Q{%~_YnOKkq3CIL4NnIPs~gh|wJ3!KXltY#tf1*)6uCxRLqiCe
zFa(AnueyZRBG%bstEKG0G%AclvRQJ?#MRJ`q=yu-kV(8+h@=qJ-Wow}P)jKecx+-G
zD*yP=SNOxv{vhx24p}<*iE(M3Vcg@Yb1sa#kw?4TqF?ATIk%Vy$EJ`A`<<7tOadjD
zSHDq3w2{TwC1cEhLaWMbicmuTwTChJ{9Ca|gz^NcW(_e#sv+Zmicn4e|LpzSvn<PY
z-iQ50UaGq1T$g=0=YSvx5ClaEG()5$Tb8Ae9FoG8?KeO8!4ZCu_2%$@<QG5K5%z-}
zitvLYte_*ruqjKFNsAOH2_$$2a4!3{E_2TAuFA~ehcPm%daldf2gF{)+1Lr3z1Etu
zySlonGV>eX_{KL_i-CO{><T)vFY3o?*!Qw$!~zT*3!jq%eVu4-`%}&GD)v32dEhn}
zUK#T|1P{uw9`sX&rRtCaeLue4@nJN877C0FWQ>8eHqbTx_s-+J?^o{ze_!`F$u*K0
z8#~|z$lGQ&PCiKY58>sj*L?Qb=e&CT8i15CF&iLhDB_g6idYpYGOA8fOlI-m)-c8Z
zTB&^uoB!Z2Q}zuoNgNT?5CTh}h2W8{0g~$aK|AH12IzpJYaD%X_PO2t`E22iLpRi~
z=l18GS=<ECV4vNASO=&ZTY%|IMoPC4Q>2z^pwZ94Qni-uC_@*xfh`nIK`rEz{A*ZW
z%>J1j9Et>Aly;za&wRD(A{$G+L-!1`P7M7)*aDT{k%17*K)Mzb<%N8~Z-xJyzu5kD
z9$r1LlncTSklQ13$^+0#pBJjccnzXE;%ffzT?6#DjIU{;+3(mk6(0jGXI>tIzxL)F
zdKdWXRn_}|NqPL2pJNP`&(f08X+SpXnD6pPX=a)xQku=%j%c84g;ImteRRo+&2tG6
zk?si^2V=CH&0Z=7tFn?9iE2j0AvK~j(28~+O(GJa*{paV8baiFyy4T+CwwNaX{?mi
zyaF|_EH-~tbrgtzd;huCx3xO_^0bM|C>&6<ZJldv3%(Rft<F<&fDt@}Y89xx&$;zP
zw{8@zRWHIQpuH8EwO_L;WMc$vT3diO+zfh(T0w&WSjm}lJMmx6e}}*H@aK5eJ~hUy
zuYDe5irDAmX(CUPNphr7YvH^uv{R$1pdpc_1FDgf8WI|{6>3@0)u2+&!W4xm21+X|
z)q?66x&+6;THhx;vtqrEaW{zuDG(JiefFSQeIL@+I;Oj_ysfQ4@a|Pjhc#Jqb7u<}
z1a*s#=6{JEa7frHrUf1Sg}bB&4-!;Cq7dum4g3B0_QwbB1oC)nc*;PcZKxjfhi$)*
zN9=@-zg`Q@ue5ECKjUqm|M*FJ1pY>aL@-9O+Z?DzkT~P^)mJb0>g7vL%bBrN=4uY`
zwR(!sBQrlv^TldPnTPv(w6>A*Y+@6$x>2S@vzF?$*F^UMY?NXZz}l25&0mwek8bfM
z?GAiI5tXa-rd=aX<yhI-RY6Da?j)VP+W^E?46<ipL_D%{1lj*xYjdF0(<HX!gy2S_
zer<ac+SA9@s;Lb?7o#oMwH9OPx=)x``Z<I)Ue|Sf_J}k@5)uN<1?u~>dh;~$=C5^~
zE>}#q)5)mupP^^51=aeA+OJ>e=lp-d|DEUZ3IE#q@9@RTMAgPK5vF9C8S%oW;Hmcg
z&>aOhp6&OsZTsKtpY1D_{Q47qp7F-3AN}C1AkqJ#k6u0kwh>wB8war~+Jg-LO8Vw&
zcCpA6GyGi`;$X&NCSk3xR;5&-)_|(GivoUagc_k$n-8s-%W7?n^>T5hs(b$h(sUrC
zh{%dIn`^NNF&~I|qP3a&poE?x^q}?!cb}s24{R-y$Wko$v72oof9q=GpkxOGQ%Zw5
z>nu}uBpAKYtx5nzC~it!YJ;_qllA?}DHEorXvltyYG%trYb~znt;1(DWGgB|gtA!r
zF=flftjo$g8xz+`rLL=iuw(!(1W40?8Vmn*`bYf5SBXiU5k)ELjyA2Yqw0)j<J2w$
zwe|S2Tv%&GLnNk|QoL9#CsY~_r-fx%2e79G@17=HE(_;n9qVI=TYvXvAC78erEAFJ
zu&W6Y$r=e>aMHaLTU*)1RCEr|)>jK`W9$#trtR+gKoA3O7HIb@hE*U}2a=`fU~V-=
zR2*pcd+#Tp9;kn}X7C@~@qqx~ZZq|t7?JVlHdlqvjKLCIH%O;H`jPxE+oq^Zua9$0
z<_(r>O9RN;Uwm~8-$s065g<>I)5$ejDkel1;nnLqe(%RW<<+ZuYBkG>Wj*sd$Nv;v
zp{xa(rG*bE8<?%ibeKsYnw`n{jIM<m#5}6Bq1B9~2iw>h%?&|IEi?yyz2E^gwz5~n
zcCAQsAHD%*gwa{U{05782&ObFTkKOeCU31Pie5!$2DW;{wxw5Vt!AdWzX`UakLXnW
z+C|sHs8~xXaX-h_>cA}0nzc3Z^o`=hF5p%Z#X+)l6VV10)BmjAVkm~b2`<J8dKGm<
zD7AX3dkZMKeT=0z7zW-TgCXn}p%;wwifswOH8_^ytrcF#7yM`Sukf!d{}z%g-4t$a
znUcBPHk7)qZj|d$IKoKHISU310o-FqC;g_3=hrvD4!uS+^_J(|Q%2o{FF4Tbwk*Vb
z)i8q*G|p)tuW>qk#%+nkUEucC6X@DNedGDNevNKHS^udzlQB*;jMsk`Mqh_oYm{bP
z7FrA5pS54xW_C*SU@p*BSOe4&X%)`*E3H)5(_V~~i8E;q<UFBGIiD_QIn(6GZJL;F
zpE*w*3EHd=xj=}GUoOAS^Dlp!vwVj3E*B0iJC)R#q`rolj3R99BgNzq*0Rzny<rQq
zO)e6MurY&GDP5Br8dNuik3wB+ZsXOmtrgQiDzs*Ofv5O{+92mhOpSH10z8S<`J$FG
zF1oE%78Irgb2LxY7!EA$%x^vW`}{`uQ#?6;L_G^@v+Cv!%1%vLmx~o{h=Cx)g2bx5
zV6L)UE}SnHiZ(7wp*ByAYM`{nNiB=9EJ`I1QL36oQl(ke!OyidTMs3gZI-&<qSQuQ
zD~W(IA*s>IvXk^wdQX$SFO+5(AbN%=Mxu8Ms-u7hFEX)UziS~a+Y%_H(hy?Eq&YL?
z0JTxdYF1^n+VhPCHvU+AKR&?t09C+PRKmspJm%-Ffx<T&p0X6=jU3Q8;$ZW<TJzDX
z#@|A>7}-vG|5|O0wG~2`4eX?d_fX<|S$TPPPgQG=bh)flff{!O4yC%?$_B#Bu{}C~
zsEl2HZD=xJm8J>BelIZ@WnUYmn!CGr(ZaGU<UA2m9PQc!i^_1e@#0mpF-CIEh*-yw
zTIv9Qy*1jViF9fno7knAPA~fP5`TSDHCvipWm(rIf-MN@=iFc4n%TAV{>CZT`J|8x
zsMW^0E_STZsJsJ1E7SnIOF$dJMQv^9M?=`mt~>DEyshk+Xk+nXMA6;P%2H{;bUtbe
znRQoi7S1;{GbxwN6udQ>16zp^JZ<ululO7B|Hc2{?tjdx7XrUbeEP(q^VY~IGpZuT
z9q0tv2IhKY_SNXY)tF%R$}14`=Hz>G==L%dA?Q1*fKBkVy|y2p&g*VquZL>y9R6bI
z!Os3yUndEnGZv4&#CK>2U1RFq7Iv(32tunnaPITc%zKqm0;O1ra9scm);Yt$RPV}Q
zlDfcJDrH%T(}5<5l#j%isA@X6IVHMi{|D0(e)G%U<je5Fyil!TV0%JUM+LDJX!Sxh
zlRDV>Tbid;XYc%bNQk8Q8N`XpvQS=s30_d5^mJ!K*JU&sSR_M!Ld-|YIFto5#NMSc
ztIfrnwv2>14~W}RL5`8u3hT1cvpFGHm!4WpvZ2kiY?rw3cc1-zZl2HFoS*XK?m6u)
z5gsyStyFC+r4ni6m^_tl*%fI9kg8fQ+*J+Kt;N&Um~~GrW)LkQ(pzdO%33S+d@)V2
z1KwJT+v_y)6lhUV%+h78g04dLs@s?vjmzdR(OKapoR^hlU1<cOtO%ZJ#?HfUn_XAM
zvrJyF)hrOMF4w4%tV%s*T1*g(bu8*@u6734zUK%)zaQWF_|^enbqxjbn1JZoGYrfG
z-n}`EwBs%vwCxylkD}|hNPnw`pc@17IK6Ara8DcZrmjh3R7QDTh1!(Y_YZvb#h2Va
zJn(Qnb2?wBL1|MRBIwR)1ld4l^fYIwm9-RdPE@(pahN#S(iv?%yKvdQ3R5DdWX-=-
zS=WVi?G?Ej2n()Bi4O8|N>(JHmLixY)2WzNu6hsOty*hiZpB)f88YyVsQ=y;0W)e&
zTT45SBJp089ne`-ULH~MvJ~^<iji_Uk1AX%>ez}-gl!S-BJ18xX@JoV07VGSGFX9$
z`<Mk8+P~mN<Ncc3CJ>?PTUw7s1ES5V?&1K4+XJ>eJAHE!8h9UOf>MiZkgwYd{>u6P
z!qeqB|5N>6a6BBCVy2ZsI7VuDiw9PJE|8rRqCeNW@fi2)fBy|JQS<p?bHcu#=%xv@
zAIUWn_edWokN(`4QyncJ-<-?)_xHKG3(_9{`z_Lqaf%&8d`I?>m=Y2i+RU(7ef<fd
z5KVVgOGCs=fJ1i$2muurXKA+`vVt**zR5~Lw08u-fn%W5z@-GLMnocEI+B8nKDlLG
zNUic0{^Y;JfAiVj;Y(g`EgOBFYHO%;LI36`pg=3?Ne!313^o&TzGZsyDf82(NC<rS
zlOH?iWoUSNnpF{+7MBN@6-gD-xLGhIV{J$|LsiPMG97M5!Ac00-3bAv6uob_Wh;uR
z1#oKqbyXpI8_U`{04khPVSRe$<<rkdEi#=B99}->aQ_@RM+09pv=k~`k8CzJ8Cxqu
zb4_dL?s;Nn$sr)p94HA<EmdlDf}ZAA>*P|_g;t8KS+x?KIjkbARasYM304(7Ri#{t
zwb=8$rZp(*VwPIi8fn_eidM1IamDU$Rn+Zbv>EGam%7_)35`Ta2}uWvL|U_ZC#`tV
zx9yLrzAp3(z;|Wx{t+J^s1wMxdGyrp@Y?G?+?ndWZ5Ru(+W;PIXq%Dr1~A8eJIM3{
zZ5z~X<JRj9Zf}$;ArMoKRC~9GMhlTrDlcEZ=BGdT8L#f|t&LAse(z8Ht&J9O5Ys=W
zn`_qA?Dad2q%%`pm^(&m7Tr6mwy&CoEv0Ke3J$W;(DG%sJ?W14s4`Cz(>$AgP`84O
z{q@ix$aWp!xOX^^60%z@nEA1F<LJBbz~0&osT<AHvICR%<FDn~pqVY*+O>SOwxRp$
zy3EehnLpK5k=r|w^sB3g%L<yOf5R2nSNl4S@7vg%y48->ZL_iO)1Max)7pQayNJgs
z3?Wo2WKnIrtY7gTzy81SzkL25^2298^v`b_ljm=5Y?dc>(fgCqZ#HlKU3m*b`0ly2
zAL2GSS8sK;{pxkQ=lLon+urzOf5Y{;J~r5X+p+y0*Wgd&oqNOMVvuiMnfyn49FIq)
zyOoEB!n#&UYgA36`o9K-mJ|XhW>SI}&G#;ZWY1DHW-!F)atcBHebUSnI>tne6NJpN
z29}}_JdpBvnk1RFY^mIQ`Xm1Q^55aF-~A1%bYB((C8f-C`;>AyAzCp%J7d;+D@zOf
zh0x46Ii-o4o99f&XS6hNI82;R3*~&a;6a3Gnyi3iX$~56Jesz)RYEJKsSR=L8$qIJ
zv10}i%39p6#NQKyoLn#Lpm3TpBEse20caGp!jq*H%c7)g(v$+V6@qsMDq*EPEiBLO
zcwK+W&3eo9aAbb?2)z@|ixmlm<k=^Jo9-FAm|{dDZk*iL-01c!)!iQg*>%|<ZuJvE
z%w*nD{gJ6zL62h9&f1h#OipuNS9GmRVatqohf`F|r?z>pAHZAO6-tF=t(01kkQ{&-
zNEF*9taA)%{^HY23p1?>wKkSk$!jIHM$pQYT%@zxT{rOk_*Tb<(ExVBjctRz0%omm
zqOzscc5iQQDf4@S^V{tT${ry3@s!O>)^%oI_jqlajt+{NU1BYG56XwriJyM{GcIfK
zm-~h}(lW-x6>Ws7B==NW%1ZUcQYL8k4mzSO&aD&+Gh2gX+LI7A{n_Z25jXvveVkSG
z(b*jF;t6L1gUD8$mXhf(bh7~H`VT9T*jSnlUSsU6oW0htn733L*KL=Ua<txkx5j-E
zdu<diXsK-g1qnOz<z^}4wODmr^^|I$x07>wnO-6L2qf&Sd@RcCQKxRqAUJ}Mu>hLU
zbMxnO2ZF|*eqGr9?1$-TY``lgO*$a!_Sr9%FZqu?`#<s@JpJD?$HZESW2rsB({b&4
ziscFs>f3?N;~$Bi(;Lj$<M#!*`uIKq(*209kGr3Dx{t2<9C!o!y)mvMe(e)@ocH?}
z0b{WqBi4T4_k8Ft(3t0$6k%-_)@3yS6+l`7S)g@GT{W;1qP5t{ksQqiMOC@1jn<N_
zFENpl5C!T|jNuhX@rH1i$<r+%W>kd~vq>&mHHlsjV#-|Fh5P$6zdHRU|LyU2`117^
zR-6XJ`5DvGUn0%-l=Cav<sRB<Man&3!UN0@I3Awy@kc-4$&*{_V38s(Up;X5>VaBU
zz@CFCPqb24b+G`D9*p{JZ))piv@~Z3LM;nm+Ugk1eoMrB_-Kuk?VMZl)U<;S#Q|DM
zw95mnG~zrXF}Q9qqA^j|Gt%|hL58V*N|BHn5BZLVn>$|V7o=t4@NmoV?qlZqoYsU=
zh0D5l`ctfeIRpaD3;(RYc&ly(EnCN^l*~)EwJ<7fjaXYP2u4V8B1@nZIG0Mq)|p7N
zVwyg`)MPmkFCdB8d^XL03UFr%2~ZTlia4yeC!+3>Ad*1E()z&)9b0n4{svU6bCksK
zAc$iDTY}p==ll2L4|ja)0MIt*&Da-bb*=!Iu!+#$v)DfdhHs>V`Z4`s8o%=g?27r@
z6A)j=6ew=U2_YHaxU8mmdwqAuXFvTJ=ko>0iAbcxb-WL8N<YeGLMlf2t539$Ov@)Z
z5u;6lR;u+30%QJ0<!dz{9k!%(5P=Zbw@({|FJl2Q_AN+Yp3PPw1QYw3|6s4UEiCJ@
zwRpnQs(y{z&R%7#8;bQDuDYe2T?kxj_xBS^2YXuAbciydJynQv9;VpA**N?=mumXE
z(pfTR`+8A|`C<9bb}XemQr4H!?u9RS%@#GqK!hF3E2SL_Yp+P`!gL4IJw+qq_&unR
zk#@qUMe}0}mSXFbnyR({)^+8p(--`oe)oUJ|LT|iLnfJQ5rtt%Q?L4^cFmo<QTX&2
z^XS!2$5a2cBW&3ATKgA-;DOCXG+wiPSDu;n13<6)f$QVnKR)kmZaxNo?>ql}jJT!l
z-}8oFpb=9b=DkKn5iPX!L|ZR3CWHiGB8Vk%W3c|YNCmOyby*wB+8|UU3Wqr{tNF&I
z(2xjIIuK)IIy@!M2h-@rV7fJ|ePxOhl}f3~!^4HZfq(h=zrpVuf16k5*R-UNo|+BG
zd_>ZW96shIo+#%NcMoS!2+=^;Jm2ui$3NuRM^8yH@o*`e&nuVvm2$oZEyQ5j<-`1x
zyY)*HFZOY^JvhrH(K>|m^~-F7g3nniJ_)su)GTV&Qkh%>Xe_D8c1l%Sqe3923^gNa
z1vKUtXj*LZG0X^7gd=@XP_?vt3eI{qgi47EmuL69eEu`O(gVlyGv=4ike9c#I)VDK
z-;}kis0310Bs5=N?D<|>MF{5lkc1EtAwo=D&m4$(=Gn(T<kKI1$}~;9ez@?{&tGwO
zf2ORBm`ul5tFt58daWV7J*#DQ>~lkmgcz+8%{)<-3$;P1Lb<Fo5+P;sG*fFC9h;=A
z<mtfSknEZ*ktt@9f{+N>*K|c{xh8PFAK&Wu_EW(7RBvbgkn0sYJcf4H<1qnvtqt0x
zj{9NyhS5QxTc1B>h2Hjecnk<Pq-qSJ*)Eh;Syee*F1)_G=WskCDRC|j{QjT*+csH)
zeYcdOmBTwwkr-)e#Rk$U1?6%&(ZO+T)tEmZWbXtLdPSXmmsTs<gb)p+<SDtGO1F#f
z^iNNPIP;fMBxf@a?EtNVpQ@H#6@=tbvbD0)D@L=YW=D@-X9P>J>fd#_pjawnnkTxq
zSGRAm6tCMCpd+1sI-LOb_fqv<qTUfDhM|w^Dc)YZ(lw5?b)DZ<a3jd5EC#G2Kozs`
z!C2k?{mnDc{TPFpe1~1da&zrfzYcrQ)t|pW+=7iA)2)uC6$Ae{*nM4=g_p}q{?9-C
z-}4{*$^VTB+Q)S8_sC9rs&5B>9bj$Jh#dpnDVKh|-!s=F8pGI}y&J3&m_ctpe%o|{
zLmT@|#y&XP2Z8kYs@MO%dfPREd@K~-%QoN}1a`k5qj`}-2$7Vn2!jS~wW{7C3n9)X
z%&&=BgB1>>K!_WoT^d~071crrD^U)_5J}Necu|BrWpbX$leLU%)&7^KK+cgI5_w9T
zPG{CuM={B-tiQ%D$6x35VdZz`|AwE`pJGrnjd^i61Ztc(FJ}6CILv0jl4fphZkgtp
zRvVY|!k1sZrj!#}A6%C=F;7Rdi81A#ZHO^il0@x)5iCgcHBTfHQa14FK)A1upriON
z<zz)()wGEr1eVJ(9?&!&JcH%MRjSs<on-0PW~?lA3&3q{t)6vgyTHc&Ui6N;d{6k%
z=Y&sZjxRpqhj)L1gWgb!8apcj(XD<wfUOz~NL?0WSx7Mv)8y=9Ktd+wndcw>fIt0f
zzs4{BsUI@W6Q}dS)8GG$fB27o%)|YiYX&a}u5WgEN(h0JE#uUKPis|LIUFW(io_Ji
z5l)wt+RX4;Nlb?u;xth&mNqvxixsJblp`?;Ed?f-NLVl9h_(3(Vtdm0boOl0cVqtk
zQ6Aqu11M}%j_-_D`XMmAz@sQ%_kVx1DRd@g0K8#t{OE&yL)>?U>+dlvx4olpChEW3
zZ?Zl0YaV*hzD@|_m^s`WnQo8Vq(}uWYUu`qk;aTMQA@F+hGgKYIhd)fv0N_hjPGe?
z|LoA2J9oApT{v)Bey6`GYbAPh*%0r$_Ru<ZSSlmsi~^VQWpp0Nd7{+9#XZ7eG_Sm}
zu4XSIE;h!}*CCpbb!&}QE32O?4lq;d1tdLvVDjGH4aNRwgC%U*21CY<z3FUQ_xaQD
zb7S<nEdkC<2pdDU2LoI6p<M?f3%j<lj?c2^z>a2&mwS6@^{PK7v33usYJ0;x8x@*+
z1BM`*P#gC-t7>gB#anl+g*GW6M%KF8^V%6u>CP-h9lDj|)zWx{`T9Ij=#GM6{P6J`
z-ZDn8qin`E_Mf2z?5SQy7M<gNtQ~yg49BE?d;+_ay=+z>_U|jDuK&Ldz^#6~QHHX+
z@81;;!8YcVWJ`pxT57&<rI`&%X-ZO)a*WJ^W?BlA)!A+hR%wgLN=ms<>x62gDddod
zDG_6~;P9bwK3B@Q^5p5n^JkGfXEPVB%{0twrPhn}xF#ZXB84MnUu3N#jg6};MLEth
zx5vtKIC675Sev=JWkC?QoF7;&F9<=I=R_?^&PS%{mb-^9XmuU;cGDGtNHib9IDuEX
zs$@bwSPQh;GbGV{L!m#Et(v9Qs@{!B)PqiAu~N>Ysk<y-3}0PoSdosB(Yl&0&X=+f
zOm4F07pyR*|J^I4%LNGw%SR8?XRr9P@-Omh)1OE5hDx_p5sE@-*6OXS#rB0|u@-SD
z5>jMUuqFbLAN=4WKKbEuKK}S4(+uW?Pd+;F;^i|$h4u1^wH9LzgXtKxH9}0>Jh|a`
zbBhA^uTI?GKR~ExjD(yC0!133U<{-jNRqjI`W)P!Hs)x+QI%G$NYM)W+@RduQw1x)
zNhv`o$XdQL?EN40@!^c1eKWfPdT+VQkN2wmMt;q>Y`@>wjooP5d-?`1*hbYGTle4L
z4(D$Or*>zeRcw)pfn{CM7%8fxn<Mcsk#pj7cZbZ`CP1^kv^i(GX=Y5eKv{$EMvRG8
zS7Y0}I@Lt=5k!YpO`?Ow);7m^=^j#+T1eAGU96yF%62~1G~O7pInVz2jZ*5+0d^ai
zWnGK~@D!<~g4eOxR{vcH!H&^3@E9HN_4G;COLl+0VHUDW9jjVpsa6B{>i4A7N^5n1
zI~i@C)cgRu4$`}%L~(()Do8T{HLgjFvT1cJX+*7sDNoe0y4`{fEn=;kZA=Hxc>3n*
z;ZsIOmAcatcBZhWqwQFpS{AY9?`j)HZAuK0r7WB;XLuI4v@+}!`bO`$ZUf=_K&n%I
zs#j0`0DWw21c5HcL)zQt-5A=f9s-GNN@%r{8lSai+IDYuqmpqw1_&ST)<^FUzkdB|
zZmi5(PtULQxCB0l%?nxG9GziB_uUwcIV(?Yl=I2#Ttpg@l+@fD*;9_Kt%Q1plF>4e
za&~R9fCQ`lZ50t?XJ{3g7*pmMgtD$Ii=}uW(oPp@JraU5>Pa}AFD#dWHleP{>EYgj
zlk|joT#-qL(b|sbYP8tR%@Y!p=g%W)nwckC8_$<w%vMux<5&n(<~*@37xHw&e0a`U
z7S{D_!KT{EWQ7ejuLzNtCo~=iaUzBZjfptT)`LEnBZ2K7z7P3>RMtXvc}j0_*^AFq
z?U^*Ugg`BxISG*|MOuthvCK}noV<9fYoslj%ehN`+K#5T`KzkdhgS+OrSP}o-{kk>
zkNLNM`0w*8x4%w|ld)rNlTvD<td@SRrMfMc$%`)4UK1p6n2j-h`RWd(v78sq4<~M>
ziQ|ucfa)_El@yJY>~>ixCXR=RCr_RbLg4=4#ETcNXspcBM4o4wDi8N(<mH*msf?ne
z!_BSFM|W&!X6Lo)ik3nX(=DcC1$n7AHi+HgXhkL49!UA;FxUQgjSm9=uZjckQgPXg
zo_&l;B;LJvclK{%2){0=wR`V&kg*Bwb?aMdYO}Q%=i|i&wxI>nEUNJO;f~rG=k?60
z3;*Dk{@Sh@6bB>XDa~S0<+Av~VlML|MLGWGEa0B$l->F5u9vv&!@3wtnB3+ic%?B|
zH5YU(T``6)nr5(+%H?tajK&Wkc-of%5IZm$t)OB=d-~T?EqxIVVipW*St-SOL(bDo
zP7?%SU02TM^Jw?hK|=E`03tXG5l1HgsvW<lYJ)ebk0EReISf;2@KpOw3ZPv#NZ9V7
zY)??vLJoT&wL33j4d1ONsi!@IeST|oq$YLuH1^N$4_pV3hE+PaX>QA5DXp>=O0A@@
zlRW5~MZez!)^syqxQ=Q1WB=ZS?W168`^VegGagyFey_@#ZxL+o+cSmx+1uy#&ppQH
z`i5G2=jr!RoxeCE1mX53GEd4YRqnbpLJKJ<*_`3cr?alqaz^R_Z40$dtgA^f8V&Or
zv5L)F6q-;glr^BoMhJ<zR@P;)7Gf3Za$zY8+6rY|2`R8Hl_-hh@rD>+tt<6%Vy!~e
zLP!Y>(JPB1fku!98o9Z>A?9Fh;1rhC9!%5FDf66}XKPiL^DR%FeoCp8`_q?btHU_H
z7N0|61s)+yG`dxi7+aN+OJpVs*p|g!-%Z=SB3jkl38r}_#AIthYiI~aF-tG&VH`)c
zqZI=;(fgXWqW0dlx>BSAN!M=otdxjpg8^`xufcj(%3Aon<sb4N-Txo>jc32fzx}8G
zNBo&j{+t2oZhzC|Doz?S2b<^8SZcAZLbY;UF8uCKe!_A-S@kcGr_Vm(>C-2qXzOW;
z;*p_^=!JgALng<-QdZ9APf#i;WMYb}wek9X;j<SfUcNrFE{cj(CdUwnX)^1alxbz*
z;c~~N6jX#c&m8AW66=P97tXAups8>pnd$HZc=6#T*&oyKfi!^h0MvG%UhMe@7{2qp
z{oizaw551dz5)2#?}8mX?m?i))<$XUlKV80rKnT{nu&*7h?H9S{AZuBtSguE#fpm+
zdQPynK$32g;vl30#8Rp^4sQ<Tu#;HIsQeAdKv}7UtvtRHK*p$ODQKz0oUP(dHnuG|
zi&koJclPLM2w_<kR`+|7ZoMK{OS)b9by!6>D`wU#yFv@kM%PlPVAZ;{l2T-zXL7d1
z{4zE!bu;N?3$I^~Wm!z;<co5@b%N;}d$)k-*K_Zy*uTRrmk4xEt<GBY6m%D5Lv>wW
zRRgWP-CYO)4Q>?cEI?~^t@`KFEgF!);C60VJc_XjW`CB&QX`!a_FVV?o;gowy?ou%
z{N54!cOT3+^t7pgBH?Rs(VOV+y)W&u2|Bp*d9*S2eJ*VQmi<7+xxLMj>E{81oqZkp
z#og!rqL0JP4TqaR(3RW617GIGS1%XN56ye4T9H62c8x`#)~(}Btp!xQBZgZaVKk|>
z##$<2S!pd1WI}QvCF`$UYcr$yx*}NH$q<DYAkeHrIhm&JkP~Nw^n(je<AI0Op7-+d
zX9QjO>I|Pe|H07z;=yn7c5(O`8&f8xTXLSbIo@(UpLw|d8CuSCPtxisXe1EggoNyI
zb0x$ClCX8#CSp_T&D&xEHK1nJDPEmgVQFUkd^pae7%6}>wKZ6v)<*KrM6B>avVb<y
zbYLl~9n)nuG(}tr(!>GbZv6+geMV&KbfN_cl=T(=_3}6Qhp&H|zx>1hF8}VY{ZF_#
z`~W2&t+_X8u!?6@N-I>wJF*nC71qmz^>SvpT$s|tliM3UdVb6CW_AO9AxCjeL;QOQ
z4Y5k$9C<Pyxw*Aw1tBDoYuIm(XKs!cUOm5Xz9^@qpaoJ49FK)52BvvpZI!#%Pq@FV
zsKAsbZsttR(Yl8OVOcA$@6XIZh|dDIH_!MU0Dk-90{}q(%g*Z$cY&9>Ez~2`+rgJ?
zT0eWyHRwatJ3d?kwyzyK;J2ehzV1zIe4`0*MU-_}_}R~X#&TI`H0HyBn8MIbb@nU{
z?>@g2aVzNPjmH7&YZ)VJabwP|0qok9os)h);J9ufkfv;(<>`x<5-JABdiAbp@&fB>
z3|(u@HGS9ZnR+owtCo(P=fg;kRPW{6L1owW35+y#%E_vYlUYcZ*>eS70Q$A8rKh?r
zot%6@-g<d%BMso^5&YG&VSn#n64@*wx>s1&>UAGMOVL#K+iDi&OB<EK`>J)bu<^nX
z-Hg4}qxu*EVdq!Wdn)%K-r{ewurX$E(-~R%dcAH1{0A71K(Oine;W(=9$>ZC-+j#(
zp-2y$wjW-DKj|-fG|_hRuY<t-od3BQ!7E_--2i_8^Zb-N-Ex>JpM1LTqaWV$*%x>G
z^d~PlofgiQg;)0r%ToNAkK`nz3?V7n3Z}VT36hXtY)uqvgV?&sd9ze!Ah9M3PM6EY
zp3N30%gX)hGa`_)fk0bsgu{G;9FAlbO1p6T^gvw9LTP?}!xumODWz2AU-}im7{m}E
z1b;p&Ua+DZ4!6JoA#gsw=Jo5JQp<y{HO6c;2ahq$g!w7j#M+vLgrp;Bx&f`I3R;zw
z!jzn;tBvB02qsIg;-=;)^rp%*Wk&z<W@~g)OMQ2Fim{PaSf&W+a6{2bt7Ui{4#RpZ
z^(**V(Fp-xn?_J5&XSKkv2Z%S;;;YSU*)$y|1JKdPyaRkh1<VO;t495`7o1GpaJJg
z;r`)b1$Uv51Ku9<$nki?A<s<H3?Xq^8l|qJ0CN;?Y$dkmgh(kXfyT{YvNWy#J_a3T
z^bMGD;-e>#_9XJKRF+Gjl*T+Kp5IO!=EPFrtHaFU)XeA_VV*N7T9+jPNN%jn3K?UB
zJY`rv^i2EXH$I#O@NFY*3-w?D9N6s_aM?+)z}P!`ZRa$;X0aDb4~@O#F$lC1*oHpZ
zs&;9!AY$H6%Z0CAe8pEUzvNOES}6RzAO3ZcdL-&hSJ%UJKwwo?(bEZ=pA?t~Hl;+(
z6LmE~ZYhPf7D!>UH<9l5Wd9We$#F!pA(|lAQk$0A5H^rq$~saB__@lsl1fa`0g=1G
z$3U%xb!h`QX%|}2ftiVwQ;dW-laraxUe1@D*|i`lG}^c>#eWyX%&P73>(&y={}&u^
z86XJTeKJHt2kCuLZQ2>Z?Iw1ZXoVPUe2!HA(>KB9T0?J(G5D+Gj3s`pB*;kbTFTu{
zqhF)osrx}Jdpc~irt6=njb$}0VEhAQK=(uay6pJ0Esgy1220=Ivvd#s*aatS&*wIo
zI<Vu-x%dbK!ux^hJAb#!2z*znfJ4m0>4xbhaySN_x60F}uX*z1OTK*ZlDpRn?WeCe
zpI=f@4mtAtS>$*$>ytU#6`7)?VFJq9nx#;S6*wHd!nn&SBt$oqZZuYB5S3O>C{Qle
zMM#d;Sr#?v#vGLMd7;gbDQE8P9{A$3m#j<Sqfb93rHN%Rxkap2Ku~M5ezTt7C>374
zddaffQPw-8t<=^``_^2;C(+u}%}<bYV;v^+1o1$KGk{sb0ATva0%HxVt7Rx6rsJ(W
z&FdQ3T9ur<qlsFOELahrs_7(ia?PY?a?Cp~5mO$r8Fi0UfS8==b1g9VTI6jp)qmdm
zrK8rJNX6Er7w6CUYd`*v`R({O`Saml<d=_sft!zhg?N0zx+<?9?l@moYN?3WbC8@2
z$RSV_&P#z3&=5H<t2sl6RX$4yZj{|PF9j_tkrOEg=6Q0qc43+(5TUe!s4(a3jRpd>
z3aTp-1Jit9J_J&QbG430W-De+6UH)HKlmtxl&u)Yx+Z-$=I$T$@j(Ef>c*d`Yc$4a
zzEOm21L$t#odq<I!}UfZ{vXkAu+Yn%SoCwcW*4uBkGCDK@0()peYo87dq4RJr}G7^
zY6hKnDmBJ2ffa@YKnEDARuLOUm2e#WZbJx6InySq4hBm(uEoQl!TPn$5Lg)L+MsIf
z;n-5H24)(uHS#o#a|3*V*<elwjs2SDWYu(5Y|>LIy6f9$v7iL&JR*2SFc@fD7Xw4V
zqSUra`2s`mAH<uVcZLlafXWtpukLk_s?N@B4@(D*ZpJl$nrnT!hSLG`ID?+90Ym~-
zyEf6FlRBd&uvyo5g@H&QbdOYj#;#K1{>=1hXGEM0_EfMvFo;;6VQmfKo^O70yt_QY
zMx(b|8>g`QOAoN#e++s2ul*0+G1uM({M1g!j^eo8fNy6acgO6kUk?P|c3%FPO-md`
zkkvm#JS$-L`RlIMFUGjPFWg-UO9?zfI8GDO@d?j9QY0Lp-IMdhP1yyFjoV}7mp;v?
zW}e<ef(Y{zsLi!&rxVL+9W$m73Dd+hAILdUiV$KV&oh!Uby+-Ms<i5?d5WYQwt1mm
z#P3C0r;}B)hq5rI%8&lk5BcbmPgu@g98nrX>z`h$^%pm0DWI+J>gDgDZ5hVOi~Fwj
zIh3YbLOv4m4KW>AYeO_35_d-V9&lAF{#aKpdcoJ3=4&Zlc%-(5p3fHx#8>sUaD^Hm
zg~_EUcD|()rkt$bH~4zxY250%K5b@e1+;3Jn&5>lgC%cDX{%|Akx}T@*N*@@FI(k{
z_BsE#e#s}V{~P}N{crN?AN{MW(+{Y{EWgg>;-?cVosYOxP6dhN?6xxYX_UsuzGb4f
zI9({~LR9lZOy@$w93o=2bb1hBvOWTo)>x{sq!KgGSep`8(?P}{%*py5Uub5xgPTYT
z!-)KRIa8~pgm1nZ)Ax_`_|^g7*cXiZz3O7H5d%Gq<bQ+>2)^&Y9Rzm$$h(H@HnA)A
z;qi%FZDRZPZ{V#e)OF?l)jdD`$xl!<#%xZmGZCx4tU*AmucTEd8ao$bbo&)McDF2v
zyL9j9<r^YlHvO3_J5Yf_h3ckL9dK#idesEArIwA&s+Ex3j%Ik~*>%}^b=tRX&63#N
zw@<wV+b~ykKpsTK;^#^@z&42L3t_OPqx33l?f1ExE4$&bhIY*+_0+uf&$7Y`I{Uc0
zPdff)(Y!65>Xo2oMBOb}jOnwDt5^T_mT|2$s#=GU9;Hh!Eb$u_tbf!1!|IOw{gLVx
z68L>+UR+X|X=wIrTnE#A6N*zt+q!{X%)({+j$c1Sjr}CZ0D=*(4i^9X_b_W=T-$BV
z>|g#S{(B7kU@)`q8;?QY&pE790tWPad{_62yZf(LwDNGi;nnNNVGbmLrjf=B6wt^#
zAGlmrmi5BrGEv)s=@6J^>yi_qnHMk1!ufom)W%_+&G>mvOxd)zYcYvLJ{+w7?K)AH
z)hg#N-ibuIy@_|cAQ1vQgfrxbNae|GBtQF@PkwMinkFu1D7AOUk&z~?RS~gyR~z25
zTv&^xW=k>MokU*;(t+vbWAc1UsfCay1WPGv^UQ@u{GDuQ#oAMfn{kHdS&n85sEP%-
zph#OxR~;jj;4R{e>C4@jAVfmLi$KC?vzewD0n4&1HkWI4)-rDE>?ZBmd>VnV9`|k~
zqX^I^V{-%hs><hl#mo9%@Z-<_34bR1CeNSz3oO&er0IaPLW;?{J%vDww(&`ri7B(L
zjcH0oR@CnQY7C}?#^G>4b0Xr`O7J#y$@`lJx9^QCMOdnZ%TqFqplV}PWm&3U5Amz7
z{yzikt!K#azRJM@l=^c>QNAC4@bSS?z<XBX3i!dsz6GBS4Kq>hFS~lv+m5Y|z@l9J
zu}|oZV~Jb=iTCY83g#o_5NpxG!{vb&U%s#jnv{S1EB}SHw~3|;Oa7U`0fx4Qcx7v?
zb%2JgOGhQgZVqg0-{gxxS<9vw@pG=VqHD1U6M8$fZpd8Rp2WuGa$!z+FptG6(1O!4
z%^5ATf!NqlOBZ`zO@vb00EA%dUFyXeot2ZmP;HgFz3`x$N~^lBpEG_|$-MO=e%t7t
zjicaT+_WvNh_Lk;Rr^oFzO&G_1(n?)WGfP>ZmS?kM)#AH;;z-6v5P$w-E}&>#ojnq
z-+ZcfFRAV$*eevRiqW{to*rsj8-e~U-^qk+<59q(yN#qN%|ReIIMSZge$Qmu>j&99
z%<OFYm3!Z0+`!+KJh!d)jn94SqQJPWcDQRD;NGth*)+iWHAP2|@J(U!AMJ5ozvR*$
zNcmHi^Tc{F@HeL;NFe0QCm;Wi`S65?hZ7H{SDa24KL7H8Cr{mqM9n*{r(g4w$r3mo
z4?O$$mZwh-91as93bhq3>&hGhIb^~-nGH%cU2pN0mw8U)DL4Rx)(ZbL`~zAro0J!y
z{fwI%Gn>A9eMhYa&?8{vY!BRQ{j;?orC!`+Mw#X$5KOz6W|T;X$=JU%Gau|4G|y7h
z*3hzAF-eTnve;a071}~b6B5G+y!&^FITCVmgMCj?Gn2fqPR!U3>n!4InyTAunN+7c
zAv7%YZflFp{fezM63p||vSMM#K|&0!e+!-!+UdR3)`wEsSci)1=TltyyWt=6kJ|6@
zt6%<Y{=)RH@%-7JM{b@I^2D4bQcg_S*1n|}TPL1*FuRqWRf3|%?%mvGOCL9-dD^;a
zBY7)Q1B6CutiHB4t%NXfND-09xvrdwaw(Sb4I(USVbz9KC5Oo5pna`{qE>JiavFQl
z_v70gAAkYWkwVi35_YTp<Gp_WZ|@^$i~P13UimkT4|#N)_w1W}TL`*4k4_-&^ig||
z&`(VTmUU$*h0j0#l5<%&pHIAQXNsp<ZDK)+yN#<>i1GUCsv;q>mf{hUZi|tay-IVZ
z{pP@_IIG95SOW);%2wUl);3tJ+RWh}ZPaEHy6e@}veLc$OoO7HMrz*j$HmU7EbC%m
zMUAyg(R{6X0Z0he`?0pR0YtY;8Q0y9-xvKJ5r-Xc?G?8>-GCj-pP|+^Yn;7StXEQ}
zXw}&k;rnIsLv=%Ft0=bXi&$fJRb{;_<1<w480)$ce8cQfzKnAU<GT0r*!$l(cj*t9
zUAy9j;37_l`~45mh1a5lx?x+1_Ibsl<PPX|kLq|ncNfds+x06iZi}hxr{CL#H`MZf
zgz>ZSDScu{?*bCg-s0pPL;W5Mz_-Vk*MGTw-wynB;J!UK{yyD*PyD*$oAh@B0Ez_C
z6nOenc`_;MT6yvE!nsuPeBf|%AkQ<OIAD79@{ZRpKBKh<&Wm|biUjiEh=#~CTPA}T
zxxGE|^x2W)@yIkE0GP8<Qa7saMPM`8Tc#Nzu{oM@V@`qPeB!f*pYppm|A<=cIeq>5
zHJ^X>IU=@(J$?EkjyFfMA5=BZOP|k88Cq!#mP%YIAc+_c#C&vr)QH5%JO^_eCjB;9
zwrc#*i&hJw=E16N=QO2h^IY{g+1wN#E%qKv8Y3Z)@=S;p$Z9iYOWaIbl9&%_vtGch
zHp;rvTCt8!PVcqm*$F2%x@T(7bo2l*Mq|ag`M)+>`_4<{bgoR<bcbtcq-bkPD~*N5
zZ^xhT)8#+mH(&fdzxvU?!SUwTfM9ygoDATdFVw1(W|iZiZN-;8W{NQ~r9_UFvMsej
z>rS3*=IcG!!dT8zD<i`Zf>6|KY3|oT4uPc<&dW+^N)CZTR_fANz0doyF7}*GnUG{N
zHT`~kYvaQ(fPGOG&p`!!6a2kvSm*@({k#bRA8P>Lx}CfgB<>iLx9k-|qoGP&tfwm&
zLwCBISS}aq$tYrJQ%^63y$%dUnl!@bzA+dj%>LvmP3^z0Rm@tYxNfD-(4EI0*urU@
z#T-QTbaiu^gAhiV!PXn~6kKl{AH90FdlL54gX!j4t8Qq@(AU&GQ>WHDbAKO#r$$X{
zvoT#$&fw_}^8$3?^;-}`XL5S6M$Qv1<@0N}v4T~bH5vEMZGK<6JYe_&?m(jiGn6fD
z=cncN4c$J60rK1vOdj1^cI>Brhd8dc1M*gFgB$ly#-3H1l7hZ*+;RLzO!e1}D=VCS
zhQgjA|2kfA_b9%BG)JIB5AL|Sbvpjk?G66jw|Tq&EMa@;?wNkX`~g>%Pu;*5PouxV
z=!=QVzYAlnjkOhmuFSI_wbAMW=W<UA&xtY-QsnmLh7@mk{_GjAK6=h_d4K@N+cPm`
zKKbD<Ti@2&sI?$5aC4YB91<~SW37NZ&)jGr1e}>N*+Wj*vkWFFF`%?k*Aovff5JbW
ze+M~6atKy&oD!GIOVgj#Mw%85hYllPT`fImetF(u31{EKM4BeRjqs-<NJ1o9afbxT
zTD&O3@3*sbeai@<mf93Rf(Lhvb-55ir1YxiYDq}}>T-{2?~Ok7$Put-GIXt?y5@1%
zT8Rgh9=KOy*-BkW(bxOdC`E~*FTBl7@$YR@t)bexEHy7~fme4YnkJTFM$jU{VX_d+
z*0Ka7Ud5IFul^SQ@b&NWuj;?Y51#)eyLTnAmcr?>A`(p9+=68au<Peva}Wew)YBn$
zLs|X%TA%5FN<%fUs>y^NbfMBcV}ofkF$3&#sa#4$vFhKYH5PZ3sMWw@_P*0GhVKF3
z2N)lQ0mKOqF>$uEU8UaR*W;VaRSMVz_PY(@8`(qjgQ@N|q`rAQ-o*V4V+Y%xnN5KZ
zgd~a9D(huot(BDvzx${D%i*tOK6k;ZJ2y75&#u8#rDW_}_uA7fsxAmkngf_N`c~49
zbH#qz!h*V{MU@n?fsM7APgPK7&KiV(?7Y&t2ClZQdE*L<bnT0=b5;PcKu*7+-&99i
zrOptxQa6G}WOyR>w6nHG%!%NYzTIrPe+O$wE^n|l&Dp@=`6t`mvaF&Vzq$ush{k@c
zrI2%`iFI7b$<45}r?0)6K-VI6FjFfuMJHU}r{fx{1LV#m8WGY)VNdgdrXV}!%F`=7
zN4-@z?+Tm5b(|r(Nd3(bVfgrMfb-F?i~mT+XYB7I`RzN$4q`nz@HYpcdo3WnNJe@W
zkVo@&i%PclU7g3zKLXC2ev>ug7ho();j4!QY0B%@2Wnlpd$=&?BU&z;YeB^-Fb@aQ
z1Los_d~=IP<Z%0ALX157_!HJuDP=JLh>~(3Bx9W}7wafdsxf9$PE6J&!qy0%Ye9TI
z)B<fm&#(AOKj-t~msG{G42d{DVZJ$0E-UAH=5jfkj7D*g<JG6Vm8Z)^&g)9mfaKZQ
zzPT+^$P*It=nynbnRP908n3R+^A19oeM{(_giv2zl~z}4uLnRff6um#9^?`tF=Zea
zb5|>pB0;S1Bc_PBy$mo6_08Kfjwd(g$(cL{m(_&uDOu~g*7~~a^V$Q@ZWUFlQmSxS
zR!R%Z#~Yf$beOHzF-nb)r-Ny_V<4mgb>)ltC4cqx|BZhoUii(A|1H*PeS1q)LTuK@
zIC>`)#Z%56NQwEQu4+}+#jU-nZdKSluT#`o;+5{>h;UjL6gW(lZjWN?{XALWh_x>?
zhE@VhF+jAY0}|8LZtwf?t&a}|08KN{w=b?4tVfai_6v8ldW#0g{?T9e_Rs9U+gCk~
zzQDb#3Ov92HU?_<vtX@$LTfB#<x-5IFP-_5&h}e6EV(6b*Z%E~*DG4B^4HG6G$j2z
z+?GTw&6d2AhcS0vS*(iXba8FU*1{}GPx;yLT5axKDuaoswYepUr5n)BHuVcD4)lt9
z2391hFW@>-HyzwOzIq0XwvD{^H8$@-)7W%~)STrDk3jOcXYO$*{@MW+WQ(l(^RQ1D
z+BLRb%qb=>ZZUc*dY6$}dup+pMmr<npnvr4)G(w7y$ziDW@o^z>vN!b3nN8rQI>?3
zVriHC^VKgk2Bd?0lin3OFt%eI{Yww<*SXQXq#^8n^Zrke?W5ROzTND7=ePoY8y(2l
zT(Ig(U@OwyfxtJ8_lOB>bhGQP`40O);TL>}B-YEBpZ@+A{1~(Dc=q&`n@^5NtkhCi
z7U6PP`Rav$M5f7lxz2}7jM=Jv(~+o^iw0sWRJD!Tv}L9nR<Y-#7dwR1r49|P;xZEl
zDC-K#OP0$k{{EAHO3hXzP@AVgC31ZJORUQUIiHA<{r&bxrIZ0HKCl4Lo}Y*Ff`*w8
z5<wz@0ZMS4at8=wu3MGqqT(4Cd+sHM!SYx4Iy5$NMXRmfEoB3aS`o~KOJXABgR`P`
zear>H*t>Qqe*V_Yq}FxFP+BW4eX-z6LPTq|>*~FMyIocf<g|HUIqo-RPG8}2sZ>cE
zZl93yf$8`J4RNrs)8RnHTG<7y=yGTFEitpA{LR<@3Ag%;U;Xqivo!bNli=Qk*t85X
zG8ABIS`;W+%v{=oxDi&yK*b9>`o1I9Ctkz~cfeBDDJEj8#tM=+Nkf)G#lG`R3SONK
zYpte}{vP~&u<@+}z}CI`6h~jzBK)r%eJ^_3Q_+c_4nW=l;<|Qkf1I8^?Az_WZIuy*
zdjRR|VY5o6mB#6Op_XDkPzxhXDOUpY>(A>#=dSIm3aM8F?>FVn7VSWer8RByY^|Zy
z^e3HF!Y{4ht#uT%R!aq|qP5!ZV?c|gCRDu)zbUmA|BPs?q*wpeVikj>7UOPI-BZv&
zRKR)!3#Ps4>>r(_a=;U#0jA<z37THBg#CMW{nq$a{+|#HEN|7i4lcx_dN#nfX$dW2
zpBy}gT`C7~=C#_Hs(WsUP<N3qFxb}?f)@eVxo#D_!I;%OkE{%qPkUe>{=D_)xC1k1
zbYlwdEP#3Y@Ib)-752w`$2-Q)zjlR{-F)efnY?v=z4=ZDiTmJeH~;n|plO2p&-d4+
z2Rqxm-;8p<+3wK0@O&4>oHA3YNU*-UhhyRgA3fuvrw6Jk5BC?&XIP6f=_vDj$8x@5
zI^1wMKY$211*DnzannfC$ay(ib!!%G4#Jb$#9^KWD;K=gqzW9ww8^5doG(1QK5>6p
z`C@v(-SKQ~*F3#!8aM0ep%SS?Xu_p5qFA>G!7N5Zz%55<)jCRqJku?gL`*6XlLwPx
zTDh`TVvek9Bc|x<Nu!j7m?9~gA7HHo)<SK?*2T6u>jqj^Pzw?zMAY2|j0H@tNp7wY
zwC6CVL~D)9<-(LLUD?mIEDP&;aYqR2$~1Zl`?FhX8*5l`gZ@ZxAnJvKVltVvDYqZ}
z0127H@yPMX6GDokcrZje7Y-9uEB9aAQwosmcxzkv8~1+|StdUH=r@>VYvmS`1^adv
zto>)JzSn7`dKRVi>fml+hONP+H3;f~v(5SJM%%UQiu;_^HA2KJU_@X_ffORgIWece
zc`cOIh!}zKJqG+R<J)HddsO)d^}S*7($hq$5H|~$s|Vo`FzaLMqj(K0w{8UcBN-*A
z(sw1<z{cZq+I>c!$VDp`w@{E9;m3dSzp@F^BR*%xdiCt6Ue&F_)zYoDpwwEepKenF
zb)CBNlyD6OQ}dJ^T=wfF<C^t<T0f7#T2|WH+Z-_>>)vm-YZOP++Nmh_tLq?ZU00UX
zDv~8^S~x+xUE9VKNpGRHOC7gved1}&s=c+EI0I)3cGvCIS`EnTw05H+xq}KZz&@D3
zVb|h4iv5mGITJ&0^JaT(^gfk?u~kOey+11<b`6oxTV*wr)&g2h+oQpd%ckkWui3@~
z+LN{g1R^FD@72Yk!41u~r?8ZQTR6U(=@Pn^3wxl7?)~%Z^LsMcWs_{c?ma+mr<)tM
zcVNGMw(kdm4vqcgyCB8?W6;3o#bXe7b(H?&pR*$f3|9HOFb>m!Po5u0DRD^f^l9ef
z=LcdCUcGuChytGa6l?odmlJpAJC5~0)O+Tv{Fy(Y+}#Jhd>yGZ5W>VfWiFS6JVicw
zn)vkN%<b)wD1oI^E^DK-N>P}n$g^YSFeU0b@x^BgzqdqwKl~V6*Vbw^@FdatjGxb}
zYXNLcDJ)1d+Z7Q5J0S=flv?3bR#t@+O)K1cb$}2;wsj%k1#?0d1)J4MrCeSSD2SLp
zpXul-tt_;<0Jc_zn9$ZN1sww^Pu@LgLR+P*i={}xjrx5Ki?A#gVn_qP^+KeWvK`mY
z8tGtrl!93ijf$s(yW2vyiV+7}_RKG(67vy}M9P_bI6#P`l-)yCOixH81ZV|LX(Aq<
z;Vb|EufuEp*7~3F-}~w#e()>5M4ThNA%SA7s7YHQ!*^GNAf}J*#fT<d+E%3~O7yHn
zZ2_&IrMX|AF}I}}SBU!FgK)^k?1lhyj?5{Vy_&e2g>3u7_v70gAAkWgKFFx&>qYz;
z-1&dTYiy#^O)6n{Iz2X>z0!>h{ImyeY`fqvkM>Uy=%xMd9AcBQRcoJf|8U2{{T*dB
zbIffEH>6u@rLK*b5;>${@$p!@HAKZhNZCC4v~xNtV`JPlKWCNti4VtL89I*EniY+7
zGiL`=O*S25X_e$n(nm3eGba6K(kg2$UIiF8I~(1AST{ZHI=5?$rm+W28(r<!3uHDM
z6-(WSvkZ7-zk!U+2z<Mz4z&F<YP0JUWAq-C*7$wS(*TiUG3h~uAxCE}{ah^nzUlj9
z3l5qC;}R;-tG%s-)g!i8dIhm=-?@VyU+8;~(7U?WDVYVtLM0L2P2ugHNA+~h$fxZu
zyfWGm;`c4=?#?$M02x>P(a*mIQ;Br8v!9!Eigp)+IYaO^5cqXt$9lhOv<`;9Be|q`
znz@ykCr^(&d6Ic@lSt9P$IY|E4<vCsTL+C+lyx!LMqLZHw?{ty$ZSA<^-msnI4SwF
zg;)255E4&hX00<sxV?=WkD22<**sV)RTql+e2@g5-b_5bolF|Bp7^`3Uh#5yfrJSy
zW{0oMs{PkxWjUV+Az6A^l~a99%TG`;ISSSt(%7X_t=!j2%!#eA+*`5vwWzfcQea&R
zdCF#o(x8=<_4E~WJ(K1mBFg3AwS%;Ux;%IgmI-OHw0MZ98p{<zB6^F;nA7;om=bk0
z;E2iqaqDt+A5_rbK(ktQ#?}NkrIyXvT|Irip9@_!qN?<783<OW(;(#pBFx7d(md@_
z%dRE#$6N#!5z^rZAt5x%{cBKT)xJt!^563R<6nOL@A5n!%yLZy5%=Wtwa-WdAp|7?
z%VM2fYE^EIldtIp{N|Vm0*kKDtTk#ailji81xW!BHD6Vf6d;9wN;I9eT9_MLJ2{0k
zFoJw<1M@+~2W<ss%!mE0SA6aFl(#O|zHvSVe;v%|_|Kca^DL4brUNB0ZQWz^`53%h
z?=x<#(i+QU<@McbUc7qE-Exl>s}Aly*AlFwg(<1yRpu<E#Imd#$iO;nG*xbHZmg<!
zS#9C*zPDRJhC<!I%XYH9FT3?gZ$SjWT9BM9N^aiQ!YlX)kxFANR!N70Kf8`wn!@m&
zZ{lv#z>X0bm6pO_+Q30xg3e>vuhgkvJpJHi*M2=>2=4#X&0^WWrii6)z<u;K6YbU-
zlSjq1R1UMXqO<ojayAyw7)ZbWZ0U=zJ123l-#@2`Y~XlA!{$?_{d?(F<sL$?l(_@z
zzL4$hbvi5Dp{{{qAy&U!+I{YL_jo55>wr)OyE%xvpJQ~~)*s++m;r|UOzHtjN9*9g
zdkbaW9eTY9^uCS>?1W<nypIFUU(_*8Gw0%+f?`BtgcwLMlMhF3pFZWhR!V_V)I9HK
zP#S3xo<4hlgvfkzL#1;2q;LTmjY$GG5~*U6jJs1|sg0aXCS$#V-GXYhASq>AFAA04
z;U7TqbYzIA^fa`kRp<F&`eD@dZztBWaG0NxQ)X=mk-&<wc()v2XpLhC)@7+X5|lzq
z(HKAxV$Rlkw_IrJ$^BIWm(y$NY89`2F2<CQkP(al3^5bqY^h_c_`~#g!C2H{M$ICD
zE*I(&ob8eBSqdoYV#e=lH|S3z3uDWtj$*9-eq%Jnfg{dXS^-E(83}ft4b#)jIZ^%E
zsM=+TKGlmCnMjDVwGq-xSx$(;S}MPr{vMy+{TbrnFCf|8XX_s>A=1IhZcBzC@i}JZ
zG#SZ&GDS<>TH!_}(O6Vi#f$7jNY3(g-&1^jRR@t$VQo;?Ml)A_iWkPzZ?@QcKmHlV
z2S@=IaJn49iZ41J;i1bg3=S4G0fSMB0ce#Sw(zk_5s5z9oaFi?s-}GlUQOK7mHzKH
z_Mi@`RIu>UAKY3vogVo7i!WL8%0K<l-vW9e1k_TtO*gryUpxD=ReZTD(qb(MOwn{F
zDJGcQx<qxbr$L1C`Ql*E(v-dO7L54|A|8cSJgkNoEzN<To(@pKZC})z#50Ua6|asJ
z@ies#o0uTMu0ya?W=?qkx1OTuDPVCQI$!{!N4IPs?9LQgHF7fzObAS~d-Hj(V6j93
zDFsXKI3lT@HjgQxC^xtB=qYTU8GW%8y6Ls0lfz)%j18#NwtBi+To|n*wQS64p{7Le
zO|dh(!8eQ!lzPj!rlyD5TjsRdhVDsgC5kXj6RM5l;Q(zx90XbxXY<$-7-=2R4ng0%
zwBWG&QIG>2f4>J9ZN2RnlWrGCJw~14?4AaYy))2YZTq3x)wJBrUcWF`M;)w2@c9!4
z%M27ZsCGX<!O2EnASO3d*LPG@|Dug)&csEBd|;X<o<BJ<&l5Q%N^N}c>Vflv>#HUq
zC&)pND9neS(3*1hP+68nQJXtCDbr*+-XMu6VPq1f6bTZ^l9+1a)vDY-th{`EA-Q4l
z@1B3k&+2D@7dRkL)jD=q-#>~n{7Ws<Vr#_OE-aUe1%h!Rq{)ILRkUt?pW|thFs!3w
zwbkWvdSKF9bUmTvOlRF{U1_d$1~0@S5JDzRH%0=CjDrQ~y}J)V-J{Tp$dY>>y18Kp
zne}{fR{%jmrnPF7!6BffjzBV`-jiJr*E04XX44Q~dxsK)6ph_mngx&&Bg5<(-?xBH
zTvSz-VlD(J1_B9Upj2z?R%#>XNDY<$O8+X;i;wvD$tT1p9Hzt^65>Xk*4@JkUIHBF
zi5zEJr*axLQ*%n-S*4T$Qx*;f>o44@l3W`(7LYObt$;*TNQgOUEXFPi<{emDWi6HC
zhnir2jK{YR0A)+J*}vb6`LCI-4k#2=vB-cuYkvK1xjxYDgLV9VEFy2A8aqI<x9_((
z-%<JoB&N~&>cvZ5zIerXS*%-$rv*lTW9gM>-9yirAlVF#yEcwslfH-$mUX0vb+_ZH
z1IrG!HftFf3|B3Mkh1~ZUL?Y!@8Y^Kb3{;2M>Ox5>ctFU1G50hJCSsNqu`CZ<6v0C
zzgut($%tzE{r2-6b_}xJn|^Kg+QAOKI<u!;mm&~D9;xDfaiR!BV^_g6DzT>zbk8`t
z70M2f_s{Fmxp^=8LP&6NGw$hS3t#W89A#rG)%zPe7P7A>IMdvpTKir(WqLnR{5Y{|
z9{Vc?nBkhmW<PpS#P*Xn2(NeGysst3`eMD^ruV+oX15v;yo$u++Vj(c!1sN;!$#X&
z-8Kqz*oUe!Ro5||eg5_YeYeJZ2s{mm7$dh&Zu#ib=X~_^3CB4RQ10&+^4*08Ew0H5
zrWHG6Zf*`7Zf+e-n;CuaKD9~GkmgJ|aC>_g2Ef`X=Zi5}DP|5ce6>`5|BDANW)r@8
zcrf$clr3YRy{*~iR4`Jh(JFxL{##*1&sJzF)N(fVOa%#+f<=P6A{l^EZ~M8P9{`~)
z_bla%(9n9ZAVV8LCSsbY^<pey%zzM5B1~5G8^yg(Tk`<gE1lIehpiS*->Ro5^Uz1<
z!;!k2sO189e{Y49C(fsZ(y;3|C8jd7#0a&Ht|6TaP!I06VT8e2Dr-?%4NNhU^Gt}D
zQXwTJ`4jF-8*4RVXi##FXfogx)<7%D<$Qsdh$&HY<?qQ~<6n>eGv;(-=HA}CU~k-y
zHiX2S1qsd8I511C7>UE+zWFuqEDnLFLiQrO;??B=(`Q>aujCkLAyBo^s(AuVF%l)P
zw)$ut`F?y`;{&CD`?7A}qGw?D0{TG`3A>x`+|Spn=rT;_{9SF+yNx{E_UhfS9hmO3
zO;;?{?p-3N1Wu<jr^|)aD`Wfd2P~7`%)4vL`T=zF+jEAms&0<(9(nHzAH*kP8=zxr
zD_3i^^fQ)5MWM9f;8LxiL<C~6Yu<Vqr86@ILW76{kex}0SN58&E&+C)A%~&0>xCr!
zyY3ecV^wYA)?U;iTdH+`Cp!BV{QK_H_yTC?HUp-U+*gGfOKPu6z6g)Pm;KGuO(%Ml
z%Is|A8$n5bZ9rNz$A4!=o4T-BtY0UcdDZ?rIWISm;11-5j((c6g~p~nnV$1M|Iz>2
zS}Cyyf8T5oyt2&sM(_BBW1RonPh~JPyI+pychB(!MVdbk`?>bEp#~TS3raIU1EY=I
zW}hT)o||@-<r`nVKl0=DcH-G5PnqT;PoF>K$<teIZw}-jXj{m+^5phlL8>;EwQ#yL
zra5su9yuQDcgMp-5an634ld>qA4nl_yqS3Z<iK&Z;-t%c;eI{w;^hKNkRy4{Xe*qT
zLQa9CnVcqC3_ic3dE<%o94^j=H7s=+y633}oQLC)x-LXf&X>hB&{_;2RRe(98X;vA
zXVvR!{jxFOxVB2j2e)t%>jNAkuyth$!K_`Z&+I~;CSuM^F>reI3dzZ{9A<5T7^t(y
z&2$+Ogy3XEo-%_i(`JUpWwm0ukWsPpbk2#zvk|Q}iWG8AqbRQJ_9<1}^nHCF@icW)
zkE7gjN<50{^RHG|tC0{<j2MqGn407;aq<4m25ptk>yP=p`rG{FC;xo}WB-g$4-nj7
zLcB1;8ARJGY$jH4yC5TIRKyZxYNq~zQc!4>Q?b3u*3P=^6A?f?0B)QwcHUDErVvQq
z7lC|`@d0}Q_Xu4FtMQnLu@$=)L2STb16c1KeNw#P0q6i>zezeN?W~>fh9vOuxvNsO
zG9M=56yd1+?(^Sr(CH~N-|)?h)R{Ctu5{qGs}LQjL)ZXepJ=@*D{kOnSlzTnUER*0
zD&*ez!izt8MfBKc-F1FYFEZ`{_y!P@8&_}V+reK5!M*Sd{N<EJQAIaheQR-#13-e`
zdy#QU{WA>sb+Fv8H9N7i8YuS8B%P5|H(y4!G-zjBdz&@Yt}_Kxj6G71K6dYkArfOE
z#zd{n0B3U@RlH^!#sAc18{FeX+q@X-4l!+?vspk`mw_Og<}r$;ZECga75jzQjjl1>
zu<M&otiVW%{=tXrKZO3Q^sgk_<=f_@?EkTy+Z7NDiZ<cvodhxaqH)-_s0964x{4>X
zUoz$EKj0T@<oUq!AAZV5Pi{Eg9tqK`jan;1D?8^vxS2SwR%6{nnC6L`6Db-vDh~Xn
zY39i-Al{eSS+Cx%Pb*x`3txWuf`9Vk2P8k^lTV&<dt()@r8T5Berf&_JjqYFmotQj
z$CRcUY?Yi7RUl+z2)jh4+pN@5%q<}Wj*}1%iOU*jO}U&OXsys{q1MG&MZ3vXu;uny
zl6>xAmAMj;keqSPwjR30RnCcWIdfhw90`Qu(G8!~7%Ww5Q&$V1qkB+KxuzVgSGZu{
zt_0$g-9n|Ytd+~T(i-HP4RjxGc-4EXw{86|b^Ckt{X+dUc5eaP&|aIbu?EIs2qc%N
zw5BXo%?R29ZizCs(VQbU$80^#Lm2n(x9<NsKbZd{e`5Z1w_Wjoy1`N#rNL4w(sqS<
zl4#2Cm5pMuoZ2d<W?e!ert8ySn#QFzuat%u)QrWoaIdyh%_$+RQHyda3mnoL`^op?
z+Z!LS2xRM?pu5~sPxYD3$0Cr8z=@qIeD9G3@l^BMe}A=6`Dl(rM07A#kKeAlKf9mX
z>3pVWrG<9=d0yQby46U>0FT(tAYytGum0`VpsM%y?G?jnI=s%ZDPsOw%W4MD(bd0-
zi>&FZhNpXW7jo>}{Tg)bo!PmhXueW76P99f?TZzU^m9)!l9FqM=mO$60GABla}5Fp
zvh;Hs6|iFFryFQ9zNZCo(1&|zjkwpm`mm<z$mUJezoT~O;NNB2IQ$rofK1o=bT2EA
z@trGWT(?#mglYnKc10b@QsU|iZx1;7*S>z5<m!9$AfaEQAVO%@Z`_4Pj~?djR&Td;
zOTE8q=+(jA{(=G1l<h(d5ZD8WuLoM!0iM7A{cpKWh4*jcpB?&}-B+V+5BT<n;sN7#
zXA_C}$nnWjZlB$FK~v*&E?m}?%VpvI{><IO8LG22BBUIQ2|Qm|mX+2TU%a^IwA!4X
zQzAq_tl(=cmCs(hMlg@R*RNjj+0S0^@|7|@nYmdH97;r}tgTX{Qp;+=ni(fiN~088
zgARu-bvST3UubP{R?C^XRw>IujDf>^BvZWDYUOS{Arg_`RwYI&?ab#Rg0+N`Xf5ni
zD@~iReQW_lUlUSHC_<hNK%uS|BxN^hH+Ic>`YJii4gx`ftr6X`HOAl`u!(7&0PAqH
zTo%@)x}A#|E$5uLEDL!)P%aOi_U<WXs9l%MeoFd!(+nZaha)j1T2qn-uYF&LK`7Ng
zV07KJ;3@vq)}os^@zF;n0Vt($TFlj9T`FHZ-0?hAVz${9e8W1I#!`*Z%h9!r1l0E}
z%X1i8y41$~YTC{0&In6YR_7vmkQ`&+CcB4ZD?=K!RFffvXh!=DZ>9M?2>c-9TL*xq
zgW74F-g0C6W}`h{Ap5lMH{uN5y|c(WemC|n9dtU6H<W$|u?PR*;hvxT^e4Q$dkvo`
zz1J<`T|%ULaf$1~>Q+cFC~dj^m<}5D0JML3pB$>a#hh+FfzCXEN0$5diCi;5A&gZ<
z#8|HoJ)N5~Q=Yt22ksJ|NO7xPRV>=G)NdHqYE;nry$f;M;rnK9vYh5#km6t8MgGXB
zcI{^y{CDOY@882;+dQ@By#Ww`7Q`cK81;1nxHtgpATS70ZF^|j#r8|N1FV8oO{;q*
zb{4c*ge;M<v2^9Z5OF^m*A_nZd0hh(4gJ{$6$10&KzX=-;~0Ko2D%>sn+;dX_?Y)q
z%I@{0ZyV!yZv%n5zjh<__sz_4-N*d?1_0DQd!)j4XVkqBTWcO5v|S;{o9FUf*+Z&C
z?$$_qT?oN|&ZQLYPYb8>YTkQI2n~ue>W#o@6+U};=99Y%zg8Px-7l09NvW}<K!~QZ
zw5}e@s6M@%&McS0)91H5|2Q+vN75X~!BT!rjip<?MVR%KHCaPSzOHx`{$(-0z8>iJ
zC#!iUq7);E%oHck$f7UIzQv2QaF$$KBLrK-f)_<dh<F1?v=u9AYjqc3N7M58+S~j1
z9*)nfyU}_frHQ@uRt%9mMOrmRFb0>H#9*F#3T0iXr7$06Gksnw+JsuYGS~MDuvBaZ
zc#5y%_z$FeC-zPt-R$3hd<a&QCsEw%wvohItBnFfAVqNoFu0E@^a)p6mtAzbnVx!v
zA1}YdpOwF0k_upH2JWpZXJE<`b1(ex>gW(6O{`U1Q{mi{D#8?$LrO%1C~!uHw_yxc
zKysVycbAf^yNa*dD@vR^SPW{w{CgXi4>G=e04TeKSo+4#*E1Eqp})^4V!U?S&3EL9
zK(~Bh5AJ;P659-i2Jkuz+r4MNMAo(N<?<!J`^$gFUJ}4k4&M4lEousECcy*Twn0zW
z5!E-X@jHPSt~c*qT-ushy2OY%1z$59!y;t|0@#^72NBdQBvMSIoJc7f&`W86R-3GQ
z5R0>Y>fm+*@-ptFU1!7>Fu&iOxoKjIUQYvyIEazQccn+{8)0a=%l+@Kg9tkoX4TSf
zd{6!EfcFN@l-kcvH$Ub6x-I4q2hBQI2ko6$wC*ZWH+^M)?1on%{n__4Kk7Y$EzPfr
zrKa|L*FIQzgFSrT=%3f`|GQuRt|08~uYC*xxAW<zD##n3cWuX{J5|Cr3*_v$I<oa5
zGMI7O25sI331*!W-n>H6AExiXxHQm<a<^=$Nfb)6;)I+NQ!p@B6qdEII3T#x!0Xiu
zL#87k1eV%(SPDU``qFe`k&v<_UDAR1sc@KtIYt9MA|9|fuy2j?abjK%JggULGei53
z^N?1Q+C1P4mQEBQq-=nvnvRdA9Ouk&Hsj~zt7M=pW;a&$9^UKfz%G%e84yaTTR;>d
z+5$}>rWw_0zJA*0rhS$~D?CW^j9xC5s-6-ECb{umzVn<}*Gh~w3)WJZri22S#OdM0
z!<}^;IiFW9mz7d22Od*4y`jS4FdO(CaiyYKhU8|f7V1X%>T3h0F)nK%Rg)jhQy^I<
zJd3P75N~dVuXM$%1?Ekyud`|VyfCL(dS4J0Ia`~|WV*b%z}c&tLogOIrCxGkT1SCh
z`%_^~fhWPcuz-PP8mZzWf>=M`U=`-gfv;5)2bbGevDd1NpuwaH-(acue*CkKZ$Aa>
ztk4jyvTamb)vvJEN$J&Q?(cigTCU?iK17DT583_xEyvhrM|TI<AK4<eRteES6rt_0
zAvtBU`_KlfFVI8B<P6fFKX$;!{iS+avYs9u9&(Ndw5fwb2J0kVP|z1+ON;fBQluMf
z250p0MC^di89r;Oz5yT8!^Pd^&W^e44BiIF`?ZtZ&-!B43Dw@BuFw08{ky)>;~)KN
zu+xF<K5)OUcQDiMpS3Qs1;00De_TDi|DgW##kjRr8fax)`*y{kn$AzJY*sdw*Bt&;
zx2oZ;2esR8S|JeEc*X%d*Ja_vn-g+qKgajo{q9==IeR?)2G7?iJQ)|{>b+kJ{=UKK
z8tCkiKo8UG8QRykx8|EtleZk#0MU10lo2EXC|S*-!!i%zoh-cVmizO?suVTATre%G
zOh-bFNU2C&ZN2JiZU_Xw)>Q?F)G1PuvGM&mOyas*ZA=OBG!t7ijbE#zX*QXL&qJ%M
zw-oVwbBhuvYoV4xE2a}nA%iG&6_%xfMDyrN6CtMI=NGaYIOl`idoQLcWg+JYkqPzc
z{q<s2Epa-~-0rHjN=U{&g<25tb=On87W`V^XU6x9rhybqYVvTpkWzGlt#LV@t?Sd;
zsAco<i$^o2uT6<jIo>|u^ztiuTRRKt?QU)k7@K){^?R+Vqs3;BMvm!9_gdSQZWYhK
zn0tYG+fH_E=&YclR&~jc@KEo$dw5`$1Gm`#xHe12rWl!HVvZ4QR`}8V{44;rKFEEq
z9qtMU=UTZe#mEL1E?0$zT4`%<ed_1ZEWn8<r79}QxfSz-{C<4v<68%S*5>Sr#u=Nz
z-qNJAG<$F;9Zl$?)%_7G)Qvoi<axB{_XSQiFd+fdqZ!3$^5Ot&53=yrh7d?<AmG`v
z=jffCQ=WX%pu<(3KB2ow@xCZyU!5#$=GOF<VEv4SQ{s641|l3l`sCBr%yhbccQ0Hp
zr})e?8PgXA@Z0o#k1KF{-^R_iZ*V*g=-%ADADi=bOtNQ)WPhC9|E>?T8KeRuqTCZ=
z!!xU!GOM~VbMtDoa+Qkh;I{+a-i4%zTQT%L#ts7amCc*aVh4f#9Mwi`D>C7&=R8#{
zTg7)j2Rn{z9ad^ZUBqoVC}k}xs}x&YyVeP|_HF%h`*rBuHO58h#VNM4NO<!e?Jn^*
z17Q9U+C9F#f){mdl7G>5C3M^(qQ~Hm-7EH3QXc)b2Muq~d5?m${qt6I)8a;FD!X9*
zyJ96#S~IYQJzFcL2kT!isS;HQYW;xEO|7?sKMUf)=wc)oE5b^Im<X8nn$4?T7!}Ad
zQiHHMbET?SV%au)OV37r`S2Tje)_`P4AfXT90;_gTrMl?T9|K+=BKH>jT;7tuohx<
z4J;So?sT`!zgkf&d(pu~9YNsch!BFU3u&6HdKe{U>sAue>}?m@c)nsxBO^0S#-MgH
zb^&6tYj7a5E~{NjKcDluf;P(K3~JV3)eAtHDlrPjn;8kwdY?B%V<P5EU5eYy_&(sS
z483aIwBDetdN&oTpslsq&(w{cz4WarEZ+Y)dwRN?M0e&>z13$(VXV73C)Txar(f`L
zdBL;8FCn2(ib6E);v6H{b>vN}1+QvZk9}96z7C70<V$U=&33f4Hbf%9Gbtfh0ZMPj
zd8vi9HO`kpL|95O*-ebB_0i(-{rFbKw+;Y<dN+ly8E<#_??|E}pZ&)74NpKHy|{-%
z<40YnzE7)en^9-2Y~NF+Im3My@VltFBNvUR82g0C5c&4{l<tq!C;hOiXljwATA{}{
zHEo-*G;DiqFE~lj9P?A6y3Iley-9dv%84mYM7KHUOjq#K;|9k1gx;=|UFvXsskUE?
zlX=^%-Ok_s^R>6CZ{VAsvK50o+P{~};`+Jfb~<LE)B6Gs|3K`T)?$liUvy)P-iaez
zv04V;gX>N@FDM%d?f;f7P|?kK7>#=d)cnNk5;b}&!=CB@hV~`OYC5H82DeFJ%H6oS
zf95`>(Q6<y{(J<bzHYv~=hu78)HmAxT-X%cOBgU;PtVzRJBTfM((CK@mXDUJzss9H
zygE4trf>O<VSnD=bZz)W8L}}&Yb}(fl(5<DM1->!axArRE(R{<DS3M>L-M7n6>v1A
zG}H2lK=dG`0--5$vU{fp#UDl?fLK7&nsxSw0!<E7pRdcZQVHaFMp2g4tg?<b2cq}s
zZlyaC*zrnRi3gy;!}=8u=X+{%Gxac8VD0N=oMu9_O4L%y@LSbtEz^2%SZhH-9Q}SJ
z1Zr7GZcJ<)Rl3h9%!e644eY6mHFTa5G05m=T+2#bt&`K`boK(2MALPoZbOKqoLnzx
zZTT*zGkwi(4p{dOxw%6G+TIO4CK^InS6V4(n^;S=!Y46ce5t}(Y{6{4Hi;@J_JR<l
zRb^Qlu{LtDbiRlIwN{}Lk(g-&S}-yt#z6L>jPtUZgP#Y3MH`n|2naa@s;_pvo5@;~
zASP96)iY(%9TJQ|tW7zum2;^?g%kv7YHXuGOpz31Yi9oa_}0b;Z3Xwfam6CZjt%f(
zn~rX;?QF{cQ`_(Nn|o)~q%S(1V(C;>|6|8S_MdfKoOQQQQc8S0f5NYS`Iq=7Km0qO
z=1FFqNE9?O1Z<CY(p@#}IP<2v^rrtu@wPy1v>*e9Cao`?!L@qHD~_v+stw?oHB0c6
zZV)&79H6gj*!(+s^cx8K)3b9_M&tG?pz$^~kU>hl!QhR%AseRHGn%j!G51?pN?}<G
zrL2@vEQ<8|7NV!OW45X~FKhvZO-pa9)ZD0fErrk;scRm6VM4nfT3}&2X{$iivC-lN
z&Vz;QZ0{&S>6fM#0RdFMt=rY;TEGH?<aZ>w?arPJ-mzh$M)`_u{TjgZXuOSF`?-f~
z=OrEN`FwQV%{z<g0Bijw4=g#p)BdpU`4T-V(BG~)BoB|}B(R%$-w_5dj6l>ba;?qu
zb?jIpplIP#l!u}y6$ycy+2*SBKHOGd(0sn97$7xbvjlrDlDc?jj-F9a%o1d*4HZtM
z(Ig-t6Ox0TwYcfCFin|me^r|WKY5CX%$!PLt%aYxe2K1wFTeVXrZ63!5z;)QG3rGN
zs@9TDBwA%`sg`#3PCYSs*Nl3htSfoGp|AbczOGQ$g%ATbH#dY}MG4v(c}m0>xSSVG
z=LMZJDcX9Ka<X)9t!7=LKunn0EplG&p;fH+uX~#Yp*3MC7n<AMw6Vr-i@tz$`|uwX
z&t}AQARL}D9S&|s13d^s&1z*?)f$ut<9Ucd4Wwg_e@p>NWpyeyC+|I<B5PSW=PwcY
zW%p+E0M-Au)W$^{H%Z9Oz?R(}yQ-~$0a#1pp*E%zIn0^6%gTb1ea|iKPg{Mjt|-S8
zU4t44YJGz7=j@OZtlz?9`hHxEZyx{#@f6a#$J-Y1J=}P69l!p;{`W()-v7M^db<}7
zEaiZ(<2H8SQ^ta6TThOOXSdII8lTx_v}^HowE?yr!{;gF;qm96u|G!<F%wkmPfFR?
zKclt=P;s<k_64!G2{Hk5PkD-TDvgmsKrk_Gv`!hpQ)QbE-v$at8he)-w!^gn9^ZB+
z9{+v+zp#JJ4*a2A)NiAzvaT!3T3E~K9R}<=ALoO6=ehr2G(W810{G2oGLTww0I0+9
ze<`cqcYmEn<bW5sbU#G<CLze!$i{Qwl$=#Ii!E3xo}PIOa(2OqUmfXSP%NF=+~aGG
z*=!~1a@A68T($n_gtyI=pL6U{_a9&kX0o?`8z9g%yB#oh7Erbi?00A!>pHjl=sX_>
za*PM3`JwwUDoT<ag)G;B;4k!04{9(SSO1RP+`Kjm?m|f9SP8XSyDfk7_DGsth^mJf
z<2i@yc1s{#w!QTRhpp2M#ttP36t=8`#)dA&R?(DNngeu3`cG8d3_T(eSki^E!s~|x
ztrZOuIUmeS85j$(7umJ7ASs*WmwR=#b)nQi${B5&v;|)e5i^MoB4||(hXbe6nGkLe
zF}s>lOc$4P;&`07EGy?rG4DJYt*rK0>T7X>kPn`kg;rPlZaET@a6X+m9A^tcOJTh{
zFrLP-_UVuu1t$y?yU#)(%}4U_mU*5@IS!kurk47>tc|6J>+_TpAx8`GOHrValQ1VU
zl0IKndL}D`z<da_R=96p87QO~>!$XMw5FU&VJVHIfsi-yB=|L`Vp+S|8V{GsQX2D<
zqb03SJQ2o}V7XMH1a5P*LKG3^V5#{j=xB1VDDTJyem}mA@j)2Cv2nilHIFwoi)7w-
z*Vs_s9FM^tusMV5h(A@3<`t&W#zJTnO7<!NdJAwtp*AAI?R?;vCyQuHA*|-IZoP85
z7hKyqPk6K!Lhu3r+;QH{SUjCu*VWFmHS0X_h%FqWYgIyUYJicBMjWtEY=iL4q6_&Y
z$i@(cO<3P`Z=EF^#FBrGxDjZSM+Fnx-Mc<ZVN|_!K(HreuI`fqKvxm7mcqKMtkv`&
zF(>An$!<a&1~BjUbBl_{zl|>N?`4}-aD(pZniV{a+c!tMHg*k4t5*QnI-}?~PS*tW
zlnXsA+`r%sVD7=&DBu}S*mluctMpzhBkk@l9u>=&9nb!nk3)PQ7H|^___g!7Eqm~{
z`~E$3K0Z{gPgRgdxNn>P_&ky8pKjm4b&nd@*xdemZyWYq7!~t=D{Wh=x~w3G{jN3l
zm21i&Sdf=pYt>t8)z%PgoEJ+WrXc30SQ=}vb*c9Q?pnJP`ZKSC!3sbkkfL%(*$XF3
zJGQii<#b|QSEidA%S?pefvEZ5&4<~5Ps)Vq&r*!!Y}LE9uH-OVFbp0nxyPSuSqjx=
z9XmoaW9in0jZ~CieQ{&raC6{v_mU{Wv!_SudSN|1Fdc7M*UI6TIiCyX%R+X2bkoM^
zVMPR5S!m0}G{F*)I3XdURas6i5v&*a@p!NxE(9JPPQ*O1t}EKL$$s9&J3EcF+rOiF
zk%j_UtO7SgU&H)sn?;OrSuCXw?v&+$YpL3&4nYhQ&U5nie>(hCM<5N{SVa-$YG8#y
zYg5j(ab8w}B-6<Iy1!PXx8#fnCud@hIg+TH7yG^`2y+l_qflZXhr}G=<a?wA2C(`w
zeXi9~%%yN%S8l((DfRvE@j+~0ex6Ls@Wk8g>U)dkH=_F8#)5*r<-=4rAgBAcX-`vm
zB&=QNYFke8?MHm0NilIe9C`ZWi5<h6liNbvZBTr(8xh?mfR)|m-8auYv!NDh^M-nU
zMC&%Itr|HDuc;0IT5E%9w}o09>;*#Fm@}*N-GOV>0RVfIYZq8+H6JglluKhF+XHrc
z@BThN-sN{-Ww#IxHfL8P0T5Fn=7d!_?ifn8&#1Z`T6Ujtb@!EFO{3uc2?A>R!N-9~
z*GX0vlH+WrvUQ%Y^tw7D;vi^O@z`x(w)E#7{CZH(Ep>ED_f~J6m3H)8H`ea`aIb*l
z?z?`|VEJ2tLwfZGo4NHPOxpK34*;$!#~*zGpB#z~=6C?t-AlFKqh=t`nsza|#I$ec
z8-MUotFji;{-JHyC?!$mph#8DrMTyt6(S@Nq6no{R&N<uYo%3nA-<5ryZET0RZ*8b
z6wHz&N}#He#XQ->Q?ZQ-2^x(*dH4<f+t>dY8mg@;s_1$lghcSpJ?Hbu*Kuoo7-FzA
zXPzi)L0501=Dos2p_Bz--yuj*i5?hXx;X8Do0uNjD{)cNvCZ>LXYPZDwTqO%M;|@s
zi!WaqK)W>WmJ)fmKXG#XV82$iu22`+<!mP+N}O*IGXq8If`}E|<!R<{JOD~<6{$*D
zF1Cj5BH+IEb_NjN2U_>+3)U;S14O}Wb*e&X%K4(Ki@IJBGzQjEu`YP#y9WjkFQt*9
z8(FtzT60e+Pjlw3eKuOOp=R4tn%lvMb%{a}q)18uwChlXR+WdcvMehx20ngj`@^f#
zGHhQ?t<co&_hqfzpW$ICEX%?nNAkfsvCP4$hieO*OZ~nG<O7WlVg&t7T=~DQV(-d~
zqHj9=J+d)v_G{g}&zYCb46$cd^y)aB!RVW=`M;SMHel6L$HReVPoE-p&bZZvmP+;4
zhOklPN@s@{pm5XUY#Uz(9G$s({LQcaI#@Z2z{a6#9qG=%*2*nss{+nb8eK|4xUS^d
zT{wDCg({=J?f6}D7Qmh8$0pvv>7Lnpgi7|`YM&A{wsd1md-}OEnyMIc!DdO*`zKr4
z%faDb0R1rf=o!%MG`HFWyC2<;u^V~!d!wD%RX3ql2K%}%2I)Ff=#{Z=e)DlzLKmV7
z(yK0$F*f5F=plT+9@kI7_1JBY<450e`sC`lzWT*xt<#GLcyu2+Sbq#=2SDr$<M8l&
z<HN6BKZeUCx+kRUU;7I_JZu1MtZP9;Oly_k>11ANwJ23B*lnub8A&}IXBVqArJ?Sv
zZcwf>X2VGV@6RfMwKbO7SnA5`a+BcmLV>12se!H>Y^|YgwGyIbNzM-sph})5M1<qb
zk)>``&4N;j*}O;yR>kY<L5#tB&U!bX5UFLQd3&<xOx@u)(^Obn&$QTDG#^brw_X-%
zHREQLz;sC5-rjOv7D7ziKb$!nCJu**ufDuPOYw?jMb-;iFGy2JGa=tXNY?9kIRR$*
zGELJ~jhqZ@olYkN?E9&=-LsX2;iKw*Hut#{v0|B8RtIEuKi8tHRe88n*1D|-+Io*&
zkhI<Bu3xX!K%qnfWNWE}=Gs6JUe+&pv8<%vtuQf`H6X;0NXHD=IaD>9m`j1BG*)MO
zBXB%StV^TOkXEf2N0nNH)7hj$OKa9H(8-1KQdwvu0!)F`Wl&M%`y!AJH2z4WfFFau
ztAG1mxbHx<e@zDh{b!FE!m<619<ZHJfu6b>uWil<I^)vIP@7lDg_MXP@iaf<SHJuV
z{KIFzWhvWgdY8uL2{$@Q_~*1@X<2&e7Dg*LTDNiW>oqQ#X(SNY&UvIJci#a#4df~2
z7|A&^=ZR^Wtokc$hSeMUa=k$hnB5Cj9$im~n-$F-qYQn}&QowF`+9UQ=zAH^MZvne
zpSyeO$hm`}hVIOiWq`u|F}*KwXGRCpim^dgEERCYMmi|7G{>&%ZU^1tS~%+@0?k;E
z-ic%|cKw=OMI75TP}!V|yJ>Fn=IvKZK`(02H#ITa%l?2U=RI%h^Gq7U3TBG~bOX@i
zPi6lcb-uFO|M2Kpkloa{#(;A5_jhaf-hM3%=<DvpzB?mGAO(e-y$>}A5em#f$cKr$
zwNO=z(aPT1u`v2&PbmO-v@z6Sl5B9WsHJ}An8rg82t;X2?rqw%13fo1mw;y-hgR1#
zw+rQbMzt{=kJQ$<xw+-C6qeO|0nH28+klDn1~vwBqW6QXZ6)P|bT31fb=36&K{(#r
zqN>b?nOe@|DYL8rEf++ry<1s4<%m_F7PXW!6!UMrbSD92J{&o{ddd0WmH7e&%t9v3
zkdK7vh{i-%3tBFoRY^>Tndoi|y4rlbJlxaD<?3;?&Pi|E3x&0;lw#{_5?C&0=6Oa!
zU@eWcig(ZmBdRs_GYo$~g^^k{d4x^ioD;RJXzRITJIB-7_}wp01_0N>G-Yn*iNlml
zLJ$Mz^9rqSoHDJ#vRsU<EQPWz6e>AIUc5Syr^qp9)?&6bty${-cFH`RXDgsl;Zh5i
zWii$|05qQDWZxAb$M}5_$OjnTIshEe>!a~?EXAJPYxrhlzqtOrzh)GBT(L{`cNI|S
zsymyF1E}oI!+^}DNfN{+KSGYlEJc7He)K6nOh4iuDMs&iz}QdWF#zbONfl~E!)5?%
z{#pSEP?y!K8~3|zcZ9Rg2qL41V=OAZfQOjO!z(3n@{S*|``-;TdX=U`y0;tVrJ+!3
z*WP&hCiN7B@BAIyMh&B3x&qRw7tJ^jW)C*E`!eop>*;33qPP{$*vS>Ipc70dXU*X)
zZKJh$MO$KBOhcJcCdEkh_I8%)wN$dUy1(%T6nE*{Zk)X@{-|zc(80PkyoHtw;A7v=
zTRL_0^caBbV;WCECkL!4MxwKs0kh&z6s>IoPs0!J`oMC1==U;T{k7ttcCT@5ZvZ&|
zPu+BqzNd`uv;XWNeeeU=YwZ=i_xBzLzH|OR{zC6~$s2j^-FXGfe{;0m`UM|~wF(r0
zlme{=OWA7V#ih`y@^rSsiI9Y+HxoGrE{k%ii+AVo;07T|qH5jz-I`c?uVz)fs@1fA
z61bTYQD9DiRZU}7YU47Pt8E?w#uA3aaz3+`mFaL~T?)5P4=n4#!~NOLA!qj;6tipz
zf!2g1HgB8Hfs`hzaC1il5i#&5*c=bS@#Y2y%tz}GBBBQDs#P|Z)rw47twaBKS**}Z
zfOUPK9gnEO`SnY)B|)iWF>n<VA!Nu$;xrlasTa`dCi{sPAm_+hY&}^nXGE0cbhlf*
z`;yVzXd1);b*a{d50})mW)@PlHsfgnZolIW0~K$HXKPn&)~2%e8E#G8_C`3L+eqIo
zmqN}FjRx4)`huUWzt7Y5BU-CGyA7;Tz2~%|Yvum_LeR>DvaX7>O4ddz3+r;Se!;DQ
zn9jYjA`PY}tktyDPYx4@oVhz!s#K&|Pvp85mI{*yA3eY2<z?l?-GztC%JY9d)9sJd
z_y7#x*Mq;WO<Czs`|!5i!SIM){rl?A{(pOLh&Z$38+<>{<}Al1E~DZ4G_ghT_IS(3
z&z`~3t~a92c=Wd-ZvKp!+_^TX0}aDdZk)Okc(a}6U_LwE*bJNd%k1yH_vY65!#u3S
z5F6@1wr8Ph2XcO0TT@zT!#X8MaHHj(u8{4tgsXq=(;0RFyXB01*Z>}QF~cLaum7tx
z9ab)7DQ4{KOoh;e=$0~CT$A&-;9x!|3()Q#JT)^j&9k=>^JmFZ5CX$q$H-Z0e-vXJ
zAeqp=gQfeabr@#%f{`~WfOj`gcF&>FECrQ73)R|dIV(H<?7PM0Z~SJe_0f1dd)_yj
z<niAhO)2>Xf9j(@yWnwX`^K}n@$;&<fv~&@-gaYq8g^0I)%TaL{er4nAn{!oug?o%
zS;#?10dD6+DK45vs6_6T!fguV0Ix4AQ%mL>=VIB~l+Pc|L`fWT!n1{}O*ogzF=vib
zB1W4RO&f7YJfNzr5w)4wbZeE;D)*;_T6F}^F=ymb&{`2y>arp^Sf`u&2P9<DG+VD*
zfps~Pr<tz%7VoRxYx81GSifR>lFGU;-&lD-N`af30}A9Mq%=|1${G`-WJM%%HlUU>
z%TkbRQkhbe^>U$>6R2tRLQJ&f!m8`=1vRII4Aa5f7eYX4rL7l4ggi~;9GP--JF39q
z-7c2%g9n((*oD;WR(c6uU;nH~0}nRCm`_dBfZ)1PHBeRq>`w84cKa-2=B*IkCjpAJ
zO%5z;qZXyCYS-|Ppzmueh%_d_+PDd<m&#HX&THX(cjgCAXINIwcMF2@<R)?|R^amd
z$w%a11$?RzX~JQi5OGa?2rQ*>E+_8pP9Tkkhm{p!spealm^szPi~9?wv$xgTxsrW9
zzMb(wQoxV(NskxiH_}@__rksDk=}f(>)Gg@Z#FR;#rWrMo1u&?i>uVz+A7m@u&DTW
z<VQdJA^bi_hz6+32<QL-002ouK~zvY8r%3dJ8;+H5W{e*ee8KifLivIzi({Q4!oRE
z>uk{OyKdfpVSowfnl$$m9H7syN!`-My{fReRs|7v?6;JXNEqkPiyzok%xN`7OT*3!
zk^T4VUqUzFr=yI*?udV{){AQ#_<`vpFYAS6SrG8z8+Vm&ZM0P@_m<wW4j_rkWnnHW
z$K#QhqAi#`py=RqhtRrBPQk5n*wV17ZPO!mX0_So+K<=uas8(0)gq=A1VlzR6Emi^
zd)i7NbvD&32j2u+e^^6xQ$_q7$G$orfnKZu+x3JVuxyFk0Tku>`QHa+?EZR{0{#Z9
z;Qsx)vYT)Cra#)_QYz)Va7fXrQ<uuklvxl;vld`cl6kka#@)k0ip7DGaH*BGR(}8G
zJ(0xIW9BAjj@cN&*UQC>q%#oA|4mfM5mK@0<E5(CXf&2{0j*3ikjs%9{fHOk$HW-P
z^MQ2*Ii1m3Sj&Z|jn@x%NKE9z>{d2~WjV2&PfRy?JW$i&03xi5Wg+IH*{$@AsvACs
zC>#zM)B!}YptF=p2$8iFPbphwq^$Sk<AKZl10iS5cdseuJG8F*CzKQt5+^IR3lYiI
zl8$sB<jI4@N-JmU`HK~yq!ftJ`Y@}u(Fw|_24_gOfIAEa2WR`Vzt*h4Lbmm}td)?n
zn?gGv^oU9@hER(Ree_!ENL^cTVjxdK3V})^q(Cc5Su0zQ;1jh}BnDc8`@08gzXvQ!
z;dq#NHYX@6%l!#NxJi*o;MvU$H^&L7wl@`pDMW%=+qWnYQF!tCf%Exn#XA8$`|8Ba
z?F~1_nGl590x!=i&u%AT3d}iu0Fpp$zpnxO0ONyK0e_vx&y^-mb$ETfZ@dfiZU16y
z%$tr&H#$pY*La?*2n}2cSlwY>gu`Lx<4-<e`g~@s6_w!3)dn2tX;$}dB81(~ckG+9
zNpI{`vi&jk$iQd;w|z4<AUldsx~Lep2MCnX%*51zlU4b;HpV0G{(&*L1%cZ)TrL-{
zAj{+_j}*nanonOEK+t<oI&4z#E)iSRbH_*O3%+ozGrU|b6i>|zpj)?oqU&(RS4)e!
zmtGEm+uK`eX`Id{>o?kT97llT&GC%~W|)rqk3*08Jy4Ngb|S5Alw2u=pgrQZ_Ezdv
z8&?nA8`PYDo6Bn3(!hd+4YgKCX#-C|dAnZd4`*x@_t$go4>I04r3EVlb1JFJ0_?Nc
zVBQY+8H=|I#GDPZ>ty?aym4u^IWLMo+uscU7(=8u^S{)_Wo;A*%-OnRlvbH@w*9gx
zm!jOCFMTmHYY;JAl{U<?ZB^PKlT%=6u+%~fi75mwRat677n{!+@BR^#LlQ(PS_>gh
zJef23Y?`!bn#ehmXY0XRTSZHumc`dlrJYYmjHZoi)jHu+lWA!8I1J|NSKY@GtUHA1
zlkIt(rewuq%R-QeH8gUF7Hg+KtBVDT_h+hSHmozph5B$uD+_hC3f^v^5_%8f`IeZo
zd5p$j=jPIu7=`tG0a0>3n9fxg>%SGDESw)6P`8uXF9w2pA%e?zHtUz(l2gU(cqF7z
zbXO~s(yZvi_mx`em?teXf964_h$LHAmbF4Ov+tOK`3TmAcHh6&Skgb?$ggp8NaP52
zclYGB^6|~gg9^vl+PSp`Pi`^+?_8CHIa*;&PLY68&S$7fbYtzcR_@P*)3Wk#T2SEi
zrEsdsvpn<U=E$W~YPDzW`7v{Ue|F2A4?NNS7>*B;0)AhH_nHm(XW5AQMB9VEx8;x?
zAJ+lM6&UO(O#?6i^OahQweHF(^U=-6{F@*D`~1JZ`hS1~vz+Jv2S)04OzIANd(ndG
z?o@QJYYafx^<>-kg6s5ezkVV+BWLLzb@tY#O7%8Ir4-h69o98_9ND~QWb_pBG|xj%
zhX|pWhu5+!*0Ca5)oLjPP;$=OecLmcZ^5>YM+c+TnZJImnww|Glt?M@>h3j%!;!m}
zF9<R5>8GDiO6BhEH6p_CaAcY$PUj={clTVD#TZL98;#^`oMc?*>uacPhY-D5y0kig
ztED&XZnV~D)(&832Gv2wxTx1RY5Tcva;-JgI{=7ay;rM$fXXg?x?_O(!(S8s;DB!{
zbn4Z*g9&8o1k!`6t~T60clVCY+Ua)eO=56e4ZLy28wlK_3@}psx?cx=u}6%FTH$3W
zOi73-aKAJ{GKqt<Mheja6;YlXXO<=M;$*E8qgcnFkDuIdS<K!ghQO2qRg|YU2U=+q
zQE~`O+4=@wmPQJ2lOrKy?#~OQTFb5=<^d;tz0kxwPndN}Af_23%ucMA2U@dE7%>O5
z8DP*DL85)XV!`$0a$%Zhx)xi+z(!wJnisH?DAZ-4t%X{ZFg4C0psU&5NIbBf&nOnG
zHiXmty|I$#JMc_L%CoQEmO_p(5T^rqo|)%Nj0UtyF_2no^>%Q=d@~yeYZeFxOEB1b
z)-!dzYz+8V7ItzO@%=_EE!_?OBv@tjloFj`?A>*Gnz%pbAXfd`psaO!_N!V;(kRrj
zr4w4S_4LUxG9?4XA^e29@K>0pAMzx83gJRj;Wqp!&dt0X^(0IZm}1~&3P@B46Q|NR
zKdjVLj4U{x(U`cLE9?2h`Eusvs|BVTf_H_viNX(m@C=EO`;+-EKDoVRS(TrDc|y`e
z${ET(pZWI3X?&n6;P-D^@2vv<Lpg!IuL@Urul?us9YqAuuzQ97ECkkNCC$dr9Ht|G
z@lX6^{<GKrPp9`gbJ6_-`)50g==!#vrnMZbRR`DZn*6x;te^Y9mR2@5jya;nU6DuA
z?w#h*WvN`w7iuY1MA3{%i(b4@Yh_&)bF~+tI=feDwMyb?CQq5159a?=tV4&?I@+;y
zHZY}3@YZQw4~S^ouK(lv;y|cJ=e-a^f~9=dvNGq1<KaNgnbJ&jeL0^fr8;<Udj)@H
za?Yfj2_eGWYnEjJOd~lMBEO&e@Id_85Mk&5YulXUI=JWyxN39l)831*-?J+^c>r{K
zINH#KrWD;~#;XHk+<n+qsl8LuJ-&i}=;NEry*K7_|JpmCHh`<_tYKg<z#9Xv<L8f$
zH;y?N+Ohk+`9Nt8qJX+*`hW39PG)MXP5Jc6EyT#{^U5(rp5zQfYJn<Hweb8HDcSnQ
zURGt!i4=sVQ-WHVb0WmR6z%y`5$?|mUz`>$3P+bgT<XfDDmezIja8NPQVElCjP|~@
zRjPWQZ3$l08OX;QG-N9+j>Xccy`OMx6g6Y#m;<#HH<yR97S_u`%oAzK<jG9y>)O13
zt*tEUd9m&sRjF--)qo+@id#~;VO<yM{hiIjRw>1M@}h*0AVk7+uwKG(CLfQMN)AHG
zc@)v56ydU37G<s0PEIUqGR=w0{fX20!aUDbt-Drtc4!;e+ww`{z6vs0zac_Q$@F^?
z5iw~<2v+%9Dh~RveFIA~$CP1LnXTd(3vVW{R5PBoHO(x8VlZ2m5Tdn&YbU<qHD4|N
z#HBzDQ6beMni3+3kK->h>j9NUYJooyf1OEY9$KT+nc9wAPG{O$xLX$1^O=_~K4W=!
zAjl2z>Bq!rB8l=Fzw$%oo0*^7+dgx9b7V^9E4ZjK9ToKZ{O^Yt|6F?kzumO?5jgAX
z|JSbNJDOdt3qHo*9;IL3LI!%eX}94HV!{^=5!TkoIrHq<Gp0Oox}51fIgRlOgkJgB
z`&jx>HX9Z}#Q?@ut+?Opys%)fD>x8!mT#Q2Z0TEHAXHpGQfx6@E*H8kFNVl`n3?iK
zYn9XK%*tXQwpN>@DVYyf2uzddTxu=DTd>rX<+8B0LX3%=?J;Vt(!BLwZ(U?*{Ktvj
z?LO>N;bE61@9Cfr1372!U=k>+eP*kTS1(^VAhQ%yXCc>hwVtC61h)bi8ENvXA{LBs
z>$oNauWl~lIz&80RmSUe+}q;!Z4dtZUhccLAeIhqYUhv=H#aw9<Lee0{TSn!7uh|7
zyTiX@{5-MKKdX1(z5vI5&=0(O%_GiIyN8T!uh>2ISMPbF5JdWwe+wAigT&o)^<5ZG
zr^M~k1B#MzL^aW}l4E4f!o#U>S)r(MDhn}Y<`{T(oVg2Stf9vofyfjSH9``zVp(eA
zkP>rTS=P!;h#Yg|I3*s|l@^4XI5CG65@cyK!8E@WLYVANF(1Zh$|!-VLOR}Bui(}w
zWkH)|3_`FqMpp~aO07)ufjrr|RLeq|ERES}qc%6QE~_)o7Esl?vMwh?qOUb-E$H$A
zBv6-=CtEyti^+P=#!Q@UiBq;>mD(T#@;rH}x4S0*BIIPL>UBL4#K036JGq3wvaFP*
zq%>33l~z}_)upfhds&ehC|#{uy|?M?V9?soc}m%z1J!PT-8Ih<yg;V;x?)R=kpL0`
zS}W_Na5-IwDUfp_9wyeJyl#c#WQ9X123rGF>D3n^D@diaGcWY-np{W}ZNl#^e;f1G
z6>6Oc?LhV7nUC`?v(&<Uc;HET#^IRw<kOE>Ya>b{#lXY4u%1k>Nm1@kXQo3S<-py`
z*PKr$m_Eq#`eQvlH~_r%c-zK5i+Sy)Gu-$Zd5jvhS79}0A>Q}-Z!&}ih$0e&wl=1i
zkp-qaasTjO2Bt6!q_-yRmbU0c3dTG=+VMIq@~%xwG40p@F1^^Gf7{Mr;hU+6p_k>t
zx-3>C5dtYjj)wy`H@AckIGxPMbaKc1^>X3%_6f(E8}9Dz?DfZ)Ac6abvx$@4w{AYn
ztfhF1C;QI5=Wg#PVhhA|nr~ol|9QVo{rY#UU&?8;uIm-qDJ5Q?UYjRcvUGxorNriW
zHd}*rVOdsG8#zx*ld+xs%8m%v??X?sTXl732m0cQS<OU?o=e-9#{Re%Zfn5i@7Qlj
zPydguDgODEew(;C9(j^)`8S{cJ8YD3fBVKL8!Ij7ABHM=2bA{U`|Gs1?>*#W8OQq$
z^9acf7%to9*n_oq0>Q5tJ@EL>U_c$65-GK@wqk*xHqb<)THo0qOgVZxGvOE&B#m|D
zCMR2ufSZ&!Mj-{|tA|2Gc|K(WgE7zqesp^Poj`>kjhhJ1bEdVz!`etWuxj9!*5BZ(
z(?9iv$h%Rbty@t{nN}(>391oRH>{WF;LKRNQ9)koNXV4xtXnBu9v*0}!v+hIO`luP
zc5xYpF^erE;xqwDIX^h#mr2vX=amqr1JNy9bXf@!UC*q(9t@eW8B&Y2g^MvT9}Wh9
zRY`f`ayqdtXWGGxnZ*OgzIW`?zWq@ywVBfakYe&~NX@#XczZju$!bJN7J$~3)+#X_
zx4_A?N@MQ<sH<JKejdHckw{=&DzykP+5OK4%en-yT19)_+<8H*=mp$QGB`P*=8B;J
z;R5oC7>x0JC7)Aoflq|{7SZsWl)vCd^6RAflr+t(7h$=-C+CTC4V*6v>0CM9WKQS1
z?^FQyM|u2nV*tOY_}q$Uy{tfVW4mZqi~I%xuW{7akff)R0#>H5t1@&zxNp@qhJCF6
z)FQo@lTAW|X|gayYum_u5t_7NoQ&G<dy21T2)haL<4tV<q7GUIxzyPt``#UB4(*?x
zd24N$y@L6F<z!4<XVd!eqMJqsF|EsGU6J!<v65nBnkQ83yRW4f3zyCPepwcJ1+j?h
z<3g|A-7%Gd$xyUgAuzsox5a6#nFWg*E2m`Vxvncx&F^oTX1Yc)rr;ekEMo8H5#3je
zaeeF>b(XC!#GQ?8sz@_4Z>^i9OFu;uS~EcIXmmVKpgIzi(QdA%`h%NFH`gzTS8X2V
ziD%PO{)f;1J$_^Qvk2;-^d0Q92txG2kZxRi^=sbBjE!f?pS4{t)$zMMBmTzczU5f&
z0E<75^~&|zcmJ>pel?-TT-^mcZTI@#f>9ha7Oig{_ZN7atM&SQHf2<V^ICZ_M-C~l
zdiDMBkSMrWy4cU1n<jZzi&y8B^D0zK7pO5%{c$edtM`xstyS(y<LMN5k`rsKJk7$9
zNGk<VVNMwdi75t-Avt3$-YsW~g(L`Ro@vRuLwFT`tA$oqZ=<N>JUPo=QMG`1T`uGx
zEXzW<T&xR7EtV24tF!i2ECCRrY4~bcskIPvCgdBYm}%=mJU%7l39Zf8y_kq8GUY&?
zlCy7Wz|MDv+6yb1TIUZf1aGI7voUr#&ph1Sp|x^)^`*6rYbIdp>|b*;TT7KYo7a~T
z37hRqU=(xIQb^N8ExnKO>Hya@!mYRz#@3w8ht$_1Uvq*lX6MstS)81h<_rR>T4xUk
zL~*d%35iaa&|6*>JO1cgWnMk8&qzsjFSHR;FP?#g*L*Jbe5U^ZVkXvGK1#pJ@=@T=
zKKXOhi;~r4Q4c4M!1R`N>ihABKR!4Be4FLV`!<cfrP)KdW6nC`_0|oqvy6MfK}J-G
z>~&S!dF(&4TUzY*`4Im9?EPtsEz6eQhy7MWWab`fIK!RZ&;z^sbvK)>ZqiM%yGc<4
zB?vSi82;c7maKs`C`f_@(=hnKGA!6IECDuULLy}iq(BLRAi#nF3kGEfFeF%zCCIQS
zkrYif$tKyahxgun_ug~QIaRgy&di8d{;(o4bMLA;=e&ORo_<5Ei@K+3?;Ii{BQyT-
zUu*rBm8)sn232L#Zuq^Ie;0r5oB!9vc?GLAONr+Y(}I#YuXtj2a-LjP^Kx!xG3PcJ
za5c#gH&rUBD)hd%PIGy1D}bykKB3B+1Smn#6?A#;OKCOBw6mBwrENPx)1ZiDXo<0M
zCa@ced%{+gIqP)oHU6X)J3qdT=_5=cWO=m(aE4F<t5fNC6N9dxt?N3P5X!1nrV?ss
zsS4tZeb54)t9uv(k3Sw;(~4Zn0%19tWNLe@B5-TLvb;Y~sDdi1QyAcSO3jE;*uRFy
zo5%ct$A6F?Kl@3X8i2KTcRn7{6JsFrsp`1DAMYIjtfR-PfU8@qU14bNvr~6(1%Y$Y
z(VMT>-<W7z0_^!iWLfub0Sb?sU$*-mux$uvW3k66L+czX@9|C;QnJN>^Te!-DYM@t
zR^CL#bJfxW4l&Y?HqKsZ&mr1-EF}oeLl9aK#5)ct(fL5<J)Jm)82NBNa@Gdcp(&Lq
zGKom9v!Z}=R+L$>G2M1;DOG2(QsNz&94<5nFr@wTo=qvzgfNw3)SN8QW9UsMRqWlG
zVok`hcaY-9<=r7J;6p<y?=kTn2@T>~6Lc<a)~r?%uZD#*zSQD1yi-J!ZO+Lxn>R<C
zqv(X1+>0+7d%*_|hdq)9avG*FYs0j78p?ZHRY)bNE-SQcKuVo;=K`^KRwSnabcGm4
zE3@YfxB_S#P3urRszR>JX^j(CHMuDfyqLyMF6BHeQN}>;Fq*!k^R?_|hMdUJ`rUL1
zGo*-WBE^1UHiHkPkbok-MWql5uAMzzQtvQ~X<*DfALdVB1AkHe5<YZv=Nmq7KTI3~
z?=@d!-Er@`?ZG}f`z>Ptmy^b;CLYzgEv3A`a;Cm(zp^@-CI&gejg|v)PiC@|ho{ag
zW#v>wqn)qm@Wybh!-}u0f67*LWmPmWEULvbDE4n~Q#@GBaJ}12bw|UT9j8{^FmIt9
ztJe0_vZTxlw#qqQCd(P{86H(2Pm(7oSp|=_?QqV~w(TV8pbFy{NkuQvmns_}lvTGl
zRA%>0R>hTjTh@>UwNEDJ`B#}hVU2mgQ!;PM%X4v`n-HK0CaGBVF^nh0eOqcImojaq
zU}}0XGGmCTMY8x;=0Q{CZH!DVk|tHMVsXsrz5`YUsUA<Jg+v<~o^GG=cRl%o{P^~_
zmr5WuHgXlGF17@tm9MTkj<d`E1eAW(R@XF3me87X`<!oh`qQ|;<I{b6*YLgQI%*qF
z@=Ob0-#C)MFh-)<(xwW#2k*+e$(baR7n-6A8blaKfG4HQVT`n12%>D-1_wmVoOdfO
zRv9$YG~O0331o<PZ%9ELJZ?Ll^@mcKsJtiDJR3RB&x`khW^<_~ReaOXG{H0gMad<u
z&s0eY0#j()hB1ze!-1p4SaquazUdHg#AJZ6Tdi@yGm7Aw7WDyCaKXT?D3+}()xD{t
zXh5#<!Z0S1iY>~Vo8EglMRK+f7>iR>`a@6KTIh+A?Y<oH(Y>p>s>@IiMU8RIdHQZ%
z;Msz;YN3)MRED#J--r(&j-;8KM<5k3Xeci)doOxVM({o$&d2~?g!WU+9P)^l7T?UC
zs+<!V4~+YlXc|xi7Xq3hnr6?Es?}<a5+~0MV*>F;IG}hJ3R~%>zVJ?QqKGR%G!G2g
zGc;Gc)Ni0I{DxoQf9-Gc-+grc**{DVc4Ti=2>SKez_&lhR3=YPYr@?Dp6v4q;O5gU
zbjT_iocyYFALa*V5sx%aR#JlcEo;*kR?b|j4SnC+@|-+XmzcmoSpgl%TP#_V8fbDy
zpG~JPD^-dv-7=;GnDqZ<7weMcnIV)?$_wS1@!A-?XpAlM!ICC|4>PbV?o8_<H@P;m
z<`Ro!=c#UBt&D&>W`^o(d(P%OC3)s+KB_BB1zvVHs@U~hfu!E{T57%2{;e#dEbsN~
z_FbHKzSrg6yK=9qj@R*6D=XU5d$brwfK9+DO#m=B$Es_2w0^`-U;eDQEKA?T9OrSh
z?3k(Js@q2=;9IXpDg9l}|0;m>o3eW_)o%v;z1CjAYSz<Cqvr&S&cA<efxw%7B(|9*
zSTaKAEju;gNzp`}L-0IqTAHG#C*D9*%;1YZwsFFv&5FjsDpZ%Tg)6Lsr!8R=PK0*V
znBJf&XPY(qVPK4cNS*-2J|_~6vu;DX4`s1wqt82A*wpt&6FlCP1PkYgaU`@Y;!K0k
zr1~Le)A)2!7_9<qbBfbqD)^E)@68hsnwHRb;&7PWKS<FIbPmm?b|59}!<CHcx_GWK
zP6#E@ti;H<G=+j#ES=&SdoMPPw~&}&u=hzEk!<eSOmSsrCeIVaJ5v+F%yVK_Z7ilB
z&I7hck7*#}Qr-`6c$@yjlH%+=nlK^ha_o}ag7}FEG~sL`@LcQhZK?CDcKvx0vbPW!
z&578XKsTDmDri5LE+4h*pP|hV0y4!VTO6V0Xf=#QpwoqCq0Czt>uMZ?y|0^4l`%3B
z0Ev9{W#;AjbA7BHjJM1HR$%h#ReGIRE0e<F-&eobXCg)xz(T1d+G?c+IdUZ*?KlBI
zW^6S9DJut8Bs+>5p$&n}dc)B7^m#B4xU6<xfQc)BZORE-0Kln&!~!^Iy`b7iS=8MX
zQ+Th8UOh$yY=$L^I|fBde1%phRl%VYpwqf(bvVByLy=c0OG}x>WolKBy24*x>_qpM
ziTsGY&CC0G3++!Y`^v&i_jLk`$Kbij>P#J70HOV=IOIq;yJBov_VMB@%hMXrm8oWD
z9YfR0Tx3y>sa#j>l<IY^fJ_{X53Jgjced~FdmjG+ZSWS0R9sCtCxQzs>1siy@yz0Q
zQ;T2wEZuLn_PhHO$NMb5)cL;L`tEi1Dt!6|hdJU{PM+WIspOAh{g~+c$kVoC>l?b(
znE8)~1Bt}8Gnp5Uu`GSgoCQxQ&PH&ytZ2M1nt-;r(H!Z=$fj)&lp!fiFhKClYh&G>
z-`wyhpCLF7X$0&Qmqbk5H)h`7{1EG~W~8^rLTs_(ywHee9E-OjTc*C2*knXXH4TWd
zM{~r5&f<~tNYJi~vuT-Q&qU5R9|-N*7JVW(-&oiPCUeqk){Nt5Krgh2c#~Nv?{6)P
zyLVQ_qt?%LrIcDq_Pjf%MXqZk#gV4OeAf`6lp+*Y=rHzn!|Lmx48&Tcse)fr$pvgm
zdAFN7fw{29VMiQ$#5GnnGG}mBg(PdROlSd3C5u)w>X?)8iZH)-zIdDr5c<ZK1QVGg
zW_(dU(BiF%<H$Jl#sua}DxpjUA}wCiY8ltiA13D3`p+`Hb56u@G&a>&!oc2<86jrK
zrG#M9Se#7Gc3)D4;XwuFt=Q)d0GDjf{P$e|VA|sNw>lr{^3!AX=>B8K5r(qGs2p!+
z6w%BM)pN=M43<sIRazdZY%hj&v*Gu>|M&18@BSaQa!kj7iRmR!;~^;VffWGi(S@8|
zl1OQuFo7!(CK-TQ8`xa2Gl<PoJ;-{gT2{M$l2n-zBTBy*o406$2qo=nYGP&1YALg+
zueJ791*XT0WbIU`E5h2*I^QcNg{@n%qZ2Scwklhb%YJpE$`Vu{JKb-w73kcr%P~n|
z*XGMvf@GOCVgjjli!JN0`d^lA<Ju=R<9EGY^X~Q?{_c1GUVirCcUstX(NRn(O)KMB
z{%QuK7J^XwaCuAhTLWJ8@=l(;R{_b><1a~1ofy|+(DfRk|J5MxTe+MH{>pw+qRX4M
z<GaMRt!KS%Sb0m7IHbrJ;bw>+YTQt8IB(K0Ss*lyAi{1KX@kQ%VbwYMA=3m;93wF)
z<CwTknZsZp!FgfjTRwc*Gsehn82E5EuvvBNhLN+jAp|96<;;7|8^LR|BqtPC7q{hM
zsZuK9O{hA?L`t!+diKCMffxsp^LXEqlF5mraiH(_q&OJ6RmyHoS>8F}Thr@<Ow)Fz
z_bDc$;)BJciStFPG~t6WV79nr%DmOGZnd&*QZzmbW5paxS%s)m0Yh;(;K(><F402x
z8(4MKqg3A~_HT8MJLizl5V}o?GqiYZ3lGY~VUPI6m_cW4&f}h>9XJE6p$UvSqA{28
zKF3Gag{803VI@A>5<cv+w(}PM=gfN*+7&vk$YXEXu*~E@%El0;Xo-nZ;;+SdBm_KC
zT;-WKMx2iX-`I7zx#oH*qStK`rm~}`f$4sTHa>E``(b#n&(wa)UBJfp)Ja+=D*k99
zP;LPLDKCFn9Dn+pyBWiYD3}g*I~H&PfD7<9AIS!5Dy*1CSo;;f@bvHI8{hak|MlU&
zB#q<TUey9<SZL4b{S)`H@zd8b0h-yBImZ$LM<w7U;8KM{vDLE^KV`>Ow$OQpQ+Kjj
zt>>F3{~5T>V6CnQ=iZrDmB;Sq3IeM;wlZ5wuxdbfa{G$K;F#NAsjl4&(+d^VDfvXF
z7kVxIHfKiHF|sVup?VAqU8#an@BOkLXK=It>9yZ$U(|smOcFRkaf5}xs%v@w{5}4z
z_y1mg=h^R?5<RR>OD0_flO@DpdIn^1o_f|geima~YUjSr^mxH>dQHbJd!2SU+8-u&
z{4pzd(utp^)q5awf6M^z_<nu2o2bL9W7iKf-h)?yc)B2r8CFdw0sM&^&DA-MBd)P>
zT;3bHS^0o#tjri58ZxO3<=GIt2>V!c>6}MhwDNIe)*iG8#H?I~h7de!fmP$#kCq+k
znPs1H(LZ%4P4L#X?2zp61fs?QVk~RfHEW!}I36&LC=hcZ4SQqig1z6IFG>J1$IMoN
zd(+8x1?Xr=bdbp*)0FlT0nMiMH;sFHEK{;9SP{x%-q=h*4P0g`sX2jL&5&bR%n7M1
z0-d<|{ka&6wcdIqH$u}D02oWjx6#xBP;x4j$h9P8NXbqX#kmeudxwYMtvsUl#>}a`
zMsiL?8+D!sG#L<1&Jo%`+c+dRTyT)h<?fpm;&vdZtg-6CUP{?ht5Bk9eG<C0x%G>t
zW3B3x#vUo5M6+i}XsATS@iQu|eeS!KHauhjzYY5>6@p&cGN^rW<1QzcX`w%ZwkZzk
zmOGE*_kRT)=&iYLj-^`sUE8xffYyvLtl+rV<gPKXl9oU4^bhjl`KSC<_upDuij)b?
z9?W^-u7K#?j$kTwCG&OI>ToJ7G`COZ`l?55KB8*OWnAO>>UrvMtqX3ytbn;5?*!uQ
zMUYEo;c~v67|QwZQ=5uX-VAI{u74T3=$vDE{1-bpVRGNg#5&Cfn#$zWRi(^VV?8Ph
za(MnuOS|hhsQ`Eh0$BiJq`KDUKGB+RA|+<Wd)g+@wJocz<H_nV|DDG_S1K0hG@*@g
zEWkW4j)P6sryy`~4Ry6V2}e(iC9h=iJ`bx{@>@OS>sZUz9q!KStm81>Gr+!!O<bny
zEarC1i3ZQM39K8(vt5r!=H`$H&YmL|JnODuH;(kf$Pl%-hC#%0h?&aPhmu6UY661L
zA4XmdkxlEkSgkpX#!PhqTCYSkS92?ZE7c{8wLV@<i<HHr3Mr>a;JAttqiXu2DBb1)
zO(SU>82cOIFyLB4F(T$BHug{O-ZFZZNo+nec!B{AF)?9r2EbfnS*pW+Fn1<G(|AH_
zF>BsSsgrJuT`r5W;05mt6suZQkI+~mNlaE%2~EaGmdx2&D}H_^7w^vUd6lZN1#`eP
zEj|RiI;#qmhIz4_Vp$v)7qJhd6io^zjf7@xEW8Ll_`<Rk86G#sg&Tl2*hRL`7}Vm1
z`@uqm+8~6cWh`R%&1y{uo;VJLu`G-2N#(%6pYvgIJ)0{&Orb<EM_i7WmShR>sqfbM
z{?sDnQx@?}VXMVG0Km6lpT7(E6tKL;Yy#FLxiK?ma8v}oMzzg<-@WdiHS%P!-wgDQ
zSV%Z(>lB|?7%p+(lH$eTa5Q3417Cjc3;cl>e?Kq(=1YFP`xUASq+Yw$D^N>NnEPi2
zEmh<1r18d5mBb5IUL>|k)yZuDHS4Ybp+#9SgEww+{m%HZ_Qwp~j?cg7SXd-xRB*M}
zzMwwEuFq*vrWZRv&77WJkb3J<mzky$lelcv+@(kDDsR>WtG7IKz0X*7^742(j~RO@
zxoJ#I*SG+OWuwy8x`~vGV5TaA&UpiVZOeMS;&JzgpT7KQJW48TOQGzT>Nw-d&wO3F
zyuL-7t#zlO)G}SzeUF+wD|dZ)d!cn3Ki#@nx7ze3rYFD4t$+3Q=LPuJvU2v6oUZU>
z0)O*8eY1DYce<^*hK$k$PmWd<W1j|A!LeyPeM)@za^$jeoOh1rH+yz{Pal;{>lva=
z>NzW|cf_1&N>;3^;R{Z1WFD;o&LvjfvkD%FCWP)3B$6ZC44IfCG9*~FbYVs3*IbPU
zVumr+a9;yEF)JzR1Uj^^XG*<it}uoI;w}EzRhP5}7Z7L34Xz5n7co8OXK5#k3P}h;
z8plEw2#4zfafrB5DX5muObY$M7&m)2mW7^XQnqc(3lXuH#^QB}R{D|{%?-O+5#vam
zrJSoVZE1QwmhYVMx}a@VH0{Rf-fCiqV@Z6e^Is_&sX!JJf!2hiksJp^6Rzn1PYA8e
zo5K7~DT)YIZ`qPhipn42T>-qxzK^8rI9oY_?-<h_k|n2vwkZUpfxK8InVgy0>Vgpr
zh?TReGpti78ivX08;fA-UZr$t(Ghg6CVT2aOaFUk0r#D3Z-D`vCplP5^2Kg}tqR6U
zU0y8{<^<3d^!ap#TeRNo7yt2eas2)3Ckg>yVS%|c97g|}db|bsf8~dMl;8c~@8@gj
zSEss+)A6;6CD)Z8h3U!Vf%Nran1npbwN>C$YF8Xx%tQn(5-g6uOnKNQsAs8Z6R57N
zSH%!Zz&bsHQi1urik*Mv7f#)&da$*iA}xSWDNED&pe$XI2Jp44svY8mg+M~df@1=L
zbMqFln)A_by+W${cm{t{L+AcJLYNi&%^<zTS=zV7HI9t~=d8U=2&}t~u5Ec|^OWCp
z{&W29N58vhnNbA7?Ks|f8`CM5BoOAlnA)@WaJt8LFREmIgEG6Ibh>{>&(0FW)%_PB
zTXeoXy4e2s;k(s@*V-uvtp6(Wv#dwo7b&OuJROBl%w+~GrmHZci{zj0cJodc6kJxu
zJTR(fjPN2RhCZ@x0_)CLpk5QBDv~lWnc%aHG1GW^8G9NL8YeW~GsMIpCi-Y~SvQ>t
zBM&jbm`%IhF~!I=-ZFX{!OElMgrvZe=8HT(JTqyJBvb9$oRySIoLx5f;hZ-C<Cu!}
zUM3DbE(CmNL6c<zYX%jQX$iKdP08x)IxqO9DcXM)U*^_pyz{7dQi_Bou-^^rZ?5fG
z&Bo&8tc-(B&!e?7mbg4AW#9@>@t`pQ87G0162Uo=2)Ps^aE%E{yEK{`v_3!1S*EME
z#ht44an=&U9omw~tT-R6RNkV3h?ak+`J-qLYSHOS0_aGLBi^_8rUTdDo6U4xQY3q7
zoUVCT%DW(|X%0r^5IkcHY}XBh4a07a4ttsaP16=2?&v~8j6L2N2&{3As>(QyrSB|B
z(22u41Bsytr6lKE$UwE6zbry#fUG=(0KOgj+yUUXx(Z){z$2G!c?fQ?eWyOC&*-Ip
z3{1=UDjv`Zn(FjqtID(OmY?~VpXD$AxBoT2y#LG4Hf7SSphm%!_?W6^RcF`&GIAaH
zm2s-SGyi&o36?;n?lQSaWpb&`@ma>E;O6fuBQ+;$Ojxj-q`Z(yP;}O+VQAY=?#J5f
zX&?JVjt+KgM}SQ&W>VDx%4?=>1-(_?p-N3u3_D%-@+>kRd(mC<YWx&eE!vS6z}9<T
zUP@K#&@pF5Gp)3Pw^Dn8(zTZK@pS!!UwHb9{PgPYFoG&(djab4Id6^Od+qaI6{{XM
zt~N~y2>Yy?FyEI2q<hUXthHffzsXK_``*BRD9UQpFb<KV!XYdBG_vxJZ4>BP&j(%L
zW>CBcD^V`Ja5H8Oab!OXWC|0PGT90Ay;X%k=(BPd21}lZwouCo-X}Jp<vJx4CFVpI
z8gl^$Vcj@l62{5|IZSdTLqBRM(`KM0r%cy*99F5rH>Ra0YKdI!`V7?C2vTa^T3JX{
zQ0KulXv{QiOK2KgXh@mLn_Gn@g%pi(+8-kQe!z(bv7`{+bkiK-e9<^`mH97KguKNi
zk7F)byXCnRWAD6!rfoP3d&_t&NhIxRMI46`2VUke=O>|lEB%&mzQOs1CNxLyPE{oh
z!<`GH>Csb}+UmrrlQ`d4C8ab%&V<mSNpX$o9mX+|%Y6|s9YfbxsDPwIj+Wg$M#pYQ
zoUb=*+lJwK&3L#X4I^FG(5@OZ82~kb=<<$i8go&Peu@hZA=o`~zRc-0oV&~&%wbT~
zvotNt^>(P%gMB{sb^xFr|2?g0irvQwF5O(ycXzziUo~O8DFD4IQRV1&SHaEk_cm~?
zG|_6k=8Nxrfxr9dA7b~lJzvXTMVeUz(^j-@0oWECmPkns$n&H&O>#>>W@nrpmL*M7
zj<sa0rZ~2m7_q#r1-n@PJ_P`kVVTamVE?ApOmp$F9pyo;*CD{f=N%&qSpdHo2o`sd
z$l~~Qm&I$b9{RXD=e8|AS7pvZ=HxP`=FIK)&XK*JR^MQ8LQcJDAgZl$U5OWRPH25&
zF@Kw<{QNt=k6(EDi&ogY0#0GRKMT<F<{<F%wF-=vkIf9Og)FDz<MaTp!H5;PFRt}8
z<NH-_|J(94GA*W-3mA~&+RWd@He?}=882|Y>IgaWa(5u8=h3!dK*<S)l<8Aq>xK2I
zL1pBtpS<9c-3|r%lpwq5{tkn&RflLvNUH|Al9?%uvgsNQBa9<lG=Y5{=wsrn2^Q#|
zVTjQfzPu2BE-%ZJEVeH8y~V9bXZnRDTsEgn90uAZOhU^U3L7@-ol@SNq6xtfS`(sA
zSs90jI14w4xww|I)j}VKgbU5gw7FWF*q%Qhgd9z8(3OR(g`U{IAqY7QXd38>gi1=v
zI1Yr+(ylke!@<-cf~OxW2|`K&NL?o;HMX%P(TQ^gM1hn>kc^52W9Fg=aPlgf#=UbM
zX;y@8OITg9>dx@(3KuLEFU4eG(|N$bbYEuHu>tQ;b)*PMfigFYeI~2D%g)ygtL-_V
zY4F|=TGNSiwmbmM#Bt=X-;oM{td*lo=TB{65Ho{1Mgm<&$v!SImfk@vmx|4!sA)h}
z!RiP5eC@3OKt0J$JOzJm0tQcif0OgR>3+*L+_|kV#opOAl>sbtV73=ZDCN9<@+W?R
z>;4b%|NO>(!1M4iP;`>iT%pGky-fJXVm)pHUOoN0Ov0+hlHGZo3lLoa<}m=AS;Er^
z`*<Rq&ozH<5341F<Wd36ki*?c2zEY2k778dgPpW_I)^Ns;U}26d~*D<tk=BQez|p?
z0oU|fxgRt3n8$$+2J~{u)3JS#xRIRDln^I0jg{l`A|?zT0`IRs;O8Fw-o-J6sYFwG
z(bf93zxA+hVV`qjba4~Eol`ACN-BF&pB=ex0DPC)VMy%uBb^h@x<Kn37s1g;=H_OP
zI*S3jTm`OsWk>^(J*{(m`RP+G)@!^7+qNU7lA#$*uVm8%Rzc`<DrFUAitXARl$#-Q
zy-&z0AnN$^dQUi8qgi>m&umu>s;P(xD~ctFb*=q7^pP|QIx2^o18vt>Y@I5xlxv%w
z2?Wa<F4cSr#qW$hHzv+E#9^@K(FH<lG9%*{8TUQTJA7fELboPmh`mX-Oq!wU=91Xe
zLqc&yj|tznq75iW1*U4HB)u16vSbnGt%O@vW6|O`mYTh35@y$&aM{YqnJcu#hZR1w
zWHCl;zw7B%E1+mxmZXSj!H8)Yw%ezizw;j2p5c5ulM!XXld|GNO{QDnz0e4>NlQ&*
zS-jtjRcJJjS1may!;sL!$W2zf1kP720ju5-O6}v|XMNO^ld+ydPmCjneP1%9{So0X
z4vCl#<m}jV=M1R_pAks}u{kt2ncUyu0sOr+dn+-3W_N?I(~Q2mVw>*M4m^!3{4BuY
z>1U^t$>jcBvU6e;HDZc*FFbzyh~NGbKf%YJ`~m)}oBtPA=g$ks&>4hO4Yd;3ehRWq
zxy~9?T|wK@rZ}x5FnyfKI``9jT@(0Qur^0cn21VUestX5QiMO}W!w=vSmO(8n=>^=
zP^ZbeHfT!tmBqPOWDqagSjWWTS_=8OI5(3{ePt1j>H@OF2*<3ZX$X3fr0KZ6G6>#-
zT2QlKpRYS(H@tH^UOnaqHa~<YM->QgLM|`%3H*JBFn^z^z3O^SF87#he#5>vVVbES
z6^oM`#SMR#+gTS_Z5pn3k*n)H?4NPD>S)(z^nHRH*$vnD&5D>K@0_2L9W>st_MRu(
z4PG2)ZJ-~ND8gYJaHShva7a}4V`kHuSoE`-171u_IC!`j1~$&~bh|3I$dV{x%Cy08
z7S5<~rXe_T0-wcXMphl<%y`(d9|wG6(@st!>-B1S512e$AmxO(){=d)W!y^1#8pjb
zQpVBa8*dBBWK}o1u4T3DSgktL35OxWSl%AaY7?tcgrw`+PYNS%EMh9}=oB-JcSyEG
zkDM~mm1L2saf*QR<k2dbpr!H;<^6`e3sZfODfqCWX;*k3rZsl#2Si~U#<ECMtNc{x
z__o_{{@xD|){miS0AbS#jTidCYBFaPx*%kyMZDSE@4gB6##Yh-Q@!I93s9`_fPiWy
z4MS<iimqwtnub84bqbmbt7~CEIcM7zKs=#salxw7q*#)0yr&H--VHy(&%XD2c-sCj
zACDjL@%W5iKm0QL;hGomn`~N>p*dfNxi0#HeeU*FgrM~`ui{s`aA7%PzltGz-LYO}
zr!jwb{UKW7O%@oUx;vb6?E5|KW<wt%mlx;!^zZl`T>qCZ`LE9Z6Y2^xyEzjM77Sjc
z0L#mDTHIBjQyIilTBE$fmb!X#cgYMs3RgB|po&bC`x4aIW-=YR_CbwToHNu;!Rm4~
zI)7Ry$3=7RnBY%9|H&;!Q4_YlwI4G|6s8yYgxM=5%}bzJk6GJdaizCCi&C0x*X_h1
zzLcT!CZ?^yRQbJggb?W3hKuHcpL_g!`1!}b&)TCUuBi5-K&@q6`*F!){#M&vkI&pU
z`fWGV;yaf^bY@)dZNM8G^K@ZyGVH#qZQBG^7aP8Ck=X5fKKivi5@FL^@Xq;`{iyWA
z$Ty$AV6|G|N)lDmS}oWiClGUw`rx>}9?;ye>N0(<pvckZM9P-EdAZs!#F0bHR!bPi
zw((?9_Cuy?TYO=;#u)j5?T_-e`#;Z+JfyOSa)Or7h^A@KID#g|zDJyuL$d{3@g{gr
z%*k}-a#Lz~E17KXL5&ti>l>P8WinhCuv-`4C*Du)SMMC!)EesMb~KXM(&=(x@Wm7i
zON_1|rO-H>B-8{|4e$&jG*%P7)(;NB7|CJiiw2`3m4uQ?<7tT@bl%gfwk0vcBU0Ex
zRr<r87~|wZ7i_WbUEs+Ve;B_y=VljSRJLa=T~J=Wcus#%n$4M&@bu7+k$!iutYL%!
z!M6=br7T{>T+>;Ntt$^n>@9XNk3EXA*UVLL#H}&TFwm$aY&ho_O9D?EM&i(ydBmPS
z?;XBrX~veP`!DdW`~+Wo|98*~XGpW*ht9vkyJO^s+TX?3UVe;RB(8=}QTYb#>WU8!
z-z4Ym?LTPWhxT?Vfh<^^rQMeqy!@^*Tc@Ajwe4N@#_}_#o858=PeFi-NzlzZN@e&=
zdA$t{KltS@vHR(t=2t%aB|hXgEDl7O0pNmhI-OW2>`}hbT!7;VJ6_`nwNR1f7G8kV
z0`@ExQ>J4dx#G&hE66cJw%BI*n*lPN5>}>(cOl(UPb4Cxh($%J6?j!7wqz5Roh#G1
z&^Zq92sp@mOX^3*8=drN0qcgn^8|j6cd3lPd_aK-^oes*jEXvomuZ5h37*UL5x@J%
z@8#zo{~{hchm?M)r320V^_>Lyj!oPgle2V<EAx20w}aonjL}tGnIL7s=DlgJ)X}pf
z3~wzg{Pzg@1A-GoGn=lZ@t*CE{v_9ZkAxN%8iFK-oVk2-$#%VB(>V4;swyU{PNWL^
zLD>&h-QoHW*(ca;0x@TXXz{EuD}!cM-fFb&vL!v8wT`aw>{29Y=F|O-RU52innsXJ
zQpcFE`mxJ*OSUQ#&8mUq&08P>WMV(yn`VmRoJ>kxT-V@oMoEZ9oS62UZ#$Z<BXljk
zX+RuVth`vtWx<>joWnZ<f5A)1I5pWB=l%3DoZq9?G-IJ+0jf=jqZ5X41gu}R09W5M
zFqG#;%e$<^!I|!%;zCDgI=c0S6yvlw9EXvZQqheqNsh(S<HWIf^pvwlPuSh`pqaL5
zaVR&tffpZtL}<47ZbQ?W0QUa+it%v3xdw58q{29iC<Z`N@bf#lCie^n#{CY_3=(jz
zfvnuzjPw~+UBGWT90M`+#(0V_42fYM@l9(kWz7(ki}-{`!+Sh_dCAlFev$(X*S+FA
zWS?*nIa_&hOgvqmp(?ao@O1TKJUJKojq=-n?BC&o_)GlNxB8@g|JhprfcH-q_left
z2ilhI;x_?y-#QyOp{BC`t#5%UNPIg^p=lc8II`NTQMID!Kk=0x<L|uw2l!8S{}G?)
zH;cPUmopbK&1CmXf91OKv`q!hi#|D-Os57<OcT=|%H5OgF{iiq{B)94CIGmE1VWLj
znabcDA8cl7>d_Y9a&lkKz)x!pywZpxW^iJ87K(7kx6eLWBvsF}P<2Dk7d~-^<#@}n
z<4UEKrfp5*E6?CDG=Wvy(Kdk}+Ws&<|K#@*1bdm5_M6^TDy6cHdtTs9cW~Y6bj+VV
zGj}nnck%7hZuf-OED7LO+pD*@ouqNxrdpMuj^w4xPj$t;&I^3!*VWIqeBXSJB!OHH
ziSz78HOav@v@6ebpBYjn1P@NhA|$bpi4-F)w0IwAgUJhNwgijEtH63Sax>U_sCABy
zhk-75T0Bi~B#B)2iI|nvK@?#(^mgA|W*A1!+zQDf&h>n4_$r4e^vTMo#h6Gjm&I<T
zX)SDG>_@uw8s{8w=o$JwH5|m(SUe24i4SNhi52Dl?fR1L(R;L;EsB&x5XHorZ46X(
z7X*UKw4s!vD=a9~^`q(FWlY?fP~}Ssdl_S?w_6r%b@5)5T*qMqv}_+pj3sWdR#~!s
za3K)d4&QVp7bDOg_Ex*N%mrEVRLi)6QQviJp1g1E^N!$^u7$?KZg261;tWg%fxhpF
zha2+PgKzM@wXE!<WE|eLXl`)c>X2(1h~vRD1#`xQ1|K>QUx<cG9y7U7hU5q?Fr<Mo
zMuHRCwqY1E5n<!kd@=kO?}i`ZEWU@+HHWjr(0DGk8}N=O#-8>m(?ucpz>}@xsvqh1
zBfCT9rVngZ4ezZzo5S0C#=c+ea|eK~sFS~`-Q|jPOIG5JZSTW*OTyxHV9-Fz@$r^>
z)VhaG7dT>yB-VM=s8&$B@eb8Q8v?)kXMPs<OD+HLr~d(;^5Il2OmhK(MJC|ZNf+0t
z=%i9_FR>XwRTcKCo~~3u%~G1FpzvA3PHrJUD_yj}p;7@yIbvyVR<)XtI>+)AlFlac
zsVv=`Tq2AALy?kUdHyo#yTv)xm?B?}qd6Bjh4}_mpjPk4^ivLf1o9Sk46zKf)VjtM
zh>CNiyp*E}jk#DtL-5|>dQwEQ5S*1_y6i4_fBha#bcz$Lz+FpyXy@HjKz(xkR?v3>
z;3t@`b6n>#%lf{>Q@#Iog)<XVtxk;h?SA_4)I1iQOC2Y7ALKN#>-2zC?p&qFYMd-w
zUI*LiQ#Z*eonB)NK>rrrJKxFntIw{uTwU|!ixzQ?n;|0N7&G*7WSxrChI6DOjH-Co
z@U@$PlrxVXk2o)^SB?-N5qamV<!n^04w<Bx_cnphIBw#BG(h7V`%#I&Mc2|Fdb+kD
zdxaq*;`#OKU*~V{{}Lbdzrjs>X$cp@NQ`;%4&<0{4w}X>4kJzLQH6dOQPeU=b0&?0
z#p`M=i!ggPIvmoS5muKp>qkhp0Vf;|HKtb=321c~NqHYzjdWqwLRE69#8{qxVbW}M
zPD+NXWzB{XvJqn}jHJ0$b9ooVob0=l2*I*tV;t<9MJ_7u_c#|!4#s<n%X5yLA~8mb
zC#EirtGn3R->%OQXF7qY$m2L*-8ja^(XKY8wI>P5AZgD^_9UN)m@Z$|K#rO0LeW@F
zWC@s-;6T#|qXwE)Fzr3f_%<+(+1^_&k)!EOrcke6a7_T`dA#`|?}o4N<>o7#ami}c
zGK_Y<55IoHMJF^}!^QcEn1vW4+s=f(`(4kdi4-Ge>lWu6hY?WWp%UENvClsVL}$k8
zK6A^J`VjyulU>g9M06|W@AiK6yvy%i4FC&&wV;wF__6?bM~5%}pJ7d9^|VYpLhz2w
zs>3y&pZv;?k^kC%kN@oI&+*)Sj3SKDR<|{9nC0~?kXU8Lw5$|e<xT1&TtHKK4O-O%
zEfDN3Kso*YWa5;$4|Hn7bae&zn6kSSMK3R5DK#t>?q=Z>q#U1hsU}iE=JABB<$U&6
z&ZY@w6$>WH&R*`Wl07rztR!o>v!*n^*yJ{ICQ~BI1S*jtUTU9NT~z~s4!|Y6b6hm%
z{H}|i;};(NKC6H-v0SrE&@|sjJ^O;~Td=sVrN3Thi&c+KiAFOVKRWswp6}kxPQQ4?
zg6z)N&(q=gDtE!@uI0S>=-5IXhIMRwFPN$@B*-G{_6MRmo*jyApM#b6j2anI;^hG@
zSDsMfPPc7CK!{2Cbl<ZM4d<<W*LA{>lo8L?XU@Bpt3L5;P|FJTl$v6&UNyv+=~Ltz
z>6iJdFa9duJbaxYkHy8Tcqv|il+ct-=g(OdZyaN(76IcBEnd!9nYWxGK6`R9@#mZ(
zpoGxSUA~8`AEDWiGx%&Zk5ep}y{_EsG74OIhq>TrTaTX>RHbAl4k-f8QlASkCeexD
zN{qIMpgE#BGma7Gg_x}T90Dm>J>afwiDR_TpU~2E>uHr`*JCd6(6j{@%9N2D#}U<B
zN^9Cee#!Pb=Q!+lI9d+7Jx#Zv@nWLYsmDtqrx6`qGA7WJX;&>sU>FlJX2vlS^FSMd
zv4x@~MlsVk<?>>~%a;e*&XH5bd5<*pEuH9sBe=*oI)-5HRfp%X`!bLC5x&&@I9WTq
zDiQ)Yj_h_LmzU?9wMuXqT&C}PzWL&YloOlvmh<3gTwr$?Xq=EV)2Ga*S35SF4MR+<
z-ujdF{bz3h0CGP?SFNVKQy=E?aRnn%84X>|Z{;~V1#c%imyIX4<b6#znndPyNO2k7
zz4rnjo#Nb_ty+i=Nbr2&op;!-w={nN{@mf;Whkx<M9`EMW!5II0LCoJ)GfwTDe4z`
zdDF6YLN*f^RXLdgmtgs*d*s$VnprfRe$oN}Qa;O*YgQ&!YX)b`p?^yk_Z?)X!cn^K
z<aRFEW;x~J!t(gCV6YXtz1XuF2%MO|2@ucM0hEM~+JFTJTQp1-*U~g5vZ|U*^REcQ
zg0a2c2|srB<NU&t--oE(bD3V^h3RA=88mTwWlAcp<h$IvzZsjK4_%Z`n`0>FYrKu!
zZk>MFKa20{wcN?dFaAHD@%DxSOhvqwWA9D6V9A0m%fFotsP7H@$K}NuX@!1F?1n_!
zS|;kS@A=}pm%QwIvOpU=H-`bu9tlF%3h!(@Pc{L~nI^!HdcLuDc!Bq~YgVmC#*BE!
z$Nk9D%?k06n3Y{Wa=r<)-g6VL`Ssy9`12qAfB10!5o0d>o$VfYv9L=Y9Kk^x?fy7d
z-U+$rpNd$mUTLkaZMFmt%dUkKOD*1Z#rb<*gs?I#!BGhw#F^%y7R@_#uGDBRb3iHC
z7J}eCIMI^5UlZ7hw?+gJ3o{XlV6_umaT5!fIQI5TJF(Jl;-(5C&UwVo$tzOkAjPhw
z34xSHx>d{Ja6r{h45fK&j0Ye3p0;fZJ2)2MWDxP}ub;DVo+umAVMiPWR?P)tggET!
z8prz48F7rXc|{0kblVNv-TR!MU2y0RtX6CO=3oD7TyEawop;apru$W%J-@+)hV3S>
z+BSUQq5yXlOA-;`llBu*jx>#e&kRWkX~X;OxAS99e}t#qm-*!Ro}eCSJYyV8Mn^{4
z#^iIp{Io;b2JbvEW!}BmvX3SJ;Xoe0H`L(BW6#AZaJ}zYuR2^7)?Lfcj~ot>iw73)
zbGNr317r4j2CmxT&0T@S(k*s^XikBnoXi5#g!Eb$@G)RKvR6z>6RT9%3s>EK3t$n{
zk)80~lSlmIkNy~c>DPW6zjpPvjY*5?SQqc);B!t!Ic{~oV&_zoSGP&NfWF$K(<?fX
zO1WdZnf0p^Y+?OMvG1HyDS~dn%?^9{8{s)$x2>+-?XFw`j<U~PU{E;eay6!Lh)T#x
z_6`xnWkCoNSo0o4g<3^uf}A?{rJf%9s6k9h_SFsSoGY$pSMNgE-dsj!%^6R+$0h`w
zOAX|y*^7Qyo@fEumrUSiYL#Z4M$a)cTY|FcDn7zt6OcRxfzz>Be82p2S&FVa7q?u|
zw^;md0jZs^%S#Yw?*l2@e2<_%;Jo#0&bJ(Hu1KYtL>q+We1i|d2hEBX`va>ku-|8{
zV&XD*_I={o30J$(+@U#>RCs=~Cxk#8mCJR@bw4MkeWH=z0*w>SJ5a9p^B?~={FN7f
znNRjR4nqb0i$!_Pj9HP!Eh;@^n&1c?+F)QOrp!1dLMvo1IM*@`k^Yc~qg4zMU~_iP
z`P28wzGF-o?*!kPTXq~RxkH?>^HUtM&RNb?%vq!e4m+C(R6ugA#XhYtsnI_c52&=k
z_8HSi8hSfVv_uMVLR7V^W>rGt%X{DA{$k1M9aEwSO^K@u?DqrUNXfhgQmT+t@}TWj
zh!a+;HT&J3;2be!7!S1DEnYhMKGIwVw)!4kl=oI&;7jlRC<%v*8<2DM`;oJES3KEv
zTx>gHO#I>({s5tAINLTHhRDn7$U$MV3Mj_XH6aisG3+xphs^c%Bf2nfwmqXyN<T)r
zv_kZZHe{~$JuWy%i4`3mzu2*U+|Y&w?;Bk348wr-k(_%P9~jcevzw9Cs>8LORo5|&
zd)nX#uHov?^W<#J&7miz@!ojl2m9>o^9O+LO<u+MgpF9}EGg8HAT^?Y-jX`C9pkqO
z1{X1UZ`j=B+_nU9cfgXPeX=I22P>7$VTw@^g~oaMn*(2X=Ux6$|4;CL`P!f5Z|N^{
z{qoAB3zk5tg4#l-<*azOWaR+OsMM+=OV`<5(2paQZ(<d1fsgufx8;2e&d%B73Sx}Y
z%h$P@QCp>Kr0lae9dcaeXYSa!ZNgC-W_7L;z#y&^BhR>Go+)vbj4~nU%mQ2;-=Fyw
z>d|MXcXYlKiyv}jcBrn~xk(mS{u5_E$uB(p{RaG`#9rn(W_1F(^{;xbCfEJvBm*-o
zjOumXM%sqI?*^bpjNm;!;Cnp4M}V>(Uyj;Rw?BTe7Ng<~&GyKZDTk>xX=U!e7i=F#
z((VRHnH(c8MkTqx?y$qTOU`_zk4n?p^E5{2`w^#!F_~nCAf#xExu@GLpIqJW(e<8{
zZn^Gl;p&3pozU><Vc?hdU**H`ukb5ZU**MNN6IA;qCB!S!K3lU<R10|8Xex*oeaSf
z8VlD*DcL(RXF@P<k2HmC(qwhNaYV^%&Mvrk`VN#bZy~r6GwK+IM2e=RDYzPzS8Do-
zy_>`UsVni&Q%UFoTJ!Y0)%w_0u|(>-OrQx;q;wK-j12vNbDnV+tc0DjER!a*RvS13
ztMrufR1F0{Qqdm76S@_MY2-O4_zbJFbDV4Nu4TR5GLDfC)?Z>c8#$K?nimZpfBb^q
zcJb3Z-CPpaLK9kU2G3^G@xwpxMWO)W_(p$28Z#H0Et|F>1kZl7_jXo~$8)&4X5Fsw
zs})Vm^f58SNYgmt?!aN}5rOSyOM87mqmXxjoIPt_LVLVp%$e;v@K_wGBUi)7m)?7a
z&RJr|bsxD{tw_avzFsxNF|q0zx`VRY9f;SN5PG&N$2c5#mi9zEp=l6RLi2$B-iCet
zNgxY%>b-Vr(mtXX7Fgx>pr}A&8Gj_p?*vXOk=#E0v6Je^6}NP8zUml9ZIY=zr3#R(
z4s6KE_Tr2X8ve)Yf0}>C|7m_H|5Xl$ega!lU)1E8tdJ5kwUm(^H>=fxT(W5TiWzje
z3rF8A+-DO&(6TT!)-B~!+`e(rta6k$Z*l2Z+@bYTb(5N0N>zg|Al=!q=ivR5=1$Db
z@%`}PEGsRQTBc+xIxfD*I@rs}iW?MB*K*`j5Muy(W@uICVDf5CN?qU5{@O#bKs*Ah
z{f4t}fz1bc+{a6VxdhyjXOgu{IG-f}I1kf0S<Adgpe#;Ud_HY`8!X-efz$P1?&TBK
z@O1xax5Y-MT75_Ry}`4#>{=5S$<@z3EeC(ITA<&l7RcOO-w+gr41FHyR)KBTk}`aB
zvtx)RzO1UWcyf+(!4n6QgUJdpl@e#7TyzaRu-gp`J?sx7zKslA@s~gS|MK<WHyAZD
zCP!!jftm;=q?mERGH8PnQc4Vm9*v%+v6$SHGhS+4T?2nsjlyKIg178c%~txX>pC92
z^OW^=OBO2^=NtrY=Zz6^aCo1~qO=fbl<*Iofhl4Of!tUcnLUCA%)GOBKYNz*(erE-
zE=+c3JdBKeZ%H01q!bb7NjX~RN6{dRF%#2ZF6`h@5e~z^s#_sGkW>L<{2I>Lo~_6^
z^WMdm+4^%{KHu@h=1Yv~*u|bF=?Q+-0m{3-eako|vI_)2X~-hD>z>W|nrGKH&^81$
zeb}~Z>AHrivFF)MA_S}A@yVxGG^-7dR*sAH$kic(3nV<_nAtzy(S(+ZO~Wu8NMq(|
zpJ>~_S`u#6fCR2~Hz<znx}lNGhtF<!|LK}Xn>9nSx$e5p_|VXV%(`u?%wA|XTX*dD
z2bxCN9d3B{gD-HnIT#!4R`{gYOFr{->)Ebac*p|&-0kyc0KL@uHgB}~vXaGepJVVc
zPXl$rnAwLixg72ArZem!4ol}5{1zj1ht(Blb&s5aEtfNG*O61EQMf$2;Gh1f{{??a
z|2V(=^}o&##}Z6zE9mNiRA0_qGRvw8Jz6HaGSN>$%~E2}3Z<e|pRhJ(0w31ac}2XK
z;5lafs+L$@^o@ZgPS_nO8hO6N%9Wz!#U<lK%>ANb!lbo_W$r^o4D@QnPEv&Y9nL8+
zni%QCFsV*@eDxT$`bpKG)N(JD;BN^4bM3pr&RNE9pvLv&Y5T=wA@JKTf100v`uocW
zQM}5Wp*n3{yGG_wTkp~0Z}fg9@LR%%3-Gf9f5%MSNxb0+3SEN3TLATS#M*S<=3|{U
zZwdHqpZS2(IDHc<@n@5HIdVaNuh`lNVH4PWdc*Gdfv^0;m-%Eja6M$Ul6ZW!=EG;#
z>~02p>*+$rWw%0z3^B1TpveP6gy+{+teTeneotrvPqxBe=>Iama{ZUlxMx&l%tA_Y
zd9Ix27l&adi^aoHs%jMXZfHxglntDe=8%fu!Sed;)?3*y=+-ORvol=i7_s$k(h_%_
z0L<;0vf6RQF%D^ZF73TmGi4PQs>*@^ziCd;nkZu9%sHVSbJ?$!`?Sg(Ng>9B_bt`?
z&~&R(t)#GFC0;Kk%ffASMNElvF4*=@d9=FV%w5nN9<fV~5YPCL%lAn-;t5F1^y`7v
zDeY>*bHB&8!Q^`chO9h(^n@2TdsG~U0i+E?M!xp^Iq#gc96otY*38BE6RvlAcKyI7
zFF)bxMbE|MhRwR=Y_;X-<rz6gc0*#nSF+0-qLtL!tTuf5Y=_r|FTeK$=Q1&NJlb}|
z6cO>HtX!O(@p6B#_k()QHXC9z22?fk>5FS5bckezXM0vYkX88FhoADnJD2PaJ>xj=
z{?kXOT1;xnnW)fs&v1P}8^`J_ymr2Qu+PRm|60I$t9Hd58!e{^H+{A=oo1HBZsq^7
z{Cy$=7oS#uv^?GsU3M~QEWqFVe~vyoYJrp_lLCko6AiUYGJ7e7rU6xMu3pl0YuYB@
zv*V9^<$uP%aPz<7>gF18o^c!}mxg#>N(R|^e2Gar0?A7Tu(F6X9uJN(Yv)&Zk#MQN
zhhq}9x_x6z#FQ*RT4(or{eB93E@1SC-I00HP}M0rZ`s{iz#wkn@@MMC87!6adhbXH
zlw6oZTQMT8gu7MHDbszZ=&ypFCHRBFL>4ftWnW(;8>Z2rYPCGA#u83o<JO$}4Mnb&
zDsEDN(=u40PEd8`vx!KknuB*SfTv(_2^4D|*Y@7W<lSX5A>qY&mgm=byCWX<+kGD|
z0ijOIjd}RqbA0>m7_6QG*jutQpEa%eTWpO4;x6M;;0y1)hpO^qwL*u1vql(W<_F&0
z@+%)+)9oV{k6Tt@WZw(x;90jVg5oujbKsrrIp6%`C13r<udu)P4fgHV*}H3oPKeRO
zoKsTLm<Zk&y%;mYkfz0)6J!2*724M7@1~S+uGG<W1$-@@Uvrb%eHq5cI1FsI=d{~1
zMs<u)=jR>@Qk%z$E2Z3GN;Iui(uiY55tE7X0M?eKarhwk!gyDytfdT-f!QKUrE15@
z=CURW_$^XOL3$4^1lp#v$2dB|VasLrF01&2_ntiETtDDzzwtG8hn{yIe~G914o|ih
zd^|p9>pj{D<Df)5h$9XoSGyZ7*DLnp0p|n5Fp~O6KO`PqZ0WiYyeH!5$qY9KuCI5j
zx{kz(%gZ&P5psesWX`u8D}RQ2`Yx*g`(ebdHi*l-_vn%#W`f*Ua!QP3VC_9|mw3GH
za9L^Fj`tr2c85KX&}33qUNY}A9V-{u^*ySYF-CHN^}6NqV#}v5_dL1WGG@@2Iqdg5
zxqO7C#AdakZB{hF<50TG3pQ=T$A<$iZU#ol-1Gy#@hmfpiOpMhw!Yu(Ex3TKT1Kv5
z<u_rPXO?Mxos?LTG9kRq?Ng;@j@X~0cAmU^rkSv^Iy$XN?wn+gN>#|jIMtets+Q@y
z?p9<nDURSAZR_~*(;whh@~_aXEZBB`IFN^t<h+#|YMKI|<~}rVnMl-(Hl~+a&I*!E
ztlC-jS$zQ>+2kX)eRG+KQ_745{7ox(mXoR$jVNOzwZz@TF_}P3VF7`{bXgCMac1E%
zMQ#BFbw`tWR$~ArhO?a3dwUV(oXDDqDbh$l5n?vouT+y_78n`w<g%PWef`&f*z{HD
z$Ej^~%%lV<;el<l;TInNB0vA=7a^IXgLjsgl5#qxz*Ir!7+1+tb>ex9)bm$xGqrm$
z49l^kqSbQ$?t?iM+)v>V3&d3?;Um{+84Fn#_j>~5GniJX=Up(rM-NgB`?$Yd6GK0Z
zWeSrwKSLrV7B<ZZPjxY1_%4^QgwkGDVm2-J9zBBK-`$opv%OgJY?o+TN9@77j3;t^
zqa0G^-L0n+r5_W&_Vt&nH)}LU&NeHaz1S0zlCtO1>jUxpim$r=n!ncmMFP;WqK}?^
zZx4a@N(d$_os|+(XZ3K$-s0t)htODTSK7rQmhKDKVv>JxVt>bIBFE#QH+QPQdb6e3
zY#DXd{zT1b6cvWSK%Fl`*gL~zmi22|g{qmc*YeJ(&$!+DCYTm$$~9a;tb|!Ag3v`_
zfLs>JMZ#ux7-&Kutj=g$B#*+%Z~4-rpMo1$J>Rlw)@<A*tLB{1!=+o{o0j;*@qB;H
z`q5)<hLPdr6@5;ux|S5-=8!n#OuK1m*XKOBxFnA^xNHws=Ni<n`1Ipvv@zq`mUYwc
z(aRm@+ZBg{<A;9ioWpL<UL0-HTFv5>BXofe9-Y&L$MpTkwrkjra2PWe>z0@!!#J{D
zcc=)vVc^NxmUqrqp!Tl)^m^pOzx7SlA=93B<bKC`d%=^-bJnfB*T*5_ykm@!^~D-N
zd2zEtyl1z2N!G}A)8M29Cv3XF`DTTB;cMS~j*6otGmayBlx^E0uBE@ZMgk1U;sx3L
z7Cms^NA{MofZey?*ST%S@j~;YaaSDCtL&BHx$_*%BAROgRd0Tx(+zcj4J*y2;H-+o
z#=-yO$Nm}q)&Bp)Cx=f6P0MP%X171E-|ZM<r0Kd6Pj^={7GSopOj8mGGs`x+re~II
zf<%>JQq$OqF%n~<X0g^XP1Qv@-2<KXufneu_&ZL4*cI5Ne1yo#yS`)?r&t=TQ+-e^
zbw-im)-BaEy3Pbw4~K&Zo#spnc;`_Pwf1oh=&s<T))lSybP4qB-p^n!>l9#AVwVz{
zlF3{+$Im_bIe!1+Uz|V|*d%FvczlNp<II8Ql;ppq4&p0qaScZYtxVu_6{nlXo$Q$|
zT&Chqn6htyH9KPA?rP(`JKpyw)$|`b*0Svrv%DPNGX^gEo@V|op5gCu>qjL;;k^&e
zX~nTS4CGPhn#PztfpM^R%DVMMka)y7;nV9KLrSEq{OZ-$_-|hRCme1*MqFlf76?tt
z5bfDswRR82gmtoJa#V2Uj%_Ru4S+TXs}5csGS@dZh$`KBg{m?PiMEWj;N4ULQLsfT
zIdSpy3Eg&U^>MXcW0kl8m*9FIX;y*Y1Sc7F5Mdt20zL?1pN>LK3`CmfaulUMjDXby
zwxR-wI2!whS{>XFJOLAZKGQRvxc9mAmplzmxOn_Uo_{>j<qNi3&%h0<RZEVUZhMX!
zBGLrTF3z}oe9rTm8(v)9kkiQH^@f|lLR7SM=;o4fNUS`ZpBcE@Hs=h($l)-u-L5zc
z86TAQ&sMzq_>!ALWV3>GNSt}c2T#rzE;oF77;q}w><9W3xmd5*kBJx8H>@{XzWTSn
z!K26TFl6P!XCHHU*0EhT3^}nojEvdS$Heu`o__2Bi^cr%m!2|`_~u6+^7!%*+x3RS
zA(EmJy1>=T8*cWIemBr`4NV)!&eQpZ^X(R#3FJ48Bd5UgtCvLGupc9?Y1ph9&bk#T
zC3c5BH@%f7JU=_*#m!4@t`F??hX)<Nw_tB22}D`A|L(HaO&&)t`(xnurUBnex8Ggg
zRq!`43@q@%E1LdBRu~f%v1sLrFVt4iTM7I#Kl9J>FaL%AJs<C%(S*Ro`58V0VvMGf
z6_z@Ii?}|`q~Zd#*cLqk1rytu4SZHjMROA~eOLAIEbcGOZfKdo@s!bB7`%x{%DISj
zOOcDoWbg68<DH)}hNZluHD_W@j4={N6O~rY=3;4E+P1Yg9%eRazTb|fb(1z=j1e%s
zKFKrlI4STg$H$Vvoc^qCV4F~KN;a9MWB|FipsF0qDs-&;aiQ|1d-;t@6uNSE@?tDI
zJy~l^kZ}fBOQ3ZUKX-fCakAafcMJATj^hMhxulYrDO_IbyfApZ337^&Uu~5Y#<ERS
zfu!Ub+xmum{@rX@GyO0!?vy+xdd+mD8qj9d5_9Hy$P5CPn>G2+vuW0R^Vyz{o?r9B
zUmOVmK573V{^B_s4duNIyTOEty=2;8&+_$dplKSs7n<^3&<t%*T&di!ZaU64p2N+^
zi}8RLtLN*yH)*Jhc?R5^AeK^MwO+A3+pt=%Nto+3sl7Ma;HHq3E+_{QngpDqX~gPj
z2Pugc<prur@Kv<g-f_+;W1k?KW@iWnLQ~9Sr!*npu+Q9~Wh+~L+xExFHyc{_i1Y0w
z7n=*RM!L?^D-5eEymvJ13g2$nbe?gHWar5_^ZaH{v+783PrF(XoUp#w@cz?Fp6xP?
z7ltA7je`kiUu@SrSvNR@O&f6P==*`2{hkKTS>su?o~V)MyO#)_&cl<(Yo71+pq{Ji
zYrgQ#Q^Y&^n?2(p5_86j(yWyn2XamX7dSuL&@>GxCbp|J-@G1qkvxxmOUwhRnKlG`
z01~+9`N|LfAZ-xhsPr+@7vLOoX0>YAZZ5dm4`j)#w;NjVj5+b?%TGx$^Z1=DS@-5r
zN7xUUu7#+|uYK|<tG1yD@X4n=aX4_^2Al*o>y<Ia{+<Wo!9Fwl{9^zM@_W@gT9<)p
zcbMUaQy?IBPHwL@ohJQzzMdz!JKV&tviqANvkMI-ceSx!Ba%H1VdWeCM_>G>`L}=V
zU*S4lBO+|q>nVG#y0Eh<<g_HSrV?T`p}|!a%k&a(;>b{5#B-+Y)Nru_RZs<5g6S*E
zBePp?!TM1q;2Ml=QE&y6Rgmg+D$RwY={%|?zA9upSFZIsWx*US?|>HFFLR-`Z41C*
zzc2DMiO@C+kZegR;^sb|*wTqvD*r2Snw3H5q@|%jO15$1Ly3WbpTGRQ{QRR|xCK6z
z_xE0#SU>C2nwhx7cavMU0=u#psh;YS!`(aZZq<b@R$aUm+*h}o-Z9eaV{nvcG<`RN
zzuVP$?(c)CpDO=c+AuTky1eCg?yp(Vqzi%g*(0z#$M=fuhlutOA6lN>T;W_oL&K_D
zSvj&nNe&*jD_-s`%a)wDTzPh@K-NSELN70gvAEzp?E6GN+H=-6HXerkfy31e?c*m1
zbAnVH2GY1E9&U(Z&pxhc!xL7W(l!BAOeV#(bCQHub}DrdnG@%aE@->9By~W_aCJSf
z>P%QUX(p7>kyErqpl=B&6PyrS5rjq<hX|T!eC9ChO)e`mX!0EP0|(4qeSX$4jw8d%
zp3CL~&YGuu<%6H1)2C?cST&9x`tp~!zB=&oYRA>f9sQWutT(vU^JLTU+rIFGPj(0P
zH<3dYaDk_n8=UuyL&x*$J>EIqedk@i_Q|J+H2Bams*=1X_6KrKG-Kj&-GOG--ZN?@
zm4)+mZ3z_X;Aq+{hy8(<FJ7{8j{Pw3Y<I(>vv*PFxw<;g`i5~BIBObG*D($~1W!zf
z!#L1&ow=UX7}AeEeSvt`o}aNF5{K)7;k@N+9YFVVP2kzhV2L&v`oYoApuo$+Kybn;
zINpEfUE(nC>Gfdo&|S-_Yk9fblTgksA2W=~bw9Fd0@8LSlSag=XXpp|n-QmQ)~*OZ
z({0FUz%?xwtCn@w6}tDqK2Liq0B~~7+%qq{spshw{N3mNy5kH-lh^I@6)+=3vu@$q
zlBH|vH)&OzIAifFuGapo07sy00w28l9{;%hPx;ed{o@QN;zPjufOq+b&CB8}IQjT2
zC>0>);wG~PS$xeFUeK@waZJa|=7KTxev9qPs+PttVw!^``?j*BHfviU^?A;>`;s-I
z$ztNF`v+!#=ba;jW|}O(;?Jgn@D-FXFCuEmB-8+L*L7IEO_OWLy+Wge>i(TZo=mvn
zs3J+}hzf-2wf}uz_y_r^i@yV>CS}ltKoBgLDa}RaEs0~6x_J}e;mUndESaRbWb2|}
zfr)N2fhYb`IT^?kIS5lW_;L%G?hf;DrjyChgE6!1uYNw4VDC*TnVdRmPtU0$mV$ZI
zHjZvtF03ufM_}3DZ{f}G-C&p7HDecec^ItJS8!;CRV&=|2cBOa_|gaOBgryjA9Wpv
zq37w7j_5t>cEz*)6P^zrnR_@XFJF#jvFq?o$SJXl!l55X{SM?Y-aB#}==Xb4-=pb3
z9(z!wzj?u@ab$COi6<f6+hPkX&$rNe1D``b;6h-#ZrN@c#$-=G>m8#eB<|UqpP`xw
zDC3w(Vx{3W>xS0nQd`~LKHe#4CW}c_#eR>I%zCpb?<}Qt4XfdTkR4ylKS0_kyKh9k
z^5Z|n<I77Pt)KF8Kd^e#aPjyY4W89@MQ8)-^_r$>2m;czG>vD}kzrI^+wg_$mfc~X
zC(|HEaJ={UjP)u&6FA>=Y`PB3i5%gg4V;~w)4E8zS_8s|FRyrZb;G(_5z|21cyvri
zvj&$r9D0T^S!I>baeYX9eATo1=zugG>-Cm(;}H*s-GS#X4!rYt%h>O^856l{xi~w+
zF>)9p`!SJI<h{!a#Cui}`N1b^Bo4re>>Bom12HP^KYqmZe&A+5ux>p!`#nQe&Q>eD
z3fDIW9$jqkyNEF2op9(6#Bt#1cEiiVfp5Ot@&4ltn^nVyFAj{r)9r>$(|~0BqlT^#
zR&C(D%QYgxVQ7c~Uw&uHYPEvv-`eNy`@uec00_5BlHIk@E9}0BWu7EoxBu&owPc4*
z+9U?xUN@P%>Ix^9TbXFxqU>8ekxQA)s^x>nPkHpwBR;+PloTUcl4zz<M|GF#cAV&s
z`TA6h)ym9nS?B$NA(=a-f+9<-s4@v+I(|8mkAU{lC24&!$6^|=adS`3eXX{QFZzHX
zSa!4Xv!sopmb6kYv0m4rZ*#I3!OAX*z-qO^iDw)K_WJ|FFciVrP_`{?Su$8UpJXca
zC&qSo2b)#L#n}dzBCR)OG7Tf7JOj!C<>m#>^5)!avS@>4O&hzN^Fp_9V#H3Ez<T5(
zx9j3udR&rky1wOjpRSh)@E*1AG$!yCD3nUI&22ckzBhOp#aLNgg12c!LB04H3<8pE
z;Z*xUQ=Y&8VDc{$u$AM4lJ5bFYPESkdig2m-5S+^Ydz~OaCMV7^f$y|PZL^J+lI4M
zOUj8CF9t%>adElg&wo9a#45XQhyB22)6jT}?du2SX5Vu-jEFdh1C4henZtNM(#~S$
za%4;ijR}<ju5WOX=sJ()k<fU2+YkbTP<?nUS(VkgV}Ixo2_|eWaJadlaS=3?iaXHf
z#Bej<WnkT|ZIPGlEt+yoG*hBRG!6u>Y|hVVeaDx=&+?cbp-pS@P2ytRa<*xCbbi4n
zyFKg88u7v~Mq2NAa=u|72VT7Bd2(?^9tOHq!>Vm~_sJQ*`q2y0pg1zBjs%={9!-um
z_`-5}-n;0)d4BoB7r4;koU&PM=z_4_uGp>uZQJqU#g3$z-}c4#*zX4R<46M5O~cT5
zKDp|F#5)%oobO3#1mJADrfI)a;wc-(oY;hxo9lsL@EiuuH$J){xJdSaZ$97Ad7;B|
zJtQ&~H|f3OW`9G-J=@iaPo7;{{9()xZ`f|m2mzv626#*n@U+enc}&|-I1C99$K$h>
zAdzv5JlS@14xVm0_UFo~ZCG`lM_b1sD@wLGt@Z5UKpR?O(NH~K2c&6ue!b&*KUyV~
z--?Ir`^-Ln0O;zZ{)!j!YZmHr#^&NE_qe!Td7M{L8EyhZ)$caps@KsVuQL&x_{n`$
zgQ{T$H`muM*=){u=kX){m%sCW$N%v!{%`r{>LZgk2!7JOqq<&aE{+N@r|VpNn$==(
z>c1zUW0Tw0gr#M2*H(bxotxZ1;+!euPwXMoyCYI^eWhki&y1fqhZh5Wvx|3z-L>9t
zt=u97BuvRe)kRrLo0a=vjGd=jwS-{V&M}T;wX${IJC4Sk9RISw&C~rxftZjKk!%t;
zXvUTP5L}66T!1;5Wnf-sRo~kH@+@QL%B{`y`&&%l+;U|GTQgg|yv`HpnC1Q_z;fx5
zUH*N%4^XqHYY2nt?KZrr-Lv_;NUD7ex4KnrVUzF5@?E(3ky~8jcdhM)h^!mF^!N$S
zZw~Y^<D&8#2auTgkuQB2(UH}v;busD{OL>fF|lqtZf^Em?=#P(cn1*T7>zaCG&H`v
zPtu6S19><=+T-Fi&3a9r6W*J<wdppb2u=??{qh&Mz8+}WhNK6&wqfY^Fpg|Dk3pb8
zScMhuZGVt)9QgR~OMG(uG3&Dn;$TT!V+7I6i?L^Oer5^;hmqa&bG&4erb*zN4!l*;
zs3SG_OtbQc3;f{vC;0L0&(rZfeavju4b8hk*En)YY@eJVhrlN<_w=K(ZdSCx@y_E7
zo(&&;dd17@9bIST%&uNu@mM>~wrlz^;gaWkW0~fCOvI#gEo`?fDFJQ6^XmgSE3OT+
z!J}kWn+-W<US1Dmi5$jEXadhZ*>T=^o?fgOGsH3SJSNW0)*SjC#PQ_u1!K&l3>$=g
zuz2w7g?-Gt>=XT%2|n}1_ckWEn<6PEhNuiN@oOJ_lP0ueg>4f^S&4_9FFt*To0MtV
zmL^!07uA+d9H~?gIa_UbyloiT%#ae7XD!dJBmJ0hlDXJ6tXC~f)A8b}r|~sb8h+q|
zmKQgB5aFHk4f{inYXm1PzxwP`9&NT<r%WC*tFGn6KH^))dew3>C}aHIVDG=v?Q;i!
zde85l$MNk0z4P<?C#lnF3}q%>lmI(*rG2X=fOu!pDFp;-NvNppdgsw&c83GIt7}%9
z4gcKF{qy`E{?h-34_|zfL*FCbl}cxpfKhAkR>$LASlDoz75TO-U7dxb7yV*!VU|E4
z&Q^*wtJfJtRh0Q1%c?5NEZ7R{Y8Te)rp5y;u63?oTr+%azYJ45ZQHqMx)XsS`!InK
zFrm`Yp7ZlF_WM13-%qU46#plcqg;-U$_VBeSgWEr(K?|C0ZkEND%Wnbh4aS8S@O#a
zB;PocK?~DYvJh>(F}tD{1nb@VFTT1Z>U@_iS-kquUE4XC-KvX+;&so`8?|BDxg5JB
zL!EwEV*!`#JZiIu?N_Q^h{*9q--FigBdaq&p$(q*&bD}g!w^}8EitRr$xewTDWTio
zJ7@%&o$&Q%J>TdlEO90dW2x2=$;tHhQjBOE>AHqNBKu(=UnxV1<P=$-KPGf5nl|yp
z)l-`Am@lj|U%vQ3*71y-Vf$!H$Qw4{1NOUXo}6t-8Je)>u-oyUefTL~JN#|NVMG;+
zZT6mcI1p7C(nvn+a3aKh&#vDy^gZj%2H$OQp|v<SAhaFMJI*(c`I+q>;D^>fNz<(O
z=*7;G3q(o%fqsB@-+e@iBPnMdU2O3x-0Vlr)*Xj~#mhatJmb^dGeTIg4xZSp`1pCx
zlk+WcR7Op_ce&>IKB8)Inf(yyRxN_(<EzO2U~!GB#?v}uIbYrwi+8^67&Y_J%LCFV
zhd%ORXrS5f{@IFaOeW{76&e@JCE5=RSvaJ`M>oBdircJMxfM5uk(455Z3|@1+lIEy
zs5b1!#5!lLZ+afDH#|B&XFo(h8HdQDiwoLD$bIDfM;q3ya5yMQ&p2Oqh=a3LOV>Ki
zHyv4peIE%9wk>Ep5X8~014D{P%)ICa9PsG!jG^z@jj(E4+Ev4g7dxJubu>-O-}w3s
zt~o=!=h@Xu+}VnE9(N>!aTs{I>G=4iX$(Ht=Wm~X7O;2ZwD8~Dg;esZCyxnOyuxnH
zmq(!O?!k1#tW0g2nXBbCM|{HK8jjquuQ0<{MId#tORXK7VxnzY2$?(%i1YlhpZe$c
z6Myxe=SBa5tKBt2Nv>#`1`)@h@A2MG-B;)AloKjiH2VCa4N7(u^c03L*^FPaXDtpe
zPhoCJE#|HE&z*e$NGYFIKM+dd$f93-DXr#;p!Jb9BP`0vd18(riWe#DX91B@87Dnr
zDr1bqaV&jfBDKL;CUOM``KbKdOk+-Cr+^q6ZYF2rQQF2E=u{zVi9;;=8$K<XXTDy~
z;=UgNqZ%n#fFC7L-eqDour6B0ikn%=xtC84e(U)6+v!#lD6Sw+b;|s{Z)c$h>B{2y
zSg?VoNg__%Vu7)^44tnZN@lvu%%`ny@)G6FU{uvu!6P}v3i9o5=dASEx6Dj_huV{i
zH5ZT82ph6SR*ffT<=IQe%_t>7sX@fE35hX-BN5ge7X)^GgLg1STTuDNv)Qi6<G^lI
zc87@Ok+W@w_=Ya7@r`hEb6~svfYrL;%iFK;c=MR|x*tL5c)B{{QP<(xmVPiDz$l3q
zFAjWqbKo#EoaGkDCQGv4?Rn~d6z~5ll4RWaoCZPr9-5XmG^C*?_d8s-!lP`?pU^ff
z?Pg;MCStMDDvl;Je0lv9-tT`m?e$~!P6>@=1PhtR+bxgZzaXZ_Z+!D(&bJ%lV4(vN
z9Bm_9tXe*~xnazStE+3?y}ZPE$BQB2x{i-t?(pkK(*=|f=Q~0J?_V}Nx?B;XS`v;%
z#%RL%O$dxh*>nz%a@Zemq2bp*xkfZ|b%4!T!`Ws{KMdULBVXS=$8{b3n24>i?pkhc
z2BNdjpBGnG#3A!&yQOOzVotQ8G@-$vL>y7!>Dii3KJD2bGG{B#cH8p)Ba2sDuUdwo
z=kdh_!#E;ddGzRvm?GzE$9}I2W#OOV$l0nRczCh90S#8|1EFmLPr7r4{=l>A9sA_D
z=pyUT5}JT!hi^9Yhdq6agwWu9gL5Wd^!(<A^{QjjG`xSdMlyW;*^Zk&@_5tYaWuhO
z_`pLF$mecvMF+5!XS$EA`)p5k+g*Eqk*#|U!OSk)A{Xugf8UB7;{i+3KpFcU7lI{4
zpon<d5cp$1{lDf<{FQ%!v+el=V6rOvz9)@|m{YlCnE-ap7B!c1iQ|)*iK|po)!LgT
z>SW@&S?5F$FOFQ8s+tijVjyh^enlM4TTKk4mTof;U?RP945QS27TuxQVh?lq-a9{m
z0=pMF$9^p|`l<|!_on-0M4^E-SI%3P0#*8A3HDA+q=J31$&%{YR`SFM-`+{fA?4_z
z?vN-c8{W%D*Mls!5xIZ-zRC^zs!>0E?ga9JNxOAjm6`nZpVea_^V$xbudm#ED_<xl
z<y02Wb>_%bJulvvq69vdcgt^)IhQ%#Gz=->{EB_w)3puWkGKRP)=~Z8K<HWyW8{3@
z@_ev3uHW;;z~A#b8~&rm?{YI<^Xt1`<2RlSe6w%qqw?wXz~$O=wjKDv^URaA@TIdK
z=X&q?VDlv!-?DPB4hK~K3g7tjIUhW}Ab4RnS|y&WN^qX>NunP|`aaq-n@61UjAP<w
zet6GUQpaBRcz;bw3CUVirAH(tG_~Ym4?fVgLTD86dveMQ{WabP+EvF>|A?Que3z@;
zFVjbObhe^v1J7O_=;OfIs^#fr$D!}pwx7^=j|&|>tWe;r3$(4{<?eurE$W&kASv-;
zH{e3UkAGQcgW$bm+cfNk$d^PoUpEAl)_Y#u^c-U1Y!%2^IK)H}Vbge=2q8EQedLqt
zkv3#=KX(RlhLmxd=)ACMIu4@=gbt&k=gN9z$M{JyyD`xet-Vh_-LrECF3tlNmm4CU
zloA(P&y%x=xWJ}$^f~fq)p8iEic1KNl+|<w#c@3joUdBCfYtI<iv{;8bgf5FcEiB+
zVPw-abY06~Oms~^G_xNP&vqlLAauc#RM1j3?nl2UJik70-a0N;fx}^-X<L5q{ehg7
zM`x=tkBy`x#H_SlX}n`UX5PE}D!YeTz@NLl9U<sae9`3378?E4-St}k&Fi;-+Y{dL
z-*1Z67Lj$!3H%=5_bta6m4i=iEm1_FZA-kGNJ&_TXbFD*W545{wfdYSy48xL!q-0e
zgg^fm{v!YB@PFgr57e??6<samUMAf@XSH};iI1xLIPXc(RN|+eQVZI(d9;8!Syo@E
z@p>|WCNF+w_fCM1sttoWS1-xILXHAVF6!FvIhFW8K*Zz3lT#v7+*WFmHBAUaqFtM>
zv4&D4W=h9bkaG)Uos1N3*XNv-n=;1EStIYMYrViAw^?DCJj@mY=mg2NG!}FFCdc&g
zD85r_g_E&<>V`d82-WSUV=s?a*~ZDeuL}@6A5Tx7!~0mkqslyT{EP?`_xIwGD;ewN
ztt2udZGk0;cuS6EQUCY2)Zh1pd1pcQAHTS!KMZ7XmQ>JeXj0<*tV6Q$@@7ZQ2E6*%
zvx}LpJUJ%~YtrF;{{A2T3&xHsjHzKz$JefeuU~mS*hZeNGv{5AW75A~YW)5ZS@ZNf
zJ16`dKMYVr+*L}6##}&@EySE?d4}wn8`WwNBSK7Vd6xtlZ=fWjc-N4NcyxXL`YHQY
zNV%bk(+gii_!>jBcT@-lSW+_Ctl)ipk?~VMY@+n$J+OD!R1yv)m4$81fm*eJzQh{V
z`scwks?4SB=kXS^RbKKXNzb0~c`;QNbn_a{o~?uuoFl1{Gm`|?Fs34)?eV^Z?nq&^
zvmQORs^%fF(&<Gyr@Uu|nCTkRL5wNW1V>gOo>YW#SiNV<4$mbK&fcfh^E1Rm@Fvh*
zuNNt0_7s#nmHjm4zmy78lhqoA0q;Cn%}e5I;+vQ-cnE=*jC~B=5yue`;n$wM{~13?
z4|ZyAO$QJsD5g#479g<Vi&xTVrykt9Uht<AUY(49Z+kkZz~$&8Ir;wn_D`;@vTgBq
zE{>>(1bg93^j#4*m01M0A9@B=#5Zi3md$z%BNkg&5S4+z<?LNSU<TamdaP&+t9^@y
z5OfjOC&g_&gR@DxMgYl^yo*f^RvIWc4_*rhE1SCVQeXx>6>FHwn(5`13b>nkrcUT7
za~OIMu~K<4Ar7|Nt+A9j7o~qUF>h10br~Ad9a;>P(mT}nKb$k1=d=JKhh`InuKi*=
zo&d7N^~}G_qi+Jec9RS6m&F#Z8a=WYKXc>n0)f+BcYs0R_!90rpH`4Oe=;|v0_l2a
zPHGX$V)0Sg=;fq!t8In(oAIxW%>b0p5k)O#FxD7NJwBIK5V&AXZ*%wGwRZEX|3CI$
z`|Hyr39xs<9&^d(V4)SMER-A&QMOt6<X0Tu%*x;R&;CJ61QU~65{xz1@-F(Nm{65c
zamUVE!&OxCw79I#mr?}a3)7ym-J9S|7`XzPIworda(%|7go6ZUKtYuPcnt8?vF)8q
z$t#xJRqwmq<1twqf-j{QQvqbY7Vrefr8>|&A1!0dasimGFwb4%8Dg3qFr{RnPfX`L
zu98v!AoHSn(f&HrCrnuWm+Q5ag<xGo$C$_p!8x3tLoVE6J_GAWEx<falc-J$%`qu$
zu;)G%V6`l<IUhz!9c8<RxoFNd!A~#Em~ts~X=5yxdBMvp%~M6Vr5w{73-(ZtV=T#p
zR#)2OytG_X#(PgpHb;5qNHN;HlLo$B3AP7&wSDdYu)q!jH#pQSK%jfG7qpX_L|}i<
zM8)mFVGP3w803^yd9~HU-)3Fp_C{}bJbPwl0Hx(c;84^-BAU1@AilB6A7U#+pAzf!
z+9q+|kTLPiv-KMH!X6^=o@KYp5+%%agv&M5n5mjz0ai)FJ5!_Y$DTBf#ieQtp@FF~
zsn?aBH}L4a$sq`}Du}7>ixZsl1aF`;Dka}ycKhXQT7mV&sao_1F?VlDiIg)bCQ?%R
zVPJRI5rQekhv4xcESN~ZIlRnFR92>>nOireyf7!7M&LjE+W#Hz96x*UGbCprs!nGh
zykrrNSLF8$fCc!kW5~1>)hZ<jND@%Ni>>0TD||k1c?Rk%x&VR8QygEPeCr8J_kzFU
zqg3!kZJC?^yV}e}<7yg?l+O&<$yDM)sz!C8Vy2^&HYvuOO!bPz_7$-l>^vPezjjgf
z-DIiXku3M&*8LVB4tMY8nZu~YyN-LuXUEIZ`8@UCyP~P<Se3>1cWtL9zbmk8dfn%)
zcD>F{!^KXI@g^w+b(_N-uE*u^@9OT~Ot{*8*ICE%vd#Ck|Gr}cAeWmii|y`;rl04h
zx%fux?>|o+ANh{YfbyIkw}nMvkK?7WZ7Yk<ULW!QV4szJ{z)LzRlJ!jK_I+h)?8Sj
zapqc~vn0hWR_4?_1jj3<*PZIr)vNadztfe}eKt4g4!%bggWoU!K47KL7QkhHy(5n!
zZD<&iJ$mb4b#a~g(?cG$QJAZ1NO6lg=kUSYHB*e7_cU!o&WZkTV1IK%93vu*w(D4}
zR+C#t5Qu_zX>zyL<dmr|l7fzm^&%<#DRaiKYNjk6OZ{P!rSZX*4XL<*Q_A=d7<Um6
z)~glgXXi9cJFWI>jHD^ZEHwqCteBUJf;wifljrj5q6*2BdQ$FzuOwjoyf6o+u3sCH
zCu0*WY}gXG&1{*tng|0fA3qjzrt4xoe8(cEnwEWTv4C~M<;=cVSt-2sd0fWv-T7TT
z=mH=v+aSVGc);8gUCa=tP{BUZoK1{-`N->>0J7Yzdifvh!5-|v9_(&=D=~nlQ;Oh<
zH?pid^6LpTo44fN3F_1pb;OcLJ;Lj4o-`H{CdUBsO@ZIhaVLdeZZETqrUR*QcCIjA
ziKrPb!a6iGUTE8fxZk6$WxZbWi~NK9=lUOVBQK}s>Mb}h)BY9jhzP+4lIdwst6+3(
zJ7x1){N0Wm6XP%-A~d0aQsdXKL&+RVw!L#6?@B^QsjQMMY9F^)+fC)*ioZ!p++Zy8
zd)s#9Si#su@$CDaCNu{A&d*t`S5rr(oaPc#g`uq3wiAz4-Mka{u`#l|<&`zG&6aMY
zngv|S>CM>&a5_4UanO-lxPl$?+RUZ-ToO=c5<is`V?MM9ljMvp+mGIQz+^tdt-syk
zIhg>TF4(&Y?kBpv{BvuZ7ce^Y;e6a9jCk_w9Whw<xUwgK^YXchA@5pp&_u&!x?fsd
z%`C1%9(aTYd$0$4uvgjJNdkd8eN>KPfN-D9S&%w0X4AYUM^lPze>DBQIr~@lo_||_
zrU?d|OlH%Bb8?m=pla=(7|+7!)Cpd-uwRq_*+K!u5&+yb4Y`kKO!PR0exN<;*seFU
zp+WmRxi3#HVTz%f$tRg`Xb2961x`zep-RczvE6FLr3g*iu-okz`;q?8qaCcgv4Ex<
zSj*a}i7{s-lco}KUYr$($3#kDE(JFME@!Ufvu(q^SH?85SvM$3%2r}3gupNkbgP#2
zdc}IZDu55XgBUYDG}FDP#Fo~7^DF^V6Q+R@oKqGdDmd)NV$1Baw4iCHSVAdaV_6ZU
zBEEW_(?y8~N)20GECwXah)>Ee*E|mp8!j@5B3l@fqNUdJu}|d~xE-OZiM_m=?Nbz|
zVvU#pTxB2i*5T#k?^C<&e6Ba-G^FD;h`oKRI+pV-$nCZCy+W#SuIdmk_Z;oP9_+y$
z?7`kb5{O%57v5v9efj5!RnRg6FXyPa#MjY>u$WlpoY>oQX02dx?=?(ExZh53=rR1$
z@^kCLsJiyXsM+rtt>4<pK=qmRs^$8{3)ZU@tF|G=k;{uSR&9$L+{A9lQJFczCs7s2
zc`A)o`5!?DC3Y}X;b$rFAfmKwi*uf~?bz@4>~?z+*)oZPCo~PKt|RBfIF78>D?o{3
zWE=-B&d;nKrdoZ@Sfc$eE;cmY0+y3PKUhV9#wkP43G8H3m_Si>hXYv?=Vu$5rooli
zIVV;ZwGGV_aJonyomsTv5}iSykTsdKOaa>7yAtdSop-d(Sq7x@7Q9Rat=4~9g3h#p
zYAJYfVxEgDxl}sIDHEK|RZb>!${bK54pLrJ2w7|gNSMcNm3UOQU;<0<&@yMYmd#t9
z=GZ-qCBMKFU$?}J%UC|GNO-Z$%sU^^eiPd~ZM6WU1qU8=K{YRy8L9#`+Z;baQj;kX
zTQr93K*d5Jq{J1EH9oM!o<7)vJ=lXi*q!$F7{FK9=N+rFoR;K{&zAq*cZ4_0SGn6x
zV-ObsN;NaBV%Iw1)>*!~j&n)~mYF3YIECOm=VxcQokLPRbYXaE(nmSY3`(jTkDc2|
z)y-o9NYzbh@raJrb@(RGt=EjhfU4>5nas+H#y2L#(gZ{xct_)f^~wW&%1rAI1FFyl
zog^1DTezMR=(FIx;Jm{NM1?VqRtL0{#0$X{$%UHOa?A%!*GrW|nAx`CNt~!F`!7$O
zQ6(Z!C69?b7~IX0OSWEG+#t*1b7f`XYC_GFU1-}l@1~l^O18=(SbU*otIs+MRLitx
zIT9lGl`*aq;WUsYFn_f8nb^9T&dF3CcOGM}uHK>d{5}n$(@*a>)Eq;**g$VTYxQU|
zI9*&@wt=mA)~%GPBeQVx)_&7$vL2zatiuO;um^jv2fN$8BLLv%ZZ#nN-udoz@Q;&;
z_cO2~5U3L{ktxB2Db5KXjf!tQF(q@m0zP=2K7NG%#91m;MOt?tYOC&X+vQhXvM|Tx
zAxpPhk#GqOO=wAN!!TGGqA^Csainb<(|MZoupHv%#FPr2o2Io`L&~Ht2L3{5mN^?m
zr%dfUvAB)BClr~Nny4{f$LyB9`(DnJpe%<$1zUxNa=r*xm+zc_7ms>FsFKARM?Kuk
z4coms79SA>;tWES4COxLQf);5<6s?%kTe#yx*Vy1V@fu2S`6Z&yIIrj?g4dLfbDSy
z9NW?pdqQa6-Ji>zD93k8;JP@Lo*ejyr7nNhob=^5RTesy6@cr!0H}~w%RQ8!NK8<f
z(+7L72YavwyVJfy0O0FZ>aTv2zp?P^9r1$o?5|Je`1WR#(}~qSJO+Wb-NKc&_$TFr
zcY(3&lXJ$qz@y8{lC-hZ@T(`2>Esi`H~nNZJJcJCWo6l%xq<})StSEEFOKP)XVnQQ
znRs-HiJ>HrG+{M?6>;8T)Tr@sg~*b*8pniJ7m!C)8HU6-Mp6-OMI0f73E!P&x8AAw
zRvEJ+N9F7o)zbgVy{6big`CViELDfhVexT#97-^q%K=FzO)r!wX<<G;Q`HrxvxXKF
zV5iy0I$2asaTudlG3-h9@5JmfGij&n-JBRP|8q2&m0NvyF959Q_=r!R+jbuYKkBo)
z4qn+powK=X+cj&^C@bI@i~BP`U-`w178LA*H`diZB!WEHgFV=T-Dlqs0PuC->o*U*
z<YY3MC(CI9o9LQXwc)d)5^h120_cu%NhWWZ{gN@Qzs3dfFo5$VZcSNr9Zd+P_olhT
zP5HS`ryzctOj9aVHE^ZmR(1Q%pbQqExU^>is5wp9z%>&%#fTQAeTq;l;w^>`Av@F2
ztH(_x!9mqTU{lV-G`DXUM`Fr2AMnmkV6C7_YIg4|_ATXB+P|j&sJb0*gC1s}sLIgy
z<Y7Rv^6$U=_xShq-=lJOwe>$6ekZ@H{W$}R-V;I~_<-|PpR`-82;Kmz_a5&9A$Xj(
zGJk0pCPq|MiBlY+v0J`WK`?K>&J0*d^O4f0wcR4JsA_Px-QlpkZKb0Y#P8ALao_&8
zVq5EdV0Qf)+j|@#c(fqW`IBWrh^Qi}MJTvF6c+yAoU<e{5tM{VCYAAzn&znF*DdNE
z?7<%F!M=a(I|2YMRoU;gZ}DBetORoVFz!UA0$w>qz6rzlS^{yw_T4)5{GEMYfU3&O
z>S@giHl|1bEeh@bgZKV0|L)=6;FIBFD4?MVK~IypQZjV26<Db99aU-G)zUturauEx
zhlph*YdOZmq7_EQ+;N`WZa*cJ07UOC!?$V!o-k36OB`IDk|k0pM^r(l#nuVVg%a{r
z?&)g)d+p;T%eK75Q)Q15UzlU$aOfdtB<E>N%5*eyIsO;?zxglt&u16p^fLn=(_D0X
zF?@-Cp!-8KO`vN#x~Apw?3~7V2#z*b?4*dJ^DT-Mm!GNz2uv}GfJjl#ua#5^Vz+7l
zb(^`L>te%XQmo3sk}A9p7~TSCN88=U{w)?(i__HmG>M=e?Ngu8S@val_S$wW>}Umn
z#sCTqUvg<?@R(gjqm9kaI%?y=9_+y$?0d<+BLLv%Wk-SZI!{?kH`E*9%DaNze=~Lq
z=-p31E`f`A0)?c>GAFYEzdA9#`_4ObU+-{HERa6uISE9dq|#Dr`>IND0T&G63}lHg
z)y$mI666FBh{Xz~Xh6*sVC21!lCfB|X0T?9SIn;H#1-HQSQQcP5$6dQqp1=|sW-rA
z3ClR=3qbW0t(MXbIRPg#$eZqzo&dlqts({*YH~@6i8Mx1j7Y6Jd~#2!0P;x=h8<sY
zaUW|vT;<o&ukvSyKVwNdVmgPxd7k(u{6iQ25UW*3+qN`aOK5^+K2u8U`QqAE1MxN4
z1O%(7SO8sN0&DD@F2>lD=%G4m<(1=Eg^-EfJsqh`KG-D_ST_d*6D6$uG_iga{3&ye
zMs;B?`?D~cqBEGS`h(M$6IXy-S@aav*CjQH7b)u@X9+qPoo=yas~q!S5B6XW_F#A0
zcLV@@Ex{&7bH{Ju1k3&1G5C9<6y)RjiFa=Crg4b(Bx&5`2A{lM72;%tEZ`OAJw9hr
zQXZe5(>5M=V5*afW)uff%4BKgNFZh=7J<1qw}~8ET}`31wEzt%fX8ZmI@9=bo|F=C
z90?&9&^bDfOqcA#oEYMrBlwzy3|T!fgB$Ib!CCD}t4&%MIhh+QQrmXiSHL8utS4X2
zZbT-xud1RcP5Os0yDq1=Nes}5l<F*)5VZYTGIyoMF=|O>Sd{Z9F?|jp5&4FEgFiX^
zNt|<BhD-jj%YTf}1uJj2-QXG%O0Pqrmhuy<G@wdJn2@>r*;FH>)*&rzKN?q78-4=x
zuUgieGJngzPs#J;B4(0jIvLKl7dXdUUyTf1E;L?O!KZ#VbELMTh$G`Bi9EgKTAXdu
zG&;Bj*My+K-GV#8-QC^YU4lb!8#K7P%iumhfDqh0xcfKHop-I-fceo^AKg`bR#zAQ
ze@(<bj}IW1^}~-SmeKtC4>(ol6$>siccJB4>g2~GgNgj-ai4j@qnfcJnuqwG{`d-v
zGvD0Bcm$_Eh4yOx!O5mp5zjoJDi^I0aidgcyA{v?ttLL@UY`<PVR2CUVMoidhz(;2
zjI%ac=(QXp@#1P)j>DD!H@0TJU3yca;ja8)R`F<VBc*N`W3|TI*Mq_<Qw2K8yf5CM
zirqEm`+#S{hK0XBK%<_*|K{e_e71ygI&NkZm>lse0<c@+YS1v0EHHwLQbt+vwk)-T
z&mMXEX67c{r_Q`HN7yQ2iZZu9GWn~Xn9yes$%Q0rOTJ7D^iKYEe2i#)XL$~o_GvEB
zZ^b+KtraC;c+7fY%rHjjv9Vvl7*&MQTFS7jH6o8S7`(gGeC*bL?^!cQ=ZjXSrtX}2
z<eK_9XLM<dQV=cx&*J<yt+-t_7POM@a=A%Gj0|1oTw&uGMUge?Zzm|xx}xHg0_#Ih
zip9<iIrXJ~VLwOrl`Y;pGnqO5Lgfh!qBF4V4fH+8%Z#S9RHo=W`x9WJI3}vkf|*yU
zD((2%P{$C1<o;FJbmt?pUF)Zs8b>MGt&p|lF;_%*c=fR)`IaKJq%u--UjH1m6Q2Ap
z>60tA=CgIz%poAE^R2-qjiAx4TBtfgudK*sIftR*6@T!}?(Yw(4#^cd@#(f96K;=t
zL>1VT8Jv*y4#+#jM~YoFttz+&`ld8G4aULt_LhbW@i?JJIU8xtjEWzK`QWV>jQBK5
zBPj|~A8;?``-jz3(>=bo#F)TX2rKev3~$B|`;s8D;vT7SDoVPClriR9iSJK7vY?no
zC8v`-As+WU+E}r{J$D6%c5;fkc_f(jsfc%ND7@0o;O~6a47|=c{LjMZG&qUwd!v&^
zU|336ESrQm(K>0%!@m;xJTLezL#kReYj+(c5DMBk7dpWA{GFui!-c4)Acxm_N7#{s
zx7fjdeDln<Q*Rfjx9G8MPy(u$6nYa23=5Qfmw(fFU;OhKEv;&6{43AS;I!-4h2}9>
z(MdSJ6?LxQ-eXT#*qEt`bX9CZGhSh9P4kd(Z<FZo>ba=TH#LVC(MofYcISGV=EbS;
z_?RQbsc`~s*wD0veb_%}7IpL%+Yz;6mRVK7H~xwr;k5>7e~UVFSS?&?X|ebtZUfC<
zcqK2^wfG#}ll{x%<QifZq}@8QT))V?9uPgUPrsomgfY3jsGh&7x@E_c?o;JPISbf)
zcV$z@$owwD$SNs~tTLHXt;KNY2Bbhfw?ZXKjFxL{Sx?I}$Nyp8(aL!h`Au(_?)UP#
z_8`GF#HTY(m7~RlSJ^u5e%(a8xNcM{XFn{7ZIE>ieSGTz1o+2S)+`~FV1B9Bj6@C_
zv_|TL^d<;1ZNsC)lfWZ?+Ng?AR(i?@7n~Qo2J_~?b-UYNT&H9XBVxI=fn4jlkN!R?
zG}asNko*$&Fro2_emcvyB&sFDPqv)V{q`PCuC_`pQx6VQK9qn9(%gBW!mA1q;^}Fl
zpstv=LFcTg$*q*zo;({xmXp<QoICqjxJ0OT=oh=(2e+%>Ao|y!#BS1%CMz%vBiZrx
znf0Oav}Z!5P#RoRfXIUqsWoMGfQkz5XntD}7OUI#OWIndgr(pd5qW^+s<h{_)JkkZ
z=n^ahX=*L@R#|xX3V?doGOCjXI9r{BN@73Yvo#nLG=~4u5!_?;K!tTdfTwt0P8D^y
zhQjFkg!i7_9#~4<DA>u19Ur0&qpbU^r7aW&B%>TLmiz0o=Ofg!Z}R^`q*Ll(h8Oo%
zJ<mLZ2R;FTM+FQo&u4{@!`{!nxDUQLwVi_r@{=(i=eF^38#d)V&q-brh5CC_-#7wK
z=6bzhsjh#;RU5fYgpHQf2b!bM8D(EFxUHjA?K_!6xgrgCJ|J-{Fsv=9>#=K}iteM%
zMvJg{lf~;*`cJj_K|Msv#~Q{l>%R07cg2@Wt17dXr{KGSbxYVt!^6AoiRA{w6$SZ^
z(@ZZa5>ucGZxgN`-h?kc*sowfOysC^^p<$02osK=G%f^65?LkR793ylF4t!tzP~@F
zztdDs`7Cp(vQ^V|#IRA{DV>SHox4tPQa#mlSnd@iJ3_UFw0%2x&>6V&Y1$g&-7K>$
zRSD(nE15}C-S-}Is*kk@FT!hWds&Y@N*t2;V1Wzq*f0Bgm$@cy=6MYO{F||zHBRf(
z1tt~D{6tR(s7|Fld*~pKbgcc*`ex9U(fBdOY4vcZR+enjapMdwu(UVtYw64U20dOm
zHpF|@C0Y-tf3@+xE!2;YO#-TEIzFDVa*pD%)o=ty>6vU(U2e_w<e31>p|PVsQ|!t8
z@`*))$K7v`+m%eG!#=DizSvK)9{Fs61jmjDtQDcE<331aYxaL~WPP)_iePn;#iv>2
zRIxa|0Q0fGPVR`G|IsRSP1qTItJ~RIkTvM?>zg>Chlr3^b>cQL`eEV6P+D8*#EzE|
zd>w{Ay%ih2r$6ijo+MyLNKIOIfRBstxV5`RS2J(lq^F*8Gxb(n*6-5R$vEi=C%AL{
z`sNBsKe>7G8N4$r{rTmybR96qqA~+p{-yT|B~>M<YvZ)u2CAOnhpw*W-aEa25NM<J
z?vXtxYZhzGeefaYS71!S`P<Hloy4Q(ejA?Lrb%}BxBI4ZF56nz0H9nFb!<Lc^4I1e
zhyFTTJ*}&dlEl*V1fC7x7J^XfbcM4)YeV(~nG4-g4_KZvG67VZaU<<cNZfvmcZ$O~
zlTIycNLx6LgU4kFM>eYWTJWO37O2x{0;P~9P@eJJ5h9WXQk4CKhib-jws9Pi`mJz`
zv5wDh?I@y^OM98tyoB12R}q&p-ua7!tg!V;8V2;r)NeDVE-~jH3#+3j8oZ^0AGjWf
z9`JjgnDGR@jgY(+2=NZ#ty2e6OTN>`saJ|u5hOoURC!R~C=O5Se_Aj+x!^g9<t<SU
z*Bj0PuCDATRG~!ejSx;zRkMGYzIE8#_c04(zqoq(UWovB_zyFB)OO#RWIDzk88mJs
zX?57f_c*%v49(wVo%h0JBT{xh^BO?CF<4YdWdl3n@B=sn+Cjzy&g+<_+b@4(0}rX+
z5rzCNf5OrpyXjPTGZ4_tCDJd5%<n3xBw~+YXH?+3>HXA-hR?r7(*%s7GEMFQQ4lM$
zd_CP&r!exu()3@QDcg5byc-_GXdPs$?e$xl>aym=e~iCnuE(kiT!O*Medf6fyFJCM
z`P4$L6la1;w2vM)i56(D!6(_af^WN~4h}U@lvFcxjpB2(4DaUZa|W@AYjvhVXFP*2
zfyb4B<7}PHUt$m1MPzcfr4qDzA9L4yEY8RzP^elr_PmtqN1tf7ve~O9R6w~fw_47J
zNsIy;3v6fCiCMc{1OS`AV#DKuL$CuGhv=EFWc3@8XwRd@LaWX&|4JP(>N*x+9-1vh
zt(5$je$}Mc4r>}kvw&AdRJdi7YDzVe$)%reclqrb475wX`V<S}^Zcu<uEkmZRFF(n
zs$>hS;kMG{wPI#n$xrD=+pGt71v+JYDwcI}cK+g)a)zb_Hr1)cC`Nkrihc3lQuA_O
zu!AE>!O4x=3uLUWTfbk>19#D{SHJ-{Ds(jsSIO5mjg#;-K~pVL-@J6Fi}e{_RkW#K
zR03qI$}&ONZ8U^E4jtbK57*#A%#n!k^5pLo7|)dR)owGims2JX>6zB0eW)xsGKo!y
zu?_fh9527c7N>>ihMY{t-*-<dDDS)^yKX=2M_Wz{?hh@Xn0%~04p&X0s;E_0j}kK@
zlK6!&n?d+1IB-w%?aQZD^3M_<F~7=_?nipB#N{-TguY1)%S__t2)I2ZZjaZda+50F
z|2$?^*bydD;J+2B<UEapRwgqyrA$V?yTO}bk`vxYf2G6^%1h;tFp=9WDzXbw6rYih
z0w=0PDigpP_fG2>Ciq=la%wW87xit-z5go!=yz!ZAj-xDv%(C*>W}oJ<PLUQirEL2
zIq2RInpV!E6mqQ@8<rY{zmQ#gvmv%vn;ee4ur#+ysjux3L;Mi805p9-?$2pDQ&$&l
zoq}qXM(v>eNY%ZrrK@MnKoXJ%X8D>Ky(Q2u=jKn;7hLrgF&UFS(lO#ms(=IFrXQ3a
z<?}+*<^FV4c}+?*{Q~PCuD(DYA76ai&zGHVy^O|Otu15T#FphY=c#fl&P1|QHR|-3
zO^4U1iES5mOxnK|W22`B<A0Owy`e(1U!GIf>C^2UJ>>i1h<!4-=GY2&mZat9zTbN!
zli8d$6<?}_f<5=&5@_qsc&mtX=|Hu5x<`{Tn^BYIf9NJ!)!Y;Lw7PlVoeF%l-hN+=
za400M=+i>$47Y`pZ7v9t>R=@w52)agh{?o?k(W$ju>0plKerz2`)tlgZX?nL409d-
zA5$M)4?&fKaz=4E7;R=x<engWAZRZq4h+!QeZ9$$JR4szbfL#?1M1*AY2n#(y9lkB
zs{A7QHnlKRol!D9uF!O7)}w=2d6DP}kGqWF)!;vQ{4HWP2?@x4yb9Y^>cA{#q%uLx
z`6I0qDIjaIREFgGl~<lr%OYv~>YaKFR<cG48E-~s{FK0Mzy!!}Hf1tD39%{msrbZa
zY0BA|sy;t+#7|AIHr8+N<Kl;T26MAfv&Je86$x39hD_OIe`!-^mE@c)-24I1Qldpb
zZb7)<)^M*>dEagGRNHFXbLn-rnRM%9rzp0cw`j4Xa8<Ye@{{A~j^IrQH6P=Gf=SkJ
zbwbw4ksS03Ut+!gIXXrzAD_!P^qWk>(HpNFOz48A62eLwLFW4B{;z|_Uk9II<1Z3Y
z=xjm%a`Jvz*9b5%Sy)=$az9z>Zo>0=hU@A$#5S?>87}c;K6*&<@z6KuSlOWI3HO8k
zGS-hA7O&c+>#5U>tjurew%FC<;n=3(7!x!CD|Xg??2eL*of>+1k3~B)LDK1C6mT46
z3euFgZ3~q`E}9e2h{RBBL#47TG!xSA8CwDh<XXh1;B1Fe`VlEsHZ&L>%jFvVF9B-y
z!hBIo&Z8+y9e|z~-5@7euZ{0R1noUN`%8_AN6$S$MeUQ>NAUeDXhdiTCb2J-y>GSk
zYYt0>=iGnqt2#M(D0MR!&O>!qg{Z0O$@i{Z47=!sh=KH+h4}i=a2Q#RtqcKa(;&Z|
z<ia1Gdeu6pba=_UB~o&kIQElNaO$qFC0~k^d5GX0JP>L;|F^h8D-zHC{QO^Ax-)7$
zt?md)cr)n0PG2cA_gF`Hz;o!}6Bj!}2=<galq9_v?eRa<=rYgCRKL$y=+EiHk<`<3
zE>PJ%ftXcetrBHxe(s}+D8!!O8*fsdKvrn)_~hDWO37LjB5ZT`HWu!>9O{m#1P=TZ
zN3z8;b<Lx2UXbX^&Z7UOALt@8m=Q^<?a_p2Kc~gG=3{fs@^PO_#@Z&x`$AD7jcMia
zSZ>L2;*8LP*8Hj9;4$U^5}@aqsCr<G7B~ev;x#wnp5)q6=&i9#)1jsYvbsbahndiL
z>8j&V+oqF)??-d?ay5{WK>-IaWn<*LXiod<bSsIU#~c`+26OKb+>e7_r7gGrt$7d&
z*)+sh?Ji>UtmNRyW_ls!bLU!c0f{p1pl?^;Gh6BZK_13CJU!4|Ed!l40had56zfk`
z|AK{P9iiEuf#=*~HwUyozsyCT+!L}RrextX*r_!yHVAZpxOk5?q2^|VP_V<=GaICM
zY0vo8frve3DQ5CbaU%#3Fg6QtE0pOO?a+ug6};E5H7c3Anb<<doP`rI#I2n>3iuZ}
zi*EhoefFE`UxCZTBQqOrh;xGq!$~)VH1bwfD|c<j!%1eqAk)_uS7?vg`tk_}<#?YP
z%Z+=zG2DvGTO0Wun5?+WTX3qR2qf)>y{FRywkP~<njNpM9#J$-Z91^wLk^N`RZ(_#
zfz-BT#fu)^CywrG0%8X2>V41aH8rSq8f4}50VPWgWN55B-&ud1V|T8N-u76w3bEV}
zR{9?}g|&|V`bX>hTbEWaLvl-x`Fc9LU-eNZKCv=Qv~}xdYo>2}t`TV4Ys?Ao7xUbx
z63}!%#Ebo%V$IjQn-QKw5fbHhrM)&5iASrQr1bON{V4~@h%YF`s=M@ZE7Cjhmq->p
zwpJikvaVX(;{=a<@!WCs{7}!tS@o+&r>(inix%(q1T3`ijd)Z16Sl{pECHC7C*ScH
z)Ps~6l_QD=;gzft&!w!7Fx*fwiPA~Cqxas=qJs1cfu(CaROuUO&hz&O1_t`|E2mOf
zvUsQb@IyRL0h>=xS&&QeFDF1l1E;q2Eqv`|a2kJFE#E66EM^FI*XYd>$9NLn4}-I9
zmzlJ@sN$L_E?&p+7I3Xse4|#ZkwyH5Tsg}Ek8U}a`{3V01Xg&+1>w%4rBnC`J?bAL
z(4;$|DE~KM5PSxEiD~?G8QFCTnV{nP@Xq_RZNLW~TuRN$yXa54dG?byI4Kj2C&Y=`
z>Hroow`tFckLT4U5m9CN$VPu<+zZ-_(#vo>dSS}kBw!NaK4eJuP8@uQb=~?<9&1Xs
zvSQh6Kk`bc#O9xV8Q}V|Elx_jwu$_*uhpfB!F|t*{j45n5?e+mkA2wTVDh$~F$O(f
z$(OzT^`00B3ZK7be7Q|U)m}y*&0DYQgU68@rrO(71*+Kel!K+Zc!Rk^86B8s*qG)$
z{$+e5FedPM@!g(S$iIK#Iz5~*_xa|~F!OmWDw{5V!7*-BS>e2zWW_F#Qm>_2n@a`Z
z<>+$`3BL=tsA>?1FEpN!N)R5PgywJ2nxJ<sS>q}2D#(TwO+`jcNlkWu0DtkXRSG`8
z=!0-lw4D~!pf&cT3l;1pY!Gn8%^7_HbO_Oqu~lBZQfDE$3C7;-`EK8rRykg?(z_C-
zFWJItyY&+Qp5%#*_+&5P_Sa`No~_YjjX#-PINx*s)O%OR({6>1L=hmm#H+8c4JP(p
z&ywtBGv{4bMi}ORcf>7zvr($yl%k!}V;LQvR1G0$HiXEx3;`^`(XLB18-91x!gbI+
z6pl>rgK}fezl5rcTG>kH`t$vypo+ue)bRcqf{VDf#q#YxcA*b_HaKtoJR^#Hg{nkH
zfJ=K{8T6^%x!#RSQ;9>{oFWE(tDoBcp0s`9n@>Y|uAje8<lfg;r_E#JuKOwtdY7iR
zI$u-!Pfx)cZBOoC;+*=xc1hO~{b~YuZ0WmoY*U8IME6Tlh7+U33)EyQUWY2M;BLcd
zVRy6a@eh@H;LnQo?zY-Fp$RQ8-5x~=Yi7g3H)8l{2mJPqv-X}NLLWH(!h%@s%9vNc
z`#&6h*4hdJfN_2&l3v~1zGfDrlKa9OFneBydHQR7QUZ5YeRUpJ$vJbU_ceCTMb<cO
zx$Tfy{dMKzST8kqPmP*|qanX<PqoRwUWY}^As@=@ZU$HXB_1cS%I|SoR?ou{d&XsM
zVY=;$<wQp;oE#&zGgX?sHklc;#K<eFCA4PA?4Yka(0f)g8Rfo)soI}Hb%QtpZZh9f
zE)fTthf37SDmx7$OFGo)c)(@<=~?{Oa$L^a!k*XMqo8_bUI#a3c!jyvP+cXd$C|;a
zNrAIZTrJ`JeV_pQM1_q{@Uy@GONP-+Vc_IcE$Yr_-ea-Q)Y9YkfwLgj_!P@H9G;K5
zF&ij<3`5?`C2Hfk`3OKmr@9v5p~0YvORi9mZ;cw9IvH8o?I-L3f~9kc&ezcZA+{Sr
z`H+18tenWY2-b}V{Qke>HSndkcB#PQa~mg~yxC-3p;Ad5xG_jL!<^udXOWS23(C6X
zReVt48crfdeQ8HBVwetG<Ycglo^!NhaqT$0S^12*qrXVW_PIAiXrhIaYWID5J&$v#
z!R*NRPwT3H33{gYS@%M?jaVY=mrN;v9~K|2f=EXZrQr#XK3Mo*O-daFm*Brb2^@*k
zKB51L`I0P`LW;>@ifzO-Bb;z}T;P-YM<4Rr$pNQi`kqHR-8bLVrTKdOUG)aHtKS3Z
z+=h75s6Ed<hemXc+%fkw|86bnVUB6xJqH2FCK1VKeh#2Y$8M?tv_<e&L=Em1p~`+S
zYlGydR#yQeRuUzZmHdF;%Y6(cn=bHy_K?Mwhd*<}#Rnlnwx(oD!ZFal^(G2LV8NXq
z&kt4`EN_iN_8(hGV{pGrou)n(?C3c)w5KHj#iy$1Cd=8v%Qnk_2>rNL%w;{2`wu+E
zUQwSXft*_dJY9?K<7R9>+q|`5oq-p(rG_K#6R*?{Bs+m2)T~ep7r7vwpV!2+mN9eu
zj-l(Sl<qRcZ(z%t(44HVT$&<5+`)Ca4h&62DA_x>S3LF6g6N@}XBkd`8BYA)<gh86
zF>|P-7|Gng;<YWyHkH-Cm-#S53|QzoY?w4&;T|fPEil^jNDbT@@OBGC{AKI32svG^
z)h$ZhMRte~$FCZ%z=XbPwAJhqb;Rd(@;2dVM}NDYdMPkIIWEq-GV);rnc3x0L>T#a
zk{2$4NW(Vixd9eRZeRvH;laU}D+fwf@;SZa-#nn!x<OhRmjk6~@Q`aOhpVW0`^1Gx
zqsSJ1GjCK&=kU{Bn3?c6LIOsA)%7CBk)K2AmW{E5IX=pc9+aPNl>}I8W>5xf)N5Xw
zbB(Akyy)(vwptmIf;DU`99vJ^N)FZx>iZ4mk_I~Q=G3h?%-(1F{9z+emB|>!+~5!R
z#AUd=^LB>z>H%{g-~tq8Dg!jF!Aw}!-siF$zjy&Eml4VupC1cy0>2st>iTAVvth$&
zB&F`-phSh)dHdnSEzkV{lhq>9D~KOpj!F-3tF;+{#Ij{Mp|x8K{sQ%dBd8LK(VjB?
zs`v4j?ix-x-DaYH!5%D3&N*TbRB^|^uBcoY$lnl@)2P!Nh0D)zSlCht9408U2(Av3
zoKO*TaX5%wquFdA4&_YdX7?d!($kWYV(m-TfH@K^Q=wt^72Bl3LsRllbob-jPhy|W
z3gt+@%i@EYxuPw8rJ`W>?1)wstL;`h=BEBVTGJV+eIhE@nHlr@b7H%{s_4F6GI5u6
z&%y(LLzPCWT>RLFGQHTj_S9I*tOKh!dV!TDxa|c2Iknma9)xp0xSwROk%%het{=92
z>KeWMe=kfxPNgc&g*^{Io$HqK1%axX6QJTy<V^<^*IrFu;*K0QH~zeyjAN}hy9F#|
zeyr#jUfaI9?PcP75AI6l@1njX%U>03{KLH>hKoYBS9G?)y5mH^3S={3>S{2kt$Y0W
zIru@vR#Vr$c|n8z?l%;vwl!<MaU0!=GEoaGxyFY`_|?O7?i|Jd?w`>&+~^<@Bip5c
zB)jk-s5BRLcYQV|9q26*2A2=duFg<)>vWrvbk~2}VE2D%{sJ_*Vrcm!d>mGwn<xv6
z$VU-6Kn0_o3p~vPf7y9+63r@X>3jhW@g_wH@R8T0Myu9R8pcQAG!Y7Tp!yryK7O1r
zH+Lqk)3toaiMCa@4QHfFq`{bx+e=CAP%3b6JQo0ZN*sa#KdOi?SG!~&;*x3=s$xvI
zLFRQ^tBI1{X&u9Oi)n^oEA2El$Oa!@940Cm2qA%k6|c#P=f&1pwlO%kkAW~^0P1-S
zZ6%<yi^_C%cq_~iZKMBs$Mq$2D=?h-#(#&cMJ4z{zQ2=(w`r1#G4AT8IX+8hv85p&
zb*fjDVWCE<d$Zvbtnbc)v)k{(WkKtSixC)5qu5`L9kKWk;k5o2R$coR&wfD9ldYC=
z|0EkCDH!jgLkpd=R05GGv2GSPI3`yN6-Y4Nv{-BZH>4o(8b7tK0M3GDa--N(X!wCq
zFC7{~57)e=l+siKD(*n$qv3vz7zs3pa5_(DMCf^4XawRS<72KN-h_8GnW5be%|3|r
zDJ9CfLC=<-lr)J<Avm*<gf_FUf}!-u84eNeuC*;wRueJJbQu6q0h!^*^BWuQW+-i+
zcims;2i-e>WlAy=U2lK6$sv1@olLW>A-bwe*lvmX`)Kv3AK<tF%Z_?g0+)1Pu<}pT
zfV@L)@Z7~M-b>8*%g$Tw+hN88v*8K@fGksJ5C_CVV@vg&R?zznDfe$4&E_;e^(0WA
zgwX&=*x`}e{Rrwvg>Tw~YJ%)QasUpsDoSjrqC?KK|B`$2jb)n~(d7!zH9V9QVKbTk
zoz2Q><Ul*d@_+F>WlrzL1{|7!yBQdw8XW_Nj8Rh$NB33Cu0>NZZZenZIS(dPOX{5)
zw4`+I4*b^A!ucAxTalG$W9Ln4jPikbbyCB18J!i(rT85ef2Jbr(taZ>Hv09}h0WI$
zi0*Y%vsUGK>IdQXd+}~9&6PepCIq4xcE8H6Uez11#CXmR=^;}f(jYQa;mb(nQL^H+
zsQwkOz@TMu{R@Kc*HZIUl46&zT*7IHoxN)PY`dkwOruxny(^KeR3V$nuwNT`{jKsW
zc4{zO?yQ6vAE-JJy{zZUg3}y#fKF(cZ!k!#LlmtSH>?MK#>8w{`S3NHaTsz#)_IJ(
zKgfUX4z9C1WBYoLT;(X;nW%o_T<xudtsVMYzy%Dcw=1t`Lf1Ez7bJm4Lc{94lHfh2
z#z(b-to-4P-_PaN=>l%Fn>o5GZ*-@5+0{~dql4oUA1p^~I>|-U-kHtiREMpl2;z~P
zjUuJ-66wh|zV3Hd`h?b^e$)G39zV0TdJX`RKv7II9IB3RFa&LwFv0s^MpphNJr`!}
z?2LM?#3NLvro7+zw+)A>ls04Q1xyB()!eQt((Mqp=`Fk9Cdg5G@+oqkw0xGulN@J)
zR8!nK(V`kfD|qS*b-N`zCj}(Wdj$UHUF?_bQlW$mfj1@vg3Zs?3y4;zlqf%O!<n45
zU3*GdmA_yp??-?8d%pej+clbkg^&B3k#AAcqQ;C+Qv*IOTRESwEn|rb7fNHlG)Rc;
zLJLdarnS<jI0&@8>|xmvzp$gV>~P=;*4t-IE&r{6!QbSHT$c&5bPdfX?4YIQzQU-B
z&vun!-P(Wp2<G7re3Kiw{F?%YVvAZ3-)v)IVzFihZ%7`(1p9*mVt4;0jb4XrkbFyz
zm{AgWx<E|Qk#PC`I4m^y{<QIufKh~C$@=?8>}4u)k@}v}W|!kI@rP6$hdeB5I#~h9
z{O>p^HY6A#7Jzn^#zqh0{a9LRIUoz$4h~!d#f}%?J4Fz;EP=HjP3+YI338D%<C*&g
z(f|6NjZn|sSw*p`_AW|{+S~qdOR)dC=+_2&!+UN+fb3E&;6nw-q5ahEL#XB*y_^c1
zX`Jtc2s>EMXlOxpss(;#Um&Z}e&_eN<%dH?7J8I{Y_quW0;}QLYXvwDQ_)zYR%V0E
zp9zA4wHvQ=Z#1u%5+9XK80qO%{;-%b#3;7>h`MgZlc-`iG&9=)UNT_pqGFWC#)c#%
zWo*^C5sD_t?<g=i9QDf~)4P7>rAA~ryzF2K0;*mdeAG0AT7tJqF9Hr>rX_xLq5<~t
zkjX2XZsGyn`Y^aut&pIR%AV)nvoqB2L_V+wwTB@cKB1A=34w5^f$@?>1nvwKZ-s&w
zNGO{Nd#73z0fxAi4%^D=rHQO^6gAVUc~$MdBUyvlO<PyCS_NP1U!Dyg&@cAxd#j@8
zOb!^(ZPmv2rQv2{)~!1WyJ}v!8>{z>C87jboyxDJGi|wGq%f<YLt+yN#pE<DDYo2j
zVi5p#K7lo3lO7eVw0FV)Ot_H$!e6Q1)S3q?+ELa_Tt%5-tDR!;tspbiXU)}58eqk_
zbSwz_W0F<ft;Z{`+ornu`^9jJ5q#hUTp8PE+JX&-=0VL^jIR<={27EJJDAgg{?Ud{
zHwTxquaXZnehA5*PTb{G2UFNbWpZ$9YNtl^leC9$c~wn|eGqZ;HB4X`*X3zHy9Smf
zg|UXRl~a<Ar%Dg$4rNP&Q0;O;U6Eh9&=GSK7RlH3q4ElmxDQAGpvP^w$&V_@q9(Bf
zBbZ5ShzV_hBs_S+322V)^m7bBi~tqzMd2b>JodWpxq#nAI5UVRVkMn9Zp+_`0c_{s
zh#r4sC9no$ry|7k6P82jd-U$f=1R{@G?7^JglIoV0uM-bgu33)FZ<qJv-BDfIhAlq
zYXkEqb&^>s$guacEH4|V@KXpQX-x^p(MB*wRL&ULLQWVONAJjz120{-6XiSq*D9MA
zi0dI$X8f-P0~S-`pGJF=UPF9eO@i`0D#n<vmPBc%l`GEe&95RcTwo|eV=V(}Pv?Nm
zX%z`HuePdRODCIMH947EVVgjU*Z6_R=M_G`o7ww|O6EYgUe70eRF{g|KnOB58QgLc
z-Hn}chzaPd)W%9(v(x%y1A4%!Q8w3cN0s?(z^CH;7)46))9dSjbLIvODXK7Bv|w3E
z4Ets&8<Tu$9Evt_IH&&bvo+sAQ99O1DQYpqxA>#PpmUg0fNTB+^uDZWA@nEr9LU~Z
zFy;6Uv2mo*t*}#vnCuf<*9f-w2r>B%YsIptFKB`Q{|(b<)dg36OGb8fM7kfJeH<lO
zxCBr9lU4dHRtESIHacGgy6(I0FM1yq-wwPQyj7T|abG31js|&o)~m9SCk#7_8zNxY
z1k;XH*C$Qreojh7vBn{l&PDDxgPgKh<h7sO`c$~?d9?eLtI68PLH~t$<nQMrK*e+U
zvG82OZ>y+Jbb$I05HwyrtPy_iZ+cTJl;};)mDZq(kjb&a*A1ZuT}13&n|qfrQ0OTV
zsAjHeHiZp%kJx9-{o&!^XTrs01j*Loc_8jP?i&Gi9L+D{$=Erk{!6cZZh0f}2d4Z=
zhqI(j);w3RusyqSIT`e+&nmxycsN|s$L_B);bc{$i^F<Ld_kk?=?=5ku4)-VYv4}+
z$Z?H2U&NZ*jALTvaSq&<%a@C-oCs5iXw1?9sF6<JUE0EV)?Cw-wEW|%2S)DG$8K3-
zGZZMAd^*c&6Bo}w^mBo`Xa6+^B1q9VXcKOnBuojeIiD>U;E<o{(z`>5QTyuXP%*j5
zU;f=><CdEU{UJ^<5GL?>-_i8FFLg*mC-CE6Pw?up%Gfj)S0}UOh2<6^2BUsye=^1o
zb6@)B^dzG!f`K1Syz&{tMWlEE@E-2O1f5#z6!WMz|8kjWg|8geq?RtSj3Idx^}%TK
zJaMP%uCj{~ZN<IK9A`Rs$rdeu$>>?8vh?Ba=6M$qdKLx1TqMXs5ILwbz4c*tC*g8-
zgk*~g-Eb(cO-RCRc6)_jj$_M<-vlTj$4W=wq98&TgQwuN>sP%S55k-G(9K%Xvvivz
z8nNU@Ljjm0pjcglb>j?Ryv3dH=(U0k-A|4cFNne3N>TT0nt_>;><C&~q-yypP^GtI
zD5ga4L}%yD+(az73Pu)F8TQ)``=)^;8mtgvT6k055ey};cH{+(x%GO#G4ghdJQ)k&
zB?C_V915}+G$Dh88B}&HpFp7PA@~@L_nG}kfL@=aN;}Cc_n==o#NRLAXo-32-0=;j
zEATeBW8)KiHy|)TOgr`4ZQd|g+E<b-+1-6o?SkJ@G&a_GGSXBu9u*wW6Y=Ut8mA6C
z5cwq3@jn*bvTW|l1yI^A_9=jpY|`qy@z-t$O85gaT%&DvI*6{{eelegf2#7j{Ftz4
zctEL9u~y5Fldv#Q4Y&(o&VZGoHNoaZgkm=A8T9OV{R}A&k`J$t1H-EC)R7#)qruuD
z;MdC_V*(3Jviy4AC7W4ZbPg1M*N)%4-=mkK<Px0P?WIj-D~Mc8pwr{Vj(_n+FnO8Y
zsC_C02p?44IURA?hEY^B`>$Bl|5|8-l?x8G`Y|~9cORIj+V8}>b+;!Aj7KmX<;YER
zZHYAp;CA^5L2L&}_`dRSO`}^TF%*WZjm%+ouAz0V)vyJx51X@5DKWG(06Ks8-uR>A
zNdQewTR+wSM53)Rk(DKCSeju$x*CqLn7#h*D}VS;FFl`oBdBi%IX){Vto8TF)<<7c
z8lE+}oPWv@YL!vXQ!>SYlhVd{k*|19r;>;VSel}CpCU{G+E~(x;p+fbS``1Npf3X8
z!Va^V2`sxY1^2?}!R<P-h4qjf3<x^bu<08#josiey_Yy!U(`Z*qx<58&F>uVoo(uM
zo;Vz7$(*~rvhm6kaJyH#&9b&L{VXKuf&{CwxX+@te{KwiJw<9^!9A<h&nD{XwMLXj
zZ3&|;g2%X#Z8+KC3Py^A!h~}gO8q3`;pEJmRJpgI=e&Zudc;6jbcE84R?d(>e?9b-
zP8y{+ipLVN{W)+M*13|N@f`|?^D=cUlLk5`PxTv1d%ZAHg$m;=nc_hRHs{y$7lhk^
zd~xwb_roV*K7biuCPyKo1PQY~<O>Fw0=D19AG;PE6$K6`D$U=DwYsg2w4x0JgLdiV
zuYyC7kjYTCXK%PeW8cPORBtHt>j#-OiE3e0f|u#&%Hvm-`yT?TfRYXEhVaZ_T8miI
zUquOteNm)H06rX%8z044zpkqaL5;V~)~o-r*HFio8rA4AS8(KOSSG91m+-ET*GA3>
zydRG%y&(WTDMr|hv!KRslPH5UCL0b`rw_zlbG9w)^J}vE`wg3sD#UviV4mHoC}crc
zMoOv>4BlO1Q0c}q!N%KJ>&2&g!=dv|A%J7USeumlBe5cN5e|txoh5uJ>hba$mu5M)
zT1}i<=MNHLuahL#VeIRNRB;l^NzE}04N~rOtL(8)r<jgas?Tq$nrxvzuq6Mm*#@QQ
zPsDvEwT#^f8`vU-#M_#C!CQ5_Db+TQlTLS6FjRRyjZvb*d*RGi(DT%P78yT<JvxBo
zAa;wxh(O`+nOJ{u!Ichh2LhPixWEgKPC|G}XC=Lxeu6d8W|x@4wUrr|P87}Z%xg~Y
z3j+8UnyJQpziK%+o4EIKn`@r2+neMaeA^pp+tKgQ3YTiibRN5>E;E<zb+HO>|6u{4
zC22}Ttsl|t?=UTe#ITbL!S^!GfUT4@S2S19Uq3_qO@jfzRrr?9e^k7BnGY*O&@>jg
zShcM5H|h(mx3L`h6yIV6dWIph`$HP4-fIy9TE=250%335Pvylo?_gT!{)x~Q{96kP
z*wl^?^E6{Jm(g-7a;z_`A@`&G+^cI_1CD_R%z1(Xj-Lat0}tN;+kXJ2!vxf;8ynWp
zecD%`Rf$t<Frjr8MhDpB@*}>@L}bCurC-z(%RDkk^>YBcYnkauq;Oi_NmSN#`?$hF
zn_yO7o@>XSR#+!<k`Ma3MIQrr%B5VQF<GgKaH_+!M1kZb(z5yO4V`5mw+VoOM>v-X
zcP$*Y?CZFXovY+D>JDnk`0DLCnL>R>Vdsp<llLcKmYGOyty80O3{}xQc8|`_CDMBV
zqde6!(2VC02b|(zSB=#oM)&n60^v#NVpEmDT7h@>8~=nAuNeA{{<Ef`UjJ*yo#>hm
zo7UOHk@lbRU7dJGq;2GA%D^<?!KVTzO1C7~>?S6=;snF7){_7q3~#u<W>H_hnMkQc
z5<&m?u>e)_YwsU!7BD$iKz!o#$i?l_ibo-%2^V3JOn58=ng_M4w1O>pge>%9v$*24
zZ7QRi3@VWk!J4V?q=GCqU%6fFTX~|*mGO+dNK#nk2r>FD--6Elu5}-72GB3vUkC%3
zU!d5mExzsl4}8473Y2h@U3>rNVE`inj3lzCV%LNF4n<23ir{0oxIN*6@d0jg%y1w^
zrWGw-&EwlN(}hzc7R*f_o7rPr9`-NA<H)QrteURF&QRX8<!$xQLK=w&5<fAN3R6%=
zr<}{#{i>(QudxTh1-tAH9`e9Rq+#uXN@dxq*ow#}`X<y7#-M=S2v40;@h~!fbXH?K
zV~H$|D42CqBFVq&D+MRMyD2lb0wu;lK_WPS%<$6B)?Fkqpif40HP&oKXI+vl7eOi5
zJwDIboB>xqnf=0;PkX`wPxd$Zeb<s9nETpH>1^&1oTV+QRZJLcqHFO=%U~4uocstP
zzt*cvR<i1uwIxhiL|gkeEv<eX1%NaOgKgBQ8{mrTTwN6o385MOewg|q=L(9L{8=vp
zx7&0@=_9V&bHs}9r*ew(K$oR(-(!FBk3-8bWC7+7I#<Z|d+)=r;byC004@@C0Ddpb
z_YMtJ8E#^9?%a5Skt5M;FvyUCu%E}6P1E@=Z=4Xy=MeO?lmh9Gc0nR^SRQ-O!gv!J
zAo^zINAs$x*i81;+Da7ee8PU{ec1h~dKbLB3>ArSjdF@|Ga2WIN&x3H8}o-U-85KW
z_73tU$BgH?Cfv)^QxbcqUfqx-91a{^I#vMrcsz0=m50{`OJ2vV6Yr6hTTaJ7XAN;|
z6}Vv($vYMRYyHc4<J6<g-`f_6y(i6|zWpO?g`G1WyNL|xujTPwV12)u{=n{M?r%hj
z)Gt>Ny|J$W38hE0Z_gwz@?9I<@1yQq-HMc%wr_5<`s|vYpg82n%_S5<jAJuO%CvW<
zx()m6=Vnvg;i0xs#amF%|BJWH!5sw*;DB>`m9Vr<>=Lek@>yov5ar!wsiD!7X~WA(
zrG*bBL>L7=PP+{w@1bKi(POF8Ok5|k9}|kf>+XfH=XJ_)n@jMO^99czcbhFz4=(Wj
z-V*{+rD*j*l(jZ7P$elTYf+f9tMU4a6mFkIl}N4TmEGGl6=nhwZ&e?wzTo0;IUeCF
zAfcLH%$sLd;N^yNTW$Bw2IUVO%=*a)2LtQkSh-9#<)71t%n9jO@flV$nuQ6MGgTqB
zqlnh=>PI4PsKRy5FzJ$rFcss-ct#c3@pEBk95giSq-SEnl7Pg}@T;xCEPYfcg#je4
zMXt8q&IlIM#-(jHUJa$kO$RzadcID`j237RiGeVCs%O@0-rp50#Y!F)wBdJ|{ZM=Q
z{`3=YVZzK|<U=;UssuLKWG;^7HU}zD6e0;x%n{)+e1*S|IqC2*qla3&dHxsJgyjTS
z0MPb4Ef&1z=mN`z?6g%Oe&W!?^|4<>B$GAV4UF}*#R)+)nP5MZL91_jG#hFS3&#LD
z1df;A$$k;LLh6zu_5!?hc83>uXBqf?qxbpqOUpYQEGj)xObYoKI>IW+S-MFtWNX>G
zwvF_}7n;<Zg>5*4qjMg?9J<;Jhz`|*LVA!cM*_2*<}WakH6@n=-Ca~`I^Whskm&Mj
zH*8p6YJH(0nA3ZKhMSxriLmTVlc10)Ea=nMx;U~ShVOdT#&q^n@)BC2ugi`tT&Ryd
zn8QC%0&y@#t5VfQmHk1ZJY++NTZ}YXg)xZyP3N9$F@<^Y)IkKLJ3;8{?||$nc<1Q0
zWk*&t7~nP4q6%2E9|kp_kF-SV*$YVZFo%?ViuWzQlOww?(zzdp;%RE1bCF?TFXp=d
zlXEC}(1l+hzKvn#Mwqh<L22ZUwRrT$5gHBb_PVB+T*^{xOl0^dM4o!^bSl(;V?W>d
zfF^*E_Zl=PdZi4>$Pr-L8D~T|qS=UyXyrBq#&;ZoV(WF_vegpF_;gv&p%WwcX+L~_
zdOE!j!AT9443__O7u)Y7dt>mt=i#&8!Yf@E6bVttHNkhNh#jZ_>x`3@q&b3muI8a3
z&Mq_KW0kM+BJOY<n_Hb|f&}b;BDohgejj9_RN^O=Mw(rkW@@j{upzw90fUr~$Y8=a
zb7zK|IGU-#MqeyGA=llkfW=DJY%km}-&=|7G2uMXwHlY@i$&b-`cA6r0)}vPho|{I
z3t_T$3)w42*}6|+xR9RD+&<9;gQ}#NtqCwy`yF~mIC>yq)_a1gHXvdD$qmZ!#F!-z
zc9e6p7`7&;YvvwQ$yHp^W1PYdsU=)ZFr*=+eEb=7`m1+Ou#7U;&KJ{cERxkYTKYsI
z0W%XWp1<L?kpjUWsCVDf*!CopfHYxO?0?g{%IZ4>z-H2Y$6$zn(Q6r;FC0PI`5om0
zy`&Z(>KNHZs2CYQ-0^56C#qj)!*}91A*dI|kz*)!;scw4bUC4@_slc;TSPP(X*y3}
zfUD>7;Gx^HH*{n9jY<dXkVOSqq~`?-a^`n%P8i0(^J~#_o;(D{>mG5je*cDpY7Uml
z4C3zV1e|kfQ~_F{(ExB<tTq}<2YMpb+0xaHN&=QpFU{p2>N<DFvL_Zf7agU3PRR)f
zS$-nX3f}nUG0J;@8LO&YxvZ_`sY|r;E3ulYj>~A6$2LYRsLsN&b&SkM&=iVM0ql#8
z7Mrs{wii3F6|{>TA3>4J?;WBGW2s-unv;-uPt1d)qVOjyPL!QZ0A$VP->0^;jseYk
zUccmk4?UY;;BvYR%`?rHbbJ7<^~mK8HIjm%DiKYHf#5<s0|*&vCpPk0jMshM`d1m!
z|K=P5*qWt@rc9r;8T3)mR>HgXrn8Ej1LUta7QU&m811kN*1FK+I;rhyee4D)LT6rz
z78j)b4jDqm`6!Bvp>$EAXo1pgwO<Q!ar4Oj@@p{=YC2b7$kDzRIT*x;P7W8y>}Y5L
zgtyOb|A3FO>7C6KRGJCLXL|#4=L)a=1=Xz53e&!Hs14IbPY-*p`C@^-{`Xf4kNZ4I
zho;l)iHswDEj3Q-!|owYX+DD?M)|4c+<SR;1dEQk)3nDvQ%qT(VeYhEk}$-tMyc{U
zv%b3oiBRT}vQdJ<YnuQS1;heXO*`zu&725f2y%E~0fj?O(`|;HQO7P}#~nas(~taj
zyi8<XQy-_>E#PASOF@g=aa7W#E^MS>;CM}Dub!2Ck<89UwbwEv@w@D>O*=hgu?*Nq
zR5`*7a+X{I9v@N|mDSF@NpWP<p{b{!A|!fJUzUQTuP=ldm;X&PNZ5U!0d*>d?4wz-
zAZtKZG)4AFwS@1={>&c+IFwtAR1x~~X<OHDqq5E)En-W2x`^6zT7WX9uJE6gHNy=`
zsc>a&!!mx!R=>vdsH(?gJvzLiun`;DS75Ej5dpjhth-46W9G~Orw12}147|;F>jyf
z9y+@_hgg6XsJ~dutpd8oAE2$eoAUHt^1sD0_Sc7NYp!{BLN6D8PJn+ftozF2#*=pX
z)pDo9D&nnYHQ<l<E<E6?)ex1EvG6nrSDFXB3)EUG$d67gb*kQjDY`?yvqnror@}sq
z%I@hyL*0-*JWCLF{GX&oo`L2muq{o{JzDlDyeaU9$$T`e0ejH1(Nb3tfXS<NNAcL@
ztOL#S{yqr6Dk0A@h8ogQYAS~a4sC4c;IVXr2Tb@Mpbq0^7=ZTt#A4QE^x!osfU;4A
zxQJLdFL6i&6f8i3drwSnWU1u4T<Th%gHcj$DJpY#SC+bRlbzz)9GF4#QDne>jGVZ+
z6%8=PF@m@L@l~9wOi)>;YK77DGDB9OZ*ib%Yv|EfHpPby7goobzOW%|3Q$b>Ft}V4
z^fz|7d<@RxZL(~$wz$#mUC=Eua~3WE?7L8T9n!g?`|_9Mh2d@KfgQ0=lWi;_$ePI6
z(=%r7C=AQHjB@BkR8aPCW2N3T>vDUdvHIXkuz5?%`~V|Q-#&jWOh;-I-$s`}A@KC#
z!fr1v&HoEV82H{^mz!M;Hk;3EXTG7T>Z7lDq4k}u=(zNrVOOpB>NbZ+hmi{}F_3x?
zN4X&k68b~lFEG{xC~=*>cw6u{0WL*ZvFiw{8*8@P-@q8kA4IBukRNuSEiNufK_0;f
zRX?g{WZ&@Sv==Xdh`9b30_6YD-RrFEFnYA4e|s{Ijy0Ad_#S3$204|<4={9go0=U1
zBmhp4l8_ElU*{q?XFn9pCTRah1y6&!Ax5TZ_%8^10|rokh1EpRT|X4-+OdcBE`58>
zS8!U>!?S(Ylg{sAg=nwK0o-LLapq|AdB#8COPOr}wz%J+ohVVBn1oAU73t0|XCV_%
z`oO!1hteHIUACaJzoTSBkt@%9M2aIQ_@P`TuY!S{PQT#Y1pYqC8chPcLy_?$t9LH<
z|L%u=6>LcW_FAoQHM+B3@WoKu{pr`tmQI>S-O)x-l6HVIWk4N^rNPwS*g3X=apAoS
z3i$~n?A!l-jmCt%`MCj6bkVqTjSjFIy9x8fJZ7rs*u)+x<2viJA3rgsXb?KFHY}H7
zE<l_>ftou%+Lqnc{i9(62+Dj{x@tWlF+gIdwKY==>XAcn$wG^hLhFagFo!s5Dbj%T
zAh$e5i<auI3hpOit>I$Sq_}UXbXW`@LG*xs$Y64v7vQGzqp45R5^&OQLN^tNzqkzl
zqHdkm@rOx-G?-!dh|O1Z;4KUEJT8ZO?_i(MQ#L6fuV-BoleSqq%!&nK?ZYg%hMI<u
zXU(_2nir6l?CIFYMlVLmNZReUH{y{U<-Sr^3@{2$0!hu_P{^WpIIhw{NJER*blFe%
zu&HxK0|uc$rw++wwFJflhlV=BB{4jEL*9sK#I0xnn<5}(^$Z~fEP#an2NDnq6as`S
zyD*6^LS@xKS-qIz^I}N#X+yA}aqAknLNe<#8mYtuRPJa)4C}}lX5Ctu+$6XBpHq)p
z(WjaLGo3)^XfK{_wWR_eMCF26!DJNdq6d|pAL)0()24P;=I>?g$8i|??MIT)N9<u%
z69Q`7pfogygvZ->RKN6q+%d;-g=JvdF3?I*`adH4F7qcmrDuADsom=6_KS?MoZPd0
zym9gb%6RzsGmpG@)`!0b<e9f#KZ$qQm}nTw8z@mOG?P^k{Slpjv+Y4E2dxjOLT{8W
z^J{@ETAThe1U1jkzL0eD@kV}%J@ErrY!ob<B{udNnitnH-@I1gHt<{c>$Ewc`3y}L
z@3(B8b28?zPk-`gYw-NV(QFJnQB2MdH?K+)?v^?^bgYh7X_?BbL^EoMJ%;=*MH}Ke
zAME+;y+8=17mlXW#ytSwI3}{rlbmSHJ5*D>VRvzLrcengio{XHT~lC%?k`!%V1?X^
zSaYc6$nl_zN{gX5!NAf;w>7U*-U~Kbq}ih_gTLTKM5KjLHH$UvXAzGz<-qr%7*WN}
zRl7b9riPmLOXM`|-tPUc@kpMYvIh;FyCX3)5%#6`RDA=@pql7wO%oJabdF${>rZch
z=)}F|pJ^=R44-?xeI1{CNtII%Asq?Jenra*c*5$v8tIJ;yxD%&H7#dj@A5RVfJnuO
z;w1&1KVjIbzdX1C!V}u3Z;?r0{Oy_bxGmON*PP3XA7p%gti3$A=kj@w*gY;YsUT8M
zd8c*aP)%!f_djHIHY>z>Pk3`>bAd5|=d<DY?%w`YS3Dy-vtK+A=QtL|StwEPSy-%U
zbfr~H&7*E+#To=imH3$8C#%3WYp<WRU}pgay-OhNm0@ED@*D_g9oTv!4uX<7XNXw>
z;1jY_$5m$T&%)Ev>|niZp;F+LC<F}(H~s0|vHqV*f+IGIzTdI{N^W9A-H=H^#Jn0S
znc?eKWbRr-r~iZ!tiI2@L2hpc1Eu$}U-=L%FGwZAy3+*wW?=6(8PHc{;`T=-i6_qt
z8n72R-=Hv9F?2zOkdaes^yb+fRPXC;?KX+vt0F3@m1smXJ@(o8nR6hwa2Th=tVzZw
z-{?>UNgn!zFT3xDkN&Xk`Ys!(r%W<u`fa&fZh&d86I7s3Kog3jAC3T1g7f<1_UF{Q
zb9mIWbqdMDPBaLI5r_TS;uxKrHWVX$`<CJC>I!on+Ta~WzsP3}(=5?2MIxJVl{N*Q
zObw@R4u0+#u_l7NApmfb;W2^XpX(^gAbaPK36(FV+#F;?s|MS}KSMAeMK!@tgtqmI
zS?^hq(ok5~VX$%GR(7;M!(`YwvP~$Kn@AF<kd=yxE)}+pLOM}~-)Wd;y3ZuXrr`Lp
z+WH?VH?=+|g?G}}sW2Q(2(JCT*C^~g+=9`Ei+2Gv!Ac?${Tx}<5jqyGZkhqVkxAob
zKjF99;gq)nEBjz&I;cgkj%Vu(t=KWjOYVmDu(t68^CFb{x^cXnlZhFkLRdP71W_C@
z5bnm>Ljlfh!Xxii9bq1DzaBlWJifr`KTN)HBWT+nyR-V-B<z9lGj~uXbAm54mxkzT
z6=CpaE8dtVJa1gmjM0$6{4Fc?xYib!+~@VTP}x%@&YUN}$IVc5j4lKoIA{GuvX!n3
zY_?I8H#e}GeknTM6cCdSayory=#;U<8am%-);9rz*LOx4^v7L5<ySOO>0FSQr|^dR
zA(#Kovb*rlBh6Sc4aU;{N7GrjMb&;^e`Xk9=<WsqkxqwB0i_g?lJ4%58oEnDVhEK`
z>6Y$pkdh7osR5<)Jzt;q_Ycf9*O|HRv-jGcwf2u_k&6nj9N}UBR>336eqb;I2ZoFC
zJ#fCq9`5bDMS+c|9VTDC4+?Lq{N4Z2PIc~n`}~N)_nmTi8d;ua>vjQwP(B<xCeBQ~
zieB$P{#X^e53HtNptLGwB#R;Kk>zscD=(k$OZ*fvjH2s&3ku@<Xf*ig+8&uKut~Fe
z$rgypZYQYHm6pk*MZHPA=#BlLj66J*;B{z>DdryYyU+Tm7>Lnuk)1`XpGbR2b|;u^
zg7tIjphhR}Mvcez3O)gxOctI7J()R7nae4OvZZ6Qpd4d75GRi==hyAhPuPMi#kloQ
z1{u9|TiRszz7rRPuO$ajswupBY53n3(e%qg3zch0$tR9TAWdzUQ7&*=tx&?3SedD7
zt<z9aw4}wh*Ns^PZo=b6EgkDZ86HKUNtoSC(Tp*SIH_e;BK3YpxF7Yt7~Vv5lsavi
zX1Kr;;Ix1`j0b>?E1n>vWMqzeJCwNs*9>ET#D8b~)&Qkjb7w8$_TSSSK0gOD4x?>b
zuAB8v85W6i({+1*eMZ8j!_9f`(|b`3X3WYQrtQm2IQwQeCK=52+4QUTvkfPU#y5_)
zQ#0dWXb;D?#WC@?&8V9jdGt~hW=E{7ZzYL4-zzq#aiuaSdB$8v#cOqtwe`QdL@FHj
z3EcR(N7_%RBe;JGCFpHd`PZ8i+iS|X*FKYrJ!&4fJ_CQmnQ49d&E<cjV$>^hFeX}%
zg_LDE0q3V*Ti7)^>izo9*@~e`_=daN^Rh|<qMuQI9)=w@bp@t{x0-P->LkyF=wCh&
z90mFBc4XS;tZ}fV6%m2EWeq?rn8?uY?pdwB?QoU}G2@rr!+PQ}pxumcdJ_EpC{q)^
z71!D2*+D$*4?~Z}`i=Q2agPfs>fA>Eum3AU>*y-1QyhNaaX7In*-{<ekQ_*Gd<YD4
z{B|=!m8(f{cD~AT`1oOf30l$Q<R$Tk`t8~6FYfz;edBIv>W83o(AW5joJhV+A2Bmz
zAM33j^zoaB&y7|x=FR%)zZ~by=kaz>4KgTUnbXNxslC_$Nkadn_?zoHJGQ5z&Qd01
z6f*VTP*D29i2Ic=a|l5>w(vM8NqBh9pk-5tIe~>%^q^kc!_ZpHm&kSMVEzdS3ocPw
ztm_z4-=M&dEj|NZJu+(If@W!Xm8DPDuxm>z@%^{kxO~*pjH0kjTaYioA24p$-APym
z86_LjBW6BdSje5W>~7bkq}`{#KC~=-u3dcUaA1l{z@o_<1#i_rev%6LGE6x{Z%0Ph
zF{j!8jwb=4em?ulW#PX8HMJk603V;JKJt$A!Y+l+lwqtYgo6CN)yd2U0?SupokLTD
zIv``50v_bXo4HWddFH8(dsDcrOhBAAN_ai-emC*>aA!66Vx^0GCvI%>Bx?G%g=13Y
zv!;oV+1`EX$=x;-#kDJ{Z{-KaP0=i$WQF9`K4PtKeCsUqr-1+HYPAol%8DzOHocz+
zF>;1OSY)dZN22+H?e0C%Yt31LC)3r-j)2o`JqOKtIUxnyK9YvT<zR9Q7PZ8YSB6Rb
z@cOzG_d;ZKWtqtuL$?khOxsnS4%-wmYglBdVN~=ap=eC!XR{!JiB|qK!8UOln22G-
z#mpSO))G3j@ui7{nGI1EVHNXl2&uwp+2T?V{s`2&e2X7w2ojv%EQmw#&EO|3nZ7ho
z&YS;U4q|>c&l&3pD?c~Ewd)JN|D-PZb(SQ@s9LWm<i?y%TCHl4QNo+rzVooy4#Np!
z9Fimzzj!%!G3=|1cHLZFf-2_Npb0vmxZHHU=MP-pvz`<+UZ6(NCUj!~Q(q4IUl^j8
zggZKh0wF5e?1)jT&-`K21wU=29RTR`_br<K?R8ti&AIZJuSlz$_HLb()0K!wjaDrJ
zOYp<^{%uub&4L4=|ChfXC6wWs9lV)L>AI=z-v-rLIn$#$x)?P~+*wjs3=9uX{v6?V
z^L<H71u@VT*?xPj{49^p1DuJzC0XR!l`0^BW8gASdbFxQzkn`7sqnloH%j)~P${bB
z4NPi$qR`E8;k_ywvX4j<Z45p3qy{r{1B1;*)Iu}jC5;4&N}~%7nr`NQl!?x~I0w2;
z`v-<{^bafdaj_Lu;vS~!<XRW60#sZ~drewma2=~g-j{LplL8za`&03yRr`ov{R}~I
z(wE)RJ>#f*s*u+gl_B7?C3udyzSXd3C2Np@*#kKihEQ`rLvtA`?&EB&#W3IOUfGon
zyz9_~D=Z_)h+$&;3NNaAne&Jndy%I(jR9r*W7Vm{Xu%_IN1R18aGo76h;!TPZt3FU
z$SG7VA6D|$Yv}T+e-06I^DK|HaiI4t#-+gd-WQxMVoTF%nS_I5a14;|tIo#<3r~|B
z+lWTF&a7%We|yNizzUcP&2;tii=5`OlVGOd@}hq|1QZ*GE{WZNT~eojUr!;=>9;K8
z<qho35s8g}P6XqxIM*0>t~#8a&4rl}x;5gm;$*H-^iWJ4M7SHxqxQiqfy*B2`@i$~
zw9@-MI&OY6j{Qc_w+CA<yt#Nz$|*;l#`oJP5)!D8U(IX6$Y*)dpC2wpl=s{cQ=yN-
zcE*j@+OMajJAoy}L|t)J&4eIWLO)5LDUyu)r<8Bn!cM>{fEB@?4gcVhxk--dZK$FT
zTvauQz0RKdkZ|E!chl+0gFJSi-z6U{5LY4oDjh37YW}4vpq%XpG_e&uLe}E=lxAsZ
zX>D*6Pb3|#FyXN7H2r(2Dnrz?+`TZaDn=TO<JU8{q$q*JF(R)UNu!mv+N`N2W1?Ul
zH(qaVZ^{UwkNYlg`jdkzcc9T{!DnH7!bsOd&u@+F!+Pj#xPR?`wrSjbC&Vb*k<jtd
zN>gzDYm@7^nkNCt1}Q&2pdx}d`M~NWITqTpm6rG={W3p~ldMW#)bq8*N4Nc};XX{S
z1mfnu)=3I|tV>+IdY<w@wG*vKjmN*@sb|pHGd;?~aly<C2neT$w1^dyX4%IUMvsWV
zWHS2ILI{stes}c;AG-D#6-Z%xvz@m#*M1!P4gIkuSoQ@Ka7-r+e7IU9jx%%2=K^bF
z=Pv3r6^6@OL?n*z4sK>C{Fj|P&Z|#@V*kd5X{22Jbb1+i5TAWH-)D^){!=%SGSSgb
z`|V|5f%?esy>0a#RUy|~-<K}xnN7=1vAlsNvELgTU%gBvmOGwDdF?i<bG{{&(Llg|
z{ovy68Y}95kF4W*JB_h}w0&?qo>XEOy$IuLVew&@aDIo307EQO<ze63TQN0Akkb%*
zB>I;4rBiGwT|_j$LSN8T>jR|y>Y%}OBxJ<mEVc%s=mS4t#G3fy_)eM#O*A$=AiEYa
z;POJ5h;dJ5mvJGqX4<bdK*j|mdg!hv#ohI?dU-Sc=JIfO>7SkDj?>=xj<fnNe)NX}
zaw2Y-9QT(x1krucAHVB92Nw(+n11eR_^65Ic+^cxrreZ7A{592KGb$M2f@a7eqe|=
zBs%3WQN#1}!PZmx{;(wGHo&!Kd<8Hg<L!W$y{B>i0=*)blrdpCs`;k5DQJueDe!F~
zzlyeOxSr3CFn{qBKsh$~?3T56<%|nu$?rE9(nm`(Vn8_5LBCaZwCw6zd>1TEzTUHb
zS2!Q}L6Z|a|MU>Q>Lp_8?$8BPxU|R#4L|sq+RNgxG6%&yyMk;<)ZfqnUxEVI_)S=h
zWztuVvQMw*yVKL=F1evl7+7mWQ6CIGPdr41qA|H*B=@JVYa9b`mMJYepLG~v5}XC9
zb^v^*VpBPoq}*=u>pmBWCukHTq@KPG^*Sbdgn|>V_~>N1G=eUQ2_fsw_dP)r;1GNn
z%wUS(moi~eYcQxB)pIDIh<j18cP+~XGhX2*A&?-K;v*wI8Qeb(7fmv>@+4{y8ETs6
zJsnWbn{C{oJu5bL7u2KP{}e4_H9ySpfe;fWEypPZ8j!M?#zlR=J{@#ysr!dkazGkQ
zHD8GH#<$YUennwjaed7~hz2tp#rgS*pUv&8cl#NKkwg&wM_?4&c{4(;o5vr=QUo8R
zcthRaOW}Z8j28a;0YvT*kXZhDTZ1bi8Hx5`D-pHGdgHrDTs_lZWNkT7DAQ6yF!K;i
z4oyen-Et;(Sh02tbM%hT``{=cj#-&8JT_EoTO^VsSb`gucC+6^FMSfuYtpc2My&)&
z@6C2cca?$@igrLhaZ+-RJT8%9bjM)NW#NEE0R|_4ZI2cNVr{S@@=@=?+T_{6#hXON
zw}W{-LTXu^Z43E0VBm5Y;K*{y3k3?!W={C(x{#s;m=e4NLBb1gZ%@4gu@eODXKx-P
z(=JIO+{M&$VV$D!L6(r}7#R%`zW6xG8rI8qXGBp?9pz=#)?1rGOJW=1qMlMzMY6t&
zvIMceB!st=I9&d_nu9zKfoP>lQ2OnJfL7v08suF0wAWtX%1SC?Y(+ZA)poVlmVyB`
z`Zg(8XUxD?cJgsxSqrEugQV@ePP&)fJ8#3fBCv^P8?@Haf8x=>hfK=EqZBz>t?~N7
zS?&J65r|{*>+;f%fA2T$^CAO+Sl3H`ag%eJym+vY?$~7UzM>1bU=s3oxfe`9pT4x?
zjOH*CC;$3Mhc0v=bL3U`Yii~aH%aCJWYvje+wnc-_Z5_VV?AUqW5FVsw&icuNtWRE
z=oEAWvf5XI3dxmQ*?9d2QN7@Fa<CxSlhL8qJNQQF{knYQs8=inuM^WJ2i8~{v#Evn
z>2Q2pnW6%!0mE6EE)I27s!QwD8!D*VW%knt9tC1bh!WKXu<3R7XL}buNAeynK8t#H
zl<feaSHAYo@CiiQ_Na=s?u6#qBf}Mk|N5K4)&Pu`FC>lBlD;_WoLk*Ly!zZ{KTl8X
zMvkMk;v6`h+c~G(Jz#3J4zd{`LLgZr(BbZ*$&I4!>X(-Q9+^YX9v;+m!|viD99y$5
z#wVDdG+OI12If-c_-5ZUut}NoIPEUmuz7Ay$7B-sRh6nHO44V@%T&<*th<Jx1J%;u
zq(Wie`Y4lX-q+;oLPEOVy-Zcyhn+W)A6`vmt*r}{S6G<LxmjR*itpT)Ps|1m{rwG!
z)4d<U0MU$5=6^F$He&!|W+NZXk}Q{ZAi%>_VOxKZ0C@sZQ<y*U0w^=Ur&FQH$N$XU
zxbKFcC2{&y&4m11aFmM#eNyqOMLMy|iSBSCPOrGnW4-Dnqi~IH_FuMyG}&%Pyo0Du
z*ad+<Y!HbavxER6)gyXDAH;2Hk7X0sQNyOnAq#Wq)Z0)JEONXcsf|(Yqve^z|IkW2
zE3wHyCE)*00aBV@0fWML`bu#;(ln}^e|MF)JSU5dU2L(-sj#j8;H9USJ}PT@h%dmP
zl*;P*p3$hrx7GF|9lnzPp+<W|b2qrw+l0t7Ko&@9UOo))ydy?%wM?9)xU(g`TwQc%
zI9@S_(|5iS3~Xqo>vI)4eq3=B>TP+t*2|JtV(3vZ)P33Ht2J`g4DzbMx%~V1jQvo8
z%gMyH_0M&YI~&SuMks3A3u`nDt{G!O?(mexe;tQm4Tr(M)9eOj#yr7Mn2AoRxFePp
z@O-fL9-+CIo(WDtaOfv_VBE&yT&HMAqr;;3Kkh@=IB#apB<1Ed7yj<R42x&vVU(K%
zZzMthNR|32DgYGGedj7CrM4I2iNDfJT|B&z1Xk4xUK<ZQ9ZyO+X=uA_Anbr~xoERw
z2nMGxD#-%;scJ?ME)#dOO_<nj*Hg_wSb7|et^FU9MU+mL>%S-4Ke!T%qq}bCc}VVu
zXy7@P>P?GKS_-N#`0FT3US%;_^CtM`s^hwklUdVte2mJFsPnqr-nAqhrvK%7<mDLq
zY7ejfx-sKRM8-DY;!G7xD1Pl3De$*+ac=g=(S=>_z@&Y7ZF`&g;sUehITf;e-`d+e
zn(p?Mpctm*&e#fyptN<!_I549`e#7iFWQ>>+$Ls~+ixK_=t1ZbWKS!i5ZQKI3mrd!
zIu&?PxdgbQF-T|*!B)(x!w5Poj2Ho`$8$CH53^HxrW*4zdkDpvac%{&m%+%;(3DaM
zkddjSAA_38r{B%oaMBVe6>wzrF_hAUEsv6f5du;5uATw$8No~>kAI^PQtf(TfY6^j
z5NFPK9w71_w(>S9r0?`@4|(qij=kG1QA~WnVM(Ga>R6i@e0^cIFS2McY{)ZaaM(nH
z%<P3gci~kZGKH;qLeDcJ1O9uqW58sAl;}Gjuf>j6bnDl3?=Y0{m!hUW$M0HRNqh-M
z{~^6yZ-U>$?)KAclDdpHSGs?$?S%c3;;D2Ak1S|qDDa99s4%t4d%?J}0$<}chTe;W
zFUWkBA-tvs*jf&2+)Yqa#|7QBTB0c>wal&8Zu>Yr<|Wm_f>OeXPpWaeoBh~1MkzlT
z#&0|z8c#(mJ5t88_?nQ=AHjR1PYDXKJrTv!2y>cSzhg&jh5|Hn5lT2B6_J`E(6Rl8
z0tVPG0KwyDLM9ne8ZpAmqoDFWAEAl%YE1mpkbpT7fGAUSYerk2F3bL~W4VY#Tcdr%
z9tUV^v}shhbOU}0<h1=nnRQ_~7hq(ae)I)6f1*HdWTez$^z-?n%kacNs@E2`%e;3t
zu-qg*lOe2bp$(5LzBnp=v}17A%s%;?0tK;X;O;B@e{+VQ=!50&i1E@F*=rIT`!}L3
zHhl%Mm>g2#`S*8X7u9CH*eCSYGx<EtntTch=9Nj%243S0KZZced-l6O8O+C>n?7)!
zrF8d<ltyftb6$@px7qGpf$j2ho<3I_g<Z@<2VTuX0?)Ub7<>u7&M36C2e!=fRzQLY
zbl>KmXSSFb#P(lLnU)#Ti#x=lwG2Em!tiy@g}O!D4+&Fd>vcRQ^7dL0@HBN&vXM(7
zpkMLzFpLBMtn6%ifXS^+tV!?|!eStjMKa61RBIdTFs|MruzvoyQdG5ai$MEuu%4im
z0I0hDH<!;U8d6BFjD@RRMz{ZA4Vz%{kklMP;J73t6=RX$6!C+{I=yKeE+BJfh}YXm
zBrFFPXX_XfnAE{d2ARF@Q$Z*MU@7svCWArlS2S(|EMoK()hUZJ_#Y_}Bi2B!g3Wta
z{$sTJDIZ#E!tOnX-!ru__x#tURRhD>){74xRx#<Dx3io3f_d9VWB2RmH##?rC?zn$
zo=o`AxyvZH3t&Lq$wD&|-ExztjsevN23hYNrnoGuxEBaKRC%a&rP&L)0Sq|n;;T12
z#MbbN(iKt`&?K%=%H5amy<|H1eRq!D;!AD*(Uj~6S}z@)dB=wDt*vj4cZ48$Es?ha
z%AFFk(}`F2!{6VsOGHdl)4K4}>~<|k3F13tz4@j`x~7B3XHh%in34QZ#tX2yOqbZZ
z#jaX&s8KwxMLL*PXC52EB1ODv^POdT5qyTrz&cLX^fHV}^w;%MMAoAp7p=2l2_zA+
z%k7Mey#Qy_X2s!cC^JIE(DEQ9C6QCv{?oi+6}W{d1~#E5P#9B<i7HjVgeV5X53-|{
zz5_!yRC8A?1eu0M%I5!T)3^_waCwn?Yp9yx6331->$BCWr1S@6*qPx8&IMxHs^&j}
zppzAmpCmzFK9yEZ4a<lDXHUfcCtdL@zQsgvm1<l$59*89RP8qV#*W8fq$fXs^k|83
zG!qz!RX2fsW-D_iX?d6E=KKW$3M+nbu3bWEFE1$V&x>3v8wc#_XSc;FO!T00Pv?N~
zC<$oRe*fvvPp!$FDXlSS>U{(Bmpi3PtbQZGH{XKoHsw|klXrn-rb)xLILc)Nq}wrv
zP551Lo9mS(`%UHjUNyZ=&Clk%K_C&VJmQDJO#Z^J)q_M;^=--GKOFUva0X1xG3aJo
zW=0iRjdG>@BGx@1qLCw1&3AQ8*(qy#fY)&UFxR~=<zl7#Y&2dQLR3-?#_T{dMWc9^
z^q26})5^bLgc3;CWWgX~T}_aSFgr)I&1^ZW?RsO*h4ScR^*o3sGQ)+M#mp+|38wa;
zbUv(SoJAS5hDD3Th`DD&q&WrI`0wGZECCHNBPnRs`j6tvM$dR2y7I*NM)7JwC2M_K
z^wY)-F=mrY$#3F39}*(;N*#W@FM|-7jlmd~B>keDZ?Ez~&$ltc_P$2L>YM3ur2?=)
zCWB3k+qH0Q&I?T$^RYN~4JR2!)z=D0wj_5LlS$3OnvVoC*~@uizFyDk^bIgW1I~_t
z$AaENP9yhQ#>?yU_UF)f*12PX6?SwIJ4m9<MkIyjTwfHD*lTX?dt+N)(cF7G!?~{=
z?y3VE*<*Rtv16hN(M2a;<ObjS;G5cHaXC<Ra@gvLxW7`F)$JZFiiEj%dv`1!<}?*O
zvx{&86jaTg(NVnWH)TXtEf4oMz?A`zJxc134*Q}&*lrW|zy-+j9dGKvI-$uVdF)%q
z`?A8lfpQiq6Efb*yKdC_<}LBV&&#JSlx#h!gcP8i;DfA5vNf(QLg@E|jzL*wA>Wvv
z<2r=j|LG5UhdGaek&728`CoY&aJE5%!fy5}MWwXRA9AHO(gnT76#8wqvC1l)SeqWS
z$2$xU=z&cS98fInuL@N8gtdD>NATlL!BSkIAQ^W(t8>3q$=R|4kOTmVwYmBAT!rxZ
z^))q=QT?^2#_W%A1~D`@tW?RVqB()IJhZguVoTu)H=U|NM@idJGXm)#J~fspQr7^_
zcwYZwbihRdBa{Ov@Zqu)*=hw_T6yHKoyb7>*=7!F2%{nQVJhe}hp`TxAu7*^?CKYt
zNK;$=uZYKuBy{4RHxmu`a?`-^FS{@q7=M*^ZA5cs(5R#JO%DpVwBCo&k2%X=jA?mK
zRR>AVc#U4@(|mbF@5H68Q7VH8PJjj_Kaj3cIlBtpdnbXC2rwX6WT&DXe*1nl+rsQ2
z`v;pc1YMx8UrK9I|7`O&I>Re%IFv;z2*~^G@ivXZ1aG@0J~YON&mi!6_abWj;rjOj
z)ru(S)2Hz<rs6Pmjoh{A;yegq#uG1N0}D|d01~oF1W`czUxGTgd{B45rtUqlG5+8X
zB^dqUGCt5f^0RTJ8+~e;xL5ujyjo9mEHk###Wo`~-e?(M@U6K3fX(FQJOM!_!W*B1
zjRlz&9QwRlb<>8Chec8C6hw60ru2M@F?eRQZd$MLsU_uKf5&sK=`+cY3qY*=-rTOF
z{#E;iQEA>mWB|D?&)4}oj8JI$a&Hg&e-19{4j1jXMYKEekE@-Yk^`7`A91%VsugSf
z2R_w2Hspv_6Aa~fq%Akb^19>hXz_CFLwm&iFXP@1Z5PtD<!>&^-4R=?9_$bUOqc46
z1m-3~vCq4_kcu+u+W-9l>wh{pw;H@NTpL?z&{|HfXrLy=*BY%G;v(1K9h<%Bx&_C`
z{k=*A>DkRhL>`K`%+w}>3o=d)eiui%e&c^r`Tg;ZWCk#-d}(Oy;2JdMXKA8KCBYze
z8h{D;y|Ltb;PsYP12djZD86HVp9UrvVG7!5(Y*o9!;IP@#^(yIe3wP?eq{)zXf@w#
z5O6nuT|^|TBkv6^+u3MA+IQhxCEAAVOg`wk+I63QfVM{lI7|ULe;UXR0Gc9xwUp}S
zm**c|R7Xa!3Bq(f?xcR&1AnqGx7PzUFs~v34ZtTjgZVmedUu`U1#Ki)BUhqRNKHF2
z3kc?#ILpzdBI=vDr#5T=hvVN_MLN}lyVDJLV}R&dP$Pw<@1MS7S8i>!`CV6LF2Ay8
zb(0_cC@^;7kg$n<Glk&f_lE~of&C><BZf~)de6UKPuVjoOEQ4?Mi=j*q?OD6$8Fwy
zSr7*yD(Y3%jwj7?c=rU!CM{CNVnena{S`k_1q)CKgpO)rJ?aCT?B18BDLZM@NLEL!
z(+d~)yjgi(+8^g$PfT}o?*Qwppcu#U3TGEFyo@54sdkR&4lwC(p;ph5UFs9i#6kp-
zVZVFp^;K!diyN(n-Z`7=-f#JFPJj^u<7x5jjsNwhxW<u#Ct#_|ArSdQvLL6R{pc`9
zXiZ;$*v{HOcM7H{M*N*sC~y^C0Ch|lk-5S))$K0;W1D*PB&%ZFSLay{>IhZG9A}bH
z<_Sefs&^ww_U&QNk2w18#{A$YKHSjiq)&4Nrm(j@tD(xmSG2x-XqmnBuj;e)K`=yy
zQ0#|6`o2=l7)*Z`mrl-ZW&ACPF00t(z(lq7-peg>|4W@9Bg38|p3iG%2Sz;d?_r4P
zpBri|>^r_k0q4}S|7ug|Tp)AAgeK1r$02p?K}k8UNpjn>ZKyOgC@mo4cYd!<kYO=&
zgL306QZkgT<LJ#wPb5PUp{DBP)+JNmlY4oe;v=3pKKPSfqGY!SR7?7mT}2PMU0D&|
z@MhSADc}GHAR<vk%hZjql6_8?ecwrA8`Dq;cFgZH$~{MYaorDy`Og8|c^{YET-%VF
z;m;oYrL}MG1?A9*OluZCtvakByS)I7*FI;5^mA~194U!e9Kclb;b&}wz30egHAk|M
zB%hzVFza7+*FMxf0$j)2J%kvvF)R5QGyu!t(sMZcwoFw%%$1hN9}^{^Tib^Me3J_0
zAw!2=u}FcVqoW^mIaKl^CV!6aFy!#Lw0jkxNP$O!Ew8U=Y#EA%&@=;XCy$<%Haxz1
z%sjwlGTGR^=s4-<_kGQ!>57oyzAkSDp4^dijw5fvkDxbmV9<ch;MLM(fL1M&G+YcZ
zu9o6%urdB42x}TD(0s@*Fb!O0q<m^h+h<TX#3gtW<uWTEQ&02-X4CLz=zR4u=E+t%
zpVQkkixRoU+8t%2<omKjpNnp$p!}g@&o?Va(7Eto0f{|S#kLb0=Sf2=9eXL;8Eegm
zWiM;F)-YO<w$no1_4{*t4n;UB*}TiDSb%vNZVJ!EEfq2O>haAsqQ#SU{;HFqv@DUU
ze(F|geAa<o@n&-4rL~UNZy~q)YEQg-##UxBT#O0_2N-x$NLF{kTN2s(#}_5!lvo4d
ziE2{Cib5vAX&x`m>9+IOJc~yb3XJg;C5h_PC+n5oA|p|fxvNsZJM`Ogx-HnEQwdeK
z$t90TP(ZlDon&ED$ub$N;Ex>tx)3X%j`OP{2!Lnb_~U0<+z8eP)f%o`v0z-`x8tNU
zM~<8wiI49;Ybq>^H?Lmz`W@sMgo-704~x?_toC}}{~TXiz1&6F3+Fzf)WEfnqad}=
zaFwHYcTirQOe~YYBoLy!7Pa2+oHN-%mL9!r5!6CzZTwHEb*yg~14ZOB{{CnkL48qC
z_E@;%aI|z57O*C3K(C<hrg}n|;SunKD}UrNEWctnu~PRqu%}o<8k}E>o84w?sz_Wi
zn@uRRU+t=z&b8LIZ~rC={_Nn~uOO8#Zc~nW+pfK0C+!@<InA1Gw($Ld-r#D-R%PXo
znZ$d2=N7i3)D?c1?zF7p$0gnB(UnQrR*RxJf1R0?FoJIPTy2Gua~l~ggXwS`59wU<
z48-D$wI<VVD)06xo-y>F@DYpX#jdUna@Z^z&9%p+n9ZPFzxU{Im0&8gXiA|Ut37p_
zQe6eElGAKFU|#bqo?ywTAw9YLaIXn3g1;*q&!C|*l`2!nqRX#>Kut88CY!5oTvmo`
z-WBXKS`(GpR{WQUZFHWUk!E<|O!EAx4(tXG|1R>fkK*24RTM(MTaWJ|GlKOI$H(n2
z{q-nDWT@qw;6Z1i@T+GpTq;B!_s6BqcKn~oJ!hc9-zk4f4tK32!RR@=Q-keh<UtOz
zQ-q)kqxMI7dZg0Q%JTn0Uh;YEMpw2$E(wIiVGbjBg57$aL?Uco_sO6j{UdJozL_&&
zw&7du4-|5MscY_{-}RMPg@w`Vluiel@%dY8v|!A4gNX~Qzxzb|v??~@?n(F)&%;S5
z*B*~dUzq20ZkC>9s>Kq?c-{zdaL##4ifrTaybBss$X2mZSO{uf?DDP9+bsCB3S`6&
zXb&>8m(ij<zI5NRyES}4b>Z!uQS`hb$6z{9bGhsZ<tU&n6rm<DxwBop{j95b_5#&M
zV6A<-Y9rJYO8W2l-rkxrKw7eI^fKGGjj9ezB%kIif8J(v!ZIj!_-+w=fqzrNd1{#W
zQQYjn>BRJM=W@{b*3dyN@i0xE7KcW!gAZgOo_Nk58{7#6;cwAtLg_0wTmP#n2%cdc
zV~VYmNUams7dFPAKYl#qVvSC36RobWt=}%;{6=ax^Q6o<utb2jju_l|@0HQl*k_L}
z1|5ZCR4yJ(c;{fOTPv=5!d5l|X?J`$FzOe%qrPY-WNb3<H`jxw!CWT0YS+7e>%43j
z&_i)_gsqM*^n?-Ar7CzZ|L>z{VwaPAq2qQVrD=Ri65cUeF&9Qu@M0j%%^fB)=M7`(
zxB=z(s&C#^Gg?~43x3d*xA`i)C;+vjdHsU2h$xr#dcxM<gyi#=itU5-q`J1_depk1
z3~dp%(4Z$g2jjAGk`PAj!%$1DHZfvM9wB%*O!Z=+(V#IXE`TTKX&I5I3Qm#RGelCF
zS4~Vrvtgpzoas^=WdmAU^Nj{kguVaO+XunZxvXOW;IjxTSB*Go*dc_AlA9niWkzKW
z?eTfy*Z(3ew3}_5=|Ex+S3a5J*pE`Z4@Y$ZCZ}ol`zLgzi&4A+7`+)aIB7Fs2%%@U
z>rWp*?PrCnU$PwE_{>qm3kMCn*)Yw7{ji`Ew1AOQq31;pcuvEKux4+Ntb&Kgv|;s)
zcdmel%x-hUGLfMPFDQ$Fe8-gtH#b)5lPNoINc$Z3JbHw|+WpFGc=oQFi1(Ka`)LxU
z7;Ut#%&TJpCfW0ns21&sV|s-)DH9i~X>`?ekB`OnoLgb}4j)6Bpo3I^ls+J84p+;F
z_$ngkilG8Qc3Ul)H!tQBw4OK25P5ASp+`k`s_q)dKs#_fazfg9Jha_)(LV!58e+hN
zY6?LL5*jprv4pAId<~H)%PGy<7l$}^By-2HKusD}zr-&2BD=t=*K<MG#`m!g98Kj)
zK{fm$TEW6^*{WmANElbz@M+zu+xExnZ2x<nKysNmNLtzt{epd}hXE^N$nUy{D^S^z
z6Us=Q1XHGbwh|I{D0%k^lrQGr(9m$4^OR7qn1(R9*I%y=y<|>b{{Y!tG}p^bv$+}C
z@73JxLQVPuFr422n4;n@bJhE9(Zsg>$#8r_*q(klSa&bBwZrdjqSUOph6`BN>!%P@
z<@AMgRbFmfw?d^k@Mz8%&q|w3Bzx94*&Gb!SENnAq`4t}A>)3V%K|ljCyc9J7^FD_
z9O_`huV$vDkCWM=e0GvW&9fA@kV5VTP$myxSLM7zLzcyRW(CT&ePAdl-Enp%T%V2S
z)y&;;5!D^Fd4eECXG}CjPB_+^FcY>k6_$|;t2RYNO`uyo#Pk`{#OiO~iJplQ99AQ<
z5fesMo=J;GdS5**$IQK3`$tpk!wA^?QSB$w@B;8FW@2Le6DTy|F9ch^P@yISEkhip
zRhlzIGV>o;MR8arh|r@94s=24eMrI7dQB1@n9SmTX%zi8d#?PK$dm~IjHxhghu7#6
zPHRr?6RBVuP0(98#WOC7r~LM+t4<AQM5Zw$QLsBj`?qSYUNqu6fP^e-_C_J@`Cc==
z;T#p=t4bY2k(J&3gim`-`&;cVzm1b)-?m>@{{22?u4^JNx)_j-==WMwvDjqkCwkS|
zj-=Y6Ob|qJI_PyRAJTPnST$#ogvxB=i&R;=qu5hoInrc>mRtuF{iYG5l|zHL>~PJy
zcGrw=!@dox%<hVDEHYt0rFtQK32K$JaTyzU-3O10GKUfhNfFU+kv?vm==7HKp9@t6
zR>a1sQ3%y6{1npQ4|r^gYO22pYh)|uGLy{Q2~fUCL1Y=QKpx?w9j|mAdL1*Zk*fR5
z-f^{^j<zQbIA94uKH?4{;E-6sL3YcV!LOz26vY4k3UF}F0he|)N-LyjVuz)AKc2yb
z;eOKmD-U{~Ma6|roo|r|EEf!n8MS)-ED#I{6}F$cq7EZSEg(jZ08QxewMFSBwJXh?
zDT@jRA3m9HUN{{<Qs8cn+<x7DdyM_qd$Fw7=FUc6<&PxeEiNNWzrOyP#GcgX{;%Nu
zpbt_uQddZ`jyQClqYI=E$H#rKBzHJ%qR;&XdY`G*nx&1c)r^T=O$djg=MD8GCjrT?
z=x#qUL$>0~t!W`9b$zLXOnV8F7`;t`1{y{mPc;&kn?HrsVJmOn3J&-pXp1sjshI&m
zjFf7{(`Ae$4BwEVw(EHu<ExV$>KUcJU8?Ru@)!X9`~9>^>ixSznPdIAIkF#-Q<zwY
zAKw@}o#v5`b;mB~%2%5^?_<L#9=6BLZnm(qOYr9Tdroc-XOipGg1zR@J6Pcq(s?KU
zN~#+b8x$-@(je-0THZJAp*BN&h{Pg2$o@T|dqgIDZQ_i+lLScuCr;u-sDVF%7#Whn
zlO<*SWgf|4(-n>Q+~2Z_q3j^a(e4mi;c(51r+NXpWgwj}XK$Ecq${m`1Q3-?E?rc(
z{;h-$8fAm0i%wY}*PU6iZ2+UcxuXv{8xA5fzU4G7l}SNnvzkiDho|XW*m5*qunA=w
z+%|rHC^6`;N^Dp$xQT^EJ*~1F*}2WVtG%0<MS5n1)0y`O#RF8#|C~=XJL9fmIiyfo
z22wXV4ePo*Iv8{;PRQ1_K}+V-C*xG=g|TGTSf3uslRv#Us++pzN+<$M^k6B*?Js}$
zWP=yI8hY>T5(y!IC>JBdx|PA-8^ysE1y_BNADN2fLSMdLwQ#iN55nP#XHlRuq<C^;
z5K&b->1R!*>QIiQ`tXt=!fKFq>IE~E{#W?|ft*fE%iH$?o|$@pZkzjlB~4EVX<oBv
zazPSS9CTMKo72q^5AP+4hW^OR79%v!AhhW5@9@HuXk2NZEd5W#ajo@MW~d9^_(f+9
z8VHLsu8Yxe<~0m_xK&l~{A&P3!^-Dt{`k_13=^I>KqYSmMXkvvW(Y=1M-`77Z~Q(b
zH4ga0iKCajk>G*R0Uf$dwSUCyvojeEf;<u5!lf;M&#({?U5^n<;vTqJIb5$)2Y;Sb
zVU1ivqy6f4a=l00pUraQ<jhG3eo2BgurU-!`spLj`YV0^W0`zwBIichgFY<1)maL*
zyH;8olXDg0Bocj9=HWez?MA(<uT23%T)xFrO>L_tHE$XmUOW*v;^KgY=|_i#cY5L{
zRrT++o%VH*9eewEmk>}$nZ%I*-V%lH?b0R<6dnNp62_d-wg%j_K(%$~6$)1ar224{
zNJ5bTJr`QV-M`67yJ8-97$5g#w3FEiEu5zFpL0HY_hp36?hTbnyM0j1482#ud%kN#
z*i~m4zYe9w)P6-b#3m>j@*Bdp|2fnf_LPYCKKp?-=y3CaH8=2<L!B4J_nd$rQifh3
zq!KL&*6i4E!{9Jhy!hWeZY3RZ->o8R=pJVbjLGnQo=G<Jan;+rlCM2@L8#5n_Zst@
zK2s!Pe*H*jy*HKxSn48*-;MQKtUK6>)|4BkO%V6vnz}-8K1&x+e|k!S7+K5En-`qc
zkQj^I<oY|1>v$er2g^>)p?Syp@2zuFhGX`byeKr8Ft{2pYXD{h;0^e?5N_J5Z-2z|
zYH-J=q*OpF1a9dfeSZ&-YD>TDc|A`EPX_u};Xe>@9DimsojWlNj1A3$>LT6S*zSRT
zQmw7`DpPH%l&a-*oN_ex?DS%l1Sm(4ursi|-xdzU>u_gKWycSJ<2Ml;-}Xgn)`W16
z4I=}mL!S|jvIcm176wOr?627rBt-Y1Ger`cSp{7GhRC%)YhOe|FVGiHTc1|l&g$Cw
z;zjlrqPi$udC$agBkWmgUkRNExfOV7xXggF?nL2#7e~T^O+%!zb{~Zbyed9keTS0n
z>R-Krdh-&j@#`Hr27icKPU{{`!OLxF_ZtoS?<k!7&%;L%+hcE5Jw|cT^@xyTA&Fn`
zZGA7zGhd?gpN^G2Tx3z_1B@1I`L9j9hB^)ddTJ<?T3W={dzYh-JLh@EDrr58DXm17
zKD45zt?2~W8(|i5onss4ZdT?mDYMJ09LO7i%A<1L{+t)OdlOhLnoX?UH827wkUpqT
zf{yWA)7lv3Q^ly=zm-ohg9}A**j$pi|HQnmbudA@rlCK^qds2CA8sG-Fg~Vs*Hg%n
z4FRaMyvt~S(V52PnHI|2tr<!oJ;$9;T}`bOscZ@WD*=5zjS3#kt~~Gx?sTC({-fQw
z8rZHhM8#ow3M5zBcBL|(>*%DhO5ufu1cIjFoyqb@<|7ooi8fy6G@lve=RVjqV$><o
zsU~<|Zib$e*=1~|4rdC9Z8tC70nPikqupm@rMRT0=3v*2@nUYw6l(}upREqq9+AmP
zMmpO>wfbIF^Ll1=Q)BfZh}jYGa9tCG(2(|zIk^X^v$UwpANm<Ad-(h*2%65U`X|nt
zTcg6j{&bi2!EXW*$ErHzwN)}!9>=^2>!1Qfr+bmOlWKwnm}2D>WnSP%LR(t$j!*vj
zE#a~B5nZMdTz<~G<Q{=1Zl4Q{VC3WY6HT6}3qH+pkK%vx;ll-fqvxcYpfZ>;RQEl8
z?MA0p^E0y<>?PC%WgmDG9?}?fid|k&(K+ICeahh$H!3Gq!&kvNu)=6{EKuZsqiTXj
zLl}abBgL?4s1R1#pKLSD&KhPzJIRVOEuSb@C<vf%ad$VDTYf_D+DumXTy6Pb1W4WD
zV59S%zH-Mv0ClqZ6X=TbJA=cE@9ph?00IrG`=+eD+y?C<WvO2*vx=jE11}rWH%p=(
zbbY2hF4T<wV^=Ym2|DzAyP#p**N6Dccl2^?Au0meqC0WrIAo_UTy@Y~w$@uh|Gpbf
z?zJcF*lwv2>#(Sb|5NGAd71Ka{Lndo@X0FJ;#K2yTi@TG-eF&SM8eT5l>XPSGu=1c
zqWMeUM*)_(v+se<k=fbwb92kt@eltlq{NSR1C0o>8UqZ{oq}`-tNA5;Cl9r}LU@);
z+D<-*DH0cRjQ%yTgtWSkt9MuI>#kUH1-Xt^iGI|Cw5$ga-MHRpxMj-r*$8vzMn%&Q
ztdJ6qFUvu7h!L}7_-E$!PrXIG4sHIh&uiyL_$a@`Lo0g+GsPEQfCPi(jaC4P@bfF#
z>9#iMUzw01FstevRgt=QBmVKcETmcKZ4@<%gMfjH-oe{y%-5&5$)~WuIC*ygRpC<C
zK6N2zkKt%t?*CZXwBDx)m=n~-&}rK5VdKyW`8Di|9b5n)^Xj@w3#LL!XmvQ~GCx{&
z0inC<%$EQVbu+$8&RRPJOdu}!62scsdO+f|s{#~JDLa2u)!G!o7F-F@B8RVIZ?#;^
zjy}2NQp1qOkrx<sdTTy8XoeQc^(xJeX(U>Cbo4nnyH75i5q%N1psubA`KzM$Y24l-
znD0k@k5DH&47cqky%a$Ne>UhfRu(t!E-t<YR&;w^uw0z|7eByaF0HwmghP&rr%b4&
zC^HxCOBAB3h$^8}{6pwQUaHPkyup-7_<4;l&c{`HZNao>m}xUrQYurbi{V6rNu}$#
z`ti#uYj(34MrZ-k;Q+F%?<<eV$`+s$$sdEE)pi7$Xm{>f9}LGxycTonUN>Y5#!~^=
zs;Ko)B?)yO-y32Lr`!(lNThK6NMl?oiAXjObqh?Px$#A_?f?93XxK4yN|79sP7c|B
zGG-J38Hy9cKCX#>LgRCvS}BvacYn&%;hQEinCa+1SAH(Ov}901|M~YNR!RO#U(jRy
zV1C8e;Y~KGXEZCGVhli|3`FVXO8P}i|Bl+w=kBORLHcYVdgNCpLX5u#K^!1Vz-uqN
z3xLfAhD?Iyw2A=dNSP+8FlXM-uGr8Xg7WLh=p<j?$$2g53Txt_yczLy7$kz&s@h{D
z>DBn-$g3&T3}>h;8ULQ&-j;ltbq5{^aiHp{^m}ZLk(~ac87?2S*KO>Aa3Hum9+nPo
zr@xmfUtL_gwhFyV>n?gozJB*Fg8L@9+WQaQg8%a8DLBYO&-L#&p2jPGE1QhbWtE8@
z0vAIEA&wr~fcg4IWFTfk{{^gSkOL%W@Edv@|K>gE$?17mV&2MnDqZeS|7GaybQF?r
zQc^*yCN8eEs;=Qhc2qta<niLiaw!dbCMaA(zck<V>Uxc7lJfHi%&;=<rQFn$^QH-X
z1Fdab23p{zat#XVh`oi}lt_Z)#<0GAhef`URj_=N{k>#-a4%U3vjTN>G}ZD+N&Tf1
zlYtldTzwjw{)x$ZozqZDq!@~{g=U_YXIO0&`17sw&jdP?$#N-9cj%f13A6FdI!5KY
z@%1GoSG5BJ95C#EnKuz9VU2bXS|ScHS!?nun?4ezF;#mg(F#)tkB`gPC1=pD*j@h)
zKvEA242ARKcM}05YQ-YQrHUT7<N{0`@k0OEBYQ<b5TPinL@1Ud#2q9`;QTksg#TMt
zI(pfhk-*%zk{Y7#&>pS(oKQAn=Q|&5%L=cM;PqJY(6L2T*5JFN@_o=0ZqN>`i{QrM
zeR~g(*hBN3yVDcuuWtGHpNxJl@N-cE^7S|oR2fhuDAUG-EdVn10?96mqdP^zqrUV}
z5TWt>-rxld?3+EKixN}T^iFn%pwOScpI!8X{*go=Oo>~*ORp~+f_WpcB4N1m8=_J@
z4E;eNMBZ5g;D{~g78apMfM7&awGY>j7LXBz?(CRn-|18zfNr8W&nQdgYz1pFLYqmu
z8mQyane+*@X<bk&J}?@iDt7GvED4q4?qigFlhw~ZLZ-b&#&ND^!f!E;IT~)Az*gJx
z{RdKc%exBr3EN`TjFf--W(?=8KBsR%mpy==uO<2K*o2T9=-V$!e}oCSpw$>)y+^e_
zK%1Af&3|POWhTGq<JGfmpv1eiHOT+`3GEAGk=BLJpDoEKPy=t?riT~;HZbr_)agB!
zC?`$2UeZ*QDJ>IWSUDQ|WEjp^lF<6F@x4Iq(!t5bRatcp7F?A?m)5_%fBPECU^j_-
znl8Om?po`UYTd;*+1;5syx<o!dJ$EB1ys68bxPmBCy%iMc748F%u7Y-`WLq5RkNCr
zSuupa!IT&!7E1DsBXbag!)4MF(|Di*XJM-!gTfvr&mPEWDJOTv;qw1}y+Hp1I*`nb
z+OOg7TvH2DTnCsm7>^mdk<THIVUJ`eH>mNGK}j78U1_hO(UA=bxP!P9Vh;6m84^we
zR1LhSla0WhL)RWAR#l$0*H`%WLiZBDTBR)`y5{$t?-?bia0c?(Hp7N~vVNjUHAG;_
zV!xKFjiUJBFj!rg*7j<0>6up*DTW1k4x5d^f>pEdSAsPm{I!weqDAu80)05zUqkF4
z+Qp2@bxB%`&1)?hA4D(x(8!NZ3@sXY!5`<sGT*~Kt4%uC$rv62(^J9B(L?*S@p7GS
zH%s(VTowo@qqB*`zQH3MXtuSERlImC(OpqLCaN_N7|l#cRBlCU$<o*S0A^Z^_01qk
z6fF#2Ja$|nR}qlF%_udt)h$&X6BUpWHQ~{oN^JT-J|lQl^w}ZBP!!$y{~(N&xzdwV
zs$ve7?%;ZHhl+BdYsqR0_dDzs9AR>P>sHL0whGsH#(2=9sUGqWBTjqr8H3-U25K>8
zTs46>^PF)$y{8rpac_7n%ym`RsyN;0zm)$b-hM@oX<zCn`5_1N>5bko?ZcnTp#Js;
z9e@$2BOdryC44@|548Y<;N2GlATS8@TtQark%<;+x!Ys4?^IIBo$B2XeiHuH=^ZX7
z$1_MWlsFhicATC5K=MebbGjRg*NJ^;CE+^*bHqrL!sU+ux}dB;e5JN<%BCOa(W9(6
z@h%)!(a(_DL%j%h#fIOcQPncNS%&C5^?H_NLMroAB_G@Mxa1c=E_0p({2arW=x}~T
zhj*jYmL<rS%XXhYUsHF#WlpI{n=NG3lo#OyjV*Eiw08;xjq`0a&X~5<Mt=-}T)&84
zeDD8i=%}%JU}uV6j(m#CK%}*Cq-I!S?5A<lfQ%{t*UOH-;Zsa_F6fI6$Vk`M7&muq
zkSQd9US)lM`ChV!;|WOkSTn3s<ySf|bUzJ<Tx=F!TA>3b<|A~fS$a5(g55Xed(!IK
zJPCmKpFo8V9SfN#;aq4$0NFeIY)cG^7HpS+iOVx2z;2jozM?O8yYiSC?i^2<Th*P{
zysfOsd_;GE7=btw@^}J^Ny_ImB$f>{$2vDlOW!SN_b5RNOG};yJpvWWT*8}N9~Bwv
zB!|Dg?qD4ioTj{__|0<rxA=hqYIe&&uRi!6WOR%V$~SA)Bc4rHwsaoC5Vna;7yl{6
z_{y1%B+Rq-5yk)(gg+x?(2LOQMHC^l{h3u_^JW-Is<pSj0j=+VQIv5QBX1~g&HlHs
zSpit7<>l;WQHC(4_pg>_sf&Nvy=?@`u#%4?qp|3HibS9JOS?K#{f5=HK&;mdV<bnp
zfZEA6h8*S$X)cYmq7VaG@Xg;*k0uMOU?W=l6FduzIujm_ALP@Wxs?7mw4X!q82xKn
zgRQ_9D>@YBUqP>Wl7&0v^qL4EMbB?|p`Mu^;uavEte@7E!0TqI$Q5_RJj=`z1UVVe
zj4WS25(IXi0tWWanN=9~?^y#oblc&!)sfY~@ti*enj03mot%Isx&QEx8yNT!nQO|r
zVU}R3GFGJ7419Gmj=_3Q<${m@F2~HZf)U+507!GA>FN0{jStUZE3k!Jb2`+EpJoFt
zj`<f_MLt<<c+SUT0=YExGaKrrSmYa8;{e$toWS|kIzWd+I?J^1ir~rXA(*k+uA_>C
zT+JVDZO!E~Em>ond+aZ@c79&DnXIXOy|+Sv6!dWqo9&?ci-#l5;_d$m^1@aYx(*1Z
z6?xh18Rb%jn(*0xsLec?Y|mTVmm=GVV0J<!<wsF+l!BZ08OQ7F7nfVVup+lQessT1
zOuW)K?!Ve`O)B1_f25;BN(^z0P>$GwZ4*7Bp6g{k##7D#8xF2`OdyF$j$VSDf5Qrk
zG+zvPXiXIKdrjb)hopVrELRQT{=s)*0^C@!lq?CwByOt0{P|O_>vAd}R@|R2V#ayF
zr?cUHV;F6H+!CP&5Nr=4tx7XhBhyUe85USj=8-^^<G1oIq*d87NA1n5M_pD8-9@V%
zn9kAEdYkJCVC!vMb*jZ>av)FUSV~v&t>t~j!h9nzJRP0Kc%n>UmE=C}@b2J32Y`MQ
z$Eg07{q);Iz$(3yFaW_cdVM4Zc*to9(9BeSN@E~0Yw!r>;F>!UP34gYLKgE3;v#-X
zVCd)(_Q%DLSJk$X*DED1r?xUkHiw4nOg$cl+R-xoZrz@FhtDTHhz9=J!l1>S_^u9a
zUvo6i@Imw4P*p*mH_XLUf==}un}$gRE9Jk$p>N*x4Y<Q84z3<;+K(e1rg__(r2gYj
zE*HejK$H0$?u1_+q_Lr~7XuuG7MWE1eBfL{q|!#zV^AcXfHzRPZ=_hW`(L~D@S|6*
zZhXFFzcjjb#C3~Blb)*)Der>h_RLFf!^W_TdZk`^+-?`f$K9F%UnuDDwau47%&{x%
z3AZuf&d-P^XAxsE!=jNBwx6Oy8&h(u@M^3N>YKB;1mZ%OzAl{IOjytIc^xWQ{!%!b
zlN11P!wdw%S!YXVUHa;Fl#kC`j7KE(65giRFb<^j9g%_L^06&6U${JzPiz$R1jphp
zj$#C?>gf=<DWZ!wR3U+c2$ZBRLv~=v(TZmOT*oRrVegOdMKd9|lvs!f>L5a0+_-qi
zhs)mLf8~}yU|~Mah07QQ3SqRCwMskXm@AAXfS6i7Shd$PQkr9b$-jhueJgfQUb$=-
z>QBJ`n`TPLsK>U$BM2&1F-V!#_4%Cvxol-+SfhHyd}#jxxtBW0i-I5X<<nty2YDNH
zY#$bj*m#Tbb4x(GQy8{^cC=ihSZa0E1fb(;O^?P$!HS2Fj=9`cub=<I-R|ZWnxny`
zRC$7+$K~pEnU)!&!f*Bc(^%lJ7=yT%lX}xH=mAXC(M35-1H~fMr~&^de=6CQeGgk5
zX^#W*o+gcD$Le&3Tc8U++X#>G!)~h7F#mQ`t_{*>`AhcWm1%+g_TYT^#tTxSk`+R(
zw$YjKZz@ak3FZ6zcxaR%Z8%LijcKn{0aXSQ6@);N07uCTS-M?X{nk2k;?#9|lAFyB
zF`ji|OYT)B@XF<5D?7FwHmFp_2@F}3;$K_*uD)nT5K~u}$s)*jJZ5{ioXYv~x0vU?
zSe7saPPgtkxb!xc&fcn8D7O{|3`9Qyv8s!i`yXohw$!G>EIhpj1vGn&F77aNZX|)$
zYk=r~%vN*rTlAta7O3$KiD+bM9e!<SH($a4$IQuB0>ZK2j<o{EqEEb35aeGFjK4-;
zGh6a{nGDGv`ZMTV0LQZ)BqQNzfKF#hjaYk&$&H7NewCHgq4f?el9iimw%!-_i+e$W
zqu?oX_!TFQuU4fls<7+-(R3DUQEp!wpBcKlyIbiN2Lz<ML%KU9b?7ds0fv<B?rxCo
z?odLIMg-qE=ltL6nosb|-YfRH@85b@{mTc10nP2U&rWJeWaFkMgTr1O{Pz@!b4JA+
z)QNAGlKQ(p>yQaS<&dLMwyf<14E;j;gjIjx7E7(_@?N~_n=oU-#TCsy{MxiCPIHg)
z853g6<mo$0@sw+$;~`ZrQM90nVUWWi=4Z*{ornbUT?Zg8=gEF%BIHK7P*e}B`qvj7
z-2<+o`d9r+pv%Rk3D_>B*}PZct2}s1nZgZ~p5Tr*$S>dxwt3bM4?2oH1Vx@*9ZklL
z(Y072U;q$3>s}l9BmL1{Isy_1zeA~$PzRi2-ZYKm@Vz6t(bsRDp$mrZ-J*r?`iG<O
zmj&5>PIs7D5&T9S&Lkq7(5;*9zkQ49{OWi(1Fr;aU)Vst<q?<{YjW9VG^{I%JRiEg
zZU_q}0Hg028g~#fFqUMh!}K|fQ87lT5&3y!df7m@!rOhbXT(^>DD)ASrV3=Pzz&v+
zucs{#kH}$EZ)14Lmz+kZbyp)dr^hRo7CjQCJdfW!T$)yAwiL}@RpFf@4YK-`<vTmk
zEXu-;$_f4{uK3CO*IO9dHAUs&=aa{Elrdyxa~;zERwbQpY>@$sLW_VlibEE%YEub)
z!%HHsT~*RAQ-YvqdqQ$^yLW&9zsQFhiG5tL0I|Q=FZ6*(D4(p)Qrr<0X!)xvn5W7Z
z%WU@~0eQbZ!<nviX(PO8szH!o&=e-xI5GA>t)(b9t#FUz4EC{mDZy)s&XAH$l0cEN
zar_<Sd(zjI&#X;<t=&u$UIEyAZvst<6nKwUFPgs_l;1kM`_B2)jO2<_^kz+_<waFF
zg~35wr`bUwy~~%I4xqe5JQ=_=jy!IS)%DnTazZla(|JJ99GLLzT;kNS(rBgzfPJyT
zO<x;FUJ(fA&0F`p%x@@$I_GL51Oevx%_wCDNtLaOeqbQH??XAOsc{YTLL%<_mfj+q
z4QK1Zl}UlyL1BHxSWvD$Y3Y3VDa;@QI1i$5K)SY#*x_t_(HZB03OmBzb_gFwI01$v
z51CvPX#ezef*vl<ES0BWCEG;|eMO6ib77p_U{qtQ{!#g)_HSV*r?3kS55LR&RuR_I
za`3xgS!PZL#A#>axLqztEnZM{>zvW76*XFKfkg~Cg#q=1miNfw>r8D_zkiK7{{n_c
zHli_hOeZEDB~n#QA-ez)@F4DZq?{=o<1$?bR{ZS#PiY_TX`fGw0p|~JL-HTKjiUtD
zlw;FK<Shi@kDr^{&80ID3$R#sl4E5wmd$N@rIYdi=#lEHX}BQa<CCcny=?6Y16&K~
zsd77A(h4M6fjpK%a;GpSLNhtym|b_)-;N?~RYp92R|GzQOuWa~uXj&z@Yo+b28TMH
zO(|&xtyVu*565RezGObHopR=qxb5m#RA;c=b;CUj#xrMYU2TS`SYOeb@1Nd;n+<Mm
za>bX-pAAo7(ufPLQ^6<hu<NFJX)`~J%dBEFI|T!m)yC(e2PHl6UE*aSA^77u=?L)t
z{vPh>mt)i69_Jjjt8_jF1rM$`AB%&x_O==un()#1SA)f?YpS){!^mh+kb?ri-~E(0
z0lqg`0C#GAh#O^61aRXcvB~jH)Xe?3%130sjH32whc99GX-MJ;s#D{we#xKOSjRN<
zcO!5-i+Cqt+0IIpZ-i6*CbbeY>;0Tw<+O!)y};_?C&h{KfRKOHl@)KSFJgh5F@yce
zTHU4Tp8cJ@waD&#7(3!nKw2mxnG$f-R~pU7d(`Js`Y)6glcA5pE_?dFUtwQit`cYT
z>&hAPA88VdDrls7CvtIwodevFOwG059|@PbQ_%+#62LWK)(u1a@pAfycF-nA9k+`i
zVY;ExqRv*v7ES^)g(uj@?@u3$bHls{lk@3<Lv(1PH2KrB>+Ob|?q{CMm-URgnM!G(
z#%_L}Dm<00>t9Wtj_-_C9_maTqT;@HJ?KOJP^i{huUEEih9ZJfYiorGxU{PN-sW8G
zQoHF4mk`Xl_*tkP%Mv9ng<%J7=XZ1sZEjL&EKO`xJf{x#;?a>|Py}$FcfsdRQ08x!
zlSZpW?%5LXYf#$xEt$pj;PUDYzJuJ~3SBYKh+^6?iowbH3Nx6<ll(~1%?;%^QH0JX
zp`wCHw1S+E5_|khIJL!iubnhK3LVP!Ev-)>Dc%TLBp#-uJ9hr*tdr<Xca64obWC?-
z@&L79q%HBc=3LURsqr2}XDxrWzmVlz<n!i6a~K!KT3*A$1?7(N+R-2G<7+(2U7#h7
z5@KQ)suAqyfB&2P1qA4r?X%D+WJsax=}ezl2e*Z@$phd<N-Fq;8lu~q_aCO&dSQNf
zcp%z1BuUgLD3?X}v8O)%ep1QQ?c=2L1R3W^js~&ZQ)~vE3Nd^;Gg#QhYu_(|kRjZx
zck!h>nC2ol&Ca5aU&hFbs@}4q*Ywz;%3H%w4uV0wdRHN!`z_e%Esty*4yT**73o6=
zlZ-8M8C-591S-5iKr`JbyfJSZ*0B$#000Tog%`J_KrR@bc-}ff71eiJQ`4N}6uQF@
zBupet_JF{+=TEv}1@A7k8&n)SG(6J7;3=?YUH?i3tVCu^G7~n%Ze2Kj;%=?PV@`>p
zWWnzwAt|57n%8wI3fB;NFU1SXY;2W%YpLxUBH}7A-vBL0?D=`AIK0!fUNxFN*pb^A
z>wAPee~Ng8SO_OD5A!oc;@&N*;__LD!ghoVL(h?=foxWTV>bv$U~5+e+zWsw>H+&#
zLwhok@LD1d>^|*uCty>cOGc0TU8T<PPJm0x{W+zAs~h=}-ZYByHt!QvpZ1>XD>OZD
zW#9Mq&rgnZ?*Cn|n&ywAfH_KwkHZhCuV2RHaQSm41TAsF)KWmQH?qj#sN!!K18$Q|
z`XWl;tbiM?jc<yPSzf#2QrsP--07}{jmn=n;xiEDUO&`SSjtsRgNw4!rQnF9f{6ug
z%$18fuU{8JZ-$@n0J!^zNmPKUO^N`$!R=IHqGr}mNK}bF$FzvB1BO0KCzaWkh)OEW
zXAfrA{y5yV+IkrK1J7>rn6Kg8<S$RuGPfSGc`eb6o2$<2@4u^~B|aJ_%=jwtm<eZd
zD<|7cBK<QlYKMEbR?X*KaNDt&N=>36?uM_X*9<pCVA>5o$C6pbe->%X#X2aN#NrLu
z^;&{UCi<jw@RN9JjT1JVBPNUA!;{szA?AQ7SvR74Q9)$AF`L7rDH5bKQ1TA<U0PpE
z<@ILrun9hYmT$*q-UR}@B>l%?07+qXychcZM5Pf+dxiRpJ#{aqpNKzwpVlDFyyPA^
zTKTbkf__4@%MP0f_jJ7<4#xS|Jej<D)MiTe`TC86kou2-iEE=*vn?{B!)?AafqH!b
zvA*StZ9j~jc4Kdd3p4}-lZYK}|5x>$Q+)?CYfvlRCw1lP<h;-T_(5#|H!4zU3rq30
z%0Awdzx-c+#_IX}L_~Qo1u$n|#yGr4`NQW2j@@8C=CHA?N9Bl{f%LCgXI7~CsA8C+
zsY8;h0`yV)jOvDmf$&t&+SLI+O<J2m?!)o3usT)^D3+FP`t^qjp;}(0Au44N!!G^#
zeusUu*VvKm9T-U^7=p_Z4u`_%wD7UP%&J%Vi-Wmnc2KZjcdb#n2l<n1wU**+Pl!k4
zGcsT%0)EI7;J;U)<5#X_C|D3ZEzVT|=e6OkF%5hna*~25Q4+2mHrYbZ3%PeYH<S+X
z%=;h5DTo^r0gf2j7w$(%_{29dpK2Ke{h}StBkiGst(J$$qK}yu&F#>&k1O6B7_sRc
z;zpjNIY@}%g9Lx~BW>sn<7)qev3wiJc~tHF{T)s*+F57t!Nd+Uu_Jco6e=$t&i;L}
zdbMxyhe<WlaqPLbgFp;L^V1)se4t6(Y0J{r9FFgiU%oABzfyDi?&q;aN42rnU5R5C
z20@;@FQW5iyY@f)4|4m#;X)d<{kiLp>ksqF8oM?c-v}sVAjQ?+{Lk-slfI8QG`m^*
zxjVG?0v1}CvHeTWib&Yaxr?L)bGYRfKrA`V&siOuL!mI<`c#&$3Kxybm2SaIyI;Ro
z6Qx(!k1;6?L{2(eOep2(k06IGNu$mxZEro6BaFfq;+^;@?F;K7`c_V#>rIfs0^vrU
z9PtY+^KFHXKwjO$(uvry17R<dr|Jf2C|u2?bf!mg3ggpIBL8f&JW)yXra6mCu&Z8i
z3l2IfspoZs01?M31@dCFBAXJJmLEnA+g<7$i7|ZP?v_{vxWcMgy)bxHCv(srE-Re*
zn3Xxhrwr#LK%fPw>RHBvY)l>;)AATw{&6I?Q{a{$sbj!=*}M7lEwLNPj9xc2vRt^G
zKr<W_uEaq&qW0dzu0BoD!(}k!O$f12rg!_GFQ@m%7_ZT}uOz;$pVeK*Pj#16*rrfR
zf&90hP*bIkheKYd93gINERW|cAL)bUo4$e!{V98kfGLo|yTKNMcJa;?j2OM_>Dx>7
z{~`T`;a05HN=o{Z@zleJP~RXIT^Fxh5k~h>!ppss%QNr6i;Wl2PscANdqbWvm{87`
z2BI?HYj0d4F9{2T@nCIo&6p`oFPo$`-X_NM`2fU=B<ZHP6@P9K;XG#71DKMLZlZ8L
z234_e{zZqM-TEZ){SRt@8H_np0=_i6_nkD!&YT?5pjHJ%VqFk*vD<s{d)Y0849VD8
z4k}5$L{7=udduI|V3zBL1z`ybjUe(09mJY!57hht>v;yKup26|;rD>5%R~=kL&t2s
z<tBxKl(nAfVLFicY6Du?AVxIGJD_V2P5Vp((c9fMRqg0OSt0lmkzb*367v48R<Z@U
z*a~w=8ooo|wu1mBaaD^Wr`ev?<{sNHIfHJ!5mITv%$ENm>p`0nz3B2zC5<{%!zQDf
z%uPf0g*zTvjEKB5fc*TEoD92rUCJn8BACq@14YgF?XKm$R1y#GSb=PRc)nYBs;{-I
zD|I~Qf|SBlG&6rvoo86mtLvgI3yiNB&kp9?N0!)Oc2g^F-H+N|AHpZJ0V0yVfJ~Xf
zA!kH9{wO`Z+HA?4|KeQ2>bea;ADjm1qaSrOxJUy)c7Z-@tI6a}X@~}k2-FO>J`%!z
zPUcPEAQIJIxAPafv?#S4nLbP}$1r(E>6}w~RgAPfGmB%HNJ&>hz%oL{Zlq*9Pf`UY
zS{{$5jGi7@N$|(GjPYQYfW(YgMEAvx3i^@;e4W57|HE(DMjvO5)STv14ghxCmVVhd
zEp7679@zsKtwt|hoWF)-O}%MVz{J@JL&-0@!{7E-awKqr$M5QQ%lE}xC4Rx7pknnW
z14$UVkN^7}<U)}Gk(_#NZ_qrFR=S`F3Vls19>@GBYqeL|Uj-?7>*m3z$a!wKz0{1H
z#R62t_sx}{@_by!&w{hW4!(V^d{r-lIe9yXNduK)U&sbM9VId&tZ5rboXqq`@H^U6
zzaUqpPbH;u{PQV^Vc`6T?GFQ9Vrt*}UePwtx+6E(hF8~h>ywf}E<}Y1{Fsly#(;Ny
zCRKMq8s}euI*>()B?lc>=gq^TC!3tF*Uz5VFQv@Hs~em+_=VRgNUqA2r)Tjw$Db*D
zx!<+|+G6u>yHFGJuUi=@e+=}ZgZ|%_g98W}2iM}dBLHAr=s8HzWQf2Rl$dwkKnAf+
zQly@n*`7fc@h2q8J!30xg*Pb6?NQ_G4m)2H(UB5472;CQ*#TwPj1CpQD@JbtR$%5h
z8w|-l!zl=Z&l0{XAL_z8U*5bue!*!9zN1d`6KkEMOgLzu>K8T@S#pSut>5Y=B!44g
z({qtAu|ghlG4tL*ktgNhdiAP8wT-VsUCP9lCPs@x&nS2ZSKRld^_Irl`trQzO7%`v
zA`s1)l231|%)TL_`deCTk4s3W+RWj9E=c8tr7B&204xVF=^mBSe)ce;Ka7R)4y*UA
z3(GFvd^1{YE>6Sf{D}OgbF7cM*bU6-Y@Y|RYd-yWVEwT=57b>>1wMp$#tY2aHg_>B
z?yW<qbMb~x$gKNP@Qb0T8J=%8j>ygYkvzqE@8y`r23HoQ8X5U)QaP>=eQ<hy>`139
zHvS&T;1kE=N7bUCEtHV96%%&Yi@e{GPR%+qoOrcNUvESQ`_99EheLd))@w3AyVhwT
ztGqx<F(OmxT6HAEx67m(dJQYpOlo-JD^fDi>QLTS`JZlLW(y5aY4Y2v&lzIyyhnl*
zDc?&6DH|}0erQU7ubKRHbL^U{%38EiXjO-4LO7yD@i;H>cU`oaw9bu-4KX_-&UaUE
z6imG*E}V4I>*N>9MjvDA?V#X6R$rgj#9}R(W?31D^z;NFUCbV>q@D4{>CsGQN9TUW
z;ZkSGw{}XxN_t7xo37Vr{RA=+tpKTqCTJtfF!-d3{P<0sU1!Q95#upS$WKz;R<3UT
zGqwwK6xMxHk+aTdAid+*V_Iz`-&UV!TDD)hS?p|}a5Be9TeI-UbtNt{P{aL+M>cl9
zuxLFq0>UhXc_1EqucN6;#t_#M6Ujyf;g%;!i3lQtq~`|zBy5~AUvQG;vHbMv*OXb$
z>Xi8eGH>?%C@1-Ry1sE;e#Avh?W!?@qc`LtQPMzk>Pxf~5$q@v>1T&pQ0H|6#fPx5
zBJu0*30VlQZ}JC*$(U58%Lh!mh1Tk-?h__%L)#8`1(Z74m+gW^&}%tdrf|&0rTTvk
zd>Q$Q!+{Kl>Dc_))_=ost-?k9FN)mo9M7Lcp0=nJ@=4vpYt>$!&&pkIp($;Yu74kw
z@6$d!(&<V(m%zpDd%=P09*5fe!(%FqjF(e88j5DPN_>NBAG=#qNg2v8XhwsgXs6(o
zT-I1QO|1+p3-bUrvu0t30bzxH%j!Di7I??x8tZp%8arTX;EH~rz-fjw6S_*X^ufYw
z(~isBT};`4RN$`qJGWm+reDXVip55!(6W6}D4Oe*fRkWaRZ1}fE%}Us$iQ*z43cMe
z$hFpVh^Q!(#3#58W-u8BrX^Nht6S>?AxAm%MpCnRAX|Y3$V#;@M?eRI<5zKbJl2ZU
zY-p%N!8*FWZe~w;t1T0&cX3Cey3*?S^iaNNG2=cOr5LwvLEdz0F+qW&YO?}KboX}L
z9n?{_uUZ=NJPbSxo^Om$0N8|na7vcv`kNgwM&c?e$GVh0QUhS3=A1o4Auq2qIzV7b
z)m|V0w{i|eprOHRNF-f!vGj0v%?5|3R*dh5m9t@wtGZ_HNX?DE;n8_etlIv4)yP1=
zhI>u{+bAzh2qoFO(jo$mj-LO#LmcPX+z?K4&d9dhyz18|i?sw6f>81LQ&%V{%{JD<
zI7<(t!v~@^#@OVglZj;2$6U}JV{OQb!^bpmd#(5jqD7oskRXdqzz4&j-ImM){-zIR
zW4@889wZpG<T4<`z?&VXrK5d-OraFA>3F{$2?ZjchCVNT@B-$$g=){Yy^4<sUlNL5
zSD6UQs*2Uedp?(nn6wT#U(zdWC!JoMxWhn<*oiIk0I{H>n$O|eebKb)=^Q#H`p8$>
z+gQ9$6|`fmhQYhsHWVMFRfk(8XIAthlJlzoL*h8JkTFt|pTtF0Wpzc)bV*XHdj${t
zNktEc`7l&Uq_0s~bY!Z;9{YnqgN@W(tD<6|c{wzhS5S~L6%Dq{Xj|OC5COv<H9Flw
zL8>n}Cn1@+Xc66hmoVcZng8Be*;-y+zf+@Mb?}Q-rn}Qr#_5u|WrbssNAcUyZ#G3&
zfI;(0jx3JFMUi|9(bK;~y@tI8W$%rHT*&TzGu9$Pj(1!#I?c98Y<BAHEs5{zZr}Mk
zq&xkad35}{x(KMibQ-nNNCuYk5(G^9t%E3jm`Xl=LOT^Sf03O_7*<Yx71V-kVq)`*
zUC56PCqA$w8U?S#d_#xSvB15OwP0a7;TokhAuHouc87xZm+3)eK_r69)z;JK<{EcV
zGPvqS#i5(tHAJrmgP4SNRMVCv`Fe|vy5M_QfqVeC)8%9>tV>rq`Sf;K8adEdCN)Np
zZ{m#d*r>r2VVWnzlB{?sYe}Zvbz}m*6p74&NyGHP-7)Rf#OBUMJhq8lHrgB0S-*_z
zfyj>NnGUirNxmOSs_6cHpq3nUf}U29o^??JKufpWIzB#Xzn}RS5-er^Htp=^3{u5)
z#m|p;vP#3Xzo&kpCrlop?}_h-^&&9>&8-I=0VG<hZ&DX$(b@Z`IXklyzgk%wtVPpI
zwUONjoWC|moG_>=sygAFuFl|c0g~HPN6G0?HgLJD#3xZz*M4r|!m{&2erW~5zCA(H
ze)Z@DkN-cjWE};fm3DYF#N|4pKyF##G7b<+;31HBgZc2Y!T-<iZ@B|+Xm$f^KCIX4
z<v#iaWh{tUj2IZ#h0|8%Hh<38FK@;x#Bhwo>nF2hTIWThq?GCbmY_0vp`{f9L4b>$
zs*`~LYDGaaFCS!@osZOXa4Mu`l^Rdn*E+Ud=bj2bk5<>!+W;oDTG>^hu#ss0bhT5w
z5h!QKf*~jEQE2SmhnA3O+Ppz5_Q+A4yZ8p+1nvyl!jR#nyM}ShYQ0B{oCgr4Du?<+
zWN19;$DN4P-~?%jJn;Qw8Z)8zK!HH2BCP28=nIRjw__~nFq+8vLv$XZ12L{`w`VGG
za1M(~a$7;0q;DxEDPUe!db0!|P{oKplObp>PfJ!DnZk>YKq%T;9dY4(`Aodvlvi+s
zO%@8}HE+D~M7EF8T4P!b2~oIsK=)Q;b`f-vzX+-GH4@-c%*}mE_@lElIsenv(z-F^
z?W^il>zCA*rnc2^sR`OH-=4-I>JD+j<~$r+eTh6<M+Y~}X$wug4PEqt|45e5b-1Jd
z=!oMdHxEFs$P5(?@qzjtVSnHlCE(%_Xn|jQP41GazmMb~Ti`%kd}I`}twr+CPA%V{
zzFNGR!Gm(oQL8qbq^y+=mivcB_S-pRmOYSEvh0;8sx)Yjg22p?kciv1M?7Qr!}8aK
zt2hz5tnam;*w{}EO-iaL-k5vsyk9+gF?r-qC!&L=`|Y)VS?cbsa6T636xtel7SGCJ
z8#p_8|FTN%kr&av=RNbVL}r|FR!>_~0<K@w+fvv&=<(6jOB2+)+mpFPj}sTi4uTSa
zWWtj<PAF;87EemLYO!tymii5B0(r%;FPIx^YqMcwO8UTVkhGZd?IZC=0mlh54@7}B
zS>vYsv~CMzbIVva4H58m{zVJi9BtMRfsxHj*%t&n;PiO=j({!r9x}Fp>M?2Qt=vR}
zhqPM6rEOk^`crG^{45GBvviNE4H4=<Mqr&7o@qPFJSc~>10Y*}o#tAP2uX{L|8n^0
z=M`6P4d7cp_R`~4{TD&U+c#OK+OPjzTHw?abEjg|>#AbK0l<w)E2Bs8L4urtz6fu`
z;i=*<bo0ER(OW5Znj+!Gqiz8+*~>Aa_+vnsVdfvWsq-gi0(9=)7}^#BTI%KM>eFCI
zdx!^KNj4VbgO;z|r=ZftQQ2I75PQpe(mt?dBhcqvh+4U6Byo*JvIXOeqmO&XOcvK1
zwzo3(ujG?ohVQ*FqPUsI%17z)-B$*=hqto}gidTUYi+c|al|cTlPVIWz>Y{c-O*8@
zNwG!tVqa`uArzyP4L>ZpI~nyAwf76KnMcD#416^DBUD7O5Kqd0$?!!jz*sPhR0IQx
zM!$784Y5c}eHWsKwLxyF&r+2gnx;xxn}}rBYZ6#H^Q_HMH_!Vq%?>NZ0#-1-(Chn9
z3m2(=_x5j{8Sr|OQloj9Q#F`n9pYeu;!$NYbyUtY54>q$E(~HOI{VQ6&Zbsu#HJZD
z_Zpk(_Ogjr&ZC0=_4uI<WT`cUDpeXM@O#B-53dbvhp5TAVZPZ@=DtpgzlF9(T{x$j
z6E)h@o$-HM(Jx&TWTfwRSZA7MBpK;22A;6xDd8hhG2R?1<_(qJ<zL^R+){(2V2>U<
z@*sXru-ji47rqUriJQ&Idk;>klkUk^N9-IjXBYt5B~poC&2&b{4#LbWbU1koT1qKr
zE~|nkb@J(&|6AfhTBbiyzp3rEqn{ViaG{rd7}7xL;Ry6zyC%L<(=}ON5;wK=9r$+j
zMjX44#~Qg)+AdvWJ-rqHI)2;UYxu_mp%r7jaTL`!9!!OAKb7ws@TPe9QeVeSL5_#L
zLYp*CFu9pNy)fR>f-lCK4m?UG$Ml2Ap&Wq@0u%CPqoCm6&h)hIWq%VO;g;^v%e=3|
znQlqbnhC}0OzZPd8A@*mTG!K?nCZJ0^UT8P1ioFq1=!`*TaitZwTTLK&L>T>Eh%m8
zX!n<-9WpH9FSzl1cVyHjyO7C6Cl*c2TZ<KCvE~~0HNtrK))cUVd1TOA?3I)56{^T4
zdEVoGA%NqlF<cmnank(_B`W;BnW8+LlG}uO{=ei=-u0~EI-L1~!#EQ0<b#qFbSG)&
z*O}|!NXz>Go8O`Z?wT>bi;5OBt@sU0O+UZ=?e+P{-ku~a4VaN$5Kcr`Cyn6H0r=9i
z7kr9N`U~J-M5S-66J#1m0J6C;E}^aBUtCjL!Bu6R!ZhZ8s(!zWbTE;|o#gh}AKu>2
zxCni(+(S;4Ng(q;c4lZ_ZxuC0fR-|ThSeFy$aG`s0au}qTA~*Yr`uPbSD}t)?DIB$
z!8~OSLK#QYF;<ClmAisCNqoIE8CBiE5*7SOL4!I^#6_H@Wnav#W~Bi`<F_E2@b5|b
z7W+Vt5Q_6wL?)T$kqqww4tZ{BOxoef_F&{_qn4oUpSK^k;m7X)AymmFv8g|DrIS!6
zZb*&Ky>0ld=<!kRm8&K|%(goI0dSl8Dp3^W_KwXo*B)Or>TAW%U!WK1BY(S}SdGc?
zMrNQQmCUpv=ZxR(34NcYi!hIurU7Z0WUB#4di#jr#UJ|k5&z9%1n!)_fX5Z+5Nn4J
zH6UscQIyUs!}lMkFE?%O9|r!+Lv6W@!Sf0P!$yviP{w*j{QgxbPirqX2fgY!K~tc!
z0jFAMo<k>Lh2^9YIZr5&mX~paof3qb5d(lDJnv(HtctcrbQZ=wSb1JfRztjELSQ9V
zq(&(s2QZx&_94eJ*6KULzt=vUY;Q-McHYk5;#vzg8nO5pHTTxGrp^;h@{~=;xdWMa
zp__B3kn?0BHYILTQGfqZAvhYy?!2-j@~zo2-$2>02BH$wq!SEXWx=7<x~mvX|1N?U
zYzd*4l%9Uu5ip%mFyv^<JmCA84R9JD)M45c(e5JPAXF3!P5!b>mv`EHxlUzN!?YQI
zN_&->I$8Wx_jm1}OWIl+#Ezbv<9Oosyi3d_Q$lZLqboG%M3`XmESJjiXOKe1vQtrf
zg&t{eo!?LK)8-8*I7z+y_Om*vOxd-x9(mwU_%l}mVVFQ=tEa^BA51;re(P-_z`qKo
zH})>MKrKs<92z?cDfT5WLsaSD*Vs9bMxtkAegM)491L#t(&ng>!}2%#CTHA{Jp@c8
z`VsIf5u=@HC#)q{XHB=(Z0VVu_H8aK8$;LwTe=<94-|0OD^2SLv6RE_6{}#(&ve&R
zLO(*(9=7;^s<=U{?Zg*UuQL_kGhrm~BeaOuHuL;MKxZau>{RP_N#fVcha|);w9&DC
z0SwbsIDM4A)s{NJms5crjUsPc9|)g~h#a;*oDQ?gW`~itxYMnTmWQp-Uvvh&bs-WV
zL{*tbh0yX&qw(G;DK5HYkl}Je8-B17v(X4bOqcs6Pa~4=&bR_qSGANozjshec(S5V
zuMTD|Zhjk=zAHaocKHZT#(Vzr!Z)`{{YBB?{Lh@k@AN$Jo6r@;Hk=6r6tXJ5Egel*
z07_v4c-kMkLDPb1o;(W}mn=j96v7VULz5F(b)gf;xZgq&QqY<#|JdK+q}ZgA{b_QG
z<!`tZL{PH8pZsIQ|1X5osR`o_F5o1+G?ODT@rGbTUJ5G*tGXn#zG3R@jWBrS44i@Z
z3ivz9i1`P0U#U0fkzRZLd?sL1GJEVLraayemlS8KZ8)m64QKNW#xdv$5dn@7v*-v3
z>dV)`&Q|0F<~vzDXo%lu2a`l(#Ido;(`rAov;(-}RrAby4^ZK5RqXiIJHx7a@JZUY
zYBr?h)_E>b7i)mki(#e!g~p(d;@+?kb9L+4vD-iZrx~{{OZ_4<GW13?nqxC51cREk
z=I161uLKTtVIpWbArZBi<nmBVW`>eZjOo2a&ddCa<HI1-@{2AmU>F&{uW!2UQ+^1C
z)D&}6ay<x4hjR7Cs7Iz{lkUclW0rWQvOqb*1=r#UQ}4&j-@mI*Syb|0ERAuWq_rkD
z-q1Icw?Ey8q<?&{HoE3%WlE=<uur=NAp(Pqu;g)fasq(Ps8dE;F&^#Z^}G!In@ZZD
z7<lU1eD|Z}-5+^@1<LIW{R5jy%74WV=2c(LR@ulBU+37ohRz=3eV%Y4E50KXaQ%$(
zCr<q)?zxD26T8%fn{MQ#flz{ac9VY8VruWxbeNnic&P!w2^;@_vjpG3cuy7)2a*!b
z<Mxy_gTr$`gBkR9fSlTWlrDm@KK-H}@a1pAd{+?RZma`dfM3_%x26ImzL)*OzsS2y
zSplZjx{_L$Ja+9)h-5P#l|SUmwXd=Prf0@>-}lsI<q0$K`4hPzO9Prrab5exRa-aN
z`(tox5B$oB&_OIUf6LL?KzOuFr;6sX1&F;MvC-Vkf=>f~P;3yX_y?gGt&5{!s0Yr$
zxSr$LY*k5c^SPJ9D;ynJ8RUkFJ!hWbmeS98A5Ib?^XskLBJqpI0NS<Yc+Un)!5U5J
zAcT->M2MObD0cCBIU2o;=)hF12vmC`$pCe}nUiztK7F?X7O4w?TE2Yx`Ch+&!O$@-
z6;+OmEjLlc2I12+8H}%&4SDN6ll~-AMD2Wb6u2St^{%E3hjiD<(f(VY2;|nxx1|7`
z38CVg%Y^CIG0I3=n-lk<tY4a-*tzTM6XgJg3*Zi6fi$Xe(uNT1&b;k_&0(uK$pAU>
zDaN=Dag)Em4(e#Z7CA_KhrK%9f&yuL4ilDSdk6OEs(LtV(Ir%9Rp|*XC;MJG+Mx@J
zWG6%@jdFmz!Jd6acz$QEps@j4`bvZ%FuU@;ajnC<t4nn(7-!-78Ay!PR?RV0LKa29
zu3f786A=3Wk8Q_S%I7HJQLo|=k6oAbqu`*DK{k`UwWJLO!F(d)a^P>#wpD=9bwt4j
zV+7LPY(CcYXsiJNiK3w68<qL3ziW}SH>`W861y-1HgT1WMn4E_{q|_Od7L|qeN+-4
z9zn+$6sO?Si_-+MZwgoh(g}D7<}ZH6UT8uIh=ZQdhJpV0p6?8fjRX3hZxsXp-`g<y
zFmsp>9)kl!LrMr(RROACg&Z7$ouD8G#|DItSq>O0!1s^f>m((SHr3MiW@28;*dQ>@
zR&Q?sl47YXrUhnwL*wvSQb?_JG>Tx8k5ISou8jz2$?O8~v*J6ZOF)Ua^GJMXmK5c0
zOVc8RLnj@_m2h;Pjk9|WtiVC|(T1Pzi!mV-2uOxpsYixPj(|Ax^3i+X*dc5N?QhLh
z&-*^C899klNL6Zabk9>XcOd9{Gz6ew-|LhW@Q-+FQMm!E3+s?Tc{eLZUb_EUkU0Vg
zw?4NCyc8{kq?wZ#>DuuDScR%pOuOI(6wacznN}=05+b%!#K#NWQLmVC?<4erLez;;
z3+oE=#jz-``zZU|XPrYON+F_suHM@XYA>B#so_y~Dz|@J`XmIDvFT-VzKVt&4rWc0
zgt(Z?=cHRT9n3Z55%UydhM26+04=VTkW1-?o*WLRHli>aFBB1iMIpWF$Q=dq@=9eg
zE3#2y))?S2nOdK5a&?e!Pzf50x$rUc$4w}IZm(~1G(BNc7!@SWu~0Vf6xqa)i))u@
zsa?um@}a;0SDhC^>TeMETDr=r(X7#|<#XWftiqi-%99h#U>_hzxkaK;nqh_Vo@^uc
zHQ>j?i_fO3BLy}&g^`NuuyFaBBSWyo@epOa_2IjJkNYQ+ZdGcq_Vw5O&~yjM%v?9^
z{+_^Y$!mfS%efsm`VW%osy;9V9w#U(MJPoXz`i&?C(9Y}bF_h%s{{E@FV9s5ydf|x
z9+ZHcoe(rzYMLa;IV_#@rlc}@A{J&XcNf}YFA!W6EDb~~b&$CoogE}gsU)dDWycdv
zVv4(i#0ze%>9_adLV`tlomD6CtUJ8V<??X_Xv(RvPCD(O7oAvC;|`K2VOy>RGM*91
zZ~0o<8`xN2Dl8M%u?8H@El53qkLD7A4nc&5hTqIa;e(Vmj=MLtxIjypgq_8$I@3R+
zV%tleV4~RFoJ3QU?DVVKoGVLfyWVI1s+-OFYCQ*mWKN#QhQ+f<KA(Df>PJs+4;K)U
z&(%gyoIKT$HvdG?o5%XP*biNd<7Z5=cPXzoOXl?7ZT-Bb=9V-w`}v;gt!b!<v6uBL
z-#q1sFwtUXZ4o~PTGbJYr*Dvtsu`ANJ_@TN7O_8OLVo71-8FoEO|e4LY%G55d~0qo
z*ouJajavHgH1MAg?=CkAzXg4bc0{)0d1L4spSyKBx6rG@F8>BRUeu0z2@gh$RX4b<
z6WiHRVDa;()~nOzm`nM|S1rx;;ycsI%rmA%GaX`<5wZ4J`q8k1?WZ2-g^X&1#NQfU
z-s~cL_<(5DMt+G4><gIFLQ4#0Ct!>U0^39*rU6>Pn+rA*CIyy3)Kd?5!2%jkMcoGk
zY>&Y}%FPtV5Q<o@Y_tuLo%PF(>=`|p#uFOiR<l%A_j+-kYw?!6FGZX_zP^1uFY7(|
zgcfy#XI}9Z#SHSdwW{95vs16ockGO8YgSO;YgAe0TE>N6lM=5XTUsm7E6rP;nP^?A
zi6w5NA)U6H64`iZBC8HxA)!t*EMHAF)D6dpP?R^vQtGrajXJBg|Il>^i*8<!j9NY2
zDp!p|_>8j6uwlpPTlvIoBKz}8mcPZt_Pwj;TU1wKB@6Dv$z{#i;Z=K{_uNsh<B;*0
z?o)4ED4j@9A*sC%61@X3GgAL;zSfSv@Q@FW``hU*H~=co(BLuOet33${1a}aPxl=?
zT@SEZ9XmQ#)p>tx{PDGctE48evuw9!@aD~Jo&eyaXqyupq6xw`L~HDelAvde{Q;%b
zt6d5fpB|le?0r9AQhOskXqra|W|ph}K9f>%mjNneFO{iFga9Y(PKa>jx~i~KMl+zu
z5B)-wW(-1)PEe?`!NvsOHguNDT$oI85SI7`#Vo~yqsUyJ_|Ko~>DEn5&L{&!+ZrKE
zR`&3SkX<`T@o5Xd0x^#ScOW5<#-H7GgEEfi%rxIdJ{~7d8;N4wW=B9||2=|-L6yXH
zeF2NEikgZ}7lSI~uL>-w7>(`NI}ul#)$97GIRv3;^^*@H|05mPv0Aw9QK{J|@QB30
zKf6PO^x%=l;=3CsI?db^VZ3BN1@4-j@n+_yuMyG|*B*afh2-XSElUdIWex7C3kic7
zDu1-nWc3h{{$Aatk=*qbJM;eaU#K@p`sYKNMlt~gQ}~t)hv2R#KtEP4;0k6rXzjcx
z*M9a%`K#<oB=%doudA{nw?C5?9;(Hz6N9AYwH5;_=y7QH%yL*N>Gdj(Qz$%JOQ9~g
zN_tD6L~>fFDr*ZzWaW#a&Ba6^rcGpM^rN51N?|0az!NWGv_lyxMrEoxvrX8_=l3!F
zn3$F|%6KR^GSvtB2F@=Q$npI!j1vjdxc77DWx*|{ixqx)Nb&W=ggCK0+pX1KJXAvV
z1m9)rgPWSxC3S87+BP%~^t}K52Ki0B9lvwvFVU=h7-zakO|c{{US(6BPn$rJgw4r4
z2Meo>S`Gz*985Cln#Y@oqSuKt>sTTd`MrlzjmS9}rnUi5qTHQ$eK#F&1J5ui8_n_B
z*3q0uktan~kc)J57niVCfyz>FTN>r83#Trj<-1t-X&p6UD@WMp`@*+qzcxdg(Ws{_
zP9k8JBb%ImYDb=MoB}V>sZ1~bJ=bp@v5R&9y&m0<r?>6qZA1tB`+t93Z5Nn2jejl3
zD%$%%1<mL>n{0m$Ndm92i!di}BMV0RXgRBCN+tdARlv+-0VAE#&+LJlcsO@sMr!f9
zLdCNu{@VFc<Xmhr2zaYirE%JaErl8+Gf8=yO65Gg(vy5&sz+z4#YZ%7S}h8(I8n&j
z^-9U;LUdMrma;TWr(G{K$J6>1Iy?jt&_ZH!yFP7@({jdA=gpgvsEIrnHZZ_d<%?)O
zBrVWz;tBj2xJ#}U8gy+PL<1Q?-oUzdb__juq~ya_)zK8OIZZ|_)T-1VFRhY52~TX3
z)Xh}cc{B&dylKK<QJYHRg!P}Tg?Jb&r_7F{$4iv2NzxLQHZ^K-!8&urf6HPok4W4K
zpl+*>t?+6qa#>DfH_i7u?j;^LH4kVi{#4HDZhEEm_Y3JdWuJne$;abHypQEn+yy65
z0}I1~>q|*g57z>ez})=_u_f0;gcquR>h#Xn+hIUNm=^&!td>wxpF`o*i~IwzyhX8+
zvR63$LxG$juflZ%`8Ec#<;ZQ*l`Uzno<-|SYW`dP@k%rxjDQ&?j#dwB9mW*#sb@9W
z@$NFq<W(u*bJhTqRXxYc6Xmhy7F%NzDY}_M`oCr3F(@Z<NJ|}bp?a72m8VK$>d127
z;{9+Q?u+*zl(1*l2@#wNMFgXTo%RYs^EEiWAXGevyEBomBdSf&lIH!e{2`dja1#{m
zy!fu!QJ3V?<(8`=>h0P{)av~71S)NQL8-Z^;&?Rs;>63D=_*+lAPB~L^$40G!GHnY
z_2#s$+Z(<M30>$@orI9d%!D>f)J?OJO2R3U2C>f}GJlxwqUp{P`{LRbpYxh7{Ao#D
zP_&(=4)MMtA(WEpnKs@JHwnx+Bj$BWeG=xL^AU##V~{E%1yvBQx<FGSE@%adZIKMh
zMv{K5fBqY16<|~GPpKF|c~u=if;<DIkwsD?(p!pdVt=`C`O6jU7ls0rlB;e+!xrpE
zGhP5u%pj0do&$F<qpLVzTM5e6+$A`}TFhqDRSWCYKXt95BLFI+B?Kv`^_<`)PexH_
zoph|1e$$R;rkkj7Ehof2Na6g&KW-NtSFbC54En{I-zvoqVN}0mFzzr$B8J0U1sT*~
z=O%ZLXW(x7iBqp{xcK{-^1wX9qe65OVWl>wFr5vq{-8Y@Q8c>su+=Bo@v*O;U92^z
zR<*N?_!$>SSS1E6;<_b4S&IDu3gX4CZYD#f#Q{#SR(uWqDHRNfB_?jyEU2UKg2^4_
z%91U%Jb2#nXrKS_gMc*`yX7G24zmg!|GeKr?y%QEpD6o@QbnDw2-87XLrPWdf#dBa
z%*Tsd=2~Cr07z0dA~UIL`tY-{ocg@WOm^LDqB~?$i)y-Jn6O{(%t^lWynaoR8B{+r
zIpo9<%McwRBd*yJ7iX9^S=7vI+xhdq92s$CTmaD}!et`FwP&h~k+Q$f0EyE6ID^fi
z9EmPg1TmUM(qn>0KYqS=kIKWoseItVDy!QLVY8gWM(khG5?0dIS!9?F)ee!-OkDm{
zWxmKF%p_OUYV0CxV(pqq+kf<Tq#kVzY!R~)trnhx(mQG^a#4s~t1hMqlBx<8eK0`E
zL&g%>`y+B<-OBnJr8O$viE4PxSm6UG)^8m^cZ_!JXsw}*!1Jws=kRujbf{T}xU}Bv
z?Z}Y5xY^g78Of_V_t^B2-@%Djp>|h<ucp%6dFLJ2kiKbb^r)Z|*ERGP*h?~q`?8aS
z6Z*q}qA99*{B%WV6RXN!y-gH%`>FCv(Gll@8BsxZ<h=~vFny}F4FWromVKy!KcMUN
z2seTqX3gO&FzO^dN^jge^vU;vD?PR!Z=TkV<|u(?C{<a#2`k(Pwi~GG96;1C?X+Jf
zyuh`uOI~6dGu276C~os`2{lfy-@mossKw(EV1|_cI)Pn^)imIfA#FpjGaBRK^%M}*
zqY(O6oezy|1_UoS25fto+8(38v{V(Ao`Hn8JcREH8&^`Y3!K9h*vN!6fNoOiKH3PX
z%HPo7m;tNcYK|LkDnjeq!XA+J@a=gxzw+=|T^%IvgtjpKul=)a4dxLZB4hRKz6SCF
zf;cX--e2*6Ug+vI8}b5ncsOn!uCv+AhzM;F^&j(diqka&bq0k6kd(dG=W4xISRU>z
zyGbE^a+jO4qYr0hVJV&T3V=j2g;)&`9-XfiFCW7(|3nUwmyrvGJsXyE+|evHF6d8&
zYVMbe4)wUl{duz0WHEw6XaLGs&iB>HAJTK!lxZY9>mF}P3+x5rg;pD4N`~<82y1?b
zD-#sMoZ=_#aP1<?o9LvatDT{a^jG!4Q__kl#5Y|P;NT`h1-o7jh{btHL^!VixO|k`
zW=&6iYvNEvR%hzCh%dpm_*b)wJ!0J8hx2{=&1hrZhu85Q1jkV$IYIw+r4fK_Ab=PU
z1{sf);auKi5B+<6{1-X1qYU4^+p$SU>xb_IHP73==jWWzIWqZ*R)-r8km%R?wuDjp
zQ~TO@+hYAI#0gAIom5o4O-4+}9HMM^bjDbzXtFd}CaM~8Bk2xixV5Q`RH0YBbHrKR
ziv0|THTt9%TfGi3R9d}5Fr<LcSTp~r4a?OU+?j%&^q`yfAvHKy+jB{klK<FoT|Tsn
zaXBruFApZvK-Jt?r&<oi(jmpF4VM@DB2WU$nkK_DGsak|FG#U}30w>+70PyZzbh+v
zhX&FR9LBL+U}xIlXULs`7L%paSFx4IuGwo7vSH0gA8?8|DCM|JYoM*RTUry=ov=*g
zyqoKTmhT0KGVoZ$J`|aKKH&~>%H^L+Y4_C=1EW`N0;phJr&TU%t3M-xoG1alwZUid
zYuh&Kf~dGNbosuVf$L}qYb&6h&BTjbK^WQn!R~+MpmFy*E>kJG;G<9tG({wdbdKn}
z#Xq+gS)hGjA?iQ{MjqZfV}#RTmLF5EEUUa5ma82Y82H)aONSIEcDR)*_CcXFc_nJ{
zZ8B#kDeCEngx{!G&~aH9Efo(anfb8@h!~2<CcshtoqT~x`An0DDqYucyS3dYZ)_^_
zqmS`Z;BUw>cC_D)U^C;kK&%^JQZSQ7y0r1;+Hbz!v3R9S+1Rq)11ng)R$sS1GS5t&
ze0Q}dv@;w2;VH;@oGR-+1?dNDn**utT;zVA$Ivj(8f>caTFEUE@6dr>Z|pgUPL3X9
z9L>Yqdz<}<LF^zD5LvIIUCi1_SzIZSEMo<m_+$f+nRH*f!qix|lwAdGomOz3J5Op7
zKMQjE^7D^j#D!F@4psG!3H=KCJ%2jVYc9wJb*x`9HIvf)nPCibe6kYceZs=`h?Z`N
zKo%z&?~kl8BCcBPj}i&%EK7TDl>Qe@KAHleE_Kk0=TSPfhf1bh9xi(REVGtFT?H9R
z_*5q><$V_ubI`#=tdr|lZ=(4p<TfTFsVI>a`5(BPD&r73q|}Ygl*BEpxh%}Nxpkj2
zZG*b7?hZypY!<vE^_{KBkqxK1VSnO9rMC?fB=LbBcQ)FJkyX)=oJEpm67&Lo=oB`a
ztoc@%7@8ETPZUcPL3~I%OsgQ&a)>zV;06jAmpR$x%PF#S{Ok&FLIBdz&<O}{@8D?*
z>P8^cTlmSZUn_HEO5kN8rlLc3cze4zKHiq0sfCjK30hN@kOWX!HNwn;zg>e%hjhQ-
zb1Tc0U=*(g(eOkgNR)xf%%)P)EWRMCfnwicsp-5f)6oXAnJEyUTB?r@fe-jqg>zMh
z5$RnkvUV*55I8qQmTv?7KaY~V9XbNc;UL_fi!(-M?>1Ua?OA&|LmZTt8XV3!0vq3m
z)x-9U8+!-`mLw~=X1>TXbiep<L;j5%@FoWvkZ#}>0x5}g@E!R4=cZzDY;twN$^KKY
zF$Y$ZVIxbSlb6GDLN{jY3UjMP>qiIiE1`&cS(#<`daUIJJ_!BZ;~UWsk}9-dV}4Ce
zH_$Z6gs-eI$zR=#5_#~bcU*01`uyUms6VHN;7%U5ryj!({uV;|uZVPOLn(ZbZ-g0p
z&rxxAFT4yI&1#0L4FuxTj=9<ES!v&KS<}<X?0v6`>pNKP5Uh8fr2-XM7Hb3GJMK$-
z1smFr4-rbWseS$Z2iA|L6Akd%_Bof0fkkBqex8>k(@JdlX+*Swm%58a-1CS9h@#ot
z*{?^FnSA2&(vG6p=RhO|Nyfshgww|98egY^WtwSW-05qu_uLAZymKjHkj*Nmd;LbL
zQ|OcNYP%vbH${-{-yG7>SjxD%q;Qws2g(aWK96K1)*(ryu@k)>j@gLKM~iFNuk`$~
z`veKw2)^^t8f^-Z%k6)o$J5m$aHqX`RCO9`y+QbJ@%echGzKMA#4>ZMR<Bkbnn+g*
z)8W@lEjJ(?i06Vg>&WzD5`ev9$QMEMG|>)xHg+k^@I7TmB}O5LLdc_?LwI>@3K225
zkC-jfX4HtTrGm$=+9!#00#t^%Km_JF)NZOv8?;1=5CZtPK3uR%sM&f_r)eNV@XJe|
z@_^Jko!LT|XRve<jBr%*Aa-S%`8>C=z7{@7FV5uSXY=|S8d|V8zHjBHz6q$iYz{Lw
z&){SXS?u*AruNWz^UmQR(YN*o>R-2I6&B%^km{t*3)o0JOwyrETDYnZ{6f%ZgEFd3
z;Ws3^9LW^Uv*NU)bva8B;(2~CVWAzl+8z2Uxf2ccXcSl@<9?Fbs>juABI_!>`I|Xi
zA}=mEU(DzkZ06Lg)E=3p&x?a1pHyx6Sg_E4e|#6%BXs0w+v#t+@?qrwbg(rNfBL)n
z<oP?5RgX&+sPnCZC93wbSRM|2beG*NM$;t8+U<L$|4PF9!z_f}<#nv%w|~iE04B~D
zy|F1as*N#0Tx1}rf?h;aG_by|P(=LW-GWjSp`GH)wyq<;ysp7_Q}k?C?)L$Dk8HxG
zXhmFz%xKURqS;=4fjb;XIiE)LnTRqJjkSb8jVljCKlL+}_B+fB^L=AZV>&?y{#Vr7
zNAh0N60o<nfPqj;QX{q2TCl;HvgbV43^<30vDx;JyW!Smol=@BBBgY>;?|qI*DiMY
zK#AVam(S*s{uxD!rw9#TrcV;^a)^<(zGu<U$Ud2$JgV7*$Z$4oYEv9wnpBRcV5*>I
zrPlnr3&d;EGf%b?TYfgbdM}44!hJ8m3^4UMBti~N3^ibMCu;H(H`}tb5FzlV1~&x`
z50+|(?bOH48j|DS&2E+TaH-yt9(bx9M}5hx*?OjGe)ep@;Xb?V%J8Yg#-HBZm53<K
zj-&A>k3c4u+yvlnlWF{&`l;r!{t`Gje2??*RKk~Yn-cJH<vwJ>vp*lRGk+|EQ`^{c
z#q~@nat+oQlA`*PA;BCY;dD}%MbE>iSj>~x2i&+&W4w=EkTgYoPXje%LZV7d4Jj}U
zZ4@~(%$2HRhkLlxGG`Gu*)5IL(=A2tTo7%4jj;kak+2ky(1S|&3<|8TMs=`eO4X>K
zU(L)H`||DG#-pjgZWIv|A$Uh7*T70zq1#Mf9SQ*7T(tPAyOhGJ;(*y#FNQ@0K;&6e
z8Ut&;9215Xo;K&VV6VWw`bxo{(DycDz4Nv12)0EzTf@CAC3yC|Z^`yy*4#E)tmPmW
zOW_Yhr2I;bL|)bt1Ti5>N$CRg=2E8EAdK^f<y|Ig#uYMVKSjsc@igyIdWhi<J`|YF
zMLPKdMy;hpT42Ja(^j{aze#wADz{l3R|uYy_koXvYfYXt!P2k+f;zF=cd4_~U-!r_
zg1Y94`X{E|)YWL=0XsjbL)uYjYvX*OSqQ*?=WU%F@X!5ejPI30|8y?;%j=$(<ut)}
zm9Mx6)QZG@Quq@qY`m1m=#UD`R&arpgQN>KicMy|k^#?bBk(dXsW3GeNzt9GCkByi
zF%>f&<0ALtwqPZ0>Ud8u&x%4+WEHj*0Smtr)X`z=x`|RYgk{}4LR&3ab%btADSZ;3
z(WrEeh9D36!h>(6A4Xc-PZ0=0SoPVHd5+hdvG-oodP8$Qgi%H6ZFO;*d0*-&NWxpk
zlobGjS8=HfRfyq&LzR#g8>^R-$nV0Gpbre5SvC-D6che>jk%=&P5u7w@NFU2p+7u_
z>0?tnr|K_Ciacq)rF8X=*UGFlgbzqV9}FKFrbCNRrE+t3!TFZ}{Kcc`QtX0(zbabf
zhBk&3$KU2jKh`>1q_aXa{*S1)V2A?Tx<+RhTDn6*kOrkYhDM}Y5b5r27((d|>F#cj
z90UfC?r!PshRZp==YD_T$-UQJ@ey;Dv8w(onF)c~fF)LhLuH|OIM}Z#c^U6xz~4l5
z3HSy)t!fQF#fG|k0*Rm;{@^Q<B~zOgs?pU`A9Z-0lI2M-;@G~(!ywDS>`L_At?V@P
zzksZt^J@d7REX>AMcwwbLhQSx#Im+aOwUr>C?j*xS1|n<f0Krc?I1UN;6wt4q|o$D
zzSodYE>eHS+_croj|l@T=#pidt0#ia)=32{GFN;H5h_{%)GG!-6Iiw)G!Do>?x5-p
zQDxx!_YtG`S%l4)sdwF<xPt2ew3F#$pPl%A78vp%MuoB6tNl%nog7(Z1?0T;eXnlq
z(3tYsQ|*yo@&O^aQvoW}x?#WsndGIXO>b-CBMV6HH3t|C#72tkA2EVh#u!v@R#v$`
z&@xIX+f2{^%>%^S+!3d%hI3-XRi!ZMM74&fkpYjeVudfuXm)H{D?r^z5bt2i786~b
zo{e3mPn$`LvHD=MN5et1Bu|mQmubgxC9SU9v-I5RB8|#B!oo<!?N+UJROsSwIBqig
zCyz$WbIH=-@Qyoo7Bt|Vs*u)&nnM1~dOD=Z!3e^B9=$JTez$s}`7Ir8;J76uW<IC2
zL_}?X6gvc)OC{|%Pni!f{&aUfXPCly-#-Y2|GYnvL0&~xe(#Gt7%70%N&n&f915nr
zGF+^daJ>wzyv6_=WtnLlE-Ln)G`Keayo4^!S^u)e9W6HO&z4^Yyvsr=H0)UAvCXml
zk(CXoY2)-0SuHEbZuy3%1j#~KvrN4zF9q#&-!WA-F;T_z;jl@UCL44t7z+fx8bw9N
z9vNIu`8UweOQ0r~H&zvkm0iy3n+b&j&C)+4tzyC}5dVCQ0Y{ncYtePH6b@<pu(VxF
z!3jI?$sl8nR#&3x4;Whp8diE$$5DaxRyn#`_LH{WE27y580_|oV-+0j(4Kqkx*I84
zpv8|}FI%mub{1|4;#mxIs4t)3IL`)4fe;PzR;sXN@~#wuK}k(~)hQ1Zbwd3%MR{B#
z2k!s6b@97vWhg>(63?L9_OGqqdQ1u8v(lGZS!mXud4`XM!vo>mqnGE{-b80z$@J&?
z=KyfP>u!XyZA{B6Ity)tFEqrtYy%5iYbm0XLuQn8zmF2LXEosu%*J~G0Xc66(Dy-;
z^el*PGW~xJd<KGHS#J|XbN~F{(6{kzEyc%XF1St3_ZM*C@UbazyjC@{A!Q;_{@o>q
zc*=pbG?&)Tjyt5l97jJT&cx{`4!f7elMx2aPd4XAc<{d)rOOw*I;iK?Cpjp>DU?UT
z=wRUK!WFP&TyqEcj7&*-G*^)c8Ag(L&A@8YI&Q7o_!fwWI-VYcs0cDVmLPNFZ|hOw
z#Fh$4RWG@p!haU4{(2y{wY-U%<U$kcJ9#;66VUz$EB2|5j<$UYJO<q!C;scL*96!g
zvrW^$KuQ|b`^DWU(Jsr|k0~nLS#7{Iz#1SnIvEM=#7dKWIQ0tYQmm|P&DXz6&i1))
zdDZ>|4U?DtbhLM#*-;uw|6=N^wtg%Hpi!cdv;LH_N)Mc6%4SjvZ`v$aQBj<p@ZeA0
zP%WcC8|6YmmQDLXL~0<nvd$@g;zxf`=-%pJ!!f(43}QK-eJJpWOxZy<S4KbP{8O6~
zIiLdc=%j}&0Q-KB<r-Vp?wsTa`cJ&V->k!CkU2}0?&Qx#b>#>kYx0~yqx!=r4RY+{
zxQA-(xRaGczz=Z>qiD|e+#}!2uI_yupk=^vcYx_4Ryal~V&NHC$dr8kNtJ%MH|^rh
z`U~+E@;oTyy2FjdMj6n0-E}*WdmupEChEuS7uo*y1eM!qxB4|<Bu7_g?p3+|-^#b$
z+X>L`@%!cK=gTnQQmo`m>Qid+`6er{#z_dYC<P7|vr&#0K+TuNj)LX-${THpIZQ|~
zwF+--6Ri$gblqkb@PwA)v}pPgrJ(ys!cUVQF)$wAnD!(96{Qej6i3}h=-L$Yva>`;
zt+qy~NRmIEvigF&5Rcm&rS<7=3OkMB8^<Z#p6o<InO#O#ZmjxEnh-90zAV;HOzpsd
z!?Ul;n9+?O?L?Yr07T1-MB_i(e@6!nE@>%E=xOrzg6<yS|8SJLw+_Xq9NLR+pA1Y5
z4oR{fn0ii!Ky2JW%EY};=9}#J#*wbf>-nYMA8fLd%#$C0#WHWX1$8mJT7~Gc`8^uH
zJJ#$7tl7p3Z;VuUBT$aBcED5|A(f{{&Oe+XnR$CH{!hEoi|#&I)>1eZD-heS^K@##
zzr3fOUe7G(^xrYG{=10e4_OYz)RzrO?%AejI6FGfXP$$n5-V|e=<_vlWuHG)HTFC|
zaRBnfmrH*|{Icm|oCqWGs7gn#*_mlM8C|235ug#-7Ex`WH>WcH%jrxj=_r=UF_K_!
z^y4dT_16k^WMl*G2JK%ZqX8jBqGtyrs(;W4c%Dc}yr_jt`f2IDw(&u*jiiAz7Ud}^
zMGXyP+iL1!iT+<mA$rh(kwdMH4nQ;fXJU?F@E5{j#&bcVmUPh`tup-*{D`2zN$J0n
z#Of8l(Hs!Vrr%WdP8<X&z7n*&)WhA|M?IPlC?hLKx8LLnT`9ol9G>>wGNF6@cFcTu
zM}4hj_3g<R;Eg_nwutfoMG~QUbyVq0LuLVlKOxU79t}}KjxUJJZwy*HKT;bIE_(_7
zd7`6+^i=!(z+JkI`}d=`{-*(Z*%NuJ68bZXLlR%a+l9djtV9-~ky!(k1jN$pGgB=y
zQRF2)L+AY-_Bujg3$Y$040It$zBbkGD3LA9teQCtN)d*!GmABeQJ%`X4zYp&8$q%f
zi@Y>HLi}m}a0q+fz9CR(_RncbV^ht9Pf1Vtyo`th#l@aQP26zeSGzz?I#O_R>yTY<
zP$R3m;v(K&>0{f#K<!_*_?jA51gO}|?-Rrnq7*Y;f0Ipms9J*XTKL>~i)OOVsGPa#
zDB)*wu6JXKh%tPrc9QMa$5R#nhqz};yHOys&OTmWKNuq2rV@W`sPQ*l;8dEuF5%E!
z7Sabi4HJ&it(JvI^ri{QXNX#_`2aRZhPWChGPx8SAh|;3WHZ`*Zg>r9<|>L55?Bo+
z{YR<eW>|T9Zi@<e)Z?9qi_b1~wlRMep1&-K@EI|7{Q8YI)CTfm%Mc%tAj|aof4BVB
zwOuSBW@3s-SU;`y`%)j%WeA#Fi0z8%-=5=QMGW%mSl_2z#Z9~Av^F29R)G+d_u}Wq
zC2Q%iLfGdml}8LJ!&z)>Ipz52ucR!Mm&ybFi0OllT7U3&q|ZGfc_i<&(RYzW%isov
z5-d_761*Z$L3~+f_xAzk*Qoud^y0WWTPRx3jp?;CW|`J+ReGttHO<wCvCrCUP#x8+
zBr;M8J~JS`ohMQ*!T)yi8<$2}7!2yNVj!5JG+dj?^9<!cHkB$$NKZgpTiGcBqBeXD
zyY{_RC!yR<w-<haB+$|0+1Y8f=@LR97n2+#gyHv*W}4ZvIF;np04)>zW79$R4sdD}
zU9?;_BBVAFN1uBr#|Cr-<KHB)uDoRXA(B^m*EF-?L8mGvt32NWTS*NI9MP||^UXR_
zzL<`%AulK6-oy{xz0UkY154%gr!6xL&~kHGJGim&rlT%>d<T8IV=M?>Qv5=yN`35Z
z|IR!fLJl!>;x23m?aqo+P7NRUhf_vxrNKd8dS5WK%6>Bya%-5Wr4Mr<V=K3q5RHAa
z(Co-TyGC1kJ`sw=MbK%?W24V|QE>v#URQ-dEN_i#B!^q>1}97|J7^WUjavPP0o7Qg
zzdsM(eWj<AjBqH?Fc1f6u?8}t_3Vbp2Bi>BJ_bc!Ak5z3&o+&YEl%HW8da>&y>4MG
zgR?O`p2Z(wS5RXqoT>f@%0)sKT0)EA7h3>>EPhB`a8wjfHhCMR=q|`3lSLkrP6oT&
zs%vLeI4=QtorazU!aUE)>p^an)<-`}(8c9*^!By{qQxdpTvR7?GWe5e-?J`%5>PmZ
zyFEGJ&$GyZSB#-0(X*|A-p@UT6CA9J2O^!jkqFPUdA2u4Ge6wyfo(`t3W99AGbec?
z+*=L3KJODr3M5pKVn`dkFyC!9F0I@;gg^Sw`_9eP1hI|yq(%B4nHA0`^qrC+pM=QT
z8S{1_s4kU^+0X@uPWM^m#b{JUP0sr3zGaS*S^MuB^NKLQff1CffRles+t$^NA;mTR
z8xYw>S!=tA{Ro-B_$LG3X^L2-zXqlU@aSAlnPDVvwj+!-;6+Dq&F?1Ade5VV!`9%a
z0$4@E>Yq&&^bL{U5HJ@*>iqXj?1tfgSf0E=PB<dwqf5`q><V=l#Qk%HPR}!YGO(kD
zrl0TcP_clYJLkSFxbB?dkQ(n_DG88ERwT}Cxa0d~$Fo%VhF!jY05ESM{AI(OYFBP;
zF<?=JayE^|hZd1ue7=cDC)Z?T-Vq0bk_iLQ1HlWw&wSmTNi@T?M&?&yA>jqJTKG2w
zhX0`O3Uo^Qg*W@OpCggYc_?Tx7f_UJpKXEnDvQ=esL37DRJS?!0X6VTg2^5x_Ua)a
zc4#m3IDDkHWh0erbv1FmJvGOO9E2th_s`^A=NLcRn+A@I_OqRxP?XDh#Ij+}z*CT-
z&Fs}$G=}zKQ{+Z#vD=fy(!&V6;Ptze38c(c8$al?smxp11s=xjFK#|UIHDB)ooLgo
zF2R3Qk4<TrCjhCAS2PTFG6wLG|K=sZV;3tiP>SmQexmvgv*1D3h0D~%vH&<c-dBZ(
ziI&sF_pJcvj}73?*tO2@7Em%Ax<DHUs4b6ts&_45Q+!3Ww0h3eh$a*4`Sn<baVz=O
zuf@h{sUV2F)Kl$j?n^~IN|@)I>yOS|lUwjV^uMf9s)(Y=O~P*wGw5WjT6~L{WMWtB
zxA;AI7m)4*qL(nO_44%mt*!qYmjL!~Rg$%(%Iae9I2K=hb4qUspu#V@KzW6$wa+};
z4ZmoFb1HDuhb?E}IF88h>A(^5C?*-wt|Iqmn<6$m71^W7k)eJOGHQ(b6Pb5$6(&g?
zNHLtS_*QM2xWT`{)u}8BYFthGQ+v2bVSRJsU6|+6kYGZ3T~t|9!~C+19lGdy*KTN!
zBw=cM^U;;f>iCz|egVF8sKh>^8aFrReGRBEV`}=S?Wm)~t;z?&?_Gnk8=T@wY@<z7
zG&*sQCS3Ht+9VAKec5^e$`I-^e?n~bjRkmv!=>!^VBah|AHur?r&s;-nXTAqO>Ixm
zZF*jzQ&ujggi=V4WoBW5_TO5ZK$#)wupw!dAwQfwjpCmt?CsVTUk*vKB@|kF`r|p=
zrpb;QW}9pZIT{(MsJKNq&EAjjiugb|huwWVM0IB}BRuNs=%6@SAP8mb>s;Jkcr<wj
zz69Tqso16W&ii-P^Chmzh<ocdhxQK;xYsvll9RgV70on$wmiMsGG3Fy+p`e`f-T^-
zP$-R1ILy*NDnhgq*twFzQwKh5MQk;0Tnyr?y#7qv7pg|{6UKadB=+aRA@1z^IP(|}
zR5$Sb9TOct4z1o&e44ZHQ4GU~ZExV=6O*t_V*^xhgWpce<3TixDkNC?b5l}rF=<v{
zyx0DF*F_jL6KNTO_s<OvJkaO`9tPNqwmDLi40q=7h&<a3+b9EF_;%6pCXoln!-ZEC
zENeXc%#CbDELfO{Pm2O-nPp~`zph(z5(f3bPl@5BCc8cPFW9MjX9`ey>y|HAxVX!V
zh}b)UXO_OJuJcW3*}Try*OInKzu_@8b{DSk_FT7k*YqoNo{Fj;05?om@lgIfYmTJ+
z$4=iHz^5+vT&DR`uLGiP@EfmBoSE$O!~ndiZMJ)AG5lhD{ID{-5aaaop!C-_7#By;
zy|u%@!|PS<iGJ%2Z+*K`LP;Yg3vY<P6eDQg#LV+}y{B1>lT<Lrza7CRFO$x?6Twhm
zJ7PvEKqvKwIGq>;S49|T5uk#|M!kb(YW5eA$jIazJ~v?YV|W^r_;tS&IYQ<RPzYDy
zYSYAR3gF7LuEMryc>stNEe1IteaY|PKF4irs|F$wXf83EBAh|-_0T}xbY5P3i$wc!
z4_lLan1gkrXL82)tFNv&0VdJ5VP}`c<EXnQN#()X=(ee9w|2deQYS5+(E@ds*K4N1
z)7zH1BLbJC8r>#@0;-JzUc^*1AB>-|w~kg>;`rY+UPo|$7k7S%{D+Et_o8`s9h%KI
z7vQF(mHsGgLU6+KUF;eugFjHb`6n!O_xf2uPR7d4%4^>`FV59oT&OCex{(q4iNh%L
zJ)J!~7c`Nfn4xZ;%W57Z2!<pDFlPEGOeK~|;?v37ply`r7U#+Uli(5p(zGg;Rfk^#
ztusx(>hl3-w|+3<S4@yJu1S|s?F3o;(Z~600uptk=-O%_qBVsj&|=ZNSKyT=-nodd
zkhOj0s{6{&)t|!0VIr1kMsI>lES_8##Q28Y!YEn-Pg}?&5_un$=VWBLlNC_Di719C
zh*e^GjsAEXQCc%JLM_nGt~WJuPh3dQ6ru_WMCU{tvqGld{s<ieXv-{mey_{^MG=Ou
zonV`!z{hg~Sd5Q`uAVN9J9^SBVjfrK`C??xZ7<xk%Cv(Z-LpA0V+M>Pe1WR_UJMK4
zT4~Q~$Mbxy!N(m3X~*nP*<x0T&Jg_-5LlzEN}fChpV{#(bJEr}dWRhu?|)ZZ`seH9
ziu5+`=>bFPD=>;;^{4%Q)a`oSY{vB|zNMWps2eDxg|4X1F6fkIhlN4K6om4rS3YAu
zY?{mI_DRwu4wx?Mbo-GY2i4Z}?>=_#(R^%XcOS8wF=RRyZXAUjpLr0qy@GxVaRC29
zP+YX1Z!;m28Eo7x`Yx#wfz^-RrJ!nB<=6Xwux#e=q6U<-l;Ow_gf>>RemXeEunA*W
z2P-aA$S$41Q0mVTq)oF@{8iPDIaZt-Q>sLRWs*L^;aMG{E~AXCR^!Z2bWu`7>u?^N
zLgqi08dT<=yRb=8Y=ES_p=C2PyaQmMg4`5^^CB1D^)^^DJJ>v`)q`orAF^o;rEE*S
z8Txr-Jqlp?toZ}qs`gu7U;80;5O?ZyaqFz|^KwIXA~H8WQi=VEB{yU3l@i(TV)Lco
zf!_y|y7}zOEAxa$JDZVq>UB@8pt@n(Vn%l|z{N`U_5ACaASYb_mC<4DF-Al*{Sw7&
zR#r;<h83#@edpxX`EnnuXXU`!ldM;JmIb5{h7!$tn>LA7;f34`v31vA1la_aUPoPJ
z*{-cU!;Z+2dxTBG@=ab49}RRq=dZ^}y-?FldzhgvVfjJWPSKkY22c@-Q);yWLenA~
zc2`AtF=H~T!Jk(u$ZrNV%P+uM2p+y<2|_+T1F#f9<;~h*E^upv@Kf@72DV8p{cH1L
z+>Ihz?aWp>24B8q3RMII1j+$=;BXr9hpJ^Xso>}is?6c1C7!t(_vw{hOHFOYksaK0
z$9gXG3uD#-^^6+13KhxB@^iT}9Q%A7x!O?K;J~SC1}x1*k9S&zHr7?S2zJz3Rr}Ar
z=ECXELNWvA#Oz#kMV1m6(S;g7<^4`8kwT@1(NnodnZLsX(tt|`Uf1wntjvR9Dop3b
zTX&WfA;sSu!4>vS))I{I`20E9O6v!_^%VZ}X?ABA=F#2N(G3}!$P2R6up$<9C2u!s
zBdG0=12}z$Sl#k7q4WRxEJEDV)7kkI>y`l~MyR*zTH#8hz>L0vr6okF(B=}?ng^X^
z7hC>3nb^GPXrLG1@`{D(%{AoA1hf!TZ9tDtKO#TjMm?j$F=+=`jwCgEe9_=#-=!5(
zH<zVUZ|i8&AHhB6H2A}@{aTj`ueU5IeeP{3r&m$lAui%GdYz#T<I1u@Qrj~YuYR<w
z*bhhS45FyAH_};7xlR@%)v%_&k`s&OCxVo!PID+j;F=Z#Mn8U3V%@%run-m%r7r`@
zEBCt8@cr>{<R0*$8!Rr41Tr_g@`b(NI7*4qK6U(z(-iX$*;0J3|A$yEqSSPOxtFL*
zrco{dEhO3td%cP{TzEpWi`+I46E5_--L}?9h3rRC8L$(n3<DqOL2#12Pso>i13ez8
znm}f>Re$pLSL6upB8X%^G<^IJsfm4*8zkmY!B#fPo$Hg9tE0f4hS`&wt#KZf9~Wax
zLEb^A_1Z&I9elzp-Cd)rZ+9Yb#p~0W{>`<1{j{@#Gi^*a51h${wf*{JnQ!Vku|h5f
zZ9iN6WyX1dUhKFgb?hCdjS~$(AsKHEqr&$n#$nrvh-3eAgvfxOPWLxmxwz5WHLMbv
z8h6f3t@)L=RhkgpKE-gv{<}uxcr(kJkQLTZj>1WlpM$-`8Lv$9?3u8prT~jSnT#f4
zvcv~hVHTMV!o$tFdexOYZSCJgcYfWCcXc%{lIsfQPW${VrKkUr7?}&87Zv1Ly?eCg
zdn06@hiBxBa(aW3&R*yN=Oo5LaH-;QI_rd3KNEK4DxOJ~pP(cQRE)2{bYaCh#LAJb
zN7VX>`pUAdPxO4l;Zi@IJYOt4k+Wu+)*$168aJl}W4(&<M`3f3gO?@e?Xuq9%tmkt
zIV|wyrR4E*l{)M)^}E#dI7zR|K|lIZFhxMGW@&TyL0VW+(GNa34gA*m-WyZs99KSh
z+Tmut@-wIR42CwSd#}n<c=x%I8>V`a^&NvA7dM8(@+i^ke=?_6u>s4plphkbJ;2Jl
zKTqU%r1|?mB|}aX4kc)9@$Om(<K6sq*pF2z994<zu8m@<B4f%^*E;<e#xY~%()pBx
zeMr#Wx8a6ZFBEoc;!ukE5F^%!3tTn4rmY{ZwOz|nbwa=)c=Y@{Luqh9B`QHhoAX>7
z;lNXJB*?>&7Bi?}YM8UO&THe@lM-$xNKjtJ11^MxxDgPJWLK_kufu$!cNMa%f}r%9
za}?23bd6DF{!jO(hB&8!8N@#@bmc%F5`YLF6RP|BV%a);9H-Cbf*Pm0<t$w)I5uC(
zJKhKW`gwd4!mVK`!<TCw48k8gnY)gv(lzAf@B^f-FUYH-iTmj^o{hU&ti}&l`GNVZ
ztuDG<tBXtfJuEe+PCr?Dpi0wXprhHg@{H&~_Rdlr<Q}X}q-DClK`^?g`60W*WEtr2
zYoh9tvqk^;I{E%mXmx-9L6S2V8fEPxs2E7QCNmqjN`H0oSZ>2%t3Ic5{^-3o!9mI3
zKVWo-@wO1b85oJfPLXdjiO7XT%L2%n-|=pNl{y!$!gNMm(ukPnZnwUO2F1yYRs1ys
zmFDkPrF=2V-WGy{2Hv$@)9TZ^vIh4%egqaK4V7IJSOPH<V|GJIUU$6=p0BS^i;AdL
zss0x`*m0;KOBkvczCH_r)fuv-0%fM)g!>QR@btwswrmAx&^Rtxx$Nni6;O>WLqR4~
z*~+Gid3-eF9Nhw~)b-S4d!F1peuAJ`5LIEpE~}ZxR~90Sc^q;TsXcx8ipJcABKdWg
z-X9G)^E_>kCre3_Upu%zcl1s9p-H6a(jcZ%r?@ioPVTBDkjxEV8&QF(#^=PVdHQ)r
zo_3;3W0w3yq*wFJKdq><Rr`lI^xRMi3!i_5Sm@C`{v#N~*k#d|S$th?ANM;dTwYsh
z#QCrh$(D_@Z(OzaIP%0~*%=uz*<lB=k`VgDVgbPP$DxibU`aWDwMfTrz^ej-^*{QX
zdQ8Tx?re*V>NElC#G#)mT;dt?fo7OPYdy&sU2mLrv{BUxQou)u7h82CR`|)ZAT=s_
zJD&Fhf*EvwjREO=4cUqPq{XPrR$aBeHNFKw33SLrO*&mREcAwD-5#(6(|kcLCF(Fl
zwF^$KC~hRH;i<O_TGq|IxcrK$h-%7eqM@YXBo~q$KZenUmOoa}Y$0;Hq_i?j7H)k=
zy~3V5zq<KNK}u^ML0brXq|xjAp#u`kC6?2<AdSDCOIfD%8%h-7RX#eITP6b1vp~D&
zp>tkPEgp|%p$?`RSG~P7;d^WMnqN=AH_fjv*R7P8gSd3m>hevR(Em!h6A-WChvz=;
zJgfi|QuVy!myp%0d+GMmvHZ`h`v9H7-6&-TC48t-@D8V|)VOymGhLQs6CC2H!eqo%
zqu77QX|pn!@gHhZKJY<*YgPPC5xr>&bA#(S8uljv$M6UCTJOn!*!irwzZ`L_uT*Xs
zI%OF3R9&ogQB=3kmX!Y```IHRM{Te{Ivva>9_pa`D%>Kqwfas^J^Cu;GK0h;JYnVV
ztNcMzlI`ce@%9<-P<7p~ri+z6<R@AJbj%TQvcjf_CvN%fLrS#aofVet;@8)>0r{gw
z-JC7UY!%v498cH%i%GZ5HSXIdec!lfbBzW>se4<j`&(9mfluwxd-X4Kq+%2SvqX#A
zEI&d5`l<bpaH~>$k5Z4`?R6P(hhL0vrSWX=wPGG`dp|h6Ak#WJdD-Ra6g3Iw{%ghs
zlU_e6m{b)P$EF7V!*|4QL)ON88{|wqh;TO~dqOj7n3pbtU!BKF$+_mC)8LWsd^Wn8
zOnp+Psnsa(2qltvB>i(yK9`DU7irC%2C^k7T_jLodciwZD-hX_St5_!ykfw6C7c+z
z6l4_?6^+4@pTrD`{h3(&Dlq_q7H(6;8~_J=5F^2`PATLfBAr&LHqNeRfXXmqzZM!s
z&c<A7s!blEVNuXZWl(qbT*5fxVsn*HhU_?!0SV*sbQ=J{!&irW{jRKQ*n}uiaWwZP
z(&g$G@<o;HH9bk-%oh>NTQH<!MZt;`z0($h6!XW}&DyDk)$Y2mAH{u$_B4SdSVQSx
z=42*(5<UCSL4U}Uwew=P=KF(SR(IP&kKb@BH3sl#wkyciZnygQ|CTK7Bp%sQpCuLT
zS#49~b0*WsSKdeaVZ!?l2}0P%ppY9GXi6dLuZ27$l8URHE}v7cWXQX|tXDTNK+YXp
zYD$wvnY7Lf90HU)y^8^8Q8v!ZnoIW?<9(Ww?Mb2v+FNW<eP7#1G6@OHqM0WpPYL+M
zFzxaPG9*<kd<m2=IQ3W3cvBP8pD-YxRC6f%+h!AE0*>XJylP{n6QV<-qrEY$@dX|A
zd!Fy5^rfDWo2SDdkL(Zx$lrt*N7$9C6So5C8h#=Sqm-i{vgV@QVDtD?CIi9bCLGNV
zrnNh6EKupk>WM8g@sa~FoE~}X!gZC764@7}g#>n9cbYfZRS{U$gE%ow0RBGR*R!Zv
z@=N|G2Tj*APO@5>Ol3bqY<#CnYc8S0Hz1d_LyR(RVQX@))xN4If7i7t3Qh<Hn)<`n
z-W)7wp_n4;+S!s1KJGne`M+eKJL#koBRJcz_rA_`seJo=6n>hSzy*3p8V|RnqTN%p
zijSAc7#vF~MXIhu6!T)9l+&iY-UAZgP#uR6gisg~4pvf<h^>WiHhH@Yo3Ugb$s`>&
zcxO@U4>&Nyy%PvD(hnAhIeq4#*>_rCt&A>62rDB<$b}Fn4fegNrqg35j440n6nH7e
zs0+UvL@~E8oCCcpPHf8qR%@u2y^|9J<X~*eGE{tB@HL_43+F9ftSUV|aj5edMduOA
zUiO`1^JBYi^|Lt&tjRMCu=8~W<q<wv-kR|0!`~k0J2dZM=lxpT(D~rt)sjo}!ke6v
zq5Zj4&~i_sL@O!nu5LutC^wMkNZ!T@^ipX}UP0tVPS+w_)p3#!vL@~@+YZYa&T=xo
zy>~-JzZ-W&{8wkr54Zs!v^nrR)_M;Xez)BsvaPVA!bUevMI$_4MUF_+-^YjG_eAjR
ztp!i{y8;jbYwWdv{`2X?030wjYC@PS?04+DQu3wo%UJ;2+BQ!Z`SnX9AH258H|X34
z`t0}LDt_aa{Vc{+kNMei?ydzg7=pynl@qWsgy2nb{^DEoe-|;&Bxcjrm=LRBQk>QX
z<ko?)Kt8D-Jvn9@B=Yy*`{|v)ro>ubHlY}<BX+HXRFpyUT4kBAo>`|`6XT4Hous2W
zJt+VY5NE&-=WOj{R@25w#ykzR)u;Jj_Id)nIns5tmurf6jaM2WUWiJL7PJk6!X31T
z{ls_vl(|zYZ?3brzSlXb(Sg)~SiaEFWRoedsfi>q!-^uE;AO<SC~73Ac4ZJXwDE7?
zadUo@2N4~nE6=!m!u0*C7%p<FEJ|(pSIonx<>@0nLgQVpD}^oIha7)#?qZ{;9ZR{d
z>IfxZhIikI?K7bTVA&dT&Ers)g6p2KE^`-xQ0{Ew?zQ@7UP{M!237#hzO~-uRafe*
z4iy1>E;$}tXiHVn<F7DC+?LNEA4oqR{;xM&5wSM@pmqijUTpwQ2PRB)pVKa-J^4FG
zduZtGkcqyne_BY&@k|TCR<S>nyKR^~waOes|CV+L37;9U8sFOp5RH91@7qHLF+xfj
z7(O9i4RpBaA>rAmnjG*Qs9xa&lp;Mia**BcHgMgqJ+e1WJc!)p;!MRs!T|$c4jU0$
z+S85~TX5S&>X(*lKy#4a6MPsQak(}Wkvpwx{gT6VY6;K<Ogq^|MeX{AHWBLozIrda
z=SRTo-%f>BxG%IBh+|AVyl>{N-X9Gh0ixB%S`o8i7l0(6rx%=$8asLJ3Bi$}5o~X^
zvBLNr4TdxdkJM_^q6=N@X{U+!T?)nW(;&lDqeSoFhU5-G3ld5kAjV+{`7w>*uJg;t
zm~O{xY*n5*P0dX8>;`ZFl>;n6-b1z`xzJdh$C8a^a6Yp)0$!?p8Y3hG4Cfvh9XY^e
z+UpxFvr-)X7DBU@>mxOie`RXi?aBa`Y~7_2{L=?vel|R}M<+|^uu1jvTASJP-^_Mu
zQ_zEWHqwc<wOh|NdFB@NQB~kqFFWy*)j7U4+jpsm$8JflRt|sBbQ&6F6DZ5n2W2J?
zr&U~iNbOc3(C16^&m{u(9iT#;m`L<pE1IhXuIF_01~a60@8Hz^8Nb+A6w3P<J^gD+
z{>ySi97MXr^@}G9A5Wy`;x;#?zP65Uza_ak1Aju#{wSQY{4{*XDHu_bmSy-gX;vb&
z@ozfYD4k3RJx}Btph|HeBJ+=Q*fX=gA?KUur##?$rH6dY`dwy~_r%;BSqR1O@nX25
zCbjTG1^P6U9QK*G`8bq8lML(LwSUNF574VVzG=q6BmP_h3+Z*P3z(*g{L~0IWHUOz
zReZYO9^IGA4<6xOyK+0Qt6k`=zBbH+?M1i-rjTYe802uTRTa5^0=4eu%{lskb{CIa
zv(mI=&1hdv=X~whe>^>&ezRPz3tF81bD9We2;-C|D0jYjs%|3ksX_>!2VaV>*a^&@
z<p#E5!1Y}iWO0fq>Q5?2#}KVH+rn2zc!jgJME{e(IeuIfMvpj5_+|L;0NV3dcuZPt
z&w1K;zB{=YN*qrj_MB>*X186ZgmBoZBS={)M<XksT<pQ)Na|`*v%&=xbLjl0TwxSf
zLU`w&r`;=Hs=x3^uKQjKef3B%2vh3~uK~zY4y;R5I<3WrBy!L~#qlabOrD<{e$|X<
zX#)L8+veb5VC|3hW{4zG2W9{hD(l3H=9wtK{mw~wL~f1;YRuv0A;8LS6rgqu9`&Q(
zk&9q^d^=B_#$f6XvwF~y-fdp?6I!K|By769JYe_6Jnq0rYai2}yzX9CXXy2QDW&v6
zw<2Av9B71H>rI@wwY{mm-yz!e(SKsCg)CG~UmM$a@%XCiyf224Jlu1T99S%-r#-;_
zH>+)(f;(m}y}Q=*F;ZTrlg4TI)}&E%jVNUK>@WD(pZG^>)A{mC;LmC!xcy9;`mQg=
zmKRo>6g3u^+0y|`Q%=`5c0p-B$Kfd8QD<bBCLZe86a>1*z0>-k$iFoig{6@=jB;0r
ztZs2<YBvhucTp1qwBef;(X%?*Abf!;hy}*)X*pzm-Ni_rhCey`{cbF*9P5tat+Zh_
zUPXP)Jw}dD{(VwW+&GFdG6s`Vefgz(Abc|(z^~)HQyYSj&Op5R^X*^hB6FmE5|7u2
zz{=TY$IF-c!bU4^x0hud&l``W8KPQlMC%&_=8%F)rh1LC3qi$A3&TkSa>x{N>54%7
zk?7_2)2=570>lcQefE737{f$cZo=!<E-Nq(J7&sW*%PY!$1#xk58hkGA=ip#n5HU(
zD=uV+3qcBzVM-z3G6p&?7iBPo5MrQ=Pm8y$Flm`#FX1iX0<9(;k7n0xso3jwiXoEO
z5?E@{$0jjILB{Ik;SAdJXp%TUh87WXwOR*FeAZ@!%j;SS>{={{?Nkpqh|?LmJ8N#%
z&mT^$uYt#))pG#UiF`9EkPPKCNL5s--T~Ho)N1#_oGq==$Ua=1MKsfHvm$7dTJL#5
z%r1UYPUS3sesz-B<$10bYMDJ<ci{;<29JNnss0)~Jj<L+FTH)3J-|G<FPy5gMsjon
zjE*8i7a?&%CVM0+IcZmRMItBj)n>w`=Q=(*GR9yUXPO7lAE0V$8e_|hj5Kq=TjJXa
z5*mNj3>vAv!V_-s_;Woy`yF`<x!M9!{qQ{Q{_cm%y!TvS=9IGS6xOyqmk`lEx~kh&
zD*&uv>6ah4N1W27F9~OiH73sK&vFGg#5+EH7WV`D<d@wapeK_|A8OQasmquy9P-^A
z624=*AW91;Ooy`jOS1T%Yt0PdtC=0~<4!-`A`&n)r(Nqgx{8C2JyH{pAnzesMyQh6
zDgd>Z>1Gaus+r&+;z@6*kva|^6h1a%{P*lX3J_TdEY_+J4sYM0UM&o7mUr|;*vuR#
zxE_eG`q@OlgQbU0E~Z-R5iT(vr3gwvTvn@RZGobY{soHm=VN1}?Q5N=A3PN6@v#>+
znIKY)VO*2I?6i-e==;iP4jm}iB{Bs7at{`&?6)k3#69gEE8m3TT;DHJo~&0U`z`y6
zkU@U=5&rfu;P_ed6BanG2{`?CB~J#MqTyUAz%-UyH5*m!xoMsfy~FK54Bd<fW~jn_
z)<Z>r^@$DMG%{7=B_!(RB!R|bHI%9%tf<i9Fsg4gOss#XYx`po^iymh8f#8vK^pJk
z#am_nV^)QWEs|=)oWLD!_T{L$wDw#-4DI^4PVH<L1Pe*i*OOF@;~*C#IGlWLY_gnR
zv@R2xp=4aG?Sl0=eq+c8z8kHMR3fYAR-&cAVW+I=OQ7&VhZHbrk#ii;>awiKypzKe
z)UQ<E79aT}E)t^k&59i$BktMzw1n~091qQA!+Zjg!AC8glAW?gncC?VAR;zJFL&1X
zj9!R)-V>HgT_>dM_O=R3x%LVB6a~9cC_fbIPC!?aznPJmPx9OYMEk_y&^3F5(2hx8
zR13oS|26EdpvP;VMkGdd9moW=K3%U%g->8?lX>{=(W<R`wwEtn)POU&SVy?;@(;8d
z>*}FDV+}>z4>)LJR`4gGc`rW!nK83Yv1dMtb@}dZeVnsDK;S@Z)BBTV_VT2alKX<Z
z3;KQW?}-xReb4WI3^ipWO(pqFJ{ll0hZlebJS?F!MQB*!2}5RbcowoU2>><7*Oc(&
zUo(T1IR$#eOmXxweWyD)4<~@)eoq`@C2A{DYUw9&eI$T6ueCWxOmL_14c`G$B%mNk
zpV#t{`<oRc#t+XNK0yqr%FBuC^rl8DTCVrxNumA;zwt$_H1QDkjf_goU#Z996!`}M
zrQ;Xk=&wgfzQPNT&e-oh@aBv1<2UAhgb@Mj?JHR;pamWD|AR39<Cl^*`3So_XpyL=
z^Y`Vk3B+KQ?ay{OewjMZ^bNm3WNoq;aZmKA&~FZKR1=#ThLY8LYU6Pg#;i5i;g1=k
znUS$d2wKtGReyI0I=&P(W|?W%2>$uW3BNQyiT9VrsI(UFmw&^;oQ&XWjX45-C2zdK
zkr@EfCEK!shJm^QNC}LnAU&3yx-0-UN8|Hfo9EPP>w{SV03D=DhA}c?TR|#fE>7Si
zxf6iw<x)j5Ey0gl9HtQpkKJ{&eL%m{Si%ZiWYQaDcnm?fU>NJDb9KM+wLR9c$vOz+
zd-Tfp@yVTOR9n@7lPDlrIBIu5@hAJ-Dl~+y<yj}P*%AAT{MN}@?+`P@0eK<HXtheZ
z^U+Lh{J&RDB*8mN8`|~h$ooZVZFZt}jSWRi;5(;q4x9JPKE=Ttb%1^zw4O$H<(fLv
zkXgKswNR^fxF?h)JHQ|!oAQfGOessVDCuW|KR;_sLVv&nY;eeq`d=0FZ*h)1vB6<a
z;y~2~VGFai<aBHm{Z!L*1;STTp09RHm3B<|sFhbgR6qcdY&rN4I&s=s6{p|zC0k75
zQ4&v#W%$0~EeNRH{?JVENQ9-Nr+u573VgW<R~vN`9aq-vG^DJ~Grrx3#3wT^`njx6
zu8ONqVwl2@fw^fEFJIVeb6hFy71TDda-VLhE^B<G8y--j);=n5XPh|!Dxb~-$WPq(
zS7sTGQT*0f9_Hxwo;Lr_W9r>13IL;}Pmg}gdk9wkgbAYREQ}?$Sbut&vj|^}2e=wM
zFAFi{Ql}K-SK2YhB8M=a(hxDpqr5fJK=?!khs%Gvs7lIBC`9w8km874GgMrS;ZwCA
zLIkR*%}@%9galtW$X;A9H=qGe2UorP01i^|Rj&lA29>JI5F0xIh(*noC%asoGd`}>
zDYGvxJ{r94dddSpx^+T>mx?n(vw1i`f5-~~5I3CTU>${Kf`$+lXG?#a+J8mC^>eOl
zAeB4yo%EK6bv*C%3qXFb{@q&4kqC%3FqVq+qMlCr7P{FEwz+Js(xF5<q#(qMw1S;H
zxSv{?EWo&=R!PSd%Ei52+D~p*3XCUrK*;|@E_AEHn3OGIQEvGR9u}IMs)Ue{Bxe@R
znT96;azR#~X#7Q@&1OGvelW3}g+AicVB=i_Og1?>5g{tym4fbvSl7O~NIo5(KcgpO
zq7723tTTjA;r>zoL%{cPgkq*_goLSN5zR;H?DB8H#687S^s}UvK7j$~b2v@Vx8QdL
zB;kADp_%D;FvmMpH30{WpU6Z$JVuc`;#0$A03>SGkO`~C2<nGf+MTE)kh2UImGX0Q
z6UxQKXK&2smJ%lk)u&mA`oi^l1KjsDddbpIyVeqNy=yMOg>)y{^Yo&i!b1ys5xe#6
zebnlBjkg?>+VAVa)!uU$AW-?7g}@uvSrYgtMVmfU@or&0A{fiJDvidg2=gCymF)dM
zub<6y@$3koimpQ6-Wf?8N=Y%zd=2fEPbCNso)CK2n6;kxW(_v?Q{>@^?2J7a8?oo~
z9qdY0M`o+ewggMPeOwk7TCQheToI8n%V><vg0G6yenSNRMblij3+uhg#aaBq!%jB$
z9P0*r?i5ii73wtV#@|sK&^{wZKC?T7FNX@6&_HUXPt&`!Y6W-``Ii`}(g&Rg{_4IL
zPBcN!s47FcUT4N2GNk!cz+b=}${w`&jaot-Q?X_EuPD<)3R%(?S!i1@dPijWIw-5h
zL8tM?mGp7^!sWs+NEp)#Ehx)ag3NB)%f4I52B3q_(wWQSudgA%da_B>S-fu}`aqZQ
z)cs@2uJvn8nvRW^;KV7G@|(Em_D#JVgYw0bbe2H03)g=WRO&lnfW7hM5A&!Rc(qBD
zZoQ-NDKpl0LsA*sW6kzXFWkL7LI*tTemZjJN*24%vn;tsM|Odt07cnFKm;(##*)-5
zzW&9RZc5Lb-p8*J$m!+E!bFLnqRARAq``=n85q+XsD7S5-9u>#iDOvk2}$iOoxbEm
zVA{scFQ6P9V9L@m!7BSG07qgeVpz$?d2328Jti9Yu6ES`U&cGrLjux@vQLQ?X+ZFi
zebN7ohbQ6w(5iltZ@*>xgO1-vz`}EJH5dNLT+jM))p5IY7T@~ZwJ<+$KEVF|)kA;V
z%i4H|`SiONDS8279?ux$0mc4Q*tZMq(PDA*T9+u-uYrP~adi+G=Aemi8FI)CA);>d
zSHi=zU1fT41OKl=Tug&l+=>f63SoT622sSvVx#EeqE@Z0{z~x5+`{ZXS)RZGwdXYk
z7a|V_e*i=7?!GvhS0_$d#ug6AFo^!9WdK#=L>ov@-1DON14w$|N?^~TGliWf*ggyR
zJ|zeYH2H$Wb5sFPt<>#2$)^-5*yB8h;D+WDG=(E(v!xv=;hI7M&-Eb!xj_vuGNN2X
zfeZB*^ji*eAK$IAenj>^J}@h%YLI31VdNpb%!tjt;S*j}owxbW2)GfN7xX%TUrhc;
zGOF+tSk>6_@Bu-bcyhyO#3*~@#Fxiagp^_qlWTZNfW6vx^hwR7F<ScC?si_;`jOJ?
z9`?ijW)KMYFA}%%^8^5JF{=y3v`2atI}Q*|{ZMBmeLGtc&4PROfa-V?B7MC(v#np*
zK^n#(LoJ>I=b!jtv766SN&zy{!s5CyYBe-~NSwkR{x#%5jXxhQ7GEx4(NogZGlaq5
z&o3bNcYkgc-`2FEGR9Ige;jn|JMChz^!0c~ty#ECMqV0;RE+Jhe%C6_H8|_xoNftW
zEz{1o3FU`6x7%#|PAy#^i8THWs+n1bsDLlBne!<`X=Y7<u;X;cPAiMaqq8ELaWAY#
z`_}qdFT4w_Yk|8Pw`-%d*EI5+0u9oIFV3h^(TzMX(GQNQ`yK1%6ubV)fuj~L+t+AG
ze?6$ve*N@ZpzkZ}*37<PC_Fko@cn<w5`SV0CuP?38=jB;C`PDra^!Ss^{Vh|@7`?v
z)#`Xt6@A!xU7zNBtDI1OHce>M^RGtBPmmPldCP@HB`o!i-{~?Zt3c_josz07Lb70B
zL;r?00r-s?f+RkdJnZe4Jp?R<5Oh4mP4OPAlJ@=Jy2RKlY$z>Fct~~v{x3Cqj9wun
zS=3SS+m#^OF3CV=OC!uWR$PxOY$RXtQ~9)Ouoyb0-2>`1AdZ#g9GTK(&TsR;{Dw@*
zI<TTXa_3<2<YX@Yp`F+IOnM@F1OtDcOl-klB=6;ZsnPG@+cCY!GF$kWytQyvBPGdv
zNbU>i(kRQ8!ryO)5qc-w<EO=Bb9u||FX0_8|1dPje=sy#Jv$X2Udp#Vt*&fMIyCmP
zyP0AO%GhVAEAMp3acOP{Uoc{;e=Xsk@OGT)L|n2_hNB)A>H~2=nay)UKfeVLbfo|&
zziV8?Qk%^(&-JL(GdF3zZ=~19@6yqIr959G#J-Zo2V|$VP-Yhw<op;XCIbEne(8Lj
zuvW%skJl;b>^k6oCKPW7m>iKKk9x>>ez^Y|Ihcc54DlzpTXgNoYb5EI*%l~%{`$&;
zxo|M%xw91=YBqg{daU{87SX_0nJftRowM4_E2&kl7C}xgc9jN8x|-m@-doAu;Y7s(
zcIUD-7p1FH9jag1H*2J?X2&-{=Q2};RPKB#y@~F&OtO>8ee%!#^;~+Zl?_5oMAD@^
zo7;Lx4x+T)noJdx`0l3o*iD-)8o$C|jsV-ISnwnPfBZmVca3eOhKK@%RV??QB-4f<
zTQ=}_iJ?0B31Bx`fx)LifhM1ybSv;1%3m$h@+&Fv!7Q436#+#>Z$A8dv&~?XS_{fU
z$^D2G0~BF{DW{ci&YwCXOv%m5p{)yVt1wnGCo90)vloM>;*2ISWH{HVL2}ge*w_z2
z)F*moPmN{r74384)N9pjv=<lM$jpb8`YMZm_U4Ye(F4d`&N`NM$|8j3#`RapZ_t*v
zFK#H#gRwTAtvNCk7hGx54t9e%3sdg7(nxvqf`xw%ca1E*CbB;k-96-hzz0&^FVr#?
z?uVI2Uc7647asrqAEP^r&}?ZWdz(GV)PON1Mgds=%c)%;)elt>^p?};{;1HmT?t&X
ztL9aR`eRUvUnS10!z2Xn50}aCE&Xwyd1beaBqjROp(*9Rr7E};s@~&ZTplhoB8>;p
zs%0s%oi;61S89S$OVwCYL{=xa0x&6r*Bph7Mo{=jtF<*m6;M87Pg#l*O2FgywlK#`
z<oO(>!1BLvhSxY!1-R&Tc{pbCmq$k%2=rzkSq=(9!msy~uBbn}8LLLZYqgo0VVV=C
z73&FHRIxGa*4g1to5Cch>G67B=TB5`^i;hpee~&M17~?ioVD1?Z;IvF!>&(2Ch_^8
zn~popI@!-gXp=HefAnlldVI3Xy3@(#xnZ;h4O~4M0`f3>&^|xn7U?$Zq;6IBa^Ks0
z`u~6I548FLJFahFiY-)B`t%~xFXe-$bXnKv+%Zz^DoBzVjWG1P?qP?FSF~a!Wu&+h
z`MW@2>-2a@1<+p#1q1R9FOT6jaxMgvYfg*ppUhSOpHZ8lBl)hpSiMU<r<_s>qzZ2|
zkPR0gwjZW#tMnyC2H3blNBcMZtmsu_5>1iyO5%|aCa3-SGi|s^8w%g>)n^3R5+O&M
zgxHXYbPx!Stp5W#_eVPnA@DG<S<x}8A>$8uwPLk2i|7R;`B;mKM5C9KsEZ=DCXuB`
z7eLw-18xp%iya_om4WIW#!V})2$DQ7VP!kaSz`;<0BJXU7yZ|l=ktYZE_|yeFdzK4
zZ<_aPjFx=JaQ&U6uKxaQz<WQCJ(OLaVK)7+)6>TEd$uYqpf58}FeI1^&<KflAa~a4
z)}pXkRVUeIDNT!b3crg4?Q|jh_lA1wc7Q;<k2y0(BB->vPqWX>Jee?MOa|Zt(0G<O
zXH2SLg$57#9eC|_ULj}Me~hgmw4FT9sL!ignvp0SAiVE`heTOU5E46ugI%Ug6E-O*
zYSl|M4f`S##}z67W0I+OuQPS4-jI!%NSX|P;UT8L;b<tA;=reO`uG?$DVU;|!ST6C
z`)un7pCMz&nF(KGzH1ELNsR&9UPDP%ix80$sGSCfF{NV(B$Uc5G(cbM6X>k#P>32N
zAuTUmOzk9^=s;5jl4cjeJR7{j;E{+y5Q&rE0+n@xEOT*sFV@-<Cst31?)AVQ4)qU>
zt@cn4^mJ><&2E7pt`7W%k6PQAH<-&5&dp2_tEDhG4rc9Cb<d(?s49#KUn0!p3$PL*
zCsX91<0F0}o%>7{2h=IOadECj8|&*3UPoG-yyq+?#9Wm7y}TN8b^~p!_wfr_`d5iB
z4K|qp7!n>eT(A(cSR%GnXHCT=a$9rl7UvEYqM&2=>YBeuRUPS{Dwh#|<dtDvb)?i9
zBDs$LjPzUqkQI(xbGkAy<Rd57sn08Y#^Ip(&M2x#+!NG0T&41BtLHPHXc;&x#jQ+j
z>IzXz(loG<LPT(VZ!c)jLhO1ixW`(Kx;FvG*6xjM!J0=45fh0Vv)wceFu4#d({!m6
z&QWB!)Sd!nS-6<s&EEv38w%%y7(umG0PYb0KhAtOV*e*C6<hb3E=vqA#EsA6T`D?O
zXvwuA`m47Yd#-7(O^12*u&zTfaqOSW?v?txW4fCs;uPqxQL%`$4)=}^lkg38=4U|L
z-z>!8!T=WrSEWxV{Y&16_&$2ktBEjIkkZk?Vn7CRWNc*p$Frx7;Jq^&iXGF_b!+SU
zJPS|wh5vJpQ0x!_?Q(|h#jfA3Fw~AD@7YJUJ0hO0;8vf#1ySRlq||a;casvy{qf~M
zq>Ww3dz&1U4(~tZz4Ep+_1savEHj560T=&Si46>g2lpdQr{LAi3i;hm5a$<!sKLpg
zgNltwyQ)_kP7|qj2MBc!<|Wy(l3xhHjLGsAStA_6zTU7qP9xb2QKNk&2*mV<$S|GU
zWYAU%6@(alRvR2_n96=GxcQpk!mdar!_)ck09Re?9LJuBh*7`i&$ZYB<ARqEl{4Ur
z8({q#N%h<7h3%Ebz75aNHv)Hr6(^4bC=SB_A|sK8*lb*Px9oflU23IvicccD5q0dD
z9~%Yyn(c@lY>SWDO^G7KN7i_WSBJ@iL}MDV0Ww9z<(v`Wjlw^lP0+JCSMnTmpD1o-
z7hQZJwp{BtzCPZtc_O|1S9E*c;f;jITAVayih-cRn|uF)fK`S-D$5{O87-S1NjbJ)
z_j2Zi22YZV7XgUlwy`FTkM$uUE{1*RB}E7HYEhwtP7fWKeuAqBP!<if&&1gdjSXcR
zC1#*2dmD#b#J7Ir+3l#qoER>!tBE8{6;S(u-lRAEHB*vpz54<;bF^-2*olP<V(YwH
zrOu}Kx}?bf+2)RAZ(p>wtseRUl5FbVoF;NkQnHBL0!@>%6Ix)&3Jsd~RCY&3^DM)f
zYV0{z1;AXgcq5{S1@Z+bXhG@$E+b{3>D3KOS8^n3TwwOH@?aZyG=2Q8J%jB3W9h1c
zqHN#pF5L}+NOyNhEh*C7-5}lF9RiB5gn&plNOyNh%hD+#-Tgi9&u@l5c7_>-eeUPJ
z;#}uC$HSlV5tE-h_{N04DZzo_j(nZ+XOJ1q9b9uH_VKBu*S=KIT$ptQbwd40%7WpW
zEx%?AkWEs9s54L1w*|Mqu|Ih!qGOJfe)_%Zcl>g*bn}nQ^+J6Y6q|P<odx&X&X_=T
zjIUT*zHi;BUCx41^2CwZ1w`z+o3>4oHy!nK011ILz}I7V*&y(2#>@RFVtV+B-`=iQ
z_kb#ASr{@^jP52csKmovDo~80$=X1@i@J|0I7Q+d`;9qS3uI<tiD$-!gn^H*$()Li
zet`;H>Wp6kdlo~CY91urymXp)G5sD(mb<CR`woi>fZi&%0EILJOrCv$n8+8u64^k6
zV>QLGv`8Jm^1XK?=V;lZe%ANIc(EybwZCSR29vk!t?HX&^5rgMTZ`(gA4B34KU|Y`
z#bS0?{w{_z1gYqEha-pm{S3L`xA-nB2Q%W8{cuw--l>cQBJ3=Veun_RL_84q0%av;
z=%yIx{HY{<^3mhya^B=5ye~?iX<OtUq44ArFr3Li`9eqw8}3MKk$d|?(i}H-bFAJ<
zPZr6Hx7LvzZn&G|<Pl4XvHPnR4h{}3*auFeq$3yad#I#tXKO#vIIiM5^K#=9_jA7=
zAYnO8wRqh$LMvb?P^Oe6bzJakF@=O8lxI62vxW%JXTd>1W6ncWk3&qM9R${-CL{Lu
z(UpIVcZ4GNX?jw$^umc~<F<DiaRF*jAh<LbVYPq?3R_t5n5ET@AwY3^B^bxb;6Veo
zN?=$i^2Vb4Z(o5J(s2Fz^>{4B4I0ha?x!>?($LqKVo(eG?d@mASYC2|$mvYfb}_7R
zEs`2jVy3a<=Gc9%^DgHS@$^X{Yf3KB)k%@!==R98oDjvmadZQM`<x(ER>Yrl<okf-
zycj$9hb3i<OP!$sGu$6XV7O@NLCE*^SpxNcIr*{U+(@z8D~1AM%!%wxPx&InZeF6W
zq~V>gD5H3CoAw<J83_R_<(C}#dXcCmpCNzXG)G3>Yrz~8ZTXk{^ZaKe5nm+}Nps4d
z{Bn_+c_qnZj9La&U9Ci{3KB->qkeskEr4B@5&P}!whE*9Us$T5Yg(P;H^xHA4`41Z
zbIFHDd6`KD5u?c>7qbmox!ewUO*&4dG+5YTx{ipbsDP@CRRHEKZb&x%WID-(U9Y-g
z=>so6PLm!-I&&E9R?&xC6Xamh^{3`Q+wei-+2bGI5^<r}m*(&%+w3hC&@JSva{gQ~
z1U!qksirgcMws5tnWSbK#bh+iGRT@Ou|DgGn^ozp@CA_Ea<Hy*I|D`H2$&x~OT5Sg
zFV8ufJP|MS=j8PM+R1Z2%5982#kQaY$jmJ`eK(Nyj4trf`+K5Yaya4tUd6kzfM`mb
z1V6G6rtp+6G3Z&;H1^t*`9%^nj)hHwr@v(tI8jQdB{mB!ap@HmB*)||4|IG$VVEm)
z0!AqrPT_9U+%KIaT`Zo6%&IyvS`c}O%K3mBO-g&Uw!_x^G;=$nl}ctVco*?<>YQSg
zT9g`A0SJkZ-tI?rouqp<=p;k=>pFHyZQxM~s_^GPamfkX5IPY8G!WA~ds@?wy4^>i
z=MjZf$@ZQxv6V@Q?@5=sTq`jk(sI#1{HX!Sw+>kYnU!P}gSYUqee}P6pJDnh{t$L*
z41W!AHchGNDZ!ROp58;T$CiHEOrpAoBJ#A~;mow!AVt2lh8H$uI1^<;df#;O?r{_n
zl1OqLNQs{Oka}-djBFPY1^-L6uI4N$us45atqT44@Xz}kU%6)9njlxf1Bj>i^-^`8
z{>S(W?=djkKm6nK$qu}&n8X}q>~%0MAM^3$P|+VPe@lCwh&#3%#oiYtj9WsLs0t91
zI=yA1CZiDU%<u+ohBLXRCz&-zQJ6kqL;fx~>-?L&v|S|YDD8pP6_8fyz=uEDJfUUs
zx+i)19p|VJdAaP$#eqQ)Kz^>pZ1OYWM-)nRHLV^LP&4s&C~;whoOH?B&3-Wh?OKL7
zvLgw}2{l3ya}AM1c~>^^-9f^|LD=FU>7NEkc?tOh-%JbyKbQDTq~$PlB)Rsq8Z<nH
zIsMf~M~h1R=ruF=#%tVfj@5rNTE42O;x_k&2z7f~_sQZkzMCaGpZ5CQJ$S?J)5&S@
z(ec;vTh--{<9d@w7wv3U?+g+#KQ0VO{S5E^G~wf6I*LSJCF+rC`VBdR)WJ3mKLvh?
znJ@yC66SqPrr{$0yw-^DJz_oA+ascNPfAeR^z5g=^~uj>f~4+0Q4J2C6=Avm>cz)b
z|Cve1_h9|*i27(6cb%i#yS?F0&M8}zFf>O8%<8j4G_1#k08!F$q;tF#Z0s$x4*g4W
zzFkeSo%d|IPDEw~g>Rue!EJUqDaEK5GW21|5Q5uBn9ZEJi;$6GC^IB%@Ag<l=C(_Z
z+J;x0DU{Mxb_+on01;TE37M=?_0{aTAXIR)*Ni6cmc_FPW4?_d!9KZ%wys$SwpZL9
z8nT*e)6BZ*q|#_%bQ4UW8s6veD3h<j^ziVDW3hJeyp{Q;YJY3xUyekh;`j;HhR^v@
z7i=Lfa^UW3OiWXsnpa64<gXQra^4u^h5Tj*Ni~V{<uUOkg@Yu%vPi{`lX*4VX*G6G
zLTGibO*3%&8}0|@P$VN%{#X+q@m<}8RyKt$gY5(a5}>f7c7djDByt)ng@?HO&=%{u
z^+qK{$LH@~kC9)0mg$3@g2q?j%|zOr7VkRVP@erSi35XfO*p$9Y)pAnVIG&oJfuHw
zG(FzkZZ6vz5UNU8_g)BXyP1#EQkEjA$C<NjhU-u^<fp<hA)OX1`BTp;ri0U<O7D5L
z-R?v7i)|Re!1jMZ_G#Zx5@UK;tEQoeG37v*q2dItu3!muc^MV);{|c-XFB7my+U1J
zJF<Dr35$r?c)cow(Py$dgQ~&HS~_%p44?foa_~mZ)8UVmePzvxWh8v#^)*QTwE35J
zsRWX6VYR^;*K*cAk(gUVDZa%f6ZwfK+$HKg=9J!jPXS6?0?b&B<V~roM6;`ypAe)w
z=+9_W6@^b{pzSeUgul8qmT7r-A+m`-kJPpjF-g0zG(zZ>q0N1-8t&i*9c$}JAyT9@
zfdVGrqGPVrr`_YfBYWd-Kc${+5Q#{RUz0(Qqt9AzX8p*W2sdvQv)qf^b$cHSV*?qt
z<J*kRbE;4My`KAC{;imL?+Qahq@;q52-cr1aC_chPzs&#(CZviQ$yIGHoZ(>25nYn
z^fhB+$`2AoXrq59t6QhYrvFTQq-V9}fyQMDEp@LWPNmx<;~B3RQ)!)Q8>hs$w*L2g
zDQen{)E2|C&l_oVW;2a6-q9tfqmMHv{$~JSbwi4fX4TWW1{ocu$mLW*t*++CvD90!
zm%TmR^o6e>pEwpYFg}IP#&a9hO-H0fvMkWnL`g;gQkIB{sX21+D(x6)J&#C)salVx
z94d`@0`?U%lYzEP`<?5H|6tG2sJF{fk8Q~8=65r=lJLdf^&U}4H&csd9I`tVP}KD9
zK_Wk%ZF0B_6Iuw;Gs=VwyZw8OnKLY#c?Yk}gV1_`6{L3sq=puKhkL8Uq|VOmSt{~>
z6OJa^4>8zLHY}qfh8+&x?2t(<7iGU~#-eY|2qOF%`sPK!LgB`s?8^7WOmW#5Yy4jk
z?>*wpgedR~4o1<;zT;66S!y+mmCoNqfzN9$)t>W=q(!2YV%glCb-sNZ<IsN5M8Sa<
zjr~*`CsnW=WmQ&$z}7*w9niY59Zf+g4^Wq?br0jjU+$)EAMbdAp6^~g?y3a$-UR^b
zwsg&QN!h#QiCi>rlzYyEe-)r6<+^2==}j!16=7XZk1T)Wp}E1`0bD=M>x_^5=u@dF
zae<zwc<k>6Gs@Q?oSGDCgb-GxiGaGyo-1p$LB#|(u+`wMTMw^FpfS!!z38>y4OGcU
z|AoOGlx+J7=XCAa$Hu#)`eHCC!F<<i!+D5u`N5x)SG$ta#U^i6<D1WM8&M%Uyo(m(
zgI(c;ImLCs64wx}ZBM0z!|*i3O-=Rsy*qvlcS8f7*1plvS0&LMWuE+#HFB#Lkad%n
z#S!!!=r^PxlwW2CzP@x_BY|vZK)WWeQ_B7AO54$S+d0bq>gxaNmb_S=8-YyAzn#TE
z(AtdikQWh(EDE`7%|4<Vsi0|U#n-TDhbhL(jD2FnV_oVCI4TjYPvU3Nso86@27RoI
z0MGD($9;SQd&Uk~#GX%aFYE?=MYr$IqYkeyJ&?*CCL3&bi(CrNcs6h7YC=%d5LsLd
zUfI&ImNm!1@opXzQ-E_;ENv(y%kw_z=EX7Qhyg%d*LU99uj*<Fv-xGv{r}b<6RQ~)
z2{C$wczomqzbZrBB<B(^O%&+E{(`@3LvxTKvkNb~arPEu$8WC(oXQjE`*BAcjo;zV
zfg8J=q9e@`g`>{oTxH_JDX&%%_Au<HO0?$@sPj-@jcJsgkrkb5j>Jw*Ue*h4>m<W~
z2KJBnpzE6m{Q6#$E=$x-Le8$4)fHRNquKJZ;fdmaw|ng7A+O<RTDIFzcVuu`2)@8{
z6<!rF1hMggtgDIJxzoPP;B--^#!+{<dLQV*p^L%T529kX|1XH||DP~>n{!uj_UTPl
zuo@@QB!cXk0LTeXb~*wJ!!SPH=NB^)OO2s3(ZDF``)sgg<fmOijG}_i{1Q~(YFYk_
zV=%PwtDwGIP6hYexq(3<qIxpea*?)n-5+Bo(j7SGQ{IvyfuU+;5PJzG#QuZ^Z9_4U
z?YnK{$Abg-pr>=B0#q&3c|ED}nBI~?YPu-7o}na)V`#%ox9NC;v>gYWy1bwY2|f>z
zakps!lLUT7u(=1hJ?`+9;?$bzE7QREQq1!*a0yUBekA4>`m=lpzCExga^-=Nm9PA&
zUaMU5OEN5+7<7+P=?~y4*F7@8Rb`sir)<y#ee`p|fQg?=DK+Eudz3PlX-n_g;X>>+
zWz10U7;`C2DY*n~oae#{N@iO~;VY$c%?{$%a-n7k^$){%GOgwC4L+GrY%Q=fKr1r;
zxOB4i@v?K^#@XY$kL?$-_XgsO2%M`4=1cD^Hl{6ddp4Dp8@x<={<p*{&br=-7vnqL
zB>qgpv*}YEf7^h@neZNtlQZnNruq-sxUqqobuX+gFLgP=0`ncOHe-D})*kQ4m$;_)
zm2o}#X!2~gw~z1;ND5k1ZBXnIwP#@%YDvy>A*_3Pq9fa-V^uXiBsoAXBAX}Zc9Z7$
zeCj2kDSvv(Syin-1m+<Zx&V;Y7(W+kzmF)snA3nxTbwrCKrO8m@xS|daBFhW$tsPE
z@AP2Y;bby#!qFm~E)=U<WWu1-kCK`a`ILAy{=Z7smugTC25lTqfR;ONlfOGOI+#e)
zMZ`Z`dfPbMqT=|0B&!NduzOuUMN>6Bi*7}mrvV@1md@Hz+1PPl4g1tWvEbfk*j+@Q
z4-3~I5-qYWUvA*M5tNGYID#;YF^514_e1JkUCi>}+WiU+LB+~K#}q05_VI`2RqO>b
ze(MwzNg_Jz8~@0-!Vt6LXd0ZTA5s?kvKV4vC)XAQkB#Ls4?L_oxA%yj-!?u(yR3eX
z$AbJ%ZVxbk@CU^Zmf<g{B6K^+uc+b>cV~t$x8PYSoE4)@Vz3du#kA8hECoz>wGh+?
zLXpAaM4M%MhEtQ0`hP3Ai23qANFBt4ac5C$ZUAlW+pnjWj}hvup>=SBzbgmE{g%N!
zK!7;%>eENx;|-NN<#>6Vy1g7@2)Mr_#p6Ebm3Ut(Z)dgRG&Z0511Xuio4Nt=xv(y!
zZUX>n_A<;zk5gUcLZbOQ5ELZfW#zSdtJt;m+4oHYF+!p<R6=tG8J1wtMemTkFf*Dz
zy{diPG19XQU-0(+Tu;2oVPh%C#t(xF=R=#amYTu<@5JEXdt?pOi+9qvs)%xgV}R6d
zcHUu_Um%|<cKNEvKE{l(a3&trE&}yHD?td#g-FIBq`@#LKWz3O0-HiSM58x-Q3u#S
zZqyrPvcTK;^f^>Y)392nB0GfF?5$Dv1!}pbZDU&mL7jmm;n~S3jxC(S=!;{9)(!l;
zX(K}jPcmWVxRT_DNOqQFJVu5&PBR9X%Er%luy({c2Y+hlZ$T)ZDIop!J=SMMh}1V}
z@S60#iUPLcKu#*WN=EnpJgxr<^07m7pCE1`2?Az91g>;Iiio%YYC>3}W&WBkY=i{u
z*qX^R?Y&v20X2k)kUDI~$H`JOZ$?+z!Ww}|@3}+H2(LjwL!M1*ER`W1MAvI428hJw
zumb|sV715H+~=^mD#iwLc@pvGTN)q_+pr$~Z$#4<om7+>WT{yVev?4UW0Awd-FJcY
zc6(OQX3lLMQC2jU!E)K={Pyj<*`C~Xa{gr1TS}F1{NQLOSmO!7kxOF3`bmMQtE_c3
zVA4r?^4K(|+k-?Dy^nxH_OReSn(^K}(}uN)P)KEv=<{9VB>wj`B3iT5MMC$<Th0le
zr*qWZ>nL|?#q0{dgm8<W_RlUtn|(>*8|T|RmP|emmn7o}`xIhZM_vQ_!4HTCy>}ji
z2LXczZb*2W+tH}A9Qf+WfVhg9eb3QVTmfoD1e21WkuqahGaJjLQq7vva5}oZRe>JW
zh^l`nh>NNaBhTrP{RXE6*&#2>KKg^huSCR%2U~rQ_J_V+Z0(-Mk1R5Q;t}t~yNyG-
zWUG=WJ^F3E+kPL`fg7->{l{oDns;fi&;Qk-hk&k|DI2<Rm9r|h>OkThevEb=+F-y3
zZE!#S4mA|XNqb&yg>(f8H~DknUNW$I04I$TjP-+#uWDw(LmDG7j#S#ZQwiy|^-c*;
zng!lNIO*mN0EOb`grbhMnrAXRp6*jrI>V(}feCKYgKs$2_*LU6x%>+wH|7;HcKhX+
z49L5wziE+fXyj$V&5xEFvG0=?9=OYfmpNSs1~^?Xi#1_T3;cxJ$cFTOJ-Q%*w{kq>
z^d{*s_y69p5+;thK5T02T`_@Fj{$%9UOAw>lYv;&w0&KYky=j3!swNv5PYGzwPawM
zIi4glB0l)){=a?~N}!_AJMuZ898H~m91sX#-W&S`!Gzy!)K8Y4B!8~}EhZ^0=IIQ0
z#vpQfK$cHlM6@$E%wZocMtzGP^q|tUv3>FG@yO&QO5J&yP;lba_$-3eAWUD?q%-6k
zV-DYE6k2!USuN}+F?@}&s;mQV=Q<rrlAy)QQ~kq>iHb|u-G1C_@^jYHQQJJoqT+KG
zKY#N>L|ImHxOY55({3VPaPnO8+8x~>9?kDHO8E8_#8(GZiT@-e*8Zb$7aA*t9>leR
zWI!c0nifb5aX`D{Rau&IdSzY*U3SsN=5eU;(XftRs(>T<PT8C5oXH)7>X|Z#^C`U|
zvj|uau=Ksl-ptxV_00rCS#^GQANdUcKXzikJOs6YpuMwm%i=y}*HK~=Rh_Le3IZ9Q
zR#%m&tzgNEo5~fp&|}-6?tQBvFURED9lc&`FdZ&2!flZ11kHM>NlKk!=|6nb54Xlc
zpu*d{VmLPF4EN7U5=byo4RHw}?-BNQXFqaV_=zPb8@_xJ`=ajgOD*))sbv19kM|op
zy7Krm435>Qw}Qp8<r}L&wsm=zA{woF_8v*`SVF^!%iXDi^;w2%+Zt<O@(+vIR0*bU
zL>6urBH@fwyQmklB;2^8=`3ZsA=rBnBoSAQTX%gtjJR>EK0!f)*~-m9zEW&xjN42q
z%?pHDh}4AYHk4L7BkE`)`XaAMR|(B@GaPYS&8oHwr|57)HKu;*mfcMh_2>1I*K;Fz
z;rL1WyrI7DOzpW8{xndjkKKdiJNfpcampod9m{M1i>T{?a`~IBdvJioHRk_M^Q!`)
z8BNhYO>_28KZ`IxU#pnfwCE;+6QGH5`D0&CKKzW`y!isH5P|83=#vT;s1{_q1a)p4
z+>s1Oz^?9*T=RgTUjjw&J?>vV2cZn?HGOoaa&TKJ-kx1AlZsLc&>-tB&g1Rjg=*B<
zS~4o9<NvYe3kCPiULcCU98NJFzD+jzk8B1Uv;;_=u%xI+RVaW^>KC&M;5Ma^G!KGn
z6V>)fIdZgA%t**^u!hNA`~~GGxjjUZ;|u<NqZCc*&+WWTx&ZmUu5c6Br2xhk*Fs_w
zNAPGq5XAkK7RE%UC|%M->Vs>D2frKT*^<A2$OK=8{)xv}Ma3rI@f7@+eRiLP$iC|!
zmiM{UAcR7-YAg<R^y4A;ZA{pfSKH?7dQi{B1rI2`m3<M>58iFEq``a=xMN2;1>?;k
zUtRs4Lk<sLDgLOnQ%9R*GE?hOe^sLCA{P4F@_i#ox1y9Wv16&lIL2BW<5o{S_3xfq
z?${EU83VB~%fGkQYu;?rUQLO2csEIPPs-kb2TY&64xevRj!to4?`Q|}q+o5vhfnVR
zCT9o#DO_fUa#U9%tuwYA#DFVt8Z?>vN0g*TzSv7A>XT|0f>0(F6O@?c$lIB%5DSn@
z79NpjQ&aX(??ko_wvpkrskQMYX`~priY()*=fao?7h!}9+5CUgO7d5nRK0H*8bQf`
z1_h9>J&jxgJE)5t#*It=D>ndbr_Xm7Wn3<ho$nSqscHlzT9m0Gp>_V|rII6~-gcEh
z-y+v+<9Cg;(Lvro!0Gi0obzp3sDWsvO=Hyh5Z19jzU`lRpJIFJ1%?yQ++K8Q_b=vp
z@Azd~9{b$PB6{>F3LpPZrMWUfM@OiqOj_7Q>I7XL(YVIo5cZVdSL;)u@VKqJJ3#VS
zV8z&2O+khYu>lQOD{i44n8l8LhwENf@@tn_w@N^I%gKx9kaueD)h*EdV&8&`FGfB&
zcXfu~B^Lv4=_7v7^J4DV`8nJqLxFS%n!Qm-R@g;EHhpS-o(NQ!iL*UiJkILEmqND|
z%IEIS8N4!lGpt5MoA%g2N4`R^+;v8icUQ$<(~SM*HStz;Wyj#vnZZu1TBOfiOWUER
z?aj$j(LdUd?fDT&x4KJCf*DH#G((425mk|7Jffirr87?GVf_)|T>M-=^pF1g6+0uv
zAC$!yNESx%Ocuco(2o^W9Qw6>-5@RNy60rNZ9=sQD<Jgs0W~0OKH!;Ee(_5p=mV2=
z+*bx0<-d9y92|lsliDg;;}zNqgt<X?0A}@CVGFfbLqKI+o97h`$>GqnbyPf84Icac
zY|&a-;sx%<WqYwNplk$l3sTfLMI*bUzpD!b?Me^0e`nepCcfX)hvPkc?WMfOx<$FL
z@Sujcbgk(1>8Hrlw6C4IihXPEpsYl<o|KzSC`0Gw3A*=w!P;#!&tWjEc82ypZru9H
zS`EO`byli$Sg1-SYg$YgKoA5WI+vV0=E^xrT(cZ{;8$ZB?;||*T*PwEE?)Lu;)|0g
zPA~U!RSP!KUpB?_KWz-<isj|tRr&Qj_93dz&m8+srxL-RDk7v>K9_NUEz)d7<a*m{
z@eTIFQ>q(Gy44Q#2w-%-Q`t0-Q<%v>PV677&0mN7Zj^eeto+e;8?^N4-DmEQ<F+z5
zE%-HqSy$(m?%@9&6tE3)17@9%MO+TyNrAJKk&qUMD;%2Uyo#(-+G8k5Ba$kV6#4Wi
zhI1FeZ8q0Fm4YX@a!1inek6l5{&^SWk}=McZB<F0aF$ny)V$>k0Q)E#sBvlb6YO#z
z4%P9RD^3Cz6vDYRYed=mpS6y4bpr11UZIn#%53pMB?(iBnQTIa;UP=q?6!wV5fnt=
zSbXIR)X>oUpd(_`mR+g)54ci_OuzU5+l`)p&ew@MrtXW{w=1h7%CYOg738}eZO<u#
zFGdDNhgNMSyg6GUDf)UHS6=WsJ+r*u*Ni1RYg?#oOl4AHU7jvc_q(1W3L?(ys6X_2
zWS+sMa!-Maz%+9|-fxWFyqKabj#IMKLtku3D6XlNHh@L%L{6kuv52U}ELNTC@^l`B
zEgsZw^77P$Y!|7Bn)`C}>hU7@ieqC8wOYJuBRy9fTyx14BKaq5S3h*@ohpEgz#$&;
z%b9n*Uab?)^MX@F>=pa={55tatf!#C(_`aZpIuMINgN}^EB~KM;$5>%=O}n5*S`nT
zxj<BRyPZq^iihjy!nzn3JIzCy-VO2pq9DnCHXnzutPDLNszQgBBM~${f^Z)kBu3AH
z?{Q^_5J}}9ENhl_mLkLvnrwgE^=d215xspRdEw01SmgCEA-!(Rr7;=%OZoi8zltEz
zs`g;vX@4O1wpIZdxG}SR3@(x@gi^AUkjZ9zcBj;am;(@n4Y*H$!APB)bvRK_i9GdW
zMDukrGqZ|O%erA#NI2iM%W;d!N*X*E#FBSY6i{6#MKePv-fMnRh34we+G*odHIWuE
zS;*(~se(>HMUVD^7l?Ehn=4#wu=O_>8)Ox)y#Eq^d*U;}w?Be%*s76QMS*Jz(DeJq
zJI6@>TwzWXqm~e+f|L`+@fTXRt{YA7R1l{q0Ey7ErrcnDSzU|@2IXXRFb-aGmV5_u
zq980?1CGWYC+VS(U;HgZG!o~tmNyt>SNrV1xr_Sgnk4x7P~5DeLk|&)`thFP1ylTK
zR(x~OBuD~I;t!g&@Q$ew)fg!>H(3zeKsmuv_Iq%S)c3Rz=@B%FlqSQ*kS{NjSr=+}
zd74{`hk63e|8e@~qRZ@wKVCRQ_zM!scG(_byNTBt63QxA{ez$X@HBt#Al$Q2723bZ
zXnIbTAA+cF3JMRk*7d-&stBw9y_mKWPJCL1+`y`>p+!&@(YgYO<)j<50<=keUZe5o
z%2FaiDP>gC;N&C59rk$Y3V!|xB}J>_Wh=ue)lHutt08g0D^26fBoR^qv8qO7aDBD5
zB@B4JBsJ~~kZ7qaA-@~zPpqV@`-AhnG=U-Oc>rI&3B}eET7p(mI+l&E{u;{ZamvfQ
zIvCYIWqr>>XH2LY7q&FVfiHAiL}uzjRFo0xJ-Fv}_PzLIxAxopv1o_CD^TLQ(J!3w
z6Xp@McdVSI5~oY+riEpLD(B(Y_b|*QloMdBKu;gGAF*yMX2FKoAE6;;(V{o!JjctG
zu4Jcn*fR}?=!ruvb_>#QL_(@+tI25TltbKXa?%N3^q=vI{_9q9Qy|UR%5v1Vn|CC!
z4DA8F14pe}yt$N&KIP~OI^Y9*H)X6n0ut2PEO;(Y;^9Q<zgIH#EySGe?(R!N?g<<0
z3q^G9<TQSED=dMJS^DcsS?SwJ5ah_f7U3VFa3d2JTanj?TSF+X|6MLX0t*Y&;Kif-
zB}M1-he!*+zW$sg!S)(A=Z7-Yw7e=FBT0BzDo~NffSL}&ns$-W!baTddjijZIY2Vr
z_W31(brjM=jA>KI7TtOni6#A<{TI3zoe-4bF%m6=F-I`2i`dtmai9THTO{N#F?R9y
z4^KlY$#4kgIvd8u#;VzLqXx8DBf*eDl9VlN${evmhTn-1KmAXeG7eW&Kob>CdqOfy
zVOB;dKf?be#<HcJf#8|sdt2PRF;)Jk6=T28uC`sHiff_(t#$F`=V6BXB^u==y=Qx&
zNh2)ozdbt$JKWFZ(&{mJxz|ypJ{=#t{3OYd@EIpa7qAPO$_?r(MJp1aDvtQ?<9TC<
zW=fBo<;W-D4_4}h%R%a%6xs@h-^Fae5Js5|b%EKcLO6u%Vv@0=C?GnNpt0}sOkN-u
zX{W+4Rm6c#CHR@+HtFT$1-E*U!{`1Ke$+9P0KACWKQe}1M&>Y=>l)p_mL|4Ti`3X&
zCU@I%tPBE8hY7uYDVx!4ZhYl~j7N$y>zL|txVTW6mlc31A^SI_X(p&q)||rW>&xp6
z0)@2OwuJM4D^_&@50p4=NahLrE_`e93gG8=t1JY`P@$!<T3;&cyw_4C1<B01nx`;^
zyp<~5fj>=QD6zxila#EBq(Pl)$@{3l2?^O0WV?#xa!sY3HuXNW&jpk*(y`EKGjrd}
zIC4~Xfzb>nZD_4)K!C|o*CT8_Ow%r(ncM-li04n-ERFO9ubCS9XfkgI0-)-XFA1OE
zDi03|aClxl_eO(f&flA|HkDDOw&Q5=ZL0i?<<}^&<DGp+G1jKTw{_sX{7tMCiShB>
zeCU^pPQ+UU#9NmPAmy+q%iSJ!J+P#?4P5#4VW;PY6my1Vp%S>m5#Qb}QdB;Dp>#%%
zL4~sj=FualoHkwd+MqHEQ+z{9pv@<~K=8AE=K-x-zkI%QAwt9oa6t8)M2(2p_3CqL
z3)AQoe|mT!QND;UdA_c`-5(2%+~4Ps*Q+&NcacX;B}hip$JfO3p|&M_$;6Z(qH$N?
zcY+(wKD0KZt@DtDS9J{FtPL0TujcZ5TmOqXsObYtu_j5qf2Hz<pHIno=Pc-4mz`H^
zg-QI3Lls?B2(<CPb7}+F<ffc@G<;l~;3JiG<lu<+#IzyWjC43E6LV6QOVSDh`shIh
zI^!7<c5rzmI1rJ(yl4k)+|)zNu`47s1=!i;<$oZCx!8o<_-ef_dVSuV58y>){dm30
z!>vd)na$-oIJVDm`*`S_h`24c^Y>YKLh_H6AVQHx2C8qJ=_i<h=ejHzG|@Ha?_w;z
zI-~xj-3qCx-OcQHYFsL@WCAs!7NHhqKd%$R()UJk+o#&nMW=jn@>@?j*e{c53R6q-
z#6XuyyqNrq{_lwV9_X!=(^(sLsi9X<3Zu8T9<rGv8F=jCj|iNbi!Y}<1w`;WD--ei
zV<g4<4AcD^N%+r~o4IGjTA>Z5zHM8(yNI4G68+c|##x!%7FQd2`xN4mny2UY_E;D}
z-Obrk(4taAsxSKgWv-3g3v`eS6N~bNgIz*DwEaS4`+64<cx|UB#2~-&W7rs4LRdD(
zQJd=Wv_tl6aMY_k;FBo36qZX8;Vo^Hk2JshoMr!o$<{<EGoKdvq(OEa`J*r5*m-fh
zx@}*BHZlmpbH6)>MR%5NHVf%W@`JnZQS2+s*Y>I-<NX&MeQDRJaCvJAhrCd%6BHff
z3NqbhcC1bKWRu4)<(%+{v<PqkF;=7*%b)$s=0!GY@MtdAa)cRdh!budoBOaZfD{M7
zHnSttAtP`!Zt^9@q|Iqz#TO)thlCoLko?=r3l5*Vb42w+9J^m%3ho{<KJjv9fOT_8
zEWgO2G<O)Ks1NGDeK4Py*p-Hi+VQ4#^nqjr-Uz0=1BLf6ePc9Ziep|@>^m@1C{h~n
z6cs>pmaCF6YgCB#plr3t^O_L*TRrt<{fCN!QQ<NV9BF)59H<Gd1uX~a@;>3d)ijeE
zk7e|TaCSEWfC{Iat8q98n_GjebLb!Mrh<Df18P53!nLG1nPuZCPf@tu7big)3MLmh
zoSTha&Hb^y>lZX(Zx#^`#oI>KLZ>Yyyx6rIHw4?OsZtjdC)nKMc6B^P&5BrTp%TGr
z;WE1WZ;!6wMj|MJZ*j+3w`N~%1p@AOb+S2LS%`fY7`LfluK6qvrj_~{sHqAcqI`N^
zfL^A5e}y$2u$IC4gpMhL*jGS`j2G)EY0==XPre$`sdOr1uqfD`%s_pjK#^sU^b4*(
z5mNu~`eS1K1&s&({lmQTKmC01dH}Njef|1V%-zN1>%;@nDzZ`*K^j0)4eo1^Qq#N<
za%6~=Q8J!c&JuMfKFw#{RUtl}?B2l5IYl#SB2=b7A+?$t2P0)v%^Z~KaPY>9i=&Gj
z0e4}oRV5!}TJ(<AvxUQ7J)6uwl^110v^ah1Ph0)Edk!!_KED3&7Y<BL8*jB~c<B^#
z9Ev1+EkN@{)l$W{Zp9Atw~I}P)r+KYC<v{wmw!(u>Bjv*`i?P)Je~XlFYkQQnp-n^
z$rN9o{SM05)M<1j;mj2Wr4bttv=R+BZ1tk1BMiIs{TJP2z5wx(=kzo29dEeM=E3&o
z-ADw^DL|a^^!hw(aq(6BF_KG5_8Bl}hy`_MA404P_VIH#Y(!dEde6PWA)H^W4HrpA
z%Y6KT!tlP!$#|L8``q;?qPB2C5hz|B!TV!~XflL7Lces=VR5W+>deo5>(BUq9A#t&
zhmyp3-M!62G&w_6-`7^4$1vE&d2o_DG{9Kc1ZjzzT{t#~E>>yByKsS}m1sW?_Fs#B
zhx51+3ZlPDCGL>^@CmN2b`sy;6TS*HCGFE+!KVBjjdz9X0H^c+==~Ug-e-RTS45}f
zB0FPIPQyWlKVF+x<b9&_y~7SA7S!OGFsqzCtscYIjIVX3DaUzV@T=P#Ajs+0umeJG
zS>BmkQ#@|+%_2tK<G{$A!zVDXua&iOOhf*7yATTS*e6x_RWH2!;h|@bj}BVQb7d-X
ze(>(Syeg`2YiWR82a>+w;fCdjhGj#%Miy5V{ndKgC-gpRf&ENsWz#u%A72BPj;azZ
zGu#D)1Tl-XIJ!kX+2K8S(m~t4Q;fKm6o)V@^T2m`FLLirVv2{O@FC#@$y(iI&Eu#f
zgy*|Rv@$s;V}2KJCl{J_TQ|-h6B^L!ke#OOdgt9ma`y0xL^j=oXbL0L(Z#xguFnJi
zsBNIR3<h+XisiZ*R5Lbg^rd!dAbp7g3eT}Iba^)2!g|US-%KiMA79^|>fi^Gkd31w
zwB$Ra#=c%&Dp&5=t6kKC*xL3r;OW$(P0nJ>#W4>Apk`o#_DF=jCPQlkD;8bNWBH#O
z(`IDvbH15zJtAvN@uB|hEEG1KPYYf`Qp2=r?`*=Q-#uzvdgfa?J^=;dN*AszR-*d&
zNOETVqd|$JI!}RQcbwyQK(aEk$(t67c|_Ck`i6ukaoQpH5R$Ya8OUgfX>#V&T0j16
zwpto#O%>AP%22YjgcPP`iDaCXJBUDw832C@EGO@zBc@D@!5}6TN*sATSKsFl*ioGL
z(}4=KXk@g-PNouo&k5^@c2cQIhAqKN1&~Rey$cfY;2?=iz$Lwm*vCqR+>qI=4$tK-
zn`;822del$E{tiY1h3-7_6_A5L0B2#XALUviMf~?&PaK+<3v^p0Z<gA0TL;a_^~$M
zdusI|5EDv`Bl$uhJp6qD)Nm+^G+j*}J#7A0%Lsi^QI58xiKS?qvaF;WQ9A^6(X$8W
zwP%yW!Q(jE<HaP4ivs@2h;k@Ri!3pyEZPOw_<I-lZaJ=P#<lF|3-ef<Og03S$yaF9
zrlX&^%(d5$#wsBq`969?@dP@(_kII@9h+CPuYTJ<2*TCiZ%8PB7SJSQsKQ*rlz~44
zfT%7VYp2@U<FOM-yjIx~jGDD;of{rvW~Ff#vk7o{QBUMZx0>zw=rXjeQ$N~-l9AhK
zU*x%2brB%wWii+Mw1Ky6NN^sb)h%^S-CMeu$0w#AX_|_QQ+lUgl}LV|FsT`q_S50p
z|Hq+b2X?meuvD#_`yA>wq;kM1=jlUld`m(!8M%`yIyrm0s+ftOUBm#VH-}5yX6}%5
zp{XfK9RHmb&drT|ryn8O!0c}6T^tZR6UJtFo)0*Jd#ZEAHa*^cO9d)?5WV6)ei?IT
zz<q>W9VGQ)7cTiJ)gE#zqICQCYkcG44vxNe3%4ben;=bITGTcQsg1byoqpq;Z^hCv
z=cOqYK3H0GDBn#g&XV8N1E$Za6--9gT`~SuB|p5w+vS1cQ=)HjJ85E@z2>Ldpqg%M
zB#AM!O2XOfZw(>R?e`IR_A-)^07U_qc^{}Wi->NH!C2VEMZU3WkDeUOvP@ni0OuOu
zJNVy!ygSAchqCAEWPM+}kFD#3ZqJsdfy7N}tG;pXHE;-HDYCO)n78BCiL9$?Hd;?k
z<=iL1XRs|5-2a75lcuOY@Q9@`Svj7q8A@-O5@%&4@^p%;vX7b@{6wfw3Kjg-(Mlbz
z+K_K-5_*7Oq{GO=i)dAZncbrGS~ydbs6Ash-1|Fc+XsPPN&c3WYy2%>JO87=G75m`
zR6969<`YlyKt(0S!!Wez9Oe`)bvl!Da?<g?Q4Qd0h^5x8R_}ojdKNAIy!HKM#UMgv
z;uWgF0;>ANn~_b*V=omgO~MscZ{Azp&43n4+|&DW#QZQcIGaHxB|tFU<E@mYs(jB7
zHU&iS_VNJV*tnCjFaqY*cX9Cp(Sml+v2+-J+au>&nOw%%c5YAz^Xc_{95W(NLhS!4
zBde+PmO+qDNbSr0*?uTpu_ZmttX4|(W|@ZvDy$mBk*U2>!JkZH3#!qM2-aqFxVp;g
zX*yXeNZ9kK=3c&CgAX=CPiQ#Cw4;pjT0PcN`lSKS-0DvbO0_}+YmZpAgvE!@|G+NF
z{>GU*woeh%eb9%Wb{NKFV;Jl|u-k`!j&E=}A@XgF2oOk6>#1bdU1t_k)aJ6(tJ>%l
z1lYZ~Q>usy)BwqB{`0%sT=vO)t_6M7MZ|34E2UGIb8ZUg@k%9FVv<K)-jqIj@>O<j
zE+oUX!}#p_I!>JpEoN@Cy_Dyt7D28!deIo_6abyw?_%n*EivwsXk-HX7oWhVunU#q
zL34aM_)teP@A5B8#H!RP89V6RJM75te&+)uBz$0fzGS2EA#0<Xy)Dzwp{o89-s>-Z
zcP(kw)-T?YqF<C^cp`9l-EjU{J)ex-yocwh!N|ISeQ<bbPCDs>W8O;gy9v~=t+XE8
zF@^iQ%{Wk3MMT0xxby6rUd-^JT3>Wy_Rdl)z(LH48bFU9A{r`yI*w62Wl6m!1XcX`
z;N6_QT4%-JHP@zBN4{K70S|~W*9M0x=>9?lG)(Y#6vt<!np2tMoOz~BAVZkVvmnzJ
z=I+I)XdsQmIUD4(mUZlkr$tJm@!=}Q^hj;gK!wXeDDN!%ST%TmHrmPSnPvusQDiTO
z=HLe$QATT<!9timGdmMiDV$dFaE69D3lGXb888iS0TT84^R<8>`>TN4GnzD*xjXBg
zj-P=ubf4tHYEVg=5iG;3iINUjoL{Vfl3SIs-6iD{Mp#+^TbzYfUK24~#}@kHY=B~T
z=qVwlbbFEel5pt}Hi=XtHp3%!ZW1D+Ce~$YL<HEw!W<>?BjmebetqY4kV%#1Y-?6e
ziFrZ<3!&n19~MGd!V)!MQqcyS_5jK&!ny2mkGJd*;avdm`L)Mp<~m34v=upck{pZD
zrK=M@Cz$oDxEB^{ivNd#GJoaa<?X4%7lm8U|66l^K1oLLdvMoe&u?IeBP%TzM&)N!
zlrRgbIF6R)#OBg4O}Ql|+!#$>e3i|;?e<o$w8<0T2kl|lhE`3+VBS3})P;%KT>=CV
zb#36mv>ThK6^1cy`UUnm>TxRZFt!}%kvJbtluI%e&5neL74fiBJSPDA6=fTXQ`J5#
zYk9wW$kO(tzinM?m5S`89*52dYg8WfagK0Fj|5<BvU{wl_iuL=Gtxk5q0pzk){8DI
z#^mFMhqS-2+fdN}_D1W0>vrIVkWzNvrGoWR`k@e+NJ+>DoO3tQ+}><N*KZSyaT9A=
z+{f!f24KtJTbaz6?62;9ikxo<hGJrT8thSbX7sd-3-dRtb@^$_^8H#&lpHVKZzp_O
z4LSEwY2oxxgP*)#FaY+Fa~HS_FsiN`WK=FD+23c_=?dsBT<RaK+uFTCR5_2J27E*~
zx>LDwxz*)$oe4cV%z>JysvDL*B^dZPqHe2R@aIhAqtq~a5A(cch3yv}{wr_cr9UB#
zpx_iYYs%9mwWh$(A-=h5e<+{>14>8{he@nQ-F<%$6~E60JZUDVUH+~2F79N(L(Iu_
z;iTZ$=8=izMXArc4Ocd7ogK)mxjiGtLCH94Zu#AH+9;?CyK$aI0N|B;5izzKc5w$x
z)wopfs}@=8E1L~-RAo*9_*S+w?tu4OjCse}igWDgW;BI_G*sDUJROj2PjQ2D=Wb-*
z)z)W${Bl%F)pBJE1E81O>gwi7II8RZww3V;=CtwG1#%gj{{5#|?^d4{idA4&@(87B
z+n!lv7VyiNT?<c_S(4;pwzi%hMlTrJn|y*po(~q~;CsmPDA7t0)zPf2$1$d<<)oo_
zW+`x-n*l-=o!Kr%P8+GLnB%Ex)rgFQZ~*nExwC_+FaelwrsK1V2Y^&o|NaW$7_VQ0
z)ZmTkpB^lloL$?8O8!@d0X-mCT32+VL^HC^lv<XJu*s^&H}mxT8c?I}t6QmIWkHF+
zo}zH^ZC&DSC*%qJJOTLkG;({5Gm1Zgw4(@mfLT}iI)Drsq-eTk;w|Vy^r;3yx+=R0
zBfpt(gas%&IUvLF4SxO=U^mo2{s9DU`yB7x@{II@t<l?I(Lb2utsGfQpF|Y@Y9XbY
zqu08uy@BAe569-^KlLyAbq3taD8%U~Ay5c3&+}&JezgmhR8lDSa56{aU%<@3m~fY=
zOhUQgsMq7$>zgAHdv-VW3Scd5xwULG>Y_mp>j3EJVu+&K!Om2<7=P(IorZaVWk?*D
zhXP%7ZYgEAF+>^=@2?n8A8$9V#-UnSh-2n_VL@Wt#d-{omoBW$FBl9P$L|`muYYK5
zB^$5cuP)38_Q%@~5m+`xlnU=DK#G6u8Ts1Nl;)LToA2=15518nU1VlgY+^00y7&ce
zfb>+w<O@nh`BN7r?K+8?9A_B117X75q(_vqeE$hQ3%_LllZX<&7yzF=2Om8trSAyV
zeN7az1|1yZW@yd7)qidK{^arb<>!;=%nX@fsTem+@9KNdH5LX)B5C+<+T3dK<ojcN
zJ@QQoTDe~MEZBHs2b%z3rXl{av-vo75$Te_zo$CXPDkoscCK!S_;?wxd3`;2()Bdh
z+B!cSOt5)=**|+1n~EnyCyA1+!gJ`jIp(+-+_CB-HX)6vsFHhokRHn^qnX-tn0V;i
zCE<0|s=Ij}uvgsmh{$V7uq7hJ?~xfZ^(yCi_mySjW7_T0A<H72gna6jP?ddh$&F69
zKwgXwxxJ&_zL7E--P~LEA3^x@Ri&>?bjC76=Ng!|t~<|OU*|?cUz(h8DzH^$shQvr
zEATyvu@>i&1HrS$i}1);Zv;=l&zlI-@+1XJ#mtp9#V_BSm^apghB|dfDqavu7qV@y
z&@|&v6y&8nKV5pYaZMB(2R@8-Jt8a56Zm8*+oHsjcq^fJit_K{G5PC6AC6T%A4k%(
zSf)M5GiLx_O+CTWh>3~o>r}-U+H8azWLt3#HB2I=b3;3~<G5I-qB>fFD(Ys4Bfy!9
zEQV5&{mZ3jnb3BaNX50|!;y6PG>az8rtugfB!RH##b(K9HfP1T0|$vEH5G+ku}rrW
zKBGjeSToWj=sDwV7C)Ps$@g|H4WZPfqw`yODc)^NiJREhJWrCpgRw_u&?GAjC`45w
zM@1!|Q{=+lS>!8+p|TCVS{iIZsse=}k+oXD_}N9J2h<)AXqHLMEhC2-#;w8w2q+tg
z{a8EEofHu@0&X^Z@(IgLy<5jRC82pj>ek*)ZmyU`H<L8AtTGE;c#~Ac{@@RmGI<@5
zztykH<bM;MQO`4eo%y1Jw53%?s(-@1;8c}hJp+HlLmKQKTuTx3WoYE4$EMP4I+3<<
z7%u$Yd!suP9=S6*Se;{o%)f;3OTkW(_)U-$g&50IgRC2O+AD1yUgZelLqBuMl!vFu
z!$Z=|{oEKTw^qa3-TkSZdEFm3!yOzfW1l!f3!nd%X<93-DYF(zPi7EBEEAwsQq{(c
z^B_uC3&hC6)Bh|_J`0d>IqAw?T`QV<tW!lD7Lmz6R^a%%de+1v&lXkE$TkoImX|l;
zQ}fjLJ=4Cb&v>&+N1En{8bQ3-iFGa?-}H;Vw{hxt`1AZ$82@$I20{6NaJbud$@dxN
z11#19Q}IZ<LGI?XMF<#fUh2*n#1VKAcL8t-unDC!clq|t04r{BirnD)*hScv2e<uC
z?EMNALjJJWH!a~%NEITs(2tell_~+J8)0BI4W%pKId=bIpl=MR*p4~^rIOGLEPtk`
z!n2^xs6btin7V#6c;G?5n0#_K8;b+)d?S4GQ0XNw0wq*uitydKo8>`yoHBWNSurNc
z*;5huYHfUyktffEePTZy^fgYe4f%re_)P~6?Y--bc{%yy7kjNAI;Vk?_DAEjrNIk(
z>&L6t6R5D5?;@Sk`=<*Znt{ZK>jWd&Q-z(-hf@f;07r*?s`3XB<AotO;U$w^5Rcl~
zVZp)>GB}P-8Ro-bs8yKSf8|A^jGrqWxHXpkD(J~1m~$W&7HIaq@P?S(M<*e*tRX5k
zYev5^&EYFVvt!ttxTayHp;azZU!Yo6vSADM5u$H(Tob~4tXHz#6Li8VLZk#aN{Yju
zC1(su8idK74PXzA<ouTXC2Z!mv^%tY4=iV+aNP=VGtkqeugSo_F6AshDaOG`FjazQ
zOSRXpbpLX(&~GQ^9^6sOj*7cEyRV{Hj89bw!0j2~L+^^qya@9)(l`rM)%WhNW%Y?>
zT0i|!g0P%ZHg7f)udJ;#2L<-Tc|QScJVL4nh;l}NiRe2Jv=;%f$Zqf80RMtFoS92`
zB~7^N>6w&DnIK_yD!2JLiA7xWB|^_S)8^RMY9t7L<{pa_ktM^FkK&t_P}Y9M!p_WS
zd8WhSeL#KO^1Jgv*}0)&Pv`TM*matp;cQcv!nQlrQ{iGv6)J<Z_rd;t?FaT6w(HMF
zZFeRAi)!K9xpJVFn^U&?)2EG(5;S216?FYWBj5EIsd98ryct5d?j?geaLp~1GWZC2
zl?l|Qyk}5+f=xe|)fe%^%=O}j5(|iT$QnvR=3~uhLax@=aBa#ZkE7L0u&K^<7nc^^
ztq&DpI(7IJ;<q|TRu$;pcOr3V^MAVeDdbh9Ixo+n%Qo*2|Io>{a-Fd{V?apeKLr>|
z^giYZB=zcljRTqiX0V71!pYAMe3#jWzAQJ9188HdE}8A<Lj=?ss>;PYpU@W<b;JVi
z5ft{E$QM`#r20ST-`!nIvCbnr9(e5Kd^6O^N3FzG0JCQO_~0lDlhxCy8tXNhn<$Y2
zu~huhGD76_zr70Z+d3%DZQlR#(5ikMTVKWcVejVU3VGjm%gv8^T|202jf<NKN%~7+
zh2aGmZgcizL`e*>ModY;=>5UwG(SVZ%Q`<@lku2QgHwlUU{$?uNFJX}gVKhm>l^0q
zp}VJ^_It54Qp^$6%b)8PtL<GUYhT~|cMY(U#Oc5Rw3O>wm0|&n+C=Gw?<FC95pj79
z)ZE2i)`K3%%FU`x^QiFl<!}*E8*&};Z1*`u``cjKlv!6Owgv$MQWaYr4eeLizlpq#
zYG;|}GaRN+`-$)2R+1J2iXH6)^yg2CKJofsZA%5X4}socs(Dd3_d;V>Qef|x3yJ$2
zPI}$2bq1K($hInlD+jN=c*9g#zvu64nVB)k67;w^EmJUW=vST&QJ6p{ZE|vRVz=04
zXX?1W(#x~&X!^8@gp-)O+@kqZSzpMo3c2T=jL-IqD5a*p_de+(Jx#WmvMr}A*`Lvq
zpuLoL8g~qppRmGsEkesW!x7kXdw)Obs8k{=sp=%e6?;B*XCsd@xsTrUa?OI*KASig
zf^>CM>W!n9F5L8xsE1Jb&Dc4y=*gUS%O;RqZmP?XaQn$#TU~5zT#k*{_iycolbBy#
z9pX5M;@K4E$B}BJjYHn;Fjav-@{<IgYFqOEg&oY`w{W*VbMKRKNVOtwI^7Zx-e%}K
z>&PqHCe)Wr6Nix^?W?ohaL;;B%k8wzJ$7;~UJa4tQ9&K7t-cP<MGpO$ve3%sH7}Z^
zv|`Xo7(3R!-B1xCL+5J#g6!G$)okJ2_s=7NauvH^zgYsFcJm+H8|G}{3fEZ<+a!{%
zVXt~FbrM5yPsC@CIHwcM5`?2}zFKuKWj=q;;?Iii4^91iQ&Fh7ZBuRhCA(w$Jva_W
zRe9DB*4W|tarBM#!j-bVwmcUQSQM-?sb9{-AJH!k`me4M58bRfI9Qw7NYVAk-o@u>
z4}RNnypq56ooCk7nj87n5X7<N#Z8s0KxH)*h)IQ~(59vu^qlmPQdbqASG{yHs9zo%
z4w;l?LKS^j)Rvy!fLD+7&1@+mZ6V;mJ6J3;itu{M)T>({jXeqILm)VM_DiDDuEuzQ
zaq1t|p9MD?*sar(>}~qj=VyPF`MDhhluN*?9eq2C+uem|@q$K5RHjCaUp_#G$BAn~
zFuUwtpqNz{|H*z6jO3aV!<Z(qr3f%lMx4kT%if)QS}rq3rqPEbPAx?y8N^JADmUV|
zh+_qtBpgpP9vQB&IbJ8eQ~B%M68VKwPNU*4Rm;oSn<n@xQ9Bky^8OsJOPEH|SDC1z
zBb4ufr}Wr&d#T=Nv)8^cSC78Mi5uEnUm{9kZ`Km=EoqS3@0Sg!5~zUMMJA#|WryzU
z`d$_#)Hn0?adyAzqo}C(_ioX(nw<jmXyHW3u?$%KxH4MSY-7C>iIzk;f}c`pE{BF$
zrVWunGY5+h-Tx8wl~GZ?U$lUrgh;p2-Q78aFm%k&Al==qNQZQH!%)%<(j5W=!VrSM
z2uPRoouB`E*IkQm^NII;-t(NZ_da`bk!o(aXMYztU)U$9bE&UHE!WpqMjkjOB#xnK
zVhm+vldMp{J9=dCI?rbCEIZNUaR^hl|KT%9RjO<6V)qJWbLaK=g|%CMbj&&TKK3b?
zA_%832rT)RS(QlqzYFsP)%NCF<hP)5pO9-GEF}=KAt^f(I~m8W1VyD9B0-Rc#Q7%t
z@NwcTEzL|SHK9lNumN-1i8r;!e8G;az<^-5)?GkbLz}BHOK+4^odJ!w@3&?r$+zwy
zGM+aoEaA1kBJly0dutM7H@ZL5N@S>`k{#(HbEFKhj^2E3d=y+wQOgzAtx@a}jVBbt
zOF#u;Pv58!4WmPzf-k@h4$jL9ZMeT)qs<3diROKEY)Nka%8qrLz$G5?G}T#ykk@2U
z_d|zRcvbn_@)hkER0u>!OiWGnwcVv8njoiJMS&T&tq)Rc%(9yX7hgE1HRY%LE9VX1
ziK&W&F@SbCM(p4(C((t$zwl&-#9Ywr^fCIyQkg>|G_c1GeYuNYmNt8IPJel&x@3yE
zoi=lYPeg&<-F)rVnACUIYcjr$nQ{023G)~JO1%C1ryq;%&vz36(9A6{`g^C>gRr5=
z&&dC(gV0$8&Bu`Yf&-VZcuG@VT(??zR3^6$PR;`n!^y@(WyX=LTTgSP@a8p<5K*)8
znsJyH6y-Y*g>Bc?dfGUxx;n8LZtqYF$!-CJw;Eijq3sdnQf50#Fdc{)QkieW($Fm*
zi5mS%xRS30s`Bg~&5-L?4$x69Q6hpQi}*cI82R~Z?`$agZCT-?I_|Tm0l(Yp=4)qb
z#$Cn1we~SL4jZhWBt~Zv7O;!^m10MY`n5)Y3Pcl~S}`Pka&Fo?>azvn>|$N@>lbkK
z9s8t`;?is!Mfb@Jb8!4b537FBy}_NV4D*J9>iX(vOh`HS6^-@I-R#H8{6V<WZmI~X
zJI_)w4Fzp@Utlop=J6E%-#dg|<{p8}2jd(~-#lGMiI5zti62dVbZhbI*dEG5vR#kl
z1La6y`+C(g*zt$lhg-?XZ};~-)1TeNkJTe?oaj$qkWwZ09Z4$R)pv0XaIe2P{@1o|
zjuu291*^yUSGY?&uycq!o6(s}X9m7oP_zphs5BO~xAirSUKls5%Sv()+79$K*T9rd
zE6#t;a1aJJ`yyjd%jjd{UhV2Uc!X+BLGz^rf{0(R<l`)1N#%N8%0j|kVy0lm(~$KP
z$-b|GC<#DhVp7oSmk`aZWnq+g3%P(|q2(+An?zuT=J?1fA-B<5&%K}CLMDYWOl?cV
zPuHKyrN`S@DdL6h8@eDrGH$ao+Pn5@f6|e3!%W>|cMD5X=U5Uw%ws+^*U3c_PO4<A
zz-4jUjA&djQ>dOVFuF1%>m0*#`QR*tjJ9z`rlx)Yehjsg$!Vgar39w6=>l1+V+nCd
z$}}-Pg)EsQM5;_Tjj;oE`B9L{4r3LaIp_y{GyO4<6py`0D!|wXN=;ihX^V;Jp}hQv
z!glWGrMD8@7Vqmpyw&eC*GTaFbV?8L>?$|O`qJkg+1olI(;&<^mTo$*-u@L^)@@n6
zYuM)DT^z^?YLDK@gctm^q%&Ic;^E0O?Z3hHZ^M6Qesgd%WaOr(!rI{e`@O%JG}dgi
z?gNBXgu#8HG&Smb0fC2>?QsEM&Ti7oBltb8i-@j&rg|GUkJFhG*1R><LbVs{P!PAD
zvhw1_6IxsUZLl?x??acomtj=AOowA^5R*@fLF)DOb&4RlaAi^pQb`0_JTM}gYrBne
z{sZrNTJZ?Mbi}XJ-stI8O9}@GZX7Y-6(;-Ay$aR_FAiVm-&tj~z^7e4)8NPWZhCjE
zr9+Ag5!}73?m9)H`9>l~aC!#Yu7F!mS0ksg&|$0An#(M;ZP!f15};Uzx5|<+PjMv^
zK6bGXUICx&=E!OxEeP(#EkVqueb2Mc<03N;mka9M!H)xz%Osl!4ToJ>VhjnZlJ;75
zHc2jeB#XPG=@^S_)yVczR3S+zw^2V>!1*&^E$~86zAbseU>|lkW!QV@tIa(#de`R@
zF*WExcY7~+*7US>MDbl{X_L##zlY-e`m82W;D2>|bF>cw<?r$w3Ga0{HpRgNw1#~#
zz6dd=jr=b9nX1S_P*Qv#Pd_be5KGIm6qU&}R7AeNq)4S1#v4{rV-Om)@nYHDp_6#4
zQT4H@Hq)G<fgrcF$3=c{LrM%+)&Z-2@oHfV)L>74P~sHAT=%+ETF&3z%H{(o%ik!`
zsVztfOz{kb`jhZ9Ken$n5a+gYDVOVAS=XF^!aQ=@H;=n}x(hF+_T5Kxl#<6L-?2CC
zOa<nTvs~O(6lB5)z;Q4at~D2-6FIHH*&<e(jrx`MNih`RHfP+8mF*mp{4WkzUX)Rr
zZRf{Hnbc)6Zv}3P$PJ7Se&<?d%b1<rVP;RyREau<s-h>cuPD{ubx(k$!*J=8y@~iE
z5K>)3xO{rG`>VzLDCk7xM!Syt`iZsyvzHEv{Jo98TE5d$a6*5uOl!0DN<IbCr#~Cd
zx^sH##~xHJMJeANZ4Y>YWEuW<S$^vyK(Sp9mfhgcpo>ie>SUvoK}@bF=-OcMUi#}K
zZ_Kh$7T$Jj>Zb3`5TYQPB~#8_4j|z%x*1P-ptT)SQAee5<gNhhg9K{*T+8a~V2Ou+
zH{LlshAhZP&Z$RSXhp+DknhBRJhK)}N4}y#{6Tu6LHgJ-c+-`ivgTAAE(b53ur(8F
zeG5}Qtf<*j#0H(-YAH1%XART{MAkwY2`%Qw5M!-cLB2?k|MqRgZ1>aCD+UIjn%>zP
zau$Z;8#KK7mwlTj0vTU|<K3~d+TgDT7h4ajCjzr>_b+C<dySN;PH%4!U&Q@qSR-#_
zR%(0zP~|4AKuYz;v{Jf~IY*XnwnWdjz^8W8_C{hVixG5k9!y#wk8FuMBDj@pCl%x4
z&){hMawikLD-_4qnv}gXVp&r!t=l!3*zN%dZpm*J0yXYdTd%u*IO|zHShewynN(&F
z)NUrJIX1Wi?{&@oZ;XHUw2u1?d?ad)<<j#Wh+#}@-tH>ucUVU;(8xI7DNKX8eboBe
zGrJ#C6Ah~8agh5|$!v{*))>{tD8N?zvH6%@wzA1)F5$T~Nvy%1=$jKy?XycU?7F+#
z^l(VE5Z2iXQokE8o*GMV>a9_5s>+Xa+@_>O#YV@#tIgYW6#CLfMn%ad=5>13uG;;(
z5POWMp|)6mTDHP`yU6}iTbs~c1t04GRo^qF<HV7tD35xR&oVBJT1o&4c;3d^<N?pX
zU&i}B<L6NQVt#i)$Zgie0>RK2Y?%U|7Cu|hz1du4h*9f<xq+=_)6W{M%EG6f67Vtx
zA%Wmw9>&K)D+dmJmPWw_a7HgI5Wp|85Q-Uf7NXhjgeejoe6YH*6)?4>;GD4k?;c&u
zw)>=Q%Gj>$L>k)r%D+(co7kladF`J2AC!z!gO9TcbZwyoDF#M^vVm<x7|4df{nkmr
z{~#{Okv<BPAMqAc`96vnou$N_sk!<Y;kblQz6)^z0QJU+k&DlKr$)ALMUq#CriRd3
zp8C-3YZ-_B&sPi;2Kd_D2z&Ju*ua$9Z3-{v*P!MP3UfVQz6U#xFdbDIiv-a3IZn$T
z5KwNzz{f^|SwTjBGw;ovC_^1#Sp6wyEv&KfXBIUtC@v;P27}s}DaI+hAANHl-89NN
zy;P67&A?9Xz`6=3IzBE!FY?kayo*3CtA}8UOX@i^pVn%62zO4zHKYsgVo2N9yKJ1g
zV#vKOAkCFW%V3-#iQ~n<?d`w?0_YGo#svX#6EpQV+-?K+@_Chk{2r%9z!plBg<7~<
zw+gfKatTi|J3)!7pa1nn-wg#~=$}s(&QqUKw<)(nWCok`_2Q~%6jxv^zC}Gz!?#Bh
z>Kc1E7LV0QTaiE>bpSzxtQu5fkA(1WHk+M}kFs*de~E3XM&RT-q}T<#*NzXs8|yN5
z65}UpB<O8||GOE#pPhsvp-?3ky7bLbDYLAd_#km+#HL7t<Wdnhl<{$4Z@a<y+^JSv
zx3wB$*uEU0m&2Rv=X&Ts6k|eE{a?ozU6(v4x~lHb>V441US=uILYR4%(9lMd3C>6C
z(Su2dPl)YSbFvWKX6xa#GA#B@0HziQY7hh|)9|k&g<4y7W1iGUk|AtXGl{1R^7D#1
z!_&m<HA6x38QL_J$;!p4Dp7dK;3HEMhlxzq8(4hX;lMinh2368K?rR|=+viu-m`S@
zk^sVNGjSz8O`9`}8KFw!#pB{*DKjC}WcTPX(O*j24`fvPelJgAk&SZk)-R|B8IVic
z764eRb#RBsGs)auW)z4NaLR!!y$rKj2=ns`vbYfEj;}ynQR8omYu`G06COPJVi(Zg
zZc_gNYHurSR(q%S`Afwdyp%tDR;myZM(Xpro8JSjFz@5oJyT-_%sz7YaP6`gOhBeH
zN;7+6=G~pc^{cxflx<eEKQ&=|7o7Oe$ENzPyd-=MkwSHq=EDsR(>H7Sl|mUYcwEW6
zrG&xL4k18cVVL4g567kbq{x|-8<->|g<+Oo)p37GzGKc@M*N0Zv8ZM0faoiSoNi=?
z1Gdp_ThRMBoZn7CE_8kgTxk<KkYH`uepz{UrWbdo`nuXeE#FBFm)`5!Gwa*!9F-g#
zy3r3dfOz$%UY3pQ%C2AOZ}6`##Shf@B7CR>W(5b{GBcQ=w51~K{Pa_=o<<~UEjbYB
z<s(S)`nB9<DsV(GV7sxh%4P&q7VYF$u!Jco-H6)oQ!^si|EMZz9{dV2O}etFPV2h1
za0EPkaS?9!rsj5pzaSCmLgaM!DotDGxW2x0SuCAb^TmI_?<CyD@=e&@Nd_;Xu5D~=
z;1ZP!%j&4~$J|ptonLV690EG)NKM4u*;ZggOG7W%hGl2=(pfnp3=3WvU>PfTz~6ti
zIg+O_L~m!Qqbqt_Zn<39rgxTJ0($}dm3x#CG{-6y@^){?wqoV{Jq5Iyie-2$=S-~W
z{6CmI*z=)_{v(ypCl}0IC|dGNuKNu84Fv?T`%ODEcFV6VDdEwx&OB)l^qsdjKdNu-
z<in}R%Ve_hA}$@Z``lha<Z3f$y2eS^=&vTQct|b(UWH*I=g5!opB5XNViO<|CnutU
ztB$&`%b3||cHxRYm`h)E(#h>Fa>YEHD|&xG^iJL+9cmclTh+gwDN*Q%cM=!+IYbZd
z#!CWPddvi`9l1yAbP;s|Z*JZ94v*rS8R?N^c(j*tqfzAJNSOfo`WytF*I-w5zKVsN
za53$j{RM)_PG8OU0`8Ov7~KNYU?p5+N5Jw_BRzt0b<>jT>DgG=9olu@=6}zWr`5h`
z`uFlSYtqWxm^k?PN_YNoL}VuLX>vAqhG~EWzx<oKsh?kvY^+;A5M4b+4Q_=Z#=!pm
zK#)G(PpOAp)2{+0Qjb&1&zto%E@XJR_*ddCLTG!-`YrfUkYf8f1%nw<u#0!)g&Cb$
z0<9rS{AC)1yHEF-bnk}=M~x8iykhYYqjJ5Tvq!mP>Ne4BzV1C`kZ$caNb=f4e*n&Z
zV9Uo}n+(VgoTcGW5g*d}I~jkz{HBkJowVS`n04Qt){L<F5-29uKNdP?WnX673WQ83
zfmRmL7vIpg7%X9^8uL3mQI+Z7eB*uNLMOTY(pug2eJIqk(fl!b#KVWEw`8KcA7RmT
z3wZi2VlVbPbf=OMTxYXAvQ*|<@tSRz<OZUJ&C}PbzfvV)VrynkX?pzGMh{BSae0xc
zrH~m-&H`3s!Lzm!E$FBa0<(d7uyS`)Is@w9E`3iE>U|G@+%+U|1*?2o8l7D{y%d7^
z!d1i0D$#vg`x-J6bi!(L(o%woKj#D@OgiCx54!|60oM8ieNXjLV|KfQOizQe&(XKM
zc!N9P@5X%t0%SFhm)Z8ceayksa`GS%<C!RQ1bT=t+Y({ueY$@2yl?tMK!Xa@VP`J;
z+iu&FV<ns+Ha_irF<~I1!JskE8tF~zsr`y-pm!5}xD~5?1$Uen(CY0*H41`%S+u*b
zh4WS>{=&x7akp>HvG@-|cl~Mc;ERfToh=W-sv$jWq~4x@u_%-ayNOUB@7|pQW}<OJ
zQ<y<oChePH6)gU~UqZ9kl*~wa$qn40b1pc@B!}a7-#AYoXtR#g@}DzDq-QtKq5`6v
znMR>=rt!P!S`ZiWcNWH}wb%jZH!ZSfZ>4bwu~EQ@g{gl`g27$6%}3I=uJ5jM>i}Sy
zRDj*qF?p{`IF?oc!4)Ta6)6j%NTgi6TUdQBg1K-TwmaZ%%Ja!DCeuXk*vGlPp55mL
zT|93#ca|w%N{VlmFY4<Wi92W5?{Zeg!K>BeiGy9jP*h!o7ZRQ|{YtKtmaT+(%VK@c
zPhzwqvr<8O7Z*0=+Vs1CiLB)eA$F2I+}p=CQ{$^@$uQJ2qd`*YLD*z$TeTS3sykCt
z5vrC6_cSibO|F@{{3>p!bJ22bOFk6r*E)Gm(NN+g5MFdu*Sa21lhQ4rKmmWh&G^`K
z!Z`BPD=ocEG|bN@%tM^9^<)WAj1W^RXD}{`@Yh1HYr}B$U8p%j%2Jw@J~|Kn_7H!U
z>MMqeYkwv<DvC$k+I+lK2f?urAWBTrLT5rcpaC^r=QSNnE>Y$&+39RZaL}?f^6qU<
zm!hd?S@_JxJ(?g$d04g1gN%4qkl3<XnP*dm50DO-x~wzW9%Zr*%GFX1x3b~WPVgl3
zoFe*Gym4HOO$}cQpl*Z95rrmBPiMkO8@hen6c7=M7vlGRYGl<zo=X_RN;BE`E~cjK
zy}kJ6^!3jJT+inu=5eUls2C<uTH+kvVqiBzq=~q9r}u=(M}Zv$g=VZ_8cH<-yZdtE
z1yP!}m$!VHFIM6XuOePr<=Y&)==MHbiy>@PxB*ePsTFXz!LAX4V4|x)5L9QLegOH~
z{adCw7hS~P-Xx#tF_sFG7O(l~t>%x;Qpd-N{jF;q7Zl)l+Mk2+`G%@cn9%P}M%b9Y
zOu0ILC>l`>y~()~-eU%0&m!W)Pn1hWoYKzp{uLSosw;#3lafOuXooRvBhh0kxQh-l
zWrKL~6=S=%2#zS7!7sFfx#D&xsKc$#)KAp3o0&GO&egXltU0qTrKC6P4UI5NLGVk(
zNMha^wAEhhTqw|U)b8G+ce9Ru$+)sJR6jY%gG^HFD__7`UF3;Qcy6_S)qJe30Zkn6
zC~DqvXf)Uu=blBN^mpw#<lXoSnd^7Vf*6zL2=d{}CnY+}8mlwc)8wdyq22xHRNZ5{
zDoCxXXdb%qsfH5^`;D0C)87|-7THXDetUr;bTSxViHn%|bN=OSUlaNd^0=vX^Tut&
zY#eVX{n5R?5bU~ddBzN_cu+t;NX{bI)uEr_)9l^;H%WHpl6eRX_yYeKtiBIF78f&@
z{`_HKHr1$o-6hcfZs;n3MD7Kdg;KQZn@k2EL&&;&y@_I4dh*X<wvL^i->s_oYTiCV
zi~@qN*o={{oJUShBiOvWf1F=R7RS_%$=Sn+bgN+x>x*}<Zj(6bzph&b%5nbh1w>Ir
zAbF?N197_;Dr-qvO5PjjIt!ht;XuN0(HDkp_e0MV48Xjf7WM(01ReBR28rAv7^U7k
zb9+5EG9O#XWD<aiX=!)AEsaJUHEy$~m1C8`pwh^ZQkOcsS;RX4qq&EsIvh5E=U_9D
zP65M$@=cN3cfpWG7A-6*Ojc*5W#YXjXV}QsG>NV&MV4WShea`i=C2MF2{amv;ko)+
z5eq}51Pol%3UM3D9Xp{T(wVfg87t8<{_tSawab-mq8#Fr=ZE`~KJ6OrS%QkmaAEm2
zvn(661(i%lfeFWsCbi8VujEWB+u4Q3&!ZO(&L3EIjBvvWq0VKZD5~0WyDFU=ir2+N
z9q8~2vKRB6{_5K>KpayjGmBNILMS?yX)yd^fr)U_l#?ur@rFFE=w+ZG^XeUx*Whp4
z-&<CJ52lYuql(909$FlJjORJ^;rRL8XYo7<Iy*dD>)r)E{Vrtw0(-0*UhwsiZBZ9R
z|M_3R2_%t1MSdHd#1B}sD5E&DJ5=Q164XnFcDY4SlV{E{5&mKb8V1tr%8BPG-iYl7
z+jOGSF8s8rSe%lUGpK}$o}|!pFA)(X9niH)=tdqFP~p@>nzJb#YU)T8pMS;lhw)b=
z@F*<FcdzStyYyH;Zs$;<+iUx+ijZTXq3YVj=rbkd51J-!TQ<eX&8QD(*d0cU$$Qi`
z4~zM~9%V<zCyf1sPcS(R>TY^De=vGz7Y~@pYO!pz3Y9A-70H{Os>pIQk^f_?(I~YC
zP-0V_S?dj^S&muo%SQZYGy|{09U;ZVmea&>W0T%5pP9zig4zd{CQp{z#xu!)=7ebo
zxXHSEc!mjXyoyQzogAo|`2%rf0L;_N1sep4e~S>hU?J=TvbyQhtz>kJA9Fa@fC_3!
zYeUNi+AA!2I7j4G=pHV&7^^PH*SOxZZ{q6L`MG-q8I!)8IBadxEW%RH8^$W92c9I}
z9Xs`^vL8C$Dw_Nc#}RwLMRhe5ZV1@jYKuQjHBy|Mz(-;HWCuP8<B>-!Q6e_3m%6z}
z712L6Z%CMabuL?^7pa1xs*0p^8XPxwzHiy4={`L1U=<PUIFJnPF)yn6q(kJQ66F$;
z!<9%QCzomz=U8&F>3sNj!&3&K&|gkc3R>rINB!;}2o}Y|b+{I~bZa)P!DGG(O-h~&
zs2xio;Q=je8w*OR+nDE$i)gzm*uw<VR`+)~8c?_RXNOqu0dlh1{lM8IPu_wKE}!#*
zQu~f|zn~z7kuiSN_1u*ZNuJSY(ic3eZd&}S>3sXIKqWte2OG-6h|hQ+TJDUs(e<V#
zNbMUzn>AS?Dq*iPQhXkwXtW6V6*$!i^omNS5aPwuKg+yaY&<hHH{qn#OeHAH8}oJ3
ze-2tmON~?fdxn*|xx;*XHpvQ1N1i4Zt*6_BGd#_ggB>!o?y!fvJuG?(%V!0U{5E$h
z5@sQ^oUog<(Tg6<cZgQz+b1f-%$qOUW{68UzWEtwtD(iXu8YbLg<JlcXdr{&rn_;C
zz@ZcG*HEXl2_KF1fcPyyx0`uDu3@{*lGT8ycmC*u_d(oW+G-l#@bZqEd{|2}5Eic(
zEsdllK|gpS()rbM82g1=4k1R&%M;6_lzyGTMu|W0XW)v#yukTE3yM~nYKU4+V@cfy
zpmDQ2;j}Q>+pFPMV3U}XZ16*}mU%WKcZr??|Kh|CX$&?l<!wTICeRr2q-1*GglD#h
zF+w;Y)f>3=<j$l{hK*q_Z#PqwA5hPaE#kpBQDYO+3CUnjSsG|G(%*1a)Zu@*&5!t3
z#8nZBB?Oir!A~w*vu??Xj1x*E<ZY5cM7T}ZvANDK3ZHuP_YlivAW#t=Mo(_g3809S
zZk;;1R!EGVWGK9l19G8?xET{%>=U>s%Ty@71b$1wey`=SvCR<`nW<DwGI&W#SMbb<
zzLfKk@^kGl%~ERjlUJ38#i7tzuia5l$!)u$S4EO$){P(RFf{hc$whW^6W>$YF6o_F
z3-ALK4Eujrn*^fi(aR{IVTTgF7`@l4{yCvxLcj`qb0DXfhb}2$%zPW3ON`4D1+F%L
zimz-GLL5}{T*m>gYMuJ-&7fNQ1u!}MqLxK_%=cUawi14a{X{8sBr#5VUD7x_4(Omq
z`|@7N%-UCqY|;%*nk1}eDU{yCC0rY2M->7+0<v-r$SMpCI5w*yrSLK(Py~>r`EwJH
z+d>0~&W*Jku*JY+sii}^IKx&T6~dUm{zpW7y5?r6XH8KVgx+#2u?q=ps4`(JL|$a;
zOtDI~Z0u;6&c>Sbhu?Y2%SUU|mlgaiGkCF^+mX{otP}2c4)`L>QXVQdi)D1e)^>;u
zvcgP)qQ-DXX4!_OAF$Z9d|*P&S7X8qzdxk#6ZOdE>ej1#XN&TC=*Zt9xbrKd)5dG}
z5&{tWGhg@KfX1fpyDe&UxHDQOw4JD+frGB_?mQmVnV;muqo!(OJ6u9_KV{4GF&&$4
zTc(MEY2vtR6<C$sGyRm(#eDw1juTf39obSl$MQ9S$Oae96T9zU_KLLTN_ge4H2Nn5
z4c}SU-D9F;#}Kv>t$m};@=+Hx!NgtG)1jbWCyhTtQ4nAy%dDAw$9M^xal;6F`?z#a
z+t?LZ3eZho!z|z3_qwWo_sUbmiVdl%^s7t#?X{BbseyB0`n2xXcyE#oieN4<de~9B
zLvVJK_Ff?}Mnw2eOJ*B2&8!l0eKGxsDsE%z`Ga%`QM6J^Zn{O+ju>#+%;2M1c8?gA
z0RLND9ab8`?rKHiC8JoTbW~+)a6REHOG=ZEuF@pXpgZ9d0egDX9&mId9!E<{t12GG
zSfjV#M1vVOV%b7@^VD+@rWZi#UbR%m(t2IYuX}N7BpbvhVs8QTk=~zw%ITCP0bzT#
zXWN9pGm~ZXxW~`1?qTS`>ikH>0Jl^dhEQtW;^GUeLVO`vkJkDvZvwtx0!pvjv+=Ew
zB2e4S(@2-<<;2#1jlJ-x(*rf?SNz6RI-=h(a`lx~7d0vCOjsg=jhwcp@}t$ZX_LDE
zweWi<87J+@8oX#Qv>FrKn#=SfTuKZwbe&XHcV$1_$}Bd$8tjh?ER4)-i*Wswd>HH@
z!c}TCHas_nmhaRdz?5Sx;8_u)e>`f_rDw>2nWTV3)a;p%srugXJ@<RJ3brHxF7B{7
zXqkq@77$%eacy}B{{{Ivo$K_-l|TG^Kl)VMVsc_P!$!?vYqGx2EG<z0kmOOfQ-QZC
z%5?B!==ZY9Yr-20@qM{E&`>fad)B_Ex#;^@8<lYJg7%HNhbP#H(PrJluGB*~?z#ha
z1R3Mj)ub-F4Sx=WGJHPZ0(hDJ*)0D%sRSY0#BIiG4Wg3{>e13syxQUD;Fij(o?nfL
zKjdkyrlUp!szF^>qYEQ6TmYy%XUF_schSKE?W%%hN)xI*??ByP=+Q722z!%=er4=_
z)JE##)&9I^pS7yxujw^z>>lcS>QDckyTpSqs_P9RRH=o7phQ&Q;PVHtqMem~Dz3!h
zt8f-IaVd$U6nH(^1b-Swx8qBQGqfRvIr|IvX2Uf36Y4xkSN`GR=z`$)FYsS;S#3v+
z>nJN2i62qrhp_O0EOkgwqkgB!r2x_Zgb{wIoVL))u92TfUU8Z4Q^`{C7jV`oN5Yr7
z{JX;r!q5nlcWYzKBj_HC#wXona_A;Ih;27_Q(-AnglALvLbPvwd!Nbe!=%kda6ms<
zO4CYi%8ekne}cHtk*Sw0Yj5-|`94@4K(F*oYwPNUs}da!k7fJP?_{g;Iw+c5#O&K+
zYa&AXs*yhRc%;BjQ*663<HEW!SkB{mdrQDNakxwp`ZE8Q_Ot{CJGyAWt-oQSF90sx
zI<MrXegC?1p!%hs@wlCrmWlnrNBx95@5p>76lYY|-=Ll#-PZT@QEDcioR8xAeJL7q
zN~mUDc%S>;o2(vpsFwX9+Ko?4ze7vHY5O0p|JMwi+S5<TK+hA)w>7QA?h{?Z!3m3k
z!ExzIb+88y;Nz|5w!0ESWIG6H8C&Y|)<za<qqv@5Dfn@1gk3W4tGEDOExt*r`A(l?
z#@6etE8lptT{~*DOHjpe6w^@~bEx1IDCoSQRJcUhvG_D>-O0l9xIPcI=JM6Nphf4P
zG)Lo#vpZ>h^z)ufCU>p@CId;^8xXjp^-V(;9H5d_AwtX3xHq)SW$MSNu7cmO3665@
z%xwtResaH3qvKdqk-?xSYCRJDL~)BAPa%TQjDjDA%a}}rSd$M%LSPLl$R)TII|!NN
z2HYp65_5T(f~zlxhE|1HE3R_2+SHKJ@^DwIK;7i8M}P$;0m698t*e$Dtcc$qMkJia
zGz)TkXq`p&Czq*Aso2?(SAsTlRr}^G;!rtv)@}_o&Y-1Of0iw%e$3vTpN)*$tavrw
z#^)80cOb5IlwfDhCT+jT0h`4Go^1XjU0-ed8m0YM%jKlqpdbj{37@bfk3T>|*~Wr*
zvLiT(!WJw2(;4iW@YFaX{9)dpqXry8(U%a$oFoM%UniM&C=SyL>JDXAC+C_(gT}X3
z5^GLpE(#-pf7bE%Y3bYK>{w&5E%^IuK7i4SDIcrpsl!_3x7-$t4Qs>sz`rfEHJHmB
z0CcdoP7XDxoi0bs`a?v*0O>u&j(Q_)+DiBNk5z76oKqKa?W}_LKeQ%PGMXZp>NnJ!
zUz<TdyrPVDs|o|k+7b??#*=6|{tdHNM?k?}h%$6hVZ;BD9vwazedrGN@4G;-P#Kfq
z_0JbR<#vq4Vg5w`jT+*d!U4osm`iKrW!eW<FVNdWkR>ML?TPyQ?#3how_#|^S{E{}
zkR7}8SV#eGXkBn}XVeDu<!sTfIW<9&8sp~uF>kc1{ISv-yAUx=>1x+^@feQK-aqhz
ze8R^2pt+|5iP;Z43*qLqqanE8{=qZE!PF?|c|@A46Yq7`K71`ZEp1UP`T;KVGS&pC
z!JB;azN(;qs;FQxF{vwUPMn_*&l3#oAR(+0t4>gBY|R~#iT+bSM%<0_#;#5JrUDiM
zbGvbxyf(qcl<(S|>Xqz3I|-zi;e)k~4Bt`IIe6g_^Wfwx*=sE;v_L|v?rWT`xdB*A
z#K0>e$E{WKP_%FsbCHy#DOh8%_6|N1J5gP|5O7@lahG7<-^O2*Km~)RnCJW|f$L|9
zE6NY9rdNAH6z{kKnVldXf>zt{BdWcu5wrjC;ef|XZa&vB1u5M{;5I^bcfrp$toXXf
zB>4}8R*^>HPbO7!r0_{N6A%$e;mFY)6jVaG)+CcT&MYTW{?qs{o4`jRBc2*cT2HX@
zrz7@A1dpyQw~rY34(`ANy5D@9cM%fexPiL61>{i=UM$tuX5C{OoDlTvUg?KyczfIq
z-fr|})c&qy`*{g;mmvJ#lV%L&MnM-d7pNpLmZXsVbe<oT_U9b{5Z9vFLu;<$VVO<0
z44cR%HXj!-onTXKQ*B0Dp;{kmRdSZaS#)lr;iog&w|GYr)PI>VP0(`(=}hY9SVK7C
z3rK@(TcdJ$BtPUBlw5VEX(b-u*{VVVg!4+SUh(4$5$CfTnvH$Ul-hfh4ngr3sLNr^
z!30P<hGsT6vO8?4SucIFlhj-iyE#NTAJ|>UVo^D+5}GFD%jcOtQtPdK?7R>56vLUX
zOM<ptemii!)8lDNl^Lr#Js;IYDA*CUoM#WWS=yPtJt4b@i!mRL#l!4pe0Zxt&wO*A
zw=eDy+=;8k<UUce9L}v;3Y&l}>z1Nd5%R|D4<++%`7xrc?17H%Xso*&G!DeaCAb!4
z+c{v`{`N!1`R4aZ;T#z18$B-7tAZyLMtxM)8%|bq{@_?S)=zbYjiNpj$UAPWQutEW
zD80`n1!I@J^ZyU#{fLTNiPiaR`9YQA2XTMi4iEFEP`F*K18sz~Ms1I0+AY6_DgI$6
zcNtM0M2V7v{$R1%Fwk>ge<Id1(x@<*0T0cc%0U|x7{yz&s8;H4wS(*Iz`9hqkfr&!
z0u=hfX)WQo?odL(@|$Mx4#?8NtIyKDJ|-M-M^TFvwiWDY4NS5pu+8^j*7AP)^$ERQ
zQZJ?`ZE<h=7nD>(`j?y%nsU~Cb{0KP0TUp%A-UaD!jj61#zbvLfCe3*3}c34)%QL3
z`*(e|Wy%}|N+h)8nRLD3_+ygL2Xbr^R}koiw4)zRrqwGKvqvNPLAY?r2}Z1Jju<k5
zMQv~dul%;3b+1w3w;+9%dY@w#pY!A~eP;5HV-i;E=<@4xVpq3j`Rl^GOe=LAnzFyI
z#7WqWcYip!8ki;j#g1-02_+e9Rujtmp{$C>JbWQ1OTBn3T+8K-gS^bc-_mh|Ab7hs
zn&W8ppTv)<Epc9h1iiX;?KQM1OaEyfUrCPOC?dVmclg@@YPu%<(GC<+fTnhD`{}R*
zZ0dpJhY)FUTzoEVX|SXC3gM>k+$v1d6&6?HrQd-~5H*e|`yASNqVc?Q9tcBg`02~t
z`NDvT7Px{~HYYK0HDuQU>9~iHzUyMS<9j*(8&bB!mWD{$R0S2CL==-kzBnpfB#*P+
zfhbS&2v#$9t@KzkE1Gq_WdWY5)4*)paL>MrYxfu0<^F*@sOk3ufmtFFQca~hV4-l}
z6GYz~G+7*R=dll06}Z&a&aqCM9wmcEX)xam9p0fW4Je^Fn26+A@A;XtjL(8Rb8;4L
zcNk|JX!k-ZeSRyph41zPh|{$a7YGpSt^t-D@N|YsHJ1=%zz}7cV%^M<suimbaaRTD
zma%@Mm9CSg5{Lu-a0Zt`qRLF@K`V+KBDb;HppmMg4H@Vm>&5Y&<^3B^l=c8`+=%Mp
zj_C#<>)}mnP<S4J0n?ZN9dn6bX%sy(r`?wo9J|Kz?MO?5iqKS2(|*L)<^TXihmr(W
zBK8#9K}6hTa+&5hXU<#e<s0K)zhK`}va=hQY^c%jSzK^Ac0<|t+AQxCP>hV5S1|`-
zoO=B2So7fG#1tf}k*6@`IH2LT!w!n;WvR7R<B()>0MS`7`4uryi>>q>?b8#lKEo$K
zE+Z2oq=i}@{o;?Yx}3g~nK8I`gG5LJY;Wbdb#xVB<RbojIfYnZ!RIBN%HS)2g4TDv
znbuD>V%rJZtet-~GS+L!#s=quy5Bby#Mm3;#c5E3LwaBt8LLbkiITOsZ~}ymU_czN
zd<>zOj6y$K;M+|g;oiOT-@V!FYnLiyb};$AkhN@Og|=-6u=r3LrXt|`qp1w(^|mgP
zVFWxU0^>UoTWrdc)|~%i%oeZ!p*EY}F_sFxn3GZj-8@V<#BIFfT0{z2Ie%yk!nsX3
z;q(F>uliVKI<A+vJ#*~8Z=3Fuv^&6wMDhKn3>(0aMY_pO$9^vC*zF9RGdG06vp40j
z?mIATYhrZ^A_{l9@dv_Q?s{f-Z17FG|0+$PAiZ&Nf|yr-X>oHqEZVBY@6hUie48t{
zLULqiY~=`Za!?T16}Vyk6Mg3NJ%VG3okK{a_24@rtt7o+N3kXB$7`lJTZwc900qV+
zTzrDF<#+$!tn=+Q>_ooEIH-2iUJa>&QuKzq_^&3OxpXG-80HNukLxD9V%)GY34A;W
z)M)Qniz3JP-enrbf)Csxq^p?Bx`(AT*Q5$0rH3)Nppo$7mQ)+Bkjv=M3$Nheg+_=b
z#1srV79!Sptc-DqOrHJ@h|dlQ`w0_f?xj$<@aoVa765k-*(MFdRaid>=XKa@r4wug
z(&jR{OAKnI2{$$7^>JR`sY>Ti6;>+Hqn=}GP}{IGG3}R;nL6~mvNLee-ai0W)5iFD
z`Mhob`x)ZA%HXK-q)2T(F%4?nBvg7f4zM5mb>qrMw&|=k;)3xg{&(RO@;|uIWh;{A
zV?gHUePMFf2)NUv<@+Epc-0Qa#7W<smceQQ8CM4jEmNJq0pSizP{HSxy{ctQ)YjRf
z@X}`s{6P;_Xh3&-T_bWXhQQl>qrIvwpR6p3Y4`oj=Zo%c;H>=>e2=0%T!2h)TEo%&
z1=(+)#eE_+)}u6%!S8l5`L>1(h~80*Rm5j5Yej6dR{)7ugJPiXzeTHF)>rma#MOp#
zWALbq(dQHRnkLqB4f#O?Ic(Bl*d4ji6$k}JoTQXy`qU(qsoq2nNknP#%9kG%Uv9Pq
zn4NdEfNWpq3?v*;`&Fwf=rdl^iig}y9Zu@HPZ*ij8Rbi9nvghi32CviI}{W4djiV&
z+CmL`#q*-zS&qIp>7>%#!gAt`$3rG%o4A*U26lb+a{P&SCxqI(t#$cWnZs!YKiFL$
zm~HiYS2Kd-_DLzt8m?8t@8VOhWA`sY#te^k>WLjIC&U1MPg@FuV#$vVeaK+Wfv+D~
z{|7lHoXX;U!?|R8kMja{8i4iQIW~h!1Xwgz&CWd(mrCG(IYllW@|y1V6u&wYpOGi+
z{A-~TaHRZF$?4{bdq|MeS7Uo5O}(?dF(i^}QAEg#fdB$2A-AU)X=P|iAkiTc+=!gJ
z=zw4{%bxe@tpBDsbTJ%j*%gavB5sIZcRql&_NC&ZLfo2OTM()A7wckH6P+K;Jm%l}
zcU|mzZr^$qmFDQaLP!Nuv&t3czEw=}F?>5ek01?$W3g=V@C^|9`3Ez?r;dVsS~R=K
z1C7C2oI@huiAC%%T#)%u4Ptc%IlrJl<}`l<W%5xCn6+GApNKZ@8q#A{p2jI}+36~@
zl}^0FJ7`B+t@F!SDrCAuL-rf{vL%{#{+ErsIq7EN=f&2jm+5>X2K6#O_EIqs5r#<z
zlRqFqG^h+CKL#&PE{?P96&Zq|1@#fBIkZO!hG>Da2(KrD_28h=@U9!5-!e%5$fNq7
zk<=e4C^2;CSmWE<H;R@zedA7sWOE1Mu;TVL5_54n20HRjfVfY>>N~k(B6#LI>-d5Z
z9Xn^Y3S+?O7d0pQTl%nF;#RnVGO3JAZs1<712a_CgGOzLRNSWaKs*Q8n}qD%*T7`y
z+p|K%TH4^FQj?SHFSAx7WujCX8<j{}+$N4(qqLG-d~@g3xAsM!lo$#QptUc5@9QYd
z4IHK0#vCo!ndhh8Ujg~`j<FbQ&Vy)d>Pv#29x%FUjrzJ>O~F!aG6+(Cs-R#Fm%8ix
z?)rQHHpzD8xg98(;yw_AF!NmU+I_ej%GTXm$kM@|%tkn+*_?AAbF14d?WP%FB6#W{
z6mw)~R1DtL&@#hWI>AB(mVfq`FS-E;1Hbdk-4?onQxTNIw@0&2EwIy<B+a%rM-_sE
zL>r_eC!>SLE;E<?$`GzS5-qS4Yq@GPR~XEBAXG#`9Z%4HY}(b6RD7`frYNN^{&f79
zt)UC=LqdAUYRQB#IaRC2KQ&)(G*ujup!kH5VHAPAyd}0r%c*sJOidn-+(EI<njr^6
z{n~4ROJb1lbkd<f(!xQIk3c)md0}N%^UoiHQF5Z=_$Lxe?<M!loQ5e4GiN=>S|SX3
zgg>siH{m;g+`=%@+F9HlFJGfEc)n@)<<6nF5$Z(wvPLA@%Pl?9L*l8g{xUR0<LWJd
zrOco{&YM21c`>-TM=VMb7bNajMgDk)r#BgYn5}Q3TNjH!1*2Vq&z*KvOt4&TC|GeT
zq*=e3)bF%dvt%b_49gCSoSpAt@x06rzLN`-EV>cfA`}VgbhW=tV-54VzA!+IpP(l=
zbn17`y3Y{e!c^LGI;T%$a5qialtyJx`gwRYLyKUL_ypw}JINjPoNiY;9Jdxy`7Dr=
z3e2hirgX|zaTtn~2M)-yXexQG(Ws>#%K)0Xyu*-3wH9CGu?|O8SDTL<P??w86+LMm
zn)3F;H9kB3aPGc~AMvGI+Pu@Oz4e_s2@(C*K?Xeh!uW|Baknh*(^}n^U}oA{VEAn^
z&)yE&r-mYH;n-f(NFb-8e-h1NGKRrV&Kk~w?tnCj)iGrMj<#oFzB$G${&kCalw+94
z-?^9+R;a=}(3A@e;-xHB{K-*?FAE$BIjJTE<EP_~W18F=p}Rrq1Ki|mt$vsGtN6<x
zMWp-y8#^jMT3HNQXa#2HA$R!3$;qn@?KhHOAcz@_Aq)tRLifv&j>DS(KWMFj_!Iy1
zc;?V}%|)qV&JJ|4r&{egsqWMZaaLqaisP9NDMneLGGo;B_1<08GtpOSX_@tGbkzHN
zMJNUc-17C7(c<eOh?Mn+%l*oCRK(?%+Qe1?6;-GF1>-JQ+2I-W^)=8%k*tQO#DTDp
ztTwCNd|l5kQx(xQNF9P-E?8~3LRTJWJkCN%n>Z?(raz}=Q={O0PF6D4I~Ca;|6B$}
zW71(t)Pr4t%DCm|J4Xwsd&VsOz+vte26Y7#Wy=q}g@zmO>$J7Sn<Uii_>5rL=Lu&}
z|KP$4JMqfR{}A;Dfd^^K4_FMxOXQ=H$IV{+9}J@~Khp|w3V5}QcYG}$BD7$zQhQqh
zXjEzVV_ElJk?rhHIQ9M6VsqVwb4}_ta79`x$Dj#2kh+b-k*y9l{3lk0?pINaDX?BF
z4ygF!OM~RTw*Ax3d^LNC9qsfA^;_$IjVZM03I}~HsqOW2DIQOa#}^DhtEojvdY9Z(
z6f2)p66ey&&oD8Cl>kObpnmgGBbT-GA@?4S*>L#6)LSMUiBto3g;*yu`-^lZ2BIa{
z!H{gYr*dTtDwdHN$)2!wG8C7)6b6V&%Eg*&bPejz0@462djkFn5tT$~4Y3t}AIskI
zHl#*B<E2{&8=8~d+QUw>qxToGBA2NbqCeWvx62a(A|NdTiN;d@^;-HE(fu2WD5N59
zb36OZ#0&V&d>>IyrtIVQ!M;`awp#!3{d)f4a<A!lUT+tb%i~iw7kMxhujKzf?*)Ij
z&hEE<j5yv=Wa}1yP5Jbe!igUUon+=(x$i?Wm!S9E>Dx;b(z&V4nea@zYt|)`ZyMt>
ziLm$Uu|9i+k$Sb+uMH2|HAmM8<(gK-+LHJ;1CLy{S-+sLg(L1EFqH>SspS~(gZUTd
zLiW6FC8|uJ@Y8K<F8Ut@+O_w!EKgl%8&nZZQi+0gaEhzBCSQT8u|-&K=*wlYT50ck
z3C*e)y&zKt(cnLIo!S0<Yk>w00%z&?ZEb1#FEZ@*Zu#-x;~bh=5y|~w$(YI$?8`#*
zFSp-zWP}I0bfk?m*3}#yaG-$Jsk`?^cXjT(403LYWWc)6*9<nlf=xRr0OIOlb^!Qq
zi@JfYf03^U><9f|<?eKvR}Z?po1)Hfq%*{M!#7T17^#2qz~*t>@{KJ^fqyV1N6fzz
zu1a7~oKd*wd56<-<cg_0Y;=qWPQo8k$yO-^&u{94A=8;sD^vyl;4Xr#f<mOHtWcio
zuZ*PY7efcLpe<WyjA~JGaS!vLX!M2r$T9<6E!!=BSB8NA+83T;ktE=Nd&JKzBNl;N
z$i_kKF7okRCAQgePp=@cvg7D>huhPe;rmN_<0wrx(0k-z*^eyR?q)1(R^ifH?pnG5
zFQL=&Vr1!Ol(q;g>oW!AhUR#6oSen4RVS*Fo=6BH%o_P{DceY|+8s@A2|6BiprH8j
zO3_i{<vNrUgNgnn&=p@#YdaltKN#N2n6E}qQ8?rW$VxP#wwlc#R1|6G=J;c9N=Au-
z5Y{U9nBlVT1{=xHResI(<1)_`-bfFJJ*x3WA+|&oQs6NPH~SYC#@e_#cP$~A{p)QF
zvX+@lt;Tpkzc+c`%KDrZ1yk7vNMd{vp4E#p;c)CY2HgEx!nUCc6`O9ISW)G0B)<#D
zb7r<Aix6f_b7FNuyBLfWvwNlv`M<oxAVy)%yXUNDQXw?g=Ug#kOB~N^ohF+b5eEn5
zNzW!?1GeyFeDW6z0dgQh0Is3}OIER!a<HzGiq#>2H?HJoOnwVxg&3#h2K007ti85Q
zu1`CVM2A^mMMFL?<DkpLYIx~2;(b_SYn)?LEfnVw<JjH1ub&@ZA8eE#HwQ^YF!!hR
z6X(9Y$9L(l$R8yUNPH~^-0V%+^><&4_JSP!Exk79EF)J|u8@MA%-iQlL~?|oi^#eU
z!o7?;dTUf@r+8dN@`TmvMwGB|a)3<rWmkPp_vi$bNh_FifZXn9o`bNNpI!opwe)QK
zGT}3^0_&R#EXr2S+G1VQ@4tW5*_)6cuMCvmrx+zb9dGJYIUPJ8d5^EieA>%obfUEB
z)$n2orUBMEPuVC!5yt`H$K1z{Z)#JXA6pg5+mgD@sypyzgDW4A9K{p=JFdc~#t*n6
z9}{90yG-x5ExQt#1a`_;!*EY4Emo?+VCOprX29r5=`%#eqY65JSqDE!{BoJyP?YTL
zb?jxLb1NX79Gv!O-fk!M`j;~s1kSAmZ3vE|MU6hbWdxeLsf#G9X(%^^ojLhE=^X)z
zgM<BlbZ~XFU_}v0v$U)m(zSs@PJn5bG~V@{-mHempLDfdVobzz8Awpk1%~`h(^ng7
z!xvUWEw~d4I-PWXiL03MTNI<lj~R`{Mw(@$`WbE}+Sl_U5}L&M)JXsmv-wT!B?FJr
z#bG9o2EeV>8g>_i??|KhZ^BGYCZ;IR5t|pl@3{Cdc%lu|_ylAtPLw&SjjlUl{IqrF
z18Q+mrmVB-a*hep!>+A{)W|B{nCYEhz`wCm^?LtuyHtyn$nB9hM*mfK)v~tsp5Z!a
ziqi7*KS^|Nd<fvp&w|GUS7?ihYailFjgH~BKy^=Vc8rb`Paz6qQ>7jT(c0hK`w<X#
z%<myWg!FNY1#gJtR!v1fniV7iOShRtUMm>jcn`U|Z2-xo1Idlc77c#P5kD^Z4NBER
zhQ_PyIc$J#R!q7S#8r;dvuZ6qn#|@;^<X-~7F@h4KSLBP!xEySngv)nHdd42;^ME2
z+TfMJZ!?&SmBtL$%O!b*xF`$!9T}!h3l;ZRqY9Lzy5r`1eX7XYax*IzLKZ{7@=_x)
zOTu)=HIMB>7(f~$E&b|2&~CYGnoz1-E2cVyHXnyMR4r+ra88(KR)8>$AS)ElUN?bN
z%4go5LAO!oO1r=MtJT<Gb$tQ?iR|;19Qi_-W6va;3q@*UpZ^IOWr24ZAC-D^E3M)k
zC2-?*@pu}A{OGIw5<(X+Z@Q0T@m>8_$&uFO7ykdjPXT`;3BDmrFC2U|Jp8@r3PI&s
zcoSX2R)*hR`+L!+#o3KlzAS!89H)hEZ>DkLSQ+R@8GL2K+8+bDV2P;2rjhZ64yY0|
z8;l#&Bx*C}nBnJcf(KXyP-}S#dCWyLUDXF2mj)dSD-v3{N_PI(m;Y8kYVlibKi4;4
z<1HfQ@;xzfa8xUdtD{!^=4dH{CQxQe#-*%VR1?#J*wCx;*Zs27lO-@@lC~OE7DK%*
zRGz<OV8%g%>k#h1f|lprGR30k&0EPi-dMivbZ9@U+~us($aY|<&5yexe!;(}0g1U+
zfGi#AAB6*Z)V!XWfvbY`t5qx24ukBDW_r+fEHFI2Pn6H=l0R0oy9A5Xv;-7zt3b>7
zn?Dfwd#1O}E?i$qUEU*6;gkfIu9L=jSQJKnwnsO-i}(Zj_K5pWj{NP#lZZSuFQ*A?
zA*4=fE@7+t$dAwecY}_=$lY@_JlnUiw^9_3hWcP<HYT8UN>e{wMGZq+79n26dYw(X
zAUo*AJlW`;rPf;?S!rNO!iJ{_%?}ZLKj^+4+^3zHgeyaYfzSkR66#iz`&}p6_I44)
z`oBuj0*x@1q8Y6>HJdxQxd`*c#f51MS8cO-f+*y_E!L{w190U-Maibp2yH>|LRNrk
zo;@0I89ZtAG@H#Sc&*|h>;Cch(N-x8SCs5L#*7Ys)ueTf#n;N~C>G9|4cj7gIH>Gk
z)y1S)#q}<upb*+shuHD9le0jd>sIDHy_8Y0>u#+1`UKl!md{mk&MWwnweL_}dtIQ;
z@?G|rCN;^*$D>GE9(ZjzsK3o!drVfj8b=f}t>=DPu+^bgG<!?bna1Pv@)}Y(z?^BP
z3KBj?U*6*!Q&Mv4Jb#f>&+b)zri|<RpAIMCOq~Vg)-~<#LQ>zE+iPVd({}T?uw1?L
zv{Z3N$Mx_I7GKJzlww6UZJVUf%=;xYJgxD%<dtC4ATd?1rg%iRnd2zp-+I&<k$PxT
zyU1M3X3$aN1#1zD`V@|I&|y!bFhQu)>i^f3WXoYo-{UEXx0R9U<I3n~B6boX^*TXL
zq~bD2(6Olvt)=GB1H2r)Dty(c_+sVGPe#?AM~=^bgIX^o7OC5C6tSkH>&~gNAcx)=
z{<2ob%bCu+EWe54RlW0y^xb{!*~%(SdH!V)b&*FE7}=1@@C&>IIvZ(<L}kmw)jR8L
z{1{Lwvl(vK=E_Q8M>V?gv^m&kMKGhnGfDZfPM!&$30-Z~_<_wh;a;&FHrV&#^6Tlf
zGpH6dSooU{Roj+PppOvM)BM<#%+tSvdH>(1O&WSrR7Q~(C}sM)Aliquv?oMMP>K)4
zkQr>_R2A!RHdkp?Qb{91Ag38t_{mCDF)nSc<<r+;4+l<jx)8}ynWxWcC_ZfK8r+}0
z$0vKJscnS%JMf~7XW`Q}B+nB|HupL)QKr48RYjhXE~X&ze=14o*TQ;p0P@}@WW~@b
zs_Z0EJqPMSY>d#$yyGi($Ky$tE)p<PzrG7T3O1*O*3V(MFMiV*>^s(l=SLQ0<CBy`
z>VHy0tuj{mKgC`7Ka}tLMfQDPvnP#`kg;bMBRfNgELjF4gp!EtOUS-&C1x0r8G}${
zr;s5thLG(0mTl%UZ+-uV@AK>P+x@y<*L|JqT<07h_vANNX#LP<L!jWs&RELVM``b|
zt|b<&f7c$jpM`iqoIK`UxjLD0k&;44YQl|{Jf8JTf)Xcx&`nzg9hGW$c=o}kI4}MA
zd)3F`-Je1JPJyL1wW)!DA)fXos|K1c{&p2~pA2INXD?Elr#9y*<grJtET^g$?Y!#Q
z0@vs4BTJhheM;i(E8BvH42?21`XD}5i+0N!Hn?;cG<eb7(h*eMY4Zi{VFW|nP%`{#
ztx~N=rNT~!<8DD1LgbD3^wD(2BX_O}gv85+#`pkof42{pKONNv3P0Q+8`Ft;KIg6~
zK%Hna4DsG~M-qL=6E%Vptqec($}+iFQeJ<?@y;`fwXZyyHw_D{IWCGvZUX+iwUg39
z6v$0|GXl{u4FFn#Kz5c)aiO+YmFk&qb%(6;nrFg&z^;{_mihWqck%~k#f001y?QAd
zqxEG4d_-cvSMhx%v~L|k9m}jJega1^9Uwv|9Mytze{d6dJ$~}|cDLM9UX6RXl#CJu
zJn#RW4m>`~(zIz%y#ws%OMUGayC-3f(O1R#B2shaUv2zsa$NTN`z`Kb;6=7=D)RCN
zr#cOM+@VH8`egTG#5afgh7+9GzZvpc1}K9@y?ZBq=0xWCY(xWCjbsqp-<h$gKFv-p
z&l4l}8JqnOUak_A0@?zW2ahw5f0gp%$_$w<shAH-C)<Ov!hC48P+dE>DtkUZQ8@tp
za)5Ao$=;Y;02Gu@eZ4^@=eZUmi7w2=3(-&-*&kHLx+{y}rQA72`P0WmiOe>AiJ};9
z|9ml+<5#nqH1AP8PaI-3JH3@3X)UR<=H=4BhAeghr#g#E8y`(GZ?&--)MZ1L-oMa~
zbr`JNG^~XMi*A(7H(75pPtMXkvnGQ0=z(u%d7oe=q{?oF7m6kJ^h+n|;>kbmRKcj5
z{E-#Yl`9YN6gWP9Dh;d$A2c5o5_0Ef@rq*7rFDC^)4Av74CU?On2Q?6k{Y2z9@gSu
z0LYLg3N(w)uJITvuwSnV<+bqE06V$S+fH`|FSxH#cq26m7AaL^^%m`7T@tt}y<be9
zQ1{$b0SI_C9^gTeOkf9O-ur|fYU%7X)lVaJ#h+TI0>~`Vu<K1Rsdu%ttrHV@C1~f=
z<n;{RY0_oB`Q#ETEtEez?pht4ZFeo>0oVFaL(w1buXOoJK1&HWR8S2Dn(WN9CzOg-
z;ze4xMt}R*`zJ_={`H6K1g61j4gKt?i_ZA7I$rC(FA5jL;x>3!#D7eg#+Wbe?(Gy0
z*_?niRONC&v59d%6;|6-&?i8YbhZU`#wsv7bjGM4%@>Pw;jW|3;nkwJ4=Gui5cNR2
z9c+52q3vkJi#~)j9BmfE*wAx-YhW#78p@ZO_mB2B)=|o7?kM=9ik8rSirudog=FC;
zXL(IHrYib}r0kF-Msk*<A2;bepAOgTf3PzP$^K|Zby@{%G=R!4D0H6=!q{A2p3+M_
z;G7D^@J&Jmr0d!qS)e{Ifm1)kj1I|lC-X-CP%{UK+{g>6m0qb~9CY2&oF2+=Qu0`n
zo7YvKtAYJyz2#M$zy|0K9Jt*|^YAe7ewU#y0KMyG>|efv>CtHloXPB~g!XHmL0&(!
zT;*R`rta=$&>OM4mR@T!-@{5vZZSX-PavdYn+7X_pEwV0eFR<MZ-P?lohV>$O<Z2D
zNWHWJuM(PTRYvWIui&;3SHnNDjq5DLFUh}Ns&VQb<HV1h9zgeCfN&~6Gl#K_CN3ua
zsr+Z|Yn1~5Q7-Z7RsK3`rXQd1$z3{KTi8*cn3hhZ#vqURfd_sOQKykUD_1twnIlZG
zGx;t<B`Wh5b&uHc-=p!pBx1hFDWc+vj>II)=%<T4ae&u4CPe18)g#)qxv#9zyjcno
zz<FCDCaXHewm`C2yjAu?R?CN$c~_T$NM9ZbsthcTcB4!~6!-v+5osFb>z1q5yxGH<
z!kVcB@O*MT_rPQzPx`WhhDNlwK)sKKksxlF!&XX^gELC=8zoa((vjLtzJTF_{naEG
z*<n$GYX_-i6~y)M+9~*3lh@*g{_@sLmBz#4EIu98i)5BUH+{O>j$K507yuFzkDTvs
zRZ+Uoww6aaXPzhMzYRUuvi&L_s~G2NAg;37tMI0y*Nf!w?wtj#uk|fG6$tHy%2WpE
zk6SWX+_Wa&_YbC+_~rpH0Q4<WRRwiuL1Y$L+L<rqt+_olI~U4otu=W2BABB{(M8&G
z@R^V6->cZLDQMUk+TmT;+bLYiNC@!Rf7en&iDY599iY5-c9w_RuyM(?wG7bSQ#=ov
z>3>Y@^&DTvmlN9XywDN&Oagh8bUo}lu2fCgYigX?DU9B>ht`(OB322Imqn&F+<A@)
z9aEoC=o;ct=#Qn`Doi)(88^5RrKKrrdF>tK4chCL*}!VEgY}?XbzOoMqllv#tA*H7
z)*B1K3So3SN3Q~xWXj8Q+3PbJyr1^sB{^FlI<@;Lqv{l|oF&Xau?}6ZDUn>bX*2fg
zZCNzOh!$6PvVYTFf|cCzq;~mBCGw$gF%g3{s0jrZOCE&w+Pfo$iBeJ^NI?H@1rqZP
zDX%)Z2Q9kua~xSoeR%9PZWdC+-clWJ6<`?}7Esl^;pU(Fd@Q+2UMuU_YrM>NarD&p
zXEww_Ta*dF-D+=j;|vgfxx2w;H@_?^`RLP~18)x`a>hg+?)<o`Ykq88;aQp=%_b0h
z2qHR(KLvN#6o9Xd1PX?H-SMYj><FQ_fdkIy5a_EfI7MLrB5AAWUZ`ub<2!)15Lbq5
z@g8|QVB}oAo{q7r9TlcjdNS<W(`wObzpg)-IV>Ho6zyN+pqx&p8+i{ocDl@)G|o?X
z$*BFi-Lo}g$*?ck+}I3@9|(FXzZIZ!F~S;S5G6usl^8KIE%k0OovToD-%7R`e=l?R
zRb2zql=OP|?iB+GUCdM42jrMvs~q>!&APhmy56L4Ccn2>$J4FMwIP`KAJtMZIg7`*
z#iXek0{BOGr$?G$VP$OELH5NpkpSwt!dfcMx8td%-iU|w4Q%g4slESf>_F~^U|gI0
zq0Y{Vott~`@OEQ9Hs{xDB*oIsxo^^^)(}!Gmh~}*;!$uA`rB0a6jry<x7S*Wi%s%b
zi_FaxN;vy%-yP`Gc*ouHqnrEJS#Vf$lzQa4Oh?AuxFxrvUz5MgqksR0wVA_Mh|wk@
z+bhQ$9p5`L8aVRaG~L1?8h3=h?GsK+P$7-eSHIkg3wv4}d~@}Ham*^;S>njUv&_0~
zpmyd7#@5uH%DcKK*gnXIr5pPp%ebpCPH&`&tNSYQ%}-22@d7smEcSP{Fim2HV1QUk
znL$Myi|VMnc(mq-nc;e6?0u(WU)%Mz#EE-w3PbC6cJDK^`)jp>uB*u^NJc)jHO0R4
zdCuz0kP>n{g6d=kv8d}vM3>vv=>JeimJ$^!Lqn=Tnx%(pzDx2;^Q+#n0fLe#B1rf&
zhrGS*p1QxF^_?$=ZU(^4-=K}3qpp2gXF@~vdy9o$&Q87$^oAVTHp?EZdHMUEjrqAZ
z#@aj*e;0Ho$!uvY?MRWescpWAt&^}`non^WKCB)+?J*L(!Kdi<+_0X*8Ry@3IR51y
z^>_OBJ?rPe#ZHpDLU5}&MA)vjX?@n@2VwGrfv^+Kv<KyBrE*Z2Gn|Tzc@MLrb~LD8
z^ip`kC15T^mx1n&4)QCC1^=-e3LgAjK|k-er{`EDm3@CJdd{e|OrhjEr_)(?+YFWf
z>@>t@!`(!!Vro$O^3p*`?Y>KQyTsb<7pTTr6HHFDSK5MWQMU5(NAxwR-?mnBYkX%}
z(CN7kwoRk|7~&B*Ru3`?y~z{Ow8@5TX*132d%E>MQNnd4Ay>w%zR0nzZCU4lY&P>d
zegXO8P-~+GmJ-!vZ~BJ^JPPiLCtw-=Y#!uTR^}xt##oYFZ??Gyw`vPT`+IKju*q2P
zhnR&_A?a1e)6kS1fK^!Ye-6;<K#FU^gipgMJ*FYEpF*5;_`+e;G(oejT6vAGjVwU_
zi}zfG0H8E`mGL<zr>>?y@XNCFyqox)g2vqt;a5z=Ee+-<e5JF`mj8inT)i?Izz(AK
z5p#x44K09PKX@&d`}7X@B`f6wzhCX2PXr<OYJtt&mX-SP_y>3!ul5MUysf2mw{^e%
zfE^EGHN+AAIkVCo$@&n#q^bS_N2{fl12xWT{Ilw83sGeG(qK*H4L3W?ep~=)DPqy8
z1)D0?wb(VQ^b8i1PuS;U?)niYWO{E|*3cnvm*oOWv>0#ekkRGU2=Ql^X0p;M(3rUR
zrY=d=zmcRhUo-Fa6&5ypae&m1RdE)cwMzht%i`V(FOuRmgly9{{&4*+JROdN#UYXW
z&znFl@LrwgnZUk@OXwW#W-61`?A7D62i0M@*wgyWEdwAjRf{Wg_}s&aWl>9S`g~uI
z>4_Pc1|#1;%<MN`qD@9#TwI)MDj5JknvsKQJukzZ6wfffCT9gx*W&vo(p?yK=0%6}
zs~}a#@KUIU<!%c}XF_S;&`uq%#kK{X$V*!9tV_}E(<i%LH)T$PtBU4PiU#Q?vobQp
za$iG28YEVeq}VASU|+kM*hX)f_)r6BqD&5Hs;b=T_pmx3$?&j~(>KxFTGReUJ<O#?
z|2Ea9sHQwF9sz@gQ@faazcL$#C+y4g&Z@pB!~WLZGDlQQf;YPPSB$jYB|wcQ?bjd4
zvCCljHsj04a<q}#mUgnZ@c4+2@gdAcFlOs0^6W*lqr+{7$sLYZ9@;FG4()^89|*#W
zMst-Nj(6Mx%fW>k)H;tCj!|s~%BUAcuwVBSG1(JyZc4E`RtNjBM==5mr-yatO!qeD
ztc0ew{;PZh6+$)Tc9=^2^p+dfHBcp^Y{AjrQ-Ub*@D0uyKF3$~$AW5O18JA)=;;^)
z`ba$`@r?z3|9M*`$_y6L9*-bP%x!{r3$L$#3$^tTj{*no^IE;#;KdTH)rNeU*@qoJ
zP)L=5E5Q;~+`0x&vdmm-!w?JRz;H7dhIG!ubwySRjl5dg5@tRQT3F@U4&a}cah$_s
z3XRwjbb1g&tKE05uK|*&coX+^2aA_x{fQsazd#vnghPLDcY&f9?~{Ui(@H&)qj^VI
zn-Ev9uQfxN3sMSZ**b^we^>N)+Ozc+qnl(?=H`Uhp<5YXX@}5d#dvrwo@I2Jxyr8%
z*2XUe2Fp;hr!Dt-yx0WUh%rzBwmM$newGS=Pp{xHo3?FUBcTg!3t_&}o4Tk#XE1!d
zH*eQBS@CtMg3jH8tqZO(=q!V_lq20MUrh}#-|qSBtxS-XaPe^*>}5*klNh6>3FvL_
zl>9vmCmdFYWUW|5?W!7LZzxxRSUjb}Zq<<?CxD1j_O!A($lif>X4x?%YTE%Bjv$41
zh{>KPVU@indeYeFDEY=4M5o=Q!?;POV|Ka?KD-jUr#!2vh(y(X=$_ovwx1EW&)CPh
z`GkEWQwC0Hyc+2(_zWPx*T{Vpbr{H(h^Sp+uv5P!E9}h@=<z*p9U^6n*mVU@Ap~!S
zuLJ{>c?MwdH-)rI#6)#vs;Xg+2TRs{SKk`L%nJl7?eW%W=ECl4=&=$U^cr>vRd6X*
zAO&phi1&+n68T-V-Ky>`L_!Jr;v*s|@fli)Qa5W_Dyth_s<#|HY&z=RaN}HQWwMbW
z-qq1nB3y6vQ`M@xk9E(OX<<r^O9-%?X6pPy{pVMNi<KRxUJ}3|EXojP=awS&XPD0_
zc8JFj1mxX|o}Dzv{oBz!TSSIR!-Sqc_dCA0$Va0D?sGnzDM}W<HQ2Fh@E;M;k#yL0
zE>G}1wih`J5ju__1z;8tPn?Krvw;~GOhrTr@o<(PKwRuN%Zb=<UVnVnMrx{N?+}tB
z><Qdwc)Evx7?KexfrcbHyYr@=1p2q@nB^DNChN|Yh-W1<6&o5fuRC%`bTh8I<k)X;
zHmXDQ->TO?CS<9nLw(>MXuB9u5$XSL6YPT5`||kTKKnmmB;)_vuWcelh)Jg?{>0`F
P8R;|BGtot8KaBbxkwwQZ

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_e2e_thumb.png b/0.16/_images/sphx_glr_plot_transforms_e2e_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..db540f4a19050fb351f21878745ea736294818e5
GIT binary patch
literal 85362
zcmV)$K#sqOP)<h;3K|Lk000e1NJLTq00EEy009^X1^@s6a$KN!00961Nkl<Zc%1D0
z*|Th0cHW2m#%$JFnS1Zk+^(u_p@0%v5g<j7ASn{EB+5kaLAGp%!{Kl^9N`H634U>e
z!~cYy{9yY*QuvJ$?XVnn&|&rzNtx&j5-3#P-TvOw>?SkUn$7UTm~&<Bea<ZaMby4!
z)|pl3p0jgju3YVCzVVH3h^q2KtRLz_uhy&eYQ0*o){o_qALc{uPkS$3tyk;SdbM7y
zpN#d&fxKF;)~oeuy;?uidgVY~tyk;SdbM7yA8NgFAg|V|^=iFZuhtK>UOA9g>(zR-
zUaeQ_hgz>3$gA~gy;`r<tMx;zR}SRWdbM7ySL@aKp;ow$FZl=m;2-e(^74lx8oyfq
z1}?!`iGI)>@h5FH`qQ|(l(*l0n_v62U;9(o*6sD~yYKQZe)BgGk+!Gp9BAdZ?}BrJ
zNIySqui8Jwx%JGUr@vRMpX=ncb&Dr2$jR?sc6#Obi5JhNy!hkEHWUTb^=tY*>Yb1L
z(5s)f_V?}W3hn=GUrx@M2!gctw{v9w$I0D2(%WZnJ+Cj@AIA$P;ZJP|wN`%p*Z&%C
zyz$15VEcc1OYWnQ@UQ=e{{hn=48w4|L@(RI$zA^^*{^<Mp=y7BhU>*!^3G3s=d~A}
z^ncCo{^K0KD*BRp|MZ;8pTc?U+tX9)^}?5a%7>*Hh9CS5f5IL5w3ihT;rX*me(slk
zh5zh7`_KPm+x=s#zxB8N7XRD-&HtL8|GA%`lnPqW=^<=h!%-13^Wlc?zw-efd~$^w
z211;0PT2;BXW{DUC+rRjE({DY5<)~Iu-%^X=<yqbaYV$SR6GSQkgB5)aUlST)Pj@B
zc62!BSW-n1R138%I1)l8p(KS;m68%AAMg&&#z?7!-QfTtMDG}5WE#$yhAj|S3XIdp
z`F6`N1lxX9vMQxO6+y(ctE-e!skN?7&I_PQ)eKb$9&$-MySyYXGqoz-fuoX3A>~9a
z$}(qCSr~%n!TAQ&%JuFEB|XQ92eyQG&UP9oweaNG&K!XXs>(DBj8h~fsA0o!@fzE+
z2aIvByF@FZ6(=B9@4~YCk=PsW9bftCH~ByOpZ*K}#^3nw{;6;OQ`Qf4AV2a9L{)kF
zi(le@_CNn$GHxac)gq=^q4d43kQJ@w@vDPUD{$fvS~PIm-U=e%oI5%Y1aX2BX&)6c
z5bOJes`b`B%IW`HyRECPQPpb9x*c32wf-Jiqaf!j-chyvui6Z$wg*KkwSty@ycKoL
zV7EU4s1vl>xf18_;y&G|pGI45J<=A<+V5&Rcl{jqC{(plYN3=uwVk!?$T6gocAkIG
z!_tlYX(U>6A>~Y=McINKznamaDpK3E5!W_<yoTZ&-Z@;`y6uDAH%i5=-dIlq@^*ZB
z;(*%kZ$<6>>M2?(;vDb1^A7L6_wJ`2<G;=dA@Ykq_htUtU;Q<5&Y+dNd5iksZJbns
zD_lMM82Oz)KptK(o}Uwj5$A+6SCGRqJ{|{lb79<^v6&`<AMkPHwKqS{7k}zY#9;zF
zh@v&OqZ=Sj_z)0PC<&*92j`CMW@5>ed9Ku&akAib&pcmowJancm%?FpATK+-tDJ8G
zkh$I+(5h@VBWIf}=i_6x+ebvdWiG-rjXZq#fH;g)RZ3Bc3Mw85xZrRCs^$Q++O?_e
z{&QL=X+}y0QBs<@yuRWvADHKvoED-KpmKGyBPZpM5;ZRjG4S~}9<tqxeDB?_adr7^
zhA>d{5y&Gx_u3g3XP&Qq_bKyS2!23uIEC$IVtc-&ga^2b&-3WD&oPY=K@loi5<_sv
z$A1XBr#})05<?)RjQ9ReecON9>s}x%rA0DTDb1MH_BYkiN~yXUumWa05kzz~z_+rJ
zmgWn#30WNoP|bLDBi~_zPYd>?hvOp9{1)|SWDzq)+MEd@h_-!b&ZHaMT5m%M0JS*+
zGuDb)0BT08*;adg>hXrGVopdwYDGoOQNHv9pJWaV<`5B7I~*e2)lpZ+Xoj}dLN0}x
zi$#I$-1cpVYtf>-c;4>p#L?mQ<5t>tw=ki$KkYnfwHHMNTxq_!Lpy?0P^UQQ$JFl?
zP^or&Js|0hq+c`4$@KH{BJ9)m(XNZb?$ICRDz%bwZVtqZVwn#J6-VXj`i5uEo>8*x
z4_c7wDO#zA8`3fp!iI4ii7{HV9vtHs2oAg?i`o%Vu4ENlR0gR83R)AcoyU2uc%N|I
zBd#J8ya%U_bjU2r9-JVxBISUq2@zp`*dtWD2+jw*oA6jxvOny|9&&LAmEGV8IkgjR
zc~-%JdYpGS5ppfmS`c)-7N9t-<^*#B6z_y-h~!c^EDK9s*dGpdZgM7-O3l@7x}wZ;
zVTy&<&L4Bwy~ncbDRrjQ8$PHvEb~Z^!oF6jWSkF(c(N)L!FfkSk+P5v32MPR+qYai
zEE#hCks%)Iv-o4a?!|$upya8cTj5A^BHHY|_Gs7j2<5nmyXyx*Hr<e0M$ooh&Cob^
z7tih1q#K;(M^2npH{Po`zJ+D8Iu@v(0+pCCcLFM)(#)G2Kd*YSUwRko`T?-s?eN>2
z$ePg{_ZzkIvD!)1mUr13%;~i2<XR@@{4uNf)USW}FYN2xk*QIe<7lslQ#y}h-FA1-
z0(URd>2f$k6v07r6z1zsvsAf-YOgz`x+bhn#G+4;mtf*Q%B48xQ6E7Is)>2Y?B_i&
zv>-)z{6P6%{ZD?EL7sDUHRD_)4g=n~cFjD~_FHVXTY|T9SrkeI6}M)uUO=e)qwn7E
z+aJvQ&7a@$=iWTCGp7ooI8kcNENM^5%eto}Wi#v|Kxs+2@(<rheEs=`*I$Fb`uUO1
zZ6k^Y6@2gv!+?0J6*wG~9xt|sy+4xxQqigwZmhwIMObz(2vwCOW#(ljEi);ZbD<rB
zJ>LZpgXe-%LQqn=VOb7TeS$V;7o6gJ#QTAqDu@#Nh;!bYsvii$h7coyP;0Ss<UE^c
z=Ig(E!SDQwe+%#b|HIJ#sMh^B5Y(dZYH48Y5Y}C%;y8hcWW8y$Xv;8cTPGIp&bRb`
z+>SKM8ll<?MQ?v@H6G3WA3v+N{(XdydQ^Ti{(b*MPL0;_d!;!D=UU<=&1lNWAC8f-
zB{<TPs9RNpyT^Phvr;?PI7`WoXq(<Xrgg`_Ip6-#;VnB>@BA&;te-ahV?;-5hI#UU
zpPVje&Ptkt?-0`ppZyW*<jUOgpO!uKWBrs5pEwcwQJT@YPY2S1l3_5zog9Z<r4%7f
zgE_wn+adAbY|sDx|M)gfpBxBr;^NUGHbbOp!3}4;`SzE2@aUY-?!TfAbrJCq#UWJ*
zR8rdUfBDY;nGfnmy!-T*_>VvL8w58Yt}=}t?_o)atE=bSTt8=?pHuRTmb%{mD$2Ys
zeC5eE`O5fhGB<249`V8Vex6@)KSyx^Cyrql*ly1V<A8Sqijb<3vyjWl{jFx4EfdZ-
z5}W98RtT~01)R4~V@`=@SI@b*zGQcOP06J_8x=2w!{LVgVWDKjyTTM_rtt#j99bt0
z*>in;Lrw?I&dzz`!DHqnb6Awk`6I^Z90?Q7ZP=bY;A}ggz+pZh0wHesrJwscj~@Nr
zrv!>WS?iu0NKZnJa8&=Z2HMTib>pj?*z%4)x%R`FQk?w$#L%9u+HU^a0j{htjlI_6
zol~IexYo_LofQH69YE}bkKHlYw>H>oH%C~k(o4<#3r>!^Z}FBBJAO~!hMgi=;WAZ0
zr9*Tldv$A<EGwczetP_#Q^>48$mSAWc8>1s*Qx)Io=}~*)_&nT6xp)RUI)5u{a@xj
zPhF>N`=uGjOZL+4x0CN!>t*-5e2k1|XLtu`w`0FgppFoy26JZS{f_6)3Sawr=CF@U
z+li|~V2Tm%M$TV<i#Ow!$+Bg)B&3+57Z-8PQ&cEP3Bda|-{!rW9qK2(^6vY5!N1F|
zJo<ISgF1)Huq-pXwBdT*a(x{+Ts}c-v5dzHDGT3t{~f-z{SZG~PzxNEi4SigZ@b47
zJcwr)N3xtVlnqFrproa;B!dc_IK21Ns^pxl7@iVZ6H};6V?Z2~Y*xUxnn~5nJWuQn
z8!m6os5O&wqLkU{NY7_3uMRjoPK4-ZF1DVFhmTmw$n%@TJUiyu@#Hyt_|cxxkGQa9
zST5L{JthukAdw}XGam+=7nZ!GXa&FJt>+h9oVh;=)BTB9_vb)XI7nCMr1WT6)`Zp8
zYLi<C?VUxv;_hzr_&F!kmw1p~oI8DQhdOA{?rGG2mx<KtCp|ei7NE8HvO8JO`gwcZ
zqFOyha4PLEgxcypwMG5f3e197>VIO)QG?C6<3v;yJ;9bLb^U!+%NE4B+whYf3+bI`
zumVk=@=&bs76waEPij)quGML#CwD(@*TRXQJ+r)>JDu<)w82TT{w$|&-~@U@*JF@Y
z!_t$COE?J@1XSyh+9}P;_3M0(UPM(HVz3CeR?77yH&<7Nzl+27{iRZC!8ymU8QE+G
zf-6A6$1N9+9`M#1512L^ZuSdhNz_v8{E8B4WloA0&xi6oikpcpkd_;M=jlJ?XCMC=
z-q?JBA=r~4S}0?rP9u~pb-rTW8}sOF+VIWyKjiD~|4TRrN(MEkb1!>BFxX974Z(S5
zPJq&CU!|RYhjSJp_L`hltKVoXEK5P85IZ>ru`DP+436P^gX+S(6mnKvoY4Zi>*p*>
z#ydr8A=knfA%tLm#xv%8pyZ6i0bInnfFCA?VPt!D#xQOW@$0izY9<bjZS>69vZ%Wx
zoS)=%Zw_RAG+W`c;}&ay9MEpVUKT`3H!`v&A!{O2J0$kU+?oD;>wcNhzMCz)%%I$Q
zP30wq(zW3BHjLJ}{bop8G~P3sR_qbAnuDrRN;w)$J$~=4SX_@;(_P2ap&jRFwVk7Z
z?9Ri9n4J~dhnK=+dUXChQxrAhdy9G1_O-)D{oH5^77X^5+fL?FLpWW}nVuwMr>mm4
zcJ!SRs{Q<)WHv9{6YG7{etpETK4o8iG)qLNC3D#CsOi9ykKgCn<#R%e4NkIJLW}_y
z1KURrAO(i#K?=?(J`QX*5157l@4!0+r%(ztFIEVtf+S(_@L~QwJ_M*40Y1EZpMUx6
zzvVYB{s%Y_f^!VhfRu<10m&0}dyW(sgF{sK246)V9jJ?=ltdUdIB_IdkO-kVsye(6
zAQ&MApcaJEjJqOuaxrEifYxIBC4v&jB_ooFAy`sc4N`T^f{U|hLpiVPQ>JR<uorH)
zA%uY#K%JsxVenSpNGTJ&V>52aOJdxfakhQLp$44ugcupeGix<aPsOdZznsDQO7tG@
z0@1%pJ^H#g2LjFLHx^PiE=M53qVZ2N7`^6j8a3<5Ha^XH{85)$RU3=zl=k`ZwdOxB
z$vjSW^u&hO+s6N7M-9SqRLq%rv47`U>;ZE}IBEU9ZuIUry2|k!O8ec~a@G<zfBfn@
zyL&XC-KgF<L+g23U+Fch(__B#Z&`mlJyYGxA2YH~D^#p--lv@Bb^9ld(Hw#WAS;dc
zPBe0)W4Ha%+Ya!)mWbo&#~<?i{Yz3xeDu-BTwPxg#*v(6i@tFzhXXk$oU`5v=OCrR
zGH28$_WLWIeE1%DUbw!#hN|S8$az_rf`TxgRi2mkaV?9F-gB7u{O$+;lrOn2@mBm9
zoGV0E$!TV{yP}pEDTM&ID*UU<KjfqM4naJrrsKV$%Dx;(Wyc^7u#%clcU6jI!YNU5
zqN+oy6#{Eks1?<W)Wqb~3fR&ed_hye1y2YA-WmNuYsEXy#o5S18JSbz<63!e{(#9V
zV<=3UK&=bA>t~d-BlrQtv)OFejuWpxIOpu_5#N70qrfm&&1mut?;S`aYXqbr148t|
zfU+H4Igq~Y$$<#9j+8u5b2eDXLDqVNZ!jpQQ*vK+iLAe)MF!2<=_!=*G6PwTRL<jz
zx3?~<{pi`vG22-otkaK3Z-nao5w!mZqQ)dK8uDrOVNS}*3Ox-Lsv7p2Yq8HuSR-$P
zxL8>wa;qrR4@)VX>U;tTt@|#5Yp{Ye7r6GNoYsNFtq$t$$x&_n8>Jo{e<wlQIiGTC
z-|YI9w$WM(xz@GEN5l-tt*g<EzceQ(g2!O5PPdlYi|_h^Gu3ua^nc%7+99iQ{Mr6E
zY8h^w+aKi;5%$lY@cw(BaD7O8^z1-UE1>6PVOB?o1M~id<*;XW^^9;fAQDg<r9wXJ
zdGhYJa98h=b~~P3U6WEqOQEy?C3sI1hx_7VmhE+m^cC@nmc+BGr~K3U|KQJE{t7u2
zDoSvc;R#S{f}*%_;2Zi)sx#QAM}b-kl*IF8$2(Ww;i5jq4+CL3!}*b-&gy$*A*TZ{
zP^(yHkb+RCB@?{x_~8W)wr4nRPV(8aE3U4d;k;)UHvj~0M3kyZPMP2wRUBoW@nSs`
z;)I$7Rgac{)(9@(T_m`XVKBDUTnb(k@J2YP#Uiaz2+^LuRE&P@MX1T)RG4>HaB(lU
z$&dZIhekqSv-yC({FnZBY^JcnHa&ajwS-Tl;LGiQ_Pz$BfAls~^uotKm5TEzUv}sF
zPEBI3MW05$r$+l0wsdPbf`jKXe6m;m;dkNPOWXINg+ig!f;b_DfQsj*F5qweY~}Ok
z2C$X-()8dM1omb<P7Si_ehqheU@Lz2=@I_%B=qRM_3o_Sa{BGJ&%>?XpJthQ?7KM4
zs9ttFr%rp#Y5LE%>OjXmxee`|ZtcbGAGb?``h;(N>x%#Kzx)qA?M(dXE-{CCSY~bx
zi6tk(bk>ZU1t(7IesEX>pYnqE06ro(M3rTK!&SK^ppXvKCE+TFvz{yG99aT`ypGm_
zmQ3&gr6LaUoOoyWm^b9xJi0tHy6PB7t+6yl&opfKpnk%YJV$D!mJH%s?}VcquDRCA
z`=7kS?)o_r9cLF0@h)<++oR%$gQuiK%7rBvwi>+Ps5o(KrhzkgL=BH|gQp&5u0DC6
z_dj|9cxr{5GvXBI4VBHMFvNj4jEFN+w#+;7G84QbgvflzEQ@k^eIUe7h{H8O#2_*s
zKtc=j>Uyt=!WaVuBO&CH30M%8lCX*I@z$@LO3?h|u6uAG5W)>#eEXjh;~K928?$cP
zKwTjNvtkZaWvPY#_FedU--2&Eg{45&lf)%}#G?aL;Dc-U-q+#Z`)T;;hwy7}LllQd
zz_|Fmt`UHWYRh1Fnb~}`{-`BVn<M%4`}slz?>)|WYE3vF8OII5Pk{C71P9YJ;6uc_
zNXdl~3T`|jI8Ss+)IF)BCIwO%LqMFPRJ%vK7rc1JFj5;ls@~&(-~**9nUQZl`h+*e
zYc_M_0T;wM;1^*Cp7Zk!zg@pZsY<|8a<a@YY>XL|GkK__U{G5Ko-qj2f~y(vp3xgb
zG#{YmnM!4F9w`NtO7tTs9oX-$*)Win12rGm4uO&jb6U_+$VG8U$;q<Wcp;#zm!>MA
z^>GZr<D+Al9rIGy&wHF4@WB&8gbJx-LJY(hx!LcIk}7o}IzjPvPlzxDqqJY{D<8i9
zA^K-7{tdHaKis+(2unm+mcT#z7yl{qvOzT)y_OS;$+3v7t%F~+Lkx9gKDY+wcyV}l
z>X44bS?$=-%ht<M9e3LN(K3$88U=zer_?K=*2Lz-Xo5Ir@(aO{MZtS6_d7oP=p(*=
z{T)7DKH+=y8@#tq9E$C;bJ7?pZiQ(?m0Aqm5rL8mzn2pK^*>wKgu+|r!q1=m5<mUm
zXZhKSUtnZlUJkrpzsDEPzQhB6POX*D3i4~Wx1%*ZS`k!EPgx_tiJnCKpHc^?ou~j5
z0~}wTfUGZdhFgb-+MlYcajR<jNT=_&>vyaZ0fYt>N~_cKLlBM`rf`&N;nsb1vX2P0
zebDX^JDyq`uRnf=FaKNr=}V9ON3j&kdSVQe)+6O<O+w;P&8F=X7rghJZMURsSxaz%
zVI0{8hZIGZf~exck(-7%oU>wlkU)$O)KQwejSmw;oT$4Up2S>dp1)amw(NOd-^Ue)
zpFP`UWa}UD-r*C5@toisi+D^jMW}T_!eG}kRHkXeu!#)AK$<h-7>RKp4gv2S(>UOQ
zr)p)0fqBk|z!0nn%wSu4CgLz~A%UC|BFdnVf=6qi44Iq~qj!wa6XIwN(>bEs8p*&Z
zV-P%WJuj3}Knu}N_%Y(Mom+>)Rdd?89orZKDiPJnW(bVI)1hboX&tpcf$QEH2~~xw
z>z`qsw^Ys4(rKJ}`*~NTc9w%0{@C@Jz?$BibHvs?+>5`x_`dp$<3Da25?%jud`aqB
zd|e@{Du#<{Z!%N}YWOh@A0pxcD#B6<_&`$S>(9Q;H$QrxKX~?Ad{EvYxdrDP-c9EC
zg6+Srel>>9`?U*Nn+}*d1)mXj;PQ}ova5VezK?V7@x{&ecyoG-jh^w_H~)msPjB<9
zkN*;X=E1M<I6ktXWIwmk$Oa<j935*Rxx;^=M`hBNG0)7_)>U4i(BZ73@qY0$rw*a6
z4#W&ZX#s>8{Z79=-VeuGl!zGs*Y-`CZkVVx$JZ&oN5`YhG2AVCDS~=wfk|EWJ7-TE
z%DqOE-Qo4xlAx553AGTSL!EN@{F1zU3|cwcOyumTVF2eC2gi0A$oqg}z=_8%P)fy1
zB=~@j4y^@I;YL3MR|(!XrjgZ>hG8T`t1Y@JgmFacLamic3g6Fr%HdsdRyLd0@i7p^
zQx1Fbyu*74;f$PT(wG?XKuQ%A>+SI3h|ycHr;4<!$^;B6l}r?&RHO9v`X4F`!vs>O
zS{a9c^8uB>k}^^&DJ&q$6dc1~{U<7pF*wH2W0|=_t49t8mSrJ&$98*ej%Wv>nPoO=
zuXEn+I~Q>wBI5DRalVNR(NU`sqUU0}r9DDF@#}tBSW?zMZe{ndKIuJvlqShw5pUy)
zyPko77d^73dQ{p2uOnpN{$Zb!{>$~d+J6<>cI$$dY8!CQincvDG4aP^H*Q9#RJ3|N
zeD;Lze)4_3w*OcB;rs_&*5}~C4UT|yIQ!rUJ`iHSJCASPql-3*^kza``$U3^oJG%=
zG9_nHTF9mF&8x5Q&C5Sv2qQi?zQ6xA?;XC!Z+-mV@R#2DtNc5U{{m-zYL3thWzyR4
z-5$HQA~wg`(os{YU$7qa3s#A-ti$#GiN0RqlviUcy}i#Ig>$AxAkwaOG2N%`khEQs
zUh5KP^_$KrbC&eoVzj9?hMbetF;u<tc#GgMp-ww3$2y9K*7<zDOHfMAR5h_h9~`a3
zfhy!>Cd+~d1Shx{@tX^3+z<oQx=`za$0F}(uo~kKt*-%5=J`NQO3}lrYgS7^1kSb_
z{CI|l!;^7!hvo$kQgY1o9QBs9)$)W|Gs?ga2X^}_&`KD#0F<<Fvzd)Pn_)=>?_Fb!
zC31G=*qwq4;5^HcTLfhn#fN|oBj8x(#9_|NhlRW>_z(yoQgMtv076UwC(1Mp?YVD3
zO0nm%RHHG+VPGRh0}de&;>a>*Bd9d$ya^XdyAA;#eCs{4yRug6AM#EZVtfT(wRNu!
z<gQZ0+JkUft`-{1&?07$)tH-#>xuD_qwxc^tVVi&gBc|0Mf-IhPMwJK(AIwLj;Oug
zBBX9~jV|k*!}-WAWj?uj!dIXFA@5$k!`BYK&!s*yWnFJKYB!@!%m9iQj6}8V(phD4
zoLA7HB5mhGRdB&Fo#-8sQj1)bR10~@1m`JI`S$#6-aow0?>zk%{Ki}VJ$~lwXNb-N
z<>&}HX`!0-QQyZF^c~@&1`n<cg6({FxN6OqR@jK+FWc+AVGj|M(lQ?fF|j>u+F0xM
zTJ+f)NLOvD?o33Ej5*ixm8PFm^q4{3{ck@&T`w|?b!Pjccaqe5Spqp%ZgvYP7shEa
zBjyy;ASrQ`5IlmAN=AuzHxS|gbzwOi@U^V>de3f$A>y5*lnk`zp(q-_$ARc7<KT#~
zQqps@ByvfVl1VvH@ucL)E)l{RVROblUvap8N-40JMw}aPg9#AM`zvq*St{a%X^Nz3
z@Z2;wQd&sKP=ui=EYC|K<;*xvjN`!Il+F1CakHhQnf-hqFNwpvaC3FTG(^rWnqspL
z1aW8xgkYEP`Q<eN#2C;DIgKDfS~7FC0IWAHS8*ow6C-7*h!4bJG?c@8#0QF2N=gK$
zpc5sVPS{~iq?Ea5)A?h&?)fZqBOq(k-J{$)Z)>S!H@eLrwFW%t!(2KB1V_W8aLPk^
z*}7{i+^rW+KI>$?>FnY5%(e2pr{Cs#&)?%;Ui~)j9KOR+7gVrDCvW0|tC*u$<W)tQ
zG_|14I<cux{RyK;ntX_J)MB#^j2u!|nQI>-n-D0)>IhmDYPA&HM~Z{5Kl=tBU;XF&
z(t}^&ue|l^d|~^AmV#SHGb>ckB}Cfs3#~hw1~HVjTXF{uIz7MAvI)86sJmld8z^Z~
z6w*|NtHT8J+!4OA6lo14PQ_;El;e{q#Ij2J?$f^Z7p~*Ubzir*I_=Z_yx*0Ja=l-u
zYS5C?N=ey*bWy8cc#EWW`vnOb%94oU2o$^uV;l*ZP%UfoY0iZuXR0=(+v<=okxL~y
z#}GlZaCQ0IdQ3I5%nPY#yU%)-xPyFy_>pOd>~5~e^9?Cjky5-zT>w0Bn8?8q&LbD@
zjQw1iLHE>JSyBQqFj7mQEQuh(F!Z@51s@E06D>^BM9Gy@3Qto-%fhrB7>3XSjoK*u
zj(J|VzS-d=aDHajvglDBNK1iI@$FjY0^$Q96z0RhnjD=&TEJLUC`)45?bwU~-HwQN
zlvJ7L#Iht>XVp*qx>pBs>w!I)TXA<uGal-!+nCxS-Hy}K@$uWAmlJPteCM5d$B89u
zS?{_nOEhG~H>)b*1A=3zg?Fys<2OJ0=luTj-{mS_lSuYiXY$lw{SuTG;ZiLd@pjWY
zF}W3K^%`rMI*H`<q`?j&3@j}&L_z%V^%%o35vxiLxd~QUkM_sQ6P|qZkND={tNi&l
z{vv<n@n7a~d<@#U+(jDXV;%6TA=aLx95aNwSLjqDYyDl1<h8cU@(3-e%BrmFy5V-_
zMAV(Ym%RjX%Uc~^E6Cb?-gP(o4z3M?f5Z}%Tb@J_#Cycuw?Pm}RhCq(KOqFXb7%wW
za!pVQP8_vXQm#1fDNDBdG6sg>jP03cU}iq-$ax_I%YKt4Q<(*ZX@finjQS?N%=oZy
zeZ8<hD2_}{3rntmGl_NqA-62h`bvDufU_2~6k?2&T5;n!)8;XGIgq5n#<P`ks$N<n
zVXJ$2acuxpZe$K+2mx0!x+oeQH6^16t1=7|dJ#x&5B?b@mgtB>M4aQWUrdcxEaULb
z<1xE34v{!UQmLeNzr^58g2#LIrBbS+71YSZh$~u@nlo+;&BoaLkCJnn8RAjY8Vz49
z;JzKm+BkS*J-w{i>sUu^Pr22@Zaa&U(HW<kIIVEJv>ow?V69QIqag>{BIFgya$L^W
zeD&!c@c#Zo{^j-W@yY&UiWX>kIo1<dyT?HgocGp{<wMg1YFUX^a?REdsgAPh0vU`|
zTcNs3nRKlzGd7u$$_nXuZx4iTnnTe=iYjTDSJqXoh4-Gn$0vus!$-S!_)ot0KjtD_
zpjz96HaQ_(9pN1NvB7(HE77pd965RE$e8QSUXQ<%?wIMXH{~sNJ8xKr3Z2O6A8!cz
zyQ>DAUc-)SiTu${K)P1eEyws0FZm;0f)%_=0zY^Jw_a26Fdr5`8RAIrkuaT;@JPvI
z&CIDHOz631j02*DIyU;M_r&D!v!j&C_R(uxoqfbGd&b~NWy>NH1W(>SA+v!PED(wt
zoX)s<gA=<yRc!*`I0iIjL=##o%bb|DuYn&-hG?iPk~r*ll$`LvisgbM1VzN7cxvfl
zoox<Da0Dl4$!Kl!LnI)<b3i%N$Y!vnK>^dbN{M-1(AMh`Tp*{+GAFzjNHeWx5*H)U
z33yLdK~tqxM+}keIFcPKxw2$?PxMVbWr(=i>W?DKv$03L@8fJ=t$TJLx-w+e^)efO
z7nAN`@oi9J{dt95I>z54%oh`o9^@Xk)1qRJl3T5#I6wk_`0OkEy?6f(@9e%uqL6B{
ziV_Sm(M}z;EXlP9=%`zzR>*S&r@qw@+Ki2L4_afL_!UagR`@<1LNBx%#O0i8$Dov4
zFhz*Uk`@!vtJSEF&5-YR2fpIJ!Y{w}HGbvn&lA02<sGW(-(&wcp1)NqOK-zG0=FDb
zKibn@bZG1-xY+9<?dR%D#=I5xjiDp!aoTOt-rIGxPVSl7CK;aIK3+~f^C_#!snfi3
zQ{3kgT98n!Nv%z%bmHjLXBPv6iXTR{XOFqsw~3ayP!Q{7STc*Iwn;k(C(>rZIJ|hS
z^@dM&-=`ctp~T3|vZ1JTv@<@&`wc8RR1;dV333uZO6yett(g!Vp(IM)k!qo;XF1GB
zapd{HVes6z1CP^4h(<Lox!{vCMwYi{;Bc5JWoC$uvDGmELUh>7jREHq*N4mpJ4iZE
zv{2RAG*2}WM)08pB*HNG2D>!{hnzvZ1uJVNTq)V^tyGlg9AgYb4?b6$Xec8QM{pG%
zyy<6cMijyOLWus}W&eMi*S$%ZSb$!Sk!g=y>uC+Z{_1)?r}7-VkiAB;eIiwN62gkv
z-`aK$X8qBqL(4ianGC6>b)!K@4vKibx%)2v*LVLmAKbi;;9Ji^FK(OJbRzgx*p&*-
zoBV`#uE}-Sv6N!R?VYzlQpMm1?;Rn8wb-m}`=u00&er6W3tE&IB47@yw5fKb<Oabs
zZJ|=Fd8-xeYb}&qK$Ur!`R(Vw#ZQet#~blAd!Sm+NiVKfYGXJx=*o$)(zHv52%lCj
z=(8QfA06vjd~ZunvcG;4IM-zRYc<Sz#S*(?=^s8iBvH^6Mmt_lk$TIa+|8csm}yr}
zj($UhRmt-nSFNV~G{mrGpiT_Auhkp{)$SD+0~|~sQ5<Zy6X$0h!2$@WYVc2!UhgtE
zh%go<E(b(2S9`}U6{Kb>C;17)Q|9MrS#aaVu8lXKP@FfyLZwl1BTF(lqLOEH{}fj$
zd6_w+3(_V6YW0MYtuQS%#(8c9|5OT8tWd3r=E4vybCNcnywu8WF3dR)#|>2zOD^bK
zZ0ce&B;q_Vn#dyJn&6^6ueB`J90=5ss9GtEC?i!JOU{f^a1MrXAk2Y%_AHH7?Zhzz
z>s=A&tnLV(NxkgHb=_kIVg{~iZ8&zv{*O`WYRr!4_U&~FYn?{3>sP#(a&*kyj1C~K
z^$?gwlx@mUoTA=QfUiIQ2LJc>{wqG2KV}-YEK5SMNoVw8wjAZwakL3>c9S`0j-fUh
zpzj%nSog1Y#}TpC++@bOw6U_fo~t&UtrgXsGl%7di1VJD4U?s1Ex~(41C)y9YTH{9
z;*1vg?Wf=3AIsn8KYH_T@G#mO3RT~V{4L8t)&|E8DLH2vK&zvBN!s{A3EMR#-Ma~A
zOHH~2eIHk={km(*xEyQqXwNWjIiow7tgid;gGDjr?!SAc^r@ft(XFa>O=E~w_|3)C
zaz!B9I9F+Pkx~`3kPmy}5INfpY$r!RaUtUT$e<gf*d#$whlnS{V9j=7(BE#ipyCP9
z6T)PpZx5H0<r+a5!UcZVpf%%zYeU{LB9599F?bY5Rhvg%(>2}?#5jU@p6d;{L^OlW
zxjhq=B^Rr+sJ86Wk><=*J|MuuN1l?c8PPkz`+#!=Tp(^Qh}AmQ_q)AmQw>HKakaE;
zwABw=pO97)CKZ9`9ZQ9pO_!~RbW%E{D*L5UG~?Vrh>=n<K18A)$>qSD66a@IVhkJ_
zZ1!r|x>pFRbV{vaMTbAeS{==aTJBMt@Fg_U(`kcpvW*T~wg_4(E2F0<)Zn-|B)<CO
z>-@i-{ntEMJ_Z$*w2+E1TwDkQ7Yssb-SHuYCcSOPQ%d1*I9P_!peAd!3Iy-lgJGd~
z%E_{gQZ1WlG|V39_qvR8N^K1({g{256<(>R*t`U^qPei7#Ju05$)F9bm7EK^-HzYV
ze}>e`zyI1_=RrKj1*@z0;LUg;NY51Yc%DUiFUHA<ZR^wOSnEW^7k;Jdwi|zo4p6QT
z)|zSZQuTnluNbNA2y49<B_2V~V+P!UkY4S-_oZutq(e}tM#eB6DwKqGXVw$aW`EaZ
z!4DI`c|0}>cfVg4#|_)FGwLEN^Nvz&yTJuYT-fCuSC`LN4$8&jFR|I48QgPq$zgtm
zqu_l+)b2$YFYw-q-Q=C=Xpx~R)8-uUy}&%9`I5MK3n#){W)LB#YNRWyj#o-O4u@))
zV9Cld&kSP#bu3FF#E1(4*3kySzIX=poKaCKNntJloG?&UHetV2oL0*`a%C8U72Yvs
zTPcoGTCGimeRj;VG7OFoJXZ(H<`jnuk@;}L&GjDF>X}B*f8{{>x@Q`1XBlY|^E`#6
zjt~9XP<ZMv^!9tYtVNGl?fu*~Sk0?+gLJeT0!N!q7U9{=fq(SQ-{p6n|0d7$Dc%Q)
zD)V9yo%4>+W_^GXLpT|VbiyjrCe_i#bb?Te75#eT;=LyfVTJEhm6S836q_U6`a{&6
z0RLijj#4VMv=PCb-rF=2b1tUjx<8=VzS~Z;!Co+e!tXr&O)lqW{JFDV<>wy!qEREY
zG62z;K%O+i+WTnMI9<hx-+z?7?J(k+UA{PP<*wt=yRZMCNbe3jR@-at)`h;iI~GWE
z7^3GZe2Sy{Q7`8_A+*{UijP7ZyfwMyj1x}`lRXE;3eKBN;Cy?*7z4q#TA3=k99W#i
z)!qjxr45!UEM?&)UE{QJw%PFD;W>h{`(?hu)dRy6K|Djq?5;i{Y#!jx9^>LjsR?Cn
z*H6F=#O*|yZ$J~G2P8jd@Ei7eMb59|-2h>*kp=UD6EpNh%}ILit%1^cri}?Af+rP)
zT&>wrO`2z49s4S5vNCv^@18Kp5tRk;_Pm7X490GSbqn;wiuTUqT%~5C@OUq1bu2nU
zC=7$+W^wFOCDg=CIUs`~exQi5+s))$xV*{i4zKz^`nun<&>){4Wp+5|MO@*<^<r<p
z53nz?e&RUmoTJe|45PO>$(3EH{PwqhkAL{&?~<d96*bjQ&p3{21TF;*%S_Ih^YimY
zWh`rLr|*yt=A8_ZvKzX#UJX?kr-2FU7WSc2dDWn!wvV;gLzGKxHHiK|sJ-3<?bZ(R
zCiPN_bpYGU_2%Bhp>6V^aqi*k`>*oB{64#U;O8Fw+f36$DupPX=sjcWWHvnenAojU
z-R|tWgJ^lg%EmhNI~=e3i`VL8(&0)-Xi_OX^H^oKTIBe$E%oma+sQt2<^?#*o&4P)
z&(9|Luv(w9Nqt2ihJg?R%e+TIWQtqDIN?MYT*Zxnhi6+JoNXAQuo*p@=uwT7Z0tSf
zBhClnvamQM#=vz>MsM9b;OyaBI4>Ob2M+ri=EHL~gX8>c;IJr54Ggi6=cgRXLfkw6
zt<-GUY^^3y;zGoQ5m|1~n$dayP3&CZ8s)VRz<Wx2E}U2+-tMr+Z#GQhIaLatV!a;@
z#EVS}6ld8zm5VX3%z-jT_H{v=1yassM6Ca099pkfO-39Dh7vfRjm4<05eh9B9D+x@
zkg_K&j+85htIE={9#Mm~mTY}9zLEQ$-z+@6T9{Lx!~SaBs{^rNgeJ$_4T9n1M_6Hf
z;H8&3o8!0xF?789*1)A!SUu84S?$AV_Mxq&P*CQQ`S9awzIF926RvX>HRW1C+bn1-
z!bvr;%nRa#C8Z{OD3so@-u`&!o6{)PwW=nS5r^379mYzLoG4n>Zgd}dL~IX9uYr`R
zMxSg2Zs&w?8V%aWc`ee{2DJ@wAny<0TD`})Mon&zn)6H{FxQ#?X7~5`cK#;6dj9Y7
z+V}?11;!XDz!0O&TR`!hk#*89a_TTnJcV5uJtb@RU%lu+N@G)1nqyi$&m9NU{!@Pt
ze9-GFZ9jU#)bF9S*x~HEM4iEytu}V=O9T~%Vj`NA5dftUrVVH3kBOtz(WYpb>ucv5
zhS8y-)~_K(ldaAq@||<5;Ao72;2pCp`0d1N+jAhY-(91%)nDR3)rBS5br{D8?g4tZ
zX4yX@EjvQoB2HV$(NUHKbqnIcTCnzhK=ABz&pX>o-l}g<#R}db3Q8ozO3ay3EAz6f
z_j*>yOKNl1Jz6tMPNdTMhk&`57-R@yEJ7+d@5xR`WyGm54?{GQX4(32I9qLhLU6d?
z);Tq~R`$v63-2L%W%QoNg!dWI1!;Ar=?osaiSvtwqTK%-^JBL@lRQk(WBvOjZoIRw
zPT5)d0uFwRXn!!$p|_bs3RJ~;VXlSUeBj&fe4BUkx2cV;=)64?ZfzJ+=EH&gZci<h
zX}dAU(I6`4ZH9I4Ul43ofz4BhYe%tfvJ|yiLseD9bEM|7ZsUm8_a!2Yp4sanp0q6F
zoXIISiel#M{G75ZxWT5W)s~r9jidD$tPCi}LgMT5SGb(7_;<oz<*l<XaDIM9R#VCy
zgC~eNXRAY<81e2rI-GQ3@%02+Jh@~q{wPO-uXo=gR--u0?)o)-*+tiS=i_vra*{<U
z4ex3*(feGi&!RADVRs9Ytmr+@3%lJ8@e%c&s@Bmx1=vU+dUH5HV=VcWeI7GH)oS$v
zJd3Y<a&bfS=PZXrUM@}HKn8+$Y|k#p%Zvn*m&w(m1ghI0evb<SetcvCU3I1|*C3vp
z7OD#3Bi;qvu%YIJ#K?QY`~1xAEgt!i7#u|m?wux&_mO$tb66IF_hf}su)$YD#K#fw
z9#?Fp`94)L4ha#|=7Gd$1ei20%=^Su$qWMXybxn04wluY#b9HN5aMKvNuZ=k$}ks$
zcfEJUNYsq4g7ez;$6He{8Rr7gSG+UXoL9@%{X3B63|76PJ|$0j=kiND%FB#Vb0B}5
z=f2CQz|r`LbD+Y9pM1>MzWYr+eE1$U_IDUHww1HVC386JQLT*I3D-NgyJKpU!WhDe
zzs3+7rO+A)?Qhpa+rTa9XyK$un24<1=GIKu;Hzo{`;-mW?%A7Z^0a7Lg!A)r#$jLw
zY(+R64wPh6T^~Z5cTkWJ8<)MwQ)}S;`VNPif5?}1zrauLexA1;z0QSBR1IumY_#CY
zDy&#f4OY8I9J7qt{@lG{FI?LGe$^G`>Qvp6bYi^+R_F19ebe!^*1OXohb{<7m&oyY
zR0RGEmZ~+6dKVamiJWJ4H#bb@4@tF>OT`InqhksK;}p>lQMICb&bbi-jsZy3gj3HF
zgx?Kc<-5D@Q`3T$h2RBaQE5X69br7f$4JSEJRd0REQ=Y9ku{#bVFS%HQ^G{e)<~!|
z1D3^vFi>j3hb^;%@8=JBxcn4Fg(X8inH(CLib8OKc}ZwhGLE7)rUCJ6$B}JXVDY%Y
z<7HqdncyqlTX40NfwCyeViP@`Gev<~Kxtx*O=Y0g%Dg15XJsh?MOH;{)q<vsI%z#s
zx!vE2$Jk~Ig*hdzb_-WG2ULew5Eh?-0}+l&sXdx?(#fGm29mvDMb=-i5HF~$e#&yK
zM?%~7y>mv_72(PAXZ*n*ewC-o6L7|qDy5zndy_QK1mifZ&`c@SRBpX%G*q$1so^-X
z*SGGdwdn6H%P=GC2+{cRwU`k|DK!|&oKQ+>wV$<7gVEzd*qK`~4xnmM7~X0maWH9&
zIWKq#YaK}2I9Z#=;aVDQ2G-$t65i)&{g~^UD{}g?y#DYJ4==U|8^E>`njqvlSrR>Q
zB1b1tg%=C$FJV(1vzfM?HXCDwq3lg<+pPb1SrPwkmMZ=Ht5T(@4K!D>Dq&TD9?;zT
z(&o_03O#Abgb<j<Gd$jUm4udgSS`zmr@DZWSLt;db70q{=7Kx(d~NzBA1v?SQ(~ES
z%~=Iw^=ZYq;Rx$ubHJ5^lXma>fFHNenhDdP;Cuve=rSX;pr;>4t)wNfxj5tf%_scy
z@;bvjvTqdhQfx4kaHQJ~-jix&cgTbR;^bJCj0-|^M&fV+9sw-~l~T;{I>jjMF(Ne^
z>nujv<PJMF=O_p{XKKlaI-(C$C#YA#VP-BFhpsioQiLSRKw$``1wGG+oD0qcJN{Q$
z2ym|s#3Jn@bE)IEfY?K!z24wX8wpv9;cEf6y>=He>F<SBD_!d<Ze=Rrpg3ir^0n`M
zpYJ_+&eg-GEPhYOgKeuqbOYXb4)ejHyp|zVt(Lve`XszfU~{cUVwyJOHoCJ9U9!Fk
z*4~bFhfj>)sf>rJl5<`)uZB1RLP`tZ&}uc6&>GMbOJtmP>~=e1j8v=`>Ab88#Ljup
zCiEyr7MD533hJnu`KJC$p6@^5=kuTC%jwJTnx|BlLNsZXR!r!u8t=XBzirm8iCEY7
zt)(w^kVwCF{ee_zwfofsKm%>M)9OZ#6USxmlA~ieIlXJIy;!6!C*Yv;0b-wl0}+gJ
zDvjMY&olG9aB=>C?e>9*{<Y>#?*p|uk{SiQXd$RFh)`>``?5L1>A}Ev-uNDu`D0!m
zx1`bWN!oLHz2Ft5akP4vfaZjj8Bko@GHfr%^8wXt#ffP1O0+#d4se6jYS4-<R?P1A
zwKYI0H*Vp(?tQ*Ie-RZUb(DIftk+!eA>lpbc_CM2yWIjd7iCf3q+S-N*8kLt*<D#z
z#4!!g=Ac*8z;|L|p`8XU-kK`C_e@R*qc^o-HKugU4nGZc3@A&^1UzS>aK3>dv}dxy
z08DtoI3RZ$oPI*r{muhv4|qTIjy8A0JHtI&4@7_XPcx4r%jf7+;6=wF?Y&(Jyf*79
z9t5R|NkqK={s(;RJKtpvmBZ!+=UT^jtvHD+^Gt}bu{tbU(At^_rSA)b7zr^_vr%;s
z2t&lh(2CkeZ7Z-StTc+Ry#~WD7!#$#VF=?m+Jn{{igRKV-{v%Ho1j)zP42x`_J>0o
zZEE!6VVasO%CXVVIln3d=Txi?u=dknv!-?7UHKNfdWBq?r2Lz&J%+(iJ&Z=dUm0Mv
zIh5mHt+@WOUZ0m_4(nmpWA&%QbRDhIQ^(u>zFTLhrxFuuZN>iMRGLOu%f?Rqx0PD%
zfAx$@93lgyIGjIYdv?w=Y{|8-9cP?)l0=F`qB<x-twJpcB$=_tf~@K*UkzX3J0E`=
zr!*l?K$NsBNDJc7N+p>SD2beA2zy)@@lIMMSV;Tlc=1Tw0*;h+1Zm=v-Xn2?czb`v
zAuhDX!}Go2JsuBdJScCFuo>;W&~MK{MsU;y!#QWgX(yfv9F{`O)#ky7Aq1kyWP_Z+
z!GjMB-lJZSV$Y7Yu5@vZDuT-eae~L{c~xOh;kp!}2ywJQRHbKf)F!W@rl(bDft+P7
zt_h{STA!gBFSPsGofL>UI*QsM5<M{yr-f8l$C>tRO5dg&F=5wbTRZM8!jdyz|He1?
z<l|3R&NI9C6fHJKI);Fcfq7n-lXU{qvJWkVRW%nS#G%o2Ju$}B;HTEWce>lEUB%O;
zo_X)gA&4{?_#?z+`|X;1Nf%bM&T^|~`&-d>S`sNOYZrV<rV_l}Y^;!26%`vg6oN^p
zlqQXzQfe8DoHR%pM)Xua;&=3)u)DnG>hRNi@%1n8;C$OyK631($i<orPYn7?L;>zz
zLe`sLeO;R-)Cn|3&pdRM>8jGdM{jTNjvUERhpS~L>rgF@&g3&((y}5ikr+3)xTUHC
zMh*+a!Lb>Q6)1j#laV>i<Ru{lMsMPrs-+!NWmgw|^YZsOIGd*tqvOHFrqO(@5zpp?
zh*?Gfuz*Gnv&~C)HgdKUMa5H9a1(@K1V3Q&$E(q<U2KZg7SxD~t$+{9ApF6@Z}2NO
zj*I0XxiO(=GjH}qO%z>hr0wWH`Vc2EEw8~f$sPrD7EHxBSODXELq%H<Bti@Zw-vR<
zWs_*P`=mA-(4i`W4gu_*ABI3wkJf}E*$nvLiP13*28`XL^clgOKc4HJ9mt9970_vy
zGy|vJuZE*rpN^5QeFV)g-L+3E-pz?#Q)4ZyqfbS^dp`Q`W8QuDU7Yh=4c{Z@3@QZg
znWkw~#jRS=)I<z}Cz!6o%FqhI<DEBr)m37M%>kT3D|U>1z*LNaZ-y>(zPjtZ5WT0v
zm^3M;6}l^>u8zb>Zl}*Ox||9$4A+Sxm&_0b1XJzxu`zWTMl6`Xafp%JB0(D#*C@Ys
z!Vw8Pt)KAE^Z%QlNxw)*e~q7d>kGJtTc#lzn@vu1yqa}hNm6$pte1_W?~UNy4xBoV
zQ@HPBOv%wP+`Yf<ynbv*#O^P(`>W6}_0QCQ*j$_K^oViu0Ouxhj-(V>#F1-eh?&>V
zt>4GFNU0tqGVc$FQ~XpJe89UvPBXbwHZt;a;Y<9>@>RP(s&aEP+kM&8m21wH^$R$x
z9wO3+8mWMq;;a*4o@eT^Fl5C&d=u$%J&gdO2;wZu@j|1FTVUZNaF`Z;U%tXGU;Zo)
z^)V$2bIL1h=|bRQyD>emQrYizBoT&jB1Xr?3n|Zdfx%US)l^M_AM=4rQGBBlYblK0
zYMI%JB~UC|FV!Gd@r@dfB3{k8RzOWl%UD^mwhYB!bm*KD!8u;3xgXd4IgpM;tk&XW
zwVUxzwz-$!>1#7dcObM(#EdiElWXNW-}^q#uWuNqiE?(Z(Unzf(wiDHEO=+dI;*_m
zYNcphx9J<?=0j++E|ip(7KI&!2m7(-+-e2tTH{oM*lKMGqjpkU8%AYCPLpIuu|~iK
zXU)kR%Nj|Y{5v&DFWwpsLzkxZ(K4K-z13Z`4}qFZ>Lk~C0v_4TVMW8jJA9K`3Qs@$
zCH}1b3U9vgnCOhna~uBYxxukJe_d5yy!ITk58&u1PCj!|);NJMk4>TLr+-jjsG!Zc
zNrO>)#`;+-JNE<A^bqkA%i=lALaDB0y^iP{R8y3#Igt+80+~1=K}>EvDC6b~pAN|W
zhS3lFoP3$DUVn?Z(NS}@Mu!!e606TyO;8NRYR#KD8}kba;~1KpiD8!7gQM2rt~psz
z-3M|mYet&0i8WF(Q5^G-_*Z9N<<C6-Sz>+E=FeFEH;f}OMzm(0Ji8_<_;JhGxjFs3
z9AHt#Nf;d5q=f>U`3;Jr79+KYD<)&&zz>luLh1^~90$9lqBsn$>dMuu&!QF6X?<!%
zl^~ECF+o*`F%rB3%4c!g{MfB~XJK_%V$B$uk?qh>LDx`K*O2r`v+S^gDbBXF!V$+C
za~;amlte2Q9X<KvDc^th1D3gPb@r4iK0%jE41=*cFk!^&{ehZGE0`PiTb&Jo!ihbA
z&1i>S-RsK2RjqiZ20Mii+6}D7n^tI4+M}bf=BML$*WL_O8#AoeVq%QsoY$<yd4s-;
z(Usa{I)NB0-d1yJ&N<V}q8ZUuu+f27ZKj<lL?F`!^)xoxyY-tq$?s7<PW%P`YrOIJ
z5wV>EG1$3H95mLG6g_cAtd!bS!)z^f_Uni_Up2;5723HzLRj?7K~J1W&we=8ayp#W
zoq4;-*89UUNlrAk?%$QT7QvUIEQ^wIBZ^dm8%y$}C6E_5>=u^A&SyA#NZf4LUp=QP
zJbddhlM~YJh8P0Fa>47Dzsq-dXN8Nrw|lMDN(QSp3DBDL`rH{GM+!DXX<3T(FAaj0
zg?zX{TtLfguu@5kTbm=ISpBe8Yu>9lk(U{*j&X1-TgUI~xA>BLfpBw9x)vl0JKd8G
zdxA3uUvQNDOiqQlRt|?ut)3a7Bv_V8aNtwHX{DA#&WlNvgaMq#FM<mrctO*_p2KRo
zc1y<kL>go))$T*@?QSyJ79m%NaU@I;E!ncBVlzWtsqr4y{dS^vd|ghp4Nsg%d#zVh
zd(oW}XVJy*grih@&o-R%x1#<|0INV$zqvI@>cEl;?|twwm)AF3UEgpyJYmTPToK%0
zG}-7p%d#Ac_N6pgbA@U~S7N}0NKR&~6Rhx&2n<cqouUkzjT!FR;Gvc}8?9JYCuB8}
zqX1*hncUQQq0Km`#Xi}bg=?y%wOTEsYVRGfMn;3RENkg=O5EyfwV`XxaD+(BnOrhx
zZHC<T)8tVCF-**9<~P$n=A#eZ;cxi=n71Fi)h5{WD@arJWxWpTS|Khy`T6!eqV4xx
zOtPNNdVy7&>zI|8va0knNLz!Tw0osH=##8;-PQ?qb{ueff6E9Bfn1u?@HQu1i{L^8
z7sz#{6v)|X5dhNYy-Ttg8O2R(wr32RM@&ORa^^1_{xa`9`GA79T0uoy2V5@267qIW
z6cisO&L6&o`p75md>d&nX(@$rxCVB(oN!@6wU{$^uJuybDfPiK1V;>xr3jtw+KbbB
zzwq^kU*qi2$Z(h#ug)2-F7R_;3S;XhF^9C2LTUsQ#Tx%o#)|F&hsCnxyd22$%rH$1
zA&`sX*+CdL1Fp=FGJ<l@Y)ZO#n=pv=fi?CqC8u_Oc&dw}xS@o}?(KTrm{>|Kx-$1(
zRYLB4pp+Rbn@wyk<n)<djl$_e&dI}m^0Px%qtT0z(!OMMO05P`+HCP$;k^$(<lPTG
zV95*5&fc|Zn-6Pdl&e8GVF)cES87TYJ%@mcv9ZN$AGH@E8)bGHr&XPIv)LG>R<#Xz
z$~F%~>#bs5cOHsbm#{WWmi7QQw$|xw>w{m14=p)a9jBc{Z=F__Ola$wsX$X&_1?8z
zwrnBS(&m%c{O#HRYNf$ufER;6hB%V)!dK^iMg2JQKRy3nFgTmnV9=1>$^shhM6U&%
z_G-vWn>pI=+WB1@3fr+A%OZW7%~@!J&E)o(x+R4p{r9K$QI}EqBW~V5?WM{zO?amq
zuJ=^+gkhvOho~d?fnoH-F#*E7I4V=CXE`KSLfqnfWKN!`Jp(QTY8Bpo_zQe)ew*)J
ze}^ao<N2G^Wk;HCY@%jWYPA65;e$80c=!f7PCR?^6s?(YFgSNfGujvtnr*JR)ybMV
zc7sJl)nr0ion;sYYAK|7hY*N^$Hj?yNF2g}<@~_@!84)`jLVkg`kdkBf(QB<UIz=@
zYTG6b=ZtAsH4}$G3?o_#r6dYQn;%**ahNmJ6=Ep_oS_iebl?!*vhPAp#YiRulO>|{
zf;gdWw<HGAJTv7=XwybNQ{3~%aNVB+X*j(X<8<RrxcGACs>t!L<9u*gw`0PFZC?7x
zdtUfTSflS+mEB?Hz4zW{Su%CAP@Rddx!}oa9lwW^OfJMZBzmM)3O4A5-jt{|@TOBF
zU5o6!3oDyxv)MGhyJ5p6CrT=oCDdYe&q*67)gT0I6Z)!YYn5UU+7#n@JXTE4Ez8KH
zTGs4ih_SJw+*%_6aI_|~+OZwM`EFRPp>U-OuS&efSsknVN>0YPcb=&5&8x5S|NP)@
z^WT5-f5g@e2BV1{=a|%^P@)-}V-R8cbu{G1+oWY4{XI@Vd*ha_y(OpiUR0>-X5@uu
zyx&JJyw?7$^$45Z*HT*lP_P~v(L~ulqcY;c86rw-^UaIG)vj<zneEmiQVC8un;tN3
zE(mcT#)0UCAr9=89fzIrs~7(s52nxav3pLAk&+T)*`tR`_WK3z9cSAMzVP<j<O&}@
zy=MRXDKU7qn+NRnS9TA0Pib?2OD>2v7NOF6Br1pnL8VxxBi`zRl+a|^dmLdpbf)m2
zNGzoBlI!sq{x#3DykXis!ase;X8!<J0=XQ}I3ifUwagiCcpq_bgOAGg>;XUbi$6yk
zN51jhkGQ<NY6`%$DH$sg0^q5QU3E52gk@%4D!PyOVFZCCCwwi8+dx?g(QgPEaaz!p
zX}wyu?ll83=iw25lrd*DQ74GTO20fp5Yin^FOD}`cp71@D2ARexhjZv>mV6o;Ns#D
z@1A{384}Tt_7E$$;0=;W$%<ysI?la891ZfY8iN@(=e#+`liB7&j8@QY&Lb@gr7F%F
zTZZ;oY2D>uSwXE8r;RqOmFnCoCRegih^^0~uFya)?0aX%CWg=;Ah&j}uZ|?NeQ&Ro
zQrjGEA?1u>4T|P4bQPjo2?@PZx7LQ$!`H69!bh)t$j`W+He=#2NU5@_0pBvs4Hmnl
zO?J9^wos2SlsSh!1gzCpUh>iIp*qZIpI-A2d#yvGuHRxWI;8trwHBPKOp}LNSq?L0
zIhX|dLD);?a7ffDEZOwfysK=+5hXB;8@vnbE@y&z-hAtvah!O5dCmSXb6y|v=iIOJ
zd)wdRefI$pCxUlOV_>@#Hq(G}kuVHgee#UUryt;5Vte)&L@0S#3**N}G9Z2=OdEqb
z#n}KR-|y$fBrJuR3&ZxD!8dw%O=umcsZbZA(Dv)1D4B(W%sksZ!*3#QEMMT~!=L9-
ze4C_7D#}un<*-oGf)5ULj`R?|@Yb9BnP2_|hB4y8hOd47yUhF2vTS<}A3l7Di09eU
zOC-Yf{ERR=%H@J{XKc?OaY%buY^FP9W(+XO29M%BxWVDVJD<Vb^5ePgw}>uS=Ft(R
z?i^Nl>S!2_^$^ydEC|)(FIT!?tCf@vo%Co|5Oq|9;-ChHn?{x~bF+Dds9WbWE7oY$
zn;D&RHVzhx+@-@p9b2l3=0c`2jU&-Vf)7L=Nx9&v^?g`#pGljPS{p8039VjZk!V<>
z|5OqwB|<Q1lrD49DZHJ1<{Pz8b3(-E#L;_#^9G4^`FG4nRyE3PX;UGkR!VKKSBu3>
zq183#Zc^B)*5Phc-B#Fc*R3inOQxtcAxhRUs=e33I?G!f2M}B*y{uHuTdU2zm6M5&
zg02><-vj-&(8kho2<3zhW*Od5n6s~{R=Dr<VFZda^2R+c5#{kifg;>o6{4?1$t?3F
zhqNOf&Uk)ifdbAUQqZ!X*#i-A%FXpP%b~D)K2plUqsM3Lb_>CcTwY&t$k&u9<HrF_
z#RL?m3&vp}c+b3KyJoHuy%OUEaoBQo`JO!kRVlgM|1KcI20xtPrx9vFv-OlXag>x9
z+O*74K#R?IkCV+tmkc_J1wMI47k-S<NGY|~JgSm4@qPCVo`)y=h1Y+bzx>62pEu6F
zKx)rJ?yA07nR6kC^8QC3vAek;&y|1s7rsmg5WLk;oH#DdH^dltet8KfaTs{|e8;=*
zUo$6|rj4m=7iCF%cIiNjBj;lz1`p10y)VS=Bg9?tstR)7o#;K2u%4XuqWyJr7;@qO
zjs?{h)=8GKV(h2yYugZKhAxK4)yGeG`s^8ng?YHdWIb%=v>D4LzuqEbCqhXX*XEX&
z(=n=IQTs3qYr5`&XGs>dg%Fqzv&}UKfi8IHTL*1y%&neXlxl3NTr$HrG7h6<1<e^4
z3}&-OqHmJbYLrCht%s!>%Tu_|iI~$<>jdt^f|Qeq__axE&iUrx?3_9ESShDLNnX&R
zs2X#qkCg2&+aG7Se!bfDspxU6@bSW(JePX+H*0-m4JP_){pXh1c1431dXG-Gr%F|6
zeK+@P3=}rgmM{O*8+`utOTPR4Cw%--A{SWFf|f+ZlM}5^DO>-6D)YRs%$^5+OYq9|
za?N4CLn&Ne?FrtDA^Mn*Z<rn_7jfe7{3$n2Kjvb1o8Zj>i8yku48z2OM_)h`F0bAr
zrM<zWwc>^|Oef2OADcL#i6Iu(Jh3zQ>Tp=tZYR9+6j$0vgGNMgjU*CB@^YY*jBD!C
zuF>V!MqO1(f+zFG{Joq18-MWRxA^aT={NX=M}Lv18{#<P1rFCUNFkTR^XJb<^TOjt
zukql~1A=#qQC6ftscgr{<45OhSXbckwF9rcw&!YJm=D$<dA1qYr@|-CZ4%+dc4Ua2
zL&^--mCt?t5|LL~2;92^SrI5!Y(~5O?Xlh}^~5OD+G-DXhih|m9;*{+t3fBU)VO1?
z@4o*oS65eDzVRM@2;?PEi)97oG}`xwNkEL-5gjK|O2*6*EE40mUDro&ddHFyltKu`
z<SNBd=i2_(wx2OXW2BYZps7kH0c@r%BA9f1kF*i!;)!9TE4a!@rXh_ctc@yLYqd_?
zMioYBGV-+<ByCQ>9$>}vmT1u$-lBKchyv!LTgKP=QxsOnkKvEL{)(*X0p*sXc=4St
zeRT!WyJOS!KGAlJaC9Q4o4Gy7@^0P5!KXNxAJtM67dDI+Z}aw-zR2g_zT!LId6(~f
z{|PtWdJph??v22k4=ce{Mz6T&NXx>0UwQcYxk+mbo;Zx;c_tkeo*gp7;5j>w3}GTn
zo?H^s1iQnYAvlI%WVbJPZ{6mWg?o0pE0)7EmSxA1jJOfT3&Q4ga#mCVK5Pu$O4Vqx
zRjE|`;M@J2$we8W2TkCEWzM-0ynu7~VZ_aWoYI=n_fhcB*qT*W7?cX{U4ECp{gwZM
zx2M0)FAjg6pLz7F$oM)>uCAF6dp6M#d}Mn*Q6=)^Dv?VeiZ<$fWOq0aDQq@l6F(iT
zb{HleKMGu39}um)@o2-GAx;;hYPHcAJcP_4H9^H!m2}`UU}4d)^K~GL9t}YMA1AL@
z!Rioq9`HWR%|6>|D&;oRv1TB^)zu~6``-7#DNER!$Y6t|z&gv#nHZ!LhiH-iG_-jh
zfmDp@m;0carhcnV*v~Vy7CcAa*d1o4;wskFnwu<pFP!&Dgk3o}r#6;TWNmm_kFlC0
zML$oz(TDpS5u?zy{tgxEcWRA~{R!=7sC3LpWSw{@y`F<R8e7*GVoM_Bj1Qst?V}b}
zpHSH9@3-K{<MnS+`MriAji!5Nxpwl_WTh9l+Xj0^b>c5<%x<T3cPG;Bs=ihy(pYq=
zx(+71Z*IScaD8*l2UkzI+HKhk0}md5f%9jc5T39sGuu&l>-ESuIzmuxt`F>PcAP(W
z$mVR*41eML@gs6t$orZ3FjHK_No0tTgHJ5AGEw>b=N{q1n;iCmoEJ*bCZ=gjs@?Vb
z<YEp$ag5ssOy_Swn9!wI23#_zN3~MYY&|<|0;hxkxv(5|)NB$Uwfc5%30iWKC=q-Z
zQ1&)nps*@3<eXb0trDCgdXot1M4R`P@AJX@4&S`~U4C);OPr0ri0VMDj%gea4^;-L
zdc+5oqL6Kf++lGfg+s-N2t&qu&t?$BSA4Xr=4!U*CkD&TQnA`(QDrF$yTgo5uiz{A
zEE)+}Yn_0uPUTcuTvopO3BGj-S;;X&>F<15XL>i{xnw?h_LQejp0SUY?6=SC#vCo1
zFlh!GnAE?cH}HwF!cBsW+^pGV2XrgB%*)#2GCEIQ_e|T2f-d9U)QD?kS+dnKytTcm
zK*=q$z#gcgM|Kodals!4PPJWKXOpz<UWF5d-ChXi=s?6$`jVBL3mTeGT`TGbQ=L39
zsAh<38}L(Vp=5i1Me&?`S95wjFF18qD^<2_+tHiXUR*2XwqM7S8ktImI{Gzk8$Hh9
zx^;f;-W1(Xq1ZK~CAas?!V+LkSIk;TheupL2ayfu4<7N$zx?NT_WT()H&3~|u`#Sh
z@`#*2ddwSdJ>v1>i#AgE8XqHZ*y7H%Mt=o18%Na2H_CTd+=2Au5+@a{kr>}tpCPRV
zX?o{DtWFn(4H72&a7Nr-Al{SfLP?2|GSvx!4PbJ?418J=VQ7rPd;m8R<3)pq?EqXm
z*%ov(5%{XcfL!O`um&f6G%YBlz~NvW^`%z6=D)>v4&UWx!_V_ekN!Fj&wrU|vtb-P
z`$J|))_a0egWH1VjW;%kLTdd~y<vB$!kiM@F|v)3=ktMGGA*qAl8DYmU(ab_q2h*P
zF!qza?l}Wd!_mDqM$G9i`+Mx~-te}vWc1WwtQ#~u{w0t{hdVN6TP`mzx!&DywS7u1
z87KZ&D`-V*up2k$Om39d;5<1AC8gFA(L&x<<jsw*UQ4!)T(QEVp;d~HejQKS$K$4a
z6fi73J4r_xZHR$f?7L#HuJfXKtvlH1YK$ytOt2=uVBPCBt3+Fd6I$1;)@m%RTF5m~
zlg%y(gOIVVV;9!6u1#Dth(}sen1ZRaR*<UrU{sn_YTdU6#5<Qn?V{4@w2C>+ml7i7
zBzS1>romEow|NR@$qQ3y1-W%yyYmz$pw@iYVM2FTcI6&cE|pXdOv9d=>pe@J7={bp
zdi!-g|5Kmi>C+t-7tffs-{<o`{WI+L3Zh(GOoTXsbDW)>nMh`Ih!2)kmx3#Mo?d>7
z?{B}2501l4;mJoI5W>XgKKHo6kA2*L?MroA+^```TU_fgftIOx%aBzI)YP&*n$V%*
z)KSv`%7WGew8?_5JzLtw2hdc5YwLGv_f$>=g54P_lvOJ!XL45dhs>M>#KayqgD3mI
zAIdwtfAwGSEAC@{W%vzT4(#@cDoyXq*(e9+OhTx({YbT%l)iTqtsF|`uuohkGl_Gl
z>{DW13ZwTNs%2m5B5@2a+OMC`b&n22SAjz5qBR^HLytHGY$_T3q370xS_c`ObQ|}q
zL{7Apjt<8mPT0?TmYlfSet;^Zl1+j{jbWlqd#X3*r6r-IrLWDAxfAX?rL_4aW<*O-
zpthzzgHwD6O_{eFCsWEj9A@i|ZxB#$HnYHLLuy(?YOq?T>MCOUB64!hjJ|tX4Jfrx
zQPZg^g%}5h!G=z0gUWJVTBc{>bQ-(~;8k4*6O~fX(wr2gGbG+2)vb<Hv?4C7Qx^M@
zlmDKUIt*z#1_JK%eYZX@ocNDVxtm(3x?+_W>kjA?tF@SZ>vJy#AtHj@w?dwnB>TJX
zJmI5vKjLgV^5*MrakG2Q^Q&vbM~3ksoAcKg#xov0ev5HTWIdo%ZuSdtuwdhC8d~?a
z5fO9F)SIV#(|?owxFALF(*+OS@XY&}+(a;YL#z{FcgU#UV)K%M!8wgl1$vzBS8W!$
zsrpuGK9Ebo`GAwAUOR5_!^XPmYr;7j3gw(N@O5=u?*lcLHg;L9ze2@S7Q7KPO3v)|
zmCVTI!6V}4oDht9Top(+T+L7U|6KoF#E<;qgMXhytHIDj*V{d}E16t#>&2?34MuCE
zwG+Wo3)3*#?2srptjB5_18xlL7vG@f`@LO$T-QB{DR$w&*714zyA!;+I}!U#hbvC{
zI@-&%y-`{=a0|m;i>;=2G^fITH<K@Rq_E%`i>TJ)Lxh@4iT1%$-SVk5_k|d(;jk2|
zD>RrbI1g=(2$<4qtzC*kSyDc-l*BR5vxycq<K~06X?Z<MP;HQ!kW!{-!8d2KI*%4f
zgCN0M<6*Ywt}0Em5kiPe!)TdLJIGpF-9ziz|Dv^0NYvUhu`V>z`TKS4Fz>^$3{x>}
zuvK>bCD&ZB=0CR@b3G9<l$ZFA_M6rJ2rtuyH#n=ke!}cJx(q%3x7|acFSw7@(bRmu
z;&`gc6bBxiKV}G(w?B8mI7Ytr?hRKrHZ)0i!ss1iEPUa$z{e8EC315(Fb<BN`uvu&
z?ZlETqJ9uO7n_0kYRB(?_mU6CCjd=dK#(xvr@)*ln>ODf1Yuc}T`KrtLqDH=@K&du
zTe`i+dBw+(H1DWsXMv75s2Mkm_&7p?ZB?s@h&q!?siF+gu5BsB==C_XLCC4q{%Tc%
z3NBhrFjZmm@O91}K4yDyMhwOrM1<?-${{UODg5@;KPD^>_}S0@YJ;b3L~Y6yrI1Tz
zPKjFDXayH=RHAG5lJp6SZQe~4#u)J8c({ozkaGrUzj7eJy=!4{G&XwrD0f)s2v-Tm
zETW>v%%c~{Emdhosx}7IU29~<VSm{3?CCRZ<dPB+p2@P1Wnmb{69Z_ew)3807)ZK+
zwodaVAC1zg<R<Rt8YD$ZXb6;MBz<!>70ZmQ=OZzvWT|vDI3h@B-QOMe7o_P|A?5@+
zl-C;qrNKG6ruzb}>4i0VqKGgJ6QN}t+G|Xz^m*vcpG=Z#Y^t1cbDDNJ6dL`vvWg%&
zGyYC(L~uDRa^E`rj;iH|OZFNN$G5)}N|ryy^XZO_qibDky>EU1M19AJNDEHxW-9lw
z#$n>kHy`lUn_Dg}0y$-VDr|Y}prpAl?`K|rZQ-?tj$iqO4PXE6%=IDgx;(-K;ru*u
zwvFuPOx`U_5x(?=3(~{LKY8{Mb6%j@%#55CmU*GnN~+JO^&FLvqQXrsWN}7UN7s96
zoy6&cw7W#h4XTl}94Ki9t%PAq7$(awyu*h9&?avp*sKkMJ#j)70j<_sA^~xks&<dL
z0$rXc7n^;vq>8waFicF_jm^$+jv?4J4{^e89#9T9?CZksrhm@Eo1f;5;d3B1CnW?&
z!1`!*yTr0o&d)Z82-^^-wJ`59OInEIfb*>p*TY2|C>3xHC!UdL9ro!5UD=<Ub#D&D
zuyebaq$<*;hCwT;ieC+-Qg1!yt%iIwM9}0POc7VabBl@D8@yGa!t<xkdH31-e6o2L
zSA|?tgRb1k-W{Xjuqx2HmRZ=!O>nKxSr9;Tp=Og7kuH@k4i!f#8GK_lh0p{L4a1dE
zaLyB4Xe^W4f{bev-Xi2xnnYP0Np9>co5K*-V!IEXAr4I?*o<RonFP+}T_et*gQixi
z+65fb3Wu%N#)%2~^?4-DbhD%ud@$*ICtYr<;pztV9pz@gem+em<Ryl?)3`c3b?5r5
zPL}Qzk3oeyd2b&WCoN00$<}T2C*DyIV$Z7XbE)Y?1sraOJbdjj7u$*bocQRI9o`Gm
z;CcM$8W+A!Cg6veQ7Xd#&PCIqQeljd9O3F_&-1I9tN9t<Dc`qV46(q+`2spO+IV5N
zTyv9lI5(JVPBrSRYxiDS%z&4g@d9PO2B0kSDm)m5En(c6(B=?K#h@V#BH&|aO{7XG
zi9SQU7MnXVY$kjyWR8Jlk0Wbg)qwLM9+h^3TUmVOT!dko$O7}to=bVgZ(aW*e$D?)
z&Zd!ZFlip|EZ{rajBEzSI7UM7txMiB#fklro1%j=EI~kVWQCiQ8*Jy;FZoph($+l>
zl&Vd_TN*UchagESI=57NO-34;E*2>oZrm(G_b>|U5D94!jIcst&I!++UGa4Plzg7(
zSc~+cTI;s9?%0A*I5Y_oX>6TiCwH5_QPil#Y7dy^R`*Z`xgtT1B81)w(9%9XPD6u#
zJVT6l-)4(=kC&soi8cq<&8t6=N<p!~IM}|07+UJzniYFTdNbBNGG7O7IZMGybAsNx
z_57xsS4wXwrDZ4vX@J#<LNGO0363vrzswuMTddULTb1t9?>&k8PkhFwWgzPoPByp7
z)tuHh<m8Qawxz9@FTM#_s%f!FyWd(8xc<4%r3%}#$ivUS&g*Z!&gVY&8k<3Q_T-ww
zTv>8P;>c@nY>08h$AMuiR9P5e<l%!2uRYk1=Y@AZxMazmLg3wxu1Q!Akp%}1v@`{R
z)=^%27FnqIx@oCdUy?l=u}Rr1)Vi$Muc|pN=M^6YT#WeOiQ`0FW`d94)B;4bsYP2M
z-Z{{GK=1}#sUr>p&MEWN@!86`QTElu9!ssb7;r9{YH`zXGlrDibDJqrf+HoXM}FYG
z#k>0Z{Q2RpA<jr7D$+Et0?By8xDbsECgwQ(VB>V>S`GGUb(`31)NZ!<AghYOtM%E?
zc+uX()*XVwHDp}eF`_(OvWfw+mM1#IvbMtAxvJV8BAoE-;zNpCaMgh~il9wT>!ps`
zoQ4_jQZg)UpIUtapw5fU)adg*Qa&D72i)5+I_FH<-CK4Qhe&Y7${IqjF}STW9XK(z
zy&<q?XRU+1Nt2YKxY~;Q&KySXjj(g&S9zC}%G;ndaYFFHGAyyiNA1?M)u@(bA;f^v
zy4btH?vB1H>cD&9L3qT5bE+<Df&3ICI_c~&hy22m_mXR^eOS~9HCu1L^OlxYr}p06
zPS4~{cBs3ueyOxJX5b5Z@uOX;%DCC^<)8l<9zQy#RORMyV4t!@+0}t>{SzoRhr+bk
zve`^@2Hwr?!1GU@^3|_@Kph_O_UA9GyITj&;{`W7p(Z1)4C8<}$9$Nrkrb=>ZHB;E
zgqy?6A<afpZ@ose44*JrIOYN(rnx4;6GEh>84c0&;%v?=&Vx{K4uZE#*AD~gK}%-8
zOXMnyLnQl_tCouM&IA_ik)_lIr8?^})Y?Q$)v*+%v?-dY1yc_nZUEts3t!v+D}H+W
z=Xfw&P^+@k%03yuJcK|M$I{r6`<z#exJ8?ce65VZnIL0T2Jb0ZsbbR_UyVe!b-$hH
zSTueb?yUlW^oPZ8(-+nYPk4WCb0XT*dAk8_5Kyh~&Xaez;<-Hx4W3%%({o<+uW(1Y
zYu^fh=1j#N1TD=eLDR~@ptxSA78`%t*-zs%t&NEBgf1LIXjw(4w%We8@`_O`ovhiJ
zr)O!j@0Zp&Zu`<@OpKaYsds8Zojxmc5y*D#+N47o6U^EWmE3f(I?dI-_Bb`e6l@J{
z3^Abde7sC6)9RRctl7_<G~ow6hwWM#WO&@}={icAoML0^299*_maW#-Rd4!Ux9c(Y
zyOio8&&$B!c{YaN&4IiWq&SKj$i?&C(;Z*hB|dmIQ&ENpj!Maik5l2Wn|brChfLF!
zX>zCrQ=3+$){IDGyBTrLbA5AQKFrpa)EIpi(>Vdh<<)nLyb%VQOO8QDDjvl(YA#J!
z)UKs-o;Zzd#*Mueg2xR3suH{qV?foBmxK$Rc}Y!8SJtruCa03<VU*U2wt%q|7~2F-
z?-akzlTn-lk82uQmDqIc(8A4f!?Wcn4>yn5=f#|d^K3>_taf7Lm0Xpq)@dFPa#jM4
zIC>t8ChxPPL{V^~SYxD*e~cpCPsX}82O@xYaJexY8vJnD>tSlN`U0Qd9{dxdw89q0
z$KKdPoOGu*7oc!0m+Usr@NP6(Vl~IX+Pu`mSGC#*E%Ns^gFqT?D@q^yY^JC5Ss>uN
zWfXB7iNnY=j)WLjsgbT<)y-;m9B`C_hm!!^wVBoKHZ$$?_v_&|m`NMlmU3!CeB4@4
z?@%0iRPJp*tjDNAld1Pd=t~-=?Oc<DsgzXkZNeedI@EoK2G<OXW*k`eUe<j*xi4OV
z1@zB!#+mK=si|*zNi9?R)cnCxOFBhD#8CtgCBl($CokWRALSA(!Tso_fMf;-mB6@-
zOwqGjGP?zym4!rr?E~f+b~zEn6Z`<dF>GJwYzijXp-KR%X3m!fy#MS&f;Xd{i$(c9
zM7;AHb~~ghn_;pbt{5dd#EDumE)4i_OHKu?85hSk5X}a((dv{{WgR}|qbEiigIjV%
zM4Jv4<dj$rd*;~&rA^a-b~rRdQdtZNmSb=O?YUOPNo6xQg0o<8$(EsbCq(baS(#gg
zu`HEzn7KY&QXV+8MR%KFU^CdP5yTo{qZh^y5fPSBsVP|v)Oij~k-CcGkZl}p%F5M}
zU*zljgsyv*bN6ItAYzXkPuJS_(Tkk0KHMUDYZ%eW*z(f)F=+JIUVJu)1%>VQ9QVu<
zV{A8$VZ9xeG8m&Z2{s4R$6YG6Kb@_kU=V||hXiN*_+cE_P8;Gl5Qm}7B9~Qa<g_@x
zJ|AmJ(drD?*PkRAx17f*HY9S)-b%KiQ>8&rRr_2KWohu6@1uCzIqWmt(Kc{PjiOsp
zzNLj_^4v!Kt<dBw3aF5hO+qXw(>9NYGr~iIl~%mC{e5bvmD7W3uW2i1c!RP~)?-;&
zRw8Q@%6jIGiS6<9o@{*eKlbjd70Je;v*hIdL&~C-?bTYC+qqH^mXx_(3YowZ0xH5m
zg}pek3wS__6_B?>fZFPa08~7YZTuXi{2@7;D5FtXVRN=6<*sI6uU%eWGv|zt6NpEG
zL*i&t)14rAldqv?Z{D)*HrClY#}GZk5a~+Oada&6%(5)xWuYXSthg+hQk7|I;%SE^
zTd<@i*janv;QDp3x?Js}nw^j;>=&y?BsDm-R+|&zyeD|a><UlQGj5tJ%P<C}7+TMX
z^;MOsco7m6^?{{In|lJHmi?3pbF!Kg+0=iRTp-^2?eb%}?is@RBYuAZYtt}y^E1cJ
z=35WCO@2DrdQYTfWp?2zZw;U0arzuL?m58)yGg4arP(>wU1}~ZqiFpVDO;TYR<5`Y
zfjAB}%+PyclXGxBtbG+#tVW@T=`?l0MB$b+2XX{;UW9Xwu+<v4tFfUD%fh@YHg48g
zb5jg~Fb>9%2XZc?T-to=rOp26bKFBy>eaRX#F$MS#UGD*(FE~IzdBa?X&xZNfYW6)
z_#GO$>o8tCx<3XI&^P2TaXnTaZw9VUr4&1XFLfR#=h<X5IMG+K%=K1qa!<<~u9B-7
zc>x$gWUiU#DN}*hICky}jsh*CwOU<O*e#jrjNufVG0qNEzy-=&D7v7)FvhjEHf|<%
zSJ%A%{3+L$PjS&0RXq#0Q3$}Qy=JS{jH6}NP0nQ6oHNGOo1v9CWn!@X$;)i>J8)DS
zOWGj~T0|W4{>B_kCp^$3M${T={rG@!du~}xRWvI%#+pZ+g17skXY@<TjH4$Gk;UIw
zAcEkXWf)5-oDZhZU23JY=W%zln6h?L&IVl@c%5@R93#U7xeh$qjL7}A4gJ`zdz?s!
zrKi#vJtx09d6+pdmZub2I2!rB!73Ih)FR+o5&2~HH<&QrYiEzx`iW=DM-Yxw!%{7Z
zbIwuein+!JDLF&VXi95+0-7*f2#EKDCh<|VX+@<B4pNS*fp#Jd5^9iLgS3M08Ic`}
z9KoM{8Z1(-M~JPfl+Me{k}cZWZZ-_w9LOm&Xm;0_W+eu4u~EIXb-DXsX?iKij=vVe
zhFkMspNhEl-KgIBcW-JG3=(td+K?~n!neY|x$~E08ZR8^54?5SF*PS*7glh3TgIlt
zT&qM)hr3RGq8SA(mFgXy&q&T)gjz9ty)|}%HN94?jN;Z==2<JH7IHQR;kDwhjJ{U;
zcMzj$J3JfjIg4BC)?Q|I*O$0qvhQpR!(kkm_Y3<ugP(9()+S%*ep%OgU>|cFy<-@R
zedTcs(NWTZ8x0aQwijGpnXE}Z?9l3Q;~C5D3N*7lzW{M8OC}vmP_bfJRq$>l(v(se
zwj<MKvv!srQf0qyS!xg*=0t}mOq+qan*ql(MbsVGMDSvP-PIwpUlKxS<C_J%uq@X5
zBx?Fs8}AssQmWX9A+0trOO&}}c4>J9VYPM78Au0YZa=K2b3d$0{iw9sp{_3bu@<(G
zR>Za3e}$L?T%+BTssx0G4<2x~IVUd}O)dI#CSy`+#YsJJVqF%!uA?yf-Xq|G!*}VA
zCbd9Sh7cLYkztI)XgISwIqr%!_0w{k=+_-;g8>>(vO#ct5}d9~s$5K|)ce3@7%Urd
z_P$(-MRu<9<Ykpk_Xyf0PmI=rtEM`fQfirk5eY&J)+^8)Y{#J6m1+lML;;63mDQtK
zurb*T8tP85!)^A_I}p|%cdd0Q`}Zn-xhnA5T8Yo?9CS48_V=}-Z>bcK)}Qp*Kv*eP
zYhZJZ*CvP4!t=R+G~?X*Gg3D7=D|4vf>fIFGAyciEvAZ4D>&sWyw(H?>v|u|`@ID&
z(?lvD-ct`dXs}T&*_?h0iZq$d)vy^57pOTCoag-FocXZFJJ0#~8PA{JaC38Dvz^%8
z?3fRGD}*n5s0A6fs0uaD#5gc*HpF3ISrTcUo0F0?=rC1Yp=z7AE_S`<rE-0!<kEVR
zFqqQ`ED5I35n{xI5*<4(xm;#E1K`MFQu5h3z_Cx6{mo3CXI{UUkdmoOW(Ukz@RqWT
zkd8xOjE;S>K<worZa;bJ-X{{SMyA6DcOqNhs9bT&c*&{#uqLt=wMqLKEQ+a4P(9AB
zn1+!r^NV~v|A9$%H0RN_c2tE8rUO;0H65)*TzZ2etr4)%X`52+FhsVSEt_dH=g}tk
z^@c-pW)+c2Rd>uTO0oKcm|=HFGmb~tq6PJ2EsZ7B=7m?AQI1xV$!<oqE6v6jtN{y?
zv#)Cwb!4&lHiTgv+NHYgcgbc{?KIWf*4|m&$w$W!BEdPRX4LvMK3$&{-5={aM|f%d
zUA=Aadv<fm@Y0qs(u0h?rIQ3i?qn$)7VP&?4?I3i8Td!K1j`<Y)*n@^q46Sm!WbEb
zKxyZ5GX#QoiYD5OMp7vZKHx<dyft(#xsq!oS1U62py6<Ji62JGJm!U31>dA-N-j+}
z(HN3l0I`;Y_l{|r5FbeMOg*$wvNe(MEGkrS%!dQHWJ*e;n`^Wjh{558hq&zpX`T@+
zjGJu}3svUBo@L$vZ59u@!qur1O0}`SB^Po~4k@$Dpm>ux;TTt_3Z+^wmgbB{xXe4g
z^4=#*dd8!R4HxGdvQ`e)2d3ajOJdpWnA3q92jdtBBIIIqna7)nL#pf&98zZVfd|9H
z3(c=T!Rwy;KwdoYcby5)W?9g%*#6FwI*5WwgIGAKwTeJ3f~&H^QAHv8$WOiT1^o9c
zYQX}K_8o$@35nvQ6}z>`FP!2lVv%?2TCQq3SS~uoVIW4o%2k(aj%_vgrHL#y=*)Rf
zR}r>uaSLaH_h#@~#??)qwluxD%ht&|Q;W%~r?ik$TG>>VNkp9U41+-c;$>|#>_*!d
zGZt+kw!fuoceV2`benx*U`wTnAg$ZoHzT(Wo>~i_-PoU8?>lz7^Cpgjg7tDA&-_X9
zVNIP!C#5GJlP<BpW*jYJJxzLFT9eUMT)n>~z&>S`B{2mUW8ir<!#@RKQ!07M_)wTj
z;oYkp+qh-)j_Z_(J{qoHmBCv_`XN{Lb7Hs89CnqQ!N-U=N1kUC;pXy^O2CCkUJeGw
zI%7iBqWB@&b61R>J&uua8i_Hmr?Nk&>2#?vtgf$T_BS^yyGwQ=)<js%@whnQ#<Qjn
zJQHJNGexGsb4W_biJH1xz4eWtCXJJF1+lETl){oKInm&1ga1-iiu0C*tKut^tjvdk
z&j34i^*Mu|C`0C1zF~iuc{F+sxj?OWojH%50OX}GP600-5y$n-j+^~~>qADpH-XZ@
z&arOvW#;TBZr!5;xgGtkuikMUx1frX)g#2ClVKJ3>p;3Rub_^msw*tH;zHmHpZ@~q
zA3x+-ergj0TZYjK$K6$Q!N6`jSGHF>gq694BF>trdKYb}g)&>gIu~Or^}`kC2*Kn=
z>av&|NboGR;H%MUtGIQvXJ@2oYo-F$0{1e{Eh(^v`R2w7k=|PY@cjIoQVNH|fy3cI
zsfF#?cFm}qKrW`U<$YK)7_F70we`t3YHpfbtA##Z%i88n1aGj3psbU${)Fpvmdo8t
z<aBkH`l9rDgBM-b2<vhxCvcf`f<-$VX;-mlns+B&-oq+Yxmhwdb7hQ53|3>y;&^&U
z9F~QPZ6cM*v;D&5-tl-FnM;K<vmHlnmJGGfDi8!M1cfu%5+pF3jVuY=ZcjSwn5*K(
zGh{hX@`4W=tCht_accLoKYHchu_02?G~23X_Pc|PyUohg?h}^%HBvJwLhzBWeatux
z)~Do0prADoh6x{?u_KpEPK!OSYV|9>4wdT9b30E3D9W-}UBV&crM5a!Aq;`RJDjgJ
z1TJ{cOz;`pdz=rCsfPnYu3Wec=i7<26oX)=5s3p{9CJ$eYK+EbmkU?>12;>;g@M-|
zZdnp+Ct(|3;a7d#^GI8_V)T8T8Vl{2$ICy>k*6dbx8ejCr)q8?sYBPMBa4jRym*UW
zee2)lAAb0ETIaT|5j&OEsN^iVuGQ3jy|?Ljv}n53D-@1(3p&kII4p@}K9H8hvM8}Z
zJEasth>X+3G)=@d?<1wuhD|l~UU92Ju--Xxt!k-!pV(Fk(>O2;ky0yHDRF*&&Te<Z
zX4-J^-~mscK4Z7rk&DfG55cbvyUSC=5RAra9u-j&-gCkC17DqJ?=+@XQ(qNnQW4%J
z%QeWT&tLqL1ZXcS*5A7Nax!7?&dE82q<R3d)?AvivCPi$zDDjl1|y$Q4J3H$_CFg(
zia4erGKj%D<Kz)3sNS%RkvA{4<T5Y}jwl0HhlQJ3sLt{Hka#pkN;adrjRDERFhpYX
zgki&Q_8RH(6RIxMdA9i;VOXJ~G|z-#Bn(zXO0&)D$aAI^laIHC+Q9Ym=Kz$PC~1!%
z#PJ;b$S_R|)5P}71QzExGw*kZz_i_v^MNH5O39RYw=(sFuBa{K+8`|#a4xiry-iOn
z1~<A8(5j?4S*G8Ht%bpQn5NChc7%ieHXrz};j(9l8;U;QS$f3e2hPI8COlv_do+8N
z{mjRQ16QAX%u+V^^K)K%I5D1Y*w2;CCUU*YE86jDeFjn{ecd$@r}d8(9Nzr|M=U~5
zUc1vDg0^V0Ige&EhA{BvqqpG0<NJGKTtx|Tlq<<8t%kE3*7kv;0dBQ|&N@1znZtfh
zEtzo~d9XdROeAG$DGZwp)XES;n+0MFn<Xu%dP=eHchylvSI5<kXmv=w(~Bz%!$63E
zo81oQ0yj4|CVApLT9p_hXXoclw&lo9Gn7eC*L1X&`31*aQZmcDHl*|sz?KJuFtXil
zdFUS>p{ialS_|fPulWzYAO5IjzD}e#g`{p}8S9I(5((DVEuCE7N&it_ych3dDas~#
zK7TRcS`WhcV0un-u84<6+X?Xl!3mGIBe{-DgUzE)xv<ZfO&lqsvFaW~VXPK~4>A$l
z2$C44f#K|&Iak~=lk)-9f|f$ci5N$04o5~9$SJcN_SWx%l9s)(_fX=nLA4nC<pMsQ
z<A;ejS%y1oHw>foVgv^z3x{+dcu$C)!%n%r*&8jqw%MI>R5?h+8V!AnXrs)p(>Y6_
zEQv6VR;Mf0)Tn9-3}s2kK^Vu#q22F(%v`&Vk*MpH|D+Qyio?}F&H>H7O=XOfSb5{&
zmw7a8cx}7ka;Ip9{jwl2bA5S*KK=}9W<Q4OGjSkyi{__mEtWRAuWIQ(HUqRi^|!a#
z3#q3@NEFnqnE81;BM_OFWvyp)9ViLbj8)bs_V#xut+}ve2VR^Fb7}?HWihGjAr5Re
zTc&BEl)`4SF{uTN&bygLd>FXd9nex3$I+VcO2xU-W_h^PadimEi7EP)VhkzgtcRnN
zN(_<niwk@!q?(j`CdR-xjcf16k-*UE9$H(opswemq;x#sCUJr@s;+NzUutE$-SC%Q
z`z!p~`Cmq*wr=E3+3ifQdQ=d2iEpSaC|H?hdMCqp$!G4e@m~17I}uQ6m+<H|P6SVs
z#yry#vcUZ<p$l{dqzZ$87a<FT!BGWLg>5vIxH(mFE!H?#G9I`Ho~2eke>M_<5hW5S
zcuKusxSq%?rk3o5VKY%eKxRj25Ejzz3oU4Y<*;zr?@9X|XyG`$1secW3t`;Y(5WyX
z!Svk5&4l$AT2rBjHR08gKojF;LkI!y0yme>&~gARpiQCMnV9Gy8x-bzurUFOF#=1W
z<kX(aLSAO78sUX%gceb4^Vu!`b-`wyY`jC9HN$Exrs7>QwPezXhFl#Yvi0~kBhTfM
zJ>TGi`Uqbh@o4xQvOhzX4P!3&S1Aba899(Ol7w5$e9#?8?+`xCI{MPJP-~<j2hM1K
z+B(?R>_CimTNJ7me*XNIcs;((_x9h#JDY~qMv=9i4QHcX>+OhjtsR752DM?v!F!WL
zQElRcLi8<SDwW^^Nzy80H%`{+TUCjJhh!4~dvjMGxTQzI)MFi{8UN59kZRrIC1*C9
z4cpD;sBxvV-j)-?c;|*+e{9<>B{$7{x!r$LtJPaX)=6?h4E)76|1$s4Yrnx+ob37(
zS-00!rj+i`^@Z<U;igj(#iz-VXfyunhhKVIo%DaKcgpcPermAN_v?NQf~qol58jij
zP4dgy$O+CU=*2lxRsd4Yr0jSC1P2et!Psx&4F9L{@vgA(0|C#W?$+<>&7#9Nl0+I^
z*OO+bd7(?zI8oBPBP~5c$vE!_!-Q&K7`LcaBn*UM3oRq=&svPZvQO0(q*SydLX%*x
zIX5IaQI|cPBqPqP_kSrB7vf5@Pbneufzs}(VKSO-I$Tk!u`NyHQ5wJ*s6a}Ed3KD!
z6CIEbr2VrsAgei<EPYcf7Uyw(VA`G&x54CB>J@u^2Aj&0DhzQXmIr*58@{+Juh>}J
zw*zUB<K5GKi)mHu!LCw~yS{^y2fbHB0m0MdQB-;{y+Igq6x2Hl8_uRJJS}GMx&sPL
z*Qwpa-Kg5WwVxdp-L!4>&R5?I!0|z9U9P1fJ}_=JY_?~F7>G@CM^*7I;6}gdOx4Es
zx1zbyh+%6YbZXz6v%Zc#Is@kj4Z1QE&K{-QXvT1)K-ZH<zaNKl4l?U|SA%QC7lZOj
zY0ZX?&pr4Yf9;K5=ge=~z^IBls!1B9moA6xj#4DOhVjR~Uygr?9wDlKon%3KT+(*+
z-cygh1P-ZCQ{mxuBzn)~A>l-M6hrH8u)2sB;lUUQn~|P`pAC`djj^<w6MhU#gNgjj
z3L+B^#&e$J=QzY9%@w1nZzf|>IpuJ@Q1bzG0ih`{rv<R7eLhYM!vqwY_tEB%Xq)X3
zT0z+~+f2u9St`X<oHQzM2>39N(n8JIUU!Y8fgH8q999H(0bIbN<UGT?2On_~tlqX}
zsudBdkvMNk!OnYzX+V5qW!5Hha&yIUbJ?CvMQd5*-A!XPAg<y=AdH*VP%9?CRPA{v
zKC#Fr{K3sre%U?d1un-=;JVkD=p8yRu6KhrI57x%@_;vE_Oifml{`Pexe@5Yk?dY<
z{R~<gq~eGY_)DMrYkcREZ!*s_(gY4&Gjts?5;@jlP9u0ZaT>iJ#5X11QnI~{-Hf4)
z-P9V_+Ji}q2%YhcHiy;DfcFNG(Hx0871TY!{JZqXaUheyA=F|KnSe->gmCtqrALOh
zE=ilb*7mRVETct~bdmukn;m213y(j~fAUlRL%y{6BGQ#n*Nw`N+t9A;A5~wt_4_*k
z!|E_@mn>dzsDG06|6}jZpDo$0`%dh$_THI!hI?-fZva6M96(ScErm!~vuum4p>4^M
zB3f;SBkYdwFaDDp;jkmzwiRu;-Ql)mTN9}!yG2PPMUf&2&H#w<0eE;*)vY_6b24-9
zz5K)4J2TI%dP70P1zR`^s8@B*$&)9iyVhRc?^@qAeO-|~uJF}&!WRT7&<$2CxbGup
zs|H_I25oSx8&4ZN>&_!;!;JcvY;J~k+#C{5Z+GV0y<-&`x)8W%*Q^dJ>xRP>8Gk{u
zBONUkGI1R7zMV?KB`5HWm9=|sO`~Fi$Xsw_^~f|;vkdBz(mz#@fiWtLvkXLYMw>*A
z1H|5@SYlsapFLL;f_?CW^%ilSIP}C}4+!EraDh=3=NoGn)N%293EmlmshV*Fnl8|E
zf=I^s2F;e)ITuPslTh)v5K2ExWatNcH_)zZL6{L-6KD!YGwdRTx;Yv5UB<g`&@+rx
zB+LuxEi8VUg|gyj>Qyk)l3~<s%q`bu!U@8vI}vM03O?}XAN?i1e*Omk^6;}oP1j_p
zfoTZT@^cg@O-c{~rap;f)=|gAdgrH1!+BT6?#}RwOyh9vbJM&IVRGh(@dK`g2eosm
z3cEgpc`iBgF=~y_(<zjGn8E?sYOglV)vj@wxE*Tsi)p=BHdlVh+a|EtuJ{Wd|BL*o
zhkwc@7m7)OXO2B^4Qi?dqO1=8L-ZexnoF1Pn`#+C#s@}_7r|e5U#ShDwaL>aD83-o
z{dSw$tc(zZejHghfk*4WkTbWv5}akS&W)&ocMU0;daXL4^PcNIA<pr5yD8Z>I2`9~
z%USDj*LIC^IbVCo$^=hIgtnvIJRlx+jE5Tt=EOxBTxe*!6~Lr%Wl)m`@uZX)hm7+|
z+u1;~VW?TFNx7Q@h(W1CzenQ$c?2@G>JUIxNJ-7Phn8-024GNHillyzxDoM=Bs)TP
zj%!vlT`1^7c^ssy3sNTL$PgT>Rcky==x8@v{5Virvd~%;8d`(nf{iDL{eh<Kz<cY>
zn%1a-c8>Kbe9H@%|IqASGZ2Ky(LN__p_p)$&RK^&_~n_DWAH=n{0xh1WZuX$lN?AE
z8iWVyEq~(&{}x|=@&=#Hze8}Q3YxW+auzJLRAkj4V`aO(XnARAdNP;gofsJ^)_-DY
z_?#I{`<h-b$E_((g#k67W@ZWXq;7%h>lAQoY^}qWfw9#u*BHC>)mkotFPT9u#H7^o
zv;#E9;&gWS-kW%#2we!ApKti_$3MoOefXyh)^jD7QpJ@}XTDuAsqY!C5~(+GCb6Xi
z$h<heWz*j*KT<Hf1z+)gcDIw`C@xe^jCtQSg4co*NB8J{EClO`7{|zsRfn=46FCdG
zM86yO_=8ANWk?B;hHcx?$HbU3o6cm~y#%5NDJ90Jj42UQU>hDYM5P~1MrVwfIOgd=
z(Ngk`SYx5;SRMs$tMn92S-NQ^8TXupF(r1-ZcyjxRx4E9q@!gE9Pp(V!<VUqecwY)
zG^;gfXY)YnkT?<RiAh;$yDe?ECYDCo+7VA4A(h2NJmA|k&NWE0#)VDMr>hG`hcwn<
z|K!Pmc+t|xnwYL^f#(F59Vw2e<|4&YATJdg6`eDoZJDMcrZhcYch_m(&35lnChrFp
zOv<~WmQ5@@M!AVE6ZV%}iHKFWR1uNy`Pj$!YoGp`eB<}u;5O{gB2rlS0aVSQq}rV<
z(|-jD49B$XR3ht4n@Z~UmsQsEtW~F8AgZMX*vu|h^zNAasDOj27Ewhh<5`)wg~}P#
zpo?A7xqBwh$6ig<s%wIa3@IbxP%N_uA<zWJX4Ua0UjGyP&Da0uJn|1o%dTya7xeDT
z_ijl}&jEh%-tkNGK`lP>vcIcZ_E9s9#lg<mU=2{qZ1*0mUZLy+^^SLk$XW0_IBU7>
zGchX<gX4BM(8tJrjGzN!Ho1}^XU;oM6NJtS`(*01>(+DD7I}0vPGpF%-z#YtK{8$A
z5Dmm}#EHi@Ejjk2;ntqVAm9RRv&Q+>I@&`+79qw&zaLTY_|SoGV1iYPR)Un_WExFH
zH<eMcV{Z$trh&uWxH50qe2gO!T6|~$(~Qf}y3{j8S}0rduJ{%g*7$a1%Me$ZV`Czw
zj0>#|mr@uGBhN&5@L)^(;A4!pmvpUX9ULhRq;V|puKjLFDNGns#ud@Z8Gg2N^Un<|
z`tG%Rg|KSM`Xc!f)mc{Dww9xeV%Y$<{CjD4vl@Z9kxXoVS&bb0Q$O^ldGpDC!cV{b
z4;h<?t9dOI*OgfxNtRBm1iDh<YOq%Au_(P9A_WSnXHtSji2~I>8Z76GRLnXl)2mh}
zNsb-<;+o|Qx%B$wTm>X+fRklB0)?D(@$>9lMNHBpjs?~TzF^3(>RN&q9&I1;Qy=^5
zyxx6`G4@mTH)kAwc(yDWn9kt^2-6F{R*nO^z4rWbQkxR9<^tPIOM`ne0mf{y8!HD}
zNi<m*_ZcUaTW&k+<o3?73W5EQIgCZVU#w(Xa;6_Mw?j7FC^9K}R>8_(GtdOb*a~A0
z<L-dA2K%HKX}pJ=K!U+jIzo;Xw$)(ejA8#FctYFKZads)iw*BB<Iq~Imo0-04z%x)
z*4NP?_;y8%1LH81j3b*MYPM;C?P?27IP7nVz@<5CQPXW(Kl(V`gV$)<E#hoWN=l|M
z?HmNP86HTTsv~qhFb;{!%bwRBZ#a7xXc}0R!B=q{>GwAb{ZMAZD`QTKw*%+R_Y%U2
zn|w*sJ<cVZD9(Q<8v9*i_e-L`KrZ5LK$zqx7OV3L?|fLPrCY!d6Kn!Ft@7?-4~-^P
z+m`>)kN;2j=96#o%kjS>`!uB&w$h&~m;@j{FSw@#kGIs>&7*M_pzPhw>Waox0(S~>
zNv92J0RTWM-4cpHVWOp5Ukco%i$Gj`rdzD`6_#@4llV(<vQ=(g1RC%0jm<XEII>=K
zoUgb1rC0wlAMZX+(pY+5N>)?$@rM~gmPzW&fxKYDI}t=NbzA$q+<^^W0PjE$zw~kA
zKP-?gp~Cwy1<sqmYmc_vK6{GquQ-4B8pEJm?)PlBD?(^EUvKG}z@}|D95P-MaYe}>
z36dO#XkFvm#+Q<HVa$og?W;6CFnCAXD7tRY!87dlWL4U(DbK0H`4$Z;(pV~#;(_Rd
z&~0$ebM~O6TXi6g{g4SU6M~f6kjcs6R7hhkP+<hgj4|U}Q@+DQ7soM9#S1ojU1-;5
zwB6Ra*Y^j;aReOV1KxQafA0^W?L(SX%i2YD`yFAm!jst_1|oulU;)QL8ckX#r_8wD
zaflHw4|x2jqtTHV2Zliz#z^dw1tJumJ&)pRe7yTfet7%mI6uGOox^*)as4KHp0a8#
zasG1z>F@P<`J-z03SoWNVs#D>`dnbS3q`!>_g$*@k>&RGK$)YJ;;aG9yz<IJ{@PFe
zb-wYtZ}7GFMI+=2c7HV8uK4CUPrJm8IWQ=*<C=&;Cj4IOdy9)uXKkb^qakyUE1dY0
zt?0DF`MPK!si)5U+PpUfrE>ZuW;<^JbX2EVuBj6t6$qkTx7M|;3Qb_uwfwo){vtp1
z+F!xh)Fm`&$q1gmne`gfVE^G2SoK3MxRHy1_u|hMj%AT)X(8L)^Se3v&YIm!5aj(Q
zAoU12>jDoy_A1xAEs20@;jC#mUpJ&=O|Ut`VKn93J|$eTpSM9+1xp~DINIR29U|t#
z2cq^2c^+r0In%8hyerRL8X4`OuxHLUNE*;0>%KaFM0fU}l(^bh*E(!$)p^s&Qcx8%
zfMyR8P32ccaMrLIiv+q9ZqeF|9O8`!^zDk(*#n#iaqJn#kvI<2lK->EA7lIQ6>e__
z8c{BvJRv1`uwHR{bH(BM8t+$(*`X2$Z3jRc55)Z~nw1b%9Bv1m_y+f&rIQ`wIG9jl
zgp2ShuZQpFQU3`p+K;h1e}y5g(G>W^`r~ZnRjzkE+tBf;@A(X$h(E*Is(iPDS@*1k
z^*k0a2ZG`_OK_MG2)VP}?0D{GBol$QfY!x1+~hRu)!;ozGmUrr(D!~H|7rYR^Z)+*
zf5|uUm!?dhP7SO(vNFYaGMi%KWWB3Wz6DwmJsRV>JcqhNgP^hvc`8JqbIx>soO=%H
zs@a9s?n3NXD3w#bsL(oWjIWc~c~;-3M^S0keiG+!In%W*ZQI!ZsyN_8`NY{L_>&L*
zBu$+UQTCy^INoxdmNbSegNMAhYutE^3O6p#ljlsziPdaxdIpX_MLPy3ViV>Tf%a}(
zs6XeUjBT#4yDxJhynTHjJ=<~7Ij%?8$B~tH3<$e1GbFfCBhi}xXRAQgz~w&in)jfI
z>ps&bW!nT&5=I^P%IzDxark}S?%!b7XNEqKQo@l5!Qp7=hlpw-$I<?%DeQ*s9N&E}
zVf_f#w)8_rH5&f_h!<8}AUMiI%UrI$_5FlU7Esw{<tVNcZYVKEG()%EQZxPieotst
zIPZ~MghH{#->g<S*=#O|u>`e7*?CB5M29O}yrgjl4&#Ag0DMb|D->va#5bi>JwVct
zH1_O=mK-u9;c@r<yt?{PzOVf>Xa0iy=(yS+c>R?#lG^u-*`wpY%E8qxL4wY|_<8+q
zvwLnNgzC<!S;{#t0nJgTLY>M4MIDo|l=togkm(;Y*Rqr)oEy=KdTh>gt>;gC`bTNP
z|Caya^Zy6l-hTtK5ot~;52Dlf*OW6(t$RE>)7f%`6hBH^7L4}tICHWQp(#e2SDi9h
z^E6_0%2F1Ncd8mvon~%iRdS5h@g7Xt8*juqWzbO4S{cRU$d|KL>Y>_n#BCYq;|LO1
z`If)(+F#?7?f0T-Eakca;&Cp`Tt(g0NBC{39ryLV*XGY<jl{GYnd$TCJ<6AfO-{Uf
zUxBh%SfnL)O)7j<`iR^FI^L_PDsAUU0$2Ni>rsu5kWA9V<sQ0r#cQj8p!9j*_K<n7
zYT0&<oUHmUjLNw{ayNY8*&pzk_kNSN`?ndRjT~;A5=0TP(ppg^H$}8l0!cG5jcCfO
zFCNogyh=<KvC|kChtV9cx9q)XJWV50!)jU208+*nD8WM-`cf5^5MSn|w=K|4!lll^
zdb6e<jB`^}Q|P5CE(=*Q`<wU3@c`-=!#OMUJU%<8$!lIY`vhlauh6=NXP3{|Y&Pt}
zH8=gpg9qJoY%wcWhi5dMBRkJy_e1>PgHO}O*J<3=7FBs<-88IPt2!IUksil`&=AyQ
zhHl8*?0WJ>@Bg9sBWU-WfpA2i+l$p{)RpV(T$XFdN`kXoS>CyU3jl24)Q$sIF4{NF
zOL54VmG}Jd@BcJ^TmKXO!598M&#s>VQf8pr0h&^<78pyX*&3F0*wnuo)T|<_47I9`
zyVNA-oEc+e7_GiPgkW+g%#fJ%oXCWgO%T;|K6Xw@k&quO^H8<fBHq<i{sL~3JN{HC
zHUDCr!p>8t*QtXaKL0eII{P7nh${ih`UkK^NAkpj{QE4#rfj5K=gIe%Kv7?tex^Q0
zR;i~YZr8QR-C<nMWgsGj->Us7r3dCa;0?Oq2%5P)jEriHdO?J<5O}+fFpL}q(|^*K
zxgH|dHv{XYC1%HimE+m)7GFDjmd`%>eZG17gfXcR*1$MM#M_lvb&fXZ)S10%t*_+v
zW<c;feD!0jE?yyKg{<uRNE*S1P!>I<+OPzO#>-T_WJ@x8CRDA5<uLXPhrRV&h^Os3
zd}j@|I4hM!9L;8fHj$JKW?ikfcxh<bH5-3M7#if@d2RDaT3ylHKH#;7uko?hKEd_g
zkz>ol&58%>mSG%eJBJtHX0YE7ahaGRF)Bm4VbwN7b@<#N2q_qhmDSO;4L$^hVKAsJ
zD>t_Xq>VIQ>C;H-&M?8>Q~Ck-T!kfGka!G7P0QHjb_mO!>FO{<igeZc{IaVq2;%zp
zolTdwJ&1K9<e2Ha<FEeEUuC^&`TM`~f8*KBmBCql7Er8eyhn<#cfMjuKBz8!{^p!)
zwKt3dvFH)aCA@|YJMX7nl3JaWQWdl-L6*7@cQ_yD`|=A|zBt$BKmpCR|Jwbpg}kr9
zHTYoqS^zjSC)R$=AHVpMY+MI<DD%&m16j=+Uh>yy=BVc5L1YRd6%1QuB{lV^?@XLW
zIiRUnqAo&?ZnoZe*y^2Ng7DlT`@L_IoY>voFzgd<vt}F<mxJj+H9oL*ASt8Sc#aNX
z7&Ak^=Z)PP?ABl9i@VQrHSEY)tO!DE&{Gg)cQ`PrvRZj?-bN~yf{G>xiNt!n;^Awr
z(X7@)g;tapm6#kU*bpZL$jUgRqk<M!5m6LPx%_@whxc&k8Anrrz-1DNm?PtmXqt{3
z1Do~%TiKFj&H3gr7wwky!!2QK5jp~E*6o^$bxT%{(r|N#xJJ3XNwz3+i8tPRhp!j-
z>}=K1dSMrZAz2VTTQyv48+2Sl@_5&9JB(~v(;7S9tT;qvAC+xWVN*d|q7BJrrQ}SX
z6;I^RHX!NlitM}7?tLBz&hdP}U~-~nwbE#6MJrl6GCLvWAaSJO1lW@`VIfEVb`?6w
zmN}pX`)JM}BY*jme}%UXZ}b2AgMUEot+w7Z4ZdlMgo&GudhW?Ej^JpPyUq@~aslHw
zG7O{599IgbQedaDnsXBtZ?KtV4&}U@X>nCKcjrsri5rPAF**S}S^yq9Mk}|><ViLq
z=WJcSBD8Ibca|CDG9vd=>mTHYw?Ay$%zRc;-Xry{6|zUPmIW`eKf5_cS>$ry)I9q|
z7XPlFTi)OIdzk7oCDYGNsVj_<B`tz^RRwkqQmw#J=9<j=^#%%b&f}7!?HX3=j($k2
z8qa&rdb}pOux91NhD9YI4g<gc&abmuzrow-EpDC-<m}LB&8_=BmTVc*ZgVzMylKSL
zXiLR~ro>?v387*A@Bv8z`)Hr@<x)8B2#o{D=warSoVQW9$BNokxur=NQ)V0@Q5syU
ztlBeH{(>g9Y}0Ez>|W>GKjPi@u6X^GPx7#R#I3?wT7r6Fjvzga4~)RGKC?ON2u<MX
zp!7N6M%b<!y2kXX+N}voO4IVn#g;=qB12>aNNREFXvN_>laOyi!{%WNNw|$RG8ctm
z9Pqv&jUy|+qHS%e=cWmSwqd;4b8~gzc8I+0IYC(5TL3AmJ+Tf~1M)v;^EptssWI<G
zB<D=nS_rAYesNp2bebLBWHS^t<M2|5;*r1n{XfMQZ~uTVU44n+kl6P<p=oGBnD$}>
zx_~ZM_g0tg7EY}OikNJCX`NF|>HDZdtx`(#heOfQ@~Cq}F&HdswxC0S@|>Ib1QAwU
zOHa?C?=ewtDP(qmgfvYs+?*Ae!^kNj8B$E-Jn#dXPw=;1{~MfZQ=F{KMGMnx^o-LY
zz*~bz33k(UFOc6{O~vAxShFZ@uGFf5W)4;bqN=YfryVMDr~mS<F4I3b-se;@Ej8$u
zu4!>(b#Tx1Gy=q&37a*j68i`_(`W0U*c~D_`#rnEEw{tSXP^BBUmJg)r|p|0bwu@e
z0!}hn3ZST&aY$^|p8Y;D^ds%6Ax7gAYT?JQ@99=8+eZ&bc;b*~g3u32Lg*S$52JH5
zjg`cb?b>;#s3s1FR3KRs*~}^7lSf9+DnH_tvrlv39^v#M7w#e5=xEzBBm^$bH@vpp
zvD^1J37oAwBY_)nyz=;rvrZTf10s%R`<{mvXCxdx3_N&sL!Yd`U`UzsRbUkyheISw
zOG=5&s%4k#d3~_<><)?R-N=5}v1$WNqqGjLZg(7pj5M7+6KNnNL4sx3Nj$#wgeEW)
zu2J2<_3a)m?)7o`BWw4Tfe5D4W1XIh)!qvhes0P%o|ou8w&M>MZ8!U6$sd4Iz5v#$
zk_Cyx$F|?Y|K>;jcl=*|_y5Sd&AYTKtLmtOnsPR&gDh3b1LnpwshFGXyYMB}q?G3B
zprVNtbDX9lrc#mBHqA5yimZ$!_~c~i^m?<whte!1)+M`gD^l9AKOAV9hPG{y+FUr5
zgc$}t^|<7W4hjxh39Py`Kl$({dFU^SLt?5BgW1t1Kzyw>lcF9V6CC-Rgh;K5s_BH(
zxgj+}vhwGg>nPgU`A_$1@f_5EfA=mfXm?6-X&=@3aB|XV)%@I|;+}07GxUku-2q2t
zvudp@Ry^Ix6G!3Qs{;eKd^7wzuJaAvwG0PXHQ31FQD~b6qS9!OrW;o7oa~jGaiD2D
z*_+0d^8z{Z;L$nj?UtxQ>=Ry0*inh}Q3zeYH_#7i!<{5C#vUB(Z+8rPr462RP#&*d
z<F&I-;M0arKK=ykwn$uaezs+`X&K{4pM)luDCEs<Pj;}~u7S)ej~>tunP>ZvCInvJ
zcC5RGo%3XkJlvkq1i0EC*n~vawcK1M&ekh#`^bJjlD%-~2f9teA!ee&^{6P0jW_k*
z&^jKjA9B8FxZaNphfLcxj7f>vv)QcZHPerYRpaPmWE`PuJWqO?9QyRxfD=#G_!D|n
z>|P-(=LDbUUI}=ACvU68MaTMcTj(8lW5)HX7kWRJAa{Psg*f(Q)<%j0pS<`4fA{<U
zSN!Ln|KGD}0+bbTs*bgM+J-u%DQJlIQwC7aFP4g!5E?6|bjCI0X=Pue;;|K7KV)1Q
zr_S~e0`0b$4!*7?Z9J<Tvv-cOvkf;lJNm=Hw0xQ{>2fW(p$cuCH|P8>e&9dj)%F3a
z;PE-LmDMbleFtis?jr_?O@cFvfHCnIOE$dT!znYIUz-Xebnz97+jw*YAF!C1j&}-D
z<v;Zrm;G7xQ6x7^M)o`63>0XCXP-v;oamZB&LcPbfrl3lQBgXlY&Q>hI^J-|4ddYG
z2kQtQ#>~yt4Wa1>t#H^qBkr#l_B)>JpV4eCXhH^h#CJ4}FpfPT1h(6isU8d?#F5oz
zWilXfqzwTlN<xivA4Z5f+7>RF*J+0btojwN^9nk&eBY}-#CH3L_Cd$`d`%ul4*QYK
zx`W_pQ%{IW*R(iM9&B4ytH9;Wf#5u=rs35GEk9)AaX9peF(p>3hB0~?r=&R0dC$|U
z15p#(^#%v@d+Q9pO^J2e5_9Bcci`dqidIJy$9{yNKXAT1V@#R-VPrQ%woPO6QDSe6
zhk4}DX2o024s>nc<`7vmo-PEkORU>K>l=2X<JHHHki&@sf&1n_oHLSb;Rjw6K(gi|
zAMzr%QA<-7d!OKuDvWlt>OI=>c#p0OuW?eE&=hn$@Fze1$N3xaZ}T%>{wG$o<BmM#
zw8FKH@JuS>L~(XjMt~dN;k~b4Rz8y_e`kFg?m4IHy(a`u(=-LP%CmoT3-Y;0z<B4l
zI6q^z+tUv{BH-J00R+|S0;)pe93NXh;`QzU5=|wQs@<jYxjlFNM2i}_l{tH)j8~Du
zn@k%{{15d$svh-@NRiR6{R9?E3M_`&MP9&F@2f~KofkEwv_!YmnO4F#rqQ&#mT=b#
zGn#L&2?oJrwJb9xVNA-V3-|zIGD-B{;CJ553@Ov^tpa5o99P!|hW(zv2r-f4EfN&p
zHniOtja#!lYY4t&d-f`;#&OZS!tU^dZ#?-N;ywMLCnrVn8N(RaU0<@@p0Q6G6o;ma
zlb*+C7yQ`6zrbt$PtwU5E8y{iHOax%ene3|_ISho(6fue_N-&OX}Rfv+_3G0oRsyd
z;bPNq=tq(S&dyhyuLJ8ga7fU{#A?%Uz6x9)B8Qw<wVv&ELmw5_ZmgSKA@zw>*Wt*F
zIdQcg*o`~9cyh`-UN>C#iD8J49j*5$nTySeb!X7*5Hk<XS2SMO4T*JQGWA}RO}het
zXZwiaY1Ol8g(#leL+0`Ol5^i(c7Fk+I&|ihsM(7gRK*_V?8BUG#qpp2+$;lcPF|#R
z3h0uguQ>Vh-?wmSxZ?1$X8!IE{-?A)@Xx;TkJ$}}dBu$l>kwD_GZZHlM5<-Asuvqm
zYt=QCShjg0^%5w~`$?qDi4=i2nMTPv=g=~E>PY6L@(4PMK^pmI#CVmQpQ@g)(q+lk
znJhZuh&X3umAdGpuf%@6)DtR&P{;0iC#0N^%oMn%3;aPPz7uZ+wGwN}>`Ek<D35xs
z3s|t;x8={3JJWIq^E1QDl_`_VbIn@n^|?<|tn~Yp^DVEQuL;ia?DjBKhk0={o0dKu
zc=v{)jGP%`;>qQK6i3d^S2!n((ebedpW=JYzn_cddw6C2n8)2ioVH}~+}sH3;AvdL
zSKs_BU%vhghE^dv;xN*Wx5#)PrNlupokkFo5&7iVr}%XEAM>dPKg{(I*{lL*EevT#
z&W=qR=o-&3^pGQ)RYS_c)BVV{YYA;b2+oLV5q4wbW*-T`v04Xqeb41?&&?sDUT9lK
z+pI}BaklaZrbPVi?GD88AgmY<BdvFwZ(F*?aeIiItpb|Yj7eFyfu?a>oI~)?kA?_r
zNNhTXh)pdV#>kLH4k-{71_|_8=>lw87?W~{%C_^2W%0Hx8QOOyZ~dcZ_uEM5tdw`k
zJKiUXzwc^rkx{(w?JFzcqwHlFJl@x2ia)0O8Ex!s>iJur{5xE%&Ut%yj}RKZ`R)@w
z^UYsl?{<{52Q3P@GofEZ;soM25({+YsMTb(U&NJw;s|9NZVe#549cks@RTiGvC>?o
zNq>};e~bw*&Z21oS^|iz`e!Nkdi|UTUaT-6cxx_O=trU=_&XD=?7me9ZqbFlxG#6F
z^Ich!yXcgCrf_4rymK{kc|Vx5E--3#RkE@&t5$`gwQ>AqN*T7b$b`NV{$SO4F3wtZ
z{fL0gpb$lKM4i%$a&gh|cYorHcYgH1H*TKr^(*CT&k_&U!Vf-nyt=xelXK3RHQVNj
z7@x6fJW)#{X;uzHUpmP%fBr`!e>r-3>G7uiWCs%i-mM5-i|Rm(iB-Gh6I~<?zr=d|
zEA0D`COFoe<LT8k&Np~(PJD=wm`%8G7&GdGHaHsRP!YDRXB;zqjCkj)WH^{=@XZj7
z-wP)4=*f5&h{>uY+r|-dLd27XOdKPdZEN#9L`fOiVCX^0N>Yf)`c^tGrZTNc*Le25
zjj^q8^<gvx-{2f7-UKL9V&yX_3z^KS2@Ek269nft-@L^)Kl|M-Vq$jhQYKOf*-4#N
zp7S_Tz1T9q+#zEZpvZHkg%7DKbW1vFR`WcBr3-0I%TNpsLabRX>%d=r{ileUk*4Kq
z?Ni=(`wia6Uo;`YsWRCDiDDT88N3UqMkFP2Oa|YSfmC+dmVF40Y1FI;UP=LNDF#rr
zUk+6zW~-`7*+k%mF>*NcbX~)CyPd=lRlQ{*6Q@xfxjEHQBhT_Buek>kFEA4ZGh#Od
zd#NS7kj0^KdXvSnV8-@Xz>rHWt3Y(3I!{0>ThBCx9`9hfKQe7O32>IIDb_5sW=<GH
zR*y}SPNmaTg5Q$2eEcQ%VB#G8lW*|WXZ}0-A(nD%Pw-M^!B|GQ5AZL)^e5<=1K@Z?
zgl#lNE;@L#RNL9;Wv0w?E`YO+^Qv@UPv`4UAkUidNr53vK;4}10p&?fB+WLKAUD(Y
zPP*w_Y|+$CbXRbWR5aIW5sXziI+S2#>x<8dYZfDmWy->=qR~C?BI<Kl78`LwmrF&X
z9v4WMJZXWE)l-YhQ&zldn1{N_^!t*E=9eooOPsWpjH{yg`?$UCaQ;K8Nxuv1egg=_
zMU>{EWk#_y=PH~bQ!uEXug*v&u=@qyBQlYubMMF;IPPXR^C8rrreth|9a@u(j&VSW
z{6*I~w#^0^Dz0y*;*gR-SVN#ML2R{J(F2#4m+WtMG@+qet?}NKQK2619U(MiN#F!u
zXN+t8<tttAWP^-E7>9wFGS{~@)8zbYOWQWnb#dO6<8w1{EHV`~U#Dw+>+-kwfz1z)
zOZT!e_uw2|beOV(dagJVg^?(DF{e}FuT2mv&iv?VSJJwAn|d3|wi9%=xH+}ia=8eo
z8ArKo#i^An$YC<=<$UThR0(t~5T`ifaB|l7d6K@JCZxkH@o+<}aZ{kLr-#uugTg!N
zQP$;CwaIc3kdI6kd-$;Tjx)U%TeH|^nT@{a-!J-}f1fM+VaN8O?LWxwp6}iBnz#9|
zqj$~%_jkM9uNp7&?UK^b%nKHOl}MddqcTHMvr{cFQT@!F9mw=z4L*1Cj^hJ6{=E50
z5gU>-<U|lh1Eic-H4Q(~{sdp_KR2z&>W=0NLI`a`jA;rY=NA{mF>=`NNhz`3Z0Op~
zoLbIhV2))UxdhQ#iIuXA2lYOX)O3Z0rY#wXvfZrNY__%<#W)rbCgH;gLXrxf**0D*
z`$;vtD`iDnPpgpoG6!r^y|t-?PRl%N6rU?WE=A^q!gnoUq~&5@4lZ>SXzL4^wy%8I
zocrkXjLbK!WFjnhj+zse5dsR9fm8_-1gEu$QJ6J+@6qJ>r}FMIYR~!ei=_|7;9usw
zkKzRN&fg#O@0Z!{NB!PE%<W%v+~3MIIT?Hh21>ajXFYQU$D7twmRj`;%+O2uBU)C$
z%a6T)<am)=n0#7ss%~;*QlKIe>vK-f1OdZxMMqo~R!!i+*%^9ePOw5K3B~1HP-QhJ
zsX$GtO4l{K^4hEHZg<@7c4U;$2G*+%)lAa_#$jOH*_hq2lx%KREn~8kv-d{mi3me7
z!Mls|ja8&bP1a?Koi#qse7KTIsvn%mxwx#PGE-ipD!?llH_fZB5-PcfMoR6qnS7hA
z=3RAdVi(`dO99HTiW%;z$5exxx3zdKoDI{l+d0ikf%*kAKV(gdG97!}rNd3L!j)9u
z5ephlJsLqqkJl5XeX^6itlck({vwkPPds1vF_CGQ(!8tZ_w-Rzx~ET-Wp^zR)@(41
zAZXI*V7+s^^5_x%n_}}6q^!JaP%7S9RSCf(Nr^=!-h0pX{EYQx!!Y!WabztWO=xKv
zPjJGzv-#?=2{^#oP+V!06oNSRhsZDtHsiT5t^=9S96KH6?(>Qk+H0nM>Yc}DVWlnV
z4YSTEO<1~=$!=9NQ3`~mYI7Va^yBcTrn!2|mF{^K6?3Uvmzq}2>E%SSPcPNm@xq%V
z6D&5thYN^vdAsY(6$^YevCpdVYJi<W>S@h+vXgxa_8ku(FC0yK=hyjYs^0>hn6U8j
z?-$FRq!+w)_xl#RE8mkd;4&_|X=sx}3b~nibn$@Bw+!(Hb+v~>CY;^=NRh#INUmen
zt&3fNRo9_e8TtVlr3>CFL5ftj6)`F?Ldw>Yp_)1LBSXx1@pvCVta_?SP~IJCb%gZR
z3_vq^gx`Pfci7!tkwo!$e#HMV9{7j&#uL0Jgoe;~TGz7iTbic9X&`GNl8ngD$p!@#
z0;K5l;--KzXB&%~YWHmUdR`psA_G|l4=LHuM9x+`lSOROx?`OJdd^lxt=T&FO&~Lp
zB_Y9`MwFiHWZ!~)#{<X<r`CPIFY}8^LJgyb7h;z$T3SBG`RPRNK2Hal14@+}07$~C
z-7Ea$qd&{fzx}f~H3eNQ-)G(Cg3CgdO03s?6l}gW1s0DPW9H`OfOlftgm*>pMT_o}
zk-ae|#uyPyiPr*1e&IE%@X)dl0aY|5hW&wk-}5xR!*BCj_C4bH@Ax^q3x#{JEW~-?
zad?Fv?S7oDX*pZ1`SkkJd~EYNp>1$wOt9$`+AJ6E1d^1Ab>%wNG^i5kFDhbDj)KUP
z!7PBQDxf=qb;r-cM8+>zvSsy|s8#l4NhFCxwKCL`o$O>UXZKE`Klj>;;>U-q2BpA7
zlS`2oXEJr$=fCqsf#x`fOe^5Y8R@*4S60sR+QY}Vw}FyzIOnMfy;`e#oIyS&ubxYp
zrR_UqgME~uRTNBJG(}r62S4|UG)4Wh9|Xi3!9NAf0^yW;;%VNr6bNhT7okjwOT=-6
zlyC^1VC9|Sh?+>6ila3b?UHaa-13$Em#shm$K&vrhwVfD-1(p5N6vnPN9T`_&Klr?
zH!a|pt&YyvoAN)IyVccMrYEYFCoUHTMrlwAssQVLuk+rW7I^aR?X82N6r3tJ$XO80
zg*Rj3X|jNdkZ|<z1Y(`+WG`#?7(lcrcV<L#anLf!qXTorOrBSbOJ~3AKsF+LA&WAX
z4b8m*B363K42yl(obGBlYpFk<J=BDhN1NsGpxd%)Z9FJyRJ;>ELs&7HVyzuaA)af0
zidA^2Rw)x%yj~$6yFNA#q+BO)1TP?BO;4I<SrgSG7;l=70@i?-5-COEII`dE&~Y>;
z3#IuE%WMiHXXVA-=8K58`6U>9^ejH(X?)5T_h00bPd~|P>(}@*=YNLRFJ9xd^Vdu(
ztI{5&gWh@`wbbeZxr*!xu85{Av%u-rN(ma}I=Q-Na3zS;ET{IDIGq&zr9gYA$?aQL
ze;I}(&N`B_aJw5m(sOaLlbvj~?=Xb5bQj;s3a~6wVd5zOs`N!AW<QcGaP<23R9Xs%
zUt}k$<T&uBFa8v7ee#?9voHQLn_wma-ewb~Tnb~8l`Ge><Ea78nX@KmoGW@<uJlfn
zV@b?>iqC@cMy{*W)%>Ciww5|qf*PjbL0PS8E#+ML9y0sgjyw)!Q=J0ra!D^GfYtKh
z-~*uvG@&5`>zWr8zB+!DH^w*kt()KG;X4oc^N;=<|HFrWnv2aD!JEik?fkY1t65&$
z=5`3E6)#jV<8pTan^7?xUAfUVX|F(7^^^-2qyk#X_mb5(P&cxMJ;*BL<QY@ucBi<H
zz?GfsWG6ee`=5zERleK}9Lh&b4i+CPy{lSla>@>B@c1w+tlFL6vB0mUeNN5|tU*u{
zt#ka1@BQ0c?RNaDH-Fxs66dFW1S4*v@gb!HzjoEudEAt#?ONm&U6Dp#D&mSQy{V<I
z8J$MdE_J+?;*rx;Rh|CC6sJjrA97)#5cnzEv_r;W`i|fmnx>`eTH3ZH_)wZ1Gf(0Z
ze)`>?;g>J}EuYwaoS%CAKjPCDpGHL()H*LG>ccUW1A~>|R+9Yq{!qt`*dI4nVU<L7
z&OlTtJA@LbQqEK|Mx3QSIXhB@!ysH<jimAL(H9#hJK4#A`wSpD6UQ=@%T!0dfWH?j
z=nvT)P~99ybFpQy?~h_fUv~!vs|Qy`MdMk(B4irZ@Kc}skNKUe-{IZEySN;{g>ovU
zs}nKp9hC*l;=yH;Mc1q(P54YvwJqOUBxEa@{oDvxKX+qK*+&gBs$@-*z+#o0_aQ8@
zjSspgFXD;pJH|LN#*zKOy6RWmiuHO!*S1u(uV>>Wmsd~t(&5W|>f%#;a{K-KN00vs
z>hi>CfRum$WXIHjQc}I6Y|t&M?VvV`PJKe<oU&MUVM`Q?JQIhdMfRgiWIQBzdNuOy
zy9dVY%}09fPIj`B&GsGWM9;RmUUn##rDXj@cUr|3^W?`SiH&)4j$DlJLI<EzbySt|
zpy+Vf#JqVNE|XKLD}PtYfHF28FJnr$73|1aNjmb0i|^%s`KkYm|LSvpkE?iHj>m-I
z6wnk+@_jj&slZrX0{AlCcIvuboS)3uhjXq*-$fZK6aSHQhN7xu2Zcb_lxdsB_M!9j
z_`sX>A#bWG!_d<YJ-hua+wB?K&6YL<D_vH{zTfkEPk)au?7jdh{M4hrOz<$&%CTHB
zGpAPSTr#ZsXw++U#A_`ESZO(qOyNGX@HizI$yp&s8zhBJ@+;#AxBJZH)1J5Az9zgf
zp5n*JPWF;^&+WI>!4>CLD|4#jn}}SSufYUKl+`0;*2JQ5j2XaL{C8ner-REBW0uMs
zviOiNzh<+B=4vndoE>MzMt|aLwVmU@k6-*4e|q~DAVbR03I@tqCN(H4r*uU4N)d{4
zP6)wInpoaDBYVZN50iDb`RBnM!&cN|U7UL^!6X;K$}~oT3j`M|pe$g+WiI!vn1~VK
zDMhZXE_v_CyWAdj)(@kF;}|m9Xa4n*pXUp=e?Y5&CfIxibtP)J5=5MqYhNm+N;%eC
zK~*MO8qs0Mlr^BtF|S-7udc|IU<C_U;lUU)hf%q@>Ur|+Ezh3r8HaC$gFD&DzPs(7
zIS})nmGE8^P96ENWd+UjR#}!$(1cdve1@>5RjY&At^E8_dTvIm*~drI)GRJqW_U~0
z1PfEnv`x#aXRjca=A@FXiKz-1X(|UU3vZYL13Is=mw{oDe4qZ5%b+t}S1XgIKr03n
z+KlJS7zct6ChsouNY{z1pDu1ay?SmR&7m&xHvsznz|$v@VLNbf@xY1$fN>m28u`WR
zU*J*mhzlOf^X|)u)zipR(E$@5q$)EH0<)^_6go?0Sk9^5`>dH56JyM!BF-^HVZV<&
zeR{)_CzlMpDaD>rkdvM4<?LQFkQt*lXhr7b_?%@-B=G#dDex@dtP(bux2{p=?zi53
zVF`Qa3n%WqpGocKm^#mG|6VLJSa(g4&4!#4F<B;3(YllpF~%}+(Hi<h%K$0n!#^U3
zmzk0!Zw5&%$nsgKw-EO^kz!;R2OL7zwn)*LQn6|*wc+T~m8?6vTRQ!+CSn7ouCA|m
zc6C{*(TbXIg5P`Z_xR26H)&|_DpRJloR8vzWnt5`b_P$0r%Zb+rB!b6$G)87`)!MZ
zn3XXpDGOur-0m_@pWO1!yU)129RZK`&9{8cCp+19tKC}$QUR-FMzE}untD8(JHqFW
zICU-ghb-Pi=5p4B|M;K^tjZV6%)Kl=$C0ds*2O(F%c*$0OlKv|;WX1V!7_s4h*L__
zvAd$|Ydsy-Q(=vC$-2oG*_b5<7Nk^Wn5e#|B4v(7&E#|nh%rT5?WctI-b5Cio36Z-
z(aKbAJO$^POH8Nr(Syj{EW}<@uCK4r438f@!k2j`I>5jA=D*>4+wbK^!l#jTh90MT
zU3y`xPo~axkR#a>w|s82=yfed4l9B>C5_6MvWZcuBdX{6X5hW2J1#G8iDMu~k0$?-
z&-Y{}JDJ&cM1oYP)9_@1BwXPebTP8kR@!;uNbI#D;y1llOEzW743>XhWKLZ9?h2CB
zH%h=L@J?1O-O7}vwK7jo1n0fY{O}%90?qu;_EUUc_er#<-3pLmB#v>K$5AzzOcTp^
zik#{>%vY$!sF~NO*NM_MG6&Kc5K~H1@MxOAs_UlVQnhl+$at}gq?8vs?-6f*2j37v
zgAZZSjGAq+0JyW*e;z2bAac9A;rjZTT=*jaE^nUl%a{Ka0+S$S<$J_0AUo%9-r4#3
zdJigBQhqM?#)%j%m}EK_Rf<+uCgT`m=Js~v$>o73mj{M}pvjY?sV<-FWG8z$yJrrh
z{E|DJW<S8f(L2igiuogRkAs^sn7eVo204|gr#t;0vkd%fd9k`a$~Z)h{9@%6YUlgZ
zeNc#OUrK9(x3RE~+6O#bUBJ$?pfZ*%q!f$Pi8<s{gCtAt!%Z{Mp-fBjb_FW8qmyrx
z#ddKDro%9ha-wM(x~`l0CM-*<7o=X}+8;tqc_s+Uc%G600XR3$R#zpbXybG5+=nF{
zUw5mTxw^ihX&bidEt(Zi;1AyYJpa7;8UCja|7R97O0P@32UF7_Hg7)H^yH4}p_cQ@
z0wgzQPE#@><Saym!;pA-HS*-ij@@k}YXA+9{d4<0*~w1!;dZY9B-gA0wTXl(rJ1h!
zu%)A$C)(N1GvqYa_!lQ%oN(5P>n|ZJog2)`XUTc;a7PZbI%p9l;6xUGtIh1I+a*l|
zd=U=Jv|UTrbx=t9X)i^NB<F0aeW{RAIi?EV)$=JqHR)2eyvTEu<IF|IL^Tr&-;uJ-
z6lvRb$~v5LOfhsKs&#rT=QLqwCP_bmSyWP8oJ~tytyj1b5U;MTN^c2^3t6u?)BaG1
zIdOe`P1mdl-lSf}e&BbX{T`nVpXSpKex%$(J1;lgyZOE^?q+#j>cXN9h2m(E;-reK
znS{;AzTNk{_q6BfvpxNofd<v8j&%wkCp+28+C7i7UDP{OFH<v-T0Nt6rN1ObnW3us
zSvlU#QZif?LFAYaz5}$VxsE)S`~Wh2|1k_!D}G8QB#INuJaSUjq2-<Hw|Ta|q&n22
zQd{q4UPP?t+G#dYTx)T&i4lsk;aj*8gwz~oVs?@hnlaeQy2O2BQX{q7yb{zF?DBLL
z)1|27&O|;e6$0wD0Mv+i&6(^Rp=s#46}#Kpx*5w`_Wbo0cKbcM{f>*XbCYd%j;q}>
ze#ZYdywbkPt7osGnqF`W%cFjOUPcN>LgB=~ip68OM#B)fzTNTka?gIqAPsd2VTHj?
zcCwSboZW8#sil|`7-U+l=ETYj3(0eK!2*$xTCfX8<bamW=F;;p5zo`NmcX)v#iS7L
zI&&p6f+15mY2CI$Clh%t7NlHpL4xq*H^0FD=d1saFYms9M~cXysq@w>z^aERRalUd
zaUw1|VoEeQ5qvOyMKY4>cj}SAWX_o}j+WJE#ydybwUa)TE8>b(f9mMu6*O7qGPgCB
z3teSwso6sfYt8~yjOBP5HroyTFc8PFbp9`{Rh=<jGlKH>p+B(QYz(SXB@U6RbjdHi
z`-}Xo&EEv|N8FC7<~pmsV$RgG?!4aDdZ<-zQU!96V$sZyBKv;i@^a7hvmI&d2riHW
zTyVH7MdIXSCp+28*!^-KVvS)kK}tF~JER5@eK8!ZrILv{cf7m^rRk6fN1xPsC)|cP
zW0^9XDqP4klC}meO0cS{_nPr!a@kh#;&53x-0u0p*FVe8y!Frc()gu9=vq~fRZ6+3
z`B4y@w~@IDC4-2h1hHfi-g`m_b5BTh=oR|QYNW6lr_wYHZQIZ^VVVXgRFYk$>MTWz
zyn7-I1Cl3}V!Agop2wn`8dPj7@0g-BIcCRpdqHR#-hS&Xax4v!SaYNgZtDDADk#Sk
z8R9_Kbf8L3O3v`h*Z-DJz4Ix4;<cYN`ITHGN9v?UnKMi~w%Qz6OPEV@rw2xdM3a8W
zS?SZrZHzp<9(np+X1FPEXJhw<riib!-||y^vXgz++C4*9)uCu1OLcP2w!v|!=Fw;C
zh*hl(WKcgmCl8Cf;|`gA=RnHla`e&%q2Si9Q7c7lSk%*ZpYZcv{W*T+^51YR*JM?8
zH#;<Ve79oK`KpIj0+REE9dSY~6<H}p;=<Xo$N(%5Fo`mH=kXyFvb@CyBe<Qoxq8Xj
zDu1e6(~J{gf9UD=`}aTB8boSrDS@TJZ;IlbryqJAJ$gh+iFe<9ckV0685L_>9L8a>
z+f3h^>`B|S(>+R~@hJcFji2U|4?f8&t5?w+r)JQ4mp|An&`N<8>l2Z)GNjD)ZO@aZ
zH(Xy2q|p(AD@mnHLY0+^JK4!j_R;NrwXjstoQjlrfl2HQDs(je{Jc}2JFshTm?A;h
z{s_DJNKJ87uH=R9F(m+aarg|H3{Lvm8(-t+-}rC&)%b52WFQTZ>zf;H_dB|?E!%EY
zf=6}8HRLI%D~XPkt%`FckQlUNCBKC{E_{d&o)A1u2vhlR2n`{4kTN3!rs^p}l5`5B
zL52Np$JNuPq?i_$sYHsSYhaz<%X?8TY|YjVho0Np9giPBMpb!sb;a2CfMqMrJKDCT
zABJh)Su=5rGw*_xIk#QQ=dV7;FMQ({`8z-GcW}OBgw8Ukr0+j*aY`Byglte$%>WgK
zLAkl<xq5cXIAoxmvdkLP=X&V(c}PxnvXi~g?kRxGE5fX~G!8FV%ToPwymfs`7BAkJ
z@v$ViWp+3)<K-`M3RC4*EnR)l_v)S3RB6Q1_Xj@v2cP5Ly!A``hWr}QCHnn=XV0#t
z6?cxQ%m<mJ(@WwqWghcH#j@CTN(h;ht*Q!z5d0*7=!;gCTafHEcv$dVbaNvWdot)~
zj9gz{^7P4j42MG@u?x5I{N=fOwF=2zvWr4WI}evvmo#m|<5yn6ImffBD~7`X&6#l=
zSg$s8T}R*d#&@YQj@F6q3Ih{kWVKl%S@>7){7b&){A2u?*Z!<62GkXn?5<wG@xnmr
ze+op3%cLx%EcCr{doyyljikX_zmY39!!IhZ^u(f{>|`%%_pHWy2X|-T=oj_>QwHK@
zuxf@F>UM6yrOU!m%duNL{<`DiDMWa6PgCBbOVLK#ad5)*lV|+;7k-VOefD#_>)&P6
zk(;YquCK3&d0y40n5>!42b5f-Iu?Bu)J5=}9vF4_oC(3>oTm+ert%?T*@Y_u!E2M9
zPT8Buf>=;A{il>OLx14%=~Hf=JwsEn5@xJ~+P=32oB0aVyER?Ba!+!~B(baI3omhb
zdCA8={yjW=^a$rYS65dy?J#BfabUe(GnRm8{7ss!eK}M+A91{U^Mrr=)<5Oo2L1=<
zKhC4>5y1uf-o?NteeN-udexDlx~N$hO5@<|^*|aNl05`x((tZsZ&i37lBk~SWG8!p
z-Ty?w%6&}y#|)L!sd`gYZBe4F(3BP0lxi)2iTs{3l}fVf;%9xoG-F{3B<ipz+qY<`
zvT}}h-+Y(f{laJXx&6=bZu1@~C7wNd#_eu*oE0JDab(;d@IK&!$EkTrb#j!E2(u)I
z^;=*PBVuzqf?w!FEi|KY`996bq)|%M)1k<ijKfIZ_r!5z7)HjS$Cb~R2;m$aYS2Eq
zMvJR*<fm0ZGdWphnP9<k=q*s3ot;mg+uiJp+sT=(TUiz&%p-l7cWh9ehnRWRU-8f0
z{%L;q=J$A|dCX5d{0Tm}{si95QX?vg!)C(Q35SB<r+^1aW*8IuL*n*!M6<JuM(eX8
zQ`O!?RGsW(CwnRT4v(~bu{kwI>r|#yqSWEnb({r@-}warD$|>bh%s>yW=C_@r|f{q
zu~}1^5PX1~`Pv(A@Grjjb9`?6J)SgAiNna{vrG1e!{Vry`-j8r4Ka?itBzH>A~X$7
zYvHVv65=9H?wp-p37pPb=eHLzP9!DLP%5^@$PgnjMq(Vvsf?*jX`aPi$Zk;!)xK0_
zO*7|ZQD9J^8_C>VhEF+PS@KsAN5$YQ?*raB_Pf1RgEbA?%?8AZe-GEUjKjd$`8hGh
zDHv%sBECkjloH|`P1g}Zi#W&E$FK9*C!gil-}yEE()a#F{_1Ogl^}P9VM$$VNZ~w6
zU~<NP#F*LbM*2RHb1f4uC*~@I>Bw1*)Co>@vXgx@`;G^YS;}LvDt9&eU~&=_X(^u2
zGK7gpFAh$#t)_*dv1HlARI$3)`Y4DnzgPbjAoc@a|Jv91>970~ek1=Hcu0NX>C-0+
z<G47ByU-74ij2F&xZl$^4Q<om+m_G<eA9q<1P7(8S+PtaYi8)JDl3g6!!S(xR=G%^
zOhS+{XEfH3Sp)9ed|08lI>@WEvzsHHN@PKBC2LDBT#ZkTy9DSI)07qTeNTVr*=#p7
zZOf+Im&>Fa_InTrp+TKD7;SO=#W`XwlM{0$=Eyc|3BjWZ@7+A%_pUz6UwY*)aURZ(
ziz0Lp^AwXZaSD=&DX~8cTwm`Phm7W6@K$YRmBN2i-s}`VPIj`Fwm%X9q{2APRf1S&
za=7v!1+%VIMS`1DTPN2)t?(<1VM}^R^EOkiA~ObF;juZ02_)ewU;YX|^Ob+XZ}J&@
z<Jj*HTt2%b#dyTk9GwqGxd<r4VIU3z?22Gv+Y*}As;NrJZO*B{PXnn)zE`-3Qkq+j
z(Lsr*GJj9`dn%2tTu#Yi45BL4UF9gXo@?2K-G&)be35C+i1!xI3a65>EX6x>`b`6y
z%?1?S%gQ)Jx~{XP!y=nr17UDt=jnuDOkCgGu-%^F8;|pjuI*_2oyW8UN6m8}$`z8z
z_81ItWVi3x?G8n*WG?Tm3dhsxq+Xhno$O>UW%o{^+aS5hYgC@1WFl_PM!=k#s}NT0
zo32Fkf{@lSYPQO=%Cs^q)jFl>>9{;tj873EI|c91q%`7?0e=0<|BnCW%Rj@H+!qMJ
zv)}D3cvStT+Rt*=mVrf%zf(1lV`7LS1G|~*fs!RmCq4((I~f|&S1<O<F$ZMvvpCk0
zm2tGn)+zW{vtNzZ72B!%FWZzxO|UNbP^93;aij?ip*0R8rdUSVf_I*7wPLkdvES`k
zZPvIHSvB2som7<&Jbgb*5Zs|3x!vuEaYVDyt-8Xuly3eaGf;<OY00WYEo04Uvp)`_
za=T9)ZUr41iz<@rypu<rL%b{3IpI!nyeB)^OWXZsAkN^WEJeu{j$}uQA_g+Yin*M2
zu{stE$Df+bl5JQ63$?076{#KW_4A6Au`UCXQ*$V)xa9ctFZ~Mt<(K~pF5Oew&~n)C
zd3t$C$`N100N<|w_AbQof`86GO^%rfPFk=zM}<+7u;lVwm_-qf_FuEFdX%axrx>F1
zL`E&aYX0(y^(~LYTYxJnzNutJDJA-GpxdlKghM~jwk^(iBF4k4*K78N!_=@>NA%Vz
ztPld1&#oB9Xc?I}D^d2I-8j;;E1>k@h>gLuha(rBrv{PA`6xqFuJ@T|mjnG?$YZek
z?XpRnj2?9!7koL@gvcp?oa|&TYxf#JoVY?*m(}x<!l+%-G6#}cnPu-8c0ZH9Or)^Q
z>F|bbmz}y?oTGY@rU}lGg`;$}=}t}ZyOY4Lf9cow`(OFL^NdSc*K#=YJh^;&RDSyJ
zGqG>B8A@96Ge__1oo#DG(dnY`qE9}T9W5F#r>vp^Skq0YpJOUFp7{~6l3`aqsTB@V
z#$m+yQ0AQ25Ua)on$Xz1leT5E-7?KkDKM9++CH;#*dOrTu{k>fF}2^rKI5Ix`lgKS
z&G-_O>>06uo1Ku#uqsN1GK`tq>pj<3w~R63e1oQ37A2+H%~@re^|t(B-P4nu?7PeE
zDS%YCWwEN27j352qTm)ztl}#Rs>2F_E1u8%ULlc}uHc24LJ1lZRODbKf#3e}Z}b27
z;@{($KBe&u{W$XU@~OdA#p+<Fe{ZW9#fz>(J*Q=WSUR}7)W_X@=-q$6IA5BAht6Bf
zkl5mW)vsUhMm_{wa*X51&>t9@hQr}NXacKMN85T77;|LRb(Zxh1m~yN7@9yo4D1fO
zGQULZT5Be$vRSWan#O{^CWMUBj825VkSiBdP9<mIa7bKV_uTCEWu$OhpuGYKngV7i
zFaS+DaqA~L*~vb--CHMmh1i!I2g}tj6<tzXfg(O=EbY|gJ;V8hN6<ng8@%JN20M=Y
zZz-WI`)Jhj#Wz3C-}~(UnfKB=w4o)X#O39sb%jeUgZ<W4sdvvq>wo79@m!23%_#M>
zjsvbvdnlXU0mzPEznLjo?3|i?+?Bhpke8|zHB9?kM3%M*tT$Vnhz+rdsq}f+{B`fF
zW4>tue()2&<;0aPbo*W(8m?}xh%wQ12HK?*Ion>)9}e_~p6zCXh{HQa<6IG{ER|>o
zPK7KE6(u_g5kK(WlN+wDdyuxkTW)#krHtAcFO^J^|9yk`$xe2%7ur4PH)#2?8G@Qn
zmKQkXX^icHSe_l*)EdaLN$}VY&K$>-nYfvxp9WH4&i&bwOa9Ss{(pHZzeVFs=<o8`
zGsZYfnZi6ecmA4VnUxPX`xo_B+zpP)jAH@o*@@-(-_vW0VN=r%3Kw9Zcs}L{(YgzL
zEso;_rL475YB^RIpj)j8-UGsFwPLkekz&M&5JE7XBqf8XL`pyrg0Ez_mQH%%>gw8p
zOBu9f0jgmb2A*C%<95F{sLA;<x1}_9){^8x{Hr(=VSk8RK0EN_>5iet-kYw!r5>sv
zp>jy`^*q_hPWEzkPXXlpoYdX%MfJbcxif!`dF4L&+7J1if017oY|xUFlHp<B^G|;3
zAM=~}Gc*L86RvNr>4$y_T1&^igyHJ^jc=RpIfAk7oCNRl-V1lK{KEv@)l94w9@ysd
zJvy4?SYB`12Ip;lNy>?Fj2sRJhGDP_BW6;xlH}UtSc6`j1Cq1ScE&Fq`a^*c%S1^`
z+Fo%CW8~^)N6awdNiLAZPfyKJf+Y}Dx!nyseY)rJ@<7g2Sh57tY{ABQHm%z*N6?Ye
zO!SkT>|3z=m2(#+EhqC^vxd<NZL#XDl|vJzKY0&DwT_6b1o8aey65^G&O0<E6f6Dw
zjo<qWzjE~}XqRbx%kA#g<Vo(PHIsj}*V?xImI?JaF1#QMktHVgf=@mN>wSdpe{K_5
zT&ao!)(qwaA1L1`PVhbu$9cwky+-Txn)P~Z-SUxXetXSKQa&C-974m>C-0hOSkqXR
zf%THa1fwbm$F3iU;z+I|NdshSFm(bsmz=_E7zVCy_dI!a%i*B7(3S72-ScJGRvl24
zF#<hv>H|61$-V`<R}#HeJxwr#)c%YGth4Z4hW(d%p#o<~t;(ul;>h6@hp2NfnaJQ$
zCYv?Wct=X`<v;ie|MYABh$r1UbkY*z$nEWInKgbKK$de#x>bi}WsKwd2fuIe8qIah
z)w$1FV9y1X#p?URc6-;1iax^Q3j9U6gtFja3UTgjCVag<!F#r6X9$p!%`M5LstG5g
z(mC(MQOCJ!&NNMnlDWFNDnq*z?>+sYr(3PL-raC>yCY^r#c|c|c&EQ+yM17j8YP;{
z&7!P1ql_tYd%NfI@`jXLne3>j2T4UAZINwFuxNfBzB}2;PWIt;uNv0Y*P%$J{Luxu
zUbbO=ZRxdj0b?mlw7|T0)~P&n>1dBovl6IshRb)K^7Fs{-||*?lOO@-9oN@4jM1=r
z^07y$Kee4r(;J4Ny#CQV&xd9zAGVt*M(QH-sTtTy?zmnQt=C$nKS-U_c>Fo<tv5r}
zD$&}e;kDOZ<H5xRZPU;;HbiO;F!mKrT<KV!gVAce;_1_;jB&JTDCZdl6Ml>_^6oqD
z((ev9A8=vK*Pp$^fBnV(im%+hNp`{dP*iY)>0alY+3yFguJ&AC9Y7j#@&;p3)OSt2
zWX`CTS>{R2{bVP5X}h;y8xaT&f<gl{V(Cv1XhaAi_O};&2`H|V*w&KX+P;@X$Xc-L
zh&d<4BE!UWFT3665vY<O_6L6FH-CfQkY6GDkv25!_InOPZ}nTb4lt@EJNBpPR){$I
zeta%l`a`-(v&i2RQ70nkgFkx2328n14@gdQ`p6ux<$3hXfv<zC#37n#L*TX7USqvl
zA*wVX(6$}UI~y^JlBERntmK#{Osi=ch8THvb!AmrEPZ2RTy5KK8{284#+=v_HMVWr
zwv)zbY&K3Nwyj1J+qT)D-E%+h{sZ%2)>_wfo(H$IB_a~<-`2milT=4n@|pe))+c!x
zc;x<5+xz-&Z|X0j&Q}B>S<zpt=35Az(EObQQDea^RwxJcu>9aqM>b%D6pMjbYy$Qk
z+Z7+vnQKGx$Mf_RW000~HC6})lozuXbBXk%g%Mx=#I$S~oa`}=IlkxfG`UrZeH;w-
zMD+UZ%fNB(Z|04^P`<rIx2UYt4i1h%Y@=N+Q}Z(FR#|BnA&J9R>&~kHGc~=pAx}ur
zZwGAy3OX|I2uSFIIpq0FFB`UqBQf3*k_rx23fD?aTDD_Zw5{zLA6{L>X@kv#lA{_>
z9T#T>5{Ug6PJkq=9Gt4UERlOcHd48$zRSz%)R$K{5(2chU@~(tf~~d>-hVqk1N#U+
zR12HLzGCJ(pG6k!dvLq>`;t<9cAjjY8U3s+PMV&Vmtz;bOZVTGY_N9qyi6^+sIF4a
zgU0Oeke&Oy`OJv65$rThI2z;ROLizHhpA5C&z!G!XKVQ4>G$0aJT94geDqXm^qffq
zNr7GvqFO?)*KtSr=#T!?^X<OT{B-E}+{sPhuq*A~o6!4NQJ;lsT~jZlaog5^akhz(
zj3aDaige(-wriv)A|7R2C7V{ObG)*v3zVcv`7<i#&W%{!nyah38y0hfb*`EI=}iF4
zhc*~om3be5-tKjXJ3>~-gB^l8jiupt`YQi7-|%0=d%!<V2YvF+SZX}<fvL;B%Ys(v
zPz|SM;387oQvT(E>vy#L|351@;yW(!3N=k`lwYqfH~yL7L;W%Rr*H4Efk3a>PHMW@
zl1#DKyQCl&p<VHPiR7KScc^zHJ(F+{=yZa9P$iDyL~bf`9ck&>VvYx=8gHK^jeWN6
zTJoK!!G}1@rx#^EbaPLoup30wxpUfqtG_(!B^afIG(aCua(|LKy5LPpr_HHhMPSm9
zWT5*;*VF(^ZBUL6VF2tW78e&OZ&q3uNiRHo9cpUvUxP1)p<B>N8n*Iw*U^?aIOId4
zDJo1S84cV2Q2+ar{;&3ZzcnCDAh<a1qv!aj<(S262tFB5ky{#cge6pZHJ>TU2OIz2
z4b<qnGypQY+-rS`yY`)BZZyG{UK*ZeUQ@VYI*=dVh#9l&dRyXqAJ6U`y0np$4Sf!-
zJeO?%RDiMQ&>C&1s+Vom3P>clvfZ_vkHcL5o&*>3o*qwMd1W6f2Hf6<_-YdmOlXGE
z$zjARQ2Oj=TR(N6bHTD6_l3fvdp)LVB5WliaippYz{eM&3`%KC7$s#bbe4KgZF_uc
z?G$+HVJ<7KTC(jz2euBz6m65>jSP_uoXJTA?cDSTV-~LXCFhWSW9z+Hsx7QItj+o7
z{)*%6u|6*;1XWI7^y$cQN+#^9wp0!N4c!mOV80!`X^Ve7dwn{f$i&xSev0%J?1E^%
z7`XY-+?(CL(60hQ{1#^Te#U&GyUWhe<aLd&dN4!6P#}$oDd_H~Nn2Kv)$Q4&FBy1}
zFBmL{LchA~XEE`q87FaJf7-7yYzX$ba^L9Emho5H+)cVmDI%)8mDl?Trv7r`2-^jK
zN=0r4I0sDPR6TAko|DshUU&Z{YA8-;pHy%kw`c>q=XIE>DFy0!-ooZUv?2w`GoVxk
zxe}k#PMbM4P?~xir2GkM4TQM-I0aY!QEGN3vS!*ar(ijQyb9?>OD@)qkV3%+bFD78
ziXIKuHl5sz3@m>|Gme84RZrtu_V9cvSSiy}iy@jm%s!v4DVRt)tIQ8lD1hJNW$m3~
z{ySDp6-X2TvaZ+w1LDH`xaEpYb;N#5S;%gaNSPk&?LreW>W1N#xgDb)ucbg+=8e&J
z;f1`?#?>eg_{9l(lT_lWM&X6wkz3<=JKY9BDgx$O7!8Ot6`Ao(i>;yRQBsn0yw$16
zk5ceojRB1gYe680&_oe$PsoZuWWG+vJ@VN~1I*0ibwA~}wsAMJ*U3KQDO7x^lT#Zy
zkO4SXBQ-A|Vej>LJono@vcKDnP+l(JpxqIZAJB(U53SP^wr6|>x}!tDzJK36YB4nI
zsrh7Z{<8SpycJX8H-q<;UskyT!ySJU+L`@yr5>I|_qS$J%XvOXx%nyW&LIz7c=7Bj
zvOeR%sF|*L=xadS^ZK;*29ugDz~9VJbM4zvy4%HpXakOAb#ZqecIdTiK2r*Nv@iwp
zr1r$;;J<ElEF@NV9(0!};_WN(u6C!wdY1CE2&}U`K=TB@p!M?5xDgh!x*L%2v#A%1
zx#i#nu&HmwaSsye_5Ot=F3#WCZ6q0OM@jlZQjyGv%vwrxYx&=qWUzI`8kBN3W1R-O
zkb0+&&3M?9V(3=@L=w8_mKj)=;En&?#};>bukH1AXW#SmfEv=6YJHQpz8RN;TIy`h
zQA0202#%`Jb#`_A32NA~lWXxKin*u5A819%XYO8d{&R6_>-qhzKq_Xk^H2HeXyM?t
zZ~Su}Fs?j{I#mu@2Vz_5xfV`sjB)j=`^7gB3FREi6(}mlQE#sU0f$#FpCZ|%0Ed$v
zi=U|x(OFO^<kt(f_SYD}twDo4iU?=`!a^{d2x6Ox&$$)D{3+CV#IoMF;||Bo&1$?3
zbkT0%lU&8%ZmIWe^yNN>?3o$O+tNjVv|AkEBINX`hQKWK)8<WV{^fGB&^PT5rP_TN
zuulL+8$Nfs`{7o>iRHa?tuQo9$uGdz(x&XiAk?WShs|J-<(hFDAJP}0<rY1G&ISs4
zaFd5~pe+6ZM2op81@h!#xY?Yw+&V4_di*XYdx_4QhX1bXxTK-1F6H)FMPY~|1bc)C
zwH-MQBUTY_M2w*@%9&$DNz7_3d_7=#@ex%acT<$cEt|8){`c%lYW6wdQ$%7nl0!2Y
z&$(WD&JDGXdG4_Z8wd5JgHl>rVT9h**y}|s!%f^@wmDk#JijasGY%caNQ4Bwlpo2#
zErh>W=SSfvWd6ru#BNeVIcyB`XIJb!EKs?A2wE|r&1@|LSWYaq%xn@GUyRX2Y%ww$
zlk2a&kFy>3C-@PZO%zxqv*pP`EK2Nkvqnai9a?oA9BPD(J6AKV7J&=^AGf;+l78{t
zz602QuR+!3Je21js+g`I&luE$KSmEw_f)iA*JLmn_M;Y!WMz|o>u83>mL`3=EgCl=
z_0g$w3|g-OGVQg`d4ntW4`*~5BKnkz!)9_WV`nC@Y0YMRtwTvAZU#~iQ>1=-ySuP5
zTfB8J>omJEyqR(L=3dTz_f2Inj<?qz(}Zdkp+<_4AI~9eW*C}dVsmuJ|2Jf`q92sl
zX?vPGtiAG2;@Wsv-C=yg`o@H;@+OibtQm4VmR#m{dioX9kRwqBS~r(LYeI7x44%o6
z)dmb@RH;~XQ^%R^!f$kK{1XMNwHK`d*B9a&D1Sbj)V@0A8W?o-Zk$R2HH1rVYmCI<
zHf_Stw@7<(tk=S)JVX%~xp<e1x>J(Iddsl@RqiL&qN)gv^P|x`g;^^OOI4k5#@p-?
zk%$=aq)qa=ly*b~w~`py^z8adnZU1V$Fc{EUb(rsw`Zfotna^%Uyjf}T=zl==ciAx
z<l&7Hbirs;L0&n6>c;=4kL6O5*?|?nAbE%y?gc5CWJ*|qMSqVa7VYj)gEJCyoVDf4
zl~X+-xed&>!#q@&x}4eBO??aQ-|W}f;wa89!vjvnY-)Y4Ww_mpnhj=uu<)G3>+C-y
zf}df7&CW&y35q&NgVTv-_8L=>eDGt;!{I?**C%g~YF0Z3nMB&tHC*|5h!kq%cfhtT
z9rICN_RKF3ifH_E-2J`l(mAd)d(_Zys4Ei_SiAMx$7P=LWUMkP86f@{Ro---bbW+(
zN8bIH+BJ_RsN(E#j;^uZl!b`_y1*Lf+g9>tDA+S_$FP2#Uz|@k8*^C#hwL~!MHeUD
z*5JzLKgOX`k&)qhM`LbFcGupfVj32~7{b$Nsvx<HM>H><%HG*3=TDA9>}YJqpRNdC
z!itVnyWcz%;8p%;R&w&H<!q>_mG^k{)<SDt-7xMzEn-P+gIv0Vg}D`^Sr)!YZ@@l(
zwsXiLlHOyE*kpsoD=dN*IMX>ccJ>E~sZ7`7()xqHV<Ue~ziu=ON|dNzf=Z#xT?7UO
z^Tl&35t^^c_(6&`<N%Of^n$SX@>aiDS&87nzq45gmyk8lYn-Ff=~}DNqDE_=>{=+u
zHK(9j<|Je72sf^I+Q7jp%e%RDf496xT=`h3#bWlbqOg)GXI_Obc>jVDDD`{~In%D{
zukl=8^sXiAt=0b&S|wX@I^6Drkbfup9y(DT`U9(*q6HPArY2n8CJhgsOOEy87A7yB
z;XplKjK(YKXCAf>+kw2TILT9LiY-VE*=dsT#J2+5_X5D{uQU&{Y40Q(*{txP=*fYB
zSL-Rqp(AerdkR#c?3d*@6rJne6q%V<FFl5E@JBGkZ^=kahz1I-rK#$WOqM~`ZQbql
z0LCAKyO7YCB#SKd(LcjHA$6VYhjrKZDy8vJ#l&BEI~ECofN0_w^)B3BD9>2xvl6`=
zi;5YxQWSj5kSHr)vZMz7q+Kn_{TmJV<~_p>5Szk)v1Tj4Z*Q{_yz%*;NNKAF`g1Wu
zV2O|vGlM>3&Zs^^J`x|Ij6}|!4m|M!U&C{3O-Xv3C;GtGo~Iw&Plm_TdWtZe&l=pM
z;7lo#R&p=Z)yJ)5S856l{Rq#vj6!!2`hy(mIsayZ`;*xsamSGF*9(Gok6!tn2VRxd
z4wCI~lAtbGjr{gsM&t-)^@9XuLp7o3OP9GWrS^oxf&r7mq41EY(q)OQaf3CMA|kY3
zBMTQ((n;gh&((DhkD|c7T0qh6P3@JFE1U7y((=L)T~B}0sq-A4o#UMGywOz;p`=*H
zD#Q}s%oWT&U9>a&QiWtHk2Fl~nhcQnqNJIHrv5wI7A^3C8~6{COwaE37Mc9Po-t%~
zZ4LDDI)H#!`X<jBBf2H4xZT)*DyhKmhAI3ZyAXfgsJ%_rW;)L-Sao<NjOJa^V9z{M
zzmME++g{g6j^U0wx7Y)!xpl5+G&U{-M&Fk7WdR|&o1n(<LDo4h#btM(BdSwG;j!D_
zodJRiCgylHP-no&M_WD*S5I4-txZ$y<ACQg#U%axe!3-o`f;RZW@v1qM0Y8x^xFtx
zZSNDo`Qv|ptPH+$T0828P~ZuwiAsMIBpBbfv!}-bFMO^b&sm;bx^Mf!0e|#2NVh&+
zsF%cE?b7(y>UX1V0Sb!Zj7Ziw4r0AbyYeq>XMdLSzKJnFgJ>XvBDG6@c-GaYt#*}D
znn;+ZJL19T52CM@fgy{Qg5q13-j9#2#JzL*|0lbBk~E1`xV21<Ur^C5%1ElbgMxLx
zUFv3_ZsE3=UX(vLPuKL@Ukzsd_BW{Zz;Lho=1b*;lB6}YDs7Q_H!_4&CyJqLXP0(4
zu12iky}_ZVT+CV@teLEJ(pbyFwzRa2{pbtm_5Fi#KxGqjZsBq~OptKwecO__Fx!UR
zL^XX$_ahtwN+)CtvV>~YE)@e6-jsg{qEMeExgd7MP5DgOEV}sn53@X&>lEyWeEPvE
z5pL{w2K@;Hu~U`vqWL#X?(cm@w56P+joi~$^|8-@;mFOPEe(O?Se(Nq=SpkMNn#!Z
z|3sdxc-PL6u3A`jk>eM5W8}(jJM_)_3*5f%jtBOx_#do8D9PN(3j^p#IXN+R`d1N}
zEEj7UlvQ9<)i2_ORCNmrG3grNlHJ!|U2k@y+*_A_&Mvdx9p1m{-EoG$p!6pyPc~-^
zwYjDRH_3~xbD@-sQv47ik?eXtDvutVWYzKXS--tlW$#dzl!4SVtd(*uE>~0>u3I6E
zi+}>=a7-PXa+sgq<3*~bug5rkkF~`o!hikB@AD9eL*aiarWy9}S7;mkuw&hQy`~ki
zr8ZW|z6PMLut?o%)HknR+Y|d#{PKN&VXZ7Zd@4O=iIr-#xZlBqpoF6HT;ej@V7fyy
z;n~$4)+r4o4C*b@8kxq%amiPkRJmIA4OqglY!mX*AFjlhE8Ty+pU`=ijstT`eSJde
z{-Y+Fe9=g8lUfmfxMFCt5ZZt$#v?h#b{H{DusFO!j8ZlGT?LS?yF4e>716ucFaEs0
z5O(1UPB8V#&%jvM*&}Zk@)Xfwc2UrfaTuRD&xjzzAoNEd>OOB#XjeD6b(_%JI(Vi<
z{g=MHUcr!Y9P)cIUP<~L3E+=$#Mv#NX|cE&(GrdPMOg-OWY!{Ty@bI@;gs?#yn9Yn
zso91iivU0@YX7%I=i%&NgcX=*BryjCo9{tj&}cQ%HE)LyJpB_T@cIss?uitCDhT;O
zl;ilCA>#m#XS>Gxa28gPxpwtuBA!N$fET10kyU5R>D$vHR|%~!Qbm69aBme{6c0TO
zm`b#H&7hPnjgF4>(AxJ}ufn3ymJTW)OP{`N=0^S>pC4D(K-mZMKn4DP@Eu)e@27@x
zHr#i$b@UP>1Zu@oJ!+bWIQkq-CwyzHS*nKf$o9kY>y5jS<h;%5n~@vw9`fFP?)&54
zXE)?laR{>-7h20r2-F!TU6_5}<<OQb#LJD3GAxocv*8sQ7VYAq<Xa&@x`J_$c%Dfn
zK9GLUxcyLK%q`Et(U^ZciJM$>PGRW5{PIu_up{?Nn0c2SYdq?EV?)Lw+?KK+8M&wy
zYe+9Q>z41UG3n?KQ0%n_vcJ|V+^H`+JKqKwpLpLLB(b7tbRDt`_(6j0FmG#c2q=jL
zhKHfiK7Kb=>W1S0ec$m&bY4@O$mCXsTqqr;xI_?!@F3bTopaW)u%fKCDgqn!;wlY~
zpPSqf3By~H?O^^2<S^}#BiFY;`PY#=k~d(^D<Wihj(6C`L>qE?1Nl6@cY@715VHs=
z>6uWh@T7j4l0a|!9rvYgZ`^aB^3O8g|9QCIxFO!top+K+O!47(lse5UN2599#{_Ks
zX^)m>oG53~ca@mXyHR@;@$284ZnZ>F%ukAT^$k1$CGmP1gziVv`hMOZAK%(k9g9@q
zC{w5af^K3xQTuhNRHN%3RE>hyc(~3{<=0tI206kt#Jpa1$ojcaM9qlEBR+y5$^rfq
z*S!12d+Dviv$K&VQ1KTkts~ahBK?V!$daONU3;gV-Pb1}L9-=v;+?_hl>WE0fL{3E
z%T<Cp{BNlU(?|-{a?A3<jy-M@NKFJ<E)5q<DLhvd4>4A=GcIXyxWr*}E4(ya?szYP
zAADX^0Vn&ukTa=j<pVAu%yj|kD;>iS)@S7*sxHsAlnKcnp*}v^w?EH^zHYDIY6OgY
zonyW;_{$GK?AK^i$(4vsnLTRUN)w4gTz<LRM^Qbr6HU1JLfiG6z+MyEBhNa=X{%(L
z<@>R?3?zDjK*diH-NI4}*V<-ZK;ls5y!`{l3rTuk*jMaJa$>|r{p62iIEIfBmRY+w
z5xDJTV~<us9iwJ}2vcy_i*s6)LidrbIgZm->zty3EDz9c{?jaIxRz#%)m~1{!NA?_
z=dETH%i|lS6c@`9L6lh5kT02uRT!uM+ibXD*GT-{_FwA;35z2zqeS?l!I0#adwmv!
z@tWFfqApO{oGl@W1~S}GJHhKj3omD+!aO}qfz-akS+DJ$CbTk>NlrSvmlpx13q(_5
z_J(Fg`n1m1uPGV0Y<Pn5eNU7c#B$U}=IDK1u|ILmfB5{ieH3&1^>iLJL_U9h?OZ7&
z(e*Pk>X>)~phqU&Y?xJ4vs|aF34u<0Z3bUbX$VSrm^)a_3k~~lg2P_cHNYWZuKD=&
zC_H6PPD`;)RTkkh#f`hp>N3LsV1@yDoDP$<til^p2+7gJDX!lduDPB3I`K9m;TI6u
zD^2>r+&xd4XZ9t0%0d+|^S<y8!Yc!bd_n_<c8VMC*&$6P5rcV_n$uPP27FDT({`<@
zJ+N>zU!u&YrYmv%w{?|NXJ#5!^Du*rPqeF-O8J(w5Ls7xCU4JSxQFq|Os9BMOFs15
z4cwg*RmQ$AG}Kq&(~a`Ao?0PtxT4Fpt5Ws7UF&IHLS~|E)i<=%+wlXTGU-sI0IqhF
z4Jk#%yKk-gS+!k!H<_M+`P=LN0M{klz!Sz4g32|a(nDNiKweb~)?yAKSRPCnm)K$*
zGf+ivK{db<`vpU<|8_!V(Hr}V+$xT>0C7}3uu}~*)HWY$9P-0{Jxg^WS}G4J(w=;t
z0SlP`Rz>)C45;vo!4;Dc2bM&xSP2H)YMEoGSGQ_M6CzML*=U*0-`xdc)VbPV;AzHL
z14#zCbQm=bBpnvv|4$>#?{|aYJdjMIxN_-fKFo{);TuDucPQ~v!gq4|RgTwDc_f}b
zga&jQjaomGtQcNLegt7dJnHlIkmhfs+sjZ%B*aA5yrTv=tiNgpj^G=YQ-1Dtk5Kwt
zDPcO<8QRa<XXK{0L`>Njit1VrXlP>f12&=IEYC7;+FfEIq^c@mA<+uIWS5%D?<&{v
zj{f=`DSz{Wk310t_ztrz^|&=L!c!GNSt%l8MuKAal~r;B=C#LCuliGH6@=b=_Os@%
z+B}5nL5wXyUbmzt{HEG~>n(HOR|QY(he#Zj(yH^hmA0ZVrR<CP&Vcm7PHK{Lal{Bm
zxyU*T8pS@aP#bZ})x!fkyvnl);!GUfBN^u5z4m^*rH6QdGRWnv@b;ym>-*GwUjs7Q
z!l_}nEFmdKxF0|#|BqO;93QaFFqHK9*C60N&Cq`L3|HQxkNDD%L*Os#<p<@{H?dZ2
zc9cwD><ON<zv6N;kYv1h0_IxYptnzZkY(lZM!e=nbk&r0C5S++RqV|Ojg*r*B?M|8
zq7{nD^iFm3e0+qwT_m7QUtA5?U-3tol}uh3;`)*cY23A*?<1YPQ_KQ?v}^8h$CS<1
z+(-VUBgSG=Y{v|iZ>Kj*v3Y{i)jKKBD6~_Z@u~?s?N=|K9|nVviPJZV$IPjULLB+)
zC&fu7<Zhga;$N{qi#&K%sRhiLVzXAD;EI}npT*pW_q^$%iy?6k{cM$?`H4LBX4YQ~
z383Y<D&|*|@QPv=uw9Go2qa_!1%mE-yg`2y<RjQJ=PmjS`!q{%-b3US=GtUJ;2}JG
zQviO7I>=fZ0eI_me3}*B*oaFf>EP%AK_#C~SZ8aUD>+56lq8+{zkC{$BESagVpJ1}
z`%}Ac<ym&?Fov|HAhDIkG7cxr?p|hf+@V-;i3gMCXh?fu)}w%6NHn8qF7o_z^Ijpm
zfS^^UAp$*ns&Q7TOn%6BOChS6vcB{!-}u145EbS<cYeup=&3Sdq?T=m>iljl8!UI_
zLNvDTw)UVN|HBX~mX530ll)QFD4zZB7FT4Sao{goUBX(x2=hV*YfdBeKnBCSa}7)a
z2W+K|NVY2ZD9je_Za=@efjV#)x2Mp4jco=n!&Oc90fHDE|IT11JCmF{40q_t-?N<o
z^o9stlyb-3_Yy|&+S}U?A>aX|!Y~8$-Z9gZE8P^i@3i6yyojQkZ9+8#qZGle&NNAZ
z@4Fk#rY$^xlKYTUvyF`_APV&65z2CvRTiUvc=1eF&?}sKXzE5B;x;%_NQT-dYGGG@
zN!BH$eMzz`;fT%>nzhlHdyD7%mQGMHznWhvY`f{kPNhtlz|V_Z_`iv?(HEI?M3O$s
zSnM}=mvY`_q*H7FlhY}6`TbqDcc8ZCQMQXS;3q9y^1|x3MeC(#!s78fRq_PbM263h
zPZjqX4Q1bIQHt#vJ_Qdxhg*GX1`@?!=DD(;ZInCBmLI(IFAYFO#eO8m0K<|x|64;K
zjZ+>!_TD&)M`2gLqsFvRd%oox*@{h1&48+cRkYagOc%^Y6;AuLZw@3s1h(sz$E-^A
zn@K(-s&RRl&!h7j&)P#bC~v*Re)g51`d^V-Tud@~h}Hs>S^HV*vRHvF)X&yfn&Y7>
zi#n!z57|zCSbMGq=_sE0mRxgK1*hz0a}9e@rcuCSUP@^%4)zvdg+AGMCJHl)B>+D$
z%gUs>tx*n9uBLQ>=HQhln_sI$-v!P&Gv|0L{Qq-pws0z(u}RoTE9o&fK#_0o)_>9+
z|3dwH&la5e{>)I?s8dI~8s4qv_OlOA#{#Pp+|c0`Y1@y=@!zrGf+sn|6Jlt!QaU*U
zX^EZ*OEbn>y?*|I0TkpDg*bhsFLde5TRil75QkEbHYj1PA@VpXEtsqsIq1qR&x4dG
zy27z=wPt;8b!&CIt)y@9=&04*?9Q+%DZvpqi2<~k@jP2q<B-VS*Sx`=<C*6nD}}Yx
zxq}gIDeXh_L80-b*bG+58ts^0#gp2Tcx8*7kq!FxRueb0L9FJlDFvH#J~x1d@0M~b
zrHRxTb$cul#kW6ZGCBCac)>UsAn{Wdb3n-;eAzkjrlCq%IIB$>Hm}B_{{o}JBnTau
zkSrYDyueK*Ft_YNm5?*=pAGfyKeJQC)Hx|V+Hh2R1g2u`x89Gd^p|$WcM2bQp!M%8
z>UV++Z=8&Amdh|bh->cvncWe*D7;<}FpWXM7F)HJR!aGm{OI>cag%(Er6K`#XZAM(
zJAM$)35WUjr{<?;8*^2p^NfuVUt$3-NV%egE3!toPCM~qW&-~Qa0VS(nq7Zg`?x7%
z&lFb#ha^_-8IiHlo5uCuWCpf3kd%oXC2`Z?mtgD=tMYLihk{f8uD=i~7~+4hg{xZD
zpbc%pl5Xr52Zta+W!XF#um70~4ZcTVP1OIzw@tnoUJ4U$OeE(ui@0y~GfLjpERab_
z_og}C5AD6#ZFzAI`jgB^*_Tf$G7Cmfa3wsB7zT{wL0Bv@<4G~RjsVldiher!x{Pzk
zsFiy`vEB`bL8w_nYQsiG%flma+vUA^&)oA=+ZG4!r-eU;JjpnHA5gU^nn#~sFaJ|Z
zk#0S5DFtJy%|d(*T68wnMPJnm`ppioISz3ZZw5#bNkVuIOiZrq%Y#0o+V8N<I-Ke>
zIB>_~Dg5TEpQQBh;ol5!U7bWTFP1%Wr7PR!u#?zLCsFDriI_L%dG$X*nq@8>isPa8
zjkJH-klo}?eMjuC*)C9XIwZ0b5zPJ(!|bF85EhxrPWORl>}KbpZn5PP3p1AEN~y4R
zzRYrPRd`}FZMnu`AZ+YO6kFE5sUyWG3lqi_u2O|_&6Lubu%VaZ6jf?Ivzb+8Uwp!c
z(G(u9(h8KqG?nwtJmi5HY_TJLRrbIBPaYO;v>p73+)KVqC6;f`n~OK#nb7erb-fn5
zn?e$MxiSg+(R1CeTMo@Fj-=T#q@qKrF-y*OQcA{6K1PP}@OOElfXp(7>ZgRglr`!d
zGg-l>2S}*dAr+kj*Sb9Zsg6c9TWxN26d2?qf?)QG1g@(RKZSm%wYo|bpM<c;8I?fz
zG>S>#SUcB(DuU*v7I-jNz>PT!6UjNrLsM#WP6XEebGd6I)?u#XeHO4xK(t_yk-ZTk
z-Jt5LZe!W@3<)#fV2rH0mJClptVG>B%c<K+(l?r1fe}^D8hHS#$c-?=`YwRxYMvcH
zLZMK611Zfc)#yktsXw2kIFad>i_iR5BiF}c7J@i5iw?28&!45+5mc_GqoDp{A-1m4
zq3N)&w}j*HQBk{Xz0av{eYx}Zqsxjf5{M9g=9{%vPqtnZ@j|^#<V?&$>N_w?2B-;O
zl7E-thD}M3s8AL_5SD%D05nCI>zjfiYp4%k-FoKUl|>`<51|2~<mZ48Bb)7DApCG}
zrPfHv7><1<$0#LF-o2WzIj%yx8LUbcp)55U<Prd$Y7MKIX|iZzoclVJsK`rc;}<41
z1aX-`?fK+Ta;)c%nJ!5n&?prf^@U_$JEj_6B4O_PggaAq2)X_Za&VW>svg6N<iR!#
z$+7MpG-gSCjw?kRQD(^LO>U<^sDqPsZn9_Rer)#<l;2=0;b0}N>9bqnHknah?3i;2
z<d|L^O)0}M(J#LjLBX2m^XLEHalLWH4P9TtSr`{fGkdc|`hJz|c>m|`GCn2w85xaU
zOH?rDep{yy1M*?J=PwO&<6w{rxr#=uGm*7!1UjlSzfdAv{Jv0`41~L7aQS1P>kZ$V
zm9u~4VOeSI7{_aRKUd<`pemy~%;T07q%KjJCX|khrVh`(`UOX&{lrUrA8L;!O=1|B
zdiO;$N=hhatihM&fZ>3$*p58J-K;iRXfomczTY%)L^><C-8GJ@i9;Tl=@$TUdb83O
z=-01(r>x?BnMus0kf;=!pwmuoFQWY;GXI0T(qZ|R3vhn10@uA{x@f#Y?yz~1suN%4
z6BcRuWP#iD*YD;X;rK46PKHlw3l<OKffa1H7!Vx+#CH3JCY1?vHLEecq3?M;2p9Bv
zW~_O~{$!E)+)$&ZMD`SaXK=a4<g~=G)@<^YVP#ZBg6XP22BU(*p<Oq(b2=&pw^oY6
zf;Hh~=Lf<XlIXSR-yC&ra<1D`x;`S^8~W|Gh+kOnKvb>(Zji~>GEeAN;V~b9Qmi<#
zCak1|Ru|SYMH=`{nsE>Vh?N$G#Wk#vJk_5KPJ!bXCQnv@9(mj4yiXY^y_~6fU}cTP
zcMrDk9kKBZr=O#2CI^(5HA>Vmke2EhRVh*4XYyT{J2TTE2UWe4rm#%Zvm0vYFaG^X
z-_dygn+KQ93mLFrn|?v0Uoq1-J;uRfG%gWT#kCH>l>WDaprwqHYTMQQa`c7-N&9o&
zdECQO;wRYQ_o-QPenf<&&Bhw*e)Y)I(P|)c6@hJ_W$#suwvvgZn`_CN;TZa^W3Jp}
zQ${O7j7^68^J|Ffy%9N4rkqI3h)S3`){vMR+qYVo3Jm%<r!?z|I+;9~U|W_7P`igg
z$#lslNIRBB!C^P5waUJ3cs<5AaEQ|boHa`{%U?*hDu+`~9y0I_Jq?f1A1=q)L-q1$
z4@d+zlA?U<T*)|flw;u^{zK^np0;@q`+Yx$-Al3HcFSxrSMJzfO*w+`3#hn5=}#g^
zwnR0k*&fMoH=`j#&m!*)*7<6Pj5xb8&OBd9P_<!h_}=`_MwEykMD~aaabgb<X(ojq
z@NKoP-E-(W<A0B*#vYWV+bu@>71n-%NPQK$7OzY_!}x+cv|y}EVb}<y^m3Cc=N$<s
zK8GUEnFDG?)u3T$K-x;kB&8G(YMcm5Y};iFC$}VIq8!IqXa%lx8X|q<4h5=R7@LRM
zUhYm02hd?i5c(;4#D^du19DNf7&L4^Qy(?1eKWCpEBG2j4io=ZDC_Uwv1|o}4UMfW
zcu1TR&nS@kwVr0zsi0&tXWoI?=@Lngv`E}TKzB)X=o>0iC{QfN^G9iaYR!%GzKn%F
z%hZUk9KNJALUXuJrwix(C!}TNGRE-@(Y<#RO*TP_RGNxK2hb{|mJFhs|93e&$zxD2
z<SI5JS-tWmj^v4ij$1$Qy4LV!|H~BNg{Uo6Xy%5skS)U`8@~Hx2vUZ`v3p!PE2g~7
zDAt57CIJ)Q_GK(ChZJobq}w>1m;;9_RD|U!BpPwwKcZtc%rhKw%UY9mbkq@DyE?Z~
zVt`p#QPpIjq|;6m@#|%dpJvZ#tEooBx-NzX)6@*27z)=4mr~y<`dw)S2&NGsuwW|c
z(>&BYHlpFg;OpyYy#2}fY8-6TdyVfZArj-L@naGH@(60XsnkvnPFY=KjSXhp5A)oJ
zh6Yk|bU)F|Avo?9Uq#krq1|GI;mGOThGeMiYRK5xY<nP6WXcrU8|}j^B3DARJSCDV
zc^)_Jni6O`hGMT!w3)5?4*cIm^L0Q)7xA@p|LnEzvgcU)RfcXk*Sy(M=yz4kQVCp|
zOo|4)c3RFNyxCMjEKC_<?7omY%@({0@oxa6GA=@2q$#%dr-(ukY~su>pvo5s6evWK
zZ{Qm6?ow>z8fkx$zPvr0*3+Iptd!oYhf9f1q_o6$suGC0v$i%SRw^;lQw0!SnWqZ4
zyNv?PQe_F<71{>w^4L-(L!Aj#O|~I279d)&p%su+FH=G{(fhUbbJ}J(gxd0ADT$NP
zDAMENCbJ38v=Pz&LH(t{qdwtMCZyX?UyK-$4Yp4zO5`<WQS2{Dx}j>3V^`iVDOb~=
zGABF(3ZSbdm>_vg4|-1b{>N!t6s>;1VK{39RKyKV^_d5R5`Iwsv_09^B1+NywZNi@
zl#1(jh$^EfdTFi@vltLYqxPf5HerG7;yLV>=8!JDyA@QAe)1ckSkumx057!)$`0Yp
zXBuMn8jQP^cv7lzU}Z=WAb%I|Px&~0(x%H>4yB&vIMOw)#`g2cn{!q*v1;*Cn$;hs
z4CxcV1*J;kuZq}mgTpA-3WKev!H6X&$vPBjv_Y7k6__$@7O<ONta;x&=aXQw$xqBf
zZfvv}tSSPq_N)XL<k#ic$mVmtRHFvOY4N_Q%1P0p{2-uICuEjy5w8v<DGe7^M@ru-
zjrt+AD_*XO;tqDB9rDr7_|00&B^)!a{{25eJDW!@MjGBE4#mb5x}W=tI+p0{&vD=e
zjO^lPS={HE(D2e7q3U8fG``b?SZZ(qs}5a7Xq}EWN&l`9WeY*+cZlr9O8PPy=WW1|
zR;ld=P~?slJcKVAwAa?8_AaIutb+8elD)nNCscA5fz*qrQF`%1s}x4#YLGjw)Lxoy
z)oOK)PGFbykl?%I57{Xfx7#E{?(kB&4**yA)6DY$FpU^zpX!F#PpAX6W|Oo>*blj_
z;(?bLO60~t7#7iV%qSL!uFBocX`s@8RC>myS$xKr?4*h*{Ee@Y^Bmw$YShnbKG9dj
zv4R9B=7-m2=TUd6rj~Kt6<6<}i(p?(QOec&PYhpjxy3h2F+hg+M&_Eiouv_|;Z2dC
zS^;W6Da1N>d=9)TYSxhi@hu`A>BLE>S8Wf}a?y(LUL?$E{#tD+5SiC@$1V~UIe<bo
z7n3Ec^Koff-2k$}YT3CM5l|S-T@==s$-thE%Bw=r%nGt8#4gC|7!wn&uPqm7XrWIi
zuT~Qq21hNIBL)LQ@}+NiS?ER*oiE~Ltex{Nh_w03MT^C=IGNnM97V#_H{0L{kUuTO
z?8$f}OiW@pi_B*|m@erR-(+fhRKQ$G7Z&G?!x*NzW<^L&A+@Wku^x}ZPALvPjA2ot
zq|H~@X<m4A&y@1qd+y!-Z!WiZCJ>U?+2Q|Kxr6_(Tyg!0t*DmWDBwIlAHAqijk|`C
zT7g-+bPLTMte2~*O*d*+q@eHY<<KmV8yXNF2QLO?MJ^$3C67%fNlGxS^o=80MnvW?
zzQ6`eRok<?Z7$oG<)>0k4!Re*RUX@yb06ZUtZG~RobJ*Rl$Oc!3@A~3a^^Dpr5K9C
zE<F~9BHL<iabqIM5nxu1Z1D3?`Z)mY#>`o$ua>(hn~}*a@;=RP82G$K8|gX!j&+-c
z8u^=&e6_lR&Pfw<Ea#rm#j0$lbka04^T6frd|%nPXCIEW%a}xWz9b}hzEI6Yny#3|
zzEa$Vj59_%<!j1*&cl?#XGp#QL}9-CX%4#E@z2p!()RWTTSWzB+oP>l$>ed%q74(h
z0m(b211j4o)vByGOr40XNoaAy2pc)~S2S%W{Rmqq-lpOJ2>ac9E#lvQ1+5}crJ@O@
z6*(jmk}6@~hVB<bqF4(3m3IcBhBcI*C+S}?ke2!80L%8=VKjzUs$uf4f~sxBE)fH&
zV3NhzL_Us{`qf!6dQ8Tik&0cZj1niZ^;035pJXEsg&X0JugPR_4~=B#CgtfxkZkyV
zh-I@jrif_l5>+ptU$lb$pNj<5CyloHES~c)G>xUC*@f|D(^xuOpevS<2BSI(Ti`U2
zb0(aUG_IZAm4nRk(N3bG`G2mvR+DEOA(YZHb^PqKY5tF|dZn$WYcY2*w#9YoWoaq~
zi-;l>9n#ox;_k%k56bkx0RFuk^=qvNw{M@qs(o@tYq{X`Q3bs5>b4NkSCdU01!E$1
zP{+r;lN5!I^Pn4J4|sO~EQqGcf=&#b%9TxLLDa`*4};>LX0gN|orrlY)^-w7^D8Lv
zDsvsz=no%CA+d7Hqc3Ec9;K1)3G+M4wgWyf$f$IL{ns!~A(f%EGNfZRb&DL-Z5znW
zBJ1vuZWV?Z=|2vsJoI8(u|M@eCu!%qGIUSmZBk_HBW&ig#`kJD_Y5XbkIr>EL4Uw<
z>b|KYI&g?3Qr)6}2Acm?4Fm8MBlH-gmg6u%VEd#u(vd)Sb=iB-Je6UqpdQ@FlHC?p
zW$v}mz*g2rF>K9P>D1+^L`14XxnOV0CXVSbKUc-hBk>#YC$#s?NnNAD6+qyMNPBHB
zmc8&0k0wL98ae0F*@Vq%s%d5rHZnP=e8C1lc%qAVeD-2TwXH+rchi#FX#IngI(#$F
znIRb_mCkDMm-6PxLhG&$MXi3HCm{l(beFs;!}fJI4UWxlOla`|UX--_nLbeuV$3Qb
zn>jRiwOCwTwWs-PmtgY9m&ej9>=@`HVWD5u-fMY_q7)&@IH<)F2c>`5<mV7)I+AFR
z#istSI=*w5e+oGTJ2$VGgMR6a#MSC`kF7^A@9TY1cUgdf#a)ryJ2)v>cniHHahJk!
zP(FjfMmch#9ElPWCw*%eDs}>6%bv?TRwAN@TZ93nykYHSd0|!zg_UhmyLL=+cKmQf
z8FE)mHX@$;f9q)yycU1?L|gcorr?NL)tcUfveC0d(~J!{l2t-w1KX8y#H!decOOu~
zCU>C+%2SJSjYR%oA^!C3tjn1RkihmGLD(W>n+=+KN>>u+mhq>MYJ^_9_fZ=#Lal$0
zKrD6ct7(}bZFNOe$ACl&wGxMMP9k`#M?)`|c_6Yh)tu6YYK3Q9geKF2JNa6NZCIMX
zu$pppt5Wp$c>Xil8ia#Q?(W^x{e<bP7e8RR^_qS$&zOhSMZi$tt)snWqC{hfxp6U5
znw7m5oTtkXEk8M*L#q2P^d2=+s|`h9sTAZStAVXFr2XgI=BF*uD&pxAmb9qwyxmB<
zxkv(~BKvne#%y=0WRP;9RO46aEku*bTgUHZa+6W~)FZ5ks#&#&xGn5+n^B^oC7*~2
zm)HxXEiKE@{6abCi}iO0ZG<bN=q^*~psR*FQ&B2FEh*O?s-?g|u~dZ_kh=PKqBy{y
z`3HUsufv#14!r>R;VLatubO%4-DMT5P7-1;Xi;6&O{J7t;asbAe@v%QmD@5}t>T}{
zSjDm@)I0>JN$&o!Ngvg-{ph!U)@c4{SF24JA?GLAe8xfJW!peSF`fjM*_!ajKiQyc
zaykPR^_;sLP~Rt)#8#+*L^y!N<2tm1SHnbx2DD^}Q)X-qks_o1f$6N%F;Us&`_ZcF
z#6u;d(N^ZRa9lBoi{vfT?`@Y4=ce5^HlQL=f?$9P9MjgAt8&`RYFVFgqgR7eNBxUP
z)=@shxb?Vlzcb|Omr91M-@i$f)!qm-FMH9`kp1JyXz>g9GR&BLl_`gjoWgBGrLowo
zLRw;Wvo(~eUVa6>D}ckQzN6F$QJ50R2)UiLOq-th@9s-)fWYPk1X~w@`IUd|vv$fr
z*=r^Gnj^{Anz@!9%++4GW2{1CPMgY)^0~j$n@!#q$EZS`SVqk&Hw$el(PrfU8xXSm
z@`#MdjVP~MlDH;j)Ry}9_)$I0n<{_O!el|h^cRe1JPL`CnB<T#LNn7T_f#h)ENXE2
zpObdG>}ZOnJ6)FJ_>@(`%W)&n42Fj8fDLjFm4%dboqviQ9Ye8@oJ6UMyvo68nBKTv
zCFQGH0AH|^+;ibmz)N8?)1$`XKm=WuRr@qAE1qoSwq|NtuIHgns>FznKaE4jzvM4E
z`avgM%x^w)9FxuxX0^X@QvUaVb0BZGGXBf^BQws+fSN@~tY@|AP1UKb0tq#L0Qp2o
z+fI>y;y+L-UuqEQ7OR9S`PQF`>0gLn<V>QE{2GM9xEio9Z;0JP9Ek$l!PHF;nyA*T
z(6^EI$u9*eWoyGc2}*k-Zm~8wBBwd|M%e0n`F46iQA`~JUl}abw04B;RDOY?=y_>=
zjAXqe4J5E#`iOS&i0Z3`NamE1gd5=+VlSvgW9Fz0{8Q*qs`17B$RKj50K}9;n4V3S
z6QSsY81lSMGH45K((`ycnXqW;b7WZURsF;rXn-%iBj7bB0^z2aB<i2+skvdknd#KY
zl|jOeAtT;}0hE-}K&_v^?MP!S_C{z$Ogrx@3igp>?fUUiUI@~$T;CGrlT%8G)G!!Z
zetr)E%8OK&6v@mt-P<gx4k0IFKJp~1Qd5a{C$1zFpwfeh19E!051sB&rYfXLU_`8}
zMDXj>C>J>`Ln4^GoH7dpSw^6>u_-aO8+59z*1xDhi0#+GPka5Gkl81i`PDnMO5={f
zY^|oiT8Ax<Od<(T*~oe6g9<bO7qbd0yv!j0%?XBYJEY6&e3r&NKb1JU)<kAi*0^ZQ
zSy5Ps-h_44gh9I+Wn3FLKWHC^4IuAO9Lo`uf<ISmUzUzH?f1mJ5-pO}X(=cbG`&jH
ztYVCK%5y|Rd6Lyb{tYt{n`bN`9r%22WRFnO^$zPpZRz_pf4#s0`aZXzKsk&dJ5IPs
z7BT;R3u!M$0qCv6cAp9}x@PYw-v~-6M1tc)3B0#p4r-SMWA!BYcgMx1DljqTTHMDY
z)oq=qi`nlmevi_f7}-cOu65jX`9a_Xf^c6o2F-<sFHvyJJ@V2yCwboHW!`fhzbO&>
z9v^)xDk?<!m|t>-gcC;&Y@kLnHIYL+hB}WyEr7%Za6v&p*(#hBZMr(TgV%kAPoXUU
zNi%U<9Rgn2$IPsgf>asHJ*%!5^nSJonYPZXvB}m6sL@$e`mRVYbLu#~AS@4eHKE+=
zu%g~XQxNU6m^zZdRpo_$fW{F;=3m|XL37o!|Bb^A+fSeNi2NQQ)+c!@QGNMzX;&qw
z>DNjo*m=u6H~^8!HCSl~#Vu2{B*0)Yc*x8o+CAJGqW8tKE4G(`#j1a<GHZHZ1{Ubo
zMjGds@v2pWY}46HAMBRs!6e<|e~+RAdPfiB#*w-bKIwl22G&tY?K$DCIxJT$(LY7>
zx8ZKrZ8{(putqTDoqtHVATHNnZr7b@m~%L^^k7<2KI37t((Nc8BNB_ulNOLsBd0a<
z^;%$Kb2pt}767;O2&v9tVL`mCD(1(?Whrc?94FK1$_>pn<GAX3naAb*rl&h)hy5V{
zOTcpVY1psilxX_ZdA)Y`Se{cT8+v%hOaj+m4b+3$RkPv7+JP~oKb#1xA(ET!D*tup
z$WNV^Xuej2g7BXct;1~TWgww@<Fc9T3B%(qHey9)EVPm;m3W9m1(V>L;Xf&ho?vsd
zu<V?Cz)Pucov{Mc3)2i2SjH3?jzqQi+8+E;dcKk;NZw?N*4j#`JY+tE-~NrD!1h})
zCUfR)o#}<IiU&m1=h1D_c;>WKuos0dvOM$WOBN@Vj!2M_!gp$OB8ox0+nxe@J{+0!
z7R+fLs*#M+#b(?P9$G9ql@LdU+|7w%1*BSKZ|aE!1`I2*8PmyPJY;QRCn~~<MN?U}
z^+ol*Zzi1=KCh~#Qef3Y)@buc_k2ibF+51=ko;bY;hTj>&mD_NKmCgiorUnlzwNE{
zDxNq>ZQt5VzE4S*iQ4-2MUrdMwx?_E@EuV0O-m}sr`cDa@qC1&PaB&}ES~V>Y?Yl2
z7QAJ)-?cn7%p?*>KeKFXfLZxymW!q3tLUgMP#ZmlFeTqFUzjk`Ot(m}RxblT8_s0i
z=h)+kJ$L_Yc53SO;IC&C_(_YKWtBfj0tm@SbQM{o8UJWzZ^tpgLr@;Or>0>BT@hJ*
zKk-DO;GsG?pS+*=dCK9O1<o-wGH`Im(;GY25l1u5UGA{mLiwp?6x;umOTFRK*`f#$
zG{LZ?^zUMstG7KM3ySDB4V*Od&rXpOC_17c2&`f*W978?f|9SkE1V1u@^9yF16Kl_
zl>WB2i!~3rRU&#vH6b1Q8e1;6hUhexs$+{l+g6_PR4XDAsuS*kx5d%I?u@RZ(a$1Q
zBwbH^ubFH9OgR*;JU-!uA8p^9hcFGJuDPWU$|a&Nhn@hz=ee?%&MMA-XINi*emz&)
zUY@WTh7f#R|BA5=O3iWI90rIYFr3R}P-9RXg6b*l#|N6-_F2^a{`w^vWK#2`>q|KB
z?&TqllV-{M@G!tK3l5wv>B-IjD@}jyXQ{1_y1nYw!GS;q(jALV68{bBFad>BW;}$V
zlGlDLf(Mb%R#7^+FbSu=H6*xJ%*g6@iPLc)8?^f9Perh$W{EArV}9J$5@Hf0ZUN%#
z)N8uOa?78dw!!gAGfLP!yvwsq6Q8P9m!Al`6rWWDmKt5@j^@uwHo4VB7S3i5$4Z5@
zo^1Ao!qz+p5m)4U91p=<Mq1O_-H5*kJ^uSEdU<*p3sN2)O<RZTkOx}Ef9;;*{$6r!
zR3yB(c|$!{bcd)9qOkBr6CVT0sxohJK5P2mt|O>#zwqW6y5w5K$Mu$(AAX<Cm_*qh
zI&~p0Ni~-rDrb_<*Er@B%ebc3->Hj224(XX1UKV7Q`K7-%l|Sq(bhlY9+INOC~D<r
zK0lK!=Ur&6ncy-i!|kb;Ct*0efUu8BKNoDHJy|&$OP9x6)!SEcq&%T%DxU{+gGA+t
zhg;C-1U+04szwiM)rj9l)uJ+zHqR-_lU(lrfN(TZ`;DoVOsl)gZcwVk7I*&mw`B)T
zZw~r{C$9puxPWcLFp9XBT8!<-))9zP`!YRi#I=qpp@np@zt5lfp9kr=(pYVQ9!SJi
z^wHYFoCxlUtZw7lYG&LQPrMq;k>b;ZRHTNr+|2?tPlpi>Ff~BB&$<1BxgTIyms6L5
zfVWS(aRJ6Dp#gLG*4Jbjeiu~5xU${t`LbkZMT7{+0WwgJ1okEXGOtHuzAv-JidHwR
zP#O79iP$E#K=OCT5+1zqTi$l~;|HUDE?e=7A)MXaE=JX%qP>31Yv8nGza#Y&Zl6~}
z%NqH9lz!IOtyTHhF;(;Ywn0)JXaN{gsN}UDtXb>#BwJW5@IQb4n_&YV6##k86g~GX
zy!O2%pFYVmN{b1rRkD@(Pq!Ig6UxCF{!N;w<P=_Erg2wh@okRgg!-N?k`|(#nK;jJ
z`~D~F(8Cf-7@o$;)j7b-m-$LENrk*zL!l3JW=lHs`50%dON8wBM>Rba%~BJFcvL#J
zvJH)9SgON45?LeVe!w&iKAyr?H#`Ahwek_~!{tp04P7_iL!-aMuTDnQfQD4wZ%ioa
z&!A~#HEj>Y14iw%EdCRgc#y5du#*2a>Wa5zQ*%c%RAA!quKVZ$n&X+b1lOVd5MNQ3
z$sD)j+AXIc@ztZHS)f5@x+7uS)zsp4XV(;uB$4xwW8`&~(Aq+)I#Fe_2iL#*xvxp~
zo*P^Rxfz?+EN@TVZyg0jse*LFAH*^vAuM9g2!_Y7*m&b&c;wVQeGcPWONWXo;GP%2
z!3DLZ^UW|2qy7Xf$X@0%=%*cE@_s-@r1;D#Gf93-Tw5j>h*g<tS3y(ra${N%$UOYB
zegscz3D*i0>S!;K8h8Bm{7gLC@k3M<)j?kych5#N`07Rr<h6J71<zws5wsPg1kh|o
zHE*#wlJ`jWAmI&AD=412E*8#^ryib!{7`?CE1;kTMCfPUxx5ls!@J$!U$fTRNm|0W
z$#i<zRmgYuAw?=lbUStHk1h2kKD~tN<-e@^HNymwP1qOoS9ddfwFhW0LBaP@c}xHI
z-XDt&hfYduk0W6zwX4nE_BS+^X{uHF&#Kx^hsx`Mln&MEQ85(k;RN2PtGYP8e{>VP
z_ow$iNg=@?-@9lj*m9((GsN-<XO;XFa%6=_OQtmWzm7gFr<_g7$c|`mHasP>safNl
zUQTqEv>LRg@pp~e+<#g66IXtg69*?k#qMYDqZ%)vl&|Gm>vFpNXfLMfhd|};tDZmc
zuDQI1*Ph|lcf~3Ng8_!Uch)wDpj`jZ3Y|G8sT!q*dpk1Pf-Fz#73!<0tRIuhg025O
z>E-=55iRNix#14~<!qOlyqzojE0N&dTc?Y!!_`8oxMVOt^spDPVuhgxIqD~o&8P4N
zAmUtvFI**SOs-72kD-H_3am}~4xcL%l98xVZEOx-b)`f?rUO!-$Z}|TU9#E5xdMvc
ztkonX0MiMLi!f)GP1Oy6=I-UGil>#t$8mX)llKAot^O=?r+x3N*Nz8;O1!4ojcx)z
z^E+|>>n;PYXO_9|Yqq_w1_E(VvO26eGgtGso#%g_My|^y-()rZA4yli5M|qRA5u~l
z7Nono1*AiyW9gRecIl8@Qo0+-B}KZsJEf%sB$v)_eZPNjUvbZM&dfP88zryYu&5uB
zP6}zKXU+y9L3@Us!uh`axPe!9)m=Qr_zD~6m$C6xtA39C_p9%ey;F$(MBmEhzL%7>
z7k~f3d(Cf#xq<%+Sq2B;{uF2De+N>wMkWNTxWQAne+f5b!!bLuDm#{#v0hfbtCu8P
zd(bDM<3+T(wPd$!5~lJ8y)F8?qJ)R7P^dg-e?oNxW5)oAYG^J9S4VZ&nQzq8#hT6@
zN!R6F+Yo%JV8}f9<ZmgZ7M@@8A-B=DfqhvYxcTxcAg{Ci;Bhn}^KT>im+~BAHINn2
zI{%DwiwUlysx!~YiP5h{JsT38RiOrGqfu-ClD+xYE~~{>Nv@_*QM_T?Pksh|2go48
z0Tn)nvFJaGu_)~{#sbWz0tB++b(|5VNcY0YTDF!L@d}a(vLRf#400KJKDV@*tcIAr
zg4XB6$Xdu1JmfE^GTC0Kxkg7j$Y@e}h_<=hBNhgB#fA9MT9)sxoOnHiooq#d<j=}@
zpi91r)eT1pO_+ivqbp0!=4~>++uWfas=o-`1LkW?eL-E=C&qG|*X;z&DivKF+_mgC
zZKXH9*g5T}a${KgG-*`^e;23)SS*DD6m1&)t=bb;BnulWlIs<!lOLc*akovlo&hKS
z3q-N-KvdVFY|A(jdm3t8px?Bw6tTdHe5B&uN7}gQejHu;o%7_PMklyrV`V&%O_&f*
znl4wha-cSu3ISJXe8ho&o=073=sp(I7IO4o(H&LJtEM_o)9#ovgMVF*y+GTL8LkvD
zrP^~;Da1dl&8WYBNaZ3m`K)YJR8=I!n+)=Mkx|q9)zGPbz!G);-Cvzh!B!yu0}?7)
z4_fQV>6p{S8G-Z^!O1TNbl*3Y%UQnX1EQuC3~tQB?EuuK{fTa>Tr?#8&8CxY7vXA~
z{$wlm2IU9#6QVzkk~8kBV09?=C3-h|Ti*n?Bf7{pr*J~65G7Ymn>bwly4Wpy=Q#@%
zIx(*1&OfO9`i{uab~AgkfNmiW<)(V2#C+t~!Of$T#EMvI*QCFWcUv#zs|QJ`^c1|#
zTT)^8X}6Yqs%E9~om46XFTDilO2mzb8yNe!H!(9!j$zm;T~j6=<b%>cKOA6nLb>Yw
zG75H1o2vZ4q%AlMOv51xXUT2OS*&h7ybU)-cg!~%qkBA7zfTdcgxhRJ#NP-dq@e(5
zbzN`!UZiVWjM-05L{HQ2JcLQkW+*vdDC+6AK75+{>1kglST_Vm%hh+)AO-|E-P=%`
zTpS?*#l=?PDxpI0i|EaBJKv+?H>=hDw-K^xXD=44923So`t$dZ)SutiDnQXEjp=c!
z+TVb!?NeFNS<@rO;6;Y^)@1ok>Qth6A~T8p4pFmgT@C8~plt0*i;~0OvLE{()2gE}
zjupmwR{d)`tM5jH+^x?JWeeS_jVmC5jqqj>|FEjy@DB$CV3rPtNMF^=ztyEC&c0O|
ziv@%0!JRwI52AjLre!7yD!cVG>iOvQr#3nMvumq!jel}1_EUDmnuK*u59Um#ecY=&
z?;74p{5wA9m@^Dh+pqwIlQ--G(E8SnmHyM;7ca3@EVdHkjYREGSr1Id8X3E@z@Jm>
z%MeU{O&AT8d2;dxLOdD&TayYZ&rZj-%Nq=EXQ73HzFHPK_4n)Nj24;udpCyughxMf
zWzSUl1ZJgD-abHlesX0pNV!3whp#rHQTwj^lCl_|EzM4~A}6lQdX?~3wX~My82jT$
zS+MMd%1<9<6ZUrJcTZK*6$1PA#gk$;BZe;DtNj<6S)+@Zw^Gs<Sk!&ppx&!_R*ybb
ziulFv@#~N0NO*4K=e>(>EDnb%%8NXT$%%I(Xgm2kIU_p_^x8Y>PJJ8=E^jfnOkzhP
z9d;f1bn<}}&ycDlyQX6{bdB2<G-?1u_XG^%plk@M>=3?{sPUc-OpLAPkTXW(7RJ}C
z2sgxF7zG;-FJ%~pI&6moHEd6}Hir>S&-x7+9G?f!Z@g^ex1RDeXI1^;oYr2|CAH<)
zKI)}LTd%P-0k;R8tO!o(utu^AcOgIn4C<j6E-m*)%gbHPCzMt2!0`R*b$>hfZwGfx
zZpi55^SRce6KS&pM98HQ4q_MYPTi@5zn@P;)`GdwbJ#Z4&dHkIEle&vXm|)KB{)Vd
zoBwWe1ZM?rcwB5qu?#mR=}?!pes1{q5&3)TrB;*ZxTB>z^RHI3<8geu%kqzjif&jo
zXoAIWKk7r&3U)*H;4}6%0*;IfN43pQVt$hF(N&Y#+ZkL_8*Id*>NHTdo-P$=%|XGo
zC;25l>RqXn`gOCZuKL$azH6+r?OI>2I~B4L1q|Pu#CoJ^#V8-QM5|*8FIN}OxAw4H
zStvV&<piXG*K2M%SMb|t)_DexB2V4Jhppa&|54MMyH2wxvyQe><+!sE7&i2>3@i=N
zd(qaqY()9x!QSzl-X<w>Ku$J_&Jxonp@ObhoQ3PXU7lKZBHBcks59#m@<keYXJUUd
z(!kEYW2M%TClz+$W|g6%veUvbS<Ki2qE5#iw#^O!H*+o$$7u&iq==TEtBi&v{*=}(
z!2V=Mkl_&fpW}v=PUl}-7V{mtn@IDff!D$vsHApI)BQHbHBx;Be!Ozsu9nBelv3C8
zwQsR#)5B7e#VmAO40+{s+Jn*Lc@0@3d<&o<0{^pO&%{40{HFiTNzrLFTrqI85cukG
z>P61(>Q~%R5*L-Mzh_gyG+$+^r6&(g0Jo$}>q`<43i3g6*=lybe7qcY16q6;SRwk}
zMggs3AY{fOvFnq7uA#Aw^~M_+Th=H%`LvsN<zrY87q5&Qy@$=P3fte*=88zo{ajIq
zdb?U*5_ZGW#`#~CKbzypGd9=Ms^2WtfXcz&-({i*oJYIk6yGZ39|^K;zWkXJipz8<
z?@KCxqEU5xFuyH5=v=oRjWKS4LoM7c(N14PTRM*)vyCeT+rK2ayDtB>qCJNb+BM$}
zRttXlp~w<G21#RE$G6R9a2qu)z)#bgQDX&^FMJ9D;f+0q)f=iC=za9|@%eqYH7@ve
zEp7Zdw^i};^zX^=<#v!7Bpu$Bdu|Uw#>SjzWz(pQVRmnB!U2XB*ziYdxU^S0;?ed>
zd!0q)7fVsV`8upCwNjwWrRmpe$|`K1!S7;1^vir~NoaMZE&u9Y2*A-ZUE6yXHBf5v
zYZ+RBHunn{>2T%3xZ@EukK*5ASUYRd&8o!OqsyM93|{Fus^EN4Z@Q%JwjxvV8vn5E
zpq1*>QO~!~X1Y<0kl}!5<#aLfbR3W@_c;p5k|g~Mt}6-EM4QR?wpyw#00;EU=1_%|
zjpha11qI$8P-d9rt2IwncxDj<u3H)d#yE0tWn*H>>#A8zjUso{2tkH5<z0u4L99Y&
zpDVPL9pir7xb1&1te!tk(cYjx;2@Zb2zslo$7zAFI@#sPb#4CZCVO$CZ9V72txJId
z#437twruxWiTu&bMHy-l%Iz~HV37Fdz}g8>dxZ>FEP}Xxq)DzOwK9OA(ZZs-bxozz
z)AxDNXax5Y*^juhzjC^+Nf?%Lu9x+aWQ+6aS&Kl2)jfvwnLZyU8_{QMv?kHiO;+4M
z6?fNb1lhIfMmjoa%{1wj7zAddCi<i$CG<!xG<xDxplt{B-ejNU_WB%-VM);3;W(ey
z@k!}%&SNVL9=QUPHMI=4H1d&v&Z|j>d(DU|d?L8P7|A3%%L3vKD8ycJUhKZKJLHAN
zL7b}n*b#0AI>y5OZZUU#Zmr>axG_3v@SRZg6NR#x?Noa-b=L3BoCYCi*)9hgU>!UL
zCr56HoHWNRHm)WPHL3%Gi2k$>b)tCT1eLdMSsH77A8j%%f@3h`Cq^QLlf_GHptJou
z6TF-(-OsBHt#>(c1;XW?taFXr)rnMSW<3Q3E6Vhm;)$Sm&=ISJlriyw;D8*_n_P-d
zFHbPcrPcAm!6Kwo%wxBq%Tc0G%Gax0$_-IKQDgKCuck^V?`>Lf#WurOKU4M;qj73O
zv_hlgXtMULSSg|6Ga1Rk`tFdxkG`fQyIXSa=Q)PR&EpjPW%R0Go2d;8Z4ZdlbmyN=
zp9O!*pVB=y#QRI1M`pcQ5P-pDkIkEBp|*brtSUr|S1Ejh{ntm^RVhpcO%!oIJ;QJD
zCdKADaJ{bpv$FXM+GnHQUJvx&*z=$7xbm{1%?Y$;QNv~ctOeBOW&>T;r<(-`GUdWw
z7$+fuJlF}6w8>H>RMHlmSZhwaw#J4jBRoPiW%QG!KyeWxq9f*puoirCqECqyPE$WQ
z-K6uAm12TiV{B5qZjCxTX^Q!ALx4NQ$!~wwo8NZn{5`rhTmGFNdwdxE@NG%0aUgI`
z;))VO)Aml^(a-hly_&M4?@8x*BaKTO-}{a7&Gr*}!m!GNU9*(8AH9KJkH6{dN&DDe
zRoknEX!o%#e_P%>*EYKDUlELZp@^%8FZ4C{t~2RR>(=HstCR=dQ>ozM_s>o1;2N<N
z=PpYbo&s}HWA)J1$E637Q_GpB_SiC~Z+ytqP|B7}Y>*p6sZ8s7^CV`|BR0*`-0QhG
zH*&KQ($ZtEn~D0xt=i>i+mJZ-;E|})YA|~t+i}h@xY}Ah-C0M0d-9CZ@9p<VS`6b>
zLBb??W?n31w%X#MOrp$l%Go5eYu9#tT~cw7UWC4p1G;IN*GA&G8Q<P!CPL%x!aFlu
zrq%IKt&O6N?h3n-6woyyGpWqd<cxy`lT9gY1CMR1wrJ%1E!!QSc&xU0zGL&vq4;KL
zctC1Hn0H!i6QO7#{CD+p_dCESI(g43IQOojHQs^e>`H5FsRS`qt&TA?zyFVxj6-C>
zgp}N>)?=cE!p%o<%gb!JmUxZAU73+Euo1I;NmcwZUY{iuTIE<Ul<Qm2qyG(LH)XC&
z#HhrW{!eusOSfO1-V)sGm9gCqd+N;u&I$hY=zeL?A{(GO=m*elAAVWoo;!cJxm^d_
zSsS#_ls}i7#B}p<6m%>-6uUh&(1Z>OHAGc&4uTtzLZ%Z3-}2gnXpKjP-_EXlX5=*{
z_;0rO@A9xyjA)DFLZA6@KabRx>#rN1e;B-)Fyi>mF`e6A=mdKGd-xUnJl|R*vHXbY
z$vPYS=+9V{)i1NU9Bo<B<3;qQBp+GAHKZ&tkrARVUpXel5ow4~I+r1+&DKGb_wuAM
zWYq6>(U@&(-yn>Jq5esMd$GRtNlxXep_(Wzjdu=k{1`~Onk}708`KLxlqS<HCzB=r
z%dx%GU*{RRm#99IBR<fW--~K`LJ*zRS}orGTgOI`Wfif7_b-d(j-GysS#YwC|IvRN
z-(3HgxbrUj(xNE9HoD4Mau!-W>Z1RM>*?vRrl}Zg6>jqf!fX`~yRvnIjkt)4yLBx8
zS%7yFc(4Th_Wc14v^Rm9O|eZox7V9$NwODVisi!V%2s^}lI$0!wsb3_Cw<m6bg8=<
zI}>pWnX{?lFh3!83da3<Gb<5Mv4p%0*u?=3GNPl4IyPQ4sv(O59krW@cJ-35BB<Av
zE*!y)A^QCEVuny@wdCh2$Y>mLJNi5+vMFa@5^-^NB@+{rO}47Dj2D7mVB=wOo*cow
zc-(M-cCy*|+gVhF{nKB&Bn9LXzt<xHln_G~Y>xJX|9n!YKF0q|`Fa}8hTIeU7?u8?
zK9yG;)ApU|i1G8E+*_No@>W16eTx<WHg3{FmkCe`Ew`%TRYR5I)U!@8SaeHsYwX#&
z>ZP<atwvJw(+hkH0(JZX(JFdb9d)bIB~gn`L!*Q4XbxV;BkLzOv>y+7DPK;eu--`+
ze$-iUkVHwwqVDhB*%21&S~CQ{N6r=ZXBhk)r(T1V{NBjeK%RRsThFBWiKhl1byD?g
z*kW-NtA(pS2m%=)tk+RYWz2I3N8LI+Ot}7_VJ17+-M16=erXov#YR4*6R0ZWt)sqb
zdFFXJO`YM0aDBjHT9McfJbQ_>%B5>~`(A-r?me3&6AFKy+aPEKWLEKx?~XDuA0avl
z<zvWaT(^!ehIq@YfnV|#56|ltK*w!k{&>rQ>oz+#Onp<<!&<`Y<`pZiQHBY<a^9k@
z1Eci?{kBcI$fd*m2snxtz3`X+V%xU0t}=flWzKQnI6zhZ+zuW|sAE`s?4=6Q742PB
zA^me68A(m8R>YP2M&*DvJz+m%&x0|^&)G@#9V7!U49tBy3D|;&cSujg{aTkmqOa1u
zx%KcC<i`4;LnbWYw8~uJmP>OW7@ww;9i9nrgFqVF)#Udw)uM>YEzGhE?#v48ExV7M
zJ4n(kLbAIdr^&6DQ_D$%iH2xARdODOsghy&Rop!$%MEXhEO=<~5i<}lk=#fUxM_L-
zVW0s9_wt3_&f~YnBanK5W5kqeSo}J?&nVr3`!bB-5uqpGsPTe=RH&=#ZTma^)4a5Y
zJlZ%om^}M-hH09TLUr`^>wL!Bqsbo@Md`lg{}maoid3BqrdbYTSi|$6n=22mM-GzU
z+%aa~G!=bZ>~F<G5vYut2bTMn=W<vHd9P2Hcu4d_;_ZRk1Bf1qGO=n{tmp(@e%Vmx
zTl*JtoG}@4=-0e9im28BQ=sLQN_iuo(Hxy7mq?KVpS{cxG0^vJsAb1vP$i~q?vz1?
zrr{^E1^}a#BoRB6xNVb)9m{OP4-j+6hYOL=e;4jbHGx@-pUXiQKGzR}VcIKjJSXyK
zYcC$hx>2H)^75c6@M*K-S6N#QoP}wMDgNn;u=>mbc*9m5(2?4(fz-rz<6%W9n4+Ho
zPEb77*&(PuAdD8{`jt)AgiKdc&4{hNK$`^Lo2J2AMJm+WG@Nl;84Q9+W~!DKxaH(T
z*8Vd-&v$KLS$*LhACKgZMBiC8Z2Yl}YEs%hK`HIgRy?Wy!b&|=+h6B!l9SP)l|tNa
zX^;K49!Bn?)TnE|Ouenpb^pyP1)xybd0*9$qyMVHY`OCu8tMb1^(3-VeOS5K+}Bc1
zvSSWszS`>grrcn?6bd4g7?=jhJK=&ndWLj)5_w!Xe$~aNks^gLv$(2-b5lw0Ub$G2
z(HuK&J-KZ4sfzzxJIl`Hz}orxHY&9H+T;70_9>^QEIF4V+miLHNPa1AOqA?}P?|p5
zHE;D=n68G*dl^K$@6w=Rz~D>^fMo5^`Y89J&ep%<!Mj4Tk@fND>{|SUmf_m?{KoZP
zMonPgmOoj6L(E1G0ArDuXH9r-+pbJ$VB}T7?{E_~U72sbdwDhd>~SEr8EN`^{n0am
zMVc(W%myo>`@>YI?Y2jUn|Cu87E*DiQ>~mz*i@calDk{pB|tsaNjUmgU;W1&5n^eQ
zYII<%=Lz5)zj~P<_d4YKeNaCIcJ2=a?c1CyEAC^0v2P5{wI<fYWQ7<XhxcuDKE@}{
z2;pk<rrKY#t55C>x8262zf+k*4^{((|E+m`I@EAxR}o68i3HF}$B1Ne7>@CVd@SK|
zd-U5eOYm;_xvQkB5(<-x3?oBfU5q8O7qd{+j6kDya*ENo(zdQxYy-8`)w_wWZx$H~
z=dsuMR6gHXe&QDp{><xe@M)I6<1OBE1Z{whUl%eBKsO<~2cJ2JW+NUGVMY$+r=PW5
zGo*z;R?8t{X0e!8h9b~w>M`(TCJa%7p}$)Yt0?*Y=~7A0dU7suj_*hHvL_VEW(y`S
z-)?(7n9B1GfOoa9^XrsAQe@e+LpL~2orCc;?tfPi`m0i-s^bf)Yz3L<W$S;uSb4Od
z1Yfpdclb+s>sp5ur#LilTq5rq@3gXoTGCg0_G@no@#Lt8WUDHr0T~9}I(Rp&&+NuB
zKs?N)xx?Zw_oa4UP)v!^IT%ak`VV<rpYJjtlv9<r<Ch|je#1vF(f8k6FaX_0*`GZn
z{~~{MB;b(!aYE72VPSCW3X}`oznzJ6>*5kEM&z(LS7_>Jj)AMh`|T=RmK%^L{}Dd^
zPS#L*A>N3j{?o_=2}Fh<2LA0rFNK2Ur#G@#L0=5~-&6$4(#wF;i&Sz;lXVp?2UJmb
z*7k)#e&-IY*`m~zNGR-s*Y~5Pc>y<z+PF#>eSGoJq!KBuF(2j&4OJL*{EN8>+j7_@
zAbo_R;E5O#XwK38{RY!c_=o`O>4ti!w1}~v!{+)*f>%m%y_6vFEm%OpU1~ogows#^
zNP!v-SI_UfddHhPM1H3Bj~%djD;pzZ;L5a*v}hyH2wuFcWKKHeH6?0p4}@Rw&=`aa
z?cTWz9C3e|eF6i_5e#avt0sGev?9m{XVBv>1(kHmZ2<0CZez+@W$$0mJ~PG6R$yrT
zWRY`k(lmb9<5NJ_Cz{h*8_^ogi^)EXBY19%xzhlVJLG;(frS=2!yPbgE!H1tob-3J
z`2)>6<}oFrlzc8u`}JNDW=m%ct;6adw4&r;`nDKv<}wyVJuaYy5%FLG)hKL8wcYPY
zCU*%#z`K_x#uQ!-li>jOuto-RTFeQ~dxtbWZst&Gk{ZA(;;>#9u^EbAk!z02tD=wg
zf}=0EZ59g<q$X>OL4D5|eGHF0p_E`YPuZ|JOEaP{{wfjiK-7!y&}2M)WPa-Oo13GO
zyoJ(5QU@y(b^1+SN@vP+3Wk!Qgu^erSQL~DsrEF*GNS^s%IF8^GR(7A2nyoD<pd>`
zn`)uVUz(@CQt2;bRyEyqDo`}goT!-~p<9Lw)f-X&48VTmuaf^jYDDq!@0~um$PogZ
z#jxlT_Moa&e8vNi0~{UO5XK@68TP3R4>EKiBeU)2u&(F+zBwdnS+f{q^?l3JcV$AB
zMDXb^x0d^SDstRnFPFSidAUFMaqB_QiC`_$bX2F7c)Q}niDETWt-&qD&A-oXx%SqY
z3X$FuF3Udlq&J&#TN}8V_pmO4n&Q9Zq3t+tQc-gdER6*Tb`0^-V3CJBE<S;>jF51J
z(Q#PLCz>UkyygHOYN<3D+s12^-L$HK=1!Aay;}^+mUX=Y;hS2wfM%|!^Yf{*yRpE>
zNcpftV>J8j2&$mMtHax+j>cTFmH#Z;tn{zDqAW9=$4g4BI+pXNe5TjPj<20~dE0#F
zTdP-aqV55^2&QleLDP_DDo=UW-(pRxpOL?{v?cN0@&5U`8-c&`(tZgeYk$_ECh13n
zf0Wn#6)}2mUmsH1&?f<gqdK}#qEGu^D+yxy%cl$A6s6exsb*Na#xuKiL<k@};rghm
zwq(^;TdM>R?_|{=T>7w?%v%`LVzFpOjtCQJ!QK8$khclkcD9Qj3~G<X&}+DvPZg2E
z9rdT4%y4HC!3w%Pqhu;@X*1RfXbJHSZg2OMF{g_Of>2V(!+M`RXgMS6-eGXXAi=lr
zk>@*!pm;VhYtQpxAEU8Yz~DS}z3S2we%RQ()F?U|uEG`#((QfJAD~6exx!DQega5H
z1P(tGMp<P(QNXHW$dRIR<dT|#y<IV#JG@w_#;Mr_H(xW@E=}?`Dg<S(GhhC@L5+8v
z=rRKao(6wkPsM)JlDy+71e!^W^kxJ5cEx^!u$057Ho{p;48=uZmzTYVe*1ncSKp|6
zi0E95tOD+@2v0e6O#BwC89^#GF=fb1n%?K#b!%KRD{e>f3}wEmaypndwHpL`>-s2r
zu#;0UN}_^3)kO|TM&|<DGlKAY4BL&Q=#!#f_Bz_k8HR>vpw!R}o;a46is+9oms??M
zYMM_z^BfJ2V=0z;L_<9X;wsiE;@R>Q329RLI@^4O+%8dP^7dvT*t?o&+%9!Mdb;|D
z>|`@TMYB$SEB}n17#oyfHcNU=R)!G9wy*Bk`wKZ{Vo3N9pWMk9zBrsZ&jzq#HHI~~
zk9~&S`^r?Q%dPnDo*{C|w8PS|*x)d+A?;tPe?<s`9pSN-3e~8y;d$&)yG>?Y<DBLN
zL=t&aORd^&u5bA4zjz*f49M(Q6!$It_ix-#f?P8N^QPL{4O=$P)IP@`mCo%Bb>Cgx
z^OlZfaKq<3<r0@9Ag(Pb*;W`Xj6}G)T!^!124C^o?1HDe!Qp9XVYX2X{_~6^l_Wgg
z>N`0$q{(K&y?#RhPYMJycksUOcr@vzW22$m$)QClMU$?8WLTSLe-a}y(A3`;z&ki5
zq0a*=C58Dx0K}YxO7O-S$>YyH)<!xdBDgXml-9X{FJINUE?781f4k(%hQs<l)a+_b
zXJ@F*x>=HK(n5<Kh`-(lm1)_K{`uZ`v)I$wIbai#{Agbnf<Vf;(EP3HIsaM4?(!F3
zf97%V{C+Gge2-SQAKj>mnHffJNZ(%3A#Oxke)BcxG;>8J=hoP#|ITzqF%?=?IW^@(
zWT}8Q<7hfT6yAN}W8Sivp<-QfD5JnSw>K`R{1wn2Ax&rVnT#gSec$xMk9V}+8?vCi
zj;O9`d8zyC#BDi<Xv`rge+G3>JligT6}!O;iwhVo_cvbl%j@J$Ny$Yx<Bm+qVT=Do
zC8Ww|em26rCeA$kvi)>QZkk;y$rOETD4I=icXtTP6BBU{Xy?kIBvK_L3N1Xs7-|97
z5oVz7-NURoTnpE<zi<zl7OCli%3(swNs9ntO4Z6G;rFCHF$I;CFl_Qkx&LIi$2xMl
z2YTw0HdwTlD=$A(H3%$?_OT!@C%z#_bOSwKiyk5!_%7)uK?$_m#*60Y|A!R&Z^LTI
z?%rsbq%WV*tx7kNe>aU!jde)t_Rm%4-^gK{#IauFr{3275v*848!W@<H#8Sf)Sqbe
zMGYk1Z<<e(P05*THcro*O*wPeE1CwKh~|_NqDAFy4nAG=$*=!nHu+{Zn`z%iO*yL^
zY{GwOsKvC87+&*OqfdW%Jz8YM3aP2=)Au$)*CxWS83FWY5^*w0#lSCROS^9wSx};a
z>?cbVR&|UyTZ>FoEK_6cV3}U|jI`Z_RJ#<4Ai=jHPsedes)T60tsKAF>o_pS>Ua*q
zj-SKk1bsZUIIv8AH%^A~6-bWc{HcDS(`uFJHdVp)zmaC$x1ewPLy!Fps-yLu`7u`U
z(&F!ufxyM6{*UG(=kog@Q+U6=cz96Szk6{zy{;IPV<u+x*73h1JnsG&7InG8{syzx
z(U{qLPlQDpYoh}gnp_9(NmcvKJX?hh=COIf7sNt>+RMjP_8=8D*D|NEoB1x_<x1(t
zZ)7Ltc%T{Az_cTKr=eZtkuQpyG0P95f_ILXMzmi%jg(J(!KYh}>e>PeCJgNEUh>Gd
zwt3bYDD)-$`3hEhS6-4sK^J`9n3WjEiBQ5wm-`EH9LWfcFBAnPFTD^P%HiVkySqQp
zuFN8$qTFE;i72>xJF4R9ZdN3pC^ouwu1*sYd2NlDf~QhN&^RfbYe~N2U)B|>kI*!%
zjtP%Qq2aR@ZJ#FihMDDi1f)z_MjV~$CkF&QYzmieWLCdVlVzPO`0Usl$Y|;Z&qn!a
zJe7Vi%^<*dZt{g^2(~hw-G|AsmbP{UJBjb#!(4=Tv5^$@;p1Jg7^TR({9b8y_`nCP
zW*eJHZl>~)_pNo+0*lG%OZ}NBn7EM``sfP|D%&hd=6w=3<Ql&1Mo?-^`oU#!V;|E3
zDaMed?Ow`TOa8j!YdaBy`NPLuOlJifju?`NVnA?#hJ~M?xw^b<aV2Y>Qy!5juoUvC
z7k`?&wYK0C$^tAgEh-vU4Sd1WAoIRGC1<t{7OcO%?_-q3ibWuQLxc(k9tk3CXghOp
z;-5omqsA86#Z~+9?_qi37h9d(A6kh@=EvAJ?o%}`IT``evtoqtiVO#*ZVRjVIgFdv
zHN<}GNKGPi28~X7?Tk>`z<=aD8m?-ny#9N;)n#;fr_k<{!k$jg#{S2}%>)Buic@FM
zb^GEX5&|REVc{HSh8^xKKvov}>#B>A{mq0AJ`Dc}z^6yUlg(p9N1$$A-O9`-;%1SK
zR}LxfDy@>LUCA|6&{}8TK#>WN)xS-Vh(IXaSUnZIHNTb?UYVyCK{L43VIE*CpD+|{
zA_|I8knC?35iOV-&c+jeej=0#n!~}VWCp4tnPN3avFP$#W)9mkOz^@S2lxB)ayM{s
zXC1AGQ0u$LV6k=8Yi`uxa)v>YfXJAGggMLYz5Bt|wT>JCD2|meP@g75l4ZkE>!^hQ
z2{Wj!L_f?z;3pn<7&^Ha0R>*?AUiJwR!Ddxqmkp;2fBlb)wM|Lf0dV;ypsJZFqy=i
z83uLRm%zuvog9<j-H7J<gpyiGO-6+SUDI{lw_vaou>8~N1JItOlEEU`thI*%&E|{u
z#eX*IGjFQg3~_s#aaW4J;0cszN2WNIu3B<cdX3efwM=+?u8C&ZS}@x$s+D#$JH2Yk
zxf=|;<Blm&c1rPcj_T6JQXDKoXuxVN(flEmYcxyGnE`=dQe*9cbgOwi`1c<>a6l`q
zmzNj3^esA3<xNT^;sN`xOd=_ZX$b{ld07deIBSZRE5_8sjDxU)xT#P<kkxH&oH-br
zeb!H-w5L=n!GdzOlfwDAXmN54I<OdCQQ(GuvQcYm*08+;Mwnq(IL0YbDZkarMlFVm
z-i$Q<m+YFj$cvZgq`2I-k6vO@4f}bipcjUx5{hB6Ud<VSIyi@9CFM%M9$0<EFgEj<
z>S$XX7}BrHor3yvPy6of=@Wt66{BJeL*;wabadvXREqLGCx4V<`;4uZE#npkm7SA*
zVTn1kq?eARHVYG*oT6_H73C3QJ?$dm?g6(*t})1=iI_9U?7f_?H^Adoyx_~LQ*z1A
zIlPIc@y*zmT7?uy2HJQTPOJuz2cZ`iK1k{y%|(9m<@6d}o|<F?HWEA}Z)GtoeUGsY
z%~Zu+hC-W@I)Prp3Z0Hu7ntNS#p6xLmBBRAAwaurvujKK-}_T>2a%^TiSDQ{HgRV|
zI<mOtM+##Cn6Ixx0ez96`Go+3Hfsnc3x_d1w_!xhM9k78Cu_>utroov{eolGt4+*9
zr`PZ@W05pb{<ujujpG*PQI4+Ajyt!XBVmVV=h!#%7n18A)#A2bLEO4#>wI?BxlCq~
z1VmdSA%gVneimNjTiJ|BAkjEylaHgQYqN7WXnZvwgkN3W#LJx<#2h1;KwbBa)^Ku{
zDqDr|G*A3HQ{)HVg3tKD^*4G@0@sdGC_$n_Z%Qs7_*0qhN6-U5!IS?fr@o`k=hq*$
zq`@~6H?XGmAwm%MztwKW1V$iVW(9y(Nz^3hXz`c^z|h3sPam<^#>Vfw7*KWYufOFz
zYkGsvD=_P5Ql?9F7>g~~gk7BlLuaqjZk!Jm>gcLz3|yq4hAU_z-@Ixt_3c@~-NQy0
zCX89AA&#{w#~VjKaRp21Q0=nfyh0vFmf7v2!G{lJqd4O2jC)dKG*<DcDV9b}PBq&t
zKjc!faI>!(?eq+4uDMWYxU_3h%Mx&@P0mzK2|Z)xR}xw0Qw_!8^ED<C>R7lmO2@Ns
z(=^%HSoe>{`MvFG$Vl)Tj+<9pDH)tuFl{5Hys|jIJ^ioV_6k7XU787LU3F-iuEQ-&
z)RCcy$nX2ps@(#h9p(WpP-zFAj3z4t(MUnxwe;XwNl{VZak+k6wK7KOL_^`nSCwjT
z9Uk#Q;FYXWlaarB%Gdm3pM}B<{&59EqrqaSl-(oCS4Dg8_OVhSowUagb))|zpyQKL
zYqSZf5=Kc>BDHm?uPi;50-MxGhM?@`eVSB)RBMe$;8SREJH1p<pZSO%sw`yq{!%Wn
z`;M!?V$@%9iQxGTT|T41g7Rb}9Qmw{#jan_Y)pY`=|5mOwAJA!JypZyjN@kzv<v-I
zc9|7tnM%|&r0^kqs6sWl)ay2F`OEN)0Yc9i<fYMZ{9ge%`rlC-NNH8@RKK>}7Xy<8
zv9S8O{koE7qwMHx=t@yygXCAF$mI}2BFvxhSVhc@oSqc*5;(VFYdH9G*a;bj?NMd}
zf4!0X%Cb;^sHd-Y{cwlfUZ_R28tr!>s6m|~6%kOTjs#D!9=MwMHM=~YC#-UI0gO@8
z>T!Y|4}_2XJ(eFE>7o<3v||7%rh<~%s=0mr2WesA9i-`+cY6mo!y3O^*CJ5{Kjfz$
zXC?mbWS7QTGdYb|XtZJmQ*TbX;V?1>qk1Me{G_Z2Jux^w`S$64q1InJVaddE(=&R#
z%PspK)ULRI$We8qG=xhFt!8E~roctfr-zXT2ZX6$D>ZscN=@N7B<dJ8ur+1|-O*yB
z)kY895z<$%&L%gxD;nmX=!`M*k!t;X&(HinI?ZH9o!i2)Yn}Vtp-!as<m0bA%DVol
zZ2_h=Qg*#C872!1^##T6+fyx@PFQi)Uvfr-?vX`QFv9@xbMxfn#Y7eKY!C4VIImA|
zxY5$Zw%g#M8un;6uDI`~z4Qu_gh;e<rW4Wo?3oG`;0YQ{?q>hkRx_<;Jb2&m#javL
z@&T1LlQ?xk0IR-;+MprGo#pG)n`^8q%XeJw^L(81s^u$;RJ^ka;j50=B!QnVqVlJ)
z@16n~XEN>zhBcfbVq%wlkDel3hAkUT=`w_$+A&LEVHZ~t&_#G53eP@M6I{T7@*#+0
zl5^PVrycmPPNCU=GQ5Z(+AcUWUZJ{X0ZMx0v}8!~(Wq+4q%CLmm=5Gt5ZWRBaMM@v
zMoVF#-%exY`hZg0Z};Jzks}SLdY0#P+b6u}O^+4CZK-L*5ZE#We$U{Vb*6-alg(l6
zoRNw`PD@Wn*Nh*71uPvy8Odk(bKifzHU32ic6ra67jWPGr88gB`~E6Jw+2(NNLQaX
zc%B%i%$N#P`gd&KnCX<_S0>uZZLx{kIuuglN#j+>6cFowWs|nBe3$h9ObaJTqVt_q
zapK$F*2;E~F2H8T&fy_0aLcCyST_~Y?)Nf|G|HvS2JXSpnw|!XyXC#QyzL|T-B}^A
z@kQ%|#1Df#h!0H`>M_^IlhXM3j+P1hw$H$z?PZa4)`;YdS|tqS#J?GC!U*HmdE5|U
z9Ar^^b@Scw6KW$lBcc)>=M&3j5$C{4!UIX73|{xNRl1eCSq^eBGZRvPEwX9E1Mg&7
zHxqj9woG0lxv+Qne{7tw!N%`gq)HKx0-ol?e*|=D<i^rTxb%#HL`@CDL;>Fu-AXJf
z@UE|x<2-8G{)&>m7T}L(tJ>c5V7IR9Md-ITuyyUI+Rc8j`$rIN?xY#xmo%9w$438U
zf^T2Q<Kl4eSu1t&44X95^Tcs`ade5)!kDVtk5fxJ<uov6Wgd@M%j|Nz8gp7B7NC8#
zvz1`ckqECh>MD`@OY0wM6-=HgiXxe!vMh9i*cwR6h26`mYMAMzrX5d_{XC??l4D?8
z@u{Q_fCT2cqJFM#F_ngSc|v)N+ko_vnXT3pJW_L+k28|cO}{S!pTHAiW0q1Cgz%5^
zbefvV4iBPyz5dk8rKYg&)8PgW8ugX%uypUYpQ~LWNR60w*n4ms+u+fyF7I67r)K|S
zvK)gUN;UUU>I}U&8GF*LxuOR$3k=?@T^x|;P-zG%&~nit&KK=CPSrB_`tL$C4N(dP
zY09>8U}0;ToO(L#boV9c&`j;|-db9sC4f75H*3z%kxP(WIlj-6JM56O@4+1iAW@OV
zb!Vffhv^)2Es|n4UY0Ucko<gMiS~zjo?_3p<!IbY84}vk(du;0T+!YG2^xj{e}<T3
zXSVIF;S#SXIjZzT4S0oRdCmG|%g@Zn>gDqFnUpf%A%imD3HW-&`MSNXMN+#<76GME
zKVHQQRZFto1CRgqXye8Lo^5uS>%l+Is$-)4=?Q^#VDEA$0S;*Rxu4GE=xR64EO51x
zi#BS%qOs)GM-zZNCRhmrzD!T8dJ1fy|C!riC+lCWK3JWKfJ#S6ik9D6NQX6y4}W~C
zxC*}><QH#LsKA~ZO}Tew_Py78QT@aBTR{*{)ygivqM{T*_|qSI_(f_bG*U5Hx5vQ0
zyy95}B4ht1rH`zb#vm4G`~90E#K3cDUN64>&Ovr9pP+^^bF*o+H4{y;zv*e^sRj1-
zz_WPSFuQ5EnG@qb^WS2e_`pe?x=PPPdZ`Gdoqh=`vSiDNN}-Fb1Dv&LPn9&_0R*#;
z3*MslH;HMBj|esRrB{($CUop@oD17V6H#_Js;mwRsgH|BvRSN)tMdHLg*qQOS*R~6
zA=pe(G^k!u>O#RWCvC+-4f4+OfrGu^PIM_3dF7n3+rXwG3$$Q&PM}&RTFXHBD<$<G
z>T!+{cR)P!krfVZ^%p<rh1-hFEz(us^z<A~z{*An+2DiNY34LDSnlRoaXv|Ie94-<
zgWRNt*I+u%)R0SAZD?m#S2FVs{*u4d=Dy$gx-w}nf1=G!1afP#MEQW6wAjxjSVX@-
zo<N;g$aOH@CKwNkkvCQ5O~)$@g6;K&#viGO7x$MQSifAtrXny7#xlnCdhh+L$QaD7
z%+oR@*yvRQ7I&v(^A)g0m9a2K>&9#G8OOC|dJMwnV1E}-6x0heg|{XOmuj<_nW;%?
z^H;-8JkyuhJVQE-wP;?N0Ukf;v>dxx7sxq5c{7RbKL<p&;e`VBms<3Gfo$wIoBG_V
zVtf;<w`szXw57H*Mk?)wXUOEtV^P@{)K0>>1cw++v#9TmRT}v&8*uBk4<%RvatYiR
zZTwAV^QjV&{R-Q?D+pPU>phY5PAV~u$Q2lfxj$-L+1}+KE0*(|7WX)SJP)|f{fxXw
z5T!!1n+^YE(Z;SvpWg$Adba8;Vr1Ol!n$l8*bt!;WnRJ`p0hc@i_>-C$uuApa{a>q
zrls<3d!oj^Pn)}MOJX%9BW8@|#VBCJ@*@+QJI&tLuml=~x52i18I^8$%=cU+s4EVg
znqzBr^|RziP4#uOvqprSiv81K*s_chx5W^~G)EY9T_nT*3QD5?cA@Fp19Wv?i?Ed7
z3<D$R+-3XH%TylI))-wnG#NcfC`je_&oVTl(dPT54|ySn%D|dx$a~0)$3kq<g;OQJ
zw?L9vN0MBnHaWp+Tyw1RBF1R{SMAj+S&B#Il+%5kar*TTj6T8IHEXf`{-n1u>J${Y
zL>xdAub1--Hv@DU4roLH1C|UMmIbQ?hW$vp$Y#rdH63XhEF+so(YQ7{V=!p$UWqyE
z-cpt~lmL~ONa<+a=l5l)C&!f_qe3`cP=l8ijXHN2Rn)sKhDuibC>`5^jWCa@D0gLj
zBcv=gg`FZk+2Bwk@XSNGK8@)ZxiVf!!&&1*2*tNhxar7u8z)-=3t9^qWdP28X8M}Q
z+w_>i#Wcrvd3D!6bj>T&b|ti6ElsaBOisAj*y4Cz+oQknaUqpqnH}UCQnsI7N?m7l
zpsdM4tOUgRQdS~f!+?sdK&2+dzg{<(_iHgq-yrlmYG539-Qe(8X1WS(G0|8e(MC?o
z(>XHHEzXq6^jyj5M2!J6a)=}??I}*46wNrX9#0&ObrdmY6Kml8Az{E1qh!=)^wcR&
zGajk(X(P>&A<U8>Q!+iVjh5p6ak;lqZ)Ykne+HB*j_!g_a`b|y_NP0{k2=u6JvCZc
zmd<>F${X~JY2KTGQV*L3KR<?$82tD}JskMT2v6+4_(qF+AaW1ACo@{H(K^x$Gc5v}
zW>`z94Vvy=y6nRaNW-McWMpWaauzo(BNj0Fho~zVgKIhoDqaW*Cauu4G+n@Gm9Y~`
zLt$!!((WdJii(Vgvq}2?OstCT&&U7qLOc~vql2h=?&D$o*}y4g7#FgbxJjfro!HxK
z3e{S={WN<~$ogs9N&f_d;SErhTz=s0eDkHZE$}5_ZVm|^503pHRBO4;?N3(;BZ$%+
zZXdD8-;|WX%Y`sl@7P$5FnpX~24c&uf=iPSpl9|d_q6Z#fv)g(!}S4hTP5BDgLZ*9
z<g#eW-$7sRQN&ak`=@``r}8uDgx1yJ<;>w<a5`{7x&yxj|M}UiY3Az>zvrpfJ#FuM
z4q)t$fb!%p><j&X2f;D(&Gn{uR0F*v%rs;A;Vj9<ji)eAc>-Q3+xBpCm_ECnqHl1J
z9)nzWrCKHEUU;1Wml|qD?Sqh5tU6@C+z`yPOV)0O+Y^pZ5|p+og%A0WD=q?~BHX6G
zJm0}d$7q_GijcGkO<2%FYg8!#8QNLo00Zxx1NTicyzjQi4`5-d{-z{voQ&HS9P9x|
zrKLX&)+YyH2f|GR+Of&pR3W^ivn(>MHg-ljU2$g5{>i~QR1?)}o0@6Po(;RfgSrNO
zMjZ5{$qpSSOC3|kqDM{5VD}l}+L&%zL(`|YJ0*C=Cs6$2jVjgjyU>a3^6CK^u*jNe
z;-kM+M1m<=_9QUR#A*@D5R)MGDgX4_5@cOd^7wXP!H*Qs&=5dUs9nr6?zxR^u7e_V
zVzu(2emH+7FU$R6vWP*FMg=+g!OJwkQDUZpkL4_QEL)e02a!9!6nLEIodG<6$xKEb
zLkKqpZu`fE_nw8fxm+I4BeN9`{3@HI?N!S_p;WRegD_Rfj5FT2v&{r&OnR9b8nz@E
zw7%}0xHDXAwx*Ov2;r>0Ze(R47dD+hFr(&(_dLBWh<uVyfqpbAbma8mthHK!IXQcV
zh^PdCHH_X-OJ>xP^@hdo4w_J1KgoL7ul-}ZwK6bG^#Cs8)A{yFgKqJ-Q)C+!BUA8T
zy}RJgshQ#$g>^jo9I%Hs#kWLz_|=ya<%z(7bN`Nn%^CUeCKlSF30Z9aaQ`5DJK@{I
zrfC!4o4+W7NB6a*()a9rMWn+7F*5>sJ~vM|Z)VKZ;b`e<ZyXNrpF6SCdQ#EqgR;^A
zrWMzwQ<3FBonn)Xy?RMw1ERjZQE6qKli@*dxP%{B<cxRjp`VpsGZl#(adn=t+--kh
zLK`{{w#AGch@Q$(<fpbbN&E-XnmYNYP$KF@u_KHfby8d@G9uKaN&Td;yn2=&#Gr=9
z#V2u_>>(#S|DB)4_ErD$QnSBc(gAGD)T&P0liJY08b(aZ&71Gcuwo<ya~Q--bQbHm
z;LXKRNc<L(&Ft}5G%wpvtR}82kqMve-LZu%qh?n_v~u#Nw3+sSCmk9}jpjmuiERKg
z|M!wc2j$0~GR_`W5ymlE@Jn3(h<=>gllkv-N++Oos$E=p*PFg->Q7#>*Eb~`!ojf&
zaOckrtHt9OdK1&1LT3QY^+}zu#PNl|G#MK1Xuyu#WF%|{Nd0!LhL=*T!^Hpt3T#5m
zMkSI`K5BSt^HHVh7hOz*EZao<yQQBj`UaUDB=vu@wK>87h(9i+l@C3So9)h%wUv36
zy~~$WgqQIVR(9OO1)rAgwBW??j5zSuW~NT9Oqw16m1f!Zlk)*WQ{2neIt~PEDKsQk
zw92t@E+x_!dIe8(d`ekgEm_U&4(^i^e<ImYjxAh8QzFLr-|ITDMB$x4r;|>xJ7dD1
zqX>Mx?Knel%`3isX~K0V+=5)4K)u*ekGji)i(fsZqst?nJ((;5yXa#aTpixK2$eA<
zksTB~w;jPMm8<UeP#F;&m;Q4`JBUCW>|2v98_Yy0{?rQ_vMzDizK6v&5YZcYNirqj
zsUnqgtXU1LCYHxqwqVn}O@CYD_I;eUF#kOjOY(1i({jAWK7f&Vr9!S3UR7z8g4+92
z`Qp*Ec_#sVvs5Fzv{3}~Ilfm&plqDrc9+kSDKO%bwdhe{n{&^PB>?J1xqF`VnaXU?
z9G^>vrJIUeWt&&kx=veKZKI8+oV}tkY5D1BtB@tAH7UyVgK}ssyW0u*sX@!Qp*m}H
zf!3WfseW?RM?Hd{zo`s1pa1|!AH5mk_3nEa%>hvT_swCrk@ZF%QV1F>o9UPrBRg-B
z4VD8o5)w5O%wYsgh7#R9!QOD-;G7ny6Y56>1>@4Ir7C^?9blBf9fcBwTwVSd2dI>w
z6;p;VcBO3)vcclGuM($u?`uFPu8Lr=uH8f>civwN<v5(BTfa!PU67cTK&$&n8Ssd^
z>?#j_`|0=*kcs^I?=fJ9`;dp1^gqAmKvpf9S6Pg*|9S*!DIrRgX-OB-nHx#WN8pC5
z$=OmPt5Vs}Z$-~nvkKVGf<7c$5JU@PqU8zh;-l7~&ia^l7$qd4mqx<-<th*e7(=r|
z7Po|NWd&%8CpGb(Vr`bdJlFUH$cwFy9>#jCpFnoEy7m)Vy<Q9*m}}JnscFJ-eZxDp
zm(-()EdlN3^xsQ%y{|pCsJTS5(>Z{ySfQr>h1N67W~u>GA4gAo4sv<rX(EG&b(#KH
zN`0z|#WEhzI{FS4_-~_u!OUi)G73gcZ+;jqKUgn8GHv?B+pnKm${r?A4X7kRqb@Yv
zd0&aAk%P1PG3&QLW9rF2u6R3et4!4;#qiozXs#D)G1%rm|K=Vv%9kM4cYwa=?Me)z
z!U}O2!vv|2F9H@e-rF@R$(OHv^~6|S>`B5T!P25eO+(CJt{h9*^N<EH7^8CZMTBCs
z;U8!q)=d=7-)L7wuU6CN>(I;_xy`H@Xy%ll(~ZTm-Yc!<VD{NcnNe*z`(+YVFOAd7
zQa~aTg7f9@_S29v(dNEZ=`f9+1XwF`UK^&gGen`$1X=V#k#3POQO)U8Xz9aOoZ7g>
z*8OWKO8H)TJ+$;XoYU1EdZGY3=ozS>8{}mc1#AT@T<piu9_W8N+60}0Ld&k^f05r+
z=>+|wL`?&p%2clx!Qh(OqKDG$6YxQ``bhcJdOLAZvK%wDI$B>uveZZ<H6M|pXIaa2
zfz1(e!<4*k8Y6Cu8Ze^VoeeMMI>5SfJ;HKtoqqNU;>RWg`>Hlu1+7N{lndL7lBR`b
z009bo@bXAvS0k}2T6#I>D054o{7KQJJ)joyW>m0t5;Jc<CCz>tFB!(grA%v<k;psN
z;nK10_oik$5?`_&V38kDN44f=rHm4ZbivmHpbq`urMKDi_nmtg1*H^o3yw}^%8K@w
zwoFuC*v?_7k{I%G62wm6?oz~fqS42B=|&G|n`1dU<Z#3D5Uc7VjQQoCe7*6m&yA`k
zU|l$A{pRPlrnGEyb#Aau*C0ffxd<%zW|}C1y0&CtRT<l4t2m*)+AK>svTAiG^WijA
zx1NEwZ<;ClDim7h=fhB(Vu*qCZ7a*0JAnYBm;px>s7=jR1Q@rLtjD|2h2fwFyj{41
z&Yu=ZY5h$p&{hDg>URVahXX@{co;_8&u#U*JEzJdHBDoX)Q?<p=aHd~1(0ma<3u^T
zep?!2SvnatofZC_h3ZX>--92$&8!US;Yw!9*fl3I`G^2RbxR^hXSYF)2&d((*>%m^
zL&v@%<@a4(q~0IN9ob#0g8xSWnLF(f4|9nF^S;l;4%iwgG$&T_ZS_KfqotXIEcq(2
zxZ@p)cO}#-99!mh%5e}sWSu7c9=LdHMB*4&bxHFc_|hGm#Wmnsl+7^e!$_bi#j*BB
z(Q+~}xemuHnhKc>c{2ziVa{8b(0)A|356lXDpg0iMj#>N#iFTc*%`HSFTUsj^RSvs
zCEQDPW#q#Q6xL=j%HLCXVXmScBA7;mLLzM=bhWzKbLLzujT~=V*w+XW=K$0BMWg2P
zd?pZDRuhn@5S>gfSK@el@-vC<%P7gFer*wydCb0w*uInXYP>+yd!D0hG3%Prm>kac
z0-x_X0jYLFIj>@-=iww;TAIn(X-0^!``Ba*YAa2kI<330|4qIaRCsZZVO9tI%I9z0
ztHH?HnGPDG1zKoXD1b<XQQM${Tr1R&X9f|3)i(7@^wv(YD2f&ugzBcIM5_<2FfJ1(
zWQ>pYrk};Y=ceud1Kfg1y>s@zhIzc+o)+pBQ2mG?kam|Gw_NPN==S!@v`VIttl0Ih
zNL=v?BV_q4l3=}<c^=iph}{svQku(0@+v|9OBoTFR*8{K){P^gxXSG=#n-7pnyr6p
zq2nw>_#}^<wjwX~=2?{Oe_et0@2b)1%IXw&H6DB!aB=bu&r$R&YRpr9uqFQmkyBS<
z;iW6~R<h4xb1Z*OS*p-|hAeoast<?3t=fpp_qFRxJ+X;pK6=*~)TD{XdXn>|Y-f{I
zGX8S9_#hWz0_+4lS`vs;CjQ6Tm9USt!gh>T@SH)K(&TcD@G4gh-?A4HVKxuiWClRA
zr>xmR<YbNXbk4Sv{8e+S02QU7+(WuUgXQl2AaP^>LRowq`+os*2#ohMILp-lbQqA~
zL^4+#)|{vq*Y-MW=dg(=)@rLkSk_>{{U^dY%IVvPx>3--mY|L};)q|1@0LJBU>G9H
zl6(*RcgY{G4_ZoLSr$YJ-dJS~;PM3~yG9`?qcNirsxP{?jmo||rzMqsLfb2v6^3zi
zFY-BKDPu**X=at+EydaTqV4N;V7gpLvXaV5Ne<Ci7P?7EF*Ib~=}RW*f`mXGqR(g4
zJ#=Yv8sgefh!gH%7<}jsBbBmoDaCs7Xw{EZXis*FZi96;db+jBT7{eT(lHH&b5kxW
zhD3+1sE^F0kh6D?X>HBw((OeM9ns-L*67?wF^t^Xlfj>-p^@d%{$Cp;nnJ*iG~cI?
z>M!5#Tl-Vr)1k(D#1Y>X-z9+<P8vTxJ#oK(bm{Cj9XkK9uO|KYr5LJKp)9WgTDz~A
z_(XIzhp}ShzUImu0Jv*jBYm4!o)hu-rWxNU4^!ar@d0FESr&4If(mUw5MK~W!3Zn?
z>^z}yB9s-<3Uh{({V~NHPFewN{^5qDLSB=bPpLvs@BFGlkgc`VG(d3<U<d;dSeC_o
zL?rYIt9O{y1h6NRbvHP<=|+jeRV``wxBNI&0wrkl#)59f<yA*cu1xIYe(|>Gk`Xhq
ziL2}CK&JVg8`>%i)!q!?9_i3Uuj*U&^Qm23jn|NtMEo&g+h?N_^Uzy>k2vCp-yYws
z0#St@-v7XdpKsphpVO~A@-^u2YtfdDw-v$m4PS&zywgVzw`5HT2rh9_GPx+ZxQUU5
z%+N@2^(sJ1?LZQjWEWG^9HwcCv;`$%jALYCz~YE4GbOE9$sjI<m`n8*kve_u;5{;k
zyCe=7%ZajPN-B`DPo3{ChZ!W_J6mI6RLB*_FCid7klMxMj(wmZFis;SdjhiJrdDFL
z`>Q6~j?`K4E+}bm)UGqDCXY@QRcI+x36Yj~sLcz-gq&P?*vzWVmSH)gDc81Wff!w?
z-pilL!$_Bu5pg&VYPAI&(sFpJW-`pL!7E&9gTdMp#ad#KS`i&_#1Y>T-zS081@Llt
zV46;(wD#_u*I4>*O&`DhI@`Kd#T5mFfQDEXUv>*$sS78hk_nO+MVR+%cH@L`C<#-7
zM4#WN)6{JcO)ILT)4-rcx@D|nhmdl?iqvXMo!@N(gw$$ca7$OBCkaUjOQ}{QrNTY!
zJy{dyt<aZ4X9O2?BvhBb5Gdn-gy=$&<H!)FzL@9a4FH?uOG^?yB)MH$!_@Cps{k5T
z;yP5mxVuhQeO_CjOCgsGhz|hb?aNkdgX@Oilm1P<rBy^C{n&_9Mc=&)$;ONH#P!!0
zjV-tJdn}bxIpT;TzAe661!4x%H1Y88!0q;en))L03dZ`@U*fN&n=LW5oi(_pwuwV5
zxyqr_X9{q=CMyITC`)o;wK1-ZOz(@mIh<5^2g|;|hoC5#sDcEc1Z7>TtDyM=x(W;-
zdL@uD1($M{TGdefxea{r9|9#8N=mNSYX;S!uv8evhvOSiV@RP_cxEf*zF;x<u-~u}
z$H6(25C}22%5R<Y?u)*LIz&h(?8V%>Lxa#Pc^h=r2<lcn2n``JQcgarQWH(d$)#eb
z(3f~pCDb-+p+a%hJ3VN>uMNg(w_1=&`)n15_Fi3~IS%&c(|C~Wv(xtzM;vj)H^q0J
zM2DxRM?QS~#r-W>_0fNqzqR(TS6}4q_bc4uLvR_vUJsy2HGwJxHLzj~L4e$M`My{O
zFhx-pN6htA>YRjX0D~ba8Zwgc&Z!t(q_IwO8&c&<0{3NQT@zUXRxNHd#7%BIINzcu
zQVfX^sZM-t7RBXEN}ovYT}=hZmCq<pmIASmmP86O8lxwXX+mQJ^|oNw4EtJ2_W6zm
z`IuDo<NnC)85^YJ-%q)ab0KIX4$+fPt`(1X((qEI8GBV8rzR8M1o*L}O^?@C5ObN7
zt$M1jUn`eRw14%Z^X4p%406N~-yGjrOwo9FICGjta$d0nr8?C?`yKvmN#qR#Ul%#)
z`bwfgYk;yMK&d!-LDa<+UFffRN|=IW;s8+-xl}0;3PIdI-FLI?IQ4a<N>o{9s6@S2
z5Q+pLPR>EZA%M7UR?s@gm{YZ<ZI{P<Ku|&a1sxJn_gBY)PqJrEdc_=mqE{wP(rYmY
z1<Oul_yws6yMV2kHO;K^6&(gbO(-!8o(wDRV6wZ~ww-g&O#L4CJ(PMc+kMGqSXl`w
z3}Zk=s+W0jxNlHO@yU7_f^$q_K(6y4)+-?IvZ{%pCJ8?+ZSTO{3#)`n!!uQ0eNvz{
z12lUQfFRan#*R4Rh;NARl0Z^Qqzvct|H0`5|LeQ*{pE)$C;O|fDShM0TXQ-Q2)%W5
zdpvKBCkX%gPu17YwuaF@@4SOp{`tM}zx<DHA(S2Q`-%bHEr4}5Zu=j7E$P1f*5ATu
zD(4`(*Zwtd@N0*Cy+m){>$5kqU9j!FZxv{3z1_}7*gG`7ef2N?CHt%RH{SHy696<A
zpZ;IS>DQV&{(VtOVO`g%!*<x`9`WlTz`Cw~$62}mxcIv#5X1PF%m2*(^S@sS;mu<5
z4;f#-X}|Se{_5iO8g26*|JsWD%`fyH6yNLf`tNz~f3No_u-t`z{@?z`zsuntA0N0b
z@PGba|MwqaVL0MHFI0r3L@t-}fA(AXn<2mR!~9K8T9?&%gn#J7am0Vu(JCJ;e;0e(
zQVJ>MfAG!I5&sY&V5PX_#P`Eif7b`|yCje!jyU3oBmSY|cO{U6Iy~ZtBaS%Ycg2xF
zjyU3oBaZl8aU_r<jyU3oBYsyL3FL?)jyU3o-xWs!IpT;TjyU3X#gRacIO2#Sj`&^i
Z{{xa#<!BcS459!4002ovPDHLkV1hrTj86an

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_getting_started_001.png b/0.16/_images/sphx_glr_plot_transforms_getting_started_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..8c3cd951f552377560a06fd534c11e74ba3acedf
GIT binary patch
literal 274415
zcmY&<bx<5p_a)BY?hZkNySoK<cXxNU;I6?T5Zv803=m*|0D}b&E`tQO<=fx4Tf2Wu
zb<KO-@4l|;K7H>w=f<h2$e|$<BSS$!p?#5;)_{V7@qvPZ76Bl9j6~D0O??~$ykzve
zG+k}He9b+qp_I+N+?-s!oa`+qeXKn^?Ok2CSov7Fm?`bNyxcqm+1Q-_?+L7~9=2>C
zajBz^MUdR&^*y1WFwFn+gRT;*w1<L*g8CvYq2*U__PU(>X+7^S?lqg8D~C$+Y?>v7
z+1;TgprX>2n3CNr^2i@y0G@4vV24&B%ykF_`%^5D+0Za*+ySPOdS$JucHx@s)byWR
zqTlUtw0_NCPOkgQiNuxfwY_vYBV2!GW@fTk9^c=Ju{Q%}rTiH%_Dh8`W3JW6e@Q*2
z3``6{kpIu$|ISfV3Pbh&&k6q>38mbJUJ?24MgM#31(p$G=;Hq_^q(;bd4DLM|7UT3
zhIjE_uuXygUqD0+^01BnZxdpv*HC94d--pukGOB)xWPFY1)hBYxADhb@tC`n4GaG>
z{ioLtqR)&k9QsDe2%&Ekt#kk1@!nIXg2A0a(7n2iCRbPE6z>dK|BYJQ67+oD8T4;n
z?*3_Jl=%LCZo2f3b#O0EQ@%<i)Qi~r`g^iOnz_<wekZ|8e(0_4OXvGc=<H$WeME&%
z<{h?ToSA$nJ?;tA4$mf1n}$46@#{%c=)V*(30r~ChZo1UbJMqE3-;~72ex;l_sP&}
zdeY)Vb>t|peChXp%$K5+I;j9EOc>!ES^%~W|7JfV=ikPF=Wh<iz4m@s2-X!k427$W
zJ7~AS@w7ERO&OpHwNScg@HuMn>Is?l7Jm&t){rP-V{|jJ8R3u=)ZKBRp}<*K+Ae9o
zAu<kD?6_^`6_x5!35Z^gjdvT|I8>q^x4mj{l%mB#PHm}Ah!5?xy<&hXHD7g@efwN^
z+CMlSx_;hfEOuxV=n8oSdtTh-V(<;l7oYb>i8C%a(j!IjqL~JkD_P2yWQtR!GKhPj
zku*XIMu8o^;BAf2a!5T|=w3H=%1Xu~+ijUAx+6)QqaK@0-|Gd6NyuwX9lj~`B56>P
zD=yd}(YKB-+8e1h%WU8u?x^P58QUAm+mY&9!QKLWmGJo9y3a-Dmm0Cn$X325#`gSs
zatLYcKPU7<O=B{UexkddEeCpE4nO41q>-$1(|OtU@J~&{<JLRHo9cOhcX4{ijo6#-
zI~rmJG-xsC<`L38AUwLBcU5Kf%^J5`(lJ~In-NaoOQx1x{bUO&L^y)o1VJX;JG;le
zsusc6%C5c{tdYZ`0N_2v%lDr2;5giR%OAAj$lt4@T_6&BKjX>-5hl|mqGO0N#`lST
zPSRmY6chifT)`DRN(El&BDX^r#mWU?X;Al#3nZi>SK~=Cmh&tse_0^R>#5-20qYT4
z+U4bWw3gQKGSp9M$yzhzwUoOM6EB#NC|Wx9q20`wscD!OjuA%Mqe_Np?G*WEa}i@O
zUPsLPO=Mscd4z<3f;h$nr)~~sCvOs$9AxxeQZ+ny^392*HuQ8(c|iJo7<<*t2tVM+
z<2;L6T<j9doR}r83krmH2A^!-3nqs-1V1;Kv}|hOMp7AA6Ix2H|KyoQxir}Z)DB>p
zBQ}9|3zIzAlY^lScWu`VC#8e>d(YN|c2hSQH%$KSJ*W=8Q^`PS{JSe{la0#{Ja)cc
z>vfG=zfTKUHMZ>ybO}Xp3kgh-hYWXyK99$Sj&z4Wy1_!Ee8r#1ZUB84_sapXAWpAr
zE<#ksYYvUmu0}_g#A`p&4*Sc^j_1})&ND>!*?adw-~RULH3Umkut6OUq2;O*@1nP|
zes-PBvX=JP?OERY%=<-XVCC(($`C8h<;`KroPnpe_e?<h?pWs*NAG58HBypms-4rI
zf0TI1u4$Yv`i{9Uk-L{yly6aMD_=ozp{DNVA$H;<hIrB#M~PI&@m4#;0Zf=d#1dMl
zIk5p0<W2aCD9qTdw7c-a%k5x2t-AJe??O+ZWe)(>Hl2I8t^!Szv_cZ@vv|%5Eav?(
zkJo@RB;RC5bZ1FyGhCSH)Nzkz*QOW#%;f_vXduBV!B`;OD@`aNjW6Zt06d!q)(IyF
zto)X7>-$T!N^nzaqj8TYhPZj>X74O+k)<zZf+2nwTwd?pIexUumCAB9=cXbZ=4Y&s
zP%(Np=lZc+8<eX1;@(q|6r~ZF#(xq6JUq(nW;}-K&LDRm*$U$F>uq@S@XT!z)lzs_
zK_@DE;;Iu`Hq)T7jLKPlQ2}#3r4sreKS){&4k|V#(EFxD46^1KH6BQ^zUZ2?@fK(~
zd=B;eMjF}F1%VBD9(uZ0|534(C<jYdhBabKsxr=<F97yKkO8vK5}|TlD}4ZuB;0j8
z^VEcQ^u0O9vOxxc%yS)(FcaBxOdx>>R~}<wKCGth+dgsJ;q^I518ZZ|uXKmz?fJ6t
zL^@4`WgLVB#?j$a33YY`<hEZUJf7jdiIrfJMp44s(Gz3|HY#=#B@`0c@|f@c`lDA2
zcjr}be5)DZHBZU_X)+`+A(;OWN@+vajk(F|`Cg3%<l!fYN(1t8gEW?Tw6!(0@idU*
zC5HVtEnXftR4GJu|Ni}U<%f!V;t@5p-M9DPCYf!`urrTbj;&|7z+F_)EYDbVDhOEK
z{GKF!4CPjz0F%?`u2PZ!O<!4AP$X)>E+~Ei6J^8Sk|;}4o>|A|SPn-7?~3>hI7b-c
zO|E*Rn1Toyij*G$Jp;~yd{K9cy!g!S2P_8=KP{Lnia}z+dV_~twu-tSqR$f<P!YX?
zo4umXcd?%`A^w}jJE2?my`j%rhQ2r}j}LHAAm7)j8!Y|oG;4t|Gu&`v{P83NT)^d~
zxeK&U+_byl5No|RSrZqq%~3{M+kDIZ<g&YAy~E$M8>dNeOd!%b`8Htebh*+0@fuIr
zTAdnEScA!}*}o4XQZAql!w#%3Wt^jD$VI##ba@tK`f^QkJ;Fu!CCUq{S+N{yzJ=0!
zGb;H)e9jqlqoK`RXu<QQPc49qRopLp-1)U3bmPKc9a|TXL0AJJZp!nyeAQamDrH~c
z{<i@HzA4zpp?8-05*Hs5)H~nhW5I89Rtxw0>9#vq!#5!UOq@#g{B3RDc#p09mUU`R
zNXM~L#OJhQ3hM6t;!RYk4vKe;iPC~6D9TZEvA<t8;d0lt_bK{$LBSjJ)jb}?A_=nr
z-sg3oyW39zW{=)Atn7~O({}f0;v+jpR3-oklwGHL$eh#I<dU4qBe4i@`sPgz5jJ3K
zL3P-hU(?KuUTz%RUz$1;4te>uV;A)+!3Xr?37M)xJmmA&i+`G?M{X>7KF!B)lKg^h
zPHCcxr@U?l>sS7oNAjVrmf{!2eKOl~8maq48efcH^#}|#sHJh|0Qtcvn^DrI%<M+~
z<WE`%I;uxQ4}y;Fa7{dfsB`jXE9|7G4|1Vz1h^VEV~3&J`Hcg+PcIQEMZvMsZo%GT
zKmDDShWwQ1@J3L06KFLQ;)UhDjJa{kQeT2JYtosgFv$zrGqvMdSrZBHfcdu-h|{k+
z6vUS79p=q?O-8KlAJ03QvZkfdO5*QUK4mFw?U&oFx8&2ubY=ZLk)W3=P0{Do%7Oy&
zn_)x>HxG|#Gxnjq>RMum$mkBBkTL!{6y+2ra=v9@DFXr}D%e-ae?{gVnqk<C{ikB&
zFAsz<hA`1nH$5-W5*LRv51c1y;FLT?8wxdhzO+v2it6Snc}lCHhA|YW@2%XQ&Wa5W
zo8FhUf^Coi%srz2mfS=4`TaR@_>n}}v*})k;6{7>#b(9TCWN1l=W##~-ok?5^MkJm
zBqr0vx%c@N0)LiRe-*V+%iuwm!6B$giLFiQ!a}8j8kwRbQ8Ml5&|1zxJ0`A<I{q?Q
z{FYd@hb6-y4Iq$6jmKf(P_}O^+kgruNi+`Fl6!`bu=kSazC{1*858nl9#dK6wdk8<
zc3whi_73v(xM-Z3u~kTacb@&thSWzDG<EUNHSbbQKe~{LKeaK#`HFU~I#nf4q$kZN
zlc|C$kA*))w%u*m7;Jq%us&^czj3z<UUH|HLX_rN1nJCOpt%pbECGV&$jWWjVY!cE
z2F5?d++oz{9ttZc6nOFxF5VFwe@n-ryeprqidXiQcOX0rR@}cI@5scqEosoH%eHmm
zwFNZ8tVVenS81-*_1#Fgjg;cro#!@Y_XXXe&5FImX2_||KT>v#-;thb{U+b}nRGQz
zQ3<8NFtld`#jioftK!9mL>LH5*i+A|+7%!c_FPJ$Z5U*B+MfPbKRO%xGtTGFX=!*l
z3e-kHE)-FGr=*;qMmS?EXee8Qoq|fNlBglYNt;%4VM3UaAAd7aM5BS6(;f)=$+aZO
z7%2k&FHyr*5jf_`q$xmBa#A%uo~`=9(M5!~Pb3G6){&j)wvh@x?1Ux1ySPPRgfPJB
zxq^#q+?3l*9{{;f6Mj<hotZB!I0HJ&kOIJbT74QlV>24iZ_`}h>ECUosZF`PZT54V
zB6ON9<WVu?Dxvu1xUx9?P#<bI%{g0xlN5~*6%iEywx&kXF$qjm;V@^Q$tS^Z6gBT6
z4W}%DxhCi%SCVDJotJtHQ!YD^IBX_w{hcyuyoZT>e3QyLuU5jl?(2|1#6js`<pojT
zW+fg?M3YeBSn!b3`(X*o_fFQY7i&d<_Z07)kc8tWh)_q0mVYPr;JZO*|CSunA6f3R
zqW1w$^ZaJi(>gnjp8M}|BjT=}UF|R2O$~-OKFg!NS%J{KKM9k7QlgK4rQZIjhR!^$
z7Tlf)VWMH`Aw!)K*VWE-|NKULD3W0hnmmrW@{?C*h!2;wrALj9ui@q=a1ts9@EUw^
z{!$yE459hF*SQ~=?=-OiW&G}nLcYSMzuEY790t6T?S!;9n)HU*<hb`GbZ!kedWnPD
z^IurQu0^Wd?gNg6EtTpNwfK;+!fkxn7{5OqX7jofZ)~uB?Q<<X=?sOcTfZ}bJXk}B
zgNK;b4v!lkouK2896G%<y$Jq=Ew}U}P%pmJ58N&@jV4v^CSwd}5WI>w96L*t-6OOm
zF@CG!!42QH1L9$1D&|eaptjo^z02wWaw^!<Dlap_+0Bsg1On5Wm%U=mM7ks9Ug(N-
zjxN}y=hxq%HjEKWij=#+KdaFQ5^~f=p{*70-p8(;O{?|1N|1&7``f4u!(BR-AILt#
z^VzH$p8@PUS>i1*mweS@g;M?mW0F`-S8cL%+f*X_vo`G(!lW_HF@Lu&Q*CF9JC_AQ
zU!&)(rjL)2DpkeIv~~0>zo~qnUV&N`Ucw`cBiZQux3Ua5=7xsqUp4IsJfJ;myU<f%
zpkR|Id*27ZuNy4SaVu-d$;z_A&{9Rr(P^=FV$vl_EO6H?JK8KS8yOj!Cz`AMO{Aj{
zeK}$*em&zV5(@=gO(OZLfAT_St;0}oj4tEII3}cx?5W1nsdHQ(z*7jEmRTc-RZ6=r
zoGWQfoVSYtc@Wd@<snL^Cj679vR8`iEPem&iJT_=R3uXE8Y5?Rd&kP<i>T`h?@ea)
zhp57E@HlE+XzQKqc2(p~G^{Q7?74l|`LrJr97<7gKj8Tc^8&s<FV#75<q?YDW_1BW
z=%u`a#FBpYhZW)y)AGm^-8@V(hCbPb{&ecF;>~Y0V8aBaIn@iJSBpE2|23Pq+WCed
z7BDg!a+Rnm`j~n=45%iR#LN6)(QeRevW$wLS&<<XJ!XcoB5>G=`XEJr#>-`RGPO=@
zK{3}w<$d;O-}A@#(~nEz8qGg{_j`x+DbgyF>Rrq329i$cppH58=S7<a#qao><U2k+
z(@$R<5S5>AGt5PFYi+%icUeRCIzQb%TTr;I1b4vcKzNUQ!>!(f)#8xn5l0Gpo(^8)
z^24N3^{R)WsNe<;)tXTbo|D<AIZIv#*6#mEMkO_2`26Nc<j}I74B)YB?HPIGQCPch
z-@6U2^YbgRyQ7rwi%D-<Wq!O^QUuLzv5WE_G%>g^Gpx^k?-<yct)de`PWhBa4L}+R
zM?Q~fs^^+)P@|@~SsnE4(e+KCh^wc;qQNegv=&?$|3!XUBP`j?k0DsenlkDyvB#_z
z=Yzf@`R+4<;kT>-qjx$lD8;Wzi?>R7`1fjW+bPwaG&KbQwbGJYL{!FPwmLuWc%F)$
zLU@>DyoM=F(!(!k<ajn@IA`$nGyd}M%9?)6bjutf*zzPViG$x2OM_kTn?+)&J+Xsh
z^+|k94OB{cgaxOpZMs8i=^v+w8&bvyG6QXGJ=-eh8IShu>!)-+S5|(0VvnbzsLI)*
zYx==Q&5*a7VlEM3@k64UCC(Oi;#})(7C{2kaU9?RV}-rhYCRd4zg(lsPOGSZPD8c*
zjJ^6x8H<Td>LDy$1|!oHBOk&7x6^fA4evai#&_@J=CmC+%w((Pv@@Unq}p6aQVIjq
zGoUI0x#e{ixaq!rd4>}4O5yD(CW*8YaQfqC;CjRAwogMdMDNKP@=VTq-ZULD*TOUS
z6nHp**h_PVsh;W*`mbTjdBfz-w`YjKGXz#;-4H2OXC_Siq3~@`y#sphuKwx#{WU&;
zPV24G;!I}>?dCDP#b~e2mXqjm$}|+Xx(YwFAF+J649Ic%DjaDJq^A%&T$nAod3mHM
z5*GM->GyN^3Q&&+CK~$Aow*dBaD$gieslDXg9bspCPh>3k5gOnOz&js*L0^!mmKyx
z@4ifGLB%ulZ?B0PGwO^Tyj{Wh&jZh!bi;pum(EL0wvp#z9xGmgY%3+d*GIF*IQ;em
zH#<)ZNq*QthF79{kk7LYVXymQ^P9vO$=E0aSvg4TrJS_ItNZVi!qy3Xz9Rg4_35qk
zYDE!#Zu6zEJL?_9JoXoa%i<;9s)_A5&1~3!NhdQG3J<G4O!y_TBO(fdO)HuYOysKL
zQ^ddd{A-4kO4)V5?=o)ws3x!DYn8P$V=CCwCj{vG!FC-@#wb;0WLHyZFH{~{B~11x
z29BS&w12@wz*YF_qe#hG{p5OD`z8YUJTp*h^$1s(<i441Z5clg7bEkKx3=75Z$~^?
zGLf4sCl3qut7`_3q5aT?x+_hB=TG6n|G}Awt7qAiwEu9#0gP^^qe!;PLnuB>R8f7f
z6Sng6`su{GH*+hH<;pJ_ES+r|E2*|Xi9mE>2cxi;QgL~dnqm5i>g_+~(0;I&B5KZ0
zO8sNRaD4Y9{)1`>ow4bcwaVL_XIwB({yjY9lWO~-`o1t|ruJ9Z|0VX=3b|ru_bhSs
z7LOwIp}^=f>+Gz6kkF{(sO}c)%A%rL^dSYsP0IVE*v{kvJ!xTq+tUNqm>D1!C{Kur
zkZRB5*bMr99kEwNgCLW96S2l$hjM<*g4<(eT+g_ySTn=qM}Qx1)yOD~AWqgFaM~k#
z0<EvZPO9wp@)wraQq7?38nuuhbJ?*15!S20eLr65Z%NNp99oT^3}KEva|d{yFIr!1
zD{do>ba4G$G|TuVIe75at1rj`@A*uRg0lO_ZsmQz^?o^g%uWt7xkBo=u5k98dR|BQ
zPt`SkQ@>*OXQ;D6CLeHqaKwfvRYu{KR8!=D_v>X?%BQSYx^Lu}3HY*Jjc{*AxSi`M
z$GxYn=kFoU-A|1h!3?<q8*()6Y-aakIc$E9=LHUygsNOxI+nqs$9tN|DHNUpz34t9
z{88Mf2<h~ys*gu|wn3M}(1<bdSLG!Wh^6V0)Jf*-40E}aHTcs6IZnTL#z{*9e+R6{
zTV*~_O|qwTBB6bTiABnn5$}PvpIR5<7a(X;(hQB2g=BS<vb5EF3J49Jx+q5nm^7uc
za?4&fDGj=JhCNb3QwO2mQy<rgxFY70OB-Dpu-KvIhmXc1uni%F-$U~=R6`NyS3uPp
z5ieG=^p6VSl<V5M-$sv}#j_2*MCdjS$N;{n&<nqABc&7<Tp#_*WTsH(i{G)6OiM}a
z843*l4G%(&(xAjMSuAaVtIUVeqXyjKQwj59g?)kJpYsfL6*3#g_oxjG$cTgT5DX8;
zETT4{Inn>%o2ca1M86;eR96QBcy<aAv4CG`ZU@nEOh4u4yGV~wbeahy_Fp)rxH^l$
zFreH6ObqCvtWvR!{>k2<byTlp9!}?&YUV3)I>(li`jO4h1WnEPqx?q7ys^VrlcF1D
zV5k+?3cg1hd>CtVOHpNvwsa!a5?$&?G@+TeM`C#bp{Fv$w)=6`SO*vAN><OkZGP86
zOpT1x>T+r7MguW+m9!{1bn3X-*-ui?5y2Afu>;|S__6fJ7<rU6OZ&$Knnd}@_-*&T
z{qs5tI2rpHFVu%|XP^-fMYBd1-P7~tDId2;OiKwHy?_Kn^ggs`N9{1vkIsp>nx`s@
zltL#jL`a!qwsiN8)5PW=rR<kOMU0&XRCALLODm`pi<H%TzNS8mOere8sk2^h2MY(@
zMhGty6cw<O;}2_s4YexxMMX+k`4%h8JGyMz_0&*NM4ui6{IlKt{Dy8)L~Vo%*-4Wo
zELRG)AJ6OyWCXtAz_e8mqR;y$6JX<f1va_V_@J-QMZ;cKD+`6!&gNk+%O46y?!<mj
z`@bSIb%`IqSVY$<&kGHt_QvO5C|DF1IC=Xw{75{0UJbfBLX!W$QFf**Uua=T$iBb4
zK>~a%hMjE=6@uC4dP5WbybLtD1`S9spP66vJo@Eai_Z1lIQPh;%nzb1_TLyd8wdWC
z2@Ln;9S&p-V$*xRP&~&J*!rLM$2atu<$d@4R^a9|euuDPJo(sb$it=TbCOxCRUE-E
zdq*9yO)FmGVDC?ES~<nBovX}Inpr<mC%<@$s2jAHm>5zrSEwnQRv1flK^m9wPmg@(
znZxw;8G=vZim_hI&hk;Ge_$^9Qqpf(#vqTGE2+uhCGEY?tAf}JKMa~W>*;I^nZEN)
z>_#b?xlk-DHwF$yAFm3ve;rR4`sYVQ2}D5UO+>aeHY3;_Vf=xb7{MhkeCjcIb`p`0
z;F*0xw0zjuJ>Vx*<Vd%GrBCPYRhRabcf&`(y&n~MY$>}ZXVfoZ+0LrI8oTq5I=VNG
zDi>p3))CFT?1u=GL@*qJK?S=`PYUAQwsz&J45owH@^r&;Hgvjoz3(*qX-J>F0xtZ7
z>y#w8q}pX>u`j)Dj-_;mf6Y#wj;80OA>I*uyqZm&v9_Mjl;^dKZ3rp!l&4ZRu?Rl;
zT=27Y8_&MKw3@Fjo51OpVhOspJ7pnE@i^A1wCl!-sOU7+aRhu3Hf8xZ(dgJC2`gyp
z`fT(NRo1dxO-}*<-Nr)%*ZZuw-eA9z7?A;~)zplE0ucO&b3Wh-_oA}ULK6Nh_sl>r
zB*yoTT#WQ1OLgaM`<jYeXw8tNu5JDzKnhv_{ypVNWqGJ<wMW~!l^XjbQ9k~?!@>&A
zH|qHyGNP1BGlwLv#-iVZDwB=^RlkuB6`d?{2Py^TN3(Jyi2Jc*sHf^+iP#I!S&k@G
zAt#rZ7af%9>E{ul#k7_v5T&J5iD;$3Ns2njg$9OW({WTc&|Ky@oyog(`0ANlwnrm7
zN`;ugI6fYO3dCs<ma^7grxjIbg%16ai7Ees5LJ#!DFMU5z87zVu}%NF?+4=9)844u
zFujq{4k>C34c$WIQ9Xr37^NWBSo_a4^3mVnJ)%SFoBw=$!t2Lu4h4OgeK(rM$l$Zp
zB7I?je!~<PU(dkeFNCd^kj6fYl?}fRiiaj&a@cz~-+LGz(+?rxGzTTmz|FgTxW5ai
zzLRJp;LBgD?h)qA1133Wjo*4*TJC#zO21;*JK{f_U*H&gTbO-VY77BgT|rA)%cpS>
zlX-aT>a!8ow0v)1LXg2)62aC|eYzjkWM9>7&ks||w}?3zJ)jXRwXluoD4FyvVPI|^
z$Zrb2$3T(Bprm6AUhk;twSn<{w&*_3VBL)QeNWcwUwEN^%wqi8QSg}#MrxM~L^PzX
zO`=n)eW~#o`NUy#b=7g;?MZRq`2v2)W^`u<x;W?|)AapXDtdINr>KCm)Ehviwv4B&
z*=o|5Ya05+nHeZp18iQ?vU-f&%NP%OIkx@K36VO~?RbU^Vz7}*Eek{`b^-n{ma8p3
zjG4*u`JqR(Z+Sf!P>|%f{F27Q)V+>#%+l6%C8#;XwKE{kUG?<y>w@m+UCQnF)BrWo
zNStRpI}Giw@k<bRVezxhjbn{>95xX$#%A>PcGVDS0A6kfTvZQMdI2=tFe~KgDQ~&a
z^rqF6IhT}Njy}HRH&*qGrO+p6TVJEf*;m^&?Uc52+6b0Br^)oDW+4bBYd>Z4)p5Jn
zz%B-}D8Iivh5}Q3ka(6goD-ICY5xtCk0FXY4$h$S)@jfo6t?i2v5<>ZV$xo}>nNu8
z9t#{V1u6mnGMV{ZCw}Y}Rg**~fwr~Q5upre)h>z^;V4L6U;Bs|qhtf+8gP_O8KI#x
zuIZNDH!=TMf(8lq+$Q}Bn&UJ`GKjH9Rgg#Um(X0Ew(&f;5wO<hBuerB_YgH235#BV
zm>4aU*g9{*f=sLMJwUPnOx0kUZLK6r(|5m8sFyq`JuDz+DH&Uz{p0-ByQ*chBbds~
z514@@H$HAx?@lg_+m*y|q%?lXhThI)byt0LSN*tk+da$LU;D?lY~ZxJ7ZohjK-k&t
zJDa=F9h7rT!BzGA%(NAg7HHZ{VR`?Y!`&_!4_)D|rL3aw%ucsrpwEU`HPOOA42zbY
z%Q4OBuo=X6xIe`Q*!RDFz>1O6%u^mc3m9D*%d3E);)IK+sS)fEiJ&F0jOu2n{z9am
zg0FTv4ZRzM4=`F=4y;2fzsZQjBEYlTn3Cy1m+p=HiFQknV@!|u74rl^%9r9-mn+SR
z980j_NJAfl3&55(K#-^rP*4x0*xhq*c7|qa+gM%8voIN3v*zX=BHOLMu!vV?bbU5&
z`9U0f6e^+eMUDugRQ{aP_(it?$_skeABBzF!^cCckwcr`QvN5?s7SZ(?ZR^)@b*8d
zwe8E(;d|`-5nzvUR?e^VN|fCfD3A(sd^9a#@$PSgcoCIS*g9D6yY5|_WO=C6LgBUX
zz!I$ckiT^wI_PS-0$&(@o)y@b*0U&Ohe1V!FowHnZltVTujgvRZ<vl0=fxz>j-}@;
zhe#f55#Mn_bW$26y<1h2{-K;18Jl`s|78>;fuj%m+xID!dA*}v(P3)iPUVKbFhe*X
zkRB#o+U?gjMhSib;V*@wWBeYBE(U|au8!{z%|v%@U^u>&{;(bAj|RT3grBQ)x5YzG
zb2fl^&U%ecw_fbwsU|M%%JH*|$MDR~>$4qubKP^Y3XZZGoFGplYYuSal~J)NhpEd)
zVKF^_Op~nYiO@4-sHT#N^tB1yLF$_{Ormv(=ckfZ;V>n~Aw72U0c8XnPFf-Ysq)Jb
zwAz`(gaodF_BsO3%(;Rh;e*xIN?aN87BeW_8~vlhxj5XhRAHOHbN={n#?=$8iG2)e
z3h=V50n5=&css5*u<Zu6>qCq76seqWD88TtLOEw<Z_8$InHOkho`aq~B~{Y0KxP6p
zYJ4@?Xut5V1vR5qEZGHT;LT);Y;wpq4MFT^1Y>`kjjs*u_)^)Uk_~FB{SV4Ye@HNh
zJro(hX{^Ra^nl7J?#K>=#aE5QhQV42*Z4c!JGoewt@x31Olf~H=BeYuGYx%=W_!D2
zA)mD3wBZy6nDcayBMYW&wx`P7t<gibo84#P&+_|aME&kaN*R{bWQcaF)7}Ht3xx?{
zNphZj=tgNrTu8y<orX_D2%BSb?X!PsDqz;H2lRN&hZRXoH4Lk=9kE6tI=N@^*ajX%
zOha!GGoyP9G;<U~-h25n_w)A3q*PRs`GemQ<x*7m1^KtP-H2iGcdtbKzM&yV^~F$A
za1v2y;KF;i1GOAkS!cWZ`r?oMNd4(J!0wNCw63Q8+X;hj>3e1L!tW0_qR(eJ6HZV}
zuInqrF|FajcIOB(Y|M(BA;*p~KD-al>}9=`7n+fXCDal9%j5@;;5r8}bW8sCip&i}
zY3M|WOC~hQNMGbqo`Gt;^}HSfHEpi)D)rwq-}|XYsI5rZ(&|0u#vJ0`AFC|1B}>Xn
zYS0LB4vmbB&GQ72mE(|-E6_v=okE%KVOj2T=R(vKoLjkABj^~@L=cOAuF-}ov=5gU
zpRxqvP`))?^MyVKR_tN<HxSyuEYvYf!B;dtgrlC^R{gnC$0O5)N}r4=K|I-?IYj)<
z=V71-W4AGmZBN8`r?)pR`Q0-qhck*uzro&N%Eq`k&vgO;UcuTKF1&qq;+n3>nsZT=
z9x6U;Eu#8L8E2^Q5?Atjl=N7td;>@3Dca4c08ENb-sq?ldIb(T5U~!7LTrljU=pKC
zUW&_Pvo*&tW^Ptsj4DmGR%xF{=)BdTelPXZPKn8@yl`79?<kJ^k2K89l0|tDRqrr2
znLZJS6K@eFV@6ih#9&cRzswme+fpxbC0fB6e35e>{KhzC9F_gnr&#<P>S}`HvbS;p
zdb?%&_9lGk<IBhg|041!;z;-uvQYDpOyu7Z(}Gcvzw#geu8Qhvuebr~d%=+ecwixJ
za#bV?(v=Vly)W9xZt*SL%r<zY!T!L(xA)*=t-MMKQmzT)B9UMie$GPkG6mS;f_7S}
z3>meQuGGJ5sIqo6yRh5A>y&)?z;PvLbDA^`H|x}@GTS-;syfMAf-@vbwCU7W!4!>T
z&0fg$xK~`_UbH1*v!g>H?i6F2g%i#xe_HTx&OlB_(=2IpojBaKy~xx3D*B_c6gjkQ
z%$;Rpj89zhNbYo$9gSGn(?pN~(Nl<;3yPQt6Q|m$H5^by3?OUV*m2V(Ap8PJeo(+V
zpN%Xp!Q1uJQwXRnR2}XNrIYlq9=RO(Ec$GXdeaU3gMMSRIi@#fqpE(_!*YojDQQ2G
zgeuBCH+P)nE-=HRV}^*IUP~GVs{|lxZ%179#`(U<M3S*2cWLL{^545c_e_i_HetZN
zqV)J37}U>u;tFxz8v^Ew@e@o!1+~>R@cFlRvGG<F+wOil>81+uSO=fLG1G`Z-)^mP
zREe1go-%Rd+5LUm)>WR<+q}p!Q9t~C0XI$O_Gi@RGlOSl)1C^5e=2}c4N%}VKLiTz
z^jp0(9EA8BE@Cj6hCI@i?W9s`qN2ockwn=|aLu|AF2&C`kIK!|-261E8<;V`o<@*%
zMfK^(dNsJ{AgkXn{s&s-$SU+=s7_U|>>DRkY^`36#oI(uawbXyTIoYmTdA-3eVz1d
z!<{de+0LVDGv1)gC)U-Kz+2<?gH-99V$wG=H-lRYW~zT73v+uXkw&2s*tb^KV#4(*
z)5NYI|1Eg>`FKRXX-J^?RGzvs(pIXQ*Aw?>CnU5?sNPE!Y7)h>E-Ov|u$uh09Xo6m
zr6G7X%xk%3dsL9%X^psY{@EAa9|EHYd3x>rWh{4lvDT>?>2~w0v(^w&AAfs1yHRvL
z{;lyH^VjoQ>FeDGxtQ249$6{L6cBEYCdpgx-i{gGeghzN>?15&WQV~~?!IO|;7NFl
z?rgLTy4e><$c5UmtMLSmMwkXYQ7Jz{%b1i<mtavhp>eXN>cHjaY+K{!nfY44&yH^3
zpX4;nMK4y&_a@mXXRZiQj1cg{#2&dS@ev8(jq-VwnCYgk!Mi;t!}g=?93bQ_&W3w`
z$(PM3W?f`KXc^PYR%D&sy3`H5Z4?j+{0iDWGE3w^W)z+cy<*Ju`P)ssOjI_{t^nj|
zJx!;coxD!m@fxuT!j#NkG0V4gW)I;7S||giYqBa_n5V{~YaA|B%#y72S?BbD$5%+T
zx6v&hc;;v_V8CHobaS$l^Fg^-7KJ_#Mee|{4Xw>>X~bO7edm%oN?D|XPz-ifGd-Q@
zP28-Fv8*3Hq-zZtmQ{5EF4nXL9Cv@2=tKq&**aR%MDzfl3V0Tm(r=%TEx}<jHKZ(_
zeXWF=yI-ZOV0IG3*Cv_x5w^?Z+(-x0rtw9aHodjoeeR-{0`L;ovgqI~@d4kLr;=>9
zoKJ&iLtwXMmp=3N9sT)6tu7^f9@yB55)o?mI7(m&C&Wcdgu6W*J;{H#+NvoF8;<SY
z6s}1QiSQi=+g@@;m86Fr2Ht&9bb>1`Z)YB(`O~qF3K@OC{#_QNm(+%*#Wb!Mmr$d*
z1K;7ES&OD@?FKCM!Wqr6i=tW?grQ!$isJ@-rOZgLX|T`B11?g5*KKf|iH3R6YL~F1
zY>iO6jT)<Y@g$>m^lDOZbQI$7=T$UoQoAMWtrW@*NF@c^H2(UlsQ+d7q3;Uj$N8Pc
zqYUFi<@CjPc~KdMbKvIyTC$Xcuek`}AqQ`mAloMmdvcM0Uv}IQ0KA3N+rEzQOvw7P
z9qbpxp3!?giODTKPx`gs^XDAVxtq2=PT!a$-@e(Dkg@m2Y~T7)$mb747F?qbld2`H
zN03*iW*DVbiY^f^xq>STIz$758Sr*#1-Pw`bB!Ar;~fW{u7x%S^ur5xo0$nvOhSE!
zm-cx+(tD8tYU}GT<qmvvNr%z^Dq)zw&MYbRX#cJRxU@yhFS|a{9hA=e_7<_{2!oSt
z@Mg<m1CEZ%h>?(%`WEoqGFNnxXjnM@L<bQ@z_gZ@t#vN?NzzPwC)?m(w)bn(>)m|<
z)R;UfBG8v25rLhf%h(+|09>`PBGGRoZ8Q={GceVDrffU~OY%3jG61p*H(Kff8xm2%
ztf*6q*WOM^mJ!J@>VCN^^g;fjNEm<hl95D-FO}rZp0OET8jHVa&FKre!rKaZ^nE#5
z9^9Enq$s;F*IPkq9!gI)^7`cwL4VTwV_H<Ptyo#H=FvI7^X=JXZ$oTV&<#u$*lAE^
zIdHeMAa&`m$$5v$KfV@SW6)|c3zYz-xg?TIgkhK?N~*<F;35{5v>y(-Ltf9M!j;}|
zk^@|Q$-dHCb`$BfXf;{jMvJTi)QnE3z|%+O=TiEJ-H*REh#1u-E5FMa2>YQ1#_qis
z$1UB@G*f8QxpQw?B#rT+uV=1Q(9Pn0rH=A&C;Ba<#4BFi^`)mo15XeT0WwTA(~iz$
zCB7eLC!3!f@lWuMu6H8kF!i+cppt~~&^BX=yiwyp^OHzRkvQhPKik{G=>te)wSxf!
zeG2(C6%9-^4mfb&sxiWof*!_55yJ(wWp5I|ih!asF*qM@)LcT{%Fl9r=qGtfGyYzL
zC!(Zfy-c@V93gz#b{z5b7mS#XG8@_eF-6~%%o`@#n`~96*V<8cS^2V1Gag7(mr|-p
zEa!|0^iL&b;kvf?NnaccwBsC;%{Gwj-Qd<qa;9D)CgTi9!*lt(es-OrdBLS4#P2hB
z`rf~eAew2YIG%Muy49N9(HLm-ZZZT4l#%x-qOy)oFImF0R7_;V`U{BOUfx!&jsI{d
zS{KNxG@0-Qy)JzQflP09*SUB<CzxS7HEWFpb?m;JeRw0sXnf<Ry;kH&<#!^#8e>e$
zV=Ym&>9Mve)I0<h&&YJl-}B+8JN!UI@Cn;sWyj&T`y;BlqNB6+_<B*X_$B3geE|Iu
zeHfNR!x>`J8Tn=)#f^x|!)fTfVaPMvJ7NRt7WEI>1=yBJXv78R6xEc;R+vKjnhJ9c
z(-s4k)jBY@HY^}HB^eWTGJo0D2y~cOsY#b%HlBXbZqy*zg)OOKeu%ay5M73~`MJhs
zv}~Nd&;OP8Uokc_7{#lCU0FuIVKM=Rg-2On@8r*QXT|O*!4{`L$j_TY3iqHOWUrzS
zY?;^%tM#`Y)m3;(Xfvdqu;TMh!(wd3p2jxW>+~*(%{HO+$FU;Oi!tUzC8u~&g6U?M
zRQ8Gr=H<+MYb#1*PH%B_(w%}a@UmX-?$Ljw?3Y<azH~OxcY;3uTb@ls?N%N#!{x>d
ztq*gNenU35@{F1+VnJpMGNzZ=ld+kLZ^m(_oP#lyq&#jK^?T)Atf$LXjSnp48(@g_
z2O}!F!SL@6j^urhu@f_Kbp}pab_1Ewmm%JO(`NEalt<ZqPY!^C#l90EiuySzDU?s{
z-lsl+g3kJGc(6r&DoxJ>a{LGvsb2XnyShqVd=yeL((mz9oU>BjGJcs&DNm4=ep29X
zL*0QdXnyFCAdHFcHpNS|79c0ESgtk`&|Stm-NzPIqStg-S6kAy0uyKcu3v~+M9yW@
z$}VjKCa{Aze@uEucNYE142MTYOX9<)I6q78S=-?N&#fTPO5I`fjXrJDI(~R@DLytK
zdmPb6=n-@cSBTeTr&p3Wvcm8!@Y0$50FzVh-_tfP^nSz1>T`B)itgZe;*rQ;m&|7@
zG54+Z>ttW&!;9x%?`KO*)PShok&6`V3#2m`e#e3_zsL;HfR-<Cf;H^cR%f{QF_`#L
zU#$TudwsV3E)ORO%ktLxERISv?j69u1l;(UHBX-bvkoKmbp)1O4~EFAmVfAMBa%cq
zJKMVR2%ebhx3o2$9l(G_v`9pgcrOYz`M*(CDMr~Fw?9&y{j#L9d{I@O{)}g1F(9v?
z8}yNEHFQD)p&E`m<uN&(tXQFyu)^d~VER(-yt7-3KJe~w7bA3v!1N`2SlluXO}ks#
zZ)#~DUJema*#DZao?AbyObbZK_AMFskrn3wOWLW~kY&rTA}0y64pHSSfqO`!J7E>}
zbNcu#8uH;BcnM>)*81vvcpV*D`yZsJR5tBxe`OQAx3Rj5!by#^U&2D35&`!u9)T?i
z!cv6pR`6J-=^M{~EDOVx#c$Zv$dvT{lqeB3!t4~hjP|~=uJI)Jye>%1=Ym7%5k(x<
zaN2DNHLm^WE)kc;mlpL&%NV!V)WpLVl2s%nOiS&xp3(xt4GhuKP2k_(@_N?hGOfEh
zi8%$I1SmwGX~6qDY$h-06AzaQVuuv3d!eA7HmzIuLtEGCy!7%wyx&-uS&2@H@@|9r
zu6p(0VOcH(6%E;@T|sshX;g#(ntwm(Hi@6lxe_SL`E$CKtMKpUITT{I327tPZNFmp
zL`X)9kI#FHOuMea4vN+3ow~O`GHKmGYFRw_$}%EM?A<PCZ&_OB9|%GwVjf=aUD<e|
zI393U6pQ(?t}p0DQ3j=lg6~$UtTB-)OaQ2^CXUqV&_=RVkaf#~<^<>`jo5ujvt__4
z<mj_qSfqk#@&p{tE2JMY!}QJ(FhyKm)|t;_h(c*PD$l9>yj8SM`pt9@hkjn&hMZUi
z8fpKo+nOZ1++k~*!;J<8diG&Jf0fZ=bNXkatW>H-RKeW5nU0rdOA(R?769PM3qT?l
zq}lLtZ@l*IlEcoo2yVRBftv_&1D5`LRf?Y2K5=2|C)W+98(nuoC%3HI*C5`Dow+yv
z&Q<AwQ-kL-M&HKOL3?)lo_>d3yp1iFz#^e^dA^1@+oJ%3@049z!Kfds2$H>JsutpJ
zILhouPGoBy6K;kyQnHAr#aLo0CT28^gYAgY*E6OnqW<MDxe2pBkp7{S#MT5_f@Ln=
zrTCyDEQY9M`9^BX-Bd=&U^r)_pK%}$YV!4tR(Z7i_@7j(I)NRPkD_@hiDdHaj(!y3
zdGdFd{bo(y-$nZu!H_lHi_kk50-D!OoG(7h&~^n#eOX#?DPs&tJ$V%-9`1K6TR}I7
zn*;sSvpgv(7`ZCwNJTBtHBA1uyN;MmM%i{*?j9Z^S63W}h=@wS=Az#j1VoX;VZ;Bz
zGBPs2d*|Cn-zx0&nDM7=xbT;``gm}e%9tGwTHzI#Dl(9iEeTO&{SN(aHyl})@kX77
zx)?A=H2G+?{go{v>eYgcISEOr>-X2$-s6wuJ)a(=GK;9K)OZ5f%HqM&p~rS^Z0R+E
z$#S-90tc4>NJLEq#<exFU7>hkB^uM`5xe7WS@}D>YkOgx6(S9pl20lH&hQaOr0L0H
zX$vIr^QD#69;BO-P;g38nl5h2N&iyOfG-DPueL92?~m_!r>CwDgRiFVDM7C>gZq;U
zVh5wDZxO$>g03Fl`2r|NqH0Es@ZA#wQYFiNj6c(H4rt3)Taj^M9{O!zI%PefSgAsN
zAFO)Z{G9aqml?AuPc`^2sY)!bVp&S_Cbl6(#MI-^^~3gSA$61Ar-?uo+dxHQ9xC(#
z%7bUL8`oA2X8h0Sl~Rc_-}9Q1JKCsp%+zrSb=$+;tWnO4ye?v43NXT(5F=9)mWd<n
zjnss&!P(3R5A~<mr1Rzy(!<K>*6_p8`cdK+8tN{Rn|frHzo|Zf^4Kr>X%++}cbXa&
zwZC#zocAy<Q!zAO64F<lAltHN0Ir2$ztcN<2x|}*9v&%0qUL_~%BCs@m@5>NO>kZd
z(@)*7>p0;8sH+@L^5|MZL91?c)sfvXzDL3+XT~s5;CVulKI0pz+n^*z)P=GLvDX<>
zGg5WR4F1z$a5*yyd4xE&B60mQl$^z;|G0pd_39>0$7-S0=B+<P*Kq^U`X0Hs`Tbfj
zwMA`%(DuPW>z_~To4Sxsl@OTj;Gu@l*WmM3__G!NiZ>QxIrJB`sSU3!mbv@n8(*@!
zr)F)4!9_QAMG`%;4+v4U9&>=w`xj{2ehG_LVfFRqX<`soeG`e5wV|u;TeDF^t`f?%
z(w_1k${_5XAYB?PS`w<j#CX-N-h0$@l{s>>%sxxslM8e;MjWY*nKtLsVI07UPWK(O
zxGjzEbH60YuWz1I0w{U~DCPzR7!tyud^5H4NaG?N+M6+W)8NOzko(@jze~Pb%_bx|
zQ%&|{D)d1QJ6zFU#my!>C9>_y89rQ-q-11YecbJ(us`heb?ws!o(K8qQ&Wlvk<_2o
zbcR7}vs@~wD*OV%Ltxt0Q=&~4|9@3x{hLiDu7pVl{I(BwW1V(|HxE}FwM!BE9^v5h
zdAme!0U$Tk*MbSPUmp`l*3J)4r?b_>qkHtl?1W62ztT9eFrA;L0OZm_AWXiD^D}J<
zU}>)24{kcml%c!t#ZNr#nxQ*x?gdjmQs7TK0fdNDKl1PS#(&|<JL!|J`5)Aeu0&TL
zb$r?MM;J>Ldslt;Bo}#(-K)e2xuFjQ<`aB;BrycOk!R$h?M;4Tp5oA#7o7zHlSbvh
z)18ph9t$#k!)h(BZ2);H)u?_`Ymf0v-(bHGW&!(ohpwl?mRAU}D>b#y28f3di{n{J
zY+)3GaqzWsP)0E+9fpzAfY)k`i`Ze?QbiS*2qbE`i~T`A#fk7$)+W8DK^(!m#S=Y2
zCKiESglR2qa+9mvjXbUzhZ<;CP7<mxw&-nd)bVaHWOr=IPZ^#a@py{|2v4T6u3-w}
zg8nOmuY>Ixj_LVZ4o&>u&X@oW>6pc~N2$1AoTNn|TB5$uGdxT+9@D}R^Iyv!ow9d_
z;hy-8QZRvsFt97!XxiH9+8<_xow`z|6EF=gdmK!rbd7i!u`Shs#z5YT1cT_~Kdui?
zIvbHHhPw%{3aK+?OkS&V-xzAU^rg$1VEZNY^v~fRlHeecxCN6y!<DyZ=3ax;-poLl
z-dN+{%Mc9(=^U?!Kmn>5b%OAI!*WB6Pm|e#4g`-egXuFpJG&M@hwkB3%b*G21sIt%
zG&U1@lh0;C6er(&?0TB&%u}5FH@Ysn2#eosEM2lhPz@=X`cFI83q5ntk)1U8d7xB}
zt$V9Jf67HiYCf&`UM;0m)x)IEY!mc6L-5gt$kS-jZ_%x%u4@h(bYt_?en58nVV_O(
zVE!0drgS92kyalcCeB9R><|WzW%8y#cy@jas+y}@BQqautN$T;)hQaB_nm=XL}cGg
zaEwIG*s!K%=j0^*F(?|d$$;gfqmh)9l#>ws3UoExO3&~6>+w-Qf${n!>^5NZ_Lher
zr*R@a;&c@cQI0;S1kcqZbo6l@MfB~Tu5idT71*F0HBnNnR+gjzpl+}vH39DLbTvp(
zIP?YX-qGDQw4mJusqta$ZDI+>^kR}>mZS80zZ=Ib$YnR86z;gGKvA9Nawx;af3&5w
zuEFtS!fV3wm;lsRM(B<st_C!S>6QCzNz)k-itaz$?)<#@c;1>avq0PcI_RXTjtjdZ
z5flcNjLph=_@9YTSA7#7^0SZNQ0M7f3(DV+M45+~$Mw`}g|Lw_ySZN+47jS~%7#K3
z5VG3+^hrPIB5Nyul?|;U;kI^9v$ElV^qoPgm~3!OPea5iQ0KLCJTG<#9fhjcvx#eT
z>~xHW#|I4@P1@0ZucfWLk-2RtEF3?CPn_V4s)5v(I)61fz8FbW-abbC$x1&u75fx7
zs~XFD5#@O<lnG_%!!=e-=*T)f%xlnxLB0ak>w`Ttl3%w)1y$Ue3mZDZo{}FMG13C{
zPYBU}xS&c}>7D8~@F`WMJZwLXe~1@c5$~bD=!>Y+?sE6ekJ)_Io{Uc>g7+IvM^e-;
z%E;fIHE6eVT$;2y+wM);9a3UfeK!n4?~@^voF!*fmce3wv~5{$?~M(pk1Jt)FnMb(
zS!ugnMI;I(Q?a+`v$7mbe-@O3Hp)GoNf_mM9j8|FKOfwnDO=9}0#f!i(*(SaS0ZYJ
znPc%*ca{9B^bOcLsW<+Yip_^O<WhN2mUz#B*E_U*u-y|(d6#fAsk2dO0Lxnbe1)!Q
z_{%}r*p(gFY|`NOe8-ZoiZ`{yBa^0{MUtI>#w%;8hO9G{S35=IN(Wz|fA#N}QQ<2l
zt+sZ1xXg{PEk7v?+~k%3IG+dn^$b&6o1=KPj7A1}XlV(r_<iFr=w^&h%X@5oo=U%h
zovzB@!$@>_1XWM#&pLSnZLQyIl9MQFNw<f@W;*bbaD0A;MmB?SY6<3)RED){AYkfb
zU$JjF=zc@-?H}KJ>_<_m+F?l`tJYU3w-r}C8r`-L)0s5@JH(2)TJbpQch(w&R6=Ok
zL0DjHle38zo6Jr1!0dNwrzB-bTxSo1d3Cuw_{u`B8anFm_@B~NL@%=5F|Ve);X}uw
zYZ5Xrm%cVpx^r)ErluXCUgk!tKw>QB%%#dg4KE93580YwXm<7dlg01avg$RW6L1h>
z7%<qb(x=+=MD@c-VT4K6YX<3nwjwQ48)+_@b8Yc+jbORxbSraJ^m;o>lPd@*%59jH
zA1SxFW>`QH$dUTZxS7@>dV9d#32E&JO8Q{RpJE0rLPuJDh&H-zcn$Mt139jF(gd_!
zCxL=TxQ6Q~$Pv-3`~#PP?3=+396~}9l`OevTjVcX1$tSQx<_hY?jP%mK9tR2*3~(p
z^>9H;4y+Yg&FIBta1SrZ*LL(MwA0*FQgU*4ub^L60b%Z<a8;08PPlz#joF_(T1~+q
z7Ss)AO9x-&twG?spS7;CpDx~6{o(6Ry^p?tyQ15Fg|k1Oi;r2iT~0|QYBya5Q)-5S
zPZzgr?N6^Xp@k$ke-muWw6y~emRad6(1B+)hE0}x`U_^(kvdeuvZ2ibtBf#?p9l^;
z8yfMMa~z{bC5~c`vN~(YBfoh^Tn3wd{hGp-b5@MKpPG2IaJQai?2S2xfTWVY>XN$N
zao2CVMqeK#njSa^8SDi8ev<MR_bbOZ+j9`8{Nv=t?Z@1@TF~hVva3O^7Ote1t6NS0
zxR|p~e-zDaI;b)51{mx>4>vbLSmHyLp^tRi*=>H+Nmg4MK*i6`Z^obN=H(?`WF5A>
zim1vsg$C5aPk_H!#dF>2+c`T+tgUr4!SIkRi`a1+vg4+ll3IzZsP5|N0r+q;S#sxb
zk|uZcb?v&o-$6do^_T0M!QIR4)V&O;UoDA!wZrS!jEH?9x3nKjHuX-ia3w?Z{3^|8
z@6e*ava?3RA)))JBo6({M>__uj<3W9ctzbF#}_LTzP<KdM3!V3RQndOV*CD^3sCCT
zF-(ZU>J$%xx5a{>Rcu}^5e(YChCYh5)`0w;pJ+jgTBq<gwVnt2@gG7KChMA<L#*fW
zvl)a=dm^JIYe-*JGs~O{9=OM-fKuZy8l*#Gbh-EweoB^2vPipc%c66X`z*QoTLhDO
z^(M-vj^G31b|0qR$H3>)i4@)63UfHr|BIxvV2iTrqA=Yl-6fqvmq<!WOXtwt9n#$h
zNJ@8imvl=r(gMQJ9RlC;Uf(a6e9qZ>?X~V5K#dpAheC3)b!E^{AJMq3D$Val)*ZXK
zz6}o{gfPUFIR5sQw)ewbUEN%#4eOs2oH2*-+=LK+;zHmge1Pe;4^@}ox*IHL#&Egt
zhjSW$Ecu2GeQl!oWStH6^DDwr_z?z2v}||;Cq+ct@nEe%w7VRQZ}nLQ8QsUtPzfvq
z+g0iqvOR~cZtp;fOo1XV1lBMRfnv_ce;NX*<bMFpmMBqH9Y@kewg!)*(tzNPz5d5w
zlC}tlL(3g1?$N{9VSZ<K;!Y=CIXYD!xAS2a1r(_tTG!!KQ<eC`U)!V8X(M38W2@~`
zamROs)uOi{IM0s-Ec`H?)XL1wNrIT4t*pQ8<iI5pcbT8#>y(;M5f0|vki@*;?hz<p
zWzH&a%+vv9v0xuXtZ#DCys0V-XSvdZtJPl&(+D*V9VswQNJPDCe!AzH!{Z~>{+=Xo
zc;k4F*cV@d>k?|O%S6TRm@}N2HEkQsN55}5?C&$}FElDboQTXdh6A@;g;-iK<~HqI
z4&(%m#+r+5*k(WaPi3!g3JMc*Q4#^AvKo1u_sw5i88two`#yUpZgLn?2*~6Uot8ao
zVF-}<h_X1fV`!%(e>Us|071BH$`2yEPUzi#dLk7pCdS7=l3IqQbOI+cM*hI?FrxA`
zp&i@8QCeMJ;DEE>?n=-<bU+zakpBWD*e~6Ect``vst1#}!kz3ou|uMv&>zCTLVuJ>
zZRa|jLi*eK`}+D%IC#5ydLgo2hN}K{P2X#YM{KBaP;duOA@-Y?NOOqik|p>pptOBb
zB|NPjtS~9>Scfmw85uc!0RyE%Wfg~lO=dzL_ols!L0fDcf*<WZkn8U-HnJ&*Gf4Go
zTfmucDTg)@mf#m%5`SZ^#-4v|pk&rGm;igiC!2kaU~J`>VquoHDxHU)$(xfkOAPq!
zK;}sGaCSG1&0aD4#MY-CGS9^JO&z6C2}Ftg-I%q~cgU4aO^=)S&%99r%JSl}7EPtE
z0ZK}hrVguxeGu+7MVn`u7;FhO^{Tb(-+Bjet8Q^`k|zWztzpydDg#w>U6o9Pcu5O5
zCxU%5Jh*4HwRP<0%!DZui1r>viwzfXx3e6@30VtPTyC0#)UL*x+<dKJ*YzY>Y#X7z
zKs8~k30PZuu{4IiXrTD#BXvD*%YZl8?zfSiL0-ZNUaFucp4SK9^73|-15@ax;~DHc
zmCSk$^T#@E2^hRetuDS5e0%2hYRkRu<$LHIb<%%3A4++N)I1ws9Zu;+u#h9IO+#yl
ztCnmJG!RIh(}H0HMGFifW_MGhn?XN|fm;mRJUz#K39gWAE0S<Gd^oL}hMt>k-mS&r
zN5)XsJfZyVhKNu{8O$vwe+VsbB3s>3dV2hLNrOMWeC~*m@FP!t>r}ljc>ltb=AJTo
z=zCTvvrQ%6lI8Tjrx3vC)}%8{tQ4U%qS4LAR58XKpAN3pb#wRbUnm<jFk;J8;#Hx{
zFC_T4OO$_k4Lr*Gv30s^%V`$My36LeBLf3KwZQ#dX+2ty9E$`)lG~dk7bb<m2u#W`
z!`a^1D~UFMtU`5tGp=zgEM5Dy9a=VJxxz+>4x~dEM-zz?u?iK?!{!8uewpt3%(x4i
z3R7I4ziw<f2l~u7y`Cf{$ppW9SM%P<I7n8Dv~r?QYrx;o?dC@6)1I2iL3M1k&N>qv
zy7MIY#<3!e<IT0QM(|nJhW33nSFLa84+3(3jH6&TlmZF&4i&Gu+!4i6bWINH9E%2>
zUz+nag<A7uVN9UgxZ9_k5hL(4npH+le!$RSkbvnI4k3{sOt|I=RrS#{ci6BAYr>^h
z0yT%j4?b#uWm-)yJPwv))fVxPd1FZmZdvwxW3@G8dxum7*Zn{{w_?CkK0zd)wK*Sa
z=4L_pbgGC-J*A%5e;3ZJI9&WMbuEdNi+H+K+$dg-o!LkPU#J8j+>S2_3?0G~f#hz^
zpgb9IhaiUro5*1(kCNHA6uA;r5c&iD6<ssCyom!Jks85S)<brGrcF&qP`exGXS4%k
z9X(OorN*`;rGR6ZB+j8+SsBdkp`nq@;GF_bE6jtyXFpX>XMUr9?Lb!A+sJ`d`n5tC
zs%D_2b<^0A%=g(1Q`%rC)vKn3#-z~$jfP-oC(3G8Z2Qk21UrId-<W}~jSV8KK5(wV
zd=64MH&I`C1MdJD3Yp|K<=s&k({pVQS$W4oM~&^1EQiCw*@F6pfxwpj?557fQ+LyH
zFO;<Hi$`L=oKuaHKx}@}#y{rC`-_;<g_S~hAIX~&=&qFt8JA!MbGNfThRoKd3U(vv
zIo|lge<vY8C7>(DFe~g~s%rXkA(||jXq7SwU5I>-B)J9s<0?=6=DbfAUc2$VVT=80
z{pseGQ_v>%-*q?eg0NKx+VFDPe$2Gl2;fxS><%2TY@sDP02l%aT%-i%Ea*a%lUZ#}
zTin@(K(fC2OYpv^DCza2vHys(&c)8P%H#Gpol%ePfNzgk)7k}O>opy)^%SS!%M7>+
zBbxA{*>8GtGtBbprZ?!-wD)#mw0c4L2K~gmpTdsZNwplQ{|JWU&?O|Q>>jXvpRV=3
zJCWkx;E)W|CE_1gvz6Bn5cRtQLZwl+AyB9gRnQI1I~5HrQv;z%Yn^(dFaDR2Ti1V!
z`gSk($}l1|!jg3G_{R_$i%$l3PgqYS8xFe+{W=YkGDglly~8%!kAu6L|CZFH{zyGr
zu)6uxQi5_?!!6+$I#E|S(LeK0+{i~`nPI4=?Z$@QZMbt1Q&n|dUSp<)JWAR<9R2x2
zj+_$>Jja4wYe=LQN9B1Wz6>suN+(%3qF7DW+;=x8ZoqQ4iP1749h^ZswZLnv4mtHZ
z)$nRL{+;V{3ogco66YCF72mQ1bBQV-FV|KN>MigRjPSyWMp;vl!OGVQv=#(~qc~9H
z7l6IMwYKvW!RmJr_;V_|p#*W)E{3Uc>f|3JMQ4bmcz$+cK#j+K5{Mc?baXcNEWXy)
za(er5ZJIUcIDuxu{ta%*5ZJ+Zs8=OX@l&J*S15*f^r_T<1aZX#JVI^Y>J=NH!k<7j
z!KVMhK?J+Y^aM3YS@R}SmzSuFWr?c7u{Oj0_<6#(FA-G!CtU&6|9WTdwTE7tVVsN&
zpYpemU0cQ8%qv%&j{;22#FPx%D7SdsI$UPhxBBV|yH*I*%KOyLy9_TG*}E!t@4r2R
zJSD>sjThw#)YNynW4<$KYL10bQ}j^O=!>q?e}pYSL6B*#xJdHMRI;u7<Qyorx9k78
zqarho-3YOD=5fYxu+3yfcCPhBj92NZSef5bfbwkl&{1Ry@yL`Y$szO?Z?hN2mLC<Z
zn>|n_#p+G4?ewD)+2du$R2sMhc#LFga|#AMQ=}wuvxPIQR%d8tV|H*PiEDk^22$*<
zH9oH=2;Ek^zy3S88}+6xbw*M_vw0t6S%Pgy)2p<i3<)rQ$`*l&E~6?dy{`{2SAqn<
zT*MbQM;hJtX%<{eRPmqUJ{6gG72zuz!m9oeQZ_1TDXo=?p+09*$rs!EYc1JPv1-o`
zoaM*DVPa8(c&SHGZ0Jgc8Uo~0`9HKG%p*WpHM}!*F1<U(uaj@sv-v?|L3jBSbo<nQ
zD%^7wG{?;IJJ_QgYCPmoq9--}9yHH~C1c3br;W(8%ido(eDnl|oc$9NO<DgX5;=Mo
zmK{~^t1BC(&XkL|O>oUq_B+dTCt5PWiwWabhTs-;_b_<XzP3ROyrC0z9+nLgHt5Cz
z%S4dLOC%W(hNzchfd~K;N(e@>Of>5>mHK;gzf3LQdH=f14R8B+h6hv6@f8{3mv5k|
zr=9j*5ZA1n#=z_Mn_xdjwO$~!&P!}06ME2@k}^Jk&~US;wZ^TgTN{eA;}0buOE2V*
zjwKrZ#KoP;U%nWQ!ln=Bu{r9&3$mpRAv%{*^(+93d(WfQI{Zre>AGFSMR5wPc6i>3
zrEEIxA5UT)e*6*AG9QK{;hdDj9d@GjuE3nEUZ?!RKmn1lLaZMBD@N3PtpI_?Y-{ao
zphrk!>@3y{lF7<XU%OlvSKo;yD=JslCt@%KM+*8l*vGHs8$!merZ18}Z}#Ci*#ADb
zLto4P@?&ykW02y*z`mH$j><GkFEsDo;@4J*Z~&1`kCR_}jb<Qf4^HAn)x<gT2c!$X
zv@qSXh0uHON%#9%JFHR?*+{x97Av7T=fPySGnV%XEXC*%J*ZW|vimeUqBFE%pHlEF
z<ik)B;AmD#1|QIh%^NPd1DZ@*_{rDbeMcPf9xH@(N%=KMm?Q@mjB<=lb#`cWWmRYn
zaVnn6wM93%C2J8$Y`uQo0rg99nhsDdw!6LG_adPATX`W@B5A-{7orERx=CG$roo{*
z%jx~H%W9c%XhLQSwmyz9%WTS7!a$~Lp0F+_!9c4|H(;iu%2(H_O2XMBE3U6jUqL}N
z4!lKveMo%Ete#8LlHXzcGsFzla{eOG?+i4U<UbnWpjPDMlxA5tgYSJtY-@a8p7<vB
zx#M7InB7NA1kDA>cM(A%GBe(yfBrsX?l_MPKN2W|g+6?spXC8E4)jB&5jZK8F`sct
zEqja_r`)g@WO-Y*Paosf%m{Vh_P>)eEoY;@p7uQkUCRdD*kL8)A^tAKUsvo#H~Koz
z5J|}E9?HXhkt@tg5)?~JR|kIg4(|wue@)4Qgfryci&#ak(otA;O$KJwF{{;O&%nGc
z3}(y12{wKa@v~er;T%&~3h#2GQC&s0aGZ8o5{v8;>9G)+>;hBT9Is`xEt+`|ZYz>0
zE(^AYD$9XA&cK|}S3IN=IPYZJuQQ|Ok1%0UtP&gq-_vxq=7^(<qJndaC=3-c_KR|@
z%{q6xuk+GB5J@83=H3XBXgV2<L|nPNKEX`Xz-mT_@5G}zlt`O<k&m6+wLmFLuo*vy
z31Vp$wzz3>!<?!m)8&|$0#?YJ=}PGec^Jk7n0SLR%K{HgMyq!Pyv#-6AU+P?5QfMk
zu4T!77n*nVu~2>ccRPr11Qcr0N+7FxT()o=ja~|O$>1v%YVULKSCiO=)={?7!^AiF
zU`;dv_+(ak+)l6pMQ)5}k9v~3Q>cZ<LS}udh{}S_#B`?xG=v!z$Bm}U+lxcko7k=*
zD`G@AC*!aGl)3AmRsO#eK(<=R$r0O|e&mxpkShTWvvU9TrpP;T(0XIf{N`MGvF8J5
zU`_?$tk*a?MUo7)a8@wp^!n`O1)FtC`<w6i^T}e%QlbC(WN($*y?@)%jq96GL%ZkQ
zRu3>7jmN?OULZ~7<4e-cHm=pyH-E!?wJ%6ChH=Uyf+A+MV&@KIM5h<d6I@hHncsL@
z`uz~gab%3cRLM~iy^y%R7PFAe%+_X3r{@4ZR|3<8uG9djRf)fl<<q58DX2k%$oRdy
z<qP=KY}*Mdy_QrUF_yPt*9#awMl#WSt?|m6woRy5HZab;-0<3}*8#i&oaL36Gs0Kh
z)RXYkna0N8I500Ui1LG#Af?(jjhp~r&S}m&flG^k8wg-9#Gd{N0p8Tvk8e4^CkRVT
z#p5TN)L~*s4aFg#igYVghP@eRvf8GIeau<1C7_+KB57I!ocrj@P`=dhFjJ5?%diO-
zN&>yY)t_-Z_w?j_61nm0rnbxp7iTURtCZ4;i%p<#!jYs29~p}nIzkB$ltB59MFGmL
z%T$szn4Se-h~yYdNQl|fA$`v;TUU_2e=GmA-%f;<J5inj%%{{*@cS5%)?eib>c^Ff
z0-0sT5UAhf=G#}m5!#G~K@+1OBcD*<|CFF?Mmlh?qS6Y!m#autKwip9ll!*QjU91l
zL1B(g!!=W20H|P#G7h#$!7nUOf@O0ZeY$T>x>xvV#8IHJLB+}odp!$4<WDAoFJAOs
zX+9P>RDX9L!Ny7pc^52eY{=f4OZQ}JM-xJ(X<}%GrBx7u7jpeRad$2nmp6=R1ZEha
z?z6NxPms>h&!KGudR)1k<n_qn>tz|;&jscq=GAI{9zNA)8uF0GB{9X>r~=`%2%*BX
zLL1w9tc-npMBZ%SO8%mL(}Bg>7X38POU8An#TyQ7G28Wa?(q5?2OLRvj)yB72$t(t
zjfO&n^&7E~E@95*uLqAneR|4FnR&Pw(OFy;B$igf9fK%Cq6$Zi6+tL<cqTU%jk2oZ
zvk8&zceK4hJ;s)(1VR4wTU~pN1!5_%*@fAj&I;ik;;{3eQ{eS~F(UY4QdNP0qRr0w
zKr6jq0STOb<dHml;*W7|xO?RaH2Qw0*)a4qFB|1iS(^{NeXP%p{tS5nYbvQZ&eD)i
zNvj1n*$#u$@x3*mjv?Q39>gQ6tq<o1UWLf`gF4tv9}s%5{4Bjn<%C~0dC;j2TP{$j
zda)D#t}omrzn{5R`nb-xc~JB#t4Fc#=6vgRzSrvH$(%|QMKMJ>#zAy0vS#D_!}rbB
zE@*RbZy9LKBARB+E8;d`llb^7nOBH?Ug}q&JO8d^C63EB<=mN3|5^SkXCMG2S_9fJ
zFd@&tKY&M(#fG=2oQ83vB{$YiB8U{417xs;ghcmGp^8Sm?J-r>tL@cwmpWIXf^v&3
z80sZ_21Z6aT}J9FvXFQ`#iP*nMbvY}@p+Qg^L+7_y>ax`pt{$`=ZR&Q(Y+qb{vmoT
z7$yY%(q9dw1K(qS%=7`Y?oHgVZq5Dx^jjwM-A-Gq%(SnD`bxy5^F*p_A_*oMQYd*1
zrJ5FNgv`j!PS}#|IB?rvJRasZ<<tr0kT>epJ`8OI0R71R#WsQ+xkob*OI6Ci51?&}
z3LOulMXyLJU9f>F;r?O-EVDgEJo&TV4QDgsM*ip~xX@huy}iY#c>ftEl$!O^0;p;~
zrwU54a*enb3eQnzm6pv{bQz|s#ba!+ut=|rmy+j)-S<@VaknTj|9Te~9qf~y!*OkD
zV)nh98edrjy&AjvK4FmQ*4JMuIzvolJfAt)(RO-@sGM*+$2anSbX69Aq4u67-e8|;
zX{R@!LI%Za#$*$o)N7*SOCr2kFNF1@UT`-z%mLgB9Tpros%tj9Q6uLVF6xx0nIg^A
z2lJ^?x->J<Ty9Eb9PZ6o;tR3&mW)3Is-dJc*Va7gd8Uvf^ZXYR$m`9aK}Mr;l!-vL
zxMA7r*@4E^tJ(`nf5N8a!pHaKFpLr&$3+c6&bF7~RyuJ^{~4-5%d{mZ;!I+F@dY;I
zv}lUlHytp{nyaD#zlrGVSi`X@=kRfK+#x|G!d;l+BN*cRYen*?gFkUTO0<9D4nC^C
zi?Y^STVv{4PG>g{Nb=6t9D@Cg8%m2j8@P5|i)H6Q9Oi<7lPPg1)w}#75Oc>@GP5XY
zZ{>OHmrNNZIGg`dQG<&Heij>*o430Hh&Xhv=7LaG9SXbEa2Z_#AvLAowHAOHnzS_;
z-Scld&~OFx=8mtovO`)69G`zVt%T&>AAN5779c&;JKN=_sy%wzb@qfCDImh9kVrvU
zyn-#D55CWXMT(}3F{_$^+(KV^wPyqD1_Jwg$|!r#92=VT5AOpxJN?f5Mm8Ys&*GJ%
zA}8np_t~c|t0HfHf5QRD*Q@s1ZP3gfBeH=ZhtKtf-DYoqqM?APi^RJtBZgLvig%)v
zq4yEv?m(#g^LHh;%FctU>Jh!fGE>16a*9Nf1hlMmw*ZoE-dw(^+6)5&;4Tu|{c9ca
z?iu-WUvqUoR_EEetyhTaUY8+RyP@++8(t@Cd8n$_A%)S{-xB&lTbA;YZ($xSf1n@#
zu^&1a0Z&A2-7GOCPZ$;mHWwZ&YO$nBHfk=qKR*sy7$>ek*d|Vl@V^@$)v;r`BKIE)
z_u=~R$JbLn&^ro0rWDJlrfMH>0(6jmQ^%J{NZGC~sb%HgkEldLJxm?7UvCb>0h;{1
z(L^CJVI(5jEsaQnqJd?`)qymRcpFlr;fT_ThMB9AH69={B_tH^6MDXB76%s1+RiCp
zBjeW^@~$RM9QDR#94!Mlm{r=G6%+Idm?W44b;dn^uD9O!5?_z-n$-~?9^p7K=4=pb
z6}N>?Mf%IXBsb$D7xRBVIP}%wDSANepJ&c%?oSJurHL?Bo>5+<<rc-r6C@du${E~@
z<^*LvkC?lM4L5)!5IMtXsgx>`%Nd6B@E{vWjbBhC<Wx$e>&xDyAKiRa5qpUX9g^y1
zd~Ym37%6Q<qGz|9t*p6s8~i&^wF`2<Zbu}Q7>hzI*;UTLVFAGwfYe}5w7l+gn04*?
zz2UVC=5FY~h>Icl%>cF!!Nd06UU8;O%Fp9w<;k73mtrITG-$Y747Hh(7g9ddq)|PR
z(Ij!Bll}1URzF)a?<0Bt2-D81aAZBI*VeC8k;J&%hKq`!KB4NhkJGvrFFNo=K$<nG
zFOjkW!WL>L;l8p$L5$Jwn7vUyW13T7<J;=tjYD&)>pF8G)_tBsj{(f;rp6j+iY{S=
z(WgL0&6fo5aya^K$jLAW$sj14)H4P3RZ%)ZFJyA-cQxL%*VYxsb7MgMR*-*JU>D5!
z-F?80C(#A!Q-tiZ{q|;=L)W9h9O`c`zk~a9ir{yS3DKt>L4F@r*+q(N{f4vUO5puZ
zvw~Qiq3}PS<5&D@!t(v^FiOblj0`V^se1SS*$4B;xL;z!svx2j#t`7okcz7ckmb|~
ziOvWHxQBqq0$2PeeYLnriTwYK|3{(%`6OQ+-10$}g!V_abIluc+yXs^%^C_=Fo$TN
zabeHXbMSf<??-Bl9Wf$)HvWbuIp^kmUTk^U6Cm!gp34qdPzV`lc1yUmoe?m8&6~Uy
zSe!-*-hc3pvEe7y&`6&w7mnz>l>1in2fscoo8UX)BK_C>zZYL0tIawLxJl#NySq(1
z1EvAgWJOgBFfs%5)($=H{ArRzS#!NnXVd{eXV3$T(ZDZ_5fSPMR@|%(DDq&YkGEN^
z4v}~)#VD|iYbj^@n)GABrFum>E&?%hCu!Jqo*_3wy-Dv|x4%n*TL|0a_;twsFO3!Y
zzNaN-EnOOt?YPovpzYxyg@^8^^gV7EUuJ)NL{Ipfhaf^nl8Fda(jV<*ssG5K*ISB+
zF)q^JcdCBe7N6T=`<8dJtUY@pFsO+xq51euvVEIkCPGz0Mh&1IHP3|ZrOYw$?QCDb
z2j0f^^$P+S6Gfw*3gY-OfRjnIs&{$XrfpzcC*Ha01;@NVj2RpM>)khWtc1^xCrz`N
zl;jcr>hscR)b)y*k*H1>yS&o`AT-mHV&J?+CwDfZz5u+vUv^%70X?tJfm~ZudWGL~
zV7~n9AGk!U9c5c0E*fi*gi<@>%Z&I3%4;D<QrC!NcW@N6Ih>-$(%OA6)Hjv<YPTjB
ztWHWpk(bofqP3KC@`0y$MvqV%WvwwZctC?)wdpf$n9XPq7p2ORmmEX_1>0P$*6sLm
z6Rd<c$SEJaJdg)kofL{|Al@rvmDRy!Yi%h}{87vAA%G=~e39zgR;!H9p;5ofwHz}L
zwU9cB5xK{&nGRy$%5xPEupjLnB}k_nzB%tXL364w@~W8?qq&lW7xe0WGdx(XmArc1
zw0pk3e>?3`xN;1lDqrVrIR;1TBRqGd8>ZhH)}3pdA)Z{ZS@CxrWY-y8p9&nNL17gz
z%-ZAM(Xj|W3vYxF(u6katk_)zA71XEj+N;cH)eAByzU_T->tbt6om}?Pfrp#Ss#~b
zd1-#zIn72VjOEj^ASR3kL1BhDl8#9O`YLab_)gov7N<sSZ*BQajc?BZD;pc`A!zh$
z+v-1~x8e0ol|%W+dS8yQyL}%C_HAX<tgj5z((STJ&3p9fjW{p)5fUll7?jHBfLiIC
zO{*ukgMKBZv_gjysH^xFwT6B$wRd)#i}f8G@4|Ikh*Iw40+3-kP_?#VOyTz52ZVEW
z-E_4KdM(4f+#(xWAnz`T0<<v85pIp;rG^|C1FxK9DTkpr3H`du*#OTfCv`!C-(noV
z8;j0na@wAp#8k3$^!6V3@*iHAjrdwz8kxoQ`v8x$Wlyb;C}{&TyKE$*LC!QWVI;tX
zq=!~Y9Kl%;9M<*i?F$&Y_`56B0_5oUmH{JW`zoZiXHFUAbKv?-EQ8@zm+KC|Ys&v1
z@pRg-LPhmV0;W3Mr`8={qM&<@#R*$}-p0-1zxh!}Fd(Iluf$jepqFs5X6;>uA=fLw
zzyO%Ij0zU%vh|vXi7@`%9*oE<R;DR}?Nj~Jaj<rc4g5@_K-S^P?5hXnU}UG<TE&i}
zDEE*FHEq4K1KD}D;>&FCkq!I6)7ITOOGQ$E$o-SJOTPGhK&;#gbHE%%6pq7B8OsUT
zC2N#2b6Rx5kQ+jofg(l~9k#=20oKiKU7qi8#0-qI=QhWs7Hgen_pn`0Nz5`Ms#N8R
zEbr(&oMB=WP$0gR*_%dt`jL$={h<Mx*4ZYV^NqYlBQ4Az<d7M7rMrYO|4|$ANlH1)
z<0W&eqYr;rhVVYv<=XTq!e9%JHX$6>ptBhY3m`H(hKiyqm%3qLCl>nZn;DMvR2Mr2
zZ+vMlNZ+~DKlx6?oO7q5ebz{_b@(c9&wCfxd+PQ4af^bf7wXg60<g^9I2@LH{mv;N
z_Y06?KggD*@Q+qIxD;}(t^lAga)>v^={=t;*kBXjlYj`)SgXH&fa&V=oH`_Xy?6xj
z_8ho@acM%!-;(u^3wOHGg#RQEyrcgxvaGp;c5a7-9SO7;40|CXya3h}l>1qS-Oi||
zd;1ouIMMh?A>3im($~{-cxhjhI9BByxDgNRzh#0yFy<>VvYK!zWlriTVel5P^`*Q{
zy%Sx-ZI1_G_M_+53O9qCzWvzQHzY8BTm%>Z$ykt(Gbm}uei;WFNab0PQ!0@pMgS&$
zCEGL{`!9fZRi}<nP`gctMES!HwhSE&9%6INeKc$5T84!f2{D+A2EB+ylbP@&mZ2S(
zT;_}PXhCUhf};7TqN1WNu6LU}z;rME6K{(IG-tr%E3$hYC{O7>Q<BrnCby6%H5Goh
z?<M=>e@z6Gf8f%b+oGdfXG)tboi96{dXi1|jN~&fvMk=r?5v<*CW0(n#hhM~o`Y)R
z2=B#4mDA(y`D=vn>($C%vk)oTG&sm`EcD(%z|`<!J>i!-bwWxoTeF4_w*E>`ORTt4
z+zB3BXROC&Ld{$153M>2VJQnAISv)>+SxJp1a+Dg{r(?$u1!0;|1e4j0INtT>H&r~
zRmdV~Vv3Nur8Oqp0veV2aT3|06-K0#*vp9(kfCJWyChDwY9HA%49$?h%-pRn5pRly
z{ml-E)q-<_6d=9ZM%nz4i%C>GM>X*90k-p%Hd}n8*Ke!s;ZNwnkY7|)94hm$!4zq-
z4$XNgg^8*PB{})m8%^bLvyS%`?y4+t8q5e2k@yD7>oeJ!?nqk$9Rs)A-tLkz-mm6+
z$bkO>!3I;KIR!ik=T)uZ&AIM^374W=@*^O<rcW${ra-k3&tBt)HJ)2LQtDT(#GEGH
zQkxGZsi#IU0m#h*k)bhLKXJtP5qJ~qWAM>9=_PF=hFlPQX|w7P>g=dUHhPc3Mr9E{
z8uqU8#=gAz`~#<V!4(Bj-ETcVQXZa(wAvNk)XH|p_Ip5>0}A$gS6;;WI_9-q9Ik9W
zKH!Fpdh0&HExZc*zqz<^npowQRN7_<@3SBr0)@_nW*EwyM8JF0&p2?=9+)yn!ssdH
zd5MCe<e=OHms(3Mw$*AfBLlX_W9&nN-WqUDBb!*U8Na@BU?A|-{P}x9z5ea@;^-N8
zXYjyT<7L?q%P+7mHm?94!HY!o`D{snBL&A_BaDk=5H-#V<^16;yRN*FZz$MOn4$@Y
zJJk`cj)W6CiH-D24p+lUd|2bf0^ON+jEE>|eB)IPSBLwCnwCb*CIjF%WXDjmU8Qtu
zGiU26DlrpP@<HE~GL=Snm4~c(?T4_vseVNsO_yT2W}U=`D7>#sUa?aG&v%-17`S=+
zPZbs66|4VyN_-1roXxwu+GmY1>$mI~j7~zc5z9{A7uqkCIF7q?mv;DK&k~z~89AzP
zWpFXj1V)k?cpO%&hScjB^W@>%*GjIU?9c;!!0sgH>sK36X-gML>Hw}<ODk7~&}Zs!
z`S*ljP-Qch$=$}FoaVi$E%Pw1Q91vWV2ND@d;xZf54bH@C*(79MY(BTg#F9h|0PSQ
za_9FPc=c5!D{$-gz_wY6`3(sezhQbX1HPx9e)6Tr*82^f1nTJ)G)ihw%FIHdVh4LF
zVmlWXQ6}f<{C|kS*tDrkl9e5ezs=^Iz8bVrP_&kK&$kYpiJrXG=6gD<d{&o+J&Poa
zgyqDgtVYRu-HZ%h%@h{SUETadC}C|r&ai+gr{nO&z;Jd|Q_GG~kQgImBNu7s{d(3m
z0-fk!iRd8BO{yCU9JwCsKv!BaUA-1g$Vpf)whh`s*55sotKKFbrIm|Q@gfngL#MYK
z{2PJta4or0Ip^P~EIJTjwKbIWcn}R(#_&D_eR7!u1uk||2U1}JjbMlflz(9>9*nv0
z;v$#PA2Jc!&VeU+sziJ)3~%2!U{*&dbo~65R(|GLgA>Ih_(~i%>u?{ib?La(MP}zk
z=Hs&GEt0$C<kYLut~_YicD9+T)m`&@VzuY+UqUrhWbef7=pm3E)#amB2T?s4elX3`
zbp8wInsxe-EYZQXm>b1TpIKia{3e&oKiLjYOHuQIXgP3<FuOLhWacQ<*A`c<RBO{a
zZA}5BKfZn8h!kha65bgyekh-lQND|Z(B9<{YSH{d!r!j!pEgpfBGmUV9UFZEc=IjU
z3MmF&itAyHjwTI)aOHTQmn0%<8o(O|oUw4>WwpzTx4axWG^IMp`->jK)KrB?+!4v+
z^9nBB&Qs1@gZvSf);|Y%0pAgblwi>VbyEXKsHhG2<_1%Nb9eszV_V&ru2to>>dkcX
z_Wrs&5cRaZ{nadKucw5K6{*OS3;#Y3$Yr_<TX26@O{e$JYBH$Bku=NMzWEbYTJc}8
zGrmv;#E?HHsK5+R^*-cWnormV#(O2qTa&v@&*)AG2r>fJDD&`9bj2lm;yAa>oPf~7
zw0V<$XD(^0^k_;|QYph_Io{OysL4ql{wl;y;)0k$4VRdg@o^iifSX0_XOhAVP`Z(g
z=ooRA@=ai0U06I5J2`EwQ$?XEofc@jDQ)0D5y3u2^Ih(d$uk22jfS-R!qlaa^jwd$
zRYYa??yVOV#JaNa{V?CrRcpi?wmQqjK>FFAKWVaw`COzbb3RCD+`Ox7_3wMHrCa7Z
zM)bOB%(_8DB3~Oy66hLM<gZj>?b^l|ieAK$6kmP`nf@b=Gc-Gu)Gki)eW$XpNW9=q
zIc%OJY3;4YV5^GZ{q%>#dwhQhkg6l5eTUqIEk!9)ErgW8mp<mO+*C`o%z0YoRZ7kC
zbg+$BX;GQWWo~(6w0_y^BI|dkRB*VF?Og61scU*K8_N%b+ZvTZOLwtpqy<t;S)~;f
zRsHq10*{>)Am#dss$=D-X_9B-TLkbS{DSc}J`WAGmE>FHS&lvLG*WXMvy{*n98{A|
z0N|v^1ptRJcQ(M0L4~_*0zv0sU*Y_2&jPYTFV=Gni@-YDO?b`UrQbV!rg@Q_`tF`<
zZv%8%&-EQMBkZ?Y2=MisiC7lSh(3N=`y%w*^Z#Ceo89MW05#|;J8Q{xy6^dv)u>Nr
z`xIsWvzi&9I!VW0;T=RhG={yLL^0o~ro_)o5voh9jcn)<_3Cq)cKq2f*Vl@cE361Y
z4@k^PC|W!C@oG6ehOQnRrPZWY5{NzCp0xCh3_cJUzUbIgIRrx~6iQNu@?T*#HuDLz
z%IM<i$*8cK5!U}$9HU&Z`B=NLvn66WEJ7>a`+p}SAyU_BQ+Gq5;p!5$66@N2_#y5c
zS&u2hP!Yz+={jLMU0OM(05rh${OdK|#F(-==giqzW`g~D%vZ3H$jH1N!N0Nq)&dri
zkG!F_xzAXGb&Y9{ZfO0V;jg!r@>)wuWq2HQGeV-G+ZPu}qO)AeL;Fhvz?`F~2;7`N
zB3E5jNt%%OS%-rt({kwFQ%V_?=>ES`FZ*T1Jkgi0lh-_1BNq>UK>KFXM+0=r*C`oJ
z#vw^*(5zBaECB_HQiNrpv>R_Q(<izwlLWwHF+z$o1go{@?e04WS#e_G9yU*L^B1+`
z(nK*$9^7N2`DAchR_-eqrtG800+p|Pk-6o+ea0l4aL-dT%Cl*JMDx%iE~DpQyMU$4
zU}0(h3`tieVM_Won#5oY?$Qg=bHP=ki2oFhCBR}=s^3U^H!o*^X>w7WB)P}DC#By{
zO<*Fa$3x@aUV^6-5s49zi1SGNO+SZ^iPbs|jfKOeUQ8#4Z&{5meMA_IBm+MmZpINg
zN;%tiQkuS$z4n{UeBAs;Yg|HzVnx)B@!(Co+pVnX@G**b7y>hSBO)Jy8-gUxMpC&L
zjOe+P0dI+>+&2dLp`E0K1&y!^Hf=?q{nFZ?mX2OVs`0>ihbvK&RaVc4CfzLZTX!{F
zuH##Pv-f4Ag<H!E#BL-#0nF{Y=(|=qqoF1i&T+}UkiIx2`JOR~Seqx<23npCsCsJX
z-Pg$^H@qz+mHfP*CTT70RVD=1lB7*AX(-()O1cG!8Q_VJw{cz$j}ri)G>e<$$`T2M
z?qr&Ap-E}Qe9=!H!JqS0D|>rpVaJ`lZqIG+eo^kMgM)-viJadD-HFv_3n=^+ryW@3
z*6>2~x>wsjgA0fB`mLOAdYzs5oHsQn?zbL#6J<hj;k{}ObLH~Mmewp90$4YE`XH?z
z<6KTdhXQ6F&Iwl8wdqqYzwV_n_r306H+)h3#wqBx0Rx{l1@A3M>py1pH(b1`pA8Ut
z_{b`KCGrjzZ(&L#tWtL5+^(z#j{vWJZ(j0v{k%^0m9B02HINLThu%hv^P=0L<dc&~
zl4hl)lsZuRS5Y22;pDx9zAt_~3d#_RobKED?l{ZP*YX(q`IfOT2Il-3;-B;&gW}=p
zd6!kX{&mG1Ev|-Z8AOgL>8ydUFb=!ShapRM+-cUM<Kiqj0Du?xfs1Z=v3-P#0<(X8
zd!mr%%08-vl3DU5u*6hY2+cNajec(@3Bb@vn`2MjW&wL$untz-#9nSR^2FYfplcy@
zeUGFII#^g(bYmi>yjjf=*QFJJLIX%q-{Tl_jB}cp6+f#`xp;gis<lhN#&%JjHpf46
z@$`&VHT^|`r>R~v5Vb#8ZzO`$5qFd*YGnmj<jjdN%4puoTNZl@9a++=NMe`b$`+Co
zl&diQ)!v(%A9v!q6eT+8o0#aQoEZz0cS|}>|NMN)9cbhKQw4S#$Nu~J*HX#m@hzDc
zD~4aJTp**caK;=C5k9%NPzD*o{P8`IX&$xQckXw7Ng-_tsHEY`LXw!yOw~>y4!PqO
z@lHJO#a5`W7RpcrTtbO?x0t^zjIe3p{ax}b2bMPX%bWM8ezM9iHlO85;HbAEkvdAu
zzhkY@<6EQB9O&b)HL}?~FLvNGLquYoNK`{6gxLszbqc^2q(8g*-$E%ZYw*~0w^jvf
zMTml-jMEB=`1*3}igI?@eMMp@rdtfaT4H=xUw~-+;-=1nO3_#uro;xZqjYZ=5jG89
zrYUTy!gx6~o0?DJU=x;*wu<hp2OB7nS&2BUnUoO~A~4~K>sMBJ<QZ7}8HUN-Pu`(#
z^$Mcf^o+8lSr>~}2(ov`-N<V=)Oq^{5sL(O`MCu3T*`ZQ-BPw?yjXsM<o9vTbxd|Z
zBuCDAo=2VF8hVvC?`wcj)uG7R#|~tj=Tq#{SwArSlt|y<zvqk%v1b}_$2MSQ{<7rs
z*1xlp**&Ie+Yd<aj}LgV9f6bkGSNh(U<qk7FIrs=7NZ_PFn1moIX;HlxU=9$C*Su8
zCZxP^i4m)hA{H&>KuVygiC|_J5Iy$7%{vTSJ%qqL>n5iIgf00<Qx4HX&m~)5H~NcS
zUV69A;GLfpI)jXryXvr>698}?P4N%574xDHiM*!Y_O#UQIe9rh(o76%$J{?@tSHIS
z#VCP~Q*K>>Y*$>RL9*gF6W$;0)n0FMb(`Y})o+;UAv_5Nhvc6VRMpio2wE@AzweG0
zmDx|EbXI1@VH0CU+As5M`M-oPs(jqG%74vSsoQe)aRFc$z)}~6RG=Kp#1`|O$!@FD
zb>RQwGqdC$>~V3m<TaH+)<|mw%IBHd{%bu5xsDrE`P|Xn`BN})xiVh9bxo6;T!|!p
zXF)|pmkj`B)g`L~j$Em1zfrS>Pbh+d#6a^d*w?-dt?N4gQg;~9D<9b9I@c0=Q1_jt
z&X;vIIw+a;?!vL6Op4$=o>#?^UuAaGe)@Wnx#_%@qJ#}rrq+Bw`RUJPd7j9FI*wy3
z&6n$er3#BqKS7N~tiI`9hfP@r3Z$g@T;N4`UdSABnnP85qVt2JMzbx7@y8b5WT42>
zORIo@Vo>O6eC$0KfobNNt&$(HgHR4v-Hh2{?5cWNaM!>Dl-uM@k|}y&UDB`#&caEz
z6#jtHx4`nXpwe+65S0dQ58xRzGoHDYcq&=FL-T4u@%9*!3VP1je6Sxdd`GWdnqbYz
z?m3O9$I|5hKJ>T4&VV5aJ9n7<Qv0UhLLB_%Wxe@pZGmIf@+T)h@`zqO{$?9BjCtYM
z%8lL)YIdD<!u4MuGm189Cl!=cBql?$Gw?yzhnKrqNPp?aTA4(PZqC|gkoX3sNpq-@
zBY4DfVDygX)Nb_5qiJS$gwHGA?b+LURzH8|gWt>Z#2xw3u|F$J$x4?QbT;6;+vAN2
zKIwk4=j+DX-rFmupje)(ziYoa_b`bvm<=UXlMvWBNL8aQd0dy%XmE8kECe^~&AV-q
za{U)q(5XE8;(35;Z+=fI@K!ctx^s8yrHaVF*s;)1vfW5G9hgIdjH}D-q>}e*q}T4h
zoL0?khR1Z>4_MN-?9@^i1<fm^y`HsrPM&!|;7iE1kLo~t*TWb!^fUTtggLo=pR9y{
zI05bK7`O6<rlCI@HQrnreY~!QBa4#ZFO(#``82Yh=21y`!~VhdDW$EzRmvo}jIsc{
zJz+r;n+(vWv>!HXGGhHA5-k;=77Jd=qMThoGrszBgqJ7gGXm_5Y0*=6__C-a)lti2
zX5ZO#?Z$6GA-d5mTMn3{##dF^^JaA#OvP4|O5uN)a#Miv>yo}(auJ)7PF^@q?TwcC
zUyVni*mz&xiKM-oI=uGW<BA{9`%mSfp!h{B+p&4kwe>;(>}+~V+cC!>#yQmgRMkh>
zk~!no-LqC^SQ`BBeMah<Lw(r+T?hT|dOSn@`-Uvj-;xOKaCmE?wD4upQ->_SMo8+|
zi#BppvO7Ng0C)0)pO=&!&!)aMMTxoX(yia--Q`11<XW&8a3QTxHwVjs1?7k-YCd-y
z(PFt<`Ljp&0of~5dRN^j{-qA(_M!OcY^fMXB;JZf){RPS_eyK>F@^npvS3U}mMaO|
zq2R^1G;=N{U{sispYH~c_Sc4m5<h>I>eIzG;#PlmN*8NMp<wYxCKw?H)igY$pU$ka
zWt*?2<x^#~4C%}|`<=aFPu4Ow!k~;S5-xe<Qvm85K<Wme1{i8-kwtPK8gJV3&x_Tb
z?atq>>4b36G!MdiEYz>rmsiF#)?RGNChlC=OCn->R^TstPfW*LdQTi3(eYbMUtYR6
zPnFw|;&GGivjFF@hjQMbz=&?XNpiWP)Mn}jrhrYv=#7ZCgjvaEw}y}D7U;rW^7~Yg
zxoh+|cQx|QVqHd^KOqP;uU=i-jzrtxi7y{iHrv14w~ms>*3QV!oz-5Seek@$HXO)3
zFHwe0P_k~chPJJBZnD{a+^BPh>#sH07zI{)7)L#qD0LInKVQWn*2TQ5Fw(#m+M9=+
zVOcIgwT<4Cq!dpY^o#z1Vm~Y3Bq*G7^)GZZjV!haisRh2u8J>pK@u5>?EkbTTV2cB
z$Jtj}i6JT1JXKEG-kswT(D9>!J<f?1y2#W9@_t5^FOxdQUN>B>nT_j4x494AcZAv6
z)Gc;q+(hQfbVEuU_8T$-?d|%Gc>DG{XS%ujhv9^D&*|W4ZtB?`FXbpH*4A0sPsBi>
ztd#j;lGqF=unIsjnWw88!^nZDvJ%!VNg=ZQj?ld}^g6sn!of&in&EP}saAjnE93Sr
ztrSd#l{Q8s%}473o;0GVIxDu?Iua@+Vv5f$Gfg@(faq$P3QDM~tvCjn%jv3XP7*A@
z7lMnDs2(yBMqR&PGvlr4%q8^a<|eGFO3@ujj~Tc~vZs;GrJ}wy^5^7`gB^gD)S`!@
zg6!eA3V_lDO-2Gn_GLlH8k&#F6xkZcv!btcbn+F^-1v2e3?0he@1!`P{I`|@Ph@X=
zVSPse;ad+P{i?YQxo}Qo|ELalbK;JN8$Eh2{QaYPso&jb?|p88GhD_?OI43s4pnpo
z1%7A^8%cBzN+Qdu0P+G~PTx4fNvl`iDKL5M6~C8K6sao;P>#Br$A+3hK))O$lSr9;
zTop)iDVs-8qrFnb--ij#1KSCOCLd{63$^$>9B|aRNRX0Jhhh4)v}7*OCn}OYRUul9
zr?0k2B*Oby?y?FthhvI@zPcFRBv>{DmlvyK`?nqV6rb~3mb@RxX`AfDh551sCjw=&
z&~glKy4;MsF;TnC63+!8A=G3EyGAj6rwRMnr5ZV4JvBeh3}9l_Sint3wN!9!k<Nk&
z9w7mp=MGblvxW_qHuU4c29bIXzs$~8kG6h{bB=S;*|tXcHc_a91yN=X^zmG8mUqx|
z;GnCv4O?f+bJOtny4DS1b@mdZ-?5h7<B`(t*JpPcsNzF3+3B~r%<mHE+ULx8OT5QZ
zU8g&ZIccvvhRBMVfWgq&ZCAX)^sUH0p#hUpenF6fQ=82)!L}&VkkjWD0K#>5m~adK
zxtYh~uBrc|`Uf6>`FMPwL-@e=)lA6P&7<YVZ{iag{Eisrc<YkB4cKhLLJ0p_VB~F3
zMz7P7Y(3)H-G1+haa-5xU6?$&c$tE$gV;vf)^tM$c)r@j$P9{>a~iTCJX%jo9cGq^
ztwTH%D`4DUR43F0>=LN;_GckU98oHh(E5Ie_)gyNOTmK2C#_nN#4@>(u@)>({)XH^
zCs(CrM1_=kP=sw0%`8y+kk!s?sbj=g+n+IW1G~#CJ+(lY7)kHGObTr?$G~P}5P64C
zOqBxmkeZIpc4}56@hFD6Zziowj|bTqN+aMedjO~aHvMsz)|W1#bxtpGdu%v~jfy&0
zNq)k1$gH7XXWDD4XLlSq|3`Sj*CN2aeC-f&vCNU0y`!vd-@=ighXT9svCP;J-ZNl8
zVq^~6$UQG<6IqNy2=+ZAJdIQ<7B&rG{yj06Sy4*Yr}ERTKgw^GVd^Q_C7-$0y*1>-
zFka7vcNvoAVVDM{_T`}v(l+it3b=(96=6642P4PL`rG|M!b$vjHkrW|9siQl*;H9T
ztWnV3X0Fxw1bje#isX{cp{h#_mdUuOgSyoH6YncgCE%PBkw&(~9gvV72bNAPwi=R1
zC>XAwELX3<yXGPz!{KKlfiV>nxNBTyMp_j^%oyXo#_uhk!&J4VLxv;U{ToY?Q)(<S
zY-HMdS(@BPQK{RaeL^P3^7!WU`MLZnAh~?rc<VfZnu`v<C%c(Cou$I=6m{wC<D4>x
zz%l+&Gp_1JI3@I1j$%SoA#$uWZHvD*+Ry#@0!~uIQvHn!wwE#2oI<-cvGwp*hr9*Y
zpB?MJo^RfqOS1fK6)SPO{}zC;Z8#H|{o~A#y7?{8EB^O~bHhPk>+kQHy)OZqkXFda
zF$ML$-|Rzg34#9^=+pUk0}y<SNfJA0HF}3>=+6~k$K7C@%TRGtV~N{hNr{0U;dR!&
z?U3<n)knZ)VWjp4S*dJ`I~R*MKBBrdAS&Jho*yf_>IWOJ=}R$d1lX#ZSHW+5@SOB2
zW^&&P`%S@wy6Zl`^1Ky#JF7f8=#ipO!>S+;K=HI%2ELOX*)>!CXi8%8Yccw@h^d=C
zD&CJ*-ICMun9%6Bn#Ol{Xc6~tMuH)mlamLKy#ZaKEHU3DFc2tWr1yoI_A!`EMYCwb
z=vHDHU4iKu?QBJk1dK8Pafn1M36`vZ;WS7K<&%ER3Y(=i29l5-vZ??%W_!-8YWTsQ
zKVcH3pe3XQ%|Cjso}LovhxT7$VQx72MITLmF5O}rB4}+S9lm1*R6RG30DvrqOKZL~
z<~lGw4s1<y9PmxH0P^}j#(=uNm_@Og&z>Kk3<$9_crd4`t-B-639tW@nd_OnqkG~1
z$qpIC5>KRWh3p9a^6=F@B^+o@^2QyI7Wtt!gan1*teA|rZp-k$99Hao-30~`v&oxo
zyh-NjpFFZ+HX|cO7EzKVG^L^eM9r6A!k_Oi82=`O-yLIIrJ_o!;nO70JWIVHP`l!7
zZ`#M6wmSBFx`$~EaBERke}j*0;fE6;b0C(2v}YpCGCU8q1`S%~jUXXJ5^)OGQyLYp
zUlm$II}JF1>AVh@c`^lPX445S@2O?V6>IqvzG|QynnRE!i(M4QIw(HWw+Y<7u&KYZ
z>`v0pc4_SREjDXkf~1IHsOwmM+}Kj4XkQRY*y0`>Nw>EKM!K#Rx7(<6ismm+Ztkk8
z*>M(@K8i~sR{G(t8m`D(pA2f|zgX2RT|fK{XUA@W_OgJc)N_3XcjgFAYYQ6|bG6In
zQaYRC^s~QTe@jhz42-bbYIx6ecc{Hp@W=4|PY1Sx=Lvq4-we`@@KG<q=VyKA>s3x&
zBfYcddf+geaewz`Ay3B@BN6tnF8=5_U!D7(-Bdlz4v#<tUpD;KkTD!(&xVV?f#zK&
zhJ1H8!j7nS<bC?B<+xeH_P>8|&?*wAEh7yF?&B77&kpf1w=43nNo!l2FzeGn#pa1}
zC}u{+3LBB#!!{EKgJ%`)lLgi8)`wP+Ku3y2k<eSi*n!^QCCoH22gpk<rP7?RwAZY8
zf@X>+I>~$fDuU2~Bqb|zJsu{s<H}M;jHdk<Wi7g9a~*u|3TH`>n4p<pCCT3$Q?Y_M
z#I4>3G<JAZo+YGgMC`Qz_X*$+xQsZryl7$PRMfKj?;pFKRh+CwrE6#lpG<*dC7pJY
zkyiB=^x9g<q<N1Hq-&B^8yya9-jNg{#K}j389#e)d#_#?7jub~U}n8ebvcy!Jvq6#
z5`#wft%0%OLMCa=2ZiuBLiGt8Z71q6SLG!HB-0)vBsvGXxek-w%z0iKlR7UmJ6x>o
z_wjpUe@EUQR6u@N;(QzigDA9}Xc<MKnk$44`#U{DmAAo@KGY%QhVvWG)}w?1)aM47
zQZry^)^*!|3*uj#9;x0U-||*CKbARi2$Bkk34KrcAoOF5Kx?DkfL*sHLvUp~M;u3o
z>TeaE$aXr6f*J3xc8t^A&h2*gu?EAa0)3+0Gt#3qPu-KKgfp$DDQT9E5%GqGE8C%m
zb#pW~@)rU}B_FR%G2WZ<SlGyU;BZH<tHD`Vi1hyAE@`n8B)|Ii2tXF<bUA=DwcWur
z$q4bMBGBGt*u3)~^=t!wCRed|H}gn#8rGpZ1gqeq`*~oiMLQlH@<wTeruy^(cb9|=
zsb-Uhp~e@xEa6q?x23FoUkbBTKCSxH*p44HP>3BO%x_-hf4!5Ag}_drbqjn+gN56Q
zWm2$`BEC{tB!(=WyO{WKjtL#<x<0c^x`o>5NqwvIq=R$D=6!6OyROOk&v?*ds6rg)
zDyJ#h`|XT`lgo*J!4_14U6VXL=uDf3!U4|rVgx6K;)Y(cUAL1CCKr|zcu9UM9e#ud
zr`fm1w|Q<s)vk$`KAa|ungomnycc&Y94J__7}C+y9n0t>gd^*^IZil(jcaTGrN_YN
z;6D>}KIf%lNE;%KD~2OK{dX7uu9K@>xJbGuN<^d$_(iYgN6Y^B91`4#CAB2-$(guo
z|5|(gcXW<b4#h4FNTsHL@;?|JQNLfM{4cX6C24ZcA}ok-!2AbkX?}UQ-l#^GBcp)5
z))GGych5#58Jnq!w3wDPLMb6}-e{L~)R{}1d0=>-Y_Y)vNZ|8-3MCp-x_+3KXWb_O
z<ESIq{1!@|B+$;_lVpk*3IW^??7SX55^_Vnd%9C~uWx@|4pS}Uvo@n&e!oPq?nWr+
ze*dQRtwewN1_k3QA{V+o&o4Q62@?m0`22Lse?&H|RO!?bA)hjFy9{9Bzvty0pB<-E
zXh|D*grNs|Z_EnakHn9Sy^~71X{?TFzn3Gb`*?oM!r;d<W7NAH`~|R*eEvK$hR&?@
zKLB?@h`y%0<>b{X2#UJ0EEX2)&ba&bTP$@$W;8DJCQI?%Bt|28M&$_SJd4F50=p4)
zRpY(m#q;OnM)&2LjN-+M=d4yMrh5gFU2$V#Na-fu13NE?;Wp-z*5}z-6~WICxTSx|
z4=Vw?p~jp5o}3+o*$I)u4M7T)t3Gm-AbAmuHh2%qvjxBX?6>@@fAz2VkH7j4w9Nvo
zV6_jWn==_zn5@RRikBxREY~M=)zpGPQAfQ(<K56#uuq=azu!Q9b!=V_!X!9AHg^FO
z=LJuapWAC@xMo$@X~4R0J*B?<q0O)WZEL%Y*Z?8x8JN&fQKCJk2;lS#DN(S_^H`N5
zrjW5Mw$*Wt#m!G^ibB|TYaMMX{%6bOGWy-sa12K8NhO&Ef@`t1K<lgncUepbEfJ?J
zT4g{#uT$AY>FYp<b&v@$wN7xNXvOv3&j&{xSvlKEB<p&;qN-LjZQVU%@|W{5-Ao{w
zU2R_{HUMZ90bMESuau^4Eoj}X7h{Yth&n?XPhrG#eXzgJ!NDPI$goXK+t!32<qk*V
zg3M>w)>73giAImQg6<}<>IpGa@<@=16a&Dk819{+`(uOn3!zg8<K{<*_Lt7>U<dni
zuF0b4ejPrCeD-<<aTi%})z&F-oCic#Ve57uqxydxlchHLwYyJ}H=S}hB#Z`sck}oB
zJ6;vEZj&oYsgAL|S_|ED=;VC&&5y&<ZQT7F(0zS-GhhcmDwmW#NUKB>LM!E`@Mtue
z3{8fLi7=J&G}iLy$xEunv7E1|)(#asfhL|~@F6F}>$(tP37I571FKcV%NM6C=4<w*
z6Y?S-2-)qyN*i+jT|db--we)E$QzSl<FdPH>yBlK3%3J*>3gNN?yo?J>2W%lb}Uu1
zww#@vF&<CJvYfIIM|JN#^ZA^mioovWa}jb`7NZy;(~5)Hj3O_{^O7<fG1(teH*1b>
z+~V7Bzht#uFrK|)GM+?M2qxn(T4~DB7*JHvCvUl$V_nP9(RG{)VipOWrWOHP<Ryj4
zsPh^_jt+`4(=1P4k?BCap3~Nbld~mxHl~_4)UBm<n#D2@Xjv>8ywfpBYar7a<sCP!
zA8~xRk8_et_w9Rcv05#8_Usu6kb|V%SS}iLkhxM<4XgDD-+cWQrza<jMhECjhHWGE
zDGC_xQ*>fVDBPP~`*|~rbkq3YC4Zjx64@D|3Cu@^D>|-+;ebc>D=5S{-G`3q7jctP
zBB*}8X%TRB)9~We3BUfEzu~vP`;4ltFean1mXjqs&Ml+8IV~CZisjh~=L2S>2{r(-
z;cBfg+DCW2J`I7m=x2CU;8g}VRB)(J3fRmXv{pQwyhbai+lKXONm-6j1axP%AXYt4
zfZ3^3_3JBwo%9)fABPw&l7?(}?(muynQh;+`GN3En@qP!b@xks)ytD~T<S?lm=0N<
zcZeZcc{^a07^l%_gl(h$T5Dq&E>D)_1m`I7oZMvKEjDu)opp7D6BCvcF|B<B8$*A^
zA+U4ZducCE>tZykjS;Vj@k=GatZivqOB*XhG$~JA+isK$LrX46>s-uL=Sb#29hEJk
zdmm^C)DZt%9miA{;l8?$^PZ}1sHz%M<`j8Js}PfKx(Bl6#ljE=Pcf(os|KWsFl8wS
zsn1qSHxJYADWIe`=|)gT55~GvR3Flm{_gt!g1Odzzqv1|t+}3_;WDe!_I2NcsoMpZ
zwi(ZhOq#=={RL7NsP6alej9sV+W+E7{rA$TAzkc(XT0q1tA2O!IsLu+ci6hlP&RDa
zc7KHSCYOIRQ>@i7bUT@^Ea+T7ol|%kytd#Gh!%2`H>hAx7(ABN1|B?o$;nAeRW&pX
zWV*#DMF<8J6go&!)l_R1q=XVmuG3VG)6*4IW#tuu6uy>l;67@s_c>ABr12Y@84NY<
z4HjYRA9S^{`-LhJbF24-c^THN9GYNdd+jG{0vaaEm`)|pZ?*DFM+NrzbLzTcRE`*r
zCwT8zuUBZTX<Wne7tdnliGapa)hnF!9PiJVmL<717?n|$6Y^rrXgUK0<Jl3Xr>{6W
zJEbghw9&M6gRL5htl;R{b)5BB=dreBU9a$1v=T<6sT(Od=42C^&(EmVD>MbTjK)?R
z9UZe=EwD{PnPpUUL$h8}2S;cX>$#&fiq;3t7A>P!3r3@oruK}-Ihi(i=QulgNl{2*
zQWFf{e*KWPZJ96Uw6-PBCC!7ip8e^JS1(R+-XT%8o~KVA@$kU|-n#R?D3ZxBoIqDB
zJt%@#cqb*r(h^F3Oq;{(g=zD$8EI``SS7&ABnxqaN@mqK6G(MMgBOsmg2yP+v%y+G
zLuaA{f|kUti}i{Jj~?>HmtUZBC`N|$S{S(2!$}2?PFAd{mOL~lS2HOM${Jh~P};@B
z&x$PbI440E&WTwr1c}~sWJp?ThxP#-N*or?5>GAFRIJQ(9|LM~lR=QaNo&R@EtgFo
zHe~OViqYmldRE6i=xZ(OUc_%P8r#4B5Wp_<+bl`8hoEA5m$0!RsZ=*N{jE~k7I|2u
z6vpInvzHU)5hHVJOiYsksd~e5v5b`))_5PNS8FEY5!1;SmltGNBkaA&DT;zTmokna
zK-;!x?YcLS{_Fc$X<_0z2Eg^e&N>^*rAZ{MbE(#A-GR7NB_c6#@-qRu09sMkV%15>
z+p{c|#mjT?fegK6q*uu4{0@iF#~}pP>zWrYUSW!79Now{+}|h54Q*Rf*K3-li8YRu
z$O+qaub)`;Ub&aQgU|+@DR$Vl6^8BaWGKV=&qG|#=^g=wfTz2A0+JiIO{g=F0M1p|
zq+N>Aa1Fc;;9X{79j?#I+Bt9+h$PPWYwuF-Uf+7)x9hx%tYa6oelOmAZ2$N6d+z|y
z-BE#U^ZO6q2B`Q7?adgwl-DXwWak~whny>A1Y~IlDTkKUohBnfPl6)a#$&Cg&}B4D
z273R5F!%?F+SSq2QP(ZY#hR*;2rH>;3f_5??z?-q5IgNVUGLnx_32G6(oS~y4(ys(
z+vf-0`%Ilj{{kMI^AttN(a|xFS1)LqhS4ZGrfaaS!LAxc<%IQm#eBYCK0iaDsA~@j
z*6W7xsAMvpvcG>omgQou$qXjbl%-_5AB{`K<1w?@zBn-#Ia+C6J$s2#n*D=oj7Agk
zvc%e!#cDy*tZ8k7caGq^^tqQvIor0Z*GuMSXT1IPl1Gof#pDX>8lFCTLXlTYr-8-N
zGM`IHE9(L_ID!M~1212maQ*r*!E0VTuL;M7<Ku~JE^WnPu@qAlp7ErV1k2u2RW(^I
zDH6t$5w^9Qo}JOO0j(TOv&Py+j!*sGPevJ=xalUrm>7;>x|_{$?oF6ObKpiAj2!}O
zs63>mk;vGz2`LhjK?H{`Mo!kORx4h-c*XwyK0p82FL?6w5sx1~W<Fow+Lo#cygFI1
zRxr*22V+f<NB=i3l`E3qx2@|G9p_}j@iG6HjY){pN(o_LvK~-aDVN>0X5Rrezzwdk
zy)K5#O#(ZK^-SOQrjtz|w{4#IE?W8Pni>EC+q$mcy^E2vA%qQiF7l85ySPDXV=y7Z
z1&6i~+$+`jie_0xSr!yUL7A5<mrIt*C5y#^`N=7xQORs?N}1&n1)k^RS;=@jjn@!g
zzhutbm|t3evep6R+2~{6NIwr5y9BDNwXCW&tJSgtaaC2Zsw=E*Ba0`Wr#YaEmU4h9
zMgbc`Q52X6?qnQ=mD5^BP#aK*N(Y*zrD@tuuD?2Y#j^e_4<4Uz=bbY?{`e#I_NL_J
zh!7m>YKhk#GDA`39k3ftHPx90d)b|2uo|$zsePZhr04<OHe=TV%^!!_8Lpkqda$)l
z`sY*~087Vje7|F_dtDnIHC*rC#gzDc+GT)k<2!ZUPj_$z7jvz2irU{ry9}3H)ZQ4_
zff@)02ceA{p^kLAWH(QVRCO$QBSod-KzoPB5`u)>XIe`3Y$DPjxQphpSPsj%hWUKW
za<S|HP<pQ6g6hpPy=f$UzBwrU1iT2;4IQ&DpYZBDdf!N?k-%MYrj>rmO+L`nR2RW1
zpw-c~bp(F}XY&(IPtNe(ayFk!!rs#b>y=|v1g6s|lgTu`v&NV{o4ryd`osjr<2}kE
zr5w<dclYofihRUqGG#QLfPzWArfDkbwxX>YG=_3i#EiI}wrS7`7V|l6+i?BX9ja={
zWHMs0nDgY>W7gFgZN@x!@R-xHCC1dW4n{?E{BAv`uTHwC#IL_vGCymW&ZI_e2x7|G
z+si1+lA_3Ini{Qj%wXRz%BNhrc8J!Vm!}K7Ry1wJv*(YI%=(JxIxXXYn5GC$%0O-9
z`r0;042L*<o;Jbl`69k!!(fd~KTBb>#4M|IM=wN%D|F&NF-6vTtaVJMQ-1l&KgHUX
z<!ZsJS1<Vb>#zCx%P%;4`4rczSynYpLzxFk3oedR0u*73WlEq{DPgA4<{n>W<NXsO
zaSovBz6!#|TG{HyZMIJShMr+k8-3LsA2*_H@b^Q5yP>`=!7h^O#yYUp<7~j$(B%~r
zvMu112OTh(#^eTNR7@ruNd7|3IlKV|RV7jaQsX5jqf&}2&(N8{dPlu>EZ0kx%LPTA
zMfVB;fz#=X$z+er2xxNNQO2m-I7(=v&^krCO8z=L-~&xvqfoJ?ucNWCyjLsWYQC7W
zSS=%vlN@z*Q$>c(4gB&#Ts0n@NB=)#$c%}W3#pT8w2F$RjB#66SntqA-dF1k7d%00
zbY75^BSO$rRZC@0SysPiwOn%ht=n9`evRY9J@)qYMA?jMq)P>(h-*4PxAud4pRimA
z-QN8=rf1jp+n}gl8?pr`uL#l3qjsFvvrzK;Rrl>(AZfS;uENwMP&-_UL-2F{S$ADy
z>$ylK?g$){EDHn48e;WgC3~{;PvlRP2cJyfD%n*?D*3K{xa=J-WA_w&WUNMrrLCh0
zsm?OVy<zK^sKm*$hhBJeZ`<RSOF`nLP4@)Dj|J?+fje}Xy0CU0rP`P*GHy(1a`^=G
zbTYC?cLgsAH;o26?10osocl~`O;8dM=7MH^wqUVX&@|1t4cN!x>&{8^o_yvv*oEzm
z9)iJ6Fm5pr!*&t4+cnpQ$EN;~W5ul9pA5!i66)VHl;e`JENGgxGfmhie6w6LUo5d!
za?Dj#OHs%sKHEQ_C<N>plcBUBD{_K@lhaeIbri*j>0~OtQG?AEC4N>q9}|>R_Q(p2
z&PsGU!e}wWPNx&eEf@mcwm56iW`qxp$>e})y+j+u&0C)9H|{WB&bfQ<10Fwl%s1bB
z!#CeP<mrnOPS57d=WBurRF%UR$aBNq9yATOvzBr^A~S(;uE=vGW?~)3HUuhfX<duP
z(AF*5D6U`M<7~BNT{k>=@(AypIAsq;zcZa8n}7~PilZ0U=UKMUMTmxh56*^ZF`1X*
zm~NX5H|B#+;zC7YI<km53g>uRiKLux@7}xIx^;`9D9AGLM||}7A;10XGk){yzvB0w
z{f0+hf6cmXSgr#_7AOlvkt>vPoJ*%dzh<2ccxP#yj%+*w;TqS2jy_XLZveUe-&T-`
zXe&FxUEg>9T)zx~TgTWXFm=hs_I=xi=<e2oM+HT-vMd&Bsd`h6Vtzr3i;2a(vdARz
z5j^|YAF`&5F$C|TZ=;m>Q^rZ1Tjwzbin8c7$fj-3T2ZYks`VPXZdp~8)RxV2R#nA#
zwTdQL39hPb(?J-cHCbi^+$q(ut}5<BZ6n9}=nvXjOI6o2u`0xRwW4WT>b8kYT0>*&
z&h(o4N=JrHWd@TCf;&ts37M4LAXP@v9P!=>$TNTnR855q0k0%oOyi)KjtQ#7dBuFb
z;K8@zicwb;!QEmu8+Vb*N`cn{Hdm$6j^eNFy#pUPWYv4KZY+pE$KpwSUf=B$YWSNP
zC=;sxJk23B_`91<^!M0AdfVKO+W5O)gQ~mwu-<J*WB>sd5^=x)tfb=yawr1O?j5)Q
ztX)8M>^d&qQb27;?B$>z5|N=or?duzxb3HBk_U;`O84Cj>@EY_SNUGtVft}Z490SJ
z9KH***}y0&^?K-CHFsbad6#WRuFZYFp9E@fS~ox)cD!nPL&i>uV~1enMmk2cg&uV)
zqL!;e2kJm~3_&+JgOA@!t}R6@jijWvyD}{C<$Q3Fb&<3wnS{-D?2|lS%tJNM>>&<d
z%mJn#GK#75;<l<bHUG3}4cUQhN_5v|B8zY{?LuF^ByC)vbcO^W2y0Z=HOtkS#e6|i
zH!PPcJWdimJ5OCV<avSjfofe*H!Uw;&N14ENxBUj><i;|ba;f;M$958R;x8HUcBV=
z<dpe*fz}z5@gDp82h3*sok^!07gChm6ez8+-pdAVazeCanj(&`lG>KmwV*PrvnZ31
zsT@7cVnnYL!^(>Z<GmI8^^&7&*SLM>4tL&u$is(E`1}uF@b%Z<vRK!wR-Vx)C(lbx
zPF6CmAp<o+=|Gl2QED)<8QIq2fYC&v3(LG<y=cfX!+38@F`4r1!xIh;57J18sYXR`
z6G6Ww;C1NCDBEqjCTXXo(dVq&_V1Dc@4EYL__Jrny!5RPVy073xWzFKK^Sl6#AmO_
zB@!6|v)PQ}<0G<6Gu_*Z<=+Blr*lqTzT&igL2aS+8tW8w?I`mkjdTQazFRwDl9dhK
zdhu@raknUjiG~M?TQU=Gnym){F^I3}uqs0OxuHD|y8I#qDn0KuL-#|Qc!u&hZI}v;
zQJVGAaeA^~y{;&VlAxqoj0+HyC1{I8nM%bLQL?kHf5eAO<8i`Z<wn4r^C%-@Z1RjU
z0%+%*uy@Abf@i&4QB{?ILZz867Zl})vMd>o#!UC7_yOq3vW&72A6kVHCf1Ahq*e-V
z1yrWF*?Pxvxny0fSyz<=+9*X8&9Kf1tCxTvF>$goM*Qfq3|)w6SSurm%Y#Y=izs`O
zYzJjLSy2#_ruCk6y+UOrL1*}2s2j^cIYp^4sx{PY&8wHMux-m~vBWklZ{K^1z3G&)
zC?vI3^#4Xhwk4XTWsXP`a9ZOkf}Qh>o9X~clI6Y0wg+9Zmiq~_DI)+CI%co~kb{Xf
z^uzAI-OYq;-al;o>?iDy$!6m_d408U)oox0oZH(H(Qy&D+fZ786igXNdw2ib>1#!Z
z>5<~TIFP5IKd1wOs(Y7<-8s^8Ts7rUlCMH$&@p03gr$Hjm8$+I1dyi5EbzJmcKuMK
zb)Uu{SspHXt}Rfv&G31W0KC$>!5ag+ez3wo%aQ5p2ioqyB;%8C+?-S-1grw2$icw`
z2pz>KEQ9z|HR!D6`i*_=y>pwxqXTiCr)O5ux&1~O?>{WSO!E0lqcI41AFu{PAa|Mh
zwfihh=povV0;M;3#n0kGlRgEnG}bzvJblK~r_V8n5N;Z&<{`CV4NcwBG?w|<0%slN
zXiRI@OeRviJkJfY*}g=`Db157PdPa`r6?v$CKK-5eTUiX0Baq~#gcCyJfx~5x^*%s
zxOV+Iw{E@7{=qR>R*GrFi=%U%8GPG<7gLO+Z}1WTU}9dc;NwP=7H^Q6m&KUYS@Pi7
z-#?&fS~62m76rF%UgKnb%IWDD#c0aO$(qx%imGaHZD8Garjv}k9FgTYXQwa0TZ|UQ
zs~F|z5YQ@Pe}9C@4C86ZqH36o8;*{yMSyBzqHJN^h*7OcYZJg9HrCHJdz5DAHtTfG
zl%a&l7z#);*krm*ASq}eX3&SI9ZDsT<X+%?plKRObF*A98VS>y=LM7Tl)c%Ez5RVA
zvl&@A#s$r)_DpNABP}c4BWA-OW?iXx5&^BJ^B=QbM<$-AY^iSPq!pip*>KIJ5@Tr%
zd+%vuWaLHQYq+k4eUrfSX8ErEoF5X*UD#q!4O*d82F7uAw&3yO=X~+i1Ma?c%ye2%
zjz<)u0<9ITYr%(@c4SkIWLW{IE*Pk8Du6Cuq%#KREXpV_u&4GY#_{@P$)cPypU-2(
zhlXmopgnoj^@WMS7_d1WY2I~}K9r=P8W#oLx-Po3b(Ym?MOD|-(RWU39k;RA-wKq`
z<T^v=nfTB}`J_z-%Am9)uJ00JOC}VwQe@JdC>5|F;9QMIW1Z{;CeOi?IB#f!$9Y9v
zS2%AtKE8$m=4YpT^Uc>hdi0R_{8R4Sxy|v>p_I9tj>%%VSnsT)jfnwu#u!nWqWSoT
zNND|<3G^dh|GPJpKn%M}@})KQy>deiw#s2ogdLx;<NNNn9wl9VyDKn@{i7u$qdS&R
zCGa+UmI(TkOxoVdDkO`IkrY+XDFgkp|1fEo#NsLLQdezn%-E@TqY!#0U}=Gb^<=Y9
zXf2o5F`1{Z6meS!@x_C}<1qx3N(flV=L0iXz@v@h_}Y|@K7N~zKYfSmH;yRFvJ+TZ
zt4?$3{p)l$gxec%uOo#RG1h1A-ZnGsGRI!fOXvRX7g4|A`DhYIAfU`f6nRNq)mSGH
z$JRq@#jh^U6ivNGMF1#)5>rf;lVz64c#3T;2m3Q-d;0>MbVgZ}+`IQ4$JcH#nM~Q+
z+ha1Bg3<WaQpb$<r)MWT_~!R~`Ni*e@Xa^8^X`Y-dh0GqX-t;Uwvz2#M_(;xYcat(
zN2V3d)+nG_ubI!!sGFL_az#_OSm(qMd%d7;SIifuJb(V2(1;(ywc|tf4<^iyX0#4a
zn&P41bl%c<&wO38Sk_oy)3!O|aZXdU7@c9gqsU8~x165L8I=<z(=qFH!`XbnWU|N2
z8*ia?hS6F;sY=>nZ!#lDG;>nEQX47^{hT?k;hvitLrst#xC}Sg<U7{Q#2{WB5<2NR
zr}ogaESGEMb770>s%E)d^6J$IMNzO^E;v0oWwD%7)fEm1T2Wi5nm}t6&Kg>837L-@
zs%Xn?E5Q?52aFL=>zuHm!nj>cs3y&{XYaW(_)5VO>72BoBtX3@0r`0sh#zMY4MMG;
zO~%=2&7;T9IG9a%_no_ZaQ_}fk>Na?oGr++oYBaTX9mnB!{2X^Au^?&x}mGD+aKF2
znd9-8(MaS%rlq<LT9L0Ss=AV@72bDIs?J%M%bZOC9*MCM<-m9Bk5P(Io_DEAx}X|s
zyL2U5YfNT31=JWrUKC`xM7d@~-X$8A%}>hCsR-OO0qs?+$*f}9D!O+x+TenXCgLcI
z1JqT+a@FwY@k^dOIpg~EyX+ra=j`lESivk~G?~y;6`y_f882Qu<L1p9yz|aIilU&<
zIYuklq&+BYFeYO(Dg?mB)H258Ty68+Vo<JdG{a@1IzsJxwKtjL+se<bzZ1zYTuVbB
zx62Z7#k+4l_uzW$_?;~gbIyDU8~f+ZJH7#^8y>He5`u65bh|7$>0qrR0PN-3Fe&-)
zdfT3QKSb-W#~Zxk8|JY~z6_hbV)A7W#>Rz?Sxt0sa^iO(3Gn`3%D0pvbYua<YbD^N
zJlX_~j%R%G={<h>v-i03_I0KcVJ%Vy{17B3(ooA!Xnfc~{e<qBL9K}XF@ui$&TTm8
zJ(}-@eA1}p9YX+TW%J7Og6U++*{cPv5ud`Qwh*LXcTtogq)N$ocQ!vG&kBq-w4vqt
z^_%2*PLUU6nN)9>&5jt2W@wesww|VLKq;(kXxavoX^xK%+28w!dR?(vw^&>8<l)y$
zX9rC8_V7*;s+yEg)Ol=EW1Yn*#cC<3P0r3vdG_KNRa2udj7C$+(FC1mj3y&co`dRu
z>14`_N6&fk^f6DKo}e<%Xgb1}jQxWN`}-v?URJ0$XBy{Ot}JEYnT#itWf9YaXzHqE
z-L5c{C<FvmO-1YB&b^Pge(eS!D7+IERGh87n5}@6mqVB{-HG~PP17_l<Yzz)YQ+xC
ztEmkwAsf{u=vkbh0YX^ZsOSU-K@!?}=Lt$!ufxNm=-Vi@YPHsVv|_bpxn8iUR#@**
zS&ny()<W9`d~OIq<CKqDXmU$%y?>GqG4fNc+ca$l26nYTZ;DRA6Fob9bzv+0F=Xo5
z03014Q5X5>T`e`kn>LkRGRd+4i$%rr7pFXV@{D!WFdmmErSZPSIswI6d6ep(ebWk{
z@Lv9wx~M&<5-__B1feJbZsfY47GsQnt60GyF?gxcLt@IUwS>0qu9N(9i3&wq6dik&
ze8t4}LMOu1z<*MJT_2fiik=OOFNjh7My!G)V^^_OYhpw;TG~fczfPK}=Je!@Z@&H-
zAD%G3TXXAH!)*VMqobqFI-r!IZClRfb5P<opit}|%$STvltoUKmr-$3I4c{n)_wJg
zs{^)e<?ae+sLeUO{r6RY+C>cYuImp!+qN8CWNo^LtsCCsviBQ&a^bO;z4P{Wo5u`T
zyRClT^=xm_r^9)h2$poK0VaoJwTOgM3Zv+h(C%8{s#D|5n(BChE4McWcH%TDX4Kxy
zDN;{q0ahs62Py=EB10jYut#bM>JY>akEG?N2bT>;YeR63fF;Y|+VvSf{n>p!{powW
zb@!OPy|NGD0E|fyr~P)hJz2Lu8@`#+A@QzV#`0v=7P%wunw(BU;$0O({&mNtH`S43
znoS!SNR*;dhP~N7d$R*hUp}I4E$da&_llNermIy;S(Z$vdyI-PuiPnlE_wfqF&rNs
z3tW!TlzCo==`GfBtecu=&!012oZ-Eb@?gGYROS@gNnJ`)pn{xVt<J~{jHeU4S7?=^
z0dK7U!=}c^pR45x?^>=M9}yH}SwT@w8BZnzEk*B@3bd|ay<YO-(F^|Y`S1A6Z~uzN
zPaj}I%VfGwUKETbBTRNmhGe6+w!&KReO^}$d*hs@X;?2U%XN*)z$mEemUY!q)t>RR
z;OOugds9jJY@M(gJ_LgEogW#IsgtES5Qb7-MRIZ>41kPhA2^Z5blYU1h+YU`ASgm#
zQ@LYNk}OE>2qEC3M{$VhDT+K}RFvc~t&0zm1h%matJQ*l=ip$*(cvMJ$pn*$<9ZvQ
zwZhneL1zK49hs3Dufc<JRhKTOTOTT5t>h?^$oDA3AQ@r1esj@}(wH=-MLx)@_>Jk`
z^{h&3N{&wE@WfySl8<Ww399q!>KvUr^*U}!;mx)7i&~Uwoo`s0FPRY~;^oUTzWVwB
zPoKSHG8r??#sYk_<kL*1O*;ThN-2pFO$==aWG0n(i^{_m@Q<%YmSs}qC4${7&nZgr
zw;bk_Ynq1B(^KlE!L}_`U1MXpuN3gpRf#}=mrr01x+KCnuGLh*MJnDHbS9?WESIu|
zI?v952b};RNm@qJD~e9!4Q5ethv`hEN>Mj8C#R=;`|Y<ZmY%cIHK(Tw-n;)Hv)LX>
z<-$hB#K+D$PEJmE_39N0&8?d^I6gk&#`S9ydD-RXtE!gF7>q73I>hM9%NVINL$5xs
zw!v2VY?mL~Rog|NY}Y+^0kw<I4@vZ;RfUXNcV#xH#%-2-^C=ki)As$-&x-)w#o#XN
z?EYQ*`Rn>=`~1Yf^@3(N8<I=TrW@IiwAzg!e;D)Y#<O-7$E%<7X232OVU9v2A;3N-
z1P!`H=aLr-ouRTQ(40b3lIbbd)wE#^I6MxACL`eR!Wck|qE(X6*okww$r|q5IpP<8
z{66o0@HTJXyTx=e0iY!TNKe}a%CgwmCEZ_J%-yhsq+P0^qo@Rf2ufGo_0oXLAu%FM
z(U}v@&BVC74L3xH$19E2IYw(nqZ!lbKFd|jV(G9>iH{Qj?IdE+7=!XHhqHoeRpYJ0
z6mtLZc+B-1$0!s<kz<UeuGbJe>$5pufAfHcPoJ?|S3G$9h`h+KZ9`EQZr{AY!DPmC
zG~(LvA%_PCq6K5Its(e=Ku%DapaScr!nQT*^@6spXlik6Zksk%L-1sIfmbh>9n2U_
zr|3)~=d>|6onxjGuHU`G2dbr*l&n^BP&rzc%%+EE{hZt=ymQjO4G;?Qd`w;%909+U
zbUS#1Hvx~uDvtwNr@4RsV{%gvyu;cy`Uaja1D0ls;O^vi;X6PU;y5Y=p-cL@Ij6RT
zQcq6AxHhE+{upviuL7a-RSXcKf7?K3Ec1+UQKC#1rErPGle?TO*O)@#e9g(LJ6t<D
zBF{8=kyEb}jZ@SPtny$S5qoszhoD3ug_v7e5EgAy;cQJnl07FSN-6_LqFrrd-f1l?
zqc)%oQCSOw;K-E06FUFjH2sqTY!IbJSV%Gbh9JSo&IzM0>}ZIUC>%y?dp&6K9<O|t
z$mWM=Lx*waJl<K3kM>!w7QB3U%K!N9{u7h&h=1@8{vPkY{~mYl-s1Z45!JdTScyD!
zjl<z^#t{lfHkpv=3=>mB1!E+t-$^ZC6~JKRJ+%bXIPXv<Cz}>1W6)VfkxF^B;^vT&
z`Zi6&WN)tndCTRJ^?JSWebY3;(m6+-=h(r#%?e3+yy1$H8AhYgM$SEDOdR{|%{#=o
zlpM=7DyE(8rDqy988ho0MCM)0zJ}lfZPTz^R;()c{g+?x?c*oBczHt8)LgrEjoEC*
zXf%@Q8&;C0uU0Gm<A3-Sw{PF#&YidL9&X&YLGT7=#YeR$43p^?tzo@h(8tDZ@6w&x
z@q^8&|4+4x7(X9Wv{KM9;@uv{AUI<H$iHjb_LIHnnDajqu=|0jumM}Q`4Br;Krewf
zUIlbiy!C)U@@jCIAA_-rkNGAF!5aa)Via(`0o5SE5=JJ4_gG8x(NU2)RMFHBa4tvt
zOv;6Um|tVCMH;ivl~$4uoU2i~<@Q_0{Oso+^2sOfar>?7%w|(07n|!!4nYc*2r)r%
z>IjIsV*G5=Of>v^yYfWr&T}Aedv1yec;IlKcC76ahJzAWKCN`DjR@9>CO)3-aW-E{
z>8>Ef$i0nPZ!VL4wl|}7mZ#63;c*;{_fSePolH48IwBAxYR`M>y2e?{um1h7`1Rj>
z#>sraXgXunR6O|hTdZvumm}`IeT$!c@(Z$K3AkwLikr7?Mr9_@+LmlIBFhYI<6=~5
z!)hftwiah|<|k)>;@R_8*jQO1FAL5VbF#7kqnXZTj3y(FkB=!$Mr&Kf<1xoaM`)#(
zFW2Cq92I1_qHUd&AwuDNpsHHR;sEdzMaI!l0cgC>P%&rMbYHPvwUnbHKK|$vGLuoQ
zD^OAkbLitX^dS)~F6$V!U5xLx{yx-_&jGuQg{attJ2Zffii4LqVU*O%?M&VQ7FA8#
zuFy*Sobod3H8~(NhH*J&RE}aKr3R(3&QUiljSV=}^O3>#6L|=@64TbU4ON^w7YgLT
z$3r)VjzNnlF17_y@+Iqd0=3Cf4f{3yGZ7-Byl>E(fzS8#cW)X5cy;%d(rKC2)QzQS
z1Clkv^EbcwJ&VPPd+&VC#~**hfAJ6h9<#}mvM9-nyi3Y)=$!A8?kc4)Iztlm+Ad$7
z)}Eo+UTcjq$-JEy@4lKt+caJIEgyp)_MOK%M{6Z=bhTd5wh`F14fVRBu4~$+!PYe@
z`IpAEIvG!*ADfYNr&Y`wSx6s8Ht&4$?&!CwBoQ_Qv`>I8c{&dQhY*HP<MBat0X@zu
z)>TDaHxh*(Ct6)s{O%9G<LT3<+`oUH_uhMt_uqe?y}doOHeC?MdcEewix;d`YtBv=
zeE8vq+`Ms<{rwr!=`4<eiwV0mJ;Sm+f4<)iy^E#0h^6{2*S_qYyGV;&4B91k=(elP
zW553z+vR7sEju?s;O6ge@q4~m8=^5lB~>~ilk_|&DSTi&Mv21qUB~Uawq)h|(SY4M
z?;fDcl!p&Jj~~56DTmSqsp}d|u}FC_2T^btw3mX_%13{U5Fl7o5ECuK!BhdIEIO;X
ze)E8z{PaGb{^SF0-a2AD9(95~%MAEb!72IUxGt2w?+6hnLqNB^nZu@rKK%T$1+h~T
z+Odh9YawpO$+<}(ncB6mQD=*WBF`D+;ullbHfmKKlNAK5c=GHyAp~L#aEviryLOH7
zcubM!WX2G*rm8BQzB&=}>vF-L{OO<aXMgec`9J>m|2^J2Zr!-f|NQ^@58QkE9v^)0
z0k57vqpcfWdW+JEgTq5clW7kKg2OwDZ!N8DJ3)H#;w8Z&k*(!~(Rjl17ccq!U;mnK
zA3lK0aBy^pGMaKU;_jV0+`D_5quCx+Rg>qkc@&MMww|)cF-Dl6EX#q6pe)9S0+8n!
zCNp>!Q-df?+qM*Cfi?xxNzNyq{E}O@Z+8h;eQ@&p184DFV|6vNNr=qTd4H7GnN5;V
z@*%`>T0NVqQ0O53q9Fv@w&irb<n(O8qeqVkA#nH3U2fjI#%PpB6PNfuWhP@X9y6JY
zsisrvRfS!ysT)t-SiCo)h(zX1o<)+!CdN4hKghhRvDWQ00R+Oie%c0S-TI544~6&e
z@9pnXjnU!f(JH=k4gN5KKJ|Z0=G*0;i_f7HRkfzI0=}{$W3jCG{EG*?I$3afwqP=z
zaR1$V93Sp8os1ce$LOfs24c=nA6c!C*iwTwA(n1SCTsypS|g<vfe;iXFDL^dH=^m+
zMY~Y&0qdnIjPqiuFN>10n&7O(Spk04YE50OSyyXTXJ=jI2vC%x5u@>lysH*ry1CW)
z(&-*-Z?n|G@+6i!g8Vo?b;@Jf<&Kb_A#&Iu$hcQ^!+KS*UW@BRVv~FjCG}<d0%f3z
z5yc;T@ByPz*hJU1j7MWy+p=7)c-THB&r5vVC$1eI0~%+o>=j1sChNMH5FhNd)2b&E
z(kb%yraXBwaCOPucFgsQOt2R{N3ui=&p%IWT>cEZ+xh!n70_;jx?W*OgLFAiev|c|
zO6?6Hqg-y2s0oNtrjQgm*+4M~i2%6rUB6#bZ*ZqK0(O7zAAF3p4pht2CogDyJIFWJ
z*LCgEGpcA#2?~Og7t=xGf+cBq3PYbdMYJfT9J2$%N1xo`!;jzQ*6m{^lM-!2SR^NH
zaYYI-?S@49sT}}q8`upK5U0g(&PQtN8oMUrkgeH1XS<zSh{^1vM94{63C+#hce#K6
zLtZ?4$ZQ<&zGbyuQ&$!r6s>hUdHNiaW#mPH8YEP{dGls0<5TvT-<4v$UUTcMw<sr5
z-g)<3_V@SMpUwDux#F`IpK<rjO+L7PpFjDNU-GMe|3CB1*I#jXv`-P8tDB~w^d4s|
zR$FY-(9~<zs}*Nwrz{o=UY?xbtmoZ#@8i6}>73vF;Y(bAr!N-hEbz{I_xZ&i{}I3Y
z-S7D9x4+|y-~X0h{PYuM(=kn5Q{*Li-cmIcWtlS`8P3iuq(zlU1nc~4&e7p5Cewn|
z;)JvHDaCli-fWE48A=ssUGj@x{1KyaM88;jw(cBLx06-6U_65Wn{8n_WSOqgWS<mZ
z7p?M=H%VzSGMx9kIyqrkJ3jmTOaAt6KcmbFKKkeb-hb~MZr!}bY&IfAryC`Hm_<=Y
ziK?=|RTZrZ!q|mq0t|jIpH(l!JQ3unm^-nSTA<eLBs6-GdXoh|--r4_Vcvtk{yN+F
z{<K6J^{yH_0Q3*V*r}+<3%Jw$2TD;lt@smaDPPrCI6XV%)vGzLPEOD&W4&B)@9np^
zapM|q-M-xwrVk-tV;$iHazvvuXagAn^d(x^6L5qWbe8hc8IwuEU!8%L^dcdSg(A>;
zkB>fd&P$2HV%+qV8N8!j3#ePKmgG_Kv(eA47?q616Jft%aK?~9+}`-p{P&6KrB@b0
z2=Of=uuD~oV)AD3MU=fEZQ0HT0e0((^}42U{dS(LJFRbd{PZ!WXQ$Z4GMP-capMM~
z(U{gbrn4zjiuJnU<m42kz99sJ`8OGp$zGga(DR8qT0)dp+DtcDt*fwgKV;i!S=a&c
zHr9w}r5oq%Qp|RlP`CHJ%bsU=&YSc>fBbG>H+;U~?{6jtdR8v6Vj&?mq>FqIYnn#N
zm3Ed6mDcZ%pL(f(|Esq*0(KvKbcbh8Pf5kFeXyc&DibyTsEsBD$|$^xTB*)ZP8@t>
zbBV$s{!Or?N(_6mf{#Cai=X}ceeS(;i~aqHXjcOt4oCtx3BTxvqxEpU%F<PvlP%>h
zx@<sRw@L4!krNvkxO}k*v)R+D0Z>LF;%?r$%P;=;&-uUo>;D1kYbN6fD1)^NZ0lIp
zlE_vCsOpBJgNeLlfa!FKF^2JY%y=wyQvvq(4>&%$&V0F|YAjElKIMb=->1kj=JOMR
z6-{@3HkVv?cixI-&ZC%)L3Qj=VKTgHv94vgTri){SuU51#$%3;uCcCaZr{1bPk!-B
zzJ2hJyeRnQ!9(8v-~;w&``mx;KF3GL{JVem|K!0}zvs7q`x&2n@*$%<LmNezmn_1X
zJX4G(1;$i3-=ejc+MSAJkw8F9Sz;Z~G&HScwW_dI^VaQmxPI*xS(bOHJwTlkcIU>3
zJ_X6}yM8W&Zj-sng5CufQ`3M~!VJaa`?5Z~CMYFN;qy6PeEXRH^qb%E^2I9-4i5PI
z4_~la%$dJ)hqrEDXFSRAt;d*5Hh&$fv=~EB3hM)n1?v=97Wb?qp9W@^xD5-ybB=1g
zrmh=FKAk0{ZIGg+D^a$&>2EMOzE{SE(Z!qfV0WyMZKcZxsY*C#-%qGtJYFh#g{f!)
zb>7q1hH^B*2T$;h)3X)hNy&6?iq?w7s^$Omul_GQfA*A*Kf2GyAKj<O3l5GBDe@w^
z5+G?ggOF|jN@tj8hV?<=oFoEvUfeHqrX}aA4`g|P%0Xor7=uzq0y~f$NCi(56lfor
z#|%@H0{kMoP_5UD)@#O-DMtr~ELRJvs-|sP&_pw`!DOPScX?GKH8K25s*||^<DAAW
zC8XDAQnYk*?$ll^EgtH+5(d!ZF{MQ4(vh^S+h9{5CV5r{=N;>6&9DFV*F1mmoR2>I
zm=8bvkWWATlzJ7}|9nJK)?!UKIpxbQzvl7dC;a?pf5i3cN0j3cO|wX&zHz7jzi28t
za_k@`w%5_$wdrxYC18^t>uMI2?{FUiK7xJ0H-dAl3m8STq=@3Zo$e%E;m2r0CH$)G
z&3x?2jOWizc=G5a!5g#=IM)zVE`mTS0dwM+9W_pc30VYo8PMm1Q9(z>P9&aEFq@6|
z_~U#0?C1A+``%j|9L%Kjm`Xz4&~3qy2^4cppO#@mo4AB|dmT0YI#37a3%hUVZaeO$
zRNOSwfXY=$z;7m?^y5$eh|fR&0}sFX8$#t+ENaf?EiNXLbpf1h8I22U)3RQ#qOXZa
zT#07O$&5tP`4Dg(Mtgf?*$59D92{`}z4x$ng==dvt$6<YIa+D<_xG9YNkU`ieFWyh
zgo@*}#aqj2xx_^$?8C!7_Vx}LThIRCA!ny2ymRj!|KsCF?C(u63T_-<gVup{+`4uh
znkAz$2QO^IY;VeBI%T<DL)&n;KjuksMpb!?Q7B2{MccI0bw!>X(Pjnfb&FkiS{KNS
z!MT9ZBWAOsxWNRn+~Az*(n;)MAC&6zEN;(@?){xl2yA0)hYshv+Ag54A`=u<mRN>I
z){Ton3|c7;_V@V7rytQamQh(!uWOz@eah*{OJ2OV!MpFg&1^bBDNSZFP{gzs4j-U#
zlFzPgJz3u8(oX=dH#@|Dj+kDgZW`*QmgLGIu4yk3v?26K!G{2ByBwcCw<>lkO+$8I
z+t)6!tqjiFzV{E+GOfGB(mnX>Yq~n;y9n0Cda~A&XOfgRFDLx%Z$IbZ!-xFlZ-2w*
zzyFfE_wI6Zbi~1I#$-0*a5iHyoiLeBm`-HOv{o@GaYmjO(ML|!m?ubf0B^A_n(mb$
z%S#AaB864V5oZi_RY}Sk@32ivMhg29ujtGaj7F1K4|dIXI_-ig8rvWhfRY(i<Zeo3
z<D68oQSp240!>qs=v%FGtZTadGFgcsCwPqyV!iP`6V^^EyfTu&I#wY<Wt`0yynK1W
z$;oLqYcg#>YZ@!5TLj<<;C}w{8TK>Fvu98F_S<jy2Y>&+Ku5)0o1FcFL#*}Gb<1ym
z_j{VgasU3iy!F;iZr(WH^z3El<CtaQqnGX>^{AUz_Ak!I&L{7@43P6Sx75p6t;+z@
zMIdX}XO|s&kxyHCgLvZbVv(qf>YI{e*4;JOjNlGK0RlVCh3Czy=b!5`vud*FZ7Y50
z+>7MErE)<@&_nQ{D?1v306Xu+EyV}%r_J+%EHB9OvMb%EqYteM9!Y=Iq0S8-?#rGd
zz4uVb-AIiUVuE<Y9PGT^d7WMDjey<Bt2ux8`~k~_L#q;{0^Y9)fH8##eG`F})`V0?
z9071rnLsE5jif;_QIKdYl@IQ{bDf|3>^|?l`!)v$`{GMwOb6^X=wY>KW(__~kin)A
z!cL%j6|H2~qIhlFqPU?i%hLM`dAh2l5#agY#pg<A+`RoZfA;tO5l^3f%ZnF}vDQm=
zcI(KE>gt<j(Xli5z-Tn?qQnx=%`-8}`4I4EoDUS`gmOHQa#nz?E9NIBRI3Gz4UkJF
z?V`vwCZ=~Dv;!qRDs5BK#{5~v7{-%5Zry&15ENNaQr8t9ee^!puOIRF@e`Jd1=H!2
z<=F|YZYZ>3I+-vYN$Q00Xv}o7hpJW-c}CT&*_(~Yb4A+*Pyu72dy!EzO^eow@n}qL
z49k^cy%we^%Z0Hki*W=yI+lOhTs*^hkUsZmK(^Y3!#<AWqcNNxyUeoNKR3?lm}HI8
zKC;)cb~G6}Xl~s);O+O`qZpOg)?(Yhg9i`!+u-=a7r&=%*UV-!GLxf%qAUtVWx;4%
zpq0iqmTJC~WW}v#<U$9`RMJdS8AQD~&u!b%)HQ7z8M}~xUx?{E+(r(z-J6g`>(>kM
z#M-64O00Sp!Q36U$<qDMZF7v031oAVNuT<FFl#|!V#=bbT4CFUx~XUz$JY-Yu(!9z
z-gJ-YY>&hJeWue1d;2qPT)Re5mK0^dcvLc)j3tdomp4oN&$KZVCP!yk=W{2e!3>RS
zqudjPFE2)z%wxPKHx}0w;tZ1CkQQxBtRWp!n&d5c(UP|{gCe9dCIb+lVxjXWb&--1
zeVInB1BM*Q-meK>(ZpzKNv+fIdVnD-#-b#8VMUg!H7{SD@alBIVkvcOvn-4K?7Qb0
z%5>{2XY(_hm88SR$Je-d;}!=82h65>anx+g;Rml?zCx+MYB?wPx1;-p%(b?)WLe%Z
zL?UDK#wN|cKk#)7&3B!j+d%D2z}@!!FEY(`$3U0fYuB~Z;4_gKy+X=hZk)U5;2C!P
zep&0~<?vdY*V!docbT8J-)Ts|gv53e72Hl9Mo=f@GP%hyCXXq9vR>8=y6rM6&Shrk
z*Y=6pJ+`lFZv^ZfeDj1SkIryyDLgz<mlMRv-P#tVbFvsgo}6;UZ%&d*irF>*8JaA9
z9>{aWowu*^lb_z_-FI(scz8&j7li0YuAs|<CiMUlhU%FT$={aBTSZ2KHz6?lwrit?
z^UTY+&HV~4j>+89AE;t|P)C-RXqEHxpZ_T+%fJ5D{{vrq{y8pqw9$BLDf3JKp0(uT
zQc7rzN0QGjNZOV>%R7H7WkkcbF?w0D)H`sFcD=$`$?qhKRe(@g2m>3E4_`OI0&P=6
z2#hBaviVtey}E6gjAxWZiFK{y(6cSmQAyJ@fMlD0`SK;za*xqSt~V)193LH#=Q*Ch
z;lVz2yJ9ja$a9U!M6hOgjEz)?W508ZMkRSZCRj_6DVDY4?%mt`#b5kGjLy;8bf(xL
zs2eWyA?WHiRW^?6nWH4k`uT8?Fu6>dj2DS>pjSr345(2iV_Z%+Jif`{(LSSbj%x!=
zZJCw@r!Qaf?YCc%OkY}yKjmmN;?A8rOlMQt+Vbd|M||`7=Q!Wev<_><MT^j<4B337
zl;$ANS}Qd}+lIW%JJIS>MT{-2bh|I&)b14ei@=!l>+tvUT7BA3wl(%2x($P`x<q#A
zIk$YNLWo|$fsTGxS|X88PnRs0U-0#}54wcDqtS@zbi#NtVsE;~TW`I^WHMoIZ^Hh;
zKG&{YV{dP-i$YCGn#oK|2pq_bMAiDp>=G>=0r5c`6;e{n1W(8uHkOSG-h+!ZxsB;6
zQb^9j7L&D@Ebsi~I{D{=R6ub~W=8CrG<8K`mkgx>?s7a{@-w<X>l`8In3GZV0`Zwu
zm`s$H);6rxHP2tY;_0&&oSd#$Rbq<GO-6041cnU9AtY!OXxduL*Vgf?U;Rfu{^%3l
zefK@~_x3QxbTUyX&1$`5_3eU_lb3+yXFvaxER$o})-oQCfq-+Cs%jXIM*th_Ed!sd
z?fv20db$EDcbT>5yMnsy>u=mEvT<S^c>VRdH6CR{zy1=d*E3ARbG|c}xkB3o;&#b}
zU5YmJDqGqYBp8e?cc>hu;ylf}NOYyMUe0wtV^0kBksn${bzu66?Tvul!$;4V&nqB@
z&=9<e30FN@IfCvBf{V#CX<0%CNWC*d3Ncqm^qJ9$TQ`pQ$tNFh|NdQ$j%E~vBx6mR
zUN>>npdx_VtO(FEcBwPALES(!T$zB|vB{j<ye{wXVFiN@Jah`6Jfw4;%w$;Wq!ieA
zk6--qFL-%6$9cnpufD)HYeH}s4Vi*GlQLO(7As8{Nl{>o<eoCx$fmC}+GvU_2NnHS
zoP>&vM`LE=am=Nb@{$GGwqmtf5|XJoNE!%hTfmcL8AYBmE=Sb0IM}N|Op5?zZg93`
zZ#u@?g1`i=HTAlJqXUeF)8z?@K$+(pA0MNXVpXjuM<q%riY%kZGqi5;E(r4$Ld<by
zY3rKFsGuxII5($h14d_j`svTPckg}jyri`jWpwu*X*_l@KWUK;b#qANt<G=k+hAt<
znj1gmfX%r@%6-@&pN;DlWl1?2QI<IdkI6N7!~Wic>(`ICd-rWd<C4i_D*kjq{4C#n
z?>(w!O}%c($awbP0rkA4Zi1MjP3X)UL@k_{zp=}~9O#%?Yg>{jR8H%wL8SM-iV)vE
zZrDdD2jnmf%|_(%7D4h|wAiaOO{!%Y87d{psX!+`NYoT!J!CKDm{_i%98DlNY*SM=
zk|Y@F!E+PW7-X4=)gv<ULfj_Gvg{PH(P)IpGKx{bWHM$lopSB?nEm~I4h|2wc6`j?
z;UUw#J@%$kM&mJ6B|dt1k5>-Wws<AatxZmr72Px7EmgIqT31x-HO{ZFP19{^N@-N4
zD5B!rw(Umnk!xB~U`z-KYXfy_Q(P<>jWz|DE>J2X1aU-nP9o}0mn%+API&q16)#U-
z@%ZU8o<2QgweIKEx~?{ipCC-BQZi0yJ~?kWTg^H9{ok@&EqVFs1$DFL?Yr+#jthbh
z%+D6A*K3MA!~2$B|N3vRw&9(3-r?G{V`j4%P1A^CoadBf*_md0hGsYsw}Dgo`|>q$
zS=%NKF8ltXzjrg@0|iWeffU4Z35W!-CUuG|Fn~p!5|}{tkCCAFs_jArddK%7zk<{?
zCXQJ&%O-T9V>wM(lLf})@>8cI-g+=6*kCb#$R?OdonuxnYi|VX4iAqgif<?iM^kx}
zH)ub?drQy`3Q0N{h!L<Fytm+66alT`0;fU~P&8<-*_%wb|L$GhzI&T%*AB@`1BgF|
z)*4Bk0wG9LEkTQkK?%fGI-pH*b_^YhmpVlzq10yWH|`;`rUw7|c24^m9M+c+87z23
z7x8u>gMs+E^wL#D!-PQxN&XuEw4#_C@QYvm1*6fF|H=Ft9(?gT9FD!Q!P}Z5gTfT#
zTIxRL+Mpa{Do1&ZF*&6v#eYYMUt^YKQYW^M9C=17OcueP%L(4GTAxUXFl)&&Be}i2
zg2qwROPm*QXWIr16ec5BOAhQ6xul>D9$jRh6fQGZ7g@LQn7yJv<vHWon6|D_T496u
zQR!T?qC6inE?Y)LMs6Bf4@D+m7Q`g&+J<JmMn|)#cX0Ra+x)Nom;W1%kKPitBX0;Q
zf;xX*;0=EQq)K*)6%er{NqQB@>3!35xbr$nb>l3f93%odYzjUB)4I#fj}kx9EH5d_
z2`Ek7211tMfz~@zmT}|8b$<Gjk8s{G+nbW<fQ#ve#`%;&mo%zoHlAX#g4#jjOB(OU
zaPm&Uq6wHFzK3y(1Z@Z^fNBIVH*=gTWv*(CG74p4&$twk?z-PM`vE{$_EaM@#Ek#v
z*^T6z;$t7D=-8Mq()WN%KJPaewZ*P|*QWLP2>I>QE=igg?`fTrhca=_SV`a!R7O^&
z{B+W9-Z`vm@y_8wAj@)WfTnH<>xyu)02JD2vP?@(!XoSX*BBiGWW*n=$cqU4MY$=<
zg12tJ#cVcXy0^#8n>RT;JYX~$Qvk+6IldN!qzCGTMsnZD<7ec>hzW*r<hsbeG>(!k
z#gZ9?4Hl0aBl-N~pXiB!K;E6Jbcrz|l*;g4v94-fou0B<EtxOQ`1--u%;#sE%~vdz
zHLInijmo+NjHL2rscJ>&6s$BJMw`xhBmQrS$4?(|Hb3R*)2IB!U;G8{-g}4r*^F|O
z(OOAWgM!o3Q~v6&e$8^Z=AC!$asU1I5TJ^TgEmnamwA!qS737w?-jf6Hs^MMv+o1&
zUU$5_o@SVQ`Lgqeq{>e4l<3>;;y+%y%oStTXBUILcwF%Kpu0$NaaRzNYnJ8Y#fZ^p
zjLGuohbhe6(AoxcMW}v^*1Ng<NWkuc5ASgD@|6GS-~F{v_u7kJhcOZzmGoH^ft}G{
zJs>o`3I@=41v)tK<qB{#o^ktl%C&<DWszaDkiIGgRP-weLaYQq%xW&1c3KRZno_^v
zrZ?EKgAdsj-!I&YnpV2&+W_6Blgz5!{kVLQ0VE@Jf_CY!^d2aM4~iStZ?per6LK7X
zJ;|9re@L}>MN?VwOi>hu{n?D!bPwBF3Csx5_f7{?E}%^*E!sK;+e+!5Kn6;9h_<zi
zD6p-gTGup9EkH|WlA0xWh-J8%ritZ}MDrOHQr@ku)-?4RlNriVKwUr)a)nlgCZ0=c
z8BN9%g<+-USnp`s2Inl+d93v)CG4YCirj!}!COZ$F2y;#X{oC<nUT7tS*|F2!N2^k
z|7+fV|09gaqnRNm(4w^K(q&vWE^z>RVYML`dd+<MI_Ivt#pH#=`=7hjP%1`n;E~{$
zEW<j-$>}0i-;m%W1soh6pp~boYqU}{ZB2-kGJ;ol7sTl*noMkf+A3P7y3J83F}w~G
za8WFR<b?ECtgUG36>U>vOcDK0d<T5{_v>pkckdE5-z~l2@OL>pDS>@0)wH(yYlMyS
z)0)?*h84PVX#PHJIF6zFPeA98qey;VCOBMBvC4=9DS3iK@d9|)J!^VSw2WbEhkX^{
ze+W41aXw)E8q?@*oq_L|%@7km=lOv3%L~5w_CW-BC9~N+lgWh9sAN1IvA=)7{{BAW
zamlzWnNG(Nc^l=8MC+?~AK&><1}FDT=G2rHr#G#po>6FB;6jjfT(we5+6J1&aynn~
z?8S2yF`E9-)2B3bP2ILMt!Gfywv(&LH6Zy}4x5Za=;UVza*nkvi{*mgAM%^u{D$B?
zckkZi@Zf;5EQEP<mf$>1)A01^)6QIb=gu9n%wU73uAAt8Hw*#VwMT3+cyBt_F8ciY
zvUeBW7ugBJ5bULWZ0mO*Qls?ieM@Gg<(YQB%Uqa#Xcvm_<#QY5OV2`!JknZ|nVh^R
zDa#Q>UZS!L9j#%iQ~Y`#J^6Dtk+bW0ze{_Q+4W`#ir>45bC#!1pYWT%{sY=HVv=av
zIJkv`-A8|hpasbF4<>A$!;oVsu}oE9kBYHfQ~DOwS}1a1;X=%Xq=$QAKdHoDAYNPb
zc5Ux{m*M>)pPtPDNV8UL9JAdL;JWxa-@heOwqQk#DRQRcV?O-oV^;H*oE}Vg@YUzE
zRnBZWX0cu|+uP^X&0CnLQ8;hAeyE!o)hY@zCDTTLPIP>C-V=i4iSy1;W^Js=S`mWO
z6qPyb@j*bNF=9rRpbbykv}kP@kH*4Cv@J#}d}Ct@7fGtD4V2oDmm}KDQDzzIwWrB4
zf)xhTA?!f#zGFm#4`^l3%F?zW`Zx%vY^`TqRjlipEHC*l|LK3lKlvwrNjV-%Y6%^a
z{i0A9f)PjZi~M1>$KgEtbn(L7J}$5+Y%Vla5<zp^><Iy7h*5~1^+L)*wQY+rn(1`H
zbUY!?GbWP>d6r2eWNW1oL8=sylx!^tW&_aLmZtI8JesrUsy=s$M#?KDCL4u|V7zTx
zw9dPyPE`<wCet+MHUMoTV75WuMjwQ*@$Xz;hTf-kc>MM?f0(7%`^_ZfEre88R7${k
zQm}&5%2j@#nCY(0jrgY7z80&=glS9ukhX0*laAI|=hK?**;+3-w4%oDeJpPlOS**s
zbqpdhMw4ZR$4{OL_zS(sGRq93QOS)PH@JQKHiriX930HJed{J=kx9<OLTb+%BMIZ_
zfSp63`bc3n_?N8nL6iySWzICMV_h{Y=4<AQnyRW<uWMeN&Uyaw1<U1<<!a5@vg*f8
z^~$b{vE*fx_wbyDb&)v@zONM0xfOs9j`{M8-~IkGjFJ3#lj(#k*Hl$Y+ccDUiBg(Z
zuU^H{YYqGR`%I=|lv31n#cVc>?|GStarkVy`}-nRY5VtGZa%wY%~gQhHrssR)294^
zgZ*>kNr#v2CO5qT7_AXQz!&x4ow9S+-`}TQ1Q@qJzv#SFxY&~dC3-?gYc2JdixH!8
zOqLholNzJLMK%nTjWFngi}#32-MN0?_J*d|J>I+jHh=ufpYrnAE550o(6*wL8Epuy
z!)X`&vNb3(&}BS2xM<GG$ed%0V_!Rp>XiEV1L~Kz2nRFFXoL=-19$NP44s+PRwr!(
zyUSYgvFeW%m-W{%b|Grz!+ZXaZTKEq!aoVY5%Ao+eTP5&^S{qGUwp=Twcy6_N30j8
z6sM;I6w~R9ESKR-RTg{@Q%gXjTaU?=lseK%SUM|d6vQm<uy&32HA;DkJjY~#RM1?}
z5>y|UVOufBn7Bz5c^-jXO`a*Vl4mhWK`dTwOfDN#5KTw?^D3!j*;?`C@@c0a0t$zB
z(kIS>vl4OV39Ks7_UqR1;YUB^fBnz?8T*F^;-{FEUD0@rA>h#3Y%sdt#ksc|n;pk)
z3-Zgp+dd}vK+uW5jR6YIVy%}oW-XB6o&4Qe+x=cu6?I+F)J<m&g$V47MrVegpmm<6
z4Ok}uNhlXVOPt#Qf|6WKkut%1-=)LIi&1yo#9j{pRswlx4yS0y*Y&9|guB1*)^x0h
zu|0OL6Xf5w_46gNQlXEEbs<%gaHzOZWGe3i6MUf2ZI>Vztwb@=QfqaTl}H4spfHDv
z<+TzgqPyINMo0fy-9Ojx`9cWT7Mul;Xjc=*ps78p<%(xdzv6Fx{Z))Chfx9Jan9a!
z%4~0+>(`IDapNWj2L~MN?{j>3%w#$u1Ocf^fop^0nQK~X)6%q!C>7RGw+$~(PdHny
zIGZnd`s5`qPv%rrOVe1I*3w!lHJ%wr+w@E4vZO$1yw%vvBL$=d0hOuwkn&VQz*6QV
zT5GDRX1+Y(Z~pezJbV6>v-vAN`S25Fvl)4|Pg7Z%x@Nszb9OdoKA&Su&U^2@!}aUe
zneETey6@u{I<WCP!<K^jhG1jY?{e&U<>mX%;j1W}5EVL=SZ2;?L>B?b%fXreJw?yL
z{Mg{`vUU+`*dd6hk+^~KDGYeVs2q#?!gxeE8e#IH-|uz0rpNbJJJ*JIWczvkF4`Lb
zyBE*S(0ao0(GBk0d6&lzPnow3nHkGsvjL2Q%!4xYcKqJ7DkZ@cC7v;!nR1Nk73-&u
zSw4M2d+i$I@dP~@cc4xsgxqyv@~`WkEp50qW?oMnxvN9d-<`%ELSzLm)~<jKC~O1C
zjTW_;%e1jw*WRhA`9NOe1n-&b9Z=*V*z0OV7Fx1fw^*&0JbC(*JTEAUQJ*HlS)9_8
zg_PLxQL9!(My}JQD*+Gq7T0=OyT<z#iBdKw1=iMh>u|vmw4`=O0U4Re7?mZ<)r$H2
zjOlbr;83*uU@%DDE$f{KVjl%#NYQ<wZR3N~x@=l0_Y-r4t(R*lq;7?C^TE?Lo?sN-
zw%C&ed69AB<{kd!fAin+_PzH&XT%~NC@%m+h2E&vGeF->sO$ptyZui3*}7rg+Xj^b
z?Kgy8#wHV_1fvM^DGuGaWXx9q*UINM0_Sep#Lt{}qKzs6A+-%yiv&;vF~tlZ3!wYL
z;NC}P;-0Z{$uzfvS=$A;IurDc?}z=F{Fk;t@y1w%ftEZF)JaKGO2E$#*1G$-fjrsx
z8V}eIAV`S>gQ2P$SsQsqp~YX+sF-^*l3}fNxJDw;U2udT=1v!4)T$OgyHQb)XSpQp
zZCbjNPtY+c(U=~=Z=f8FJ5!PO9^bYSwyUKNb<72+fTo6~X*oGLL$N^r{t0FATgD^9
zWSntyaL8mb;rRHN<HKXL5e8jy;59UL(*Zl%w5*$&N6((KSXZp7mc_~m!x)5t6Vt6h
zX~;BWnPOeHr1+r#(VVWLs@c2jCq5|wCjoe#T{^sVE!u00Qn&!C^@0bFDo#$GvtF<H
z#m|1htsA$x+Oo2tOGbg;{N^{TRtxUEbB~Wc`XDl_N!CkNhMvw!KZO(+u8E7}%q}zQ
zWtM^OVuHQw?*wji(x27q&$+C1>olDwC%;zC{!pOqs@K_&y-Lzz`7ZaZK^cS58F^MR
z9!)4lqptq&Fuh4yYnv8^{vJOD;nP_xUPC+$+ZzG9|NFoBPaGfIVt!V!UIp+aMo(g<
zzd&Y)#alsVSxlX7z==<n@{T+>GT$;)FjJbT(u{&-{o)lbzWJ8pYqyw8X2JxS5y}k8
zju{BiFGz<JskSNLQ`=4)JZ#&OE@|7T6VoxnCfWQq&)?aiLU-BV4ky4aK-W{kqRC6b
zO+z-CapSFbsMbrG<twfq&Nw@LC2U<LQC!|jg@`bUi4&D*lFA2E2;@Zp#$bvJS7$WK
zGhA!&wq?CuptZs_HCbi^Xf?6?PVnelOq)1)uIYHfdby%rt*Mc$TW7CcksTdTnv5*Z
z2}+R}MbJ_XO&Rf7an{qc4J5|L2kNR46K@C@rCBZ(F~VBP5GkW+oWoo2K{J|8I6OMy
zzy06+cYO58&&kUP-h(MJVTP50pyLM3#+VNm=CJJ=|7|d_4eAp+wKd4witxEPAagWE
zqpH4)l{T6pm;8R@1w}c65IS?43gEmUIB{xs&SBf8i>jQ@=lGZ&BB*F?P_j&|Q&=ni
zbgnuEBF?xtd6Uj0Lc6<%wP+Lb{doyS3L!-e`*ZWRGgX9(*-LqMO$ZWvA@g3$$3yl}
zDYX&x-sOZG^!pDYP(#0Q4}ocXjZM*-HplcY@@xT#cD6y%jSzyQWY^J+A;i3FA#~{`
zq!z2*U{S2~c;Cu#-M_6;c=1@2mY@x>L#88y9So2e1zYS_ms2(R)V0piG_YF1X{I<m
zttg6|$)lHy$5Wy~pi3=cTQSwT=oi~K%jt3{%$y6fT|U)5FQ1P=Gs;77vBxys#_2bb
z0YjeU1Y9>il22WS1P5=V6<r^@I6o!zmCTj1#ey%s`~yXvvtF&ab^R^IlWA1Ke5{Y$
z@c8l5E?DLM{dXzLvU|SGSukL$(t8fwgoeI<7XYY>J{x{6KR54r88Hxl+4RX8{GEW9
z><zo0W0%sly>5r!tN5Z<U4wG1ZB#xS`S`o)2*68y%y#X@vKkagPKkR{%2g*N>{8Zy
zqohP%UX<iTPL}6rlhUzuPtC@pOLA?v-~TA2TymEAk$~O*<Nx~COeSAJ(44$HrKx4Z
z)yjyD6<lOr0^Ud#XQfok;M{_1$wR}4mP4(WnSv69cHk?|i*KKB{l+6Edwb|SBP%jg
zh@eg-`4uEG773qlgO?p{xZ4}?MmJqBIidGaIKTPyIefOy-Mo)%ROy;&(!UANOB#ZS
zk*pxmY$|{$CfvI7F70B*v=r+2+O=!c>q^)$JIqckl|cxxVuyG5rbdrO5In)vEYDuj
zF6J!fr!1>Ad$SqZ$bH+!;~W^B5rU3E4bdFx6)NT=wARwLO~<^9$EElvj7qZ1;9_4G
z#4PAiXkIHmbvF8j32RpqoTI8LTHBK68oz4sL8?mFR#GaACu81w?^FKyzxWsY<mbO6
z&-Z$^DFiV!5F)cHn~GAq!N>O8xahMqmcwMz+bq+@c{>*3aE|KESv@gU2nf(pIix5{
zbY4p35Nl&pq=|m4;w(QJ38Um}!+KrzdCpS)u?QX_VM-f8vC<hSW1WZ$9W{700Kq*I
z<Ps~`nW1;Dvu&kkxFrqyX>;w#xMW$j!L|?ktmAHWFdONd1Ps4VI}iSeOw#~x48dO9
zBz!6XmS9H+SSR_{Ai5-&OcbMdzV}{C<uQpiT0of(0T;ZuJ7CdDvj2y38SNvx7a1jq
zyiJ>>fC`1iXh_?7s*@Usev&%n;ggNdIjK}q+rYY7qxBl+&IlgzjA)vS$-d*QT!`}v
zi54D09|<BC;0a3f{?x}2fSPC;21RQJ*|HJ;K&e`h{5UUWX@Zg{SsSB{A&}`j`j`gV
z%JSghL#*{I7Aq#d+~er*8X-uf4{O&br8zk{?b60vyLOGkL-9jYBp~ET6)2SIRG|1i
z(Mptd_-+A20lxX+rLA8FJ>W@uOltjd=&v18M|FK4=j4z?NRPA=*qv8OckiL+-(y?8
zZvXwdmdx4RJE>$<OMmY^Z=lL-BwSJ5r*c$?Nxz+UvJb^Fb=qWPSwWVUn4*XYyiM#Y
z>8MTur11>>)zg2w-u0rrf0vbG^Tc#q%p&k(0lURqQLSpcZ)mEDL`?_Z5JDhBp~)e7
zHpz2|KngLz<yblPL&Lr*87ocZOB9AoYwCH!(=Q*ew|_uBn=#s(c4jmW;$-eY1!3%z
zmd+fO<VXoAFY~k8*7`g3h221K8_Zqw6Z$FiUE2`c87<~)haxCLP?}7`cyAx=BtO<{
zwvSe@&Xz2fOKjUlFf9VoIx9YA-hi_-)e>JVY3en$T5<OL8LL;%SuV~f_GS=b{wXra
zI+#g5pAf*P43jA`BiZp6%emCB)S6;EVQ(_wY<|X*r%wn<F`kZbHlU408_AlV6$Ls*
zc<MUhE+3-i?`Uc<`?XDjiY8iuj}Fixud=M<PyX~T_-Ft8f6FJIeoB^4u_0D4F-jsx
zL%^^(E{UDG3SGWUx9?({wio2Nd2`A5cIO4v0lE-E><a<V#t3^CTtAmXq=&4tku~c-
zw^kBG2k&FP+|Y{yB^z*S1J=f}IXb1NAUa6M1|`qoqwi95<ZoNrwgr{-eKG)H3G8kI
z-isK~VSlAfA}Lkpn7YtMKJHT7wi(0k(^}htQyc8OQksCqwicHPImF|=Oiu`Lt;C3a
zTuTUj=$Hb$ZILAH#0oNTo;fFJQ+BkdD#rKB*yS59@~(d>)olprr<f6?+*S04Bwz!0
zCqOf?p<N(Mx_EkDC4ev3Ezp*pVes#q<PA>##WTXz5t1u?*sJlzXx7zg*A>tA0g`g1
z-YC1Is9TG&E;4SK#!5AiSEqB@*7KMD=$|l|Oi|iUH7(<E1WM{~|KW?T2}<+lfBxs(
zx$_p@d!~~SD35ck<ism8Nv!T2tJRvakhDN)U56k(=jr$1^Y-I4lt<mnlD`5SYqT-F
zJA4<xRRrtKxwxIhe5nZx4(}1!$6l%2Z0A9p8LYSTj3FvyCjK6x;=B>*xI-57KC3Dc
z0V>4)8N6B)bJ7!G(<r~$37fq(FxnI8a8f?pG&fy<#%sJ&<k^U#l&VcdQBoEo^1MW6
zlEhnUVeLAc5Fbnhu?T7M2>9h)hQ!KH9wo9SeU@CtQXA%n?Br=TUNPm)8;G|z0(RbJ
zD3go-miK7mQM$$X7L#d|%4LI!VZ%i2LIiUlH;O$JhXJNqQKT|s7@F2HGKN=Ap7HRD
zFH!q@WRnr&OiLl`%yi?Z;>ZjA+jruSF-V&2t|*})i<V40iDetMoz%42{QI0KHEdr0
zP1+FDi8)Gqo?_Zm33~7jlNFTX3CcoY0-07!#}mfmF^?ZTqHP-LEQ@th9a)iKl&4v*
zpfyyjq%)}(r<}cd&dHO9jPioL=~OgiFNuE3Legz0CDFDAhld!Wsn!efEJtg2`QkYc
z^Nb<p)F)4$5VXP+83+4^6uOK)4FN+Cwcn^N5-9{9li@a?B;qzRhVf{G){5ygr*V$;
z+A^8#@kf9BC;WH+`~L%X@4bh~3!2tN4GV~V6G=nsek7Jcoge>QV?LZiyUcx80(BRR
z|0c66ntsPbBrp}Pvds=9V3(Me{#kprz(?TcViL}je=!8vMBCQUv@J$SnI*r`M=5_C
z-L*mpou8f3F(Nvaz#G0(nzzF!*<CI5|BwYu&z`QeU2Achu^W!hHrxJvno2Gg1JE55
zk*Q26P=+sqeC|_f4ljXSLz8Lxe^7*!PxF%g*>197p9ikHG=8mnb71=H9LQFQZP0Y-
zL{v;;qXrw};QpOE!+@koFlG2umqV~OO<wyQlHF?6DNGO}k&Q$mqvD2YJ<HXSzy9^F
zDe{6(KlzkfH*azK_H9m2PSIH|k&sIB`S1V0dcEfU{rC8jKlx+wqF}i^!`g<jEXa!t
z=UPGtOs9Jiyb^jPQY;0;-%SVZ6l%i-2)4z=a9<gI9&<KsRLDtT9V%`UJ<FsbA{erD
zf=}Z2gL_@v`t=H%b86$+x3Kx8^IHi)gQ8#C-8m8XMVZyRS)IS$@RJY**V1BwPkoe_
zV(TSuol+SlH<)}xR*W&Z1WD+;5KyP3O4g+>fExUs`oToNyY-wq6xv;E=H?o|m~47o
zdm~^cfK9^ll@`CLl(WugjrI-|grRGl2dxR-k-L`CRZLaQp$Y7(5jy56Q*jZv)}w+(
z)q#_*A2U6iaWETE7KW_M@LGw3qcUg@hG^m)AbX{I?cI4FfyDXEcL<<D2%wX8v)Re#
zH|@dsgWrkIpp_m7zT`Kh`{z^1Jdzy7<M{(^ENPaY@MuD}_#}E)>x>Wz^yr9cX|U~*
z8wXR0Oj8sEPoF%YZ5nKz(Y6`d7#i0|#9bqsgR_>lu2`L&a&q#5d9~*F#tp_rf%1X6
zZc&95vt9_awn1fxx~_7GRP`S3V7Xk!Tu++Cd_@S*)-5ldJwt^+_KxW^Mr|^hTA1G8
zJXlL`mbPtatiyEx2Z}uB+TkJD2Z|RbXFPgx%FTCg@!p4@^3VRof5SWX-^Y4QYXin)
z!caPZ&^fDjf+z{r0qEWy<4&U@X~}Vu552(I5L7CSeF7cZD><cz7=9iGnzA&n99~CO
zI<h0VW<zKv?J$WFxtoeYaTv0nz*gwyOo$g%aSB-^O^mgHkjHB#Wt6muCzG8fcpdX1
zwm7>Ym{GJnLhvpsa$eSd(khd_rBtf46*rTx(M?IaPQBV`@^L+=jboF>n%*x$lHlcc
zMW?HOpC$k<ZD6qV$Q#dou64}AKqw8~FM<8UK<?TgcRtfVu<x3?>GzA~_VzI$^fM%}
zJ|Rtd`Z>0J?qKq5Djd*l!vDFnA^c8$t4c}w9#I}0>uSaCzxbT8EI`3K@4QQ1WHCo)
z0444WPoF+TDL6Vhq%5*7Gk~lw+06T(3i-D^4juD6_?h;Aew^aU9_$6ZDR#>ZNYRbg
zaNLtwe)tTbzhKAar9JA({kg3iT=YGvQ%)~BUd72W6pj7-*?qixpH2un=9=`ACtzX<
zp#<!b1wa@vnHy3sxggKeKgk=KWwK{#Wjc%??{WT}cgcVrtX<qk7quaH?#6iV{%->A
z-U!%<Ty^44p>&Ku%m~gCypGmuF{=eDWps?I8Ha`gEJw;R1~ggRJd>%&VYDPOHX+B)
zYhHf-2aaZArjrTzWP~z?pbcIJl#7|jV`5MpC5r$iVGUgJM}ZA`bJ&UfLP_s;K7e7k
zm_is>(>npJyG_2+F7)O-Hb7gvUk~hv!d~^emr^*Tq7K$iX5|AxDLe%_D^cTu#mk!Z
z=?MpWBV4<tZCWw~&bBm7h0b!Ux8z1kKUs%e*Q}Nc9zJ@+*?i82pMK12{{W*j&AOtg
zYF;%JuTIZ!-ZP#~IXFDTx&|K<b=|UDE_n9zDNmj}1r->LN}8sQCKZPcp2gWI55D<|
zJ9povEXU-<2#bx(k;S!*u)ns&+Lp#k$^;)gMP6|I+I2>S;p}wIKmE&p$nAUgx%1v9
zeEjK8FhxOITPdaG{YJLLzQ=|?`|;d0#@**WZFvK~O5L0Hd+@QjG45I|K5Vily?~<I
zI66ColBy%136V9D*%g7t=F5n-KLAdiblp_+N*vq0OXcI_#tvv*z_}oafW3+glj=aC
z__fG6!Re^1IlOBK-bgB<AOf5q(bY!kJ8kH(S|Q3`e8rSstz+xxBg$lR+WdVe$24>s
zfe$elOG_kd$If3}?p@fl))8d;Zr*GIwPEX+k_6I%uwTRDFWTg<%-9W%-6h;FGJWp?
zejzEtaBkWT*Tt@9?%w!<$NtX+++74F`uB_Fq?F;&;|GjKCE6H{j*ls`9OrB#2P7-O
zx~h2c<O!dB_8Er<Gj86z#%NSvU4wH{!y3T0KJEu8!hKgC_WQ{&;O{)R>wT?-<=eI}
zY%P%35<xM%SGwwO?!6Aw?drqVeijCD>J8cu&}@I!zekdt>Di+qLsC)(QeSQMRU&%i
z+~~V$40%zI#X8QC<XikPjWKbLe{&1Kw)J7VUG(`mP`CN**4V$ky_qRCSe(~l^A4bb
zmgrQk2rA%=5k1=0D7z*PEk%3Ao@zMG75h3v<7lFTFhSB`gb2z~ezV{M%NHj+{M{Fn
zvjYx_oRQ8jQ$xrUP9+oWd5v^?K>aYU*N%?sCQjJww4J8btqx9uVPf8@qz#`l&knAc
zSfRvBkhH595h)wIOsb7@2k)2!LRpR|uis|=>M>tGc>(R3%6hc6=(-l|C9@c<X+xl`
zYjB==x#r=+hb&i1KKkSnj;~*rYg&aa1JlunlhackK7CA8H|))3oUPW3CSz1k)K$&P
zS1)+}>>2aL8P>Lp#v{t2kcgs~96B=@%jJ@jS0@}DXP`_hDDJUsi)$*@^@_&UG;NKw
z4XthQ1jgeLlhKI1{R8g4^D(2@5ykX?pZxTX8ISj9eTV>AwCbz+Hg1|hFGx#s!SVfh
zey3dA6y(H8D;o!Nj9>x`0WYQGyu6DK(N8Nm#BTt@p#a!4(S#x2Hr&L0Z!%2)&;_tD
z$!#BO)t{eC!GRuxisQZjMX3xyH`j;#dEQ*3JAu3lI9wInr2gFo9mCImh_*T2S9?AI
zO9?c2Nz)YfAGLLy{De()<Fz;Fe<kS>9SR<>C8^&xPTm-s5T9*8ew=Gp>EDa?7<Hkd
zv;9gx4g%cvt#;9$H+`Pi8)Gz+$%y%U!B<~@$$Wl>wU$5mqdz8(dCQunVK$pF8IP!{
ziog1+zrtwvvp@S&-hTTIqtS%*dWo|Gf4D7jVAm!MyZbM7$@X`<_M&r|&(`%fKwU`k
zVXy~cE?k2bF})YHur)u2^5?>)dKG*__MYoIQF_&m=SuH3I9dGObia2x=el%IKm=}?
z_|ugoS=rZTPK;f8z+H0ovNn`Ym&vzXAa>{VND6@9fj4nAc{5WimJkF`sc3Ie2on^%
z$C_MdPPe9XYbJih)Gs+QmZKcXplCg`h#!w0R;f_30eONr*;~b{Z=W#Q|B`GvB_B_s
zL*oQx3{EQvrbr9`bTo6KdYyClK^~gz1(HCLQYw}d+v?cu1-140u7#Thp{LvR#&HSQ
zZGgH6Zc_p_c_y=Sz&9q0QmA6WjeGBNaP0=FUGn8`8?<tuwUpkn!q(*~qpdCTlT#i)
zevEgH4?g~cx8Hk@0O-tMgQa!yypw|ie)h{hA<J`|_ne%}dGYcEtL1|Aa!FNJkim3+
zj}Ro8u2&ux9OJyCC`<PC_6SNdpU)}Eu`pk;BehX1n>CB&8SC|ewq2u*e4Z6K<#@t)
zIzv<7eTFV3eDc#j;o9}LXd<Y~a}zV;pZAfx&;&5}ej8wfsAz0&Oxx$B?Bf?56F-Um
zkvrN@vrX(>+<;Mte}_&c?14hCp)k-rXSZ1=5Eg&>E&yU+8X@|d#F9`6Xol`NJcEmM
zKrtn%cHR+U3Yj4)82ZEwf2JtPZBVxPI~&*9-Fg8#RM0k<sMjPWze?L=aT#QqzvP?&
zcygYuZlqY}67wz_*-(ABs{S3+kP1_sf*{|?XK6$AJII*Mj}AK56(1xLHTljB-~IoW
zS_l&O1A{q;MA<GD3-T-@c+c6{3IFj|{{gMx!w)~=+VOS9;}It(C*+y1I@9SMpa0=Y
zP>Pe&Q~u}|KVv+avRbczz<4~NZPFa<*T6Z2Pi-k>8z5+t9QYn@PA4CB%-%q_^87n@
z?q#6u0tN3Xb4$i&$LEQcQwloS16{#_C}*`&D5JaUxBxa<CuA=+Vs<S?jLH#tQIZvf
z)PIdkUH=}vT!tSoKjzl^!nVxWpcO9o{Ppe4OtC5@Mo_UPDX4C9@G(lyk-0TfKj$bc
z*;f@unPNZ7!NELw+$JlOjsn}Fq5%oA0GSm5_Tr0g**n~4G#N_@l)NAqDVwDUo$pcF
z7*jfiK(C!8@OG}a_dr-~Ynt0&EM0V)fg954$qc+bAm<svbcnoTV(5teiljuV1Fa^t
z^e$PJ5s+#E%hpqtd*W-Q3v}kunbc=cc%kRL^wrs7&UAl|4?p~n!{cKDkQKRTJGo+1
z6j<wMTSrkyT8!Kn_JYPK$c#qoKvu*m4)|z-H55gL)|#R$m`?U2e7wjZ2)GMMF`mS7
zJ<XcsdO_V(xL|R<?fk@S2s9qnt?ZKfvm<`__x?WDZr-M{4s<4&Ata$4naPH}>RV&m
z4RMU_^<7$wZNQce+~ughEu78+`fY7WvCRT+2)qc4L>L({d#dQeI+!<`<G%~s4X!;z
zi|I2zyk|_R>%<&uqY1nFHiQm(bl5ZWOpq|(5xI2if;_}9kYTg0dK1dFxuMf;!>+C8
z;<nA0ccA<S2%hNIy=cm*UthN|rT$TxSU53sj7s`{k6(Lg=1Ky4&_CVL1DVvJbG4P~
zV(@Gs&<88UL&V&yy79d#u-m|y=RUjI2J}O+th=r~oR_KfY?aJOANQ<OE^WPD@`o=z
zXEYvRjOO;uTQLQXuzp2R(lixMo;*b>&15|0gAd-PEXMNw{=5Q{?%V&pOXgh8_PlP2
zZ1&{<&<*={TM@nR{PWLt(PtY>U9TOSQ}C|RE}jo*^(6m5;w#E~CJ+}QJ>KQF3=&W_
zbi*<xi%H%~@<ROTOayi3u|Xf4c2!Ju)%U*2+)qmVxzDcH-jK12CaC0V7C;40t|d)?
zvMq)7?6GD)tT<E^N5-+Ipp2%UTx)bHWbM5;-UpJ^oHjUeGP0miUQwMbc=XxljK*WK
zvLG)@NgSBvLc1yzqyL0vb@T#039%t9J4EF}$~N`lcDja(*tuN*ubX6Hcj|?0D!m)p
zSOp+UH6zh6D(5_q*kEn|lNSV~u&x4SXk4JsFj9s*Gh{|ln2cQISnD`gt~fe6qV+J3
z;rU8AOqL0t)Ed`%Mw2N;UN9=hD5V({W5(r}YHttM-lA<A0cFmySj@4lM6}wr#h8qy
zu_zkqre!jjlIJ<|`57n~^YwZ~)3#J~#cI72veHR_#o^%*`-e9`<-C38L+-!#5!3x6
z+7QUgQk<KS(0UUSl5RtT^W&BrQ1zt?>vf5(J3w`XMVFa=F7?`9|F8cn&LO3MlshT1
zK3buJ?s5SRZ`?;u_xYAoOiklbDwHS)LlBMD2Sw#PjrF)-I#x-7J%;R1i2W?gx6ZSe
z8_&fw9U}M=NZyWwz04GGZqw7M%ZWCuJ)6Go|30snU99bXx0cvXm6$Olse{6Rk-Xr{
z4nRQ<vR~H^SK0iw+2xC_chPs7fV-c82{2v-*qsL|yN>C8d)+JjsGx4!mwNkNX{{;c
zvEGpz6BR|rqel<fo6RVSg5$$Oin5?yS9tFz%Msq^%;yW7dx$ZbYu9dY<Hj}eJnO1h
zT=c$bAUKBdYI85Scmsdk9JsJYcU^7=fc!3P<J>m@b!tn2-Ze&Yp^YE(?hR`2X(CIc
ze}@cy>N8^ud7e=e1x0ch%5#kAf!($ue^n)A*t#`#!Fn1h?Yn^e#=Gyh_Ep=P0lNU9
zqdmo~TapqtC|^;!hFR`8&@BfzromDMM}|0HC&8>#%+Vb}%-J~b(}7q^P>}^i5ft@{
zS3LXt59H$lQx=puBio-zb&I%=QxxpbbT#bUVdCNQAZgPSyTO<YI$Xynn{!OwM#uLP
z;xcg83DY=JWtzrXSUTp=Q9WJ0d$1#8sI~Y^#7x;9F~w?Q$O=PQlw>NWREp!aVKSZK
zz31fBD>RDPc*=M@rUA-w%%q&6m10s(Daw+(95WqFIhq~PHWh7C;hd#fubD5-80Qi=
zVXbAoUQ0=^;Ak4lx~@1qJ>$;XcO_Me5st`v%W}PrMb0ynlJZO*&wKYj;vf8@|BBXY
zG$V{DXuV7v>%0W+1mDdaxIzxIlw@|tg4_IiizC~if$su#!{fVo^BV0u-4CL_Akl1h
z;k$$#=Wc@s@c|5eu$d3uO*mZfv2$MXLH5!!&fn0_&)y`_zelG-tK9-@d-K^K4t6L?
zA<$iG7sI_v2yWl+hXZyVg6?u;Dy4|@e}BIngWrQAgj-1a(j`FmO-;vd0s=4ktlM~0
z1m{ZkW4(2cH}B8y-~T^EdtJW<<<Z74E=SB4YucEKsVocD>os3|@j2_&5@QrU|M?$*
z@-%fr-Byf86RN6WK40)xfBiQc93J!YpZ{ExWh*tdmGbAt9nJ%@)cVhMGjvH=I0t~v
z&8bb6CtX*o^X$)?jMY`x>hsTg{&RVp^8jYooa;W@P!?48tBN;D#7$?n^C8x7PI5)|
z!MrFa%aU?5p(vsmHZL(I8;~+rHydBx7rW2Dj4;@FJZwDIo3uA%>~yR+Fl0a+-q#pv
z_WYb{K5$rQCc#rWsnZFDpt6`FX)x&nT`C73|2aGsgH;;sJw_{h({TFm5u@3J$-x1p
z7z-HKo1pTn>zFQ+dt@?ftY$FWFm^>E4Yj0gpE%eAV~WiMx@+O?+MG6yjhlr`feyOG
z2a$=WjX?$14Q@JrxVXEG+_&V)g8`!g-ddE(r7uh-xp}l`E7#0TOlD}Rnn_jjZ~o8!
zn#bQh;FFI&;riXTnH?N59Zx9og4uLMW^!hG2eCAdqRdOItEuJ{tJRv-VueD}tQzL?
zIaOWL+J>sGX_|(;{XOp9y~DL@$Jl2rqtS?}T1$Eol@Z9$+DVSSrXd8wwd-%;Rf$8R
zvK&hw=*Y}!EeVL@{7DqVg^MHg*A>~3ZBV!S=hlH+W1j?t^Fx8vUwxS8DEJgaL042-
zuCE5*>I#c`L(?B9Hol+Z@i#osI!SwC^boxD^E^>GLvWYaG~M;ZW}gld6sP|3zMrex
zpswq)I6mpzVJ^JkScGsMXb;EzhcLx%Hvx8Uwh4CX-36z;PBXvg_;b&7g=06_x;Pf{
zMYkrxCFlMSVEON|UB;56@4XK+-r;P;WKz&H7Hd80s%A1C<7~_G=TG_7uYSex@imSP
z4jGTfSZi4>mxK_=qTk(be)AdYb<O+lzt7Kq{&Sjoi6rjtpXYVV*yet59+aLhy>;Q+
zvzZb*&XXulw(fcf#dpQKT_~5n=Y`L;i?rzU&A|OCNO0bet;_Q~rZg#|Kb`IZCANbS
zeh4Oc7eL%5l!l*wKV@ondqc)f6Jmyc37!aIE>w;z^BmG}T|4%4fe8V;!js{24$9!w
z&<`lS2+`?LX#x%(zy>eCE+~rNDYd3uS3LjvTSohb=<y!;Xo@THE_a;(9I5LwMBphR
zX1Ku41yi8nyPe-KHi6rAf1Dq<UH{HsPl1jbnm7iCvwX}C7<GOrV9&)MQtBo;unibb
zt^!4#;VQ{3r;V7;toM}T0$t>ci;T8yDV5@_@-{#F$xr#_cc1g@(PPT-4CNJh(7Zf*
z&T={<(-|-G7ig5!c2!`##d}NLG%S~MUcP!sUDvEuOO}fT-#qx1-+l2V_wK#LzxbE`
zoO}1~5meydV4uCcDFIK}l;n8{YK25jYP8BIi-PI&5LAiM28T!MAn#H&F+EP`{j$Pt
zC-n17Pn@&rd?~Qa`FMWZ&H;8&7kA!o^W4oV4^0Q>H}}w+Mb+S)!WG4282Y*OS<Ht?
zgcEoRD69*#Hc;ff*bE`WR2|U6%@`Em9M*!-noMW0Pd3@0?e`xlPMd$9&*_zZ4&5k*
z*W3nmyT<Q_5XwVf-i^X*=1Y=kvK}KQR^ZdzJg3FW#_zTw9X966g#tc0m~Vr$!S_-7
z955&HFJa{_-AV8Sz3CW!8Th#DyYw>^N!i|@wprir(}we8SVk_@7fy3#7zwGA>hjac
z_(E_#=H%>^4{N1zPt!J>o}TbmfAuRq`S??A-+qhf-X2XQ6<Cx~j3-kTi#cC^^)&#u
zZr$Qwf6`Ts$@84LuDka&#!yu?MIp(L{h$)X>)!WVi)(c2qm8-HDR(=s+AECPM8_`n
zcH_*!dgxs+#Q%QRtnIHS_Z3C^JX{lb(m(s)8@i;=YU9)Hx;~`Rhmu<`C15t0lwK?f
zsVq?xLzeFR`&{<EoBxK--OKi&0)PJfFJ1$~Yxi<IUVp=7>}u_efSrzhZwig}3PV7L
z78TZv)tY17aG(PPQU=SR2^c(vOevHKm=Mq*${wY<toV@wjHo-%0=bD1FkYOItJ4)v
zzkWczzfU<ClaCB0)0iyBc#U@<mf|pC)(TRR$~%wBH{Ye}w3q}Kwl${>{z3q#PS2`(
z)@=(6s%Y9!Dlu?saIx-1r4Lm!#l{4hIsrQ^K?zDp;$*q5T*kTHNj(|+0us<-OwMRL
zX5B8)8DyEl$H>^$TMQZHXoSgf){7Nck@M+Ke})S1^4SZ1`R9LzG8s2-zr{D-JmC2D
zT^@e%6)MZYE9UbBst`ZCRaH|}E769^lJR7MwGF>}^nh<4JmBMx@AH@c^e=h;;}0>T
z5!PF79NwYGGghlL1ZZtT2)+wAvDPx09WpyO#(HoVGNWnR6(*(u@m9)(g+OPD4TI6@
z7b}Cm)!+z_dW2!)xWUZm$18%kA=vAH-PU<Z>+bpqjO}7`w|^fdDHZiZ$9kix8{6TM
zzD(GrvTQC!u=-eNe(OmILE%Ecdaxd{&K%pjBnVR{;Y?B_Ug4c+!x|fz_w>3Rr2|s>
zChqK^ACFBT*pWqBKR4IlwjdiGw=E!h!Tdu2yFsAUjuc2LBxaQk_@$BZp`SKBq=d>B
zr3yJOs5kmy)27xR!{r-9zsJN#*bda42YLwrGU)qVo7}dNbH2~cD>J(`qVEIlhWFmZ
z3~Mb3+N-KMXIe|&_rFW;I-qIYKrTRQEm@}VUQtzRe*4?c*x%o0Z#Lr~`rKLPKuG{w
z+ghGKe}UG9ukU=##~<I9!aC7!D*>PFsx0DHy4^<?^~q+SlpMLK6l`82N{()QZVD{5
zy(l>kZZ4BE+rJOjpW<>QVAu67mc_{xBJHE+(O@Dt(!W&oN-2fXT54`bKvxtcd0q(6
zj|$CI-g~1GPv>j9YfXgynFrU~C9{U#4cm4#9Z1C21pxno_C~<Y5Xij71aVl;42;Jx
zQI=74#<8lIXcHxe!H1Z+-v^9=9ES^@97U@%P6b*m9io-OD1{9kucU~(C&LAeWc<zW
zjpOX;bBZtjz+_r*uvbdy7q3xdxHh1pLw!o18aycPF)EekHfh%-r5Ay?=yf8(Jt~G$
z>j7mFbKG}K5h$tUnQA_g=AXf<=nStED((^8_NkQkHYttu4yA$w!fM$VwdrDjV%m#d
zGZIh|6Uf3q>&ghqFq$5)TE1XaoiQHmMJ7;yV({X7WymSAF)o0fOnC2SKc#r^EstJ2
z;^_KyvZFn&kKd;}Jm%VT%v-nL#+Z!H{^NgOJezWGbin-Vgjdg>ayXsv^6^7LTl4(n
zF$dT7_+R|=FZuYVKjGloHLTIpMpI1p8Sn2?&*u;{u4!qT6($ouueS9#7dSe;A?4Dt
zg0_J?7;;q#IQM}Juo1BIfUfVaA;3!uP}YM+GtiJ#8nbP!cTksb0PYe<A1rCMPTVPH
zhb%G3I1X!@ieQp)Q3x0?KKh`-hL*am&2F=j0!pODaQC@lOGprls%xp-;XSB;Nt4qn
zlo2MxV{xIvhX$jjKt}KiuN@&+tdkN{KE(ABS_oFwS?F0ddGF*Bp$B6#cy6Mgi|;Lg
z9-&)9GFCp0r>udY!Xtfrb-{O0(>m2@Oy!8=ITAQ<F~t&aZtkQuz)CwdtMAudcfEeo
zx%k;aQapA7ga7<m=h_h2Ip^F4Q<r5KqtS@Hy*-{id)9qd6b07W4W@299#hvft+m};
z+qHBPfDR$>?Y9q5N>da$hqFU`2((Sh%a<=@USv6o#gc#j@Bcl<cs~5_1In^wxm<S7
zp_HO+TlV+&<^F9O{f+-F+U6U|I=dKreHS)-7g*c|IGdmE0#w`g2!zD^T%ZVr&{y7(
zb!RX}fU`D60wwY?{*(m7T4jjkMr~d1vNiL1R_`L*^*W||`}eD!e_4C81jXJHRD&{5
zl!2SqZ*Y6>n0!9t^h<|+^^{J3joI#dtu~+{Z0a^G)oUIqZP3HPBE%>?@zu*!pkCKJ
z|K=h4hX<I$V@xrj*v}ygxMb>$g4m!Wzg`psA|PVnVwHk9q5zL&oT3pl&ZFdk85jCS
zZ>N{g-5;F;X9R;j9-_QTRl+(ZPsQk5qof9`LDGWwmRO?6fKri#l+Ca2I~5lupuHUD
zHLbHur~5n)8s{C>dJ3ImbVf?^VXY&Jeg)220+1C2$Jej1eD#W#&t5S3_93_4xlfUm
zsQp9oynw9WoqP8g?N2b7CRdu=c?^y`%X#?dA*1n#fBcXCG1I*XMDvHqa>{bV;o%Vk
zPiv)$cvaUiCz{7Os2hhz^VVGfkuC%>V+7D8O)N!B4z)ba1Fc=AYY+uvv-P8zK+v!l
z=8zHCZ09xT^!fI0mqmVUpSe?iKQAxOJwx|=D&BDiU`@>jVFtx)5%317<^tUsO#Lev
z@BQGq5Pa`z*E0fvAsAM1!F0UiYqU)PYUjTnvUkI+X~^7dvo_zib@!CA8GwLHW6VZ@
z?~7Wm%n#pZ2()%@QvG+^=f3ud)15E*ev`%NfZhLh;JVjts)-HD^PD?(?(pKp3zo|z
zP1AJXE-~13U2lNGO|C`ctM>t=94=V2S3G<E1m`S8QSggj{4t~P7?Wiz&KBf(j@B8i
zwS4i#mlTC&G8uE{&Rb+z*6lk(X4x1M*}5MX_5aB2JVW<)bl+V}7zhF6AyzhwYfr3S
z;zO4+8jTo@r&47i%Vgh8?B5k$;VO#MMeOwJ`suv8(B;qZCVl>9jGfkazh<;o^3MC$
z`Q+nU+&;)?o*nVEsX6)08oOMOYa>jTcVbEmA|(68lhj!{5T*&SW~KN~`IxXcr0pRD
zG7JPob-L!+7Z1s1M~o*kCZjRlXuR<_?FdoG8wB8p3CW>!CUiyC`&%*CJUUx>7(5@a
znIvr>e7icK=_A!~NAl;E&lS33ARIyJ<;E%v1cmmJ6<axlZ#_+Dsj4;Ax}vHps&&PB
zy{2wz>Pob}Hb}AeT<4UTW|XxUwPutTI3LSX1v+3Sk#HVuG`5w^I|Rq5ETEjAtl{h5
zeZi}jORn9#O_q&d<tg$E(>na3#x@mpIY-x?s#@{*n@8Z_oxAVSG!?5=O*tKr=LK48
z$uk!MO;xj6tvFlEC26QqXp_^}mQ~%7<zuejxQ+El0Ejj?XVLfpnAnW)lVM6JVEeXf
z5$JQ~y?#*%BO3Ze|DSnF)co+@M!n+;Wq`b+^PZ-4lty9bD%n9)j>POBKU1DB@!bR5
zrb|FCRpT=yFh|b{z0RC>{{6Om8#_k4ADbT{A@PnIZ@fbYY|4nFxL(AFG1Tyf3cz{g
z;DQG@_x<_)8Y(eEmTuQE=NY`D&iJ%$2Iivk{zHRz8;tR`(sR)U_d~Q@rs1Ivowb&x
zX;`gR93LOEzrT-Cip64qwYD>hwr#tS^2%)i_5L;kf2ZJM(8PiV55DE*&0Ac*ev_go
zm`o<orw>q|jppgoCw%$k7bpeqyz@>6?o$6$RYh5rn9N{pyY-G2fxXQ!iM_CM4)$i(
ze(r_%a@)Y@qW<0f?E9F|yQkfn8;SfFelM3kKh|OExbz$$s11L#0A0R1F~JsH&blni
z`n{I(fNlG|!mfjM9N){S;pbPAZ@YfqeXlnM({JWurvlSG%?BUf<QISTKKI@|W|CQ!
zdx2(khFQ*e^@lHM12J8IkDOQxZcs#jHr1s&AtZbi`GWWo-g~_E_*j8fqX^bvf+9z=
zc>Idz)32HA%_zzeeSCl&<pd?>o=hoB5Yve#Ix=hVr_(X5h%h3mmtIOm(CNjY&i6u4
zIyEu^19~|1wbn6hv(ltO%zcWLXd*MQqJ$A(7lI-vEr}Fck8drj#R+Gprz{ss=8IP>
zmS;3|L(?`?Rn?gmoD~h!Do54SXbj3Za#eHxokPm(7SpmMBPUSfb%x3{-aD)pAnCk~
zK^6{8M)zxzrT_r|07*naR56{QeFikxWrbVS6w?vwlT%*K=lHfI1L|tca=D~!E0*g8
zSkG#)qKcX9N0X8)m%x@R%R~USmgQo>dc9(~T(YifOqNl#ma28wpt*bJ9rpH*0dayR
zGz4dNIvH>HDZsgXZP#MiwS0yEVi)_jd&4>}Xn(9t?c#PGqz;AN>zenHsMvXhk4bfh
zm^J~qj+IO6GI;Qk+%>gbqD+*Pi?^!FThIP=f-*6I^kcTog#3^|FGK}zD8MiBuiW+Y
z7irkt80|Q3lUqA~%?sPM$t#47&n|n;A*kDV%nnd>8A}n5dhNZxYui?4(zrKG1Hkd|
zG1KW3t#xPeOqB(aNmi5~#jenW0G+R-k7e{i2-Ho@lP8aP@Zej<<1yvYHKx-Ei^US>
z<(frVa&~scmtTHGmSx<!b*qb}P4l&B8ph)>CKIJ~6ClgzUx~rlGOwm*AI_luGD%4p
z+-n=iY$?m<?so+SYS^~H|8UMEQ?S~~P@#4x)twya)?ZRUROe4OI6CgLrmOv$KwXw)
zoiFzge4PVyoxp?P{=VH0yXN@oOtriEXxC>$S>8X)n@Q0(19rx7=k5)D^0RmN=qGQp
zcbH-8C2h6G(Y?3H7IUikoU`XAsK^TGxbY543-wX8$rzfP1L;y&c)a+th#e9gL~HjR
zN@cLM<K&x1OlEtG_9l!A$VNw!B16SeW7?p+7({&_R(LS|lIU)_)tSkJY*Z|S4q}q(
z0bLx1I2Z}gMFudI6_aZwP$!|)k_gtIjV$uIuBlcPub!Q7^6HfN>74a)P1`m&U(>Xe
zXqu6+a?VlLH9k7Yx1ptR79Dajz$?$=$1k{XZNkCch|J^^IzzdZJj*&^VoXk!8Eo62
z2;wvx0*42O93Pjg8q4wFF?-jp;i`tyw#FG)y*k4#mJ}L_vZSq7EY}rfQLt_*vOFU%
z3rvxb<vG*wm=HDUYPFVpP)&pN4zCq;YpESH&hXv`Kf&bU`xm?u@DRKN+YA_mZn|Hx
zkalTAnv2g*!Y-f~rWe?~>Gk~JE<cjatULI8*L{2N6c=tdFK~x<yGS?;H@WoBC7(9Z
zcSNKlBEltLX9He=H==B%=k6Xqq(GhS&j`BvXV9u=eG;n`QpD%ryNjCicVFpYpiXs5
zRFfPz`XtiB)>nP`VOnAj-c(cWK%Kf=6z}SdA*eei*0<;Ch0h%}uRl;)<yvpH`Rr;J
z-Tyn1Id9aqP49`Zan7+`uX*z12{&)v<l41s6h*;TUwuVg*8;qXq62%ddklPfk0^Jn
z3lxRNI?H0Y<afXO9i!2h%;em@eTP+DQ!m#L0+Y#vvu4RR-+WEmHXI%v^3$LGbmRI(
zQOJ9AKGUyP#&$7TTp*X^$nEvEIiosY@)~frZF;@v@2j*Sa2-DGrd%J8C*3(Z0`1;y
zDuh%jF|G|Iu2WiRGLw^;f-En|ijn-ximhn--9bXT|L#HZ>ww#?W$`9pZ)nv>^ZUEX
zv^N5F`-cUeeEKdQfATIjZXROwnx+9&mdvi-puE4PS}mCW>esZd7G%MZD~bM8D*DTL
zVxeAz>XPHju$fd90HQg_N0%oN>MABh_CcKJo7I|=Zyzu{+#`$(d*hr;?_spTdjWRZ
z5R-5UOBIxgWxJ$&S67pB(6MMQ36*2SgVG`5Iz!Ta(vyb(TBq=7-Mj9o6wzvs;0)E}
z0JN=RHD9urFL?Uo31_d)SS?r7RgG&MAwb(SRP9Pko9i|6`JCl)$$UO%v0Mm?SJ%{`
z#-mW)kfAAb%Ud_cy!*}}C#MCo=@cIv-b33q7@bpQqgeheI-%>RJ!wx+0h4P+x}dFA
zJpJm7x~{=Ha-#u@kypYik2WwGmn`cwWSVj^rWlPPz?C4IrmCr`isfR#a<yb#SCW@5
zIBMruRxNkm{g}7zy^r-K08v4%zOq44fQO#_*j|jUV;VNW9i8U44YJOgm3Cimc&+5~
zxDD(s5{lbkq&M{p&6%$!TKf0upJ$jgUDi+N%s)wZ0%E+;at%*n?E=;)j2UYDbPo~`
zldMWP1Vb|44@!v9r}{`5_M2?-NvKOVq_OXToz|k5sR1GPChX*QZJj^iaPEbR{TI9Q
z=->Y$cI(G#7nyjp)}7gK7f$HsYya$j=Opw6ZJ4eE=z!ez#^=4~?CcEhJ^TCnT)%#u
z_uqe?Cr_R*pU-KUX2Udk9@_PN;61pgN;FMNmIcZ(XStm7;K4T-V;E1yj7B58wX{u3
z(@667<zmjWXV3Vz|MuUqzrW9o8#gG50`EQ7u3ck3pHtTslWkYX=$~QNb8UPQb_t8X
z?t~+|F8Kz^${T{XtF(^&9{e3b^6O1=H!9>(5=TH?_dN89YYKSMh9Z~B5=GtzOQd%g
zzRT6$XV*L#e0#zAe%(E`+00=-^q_8#dg%=p;~N3HyKi6P{`+^ias3)aZfM&I=RujA
za&LyYb(3q=eGZ<#;KlkYaF*a=GS=vbtU_Y#bhil*`uejPNyU(KOCJI*DV;$N^$xTP
z%%48v*%u}GY{Y0fVw~k@mEnU1BNYvN?8t!XqurE(2>7}J@=8U|g@IX4>6pqvp}Jzg
zJ!2OF!oVelP6LL>lq4oECk4~3S2bs6XS{m(lIKsKvsy1{>V|c-CV1ERb3A+gj0cY&
z^5W$S=JPr0b=8&d+AxDDC+`q4G7LE_!A=Pw(6lv+)ttTQn4rZ3?J`Tq0^UhfSTYHi
zOcU}z>sq{B0~y+D+O}q0Ezk-Y8*rh;l(Np8bL69(LYGvkMXQWrQlg6tof(u?l7gdk
ztd}dQ)tdEsjq}2Ou2yT-RZU(_`N_|I3A&IJIf2N!N)-SRwC78G?E*O$w!uc5ex3(n
zuiLWwOYH)f=Rh4jsN1FSrgOcQh-ITp%wzcGEeV4TSVjN1#3akdLHe{?TVTDyIq(KB
zl0e*V0qLY6D<7on*bvkWLGgyh8Q*Qd4D9m9O22#G=bP-VLoZf^_3u+wY|jG;40?TV
z!&UlmyKRi&6|d86JU;?D+I8J+pu1}`kk67YGhOFitQ;l6b#ty?^i1)HH-4Yjwc(f#
zpT&FMfj$rb7($?`DnQt={r!Co4h{$*uvjdjxn5?}fQM!%Llix@Q%cx6tyRaiot~cZ
z<jG@Zvjg6__b%mVgwmRNwZ`O1Qlhji&z?Qw_rL!=N-2(yk15Ntdv1y9?sEuURbjof
zpSu4m2Fqx}<bF~AUiR6xT)S#JZ_SVuEwW-aTXvplj8TU=NdblI==~n3mFaxBB^hyF
zE=<R?CO#=;gEhQyr>%Fp=uNl#qBk=Szkhu{rSo?sV(<U;Ze4E*?%oL4ee}s)-oAH>
z*>sB10ZmP&M+EH%%9HKyF}ih~JD+^W`s7q7@c5#tV<l8q<5hP;f3N`s70sta?N~~T
zfz+~;Knj6hnbNqnV*d31&E0=~NtT>vp5V{yD%YVSBUNP;Py(u;0cz0{1Crg-J*TJH
zv%NcK_tTvHulB?4$343}+dCwi<WK-4&=dd)C=`G~p{lY{hK}%Xd6k*%2Q$0(x`#(r
zW(hWXEstkJxL?1*#B0y{yy}Sa_u8~NHdl=l0Xj<1L0FZa@<};G;C$<xt;7p1QHQO%
z>1VT532Afp8KEn1>V^RDPV1t+f0tT>#^8ABcREcu*gxd>;E+7cDT@-7D^yvMXLF8@
z4|(t2T^>BT&-ukUvn-=js$uxEo0~Rz34+v<$pB$25rL%N?=cz=Ns<5~B%z3`UtU?*
zNLe5ZEZr@HL<>QYmndBzF_t})hITKeRE1?PN@cNu5aeaST+NvlGejhalZdDlBAhqA
zF@~ZjDbj-R@PgT7N}AcyW0}gy%A8VbUV7zau3f(gVM3_^v@_u=w?ST>++Nswt?g;r
zlR@26!NaP%y}4r*LswmcMNqd=SbApCyCZx=8>8<ZDH_qDO_by|F0Bkof-_WCjE#*#
zt^+78Z`g}L8_k5D;DMg36uFh-=D#336Q=zVbUo9|(<m+TZWev%mdzI*x;5|B-$eu1
zE%iY4`$|XlRiw(t&{nZ@S1Up8IBEsu$)?@a&#(%7e6*ZS^B%5dKbC>M5TY{64u?aV
zgRa-Jif)#X7X|eqc+y4f^Jwi-|M+^$!Ukms5zy`SxxROUR;!H=g86ic(hx@p7#NK%
zc;}sW==TRCN#atiSYN<q(%D6j@GKqUeu|~P0mNk=yG!5WvOK!-Qcc;_{2usC6spy_
zqB#Y;8C>Yk>rcIK4~<(Eg334FTLZk}WIwklo`NQ@bD*xW09JFTdGAYepwf2SkL5g;
z{t1`+uOTn4YRka&(%M-9_Ld~7UAnxyecJYHjNNnB*xFeqX|>37iqH{83G0(4B}EXB
zY_79)`zGnxIS(#I=!-Ed2tb_awQ-|{8VFU|lw%d#PutI46RA5t4np$joa6fsNqSvt
z77jy9e+|=$ZB(dkPA0$6Ifg?9)xd~~v1<rOWAU6IK&xfJPJ@+mY>V(P)BzF%Wl=I7
zP1)Z+<mmW>BFhn_VLqQzlqq-Ldxsyq@dF+mJY+H(qm9jFS88`J6`r)dhght-cpp&p
zR$eF&u(`Ft*~t;J>6C7-LuOKh2oOk;NKzIB#GzXR0v#GeAVEt45u!_%=+@XgWVy~z
zqC_d{^H=DS!W4vYKo~~^VL%WC1TIZUp63*~VltXC8ILJ->CC=Rqy<@_*x9|#=RW@>
z41ooWBCvfY?C4ijDHh7UGy%DAn%MhqqF!^oT0=FSrmbe`vi7?$m(A7S&R`bnwO(=E
zW&%^M06e*g#?Qrd)fA~s@U+Nw=<0h0>uaZs4KDJ)PMbQDyKz5-R#r(7KHZ6OelE@D
zv-eeL<jca@X}0zHsP}}X56mh?4<{Qei?Rr0nx)7-Y-=XBt^3wKv!cOJ`X}>KWo7fS
z%pbqvz*U$Tn!AVl^YYj(|9<6NH2TQ??Kk8ne(fV`AB25c{(jT++64DLTHep`W@6y;
z^K*hgaBX*wz1>~nDB%A6hjq0Cz<0uOSNJ{NGIxfuRMs~y5S4GV5b)xQFVg9DNP+|c
zNtUMcdOh0h2vx#QfBG}l*Ei|+2kh<L;OOXxwY34QICg1Q>bvm!&MF0P`R1&QTs}rD
z98*ovYP<BuNq3H=$!8Wp*_B}0|FQrmSE5U|fnLoN8VA;N{f$~acvjVFy!Duf##N+1
z1|S1L7}=CJHUd@>RLMDwBWjxWCaX2pl%l?`s&bW$XYw-6a)yiW)>PQ#$*tL?S1DDM
z%%-WZu=3ce+A{&W8@IPe`Uxt~2qXaz3V{ls3>3x$kaP%lx7c|3d1e=9oW1@NOj;rW
zNvd*;Oc2VN*@Q99ug;r7C5QlxCKrN03L?jNRZ+C)s#1gz5kgBM6y<Qn>D%{cx7)OP
z17w^)y9ER`mt9_<5=|1>1cgSpT}8OG!`fb72nf7(AV6DV(n3v7QPT?GB5k#>!bJ-M
zp&$|oWeS04k&S0OxO>3S@d4Tt7-`5%!Gn{By!P7n_}N=;GM!H>W8o)ES<e8ykbi9n
zTpFJ~#+tpV9A(;&L_KykB90ID31mUgjVQ~4bUp>G2}DR)7(zk^#^!`FwEZnYqE<kW
zXXK?Jib6nBVkz%Zm54$j0vq`%qmZ)9U0Q(>T^i=aoWhxw3tdv`f;5{molUry%*af_
z=U)9~)&@JwGn+d~8iO*WP01n!p$HIob4x8eJP`A2l5vZ&GKsr)Wu+sETDxra<^qWD
z0;W+I-T2@D@B83{6}n0<QkljDHJDpcr-cKI%goOzLGHn74MMHvX48!r=X2tPgbP0s
zw$^f~wT*CfneD?MAe2D`10}+p!#-;vATUrELuh<dxIrT*l%a|qmjWY<EtQ3$)CHid
zLZIEuGhVwimN7tCKUQN<E~$3)tfB^c-aPCBth_ef1b4=bEoj0(I^Q>Ip{blFJ|uu>
z42lI4MpI~;RNbdHtdvY@rg#BRdo(Km5m=?rIo=BCs_wYCV3cxWds%QBizvL^VBz*w
zF_!*!|97*D;xdcplex;^vY)80%HsRH8m_0_Q?A~`{m%~y^j1YZdu7}&h_cYMS_x7{
zOlNb>&xWX?WOIF;TQ~L)T5)`G#%wmX7N8)g>WK#~scl)h1jkWGmKQdkriP1)F}+@o
zB(XI|$I}r%dgFDnBIWwtb)J9W4iD}=1d-T#6%jXX+~MB6M_i1i937wXFaOPd&uDnc
zJWXvKbfXA^z~+rGRv8Y0fFO{_K#-*=fs7pU)RYrcfXIp|+ppfItU6|OPgY<F$6NVy
zLjF8nNi%lg)pOnF-}9KQS`mH@TiIwWYt~Ou8fOjgX(qf@=K>{!^!9@0IqMdLp{;+T
zG+LLGT7fW#0Av&pB^JzuQAE^=To9u{>k@D6Y0!cWlv$as4KC-d25t8c>q+nb)y?ny
zM!J1K3TwoZ{u-`-v{;a1&HdZ2^@fmew`JwBc7DA@QQB8tR92H`SvSuXJh3rA1XQNL
z>L#7QH-L}<QLjVy+7>&{-y+*Tpm=<O5fBCexpp|*3moA)uNFQ=2$!Ik>Jz&u`Rf>`
z1lSErYJ(~=W@kf=?>{2kSf|%%6SNYDtV5V_11E(<+RSKfuxw{ZSDC^)>sRzAwNMg`
zOyPby2hs;*xSxT<2q;TU7_^y<Qr>&_0fz_2C|#mV!6cpV#?OArfBF0W!tnf@Qk6VG
z?O)j{GZ@-w7cOo{0@BzVcSS+J)1z41<aBscS3HPqKeUn%V<bidL}3d|X~#7PNRl>j
zoRH@kd7e?0IeC~;$^v7oFP2gTd0ALLI-^OFHfcJi$P0wA36p2D8MAa|7aEGx1%s4j
z$=c>^);D%3OP9I=Tg17NIc}2_ORU){mZA=3@FqcH^Cvb&>`7+R=D?bYIrAI$f==fl
z;w1r8F?CJASzTlCIu+aJaw~YCAY{!}FFj#>KgN%L?O5xdzlM~neKQ1xKtj@r!Dy5!
zQN}WNTH8Qy?;lzXfvCZZ7s{6lll%43*f0e4d5jfYI8eFR51QpD<Ywt5--pJ#PAvBL
z0(12Y3hkB61A1c?b^KMUvkC5+-}eXmz^_V<U&?gv!EOB|;eR*OySK3N$=}=R`F~kq
zY5u&*RNVx9tG@p%tvbUM1My**yM=2tuOaGd3t@vS0x4<7Az5CcRY{hm&NSPnEK8(-
z>1<9MNR0CZE|mhMow5L>QZ9F^!Pj3FLR9Hflv13Yop5rpL8sGU&|jn7>QWRXWvMVo
zN|!@*GM;gGc*ODHkqaga+?>$Wa6=v4ExmHue9j7+IN9Wvqx;O`rB&92OU)Q+Dz8?U
z7Z3501gGryLmh|_UdD;)T-Ew_#|p0`Bf!X-WfTnw=jFOXcEaDEzm%Kr2CHmoqpN%J
z^1_2*%LADO+*DCA#>;4{Bq(k3ajK$p0?;OJw*jBBvMP#ta#>QTijZ7x)q&M_d@jpH
znb!o}SFPLn^+Dv?GXXoRx!9@*^~7sTKI6?w0dc>_;N~7XpM0L9X-YPp5g-Y%`GW!>
z@P6X;7+kSYRvN@o(~7{|M<gU=nsd5;K)A6^Fj%A8?jXAsm206Wk`_vpRn>>)wQY$J
zKh<oHR+_bN8+x-IsB6ur6(IzLDv86Gv{0<C_jvt#-{Iu=3}ZA(m&~&n-~85J^SvMb
zfV{{X6X)p@w^n^qEj)vu#4?VCqEslIp-cfnv9-O)yqNNMf1ll*ZH(4P84|}Dj6$i>
zE*2T0io!afn*t$1q=*QlrZA;ln7&M$uz@MIEA4faE}2eeR#s`nWHu&EQ$%2&ZJuRJ
zr!%rVM+6ar!3OP44}%~GLW<IYx&Y~{1rD5j1RJ(*&3;>D^6=X4^5+ZRU;57FWBf}m
zo%^y`c6s9!3*hHT^Pv{}Px3DlF1o6cwn7jmAsd^Ulx4wWICPtuMk#5>vmxibpPlsg
z?E1%>!<uW;&*N$%fI8~8aj(8L*~?|K@p8RSqwP9g9*c!%{RohqH$7J-=x08s-|XBo
znOY67vcxP5t1-17p)r=xtIy-VwfJnyCY)tf?pXlTRpy`uWcWziM+xp$H6g4|oY9(2
zE3Oi0XL-i?`8hXlTxV_22V<DcGbh6WlonQQ8p{X^pj0+7ds(U~>bt2h8DkiaN1UIZ
z(d+fuSl^=EZc`K!l+p+x$?}Xq1{7t<`T3B0ci-ds^*uVBmIIJYv-)C$d*xnWSdba6
zC#lvNtLOe|xW*OlKIJ_3)1p#H{8wv*RJc9UDqbGwnaa{u-+tr%@rqEr9{dMY@ME<$
zsT6)MEBvcwzZ-zac(v42`jXPxnb#HhJhkk(Hbh8t5E4j9ARNH<1e6dENPLj4#|(>I
zDVo~aGs}<F>$gHFPu-p|k6lo8psjJb$O_m2vH9mr7!s@vSbyOT^Ybz1Z@-70W)2ht
zSSC|cH8Sf>%Ga}8GzDDh0B~kVY*mUdLQpcF%sGAdh}Omiai>Ek3@M~UqtIo6O-g4L
z&QpUb53*|Z!%kpPO<tCsxw;p(Ambn;on>6Vc9Vbn&i7pSy=B6tvk8Ck%|GY$pS(^{
z7WQ-!v}&e!c%QbQMKy=q*8J4i9A<f*VjQrGqKLigH^_>dqvI3S1_R1cBSc6NwGkq)
z{bvlu_{9<0ntFhg5mIJI84*X)YB^egu{C-FDKQczlR0H!h2V5DWip!}g&-0UX^}I!
z7&A-f$RHpHV!Hixdz~PvSTX5Ys|78`usE?71)RYG5Fvav`dVqYG;fwR$mO3+P-j*y
z&RUvz;MO$dE|7m$fXL?OqNXP+FxNDNUA>&Wka{vrUo(65Z*2@w9J9W$L7C@d(+Op&
z&<d0`7{?fDZBSa<mr?|F3&NPie0hu4b?e6~Y~u2BH#e<jKP>lGo!tA$KxOq9eemYb
z*DO2bQ-wU$_uc;jacK@!eb!)SJ>BAu^T~1j<H(93@N0g>2L1%rZPk3QV1DEC%V6=T
z9(VQgFP-<FvQ_=44SAl^?{%5YQ_?h}+i8)eDTBd)R;$g$Xv$<drPFRLm@usZFF@&P
zQ}+wIEQ=}+o|KZ=Y{uc?AyNc%+C4t?$<L5wg;QDr@**V+0*p4Co}TifAN`1Kw{3x)
zlxS_)LIExUvXOKf=iXzC%M00Tjj?J_mR=?-1M^GYURawAl(u5r{g^%zxDbn9xikw_
zO)yi_Q-1vja72MXD_1_u1@93^0vRp5{GKHcjf+}~oiR?C1j?cy%?jr8l-YQK$_r4M
zNCvdxgeVBDiColvn<5AZqL2WCbkNBN3ChY%Yw=s<-moY$pQ`&BOrejAeU|o2z|J@@
zXaAr^u^t{dC@Tmc2$9_$og3HLon#b~l<~dCh|-~H+!#7a)`#Ic7jl2a#%6o@!K4sK
zV{Pv?V%8vZ$>ivi;e&^)4f?d&EljJ07Lqd0D9eH{PHJIkya1}7rSem0Xi~=iB7FY4
zB`0fP8NJyoWp`_jci(=G$B!OU<^|elX7dT(`u1P*!#93tQ;HaaK)QT#i_OcC<tO&+
zsLiZxsb{SUj8;TpOcX}Mp<=CX9Y>2IL+g?tkhH@#&)vSmK@@R%dP<(#$k;(&B3vz5
zOJ`g2Y#b#7L4*y<FOf1Nia>-3Wl@slsdK8gj>T!3GM&yS3e7y7F&bZx=Q*u*n`|~?
zJe}C0*isP35rr14t#5%0D^O>hU7a>efHtmNPX&5wO<4$C%@5Z*3v8NyZ?XUC4bW@D
zl_Pbv=oSIfWgj@<Obn)yL-v^Jhzj5>1D56jUl{A^+*X*&rw{=4JbK~2{j>Qc+SF!!
zU-;7|t=ReTglLmFQ>;Ln0E16G(k$iHyzVj(YitT$3wMm2_kn6Uy<Y`#n?EmslqdE^
z{k$K#8B-+#H)55!<tb0o0C)DZZv3ekiBN{Me5>Y&XCmCZk`1MSCo2fctmjp8_(|XY
zFl~`z{HF(XzDc*q7Nw@$?ogH`smjRnoc;ZMZr;4f+WI<Q{Nfk*FMs#<%;zaV5YTD2
zn9WnlqC^Nm7)bKGKuAI0*r3UDMp>4$+im*&K2l1X82RW3Awurle1U$yN0P*h$73Q_
zfk$gap69&#u4U}PFyPwmj++xc2n51VRw=EFu}b;seysbsDMMC$zVh3qPSN2Jn7Q&`
zd#bDUVGkld8r5p`<txTc`!qM9C5WUX3?ugnfx33Ped&fgw)bkZL6wFgFPKed4A0KU
z<|(SQd3SrQ4oMsnIg`83hZlq)Q5+Mu6C{QJYqhdWl<%5n1i)NgCr!n^*$dCIo}LNV
zc~+_!h2#_I8L;LvW&O7@WiVk(w6;d~_Dwb~Cd`Kuis8hXVYP8Sb565xl1m7erlIkG
z)_Idllg@6gnlKDpM@wQtC^WOfV@7KO;$8=Kwh3A-n^#Zkis`C+NQFR4S#3;aiAt^5
zND<f)TuZ4Pz~@<$Bw>pp(>%EM$og3rgVH%adgF(@_WEmNWrnCkd&%-@5(1l4*@!A(
za~N6^cpS$roZE(SYpqaaiqQpuY}qK*vP6j33KQqAhJY|i==1_^KldC-E8)?Dha4Oo
zB4tDrCFEs+09r}Q=A(-fWDwfmi9D~yT$LqS+X%HJOi0r?S}U?NLu<`sI$<`OkY*_u
zm}e>Dv1RIlARrDCgp81cZ0}r01Tk7E;yA=Okn0z4Gh*2TuzK_MLakoJLiqF^eoigT
zmnC+>|Lqf&`b}0hHbj5?BBSBh#f9Y7b{y@rSp=P{7%6{0s_9?m{^2xq!IK#)Z;Gf{
zIqC9xNdjr>B!_`7r(=ESi~yyrcGd25NZM`Y)2Xe`Yk31$aG-arg%dQI8kp66SUV3x
zt>87Ec@+bCm0#ClmUwTnt<9UhU5cdr2%=(_+1n2Tc$Vi7c-E~xrs~`3_v(}C|7v4f
zy1(lCjn<5QU3DMA%i9&7UDZAaUcQQv{WzQ7yrr!eNfZUlW>dP|F70-Q^Ye2i(>WIx
zBjPw=d*>QEJKLO}4=M88M&1gX$-f3W)_2(|)Mb{F=LJ#*wA*Esn=Z>z&dyFbJlN;P
zjoY+ZF{8023`2Bj89XT^<MEgW4<68NC#<gxh~v<uuv$!w<P|IbV9Q;%z{D>00#6Gp
zno6;UfKMEu%Ocg?i#<pE;nR=vs=2!Sw>4KUf-|t0C~U5}u*z8{T!2*#?EK3q?Y>N)
zld-xi$x~Yusmu#<ArVTWip1`<VMyu9(uGlE=R+6}IxG}OyJ6S(<g((sTK+cyr@<zD
z_sY6^xkv*$A3jE4S8t}G(#)-<VbQ;mxGeP$C`q9OI!cJPHmynaV$7o-y-Ap97o?yN
z1a(OwaDI=Chwz=VLb~`UQoq3kS`rF_R*LN6g7Zg@2-do^TM?}wpi0haOvNI2eD9qi
z?XnPdb90YuJZO-i1-DDt&TZNh!E`ofdwq|e{P0ai=OY_!V>AZ``+Vo`{+2YKI~_t-
zPvDy`rxe8jS{t5!{x&zR?Geh5(ddHx{e4>_R+SWu8iNQsA<800m5L<tffFSdA9N5P
zf)MHa=yrB?0N8)L&%xmVLP*L|(eHE#q#y_kq7azef*z$yd`57ka%(PfGg;}9G@Ub^
zTrj*i=i*{Wmgf|zB+E1A^Av-j-RTiUv5h)YhIYSCyWJ-Y6T$@YJahge_PIi@$vRDO
zd9$x(q-e4tHVhplaIPsUO-?l4^GQvc;7v5ny^0ADh)eW%&1@|S<yDJ(;ap}p+Rb<o
zWbfY?t`fJ+cZH}a`yjAMxV6^gMegpQc785hV%J{3Pp>~<JUq9Nfm)(8Hu}}y?^4PZ
zsNZW|8+XsuMH+Xw%vAcneNy2jYij41y5wtQv`uxgEG&KBLk;vkY+D9VA7?X-@inIQ
z<EmLSHy2Dj3zk9Cvw}}M_go)df!gZ1{n7NnKP`J_sx>;F&skqzCyHW5qcM42FdC0Z
zl7!8zZEjq@j#irCa9GVdqw9c%CLj*O5LuKq)l*?BO(aQT<!@OsnT~n9|CpVfT{@jE
zLRkO3BF}6B@*t>K!(Ok$?OQk5*jOVF5~NvN2MfyNA{e*Fu<%|rH>zXB%FnCjqXerX
z8%JE463f7HQ88=a!>67uS254azc*k&rQCT%(x$-*ovAcHh+yG!yN>GbRG`ii0rt@_
zTBAxuSrjM(nG9_AXbsBdZq>1^Kx3n(BV=gHf}$6h%4PYqEZZ*sY^IFTD_)lq+&@iw
zHpWggHqwQ`cMstd^}v8E0VOCyL95##++3r7=LUo0Q|9*%F&G<hqzpb$M@1YrCU&)1
ztk6RI#_w*-B{)_D26SF9J3ZxWf1mZub)rretqPPXoOv>CY~tShx6BCD<~Cz2ljF@~
z%?;legOmZY*$lA``;QLEvJ#~<Ws&in@BA$n;~`Rj@@H-gl9ZN$FO@<m&0x^yOJDo~
zy>^duK4&(YQkEq}VKqOcl$$Wd`sis}NkVH)C?rNpY?<8vYhn&a;ua=}0E1GnxwS=>
z<s2WLaPR(m>|MJ~na#=5lq7Bug^4X*4b}uN0hHBplyycf%bf9a%xE;?^z4Muc*J--
zv8BRtTN73W0ZGz9h`^<%2tXLR-9DY}8bXBRMP}>H<~9X^bm=<~0wpwBSWOG#9olgy
z&iSF>OUAkBQ>m8D^d&9}jjNc7X89;Pu9v}rWmal|tF^IV^IFuuI}qfrw=i#;*A`C1
z`0t3N`TvBqyeLp;2mvclX0sWi(TJkR7rvvE?Z<Yj#rpa>XGce*qY0%plu9G2Ty@6f
zv$I0YCK<O;Y0Y`4wWf3q>$L{D2t*c{h}D{GGet+!DquhNPUMUI>8%DI6>wNQ&$DaF
zSGDGN`&RqyHNl<zTL`<Zec;yW`>Wq;DhsPVU$_T%AFHJ5hxUzg>+ipoj~#$d3ZG>q
zQE2-lNn&<)x7mMuz-&6@*jRJTm%sc~qA242{rl`69MbLBB*I=6$5Cj1cRsxBRzjNP
zl&Y*ut#KS9r5(@v_uk{V+s_k4F-ejjgdj_;n@*Cn2!aVGCug*h7B_F)q}6KK3=S?Z
z5ghA^0i{-sX{BI!&_ks{+21Nww%SoIwMAfPe8;Wq&!wYQ^tY3R#^6-zOW%Fc_pdz0
z0Im{|PX}TjtIloeYp~aa06a#yypK1Lr>|5Ihas&fM#ZsJqLr@pXP<fmK$Moysbrqg
z_Sg_h@%kL_3v$?0YkYaXSzeQiq`;z5ZLg`SdwH7nY>Zt)dok5qu$^i!b^`{>*lCc?
zyi-a^FzC_0agCdwoN#b4XL@?<y4?e}dTQeSGG34iFZvo2ulc^fnrpQMYJmf4#dONa
z{yt`Bo7P~BxW8uEC#@kc-tc9?k1H@-qx}jmHnb+um_!XGLTkp8F`agYci+9oe4e_R
zuZD*YAMoCTyBMR<!fufMF|4Lxo$!^iJ{Ca$&p-d1b?zOWGoQ_vPR6LRK&g_lSS*`V
z6!v@diVzfqrY!4nTplDvQG)3x^3>kK&dv^j3^_bF;_bKJX1(8|-|5lsuhZ`hFvjM(
zl0p!K02-Csv;b*3XP(YEAD(e>alzTekm+nfI-jGAK^aJrgrwa;%78G6kTS#wL6&8l
zpI`6~|M*>A{o<GD_gWN1MifP8Rl104$2wuEd~vq;d}BNEpa@$ot(rrkX1$gfJ6Hq~
zuo#`$R63VUhD{+`u{x`K#wsClSwr)!5*Sh}%xq(t+k6wG`V(Kpu=s2G1l*QINYm6s
z=q(<vwWif-v3qTo^OIu^r&Aj+p&$rM#S$%oOkDxcX8*C!56fa&RAW$mPpOI#Tg7NL
zg`O0($E~1Vcm6w`b@c%AqiFuR7ND>834GskESSl~!u!jBzm_qsa#CZ{^@@yls$a#<
zT?Oct+mk@yN7B4fwKVttsaqwkfmW-<#l;0>S#tgQbyQh08jYAtXXK9_@zP5#v$eHB
z5D1hin9t|5S~1#a%A%k)ktwn)B}ro9NK%$IIAb~;BSa4=69j^@^Hc8KdlxC;=FOWN
z9v+g!F~-;&bFEg3qR1HzFZj{xZ?L|;PPn^697is{WZ?2SY7~BaJP=#mOx5qkf}W@9
zrOWH|!?uRHRsF0KmSs9`;XX~ZpVz>mK8;aLVCJN~bBpqYNWFEcVp<n&On9(>Ux9%L
zNTP^dr;Fa$pw()TXBny}k;d-LK_lAM2aYzGY%c?a5|wN+u%PsMH=k;KFGc$|Em6y+
zc&|WPB|}v{|4-ANEh<)kt<YtWIl^x!cJc*AqA?VvAjdK-J}pYv?jm+}5uYfTo}RNm
zyFkt6gn_W<G<NVM0Re_Wlpri4r-X4S4=roZ6!sc}77BzyNKK$^_9<xsOc{`$Oh_Lc
zkZ*62Zw@HR45JN!4(x^%XtWd<9e~dtBycr5?E?zO3QFNK+`DLAA<?BmAbIfKeO~?i
z=lIDF-y+LX$}%O44EG+tM;M3bxwgfMofZ-VA+1(Ip3iMkSq)(b&pmgO-~O%N;=%p<
zTujcGjK>Ve=afd1m!+Gy0*qxx$|lY+AZ19Am0S#Gh#;UeIVMm@X$S&I9Jg(rCm~2W
ziVb6<ogY7b%%g*S%A%mt>9W4QNw?cWN=Y1tl~b=N6}l*yWhv+9=Uj|0xEPO_&Zai+
zQy__w7D>`33L^voQ4k=7hM;7gYsw%b7cuX?d!M6|3%>ltSGj%ry6bO{(m()ODVuQF
zX<X7(KCt2Rq5^M#O$=Q{j@9|atha@zAcj>07E^H;aPL_Ym2hm)s%9K$t*HE}6qi4b
z&5LEJ<T@~>xv&@5Lv0z#Dsl*H`8tf7?;fn!`iK^&H(10<;7KRqWR;Xy0#P98bvq3D
zJ;vh+MnJFEqtog#=&vyt3>clCF}XOSoKGomxjHd64UW>Fv59q!1<iKtYov5RPts<7
zuhFAKPzPN2M0v&ob8o&-w*GAOa;I)`1g6ri>aq4fMyVReX|2%e!`F)St@bPYXc)Wd
zoSw}x&1b=h8Ydrw2fBK}qOfX4E`Y`>*t+)vZYvqRC!F(X+A6c!$K8am^<nL%Y6wF~
zmf0#fQDn<0?(JP8&kN?M^{Kmm{~kAQ-ehZQi!@Dn>#escXL+3@TwIJ@guSrxtt=~+
zxGWc0YlA|Snxms5Ha0eR>7^GjMk52e*LYKJ6h-7k!NZ4-xfqVvSX;M&a9UetnnmTw
zy8<otY2CNY3KSn?xIP+K=USy;E`t1(R}t0hu9e%R=d|#`p*1geh4-gkbk`8ZN78yr
zO_P9f*D_v#5`-=SR&~3CVMq}t6nSptlDikL<k?IP4%{^!!5EjK$c3}lY+p0*tY*}k
z%Gp(~ifE`5Vf``t*f|hqH3KWuy)+g=Al*KeX+<tdkcKD_L^4E32zq^zYdfsJ^gP{*
zGv@ampo?bFFgsAv?{H3-Ye0#5?~`tzlox`+?mWUMgeeh7v=K;R#4Kn2=#Y!ub)swQ
z^ve{ZG%~OtR0jqnU2Rh1HmauC#8s%U`->O04&Y%3f{4R|1Ma@_E?@q_mzZ2!pmdIw
zii_!(x9`5qWInb|@>(EeKolkPdws^^3zRAl0+K|qx3|mx<^TMDVtad+H-Gw5=Gl~a
zHf1`SQz~W4J~<Pho$K0x2|KX@z!-_v0q?x?kh9^K!P*YF%8+@6a6+Zs?viG6Vka~P
z0R%}voFwc&dd$(;33ngfw+l+(>hM`sBncx>iZss{PbOB2kb)ozN!l%1?KVLaBV|M&
zLP9NwBoHV{Q&4J6LCE-g!q(O<^Vytl{mr*|<>eRo>}OtKePcjbq%P62KszeC)&U)J
zO1()MEHGe6vo(#M9^B!73d@w%^1_}^Hu9u+lVG!1b7{@sB5EzI0(Zu3*hX9uw2Q1C
z2#xdG1mkvf;j26tP2k02mC(~MXZU9o)q{F>hERqial)YAr_;94(0N`kpQRXMXmz?=
zzj1@x&pn3_hByqkI6vmm{ri+gp|$nHQ(9ZMRPEM=zrMiUp96SaANOLqUc8pg)=Ie=
z!;8Ysv)v0XH-H`h8=O`vjZ11;_oJ_R;X%;lq{+|TDh9Fndvg=L;`mGFX($)|eis85
z8goO`pD#`qA7NadV^w=UKz11r)U?KGGaoD8Ti%#1Uvp{weT?940gN^<LlDS{6_&z+
zF|akO^E~I~jq40A#?0m^2F3pVKI`l2?CflF<HmLF-Mi-kC2Yi_an}%7g{LS=!q5gy
z1);r`QYB>>5Qf%W;9@l7^z4-BbV|3|V?Lc#0ccW4WGKn=oY82^<H!5#Y;BRmb}RN(
zVHQl;SCr(cJEK;5p1uPcijx1&gTw}aayPLg_pTm~W_uRj&t(>V@tmvY_wwti^Det?
z;K(6_Bnm@Xtrmfllwn8_Md;Ehp#EDDE2BNQ6AR28;6bH(OF4$k`o!D$XPLiXK?XdX
z!sU;BHYMxX(w#iTAT<TIPwwYYu9Ol6J=ShrXZPeb)8UZvY-|y)r<?`0OowhbT>JU7
zbom<Z+yAAo8?!aZx&R3S>1e{);{)2a_9(B<F@0sjE~G?DySrLa?(*bX>;MaLZ1f>W
zW1Vwfd+jv>88V$r$+Mgy&kzBepPzGla!g(n)xgAY#NM?X&d-N7$5A9`w?nSq*y9iX
z-M`}tUwD=O^q>AWN|lpkIcb)Y<t0U40$!^V_-b_qszQ6Uua@NPw;%Av8*g*t`gQU|
z5ELbWEC_>;C<@Uk!f2bEIF4duFF;(k6Sp_$Ga8Q>U0iT}e$I3{B+qm5JSP;wX6OzA
z!pMeyw_7dRoeu48mmsuIL`XwmjAdj+JvV|N;PHF={Oxys#HT*{b-w)NuaZO&Kl|BR
zoSdBSr7wP-+qZ6@RB3%Eq(bRpY1CIObpP3EzqIC<HaDo1t=eQQo4hdZni^@f@z<(J
zt$AreSzoVtAI*D~LExxxumL;5ncb*zS6zLZ4d8$UI?{pgcDqd&N|GdDXXhH7P8Y45
z`BX}}{Q-M>*NNkht<4RlqapL@g!8i#N^K|#Tb2%R=|2o;jgVe@Uh0F!v(wO6y^!<F
z#<Izw*)Pra7R<cvoPMKSZXcx?w&@eoe5S>+epk2VP-&wbxEBj^&nry!cn5&-&)y8)
zkyXjGW<M{tRRGGHsJ$u3{#*oJ%l@OQ*vVDrUS2yj8}UA!^D(q#mTERjX|)ofFl0WP
z({4o+g<?L>m`o;IyLOGTEQ#ZQ2M-@|dN$;pci*8@1z-5W7r6cC76%6h%w|*P!(}7%
z4G_nYEVGH`!%))cSU=?HY>Lu~t&I+06j2lz_wT>QU;lN%|MI{5UwPy8H#j^ztkQHC
zG)bHg1p$BY7k|Ol<~rSOn|9kK(2t_n%AF!33`5G|vO7REjU4EFR%W$1r>rGFIAxbh
zibOT1{Ci8+(iO|@=5K`uua>c14m7W`nsN4EJy`I&t^1*YF@&hr?^R5;F*bT$C|gh6
zGpK=-=qN@f>N3u0loPe-Rb0SEP-jIj?q;>o<V8lBrnXjkQB>E6qKHH$Ht})Ld=6o)
z8p20m3u}rK2CbFduXRiuMJxMqReNS&_at!V2g?f+mO3<$650vT&N^Gq-)3@l#>xCH
zW||=#Sh0o1{b2ZwqW*iaiHf`NrI4hM1j1t-TPrkAOQxqo(z78Z%|I&)i=@B=-UQdY
zSKLO@MCBl4Kwxu`AcP<<a`qqZ^TQwhkYD}fUuASYv8;#&(43#0Q53mlUfhJZbLTcE
zCnt<27YJ!+w-Y|~>6iH3-~Ana<yU`&Kl-CTW`F+?X=-yTd3LZ`2vrniLq~J_L=eb|
zq8?9D{^NiAGiLLQfBmn1kKVO)v@+yrMyJyziCd_$BorYsmK3re+=vOnkXE<LW|om=
z8ME1pvomWF9*ss!M<enqquuGy>-UJ_m|nMQ6E}uV>og@$B8?%yGh@avTZYkS#L>|S
z2Pc2Y!$*(!^<V#WZtPv>@#6zV7bCv-g)i~)OD_||5rzy<e&ui()Lmv1{CvBrtr8CY
zyVgAGR0ZiP&A$Jfzf7a+g|azU2`_K*H(2&b1G-Q|%Ef#?wY;|XZ}S@lc5@JhPN&1x
z))t!^TNvY+MF^7?8ylOnS_xqI?5m&W=y0E({AkX2GNC99dExSDVNIRF1u$5`e#s}U
zp<Vja4OaoXRe)|)zxw6{x=mHo7^@ZM|EQXut3JQjhd$qv6p3fJr>AL6mg2H9!@}nM
z|8GqQoD0D+_)yGdGsfdF{eGXF?H%qvct9B1i2Sp&Gw$BK%V$3G8H_O;9v-?hL6XU2
z?#hH|q)Vd`MK+C4nx-x*ge1!{L}a69l`1$q+UNM>kRX!u`aRNlN|t4A9U7D}v|1e=
zJ=~|;>F~)<zJQQ|!uj@Lq*Ew9dc<5?FzjDX?$f7b-d63=AEw#+(k@`@Nv)|6u9CG)
zIb)1X(w+!`DQ&L6N`{mQrPadmfv=bCa7h^2y>20LAFBZ8g#~kInlhiyNz)XtzMidC
ztGcf+3>ym3Vvt__cMV#-!v3`FBW3KYX=PFK_D+aODG&l31r(i_-i;k@jXz0#I%1R_
zpi5gh!P=($`KW^rmzsUcI_FtU!XjO7VHd(~ml6oHK#eozC+FnD5oWJI3x#&nsJV2J
zD<q$`DJ4o(f-z+CIq%(jmxKMsbUR(9(;3!@(ngPth8H#=av+JKfW7N`1i~;Gj|d~f
z#>Rlpe)co`_V4@_U;EnE_{mRx!jFIa25Fj-rPjfAI2@CvWmV?J8+exQP)h7EBBa!c
zQVULoQ~qE7@y~es?R)(DfBP@_%2&PwG9b+rK^PDvT}mxT)08l>AZ9IYA>)=aFO+0i
z&i1uE%F;%GogN-=es)R_hV=V=qBsKT^dHt_Y&%P8Nhl-A($-E?CSX3#IX%6gP&ReS
zPu~17C#NU;)^GkcFTMB@^R(dG-~Kz!&&Pc63tyzu4Qh1bnIf?W4E&6EKU2}d{JCP2
zm98_r)9k7!PXGN??}KBTjJCd?ELKstd~M6pX$;y3TiVMg5{;{ocr|0UY+^HpMxI4c
z--|>3(&e|aR%_uZhya9aZtn8ASHHw;Hs{SZe@w0nSy56t6Sp=BVT6kc2V*V^P5XW=
zJe%OIsqn4puO`#Ka(uM0E*FhHtRixq$>XC0b~WqxVNY}Ud8_$poO<T#W7==f+#sEf
zSuu!D+g6#NgjhU-XC#;Jckyrk{j0C_QGnKuqSfa4Wp@QIMw6ztUhMAfE}L7M^m;v}
z(<$@04dxmQ&bW5%8hd+t2q8E)IH1$%V2lm;D$BwKrLkxVG{#UCig`LC2_wof2W+s!
zPk!<fZeG91`uaMLr;mxEh%C#hxf(^0yH1O}y=(OPZMROGGE=3)SbWwG&5k;GU{7Y>
zXr9tr9h!Yx9ePQ|`m7RW={(KgiKY9mPu!UCwP~*YyuKIhEKExK^<kWUzOMb>nle_@
z{I|agV?Y{%GN!IUD*$Qz`-KolEf?-lx_k3hw;&8HTUR9yck)kZTbU%!NYfdU$%OHE
zgw}>`w@a_rB?!D_#lJj&v&w-GHc`9~b^0nlub(BI)3f>5IV5RZ*m30(BN}<B93U`S
zBcrGiz<%HrV3L@y-(&5@HFjRO&3rs&el|v;oijNLiG;0CTb|k2jRxuba}3fIbhf^D
z#-`4Y0YM2Aipk*#`Djd@&Iqv=64=d7c$P*8dmfxH^Akl%q%*7JSw<L24h|pj=1+fw
z#L(+@P^Cg3(4{7Bh4wgML+z7x%#G{U_|Y3bMk~ekwROJwwJ-C_zx>O5@e5xfj#~Wa
z^&fLF91#Qo<MD_*FHxoIYhTK#qP%SrKp5CMzQVCSJ`wgbEePWluf6_L9zDFzm%sFR
zUU=boZrr%R?b|l{dbisqOgdE{K@<caWAZ!$At-X2uSpwCe=TKwV~fdX$ZS5dOU4UM
zZJ8Z4pCg6Fh!80g@-jdfNr@p>ii>f|<NcF*5*x+g@d5weKl)F6?JHmB7r*{XwBj~z
zyzx^;qbXni+UMEW?BRW=?0oW!rGM{oj=N>nVi{B|vvw8ZW}IcznH;L9B@aH_hTQzl
z^6&P%0-xW}&;MpbR3#{d&lLzoX;4~_blSwRO_iWgh>B@53&6E`9>31~hGNZdbu_VE
zx7hh4BvBHB19zp;#H|iDZr|bM(FNzD3-%8mk>(}yG^0=@A(8d>bIfT`7S)<J#+I@3
zLbwJ+%enm)`>P4$7Ss6P5-l$Y&W4Go3GDoF&0vZT)0#l_qiS9Y_uEWmlC^(cO`23p
z^|QC;9A5-qHM8nJTfEozKkud6T=>#|J&<RE8mwX+2%LPdV_6hA7Z(>?yLOF!f56t(
zmWymnF=oneI3!I|Ha9l`I5;>UNs=mnY&M%w6scp|?A*1!$Cgz}vlQLw+JH8|m=-_(
z`Omqrcf(pW+?>wyyy~wo3^_YHqucFpadAPXYZ-L=EZ&E7)tIapgX)(><zV5n>g!FD
z>C$=J_p31#qVaR-GxxSJ%PgRWXHS?5&3RN^!~JJZ2wpC0TgI|fr2}|@{oO^5hM|p;
zwbdu=b(dv)bt!|ZF~4Np5!QktM6lRDdeN;+1_2^iQl4GVW~ppiviW?@bUNYu{M^a5
z5-=o5LKud`aRdYvSQH|_S5sO9e*QV+#|P}18adSHz44<KiY%gh4+89G0ue_9Yd!im
zcj+7-GfOj6rU;P~d9gShuACnNkT!BqTfZ)euo0O4HjRz76{V&aO({lWbdiJ6)!es!
z`n5^6Y5T4=P7GO|^1~nggipM1#}*IH#_aB_Gahb{rc?Ueb+p!W+HLY;UfuG>+JM1E
zpER9Q<SFYLUB349FY)!S{{k<)^a}mn8b5pM=V&cxwK`;3#&nvZN|?_xyNM~U(NsMT
zSP&8jN$3j}OY2MHO^Ugze|LB?<d6RIUlB$TaTL<;cNwhpS>IS^Z7`tQ>(cG_*xK4+
zdwZMB%?<kf0Z|kYCLI8xq)VsQr`_&yadBb&2ZZ%&DDuKa6j7qIq0|N;LIer(qTpgY
z=HcT<{N!iva&UYBxZ0$E6o&D1#NYmxzhyib^BceV8>|i1dGEb<nNH97#b5XeH*Ve_
zif{|01YsHi?{ZZF8uRZ-z^)0#jB%+|giW>LxlyGxwv?Pk>N<I3y)LfW32Sh$l0_7@
z)Q}O<GIbils71FsAWmA=|4s<2#4Lf(<z_X!I{%*s62@q<ETbq2;wV7`P$=#EeIi=z
z9<3~=-RaTk&bfX26HI0^3ay#Wb1tS6wzG^lj;$%c2~vMg{<)OfO#SR!26L4w0kw=+
zsQ7US7%w)h4Vv12#sj<Nbv{y4>_>^}8f}%Uz-l*>g>O|#?$e(CNuM<t!>4Ue!^nM1
zUG{;R5D>><!{^EZ|C5swZr!@Y?(QzHyz&ai$H#ym&kIgZPkH$8A)o&Ar)jkk-hTUS
z0M^&nF~)FqcE)U$Q4|_uY!F4O)glOrDz#9WrM4Vn7+{Rzop*lD7eD`bdc7VS8ylRR
z4G|(Bkhn@ahRI~c@zDwQ?>}H`dz~b)RTL0jVP4huS8a+-P}gLup43;X{39FZci`iR
zGjx@Fd&YLvIqce2lx5+3*HV0yoivTfvTo&qvdDgUmMv&tQwxs5ARrX7mSfVo-bfsH
z6S4w+HszKDStgJe&$>CVrIf;8T=I8CmRlF5i;D|RPfx2r3S$hdR*N`}X|-CFvgMQH
zd-<)ku73Bx?_*``s+=rWh!}5*)mqmOq=EX4F`#X!EF(j7E2e#Ii>*&SXFKBIA&IgH
zj+>yc=~&!sK3JgM*r|}UjpDPKl#u~wO);4<J~<_u&1kEV5P@_Ip$MNmaRJ`DdzaUL
z_<cU{i5qNfuA|isuYGqyzu)5MXrIkbZ4(C4nZv9<-tP7`tyat@Ubw?Q{NwjYl9;c3
z^-KKfulx#kp8q798(W;5ob$c!zD8LXv=U6GbEeZ7t)xd$*eG0Q0HC={0|~K|gfbuy
z0W$C=J#ACOC{0@Cs4yV193^FG$W*~N8?t{g#F+aa3{ezR3&<d8B{AJzm-V%EHaEA}
z+1cXuty}Ev?a*z<gkg(V#<ql+wl!T<S)i4qC<8KOm`qZJ!*kxff1h{m-sNmKWH$GJ
z+Lx{ZWi)}16h+GSfAC#~!!!QXAN(u2?JoQK`+W18f6Fib@-K4x_DyTJ6g9=)Bt91A
zMq?vtnoXLYc@3<Zzku<{jvE@ARU!>uUl3qTgJv;|eYrJf?6U7iHNPcR6D<`%*kQ1?
zL2s}|6vtMuTgmf2Gt=IsbpCqHYuX~?jy+QfN~&}tK@bu}F@wQ6S(aJmuvENk6j@Pn
zcyPdQc)@r&q18@oG@t_n0*NY@&7YNacL}g9Gn-9DugTo4sxaaeMWq%^vGp_RMzY@z
z*A`jc56{|NWr`IJ)L-4~S{IJNct7a(e(j^!c>h~yevb;A!+KGa)utE(0cU4tJbwI`
zR=dSZFTKbcZ@i8WFrTLkhZByD4jBxrg8PfV_zQgJJKsj@k}wSE^*XGtcb(6jVm?n<
zTkFwk#gwIFKASV1jM&@VBML)CqY?f78b5vWO+fO+FMOFl`IG-lx7#C$6S6cV%jYCX
zLY`;5`R1EE|H3Wy_I7K(FXI)at0O=ETW`y<x@ncL&7gHjS!T>;Q^w;9ilTIksif6v
zJD*2eeZoc+d%rY65Ja{vc9qar*!{ARj}~cJvA_W>AzV5YY15rVuAFBqVhC*QU>R73
zP$-EhKsqaiC1$2fCbm9$S(a7J*?zxIx7(#CiYf(9<bopn^Ln$gpSK?SJWI*MX99N3
zoGe$VkCh;6ZeBR;P}n?8Qb=;3lmfZdW%JGr^4Xm1VuC)K5SZF;&a-^0u2*egp1IKm
z5lN)3i^pr>Hai2REV(#4XEGYm&NF)*<3ri4<9fB}dUIT3cMd{<5Sk<k==WRv!#}*u
z>p%KFFTeO1#*=fBR!FG|vaCb~hQXlEbGM%(D<^c?9keR=#b5X;zxkWL!HX}xNVm5}
zuQy<LKIQy;?0niv&d)9gWK5P7PO}!(^SU9nA^9OT_`x(voEaOnrH!W0kd--z2oc5z
z8KAVPCa%F!#Kste@j)T>=R7T$&dxbL8lv^Pq*+NI16oNyuiIgJdz)J~cDZ?dkFCv3
zw9-sQV`kGSr^6w4A06=U@fqVuN}3laG=<VO=a~pl%KDG^Qfk`e?2<Cz-Meq||NdY9
zpZv~m{SLQpJje0zF@N<}-{hBm=@)qE<xfxxIq!pJu(bp0B(BSB+*N*V%kP(Fu>y%#
z67X}fntO|8tS14BCa9^-DQtMWE;Y&s!YHEOU!y-*BTm{N1*I-94)iWFJ*&-O{^lAc
zWJ9IBYD?u}L10jo$LsWZY-yjY*^sM(7higrD2|y=r<@PZXtiun>_~<cJE|Ih%>Ug~
z>Y9Lh*(})Xn<lXHW3pU%LK%ZL&Zo9<9`E$L3P633)&$_LD?eIrS1|`orNUJ5Ml?j7
zHN}6h>ou?Y;hB`DyPiAxLtX2`G;col3`fmG)n>Fj&pA6gV{L7X+t1zNx#ynaop;_L
zP3I&@z;JlZ!NCD>9P_D9eTsg+UoqZ72-erv34)N>EG11-%Y?@<VHgs{5vnLDl_D}4
z8CYw^$;k<)r>6kKaY7hI3v=6+C9}t!oSd++u}&04m&R?mpD&F|MZz{&yVVMTsQ$ZZ
zelLQptM6k46Z{NqSspK+$Ftdft>^ii(ddHNY(|#lXsrl>uyQ|%;}~zf@L;}*VwHh2
z*-8Q#5{LlfnLP)h!XPA!Lc$7MN#Z!6-FCqbLXd>cRO=#Rn~Cnd8P?AYs8ctUWm#pH
z@P5I~DBdQhU6z^8+MdxA+bD717K{y9aAqrGz$lDSj`DO4nAOesB&2S_1tDRlOSmQ3
zeJ*4C_=vOd3{w_1mz^Uks>skLP()Phfsz7+KnVzi3z#q<O@MT4SBOL^O*xuSj7FGr
zj!-2=glJ(m5`+zX5bmCvIqyMm^X3hnzjKF!gU6g4KjQ75y~)M!3}Y0>M~4JbI=>v9
zWmHsO8^t9hl@3Aa?(Xg`>2_#_?rua{QgW28p}RpkhLY|Ek?wx)|6Pku@MYF;&$;J3
zd+*<7%jyP-Cq<;y6A7@b_rDq~8}2iiKH~t=RKd*2q#_P`C1Ws0tQ-|ET-moK1cNKx
zJO%6V9AuCTAsAu!A42dDKPcO-6tR?Ds4a!8O4`DgVh<qu%TtIGr!azntn#?M%!&V{
zd;UpL^^BCf0+9jN0k^g?5S7k%GXcf|-|`cw?Xcz53yYZX^-F)?7nU=S2+C=LK*r(K
zw|ISK;7V?`P)Wiclo2~qJ)oBZwb!Wi=W(aAr(;4QwIcr`r_Q+PXtjX9mV}7v<l*&)
zD9UCCUKN*x9jA&m%`Lh$&es!YeJwJ;k$%$q%$hMsfTEf~tmOC6vk#Q{!d4(QCce_^
zZ;?X(yB6|MObODdQ|Q!3jHoY1Gx>2U7_r!uq&+w@Uc>JCKGj?KzXLSDr@<_R-QoN}
zn%g);Ur~rB58YX2bPSTe^76tzo;gFf$1{|R>J3@s4Hjs?ScrE+B85lN&=4-FyZsWa
zz-o5ZYlBL@n4&~5noyUFzUAn3Nab$a$H)I(FUXrO&+olO(=HZkTFDsA<Zu^Q7@8)$
zUq|1x-p`l@P2Ik`^}6(rxqYCnKA#AA&mGD5e&wRhmuuczc<<l8*mG|q?XN`dzSbLH
z12yW&$y#%Bb5Ywrpiq`~@7@737XJ7{&$ii?Xi7jG$)m4*LmoY>WV!gGnj9ma@c_(E
z%K1s{;Px@tos}ky2JLqurLHn9uBxhPi^q`_KGY<vYSip+()*P!YG>4m#@|`4iPK@v
zq9fZ{4dp(2{ccsA6A#3gW>g?0ZJpQ!FfYw$MOR*8Z%uuTAzyq5y3vaO<SPs*UB+CQ
zD4b|;$p?#22HPB*y1<TtB_ZCmTxfW8V5>qTBMn77P54?)T9A_G-0dI=X*0{Jg-`pm
zf-kcfih4YXa7WvtK6?MYk~$!`L~nUmW#-%UTW9ehn=Ca|qF4qN1ZoE~+^+M@?ziFO
z2(8Zqu5D(g_shy^YVrxNsL!z)Yu2zpQz0x1c5PqW@SpA}w)B|rGB$XMU!^G}(Bjym
z7s3xJf-i{|r%-nfNIPmmzO~wbLtGf*5QO;Y6jvghMx&6-l{0s<Ng*yr1Ml)LDRN5s
z^WP@0Q>PeA*){zja1)<AyipiHbtxX78cqbZ(fEezPedTMLAb{RGWIOk!m9rEnxEMI
z-5C}GhNa2Rk!f1*cZJs!D9|c#{RhmkhrnShMav<j^m0YLK~02|h-zD7Ww>T}a_oh3
z4Ky`wRG}XWk|Vr$T}$#V6$uy`vg$VWqC5s%6WqR{JqP_XMjV&}+6ZS_Zzr5BuiIV4
zQbEbF!TcBkf+E{$FSlabrY~MW-Qy%p@^!IJ0Y^gg4tKr#VsSsoV+$-_-J?qFvIPg`
zoMOGzt&)Sb2)<YPKT!{8tlL07wY7X#s92W;9zH8s3UOpiQ%+EiCqh#5HnUdI$Ah_A
znWj1kFNd&#)l$s?GjSYm#DjsKR>Y&V?h_kCw402cpWpDl0c(Jdr%#FSA{em&d}16!
zs)r0#616-^8&0Epj+ol=&%ERr5UVV1cuJZ|l+^7C9IsZVGq@b7qjxvyO8MJI^kqVA
zyb`8HiZuGta{oQygBq=&!qD@<JHdvy%x1FsS%DUSYcLtRT4n{*1X=y<@w?W4Zu64y
z$z0a!pYx%$zi&x9-Hhg9UGqXvf4by4_I+i1RF+8<tGDus2nO;K!-uG+9k)Pl$rCUF
zSuo=#r3~NL`Z8d{y$8giA8$`D0Vj+$Ndo{$K}fMIWm>cjOYWsd#7{0{(3GE>GZ?dG
zu$=zF{ox!Y<W$$xG&nrWkfD@}6Tv|gGTn%=)AK^{d!Nmw0cVS!vUys@2H)Ax;%aQ#
zn~m)6b9jV`Ll-~UYNu|%Je%n6ULTP+rTa3Mxw#mugcB|qrRC?5E*Il-pWjnGCWUmH
zDH-Kqv~a}I<s%hZ;Wra5pebwMvFQnfSKG>OQP&i$l1oRzwJi%(I6119Og{`ZEUPve
zBJnYy6&+>|yZ2gfnsa~PfJ{so%s4icE7Ix^(=*Po#)H1-I-ZmT-v9i!3M8)${H{sT
zN=&AOPT6eyM*03NVvO3@f|xnqQ7whRE%Q0C3kzFtf507sTCp}Dk#ok{q}jMWh+uur
zD)NL^ZtctEFY4wBm%vJb?HQ3ms(nNY+R|}lc-h6#CNz<V%pI;}`P&-ZN)m8^6Fkga
z)LazklIeiG-hGAap1yASWkV*spAIOFju2pKk<ze5VHBw?&-C$e9?|CvwYttm)wWfN
zcwE3K2v}1JE@O}8Zvnx|LNI6Q)Qce|RH93YC_+1LYI^a^o=#~LSuaxh%jZM8KGygU
z-T_%+5W8k&g_E5gw-n2xus#_}95&o+<-hzVQYu|G93A<jTc=O(#IZ+S{byUTV)u1o
zdtI;TubIc=EYY<ahU*UmF3f;NZwjAJicUH8&R^khfnkuy&!Gb6VcM=hVq<_Hguo+Z
zgR-Rju?B}*iI!EUn5CT~)xRR=OL?4jFh-%}-)K?_c)&{8@e-_BY_C);fL!ai7NVnL
z=Bk|-{8LpGtb)b6w2Y~)a!y4b1OKrN#qn4%jDnF2%TMw#cO64b4QRM3<_+<v3Y-7K
zXvzmF(c)rO?pEq;Whf~~p|2T&v1D0QeTcCX>qzLGfbCPfpx_q&Au-qPxB8BJ6<hv{
zz~r^Ej}N+Ko;Umy22_$TcfaL%`#i(5Y0+U!$B$Kvd8f#^q(`Pjj7((yqnPXF_Px(x
zVJr!s<M;jO(}2^rV8?eqRQH8_-Q|<oVPr3xfb8h|H_y5l&qsn(#@)aHxkI#!-hnf~
zGBG5<|EsVXaJJihxF)f%9aua|NuH6)^*a#&%<UQmzhaB<n|@3b??lkX8Ulj=z#n#S
z-Fo$C+*=(d9w5$Jf6lU*N@>3Tt0`A|a5lPsnvZbJkl<J)QumKlOhP>|RU#$zH$#kv
zBQxqFdlCzMlMJDbe<l7A#@&({cz7}{4CzIhjUT>gMl(;46mG^s3wGe?Er`sFR5ch7
z<{l7q=R4B)Y%^K!{-_j#YMJ|qRElTvR~>cJJ_ebN<Ef;+H;#amJs5b8P7b2|D~r*s
z>oYb`9CT*Ue9)zwik7HhZ}0y3JJ<#~F?ZQS96vzX6(AmLSsu=bl21aZFn52<EVGN)
zahyPo0{+I3{~ZNhcS1`st88cME@k+T*szY++zXQ`0s*1&paVlfURUgo<IA*wPT<Wf
z+XdwHAZQl=4*O)*;75+l#OCKqbzbH;U9LZ$kU770ytOH`X;^Y!GPYoD2{;n)^tvIc
zfL?AXv6%-8|KDN9PE_II!m+jWMY?<zn196#TNv18YSX+=_foSrMAxv7V^qSM(^W)M
zAd*)+9(CrKo|HFb9WCB$CWQyZ-a6D*WLWx<hR-$Pgm@Mfux25rz4bYdE1o@hhR+!X
z3#N+8!@klz9YUTt*6#*R>)INmzVAPculr|FfAcx)SHp_Lfifo>87mRBqn7&|MsInd
zMT&L)YSw<%321!7_%BG*6m;p*ao_t||F4+K(5`D({-X8`VvcG;kC}M!jEnih8lKOh
zf@e!mmVqN?l3g`*IFAUC76RELlAo9;&y-}uTL+ZTM3br*4xzZ6E*-460YZ$ggf<Rb
z`{>8@f!i~wceQpNkSb0JZPYQ6Q=fsdll{@l!eMb1x`>yOOl!Ja$JR<KW$8<;%@o-R
zj5_jhK?~hf+>S|;PZq`U-qy|tvsM+W`}o-V-4Kz49z%X3t1?B~vato`>l%pK%gO3^
zjN*Wa?^wCeUkKjY=)X7Nj}vW_h`&lM0MCF-NHsAY5Q6*q{#HOhv~cFcHQ-cmdHFq2
zoa{{neB?f1yVvcfHlcYel$V#&*ASEP81_(V#s9AQS=M8F#0JF3&y~;-s^h?9d}3Lf
zUVd>cXZKpQOf`iij-0%>X=Dn7ypmm_r=%(Gfz2s~jzCDmwMUlR^i0#ZHF61K$^15`
zjDuHu^|7D(C+l)~Pb+7L(o35SJf14Ujayk#zLm=ui%Z_o!!i?NeMRQdivYnopOV<K
zf8!h_abKT8RuB2l<&O(sB+Q8A6WuaOiYuHbJNVKH`k~9-`@TY$bQw4}%{{d#VpZ2~
z6)vYhiOCTcnFGFUy+rdDTGpMfV1vMs>~FW}dyIS~VEFcpYKxxyR<UaCK(i&i@!pSz
z|7}ou*?5slvYsv{$lFTbHntmv>FTCsj}m?>l?pGE0gk7zmFf&JVv;5ze{5V?XgG{r
zoceJE-g=WN;cK=p{x#=cE={#UBy^>omM7$bX%(Il)bQyS522hK2HD5z+up<zU;`e7
zXUSoho<m26vg3zfe(uI13em{Cyrk&5ja%QVdf5W>rC4f(=4PUchxc6*r>|>W)EEdO
z)6GL4C9wMHmpLTm8=JfiJpIxd6qI7W3;Wgsu*L*n+)4m+4FK-k^P6R-x~|s+D~<UV
zfq;(QUKuiy@KFd=tOO$O-(sWaMy;he(uNnEn5pdXy^tKdm_sRo1#Ize{gO94xOME)
zz`1!$1yHAbYvhW*0{54f==IkQv7h)&@-x|hTI4ikcirr9gq)gz&t_Er*g0YUr~Unm
z^cPH((wdqYl)srjzhu;ucP;|qyCfaI^mvB)mk2ZvBGfbe#ViNNfn?X8y@=)d=$*9f
z4oxW)y6GI(6jdT_M`T)ir|RseYF+RD8F(-oaOpK8XJVNl(n1}a@*c9`{xzc`P1#NG
zrSz!5fUIqRh1<_&_|KMU2nInODE&D<?QwHt#am7P=1AfAkSXBKGh^EZ7TA_6j#6%;
z1L1zVXs4Bq4|@86c50>YQQVTzS=Zre;7>3{HP}61d<pVB0y19QjCiI;`a5FJ_8=(q
zW~cm*nKu7Tr6lfJ(nb_wmHUoQkGq``?OMA}%=NWL<*D)ZzpHG1RFnepoZ^)(FUW-d
z!^}VMF&2_WYI%KX{lU9d;253=bE}_zr@qwJFrckOKax$aIGKXM29vKN*1$`w|560x
z&`2z>C=18!)Ao?)iub_|PA+otiiO~`2&K%NuYX5t<I31F*+|bgn?K8-ztySkn##)4
zB#qU(_VLgVE~&M*o&}etPDXg8@A(Bty}0@LW?A-PAx84^hMR9ObFnsMopTPED3I){
z=4GYYwujoMj<hom<><Tf=J3IPsAx+nsC=sCU8vruoNZvpuWs?zww(}xOk|7~jTLDX
z0r}$ZRL<WQe>G$yvbr?lBP^wPeZCP5Ffn_)>wTamu{BLidIi@WX9VYf`CZy5yi}h>
z&+SCSRp<@`Ez%=YrU<uYiBPEXRw>u?tgF4^;yMC8CD74B88vGeBs+L1!^gFB;e?Nv
zt)JskT#P=#jO_Vx?!8COrbPd{9agTB+ccm^tD~yK-Ik*>ItV4<lq$Y#slEPyJnsxX
zt(`uhdp3FvUxvPzi8sw}c0775t_KY*89-)cur2j#YdU2Na;Hw60Pd9rQL;4>XVjs)
zum43d+d0%JAlZKYrrBki*2e@3F-h0NC!?+o*NQ8vUwD+0kR@tXHk0D)Q=&q;pS7*l
zpQWEUlvSw{*{a+7TcKKtY}ryI+kaz2WrZLeGQrIUQN{^#$>WlNCc_avUt%dRq|)1l
z9<<ZScFtcAI`5%fTeqF(Z;4oj(EA}z!w^6vKkhR3vv=4O4egF^F9}!CI>?9?I47#&
zO)w6a_|)!gJ_pY89js#u{+_BoPDG07gp&UwFet~QYCo@;;4*slXJN}L8?q|%T5>5~
zxz&w_P@0ZCR3g!mbwsnp$07paEO2UPNpHcfCpYp&n=zWQ3dZp%fhZlcq@j%W)<#QE
zODz3FgFC#{iz?Vabiof9lo==(6R_qFD>oT^zK}5VKo$!>^&1(z+~7nxzuX-@vmpMX
zQ#WSsu3j^C<Mq0-goMRYNm0S3;if`NMJlAlgi^kgJg`~T^?Jk|^y*RdRU({2=iBfl
z4zXTW1rU>l6CYnwpa6EWx|WRTV^5#W2(^oowo~}V(U`oAv$No%`N}C<ue%*|wzD}y
zr>R5tsNN{Yw$YnC-bG<hZc$31>*ab9x(4FRVtAmbIwR49Y{4P0j>H!%8q1gy5eYVi
zv7kz-s<4Gcfu=VFLke3uyM$n}CP@uu7}NGnqJE~`7oSWW_)+f8E;~;&x^Rq_yHN-8
zrSl4`oo7R=aJ>xb)l(hw_rE8REBR#}k}g)Ar_R6`Qv+L=6|v(Zq4yP5|G=|v+dTbw
z;ktF=4{m|84Qnz2Z@NRS;DF-*STvuWQ;t(!cWXv8EPd2t>ssbjx9#S***CB3142)4
zhp>vDtl50j${FGHx67wa?}$X>r)7<vUd=rN`xF->&o&X?Wkc}F!tUKc;+4O}42VdA
zzCg=2M`y3Z#`&YU-75!04Xsgj+EK_)xbM3dWT#vL!ju8#%C(}^LknRzVNg4YSj)fa
z@`c1FVq1ReFZPmJNvg8pA!gn#5<!zvrAd?%cWqsd@4BAZ0wLSUZUq>V2H40oNR%zK
z6X*b0s1I5>&XsNjWboZwT@}05=@h<n({ipXFYliC;eOWm5?e%HadTV~VGV&V?m(2V
ziSWT9PpR}DsIOro!?gA5<5E^B>bA92Vw!0_WzyX%!=b{wDO}9l8`5?&Kj+uly#TU!
z`%h1#)xp##@L6;daQ_i-{I7yfyKY1}Z&FRk)zB~6cP3u0ChqxO_ik;QmSKZqvS+M!
ztS$SOS*9QVZ0y^M@IAYfo4zQ=?DPKFHqW}Azfw%BV#SsLWS|6-pg8KojWF%MK#{Fl
z5X-61AJWVIPUD8T`ng$=Ls^(eIN$F4uUmy_&!rZZrS@yB=zM~XTr>Ofl6sRYS8{94
zPstOStsT4rDHc8^#)FT%ybw$6-EVg9qGXVr?tfTb+Pl5@??7u5h806vfXq4!`tX{t
z!JC<t;>c(a__FojbN*6h^Tms6yggXOl7#hkUHj+?@;1t^)rvS{iAMYmbNxEYveF?4
zNus-_2m9!V#J$bf#P^UU@GNW|;PKIq;!tcJ93(zJ%K<Fltci{XVw0d}K#i_9U;%Pm
zupA*G5WlDZ&`S3?l`*{DBMU(M<fwiXUjoDL|910iyp76_;>XC?m)RBl<Jd5CnKT0g
zb8a4aHJftC8Tn0_ASQ76=baOuB({ywho1b+H|vy~Os2;2A_2C)Xyjg<3yS%3p2$yt
zeL<}dmEDi|w_a;5&RN0n?iMli8!r?363bD!^)OEQZ`DZNgdDB3WNHf4F9uK|2ASPF
zz!sJf3nlghzQdx@n98_s#*qls;-AS;etVm*+jok8=KNDAUGmUfWaIN7?5P$+O~At%
z#ac(7NciKu&U<$nXLv&A5G6$UH}awMMIF3m&Td~{)@T`V=xx;{bwe>6>=HG7f_dR`
z9}V^D@~%SF_14p*5iP8xV7Ir#Il1aP`rF76N{T4GDdDtqzsNb<YU=_b6V^d2X)8gu
z)<)~AdqD0rsjtr^JP~gFo^4%19?XdA{Dw|4D)G=A2B>WJ_egIa<w!r~{0FGhH3>w*
z!1yW!xRd~cBq`~?y3QKe@>F%kBl5Bom#lJGhM6CTpTFUCu$%H09w_Gbl)ph=>Jo#|
z?L!~yKJa+uj%E(fRG+5^8YZ7#7IoUsSHkfRz|!+ByET({8N{9s6MP<5n+|RShsZZJ
z^X2>>m)A=PVl<=KFJe0GW0it`t$jLbS$d;3+Wn=4<*tdPB5cW85#`%7ydk5&&)E}B
zz2%=X#}vOy;82bpORI)QI!%!IR1)_gU5AESmvq1kADu6M_+n|Qd6<atc%KlF^9#G4
zH4q>GK1f|h2jGq4%~4IS{S3eeF8&JZz@0n7>xOI>4L6Kj1R|*i;Lsc8Dv=EiT(}n-
zWm~7>S8IAgfQY+TF%nIAeAmN3=DK0Cl9(>Zp-tNUKhEk-cQ!IDv7qX?m8XAJC2_9G
z9V)FS<=8|H^h)-#H4?JCNd4ZzR=ipHDYObWo?c$^Nk3Y3%Ksbe<ftCrK5FBmLCV1a
z0jqPls|zJ3j3@<K>?B9iA>%QdNaFYm#NoAyxkq7;=sXL4QrFY9Fh%_;>98PMb5mLF
zz^c!l1vPA&>+=gJ-f`Dr^W!{?zN=8j^2Ljwvc_08)-b9d(pl2m;gGB<Y6Y7H1AxjP
zn4_va+OU3F+t@g;WY%rQZU%@sCRK86tCR}hPDwGg6vpA0SZ*bSnONqgH&do;;5Vua
zG$A13=Z>n@SM7jhqb9ZG+{Hl{9r-M)^+M>qwE}+yp3z#1wm~GOc^HZG25v5zG*J@x
zX0h2LZm~aV%NtSA3=gI;BM|(hgu(W$G^5S{tK(yQ8(sP~D{?LOzQt?s>J*Y9{!QnH
z4erFqZksQ^zI|2{CFT@tRjb$&_Pu2ZY_zINw%6UGj-~2VcXjt40(QEbeaI^wptbIY
z|FU%KWtY@)<{<s`K%uyZ2OMOoo4j&zaz4gRt+WLtw0W&dlmeknld4g7L7JR>#E_rN
z(4sy4nttmh+8F~WhuAC^)hEe{V8N4HT5;U*I7L0hwK*QQLElLOg!%_zHNNyT+$j?y
zBP<#0{~FAI%JkU)zty_Wk$GboFuJD$xhgk@TfvT#H~CWkZC9S2jXu}2x`59jn$t;r
z?Im}GRy%{jPGo`m6WE8!1i$c*n0vFEq?a=<O?Dmy`=>`RL8VSDM$~q=YO!4Ts>Dz=
z{P2t5fKlhm-|t@}qI@=IpT>%fY%@hWP9djgPI<EiRB=7apN?9?Ju4&3`^>7X!sv}o
zDr<){d`1@UQ7suro$D)dvC3TYOs<*BA1;!D7Lzs%Urrd$ZoeaHa9_`%IkVvs=++Rf
zWq>^F*B&2h7RN=gbV=)+ExklHLw<@VKuPc2pl0(~w>yt#5W--;E~|Zp=KVKWD62BP
zQX+xoc#9kSRP<chG)dQ@B|<_%q!3czS%y1W=gAuXp*(;>(d_Av4v-H%xBLNI`u}C7
zlzi&DPAx*+E8=;m?Dn;2BFwly=`*QXWMq*ieuhIWN;^(IOaV&u&RZq|(k%PUZId)z
z++>;k|LzCt5e-Ccn7rv9z0@!=0Nw2WZF#L9dHQOr;sx2uyzT6MA*xPX*R?WHH>zH>
zu<uk<Z*=-SS>&8FZFQJPSrIxpQd{o>`uOHn^mJr4q9H9IwwGZ7s$)c>eA5AP(d2Sz
z?|L<W>BlVE{i|oDTCWnp1%2fpf>tmLmi&!$_{yW7yGvqID;Fv_@`qCi$8d`ZMN;<z
z6Xjpe#deptV-i0s#o|Q0$7YwT4i{P-vd&}28pbXf;bz=MYwCp&!VxSAqEDZ=L`afU
z>Fxqlnr-lo{Pl4TVy~`(<-}g#_Py4(B62VX2KW6h*}8@yuZs{|l;QWCuK{y&$l41)
zrsF=y3P|)G)<Y9#Lz!8kdJ&;`Wz0htv3bn;rt4;Y*6Bb(zw@TSbYuNF;Wb{!E0=B*
zt(H3<hzB<ElE(*7O4+0hop&N@^Pqz0Q(b|}afGi_R~}1n$3Q`u9zFxpTgUg=0YRpj
zv=c)MJaD^*hYT{gZh!{4&u)ay<M5LioHVj|mVFxV+C~i}LKq_7$y$FRiI0yTj!A0O
z*_bwRZ*puWza{~mw3uFS2v4cth{HPipV+q^FK%!l6?k#v7);~r82KZl1~n0m2vsI8
z5BT}bCHj2&Y;4k*7_mX|N}^6DTu^*lK{_^0FEZAT@jYo8H}IOvm{ieU08wmlaS`Nn
zifm}Oj4?NV=<a6$D==Km$MqL=ZDR<I<G`0{iq<#iCuO#;R)d!!v$TKU?g}cOmv%ES
zqKR2gs4+YUi<o$J^Yw}@OxLZ&)Q0c{Fdsj&q)r?Dou}jQczx0nT0eyu$#v{KyP=!&
zod9llU`GT6<VbWSq`+BfWOV~dYLnBtRe(bddo4)-Uc}gp!0%1qhq%iBuO+g5zzFC5
z)fFO$+t3g&-WV6JpKi|Z5FuEW+R`{EmvWR*$A5uMIQ+%4xr5k;xmL95L{i&+hwVSz
z4T^b&1g`}dS9l8$gJu@y*+6jl(gu4ll<>weV|?s3%@Cu%s}T(ckAD3eWZUj%emU!>
znVjRr+_tivjM|Uyt9(V@*D}{Jexo$x>Dy0LO8TKAR5Yt<(*Wu%vC>|}T@LSypz&b#
z6QUc{YLa_&HRBg3d2__XiomY-w6lt=Gc5n;PnI{4dw5_Z5R`zE?^e&iP7iK0L!}ju
z9#Z9{s_|e9iG0&5Pu@qA#l(sR7U83S@>(8Q^dmyypgja6e)9*8SHr{f(h~YIJ&`up
zKWWJEByCcr@DTCBiHNaK6KzteZX?O(LKjeQ5PSuxL_{SLiijLgP`{}t)8iz5=dp*e
z(XZcg46B(+wcmfJa(uU7>mg4^0}M_{Fa$+fU50Mv1fQWQtw;7iZSL8zQ&2KtZdaT}
zW@6N#Ogs#e9*I^ehS*ks!cNYpEieq4Ln5t^Fb+L>QhR*~+5@8eT7OZ98;L8NW<Flo
zcz_HsC?|<Zu|g=Ks5#L{qk?-Ep5#9v+B<FF!b`p3>`lY0TMBUpXS2z#{t0h?Q^ie5
z!3a?Vln0*s%-(B-P*8{r`v<5ZJOjU|Aa&o_%J>oWFLG3x?|9+Uu)2F;IySUVWWrm;
zPR~CY;xG#il6KGbUmy7t+${66>XD(S)h|iY&$vsqT?PW1R>j;Yp8nxF{JslorsT)b
zDR5C(*wRDnvWOg#vgA~;QQ`wAi|~uRLqxdpXvSvn@=F|0M@b^3df4&96q3((uAK27
z@w9l`48EDY{_*r^@V~G_TuQ~kriPT2$Vt3pKt+*arWtH>X>ZTu9wN_<mtrTY5kEzX
z1a2V8M<BTeqorurq<AzF!YP!GE<MmLFV8Qw1!j%s*n`2RiS>SXg}>A@^FE{B+Km4R
zi6fAnBY{+vX|$*^XR7))-pGvKZab;LgHEF80xtpuuV;euEC*aUPIu}Q9xY`y6n@@A
z$ZY)Ptj4QQv|4a?T9DL^J+tECFac5vU{AAjbIVd~i@}0(`9qmVK}ZOyG;V{fT_3^f
z%NSBEL4bkk_GESY=6D$}$|uzu0OvDSR@MXeB>;voaGccyUSudC>gJC>FAWB`QCnM~
z_x_QZ5goeSy4q*t?(BM07$9J40!*kW$>ZseMp%lu=~{=C21Vmrzc)`WBhp$fRj)5q
z>pB_cSpb4{MeVcC@W;Z>@19ZxeA0mPI}n*?h4WC%wccrY&af2FPy4>#lB{HQE2pBd
zQ?QtQ=-%7kFR*=%G;<g(JW7Z{Ko((mT8ybx{-|6Ezk8y@)bQqlU~p$~U^64wSbv9T
z7bmNAILg^$K6Gd?6;x0kh*|Esw9s~?QG6|~U7?P(V9^HvUr*Q<mcCTiDI`i0^eK=m
z(tNzsHSH<f>Q;4b*V&@2qq&K-{1@9Gk;M%1_m_0xpd4a3u9Tdlq+{F%XNGR<B%1jn
zUsA}OuF?p#ZS3wlX(JV`2izL-^37$a(XK2kgC>t=j_U6pfm?5HO2xGwQQ=g?7k5v9
z?`S3b0WMQm4es~A12R-+S=%qse=CCU{{8D_5U|_D2(x^XwC_Fg6ao;Jhc`>|hQ7a5
z1ewT3F8)fVVTP8e_Swg>V4RMzX)-^oFcVr3R+AIvrhWHbLS+b_txe|N0d!c$;{LS;
zdzfuKa!lD3dea7-*gXfJ0`@6{a)fSg<96OAtk)Sh%Czt!D?gTD@!VZ$6<8%OQ{NE8
z{PlTKLb**MVwjwMQN^MRebYPn!irEokO8C7NWqdpT~pJxNB=>WjG90uvi0J3>e3w4
zeLsx4y@_=1AJ3{ZnkW<mVQlaE6xVvn*to*~rfF!w7FuqHFFK)9>RL3zQFouxK183h
zXa6B|NUUwuof@k1b$lE!<vBV#oC*Rj8vy4<ZmE}5|5@vY@u#5odxUyz0~{(v9zC<5
z!%NGq_Jl78I1NK3CZgVhonyPYZrpvA-nW4SDRgiZwCwgb#Z#>p%XvQFT`xZ$TvO5c
z<R@Q(K+|*J4*l8+4V=?l6!;Vg2i55{o6Mg-0>(T5<=S0H<N=WRd59`>9}_tunU%05
zKfkG-wbL#-tYvO#=9fc}>RJcf(&QLYz?x83oH+(uYnQb4_V&ELL$Rqfz__aqWxPbz
zH4NA&z!~3ox4%aImwUuPAvGGMY68%q-78Y)_;jg*ICXcJA!I*X)?<=9U%~Q*$?pQK
zu>|#obs~MVQjE;9?*8f@db->!mx*F}N||_RXf3l*2H)!lLASxPt8Feu+$Vc+a}bt|
zwa65JlC_*BDx{Ce2Iq1DxDS;gAf6<X3#zBYfPVQE=SVG5sa5H{3(S%Brnk8vGike}
zk5BqWD!FkrZlXHKA)4@oNRo(%64K-j^0h7>FAF?V&2AtH4Bs`P(n3pS%r754<&1kM
zQWznAcQ>=R7sMR@V_ME}{me1TIT<~C&#-6{Di57$N@Dm$RA(z&|Na!4Of-)U|2U4-
zg=ur62Xxbj{B-2>gx<*e@boVSf;)iy_Ul*hyE%C}ty*`0Fp9eVa<dKu%;T_ba_hLQ
zy#sr^xf@^P3%_`ay8k+-0G2to3PZzQjOhdC>!#Qf+2Q*+U~j39yuMbsxv6wJnq%?w
z@~rS7DBG!=^9k@+_e+L&Qmpeh*Ubnd_7ox=@omV!b@J<!^FsG1g9v(@R*)GZ0PMy2
z-FZlP)`F`~@V^c@jUfw88Wk>JyzKOJ=K;)KG)Oge1lAwdZUy-HOZ+1<MY(6)cus)I
z@^Q%Q1i21)1=Ievcc3H%)B8IorNR;`<p?q=G1a7EVzI#BTe+2hedCk$w!xn50HDT$
zI6)RqKVBovqkhTca@Ve)b^V#vRE#bal@Z0FaMgG7XNpRNkE`pLGyl%Ns}#`fqq`tT
zFbCKPnIv|)-X~I{0#DaJ@M&#F1CSBqJ@zCyMogb>z_LfyxC_+3$*_lyjc{}bZq1~r
zOig&7CoHUDxF;O0cfQtZkGdk#PkYOKYlu*Oix7l4>B*4it1-u#<zET!@e1XI@<@;a
z_oKP_w_pbccA~5DX;okg5*WD!*!HOzS`1k=Kbe*ObDd|A0HYr|1eWF><5qYwfR`X|
z{S>z!@J!(q;wdCmcXR-YqN%VXk=i9TIFfh*hFeZ8(Kg%L+-@g%=l%s=dntf7{)=jS
z!tiA*ZMa+G=PXmzA2QDK<UQq((JIA|fD3+8&7oZHv$p_t+`hb?r8wB$0bha1-k<Xt
zUi%@lB7$p{U{0*L9`4;phD13YuIT3_qQ~4hvy=u4pIwEck<_vwF2LODCYrg#Fk^0`
zK#-p`GDGQiyVLLw|F5><;*8f~U7?T$ji&wQZ%VmeJvb5V9ZOg6kFr`Wj4;XGL+ppq
z9T{lGkRmvS*nf7_{hm9<zUP)RE}%x9r0G5A56>$?X74mb<R5f~Y!A|4^kbn8$x0Pj
zTU#VE<Q&7*Fv!<ppC80N^(=?aP_+J-I<9YS?0|WvosZBt%{=rK(;J@6J-_vTZ8*I=
z+r$=H))MvE=Y0>X5_+^X@9nSt*G#5>12!x<nF@m)vWw^v4cOs?jA`?+e_wJndQ~na
z9v+4}PV9}{H<&(pz+kZZu2-=ar^cIXoC*i0X}bn}!z{`Z5i@@d_}{(APX$&gWw6pE
zuManmqGGFMHFZJrorY@Q$Xommx@%;iwV#dJ8exaS48RB202URznD_BpZDUywh6~_l
zOR#D9@=3IPVrmNR*@O&=_FvHTadjS0tzQT<b}q5?|1@`x>3>${64WEv2m7Pe_$C?_
zf3rQPPVEhIc6|eJ*26)@HpfDMi%0xDr#IVY-uH94@P~5G!iqj;vQ9ZOnjsgl{Rl&J
zYe6;{ncB9W+!`$jEz@gTHL`jUvM11(HSCToef7Yg*%vM$Iqmq%Lje%v0|GjtcWQ4P
z=50xv4sPf4443PbBQ=J|E5swk=CbMt2fUH!{c!B(E#mDlZ#mCiO5lbcGc42D=+YRD
z-%k<bTK<CgJ8VRjhj9QkqFpaprKSFcYoF6p4AJLbyC=@UB$I<v0EvKv<Y=WG;E5=Z
zN@)qUV9y-ibR&)pAPu05Xc~G&8vQpXsuuz~I0BsGp^MUMH@CU)b8ksfIgB}9K$-%)
zH9onxh!jFXZ52=lAh{o}S%?rBO&^OQS<e5}9MPr#U?k<aYuorkY@t8KU88G%&MJCe
z-NCU52a!k+Rx8Jvs57ioy?9l~z-5B$2&BDKeEkzWecqoL{_f4Ra5)jVUL+Bne>t@b
znzZ!!b*|4zkowE&hZ&bQJb5>AyFR)U6h+~%0xv}3L%N@;Y_J50cA?)$?Vn7i7`#HX
ze?OvILlY5VUS87i7nEZ6lb#G6O>F&u<`a#YLk#^L8{C#M9^P+N-Dw*or1!rQ^Tw*$
z7iqWa6ve}s+$fP4F`^zscpX<P1hbt51LXBns42}{K4_@0LNR4^W=I(-GUn{Ka40AM
zKOoV?aMa;F_9eK_q(+uoh)Se&bn%R1*k3w$0MG3~qWffz*QxV5@#-m<?4^JC=^)O_
zTNs9=@KvH|bnoXxy2&#s@L<*poJ7(+J+B94nZ91Hn+XzgL6|;kuxAyx&+{k%wX;ap
zbaH(feK&KoU*+ho9uWYpo}r<T|29e21ekYw;`K>wO16ggAPh)z0faWnoN<QQJe-5k
z*j?9hJ>H4<;ci-1Pb&|G^u<zNa3;xtE8g^lY|T>+9ZyxwQg}STlmqObYMpneYraSO
z%Eh=QKD#t)9gnQQZ-v>69<X|ZbFh&(oI3Au1Go7Clifw-Xi%MAu2@8$N0?Sx+>*(w
zTb2u$&XEQf?U;xLWy}|_2zs4TOcG}U!RlH4&yo#(?I0-x!~VWaG5_lC($n#{m8Q%2
z^MBQ?T4v4=^4M$Vbs9rubN9#g-NNx2v(x7Xa-8C)?cn|_a}Z#f%IlVf3-SPSS^)8i
zm5n3g(5tRa5Xn6{6q5o>5X)!FC<ijP87#%iXMe5!#TsOZ9ZAeHZ}plEoF%mht+P~b
zw2SXBeIzFt(q@VN_an7X<$(mTk<CLJ2gX|-!FSq&R4Y&D(*$3@+Lb0|$eW-uK}#d~
zxZ>8JZ_p<2wKPP6-alS8o%R4tMh)d$nWTTFAz_f*`u7ncvP0dS5M;D{=zFhV7Zv#^
zSiP69^Ng|U4Bd3U-1M38^)BIcXtm9YI8GJ=86lP<C4w@iIJ?y-Kdh7G==NX=VPuPU
zH8AV?ny0am?+1((CFe*|W6#Ja8(4|-O8sGe2@6X`z%@Swa3Z+vBLOi_B;cxIZA~Xn
zZOd!z;D^rsAGZ0)4MMKAkLNQm*YFW5Q(T(JGC8toNuf|ZMe*$H1>vG<3?*C=&Crwi
z9f&)rcXvn0;A1_<SY7rc%^5z-vaeR7lB|##1%a^o7embS$2ObhOU^NpE5@9)gh53C
z)6vG_4202ry?1$j)16aBLId9VZQilzmoXEc#bh;hi1Lk09|I(QYTFtY_a?HVpyZ?E
zD-F7)1Hn9R3l>t~vu2ZiOor{F|AY}`p~>(uzvDgWHqcFBtT?rz$S6+oCYW)w-{Pl;
z-WuuM3JVG)C!Y`-NcwWWh<Y`jcpN_-Lj^9y4T0AD9W2X`uZmxD(-5aD@z3ck><e96
z$!n1OVEqCYN}bo%G=3s%DVm-5@g1r<qTVkZnXegpJjbWgA_WPLmsIxa^A?v3$ukI6
z+W{j~Vf=ov?cUP##Y5H9GicxPK-9}w2sfUB)c4^w{*v6eX0(w&#rMUm>611v{RRbe
zE3tKUKfisiFcV7FasB~FSplJ2?US_}c#2EgfjfWYu4TFuxK?K3T*#E^vY9Bv&DS>)
zgL13+w^=oe4^W=}oszjYpZ&P^kCnic)%l64s{`e1V2Eq&ZK}y#(NX?zo7;^nbAV4$
z4f&DFa0!;+V(E1)_n*(LSyTYOwGqy(ZmatZI8VJ!JAco2y?C(7N>yCiedEpaWHRT>
zm^{5&5c3}DKgjlMvy_^wOF0d=<*77Xr%`|pPmJ5Y6bD}id{$1@(#H~&`&q*0rh$%%
z0Mi{f{MJho?#V0Mh3wuSFb`HE9YlN<RUmB>c6;X^0bJ1|QbOD5iQ|BJQ3%%HcP6EI
zkLQZ-g&kH&)O{{SyL}f8h7G)x7&!<Sweba*p>G%pluojLrm7&27nDv<-C0bHyWmJ-
z4LanN$Pm$v%@A9bY7|N?3xb%5=t#hZTu5^Ib0-L+8|RExRf~}EQ-3`1{!l%q+WH#R
zaceAe5~F9ooRN(~x|f+Z)6teC-5Y;T|Lk+7`qXSCrxR>HI%Jc{86qL)HWJ?|i+V)a
z-X$*88e8%Zc%7_l?WUXV;X>}hp;*{5?t(dXVa%gR+@n8~tpKY!&|>smLjL>mblj}f
zhscVms>_4#?vF76k&FIT__@?Yvx<xnOrx&7<az64lc%=|GsFh?+E`dNdF<Vs3u&u|
z?n|Cf%E^-#rsO;>CRh<JD=a?a2Ddebzy%LiUaAzokdYo;dsjV<if&t41YM=@r@>4i
z3v0dzS{-Qn<#mHAk04S9HSnU{gcj-PHm%3lqmv1yaV$45`lmP`bScSf0yHW-H)i`y
zea_SrCFFbU#UadSQYI5-RF+LTnl~SqGJ0X>Qe5p<%#uuiHMmP7(0I3sOniv}xIb3!
zdICJ>=CH+ynkF@*$s-zc(K|vneQJLiUt^gpr?dVk=1#1~*;4&_Z5VxcDDZVgz&A}_
z#Bo+n5K|J3q}m>XJ{(CR6{<Xw1cZBrH>LjE2XN9ldk&G)qX<TFV$maVqM)EIt}I%4
zc_jy8RDrL&7gYF3ivFz!^o_r63R0%T(t=E$SBf<(Vcy)^XJ=7P&)24$2RA@8Abxa$
zRzOHJKec{gX;GS3=DT&Q9ASz~&(su$rYvxC5CMiNh3O`H>#v)yA%F(^wC(kpEB3Pe
z`hMQ&pQ{T;h2(y~x-rt<pL4|#?nC*$gBnvx%Ilov<N$IN;m`fHj`9AYJ{|*&f&MRD
zy#M^Z?N?hd{*OXIgU^q*_gBSY<40AmiNJ`pr-x%e&(ofpcq$s##19}%yzsr?*Xb+f
z?za2gihL6F??Y!>F~aUyzIqFV7^3YLYwqGsH{RqIFMo)d{ByRz>CMfnVjhM|o6J$S
zn8K`PFL81~g-V{(n398n@Fo4TJ~>MMr0h3@nc=9*OD^D5s**%afkJ)kxz+sxxY15V
z96yF<S!M_oe`!NS4*5xcyso)WGUM46hariK&_>)~5P(R*i9tUKPr-avzv{kN#f0%?
z>G;#J&R2KdJZAht*Pc(2cF$}Gwc0^!rrQeZmk^#sXt^edusynzZj%mbeHB+U&=tb|
zbgg#V<{tHId%FD$FP^zaMT|I0tHE(G%h`sJcO+yOc>0^~9B%GLZmJ-WeqR^O<CAg~
zJ#cVfs%js*oLDypfq+F*c6%vYnGz*DnlZiyK-cFhSH+`)x$9E3qHD#OREQ8OpwPt<
zowuYBSY!KKZ<45sRvo#Bmxo4#C%$w3oImC`s1o2hj-roV5ytLgJ+J9`fb$g8aluU`
z#ZPs0Oc__m1bpAwpSVuwq^;r{gTzY^*s%_9KEt)vI}&OJd#05}CZY07&N!X|TWoF_
zkMppd>RjQhLF-&EHd&@NUwYteTCp|++Adeq%=C@s%xoTl<f}OUo~57yo@*LxQp`83
znZgK{Rs1IndL8I`(iZRE>zxrX%4wc@O0DoAuqIulev*0S{_4M_(=)MdVGl}vhcm~A
zLh0nkDsE#zU+%n4SHn9x<*Z`ym*)qHxxP^TRETsv&e9P=qoav+pzFZhCS2K{a3j?S
zvz}izpA*>^1WXga<VXW4(y~*m`~ON&p~`SvuEt4taNkkBMX~>~D#DPc8ecN_I?iuF
zF64iG^X~UB_Ir=M4zIOczMxmH7Qa(t?q~q0VbhTyOn_$zGkv)>y>NQ{DrNzUFO%zV
zM<{nU&PC1fCQm#8OLVwPXwTW)L=fz~X;4`5Ht3-PZuKNswJi+mzqqya{Ryy$-o8K4
z&Ps&hsLvC0l&uGtqEi?<sDse|fM!C38Pba2eRSE~(F0!Lkh&(0b{Zgn5O&EH_{g`o
zy1KLZTMp>&4qI+=a&vbdnw$IZdlM*!kjX8yf$|Fqx+}zklBAQz>7LHsifzDhNLY*2
zJqEpXzoRYaOw>Ls#^KtTNL7gNuHNf$tC4`{qXC0*05Bs35C8OB-cybSR~FDKMj7Mz
z0OSD$Wzv4YkHO2!C!ZA0(!w{Fos;9@?k=4~{U3DU_}Bpeq)T8e2=<0p>I4k9osowz
z!t<$UC6&n7<VE*H<q4t(HqB6=;0&&{X1l$MdQhoBHZ<e@uYsXIC4xt5p=4?3X8!RF
zcNYU+g3pIHP0=-vB?fqi^cVxS;cLZ9Tn%$bg}HQr<EUYntcDz3m3HN!;+YU01i6C8
zcGaf}62B$O5=#uGrsns{x}$+gqmTgM0lh&!9%T(p#Uy_7To_zbUSdv;O5pK)VBCJL
z-#Z2*05eXMU(SdNFmI??YTK(vOii2vPLU<$i;{_D%-z1)L2VmAjZJ)q*S^YhRIzom
zz{I&bvEBdxw8ST<=3H2vf<OUb0HdhZT~!~3+Uwg{Z=3?1n^S9;KctK!Q(AT^qi3mh
zL*R#~@UI5-$%5to#1dpm+k7E55C4^dWs4-4hY>EB7oM#4PTiotzUE!3THKYpYb(QH
z=22~$S9*dejXdPzO2$AxiCKK14|56LRnc@qA2wUg;Jv<3vK!`rU3qNW=LY*FI?7zt
z;zpet?=gQ-DmGW$n1xC$`^5_Z3FB!e&}C$rWpGN__niesSlQ=vi_8g^dH)mZ!!97$
z`#$J~cJ)P&FF)L5n$o6eaw$_aqB@z!eJ`ho*SBRhxeWF1-wo7qm4&71NTWp~cc^0)
zl*Erc!_|wMRt%y8a#Gt0cZOCibiecB(Z_6b`Io#k(BRPd&_8iL?1=Kc_um!$MXWq(
z=$K%ZwPqqxq&%|G@~X^7vRZtP=DNm#{Znc85hCLqfMZuv*Ea4hxOa0*0vJkgWhgYe
zErfTgULUfaPEP&99E(RSasg=YLeSNZ)ed6|JoCi*HhZw6BLJ!B?8$QK>@POC$J;(n
z5oU6z-vY-=07V5^SqOqTyxUV_1nhwZMQ@x8>JL9qH!=>EU2&UWX=f#HDb0P50eybX
zdkv1wnFn0@X-2WE-T(&yz*lerLenik@`U_{eNh5%Q_jartVB|oh4o3{_x`e^4c@@t
z069AFyL%c=BnOZ-3f|7gtvq(n?LJ8=n)Ak4eN*K5ZuEgji>}QIu6xcP)!D89Vr@WZ
z6bs8xrse(Hx`|E~P91MR1i~F{Tw-Zm{lwOJM+tczvp&E?1`1gh?*#@2<yd{JdjAMG
zyhjcz`((~V;l>3&-Npd4m0567D7l4BX4=j$7k<t>_eMsv$e=`9hWXD_Tt_2#HrUS7
z$>*rZ0;fW~#8rjtM9w&YT7JH#({Xyk*EPz6^2Vfrb5<5{s->~I@$m~^vLYX?&W~ge
zy3tI_XDgR)zZ`@07<Av{7P_1}2+&(hT}{giQG*}Q$yDz9_Wegh@5lqLxs_LELfeY@
z-lDhn9zucHih#~~{EXS;_8gStX{qX&5pvgu4jkq_iTDlye8-tX0N5Bgm$|wMAmcwB
zKEX+_|0nkVkw)2V3ZMyrd;G^Z%jmpaLljj{T$Z13b2<k^1yu@o|A5V;ri<o}5-k~$
zhf{_v#@q#Wfs8BQl*yQ|jvbInQN@PUbgnuIOj?D}<<YxKe5^$)s~v%gM`4A~j-v0;
z4aQRSHxvD~2wz$&6w%9{Aps6piHW7L?ewTgtOtTkA6ajU-Enp8K{K2phBwC%jTk5G
ze(xqM-Ua+EZ)uqhPcL5%Ww;H1weZUwZ&ex-J-PjH9mz5q+(rjz3AD-Bm-l#3r2=Kf
zvF*{Ixm%E>O)0c=onJhi`s0Wuwh&>qHbYc5Qd(Mrk@phNjpKN6z;T;2FFz+J9Tx7$
z7d+V#5<~FoPdF!a3fwQ09{O<4o`t}Tnv71Q&J7mvj{;v}#YsX?#f2ys;v~|MEu^hM
zxR>pFcUPgwi_1|)Qs#?a;lsWmDxy%reNU6e22x_7Lgtg25;p3bRENm^&$^<6qg5-&
z$somVu{cXE(#pY{#E0(jXo#(NHb3Y8UY_(DE9>TsVox~rlZuGaiXC5s1wDN-z3;D@
ziH5JA@_*ohZ2a4a^~Xsp%6xme29z?KXZatecPi?H<7HR~lDb6xYT^~gk5+9Oe2`5r
z>59WpEV@L0mq12^lB3@vI2W{aPe@eZv9PwZdHQnVmqOe0ueVU6JS6fXd{*EJJe+=^
zBDMjpykDr5u|A|prKL5zb;*ARe&{^0QdVt(v#uI_Ue7d{JS9bzB5niPw6@n2bkua{
z@tB>P%_U57(VIiazp%bPcb(OKd14F%<o_T=hYrf!`dnXD;q+k}{PyPDUA_HO-Oy%S
zh+DE*oaKaX=&1kY7iM!O2PRj;{2d1OLys8v_doN0Z_WQ~JzVcEEG$6ltMllG0fSSW
z`NfRShz+-uU<L<dmxzZdhp?nK(He=65C~0O1;=a>Kbw4|R;9mO;Y2T#mBvwsavy8p
zQ$h{li6P)^O2Ptq2!F?DMG{IyO9F&H%Tm<P@YS57e<%OG*9+nceM4B(=DkINPT$P?
z^X?wlkSQw^<GMUA#m}e*Tw`=pV{@DG7$SAreJw3Y{8?#Yel`p{R(s?O%r@$>e+DmR
z=`=Vt$)y#Zuo=>h6E%M>nmXn}2c3xOd79h2tl`1RyT{A63A-M5#vSx6Z9`%2XVN5A
z75BX+ScUtyg3|nI%SVyf!_5yEr}bzUP!ItZ!`7^>*x)_@bb(Dw7^g;=r;0bv+SPS-
z!q(P2RMjTR!~5-l7HI$lqL8+$En<0S7eG*C=H-<>vrXliz769%=QWm1q7DX-?%XQI
zyV3i`jH#l>1u{A|^&_N-F+<w6mw#G(?CqJB@16X{N44W7acsFQxPC<Z;24W0)yFHj
zC)G#^&w=CQ=v(O9vdN;FC|5QBU4nH4OG>#Hi+M`@3L77MrQmGeY3}N5tlN$4Mx6W8
z?p3DMp;PCW3e5_F?^d);jeqG060^1gCuQ=WtpQ0dP=0hM=frSbtUGU$&rvO@F{3We
zDAQ{ZMUeePKC?$EbIZ{2gkzJmqC}=AaQ>+sL(P9v+E1guWw|qm{7*W~#r0HV>l#p+
z;56D~q=V6xh~Nt_P%So0YF(SF)Xd0Cn&W<d*541vB^%@>3d0$or5*_8Br5!FE;(uY
z9o(~Y=>@c8B#8<au+Ye+5}B+2>Xrw`i(r;9f9u6E=PqOwr;lY`?Cczv58Tyu6mAN8
z=N8T+$aV=&fq<4zj`u@8?*jfC1r9c4)TD$YS|3XYzt|?sx7YAD)yNv}`}KgKm@47H
zj<7cJ&4)OVwZLZ%=+7`T#lVZb_o!`=6YY00(G}!5`^s#9(j)7~0&)n*@of7|C+b5|
zX#L+3$}mdo`*^bA6Jp=C?EhB_viH&fscBiHi`{d76e+qCm=!in$}rh)fS+DW`+HLP
zKx(|deUZf_Np6YKjD|N#qJ%K&Q#Z*qUivo(j<vdo9$x{+3}6?4U=wowH+7I*AXIn?
zyeXURQcWK$*Pic%guNaG9FHxtvvY8yDQrbSoFtsN7^uC$8PoE_=W`S5o3j%^k*aM+
ztaeUIBZyraIpNL|RAuXNk3%v-cA!f+o?Nrp(7GNZ7G7SerDK9)on?~&nll!q$2xj+
z*<{o@{YWf|5CFyE>)sYH?K3bng$<w)P`DLHab!H>4mE!%9xJX3BoI@OFem0b%IKSw
zQ`rNvI<fmjwY&R?b>Q|Ef-$uOi)W8IyahI%W%mZKgL(w74n2*6rd&s>XkDObrtoMK
z=<vdyAsmIHsN(7I-!t?56l;C!XSGYq*tc%HWIuJ@J^kmuoHJ;|j>2~&o>*EsMr1fb
z(IyKyR=HQn^E5YgBT;ieYZZi_d+jWP3VM3j2Mg1wu?c0*uXja@Jd+=n#XW3u_WO2e
zft0UjiUStc)&O_&J}_K-*I_at+a1sLLm8|(DLes$n=JwY!hnD2{h~2YaO2pdf9jw?
z5KYO(U=#rl;-0Q4cms%Z?t&j_w>jo2NEU@{N1&+II-{4_!xoZQ3)7Yzj`Mq-+(<|?
znyv08dDPCdo=}hED^o;T<k-kJSTgZrdx-i`BW`<&QW`i!@@O)DxT5GO(rSc?OCBoz
z3?wnY)7^qGXGbfkb(_lPkL%(#zuIr%7_y{In#>l@^68BEz5Qn=vl>L`ugD!#p?fe&
zSv=%co>=U5!7P<Jm;F&*|8Y*xs;~vQozm=R*g)%l6rE*QTiq6fD`=78?gTBa!QCB-
zyA%%&MFO-q!JXpHSF|`qi(7Fm?iSpgd+uL;<sm2g>@_oM-arjJHW!-4S=JAktMr-)
zl@nW)qAA@EN6#;g2dY-$0!ltlbzk+?7HMI~j!5Lva#V!`inhvT<Xxza`kZkqMaNCU
z(q96-sT&yw+n8iqWQ>wSF^*H;;UXbq6FAI!ttWj2EAMR&&@OU;rbzVd<ALh?(D;z!
zZ!Ee0M!`W{zTxc9YxlN?VqM$o_w|2&QIw`~{;iE>jE;C$PD~FJupw;ws3z>lGXV;T
zIn?*>+%7>~+x8PpX|dw&9?%tZOHmDd_#*mvn07xIXKItE`h&o-C1E9mkFRlXDzT>!
z7jT3DHJn9lNP9?l+^#{$@zmI+Zhm7LnCOGYM1h#eeq~Ju-}>$-+|k+D)g|5xoh^Lo
z(b-nT+WcSX0FmWolR*B&%mmbg{}jFz=lt*<dbj(J`is&?C|{Ei9$mVo8`J|3x9^z6
zUe4nHQ}pki7GD!z`nV1B<uZ5Ax#PKMkmHT+$<EPJa`x!b-{5!TvckT1>4k;3Ie}F+
zd!{mQ(r{;$QB71r-6_4N6u0m%{<#Eg21&3uQO}oBX*Qo=%h~z@p?xiaMcH(u#$;n4
z@n%g6Sbs#A<B?nG9O&gOVWV7vI_W>k@N`#7KWC<3-Q0rh&z~-H0fjc0QL${8$bfhb
z2`+rQeCc?4lmLQ=>y>WQ7Nvl+RV?N$QzvIKXn-P6fjoQ#Xt(<JFN}n;K6b13UIP2}
zq3qcv6(aD?CdcllU!w$(AImQPEr07!&gdNQ6`4IiC*V&niCbnaand$KJS=Ol``hi#
z?{DmHHcsPhGJ4F~q9p9!s_3H+-8#_E+ih=qZ7_xrBFg-HqfN*8dZ^pkT`KncAv2Ry
z*!%1opl(@Jfo41eivfa1t6$9ZHAZZl+TDx8nvFjssdChzqT1Qb%}qALT3egotSC6!
zuiYdKXeqMw=z;>Pq6bS<)X13aN30h?leQtriK?wifxr6AIH<<=#LXpwe_bnx!1YqU
zeCfY(jMFaptR2MzbZ!%CP)zASYPlFj+=l;US@1DXDDR(c*FV5P?Ytglg{%$nw$u<=
z6v&DPGzBdyR8nk5Z<eL?`C|CKwaBP~V&G$=D4}vic`~6ZaqmWj{K%tGO8XxcLN+Wk
z_j$4E@mqn+(Pan4)PE0)k_%>^93V4Q$`<~{mdYu&Y${r{JeHzO!?HxOy{?*w=D<Ua
z*T={1@}>RC;r^e0WG?CgQz)+nW+~ar9cxP^5YH;{d`7ye{Gnz3gZtfC3Ge)~d{aK3
zl7R%qQHG=@K>J&Wkf5}aBg&IhAXZ2m6<uOqmfUuy_!|(G!&IabGIh^^4C8u7rzG#V
zJ-j*!hd!rubYHTI$}J~Y*~3tdR(^D24ZkTy`>bKW@cHYS23~><ACozmEbzc^h>Fq0
z;@&iV1(ToF;f61Z2o%k(Z5%h9T<(~@;^aPW68`w>j!E%c9Ss!v^)iSKl>%i_$ymT3
zMB5X8Jqii48L-s`!Vqz%V&ZAA_Dp`g)7!|LH~|4-oUlYNCNbu?&*`dP{UUj-@YeQr
z@?_tCs~%^Wsv-(>*RnHK>QK|i(J3KYznF0cev>!6gtOgfb&L~Y6Ym;#&42AS#>DV?
zRYTle1>b{Qvs@l;ASdlW(+g$W0MH-y_haX)Ai#@T?zkq4!=wOELbg5s4w&W}o!>V$
zH_n9t<VyFwIdTT+?0#l4U90*LT<b&7XzWA1@p`k!$Ge=1d*}mNF`iv2j(jEsqsM`K
z5tc1PD{H}G=B)<NVvI~$w5(KY9N>NkR2Id}(HVq)Yy2@Yk~O70dsZK)^@fDyY~BEP
zK>$dc%29~Zif?QtzZ3QVaG#46;Q>&iP5|^WhID-LxQZHqt!S+l*ApLYZ#n9ZjZNUy
zx`=EcF2a<P{8Y>BjP>(~K_Kp+o>$deV<tY1qs7ihfNfgHja;4;p1vZUjE4|wBBLx#
z88g|JpXzM32$q$6IU7~mb2#aSY~hVhM<b+j+CzFjA5MF?on9+>Pd_a>0yM%tLeuU0
zy;CfNx8dQ)%YiO&wJeo5Cdot?OG``2rCLCFjn8qzJP@z6cDEdmiA{@^vz+3QaexFz
zSeZG5?nd$sq#^=ux@cKiRn^#qe;t^Ze*iD9fKifZ#D09wU6L@QzNQg4r!bQa5GCw5
z`jII0cmC?2XDF9o5IgQjlZEho(bgz{WqD!YSR=Zoa&u(h^?))9VTl!tKcY*N8MqT)
z1dkgqkc9jObx4UB09B7br|a`#9@!RCu{8Byt{Vk$C8_<NFNl3SHyD;VCZYjZqp(9d
z2MM}4|NMr-LB(A(XbSnOCh!4)t(l~u-rZOgEv&|2X*?OeX61i)Mw*JcCx2t@fG)Nj
zT1!U`1EMg#W7P~UATh6MQ*Qw2p3phFU`l&hjc&E6WJ9(wo&{i(DM;O#)6_t+W@me}
z3AH<2rO<Y*?Yn)bt+8210nEUqvh0K87DLXnUghi}nRxe(^)zE4(<I=;do6i5pGX2j
z&0nzJC0pdX%y~oFq7rm!nE8a=`9-P8<}7=gVqkj|4c#YdWze(-j13<fTZ3rw^g4Zf
z7x+3BtMcmDDy#E6J_GQQ3TKGBZ}a$NlXFs9;?ax1-cB?C{{s^_y441TAMZ7K-)mK$
zJw-M7SU%vIyv#5}21V3ObH}R23V;RI;iiutNSBt3va%%;l2l8{L;|E4(Suh+9#$}A
zbn-5uFQ;i=nqIFUW2$Nk?TvlyeviPaFh}la!i;jzvFt}_w(7bal|~C0BcjI20SA7L
z@=*QCdOHK>Ai=K<VP}=q&n%-v$6WnuROoc`RmYRgrOBZdeA$;ngknQ4TaU{90LQU_
zf1fkA6nf%b?G_NAN`%(Uk;mfJa!kN*xGXF7sSC(TMKiqy)<5qhSsA9TAis<W=(eDz
zBp9?q3(8Ei8~!_##l`UApNQc@2`cIe`;+z~(D5u``<o@7r6pNB&jgrf5Xm5At~y6P
zn|N5n$jHAJ3f^n++xDw&p(t}5rJ<$m?pfJdb-(@R2Rz|H>O(ER#Iw&>J5>z5>SP}o
z|Lk%;!#s1pTC536yZHPfZ#?GX&9`bjP?|MFQHV~KC)1QaY}wpAiXQogsCjog=SeR4
zpP3^$dR1lrhG9<oe_zCu^IK#4EWS5{J^P<hCD6SO`<GrA?11j3w$3YIKt0NJ(~nz{
z#I$uT`c8uQ(XI=mz=yC=CHEn_kZ}&Y^uW{yrb}#(mekWaqA|`5pYta{^se(>5cM)d
zFqzcZuJ$s%{}S6cjlu-H{y{PF(;mUmAE@)VS4Um9#AkG0{cQ^7ebL*lUucg=T5bN&
z_WDolVyg!VcH7r~LrHm!_4xcol`lE+<@dXOOx=^7SF+5b*c?+vnpbzV-VfCB1tE<N
zd_$2~u|RDp0ot0V+H|sRwTXEpohEkAMtapW^ll?OD?8`%aK~ZQXOA4n3(FSK4A-7L
z(buW86ybP4<mMaEFQ{eyC0H~0<V^8kXl(QIGST^@Is@zbGlj`CsyI`^mS8fRK2Ax!
z=<7scu6M*t?2Ja*dRrW^d4@Jrs!tDxx$Llss$w*)3C<vvBD!wkxgD6n*O_vqximuQ
z51W%AU{b8T_f}an*{I~i$XtU0kY+H9!Uo@eV=?zD_Q~3M&k3Ph?we3agmEP#VG{a2
zYs8z|t{cId5gErQ>#7&%t^4bTj{UvioD2P@&nVwQU&nXhD(2gb=dH*|M`4(AYjV~}
zR(|}HD#p4JtyKkfB!bh}GC0(2Jb3v+<`o!H5zp>d?@diE`C}0^`>)=c88=-w&MpS(
zWv=f>?HLDL5PsV4AhNI_A4b(g`|0($+Vqk+wb{s(Rr|jlf0~2J77IF#{@dGZUU0s;
zWM~Cn2=#p2WmNnSbIX`F@;kc8F<*VmN`83H=6|iRpji0}5lpNZc$k;AhOm!>BTKg(
z841R6n}3pXz@@`-yC3Nmk4h8L7tWY29tzaI)9A(L!>@CQ&YB8Ze#vx^mWf?&%<DT3
z8qC|8#&`ZPrCkbqjf=BuiMT)FtTqv25Ea}#I&ilxCocCLnh!WGpMmrGP}Jxdxjj6D
zT-z&<F+BateI^b(^LwT9Lk_QC+e=r~szx&r@t4>G@t-W4JM%~z4)onKr9}}_23%3E
zJ+C%#j;RwkB!0f1Yq*hCIm^lu?MkCI-+%3l!aO<@hhS-K@Y{TnXfC&nT&sBXeb>B)
zdlypG`SO2Y3NBFA4Y)fyrUh2L3N>cVzJDxgwpW7zefnfl2Ok0?#^QdAT0cEh2PSPf
z@awUSM{0*8#zNzfm<Zo=!PD4(UDMs$%ELjb^h9KQ2$T%sKL?loEOg*BP`@#$RDJIE
z#g|%mzCXENp*y}wxp!Kx<s*U$9ek>tU~bMbJv-dto-0SxPequvH$88FsA#9<^YWan
z4jExU6`ATe0r#cGh)v@VNG0EJWal$8AC_735K)Kr6lmaNO(yrjswyW0<rHO>?)iG{
zkNGP*csG`C@xSQ*GGW6a>hDROoHS8yyak=)T-Nzr?b-7GSQ_?rL><fx9XnpUY}3y>
zDlmTiDQCly&@jl6O~k<YV~Ar;oq$f#yTO;Py5L<V(9@l1Fww9C@^Z3~TGIL5ZS#;h
z+Bfp|N+#o028j>CLs$~{7-M=RR`*=*7O2`wYBVvrO0a6$$@hBLs^50bh<!phuIQ54
zaJ)d%r<+l0@yK%25nkS#1%&Q=9p@!ZGibgwIshXKNI@A-uFKjb6lnh$!IJ!?$Kri`
zh)+o9Uck6JmfgC2k1{QxU&W#G_$YUGNj88jS!-ALT@DC9oXZ=6Z8Sej{T<Odrg6e#
zpUW*_+8P4kNn!7wK{c`HP$d<IYk@>iwIBjB?8y^jPm?*6H)v6S8EzXdJ%&w-HC7OP
z6EM|M;_#uxdi0Qmwu&AU%z5vkESOPHFi1P(O|EH4CMyl5887y{o9kG@{C9n?I<NE<
z#bt`{w|@Y`Ib7eVs(hh$T4kpOmmzyWUn#RdfQ5Rp8m;E{s$r-4YZ#x{!D;@llaQtM
zLa$+xCimP>&zX7na~*0Extoi*$RnH>A#y2<zIUmX&9<fAVNU6LyeMUYL{-1v`~$p{
zKO*;%(I{j655tA*^Z^021U$Bd>1XC!g$yhPjR_4z2NnpSuyf4b_m|oM!mR}Xv6{jV
zDx{2p^{b?~eGBy(3ucNxF5*XIk?}+aYa9I%7P#JD7Y&>YzdNgSF4Vt&g#ED#Ud7%@
zQKX(u-sjjngQytJ?$_oC3DMz)wL4g@-d}(9*;D#RYz1pEOxTGW|1VnttW;)E@&2H#
z7lsDXvthMMG!ty+M0UzRDiq2^M~zg(xrvR{%W2<zIQOJjZ3&w@yD${n`8VF7mhs7Y
z?Be$IJa9DFu<8E%Ylo6Ht3Inva=8-C#-P^e);+M`lezZF9{7<CPgN+REYCo<4SIu4
zO*b%Kodn{P(RMjKfqbSRj+6U;n-Q_tQ`argE^z{k5tcaIK(w|(n+|p~+gpk8NMo04
z=1aI}@fM_FUGh3o-MJKy3w%6L9XRgT-FvP=%TPrb0EFtmuIlH-d7qmj;H{yRf}uU@
z@;H}%?JJWA0^HW5(m1ga#`*u}@7jgUdC1PsHVmP6<_ctwA&!C+G%k9UHH#c#nU`0O
zfn$EAzHwdNxvKj0iky-b#p2A7HApoIf2-(e4Fi%E|FzvI7x_FkA%SEaapU=mb$neZ
z99Uc2KYwSqd$?8ty0vN=r3QO-J-*F_wcJ>gAwPz{ZJ=ozBlvxY+6-CJObxyLLx#Q|
zByi(aV+JNFNG^g4Z_<N!AYHx!r|g12d#bSC%d&3ScZ4I{_J%ZLCY{e;y3QM<fHnbk
z+K;9A7AG3tjWg@Uy;u(JSz1wy#Ufju?@KA&uWvikcgFU}`vnEzws0gSDmAT55RD?9
zr?C@G1Nb6T>toUN1sjFy@%`s#Yy9YKSMcBdx6+E^ldh`X>}JQMtSUrALtK3>T<14#
zCao3&3(L4wG46VnzZNGJDHM+8jnj8m-Vu0v@Yv$skPT*IoN*@s{1iu3A29NBs8{oZ
zTR0f@-l8=Ixm7EuwJ(49x3kye>$qSRx9tj&0?TXrj$4;md=dy7Q%T>K|C{h_up(sX
zVzUp8Ac_}gn4RkZcD}r#=Tfz`H1QC}^koa*#@Dd10U_@VDh3#S`D`Lcg)vmlB`z-R
zed#R1_l>u*ncx1TPrSQv!qv}lfA07EjE7k5CQ*tpCWsWw$*lN;T}Q(8j}UgW?EBOa
z#6|m}#9mib9|0RSGWSpVdPBWL7VwB{x<>rQPsIO)Eb`)AXGAd|f}|3=(kHj$dzNmA
zR6a%ko$fFuqwqPLm_9`X7JTe?6<x$CAyR!G82*^q#q`U3!_nKp6*97w)r+cq+wgwe
z;g-M0>$_v(!|~yvh(Die-5%?0<m~RN=*CsYB36I+Gx;nbCmVYB)#E}6&6M#^rf(`W
zv{dNX_dSF#f_x}h{n$i(!#ty-VrpN|ti8-q!*<Vi!eD}+MhA6Cq+5Bpr3RFp6mZL!
z1;&v|Y0<$OYH1f{rZ&;4x*6jh3lWAPT6N4QY8BG9h7!VQr^Ur42t32|=FUQ^rrUir
z>+D?#nB&Vp`A0FTAsVZk#I-intm0nG7OD)}_lssJ@BT4t`nLpJ@2+(aMR2&%kfT^~
zst_8q5LEOG5oPv*=3^K9;1t(8(NS*wFW!GnUsm~j&UPMWAQBbOl&Mf^Sfc|mW`A^;
zWm!uh#Gt7AqXN-ZF({f&MRRV14l<1+hutGw`ky_lj7CXI*3a*+nkiRi_J(8c$`(`S
z<%^Wb%YB^Bc@kyPTEL&C*KNaPAD1NEanp;VqyeBeB2K6&mJtq<C6(>rggD%feeE1r
zyAST{S@n(^pbU`L#!r?RVxaVwpZ*NoYBE_=h|o1xeI6mJ0ed18!qDE=XE)jT)aU_R
zz!CZ7rLffk(Fgb%pp<1-k!Ig2zi07=%w$KL7Jj~sag>f$g_V=Nm96Q>>^*H6#qYnM
zJD*H9PoLYxc88MG?zWQ#j5-5??;VE#Nd2Kci*AwKa8DD(*~5-jMu~;nuK|miB0ED0
zQSYdvBpW@${35E@nf^ON#ZWtY0?MavN%mtgO`YBQcWj<k2L&Oj-;!`VcTU?LRv#%*
z@Banvl^VR)rLTAxa7?{X_4{)v@-)~cgN&RwJzuTID$-QD27Uu;*t7DUeaa;Big8W}
zmv$0bU)La}?hPBCMEU`+q$P6bq!mL(!4_QPepZ61JjWz;`+C*$(6cV&r!ye#(r!}N
zu)6SuuS;!?DCU6YaT=XkYUfbc;>w3!Cb#6>T^vTPQ@G*nF0VKD_azmA7eSm>&si6@
zfUbrfk+5KNh=?R11MJ`>#C*xTKHq_7<O>GorhI}8lU>Htk2z-^p~#z(h&~1%P2Z4O
zJBQCa;$`Anm{UCZWJw4kWx^v0iP2!y8l<%<RI5W1yas1DV2d9dAgUCiQWIKwRexW&
zZ}w&F*3I=Z@j|}j1H+d4v5BjKPg*`zPr?~*;<$z0^O{;s6a-T2p)cxHOtYkQ$n5#3
zzO&>7m<ZC$C2*cNwVYo@RlftE!9rh8E_T>EKgtuN?XjelV~BlVR~WV3o^-}##EAXY
z%&3qEveIwFese0*`uHfJ*dU>44CdD$OZtf9(b5%(G+7T~rcvO+N-QW7G1wy`kRzRg
zis|jR)Q|mYjQzov#mibXjb16cE{~7S0sbOS6M;B0tlp@UQIeN2t`A^O(<$OrO12;U
zqMc1_Y6Bw(n7$PLM|w;bcP82cmNQNv+Hg_C+2PD6UuN4tVKi-KJJHH(GgZiS^%2t+
zjVO13bL(zd$2>S(-#cs%vqJ(WbjcmX8ewFuGJbl6{S%(7kR``3Qr}3@J#K>sRZL95
z^5EC<5?+<OUT{`Quf}2MeiBcwC1vJ_J#iSly`ZtyJh0-JeC{1L=^}ah!(<*qM#HhB
zD)8SH8l4xHw>Sw|<-spWs;vBsai8<ck7hi(Bpe-H5lM~{aia1JToyuACthp|nUUE`
zR9*#gsnLCn<o$0G(68tanI$pJp>V+h?+9jg@6$QzrV}#Y*cM*?XPb;;<rBmg#uIyx
zkt)kq<SLuUFmr8hSq6#jw^b30mEQ#3%_=PT+FCRrRvcs&6^Km4Dza?>&+q6or=Y`r
zw{iEujXP^kcXTUXy6&U6frn2<dbYynX=wb1^388A=wA=6VtZBvoN2GH*SpfdvD4?1
zwM`%rYzVzwT9f2Qn=7Zwg)Xb}^L{-M6PK6B*)w8e`|Wp2@p?x2^!{c3^*Rt~4=gAw
z{sy>~wv08|J@93th_<yx8msLdy>`k3F4PklnNmM%N34(zjj>Nx;dg)RU+(bC7P|{x
z+e;dt1p0&|i*TZAp0*jpb|#;1eO`{fEVPN}I)X_yvuH=Sv)n6EKT#|P4sLW&mtPCm
z;rOZIee)!Os4sFD8Fwr7jO=7|9i!b)w#jt_N+%Za7;OEa_10Z&nIE}$kk!Tgy;894
z{?eWO?UpzqslmL3lS*)IY6iomlmVdKZC231EIj5PsMb7L2ptHb=%{|TWob}QCKbQ;
zdJbA@#L%iE_cX54WAOp5-73>AuD)YXK1<Qo_4_&-J%``^k4T+24xRlj@th>qDEW3m
zqxPO%jse}iB0WS(WS?sP;)WsjM3lJ^kxtH4svSPSd}#wn#{StL-eqq)C2e&i<Q+FQ
zG{v>U7W@f<Zp+EyC`8wH%F~EeoO<)xtr8hT7!uNw_Y-4PTRGkyX{32-V9_z(k&nda
zxh^R3-RiP$OwCY9;~Kd<J;o*VGI-hnpttS&Z#<(BTj+(Xi{1y+r&fsxpmX{f!|HjW
zEk+ou@fxw5bK_~D@;Q6Ow$$fon~+?DP(cA3XKv-x8TZm8$z#iRO3Vp&=~>sLw7k>W
zP|Z2;i0WB)4G~KnAKHINHscwEr8G73QSd)`>i8VZ?=&NBwT^=;ejl4XxA+vYT)n&`
zG<1k8hicK`EZRH#e}}xIH9y!ue!;kGkJ0KGVcl;MXkZpk6A7!z`L_{}*IRk&$lxet
z?FVW$>RxLU)V=?PKiEhXdA#seHe{+>a@NaAQJ5!1D1z<Z*0}V7#h$3L0z3I86dgGP
zcnNG+Vvv?d77Gg4l%i*?zDNl#(2kZ+70Ydh$kj#Hc)V@HPLPzvmh_AIGpGo{vGM0(
zU`3gSe~>roL~cITe*8Pp%;X^w=Rr8`e!|@&bC*5dG4MQvV{Uvz5R;$22wj8B|AKq+
ziH&ZoWLs61!36dCuY1szt{wS<xsB&14pU1Ug4mltFwI17fXpkcQCl}p0-O5!Hmw{H
z7{SYD0so`dyKTUm2O~7NxXd0_AntH_bU8v{%2I;{;R{}=*DmHQn$yS~7kE8;zxg>o
zqzj$fF~gkJnH`jM%c<?wO6}zI&d1Z;L((jT;rvsZU`pMhuISVQF5UABbus8|tr4d%
zh5Ie3=|*J_=h$p>m8EedaZEotl2rY4PTVG_eCI>76_+N;jzT<y<I3~fuhcS@q06@_
z+S5F|g~FwEsn;!kh@?tD)9vre`Z64l<@8GRW|aaivs-Nm&jX*<rI3=cYM1wNpVnTg
zpDteyI!g`4^jOB2%6M;2M9p}F%{rK|f2@1BcD>#+yaMHq$DLhkPuXjmVo!a!?_z7K
zydTajVFlYK!N|#83fd{X2>|2hE>0jf`;y`HWi9C=*Ujzir18LxMJTs<5<V6aHZ$})
zhF(i2A<Nu-YWyGfzg7#sq8rWQR@-MqRT59iM=1Qi0<e>DKKax;kJdhSU^>RRoE)I|
zVN^YTr1J6p<u+ij^)!HvN(!S#Lvg;Lx5iAEf*K1@o#Sw9M_;<$%FV}4lz@hiE23ay
zcMf(H)HsPH)9BGXtkx`PXgYh<3xRhOe*C8oOqNUs(qUNG!z4u^1i=YN3=;Y3bC)U=
zAUtvlB}CM{{(-#88=bWeJ-;FG2l;^y>lPE)=D%CEXW?lqb7*Qa1QUk@mKYFA`RTWh
zkEn~byJ>_Kb7@p$D$~B@MP&tL@O78Bo+})=?D|!6!u~gjRR^{isdfL}F^NeiD9rnZ
zRtH==Sm_hc*Gd{neF*dEA0?e{w^AZ=Mv91@w9=QH=G&i98@++sl@i_T^+Gn;3?+Z8
zfZk_rJ6-bTH^LtyHni6Imn04zJg=f~k4#^@A17)5Tx_2+h;n|Y<lB>WnTza_!0XG$
zcc|$vXfXe6lvIw#gf9v9S`lZ#wE25;+DE&CU~bKXub-G07e*?oo4NyB`8L54D50rH
zc26P|$rfO;_>zMZs{mmN+G6xxC_9D)hDnHDoj?meEX@>(a$w5Mvf(n;B(|n9R(yme
z)Xlrx(%?8bJjy-oezhKR^I+2!w=zm1=Yn8+;LqQVrnH@uq-q(nwZ8tXu*X69_U5gG
z{+!<NeL<w6i{2rLOVEZ3UVIZppDWR{`HEMDhN4<7_LN%SCpu19r_Ck$AMuHL{eW*b
zKbJ#_VlpM^%Tt0{*<cb?vv(CaG3Va@&=m_F=!Xpx8*J<siEmB33%S0{s>xZij*->-
zUW;QFdYf1YW#MV*Ta2~U=UA>LqD0;F*q^0$v9&n}>7&otiq)j}E%l(m{V9R1ez`m6
z#4<+>!S8JXMnS9o``o%;UWv^rZrA<ut>xzL1&=@$nQM!0uyiM%R!pz*y&nai+wXQa
z7|Mk*O7hFfVzdi9oh(-t#5$WkT*g5bGg>Y?X=Nl+6?z8?e=~cG9L365YHPEkt#(<|
zoJE|ufNj7;k8DYtpHl)m*cTJ-EXag}QS3r`J_IH)AIp=c>rA&nySS2ihjG^0Q00cK
zgIL_&SxO-KNLNu&IPZ&~e%HVuWtO3PV?Jq{Ma9A$j8Bm=q&zC^Wi_2XbQb^D&V%(=
zj9Ge&IP+o2dnScC>@h(a?P)wj85^QwwFIB&=s%t$=IC>_i;dK=G_pN!fn`w7;|U#H
zr9rC@h|rW{Zv*_Ro-y0t6kZ0>tjU?*>2i??>)qHTCH1Yv>V3cNee3If=SJXt;IsR_
zeXaRe`5wD`uK+T?)8xO5DgFO=UONMyJ1>D;TjzS*>sj1glGt-xoA-U9r~cW)Jz7TX
z=4ub}puFqLA{Wu_Qy1m4h${Oo^wn?8T#%USA0;gaZv>~D+%AJE8LjKzMn(AAG*6cY
zxk749XXk)zGoeQ-l$f_Q-QAaOz`nD_N87{H^(f4Qa4}XA)ijiNJZfLY{dw#0$3!;i
zkPp}SeIK#db0n<-ZT}WiF7zV(>*@>Fz`WDh(U5~|Pu4k)tU_GL$*~GkR0gQtY_XAU
zR1*_3F5NhXn@CwN4P(*f1x)>4$d&>!e@3k3d;(qS$Vj5@THxqW5-HiTU+>+9f3MlM
z6S9QMUY!I*b??nzUh;79?<pN;%`{g`59`<&%46k=g79*C?4a0vT;ZU+;&XWGG)CRM
z0Abo5(LA;sv9_TWKceRn&QEIos<k4x0h@L8-C0uZhV;P321Z%g#>GWq6&!j&3sRLA
z+}e*WcH)gq74`61WJpQ4662;1aYj>Q4hh!lk4B=sDS4!*K}Sp{Jb}0}iet+W{Ip;R
zaCdbhmHtXjXV%2*wbaW7y-5{9Sk17pZ?~(tjCwME6;y8pfL(A#CZH2A4(#LE-PP40
z7baC)Z54ukZVx^~QJc5ZDd&yS{%zNDrODr+-2@T3IW>L*tWm3Q;SF#CC{J`*+O-VN
zSe6k1VPOOW&lRdr$uQN=aUGpQ3S{UPqwpE+Rsqv@Sh7Yyy2nvH*PpGCwAM!ExnSTr
z?pP%3A6`hn@RHacp_wHYN=AGMF&4A62n=1!uRAu(XA)nN9#^9TRNQwjf0x?Muzo!i
zwGak?t^(TH$lC&RkzXsZpIgW<!*ebp&{!R-=@{N4>|A4r&@f{0cuJlAsZ&9c5eL&T
zUkJC=>OsB%&B`ugYR&vAQewi*H4A8n5~lHu29}mGrJzqMoY=7M$+pzL^pfL=$odT&
zZ5<R#)oEn=VDi3pDOC=(i!cj;!YIvtIc|>?SMOU21;GN(*cJyn4$wUc@xV9Avv}3~
zEOF|DxpJ4yZs|s>RbBnf;(~6$I~uyV%-iFX>g?89mMV~^6Pm!)@`-W4WlEJ<Zm<6s
zDct=du7x3Bg>W|%IFUpj{LxQV>~=6cj{oergS?oLaa~3g>HY+g{0MbigK>t{vgorA
z7#SJ;()-H~{}hXR;rMGB2H{skHjJ6-e7KX0%&rrmpsqr2lt?eAFti>woh*xZW~3eu
z|4$*vHEawOXG6H=BBrofoL~LRH!vSc{C;qVI0CP)Idu*B52wfs-dZLsNQ^sc%z1xX
z!r&k87wLJg-pF|-O+_vebp5g2nf}gFP>!#&!H$S!OvA#MdlG#Q_M4~?C15+L@H_(O
zRb*}}owN#SNVRCN<{A8N1j(@CuE!agJtFqZ7kB~iy6?<}`CsNvFVe#Ky)U+LNpuqo
z`kpS5`+P@_o7MnPuCE~fKi=2Pz{fP`rSo4v%D4=N-U_X4w2Iw{mS4m6#!b#1wxa>4
zxJ%FK`u6=N%<8rE&Ijbz_<8nRcY#(a1w1*qaOKj+eI0i%FV$%0K7&8@F<3eE{@&ib
zR?fp48`Q4JlZ~vw`Fn|3%BmU(n@1k9OC)0b&tW#XD7JZWM~ypeI7_Dr#2?)E-m3@d
zF<>DL*KQsiWdURJ#7stE4{6)u$zvPzJfq7jf|nfm@zJ($vV>kmtKuB8433qe!*_LY
zk$(cwk0ay3-)$gdpI~)sWTeU@x>KWDZbW1u9Y+|1|1@#!&~v}pL>)MNMw)JQ)-c=O
zBx1mKWr$SXHYC$T`tXi)>=a3Q|NSVNVv~E03&n=(oLZ-#BbYs4!MvRfEG7oP7eBMq
z=xE_Nj9J@M=+NDsbSP!)$}|0LZeA~p3+%t`xD~l63ID9EG|dC7+b7AzQE>iH?v~vv
zri%W3q0@9*kM^Ek$l{O4@NTezG7dvJR<9qLAiiVTLrAx?m*@{aEq5n=^-h8Q)6TE6
zao)E!t@6t@bWXYfH}mle<jSQ_i;t$o4)!l%ucxm+Kqj=sO%aD4O#<JUVd<*30t;eP
zFp*ODUB73Jn7^WaMNR*NeYmUjQj2R4qbk3IuP}ul@sEj%;Y(p<_n-zOYFa^vh9>q;
z1SXa#6!uE9=vzv!*s4p?1WF_<9u<}8&-KPC^S?1Hm{Z=&&<`DjGptH~=8H7A>u{g&
zTyp=kb24!AeQvWAfpkXD=+&6XXmU$1WZ(z??a8OpA%;E*V*a_N_D6=xr2i&(G8hkS
zmts!?ZXbC(l`^c`o^M*)UQ?<$iK03BYmJh7HNupkJCh9oJ*mS=4j4e@LY_t@9ibtp
z@j4^=`E!dW5?#CWpndzIH|lND^D|YRCQM$Egkiny4H!HH&i`o;?X~(W!vyNc{8G7;
z*S?xkQtTO401+^C7?w=Jlm+Qa;xSFh<CLS*wd=Oxtw)R7qLs)PC{?g%(iNd-+Awm~
zmlTT}c-I(Nv<3}T8phk`(bZ^3E0M;F%apSIh)7@rqGit=lvU}7(Rg5{s=>_XQi?$1
zWZ0)aJ@n;^{wIGdJ(f`Y*PajlE93G;C+t!w?Lw>{yG%Qlbt;1kuk?vE=KZHMS01z>
z9{(i(rtJ`%O#aT+!1P;g_WxUe`~Uqzyc|KRKRMuL>(vacoA4BjxRJ}u&okY<vxv-g
z*IiPbyRZ24d7dfJSKIaF4v~-4O*o57zFj0C1t+G&6Q`k_ziGVHJt*|DQk_^+R^7yC
zGSuqNhay)il`e7H#40Ngj~&Yd9O;1P+)O+}Zm>~CaKFBUMag4N`>GBkQ%bC&G5Hfo
z<1^@Ynm~ev@;n^P^JP<|AD3;9(2TsrZfFysJ5;eSebP#Db_ROeDz@ox+Mz-sho{oD
z_AtEEW(1A1S$o7>dwS6Q?Y>1Ic7Gr@qb*}k?T%?!=>tRt&&TFFhYOr)<P8u4UyhyK
z1#{2iUf_Yflh>9nvz;&VfD1X9%q#Zt64xx8d3o~qt`xCQVNC1}=+y+woF__Q8=6F;
zPS?McOHF{LfhhKRptc|MaTAEtgGo4ls4CP&gH*)t2&A&awP#rf((w9VInC@Lw3=M?
z)+;3<w8d%{hy=w?qZajqT32ux$Jz`?gG{>QO;bigY{?`G8(NrD-`Y#?TAxP{h{DWM
zf$S6gH0K**IAVCHI-X4A+#&<9MlaQf>l|&1j0O%<C?-&BBa~Au=+vAAPs-9D!&Z`S
zO`Y)#z1b;OGGQUucT%`mf16EW8AH{8`cuOeO4@tlYw0MYkV}Gs@?(Wh{g{gHMeK?}
z3gy&=Bhwg&q15Ok{pZ*dCnO+{UuJN!%*W$7y0e)WCl>G*ner)Xw(L#mul{_RKIfNR
z=d-g7sVSb7GoP0-f}Ilow&0A`Z#+&VTE)u}3u4)}`tSz&HR+kv?e1&s6a4#h{<5<G
zs&y6pobhNmq471f$rt8uJE0qTzo*^#!0a@&+{;M@O77NX;Ibk)tdi-`qph9D7^zU3
zRk&rTGeY-ne=)2C&bL>tmbk@3t?-15f)t!EFPJ1UKD&g6iU(QZ%p8sUES*tA!RpDk
z?@Q8V*06$(%2dqfCbu-da?#Npxrnd^x!8WC-PXYzS}Sb-=o~lqTcy0bGyHqD^va0V
z^%S%{Bxa!^^VslAY)!HCh{AvHj7;Wta9C)!Ic<!Tw?F&kjAtxpOhCAKEQXZ-zTY_@
z>G77bw%Lf&&~%E2Tw(LsxZxy--%%x_cl*8AGYfuZ*k)K73xcM1{aE_9e!+KXH5fWk
z7M7fi$D-E`i-c^Nj8i;?c5UBkISE2OU^T4arCg9>k~ViqqGV~-k69eTXcTQ}WIiD-
zqC}(TOmuG)ffj#%O*x5(@6`;7@1_138(KWaJG$s{e}SswWfdL{dVEBmN?x?D_te|_
zViH1dRb>S}FjoJgRf0e&cCK>$phVV*w|TOxs@cCMdfRyYY4z?z+oQ#_M>pUr117P9
zZH9H!Oz3xRn-$e=HZPw2CldRNu#&2uhxV4lNRDCn4ReFPV&N_3HD{?`wbxIGA^U>3
z!e1ovu)7`S+OhH)iof;(*%|iq1&RI84p+A~pee8$-A_WqJxUQ;iWBN-G67=TIJbDr
zPe=VcO=4~GW&gxuB)@qO5XYkr@a6Y_Lw+$%bJ~(1RY~&D<#l;(+l4gXd}{lz6t_b*
zrZ7Ty6#9lYwWU&FGb9yOWoK3?|2ae%hvp}if@CFzrea2kdHh~9GA+R#cP)IL7$ZCx
z(5ea+KI)?%Vr!6G9&Y11w_jsw%)8n+4^qDL8$K<4q}<ZsuA(%e9i<0;Tdw_vL<a=A
zy3n(BO4rA(_nTc2UEY**8`nb)FaN|IZJu6EU*EpYiai1K&Fr0hw;it{8Dke$Tcs?!
z4ip`i5v{kUE<7uHr?2Ofm(P^`lnj6g|1%@D%eHsauu0?}e=Kk|bh*ec*<;3Hx{kSe
z1*mHI#G04Fv5cTQBr;vh!;4r@1iK7QB5(d+7Nz&+^M1LFF(sA$r2l>Nc%h%j%u`;!
zmp_2abzH6L-VLX94nIFS7YdCyHd+Y_u0I_mpL-0tq+1&IxPbdfdfM~>r5_L=CS4*D
zr)a;*C&To|&^&COjv+$HIG|wa7$+t;!A~JamWWCVM!aFY3P>c$$Pq)}(6Q@Z@s&bs
zlk^I$sY`?IH;4Tc>`%EJu;cfWrSS}uZ`wnmKL&}8-MJg#A=&_7r5zG{rV3sbT{MCb
zrPL@y1wpD<t;5CEm<pXwEvm@#C~A`PvULWPL59+|Kkcf=(W5Zqz<R9-8w=1g*`VF3
zM#*ZB_FEaQ{t=(6>ej~v{)Roi$ee4dmzQ&Ade_{6TtD~p?sQXs=Ek2SiHsJ4IcKxC
zo!xHjz-?@%F!4@W>L*HZn;_H%eOOP;HqLT<=UC977^}>A_Bt~NG|D~2LpxHtzMqg^
zU|T0dC$aUHBp4M{yy&ki!64$ypaD6`%Tm$6nuDQ=BKtRTocz4`Swj4FGSyB>G-|#<
z?dng9Xqva<vxGi|G=g1+mCj1{2!9?wE|ve6xc~rFHEZGD@9)J&Q^W6fEejcG6mXF4
zzjc`ndeja)w;~@=lDR>T1%WLj8I<anG9f8EI;)@apc!L1V_m%^VDgQY?ZEgHva;Qw
z`S2Ent}UW@Z_P8)(?DB0uj(7hyP+KZWxvrIykH1iAQpEmtWrpAT-?e3x0Eo?Sm+(x
z(8V>8Dn&vUlN^bzzQzMwpIkrJDiSFl6dj*$5gP)+i3Ka$DcNw2EkO)|LU%W_Eb_28
zi+YkB!DkA_l#g4E$f2mhN|Tp|M)F6^K{Jjoa8ZW&$G^g(?dR<(q}znm32#_$SAtfg
zAtPN|iV}7HFUOg1w4MOlg50p#7Bu#{S)bXi;8sXx%%ID3qsfWvS4U5M5r%1lD83hp
zlf~HUmW8t$<$cDJ4wcu{$J_b9@k?*OP>Evjs;!m{5?e&dKs2XHP5dmBx|f#q^aP}+
z;@iWqxuUQYwu3v+S;;5ey_CgG09XQzyz`X-ujXcj1giR2qM*tmFFGS}Bur0VlP)5{
zF5?3Pz2BasQXxcAmQxbtLgFN)Ei<*i1bb^q0kULK7N(ObttEECmLl1QJw}wP(M0~}
z=LbYH?{AsuzGyf`LtCpXxw751#&{1s6^_YDFNyo^QBOr%)H1<W7s)+C5zj;EhPM<J
znXtN2O721K#WG3T(aYSxkP*4Tu80GxL0b-xChT?Z6(AS@n0Ud?ThkW`{@Civ)0dT8
z7hm7up^^E&Y}Rew4<WamFZd^G?E{nN*edgrBqJlb)%vWuN5_=$CB$3jA=LrL4$B?x
z;rBa<mr=xG50h~UN!TrwBGkW6^awxcl*<sAEL;y?$s3Z%0dBy72M(1*s~xuf7F3f?
zTK-A+;p$>XiH&WyzVf_M;%(t%{OziL+}eG1=g_g??Jrp_c0xwt$qX1+hm8DW>z%0`
zhdegsp{BblA2LGHIYJ=@WIF$I$n2QXoL(d+xr$Z-;}Fl-4<Z>y)`WPygKngQ((Q_w
z=pvAbVHqVf?m#HsByf-!pXYb?W>gSL1GeHyNJNf&jeF(;2G!>iwq<TEDyxkS2^5cc
zAko!q<Xx}Y4=Z5Ns|0}PZ&5HYW7AbjzL|-37FCf{xRTnHiq^LWJZJo$)gJKfrkmj#
z;M&?C4N%CypSd1YJksXPqtCVd#K(Q-!&SQJ*N9SG4Fil`vsxoo@Wpyn&|f`1HHvv9
zRFC6uyujlH-!ketC@)#mK=0exHMHr}azuBoHu0zt^!cbT|I2obtt~sKr-RWP)`C=U
z`?pY45)%<9ZGPf$DVN^L+oygp`C!k0N)b1J9`-)^I>QlOi;h}USe5y0*L&u#`@g?o
zBTXybzV%8c=-u8Ttgo6vk-XF!iH13*WCLC9dJrp-L=b&!;b;S#xAQL&C%QtrSG`I&
z$K3Rnpapm-WVYA(@?yMC75YzR6D98cswL)guPu0Uu>q+3^<EtEXc;(t8e0QJa%Wv@
z)s5%27nPA@;U1lmQ+<&(PWxPy$2f;#*X#xT^YTLBTEVBug_#?%qj7@fkKmf6K2I(w
zUr~oAgT2i7gd|FY4%KAvTQyF<;CuN9xQQ$y{|%>>IMpmtpFJw&$}3O{Q<;m58!~m-
z3R9<M6c|x!n9S&r8d&bDn^;JJi?jy^A&D)Z#G(($XY(T@F&verRdmF5B<}j`c6&j5
z1p5-QqGSn?L>qV{QEr%9`Mwoc9&dYVydw#=TBkP3oWz_#>JQ`26uC%ye4Vc&jh20T
zh&+uoPPd#QZu6aHr(GW?MCQKV>ZKOTtu2y31f}y}*Y67hn!7(Lc^L1tuP#{^S~3Qm
zm?wgynfpmT+Fr}A*=u``v}WJ!MEzt%lnm_>E@W*F+deTHH1&kmY@kv<lHoKdO!M1J
z2+4vaw#8>NW}~!J!{Z_^qZKORYpUP;`-cEkzp>KHT?uP+6qji82Liw`wPBwn%Y<ZW
zXlLnuI?8WF=qZBURpN1JdW^*dmaBwSYS06D40d8JGK<WyRzoVR%91H10FgPY6@8u5
zDXUm9?EwKC!IDS6Hc_9^7&t$X8ojilM#cPR8C0LK!lpIz-)uhj7X2RsZj!bCT?ir5
zuD1FNPFsZWVA$#Ax1@HSFLjPyZb{_?oN-@(i<y}dc`R3Fk|8R_f5B2N-Y)WFHc$7t
zuS}EyXKw@V<N|xe?n__EUoZYIj9ew;rux5J39y?8Z&lt>obRv59i5&g6$zO7zrvrM
zf<NAEd<^(=4d=h>eERwtQr-5djfG>GSK$!wFb#l+UamVggw`I=qEBoM2X>$;(~YDk
z&2r|+FyX^AbaiIgAtEN)L=dL7%pUH=40}H~RvxGG$E>H+G>?%<RS(i^rcJdzn~vUP
z;<+tO-r>6yJEyhZU8@0MH7m0AR4QDX;Enssv<`bU+?K{KqOIE}*??RU@J=<TMsG!<
zg3wAKobf}LzT$^o)dFp*zl{X6JTVRm%A*pAG}hcLl&>g;pV2!m@U2z1<ca>nGp}f(
zXvUG3M4Mjx^bSYF&X*fp8@%ju@J31>KPx*|6EBo-pgzo^y?{~YJC+!*YL*H7Cx8HT
zbl;9eUDt;LY`Z`o1VTdYj_XS2^X(yAopO92M<&5DTPV`8-%5fm=0KIN_PMF(q&%s#
zH8P{&X4vv;X9uj_Nami;6fMw86n@?scbt^=;@3K~;Z%OAE3nw|CH{-FYcJGm=+V?C
z_v@NZ_;eJvb(%YDN-5IASZ?}<O6y{-TzE9i)ZeB*3`vn+YcxI-<vI9oRwUCz;(o@t
z-0V{-E0a|mx+eV!27~h$Kd|k68ZP+a?7V5%YAhF5a7)uSM~q^B8&LQ2{oy)8?vl%^
zAabi;H{Xd<(lA9nJV_{Ujh=c4{`_e)WmJB!+_r*ug7*QVqeN`g1Cg_zBX|5O>_c57
z>dQmGO%AzCqA-vPpkHqB31<(y_jg+B=m9#A9{nEiWTeO#z-Z;#kqsPVcXqm!yit9y
z4x~@8AW#Z=n80*9>KO526&ggVvjezF#kNg(BwBq-YwJd29cYwMj9cIJ2`FAcr-(ek
zo(98!{K|C;CaY15cxdN}Q{n|HmwlvFFOVfii%I3ge%C;YO=CQfjGNQp^zP$Kez(ZJ
zOZIK|aTcPT0!=84mD$y{&~6S*@NK9kdO`B!3-1MEsaNc^#rNSbsvaxe4@#%?6Iyjp
zQhi7Itk>zdPSMyE)lXiybqmlqehZDbGN0oXw?bo^N*Ts+zvYET;yU7h10Sy!TD|F?
zA)xa`&kl|gG@rn6AJBwa`R(|QF>|AHoH3i_-D)6TuT!Q*xLxPySO2^TaLBdQ8rgR{
zLAKm>D-Pl$Ozr~d`^GYuKQAG(GxZFn#jhO{Dq+dIqO(T27JBWVNKGqaiEWb4n|=5)
z&0s~w&jnchc{DOOeP-mfkY}3DFCy|AlZ4nz%1hX!-wDF5Er#TesW(yZWvqVWd9Dk_
zxTZcAO67?9^#g~0*D6K%EKH9%Ilh985Lil?@)+#f#IKVH=gk;ND-|aDxZofZ@NoI@
zo;2>{ygKkHJx=tF_#<We=<3}!V9Py%#el=`FOf7R{`W@KG1SyF!svdj*lKd~=q=MH
z_E*!@fGf7nXW+;Q4v)Kw>bxT;wW+o(gHgV&%eDDF!<s1CM+*VCy_$Q+edgNs|7!`_
z?;lqlzvWHyu_zC;w>qB#5(vi2&FL%o%X8;r6OdegMIIYxBf`)ERu%rgz5r}8aJQz{
zu_?-KVf*BS=H+fy6hZ34^HOMF_<CY1J4;|}S3FlJG`R6t(hSbG{QbJctVU0>f^?=b
z;xbALh#~?V#6y5|{36_lu^?>OV4yWmk<jA$OrXeQO~@^Jzi(MZPOh1P(cSL>3Fpix
z&`Xf{kWXqsGeJTa>^=;qDG=|Ku6grQeGDgxK*Qu$K>p8y)C0p)^TfzpN%s$}BVYH;
zUq$axZc7Ups!k;jg}UhLz1i6J0MVzwQ;AIOFM<1RntS~mDAGCm^prAwPs)sfZ#@~T
z%2=K%&xHB$;}9~_<$d&gwHjDJV8yY;&rhq$v{O|BuT4;()WOh9NMnuxX{mQ46r}DP
zFJQ^4C~f$oken?c*&OB^y!%UG#w(8hx5wBqlD_Em0zYo+%KVDVvzJJ8;Otg#?sGKI
zdF1-M%UL*y0%>3Fvl6CO;2x(&s~iHtF#n)qzU?Iy%oR5U$BKRC7hE@Sgx|2p$dBFk
zRz<4Tn#ZoU;5-m1ZN3>5DQ(>f9bh)IG5Tu_sKR!sNkzdN`KSIt&wqkS3&EBXq@1&<
zEo=vdQyrt0)y<YDH8NdP%0q_LO3Vuz;PBXJO~X%>6wmOa)*PwDY3`y@$wFxlI*HSS
zo|U4S&KG24WR|oMjJ5x&NaL~sktFB0-umQ($Grvqjs^ZYr?;q6XTYkEL{?m&QoG0y
zE5h4rj)j91oYi{k6klu?pKPm8hronOWm`>;YEsfi>jv4%rmH^>_H0jGF^lIWiHYxS
zn#_KiyP&T#f^R(UcDzs0a%@{3)_CyidkYCuDCdM%L)v6bb`u869GP*Z*<zTUzzq^D
zy`3V=xtbnO1309?I|7Yk3ml=R9IB@-(;~z0-#?JdK?xe@MYP?L#$As61|oEZP10K;
zwHp+6qSg>zastx62<f0Jg~B?%J0l_XN}A{GBR0WDpQlwIL9<xmY;?unc@nj)CNe)Z
z?&o@ALOW^v6RR;)evPd&@xNJKncpQ9iq(dMpz6=t`8*D+DodKl@c66)3`JxUE@i<X
zKpf=y!9f9(GLt#rkHM&DOlpj({wpJav^D07pR}32s`~3n78G5m>YM&@c@A=8$ebCo
z$s@&#eFIXxkEp5uO~!krUbv0kC)pGwWr5WOryaF61mVH*WaXo~Q%w|Y!XaY!it-jr
z>as}&gDO!K_M1N`h34twMPklhd@H5ksf_26@m#SHMb*CZDxn8FQ-YfK2{B=K5addF
z7|2uKvXCx#hzMVf@gQa=#xrLi4-|YJ@y}w+1#A_+g96`|4))4}hqWYRbr?+WB^Dk~
z4D`?*^+mV(`B|Ncu9v#^gokU1KHQIEDfBx(E%HYW`CM;=>s9{HC9H=;C3eJAEM<)7
zD=h(ESnJrRA%8ln_0#R^Q{cLT>Agw-yeS|`EeF~=F7`wWckAmr5E$u$?Wvib?Nqmo
z^S}1TFMT|v{oLy)$}k=-PfT0XZ5t<yg6+C@BlWWMx|?1N-6@B{mjYFnbbvLMq+Zmh
zKAcKcg~WoGjcd3r;FFnE_D^1_1g&T+;mC0UQ}dZox{3JAKI_lLwQnZjC$r`AKxQ>+
z#|_xIAd1nv`4}ecsy<Gb12WR)^<25Q8sZ)<0LuMQ-Wp})(tews=H6s1{s}cRschTP
z_Gk{rC&``Hw@S|2Os?VtL$Fn;X(d2fBrNoY>X9IOYdkT1voZF6fmgg!4qyM<d^>8r
z65hCFbJpsm!Gz7%ExG`PXA;ZkySTcJx&4+_6zgj>sb(5^2Q0;AaIGxu^~aDN9Uq_X
zYlJnV!uc@;1ls$y2GnfrM~|1j{!k9j%4*vM&i(jVd&Ik#Yk(pFy)HN1xjKseSG;EB
z3yX9QK`>h~;H6I3|4r2nl}ZE^>9;7XKXj$Mklbil>1W{lgxLBd-PZa!L3!@4?jKkv
z)=pm6$krERE(214WnqL3vc-BbGIAQ)$oL?TYCQ9zI!_|#i|{{79+F3<rYLtdn@fxh
zkyuNs1uKAiBusI+KPd$8tH^;WcEJD1!+?-hVtmm)zdCT=1j{h9^KSGdw9{u!5{&;N
zylR0^aV|3Oc+RTpbQr^{WBN`&9=dT<HIoc81X)K3Z`qZ$$4k>p&m#5NJ!dj=ZX=Au
zGoPv@U3wN|^d27rX+dOTN_aAvd{uJP*JGY*9k(kO&bK6fcw+@ZohDHQS1;vrrqwUM
z1y>L$?Bc2(F#Nd4fkHX)ENO3N^-=NNmyTcSMq&-aNDY54-}*a1+u8U+!){AQ{WbAX
zn#Tkd8<(g8dnP#|R=o74KO$c<HlNSA=vdURf+>4zkSVf)lx3{`$I)2_McKAdTtOO<
z?p&5`q`SL&X_oHpQo5uCq@|_1L%O7;OF&q<yZoN_o8do(nVpB7`?>Dxobx-JpnBpO
z0mkUCbP-qK!<hX+TNf*8t<YX52O;cb=IhB%FR;XS7rw&p)$9%YE7Ec&YC7Kw!sMj&
zu*|}5$s76}UWH#%Ts9)bTXVkG^Covp73y9Op9-%BcU_55*oq!;6r|5U=lG8o?`eNt
zEQ@WkmM*qvWlj8)5>5P@Wcx3BBgtaC>WJS>m|N-xG75Smd>#opVo=X9U{|HHAz3N8
zm~vjf4U7ohm0^r|Z(94}|L^1BQaE3&E$_`Cfw6b%R0u+!54olVS$e6pl&ah1L6n7%
zMA46bfZAYJ)u_wkZ_jm&lcMk^RoZr3On`HBwmQ9aWa~Jp1g*9GVirq2MAWreas{^#
zom2Tm${-VaaO0~Ozg*Gfqo1~NQ>&QXXE{)2Jy>2-G^YflrAExFC$L^0?DKQygXrGl
z7WYt!@Iow&jQu5ZX$48<*yn|~hPbZYMESIjWZ=%Ao0uJYZ%kRuJ(-KgzhM-?cY=CV
ze1ykI33opZX?{TFn#qWf7LveNcc-F%LK8%+jBvpZC=aGe>S@n|Ca?cFJ%C?r?vnJ}
z7zAA2Q|I>hsH}Q)`duVH94CJex&PGj;2)f_`LMoe`ENg?-Kp#lp9)t#JEek9G3=B}
zT3vmSrNw!jCI+l+<QBxy&p((W=-u}357lCgiR_Q6-rMD628w|5w`?Zerd`JBOK`<t
zjWQSb#_u=w3^CZ&WRSGaSK@8<2F9H}Lofb!n6oY(T$Y?fRAF>!kGLaDgZ5Pr=5*k5
zaj28JAKf~xRPSCt3xV&_nk#DcXL+~@VAja6R<GJAb_ODJb1BRHN=aU_k|yMnsBlV9
zFH!U;$PiSQM8zGiu0Aub3gMkzGx`{Cq?q9o_O%a<pAV^bPxcMZJK@a`lznN!WaO75
zcCP$yY<vIRm3V2?N$U(*00+U_Bn=>PPTqCHb9cI7?d<n!#%#cb%c7}!3XtOIRP$Ue
zAG5_pEQE+SzbTa+WvYri5UzU`L54z+?%KT1SS64XKr=AU*I+Mm%Ck_Vx!;uwfYnwp
z!GiDX({~V?{BNYDb$q9}-v9?M1xa=6!Uk99Q%i$g!R9%NO-X{zxy<Sb0gYE)I9&H~
z$%{5H=a14IaZ>~X|2B#Z*j8brQb~$(g!)$+735hZ_q$2q)6HNx$XB7@*KFs&BZKTJ
zw92LdSEWTy@jRzIw!nk&=NXC@qxkrQ+DU_cGE(#ahoszs%Y}Y#pEE_1y}{V@fai;0
zKrY<eU2d@E(vDu7Mh!NXYkwZdfu-qaD{6aMK*BiUS<j(`WdvhG9W%&L=0H5J{!-x<
zWMR30P0IMZC_L8u^#YR~&v~IPPs))uRC_h;Ui;NNILAwU;ng*;FBLp$rF?Vi`-iuk
zy!T_or+DP@SvsA<nbhVRIXMUqmvSU7*5obRkn+BgPp_x>dkw$F%lrGNUYAd>3)}1~
zLpfM>i#GC4$;_f4u~F<2eL+!Me-1kN*=d#A_I!c+=%=fL4y>oA$Rb;bwD&%)w=vZx
zJh+7{<$r6waF7AkRnMTSVM2#hKRUXKBz1|uoiFK7`;+-AxI(@m<WP31Hg_{dg4m&H
zv%XKU?^dxC#X%C$cFfh4nA_NeWX<zS)l^-cXhcle?3K#9p2X5&$sD$Sxlf-1hT>CS
zoRXVDf87n^4egSIZ)Vv?PAsl&uF?7m;xXpUjN!ddv?#+3suZ-Y>fA`;1fnevFmtYD
z313>THzOYy0OwA0Y9ESt?$dx1A7Qd3Lq%>Z!9u}%l=ZJYBza(_8KC9p7a?Es;$6CM
z_tN=09%SnFKWPj7nV8fgIdJc*Ts;aFi$3z1^95f3j8c?`jWCCdC!7i@6+~ki13dtR
z&b81ewP0HOeKMdpNr&divA@j7Y*q1Bkc>UHTau|@sH0$eiHu_O5myqnRytEb?;CDn
zP%<#=3`F0J9A`$f+qoE(L5Hwcy#^!_31xXr5>cY#SwQi&pXK}{VKb&m<=8q`G?HJp
zNQ;&~wL9A6SP8{EtfYUSk3pSl);fvnFC?*8{xmp>hEc;Gbhh{IX#wls(}%qCd;j3i
zPqT}Jad(8fAEL(L%T<G9v`s)#k-sAv<`$*;(9{-=xQz-&ktE})`0|8d4zDo*e0!5!
zPo4klH~-vJ=w|0+tLfs^<caK=Jd28KvjKOm?TXEN#Ix>yo*30+DI;;I8RDrN<niBt
zYOyI37k<&JtS&7+0-Qm~`hcG{|A<3pQUlmQ@zg^pfu|e(Q40}(kA@|ILUb=u^qGnP
zRto;HXqGi$1xyjQHY!F`spj>(R(ynNc=m_&5-kW{xa8ESv%&Gq%1xkgO#yo>^^jy^
zeQ2HGWq)2kUhD%uU{~8azb6KE<~7zZEt$yX&|XNt!)xZ?aTwB_lN%wdvx+!E`TZoc
zN~#G%v&F_f`jXWA?NL+}B`1f5Pw@?(t6mq#SX4T|Ey7IM?F=Yw63H2JB?Re%=I%*V
zEeNrw4Kpx3doBC;-|;M|@s}6;3E<hec8f@gFM)hCF;vsB2YqEHQfUi(5&3Y|J-)I3
z$&9CHJ}opf5V0_O#zG-Go)A4(6Q9O6`J<d&H5sW}&?BljsM<(y`*q#mSasMi+j3wm
zJ&aS094u~t5iJd^%(o+9)pZAc4y)h3YrcYq%Ly$h=g7KvKA(XA_uP@Ww<$BJ;tR1|
zeoh!Z*FT*(P0zO@nsve58u=zwW1HCnU42*BtDj?n?ss(gS=rdI;It}zp6@y6VAjVc
zXRb%L%H$$`u}8zawUc#>s4B)iR>8q}Gd_KboBsq%UM_V3#8S}DonavAwT&R3e9_VK
z)DduFZ?eO57W|d@N<hn%)cs^1u>z{RYUcq{XZrp-g6(#pzr`1J;_dtL7zV5vHM{R}
zoaekw;)8U??V6xX#*Vai7`Jz2<*1*jq6=M4JHr&th>l1zOjEv=9LV?;-$|@vn2W92
zxO|m821%1L`H34#K$F1-^G93FBv(#~&CZ3MIYIEdW9&mMIbvZvZXYNN$;#irDQsg_
zew$+B7ZAYc>rby-MDTXngwL<85W8qyV?eZ!Luc{D2l|4P>kT6&yu2@02uK1I+tUpX
z+T%wp5*`A!4E6{J062XZej9kaGc{U+Ji4<p`vQ<K$Txb1Erh-waluLf@s3UdrO%fU
z(F|k+*F49iExqcatg|IxNGBFj9}o7F`g}l}b7d-}rsx{MPDH{;bF^Qo`-Pnl4s++d
zYCr`<K;vL1yCJ^$_Ey`-$ZQPHs0|vNnoo<f%|(^M@%4&8CTxk6U$x71Nc?z8JMZ^P
z!RZ$4r83WI9c-Ir;XI!8lx?#19KhRqRc7#K$9QWA5_dpf!F_a-ynC5EZ-=)Kx?ZVn
zE|B*Iaq1+<&MWIT7T=y}wz<$oGjxv2`hfVb-Li`{<Rots-YdxtM{33Yuo>5wWSSr5
zi4*SKc5jLF?;j4PV6a^r+P-Aj)02LiXj7nAx&(A3&8u$v?=o5hZw=M0j8uwnSez?%
z<@FQ;8Tn1^Yx>voM<~Xy9>Z;s)K@$Rh-i>wRzI7^R5+=~o|qKFh?ic>n@oKU)01qJ
za@Z0@FEj=OqOpaCqXVN4OO9E7cG3somdyXI2=frS3~JQyrGrBX1x^btXqWzs$}J)K
zGuU9gX+-Y3(Qu)xciqta=)hpe!!FGe5@4{mu4FTI_mI3CHVH;!U=Du0Cxv9sDMVdy
zgpP6ty<{yuBn00E*R$#j%+_;_yY$l3VS3yo4*|_VDitj8QG=tI>2!2)5jdwC-4lY&
z!cbR!^RCtb;)&4|94ofh4?)iq-ThTacdd7mt7`y^(uPy25vx*v(zqv|fxzP7&7AE@
zIXISvH{q@|FyQBgk6hYN_X<iVw!ZzsU}p1=HZPBns;oueD+vWEoe2N__8g2QnBwOT
zobdZkEsmpj{qrRBa1ZZMw6sbBG=3_JY2u=ZQYO(NV}mGlQ1KWJ>;K4ZzEG)(`k~6W
z9X*o8NDuB>iqGj)#y@xJECP-4{J}R^9C*%~>(*ccjCw*_!If&sWK5Lk0aY@2=?p^;
zDjVO8sB>ch!a2+)%-HzlVNoj8lE2=Z0;TMXO24)TeOZX<ENa(K<kk_}uEG;J;_K=8
z1x2s=T3V+=Sdui@iQ`)RpIFbjU&vdWHbC+p8&BC+oYcZpyaA(r#10jZ@@=-|B`@gT
zN-Sacuh}EAKmVqbN5@(B^gwwVS1rD|xZJ*cB_;If4Gcc_gTqP&6G@Bgi9Rda3!=YI
zZAEYkPM>7cnmal|H|Gh{{)E3D^=ES!tb??R7l}1!-aPR`fX3WBGaG@Hwz76wYB=mO
z8O5e47jcROLy}f+py?MWPSOM)SdKS9*2f3-lruJc{%2joU$R;NyB{2dY@`(0Vfn>%
zxG|Gjnd$PLTyiQ2uhYkG$H4Sv)cr8we7QsM8b$GP9|>R@YmIB*AtXX<kZI2dt_dOx
zB!EL*&urahsGdgSx&Eheas9f+mMA`gWWxYA6+R!f7)?2zT)!(EdKxmzjgh@%vw0K1
zW8J*iKWF!tQ?@vCM5Fai9XADr!9ZOwQgQA7g*TStqy)hqq1Cl3v9j1v*l#g!aeN#B
z80y!FXfJ3_$x%GAJhMx%<ngfvdkNNT;^xU)ryU7j>#}Ye2=ZBGRxqVGEFp-l(1gwe
z8S048w;PPVOvVfJt^oaD;iySA6#9>}OX`bzu0T;!NsB=ez{WgreDr8hwh3Wxz{Q{m
z41l^%jZJis)KlfnK>}dR+5SDvhKWD1<cJ<#L;4HD35zfXBzy)2xmtC^P{tRh)F(vj
z+J{}7H?JNm`ypRAvCV3_U#)38bv5dlSmttu1Mu~SwLP4&j*OIVMOxOMh6q+)Kc}zG
z4*A!nY0Imp{q{VG)F1H5FZ)5VLTRgkJ1EnE*=!$tWeou=jzKO^q?1$guT5$68Xo!)
zbC}SqoBS_}zpmg36$X0?`4AmylMI6nou4ar(CEBg`YuQXdB}a<WpN-8ZaEJ7wQ}fS
zt$6)XWoz|$O};kmW1mQb4^P6U4z$GD36Bg<(_@OVRI=h8LoLz3=4GMA<4~e0IEu1}
zJ8$L0j>Mjxf3Ft`Z}PAoKcLS}Y|UAV_|5;M46OXd8K<~XT$W|9{q}20s9PHxO*BFo
zpA1TP$_B9=JE6<#W2z@4(W1MFbeKb)8N#|}$f`?9LFO7j8jADF%AGXiU_%5}TEk#0
zx_5sX`1<PHIT5<b`|w~no-!D6@%lDaMU5xz;4kvJ$Hvnf;pG}h5EDb<*vvcv)4I#q
z%<3<ECLK%pIXj(kgxD(_jJw-b?<f2Z8-63w+I8KArZLG^i>1F*R3B#Udir12kJf_!
zkvXk($*#zSd7*QLyX5XrIVk3Sh_aQFlB`4@YBY$3H*f9i{CoX3Y_0*Qw_^aD>HuIx
zj`T@>nTv>%{N&=cknOVza1D{IOh<@rM33Z88THvo{IjT3rlN#o)K59S-=20Bd?fgM
zJWMg277^NIPkOfg_jnK;2F=!9nxrf)HvGjOc#Bfd{nv)Hp;SQ%q6w4_T3+Mi<(vxn
z;rnp8lA7gQeP$Fx=<|)O0>`uKQ)@<+>14lLcaRp0oOa6E;#6WbN?|TB>vcIc==DfG
zj6$bm>xXc<Vxb8Cr6(%}QRi$gxnK)lvN~`6PyvUKDf@`a_oftDELT!>UKYm+%t_NG
z3Rc^=x1|7jV3pT738-g?ui?m1k_-Ksv>*8GsSl4QWk2hu&0D&cvhJJTh}hA7=BV{B
zDe*t83~d*p%!-@(X?>!L^O$aUuB(_oyn<eJFHRn(ba%D7UAUfpQ(2fYKq8wFzY83W
zsWP@04f|>Pm~0o?E758;{`MGafn0ue?v0}}2gJtl4UfE+(9kbdIh*-l4Dr>Q%=kda
zF@mjNi{qDJ*FK-h43*NgPFI|8XGpi-O-#z%VrRfZc0&VBd3n(sOHxdRQYk4JdDv0+
z@bhIdkhQI2onGnCu8d7CFOMz=1FP^{vZ((azw$A$#Y0_DhQijos!WBy$~>51;PLs<
z=9iEkXsjM7u?-gNhfg3F(@j39KQPSkDX5kEy2~F2bnSw_<-cFx5+OQ5Fe#*8p`_|U
zmu8C*CVnqJn=PcePxh{IT_xlC#Pl4SeL-MlTH*>NCB1+l7W=2*O;~SqP}wlCq>(mo
zP~P9^^2$*`YGS^SeRGCiQO0K2H`#Pg?fQbtM{rHUOMcWX*SOW+Cgz>!17WA@{NQqb
z+}r8c$6AD(=735CUqEhbf;y$H+ZQs;w@N8cW?f1@mn<NeI{;RX!NbDX?A4g0lc#Nz
z^2Z6~MXOssJ0W*2uz$G|VS_wL4`&UU^*|q|wv57-j?MKpBODJ`t`e3t1dFaoUz;6y
zkMa}z1HadPr+A|<U`EWdF-@}#Ou}Nj4{wgR7Hk_Y+S4RF#HSXiKasDEgakXp_s=#@
z1m_Mot?rx-vU>Y%f-t$lfBsRnsC3smSrwT@RjT3-2Jn)v8C^3O^s2%=Di9ZuZz($4
zm11Z;_Zm#mu3)3sW5ywfl9Zxui=KeR_Gk$zp#{J6Dq=8194Bg^x!ksD^9=`Lw4?<C
zr$(9H?EL$_dlc$(3pW}(NA0)l9+;L4tXkjnq6#z0tM?mujT3gF_@Z5PIKMsGE?+Ij
z74}odR~r@D3r@1K0!<T8A#!z-T+ZOzj2FSX`=NE=>NAq<yq;};(ML3uj;KO|4?eJ0
zL8VBXTp@p%d7a-?)GFAS%|m91aLf?|HL3xi)!)Cr+-j%KFi%e19~@_IdGt-Rkh>lY
zac7nn=&>;9plOwU6Al^z7!E~{$=<3C(#`gX-Yu$Sc3DPokzbg6Rs^M5yD?i8i|BMn
z7G{ji)%M0+0izFt2*3bfANdp)!5=F0K`0m1ULfu!YYfU^-DK!?wjq2=$6?uqp${1)
zVadic%qi4ky>Afx=idD<lwyV}qQjf~a<Y&uG<sj424Oc&2CB!x>hp4JUOui_tqLM1
zRcXCHw{9Fe7m6J@U8jk~4%47g^9w3N3vns;{#uJ@{SM<G_A@5^N;`;idXm+5rg1I)
zV?kjDmwH&<R5LP+qUR!}Se8Q~1V-9^L?6T5?l@zlsEyy~DTz%(N?`;$0eI)9If8M+
zh56Uj>YNe`;uY_WGyzAzFBt3>>m;9Ou(GbFR1!FzdL#JXIKZ~Z+H(1iwZHJ_C4NmL
z)R=1@a;5+I!E~gz7b~>arZ94mnqtVWf_mL3hfC}2`{QE1YL(X2_bqn*f!-l)Y{fNT
z=RkkYyPd1#j`gXYL$kP{unA##?Ck;Fb)&h9>um<Q8daco^v}l2Q|HaXe;Kj5n7SO9
z7M3A;nG~$WY7<~F_GdQ`pDB#~c_1L|$>RHokhA&&e%3+|yG>{r!ovrecM?7Xxpy1>
zd2GSEficj*QNS+x((|tc&?Gj2xihRlF9pT`^h0e8y%#n0JQ-1ys9CfJ0#(}cfJm>$
zoMvdus#moL&=!d75)~XOYy74(%(G$R3gO=E`$<SsTZQf<=YWH7bjjI=+!NVN8%_%h
zhdJG1WRy`!utdm)Sh;=fGnzKqrF_}E54e|O%NY+sB-J2k8&>R?+k)#Fyv1kD$h{Vs
zABio|W#QuvYA2~ZCH!8$n#>_m3H7X+q(_w4VoAKy%|(#j+R=dkAP{pP!rT9NFl{8O
zoeGE^93%A;KXoAs=vP#eqn}o8CBt5f?S7Ed<XI`nR(AeM^|boPwPVj2*?@?%9ixT)
z9Xx;)M0cHVKql75`y3c9dy*QL$k_+luN*|(xtox~M2Sqqr{hlgJYOlsg-~6gzJQ%@
z8#PD8NmPJci`{rp<uyJZc6xO>!+W;abNKqwIeg~k6+AQLt@74IyQ87J{13_zpqv(A
zR9~J0cIEvW4Y8La`X6%jaY9qPk`f9G5usEQBZXs@5U7%DGGjd>)p+NouR8Ac!PN+}
ztltFDJ3wR&r*dA)b8ya@EojU4Fo7+839B++%9$GbJ{m6j`dbHHmt`}I&J=k&I@zup
z`udk&R(3IJ2wp1r6E-pyx_ge8fjARF5^hk@s-2-hG8m2yJ!c;C$D2VWhYVAB>^v;R
zXDe))A9Up;lDf;aDB;TQWO>N3N>OA;G7F?eqh?&<*vTzzvuKpnF=bI{NI|2Rv#LYD
zYwdKcLowOX#y%=ag63~HqU=r`MP4;yJ$QHj(X+gL3n~J03643S#no1Ja~*QobPZ;Q
zo);S&m!bq;7AE+<&h=QFoeCO?c8}V$4nJ%sm$Kz)a^Y#v>aBvp^8yTvCS6qWn<V{T
ztSh;t>Og79K1S1l|74jF`A26`FE%oI_lZ5%)r_kXs|pQke0^`t+bNzi4qipP$=9s?
zYNvqI)YF`>Oz0MEQpFKY<4o2>i!E_c!>_5yds|KSs|A|5E=f`8-=5+lza2!v_5msF
zkkgEQC@XDP=W^&FZPs)kKm=wG{P5r28zpom<8C83t80`>p+pbnxr0mWFd7ildvFBv
zI6~NucjbbrzRwUAq#N%g{m>~DW^emRYEVi60T*o(L;L~1oTImJNfl#N?Af?Rr_r)k
zjL}$>0YXadZeAamX5N^n_b$o(I*q-X>zL^>afvV37&vjO6;{d%-fO>X_~f-vh-~+E
z`5&v6=Ej&dIS<l{FWOi&@YkrQ_yk_YES9qQKK|uM5Pb-1cEbGmg=Ep!<sU1sBR3BY
zrYOdR!3g96Iw0>hlqn95RTMQ9MornlF*rrn$k`leSMCWxTAuQ&JbJLv`$^y7aKIE(
zqL~a-kAL$qfs$}xw#8I-J@`fvOV-phGub`Gs&(-G`&U68DJz?L)OTga9;62+n_Loy
zC>eI?<th3!I^}bDPXypw);c|7f%q0+olwg5&s`aOKo}nsc=r&-r=rJ<KcJ1n=g6QT
zBi=uz>iO85Ej#9CoIddM#R?*$Isf}BUCoC`^X`i;o7;dw2xQ`c8m%K9u<eYD+EU@$
z81QJ>QQFSVQI=WwjE{u2HsIR^Hdg={gXZc%$mJF+pPgJZgP7Bl_DRSIt%kQ$|7&V*
zf&rHFo`S@tb41{i0y0+RyijlzW7enk2Zn`{n|0$nW%LjBW6n{=HB*YpG>t}z5d2?z
z9~+&BzPKe@C}h$A+5LJU8HLD?T|A{LLf%tz#|B=O2Pu888$J)0G^KRb%mhd8tIV98
zol}-P9sAoqR>s2FlL3*wUaxWq@66GS)Ov@2Qc5sp&Wv6mM~P+Ar3ZR3sXXyoX7+7A
zM$!Po&#3pDbfB|9l_JcyW_M>oF1jsdNSKs4cPZMa#n|kKCos#TNi4<BQ33U_ALZtf
zxHGG5f27%$UnxdJ@)8~!nb&%M@eu80Jjmez$v+{SHksUZclk{PKX%`}bWf?iR-Sw`
zG^ypYK4`bvsaXsDt=-i&3BrAHv~IpNGwKlc@J?zEVkxy9+5&u?<m8;hUq-I7!IMOP
z=l=ZW0xN6A$73*z(I{)qHqxZ0f5b<dIx}E?20y#-S#bdoBBczKVLj`z^2>CQ78CZm
zhEM2@dYdf-eY(7$GOz&5SH>U@i^$X|dIrvxdpsRxxZyML$28A@T}%cQi$S$h+BERr
z_}#N$r~@Om?b0d*b!J~hmZJEg2Qrt2*c{P{4zSaWq4rqBE+*>J^H}{}CGMTpJe$uY
zMZAqqBwI#5_yfb9+xC>sV6|L^L4L6*bsV8C8O(V(aAvw8w7w~cD!<sc?^5XZ7euS@
zya{crX-*FKxpl!&W71{S%}j`af$1~~Rmc8;5Q?%!agx&GInwm`uONlUjscsfSy!jU
zzk5V)=QXJzl+p$w>jw8j74sY#3|^m#E(yU;UK@Uhcvy5;lICrWRDHEVf`#p!MT#Gj
z56DM1u#`(71bF3@HklC(i!eh&RxH<FrhMD``@}8IK$pOTfzp2)gagatmogxQbZe97
z65lYVYxql5elAsL>RC!Jw?Jz<DRF3~nIkknjHyE`|4w<V+A+F-&>8_~AO8SgQ#dpj
zH(tkdiBxa9qFdsA@?{@jE)?9vTs8a$#ayZeI#G)G2$lJK-4qN%#i%Nso4|E2PNKrD
zG_4|+p~QFgboP#7M*(V>3*RXl71<7AxP3DdlWYARHRDe===i%;nHtEaBJve~yX!sb
zRHpWg+j_j$hNv+U;2v=+&$BBb33dKUjJ%_8^KcoVR7#-EZ^&#u$e8-Sas@!|LdJJ#
zt_Bert`?aIV7TKU1vc#MyNkgmi_sQRX%??CAByhj9iSWQiwNM$Lzj?bEcs*JxRMLU
z%KFN(@{mF3Wc&GZb^ArZ_S4hTT32ACeWWb<WwtHm-n#Oy4e+dqV?&g3!Suh$>gO>h
z(|?r%=L#ebx<VU4Ph?}ziRA}|h6e=O%o(p-gOn16F==Gx9M8t4^s2J*%z3woD;V3u
z3Xt%%=dJ!o=-UR5hl&(?_sKDn)};y^!h@Xip~o&6EIsJpv0q+Z;-v545b%q;yKb-Y
z3MY36YnY*MCX8$RO+vNp$xQgeGj^4~H8M#_rK}RIIbhJX6`_=e`Vo9D&~tU=xMHPG
zkEQHyIOpWGRe)#%PtI}CA(?E`=_Ejc%0cX_w%nh`upd3I_*6<ZgN)mJ5(+pm{R`A-
z%186Jl765k9%=*BPr*H`N;%b#8m6#BfPEMX{xVatnjhnv3)D^%Vz_S)xPJw(eDCX$
zsd^6=Q%pj;Iu&t$B-TE@{a$k5^Rj{#ZC@2#n9<)a86kOgXJ1wPk?g2FpHVTQ-;D2{
z_^SG&S3~;9c|T&;t>LOyW0h!6$sE}AmkSr=lxEmICET}93e>3$$XoIdoe-c5s!u=e
zDyM#$Wc^bvsBx)SlfH8!lOpk}-tJVLCS2-zw(}NqT{vm%Y-Y*9bW?wOSTjx(zK1nd
zAp?kna*hkP?mN>ofnz|Xoj9S|19SEIsVleH({Z25wa+@N(RYwmq?O(E#HJ&yyqsR~
zOiW`D@4v^zb58G9%I&HBgzr#VgIo6EFNb}`$w3+qr1vl3wF@~Q0^n%3)UhbgW`VZC
zjOUZqO-u>PL3_-uTf!g|3$>C)dpQvtNa8i(AuwbnH}{tyw_<$<O+fV?K_bHk4-@%Y
z7u%o7R`+`ZllglT`5n@va6e<W%bAZ6{{AAZa^WMOMt40b*I6o#qRGT2mbS11N@`_X
z9k2}wkWIH8g*taucHk3GUdt(n%G<`J8%2j|blwIPx7(<xB%(_5T!Mz=MNA<kv&1_S
zrwB#9sC_f@c)M=VQP>v8FV;!eJDupmrd&nuo1yBYg@Mgtsd8~Y_A9BCu}hwiQnmY`
zZTW{>@*)Kg7C^v*nHGSY&?A+|4R;=mhnlRJUjGi~oj=n7u#ok?D>?q8WStY1o>~9T
z5Il1mEkciP$q&w8D|&|7nk_6}-X$?$U=NB}6|w9~P%DYkDAWsCauH*y-nAOmWDo!y
z_gp2G>Jtq3G&Q;1+en8<z?E&T3(MdC3Fw1s=xI?+vj(e&D8WT*=}jTjWmZ=Kb_(zJ
z^4QPZ{Y?tVq=tV7wX-tI%Nb%A6T9r=>VR!V9-lHMBi!>|iJ&NHVO(+cRK`SZL-piL
zj#bwwX3p;87kk1T8~ZPI-*Jd(G{UYv=NS8s_iW{pX#2m_{-;dAxBOV9zdmy#6k8$#
zMj@bh`2^fITZVBid%1~IVPAn!)j8&u8n+)@oGiWPuwR=+lRo>%@LES;y6lI|y~mmN
z^;~yVxD{AQdBlrG6L$R8A|3fZ6vIA(lfa6gdT;?|cWY~>Az&N;T*M(l4jWmwWlMFt
zsKMPcR)w3txmYm5C4sNt)dNlJpq(UT-h`DIa!HZxi^cKDYD=s8Y6tD{*~tTbNsI)P
z)Z=i|rlF5j{Be#vq2u3K&xSHG;5Acdd>Hf}U<1VZF@To2$H1t;tgARy|MOk{<NM^l
z&n24S;HvV@wb(dt3#&CL;EK!<yc<R0Pm&Wuj>Fs7X_#v?YMH4^(eV8TAqq{D>pT`O
zfLZ`3k2zm(2Z1z;<kLvSOzcnP)x)TVw<w#J+-5Aau1d)zVEH|)k(rDwmp|&=^h0_R
zRR-Oj9fqv@f`aBEd5)7OySp^NScqG^iR`)fu$1tIj>qcYW%D<%Newzm>8V1#)KMsq
z9sd=-<T3qI6Ps!c@rL{J(@M+zQF{sV;($7~P=%mNnw&hMFEAXTN48PGVTn+A3ICxS
zn5-hAZG<l;l*Iy~0pek;Cjy5hW+Hh{D(XL3{&S<&do++&HC2v>wA;!x#MLvmtg%v+
z9cFH5|0G5lJCZqcvZr+h$l@{LxhbU#V}#HHxm0j8b?S^9*hbW>x@wRpavF}Q?@eso
z2&GvgnFFreI|U!iL1Gw|7=t)UHXIWIa*YqwCwg3(N&W0*xm8dFDqT94&NQK<9ZR%}
zR5WZ2LAyh_cXrCgkpIzXsMIaub$ENsyhf?Psk=w|m}jaj@bYBG8@uDR;q<zsX{FJV
zUgvtRU4PMk81Si*85%ZcLt)LMme_Ov-b!YNBjqA&#$|w#A`wiKq#&1)ee$~{UH!*l
zYBO;9WHL88NA_bucuboZ)n_}HEx*Y==I4?A+l^;C6{0$~$;SD9LZ7HW{mkA(fOkqU
zu5&fXH`NuJi0sHPiJu*O4bX<9h=&IlBan#$dPG~|zt1RDTC&9xp(%n`c!(1*hA5%f
z;IW%1Nw(yJpMKInSudd|-*-Xf6)qcVad|0Rs*9LSouBL!(>yG(4>X+7BN6<BUy{{`
zBn=kJ3QXB^9&bZ1!L|xUivzv0<xX}5;EMnxSYpyy8?SbFVhCUKvZzr095<jGg2zXZ
zazm+7#jBazh(3<EZ?wAty^z~g_WZ%ODZ>Fz7mB%Ttc0YSP!Eq#@3VDl%;9XdNj={1
zmOhlS^q*U)z%~pE4}a2NT#k(|ULQkyY+AmA6$%2EIFu2<Qt>fLG@Uj@N)$EHs;rt>
z%j^=o0P^O1Q>Ct=ig6A!M;dhsiSSE=8LF3DHq9%wPPZ_*tdx4fn};7`065HJx{3VI
zDKy$mHAI(_Mi~Uc`mFt37@-+}!4Bf1;REB^1H!_B-jN;)3Gw^kxwwEt*C4OSX-`7f
z&6-T1`m~w7Eml!oI+wae^38sRXV$<D>g*e{{+qF*A!3bPxlEddgy4}20}~L%@l5za
zB<?K1_~a%6l&;oi7|&#)S{LoG{%`Mki_hIU8?Y#2L0o>b)hV3VL_9}Cysm-EZCN*v
zfk8th4?^Uz6d;--qu?Ok)7w#-RduLNw@+IJp0Zwz4%gRBi*3FV+brc7LtbtI0)k&w
z4L{SrU~zu0R-Byb+^i(XypeiruZFNLzMJs{YB&Hmp#PTUuJ%FGf=wa3<&GVxuV$K*
z-c<h$Y>nDK5OaNZv1@s9-7|YX`laWPqaz!zo;o_M`K6C~jHS@!vG-)B6fqply0^K6
z&?BGeo0o;KpBkd7s2IEXoFukv1`RmBOrE(MO_BlMaM#RP&+b9SuOM}lhS=qZGu%;R
z26NEC!P}&3k`Cf8Ht-3$BsdmAI);;MlB<pDRIdI-;*_CZ@?<lGOjMHIM-IrxxW<<|
z@r#iAh>p|WhT^J^`3NmHD&)8~I*vPd^#Z;$AlibC2)q~o{Mq;?Lr;%#$sd3!*638D
zG}@QO)<%;Z`^PSc3SQu^sIy}eIPsmq$0MQeexR58OOH2Pffg^GvJvQ`Y}#Pz5e>3<
zri{e_x%#0X%7V>diG*qN_GcJQy`p3-`2lu5XcsBi!i8j?c&r_LCEh*mqHm2+!>}Vr
zDbq4*K$C<mTyvfq@VB+;u@j<qx+AHjAwV+WLniG~zk0M6+md9id<drB1(vbmBG6Rn
zk#;Nk_l0uFgUNDCMf2hblTK>_VOJjX0I}G|QST_=JA(!0Q#vY5V3I@JdlpEzTAyY?
zfHtvMmhqcqOcYV2?NqT=49%(+Q-wOI)(9y`GIL3a@yCVjJGs~b)lDxK-cGnOih#8w
z!dnhXu=wj;psr0y=S{rUYxHv_W?*^|MItgvHkwdBCdK2KptY^1?D~S1vhq4tqfumY
zrks5C`D^zIVP^G>jqxf};xn7XU4JCGeA*sB^rhWl*x3r}HT{c(QHaa%f+=<OzPgDv
z9?uqUrkTh{H^qtpNFGvD%Fo4d89YcC$!Xz+4@?GrAG<;Opc*v|Jz<MTg$fcKJxe1u
z+jXz+bdn{lx3$*E<-2AJ1<CLzMbCw)7nuqT{6G$)`AB99;ixpYtfqE*uL6lnC1mI<
zdWUWH^rzdL&q=xUo+?p_16!Vv+Knc9?j3=2BE3?XTT?@kugU_KhofXN?f!}>fW{20
zsAg4A!y2WOwah&>;pNUDcf!3NFQ-HSRlO@HS7X$Ih4k8t7z@krq~nf%r1f7Mxup8y
z8nplth>Yf{b6F#{!8nWKC+_O+7vz1I12*ENHF=!CtI)&aSI!P~^(No82e8Q*L^vwH
zMydK}W6eHvb&`spt%wTqVN{sZmX80EZ&X`@`&0)dWx&SLWMiYx<aPt7<HW83Pqd6g
zrd1YgdRLEblL1`?Xef!jYKJlPWqg9dck=4@i3*4~9(nW(yG;0O#>`2*lx;0+a6KsZ
z2hi|Sf$UH%1{-FX7P&d~8kk3{yG4VQDPwubZoS=poO9vv2^lBAXgs39iVs!qGsWJ?
zR{iZi$7bHijFKaH6a`icsA#047o3_Zf+cUe$Y^qt+2Z5K_x7?Q7Pkdlpe!bwr4;*f
z39N-pI#x^odJtB>oWk~qfeINJ*~NJ;d_2Ynbq#f0?ghW^8ji#Qg|?oO+TN*y#581~
z;t?EJ@*w%ysQyp}WdmwUiAqkHLgAHWfb50s0tDw{3Xm7Y!>>0fJlD?wb-82*1yJ6=
z<HJP<$mW8FE2;*lq_h(wNc@mAtYI1>xoH4J1b7(N+I&f(Vag0dYZlqdT0I>?-9}&)
z0nygKE?EbIQ?u3X&7zSn{9VL(E2pvS0U}TTi2#~G%{~9=^?-kq{KH02y5Ndxkcpw;
zf9`X?-D1`K>mgFZZT3k@f$_lx?%2hthx<~|zmuU)YpsE?26v~fFSlff{V1-P-vLlj
z*I>#=Vi>qC3X@ajM~0O2KotT_6b(AGx3BT{1P~}y*HLV*Ud~nxerINgWMn-p^GOa1
zwJ1ic1QR9&QAzuN$tY326#PzAar5M6DLw}1CZZmXTx<1>G|u+xGX9u1<qH_60J9oK
z^hQWMa}@5DFPEoLy&<c&l!<=}!X_tqLaF$csLOrGwOXo=#-oqWt+X5j@Xo%G5hrDs
zm7)sYy1w}clz_1l7IwgW{qbqw;caNO?SbDrc1q&~Klwa*d3H_-xtelH={T3P1w#Gx
z@Q}$`*2V1XuI<ndqNdu_Jkjnvr0kh08RW+)?;|V6lIP1!AyolcHH^`+ZFw0ji1wtM
z(Osxv9C|YaWXy{n@qzT;4QhoT0K=NwN|hI&<z@+8a7h(79psqtQ`Xgy62l%Qgmvil
zy>m5itaFT<7vUFYkRrvAHr^ZHlQozhp})+*j8p|>i}9m42~Hft8K&&Q>+h~BlX7z;
z>v<J{O6=36tnEEKWt#Y)F8>s@?4r_=ynm?4->XrA-}w?bhn2hyZL|s<@cr72Oh=s{
z(nss%)Oqn}hO>Bfw^|hcCyeT`U=8sg`O`J8Bf&}yjkJ+_U8_a2p=o8F1b%eEJURaf
z_t9ASxm*_-zAq<g=W^<n>Bz=@RM!CjyhI{3>5#{V6){pr3->H+MVTWM$?<<b5(pzr
zvLssIzibChELuE7fr%0=6j#^-k0M(+D|zG=pJ!c<0f@3)3d8_pCRYc>PgUD+8wej=
za!GnOnT3nyGHDt%h`AF6a*J_-xW4{HSoVm@Ko6*3#j}{?u(Oh?AZliG)p-(s8<(yj
zk4_ttxKLqbZX^eHTryt_az-loA}$zvAOQLvRgY`bq_+&}ct_xWX;)}abp2?m3?P21
z<olq%kR)qE+U;!8n8IAj<o;w+sdPFSB@dVJ^m&ps|LE-2#Qk1Hqe9)T)4+<V8kK!o
zcmyQy@u1qSCo7%>I{r3YPVAH0#GkQ@<)2?I#*BG(*~h<)e0bP8QZ(+m|1Txely&Fh
zafOBSAR^uLcX!miyt&A8nx9>`(cyWZG);|ImXR#|3k^Z0IQ$OR7JIasis=}lu8FA%
ztItee*chh_5|Uw0U~EUvD-CU=Y<70`#pAsFSiOZ2zKwY(j$~rq%%8X-9MsUfm1G5M
zl;`8Jk7jP|y70PEV;`pXhYB1eD{)(iSF1*Ar98upSw`AX&FvP9dJoF~Xxq2pZn$5J
zx14f*<+!@~=rn}#q|rdZzg-P>v&DSR6D5Z&T}KdeOnNkkePGH*Otn2SZsZY2GQJgd
zF3Bk$SJL17zAh@d*n;^T^OV^|<~?O!6gmrtl$IXiGf{3_wzP$*!I2y1dw%{Fl3O3l
zKA0K5e9YBq&Bc|YdML0SF0U***KAYDp;8FnB`%E;N)Kfx#*sx)PtDhovL{Zn&l6%M
zKwAjsl}=|I@^k2<t8;PV**UmMm;27g=Yk%wp};0D6j^aZT%@5i<gYaJ^jka9oGgX`
zC?wo@sDPE}aS$c&B1?-_Exjw3V0%5tUd5FpS3l>lYQgK6Soh*=Wg5-?SCChJ)34RD
zLKU`i={D&QG#abR_D$t>X;t*k{t7q7hXlE~xO#fM8Xz*qqy$s9istq8Zod$kYTm0n
zDT@h7pE5b0nG)T8JXtmD@y!C%kdy=RWz&5mAB%UOzLm5Efg-g5Myu2ha#z{ywyn%M
zl}91FyUF!>bKGB5Wk8~pl*N}vaaG`G%%GJuMZsdFQhz|_0iv|3bL?1I4u6RPT#u(J
z+&8-5+Lf8+>5{NQ^OsAs20R2T+YAu|+d{zjx$dd7ML%FUO;r>z8B?(AY@Zl02!5lZ
zxAeX?gS^Pdf#XO$^C~*apzE>c#tJ?R-BGh{Xr}3lb6W|PFaBKfe2LNT4|~X+Vpabn
zaY2o5t2}eP39#Q3a=T(D(V_sfbs$2Wb9Ft6W8C8>UB}9ER73!t<eV~dTpenuC9pp%
zDW1#KW%Wm%#ld+46O^CqFsV!<;JP6$v=4Js?_(ziL_kSITe#iiq+})PQ8@)%W1ULx
zEU|YF*jt_^RSj9+Wp)C)pMQQbEfxEEyaw&IqzPM$Ke5HWJ|huu-1ms&A<N<3z*+J#
z{(zt884FMc=azcOkB#WYBBbhCt{uMIggRkws^IpSz$4f1Q_*e@7Q8bavz#LcS0*}a
zj&;3)I6y;I^L>Ql{xT>Cx(S<2mpa*vl@~d!I+AJ`1UZ)q>aWiprYYO=HZp&Wkxz*e
zB4H;xt7~Z)-mpj*wNb!^8RKZ&F8Jf`_?ew-Kn6kDVaSsloH0>_so<u+@FpHU9MI6Y
zZpXM6sL(cB#8R8fXt*?rGKjhEJgXiL3lT@F^YPWMmKf)cd<9P#rawP6dZh}R&&@r9
zoAWQTTr4!{V|3fLfoj&;9GL+#n_fFDY_l&z`kUAX(T#nCqWg~U>hqo%pYRoEk{0No
z^W|dFSRf5TqZxfk7}o+(J#*6*-^Vv^-@K8R64zR{QD@ovsBcbK3ZLp?(I(53)m&Vq
zP{1WY17N*v?PtNXqtG78RSXh;{`clD8p?1V3ajmB`K&s<f4Q_<ahtH`G@{pMmpIi{
zZMu-yUo<4rK$6*s&>xxvk@(!RxYC8HnMH-${uX{%RJwBlq^6N|gBHFe0Ldi;1Kb78
zhSNl7V+lpx3lh@=EZkh!s3?6Bhg{Z!$)7R78p#zI;Fw+_{Pd5xc|ChD)!_KCNJ<b;
zGeJf!<TKJ#Xdm=v0HUKkxO^P4`<(1#r0JchFUW<>R7}9NKvsRDsXMt$*;ke*Lmfx;
zHCaew;EfTAUpNbXM!giV;xTDO`_``AU`4z6ZvK1u-U)lzPR}Qdgr2*=_!O+RAET)V
zjw20kBcC+en=gyD===1X{b}LuzB4Km@X@Z{Lj{yD_t!BAi^Wi52Ph50<~Dv5FfCT9
zny}}RQq6K9``q=GsHzHuzXb!g6&Ih8)w9kWOM=Qhs8@uO@LeGqLi9=K*9$K><wdpB
z;`9}idcs^@SgF;{(vi;~Ra<NUp*@*2L_<eguIMpZvu8^W#%`w3QVq0M(Q2S@s>*|;
zD-kmQC<v~O{W})`KnZO_ZdWy?_@&?P*7S3WzgR=Q(d)C|Z?!bz#E0$YkOKPWFel1k
zlO!fg!;4ISuD6)AiDTA}eLwJ|D-xD|tVcj&X{1@H7&EJCXJxzRxmluKs}f5O(3YB)
zM+WJM9`KV9Rv`<qA?*&330>S+?xW?Hu9v`D3=BRY5i54WL`(SkHIhXG5+3S`GD}&;
zfL-GzJWp-4w2i0M>Mm0!P!~qFwZ!KGbeSjcAS5(7nJa$eDs_v}JkdIc5OjNIq?)AG
z#o4n|P`L02SE*VXex6;GTf=}^B;RMl2e5FM6-Pc!d5&Mk&DiYU;J}yKHtD;%9pZHC
zx{W?86CD-gJ04Ne8v28|NQd-*5`vN|fiFKJ5|FwLxCs9Oo;crAPxhns?s^`fO;wQO
zDl;=1TQ5zsr1XNX4um<7WQ|Z+b3QvS*WJ}?ayI}_wfK%EHs-+QT%>^PXH;>ZKkf%n
z!{g4WBRs01s;yhc{p_${zsPT{#{%P;CtlSyCBdf2i03TXiDORM*H*+=te3+=!QAq$
zqIP)g!fZ!=GB#~Sfm`o?5H;8~*ZmjtxJ3gi9%?Sj=t#bQa3}993DJym{wG+3{2R^j
zg632Hx11yz{^%Gr*O=ZJ{VDj9r(@w_CJ)7y=a%!c`F{^7bk$)<Ew(kcCb1L#o~v{i
zWF2BEeUMmSA8Mql?J4%q@|Y|9zUZja;mg7XS{DTzus9ju@Xh2PM*kK^2e{r!Qf&1;
zh@6g*ot8C`3!jGnd0DxUECa^fO0B<N5wOn?=Nh>SaP+JX8aGB9U+Ra0L)PTWK1rRC
zl?Ch_z?t;=j;6f22=`i*#?%k3X3<jtWcK4-lnajmQ$EuoY)*7_h?zm?8-4RyLRA}W
zR2ceLbpD?+gSzHLVNFhE2O_z%TxYu8p6Uf-dwGV0IC=3@8NK4D5cR-pS};ULSLf}?
zWmb6%HFRsg@jEkW#3(KeP-au`QIuvrI<5Z4%w+Ok%AF|Y3HBBL9Q=x#A<0YNoePBi
zjy!ueEMk&TPoDR{?*^hNm2or}Tz^aTxxjpW_3;!{WL6u^8F~Clo8mX=GGzxs5BCDH
zyl1XHC$vYqAy;#YhRpbt9xd|q(mrpg9#N5q?Y{}?p+z-$>&bn+jP@v}oS06!>?tG(
zy6-wW-vM|<JB3HbWXJAT8^W|mPr&<%79v9nLhQ$4oR}VJc@uZAe~7*Qdn-a-sIB9u
zwtbL?rd@Alf^}vdd#0(IvoUQ_-^z2i(vII=fVI-R7`rh>&ZECd>2jfo0jSNgz(o<|
zqb6(OP>Pgv67l(@l{tTy%%uQNClCEXqsZwjM_y#0MVeHyJme_OC@OtaI@TFTj^R*B
zgiiMt5|0zJJD5EKA1W8}UVW1lIV%R&XYw2OfO_ACMP5lq@AGB$jh(d7f*f5}x!~tK
zM(IC!4`5)U;^yZywhP7Lni3#|EA9+3`Y+i#<iTkPkdaG7_DRfN{{d1+f&-*EH&%F#
znSf`4r^!Cuc5^uKm`MZsGum5h%KgGqXS{1V3*_y@NZ{aCepq2uZrv@Ji{x00S(=}Z
zG)fl&zrA6!*i|7QYJ+BSY+eu-d?-tO$34-Vg1W`&nD#!)QqqjO_vX3_k#vF3wOn@=
z&Y&hOwRr0w#UKDVZfCirk{)#n=V0v+FIP~(Hl|+c1AL(9`QeSdaZivUXcVq2sHVb?
zs*;b50w<F_uptu@*>f`JnXRj&Z;`N{FaG$0;BBOE&(n$X!~U+Kad*Jzyw&*LUPL>y
zZ*AJ#g_yt4)=;jr5v)_4UuqCDO0wukB`l?|NAS|C&a@74ahZ~^ytY=`*0y~($})b$
zGHSOB*6e$abe&(OEnbMa#>ak5*M2MJFCdZ31s7hXE|1Eqg3uEdN1~~rXQQC33CWyQ
z!_6$f%*~&+vDk?oO_5+dj?w^L5iUcN%_Nd%`%h`j_RGTLd`(E5R|U`D+242!3?Ztm
zR2ulnVH(a%#@){udH<f%`CxoGDgmz#*2j1L5$ezV1CiC{gnlc{JL?@S0XKuNdo$Bt
zuV$07M6uz!hjvWgvX{>k9n2%zCfLT-u1bVLt=sk+6wEk{Bnmln^#6>ao$Q@X7<4~e
z+LND=AFLXO5xgC1hc@ZC``^5)KwU8(=JUHvK#CASeXhinN{O9L^$mS6B?!~OHk{gd
z5$s#SdXYn=o`CkA+=OH~P82AS?$Ae8^yPC695<vx2WV`K6yZhCsDBMK{K}iMxgSWR
z3fBJ7*(Ze{5M#%4HMRG|JtfUhP&I!q*IYGQRKy{pw)}@ol5eKVU)gTK#y}(WD=BG+
zkntGewZ?DxmWV6oByK!8g`s4v!W5J7Fg@jvABulZ;U?Pk;H|!6z!?uz@=!!3H9PV*
zC4)HT#?*++$cZA8TwU&G6xFMxko5_fq>Y+EZQb2_dxx1P^LVklF(BBis>dhD+Z<_8
z>jWVd`+&G_Pi5951@+7yjUtYR`B1fY^U7<p4q<JFd%semiB=|0oP0I?oYF>yB_?sh
zXuqcv)FTAKmV<9%h6j_aEtQS%&VBM#Ok?G%sIg3e@g}k8#*|Bhz$?9ujJ6^mbyklW
z5;5v4ft?#EK1~o#$u9p{f9V?zkE7aCwRx~mt&?3N$3SY?Au*oNI{Wm!m@QgF_3%w-
zBL=icu!&2-Z2S`^aUt%cBD0dV_PyBv(v)Ar*5h8p`U~|_|7W0^Jtkn7v5+0@Vjs;Z
zRh|G|G3P#?;~mls9q;d~q<=RP@kTJ)4q=l3?#qdmGas+O{)VIXszIG4WCGoOt>)fK
z?gCA%f2QcJ14!=}I0(fHJ>O(uTVC#~Kd%+c?Bha8qKM#mS6LWjGG{T@eomA7!u+}e
zuOiq)pW+x9%Iixii7OcZL&dzCyb(lWQcYQ1gCER$^V9VXV+aQ?9$VbhI~2N!gke5K
z_3jo$|9w@6oK(d?J?HcD!~4_M&jPg1ODN^}*u0C}qA$y(zhh(OY$NY-OB~z`<O@8<
zG#vD{*)g;G6M&q{x3z(;JBic;^_W7+AXQo%>xu#(y>#-&r4%DG)j(7ogC{zU@TbK;
zoKA<de)e-0@8SM+;a=bJdVJ~c9PHmkALw#OY*l4X&rVxAdI+wUzO;_|!J%lIROtjf
z&kS?TCzm))>!)spHzCUkKs}x2;EU%S7MZH6n=3OBnx_+moRsc6%i!s&!lsRxud>to
z_!e_)s_4hujFIV#jg;i+u^x63%}XVe?jHhPpVA+@68^bLo%VEAhOTcV70^Sf@du<P
zsqo-eH=p+??vGn~%$oGhkC#6NJzSZb@2HB%B-yN{d7o`u)NamvC~WLZ!7W@MM6bt)
zD^1&4x;)=w$$RN{)-$ZqGjeVVx()jfd`Vp`8ZZg%DA4Ct!wt*uSvq^TJ;B%+@MvkQ
zHDa4qkdmgCn?Yk^p3@nVP-jqF|ILigePr1`UvAaqL|;K{kOsBYv;gtFJsZwu$TG+n
zn?w);VHcKC=zBfMdUeQdbafCk%-2Kf^y-%#d@gzwV+ykIY7*ttIna~ADQ*PC<-|cZ
z6~Xe0IeRw-_JbEb+pn*=yv`fvyunv_K2MJ<mJrL{#4%*+ejq#5|3TEE&2Z9xnms$V
zeZ0jtpeS04LR~(wnjIZ?NCUq>j5IQS5%}B38~Y$2V14np%6mU!o{V+$4cW8quh&x;
zfvU-N`_G+qvYq{C^|uJbk_wuNs>iLBSn(!N)Is}1?bM|&@awPSIYLsA3h~yfB)LZ1
z4t7!f)XG@&aT3L1F24rFCCE+eg`HbPqD%i4YbPq8<b>m;ciSgI6c1ng3%jkcZE3hd
zM>{5rqVh^WG*awT?8j&PoevTF!ZZA171mmV*a`P@xz@E~5~9j#ItycpV3p{E+{*NS
z_rFvQ`i7k(F1oKGm703_;ynV>PqNy79nf6);cN(UFr7{*=(Sm=&Ehj3BCJ;7dgvNl
z_^ul}N*fucXY`?FX3~n>c;po|{D{L>mB6I}kfz5cWg{d0pB|i~3huxauMKxwmyc#q
zM&!s>THg$mqWrzGqbB|b073u0%;2b;C*j=okqX=xgUy7*6k&L`<DvMBakHYl+0wW`
zSqKz~23?9bolcrU8Sp8G^Lmsv*`p{AUMm}8j5&BcC6p>;kmoTYCaVcs6eJVpn;SE%
zZHt1ksoB4~qqc@+y&zc2m=PQk3}F~4ni`eBjU&s&f|N~p3tM26rf)l95<7wvM8+O~
zQ`^#1HBX;CV`w{e`yJb>EsNz+#3m}y^;yWFG*wZdB3S47V6mdHCH4IsC+B!WVo_=C
z`@j$#A!@$)_AB1Lg6n0&YP;r>7cYq~KIC{f^00g4;qg&I#iMLmDFnQc&0gf~GVxQ6
zI58l^V&_>G7OM@{H`gqh1x3@~hK_H)c+H_}sjQ|!qoOAsc683sr-WNJtkx@RA;_!q
zieuX{IG^(#d#p+5<nTjJ-$iz*W>3kON}P5CBQ38&L76;P1$=aMLuvzu{SIvu#t&4$
z^Q(ft{OJ?6<%m1%>DwI;sp@Qz<2^P*@YdQnq9{2}3UOWt`^*cAQ9KdT=-6CuS*(_n
zbu9+VAz-5;oQ~*gM^pxl%7tB?IMd`n;f8_J;l$mWw|w#W=iI${L-d}aD6+|HjV<e$
z*sfG!9EHHFF_3g_?IgS#8Ams(LY;Hm-ag6d+w%!gYstpdDq)mF@TDYKyvs$yX1!v)
zXymYy=dakJ=yRbOGFe|q%PQrx36R!V?E2ZPw(EON<H)HOxBiER2k!3f*c}e|l)h7l
z{eBsD{`w!cprwod|8KntTg-0sDJ8nD$9oZUl*WkHVTwH5zvHj}{fX^n#b&+c=K7kC
zKmC~IaK*9f=|@L*I5PHPOsj$yWg*N6OKCL2@yO7hc({ARo7dm4eg1^a?G;x~Z@7B;
zl%{OJMeq)d5kqZcqmD61VU3tx69aziW<iCq1;ml)6|PtGt)(iIL`UfaMVS$rvx)G8
zh@_MVVI)Q|Pad+(kMQY^wCxF{V7XkdTCLe^wp?9Zi*1W3D07UGpegwkoshC*i77Jl
zBf&EtWQ5ZhsOyHNS+HK`$5avT*Q}NeRV6Ckq7V+=lxiV!c#18%Xw&D-Q3@>xL8GKh
zC8S7QRW$jzsuX#Ac;xlBulVNMZ+QFWEq&XIf13}OWT>l}RkLPMEwQ>_=thRqNQ`1A
z9Gr+ZWD^t$kkJYqA*I03www-oy5o-H{+?1NOx_bUaJ4Kb;($K|`qP1}iTJ@Y1V!fz
zRrW&DN{O0Wi*hw(vA~p?lJfi#WGL^*iJAT&(TOpmSqaoxNjgzZ<VDL}vHWkHxr)(y
zTuPL=i>8ypAe5fr;Vr>A#(v;R8Oke*Dg;akN$1Y52sXk6UN$lFqhc;<zFW*(B6HtY
zgwEr4GHTAKeoT=NJ);}xhmH_Nhyj(R8brEsV?QwTBYyN*upsDSVGMOyv0kp&=GN<W
zvt_edQ&%<CW}|xzg#6s5qRvc7DmhP_B$>-TQCLIp8kgxPAq4Th4Po|(b#9ba><^%t
z{9qruNnV^TKAvBOltiV2o{PH)$qOk7>Z>Ps)u52mhW3F<LHqg*Y4<?6v{Wil8bd6k
zs_@=rv_zmu6~L5+NiY$Vgif2%VvB-)7_fP=g0#QNs$vXLyn+w}BBUrDY?;UByw7y1
zNL9<pwjV}`4T=#LgBVz9&CvA}wxBLb_J;$T^%|9+EXDq$tQ*R*#Jdrd6m?lK_5%sT
zNq!ip%aZ3Ge887qe8K(017sEL;6_CCD@f2oyO+hKD6n<O3d@hSH!Kz-WmNq2Z{JZ7
zX-!}sdYm<k!E<*Scz1Z9eD#X!>l<!vZ+QCCpMrwjZpYiZcRcQPoQ5Gs*9c!nDJkwu
z<|qUgWe|plOC#UDdc|S)K+`n17+Gz$RLg}#U<FufO|1>edx`{w%EMYEeDpX!FnGtQ
z?|3|QIA!TW!1Z#<4ME0Kn<As~NG&79MUfKMh33{8Zi*6n8aSMKs(iZHTKMtPil4vS
zaN9U^e`FkboC`F|<!qqpozHew=i<(fVmh0mkXntk*&0L+2>3u80`309=sn9T#nKPd
z#ezi)#N#6-v?!xUDVuI5F^M%=;_pZaH}>rJvWdL<`Ws$<`!!>K%Jwle)>bH6Q<M#5
zRmxb_hH(rK4cD7nMn7`u4zhs79Q2;RkB;^z#+!@9f~X^I808e5IV2LDl=q)>QN$@V
zb7^9>fJiQ%A7fUyhd5I%#7x&o&S|!-5k;Ux^z2U^$5y=2`hJ+BvZiZ7^nbsw`_95M
zfA2qb{g-2QMIlx;&dFvSyq}9Flq8*m<cNM?Y+GD+VrWmqam1&DSu`v+YYHDI%YyD$
z;`^R3402$|bQML!8Y|IKqhri|T5Wq^cei5qcFW=UJ<DcAQ!Q8&4Hz*fom!<Sh0K+b
zzkP16O($?7F*wE{K@4b-Kr-|VRny}`!sLXJOG^>~@pE=<Mljlwv|snTeU4so7_G&B
zYq1dTzpBI<;RuOpu#kwh-%~`3S6@gW^N^AV$70A?oB^dYb=`1vbxl<&Vi;L2B>L>T
zNZjS&<??f9@3QmQjxmsQrU?)zjT};mk>lya>u<m1^{dysdwa)mf1n>6&Uv)btd=Xb
z+iSM#Emctyy$DRA7w(}>8XwQc3zft&M4Vg#K8$py6YXJ7cib~}M=FD$|3+hV!NMv+
zf28Xo-Eq%q)zBlpe7dk%C?hTOiOL`ct*ICfb5S>qmGgn}d~ekJeuC^^;-*bV-+9c-
zx-Hy0#jIM7aaOUDGEF6Ux^|Cu9`6Fxa*eH)g5ui>g+W?mQ|Ts$P2ij{OMWg>J2rjK
z%$}a_bey~R_w)@ZO~oDa8tlCj@7HkAD@apO95T1g&z#xG_NU06a?9n4^=i#}wPLwk
z(kvuCOlvuCou8A;L8Y|BRB9<+L29Hp1v})!haA!}ug{HA??mDLkAp)c$u<6Cum4RP
zJGkgyEAgATT$Z!TP!>@jDxJJ|OHn9@R8(4c0SH9>^lHP)+bzoiN+sML5Rz$f195Pu
zXwXpxQgnIZsthaJBuboo5)s4#l-4AZ@;K4b2R9y{gm4^<@J;5n>`X+1#d1lv-_!LS
zWmQ2|+L96mP4Yns&}?`psE81-V+a&gi8i9L9EX8%7)j*k3AkaTSu}j~(MP;_^9EB`
z*4r&9CBFFlOCIhHblpe_Vq;*erl<_-?UL1Y!A)K8KmX)oK3p&Pt6zW1Z{D_)+R_@$
z2}QI<ysX^F-K%f7fA^a8X2aF(4OiD!Jb(UYvH>4YynVRmakuAqIOVoUeyx#HTy$dK
zsI)=^Pl9N;7#Tw%?RT7p9zO)uO0zbarB>8R(HKKvG|6Z@i9sbE2;7}|j$`2Dz-hyn
zASpvM22!BHQKgX(#rC18EbC%P(T^;mr5X(;1sba;iF~lB`RTJIFRu&M#-rUH+yE|O
zbV<`R!e#VcUi-W_S!<<bX2^w|sE9N5LwF8oh>5}!R7FkD67gfLqSg?0_r%>>8vik-
zPy`x6%qL|HVz(kK^<n5Z93I&{-m~BB=-U&1Y>D2}wT6@|V^AatL657Ns#;LkhNP@4
zO3F~#hOv}<(J;z_6h_D6{XO3l6{o`qqYYi#^XBy%_WL8JZp4Kt_78eD^5*S3HmeP%
z{SlAHrGU?II^M?|eU(4wCGyE<&Wg(<<8d4rbNr4QT~?&0+?F_(aKB%E{g)SH|Fs^+
z6b9{7MrksVBKwdi??b>SL18Lmaec?b(DQhANBrGy`Ngk(#rDZ9t3}OrxnNwa=)0Eg
zaG*P!2yPVKT}I_ip)skL+eC-AcO38TxckcT=Cd!Unk7%4KIi4jkEyCA7x))N4SN1q
zXI|N)I93XL-Yg`(PLd6(x~8maR-2BZZYau<vMAo;lEx&e@L?D@91rYwd-jI|$HPH3
zbFI1BZlyKa)FP~jfyKJX4*NnNpM(}J8!S($4qB_ZZFn9#J%_^~D`iLOy5`A~Tin=?
z;t1)iDxBUfBg)_$4|eHq_}za`UYR5)t#oqQQ_I(1eZ_D7;n%!+^OpU7PkZX<GpDa<
z8b0{oLq2%^Qli$1ioP3(0a8?%ubYTDHDlUBjMAv&34UbgPaJoT9Ci<k-GS(O)~k}*
zS}JQ<REkO^x`%s$8yUM3ecEscUM#+<QsU*xYTn<7Nj&Hv+fTiJO>NuDHl>UkvSRoA
zv78G8y_;kGBIm+Tr}tXAz&Y<$iPE)98oHjb?HIZqTQ}HxNs$t&0j-5(lbu!OTH##1
z%_ZrK<)vv6=6lwU$pU5i(KDczzc&kk!Z{mu%xOAP5j#c6NzvzKn-Rs-CRNlq;&Qva
zV!Pe4*=$&^S2RtP6I8O&cQVk<g`JE|<~b@|=3GF>Am-Fw4E)E;B^5uzo^kA@v(ujN
zz9#;YA$If4RWq4z8aUHSary7*yVLxZYve4yj`_qey)@6C-mqR+QtC-nL5P9i28f>6
z4=5i<$&i#qDTC`oog<G4rW784IG`gA+*(MHlc}I845lyyr{r-{A{9B(Y#Ok+80?3B
z&h$GTk2EBTvJ{S<(o%5xKv|ZvJG{-(^Ekc0Q5G{F!<i^)EEuo}l9Q;Y)q0&dNI8-X
zIPUk1{Xo|a^g|>fEwFl_ICLkLN6Y$pjj1a>TGu@J-~EV}pS|L*KilzU9BIme``%Mk
z4L&$92IoD8$2)fW9bbQKXqHPh+YO(5@(CZkd@;uz-9J9^%`34p>DnG^#4<%$QFKmf
zQj)Z<y3yf?B&{h5NJ9`;a)m79!J~|pHn~a!6s>}zPYg*hOxTSU(TfTRqrsA>3@mDc
zMWI}T?#Pu<JgqD%162&H%E(90uX%afu(Tc<J6zw%n8_`G#d0Yo#YPi+pe)K+eJQ74
zS#8`Xo3F_eQ)y|bSH=>Ylh?GYDc2RYDH(iEsgCHe<Ly_!VfBkop(1D5YT;-hM0%4s
zF-iprD_(Y1D<GhVStTWinM#V70>K+G3Dt?hSc;-SX)S@{-eFWhWlKgqWP6do+c$R{
z_Q!dX4L))>o;VDFG5bnRt$}{@eEIdabf=EOT1G$O5H4KsVUD<paWZX|nH*<CL>A>~
z+)jAWnZNUUOUNHyT>d*>_xJZm5=p@4b<`*&rlLhjX{=DXqsMn0KBiox7%E-y_y6|a
zQq(out1F&9dq%TZP!)#iYRz)hFbtivgZiGYzxtA<u4yWH+t!ts%X&}fN77-(o=4vP
z=4<|){wp?5p7QkBGqG2=y5j2kin5SYix?xL8(D2Pb8!RN@Hxa;k>_I|4LyT%JRVw#
zs-etTzuFr4`%LzN3xVE^bZtl1o*0Lo<V7rFjiS&>#1Un|vaTqbhORvkyl_S9MpU>}
zQQ$+!<5h|r7ekJ!YcmaFr0ZJthr_H$Ou<uC1y#8qc*}lwKdWXZqWKhEHx<U0sNJP?
z&SaP@xyF{lNRIYy|G?vJ&)au*{N^{m;dj6LjD8qNF;NyZo4Vv`bH&ZgEiYcYWV^W`
z`bhAR(Pe=NlIuGTBgqTWl+@I-O7#7St~+tuKXQ0@iyK>#AJH-5yPj&h;C8d*`Spg-
z9eLb8K=d@F<?!aMWUyNUO~rPzqOgjP1QbZYqf{e(mJl!|11R}#S4PsIMA0m6{0T|e
zn7Okmp;VmbD=E!1kE9gZOdNViO(={OA%s$pB7XFQD4bbRq#8ZmC%X2?U6e@8&GU~~
zJ^he!b)B6e#7Zm^vzUYu3eRQrAm;8F?Mea)P5Ax$WR8m@q4}%Q?j@ypuMVTbjU!<e
z?fLnz6`XvID8ExHK~a~B)tm?X{OL2E+}yHSt!bJCby>|ZjfpgO5GN?hkkBdSZU=ZD
zB*mncn6iH8IUe^M+assk+1TwKvL#R}Ikx*lHnslauOBR)n1wA$W%1A1ljgf^mAMk1
zKQBdHyeP(^H11if7JU5i3#^qGlb|y^=^e&<^yqWMlCYAKfjD}y(Y?f~q{OH-rVQAs
z%%oz4*NMXB4LqxOl}hNOQASZN7r76f4<=gWjWb9~wJaFg4wQ%=@X|u?A>f^(j1@*p
zL1~SpFqYoQsX)ic{8CQVLYCAx0$tZ(m8PmIymK^7Lsa5?j)KjqA-RALiREU=`s#-3
zC)X^N1<oH)CXiHOf829=d_z%M{_;h`lU2cg{A$NnCy$MW<LH<?kBYj&1y4ZCl}}yA
z{lh(<eeo67+bz#O_<#>yJm>26nvZ|{F^A)k`}+sFuBB@`YM@FAHwr4FQ0Ssy*N?Q$
z3!<n*R^KQ^T@(~V3KBL80bGdqn7A7w2bb_Ew-iVeQPjJoQB;M-)N6bI*Y`x{Xd--6
zDt>fT^Q1~Fjpy0A;n}vv8jnr~Ocd^=kCC7XhA1Z0tHrquYqVjxSPF^yVyQ40^@y2s
zVXdTQoSZ{X5lt>+j)X%cG5@qGakH#xV<7YoEQ%Z6>{{x@S|I)0(uh&oI(1$0?Aa}U
z{%5~rQ5$~uo0=~_|1F38o{$14Df$XsVT%P-y=1xAvRGbI77eB-==u}-B<pKc7Wn{j
z$i<yxjkax<TP2#&1$@$kyyWC#KPe?n{Yd}njkL(}8{}5Od##XZekZ;&@xQZHy#M*j
zn6U4?{;^K%ANU>r6xWoqnbKK*r*eyHOe2_p$$MInGlz9X@NAN~VS0yp_rUJd(l#|^
zQ&X*%G@BK+E-6bxSsIETeJa)@F34-v_t>Q5K#>fnh{-v{M7X=V=i%{DEO|1T_2k(z
zR_is*Vu39SQZ8VWQdlM4kvdaH&c@Kf_t8;7ls5f1jI(E)np%DN@1D4)R%@Kr4C6@G
zp4dM;a(90xe537_?bQ`Gx6c6yS>GQXX8%I*@e)KN#E_Mhf&@bzQAtLgwL(_c21ZxY
zw~mkt%H?u3qkbt}TAEBMSd)tyIbKX0Laj|M1}n-!4pS*azWMfBzW(-G-n@Ow>sPO7
zdoh76tD39pTb@6A!FF>+T{jq8&~`1tZ<TdJfUvAWRSIRK{hcPGc;`4B_q3-2`~3rc
zXt7%ISYz<?T|%uA%cA1R_J(gC9_V+D(kd2B&9Hx9wW`piye4+?EDRAs#3NiVm0Rg$
zS-v;NX0T#HDx<VFRhG<NdPWixDWh`apUIxlkokKd1VMHu0Y)yUUACmrBuWfEB{8$U
zf6ZaBqO5C{n>F=n!#E+P7Zi(`2Z)-9T#!&nzxRFLFCaJ<b?;|TlZ(4cq&Cf6n#mSu
z8snD{g;RT9Da~fHmhqj(`D(SwChY}Q=^Rrz9cnW@43L|gIj|Ycb)#b#hgrPQb|-eb
zN80u{i#nvZ+i^JTb7D}>Q1w&a_K&)LP+=FzzR-Emm}R{wqEbL-or}s@b4kT~(<gjD
zVWT8KXeD`+K}FCb%Vou8RiM%+uF?G{r`ORDoF_W5ED14yju6DPS}DWSk54fYMkj4j
zC8XS>MTsN6MG9p!AqApf%GknStf*)wZbCxZaa!rDkSr*SVQ706b%X0i3L}JH=OnJo
zYC{}7YE_eLLFfm<=%f%NN)m(8#4!<Fz=eU}9Yt1O8f$U{Rgl6f1b+I<pK>~O^kV>1
zN&HJ;F$I*V!MUE^^*J$N2{F*NJ$`rK$5#zYtNAz<JO;}bN5|X4fs~9a^hJR-Iv3A@
zFb+Jte#6_>Z}<=YsbGDx;>q)8Jbm(nPd@pW+LUN7r@nCz#9<s9K6^H8pL^PV$S9A;
zrT|0BNMO)e!3rcaV@S9F1&&e;K$O6At1%&BL!gX647@!l4x?w8JWqh1FBSj%#gae!
z;D#3)f%uc_2<}9TBO#0oZV*IO8B&tdR8=*sss)v)h(2PhxQE9e8Hh0k5j#M>FA0LQ
zdI~M-WEVU!j@VMO+AJ}R#ZnMkPp5ifQL-ou{p)YIy1wG{=&|-ALa-Q9p^`zTgi3<e
zU0rP{?T^{4mQ-cM<G$z3eTVZwHe{_Z7E}SIv^2{F>+PDVYBKd864eq@Bz&--9U%?q
zK|JkJ5-w4Q5*V+HMn#2Axjl0(T@YR5d)*9KsQ$5n>L2@>I)Bf17u)}#uXI-SPnURT
zCP+ax9jz#g77-B4#hs0bCP5hkR*c3-<B0D&7<-f-QDs5Qf}N}FHrp!5d>VW<FHPbY
zpc43m`12<iy~72eLl|orM@Kh|EY~Y3?5c{@a!FNHlvTyDksR34nrwn90<#pS__aw*
z80U5rWL}s)D=(+ECrM0%t<j`laM3aJJqM*`tC0|sbR-ZI&3T{gIb@BF>6|D6A^@2@
z#FR=Dg>afeevZb-Sk_w1vL#n{>Wav#qm+my#ANxb#txJ4%a|GAvsU;}qZ@hm_AOt2
z{SDu~`j-30M^0@=wOG*94U6T9>+2_6-9Diz<o8a3GCfm^U@kz&vnY*d<%vp=p#A=l
z(_zohcLdj=Re%_XQBbaBRk3Ool%~Y>4%azxmXC_qM{qe6LdoASA^$0yi;%fHG__3A
zBw{?LlAH^b%&(goo#|XQN&8Y(YDJvQm4!MGX3Gwy0#T{l+8l(p<{hPyL$r<JtdK^B
zJMD0%9lkxFHa%8Z+53~kb4{FG$n=*u%{p?ynsUqayDfmcTFsL+7mSK5T$8aq9VDi<
zryGUC7J|<niD7Qlx=}hZg|(DLNmDP_Y_8a>H>{T{7EME4<->u=et+-fj_2`^$FtPO
zQ|_#zYg<lT$MJZgZCl#DWxwCibuB|L+~oFj$Ztp5(~+T*7VSTs-H8fVrKyehxlD?r
z6s29NVs_I}lg35bODe_8xYHC-Q-wk*Tcc6NW6Z#ZAKXye5gowwBRT|W;D(rq-gyFR
zF-T5HPCa2Fcq3>?N}A{szKa+waVpvtgwYd?LK#IsV!}*eDM~96WUUD?Tnu<IS#6W}
z?$t%f{o|grxk902p+}*#!Mi|R)CA{I%0f}1v>}WRKMp7zQMSl!I8o-7)(A%{d-RRY
zQPvfeEx}3zSwK;)8|=&h=q1;&D5z}3=rn536s4!vR@Ata<>tz;+Z}jWC1M(Bm1bEq
ztd-;IV^6PO%z^Nw)|6JGEOa`GfFT9i#~r`B-|@Tj8%kqYRt?K~i7GASqGq+(u-<I3
zvg`4*S>r|zQQ<<s))s|FDMwXmhW>=>M^aRnpjbu4kA|K@-*Ol_`s`!X`#|bEDhH8{
zN>eLXk@&@p<$wOP;g`2-t_w|(JVO_=?MIgt(hBc$TU{%RPOR#J&9b64B^HB=p1Kl;
zaF-lW#aXQ=X%Sk)6G;}pswlBqF?2^#as&eXaHNU_g<qhlVPOeq{LoW8+>yThlBQl!
zn30$&QYs+?O!9z3rzl=r+bf2_VTuj=&hYjCqgNOmvBqIdzz!o_@OU&%K~sy}$T0Tw
z!-)_b=qQ2lg~CKlQUz`dd2Z<3eqoN>nHTBH`3?UWLy_sb{)h$Uvi<tUEJ*)@)O}&v
zyDmOA-ItUSCPu6h)ZZ9`$wi%3P>!Beh%C`Ga<ZgzUUEQ>d%DMa%yNM$D@<LpUMwk^
zMkoPANl2Qp?Qvcx8AK`Qd`h#5cuwMuiqo;B9Y(Yj8wHf&=H>~T&4#P1E1o`m$|@;h
ztx=|+G6f{?$>+5)hrMUsY*bVGGR*~mDCC0_6G>Ej*3_iBdapnm`i_3+c-ZX(@zW}2
z&{}L^rTtlmJ&T}%N@x&5;$q5WwuR0ca9|ugN(X6mRn@FqzZ7D?WVW3-W?@<@1I~r~
zJ54rztpLS1IvyS#_z(Z#Kl1L~J0ACYhH=0c%hTr{@Z|OhtM!&<u|#W2-}a=@QJ9jZ
zta4!oF}Gc3s+vl~*kN@fCdW7&xO?|C-RVS1fx0ZPN-=a}PPKvOPqy4#U162s-K%$W
zhXIp7`$YTLQ%{Q1oW2o)2uxHub1Gwu^LUM!9~P6)og~xN&fbidZTjEmNQf!5GGb9f
z=ABQ;AQNYB5&*>*8M9j24FkcC5?z+WQmjadGJzUS_~VXYcTcmup;&AoX;Kt!yUQlN
zkmS%n$~^gmu|<yaQr}r4-k+!{|2ZLwHI;Noz=uT0f{Ae)a#ndOhY;^@<3NmLZn9_S
z1hK74@v2*|wmi9c!e+f@wOq1j8k)M!zav8dNe+7|x2De~Z+Q;Ih|8j~{a(zuce}j^
z%*I}XWy9p(c;c|%aXKF8+9Q46O2lNCc&dNg^}`~?RHC#{7{moT{{gvsadF?%cix|G
zK1$(ZL@7bU5|O$p`1GTflt$zF4*C(B(%Ee}#5|DCr(<;H6y!FD3R!*Y1FDd=my2E`
z!{}M&Ekh!R({pxQkH8c%O-gd2)=FZ@ih{B4Syna9jU4uSR;v|#-_H{z#we#LYjaCL
z#1d{CF}<O%Vm%OJ%>I;e8foepH;l7VGKI((M^qG(WHEn~38dVJ?|a%|q*h}2qO_%|
z7F2*W!ml!>N1KACsYxk`$JF|oA00;i_N&CzUE;ToBX>@cEm{w$ZiItV7Zl3SB!v}n
z1`ap{&)xCBH;26-l^I1S^J2SPtynEql(u5ISfF)*b{0h>#esv0#N=@y(2tS9dq(Fu
z9Uo|qEnyr{-plJejg-Qov|-^oDk8snx#Iu)ix2q4b;%|=V&8FUPqeMW`;izW+R0jr
z))PV7P?iPj^_tacjlvK^#28CeRcG-JX>L_3CUi-BjmjJyR1^iLgC_<rr?skp9Maq@
z7O1kM8+r(V{o8kpMMe1dW5U>jS`bq}MM-$jDyyxOVhoP`spZwXcYO8jD^7zW04fBu
z%e7di*q>VN9}m>^YpPPrlVcK7S*;UF2S(R23<JTZEO?0L1c8`yo7JUH`=74wUW5Do
zu5bndWW4C}KBJ7z`H@g*MWa2-q*xh)Dh*B2jLFc)NT0y>VtVS19lv??mU_LWS*~bS
zD_%T#MogK9GdgyMBd4~-#7Kdn_ao6oQI)5}(2wZSX7#oKYxwfZucYl*Sg{SM7Oa*_
zw%aWqe)u6DJblJ$wW4WOf{q}W$y1i=<c&6=Tty+g<%xJc5z8miJe<AEwAR?d$l*aN
zi~(DKwV+Kx>4Y{BoxQuN$}s1W=0YK*gby0;C*`GZqx{^mHM1gg5-CieP0Id|&Ph?1
zB3Z?_!CK3yJ@Oy_*T3QO&p+eg@g9%}!4E(Bn8k9%a=oIi7AS2<C`Ko|z@}~}jFnm<
z1iZ^Us+289nD)YecYE$0-*G(dIUEisWmzr@RI+$Kay%aKPTK1$qbN;DbdkPubdLwx
zea9m2JC!ccNeNKl&gx>xCeDby)`DOnitPz?o!YE(Ab?gVuPCgzMx@*-{k}_0C|;B`
zu`o8P;Nx82g%EPF;d1w;C%KWuqQ=Kwgpo;8MosJv47<0Ss*=Ta%c8DHO7`lgCChwz
z&IS=<CWLmCf){6jAFyUg=k5Eye^T?hLAay+&@-*|-urxbu~b#Vav?3)tL;{by2XN~
zY33|!#re=W84iaOA&GN90$dsy1~KFokH!Zc_mAv%k|*rE6nujl8QnnNcl2G$ICP9d
z&p31t!+f~<W3L}n*jXJYZKBAb$%@=g{hs#4_gx8y3OtFbFrX4CI<_0h#T&a0J&u@|
zvH~Ov88UKmW;Upq6F??{&P9QWC@Tvtc~Gam2V+P^5mY2uP(~IG6#_wlR(6*BYo$@f
zg7%bU$>Dgw8%ZOGF}_FGy$ma#)MR-Fm?xFik{#f@%Mmt;K)@CSV?WG`=@hc90bMD)
zm*-NHHKwew^+NJiZ7~}u=FSk6t!T=IWxb&5T6Vifj;9l0Of(Gq#gmFpZhp+)y?)^7
zH}Cjj=XibaJoX8#N@9wXd&#K{O5sex8%SDF8mLTxLemEiR?t?j664r0JoLPM`%3;+
zr7&1@sxTU9nJ39H3xHD+Jrh;L;3$1!5upKzr$eI7Iu&Sx=b1|UpMJ9CU;p`!d9f&1
z3?2T|(;W|VW6zLdwF+%0tVA+dYiJsogg|cL$mpq>hQUi%{Peoa{1cT=Ul%7~kj=cv
zi(yJ)-P4aFRZ&WtE=BNOgbg9^u;1aU2h0Z_5XS*kdSY;pkOD9Bf?^7s`kt>}eanCP
z`@iGO<0C-{N&^>cwwOpm;5Zyn`(x%v<;7QLexjK&Lh~VOa=n<}P3E7ME!aO@-@T@W
z&&=7$?<dsG#Ypsdld}e&3|1@gLrjTEMd9n(NLeaS2AdLfOxTFf5PV{cj;5%j)pqQ0
zeWWu-k})V_(S_l9yWwiHp>JCbk2{pM6s69Y++rUUl@RH@^MpZAex(A&1yWQ*FB!Ur
z{gGGSzNWM#g)x++<>~D$+p9I}wY)8t%eh?}lW?boVK^hMNfH{AR#;`3(;aegtBk@&
zhj62iT;?FCpGr}<qfvNEK`T=5`M?dAz6vV)j~Xq;zEWUT7$c&ElnacQElZ|s+!$fE
zJ5Uscx@stjiuTm;`qev*$0LWsj$v@DRvT8U4Ur@pbybNV%vgpo2<V_S)|S+DBl9){
zQ9QaJAOlLcmL#Hc#NY^_=l<baj;BZZp(6%IS=OW^f~M{?FuH_E!il}!O4~d}2%{5U
zKNqteLy^1^+E5TKq!I~kMu;X8>HJ1QX&DET9g3RzX2COM#)e?AHVe4kTN$SDKH;=#
zUVEg~oTZT|%6=ciY@Q$7h#yCyA8}4NjsZnb8lx<(-P0zqMk?1E>h%U)EitAdMN42(
z?w;GkG4Js)PL?O%S+i!e?mO6elmo=Mvv5YA<pdBi*Ym7E%|tCGHTj}$*sNAuZMWQP
zZ`iK0`n#&B${MQ+(Hdmr5*24)z!){RgJVe9Nnl{;duhG4M^1+$!_d!)aY#Ad?kq~O
zT2UB9VKr?pCh7lVs#s$X16d*}%1fU=663+CsQE{(I4ww8s_|u0@Mk~&DNSjJrxBb(
zMMa8U;7ziM2N}UBn@y4u+Ds?5C`E+Mq)^0+Fvvf#s4Ub-_!O{Of}bZZo1}$*VAMIb
zV^YUX#>ZXX(=;{C`8kiUZQJZ&qt7njsBE@r(ZLHp(0f!-WIrHTG{wUyiZ56cZ@qCG
zDXWU6tRN(TpbLXh((;x7<49Q-l-ddklMQ<%q+REPguvC+mbxj~?{^#?-f?w%!w2gv
z>z{wZ&CM<U-@pAGFRn^-bH&%M9~s(_dZ-x2fqfh~;OL?(%tcaEQ3gw|6hln(L%>)~
zRTX$5!6tO6g_Nws&q+sMOqd|%l_|+98I9cE1}PesRmr-lD7EI)w;YENH6}jT82)d+
zc*_6$lNU5mGv0{>L3enh8(X50qC7<?wJ7Bliv<8hQLx!;=2IBJX0rif7>+Izu+4Yu
z0OYmNna|>piK~^G)rTQ?7ON$T%?g8-Hh$)YG)09Aky=>}yB$|yMEikcJwAFwuy9_8
zLP*^2c6|QL*ZlU2FGMN$ox*OuE-H+aIUNad?=;OBz^LC}1Ap%U{&fApYl>llKhTm*
z=gkhp*U#jIIAm^bAmUPDkqlN%zGG4Vk`g)wbV?M;<l-xvtrJPcKvAGlDc)t;P}Vhz
z8-sJgsT;j#4D#kL_(logB&7t&2V6g*q9QoQfSA}CB__vZX&Bm$$A<;WmBeyQq;L@o
z$s3#%rKtBae5Ve`ER4x}0&}q;H!;mB>_|#yFEl~e&mT|M#D|I@WJS5g8Y{f$6tnOm
zpj676=VX5(Xq9sT?*nZ+&~+!;R+0p~k62rt1ymWmngn%n7@z;TY?G1sY*7q-O`088
zF)0$<i0eDt*yF|%r{fOq22_d|qlHW7BI6hsoQUqMg0i%%*K0XI`i$_Tl$H4tf)vm(
zpeSYsiZdThim#NjUeEV$8k=gK^1`81nMdaFe!AxuKl0uvXGTHgH9K1tP01Z+PG8FZ
zmUe205%0XT^|eM@MH+j?ARe&IyIZcFJ;$Q3wG?bVtBRvKFOTyw`Cb8c!PU*B#`~pd
z#_95(KY80tNc)7aO-9aBdsZ<~1T1w?vRba#tT$}eVr#N!q^K*)vx~wsu7ztTMY9+b
zOGj&dUMX|?f(U>lf7Xff11Y1|8o_-HMKIVxKnltHwt0>v)O#DkKYsl%j-3&!kT^PN
z$!W4rmFzEq+NBv$n4IuITBvCvavCAH)`cPZjuZwS?^+t|F+QTS$)^LkIa3=o<)0O@
zWgrHDO(#E}<UK`GXNiBprHG24l*Nw_RK%2aUUY(Ug3?4a``}3d7;=B#psk^(D!Su|
zgre&@nx<jBUe9fcnd4y$J_c#wT7w_@d>ZqZDsw`D+!t-L{fbg@YBU9h!vQyp!bd49
zq6^|4uhh&XJo6-EC|FbEIV=Nf=sU`yKoLkOu-$BMZO8ro#KzV%*DHSU;WO4v!(aWU
ze_-Fg<In!tkLZTLn}<Db-@N5&NN6l>9OdMf5^nSiKH{{dOOaN?&N)1Z_0t>HR~vj7
zd3?C%_3lKk6t&eXYs-u66PC*0jz`kiqoboXhU;?6X0=3>rEu*|M*;z@;1^HU{Oe!+
zlwW<gB0aw2aCe~JkMyUWu5U>Oid8|`Xqu{GQA%s|_Vx*o*zNW?ibxiwx?ZqcuDIVl
z5Q0B*O>+A+#t3@i=IL2UHho-_5}JNGiVJ;rYEdYPjo3s?Hamr}tg42)-5u_<r`WVm
z3@9Z8Y@I!?e3T;n?*5VAe)c)<?jNP-&Ce+>3xBW<OW!lefA0V*f8fRYKNWTVxyxj}
zSxO0ial59wCtN%EM@$J%$ed!?u)Tp(f<`&k8fsfm>4HjYN^KF^xWc3?Ktne2pumJA
z{I?|NO!OlW*le~udHEryQ_JIF&#CJ;bv>uxQ7U0jL}duZpp3zXC>-5!lnn2z0@q1X
zloeH3QkIteZpY*9VMf3J7K;U|)rza_70*9-&UU+{s;ar8F!{z!h1sQ|FoZx^Tk-6=
zEROSLI(0GNY_&ns=kY!l2yqT-@B2a4%WM=Y-zrgkA;h_e^Inh`YfDb2V@BwXC^gOf
zHP_cSbBabvuxwVsA<P{XLFvql6h$BC`d*Mp6ee?$t=8z2NWtN|Bi&(7-<{~X1A{+d
ztfnk1DJf1}N7n|NlYO!RY1?Oth|0xQ%xY6pmXQDcF-8gkMptuM&)i;3Qbb1+C0q4^
zXLwe+rkM{YMa{WYJ2?Z$*S@ouM5I_rh>OL<*?W(ZclNZKA45t0>*$@-0Xcu2;P6Ao
zICNBRt~fsbfR&Ym9gTE|($s;=HN(7WT@-cibAr?Ft#{{Q>iu=@tjN#BUM|=}HdI#_
zs!CH>O;goeU0ri^b;H%wHJgoOffq$F^Xf$0lUuI2uoD$__LRJ6o22O=G6_)fb2E}N
zC#`&i&}md8MoW>o|D`W*uD}>67$ZRc$8}qNm@3v5^f6+Q7?cb57ABv%@83Ss-1AqN
zBd3D=yj2N9B*uZVDN#mCjFmB<bw;=HO#)O@bNf@;dtOo{w4?+iM5QU3*+dEwc5btU
zNXUz3q2+OW-xHO>6jg3%nH-+2!CI^>SS%NGZ7aEdd4UtPD;y37wAPemDO|Z2A-5*v
z1)lf_0j2YzXCx}k=t(&iD2!pXSa3R>*xf&Jb$v};)WVzSdmNESq9`QNNT)=lbCIW%
zL`4}5F=sw1LkfxQYRgHNv~ACDXjz+r<r+S@UGeN+|1<vf?|;MRpZ$Amv*Z__{)j*S
z=`F{15A=_FYHd-u#qFc#{&C0dG_oH8cP^s5;>n8-_`m)y|96(R*W5kc^Kbsm|A+C{
zf5+naEt}PnrK$LE`-H!I`6;Kbzv8oh|5q&hz!HTWBmLpXtM<e$dJaCZEDitrpMSvr
z<zM_IYa(~Q{Wbk*&(H_@Q=scd%DUv`ryo)-4X1uj)0Ax2Tl%ic+?|TUK^FFu63gY1
z#bU`gjvNn1Y18McuoiDL$%UK_@jg;EnV^<sDfzJ=_7~aDuJ3zB=TKVOz(#8pb&cJ@
z{@p8@+n4BtN2!3u&gftafxhc`czER1t2cDL&&8Vh{`zbB+;<DQ?|$!PAMw4F?@vYD
zA9E$?){wZcJK>B#Wnd?(-Ok_R6AVd;Ivt@1&<q{TIMQg#(w1zCnktEpqtd`QP*+ft
zrJT%5@w@afGLAj%zGrYFk8fVF|NM8<%N48bmfPEF%BCh2R`RUdp53WscRUgh9%We+
zC@nlYrDRPThJn64GPJ@Wj^y>s=$vSvI2~G!`xD=O{ffW;yT8vWR6(%oS_tK<)sn?>
z$$GV>X)ZQBt2mzenWUw&h(b>t5i@y(iltOpmiSDPlXcL|`JCQMXQXR;CO6v{6W&Q7
z6rvYCotLDF-~%Cg4tpuAthGG3eMU@@fZ)a;9g(7DwcJ30u^#~0G>tWsrj)E>Ab3yj
zdj{82l@*0C7$Q0ZLf_)XfwApqcMlAw6T{dsxSq}JnlX6>H_~+j$78@}bg3}Vlu#N=
z)09+Y!SU_^e>$R_r%GT8P3$G=F(l1uWoZ^QO)beDQ-Pb}nf{~lx}%iXp-kg=dJRkl
zAaoYkgpdu7^o5Y(_vXR5L3_!koQm+vz1%2qe9+2(u|%awDPoaMN^M|tBdH%~@4w~Y
zo6j&sLA_W~)GLD4#0mV1nL`rER#f`E=cV2|>wdQg6pF%3s7MofncX?mIa7Wd2io>T
zcRWeUy{Kraikq7oUcC5#C%4boZmw7^RuUODiGBbTGh#Zcgr$8Rb6#^up5Vm?bR5MT
zd>lHQ>j}X@63`{ZbI$!G)y!(~Ta*lQPugx|jR}GK!^1zieofa8D(tGOTe`;wRz{-g
z&(T8{RkAw6p6Yw5JjI!Cnu24ow4~^8V^49tq**lrt?Nixl{1%Z@<N&xh_V!ZlN$z1
zgoQ2Xk1Zj1L060>N-0bX67H+CqA0DH78@l<ofoxjaBeP=loqnNM))13C@9L3!^3`_
zIKWIQ4k1w2wdAhmpI3Tf1BwNUcOFw}j21L6q(oU1gtBC8I}(aT-C&evci8iI|G@Rl
zErnfx(b=qd<m3ii*P=s0MKMfPM$#|}iY(JEu%@6pby!n@_UMHX`Tb!>-|o5EZfKf@
zUw-<G#s(gD2jbngcr;J1uULKjV+6$1*Z|f=eB1Nwx3Bp6{*J35Q1|ewfB7%?m%sWI
zdcEd2bl9rq=EDzo{>d|*zIaKDiogChe?{~3nt$>0PqF(qy!%i8kun6LS9lj`G;}_(
z-8TFm|HV)F)sH{MzkSWCZ@%Jm*fEBHa}jL{ZeLvU{DWtdbwM{AQ7KZFC0Cm*do*o(
z0z|;F+wIwIx2)G2s;c7gVMo_>Y_G4*n@slYis`I4O5`z|WKlE5&O~yhC4E9i#Oh^~
zO$aqeIz}mxN>y3#`rR8|{LG=F13W=nQHU!5&Cm_py}RS??oQUAS?YgTz2y};b6>vy
z`U7A8&lQ1xx_<A<KKl8tpS|yA?kuvINkUDUQPpYtNckm<Vl&1N11*u0_6%j9#$Y3u
znAlbg3W-V#p`UXrwGz&0Q);{|>73)#9_YpacRFCI5?j|)>ow(a$+OLh=TC0HSRPL&
z+J2zx2M%qA8$CXZ1RueN>}H_R){wN7s3>(Vye@sdB--o3v40#nold-cdp7%>ItVIz
zf3CM1F*w$ON{E>0tjHAl%;Y~eC1nV*;j*vT#9ekd19>j&0vUmyc+KR_&D4sOb$Ie?
z>)jx&N^%jRg;o<>Kr2mCEg@5gQcCzC;D&B4019oXS6g(95ItcW7~3PRYYBZv=sKe7
zDe;trVNoq|WTm4U9m6<{raU$xtc#MX>noOvCEd`oKOOML78^XlC>D9outrf-mesnU
zEDMY^vlpQ_DP*xph(s0Vb|X4-gB7AW&6Jb5wVuHQQFMO)bs5hkX8-5!oC>3PDM`%t
zPtsHj$`oiK#!7x~oiwpDXcIBX;oCjqamRStQ`c*3WwBaI)K&~L8km`kVjCn=Z(iws
zZz)TYH(vg5@}$c@_+DPN))^7*ac-n6E2_F;xm<C5ea&`z#d@`7(JaqGiJ5MY)2g!l
zi}2o(&;XN|goh+nJ3dHq&oFdiikxXd)3{d(5yzy2)`^T|kqBt3X_|_0tmziQ6%Ri0
z4`2Kf6n4))`4R6%hwoYnB?wYVlR{}40P2i9obT*>X+XzBX*4PZnySDWX|6{t=EYb;
zG>T9NXD!Ag1IB7hAt+i|SGW}Mg`z+chQQtKfzn#a6j6o2HZ|p<$=QGjAH+wdG{wvn
z(26+)BgRPb9&M#4vqsdZm8;lZZRt-PesD8lIPvPbuFJvg(iWYh<y!06*fIKmE>bRh
z_3WEdlm)&Y<ix21i>5(q&BNm(4{zVHy4uoQZKz6v(pmZ)Bgs4Pk<-`_5U!pvHgo?c
z?;?XW*>WI_RHY>hJ<*Tczw1~mSJbNwi)8~!QKy2{_L_&W<JIf0F-^nmlV>#bidtC;
zAK7e6wy!L2y`tRp{OBhi^X%md>L<@Q3?rdxNXGK^_1AoM_>$GrryP=|t1A{Sp7X&^
zKV^LV84Evh-#HkeaPVi(ZunRK(?4Uou)O`^cf9`MGn5~wih^>{uxe@+tA^EPfmWVw
zI8fDARBTF!({<g5KhUWo_(*%|c>e4IntDO-k<;l!j8PP^)9B0`kr<;mya(yCDK&Y)
zr5Uv=$`WggoFSbANNcT#KG26AWlKz^A}pJR;q)36v*k@ChNUS)k`YeH>2%`V+qZ%;
zXvuodeW8mPGZl8yy655mb)H)K{cr#7^#iyW|Nnj&WiHPA>de)d2>Zaq`%&NBXc9vG
zND3u28e@>q@1O#tkpV3{tt4FII%z7cD7B(U($2L?)Q;j83g}4GiZKS#*b}3}1&7ko
z0xD9(=ELgEddYgRV03{~KQQ#7KzF0Z^#h~x;x*|Usw@b3qtZ(L;>n{;D=o!qzAq`I
z-17A2A*ck6QxpG7D~@d^>z)!-tT8qtf?8UQb8hAt4JUKPrkqpe-zipCDkl@9m<Zuq
zbja(ICYl#1MT|+D#TDt?K2`$7hur!!N@EMzv;`y=6wHO(3Ma{-&uF4|L^t4v4tF}@
zyN)>aFb>!hum}PhwPxS;^rOT1NDOMuT%E>;wIz#YiLsV_wz-I2@~LftU^JxxYc*wM
zC`yy3cbo^?gp|cIkuh`aF1lF>R4z8;a}fy`&~s72#7Uc}A2PNlmw=RCgUiBh`u2I1
z%0+p~mL)C(lF~WDR({TEsB*efvI_i&>rQlsJ$8MADwh~EMyUy``yN2(qLn&5+x$0h
z@wv;4;EDS;J=e)mWAgSB%N#{j)hriF*6S6km3Y;ar8qtauq5+eo<#Oyj8kh>xbi6B
zXFE6IQ;-bkOo<q?Z=R@H<##F!Q42>>igY!5Hln1%=zN9qHFZ@13FrTb#1k)m{%6E~
z;P~a|#IesvoeY?m8?yQmrn~sJN;z0Krfl~hMy;l(sG6FhP}q6_rl2S)%0{dzw2~J*
zI5EJ~R#a3;3wYU>1``se(*sdOZf>vTW_nLy4OLw-PNz0R8C#YGs<0$4aP(}dnKyAA
z<@D*joCb@kAw{@-@`T5`dwg*7DN`$O<4D_{xVpL$Qe;N{<a7jjBAmu)Ll7Jw#6(@y
zgxb>`Te@*TX;E|<V|n-XHT#D<y75H0T2M4Kr4cE9GNMS1$qT<C$h{Uv?TF7FX6>;h
zB~M++Vm*$LqN!<59j8-EvK44cwOC4YNgDa!YD=Rn`?iDqkseK0uSf-4mm97bF{<Ek
zd1A3?SZtT9p4<@Hkr%e+o3Fm*%~#*hwgbE2HO+d<esqjg$?C;3TwSnOZg9a<bR#$Q
zn!ot*Ptn83=l}8Vcz5@f>)Qn%e)NJ>wW2T;sECgPAr73{1BFstU0vaak-NJcd;2K9
zQ%2)P$Ej`UhmoqO(bmv)9o^82$CkDDoDVxKTHZTR;^mE1<pMoqE{n}5aZyx|6hqgG
z<wZ;kLys~Aby-WUs8%eRiu5qxPDg@0QPj6Gfe~+|aTwX{_Pl#{Pd_->cw+kfN%sd1
zNaqj#h^p;RNZlWORYgrmVoL+a%9bJ_<^ya@1a-znwK|v8DuMOF#nR+LBIdkY1tEF*
z{QWAy!8w)%G=*YW7#7OV<ftZ{sT3(G3T4<d6+;A%2oXZ?gs!Cx9!^KFmS`<6KKg{k
zVu`UO+7$R8o|WU^IPOo}-#u_R9yxWL*gi}K!U?JnC5tu}cN0>f@<Bb$Cb1Vz{<*lC
z3gCyk$BcMOVKH&O&hO8{3GaPI`{YiiD33Fvd%5kXCqw&`5VM4gGAe^KQ1o1c_uVl6
zUZX8lSzwHn!*ulcG3P2LMUk9rrq-ae5p7dUg2ab_kE6_4H^|%2;@cC+4XEI-DpIPv
zz9XjV``sz4{^fPdm@XP9j1dxkQ4xdZ{_z1d4(KS~6CxxNv8ACZ4RvEtR$6=^C}Jv@
z3IoYcz2GX!9eiIo8O1<&4m(ze;&R%b;steTs+^|xoKd*<V%jdK9kYj9E*xWu^uvfY
zaxl?KW$+|YT8-5fDzU+G-HFrVJ>~UN>P9*!HUkn|B7)};C+{!Hi_0Q!Dox)*)}}q0
zNyV(CZG6EOt?OEnC9bwyZLip@)>LKnUgofKZhrhUBg9_f9`iv!Q2i*z$4U0y5E9P$
zb72?7;M{5jMw2l4HH*ZkiAiIk!CIZw(n2p77NQ#%-9J%bxBmH`@lwGTLr?qV7o=SH
zqy#}FRPtz*NXp79po|n9O3CL_5V^e4gwTOKQLNY4^_r{gM{KUQSeqysi`Oy{+@ZrJ
z#n+$zhQIrpzva`PzU1YnA7UykszekqdJe}U+uIw;MT0vA;us0ThzfzSFpMz}Qoybj
zP*|K&=&FJUCKF{R2snD-=@mJhVGIMQD5*C~)_B_Ek$z|)M8>}5_IktaaKQP2GSxB!
zw8CppS^gXck1i6*0@`3riBg(m3skd^>^kpB&Y^5UvsiI`>v(*)V>pO&_hPlATCTvB
zm}*H>8k-ViPKs8F!b*~YH%_90PDh3~;5*M|vmpk>W^+xu7oxEVhB8VtS<s5dH?Jw5
zCa#+r8y%bL73F%#eS75Lu*VfE%=S5^sn}FIzIpo<|J)x4u4C0~@S)=R`Ab;-HR|ZG
z!DBImtW5j*?t%L;^1;(J&uh)PskvP=9N&D++b_SNvWEY||MVBEw%0r!_PlxfhUZUj
zSS}X$(+FO3yM4~_u&1fF?1w#=f_`jq!HdexmVEi`8&<0gi_He7Jg4zQNC|C=*_tO3
zGuOeR2m}`-%hptL3nxa&MkHCvq7vu&Y(W`Aq&prdtYy8eXmYy*1xs7-yj<{`zxpd)
z{;L(m)iYes5X3gd_amp>j(7KuVl+QXS*OKHUA#eV=6uhr6om0l*MFVsdUKVzUc#LW
zK~Cd!SrU+iHfC=?5S6UhC5iCDXUmxa=9f-vNn}a%>WsfS1!IE4kg%Pn44zs;V_{Xn
zrm3*H$eW~a^YjHk5eM<a3rezme1b23{TZexS*+LG+&*D@eam)x#nUG@JpJ=u3KuxR
zvFrJN{kwn1!+yu{bfW7z$*OiE6DDTJ8_>lDFQj`>>vCCC<jr{1)bq!aZSH)oMcPRT
zIs8~#VXQiTfAm6Z&-X`K+)5BXW3mTiu0N*su@$5I>1I#QQ(O)--Y3R!lmmB?bc$x#
z$iccF8Qe*fuTkt(N@J)MsJw|2M!MskICSXXsM09je5IwpG7cT5<B3y0vU6}k*8YUV
zTza4cHqC-(PoHqLSkNCk+QWh4eam$T3kyY(w@Smhwk)fH+A6|u!drtYN`f<D{;PGa
zn$HJNHQmEBr9Wjmt})`B!&r;Yabvo!g_HM>%!9<aaGE%FDuWgC>pk(q@*W?NeAs4D
zVYG1Vyid|vwpLP;%95a9>`t@~@7O;35VPsCGFhQQ7AH)IA!If5#glvQ<L{@ZTt5Hl
zxnCA^Wm(cR3#>H~&G+Pn>zgY{Kr%WX<{T->zAztTvRz6DQ#4){zvRL$1kYr>6Vo|y
z!a0|}77&YwsfLz4D5i-{AfsRBLzOlvbL}ESKmH@H*!$NHE9_o;$d%EI@80pn{XKs7
zm``6G6@x51Y9cU_VdL}9Pfc6r03Jc%z5})t((W&Q{tJHovtOc=p|qOdd;I9|Hi3^=
zZRw9KcW>@^xO?Q;2iL?9C{2OYmKYM>e*HCNU9s6-$;)iBXI)mL6t=*al2Rj%nP!4w
z5i*RAvkGz=tSKd26kj%NMG{YREY?e62po<FZZ=|Gcx+E-1;@jIC!4Ez;hO*WsaFsu
zNqaUi&i$OSt|-Tvq3!38U~3G^&4#2C`^P)F(-H4IML$rlHp0^>8&OeZGvx^tRVDOY
z1`Q32{m3wk+}}N*W1_8lilUNMtu|CeK|ecl4}A0OH~jo(KcgrteS5-|1s^|q0>SZa
z-;zA|?!>e07GGB!zy1;x4XmD0R$I241*_GPaq#q`$7;=ZI`R1W4X?iZia-Cw&nf$s
zXIC3amAL!%8;l$I%Rm1a2F=@d_x$ejFIa9iynOiqPj0R-x#e9{4Xfpf!|p)L9K*6K
zvmvVGYWtLLU%lmciv0MeFDS~2!5ulC4q3I=$VM<h_PLm!cscPxoFkEV8SB-CqL4@;
z<pX4{&$MV*E8-n(WWlw`u&OJ5{nb}&e)d51V#E|nQW9~{(DxkndroajNRdph`kqVi
zhl{#DUH`SO53Zk}6?i}J=G|+Kt;YwC_Sx1TW&4B+lVXs<jU>xU$Spk*l%==_K%yO_
zEo2Oo2Ar1#*osbz-aw1AWA8jvy{Ax#g@sjNFmc+jnA3IyZP6KxP%M)s435!1@@lu^
zmHr%47L--R`ud8it1GIep<1r^SHJp8u5WMAR;*M$|Lik9`|aoK4tpu|Mvn_l+O;Z?
zRG7_prSKs~LCL+)3H<0JczrGc<R>N+5T%*Kb|FFVXY#l;rNjy<HG@n(dO;^swoJ)!
zcqD}<VZ89T&OAO@$72!$+?40(<UJRN5JyN7qgZGtt->lrX%uA~!HvXm%v`#bWs>4V
zX-QCWK`bkd!@#~B=*K`8;0TNwh7{+cmI2DDVZGk+^y(=Kt+~5D5RU^ju%=*REtN{x
z=&6;bj-J9N3JjzXn@TF9Y03hl!Ml+uMy#+FV-z+ESX|BpPib<ZlVaC9Pl{1m(8z#O
zTA_@HW~L&@UvM1Hs4R2KR-8`q{Y@eiEUJ={geP$tTj4`vQBe~VV*R26#>1ivGz%qr
zy^5UPea-dLr>H`ca11AhU98Z?5K}fQ&f<j2w(*QA#yGc*=gE=3J|U4QApn!nREvhI
zt1F%j&ncS%tzo%YQ<P<na1_DjIJAttI9N=s2@{Xg<)SX+kHZDJ>Bn4r%}*zPK8vCP
zNdSd-wvdS-NFkf%JDr_D3`&XM&R9t54~t#Cd;Ktj;?<gMQSi|(|D5C7H@x}vKscQs
zMJ$miG<Px(mg##Y%1)5JwgrjEM<0F4Pk!<<N+WSxF(mQeGU8NDjC5_!{_c+B;lOsg
z;`-`}!djZLrtN!PzkWm1imRtjnDRZN)g&Vm-$&s<2;^=_)|`Rc+|N%Ml;v!jqCn+V
zO3Eja7$n|ArO0M`#o_*e<Ee!t!SLJd6^GMFxNb0;M9w%&6qSt09zc_safo7QT9g&h
z2b>qe<!CIWDOjwxc^TETUCVH4@e#~0P&Nx}vqTrgj1)}+xu`1<1pp-h#+DEQRoUR2
zqbv)C!Ere5#pXjRRAE?FD_lR)4n4NCqzJo*2b6+s)u68`n$?z9Z(ecu_AB--@>g%)
zQ9t{b)ytnyTt7oqCCjR!txJ@(1Rv1e^VL86niYy~fBhNZ;f}*?#dcA%u1btjynXd8
z&UtKA@#CL-%JUCjP*{adiLPtu+LqOFF>`aeuES-~OS4!~*d@Eij<3JEqbe3Wd-{?P
zf&?rNClYc}o7xvsKH=pAR3Xe9oryT_yr<BLsw%0f3LiQ#my7`)B!Vnz8E;djVIa~}
z5=9o=KnxyLCPH$k1cU22bw`e;;~(&y`_uK`<JzuPSfg>S$Mr3BY3YZaerO2@DkcWO
z+4^KE?s6@Yko9DW^7M!yEif7Rizo;Ao~Yz8ExBVxE`=i4kSJB8PaYRXiD$|*4rVh~
zGPOn49Glx;!8-=;B!5;xGKTxzJrCP8^<qJ@+7Ptjc-qfy`rGx2U;g5!qB4y^-1+;S
z<LM-^N2Az}_z;C44mqrDI=D&!1i~{+Q~O$pPnVkZ2bG&=7l$kIeUpx%?25u0o7%tf
zPEseP8=mp%%o9xckTz}bDkW58E=Dy86%r{#jEYp1#w3T12^}IhG2F#D(m0}q0X_C&
z?WG_Q8G~bR1D$iUgE$0?0Y+dzl+jL|9YdB?#nUHGdAhlxscPE02gbHX`@q^%RA`Cb
zLJZm~r7+X;Q3<UhRZ&ovC1rL*5POvDWvMku$GIK&9*}bGR6-Ux2IuhE8_gD$spWc^
z=X_b%k?-r}15d@&w2z4ZZ4w<iDaaBeg!7$z1RYRX4#Xy<Y+mhfrvqtd(M5^Y38HlU
z!Zeqd4o~SlplJI2mj~X-n`8d^)Voyzw1|5}LESX0H*1oXd)zej?3$5sJC#Y06S6=w
zUWkgOgU0-p_mCK8bnkrQvn^5-k)`tvpI^V2gnkg>l%brT`Gkwj`+3a8H-Gf?g9<yU
z1$MFE$<KblM~@HmZOi><pP@p)<Pde0&vR1<h*FcUb1tTo*pQTU#XtMYzhqf034_n3
z#3EZaB@`v3;OP$s_74xddHs&%vLyO|9~@)f^YvF>v)XQW`t*}rEjz|xKvxAOR8s8d
z%nb?Q+!N3uKaoXLk&Mb}BH<lONE(URFD%3y=&LohC@89${?xKvud%jZ7zScCvP7M&
zU8abt%Rz89LCr}OF=DKc^W*68&f~m8MN3syEH+mtYvJLJ)9FaxcNp&>$w{DCEYWqH
ziJ`(5F$vI`K%&rz&34P~VTbn~g{CTNw6+8y{Wvm=PErgCgNc^=$9rzBZ=}#N1#xhU
z&XGc-s4Je9mM6uMcaM91_nW`tZ~y&Yv-!!-sh+*y=)oz=&Fw9%j~HcG#mN3Qf6x7I
zf5&YcSgtB=>WXz;(i8>Dx@Orl6lIAiOU4+{Rtl?_BD>v=VHjAimUDLDFpMk~4L7&9
zjALNG8~OUHx4e71=O;h=8H;8`KiqRX9&uw&Wj2(#9g(s_`Q_pjLy(1X+Ca6&kloL-
zoz7GgCWK6$ViI+rmDW@`4|?jt(iE2CZpW=3F&WEFfQuvD&@zlep1|r4aP0nc{kObU
z&4R)j#!-{5Z%8`Qb|>DvdxHx>Oo3y-#UNgG7pUF5_a#t9mKztO9j6jZRHX#CkZrqi
z;o&bfEg`2XbW~JIY!1e;qfX4+C5`yt8I=R#bwcL?M1s_(<bZ%nayV(nfo?cqs}gNX
z{^mEoCZJd@S3J4B<)8nvU-1`z@fY0Q+^|?KN#4^9BX8cm<<+ZKJnnbg-QUx8J^eW1
zV!&t7LkNN3JSMM!DKaM#&3LmSo5)Y=auzLRa(ajfJ)tv-tcFbakc`Ynb;b-~@;}Qy
zVF{8CGndL(MWH3nS0Ttz3W1ayg;iWHYkb$^#-1?r^lb~{fXN4N6B7kVJnQm3-1jYq
ze&pl=Z4iW2thK;vu^BUy-XX#DX2Yi+f6Vn_g*&vo{`wnmJ+;zYl{IB_DCbBzir7T!
zIm<akFoj~VXjn82g*Ac@1+iMPTDEWDdcjD<s0#z>3H{;RIt||A#!=`|)?|Jl@3mFW
zzHb*rQ~vsdM3K&g<0MuPKB@=AD^_bM@aH6)C{7M4(}!a0@w#AakCa7)sx*a_lTS<$
zAGM&QiusqGpVG9F&R?8Cq)QXyX+nudp`|FwtbkpDPDBD#A(4?viF-y$Qh23Y+>PTn
zd-;W2+{H{ah@|&^pYUSy{YY69MCQIn5g$2!7GTC0@!<j;%>~EV{9Ep;q*DD8WyPkV
zAeIIC>W1y7KjG%h8+P~i3}esKd6Ws1$*oDv5cv7!NT3t9x3_%q@y8evVHhd26m7vd
zibBCSFmx@$>Bz&~9mm6-4?g&m;5|Aj9`7DWA#k<c%tS|J4K|A6wN446<RsB|9VvBa
ztI@VZTgg{c+RVm|lM!%KaXvxD+#i!(w}xa0-brz04Z}F{%{Sk&Tr5~DmguFum>Bs)
zI-NdadM|iAT}+^9;F%(O@nyqJ6#{q(HC}GE!VRcu_PYnX8yQ*&B~MDiW307c3yhM<
zWC}~E8)6TYij-wX7)2GAe{Gh9%hDTyct<G|PHArSEp=Vd?oTYr3KJtbIE)`CT#xH}
zUM_2F*iaNFzIydJRgqXOuIcoOGKY7Eaika{O^R?lvRaqCxVqxWbwg8^Jh{1*Ji-td
zoa15l09LbD)}ku(5_~<`%Q%;)>zcBZ*SbCRJnmZFyn5t!zkN+vHavUw91>Y>-s8Pz
zRoQH>kP{~Ik^i#oFiomS<)@I1a@uTTO!z4yap<YtIP)VyRy`$z3uTjHvsf@@(roZv
zj3%R;;Ck1Kr<D3(r?Ed>|8&=5yXX39OSN7K@}C0Ddd=qMmQ&kuIP5vKCywoj&i4d|
zAW0gOkw9~$6lEb?A?G47`{LQ$Nr{o!+bm7RojwB)^U)zj2INrL2Qjxb>dfaWw80{V
zxi(u`C?)NyU^Gb;GKQm<=)?ql45)ELg(NhEuBAO3c>C=){QB?yf&b_KL(g82FJFAf
z%a1<cM?d-zKl<cjR-3<MeYM3DmY86FJn-h-J05pC4#xxS@kINuBe+2Hk<mK_=f$qb
zM@AnAQwu8REQjc_Yk+viB|j30b6#mAF%CT;iQk+x25anWK{Q)fAVPv>SxYpNQWQp0
zYT<V)A}SzU>Ikks@$l<kqr)hLWfWw!P;xJ|&cZ5f*q>TX!@y~D+_gQ2nCKGp2ue3?
zd|Hd+g9*}3FSO!=n+@AVg^8ZK`?s{?j+<h^wrp6{6|oBt<TZ<0F(k=(4oYFmlJ)h9
z&Gw4LQV^}v=|tc6k`-HC<ez8Kr<y_lIPWi0ROaO5FggcXW30(V^jRr7-OuUqE+2Q=
z#otqlPge#k3S+5kL1itau@EF7${0ggi1@6uB|Qr79DQopKRmLqHKnO3%LRUF`No*r
zk2)j47jWcz?3BymF3+75FFvO+rl2Tbu~;$s0c~M9F7SSYNKX8KuIr?b3leew%&~fN
zhc8WH(KGNf{cb%0ITwYS%LQEsBBXH6F}e|-Rp96LcW!%M0z*p8Vj%xdl@*(mAr=*J
zxn}v{M?CxGU$DEs<D27NT))-%NUE8n^J#8B$T=%g*u;-MeL-D<A6oE%F$lR(SvkRl
zwxxZ1;JAC>?dw-877b5tZzYw%TE6-AouoUoJx*y%RZ`eO7CEi)X#6lTv>m7Rz&MUl
zv6r<NBUTl*s!-NSA3vV?B09G<waQj2nUIRoBxQ1)u2I_Z<(FUZ;fK#Lra&oC(wh7@
z#u&6MWWj=&YA2&0L*G&fqm0G`3GYrJ2VVyVKHCFmP0=h_<%puwVb9oigmEOP69g}2
zq*b1H#>@lH76l3!)O|QfF*C{3^8^@&j^I4OJDR2;#l&v^$Y!}ArHC6xmUY9aJu)1R
zTrC^g;2=6y7M@fJmmqaJcJ0VxIP&lQ{_hwoOMg1CY6sT$cU%u0<=SACK=EDc;N}*k
zG@~DI)8I{!zHf;^#4M}TN}Sr07)@FuyrARpNP7yrd2`_YZcpD2ym<Km>vhHGT2AeO
zz8?VL$jzQx>Kqy5GXnUYYy13qWsXIDtrQS^U>L_)HQjeT${s0Ppr|UeHCUsljb&9Y
zcno0Ba)?M$Eclf0d6Sza^q;Q({?~6m{|$?`&1`|NTrH{Vnr5}3TC7+smULY!#ohiu
z*NaD$_a5(}pc6{eiIds&<cDpHtS8fYI2C%A?X$~kvRCoxLWMiMB^Z6csECOQ8v`ad
zp^IeXI2uhd7M%<x3ICZXIYJDxSS=PzU9B)F_#F6g<c)gG&<}j`^_SHDwq|pE#rFD!
zvZ?1NuClHu>zb#xw=^HV#722;DsROhiQl6SfqwM#gPU8uNilDD(L2c-4L+l4lYy(k
z*b=3!sEfmd@Z~Ca+E<d~(c(t=Iq!+C#SJ~t50c{H9LYI|o)kvZ*kP5}6KOT22EgD)
zL1&$(^NG_i()mCe6Z??pfH5C_y*8xDUMhN4#-k%N*7Edf$?bZ{@%}Btv19-4p6hza
zrfw*eW*mB<30Wzcd=xaew27vySZ%i4K6^shNED&Z2F~L+Vzs3#%J;n4j4_0)oYYzU
zDG`(BqVBQ~6v5s5%H`?bm*00;)WNxon!G+ITu>!3UpC0v8hns8caCT*GauR*i%HUQ
zCMMd`3BBwvi$Kwogrw)9O{sk1Im33}nIsp_D<8fmba6_)n$~-3Emc*qsF$*bSi|5t
zT<^|pQYS{%DNLY_e&3*(^Bz6}^y>TsCY1EO;y47xabO%r`o5R9VVIG-3Gq%D;l3>D
z#6Vo$Pk_uHbN!&gPNzgr8ei7vt6Q#r{1ZMtJaT%x<KeU4k$gaTnRIjJ+x(w=WK?9c
zF8J}s&lpbkDA!Zlig6UKQq`11JYnb<PDgh4cN}*Io;`bp1{QTq41v@Bz@n--ynDb6
zBX&_!)*@5Z36u}GQ^zp0j9p9Hwy2~;M%<4SbwjymuyuthEZT}!jJyz8Ore;eYNa$T
zfi4QX3nXP&t~a19ufBavVG3?;Z?VRTrG{4L$IyC)h-XwtYgG2INn{>~kv4A%lC}}T
z2rkRptysSlb;G8x6t-mda7Q<^(r)n{lcr)cqDyJF7;R8iq~5FLh5|zhAw$k*ONST(
z(L3?+19Vw%>Q1=PQQ3;)ekUm%Dbn{nCw0!dH9AnLNbLvQZpW&zY(IQKTNvsWPx<V(
zzoC2kmTM)68KdiH2UsqpNbkoH=M&y1BAWGji?TUG&uFlDF^Lg34mdZ^cOB!{b2uE?
zJvtsA0&VM9u1Y@s<SET6aq4z--2vx4RVj*Abusxay|bi%d;b)Axll}vjfAwuhS4#O
zaxZrG_c-Ti7M{BCxt(mNY{8;ja6}_wsO-v)6afAY3MT$^{r9>Kr#;=!QB}358dGBQ
zfkh*mMOii!*^6vdEjgVI^j%BWcC<rJ@Bx>_FfpAMf0Q6|q|Bu=S_-ZC`u&kGBLqOq
zVvV`5Be!>xm?qe4U15U~@qz|0Ih#A3J<k+bbt+M0;z5=(SG_~*DJ0-N1yA3%jAPI7
zXc*nV;2ia0A-<heiD`?O43uIJb*DBIrl6>6%BmJ&m$n$R#7G)V!ph+`PLWGe+y(Em
zN>$P&QdB5qK}l4abJN$uIm<Ol`cMiU*Plq<$zu&2zVC>BAbCfM9vve(#o2a3XT&rn
zA+Y-p>Ah!gfzAa^A=3Lu9}@$jdL2=CtqHmy8bkIynp>2$flY0BvRP4EO?x;n9FN3t
zU}*~si5SG+QX4WtDaEYOiL$9#tXHfyTNbMoO7)_s%*c-z0iQW=li{rtc4tA?WwACD
z855^*TKkmNQpBj7v;IBp*vnS#IYR8*2!-6Gi_)o(c+Wskff%8%QaBrHWj{h%&V^N&
zq#1`1H;e?A!J)~rFF$|jDovu9bTM1hJW(!d43(2F=CwQX?6i`E6>AG@D6nzFg{*)L
z9`77(@YxhwqW`q9^FDdL&U*phc^>l5G=DYaPM8}Veb;kpPqeL2aHiMNT?kgDsJN*Z
zmDhH94gdJv;s+IWG9V*EQmC?`ynV{EpZ=W3SKn}Wcp%)}W3xCwZg!e2NyM!^q40e8
z{EDWG93J1YEG#ZMzz~(9cOy1MQs3g*mgD0Nt>EfvLx_>8tT^ot=md?i43B%F_t;^;
zH5H)n!-#KtVm}g-!$hdGAYs}ELfb=#((6wN6G~EfQH)J5%f!vd91mMq#&OIlR6}x#
zmoK05#qT~R#l-P&#I6?TOGQ!22w)y$XQRk72V(ktD)SzUF(j>tagZX<YtWh`I|52;
zswS~bP#hF}-x9`w0crW_vciT$QEM<(?!U1j?^arnzU;+^o}B091QkLgdQa0dP*o7l
z4)QK|T!;+WwcE#jULd;DfIpsa?ufZwU_-<xc>4S)Woft@NB**|7=QZ@{QO5xxqY&v
zDNCxd#%N23n#vTAB1$K;5|hxFGa+LL`S(h&ch|NIL(lPe<akH~C&tB}{^&zqyu2cE
zpzH2&zC$ZRU9GZ#b4FlAuEnG*6m#k@yYJ5vRss>{i5#Qw<YX)i^j%L|oG6Q$pay&d
z29GL@c>I+Mj4DVzW$C2&6(uy0Keb^0N4uOKg&Z9N%34l+Pgztf8;PY`FB<Bq5hQ!L
zrai4_Pe+dXJ-g!p)wOiqGt3J3X=8*C(u|N4R${(pGuZC~BQj#AWZ2BjD4pbOLXb6L
zc&rnc6vQMs0xA+zngt#rPRP#pTG5yiHw<{^1YJSpgaSiZ6qKc<EH08Gw8m(K$*B^G
z$m7EuclQsxdw0jHZ@)&3fzn!<s%Epj;_CXE<!Vi{Sh8rA6jeQY)RkqK*V~M?WiE6I
z*;_AoIRLo<?=!#21>#)DMVuRmZWP4N4-BV+n72m{ek8d89fKqxsFa_-Z19DtaM^Tu
zbdKIjVniPTLx3)G=tjWh_dO7lCTc@6R^ZqvEIq4Y*0tt#QE;<dqMRcRJ@}DQC#*^t
zJ@ojorzov>6Q+ntk!opJZ`N$Duh?8)Q&c6PZ*vrte^1mfjgh$k&<3Tvtmo>C;sLWb
zB7_{emBj#37*Fw7dM@nd$G`X;({(AdNWyDj$^_F1QV`BxLI57C#YfO+i_u~<ZLAja
z`KTF3kLyRm=&}vFKpj(N_f*8G%Z^u?4&j%;(fRqkNB&CFUaK+6%<*d^LDQTY8HWKk
zI^5tg4_(dQmtx2T+eDqx=b|pX_wKzEuF1GLIFB0!hQ8yJg-GK#&Kb)t`1#*6_;XQ<
zOx}AxWUu&Pg<ZbM$!L_Wutw3`-0<S3KV$diHE;F@k{k0M8c9i?$>jOg7DiE*uzPsJ
zvXV)vjYb!mU=$vS;E1OqkN0;3?|A<FmhEOOUN=h9o=z0nvR$s(?H-B4h*wH(-YAnM
z4Gvr&C66jJm9fN#D0FZnrSQ7XC+V!7%6WQo?VoS1sM@kro(vD?Q_}724G-@g_~OfN
z_~et1q(Hh1%D&*-or}HPg5v^qP2>}N0X`Cf!(`TVN<rEa7@`(b&Xy(CZm9}Ido-L*
zojAjH13Jp{Rmz|WaZsNkw92wXkqF-4{J<DSDe$tnn^ihT97+6;GO{rhrG)@tl+bgm
zB07gpkui86QfY%CaXKF9?(aAqcl7OvufF`8CJ9Qo+${O@CqH7fv_cwpftWNF15r_!
z0+qZJcRDfph*g?^7(wrMJDeK`-ZS<C<KRfKAo|414?p0iKl>C@1P<LJ$M!+Yr?sW9
z)oi{xDYRp57krmnH(g4-fAS?s8`gV|v6gBfJiQq?i3#N+L?;NqSWuNsMoBZpKtakf
zesLhbI7oc&^{2}3zvZ>8HQE%Iq9z2-Znr0dL|Ig<nibpCiq)cFy<D(vY8G|FvTAsG
zb&F4)cgH>Zw&l>Zw5QV?g*c8QfjrH!1xKOH`{e73Yv%N%vj^S`y5;kR92YZ!r=U<G
zSYUq4Y^@-t)BM0<C1RtZF^Lo$8nF#pE*5i(&ba}tmOOv@M2eoSla1JDbV_(PFuKIh
zo^V4?@FOHqf*T-t$L{Vek9T*>P$B1y*%-rmy{0INOs_E*D-Mz818@|?4l8R}rlAzZ
z<^&6A8N0E^4FltF;<SH66R`>^qo_=xC^c3ajFN;7Fbpp7`rRXgS2!;%&@lzZC<UD7
ztaSD2T-eF(7S8Z%(x41zF~YqmG(URrob95)cOCEUzGhu7SvR%t_1@i~tQC}1#eCnO
z(9o2MXHT#B=;d>^+ciy7G7cxgFhCqoxv;xzzfL@^7$tpZ;#nGPu-d*?*v&lPtRjv|
z5tNz}lu}Hy+xB<cvC|lxvue56%O(|V6tuTc2BW3|5aHF8C1q7gI+HO3V<a<|f+44(
zxPE|;rn7XWa=l+je&6*6m<;oE$rp4^RH-hvMcd<v!+uZS_Y(W%<Q^MkFve08etr(}
z_gi`WB{wu1CTA2h-_vRSUbb0<8!o`aFbwp)@V-Tm5$3{fdLJ*={WNtD|FeJT`T9YH
zT}Tl_Kx+vRPD+v1Ys#0Ou=>R#$A=?_U;mnd2;)f5iBTK6B$<cWc}yHh;{nSptEQ%E
z47xU8BE*34p5fRs9FF+5rBI3&&!16N78gC3NIRTZZfaDivQ3Etmz5{Okl}opXec$S
znu<c_uMNaeZl2NxV=9skObbvX3g03sF&8DE&_t^#VhKvoSx@wdQV(plD_%Wl-aejq
z76wwNavN5>ctFo>6d42}ZMK;QC-+kRl}sX7O|+KeEjmdqsZLQQc95irWHhEwB)djg
zO{tHVLkEM$Mkm4*t%%wK1xD4Gd!tbl=)zD0OU$w^N5QDgIUqVgkrGB55uqTGYoo-U
zD>)5LW244KOX~+LiC!g&MpG6efAPT;fBSF#-`sus8as}RG4PkqZy9xoreu*K$22e+
zO+sNpq>NG+k2cWR+@21?3GBN)-SH6>A^Jq>Be6}`7<hJT`1#MCakZ^!#{-AMKtCix
zR9GV!uStp0HN`}5UW(>P$*h&2gDK8CsLl#Cg_1a!7|{vwiVBI%W+MUNIZ7_6C=z<3
z8GMg&k+QZ_4z%lWG194o*AR7}Fdk(xe@dyD<C&*3e=6$!+g}AR^dnv8shcIsx+W%Y
zJ~0df_kQGXFgz5Nb<?m~NO*Wt)f831dbOr(8aDolHn(V}g}fhzd1DE2a`ZpvZBMQL
z2_1+zWdf!#a&Ez*WGT_)PT&M4#5`{Dh9e+MHayj!m1W{O`(cz^+U0`Hdc(^XA5xYT
zUDt9r95_54_~P@=<hhBT3NfSH&U=n+%ituYY#beRRbT`8urr~MULw}cjWR$)2J|f|
z7U)!BR6%JAnRD619+mRHEg+fSG4tnq5(IC|NXRfq(0Wr>7?rYxcwmf9$nU}9UBHdt
zP&hO$8U_kN5_bSkV8k&Ja3tJx-!D81)kL>S2^GAEB#fodnnqhbxw&OoHH5+Q`0kOz
zzN1POTUb=^xy2_xhm!Hul(47`SDTvWw`<l5OQ}7{d0clSjvXnED8ii9F=h2yYiA@5
zz~(&M%fssQeg+$Idv!{JiaDD-nkfEhYC1*8&xR0WeIjAvIgTqO5w^*jC;Es*(Uc`+
zQDUtK)vT33`@+ieo%r8AL<(y}Av=!vVF2&thHFhSoHJi9F}NwgY5vTlhGphI%$)eF
zTuRwUcpMyKKhm{Q)Ey589v&X(yAC%x0fmTYWn}NOhQbQssIoD)$^es^3cKttr(mAb
zQ<R&};Q=ALau{VOv}Ksv#Ayaqq(z>KyiAdhj##=tF8{de2Nia*xFjZn%Ag{lC}6!}
z_3~q$|6)%#>^Z#p2BmU4O(#^+=%}$qvuFyQKD%bMT~Rd!w$Y>_;Y}nCUbu1H$YH<3
zk0YDSmi2bczHM1;Rv4q{$DY-y#zjZc31hT8rxdYTp^UUGCtp3tEd(VN1t}*E1UKS(
zgEA6AoYdsUqi0kNEr0m5Ko%*7Rx3qNV3lWoI`QTqU~D8zA;wDO!C>TZ=JtD5>`W8y
zJTzrMDupgACKd!&WU+w!eVUkE&BYutL`5_jvslq2ON<sDPekX5Y6JsGM|1&7m7v7e
zE+|mi2=Zr3@L;qUCeJD<6b3mRlY}SWLm>Ep5WxkF8v{g7=#O+mhi(=;dG?&XSYlQS
z%B1+!r$0fBiBsEBm70gr@asd*pe!FK7y}H!p|u2~hbWNq2zaBjvRT%c6oXhB_=N8R
zVepuwS*>ece00Oprz@0BblpJTkAyhI3YBw;NFrrxgdnF%E>5RPRCuVV6+f*c`L#-9
z@^+ljyHuncZmx0Bqm!ui#?cW@k<|*y7#Mbs*y~8LkV|I@C9e73(R=;r`tNwH3Qa+x
z^I%iR2QQ7Xh8UtK7Tt&&zzrQkUr|>jO;u6UC8nw{MZvNusmqeKtfcrDg!JC^9sMxS
z^}Q5j`QV`QZ<;q^tw=$#U<Djdx&O*s^XFw{{?w${n?HecA|^~&<wNR(iUlGz&xfHO
zxPN$L(THtQQ`eMFN@=HQjW$;JUPjaR!f`zv4~$-l)DXo=L{^PyPQlbdlg%}qH(C;-
zci;+?FHxaH$1HBjb6)(7E#l0N#53hIkcp4vyc6+?6Sbl6iPOAZ%6cJL-QES9%U*M6
zoU(K%K~hopr0|HTZXilZ<3+MTLe$8UV8EnA3Di1KTgAd!wu>dc9~s-0p&KcTA$lik
zgAZ6G#dktp&ceX5(QMZxPp+42mL&xu^6)7_81QZ&<rD<({fyR0+uY{--i(}0U8M`t
z;qQ|1-}9}zcuOgr<MF2VO3m&N^0?<Qc8S)BiZq~N6w~`Gl!#gKJqZ%$$BvN{J;s6x
zp5*0?=s{(3-XvV}DML1W=dAmk25hG8Oy0Q{DBpyh<!+15t<k>k={iXc>DyKiJ1-8E
zlb}Y_yhZlvlVVR>n^n`fb*;}TC@E$ytW2^mtsrqJ7k_51OCaXf`DJ%+dOapTLC*Ea
zA9?*SAG`dj%suQV=8(!5Zk|42^XI=rjVJ#0w8K5_D6J(Xg^xitZvxxfCAS~mQZ6fW
zZBYe80|}{AeHS<#2ik7HmK9ew*Q_=h-tBi(bwkKiaA6A}`Q{Tw4D!N}m;;|h&6DHJ
zB6(9MlQScUEU$Mx-bfoUsREVdYH8ZxjS|j^90~G*BOIwADh|Ep?l4d`nlZxiqTMI2
zw|ejWB(J}`F3G7%Ns<Ibfy(v?DHmm#jCXz<#C^P2E{V!87MA{W6ghR<QIasVA%p=_
zme|5zY=M}Z1VT}Wa;YrPwh)d>em)^Z@D3*KT@YS~G8U~%d~~$OBPxw-Zf+^Ao)8vS
zy!qyyzxe5gJPeK<@aKR2D^Ag4OUvWkYyN6?<nJGL{Kcx_NnNlh4K{gH^f;{uK{1Ys
z<O)n70t&4QlrnPLV!|hnvaqTaT;JSqb9;j=3x;u!&2t<GF;EuOd{Uf9#6EL)f{-kj
zR*M)TY4)99o>21%7G|hiqw$&ZcP6=-8BW*Q5(bTO1{)Lmhj-k5++$TiNClP(tJhdv
zB9!buUH>Con?e%`P3IFsi1f~bw$!D?Av`$|D+Rpk8TuA$G=<UV!m_=-Ww~CnT&-Bv
z4TZ6UW<l_tki@Rycsj8=>~Z}d#-%DJo^TP-#I(+yIdLi5?@Xur85Nrj{3yb7`kz3T
zg!s3m1WA!nKqV<kf)=H2-*>d9j#sblSS?DfueRLY-tzqUbDE|>8$0{Yo!T~YkmRwA
z^vr~>l)x4SE8FsHp`o&IypK`@PMj|19XeE~FoG|@Nnt0OrNk*|WeHL8l?Ny7Gv~AG
zzY-*H;?0H>ab3q$#A+pn{gl1(5^|`WqTG&A(IN%z^d?W($eR%pU^&bQoz7}$E?K1Y
z-Dp@?MO_#cMM0$t?)DF~$0N~uw#yp4V;lp<2TB7;fIDn%zh12hZnrg0u9hrIMe<#a
zuJc%<h+)J#p?6H{(Nx$?oZ#;&Ulod^&wI@DGt>897CjT!oXdUbq8N;m0W_Cq-Q0?u
zIC{CTlTKMq>am5i8S_KND9=yZAQnJTA~AK8eI~gP6{L%ylGtCzlrXBe_&d(+d?x<f
zOrN<x^-`KYy0oL3Xc5COFmxSl+tMD7bZyI+1r6lHDWPI%>at|wwp(LGlxXaXZfb4k
zF<TVooG6j^&0)y-#3AJ8tI^tGbR>~quM7nXa-WNo|AOm>6?Pfo)JmZC>86Dgsn;to
zAM#{4arNpeUjNPCpk1IKP&x+$bdmV*<0q_d8_Gsgm6k}v=V0A1dd9A&JGDd-^=e63
zHw0jGk^?#nnUhcys36sOa8m4r7?>0#I#=8&X;N0jq+}2cS+@EJvN(j%h)>cK(n}#x
zD{A7zKuQuRlR^&0)|#j-J}O2Ex)?Z&kuhXKe0tx);yunn5@pt9!18p$xa2Y@TR=RU
z+D_($Iv3U{CE*c92*#l6ioyz4aXjqtrxuPK%7bk*GkegSAIB6W35}^oioT*BdWK=d
znx2#rWl?4{QOu9+|Bt=@e6nr5?>n*26IR-x^ErLe%>e)r34kELEK(Aw5hZzS)mR=$
zL%ZyM!Y)^N-pHCaQ!}qr<5Xprs%(v<l4fc~aYWG)L4u?Rh=@eI>D+ry>OLKISm8;2
z@q5<ZeGUK-S2QJ2;sW}fj=Oj7wbt|ee&63W6jrPYTuW4n+1{L;*_6rd9<H47&cgA|
z!^f0g|29k}-1xu;c<I^a__Z&8k<--zd+jE{J6!U-vAm#LdtOYMokA4T!6<yvB&X00
zViie^%KVxF&<w68<XqU@-6==c518-ninrEk$*Nrwg1ollxB|9CSqvF5RXH0tKB@Q`
z3)>A+mjfy{i)!{fBBdPg?8T_q7MsPSFcm6U7$UVzoLoG_FCJj#3ah3Rs-P-oV&yQ7
zM}7}3$sZkm@JG5FC`v_D6*!|=D@Er6-Rcrk)tH?B6{9A`AdN)MnXamec5s-J6Uwq=
zZ*Py?-CeOunNA5Q($p0P2m6E=#4e>5)L`AVo1!Uhcs4oRf`m>Ajf~(%<Ub@&@}oVF
z%=X3&%>YK5Ek7y<Qedp)jq2d>K2oT(nYtxlwH|Q82^SY<+`W5O)+l2zg{5v9X7e5D
z$%MVVU5;-aXFpIG@tLF;mMaKB3MwBMx*k6`!r<`E38jS06W;8@Th3M_WtD6q3<*cV
z$4EqpRf|;{w|-EVx&bY8o8ZJZFGMf#O+Mm-Bl^hI+bSR&x(;xpr!3LN_>1u~^1AAj
zu)stK%2pu0f>RAMZD~wNUDUk$*4qqiFJbUR)WwRKg~lqXN}&eX7n%xg?7{KDl!JN0
zq$+5PLk~R)QToQ<8Cr)mmOx^??wC#+rc<%tan1p3kg;*hUlhe=RQt5(VN(c(fOk@$
zZ{xDWwLoPW$~fj-wQKoZDN$GTW<3>pN;Vyik>q3+(XKkNf$DmambkN|H9C0lE$jhZ
zQrnWUwCvOst#@=6=cwpVg3icZjwU9buYWloWA7~jkP*z;qFsI8i-~vMC!_b7g5~jU
z;PUdE^K*H-JU{31;#`tdx)#^>q~M-t)=OKUjZ9aJk=j7z!p@jNGS_o7+xXgvv87C(
zg3o!gQK3~qVHLZxLJH0h8N$HedYl{3Lk}@YemGJT$C#~9-b1bTo)J4)G{mE5B>5$y
zF=dUNPH2XfYwvrN#n~y{yN@t~$Ap2pgkn~5^Y#J8dJ+RDlNC%6<(wF2b}gyzD5{!z
zI;EOUH<291T2`y2j1-e>)V&iQvXQeQK4m%!Eo3ny37IpmD7Zn9;z0024)BfWkT6QT
zpQ2LYR-Vml!)R3O6Qn3aUTsj?5|Soh>7$|xnuy*OXiuSVN^J{-t8Z5?q7BRqL~C+U
zDGQ&1jR+zHN$Jo@7P=vL3DhVI#eBw0YpgQ3)tY{}#s?^ar=&p@g^a6nfvU>_Q<Y3+
zQ~JK8ZQC4kBB<Ty_C6Lh)+p-2GAU}Bx<+dV+Muf`mk&=kxma_sI>#!{H@@~o!kgbA
zHZ{wOGv>uLqAo~9!(c1!-+z~8eaX&VMN=h`kpj}Bg2ENJC`G?jTKGZA3M6e5RZ}tB
zpE2E?fi-llW4T_@4FlR}td((qwXzw8?4KootDIvLD8J>Oj0TfWT=Pnkl8ifahUTM>
zA~MN@;lWr|%O!PPW3?ef;T91Fno5x_&sd$lPTABLtFR_glvXl`|LFL4<~XWh@QGpV
zP=&{slExbP(z14gG>1xohQakzwZ)nmZ49gBQVei`ZVl~n#o6fzb=@$TOqlLWna_5Z
zP9{`M#lhY_-VY3eqw9MX%LS{|T8h1{rSIJ)<zk4!)p04KWgsfg9MMHavW*fIoz3f@
zi;_~{N$5d1%A3J$6b*pM`;akWtXq~QdyZAa7-_qn)w-qaI#%meMo~u0ZF}$VKH_s6
z<l&^Iv4+YB1){bER!gC5>QW4V^#~5Y7MYMWZcM7!gsPLoe#CqF80f=*PoiM$oyV<K
z_#}mz%VWEg=P@yFhG~q4%*|Xvin}2VQZVMZuO2hRbK_6+kS#n6vRT%~FtLVOYZ`5s
zC`F+(#z%r{>DDU>ZD{I>sxWBR6P?HPGQZs~VK#-`spa}!!``f9UKuKzu*ykclU2V2
z58e?|LCCPzRbk%KMVr=UGnT!I;$8hd7)1#}bJ0m{kUBM@Tvvb_uKb>>k2w}}8)Pgg
zNSYKpVes?`hQZ@i#MUJfl`uKNtSl^OiSUZ9Cv+{C3SE1d+@vTJf`m-mqVS)`<^{&_
zB#e=BX+veuQd4CRY8;D(;6?p93>#i%J9^4>9YZey96xk;KWr513IyEH2A%hP7zHzC
zQ`~7Sbey6v8#E6+f=hz*hmfS%h#_-H<y_j_L>O&0=U2@5v&UoXWu(2y%lo~L_hwKW
zw`!}g(s$mYtcJo6O-Xrp#Ln~2aq#E?_b!%1*D>C_hljiDAMI1sf|PhC5nxJzYuC8t
zg3vCovBylORI@3zt_TsPvmLaxv|UG0$_rgK6*1*-c-WvbPmD9tm|r`NjTOo&`YxfO
z_#yeO#{`8-39IBL4kVPZnbRg?ftU%H#+vQ#(3)m41Kl(Djv;d(7|kKAK-GFAUS@UW
zCOrE}5%J^=ls4Q1l_B;R<H%siCs^m)z6@I{AzWKyXl64iThJ=Qa<!!ITYPjx63ui<
zQ8u`%MP*T9%ZkZt&U)QZlr>c|l}H|Aa&ec?v62PX6lhz@v#=W5Oc;ua`;X38^not5
zlu@BWU}`FoDY&&W5mX{-d_CbYfrDno-FHr@!<ys6nn_jXPAy4tg-*EOC~QegiNOVi
zY};?EirMa*qNy01W7Vzcox^!YRo4_nc8||p6I!lWl@|cnSF$m2RGY$X!;#dIy&0o)
zrp`H1j8tVw(h@Z@E`Y%cZ)1|9-4qd}Jc^D|L8V$A-TgW@4vw&U4N3I`)e8sp%E;r7
zP`iJ-$ML*kwRT)|k=~`u`4OIwtqgsP3?bruqACQz79LU%RcDSRj3MIto^^D2BkAZ}
zi+7$$KcR6AT3c*km`*3mW;15<35&%ldtI%BCz=xN(6MaS^n>tR<CSsi=zF5_x||YE
z@Sv|Gt7hablClNLlZPZU<ixzZ8C|hpK<*tsW}zvWkaTX4d(K5d0Q!#F7%FXO%7RH*
zP-u-+T7(8lBGC+Bwcb8IrAR6gg-N_A%)CnUG1AAt+I!YrN9%H-7bK_E<-$%pB2zBx
za_rcK;64KI+s)CaAod(e#9JzkBXB8xRklZ|%TiK4QlLyp(ms;L#L*5zP{kA@P3CJC
zMpKd)!)mP&)b3zjus1dAP73yCHM2@noAJ4L2!qf=Vi4pr%BvyX(%bRtRWtQXI>hGO
zJqa9a&hQ8@AdM~hst{rv*FQO3N>}IsBbc+vR3EFWwB;RIBVKhnzor`ph$P>j^Y~xq
z>TB?hIP@f&NWO<Kps<2YtFSHZ@^eW^3C5b_n$6qN=Df<Jj=aXp=VJ_l;s%$CyPlyR
z-v-<;$f$JmlM69h>SUxh^8^*bjf*KK4=X9~wAzA6Wm)9HPChnThzX;$Rg4*F)G-&)
z00LT|A)2@m#|XzxZ<@EOX&S%R@vnv0jjv<Oo@`0Wm8)%qt|!cH+~E5AUZKBO@b(wJ
zF5@N>IXFHhn38o=)RQT0J&=^4A6nL}1Mh?=nlc5{Sjwj1@%aU1(-4zlxmvToyGK=t
zJA65rGRwR$tyCsYZZBkVl-IaWd2dn9f;E`l;=2|%^r+woT}z+;M!P|DB=j+=k((fo
zt+mEzNx7)&GW${mNtw8Ux^0St%(+vUlOn1b^~6my-k{@nHe~Wl%6E?ZePXSl5HF@N
zta6MjOA53FRTPwy9G_(kalOQO$MW(*o_kTsrld77M2ey!mK9}HF`ds)S}~bSu5{M2
z0Lbr*L~T&Ez#5IU8e3JIE!Ui$oUvH9w0@vaHOY6l<r$SuT%YXn;8HfXa$+%Mflrb9
z-s2yyP?Kxim>4QCuOeClCSVPO=(BZ1pdSKtQ?h?}NZm|?yVS0^%t7Tk$1CaVn7*}7
z*cQ?$Mxy44jb;2B$6IIgNixQ4qhvD2Q$;o-36~p`BD=f`&qa)z6_!#J=waaG@jJNL
zcQ|}yj*FUQcS$#N-?O;Cdi+tL_iygFb2wp9wwUvQvsGm6JW+wMFx%T<(R-rv3_j9K
zrX;QKKFIkmN{VEJlR73*1kVsWL*KJ*7pxa6s-~uHY9^B@vz-~!>6GboPBST~s)||H
z(6&?h-Vst_*{&#ZR&v*MI2Z7KyTQuNx?P(EsR|ebg&7I)<9ak6;u^5s$aO9fQxe5u
zjAQH(`F=8a?5eZKN}5%pBhe&MOsu+|0VshMSQ&};GD&<;O-eE*R*Kr%?FdaPP-1{D
zc%mS7PJ_3a0mae{GNFlL?(LAb02>l}7hE2#=4wu9QMFz<$0~V@id05M)FMRN2rAJg
z^XZIa&F;)5RHh&ej@0!ipNkb02}(=6Mg|G?W)n!B6g@bPO@X>pOiIJ<)No^e%5GCK
ztqjvjQ>%zh9u*ygUIYXqFEGUHM<{E5QP|CxRcAD49HGjo-Tswn^r_3B!d3BvN&<dq
zJr-)|3TVT3?Mqx)8$$?$@iCJ8+}KnorKpO6sw`1f6SWjT#^#)DE8|M76gFp&4$hIr
zXvKa_=v!1>OXjVLs2EUc<h*a!@=<3!QV8-!iBh9SY<4q1l%_HGjE?u4H#hWB)D4c{
z-KLS*5JvNB6G%B-L=?M0`Xgde#FUXYrDWeJi#)Ow<j$G`V=6{_lyRWEjrCM0oi>Q4
zKno*ylW71-Wf<oNKi*@c*wx}#PO8Yk70Md4vS?Lc=W}MaZ?g0JGt}>#(mgmODMvAx
z@}+OyrKwAvJ-SX<uDNzFXHxGFi@=aN7VDPETB^weTb9(5hSj|X6h%otNOw4$&ScDx
zBFY*}u_ae(#Rf)|qC{N1LrM?~c#A45#ugN|z-c3M{GlUyFP=qKVAhzVK$5(=NYFt<
z1SX=bm~<9KQCp487h&|n6TUf5LgBe196One(2Q?azjs88HXRXWT%!aL$j<8{!CJ?W
zD;9}S#uCA3m{d%fg0ihxE|#noYr6FsV+(9OL6s#$F_BC~MVib|Xx6JW&PS$`W;5nV
zSu}uxq!h-K6h*#$V&uWS`<$PiftM7BNfS6a+-2Q&tj{mFad?9Z59Prwr;9bK#|thV
zz02zSjB_2>Yd!m3QxHkP5t{;KqbQ+Qt;9EZF)f|%?6SXqjiM?UoMg;)Lyr#uTiQ)#
zqAWg<kdPbZ3ZE#&5z)AU8pJ0%-LaT|dStGV#>Oa3(nfeuUdH3m2g%b^;C;kKOUZ=c
zqUGe?Icoosl_@!0oU`uQCl};DD&qdF9`Czz8}A2B&M*1)-TT}>gO&80oZh>nnZiMB
zIoz4??#Y5=z?LNkyE`lvD}3;qeC(>US=A|O+QAK6I)`3LPUmDY#ocviqbbXpYsW|2
zxN(#Hy<KkHxK7*ltX3<or2}4i;dz`3tX3;ls}+}*mpp#_nB{WC;Nufote8Y>P-F`X
zrDVjKq;}dM2ik%)TAb~b+H5)zAc&M;@Noo3AP9;yMv^HjoW06wid2G+lo4!9=$Ks!
zQj#|fg;LbUP~@grqr_KkR0+zC%t##%33>w^i6Q0`kc3Poj94BGDf1B12!0e#Ao5QR
zuhRD4lu#ifO;M&aNl8IhS_E+`N)Z6A*9^{skEkGpsnUj`sF{@oRb{bCp$TX_Zs-Zl
zkpdjd;rhXhgWV|yyAx(sqhgQ2Qz}RD4j(!hyNau?q)=qP=EP`dE&J7EGT9V&GD^>r
z1GP0JzKZ637ZoU<N-^1$>m%g7XsfzZPmEwkmkFO62|ZF);#U1B1s+sY!(`H6>jG6+
zdOWI7lvRnX3L=qV=twqUO#&YX-I_Ke2#(9s`{=_%OeGD39|rMI#a{WFeEus&Bn!f5
zC1W9FRjYinAf7P=;cliZoXaU5&WoDbyMgG3{IdyT6(n<IPnH{K1Sw-eL_sMCg+PhW
zBZPo5CKq&(U=fc-BS3@X3u?wVPo?C+65^=gLdXKG?Y^_Qum7NuC%)?t2r5I~3X73S
zyC|@`JM7-M&C&B`ytC|3v1k9<F)QHaofmoV=3T!0FMfxnfDeD@1+MQ^?Cf5{UM_L%
zl3CL*ozF;U)@{qR*^K3KnSH}VmEx6jWLNDf6e>quiG*FNNDM<Z2RA6ADT;<@EHQT;
zDuRnywc-PP+v20d^jH+iNcLjzGI}nwAz3LJQ*cy;qB4r)Bi@POuZGQqzwMaVKF=#7
zw~fEhQ!uM|u1N=>jo!w!JcTkz$s!(WDFHzo=YvsDSn*j>k(t&QV_B~{hOP%?vD#p@
z#HAHQfyjsD0`2=g`wv-+5tHB)1%e(s5~XA`bMoK;_wK%pNs*np#)JyxOOE!AdFSD~
zxTuL(Za)7K@B7q`akf}<_v>HfZ~vXIbJllU3@z20aA+!?nUoZT$D1K{tCr>JlI7x(
zesGjk!;KraD65(fBg=Iw1YhsBql0W)qjG0UlDtKwvxp?a^jDl`o;bF}-B#T^c8m(-
z@x>1Ttu?O8YOAiJuvX9rZPB_$Q&XEACMDc?;Um<?&-2AM-{fMkW*Gbv*TU7V`HxDx
zfAhzz)->ge{drAO8LmHG@ZhZD-lKtL3d;f9;EBbWTYD2O+m3bbxabxf9q!StT2a=e
zCrb*Gi>r-UxYay5zoe=ZbyWe8M~@!z;K2i|wM-@rw{G9z+VyK#Q{a5y^74|>80>t;
z{_YO1zWfs2c^0cBkI&C}^zb1G#DLX#T<>#y*p{B;^YyNg(H_~xL2E%%#x-;^vMbVN
zy5NFm@B!z<^({KK^FKhPL>zKHXxdU*%KUYkiOwmHJR^~)6;fd-l+oEYR1iC#U<ev-
zEm0Y~E(o}dc+=<7L{`WI!DF-HKV@&gP4lL<HW)_9^NT^AV<lS_Ws#Bo%sY$TLqFi1
zJWiQlu3%aeG=-(MhFS%53OV^B;MX0qGBTN2c4iHS`xAC1C3UH(jK}7DZH*KcI7f_`
zXPyvVZrYjv2Ly?<MS-nq%Cg+@fz39@bo|&T`$Tc~1W+)(B{uOTF~~StC9-jL+AfIs
z`)x)DQ;IkrB|SnZRDO+O*orUuzNaV(lGSLdHy(=tG2$J(!{AXdqHV;WumsZ3)B1tO
zkM1*{lr-}iYf8Ldqg4sI1adK(3t#|k<bsLbO9M4hXLQa+&U6H$6xL`;YpJV>x-7+#
z#THm?a7t`$QuZqyHwKlF=?y%XZBDfwi%b#rk$vqX+qDqYtFahUY~HlCXe*iW<&`}p
z1V@O2fMv>%lm!!o&#pDC4;cjd!Pl|(RM;u<xO3z$WC%|ygGy$TNvJ6>x?-|-!0vOm
zc=Y&$rYN}f!pjtkCFkc0mg|nA>o>7!pl>6uJ${#42XjK2Kxv87eay}wPHVb;z#7fv
zYQbbuV@v4<RKEDR`UmF%YTTGZO4%fMtDK2Jv}V>8m}V-~wKjB}c-cfZ2(LjU;RBQa
zMht>+_@uyFmS;=a^K%L_VOGIq5aPAUJR74;_AJXC<i;ddI0#}2AVqORP%z<{<c^Z;
zyA~q`h+2Z8vmS($6vhWs$ndd>s3?KkL?dRBAqfAacUWzhPG<x^!t^pS)Y(lwVbLL?
zO{A`;xS^+Sd+gMpO+LR`Zc<Yyw3Sh{^Pc<nAJ8q&IBZ17v0EyZ3SN8dO(KqBvdh_{
z#~dDCXLc~-@P(In`GsfDvy#95|Nei`pFHB-!Sl68Csenuv#TPdOGFbOLVI>eyI2vm
z<@lNBxPJW(Mmc)lvTn~A`i>MGRbe-apSA{VB%f8IF;SsYj*1}>f|3OlgP1fZ2HoRS
zB5Fwh<v<$0(~Q}xC>cQzM6&L4_b0Oc7$A9t>jFbD;EI8&YDlV}t|}~$qGI>x4)g0T
zaOz9WmmO!1&v1S;#vX6;alzmU@8Rm#pS;%DkcXaX3a%83-+dQfJ-%Q0_<OA)|9+O8
zzFUE+sg34nUSoBnDJ@Mg@bGep3X(<;t;8`EDIr0I5W0RVy2|z%<1(p|CZrsuo<z7%
zTFHxcgChj+IWyH-P1kqWvcUO)s;p(L(uV!LT^Vg<(M{WR!sXVAvap00==*_n*Roo!
z<@(Cp94QuZvx&sw30gNA0;>%bVf5F795YrFq@l+JLGVOeMQ)z1AVNwupp2(rb3tp#
zF=PUjlaQ>|&A>@*ulo&(o)L@QIaU!`7a2kt)2lW%8e_qxwUFgi9xJIN9tiTjtp6gK
zlILcuaQzA^iAbd}Qds%Sv&#S4AkV+iuvZ!;R#O>8p*%LmTq8-6#Qsb(Yb?`A!Opy4
zcV1DB9J=T*$)U*EurcS{UO~!Iw)l`iP1=l@wU#mL#zkQ)QnTQ0RKaQ`dx}<D)b8nj
zWsgjo!fiYU<25p_CknQL#N246Of^AkF^bRoNXo^Xcnx}z)hGkOBvjC-WUv`lmGdo#
zZjr5oN;e29+pkH!M-wPaqA-HSsU&-l*0P4j@gk7<*9N&uDHlPo@!V62ET*$mRY}v-
zOs6xP>m`P6y~Y^mI)&>cRV9XTas!%_gImT%Kc@yMh04hzqo7i#30XyJtRSsL^~C3{
zCPAeb8N3ERdK*TJ5%WnM11TANPQQ?=#Qi|)*n30l(v^!NLGlR|a!4(Ps6ZJ*p-Wt{
z=%S=L+~w$f&vWbgG0b<cea~KD*w=@g7ar|;R%fR$4D3I5NM~z`$}+!sgNtrOdv-!X
zQ5ZwJUU2RDH89XepHmUW=a$2)$q`Xk_yyKhIRiBiqsuqEaO6@^qKu48g0=(}q#&HT
z<XTcM293&Ctg=KkFoeMJ!tuLb{3f&6z-%{C8BZY6JA^<TWW4EBHrzC#1WHCuJ|iNd
zG+}J3uXL+QrCf;U+&N?eNh2tfK)-PWO*hajDgO+45p^;eOi`YLv6{LLtXgq!l3}{f
z-GCStr=(Cx7KfA<xR8IBBaE~@2(q9&e0;{)<vDc`XiAMmvww5W-J=`yZ{BA*DNsQ%
zEILXP_|g}DgFo@-{@0j;Ie&cj7kU3jKg@$SU*{{I{pXzi?pOHAqZ8hLu!FtmnH)5n
zKfdJR(FM*ccAt5HofluFucy$R^Jux?yl+V{Vv15Sd_%To!RBa~0&9z$GnwZv(vhHS
zxlx71e0@t7tFX#2BoWmpt?&ppLYaip0p&C#i;m(q7zYS05uG5$(zz*8a=189&S#|Q
z4(lOtv23|`d;#GqPCGUMYE#suCx%H|lh2V3k#9;_@bkwTBsf1#O5+VH*PDFL_#oe_
zfK%V+^H+}_oXGo!=~yo=C_+b3RZI&_F>lyuDh~HN;hQZ+MTwaz6sM~J8=*0pS{qi2
zC1ulKw85wBkEmrNsFlQf#VANX3Xz>jMceoEgQNEY2`MtbF!Vj|zI%^*_wLg)4g0%$
z9N)ObPPR|cTC-R#Hf{m)ojLo52RP?gcP-21iu3bx`o0%lR31GU=OH(WRb2~;h}`cZ
zhia?p&keEB6lK8>CG{pE4Rb;eFWH3z3D;UuA4El}C^eK?QKBdnm@Ei@B$0(|(e+KH
zEqO$muj8(FTn?Uf0vDgW>PE0(+$`NN<aT&_6>DV5D_ODRyjYcmFZt^zf^a7!eZUKv
zuYd-zHJKN%-&h*0u|!Prs1&jgME=%ij^|8kDOQ@Mq_jYC!nunBMun{0jf5yE2eR}E
z2P;O6AMekMVv6E&YhN;=c~^butd)jVrCj)3N&6V5Mw??ZiO7b<F~d-aY<Utl;+pYy
zY<S<9hc-IGY{drIK1>S-87C&wp%9;29W6;Ew2{a$tF)-Gm1f$Mw87(jfZ&K>K;tP4
zG?nFY<<VGieK0!jZ^D}&{pMt}IgW?(`-~GYuvU{&K{KhCb`wI_$&Uv_PrWEuEeqBQ
z#cCad$`nH`7NejlDgok(p%HIcje#tX(AuDFu_@w=wG_6bC<J{pqfLrZ6h(<rS~7sO
zG(?c6d?satRQBf({HAdIgDLE;98dUs<#~2eiKJ6Tp5$hq?e1Zf;?~hMP)aiLlA|mX
z#ZJS;qXns7^Ww9&INF`k*P3QF;qc}SeBbd-yXMZF+b9LzdFJzZcI!_9ho&pQMt)(&
zW6L#os&hf*k*0iZB-*3J9I#z^f*Zhzy3+=*Wra2nwZXZ>$;lb_@1OGi4_pH!90&rr
z`}P7;k79;A5*)2q&`9)@8b`ARl7^&E$>zEvp^|izr0rPO5FzB(H+QsAMUqb0z*5pa
zM1Yb#p_1G{pfjvrVw{r*tspnGHe$GJv_@G$>Wx;p7!J53NP;|R`8Oem!O6)9!_ZUL
zh44R;<@okZ%nPsb?YAFL&u7G;VnA{G<}Lc^gy=n0U9i7@L|Ih4^vrX-aD0Qm_1EsR
zeDDt6e&-&=ts6K4r|&%E^zn*=7hmNwfAKHl${80API-9uHQrdB6B<p0$EHM6luXJ>
zlxamlV+wSVOxnSMk4cnSicENRC==z}r$pmHcb;<aRF#;~Dy_s=m55i;p;x9L21VaR
zf&(|&<&54+h)NQb1wAR!w@X~RqH8^u=Vz=}ZN4<qcC?^UMsww*j-%)Uwncp2wXUjZ
zQ-~Rn9=qbPBOmYAJYi5*>?Qt{b=mhT`u_bs{Ls>+4&P@~Oc^YinbLgh)myy#c){sq
z%gM5*+L^I(iPaETbe^52Vb%BaA&5?0XKqm;O1Yx4IW9<q0e#=0fT~m!rOGIjaBY^W
zz~F^HX4Hav?>yv-U->45g`J%l*N?CB(#tP%{rVBM5F^;dYK78@9q;LffosP{S8^&H
zZQF_;Sxo4ZC{d`8Hs-R!FyNhx*2WRv7(TvUcdXlv)vBdW37=mtgG3D~Bj{Wl;?^1j
z*|V&*l8>r2l~(AOlY+>$Cu-ZUD9EJsNQ?;|k`UBwiBXy`4B}D8_WM=#B=x~lT|uCv
zm?IZ+!Y33?pM|2vOm=8pq;ETX-{ZP9<f635&%Xh7G|VmRmWHF+FfTQY74b)7VNw><
zg{3YH*Z22HZpc}>YhpJ5G8VKZZJJ6k%^m-W?0JUGwk;ViQ<fro2o_~URaL-t8ueX>
zoh^ivv*EmLy-}}vrlve9SBNzZnoY)O2!RxY@0b)QqcO>D{1Z2Mv*S2BZu3x6O5)jA
z*XX*0%HWg(vlSDF!-Cz!&XEF>vSe?junDk=p<OU^E2^-=q^Wpxx<H!*%3w^5+C;`>
zp@fpxP$|S9StZn1C}lLy7>TA+s=@mms;XpXXNSGLIs3bFE-%knT%L1w@`#5IHOu9K
zt`(t@)V&d<qKrxNm8ZsQS8KE~S>-N;T~SmNwxld;%4&|ag_s)~D|@jvBof|R0T?uy
z4xuqYi7GdZ2F8-EC4cDp2U^E|UmUx5m6*A~SdrpRE0WfjvXJzM*$m$g6jozhKoL;M
zk$lJAPQ$gs9kdQKwIRiU=z0tZ=Q<Ae_qjZMKwZnl9b)!^VGBu*7yQ%yKASEy7g}n(
zcynF4wJlMUb%|n1jGiuuvQ0&g#-WrU1Vyxps3NOX3n-?O88Ja&N)$Qv#(UwHY-H$}
z|6s)1CMhKmWhByQ5IS;Sl0hc}D3WqGl{fupGBj^C#jI@dQeZ_D#6vG5DN%XGH02m0
zxPXsd7E+@rO3A!UQ3^$)bw;ejPbcY+Ny+(>D*ruN6NM6kwVJcDGnUIslnPX}7*3{W
zDCbkI-?_!NtCH#dKJ6uhZlJD8?!53ku^-S?NyJdsGgMjg@`paghd=che(mr6x0n!l
z{lNovFKf>3o}hHes~`Shnh$)KN9T^6XFkF+Kl6+HP=C(J!@IN(?lD|k;MZ&V%LPNb
z#t#E+2#5u9AjKZ7JhHw9QW&slyTIs_u+eVpDpHIRPoxqlcytux(uaWZFm%v%;0KNO
zNz}UuNeIviZ8K*m^L%2rMh`38vSac1G5xyR6ma8aB#J?CZy#0tQOO&OoDa3}zmrQn
zE4MdRE<AA`j9fvC!V@<frTBgLg-^|a|ARW_^EtCZ3S{xT>R5N4%cY~)n{za6sEd-y
zu6cCXQx+xrbwk=KS%%0(SFju$Yv<^CN9YHV_abC6cEfekTGKR5CaueQ=6!ai7ciMp
z{J4oxxb>w05-cv)ELU&y_B)T5@6=4^Q>Hs}=JPrG2L}{I!L*ujcz7th*_80!vs$be
z`XPHRB!+e^W5?_j8NAOMS#Bb=LFc^B5F(eWHRa-xt`n5e2hX^D0uU3ENr{dT6=ba3
zXR|!-1CGd`g<zsd;Jp-uQFiB~6fQ<7+(KgTiIq~Ulfo%YDz+K2<AqxmC1YGw2%aJv
z%L}Jb#2Gf*7TF9e1Z8N0=KiA#hNwxI_ht~&;U>RLfVqaFsbRmd985}XAMCI<t(caU
zx=@r#Ont|{DXcCbd$E-x$dRPc79UNQz4g?_0AG&0jz>0Qe4Rmus;H>TipmzmC>+YG
zx8??6Rb<~Zc|YX`AH{r|-xNoOHkJRsjkL-?z{nXhN{}g&CqP@JtrSfu1sQ*)KvIdx
zY(g`gig=<l_~c1eiu(c?Ir`{HD121-&=F0EGEgdwC(vG;)1Ob7Oea*69lFJ1R52l`
zC?@tH=25BS(5lg~A_^@`)XigoFG3Hcw-zyVU1pJtp|G0L80xyFt{YN>%VkU32T~t#
zsom^%!Zp1*Z6g<%<dhHLa2qSdxGf5bs-&z7in7X-FHx7Ld|r%Ipd;FVwF+av2Z+&U
zphiG~5KIPEez4>Bh1gvsdXGjn0z4@h6(Xd<$)uqu3M#E}(<x>h7<$iQ)e+Df-?&LN
z+hOoM2?=PwJUgQ-G~IelY)UTA&$)JZjiM;%y?8k}?}aO24BxAu8=u>lr$^W-Rr6Vq
zjw1%5PE=(<QB^dv8KgwtwSw#gPw0EF3Lg^af*3@3Hl{&TWrbBb^I6!sv1eaBt#wux
zWiE@-n@w1&07?j<LVg95qB8?Sl7#>(+=-Oqd2|v&sL=`&g(Ejc&ZN!FJGvoc3j?wk
zL`$q#NaD^;NHRvTBGo1(f>JpvPfONa6hEebK^GMyMc?<FpPy$BFNG~M#zIJ%YChxE
z`(EV>dwYE2&AaSRk4f6_@Zlrwo}KZtfBFL`Wzj}qRKd`<6ivxTKK&E?!@v1Aq|<ep
zM~@y;N5u;-y@a3caH$*qum9cO=h54jyziM8x%JF3$9E1ne&*v``<QHMrAjm@pq3Y`
z7iYAq3*2gfx>%sPfzY<B7Z>#F1u^s_-{Rex6nc84(I$}!%U}|fHJBLCL&PM>-gGV!
zU8L_7A(r$$5j^>Z7aN+QuvB#g*5dmC=Ld%MIca^yaK2#q_>{i)n^6#CLc4;>m^`8q
z(sodZG5D+~EM-|?i-P%V%4{|hf4{!R5B>If&ewU!=uON)-YP+eIqrrnlJ~^DedWLZ
zp&eq&AjQ7c38|k0Zg8+zldO=<=QcrOgo~*X5l=~^nKvw4#pC4~oQQdZCIUV>FeCTV
zJdyoqkadcbHchCI*`-<VN+zRXwGi=PpmPJQUs0}?)Qd}+$&|y@lB$rg&$MosG!qJA
z#f;V`rdsF<S}QtR3TG;#F5bB<|2E0zN6$P{z<x7j_uvTcr3m#wxN<Q+U+<g*rl&v(
zp6DF}C*p#PT8Vm7${4Y-P>laeN@1h$)nWu2Wnz&A2QFYW^m;(YAg1wJQP+)}Gb0oO
ztA$rAxuvq{jwbHT2OL3CCsM*WMIR$7MGS&q7D&QGfuM9ZW^jFO*`E~bRhFZvVW+V)
zWkG4Q#9?K}0HUW49+HPK-6*Avh$BQXsr7Mej16)0(Ig`>F$ij}Ew<2<Ry+U=VkT^i
z6(lYuxhC^9YjQHe2y~>;SZ>>wKZ(C50Ts5&nrsP`ImhEUiAhi}g`%()trfxhu}_p=
zpk+#g&QvG@DT$|}P7%D$yvETVP7p$^WORv<?j@y29C{IsuGYlM3rbs{+)C<;P(Ug-
zuT@Ip(BI^JSPFiVj3AOx#;8}F542V^O+#4}1m`Gh-tR)f4Uw{5Vu~qAt#GlU9aKi_
z6r+epgcB+^Xqkf-gGA}YfJ!2$iR6*?s7}eRQPSBPQsr~7c@x}5=c$;=Q<WlF0u~p4
za5MY&CZ5<RK}M^Nr^cNr=cpp%C#<2WC(Nf4T)SpwE#+j2u?1y0!PtuK@{HZRIiWH*
zr6@~7VZ=A;!MpFGlLXDj5UA@iX9#KvRmzwwZIQYsp6gQ|%ZSnDuFz#E`Bp@bQos)Z
zYm6)w%8(-T>orP6$w-ZeB}fo&9?-Piz+~DGeTP<wrmj#Xt6WFKM9GLn<qHdK(1-#x
zB~MJAET7oqA}SP)mXHEbSt4emCL>5kTxdnmBA8Iw0NJR5l(Ou?%O<W<Mhj9T<+wE!
z1Igu3X_e7cTgZ)_A^vDGN*H`D03?DfdPiB+TwGkxwku-v)U|}6YXd20qSj3I_SiW-
zW*vHd=%b%t+4U%`dHuC-6I~C!Koup*S_(ZQ_?{atyv(f^Ugovm`Yfh0q`v3n+qY4Z
zhIdX*xg9i~nv>I0+DC8jJD-1zmY!<faIm*WRTa!<6OQ(0%%+y(<6ZU-XG{<GnctW)
zDQjw5vZoZKN@yIZU$R=9uwGse&L6WrKjCqE&e}#Meqax<%3y*5pU|lwhCql4A1s3x
z8ysC2LdpxRHJC!wkFLw)^<l-uyRWl6>*y}eq>w{4Yoo@|hzK&uvZO9b=JPqb2m9Q(
zbBkx5eU{_nV~(y}<Hn5}+_`gyoxQy*kZ~**7pzt*x~^lLEisnMB^MVLa<00Llk-zv
zd*gLpfBj7^E-vssjXtFt1W#*42>*dboGC%eqXK0WHW`W{VQfpc>ItbsX-!>Nrn4Fs
z5<WPFp+zUn&ScIcD6tFaJA8_Gz$a1Oqb(qm4Bv@or^=gTX^Skh5kE{R;3TBp*lbP_
z8GBNpr-aC`Zn;>u#Ppcc$ER2=VxguinNQ|StD0%mFso<mG!sl!N!pAlNQI~yb&M$M
zuu0bL7$WU3;M$IuBynYbZ;#n*2dyQ4d$eB&Ij6hrIw7nR#a6;Qyz2;IU<|P@5Q>LZ
zqNtQns$zsd%4D`^5&{s|#gcNhrVF0c(26AliDVQ-Q;T9W-*YM<8N9v2_l^)Dg*>t7
zJb?j+#!-Nokdi~8kQF&nlE<(Hb}P7ZpgG<t*{w^aR<T<pCMHr4u~89&B6=xALx}h|
z7V=`TF?y(u4O9pUA5-4P#3pGJItbS>P$-49in=ryqbL-Z1b8T$0-ePFt?vejyeex7
zTVhSI;g^%a7Io!X+2B37Ad4gSPGz7b3YScwvMH|c?_!YYSXGp?M&r6(Xg)?!7%e92
zMv&RT_gHVFs*1u9)5?eee9$B#MY+V($#YL+ODgXPeWI)^x?2-3&M77hWm*%_7}aC6
z2TCF%dM2>8aL|QPU_h>C^0@nov9H#ex~?~>XKO6Z1-jB>?SQcjN!Pek(Rs_NQ=98i
z+!nG3NRMJ15jiNO@CmR3K{k4nF_`GE(V?T335?3(%0x!$BR)Aoa^mK}w%{JG39Ai-
zG8EwtsIVK4C#@i^AZ1BTCP@^XDjV)qMFq}FIz-j5cXTXWUEk3!7c_+?hK?8=lWC1g
zp7rB*xI8~)Z`z2NAyLM#K5R(tO6zR#vE>|G^@e-m*o<7Y!2c=bZbs#Z6_HF2KFT6d
z6x0*pf<@P(N%)}ytq2~Pra~J{KXf#ck`Q_{k)|$1?j7VLg%q)Ro2RI9l0^zg?2n4X
z;Bc-dc*&DX5t2)o2qsz31u01`tua=LGVh2g5lT-mWVKL4#Q~!v->xv07`==)#_?MM
zZ4DtNaoctQr47j>dEaDzDy?OFloJ_JCfUXmAv(G)adC0Z;6&bD*EMBX2?y8~=)z*l
zid)Y;&)W~);XCiXi#7$zYRz;uWf(fj$wWqpN|Cf7sX()T$VWf@W4!)_-=b^^qE$Rz
ztxzW?sA|UQ{G7dN%8qWZgXMC$Koup)m#jO*xnJ`R7ZeqgP2!jSH-CZEu;jA${Q4W;
z=8Iqa3bn0i%7){kBM$d>xPG+D;o)@-Zoka_D+RYFh7d1tr*HA_cYcfh;X^ErICzqe
zq^JnN;Ju=E&^bkn8k?k;H-!=JB_o}wB++?Nw`6huE!<iW+BGEcgen!AhT{18bw2Wu
z5A%sne1ac({|9*H*=IO7I$|=P(o81Utg1>Wq0E(vY$V@lMRbmk(V~q5eH32LVzuDn
z@`7)D`&)eWS3k?=zwiaV`HgR~S}hp{Cx|pc+!bFel?l`TVc>Yi7ATWn%P55Yx|wkM
zR>N|+WW8!}&f|xHFbt?9g+QU8t{UDwe}GaYH>Y#<>O=0IKIXi2bPmc&$lozVT)U=i
z9YtZp=So{*2n>Eini8|BLKPNIqzj&bhz*h-8PeA8ZEODltMfIZZleIs%wq~|D2$;n
zC6y_#g&4GV!BI{c8ACNSv-zCqbVgZK*s5kenNm+3ZQIHCHALLylHep~bQqjW17fza
zaE^m(N0`EZE}$|LRYj$uY={=hQk1yf4cvX}&0N^!7e5I&AQOgM?4?LKopWuzgO8pe
z1zhmB80dqb(SviW`_;y-%DW)>utN~<xik(sB}+V^#V0R#CPg92)+qdB3tVsDc&d2r
z&~SUF<ZxOsFAEy2m>7wG(ul8J@(J%`Jm`WW;;%$O2^22eHbo-G(UJ4LC2yWU1fx(&
zY_hB|l-5Wjnby=*MOl<sZ8PUtK!VB^*h);TN5$;8|BXk=|9N^l%p{NezIx)*8{=V>
zFd3-})?$sJ&<3w$IwYNj#s^1;lF>W39y3Uea^X!*%qVg|ikA14#@G^DmPDX;0q3O%
zR!O1y4px?E48x?MQh`)N3LCLTl-)YzWD9{yHk2YA^9XiSu_m!p31h)D-sf7Qv%1`v
z3T+C|78f+D*0F3I%QnzDwPmI#5#WGO2*y}SB~Jlu6fSy-VPL2{mG>KJL0P(-j8$&V
z&&W^oe!zQ2j6MrCwkAF(K{>aR#vknXeR1qY#BOx>AIFv|CX5p4cnXp!XA-#eii^c1
z{c6e4!9Ip4Y89eL@2z3*1EF6tv@0e}MQJtb^@{mymgB-Cw#0Ab&UwSRqUd_s^mlXQ
z)=41MEy2hr>00~2<6|P&C>f(=g{>-D*D~4!s8|p^RP_W~8oHqcrYIt1p_oi-@fV5#
zZ8Rz6{5K=E05Rj^CgnvaWU&PW#G2p?F<7EAL|2khK`=EzRdjirpaGQ}Rwumg@MfU2
znnFh`fzm|cAbvqsfyo74u4PGpDGIEQn3VnTd_olh1FO7HZ+7vF<tIhH=oE|PlFQ38
zQVP`dBxkaYsx3=Vlq79<{*_nw!k2%CZ@u{@2Zu*ArRIGfd_Ugz8-51LGRFT96`%a{
zXZT0|&;JX?Bqnvmoom;5aQB>(v#;^eYu{q8(5QYz=nIUN@N1(cQdD}$2kyI$pZn>b
zW2f2WUw-N9C<8A)`#$c!b<S75_;rjb`OeoKWNwTnrJkxTX(kOvH}|-HtLB&flRr*<
zbe{*`e1qo-iG}ebd`$SD@IlcJioq+iwp%M3t+Pj%e7}ogPOC#t8G8<AyByAERK*3?
zt{wBCk9>q5{^Te4<R^cGkA3W8>>eFpjg`(Hgm9I}>cLF38c${>@5Uh4TgX0R))rVR
z=#wlYLXxj%Q;x1(=h<hT<EMY}XK>E(+H0@zo4@&+{Kjv5jxT)SUvl^DcZIuo<=(j>
zx|Lh$_p@;PJA9O7O=-rz-DMt++So}<r!z659$$Zidt;12l~WF#=DZzvbh#vSD{kym
z>`pD0z2p9wr*#et*a8|Wi~G<;+UQUTnyR2GO35P*k#*Y=twuF9B#-tYd?+HD2rM$c
zC}hL-s1rU2Z!aXp5F%X&I0oSyWvhk|BgLh7w^=Kqs>yUp-AJ)llqHa`rl2TFp<@^^
zf73}*nUd008!C)XQ3M*<KUQZx9hrEpwRixUyeZzf^SnIwvFVJ77_MZz<|VpaFIjbK
zoFC}AZd1$+gTz7k;5WwL<GvsT<C6j*jm9F+)sU#tfbtzCc`T6{xIKsGjwamNYq+yN
z;h<9Nl$Kg6ij=TY5#pmK_=Fo0ZU_?fmOKeh2u@6JL)c6|u9^lv4Lpn=i-Uuuu4}5Q
zqG=kmHd{ktuBcKoZ+k52#;<Q44{=wJ%Oh<@e^=|f?*c!@V{@YmGF4bhC<@#d*{Nl6
z5i%+n$9-2Tn%RV=syEj}^ddeR)wN|=%5<m@PdArDMH@nPXb6g->xl_mO!&H%d+mzI
zhh(oj)sP#3w7r&)@wm!`oZ7s`+jVQ>uQ&2rL&SN<dcEf2@{+}3LAw@o_{y@EJy5q6
zE91v#x!#@gm_k$Z17*>%ZhO{sOHou=Kv77JeSS=#&bTYZTa^2Slo8j6k3G%}^uzG4
zTDqP%et!nVJav(#I5Pjx<BZ-%;^3(gkP@XzT%6vgJ$uNV8`n|6F}RMD20|E!DX=rE
zSavN|E1J4y@pvV1O(9TJ1uA>0p%9yixQ$hT?Wp*>jw_7u4fLGE<49#!^%#U$s*Nmm
zWnB?MMRWs32V8Q(i!eg8Zrck|j;N_i!=w?@+>pnk*2)dI@c|k~vdQ6dL{<oz5G;y0
z(a-4ChTavNwI!E5Ec(cK7rE>_y$=|jPy{TI{rQAxHKnwX-Kk=44kn&6DHBZvh8~qX
z*66Holj5+bN<s88!8?Y49|o*0F-m;)thKn{b9_)n8)F~@&th>-*SA;l3rC)Zh9nno
zOmO|iO@8j@f1Y>WIHm1-KK$vI>4uiBZ<!^BN`)w}Vph-fJ=dOnju&5ek%wRZBB_E(
zCw};Y&-0zLj(_mC{y*FsW}Kcl>S~X^(DcR<tA<V`906mpoq$RF&R4$3tvffkxH#o&
zU-~>h{K5C}`tN?5cI8mXf-<-efdZ0dz0%w}IfHLJ<|jV>af)-tyYD{c_U&VmwhT(q
zXE}fH7UvTo8p$A53Z-)wm>U)!gbUhtEv{Q*))&-!Gk)%8ewttW%kSqiKl}6Ce(pJD
zvzZXp72+>w#L#&o+0To0Vyko&zr#$2FiN7%l%bx?HiYe~lmM(j$AC|cs-92*!8v~T
zV;|=uANd&ntN-*b^41$~^7+qyfq(X^pXJwn?bmtm;67xtSJ^CpZB6s<-C=D(VMoGy
z9tXy4f=p3Zk{wMx)Mn!@ijtz9GD((M>!?dZp%W|TSi6K#8aFjux<DT#$wIbHK}Ap!
z6z*I=Aq}BQX;UPb%$HYLNuvRi#%Y6G8BrySv4X~{Kp(Xv>7-E|FG_ipf6nCfU>FAa
z0g~=nb)6s!rKtFnf}#*qsHq#VX~?U)(MGt%Ic_Xwx(4I5no;7>wgDn~$mmg~9*6~u
z6m`-t39oPDfyNl=x)puDmd&jlHiba+UM3^sIv(Rj?V0uaa=qs1j^uU<7f+>#4lyG-
zk;cGLqj~oFoZAOeuFq@sW+gj?sACHvCS?E05G2gq2b_<%5Oc9F)=}SUKf22K9*exO
z7#p)e$G5VS*IQJuYO{BYWmp>Td%kzdME*tJeQ|%koAwFpV~g^W(#GU{d~O>v{d}$!
zY`CJEBGzcx^H3xmH@T@gM~h)}_H$E`U3^u=6(cV5H$wuwS`mwq<Q>XM^jB1tl+0$t
zdexz0w1OFJqSAQ&$D%fa6k{~x-~{z;`;JxHvRExyESI!h#}NDm*WA8DR~v`XnoNP{
zO$affqr&;fFgW_b(RG9PP-eqoou`H%HZ5{*jwl`ST=Op&3%lSQgZsg(PT==l*gZLl
z9hE5~CsW+Vl^E#yhz?N1NQw@eXL0g?Qbi%f4r@>WTUiL2ZndN?#1*=4d*-w0=5>|W
z4!Jl|R$%F<bB87i!e~w`Cn1frf=tpJ`2_ieOlgBGswAHe&XJPH_)Ssb{D4u4;0Lr(
zXcO7nopJxcNnU&%wN2DwT$CSAZLin-1Y}x05Z4!|WbvV3a0Nb0i0pH*sJQ#?iic+l
z&Z;R33{F)PlRdf$TJ14LK^zD}$Lu1Z)=Tg$^NHeMmbiIPb2K;XHXZXC3Iooyl-5fA
zZH@pFUB5zmMe-YXZhLc#1>KcJB1TWwty!;^gy1Pl@$J&uVss(IY*k7AY*`U)&x^0V
zkK-f9vR(7yt2cP(?wb^4nMt+*o7<NJM6D<$H6Q%=$9V9qud!UO=+_<dy@ET36)%6}
zN4d`df9<dRE$%(Kpf#{}=O#gCu0@eIyXbh_p7P2wcQ}6bI?t9jvDLST$wLjU^N=(l
zNatw_%McT(SJ<k?nmJ$p`Wt-gV9ILU;$xuqk={e+G^sZXUeLObK>J7vIq9OPps49}
zMIRLxR4j`f%9}g9@R5(R|I!C(p8Wv!@D8ykd2sI?-gW~;U4c<lbxkvEK#RJnEGjBn
z=8Yhi17pz!8*2(<i+t{cRHK#IjAKQ$o{JNEC4#TWxd1A%w|mI$hxd8))%Wv@zxZeP
z#y7ssFaPo{^FRELe~&j_e~tCJl_<6GTKN4f9{)}r)*6aJl%Pt<M&^crzVEMW&fB6+
zXPb$-s+de0rWoc;#on}}DHIRS&beqKmWbWgG%L91A{TvPh+wTep3#~<ixOf~1TTeG
zhzjLF_n9k{MxJ#p&awy#WE(TO$VzTaiiXh$RwX4PLX{Clt*Nw#HG=oJAfwbiZysau
z08)4qrKPTF>axz8x1lsrfNXuvG_&cBkoY$!u_xsyrxdb<gOZ8BYPBMylpA5OHgThv
zBMM(F#z2e%DM;*->*I!jXRf^I{ASq}brSLgW8|eNtO$mJM~8sUQ;AB!Zl!tVu;!)P
z*SR*U*r^P23sVcmJCqj`$om05NOYhd&EG>HWC2AU`sM57sl73C+Qzr>b+dUSINFhv
z+1jM4Dk*sL_;sU%=eujPr&Em7)z6Q(%lN!DuV60X^7meyay{{QBk%f(u`UQlOhq<D
zmCCVNV-u(|%?2%RQt(OolX%a;n0Gv)YAHsBzK6jPLPD?Bgqg}Y@`EHhnIfNc*=wc1
z%gscJL^Vc5vS~8LPGbC9LyQd0)Aa+ZcFl6N;&QRz{NjSeV!^r}a91L{#`m5cUXRGI
zJSY+~wy{TX7(JkMFJ8Y|C>P_pp)!4D)A;0OM`sS7(g{Vx$AK7yV(^0=@7bVuGlNf~
zqqEr{K`|*}<}aiaAUbsPOv;KfB(yG3$YP>_(g&JSL<D6$ArPR*DFNP7Sp$P`8LTow
zM2!(sDuN0GBh^1zVPe80O^Qj97Yun^70@YW%M&>*A$U=jX8)s*0#+%MF+%tZ12K6t
z1w~~^WyLV`_~7N&rG(XvohiK02i9H1e5#mfm=~Hrz$)<+3Q6Kkf)<60_lnRDSQQw0
zO^Q2o?ud)^KHqw$=V1qZaf9VF;&vjl!(9$%d;HW-{RHp4eZjB%%0EMC#ZUkAPx6&7
ze~piR_+=ix`;f1F=^LDMfd`8v-#NMDS{?Z43)i^0lPJqmnwiIhNSPu@MVyYL$_SDd
zEy{ZwqNr5);!8<#TD4P|xSdJW>or5)3%|%n7d#@gMp<l8ijvDE25X5SGM!Cml>$?6
zFxe5}wg#d{`G^h*U06c06w&hBkNznA-~aC!&U*%LnV1UGuNnH57eDp^uHAfwohi5!
z`H%mX{|lUQ{PyR+z_ZUh$Hm1NZs^$Gt2x-45rd~JCp>)ckmsLyg-KI!cCsXR(5l1@
z0i#Rt;<JusV(DDZqx%n-KJy#{J<d3K=NVRt_0oc`=>j+<8?H5C4S*>b%9;ns^3H|h
z!oh``VxRv6dbZ22t$X?}ejEL}Z_o#i!_WsY`7}mid#p9oRVBu=^Evx_*Eu>o=GwLE
z93CBV{rYv9y2ch|7Aqm?DoK-qa6MwaKgsL5ETrQH<5IV2dBl1lCX&+FvgYV`!hCm^
zPk!=~{Dr^p7x*Xt(?90#{+++eSHApZma8>_8Rud=(;K!~hTDzf4_o<HnvyzaPe&W^
zw!jx9O;yu%9c{Z}7-S3>W5gIkRn?i!QDH)4sx`Z_io?lu9_-YdE>=8RtT<^M^Tvtq
zT_1RK+0g|h&4QOj+v<YSmJ~%rN(t`*+DB~fu)@Y7mvzF6H7dIhWL-scl35wUsI#EZ
zU@(-K3^ek6*Ig?r^0H*26d@(VGFOnY-VHbxaUsyRfmPe1xd6uYg8Xc@N28~hoc$`7
z*_g5rGmS%jS}7?QJvpm131v&xpe=6@vTak^ZskROt9ZJQlaI&e6Ac6f(L&M*uLcGm
zNJGQ~u`}5RZtN;<9yPrD+#R03cF4p)kpiV3C_*H;K=h9228O;DmjExBo}>Md_|nOf
zHd=3TTj6QMj$G7<p)l}7qC|{wGXWVXLm_)qno{_f?h1eJNk7@|JzBlWLBI0Z&3mI&
z+0#YnhKjMNk+x&yTt&%skWt74TwGo<G)?A{*(})5n_@j8$SV1qgN%B~;ke!hT-%e7
zoNcpeNlhc*Q1BwS(Lu(QIR|#b6B<#v6p4s)gfiRbKF&N+0v`f{_pJMl^NR}}ot$w0
z;Un&!JmP${6z2^Dg-chKu=Lcwnx|H|1cb4O9>3!9>nwIGo7Y`!pYvAelASViwyn~m
z!1KG^3;tEddq(Wil?&qO-XW#2b4Z)-3t93jd2xWzC?k1(L1{YgDQqsKjn2sj0+VWM
zH&G=<Ylw+{-I98w1FLFra+a%WlZc>{p-@`j`~W(`q;V`tb0JDXs{1igM;lB^I45v_
zi~+4Gs-h%R6S^qG<mdx#=&9?Hs;)UdzhpY8(8>{A5SB1XI4=f)E~AHJ(9VEw2*Z^1
zYC<1w^V+)w?_Sodi({5GT%;8*f8=HEym*Zde)xkdmpwxdFT8q--};vgw{G3w<3IW_
zCc6zcZ|<;vyWyGl-{j%D3ts=)YotZX;^YCp{o0B<dlO!`^8$@8IH=debVgl+P9?-C
zB!yBdbd`yxYKz!Gk^~21#2O`lcMj*=6W!<4Zgb316;oz|GLi;pii+KXBi6$}Sv4%%
zmNHJzN~{qK<z|FomFC8aFVgJp(w(1D6h@HJ7+5VYnbb8qd;2_EE}7O9^XZHa{_v~(
z>>vAyOdt;I?d@^@!F^61JtX;(Z++t~ANrvWF`Mr4(u>a#Jv=^LVy$Hu7K9kErDFeJ
zkF(25d`P_W?ghzCiB{uMkM|wPMM_GdQiSB_QXpFJR?!!R%VEXW-@Ih~wkGY~gk}db
z-NUa3R*Oe?rSQqq_bVx@j1?2sjX#tu_*GF-mIcepB@gbMqV;#g<hCf--QD5n=o+_f
z-QxQ7>+J3AvAe&|WH!r1T@c%ve22z_C<7s%kE?fZwu)6^?;+>aqBLUH5&|#3{4&oz
zbB90m3%|fW{>T4_zx}uV7GL<{SLg=G?KEm!q^}lse>jh;*Ynk=CMn`HMPYE6-_`d$
z39))`uBTgHvfE5BDiQhxl`VPh+CImJ`#f$tzWw$C9xm2Q%gD}DGo8#>_kr~gSq+ia
zMbbv?CunkEN(_NAN-5<TMHAUNBMk(^4kf67)fs`!dDTW6Modpeuu@z^h4^h|k1}ly
zS}6*h&1towX~Z*hY~Dlk*%HMQ@(4|788eHiwxw?ep(ZFJ#V&E&7*L~FGb0rtBM{>5
zq&8eEF~`+n_K(J7|1uQ<I)I5_6BJPjysL;^WyHqGrzpHBBV%tzU`UD~C5j|P;UVzR
z=SrTxea!WPJ@%(lY64|Q6e(i7#7m{Y5xpZg$Iy#^Ur3_vj4|T!*f@m|O-+Kfd>;ht
z>f4wYF$y1sVc1ZK#&Mz6hQel%L|M=d{bn3J7FFY@bR(X)QpAm@{nN+ha~Z`I*9*Pj
zJC0vZX)6LT#>h35$FY)&o@7E`G*wemHMQi6YXM`lF`LHHIVbPuqzzM8tTGS+DhN%)
zB=9kkoTp!}DfOYKOhIgO^7S7x@++1+YUBTS_3zG4jZLkRAjFUr=Yif0T&|X!o;~K#
z=_AfA&gq?uffHli@uvLR9AhK#bgiZG_ixAM+u{!prS$mqv~7^Kg<>MA?2;g-gOus+
z-{Ta1-{bc!o*18zj5o&s-TXN%FMc%OjeI<$M9jrPs!BpekhD(NLZix(7<ya?L=D75
z2m;Ug<v`o_vgtH6)7czbRg&OrEm0>?h^D-hc`ud%!R2@*CHz07<&5QvQ53Qsr4@#P
zkOHn3%K?L;(1yAwaJ|Llf=|Uj)6ALAXPliam>lj=O{RjF<i<3ph_;CsdcxpH-h!|2
z-JH(f;_+q4H}3`RbtMZk!|v>(_Y9x@_$PVe?eB2o&a<4JuDCp3@b)_ou|~64Typ;b
z{Effyx7pv{qbObiwZqLj2VB1eFTejqOf-D<pZrrEz41D~yS(J&ZsJ3?US#Or<i@;%
zVMP%G#u-dmk(9%$g-FPATOeLnf;bv$vuaFkOz(XjRc#lQ4KY}6JJqqHOXwW2qR@q+
zNSa|08JuHU)WS0|V+4!{O!lrFbM4k0F5h}xOetdr(J?F+%<GEDd`7Tqmi@qQ{MNtV
zr5}2hdScnxZCJ0DL^{0h`Rdoc!6!ccqo`D~v%8CPJwNkvKf#~;g+Iw^)${1#BR>0U
zzlK(dANrvWP!|n<^KbusmYw0@Ss<DnirJh#b~xV(hctM+cVIPYQjn^Wg^t`$9joPv
zpfqM@#vDy!O4@bDdf5||#V1igMK3C9V@J8ZyiF#hsBWVoM1yl+3nhHdvZQGm>bfC>
zz?*Nr$y;x|B@2Ef=<>@izrw-(0egG<6h(yuQ;T(v8qdqq_wjf$t8rv2Vhm%fWF5zW
zqO7>{?6dr(zx<!^<3Il6{DXh^kN6w^?cd@X-~5&=ZZSTwKz;YH=nu=WC9-cfHW>*j
z#fCg?tQHlV>%~*k4Xl@ri2_@e)JCjsqDm;M*`GGNaO(#3*(ImTmdA^h4o5ku*qs$b
zRkC)G#j>OA2D-svm0>!aU_(T811cI(g=S>U5i>zN3l4}9T_rH8C=#Qn0yUtccwZ6}
zQ7NWl!byzTs95Zblo8KOYYj$kk=K-`A1JZN5&o5ok&V`3gMuK(W9*VjYRhfjjIgEn
z5k9l*ywd#3^UBqq7=05dbaqzDyJ)rpqf}6tY(9*Zm@s2;-Z6xP4r!Af7;?ci(J-sv
zpin&D6ufeCkK5Pw*qhBLl_IQLRFvd~<O9ARh(k~C1HlVtDr`K;;zsC^CyEqT3ya(s
zTov<;qtdak8xd2jbyi6F&CiZ+*2?#kWhn)?-*T<4uEAHY*{jcCGltwyGM+~IuApvX
zme`eIaMP|`DM+s>I~9m{K#50ORo7UPofUi#A?27wEKiDr<gXfIHdLGuz$vt3Eqfp6
z+m<?I<SPYK^pJV#N~LYlI#ym8N&N0IY<n-K1kQW9p{E;q7ON!}i%TvpFIlbETYt%n
zf=Ax_YIYUTe)`$Or@nUeeaJ}l)rv1(0VIFG<GmGjYJA<U@^-Gkw=pjOWaCPbNO)DE
zC=EK66lo&L(%^6;CZ$0eC<~3X$fDsB)ufR}Nn2Zo6caXq9~^z}(5<Iv8qBPLNkuGV
zG@B4qtzWkc>otn(*OL-41ljDa+-&liNnxR_rAQS-Q9KQUsMYGKq9{s+p#zjem&C|)
zHsk(-3oaH*Ci}aR91#*;1)_-rJjpvk=ZJ$sc|||$a`$YP*G`vQmNmmQi=JtI?mzx9
ze&yGGgX3ooI6B_vi~sVQeB<k{b9u4YctxRLxpaj1fT}9_7r*r-Cew!5?vxLF=v6-P
z$q%x)?D*lI`Dq@$^D_VPx4z87$7{ZF*YWDD>xA=5j;D^DT0!iv8j?-Kq7-O5cZ&vb
z*d}d>`!RPpW8US{1l^H`p>qhI(m5PirQD%`>j$dL{aUV;%x60}woCk;V&;e$EYH94
z0q%YAw~4W5h%H^aqCbC389n>^yM*EdA0yxV#%uh&|M8dkvw!Xv^7Z0bon4|$<i?F7
z&d<---PvW`u6Xm!*O||E_|S(w%nL8w=E405Kl14hGMzSbUB^3b-NRHGcUf?@NSv<|
z<#d;|d&Ce1k)xXuRT+X-Tqe)i(BlhBYha-aUKKQUN>DWq&K9_+>0Dsl_VitgA6l%6
zOzM)s2g<tU=<qs4Q4#~}@9p#Ai!by13omiw=1q3yJ4~k&s;Z`{O5xgUZt}$7JeQY`
z`6s{p_xa)%KTnKs`}Q3^_`whJ!V52PczA>{C0h#i)x{gGjswOUU2+U%)TmV{>NcgQ
z8^bHFzK_GBYkcxYKE+@E>wlep@DKkXi&ZQ37mSbh9#^7&M~;mrC*%5%azA)Q;b;^^
zA*05W!Uo~*T%grZGG#iSV+zCCJB%_+i;Bt3W6HXrJUPepj?0TnimIlXPN*llcr*`B
z&N;hSaDK4_4g1p?>*e0Si^a!4WJrMlP&x|+GEzQzP7<Z~DWZ6w>BPo2aYPzs+RDZi
zGrHv^NkS}-G$x5}Xdq>D$}`ev#)3|d|0)eeQ8xuCXH{zu50-Jxy5V1|yxQjCB2QYb
zX2p*GRX~~Yh&BIRBHU2aT2YmjQd_JRS_ve&A)*JdIoT8q$WxdXaBXI|wOjCkqd6}g
zA9A=er7A>7w(bX#7jZ-k0oM=sz9+Z=9~>??8Hg(x^NH9AvJL}NK8aV4#oN`wYmE0A
z&GW~?&U;VSb>B5o9gDxBC^i|cSI4VYi><4B+V6F2t-+ow>_Xa(d9R8-o>cg5N5@k8
z%Wsq>v&3dp4+JSz87dyVaK&aUXWar(=j>Z;D6JJwK<^pWYY1Ml+w+?m$8h9^A^)`r
zB^P-%)qeU&uqnd(K{A!sZO3A@;$pGjV!5F0yUhty>iZXU-*06blVwIUPKxO54~#!|
z_2-{1`ZoW5|N8XzJl^{#Ha%I%UoG^uDiz5%OL?;>s)A`#)6D0r&MiYHTq&(#c5R>e
z&Fj!ES)bfvxwvFn2!R|w5Qx;%3C(QA&U{Yjd%`fVUajbtD}r;F<i*H%Qe(<elvvI)
z^qnX@jo!Gq%ZPOgEo-f_Yq?zHR%?{8*s>(W#L#tgzQY=^@UTU};D_vQH_+5I2{^x8
zak#T9v17esunMOW(L0O}L}zh*jaCP|bEf#}gO0A5a$fZO<S%}j{b!H4TwU^K|MVZ{
zcRu%JzVzF_%h_d1(gvkU;aO;n9~@d$41JGLHLhQf{EE}Zk2pQM&l`8Y!`{IWFTeCE
zmp-u4Q_QU7<oprpQRJ0tckpL#Vh%miY5{&hSv4rtpra$^jxpvAJI3sso<BEA+cQOf
zbM9qykWFm?WvrxZXqA1n<Z3xPJw;8Yq*&0dIw`af1Qf)lPn!x-;F;%N;fs?gF?ohK
z01oRq){jrQetbx>uVJDk1;xMkjjuEGiXZ*Vhq-=ZPBYQ;!<t*Sk2pVD^2S^Dn9mya
z4-PoGcEsiK0#krX9ojaWFHV@x=FE3z)OE#VY6;$P|KS<?M>kpBzt4HwL(|ZQf>Yme
z>O5x(*2-|X>TyP+=M%J^vRr9S@84s&YFRAT^qml3>!K7NtmHXO5j&~4cH@{kx9{-B
z{@Bm)!4G|m7hZgsx|yPtmDf^0wDAVqF1A;Y(Vg8XbyM-+;k&f!j=Oi?<Trowa~xm4
z$%j7t5q{_cKg3Hfy~?DSXYUg7hMJK!m9k}uD3p~lXq$%!8cb=>!Q<V4(wd`d*Z9P2
z#!D}~$WQ&$Pw{{IzyCk@;+MWmOxco0U*+8W;W@t7mROc0lSv~HV?&1z9`8lOB$hIT
zpjN3Mgv5HiCWgRtXP0_9LlqTX85VuZytLG}u5<JF8o&A*zs>pboYnG@t~}+|voCVz
z{V(7{;^K0}n{T|u+2b=}(#$3`DJZN@lo5uQ@ChokpujQV#$Ib2@kBw842(3jjY(~U
z973N_B&}(z6^<L(CvR~0OzPGdxii_yN1Jf~;)x;KL;+se&BdydjFH4pJ)y!_SPCcC
zYAvJM>}?W#Mjv^Sq&|*WYhYSxn!2X0YbH&Td0bocXc#)I3KS|(Y2itg4w@iLr&~Lg
z7j7Q$^35YY{@gL9y&$=s)V64!sC}ec_cC(rWi9O2J;Truf|IeF24y9U13OZDGGbSh
zS>-9u-{iQevJ~-?_nsl!R*Yl*s;ZtSoSrDql-Y<Wj4@(ilgF@QWaZNh_S47r8pUqt
zbLnbkZ+@JZHf#KNPdtqbUj3CZiSRIl$ZEBsjAHa}l%SASi^WIDmcE0^x#bz18^xA|
zHk3t)bAdiM`c+HlJ5(k1S;6*bJ7m?7hBR{LWRxs6Bx?Hys%^tC{=G3py0+!=a>3<t
z#pPnj<#Ne#)zS?yJ4j`}S)aS2@3WL$edWisIz9abzWc|YK4tG^%=&{I@42uWCFuF-
zY(AgzC}1ozV{Tzml%QTW^pwhoSRo=L*5GLSp1B)XouARJRz&YmIVe|KOG=4mHbIwG
z)LWT*Fr7{*l&0-Ey5)*eDGDXf{}^1Y^jM5Ul*t?tV{u55g^Uz)$C#vWDy^k75_Mur
zQH*)#=!Sv1oXQw%=!D+@5|>t1oS&ZKqGs>`V*@TqgqKPQohm{o!5(w>wBQ?$;c_y;
z9`5q~_rFN9{|pzGkvG5dF8}09zro4t52$sGElXBIzv=j5Oc+xFhO(GKvJ|Go<wS?*
z6lbUBC=>bJZ!Fo_J>+NqqYqPf!}6U6-22AsytO#xmHmd-@BcQ>-|VQ>fK!fga)<;c
z`%R&yvfmXW8URm{c1K7bHj9E%3Lhi75O;9$=*4KTBrYF6L|<s;w_fF<AM#_z9O0yh
zTH$@7)Fu1ZZZe(C@%}NF{SqG>s%tsF_cqs$j|f_zdmSu8XZX#}eVMO)^{agNV=wcw
zKld4?vw{bY@8X=Js1j9Eb9R0~S=CgvGyrv7(+>e-CM=hpZ-479i_0}C)mZI#_wHMK
z`uXSh(w9EV!^bNg4}sHu;7kQBRbqi+<s30;f&&#j!{UPTRZDOQ=Ve@@m5h^;ALv4$
z)?!v`Ck6ZSL;li#`b+%Cr#?+tPtc|yp*IM*+)^<ENl&Vj^NWsBind+z`0)v#GRIEy
z*4wY~*4@|m?63U_&pr17Kk|vs@SzWVg#Ckqydg@7u28wC+m1=(R|`>J5DJ5(WOSsT
zG&Nf7^XLBDpXG%YUf^&1&A-Xt`Mdu;i{+9K0><c#V)qZrQIrK`QKA$KLr>rLfEeqJ
zNcD(16}G@M@;5r?@WX%)kv>4|BXK=Y(PL|cscL3TL#4EM%f-Zxf8wLOeg7d39-Z>w
z<T3ZZ^%df+I~*Jy@ye}3Ub?=^#d5{j<$}}mb5^UCuo}PzOp27oqLrk@1n22pN1?Zc
zf#in9Y|jyJL6D`8gk!hUR4mqme6KZh!Lt}7DMRPt#FhqKh+(_2hTZ*rVhVWg&|0%x
zUgotl@bF?lH=aHyVp1IK*Q_o&tV)!XVOC8!nC+lrLWPKq5ZyrV11d)9(z3d^5DO(O
zkxI2OOv{R<C@76(GHI}RBpFbw`;NBj=!c$u7=-2!#h5tt&;nJ0!^&{$#+>)R`~o)*
z<{V8+c9q<p?W(0~JNmXKbP!#(GjX2a#V0d_C@c&><rFOPUQ?C^V-40=YHP8k5KMA2
zN-b%c3A5RZVQ^etUeLB3>vhNe{vn|8K5T3;QW~vqJiEJl%x61TE7xDQUeWhGJ_H$K
zW<~AOTt0Fk7V~*bj7XtyZ>4bAqI@H=l*Z~LuL&dfQR;75`^&1La+TyLw+@xWRlsI6
zE_#m(avjAOQKd}L$maJbSyPl!Ij&NkvNnc7E2^R(SO7~ww7!os4%e-D@a{WYZz}eV
zjyPRha((}pK1hzdttti=GukKXmx4TLP{ibM-b)jl4cEK%T7+<e=i=gmd-v|~=+PrC
zFE6*~t{>-)e2+?P3v6w_p1+d#uA=UH9p6>Q{U8wQAJ{0CZT0`X$(SO6PPqdkMvPL}
zq;PFV=OeW-H1l0%O~uK{1I|uPn3jrmc|i)E%1CilSR+XcN|CIlm^8$pC#J}9c}Zmp
zX8Uu(LeZ~Q49g`dL`+DSx&|vDyv}M;FMc%Ii0ipl1x_iP_gnI1OcH8sJ-V(b3oFrQ
zlLmTC?>*WF%A%mEO8mMdr9f3zG?NMMymJnwARx%4PvXmDq9*nhH%z&Vd%V8XoD>c3
zxHW(3$NnS_F4kNu;j_Q?dCnf45SB}{jrg==w!h2uy?q8ZuwJcMt=81l6s=*sawt<1
zFa%?m@9s05CqDC&AL7S;@?$_pQ8q%v4xZ)7b9`n0I_rCvoZS5y`}Iq_{ou<yce?^F
zqKIi(5Huo=V9UoCJNi)?KXMwbtRYW~17d=rY&N*GQD9QQT1{AA@ZkK6+0Ex!uNK58
zzI>8&V6t3ZI<Vbq$4q8>bdOJ1`2iC=satdL;2yU=_5qrirGL~)XP6B$PaduK{BM1Y
zhYwEp#Xs}MnN9ck&g)<2h38*{STYO|Yirtez`Ks-_<-+x>ppM1ewTaq9`TK@euoq^
zg>D%xAMow3oicq?@bzy!;_PzG*^ua}niFq0?FTM>q;rBym;|MAXk#-sG-Q>YteQrv
zO)Q{H5EO^|$NbO!?4Rd@ANUAHm5?O&FtJ@EWZnJV{J8N@TGMwui^XEY$ML~$#w@3g
zSDZe+$2Y$IRc_wA&8I%~X+HI-ALZ!qTIToJ%$1RXaNDbA=3{;`${4It+36iZA?Vyk
zKl%}lkB@ob`4{+W|IL5H!-o&?<BI)<=rBs>L>NWT9;401Y!;AnlSXclx?)vfGAicr
zdHGAu69-4B4TZ5evsG<2p@Z23t{*ZfEYs3*c791}FK8cojBlx{hBCp?e#8FJoRhP2
z&QDM2)@yv<QzYU4>SR!*CHPUWKz^<%t2H$~YK9OPd_>}*6@^ivOxK`wf)ME#$EC8@
zRAtHj(INZSjwtGy`R-g4g-!_kCyySoTCG^E)>t)UbVRJCTyUUe({HMZ*<`}*&YW4@
z(AWYK!L62bt0k^&u|y`*hRPbM!lH>(+E6A<p)^$?sRhz(i^8=Za#80UgY)<hq&dm_
zy8Q~S&u84&-{FO4u5tU?5siV;4Rq@zezl-)dt5i*J4YNKxQNe%-7xsA@x9SBbwyEI
zOsOyyl8I=mGkR;OY)P5Z1!9z<P-l9IC}CBm^lX*YqYy%_F)?o*e?}CuQnZa|pL1TU
zMxKiJylT#`Ghb0{Af%^8t8AO`r;Cc*KW;yMiX%AM-Dqv(cjjhCp|W#@%*+#l;sqV@
z-!}@x^jJIQ>vJ=aM=O=b>LSuG3A)u1-**VdE|P*1ssWZw;8%XUE8!Ab^g3+;h?s=C
z=bdM8E*ID>UGH$2@{q<N_q$%3aTa=ICi?y6v`-gmKS<&B|HtFKizm{R`TkR1iYX$r
z(zq$`bPR-{rw9oXB2Cp0kf>mzL~zk}YZ*^z2wg9xhPjJ~3ff_yPl2MeRC<D0lq}aR
zc3M&I&E$J7R}foLjv6sDgDHiZfE$CYJqfcp7up!fx=TQac$|!1^9^UMMi&Kz2Dc6j
zZlKKSq}~m5!9$GHO)Xuz7AI}zTY3~BDa@epYm2t~yzwyd=6Z*_&azf(KKDDH=f{5f
zCwcSUw|KmI4JMxHjv@Am54`^se(8Vv=V{vxbmYrl`YNks%k^uw_|`Yy<Xhi(11LFu
z;U>?%?>4Wze4D1)p)3MMl{|jr`1Q|zmN&lh7FBI|`K9N$^~?`ZPHyn+2PaHb#9dr)
z^6)W7*CR<^qC!bkh?99t5j6NXZfJS&h;e(-=OTZ+VMh-^7XxM8WOO%SQj}477zphp
zm-pYI_G?_%;@v<|Rr32%5br!~NCB$(KJ{eI`Z2g%FvqrKefF4VcPC7yCCj%TVHk)?
z<71+*HQrgi^UVkRz5n5}{P92e6U>_Hw5!PB(Jr%D&uZPHl;PonQ$F|WU*>CHeT}#8
zJ|IOyZD$mv!I%q<=8pd2f?xge8@OdpS5`dI9cR7c;VN?BBb^om#73wUfFd}b(d*1>
z%H91K%&n}(MYwh2CcpGc|2zKt|NK9pvJ;G{;0nqKTdu|LF>Qea=jWH4on0`UHq3Ws
zfS5^#=rhS)F$`-im#2K|J74A3Klji0iJ$msKJ_D?;qc%(<R~$bPD&w}aadxKd{3$L
z#XL6xh0$1JvDR|)<_-R{zx<atJUrmP{IC9NUVr_K98va%=h!M1)dnT;UP<GjWwR7S
zML2h5S>zE~f{<`YxQRm))n~Gfs3mtfDupe;7~!QgMsd8q!|tTwaA(TfZ@<II**WX8
zC9AWC%%(HUWXkUKV-9W~Gnp04YQ_3;LAP8Hy8#nrQel<B(rkFLX|$I}!gh`+(gjD?
z4-7uyLe5^-khHk;#}sLib8Zq8R<Sdg@cgaYJpal|)U!EdQ{sZB?|WA3HKjE?dh`hI
zJawsXCQ(~UQW9csW`<c+F>e}nCUa)93C9OV9PaE=q3JHpxjZ>xd45K}xMbecOiD{#
z7z!oo*Vub-fl^0Y*VDU!^)Rq-16}eAQB<%)NQ4CBCaWnlw|6Zs-ah1+8^^qG^CpFY
zetF5Tx};q`Cb$;o0>MR+OT-)z=6zroJVWng)TkA8ZQ0q|p_x|L!ia5D=!i<tbZfK}
z<Aq!kV?nqPlflM-ctnDoa~lP02w~GGjAPpI8jLxzZ5V{lnMjyyl$~Of^qFjxqO~Sn
z;oObuw1Q1x1>e1Yf3IVca`BX!*Iap@Ou<nJVu&OYH#5l?@*3p*W`qpVh>ixoS2<`S
z@6H&5)f%e}ZQn9<YeL^)wZRP?No%yqHYvH#dKy!^I&L1}H9@caM%mtW9eqFGTo|(z
zHbhiu)Shmr)s^4y2Y8GR|F2mP{%dpWzPqp+6)HIvBZ|*Vp)|hhsAD7!J;~Y}jNTHX
zBsL^BKy*~b;<U#1{f1Z?Vxa9iE*F=~rxWVDGWr-9RHP|NCesEz1mfU{T}vn|s;)3r
zQ&ttpCHk(zIS0`*y77<ldW}S%X(PYSJK@k2WkFS!n4-cIGR}w~BzEWNI_XZTvc#bo
zd}0WJm<EySM@4Kkx|`7^#k~u76f2gxrmPeXPw(;1e*N!qbi3gLA1^RvgQB7^Gl+_d
z?j2O>D2sxlW6w*kUZ-mvKXvDWT)R>6$}8{VVYi~y1@AmwQ0g@&_a5-}8*g&w<}E(?
zsSmQdU$VSh@x?EHjeq#dzfEXs_I$+(&4+hC!TI@bP*2X-DLdLWP?h3AG`JprC8KWS
zD8>zsa@z@c@AHBpH;qpU)l9dZJt+~Br%;gkC2xKGE4=i+k1;*ILkv;IIz_n|C!nN|
zC}%U~ySudSX!=Z=Rl#%q@E)%enrlb<#FLgA2M1jE6>r^r#L@mf!5fyVHQ)H!n>>1O
z!tG}c_|&IA!E_dIL*li2Z?m{u@zt+<gVU2sRH_NCBnFK#6_pv-n-zrfa~|A%lXTW%
zO~s=wa4#hucAh?I)(GE*Bph*v+VEb=qM*_Vo5$JWEhcEm5F*WV!XNwjKgo}L`p0oY
z0t|gSpbJYRVygmUH^jQ%^S<=n^Y%OM@b0_!m`rMp`hCjEB039CN<orPjKdm-QkJj%
z?w7cG_b#9N_0REhKlk%|{D(iq&d#1}65~R-`PFhx1WS}_K*lth6aq0s%A&w%!=L%Z
zUu188pTGKF{1<%r%U|27%>ED@g*Mpi!B!YUGA1wW7Gq>|lzd>wURAa-SX%%J=K@9-
z5^XgMQg95OK6R8aPz(cg?Ka5*vqnZ~VPe^9EXKDKaS(N0yTV;|gki~p)no3x^BT9F
zd!7%!{4Av{h|Y8W?RPjkIUx)lt2EQegyg(zw0VqHSi9lE_I=OA#RaQ%OWO_j1Vt&$
za}>Q3gM%IjT2W0Jp1FOC7oT~C=kGkr&d~ubNwepI<Kp6i6lKG1+ck%HF5QybFWlzY
z7oKBpXP?I>XG9+{N;54QVm~n3o3nqg&%tCyT<vl`tvD^<?ej<Ip(D9O&`@M#TL*<s
z2KoUXBmL0RIZq!U7~!`01bvj}zH8xdVtDq(4zE7*40jF>IhxIwmzMSUIo<gg!{Us^
z<zwKAp?yerC;n-{3*wGR6vj|iCG*|6ct}obl!3wZT&|KtLaKx{hUs*=DeOi*_lVvJ
zFLTSQTdh_bvBbEi7DcfUOKb!aQ5uP1=r^E)&JERwp24;$F-DLylMBX_jhMwaIBjS&
z->YWWqFh%jm<qF@IF0dh#%xWMM>`Y%qIkoiv)e?BMCS?C=cJ`kU?qF7yzb+0M{UF=
zMd9Uju$uMilI7x@o%JQ<^oV}F!ZsyEJ;86P!g%FPjwikNq;YZ1G4um%+p%1(xVX4r
zxm?oso)EXMiONm-dtbTUgTmqWd>sC@AM3ppc77l#aT6y+Nq^Ado#SI5CPxedDGsRQ
zQE{NMiP{9hYJv5hZrNh$8dF5(`xA;3N$WMrdA!xwW=iK*^aPX_Pn!}=@{z7{lubcp
zE4(g=Lm-r*Dzvjmw^-pq!uJD{-U%PbXnbQOY&m#TNLj5jq9++r){-)zgDm6Ti{*(X
zNe-$u1fy_4QC1Ulr7^xIDyT~7@CTz9Y>5dK^QsV3!|t*w@9?eHYaaSJt%;cWg1u{r
z8@Fz-Sg(2G>yPPOPuun+rzy+?KX?{zKjx=?=EGoCJUV%ox}I_K=F5a`iKzqcKKdr#
zdGnmJv&8eyJ<GGVpW&cB;@$oMacxPR=U`rQ^I*nqG2@^7^DnR-3NDtO2T^fjRxv9M
zxL7=9GBwz$Bw#3QNgM`Ix~W7`B8CxTOeyllqAmn^KL$ufHvh}P;mZ=!d8#N}g-Z>U
zj}+^Ox_83*8{cB@*7MS-x`-{5+*n2tLIR^`tmgW)ea_Idt&>=mu0<_RSw4Q77oWY&
zy>FlK?5iJU)t=C|o}c)MAHnL9i}MS<^PO+=)|;>Mj$iUWou0FQ^A4}Q_B!3FgOKop
zCq<31iGGj;_UN$SrQ@2j*B|ic?p@Z4b5<sDk$N6>o>fu=Yw-&3QT$U8gg|9KH={(K
z5R}44m^2j{7`hG@14UKwBR}#<e(c9SgE1NxJnOa<3yneJgQu)Ys;Z=@N?hn@yA{jj
zg1#T<`kuaRS!JW>!8tB3&UpRXukrZNIqG1}gJHm0BT-NjQFLpq7}Of2+Z=6E5HAB?
z{n~=J?!Li4|J8rWPyf^(<CCBK4Aa@J5Z{$12H`)7fCuqbQi`aB3#1U!?s42~n}(nJ
z>7SyhOa9CM>c8f5pZhIQ6LZBR`AV1m@6r)+ypI_(Mf2t*T&j}3Z;4SdQDxH*gf%4y
zKB8=i(H4tPJ(7~g9EOA+di>B)78X}k?Ci|(z9Ytj4uSpo1as>e`+HMXt2OKOnjr$#
zqD#x;H(%%R8*flI4LdtK+&Dhw+1tl>=UA;)JbwI`+1{?$OQ5hw?3V#lOt^kvudeC(
zo^{)Cc5z9mJVEJfL1ppD(?;+{vwN_|D=)slt=qRaIN0a#_?XLe%V-PROlR1##1<9x
zWQr&7@Zupaz4{U_z4|h?sJOrTkfCz~FY4WQ-gukk(6J7V<4MiI!5)XZa}MWI!t$KO
zyYCRYp411T%4%L^P$@7B;u)t^qAtJ~#VRC{7w%vI9Mo|8c$YiJhrIgy9d7LGV!gwy
zmppiINq2FM?^+Q<WMn-?2qEHpl(i>G!%|L5tg*~?<}{Orx-O|FB}z*%*AFdz=mA0W
zv@%q=Xdn4wZQE`}sN*#>7JEfeY<%lRAj4WqRaKkf-x$N-1!x%hK?=JRvD$7;fc4fx
zbUd!6UVPV7P^s@xvZ$=`RvQ7_M!}m#^pA|*nJl0gZBO!V$Y75lqj%#Tk<rK)!j_}2
zWP&DHsoNA6r4$C8#fqXBj;*6xt!S4^uJ7I8vhOjKL#sUHNqVy%3Fj?v)v-hVp1kOI
zak}XHj<#L1xV)ro*Ez5KDo~b9nB(X*@>C_9{#AvSy7EBpaglp^{+qV7%6mE9Q(>2q
z#NDJ6o|wDK*Txxn%Ch=|Ptxtg5YQ=*2oe_(9I-Y`cPEr2y44DUV(123bQFzY-L9Ex
zP4u49SPGjJg2phN%}Jf33%!g)$`V_Z7$*k7Aw~S)sEb0BnO0+~f)u@U56%gMs}&GY
zCZW|B^`WqcXu~;2-*srEsLGO92)rIK*s7t78lMbhVJQtFf7con4arA{n$;Q}tQ2=I
z0&!<f^lSES7;e2-^46Pga(1dH%#6C;W9ORU&W+pbPOtIWx8J5)p7Z#0!56;tD}3@t
zKf-)(&UDtWZap_{9dd9u<A;Cb6P%p{9zJ-$*~y!H_E&$KYX`TvaeRaIs^|K(>kvF#
zoA4j~sXxK5{rXqAyL?En4R1U;<9&zwJUVwA->C6X(K}CBMT{x3$~A5-xHO`7<HsBH
zHai-GNZ$<%$)c(nn^6Oo6hSN5KsrY`c>L)(jj^J_HRQAqF_c!hvs6{dWN#nu6D}lb
z1vUieSA_K`uRe3k=lti{-JcWGf}77AQ`Cv7QrKpOuAk@r>0RzT{|q~O2mJHj{5+Fp
z#*hERPjUB+H`$rZxqtsYr)Ou_=(%>Z&qsgg8t1Qlm3!a*I{o6D!6Y81z&o919icQ0
zD9sSojNAIwhbg<`YY5)aW^tTLNeVb02r2QtmtW#XKmBP!@H}{UPZqx9s46kSJ-;|Z
zD|qzi5wE}TI<LL;9ZnuU5VU2rqHWuavJPOmTyl1H#(Lc{ola=G9_L5V!ivHcG?PYz
zK8@rYL%<qOECM|}ec$r<@k74#t>5K~U;G_@;TQf4&p-E~h;@=8CA&$+ASvOUWdt{Q
zk50nVG{%5}rY`x+r$5Di{-6J!_$z<qukx$E`dJdvFl`$AyjYps9F7~@_u}vVn>w5y
z@cz5I?k#^wtRBXOR%?^r&#crDd9X(4CAor4DToR+IilAD*HhT7J&-Ym$)qI2T1KZ)
z;`XYfK*{eYE%=mBG2qrqPEQr}WFmp;wdL^EbukJ~prS$tDW*&W9TJ7nOp1~if?Pk*
zVNAp&F^kuQ#%WFAB7?T<%o>i5k2pL!V0U+iz5N3ymf~EZpd1=7hTiAmkU+ADX*ESh
zja8NxpMQy>tRX2*9zEi%H{WI14=nqhU=8)ooTjwc<hgcy9e??lfcTo$<c0~gSznTt
z=FjS!u96EnTfk0fdH&8dZXE4%{cx9SyE{}$F$^93>XOyPV}`{AA)71@-e*g$u>g^^
zO<O@ytEQ%`N_O{V)J>i3FGNt`gCm3i5W5ef#r%1+>`+=1o7NQgEXEiHw=EcwH~@@M
zb=FvPDd?}{6_;2eLJA-J#t_>%x3T|v!W4e9504HNnM;>B&sSBVBkyh#OE@m*jGVk}
z74c*W#mKXh`(RAh2%B%09OALm8$ATC);1d#g&;9nDJ4S6{Ok>qxaAm*X2^c%>DDU>
ztpr6FjRP}@E~Q8u$0wO4lG$9SjS_6@jNt}H-}TZw4Q}Jdcjcwvi2`l>{O0&x$lR48
z_q%`KJswY0a_@0Xd{2d4hyfMU_q>pl0;3F3Srq*S9j)pLYb%oONJ@+A^Dr>DKs~9k
zl_iqs-N0m4(=?XF>YR45WHy`Qdq-6ilx0Q8gr%aXF=c^oTg>1{R^;Td37{N<6F!0s
zK}ep)P*k;at+7MJ9MO>@OkzqDW?Q6cox{G9IDxk<7^5kxI$tP>x~_<5)_ouZM^$Jh
zwZ>{iS!m%J_>#yT@1FLU-2+yvnH-fI+??}p^)AOR-r^HK{yuKsyu;z%Au&3dT8Q@_
z{mcis`}*&2d~?Q!(g!I@L)$I6Se$Tp?FNuAMdI*y!u2~fH_r;b^X<Dle}9j|y#w0e
zF}u4n9z8zc`r&nc_nTkmqo4dZ*KXFFpBVbnz@o2k9N>M!>G_Il*XM*7==)xjNmm4o
zLd1T`406L;h?`<8ggjQY79SHz8Cd|7LX&XAfcK8Y<pq`2s2~wqK1!^c&Lh5%Nyn3;
zW0ph0^^RH_rq+<+!08)b=ebvZoWJ;|f1JK(iTefDj$cBXgbOQ9&Q9rjkJpa9oBPbB
zbLKlG3eUG*{~Ap_p`2Onyl{gqUh={#FH%`}>sw#q?iYWX_2L|*45!_|eV<qZ%4i~y
z-t~;iQH888DnV5U={rS<tBDHAvH+!MofoHmrFiGT1HS(4ZxfYb-Su27&Ux!?%h~xE
z_wV21cfa;k-o5t@=jZ1%lLlLwyxB)lq^2z=$2m{ewS?d)iqU!Ai05opeXrLY%hgK8
zJB?yCpE8+D2;Rx~&{zi5Q<fEv&mQs1|MVa5m9Kn>pZ(cC!O#EvpJspWNVs=-L>aPq
zu~PE4_eoUd<WY2p2?LZx#Ros|0sfEwhyOi)<^TLw`IZ0guVS=*QdFT7##&NNGx@h&
z*bRNhV8%_)Y{pSQ9<7ZVT(<qtR*ZR*gq?4=l~(1ESP=V&7>OzSqb2ZzqOgMC6lFzO
z8LFzF9}0@WVvVM(3dxp?3Kd{~J|S7l5In1)=j7o7RB37EGj{g(IXb@1FbGdrMbJK?
zoLG%$lPD;tCIv<*VsNZi=U6LzK&KSOCqij4eV|R6>9pqXaG$;XU8eInv-u9*Scw2r
ziPC#iS<*T$BU<lh`!zagRu?OVA@ToX?>~dB%g*~w?6=b1`-FRLcyssLjc$Z)L=q&y
z00JaHQmHi3Xf%>(Yf9r#R;k8vP5F!M@wm$6Q1iu8uKDDeZ{x95uChj>fk=u9q)1UD
zNFop<0uj2=IlTSGo6k8rtmF^R+UMN2(E!<E0EL;=b^G4;-V^uPd#z{vpMQAQ^_v`A
zx`K5TF)4!5tV3kwJ>F`}bV5~IR2ta7dWGX#FB8*%b(tn7RD_TfDaxiVCyC5MDO8ZZ
zPn@QiSoZg3eDK{*u)8y3*3>k{FszoW&reyOp0GSS!LL{397r(}HW2}35iNP9m2*^e
zMKhZ)nbgb{4OQjD_}KTvC>Af|fHhL+7$dwhpcFj@jM_wl#TX<WPK8Y|Wwa&uweawi
z<cK<3Nl{lq<8z{jIED?s&S|$1v;k7s@Nf|}UO_pJ8;c@_8;7EFUUCtyl|o3B0;@bi
zDm=~2-7fI$M%RRkDt!$X)`lXiDF0D}J)@}SBFdJOk`C8q10~mwH6lEU-qWquB0kDU
z%Qr<4{ZJe@HZ-Bqa!%t$Y+HoIv~9=sT~E7S<9*!JFr(PwTSXJ&*Tq)DMbvKm`$a>c
zx47wd8W(6U4?o{SfBZv@w^i7sl!(N)EbOFbUVLSek~Y;4=yFTFf518`ak1A16#^-E
z{CbV4cMGcF3Cj)_9cfrIbrp?u^nOV_oydt)Tar<Pl*!IO3%>UR=do3Tt80<-E6F|V
z{eUyljy1-RjFmjtsL4rGIx#0yRA9F!?fAW_94SS-?>Rq@%(JM(Bmze!n7&`ruFjd(
zj=k9wodYR)VizIqvKn?c@rG+3csGlK3Geyg53snl!(?AG^qRY`pP{4W=;)mL_aE?$
zXTHwh12-RigoCNam;;`C@@Y=a?y<AGpz8<z^Z)yQ&b8~;`MdwCzk_y~Y&sT~Yd(DN
zUY>scQ>;%SU;4tcoUczfUmc@##k-z*KabqJ!sG8)V6x_GzkM4$y~0s@!j;84xOMll
z>>tc9DsbNSnB7Hz=HIkyyXYc1?#RaDp%o!Uth4B<B7#WZlM)_dkod5n?+ATIST33E
zUlC%z%4DKAT$PI}ro`TjM`)?&R*9KS7+2xLniszCo2>7ia`pOCOfEgf<i=CDW+Fl2
zA+c&7=X}-j<fB(WX}WIU$G`tQeB;^Yc-K3w^5E!%$rQp6`Oy2H;LaPb^7XHNjotNq
z!uph?;KT>+hRhPh0OXi43}mvPy*Y0-jg2!jlLoC6Lnm!VQyIDtq+O{-R6BG3=!oC@
z@|V!oa_iP>+`aP#uitu|v(r<27*GnDMno+AkT_rYO}^*Fs3DLdtLr+3L0Z*)ABx$j
zM;m=1A34(ZYw{{_db(yhZCLE?Fq=*2`=JztVhd2Yio5p`pZdrDKfe04FY{Ob>R;z0
zKlr0ENos?!65D31*b?PBGvrR%T46-<4Jjw8s^UW*{2>2_|KvYGYs0Vo`ftF+!Y+%Y
z(8e3{yUfRH+YXndaD{>rmHF<XicuECV#}m7+G-|~3FqyKRJ=Q_)<i#AKLO5_FndvO
zu2+h_x45dJ4xYN1$Res_6QPaJB6Jd=!(u+Cna_zv(Z#^MlM{}XXROc9IXye$^;@rV
z_4*B(W+EtURa4bYTC$0P<Ow2f6Wq9QiRr1Pi-932W?AEgh+B1>4UuM2vAeg+bT-3Q
z7F{`Zs$G1N{8*bZ%GK<3p58}JjvsLC;2P`ol6$Y;<>-9H3(vpC;?e=lY{tp*j5a0Q
zWCD{ZG!ufeXqCvdV}5CmK*m&-sxnl;69!LAnL%Y@C2LM9o@qY7q=G9~cDa7-3h#RI
z39eq=hv*4I$N7Up4)49e`N;#?<%*$=f~XszN+<=^8mx1eViasE5jRwIMLnshDvQ#Y
zA+_{FyJ@Lft7)1VO<j29vIY*rAjO%}8==I-m@x1CL!7#<>ozUgi@Z8hcypn&VMpaG
zfU(ks9l3m5=$^fmy-+Gfd%}Sq<21J$7M0RA$R3%;#p72dt<eJMq9L*rUBa`=u?Q)~
z+|HN^@+BaE9*e=U6O1_)_XEI45qz<}9_bq6ahXjT)*-MwJ0-<{bDER_zV9#-5e$hD
zO=-!dlAxlu5o3Z9Cj1};-MU?~S}s|gF9}0wA!`K8DA$PqmzsyGvOg^MZ446ob{5Cq
z>3(lU{@(t0o2}TbtQvS|cdZJ%oU<awgpR_It1HWNKBqf;fDay>0#;|7Rt)DW)<!18
zvu;W6C6(a(_>epd#LDA`ft|fwqLKDSHih$3yAoL<I~nJsfKyHkD`OPtyC6cYK+tmX
zrIN>}blR9(ju+gh$Qj?CbFM6kJ<*>qpU<%+;iB`A)$*Kvy=E~nOr4d;ofya=qt%@I
zM+T<XdF1_1Q$I21{_+7YfAvN74i;SAeFy9Nfp2{ID>xguc4NWiMZ^8WmaqKoH!<lL
zXAB?v$scAmze3$OUViyyUV7;+PdxD<3=3AvjB6au$!nj^6qCyEv7i1hufP0&fBrAN
z!K<&_Lxls@!+jom{4$ph9_8?L&DqNX57wHiJJ&dUU^!p5s98nu1G8yUyu0Fs9absu
zV%gP{B5#DwAxoyKHP}g0is)F1i!5!|m`EX@bD}#rW#^GVOrBs2IYcxGn<c+D6PaJW
zMy}=z{vK@`NDIYmV(G(*SHJMD`P}_q<S-h{&H>t1T)KRjM;?2Et2b`4f9Wdk+uP;R
zm8(pqGraHV|NKwV2F1y`<M@2VGtYb-OX73SehJ$zQT;iyS;et$x#tJY(R65XX#qsa
z#F@6cVxz%2NAG1PFP>MKl#7+lIJq$1HS@wtuW)>Nif>m$-(#&32YRDQTfw6`IR|}6
z1sN(-fQ<q<T~OXx>SoTmJ1yjI@#9j6LQMoPLXZzW()TOY>z?VfW@j;@t{sqYl_i9M
z7(M;a^3|_>kz2Q3=VyNU8UDgw_;<K;=?X^6w2?~Sc{OPO=-Q5?B%Oc|FvU;OTFVDN
z@B#jh|LK29+jV^Av!5;8OgT0e7o+cXVV6mQ$drv=C@pPkT{-r-1i(jdh7g3Xt(17%
zk)@4TJB2ldHuMyiUYg30l0~a5(O(++zM~rk+QBo+8I(oa3fn}YGGg={6JAADc_4)j
zt14<|xw^kk<0{VAYtB|HmS-z`b%vP)Xd3L?;HC}9iQp`z$kKZldQA3QzPt-+py~&@
z1eL8YL6fUBaothXj%rfV%o?iM6xUQtXFGJ>vmOQ-Ww1j}GzL{wLXKaqsZ_<(%`kgY
z4wvV&{yw@oB`D2wXAi8QO_}5@sj_INnW$Z^K7JE#g!gQnp~@9DN!yGNF|&fw`Gk%z
zom;N#?{n?SRjyvT%(a8dERPOp+ZAoQ<m~hThlls*yC6wH8YT;vHWf|PP?;&txQ$1j
zH5%<C2}x_&xNDv0hYmmV1m8o>RIZ_E9FwL-XYhl^4}oFute58~G)<GK>v~fJjo0mX
zu8&1t*L74?#VDNE+UE%J<4ZR8c>i%t8VkUS{-I-GHzrS91Yc4db9u{LeW9?^7Z5tF
z$5wuNs2<t+-HjZ(?N`>Ov}a3Ud0~^5QUX_R%9I!*As1;@@s-m?ZSKFJJZO!vhALYo
z)0*?;Dd(pr^nHu1cIjfmcOCPo)E=Xh^I?8q7LX_A4N^C%xmT-ovFTY6bHW(u+p3tq
zO&Rv);_hwEQ{N7``xBb(-gaR(_UJE4xCJ4Q1&f?fG2(K>hYpuC&P1H@WHpd;PfnhR
z(HJEjbdy;_voj+HjnkQ2cn7+udz#6FRq`9CTsvUAV7#>Vd=q5nRXH((4HpW}axr8d
z=;<VnP>b3vC*i)RoQiC_oG3XJ1(QYV39D5i^q%9RBdVstRF*!<dmQ?n`NUw7hq7SR
zuEq~FU6|3;Gyd_f{t{Q-zfUu5@T-bDFZ8&4m!J5N??*p=iTk(ju@h$WK6B^fjMd>0
z)@7dg`aLck9Fo<*t=r$=Y<0{B-~Z!0|IG*dXaAS~imTVI@yO#>c-MPw@XmKV!p_c|
z*>px#+7o@>@`E4w2|o4T{3Z_`yuiQv?|qWJ-5svoyw1H>U&Un0I!=k~a&mIU&P?)1
zm&+9g2a8SXZKLqYML|4rCsHOy$&gf8dL?yJiy}*FvM<&(F{85xTPkZXInlSBC@M(|
zp{H>*CR<bzv+JlbcCkw}-$!T1df73|6APo+)d@pSbZc}>_`@}AJ8<W@$Qz&eEo@!m
zqXK1_?JStiW=wZyOfOwye(f=KZamK7>J6^!O!(sO{5FTLy}<jPc$6FrcOIN_dlfhb
zI<3h`eG9UFapx8joUgmBqN?03$GrK+wly-gA_DjCKcF(g$;%O9-vLrEX)TM3a83*g
z|4>Urmq|302`q%ja&@@nu%wZ%^6;xpar-^RK;H#ctB(0j!+bs^#*B4RxKyE}jHJXb
z{F6`d(n~M#H~z-|n5Un9KNMoV7+Gto$%KUPo`Uxq616V;sHU!Y|NGv{|NejbzvKV*
zpZ#Zi@k?LcDz5(wsVHnQeU*i5G~<@Pr4s8W_CArQ0)FU)L#(9>pwt*6RXppEMTMM-
zho&N#v_&vVz!qhrlwgxTs)|BzNImPmMQMerD-2eM`o=J?96B{P6jSS{v_+>(9z4En
zX_dmJ3garUnqWY=lI3bGx~^ogj~Q1rG+64Ca4Dgz#yW$o9IkRGtErlrfehY*)@WN%
zPi9Q#1KL$MSF?&s`mRS+HP%*OENDl}iOF=j$?ZNpKWDwSpspNM(=gxNqc6uQD2sL$
zXKM!Ui6LT9gnl4<$-mv5?(o=+8|?4wGHYslyJU5C!ui=L%jG$zrzf=CfVGOr%rV=U
zvbSF`X&P$RP@5?la{-jub`C`HRl?$fCq^$Qb!ovnBdD25MTAvoJwCQrR~PeO^H3KE
zq@a^dVifM2j~FYzQ@W%eZKz$D5ao70$k)Y<Pe!^Ls$BEq^NedAP`V#u$0d=JpzUlr
zZOHXk9#(8iwnf&(A*kB7TAyq^OvQ3V6@+h0EE?Ov!b6sg6H2m{^~l>(TZ^ahxunG{
z5o5%J_Z@K<a5dy4evzsW_|%2hw~1XtU6>C?>WZ{shkl?R2KvD>K7tg5Z)K?d2lZ3`
zXz1FX#CThUo%exk@&?hmIFyo%xrF%Bsnk=Y78er!{wt*R5WxgIDhG5hvQy9%gAepE
z&^SkISA=$8QdP8FN7dAfo;^ZuQJa=j^nPOm>qEefx5bHSFA7=RB*?lztLg&OsYG%u
z4KgW8FCvl`eTAmeDXm}8^&RV#M|(?BiV!@CAUr^=a7v3sLehkwX?sIkSz@KQb^mp`
z(~f=n7-vVXv#569=@0X}zxrD|{@9$|>y9Ts_#_9rm$?1Pn%1xQ`q$pzwO4+Hb$?3M
z9s36}s-|LSGwX9nnt0)xhfLi8cVGJktJMj&Z{OyzM{csWzu?mD6?C>7>|dchzl8}C
zUU~U-?moCj%NlDUZLrK|kFj2Vkzwc|SxB+$6#eG&l@yPRD5tg%wbGfv59kPf7=++j
zIkFL#agCO_GeZjE&|e#iIy=Y3fRCO)30lt@oy6xbYEV_hY`(*hhe)RFh4ehFYo?7O
z%`8gwQ2eZ-D;yjxPVpf|{PHftk>}*L;x!y?y`XLOXidfL!9MqIzrs5oxyrmYyu4iV
zV2GS6SRn|`mNx<4dJ%7TZ*R$lc&nOa>^o!#QRMzAlmk>oF-n)8iCcfNGG~uTB&jHr
zrQ0_y8W%Y;Z+h;{1P)4Qa|}bIYgepS1M~Tmx-ryEMNX<bj>1^Om%sdb+`D(5zy8<%
zCO`hsk7Jq{T8pR&t*9mwls2r-mke!3Hc<>`>zeg?jWw2!{Ll~aAOF3-$N%a7{D0-O
z*I#>!gY(@T##(GKm^CJ0v=Na=$|8s=n^&@I3jHvUbS(Tj*&u3VHj!DQ8LMP!O3OCK
zM6dLg1EhstrO>5#Or($s*Vhq-LC_u5(bP4SwNjL2Xsp3BHI23SW>SO=iliWifbZ7C
z<k7)nYDd<^kI-t8v*fBFWr=m7ZZH_^;ITtMRgOdwL0N4J1p(n!X-zJiOKWSYdWLb0
zDC={k?R%=yiX1t&M31tP6Rl93o}RJ3bimw7x5G3I9%&VOAZSfJnJ}qqREoq5gVCsz
zsg32@{v{r{ewj)eviF>woN<0~$l2)$%jJ^gvco#deCAl}O}KRB5{tzQYsKhTMUBCb
zOPIVATfV3_rFc&%6vN%}S(Ocq2pnpn_hgiyLn+Y@k=pmT;Be7Cw4PkNAY;z-gU5$V
zgLI>ev!Jy!LX=#?Aq9L4gcxLlP_E}x9wW!JL8&&&+V;n>wZdpaChtpV9OSeW7F;Z<
zO81R7Z@DfMsTaFjag%l;)Q?mOsyNPzNz$KUj3_NF<r07)AWRXSjGIug>XA08(}r4G
zY=S&^+Osp<WFI0WWg>2P$GLdg<?&!BC1THF!4gB<_}6t^PuGdbb;^`ketQt|!~gpZ
z<84B4f8ZWA^R0;B+u!)Uy~1u7&~^O~1pMZg6jMYK!FN=tNB2F;yFIB}i>fXq6b79H
zS<6A=gZL0xThsSvXf)HEIlgVFrxSb#1OrKDT#~q!6e7usLGVbx%t}Flb*+M+RM|C@
z{J|oakg#CAaafm+RY~it@W^sb<NzT`)}M0@ng(*F4ML#JTA&tZVb(Y-#luJGGU<DU
zsPIZNsKD-Gf<Foz-)k8<$CVq7<HMJD`rT9ZccEL}0e_twJXU8^aO9|ITZ0c1s>afF
zncclT&h?TMEGk=K9QfupUg1L@dV+U9d7W3Dzs0ZrtIu-z(j|WKCx49PspsT)$%F}0
zb-=w_uW;-3>kK*(^uVeQEc70$RWE5DMho|8G?N@*`0~3PvrLRijuR;p{7?wuiFs3_
zY*ox$b4gzSa;EP@m{D89aB@<FEQz23DMxbD5Fx2VP>HNE^PL3*A#AUGAY_>2C{Fd2
zVxARFYnaw$cPq$YOofjVz?)3anTo`E=ID3l{MJ{WMc;peOS>~pmZzMb9`gFhn$wh7
zj~f9*fn^ZMeEJr{%C|^Y_ydsov3^r*YJ`XXE*-tnV%P!-7ww{ANL<Wfjahd5W$y!Q
z70VZg-YJDBM*sl;07*naRM>MAbRv%)UXYS!=vPdqj-6cvNtafmEEBVt<@D^BfA|mo
z0T1pS@)!T&UuLnlM=X3hw5D#F4U!xCp1vPM(djBd7nJ4`pZEm-(ck^M{6GGS|ALc~
zlg$F~XJKTF6fA&JrYt&H5>zWg<=kfXpL`$=LpcL7SwTo0VUVb-D_5?tGu@%ZupG|O
zIHt`MLq)stEKggkbGXW)tzlSuv4x2dhr%d>%Zh3`XXYwieEtP=i0Iu&Q#RMN)tH>X
zM+V<Ao6c#bQ<~Wn(=@Dyj#byOIzM538qvkxDcOqi`5wdUfYb;P-2=;hU~m>YpHlD5
zxc%TZlgs<;O|Gz99TODP(;Y7DAF%2IZ5L6!LNTzjcbT)}doZ!+nWakuIct)S=wzAI
z6M8?83*Ew`Ot)OqSC%$g+LIH^WQKN*U=)X4hiWFwW>e<QNCvOgOq-gW>69ybyHqOB
zo*%P1JLBZ=i1pbSeb+-uT;8v^e*HRo2bY*H7R-0%Bw*dHI5|G#<rkjg(xpr6Ef!2B
zjkFbm6xOj6A}L3ltEt_DSZrmqs;R3v^Vu%P$H!zG=!CYQhL%JkBu_WA^qr^gi|SG<
zrjt3w8cxs7I6S_^5MZ{m;Na3F_7C<MiYUWi&_=WB*Z3I7N?|M6aG))@2udDmV<HfR
ztYf1JJE9Pa7(w4XVc4cy7^B2yNdrjHmy3#Ev?_5Fk5QGeO4GJ2pmDBZHl5SAy=(xj
zE`pDeK2heUq3cLdgou;LM7WG4!Nw^qqB*sdcBoMVp+xaKSB}e*9aLKJwa<Tn_kF~n
zX1ln0MvfiE)bv4|FrbzcAXR*7k*xCQBXRJOfVCF?x|7p$(2k^fd`!YMRO8+)y=95@
zhnYbh{`u|BZJDti`8R)zw=JHCTeoT61f41=xl<0LIG{pDMSw6ctd^)0@Lgnx#oF1-
zXq+M&i&7TrDyq7n&#*IJV4NldF-9C>pbwFp49+?#Wu=8zP9AG4)`_~7A~V<4;FK0|
zgfi0R$Z4~<$Yd>uomK^z*`Ojth8-fo4?spWV60)%Oh_JD-xK_R_dS)-%qBHfK}3lo
zHW3t9*AQ|h8<@_fEY*Nwiq~-M#ucvJ*rzcIbUq;|#m=53y}Fb*9+e3`62gG8nw&Ku
z^l0Pc7-n0{KPBDc>G!;wu3PcRpZF=h^m||9+LdeUOs?_9onzL^mYg!@$AP1JOZtAz
zkO!y)La+2L=R<L;kBQ(1x~^j~Y0APMH(vt#n^J6KDdx2iFCeXHc6LZw6LT(wn-t3`
zC5Cp*?qUaP6y5oGIfcuKo3j{NCfTt@mC#DD*qKR7FT$!FSoZ_7S|$Xe6}nPXR^CUQ
z6v2nGGnGQd7(+(U4~m(avz+hoJFnlt&1&k!f>-anf$;;i$@JQBmK1}EQc)!d*s^32
zn#71!zsa+M{O~^&C1&TX#~8trxZtt0`6+J>M;=k?e9;K>;;qIBQTVOA>6-jj&YIr7
z+LE0kR|2G<=wl)Tu}hiHJ=0me`MiyGI8#%(hEIR`H#j;x;jjLCe~s%mZV*xw?wx|V
zX~e#0y)HgsNurUICMC~oHsi1V`+uD`?%d(O{;&UQmh08GIZA)}!xiollm$l1@zWH;
z*$_kptunbRh$9*iV<6=~43XaV*g_@o{lL%-q<O*`hn`p#i$vEwz!^tnYnp0;YidCi
zLZBZ8e2RoYHj#bTF`qWX;5j)yVzoTy@}*1c?CcZ*I#kx6NRT`#c`Dsd%_hty4gF4F
zxms~@cEWNkVdvqx71E`(=(=HlXNS4g%$7^;pPq8}^n|Z{?aP?TvAe&In=Yu0LdQtV
zJt}LOszxh`U(7jh_l@VVjia7TsIy@hJl+R<-;<-nOj&2Cj3y0Ul)=uCLg3kFzsapP
z?qYn$-lXQqYuCAY?HcWRN#C|C<`XXO?K7=wtk$Hy<?!AeP9Hqr{N$X~Y0ts_gx#He
zX0sVru3qEPr7Oe){opx#aKh2i5hrISgy^|^`3j50ocVlCQ#|APzNdsj2!f+3_I4$!
zb-iBEtp`#pCdM%_+TT#VH&by==zB2}&Izj0l6J15?+4D$SDc(IY5R<;p=xSuRfDz|
z>CO{#!utUqhr;<R*Zu`X>ZlAzl(3kKlTF^7x{Py69_N>QK|}#%4!CgFQq*k5D9!*`
zI}43K-n(HFTUPjS7sVV<(j>IrBG`pSRauL3qI^|F&7!p^9#Wx2B{b7iGcgvIB1gx^
z+<W77>Vtu5FEjIfF%h>#@T7|ovqp`Z*d)ki^b$oEN-UdC(s&)`*zK`;c#KG5-fXkq
z>J}IKHGkBN>L2|v-d15}V!)!X%0Y?ll0zR+ITl_GRG~-pOPbVESF)?H+LB@*M^7Jn
zPKOm|95Fw*Ol4+Fbb}fqp<AJKBF9V^JjOWk5QsxZ)=FYOW;2Em7-Gh0i?KD<)F@>k
z3q+05C<`W+sPhX&2^w-yt!zn~NiL`u<>xHq(dYwx>L4ajs5yhf()BH&AJ8FVP*l#z
zn-8h*(h^Z=0ve+lau2m}h|zCLUGMVZiw~GJ6?=<a`XR8pf0@hbI{P<geB|S2v}X@E
zd(d*)Sx)+aj7Pa7Ji)HTMDg$QVa44$5BU7&zQp@K@E#`9IUoGK53;-1rLG;TwnB4>
zA>Cp*tQpYsDWR)~vzk0;X>oRe*hfs$kPC@l=}bz3OoU;O!We~)kW411L?Ym_B8Nn3
zCe*u^A!};y8B(U#8i=?!5YN}lCp+Xg5W8by&ctZRDx;zYr3u;*QJCb>80y6i*i2r-
za-CRqiT!ybJ8-FDlE77#WDIp$feCmf3{f#P3R@XErkFHmH7L%#V%2vvwdL@^8yFoK
zQl@Ge+6a9yxXe*_D+RA97xq|Ojo;sN0p_>-@h!#{N!}``3J^aQbd2-m#piwNM-?hT
z`~&BMbm4I>mWDa!jSYzplnF2sydp0wN_8kKlgSJpT9itB{TpBAe0{=S{&)Tg?|$m(
zQU^=x-8K`9wshT+OvcB6R<bb~yyxoWEBw#?!~c*Mo_mg8`sH6nX^3Tqugb*!-5A=4
z`Rb;(rDU-(Bl<6<XF?g!R$7Eg=98R8(U0(m+EzG|u_W0@KIqnY-x@1Ecm@5|O0sxU
z7P?C;mJmX+Z-s)<53Jgrc(!CVo3XQ4h*4}=EJ4h3hrUN!M?Ia;lms0+s`%TM6g*wq
zlC=oI)F3U=YBr~Fj)TPxS!;r|R2rFYnklheGpK=pCEE$ssdC;0Ac+W}q!Z*g@(9Os
zIF~F?nUh!N=Y(OPTdi5Hm%Q-J=a@|=T%H|peSeon4=&NJ&#^``+g&i7)MCTr#b9=Q
zddlkLgp?w)*@WHME;k>!f-M$TRaJ3%ddm6visfp}vTbES51y(r?CtF_oi<DhilnvP
zDpv`h#5B07Dy-dl&TUL~5lV!y7F&48W8w#l?RT;2s_O|=)iC%o`rgyF1H%A|of(tK
zjH+tnGsMW|AbN(OFKXCrn{nF;9j#Tie7L+xhI#Y2Fqt46jf>o~(f4lSe4sAehb?yv
zq$_YS;bO}bE%??2s=~t$z~<P<xqgA0H(Iw8lnp(4;VE?1IA?HHu9L29IXOONr=Ftc
zd(vKx5wzVLm#oH|b167SggtDc#6*xXDvd{^bKA!Kw|~5?!mid(71O|wmH1UDS(Y>+
zC)P3Jh7R2>vA#vSii9P^h>n?BYcU&cblK&%XdjtWiFm$bb$&=rkxR`U$p^G?!XxN~
zI65To#!^ow4AUub2xMh&wnke^7$a1aB9TND3-~O*PdSTf7C8Y`;ZA5%A!lGf6H8WV
zCTl|qfi6ZuEtV42NLxWgMV|vfCv4@w8gx|HtjNYOC?t9)SYoT0H**#T6Vf`fK0U_|
z4Zrc(7ePf_1KPdFNBc`CGw|*YT;pfH?}z#PzxWL1%DefOzw;&Dc;ke1Uecx-qXXzb
zKLm_+_+Ihp-}oX&M@xR-2i}9R73;3!e7)j#fA^c5pRf3_AA6EiMb5fR$PHHYI0xNo
zO-MaNMe>DkfS9W09Eo8NRah}{w90@lfR6!8LWO}US%RH1nH`{$q7I&zVQn;)0UbJq
z(={q-6dp$lF%qK&l_2zFXYsZnCXF33B#oU;h$f*E3|`ZE&5$C&M(Vm@hlZ>YQ73Ax
zvD&eY70qOU&WUU?QO$U8T5)F`In0reHLLSuo_zEQ>*bnT{VC@|=G1%1FHEwyPMay`
z{v2VYWOz$*?YAG_Zk~Af_vt(Q?PAIIR=0S&Gtsxc{<jh@<fK`zBg3#FctKp9tynJ4
zfry$fsOpM4cV6co{kQ*sPk!>J`OzQ!D0MyEEEw83OmbV!65?wm0pFH)JoY&M;ott-
zeDm3F@XD*N;>Oew{!LmSe>`KfW+>ZuDeTIEs*Rkh;>}gsJ*mVK<_u94?w0UThklSn
zw=ACCd%C`pAp00LB8`lo-l@3UE7Zo`24rFChk?anE_tEevtG8eeJ3bPA>Es7$Vw7I
z`o1SCjd6y?iKU9xn)Bs^b=%PovKbe@!$fQoecv(<5mVR9stR{dkzB>p7>4sR>Uv5D
znIUUDhGxEx(S{roDG1NCt{j!MSZxS#3~1kaz2&l?lQUh{(yu#KKzn+`y*szLcKH(T
zxO$oE`+F>$<C))om8)0w*;_1dMv-HnTP|6jpL2eEOb!F{X2N_r<zTVk`qk_BLAnjg
zRm;8mha4RrvtD;Z6pO`z$)sjJpR%{NyGdsmtwToR@3F`+CG7qoh3N&}+!(VJg4hf_
zEMfqo9hIx8s)oun<g{eyBYhVLUNN2SFrCgMt)hJHDM(CX*L8vfr&y>N7ljMo>LE7@
zk-i0`&)dTK;-4P!y?d+sd>iy?H0Zuqn2o~2oJ$KfZ;#Q%P3q=xF244Wd$`$8m7)SA
zR#_#t*&0Lep0l$v+}>qoWs{Y%P<A$S!3&LxQT%nWu=B&fP}Ig_fu73?{T`|7-li&c
zjN>Z1mY8xei`1yBOD^0%=zFy9alu0_xp^^SQlc`#1G3ICsiwkHL!fJev`edECz7>7
zS+h|Rg%Aj9Q7X-McDQl#QHFky3cs>K7*?e%tW}9Z5jl4%A;cHeK&tFuvXY5a5&)!K
zpcE);FbHWlgh=w9VF)yJgEb;K?|m#>EZGGaYse{LbxrVMJlVH_Jkgw-EV=oPM_C`;
zVYyt9)QZp>7K=ydx+TV(v3I%Wx#wQwUGMl9!*azdU;J&J_~DxUs^cSn@uQsTbIw%c
zi}$}m&<5ovsHhpdr>YcR|Jt+MzI~e;H?DF2{yo~Z<M?>RrK<;g==*+vFa7Q}x%KKf
zO3jGjl&UhsxWZMDE_)zvjGJT002LG5UXI^5af}<gvdhXTQB^g%nJ}Hr3QsH+!^%X?
zku>zQ>lLX=)OCfbD%8+(f!H7gfXc<^D4{Y`bu9&X5|KdHd4|D&)6{hXuEJ!G&F2hx
zPErl?YzfhkCYSIk)14gh%KiJeOE<Xn>T9g~o)7=PhnUYMeEsX+pspRuAxrDNOwiW8
z?I`8%iJ=)0fsFTo)3bA-Smd}NDJx|V-K*i~=!jqXrC*|Ldw%?*A7`<<S5Ux&t15<)
zhpE*zs61!IWHR9sKlus%=HK}D`G5YG|0ToV(Yhe+-^DSutS<gL7MFg#as8LlQz>cq
zn4GCp&5%4gOYt0Iq^hcdMrc9^tk-K4qG)!7uNOif#Khn|rif8e$sW~4)Ax2mb#YaV
zs|R%7vhG?2Kj5q@Vw!}pmYC$R+J3;tz<lt`b|n65Z@z=omYun$o-M#Q)~#f5-+l0a
zv*Sam$%J}7XS&#9wy0R&zsE~Qr(C^xj+rmOO|Z?3Su;b&0h?iB74BfcgS}n4)d}5t
zh4*VvS)!q8O<gr&2~^bei`ks@`5C%fqlcEKo_vy<d%JWeN4)mZOZd}My7N=!wV|yo
zt&pCH!vG<$Go8@59p;lMv&n>=>6BZqzs~vjIjhx*v-5M7>z174xLv+_g_}2T(o7m$
zWoew8<JOo>p)o2a1+|mHPS71TRwBTLp0eRv7Ot=lo~D^l*A11c0fR9WF~Hgntk*qV
z=kdiYV1NHII{*<u?!NO~>Z&Q7BH^if-_v$05pcu^S#MOLBT}|4jK&l7I~OJ!@l`24
zw`Rs~D)7d3YB~GIR0fHc94$)p7R@x4tWt5uh+=6~Xisvj*v;A+Q$p<4N8Pb7m*uJC
z99I|YoW?I}G?TibvWB7WAV(@EbfA<Y5+ha~aRV62ni6s#M2`;x!_d(U9ev+23_Wr5
z)O-`j_;34odmOu=M`t<FVuGCYrpiuH+MPKJV%y+-afU9@CuVeN*TftpT1a_vsrvm8
zm^Kq3ST+j@rq&8&Eh;69vr=J)NVi%`yK1q6n>KPnm7rrEg7nx+w220COblI*?*`d%
z<h*g+rjSBI&V-yK_&!GAKoycywg#=FB@&8fQBya7qM1y@R@aDnPHSnyN{?TX2&&|$
z%#>GNy2U$w;)l6>W6kY5hg`j}%e$U@g4=JLFx**Tv|&EkBl(W~D+<*<K>IUnJmDih
z^ls`|#m9f-`w8;~ectlLue^d)4cGTCGn+L$`Oce6rVfZa_QW;r-9E-S&2RnIZ{u9z
znP<PiYcD5sTHr&=V$Wf8#_B|jYjTblGaBIvhjA=whrvrr0I_2b@1qflw-^h}WJXsv
z%yxFftax<6R~cfWTdnEaHEy;e=FeIa)~#?^N_!)wNKz$`IVCiidNLu6RuEANyRM7)
z%3__t>6%(8rrB|vDpu?=MWbEAkv`zrS6}7y!EIhWY3Y_7ow9u7qaS7eV3)6a>G!yQ
z?*vmdq$K2;vG5;vc>jGHnQ{)79dHPl^Yb;{4<&WNpp+#gP4J1yWJXHxOaJ0ukTZPZ
z6Q8868bKA1-F6z4rAdm#s%~(=&dv@$`?EjGum8rc^M&905?Tvaa5Le4SKfVQLsu+&
zj+7-2)HsW_az2L;$g0pjjKyh+XId`zDf~hs9Aaz5)K#f%8+I5xAxe5p%n@V63L=9<
zM=9a<*3*Wrkk9ws9+PQ9j+y0pP1DF6m{q2#t4$j_4T7|<R;9?SYhq3~rJ2@_uIt#_
zpE7SM_GS%@fz^7Aj}d=z%$<Ap=`&Q*Ig`aM?db`zYRI-h)l;Simod{hu9?xy=3Kdc
z4WlB?XzI!lV}n+KkP<0NvnvJ97-fZ346Byop(U-)QG@6B_8lI){u)Pj?$Mtnu3t{P
z{^HBrdG%GQ3U($_CUq_Isx|no<7m0!Fh-&u==*Jg$z<Aa>B?p5dctHnWwC#Wot?RK
zb#lTQ$q&ytZ|t19uG>U+S+iAuD#gY>ZY*Hhw%ugGh9IHQm6MEDK>;Vk0Kq4Qp`~qm
z?%zA+_~@Kr@a)W{>@E(NOm;9v6T(1NzVQ4y`o7z2a5WVBEmy8T{qQNP#tq50LEvCh
z)MeOm_Skk#FWzTFTBXn|$wHD9x>3aDl5K0PAw{*Jt)whG?Nm6yF-1(_?Tsy5v?#X?
z#lm-N;c8XR|IFwzl~cqS!*tRxolRKfNQ#~my>Q^Qcp)mS2{9JUMlM3Jw6Qb`$>U>S
z2%f>0+}#VbWyKa<{I`9)t-`K7JR~jl(G!PKcH4W;qSDL3lZTEtwBS48gca_Owgwdq
zA^OrfFc0l|G-4ina(v9e!G3{PMQNjwp_(-K;E7$2N-(TjbW;<Zp)wVE+_5@K4kA(C
zm^y}7B<qMX{`iK*Gg@Ur7Si(IJt250Qj2k+)y0!2CD4hs>*+gBT{Z0N?$OM4aCIYw
z$SRUmM#oH3TV{=>89)ca?&2CJCr7;NsYkef^ngh{;gf&<<IHA`JGbuf#m_&(D=*yT
zk!v@Y&jNRE-(~;iRYDcGeq+JczwuR`GEV`XPki!+dGsCc;63ko8kIbIdkgjs7Mzt<
zcFch`w0!VGALNDCe}_;1+OPA4-}*9-UHK^vF8>gfZaFxp@#kmElBHX<OzIlzjJ)sS
zp*2dxyLOQ1m691FlY~NJsH|gFH%zBH_@<$r&rwFw4>C&BVmZ;bYf_9%rxUC(^fA)+
zo!FzqBvc(UW(Xqakjgk$lf(mnm=xY8hCVY43R?%PiBwt>q9sHFwq`IhT3_>-FF((n
zlVebk!y#g)mXH4APqMRriO+rJw|V8|SA=KbgNQu}@~X9dOBeV5K8$g<n^VRVWGW>X
zhQQ&26LNxKNc4SRcacgV7MV<@{Ng|P1#(LK%+LG`);eifI*STo9U&-#2oc5r^LM@b
z-TbY;^|$!ubI%Jp_}$z)D=r9N)P`=PvT)!w+$5DmT#=Kg2(?0GnI~*=R5=nuL3BtM
zBdhPYDC**mXiRbc-;m*T**qmtXlC(ww9bm(pqW!Shjo_o(<NO&Lj@U>&d4?@N@GfP
zYz~n?FL7XHW7{rUW|Ik3<!~Bynwq*i#^3{lC1JIuZ97&cCmh|s50f40`7X`wK4`^o
zdO{yF0Y&}Xlo&nB^D~CN7oHtJi1;82y(w*Yt049R*%+2b$1ES*#h)IrP@rO9F|Fx_
z4iz&AUmOi!PzoIb*&3oTC1R@hYAVJ4-a>>v*3wKSq9ip|C^Sx$0!U1yVe9*rQ`+RQ
zjzu2kf-fA~_Th-KjuBs4YboUFs&EV)MmxqXj`xwS^DLJw%jKGO?a{`vSnM&GOmWts
zRLL<-;esi#7eQ2JBP`gWbq^O}TTeRP;&&I2xebb@FWhFl*S3hwY+JwMeRUddueNCt
zBS$W6kI_bMaIqCzIy|QEA9cyo%|*!%6gx8c%4+<G<c`KDDrc#yinBfxL6Q`}YBDJ*
zUnOhQ1-xB&%&C~gi@+(yu<1x`-}(6e?~TgtZ54LQ+qXyumoc_N>k4eWIUp$~V&9{~
zfL|@i?HV)H;=fh2e=&=Bp;E;OTT+t&AuSJnz*QAZU9%qA5-6_7TE1Le7dSyCh?KJ+
zbSZ65G*r1?u^P-`^60%-380F(YY>x0ZG<!&iU*&xA?4yqvB<@hqIMOvtHmZoYl4pq
zs}(UT4z66qRkf&XqIm16&|~PC*p|w7yz}N`{McXm0DZd8^Uwc2ANqkOc>b9q{=a|!
z7kK);H~8Tnc@KZ*fBjcEda&eoKmR%Y=|BA?e)Rj_$>Tr%qnu{V{N`1jo_?71x+7%E
zw5s^?Kl3pr^M+wq(oAcV0XJ<hT66pM9X|ZS?<a-8<4@dV)eS@i^WBPSa>m@QsI2Dn
z@R%p2H|Xv>%g#=Pbq1v*csl1y-w&Lfp40b||C^I|!#OSfey)U8H}w=-Pq6ij)J%$!
zucfML2nmy6d3Fk8VdpFXh$YiEhe*=W29&f5f!I@w(%MCXGQ_0u9@;+8_L;0(_GT5<
zRkUi4?0blTlicvNSC4t=jXTuyhNDoS_Z&a+lRwJN!2w_X%2#>e`4{kgKxOC$F}&PJ
zBhq&ZG5WJLMnQu?5ksbnfuo}{yiesCf)FC}`A%uGI{w)||3xy1Pk!>Jm=xFfT<91o
zm+N!nsTqSd6@TTg{1yKBzxWsY%U}65M*E`g#t@|9;?G6>Us_?A!kJT`l5F-AprWAv
z);Vec9q87rAfGu9(VN!3Qu1>Q(&|%MiW3njtHP-gtBK8`YIITbOSZ1?zE<=@zbQDJ
zbA-Yhie*l;T4RfSODc<59H}&YCtoUIl%iIOrl|{p6~PDc&~v(MIc?8arw63ToZ0>*
zc7jJ&mNvoZy5n@UrX4(2uU^9!7lYZ(99>sr6&PL3jIL-Zhf^?|oMCdNTP`_y@PKf7
zjCGcs+A!al;)!GrRi&t$!WnQ{OShnG+U*#xRoa5PJ3As~aE`j35J(I@$lRWSbYxO2
zWch937Gr#<u-hacltS(TmF*!iJ_wRGB5AgA#FA?p<Jj#8bZyUS+0w3iyw6Olirw9P
z>S}`476_$a+!Fg!k}%uMHs;cpE#D6wF78IOdHXXD-EM=5z4?B5d){x8EGQcrwY?A9
zzg^_}l>}IF{#K*mi|yXceT<QmC|$%7Bgby@BG{mil>Ls;8f!GQbL4d|Jh`r8=sW6Z
zW-AiTBj&#OZ(P^Yh}@OpZWF1-_I-bk)OByKuseDAMb7pP*sE&n&OW{vTV{|pjqga^
zieYumP70zNE2o3NvW35vWS5(PoP>u{RgP}mF=-|;*`ugjg>}-p2ue;GXC-2XoJmSc
zmf%QOEhOiSGE2K+e<Fd^zNp1erdR~X?NbJ)Es9JCMQv2RLX4tn>jz>J(a4S?MJY;F
z>y~P@B55%!*2YjdgGS?hi{X$(v!K(KH(oks_e$pVSMG52+LOHVosaXvb1(9HzxNex
z-Fk_i_{2SKK7PQT|EoVv@E_-&{L_C-dvwJA=zsn{#qG?f<}>!M9#GAusGSxwX$T5p
zG@PBZXbq>Qr?{%(`i*z;+n@d{pZ<;C;x~WucewTX1K#((52KRDHs@S>WWn*BJGf9$
zC&juwVtUmUMXxVz>q6jNE?2@aEU-LD3h}zCshf$UUAP&_POwdlvQ`R1$s&LlQ908s
zS7;5h*;Eu{*2u0sMv3bR;-e%eg%UJm44gNnM2uScA<}gNU1QL8K;?+D6`kEBw)YtZ
z&+TF0i!Z*$WIkbFV68PD{pb&H{n49z?He!hr7wP&+z;YrKTcfgdq)-ZM;{yF4!{LM
zc%W-NDJ_Ab{*I+kxWF6x=l}elqP5}^KmL<UrZb}VOlNbJXXk}{T~vcw6B9i0=p+1%
zzxg-$!tZ{Oqmz?D1Nd$XLE%Q^VbhA!nv3lsg~mE3-G{Qs$K-Lg##W9hil3IW>Jln>
zL93MT5{=#x?*#!Y?f)2Ma}kA?nv+;xgcQ(L(TAa+atRFhFpvv(>GI{v^j)__oU%Or
zSm2GmWR%#a%lmuGr!^`keBUuF&u}Am%2?dIVNp4TOPS^1dG?i8dF{?Ej_=(?zxE1y
zcW-j}vB%h%@36SKN9!Xer>FEGqO7GpIKVX%nw0RRC{<{tm7_5Rv)F~zl2ujFRW;S_
zE_L6L)-8P)m{`N5>z9~KD(c$MRF<Zyi}*%k6j)udU7e$@Dvpnj(Iq{^7*#eV!d+7Z
z`C&u-D4e;%=Nj9vp~T@`jP)AZ$0EKf{Qh#AjW?T<<+SU>biMC9U7I*RTXA~2;`sQK
zzW3Bs#qMI4%a^X<T)h!mkoAV#?}Oi3OHpIYMYK+?&5?hm^j3&4wp8D09=m97^3WJJ
zGUYKwIz&=pRg%kX)P>t!6hCaD4W$HSL@GDkII~5Of$jU)6pC9dZ!TSvi^C|*$mngH
z6OLX8q8c7NQz3Lqq6MndhHBG&83h(0M||}7;PEL4(6o7P<BI>i(WQA?h27!v&#^aO
zP%n0I)}o98DiRqVdZHiDeNVqyvWF<(wAw<!WHxa!BN=%7K(*F%T}NG4auWDZ42|;!
zniZ~_gvm!A$RQ942TE6xHes7aPR5aZufR<jnQSKwT`NJ(z8{JKrI;a(<b93W`0i*V
zfdM5aK;I9L!C*>TFwypF`aAcScLP_hUBkHwog+GXtXeX0$JlO%{&dRa`3F#(@ZKjs
z!e>7FBA0HQ@sS^Snirpco#&r_l^0&P%cW~O+&g-mr=I?Pe)ezwZC-o+MXo&d2+1f`
z{)ikDZR_cd4hcCBT|*2qNzLX{b{AK9;e{9Ztxtc6|JQ%@|K{GkBX)K#^W=LzKy4IP
z_pUOV?Q;L#J@TsNN__zRF($uG(^SQkJZ*R*=jZ3NZC4IY;YBEL1s1RChDklaXh+Z*
z?W&TI={JtwSqr)K=<oqeT{D?9Bs@lILf@gYpbKOBTNWVk@DejRT?n`ew6uz2lpt_p
zEUup5>M2$`qIRre=J%ibCOyLUIa#mxkxzb{cRcwxZ`{7a*T4EThIPlxiDP|C6x$Z_
zy|Q8ZV;r)5_C8{?q$B|NA#!qZ%4q*Ewm5cocNqpn2m?R=^Pd7A`RSke3qnHgIwq3|
zzJ#JnvWxJBVha4ZKlkVO)TchhFaOG~e%Icun5)XhbF{yZP5_$(icL$ka+qwzNk15d
zLDBcE6ao4|l$5$q7{i7bJ|cLV-;Il5+_uZK)g<5JhaTq~eb;X~3QbiLfjD?V5U)U2
zY&qJt#dlI97$tTs3I!pGkSs^E)#wxiWzh<oBe5IE3X&F&vK{AWhd|1Sn+JRB%;%i>
zz<CIq9p1$`&EoPUrn|e;%5r&chao^zhQ-bflf@3%h+0=E!_f6uGBwD&(Ny#gc4)2P
z=#2-=i7YC|Vm{~c-h$Y#sI6vJ*HlhZ*N)m%IBTh_yI{8=Ji_U8DjdNO==zR!=!LNF
zELG#s1zbqwcw1|WdHps^G^Hd-8EoyZMwDudoEuTPjcCGqi4+Tot`pkI&?nZbp2MRP
zj*recI$U9$W`BQ&ot*`<+4Lc0sM!1zFVPE1>e^HfBV?slx-Hb;O|{T>IKjUau~QeS
zw$X=fERe{ex*lP@O#w6(k~wd<q_V(U;oy&7DPc-(`zWT^kUhrNQV{T*DVf}9%lXUW
zHJMQv>dH}DC&D7F8O=gN2t+d4H05xV_fXzuLipPu37>VeEgG-0w+dYNw|Ts+!tU(W
zYrH<4(OkX8qN!=>3R?>ScnCc?1)SCB6tPC^0JN4Us1YDV6?I(M4VZF~D5Wvl&`c(>
z8$-PB##Zl`Dv`#Pg#tpNMxzRELzlu&3P)XHSwxM4R@kOuR#~Po({~+x*D(wO(GSu-
z9eDt%n8qql#-b}ZO?;HeJ>-Nggn*bb$A@REy1-(8AL|_14^%nf)SAYgvoi@i7+S(A
zGHGgdrh81Q1-D*&gOkIrar5yjeB?*Ik4GN8%=PP!^VpM@v2|pc8>YJA=#4u>m+>cS
z4i68RUfySBGtI=|RpR96jF(@2i7$NN3%vNki=3RC5kltD)ji(v#N)W>F7w?hT-urN
z+Ov0g`Q_UfH6a{o-nF>O^7dC~n%gXPp4@n0fpAXF&dy6~qBxyb3hNYgQ!|}Tsi#v(
z=ExNxX^az|Ne~$sQ3bG>7(91x-{$h=OL8F<D1I13(4w^xC;t$Ls(3dQq%Q+X8<aMr
ztOx-HpSP(9Zi23AOy81h!>#jkPR>ubcIhfF-@nKEe&|EI>z!9QJw4(pU;YYr-ndI+
zENSql3`T=7PI4oIa6<ow0dHO{a#kbwv6;}%&(9h2A48Bl&Aq)ntg|eak)QwhpQo;;
z{P@Q{hSsG8tJFq>k;^rgQsTyq8~p6g{w$yU{1-SrKKX9EVKD^Oxp3Yn3s}C;qS_*v
zG0TFJHs`3g@rzAR7G9Tia!yB4np?A7Og3$*R5(CojvUw4F@JRAdd4scGm8IaK|NLB
z7p>My_7-~^JCWpNb2K2R$E2B%tu3#`<NJYbwW4wkgJM!w45=f>K#l=p9kr>5I^&4U
zoRR!zqrfY&Ni;g4eMeX+)&aEJrHc@h;ryiGY_-B8v>0tI-D-`?P?4#I$ixpE-@i}4
zTA@=SM~`bN8YdosO=YQ_!x~NW14GVOrHZ+&+wvYuF@RDV8=06QLq8Cal$sDd-is14
zMX{C{hJjE}vXl}{(*SG`vQV4=MowGPG@BgQhl{!ZXoextb_0DU3fQyLHK!*_*6W^}
z6IIpFG!2tULsf|pwjf(x42iubcrTxI8olebO4#jYVOv<0)3FqM7ex^w3K_Yc<NadE
zHr}9aZ>njeo_e@i94%|cw)BXq#Td6KL}_beA7UsdcEZ{Vg<knWmo{?<#aW{OJ?UXh
zy3`Pcq#{NsoH5L%QxUL~wZrIQYMe(t<n}$KBIX$1Q_fk6yOcx_r6ir`Vw~gm!gyPS
zU0nCvdFdsp%a?g#K9iPU<p|wC@R_U(bu*zdnhL12xIc@wF>RxfN=r{GjVmXZ<i}}}
zXVZpJMqH%DibRyEqbF8IEK{6u+kU+@BStQ0d&;HdD<`;Q%L#W&Tf<~F!__su>xf;8
z_lo2_VT2M(@umP0s3>Iqtd)2_1xcmwQL!F0%2b=(U*#g?6{<Vtzy*$d=J!AQyL{~5
z{b8=&*k^aou{^s%II}$W)q8aLOFaF7>x6tlHb2VlV$S-sV>+4Ow4$yp>e7OkGH1g%
zzyHj$Jo~~keB;>{c<$L(I66Efdd2?U72f;w`#9LYgmsqLY(nqh)}0gXzVtd%vyV;_
zhI7r;`hc17oSZ$+rGrG($POq3F;@*SNsLbtcl{JIHY@7N(99e=iz$=&u6SgHC<toG
zkTod=VhU)i@Q2zptD^@zapkhu(x5O2!qAe_AfIzkxR6O&xGSpcxKaqxQ8x<Z#Wh_-
z&{EWAZ9rK}?1|L8esl+2S?-_SW%tpmyz_le<NGzwf8zz-c=fiBxk*@6ESFFM^{0r=
z>3al>|D%j=ZG(>1r#X$L-cp#fZ6L<w#<W!%A$Ct^Q<}PFna=sgzwjw`_jdW<2R<Y`
zD`oI$kcA=(@6H;B4}qWh<WKRDAN)an<zM|8d>6QO!Fz%MZ3UG_E5w5QLZ<-=2ehCS
zg`bO7xXO`}DdG@^a}^}ydWh*E?$$<#Q7mLOHcN$Tnqw*)VrdzUi7KLwv}JA2g}0<w
zt)#HCR($J>Y4QDlN8yaaOf}e15T)q5o=z)N(_pKLX;Y(v#}|^hP8thXm9Z*Q8^u6o
zk|kPCSxaRUDh|YTq-{N>)~sWuPcY=f`OvWr0iP4DuIZL5T!xyA?L0dv(ml9Oe|Cn;
z8I==OgEdlQ&vt66;!4mB12F`gH8gc4#f~bO;N|m6N%)AyT67d%b<7c^6)|{*O<^Y{
z)4mi@F^I5dT8ai$>iUb~g0ZkurP#k{calrLvg<PIbx*q%3d_mKlGSox7^EN?i@Lh5
zv36@H9ipTd4J8*igjmuS1n?L|2is!oLP2(+2D>Pj7)2gqOZOr&_g1RehsFg&aO2IV
z3(u1+&o1Y@X$6nhLrmKOE){foM8f4YNjP?qs>#)*g=@NbUZpl5i_(UqL<nR`>)R`Z
zA2;SD`6Hq5V_{bgNGjK9zVI1t6#65YSCGN)rSbNB>~Q3D$GvYp%l^SWyYm@JIaGF-
zT$7sx)(%*sX-c?pj1a8@G&Xy5O!)EvrYwe}N@bjrlQzTvMl+~fPJH3<P>KtsB;PLU
zOg5Tcfi(s<wl&Iyn1b+LL==$<hanepQE|M{fU_3cG$=IQRubWrI!sleawZnED*6G{
zMv@D3wZ&OWT`7hk67__mbIWQ_*g9dGnqbe+E^?qW?d=ELtRExzK-0|m^zVP2n~y(A
ze}13CdjpleLFF7Lx6kRHNc`?+zs&dl!1r<W>Sd<0hF4#CiQCV-iXt(a&-hpW>eu-1
z{?X6VwVwUmC#lpmuFXHhytzWZ*4%kUap*KLubAz0_@O1H$Y1+we}zj2kMPakf0fVw
z%g=Cq+R;AvGK$yOzkCT(Ig*a7)@ueIS+CCtKBJ<N<PBq)Xv=hJneJJ3<_*ryNm_$m
zqEb)PkZn!u20AiWMKcVz<r3R1*?HuEpoK&&_CqbiHHis|DN0r3>}5@jnjAcF=$KZD
zH6Dc(iF&eBZh|!xDh%K~=snuiny$;78_TJQ{M3(qKUL*;<+&qXeC9UoS&vl;pE3hw
z5j7eU6Lku-e6O%9{;1<yc{rneK`zeJMv3wG&?lC3Y$nr`Hne3|2VT4VGXL-&{R0-W
z9iDjXNg(62CZ-@rNK`ls7Q>axSNY3-=`Zn_&wQ43*Otfo(<|^&h~#LBsH+rVNxUu}
zN)u9J#L@*feJu*qvGr<;9Rx|(rb=yjKNrP08#M4DCusEWioUevwuY~Yt&#HhyOI2V
zc6KK1#e&KhD@DPGiaKqeuE}JUNB6N#Q#*^sQCo{{8V05CJu*l84xPk*w2t6YB9Mqk
z!DM5P@|mbD&PPlQxeu&XEq9KNxqW&_8$3xVraN=G)fy8bRfaufIFFuqdP-g{*>~Wa
zKmj=flFDdZ1Cm6Pvxx3S!{=>*Q;?{X==z}`Kw^C*?|l^Pu#H7brHiN3ra&B{<b?kP
zBlByEpl&&8Bb8t*oKlp$<@2Ryz3NNId{6N5`w>X18^g|G%4}Yvv<M4R5(TVqn=iI4
z6;M>CYK-wJ2HF>BOmF79<(!2AGrnFma?aR9l8yYqKh$`W@Zvk%kPu_Bx7l*)E+UOF
zMxqx>8Xd(mG*wx6)T5AcRNCroiKdm3!cJMNN=V|o5?ix}UX~oQ?ayjsn-a4G6Xc}6
z=L)$ukGEOaDQt%R_=G#pJ;%=OKH4?J=`Kz?R9$2GnyfWZNxYCuR%O4RMICF6MO%v+
zw`N)qNc7QT>q;y&ih71D%2kB|Jx++_f-9<8S1!yD0-*#h#}J6#ixx|>xhO_z<|xmm
zOtz+~uyw=Ha>>!@8H=5{+%7A$GOT;ga2^?I@w{_JQEQ8GkW@uGI1bM}dwV-HjlpVB
zdQRpO_7{qwKf(sj*T3);{_<b>37mpTeT}bt;WgT|Ci-2{y5hBGm;BI=KEk4z({(HU
z(SQ37SuA$>$VWcHmCKhIdc_+r|7%QqCpWKNXI9;SG$VLV*Fs&1%GXsjdC&~KrR@iP
z^kYwP{n1^n9qjV;-#%rp3M||iw~xNbwME75{sQM3e2lEyH7O<9uEm!&wKh;yx^Uzw
zrn8B-rE5bhkpD0YsA6G}LLg)qh5;7_&X13nRF1lG7^TTs5C}g=i-Pjm3_jrT!sQFn
z63Qu3RR-@gt&fH7p)k6LWQ<tl_yoOA1Yp&5eCVSe=F;U$yz$!WJoDTOJUBU#&knFk
z+AUdYvN6TO@O%Dx|2H0Q-r>svLaF6MWxiU87h-|C$z-NlitiU*c#i+(AN~XW?*H;X
zVlm$>ZB9Webk->6FiJs+{M09Zig!Nw|KXYEzR587u^alQ^4X;5$+4^`P8{{onG`b=
zW%hO<mG;yIX3OHqktt|~gPasW4@5m|ijlEJW~?F?&ew%kz(|EiDQr-ooDwO9QaH*1
zxoBCDa-t7C503BSjKdj=Mi&c>fi^}onaa7+N-uUFV~%D@C}lCuVsj)DrN~thD`<_v
zY9;etf)K%{fDXmyQAP9+No~i}HK^T+_3<IKaa_G|gT1R)aFd$nzWxkhy+#k7(DmX}
zkRnb)Z8Y<y#%d+4$s9;&A}A`EzVFdm(+>l#A85M{ZER8dN<rj9M2+F82)dJ%pm<_2
zlLQSd=dCdYRSI9=;escPEzE7pd=#b#Qs)_lp|o=&-oxqXK-+r!5b<76+LXit&Kkwe
z&Xj{o3wHOWXdB5n5Ms~Zg%;$)Ae4qYW|~`!aTsI2g$wp(5d&N(><W_jR^wYZWq-hh
zl}7$~erqb<*o8|u6Mfhqjxpv<=PV{Ef=3Rcu~GwQYq3H8J(@3PV?YbH8j~?-5y==!
zZ4*Xkq0MAzx#TQi1VZ4Vtnu4dD%S@^9jgfCs{39ZZ?muilR*1KcjrF0pM4fxH#9eI
zqUJkTGm#6;I6^c;tuYx?E}l=wHp!O!PE|SKYm_i!<rL9JhUB+9U?@={K`u(AL5oL|
z&J_Q%NC-Vc*9$)(Mv?soVfUdfL<Mbj1;X_k4U+qmah1cT%;CwBlo4MyK*h|^CDvUg
zWZ~{bolq7+$|U7DZU<h!cfu1l7j#{RAycV}!S~FkEw|5}=gQ(KZoTvZZ+!I%ojv2@
zAOAic+&kulZyur29KWu4`Pus%-u+Epdf{vQxqt6tD81sxe)JPuzwtPCZ=LfC|M=H<
z<CVk{H$KK{83?Nx&P-6T6DCV-3BDut#cClX-ur#;<%fUr{ao9Bgs*@08QysQ8$7xQ
ztnYsflWz0KBl}EeJ7}XB+B4yUq$sVQoP~pC6|T}urxSK|W=y9us!B3lGm#+#+P=qD
z6<JA}hpL7ok#6@N+-EVJmbf7~ZheXV5wbZbrNtOoZI?Q6UiVbaGFXLDY13`jN@296
zc6C8vVIXsKe!|Ts9_PL9dmo2Khupe#n^$k&!KYYSEJ~usjK!1bGV~-m{s*+~4;our
zfU+Ae+#3Kt1lDAX(S=-IfmTdvMP(|!^p!91b3gYf{?>o+x6x>f)103#+27q4gsE_&
z9)HK<eDtF~#xvi1wkY&Qj?<s~`%69$t!yNfiYOx}zf@Ftvza7I3Xk9_s!U-LjhD4-
zT&ykaX&|DAF>G4cDJ4`=8(PU0e!2LM@lXp_m{LsVV(O+*LPfZM01o{?GKpLzoT;`P
zzL@A#OY|O;bSy-XD|R(~*P~Jt56q1sQ#=Y&mO@iRWJMI0Gde{{j)@5{s3CE1b%)%{
zS)=JS<fdZ2zt4PU2VEJONyBoz!Veub1dJ7dh}IhKGa-0}A(o{g(e^D`*PHxTQv%~N
z6#JZ!CXyuL(%Q{2R9fMjm5)WUT~kU?o}$_u@{8!}cs#e?Q{g*?&H5t6ofIlk+y(pq
zeV+(E6GA42kq4ZyPBWX;EOuv1XEjx=K}%d?j6EUtgwT^xkmpN6MH;;xFY@ZX-MH8~
zeX~jN=I8YA@7}5<s|ui{{*V`4E+_ECw1^lJp-iUZar47KZwxU6G3<636D5U2D#Kz{
zm=a1Gw2|YWt>jM|ZLr4SbR-mm=CQ26h~io+eS!2fB6Fi?sn`|e&9A>_$J@(H${Le1
zO$5DKbM)d1;51ilX7=c#<Vq7$CTUBwj;IYP8&pz-FC!|{;QJz;$OTXp)UKILIXONi
zA}UW!Mi<W~NsA~GVMwCv>OG+?uy@<icP)KC5Pjhlq_oL(+*}x1e4wh4qv0&JuEADB
zWk?u8P@J8$*gZo=5&NFL4@4{(Me3u(qtuyXz^jZ09-N-xLr-lpN;RZtAbON~z~0=k
zTs=>tp5W`B{#}0buYH25a{R<6KfuZP?{NQafHcA6J^H?<a#O7FJo?BUmoG2)=GVW$
z&;NJ-ny-H4b)3FTw~DA(;fD^_Xvm5b8`@T8v7FXq*1Y$rC-}&ZK1Db5JoojN`QmSU
ziKgpN{xweC_yUh!_8jb5(3)WgoSiP|+MbkT*Qhl(t7sa_d_EC1*K|r%h|964i+u33
z!@!QK2u1x?xf&HCIS%y8a}E{@jFN@P_Y%fD^u3T+QN?Lr(1|emB4#nRH6~-NqL`MK
z!cG_Rw89UGs5SS`Pq6a|?|I+T^j*vG(Gf4c{2Hs?qjJV6aN5u)Eroy1^dwp`gL-J*
z`ya+zj4d}z{vA_-5;K%D1Rq(gI+QvGghS_yCHcUlp7XDN^IvoG9gp*gPkcfg>9xiW
z1C^}`&rIC=fBL6?nos@Q&+*{!5ctz6>Od@vqSd9aD=0uwc&H4RZRxKH7wf`GxGgGF
zhLRL8wvj>*g8A6mO6fwB)0P(W&;q{sb<U__5Igdf<f(-hy9uz*g;$mg)@(&<Ib{)w
zY1!OtRlk{`qy+h(#CJ5s;-MxBL?Vj?Q7QaL2_;5t<Y5360Ye@#SN8UZ^_(hZRzyNA
z?cMP>O=nZO(-Vdy{mG_oFfl*~#AUzD70!~~6H_3o8e_%#%o&TTYP6A}R1ywyF^SeV
z>qNYQ6yCP7B|$}sW2Ht^P7}t?&xJ$3EsDkhWov93LckBh7Ddf*qonmiWauGyLyEfC
zsVG9olFCzCc6Vnic4ssbi?tAmq%;s>Uy3^sAC1aut+a@`v=LtGh1XbcvJ3xa+n&Dg
zyNe?gD$=+6`^7PGYX30B-PZMD<o9jrv(3gK_<%Qq@bp4JE~#;YWTY*gvUJjtLF)uj
z6OEE&2{c+6oKaXKb~GuM>nUYQmMALgEthaC<dC1sWwd!a-^)||+bZm|Hq>MqrBH+C
z^!|NrtAv_0n8q=6H4%??mYgj~8B$VI$`aE+&f>|Wi=ZJRljYzCX7f4DIZlsHASJr>
ziq2WAvEUrhOU~=i_YD0&-*$vmPj|jzwOq1Zt?Al!W5P+907>$E3h`dc3T)+YO^wnT
z;~dR=Mm1ZYyB@7H2M2o`ADysXx6G=V5F^%V^wbcn?8XR*5HqQEbjsribn6v;d&b`G
zE}BaCgL%N&fyLCY=oLm?=Jv@szw@iV&yWAbnxs8H{!<^|^S|=~&wukCR%M3ufUygH
z?O%S5@Bi=z_|0GYBER@^zs}+PfR1y9E|NT`9BJx=)@z)Z(EBJ%C1cRe@Pj}2AwKr;
z5Ax_^*Ld!k*ZK56{WZ9I&Z91K`=zfiRfjzO$Q5SujIoJs?OCr^;-wWMN@d|oHjde>
zVP|K?Y&NH!G-y38E~0qthk;ewvg;Zt?xVDLQW}Uu2f?$yGlLLFB!)qJh<peeZjs6l
z^H{907$hgqT219N&Lo18wuUv5QLM5hpg3+<ym<E(kG%Wc%ocNwPfmFGg%>%zcO+zK
zpmLT<8JyA#gPemgL!TjxGyM1R75^6Fq7my`{cZd{q8L)#r3-{@`95-Dhd!{P#n}p1
zTZ~?^yVzyjEopPh&;7!uc=vnW#f|GXs9nv_4^*xu#e^=Lun&Lu!#w%UC%OONu!t=F
zbUu_60@3SD>(@jB+MtZ^flLAKv=znO7=Lu(fNk=$wbh0Eu1TuIDULQkMNymC7GIl(
zU__VXE0+aQQ`%x_L`EW&6}9--i7?8Pf<TPMjg>`H=MmvZgi;J)t>p6hjAl$38S^$v
zVRwPM-1Y-~knHF%cw*ZLQ(0RQn&cIi77Nak;iO$ckZbpRy~MZYEG`{z{gE5!;5qW1
zN!xMt$`xGO5~@?qyW7mB4U2irbW%~(n%$iVlV&2yU1O+iEwOIKptI0GMn;@)<!s3`
z7D|9~R#K|S#pjHYh#~J)KJ@eudbPFii5FC_eydn5#a>EM*a>168G1$V2C|icI}3kq
z;)GXs{rVLyU0P7p3bZFDPfQ&lwqh5Sw(WZ1l2<ral^8m+*;L7p;%;j^j1>P`OF|M4
z7jGj!u{5wADggg*{JD$cP3YmcQOOzYEG7iB&L|a2VHfazYi69SMJqwyG>~<aWTBYJ
zI$=yfy|uK=vr#CcFv<v@-b4(B%$SOCaoy0H>oBV=LPsg?wuZ^y+v9B&b}D7G8Dn;Y
zXt6##=I-+^GMVq->K$CIFbb44oSZH>d-Xo9o-m(R>`XJ{$gDP?G+_v&5SY$qtWM9d
z)^UDx!j7$pZO<y7;;h3P!_W^5gA{e$x@CEOPJ6bdTegDw6c~4mOfgy}XO$&V1d57n
zBzTrMk*p*iQCH_wO@pf|u3z3~)>sbjKOj0w-859zQdbSJ6hP_B;~5CGV_0TjU|wfb
zzec4U$)8P;%4D+Ym^Kwx7lD`Vyu|hWeZKk19e(fEKF1G!;`?~VW0zPBEy+Y~-?~d!
zSuS6DoL634^V`3Ai!Xfst5kNM{?udC46Q9zNoyx%$%kzH8drm^S2PpD```NxKJwwG
z(L>_7-+h74|LZRi?w<0l-5Gbk@j0q+$h)50=dnlji8-)b9nvm6tJOIn3>cHeE6N!r
zlZNSR#%w;Pt|wSmp;TWi29g+RdQa~I+Ew@<5PJ3zWgzt(_@2fZ?BE#+;u`#xLllTq
zdb?8`l}s@NAfR=^IxtrFLDpKT;@bs^J|@2T+N;Eg<LVPPX?@4>;UV|mcmM_NZ>+*;
zMGOJ2G=W4Q6V-*n;d}jxe~U4an>SAEquT8Pf9GPnkWymAEj+NXC>e!N!l`N79;MH5
zm0W{$RZ%w$F$7+H{bfG&k3YrV`Ct4U8rMJ;HL59Qy-F#rUAxAQ{K${+d%yR4C42c#
z=c5_9Y~!<~!okwfqO91n=z_XwYe1!OrDViW$^k=B%vgcZ!a;LE7?HzfLB9Cj9(tc6
z8kn{PjLKP(T1Ep~B?ihO-cXyx8|1S!YBWX`R{<5JxZA|>iVPR!3^|R-6%bRvCz&Tx
z5hjF~2qBStBzaFt3Z()Gg9?@xp7|zuzRNPfOE15K+MDs@hu+WQPrrxhe8x-9KhM-C
zuFPk|`ww{HvB#;}HN(*%cm0|t-|+~S_UFt?yK*+2Y+A#{7_1g+tc)VZ3{jM$QV7X=
zm18SCI;D!zm|T#rVl{&*9rcVVT<-C6kq=?3D!r&Iz1V7uag6_p1RseJydO8HQv5WH
zV{d=KmCO5Fxq6A|yaFW&2tM@m!;-%5B<ZM7Ma00{V6?;82CX0F*KHN9<2k!keHy-f
zD>h|8#72?AcSP;f7OfjaKWap+FO1Q5Pq>7Vi|m{$kDUl1Z|D>k=NGNiNVApH7Lq)6
z%Ef*}Z3?^KEebKIuZ+Rjh{>9yjMy*53`S}?sc{vgi%4BLUz8?m$=Ow#HtqN3czcCi
zX-{M=r;#xR<s&Dr-{D@fK-($q#x5#&j5Q2C^5x(AePV#ep18(4p1eZs4C@YJ&KT#I
z&F4J0`x?=EP7aT#jAdv#;<~5pdZKl-Inl0Ly0&GxT(MqCMq&&yl{#ygR1=Aoag`Wy
zYH=rrOo)Lo&_1O^KlGcOSPYSndsOd<T}x{;)udr}KBF><<KrWUL4v1i&`pKa4QK0t
z^{}E#4X15~QgCBuf*E?^;L*;bt?b%OcKF`YxSnf!np;O-;qloAIeGDj-#C1QANZM1
z^1gSygU4?^$uIx%uky-Er`)-9!tZ?MMQXD{SRW8lgJX{vKs!s@p3+Pm&NTGhfT<#v
zZp?W1Q`h;>hu+WS{XK>=&+!|l{OT|KGLvrLiJ9i!^S_SqFY?}}4tUo)XY5RMv7Q*X
zfA;}CbSRxm%R*7t71QaI`D};jWJYD1vRe}vJj8+C_r+EqldUDhyg^Kr%J^<gr4`CY
zhOVXaf%Up$y>5${Wnc(B+B&j|;)$q=v%2pIVZa(i)L^ur4V4qihsxDtr8qo4=hX)f
zxc1chNR8w8@Q7Dme3|v>5}_U%W=3O*A@z_01DOs%>x3rq-$S?#Hk0nwd44<bUtCPa
z;%~IvAhS^m#wctvYsVbaOxF&a9G*xE+*u+~BIdH<w}1OL`Oy15z$bp<lNe)Y+g76E
zvM$An=3_tpG5-GF|NESuFaOjycDhXXxu9YSS$wilrN$D*99j&u$4I2neE6-nP-Ak1
zG1AsmO6*HC<s8qVIF=eu&d-P2TX}O##>HHXMMjL9#~!UGMwkCo+T_IzLbQTele4(^
zi=74=WY8Ex(wP)8<V4QmH=2qUXviaC3r0(OS&B{h?zHCk!2=rQ(32@$*P(k+22ZC`
zdOvXI-W}SL6LQ~U5pO;zoWvM?HlHz@&zMeYtOHlOqFNu1S!uzQ*0(mg6w&hgvB)*0
z2+lc<b$3DZ*r4(o@j=d;&u~PgW8Ah`MX?$;#nOnDj;h#ENR&5u)-s_OZJF=PxO8cc
zz5NA~Nrh60L?EReA6f?A5kglsxa0YvH(yeWYQ9DA@X!gyIDu_m-@_{EhoSX13ogFX
z7?01`N-wiPdB3BmWNT@XR4Si~!MH?LB38M`%^g1iL6VD8MM0v+B3#~+z|<+kq69V=
zD_lFB%VUmRo`PC#(6~2s6DWo7MFZyV<?*%(JA;C<z{adZa;!9dHE{pM*NE0KcTeKN
zKuiOR`GTFvoV$09xckN-lbPkp)k{pYB19>;+SMWFr)Qj>oYJ1Jc+V40vb=vn8hY^@
za*nR;Se~ELcO6~VqjDA^{&db{I%irqUv|6*phUkY(BNb7uq(wv7(8v;vN~VVbuB3*
zVn5LLE3`9oVZirmTvKuR(mtmrC#<_QMmd~r82p4)XE|Ow?)NQ+XPJebaD6(b3JKqJ
z5_?lwQqBxR!YGHPXLkWHuQ@*XJeM!Ok7sYUeC`)M&m-U0vU6j>>Aib={D*%CqclJA
zk?-f#SMCt}lA#M2?Z{~Wm6^;uv$-M;ft!y#%8*<B#{cBc(+z8Onu@!xJmA?ceS<fi
zzr}3uJhkt*{o3cC{SxncYQf_-CN!0!YXezpR_800t1~i@s<Id@k#3X8l<9O%(@e3p
zE(s7(g3VLf5S~K}^dSiMMN#~v0w_i7dJI4egkj+H^qkl4-ly+9JC8pu7h5Xrz5?BA
zXN%vIaC)lB60Jh3gw>j=wCSy}q^!7g{{d<?VSfEGC&QY15AJjOjeFP#0Yzgfs$9s{
z*3id5r=g25jD_8n#q!5`k^Na8`Arl4xQJZbVUG&CZJ{UqL0wkL@wPF>vLi07s3_d7
zwq0Xtjdjw7>AIF*{Ka47Jx@N(&Bq_xETrgdEZ_$}_yKO-yvaAd@vO|ne+re|NZTm+
zd1bLsg3LulR*|HXa7R^S%<;_Q!ez{OP`58vNveP;D5r`A6-5LZl44d+my4~j&0-~w
zH=3_+k(v@IRlY|5HPrZ-ZqH{`7B!s#bjiAn8@B*i@U*}l<AW5T8gaW23fD`45w5C~
z^M<Spl{T!p7V8I8Gr?%deO@omd2sJ84^EDG{k7M~>y|1+CSA{Rd5&$@#MP1^1YA?$
zYKdjjS_)@tEm<2WenTl<*cQxEz=?sn**G8whbu-A?5R=&s<HUt&8E}EpnhB<ippQa
zU2*f7YzmkpS;;AumU<q64@%ca0EfN31^fGZEEZE-Wl1@Z5s%LhWYg-Sm)9wF5Sv!)
zrnq~vIWcUCl#Bn~wD$9szd6Fl7dIT=v8YpT`neR)W8qsKVSBR+=-`-~lF4!%8-oE<
zx>(p<_$vP%i#(;W*q5a3`KpvymthQ;R0MNcM4T6%fAc!nUPoigcw6X6Kj^LA{^nBs
z_DALqG5+kV8*dM>D==|ZrKOn@DS^e&9k00k{LAd^XdXGx1Q%%LbDns|<D4EX>AQi|
zvZtQ*JUHxFE?b;64DG<#$q7y+-u2j{_*G9mX-GbBc6@>@ZP>nRsq2QzHx8K1=S*jF
zblqU9S~xhRU6Yc;SoM8}FN;jlh&#J6SfH*dCesPKy9-I5Sa-CiXRO*aLlS{U^gXD^
zd}q$V!7isKr<|UwVeM%L#c^A6JXE}PwxWxU>$3?nHz5z_q!6%LYzD}h+BW1NF>Neu
z>Nr^ROyfDPymHPvF8?I=W6evSe}=9JR88io$KJ_}M-F)T*0X%{CqBU8y(K5dEq$M;
z>j}U0+rPnI{dfNzavqpA3toHm0l)k&Kh0-;{i{57^9G;(@|RFUOMh~Y{n~J2QE~Xp
zZ_*uqiT6F_xN$vF)ji%riiY#`n%j5pVhp%Si9x6}OeYg&(>asLjJj%szmgzOlyNB&
zqZIKn5k{1jh1(Q#CnrHRqWAP&%iTM7c=6R+oSv;YxN-@jiY3DZX?~0mWrSoaE%;$0
zm@q<^5j9;2H$Pvu9Jgz(z2hCMwc`Bbj8|TLmA31tGfZ@aR)(SP&^n{7!{<z&p$AIB
zAR6^OFN1%pq4ikMna$5*2K4xMt)xw=H{OzAvv4XUT&$F`sMw4kej!A>&#abfl+{#|
zin<mfSY<RXzVrgW`fI<!-}+ns0a`19_o4zc67zKV(q*20`e~kd=Gl_@`6oY=-l9+{
zB$QF85%o!e=#8<1r~rt0>bN49O=0)&<7N;ebqpxbNg=X&QK<@G;VB&~K`+X8;%4Vj
z+K6RwF4U78lN6acOMa~pt0^do!V3k!`1ztF1C=C&t8CQ8!Xp-&4~_7evdQSI$Tk%b
z8zh~HRVJij@2xCRIo$4?s3OC#;<4+OIP@J~`P^stm0$j69G;zW{n|AeZKz_V?K*Dx
z9@Dnu<&v;&#V*M@lrmz4k`2xX>gc_vPk~sB=XJ5j$fX@O3LcEsSS3ZA8u1|w+KC}>
zi1HZ9P(DpLxZ`8Q;PFGpdc9=5T9wwJ-ngTs7|9_)$kGN(ND`6K)JBuMY-sDM;mYMb
zE?wGVI;+rn+`LHp()-8|Ml<4amTW5XZ85D>nf$&@VJCYYvGf?ZbW(7p&AFF1#T^f2
z%f^QpqMLJT%T0Uuc^<k}V8e?Zk-`|m2Jzp-=!I<&{m?a05II4q<(iY{A88=+x0pPn
zqIxaH@|H{}A&@F>Ig(Lr+u$fgr1;pWEY>|l#92WKzm0M6n_t^o{BHY~@ASJr8|%m0
zE9^|dW>8)i!%ag_nIUE&u^uludf_&E`YN+aj%eZL6IWQCUggD?-r)5Y?(pKPha9z;
z<KvbJ!^A3{dh&p$o_dUVZ7|Vsesa#y(Jd-xm^Bj`S24MImFe!BW;Vk#VyU6W3o2){
zE+1?Xr}j+5#{m_`9zUcgF!>lcQ`Ls5nvte8yL&q<mrMCtuju^%eWVRb>Zan-VxRl!
zoV%w3$LqvV(!ADs&SGX_d+KUQtENn=2{r{>iKkLn3TKrV2DBFLn%*;9%{{m7|0=WQ
zG2XTFI45z!y$6cl__^QU=@0(^m!8-~QQ;<zAAIkVBrNyt-{+A#mw4~@KhEbr`z79Z
z<CuH5R;bSK`tQF+`@#|7_=u}>xVqPH|JK)e?&Z&OrMkn1-gAi?*B3Y^{FEUKfT9~#
zOr{MkdVCO+rLGKniz&0FVcJZnY=y?+fgx)M1|M76e9piM9|IPLjS7=I<Q}gSHXBqs
zkk)}0U%kbP&%ea-sO8FZ%9YsyV=M+u2%eY{L&@_~UKFe$R>VG2kyxcHqC^)^6N}Xq
zItK18&spv4GQDz@c-9ebpK$-)32ou*?HI>tjC9#jqnTxgj{%IFw*j_j#do-*-}V?E
z?yat+aehvl=Ek4elyGs1HAcQ<?GURHtu;e8P&>1=<3Lg+Vv-1<m}GM&d70Z*u0;4C
zX0OT;g2JyN+Ic3E0nL*6e8w;gRJP_<f9sd{k&pc_Pd@%`DV8D-pt2QB)$kJ^|2Y5T
z7k{x3;{W82P2EhX6=WYFL~<16uC;;~D57}3<pj!-mB_S(sI5s+)O#B=MGNeavyz2Y
zSxAKpq!h*$vU>>R9#l4D5>{DsP?(??jKaf~>!oDgvs$_o))?%F=tTJ(OnH7&kPvcY
zi?@g)LC!-lgSBXF3b<p-21knmwwHy}WsSi|@Oq3TMI*6XtwCFqt?^YNg(WH|ss`Tm
z=q0Z71N)8Pa^vW)T;h#4?vUju()KO4)+=1}OvqfjdO(|?&x%T0Y}Pnc;bKC?gpMLY
zQ3irmL@S<Ts#F+RXGzM)8Iwc|;;bdAMAU_Tp)=WmN*b%BQ-CT33Q}DA&@v2bF=x(E
zio~2t;U}nPjtUhNu}#D<wg3~spR+q}nD0#4zr4@o%M+$kg)42z<P-fc(DwuDbz~Sc
zF&f5{B!!3+jj_d!W}9mpb3*GaBAiNQx;8~<qr?YP7nCh(DKf_bPlDSsAtkiR7{&Iy
zxq#;Bk{G2_>7F1h(>zA873Ts<l9OZ+Ya)u(x|fbh<!I`L$#jCV?!v}GY^G2UbHFJh
zne!TQp#WjFsV*`0_#sid28SqZr<J4c6W;gqC&$z`GeZ{p8MLFzX+x_?8lui*6;aA#
zb)a?%Yn7Zwv0M+O7~Rfy-){WbDBS*_@%HRaw7}K6EJg_kS&+L7xCHA5ryTBR=F<aW
znlPU?y!XBD<Y51R!?TWi=N-f8iqL6nvOIQUm#5xy17p^#*DG?b+1uM^|I)juD#Ls_
zL!-nWO+&T@PhyB3o5WX7T&@eJzzAYsFsP_VA&~u6&6jc}L@&Mm!c75D%Fg!}Om=2W
z*K5wt&si?dH$Fx6q+v3jaHTomjpt5z<LrdP5J}2X(}NxeL?ViHyJFJrP*oE|Rxhtf
zvTMOA4LWeJYnU{eJNIAa!R^~D_TI&Oag*cqeO~;+Z}H5R6nzTpUq0YV`YY7a3Den(
z8_f;A{^=LF_xvHRzjU9~@xV^6*fXBXv|M_`a(eIgdFACV<J()j^G3y!H?DKAzl$-N
zlp`Srl!nu@6Z&CYg4j*rh8QOGg!yd2w4PDB8ePzt6hoQB6v``7fEX1jRaDAhs8Ash
zLn2ti5E9*kC9gj75^uct62~WNuJ6ux?D|dSJG+ERT=4te6DC`00i88N7wLP?&~^|9
z$Qg^mI0Yu7(F}gzLAyj9Jj$9(ziPSv+HHI{5E58dGsH-z6SIS>#MPQU^qcmWG?h!C
z^~Zh9zDZ=N^oBBZp-A}-Px>d+86LYb#;B5eIp#IXL9p6kmBCsEjQlxS_PrknAyPR@
z*w}TX3vHoouSMEl6)6SU^}zYr3P@OMO89!<&i&f}{Xhc0{Ng|Rr~Jo%?|)M`ZQ^B_
zm8LS5@B82f*x8wLe0=)Hdl0_OFs{PnOpdB7U`T5y#Lb5$ttO;3tFshz0Lq}qqb=vR
zj7W(OLyBT6Ie8ntm)1rsO-gImj3jSH;aA!q%A*)#L{gOe5~H^npaRg0B=v{44ml^3
zz;)yVF=b+A6MHr)-gTzTt&!{lq3b1bth98irbgLXvYKs;S<I+*b}*AEt9C`D4cB%T
z%$tVS@7-f=9ev*uLZY?~v?3`n75Cb3?mZJP&I*lEG)bcqn1TkUEXOgXEIw(aOD;UE
zRFI~zU9DuzC>^FqE?s32PNd?2yKy}m58?#`Naa+n<AP$B6p0+AEv%wEKUp4gHnq&>
zj`>c-&b*>&49?2mLe8W(3NpmjBxOa)dfN(A;?p-8Dof$5wsS%G%r6Qg<Vy+@bzyGV
z+@)NWBM)@!&_0ASjUeH8hwn5WzWG54YB%!d#@lRK!53cl*b*OySm_8yQD==dvM#4m
zNXBtJ2o@a!Svzq=P+)Q-`kw3ul&(?QlC|1ce52Gx4=`Ra8mVtbFT?Sw&}@&-n~(Fq
zjpOavos1j;Df~BlgPg=2y3&SjwdUR%9hVO*`O2JZ6nlFMX43^%`pBiTCDp5M@LJcB
zLu4^kT;7|}R25Exr9xrBMsf^zfDq_YlD@r@4{tnyNBvsis|sYUHM&yhR3^d@%Whf7
zv>BPma`dQTVw`g?O3SKZzPrP8XU6_;z-qOk>v}SY+1{L+i|f2_`x#dE4^et9r;Czk
zB;tTE6+UD_7?{rHfFbn$p~^v#sZF8=rpD4-ZP=Y;UO#%4`)@qM^z!3edGy_^{XS=H
z=J=~Gu<AOrby(*}A@WLbmftls?>gA$V6P#pA8>r{Ydrt;FQek??9JCa^6rZ3S9h6I
zGmJ%e9Vrl!$A`escN|<gpzk}nZi&v0rkS$XIbhn%F<KnHeT=f;rWUfaY8V1B0Lr0~
z<WW*=5>gN}Bgf2(&%eU&f8ncCx#i0BU7mj82_D<OMs5t<&MsOR+U}f1hzw!C50XF<
zLX?Zy_cEKN1S!%uO<ieh<v445&dD@a4!GBK9G#tU@BSg#i?vDChGpom(;1U#!@BSA
zt@sf3LAqS1aiRO8F)Y62deLgrzR8&lWnj}CQo;-RR+EuGxuKAP&N(^PU9picM$kFo
zsEYVu3?5fX_(npqH^ewlH#I5?e@@=(!|j)i1Ag%i>-&M#3hKtOUU$r96TDA2SM#~g
zeU?vt{HOTv_kEat=&4;pP9h4pdGj$IdE^mJPEW^E>rZrOYq1&0A~un{DSnDFckDlD
zB|dJt6nbSbl=dDQZqr3O2AQX}e54KkM~{Wrn4vT-O?vZSo5$I-7sthk&EhJktA42U
zI*y0M7&3ROl5-^mMp43!QB<2nIztTP=<#8|4}*A2DvfrQ>CO!8#0OOCVjVS|f~z@O
zF40xP?sUp*I^*c*gvD&eY0M0<r*5WHR->XvrO2Q)XFkvj0qr8mTB-yjg-eQ*wy{u=
zOw2)!T^1^!84FM)f$-yu8!1J#%52}&1@6_utGRs3+qrryP;!!DE@kApEm4=ktrOI2
zw$m`3*371rShhHWF#-wZO#*|sGK|;Arsaw%YS?X2C)d-qSZ6$!9%_MtD1g<)9pk2$
z`<DBbv*_CwEqPPrZ7pEbhW9qU;;{vsFV+DUTDGGP;D`dgl?Wu4&CaGs94qYdI%5*+
znh+w^N3<OgLWL%a1x??R?F6lB`Bcl!b>m?Nn^!zGbjP;zL+|>+@%Y1vOa82mw=bSh
zilhoWthOXpt4-lGD3+(^oF1PL{57<-!ddWv5ED9gG<i*xmN;k7@r=GbWq)TEqcWAP
zIX*cfdV%^2aR6%!RaJ>Ie1T~<+bzmDp_Igw8H2Wl+G;9aK?%<W;YR=^pg29`SJVyz
zeGv0QrNtO|_h6sB!q=#4O>~Ycmv*RMJ7Se0N*S!FBu&A0kPI;<x~^w74@@RCAq9Fr
z6qLk}QxHLb)r6sA>NJx}4U4(uW(W6AU*fedd=d0bE?s+qM=$LYGx*q}R3Im?I$^b?
z8xFbm(r<C+>>eteWA&12yKwdLgewQTOdBB|2Ld5@K?QQ8>spRaj=6gEs-SpD&<Ud)
z^XU$=$xLpmb%DeKAq7%C$LE$1G`UZ#ho0U_o@_FJ>%az9t0ihWC1u6o{X<r(mM7nH
ziFZH!I9C_@sHpHLlyN9yi9yKW!4C|5N9`<%O!6(hJw>G;f)1-`Dvi;ZAxG}6*64$M
zRw{FHe$KtaL%QDMvYg(1h|o05clS6sJ0tl)EGLu}<Y9w~jZ^&}PfJvd*Gq=Fu24$a
zLK#^gd~su^OoS=_@jdr<8d^)6w5}_vsulxXt#QthQlhGAtg#HWwCRk|^!<P_UCGiE
z+X(9{y&o<tPCOK0G^%K03Z)>_c%N7;TeMb8CJi}fydOCCr~JyV{!8BTu6HqMW*BX7
zR*ZL>rs3W1emCFv#xwkBv_a~+!BH%5pk&92Jp~0>kXA*IYC>gHHpTiRlVdE4pe{RL
zHVeqb1y*19cg~qKTFyuxjhu^*m)6^2<H90+ap4&kTwtTp&9HIX9*fD*-;+{s2R{_}
z!jQ69xx~WT6ONXc#7Ex~hK@{P2qLypM&l+E=9jLJbcN4~I7F<fXwO^v)UufG(#4F%
zQW?j!tB;_oDXux8Z99o2be0@Egn^hM$04DHKyL$c41*5r0F#{12|QA$B`9XL8H2F1
zEMtKx;*e7B5Hv(xioJr=tA`+~@%ZXZiM1)lliC#MF{L69C_Z;Nfzk5(Wxd$jTQHqY
zm`o=k>T7D0QL@&=RP0gwLl!F1O_I;ws*<pv&1S=8jNWXr#%nDVKH*l>!9#_edgy%7
zjIn6j@z&SUrf5_b#6NEuib6fOut|t9is5+afQ_6x?<I2XVqrJV4dc0gQJD2+CBC4^
z;QNfRnWj>h%3#TuoCrhDdbOmgXXr{kRD&vrGKKMA1$>}<DO<h_V_W6o^L{Ukw<(^u
zP-%})x5^umEjeS1BlyVa>59Jhq)KBHXagz+RO(q&aCIT+53`x(@a`>4>Nq&q!)VL?
zr3Knm#1t8N*(DBxr}utyf*WH=rbPPaEvH0VyCFau9I8%I*u^m-Myh|Ew;$*x=QNZ_
zR*C8E;D-W<i%ecC!|BOw<_+A~pK<n}Ct;{`g&}UF`v4&fq~xh<C!TiEOFWkmC0Wkk
zmBLht+;%t%GnHv-%l?k##=(NsddY*s&+ytCr=Sg8NW>tmGLj+;fqLSp+`!FC4LkFS
z-JL1(NzH6B#Udu2N-2C=$|9R1XJ==0L(lc=*O^Qw4CS~Qt*EM+#bSratPw;hC43C@
zLr>Rt4B?bEwfJ6>){14{voaAqC}OTz#g;wo@qUfV2H$zQ^$AxFCOmfYDzkaTvOQtl
z!s!6+#$z1x5;v1mAVrDksp}dWBI^H7-k&yGmYw;1*l!JcI>Q}vsH_<U6wpn8=oT5>
zBrTbeDTy>mc?dhKfeQPjtdHPtAUpiR?<`vpvK`@&BjixXj!;n4K+;r`Ej4Aa*&usv
zpn(FaGP5%CPG{KDn*8Efd!KU?1vIL=;NY;)w{O<H=bmBjwby#q|M?GLB=kEmj?If~
zX*3}W-0XK;jRAFf!PRcZcC+K=ZpV;`-vsE1ELJB_6m-LN9)sB=nS5K%T(}v;|05rh
z7~|b{-{l+M_y*V4*SvV~f^i(@iLmc``f*@z4(8zRpKkGTuuu?Qm9+&`RZ&$H&N;N!
z)MbrPhwMgZ0YzC#MAbO@c~YMya_{{;w9b3kX%|+K7m~7s5NO+;X3@~L9d%s+iP81^
zy}$Q&`OfeEOZ@S#f0Ga#CKV{4X==Xu)vt*9@keA|evwC6S5z8;F?eH$ql`)CRrYPn
zv=1+8OPxSjx!%TFaGszND2>i(35r}yps-fwR&2f=DYsrl8I&zhGCGpmamvVZMIUEm
z^nQVNn3pGn1Q2n>QM)c=g*nJ#?EHYSrBGLlw0@J4NURtlqZ`FTZtRJ{QI#civ%r)k
zx~egY6{i=E*pH6=cF%U#p-acZH{N7@cFFC&r6;lL2R6G7KX~4H^Bpi1+jh@4{`l9?
zMl<vs-G0aB=8CJn<@RQSxw*rnK<OM`y13xU$(ouZM)8?XrnSPPgicZB^PGPhksfUf
z8Qlv?vdgt_IE5B@Q~;}knpl#MCZ99;dz(8|0Ewvx0d2(M#hPMHGg>Sb)J;WM*O;=H
z&$W*tlo*|~VV(DkVOlS=IISslRLq%qt$Q8inmc7_va~s<*Ht_ycwa~4jztb=DM;lr
zacG_X$m8A)<-ih2Y3A655SSz;$DV)F++d86jO3%wB}UrENMxD|UO-$Jf+7Y@7!^b@
zr5E+OB8~&yZi`u+QZCjMDKKWRTYT*D<C8&CpjCy@6;@e{()^Qpd~P4R3HlYj$TWbE
z?@BX9PsDP&-Lr28;;P1wceW~08pXOQ@TV2Fw4~(e_5<H<J8rITSgdLuK6*%VazV3L
zvM6iZI8x*|7nP8JYokep!NoXxFddJ*xxb%8EhaHr;vR2Jakx=a!g$Z!cEd1;>8`O>
z5FDjYQ3Dc3)ow<5=TJWJ-lG*at>?CNRIx^d#Na%G57c#mb0clHV>v8o7E92u*=+IN
zqf~)QQg|2@QDp&EVRWR>i6Ta-(r|f}c=VvA?V<Oc(I=b}S-c-SMwir8Nm11-79~w1
z(LGwse@Ln%QsIUkuO#~T>go!#;>nXIlEi_Kbrr&6Tr3yZx+E%v^MS4(+3j2UeqguT
zv+sI#F5yOl84RscZ1EI9;bP?dE6*Fz5@SaU5&@I^p09lHCKu-`#&Iv!Hdh_bZUaxg
z`m?yqb?{!!a}0jwTPXtFZX;%(AqWVO#DMhGzQwN|vQ?J0+p@jcu-oq$0jsR+?5(9d
zIYBE!$eG27gqRK-^GYVh#UK7S-f<dZSS?okmA~>=__zM8e~aC2$8Y`CZ}Drt_G|p=
zul{{rzI?&;^)>shW#7s9jpHEi;tS+;9P>o9)|90nbyZbS*ELmDQ`Z&NTB@?5tV%Rl
z)n3(<Rf)Bh>+5UEq67u2)rx-Tg{zofV>X)36Spyk0_4zY(!>aTHwapvk-`u>FJC_A
z@BZEYny-HOt5hY#;DzhiEO_UgcPYyf@47GSy!@zR$|IFHBH_k*r_oCIKFNoS+(;o|
zGSaG*FnAKe##RE!F$pMC=6wl>cTh4;JyfMMN*h7@rk0VhC~HyC%7#RTTwKbZO$%yN
zk&xS6DozSRX(t}H>}JZ-^t!;zP0q^1lrkDLedQv+nu2P%pl%k}sz#R;-joz|gC7IC
zeq_7vXqFzG49jYPEgOF4qwg_<NK-HP@Q?j*e*3%M<H5s6oRQeH1Ew&tp4bw#SQ9g4
zMR<mwhMt?0P;JLj!J@R()=(Nv4LP$mx0Rz8v*C=^rKD$ZR76_2Dk*e83TZpFzGZa2
zZZkq2we$LxTj<l-(?$e@WzI#eYPqi_B;4frHRSu^7^$0l@M6r#NK-prs~;-lrsKFU
zo4w<v_IgGYr_>43{i6QRQoY}jpT^*iY|73^)<I2u{P=s%{~1qm?>$dzs**0!f$DLj
z(mg`&WFECd4=Nc^#zra5b;>~qF*0_0!qB0@h{lT012H&OQORR5lFFn?v?|cb&LXFD
z?_B(oaeU6gPL8~9mCLKLG8G7D#-JFL;ch>&-48ggu-SQ9D=4g`Sf65aU>NrlMTNBs
z_WObDcFWcE3tnBl;OycFkKcHciwBq3szMb7MP28@PA*7D5=1>!bSd}Tg_DxoK}mi$
zSuCKGSasw96*5v1oM(5p0i;!R43mc}AJ<Xgqa(z`MX}<mmkWNc+fapoj+(BIj4o0X
zNeJUz%Wl8HlxA){b=`=M3T-Rdttl|p;#HPpBanYeVue~4mZDs;O2RJzjnM_c2N9KM
zNO+Wr7^6s&dNCy0{)U*u-X*#~KlGfRpVKr8logM?7$R<rfTdZU%0wLt+Reyz*KvEd
z<MwvTzU>(X&oC_LT|@AS!huuJD#sFoDjGt!Mx`YojA-BEyrQy(g*Dt<-|*t@1^Z3Q
zaO1cdEFUaZjH8oiqm+ZybKrJ~#V8;KTu2N~i0Vp1?;YFG!TN%`KGNOn7_awqont__
zmLVyEHk8d0>xHyGjKY`n!0Y(WA5{}OwHS;sltsx0U-^*#`Tycy<-x<pJbLmbU-{bC
z_zQpOFLQHq!*BlPZ}NA3>2LGj{K_x$yWjpcFJHdQCY)+UeZR<KnoOH|A#K>AqOKM+
zjkJswi-m}Us*0v+=-QULZs_}-)01<WdWqJW{eI6lc#QE_V+ktE?UkwRGVS7x(S+bg
zNitJq2c2+Zplf>|l2R$=wnfQ*^DDp1U;GPyiMO7-gL4C>TwqPXJMX+p(^RzkpMGJd
zbf$77FeIYUm`P-lQM;JAQWNYZrHHgoC%&9i7dk5kb9Ss!_XpRspl~QCI5Ire)}Xz~
zln6=dNo4+_%I&WZk!?14uH(SLo7+*xFFK)5Qvs2SyUa%paWYMo_p6P@6eUGrSS(jG
z%N4e&QAJ4`6}qf&am%h7*tR{VgQsr?x_ysrYKFFFOmMP3<(uF9IbLq|oRDxNzWvch
zm{Qu<N?TO3$c0vJO&Sel$=*AXbBqeHugJx3l`Tkgn%c5rPm?kdmsBK3w>ERIC3zvK
zL^7hH6(PJjD(<GD?loRoJRnfVd^2keMOEZvm5QpC%~4SlXk)TiFW|hxWxvl5{DI4O
zbO=B4Sp0I-Av=1k6qv^`Xh+<zqoMV~0*fC;@bUoUwSjqm$IHwHm>>R`bL3Q9zD8dA
zQLmlS!J%S0rowLG78+v-3Uo60eo_!*UAM?q%J+N1(2`t_Qia@aNs2m^bhaqcXk{_F
z#Fzp#B~Yjzy^;Q>Mm(YN#FR*xjLKnGnp=y5);KKA^<3TV=pJ2CRuXk%jfHZ7Rt-g^
z*lfYtl0|*adR1`YI@<1<e%!IU+n_NtO~dNqEW_kL68=jH;ZT(MBMx0UqD$u8vNq=4
z%`#atNYN%-32Q8cu_Tgst0f689)=+=Sc()&QgkF7O%Zu>v0_{|Z1$1pU@R>zDtzz~
zb4H}DE0!lq3RTfG4Jj#ZZ}%uwQXGZeyU9CBTNKh3ozjY?$xiBMow<5C7eW%EthJ~b
zr%c17BDRQlKhlpKWwGLN`GBgb8J%Mo9U;Itdb+-s76FRwu4i;Dn@z{vrlsw?m=Fg;
z2nH1!VyZ|!Kpd%xK(n%3)=Mrg-{kG{6_%_@h)T{&7kJg~h;GAX+%gW4)G4afnz!D1
zmvI<L`8pocU5&9B`ALMN7^9}LC5cEojEwq>P%OA>cZ5w(-1ayZ>9s_M4c=jzHJ?0t
zPU9k@8)gM}7P9f$9Q21irc{Qhuv@Ly{LBBtzrY6{eg$1l^8SRa3LZUq%;lp8{HZ_l
zr+EJKukg43_TT3J^zZ$j`PR3-mD{mj+;NQ79Z)(!78Z*I%jJ@$Y346$EoUcZY-};3
z4lzoE+qBs!%aX;iL0e7R?g=5xzb}d+3o7!2sIpnOHsaJD0$n#ylqSa!1*|QI(ev#2
zGrsi?zQyB5Pf!WUob2%A$s4RzE3RJMexc{)vkvh{1FS}y0u@Ydk;=wY#UQA!mx2iA
zMKv1|(MC+!psgSa;xRYPnatK2_Zsf{FfSQnNLs?<QA*G{Yf!;pf*}Nb4*_u;3igBW
zWwNW7w!4A?kpzvL&^nbA6oqicVj={GOOes%hbTo|mnc<I*A-1&Q<NoTv%r)UQ5)Lc
zk+fy-kzL!f>w3Db<9pxz9v^-GW9s#a)#W8_-_o}o7Z;cO#lQF$`SwR2@q6F>E=5&}
zU5T}rqM%X@DQ~JH;EgCrQJUlpIt5e^Tb@G6brP!=a^WnTYw;{QQb$IWqpKz?RIWv`
z(tS!&A>_GrYQ+k3Z$_g-xVO#-r2&e<QdO3^E~%TE<!Xs33XC;qV+bkWon&eE<H+bm
zw03O%J-IWy-h!Qu*U`3nNZq|-+C2Q|Vq~^6c^wga9oaqN*uCEFzgI}ADbrYr($A{+
zrq4QqB)M%n%`3;n-EqNZj6!2YJ&r~t$vuwVqe&D>QJKW9+c4~JsM>{awpE!sR{%CE
zVa4d#=mKjhw6$=g&iyCt_}qxyRIN|QtD3k935FPP+Aw%2JYHUJ*mna9Qwq7&7;IIM
z6cj3f4r0<c1`1oVTAs2#Dd>j<ZnW64Cd9xn3{tF<%^`#2lO`W>5hkQ&J3+gd^gZ!Y
z)RZHmrW#3$LF6c+2JbyuDVB@HJP3_al-5#O%Q%j?EolWAOAh4+R;J|fvcb1K$%BiQ
zyPc#g#Ndc{w%ZM>lXc!Dlo)GSHjU({>PXwQ@}{OJplKD*8ihdiq=eQPO;BP;s+9ap
zCnuS)-he8IL1RooNP(_vscOp`Z#*PYu-molc6&w_=*E$*8)(~}ei-R`$8LaO5Y%yW
z;G8B=03{?%LQ|-SDF$p4SzFI?4QEeQJgghm>yop@En4j$D%_ALoMG(X=2c4y1B22S
zUGezQLmqtiA#Z=_1HRq1^WqkAYhqG2ON%xTBKQckDToYoE>fJGa5H+^{Yc#Os9uC6
z(ZWa~YC}<!^j$|Wh*&{ud7c1y8=Ah5;%?e3m@M*n_nmk6bARq%q$n#U|BIMaV<_Qc
zRa*1j`|tDM!2^EjZ~qd%^_#!NcE9_hznRw#1(Zd>qG_n>n!2v<A$4_K3%9PW=a$iW
zeF`KNizVae*zfo3_dDvkA;v`4w^U_`a}r!Wj^kX|O-0_cK*b=1-8f1#lL#q1;|SV_
zPnuHFlKtgh{yY2&fA-H|ONDbDYsCKO;^KmjzW4DLer{erd<?jFaR0|Dl!k&zlqX8f
zNdRGRvw?63j^I2*O)4V61Z*^@X#q)MFgpd4%RPVavs1bC$dopqBFY4mEwD-We4-BY
zIaf4_A6W?V-zzm4Z0BPBsJ#2EQK6Kmwv$RC+6ZDyuC&2uLs=FSg`p}-%A&*tPgL2K
z#g@WjG6kKB-0V7Tw;kuJ9p`5Sbycw6@A&xT3r;RCdE>za-~aBn`Rjk@mpOg#fO@s0
z>-KEB9$S~xiv{bG6KvJUcrZppM{Y#-J$mdJx;?7vP=f~-XhOlF5??#3G|E^UKA&4T
zzbYC`M11Fx!A6tUuLLfVteR7b1gB7Q?(Ve7PT2(@<`$$jxk#{<vM#6>6?Ie5EE;T8
z3U4eIW}|luZe$pS86@zTt2xEM+3Z<2eOJ@`m&F|MXtQ@H>W;2zp2w6DuDK{*Bk>4z
zn+m$ANW5RB-M7N|p^=h{H&MjSF!0gu^ZA+zrHtAgvR|dGFX>N8Q0ye*(+xten%8Yv
z=)_J2O;j;=nse+}&ZxImQCp}(<aU2cdt0Jw%js%`PF|oVoq4cCv;w0vrm%DD;k|S4
zS~2{OyO82@7Iv}|3h1e=F9b#s126(EK=6h_Rova|+3xl%tY=YKbjUPOlTc`iVu8XD
zQ25w`m(b^v<pN@EJtW1r@1*b{;ho3l1%i}F)?x}vX$wr5xpV5^1BA)R6Gzug#YFN!
zo@ekxC)NsOErn#tVazcG$)+<vKLDEKwd}$}0@t%HEGJc@^$F(}?7B#;J*!1c@;&`_
zM|*R}qF!LC0%a}bqNcQ(aU2+v!1L1?7T#j^?kPr+lJo`@ORN&^&oH!TBP4F;9aalT
zTM`^RLqB3`!-K~USS)MC;Mi=hxx15A<#y-T?LA!^*|&~iOblbfQ4*5{#N5+pmHnX{
zDg~_7tZYZKhQ-;E%hiI@vw~`=S$ad%BV+7HWu#|h;SF({`H3N5Nt`Zfo;*0`{rBD{
zK6wkXUa)`hoGxmd)tFGRq^4E{L3>6k_9wa%6h0LLYSiq?6~)Dxo!i6Rj&Z-Ib54++
zq#2^;^z?$10;%7F(o|Z=Pc8;=T%Y?E@=biq|6l>b@wlIVri74PnSzVuDc}6rZ}7*z
z@pWt=g-uLJ7R{siLMxp&H$6AkSHRSYK5mix;g4hFy0isNRnt@ri)O*e$q7v(`O1rC
z!D_W)%787)3h#U_z#NN(MNGaA?IY&~TnIE(gGvhT0;U%3tK>2Z(y?4F30}-YmCkku
zD$(~n;tf)vRRUWO+`w=D&TsI8kH60cAAAMxM@%%Fove9qd3pb{{gm1yA-CH&MmDrY
z7X>z-Q<E)4LQ=B<Y>JWSBi_h(SJnVRp4UtrgV(~_(`!Z7%w>Y4WT8`8D5i)6C%mKy
z?U?@N^_Ja~aT}+0;_-P(DJw#cvP88OyD9@H30pVDX6~59<PDbdo}fLdOc-slqP@Te
z=z7n-^O&+E5a~uSF21-t!x4D;^auQ{zxmhs^y(U)0!w#FV-4Hjh$DDoM0_^!Xtfwp
ztKg}OMOO`O90`6;^o~TL@t%efRVpwEqvDL#rKww!xsRmmbda)QRwZ1N^pwMmmJ8%e
zk`x4QcDc)Aff<#z)?!PGEendWps0!&ktK`sLJ*;Z3mzXs_G~+hFE~<+PGE#mbF`g^
zICAlouPujnd&GO2_6m9)zq3+$Zu=Y+EE%2oQQWxW<CtT24C+k3Q~5lbOeILK6{3e0
zcA5*!V`0VgGh^~b=;&`%a-?A75+)J<Bo)z0V>OgI5?#yKZ*c7j9eZ@HQ<C89pdv~~
z$w1El$#Kj3pTy(y85AQIR7@4s+{+3`K(r&`1EM}jiLo2m?)RK5165IDbxBAggC9{b
zp_9hw5?vx(Ih9D-NEm#C;0693TUp$vn@%A+OH+y(sVEC<S!T-txv3{IYSd~D6j0MH
zCrO~T(OOVG3e<8Ti4{StJwoY-KIGy~VG4sA14E|>qd^%kQ4`Y0=~Cmi;JjhbmJkzV
z*-+^cg=cJAhW(z}Rs^GoR?&qKTNRuy9^jm#YdiX`leTYez3%%~1Rtru#E2$GSP&>D
zr+8D~W5NvsWmR!<c}h`NRCUeZN1k52;_hz8&27u|&6ck7^nJ>X=LJ3(QYul}5~prP
z^62C!lozLNo2X08$+G5h)v!1#C{9*1RY_GDNFFmPT-3A<F3U!2M7j=tyT^|$1CF9L
zJbZG-J5MfHo@(~%il`#}&=Q*!PAiJ2S(*hpX`BfRR`l+AkU~FdPZ(<4>H+<-qI<DL
z?OQpAxy`CWM4OVb5n;f%-=JckR0<c5Vun)Wu&!z%1wys{@f^3z*gzy0@>D8wvFoDX
zY<a>j{QRHf{OoK7W>iu?WL&Fc+sgB2PYHg+nh9a~Vh@|gx=zAzT`ra^8bRu&cC0am
zx~?h9l3^GiiPeZw3tqi?#W3`_?NkDZaa^E#82Ul-dP77TgBv}g^LfG1wEIpLkBm53
zD;6jzfHqR>+5%CK>xSKa%isUCU*)~`z69Q(M}t<1hYudi3)Hkg{4@$X@pMa>4wnlr
zrBOyr%Y<abq69@oio1~0D1!GHSxD20ia|O6GAHF1N}%S%isQl}$0`x!@l$b!nbA-m
zaw^8m9~J~@%1xb$M}@w(xKGH3ShR>*Pyl3uRt9S{D1$O~PChU?Qz2r=oeu#il~NRC
z&FF%N#$2Eu6Bvz8f+|*v1?!U~x4RwJcQ<@`^NMnPO26NNPBhj~8!ZgM;Kh#&B}V8|
zK@+@)%Q8ynl}1G{-m`Iz1wrQTQsI+>O6A2|3cD$TT4e-mDn6o0cn&=3V`BeQ$Q-(z
z>hPrw#f-5UTND&!UV{poub;Frr<S9SVcsyz{OQjs?2e0#qt>Jhm-{@q8H=5-@9PE1
z(a+NX(fh39_%50X#TmUk9<cY?wMT-8-2NB;J#$dCPLGL4hR@SyN(cVBQVKiKrH(&K
z<~HW4PUh?AXRk=3z$lMqB#bRXx4{oPRFMnZWMn*N;il3_43MWw`C4K3Pm+-0a}{<m
zMFLUcX7Y3-sv~I&_!LM{XOlwVn%!=0IbEzU(P37nOuPkDjw1?@zUxGZYNl9^gzH9z
zp=Wf$Ntk#%WnEF0B~@Kx%K~MLggZ|Yn^yCLh?)#ybEq(xAg{!xp(up7okEeh9gg6I
z+?q0S7bcqx5^nT#Rm<2K{4g?(iL$UP8cVSs7~BBH$?mNvI6XN*MMvB3xw*aL;bILK
zw6&znU5ds~)-_dK<K3EJ=w;U$0!3RL(4^o=A%F&J3$c$;3T+goEUBsnV=H{nbR#_d
zWW&o>x4e9H&CP8`*I9<qkaAY#Jg7}VsS&LlRU9d-!l;DRiLwyip!Kq0y=pjJEjXzZ
zx;D^M*aRthLImX_y&D*>Zuqh)Sk`BR?Je2`!q5_9Pt~kgU7n-YYsRXkJUJ!B#OMOW
zx{%#iN|<a}6TBmhff&GrNC;w-ym2F@UNDBlX1gQyJ);*l{D3SxRW4G>q98a=N}j@K
zT#{@{F&@3oXH)#-y4ZBY1w=`PZCRAmb<Ly4kNJf^{fpRAZZA34pYbFLVu<|y@BcpI
zC_F<kG5+Ev6$P^qvG9csNKH``axYaC0E@+9Zl9KAIggRLu8A?y_dPc6sH?I>5z*H0
z{P}Z=q7ZMsOnyFA$|dAJ83sR}U!}nN$T)g*7g?`zi?{3e)nEO4{40O?e+()MNsQu+
zH{QrQ{h#i{D+rCA=Qf>=!s3k)_LVkSIAU>Xz$-~biDZAzEa*x@AO?v^EK)_Gv-OFB
zq!d-vFy^@}OrIy)qDZ?|cv>+8e1xi55D|~Aei#WZP*n<J3u*ZS*4Wvgc(Rh1s11ih
z5*1XeNCZ)gVq^zmq&OizKEx;<e!%JZB`!k9o51ziIp6vCIXAZ(ypOm5+ipix6-8+Y
zc=qiEYc&rpPY6n3>IPdXR2*rn;_>+jeF%&p;dXmYPS5Cr$N7M@iP0zS#yv(WE|)9p
z%3<`L(C&G4dyU^k(!(<nC4Il;q*!M|;Aytc9M2R{=$Wn)RU)XskVd?hd)XESqXm~u
zlae*~tb(4{zml}1vX5%j)YOZbs;;nQfwd(*+n6}-@Y#MOaj*oLkny4{b5UpJHs{2Z
zoEoIw`-4~OYek37n;c^gh}tZ?%Eh-7I8$MIJkH_umc_loR%vt+izBgUQnS)MW<K<!
z@&(~TTW!#ong8yy{nupfJr%Pv2AYy2LuFLlC?#Oa%mK{GXdRhq-4Hs;rj`wU^rX<U
zYApNEvA=u4)u(Gtp1hATniPdIX^cYY2s)yzA|f5RNdZ5hhQ}QIV{Y<KI@kQ8R1AKu
z!cG)M5uYSW(5Gx05d%I&@n7Ol5u?C+$7Z)7ItvL=*fovZNJ=wyy+Fr7PIicaFh+*H
zqi;LOqby6B<$|)VDXJ1%2-$WXgp(<!z7HA0RLo5~fsEQIfn$@k*+(jn(5N6c?<{Fg
z5D84ywaQxR#e%vjaa~Wl9~lQl6Ai1zpgK>XlME!Ksp^{b;*@G%akt;I-E3H&m#74$
zu(Eqo%q=}-HBGsa1=s~H)+IVAB94$oLXy_(v<p+(fK~K<WIsgueam+1`N1dGynMOk
z_O54i3GXbVB}OZ>iHOaom>m};pj5!9K&?DYWvI%6sw${!%WB!MS~WrrE-bY%Qk2>P
z<pbR?vT3)NZDQ9O+E1Rs5Lwix?1OkwElW+au$-<ds-`4bORQ>E7nh7t3Xe3|(HMm;
z4P{fw!XGp$T1KaEUasR96ZPo@8z=sT<1o-W&#aHuM*h}$f*T3W$>gVmP#q>DCTWgy
zzta2+KmNGC!<9AAXiZg$AmL~K*w6Crm)<9acrS1}rPp@N3Hq+%<BvaPv)M2@mw*37
z7IrFIoXFQ=v7ju=Ib~&<)K;q%RaM<1aEpq&n;#!zWE_Q)I@w`(?`N`e+jZ1cg|TL~
zb@0xiRKj~%ID(Ja%PmU5K6=2gUTVA#sHFJhlTX-e?>Ild#0O7R)I50bP^=H4`@+u6
zXC4zwlLXP74A_k^Xf(!Z*;Lsm=#diTs8R`>X2?-dqWGN*b5jQJ6t*nTR?-ECf&h~L
zEpb;lUMeNfV)uZJNY?8?oKsjUf1lBwITkJ+#@K8n@%p09w1^uSA&HDM9a()GAwgOf
zb8#`XhS3GidHQ~!A01;5;f1XVY-LdfhR`$Y9okrQVJYg0+SU{nw3Txfe8d)(SQ@&)
zvAKGM3t&t^vsmGxr?3@CB|dreiZ?6aIu0=qhJi;<9&+7w?1z>&-?(6i0TUx8N{m;E
z$c8b~0}@75vKhIYP}nKVK8BE0&zTz*z&jb|F;jHP+G1=$S=E@LK$`-L1+5s|m=&;6
z_zZ5Cy((QsvW^7|)4zX|#mNs3&mW!}<VT4g_16ak@=(C#L6)E3$G_(Bdz4Mf;g|$a
zF$Pr3M(Y!u;plIt*N&g@yLg1?O^=z*jjY55rKWj0VRVpdOTwxs=GI=2<D09O6ekyy
zx}a=oRBeb|#D{^t-_ebGd>jbT5#ymVbMM^z6L5U4!p<XxoD;<?C^cscdco)C!KIbL
z7M5nw(2raE*fNfevTCrUAO*?i)>)Q?tK3Qp-lI{n(3A!BqQO+<+<J{_8Wg$8KDtSy
zSUTJ=xvMEhZn4BT#kt6!^g)RA!8=*F#!&<bA>h4}fcmV!O;OB^qZ4FLYs+d;LMnI_
z_tcH0X)2V4(Rs@1jElM^D#K>8p=y>a%aRliaQwUzQ$P~_fHKlLBLyO!WTN<7VM)mn
zy`gU-`?2HI?Ut9XZn(bQa(%t0>k}askP09h(GrA^tv!W_7!xQ=pems(ErqqL%bG=1
zQsu9*si><;7A`F@R8|?r7;%GR+_c<muh<M*iZ;@9hW)26h@+!gHEeajR`B*a7la*D
zrD0(;tFpueMRoareTWpcCgyN$TUaT;GTKuli^4J_jSCu|6uK;ERwwMXpU`#Uv!_BN
z2o#C2C0(|?8O8x?gcDUXHHHAep_HL23dYc8xIc1S<o&o~ZqXs_RgL6+)@8|hz2=)g
z|8tz3o#%0R_`A<4@a}GJdG_on*VorF$+<7CaG6?if=(B6t97wh%*t46?W~jqDF#-n
z6(NXdX|Y()_o8+b-?@VQeown^Q7X~4EvwavecRGB4L*d~gxOk~3p**=s!9~IF}Jt8
zPZ@DYZ0=gt%?eWD>gp9wpMJu{#U<k~P}Mc3r)T$+f<LVz<>ZBw4h6Q>kW`d*;8e_M
zgU!1?xs*a9@v0X}M@I072WyIHcUcHrAccRvPCDm-CJx+9r47cU%!dmSfv7|UTa;zK
z$cOe|S(d_Y%ER-JqcHvdX(5+?YoqdJJpVou*U=Lv@bD6%^N`j#))KU#?R)n7j-em$
zE)am*yE{+|nq@;(RDjAM^+;@3*Jl23iLDpZMx$&&(w4#$ynJzmEo-W(p|l014LFsE
zD$*<#f*{Ii#izvgU%q5rDAr9uy>DrnQn<9)Rba~cmDZo|vp_6QQ1L*O$Wathc!Ya)
zB2v60LI7hBPrpiB4yvlcSbNVtYZADOZe$n-={RL2>@3bWHXr`1l5A>M9$&U2#O}Bi
zI~PP8n#O<d=J2yEP^Q;D%yCnz_-IX?t?~|@cyo(3aEN2e8=|Rj`z)^A@$<ZH%W|)M
zi^|tb3rztnf(QWb@x#cF9PaLx=JuAFf)Yz*YXN;yAS90uW9B>hT#)|wIsd0c`S@uc
zpAWGE$i<vXkufC37#UL}D%sS<5YVMTYp5G3%%X#kBKyAO=I(}OvA~8v^nrdDh%r)B
zC0J1!6-9xqMWCUL##ANBYAGmHI^s2D@9d1z#+2sRrn#3NlgRv|s4Go|TAg$BT~FV2
zbZtks7mnNH>6B6s6|dD;EqoJQiVcZIS~R=cEnS-(_mg6K7l=M`esRXKE*ZL(-gy>H
z&FT3$z4N%fC)5oXBdr^qo#M65oEz{VV6?{3V+tu&RZ%bo=(@<}cHr*1<He`9+}!Qh
zwjEvH=kzt%krER{2`PC>Gq5lNRbenXQ5dKzOH)_Wm8B{~0oRoBGi|K!Q3@+2#wzdV
zqr`aiVPtc=r@h*+>$mI~u=|LP6{+hO`aSE*6))?AO@`Iw3Ag<f^`fHGhBAn#WqEPI
zRX<Xc6)piWkgSo(S1D9d)Wrg?N;*~0r-U05rw`5vWli5ZhJIwz4lK1|B+YS0T1jiP
z-P}p)2%6jNj=?LS<nrPXf8tO69DnPV{yO8>k`U$SF!n#;-Js@8L!k|IC3X-G9z5Wi
zKlgJOYvxBDPYlyUp_KUby?XVE=g*(hb)8TEzQ8#ok>*9HEK7_P9^FLzo(j7uR%|NV
zj;%}jzMnTA(?neq1*`Rn-FCxrxuoxV)+Z-)UB~lJp9>nEL@6s#WdUO}eLn~)?+-{p
z3^4S8LR);9D70oAM_#;mk%{z)5F)G9`d(qDwZ0c&_643aDfFj#h6%tEr6I;JwIr00
zyi}bMCS@)o0pWO0)($CeB&D#Mj9HCM5<jNY!IYLf$IiM4F)hZ3k-Lw6M(3tNTpbZN
z#V$q9i=ldLEFF(6#Uw1qC{oJYw3(kaxe17nO6S+HCchsPUAt$$+cR_n!HY%7u5GCo
zmU`L9COk&zccduB<GEFBV!#+H!U$_grodQ*>wAi_qAUvW8&%Q;GTN|Mo}%}AoK_Mo
zW(-enu6S~BLaA!*x)EEI*rYH?L&`+{C;~K@SE59<<$q^A6)G35Q&Be+bBFGYWW-vf
zu!T@xnnuv?<Mae+w`OjglU8K#e&*KA=<hK?H&J8mT_}|cEPkl4n|?H{aZ_O~*D>8+
z|9-L||C#@NjA9)SG=8WMO;QsxS*7HUbm*vYK;fq6J<gH+tYYr|`-x21wkZkUPDX1M
zY(xQCTu6lNj?L{YBL!77q+&tP6*2$M2Zsw|?kN1Qm*78+Mcii{pO0ggFQ~M$go6_D
zzmt-JN7|6Gs8n@DNS+WK#z63q&2G!den(lH%h*-9Et&Z&Wm(QDE%Ky?k@iG9;vh`&
z@JRutRd!7eA}~<7XCG4}jDms$@7Z@9UDt7Yd&}+Zt%x#!Rkh+|dBSqJpeze9RR*GW
zdABuH{Eid@h0&axp0HkP+?W{a#BOi7*=%s*$jKWI*^ualk!G<#Ys-4IX1Ci>mlefg
zDcz0Sj#VZhL{B|WqS1(>yf>D88+rEW9nYTL@cff&cDEg2NTg&@#-db(!lPBhnuxNV
zs!kMzVO1wqW}q-x9a-1ZRkppTYmAn~#ug<?=~<Oo6eY@NhQW!?UWg2S$F|$Czv&3u
z78eKfLbF&Etn4Yn=?b^Krnz`Ty<XCd9qIIp_G-;~b&9W+j95&yL@!qCcdt;|QYwuf
z9aT|EBoneYT3zCcirbL59Rk|Qd2dl{H#^#G#~31&F7j-m@gbqLWxv}obRDHJC?Zu=
z(MM0A8vfP)*}ulW{OA7%eDu+G_`UD^7BNJ;8~91vps%<3O@h)G-hcmn-uu$~bL$H3
zBXudIxloER^6c3&uCK1=mgX0EC{!lDOF>cCVlIZ7ra>t+7k5(|cH-RubJSjparU&E
zet&*`!8r6>-&|vhg6(d{X1f(GS4J)Xe3I`HJ$B4B%mRfZv8&!g%$|~c-}1>PAJ6B`
zImg2X4{4f)uKVel6swp`Y^Oz)MAVsmsiKI!auEy=wSv-$r&(#JEfT}!!<-8mGUq4_
z1J)FR_E}4aP9hM|xwRynkd%ZpZjvB-ib#qDQ9H5N$met7R81uGl+qmWMg|Bch2lLL
zK>nT9a!sb9Wlmi{qt%oik}z84*9aJENLmqH;O_R0+nXEOw#$a;019IYilRa(H4AZ+
z(!xD1D@rS>+ithz^x`2DmZ&w}$+<0y3R@K1ZTAc*ph`<s*IZs+63UXRcE_kS#d^)v
z)pH(JHRj}m-i<hAsgxyQq_yNij+0avt4Jma7e6C*I;;P6R$Yc5_EBL{u*Rd}hJ>{R
zb+e#Zu2`(sl(kTTP)Z)JgzDjA#AV+`AER(6wVJQ}aWOa@Qwx3`cWJWvP`p-4-T%9z
zR_t8V{iyLh6==s5BZq?QP$(Whf6559{KX$MOU*^y^h`$_!Ncp{NAM1>b1V><K6lFR
zQD<>e#H4Hsgq42>MHm8m-xA6z%sx<VN6u44b9#;say`f3+4n8|IOO*F7wa4PkIL~m
z5j#2Q9+itaX~RxV-zY=|pLe%KVduQTswgO`lEIDKZ8kJ@gH?JaUlwI?gjT6p1c4G>
zA4suQ8Y||HdQ!?{Sa*5@$Yd6rqZ<c?eqguTvDw@)41;*mjgF!!IX^$=^74{Jy+k{(
z%CO(>xW0PDZnI$+dV+V^OAca^2x?ncio&9_l0aw=F>2~Yac*kbc8B+#le05++YN*B
zC?bodq1|rr<49pklrcigCEFu}AQQdOI3MsR+A;9r>W<&}&NDv#<Qmrpl24RMQCWpf
z9&14Bh{91+P?U+~x}s?;)<R_hjap!}q9`kxW<^~!!j(bd#I&`HUN~;HkQ}`LcsJsL
za7M<_(GFWSyE}aASZGVRToFz!r>f-4p0Zyx#O;Q(yx{EO5!+F5ZC|0@{TUwq%vVvP
z<LcE7AAIm3tt|-J%-2h)p4ECKh9ep6@F=djo@e_V8z*h7_a8lG8x!4bU>qVY3hxeM
ziBaK$C@{C19nlALA+3+IM-S<qUQ?6{{=%RCmwDsK8w_m^5mu`OZQK2D9K}y?C<Mt^
zGz-4=wXbn<c1j3QctppK%V_a*c<=f6<Bz$yxycLjeN^R-a!ix!Y<~dv+dsMZn#2$j
ziZ^}V@Aq^2b+uY?b8~~snZnjux~`qQr*^v?D8=dNDYtjGw7K{ar5vO*ab~7>Cf`$T
zWs<--F)pr4OVXO>&!0+qiLDq0N1Z+Rj(Ipgm1FA4%R8LkpL`$)9NJ20B3N35vXZ^I
ztP9CCc8<OuFnPXlqtEsZ4majNcWtmKp;Ljz<d#Q%59!eAS3=?>XiHIOj4kE!Cr#J2
zq@-{@(hrVBBW=M9*?`f)p_>=XqeWA^JET}e+I~jM{G*M=DmAw#nb5O@_!_n&x7XKf
zZtvtC3Q#H7+_jWtiPna?Dv2qwY#J6#gSLjY?-)i0T3WixQ_rfZh^FA9ryudbSHDWq
z1uiQ3p`#rKLbu0Xz2bYn_d7I&;p+J_b~jhlDsZ-3@$HvaytupL%TFG05)@Swbv;^>
zLLz$M#wDdlHf7$cpj%U$m|Pr><0xyy<Q14A&If{zXpJcC8$s*pYJstu7zK$Gr1oI%
z5sqTRrkr6xVP$=b8FhlAqDD>QEke}N37cC~hhh3z?c2mLV)w(pzh6imA!F&a|9<T_
zaN?NTuE&Sg;y$4f1^aZ*Oq(Maf0-JPfeUHdM$q#8-j|NE5c23w6uYYg!6QOTr73iY
z)e8!h&~cz^2h=i9*o3YUuJ@>Fr0aUxu4mh|?AkruIN-v4_k-6t(Em8|?LL2DhY%z~
z5_Q(hOq$?iSy5&#TL_{K8XZ)NiiFB$r+e1RC1@qd16pB}6-2}4mcledbCqv0Qm266
zLiW!Al87{d^F$xm?{;kN?zp+Wq3=5Cy5{1+1D-s7V^$&=Yq5pJd(W`%X|MJ?d-{}@
zFJ98MEsBUW29t^LG6BdW7kr@I?`0xZij)*7Dy%M8EGt6ryn6MDhmRg*1Sw!N_`avK
zmUg?P&?TlQBw{8>d%%-K4VogGw&TUd^U>2AK7Mw`riHRT$+2uB#`Kg$U}+Vl_7tUN
zu_#$AEX!p<QEI$zQ94qnf~L5ju4-&qQj{ejO4Lorb^z8GNk~8@z%h9GzGvuraa|9Q
zaU9qW8%EdDloe;i38{v@ffC7Y3>MnWme_P$oIc_9%Jawn-2arX{Hb4HeYPfU@A#d+
z`8RQ^B~E3Y=HPI{Am@^lGhMUErLpW&;?73~EJ+)h%LjbC8;B8%GI(GN0j$kqQ6j3`
z(2EK>M3U0n-fbD2r(Re-{{E-@#;^YtH`h0;mkTZ)Uhv5$&+%dW$!>;eTbw7Y6e8=@
zny-KT>lD_a><{z7?r~(4;_mjAPd@pCyUk{f!uukQIe8$7@Mz+>C`1Wc6ymQowP26m
z-_#<Rkh;}sMbk9m$K@P%cX#*16>b)Mz;3r=zu%LPLeV)#y{v_=den+iD)UOGLN*CO
zeNax&RipUy(@*KTj>W1Z#DLL;<#H*f{-?-$$>q-dB0JMe<o$|BY;ISYNr|74TSQ!>
z6ckhIO&hdwl0}`Ed00v-LKbe3)wm?kDI*&BHDwOebE{QPXh%`tTrr<p=Nz%BerR!;
ziu~#N&ewbPa1%62cvP>)TIGTxMoeLG<H){uZ0=gxeJdOD7%>X!Qb}Z_sVGcMRn)9j
z3m!hW<n;8Em;yg|`U!2{GX@dLjDwRgS}*Cwk#Bze>vS%%-S?ch$Y!@;=v(@B&$S=;
z?nfUH#vY|L`{-F$CA;l_Zyk>Y&#oUS(3nzDSn1NFkjeJRh-s~s=tZRxlTDpi)r^c|
z&o~akxeik?4kU>WTQm!*s+Mj><|HO1q5>a1qtEeS!5@;5QZ_f9+Pbd~xZ@*z#?M|K
zqti8IYIPn4>Orl4m}@@k@8?Io*6z(eoY&qTcb^?vt&_!$*oZM1?dRv3IzY!limBk8
zi{L3qDjS%?WD%r~wjw&eCk=5z><X+du{vOtLMO+`>YT;JW190znv-(|G(JQT+Bu&c
z79<+*^^bS2-Tm7A@{iN;xeB`!wWwjqXdQX-AsulxvgvnHf|ek=#lGt)ZAoQHM%UB!
zdsgcewi4vZ`v6|zv7|H?6G?LQE^UEG^ds6wbO5h3Lx^}c((YUCa(i_cdaA18(Zh$l
z^T7#qQ;P^7J9!5$!icu(*xqfpdU?gmPhN3*b4OK`Tt2wOS}9Je!idL?7T=tN!0lu3
zm_jT9v@z6GBVJM%tQB#@=pEW>y1vCK=-NHYx?<bz=|Z5;;C(N7nn33~+s^aqX2<iZ
z4No_kyEXwOM2+u88XH+ufpr5XOU?PJ#;8PD6`Y-}C<{eVYT9<gIQA@?hUMav!kkf-
zf_S(n{%u8FQ%;K$FuIOm><P|iGf<JUyK%_Ho!GF{b;Egm%F@(yalojEQU+ZWteOV3
zZwb2=w_fq!?Jx6{U;Hyv7bo~QQoXU}!+-J5^ZUQ~oAfSlb9)0_O9~!cOwhf~D344o
z&XnvFY}3ejwLmvD!}FK2U?R?QZGt*MkpLBG>XIAZ;gUqp7$3R2y(2`!c6ZDF{{Qgr
zkO*ws9Vm}aIsfw~In)uX1CT^Hcz%Atmp}Ya#620|IG(8Qsl8I<(@#I;`HN@l+6`kE
z1-Vs!c+_tC+rz|9HkmE*@1%_(Do#OW@3rp~_~1!{yn!i^qU$<_VVKc9?<H!_$3R`z
zv~A0Fw<D#*YPDi?PV9P`CQsTa^I6D6-Teh4h6K~5O>067ynOkBu4`E=)>u=pS}Zs_
zIh{Wjd{JfaYv*wKO0q~MRL-45vPGwbFmGOyCeuZx1x}V%6@)K*Ac?Elh<vh(FjfQ<
z=zt1FkfaGq%Nxe2u$y9=W<q)9YT2TgiTRTf`JQbB_xW3MAuzYrCM0P3o7aq66(T6f
z8?CM%*lqS~Znq3wC;6;O?0%FnR7FK$3bZO9YD!zPXjYu8PH|zB-)WbF+BJjopbSM-
zqin%j@4ZLw6PtaH$8dXhO?!9C{_Ym5BERt)zmCyR7KRWL`)o-yde7B%$9YvzCPfjU
zt`LD<in(-+#EJQ@rt>IgAw@BJ7qm{4t#k2_q+qG4n#FQOT{l=;NHIQ9fKtNwfOB4K
zfBa;V@|sBD7<J1n+~eyt6>BO@KVu$OymmcbFYNB)aU4w~pK+($D;|&er1uM(*Z<_W
z?mT?1e5WL4p7@WN#16C+2q9uj{GkGGHff$Vs2qwyQF0$*-?UO#Rbo^Si!Nh1KY76V
z@-g-KC3<zj(C+a*Nm1v;N@wCcf;xInKm4BmaaGHHP5@IutiHPwF^P3VhHL~9l>}T%
zNu#lVB~heA5urq5Y(YPE?AsQ-(5T?K*<7<aIl=pgRhFV`WWDq<iA5xGD_W_nqf7Xh
zP-7r=fiMJa_9Oez^Wx=8?(Q~}wdK(pk9p_)FLQQ&infNRBwEC4g$3d$0)}DVvbnn9
z?&g-4FJ3VWibqf0;Jpvt<?P~=u5anOy%-`AbHrG-eE^bCB6TkeDhmuklnhCNCWc7x
z1JNatcZ3wtHsNseVIYn@B?=W<{Mgd>j?F&u^m^ozR|B`5VaI}mp-7H}^(>9!v`Ji?
z7M!d~mg|DDsPH~f*A*vcXN+SflKZGQyLe1l6(}q|ItY4JM^4JVl!Q04+1#<cy&^^-
zmG*td1kcOFD3kZ9Jdw$5vdZv=8cS8LP^%M4Ti~v)xP1Ss{O0$5k8l3-f0_FHjQ!Bl
zk2}s!mJCVr{g*eGzUSq$kExX>`jN6Jg_9JO?6lp;!Wgz=;yMO8lX(2*13EtvHrL{#
z=N&;otD!=%jEOe&Y;Fg3DWVMMpjetxd<k&`cgOF4e3wVHAV=HX{tw(K&xeA^q(+)4
z8lx>w-guLTj~-_5MIHDtQ$ONf+a>aYkH62Wo0qiX4j=s?zcwAe&=EBF6COGj2;$!)
zyjN=ts7O&5wACb?7~DX;sB;*4!Ub8_md%o?s^&KC#H-uo;%ywqxww%QzF4LdMUh*C
zFoZ}|R_wbLl*#re%tpS`c~DAF_o5KbwlO$#NNjdnHk%C(AG|?uBm28snzE#{g{V%m
z5$qS8t*i2qk*1W0d=ChBH($T3pUC`tIS&(Bl9G_QQ_RJfoe@qS0)7lLVxe?I#YhUW
zflWc7v_+y_RsKG4r}QjpnbA*c#LH8wY^EOZLCk!#K6tjN+(MO0rVf0m1S#pmwNwXB
zH34E?Yu_rxZnxq3>Xuhm8^+#awWhE}ToG(RWh!*kjP1zocEk2=N5Aj6yVwv3TwULw
zZ9!4icx4&JKsPvSf{RCQusl8E^5mR1ZGpqEzq{q`)9382uXy*(H~9PC`ulA5TUswJ
z2H$)3oH`{I34Z6Jk2z6>Vhj{ZPZJwVj3^bw*+^R=xow&vCgZay$$1;yFh_qK+OY|x
z6GdrRot|-icENJBpeQAwXWF!Q9~ei+FpP|DoQv#462v$A0;<;wId$LiM8JnU3e%*j
zKNMc~3P7fs{I#vlkC`eyQ-S+qj%h6EY$-N>9UUSNlCeMC|Hsd8pJU8C_Ck<NqP12y
zsmjftahiKnGTG-&l1_{*F*<J?l-zZi5{<`bI5}x}c=?3&<rD1b8J*IEZOiB)gNtIQ
zJ&br4iIG`+^oQvH9RI~XYGvOaB6%Vmb|J_IGAL1m<(5T`Po7(R#+1Y)8F^jbW0Ydp
z?O1JYSuK`$AE}#y=mSHyBgK+bB+{6ef&ikAbfaU__Pn^e<@)xHe%q0P<MjNTpZ(bn
zdE?E;6jg<`dAFS;hmr_!9O*~T*md;#j+>XSxVye#zuVE&3*P_IyS(}C+cc{M7+9XK
zaBgHAdQ_CQtSGdkRhmFzyz<l=m`p2ks)aFF13_z|0m8vfW$1cRyu~aQ^rK_nC$6`F
z7q<gXuSRxbNze^igEb?iax4qa*;4W3;Rz2fR;-qlEZj-a)C(4ihGFcvy}f3)yTe#b
zRh8lgRWBhGXf6C3H;NqJ+JdsIFxudTk==HKqyt!4rg96$+9EsA7ZjBUFRWH#_L~f4
z)zB>0R82!tnr?f?N#C-)x#nhf$AwbViv^2Sg^HeFEKysoo;_!CeT{Am&JALkn!x1E
zfYyr2OY3Ct2@G63e9YbL1~&}&VG#FmkO`Im7)&-GCL>srvh8>j-#b3<U3s)^l~^$0
zntkx)FH=>u&=9gA>b;lAe=}uIe(!tV<JHwG+P1}qABWU^{z*qVg*k~}LU(L2d(2Hp
z-Beulea|eWC@k&JN-MW0@ZO8XN!tR<T(_G$;Rdc&D~h6^>$-c3)I^(b`Sp%DhW92?
z1w%g)gQqMFS|^@9d&cumKjnijeTdeYk3asH>sMFQRmHyDgHk`8Hc9S?=7A&c6^ZIo
zDJL_i>|2+}LW@X<l3n6rBzcebL6SZSD<<MfQxv1b%f+cZmF9UWMvq3Llo;P?^2^V~
zuT0f3PAwvNvqe#&l_4t6I1Yp$rWYwmHuqsX9HFGseE||#n1yOd2`V|aCjnAUU{LaU
z`+dvZ-7TBDJCZ3`l$zinNxzaR7#h7~)ztLMh6D^TaktsB-*0G#j+65<nzCj;dZJI<
zT-`9Bc;m@qKKjA;`0yJiJbe5nFYj7TiU*vn*Tn8kHa9Q%SO3j_lfU+#|7U#b*S|&6
z6zHlJnn(z=&hvb;q0qt+tg4bmO;J}O{)+*;$nZ!;qmp!gk`Khe;fH}RIHGexSBZ(T
zs#&d0IKO<z#pMI>5L8h*Eh!MA!}*?Z>;w%8Q&E@EO!T}*D~!z@pd(~dPsm-KOVsSV
zDoA69^At60Jm$hKBYKeUshCYq=ATDYwDVX~D(^;8AgUzIvHTTKvR)pQjR{aXqsc}C
zjue<PkEc!Qlr9mKoBw)xU&mdDl*qhC;Up$4Yr7}NW>jf{4FoUr3)y!o#H?G~Day>T
zPiQnD$2DuvC5!bXx;ep7(GPGlN`~>i8>Q&(2ez9#hH?0#&J_PhAD_RlOU&k^f=Ea&
z+$t?CFA{nht2$TT#!{&o29FOBRg@T|xVyVy+0<FNRgj`%a31fnd2iJ8ePX}$Y}=8W
z&5o<P4cmTCQi*q;ob%xaU*+w$-=eBY3A0urbTHP;+k(*bY&RRW+b!*G&wjh5-M3g{
zc>MS=@4Wjii}eX*U6E7}Vz1E{T~n6@gfy$m;t}EyrxTD81eN`Lq%~``hG+;zGY-PB
zz*<NVqcyviSB#Bg9}O>dfoHdYyWY};8m~)Yh%{--Ne#;;vToq)#BzGBC>s!NuqcC<
z%N1o=;Jst9SW?#|DFvJ#*>3i9?T9rMb-m1mofD+a*m*Y<QpAUm7(K>nio!xREVWk9
zo5Bhy+Mp#%wyIH8MNw8F1PKXrqEv?A?wa@BenN1)@Q<`61W%z1i`6-AzVju1_iz3t
zzVC^qV1Ktk$0$z!6TvA*3niVs3$<poTyk}FC5mVtQBfh<J_&G-O0eUY+|0iW`FB5^
zL)x24Dc=9m`?7#WxL?!BtfuvZ0BLS+u6X+F6K<|=aV|{Jl7Hwn><=H)WPd-aF~x&|
zazc)rb7-wuuh$esL5gWc-zIl?V}#?jTCI5g{5fR~J0G%L(e&6!^=pjb?(S~p7X7$(
zM@$K8q&N;CQWRoXz1eQ@A)uAvowwfwEectq4K9Y<Ui*R@vf7jMQ~*u3AyH~Qolq*s
znN124GvYH@qojz>ZRn|W8AC3zm7%H{NP1S|r4&fX2scTK-<vWZD2I54PW+^~{j0M<
zu#(kS_=lhj6a{!EmP^JOtO`;rn#@O16ZRt{{bZEPXXLPSYtSYmzLT+b(xm*JuW#?z
zZg+U+Fj`{Z3KYvCQe+gnv4Ms0R7D`3oN}|>VS>Xr>FO*O4T{9j_24v%bw%hMW7qM{
z*S^BHe)}JA_UM9#Z$3sDMboTNRY|G~>XVwUs>Ic9Lvy;~>9bGh?rteUpp20wMTV;4
z<v8HB2~}#IoUG6xVunCS1LtLhK@kWHPE__ONN)JBZSn0u7)IO>7{^4lT5@`R!Fqjy
zsTOEepb;LKcOGQ|TFF-kJu$>=ax8XUNu^n|BoSm{AvG1()4igzy^NSA$4qWg=%gjl
zAp@sVyHjP9ct-Ow8WzH|nVV(-ldrpc871R2YDXgCRKnzrjMW;YEwYN|yj>)QF(7({
zIB^P55>Vj%TzCpcQ#L6ndkB&h){;13GT#s}`_7Yda3bD`IuU)8^QI*iR-ukkKNMDL
zbiNN`3X)urWb<$9hKKKeKz;gvP%jy^B^X82B`!ceIJ$O^AKlS<eq4C|=pFunZ~OXV
z()54w_kRd3&BRljp6`d}$j?8=eZImjx0F(v{Cz}?s!?-MCxxAw5!*<_W+cEU%it0|
zCCaj3bPn%FaDidajKfIRcK9JO?h@NR@O(e;qIGQg5!VkiN%8jik}o}Zz>{|`Ie&Oc
zRTn5F(KiY~u?SHh^<77QyW!^Qn%#bn8%I!Zd49pg#RU%@K4h_6h%iJ)lrhBYYOSnA
zMNLQ|SqEviPP^zBCipq;T5_GIjLJS{Q%)U>3Un0Ts%CY<?W>k&&#q|oicfcmjk5$*
zqOGQi5u-ilh3C;}!D?-3>d0~lRq4<;NF_#<G?n;8b;H2E>xc=Kt5fQx%wsIlwSC@w
z2F9+#7{g+*q^TQ}5(Ls1Y0>%K5o1YVB_79^lCr1)g|%fi`;}Z&j1=*cuT{*vG6QTr
z{glT)_W||e%h@B!C@Cb3g`4e;yPF#-5@Xl#=_jA?!G}MC0ZL;bDQPREB#9y#Y*BG?
za>B>o|A=)_(RYHTE;Q^zw%^f`<(VQBT97K{D620JFwZk(NRcsln&pBgPo4nB9GK~W
z$jv?}42gX7z3=hz#S3n3?(lBnpYhr~{D&`ECd5v0-&$aFBUNeuW)tEVBPS;((qc`C
zlhp|+9)uJrr8%i$x7!gyV87qvoMX9M(zY#kcX#af`x%wH#{;u=RvAu0j>G#@Gl$jr
zh|$tswF#a-f5C3IXS><)^3@f;_zS-%2+%Nab#u+#W;3H|U(_+RrGZ1)nAB|dF}21N
zhYiS-T9A+=g*c@@C|OY_!}P)y#6mv%bUxH1gH^~}F|FQ^3(Is^G?GFogU)=uG<mOS
zQBGQ`82f>!5<2K?ww#N&Oa+*SST6E{%H0PgpE2cyUrnfw(xf1fk+<6|ZP($0N9!bQ
zP7S7jT0kL|7OE0VWl)zVeA@5nL!kChDMeWqVrV~(=+vNt!jB?wcyMvXCqMXoo;~}3
zcfR}qWpzo})KF+tWwCb28!7V5@t<LJdcyDi?r(GV>J_Q)U>s3HN9<bmZUFCD4g<|t
zQKmpiM0;>!KpREU(rxh(w2!#nF}6LvAMvB8U_(-vdcorKg2l-x+Lk0!O8YJsE|DPF
zxM9Tk0q-*<XDXO;k)<d8B4nRL=Hl(ZZ4|gT&hVU+KRRl3WG)yE(ucVPz_b@uDqo2C
z5#s4N;lm0;QUOJnmAQf~PGe9I=`_8HQiSBQM{0g`b+8|s{rM0uN9#yTAtTWsYhXqo
zm7X0m1aA}AIC;762#I7kjNAF%p8OT1B`o`6xd5U_4kQJ-EI7G*K(SciY>7*Xpdn!J
zAu^03eb?b#yibw-1B$x)$MNH4DEmkJ{wICPV-!F!@AB1z8~v0EyXh^*X&_|i)GCch
zib9ifM&(hch|f~`LK{q3NFqb<($-VpoX3wN&Ibk;x!bnvx1LuoBTsMld^(QY1^Mi6
zUetW}&SQS|(K%0+4WZVIZeY8M)K!JG;=MGEJ=^Vu>+5U!yFFnioC#YLJb3Vcle1G!
zPtRx;jgZ6p9;}kxn3{}&m8g1sl&sP0NFTC2i&6rSPyapLbV2~ngI6KNAqkxL&7}x-
zxuiJ1;OhE0H^ZLm!JzDtLPsnei*n?w)|{1=2d5>?Qc+ZadSOw@WK-FS$~G8n>7C^2
zMxbt%EOKisW#>U#HthF1u3p^`V&viF0~V`go=9ZZZVDs0fPKd}jw}|d+^TJ|_m;()
zQb^JS*(IegEBBO<o7R)Wh_r2SL(dy;JSLEM{`>`xFE21kvI{?bb;WLThba@E-oE7f
z-}^3Kd-V#dbq@AYSYt^9qEh$(i>juoYPOpl^-2>4kM_`*f+9E(^qB0c7zFW<1<#6U
z=uem|@dqA9HDV;;Q()Wf`K4d}JJ`D7t#{wy^z>9RTFX*I6iO!f7-XmU(MKQg>g7vr
zZ#Ou1n3zBN#XkQbMSYszXYiuL?7E)Ca&^zTB*r+4BZNn_qieglI13@L+wGWq$&Qt@
z$Np(kVnx?=a}gIZ=j%8&=!Xc}N{c$&M|34bju0ci@^}9(N*R9hH-3lT`Q6{;7k=Rv
z`M3Y=e;chezx(ZP^Y8y({(W}4{TGYcX-%3;<)%(%lBGze>B3I}7*G?ZXtJ|Fp_Q2z
zYk|O2MrRHlcv@@89Gzy<6o^q;yvOC-T+|_aJ;>3GT48ciLUG{LQBLUaMC{MPm1DQ!
zjNlA@I2a@=jgQh6JG9e<y3r4VwBoy#!Hu$kR*BM1tu}c=oiyvRq&{h|7i-+RZ}QRY
z9dSPpe9RwXYAtJOv*7M_i#=OXmldb$6}#;X-}%mObN2W#zwqDti+BvSvS<}Z(Q|fj
z$=Aw~ci(@XfA*jI@6+!#Ts?ortLM*n_Py`&`@jAVu%SV##Kw8P`{}1VUaz=ZHkhK~
z=4MCJl$48#dbMEadg8vvwLNb1jDy3uKt!XP3RN%Bs}s!n6l+STmME(rjOgUq?OQg(
zp8c?=9eR@27|e{sseG@be6PqRPEfqED#apY?wl%&w%O-Ts1t=P81_A3^c-^@l?*<q
z0gmpgx?arp)kLM4=9U-}Aq9xihE<c)J0?_&sF35;a-kt^57Qu=wyG(&Zl%Bz<WW$n
zL_!(n4vLI81CtPG>TF>%>Yab4r9C{)fm0Qkf3K{d`Lm7BbVgGoCQnF`wJe)#xqCbc
zk7f)ogg`%z41Legb&NwIO6+>RbAHsJj$ZG8g3j;Y9wGW@1ONJS%-@cl;E21L?sb!-
zm=<oV)IBN4`JE_mKgsbq3%k4n&0jk2@(c=_MBZ=m&Qnl=KuDr;OHq7otX3FXQ5b_&
zFuFj9n(Y|q9K75`zW-v!M>nwT9U3SVJYLuQ-1~3x^)J1_qqU{f0f!VilX)t@&~@D0
zUi0eJE85+TG)C6Vk~iOZo0F3hszoiClPNLAK=h8JpeibKSqNe`qa2A8LUy)xs3a#W
z0nvwSL@FrZ6!T+Cn<4c`3^LeVh`5ySN#hd)V_80V%xdR(_WdhC3*DAwS#Y^boYkI-
zg`qYzO;e+dw1<_h^5kA%bwN{AM5S<X#D|CviMlC48Tw%&qGl_Ym{_gXa?#6D9N>e;
z7%Pj2Qj}$l(R-Yi)?iuI8M!O4Cd>O%LZpBPQtcRI(hf@QNfjbE&&wCjc`JC(nzQpW
zK`e%mqAXdhR=A<#X1n3VCm-?h>IHY3J8V(VjbfOp5#^MRks*RDEnPn_4kJS`ic&Eo
zY+)!>WJrNAZ#e>)!x00@T6~%|BVW*w4&M^+>iUZR<G=gw@*n@d|6hFY<uCJ<uY85~
z-g}R?-+qg?-g=V<4=-7+RxB3{ec$t)Z~q=Y_`&z-yIyE9U(j2d3cA_vExW5vt?((_
zyzeEW*;-4}G&A20pzT`D&(87Qb9Z-#bB>}YxV^n)yWP%zJB}k=Mt~=?+zA1k$j`@<
z(d%g5@z3InVn|MB7(BoETfa@$IezzdeviH%`1HjKe)(7aYaTv$$l1jOC+icjSNQ@j
zt5Ow875Sbx*gni0wduU(_nvZL7W2fb5UZ11ip^hID}qrNGcB}A46&oMYoivRE=2(W
zChpTQ*9piQPwA2blix#>!1EX-j_?@Si7~%F4vU4X9pb^MY<`dPChvF*NzpzTy6<*7
z#&KZEOf3pK&%4qZ)bw44je*r-!Mh)Rh2Q@EQ)0M<-U%{mP3C!7G+A_!yC8L4^VXYh
z5}fD7^G{JKVU>aCM2w;oRV9klv(qz-Q6%TsK7NDEJMVJ!_S@`Uy`;Uq!gsAK=-WNo
zdy;dM56&?@V5203=&0G<?Qq)`zw1Q6*7Z=<oSt3q@JnCj?45V<+7gmtNbvE#CyfKq
zk0jr*>qiC@s#;R5&)7V>qCncG)>^DJXl?RZDJ^-4lAPL#d3>Ny0ip|}-e+pLqSZiQ
zNJbIJ%6ffRdqo&?*b409!N5C_?twwmeW;UAUPOGBuqj}5z*tWag$57-of0mART(9W
zxePj*`yymR`J8*KG{zKYC1W}SMf3+0E=DQfLOR%>L}b37fc%+0lIIScT``2pqUX<Y
zY#)_!vPr++V}mCl?O;FljJ>2vwQWn=wwZpwYenLZo)h%z72P~nWMacS0j1Y|ua5qG
zd{vIG$?Jb7?<Hn~_X$adgjV7k{F5A?lVg|W$~fhBtCbdI6bh?EtPo-#sT>eaK?o72
zJT^N;TVpA-0V2Z~=$+!_rsw%>&$HW(PwqypQbC~&Zx)7k8_QQ7p71jlr#xPks4CJM
zg-HcvDltYJshu0y+}yC)Zcrg{d49>s$(a;%If5qvE{B)L6vYj_s<1_s{qulqF6oB>
zKMvp{ecR%Pk+KwWxfoE&B2|<nRcXckWGc{=A}ZM}J1^WqaUfSDqp{ZT<n4!CZ(G8v
zJt`?qYs2YMv#JthlPDG|QnkS6IFyhImSsa#H5gmqRAf>HDrK^1W`lQPs(C<3JXKwC
ze!hYPUC#cLVlEfw#!{6_wB8bYi}#5|v%nY;PRK%?!?BYL&#CxI){v|rpztYzhXO;;
z(w7M;u~^hFMzk$RKCoQYbp4+G`URWq9pm8G?zZ@tIT%L3`k)jc!}qK8ip_RQ@QHrk
z<%Lit=|UTs$b+8|Q^9`Z?ulP4u5<5*$e7>mcbJzi3EuPclTT<GDa==^6=!EBtk-Ma
ze)}yJ%LU*1wZG5H7ccNWujF6gyNfXrAkz~j8v&2^2A7LEV+5g{5V;9ybIwtf)m*@V
zv_ALyJ;03W-QL}zl%lR{;E+-=5yE4NSYsJSH~;K)q;5`ch)H(8D&c%2gkIW5amV+*
z_dVLaXXr<|p=Yz*^6c3&-g^6;d3=A7N2DYvn|X5+KQLv~EzQWpM8`-uiYqFJDF{mc
zBib=4bL5mZC=*d$P&Jei)Fv;ga)OwRY*SLlqH1ZC>7#{>G>`77Ej%p{ub=aoY@gqY
z)`y}FIH+jz^_q(HVHjo(<<#c4R^NNPN{Vv1V0E(Q^x~W+Z@)vmTClsir5i`;vcMw3
z28F>5p5<!AsG-l6O5gZnUx%vZ^zxGJ?H%j0hj~M3utq!q$1&kX@mka=Qq>I)9zNo1
zx!}uR`6}P}&0lA;-=j=H*);5jp5z>rR-7*yd`yg^$I-IewfKF9Z+rTFWY<U5>m_Gz
zyvbL7{^wb~^ByiLe3UN2=F=;7H`jFg4MS(R?RMPtEg_BKD_XB0N`lA`!Fn}&p%z9{
zm;$3Re=UK<V2#CSlh;R;H_viiJ%Le2f=UFUSb&kxDVKvv7W{mHiD^QA(p-3C_5jBj
zb0&dJQpjr}+6S=_i%B?a`F(^KvU*saC*~HY+zW9Ydx<1j-Fa*=(Xq&Wx0B_O#8hSU
zP~MV`x$sNJ1>3}BpJLyp=RI^$C-Fvhw}TWI`aKIb$UPYpejLPPe;gQxLHKnXr)%VM
z@)=XGnmD%kmr5P}J=KD*!v)iOn-EE*69*C9ltCTmIYNqb!|@T*)XImc%`RvyKk@PT
zeC$#>noKi<kIupgja31Y7Zart&W|W1?&BdOj27v+jpA?TV`A5iyxfd@|ElA=uX=91
z!ckEYIX9Xo%YrYRH@yF#;iUB72UMzoGQWL+ah>;cZO<?`mi3b6?2^@LMY*V<l>Ai5
z;!NV3WetTX1?{sX=#q%oY>dv~-GJ*meAj{Vgkd0#p5P_VG$K)GRaH?nQtX8kNJ&~I
zAx64kpzEZyDI|3-Efb?KvFE|M;45!l5_cWW1u7M3ngu84CoEP4u0F+=i~KnaQCWHo
zNkw`;QrSRdEaM;xYO`46g+Nm`iy0XyOM^Bcb1;Pe=e&m`3$o6OO<^mHsqns~ENY6P
z%B>ua_c0TD#gf6=A}h@#J}0GwQzBw6v_UCL-K?N22_$p~Y@dFL)r#ukjI+}d`k`gF
z-?CUVD5KczceF!`*R{A5S%VHHb2DIla>DKH9a<N-K}hrg;h$u1AD1FJ$!%jfD9C==
zhYOLg-w~W==z7Ysps0!*6kcQxGMNaq7V*U0?OjH7r-k(Z&;MbMX+aJ#Vp5!oC+8eh
zRTHyVVrrpIJUf^KE&(F6`<BIGG27!zO4wl-=AVsT#`Pu}EKdvkY`~k(`~98u;qx81
zc6lL4%mMFOCr(aJ_{xVL@~v<Ey7)vY7`>xyTfXz1??|EXM><<Sd`MAe=WBJ~Mdku(
zYKN(q5j!o2n93?aWRXeJ#6NxJBZN}vLs4~{8G4KeqR191`CO&>G)d8y+cl&loTtnS
z3^CiI93cVI;(qU3CGJ_9OmA^2^aSbma}t9nZiT2W+({)Uiph7F0+Oas1!Y~)ESH>K
zT+l3+w1p)qMNw5OPuEoS0;1vNCmZV3iZS-$eHbI}z56b@SyHafXm>kSCzshA-bi8r
z%p9=bB_6CSD$1(Hu2y{T;a4S@;`)jd1GcK^yO!-Ra(!p0voU>Ci7-amw#V;#TsJWG
zjy@{R7Asbd9`X3gU!i#N7QIV^C>gQE>JoePifXr|@3st!1>Nlnx?zV)QIeQk5Ye6w
zn3S+s3T24KfGJ31AxlaqWzkBE*U?jQLmmq;N<&dcv<jKAcYH<F;onm$bQ&MiI0R|i
z&M@0VQPEhG7}TeP3KFHKvXDYgo0<uU<G8TP&~kqMY`>i2@YFnwrCc~lQ76s?F@Q>H
z&f!%FVp20$Fl|u|_ukP*ow<VZ5YG1$Gw*Kf#S<0nNdbZz2yPU*SGKU3PT7p2ea0lB
zj=rm0Sk37f(_MaayC&}6XX>@*nm*NhA5LA1e4kGzStSKZ*4l$b(b4AYhu`XF9-ptU
zLli0!>nwGsRXG@0saZH72!!x!jMj6@e3EJ#1t|oEK5%{0^6aMNlWWhN(~PDeg~VGa
z@b*c?dyf`8c~G-HFLBDTZ(G*of-=fvLFSpn>?FNjt!bKurmnEY3g;^CwtbMo)L1P$
zBx|unfzbt{lLDq6TDtuXKX!y+!1pb295FzlO9&DYZOVd0QwzB_lTgvZuMvBQmNp|G
zqmyv=5JOg%8VV(5%Ws@CjE~NEe!XFFy2hMbu(OIET;1{e*B`MRJ<dmpqGC}lIB7~A
ztSc^7mIr4G9yFG+Xs8Myw`&D<VTnnjjm2mScv9>#s_9Tl(=-h+B;hz_!)9X(%A!Q0
zS+CY;ZRp)VNK%MfZBeE`*#e~#&JA>ZM;`*Jte~jD6li0)8yuI56(}jMwwpcM+gtk0
zEtl#&9zT2_=>)-ZesKy$F$6L7R8~o0Vx*0ckTF!2HJi;w6uaIF2TzT8F*_<4=9}%X
z3q0<2{?N_w1yl!uQi(SldniYX=?!<cce6TtYKtclyKVOx?|+`A{?Kn|TAYs;b>~L9
zPNA9x=SGqe|HZN_>HB_;5z|`d{8&A=S3?MrK9Uiewr#Q2&I|q2vWqcNmJ)N-^&K%q
z%AzDC32&c5)u;1xoRxe0`R7S8w+bIzK48jNK0Q4l#mI8C;%;-ti<d8Wb$$JZI!{0T
zm?U8HyP6>nK_x<-?WY!`aNd$qA{z{O!HQ(V=>ys^f3{;gm~55v!>VCfAT{d143$9F
z%xOy*Nl|mulKlOYZ#u6EhnCQUb|jh0jgik8^8_v9RL|{I?|a&|rR!QrWAKiYBF0P#
z&!aH=1Rz#QJ}2oERmtjfO?R?J2YBPnH+lQZ?{I#8i5nvS#rOXfrYK26FA7toI6ph%
z{Eas$R_A>Dd{6X&LR*Yd;3u<SLs?X!(pP!gt3VO>#?So;UOjup?&g;5r=L)14c6lQ
z$n|zhVKk9Q?MvJ^vfH!_+YaA5f{&!aP+wfIeB&*Ocix41MZ4<>%5rh}h(~XHiAO$2
zVWSiM%?p0;=?}QMyJEk&<NLqy5AkhBbPn%4CMK$~AXy96pp?O<gat~YW$gldQVd>U
zh{D&5QT!SSRLh#mDvA_{!A}=Zx+Rl}BX`kEHuV){hal%F7h|FXjp*zoV6mCtm*$LQ
zrteBL=Q&Myu3Sv+*ok=PFs5RDJs*^qPiGX~dvGE9{dp<IMJ*l)f)t=k#9(r5!JHm6
z7guT$nUKX7xuBdW2e~_wB7SI5(W9eBheQ|${5UcW14G|SS0Nn`yVq;G!)uZYTp)AR
z^W6s1XFvMeX?@AMn%4@u$>IqP!k4`13!w0MqXm;Ara5|oqL@_Fk#79S)`rhj*hOGS
z1uhx1ij*Xl+OyCJn@J+hLy8qj7s6tX60$7ZJA)}L(FXbsu6KzSTg|h5$x9dLmBxpj
zGp%{QDR}Et@%CBdavi9Oh*mX9l?-E#Q3Fa@P^A!nEi6x07_BHu`I$o|QYrcZ8ig$j
zfxa_wJS8bmeCO!fj=tT~w>!q1hLCa~bWs+RRYBC2s;OyKOH5Tre2^8Fc^^D}bhNvc
z+uIww4-|!^vYE%06+%(Npi8Xttcu9v%ZA(7aaCx3_1OzP-E=HY9`pXsev?1*&2RGL
zop(9CxI}9h_B(E#f6OPp|2us5laDB$-SENL8DD<bVBE-QqtJT98;eT{mkO-1=%@+f
zKyU%H7NJCP^!+ZkSp!X_uvSqfSe7-aC{cBZ(g_=(4oUbXMsvHl;?=9?^g~a*SkV}R
z(uuAc2+Nx0OYd?1m9L{+MVSo7)?6oOQ{dgR6_qb2O-b8&+O;9s75h$59vu_b6sRPA
zbX8#}jG`(G*I5bM)Fop-ir<&c3bZJEBcrwC=SB(N9)iblPYCseA5)%jNQrjrX-BZe
z%FZwsYcU;KhcO+OZ(rOyi<9?SIz(4RYltb(_bs(njIJkyCVSUSZtj}0Oj4lvh&C`f
zDe5+x4SnC|Xuts<JTVF?Q5A;LYMH>DV{~KwtRZh*^-;?;&V|5)CLgyAr$Z&Uym)}p
zhEG5Jl(*l0i@GfN)4%u&eDLKj6H?@EbI1SvKll&$hrjU~OhsY3|1^K3V-mX^cH+m@
zDpR490F5%)Z9ilsUyf8#9K7}>p3dC1%!xI#$N>(d@eDyItq&_kOdPT^nW4L>1$ubn
zKjt$Xp@)$~%=Q}7=0~SPvPF!MzVBzXvvZD^ycB<$jM!xp>`dHK$}oh4E(%OtlZ@iz
z;*uBYBR+j~MSFY0x1WAQT`#bvLRSUb{SKErjV)O$7W{)>`-glf!CUVyIX^iiIZsds
zf{?@pL_{4%1X00dq%2XW#Om=IeEI9&pxtiyufF>o?zVS4K3#KivBd4~czJt+DJ+c-
z>^2*^+Z}G#LsBeHPI!27#^u}ZvVP-DSf6pz4m@oK_G8B<y`s@ItEyquG@PESsK4}x
zFTeXC(GU30^UwZ|{x#@3Qr{8Vmi=}|-|iXOmUgpacYV#!wS=4@ocu`i9%4X8v5hH=
zaM7*NSOxq2mN1MNc@v)H!BR}}aD`Cpak-G23J5)28(>OpKtoEJ5VfGwS`oCtE5(>1
zCI!UCN5+}XRgKE+V3krXj^d0i#{9Do^z79byyUuiFI>A25^3U;MnUux>44bfPT<7d
zR$8MB((<=Df8bnZ<;Tk&Hak_MNEmu-a>Q7G3xs~h*zW1}Te{tjen0*wNNq+tryos7
zrI~FA5AK5V{dDx$1VTUmjC)Y%fzx$R*&9~R=-qKwqsWh$-bmjL`FB71-hTb~+=X4<
zcgv1b1*}Tw++7!BL@bI0hLw+OttGmN)~*OqvFjW!HzQYF<hl>EQQY1O5+|i$RViv4
zaKjd>3yfAY)e>DS=*GQd#S$nI)J=`C1u-Rzwdfp`G?TU=o3mQU@jO~+ob!0+7`i>(
zzGZA%!Z4s>!dio^N{X_eTGZHLMN%4DNV`E<Gpj*Eh!Qls-O=s##1N^<a!z@eE>w(>
zA$o#VXqzx5a(cGL>_>k4qZinB-sRu;U;eN8=l}fwgvVd{64iPwzDGtAy`*5IFc621
z&8JU!^__3?-~N~XIse_af19s8JYmp^CV_GhqNmI|kHHP>+ue*3IOilQHn-wdt0h$}
z!=WfFWo}pK!lJFF2nm}#xo+>S`S|G%c=qB2<*MdveS*^kL6toI@|U^%>d)|O*Ai@r
zGFGfgs**4L%-2vRfJ&q;v8b0c&5E*Kl1!12&4is8-UymeR~1$(n!2X%2Zr5FvV}7O
z8kKNCCh0ygQCdVbC=%Pe=lY_S{U3cCzrv65&ikTyVL$p1lD4Q}JM>Z%bKZM~aiA!R
zIjYY&$9~_@_X8T?=ZvE`=zH%e%Mt*sdz>55TCv~lNV#RVZ(C{Oriq)D(Zd+;@#}s_
z5&PPqGjI3px8Grt3Hxt8dBUIig<s&C-}r{KU$x@#<H!7!fBoO!-}&GDZ@JrUzp$cC
ztBBSIeD@&yAd4d=>~03~@`uTb&C!m0V&BZ|8I@;!H7#D|01dz_AVEMB5)ow%);XfM
zo&F0^@E*-^3TDxYw2BT&<U`XtX4FIDqeAI}_+fqzggGy`?|aFE&7VbU@hh9Y)PW`_
z3!4H{l~jucQ&+tE!3XT0y~1B_NbQc!Ifk&KFrKseA^j+m%H+-Vz3+aPs12rE^5%Qr
zMAGIo+Gg2WMoLo@-%THKp<^_vuqfa0@X;H*_3nF=ixn@QKVwl9Se-4Kv}N0N!ZY6P
zh@-4aRavn<zo0n3K-CMpl1@RY7kuaGclh>4&)DuA+G#G<XIz||^62t{i$|BNPnXn-
zlBy}FC(d?rMs)!;!76xy^Y~%JbuGa;l5?b?hpxr7dtBc#_ARdKq<xr$dC@zRDX`v)
zay6y-T8k<)kW>)wxZs2zmkT?M_|2jMDi>s^B%C(H@+BCl5R<VPNfu`drPHjIo|1<E
zR7^*Hqs;z#f;<Y34j*&T4Z$m7JVN6lGdDfuzn|{5G$V7_cQME4nMs6Vv&vY-EVxLy
z8z7457$b4$NM1O0$pw7h;q!SN`+;$s?t{Z|{O@V+bd2bkY*RB8Whx;m^Bj$+@_j$4
zh1F|(xP!|4K*4Y>(^XVPS+$z$gjvmu@C-|9=2^$Np9W%=d+Wz7SVP_wXeGjcY$li@
z_z-bEQB{_tBet+8UC@t_+nwXpZs2y*>=PtqDU@QZ;i4$1%B%}AmZDyelmuX-G}e}k
zVITy@I0|oVy%vj)3BsK|zOfc#iX((TLCm4eKIHbSaI=zCRLh29wGj4Rp1DmSnT(hk
zTUHcRNrIV}9&@6`ZntCTI+WI&o}JCa^22NALefK}BV{G#pDAhHd9UQpyz>r!_Fw;B
z@y&nXFJn$0qYDvy2(L|(ERX`)HfW`|c<WtGA3fy#Kk;+?;9K9~@Bg*G%8Qqu@Z@ws
z69z7f!TJp`1~y|)h=HoEQCidHg?PPQOUpc)m`0E&HllD7;)tzEbXF2?Hyb{E_Kfd+
z^qgKt&L1w>n3C1{oUeT27y0CB%m3rw`@hl=`QQABf0ksS^?{<$SR>7N5)5rGPU2;q
zBZf-uHe03$D`hLdqJn0%l7hOdEBbL@^aCcrBAI-C4z1SXVK1B^P(Lki;y>W{|6c+3
z*@rC7S*;ucbzMowd2XqBFUo1}1E;4JYXi=WSj}7%%+?%7qKm26Ynq1bUVQC>50cJs
z)E1N@9emw5LXD1&<H9edM3J54udc7C>zcE(Grs=yZ*YBm&Bq^q%vXQrYYfA{yYIfs
z&;GGL#;^R{|CTSRXjeL-RA7o8p9?i5)z-v2N;7J}0V+&zRL0FelCNXGGI^NNM|`6B
z+{mj=mLN(+$oZ{P(R(<vM+jj4--L+9+2$hM-{k7NASnnDj5=zk5}E8#CbZZ4?6;?+
zkdrYneU_=HJAyfkEX1ZPNJ??}=pknhF1b2c)BBMTPsr$wDJx9!sBDRoQsULim)N@D
z^x`q^f8|dyjLeJu%<Y~KO=Gg<h7>6n3&F5HyWsNio1C0o@JUQ;_8lrjYCP+rWZMly
z=ZJ2gh$0*->YByL2{cQF2>Vgi7QH&-W<T)VPd?=bpS&XM0}oCwxjZ@J!PyxX7iXM4
zIAeLbWVx<cEh-jGMblI)mJ5oyLE8pZfNr2UmvzO&hzTCmw}ftw>pS{>&(O6DLyH?l
zxR-_j)eR^g&>=vGs1Qj(+OlJAHF`f1{FvLW(kf0Vkb)!PAO=!y?JAY9nHwLCk_07V
zC5c1j4qGnZ3$14)Stag&k<qCn{l_FDNNm7D!XF`X6OTB~3i4?ukmj0U!pIKZep6v5
z=(foT7;`I@l=FDAA#{lNp(DA0lr+(K;?NU@fiRAE=g9^R^KgB=CHwmCWJs%d6E}@}
z(lj5)IXEs(r<r3qv=%<<M0t_#mXx5dn!2uLXOGF}a}u=0kX!8tioSnw^ApWupOa%J
zg`GiD09GaW(3#AdyAVuVycA$`BDFWxVr;=M1nxGDtGj`nSL~9eOPWBSiHTEFuu2h?
z2CBN^^1(yu`UH}=T}SUQMTw1|7>OY=_B}&Uuv`}c-%9^0b61R<7o`t7izFv*bYlO2
zf_m9dlvZxIn0F#kl%+Q3SjNI)nhI@9&NR&gbT2Jf4J?<Ev)MEa)>?*PIBIVwOiI*R
ziHUI$Nv0x&$i+#GD;kE;(TxLZt4SFRM2RRN86(6~?<mTKWGtgrIHh^`vp>&YdiyJU
z>+k$m{MN7hEgsu~B91JT@D7Tq=KS;$qYeB0j;gBW4A2-O`~8l-AJMje!Z7-gq5$Ou
zHPE(T+qS&8-twJ~Uhrz`NlQbw8Tiw`_;pU+c#r@1zx*5g%76DQ6pAl>>3wWn;t4#x
zy5ix(2dqks4w2*|pM3i}y#3C*{D1%3U%?T%yW6rlJrgvvs)$LFEVR}<c=&*-ZYb*t
z=d<cH3Gp|kKvY^h?3jwYgb$*oO7{?uf4UA?*rRYQjm?`Gk15pbUDtIjRaLXu+|hL{
zO|uYXGh#}tr?!129#0eQI6C^iW53^v!WtRl#+ZX|0r%}qCN$~w!tNLq1I2p1LL0>!
zZ@j^Jwc_W0{^zmQaCLRXcfb2x-hS_0ii{xs+&90;-}}{H&6YlY^mC@Q7hA1#RIKD;
zP3e^RXouo8W<HUMF}IfT2RTM2$>UH<+f+@xt?d6bx0R+ZWC4jG$oV$1tG~yWJ35ZV
zK@)!wq?kYU54`_zBrDU-Xu?4m8?z;j<ehp@5o2Xe2YnB%J67gGHg&hs#@lxtwk)w_
z0Y>}*m6Zs?(l*$tVp%QdcLDFkKz84@y!iA}&M%(=F(V`r(-xc~DTTHr%36xDlC>rT
zQVIm;#0Ki@g7YVD^5#45@_WDj54gJBvW*)qSBA5<-sHt~OS|23x`d)$P}+)my{0)i
zVH*`!n+@*jnn&A~(}(X+ug@{nif#X#emi0qaQ&8-pKjT=zlkce_<+_0rw`YhoUA!L
zIpOl+f{U|Ls<Nak3aYZ;WU*jTmn@o!lSPBp4duy#>Rizj21=d#dg3>yMhD#pV*ux2
z9HAcw!@$^e>^66_yB*`u<Hi9ubdUluj6^@;hL+LwcsCHdC`7?iXia4-C?`czVQ_`T
z8$+S>+zM94pnO6{DW#_2qmeaIioB`l%i^g>7E5qcuo)CGZ_jPl1OZJ*lU=cR<0*ZN
z*eSHpb77>6ptVtn+SbQ7Q>~&;xNb)pT2e@)!4ZZ%VeIj4l>G60d(K?Abht-9b3Py=
zcCT?|)l`(`|7nElnoEzv<CSn(r@~Dva4ctMXRKB$7K;TJ7Z<G8YwEh5*JJ10oceQn
z{YvKRapVVo<8S|ncY6Q$oQPdSC5<IwfkI{V3^KvZ$VpO?{FV$ZBz!`M@-fhMf!n)*
z%`Py+njz{8!6s@AmsP=enJ^}Ca=K={z93Li+7fLFob#BZFsi^<g=fU&EK7i@spnR!
zn&8h|<V}_E<oXWeDp^!zRtt2lBj}_ciA{tt28_YhHE7M?<%Ba8b@};9TTnN(n9}C=
zFa>u{6Sgr1ZS*{uWAl3iaGsC8^KBmf%u||2Z(@ontL@|*%7D~h%$(IXT}EG8TvPGY
zf9}t-di03@=0Eul@qQ%ffpu+JpPlByx~HmZiK)?Ak~o0E7HqdYWnE+Ig28t*w!#de
zk{o>E`Q4u1{rHyOdp^)9P45hU?$7=LL$dsb|Ih!3A3S}58U~4Z8G7Pwi&2U5vn6$*
zal;;C3!Z)Seg4{i@}KZ8{!9M?U;f%x_)q`SzskPr+3q`h2=f9q!TskKmo!aHSyegt
z1ESGleyx?D1g1cB4sxVhiaEwD-G|!$X*(vqRE&YP?MNsVMOJ-g)UIuJ)b%2_+@P*X
z6wvoQlM)uBXy|(()c1Xdck=j<3y_rV8{-}jmIJc%<BsX~%hiIu@Axx+`cLu38*lLD
zn{SX(;_1_;ynOXCw=x{ID0uU&Hz|rzicbFM#rrsJ3A)VwVwPhokcg1S<K&8;#6wK(
zbZKcTZJ0jh>shvXx^I1&Jp@At_jc%05v;V%c0N&J#K^6zBY!K%1z(DSP-J9n{=E|P
z{XS|J=egS2GK*j|MIn)KDFxbbV7uM2*=!_-)j5d+ETqUUih`oB*uu<cT}t4iCV4TX
z_Yrhiv0gPieDf`?-+Ib#J%3KsELf~gcy#dwf9B`@IjUyCfA<gn9<O#+oS&bgmE!j1
zhG);7GVFIyEQ#KuN!YTKl$$iIHws%;QoP1MaDzlS8q4bZoPXvQev!ZRU;H(aN$A2*
zG%KEc`Y9=Tss#k27<J-}hnHNOUl26h?sj;f8Xd}5Ztk{J)q=%h&E}@%(fSF=YueFa
zlcKh3yo>bRz>62xJbiWx)?lrnE(N(W6B4GNSu`x0nx-zO>w>DZTwb1H42*rtyYD^$
zBF+yuH}JC`en^RBSuZ(1y`+3lGwxfIu{5hSbydOjHOY;+AdIBsA$kZ7hCO||rSCiX
zp{4J3q;Vj$9d5g2*zah&9lam$L2_~*uUFt5zHNE?!KGxSC$T*-N{fn@fJs0I!tsmQ
z`pi@!Xv;S9WSK^Y;6uU%h<U8W7_mmsys1c=#=jJDxjac(*kcvSYQpFVDT;_=<{~Dn
z5+m;I%}WMdvKs=!_Kx=M8rSZ@J4y|GRySmjTJFvK-!=o3qJR~&Ev7gl+f)8>nxo_7
z-i<&}6`WmMaCRxFR%d5tTwY#sc6P>ky=J*wvR<!QtyY|$pR-=CIXOAu(W6Jyiv^}A
zWTrWq3jn(PmSGrpb#=vG`x}4uCz{7TS7GO+1gEf;QbU2LH>B_ok2{e)C!*5mWEeff
zB>r&y;OK|MwhfHl5GhDOYI&<Xs}fc=aoT83&l}dO6*sqgKK=9uEEWy#z4xZ{+D(Zc
zJY{9ER%4vQc~9F)aJe$sMnG#6BPNq=eJ;X66thcXZ5}s?q$FF^nu5ZJTlO#v5-Vd1
zY*mVBq?atnFe#Uke0Ea^f3K`-;l$<1RW`@s`Z-0r5CTRiqAhUIlTyIOK;hx$*;Bs%
zyT8T5_upq#HekfaV={?MM3RbxBpjV06|$%r7_`C{iOUba!Jqs~9l!J+{ohcllD8H$
zDtIZxjiIns(ll}rHD+5DH+t$?gdw|b&)L}tRvVJCJiFfV55M=6-~PddS3|{K2mYBq
z^>aLY|7-m9zxiMA>9eQU(Q}eKqf1;reahearN71FuYAba**QA248xA!{f+PP|M{!`
zCA;k{fAv58&+yUnyTA7xZZ|usrj{Z>P9B1iXdXX)B!zj3L?yhD!9}7^ER4ZaC9A6B
zs=J=i5v?=FE&~3a-r_Dsa4`@<q^fLA4Nwe&6m5$|lMAy%O45oQhn}*KaW#3Dm1T*w
zmfdd0e!rj5x4N!ncNsiB1W_&K-+6zq*tu_&bpMzDuBn9>g6E@;zR%UmSNtn~`7g6x
zui4z*@$A_%VvY)2E|;L-<m80qazWS0d-8uo*e<4nT35EGQosqnP8KhtwX~~aMkM1D
zE0~+Axlc6}Q^zRUv}rl^51bacn2mZTF#=kk!8zWI3E7m<JZ&E3`|AbV(TSbUv%1$N
zpB}4JE)0_d*b7=P%3PNuWkR@hdTtkKJ+*KrK?Z1~@DmS8$)0s7Q8x`Ir)P3QT7xbv
zIt$M_H&Qk=YjaB5cbuG^(L0B=mbPtKpRJ^cH*e7HyO2RnKghU=C_5}gv*e98-{$<{
zlIy-Dx={*bTM}Z#jgfYrsLI3`J!1%%m}r`YO-k(BJ-634y!rMAoUBe*E!H$;Lv2b-
z%IfFQixIq0Xk}RzYkU^em}F3)AdaQ@zGg01*LCbGM^zS7mErlPpKx}*=InIE<%4so
zn;on50x0N%<M*GwVz+5Cr(6&pqcvq&u~@EItk<lIlJ#=I>FJun>baF_R02G87%8J?
z85|+6H)$A%?Vix>@coXV-_v(n`hJV@Bd*<%l;OJVSR-~bDJHa<TD*}IlC%n^K4czl
zAt_?gnJ1PFiDO3cvPp3s>N9Wu*tI|eI+Ag0?5tdv#S>}%n`uM{r;y-j`&;%mSM;W$
zbqSw^D#4GUC{D6AOk}#K^X6Y2ZkL!lc9SM)_MFQbuDYsNEmxeJta<qGA@98NE@$Uw
zEY~ZV#e&u8NuHM^TU#uzD$24H;G`(9xkzl9hI+9;=k7>KNz^8Xo0ht+i7`+#i=S)~
z<MSF6PyKfijm{y{N#Z+VHtC$;RB@743@#9nn@LcaNIxcGFlc2d160vrm1bcRmGM-C
zX0@zvE^>X_@$C5>MYUsnw&2}&-T=H*`k5PMb6;93;R30kB+y!CM{fcyb4KRLPiY8F
zEK^9hagR!o(UuSsUDp#)tk2JAs*0e%jeaiXris{xfVN^XIR`wywssXoL5RXb0x=|1
zK@pWFB6+Jt3M^xUn^%1QH-3#LAH2_pCzqJ20VR)_RFzs8Bu*xxQ=~|MHbiUjv0zLk
z55M+x{^b9!z5jf+EKAQjvEMS=9edC9=_@nKJG2APO^{$?0Ahw1isTGuMie77GNF(r
z^o9I8^hpt#P;eBCksngx%s^CdAi@wpgAF#&-c?msR%T^Z`n>+wZI;uAwa>Y?s?k6d
zngl@Lz7;33GV|uW$Lzh<yPo$kzx22H^8fdrlJy2CW5`1wPg0Dv7}JpI1VX?XL#i}M
zw@;9sq{uM!oW>c1%rQ=J`{6#{zW0<TQ%NaudTVRE@v}e2uYcut*q_dkIwkiI8i#NJ
zGnw+w|C|2@PxcOY{*5;oZ47ay=AZmO|1n!z7x?S{;os!h(IMab_P4PvP&J0E)1j{F
z*ub@;wKPq*c<~ZxnjwT_ZEe8pL{N@r@s>*!qs<KxtqFTavC;M(-69D}DO89)O@C1<
zat%d`KD8((0_W_R_%GRxYoi@Onk21iHZE!>Cnu{c*=5RvQi{4+QVmYg>E+0FiP*s(
z%&l8i2ur3!q&*Zxjxh~u!#<ZUUu3?Rad>#ZWIExkx8CCA*Iwn#H{Ybw?V#GlF&^B0
zd@`jEr-rHow^AY!mLtnn;34CrK_XgH*kyk9^7cPf>Mj+~tC!#UMw&LIBQ~?qu1T)m
zvalLgv22Pt<ti>=L2;&0T=@{LKKC;X(CO57wLC2iV3%{3$XPZtO^q?l8Q(hExW;8u
zck00x`GD<fmC=A#N)<&EO5t4~%M1F$Av#H;*d_!jV|cQ^hjW2NRTC6Ee)^O&OOc5t
z%Mz-pWUxYd+h_ifODImW<Mq6nVn~I~a@MwY+1lA<c6dP3G$193N(tT|oks>mkb>!=
z1jh7TAQ`R`YD=|PaIm+>RUt{GB9)3vX>^dJG6fY=8R|Mdca%iyq#a2rRL~?OXrJJG
z9BF$SA~mV8*c43TdG_=eV;p(bWi;BM*Iy^iG}hGwVQGS5Tv?8e#vC7wDHjX!EGJ8I
zI$1%M<*W^dtPKXNkJc!%lq^k16HS(B(oCTf)d~<&k~~2vfea394boMJU@`Tas+>?a
zGs>eq>d6UdUa)xfghIq4PTZ?n<+P7ny0GlfX!m~KUhlp=__VaFv*X#*l!=zIwY)dl
z?kf5`w){F3lU5LCIIGJ(gHzr=Ew|7$4W^u7q`_G5rUqvTt)*9J8@p8#$?0fz3Hb|%
zk$o~E*=xf!Iz>Uz>995&vbnLz`uaNS8|z%Uc##`7Zm_<wL8sd#$x^Z+C(jGgEG13T
zc;Co4sS_#+rkpVlBHab4Rsg&}L%%=!%*`9uP!atTk<e*5(I2Niu|cttvW?{RB$Cq;
zWjM``4X9P5R#`S!m!M@#V5p3vHUbec60gX7<botxQYeSkHoiG|!fdwS;9x=B1d5)f
zG!>Oukf{Vg<j#5TVk}ag(-;#+79m6x64CNPEYSq#BFdE{K+J({ng(YKzA2+CbqJAE
zoFuI8>@paRK<Vh=<lM?5O0*lCPLo!#qd^F~5XjY46L~c8q`y?vwI}}|Aha2e!jlVF
zlakB54)-5F;Dhgdlg-^LY+k-b2#(2Y%*n|yCnqNqMUS1Gi*$N}xR?cx)S#1SJJUFY
zPdYsJ`cLxs!87iB@9S)B7t|(h%+44(SrJd>b%S%BJk9VH=Jf=XCAgqj)D>BhGk1Xx
z_9r}?SdN^amYy&E!Y^^>@F``HvU&3cJs??>Jbe2HREr6NL?EUUe&hf3PxzJp_djKO
zcay*IH~$7NU%kYazxGXj;g^1in=iaXr$6AqqX*0ua~6w=*`lOrYNSX}Ey~*Mb?NsD
z5U{<oO*t<KfC+)11f$IjOkKBVNgGimqT3EKE=J-*XTraD6%jK>k|cC{9nv(zyU=nz
z5}b2nSx&FtB~NpL4=s0&RYRBO`HFkh>2#3<7W-xNjZ|?=Tb5<yL7r(U{;;O*Y<Uk!
zBHQPZ#l^DI;qKkLq&ndXpZ@}vFJ0!Hci!R3l`B+r#lg`bX`1{Sd!L+F7#mYoqA!C(
zB1yE4$Q{duDVnn{5xOAa7#8nV8`l!|TXBmZiYKH%DzL6>5x+KmaMgIlbtAMY^!WXk
z&i5^NvEulAxPmx+eCW?w^3O6~oDlh`5dn!POw$Cas)}Q~s$3m}mm6%7BrA^4lEMPZ
z&le&MA}Xz;kV6x!r|9)q-`u9?^q9H@loiGt@a%)HAf-kXilm?Lm9Kt{x`|oFd684q
zWjhaxPdEezs8s{7nt3l@uZ*arK%n!S!R0I5c>V?U??2#hGC@j7mTH7Iq^Y9c@6uE?
zPxp>^Vh&kb>+s^6pCvfUVmjl&y}P`4a!fs+<Et8JEJ8T)EI})U@d{H`)YinNM<+;;
zpp&TLOi0inT1^Qj@J&n*@+RPn#~MK|TVrQ^msehRoxk>TzsPV{aR1R=_V%ArQFHV9
zOALk+^!>-|9lpb@yH7}y3?&r83nr6kq}8+uQ2kzqZl|Ez$r<#!Y^)C$414r@U55Q0
zqxB)(Jfo8*^zwvmo}o9k(bCiNCA~r+4j$vihg3)V+^K8KWQul<w7rhwo*%hw!JYm5
zqdP=c3MPWUg_sQ<#S}gYDu58=%8>l@0C>i@yEVU7N-YiEfHNY6mP_rHzKz}?yd_f-
z87$HXqyyz0%16(YWltbNL=J6>7{-iyxuP#5N|2`MiYx!=Pk)*#SFUjR@?~Cs{dHb>
z>1Bq)VNA0UB2EEY<L)Gn9Q;*I@bUn*TmxI>yVA+3SzX?b?NZV<dSb;757x4@r24q^
zi5fedBnVF%A4VJ}`q;26C+hLpE#>r3H5OwW5*JT{Cg6fb35E8G)ce?MX-%daNhZh&
z)$-;X<-#CkLec3#2=Qi59I40%4$>rc#Vhi!phXQDjIju3(74uRK%isf1;Iz|NV!;0
zFG_sNBN9?ESR1i*;R1urO?;v$Z5-c192<K9VW|w$Ex}a9u3H4bX`owNe0=+%YZWa;
zLkQpmE_keKW$G4E1Op}5$}|V}-s7vk^IMEAT&65%oSYnSJU(GjmS~l;d*L!Kz5EJS
zZr(trF@8-ckCqXAabAP&^AkV$bNtG;zsteNn6)B}V<fGSQc%`4rjE#+wHAC-Na-}i
z7=@4;gkn(}_QxeB23#&#-`!y6#t!4A;^m8<Vtp{AFBJChga;4y;ABQOk(Bc}EF6i}
z9N(SufBP^0kGND*XBoF{{eYhsZZaH=_=%tR6kqwhue98_Kw}J95|zv@L@v_hE0<ey
zX-%4Ix`TpdQsTTP%`%$#jF$EkZ~74V#z_dU^xFCh*w-!B67K^(*i}SSj1<ervtq?%
zS;ma@`aM#eG8ha_8#Q2mf1eOSjOa^K#wW*w5TX&cvsk+vD}oTG@l_w?n*FHf380mt
zu4|^#8TpGT)7g~Y{Fncd;5@yePp8|ZC<=zdA;0z;zfRHVP*pX~I;4z__CHP;-^Y{2
zy+lO>ZF#DPsP=ON1P&SE2wMPB>c}^9v3pt~87m6|!H1gL<)*bfl?wzhX&{;sE26Ce
z&PO7-j_c0Zq>9*_EyuWS)zqAAs#o?5AG-L%O1H86nh@e4cd?kUDCg8oiSuSf{!Y{6
zLx`PH3W1Eu-Vo85<z_CWLMJWFMF{euU~O}qOV3^9?K|&sG@0Pc1m=p-XoEb-FqPxr
z?mok{l-{7vU~L$mQ?($?@|G(Y55p17IJ?P-c7CU@u11H}dn`@))Tckg-S2;o<EIZ+
zUtM$xyf@6s8J~LN4d#bq_8uMa?%Vr3+WQ@PyIXACxWTYnu-@;nyFTLL<~lnYLuPw3
zW|cvDg{>VTv^=p0<>D*fMnc8$U<kgo2$3j~7~d5H^}J$#Z_LilI-h;>bA0;KZ*XyU
zoA>YD;jQn!!{PA(QU=oG4p*OhffsKMQKG}avtteq#x%ASiloB`RD^(S94E(y+1R4w
z6eaf~Qs^5(un0no&`cElUYCv08l7IwmFpMDJDMa7Jb(QH&p&^OV!K0@bt&$Az{&mz
zrfHB`(kC8iXiejnVm04_9RcuenG@ZPW^Gi+8rP!X?F}E(<6)`#UN%;z3h?&RZ3%Eo
zA+1|JLTJCYA$S`V!OkL`Ax#wBG$Et~QxoVPkI5%9s2jY8BFpF{y0ym==v33`blBaw
zz~_GQGral6Pw>)9FY)3_FVG#XBU>)Kk{YQ5xFr?KQ4}3ovj@C&Z5$y>QZ7PhR%lO~
zED>CYK#>SB8g3Pt7N>D790|d<mSN%(@rgcBV^>_*rP_N&5+Vsv1c4EPQYtE;(9WVm
zj21CK6C|Y*n2<tKVU47&E!HO>6kd2T<;i8BlL&GgSt3b08MTMQaYI>3vLc|=0))r;
zJ~n8QG-<o~m_;k4qR0xoxA?|j8jCMW8dJp^L?@BU7esW<FDq<guysRlF>AFm=+oa^
zXSlsZHt5j^f%PsDHLXRq$Rr{fbP64iDiDNgNyGvf5`0jgB?gBjo?x{1Bpxy=@Yb>L
z6;(ZBS}$1GhQ>H*8>ocDBssJ3Q@(Ti4z3td^ai9w7gP!;lx!Fuy+bvfv6xS}`O@>~
zPQO*!3bLRv-r*C^+Qmy;di71d`#b-fpMLo{a_`X6<0^whP;`f+Ne01_1d9%u(l#j3
zkja!(cuZX}ZyII@oCw_9-C$#To2zSUtZ#0SC`A?o2lpRPtAx@8RFLFKqg@Q}w$>38
z%&UrpF1dU6E^q$K7m-e~zBc00_707yakfD!fnU0DCkY7H-QJ0db?|h$1-(wr<M|_`
zR7|Hc7Iob&3T;zIR1zB>P`Ea7=966Q|L)gn?6zDJWE3;aW(%ATbi1+P%d$L<GZiZ`
zlTK5TEW<cQZA>K6hnRP4gGX4)$!x}=s$1H^a`brCczF3N{^(lD{0D<U^u6?+Ua!xr
z&4zva_$kldyvYmCzsQ$<`(N?k;Ukn*Y;A92Y7^-R0a~2)k6mTcClLu!1SD}YR3b#f
zYaGJIZKd@ASQ*8xBo=McA(t=xjA+ESLWt#`=_H9fBp(PWT3F!2s=14YKvttKA(!1)
z`+<hAGK*c+a-6i>2$$!Xvr`1(j(gUIDTF{{YU(DcE0;o=rI*-g{Mp%j*f!;$&C-=Y
z;zDS7ylLc95mCxC&A4{`I!`A@h?5C*X-HH-cQ8O^0+9yR*EZPP>M<OR=yW?sC0GvN
z0-+U>tOjz>+B5m~B9~Gkyo$ns5J188=bmGGdz%MEC;IYvOE-@u^>yPpIXtAJQm$Ou
zW^-+mr-uhTnJ#$$-S?RZ#g&)dU~gh6rzfORpj(lhkO5~sT1z}3=8ro|ns(dKrveoZ
z_0C5wrEAC9&c#DWx6>s_5-wf5NWa(TgS+?n(r^7Xi#qb`c6Yb<{ongNCdV_jwk~jG
zdz+tn<I{Zidv9^De@L}xP$D6?Xd{()33h3D(>C(~@Jk_3K-vT$HPaKr$<ZE8I3Dev
zAeF;Yap(FD!@*zY!g?31q_KkWq$IZuS)S5Kl87G3sB-qA%}H+Q1I{n6+2yFTO{?;u
zov;z7TA~#hwPV&8ntH|*u|=zve}#w`oo_jN(Sakx{kfgS1#gk<flCFCmVrzPiY%r?
zb@QBF6>v(^>Gc_`t#R?nWnO>vbuM4I%=XR>>svc)Y>pVMjp+CKbbAF^(c$zQApnnW
zfuu<EBHo_@q)za-h+4M%hv-7LT!dSmYwNujL_mZyJi=wxg<noWeY9F5oTBX?qdrk%
zxAV#iJT)`I@nO6`y~GAd69i4W6seZrZV{Lus0<jVz)Nf+utpL>%y^X&QVppJbd{%2
zfld~U5*MYTsuhr+RfVL&S<hloVqJnvQ~G%UPSBVJAtYMGQE{p@m35?2A(IpvY*g?`
z#iCiTn9iuCbCeJyQj?__#b}-3_9nfpbwoF#A(CpB4*U2(kZr>)1)+7rj_S1~+!_(D
zNTQDMYTh|gtr2*f_0)}JUROMwPPo6f&$IEA<I*r~3{_*$T2N$~!TN}2Cg6J`)_1Qk
zOuB4s@3OwWMX%FkdUDKsGU5L1+hmF4%C+n0G+z-)l~&OpT5DeT^k?|JU-|!V;c&`}
z>lxbDWI83!yU4T~KMMiUEeMUkdx=sR_<;8f^Vx*5Zm2?Bq&If9xq0ISWRfym8{v&Z
zd5_=RW^H?ir|;gzc!>=vD&;~PQ8(7%Rg4RJ_nlkJ$_06na{q%5xOwdwyBBvOlOs5g
zQMoBpTmUxKH>elUC(t<)Pv>n!+ITXhtgDv%8YNE=BaSD@xFLL!EB((`EkP_1Ak8v#
zl14s|P*~^aXhoW(QO&Ic)>#gZk3Ph!Q%X@a4P{+1olaMY1aZR(D@Epq&4}S+chl`&
zon;+n(=kFwhQl?ibx|lmpl%u(YvKl7*Zc=R_a87Eu5qw;#Noja@!0(1HFk?SM~4I@
zTP55w|Jr$^2yJX)M7)-cTG0SF+9sWPyq%&-A2J>d5rIPpbfRNZ=Y8ZW2@#P#&&opL
zbnF$=0A!1jt;UdHiEu2rPiJ2LnNiQ`b7?o1_~9&#j3e?FryLE&niW@XWo;rp`nt||
z*|E&Yjej166>3G04<Vp*!lg@>+1-DFN^{D2jc#&^P8Y8%su1j4-e9!eAxTq`L`O8T
z9l8mkQ7*0d?&)h<_0OD|+eXw>IM3S74jbFs42J{i@ga?aB8iQ+HI8yIV=&xc*z0j|
zD@SP0+@7#s);xIsHoyP7|B6O-IXJ1P<`ca2h%AZkiHg&J;A7^p@s>1$pansLm+`u>
z-p1DwmK@1=tt4rJ51y>Z@!s?3(F5MUbBC?nb%ujJySqE=?>z-8sGqBO?X@?!dF>ic
zA3UZU&!{G4H0G}@LP(TQC{)bK4)OXFLA78}dk8@zLqcObv(m6=YHELi3l)u>vMA^L
z;?I?AbUJ7Yb>o@NN~~)bNJ-F1d;PZR_m+O-mL7;L__3nCg)>uw_JLSZ7*4O#<@Y=5
z%eeZh72@rQwr5NGuy96z<W}B(-Wi-wco9G=vNU0BFksgSilIk!`s`l1%!NyrdG5w_
zKK<#>a`oC(Hn+Fv_J?tyY!6e`nfA4uUURGFGs2B6Qr#Z7ltM(>YI|XZGcWY?>*dNq
zWqB`$<?YP!;1=4>Hx4|)YHIj#>Jv3~+t;tc{Dh-h4U5?fP3(DS<GN&fI*cd4*iko)
zW4>tU%0N{$1Sio;BSI0!9a$ntm8VE0olJxCRAqy+AXqA}Jk~axoE#CDVcdxBg*I`*
zcxqGAG%+n=IvF#aj_Gx~q)CeQlF59`tZrzk8dKNgNlK>^J?DDEA^puw^5Fp0Eog*a
zNw0%%-M*!OG{h0C)F3i~Nbqg+ifChmJPlSgNCyduzDj9Kpmc(xs^R^oPk8UaKDQqo
zvv*RmFp?%H8W%5CEi9Xb=VGu-q$A55{oa5teBm$g(|`Gkbo*;aA($TS^XP-S+<$P7
z`}gmY6&)^Iyn<3GPKK6H97i`-uHRt$>J7es`&}-6>N&buiqdhk;)BDr3y*U#-;m&A
z9&7N_HgGteQ5mzcb=cm%K!32t<7fMnwL=O?D&xX_^QD)0=d0gfY!xENsE8&?-J5o!
zPZGt0JMZJmipg}wTW@`z&;Qit=nr~~)<;o!*k&sF5Jwk-exKu}Y)`<R;o2~c-kfJV
z9y4>z3gI|gtL^JA7+SZiCAt!`Nn^%lmStpFzVfF#K0d)YPgT_%A0KmgaEww3o12>y
zMZsdR;PK<f%w{v1reQLfaD05ciU>Q)q5I>SGA}Jtl#)y)Q`Xnl>GgW6@h}07j-rp9
zb&mJnf1h9e<zMDM|Ihzt{N2C%clp7)KcFn@AN&0}9&22#NtGbWQW9m67`!b>RUl0w
zr9!JHB=aoIQ5`-+{*^?W9`596gjyU)$EHFl6{8PZj$N=Gl)_mX$2=|c5jT~n0FRM`
zXWj1kNW&%Y<|7CmXPdfHI)V^_WgbN|C3exgx@p=>Soa~M?!&0vS?{~$4*`Wjg=A$U
zD@1Gg;ysyGT)B3Y7pn=I$7AM`lKR9^mNi~kI%_$vzVRAq8kkI{)OAe6X{<q|3DKqn
zWb3iGf=E}rYTSQ>R$yIJ7McduH`u*)l}i^d^6s<87$q=?z)8VkVUZ$673QfzOUd^7
zI>m6rv+06wKX}Z4_fP&Y`DmA*x*U&d5_;%$MV564(%_Y%p`rGcMtZce1Z8m&l=K+q
zaQN6{EGIo}lj5AkSBB|g#@><S&fPnB;o02W;KGGnwzs#~*j(q<58h#Xe8OmL$aB}8
zXMHr}@ZgZ~@t7n{sOA;k1*}kX^6qL6_TEKr#}++`A|8iz5+O6X#Q-m=nBw6Bjj1?1
zTyQe3Sd2WmSIo<X@nnVzj!LKayg)69k3tY6I7GDLYMW^nsi{#s5m+_&zUAD>a7s^z
z&rG)HZ}gp&;#53wHga!yoi6DIXHB$S`)I0~s+?nFLrp@HIy&7=p1XRLOE(HS>$|L9
zyu#<d@H4#p>T9SXZ)LJ=wGI|YiO>SA70xuP<QAm_QY}-OA>whZh@-u^WSi#l2bPEm
z>Qi!r_@Om(slHW8LC`c+O~8L5ao8tn?9%li7hZZ1UzhCNx{YgmR6qGZ8eAOZNC?5L
zE-qsOC*zW&m*9LHF(mDYQOY<LOr<50mOKe)>F_>aYKsUNQmA+mkq*)3T`tNANW=EV
z7D<t!k_2ZBrm69z<@orRZjm!s8=!@xoX$yPg2j^OQF_0&Hl*8b9R1;tkmryptPCyb
zK7MqfRn{sYu!)Uky`+ecK}cJ^fyYZr2o@(CLdVw`jHEV#z43x~9zW%~4<7RFvjZL;
zH`K1ZC}cs9Sv>K(8X4x~L2|fg$amJ*+}dV3o$>hbW1Nr3$-UdRxp(^(JDcmg`s%B^
zed`?_KYq+$xIw2EWz)_^iwXs(PM1qBzsA4#>Q^YafmfQcZqN>NO1#LtqiO1v7vaz%
zieRQ?#r}9s)x;anIZwaWM+!}++ehexG)+hl<ozL=yBDc~;-rCG#N0xSM+wkMF!ur7
zc$$+bi-UcB?YDlLZnw+zn>RQfAERU(sri<BNsB_HiAD;+#`*>a$H(;h1Efw6K`@&y
zsM}b(_SRVa8U^=%aT~i2H(xNFPN|!UG|eblF+^FF+`W4*a@Cy2S%(mkZnwM2Q?BcJ
z6*0E*S`(s;#rZH&_s6aH{o)C{(}@-V!N-QPF_zRsOCwr@Y(AeOgy6HE{Ve~B|M~xt
zuYT=oeC2EZ`p0?>_a2pO^d$Xm#%P$($pK$cEh^GPkhwsjAxl8(7~3Qw#1?r@tzA#>
zJ{l6A8O1uHkuktBVkpQs)Qw}i)*7WfiLCx>QIB|V`Qi8Y6p8!rd%65|Wy3;5wW4Ew
zYg^H7md04>x}s^SRocWyo4C{D40YzQe0_1#&SXk(0oxcloesy7S^T+Dk|Y_|p1Z;J
zZjGruvPrmq{~^U7V`F!Xo3A~`WV}ahEoC#qILmA{LupNsv?HqMzYx)=C6KV9ADwAn
zmz%YcJpaOrln?Lo&ey+6mgh80gA|TI7lJn|W>b`L^s*kQkc@gAvcZ64bCbXQ`e(WG
z_=vlY_Ob6j;r6Wu93G#bR7zzWo#6o4$xuZ`I_z*Tn~=!{Ei$x7aMBSC-vml3w2YbB
z))|_*W|$W&d`V9g^tXrn?2RvS;ldWZUV)K@|LpJm9p1Zjm#S<a4fw>Oy9$v*6)Li<
zOK>uRfI?Ht)~q*pTzg&0q`eL$D}ke{J+5w$IwzRuQJD2Il$!JYoky%?88>zYR13>u
z-k@>Vj*C_$M6R51(Sg9=vG_Q-GLB%uyCwf#V%h{M=K{2!9z-w%MLSt(5h@X<UP`G|
zrn`$5Im@&c@0PYHt^Z?dK^C0HHkMRIUgse1k#1aI{ranPHg@T5UgW~fS4np+qLq%G
zut5@46s=rWia16*Qz}VOr&~&E+tLbQ&fIGf(H=ZLd=H-e{Sxep&F&8$hQ$i?KfR@s
zESj;WS@T2p{*S6pW_QxbS-W-x*VN2sC6h<bkS>}>$~aVRIUVto68~Q_pBmPS93hi<
zstQpFs$?Kl9!*W59eEN^BHAqkPwZ@^jwDN|NRrS-f|v-mT1&vAQiV>Ur_$m6gdjZK
zK|d1hQH<vcv{Y<vZBleP6uo$=ky<0Q!ljy!MvtW+MO^Ss<u~o(rebd%6iOt>puh)|
z40tR;)OgWEuR5DiTbP!fciw-<H{N-lTTl0SJhxOz;DjPcG*ZICO>uIHN)@Jx*{-1n
zGvAO^6-{l}-+RPp=P`9-n9b&V|NGzP(Fb>#Pj-0ul@}Ncha8WmoSdAH6<wr=zJ1<>
z);V5q?UmO!EE^7IC0l8iVWLTOic%tuxatamL~DsP(O%Md!Tw~y!MLRHk&rBvq{zE~
z;L@clpp%G9gs49L)KC2+Nw?2@c0?ir8J0py(k!L7-p6W%NF^c5_|1Rz%iMeVlwbN=
zf0N#D!2Zz@5=AOB*12|^Y*Fp~=B;Ho=uy`Tj!#YqLSSvcEjKFPQWa$6bS$g&qx=h7
zr<NMAd573oL^Y{sJ?Bi!d2IjAwVUcv_;9wFTcS_@rW(3sDWzzdnk3OH7xt6ms04@j
z^V2kKN0>05&-u;Y{7wGq&wi1=|M&ks@4tVSU-;`k|6@Iehle%esi(*m^bZAj66kgm
zqrm`YYkV-#I>|XQ4T%ON6-ow_b~tYcA$o-=qR=32Ndn<pqt!qgL*!W+q=eR7H4qiB
zh%*sRg4LL6C7uYZ#=xP?G?Hh196nO3VbAJ+!YYsQ6#1|#ixYseP3hS&toX=t5iO!3
zLo|={J~|+f7pPf<R1PG}t2%B{N|Wfw#g<AkSX*OrXD5#6yeCT%2E9J(qY*`3wCdZI
zuNRghv3B&je8EA01m~kgjT8cO<lAj*ZLo9c0$xJe=~9<7%E}OcjiD!GpoOB7<pd#_
zg@88(;RDyNUE{)ySGcw}Ch2^KM~}u31jg8S;J{<&kyDg*3i9q6sx`J2GUi2x2$|u6
z#CaJT8ay%yEUGz+W=`XJs6=9X6Vny&RCPmEWL&*|h5fxl9zS}@#^x4Do{;7#0gv$(
z>n*{@ga%DidpaLI=7InnmgCLVGRw8RP6UZgs9l4S0i9|<@aW-Vu3y|FxpskWr`L{H
z9WI<Uj4t?7^|z0Cup&sDlX%zKo_G<h&|2TR$TA4Xb~YPC<c$e0Ptz+}5ezbh0tA&{
zrEW*7kyB};ppuSKq|8Z}i=0YIT%NJBbD8UJ`~;oR7TI7#XJdzK=Ne>PMAilAX=Y=Z
zWm*;jQ#Ys4c``;<E;HbLh!K}M(W_J>DL^RE9xqNutxEvm>?rw%Zo=i)wfDUcV&$m^
ztoXEN#ZDhW#y@6#vgC<GBh!rD<;(0$7g!gV+<$@!HY$-+LKDFHCO$1Cks-qv%c692
zlAI<iNK`~{QsqgtLrY7dENP;UGMtiyi668C@E#=<d8TPh{2F7MmHMn&ESOCubn~38
z(Hg1N1ZO!KpD><GxpMU?I?He(P+K4Qa?aAtGmMIhf|TGwG#`$}i*d9Y^<R;gt3-<Q
zf?y3S8(L*S7>uvcu1~edc=yRZU%UMQ?>sx=-r<x<t;u>LE^h3w-szKCPZP#Go!n#V
zYM;;jrBCsLTX)f-;OPfP>_0u`WbcHd*)3jr<I@cKL%#9*U+40*s~jI6Gn>yi8E4#j
z?>&ls7i$gU@d>*ZFUJ$OHFz5{0$q^wHnt({aBwmutoI2an$OmCjdL-kOX?IsAk`96
zH<WeYc-Am!qE{gZL=f1jVRka1HyjZJG}aMXP<nlHgUi>i^4`;ZQVFkLUgza&7wPr~
z(CIKY4OO$??xQEX^Y#z;pZ;(DS4@Rx-c)q^eVlQ4V^CRwP!e49C-i=)yN<RS_a8in
z=>tv@lxWHBODX;Gra0v-`~?N<K2)c+1konwM?w3`25(8AKl|_hu;2f2YuOy8X-b~w
zY;0_>zP^qSg4t}wa5!LpKYEcBMM3a^G>tjR4<0<=_U+sBdOebV1GXnqt8hY7p%_;c
zsdDr?35`!lgk#vzbc+m?2*#5G`rU*q6-YczGzekaEKG-%twXkpqb1d%Hi3X?YJx~2
z+O)K}SDbDLA$DNP&EoWmJ{3JI-#a)}$1eYDsVw{<R!&%62l1P_HfAmw5?huTvx2kD
zoLF^*OV#ai^IB%#Mulu7)klh3XeXv^tfTPM&JeUfYE2SACXi-1DoF_{ZsuBP@+3ug
zO(*ZPyw6ikRulxpM^yZ_2Rae`L2cXcdI>@(Xex?9pTW*Hy^VFO6c_|eYr#=Aj^G7a
zCS*y95R$TPz&b<_WNE_sg-huE2Dd)gqtor-ox(c-BB68^Q$?CeHb}U3<r0(G3~OzB
zxUP{%lvX5(ib+2$Y9I($Yp9#ZVN*)bG-Wgi*I;bL+F*?&Nif!MaCpGY>ozJ|(|DjZ
zrfClYD!NnH=-Da|D3wq(HBu)iA>%mIh3L`f90Et0DNG%f7Y5#Y#>Xd=iv>z+dc7V}
zEmf|5)jWG2`E-r(1Rvjz;3PP;YUEsq`SCW;#yz%XUvZoV#C=$JfoP|BQFWRiR6<Z1
zr!+<;I2D^a7X+2Jlp;_nsPlkOT~gv0zB3#W)^@SoA>=)ROsN`!vvb^{!kkQSLbtwr
z@j2Nh8mxO4lPuCSU8zuA2xMs*#bjwhmgRArz7!Ja=q{kOj^PGL8ikrG6Yq~Uc*}!G
zd^hBhTgd4?wG=b0)>nCIEA(;elQniiVw57=*yO@18l@6$%`57s`{;I#ze4PS$A>8K
z_Eu0=khSA%84MDSlAct1vNXi8hV)qP;`mGiq-^a5S^^~@poF56cc8T*@WwEk%{biK
zV?LX)IT}R{nGj5;6Algz7!F6|-5x^5Ao^KdA+&14k7XN^a%wLSY*b_}7h)}=Fhqn5
zg#vE~#0*8D1x|QM>!`ir;OUHaZ$II!C;Qx*OnKA<F1+$Ne*V)xNg5JVWyzYr@q<0a
z<zt*|I5~XGx4-ocD(x_xS2V6*eRG2ck8dHfz=aE!xOC|X&z?PFQI$C7xOVLt^?b?)
zAKb$dD7t;tMq6<~5jrl~E;dQd`AB+AGG?<X!ob9hTS!Tk<#Cbl5)lk|ixkk<z~Q{+
zq;z<xNrXU37#|++&iB8^+U6!9Nsx((9(ksrDd&9ZGoR*zZ+(x8n<;O;c!8Z>6tFxy
z+2?pVV{iX}hfinx;@|#{_=%Ta;{W&cuQ0i1xN+?UMOKhV9h*gK;>{zXwT`t0C1o_O
zKB|x&GAUyc$Z`|pR4i~_|8CT2wrO0@dc7Vmyzm0fI_})LL({a^L7TNGrJ!kQdc8iw
z!I1s^eWueHRrTX@>{#1^v_n!996Wo1HkRX~h6j5`q@v{ctJ_??u*Rq}<o#O@xprv-
zMG|9_LQOHypdGHRND_@}T9aRukZVnvrY!1`lgXIQT0h1lDS^_F)1zEeQz9JL5aY)}
zXt`33;6Tc-vPy|)*_n|}i)2YDQC?!(aqC&Na%c^a0jx39RYg@*SYzUWx<wt7kkKQJ
zI35;l^x+aojFD<B5nM`wOh9V1(j-E{WEPzSguqFOOj8<PBLF2ed6BcWzDel}&yJ5+
zzpzHX*T>8%L{p<e<b?_;SE4n7BRCrk=F<%6X%&eAmFUP*Z23-FN-zex9g0hrxcTOr
zy#0-@BV<4p3EnhJ77cbF(K00!(O_PM2I(D2X%3DKNVcya(hQNLboy(gszai3WKcZa
z-(ynE=#2)v@#dTS)nE7`|MItflShvqQzU}L(VU{&qt`7M4hKAW@;Hibyu&mNiI&)U
zfwvV|swm5bvYZoqjq`@?g-g7D`yLNJc)+tK&q!24k>+G={8Up{q`Ct^V;$6WLt`T4
zsMjAbn?5B|Ig+?1)J<#B?<^{KI$ebo5*K1RgE2Kt6_wG#TZU_WR2q>~CoE`iwZjI7
zbAkD^2B-1HV`@+EDZxjpGwVIpNAxg21W%TSh7i$LVFSWO?^h)hNs^MJ1v1U>I;EB=
zrOdE8!>JS-1Y=V&M^fnwHnv#X+@L#LL#7$e4i5S5amnoQL&~PXgg{bsP-%+PI`%L=
zP+N!f0imPs;*$Qe%2r;X>8vudPZ8?)XK9+z9rW1T*kaHh((e!07;Vw-#ng$sC`ht+
z09{RTm*eJW02rxUXArq~po-T=OtS$I`BHqyobr#TPZlX=93~SK!!`Q7Aqv64qbEE*
zImR2iIt}{Z@WJAIbRqXaQrCt?wQ_aeAf-d;fYy*C0iDECtp|__A)0s-2TSm-jWdx5
zsY%fU>5u^y(-|j+hs>u_iZo@d-y>Mdv|P~CHJzeEe{Bup+cSgG<eh?~$Y`vMF=*bA
zI29Kw>%m3zN9neyYo*PuiDOrM4b&i^HUX<T>`w(hc>gi)-aF>WiR0;_;z<>F{xd(t
zfAJ6g0YCGzKf~dJXZ-H3{u+0_{|=8A$J~4Lh{N)Tle)y-zmKsn=;Uahvp!g-*B|ig
z=#V>i-bZFVuHLxGr#|&*oT>TN*MFbm<71xfKjHS>_j&P^SGjua2G&{<z0AdpHZ8_j
zLWo{lb#0JAk|!xis&K*6_&Am@&ZC0GTZ6ZOMP-?kmW6{L6at3|g1tu%x%S*Oj`#Nv
zMMmQsv)P2;4Z=CDT))cI7oNwR+{3#PQ#BmV4G#}y?2Su~juyP~;x0EY?sBrXN4MMM
zOW*z$qu~ZUnS*f@qdwL<q=*A6?=17#j6^5#29wL!mniiQABvJXg~wNS%Xzc+XSU9c
zcjvP?508&2%ZmTxKl<BTy?T|KH*ZknIq%=QAH5FSb>`y5i;PAi_V)Hz2HyX;)f5RX
zL@Ua4V<+ZvT0<HfLezZcy(heN>mfU91%K-oe*qn87Ly~6_8)Wg>O~HY9*`G`t*wh3
zJ>8?1=P{ka7?Lz$@99%imauzamxZYbN)lwWOi)CGAR3yBHXTd02vk%bo*j<@D@CsV
zh!?_&_ag*^HlTd<kCv3cAC7qwoHmL89~Dh}^yh!*XT~3R#%oWYh-h7q%W;)L28j@H
zTv>E_$iyJh$d&DSNTecFk=t386{*ZfbfgU|8`I^7Btx`@SdwU^(yb7&B{St2h?I0~
z#P)O7dFSo72qqo|y>-YogB%gpOq|Cz4c?eI?w`%b^9<D=FxuKCO>>&kV?#qC64v|c
zT)K3H%h#{+Ghh5mJoo%Hp1*O8gTn*fdFKaw{p(+6G}_?#n=kPB&;K<4@;85-<KyW6
z*zb4Q+1X-idxK71Fd7ZHdgUU<HspEA#>NN`*v6riW@l@gqA2L~x(o(=`u!e@xxq9I
z##ETbv$?TDV=T^k#^YnEa)EB&%cZ4|_nt&)I$4VK6*>VC6y8TJ?szg`dNQG!En15s
zi?=nEEpfs|HE?S?6MVp#fNeaciIjj4mgt<0<5m?`M%;0%>nXB?ZeE~-M`}+<G)4*L
z#60HOd$f;fEO#G2V`>7Ghst<LAMjE~`>)i(!h=30D2+^VoP?kYNIEDbutLxThi}T2
zvR<b-LdSHSx~g$L#-(bhVw~dApD{!V%hH%T&oZQhriS@!%;CWk9^Sj#+P7#3nxg1L
z%2%G#>-AV$TVu4o9#bneH&@e$rElLU9N*5i#hExY;^9lRf{l+`pUj|GNb<Z(-XD?z
z{U=S5egFV3EJ;K`RI=jI3pcSRhnzfogsa-9t*|tjfM{!?S~X*IbxsmNktTG~K%RJ%
zG^pr+f^ShWr9~Xenn;ec))HKdl@cNPBBd&2Hkne*7igg0=}@F8&RV7?V<6D&k5H+m
zaTX;tcu%L>!8#lBUDK4=e2xIL(ufeyw4;-jN8)M8k}^uhkq+-I4p{hv`%h<l{d-S%
zu-|Zi=G~)x^3H(IeE#$N{4f3zoB5E*<9&=<@Xo#O^DqAOKj-1z9+P>A2_A<g1VJLe
zEgJfXVX~;{b~Cah;XB{{E;8%$nV<SRH(z)WrQrMD`6eISzsEPe`E{It;rd4Fj}#&I
zXhd7#ni^|8<zj&`21OqEJvMsrgdiesR799wiU2-xh$gc+C(}8#1B0h=hTWYVu3x^4
z44zx>yv5Oc#;mEL@WM9CremHxI^_A+-r(Ms?o!q@PsURYXO`ny@o;~^OE2v3>Km_-
zAUHWZpx5v6>g%uZFMjP;`49if&-3DQHz*ei(0SxBikSRSmJ7OT1C&y$)@_-FlE^5A
zYX`=se4_LEGhb&%v00uILd=65PsUuia)l((%x7~RJ$^#DC?ogM#I<H)V`G(wVLn0R
z&X20yE9>}1u(`g)@zcW??bJA=O6e7A0QhRo!a9EID{r&AKBSi?q{S7cO~8wqvYGRA
zucn`G(wLGUJVl~8JUS#zH7ZRob%W9x&(e4|T6f6c5TTt0sMcyEzDpl6C;m|)inF{&
zDdQv}h_j5J_TRVMI`904kc#E!@uMDRKeG@0+_3!aOM<wF8<bq~h-4f;3Bii@BHK7P
zo#3=%z9{JwIwm;iXlGG34KmkgbnChhSQ$XKhR#cl-wJJMN43!=C6e*eZikDPFQF1e
z)5J9{1dmp#4K`>^o?BB`B|u3*wU}e-8leNdPR#Z6)(~uK?;>J~fNGP6dc7{z8`4a(
zzP`o_FWe;0bGn_Fp7Z(7|77dN5RLVVB4dAlkFr`YnM}ES`4V+ys7l9t+Ay9>=yZBq
zy?Tx9t(qiJ?Wisq(aZMzjhU>1>HLH&$qB(x)g^hZNi#{3fDj(#BA+nRF>-9Sm=Ge@
zi4X$T!gN|OnN}<+N1At0Ns5;;dZQ9YxS@4Z2(AS}JZL1TilbN;+GL>CD#-cx@F9wg
ztZ`JKAx#C{G>tY`O3*mN@%RWg^;qGkkl1{UT4<Vftm}1-OcGR*A+j7JMMMXk!8W!H
z43Mp9uy3>7P4u|6k;|%G1bG_kR+dg7fD-~^s+C|V1W%HrI2WS8sA^bx)-F~2X(AC?
z5rU(xV-#l73cbdYDSLbSF()=l=@cFEB4>SlgMPnHx7%TDG@>Yq=-{E$X>x@EMC^sy
z+v#K1Cu;1JN=VWSsgigTr73F{E?{4Ig~nRO_n#4*i}0clE7MS^;`|)1k+SuOk~R{A
zrAMha)(tYo+z1Sr_EgGXr6Pm`(?D%Jxwaq$S=vPykDo12rXi8Ui^_qq5Ts@>=%YG0
zO-l6cO0szIFG}*FXiw6Pe$k<7YMi%dp;0o4r(6%#2b}U)=}3Hyuu%!LXaeI3esF(}
zuYB(z6F=mr(mYxmGrD}8*I#~(mu|d3;w?u{9`b+xzx+eqz4rsY^UilTsi%|#tOUG-
zppYm;h}o}I0|bW<pkzT~8t&b>#rMATHMTbUy!Y<A{Q9r{D(}AYcJ%O5P|Zt<EJMj?
zW8j>jDi_$M!a2`$d`w*~P%?+s6xBFGSdLL#gYxl|V6>vPfpP7bRiMZyLK9K4<3qmj
zYrn?xpZP2&Uh?~IzeVjVjji$Cpd{p3hwhsl)-PV?_TEDVTCoU{heuN`TwdpMufNKA
zlCu~e<2DA!q{nN|zrvdC^QGVYS3G!lpP%_Fe;HM4vLuZgy{(y5b4KYFoiuObVjxeW
zNopk4#>2!Zb8N-B`y`k7dHo63GN0C(Xg|^K_n6OT?C<ZBB%0xHKv|c({`%`&zI=tj
zpvTv~_6=sU8Q=ZhcNvXFQ2=6o{KoG3^H=ajadGD&-<r)>*Xso95XK{=KzD~Yl@Ofa
z$<d73NJfJmTkAt)?a0%0f>V^`5&5V<c!v;%##)TE<lP)01a(!Bb_a+M)ul_tp9sri
ziC=kPi9g_!a<;JxXC&WC0N_InnG`>)3UAf0u0=XO8vFLqhVDlpc4zs1%O?<(w;vW=
zw7+5Lfk(^FOVSLl6=*|a8bU8e>J+7-1&0Zi%(>I1lO;!R$y00XLpbFo(JB^Q@Nt6^
zf;7w7-q}GXDRr3Ror{=K+x)^(faO~5Q&jOcRYg-(B)KKebFw0*saj+;NSqJU#xS4H
zIXpbVm>L&s%orW*u)Vz*v%;NYHk)zz^2K%|&(TU_on<<kGM-Erk59OC@d}`*%Z8_m
zL!RwDW7zL791K|>X|g<{)5#f)hFrX~%Va#Kt}S(CD9f0t(CuZ^Rl}mPE$`Gs5gh=I
zBuObcDg92tlf7wdf&}fzU$7_*^M#?T9YxV6Ejl!%jz;~#q1vX-IkWOzR7;ep<zads
zJqNvyQ!z)ti-<f!6buB}Zhyj42Zu5NQJTnk3zo)PssP0(#p?{;a_^+ppmh|-h!}=o
zLmd5<wr%XBAc!dXFinkjQP3m+sTIOH5EAD;j9Q=C5-qRM_Q17Nh$Ew}LoPXVOQ0<_
zdx1!Ss_PnK3?VF5hpeS_Q<|n69v;!__2_oHY;JDS>-FOGm1V1kP7-vQQgk|JuKABw
zpUmzg%T_scJ_LeFN!QltU%kq3F=ujgj6JSFXq1mFrVOncv<L{NTUG1=FbLU-6J&x2
zIRM^IP#6;7P*`k`;4_+9u~;lfiW;57Ts*g^2#W>6)})<+BG2(clN1HRPM6`Rht&?N
z1!-EKL`raw*_M1OC90L_YOQFh7#&umI%bWEc7ed-F=)HMHG)~Gc(C8_?c4i&^Zk9A
z?j|QF4qeT2FTKL`E0-7z2TaFDSUcmN{mMV(cfav_jLQX;w^3lynoS4qQCLz0iAX4P
zpH5Pcrv+K6@S#DN175oE9D9%M@W21xeHnz{+0%WpG+ix7gHDg#t!)&7+B7)V;9Wyq
z&vC}{<b!+ovZRwOPX!_x5Q~5hCK72~95vS>aA+;hW;JCkNW7#I47-Kk!cbB*3l8tz
z=Y_xW3;gnLeVHWB7_IeLA9d;V3;MkQc_%~k*ZIMFAF$C&smubOC|<pJkr9?&Sz^fW
zy%_r46DdFa#;3S^<s!fGFMfs1TkrAuXFrFt9x{muhNI~**VE@&AFR`rPY~XqgF}XZ
z4iYIOHdtC&OMBI}&D|g6cXHnQ{i&5epp|Y-apNLbRTYn)J|XX9q(#!2yMoqHvFn_}
zIveAZ+Rt`x@7a(2Tt4$tpCQ+Mwl{bA{@Xv`rB_}?qu6`+gl7*Q(rAS&IwV@sPq)~A
z{E(yah?#*4+dVed252<K298eVtoM5uKVv>UVS8hMR8h24RSiitKnO_?3W<)r(2_?Z
z0zuWS&m9qgI@5fu{OpzrycQ)}+WWw1nn(2YJ>3+R=B`Vz#7D=m%kO{I?&L#?+>d^K
z%aPLR|4&tcVfm8U(d;rlEt2~cK?qbT@k(G_14)KT8&nc`%MOWa52CB@AL4fwv<Bv_
zq4P3Ujdl}s-lAiOENPZ;;o>E-qQmUu80#EqScV-eyXQD<ih^26mS{{}Qk4rt*U-sx
z@}i(zM8r&bMP1gMOeR<?X*cEC^A|`mMc&EV%ved5cQ6Kwt;zBplgTmjc|<EuPEMH5
zXN<>V?!JGI7oLBKMCVK=C3o(;&(mj*c=3fBT)ME!#>R-E=y37U1v=dh>+9<r9*o)F
zKji4>grmb_ilR%eH{j%COjXYa0jg?>^A_;rc|qRESl{fivA)UE{YO~WPHqr5uPB?q
zY+;#~j*Yc-dczUZ@)$>r#}a(>j<Y7xr9?>3LLsyU7diG3(Q6YoAU1SL#*9=2Nutq-
z#?df#HMXj-wZkqfB9ovcBod#b%<Bpj6p2d6@&b{h7#TCAjjQl7Dy^M>+C*e-*`OKI
z&@@$R+#S$bM*mJ@u`UWSmywF`>*I7#shFXiWf}AN9Hlf#l8~nviOyDe$wEXCj&lYf
z6k2ByG%;p%Z!OjQO2wS|PN&21@iF`R`*FP6Ish!QY1h`)7z`p$uh;8w`Px+g8dH7j
zb?lRkVwFZK6*KWfzzL5m5{l7~^-C9+J>6x$X=oNr<T}`Z#*;?sW*h4y5F~WFDVysG
zEghzcZ-ldPp09Ky%_fPUsUl=<ZG6<VAx%1{M6;NdOq&HkD~iz?y5B`5IeE8>N+e2y
zbTz39u&7~$4yx8@IADqDKwn4UMq?|C@@>;6Nv%Yd7H17d*70n*;Jtf?eEaqu?;lk3
zc5W~WiVDf+zVHQhwl?tfoFvyQ7Gr*J`}=(PtG~;nUQh*(X%#}zuTn){Kar5ioKzO1
zx<IP}&`1&I_1#`ZT^bILp3%698_(aw7{mVFAw`js<{f_W3xA27OP7&4!Il-}B8GUE
zixS}i@4x#4@UXM75f`%1I)pDzs=*_OS*;;>%Gr$L!$a!QGEBQ{tgo@YAsLKvJ~)_h
z`BN{lb!D5M`l-)?1T8GKDJds2W(Q;DofFK#5!2}mMGn=1o0kWyX+^39foOpc1JPCM
zDFn|wcb&iUpZp!P(>UuuE9$BN7nmF#AqWgdBeG7;d}1iQqma?mQUZw+L_5djsq4IB
z`?Fp`#D%fQa>nCh7K@S)0{8BHz-Y9_*4B0$%eALqV+=}3>Z)RGIBc7|hPU2+>&JdB
zTich&l7hYcDZSAKKlAUs&ep~zAKabt_V>QUTi^K(^YIvwXg+xQ7~PBN`}Zc#IJPq`
z*PCn)2W$*Bn4a9BF(-5q!LYNANMJlYh7e=iOc}$yV=k?NB(4W>h!~G2E;wAPCImiY
zGqY;6<S8=tq2vkL22Du3ST==WwdpM_aF+kS{Q8gny8KyZ|NQLtKjjOFA8J6)Dsbap
zhn9;i5K<G`q!1yaGF2#zN;6dAv5iM03g<kHX-Kl9&67Th5K2^=*XxN$N)R65J;DoU
zb2e3)GrVw-&7F(Pj*qd6If<5}N+NK0ZwMq<8z&q=C=`L|bc*nfUXio4xlWd)7(b!1
zmRu=xp5ZMt*0X<j#DfP9`NCiRY1T$Vj*bp6#*?IyUa!l`uRPEF`yU{cBrh@sgAVig
zlswOH)^hLO2i&=Hmzy_UWVE);wX1>`UU&(uJiTs23-=HASS(7kN_hT-8@%>Lm*Asl
zqAV-6w|6-_Jml@S-v-ricz8fkd5XNx=GF$Uz4kgcu3slhQl^tRkDlCT?_fsVc)S+`
z3L0k_Pc6?5rrg}UOs7AhJkDv#8P;2LummhVSXz}HF@3>NmkWFlSP}WpQVWt)<Fz70
zLOp1OAy5WS?VuE%I)s*%MBIC`1eK@gG>4!`v?i4?5Wu@YT{qOmQ5uJr3aK?hs_2oo
z%4LlP_d&MzPusW|W6((w(@mUj(ah6>Qf#vFRn%f#*Q|mPlp-KwbQ)oK$pZrKEoN5#
z5YjD|O{g)HWn9~q9--D+nx<hspRas54-O7ijib^MsWoYqP!z?-UdKM+C|06W`~ZzJ
zG_Jt~i`9^<^;o}tg|ewBA3h_@+Y8?ZGCavrwI~$cB{&=KLE(cTC=0C&J4P{TbRx++
zf~q76D1t|+1l!mcvEe=A#h5Z!*4DP@tZ#9&sQBQ~JvO&SJa_Ffc3wp!%*U51lK{EE
z35Uv49F|!*!_=0{7NkXj)&V1HoHb~xa8|+0vzRx0cRuC!?tH+lhbQb+uzuk>2O_0A
z+T`#3t^b%_uh0E^?@_E}ynp8wx9{HK?c47%DQ8#$HLNZ&g^UXi3K3F7$Pqpx*p#vf
zc-JrybGo@^v|i9Dwi&H&@y@O9aCmS`r_<rmrHsyC#A~mAiZA|qzd)Y%DC_8bSC<V<
zStA3~^BMQweV0+L*cf(6WXy+@GD6v{R3yUJE>M>>)1wp0<A$An#%r(KU}H0*Hv24O
z!)|w-pehcI9#NGiJllT;#?Z{m7#1xP9(qAn3$jE|Im>pc89B*zmXqp$!Bg9YBuPlJ
z6s6;dNU3<D%E?%m2J2e#b>pZe6`^V9baTj4s%bQfloC?!B4jQ=hG-=amd>Srm`q%J
z<k#~)@K3yi6lg8^+~+>U$#~3L-+u>y*>uiqHV2|dm(uYer6kX@*6tymuosJxZ-4un
zKlXFEefuHZPM?#L34i&oejbqwnKy#XivzA!H&ChO;MpE~Paab~eaf)kLraNj4C8r4
zXCCjZrnZcFo0N+RM4%%y&khdgWNEZ@5(?Y2yk8ZK7qpb8$g&hTA9FQZz%EW1g!oX?
z1k2H9>vwggu{%r9J{736kFad&P8+d5%1+^<P2bs{A=|7^FFt%jEgQbjZh+!6RTrSO
zhJhd?iB6$zkO8z#2s9vMwC`vd3SY!**0wRk`<2YH4R&D36<b=i1OnJLQ#b3dwX@69
z5AIS*MV7SbA<Og~;sL=l1}hXHNvP}$U)N{^gKjT+KMJC#KoSY_-r+-FF)#Vvci*Dh
z?{e+AEA$6lI-M@w1t#MOCnv}F;OKR`2mw!@Jq9625=FPyVQYJfvMkx(-zQCTgou8P
zNU+YbSj-uZ#~dFY)9G~S4+dnJCd)Fs@6gnSG*fiC30JROAW6PJ(J59)QJ$rAd);W~
z;UUixy1g8`cS3C(frJngfrMG<*gu#eR7#R}K_%4IMnPMYv$2#%TW9euBB-?1JyOXC
z0x6Bx5-$ZtgxE89B7Mg>oDO&;aWcUwjZqFOqvcYlYG^zcwl+a`up(jZmf}o-Q#$q&
zOXaPzSl=dkv=js93^CI=?sChqyYn$1!8!{;Mz6Y7wZ25NtzA%6HF;LFhfjww4lt|H
zYLX;mX^z&KEYDdRu94@lAzc2NWm#;}lv*`<OSJE!_gx5KWkY6+!Po}t4TnejRMo#F
z#7-%NkYHRxV@eiHNmDQIw!vo!#f2@l+Y{#Ty(h@pllnjsv7naZsW3)yaxzCK&tM2j
zI)d^DFOg0G8kGq0t|pyZnkHtwdaEJTNEB6BQr9J0TU+$jcR4tzdHePUJa}}(tFJhE
z-GWY*(RdLRSOh|2aPtPKC0@)(i#U>u4ZB96(K1jsCFOL?tO;0?GC48a`QRB}etOJ%
zlZrVBgYC<dopr9=e3`%T*Z)U!(v11Z3F~VEe*4S6##g`fRc4DRwTmJPLu~3?XbHwa
zBGG7a3OyiGU9cK&6^V?b+DvG4BAA{`x%KuQSDxGA`RkuyZL>p`D{kDp&h-~x<kdGn
z#bCG=4Imc_=Ce88SeiwJu#Tti-{sl85BSvOExKBgB??URu3L37OI4XrXcQf-*cu3S
zFRb(0jSYlYU|hz;33Aop)|2~u^DEzFZ(4D5GGWk581y@AZCzz^bB8<cz0boZduV6r
z3(t;#PEb{yWmHsO`0o`->5!I2QY42OMrno^kZzEY&H;u{q>&Z|L^_9-7(x+L1P1A5
z2thgoB!*BB=AQfeue<KOXYF-foOSluXTLamo&7xD=kuj7Cio_arqnp!4OT$9vFhs>
zmKqQLt~f#Lks8{+h_(|=cTOot+1(&h&~H>=Wq5v@!H-`*i>KX>&yg&`RiEjQZ&aNz
zV(?x#m(IPtGKf_kF9d<-m~wk96IYp5<X|V|_v69ZADBQctUWpy;-8zpIZ;Le2u)^z
zO^Wx=k3)b}o!Scdqt6($K>zSy?nIU9`u)YXVwoF6xQ^96*?_Gn`Lj~Y)73m=tPTzC
zRW<K~(#)NFhGZrqCwi?+LOuB1BQdThRZ?<Wb@dW>ZzFU;{#l+9%ZqZ;s&ATkP3V&Z
zG!WH{2u_@9C2r+Jmt2*M?Ui*Zz}P;_Ijg6hop5I-l$k^F_&eIey@w=f%L~V#CQ;+L
zJ7lDoIn%~1{a6;AKu=Q6=194-TOLypx0-ZOX1OUNUWu@`O_O{IiPTX(vA24U>s$3a
z_d9#$=!~)nrwp7_6TzwGPl+GFc$MpMNawyX%`wCUNxR=N?quju#&)TfIO{J|`Hs`^
zmDNzV1pk=At4cx1wsu*vZmFWNBkpmbP%f}xp9IdK$-c?FvAtch0qgjp1y)Mmtefc~
z<W@tgQY=$ff3|bs)A;ctG~4-2sOnc%9gA$aJ+kEtCb!cTI2E^1={pA%3wf%vO2U$X
zL^0u!Divw#Y@G%(&KdWo;w(#KPNfS|t$bBMy<!BEkqac+hi*qtr$yI#k`j(A2q+22
zA8J2*qvB{1Zre?}-W9_~%JK$%3lmz0d!zqcsm+Eyzf}&~Li@6~BH4}6&>$;tNa)nH
z;QhEbw?oGyU3U=3urGrzWyH^Rd>1}|ixM5%4N^D_1t#$wmw$HlH0f0RvvuKSxkZQj
zuk0UF{Lc1)hxs4de|419lyNn=37=CJH3c6bGc(uIUZlNj-XNpB^DU)XaEK)-RKsM{
z52buNbo{x}1Pz~b>fh_iiDw;@w6w}}om@0s`MuH^uLBLa8T@#CN=&!!C}fr2S9WB}
zN6%?wxZLf0LS}Bq)(BueK4kp3UP)iPNENyyik@IhnxsvxIL%;5t*NWtx`!ShG{Pc0
z=9gHH{*Bba6RBJyeMF(+)A0hjsj2gJ)B7@s!G*AptK%4@N<aUvTd^L0I{R@Ap9g;8
zN${^~nSwuDT?fz0%HXdf@U`J*S6}q!k)rwhqK|A#pE<VdPai}GRqszhHQ+cIU$Sm!
zVoUMjo|LGDYkkhcR0(elr9<D{6LoLv#Z7_7cp=+RoO&~l@;!My8+6V{PJXgbrBv?U
zd;=50#iMiC&rm~bX9(`@wJ3KB_fHUE6ph?ZE=xNe4?Szo`t{2S4(xlci~%~8STDw$
zb+UMYF>~WYCvVz-9t|^e*C*wP10FQ8xmpVs(^5N8g+NO)R?`nQKR9#~E#%V~IH}tt
zgVssx(D2Zjsg%-ddyMNvn-2@=rpQB3bIT>Ng7%MHJU|ehZ`wMEYP?9U={Vt*$=9IQ
zYx|~fs^RP1*?EeYiAH-ltq@<lSVQMOVY7z)ljVajb9y=O-1d1`d!v|lC$}6j<Jhd8
zIb~MMXaw*ivvHZ)O}p*{OwbgqKl^9ZnmwA}WConc$k6YcpZc5t{^naoKH+g^OGEp}
z>_<2I+kc;b`M1}3HR^?FeeUrgR*3CE-~imz(R#$=k!Q3H;`9#-lZMikNxmGs8KeSG
zq+g|>CR7x@_C;rF(`dzbEL}$bDVe%<AD&OmxTzw)@{|f5)Vzw$`;-QH`TY85rN?nj
za57y)!j#V_$y5uQ`!O*`gQkFzfy3c9icpPjj+5g)8pQtQ-iwwt=0k}#w})xgLs)Vz
z)z$RT{)^bEeLJP+JmH+_gg#{nL<6IJI;U|Sd~O=KYED1H*w_?T{pJwqYP_8SIk|mK
zx-xkgJ`Gcc7vB|$a0|3i$qXf?fo?W;B#%ZF+pF37w=mu!F5j(sHH;XD^b5h5oH*}x
z9Y^<MFD<E&ED<IlCXjz&Y~|DduXC(icXdr-$9u%dhD3<%_~C_6Es-J7>}(ud!6rj3
z0O2&?uima`FTT?|JOL``e{g%9P=zqVIGxGD<1PKTp{f?yWRoCe7>U4KF;LpIq5k$A
zjr3?#>vUU56Uf%vu$Kj?4LvFM8dp)Q@E7f@gclBmF?Jh;>Cf(_2P(+<2zlu_QwBCl
z)0OK&Pj~<%tCFw)+3Z?!r6eV!McBTyu|#4b{ik`n6qQGwNl;z8gerI<O07BvpU(%N
zT?RWegwz{k6ZDE(`4moY$k`n}{xL=7^_`#q5C~LuJ(;rozo|wopGRG#9mLsR{wWTK
z`_<@4^ERd653La0P}uS5D~2&;2s41TZaCUGegW~%zEs`IWMPPBRx&o@pK;JOdMF-v
z4?3ouSuVfiou(43mA|4q(}?KMpHw;ivqS1lV4i27M+xnhTKEn%j;_%7#eDV@bk8z!
zx)}bIRgj5NMX*<^8N{6aI=`a{{IUb^7o|s~*atY?$va2D5o%J9wO&gOWB@9Y)r>{$
z0IGCqq+CT|wRF26NnWqYwI<^iV))NdLRm;3B(VXaxUm<;Z^S=v`l_IQ`#Vr)`pA0i
z%O5+uPY5oBi4=owm0!bO+Ff)0xs?4fI5+AV)uRG-rtdGHOW{(EM<S8d7vD3v3p@*q
zgPuzAA8c$`z%t6UCb;!G;2=%G1Hq1Q<D*e4YB@02McQ)va!=M3Al9xsVl)LNrhLMa
z?S|7)TQ>(*oo6G5C-Zqsh}}V|`>Y*qaWahbgqXz~zwPeY4<QUSLm1bBPQe|x;V=KF
zS%~l-)&Y%KUK^|J#+MpxpwjHf@;sx^crRh$)%4dJ-zx4iM1?<nw#}aec6`G}O5Ihe
zxkF83=#bwR8vX0M><cnji>Y6rTfULlMQWBbr%P$JUDjex(*$immLju3N5?-w&@3pe
z5h|#Z(^hSiFA<=GlWTZ9ui(-DW}D^lML<N>@i9m6?;oqz7l(~#^uWyxN0PewQ@$Tx
zBB=L7nI74gAMed#HESiX4iei|KQBiG3{$SF7Q)Vv1Bh?H9`59v`t#V>%j4}2;Rlj|
zUUP~HPL^*wPru!R;4&BgVevaU?0T#!la9|G=+;-J*O$Bx_%7*tH4#(Lnw%sH=~<3G
zlsuR5ut{0oFSZ1^CUhH*6PIiD*(dx{ETQEAwS1UDv^%Tc6GW_9u@ygNkF)|oYO6kQ
z`jLcNGaQPhk#aADThC*Zg=0L?iRvN_hIIdgK3BD3YF<%N`H1$%le}5^$ywTp$s*51
zB|5rcPh5H=g9dn`Qbw$%72DWcTW-cxipObLnrLkoyL{iZw(1q$vq9Q%b=vhglLZ1=
zWdci;2ZU|b_O%&2db`wN@KB<u<`VC<c33kqc}`r{g}x?K*QQI~CgkPuPrN)3$E=(x
zDGLLK`@U_Rpy4I}glzypRf5$^+>*2T6T0W-hE{LgTvT|u>r4rT=|@<VwU%S|d=D06
z3d4TO6Gz{E)OG5hSeSJ2WC<B_3y=|_1n+`C*e8TFq01V5eba^BvOS+D_~}6)jngYy
zmT72O2f^OMq=m)^go=q3aZey|53K-HdA-||5?AB0EG>6i5=twn$uDneIk+NN@BHkc
zqlb`cF|8W!2Hy!^zn;2$!AjOQp~y^v27VO*QX4MgR0ARd@ghw-lWl@PuN*Vyjky_0
zlX<nKVxRi3I$nJvx_Ye^6_MNBPqEp9VJ;8x3S2anB?@FvI%<P%O?aT3_4-LxEs$6P
z#5_gYQ$;tTz^WiRqLoxyAzq!u<mB5(Si(D1yqO|$Hj(GPxydxrOH|Y4GwN?_=7?fk
z<w(&G6v7|876`xP!CK=F>&kH)r1SA&BjTBWY8Wy&xiPW!J4lvPiwqtUvJh5lAE_8z
zJz1AIv!+W0&o`=?kyG5WM0ENX$aS_<sg+9IibWTq;0ZK$2J@P?b|)mLsg=?&aq5%L
zbukye4<%j(ef?MLv;}r16};V}+tfPm94xo85KAfMNwLG1(YHY=0^Y21I-GynBI%bH
zoQeuRta3up3m1s*cu(E;ehR<(r+<U~(k}u;H^oH5^ox$D@e>gP_@fGZB)wEc;_=V{
zS%#^2sR5s$A)TaT2ZMWxE|oz^5<qxo0G5_nmG)|F=p?c;vs8xAq6xn*wBE=2`M92>
z&gbk8=iyDI@%xRP6|P)uSmj(FMNP*1D#Y&Z!J{`FabHd`NpSVzbOLOzL106M-14L(
zv8)bTS3p#`pu$E0J109vmZ6Uo<H6{yA!G<7Sf9x<fQ6Xk%-%Q&jgnZ?NYC=Xi{D{I
zQRHeC66BHa_jA%RW3giQD<y6ZcBqsu<!}r34gR%-#-0>{^~STxAmA1E$IrAzo`ixr
zhyWdDa%XUrak?|~hd~$rYrlKU>k^Q{oED$5x0ZP^t~tKT3)fKWyx*EgL+7`lM_+~u
ztRmy7IgXFIRe{j|5-buOCpr~F!&VHKA@77-0L<i!=ki~!7wy(=qDjmz)@6e%QBKfQ
zhDhq)kQMBnUD*y}4A;&Di8<DH`}dufhZh_Hd)JH7Sv_HMKbE-e75`FLzA_!R`Dq`~
ztZ6|~C$e)Yw(g=|T~Ox9Flh=`r1O=dSL(jIr2d+){M&~rM%#rh9>|ZDg8S`pH1ba>
z-v<1<=9Kofxmhk~qP=Z;cL6>7;Q1Spis>4mWIU`iQ2NVR_j-MwwjP(IHWv{pMNqZ(
zb+U)a{<kDo`BFZVTVr#{;>?dP!Ea&hLkk!ecB6I>2+~4A;BgFE>2j6bs34H_XG;nE
zKFcSc%VHG+TSHlxFf2g|ZeEp6sb|ZmSjHrO>_YoTl2bW>4NP$?Zj$yv{=);^5VkQE
z4G-W4CyG23qu?}NuFb9Tbvh0v(Hr~K;dcmrRQ2W){=6vQ?+I1jqKBt96;!HwVZjpq
z%$1^NRoo@+kxO!(BWg;1mbflZ@L=~`ayHJ2K^>OCH;r2T$cEj8e>%PnOg;A0eY$Z9
z)CvUGdDWF5OY3ypjf}8?r;(I}C9huk+_E(oeR%PEDQ(1U=swageOt&<zQ(`1E9`=W
zU&vyaySR!?Q%nX!&e$3$eW%t>Fn2{RA66Y($vsXeNHlFY{$!{r>0NKiSq_)4SR#Kp
zAe%sze<#ykWzCT#$Gk5ul#ydwyYT_&5RJ2aZ>n*%GRiaoeys_D?P%gXPxwZ$dZkh^
zf%09$g66KKk#LDht5YEVOc!y$glClnqz#o*6+3VXHm^nyYhRU=&>NkerR>{{8=o=k
zjs}rNh5MS=E~$%s&0UJeiE9sE!P;-py5>e$r4#_Q?oj!nXEllT6abV(^`UgNFbz)+
znW_S7GwvUy+R^^?iqgIl$EUUN+@(;xb;V?+xOFJk0G%iucepe$Y)v#@E#1ZMGlzBf
z-<eXB=r+5)J_2NS1b-jmP7uI0ceO$`J6nY*>2gw<`N8ZCnBdM?vo&@}ICTB2$W%co
zjUq);Q*%a8x*lUa>V28oC-#hp!*U01|0o@nK<xZScx>iz)Rz<7<FNqjdr8H2;__~P
zH1XRJ-y?8C_?%1^-G|_yfknDr-r_Vp8!*<nyp+pKRdPmk$lHT8pIapnNoHX>zPnMb
zU25_~NPo$Acq<;f@M-9HAL5Kl4*q$~(Z2lRo`BM(CtwFM9G+U|p(4L7@&(7g9J0rV
zzp1oy>iVwr`l;qDOGaJ7OKZu(uXWXKb$3eq9@KOu28A2IV1oHhw&Wb69>c{wH}FA;
zLT!<lQE@Ti2#<njM#9}D0m`XNA%Yy@v4JUGHCE=%OYn6!tAdF6_z7rXu1zcaY+`_o
zh9`aSt9X{D-Etp_;DD_y-X>_4xD$RYvwZVAxN5|cjJYLl(yQim3~KW^^hv2Dd7czv
zNq+3Nex2!SQo{^q79h6IYLP3obG94UF$aXm3kV{?Q+#*7Ho`~WN~W&BV5xsoe*U4H
z#mZLYDU^2l6S-d>7jbr|&%Q8LgH*IS`+_5I>gVSicQT%b#p`Yh@F`R*9!jykUmv_{
zVm23B4pTA}a(Q51P?keX{Auw+2g=_lRX`@Tx%?gBDots(d~~a1jRMq)=aZ!p`f=4L
zXkh4(Rf@CzmtVu!qMLcnxWN}l)eBX@)ii|I=+!+3Wkq4~15{Yh>5zL4vFP=B%*Kvy
zQ6HQ_Pp`&#uYAR-s1c2SDc@x2Pa%KJLJR0W`nV0DF^O+1p|<Hv_fI-s)p#YU50f_N
zCevA!7LH2qFx31_{tJ%+SVUivU<`#jZ`xx1s3q9C<Pt{pvyQ<H{lek#>t(#4_BhYO
z_kcN#Ls(eSO>vST*^>9$X3^>z`KeuLjMC-z>VR%8lj5nAPO%CG{UV67qvIWJKAbvB
zh;u#QCey$55NZ@d!|XAxfO01lvPv`K{5UsjWFoVET<$f^PnblLs|H0Tbl-}<BH6dy
zo7|;@PUhFVKVEKj6|L|yFL5Jfe(n}!KPf#eY#;lT)HSA@mhR5OF|kW}1w!REAxYcb
ze`h-qaUKM%GDwWUEU}S7eu`dx6XW_nw;zm8X=r!B2Z1ZRl3O<LPOa?WSLc^zhP~Df
z)sDF)=@eJ0?`=L}wjozW;Na`f&bSz!J^0(;-go`;S22ta<O1~nC0uW4vV`NS8Z1VU
z$V7`>y{UhZ_$}(1w3)ewKINJ7IrhRXnY_87ksi6scN^$BnP?cNA6))$JsgnL@XW21
z&Xn&@`lLHV+`u2OEmTkJpNqkDFEdXvtg^p*r3<a2-^o^Vi216etR-#I2%b~4r)3=e
zm)!Mn<Hopo`;L+5OIu!|(-<f8<F(5^J=?{BIdirQzC`d8>2CFlo>I|wdTDZUp>gkI
z*{u-f-Li!t9J2P#ce$1S#teBdA<jM1$jurgez5!FSCw3Fj-O9Ae{SGU?eJ#9!!eV8
zjTC>R4gY1xHFkk#QoZ~2W|zZaQngmhZBnzApq@UFhkuCP`0nysB!)KkE5iR9`Z}Jy
z5L|q$$rsWyfGCQ%&WecDT`L9Be{hZ^>CUsWhU3jgc*rB0g{{F&a%AAh4|z586M?%@
zR$NI`ql$y`6t9(C$d;97%vnoZ$B(xf*lQx=H3QD`p9CuGr|Hk-$+E*9cqYg3i74BC
z9YRF0P{fPE+|Mym6paRvR{H(mpJCd7H@i#RM1WZiQ>65O1(F{_T+_@$$n>Hvh960;
zas)p~4rydRILJH$%`;OWw(Hvkw*gP&QBj>~Ln^F`7UtroWs*jTn&+XXFDK;wwfybh
z+3q492b|Y^5qAU6%R4@O(rLh!y{EtsCYJUe4@d+q-2y%iO<N~HemTfHjC2<2**jf{
zf>9uMxII7j!ztqM)vjmLB}+v9Hj%*$rMAg;K7$#4rjEspg5<vrgI!mL$2$Wg6xi9h
zE??S5Tc?^ffGTd^(@Xp@OXeQ}tIgq5{)^?$lo0#$osa9Kzl?To&mb^m&#$cK2y6M(
ze`_K8LVCg#)PLR0Vn`^DKbK&eX<t3Q;T6pD95H{JEpX#WzkPpp!Buq8GS3(@!K~>B
zCTIIN3@|0g%{A5bRh}f#yy0beSSx+|Uie~vOUOI5^`HZyZ)EI|tV!;oAznGUF1k9Z
z;*6`F8|QTCw#CsHf=SjI1H(i*kNcYO@$5<=gNOi<nWfSi#t9ykP|OtBD53QgxD}KT
zB+8hO_ye_RfA@B`4C(VSpXa18{<K)5tXD8Mc4H1?TfL8v)}AYCw4Z&&H`(khgl>?p
zmV6#hxqwRvn{L>OxNX}vm+o+hv=FBH$xk(`g$D;x1%*K3_3P>?Yvk8Wkx%>k-WXyp
z$o9rYLT08Tu&v=H9BkY52dcwR056&w@oz(7{Wu3aw%FuO!cC~>RrTw^B{(#mVZx>u
zdmktvOLX?h8POh`oE-KjR8lH_Pr2y;bL3Aib&8{2i7AorZNyVttPSn_zV$9>o-OTo
zGym^E{s2#EccqD{^PLnX(U@{7+9s)2$6}XliCW!~9ChR(ry{7LTUpU47u$~SBT|%E
zURA<LbcrjD>z-D|Au}$o5nY3z=23+-ay^lqgrSbqE=c_oaOgBYi=i78RoYHa=>Bip
zjfDX0ZqTK!01h;-ld-r7_4`Xd=DPt4d_1J+O}2W?MsSw^Vh)30lJzmVm1ydElL+(I
zM-rcZt@gwNqU6o~h{EkaTcF{y_qwA=fL+pG`s+NL%<s<8z0Mn(sx<WH7j>A?u(9?2
zwB3uV$8;*-2Ao0z?sBS*rnF?gFNw)4u7k6E6#L=7t7z*Yrx#b|DN7***#FzE=Nt5c
z>XiQL2=x80Jw^vE&;5Tl$yehr{|yU_{9hY`7RUzu&)faaIRf^pvq|y)dCX;i;r9PO
nR#B#c&;I{;z<+bdpT={#t1)lERr=Fz0S8naruJUNKI(q~B$zD@

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_getting_started_002.png b/0.16/_images/sphx_glr_plot_transforms_getting_started_002.png
new file mode 100644
index 0000000000000000000000000000000000000000..7530382b7cafed719dab6090faa81ed66bb4ccfd
GIT binary patch
literal 285810
zcmY&<Wl&sEvn{T{HTd8JcXxMBa1SuJyE}p4Aq01VySqaOE`vK6+<jnp-0$7{Ufmz3
zPF0`YyZcma>F%{URzpo51C0a?1_lO0Q9(uv1_sUx1_o9X1?gQ9MZ-Sv{u1<%)$`DH
zw)XHgcl!#XYVP6U;OyZ5w4n0(>gEn~cH(B^XX9p}vi0zAaTj7|cl>_}Y|d^r>{Alb
zhwqc1x+v(o!@yvf|JMWiTf7Vi0}BJAC?l!klYROMs&Smiyg+>&-~mm#%ulzkw*Z-M
z(gbx873h?H`yiRX_@L7|wSjQ~S4c)75-};*D502Ao+{P{E8zX9=ASu1%ja2fC~VY)
zzr>z1?5DaPVAbq~3@p7sDoq?)4{u#AQ-bGfft}6|XFVIO5U;KCwcZC+=vxQ;{}=q{
zYvTNQ=*Iuj6HoENI{jZ`p~+4DX5Q~p{6D?>#NrM3&i_pNKeAxjGm7Q^HuZmHuQb4i
zzVrVv{~sBpVkffK|MK8XArqPrUe(+c^qTn;kbu3r<(c>2hElWK6Bg};Z@|;Z*Yj3O
z{v`i#0rCdEXSzKr3fVXT)jqx4OusSa{P%AL|2!L|4Bl^?7K0SKp99zbV{$9{p7THI
z0b<D8Z_lQ0relo}Esj19(3Idic4)%e^V>`C-bt{JqPC%udt`x(M$*y8UqVs>-y7ko
z@wEoN^a}0ey^NaXr>1QMzKXt`ZM{UXOh2i=H3t6)zNCElSZ1T023NqV^_xl*J^Lnx
zfL0160w($m&j)oac&nGP<#DaRa@Sic%3c_K3NS#9z_$L^th(P|?aro`Itv>Ut2D5;
z`7=uS;2nX|G0rMsj4&?(#8eo$8d$$%QGNEzRzf?AYQUVf^y(fDwE~%14sb`D4B50P
zYwldMzGYu+U$Us(TmGh%g4Oa0)ZA$fCXANaMoypg!F=1TpK3R};io8beXDBT=+1F3
z0<Y(BgPnuy#X9{_L$&NU)Nn5?kAuAwK7x`Ym_j9}G(!%N`HZ#<;!rJ}RpY+j-~{uJ
z^hR0IVY=pDn`X<~vySi(uhki|`n>%kH3@v<clf+Ty?|@<-5!6$*1%@v+@OWdfiLV}
zHJJeGaiSNh{*)>XVOFCQkt<=s1bx48QL%V)#jI-QRd-#_b{DXh9I|QDBZAn~FU_<N
z+F(>8f@zml!a*lX)s1`5Q*y&a-nWUj#kbO;FL`sjf^Us)Pj6u@oII+NRzhF3RG8?s
zt#2&<H8Nqw)kk&IFtzeJ&TE}{IV|yaJzrfsZ8*OrRK|Iq@}bkh9pTRjxe&(60<o2-
zn({fn9dvwcbcard;86Xbo!0Joe)<`3^+ZHQNy(X`GNY7wfH-e4GW!zK*ch#nn8^4I
zb=XQOm&mqyR!EZRJ3cNG9DzHgJ$$zY2_3;y3(sUigOT<RQm-uDCB7Jmn~1bA1!o)u
zB9N%L@u;8xNB#hy&U-FRcx~I@ePyKOcA$qQc^j!<-v9I@ckJ>!YWs{Ga(aJYEMnQK
ztcE0Hhan}<l%jKx4iXO-9bWsA-TRt+a?`qQl)dLLI6fOu^u#0u6w$^$qy;E)d!!`M
zXX=Z6HNgM-Eo{X>DVOLsi>y*4HZGH>W5bk%p|tT279x;C>$`KTUv)Kq@{T<ECTSIT
zgki&-u&8cu<dJiV3++BWW~_LqrO-2$VB=N+ffe66Dt`Y#vo~m3qK&0j#mnX_R{S=X
zTl1yGym_V0fMtc{=OOLE?S4dIMv^s?kg6o4-c!oVRqq5C+L+$rSpGiV$<{DjgDEjr
zOu7Pp4`HpM;0GdCPSQSJlbZi3JEFG=>lC5^{(YurZp+M7b>Cq9hTk2;!r75*8>6+d
zX0ue&J7=v3U(ViNf%)|WO3vb15+&TQ;q;jIqW(5YDZ3?xwmh6A9t_2<<^_TH9r#I2
zppQ|*lL<1DwnpJHmN0eDj}a321C>MiaV{J=sse(7{raO%Pr__d?7u59X1IQ-j5XlL
zok?hj|0>Kv!6BUTagk|7iQ2jd55o!-Rv!|fA%_#*+1;%tw(K}Duz#reiZd{%dKn_Y
zM9x>v&=URZMoL9-UddT%c{sTB8u+qx38B~@pE1R&!O1v7S>8ZWC8wbbNO0dt{IZkE
zV5Fs3oPu%yJ1I7HdJaF<UrHhtd^(SPN=EWCXi8LL`Qib}{RCd#DlS|Lf#UNUSmM<A
znjveZg<L%+b8!`-SkMiS=YmmczH=_Oe1yH*aer{%OCmQvUc=tcYAG%M>_A9bv4Ob6
zqM#G5gPGk&orG(&H`bK@gD_Pw6RE;)vhwHF)#Es~+Vm7-gT9%}tDe#dIuZ_wB_xnz
z-Ft&`U?&@ZJFv4#-s24aN^TO|HVT8~h6U3rjd1_KSu)~&EmEe7#6;(tn@S`otBDZ8
zf7LhD6qHv0tR#%hcVup$rJE(GM)))%LLiY`N9D>0tA&Q0@>!-}VMnNM{~<-9u$Wqr
zM=>R-n8|>y7$}!+W92JpqQXe8O`nRI8jFmmnBrqr##@0GKlrH6p2k>`q!r4&;3<`&
zfXkW;m+koChaMQWJ@ZEN&_O(TcSd+v8dK}v3$^lf<0SOCz&`9a@oZxY$%j>qb6=Z>
zA!<nuQRobUTlQ3^^{1erv#rjUwy`pB<zby6&J`2cIeum8oJ4b&xEg0lfN6+_1<bA{
zdKVJ-RxmY0h$xR~tEMCDbnSiIfi9WJsu~Jb0dba+=}?uzWDxw=T$Y}AS4mb|N*q5N
zIE8;s(P93^BF}tul$NB2?6_99;iS+0(}9naUUq7H9JTQ%!hug#IaUX)oHo*pEQicJ
z2i)bIem5Cb`bVKzsyI_86bBOkk1f58ugBmz0v`x=z~*3S4sCG4(c0Ef$Z-w0h54bX
z2~;a?6kp7~d8s00KJn#U^77F32>FYnxD0RsnciS8cVgV$gVr8jVgS4CdzR-WZk8qo
zNDr^X&-a>rHJv{kIhWq|jSu9WwbrD&wETloamdZkfDjp*ZI^^nv};hk{=625Ze&4)
zo=H6~=PJ%LNG?<GI#x!3l!q#JEJh7|7kSJ16Q^IFbe7xWYc#KDYhv8sG9|)5Y)C%j
zCJJ;`J8ZifH6)`1CfcH)5{P*8-vvC>u+JD!BGgHrE%;2qUE-IfTEDN{gjfTIj7pTb
zs8>M*fmi$K$L1xU>pA<j7iRBPfhCARZs=$GV3B)BOD!x}&w1WOb1{X3hw))XIl9Z_
zQ;+w<F}8_7>qTv^xBwA&xmLT1$UdTuypD8g(lJzOmpAB|P~dn9fLkOwnQ=oZt?;q<
zBIWE~*xfVl2pdkrNF0d?sdd#)+h#3V)P-tkd-+Aau0RM)t*Iz@wn!d774_3(hbR$*
zIf6Je0H<vVDuNM-xfXiB$b=qIce}kz6<!=jB;daB+{NlK1R6}aE919kbyCovfp@{Q
z>9-K(*uXjO0}QZuexM0Bj`8!J>~B5`>HiS?u;)YicjPAOf!w}*cMJLtFC0EPOi^Ho
zrroJL(I0QgRm8ilx${m)ND-mW4$F8Z&mC+OdmePPoM&a_^ak~A#oBykn7W~AHoY0q
zOT$*N85OWxuIA6~^&L-TFb%N>s$%G6mKNCAtu}l(8MMghH;`ORqx~+!6gmf2j{jM~
zs!qTMb|CymspE>nlHQ4iLD7$a^AEtxnU?e=JrcwD7NmK<epCQ(HR{e<c7rd?Q`Kbu
zcc1)=(bMdNwRQ?MgEAg{BNH&WC}jw%E-nAZ=z!zI>=h?4<y!TL_hQR5iHI~IF;P%R
zOo}m930J2u*<jXS?J9RkZ*^&gME(S(oKvdEBT9+e`Ud$hAPNudMC5hZlEC|KB4|L!
zMp8i?OfK5(y?6H!F#+(Mv<2r?v@aNQ_6X}X!31zwA9n#m_!M!k*jK~Li#grh!19$Z
zs}L{pOXmZEs+7buE1NA^8m@rdv6t$6x>Y+hbLq=U-dlX?JgWX@st61!1z&EHGT^sx
zg;AJLLz2$Npx@+A*ye2*e}&v+7dq|~*2s<YEg41zwJbmkLx+$+7dIl#wP5;^5t!2%
zLE}U?AL+JSfb1?8kNR^NfbDmn^+(g!$9bSm@vVUjldOD9(m>_jeF!+H7qj^llECZ!
z>TH?6@Nkusj@4(d_W-e-+1*3>NBVKi?88fQ_mRh&4F3%6!>aNQYHgaHy)#gwlY7@&
z2?=ig0feCKa2kG?A;|&!FhZ43?O0I5=dc7m^#Y^R!>%Am>_YY*ihmx9QnoT88VILn
z@z$EW_1zHVx!zx*3QZ3B#6Rdcr+A?-J5?v{2q7j(!D8AYD*4W*m<kuZrApX4!43wd
zW)fE=zTQJ@k1%<OVwNBMMC%eDBbXo(_*IS+RDZnZY6{_6l{(C-o2q7i1F(<_*dZc%
z@B}MqdwbajwDak~lH_@=dN3%*zklFO5l~;VrB0qf=<&g=PE;;m&Bgzv1Z7|W>OAGy
z1o=#=@+2QB<fq|H^u0$``NekKw|nrxV<R;NrsZV@B`saWhC#&lUU0F-y0xR9M>jtb
zDNaH-dak-m`qMR0Ts?-WwkEUWoPyhzB!#e&5U1{DM`>N%CROD^UMc(R@KVP*lz96(
z`Zlmas8Sm8XD!73Idt!}4Z^)HU*2)9>+uLh%_Mq~651u70v-Q>5IU5rP1;Z$s4mm@
zrA#8zfT^`F&WsZ&CnxsV?kfMdxtNl*%sGfj)6iURb$2&#S*(V$_mC4cfv{YXHh5Lq
zKH7pNE@!q$Us65*Z#Vx&J|+D50P8C1Js}h(gd>|CesYj*CfJ9xiRS7*d74M02HnmS
zdLaw1y}<Sgt_Z}fiz6u-Z`zJyH@O4{L+#n7emd|wOYnq%v|xzW%p4-s=o}3#b1eDP
z1^qcJg;>Y9WYg_E+~iCcD;kDZ(@V>{+qGtYvA&v(l%4&G^7MDxWpx}Ih3(j}N2U<q
zP(*H`hddmGj%3V|KMOC^22wZGy1r!C_DX{eL$l8^Uc^F<d0>%ih!XAB_-^UA5O6}Q
zgV~uPK>yY`M>ZPQL_G&L^#cBKOWnefLHq-^|Mv4nh3I-r&zo+2PdJ*6TOZ9@D-@Wg
z{P%IRz?UK+R12-jn`znf%ib4oB&mIU2tI|%g8{y?nm%id9^2L|A3q8oin)v0Zdh&J
zU`X7nC(2Q5@my|3UT^T=#Lumpmz%s)RqI3AcEr+7-ZtdA=o_v8YbLAFM@M|FftxrK
zmeXY!97*GVv*tEyo~Wz9msf2XbMX#Osou_4exqK99e;3&4V(0V!Q&)mpE3S7NGjxs
z5=cK7xtn9$g&gjR^ZNLFaWUf+0~>hUbQXeVNhRP{g|a(<i2Ts*`6S$OiM;$#jqLbv
z0~G5lxBWa*Uxau+58B*nO@R7}{7$6_DM-!UON|x@XxbSNYB`}RWqNIw+Kq@;NK(`2
zv#199kMQuQO0Tu(fnV)t%$(PvN^8n1Nj|=&Ur{kAYQ=jh*lYfQ`6FxxhZ2iPZwwe9
z!8gy2Aex)wp<_pDV4+duS$Y&hJQ|s<O*45+fp+7Xxct;|3-FUN)?GJZUv^kJ{gdLr
zbtP3>WV#MNS(}8m!=i;!tTz7bBNuC(ZT7<ESE{NC6c>o^$KLxJ=CDf~SXXS9-O!Cf
z(yE?N7;i)U%X#}fHrCC}S!BrR^lQ?#n*uWQ0%cjTKvOHVgej3ie0%!sT3wH2iG@&^
zZ;ha^S;mw(TK3b-Dwpsl0nyy+-Nw<EQ=;j|-8iDZiS!CzaXuBX5M)h8owd+Uga6u$
z;_HjSc`bD$!}q$gVgKkB`xE#tppzVqWDk}Q`AUu<4Xc^STSbLu@kwkL@zvagZ$U$5
zgzr`p!ZhMZ8gl#E)GzL0D@VfRK+WW886w!Vl;r!ojX=pb6WYGK6}q&|HV&?#6UvUm
zDsLBdq3k;z{k3x+x_{ZtAM00$9qLkB(jNGxu+F;qFjlMnQO1`a<@Yl3&0rr94o^PK
zzrgKwS+ft)b$GS2Ev&lw4I3=Wjy3g8aVe$2MnsaZuu1wT31^#j=9;)(oxIK_y}P@+
z<nu+^35}<OdKBVu9T+0_&+Ia{MJ@_q%wb7XER5noDD#M^r|5KCwqnr_2@lVDQZMDz
zHv``u3W?x${o35S$>&I9N9d8ng`?}C)90{j%h3Yrtq_nY$TE(i&vO~jGzgbPd^S7w
zCK40J#tDCB^632Gcx2lT#m$0!&l}R{{>4Qac_zw~JY#FkGT8s`#Xnec3_;gu1>ZK$
zKK}Ko;3MB0S3;)u`!{X6=S6S|B|CN9sCV>N)QH5%i^_dZF(-_=Sq^^|pQGx=!eQyi
zto`UoMPZFyW(Qyb4co+B-CiLjZONdwk+z8*fevvysYD<7w99K2;$ZD7si`u4O^x7Q
z{99(nR#VGELjTAN6Ej}{fx8UdgP$FZLNrh+#pY0X(yjUUp)N0G>JJT>Kmqn=1ao<_
zs=hx6fcKM;=d({ALs#Ch3+Agqo8{6#1HRO@-xh0jD%D@J&8U+WQ@&{1alIqX(ok<E
zFQ@8OQH_V$rcZzLVhrsema7b~os)i}ZtbNkt$KSIlWsvzC$$3Ny;j{OSNFR}=3n_?
zeUmST!P8IIm0*!zXwv9=lp&Vm;vpnyA6d_8ra|(@Zt!3@N=DuHaT^EkTFoFnVHlEP
zQ#*O`@t(L?_U`GdU!=G^PE`pfjLr!!Nt&c{jj;|yy<So-?6I|$*$aIgwXKs2BvnEl
zVl3E<-^T<_z;*X{=HxS7hSHeKxnUa2YkUdnn&V+mZIo#0uIO1)pccB5n{PfObmBKb
z-hULPgTk9x1h0!>9&T)=b$i)<;KAH5jd>Ii4Lc_D>5lQQRKHp!$N9)8zFR+tvi`fo
zRWHK54fe(hmaeJgYo}qOfr<8cI?4*>eH^;VU(ZKsvv|e``0Z-Bgqt|`&90!y!BO`2
z#6`x&--6#bK&@6VJ`kF5MPZS<Ble!Yr%|?!-4>O<)wGjr-t6%-Q-VfF225;Y49rPY
zcVTr_Ff#-VEam!!w99r3ux_GlE=P6)zq2$+i)LUKb#wd45gkb`z#z&oc7JYxKogc`
zQ(O|;(q-Cy>g4WFz}dZ8kUMQ+9$<G;yq_axfneuZ&@Dp~9u4aiGSzBNABFFTSTEV^
zf%G1~PYo3CYFgOHQ5F^8^_Qskg|{q`;_7@-*`)iGP}@<ViGITU^8_<H2qhHXOhs82
z(w?M+lxiQ_OjwPV&71Z$mbYcXN8=0cNE5{eKBQ3E*^f$e`^H<{GXhNX0KE}p30g(M
z`=-RH$!}p6TmV%(tsJwQ5~t^rR!32kM|As5A2IB=<E5knJPI*2#kuTp4>_Es5H&qJ
zkO!m~qv4z9jNV24Uh%boDpNAGtY9=5c4ac68T)U0jz*uNsJnAYt_aHI!)eH90oGu1
z!&Y>sL0YBQ(17?@>wW07e3Lqy*H!=|<%>95xfkoaZ<vx@yuHumA#7P$Vw=IbM)Rh*
z&RDw0Pm!SOkZG~tk@+%tGIHk9#(0*=R|M2p2g7R-X@-OfgL)(5dgj(_w=fEYlicJn
zs;lrb1-c;=T=mMrQs(77av$y$bC^X$B9BaZ{<a!#lTyn^Bq43wa8PcWcZJ(*uUr`J
z;vXnkC;MX*VN9kAU?a}T_P&eMK?vIf?G)&*kQF){Gr}GKaG8QfrXw&5PapTpbk9B;
zozsuM>j6xWICNV!xBXjl?%laNp~Qjvck@v_SM1;H9<oTlWe9DK2LCq;E*fP9g`Ayi
ziK7qp2vST*Xbi_xchC8x(^haOo?<01k{zU(-BqN2lMk2Y%veTnwQA`+LV<VOXAZPv
zp}(ROrZU2cvr-T9iK171-D4W3lq05;Cxf8P3{RS=fM)^%we!9PQ~%d#Orn6)L*n*E
zwnwpMlbIj54l#%B7^UUQKZsaBTiUH(Xgf-@?*+)$X$BsZT#$+h7*A#}o5E$JD)Xz-
zyL@lpdmo2u>k=uF2((p<baIwOVQOLbJ7WlqFOZjkzjx~p8>`SEN{k1iTOZSoVH>cL
zu0eNVtzrU*C%j4t6#Mb?!yBtA!5VP}iBR-s<YTPkhhq2Ep!55iL(&c#Zp;r?+4<vS
z<SP#)YhZXu!Vopf5g3=#EzTFRH@QAX6uTF&^p~Ryhv4Oq+`9|^BwF!YfMD1!j_ix)
zll8zUmk&tJ!N(%~C|#Qk+e6WUGozf{ut~(=u=Mfp6P5?FPpobfBK|R)1gg!J-5vk7
zpc}nX%OmrAfS8Z8@$OBfB|uWGpEw{7ur+SRuv>evBZ)+WD4ub;1Zw1xknM6;;#zS=
zQfaw*cxrJmFiHol8doQ!U_OmUHINS;0L<)wU&3YQ!G=j~L}KkD3FoZF8d=r_pod8*
zk9&w+sB{9;{gjh&nnuFva*^T=!5An9^4dxWXas+o1i+^9(&EJ4t*!h*&)>_`H6P8V
z_sa0Bh(oIr4NF_Go`IF2A@GEG0QRj`nW2jp`_sI`utl-b_wW`H-TRgteE;^Cd7vuG
z8TnNHRcu<H-SAk6hB2wib)d|gu`+7dVsMTBOOxvOxPfbc5cqM3GsllImnF;0c<}0q
zlc7wwuWrpTpO-NQdzo`@T{>u1-+`aTau>HZ?gQ#Q-5BTbJ!g)0bD$LY#h#I~)YUQv
z+s73|T#vaqsxreau%11y#EJQ$=cilv%pUVC`3PYVe@f_?%K+kJ>m{jT%=X~#x{=)o
z?`y9|1(x|1;x%elhY8-e6J<7bc0G2Yb)&9>f7j%|ieHlqw6Xg;<#HXQY@5b&Y1|3_
z5;IgN-Cm*DlPxlIc;AjVN6Qi!=|^)lv(jl7gRlx+2Ot-KV$eNm9504|pxHl<X@pRm
z4;eE#xIP(qQ{b4;izH&D(9J|<+!_H&O=gA;RP>?1K9m=~7%?+|0ze369x~w(+?M+b
z<JZn%2em>z?g8z{Q~Q8YwiXHlM@>n&x5LuU(?87l_ks0P)o?$f2v#C&H#o%Du$@js
zFVkQE<SmZvT|T1+!~VJ5K5AmWs3QkZ0?_hjt+6+RC=xmkiAvR7X=ReKk$b(#QFnz!
z>rP;EFKWdX0mhdHO%yL%=~izFedmO)gd12+j}9CB$N`+44~41AG)SIonFkgv70`>%
zUyw38aO_USumqOMM6|rmd^$|M*}U95+Jo{uy-8p?T%K8z_`Ajclh3wYNRXSlKL71(
z$v3em&MoU6<iq3WrX1uR*NsY;qla*vZ!-V>;>vPP^nE<GQ_l2b{Z-D29bknFbsq7n
zc2=3dmuyIa+X)~ksi@#^Y|8wd8?_S}pU#<Bpx*d{mPtP|wG_&$sGWx<&u};t-bD<s
z&$;pFk2MK$r<Pgf%V;U6kE&oP&x#spqC!Yr;V~LaG)+ICRSZ+=>dhN4=M=*p1g{@0
zb*6R4Y02kB9?J10JXhCa`EOw)qJZ0NkMD|>CpCL<+K2u30>!Of+b8WWkUaM;WW6^>
z4Ub3kiGq|quQ=x+oc6pvhqcyz=4sx3_KObm4KEpJ$-JAh_%1h2W@#U9ug~@==*D~N
z<{H9Wni`QhpBAlN`tm&8yZfd0S+J~3UQ9}^ScQHxJ4djkGOFSh74L=x{T~HtMVY-x
z;h`!WIbM=$5hn7i@|);2xpoL9Pdf#^=LF!K1S}Wf98>Xd<L{ZJd8=6%?bc3Ct>OL4
zLi~2aoH7L(`0_<a^<pdH=e?0fTC>h#!lZS0S_<{V?@Dqo*!c+xarV9WN4MxSdUb_k
z+T}A)8~hYjcxeD?Y{YB2KjJ&?C<NNItg%jHIhPws+q39euAF{K@fbK-p%Ht#X3iCV
zgPUh;`Xjvx6FCD=&x=bAGr2x(^5XoFK-?h7zs`>gdkM{K!+1u574$ldM(N(NMhKaz
zoOCkYOBxTb;&Mv}yg8pS3jT>zGK9{sRM96z`w83AhD^2Fys@ibHJvdz-(@>q9W!BI
z2U`tMiTtBpuS|eUqr~>>wUg6-YM0`W;mnmQAWC7~?XEhl$E@bch5_j6rsv3NY&+Uo
z5<u55$ds$7Z-c`Jm0TDZ?Hzm~coT3`@fjqEot^<swtXMF@5F)C^P%id>*u~!s~0l+
z3N77fDxkU?^=py*i3wE@G3zrC53Tp$;RskZy?~x{ckd$ksI80U(flL5zbjJ>gekXB
zpt*l`Go$o|=WZZ;{6fyct@h|4uyLdKz3CSg<Bt@ec5*RdOL=!+_^m&ChF(?41p9Hd
zI9%Ct8bhp-&_N=n*q0psdL&smswxS=Si(%1va!p&bW~tUGUNBTa%3+dL~xR%a6<Dk
zJ*`TuVP?m<N~5aw{1U;L)5PK;fv)jR#P)0>g9%<n2?~><mUM+BoK+~nW&|@V(DFKH
z4VBKoFSQA6&@s#wAW$b4)yRq9N&Ou1yr3ougPs!A+0824ZdsQml0}J5BffYp1hrh(
z#%xbN4evK>IFWBrpI?5xCs3_UqAqh}(M{F*EWG+&ll{nm&DfPLJKHovrQgIFH<9V>
zoZjVkhgm?DNGZDiXr<O@ueW~M*0VmRLnRX?p`lAK+w_r_gl`|EF;w8WYCXCBar`JH
zRaj(3+H4O2vqWn53a4Y#%)!D%Rj1_<fH|kIA0thL{j0$=1`V*lim_3~k^!)ztm&D;
z=e85S`xuc5un%}-Az#u7DOg{334$iv+_Ww%raw7I>7KQ936!ln0*}8be)7G)#F@?$
z-;0|zO)AaoVpK?tI8^D}Z~~G@XWCJ6UGVX!DjKK1XV`9E+&nd}?k7QAzON4f_bXps
zp22Ss<NFayQkiNl1JReQj>`6|3Zb_oFQHM~`}pzOQ~vw7%qgddchI(TN^S&0h}zZQ
zcRs$O`y*V;S22;dq&E;WiZBf4v3HT${5T^7scXEjnOC>jywlK3Rgs!@?h{sj1jx=`
zbwh$<B^)(5p9zqxS5dM5@<I;c|B@{l6sG2u$IOnzjvAKkm<?u1IDq<$29hPl;&g|@
z!>)W4NYe5<%WocYQO)}U_*_HGAlGC661>Hi+Ze>&vCUw0Cg971b0~PoGx4e~GZ_3b
zRi<<WmFtO#J~UE(R)eYWzG_>9CfSe!pLez8#y01PXJ_*`4@vA!s5eB{bH=ityYB%}
zM89s&&M7v<yUSm+1-3Tf$+Sd+tl})cW}j8b|I+Fn|BF_<U^14HK%13B#9eVl0$%N7
zraJtKGZ)TUi+x%8xtd^#IPD!n&{na3yc+#!lxyl1==r^oBTH?(F<xX^F*S{4di*;3
z9WJyHbOiq9<6W~-&a}ZFZi-8HJSHp;xC$wCWSk&h`gmMgT4kTvV%@k=x8bl{^Lafx
z?`AM@OY?7O*W)#M(0%25$yrcHu}e_*(N5VstcwHqRfF+S^!x@1vhJmXYK85LK;O3i
zLNR*=zj@ktdJCP}wBuKK6wcn;qU;L%eK-2OfAjKH<M&0xV)DxU1G4w~k-pi}ov)Aw
zvuWPsgzJK;74i;6V<xtf=DCVyz?Ucl?he1ZW74SSpy#Y~Tw&f>`3PO}4!G&0bXBl!
zWNO?UD_b;UT$E;?TT7T3>z9_#|0nfXdgb7&dpG{__iFS5H;Ye51621itZW1f5$fYE
zFw3&vhg{sELY#Urm_Gz4NP|S(<!!rL4yIxn9(5AfSY@1BbaHBBQQ)U$cno`<{<Q|Y
zZCq%OykDBeHBV>Go0{o{YU&BTE?xuS;Ax-7%lOnHnuG)}3c+L{ukJ<)7;#E6o$znz
zsFl8)u_LgO_2(BNBV%I-uE<mheaezv&1UcJ>&ef68|R*P2$yqO{4K1u=b4UcT(V??
zIBUs5l>@`$^P^;1UQC~ph?96T&IQTCtMo&&i>L<9=9WDc?It2yR3Py`b9~86QJ#mv
z?-$MRn6X48*}7thy4v-;yq?X46@&_!Ma9xXM(P7pR(kA`hD24&5jdxaG{{OQY4|#x
z69<ckAP}fYbpyyBz#ncu*hepHG%+YM;mzq<vcNdE*qMb?y+PH-V0T76``MVwc1?C-
z6DcOxCoU*tP+3TsF?Q)EK@+_k--xlF!WfC&ZWza5!1fRolNfo9+bekIMc~Axf=EnE
z4EMz=G0FxbRKK`Y)1!a*BD`~O^K!(q{;horYe#UvZ&`gnv}bu+6`RNDfV0a|^p!SW
zylqRiwFdJoC_kJ$iR_*Tsg6kJgU)pgt3aNj)TcQ7cX%QHM=37``52iCX}O1j*y+ts
z32>>?7F9r0^lRrzy}qL!OZv;>tFl@`qM)eIzthwBBM-TBUD~5`&yIpI^luis^S_Fj
z{Inai?29#iZ~NEaeRl`A);V&PmT+z^7^tAzm5+ObsdD3o0ZlFsXK8}&4}NB6XXmHl
zH%*%!!;8G=$%>Ill08yU_Lg?V{IWY!rjOE5BSu%^UgxLjAc-gg1Re=`X4r%5ts-@D
zDWb;KfB!xL`;R4#ikP(pMW5^?Aw5qmnWu8rwrK;g3oB)Q!Z&<5mdX2*2z=iTUTdsM
z4&&UBK*K&zDR;}rES1c6cMLThVw+h3GP)Qh;}t%dawXW$n(3^Svcuk6xi{=@S+&VZ
zH~U9SzA0Cy<Cc9XhX{Chm_ky1R<X5KIK{?9gemCSB0NArkm`O99(S7tQy({OUd3RR
z2VMWPgb#-1PbfO4fsV8pktq#MoVA{~+FnNqB^=@_WDG?GT!}wyjkOtuWHTg!f}<;B
z%ZpgrETB!};?VLQ#BW`lwe}wDzOYUl5a_4Z>9f09PZW>&xIBl5C_n8ETmHMDDNW7h
z33Y3fv5nLY3Rk1oP8s@=Q-_pUYTmeFmPKe|kgZ9psFsVLF;W;opL0;mWnQn>Xl&YO
zVC3%llnJWs0k6cd`@Gz8myVVx!0lnu9aS|7;PGW4z|FWssBUA1ckGNbP9@AW5BVOC
zc47U+mRi3~JWY@t*rY#Ey%N<HMp=3gldqoc*(VJCsb%319B<l8ZTo|DX-Q}C+r~YK
zjI4*Yi(YmeC)+RsaWn2-z`(-N9ja@|Ex>9sOOq;c@A&xJ@NU$Z*Tn2NvCFN$S=EpA
zpAvg*r{Zspld6~U@n%YB_;Ac{&;4QC0xQ#NF6Z|T3zQ4zZ88YjnrnPZBXa%BXt|pH
zVDXU*=imaDTgKoG7>Cm>ej}Qt6>>T`uV5B<M+B$t+EfPa*<jDhCm0kVo^x*?gAfcP
zw5QMK`#Mw1_o>IrJL-*F9!r!%C-Fh^YG33#NLDA6o|lvawusOSIXMFYN~A`n2zwY|
zqMEjt0h#0=X*gvA%Y!7}NT(ESxph#$(A5ztLm5NY%MRViB1t@07MPw~T24d}U*L%0
zId6f(Gm=tD6>;>KmGVvHkzmyTT`EO|USXbUEpsmwuNdD7D3wm{;Oyj6n0vkd#7!aY
z`%cIqKEfkbW}S_VYLn}9_|Wh0Nt&yP?USAdH1JjBG>*dj@2i>TC<ShT&^qh=5pZTX
z_K<mF@W`hDn=?l}xSOs0%FH;z3ETJ+l04p#MXz;H`*QEYgys58VD4fsB?@}4JY?t(
z9+It#j>0e%r(;}_D%glM0<<cKkj|>6qFI5`C!_5|R{3?(u6^%y4N*M6zuyiZp_h3a
z228$+V>g{iQy-uE3vR?ACw`!daUll01g!iXjx~tf`FBA+f|Zz>r((LwqmYi^T==`&
zU5s7-?|>++P%E$3r()+r$^N9BVarkQI+WrtBg(z?qZNbsBfLLqvWY!zr4$O;*Y<zM
zyX3(S-ft;4H=siVwS>})&w$9iC-xrN*>oP=Vj*)bCXqoN-ngt37FET$2n#VwKGExP
z)%b@=?h*5N^b!XaxX{m;9h_BPgexlg^)Dhi*}fdo?(`ph%sNYGsA8NTHR;~0Gcvt=
zzH1fneHNTH4UjeY>jG!GO%lBTD3M07zP!O1-7f=O0O~j?ruD$hwS=yWmvosy@|CtQ
zC=Y?3#1RCnF3FdtdzTf9>TBI#j^SqON=s^%99lt5XM`LIKMKEoYm91PbseC_4Rw*c
z{`o^)B_188-~6>Bfu1@tzvk1_^j7_f6UP#D8a=%wVT=T$vp58*_l7@@%pMc_JC-Pr
zI5nn=zajzC$c7T(ZD-OGvdJAMfV>0=L<-wvPKAp(r~XG%NQy=*8tWY~G<1u)Bu;xf
zD&}D}*KzJ1zg$N8cGXU~i3e|uZ&HF6esy8~7v<E|fBm7i(=%?z7@_rWWjp`MeiRpZ
zw4-|uexvuqywfCZ2l5*b*W~Xp5B=QQ(FUHoh-IHzff>=6QIi#_k`fe0p0eL7o&U(S
zkaI7UfpW0^BGgcz6n^oOdICM^of-wqB4nnUFZwubA|kKU@6`D3Zh}AUE@AjMieWl8
zNgchROM>aUl2(M8ZD^_EEFLAVebpdel#%^{)h^{2u@$vo;4QX2$?-;}JQig#!ADhH
z%^!4~lbJTD(>|A%02?gIg0NlLPqP0-=L(Gcgg)~8O_cB7WLgAQ90EU6tQv0d3{_X6
z<8KFf1U}abPpPPwLb>R>x?8w(Y!k>HS@#_X)!8v_`*LX&^<*}-&Jj6mMI(vW&Y@Fm
z(`1x9%M43`%ctd)mk7cQ$IQ9Pzl<#=`3*vAR;Ej`ut!*})$}L1P)6f<oLVvoFeA{V
z-Q&qRvcPyy<jRuS;KAb5Vzq*?6$puXxZpGUC-C89@ms4iTYFDO0gLj#QTO6_oI1|$
zVOUv9QW#o3+p1QadpsU75X=2lg=kAuDrGPXeEdy9AWDX;8UIVr7RU(&Nqpb%tI8M*
z1>~^Mc?V`JAzL;uZt6|35V)4^*Shms>HL({`vmR^xWkYwU{a|2e0_TLkztepKNZf_
zWvZ~Gn5DZs08CbT_tk+y%r62vrS!@-Hk$xHk^x_FmpA&We#45x@9(g{U*xJxrHE^1
z?-_`DHui$9s(v~O>lj+3?T5~dZzmHNMBrMVyKCW~$9u}h#oD*9XmxS<vPk4g;xr|p
zUzI;^$fSSxk*&S*E#Q(bmCDY)EQ1{%d$!iSIZ`mQ@3uC8LDYW>LZtgdrz-F`Z6llt
z&|pV+bQaUkykLo&azm7{Ey9v8OYlex0cqAL;xfQ~w=0tQwBIt)8}Zx`9vLI%4ug}s
znHyfp^E~^C|0WUGPLq@L?#}DUdbcS30e;$s>b9aTr(t*lLX;g6>%5zkZo!YmgnuCk
zl5plWu8*H6#F7}oDC3HT=gf@155^OBHWp48A_uX34_`u!ztN1b)=LX&1H}@?go>$0
zG&mH*KxT-5jN9vkPLMlBImn2A>yS9C0%?d2Pcs&9EuRu%FpmZM1RpMPxMCa~adlYy
z<qg{$RZfLhkKf-usS`5KF#D#l6;+LVFt{bFSH{)WfdcLek@A(<vC?pQ)M3=3|C^|;
z*cj&4cw0hzlR+{LC?QHS=7ild7BgT_VBjA6FzEz-P5eQfqw4eg&J4|Y+wGl5+zOmt
z9Hy&0Kuhks3a3$<rr;733Ztd@rDAI!t<M{9b(1&G7j@=0Xx;p8=SZ4DBp4I!kK2S_
zD}A22Ybjtw6$cpBUhi<snzfPnH2)3-KgL)H$?hYn9b&`m1o~fnY!c$wjh^u9qhs;Z
z*32Sdpc5LO8AkwUSEa(v_&ro7TQ6a}o=16b)tpdiAYQT!_327RlwZ#2hTPDQ`vupB
zO6&;nqhi$vZE5p+gdg(h+WK!{nOqb@2Pzq5zDTh`aVMgpx&?-KDQCq{mj%l%ho|t0
zRWxv$KG~`&uBD@BAt1J~3YaqR2`LMIbOY)2du?34de;i}`b<nvPOmtA2>bOgtVJ*}
z92Ixt&X<LObp~%$C`waL6?sG91<yef`7=zo2NbJ|v;E3b*C@HR`~;FgHRI5X4%ran
zz@z?$-9e5@iaG>+f~b^R9&;K1FO93LXIR;e4$+nGur9KTYV7~GjrKl)f+x11H*dnA
zFBi#kNr%~1UPH4Y;Ms8PIa+Ck%nG0$E2xL(ueGwauXxgYDxM6R@56m|<w43(z?1_Y
z&BXP&$;sVF8FN1fzolkaUzzjQc6Onimq>ueNN4NC>kCuAU8fYox0Z6o307cwK}jhS
zk&!7ew5R$JdXV^6jq>{C$tF8j42Fl3KVN1i>Ctv-aBe<&RSf1T!s0In#Pux8)Gzvi
zU@}bXsr}CWK4i(ur+Lj?WK5r!r=HV2n<xlT&vq@dVO{Hd3EIO(qbYB-g1SVa5*W@&
zs-~8Exj6my&5hyd-%0RDCunOgo1Zr;UCogHG*ze0q*mW~iAPlxAPtjnCI=&lRVt3~
zFDjo+Y-`McJh7S2Kv_stSKb9{rpXZIDiQXS`t_V*emWE*O^hoh2!B8rhu7OOH3zvg
zD1@fZEn!<?(;fyA{TzEvzROTV7@3<F^mok$qvj)}`-K$c+@Ay!@iqcsxWGqnuBA6f
z*ujfp^WmPlUE7dC;yCBd&KXv{o}aLvmZ-N8lL7g!u!*<{b(J|2UFuOh%XG3(e>0&y
znO27r`C3a!Gg2G~j^yN>*lar~qtKG&$p~V^`2#5qemWo$@w*JU^HbGrP1Aj+fRU9k
zDRKg`R5&remG3?7<42nbWpU>@jx)H|)(<A5F-FiS;zr=w+N@OnWcjK5XM+%6=EcLn
zjHZ(kg57U7r-)~U-2RZgHLazT%}WgCSn|Ory;MB?p~MljeZ4RF3qMTnV=*3;h=*8R
zLs5Mvvaw0qwIzGx5E@)ve2oD$t-p78_V4h(jb4oXz71>k4TJ~#uT^GGOfC#a=wped
z#iYsScmA`kwt>OF1F7q7wyWx+mYVgBq-e8b^21T*d%YPy5CT515=6mOjR=i;tzB;L
zJ176C(+4$uAIc;Ff6!v|G8}yx{x<9<?MTvei3e`x!#!eABX)?#s599;O%aT74Sof8
zqW(_oi~Dsxh=_V<+5t6NF*W_h$Qyk?j}+&U6w@}E4$Fu}%)q)i?aXf$0tmHdKvd{u
ztZp{1e<qc=;plhRe?9r|wf$?S@%7{dG}pQ3iQKmam_@GJE+;T}w@AkzFE0}GGKQDP
za<?)uAOB?a$GWiqXwdYzSbxNd3y}9t`c&fl#g#L~_$|8>IHR~*sih#f_POfwm$a|4
z8#)=Q8b1tMvd9&2Cm$02S}l0WwlkMk;_iI-Y>%5Wy8trEtsSc*sND(}P27T#LSNY5
zGTD1y=(N5{D8)+<`e?Uf5puQ#(Z$nudD~&i5zkH=^4E8Be}&N%&59R{9I<YpxWxib
z5#!Zd2%+kDDAq;$AJ2G^J6uy7Y)3E{2(0|1s_^3ZdUR2ZM$aG)9&QvjH{86uK5!0+
zATos#yzPraYnRibDth*JUqp4akqmYMLLy@@Q6%;1>(<Vf9a=W!xfRs3kYX3_%?x6a
zJ_M*AqDqSxKsLD1*^6&I0pU954)QZ4WfuxN90NGlauKUpHI$S9<n0G6XQzw7<wTb*
z`Dn1m_8!dd*j~@rpXA0KZ`)|%Oj$CMk_dx?d<zwk&gTI=SoJ7ZIHNBGXKdI|rot5s
zZovg9%nMg=px7lTS<>1)k!)+mYjzCm<!O^}@~_dY2&!rHITrcyhz>T+kpzTsMq-jZ
z)r^rx+8Zrk3PRePB!odYQwv{=ie|z+y^>c7W%WIf{nRiXFT+49l_DTX%`3%}hdT_k
z%JkFQ7m=8g0>lqCMRio^v!*Y6#lSQbrI|?TAQH(rnNS>o%9}16vLS5ZlZYaBIhTVJ
zNcNV0h-r5J)c+2ymXxN8koa`K8<rgPjbp1Y&xYMFcJp9%B_VzED|K-<Yt`w~>~|K{
zTYs^s3KVzo?2D^aY=|kDEe>%E4kMv?c8AS3&^ba*uza7I0$xh7K~goc1=H`(1@R-X
zNOTtY;rw|mg>o@y(&1@LdKb36ZTHr4V4iL4IpGIk23ysEr-RIiEW}PbX#)WSv>l$g
zatRT~v0l87eB?2Bp`2c;{D9{zOVX|%dbE~}Hr^SexG1xeu{w~12XmZnhhPwQgI@ST
zZ(^{WhYA14;Ix{Vbm<tT#28cv=DTC}rqU-NFkF{*pmuG$$XhaH&?EUfi8Qq|Ga4)5
z*L1>Fas#Wb?)CG*1C?1f+w7;E_sYrqymmrDLWZiecnQrU?zIR*9;K_7R}?0lQ-Ycr
zFf;1CJ;<Wo%(JoYSGo#a+Sf_9dMU->0KOui!F51bV*<!|)2+FhF(n(A4u~jnQmVL)
zDs|K_(B4Yhf8v}Y4tlx2$WZ6OqUe@55|r}OZgl6$uwBDuCnr0X<<X74vt4s?)-z(1
zsh0nz0I-xg38Lvt7<*jA?f)}_>||SUmeB_6y!fHjqoTIZA`?o55IuA;@aH4TF!szm
zj!Imk30#Y#wDx(O>UWQ%eQoHj_zU~f#oL!R8Yicq?YGO`H`ANf_XA>79hydD`ojCw
z@M;qDc7YwZPOWqH;Fe_D9h@Vl9PO?;^Z1$AvTlUBdzTDMlFdy%b~K6A3hRulN3DbL
z@J(h5%R7U{zJ8#r2q_>EgGJLY(Ia2P#w9^xmw8h*?1k3Ns}AFr+e4nav0#rmOk{9A
zBfngE5Y_!C`~b&Lw&zk5nb&<1%yno_<lQ_gM&fxG%q(7pG!>XqTKMoMG8vplv`Oa)
znE?3V5Myw^ez50BcYeB>zxfH}m}O%8231ILrG~Y-p7Bf?8i}uGgyM&y%cv60TdM3;
z8}`=n1GlIcIb>=6tuP~{SkTkaHY@$$E@k|d_&`GGJ1*Q8z9gUCMisiCU!O6iNQ~yU
zNhn=3oj_GJwR$E-Qud0?^b=YsLpH33uf3z`F~qQIx_Pbx+^nPz^+Wo-d88~Ss<M`A
zNv-%FSd|c9xe<~W&+-hFV&~Avwf=Z~!y)jYq{>FFCNzbzV`HjkMIKr+eDaH0J-9k1
zK|c+pj@-L&IVUNU!|5eP{^}9JMz$B410rsRCs7RUM1cwR(;c~wP2*tG!Q!iGe3fc+
zb&Z4S;b_*rBcR5*5kBiP7p2IWQ)q1Lwe1~7btcZgCb{Qi_Et?QzRNV7BvZ+tDfy|L
z#b~dz1`+PHE<MfZL-tzKv5QY*4>jibo{Qvny`PBg#N52MCKB^M@Df?ng<NG3ugIh9
z*3}HRh_zR7vX-*y*RVaEuWemtZM|UBZgrV=I<8caM&OZ6+jRL{#Z%*<7|iJ;(ULH6
z(Wf#oFfare!ExhBar{!@7^%}^(?3_Lsi_&eiC=L@(Au14SH+{q&T;h$h$vSk<VVjo
zzPU6X@#?hf+~EARueWx+H$gcobTC__&6h=Ra)(x3GQDV}z1ng`w=WZ>l^~xu#dwl_
z(fjru;Q8J@qF+lD=o~Me0h%@wdp~RgxCF$VAz4@X`b;d7zjcrKf-`UqOO7AK5*A%@
zyMYU)f2UhP#d6#orpTLslC`$*W-ho`99yA|z<=>2yx-NyR$By4#u~$VpRwNF**uW=
zc?NFi?ld&To86QvkHVctoDS%%Vxw&;Afgyx(vgFT>v9Rkt?@Sn5v=7m|4@pA!35u!
zpyp%TX+Fe2j2$22rEYPoIV1mIlVKGZS^D9<LCmHVW9v8%VqV8}Bq$fDV~<RXDPvbT
zD*+i8Wdi(5HG>L1iCYHf)t*fwKk?$mW5qlCcc95kr}GdKmzTk)ovmIz6T@Kq#+T^T
zki|22HDNeU>VQ3aQS_2u<T)HUl<w|<ii<)XpSphL-lk`FI5JJ(mgDJU(s(MO{iX%X
zW(xhSO%a<m*5z+<i~~9-NU_ut{+UcT80IZ)I;6epC(HUqJDFS{R_RFypsQ}?sW$)O
zSYFBcp4e~r#oB+O|3$xuTCS#uOTeP*N*q=~YnxukvETfiRUn#olWrJE;>#jN+(G8>
zwdmT`i-JDheOQqvYpjM?S!tSWPEw@ZN3mCBsm$lGEn3y_29`8cW`(&AIkdS`N*5qv
zW4*44l*@1B8NZEvOhPY@KKspp4G~Uf8?fnxUuG=hdt&n*$$Cj0=4$>mp7I_fCn7*`
zfCoW+o6)NbaGfsZ4A}3Kw0xjp_RS}YXf#ds*`{SY;E7a@z{-y9TZX1Gr<t{sg4bHq
z!~|m29j3jStJpWcx}=d(T^wYDuBQb=ddf+OZm1KjuK$3ebWo7fo&5qP_IE5<gambR
zr9?_;kFJ(+%Hz;7kNWr28KBrSAgFAt%i|f&8P)MgznbvnaghJLdsMX6wAZ?ORpU!8
zY>$6yfn$%n+0y~L7<ea~ra@)rq(~`+R#NFL6YQ(EoWiq!zwqk;D_KeN+^fs?i6(9s
zj(p)3i$Y}ER&2C6y@O9jcku3xcj5pJVAiPLXizO6Bt*}Smfb88cu#Ck>Ei0TW6x45
z@6amzj!?IC`KhLSU1zBg6cXGk|J>ydfhW3Pe|niPy*j)?zldPL6N>#L$A<UHx638K
zawn6IiQqURus8tVek$NY^54uGx`6Hd`u(WrH;vxl(d_FeJqG{GoJ_t={b!h-Q~fnQ
zSjI5mhe>O2H8#@fpAZ1;Z$cF2zGs6bC191qpy*&qyx?W_O;Ba^*VZ)+{vXH&T&20O
zJPaAt$hH>2@rCD=^K6l%XcA*PVz|b?kBnDmUPIEQeW#$8d`lcrn31;_dqH=*z+meI
zi(7-{P$YhlGUP)id>II&X$pLZIl0j5d{*O1oZ#zVY0fP21{?ZZ^KWOV?9OS(Sq?;T
z6F$X{A00QK(P}_@Q-qh0FzxfRt3QLh`fiCKRzmUM$Y(8|G>@vShBRVM?3xFLsOZHc
zO$pJDp>C;u<A=GjQd-!CU3j%IBX=pVg+M<O@PmKc09NSk2V;s1%{3X&AY09)U|%oU
zZtkyq_Z-28m_*`26KP|waPRrcuO9;=oLN$>s|spJ`A*0|AbYmjma5hnJuIhee7z6l
zN}6L+8;Z;uy33c_Y#6+*-52KRM8!b0%zsuuHBP3n>v&pa1+DKha0N1TUUz0nF-at#
zC|@^BplC2T+Fec((_!jC^pO2)$2z=gcj<U-xYPGJ43jUxB6j%5_O>$WgJAWQt={Of
zLJ;3D^@B0(Z(S6N9f)m1!v<vmefb$$?L&yun1W=(8p+g_aih_>ElK!iBthb?o!+5-
zG!o=XZu&5MQL^htF(-+OMk`^*-9yC(B)4LxElVND=$XCA$k}~2W@uCfVSzTK&o9bI
zN>B@5+G2Xg=4?y)dldunaPiK)*LQXrYlx^=uQxg47!mK(dQWzYOj&*SCv3AY)N@57
zn0iAiDy|WW29Z8}gA1v+{u|i2W>Hm(si{jaalnP3Wd&noYI<`g9x#cu^&+ABB|Doe
zmG)apb^X$NZ|5k3ye+GuwqoYLw6DWnhVOlozf^7B835l}nzPBot+=upH>)_7QcA+I
zP97Xgx(5U&v-oQtZ+agYUQWTmEnlbvGt8W)50N?VG&Z^^vFo;?U|}E<L)_O+SzP!o
zyY25`bPGl)<n{O3)#?p#9WugF7`K>q!2B~wf|d{SLGROA@4@M~X`O!WZdPqDdg<^2
zud1#VNcfK%IJV?-lcfLY14fK(`f%A>{nRWEhU5Grqx~<cAGfhZId`RN-QRGdi7x^3
z%X-p^><+^=6bU7ySph*xb<ai3YRot$)t$|y79ip3<IQ%QHdzTU8HLZN3LRy~Yc4i1
zXwkMUh!SHF8Ou{slS4;oqYwz$!hnbvz<Y-k7<!*!QJNnohIeb$QEr|{OXGR~_AbP(
zO}vfIl6oD#T;qaY3v;L@%JAoD<^l%1x{G)*38?GvKF&oTm|3k7&hur+S;8AHDa|A0
zGZddf$&yPIE_TgWOC9U~lu-<#t{TTWeE6iQQLfIyl0&DNLOW5%7FrrwDoHF&ojomD
z1viE&wYAIvg5(OBhXv=5@0;8L)eS6xndxUO7M)BhPF|-<vM1Da@C+Z+1hGyp`)d9V
zMQ0fh)fPozkQ%x}8l+3QJETLpTe=%0Bn0V{?idv5ZUm&IVThp{M!Ngm_h<gk%)Mvt
zv-euxN{$!kM{aP^6381+aa3(*ofdvK{nkGFqYByuEh8&8@WHN`qEgbhz(0s4ZhCI@
z`R|uli_N@R@?Mjb;k-kBxO%4Hcs<J=hQalF&xoWyQdgrnrh+1!odLaPR?zbMMcllr
zXId_ZNHAJw;|7&6jRwNC#kV+TvN)%y(efb@JQ$Qr$=q;ccOWGP26Pwc#lv0546?h&
zIYZRid1vo>ry}M*&Sj#r1j~CBQ>i0Nac$w^zol!Uc<C;Xa9pQrnf&%7IvU=Cy-o&W
z-H$#a0Y!P4KDHXXXZ96g!^G;htlN3+@XW4|W*G`^z;A}?z_lb1Gv|26Lt=`Dr-aL$
zG3DilXo}pI;p65Vb{(HldoAk<D_A0M$SEj3{<Q%f`eX)FYQLK+)K6_C&X~XEV|KT)
zZ%^0UgM%louQ6RO(?;k@m^>I#bLr*@<fLY`ftSWIPwrR5h_%{BmDcvoz0=nhs^Lr2
zo~PX!L2==|tACmGY~gE716g>?o@?DN=?)#6fX%w3Vc*&SH%tsK_c}uCL0XZyHvIQD
zk<gq2`q$0e)h69rsa3n<9|uPE2uuy3z1Y0VG;={_GJJC{1GYHQ7J&bR6r@?#B(DFb
zExmc|d{v<5(__!$di~D{|JNJB=Bl5~tmzBHY-5-?*LXBeXNgyD*PPD`Qt?UwH;5r>
zr8}?A+mw<%FJXCmDV<||4u=SD@r<(T{CMD!TokdSSR8(8c|4{dl!vdNa{Kqu=6LZ2
zc8hg5OjUyvCh|t7x+5ei5TX4@>Pgaz;G<zwX7-TdbAY+HipI*S&thzl;W{dESJHmd
z48qO1l-mMAa|ia@?38Ws!^c^kwr<RKDXv*NDD~qsMZ1vB<IrsyObTDmKnfa{@W$tE
z>_qGWEYr(ddx`>!#GCE*T(4l)yHCk`_%TbksrOuC^L$m_Oxqfx5*;YgfNRlAL`#6A
zj7!PauuE3nZU1>f#WBe3i(|6kS2#qRiDV|Zbi2m6MTn*I_IPWXB)t^uJA?JtmZ{cu
zX+s%WYGpcbRb6%OA`x*6JPxEQg)-28ECY2UD@c~DPBYe%a<QpuVbR&giC@I0BL=K!
zpNm20POjESVy}VV>9hhfw6DBei0-R{WH)mu7`)rXV#n8oMevRR1?AAG@5x$?Yeeyq
zAdlgoR9(@=IpWTg*&4;1I;<nO|F+vb2tD|B3y!Ry?I|K|nnfO@8o}q;UFMS%vBr5c
z>~c2fUpuGuS0cP$y+?>uc$L=<n;b_CWH@9`{ejD=MUh)>gMMFAgvR86HwhDZJmI)z
zDM^oqyRu^<fIfx)4vIhC3W{1<#<rhYQCNFnLKQS+e1MbVxu7wbz7vhb2||8`MvU?d
ztD)~cMGgQ~W}6lRtK}4r7K^rSi#F3OHkERr9aQPxJ`KgrEHm0#1f#@J8R@dh*@I_K
z1Mk+lf<{I@PIzUzdw59u@NRDQM_C&T26Wl98NV?V$@uj4lYu6L54ws~#RoC0s3>Mr
zb`}5ONEC|%R}k{@R^stquju69Gte=8v8oq+H;LX1O=;&)oJGv|y}L`VdPzA;x_2qG
z9`c$}oOuS5ct(P*BA3zS97lTSxd%LaS#93DvWqF+j5_V!BObt&mM->m1<Pzy!k(bN
zy?=**j&8F4zEqx0PMbK|Q7yJj*m|<{V)jN0WcV?DDdIm(N1ocE`1T#`MZm}|$NZ>?
znyv4xN`pm5_tw_4vd=h@YS*7;GRT<cTGxHTuFLrDfY2a`wlCGX98Fg`M=};nvWe2C
zS@pD2EKBWUbe|yXmjy`_LCnDwD?;6c5aTockf^ifw>4EOxLk_hgg@}FtsN?EG8-Kd
z0wI26=wG?oCnaK~K!m^j;UsW}2Vdf=d4GTLqh9a{#sBz@>dPjHOr$Lg!Mn0a-~P}|
zFR)G7N??O$o8yJRnKTgRnK*xs3a@Y5|MQ!p?)J-fQ1#GusB`7~Q>KskkJk;Nopk7%
z5>I%#R*ot8A8AX|Zwo@M(SgmWxv`8DJIgR8#Q*Y>!NNg3Vv&1<hcVo=yq;gL)FEOX
zTJHC?e}|oidi10}12=MTOx<Dy-)#EFG^89igDZUwu4fP!0_g%$SW1s#duru*o--V3
ziJv^1onf=t5owGYK|Ny?)_S_erl#lH*slK9doila_LWzUl7>kT^z=T2>4P_EpWYgz
zC#$2){9znZ$2)b^YhgDmqhslHP_}aL>0sAHC5lb^$5N8%hkyKJH}SNS$*VY}bl=`w
z9lrY_g#10O-9sMNcu6w++b~#fDlVb}aPrm;p@@42(OFA0zq?P$v|%K_MMYhd_-I9J
z7>AeEZP%K(IIU2E<CD3-POG}#i`v_3pnsa$QeH-~IzsORF`AgHS#(h<(3D}Zmh_!K
z6%KESRY@)EA}Pglqan__{Bp!$-g0Uf@~se~tdC~%E#&pjy*6FRrAw%+l**3|*Yu+%
z@}ky8PX)H-D*Nmd@UrG{Gn*@iKj5(cH69kI%b7Fb4!)i-E}r&?v2V#Gr#jymVb;rF
zwc#1xmdBn#3qFsnxx1gtzdYK#t<*Idtp~F?rqFF39DJ*1%Nn<#C}N2T6TKbS_yl_M
zE!a;L>gMl0ay`zc3Cb7u|C2hT9_H1&cR0()MkG+F%L+8Tw&mDaY6;TbSUA#FRqvy|
zgPAYy<L(1TdtMe^(_??3zihpKMa=rdL^G{)@p4R7QD0VpVfcxyg^y9)lru}M(`I6^
zkT4Y<5xA02T!rvvNUvn?L#y9ByFP|I3k92C5=k)%wkFLWIO0l&A2=4An2A(p$lCtU
zp3?7kXafI~o{~i&GmK!ehXK1$r1-YpMzTr+PxluDv)BGOxwgHhEq2fp%Q+w0*9<`<
zDJw?gnm!AaSnmYhF@XP&rt9QuRO9cU3d)}elAtq})?6pipg+T{-o%*{d`i8pM&2eu
z@Qqxau@DlZ!W1L=^Fp$>=t747OokFxL=f8T=P->_AGhZK!a$Myx|ZfbP7?Xc3JyLG
zb_jbhY)D2xqNbRod&bQd;ir*vWg)U!^KALL>*MLwj67M0{CU5y`RO;k1mzFxEW3~s
z6IgF=aC&T_!}IP*tIH10M;XDL{sgtpu%rXt{i|N9gaoQugS=T@LQRG2MtU9lsk67s
z?L|6#RWprE`=J{P@A~9t(CC9MUgwGozw(GH=`y0)r_+9iv+vTBd)%)j84ZXkz9xd(
zyd2#%pd6JA2wpXe?Y;Ylp-F9p=&J9FbsC1d)a48{|KP#}F@p+A&CaFiL=OPV=hoKF
zoq3qRKNiJs5bm}(G}4XsZt>@dxKcpR2HTjhY1+O}!(rTE;2>QA_Xwj(3@u^|s5DuB
zVJ17fony(lxl+&3Cnx35Nr{V@>4H2jqTc2WA|B25%XJ{4?Ki1pldWjUzn07RxEGaI
zj@GRmk5~5Hm@}rw_OK@;OP6IklbzJt`%_#(R)nfchBO-7qi0y3?Sg9|lee&iJZVhP
zavc?FsN^hfLhPk7aMFLe|HgYb{`F%FgJ`!NJo;a_KXUWF@=ouzMG;TLIc**d^DNeH
z$LoW|GQ;;1C8&_SR_iS6s+`^R<9;%aVRp=ijm7!`7wD8#8yU?~rp+lO&Ma%-xFG2V
zSNgxch{X<IrYerwY&Yc&F=_?h$JQ{##1T*S>>O@K0TY0fW1oJPmTlM98&@H5@efLh
z>7V}Yd%`i`*No>(fJ>tp6!CzqsKA2W+O126oZ!$bV`A3izF#UCNC*>l9x*%H@H>}2
zc7Sg(ziZ>-VrC%&Q<0rGhfj6&?B9|Vc=pXUqJV(1`0GC;Jb5jhU=xG>(&5Ew)@}ea
z<bSucets~GA7hz-gEt#+tbu5td1LR6)wCAq(Q*7ZOL4&$<5*a*-P=z3MKlUV*X_a{
zZE7SicQD|nyg6h<A2wGixc{SKhB3lt+8P?xMAA@Holc@F*wCAexv=2+Bi>Ty@i%F1
zZ_Jt~`F(K`WrjVwNN>LZ+230vgY>U2pJx>*G-rPNag?`;u4LN4rvLcb#Tcx47esoi
zT%Ig^25=;{9K8q^3(6%Lqb2R40t&H8F$o>RuV+TmDNGJGZ#1m-D`5%8ehB?J82L>l
z*vaDc0z?HB<qzNZr(|(%M%nRgHB?uT+CLD1wpJ41H^l|XvpFQX{2A~kySn6i+ejpL
z+xMTYj8;=1uD5g?k#`N37!qi2`gQ1aq;*w<KHnv^8)|+htm0~f6o2~r$+lBkhR;rC
z){cshHS&9u^*1a%tSlz&;<`Th_Ts>)FYhzA-1_b({mfchrLDNF3ykUpWP^_}u)+tt
zgtSGXRsMK<v&ufQcUtTj3~*bPo`p0zQ*hhqAGH^;D32OWQ6t1PQXBT%tWK#5gM51X
zlFQyIPVErN^?C_^;?LsBnE)De8{xeBA4{?&CSknJ3kA)53It$2Uhmi}QifSn#VwOR
zmJ~+BvZ-A3pm#)Hugois5i?y>F_179Ol1CirFnq0!yO4P4^mrwn8RDs6nXY~<c-hM
z$lr)1W#(04snjaM`%3%4UePi-qqXi>u!jT}B=`A}Q}nNArmRE72RHY+jSEY~wqKd5
zm}>scdEQ(av`UbVmZUhnMLR2UGdE0rr<*5%CkfR0PkW09)>aw;Mgi8RWKicj=yV3x
zVO>V1vs-S}k5S-R-tYYUoDzHMnsk?9@g}N?z2i9Bq2=BDTf83?c$K%I3Qc?W4LoiM
zBeIvVW`QF%O@AyhC%~egZ^(ZlI?vXwIn@r{l_t%UQ)Q_zYU>Pi_@1`jM<My=U+#?H
z=X$1nLiO7a$WjUOtSSD<3Rlltrs&uRnQOYObAz1i07M}*o3xIYRhJ=$rs8VWii#7+
zrIR>~VkcmYCsR|OoMMrxFLZW3Yg@E5>6duje|?`X5z-rSMU_YUDPfMQtpStt?4W@8
z!5=T~n<m}hbm#kaRYrMI!5*LO_OsiAS7U+1=j=iWakfOeW-tEqUonN$)Rj{6xJ3j?
zjFf&~AO5P&`WXI7s1k}TmTSj@kw)+c8v1mQ#VIfHBX0(5eo$HxE64kyneSp0O2BVt
zkEodPkpMqUw-GEis?pT9XpTh*lE6UBzk8;{r>cGg9R+>}AUje<Hp|1J{<@DZ7M?ch
zDBN+nN`<;O5^P2lsb)p1F0US-YJzgHbq?QPK>XXpYviq|2R4&jSDhI{)SR;A*(p^6
z<`XSVWsO_81xMi-8G@l;{ckrCCrl`f+wG&M_7I9yh&-=>*VPX7`ZL1vnfuoDXE)-Z
zsmNRR5JV1U$-}hr$^3>+e|0Ta&cXdSUT(W@VPIAVNYk9PeaxJ=NEjm__l_?C2tW>~
zvOeWE_F2}hRWRh%B+x{tb8v_q$mnZt+2^L=v9eV;hg<%E8(hfIrA@1c@%hNS0@}kP
z-bE>m7Ko+*CQx(7X8zt9aT?lsiFCcTtRI`k)FDcAHGgn=*ONvUZDXjCsctLkca{`o
zi=8>X2Z;MGim4Sf6O|&~qqds3+UKNu?B8R87_`iqHGLYz(~bx}OGo2(nd2P(`Lcf-
zN@5|3ycHjkp48x?V2Zp^-Ju<&^RU^b%<TRnD0^^1ai($4F5k^VwCJZMWNcI!B;{Oj
zHqAOyVakoE^0BOYLli<C@gW6Ir(60?65b4^Io*y{YTe|~8Pa0a;4H*X&wZY7o)jUx
z6PDl9mQTX%CfqovOhvOk!Q{w`tNaW1_YtK(6a6gyibVj18<!+Qev^RP<2MnAFJX~F
zi42R}Ebm6xHfy1A><^Y@k8vOW5opGGU28kstVRpF>S_`pGfYl3S$&G7sEA>bhci-}
znKDnju!6svy3;Yauhw9lnlA8h_4IC5;#n(2MN5}HD<_%Avzd&lGVDw7Fmnv~cWaWC
zjf1<)=5uBhOY2HVc8lwJz3F4+zLEmKK_f<z`?!HvBrG@TW$5M+p!AWz*wlEIHxdP#
zG6{_8y4G3Dgcp&dtpf_Pb8Lq4!FYG=e5AF|*UZ@TD~Z+wl!a{MV~$T!UT@5vB>=+S
zJ)b<|q-c>2+A-Zkv;OP@9~y<-0kJKjN0*THsNGSD34!k%(P!0CITF{4TQw%GZ@tJd
zHB-}!$ilCN@aDolq%#C2kdMFP+!>aN2jSS2PSi>lH+}g)lUB+F=#G_F=YqQoziKnj
z%xk>=sGt3zC<1K1@Pk2ypVYzu2f0Xd9i^<j(`eMc*h(%|Q(Qe=TS)jiS{;|D3h}2l
zoo2v#p|0hKwB<gvz4AP&?#p9cO}3}kvL{W1Z+1EsYfuacF?xni!ZKxA4v_}xo3D;W
z61t-ud;lAe!FRSg(Ww}^xlI%u$oR!4%<7f2$PQb92G5uHOCD+{KCT(nKN<4?E>u|5
za=Fq46*ST}{;w{%5;5<TaY7&%B|Gl<hKA?x)pr>O?Zw>?g|bAF_=_3=^^oXqbXREy
zO7Y;)-GO(84WIlQO6<w-eg<_J?B+c}!DYqL)Eb7#HsVdGbu>|>e;BX7zllQLWL@V4
zTP%F}UQ}9+em(!`3<;a<FC!dQH5UKQN~+i=*{>_!N?)6nYLv<F;D#how*%3@{F#<P
zF{u<-%GqT<cu{{3-ZL%s#{F}2>X?LQTk5_2NaVrJ3kFyCtgzMiJ>SSP45X!QYfY$8
zoo(IQiUvKblqX9CgqjIu*B4b8`E=k}u@8;oHY}x$@S=OYR$tRd4&vER(kas%t~g>f
zYrvN@3`T<(NE}WhZ%JknPr3_gO(jqlRTB&Wz9B;Q(*=^;OVXtP{rbXwnoevI(OXo!
zI??1mLbJPaBI5a4rCPcGFoQ_29B%h=4O9?=u<6!}WGJcm9ndx|un{><Q0~!X*#<oB
zO^1(r;zV|PM<w`9A?f|gk;T4G`<FwwC3nJ2Pkyx!d#Karsyh1s`WSIq<Nc3cdD&rX
z{94F$?di9JcfU_ZafOe*mEvq1-F2+Fzo!h;kl{Y!a7SpsOIl~5ZV$J^q}MuyHbR~D
z>EW78{PCX<^f<ky=jDG5OISk#kB>;dHj_=my9?3&M_J?oR7byue{TiH9NMfKmJq#t
z`z(RddI2<oV@~`f!ycuYJN_pK%4s|2+d~I*haNOtA|dKV1_=p!2eLD~guv7corXZu
z&Tz?1w4gB27G&Xne<4CNhS_i*0oz3yB#9Aa9OLonAbdT8_Uya9?wMVfJ!MP={mcj5
zne=c_oJe9dc97t-B9QAVQDyBRwb%um1v98&K>jd^4VpOim&1E#ejzjwFFV*}`zon)
ze5t1RzDW7Id0t5cOGW)n{Z|~IA94bg0#>9)&B%vMW}mux8&L=66n4KEUXWlSe>B>C
zExnio>-0@e1J{XnWrItDjIS70`U`xjvq^LYoCfm81Cq+`277U38k%s#F-C>qhm!<#
z{D)xGOnxL1X5$^RGtn9k7;o8-l+L4?Wr9=PgTI5seZIM@ea@Z<Z&R~JFqsaBJy4l{
zx$NzT(woMA2Bq9M)pXg_t_R25z0RGAnZT^)6`&`ay)kOxbs=Rq^SAht9O5IC-zl!k
z$tGnc8=_{h?)!X}aR#%S6YaY^iY46r;%RCQ9#|f{r_Jcal{v43qSUza#XpDQM1q%R
z8VRd-nJhF|K14qjw#^v08gOOZ=HO#F8kY;D^c6L>G~Si|Mz-igg$QC_#qpThVRj#x
zl7XegtfHp4OF!a*HaRC85#<zcX%TRuKesxQrOin}*E9w4DI389BsCU}WZ{tP(qYa&
z3W1zpsvMBf$=2TL?dR6Foy}9osVz+j*FLHSI>xIjXD?g1<*w@&PlxD~Vu63Br<ox_
zr_04b)QcB3p*+a5yyl5aVEi<^wSWIC=Nu36NbHVlagH`VSoidZy|HK5tAwSGztH}J
zWWYj*OwpnDTETsN&vfRu1+TZkBL%O!dMf2wy6C+Z;J)^3I@~!`Kf8c(@ArguxlKF%
zquY$wf+E4cA6xV~1N<o<?Y72)^MV-mM+kIBhGEwLq)JbK%>^=dtw<h<{ifXr>$v)M
zv+Xjs7aZ%nZuYL31WBZG;p|SvS=P?2btCERtOp1lE!3X0M{yY$s{8*g6qurk90iQI
zv|t=^kWN}xnm1J}ohTi(e=_{*|8$0%3XI>h8Uqw`oSEt)wzuEr0P^w4{EO=avj$Ta
z!F-Aa1J}D%iNO9q@%_64j9J4YaIrJlXa(IQ&%}h9VD|WM3P8{W?oN3KznaAT!1kp7
zyN7pE2FAmdtd+7$8zCYyYb$I?qr}!Xib;g2_i5t+z!W1aXE)|>8dpyMeRMT_#A~p+
z!l2p(s83+7VU0KYdLb8#fj9S_rrl4W6)RhnD1#jpoB!YC0FdkhAZ_1x%V5_lx}9M@
zWG{EK@2c=kjEAax{xB&2AxJ3qF<J5q-_fhb0J@n@y<?P=|GUDRh}f_E0I-19#Pnxs
z_``=1Zdfw~RPk{ua%$((L<x^YH0_7e8shK^`agQ1OFw9<<*%`VFNJ1DFlhRp4GUCF
z72|*J3sk~Lt7eme>BRd54|bAX%>`a$Oa`g(v_Yo=WKt$i!L0?6FNsHqZ~h@?e?K@1
zz1=)6S$!18ZsM#~w%}&?jeB6Qk37M2&K6KBY!{}VuMr0^FUP$?3xBx_Jru0o`Z3>{
zW3<+Ns<Ghrcm{oQ5*1N=B8-A{&ak$LfCE<ws<HYadQwA@>t*!b57@KZJcdgaBZMUU
z;5xDK_cJ%|18&qsBcg`Zw6%;NP=r$@?y_=G=D7OLQFgrs^y>F(8nFz2d;SF}%2ZU>
zvq(Mg%6R;s`7X{_ri(@(c<iQYmBXK@fpC{v&l<%{N8?^Y@>Ja?)}Ms(n^>w)(s$BQ
zrpwN%^ANJgFhxivSx{ER{zk6Yv%a#>No5hCMQ?ey$7yQoF6f6`HiEg4%&cJrxU9l4
zdr(7w9a^!IWiKQo)X%Iy0b>05D<vMsTxu+Wm;HnP$rMYa*t!+&jTs6z3d7uH`zg=C
z2TBQjR`^}in}tlqLYLY<B^r#cCOY|dR8RlwLBzzj9vsFy1YCg`K-Xjpd#BIDLRZ|f
zuvw2PD5_+jEqzIJ|Nj3rRjZ<X@2GQ-B&wt+Q$bCwoFFw0?6YEg&!p;TNF?CM@!oZ@
zn##k2TPFVNLU`3I8iope6nS0$bctuw@^9?Wjv}*>p)ldIpq7cPh%4NVZ|2z-Hum%u
zMpgGOz9EwX9Shp?kvTb@402&p;*5n&Q_TtW4nK^>oP_r`mzXb?=9E44!Kr|s@{Zif
zu$~tQ#e3e3S7Oh`T`nLnjzD~Kdv}07=Q;sw-(?o1X>AabVjf>q&FZt804!+tt^QPc
za<M2SeR`WnDYkXBLr{77g5GpX<}R3fhRQqE4t6HJaFU;$BXRV*`S%uZb>@os4MPP^
z?TyB~xcisW>zx*LFlT>S3g>N+=l<WVvY7S+%jr~qtT5MPTI&uTSm=2c5Ph1Y0w9ni
zqLO$loU#H+>YmWaA#X?4ht7C9_hL66{YNO*7Ph?r7O>qF%yVNpkj(Rhe?nC(21Xk*
zpUg8fABK5;DJW*Ds&VT3R0;Gd7b#o{@!`$Wk<3QA-y*hYVJJpvzbT{FBmtav-FFwP
z2?>erSf?wi0X2F0?Y13Ii#XpuC!s4A8XXdQc@Pdx5h?3n#HuP1oa^AY$ZdYaUBlPS
zCd+NMD-~f*E?YO#0yX)du%3@op92i{iY0VFvLtg2{E{$f*?*Mv->9r~QE9?Ne&eh}
zuPF<051OW&eoa^$0C7_Ed$*_yVgDd;2|=%3Fp7X;NnlD&Zy-vQH+5KL%16n|d%(6{
z&`{do1W^<qlU)aERP95~q!QMZ&yM8s@+;6{+iI6U(-2*3hwYc+2eQzjGu&GKy)_T4
z=xB+r49qcaFH53wbH#U`p)K0gp-Fhk4sX#q?^tAXFcsvB`IP*m22sTBM=>;ixrl*K
z;{RRt4j8*}Mdc)|HL`w}{Yr~qcuzNs71hM>CAyWVw5ICVX`ht^XKly9a9INt1qC9w
zJUFF27~bwBolf-csOfzzKMu=^Ie&4m)10`pIAwL$9mG(^=({%Yer2-}hfDS1JfXyF
z3O*(`t{tzcl^c8JGEJ@xztGKLFJ5$iz-2(xJCLDfh!*Qvb{B|mtD6Kw(?5^TmoHY7
z*hkUIp_)d)wg7h|tH5}je4NU(No*!2!#oCC08GLiRBeK07SO*U2Eqkr+J)4a`cf9E
z=%6E;C~v-@a#U=}{OM+q;Mu2PA1&@Io4}U=8PmX<Z@kKWw>Oahf7N6E@*gne=9s84
zsPfVNq?}B@t#L7RL*W;yW+mX%k~xvj@@%!-9l2M_3k?ZTwvu-f#vZY!r=2Tpa1u}k
z%rq^=CY>cNJ|K+pa?fR(eVvP4;mm|07jn%qCLt7n!YfmZ<c(@BC?sy#_CvpBL4^Uo
zx|+jaW##PV*yyODn8^dc9l^!{oXLonFo6114l$%H`(M$sR2jzOu=YtVak^5b+?$zR
z6<P~ToAVy8W(#*%yX#caeAvkutY0<dVdlOlHg7WZmfJ6zbK<A3F*P->0<OU`yjER4
zbKWC0!Pj3ZDwB2=b+sUT^5v+i=+NuzkjI>!?e{PLYJ`M7{|nkEl~s(cb6yxKVs>lg
zpk@6BFH<?eX_Mc=dx!#=kZS0GP8JdSy7ia__+Fb|43xX|-Vaxk+!TO?49i0`9MiJh
zy`Iv{p0{f3oyP>$l}&qoWvc80IwW+JD!^>-=;Fdp?VKqRn&uw=z$9Hp&q7F^C%W;g
zrrR>u>mRIN_#3>sl4BZVqx03eAn}4ToFdz7n_RY_lEbc)Y@MkKYj8EbMj(*t^jZEA
z+AhaUyMZD3{zb%)s2IUuyNmK-en0}W7f{eHA3=NKQxMDv=iSdm`0OnPhdDALoP=9R
z?<=}o_12@y=WhWOd@30w-o&xgM3I=Ao`@`hD}`rmdG%}W@$1dGGg}hHNy}o$ds*2(
zn{+qC1M<z{aevm|{n(SwfY2RAlx(Av+J`#p@~Xz;VCB()n57lfqCuow0PRRIEX6d3
z0H#@Puq+n-yS!9$CcK#>f`<C;FRFHm*Uv=2VHL^_CUD3};V_9xvdfxjCz22lZ|x%?
zlQcOlQH3&S15ERy^we$nQvb+-eOd23Y7%+7R`Dmlr4wkE;?n^nvDH0t-ev&2;^7@M
z@Hplfd<|%~wi<}Z)%Th3a?ik<?c6&{X$?_~D*H%w(LBz>uce)-PAw?!zMb-LTsc;u
zI9s_}n*Anyrx}WJ$Ztb`C${hmAnte-4cJ`nHal7Xg&NW>zkIZ4DU``xPVUsKTx5#2
z^me`i#%_P&cV&lsfjaoTGTkx1$8S{6OsqW9x-#9!v@1|vOMvEGcVUI$T2F^{)3T-x
zyBrapjacC#il&&Bm=jCa81}XXVf6EY#rWvveN=VFq*rG{E8~e~D^?Y{w0D9JU;%he
z06fekD^A1Qs@W`~9?G+6P6DjJP+{V3r|#Z*Fi@OkQY)8F`x+}fQkyj!Y}wIm(E|Ob
zhv}d09^^gX;opC~OE1{(CEUA|e&`fa`l-Vp_0wC%a$7-Lg_MYPwx36&w(P1}JbB|d
z(f|RQYmh1xk>@y|>*ggSx@NuHOWF|=x|*=-F|rhn0suI21~BcjYmMhK2l#Y!JUl)7
z$1O+Zcf*bcH*!mzks!>vzu2M9<L4t0CicKfy?qfVl}vGYblu|g3%QJ^@}qHQQ2#Hg
z?(M6cG;Ky+EK)1spSE_S<dbG|9C1PA59`l^am?pfSr$?A_yF1~bGZmEkLDIFdMf&)
z!L(QxL?Z+5NQGOW9R^Y4oJK31Qb^`15ZgE}nXDBg`sKK0*12o<@{KW3VyHQL{)j9+
z7QM5-Ap=<i`Q39XQ;~w!c*FjCzn3jfph>h(w{i-1>weGu<<ov=joEDDP8$m8n&`ii
z73KcPH95=mKYQ!9R`VGr(Vtz{)lWGvk4_i*I(@TW6wF=GzsdL}%AC>ZW^+!fpey`L
zQcAUctFt;ahgog!YLPLa_>6GXx$tt8y^fuw5T#Bc_c!^B%#|44OaLr}AY|V6F80Gh
zQNQ?zdneZP-_2RsabxTgl%Q^e!quiBtd999_=RNXAM-!PaNZ`yt~Cjdl4RkVcdCfd
zVj{%Pi4phGNe2-lPHHK8jmHdj^Fc)L%mjA0-0`5J_8W1p!6aMpVO2!E5?<v;*+HVl
zInYmOB!D1g1-5pzzl6jc4E-uSY19Psk>JnD-Om&$m?aN>?QAr4t7{z8Xc5;8X;_+>
z<AVtA$usWwze(X}$#bvLa{}!@A31wzo%PM?58uh$TuGW^EU^5F$*3{a)FI3uCVo1s
z2^rr;QI}EVuR-r=^-FBw&c8TVdZ#;&iMN0<==FOgG?v76LmvNOt?H_%*uIuO0JFx-
zvCE)YWD8#|vl~E>{c;@hXxsKRd;NVqPr3<K$g*@2N;2=-?nql+NFc<cr>Oe*vjTHs
zo}4Obg=aPn>@4`%tuIjAs-fEd`6#8uwT5Yol$j8Gj_(;9JbpbtBN98g1*T+m#?ChU
z>RV5_PXhHkh)?N*A72F}Aq{f#G6%(h_aL@s$szyXwHyHu*Q9{itI@L0P-!N2jyjVv
z73Va+H+69Wgum3j?|eG<u(Xh+MLb&x<dee}uM50_mDieSe38u)%wv4us*y`T!3T#4
zfzTNymkTXMj4x*(H$cj@fk|Q>lUrn#;0Haz)RBnPRA4A?vK(K`l&Qi1Y%3tl{`uQ*
zc?>GHJZn86lO310e-jcFF5?#7hX5!|k08(4pINpILNJ{Hhc=6*>ZM~uMN0)ud9GRV
zyuYojg8K({dkCnd(vIUc4KgFy&7lG{Av-xCcP&pB89g6tdD!rZ!W!kMrUQ*gYQ-Hy
z>N(|y?gEIWdNB4~?gXrFqQ0#7iODj^`1UZh|1>93^a@kbEWH#e57gX<kS%B0RqpUN
z-emF>;dnps5uhL%&+@P3!S72SD@3I*YckkjhBf1hmtI{HxTDH_JloZi*dCc>1TM|8
z+2&+X4VY>V!wdQ|V+bAKS-6vyP0FL*6+tOqIuGdx#7Yh{m*QWPcv^^<LD!Mg&r@9z
zUGdtMxDQ_IZBoMFSc}sQC7XXg2%qM{9Qq&jKhha|5YO6J#9e)BdDzOyiLv=x-TM}@
zKyh%NX^S;^aFaELGC>m79)vS*q7(P9xzI}0L$kDUYQ?y>GJXI2nX*1%MKx#h_X3JA
z&i>Oa0@1r#rU!p=DbZ<k4oz5Ls5+<SR%IcdxZucRgHpszFnBOCX><fhHAh{NIrp|r
zhvl3>8?}z#Ub{7p??ZT_bsikObOl~47nYUiU5wIL^^4<Qp6J^JXJ|Uu{>M>_$>FLu
z9v~>cFWp<aTBdTXJdOdiawzvsow`iEUY7j6Mqzk2<p~R%n2(M1bb@lTdobLm*>f=}
z(m_ZoDy*9flS{KxqZsYJZ$$VSoo_Q^ZQi*=K7Pi^L4S>%{wdZX)W{SP5&&@3C-h}>
z=6Rb?Z@g=(d)v|Ek!YriQqZEY!*CShoQLtIXw1~I)da=!ISbV5ni<}60}Aep49D)r
zRV2MrSW-C~YK~T&er0QGt7Vg^r$=CfriIig_5eJ4E+e>Q9H{+*ZZF;Gm;sV!TAF{l
z`7IWS_l>_{`AV%i;OI^X$GW9ch1?}nENkunu|K1Lh^WKBf)wW2*`~pWPgsHMcWFJV
z|7R|_l++eeRR@^WDWPg_@%X;OFaW!A$PK^k@Q^of(dTgeQ9+SNDQd-^!<C~7_<=@B
z&)U98S7cUHOm&#p^2aFhLYt3CZhSg0KWe-MWQW_cT9o;@G~c%tLLO{IgKy&WYD`l<
z33VJZ>==5m5u>N@maXt90KdNa4=85G=!SLw`C%?Ep=gkgk|~IT!_4Ae!uU<(A)A%2
ziB-Im;0P3mAiBu`=+~IET>3Og6B*Tg@%!(ES3V=NCzF6iKEK2Fs4iIORT64dYxlmv
zUyd&N(1sAsc?NU0D+{B_TbQt*n0XM6HoW%`F+6ljk<+LLh=c+?Ka-W}7dsI;KDGG`
zVXwa$9cj9G`+xOJk=?i3<+RyWPNy##Zr0Xk6cQKPy?o!jad-De`HCh_dhSXHlr6id
zJpO%GAvPF&ha1a_Bs{bZ`BSTF<UZzD_3@)(4E68lx@G;;c^QG}XQEJwAdfx$SYZ`w
zQL${pceR%<zuu;f{0Voq`XS4=VZ>#aoG3}tBD`h~6+b!?qp_CAl}0?)>GTSul3J$G
zAZ{}5(zX1XaqEATq=tRs)mlIf=ORVjuFNACoPXy_a&<q3xqr)VPLb<gIO+1_dB&4S
zs!v|xf8+Tt93z!ZjuAC}hD2*ynLhvtrO)xG31^LP(7&DaEA1Fw>E%m4xak|6!~T{Z
zV$WoCrsNz1Q#s@&1QN-j#*e0nKy9UxF-Uh2DGb+MU$N*Spsk~rzDqERI=eJi?px2k
zhKKhh@B9;L7ubbP{95J~#U2)^N9qLHWW{yncU%Tby|HD)PEP!eG7q_|Kemb24%^=4
zvP}g3_84%L%*#iinLV`q2LcC^c)QmcTbk+u*e2T2W9MtD?`dR#3&6>_`0M!Xe{oIP
zZqwdY9>CeqYyw#cGaD&=^>Vims@J|;vQ50-8Y2cyv{Hp>7AqhA{>7AvwffaGV28Ca
zpD2(9O;YL;PMgtRsE4}z?~sQ5LQ15}<;qeC!85=9Q;cGR;y5cD_J<qU<FSUz(Z`y_
zM6sZYC~L|YnJD{u<Q;D0@OTprpaL@O35@RhsAqH(K-q{dUap^B3V1^EiGCPI@^Uj9
zGAAUeF<+p9?X%PvDwiXH62g!g*Ur%^1W^Ant({s>TbT!&jg50Z_jC;KDS_Galz~Wx
zPK2rkKpC67%RKRRw4Szr09gN3dIH%dfIzQdDXyOF=dbMRfN-1?2pV81n&#yb5D)-3
zxDnS5151Z*sqa<=%z%#2fX&qn0%*o%s+Is#J`hVVYp<3fl*_3-o)DOQaq@&hj-9z9
zBxhQP-><Tt=B#W2QQ&PDiTL%)(<g^eIH$0_Qoub1x_4vealK5Xeidoc9E3UjLK++5
zE;d}4P8@!T&9_hf?FW1Gp9aP^O8yf3RcW_h(PfvyS)$bfaordZyESDMXW7mzUWu$A
z2c=~!HU5u-mZYjW>3=#If>w3Sy5*I=3b6lPT0#KW4}YzNWy(>R6sFAaT$|m3=qgZq
z`36bSgjF;nRebHh*P1;glh*_SbYAxQh@s7K;orA9ucSyqFydGJf>#PBtuD=RT>esz
zbyLt)+^Emm!{Sh*N6IAaRrB{%!1zbt59u-*3kZp8jR7Rzyn@Ip*YDPuhPSQq`CgVN
z|J+a<-Ocwt-ab>pRVQl3Q{ec0(|_1rT8etZ8$kLpodMg2o}fRMm~unLFkPo$dDPP7
zVyY7wU>dvM-b(3{KNj@#JB6BDFeY>|Ar@8d)GyRZ_O#v@<mLg;s4~6|ldqc02DA!&
zrUX~iYRspO#LCI=MT=i<q-nW99kil7n7mw|0YnRNL?s|vE1>J-4Q#X>m?OD+dHp><
zuPeu%U@U8C9zcE-uA)}%Le})@Q!ha0sNQHr33=~@kFxCFGnhbyB59yQ4_MT;3SMJh
zBPCuNdUk%*1P`7+&FPJChFu2?NIc!NJk2(Q91-gHIZB_tA*vBGwd(F^hGu+LbU|?r
z8OJb1fwnB11wBy*;?*iQ?u^Xxy!?Cbzz<`8eQ-6qyxB#wWng-QqbKT4OU5k1S1T+t
z=U2V|*B~$Y_%E*VOT}6e9jY9coT8ML^U^U32S>I3iW5%h25IS;T5KOnwyyij|D9cW
zy@SFf8uI-=(ZZ`!u_EkXIDXDpNt&`WlL=vBDZNEwG4`Wz85z>uxXHnqH!^7_a3#JI
z`{W5iEpczhlW^Afv>&*MHrWcyoZW+ZJ={kQKSsI{S&-JtYNE<#mjU(+6P=>(*4$ZM
zEN<i~F{pz=qhJPB5Ap=IaUpTx;-5Te5N)0ChNN<OZ<FLwmLE`~<_>m{O!lsmuVY1p
zjd+2(cTg{Y=^PpwDz%q}@*10%@#KsH0akJ_jnF@W6&0ITJ8{<ma?pSJYsTk0hxi<8
zJr*q{mv<vt)f^Lm13=z3kx9>pl3dKXgU$c?g%G6JR%H*kDoY6Hl*}vT6r&uC>PI5R
z1Nxn~i@4bRlePHLqp&CZUT-K2iZjPG)zr&(zDk3_39uZ`gtoRsks{w_rNc_w>uk(H
ze^0r4)-Q_UhL#Sz)_K0*zDs+v9Y|oVpiDUTghun-7ovm4Oo56lgWh+CC#5un#z~7~
z<jC1qEC&k`D3j&IGeT$-<QFfF=N4^*H!SVb0}~U`<o1*i{*AG^*rK7$E7lC>V;==&
z>)w;-E?MSc!#l*4Ibg)sS#u{55SO}(a5QWfn+3$xq6`!!IMGE)`^-*r-Q%WW9-2Mf
zV1B+%w2X<T=(>(HifB5emd)!gtif8+n{wmYn%{?N8Tw*)^`HhwNnsr+C1Chd`Dv>|
z#jzmqU*^<9e-?2c!MQ*MCO?GwFQyjRxU>>SHu`|ruu-*Jn(SUIZdDupK;MZAnnoGP
zlOW&TYe233p08CGOrFh2HSly|Qxj;GTl@DaF_EuXt240@^^$9ari)8MNPy%Lg`Wx(
zbXaL~bKa|=imSV&=x=i8x~o4ccy1l+{}C#vY=zn5c^H3?56%@RiX_lgg=*^3LsuhB
z3}&*^54}c#qbX)2f^J-z5B+_n0o2l2Ih*A|nkS2H*wB(q>a#7UGt4nLY9>{@EcMWs
zuyf;T(;k`=c_x%AlBva7*u;K1vChqm7UB8ypckT)uij3;0WkpT#dt}1VkQAkI*i@`
z(!mNU62_p}Djg~vr%)v^|MGl-)mRa9BJ)a}M>G}m_kpYhAfdxlN1U=f1NirW#_5d}
zqRxHFWea9fiS<!0KHerBS}1;Bls>B>6at*>6{g8xX79Wb-UC@p6<%9xKt;m4y?cv}
zclb7*t1iFHrmM28+puBvBy$`e1It);tJ}OrVDitm6zZ-I5n{CMD5Y}ytvY&~4vNNU
z7>_pCocFkwiJEG6Y|>Q7PLyW#e?*+~3wp|a`fTt<he5O(2Q^hZvIjH70Sh^2b5W+s
z4_M5xL;v(4h81Z=`=<_OF6Ve>!@X4Uxbq@5VH#DmICq83hFez5%cbZ^xt3N-LjrOt
z3YxH++7*{9wQ_R{|Aw8h_dPo!m&7Y<x`D6#w~g_DK3DKdCl5oa!M@-+KjCEw&~Wkr
zvdh>pbe~Tn;V9T`-3A&9eC0z!i(ni6^IIUxK$@v%5em_Jm$icr`IUYQNh#+hVDil^
zF!Mp>ytF)lC6dY<vmAjfy9{cV+5EU~%jO!Dejm%LW072JbS>sgZgyfpy9*{2SU)-c
zI6M$e-8>6$4~kHy+`tLSm3{q;7)KH)Yx3kN{<*_VJoE>sPxhadf-1G#jJ+Fq)jA`{
zvJB&WXc!JnR32iyG>o9ggYDy%i<@(p;~RJ%=UX;aSacvp*cpiwMi?#fb~MVT>dU&e
za0hS8qm5gIRFQ<YH)(0&@E7Htr`XNV&roz1leLOo8M#7I3GC;LN|@@gCQI54(sCJw
z3mhUfzXlxYv^-WKJ0NbFxN9RT4J{4%F;MV%ch(L2&)U_HaLc}#Fll1{V^Ijjv+$8C
z_Rhmd{FMMlvCQX_sUQ@V`02%ekIie4fCSmE7V4i>+IzwV{Ia5_Z5M(BI&Dgy^Y-R3
z)hI!qH63C`*wV&QSM!@8QzYTPSWXwROF<DRGR<;GYjEP>l8DkDmXg-!w8LYr>}wSe
z#k%St3+>Ut=LZ^+aIW3t8819Qee-E4s${8A80H4R0?c6i`i=KY-#Tl}41%3w3sEtI
zxXNHk&Maj(O55y=#XPIaNvgdhv)AJ7NOGj&*;PhVq9)`)Fmt_=*4B3FNe@b~fWO;A
zSwps1oRPbh1Lq^C{-%Cadust5eEZ>&Xq%m}ktD|zo66b`t)@60QV7Qf5ew?hK>PS|
z;u-?rnA3QbHCX=7APao?TDH=?e^mb`x_Ke&Tjq&uy{S%Tu>XXUaLF)FjJ76$R@`dJ
zj|ev3zX^EV4@(IAzW{_yo#488l>n|k1A{@$|F*wH7atPOo#(-WP<%I5)FTt{3^XqN
z<Vj|q@{L=RVdv|V%I6>0QQd9bZ66e*3|tf$-uh?PuN}XkZWI+1kzwv_UIB(f;OZ=O
z2J2{Wm2xwTLO=k$!b66IK2<hJsX%hkB3VgGk6mYpntl<Ia+J2cgsQ}HRNu#&K=)BJ
z_%6c4KvR#f;Se{q*u16ce)BN{XbRH`g6;taRM61;F5lQd44q=;xDC)w=8FeSblkcw
z{qrfcAHDYYT6cDIO_mO$P)x5k<Y2}V6v{e(q1^;x3V;eYZ5Rcq7*hmnUE~zwSt^@j
z@2P4H*K3W}sLQPRa~`>T)8`#xaooIk%a}YkINk%m>r?-it&Bxb?WsO`^3S{XF0%+!
zO9#UHH$hb6;ht>teK&<CNzLz-=4sK6DW-2D1lqav3+2pd5r{b8xb~IdF&DDv6oM%w
znaj8!TMg;lF>82!Z#pJHA|I%`7?wmF$#QkxHI#&?@R_Tm=!fu&J0d<j*OK38iSdG?
zmT^S??GHR!(J`%UkCfbdK#=GkGks)3IzbV%`iyt{LJLT=wBMx}yT7zi#RjXDFffK8
z705QDFa`(84Z9M*eFcG!B?fvSb=P-JF0RwQZ~><kAAfrX)v9-{>yuETiCLccgO2_j
zbedWme0cm6_$u=;WedXPZY<89xi>{WK(UTJWEnC>g?ANMc4_&E!Hi?hj2eQC9F>&4
z^bb^qFYO_mcEwi{&!=4l@6)<+f9eiG@iy^&akI}xqAsn4%xVzQE_Q|W2YgL&6UWZS
zF~zlH@5_8yDn+E>g49^Op|DfkmNHQ=?=h{Bu^)@Op9W%+@KbnY91En8U=6rlXJQjg
z*Pr03;Ay;jz#_zq(V%<zrW2FGxH{gZ_az6g5cyY~_e{+U+Hj<3-x26lsoW0jtB@)k
z+YD9;%l4*A767)B<j6`5JZ!PD#R(#EVJV$$-B^t^PyJ*r_aJ5pL&#pYKM51IW%@hl
zDWWLt_u!&>+a=aW6jJ<uB!R>;)$3beTckFb0LT^>b!QuqBcxcuHKZPH*$~VgSj*TJ
z0L?|HBv;-J`M4&f_!WYoz{^vog5TXirrC2e@Z8=lyl&A8=yl-W?$_w0)9dLiu<4N<
zwbLulNP-5K48857$&`|U6IMfS#ig0C*1D<Ia0=y6ZC?!^onKqHwyE%}UYDEs#^&Et
z(L>&*5#^e7$}s}2j9fNV1w7SNZ959?6olw6U^e?@_TB2@DrHi5P1bCWs7yl8p+38w
zAfvs5j22n_U%hy{9Avh?m+v4p@tUOjh^kh@Gvu~N^VQ%g%+y1IPlOVIw=qNNl$Sf_
zJb?THkZ?G(6jhUsI?EAx%;uRH*Rnb;L(Fq!enbp7jL(g?8gMMuTD8sn-4mMP%u(0p
z5EzzC-N#=Z)F4;)zdC?73p{^AW3<*460RAx>!CQy%T#6u7`J&;@zvB4><yLMR?h(l
z1x3j9!mH2#+tB9t#m5qQ<o<OF;eEHgn_R=~u(#BK;KZq?DN>T^ug?s~h%|WqVS^lA
zvo%JbFKSbGsp|ojMY2EkNqJ{*Mg2IblR0!lr$Js*Xw11hb@(mcrjocuGF{$r2QNIj
z2|*u|9ga<C2fM#v+(0afejub=z3}?prwnZ0jt8B0i~V>)$T@gyTDM)mJ%AUB&JJ^V
zda%7k6r%fy$&*3%)0eG41&75!CQvEu+4g!FH8#49_ABVx%C+3+LB^2$F<fHBdt#$S
zX7vDx-pt3I<e!HgX)Z`xyfNnnspNS*Aa%Pueq5tq1_;9VrF7%}^_%}TGF=aP9O_V+
z{SmzBDuBUGQ<L)Z{H2W=t-%C)UIv<Y9i5^yIz=zHyA~iPY?i-QX+tU%FNIgtjpNuL
zp&#8|kZN_U43Ar%)h>th9r72bVchvbbxw?ljwXb1iakTQc{kxOaCAD^cOg344tM@F
zT;UWXp3_{$6$vq{BqW)QI3n&&1V)Rk>s}{>l<;3PQ=C#Bg?Ca>XGahnDL$D1oJ{}Q
z6EQA#w29++>l38-AZKTt@5ZJkXE`CyzJWJe=yTC@_(E#Jqs}9{kF0Jf#Af1FElWO<
zy!kx^`U~aK?(lFdJ=nrDYH==9NKyh%yu<WMjLt82*LRA)PoS>4ofAr>0m|uRDva{_
zC5!3KdP#zK9zC^$6N-pg3pY<^38~vclLg|%D{m{89wIvC6WCaP(*u#p=ev3Zb#0KR
z=BNA$hx2G${0ijiYN8e6Dmw>+*`VKZ;hi14eo+FBJ;_lRR%S5%xj6ERXObHl%7%j9
zzHS6utI?io7&cXtKmf>?L(|5EO|Mt5?8nmF+<ZUUBzqN@>YAbB(RhlivWpARztNbS
z97jqqzi$(^Ab2G-6|Dxq^T}Kw*<L})S*3SR7RytpbsYXa?%d{bnHn#eC&9U^gb6Lz
z{eVphPH%sK;OlMCuveHz*zS1P{5U2+=>)zML|7L_4~*a2T>`T*+QAiM03*5PeRtU~
z&qd%Cy#l?RVZm;QxO$cvcY@83sGr;5^);a|d=&vODOsH;k7kMqt9KC+5gEvgybql4
zG13$`TlHP{3_i~%9^le+tUQFle~#B(Q4_Q3?lN_mtn-{+*yblW3c)6O#2S}rK@M<F
zOt?tMu-T77f!CgquSYGCv}0SQiu(Ok)N^9^goj(IRbhC1gTu1%=G~X`XN;g#E2XgD
z0y$OZ{M{0kxA|PM>kPd`Vk9-Vte7ihO(LbJq$iI>lDAR(WN?mG{Y`@wpWv>s9+rEh
zW;eH7PXO@&#8IBsE!8lU`0YI%<9L07&Dtjyr|j%{wk};lsTApd!>0(s9~R2Z3S?kO
zp64chqt9I4;g{#IqVmU}b-5`-)XqmlS~S_>N46VC@SGDyO~d+!P=!1^i<JOX1_bs;
zKVav)e%cX?b^HFPK-l4^*c@^id;YwQ?@gJKXULGy)sv*TEMF)bs>?jPglk}0=pDm{
zIqgL1DX&x}_RWzT5I^o?m1BGu<`4<J?Q!gV_>GdQE><i(k4i7SYH5Y5h7_M$MA$v(
z*huS_G;bMK^J>UP_{qe*tJS7qNh958hIbb*+#Nc7NGV-YDiqsDQ`*MKWT0;5$WtP>
zwKF=+Asfpc(VygG`23Y)+2k9}QxTn^BRG{A$QY1;MGQ7qLRd#Dtn2h~v__^-bKUQT
zzHl%N!^zJ=jMt6N`?1%H%4ma<La(<wjjN{YM`v%CRbk&jR55*tlLDP3WKWP=^V%=Q
zi23<nVl4dojliYYE$Yg?n(Fo@VeWi6pT>ljwj6tv0;y$SCFYeMgyS`tN6Q+)y)^cu
zKoUgce4pkqK`amhYSy*6+M#TdYDFm;s8UP_CbYJn@DeVmCcY~zXD`%$-(qZ5!{p1p
zM^dl-HRVIFj8D_`0ji41n0rgWYcqo@;z98}=)=Z|#_`yu_aiKkk06=p222N?pACAo
ziB@F)WfSX-r=@3_NAYOOE|>l;qFBLG*LiHrcfD@aMX0CKqM^#~N|HNIC2O+lC$}mO
zT}pf9Qz;BsWC7^9WTj*Wvp+&$Ao#Mh(U~le26p1v1Wg7bbsQ(aVsEYe_vW8wlIc@c
zUB<TBX$y80Pyxrfo3+8j{CwGhss32?O8u%+TX!ejh==5Mt_*9hrOrh^c1@;gxf9qZ
z8~g@iY`(0<l@tKhFEh$dpr!E2zllysn@^Z`kaOe(1oV67v+t**QGRnNXgin;EAI(M
zQmERxQ{4H-PHW&k-E+Qfa`pQV*OH+DE7ZN>OeJmTtm03uVaW&w3amg5@qIfZ8H47u
z?<Ed&kh1){*p*Xxdg0z$qe(U7Y}^XI2*w2aaN12LOd9C3+p3ZotDGOEmho6Ba1aCb
z*6^-(D6kG&e|Pdo+`2j{4_65a?hec#!rR-Ws3XY`C9*d+9GW#b%9Efx5{JFM&WBVK
zxT)8H7<?@0?9^N!E@3vU{^DNZ)KEs;GekA@&ReY|0Sld+>NRP50kr$LZyup7{4^s5
zOIp3c{aGjSA4TWDUe^|c;n-@7#&(m&Mw2wQ)wpSFqp{Q2wrw}Id3s{o#@+WPob&9x
z_L`Y*-dA~)U%>Cf_stGkeQte}^6TTe5TSwI=P+ynE7wplxR&Zqb!S-x1gxxRub&#f
zg6F@<cODAI!u|bfgO|<O?fFF<Au;R*GE7Nj;<>$Eh9Q%EWz$MS6Gx*=!HE>R0|CB#
zzBo``rSUp^7je-VhRk|~nqTq7^yDZVCAY)I3P^@p{wi-b?gG^tRQVjE(~pWzUIJ38
z&EVVj0vT$AyjqGG=4Ipp11$W9(0;9hU<G82_U_YUWD~-pl-n{b<6xST&be<^!O4gS
z0A$CwO9vh~@+3$}_?i?#>hHxxIRFnmM@1(U*giajshU@c%g_6lY|X998sLl@+0R+|
zGRWB)9M)+@r!=rE%JH=Ma_*SZx-4ARY7Aw%POi&=e`3|o7eeijKg9Bd4lj&rv$BW<
zb#JLoXQ`5#|9#)^HO9{OA)QIkSm-0}Nx*`ge&rue4$(M|Hwzi5s75kC&*qdj0hV1j
zl=ZtN(9<M#PW0FO-!{<LFxcFM7f!)0U0Zcm{C-(KRZcQg;`2w!%88q+;;JQ2J`*1=
z+BC1HkW0_~L2wUY3%PN#qdNf_J5CPD&q_8YBaw&}=p;%h)WGm_clIj_%hamvaw&&d
zodFrO9IaDOZJ~~`MsoST_#*lwn@=LD9!|R1zaotP-rdDFUyuH*I7?pKP8U#|T?=eC
zj(2L(%VnQjT_y0yx2zzxQH}dG`t(FCon8;B6qxDxclCG_bdmwPgjC*6F7I=z0bT$O
zD%E7&pi6bmcr|-hSFvPy{@QG6M0Nc*3wq6EO*OM{Hs%g+$ev?(#2O!e>Ud1ffJjDf
z?-{vpwy)~5)0v1Wn>odo8O#g^z8LqcK6Nk1Uf5@n%;63em~E<0LOt~f3uZSTd1pP@
z%+k#q!lwu4<Sjf5>Xr^COE}Et?L;EbWC)@HGvA2<AW-0Tuzn8t#hJVN|K@@T;h^;Y
z)GPNO_kcs)pmGqW8*JYfx<~TKD<{E&@nt=xUX^xwom^1Q`$X5uCvusU_3`<H*8AZO
zD_ginG_+T`XtXINp{);kO4k#Vr-qjLd++<w>53fz?m5hRab#QLGJa1Yeo8%4FulkI
zA@U!Qe>VPJ9nAM2z9i0e<XSMNhAQ*KhFb<y@ovPL>`(;O1-U&b<n*rF55-g-R-|KX
zdHrR|TI8*0tLU(9SFMbl*F&LLw`VWmlXc$Wu-j@utY?q;w7>dMM|p>113#gZF=nuo
z9=wF+9@lqG$qOI;h}od^fiG)xA5gvep%+G$P)Uges(-|~pF*78{t%(2VE7%K>+HOU
zI1W5n1K-`UOn>vy{-8H9z;saFY#h87)L-6eBs}gz;~#DnG~qysWn}U&zz~xwy+ks@
z*5Rj`r60TcLeor#190~AJBrQ-*Q)o-yNb{F>L-*Yf9*%|nrX|N=}-!LWEV6ZT0KLm
zi5hQpqSq}|-j;3lRgi68k^~@%zlh>9Nbi_2-x`cJwB%>uiRG5cL;$_T)5H{3gE^^L
z?>T;16fO2{nhSl&$_dqVeuHlJsZdhyiLM6tiGz3uV}7XytJdyzV+RdcOz8vqvFj=y
zm1_O~4^B?wR+~-~TybY`c^BL54*Uo|B)?*$Y;aKl>`B78^5KIm6V8<eVdA$i%m9=B
zEtysBZ_hZIDa)PS9ughnN6qOckvK9v#dIQ1ncdJ5GzW<36T8mfn8--Rr3&UwH!%dG
zA8BYrzpN|8k(ocguKP`Kw>#5_oU^3N=r4D@ZGUHUe|CQC643c$23LU7#xf(Rx~n2;
z+_I{q&gqeyU_qd+J=a96Z`WpEx6(w+Fu!0XJp)&R0)VZKwWX4976t>@89iS6TTFHQ
z({0>Z_1H}TtTABG%EX&la+}29PcjQxt}%=^<GoppZ(iNflhTf$iGLsl$&Z^iY+u)x
zxW(8sz`hVvUuf5G-!3U9QlDhV*3Eg2FYd{hs-%nPGiz0GIcMP`N=flX4#sanX{Ko2
zHdBs64SknDu4+0wR*&c$`T3IFYy+GXY++0BI1w3(oIb1pU40e6tQqZjA(pa$C@>h_
zZERtKUGA~&^Rj7fk$VHou6HG6l86Bmsn;m9WnE{W&38Ek5floXGSX$2|C>x7_dNi*
zHOKE-`0{ns?&#S`!8}`W3U?PhmJV4QbXUo3#RcMAe{io+GIeLYBOZPs{H+`I!?cjL
z(2&MIVa%^h+BH49KGKa{aff%Vff$%w7i5qVyNw*bi5j2M9FQ2k?dta;_?LhnZg+F0
zN`7HpyO?7~?!_m}e+9a5Wyk3mcaypo<gh`p^C1Iwhjc6%Q3GxuOn)!X!@uJc=yNZN
zTbIan!WH?%?*!m<ad>-YkjQ#uooY+Ydo|(Ak<mIw73sRo^YF~Mju5`%3OdBP`$#KS
z@6Sq9N@>)8C_8**KfQ+U?jmS~1-7o1dy4gIYP)_eqzNSWxTs}}uoH~eIo;>wxy1;5
zhpBdfx=;5Wwr#<&X&a1h#|`{@pM*g^^uv)}6F-2?xfS2=w34`!yW$QNsJM&4v-iX-
zFhIyzn}pQC+54Md4PAxIueo%Tq`i3Mn!?*uh?RvCDyqB|5`O%(9f4zUfN`yru3j{)
zn&V{LSthbUv}QgG&{nLz%IK=9;gI^h0wOEHetjO-4%b8Bp>M2+g#1)~sA1aYsBxtm
zw%h<~!`{WE&$^k6CsLDQ;QjT9@%`{E$I;10e&ZoC$?Vj?1YvFEAfFb{f{sZu8LydT
zN-fw=fPJpPLr=iVX112MQi~p@3uVMF;M3V@jBviFa+Q07M(Ua9c<O4UQDMcxf&`4t
zTHCa-UIR8^o}M>X&Ilc0gY<sxrpA>vpqxM-9HanU?918V2RbCdw6G1d`g3DcO#Cc%
zUprIGxr;z)bH34yJu#V_Xyi^lcRxrgFY^Y12JL7_1RFVfjD6`}fig{~R?MKu<U3Dh
z8wU@0?a5SD*|Ge5Oa7UV&Pwgf?08uDpMoyxMp4_Z+pzm;4n98eNJ^ZNc{fq;57Thu
z7FNtMC!$i8UF-H9vw4Ym)w1Q~f0GFGTYz?yE1niuk6o9kt&Qg(TmuO7E-s>akQa?5
z0zqN}F{Q?6v#w?>xBgJlu4zd(Q^}pbhxLg(RHJCBW2#BOI@r>Q3ydp{&HQ%P&g-3(
zBjrT2oIwPj*RXEU-H(J)*8F#ZpMc2=b=ry0e|AS(W!KQa5RgM}FDqQ18{5mm%%$p2
zD;27h!tU5go{Lx?kJ78?{3)8t9X%t1v*&%W2B5#b_)UD=vwmnk+eT~D3I@zorK+RS
zSUFU0Z68NWo=4=R>wKYG`>A<Yw{L>MMftn->S+kp$Mi;0X|t>30)!8r3`RVSciXa`
zkQ+YDU3Okh?JyB2YJL^(FcRJFQlDVbU^sGz6-!M0^aQ6Qx~0<Lj==KE8#cCoN354|
zmOu@t>8MLAXfw03Y6bYnr)8Tnm}$qc0`5%x&ji!h)Lcf5+kR*)W2ccvM#-78&(5e(
z=6uq%y#5M~-?Z`!=VSSjkP;C@2}7JS{mtQQfdBalDg2DMy8R%!wSiNVz?+XH>M9L!
zlWsu^pFQbodTn06CcM+><pnvi7mU0(W?Z^|+ZE~O`(%F4W&06?B@t=9lMyXNk<oRc
zRQwnU<-}CmS+fMRVhPD#O<Sw;c))XJ@9X>f4ygpW^XRf~wj-HBHI;|UXC{EMacDOr
zmviin$LeaS{)|s;8PTa3RCmI!k%6kN&XzN&HG!9JV?otMPQByKx<xz{nbC6LwbmSm
zgxZXkVxe3sVBqNO3IqTPDgjJH*tve<z(2GkWBXiUx^QAQ(Xm7uO=kTaGKBuaaF_Ub
z_m~<olnYO6))rCx&UGEvv*bxz>N^KyaT{|$67Z+zF;e(ox#yiTw#RRx`e+;fP9+D7
ze_UT2OIRzdBydNM_qljv)rG%$d07BOkUnAM;g1bxt4ITQ?v!x<PA?4ZYrCmx4|0UE
z?pFc{$}yZ>;ny?ePkNZ2MMr&(FnTT3lzx8fpf0;XyOym!@Y!PGEe6fVKG6$MRHfyO
zDvADUlJVfP^0?m)D0Z^e;7`vIhgHmt9LDz^2_{6{40!w}b+y&&3bJ?bo%C!nVatib
zdm?MF_i`kv%D4q@q9N=E_4lA@wFwT{Tr|E6qr6a!3+WUV_zC;gEcD&slod{RbCGJ5
z+gh}>wdGX<_%a3Uk=(0w;M=@Xig3zefE7SHT`vknJ|D5O7f-=Daud&9kxIPS2##rh
zkA2(<wOc)3S{Ro#zq3y5wQz~nNUtx4l)y5%1|;-LcbVDiDfu&D8b6ksgEr}}<NumA
zdPI^-!AqoBaxJHr^39xU+*WNJn~P^Ja#X}Lw=^m=tiwPFp16d%QP>rzk<AnC5u=~y
zIJ!CpIZ+h9%nn~PfNkCHj&TtG1diBRFA?*7H?+)%IXjOS-tZf~13Zc@QB(J{Cy!!q
zk&7LzGQ-{R5Th%0l3((Y!lsW$_<l+@EBr&LA3<l+ISQ7#XQV4c`GaoB8HQ}DLZ1UO
z^$X$q_UO|d&d2k{E0OP8hT&`@>X51Pl4H~FgSH}_%EQrKZfo0S9?~yKG%FD9o>sFu
zMvbvso<{DH$|cj6<-HXs_NALjd5ME`PBXe?d`Fx*q9C`^W}>lr$xp_4*BBr=D9wp1
z$wi)S8iApp;5nJ$GTTByG=j{smlt%{OPB$<KJ=ZQ&;!-_$V2Myq)3h}LIbgs_S|WC
z>FSx_p5eC_bS)C%;Z66*GiCq;bmke|5X1Bp+e})pB6a##1VzF3>T0zK!8*auUtEn^
zU#=7-iHu5j<8bjMTdopy;<ZRaCOUtsw)U@g*_=jI@r)r8uTxS8)WD%xf2lS5?RCQi
zQ>oZRYUJp^Nx%*NfPcU5I9P=&-?w9LjsH%#6dc$Ci!ki=DTDnDEvU6ss$WvJUd#@-
z{`5DLOnY3t;#4fASD(~r40Lzh;TPG9Ld^7HGDK_g#e5AfL*{EP(RftY4a+>M(_pt;
zUQ^mt`S4J#0bs^K0rU$>vS4i@iNOTA+J(3&c5!vwg&TtL9qPNz;!>G<ieXca1@q7P
zb$@=1N#=$PGlm!GR&iN#GyC94cqjQk^_51+A43w;oZS284dS<;wXe_J5s!&%xCOHR
zO%Z-kL74(41Kfq3)1Z|rlB7=w(kS}Nk~`Fc!*HI<0O7LLK!5%Y)x#j6KIEmjPn)1+
zvU=-yJN`lVQ3L??D;2QO#bO;ea(Z-z&k9U%jv6R9?P9#3J{DR%G{@ETbY(Cu{c?yz
zJ7+}Z!&5rI&PJ=Lg7M5}BbV|BM1l+MGdZ%^9Gyq5uJ%kXE;u2Z7nGzI`w}1*7*<}a
z(?CX@sN|Sk?Vx99XGzEyi}BT~k*znZJK@;Rm~sxPTk(jtmnUL*{SI20H+$6E9*>qZ
z$Ro{sLQcQ79yf}Z84@)CK;Tf^f~*B^%6q$g<w0C(W)k|-m+N&Pn^~RtwwUFE8sk00
zPijBcn5R6aVv;Au<MoK5#^}ZgJd{-@Q*v_+vK@VVMkgnm76>@fyU%LnrQ&ch*E?b(
zo!goOJ6t1&^KovwimdqQdH0$0m+N%EGpEfzk?aYwYycmgNrS#gM-2P;&$SramIBn6
z8Aew|DLd>g^naCJ<>SB#z+6*y`_yP-AOG;=V(<v0dn{rIX}6!u#zQ3+qbErAOD(48
zPQg?B_>EP#lnSJY0IQ+U&QStw=SQESDb%c`vfNVqD-3Hjw1<uAro&_UU_FvSK6@jV
zgOTDW^W5k+FNS&}EPJxF0Jg7k^lF#*W<$Z>hFd>#PME?F*RD6L+jv}`(gTurRKC!B
z_EXoty>}$lgzm0QW}w^MJn^<<ZA+!4&36Ae79?TDdbvsqD}AQhDg;ri&Y?K+`X#h8
zXud=qrfk$~jq)<BqZ1*Tg<Tq=$|(->*PzV|sJec9bz(dVQmKxJzS(QE+Zo?^k24mo
z1N7m4)Y(=@-rK|QWIpb_A~5%CWZn>}fVWhODMi5ZLU8?M6FNDvfE>4sfw4IQ)uy4J
zKX#N#l6{;ViX=dsUXUsG*WZ<*%4au^YKNsrhgsDx1xX`U3BBW&GuNxSOG*62qAF3U
z)?r3l^Pf5u>sG-B|F}!<QO@Z9WKc}&(t<XGQ9eKy;&xui{E%k|7x$1??n$FpVk+*W
zU()(?{?}$4F9JE=b1#hw)G3cfE^9abC2pd`Uun}EH0}bB;&UEYeJ2BtKRHU}hU{qA
z8hw_HC=V?`a2Hggxg$>H`2(i(?@PmSX2i=kt*#7h8s;c>-cyz7Y=$g6LcIHjw*VQ_
zr2d<bno^vQ?!Zi@veT4KnbJ-i5@(HpVg35);S0j)x$(~4-oV(6iC5^z8Z~Qp6Ot8w
zBhHUy7vd=$p_iu=XMKS5v~_$O53a^3<`@{@X03COvFvnOnL*!x<^Xz(i>Nmlvdxpp
zW;98;0ZuF3!_CCiZjJKe6<}E*_m9B7?wNUum9Iz}jlo?ts(3OIh;Iu(wPA_Y>}GNr
zU#;$qgm{?5r1puA6RMsgs4%bS?h68D!XgK%&q3Xi>LGhO4c1te0!@EdFvOZEe!_gS
z;QxD_N3}~s?ZqlB%10pFr9&DMfsHB)qCNR@$3YSWCbu@Dk{5aIQO5{mKOXNRD3Z(@
zY>CmMBZBFc1803H^}$|e)Fqb{8!7dvm9yqb4d60PgdNe0Q(@6^Qd$wOEqHKvsHHO1
zgdmVEr<yD=VaiZ;lD>ES9nL$hAZh=45mv@BCgws}{p)r8z*;u2=DvJ(XcOudv)S3$
zmh#*9B^zwpf4m<OE3|fb`h=dUXQV$ouE|Pw9r7elyRN0dNN;JT-wP*Y|HEf<dpqLg
zVH&}}>AjsglVED9Fxh;WyK<SxA!WPFqUbx7hvIyrsYi@Tx~GPb*ZAEesy^w=*sKdb
z!K-IJEQfYsLNjI*!NT0&!=CN^#?9g1%y#G!+tt47^v~R!By+`aWWR9ac_qquddcLD
zyHXe^LCio=0pv}uX#HMGI^Cb54yJQ}`-Ov#;yesX&C)6Mg7#+X61GQ_-;GKe#cxtj
zN8gNtLMq}Y2NuRr7OA8R0#=F({;=5?3cu+lTrXVfC@<RZ;NkJQVr}Y>G^&U_dj)ZO
z_XS~S&ooz4=KKbE6q?+9HK}X}ecvsn?y$&fvdie`DgmkAHNEi19_iKWaE{9IR}mop
zNOhN}Th>s4jg=B}u>YKX<{I5}U45;$Z5QfsXY>7J39HD1tzOmvjkKd&3?KXHN9uu`
z@%-yLf2yrrI`UqFDcVJYmtBk^32OS=O-br7A~-H}JP7@MrC&O8SD*Ph=524d<<1_4
z4mzxk>IQLT>gM$c*`U*XG@(SXtxdZ?KEx>o7$)^RXY_1eUn7otOeyv}{&}GtTxTzB
z^cM)<kPXQRz!(gRCZA@1t}a=u3Y&sg+_*b+<!8wvN>#yw8dw{YJi&)kiu6`iSDpT4
z-jKEcZOEccB3)^aoHjCbgik_Mx0J>uzHz587N~u4RF!XI)Nu58TyfFby*A?_C^|Tk
z8t85oB1Erjw;7s=Q2@OpAvVCUR!s?^Zdup76FMCGbV{3ZJ*epScl78J=q&xB$7{m1
zqu_#R3C<Iwk_amir@A|=r<Z-PRk40=&BbV24SaA4cUsTGw|c;*g{VH8R}ZmIuaT@;
zB+R3ca$;O7UWvLN?M|St^(x=8BgPOzR~vi0yexjC(CuN&vgXAi=iyrES{mJML*P98
z*pA$R+01%egvQ*hF2X$&^2`b7IAn?3tIa&os&chn%7BDGRpWla?PS7aC!oPRtM%Qp
zH?&K&o}&<c^Ll>yVP~sc>=jRl_%GV>5o1K21Xb4~kG@$#q@v&ggz%1JtD&i92;S`4
zGjGD|TR-9Ks48H90!I0()(Ue%?2=!HF!8OKKGc?x$MU0AgVL}2Vb{guG}Rh-)ditp
z8(gK9IU9Y%H(Y<Bn5hp;Fe*bWBP~_n_-r_Q%))a_hOAZ^cNtu5FFLtSqS^e)7Vwz5
zn}jpb@q||=znHwsJ{f0}4nq8W(F3Y&KT2gSlB^!6?lX>)h-51{(P>=L0?pqMs|y{h
zcL5T2J-wPMo!L9o8UtaevF6e3Bx|>)X+%a$yc!3{&oBK|5+c#L{k*i5N1}dlD|MFj
zL%x|sUv#z!XZWI`#Eyx|9)=%PtOS9*ts7pydkeqcO@bL`>Eq*0-5>8>-5>Wvwk!tQ
z?Zg?ygO+qRB9^e$z4)^1L-W*sct73Ui&k25n;V@Ea<h@Nz$+DGTrO$x{u~x>Rv?zs
z*|C5>HE3Kx@7vfwZtV!1CBSC(MrlBs1rPkJl@SYX&CE=x>ktsy+VYBhJe*g2+I&L?
z681u7+ET{ce=G>9Q?zo_)G1&P15%>&q?TRWNkekoApM?ueJ&|m0*~mcXvH;5@gGmS
zy2(8YFaCzbAky|1T|oG|MYB32bR*u53d+}_K_6xS28c&y@fWH41a1O~mxr5TuLXbd
z`QR^_%HJN35=y)uz-~ayXnp|Gh`Ry@@8AINGo^04@7pPH1ceSepZO$~lhwO2{QazL
zx3dBL2cy_fB$+&RQQlSgK|v(BX!UHMSv}$({<X|X3lNPzi5PSot;`0xduGZ?8aOwD
z0jrhf8q0qtB)Vh*U$JQ{ccuT?ZruIF%pa|fKidOzh!2T9Zy6icEq*sm?x<m&DF`M#
zQk_AgHI2o?AN)j0d}gYVFmlRazS&O6#o+6To+l#T>-n6^^_PpQo{!jf@%N*TYekRi
zek_(LR)-9Yd#h$TsPO`7DJfZCQteZ@3Ave7V8svA`h8D%)!rMxr9xlshf1gCQ#{bn
z&22JvXe1kN_}wMDXqvfUNNQ`>@=^fRX3nnn8kF|_2?Vr_5CP9^5ldy3RGCma;@r15
z_9M)h`O#4^k;;!kis&q;*2jX@yR@bB+@NY#(S~;I27{WS%&|VG%m`T7O#j%NCqdUO
zdnJXwO^koOy&`fg+`l_I2>seGHuFV9z@@*bVt>hnG_j|pxJYZ`I_nF6_tF9(%cfar
zoS-g64C&3^Vh1yAj~AFki1k*prTs+4;T7#$yaUz@Bl3-fGGwh6OzV<**HN;TVW=ba
zwDV@Q!_OI`aJf&tS%TvcA??hzXNv#A4_mv+HGp!Xx3?&#-{47W*D4<`cmCh!<c%d2
z-oHXu^ziT<yqk<z{3WIIyVwDra$>+3QXl@jxTKPi9jnwyBAd|x>*Uc^JohSj{q84<
z52zq|`~3};8${gEyfYYS;eJu*Jh^*@Uak6PAQ&PLL+$mQYn;nGjD{dew|Wr!0dM0d
z4`g2>{J49r8kZWwlpo_yZNN4D&31n$Rv7TENC9B+WNCrzh6BZ?yBgoYjn|W&?UtTr
zqHLcJ!H#B=24rmQTqbOZYNas;S%qMS2|l9BsX2bZ5v!;+e+Pj>tv!h&K{D?5&s1q)
zO^&p!M4+jSUn%tVMED(uo9(c0?;^P@ZBwCa>F>0MHvZVE+%{{5<=?{ix#%RcfrPo>
z&gnX|!t}ZF^gebGOr{q=Gykla7{p{gDY3r<A6u6{owP>u*I+T+t*X-ggqPMW=UZA<
zf`_)0x!iYC9D>e{gXWGkMc<NUQ?jk%i^p@*Fm8ba89lC-Fy=5sj`rgDWy)55Q$*U(
zZU=sW^XACAO`-nV_r=F)GFw%^)>pmrx*aean`7_c!mMIIJ_E`wQOp6rn_sg9o_?R4
zFphm*D=r3f5DzN#Neieu_)&O>acXgjg?{)Z`MQS3RmGeI7wxGyHD9z9l+`CvkjLr^
znZ|!tF^)SfATR?M0ATVFml8bCK}5N;@w1Vw?luL26K~MJ3;nih$Fpf$v*$Ua#`gua
z=k2WJ-#_cRrD6BsI;Lz2X#}t@X+asW=o<ze@u3$Zxwwy_1z)2^r}tCle1l1{6td9A
z=mCIlL!rGsk1syD1WHTh<{B-Q{mo#rTPnqnAac$QNt8;p<yDXN=8|&RC@x@)rkaJ7
zm(M_Dky@Lk-g@5map`O`Ey2#|yP=8<+-`2$^G>sy=_$&P2!h6CN(_Mw`SW;OuV<A!
zlU(-k{Lyc6e8VZa^MlUAAC@?Q^nnp|-X>Im=q&Xt#)IbZT>Q0)j(*>Yw|eDw!t1}3
zb;{z?o;FsN`-}@S>!`YJ&Zv7W8|TAy{Q76QB{NM*iw02aCA12Fk%F}wz=~}Lp0m|$
z6VDc>?B=tP8>kED$?Ci_PZkype47DkFnmmGkORLUuY{HOhOS5*_#8+&Z$2NVFaC;N
zB53gJo1JAF-}~e;aY}z8#(oi{C9?Fw<L@MPYnD#YZJ1P$__v)zjV5A`hEkQ+GkY+~
zfL=cV@9Xhka)E3AQI}WeLm#~N<?R32&^}E=zy61fk6zE^da!3X1MReIb#*h_IXSd?
zo-^5aE*)no&!uN1ba~p(mn4rqTV_JpyZqwR$)~L&oMGDGBl%4mMfNf=m^Z>QmOqt+
z{b#i_Wt|hV7^(MhM{H@c{ciIcl>9W*fdS|MK*`iHIO0M8#;7wZA+?JlMa~ofkQ)M|
zqfz!r@4B$P4Jhtq#%dOpHrr`SU?^sim2X3AMlx)e86?$~zZ;)qihe@f{fX*deku~G
zB1+3H&AOZJ1#6+qNuI&Jv-cV6X~NYF95e&?f?V7UZ{)?NS@(Qsbbts7+9l;*?YEPE
zu(2)*{=|QOup~=mg8fFiUeE!2E%o>~kE`kVoQw%OWq87$T0w?VBS?j14F>5fIoYnd
z9=d;{j^MY%Z@t;pL<dk)o@Ks?_$SW7pBBI_@1P5cW@DwtR+iG+9|><4p3`o@O>+EF
zj#ba2P8C}ue%z`LCe7T$HE{KibNzDC0deXwexolyU&f}UMpNMZ9T5<RBshdd<p8qV
z?x%gLYylsDl*ilmfNyrlH|`3G_*9Kb-oC3ei<8qS0{tbvy<A7aB~)&|SYX;=YEBBw
zjRz%ZIo598t>@_NNyFzXA*wwlkd8i48+5gB7bP$~I3p6D9>t|%ITU7Yju38%8=h0`
zX3eGB{qm0Z0W1*V03~@U9;?B;Ys@-q<J31P*88JfMc&V2^W3SuD(quf2}+rrBNVPm
zsd7#Z4t<e$tUP=|`v*Obu!ira2ULCA)hAa*)c!NB939NGv==5X+1(e19T?)I|9b78
zG1BJT!QCwr=A1fKYt38N+W}q=(^^dXm`kRZGuilr!p1R#!@WNTg2;B1(9?8;ZM~#^
zl!Pb{(U|u*A6lXe4N&ixeRQ2N2SD%y$R*M{*#|g^>j+Seld-82^PRMPY=*@Od&DNp
zn;>Rem}<<`?T8W9)TlI*%Q-kerMHy5zNL1wv>Ap@Lj}U}3Rrq87w87gvMU!f9VHFv
z3`eHaxIbNU!qLw!xo{;bhlf+zC@yPjaK}q%YH)Ue*)sOn$UwDvBt}&djFZ6cZfkEZ
zrZb-~|0?u&iIqHVQt4v7MC9v!omd=qy0lukx&HPd;CcH=YS_Y5I~+NmI}wl9!~D_*
zeSMHjjTS=C{nOtl&Z|Z?%r8V6fo!}I<n;R<mp~yl?;@B+PZt>=g?{1C)V}}x9Fm{z
z<kGbBT%2d=sYklvlKsR)$;wzS02D<~or41uGZIM!;vyAM1qDb61jinuz=d^@xTNHP
zLWEP9PFIWDSi>qT${*srmKzRzwO~OUUAv^eM;V0$?vFH=y2L6pm8B0_ynGq?kY$uv
z6}EAfYgQ}VQU<3wwVYKp-o!C4Fe@CK{nW4*Us>rO+F&DigKJ%6R_KMmV4kGy;npW5
zauV(cPjcd7x=lbcA^7lp+lwaBhdsnppnNOs1$+J`B;1Cty4dFHcbZ3^FDdtL;sI@4
zHc)!#ejx`N&?y5<V&OPR=_3RZ>X+|Qq&GnQcey$1rFJ9nj}_J*x?e9n-oq4A(RvxX
z4fPDWfi#r&QJ>WA`Lm-jujZe@MF||FEP*b1{>K=KYB$Kcfw?ngCA-#GK}FGDG2}k>
zP{#`^!gIDLYmPhWXa|=)J<R_gozbi5b?`VkVrqJx(}54&_4c5X_HA_Wsx1I_qW^d`
zeEn+B<qM>6zv1C6!KY6FZIN09qa9CMH4&*Z-lDm5GlauDy&uI5qQ)lftfC09zUFCw
zX;NKYeYzA68CPmpd2?gnDnl82aHN;$>jdXO8$ywicFgD?0=#t8p-J(fdBfIVB%b%f
z*~9C{dx_tWUH9|vv+AOYB5!<rX{7JdMbsXsf0c^%CNl!Go(9`QKw4ue<t{uHY?3=&
zJ1-YK!(P)s=afMB?MRV+JiefS@|x-3q^29-bjgF9IXZ}aHe~)a^X@-ir(<{`AYFX6
znIE`f*<l|?(Vy?DQ#*24y1pGXnVO2xW;s}B(AoFRx{QC<M^7Xh>64nhjIavVG@09r
zU0qMBOXw3y{<~ICw$mS)nqAS9X}~nmTIxYljb?NpKq9^n9l=pZrxt5l>n}w#yxPSJ
z8~~KQ_mPUNADfBwOz_+lOAG-!;onb{suUesA<z?Owp$bv<c=3YKi(a%hPH&}$<`hE
zWN|&9Ak8$Ccd2dWB_Smuqm5%El>b;nRD<&<AQHcY^U{ouvX570n|O_1&BpI-fA>$A
zqKk}=6@HIu@QMXu=1jQA<_)~~(HG1%l+aht`w_IUe&a10uOYOtq&K4_!0qmZl!Cy6
zz$Zr<MJa00u^9)i(#ce3%Q_M&9UQZGEIBr9-Rx`=>jLBhfZW6Uy{^WI6@c%OKcq4^
zT3<z!8vd$LBI^lHCx5kI-*tS{Y0+e*ZKqRdcl46kK6R8%Qt-9X4cjC9Ci$cSN1fM%
zt7i3#&sM|Aya5)Sv@>=IlPt>3Jt>~lyOE<HNHG(pq~GG#mna`|Qd2oU#cq7u!dsu`
zv=CUTKw3yR#juKQf2ITCypvUHSD+2$+7Z%hxoXV?jGmRTVS%CdNrlxCQCZTVe=bf{
za53_&##UwCL-HRF>qb{`|L&AQy%T>n%fHxMYWOXr7B=j)yWw3q?1Xk3@fB>u3I6&}
z3vP&iKJ<5reV3Eh<&D|H<??RO#D%zmp<;*mWbKM#%t5+P6nM^#-edg_>j%51LN|k-
zBn&9_Zi`oQ)5$5>M#PvY%7xhT>4kWP5Gv}Rrn}?H@1vq`6YdG$HGZ`!cyma9A4jSe
z(;`0L^F~O$B{Z>Y@7F$KDYkVs+2~3J@5O9n#3nJVmEyx!B$P6}On_ko6eRpN%)K|z
zAb+%8^q<X8CW!qUDk0oKBVWY8v#(UjMV;bv1>y8OO+EnZ?94mQ=y17}8@z4Tw>vvu
z;60i?)K5f2&X$YhP-?L9c+-*jw!iw>=G)x3C-Z5t0t1HjL{zkub@BzZ4*@d-FSwg6
zq`aDLEIq%n?3h>EJj8`1GTfQ|G=Bv7nj*zvGl2?A>%bBeqS+y<QGv6(%ni)09p0Za
zmk$m)eecp!SPidEpLnz8>yteo{dF?J6=m1v@T1X?BC!aHw?L!F5!lyc{aA&`4NZ_m
zpLJwOmt<z$0F(Ui;1g_)_ci@R*F$&M0z8{Fsow*1jiLTbiy*+n&&<kFAtjx)cH5+G
zi`qn-l5&Z0h*=_JFQBiFYG(qb>W&Ui5d{gbcb&#{T0gp|684|EX?K#@iko;KZ+Xn)
z3^-G5N3n0VE+Xs1(Z4%4GR&ul(h%dpkBb}@4ZubtTSAfgK^0JaGp;ekNZ*P}y#9&B
zBdB|QW9U6W`gUajT>E_Xzt&O3L-``8G&uj-Q=|=caFAcdXEv!SbSQaw1GkhB^XlB2
zplBHL>=KAW;91@7dGvX69qc<spO|7eR|VHXV@oI-%Soz7?1LwdLj)MVtj&aGzdVq}
z9{@|{f6P?6jM@Z>%9DR4%|2vTpAE1#66<dm{U4h{l!|1vbVCK{$2Os|X8TTidrWxO
z_3|^*pX&@ea+83Xv@9HC>cC-*z#N3LX3B8ZnHfo3eUiR{fN#3g9EWb&!3h|W9=xtv
z8FYG2MW$$$PN+^YMve0DWji>lIr`4#8s{8kS~N5z23u%M(~VP%1^rwr1~dw6iF7b9
zociJyTY<4XyYF4_f4}n(!SYBf>KikV9XXB|=fVH=(wXPDLw+JwO3<lOefYH|b_|zr
zBZr4nb+?|2GtNiIr`$0nRX39xbjQcb8(kNzQlVkR%y1*$-@ID4TD8ihTlrARG|Vq$
zmv;dz#2u9DFs8We7=}d=CxOWw)a1Bcoxnh47a;e07AK48Jud70YGRbboT*6}(}P6h
z3#EM<ae-UM+ebA>NypalgVKZMM!k`Q`g1R@r`1t=_PXaaP-n@U3%d7?9toCvDj2Al
zwhb&54GKOY9AOocLndO^9KgA~`s&JuSNaH#R9Up$S-;$He)^tHryh3zo7EDunIEI2
zGqDS|*vEUHw!rpo-?`iZU8x%dzAb~ZlbpcUOem3W@>EYuHb#E{BR+5Z<Ce7NhFFJo
zb@#MG3@haIr9xF%iTisG#7V}18<EHRD?ogMJ7FpXtU^F(?>CJuEU<s~#JJW?yv9-$
zMh>$!<>=D2f7GC<b)v;$zK|4@fnPNnl;<cOQ76lyZBs8wNd1*21VY9u`TwygeMCfO
z=@$>v12spdNS2a@d=rbaXsQyI*(4AMzlJu+m+J%-Xd0bl01O014aO(nqy#K8=}Vu~
ze&3)NA+ez{IH=IY^NEDnu6DE?F5qDi2S)4+3`IFIe{f0sjtMvi<S^@g8!8$&B_$S}
z59?mQIXlzDymYVE*`9ADQKR1YE~BQz(H*@U-(+;D_a_QUv3Hy=F>rR%C-#iUyL8ot
zNZOT&@z^7)I7+a<h%mo*1T@T2{k^SD8oWPDuu|XkwN^E-Msu64uzvs#JF>6fG_@21
zB?5(7M!?-VlOteYE3WeOHmZK}rwzi?r&?2je;qx$vG0$-f`U}e)qAJ~6YgH>03?0_
zyTb6Eqp;+*O=NWZOEQjK=|jrlG6p_eZL7}CCh4z%PBTySmnbPm<*2^t**R6rga$1V
zC!edR=9bfD@Cw7)p95ZyjiPq)!~t@7{{EbP!a@FvMb!V6XErjv?|KG*HdWHT=m6KA
z$zMK!4?sWqDZ6}vzOi7$I7<J|PR1E(+HX&xey?DSHmY3RaOJ~xTv9(fZr#q4HaAxp
zr9Wn{61*kZli$eoWetz?fdJRm>ng9I-yPBW{<7cTO^RY|o&}dM1uIgycJOHR=ogrx
zy5>HeOEg#c;rywi_89O`{c2MwN>o7e*(#@?{7KmXYzggy)`Z0EYP-QVZgDam;ej~n
z2yz9`PDg}gfG>Vf90`KS_BYRx(7j4*HY9Rl%$kI;9Tb*FqpQ!6lP|d#nw6OyBVWt&
z6A7baWCed599j>%cF=6{3K=+hI{vlh>Nkz&U4=ia`r`UyM!VNa%G@*nog~u=t}WBZ
zyVSpl*oQQ*3X8G7@0yrO;rp|}@jSb5(1^2q$Hz>)WqhO;t($xLDHQ#Xi>a~7>_ua1
z>=wF19;zK7=%V>aH{oDl&6S+cuRa!nf7a1@4ebD`;pzW-8cFwq&@ioq+FevW-DqTd
zXXS_qZhMPLnUomR-^)yLGEA{0Ne#b$L&-Top`AOF7m#GkSDD($SV*xgu+?^hsR#zn
zu&Q2#hFHb6tu+JQ(Zqpn-F41EgreZw(%eFuC>c1FbjKYXY)P7<;rrh{Kc4R2fO%oA
zDuX_4(HE*~v3TSb5HX-?-3QtA?7Y5;Q88jvHC>SU4KHuJCId8{da%uRNialg*Dxb>
z5rw91_tQJQ&9W2lhmF=9?3U)LtQLmEMDc&6Q$%noF}W`vtv8xf(E+k^5}cIN5}{Kb
zhY5K^U)^51OKiRw9PRIi013&ANsU`vx2gg67G^+t0k&N|(^+dSu%aef8e|a4ozaGI
zF+!O{44)Cbh7vQy_KB)mMr@6>`)%oW?VzA8idctH6i}&gtFNeyASyBge2g0m>{2H!
zfamyJ*u8i7Bo=n%PB4RLG(>)P##l%kKKdlfzinkpP7cFiJ)ExV%%QrZlzsmO2ou&2
z_CdwmJ%>zs%4#HfUmumY$tny{QyU?Ww;0DF1)fpAaRiK#c61TdB1@(g%;a8eI^`9(
z#7<FXz+~mIJM6%e%p{t6#<(#1to+9Q*Oxbn!1IbRO->Fc0~4JElaf)n6Ns+xH1GF_
z<$FCS<$Zfax!*tlA&4mA`7ETAJXn<sYG^@A?Kq-|bgJ)^Y?k-l#nLC36HCIy#c5c?
z2Xlue=;(oqktqh=l7Iwz0jGMCsgSU7gN?S`=cTfhHmEh}p?o&DJuii1*vaNW-B9}7
zXBT~^k3Iq<3FTe<0!t7v%=EeW;6rTA<?S9LS4Rl#G!Qo;XuhCB>i0q?wHMH3uVFaP
z%CdL#MnU?F3nB_&Mc7qo)q!=@4?F`1_nbZ;;OtAmOS#o2tsr-=@9eE`N*<OifBKwd
zsYb&_p>UA;6!k*t+lyCTlA2`Q!l=ZnlfraB`|~aR-}lNul}&y!>|GCe1H#XcJy>nT
zoqa^~iY7_JkX`fM?@!Ys@}ie$R$lQUd-_+Z@oBJujNhLi{{%mhBt^*mPFJ|3bV_`V
zk4mu=j=D&9$bs0#B{9Z<4#MBy(jFKWC}5FE9Nqq`X?(TILJKh4lz28AU7jEx6ck6l
z${Xx9>Pj47S3y?@;7eh5<Bity7!^Lrj?Ky#o^`-k0QcGZ%qY4t9LBU9X(xFrBZ8*z
zq~JMYoX=_OD$z!#_p^>kM#dNgl3L##-a^RO7^X&{$WDKTCDCOv8UHekN_=qNgD1}r
zUf3_l&iUpZMnu~NsVcawKVQ3gwqo6atv)z9iU{ofZ3dzX0W7G*r|qy+ZxTP*p**PP
zyiZ~aDyFHU3D16EXM(4EuC$n~UJp|1!o341QU%f9RrsZrt?PWC8Bj<FeUMk84-mpv
ze;^6@4eS3i`$@~kt~?S)O(1$oMds5A8EDLBaK1k0LtG!+f?44E5A3uOvyFQTtYdV)
zJ?jDktTEsNENbPL0QCQ#_A_dpN<L2g;@I`8Exuq#qW!wW4T7k4^1eA^1j72JTowO&
zGH(=t;wTWFX`j=~Y$P&(3DvmfN;29sBAHY=WQJ`PGQHTM=@*d2KEO-)|Hab*+XK+Z
z!6)A3WgwIZD&=O}>!t9PV3a4t!WI8PB^shxia3Un`PqQo%<JH=E7BrD!#~10D-NGO
zDGU>O%u5gd8}j-k8vc2|2fEL+?XpU;Q>W_6`pxy8RrlLI8>!DLIc*7TW@hHqs73a~
z&+__0aFh4Lq)B*Fg%ydAcdye6l7(o6^%s@_b>=ZrKGjBj%KTTtM6`v5fSm+bi-(xe
z@qlLV_3<Kda|WREAXQn|(=s?6zOCI|QhcKb(hXS$HM>E%O5oUcLobRh^y8w(59=65
z#Sk(xB}lZ0#oVVQA_s;o`FP%mHKG<J?Lr^uB%XU*ynTItRP-I21~iHnU4A)xjxj&$
z+VXal>ODKrm2=%QDGSBV$5Z4!vfhO%1wLgw;35co{*ZQKh!V``67?Io>Gz*$nxjyt
zpf)ETcRGFw$XJvNs1)yo4{b4CE!Cs>vu7-Q!Nh9V^LK}*8jr3hb$UNDI~x-LTSEF0
z2%7`NI_t+^&#vJ|hM_}~4iF2n2HtGpSKt%nqJ4EG75li8^FhCwMy2O$)88+JE`6T2
zes8KSkxcMBHCVY218BL-8_azgj9pxy=?uEM2P8*BJDFOwYevSPqJSo$j7L$!SDA!a
zP<DGYFTbD!<KQn!HZ(K!KYJ0~ZpO@c_Chlpf@TgV?hzC_w~Y`S>%&X{y@PVl7Nn|C
z@_IU+laLl=zU+_LRcjdq;>b|fYE%wAWEpz+6Pe0_+2Rl0`sF>`x&*;Th-UB{q^!c~
zX72&71%$(TBpA~sZrF*mOkwje4VQpqRI&x&{~Y3z0^K=tZ7z&Rg^Wh=!s>-vp1j1+
z+-GsKcqYV_7@<K2;#PQmCRDE9ScyMMr9a_ysbMGz*#qH!?Q9Tt*cD#jf_;7eF{>CA
z=%S#v35x!%`Ph4OBuFCQK6125FkXO#s9;FKmcyFHnzFDchmU=sJjULBVeDF`32g6B
zyp+wU2zyB0b$qUE$)P6Z56b&;Z-R3`nk5wswLBcWyyFvkQ|4Poh?10-qsTA9^Z3FH
zy6r*}W)n)c6lnPO<JDJgd)0&mI?$+!<^k8vc>|Rw7MwHv0Dg3f-l{X)Fl!xu`D5D1
zY4~=}NFx0Hd+`^uk2-9&px<90bny89;K{XP$cxC1<_oGb{<`e<G)5H(mwByp+wsZv
zooCPC*==}pmr-x!tHUj}5-C}8b!q)%9->;b`hO;lQV9HO)mYG`4O`pW@ptC^uFJKn
zje7bu16Dg!1XH~Hl@>xC>@?SgUC8dN$$4`!0&G{GydmsNPHVa1Kauu&hTh?xWSQ=5
z52|kUe`zU^uh8;aN2o@K3CA;|(t{JH3xh0e#Q0(}KhGuTV`IYP+3(rZcb_c)bZ3HZ
zq}(4MhfLWDzsCS|$<7WIi=m-OGH)7v*;qV;3LOM(63z`uXptMUj>nW?*N?i&(&K>8
z3vdYodlZ}%pmDFRt)@liJ=UnYUP<X+(Zup1b*KsJWGp<ESnhF<BeuPOITd=k#W&j*
z@HzH)KZPp#EF-IdM&525cWsO&I<G_2$Q|I?6`tVh|3g_+uThRhC48~ruyXIc-^BMx
zDhJSS`S$fol`Ob$qBJs%!Gz6NL<~EKSN8sx9s11;>l_-%8>K=MnnS7pHE%OECnwa|
zG+o6!u;*SYXLgCtY`rkqo^Z|DHbB@^6|GS>YziyNXaIQU5q@i>qnn^SoP!Ihts(?c
zR*(QIC4gzv-1Y@mJ^><b<oBwcSGOm>_m&Ulb$$}T)@k7niw{Opzgsq7;~zGq#&EWU
z7;inlar<#M=oc4hsG7aI#Z?zf^o1+`)bw;4Uk{l(RYOyz^5RzmAoULdyppe%w}_@8
zJM&nHvx(}1z}`>qFlqNNAXlM3X+to>%_?fjt{-8OOw9%6Z&A5SLt&7A4$=~b9Fz&a
zSw45C1k4ow5E_4an*NBAy6EE8;>XMt7_ud{4NZ6<+AsuAc7>l!`@8Q_eTI5NF~%*=
znAH!-uJXxYp1Th`o3@_KCmSrL5y>yN;7PA9w*<=>NGpGuprnnaE^;<#%YV(8w<S6?
zmjZdLO>K955~pGpb!0_i9d)OAlbWg92><sNtk3iBSBbbiS9|qBZ&cxZ*1OMRS7ch_
z{B(l*xxUHexmH!ri7SLVZ7%m9eE5Dw(b*b)=;262Ccr)d{-Qk<>}QJxf;(>o!5n4v
zriDfgwnrah4oUutYF)CwNuM@R_=6mjTTkeH@t|>DGEFsy+NQ%+D%`019vW>}LzMm*
zN=a8Z-|1?cS0$rdN!vl^iz}swSb0wAe#8tT5Y+lfR+Z$))d&Z4y}1{UrgZvrD*gr|
z)F$v!yKQD#bbzmcfKaxe&y;7^Cv~W*Hf_TjVBh!#;DA-zRm1R5vl}}K@peR|>}n|(
zA+n6D+;;6fn~TpoV|z$yn^$UbXOgg{F5}#^=a(}?)DP^&AFVa=FJ&-;s$RR>wC9~W
z@V{DE>L9i#1+w*t>LQ-dWH-V*N#$0;b(l<S%T0@7{QV0XD(lifE+iaAD;H}+P$SqX
z0zXIHJUIj2New$6?{>=%ui&%+%{DXZTO4`Gf}9<E=L5qWwG)9BX=f7$sn0;91rP;3
zEgDgzm(a#Zl#q-Y$JZF(Ie}pKz9J2Ban=uHizo&=csNX5jm9_Iq)sxe(Uq42F0C+u
zX<n4Y-=7xB5KD=kjS&jSVWAJ->wtJdhu^`*3&9efk|3*L=KPCqJ~#K03-PVDnsW&Z
ztmxS0VXlKH6K+|CB~3z*xG~J<ag+WO;Sc{DpjAxL<u+ukk#7>Bfi#z_RES~5O*MZ}
zyAY$&@*Aw?&%=z+VqAze<*@wCLzC<wqgfYA(vhG3s1P%=%;RxA6u49>^7`E+nh{@i
z18##^E?m+`sUG(y!UVHjiQnl648q=z7$ztZ$`NqNiOVZzt^%GPK4Ytyib8LRKq(fJ
zEZe<i{j~U0^LQB;>V5Te@cNYVLGS@chMY+JK8R^cfJnga2261^PkKz7ra@O?{SK2B
zFo4YlAbtDW$4C}}Ilj-lQ;eBYPm{|V+fbbXh6N41v^hRgIq&^`y>`AgM80<?8<&%I
zev>)xv(wsBz<jr8ME1I~<cM&Ly`&n2n;SPxTR@wJpt2$Zr-T;hWg|DAFxt8<W4jC*
zYex#jGt9K_Dt_FODkUY|CSlyKwCl=&iD+*5Tf+B*VX-d=2WqgPYa<OjX1}PWN1+@D
zrf6zf>kQX)KPUIRfee++VGVtsV1=Jf0f|TFu%mWd{d{yh&jiIU6&$nQR)n5-L3PLu
zmT2?2uWYhk-M6Vq^IeS@2b-!uhS;H8D;J>9X`%P34FnF!OYZCgr3*#b=2`r52YCPY
zKsHw|6!4>gAPG*PZPlH^_Mf%n?h8b^^7ifB+r)w$6WnoX7eN5i^)+L<?1aEfJw8ou
zDzy_jDkW{eydUx>i3k|t3uZ(6^nEADX~^U+f?5=psfWUrY(NX9(0s=TlW1~)FWSY$
z(xR04-epzcgZD#h_RpqhPoGfX&?|=)gEu1(*z@OoxfU4g7T=fiY=OVj632m&Txv<g
z!dmmVb%hRics#@APusH(ShKIdI~V`2CD<&ZYy^)^lYDSP6hPj`=_Nz{^pdq8CRC1V
zZErW`tw~jf1sK91H+p~Vl_C~BCx3OBFFQBN$KFvCaR~OX5hffRwws}#ewW9NHxB+%
z5~z#^voC>64)wP%j((W;0+@xuQ~0vfE99FH760uprdOm&o@-A(r&T{$z3^(jxRQ{B
z0CIwsI_<>;R?I%D7u2EnCV;lQ9eBYIUb6QB(KjqQ)@!>zaW$38M6r}1Nh`@|QNt)1
z6m~sV7lF@abY>rjH8t@%eFZJL+VhwSb-wFssx9g<m{<ya$$i-|6Z<8<pPlDbJl%!C
zAnfWz`{Q9Sh4k`x?{0JXzAwM=(=$iuA(EMCZq#3XQ_`%wOqj=!ZA{vONAA9O1~#T-
zSmBq!m)EPD*C1eqIxstnVzaBP9)A-Gxf{+Q7S%uNX_w15G)iToB6f17adx3oSaRA=
zQcJ|N1&W-qSE!Q8kQ?+Ph!X=?KH~4Ur0*cW9`o4K|D3EX2yIsD8hu9Gp%G-Q?q7Vy
zBbeNt0Fyd>fNxTIlHQ)&UIrm0HJq8XZqc+l=bVhkY6wIceZ}I;Dm64=K&qOY6&XoC
z!9JJ1&Q94fC<@Np4?c=U?42id)fNrf=mbjY`cAH%vHJ7Z>+qyE<HB!W|DVBj(z(Cx
z(hEOrBgZ5IfUnhq%&hDy;g9JLEXD2@=AL(u;nhvfM<T=YjHL2D^+_$|H1=I(Zz|k@
zS0k=whboHYhDWw)BJ~oazSOdAGYHO#^P=Yg@!>*{V*FZk%_Q^B!^YM+zi}d2nlJZ#
zFuCWhTLT6B+vcSm>SJpzL0HlO_S!hqLFLY`rtJ*U?3)^(5vG@E=;-Y|@qf6&wafT!
z6u$Fg-RD?4=EA3PgHFRzDW4WU@zB@>w^G%VU}r$qI3}b6YX~eWon_uqV8j0@XDWV{
zEAA(OOyXIGe94f)WdR%4B+GgYyw@;3FGsugpOwo|E@C&oFGIHY%hHM3SW16+-09C2
z9|Su9bzWmoF8z4gKufmJ=Fk5-?51@*#%%&dyoM3K{0wpf8zmx4ZrL0t&<vs2_=Ule
ze+)3mL%#5c^R~8@DsWv4*2*tBotq*G>`-)uPC*up){6;g{s=62h^NEp48t4N&iIu5
zFZgM1AFIElk#K*D{tGsjEqk}~8u=2QFodgR2khxQ?^~)<8~nM-*Kc}+G-{_X!lyQf
zKam0x(OkckDTwTLt_@x9!j~1_Otx%bLDgX@_5Xh*w)l9>ad2`HCrvOz)35o`<$Db4
zdkS&&Ngsg+Dt8w0$?ej)%3{9S1RRbyBKVlKBtq<hvDdaM#i3F#H%Jv_wZ=kTZ_>B6
zl@+}aoW}_Js&qyVj*6AGJNlO|Msu+sMWYzJFzyodktGnTyuBtdXzjpF1^-Oulyt(p
z9(G{du&#tlt*&3Ix0D#1cR|E;0_0m{*?t(CMXy>91;Rf7{CeJ{v%}Nmau&K9UY9o0
zHUW|s+0z;c?RpTWp1yt^`0Oy9c<e#1(|eS==Pe=<IH6scA^)T39HZ;nq9`0TNgFk`
zZQHhOqp^*~cGB2tY_qYGCXH?1*m&ptz5nhQ=bXLvTyuUCSbO}A5lzJ!@!Q6b+B5uc
zfsCf)ecC40I@S@HU%xDOSl*-`u9@_FKaIkiVSwGtck649%XgNe3owoOGs20Ql7huT
z9Nott`o-=Ij)%qF;7e@KLBp=4;A^_zNAk?`@!lgKbV{6&%HL4X(?X(BU1mCso-^CR
zp#I6SdhEnF*@8Yb668?4{PFq3yP@vHEN?x@@*7*4LFuZ|@C~x(<<uru;Da+{xZa?3
zy~<k;E!hn``{MU0#2R34*2Rb(VoJ<PKS|4osi?RT{-LQ64Xv%Q^K);{6ewtBwO!kO
zNs|M}$o@dy3^-}izI|tN-;^|BNUmy3+os#p0b&)?+&6qj`)2(nP9k>vl9ZI|dWP3y
zBZRlTd;!CACUMGF+vfy#h*f)Y-LJED<>Pq1ed#d=B3N-#_Aq`UW6D#}yJ4l*SVd$Q
zQr}fCaBt321ZbBpyw6wXJB2LZr=eqNv=|h_#hE4$rlq~|M3GtiIHfJG>m1-`KyOT^
zr6BhT56`8C;ZW%ka*bFwp48!?Aw$}s>$xDqk6qc9X+-c|M#G-{m(!rzTNvM$F-f{3
z)|9_y6p-R{+WS}FJe1z<cA&8vHB548j964pXGC(Z#FjIwP7Dw~#US&AX>%x3RFES0
zvN5+nfQ{O^9b3t5aZB!H*0Ki02)@B9Qumln?z&0Rdvs6@x`YwX+Th`5CRIsA>k2gC
z!QDP59o$45i0qyw;U&rt(}Jf!VSkTcpm88)aOOFHk@|BQ#r61m(Ed`~d7VY#(w8Fj
z0S|dm2)l%1Vc3XjR5?ZAMUD{1*Q&lh8XSd@ppcn=DC~6!q6BqqfGK<el!y`~Hq|d(
zN=UY7%lDb0h7g>d^W`X9obBgfLBNxlD{C7%%!31{Yi@#7O;jYef7IbSH_Q@m&&MgQ
z|La<9ds~~BwP>Pc)uIh>YaKRx2W*lzhhN0|%(U4M-$sYm>o>-IyPAu7)dPss?UZ0%
zqb=RW?`YU`Gz{5)9-X9P9bKAQ$L}ZMLxQ0+x-P{I#i2o;nPYlU3JiL&%Q2uzkeVOQ
z$V6(@<Cr2z)>bk3>SPr>w8Q#$Y89>gZBCjse@)wPpK@1C>V8b<n09J0ck_!}%&ov1
zG2MEki^t5)1?)4p*&yn1ZMoUM>3jpnwoGOXA~8o<XSfD@0sTMxq%+Xf#K)>Kotn6S
zTUW}z7i_z&f=0R)IW_pmJ%2Of=0MKWOm>++gCDpF#nMacPaB5_-G>hk3VoNS&O=Lv
z{Kk#$R_Zd8kGv2@a>U6!kwUU(;051O5|b#Q3gFq2+D#+OotxAIo}Q9l-)`askLDEx
zcb}f5On%2;Q-q#CB+$)(t35X@D`3Pj>G02AI5K-CSKixgZM9!asWiybTSa2tz^luW
z_T4HMtDVJ!gZu`2??G=cfaeiNf6vq}?tM4jT`wnArxJ~1`-!9mxK7lS#r1gnb0c-9
z!viJveDkR+Q|{$<;lS27s%mPiT5Nc5i8!qw#1k)=NjAM^5mSb>Ww)B|>|~eR{-8`_
z0HG&85hXKm=`{}D=L5yIfX|ep8N)j5TA*Uk*1pEkPO<>~b|t1`e5?@|RWfEU$}MCQ
zE!j5Wh8iel(ES)GVwfA!t&5I32&&QD)RS=sEY6lM2E8YojnBsm<BTL_abL}v35I{h
z=DpsDAhvV|`en=A!EA#VTF-Yt-KJsUl0{AN=gl$r@?nUXlJRlR$Bi>o6)0=EwuGKH
z`_uurc!mMDx-b4C&h2QNhDs^o!s-}mZp7DmIVIY5VpVOAY7FpwblN8ZLu1G9{`RKW
zg+1{x`ci$06lyjbvkVRp%`DiU_@Yiz^%dN$NOeM%^ziow?T+4Gc7u4u#Z}1c2V#o_
zMW7NkeQk|N3zx&&C)TXSh);$I8JRnvzo=bf7h2`P#&du#RcM^pr+MjD3QxB4uw({K
zTJi4_1SO_=9xN6jMwzHWoQft-%H1Uj98%H^b`f;;XN3)(N2ox#OdWAjza&N0_PvwF
z;t}~7pOM!korVdx;=f-8dam0B>`^JInX$C35u}$g!SJfta+(3L_4?`Bw5GQ9qyw=8
zt^303`4n0)_XAj}wlCaIlV!2|oa!9M_~=>FCnb)>Cf1gKWXTU>cEh6m_Ja;w3Zfeh
ztjDj5x>>L%*@0BwW+q)q@S32k<YvF4btVU;rh)D{dAplTg~EkAQoPj;*w&Fw$kR<&
z1LB3VCXoCT<(y!-p)0GgQBY7kUd6XtQ#FI8F(b*u)1Qe%P$XY<=N-eDZis<F5+s>H
zkQ|W==~L?E029}SyOd6?RpAo-ibH{H>S*Fag^hIb%nx-0JVLFnJ|^e6eE$8>mYi)p
zXbaMuXx;b1`(!E8W4o-LdA5;iT|Dh}kcH_viKm+f?YqJ8m%t?VlC3D{Hweh<UIt=n
zghOgYh&ngzl+o_xXE=s|DXmd4P#!bCdTlyw3c~z^JZ%aPJ-Bb|9b`&8Lw1t22p6RJ
zrBiPy18y+$={}zbKRtYJ|7o^u&q?f6mrSyPdoCs|m88s8(9Sh}BUwwbd`5OCi6Nel
zwVgD;2=#)Qo&&XTttDq@4Y0r>(|j{F5{=ry6_Uya>A?Q<<{;JGfIqT%&DHJ(7Vu+G
z<KMb2GBnBcm%xph0QD`%qA|Z>-?y=0_xvt}Z{`|GAuitPmA+A1IM+sw;+tXG$huT0
z)B~@idMdCmsRM%bz_l$L&~Xeu4tnF~xOKAn(RrcWRE1hGXPODmn;t)qD62W|f1dPi
zjng3HxUfl(C<d}%yK(RBp97M)q?%jY8Kk~V#xLk4pwH)Oz=Q0UvWZJY#V)XtFzU&k
zZ%wm|&%1uAtaa=uJHuZLnhs==2|M3X$sC33M(2+20gX=hYVrA2L^vg+dM{(t+kogt
z9a!?eEIO_Zw+fBfX@Us7aTP1WdZG1XR3w+&U6mhcSlb8xP905R-ojpCi0bJpqs>vp
z`We0Wo?!c#I8k6vBoApU*Tb?sRP4x^%>Kpi9n!+@?ZjP=QLKicU_yd6SN_U+gkNoA
zh8G&Byt{bI0A9$(dA6gEV09ViBc;frgiZthZZz08&+U;tHge6zB7O4ni_&kR@y+q0
z;2jbl1in23!sy%eC$T-g#6BBj<W)BOnOy#(mHD!}ZrtJFAKklS$%KHP4-OzHA!k<^
z%e|a~);t`h$XiXZ!{5I+di%t`52iQehQT3c(J3d-Ty`xuQVv|2K$Y^GIGj6J>z$1)
z6U=x%T?+3Kl!ck4Mbaq?$@s5q&OY)_)nGTVch-#P$SNOBV!Fs#P*p~C<j%S4#>Fev
z{g2v3$b{%hIbIkae4b9Wp8K?Q2QCl@c1sO<i@s0Sl2p(X6Jj$r-r^uUFJ<L_jj-z`
zN3w3G>R2BVUd7H4gb20PgieTSEdc?2mnV%&RT>BhqQ|fXLq<yR8gcpWX1j^vo*{>S
zR(}!?g^ik#-<=h~l)c>?8D2df5juN!6jO=CxYeCdP}!BkZa>gM#KhoV-|BK~UA>N=
zI(pEtcx0eDtdyd|f4O3c!E2X3;>#GTTa>1*^X||4STS^V>i*+3kPPbS_uv(A#6rjj
z;zQ*inlUMXGY#9FR98nRj8dUgS2g3NqC`zhPTB3K&|Y*~P?7C0``Lp@9iE^!1s?NP
zYYj4&ZbIyLXO)#MT%D%yu53yW3n$lFWBQV(E^>N7bLj@Vqj$`lV|TmAcdFZLpjtXb
z#WsuHT)i-*aFivAPxOoc6vX5}EEdZuyVWvT5!6uSXfM756EpoZ#P|W+Pd|W9!e8BJ
z7H`kEJAq3Re>Gps*JK?U2lk-$y9?s<cG^RmnSU<d#A5B^>j`6z=*E*8DySy#f=Bjo
zx^}FX_decL1>UYmD>?v#u`Hi{0@sA1;A>3j+T5i#3yT52mq_QSbtloKRWb|Bf#e|y
zf7emd6{W?Z&RCpWqnnHfHs$pi9bPiXbHIO|Lk1^76+5$g@*oinXY1eAKno0A&WN#i
zH0{$*XvgUyw_)ENqk63dEjlfRcx{GwExP|ri)!lpECZ00=-7Q@x3Q5VCtn^w#Ne%D
zRyA&?gu5;HE0y*7n~NL{b}g`VW`gn-@ap=7ZhcmX`KLaKLg3i()E^q+?v$Y!fjH^E
zLl`$_w@pbz*_@M%3)15%IUAEy_mLk{&JaKITu%pV6om%bCIk`)*0*$m-fL@Wfvl41
z$4l)e@I;O@Gt9aoe4f^Ra01&1zza!Xu>;fbF*2d{R?x!PtUf*{x6*RcMeHp<zh4%X
z97B7!zwVLE2`={BiW8H(-_m2_o*woEN-TPsTUt#-_EolbD;{&h(jM9-VPZ?0!c2L-
z>29&Xz@e?Z2m8*x`<Bo1DdE{yIL&;)TztuaOF%&Q`z(*8&wLRTbwg$Y)iK}}&T(??
z+W!C5z24>i^$XL=FDR}-zw;t3OJ7vT#rYv&Z{CM4Ua_bqFN$O!i6}~@eWoGz`^uEq
z`m;1C%-XzzhihdorRMqA7`6CydZD9hERghnK;P62a#xB_vLdqR`ytji63vmM0WD(=
zm5G(KUeIm9oOS(NMtCz94srr*!g*PPeJj@a`P(<y>>mnS9M%~cpA1Zs@ocsX-R80v
zgqZc9(@rjn&5s*a>)C-&&xR}lirEl?gH|Z`tsRC8TLF1C(UC8~!Z6=I5x{hmC*z*y
z1z%+=eoPCpf08QpJR=~E6mod3shBCH2kFsP@KG*(IUrKy9=C}Zqn3|kk-@MJ@gKJG
zCC2fRpq0bu&zpeTtl!X=j;uSDm<_!C$|E^1g9i~2m-FRoGSC1CgCWj<vd~`?4WIJI
zT%!C0dGs_>Tsi+H+gp3Sm0pI{Hi3&KM^bsJtCYc~f`pV4zdM`+mFe4VlDnT|>smpP
z?gH=A)j<4{AV6+r%+BeFU&KlFfI=7E__=zJ(u^w*oq9eDhnyr;OILRTESI?aM#63F
zy=QLHioP8Zg-P@^5$qZ(p~$8lkc(3<8mz&Hr0x9N<W)<%-tPDNIR0h0+hy9Go_rD*
zeIx27P&P}R^Fw>lMGokoI`)-5!pBy~rPg8tM<shu(i!>5!EFB<wQs%Ju_mRbf_<DS
zC6azmMWO@uqk9VxjCq;{gW(rf;o3pNmp&{LWM3@d_|c|f<K$s4o!PR_kN1{A9BG4K
zXv*PAghE|QpXkM~it_X+u`>I1K4AV%FDf#{h!%&B9F@cn@t$Vt^aRyw$gqCj{fp6i
zCwll$%nv#=rC7W0I%EhYeSfv<n;F59DN&oOV<aj6x$VP&D;OaSvp~hj9yky6Rmce!
zaoytLVa=a!=E6>I_EE8*)HeNGnq?r&r)}a&i6)J1ov_u^&xu5tdHUbkedgB3i_QP*
zz6*QKCO|&k0qi%souD4`);0Kg*C^>%T6U{H^4e=2tCv=I8JvG|{hlZoAqLaCV26Xh
zKJ9<c!s?x`#KXby*tkF5i3&hPNKaTi7#P9gCj%E$GlENg>LP<oTD-9~*|$QLwjQFJ
z0H9ug^)f$Y7qH*}l+XWF|A#wl0nZp_V8V?mETplySvP93=+OeWk+VESld|qS$W<r_
z*uxK#ETa6C&u6_Km|fo2K~qziMF)Cj8Ybptl7w9&KQ^SAGph3GXkJ`V$ozE@>_s}k
zyHbDP(rh;mM*avL4B?4hJGZLUw5_S#+T^5I>ogzwxPz==r~$kPw}z$-1V!TbJe*(V
zUgG}@&_M7O^?Ub4ietj(Uw?H?NVy>Cv+mySZsdN9vrG{Wi|i1ip->xzRd*`*M|R>H
z8XxU`O4Z6sEH@(3#y|U|1r&(VW-9x$X^>x4k|0PfT-b?Ad401Kymmg)udXlk56n>T
zE6C~2VQSRr?Vir^{)Kw#C8?%<KckQeUUGtb;#mX(KSj!dOK}!FwQP4M>AI9@*`NUg
zflXDP@Up3*Y~%xQ^iDc70@J3=P+$p6uxo10>K2pM)zpa(HdZYpmo7sN85*swsYPmL
z$7hF3Pw4+ul~Td$HvjRKT=G!Kn)Rx>ry@Tu(ui6QJnPq$RQ{zh({yH?g`bhwQL0@-
z<=Y={ZTRu)ZJ*N6M2^68*%^Xgrc;<<&kjW1)0Zf`tP3lNIlJR!pYv#BjIOfd{;4t$
z`WGF-lkAy%qHfyBtvl`?CmFXiIFPrpIyhOys7BM@LNO_c*+$=lmPpn!hHTQ9<7Kal
zr_qij6+jsvfvvmJs<Jy?rT1Ghk^9a%_Xg%Wa)fKchBSlnkVntIbGOyt!vay{GZU|Z
zvB&+;9?S-`c0dn4A3m;sDJ7<$k={)#^WgZAjD6FUUjcaA+w0t-OH!UHEkx;$_`{6<
zel-=39Fz*1V@i?0t@-4jaP_rrZ9UoLGgr^J8@Y}ndE+)hKy|Z;Ar8q#=4+uE>qtZ9
zQ9(y3xhHDyvk{X&Y<hp?5wb9DiynI?=eT%q+34s)TP__!VJ;NOyq$Lodyu1BAA}RA
zbA!M2!U6RMyTwr|=HA>K&}wkfcKtG|wrkwGxyc679J6e?RgLE(kr!K!Tih|*u`Zf2
zu5B9zsv5rQyi?zdpFrCp1M9yoHCI8GuL?HCnKp=`w3}B<#gA!+!@s|!Ba=L=)S{&5
zRYHPknt)-L_dB=m11?Fr87&4(HhIVhTn%e+ZFN4+D@Q8s(q>}TToOi8gd@4)Kdwj=
z(I&??awu!`eK-Fmjzp<P_-4osmwDvUoB-AL`I%kwojim3xD!KP4Qe6ZkQy~B>EY{c
z7`n&#@?(hn1B&5NQ&Segh4*_{OILSyme92&lG8Ns55IjJ*7izd#R>J@`*meN_{uqG
zX68eW#@4DV^Mw<SsF-9__63ElUQ+qRpVFhX*EtINeva|uZ*f`Y%R$DK)pKsXN{#jC
z2iV_?ZF+^f5Zb~yc0cp=ex~WIu(Yt1j%F-bzTHja)$8j!`tj|Jc^J-da!+_odAFLk
zbM4LzNQ&aL-nv_bx7a9m7c#l9*g+i+_IVt2*P*}~(KZS!n%Th$|A-oY*}oPRDq`N1
zNGnEAB-H=zroNXm@W3xY>uXsQ5*<Nf16jhcZ!Vt4Jx*uZdh4n!QTHd1XYl=32uZ36
z6hN@ZPb~ojThUX4r*_|y@XocEd5s0U<-L`?g@5h@h-oh+;guCFGky&v#_Hv4@3jl+
zrscS_v#%L=|Fe(h(5XQbblAl7WvpjjXD{2bb+c~m;`o)olnuZzdGV9ZumTi3ue3P{
zir=7)dKEDpx@IXbsQgxb{iFKl1Tao29dc39kn2tK`aw0yF$kTb0ewp4mERL=_wD>b
zxNzg5v#m>Ck|SShI26APxz@LRllmYIGhxvvWib9dLF>m*>0D7!T5}Xk>)xZqlt)iR
ztIedgaD%fN^_?k<Y{7ebIt(u2o9S~4)aXMSWAhfK!_a=Wv8;sMjmhZm5%H%Q(^L+U
zuLRC|%1KA&XcHHOVBKwcsyh*Pbn`y)y`WqVq&S>1SL%(C9bZ~Sl4aJIMJJQxK^`wr
zL*A}htArvH3gb*Jb~hK^qva|OAXQ{q!%!QRZd`)SS!n|}ddDi480%{YOMF$d{_TJG
z*DlY(apl>=Eo@jY%o}?!**g)Wl;?9v64Pa<KuX{WIYH?EI4(-Q9r+CJQ*vgEUHV?N
z@sH6|O1N;BGMKOUc}dcW;4k}rT+^bIe=1Xry0)`x)Be-G6dFdFY?ljjrJ&T<bi?jW
z(IZt93Of0!P@?%W5qc^nxdf+`dO`*t;6eP_XVV4px_=?JTT3txE2R6PSil@sbnrdv
zK5WWDa!b3V5;qQ{?%BLGBf2p4{b<nmp0r?8<Z3s}4rlc~%>DiGFiQVw<%sYa>GM3k
z&C_jc?-1H=%QR=mO7s|Cp%d#@HE|kDqUP~jf23VCE6Y*26P~r?oZftozlhIugk#sB
z8ahu1`1%lP-{3~!AJ>FS81*f6b)ydXGsse?`0|>^#l1g92t9s@pPlyeN3$j*$U&sr
zz{(lsZK(Ghbn}#DSie@|<@i9}dCxv{&kjE1+e0>S;8sfWO@J#qI8fq|AeokhG?SP>
z(IID)4q1t-YGcPQS|yichFM!>#e>Q_2nUtb7Brm1B%{3V1?5ifRXQtkIYb3LJye{g
z*C&KbtGs)Q<0N~pVbboK`8^`=&rVKLWaG9iQT{bli@faGV^jBDnc>Ix&!d35crG69
zG_Ij4yr^Ltc<}H=3x}jvW+!8h?jlj{xD$_7?8?qw-gU|+At{q`CN<7Fjjt}cHsI7h
z=O4QlY+^p`Gs9YdQ^DSGOIVL1R>CqYKi?y_21$z&PMT_klfAtvl0+2ww^i4pYYrLw
zzBqd}XZ@0DU@DmM>A>bI_;Po(e6nkQ)~fEMGK~pdPbVRy)uy@2^pME0%4-J*Z&5nI
zOMPkb6dJ&G3_17T9H+0?5)0l!n`?_JaY6(c3&-3_K-dZ_ZlWIoQp<h=;o5>%xQ9YR
zKV_fQI9Z7ul^=)p+dGsL{)Yh67>cBK-|CRhqE1mmOhZLI2L&XQV>g<gCPkV=HdCl`
zRbH-bLu}=rYP)g;q8;ki9)h&sr)*WW^z_1ojK|rLkanMeG(JXE4r#>u-Jtg=&i}y^
z4YiAAGL3dEzONTc0C$3aZ9_u@Ma2;DM>!G?^vpcB!LVD$BO%huLrmT^F=!p+$d9*5
zMRXR7eO-^^zRWu|7=cymdgf0Vqx<9C*uLE*`NJY+xHH&vs3V&%Z=o#Cx-cP<!ydu{
z`2Tx*d%~hh_PVNQ<oa8CYiWEFC4a-dEgLM3)NDW)xk}`shlrToeJT3Jq8fD+Ol0eu
z>f+}&1W<xtH&y@178iz6O#CnDczO+5xJ@B#2_C-O)B0Z@U}ooRh|<DwuCnF_ulisz
zH@H}JZ{xSvE$eU9b+#3E8+89BcCZ5sRMUnV`S&r6^FYYEzl<-yd#VC7Do3B3h+u?g
zMi7R!h2;{~rcpxHRye4-r%faI&wi!bclz+)+l$X*`5AUc1&k8qo`(!}*;B{$W+|k3
zg<b$=W!eV;-G^B_-d*3o9*PndRFaP_(ci09`X*6|-5Q$<a0$ZE=<IF&#vx{LdnTM!
z3t0^B2Fdv4OAdfDl*GqK74;(}(h;MK&77FLMNB}lASy<$Ok5L%;EKQI@y<Q`kw2+~
zj{0S>#q-RuAB9cXOntpGW2}5UY07=#hueU!0HtHlBO)MW%o#Usj^*Q3)e70Oq}&u@
z?n7>JztoI~_w9CDwyx~(gg2VRzHFcHDl&a_@HzjN*){5Ubp`nm{dwAZSQIZ9T%Mb|
zY*+2tL^5z?uz9)b{pyaH-<#Z?Bj-Ht;qy&x_#+61J!crC9^-^^$U5UP&hpEO9^J8<
z0E--4t$uBACi|BbRqY_{+G=346oj}^{o4#vESR#(SF%0!@l3MeX>qUQ&h+L^uqAzl
zjLu;#=UTj=-EI%`wUY!tDTz!K74DtKjhD^8kNo<U*N7d{K`Q&|;@XRohtXK79)C_w
zsDbtP!aZMyV}r4H7S7M>NZ9X_+&1A4@ZaSSZpZpW6xKxa+9>4mK1c4i1|A4rmx}{7
zUj*7WDf~D)@+T??ur~CUgMM}7c`B`FWeF=Ja2Y?ma*X-5+pNJ<F&DpG{YJdZg!68L
zT#%osLEDL|9qy?Ln?)}XYrO&P55P!uTiN@5ZxjeSfAD--xpf0Pc<!CmOmuL4BR+p)
zmSmU8&v-i8AkXf-q4ouxJMk5gvd3DNiUzKeT2dNI(34-+9GdyH*qJr{vP%ju8u&c%
z>HGB2`Q%kny_duu7rlV#KxfRq!JeavR;H{wn)oFr?gdL0aq$?i7*3C}3|Nqgd_Y)1
zR^CVToB>$7qKYFbwzUfeig(Ydi{dUF%M+gH3p@(iJ{x8oH3`^UZwh9wE#G&W%!@ca
zP8AMYiv185jI~ES!(7#JFu#ezQ}(U*7l)@J!4L9|awZuysEcB@Ji^ttz_B=6{noX)
z&GowOj?(UDALpTu<j-qEC-=9AfARzNxP8B{_@e=aKui}H-!WaXK@kF%7xp;s&GxA8
zj9$Nr^`VR^^u*;`((Q<#qdVw(V@A0st!lw!@p(G&-&=6}!TKzU4C_+TbqvH@`|xO=
zhKlVe_`vdKAbxR#@%o~xPx6BhPeCwiGdK>I=A<~BH0ZsqJ}B8b5jGB!UMXemYZybT
zC!EJs*2(;~P-zsuYG}$(*`Y&tDF-sAlMFOtZyZXfb=u9?w@TQAUg$JT!p@$O>FdM9
zPY*Ry4}F8Y^&YlwEZMlU`X2j{H6hzDuPT$Ac)%@8%~AxbGhRSVsy#RD-O&>~`1B;S
z#<shQ0A9qFb<bm=<&&WmQ|h9SG8tT&IRl_d4thU&-#LAsZv}}NEj^J+M?<hRzKYQN
zY!n=m*AJM#sCf+F;wA`s>8vtY*b(#WM3N$u&+uU0V;(NWXg_{U#6&ZPCXO&q$5sP7
ziiKL#0s5ExB-PbZ%*TE-|M9`3{=|B&FiY1dtJCmz&E1Xu>zvY*Haw=4{?`+MmK}LS
zhVjRrhbP^4566L!wJ+2DK0ibf%)n0`d44FicfXfvgMulFw>(VWda{rZkDxuhewT)h
zu_!mx{;;T5TLz>!jhXWCD3Y0?7*tdlQgYW~LDXBWhV9=$GAz3DT34G$1&F)T3Ut5|
zVZ@k~38(b(-7X;}%=O|sayj_I@39`zHdM{jbxE_NPK7{9lRE?$8mz{thgRwcnaQiH
zOI50$1MTk4p!osRsO2?3LHn}@?gx-85_x?H@#SkSz$d@meRf&QYW4(UiFM;&TGIR<
z1a#rf%2Z35fbGDXqf0%uie@T5$2P^qkiyf}+v?C7GcktRZH-6f=m+~V{4Cyb>y<sI
z^CS;$;X=79*>3=M=Uys~sjK(5s}<}(#B#3#dy6^_rp}R?vUtInPB!jdz8^gLc%w*6
zeUT|_-Nxmv00Btcas5F<Yk){DZ*;!A--t~Dpfpe_`sbhd+=!T5B}wE;)d@A~2cOLu
zz_(^CG0xTuK+Rp>M<|>Vq1IBE62)8)bdB8eZ`jK=YR}toJhgP+c5gO!qt?BCe)N6<
zO=rBYnU&Et9U2&V)npsG(3mT~0y)764fBz}LEeu1UdQFf+;dSS)Ktn=Fg@4VpI1Sz
zNV&|qbhBO(8fN9l!usC+_++dY(s_>aV|ESO1Pp`o7xK}ZF@MQgG4sf>J+0_3471QU
z`(R=Hf2}<rACt!tpt5TBZ-*B5qp=eBT~WV#k9Dc-dkuH{JX(3g)|tBNC@RndBNzVY
z_IRoh*HoEcXiJ-#f`0cN0?yLXm3l1(aw#<VEw60UOdcrwWSm0Y1zNQQvZlqqj~jhm
z3vSfxl$A+ztZmB}U!1;mq)cqc%H_n?8~<5thN8rd=)M?piQRg5BKkSrg4EPqlKap1
zVTJK&=HvD=+wfy7FYzck(eA6Tj2p4Wc7kou<rl(mzdiZ95)W?lEdS&)b)C<LCXS1u
zr0X1v-nrP~W;I0Wag>~2D18pq3$APw!}S&@gNE&+5N_x0=4K*8I;lLd#6i6E?JuxV
zsncQPs=MdAK^WMvgDvDtyC}3=Z3Q3#%W#RAh0?s1`H$t>fkIC6wK6*nu~R%>fi~Yg
z{*NAA2z|e;oo<qeuf`SQQ7d$d0Xb6{Qy6N&8aZkU*r>Upu2f=mA#c3xvVIXffyNEK
zB`tAF9;g-zsKd4^JKck!9My5_{CNl$n3^LgRjh&bMCQH|PVnz04%0Y+(h5{i%LiV#
z+LGXGo1|m>Kb$AQO-m^J{4>RzMlrfixa~m+t6m|rK0*%h!Z`gSA4Rct=_b0WYYwgd
z@93wak5JrGP8}aN{!gSidU*-Yvyt-lz?@4=(@x>K0m((LY}U^@7Aha0lMau<k?oD2
zk2jg|Ve4@tU+yDdX`RDYAcGM!SaCw|_)@*v+3y#Khlq&Z%O<hAE^^giai;=Ded3KF
z{y#1UNPW`9$2m?nZg3=CA-CE$q8ed_o&RH2<$<D#!h--%egF-Jq(EB7w8551=$4Lj
zK6Ar8*2cq#_t?js^B2{c)O|G(IkUcAnoz^|JKFm1K@IP!3u|}X%==9L&H)d>p589h
zI^9|{Ah7isS7R;JtRD=5B~B=q-M>zaWk(GDb&e>WuMt-?>r^cBqht&)U?*71cb<Aa
zU!VdUiTw%ke<lKLb7bOcl8jYM?!ON=nl8->+qN()J*@6o<q|v~`$d@^OShQnvp?h3
z(AKf(FKxq7=<!O%*cPHmn{0lj9o)cH8~w6dw&hsvi8s<ovpK68O!tDE{>zT*+gM=Z
zv1~Chb*5>{XxON|r9_v89AR$BqKrQ&LAzKEpG?h?T##zft^M7XXg;$rw>}*xvfH@Y
zJ@^)+vT%!I3At+3G{z*8hz`wT;d#oj%Ru<WE4=u0%LAXL!wODCg>ILip=y8TbC>a1
z@^j@gHScp5)9e`!*9021h%A^53jP&DUJtOhtp)_rd7&J`hX=B>Halw^qcd^_wIk3o
zQIN?V#Bk{|gOguLMt#s1ZQs^)zDFT$jySs+8#PCw$~ZtS%2-q_F>`z~34|ge{<p1~
z(7XO?Fy}tMp<bSXi0<cxXx}S^gzKt7B)yEpKl)Mz7EYGWjV~$TF*~5fmCe7+{C4wE
zdb3r?LY#(<u4G9RK=U&|A3F68V}_QI*Y7)dC#bT5qmyi*Rd2xRC0MihAr^B1Xj^je
z$rb!#Iew2E{7WgO*u9D#K5N|qEdjn9c1+sB8bY}qT4&)QltLyKQfF#H@D-k$1vv{L
zIZya|ju5^9M;uKP`g(I~iuok4=-TiZpw*V6vn~Ij$E|>wp`q0BuUPVyyfw8oPamD2
z<;z7^zi2m%Glv0u!eqEupuqb7ycxhH0I>K~&8_fL9_E*bc{_Dm=j`8%L0djEAEYDG
z`tOvvq4k)|lm9&O%*=9{y&Joxs$F25z+?o`GHN;)pz*vE3B9wwK^(@B>+SA8p#`q^
z{#*>ZaF^9yge9d-QY_*q5e^lb9PK*KIe40b1nmF+Du40$y{IMK#0)N~s|&BOEl5h5
z$v9o_Lf7W;8@pU*%2CH`{H<YNmM|X2$vKhYA^MlU>-!z0O;&z?DBYojcTsc6b*)<K
zK+qX*rXJaav9u35c*S!Nl=x(Wm@_1P>ltJs`79R8!T(&R!k|)sm~1YTG9PAg?Ji*n
zo4~B~e3@!3FM?jFDx7BHnNL7KFF$Dd4XIe&4`-S{(a-zMIQ*T+EYhXDdyANt=yUn=
zHI0`2JnvJ#g}NNoBV~{;QHIYZ^brM8X^}N0^bZtEb^4M@#M^VD<(<A{mU+E>Me>XL
z`20o7EN-U3zTE0PJLj+3cJJ#H8b+<jBeRT1Z82<fVrN_#>6QziXm;2tb0d>)Vh)F4
z+7MFN65gq-$MZQTAH8%G$ki3hk_-GW#14w+$7WA-jc}6Ghmm2iXEiOa{928E>!omn
z*O_D9qIZS#v6FYZ^|tj%2&{!I)4TG4_zmd~ss5!smSLuO*gfyp_pSTsGQVR6kF9tJ
zHwVwp*U5&Y<$(kM2qQu(&tQ%FDE*iwpjgvBw_&-5g%HwDiV*}Q``BG{<`a_y1>U<P
zAuVWcb|wNVnw|5PnUp83FH1o6v~`|trL%?F2n(}tt>%K?(W7r!ertw!?c>!Jt}iUG
z%^<mW>?R^PnKe#PL5nVJ^3Y|h%Q7GiOYC2ny;d<ObaYtT=A>#c8AaPHG|Ki^z@`vM
zaL*{0i8No~;)cTkn)o5ym>AJ#eT5io)kiaiNMNP}(d{Zuf!ax5^1Z*mv%zBvQgC2&
z`psUc3?-BqyER!dQB{6Uq78uxWvtQTkjeaXxSih<7N?_S?;!fu+o?mqkar{Y>R+?4
zBS7*6;uacbsdc;8mR*;3D`dHCHI{gJc>fdeFIrSeq+=x(K+OpGR}Xx59E*shXBH`l
z*B=iG^IzWU!I;iB77cC{5kzCijIHU71k(q7%i0+g<RHP&lu|LV-SkzKApVBTGMQ=j
zQ|9>(V;cJQ1NG7XR3N(-Sm7d1jdf9lBNLsxGPLxfpbWhxBC8q64bNSF2hb$FJDfm^
z%apyOk_zbxH}D#h`z;|8@bB|L`QK}PMlC7kCP2@;dG=5CMT2?|Jv)uveaFx52q>};
zti?&ehq;bWk8>789XN0`G?$i3W19&B)&a|z0$7&S0yVvps&wPDRqO*IdPu=p4&hhv
zqna{T1ebr#CJ>SpB9b{Q=m%7KAZ6jA)xLj^GZ}ew+y)vcp~SSTT^dxTOyl|dSPN2q
zWr)R+XB*<Jp*s(|yg68+u*eg)eLF&2dQFrgiq`v{NIECV;yGxX>@tL-Zlrm?fnWPN
z-sqFtC6U-al5TPqv5C=+Pg5*R>)yj)MqHxgRMDUr4ZHF|Jhr)WkJzhNJWD%{PYM{{
zE$RK%>blzxPfxSe8~4upAx{g1o`u#P%Kt>^GC~2(P%K1RO6_J9-tiuFE9YX*xIi-k
zIn?o5?yY965^;i`$u28()7Y_&v3KyT5q<xVksIxbnY9SjO|}murI{hsbS=PX+RY5l
z6WDFc`v?lS%lp*yy?IuwGniisWX_<YK_oItadd-xl3{&?awY89+y98;2<|F-y8PSz
z5Ga!=Hn82sgGDFO7~b8{NL=HyXKBFQ6<6iDP&LV(45wBgf;MQ8O{T6VgbD!LS+OYS
zPFRU^)rJ_%zGq<BKoZ9v!K2dh3d9Cd$Y1fCl00yAlBdg1V8y3oMxi#%ku+qR9hkwE
zkK$2$6F6sAb%saelkIqff`533JtF9fcnd+Ea6*8e1GSdK3|CQ=pJM;BmEg~ct{U?d
z)*r#C%kZQyyGYcSzqU)TM@5fLFMCDtB0lozI=fb$aPjPe)8+hEpM?@ATDxqBKtYt}
z6p*^=3L&7HW<g_(wokpl!~pe#u%S`e?fwt)48bdR{)l^8wlJ>o@26UdP-6b^iAl;I
zy`+fOrh|E?a!N))?zs8J(39NP2xF_jff<S-7f|2alWdTbwd5J!*;E>^VrkH^y9SRL
zWA>7mzijL3*ans?H8$k3p1T1Xlr}@bOnCrO1!e6q5#QMd<n8eK-F71@W^Aij9vgJs
z2pP!Ff0Xl>^a8K!_zrd#I=08^mQ>lT=y&zh7KU3=#Fk0{#p;RlfUOng_7_Y0Y`lVk
zkp|n#AQ7D#iGXhxv-gc0iLggfeJLcrp7Li!4K=Cqb}*A63)?#r%x<4MB6_oP^WO!x
z&}RNxgYNB>TT2z4suX84@x=Ny`*d~ksTSkb&8zbh+VhSb+qA|>Hc5(E{s569{?Nvd
z1~K~R<+I}2Go}i5GV5Qnk8Pzx_8xa@_*(H^;U^b4fjMj~$)5Nnre91b?4fJ-$jI%R
zjmT$b+R}Jj@l3hZ({N_ZoFHUsNW$lMHhtp^oArF}*~=y}pKT=sDwT$joBGcJoeQ(X
z=<uq8U1`AOOuBglPsI>Qx)7s!<AoUcw=ST<-qS;^v{_jgME!)xF~Sxzu^_nW!??IG
zQH>}+zfoUtRb=y~mS#D9(G);V`J<ZOh@y^J#i;Z0OB0WgMz6OB<jn#-{QzXPrk$%=
z_a$U`wZk`>Z8-tww$Jd?4*&{%&6*}Tf>m<jj-Ws9yuz+$u(jP)mOH~Oz}?@0NQh>)
z*0@sTb$>(7XwU_PUS8o0YXBXNn0r6Pu1F9t0z>F8raey7v^<KZt+xgDYU5LNane<E
z!Rvd4=GFi1^Z5Jgz|?)z=f2>}qZepz=6Sm@?-nqRqo|_YRi~=xY=VB16I6L!KiXmJ
z)jTo14PKySkd#P1e!vPy(0*c+M;kuzs$I<MQ4vvogn-G6Y#Yf(MDqOYB!P708X59W
z3QVTV;fxB|6MGmgxp8tfhPh*isOZo{p~V(~D*sI*Q}wZp-hX0M#ocuVyBUF|tV9NV
zzg-dpM||#tu0uR*=qpBu+2{8qVn-kE!>41|N}_H@RBre$$}(TwEvRLBqkXKB39TKh
z)4Rop!;r-gk2#C6yuurqSYcG8FDKrqZca(I&)(De#=sZKZQ3ow2<|r%ZP7JRJ_Gf&
zWE1o5n{;Q4lbmYUKJt0LzpjF!q+A6~pb;Ai1ToJ1k6F5y1beQ$F*fbooh~a&9vA|~
z0>Y>^Aq55=J<sz_xg#!Mv6{m6d9m^IF`gzh=g1@=LBz#v=gm3K5n;-_@(pF8+IlVR
z><k@XL6)82Y3CM^)^)b~0)<M@m0F~_YFPU(I^>RA0*c!Ela}nOSc_)}`c;kekLM%5
zZb`v5u9SKqXXX3Xq-f4({C?M8W+Ki1R2^~le)XAY4fuC0Qmm9fM$93;HkJ!3iZ$w(
zy~xtKJx2_WlXY?zNi}v8(?v%;j+2Zo!jTcvk$ZbmmUlg$_Yu=oe(WQw?2d<T1AwwT
z9UQLv6-w>Gqe6+gzAY<I>4t{5(POXzSzaKufc{JKSGk;NRFJoUsyaDv9B}a$tgr*8
z=>E%(DLPLJ?O~_1`YZmA1J$G>GP++(ybw7EreZpalJiPCd?C}j9`x>wL+l8v&-JkZ
zgAevK%SKKHBRHP<L22h|wXihXa#fKQx+xop;kkvq#sxgn&O?dQvs^rkjHyJ<Si&Vw
zUryxuVK|O7`2d+eWx_NrI$Opvy`9+XYgy7aARvR+Arz{fdF;Jmd;*UlyX?@!t6zQI
z3Hl4hG8_}F&c$`Otz#V|r{T~br#=~x0_|Ts9BxWElb$}@P(=mM_fBuPw|cq4{UuHL
zb#z_VcHz<|O`1xPe)ib)d#z5z0fK46iv82gk>Jt!haY#D`Y{VrElmwzyn4CemR|$J
zhW3@qatn4Es@2ki-J9&CT7!C>Eh9Tcb<Gs<?=T5aSC$k22iDQeYv@G#%lM#oph!+<
zjq#HGi7l-YO-oTLf3A?e=43_G&>S<N{|#%!0rk*Zy5Yy1U~vuR9rL@sjWo(reSPe!
z{VeC`-!9|NYp#GJEM8Ys>5|lLd=%zjIc*f^vfwttCd8+F6Za~dm-YRtvEuF-PqMWq
zCQmQ3YYDJp2rWXl0+VY<iV;dpM<b|$qzq~yWg1bYyL)RNg^|oR-4QpN-fEDZoc{=i
zACiaPNBA<?Eo)Jg4Thv7ouUH3P~Cfaig|vEXs8&wF#&ts%Z7mDMWeu50!!!niFk%n
zWph<P*=`(qf+s3fS~#=k%N*2tQsfQXAoGzYKmK&K!YcM9<Mb!oDCnInil3lR!&r5f
z4J4;RYa5#}%M)Z;UFG-hhfU&U&lG#dJkA}D4pK#;u&F|NAZ?cnubIXn_9`L`%yK>s
zvLk9qd%+A4Cm#_Ci^#}kMQy!|_T=Rm-t3RD<_h|di!{){3<f@Zyw<k&^h8~klEf$H
z5?fbh${|H6BVFbcG#A?GW?o<vk<j9OKj0#Kd$(6LAx%WK#vM_})ZFCwMbH*kMz_IK
z%C36yYt1r3Q050ymUWl8*+-7c06g_B&TbI>M9`40wEEQO?bjppaWlGKWxj7?ly9$}
zou3KXelgLLP6mrqrp22r%2lnLt~JE6HVzK;`m#Xuurs#?-B_?1O+CZqbCddWKkv<y
zfpMDpHv4%FlE*=N5|i%8<z*vGTmANYOcqco6xJU2cq*Q^?^41WlO9%3eH(=<iB=YF
zx&Mv%>1Ck0GQgGU3^j|4cS3{U^y9ZKk2P&X=1sCuO%5gPL5ER8<NN(m`m`Z%UpGzV
z85sXuP;JCzY_UAzD#RYJg|kq3Bk`k*6qTvdq|gcF5(z(OI6V)1t0;09J9N464x5Hw
zom3j$Zb~wX8U#x?tVcV;V0|vsm?kDlM#KsO1@&%ZIQ;Ako&o8Yxq9a^4HwBh%A(#A
zl&eyEss!e&C%3x<UeHI#6hvVo34FkfJQC!;AZ&efh|<F7$ez&C+UnP^yeu|!bR0h4
z9(f);0b!@3q8R~FFdo)rqm!I<(D@UPeQFe>oN&p0)YGr#>gBTR)NWj}wOakcl1@ku
zD?I4hebcsGew$QCr%i_eRia(3fo4u0ZEP!Lhn8hO;_U*onv{Y(tVr`L#TB$?%8z8(
zpf@d#O@?Pak7wFFKK*#|^n@kYZrkuly_Q4*j)VB2u!wPR!pkqN+SIW3<zELCO?>z9
zbxov%D27HMC^SnJEk;jeY_E(z)ssUR2X7IG=sq4qAqKH(F(RVGaz}*}!9|1oNrR+>
zoF*4tS)2&9cIMODP8|5lL6y1}{2y7}E*sskeMXEAmtcc{Xs5hyr3ycWQbt*ZmzVAA
z>MF*cm1w>K71Hi%nfw|h%_H+YdmvhgOh9WD7<|?mY}}Q<8=6R$jUY8UZ_VP(S77}P
z=$b0c@J1M}OMJR4EDkyoYZ-JW@!h!R&$+P5Huv`(v}FeHm7`HAgGZ$E`Ke|syqftw
zP8&XylO>DbF`4VS9ZNgph=hvjDw{2_R~Hyfou)H>m=Bx&k<2$Hp76!xDz+N9KzcfT
zWJ_?^E1Uz{TzfZ2P0#!Ol)ZC9VV4Z40d2^d-gGWK=Gk}Hq#hV1lQtBZ=EaQOF;0EU
zE`Yq8pmXcHOU3sqmf1n4Df88;zhO5SgYb$Y`&oS(t}dIh*mF=<MZ$&3fV=rLhz@lu
zaUn03Ih}4`O-N(k_am9{sZsFL1>o_TzHByZT$l;KDrvIbi0lQIobfQLpoeUqEb_Kg
z;!l;9IM9EUzxuh-WulQ<Wnbg<w&7vBWP_#xpiy0=bEV%pRHFV_Og%r_)YR0B##ksX
zEr`e6<W_1fts83O<7YG+TgW%-C!Yl&PH8}Aj~BmLA``~ef9*Xx=D=;?Tw1Of2)Qa6
zmRjAMLa$6>60yA_TKr9iD4^gPH+hl%ie6?skobdZ3QbZQ5mAFVk<qKbHM@?Uh9`mO
zTY|RN*wNNo-|l|I4~=Lj83W`PrO!;JL4uM{C#r_Ybp2m_YsQZEZBHP^l#@<TqhCV%
z%LK14UZ%*muHZdWtIT>d3}A@`q{Flnw3>&1M7ODClFYHee^b<*@SC?ITgEfYW=LA9
z<O%PP`_IbiZd7Sb#$&wQ0hrYN^B6EQaRx|gO*QeK&2c>*XrZGf`@?mxIAuhvIMtBa
z-?6{>y7uEliTp4Fd{MqFL=#`RJ&+wSQwnqcQI%MA;%9ASw%9mQUs5sHazODQb#96(
z;n6ETHv))&gUx|Qt18Hx0(8&sO{A<V{!}xYo%jY$lb83wbi?}f?!{wXH}Ax_Zl1Q@
zjI9iwVCN55)m-g1<n={ji@Z?XOAhwX`CAYk$u5d>CNGnUwbXS}JWeC4$A1WV5RCO0
z2DoZrD1_$5c(F(0JyGYy!rohCE{_tg<wg+TeV&--Ka0^R?3lVY2}!(2X1?Auphzb_
z15^prM;q<&KYs)q<)gh|iN6?7J{8pYDMrXVH<X_*+qVJ&aEFGTv%o~yBL}qIPwBj`
zd$qcDd`)gIwko0CA~3%roWnLTgHkZgeO_8h*4UfmMVIc`^kwIpT{jC{jG6SccW+@0
zL{7t1jEAiR7dBm|C7E5nAX9NN1%YS0Y&OaL4&EAz*nkNjmP5lD={wvR-O-T|f3SyA
zi>$xtmlU;UWB4D*_guNOuD698M9KNZ6z!r0(Bd>*^<^D~px1P1uS1fVOH(`z#4PqA
zdd^OKoR%!KB^*8-Q79d5SmateGrhnbUh_7uAGz|#_RcwYbA=4qz#w6^KI1H}P?RxK
zV0US&#b7jTz}?jQZ4Hvmy=`qme!DLk%Mo8LG4}2!D+&&nVcV^VZ_}rGS~PyWV&3ac
zj!}i^*9IVN@;vSWUdRQbtV&mJE2@i?xE(dA_g{v(G6Ad!u7mb84Zy_!=2#NW_1;H<
zJbs>{!lJEeC;fP8mWK5N>`}Y$j&>4N>`920Dr>bL!LXOMeIEQR)+g?~)MP?Xa)3}M
zIn2<JuTrHAhR)ZX#*;GsfVSFbG@`uhT!^C(Yg&*WQ<JA>;|i~Y@l)7|uvmB%Mp8in
zgKeBPGmCx3{~ldk9~c<Cv<v0K#+{|_HGe2oatv>C#DwjDfrEibiwUb?Q9k`_iSJ(H
z=2Wu?{7yDj8aDHwPv_C-;sM^pW=y7EGhl~ARq?x1qEEk+T_@TqW>I3H4aDU>PL$b2
zL>20WKqYzv5gI0geKVPWEpE)>oN8LV=h=90cbBp`@T0ScP9`M@Vpt@@S`YNw!X=WG
z#wO*jnz+3FB3m?cO@P>z2gG=`X}5u~&zM!SUpzCmzNNBjlm3h?2jz?l&_JbhGwsny
zufH|lgs`?DF)Bj|hd~K^{B!ugQg7#A1P^X!dd&P?We6d>w4<=7NG@%f2<WalSG|25
zBAaJD<P`QEIzy^qgzkB<dBOW%g)BcUdHO4sfAz;yF~AECNq=+n-eLCP!B1W`ag_dQ
zp6w!zK}3afbW$|}jmr8JIz+}KGpbTA;(~6nRbi;6J9#iNojXhD#+c%KO|LpUP{*1*
zMK-&Gc>Zy3s}-TK8~R?qJd)GaOIb-B-Ax}UzlLstxJw932J2{fxkn(v$YESVgQ}x+
z2qroOy6Mgj_2))fi``fw8Ghr;qI4elj~?a*T_3xKnJ6-)U#EuS_~lkKv|UTP@(HZ4
zvrhh6U~8oXE0<QQ?qKeISCc+-A1GiMF{w^qfXTMuPFS8MnQb361P%^0k6iAsb|>S(
zK>vmG9G|nnZ?Dz(&Gdzb!ZI`4G^AOSSE09g@n-k=8$gO!z7QMaV7-Z3{h!a<er*4M
zfh*owqG69!TU)$uqOLx9<O1-D{`kZ%{$c4x^?P`1yO<$+J)3M&jLTC*RKdgW$2m;5
z#{E=qp`TvcYfj>n{iEDKV(HanCXiPYyRNx)@7g_@iG|O_lX=^i_Zc)3a087*Ab_Uw
z=K-Sc6%jUFZr+2$cwq#mcbZ#|vabfXWXFC9f=G`18sY)mBmDENTev7CGQVb$Of1_i
zvT<3MPkT2@ob-*2B{2%Tr!#Kynm*n!UsQ<hP@g4rjgB@I*72VnqS(R9Kv(F^;ldTc
z4$??>541v@zqAN@Jx&jUI>9@^s3#O@;m$CWcM%V8WtQqE`A0gAutxZ_#<xckz|NUz
zX=K2{*n|~!UBQ0GRg4A|!bUy3C21P^$s1s%H{>s1ugs}b=!z+&HW-!FRGth9#xk}c
zBt7WY`^1bvK^>>pyG4)Tu2Rx3cmxQ(+t&3IQPb>EEdKc`F<vbfA__8SA~7vz{~8<x
zJ<7iu^=~)<J*<IhOikf8nx)@}tpLC_|93!+Gt`i)y2OqzwNmUHrxB&iT}CB)(_mpR
z#jT$^>w(}$Nv`%kyZa@KhtV`F-3r3%s07i_a-|soSd0KP<8YFHCLyFZn++=$b<^L{
z)dzLFb7urTlLA13Pl_kq5~{{>!l05pts`<L;ENCHi=SnR)_5nr<>+axmw)7HfNNy)
z=jhty-2%-zAZmvPyOf^6MH8LhUj$xv^LFnoiz1_AK6Qrz`<@*~^5JQ(xnMla@#O$Y
zI(!&ZUlSP=rJhQDht4jC0FUFSoCfc3iFsmVLEF5}^y}U;jWl$1_M#e#$<t7>_x&L%
zI~MKl^S;}`gRz(HNnL0pS}k^zaFUxS2^}d*nwm}3W)c+@QE&#=_y`hUkmYsD$^t(c
zLbr?33aoa$80+IVdsnFDl&J&Kf0CRT7Y)jS&WW<CnoC)@-3*k(IP85tm?dOm;DQ&u
z@J8IhJeY85RjGcZT|sV!e`mTJX#6Qm_+fEg3CWg~tE;0%46pJPFCAx)lqK<@VbR5<
zdlv{Qpn^maY+t)3m#dca^s-(2T*Q~|fpc!l&7`e_S7S^(-Rbj#|L_)jTF2hrK7Go<
zO7(BcMXuZ&sPiJ0m)$y#cX-?mG<cyDvBmHIM=8qoWhEWzX4v{V&WBKVW0rB3FF)4P
z&*ByvQ2g>^_KHC~9fuZI_WNG_grC=f+gl$Kz58=T;lWq8_Di#DB)R-~GfGsz`|93F
z!W9{Lz9vBN^})4!h!@%_29Yif!>U@VdhvfJ-(u90f9?c7`x(N~5fG!r*Y7mvsP+8p
zW+lkZz8Num3(3Qq@-$x3YhLo~HEvH=MGJ$Myu{VY+pFKk9VeOn4Xu;<M3hUp*3`5{
z4>Zqh*fIAPB{DSwd{9!;GSi`-)R4PZJwvk%SG1SUebNOP(26In@Ko&o0fs?&z66!U
z+LRN0s=y%S;Ny?OV=-8z(b4zryN>Ip&#0C&X2-{vqLxWyF+=6q&>_8-g4Bu0u(6uK
zbitUqPDg8h?D&T<aq|6i(%4^=Zm|z{EHbzDjV5dyUTLp*Be8~_F{t2%Nh2LUUF3#^
z-1p?lg8`!g-ddE(r7cW`&I`1bQa1O@LriAqnwDA9@<0B+{xQ#gc*4)W{5khO{D|f8
z3G;eJkryoITQZZgTpSPa4P{<pT}!iS*lu@h*IN{tZrib1t!Ubo-gY!?OV@QQjuw3Q
z;RoEicZPk{Q&kmBvlDp|l@Z9$+KEPA*Aar@-u?ISs>GpDS&k(TbYy0=76Ibef8q@B
z#z|>tueY#td;1?8d-rc{&0vRpOer-M15d{c6WCbSRa^_vy^wl@yz~jU`Vj+KP1|5%
z*!Z!Jr{&QDy%X6Jqo?3)!fPaqHvxBvP18eL?6>Jupg6Txs;#*X<R-$mDh8cT#^zKD
zZ~870;Ca*d`$M>5vn(Uaa`HlI!xni#X7ZtJ{3Nbd2^x*U+AW*)I+lq-#Ki+R*T<x@
zp7qT&H#aMs>!X1)o<}Q%g}t`ZT4930BPGSAb3q7_G~v9%1vdbOL&kX;`<R6=xE@s)
zgW#3c=qwu)`-`dqrO2z2qoZT0swUGZ3s~LJ9fc`25&Q|m7T1WBsyylcFa>LegX!JC
z`P)W@*!fA<@^HY8^SSqd-LB!u56`$fzr;CD7TY7wGcK<#dHL!EUAv?2o0wS{rK6mc
zy4jhbEDP#-#=U!|1OiW=Jf$cL>RF9*j_d1d6!Kn2xB1^;3;RIz@4nq_;11^vcfA)c
zNik`DN)5?iqRKKB5@Y=4H<ae(^l$GyCKFFnU00$Xd31zU8P@d>5F;xuIXk=0r=Nbt
zfBc{PA9(-$4`og3ESvRK;^?B#^sFw}Y*uVmYnp9?(HRO)@4@w+uI&kK^mb$><7heO
z)ytPW`QbZmuFfNH_2fm#r=NVx)x~R+_jFCis~4{+%be3ijoxkeAv7GlcOTbmD3qdC
z8K;XQW_gam5$qPN0^PdB{{NZ#&nHWg?8+1Tsj9KZoQ4O>j3iZ+SqufBz`7d^>1k;)
z&4#S(&g_>J+5H1oc4uUEF$BBQJ&kT)nxP9-C;$bZDrpc8=Z_jy)%|c)?b-A2$jm@h
zi>#%1dbs<0_RQ2&uiks^x#zY^e7lDI6Y|j^-i&EoLeo~%n<dS9iL1o1GrZZ#ctgZa
z<AcJ#L~lMc@3ka}<A8>H+A`G{#(QuIM}pNUD1%e4!n2B?(WNvVi}TQVCyt$0WZsc$
zO<Pu6eD;)L?*KiTk`@zemiF3p;@~*Zq|@7UTWpX%+zehRra*;zdv(Lub=+=`$8Kcw
zy02Hyr$iqrZl+pNuTdW`M!H%B4CggYr1K)MNY{XIq{@?}3APe#Ic+3;);UK$%FtQL
zC`)MDmRu<w<q!Gpx4*;ZfAurYU!0PUW+<mfz2<6h!Fn<y(Fs@SB^o8djtV;0;ao@E
zG_2Q4uCA}B>zd7G&3d)s^Jh=_tB*h7$&*L?*Z+tAk|$4|;FV{8e~;;Og2$0JIcb`M
z+8~va8m$uYEMqb`0F|S(!Q#-`%e_<$5#Q*cY5$7J-gzE7Trb3Nz6)hy_^Mma?G=Yz
zcCI@=Z}+{sXWqU?y=p}9+q9?}+>`&}&8t8Bkmt6otQ?96N9WC>=&YyhJXyLM>BqFv
z*4+&JF>C2MFj|x7gqSY8D;>IvAa2KD_`2K9UHm?zQ4G&{+o`+j`Mn9Z=XHl{yar<u
z5|fJXbsyN>HUMFq2ExOAOTQl3*g9k}eP`{KEUwgc0~|WfcXiTG5>*0`gS3T6WZHyz
zs<%#E+ctDowAQjblN6Vh6uclzT6#KrY1!8cr)o`-XB0(FlBBn2E4Ri>?cPfqNz|_0
zde!pBo4Rua_m1;6B<TlzXMPTFxLe0C(faE7<6%3?dbQ@+v*(g_R)TJ3Sw>w~td?`u
zs}<gdKHOpM_r)_>P^KhHI66K8h42;IhLd~uKq)rs&6e96{}*dy{+jc4n@QTWfv*A=
zyX#RtUdGlrznwSSUbnA~oC=QH@JzS={$&=+_kkiPvz_-^3*ww=Lz*V!c}AKFN81=f
zmgX3p;~YGB@-82I_)Q)^ev2&2`(l=|s(NJ9Xc5jX%Z*Tlyc4>K7o@svTH2-)DJUh~
zK}MqyRaJ6%amM-CDRomxL0^(F8IL*G-@_p2VUcH~S;FPnDVwfk+0@+7P!&ME#$Yj-
zVU(q0N};_C_P7?=qq@V*J-!2G9+kByJ)&zjH1(Rc+TfZx6i&QZM&gYeI~{m$3XOIO
zgGc)o<x57Y<VZK{YflzhWr+NyG`>xgLMe~&9__<6r&Leb2=N^ru-?-GsR<yu6NQ&*
zzTxb%XQX?3<l_-(VK9ltBq_#eoRw%_D<f$wFD<E@byBf@YjOdoAOZ}7D4h5=7#fvW
z7onoj@~@(I;-H2_5x@GA^*1VGs6r}spdhcb0`2icDTHm4cwncGVNCXUq@cx^lwvfZ
zY}e=nl0@M`Wo+v@3<-HrV3L$_wIRt;zV)5&qC8xkU-GNJ@#`p)aR1>WKL7j~M-SiP
z`NyB4k`$ak(<&1l-lnRls*OaodCq7&rt2F1>cumjK6}Q8AHL7u`MZCIuYLFpOi|Fe
zj+2AOWNE@?Q-X)KlgKI-oOE4BF*{(ke?;fNVn~dpZ8w-81LC^CHwb8Nbb{m4+By*o
ze%>|xO8J5DyFJSX;b(B-hR$B^*xh<x&xZ=9xE)d7g;>E>&(Qm*`lO|xs5g<j)ljc<
zUc@-szRzgv?fU8s3i;i7Pv@X>ko4);EhRyaIw@xo{WT>Bk#%UgfV{`sbtvr-`Hm>i
zyL}!z8Kdy?tKWCm;9XA5(77Al_tv@m`5n6`Oq!SklO%n>HB2er4))yO8v6UtTG}5f
zF?MR8q%LM5Xl@SHc07lqRfokWgU^lc>$#QTe45^=lWUOk)U^~ube(1FM`Ve#1&oDS
zNg`W-)*ABdqET95(o}k*CZ<RFAkN8%gcEmaxmMyP1Us94w3l%{_s&RkPH#W&up21F
zmPWC?f`R9J=b?uG^W1uE9B}z<T{kQiOJ2Tw*?+dthAdCHy1rz!T2fVOoU@_7>o6E0
zILA^wdvJKbav?3ly!RX*A2XlNrI)N5&fDQV{{rSMys|Ft`ggoPLXhX2RIU%#-4|yG
z?r<)`IN$!u7om0WbA3dPweUcFxE5_RCK2AIa~@+-k~C#9-UAZeedm3?_Wn26KR5)h
zXgdKKs;Z`{DvVa>)KHftO(oQ!7(#T>sjFMM)=ATYu;xdj5f|q#d3koq<<$kwTe8$J
z9vAHGP1&1G$rFRoip^#Z?<|*3pV8H|$abY2ZZpHQ6{&Wllbk$B#L@B|OGoU>jkU0G
z&?<bVVASFBj<#9RRx7${gKz85cS^m{EO{fx&frO$#&{vDCk93%7^{w=T5zOl#@d91
zZbXyCi%~NMQY_XxQWUMySmkMhGP_m^qZFNYI3-Qo9SPQJMDaJlHI~KM1=%MbF`i`X
zPjiwqLpzNk!L}Y92=!4w)jLqmiMllkQA1~74Q1H8_tGv&rkxmMf|`67(V%hPBSjz+
zP_#sSe_BB&0~uZ^ln7b({(w?SxJ^paIg3(W3SqTujM_vOR|VONEjeAB61Ro%dj}=K
zD?>5aXS2R!Q!N-3)37m#qbPkUJ%*Gl8DTx>@tF6%`yH}pPkC|qg2Q|FNDimm8-0!Z
z;E3bNh({0K#+Zb^_~RcinoZb0+-JGC;ril?gUOhy)93iM=Hlj*{o^VB@H>Bp55Mzm
z_K%P0jHWi4Y_i8_Z;yJp1h278OWSNPiSWGI)?uyZ@aRO^r6n0{1F1KpDi`P6c@m%t
z4$Ib|+m6@J!HWo$Y%&@tK?sMU--w1ssF&)`5uLlZgKiOkTW0Jv(7+HS20~=I?QJT^
zt2`Hli2540Z9^S7n1e_CiW9Vrum$-vdaJ95+BNK239;Q;EuA}@1LZMsayo@FLN0c6
zSYP3MgHaP8!8?W17VkSc>qIcyhxOuH@LfOywkVt2ccet<!Lu1$H^Jcz_br7UzF$M~
ztXz1WvId3$kBsqGk7!?aC?y3<Ns^PM1xcEtb%HobzE7ixL*jeqS!+EgvJeDbjEN`(
zSl5vi1$mugtszN;?-upW;xz0x)jmxZ>`)-zr)d(tQjC@jNe5@oB$_0N{dNVdo6qNL
zHf!3p<-i`2rzu7Wx|9ZoEwL7t*<II1DU>mSn1le_Ix7v!8C)y$?Eu@h!8sdJ>ch0S
zJk^Mwg;hDguea-V`+7h4dmM>f^lpd_4g;W9Lz#HJ?7YHzsjR=bx#s-*g4Jp*dXmAZ
z$g-T5FP^hn&1ss7x~XWpHjKT(y3hkQqzw-bj~I<doSvStUauLA3TAsVE-$aBt2)e~
zctO7yXKs7E{(L?-wmX64U|{+z((S1D3qNU>`8bSqcQYb)M8V+@N7rH!v+e7xBp~U!
zj;?KS*0Wxhcwdrd1@FH1bsj%{iw{2hkSFiHk5(f5yeUh{YD3estX3=1)R5VfR!LC|
z&h;6xYF$yTN~%o-R;0d)ObVtFt@-00{1F!y=T!BEQJ#{IQjU)gxPSkc*=$0draXA?
zfVwLA^B?_?#Ci%~PiY>GQ}z=e)0i}4e>`E7Wg?`mpj95Ur0<Pa(6)5V0^4}HyrP{}
zwChXi<rVFEf$BO8(VzM1bLPziik-u&24x`2J@@aO@NjxWx|}ip#G<dyh|4Xi-EZSE
zQIv+krlsf?dtBU6h(GG*OO>8*^-|@j%bJVNpR;$ck2yHPWMi_u6cUe(sooGd8<gnl
zSsN;tG)g&?GQC4b;0pxeP%t-9rz{{I>W=s@ZscC~KlD&`a0Ywf0jFY@us-Tmp*q(n
z=>cmH$l0}o7EMw?B`Bn9INPxc=^GDdNgX(+X>G@3vd4wjSZC=vN2XJZPKYueU1v#x
zR-WxTJkmbt=-x5w>uavgFBw05&VzT}CrfhF-T`TvL6Y&#llLk1#+XEtDott~21}Zx
zJb&?=VpQ-uzw<jxreg?|nkGrf^MZqeL-3Ba>%c=**MX?*uomjZ;?O*LOPoj>Q$yAT
zK9CP0xkE+%p+B|V81714LqJV-j_p(3q7Q2KcXym;>nz?r)GPnmI*DJ*@w>gX#jE4B
z>v?+({q=<n<1U9a9xfn*&WG+32I=O)V1w<EA`7t+J|HIEZPSHY#LhExhNGj><BsaC
zF&x~9^6a)=+um32ybi`^h}`XR(tdu9oids<%gFNrqZ2`OobUHNNnNQPLDX77l8^$k
z*(_XJ;2K(wM`#OeTQM3HwCxDnH7KQew9JJ7G<GuRLb}%4AfXb*F$oA-mgT~SGlCo#
zBgk8tNy^k(i|smA%Oz)LFIg<4OM+6cH=9uug=9K{jlqgIx$nfeizte-9aU4~2`%e<
z)J61R-OBf_?P%JD)oKmilNSY}Q9(v7c5CO^-;RLpa_WZ1SKrLGXMP=;w{_?uYFd^x
zr!QXe;@Jzj&}Bk+l?hE#^YZMJ)p|+O)*@+<WdT+77;QMdcSM>cOs6yM-@ni0#TCoN
z3Z*rVA3w%9M_o&L5QxAfF);XrfXw0h7lk#^xr>Jih-n;$9hCew%>JdHzICwUhi&c8
zo84eccQ<uSS(Pjo3zn;r_uv0IAAa*&eDJ}yI5;?@C`Q~|-|(k@@?&mp=AmU+Mpc$n
zb;+2V#3)*9=x`+B>bC2cU*E7<m2{0olcLb%S%$S8Kls5P@#5K2I@^$>3441}Cesm*
z9zNjs_!yLD|6q^QB-HB-FQ1-LSjS<ic#tPN*&B0|8)~mmDq(iCj}DYEl*edrN&+hF
zL3xXBE83>R)*D<`!s?oKaYeUYkhqRaWoYGwu*8=hZ-&@ujdLZ%H0PbK-Q$}dKH%Yg
zLUVq|XQt-n&r7=Xic}jxx}23%q8H~ZPJupYC00a?GCs_6;ZM0BSnT8W;5`WjyrP<y
zoPYeBWOm4CJY!spa7N>d!)l9<5iH{1ND9fKbRuv?vgKPbnA9<a(UV5|xj7`c4kmHm
zCp7*l@@3+Ww>Vcp)DxvdmPLukq6+B-?S$K-tirX9#<x^eNmW)<bwyQHl%<H9)>SQ)
zY+jnZr#dB1G)2;4R7sI$SQpw;d3wjrdxuvJZ8Tjgo45CtBF`WnqdLQ9fAulfS8I;%
zKO{*C*f_E@!L$~)s_B}FZoNd;j;h*l`uPj6@XlNB(liyDO-(*2NYf0hbuSsvR5hE;
zhQ(?bl-m{Bq%>X2rfx~n5%*3W(mA96L>nO|;|9*eFpav69d9QB?sB5E+I`K?*%CxF
zSo|+Ek^ld3+&0KDZ4VLkdEGnZ9<6gUttB@KL*K~`LU0r}FCS2|Eq4#FyOe;Kij(Ua
zICE@KmD`-fFTcNA+{PXe-#(jPgdp+lLpJ@e4nrD9Qq0>izs<SS8j{4I$OiT#ICfss
zS+G{nIisaDVni*S_tKKl`<{chGtc9OthUZ;{1VhfN%6x#_QrOW&e|>Sb-ki(JJxGK
z?mRveXeHnswsX{7OTDS+T(}172K?mEks=@|Hgq8A!e-gFZIGB~7>~xF99bqP=c~`O
z|LOJ<AH1%A8n^zsgDLiwcF3grDo&0BDe|1wt2H+_bD^1pKs3#Al!BX^YpSZG>)Oy4
zS>&?1PGnB9JSWRjMxznqQGqdv`TPdwJ0_zE2Z#HtRx9ed>G^d=Yq~)}%P;2eulUPf
zWu4s)wubZiu6N4bz58)nCpCf%UzC9Tg3I?F5c^^xN7OpziyKCxJq`{II6AsVl4K}l
zXj@BJmVEx%Q(m5)GMbDTPsVgz$KKwAB2S65hFC=Dx{l3yL%H72wW3L<lx8#@)3zOp
z#SNc-_L<0}C?6_e4b$mF0+l=^NfQo^4_V(Vsml$<dG->+$t2;?-iVX&h*@gTRwzsP
zsK6QB_g^;32&K&fZri3(d53n8TEoQE<gOv{L95sM-ll%l<IOmB$}^d2zW(8TzW3{2
z<H@^6jFXP_)YEJhnDvtDk3OOG2-gD_FtJeFpa^_6)ytjW+44gOL5vTmfR@gKf}u%7
zX}lsuvpT)zV)7Z|>5MGT(MS7qMJghHN|Pvs@fz!$^tSb40KE$RHv<>0x7gv|VlJH!
z>fAQ=lZh7vL*wSZHB@3~LlJ;hy%Cnvh7_3z9VLu7cHS#|*gRb8aBatCb;DvlXT4gp
zTwSwTFKFt9rfsOIswZgMPK<7+ELBsZF(_+ERn7bF9FQjunB+MLDV`dq6I7~YqC1&o
z)^(xC!lFsYCNs24fQD{eVK+6|q@cW+bG2OJ+Li>Ut1@g773*>Zony1wP#H0@#W*KP
zB}z$>gz<QcZ9CSh6=k_$y<St+H6}@@+K#HVbYAn;<9C?u9RVgZ%5Lzs!}(WZm7#Md
z8{F>S?<W4fdh8JL``2t=Ipj?ZKK+$|_DzVbznJ4yC-p9p^y?1iM5x$0h0}tN4}IF`
z$n~gPT$kQ~3)P|FxC<XqDx!9O9Z+J6{`D!_hy=2IW<$-%XfuA^!*$!D6X8^KXsZUj
z>2EpHdeAa$AgKg`=GoAG&4#Q&Jh$?GM=0AnPt!DX-L}GYi-ZpC%+|S!PNbw_g%=jb
zqU$>9T2Lkt$!*zeBq+4j%6KLQ*FxKMY^sv^{D!KodYv}W<+H?MoUV%+VqMrU1=Wkr
z9>DSOF?o?89bvXdbn0GRCa=z8hRO-?e}3m5Y_ae-K0_xka{bu8_7%iRUa?#*SS}Y^
z-MUZ>+jTA1S64Cywv*JqmT{IAS0)JO8pb2hI&7MT<#K`1it%{FWHRCTv(r$ec?bIc
zi}dXVHt9`+sO`lJo$I^p=iOFL_RyUneuI1SgKLj;4WU(dmdhoF2Pcfj6N+L)*IJso
zMJdZ>U2<`L!RMcU&S*ShGMkZQDbvX)IEyaK<rdd<v`x!uxuUKr@EW6q&XDH?*O!-^
zy*%Uc;vA0`BwTB>QA{QivOE(iM4mF8%~)Q|X{wrOnv#!l9?Wtc%*GrQIg`ZTQiDqp
zvLeSC)$3kr6v_t2H3ZI13*{>{7O2_Rj=fB?&Jt>)r3d7fWIb*lZ|2w;%j36B`1W_-
z;e&6#&GaCltJk#El*1=)ldP6h%O#788&p7rbl7-@t%bH#wJ{m0wh?ha@f^~#FytZ8
zUZQsAAXkR2w%mOFg7IuhF&#6?ASn(-B0~j90&P$smEk<0!-LsQ;cXbc3q{8ivQeQC
zIs_&(yVE-^py<#AV;kCv#mHTpIza%0oNdrXi!f+iQ*A1)&u_T7p0k`UDc2=!+hAQy
z(^j-?D>{9F6kpf1z{<|i`j*CaXrGb*PB~6bFF83Lvp+3JOiHE`lx<0q1Zz9I4`kHD
z(6tQ;Puq1Q-gB^jz|m38rs+63IAVHyjIA2xZH+asxn9t%)?^y8Jg2QUtjmf#%P5;F
zG-%F(+IGrhG{OfF)M``G)D3mh&^e3Kin{HnEi~5f-q*j4Nrn68oek0wP9W=BskdkZ
zUUv)^LBG-R`u9gdu}3kKA-LN(4%j_|&)Go|hxYq5*S&R~g7fkk9z~zxzw2*<xjS@n
zq8?c+{O|&I_0hJCj$P+*3Y?K(D_-{&G>L^e{g1)v-d=+aW7l5`D#)y;=$(C`QHtFX
zw*0m2S6rf+NaWZ`dJK<k*O#B~*pwUc)r!S@j?tN{Tdhz%YPZ`NM27+f7#4hNS!33A
z)J@4|z2N%ln$2d3a~)biQ`gc`EIhlWZDbP)$a+8Ez)%r+CkqhlO{`jVLEo-vYU;WY
zXC)OgzuBx=tyautdrW2>CeN|XvMx7VTwU_{XP<F%eM1$5T%(w)cNXV5x~{?cq0_Z{
z4dd~cz5TsVEjmDORm5&@mjcCO%e#v|$LAkybzGmYT_ao9Zr}tmbWw)uWv~XrDQw$v
z_VS$Nd_~)|!lw*G`mSraxVWIRf<{_rNs@%lTC!Y3toQf#XuFo_Y>F`nr>Cbhb;Hro
zA=Bv;Yb{q-7c@=Xr#<~&Kic`~$G}Is%|q>fH+*^jJY#UA?hKY0%<}+$$ge#3zs60e
z@D^vKctR;fVhm}P3Xjofn!3R`u$>ce;-p2Z8e<Gi-B6Y_SC?0$tBk|8<NdF{&tx(t
zNes3VFrjNIx*)<mzrMy<k1<koIW8u2O~>=+&-vt&PXZbnz?o3oad>pVcwB^(`vhk@
z+PX$L$AjZzCYk5)Udn@6&VHJZ<LDBD&vQnT5sj4MVjN_hIB{Y}3_cU?T^d>yYU>$|
zQ@V#I<c-5GYc|)Le*S#b<Bc4<y@QNze(PO6{N}ry+&`exB~1e=&zapjA%DN5+N@aq
z@K0&4S0vt&DyjZdD)8kTeT1ocaeUcSOy6KSdhsq$p2VoDAd2j~km#FD$<5PeOb({_
z!Z01BBzh`DMJJA(HsF<(i5EINNZT$QQW$2(NCHyC;I|bIO8el~1vQy|fMvtfIthkV
zZ>f7KR<s(VI79UsfL&|ZEZ3}-E6!fNWO2P<v))iwHMX^Q4{g&>wHrxomL<#OlJ$Dc
za=B!+UI~g<*VMk2M#9dJpviR0qx&P?edmCidB$uq!Fh{w(6$Xmr{qb|8{Zg-Dy{L_
z;g!dvnnGu^)rPZA7s3y5megpV!^kb+ltUXRMmg)cghZ2%N4*Y~3SA+ZP^q+Dtypi?
zlyxP#bly^1%ercL>)j7|^yF)F&IB|l7#qJui*7qeUlftvM%tngx&5BoJT3X%-RB#g
zD{>xhJ9e+6i3e!XU9XK5l|=OVOBV0;zP8shl%kOJ<NJ1c!$?oz@xfR*B5LR9jKUcD
zZE`zcvJRc|sHh?6WBPtjwFyCbCvL!ZkKq^HQ$!qm|LnJpoz@bVsR1DOCg|jw950@~
zU_M`QeRaeB{!yrmRrHb&{THAHT|Xjv2mKIgrD<(LU2j+|7o5F(PF-(sUZfbZR8l)8
zRK@36u_LeI+;-kX2Yq<R?_)drtwSwR4joP1lIVm{p3^0Ux~iGq+%O)E8I8sPnX=S%
z%X(RId4A2ym#4iBm_m^kIcb`Z<ptx=D>o|hi>6pC7Ez>W%4~0s$z+O2q71>UYlbh_
z{)dj+)>(VClhSs6L`*Z1G30BA<?Y~f{clB3y5jueignpwy(ZNu+9<4RsOmMVazSU?
zpj+pID8FYio-iJbFvc(*%UD)r#o5bqyjL6?98rwMoSmIZUs>DXqDrL?=v3tL{wH(r
zzSxj#M|k^>*Y)fjhAYFMLUY$T9|mS$eE7lbOXaIVC5Fduzr|=2v=oz32zpwos-f#v
zY)pmLP*xkNdV_Thssrac_V;IGMT&DBO<hth7kJwVjVS;NnGp_UmL+6k&0qc5U-10t
zQ<m2YTqpaPNj;-+&ZEaCOs2xuOBCd3!6zSo!u;h6>g6R=S3KNL*-JgfH?$3Gl%_L=
zq)13rLesS}icVTNDikIa+o3Z9I@lKL@U|fV9^OABGX~eT{Ol(m(N-4s2Hoi1$gz9t
z?PK2m+FP95J0?p_5NiZwQu65x^WZ+m)%)yUTyj}{3bw=BAjTT1q@sROl<^3&z3)G(
z5h;ckhdJ-DG5YWWNeMjKdX{JBoPV5?&I*c2!6-@5D#3XTMmice!h%BeYVq<V4qx9u
zUTvewEvl!3l!HRpNHCo0-qS~U61f5<E`o^UrNqA8<)&t_Sa5xH#l_hLn{q9YM<qSs
zx~^lrUU7bL&a>0!TwYzWTrMfgs&C=7lLk{(?!hM{7*bk%HwlZOX0=%|os95WQqVT(
z@QKHT@^6y}kIf_+pL$x`3W1XZ?KEv$Q&uZM7&?#jEhZNuV67!BQZk)WsTQpgvT=^i
z5_DovT8RWlYbn<os!d5*mO=<uip{2^tZLGH!neQstDrNH$ngZ!RT@rue|M{|yPTZY
z9s{U9et*>w`)d4<XrSY98%-Q?|DscO`|m?%(mAPEHp*-TUElE7rUipz6=42|B==}R
zaGYdqLFW|Kf-`^-0rA-Har-44E9HF9fpV{mpNeaWElQx2bh=>p8g`xT`0t^_M}Ho+
zo3mn9JpA@Nh&pmV?{RT)hQ7Js>iU}TWY&*ebXwzmMwBknF8cJ*a9%)<oNYEMZmutA
zJ1IucT3XH}i6KohMx%n0dk+|orzA-l#>9)fjF$e)aa?!C9VK?rNg(PS(uu-a2-;vt
zBB)}`hH|rFwOX=Tt(Z)v;1HRIe9ZpdA>J#xt`&sAXvX6ad7hD_Nf2*NFvbK2FAb6u
zTU5v-DOr}0BpE2ZGZwq8#CC(L?azZT41r0cJS)G&g<`OW0vD9-<6O~R&c6+3;;Dn%
z=SAPNENkZT6>V!l89WM;8k(-Atk-n5mJA0mU{T~r!elfd53LRNj`!)Bj`eCy*L93W
zl7cnbaCLPf{Jg#ZNrmMXs7Zcz9RAve-;QbzJovcY2HsluyqHkv5w~6<N2HJfmLg+$
z8@=oI&X8VIKA`i_OLIZncF<Oi;R#;7-*_EbL7L@K95Kp+Uyv}J?X$PH&vZJavtClE
z%2CxNZC8<IhUwmfgX4YDT-s=MbxT`Sv{gXs%8IsXB=Z+&QDC{eKIO&fQ`V~$SVg8X
zlyek$&i?+4!^1tiZ^@F3Jj?K%V|jH$wU|S>WaLYxCL_~gYpqJyG!9)PWXhl%s79y{
z9<bKa8H3UgJItAosqsGame61_nC#^|dgmTD^DC~-7L?2SD^H0xa_m0%=3Bh|<N>qE
z1f@NinnV|PZSl&H>`f^i+~e^#zd?B;jd4|2w5mrXRNv!O52CYv1M(`QPm!BNT8x2k
z8^n-_`;{n-Z7Y^%XIy_Wq8LlXm{tat<~Wn|?Wf`-Q=tt^Y-An{rP`LQhIJ7X%e83t
zdF6WNR1R4PeQG&ge~=bIZ4G`jChK*@>5G?~pT4B28oIW_Hd3k6)Fo%<FZtxtkNN!B
zXI#&(+0-?iwSyFeY;MlGTQ11fFnS=#G|FgZvnh+^4S8<xN|UIx=gJtJ$ZJhlzDnbi
zlIk4SqVa+rYR_nz(OKz(={hUMNhzAHrL-mMc7sY4S)P&>2`aGhz4x?jM_aWlZ{}=P
zYpPmWkJYxJ?ixDh_~4r#^5D^1kmPg@z=agP4F-Id*CbP8pe6Ms(P=+QU+C1mfr<6v
z8V%9EJCQneow_^Iu{|;ooT0CHFJF3zVSJ(#M!fFcOY(x)IqR_+V&kQLnK~aTVB>Y)
zzOF$!a^O7ne+_|C4CL;7Zt)EPmW>;rQgLL4I*nS8cfZS}+et~jNv`5G$EMl<HFeXn
zDL313BkpH|4V10$x$X-h!W+B<Pi0k7RTWjap)8kJ*FkXXRM-rPJYzPSF)GI55TSx3
z&h}n?f6vjm9Bkat`w8B9tRM$4VSR%X!PHg7X1!v)UQt)IbfiggrqdZG_wF;B?c<zq
z%8XG=repFfBTqAtr7#g;G$><w-Mq-_>pCkMcqV2-wsWT+yD*>bs+^Sd?0e@<oDjc#
znIde0zIXgydEM?Q2ZxN4^P&#0-ju9X8#?Qe%!(&THRY<NtTtFDDcBgK)1!6ckP57s
zPA1&U=Y2)+Y&xYVMzmeWV!jOA!!W7)+qzwx@bfx%yYt@r7;J{M<Zr#!4+@du9@Tqq
zk$R!}mk*a!|GxM))Sy@HR$EmmdZhIW?{2u9loX_lCKJY^Tw30>aCmsg;o%Yc`v<Hy
zCFN#~bsbGx;e5+@QgC>*&(ZM#X=<c`v~6jslD4j?%aXQkv8_e-eb^Ok*YNV}8CO^5
zbhXDQgHZ|2mgHH=bUJ1_8<QAKQRL)VF5tvsMYmp~>kVVmkt<K46<VjD6=l;g4sG_7
z1%>Nf2ZzULL!S@~!AieR5Znyo$;O7`2M0X5xsSD;+Iu+vYBS=^5W9CCaC~w=UKG@>
zLb()gm2lFurZtBA=z!xVZ&6)c^Z9&%n=ctoJXxD6@aojyms@^LT=YH+M08txKUudk
zCNl}my5#(`r{vR#q=l0NKikI_nN)?k;X;ZVT|hWu$v<IpV7K~Cc^PPh!P(s@DswTS
zash2oc3^+xe!8|}xmfe!#Y@i4FKFuq)p_YH*;Rb}$<O%XAOA7WPM@;cEOB16>^c{&
zrS~U=dx&PIP&b<QI9g3B5yH{&Ay*e?Y}RWg(=oNLP|Bdt<f*1>TgVa;Bk3lWc$CrL
zv`p|W2;F+oA!}TXRUOs}=dX1gt#3&(Ly}4KZh}6SvyP@|XdBCFv1YYg(z!0AeW9&d
z>eg~{{}I3R%iqUi#A#F}M2zzE`>qrN{!}zry>k%K8(+4s`|#$w(sw;-Ffed&WOvWK
zixj=ix%2qleyy+h+_-VaIr-(O#T`qyb8kauf9O07Ie3FOU4kYj++62l_XCNvMc!hY
zE>_Nl4Is*$*n#v#{F3WZ(W(!?g6;F#*`Ir7XxMS$E<}%D8wBiCJNG$kMfUR^j~+fI
z%?i>qM;o&hd5$J0K3gw{b=wNMrH$^37laGjvRW=EHybvaHC46g=?_YWvF|KhC)MrF
z!=@~mOj?qdX>jrxag&V#w@PgtC-H%|$nX&POp=5&Nr>c@EYE1$MkpPN1?6GMWHO@|
z6=;)?=Y>=dI;oTgPch2KQsF$MMyMz%BEwp31q7U9Q`R(1H3$U$nvQlKrw#)y-_MyI
zvJ3aF`)NN+$M&Ue+v|uI>!7ZMLswQ6TBVo)8Z3Q9S2S&n=p|~5Qh4oo_~;QyVyK%M
z?>%{uvVOi~xtyc5=F#Ivq-iFFDpkc0EO0S!`bD60Uvk{GO*<*nLEFDrsV(nu+l?hU
z-`)r9!mwDK3taW>`W${{3~Zuv*PlxMwR@4nKoFVJqsd5~snVEKlVv$4Cnr35^pJ-S
zA8>Vj!)m#p-3Sjf%~IZa@|fe}V~!3F8D}Y7)ljZhtQHGa%Oy=!Asvc@ewF7V+OFni
zdCjMvd`eYU7&44ecx&l=hgV@uXDJUKK4g;5Xe-OlKKU3|R=9FQ(l+cDDaJP>Iwi{z
z@=?L6UWjB;k>hcJsxQV+YvB+^(etkB!26!N>IwHAGG1uScYf)c93C9=$)~6A@uy$u
z`M;54_xQ;%`7Fm8heBh3L@BI+&RD!RkdI03A9MKOdu--cT>a?J@KuL0n#wkKoujNw
zAmx4F)x}h(1|_`_8>KK>kp{%8ucF2EkiJ(b%4t$8-OYx}pMJ_{G-5Q}LuWaR3c!e#
zUDIN7N1htAO2lzc3RI9CcJld3ff+#9PLfGGJw4Shj0u&tPD!fJDG!MvRXJS+Mitb{
z4WEB}%GvoT&bN5&seQ}m7f<=Y5B`Y1_{*QLE?0tB#AV-Y7r{`m{PyBdKFY^P+N;+n
zbDlh%a&nk*{_+Jzx0s1kdsXEcoWrPuuJt4&!5J(2o{(B4q=lhvYMRcIrU~Ha5c!hM
zcBodPj8u~8Gzra;HErABI#1b_v>~<Jx{l7ZRCUREvtqv5Q2U%;`rdD|zjs1edz7+h
z@3FoUNfxb0ltB@rf7Oq{mY21S{-y7*?l(Mz52=Ls8eZ}KcAbbQP_4t>Asf05X&7-L
zVkqi*d67OfuGl(r{d0;zgZz8gyHkGuD(9~!iu>*TDW#F>L?2Rq(O`!1BPG3-J11-&
zr4`Bv89C8ru-R?Z<{i)~Fdka(Nqnpd_c#=-_4L)_S|W3=z+vfJ3)s*GD9k(`xqco=
zjalKb_J(V4?-9fhc#w!<^@!mu=gx;`3y#F-P+cnNo1Q%JrB3meAK(7gcR4sbW;B_E
zG4G#Ld_uhw)pael6QZy-nl#PAGZJO{>+1_H&R>Sk1sfP3k|Gq{u_-q+bwg*n;LvTo
z;cXP}dlWOynK;kmJQX<-M`yQPN0cHIuxXJ}ND`WQLsgcnRx7HiWIUdbB$?<<X1Ulr
zXC-*Eoir~`l3ols&UN8w%KC^>MDWE$GH-U=5q!~OO55M~Eix2ivLRt-HxSyst=d$4
z`pIWF8wBB%!Xz@LRaH}NO1yJu9g6LYCQDNuKYql;#W~(vj*gFLLQib(9C==FeD5B0
z-BH&qo%Ix>anKdCvX^Mr+lbfQ=P&BG9VFg%?qpQdz+MGs@AW5?z3vWL%-}@~mUx`8
zUw*A&U3urIn}$5kP-)tCM=3@lrh8L1Wl51|RF$lwdq+p??M<0Z#*FeD-?gkZYnC@R
zEawZVO(|=~30HSCo}jd5T~?H3&4UMzG4_ILQPET_4#&D#vuDRRtr$&a+<WjCS3|X$
zQ!X}i%O$ERF>Qm<j?pM%Z#rc%E^r?5(U^QRB1u!J9!H85BR5s|%@Sm;wzic*QB>%7
zBiR$3Xp-DOuKD_hd+gtP0KfMyzjQ3&O^YWE_fUpD6{w<r6AZqfONMkhX7b>ellR`H
zesN0s`~t5aF^0wkxHe9V7<Kj=pcFW`;O4}?$fguO$A{ocHZ1KtwyoJ*-EjWdGm^st
zrsEN&$RQO%n0HQElPQfBWwd@m0sUY{vVL?VBuWg^(+p6f@MUx?$OzwqW=+>Q5;J16
zsQBcg&v|)zj&&W@x2&oaKmLn9=YRjV|2sF=*L1eyO&55d2E&C)VsYxU2DFGgwr$I7
zJf+<~;_~Kf+wmZiaVYWx?=@Z-(xd?2$#XSE8lY!6O;gh}HC@-xBo&=*@m{!C*0waA
zNFrG8$aCql*ETK6i(v9*vyt?c@;EG2C<f`ej{T!2930-K>w<IzV11}Qj89%g<Fob2
z?kbwt7BlF66I45*w|AsoZ+(vs*7wL+M7D;=SacL$jl6p_XXtPaQc8C@a8dl$hchd^
zMXPV!;r4obeE!1<P2FEpG^V%H*YaBt$;^ukyu;cK>jk-UP71-dsVBA$SiIi=o4ne<
zJg%4c@Pd8ND%^VrK6W-W*y*Ce7J=KQTO)r_F5;G3ehZP{=RQvEKj8S@eMaLcF&OCY
zEKW=>!oeatBzjoVsUU?Ag*Q!2mWhDybULA~D^PyhLQI3TE$h{a&1OwmmO=N?;C*O&
z)CybL1Tpg+Ug2;QB3c(kSe+HrEd+|a#EO#fZrY|#$IfPZ#J<udiT2G&hmqiTdnZL~
zLj<?~j_*^(-ns4bAH=>t@6pd_znuUP(37clOJF;E_>TIhM0J2@FgA6~;$|*gTEZ9#
zB5di}nzpS&!JMQivoz!A@Q@<ULm;C#I1rt>wyDutGagSEk0!M33eaR}&i=uKrm3l_
zn$5Zjg46L?yt>MNA>(z2LpS{W?mUVdIr&MP<6DFf->?3@bI$Ed#qeHYTnhBa-L8Xo
zhtnPcgt9CtHx)%LJ&KbwWw~4iNt#VxXym=nyY}{GOs8XtEJXwLX2at8n)PZ$Q`fyc
zH%UsG=9na5xmr<GH7L#A{vpOCT<J@0E^jDTYev(82ag}{&in6j|M4TzY|PEoHS_C*
z^ebPlD4b<%G`aRniV@?{h%8B|nwH6QN}6RTC3>OOI}FL7az9cDAkua4euvf&3orw6
zgVBa!l7Hnn@J5bZ5>gW;<f!9A+on`wTFH>jrtH1-kdv>!$Jx4~UT!dG5)>9e#Uf99
zyBuG$QT7n;t;1PNxQ<k5x~k#w#VN_*0cLNX$!Lt8NJX7fmNqY>XPhzpDf;a)MCWnK
z#>XgG1!q$2_5%$^oaFJVwH;ZKQMHzXgDF4yqd(;0{0i?K)^(KihJX3{{~Lew!#}2J
zo56<k)rDL8v55;6kD@~mM^D>XtgErU1!XzDcSPB)dH&)B_fPJHNHihK9C$0bRNkR<
zf(`2T_!gy-kfw07zC#KbVsV3{i)1L}bFJ%GuQx*Ac9zv<NmW%SBllL;HS6_;x@k}*
zWpD40(ReBg&m^>+ICTas>E`I1{h~mCTjO>o=gn_7lGmkd{jK-B`kMIVc3yW^TK0CT
zMC~{~Uo;=$`g_%R+j*6anoto*Q)GF<;n5LY*Rr~~k<H9Htkr~UTeuN=o}Ko)#wa=y
z!?hXbaoij{TW4T+kMS6ym$%c#!_7mS3t7jvpT*9-{sKISz3Gg-**?W+LN8zrJy$Vu
zmz)z>k)dN(<XLdaTB>S8+mv)oLzbr0O^pue9l1wOYvpfAlAzFoq6u*to4P?8Lt>I}
ze*@<|qMyBu+=_U|l<f8yxgen;<7GQ(jJ{c~X_}fONu>o^v_S)O)mbYXXKzr-g}Etz
zv#&UM(w;6p|1HM$6_4;Z<2<|59f$$aaDBvY3gKU?xSI@1;kGZdLF2D1ON>e-Xo-QA
zx4pxzkPX%2De|1-qhqv@bh|d1$#l%s)dg+WVUmRX{r#|^3pJ-GMjRX*QC4fTS&QzU
z_Yj)F|Dsc}JN9gN?!mYY=?b?4<e0LL0d@ShXFJa#o;O6)<Ik^{!@+6Pt*GoS)9IY2
z+?14?G8F9;q*=~tx#05RTw1o}Bk+=C(^@f|j2IO;#*3(ZxmpQQSC(|0_y;IMn&zZg
zPUjp~H#c;)!x&AT7v#qoZQW9CHe4@ijt}nh=&eV*{mv7Pj_-lhtd<)VS97|mhRudN
zu}m_<I8jW;WAZFVD+$Q<$CJ>CFW5Ld45q8xzw7M?PBbukz2AY3jKne;uPpivytOyf
zu`|8bDLqbiQ2|-X;-UH^!R+sG@cz4$*GsN{`U!4b2dBV@$YVe(ddDs}Il7My@16j}
z*GI?JdohZ1l&g}<r_U%356Q-3#z{h}H4cmGTJT$qrFfpLQx%=8tpP;^v>|L-+`8xf
zT7uEfOhUD(dGz2d{{0{R5sSq_$jErsn-%}-zx@C3qd)r*ZP&`_bkOm+1Eaa~=$$$G
zg(FuXM%5u2X2Nw5$1Y7%9zJ?Z-8P(^U$DQoN7p%&O32d@N*NhH@9{oV0jdP8VuSD$
zt!uPS$x<!tj+_NAJ$j9n=FO{BN!N<eTCdiuHY>DJq$;Iq8y52=o2rynNG4-4JCM&Y
zX^)EOh+3(kp}M_6?BEqLa?bagB5Z?(SLV$vV|4rb(5bsC!rgxMR_|_@%H09~UgJa#
z|2|B;4b#ifP`-X!^!<i3{G7Y)cZ*`mx=OQ*gTq6q7F)0ADvPsVoyP~n&^bY+Y|u_L
zahvqJdU?B_8`e(@*wm}{{W@;W?eW@*dw+4jbN4g)`3~p2wBAdFcA%9|0TLrc0{}jZ
z%hs_=q>>L$*R^P^m`ui0n=wj3mStkwbQBp@bgsj7j<)Wo(*|Rt^<UGpq03A~p64Rp
z<Ohzi--?_1zEeo#*V(=&v^a>i?`UB|m2B5ps!hpqKBqd~kR=&u<QhANw;kSDyz2z@
z^1*(3w<FKvwqfu9DUQ!#>jZqIV-PT@{;d5UU+W{rE7+NZArmC)Cpad`2Zw-OB;9ou
zMG~HgJeQ>B1IT`trE>q%=^l?BzeQDTND_ldgcDX)CEhzmqcJD<@5|L#k1+|ey#tQ#
zov>PEWLZX0j94z#%;$6Jy7@}x{woe1{>v;ym_I|=rQ2!QZO|5q|AKJ&*06b`JH(&@
zfA~4K*SdVjwsV@Nw?jizuYav`)TPK`<#|q)=A7I=;O1(<Pyg~SSk4!G?|a{)YbvyY
zBFmVJMkobcT~e)9EUqqDFXqxCUMX~vkfu4scm&FDc7DOrXV1v<oXKQNnrM#p_c%P-
zqiHw1ynM>ZqkBAj`w=G(9*~-x#nqDe)tvd|4fXY$ys4NN#eSM{vY&JR=um>WMx@6^
zlZj;Rych7v7%6J9(sV(E!a{2uP1};^8A<5LY^}w5`SRfz6?rE1#(#CalQ(kgd~oLC
zB#2H<?;dG|kQ5D=1U;EDe*B30tD1IIvHbKostYJzc(@^th*Ey*s0|{l?PmMx#Fc()
zUU(8J#AvQ#b#}?k=TF(+6JoG03Y^k(O-<LeBw0?ZNQmjWSa}Isht2hyMwpnATk2c*
z>})zp5WUT&;{NeNKKkh=Jb(6_u4!@Ju_;&l{vZ5r{P4#=6iE^9QRqOZ-aX=^zkSCM
z9Tm9&gE1OsrQJ=MWTZ(-mRR;@8e<e~TjQKWCZlA;J5S!_G)=j@yrgNQGInpKQK6Tt
zF(zy-l14BlMaaV)S|_9_s3fOrJL;wiq<Yx|tEytX-q5y=vf8j%&S{#4Vl<-OY*?;W
ztk-MMiY!ZMonrss7<AGbC8^*Ga5M(zZ7>AeNOD^g;QILyMrQ|2i@)0)zpX)w(Q99I
z*gF-w4%4d~I2EJ-g5#y)`TAE35W8K6Wr#rSJnR149v>=#zYM*V^SGVsj`y}To`Wt0
zak~ts&5I0UG}<IMUx!qz!g+&d&`RodU)WpD>JG9P7WELZlYnpACE_;6aCkq%@2_kc
zUpN}u`~7*3i}MRMWkZ_f6va45cStvaEEC-}9ngL2Ius&OEFMrAQlcHMYiO$qYg_QH
z4>Wp30NFKZ>$b1h^Qkl}FUyLXn;Vw%C5K0c(yQ4Aov%2DdgTBac0pwwZ7T%~z1Xpn
zVh9_)R8aKqx`vyZE9UbfnFkuuBn^&W>@Lv-DG`Xm7jbO>p#e{vY};oLk!Sh!z|-G8
zsMkmAzW?7Vzm6>@2PcwKhZJ%|I(+DR<q>KD-hy`|S~l8pJ!iGJp|h661c1N`twsl6
z!DuD&7Pg@nO*uF`;_&d0k3PB-gjI|9w6h&qmN1=-IXpb1t^#<00jDrp`pE9>?NN+I
zoSdBS@BYod=kw2>vREv3kS~$M`icSm?XmgN*ZBUlK5Q<^ZwrrZ7bM+z!noJpVV~l2
z_kLTP;p2rxKYw@{`ZxlUsKM_&^TmR;vn-b@&d)Ev8}?>vyz{*M_S;z3k|l=B<RsdN
zgwM?l>-i1q`2yEjv=%v@G|8Cl?Nc=kH#awY^wG~kR~1<&MVhcl*CdJN;9$nd{X_1Z
z9I;t%c>38h?%%u5>U>ExU(zj?l$S5rcQyOzn0wO^Z=D>HBt}xdZOeE%MH$lvI!bG@
zJjbL7Xf1S!HXyf3k!RAI*b~sbfj=OqxTW+Pl&7a=e3j#k9J`204at*HrP+g|&#dsb
zYU}YyM!LVx<jGqc%~zB+E83ftq+y*8oVoDSeT*uCd;o)uL&xf`Th|k!izJcWv<6K^
z0<B~7@|?x~9@%sZC-*Q#A-a0b_4Xh(lTb?GwC*=1KQKHyXrwa2N4X{K08wX6lP3jj
z?fCrDXLOyUKdfu`lOO+(AN=SC)Ll))Ha*=dhD<3$RN1RuusKO21)gP@RHvm0S{pIg
zU4?foMi)}W+I1+EiH2O@tDzuCbH-D{lXu=BFLIuJ{*=?RmuM~AgQjayK#>=sN0;Sj
zlSuJI)AY~WhN?NOHF=U#RVB_@>Z-;$$7;P|vsqEq6?iD?ise#}I%5o3lB0BrCgI-6
zBb3Q-){<pOKdHq5V5ed^I$&D^6pg{wdZ@T*>Nuxv&6fdMG(=pZm~RXsA}Vn^?S1>R
zcg3ycIm)!zbvo}tq~i71A-^pAhxc0~`wI}O$otx&a#1i>W3=>1PK;=eX)Q+7D{)9r
ziqTj|zU6u?{dvVj05wG9LWCa-wHvm3Qn!Mk=--6^Z+OplA)v2|_=gp;5L331tG6AC
zS1M_L0adZ5r_V{VYmzi$G@g)UIa!vHXVPsTwiZ%Bdc_2#{Gu3<WnvrKrlAcyE$1w~
zcA$5_Hs#s1(vhWYTbYB-(RE0dma<~A*)X0=`qpE^F$p8RbM4MK5hre2nnoJWw{5$1
z)OwjDxt6M2vtF$j6$M$ELE*7o;9c4d?`?07B>*@5M{eVo2*apgOShLc%xJt~F#Ep@
z?&nMWnPC7k2#6wL)hkwoAP|qn;a!XNmc%IPa>ZtT#d>~&?JPz)2<h=O(Y=!!bq3QU
zWjdWPo9&TiIdxTqR*4#GMWZgu<@pyy&dv2Y42e(&nz|;<1<gG;*yo*h-r@Z0oF-H!
zcU`yjaeH*@O9tYv8NdwB8!0Y3^GSumlNg-Eu^u4tA|K{^=kE3z9@;DUlR~S&mk%+*
z@FTT?B=F`U(%ZHzRj4RV(~LCH<ax?yl+(2p%DY~s%XJ;)dd+6FqFisVokby)zG;#&
zo=j2Nuv{#;yu85Kjz~3Vn~LkJ3ySm}X{yL_!~F;Mc;}sWn9TN3PVw2#J|n3*))(hg
z*BAJ5$-#cgy}b#?`%|W)g0|jB&}5|VEQ0`EsV)Cg1;{?`rCLj-Ond~<3iVFX{HO#)
zXl-b1;QoL0mWglV*ljn{?ZcOdUifHMONu}_Fjmt#g-dhN<0DDQ&X+v<lb?{(E?lTZ
zVYV%aAgCk^Ho&+k+%cIWIXDiQahgQQK-BX&*Uz3|_9u*rl){*Sf%QF#AX#nX{GkF+
z2*6<(sPG9&p>-n8?O+4x$)sAb-jv)sc*vjq@Fy&;7x+;9dwTkUKm0fUnyM*f%z9ey
zaFdeBkY)zwJny~tgvSpalIVoRV$O>fFG5PCqYZ7r<YCybNi^8blBcn7q5~g`4h+g9
z=)j{pIXMADW&iZ$DN1R&&N3TMq(`3ds8->d(BsmHkq5B0327m@e(O4_s${vEb2GnY
zKEI)E8d}>?H#KEh;qi>dQz3^NLuWms*^JR>Mv~+txd^%Hm<B<rtvv{V8~@5?-_OW_
zArZ&JgO$eY(dE1Cq%ruNG28x{y~8&+_ZA@>QVzDlw8ME2(Wr=a-Tu6t&xl`M?HJsM
zI`$Ete?1pxXNy<#*JO-v!<-YMe54t8FaB2N1f8AEW=v;$EN`xbgXJ{VA?dw%y|*M=
zVEen_XG25<eDBl^QN-Jhounc{gKoDwR@T~`RF7PTNS)l30^XI3`T36Xvr{|<r48fB
zge)&4jh1JOM`Lm9lAz_Lwft=~!tKq-LN#04HZ)BYn#ad7Fc(^LZO5jm>rhp>b?mHn
zbZy(W8|;EZ7DhUb@?hMfqbA?26=$(5tI)u{k<JM&b_<DP>Zq%d&1%JFBzz`qR7l}U
zN>`~ZQY(A(_6NBV)gv-W^+w^fe#g!lR4S+(gx`gG83ypOR_Gmq*BxTNMH5iP>+^!@
z8}Gn%DA%H$r&(OFzP?~RzmdMc#s%?bLdHV(sd8;JS!S3`XUwKEjMkDKFQm|<>splZ
z<ax$uEc2vo=Rp=rcy?{ukrx?RD(TqCWXxL>on=&1Z5M_u<OKx<>F!3lhHi%LF6r(V
zTIudky1N;=1RPR^?(PzX?)c94gMTd6V!=7HpJ(6qbpbgSo7T<CJNl6shUl~Ll>5Lf
z-0!j@d%^6r8z77IU!R0%-v1SeC-Uh;J4w(j^4N_HHN1rE-Mmlyb4jdi)+NW_Bk;?~
z*Y{Hur_Y0reDRk``w%Ts?vru`9hS7@BQ%8>83F+?crsdmkc~h|+mV}MxKlq7kNH(#
z-2QQ7Q#M7qMe2ds-=`#Gyreo^2guoinuTFQ8*hQ3enMON)<X2w6G~VEidU+#1A%DR
zRI(lCq4~n5AZX7F5mVZHW5w)R)op1j{v+uy%l1_a3Gagu@sJ<KA4N2hP52{fg8P5d
zF~}>gH7BSeP1tGp7A>z&Ot0ZA^uj(71E$b^9VM*8{gP`kihZ6a{#1|UThmgyHv^jC
z$qFO8=ZO`#EWErRGGZmCCfFi(A#Fns4arnMGe(|$+o+GIi%Gg&%f<k5<AR+s^6BP+
z++kavuJqjO;|A%o?d}L6^bO%yN05L>q7}%tZnmzOXX~N+o>5*}t-Ri>LFZGti{s8m
z5ze~IFKp_2L=lsak_7Jd_h2@{^UB|Q^d^rR`?o*a9C!n~dP;Kqdi#*$_74+_%Lf|*
zIWf%LKP3jvz0&;HeKnn@>d|Tl%sT*Z*tkx=R=>JXLz`0YTO-@{9bS=qd`bNBfEl}~
zz=CvHaB)=;!cI+=fojk*7fX{b6Vm~A1HaUYIJ598cAdHYPa#2F+`*}4iYYCOmSc>G
zwp;hK2D1I+!@Y1<!*{h=i_c4rxSRIkEf)(6czw4d?zYbXJ-Q@@ei98Pi_?4q77gF(
zw8xu9;Xaiq&P3Qq$jHe0Sy?K~nr`qZov_2kL*Ma+CGKikw>&(!I6c*W%U^|^x*TF|
zSh0PyWO<&CJv2q0tC5Sp?W;RXfxE-E9v_p-9ri&>NzGH71%rT#^*@Tv^OR!xt)ZWt
zoqi{4!lFL7Pz30(?N2LupxR#MJ9F-FosJtj?xS31T8+(FIdki!MraRtnurz}C^7iE
zF`mBnLE$SyiM|oEksSauiJq5Lpy~{lt}`h;ip;p<vZk@#-JcF_c+ZJ$Ml$fZh4l7E
z9L78QFyVk7FK4Kz#6Q}F#9D~|^F8B8I{ObEhKxOQHJi9GzAC0QQN7EQ?f>`jPHk3;
zhmLg0s9w`uWHUP9ue=TBIduJY|7Ty2c;-dr9Aji%;aa&(kSc4I!u;AAB@26bTd{rN
zzAY##4@y450i94chmE9JE(bV~?h)<b?mfbkp37O3>+8DDj~t=hYg*5oc*3v0dc5tg
z@%VFjaB}y6lgrkJVdTdsLo5`R_jDwh0qMF0q3I+lZ_`Vv#8X9t2Qv2YBLp`1*Pcka
zFFO%uz|A3Ccf*BWE^+TRo*ES3)jtE{YqH66E$ok{!Jj&dKC_e3+<x3KdQRbWkDyU-
zNp_i^9p)g2t26k*X>R*lt&mS+p{`B8mbcTmQ&%NBC!br6BpMAN&^8cMfIkJM2?=R}
z6`~B4{w_q}23bj&iro}7A=!s8l08J`rp($0>3PLDL9(#l+uE>W797~Uy$Z;Of9Z>u
z@hK!n+CHm)9Yp4V?k0n9(>1*={!1dfUR6`Wo0V3{A}QuoA&f2f%!<2z(_Hmu$Z7IZ
zKux?`)fef}cUyDSbNU^SAHuEDb<IlLOSqRqe5^t=U4QP#p#-M<U{jlKBb3X<9Zg$g
zzG*u)Ll92DW+qewyqzy*E-nxD04n+b1mVQ6JG+DP=lg#sYr3?)Z{GqRo9L2kCjPDp
ztqZk=%yr~hla1~6NsF4R%?5puP3}M`fLO~daguW&j#+^^&B?_jZ~}!zY3%kuDPXwA
zZrD!1oSWa6d!_FoH>e0p_2*t~Oi@E>IRkw~;?e&PP<&b!gusQ8hK@<+fH!co_vOSH
z<Wrac^)N~r%;_7A-Fdz|3!ioxJ#kq~1vXWIr?4#Bm7R%|X?<&OX-D^v`&mGebeG0k
zDmGExD^mZA=MuiJ9LmEy=p0KvhP|(po;}%&Mx#9aaltNdKV7fM&$0Bttcxs8%awiW
z&__B&@)@Z`(xaf9!Orf;^XMWffBiY=LEk9z!lg5H6c)(lg4AX7eF<vkCK9$a0O$RP
zP^;z<3VQaN4ZV6Qcl9J(n6epZ(Xg8{@Kus9Xx6n6UdwA#|H(KrsS^eO?!M;Pe&II9
zglKGD5)}MWq$MIDo`4cgs<RQq1U2PY{a#SKm+-7IZO~QKY*eOF1^}&J4Z8}<Z8}u9
zy+Q}3rWHOjdvuEUJm$>iBU;5@p9<yvAa-y3)~E{v42Q0+sDaGd4!sVI!c3>D-n+j(
z<yVpJhOF=pQeIN2dzXTa*x^CAfy;w|-GaYD+pkx;Z*7U34eipVj^L%4A5`fQhdCk9
zH&pP1S;Dp%uB-EO*vQ^AE{hY-Txq!^+rgfC=9=t5d>|JjhZTQ}S3tmm{a3yA@_3V&
zj*9HmVk#9IaK-B6<n*15-hN7T2B!+B!Yc01*9JIAq~fQJ@QvjN?rcB%(M+Lq-|2M^
zUCp3rY0YuEM`<#gwuNc#3$Hj6HeLkT!H5f%$6A+uOpmuFRm}O3VGyTkefFou5A7KI
zHF}K~^I<4^-RtqX{c)1z1$FHi&JywU*h*rK+iGX1KDn)2JM8z)=D+CbV^+n4|HB-1
z_g7IK!ld{AoTdpsFaoG+xEH4BZyl*JsXa>heZc^~D^;jwMkv*Y*Yn*6%{f8`c+F`_
zQAe8?7a_V-sAJbnc!ACj`&p(#LafwL@6p@Scwd1)syu5wr&#ro^jUDA__r+R9Z^jG
zz63~^%bb`1Q|7(M?khUdYXkei$#*e*r!#Rp6_Fa7=5oOeOV85utcF{s(5@oU_*A;6
z#Auq*q=(WP_ez!;uex8RJk${w+8Q{ugCv9kRfi%d6qYIL?5Lt>q5Z2H&clKAbjPfD
z9ojxYh>xSN$ioHt(N4<r0>OiseSv;_w($N9XfOZUZ^kJU5`#8`zZkG(rcCsUP}RbQ
z8AkVVeEV<w@6v7?NS~?Q)h67r!H0L3ceGSGfs922QtFdj{Zs(mWGb10;fM^p4xCLk
z5igzH1A94s{l4zNJhW}$rLae=Foz@|D@vDk66zm|T=iwK&D6g^6kmLgu%ph3nNTG)
zX1@)+0PAkvWxweD2te%$@cW6*H34lJbo5vtd`Z5N{7x|f=7}@sVRsk_bb)tA_%47L
zx`l!zbHCMO2icf(Avm%(p1}Gd@PLj`+7go><!nMa$Jv8zZ_fY92SAyGTXi;G8~@f}
z3>+a(oX)%WAqI(`@Om4(-B~bD*!%ch<~~-%&et&xE_$0as9Y+#Gk@HNT$1wq#e}gM
zLJ~9gZL@wysuND<P5Ym@5#PJXuGrk0`H~GW(AX3*&W2{+yYala`V+|ru6Tk`H+Bn0
zZeUTifh}L6)(;B?mLfhrK1|g$+M_l5!(7b0FxT7Fh#r(y{d#-n;{mMH-2TPO{_kW#
z9g>7W%Gq(TUhAJsieBtquoXL3rr#im4|Rl(fgC7i9f<k%Do-u_;o`Q|j75QTHEea#
z<?2-jnx-jaW^n$+ru=Gd^_|EzdN7B0`A1ZxRO<P!D3Um_cfRnQ9iU&cDG0=&zJC2`
zl}__cSPThpjM#A#E*2+Q2^CtKAhd16=wFys30uMDI`s>7bUfxx9tEa3m;1G1+?wDe
z2kj~?U?`QRV3)#piZne*!0JGKa%v`9YEq^kF{0L^D~bEhfb>0@5r#iW?rgIn6Aq~Q
zi+7LDR%b4yJM8{2@cup;sK@&FDQY^%WDu<A3`u;%*o{XU*G&eFgGmR;2X?ZldxD6O
zpM`Tc@~Wa4=K{2$Nx+0{wXv6y5egg4oG2ty;Lxm~K0FVcu#!kM@3??}IV{h!-`&*4
z9N2LfI8lUFla3$X-^!L6K)Rmz1SR~BPn_g(_X5>~B__k>8ky6>_dZMW$z}~s#+g>i
zOiGw3M0S*H0Ed&V=c}5F+X>VDEjERzcq$sRyuw<`)}I{Rz}f$#{108wEWVeflN03c
z6LK8gQF_#c+%C_D1%4<>X+5QqtKHl<7xu_idcA%KI{T)KMn$BJhH4F?{>=Ij+6@h#
zBXO)3#F5m}wypS&JN9@4T;D{DJUtx4c57VpNP_^+%I-h^#ss!HP2RJ~0l{dua5K!T
z+`!7rV<H1akM#-?$MAVr35%v-$nMDbD{PRj!!c9Lp@@%aflQ-1#;=3u#~u`9(vAT$
zf62Rpy%vvoPI*-)(5eONE_$s^`^^Dsf}N91M`D(=SJXK%YRm<yLg*rA@|YoecYw(K
z`ga|zjVj0>dK?Eyg}9?C=@gEh1l>Q2*R3B^h<kd(Nj68O)P$ZFq#qY+d~;@eW8Kjd
z^>&Y_g`aLqDu;{Wx3fOeOmSLrXVvM$>zX9#2VG%(VK;Z(Pev~+0KqLM=kYE_ZPs?!
zd(>MkG)4j=#Bk@XeO>+)-1HTK3y$x9-B6;x!nCj?oL9%yR;HqxR7_tm5nTRtD6_9%
zp2OHhJaw-Fa-G>w@j=M{UMD9q0ZmakK>*yNTV8C@stXP-!=_2`V1QW#gE^pzYz*Yi
zk?L>zC+||>e7cyabBwWZ2q2H*|6=s?dhPL)(D#OaZF}7*uDU94y#A2Z70QN@7llNJ
zNgZkWd$tL~(&mvQi+|ITTE-gKSzNNoNKTZz3zObPzA6p=7qEmyc}Z7xewd=*fO{pL
zE@8{u6N9cr&@KQ8L#YrigA~2yqg<^xbmW}G$IObi<nw93-%3yy=TVSrP4G~HL3#@Y
z;=In`{B%1*n0=4;nU*<zQhUsBluu(ITBthitkk2gWc<QsWEKa>!`UQ?+y(gJxG6;2
zoRH7g;f7J*w|^2@VV}E5^!qKmc0&p1j=~_UhCh0uu#8-Ve1#wrbg+TS&GmzjZmm&8
zgo)ScOtGa3bpyS~nqaa#7PzZ8h*BF?{T$p@1|ZzUHhJD>$vTT^%uFT|F+UaB6Y&!Y
z4P3Xmovl_ibed{ush*luq0QN_8O-rBi$)rlRkQ)K)<UIL@(xMCgv_J^=p7;ZRCw}F
z-sT5|Cv0g`{tpf>ur+^4kyk!$tW4R@^cCT5$butT__L<V`=><$g~J7)C`X&NnanHT
z<7br5fh%k@ak>d_DX;RwGK}{V<r`*6-u}UUP?mfVe;Bwaa3Z3IYIDBGS}fI)s!TLe
z0Sn%yj75E9B4jiU<$6Ho7*>VE%|L)a{%hcpecS6RbFtXnBrb6~J)#GJ{QCJ6zY{Fm
zBz@5Q-hA((>De9fCg>;}uFDd(U1*tf@+0^3d(fSHH2DZS(4ht=UfSNA?uXy3`s5UB
znL<_7H5LfnN%yHQGHh`!_%f~tESaqn-#u>z0|35rioJh_k!hBel{|!JOzMx5bW93!
z$|;P%pkgqW>jt!Qh=_^hio7*UE7Gf%mFWh8f&^EE7AynOw|L|d-ndNupe2r@BAaB8
zK?VOPdYyTsY}m37p8~L#7cGh{m2^}5oxzsHjR$(2T=oh}cJ0f+<@D&bqC-jM32$hq
zUWNTmr9?Qvb@9LMr|<_{CGEB4aRX)~7;1KhmP~6K;VIK*de_yN^N)clL&k-l)YOEi
zR08Tz*ji&s_?=JjCn^y+I`r-11SRI4*velWT0AdmPB*t@uE?!w=tdGrAJ>ubvT6;h
zsADFl>E6!G&1D*8PC@N+CP(X&!#*jtbamxfGbRFaC$)h0WBki?_nqTyiqRd(OW(`s
zOK}65`-<`mP^6sTCeD38?fGsx6n{xvC6;{5CH!=-_jo%XymM9{&Dq)uC6<yS9p+SU
z;)aFAP{?syPUn&gS#Ymp0G<k+zkB*dMh%8*<D4-PO0r<&Pn5xPoFl;PYr)yi39~@(
zA8Zv+p*T<;cK!i<c}&86JQ9BV(si}6wmIj2%riKZ+<hC>y}h{^h<G_<x!67)&|Vta
z9LG5dHDC3#$DbG+q-wQIN6$HkC-vIHC{S&pT3G=%XVISdAUx2xmn0m89U;VhwzK{3
z=!hax6WqFQCdj%-UN3Ake*SN*2L(5%Z0I60DG3>vPBPIl(q`o?fR*8GIFlS8VKeJ&
z$<wcC>1xY$&2EaxBrP<NBY0l`wIo#d*l+BFfUGV3*w>d3TD8w!ancX%UIJf81(g`G
zy?(L{z_8Z-+}S!`qB3|P`dS$L)`UoC3S>unQBNmjR@M_dDh?ZnpioHc64|>4^vGk`
z&aR!g)<sLvjoj~)(iK*|=hxT~#|&)FhnS>PH@xnVv04(?-*1HK=5-d8&)E%gX&7u|
znhN~74&k)7eNC$v!EW_&*s85KuabK|#DYD8H7j<4K;6cBHTBTG=55K@;?2=p(o$9o
z6!mWorQs!P(Lc7mR=IMJRkZ)HehZSyDbz)=o;BWj(~FM26C!Py**6~xTpAm~bqq4=
zy{=Jo)Dl|JP5BSM3rHRv#OD^dYU+}M7D^I1-*PrQ{ePBU9coaOMN%{&cypdim2e1I
z0>mG>h05h^yTFsUHzo}HmWeYqk+a#b4jKhq;3)6i*-vRLR#C)qQ<3D`GKH2|bqYdI
z@8fI#;<sOC1jMsPEFXPqO%?e}D97e&pv@6=9I>D1%2fSc<;LG3I%{{GZD<(-Zvc6=
zl8RY$KV@Y8Z!bdLDn1~}-QCM0^!OPe2s9gK9^bElg)Nrc^|N!Fz5GSzIM)Te(vwQ?
zkso}PyFGKl;t&^GIgi05EQ+mdtr+c7vv>lFbp2DA7Y%z%yFBVUx>SD3@-&nr0=`je
zkE7kNs}V-K#(w+VWUUh4qaku5AK$-yL2ssJuykZqsd#xz8VwV}dPPF=Z3zUJy;4n7
z#j|mFk(*NAM4drC4ib=pY`(s4{rqkOw))1+U9yp0^5K+`#PQ#}UxF)cj7M_Z`fR8L
zG*m^=+vWF}kCL{5fwPK#w9(xDdgULSuXUGvVo9=S7A^s$B<&GZ@zBXhH!UITn0(-h
zK8;6Ri|4ldy`}j#+YlZ$B%Th<UMcok5fb&w{JypHrVZEy?&E2g@V2M_gQvEB%{y%8
zSFuMG6+*f^d3!7E2%!z&0o$>Ooy?w~+tXWn{vL4q>E(66gXc4M_e1*&C5xc|a2jUA
zwA--9y<}+mSl`n~ZWV+|zqM1o+5Ld<k5kaOFnYp$`FRlZHuH*8gZb1`)3`lWf_h(m
z>M(U%nc+EdD7c`lxZIXI132&{Vj42xRN{U}Zv$8p99iI_<OASM01Ts3LwD%=<Ey6O
zgW9YF@CI!rUjFv=%gvmN%LR^S_}U%6fd4J<|D*(b=sg|+6U-;!+s{cw5S!nj(!&4A
zg?$Ru%4%duFR7t?V8BLGnNM|}J^_R|0qw6|A>Q;tJG(Y+o*UekUm^Lqe79vTL*rUA
z!s-3gFG`Pni57!Lu!A2rkSBUwClNyKm$r=*7AB1q5`O4^<J0|jVFdSaAt#whi*XS?
z1?848lPOmVc@x@lcfxBLB(vrhVd|MC?m5@eU$A$;i_2YL{{wb>@zN0nR)r3WDdz=w
zWVwa0O5M|qAicMK6D-XEr=<e*1iw&m?B3&quF|LXtIjZ+PmM!G7cG6#l8@U0Qgc~<
z6*O)d^9TEnyAgL`-Ak=-Cu@My!BS>*EgL&`t{Cr?&<aeXviG3&3Od!m6gv&3{a}ij
zJh?A1u^77J-DAdP5#SighME-pU37kVtdDt=T!8L=I)d9zkIF<;%`OHjj{kMn=LtRO
z%_8HHz76Prwe$Oqx8}qVDRbUMwYotnR%?0fn9YXHSDSoCM;z%hTy47bT8@KyVfEv5
zeJMdl@H)3#%qdrDmC*8oQ-&U@$s!TLzUK|0f{F0@N-KWnen~MI-Y7{na_~pzrx<Cz
ztHOdZ$!nSs$$5xS_a^Ls18CjB-64pUwyxqb!r{^Ms_e3MF6Cym6`Q4k^o_QQr7|yD
zM~M6Hu!Cu3JZ!_#lzcq%Z7-S*`KrFo$P3+OpuLPb7qRc3TBvu@a1&nJ=>+T4)UI&;
zFY>^N>K6^+9)Mi{MuI9^`ijt}KFU?E8^Nxdr?uZc12(P3Uaksy)nKSar+!__FW2Yu
zk4vZ`2}Sr`i5Gw~^=$tc#Vh^1kMtQ?2;>JZuP*gNhO6P_T9=MQ?wAkN>j&0TwT`E{
zrJp~0cJ<8(KXUsF&}RBQ^UYkj{>(W$nLoD$0y1fgiN${!vCZ%e=$Pc9P^K1MgTMgf
z>J>=E%|7H9%bPvVmVcw4Ks=qGZTzIcmHRu{d(04a0`rd`g$Z=uDdid-P6KA>4;Gkf
z^8gE0T~z$_xHh4wgxZL%GHM#ZqI3V{8HyuIx1wvSu}an^=pd7<wG1MJM|R2|_9%<I
z&z?*AElB<LGnWsiac1vV+u<nN^bDluZ&7MExNd`wjHE7UZ(r9XI9c+aX16i@zwXBC
zca}K>>Gkdut=~}<Y-{-yMdl)d5b(YHTp&=f6G(MML~?Srg76RPe);rF(tVfFJ(Tmb
zkaJcNaK9C=Xx-BK%hk1^upET>WvO9+YWoCRYK6K86J&Xixw&aE`m)viAn1QTrF19!
zLi&Q={d}~BY+p*iNR;R+wuhymfl~7y%wIE?!@6_Uo!EWF3DF}Clw@1=UGC14k1FKk
z6InAxWg4*NH5I_a+VAD~74p^Estb;oW(kMuH_lxYZ6!fFWB%tQ`uc{aH_s<)dqIxA
zB07zaJ*1C>)ir<xwC0gvD`LT$2ZLU<JytXmqnR<KdmpGwmh>4*eOqfU=iPch+zY;g
z4)JgJ*DpzK%#7hA!#A7_!L}@pL&Sayk4HdA1sUAc1d6ipVeZ*K%5|jGvBd_3zoH@{
zUfE0Qy7A-ANX&WMR<f=vao}%@-}EpnA9Rx$xcf2CN*rc@<+{I&_O`3N-!}Ym?ant7
zcNk{;@gZ_~TxSDURp479t7b8-Qz{!*4}QKhY@+u-R@M5nN`1OC>ZbANv0v5{ae|KR
zP|#Y(`sXKAp*N?(l4`1X97)W8=_He+?a#5A7GB{2UFk%ue~2A3FN{Skdf;sf&JVl{
zReu<6NGZRsXDTaC6xhEy_#KC-_k}t3;DNB>R=`29f&}R>LVVd;5gD+#g@g0P4Zn4W
zJs8iVyxYwB0Izq!+fgwHG8nT?<>Ayys~j^Nx7)q-5EqMaCLWg4t-Y?jj#L$`_|3Sc
z<)cSS@V)nktjr`6uLGs?zDA|R@|#cAiHYXqKiy4k>IEpl5>p&twe&V~f%N=jpaeL@
zi7gw|G40(6bScZq8TJ>B-^=h*ViOq?PdxcwzHuqr<Ur9YU0e*#DsKIA)%v7E@4O#9
zfmM*X<%29wV+t%kB>hx>7egtkko>BTBAZYc&<voixo&hGgqqx8&T0ga&Wh<H-WcAP
z+puuyIFgIT8ow;2v7)MZZ-q{JfWwk8lRhK1Et}qkf$p5hP7>$!a7?=#aQfY)VPj!W
z>27~3PTPaI3blz!KcK--rAmv_9Yu%k?*&DkTdn8sp)-`Q2jDKKy2bbTW2d3hPg@}Y
zuR+lSV>`4aj8-fPEoC(<Ajj~*;O_@o5Ldkiks(!7di7kchAfIBw+HiNTTMOgz*j6(
z59Vs?eF6*uEvW=5qZI=mj#vYT$SO+YtgFeiNE!q0yH!57wqrlMR%M)1FDdqn`j^|+
zv<Trmpfg^RDk{&8HQQbl@3g)6&+tS2MD7C)Xz@R1gBFPDWKCm_n~1GWXGg9MlXW>7
z%DE_oMX3+EG?N;-PWZ7@J{QQX#Og4eY9h1Cw_oY`QE41foaFpjTD=9HfF04H2$sl+
zV<!>z{+;u9tK>ZEg6MhOeU{PP@Az`<cu~=P?pF_8SMt9s`Rj;-X9^H`I2FQ<d$k8b
zdQW<_L`;Gka^D;lyfta9)=jKjM;5+!%@%x$HhOqwxp;m+oZX(8W(#;lJq5j`Nx}dN
zoseHl<Bj7hs0FD)W2|iW5I>IrQ~^nY?s6?XkBPPki0t=Z6Vt@-hDP8lpMoXeYTdA|
zH1Dc}JZ}ch?+$y7j9jQa8+AVjpJsGF>phOH-Rog3gp<D9j_EMXKF&HmvxiBmQH06v
zs!1?yAq;0)2tl6W!6((IbGL>Fzj(6<`u8*Z=7PdZWP^P-%g>&b`LXx0dZNSj&50yB
z>MD$%7vZK^6r^#3_4+5+?J4G`T>*&GwKI-_FSuwSjY~ErI?aQ=s_C3#WW1hP?q;x`
z13<#A50-fZ%fw~LQQ;ys*O*$24OYMPf6Zs#-eh#KDo4ma^f!XM`?zvN>d0eUVcz-G
z4-#^`So=%rt%j&QggM^7F~ocs8S2+qKet5p7a#2N%7iI;jdJgSUnIr7q3Dh;SNBSm
zmiEYcSSs2@fQ}Q@TPR0ObCOALRN}jLT??=jNHj-!co=x?E+6<W0n6Nq*-xo6?b0z>
z*clGQgfukR9Be9y^S&(nF%_YIiE~(;)l&;<frM?Y_t_?|UHg)(J`%&T!~5hkd@<cW
zX#D3FoqrJzVf<Pg=Fpf69E`ezN?P0UTU&dVFeRfD+di`obxp17cqxUHXi#iX4mNzy
zAD@TG=nnl7F=vfHNh2X@UvH5L^KVDzwBA=Y$)X%lZG$Hi2R;)M5xIKb%cJ&I7L*Zk
zA1SbyV(-a6Rvrt)uc3#eJ#=5w&mXHrDjK}tImA6M?<rzL;$!xiv70zL_S<ora*|BE
zBT(Lk(v*Ft2fBuhuRhzv|N2fgy4CX`RopP#-#XE%o}!ORmre`5)v>cWgaY&{H_x^P
zj0JbR#T&ntaZ?}dq>n^>IIz{9+5B$7mvk1~*GCD2l!L{b{k>q80PP;IF$*b$$mU|V
z#1>c1SEkx&GTO~0($=qm+a>06h--O%#nZzsrf`|y@RmKjlToP-*O#AS5lsb`+G!O3
z6%`riWQW-p?vb4aD}2^(f=8IndE^!4Mn|E6yb~8^JD)z6hlS>vTbthegeMl_4+Vh!
zoZ!f^{v>V>Wy0*zcxl^zcl%keoC)?%=~t(F&e<JbMtT2i>N|xcXnXT%2u4oouqbWE
z*O)}qzmQ*sHuf5OR=Bo{Fw8mb*Gp-xACv+z5?2WJe~5oCabWPj?GxUzMRdz8>85)R
zJte(O<rz?)Nnh9&X<;tFP|&+h^KI!KSlx&&XRywQ4C?#;re<xg+5gD#v<G?3j2mcn
zJ)hS&DXL(Z9=2I@R@V|In=@zpj2USn^&*3ItE5YWL>k(kWeXL}YfG1-v9LJYuDskC
z#TU}+q_buy%rE6Q0|@5l>~LX43$@s|xK>XDiL`WBep7ZaRqN>+fJy-YnR2EBT3~BE
z`ibV*Np5EQsaL%yhnc_6BQxvmf3W}>`R8cihp0GS!JM~wJoLFKmaT@TH7_>-qb`lx
z!Vg(5(NCRkh3>W~XOC>3ch_#+NxH9oa$=vLy2)H=W_li`v<?|R`z4t2xb+9)65ymQ
z;#g_##cR5tADaYS>JioQ#~r^9pp7h**cxPHmdOgDV*V+AJ$Np5>)C_X5nIB;J2Tsg
zm-}J0z3Z)L>Mq^eD)%3VpQ_QtSnhpR0_FnrUgz7kmE457d)jGe7gv!!?!<0ziR~rH
zq39mGMfi25$XNAOlQ^}pEu{y@zhe^*K?3*EFJ&hIdGw9lNs@UmfNcAN75Qtf|2*bX
z1}b4V6F&Q&M%K7}t(wV&P)gb=EOF`^7@RGX<ha~#O3ve9Ez+eIOIX8tdNANNBtmvW
zu5<j0Q7}xU5i@~)AJI&Ji_=-KY^PnNNx{ZOl0;LcfA(AMRJ6P(uFZUU+q3yZ%dTs!
z%Y)Z3rL8ZWUjO6YtUg}Kz{N!qF1N)37bN=+_EFZLYT+mPj5MXX5kje~g43!>deA<o
z<6OI7Ke-s1eN=6FOpJ!9PSZmlaZzI(%tL=*L)=nc_BTZ6Ka9vgw)z>5EXihKT|$7U
zmTrH#dM$~OJ5(Q-loTf5ch5K5=w=^v6WN3&W8l_xALGKGSofb{D365%US%5VAMc0j
z9@deIPE&u$RFg^@CH6imidb1+({x#Ph(u+SifB5Oe1Gq2qpzvc8q{U5CbTB$E8rgC
zqckmma6<yylpZla+ma4XY}C$TY67xuR=vUsI{k*PLT9crXH0rr>elZXESYfSPV+YT
z6!5VrfzB{DV9ou;o_*^0GTOa!H)RPRNx!pmx-0B{!4T{Rpe_9tt;P0w@`;lvXj!<^
zOj?mZj=`m<w5A$zmKGrD<2O3dC!W2`zitNVPjgj8VqlwA5J#53F3wYC&1<Z4`nt%+
zX!1zMU1fHI$t;V+200FrHf5DX0epGQoBx^=f~p7;c+@6O#yUk2J3!VvpWNS|ny1(#
zYf1CwGDCTJJ`*~84(h?iFF+n33V0$e^YFS)T$ZKcc0+BR0k!8;+~X7Sc-#k~VN(G6
zIzvLISdYeUUe4go1=f0E6jb8RVkxC8T+KPYk8HyC8v)z4Mm~SfCSJp{PlH}+O}H6(
z`F(mOnoqz5r!AkM8_%6SSd#s7s{v^NH;5NXuWVofIP>%0b9}L1^+OyH5tFz&JCBk+
zg^`Xol2~FN0VUqfyulJ^mHNi}_0$baE&}!{up5~~#?MJt3Jg<XjXC2l;tY7y276`p
zBlTSyFc{iu!298(;^LfS5`_dyU@wJ!{fQR@+TGs%Nej^J`od`CrSQS3@dxs{Klu(0
zY}#=wx$G@Vt1KJKTL5r0Ao1L?Bn|jGGm4hM4w|lW$JR!$%1$+5znsOryyfB1dc=LX
z>^5dkKi!0Vc;4xLoO`@>d{F>~4>`;<8^VVE{mX0T4DlP#o&%i-lPj}R&~&O6)$0b@
z{tu{BTzCc5?U~#jTM6X#oE<nd1u{i2D%0!hPrpu%lPtdVnSb^U*uc;$E6W;lbBMCz
zv&}0=BV?li=rq^Y=%q)X_%miwZ@CqZ4gtZPT?@ENXFlGJ80M$&A~DnUrGqj50Wa9M
z_82>Kj81YFPpz;7JUZR>jL4PMf+g$rs|zAUiNbpZuX`^X#;rcCkRrO?`VErIb|3c5
z1GH(ai-V@vwgl9_oC-2Y?+}7J7JaZgB&8gorG@2`Yb?pwV=@IPeIo-?H_UQ((~4s3
z>f>>%dQLPfR16x2rPLi4unHCyEL=tCa7m_jB~2*&hvUk5m6y-M-W-vtbLVl9?PqMt
zqn^opzo0t8f8ISa4VIO_{yfp8Z@vkOZl8caw=z+vc2({=-y_@nh}Wn}yD@u~)|lT-
z(TMsZz!*YB!lJPo6up)r1fCR5W{sJ5sIo22q{>o7LoNzz!%o$CM<AL8zs#xa;`Q4z
z6oskNiLvLziV8WCW<GMb;{eBEQVaAC{2kN*85>=!g+Mf3jH|)<Ux%i*kaTP2k9TUz
zN`YAsDF^NbNIt7z+%T$?YPzO8YeMMIU}Q2CvC6jT@MtIqnq4`fs%wlNGQ<ubQN3;m
zjVRedi>q5&cuUF|0k&h_ZU0<XKQba9%bLx?*{jdHabxpRERyOgT&D_0lnF%5=yD?L
zJ6Q8L%)*f^54ceR9>!kyN=jI)({V})txS{6@phcuVUPFeHgzI8O2wS^dvx|8lAn(z
z=(e2Nlq{H!y<xlmsMlI}wlY4&dF??yO@W0MnofYXvG4O^baU79PtGy={$BCM*Xpdg
zk+Yd;VsX_+`|0Ut5E^TB>evplGtqqk@VKZh3bl}?9?yI+%)J-muQZ=^;<%fUppzfv
zfJ5oW7m#`_&t~5!qZFB}+cI>LQ=|mjJD_)>J<gh&`pP$?UH97~j%qC3nnBK-<2Q^}
zXOa{f=n{9FC;7QDiSHa+cbi3{Hmi-zT_%~r=ZGWzgVgV-f8y98#OMK#fJH+f@aX#w
z$31TgpKtJV^|`n!7*e@46;<;?&;A{cLiUFEr&l7+tfCv311=mN-vi(MZ0Wm2xMs}0
zVhLK=+Vvj2x?*=)1Xb4nIuez+#FI$AQo02UMlyv!EaBTc)Xon@GqJOxD_ps%>WO*K
z9?(*WyKlVp>1u&56KVp&e5P(y-__qE$dy!OhFFR)a+x(N`u~H2qj1-lUL1*wuUl|K
z?2HM3P24@-4UNmmI={Tb9?_7Tfoc;~xnmMD1T~5+$5c)0k2_5+o=yO}i<i5#m$f_O
z00imj_8EU>PR<3(Gpu{t@u{l22Qb&iR82$@amGPU7r3EaG5wvfdmaxrS!>54m_hG0
zGym<a&Z@lb8n43xaXs;+^fDH918_fp+txH@=a|Dp`U9wR!H`RCJX8h^=u{DEA5XOU
zX)hr-bhyrM6xK)_0c^s*eC>vkI_9<8b@5520)4<8-I`CwK1oGB1CEPb{2`7LK=R&Q
zJPa81J9nUH#O=584xZVyt6Lo>YIskUp}i69Ej|HNrt7|U0Hr{B`|MioiSfUo>#PKj
zsIXyO(tuX$U;E<kB+f^6|K1WV#<;JnUtQ8a48|DL%>6qi&J}BCHpRTtn^M+D^cbzL
zthsgPJuJ>=;z`!PSG{cxvPR?YztsTe_{W9Amx)F1;Cy}`%QqJ<c$vkjNZ0~&U9=0K
zr7n`qE%{x(qKlPadr2Gq$z1P)T8N|b>;x#xGjx^##JG@81(xZ^t7_^jD_~Mu;-tfh
zz|bRk3r;ALeMq&pr`2cx!b*F9rC_5|*#v-jRbMfsQ3J=7LU8N^{cNvZoTTsb?r!8r
z6i2q;LuSU2Z8_-tU#P3`gJAIy)fJ;FQ)~l@po2nAeH`>PvD+v6bv{ohnkX_Gd}^?6
zYC{Fgo!>xQKBL-VYX1D)>AUn6Ml1NY4?ttyiXdl^g4&CxV7JtkSQX(b(YO#BM)uS*
z^*yj9CFBo&SPgjb!>lUf!FFG8a`xz1h{6S8SO?qNlAoLc9CLhVX**u7YBmG~o*rG{
zWAK)?wq7cfR<P6Tw?zU++w}Uvh06KAX~tQn2xOo4rDV?R7Rh^7a~tVg4TW2s$h6B#
zjC7bWCj?NOp@jOcqWtv~w4)sI>FZgSzSlBMtZeW1sKXBr-_P-#lihcwOh;h%2hJbN
zyN?VJH7i8fg39>6k2{NRt8^Wng_g~WMg_6*E<tFSj9H=>TzERxq!vC)%Jf+nV<w&j
zZ)>~0WTo$1a8#!f`p!kB8pvD!86M9o2tS;7TY-%eq7h%J%(@}!gU^XfWd(J2i^4uO
z`B5XAoWej?Vn!*7)PP;>RqFLAvU%0}uMeVY<lw;dD?I0Hs{0n@B7?D`IWwm-8TtEB
z-ahgtl6-l&nPPo|?(LqwxBgE&z;o%?34?GJ54xU;3QYT?Mj-(!nWGB6Fll6cgVn(|
zJz-5>RZ)+ftte41`+p#O=b;*F&qGL-&5h^<#KRw`DpTGZ)it5n%q1GyX8a!gK-du<
zZ0=201;zviRWkwzNr<z@Xx#H<&7i7vyD29{;k?cvyxzKfTuoEA@|H|&KPAzY9Un3w
z8^vpeiK+&j1ae4l#4}6xi11_kOZZK8$Cyjw?#;^#sd3hh-r7AY)@vwT5k~HxcdBh<
zUs>JYy~{@`Eyp*j5g!zTV-?;0ek03oOklxHNGNT^1OL<8W*v6<L+4AF{1FLdJAv`A
z-?Tn?!7XPfo0DKp_d_dpDh%CB&EW#xG`Gm19Wk?^J794o>Mz6SE4H3=PiD;IZLoDF
z5#OXJA(EAb`Y(;eNLd4|VkTWhMZ_??Z%YvMts(NX&645YeP$&Eh`bH!b-SXqgkhL|
z`>o&p{2gzg?A4E0CqLNgV8hFuN;`2d8<~gm7J_)d*z+E%sO`zWb@q00(C9d;jptOj
zsu5~{D-q4LV<E1S$~jtnDLNK%Ap~XrD2bCr4W^DsEfS3y^Waoam{R8)m0|z*1G>Vg
zu)dh$me&wC-y5zUb_H!{3MC(7eCtiC%Hu07O!RF9gd`h;mRI>MJ9b*9zNt_T?#z?c
z0>l1Q1NNm${2k0?H!=JkJ>|41i)+0wpGys;YHqsvD5s8WF}CSIMh$IoeF^lJP3hl<
zfBNgf?yBEY&5?8rjpfCbhPx8>@HJ*9-3<`Rw)VAIxEnVM$y+`*^J`1eZRk470{c?;
zKWaVk)~GNd%4L41<!ADehTPyZUL{hId$5XF+wFOmF%%^a+5u`K3UnVc$AQ_D%~$l$
zD@cL2-W${^%!9m43p(>bcG1hl;%}|`<Be7SW%vN6JG(-hWE)2y+Cy6i?sa3|>3KqS
zQ;;^O)`1rVoa`*%+Z$|kx(zI390R(_x4=&D_o`jo-dN5V5R(NyKDIKG{=HYF{+C(8
z%y>AHZqu31EI0J(Fc-gu(<|}Y4^A*$B!*;quDzUztzt2K9AXr&gusq6QOb*<@of|3
z=$G?+GH_A(GI`3RssU4yPMCA7K|Tc_b)7>--JG{CS`Y7$oI@&sqbg$?9#s;nD+b)1
z&6P%8u?GiDN6b2(`J5_VZYK{sD1OdW22T@qtow?*S!Df%z3;>PY0ani{ygyTU*xm}
zgpkkm+#ffSEDo`wpxm3IteapJ*P955az`l}o*2&SpuRw<cG&P1_wx{KZs{i`*_?_<
zHsxRV4O<{jKK{b(yRrC_s-T?kE4&8saE20fMA&cL(=e8^F?5hwgoj<ZgsB`&%W-b5
z$W7wx;oQd~1POZzqKB|7I{%K^%W2SYYze3b^ns7qdn`uzDJ8bLc5@NGT`~Wgju@nK
zTgl}#w03)wb;X%=dep)0rg`7tQH804yEgnz9knJ}fYdfZBq=d0LN(zdQ9-NKZ!TF{
z?HVg$KckU)j~15@&-Mq6PXrI&87>KRC0d&N5StGsH4-(2Ve1P!*e_ErH#fJIFL%N_
zRRfM{GhE#-4*}us*+zFejs;d=Rp$L1xx(Smh<Mw}ktW3~ve#l>C@!xB2b8wl_md9t
z1+G6EOuoLt%XZ`$Eyc3pSWkq-_#Pz&&!MR922LHj^_ck%n!UvPa-L9|p60Smy!|+4
z9K|jLEmk@!AFnImA$tE`Z<?t8fT^{8m@3LD$k#F?GMBS;2uwo|Yl@IpR*a1N`gQCV
zizZzaP~s#TO>|m&kATfHHfcCBSb?t-%H9Oh=(u%=f5DjG=`uGx+5Ai2SoNM(;^oQK
zBxmuu;z*$L8c#^=eJqI{F$X>bzaZRSZf4CvG`2%N*vXuvRkZf@o)XSs%dDZ%pj!n*
zKWG#%0a4ZvAj-{9h?m08gtx~NG{qJl%-eBm+YX&4=MVl;v|L^;Y#Gm~ueLob{wQby
zJ8>7?hBPcF&vEIbf4h$$#Z5WkDcYY$H}~p_PJ%oj$^&XqaX~nFziDMT4;Nz#ZRy$w
zs;jLGyM7yFC(g;a%xwC_crs_rS(!UAxRFi<{}xGrNJf53q=70<xaEt}9a-un&&;A(
z>*l5u+aBjKsIxIhVKOxnH#J}tMDV`XA*m;yT;mzRj1-_8%RfDN^-3i@ACH3|hF1Xe
z$JBbUZ7h3QVT#{la~hu}iEZj2xwBL7oT9|Gy$f^MC&5-l3e`SeLhYlx<fAT;AVPo(
zEoc;COA3;%(4EcuNjc_vZX%H#%PPt6P^+j+)1y0Bj{$%Q&$fr!iHQDfRI1<k-%m&<
z6K*uPXAMTW8w^;nTY!E!wKRuK{2xMo9{AVnFN<V9P!QnT+e>R{q6B~RN=@A3BlSIf
z9k=HF*NSr!a5sPP2XZsN7Zq<eCE+fu#%eR_=92C7)^i^)=Mo@tv10fe4*e1Lw^hwn
zVzglgW)cZjm|FQ7BRtuePA^WfJ{)@%jpS<SOv1kjLf~JDB`N3%XQ2jiYCJ!w4iYAc
z?Y&jbo9y38&%Q%<|F0wEL@zwT&-vUJB&WrFTMW~qPv;=7t%q+u22c!qt(IPM)Lw#B
zQ+mB<Ubgv2aEgaS;;khGN&EOL`y8?HuowaD8MY#UcgKHTPk@JOCDp=@<Kf@gvS=~t
z^gn!#53IsaEaj%YxcCrT?f6r`PDYLXV?3(v1jo5-;RoOEawx&~Oj`0c=-O$Fz<O@`
znetib<=FGC`Q;?8rEB&7ws&;&>z1w-ax^p7Br3B46lv)P-v|D+UE`PU6ISx6J3yMX
zx~8UVI2~BqoS*N=0kQ3dh2U!Yg^ZYxS2p>JesWYWY4Jj3KlhAggMEzlhmZNU<dBY8
z&ug9(=u|4QN5%yfrEbl~r)q&{?rAdEG~po+fBuK<vv14H0EzUuWbJ|Uj^$+`0M=Px
zx>&t)6Yz2~#6IfT{qP#HL@1A&$x`(egiTRKXE5oBJ@IBB%0GpdMiFTT@%-Ln4}n8|
z!Ji=<Z1-hz)$TRw%!jFKy4W{aGKldud*xx)Uw?nXukqq&Cj*(u;+H60+C3gvR)w(i
zoK334ZKNC*ru`xbs4?7e7+MUh&_XE`)GQiwc=H|=Gx1wG1T0(iqrP9c$uARmS7)r{
z|Htoogj3HepCTPLSZ?5U;$d^`KDNoNguv}-xFA1;!mmS{Nw>&N$|3F~ABh(rUCMQ9
zYiS*)Cs%B*UNSH2xTyiPkFo|e5CEg1+1MFTtFEbDup3@!w!z*-7L{){?OX&<>IG-v
zLvF)oou<`}@X`@XgrJ5&LaJ!RUp#UvJ;qL6Ejeq55mp}3<Yz9DKqrRPJDjIf+>^(w
z+b;i1M2Fu7H-qyDfej`TJ(qQRPq%M`&l&x}W@4f4#u>D!*aHpjUWg}e#bykZqGmV^
z0G#))v1SXF(6yR5-N8$#RPVh=F*!XAq|*tmEs)hS($7j?cc^>XjM{Y4jH;oF&4qK0
zoXLIzMKvZ8WXTT*_-2lC0!kWyNDhEh<Jf)a_vI`J0T{LQBH0hNOk3k<6HW%8MP%93
zJYdJ&hsR8DYp_(}@RpZ0D<l!>NQKBMQDJVgR%v^SJq4whDKMthuFER<s+ZnB(G=*i
zi~J>1<8@kKL6CUY)v@E*>I359q#Yp>r(botI?2hE<9Cnpa{cKgKCPsrQd17f=KB=%
z2}p~Xu;Vu@RMy!C9Wm{A_tB0CZC(P7On0zaBeaoP=?hH(+rUliUE|s<mxYg7ws1f{
zz`FiBCiIMtT8D5qaJe_ZALooZCUPTIGnLT@M7JkyP>GdQO`Q0d)SS&KD6o48qaVp&
zL^c0KFrdv}uiv2cZ^|LH1L6UHbm^gz>!FI4GCcvRn9}stsn^V10`Zn*;m^qJ8k&ul
zQJ$%NPn0O1J6@rq%cv3;276ybAA#E1@(j}QX7c3&Wu(W!la@(c!)F#Qv=dHw_bK!V
z_Z1U|xXy;&S$qp2)@Ln$;!nnS{ShcUQPg|Ff-o|xl=?R?$Fb_>dyh9N6G;m0;rRw4
zpHzO1kEz)vF#O6fs&DJqgfx)<`X^Vz&5V%^nM9!@o*p&7z}+x~d3N7aJdSlgthxHR
zi|Wh}(W01ZxO!YC3P1J<uL}d3*<=8?NMy+jG>in&b2&znn%6d-hs1#^x7&VX()Az3
z)^|yyPd|aO+wKL@#a&=NU5h8;C!mf)?=vo#67?M8BOI)qw1rb7q@$x%^5ZT!p_y5^
z*`;Bd^BOY$h@g|)4ghoFs+oz>aX}N?p6rZJCLNzPE6#H^ce@tCGb2^fk1{?yr~Uu+
zApNqR<MeVH++=;L^>X-W@FMzC*BMu$#8xF4#gk>3eVZLmBL)w_{q#Qf4DRc2W?{s&
z0L~gfow+*ofDM*Ew-Z1kRYc!na4OK2#hz*BpPoUMZLwN;Bf25H$TElH;6;!eCN=>W
zqeav*bp5xKb$%B5C%CD6uN$c6A?e3$4cWV>oy?33J~-bwQ<_QQtC@a=X1q&So|k)e
zY1Q|;bx+xVJ70W?!e+_50F?oQhRc)Ic7^Vbjr^wGOrWW=M(56crTMGlKWAX^gm-qW
zE+adts`^pXx`aUTUV*v(6F4<WBFvMQ6cYNi7(3+m3ioSCSy!4QdPx#qWpejbK%}7>
z^<NdbKzVi2b6o|sk55Atuj_xSp0VrOUXAW{J~zxK*la0;Dq3TVO|+E~(Vltt3*UU@
zWP{36=kL%~cIbyTLkp{w;hQNXXzIw0Sa8_CSTlMo)O={LCth+P09ILgZ>!f<ELwsO
zI=h8$U~#6nfq;GwU?HZ9bXL-1qse0cLSyU#Q^&5+O;YqS9DV9aEp56}oWyjJOZevk
zl7tF#Tz1@mD5<#ICcVuMDQ|X>l_U;7jutDV(5Dkubcsp@FDv*g?wSdU`#pxw-fw;s
z?-<1)jyTewYTLH2X2F=^iJ_aBoPaltXx96%yof<h3MMpgqX;=Y#X$HU+jam^Vj^Nv
zXQx^gqCU=p)=5dWd|Q5@m&-5zw<@q_oUQi+<(e;<Bo~(x?#}V(8yPxZKT>DFblD{g
z<2xQc{r?9E-5_#up_^gCkEUPYW5Attg5jC&xAxRxd#gjtJ&=HGJ!H6`@?BMgqD*}t
z)fk-xNB;vdPz+O>K{IIVnWHYK+dOZ``9_m&o%btR)nQjI^@1zPOEOC@b|m%8uGfma
znC>GtSuxxzkrSl~sVK+Wk~w2Ey~#G@K?`fXmOCZ!yEwy%xdCKP1wO%5g{1(sH(S19
zO8|ca$$_r|DE@p^Y{^n!c3`#R`8n5xE5S1uGPy%$w#FMJ`_Y@`xSynw#}1yVu2r$F
z?B(Y6FK~*S1C;-+?hk=LqN<v<$y>oHwBabvs!UUVkN(jHHIlA?zU@<9&aPO(e2fJ<
zW<Ix?Xe@?xHp7jot^51XfXBFJp_k2oCqY1fKIXc6LFj+k_cADaH}E2#I|mO_^$20Z
zFHi|pP4~G`1PViei|WLyw%LL{J%GgnSgqNw`ahsZs+>;cJSXF_WXn-rXRDFddI8GG
z?<0k$GDtb>Kgj$TaUHWRiLkq7UUjzi-KoAi!0lcc_ZY|cU9Oc7{WF$G;6Mxh)E^c|
zh2h_`up{^XQo)DkA>ro|EFDn|uE&dj=LalUa`T$Me2xV*sNidd@VAzb>(Zs!FH532
zp(iH*N@}#zo6AWhNUi_3^nXJjvXEXDD5Qo?>Wc*sO|j&t?;vwwZ7JyfXOAT(-QyVg
zA2`PVzpGBV{+|8e`>C$wgMa>6CR}cqe#GVa+KqJ72k|gHX5!UnQfDGiUpPz#3tIB8
z>a8Qx?{x~Eu2>8-`S(}R{okp(XH1`M<Jq#2Y3z#gWe;P!M^#Pkbsb&r&5;?zz?s=@
z5AGE0fF}F$E1hZVm!cFMb=mcuucC$qXAG8DF6XT5Qn|nwAW>{;gxXUc2+$IQMxdyE
z64mCxd;y%aUMl163OS*6OQ~%RGKuK|qcU!RuBpY<lKhp{gY{x>Mm9PwF^Y;wOE&(*
zl)68@9E)+*akYOjKth(^I_(l~eLq#@JYu^%io{ein2y%ZO-K62@eMCouwj|ijbF{#
z{&K{|v|&Lf(%go9SVJ$Z9290dB$KN#ZXcBIw7mS6Z(jJ2xPl5SVOo`Jf-LGlWJUZB
z>I<DENTQt#$*r!<D&R`5CrT(rQP4xKnhb49fbN_;V*d$Hj}{wg6Ue>WgjyB#?2Fj{
z_bo6^f4x@g9Ynk0T?8SQZplw7fD&Rir@GDOl>lUNV03`qt4_BHW({U9Zg>l<`lEn9
zb(rU>PaN<C6bC3HRlWwcm#$CCPzL4?-re&EeRf#xayuTJj2`t?Xw?B4pZW&+&R)(v
zJ9kJIBerG!vlN}KKqM$A>{F|!%O42)=HX$WlF*&ZeJ_YMi?B^|*VkaP(#<0hU%WuD
z?E)_=K0P`gFA1UWi<p#sb!mVKr2<Gx7_g~kcdAdpS=cP(6Jlhl@>kS3HRKHRDyQtq
z(Zis>ICNFg2rH97ff8~*B4Z_lsKwUNl~RX6c8lLD54F)Md_eIx`Gyl*%uUwh^H%Ed
zMpJJF7z!v$b{UBQ3I`08zN{4MLy=FI)f_%vVy>quL68C)rT{TX^z+Nab<Ro?)t(>i
zZB#gPi8#RC3rqR*6*!(kGU8ol!ch@ExIiaeiOsfagvt7^?%~1AmZAKZTSbwDoPz8d
zq{LbzExK1b)xytqX<z?#Kqy=hz|^WO85rU2q(t28YAGcpa2D|VUIWL>r~R+PiWy>-
zlH_aZB<6;9+2?rTe36$;H6AR8Sn8Lqgbx@MQO9d!QHqV)&Ufn=BU2n64r@JFNK(J@
zvepm0+@*A(?^LTM;OB^!mF|tOwHK1^m!<BWi4C;HCR0!bTXr;Lud~_fa$ks_Ukwi5
z)wW$&63lJdlP~cxvsO$ugDQk^;{&~<IKO{ej{`5aw96e{QT(1KlTF0^R)#*uUx3@3
zdP6zFT<i2w94uEk(mSipdpi-uoVs2g(%}D5bXHMqbz2y%;FRL-?oiyJIKkbexVslC
z6n6qYP~6?!Dems>t|7(kp8K4a3<f!8@3p>f&Um7I&jXUW+pAl|#n!VXlwEmNf$wo~
z8J%aayB_KG&knDg`yL!s{~Db!#K=J4VOR{qLigmbxJwA_J0xJ|KHv(KI<M6ACGLIi
zy<g;Q=smpecc$%;@oqu!&kCRRmZ);o@L}jLk|EO99M{XJ6wAs<LKrC&0jr>#E+JX+
zi+|t20H?EEo+dF+n-O;J(EVrM?1jnWZ}q}giqfX6fr(BxXU|p*U)T}iTZg$`^%<x`
zJ-%$}ga_q`US!ytBNL7*j)&+Ui-JEr@PvA=9alundR{^?AGo2vGdO~4T0NDFpcnU#
zqNGYjDcz?tBGR2HON02r`0S$3XE;+%9DJTZLC4H6F3)NA$N|SjGXBI#9fx+@dxJ5Q
z;C|&LAZQf-DKE;ASpjpj&dk#@?8GhXw#}kdyGDEbBkr0<`}yU0eS7<m8^1Xx!$>3%
zt}KwyZL+nZknG`;o13?)1gas{EQGm}k+%RHbKe_a<XPD_(eV!3;rEi+Q2ukpP>R_o
z9y(v7eSN`9xDNgWAvs`fGMq6GEE4h#?iRFMuV5>KW&U_R%cAs+KXM~z&?iPK;Q{p<
z!s;%s%EzE6gyLdr25^BydUOuwC}9UOI<>p@kq$A_H>t#;puhSPtR>}<3pSEEIA)T8
z>@<fE*@UY99!it*lMZ$P7qQCd^t@??9FUnC@D*Ki(>{amy0Z9!M4^Rc8rc!O()eBf
z=qX0RvYog6HgSl?Ps&ytj2EPZmk2IWeOQ|u;9GQZbNlPYZO;BjIav~mM)2>-UoXJ}
zx$cMNhxuaRF#y^*mr48y=_uEp7)CO|GM$1QMt_Oz&!pGMfmVz68I5Iev3>?%MrLXi
zY~SB(zjC_D^aSz^NPI*Qiu9vFnJXmK9XAZpJgBX0ZQGqZn>!<WL&4j43Q|^uRsa&p
z7c{5tW<Xo%VIr+0T7;=uNk1D@ho_?G*#hZ>j(@$M`LojrDCKru@0Tak94@YIX5Ugj
ztGW8<*Q8sW`v#tTf+O}$4qT|qtt4)OrhYJ(L&1^INuF`QFZa39snC+QRJ*B*CjqNE
zE21Sto$4Ksqjlt061J7~c;aA6<_D6swa)+eJ>=2b{Zca|&2-Mah}NfQPd0Nmby`om
za`sLf`oWkOyOd>kOKA|3CJg?3qqS!Qf_WO%x@-y~nb$8A@b_&c)jTxEjQqm$mB4z3
zB5F(-0Zv(A?4VDYDVs>GyMt^iGdEmf+rJhSsWOPnowot1FZ}cGQuoaeRgWs`5tWp)
zxBGw_H}KE4#O-@q9x&?oI!Nl|@-uJ|?us~K#6+{8*j|8Tl654q_F;Sh#nKWNkWI5?
zDvm(EAYe)MRO<0~J!eu^CEE2=VNy64>HRxSRCzT?%8AFxaPi-^dwAHswHXc<???0^
zY*nkh?(*MyhF|gy6*GU%>dOAoX)P^m%cTF_isorT-`?J?udn|>H6G8jpO%#cV^QZ)
zq-BVw>(SjGg8PPJzcJI4{C=|aivE5~`ufpaM=JIY-TTkR<3+~V=27mvP?WOSxQVx;
ztk;?M!n1&OX(w(Y%r1*=#M3$ISx%rG^zF}O{1N5P+EMsC0XtrWHNs-cQd6FL+rA?t
zanYGItlLUofs%upj%d0hbz%moTXX9kvpinjV~SH;#9n(wi11k&q<{+Mk6S1LGdF_k
zyJ6OUf^5+r-HCgAo-DH*x*g)Ovu}5ThHCtTZ@j^HvxXZuDwAEyE|UQ7=*%Vt)PG6F
z2A61eF-qwN<e8hL<HU9|?(RHV{hpgj7|zlX_b^3|@|ojR(=s!Ofz5o2GZb#_BzmFF
z1gKK$vd|CV^EhsrSz5-e*u|}!$FDUzaAYdbjV5UnwRl6}bm3W8KPebsiv9H<jsaWs
zFo6nxAx9p#jF_;+*^Ga&|9SO1!B5r*p5|zxhgou%SiJtSL^?EaZxUlE_I-)Vm}`h+
z%kuV`SzU<smWu|ZWgatWz>{V24Gufn37i+8r!3ps*GM+XLgCUUZpIBl9vi?=2&AEA
zrM}{@TKU3bFK45gR+guzhLOMWXM&f{yVs?0cVv%t+=64ZDaL^G5vuE$*Bwb={{ADy
zT|=$XM)M3dapcPxQ~5!pMq~8AidwoeEpc{Rm5vq{h*pqYCF5=?wnV$K{+AeSnxNbO
zymGPGA1@^uHI)FxA~lDtE^(b$1oCNal(7M#LbJYj3wj{fnv`Pr6WdgEFdvb%>mQOd
zkAqq=C77!4MAvOfR9gA1seam>EwdPNz+Fv^2LKNJ|5?%UeknvE<TX5SfN!9NY_tX+
z*|GiODm*`*@cfE2N253U{W05>OQiV1CN?%!8D_DZy8Iwx;04K2FWjJnQ)vKCgqJPO
zk=a|ga2QGIAbO*3ZTzc)Y9k76adF4}WjkP=WF2>K@(dWf5QFSql9U?cr^wY!*7Izd
z3v14w1rM~DJL}Z9m03yit<d$5%|!CXHD}cgBO>Xp)#@`lPl6Bdy>+(3o$YFC&I0jp
zKM6mDYMycB)?49*a8Kq#VKObT_RvX=Gx9ImP=(_BBq052BvLz1!M-Su+km@T%!b__
zi(CR8w=m&^^T%wki{;??CHmaZb43cWKJIQXAuYxhrMR)J%QDOLZa_KOT4cftKI0LQ
zxU2ujK)&>OW+r=^b2w=3O+mJ{>^rPn5TzOkrfwQkT3^^VOn|lHuaDzv{F-O>-p2jr
z!S;4Mlbp)xb@BUw(&NR})8|ORVrw4+ZpqT%g|gk7n@Ik97>NZB=peJ<bH8l|WDvEb
zqr;3HUroan++Wk)EDq6@!1<e_{$FpEv{>&VAM!3QnWs2}BIM5{?p#-Qz^jA9vWOn$
zWhlq|dnOuG?74T3pHJBTptSqt-1PS~Tf>vXX2(b(trAtTWjQw8RKkPZ>o)wBsiR|@
zlHV>z#S+Rrso*xF(W9FGS+l>Z^1$7C*c(yvmT|a~Slq$j#Njl{Q%>*hiD^5TW;oyB
zq2E1-hE1b4P&G7<h+}O9fbjMA29z(a=$;;MJrmYiKMMI(R--@l72U2rYu;uzSK<|w
z9mC>YbJ}w>SWjrmEqTWQnv{6<bWmPojBh(jQ}!XC!Nx(ZlSkDTjIc!d%mXdXwd-pQ
zBlF;$YFYe0l+9nzmg$uVP2lJ$IaNO{ruK~c-5Qe>rvrhxO>@4$nnA_hjGkuXIKlXe
z$9KV!ST>@1$&1~BlL0?Qwy2(d;V0~j1;dz&hlj=a#Z`TGz+W?V>M;i3Jh8FKBH*Lu
zu--D_*4AexXezh~khOogBqdr9*c18qz`NPYEpdK+pY(<dm6vgu12~{rSnJiag`Wdf
zf;c@|hB7IVVU_sC?VU{F+FM8(|00z|3b=<G3VG69CHIx!!9P?cA}?>Se4HZ%1juz*
zWPBC!h;`|A9tImJKjcD>f9#Hp1>tApgex+xY?pnIN}v-DEUG~?C!ve_yD$*MUnv95
z;AH(ivaCg(6GOML<o<hvp^m~7u*x3sxvM#Qdym+37UBJe1rkdu5?@mo^!!lZrwx|7
z`E`qjiO^$gFV|?eTlwFTTK_k1e9YZ2nkn6HbtgAf(l?Zzlc_C(R$!)lCOIk7olAfK
zTHD?{WW{Nr9t%vP<%`sO5$4YT#WVv$wQa`diyxc>Dp(2CnZS{l(Y|*+LL(Xk>v!{U
z%e|sKG{@KOx7c5`YAt<(3}=^@VA5^~<>upvo_>~5O$M?ECt7eo@*yyys1;Zfw;ce)
zA!Co=xjEA^sab)^sj1qgTw9yAqCpSlb^(r#|31y?_{4mTFQNw}Cmg~S(Ot{Bx;JjS
zc=rrEA_BqduTuCtMX)?WeA1MX7KA0XNoRNmruqPMD_Bs@s#3iC!{iQsRC67oT0-AR
zMd5T|IAw)pqi#aM2o0AEsAmHjz7Wbv<tl9#aX;h1?J*XuIN->pP*|VDdlUDq+s`c4
z!14QOzU6Y`-IQAUrKRFu75m~wGCMkS9=bWgECzLO{FTy*&m|2IPS_YB%AjcR5>@{a
zO7o8cR@|5;hn>91n#aBjP3V$6r&LdoROoQ<MK)t)Vem-aOGIpcB8k?W>*5Cd2=Oti
z+U?PEyy+v3*t*fA`9~!O!kB=JV0J;D&N}|h(mwVOico-HT0vkerJ;obA|3@NMJD;x
zroF|=HK1pX9(45o$I3->0EK=U46wl}Ux;k8VUsepmsnSKY-oAiEHq!ur&$8FI#h4W
z{U}v?I!lqEm7fggsWI#K(+09HqH@n*b>BNtT?OqWeB9k7(~1P|*In1U0|I!jx@H?#
zrdN_R8utaZyff2z960c&5QKMbmyC{%kK?EOE?oigi{)DG!_|~K-MLBUJgX9<I5crf
zYA`DVYlLfO6^0R6eg7w1oGRQ^OiQj$+-xY*7=8>5)&W9cV$W7ZKqlJe`_ojHTDqo1
z0I@>w{qxmrGUHn1RIXD@K$YXn&#S^?t(zyXM9#^~7!~NxgolL0Oru-o%h_rgG)k>P
zqt*#aU5UO~W4ZIaQ}9ezSk7y~=-KvDoF$g$)KhEsDY;3omysY$=K!m_;RBno+A;!2
z=tpLuL;tlFVZ_6zxZ>7jz}B-rQmg9sv?4%T=f&CUGd1bf`$Ey@V@-JIJ!mFq%>U;b
zi4yRoo$HnJu+K>^oxQw%k7%2`kD34jSMo76`A?Ctz!4fSqW)|(Yt_!zAWT$(A+~9U
z!vbp?3!TaOLerf4hWQW~P}407NtIRe?d@}d^xJb4xbettH!wfbiSGTDhL39Evhj}X
z792Fl=sk@rC<i4`&U}oKQZ;F$TU)T$E1iKk7#M_K>4z`R3&DU0!AePSD1IvEG*2ST
zXwR``nB^>XGL6RgB^r`yvM;2eW_k+qAHiTE;KuGF0>sqNfeeV7e6nSMPS9-K?mtzL
zKgge~8^&r@^OE2z$Tk)Tu0rN48yb8*_bx#dVL+1t5nAF|mO{#(VMMWGwZt^B&Mz+N
zySt}0--N@$!ib27+3;h4c@Q9B(OvxuHB+=?)-jyr+}QTBk&m1i`E{pt7TH!cT&EsM
zoEWJ+qzeEYyZS<7>7<IzY^o+FdA(o4;ejn_{#7?{QU~DnKQb#wtyPrG`0`!)&G=za
z#%{p}zC8;E*k?8&(AZzNBqsGVytHSiS%Px43_tAU&Ym<Y#BxU>!4(<tNtABMd%woU
zkXcw2we>T8q(!btORp0g5#b=J?ICEuYHo^WnbT<`Qh#F^*tJ#114|4xyME0~Jzhm~
z9TPH9w_{e=h%d;K$GNTjpe4R0N&e<0{LB5JNzz*+XHyvN{O<J9zmli6^36CdPIDRd
z1QFR~&Zo|f-n?0*-gb!lCsylsi1R;jz3E?|Z%9EP(2(if)p|jz_Vuv|{(^LEYIKy}
zv1Pw;Be=CFZcBAooSInETY>lkZAR-DEp5dlYpIsp=!(O-r41(yftqf#P3aT~^mxe8
z338X-;oC4|<a=cg1Snq+%LnTMpg3DkgFMSCe`X}upSqv(1IFHW0^X^m^1(%we}P!4
zDdz~~%j%39TKw>TMfQ-%o82hZn<xN=@FMbV^p5KPFQ3%7!jhAqzMiA6opI6DxhHv4
zqUuN9z_O=IB0^BbFn?!APKf^>|2yYO#yX9X+YldwkCpnI!X!<{;?AD-slF2Z4uqY2
z0s?!7F@)=#K4I|pq&1BNd!LslO*zI64q%Pm2bUf;wjTXfi=@(nMIIC1`zO+yTil-y
z8+-S4iD%*xD<$b_7BpA+?`;v(-pkQllBb%tOsCjlvd3gGW;9lEW2Gnu_iC{oookDX
z9>j^3AmUl?-}DFH9PbBuJVfor2M^7DZyQM8RZ;rQ5>x=~J=?sp@qg}_9dFY#DQ5{q
z_yPj$svVEtm$;X!iX3c{$B<W#yRkp1=(}ek&j)<@hs+(aJeM9-9M#la@$F@->rExL
zOLD(MT(YtC-kIE1Pe$z+d4wz$+AHc_!6*r)E|D<Cpqvju8~XIK#-Qyj%V(g(D;e8+
zu!Knk=_{&dYlk2#HS$Xd9bTLimNy8X@FgbTDibi<JP}HtUy&%|P}c6qxZY4u%3(&U
zlzcO>Y*Nx|<~)mC=!pE3Ci^8u<A(um4*8fFi2;CCDaQsTOeZag$N6Mea)aG*&>9N5
zmFel{>D$=KTygj))xaDS#X^{V0Swl|lR?en1=u%;UWuIOso2uC7^B7cU`|H@OrZ9^
zZBup4_hEvNE2f?iu`Sl$Lho|!%+Inaxg{FCW#ivqPHhmsv#|_Mq;9&1=GPb?FA!9V
z9;MBg$+NMjM3?|QYsg+$bWvgOnk7Cz=s5ji_6J0zggis@PqHoovy{)NjiM7V4b9EI
zP9(@R1NBCpen+6p3PV3yZg4MR|1p``nnJ`D4B%?V8A*NqYxlfzn7^~;AKwzLLMH#N
z=?1jid^roPGs%o(cRER;Pdhy;7Lhf}8E?kFO{%eDxi9tSmE{l2hJ3ZvOS705$%pTF
z9YpnZc^58@i_>9VTF!xpA^}GsUS2^HK@#av(E)(lF{^HCw}e5#UiDuuykx!Mo}B<z
ztgL9GA?WC@$M5EJ0LBQLpGhe1i8@U~dmOCj&7ze3O~fb2$iTT-+4t$kc)0){kIT0O
zu#LCmAIkzs#E6~{F3B0^9IWTI0y<R>Pxd-7V`n_-*6Pm6=nE$Mr5R1;rds(SwOK0M
z(;)6%$2cx<YzB=vGbgcwMgrRP=#Kke>fOS#Mr|A}4cNgP@ELT-$iugxx5_<lTsdod
zEw=?Tvo`d{Gy`>TC>keN4WG8!KQN+cmz?6WHIds_nFjeZD*U2Um2rYQ2j)}}U$NJH
zb()wQP!L5hOm)A0-95W&&m7__TH!$H`mVJ*hP@J2L`uj!tajDnJZ<sAW26&De0+{L
z9K(<`e3z(})^fIQ+zWhl{^rBi?6qu&BHONXifz|j<JC>~8*MfFBt{W@yFBV$=@Lp@
zw%{Rf{>SXP4q=7-ChZU{Q`6L9mP`(GMI-2NnQt<06WBSt5NL>@o}iVdHGNwVc?I|@
z0j~iE8_)OC5hhJOGPcBq%%x`@+vqmbMFYC`Nc6`+UWbA)D-rHdADL-}{2D1S6|i7S
zVm?^35Ab*q(^Dt>5Xz8e-EHx=W8C*&HSwujMhFToupR}*I9c63VYru@hdq7Lzg9b9
z0WyzVvYVjq%P-Q7M<HS8$W(oY_FRMQ+i}1fCa{WmHcBWas1)qS>Cz$*i!)?Pg-3SI
zZXra-f$R1Ox5{SU6{2m26HRgFe5bbKHMiQGl+zUz6}Kb|3hWrC8{`8OK@4STgkdUv
zS2a%au_uLlB{gWu+aO5gPs6a@4?cJCCZZ&{k~2{H6qhO_5FFdftWPH^G$czexTAkG
z*zQA+QD(qa*8G?om0n<ln~UPjW?{*b<!X7NFMX8xV*5E0>b97xfx+YHj9EPX*EvSR
z?VQCyqtcEWot>qcF(5!wrzT*f7DbAlIwT8OS-MAJ*8N|{$QfUAT_9G7wN!Gec1KwF
z?r7)P)!1B^SOQ)xqb%mmkwTG})zxb9B>xRNkVpbkg)GA%^&}SX<&UI7fay4HfsP{h
zu#V5^Yw>sEX;}_@B}Di<w2RfyNe?u^M$g*i>M&Rkq4)++Ns=2pNPW%pt8FD0-h4&5
z5Qz)PB6=kv4=f>1;|N2`F64!`-bz@;F{soQ`ibiIlT?j$CbJ~4vN<&#L#+E^^Av7`
zMU30j@3YL<618ZFv^sT&80-}97{2Wgrv|6a;3e;YOKqM`vwkMy*ShvuF)hR%#937>
zoTEkFvtjm$lGJYugTPhc`k{vs^y516#0@^*$7o?$7S=2)38fS2SH8%ewI3}7Yp{fL
zT4eUQ(bKEnj6ES{+pO1%)0q8FO?~qabgKLB<Ym=<bLig}n;w_eUVd~iU{h4a8Q9JR
z9AwjLq}G_Q@qx9B0}Nx0bVO$hX-!qN_8}u2twTMk)wWeEoWVHK6OzfmpDuQrXu)OV
zB#|~FJHzQ|A2IH7!_?n9&|WX&-im5WZ~|Fs53&8(htpu_@zGmeho5|g{Vqn+YbidB
z^Aj1)ntG}X$Lz--&x+)+R0F~RcU@OEul}7K`@4Gw<z|BdO4znr?27}_L64D13CIAh
z!*xje^SK4Vo+cA6u2l{l(oM$?jml&FF|Uleb`gM>$xaYQH8G&mridn%X=VLPuyPE9
z(bbkpJ|sd?^M*n46yeSV9v|7S+`le0EEf$nS6+Yrja#OjH!`X}{dfC#6)w@7ipHDt
z*UemPocc0I@?sja?hFqkjQAtU(vw)wZ`B>=AmgUtc&iyF9(e6v=?mzs`Aj^V{+?T9
zU)~@YAxA93E6FmU!A}e5I+KBsFqe@au|6ee;fF6&->nF6Xp7ynT;-A5I9p(nY(Vy7
zS*K3LO5dVfOIY~)kL=g?;j+C>_%@#=m3T!&jw+isO#&QK9$iK<^_qEwgol=rMV?Qy
zZw96U5;9MM7i<DiV;>)}OTT&vj<gim+mh8Zw+}5Wu)7>{R(HHGyuW0>BfrJH-x%FI
zKt9hd_d++-u)N*=bA&UKJZo{Sx%?9aXGbYqH?#<e+Vp<ZM^8vPi#l(Dl|+rb8&V@)
zD>h^mghI1D=%_DxR1%zheaGckg{LahdHDGFVyxm`GU|7GEbNEwyWfQ_*L|~`h4UR4
z6crU8pNxDnjb7KfsYpbBrC!txSVj>^G&aWHJKBB!RZZ#FCv%Fd`EM|ZkB-eYcGz&M
zED<NAL1|VUEh($`l#<7`wSrfGudm9s$Q$%0BIdg;4pmva6{UOU@yeml*CmJVW6wEP
zXW<4{QORb`@JfGMlNW5rnS~O;R!)jU=Y3oiSOC4q6o>>rKhMO^YdRMj0{nU$q{*Ge
zD8bw_*~be=jh~@XzRMc4cW&?c<Vs{>?fHyU2!na5G_r)UBsFt&PLX@l&ky@}gNC*F
zNbD9iJestdUS@f{$8dOlDuSQlTGH?vz7myYMI#uLCV^?#Up&dUvxP{iT7d<o!U>xx
z(Y5+4th-V$d%6MO{zz){4R`#)?lqaHbZdn{@?wU=yIY(nXEw|TU{t^QAHR7~9WC0G
zQUBG~k~V=p2TElnY|H3}8L=K87OUe9B`IL<^%PjiMU%YXYrF`F$eR%|pvfMxheR(-
z=#oVh#ibi`^lcBj<tP)kn~FEqRg}fr<XY1R?B~<dEIg21m}Hn0RM~|ve(~1*MsJxK
zKfOSVTa3GJJ!!|jcUUP9bc1o6LCEbtJrojh{2>P~Mhb(Kk3^-CakYwa!sRQfqto-G
zL6*;(_l{#RzD|}}tkigs0L8B8PR7u22GvNI?^YDlmF!zB^B#mM67og%cE+EAWlXiN
zMBeHwiAL$VlEUrd6+uTL-&n~9njh_Hw@*lpOg~j;8?h%`CZ0?tSNdr6h50SGjA|k&
zNY;LK$WE8`*ZgM=GEtpQgU+e3<i#k>HS${S@zZ>eOe=JAr<~DUG)Xcu=Mdcb=xCbG
z|AaupFM9ID-bG*83>0}q;H#J5f$ixjJ?=I*Gkm@3Dt(TPltyhYfjE?pPA6$Auy__p
z1)`It9904F5lDB{Eq^Zb4rPbqpo;^HsWTqLnb?eZoS?id)8pp^@mJCPg+_(n_PW<>
zlJLrKJ8VVtqWEHV2R4)->{G`twNXUnzk2yXpzI&5DF=x4P%o4On$~=8_LfHGb#2|!
zwi_vkW0t%9qE63)4)T%NV}7`74t1U0%`kqr{QzfI1pf9iu6*o0bh8_6(Iv5v^$msB
zK#gw5--<4lbVk_wuY|y9C!0<J^sX~({b~K^F|7NZ@MW#~jZfrdB<`V*^_3_N*sQ<4
z>a7Qyk@WDhR2%_E5~Y_L0Nm@?N$v7jRMpzi(rZ1%IpwtI>g_yYS0|lDv->a0a`_~f
z^=*9$I`#heo(#mdAB|j`okQ<y+S^Hqt3IdTy0Wkysdn^J!5%T~l)?tMnsaAR7s7<7
z+WZd2XBM4MV#L+h54XpeIJ=iEz27>!&^(vcu@gswQDKx!J^~<bsY?E*YD_qCFS;Iu
zV(|*vELsKDgEo!(?OQ&XTQ7}Uxn)`nw`tV@uY|tm|KYayozYpG>xh>62H2Y*d!4<6
z*8i)MF%lPOMk>uJ_UP=rcxPCu%5<gh0)dkRSwUbX9FJNjr^qG;K5yUggAw}nJw$;$
z`hgO&w)Vb$D~!aU=knvvZQ^k?VlR$7{hGJ_?&LVHCM@Of`&Ets{6{6HX>)U!vke8-
zM6!ie|0UqUe#Awdb)&0z0?q{<9v-}9j??P4X+^!;l(fS;(UdI{js8F7?QuQUURv&5
zujaqKHASsnc>GhsRDby|RwirRAEiOt`b#x&ApO;EEPe)x%2m5#RJ~y%k=`cyFH1%o
zZ${@PR+Jg*5U<|Nw=S>Z1r(YQnYk9%`>%wbdJrPXD-$;TWf3;5!9V_m#(%_6%|62S
z7qqan49i?$vKU;{UNaUz4{`&^B_EjVVMLr7EX^K;i1@X_qmMfw^7JD9LFZ;ZL3>hY
zl5E7j*bO>-!E5jMu!J&gfsn_t*}`HM--edwNgIw=QBB!AEy&|Lg?c~X!Q8{|lyfc5
zE}2XGpJVN9<gH1pDk44{F*rAB$B${ggHi?GT)1dhs-hgW{4o-`pOJWidFOFQKZa4M
zUeMp)(q5V24*UCP?X5@HIvFXh)*1ot?j99bA91sGT(bk~C5&q-wt`YJp4ZTxRJ0lA
zR&Il(8eJAPxDH$xqZ9fFP$=3*eKgy=A0bP}0-QMcL{gF;lM<f`8mZJ_mS)rg7n_$4
zfiE?Ax(j|xbe$v>H1-pYJ8V-_&jCe()Z~*@&FEX%ypBrB)1dXPELAMlmDf*og4hu;
z;EXlSnfY?+F53jJ*%oVJ?r;XAMzb#j{~>UZGY2=HfBJ(&x0Guyjzi#ZuEfiP#WRTL
z=^?#el~<+Zwm^VDWgQ94YQjv=VIxpUR$ZnB$zneS%36z#xjTMi@|@ST9?-Q_(Y3<k
zcaVT<wQeZ7UNm0QM4Ol#&6qvIZXk~kPBN5UBcTS-1bu%)!N;<-oY~xCDj|Ct)}`(>
zP=o7)$iw+RnfajPMqMmt(H1m@%FcxRQv5QPPx&<~d#*}%T|<H7gQH51(HLD)UVT8#
zfm~z>Tug}q+1)8(0aDr{bve{E?I3byCz{D^e<H)qU(x4kh^k!9x7k9Fp~bsp%w*$w
zbttRY>mYIY(<WdL-g-FXX$g2w&F{k|^?`U_ZUuZ_a`(M&-#V1*zD5S9Q_eqs%6}+x
zj~cOY>&hdF(V0BbH_UMLa!VXAVX==JNgZ+P-Fj{B9(;U+eShkH>sHpl-lX^VdQTq-
z&dkI<-2fmxhiG`Z8Ugr$`qQD`!hbv=A7l`|oMgFQaSu#{X@Bp-`E-Yl8q%vfG?Ct(
zk(mW_e1y6lT#dH%wq7*V9YH>fV;DG5e3A*k2R=lkB0%<k*I?+yW^$?PPQm}m(?5~(
zKW>R?l1)Ghw~M8q@Z1+R?CwK~wF3Rf!_Ag%&i{oliv;$|ztFz4{sJl~kIBwm1zou;
zCeWXlA-6GX7`E)i7@SMBC6EPs(3gl?$(BQ8dPPX5Zex3wD8Ym0NYO7Oy$}P9TK&qv
zq>%=x^638k-R^y6NfqFEC{SQB>^<h1W)3?y@>M|K8DK-?1hrH<Qo7w-)gw1qV*4oN
zd-y@_-Bj+~&>!lVE}wz|ykfs>V_TeM_+Tm7(I%bO)n^D!4cBM%7koZyWcB))U1yn@
zN5QZv{?|%88ENYB_&S^uq*K|?*;Hmj?ul7r@cTC2jKC7Jt!lu7IMOh;WO$ZSmAL?0
z=HHVnhPcZAhm5uQM6>Q-?-foiVFq8q4e-%dQzBs+?FI!7v`2XNEe{obU(v%gV_}kM
zL0C~qN)vj9^P8Xr$mcTJs{{?)W2!s?#Yhce*iYiUKRd!l<w}>!Tf=r+w`3lEshFq}
z2IY9g{&i_J+i;s-R9X36Dwv41N>$T@0<FjF=ay4byxF;cDGX-Z04?6oWjnP5{0@uL
zD=X{C%*qSQJo!2$bWh}+mESQW09ij%lMcq}hOlu+sl`Sb>o*&0A0NZjpRjmq_Qz<j
zyA2ko9}j^({VDNGlsIn%X!_;l$YnMP1{%L`2}>j*1=c_|;%e18tW_<7B1MK(PWGRB
zg09e3Bgnmo*B)_f30J+xMh4hul}KuNyUqM+e@z;rorc_hPXkZT?C-I8>wvdgTi)Xz
z4}c_dM%$TH=Frcu+pKNopGiOqZf*tUia*01u`7xQM4g6Mn4l1tSC4blFJ&f(AN9kE
zopO$T72da<rkGkukn-T6zf#B{R_)GZAUO8)6r<e`gI~^2{+qQdc71|MB*aXLk3g=+
za=-$P8IrS#mz3~0E!mC%`fFvihLcI<%t1VWDI1Y9B<+*flwZl%R@9@H@aa)z&$#_9
zV)sk*VCrV!p5t&%>d%-<U0)eW5lwktQ$N<5b*dgJ#7z4w9-aFk54nTix*j(c*eq-X
zbBI2kmvwFWE=_H57fh|}dWC&Ok!OfoqWA6jzcfEhz=zP;aHVb!i{%&qm{B4%E>{P4
zfr2yF)4gWCZeJP02{tb=EjF%D!soCK$f4h9)R6D<)_dxe?>()cK4*z<zn!2(0QbR>
z@V(CERyQk&XG$?ZX!Hv)dTunbY}8lQ_`2j?k;1iSpp<3CsW9|-!lskv;#D)1xTjV$
zhYzI2fB&ZRzq)_>1q}O1UuU}e;D;vdkQm~R{s_{YMLICXT3Ffp{bfG?ulrK~rW)Vw
zt>(<K_nBq$)x@04=G~NCz=cjXbnfe;fytUoMu%287pc!J_NFgX8t4Q78M3{O+dI7q
zP21I`s5|n=GQ-5QXr9dnVSssOR8+;Fp9LL6-P~8J?i`F_-5kJ!ILxu_A_s*NJlR6#
zTfdmvq|=uCysB#9b%ZR3!){mz4oR}yj7KM6CHfl>gg01?G9BGb=mBcwCTnvZQdfbG
z<d}?BYsqD0%DRYCpj+Dv($%&l%8312^ZEvKv44^hQ+j$;i+HY5X~|R(y&i#W)>iFl
zBW7YMSX7{!u3b4>>#^hk88D(grNhJPZ6sl!WlZ43?)gdHxsK^MtAOzNlwVSJV`ecE
zM7>i*7w^lD7`8lN;{T$D_Ih}j6x1S_Xh9gORf(ErUIBpu9TvVodDQfm@1z%dE}e_V
zy+`2(BPQ>E*XUYMzO%B(qM77|;Z#kvjC!#SEr1fb&6M^%tB?*md>^0Uzhq9faT-iu
zbBliTF{F_hSaxjtN?;R2(8=rW>@`-RU1M6G*@t`d{SCcuw0zYh)I@^aBOcW$In?if
z(k$dRLsEvY+WIUkhI6?2pTK!|4;12o1d`5aR>b6{iniE7=lfV4Dcou9U#oxkyZ0=(
zOmtjoyMN4%UY=n8Cjs3y>uKxS9TnG;TWcknpdc2L9w_ga5+sH8Mpehh$tq680!Ro!
zf8U_SEnq$5w6~~WA;5u=_@fci-S=f<qDazMrdD&jJBIE2{DM6z8iALEbrNAy_NQ1@
zq6NM+*hD^376Yp@RXO?rli8r8sLZM?vZ|E4<@k#JV!WzZ*oIJFZ)V;$1NTFr49ez_
zNx+`arhCO9GRq{3w!T4KTU&o+GK?i65=No3?i`F<TGS}llw9=evdzTJ3*hPE14v_4
zXEloJA2FDF5enn)%rSY}d5=u|C9h+tLNR!^KC=-NqU6*rCelz!nM)<45D^pAwAafy
zWm)+5B+{bdpTxGeQ#ueSdS48ruWyXt(?`{EO}s=d9E9ZbfTT2&)z4(d1O79ybWEA?
z3YMjMa&FZhe9nwY$>%<%3Jl+k6e*1o<2j+B-|e_aAQ(hKx~S~aM450=xDKkKjp*@~
z3|B`2WjcH9k)va&SyK<(ZY7C2`}iZx5xen$&*d6n`J`CjA%W*{DAIDmOQT=BLqX9u
zpWxmi*-TE$#rJDI&?w<=@$XA%s2Q+WXX%?b<?SJNmTAs-!ai$aL0XkQ$Q{2jHg{cG
zmtl|94^`mC)d1YHH#EQ7UlP^R5rvSiF{W6YVb2^6#4*Eb)F-=1ZnacQpFJJ7fc)Ky
z=Q01-=4EtT_Y?coM!S-Mr|(hwRskKm|JzYl|C(d<uV9;nnilPE#7dxs9=)-H0lv9R
zAiutQeokbsT4&d9*EDwZx=DI-Ji``8DpF8}10x73Gu5N9!?`c%a3P(JB1N)k+W$y?
z^P!;)M{MKWcZ&Wrq0UeO($Kto=e!Yl&_~MNI)b9hl)TQ%zx(eTh@QV8cK?&>?q7Jp
zV6$m@ARCMa`tv7N0C@hsHO&ZNrQ^&h9k`xcq_21R{T57tw}8GqiJh+;v5_eVrIO6B
zdT%Yg1-}T|Vf{*Ar6%To%)LNhY858jrx{}(cazOB*S4j+$(;sB2uKZ@4VmU$E&KcX
zNlbF4OdF|&u$7e^7FkDZ<lm*UY~S0+2d4*i#T=6w1`$dPY2JyhD6+-Fa$(ofB^|eD
z3$b))nB%ENN>8~Ke`dIQjo_QQhE(O-zg~r*nZ!TVC$yuA-DP3Z<A~u+4|h-IQVR79
z-Tk>S6r1GYT}nr#O_lI1x)B6cSsa;{62gh&0qo>w#=q>g-YtpiCgmEd8fm83>maFh
zbUH`2>dJzeaXKu?OaMH%D5A0=y2^G&R^^b7fy_RuN!&XwD;9)Ff9W%p&=vBB3K8JA
zu{4DO-~s!&*}<d)VfHZ^a4ExKi-XKP{-px~`U-eJVZf*Q$wbvjLhQSpT#$Hsa!p$v
zzhAIHhVd!vs8^ce&YWG&hkq<zKCXm&v%VP_AM^fW?Or&)D8&+efzNbr**^YHwac|-
z6@$PDB&``uhxuEDF{pXjLZMLIO#`yKm0}i5rb*>e&B*`)bf~#=!3Q>z3k!VE1+eGU
zhUv!qy5A+A$K8E@6cZWLB|NlIi2b*nm5=xp6^JU5K_!PsL2sv{u*zpNag)po9lPud
z`)mO`0NgfSY_|J;Isz&=H+xLSsXNzw1la@}*yUsQl}6yLv$zmhR8|VuK@|J-Cb?1E
zi2t}#z<fgje^=edvZnA=#Lm&3skG7G$T<xyC}Z2)Yv}FgTh<|k#+akjP41mcHzoR%
z#P_m#2xAfGpNpktoY;jq^O>k8jSph#+m;@sxQ#K=bnUiou(6UTdc8S~PB@Q-f8AJU
zDFVlI!zWs#G7Z6(q{9_*c>|-ELA=E6pcFbqQ<>RlPPwqNji}=9!F;P!SovU1T<63G
zJ-F$FIa0lvD*KZ({-0M7m3rW!SryLN-iy{mhP>aMq=wbw??76V;T-3FbI;0BizcMv
zZ!x_TBP2uR4Ig1jDcdSLt0t^E<X0@A86g>~OWjQP#t{*arq8NicxgjNdZ|YWC$n$z
zhvIzV4MxQYHT+XNJ>3?2koepZb#JBJ`-o{#)EE6}k&Xy_l^kCa;5HmX_$<%iW$)Pg
z>OA>qK9S!{>U7`T79-0EJqVieyHn#`t4pm0iYTe?4@2*PD_*0nQ=NFPE1XBPN4sr)
z4RKMYDL`ay&7FueJI8%=O~gBlip}?BFMB)tE%6;Cjznmiwd>CEB|QMT<@`K}QUzQL
zo68>1df{>0|4x%==%Q4{<c6T@5v_H1#*8A~tIYSO=T*J8T6RuKORF?2&R~3>aYw<$
z#r1tz(0Uxr_WuuPC;^$TQ?-t_r)K;ZDc}tyh7bf8U0oxYj?%YY_yewe0<v!bqnhsK
z&u3!bcJ+6Rbt|T1;jY{sD#~-ESH4LqhhnPjaVjL(T#C?FBv^EGfKrl)q`g$9&!Sze
z#c4lbrT$g9UA698Q9?=~hIQpXD?{+8IBJdVWNQoW(FD&H*_5mGR+2Z<Al9UNUQx8O
zTK)`ztzN|$9C-*^<m70;*{~&WEw{(8sK*DI*?FmXGSd<JYz)*P#rj3p^YBzO`AoD)
z1K~e~hIDC%Wv<!j;R7x_x3t2AJ~uvFhpf+S`Di+v8{I3*f~l!MeKS$`81?AR51jGz
zq9!ru!7NkAkrcS<>h|^y&qoaZpc{>}6=7$;4Ebogvs2A5UoJnNxrKPf9+>n1i|qw~
z)&gcN=LAftB}BRCW2jp?7_g-uM}wH;c8dayA@U7HOi1Tm@=NEy+i}dKD?V>h2T_2B
zd68<lDslUvrH62HJ;%v>&R-)ku*9kIZwgKsqNaNFm}}brzB;m-7G_31N;oW1y!i~L
z;kNtEFkCk^>*auoo~we0)RXNc&$|%TB8eC-LQHQHaVh$yM>17#Tki_o)(3){^$(b;
zQmJMzWz8H3mcp#E>L%a3Wv7D>SQfD}G84`C_!WTZlHq@+cDPr!P4{MwHB3aqoC^h|
zG$1Tl?+!<ZVc^A*bfdchw!MMck}c5jYqICz{6)s>v$KX?AADQ0gCNRcz97T%qfr;g
zK~rxO0-%b<-J5^WRn{{f>y91&6wT<4Sj|#VbHN=DEhj5;)#xX35`$==qKRn8&mR_t
z@<5ky@D(W~EUcN*8ME8@F2XLa4YE`w%&dPD?(tG!CDZbJa|~IW${A-pMq-?Mqu?<O
zm$RLGH0u)S!Bu_SC<X>|O~j(KbcunA`F@GCQ!Y1k(Nl@Zch+d>Wt<9ZLQ1t;oMh#p
zsD|7-g$;A{Or%-fyPc|=n1FCZK%cP+M~JlfJ)J+oyGEsf*5D&Y1+7rPV;>1=X^qcI
zJ`eBmnq_e4MrZ0G+I8;dgPDk~LFUICZx0{Zw|`#^Y>iI~!#h^s<9G-(_LJH<zFuw?
z@e5fxzq3vL#MpFPT|#|R2nufb`OKU;U?qu=%*iXO=RC$tPXJ$G-AR4*rBqf1gcxCY
z@ySB9uAak0L@xCN-RS3A_}Bs4z29iNh<6n0d(YMNUAyCs^Lm>*<`rqK5>}@DBWl14
zDe$0J5HSvTkDr68*=a!^b|}H~%CXYw3z1Y~h%lQkGjR`hXIlsV_JdE%2E7DL)nk<z
zcGR+;z8F2PW;<VPwo^q!gpc0Z&7ihjewL=Y#1lurP>#FiXrc_va-aGY!*{~b#iRe+
zEv5qhh06O64E9ymGf6i@&**S425>?K{B@n$*m*q4K6-dSgohW<H=M+eLtY(5;TQJD
zP>AKYS)Wsv>w2wpgzQFnayQYq3=&>OXWU-b2qO=yEr&_;vtLs2L^b$!86bp_nH}!!
zNfVKh4mD-M{2F^SFR)g`(bxxp;}7t4(%4AiR@SyCr=;`L4ZVOMAfK=>(DD0!QgvsC
zd=G+W8oiGo%_g5#lz>?rI#B6|Kku%Pg+oP7tSe$Io6+)RZw3K#fB;;GA1c-$y~sU~
zTv2OH;jGq#d0Wt!yggSbc32F|IESz54}&eSou<Uj{WZf>)7q?4rvMT@rM)~e!yIt=
z&>J5-8kW$oix%g!Rxg!uoO$aUCN%z))nw%OJ<cbn!Kx{S8_yhpUt*y>&mLe@umI@n
z;xdL&_P%E@CszeD+FKK?qEJU`z8rh93GXwx{7hSP6M7IY{eDK2;7}Yb?-5@a0}d2+
zl0UgTM1qUF^^ZcDoMBC({ts*()|+n<Sd9N%Wx5+(^^|1DT7E5lG$mB8FW~o+(Kawn
zYb0anQEiVQlVtjq=V-VU5@EHWTp)Fj1Alxjm2m!gy}BDQNK^Rtw+n%;m0g$`m+)9>
zGPTm7+^3I~)V+|Kx`8(Oe8|_mS~xYC8RMtLI8jUd{Q3AeX?Y8d#d28px$DKXt8YeC
zt*%Ghi6bzks;f<ex?6dg8lQE`^F^K0sP{v1i3%xv!%r|az0qs?)&Dg8hbOnh7lgb`
z9jW@oo+tYLW93<PVrs#xNAlPscCn?U{zNr^xZ1zj6+4`(!Zf$B4S6f%8pZQXHUXZQ
z*tx%vz6uy&q`@sT6&Q>yi$@m7iT=p6V#+Zf+lFhX3iHcI%&TgKiQt1lxRu8cm(uBJ
zA*$`J?|Zw<^6GX+(K@o^gd*QX9~^S-sClA#?;Hw;)d%S$#D^qTRn}@K=ruNE($Y%>
zQx%G<EO60KI2?Edf$FTr^S@hzITM;Z3lKkw57h>s>d2=dz<@dN`7sz^*fg7Pwz!vp
z3~E{`C6nW#59_fpSY|DG8F2xQf|;I|6}IG+aCdp*eTOi*xP4v0mBu@e9=U_Z*^;r~
zv_I8$V52^6Vz=91-o77`VN+S*I&cv&NiSON>*adTL1gn0@+vZU25L1%6oH65(@YX?
zR~pG_8`=!*P-c3V?@3hlf$1KMsx)ufbR$ZsTB;P;o)jYircC1wkhE$^q$Mh#G+bZV
zfwC4w+l5}~van3@wK)&lG;Tj0U0pvwC}%lE8MRq7Um&FKh~3b&?jbKu3!O&wGp;xi
z!TrHX!ee!LficL}=VySIv9@&(!?U@cSnUTMFp*r-)-2~LcbEJQG%%i@Ie;+PfG@M(
zBWmm4TR}$k@E1bRfmgCn3{!z#ROKpKXSQKs=)gmD_%e#Z?)Jxtg_p?Z`zt6UE+DAd
z?}D)7@i6XTE#O(p>t>$~n3exRD@jkIz-<jk#sNM;eLT19lIg4hfulV)TZcv<^gNEl
z4&>2FnfHqV5WSGnXyeYuKQ(Oa{-hY_Dk+;ET1YK7J%M_aHcQHH$VCx#`N(V#8kVeW
zWZe(Jtn1VA>YG1G0cj(4Qnmt<tSr2Id^BS)r$&{U6-&0jnL#Jn5$&(-OXf0r%#m3o
zpyQjJuS`ASSK62d%3Dl}zjCdKtW`%})z);X`+FVcR#Di@O*=DOU0fI>@90uAD^T@J
z_n(%JMXvNZ){NQ>|Jh$eg=MGQ9S>gf=2aT6jUHBeYTLH9b?m>rK8uBNM^eq~Tg&Ps
zk9&CF$I;LgP-`v$TvO46>i*e}5b`e!X!EyTJXafVxoiAOfk-o~I4UQ0nfcbbW$pt2
zPWI&`<G@!^jpv_@RV5iZp<F}qBEZa1$JRF>C`tr#s_g$6#9FDz`FAnfk2ji!mSumW
z&8#`|ur%Aiwa_LR@;$*E=8F5no`qKbG&dp=qW#5|QBE@=DiUYlI9Kp3?Cr{D#jQ8;
zyVKN$O6ASSt8^%yy_2=<Fm!n}2KJkfZ1aW=0t{=G<s`TBWqjPn2|^@gc;WC$=`%>?
zU?YhmQ)F-+6WvV);2Ys&ZoSI^_{X-Zw-`pA(X?U56w=1mqAe<_bYrp{Y3ItxrutLn
zXH2Lhf;3YD-&>fFE07ATKgiqhMJdF9z5wt(s}P!St&C4+ZN{!N$zpZO<v&cn=wRrn
z4L>8NOU`&IQuI`q39{lKAEvnCL{S5p$Az-ITVh~2)PQ2Ymg7f89{-^fL{}e8;ZNl)
zb<AI)CuRxiptL4Pst3;}nb8r~Y|d(rt?OtK&kIjKezzxHbWEqW57b*jHBq+u?;FVk
zFF&stKMBXEtJL83qHEP*%KR8#t`4NP=W8|5<@N4PcnYfJc`UNa=LpBzGwfR%$vIB9
zNsJ^ZOV~wt6=p_BJ`#>Jd3tG6HxsM&ued=CYYGfxGGEltpcc4lb!0KbD1`K%`DHS9
zD*Gs>itPzj%Zh9Xf;>})2Q;LNe;2GQ`|ss(=_Kbzo4xTL9os}z3Zy+b(lL5GZj)af
zZc%OFCrr8wntlcq1eOO$_y8g~EAc!G1P5i-+qHf)=x1yO#la8hZokojwei1r5vz2N
zZE_ta&1?$qP*07lU>Bc}=If$n=AaS>#xrE779gos&x}s<t+>-+U$2T{IJJ?J68fKL
ziHy-2`jCjp)evMJCh5qA_JoA~)Yysc)my<<jHGuo=v2V7?Z|}AiVi~hFjwo@ww|J@
zp~&4Ya$;gASxSCS*po_A)+wQFcL4WCsOt$gpqKS+G{vedxN4??<tPI)^?ckZ6BBc&
z6EV>`f;(BmeE`5V(-kPB9Enu~92BPfxrpW!<lQYddgbSCs2lU%GFZbRa6#_>p^Fqq
z7Z^574HlU$mJ43iM3k+tM<680Qhm}*m`ia>KRwBhW<sM*#^4p6S>B!&g&d}f95(wy
zn^U8UGcpaGy!}VUSi8peq~p9U?+pBJ7T18q*DY~uY}n8hq;)FbE8zY?jRGFXzyqi!
zrmsa#v#&KWNYPO3iyJJprm*mQEi56NRp+}<Q<eC;1&`Pl9S&@euG8}!^7+L^_&@~O
z=2}~>wGAK{(O4KiLe?Auety&~T<!}yZQs}2fGx`Pc8^dMfaOw@+SiG-*Gp^b(TkJi
zN`oOzI8&g2oyEDxkf)H9=VYy(1T^CXXGHFEKPrb(GYF-dTb<paI`B}AscMHWp=Q;T
z<}RJU>Hs)5p)uvFj9#Qmt7b@`_Snz-odW<Etznz}o}ZaYwEIIIW{xV$b}0Ydmm2lI
zz*A0=-j$Tg;&eZLIy(PBJX=(!!z^u(F;9u4%T*9z$1qz#<keSq)Xav~nQKk3s)=)~
zpwb-SIH!xAkvQC>sb1y<_>uuAcBGRnNL4|l2o2+AT^UP9;-^|&zr=f6OXW4jt3^B9
zT4O;(?;q`Rcq*P{1oqTIzQmEF-@nDt372E{5@|6~N;o~%?U9alXa*=QSL;cl+L*uV
z&0xL=MO>BOrc@}A)(QtmY*qpJuSK3n<8mBmf<S2=KfizH7BBjm`YVl8c_2vUitxDE
z&L1q1_oo5p_tD49J24%)sOVK4yCsz+Vp(bo^F^+TYQ-7uc%|ahJc(j_mY<8cSfwD*
zz`h(dy#I!`Vq7KEnrF>rRogfRR=#HmfWzyGNpeK3Ynt+H1;Ctyi&83sdO?_PDJAS7
z1zSt?lRB$SnAon2vPW&n5UnLH1%o-x%H>sUxivR%_ZjgjERww$<qyt$D&r)o0-4A#
z#INlci4VW-KJI-6{h*r<Qk{ocY%07Vwq2`rifx3){ltRMIEwlqn&|U322813134SL
zq^3(lz%cM@{WD|bSXy0G!ZNA6krv^y-o<yYAOOF<LFjCoTrV&>5+f#^Jy#Tg9<Dg3
zl~(9XNj7Ype1w(n9Q|qjU)gC_d10dPS38a_pUb*gKp%0OLA<)wdH9_^Sno*CfcYzY
zk3cQ?sL6h6@;_{)_-JirV%$V3FX~#C#SuY=%IhPMb4UV`zyfDR*Ynqv%ix)}=msVk
zHeNNPAJv%F<W}U~hDWAo$n9vQc;qaH-@Fd{%a6qWmi`)_6YcQKkf!CcP>(15J1NIn
zCoGcR^&YHxRx*#+LWPmpDdruxHAf{18+S`gCq13qw<hqiTkRh!z)G}}i1QypaM{||
zLKFk1csj8%XZtXk_%bYW)sB(UzzXG(C^f_N32*2Po&eweo8G<mZ6mJNJ!CCs=Ppz4
z<>59A%OaQ~`KlXY4|d+3tM*T*Iw>hnURcPevZnJS@bK`YA49M9J@?Kqa`pJx<Kiuu
z72^dNNNM%ni&yf$Q3INWs4H99-~#NulVA-E`@^~WMj!5yLW&zT?i8Pv!-t9aRo2N_
z3YiZ&!&L$u&XviOf7R+=jD(9q-?h*}xH8@MTgkUmr*QT4#~;Tnh5LZkqn0;xg6DC;
z=y9#%Hb48i7s`6QUKu(E1-Pur&9=#BayVmh@+R5>af|ehZ^yL0m(XmWRJ9j%C-C0&
zE3opziWl+ol=zEg{}*4pPiN1b*Gy)*{jjh;&-CUx$~c7`(XJ0WqV8@@rd|4VhCC(1
zlz;k}rrT_Ig@q6I9}tn0vZxD61x<y;pgm*q^hmcT<N`4G&zTqTZe~-y!bi^{FTx^E
zGwIrA-}uolzozw@$rlR%P5L1wJ#>u+GO8!RDW0+=uZaT5ee3j4ra9oy<ZPjZNWN?<
zHFC8#V<_+x#H9ax%#786EAsuy1ahPyzdMb-b8$5lHl_NKlV1kR3b@=iJ_t88V!=>e
z6>sGa?v|c00R~77F7u7}B0rkyFtYVRcZ~ttxq(j>YWM$$gwgF=T5_6qtq;YR5cD$K
z3qYf8{4`TvBt9+xeKVljAC<wW(EoQ?hDrUJ#1b~<WHq$^15JWQMGd6D0Jq>a{fKXF
zCAzJu<diEV1?UN??(?2;+xNB56C4KB+#bWjwe`YTCT1%jSgiqX_OwU*E<=-=;}Y(>
zV_WU@8kDF2kq6ttEerht#?|J!W~EW-@)Um+tf#;T{7qsQ>SR#^WbXgQmzr{GlbfIg
z5wtTHq(^O$v5XB!dHgIx@fP9&+JwwD#@v;f)z4Ttsn4hXqv)K%<7%TQ+%#-#+g4*X
zw(T~y&BnHE+fEuMjcqrX*yex!+j#~zXU=@H_ge4ElVyhNqIoK*OqX3V_!CuZQ#HOd
zJ>C>&I1G0F85_%@6_yutt%s~)N+pHnLcdj1mKfV&W0A?BOtDvj4k<62hh{U%Dux-^
zv0)BdO>G<}1Ci34)<U9$izb0AG#3N(X54Z4Qd>;Ud1x=oj2z|!7YSYXgH4?%_rT2H
zO_bxzC<+Y+r;LS5*qs^lx=XSR@-iUj$?ch^fVD1xeh>Hl&xA&m5_IvxY+}n6r>5+I
z4rb#RD)FZQcN8mAEph#SX>4H46c23TQU0w#dHv<<nvviBX3S{vIyWrTJ%?pO&}BAH
zUZ#aDrJJweLQ;g44U?A-&LMnpLCV5Odbv7SlW63fh3c&FcpAv9^Rka*%dEE(FN+s0
zg_QRQKq2K4pT~oQHYBL;p8x8J##8*#vV8dKUI6}@CHwp+YrslO65dp`&gm9L;DJ=(
zB;vtXz8p>Ya&{>YGpt!9qlk^|TpM#A-L#Xe=c;)6rajjy+R*Y!o6^1j%wnD`9+~!H
z)*~koTV%aOm~=5Bu#gPXj0*pZsLqxFOM@Y}Q56eFY@O=qzb;TcO@MrJF1;^ZQtl(v
zT7PE!C2eYT`=O$fgP-fTe-zeI2^2Y(v?^MPsU`5YIwn7i2TS!|As)qJfo;jEap-0?
zj@#wd;nmgekys~TWV|Swp*?7Vb@RFTE^;MwC;%uF^vCr2+-EaKUf@X!bqQyNBy=v0
ze2*cuguBmDK1SiWA4_oLvFDLs>6Bfq+3@pcY=!pi^R!>i=Uwt=lsr6y!g4bpT6Jjp
z<V!xjlKLC?iUiC+4vMiT9n4mkDp@?K3mi#=!FqCC`j-mk3Xv)dQa*1)@Nd2DE&k;u
z3~xk+Rd|fJa}2raqn9`==azgsz0bP61;TDjMPiBM@qvi~i+hoTTEo^JQ)A!852fG6
zvz&}A`ec}VqPjQv=h#NHJ$Sm|&L&(vc7s#7_&UULQhz514`@g55xu4FVW5f5dv&xN
zoZYdARqT_};(fWSP4p2ALwcmw$$oS=_Kw}MaraQDH1)%5p#t_z_Z<mgRFfs|^5{y%
zG#09ZmJDo3E$3ww8TTEea)a1QUXgAHkwaNPhqDGCeP)`9>|B19TomRQlWnk9H@RtG
z=#g#(>HRY~KfCxT8<*Cx^W&L%jT*SUiz&g}i{Q!L*`9R(d?C%!kR1lCjD$giHen-p
zLyCO#aQU>D66a0^n`L`f*Z$RKDbbLstE!N{6$TZdoozmaWC^{S3|B>5y4`s?`a9|x
zNy94CVpmX?<oDzdoK-IYp2KE~fg@6|{#z|H%GX^%xkajE%YN&1yp9*1?Cu$*#=#V0
zyxyC$O$61a_G^jLrp7ig7x*<+7I<{+jp3x@Y!&*mD={tFiSjyg$AlL8s54JXC^O{J
zRonqL&iPoa$C3(PmOZRbVe*AeBLuX-DQXK{un*zTxLpkw81`qiuyQ}2W}Vy2<E++g
z>vrsQya)5aQ9ynegWK%UT|M*)G!aS1)P<64f3>~UW_9&v7*c0iut902hMWJuSCGy9
zO_r6oN7i`N`|9meFdYi)^K&;>wRQF;ReUJH^M+)7lvB=P+^{jo{&p3~s-%EQn<^oD
z^y51-*5{KzNR(anTnA`wNUOQV=vq;aj*>Nv_f6TQ&4i~yRb4+Sf7>h{4aOA#D5Qwj
z-zY;~%UV@##!|JCoA4Gm3JqFI!(ka16kCPe)bez@yih_yKu5*}b-lEiR8Bc5{;MKR
zpfqZ@m<m*4iLnP`7(u_0TomaTdKnvKzIWR`8HVc=3B%uU427OY%6&ueloa&V%Rj1X
zS%^YY%rmV^VVvqC&^|i>HqDJM{rS07$q6Ug`B23K;e8ainU**2G@(p_p_X(P)qY_H
zyiIQ8+Ep>kh?xQNn2vDnwwv9POz(f0uI4J4;-cB2b@HRK^p3uUb8v0%^3W0TUv~jm
zIC#EB@&iCSJ2V6g4V+G|!WxIjVj6;46>=42W1p@wFRyQVADp}oBMpqTn)!UMK7aYT
z_R-aPM=AJR&vO-Xaku}V0=fX9fHM~8rs=fu0r9jg9^j(~E}W%bl!uRxDWnR<8iIhD
zA|R9`h;S60=lCBH(wl)4Eb+6GA3HaF*Xmt?S$F?jOx3O;wOX%LQ<+cC-QX3Q@@L5Q
z=TRupll2xI0Kx!(7jc!1?8IGxpj^Kkx9#U|mz!Rpv3_r8-miGIPR`Dwx6|E!0!NUf
z=Hb0}-p-;jL#M%6T*sgZ1>8p?ft3?gY7RbLeSx_Z*e|0w{CtIW2o$#?xE`eEk0Rt4
zhR%`{Y}{hWmmysmr7Fnu(RtN9^RNCp47bk|R8$IOrxB#N()BxNe@0x>H5IMnu203}
zf57@(5}0Vkfw-YxLxJuv+(!2;zLq-k6J`6KvaG9(TP8=)6kOB@FEyl#-n<KkY0?7?
zrgBqwzwVkQ(na@xw{L{-ybD2ijBa8n2kLt8rc3?Uu&~7NpOj#Qe>wbmgZ$mx&sx^3
zj7*H>OQ%nqeb{0d*bVe3a0&DjlcLAe6)SAo<6IdX`Dz4PS{*ed6_~nNxlXRWQ-Ex}
zJOgSuQVeRN{1F3P<z!J?yKk7hNJ{=Hr^QN1+6gd9U2Kt8Wb^*{6@`38F~>s2VnJ>&
zE^KNe(Og3ek<k%hfnNXfPh3g+6a9%fPgOoW+=HtdfFi++9D*c)5CTueT;PDgEt^n3
z<#jh^?GBL;j&q3nvert^Brn!?gy%Y*G{eq?J_U`t{_T9YQ>bVWi-f-er==z<Dv(P*
ztI&qF-qHc(P^^WRhjJ7evQH=Aa%PQ%kc!?qvW2v*EG37Yfr8b#sC<o1`_dtr!a1kQ
zDXCEQ3v}fboWeSHf@8!KY|~hQM|C89FY)lO4|zt)ZIi9hh)Fh@kaU()(uvN9ZCI&_
zrAQ+-4|%=@6cW;xf!l4Tf>1B;K3?*?K_z4E&UqS%&d#&TDHx%JM!9Z7AD5{`Pq6Bh
z*rO7X`<Ov$<-{)xr&A3rImP$%fo8{F<>+s!pikrE0_WP?83rwg2Qv(86pe~$=CVkq
zaWEiXy<X3vh9ORmDk2j1C7u=&WVuVy4~<$Mo+scq+q!k^+2YgcWVN!8oYiN!#p)7B
z*0o|?e#=Xu?K0FXJ6X4yW+?#cpu({rHaT**rO4%98K(8kxJR?1$GR{tOY>ii6Y<PD
zO|LQ|XKvycXVnq7sLHxxlwy{CHHgmekl!#)LS(D)P^OqO_#;6@wH3{m5c&GPnrqdA
zp`PNz8|G#?ybQ;(NDceiC%IkjL_l2f1~|f31iRm5h5P2&2*fk&%o|1uAWB(bbgzn1
z+FUzR=8aJ7cXIyWP#tFn*>%ZXh(_;UQ=iv=E5ex&hs=HYH}brJB{1Ub7FAeMANMaq
zt-D`GBI0bYH1XS_cG0$p(wDEL6_PNI`T-|yNQm5CH({wcemP5)8IO!@bykdod^^|H
zx$_cCj!*qyz`I#>-JRTeyWRT;T#v=ZXCwjn-xvi2ckj0}gO?$LpO;s0mABuIv`-cz
z_Ct{*Re!qfW6ocP)g0T~3=FNZe7=7^(q-jlJMJHX0gIfl`9zfp7Yg!&9=U~984j1V
z=KW(a*B@er*09>D#(g6k(pj1<I;ZPmdfv^|d$zZ0L<cZ^jjgTYo1JcwsFOfzH!X#|
z#i>dI2-n^l@p(QGjt^zg%cM_Gs;W(z3BGbOGBQfCvUoS?jknQrxnLFex<-s35upS?
z7GFzCRMkv-Z8k;K$R+XP6Z(}9)$ax}p3O_BO7d*)uE+TLRUDriv8_>OxgZaosYpon
z)X+`I%L8G^#1`B=vC^bI?MNp;*k~7jMcy8|v?e6`t23sulCI0`>|829G(H*|O9k;U
z7CV_tNvHj{n|~MeO7_%B`Z;s<MZRK{iUo<O7Drftr&<n@|DvVTRkeF<u?M+t<*s+j
z>=cst>DKnt2l1l!r(LZ5Js$egxQd3r#4kPI#AZ?;Dyt%r6OP+LEpcO~Bt9_5hpDc7
z#7!ku<3gafUZ-8tp23vPltRC3RrRRkAc1eDeM20NQ0<aaVgqFFyDIFTL&!>+N~x8z
zmfds3g}Xaje!Pr^qC2^Ub2YXUT9|BASTR_RZ4D^m&E+%8LIw46M@q2)6y^5;aN*y_
z?&6pjsb!<FxD#O}bF*P%9@B*>59sE0v-xBWbZ;lH^Gn8CGGB$(#|qBARyQYU#c=dn
zIm+iOBLtL+QWVIl-ip|RVR4(tQP_UWU`fp)j*n<EQQ(53;fJ>xZ?VKkAwyYQx>bf4
zvJOQ#T4Hf!8_YtDec{jxEe9_~GEOg51lzPsYCYnR08x-k57mJ0dKN~Ra~$ER&J~DB
z)YP0~XT1oAc@!kr7R<Kgab%Jy6pfGmqi@0ud4!9vU2>XFPD_tO6{MIRx>_?AtHab#
z)8+H{<JsKL!Cq5dFDEDbV`#GG22n0AeEQ#-*3!}(m%=_cyZ^9=#J_2<zek#Go&ks2
zr;BCPNYyw8WTx|a5miN_aKi^dp4~<IOpwKuhZ9X^wUTYzbkBF^KKfg^^}NzJrLa|n
z_h3X>_P2V-C+cQ>-088ZumB=p1b|>ji6=yl{bi9#fpeSH&T^Z3#z9a0weuI!tWWD0
zpuvp6(_*aKg}<Sih3_zkPV+1`p+<*1p=GcM;RLEk@A2|RA`oVe9x8&_ETio4)fQ}q
zyMa}WaQcbteFK40Qad8GBdEp_-BK?G&Dkg6P4p3lr<Bz!P>(qcM2;1qCe$DiRvQT+
z--Q?zuuB;p@sMEC7S{K{_0^@8lXd2GoQ}=~Rc)9MiFI3>V_{Uf5ygo&EB-O`Uw5;+
zEoSnG0p<2a!ljhx1YEZb@8Sx59q;<1IruCnJsh24tr2c3H6~65jJpfX3FbHX85LrH
zI?T>99w#t=jl6Ypzk3om-bIk#>YRSw%M#-UIT<|B-#bcaQ3B&h$G4|%z&Dv~@~Khq
z^?T2om0#-SJ2+-4R@;;^lLnvXC79mKh9?j(qB5#N%E~5m@^G1o^}8|ppKHAnVAQRh
za3HR|%<;33T-!=W^T3G)`(?VdFrx**2zpRe3@Qc+qo$CeLGzP-tyXmFx9peq_fNCW
zvs}m1B{`sPT>HzpKXDPG^Z&g0q{nWt&Kc3!oH~Y$&i71tZlD(!Y4C>18Dq<mo12gU
zQ3!rOO+@LZ2ngK27#Jwa4>t*+Ifsz#cqzpSx6uQZr&!q+%`dhQ5x=OaYJMdx_w@Ss
za0cD>{NB>#hjiZ_+Mag@LHYFoCLluot`SEEF(u>pZHgs48y%I!xdg&!@jY6k<p>KS
z_$6+j+y^X!v2S4kzFOOqQZ?;Kg(+)Bp4uNH7TkcwDh;@fmi{%=Y6Kgqx4I!^2x%X5
zEj6_&V83%8aqtEkY{WB_j(;I;q{Xs(SKr<1bGK~XnRcpH_l~_T{;x&wvi7*q(XjQI
z*>6##>&wt11B6)0w`B|67@XMBBonEWgZ~ke2jo0Lcr4}!B!drWO$wQ1HC34@EzPXy
zNNznu;P3LX2o)n7sM}sr7BGyomYfyVLaibYC8S<kG*PZ|5sM*7K}&Zali68uN#UVj
zk@VpfYr~*iDzRi@N39gr^-ZCCoK&<1(pX7xZ_^xK&Vtb+yPG90H|k~K&VCIU)#U^T
z2i~jMWm&~l#NtRf#L~yP>*JBWI+6$`e`3X!#QrH49XNw+RzxtAT}|QNqIAlIkV=R#
zvf*~xiV>L6HUw)zIRH@U%&|rPm!1ULS2!Z(>LDoX&@x=erbg=J0+C20?mqG$>Mx2s
zXpMTLv;-az{P#v)M)3W~D&T9>sow`TJLdAP%61O)Q5l4tn;xX$wGt96X?Et#>lb*{
zu#HA<gKAF8GTAEua%!VW#SrSYtGQWlaYhmA)Sk~b>6RG1tBLNyhL|a6>n}3&N5Z+|
zq-927`y%&aPACIXNu&y>tceK}IE`ab*~c}++vd6;O~%8fQ^<V1?<_g*6ZTFSw;Nkw
zSdV7-QLM0q%v-=XJQ04mUJ&IpNN{DHojbBQrpC$mAdH+S-niJ>_a!Dzc)0aM0t=oI
zVG5Qh_KUd0HaE)F9edPU2KKkkC;!JAKJgbQaZ6$lIAJJ44k^<fevQ4-eO1!aA&$+z
zcdqOHS0JCZzarL4T8K_FQ>BjI`;OLuE^E1>#MI|5iu*Nvj*EfJT`~@va;v&U9cj`w
zoJu(BOqHJ5`Nz&8!*_ifc%Eq~L{vom<|Z}3&vhgpGnQ=hX#RHu&@2gjjHTjWgmmaA
zCZ(@<LWIKRY!rnFIja^w|G-0lxTPO(ab+H`&VVv2PQ6GTHpb};4V_RRbh;gu)9-N#
zYQDNUwX?qrtxZe~x!rjV^m{|{IG)n$`JA*<Qgg!808VBAf1KLMBK&_@+s^XmMQylU
zG6LITgr(N?)fRu!(ngy{BE5{38lArT2P5UYY$;mLi<#fBTkoySu!)c%Dlj)xGk4%<
zi4Wm;av22Yi2T`0jaw}h^|ZH|v%}wH^m(JMVQ^@z+ur_=L3Gz?m|^h)()V7&FPO=D
z_jaT6k<dg-xf+1??_Olnj2{HhkGBE2h0E?+%*uP9xq}0kdjI!u-i+5??Gx2XsnCqN
zmKg1NaECj#bG7euFMnn&3q0c%9usC?Ym#+Z5n=IS>HOf2X!5ak*E2PDdbqr_DDt?~
zH)*&yS1Zc%c;Oi>`qw?;>Q8$EK_uuqIyT0#<wl(?wT@A@q_oVi1^w`dNNOqOosHnQ
zxL(s_Gw#`K*qlIG*VsRlZlK3P7|kLs<LIAK)?K1kpJi~znZ3Vnbg|jX<O)`Q6eny)
z(?VAX2~fnM3&8~*j}`KfTthsRBK8m>N){T?WH|n=OhOfC?Ai8we&n`)ybAKK?hX)k
z^76KYui(?5Yvssvh;ke9q_YyrQF{$Cj_s{P%gSBa*WkIc>#a+dy}Z6X?9p=m))PiJ
z;@LfIuD0Ju+S<c|qT<d7#GONY?CIq|4)NX@408#2M6@EMQalDjI1Q~*>K{o286Y9P
zVzbC?Zf;Vrgo~CAO(+#_e|lSQsblw0wdtYDLCCaf|I~n!T{isM;2-fbalx^kEo|tN
zS$<DKQF=|;(rLRIgRsDcqD2ZJbDkM}kdz*ooPAs-f|P2S0G4Ho9c(U-#zp!RqozV?
zE(jnsUR~WpiD94+6dz@o?P?EKo76Vq$9xq#QS2FVv(hal8E5v8*N(aB^B;qePKuF>
zYhx4+OUEbSSR-}VUYm@;M=`=?h}YJ(a$x#n6bDmJ&vHahGO2Q-G<IQ~s-zTGT8^$L
zCA_{+rB27l$n-qb+B3fR2D{BpZD+BJ!KF|FUyP!12iW<hW_cnhW`ci+{4D>v=}Lk{
zdM_ofap865dPL34>M%7=4V%b}2mDVl5<f^gHWyDxZju}ERC99h7<2THEun9ez1{c-
zS`R*)vYv4A`hIZ{JErzWC%nySO+KM>5Bo&RT-YJLUmizfp5xjK;n`eTKZeDE5Rw70
zxH&9FxzR)4C#jo@%GX(Gu>$q`j5W`)nYLJ5znAXvUIC~uyv%|rLUB!!SZH}q_y<LQ
zrAz|)EE-qGQbAmFnua(vGBJ)7(an>2g(gK-YhIBu#cv%qzL)DvW3d26r@!l*Qowxu
zM`;r|hG<hCo<H{h^KB)9@yEi~RgGzFF*sO|gB+GlhyIRI_FiMs*-_l4hJIcTAIREg
zYRh@B_G2y8{aA%bxs^_ttfGT4f_kmSgfD{?N1@!Z9&<)qw|9Ls`C5LXEcKS#W<G)P
z+kS(w+G15j1nG48*@dISTiop}T7ZK4$1q~=TY9R;6KEUMht#9s@kr5gom_jnA=B#N
z{7Le-<r~`RX0X%F8w${UhIuIWX>3lP4_kXC?R=j(SF<l}fa8d)TD4~FQrCMT(L5VQ
z>*ZGQD3Dvp0&@7YwZ3oc`kv25Z{t9;(sk0AXaDI5;u6G`kGF&<bUu)Gb*$8WZngOe
zopD}Z+<4v76D9y&z0P>j^0_InWw0`a$iV`Y1Oh1W<bdm`cf9kmXL8-Hci?}B5m&C$
zYYR5B!6eo{kE!w6uEG7#%UL#z#&^A%nnO05@o}c^{zR(BtDb)IU9t92z1g*$jEqXo
z-hsrE0G(RjjIn4?IhsPMu`_JIQkvD}?Mpk-rpEi(<1-rw;G%30TfE28wP4X(OyN+7
z#YKN3ABiQZPlIb^y(78!W7(eawWp#TD4{~@kyYPlj@m#V&UmeBSwWlVs*G<lMMe$6
zr1Qw3n)0d$m8i(gGdgG_Bt9Kx+TCnk`lcQ@hBv#`-I^QbR7HM+^x}oA(D(vn==BO7
z@%*R+_RnZ%-p1iHrw3XA%U&Up;iF?@vvYckk$bC*O>RhdP}JyHmX&viPXmE26G<N3
z-iC^{twxLv0PT?~&^XnTS#(6(gk7B){sIruoOrNtwGqvlBlA$}E{LX_Nq|T3+p=p=
z7Q>+XFgH<R&78tcrl_DEr2DRKjbxaI%dLoM-Ym|^Ur;(05w&V8os$|h{;hD<7#S2q
zyk9AS$GNEfd3<;ZYcW|}bJ$ixCKgYL<Doy(jM3-i{F`m+k=EP3V8(T%?~$QN-vK*P
z151=Si1g)QY;S`N!6HK`g^V8ZR@<qiV-OX=(VtbOOgB`A$pS><=1R%;K#;oC_z=-7
z&)k>jHQi;It}`OfMonlE!bZM>TZZ-29$+0M?c|&g4}es>Av_SJq?u`i((~gsu}{`D
zv2mO;;*Ep5IIG7BDmd9QM928=Xm^}9NO?j5T<cKQNNQ@9fbYcWF-_$?fI=&-<7B4B
zX;dBdpKA)M&A0@JXquPD!hfKI&gf@`?nsW7uiI)@b6#E?kij-srGQRl19jTPSzTkN
zLPErxTf1&O7k|EXt;Goxh#qaTK?6I6@|zoyv@OQBR#PKHGsyJofEZeZ1lj<Wel6!~
zoV-{IG_x4Cc;RN7Ar0nfr;_<{a?!jH4|XYveL{*b-6IPOf;ROZ*~o*-3RIM^N?%*f
z$@SUeBX@aZ=3%(VeY@s-?)qe`R>+mO86J5jZ_+-kV1#}kmS#0IWaq8=xRsrMgftks
zO#F6+&|dylg^#~D1PRc^C}m6V!R8CW15AGizDV)RO9gQZnEr|s+4ZbPff}?86Liog
z<^CFXBx%4(x=l22Hzl`lVlwd(wS0kZ4{8@lsr-HoVM#ws9l<)lBau_$EGJWlchf{n
z_x;4vKUS>7>1*NPd=R~y7GrABnj*VZ(p=QA5s8r1q%p_t^*%Iqa$+iP*TWt{&qmAJ
z*dx)LUJaLzeZRZvPsy?jip#EdGSf!>7~74XW-TI%(&L5a_@4)rep9vYC)>Lq-(9Td
zCC;80n;PI)*!8Xs&~W%3_hhK6smBw>D`j&0J|I7~h(1U@2R^U1g<$Xr&kBDxHo|qt
z!}GjyWJefJpZ7Cb1H%sKhVs*v`1sS40W$&xEMB4#J-RN^kxgFC=bKc&?AC3MNGgS#
zxHNbWpWss;zz7(cw*TB3Xmf|tu;#--6V3}`VT^nZNY(dF4+Q>{jjP<yWAs6Xjd!?U
zpJFpH?&k0BdXF7l;CL~0i?_Q;rW~*>p~kgLxa69$U|cArTV1+Ysoa2F+E>Co!w-&0
zK>PLAty`UhB<@pX<=~@rx$CdiP#~Lk#yl?q#$@h10A6P1vkxUEV93kVX0u+Ffp!`Q
zJPwo%3M_D^VielB1VMXmQLm2|yWOCRO&#xDID?;jv}sXrNew{zr=Dy7ZSULo0r!;0
z+=3#yGF#N}H#H1<f&6M6niEgqR-`-p2C#I`leVeF4Qt{Y`9k*;!v_nN;c00uu)6Q-
zBr=V^-X3zyepqW|Xa;#y*&UO?7Bge;Mk3QoL(Y-skADAK0B~^fGc@UXehJCWc@+uK
z2(5dPp<WLp%&qW-<Lfe}RHeu$9l2%ax$2nbDMu{aiJ1%YS{4uC&hny=KKV{Y@1$vR
zQ5XGa7E(7V#$a?#Feo?l&fplZWR#h|f5j6%Yl0ZsNyLGh+TpWVgb~Jek{=48ojRBI
zrb~|HYpm;Lfn6T%e{edt`KM(W9A1E;D@ge7^nG~VOTmgaDJ3a2JJ3fS#14R9TwL?A
z%pw2McsCm+Shz;OpYl}>?>6<#uV1+bxUlu)pP}!oe`gKhlLGdq=FXbC>$I*=s3%Q7
ziXnOEByUhCnF>qOQ7%%Ts(oas!i)x7AeO^n7kN;Ww00fT$02R(kLU}tVpiAFcXedC
z^&8u|ipl|a!d^86SELhIGHyMDN`bCgUt#yiU_gh)f;JOmwn{x73_MN#fQ<%xhCec|
zz4k@#YzFK23XvRk^f3BCSQGV;bza31@UzvSt;~QGh)pdMRuC36pwzFQzd}$f3n!9J
ztBTxh;|wcA6L^ig?Gb=*g#6ABz2R2Pe(#B4lZ#$>okl-xJQ>+UiniZv2p&%_RY!u0
zJtCKU&VwFjY(cMU(zsJakU*}A3>`ZI+8P0I4{pXF^)H@TOAzPD2`5mpWa@|%ky~uq
zS%0lb{HjA2eEAVVI^=BK$o)2>zPJ_t`oqVw?>Kgwd=f#Nd!y|S2vQ*4odkn?WZ`Hs
zU}btJaHG>NF1a6%<Qi`<o7SC<)$d#?L3W{5NIL>YF)8Fanr8a5ztrR4r!)+E{;8(A
z44^0~oW;f~2vwM&KZLW-SRKqW#Mp9WAd&a}>s4_NSVCr1+2vHe)5!2#YB6eXfvBTQ
z&D?TNEXQU)NCdDk@3VmY-0e!-mBjWlOQ9BEtRE44-v$&g>z}WFa_oYjlanNXl^P5(
zFk}UqnDycq5&M<jKi)MyuzcTBKQR4XQ~lxv-y>_k`n%{|Z|PjtX;*uE6sKywa;r5{
z%CYsnclM4SKhqdIPk*L6s@Ao(hMw~<)?$iGz{8>zGK;r;F)TZV_RU0U_)QaJRAM7&
z?xS1Ss&RN)tsW=ml?VD9h1mAWL6A&|K)X}1(XM9GO1VjFg-F+BOg+26zj|ry9j6kw
zJ`RItkD@}gNGbsUTLHo>>9QfN6aRxfB|<e@Dd{BWH88Rhy^aA+Mn*|vlrFJu_V-+a
zdM=;At?}A=uU8Kx94om}*d%G@t$g+-t}Iswh7<DrV8K0h@NMd{(`E6Tu^bJu#^9>&
z4d(Y^dN95}E`{w^rdSMgZ;q}-YEm?2t>CoR_qqyg@I3?!?U)95#sRK)<N-w)agQ1>
zx2q^S;i1(1p~ng2Y=BW~m)j+Y9$85?6-dMUokQHJ<6_?Xp)ogdNP-nzN}O)kr0-mZ
z(*!!JB=Gkrw~Uf8n@!-P4Sa9PHCtTqs}(A?yBN4bl)$WUW!r5gX)GbiiP?H}|K;)0
z>35XR-aREuv5m>`e6PsZvb75nj(2&%JN7ChmIXf2I_)P<j~t~YRFc%Y4@XC>8QQ@i
z-aI3^ZV=vSn0_!<NL90HB?eHxFR3y5+DB<4PI8yThUv(G;bOhc<}Bv#gn0?SPMSwD
z&58MsL=?lLD|9JbnQ1!09EbFrM(D(|NFQMmt6=QOMc#z|tZIOrAsuy^t<Xgk6EAAs
z{h%<lq<0Ax?=ZS<X#8wM_AV&ttD0DG!EoU$BQD<fm;CLlwHlF98fmJiVTwcxsdaAc
zQ<xe&WZz5l9-G-UW_h#>(*1t+fCvvTPF&G?tL#G88BH-GK9h=30(2>27ehZZMJ^k1
zx`l-@HmXUTn|KVsNVq)lV`gu+BH6!W*kjk$RU|`9X1TdZZql<k?>eSw6gb1@Wr;zX
zSMp%G5Qq!qP>`1t*<_|I(n9JiW6cUc0nyms9d&98<qRKLmr9CIs%m0Z*~nlSY;t6C
z*{pOj58`MI^dm(yOMR9UPO6(r&U$joOW5aQ@KMT>edGsqvRw3NuF3=SGT0O$s$lv@
zO8F9iBb^+IAlQD5rYnYW<9BOL3E+1Bg(||0i(siv#IfP%krK)<Z90HGYj1XbbuXY+
zYIj{~!7;Hv#Y3~Uhn+ef^&>)A)9kn^7`vBtA;dg!?-e0F!EEl^HY*pC5ExND+4U@!
zE6m1e(w|=Kz<Pha-&PEh1JDz2a+Z9@=j--qyC?y6sOVEND>tsF!OEGO-}&04ccX^s
zSfJUZ#gm4H&2bU9x=>Y1?AgC>DxDQg3`3*$8CnI~q3;`hT~p*FFqhMpg%OBL%P?&=
zT;!(o;yq#&qeAA1j6l6>v5`y+8gm~U8iE|xE>KoR2$sLHk<gdr>UhSF9H)!m3oYqu
zGJ9JCw6jb#PlxZ9`EDu#A8@$`ZXbPud%d52pRF^C==%v0k3@0hDi3y>`M2z+JNXT7
z(|&K4PtgW%7`~6KzL~Y&_q|s;yCNy47_pavd0HA>S2r?8gZ3{g-*HpDpIiNM_Z#8#
zKOZ<()1%YR<&<#MZ`rq?ZumrS;@_b~+l{KJroZbh$74Wxd3DMTcRK5FHnB`iGja*E
z|Is1B3L@NG$DaNgwBo8NJ_dgA)sPTSOFetA(??_)IOhQ^5UQ^0z#l18K<=#T{ec&#
z9x6EO9|4t2p9y1K)=eoE?U{wmO@gJ$40^d)?Qetn|IUc;9eq9inzI?T+tw*Rqa_t6
z`@8YBEjn+&5QrVNcj?|9-H&7e;Hh^kf#JIYEG57L6B%>TB69F<9C-%g^95XAHoh$d
zr}8DZhicAJYr&gbY$RCx9#UVmdF<fYxn=ilFR5`{MfXTDF`%NlJQKrZJt2*y)po1o
zRC{WQ=z71^Pg0Qp8oITtB4m7gJo;M=>W~SDyrkTg!BCPo(b259Ot#5;x}A7B8OvB2
zQqtwGG21UUGX>=FHn7_*I4rfdBPq04j&fog>3cigf}Tv#Z!5$d33lGydAIsWem+mF
zhDKg<Seb$hL=R^EYdDFL=3acuZN4*3kRmX%U0%1_nZl}~ncWHBXJL^$6G+EJ<118a
zaWI<XM+#y9hzWX?9Q>D9%KkoRYcs9dr3ogca>HNCUX90+DbsW4s0vqehd0F|S`Isx
z9f)Y4HT^F|0aQ-{25ND6HOX?KU*=pe`LC64ii!x_iY!7n-nWlCYjl#mk1idWs$s#Z
zD4YrM&zU7vjSCP2<bt_s*uu~Mu!DPoykd$^MkKH!&XD&EPx1Stlq%3g{&YV;RU50E
z_|T^rr8urPnb*#S7mTvTJW#bwNjtPy&-S2QWQNW(t45O%#jfB*YjRDCZO(aM$=&6v
z&1G2gsUoT)I%16`Q7y+v?Tw<<Pbx`~!tPA%Wny;M;H<{bC!a>mnWM#(lt$56y;E`*
zSXgv0>H}aPqh$BDV7l0EIZ8^f{C~bRq*8n<Y#^3WIlQGitPF0dix&;+`FY7;_*HqL
z{i*iS&Mq$C7aS+SFw5wh;y=hl^9+7k8A=l=OmrK%_%w}3jN1wnHMctRbE%|K*`G;P
zrtQ0jYzN{wb-5Jy=s1mh#S0OG&_@v35sg$rcyoM0#iXW+F<|HSTaA|L3u<DDaafT1
z;lIC5jpL@4c*tfEYd=}@xC*%XBpFI_qQuHeO`g1?M56{x0y=DXyv810%;X~hhzj+#
zzgo(1cV*36H9p2Q&e_3J@esQ(d*FY4o}SNvI1GQfBH`L(Q1uH<nJstkXV?<?83m3B
z_wWQO2^^GI$ybrN%ccD0jwMQXhg44D9~BbO5QmTYu88TE)YdX{^s|$&i%UT>b8RW)
zv*Z<Dr~>n>cuAl}bbH3kC*Ixy*2cI^j$@4HamS)uw)S~<DJO@~zSJ->rB9Z~mY^}p
zPcvU<Q8uQ-7YrqgmF4MkJ}m}pzXVS^Juli_CHFqTmAvx4^>06`^|;zcx^4SpQmX#j
zvUl*90C1U`8U`d|@RgMzYBtZG@0ot%sUJ&_1|O5#djJ!Oa{FcKls*OSV%zJ_J{48h
z-Hpt~O{=@(CkiyD-Sy+=)#dfmt`Fn%^eF4V?k*TFuh+_wqoboVaNY44?vy2+Z6P5d
zrk`Kf+uQ%hKnEtR=DL6$K*4k*ZP4D)TYeoDWfkeL$FVF73f7RbYO8-a(qO(r0=j=^
zA2LjPqR&&$!O;=uUGqz*=pZ85UWY4qhlC;#0y6098Y4eDg(SHepmx}m$YGmN@<d$s
z@bLk5+(~1?ZP=`vA5hRpAPu=xb3jYStrAlXOWqfNKyLF#EtacW+wo>c&$jEq>5T6a
zmJVMp<L}_C1#0T3Y^h5TaoukCNGrk1!z_n%t4lV_Ai>XxszIc(aAiO{Gjx3Xt#T1I
z(Xy)*0KmvkDQMC!bzQwtjozYGvzwWjz5S~ux;eM=%iPQcuaKUD1&l6}BLWXBK()z)
zhTN|j!y($+AdoLej(i>pGSN&2;H}Cure&9f3<l?nEJor-N$qVTUEXG`Gq!mq?w%TS
z=16qCcbC4S<qFjJT$(Aa+(QOakC_xCTajXzeZmH=SaE^cG}L_d70l^`gl1d*c}OQ5
z$$2ZkX$-NTv4DFhSi}C!AF|cc-N>Z=7r~I1w|6S-YLj71F8;wTtvcbNSBTDbO^6^q
zJ+)^;X92_FlYhQerl@W#MhN#qqy)~DGTI)#b$0e6psd>Oh4xD<EI}LQnzpQMw}=N8
zr&pXP`U<sNw`~yxE|!{Ef~Ms7ZM;pd>0O1&;BU|F{PVCAMfY<4g2bmBJZ+;YDk&b+
z^*Rmmw1EI6OhqnvOhh7{G0pp!;lNwm<l+qmIM(F`ME|uhj&krU^(O7WW%__&cew~s
zH~(~-v8ai7p6h~zKLIC-6~!moG)K{tGH8GKA`Jt>@?0^On}U`Ms|puz2}D52@hO(2
ze@6E5s%agogXcLiSr{EpDu4^Z(bJ>PvUwOX6i#T0fS=FH@T=Um8J4PSB@7JlCJ|P_
zP)yW4Ib{l9e=;ZkTF56uRy^k90ysJZQqvlz=AVff%rGPiE$*pR{z1T=u1vFWHU{CD
z4Ex5{3*ssbYkoiYVlqQX(1C9QI?|<MF<y%scI|rX@71kQcfX_=CL3;~(aM=~RkHuZ
zW+6QxeHm(#R}*v^`dFlme5D_S>CI+p=e0;zM!e;*#jW*PU=!y~Th>QO>n^d}LR~F2
zhn`qGFgzsjiWkm;w!gQ7Q4?<Svp5L#j^uyD{ZM#?Ur=#1PCP!bgbEX_{I<t$rr{0j
zOXj~6Iipt5Lvv8kN&|d#&z(Aoov?GTi}l73WUvXykzlpt%Qado$|+Sk-<4AM)K*La
z5mTJmNSV-_>Hd~ZX#>&M7BsAP((!r^p~i|M*+6YI<KmU9Bl2RWy^M1nCB4a#?*RJd
z5482XauMpsdJ}?20&DtYSw@DP&#9O91zK%F>9<PLq3!3pO1`caj>n#Nxt_b{>6?ve
z-nX6TTty9x_E1s?@8*9WvA&}tf}dJ7Cz!FwTAy#glIm$M%kK*56D*X|?y46kNdhH5
z;l`ThkJIh_n|}cc-r^IiRE>a4IC~q^`)v1N7X%^6#%Z^?Gqu^d(fjzDsQVh6$7*(%
zIb9Y<ff6YAjYbC0PyZ0Jm4{MOb8>MRvuw_zx}d|5jEbq&!rAEXI|>xuHPNVu!?eA_
zJ}rKE7lpjZEf9{G??qyZJ~(k4a<_hpX8NS9Q87M-v0E-G|JCvxGSA9pBN%Y(vM;Uq
z*M6s!4!{z1Yu1*Lp(Tw=K5y$iYQvDWosy=s5UwD()OM_u`u$ueL#V6wtCM?GhG{mP
zW{bu#H=dX5gt>Z;95*+hN99QsW!_GnV`#MG%Q02lhaok^#Q_LprkiE3(@pVx=ufQI
z&A)GIi#Hp35-Q8O`g%7@I+s^KoyxFYQ;69ei#dE3<o5w`e?CYZE6ixb;(c2Hb=VQv
z<sbUcSu2)`<NZdWZR)zjnm-L&o7Zj2fp(u1FvZ5g+X9KNcGe$w9C|n1KYF3Q8-ReK
z6X|s3%90F}6~fTcTM@@l61oLRPO0IO0sU1-)%q^q>r(ZT4G*Z;hVbq884n$czPA6l
z^~5_mMqb+GgI`fBb;4dJz4VRjoD3@2w_LqbCCR>`<bGZ}wdVf5yA~3z$VW`8;vwj~
zCK9G#EHz^a6BeMOHS8)X2`?0xoy=moGB~=yu!*i!jPD-Vz|hBmy&Ql1Tge}pZ%L=3
zqp!L8X_Xzb2?b^%G%XxgM`sxR8~hziVq&zC!GaB!JBrZt9t-A1iP+JGaNO<B*`5J%
z?nL-I^1jyqM%4*Cg@Zxq0ctV`50KPt5Px<;X+*PY6|llG$-b(Gxe?kAfI;jawnYvH
zKL=mqjZ&X=rTm#h)f}@xee+WiRZBsv9AA&5A<7{ERh)I$CeBt!G2*AH>dFh49vn%X
zE`!!ot$j9n*%t4$gzmCCRhThMCSrtW^O&x@-3XhtHbrXsQ?5mm>eze|5W$NbksUT_
zP(|OH;yH1_aIydl#|d(QSQvFIP=#ISuI{ks)Y(Q^)UFXpsWBxR*>WqB%9K*VxNOdI
z7Z+KjH-t6?f2dgYctfcjs@)(cEO5JjT6b#$eTIXwB5K&(Nuj$OG^{Y>NV(!c7XU5O
zNJBR2H4V6BDhE@1ms<u(CcNsoQ5Ni+<b}E8+sAoAAN`9Yq4Q&pEMDb<_~zG+++z=P
zUC^B%w*;YmXRcA;c-v>!hX-a_X@iy(apioP_wJ}3B}Y1sx<sc-iV{(s|IiU({LUO0
z4olK^C`LcpnRO_|TtRnk`>jm4&m%53K&@J4il7NYV3M=bjajVtF9I9gB0=#`Eb8rQ
z!A35{v`N)j4qQgW6xA6hod*78J%Op@kQ*qYPmiJ0&CXHT=o&NTGu2aTKrSY2z+Oa4
z;5(R2l<2zoJ0jH)19BaCTbl0f_j|kM4I1L+e2g89T&;7Go<q0M5!SmtZ0#M{Ow}hP
zki(XHWX2^F^qA;<(fEuo_`GMjef-=K+`%(=<=OUqUuVk(b+>NqpBRpf6MbS&_del=
z@_N1XU4Agu`dzxzLrE%Elo+w4|4$=UVlP3>2RPprI(P_@;8dlhY{~ZUygFp+VEt8+
z3D_7IyH-Z%9{S9<7!Iu|aG@hI4;Jg1YaK&gf_Zg@$-kg`S$DZc4cl7X{!kgbvgLj(
zn0Z_eY1C(f-n^%r8>WxPSmv`?_8r2J5caRpFY&P^fdNK9iA8$U2sT5y7j85Q&21z`
zixVs7Golgts=2I@&XP-H3xR-I(H65h)IQ%<D|s~N4ccJ;g3ZKOKQFwZdCHs~3=gtQ
zgirr1YX-~dTF8Uw2IAKaj8r`xTAA+tqU8{<Ew8*sO_Wq!^mZ6pd`oy<#LyBTm2*1L
zo)gE63?AGbv_C@r!!vGJA7+8F$&Q$zdu_(QApC-~zEScKSPI)@-;Lo}U3Ozj{dHjJ
z)b{gie#=QDAJh$jRVIh}_x|}{^{u^h2>pU=1<ZQ}AbyIcfKZ7{QDyI7PN^RJH#X?F
z$dn7z;({sXq=Ulp3MYj{*24sox4&YF5U-^^9n~*NNqQwK;RH68=Q#t`nu?<}TXX&s
zcfGqEq15<cq{3hW5P@?vF<dmF$e@y4vY`!JO&~E$hEsJOvUH(bfP_-@vg#?jOEac#
zN+U$>AOfeoO+5<aW66rk5ix^$N5-e1m)}Q`R?61nuiVDX13kyes_x$GjIb|YU%p6*
z3aM6hczI%0k&`D72PTxypvGJ&xKh(UBQ*^)_%-7ra^oY)q);b*A%VgEf<GZUeW$Oc
z)!u!oH+4vq<Fq|dfdgM6#UeY}MzESZp80J1m~+dw4eH#1mW6~`{ihOBaDgDyVD8k8
zkLo67o8irUxr)5$HH~r5?LIwlrI;lPLOz-KSB5^G9X495jHyPqr!-_Md!+XRDV$ll
zuPEX@@j3#NL0Y}q7_}IpJyHwyUU%j~l(I}Vl$Zy$+(|B>V7M%3HX=ecd+=@$qD&Uk
zPLu)^Yd>+R*UK%_Rf4LKL?ZTr{mr-`#JDPqTzs1Au8dml30*L4kxgziv^au~oiUg)
z`SOG(t_WY{##l#VRvlnfGh3N#oMfTRiiN>%=A4_3&$z)4{0IY6i^H!clUjhKPmdU+
zr;5qivsQQLvG#E2_}5gPW)V0UF%ui5PhVNki$;fC$ugzPj<IB+YhqGmL4IP54HeQu
zRXk`dP+DU}!#58heYDrA6~r4+;LOm~`IY&A-u(G}m=OztKbYDam@p+GVm)p$j`(-5
zus1EWd--~pkf#5|<yD$@UL~XS4OMmTec*XGhwB{R(`4KmkZj3nleapF(=yn)IiwOS
z%$VeUUMau}tvK_?V&%8rJsS+@8XaJ%(6p09u3!3x9%<!XFy+67G_lgCNH9fiSy;yv
zl20}$DeF%HL2ELjp112tDIog`d-Wa<pH&`rBz^@U$}^UKX)tLpds8buHZ+Wgzm$qf
zx7&v<R?X_6lR!<4fUhSeCf4JoPvqr1N5G8Hwq{kYXZJPNdsNe;|0Mug0!}_Qh0p7m
z#p1~Qb!a-R8Np~dUwEf7*IU{a>u5D#zmPj}72gW}6ZARn*LV4zy}h6N#`N(`fsp^O
z_ucN31^`C^{08=Ajb-omA<)~y_A`Tk_jSUp1PFfX`|u!Vt$w4H9Tzu$VGUl&IGHcV
z%`X*1<o##zJY&4l>f>$tGoV8odALWE%%i0fA#<`K42#qaR7GrK_zzgnYl{2Mr4J@d
zrWKbdpM05r^qJ<|_8wSVMA*lsDPsqkLPXx5&~Jr)cd;HHXB0G+whwesEm8FY+yl>n
zDy)VFESZaC@b3cl_YY8t_zbaOB&>*pM)Wks^;J?C*!%-CN;s4CR@@O#si^7}2`RGE
z*c5nd$A5&aHaKSAuN1gG2p3`nO;zYZ)EDfzDd|fFe{3?ep_ONTdpD$2h3!yD3edVm
ze1tEC*=p3UZPWXvU7G!?b-RAmdgapXH^B1C+AlQv;o#&XvyISyOno`N;M42-o<a0}
zBvJ6YJK^gUA=i^kY4Usl+679`FWjNO>^Oxhpvr(8xpJh?0J&bOlrgurwqv_3i8IQ=
z-Q$)=yIYNDU8;gxrti!jx&0>)JGV@|e1pN<R*`aXJW=C!#t6>LK8w)**0`w*QjZuS
ze+j}{x8QNCGZ-o9m(0Nr0-9ifJnYr;2UH|%GGA=wCG^Zh{`JX_`qMFz0_FVPBdK4O
zL##r)swqW*ny@v@P&69}X7JtiM-d*Df5A#or74<5Rf;v|Lg1f%{XFt~sX@aq>ZKZt
zkK%26-o|k4@1BGdnx`#n7-7y6mN;t-vZZSw;)##q8T8I)7uTu75iB<!_dspTqC;bn
z;EHg>Oa5c&vm!wa68d{goRyZLQmv*0qvMQCAx3PXh+~7FT%R>TJ4Ytdc{|ZU*i5L2
zL`o`2y+;jdA5Vapr424;@PUPAkxUSmEry8;Ju^x1M=xUa<kX6?Ep8iwsDzP1Ym_LB
zh#EEt4&Hm@BvB40p~L|LO~kQwVy}Wk>4V3rESROk{}4+W@$T|#(={aMn!}P#i#jTw
zAn#L7{PO5dUINPQg-b05&h14uco`if6;_U6)|_}9?>X<bBc9Ee3Vp&X_Ueg~{K0lt
z6V1aVL6Pyr{#X*<bNDrH{+)$|N_?++q$au+sNcR=wete=3qW^=9dwlzQPVn%Oxd5O
z7%zleQh!2@2=`P_)1Z%yVNRr$rKxIkGkQ(?#`1F%dSTg;4AQpa`}7nImQW&P2x&st
zH>2#UZRN3ThPO^|1Nri=DH{|znx}|KDI&d|wRYL?9kX%Kv^3-=JF}0c_}lV)EYXC2
zhn+d2rcGMtLn1f=kHIr*ncC9)Lagv&vV=On7H_XZ7%NRtNG{_azN5jiGW?{4l#N)x
zJ=xJ}lO=dq@XT{n=ZjN>AiJuXPN4mn<V+_3A=|{}!jM98%WZL$VCyd&?ax?6fmUBl
zAuKL4q4Jj%(P7Ho+q3_a6M)(QiQ+=CDweYrosLbywG@S(SN^4S{%vocNu<vkIF8qN
zKj)SmXa%CHp$l9BXv^(iZ=-^5@7thC38bgJ5kIL@%MLvWHO>uq{@41_ZM|E}p}<CH
zuH`FezpLq|<5<55zqi~ssyEPQ=V#RC$>)Q?^>JxI%KJ%V?bFLC5amJ=*iF3j>pwoG
zQQN%X?R|H98UtRG`+gP5=ld4)dG7a+d#ls?&YQ=^(%*CWJRAFw@Ojw#VfVBz_#$F>
z6=ihzU^>nm6GS|<iUw;G5pJS4$t=!+R?VPxVBCe;h5oPkDkiH@H8~2B3iayzJS9B!
zpTx9hw@LR}+#V-A7J%=iYgs4ggU<203QC+0N#kqLMR!@;(H+t-@Cj@-)Y}<}&pg%9
z_r4<8_P&Rjy`@sVI%}cJ_*>}8dnhLtjhz5R1~R}$Q-TcPQy$7`tPs~T*#F#{mg<aI
zwR^{0=8h{*OUK1f`8BrP$#&u5bliQp&D2#Ash1qXS326We*JJ0uxU%>kwDY1Zo6{M
zF^HW$Y0BsGk~BZBqVMB0dE)9)hpeoPe5^XkqM@<;HcRx5lI!=uojDm283{LL+q{>E
zUZC6i7;8W&nFhO9j@&6&66^-oOfzsp@Y(KlwB_6tdqdD85XInk{qSh_+7oc@>%bK)
z@#zv<+v72WKIV$u?wAVVF{zUd7(n-;&T%a<KDMl`H_t*@&$h+Zylgq^eATMI*)3$l
z67G6<%HMCPDXV}jMyU+mx%M^gT}Okfm2EScgBbTvEkn>QTp%b&QF(qRF6SnJFQ!%~
z!ylte4NB1!KQF!x6XJ5GNwP8^ye=x3)RN4kF-BCA^{;agj}vn(TbZ|GI<C=2X>C<x
z`bLSGD4`|Ssrx5R(&%TH*V_ycS1@*XZ6ds!&kFtJaA?8dNE$1Y95lLumE7MG<{S-a
zl|}a3=cu~Sf6&UXq-T;~1w!eg4tn1!b*5s?`1rngKbsD;z(uhnI!~6t>y4I?Q#J}j
zHL1^Ii!I>Vk-t-sOBtuljF;1>Q!55&h4g=4a1<7iq{k1X6eAHMU?HrEcixo<NxFt9
z*l!X=Gp)X6K{fOpspB+hS1^KG)soUbSA9$Irzauv7ClXCWdLHWL*DZ9Efb~;x2y1@
z!2Rf<6Tyy2zh0{v=S8Vt16JEEnjd{_p29-8|8P{6?u?67#a<bw=heE8YMPLxmO)-<
zqh{5{75>#fAMcnamWVX?CPa9*s^1)CeDEjCuR(1wJzn7c<hzZc($6yje07MsWE%Rp
zP$ZHY%hG<snFuRW!1}r3M*l?`=O!0hHujV&A$BE2ozt3X6k%kULq;m38vG)?y9Pzn
zWlc1MdO)DBO;ggyLQWt?5WOI~ZR+#t3N>`F%~$?@AupJuyceE`DvqZe^yKz>z&R0<
zvTzHfBSn=KB~o61t_rIr*%N@haqXoTfQ-!m))SPoMjaECf6Z%6$gFl8`j^ruDIUL-
z`sT1gb-@$An3wGay9xD9OiKixpOH2o!dk-$^AJJ}9xhXy#dK5v*=g_Y4a~s%nPOyw
zDl0pyZq?dedt;rml`bYWw)~V^qf$e+mK^!*ArnqOe~Jr-U+`s2;W;$7>tF9j<n+_s
zh~L;qu6JtU0gl1@KNZ~0Jx`a(3DYPs?}i3LAFm-pkBP!i&|um#(`W4`R?q9$<MaFW
zr;otP6WRwbVhW3Ft;Oz$>3!PGy*;_7>$x3sd;aIOU0AL2@zMJVJlF5)74_p|PvpmL
zBs^cx#5mUWuHd^u?iJAo^e4*arr+!5VeV(2-}J#c9p-JurOzeW;Vnh2e)r+aNyxeF
zv|)RmO=kaP`IyvqME@+v<EEN2cSOAWJR%^q^9!GU-DaBWrQZViHEq#y(w6sUjuvC>
z|1?ejz31TUEGm7&69Zn<@~d%m?lC};-{TG52p%}py}ixnUSGX|ZTkM82qg+e?^(Ao
zuM!Fd3r!x{5Sl7Uv%ytFJWcv1ei)=eX6ZNdaQeecBBu8i+|Y%*Z)r8|uKS}L9W!(<
zaDc;ohcG#_JCHS>oa*7>A*9ug3Aom{$L~)<GS~5hVNfjq=15M?Y>U%Q2@IBHR^^Ht
z*%CG(pZgEq%J%m5!~0XtE)ORMYEOCj7>~Cnmd?!>4Lio`{n3T;Kx=Z6Ag^EzMzNn<
z#7-;}Q4MY!5|3}l6dNy7Ke<v7&VA6&eHb_M!^Y=_A1`dSna(b@FMSp+eM7E$FHj2p
z35)PWts&vDK3b-MPhWg+{Rc`xwZ2v#OlC#>DN48yBfTGs=iJ@jgtGW&ASH~Cbo-1S
z2Pwpv0^!p^5HMMySYxQGn!0YVuEv@QYbxrh#hMDOZ5bJj2Q{ujibAbCDfyCrJ*C%9
zb5Jl=qSWR`N1UBK;PmX2<D(<mS&g-K%G3$@5rm<f7fSYqM8K(t$|HH~NW-4SDy$Zi
zERaZ~;uV%etl-1Jb27z94)Q+BWvLIQu#*|{RN!(bBFg(<G*$`PI{JYLMbA0os)|&s
zVtj(~aO6E#BUUN2isD%~Gq^`{w)>H7AK7$)eGJqN8k4EC;o1lK5D8HV{mN*J6L2GD
zuu+nHideI}M(Xg`bC#UkoTd2K4}rEcf76*q&gg<-PN>ktMZcc_5tH!D4B}_#DknvD
z62CSDqG+~;7!pngl9t4gNezB<yriic+9HOSx2@Ei-m_lavfFQ2uU7P3FE%12M|~0j
zQTA*3Ppm;y6y<BJGE1E}ZZ!|)6_1W;J~*E<vmWI~@oNLxStDMgv&d}BkSxJx{21uF
zo;VIroEeOU$|{^QQse{L8k)8(*H<ozV+sZ(>xHoftt=R^*}2croYtIF)}{MRrDB@Q
zP_B7%I42K$Ph*Q?f;8!U-xnnO&M;mTL*m1NQWgfKWJ0N(K_!Py`7kEe-(rdTmUbtH
z(oRN62M&(lhS|u9Zjz=7Q2WIGcFX1GUs4~nxVECLo4cr>G?7}(VRJD>2_+}7Iw+=`
zR+t#56SQ$ecRlV{439$$gd~fgHik52sp2au7lpIf(*16SPKd-QQyWXOn6n!_i~(yk
z{k|vJHM657#tFy9TA9sfB_zw)a>4cLhWUKXtZ7jx<JBP9dHp~?Xf9r@7zfSy@d=OS
z=cKD^UYwlJeE%bUbQAgMgJVAZ$zP(XhJWYp{(T%RuYUSdQV861SB&E*)(6%R#-7{j
zk#X!ffB2ZE&z^9-y5@iXU;H2V-QW6m`3L{tAMo?fKjkMs{Y$p{4ObVJtXCVp`r>PT
z^S6GRzyA;CJpcGZl(Agj+|agjw00725<G*CC}sKJ`N#Zk{@0KAt>5|`{^)=B1IDrA
z?7=COt5~hpgpjDJnh+zeU%%q&>Pk9#GF{gd&X<r%opsV_BLT?HIgB->^LiJR2_f+6
z)oUI<e#UyeV!z+8Sk#=J&9P3bF;KGUc3n^3k9_g@m+W_Y&d<+Lz}4j?3b?p<i?cOD
zCmZ5sBav73!3jwPq^?V7TVY&{A0jy<R5oa<u)1a-5v|8m8arIG`<E-?<t?Y5{4Vvw
z59r*C;c(uUvEwh{cil&&R2d&l3xzDUiQRt3)zu~I^%|{3#k<{Z+3$BH+p*^O<di2*
zo^Wz<!Y7|Jq?CC4<N<Fkp7GgdpELA<=g(hobUf$Ho7ep0CqHGg*#J@qJ%9e3-}#;2
z;o|LEu5UL&kS`BE=gC<2ZnPR>C<XYAVHgA*(weHOsGIswz)ixH={?--cI@|in!3iC
zyQSqm%J<7e6Zx0CUcZdTcmMPE1swO$ynn4D-wo6AJO8^Y*ALI@@V{wNco<-b<V#Kb
zU>TzwYR|*8<8YTnaG1rwI7LE`B$7j+UGBq}!gsP(-jAxm*a~ARDpyfeHO@9zQwgUo
zSiF&Z)l7t(P%0CHklB<HG%gpkOKVN#D(3SA=jZ1<eE5*lv$I0EXs~8N4(<v;$|f$C
zgdmV)kG2uQ0C7a;NK=^uzfMgrp1S{vPUCwa{HrLV)d_)nw;^~pZ#doiD_UMC8Ji`4
zcV1_3*5Qo<9SMNR8F9W76iz9RQ4y^S)>OE;W??J#b!6oNJD(U)^hwi<iL{H{hD4V#
zF>84o_pC0g3Pm83i8<ftUer`+NCB9Wd~fuTV8k{kDM5ZzzJJb5!kl8eQ;NHo#C~N`
zG&`#$pH!C&>T;LXTCxV4GcF~thRRh~=Wvzd=tvT1CS&WA6Z_r1j9RyBcRRZMUbvbu
zmc3cj-r7u)2o5IGPAQyL)JoB4&9TcoY8~fG`B+v8oyJ3Qgso~SS4q*@W}2WeswRZY
zI0SZ^9pipaiXN>ZO=Fq06?NqbH@Vp2*gLdPYkja-oYpr{=9X9VUV}C*=@S7lr&PGx
zg2qjPhx=D>im{Bf%ilS?zw-5@>OP#S1ue}?jmf=rq+Aoq7_`f%Xo~t<^KB}-Z|!53
zkq%+Hq$l?aeQ#uu^Him4Y|7kTyv2Ur(jG6UmJ3uZ2B0}fmz#8YXR}XHWW<8*DY8kJ
zkeKOAlSanvnrt&UEg1ZSc}h1^RWJ?%F?n>Vi6)|J39qi|TC(s$z+|WjK{Iwe-RcUJ
z6soce`#n`%iweeAjCClbi6nd&Iel=(XFvTJj~_f_d33^dwI+r@+qC$8VCXW0z_MvL
znaw$BT8?KGuV205(Zdf==f~Vc!|OM1_>2GXk4eFBHanqRp3?Pe>bPYbJps9DtIG27
z!;kp!r$1x6+flV|XlF}y`#pd7Fa8-n|M{o9{O|>T?;rdFe*W2~eD(ELjDzRZ*WYk+
zeanx3{A2#{Kl#V}qkr$;<MjLyeLv9kBT5-6*U}nC!V-L7x9jNpk>}4|V6CR>_gr3G
zle6K24_@%$M<4U}@neiptX3;ty?Vvlw-@~Q$3NzqZ@y-`-LOl$!Zk8wlr)ynZ(_IG
zlT)T?8p!FO`uqCpFZuB0OO%eBpC5B{v_PrJuPBp9qG}3()q2g(Kl=>p9M7IVXY?Z%
zmzSluPmI1J1<g40TwT6J<;=sgQ|8s4+1#PJ5@J1g;+RlD79dMSj+)q5a-~T&Qpe0<
z*Wvpw*bXqf*fRglZxG#rqzeX;v#9?{SDXrZky0*Wt7+vfD@GY##uT_&-SGDEf_~Uz
zon^gQvs$eRB4t=EmVEM^@AC4+hnyTg;{1VSv6!>je#p)31)qQMId5KHO77y(oTH;7
z`o8DqKmQrQd+Mg)qmMr1<;xc+rKp@ctX<Q1cv5ao^on8V*=)D0R=0FrN6M0bVM<1#
zyr<!v5$-VbTwY$%HVw`?mWxG^{ogIHsvvCB=tpWM+3D^BZojIq`(-HL`(xUF?c@EC
z$H59<T7mD1tvi0;y)U>+$-VckzxOvNo%&s~Oxc(Z2%0Lz3x%VkwL~8ogL9T<)`}0D
ztuVTxu4Xh%OXV7ztx<pv1LI&Aj3yDtIgoQCCnGUvcRmlYu{z6azTm-wb6&puh$qiq
zuv{J&llK;5#f(<0Omg8|PY7)!>r4zIISv@(Ny82m2P&OuYbV!164c2Q<*{)|M#*&r
z{a(<(6d>Lsvl~;)vG2y3cZysQ-=urzf+?gf?~C#=3sB%(g>?b%Gs#2F5rGF}G)?ld
zfr}9cPCJek6?suHo+`Fu5WhN~xZVee|JrBrm>FWC3yExN5!BU=xgNntLdcjhA!Pze
z92YXedCP*Q88kk@m@`i2!?bD=5N0jJvYbi*Od=^wL=jOXF-L>8DC01;qHU_fWK3&K
z)zlK5Wi{CvbmeH;wj>}Jw9y0~2;<1U+jIHmlIzPW5h8?8EIf*rvMR>+`VN56K=#w9
zG7Go4va~ugGnumno}bh#D@Cg!I`qXbU!$#L-nz<46c-w;9L6-{s0cnWU(6Wy9ntrs
zIMO)9V%g9X<Kd7ZN@+5}XSG_^t!W<#MRl&UH1aBO7v-kZy#`@gZ$z9nc}wb&r#@Z2
zP}H#Z&{)wC9Bv@sQsL*NQsCS4?xJ_!ZA57qiRTQ_C3!u6*}C*Aj&CQPAQ8%wmp53R
zyjI{)wAny_F=o7)Jn1Bs<MQiQoSYodo*dE4TUj0hxN3$;x{L%wC6(fcO_8J{eUnH;
z>Tb{JIItKur16$`+Azk7R#hVD9zA~O30jkVz#2t&yTQ(8n8h4bIgGO<=TYH@fxa8a
zUI@BrsY#>9DMb}C{$@p^G{Gg5bBrKrrJPyJmei(VbF*e{8ghoZnbD5}V-74Aj^F*w
zGj_X<IF5AVH4o08&@?kvZ@%Wi!^}T<dPMrpIsf-R`<&1I<PXWe@w?2HCoI~JaQid5
zpZ_IYe~VHc9U^ZpUSpMXXZ?1I9|G-cPD~w_*I)7W<`wOH$p<e#<hTFs-{bYGSG;}m
zhQ8}~{l#bem;c94`NRM2|Hbe8y}!rzfBWxn_V5vvZm4n0Y{PQC<a*e0dHFTBH*aOL
z*M^)O)`{)G;`AZ2#WC${fpeDS(J3!peoP3F|L8v+`RudL__IIzGp;T#sA}<HyS~2W
z&70SP=JXxPWR6aj%x5#Y-Ijje5yl;Zj%d}Rvc}oO&~0ei8lMu%R<ghu&1${o&;I3~
za&>dblgH0_@aQ2IH`lzrzNCwxj3*@L@A~F7tLry>^ztbWT+4UOz<lT-CsNXwF%x?k
z0ob;cjOCPZNnxWC5`5OUDwDznxBUY3^S4B{R4@MyV|$7>Et!aA#Ku8tJk=1D6An|F
zI7z9DD<+f6Ot;%{d40{r?KM$nhGEOi>WY20#hQwv`7zHQf5^u#zR$_xjCnI>42iR|
zQ)Y{n#nCaYUND<|#eTP^ty*SP&%?6^Twc5)M$5y8=Y06$8K=ifuHH5{Bcn21C=Vjl
z&!mD|#H8rPf!oaux0@S!KLCp46V_UMhN0^v>L#SS51+VvdqrE%I9ePtYvyH~SgatR
z*es+HPb7SwbVxq9TW5Y@B=Wvm_Z~`jUr;fLKmN6kUucZq|Hfb6Ij8sj{pG)3y7R+_
zK4cM#@>WKW07_+yQCOp~PE*yEcJ8PvDeAO#V&UKv+Jd$P;Zq8elsJ!&8B<`05nlvR
zF(`7<q%0fBan+#RF^l8JJbwI~r;na7U(Bf*F}O7da9|Q-3^5rpN{noCM1_Im*W_`7
z@qu)8N#1V6%TB=<6FCS)!lX=_q&bRO;SHb@nLtd6?4uy-N+GO4SpZ{}9cV%h?_SNR
zAVyM)jKOG243hnAF<7mL){&i<_lD7z>`pBzU7t`0uh>})S(N<RIOcZ7a<%}a8B$=~
z_Z+vG<5kBu8_(zKgio*wsc@?l+GHxkd#!e+7$ZCJ!z7ECWen(q*k;uYP33S#-v6LW
zNm3Buj6r8;Fx1-eplU%Gl#-&ul|%zoN|}@b))+Z&tmMoZW2FdFiqU(9<mtA1cKx1h
zw-;sozLV>`_t-2}<dp)4j5<{@gHBlzPo&1lq|sw@qBfSft9f?Za;ipZ6{!rgCZqd~
z;1!`Z==nlKN?MF~C)Dr0jgJz=898cZZ3CyZG|^ekfw~sTgMgk2T`5d-GOkx*I4oj?
zOo|d6nL{8eg~k*>UQL8V4)9e@hse6D3`toshU^D)R-kNYD9UN46iOS$VH72^M%3^=
zQq>Nl#D--OS81h%i)tr%OF0S^Ql^Ck1&;Y0RQj(w{uK~AxsYO>UI0}FvD3wM*P$jQ
zoQpV)jH?ZAzWkc`(}$>&1@`C!rJYE$i#eapa&f98zigEFi)AuPV>r<ov+rms&B8f!
zQT~u*#Nda4OeXXL2H^$RVZeG%HEYRD1u0^z#Tw065ICh_zMQjP?+C+)#Sq6pzp1E~
z3(2F^Mv@3J%-R`?#X^*7M<>i?Gh#~geneYC$evF=c|nR%l!>bqH@CN(o}IDXeaZIf
zHRn$>|I5Gg9kfdP!DpA~-6f+2*DjeIJ>>WP{$u_>|NQ^ScDKRl#5%6!Jy#m16uw_G
z<dLRr8ABptS-siv&6_tY=107E{+#c9=Mz@9*KAj}q!@Vf`5Ugk{(tZn|NS5H?4y^w
z{N8tY_Q^-sqd97{uo-##`fG;$il&_rePHbNG*0vM(FxBVopG|9)6`-R8H-RNgur&U
zWgL1Qo<HE()5n~jpL2YC%xu=u_Z?q<{S|L6UbEe9c>DG(fA+&a<@2BYm|zVjCrh3_
zeF_A2yOjte>Iy}L1<Aj(hJL@}&;Q3i<BLzffT(%)_ycyUj@#QE!;nb9qDf@$*{v?Q
zeDx`gj_<xWWuZsT8i!7v(R(x%R%x_2Orp)Si3*~oQbs3PT+xXb2kZz=d9Hu<FKMe|
z+7B9hHbhetv(tElKvuepPzq*qK<=awiZQd_?Rk6ow&WP*GJ@TswPG<_@Z$LkzW?1H
z@a*X`s=6TrPYlw1*Hw)Vp5^k0`Et%?vz7(L7}~aBKA$rTdlriY^ZC4Z{h345+dJ3v
zp>UeWz9U`V)AbTKKP|Kp&C}EOeeq;^*S=&JMs9C!*=~0{IJ@tEcdymF$1j>bpNQ~(
zmC)jsja}bS<mO+0!TVP~zWsgt>Y`3=@xGPIG?rCKUb;JBh$^PEOydzUqT0(ztU?kB
zdPT4iYds2u58@}b?{;kWTlTxXWX+DE3Z1+M1?^IpKCvG&W7H^9<EomfswLH;yk9z{
z;#XG?q{BE#VRR(xK+YrVw&cy4+;`;Q2??Aw^1M?*druyU^-2tc>`4S^u%ak9UDb%)
zhSH@Op&%XkP}tpBuax6}avEG!6$_rCT2zM0T5{#cDbNKk1!01i6e*S1x1=Csa`uuA
zk`yTgDm=z&YMnW56zQ}<Rp4raS0nv85)%w%G(Gwx-1Cw<nxjVNjJD#FR=WyYw>V>I
zY{k58a8A+)vVy3E;IE8ATSILdY|K*lVn91mC_F(WJU?S)9ZOkwiprW|0pb~MR}v;)
z{1C@+6pN}bG6av0LGq4Ml>Atuuv%f2!YNHvH~~{$E;*O6XhC6fVrex;wdJ^RoL6wB
zGp+zZMrWGJQdf?;c4$+ft)S4l@B*#2SSugK8gj@oswGKU#evHzigZdum?n!+dKR>M
zvL&IoTpVnR3Y|ft3K6ngc}gs6RE)(W+aHo9D21nFx=MnUz|``3Ol#1DD66Rn@xG|o
zCsaF$!Sp*ZsuO}=>L7L3)M*@1DUuTZzF+tF)`i_2Pe|Q8i1MIPd0<gV>==a);EW|C
z&&|aZS6_WYeRjmWZqPGB)QPACigt+cunL-#6iz#u6fj9~tTk>JSs2T_)+q0>+My5!
za!tl4@cjTK5@+Z-#^9->H6_OeQbe~6rl=`Fltv~+f*sjydqFWW?AIGG8mGWGhf#*C
zB^qirn{oO2g6)1s)yQbsIY;F*>(vdn>kW_2&w2F7K~`K}-;hxppPX`e@tWQ1&v^9W
zL;hF)&Ug9f`BQ#!JM#Hz#q#uwl~?SC#PQKNVepdu86$1$L?EK26ZHL#O&ECj$#=P4
zZP@Jh3}axkUa`7-!-JC(UOs=uN8kCFtFB{vyCUxPT)w{G`r?8=|I<I_^u=R7{O#}Y
z<i$(sL8C%u))-V4Q&g2k+IEIxpi%?Q6u+x@=UaArd&|X}H(b1Z%kl9sFJ64a$?*f6
z6Qyij%~>u^`0gja0V?yepMA=9y`$2J`Ml-mXn9CTm<Xg|o#Md{%$f%8NB;DW{+Pe~
ziyt%YBWEWMd35rWw_jhAvy;)1k7%9PuWq>b^5>}jmh+|K#OzV~o-P~OreRWi`su<g
z!S2pkA<`BPEpjfMlc91}Je2B;>NdRn(f`N;49y2WV4o_IQjA!{Xd*!h2R_|oUMA{7
zA+V<y*lgD9b{j~Ep&!_+Hbft2>y~Fvp7NcKKIYleC(LGzpdu+zR}%2v4<qAfP&t+b
zNRSwEplutTKYd2q);xUlkmI8x>L~<%T2Q7D;$bnmCuT@M)>z|6NvI7}8l&?&1=S?D
zfN4|FbX~`Ky=K3c(Xuqk?|F>9hq&F5hwpy=*YWM9e}5s`>u=JL^WAe-WsjH$JY*S9
zmiH_xg)U=jl86ol?@4(;5dnizo<WP{k1>`E1V1ngJ$=7tx8Krtorn~?6mxNUKg)~P
z4Uw+*bp1$(1r1Z`kc)iadrj{FLcI}m1R$bvASaLC?+CjcaS)r9lx2RWwE>-^8Onv<
zN0uf^NkJlE%+n@(XHI+{wYbCT^8Hzs9O@4FG1`#Soxh@7&kj-~41+WWB*^-XD~J+P
z-6)yS5-vX;;<L0>R0fV(M>Gnoqsx}t&=GpbF-bPL%&g@44+)Yg$fj85DCaO%k&W!1
zl44<qrl_wAKhhYDa)xYb8P#P)&SIJ}c^)Z>pJWi#b0}6%N{a}^4<q~SrWjpIV-`n=
zrb}hAASmxW1lbg{coN`5C}cHx-Bf9W3p(5yIBg84GsoGi=E<UBiALuji8~0jsHz-Q
zU14lh=mitSK#(nSh}DzVD;C~mqIjbkjW)WZju=WPm{_{pi{8rj#_(hr=bJLtzke2$
zoFoCRQi-%H{<>2Ju4KB(buj&Yp03IA`9VKW)_XtryNN7CJ1`k6-%H^-^eZKER~N+3
zi08=xef)KgZ>Nft3Qk^zd|=5G9%z;=O(~)P<$_)W8T;s@h<)Jl^RHQ)E;(5&v288<
zIwKcVN(!Asx*wFhVRh0tGL;|EzGK-qEF-=jNqBUni6LNsS<|u`$C59rgjbRhsp}b&
zY=A`Z5K>Kzwg%%gz8_E#Y*9MKD9YZk>j0Q%uuV<YN~F?BNVr|!;e!OPJ7;l~BPY$#
zv8ZJ)udg|OaL)O|hxpO+_UelB^K+KV1^w*>H(!3n=?5?QdoNCSd>U~-dBvan`5$uO
zE$7cZ;(H%{kK_3fpMUu!-PqBOqUsvPo{X3(CL&+Iej|nK(9;hArxe7%>hdi=y1rz7
ze8ki5e8RIQ4;i|S-Db^hv%>hm^=Dsl{qxVLo0hX@FL?CqIcE<Z(9B!3QMB_0MdteI
zf^NTI{^~XJqcfVKitYQJSKoZY<<%AQ`J9IjPFc*wxY8KIzUxTQ6H{cn-LhKUuvxEp
z`SK;FM~?k|E8LG#c)(<PQH)7*X18AP&;Q^L_|qT$Il~Zn@xjM@?~~tzXz^a6fN~6I
zBAe9}ufO~m-R%XBk1QWOIO6K3f5~VfUp1EI<qH~Hi92{%z_b#c4mlT23(_1!MMg_A
zP%hT7@30ZZ>s!7aMxMI{d;WsK+TvFwTph76nAGEUBzVe38oXz>+Y(Bscemd$_5+oz
zcy#`l?|$dIy!hY)mdk}Gz6ls@F-9?lh#xw3+Z8uAmt0@H<?`}1zVleCanADFzx|s$
zfBu~Le9r0VDJel!37Pu+<FFuQ@d|8bbI#7rnavglKfJ#02*r!Zhmqi;m?I}yXiY%_
zL&+w6SK*!x3f8;#_ipi>2*dBIUneV-`+2dythl@X@87iJP8lqTbx?fn6if+>N|kky
zgevMwAH`C`+K57vlW?!Jva%s7Q$!dZ?*~RdFpfRJ%V;sj{7d+M&<!K&-HzM!ntiut
z7(GphxIB5YNmGy!v@aLC2x)>OZ#IQM8V9<~7Qf!0yo_B_0rsqwYcEDG$O{^yDF(uF
zEv6XJRu_w-66+){*gH+Y^xECck|@{py^)>~6?=#g4n>1x_m1Nz=XH#SM(lp`GNG9&
ziQ#b00f)h2aN5$=HM(x8TA=o9(>3Akp4@qQq9`O)@nnq=6=Y8Yiam;^H{!rxYfBeK
z>Sm6$4r?@(H8`U&5sZhC4#Bq#O4+GwEX7@<8+v^3geVp{o6QCv#gy8Qf<h^zHjCxj
ztd*hCnnoEKEluu}rV<g4JW7A+y(lzeiuf2ML4%A{P#eX0V|aGtc(`mhU(R{{-~<;t
zf*<gqAc1;P;Z9aq_TUTYtSf-g9Zxw#X=G9qwX&Jeyz<^2nxmW~9<P{P-**E~S(L@b
z$nPsAlx;%4ITXjKj9@1d;`<!l$-i{s3s23$bS;MxYLJug$NRpgX{yqXiD*I_QwmaD
zc(wPaD5?~r6ArCpB%9PU3;p$uZ;#kf{F0P9K=6mATNjm%n!+!{4~mEcGh0n}v*G6R
zuQ)${h+WQ5RRu<&lz7XitO!XlYJq?&4K@YxxTmTVSw(ybU=8Ew0V8wsoMq#(PPl(5
zVXIn*>M`QGR5qdMjm$L51+K2dF?+YiXv<>0l(5~D8HY8)Zill5qcW;$P+Gj&mPboI
zc=>{=X$UD|v4_k;tzofTGW0#WuA`}H&L2Of>j$oGZg_BdO!P<Wx}NIlE%VdNhqDE<
z-+0DpUGY!<{AYaiqkqB4i|_IH(Nh-l6E0V`eDTd!?1m1lD^eOI=sLmX_L`g(A!LI1
zKa*0#I)l?1x833Y^v_w%TB`G7mM2SQi-+|6!2WiF9y9yxmT%tv1^$PB!SU%SAAIy7
z$7iQJc<_MJ;}aIM`Js^O`!|Fb4x8^V4CtJ=dHa?h4Lwhv-Ew^LfS8h~r&2g5-BMAW
zEfzKXzGAMH(v2#0&+;Zy*Ddqef}5Kg{^(!)3x59D=gbyMe(>Ah<>crAM%Sb)L{&q8
zwCCpLii_7@;kPRYo~2!IVrE=_c}dekd^97+NXR1tP#Gs1##GGZBwp__lPq3Dq8v`e
zUSqr2Fz$EEwBhyVdsYt~@#N@)pqB)tNhoa2BK(-zuq?ct>B3A|UYDWk#L>U&82bUO
zG$+ebzWebfeDv}qr^m-aMMyzLOGe>`!1eVFuim`j^_y3`zW9bOKL3o*e*SaDen0_F
zo;>EG4`1-|#Y;|4Pcg=Dd3hzIXCspR$$@@a^rpf|Yt3vn$GQrw1%Zn(ad{aS$APYw
z1Oi#)<#(KOIP0j6kExs!L(oDp*18av?;&gN|9gMUn#_hLRpE4-zY4W`|MS0zN2yCB
zlpPMKCesKfs|-G(N4d7A!bR%=t@koLnVjo$k?<Uf6d~CkL?*wiCMt&l#u!*_x4gQz
z;EQj*=Is0dwytosI|y%J+ApP;I9TT^@r_J=Bn~5Svt?XwsgP-gbwc5|KcY^6FNrnX
zINTdWOQfXMg@0A<I2Dxh;0yWAbIApFxfAS6PfTfX{MWf)>Ph3OimDddpb3E<$Kjnq
z52Z_y7f1vWBOw`VkYtb=MO|4;RZ%^d)6Qm`v~T(Ra>Hj=J-68lrAP~f%L(&?oF#Q6
zq{Qe0gM!^2ZZ;cG4uyD}RZ7O7233$W@l!+@(1x5fA&tZ=hSOncfSBw$M8s&cmb`H5
z1T~^Wh-ujq4aHoyyv907Q*C0xOlcx$qOnD!ltF14m032L<JxdGvphL!_|Ai4&X+AG
zZG%;r-CnGd(m06WaV?2ORb5fnO|cQ-AQYNFFr`dMN}9|1B(q<fwuiz!7d5{s=VXeq
zwo1r>sK-MHhf_ssjWt%98^q;enp((nQg>93IMHV~Jf>V8=VV87zpztE95k#GeqRWJ
z*jj5#vm(Fmj6E2gPv+a1{LVQm&zE9zBTd1iZvJ(SZ&}zSBsQij9?GCpl)_Hj`R-u-
zgQQ=}MUN&1g-(I>H<#Q#dClzXlx8+7p|?iDbWtRkf1#Da`G_7mDjhjFU805^;~<@$
zA3TbHRxtR;I0jS}<yi0|wr)@{5k^l)LW~`|049J9;6}8w7+aH#C#68$in4X=dW_Lx
z47y(7jHc37h~FV%jF=T`rSZck{#QyB<icSqc`h*slv}kk9z1%)>sPP2S?`$7&QN*H
zYPH9j6|*DwaB)ieod-OeY5w@9U-C!a{G9FjmeWU1c=6;Z+ErX!U-S0*lHLbWPPi;;
zs4?WpN0E?4{e!W%$}*3M=Lsy4_1m{}x3~DZp*lXHIyq<Xf$grt>?H=~?dwafZr3R1
zn9UcQpFiaJ(`S6}^cly?B~4RtFm24j8LOS3K`Lk7UcBP!=7PGG=$X2%XxbWMWC5*)
z%s4t`?Tk_TLZFu+r5{JEb>yV@lRx=0KKuMLv^ISAw|<A^(J9V0Xl)^8a`dQ_2w}tJ
z+n;fLeNFOSxMjw0wm9O==NGsh6b6+f_pfpS#m5*(QOLkKMk;F$9h9JTMWvX8JS^kV
zTl{WITUXRnY`*vr=Py2_dh{WqZYDe%;nbCtprC=dG!`mnf*<JiJ9g_8yX}^t8<^KK
zKK}4yKK}4y&d(k&E9!9+_!Jli&&|!6pZ)A-eEHSa+^%oASzU2``I`0WlHGd8ZZ{Ca
z$nX8#-@|H4Q#Ep-m&MqY97E9o=7L1ND^4)lQaMRg7zP=;@An;Dw_~-sVY}PneV`x2
zRQ&Yxlo!uG;NgRFW^K#y(Gkrw@`eNY@%{+*t{_tfgm(I!`}Qk;mF3C(@BNJ}?(UB?
z?|r*WW6N#I_wqfu;=xq-ZpX=Ye~ts&xr}$FKlA=D=l%QCz%Y)iw_7eQ-m>0qSRO5y
z&s&^z(|G1kL}gNOJxQ2CQ=#zw#=Zw11YuEv#%N91q-3b7jln7r6U3D0_NJ%|^8s<2
zOunaSt5PDQN}&jEWQKd#fMu{$ez!O;7$w=FMjOTvyqEYytu@*f-jp%L!Y4Bs=N~91
zDG_oeD=}wI(a8NLi>(b*P*;kFwZlF?<8)S0sY`ye8W>7iNKK{&>SEv<a%N1Co=i{>
zHAE|l*_>r<HL;*ih(M)mwxB{0U?@-#ts+WAP*Dufi+|`8jisV^%-X_>v&i0|@&v|6
zK^K*}OHs&*Dno_AC{fIg){%&~s->wH%qvBsJSVN;(X!#AhbKH+*36B<Wyz@Z!=6mU
z)t0(ysOyHRs&K9@lbXAeaYDr=RL@lvx{Q+Jgr1exPXRgRGU6?0u-0TPKB|*2D|jy!
zG&!S<7G7!=q<I>L-tnLB(F00Bs}JbfL7sC!>)u^|)B$0Rg`4b3eAq8$9!ukDv_2Gx
zfFObQQ++bzk_huH0K0FSJRyi0lPfz&Mxwq`D5>{;K_FF8NQ%rC`kOs(KK+vU$q98k
zqtXV|SVGPurO7&@iPSmLj3cdv<5@#p8Ir098gf=tI?=cqtu?`qLLyf3<|<psMG+I$
zdC6f6l2;mb112kKWoQ;+VqCQ|*8MGGj1qRQpl)kW8Q*v8*K3rm&@LTfaAFo4j@e?)
zSYpMja|Edtu$3mp4xci0RdIZJ#=aX^tygqDP_+w&&5GTwBWE}=mM8N$M;|@p@kztO
zFE{+tzxX*{{mWM@&!6(}gO{A2eV5zrIY0m63ob5S<NZiXQzub8jetHTa@W(3BZCIj
z7|trgOgGdcgk3<z9i7_Kv^9&x40n9O?RLY>ZpSW0hB&h7ZrH8%y!zr5KQupNF>g6O
zS#WZ4%Guc|rzfX0Z9`=%bXDB$m3U>50)Fhp{oW5$l_lr0sHR9&wU902IMNS2UDvbQ
zb?o<h`o1T|#M37q;_6C98i=WM@FI>eF|oh7=KA_8w%sMc2ew;}@`=Y!9ui|_>;q;5
zuL4HZsL3_|(1ik7;l20-x&ta9sBPjdM>Pd6F-yZh7<yDyarNmBnLT*G^W_ok>>&e2
zP?aQPsY7u$?K)G8k5a_o*{;`YHY;ZBl9$h4@X5#D<?;Cwnz}6+gaKtmH6DD=W_`oW
z^%X<EN99OmHBDt{D$7nub8&mS;?3*VtX3=9c1BgXyYZ;H7t2+iehgxwIgTU4Fe1Yh
zP2YFi+}yHSN&f6^w-=T3#cQ_PJ;Nx<*yqom@xk*CI6FJz^yEYq?ZWXATZVUwugR9=
zjx+MEFWvi}Pha0J2q$Fkoz>>or=a{>c-))C+zl1qy~P1CxOemElVA9S{O+fRf{;VK
z_X|fENrjZ?hk={bip_S*zV8{mr>)y}sch=rBh8}tRw+P~k1!0Fl*O!BYi#j5OG>y|
zcZHfpY1#A+&r=Cs>V4$)?p%@mC*3c0?=H3jX7`SPvC=r_1bs`fr0Qsy@+k2!GrcK>
z+hUzG6b_y$+^v|hOo7ORF+&(|!C*Y7sF<}iRa;T1InjHzdBf#;!23v*p;e03Nr8!0
zj49E_h|dX6!UG;zcgTX+2!fjtP$d&pB$r%MiJea3Dxp&GPMpStf)c7?yQE^GLJCHg
zGZpgpqe4C?JWo;)OBmX0Xq3T~aj~x~rfO(vOIsV9^0ZoUzHE7Vy5Q;Q5vP^H#*s8e
z(%6$y5LD1u>ZYcyDy*#z<J2^n`za(^?ogP@cvemtDXPib%iPX~=366sjxI?oF$duh
z9uU2JI7OzGgnYl5y1OU@Axp)ZP93sk-$_1suc4gQqLh@NhNU5x);d`S!8!d4X2^H+
z0x7iRODQf@;=mL91q#BiIKF-I#5ADK(-T%`6A%J#F1coR>&WEg6ti&eEE*LOySFPY
zKmD5dd_nCT&N#3zkVq=hrp!DBj#A>pIhHC>=|r+K>e&*Mqe5fyggD#_pH8vN<b$Yx
zVj%cHJq(1Lz-W}UXpE$_D44f1#$jZ2bBnGl&Af#eFdDij-KW{@m>nM%!$U<?*Ytgl
zE0I?H&||9#HCaknL)A14oe)@E<v4wCh7Uda^(|G?(#{&Xu1A@O-gdOwF>71C^W==9
z<$~`#Z29N^@-zO2&%WWS)fKZx&sdy4;rlN?Wab=~tG8^29i#Wfg}j)B8YNv|OzaX|
zj)_HB@yL!mFg5e6InI_NrHT82es#${!P)UK&!0cSS2gP~aJ}8p^@4WBp=0AmwwpC?
zUR?-ZKxyi_X0ezT73>_>RPy1BjK`hD+6e&?RpoZOBYN<ogrAS&$T*INPRi(-y0Tbh
zA&Ep^DM;Sqaw6?ktgbKVZ>~tGC#B40Jun1#bllL+8ZO^nGS8lv#m{ZASQH<&kr74h
z#L~oQO*5fm#o$sd#tCUk2Q<ToT#1+>waHY&n(Lqa8IQj68?>_%e3`gTMwYTz6wj||
zK}fO)DFxfjihjT6qZi-d```Z!o;`WSteKTiZ9$Qvj|^dCx9i#O*2E~0G*xAprJCcT
zCH~Pli+M|m8KY~gHf+~h*3Al6S+<)EL$^mEW3gcz4r^II^juwCaeaMV02znU6@9<w
z@~z~}$_6Nzmwn%}UT+CP%j3<P3vO?3NI4z6#X<;#7z$<M-F4)BTN0IohbymnDqNLt
zK>DWdy+6*qpEw~u{~L<h{lf9>_%w%Kxqtiu{CDr$_y4@QyFmufQdEq=bF*G?yIym)
z-O(Q%aWtQZ=9F>hy-_hmpcY1Q7)gG>cb$w9wZWEj9-~2JeaIG$SqT4<?>dbS<zA*t
z-ovIkZEQ-R^1aQmJg9e`5_gS&jOpZlthJ<63wJAcF(S5ii7MjPC={X;g+~fNQiLp`
zgqX4jOME7(o@&szI^o7hO`>ic&8%Yi;bWq%_({IucGHo(SboeZNe!_D0Uf<(@Sc7c
zrJ>GI(nxYinJ`91!b%g!@^hVpQ?C^&N0iDq4c1m=G%tlIA)#W%q$Kt-Mq{%0s8v~5
z!NpCam?FyyXtlyAMU9~`mdX{w`Py=Pc0}zohDf)*=5$u^^z?`i9-Z-E=BQ#%3Ze)d
z$5GZ!BMRlJlD{KjiF;`!3KSY=t1@nv;#?IpHp4;Sb3nOEJrqL(RRkQS6#PC&27HV(
zZG$$FL~@^tkc)Dbdur31vheC+8T}^ER)iXp_0R;3-=7kQX}j!Q5|KLMy)S;Hg7_Yg
zRHF|kX`234x+^wgnC`nEg7MB5@Yg-QwRl41saq<DT`?z|Mh9x@Fmf69<RYo(ZrtEh
z1w-b|XJ69HYg}Eium+ZMvaz5NP4v|Jj(I=QdXLQm*(9RYM6JY90*y-3;(jl1JjHTi
z6gb~HG@4plTBC7wMYa|yLo~v#3OP}?EzL0C$APM@2z@M>b26&ycY83FYSv;JA^Z13
zj}P;M1OL9;Q5BO`$OnI((R-9q%ocM_`Xzd}Ww(CI(aA%acFqtpYD_4fi8kO==GiRp
z;DaUSi|_IA2T%EDKY7KEUVp*$+qW#vAMwGXa~?lA=i=syi>oWP-ChdX90@rRW3lH+
zV0z6iXTDH-E=@;^;%r=TY7D2&u(XCsIr{Y#yJ3r79<f*~c=Y@!rkatGX1&_-`u2vK
z?FL^8ph#pNI`+Gcw-=XE6qOHsP_)SfVT?fPhoE%hNKweRrgZCC>!L0;1n<ciYGYuK
zcqAP?{q_dGx<n6KX0f9iMsB+SABBK@a`J$i)ta<jW67K?DrQHHoTX5oLcd}ZTM~RG
z1WVOaMTTCwx<cY+;*yfdG0+X3x^Y;oIGZ~#f!nWt&f`ZP($psmQ-0>e&zR6UCWt>`
zwZ>V2>Cey4`S_!cIDhZ}qYZu6lai3$LmcoiFouDy_w09DjLytv4KZ4(N<7UToE<Za
zJz7^d)38{aFpdMax3@rMy;-r_?aKRLO8GBG-w%BL`R5!jk9hX<DNVbit{24Q>HAT#
zAIZeJp`EofO@niLd@0cO`;ITa{E`ozK1ZXds!EiIN%AmL&iCTR@;h;4AT~#~aPeZH
zH2gB}x+%0gA70aU*Qe2s-)1@cxBU23mD<1N@hkpREUMhmN3O1JxV*gN!RZ+fPR_`u
z<#jJ+t5H4`rD_&MHYq?DNJGc4+j3ejsjb5qa{#P!Q7mO3gptwvqUH>TjZkYb6E-DE
z?A{C-rV+FfU_t(FE@N0#5Wsg|zmyYx99iZ$bzPw~3`2+HlTQeVjBgi<CDvNHeRrU@
zq#WtSfpPRC4{^+lV`S(9Vd${dQ#B)T=&5HF)x6>1(Gmat$1i!gIOeO1TR#2jg5BVW
z!@w8=ILGm9&cnrAOp-%jKlF4tFvP?dAz6!QXV|tyHw|k)aIwB-(|7FsNFOp)%2dvB
z+{}3N<Oyvnjp5k$y#4wW{d$Xvi5fUn;Eci=@t&<T)J}>lqs?L5C{kRF&xvZ@usB+B
za&pYVT5=rdb{qQLhVzq_?|k?HXGe49jY9`dzw?YG>ouThXW~L(tT{Bul%`o%24^kP
z`L4^zbozV}N2F{I)><h~(T(z<nCcD0MPj?fdk-1r^BHwr6S+(NxVyH>FDYpPCe`Yc
zc^!g~;;gC<T)0>i!0*#ZT;U|QZOb@{5<U1JsP8D$F>6cATNyKJ{hohah8U7)28xB7
zOl{<Ge4B~Hw<_$+0pLzpCEZDtz!dYol#4Pp-${sdo>YgCs1gxJRtB35e%*2T`PVeZ
zbJWpHMs>BK$(hA)jon<6x;;4s5eg8+N~MZKYtczj6+B*wfIGK9V<iqO#7M}A=si{$
zY*Q1uk)%hougFg0V_*m)T8lr}(D$r1TdLA2wX-=SMc)k!>kUd}k*Md0tHigiY8_^@
zWba8sCq6XR5>mp|7C#2IyB&+gf}`V8LcizoYJ=KzJbd^VI|c}e)DOY|7!*-~b&d~@
zjyY+LdC;^R>7VjvuWxzv^`~5~Zg}|gC7(R~gtPXPUAN|&H($|>d&Uet3&f71AUTSl
zVIMQK&fE@(Rt07QEbEE~mBnO5+z;gbiumS=ul=W(reSe*#^dMDc=3C`!O??r-n@Cs
z>sPPY@Aqu?I|-H6s#tv_bbuT+UKt{H@^q^vAIMmg)1^sJ25Uq>6NW@c1I`%8J;`rr
zNZ7t3Y*sYBBgUSs-*K_o@oG%mDhXu%&1Ww-){dLYw>)(fr#f(IGtTd^xh|%`5oKir
zlVT+L%yzX#XP7S*C|4E{?@MTVCMzW(naUCHq;bG0htry+Qn0>+)oa|DC8~x{6l;aV
zCIw<amc`n^usmAw(Z?Urv^A$^$E4`FzP>2w83TitqRt0jHoeR+j)WYDIS>lH!PttK
zwagYvoRu+$F_tk7thZtnuwJd$?)IdF6x4!^mwhtx=Iu3KeDw`l8P3i!v$-rpRaLWC
z9O3<dHio0+2}jFQ`p%QtmS#$Eb92pCufD=K!_iVU#te8Lga%Ql3e#vS=X_{}?nRBh
zGm5>d-QF83zW2qXsueNS{~V9}i*?>#cl=ei|AnGX{lb?MIx-A_^=8L*yQAxSd<aE=
zkOieHZ`u@>C)PV7e&`sx9mH6?$n1e<q?Bw_F$z*2d??C8@p)6ZxF~3S7}XsRyLU$K
z_s)^B-{hPRA?{O=BGsNeso(>{Fj7|)T1#AC@Ikn?Wka{tNt9!u8#Rrn|6@*Mt<c(G
zXB8@G!aQ+%c`F?5;88;sj#~=EaU}F3&5`CiCri%inxoqC)34uh(+!McAp1z?g8&cO
zFf#_112dKJ1_lGEtvP$~0qyAts%`n~@&bF@(1n391-fyhyV=la%fr)Ce&Yu};N<8O
zXAH3)`LX^XuYUd+e!Ih^z^Q>+L2WdRvn=NgZByU9UWn67Qc7GP<}KBHCS0^M5Y}%=
zL(j|^9zIy`?8zfej%PTdNPeX6JNjW@^pTh~+BMWwD+Ra`L`>{PglAS+5wJ~HvU9eG
z7N&~2P&tyEBe4`bcN<&A70=Kd0{wnZ@IizVuEe4x@q0{4>_hGV>$?$sIfG=4E@Rk<
zzIuP0FQeagt%9d0)ahKEta-{`B&8@eF3w^L=T3wZQ!&W_1)Rj5S+Y0Fbu|@|fBj`Y
z9p4VIE7h*d@&lDhI3=5MQkF<8E?!f((j?ahC~sB~Sr{cl{msP{uRi+%eX`_1+u)SJ
z<w(6>lQvf*A5gA{7E-`emJ|b$M_AsJEC!IKoCrCC%D%w-Nnq3@g;=-Q$wFfdF?5V6
z;&UM7NYm7`vl;XGoYl=WV~i~88AfHAR?x~}92vVkwzil`e7c6Qr>-1T<Aks|dXg)W
z{v-uV)zl~@W31XamM4#xA8*;*tT;XznKdm#KafV*aEFu-E##i2syLd>`MV#Sk>hjf
zR`Xex_|cnNKK<#I<CAARdh~!7&z^C1y5#!ylB?AX*XtENCNOfb`jkMxO!ff=4Lgkx
zdvoCJZp}<7Dy?aaqp}WiWOaMZSC_B($rnGtv~y6F?|t`seDu-BJUo9Wf|H!sbRBQ6
zuUOyQ;<r7$5A-1u&?K$J&^UPV7|7*(Fvel6LC1iKky>k7Yj6O54`D;vt%&<QIwoAs
zZ2Ez3HXB~`o=t|a6a`Nop7WjOAMn|q{3(yB1#RqUOo9+;n;8#IkMUy%r4x=xV$T$U
zBzC*KsIbgkF6BgmJrEeRLaT%(qOv6iNviPsE!&H)SstCE>be+sQx=P<JP`^5XsmOb
zJ$S%(zxO?2l#%Pz^;^OiqzEbvNh;M%j)G(ieqguXi`9+~(p9R75}|Nzlx*Ho+*#{L
zDYM(}*>_zjq^2>JEOFvzCM4s&>*%{)g5=w}*e%S6mBU#fwkzPlgL78vmE_K%SS;qG
z6j^UpWQ0?yjV6T5FboX+z<w`=%vYCJ>~=d*=oT)}JFGIrJ6W?J3!8b*T=}rrDQONu
z`t?;~|Mni=vWR=XQ2T2Oy7$JsvU9!n>w<oT5ZUkdblqNJ8bc&oSAI`~6Y}QC7N)ZJ
zH^y<m4?PMI>liJ#N=lJ<K*9wTNz=%y%+=)W{3T;p?ndyxOc8tc`*&#YRM$=CdrFCM
z91r#>RaMj5fzS(8C#F>V9`82qx|Rs8m?EPehyg0+Fxr6=0bjT8ij7Rd`vf@<gA}b+
z)zn9)oHY&}pE1S=Z*N&|d;BQEu^a<vgEE%NWUNM1v|3XgFL?g^G4<&gz1B4SmdB@a
zvKGJCuG@26zok-!M`sUt@!};XM<>+Al1EQ>`IhnWk}iyxF>+jMTBVrPmgTJB_;^9x
zG$KYpmZzkok<A*kb6^xH24d{-VUJUhWm9o>JmbMpi#3@P2jNQh1A~{nHya~)>qYF4
zNhFoA#uX*3I*h;%Cu1(fRiVImGEwI{@k9z_Ws2ZHivxy&6lDDx#}UX_>u`l?6Q?mJ
z>Yzpy(NAekltL8VchryRT%3wOrU`>m+|Py0rN%J!PKcrPL`It?G@z6|2q*6NdUxmb
zeO_IPQK0jZRwSB+UvX{!lH*^&?qq_yRUt%*8*?ad_!LZhcfG0Umkv)N#VD4VN)x>2
z`kObn;{~(%jD@Lah}eBc8Y2oAV-zJ$>ELpXn5>F}zl>tj<a=f&>9j7Scj4P<D5isf
z;63VI7iMcqG9ptp*5K-js%h~1uAp7AsTyZ-D&s?BzuPgdj>QXU7zs^~fN3?dA9_Mg
zR4r&{Y3ha;B<pmy5t!IH$N9sDY*%Zx+Z`uI#~5u0`7W^8Rf@W<sI?~3V^!pXM@v>|
zpx*TOaf`q5T(7_4a`QE_`I3i^9&`TmB@Z4xWxHN+dGVU7)rw8<hay{*J5Z%1$r*-}
zxXG|p;DRJ9)W%35T^pzyf;I9xVc^gH<d6A_AN~oC&L4AfcFNP2FF86p<#;jYsA*`F
zWKNp8Aw(rbcJ%muPZ;_l1+N%KPxPJ?M$)h+>^Au8TSB)+`4N*mCMA4IT=$;qZshIY
zxk{OhsR+hWg@I>{<~t9Mx%lj-EUXgYjL!5iay*~&^rMf^Ca~GuQro(y?KCMSnyL{c
zQ?YI_&QjOSA#GvGsZ>g#vy3pv!sD3Dk7%2gtPJa`8x|i$RHg5_z(ZVCakN*8oFdLv
zoE)7p_$S=lUb5Y6==VK3PYY4ur)qWRF=Gg<S8FcbUU0iwiQ>$AG74kFzpkn(>ZalN
z=$Pf=n7Wn)Ek^NVQdvBwO!1vmK;<k)^OmD|i?f>Od-`4j9$R95v;t=h%f+0kGCVwg
zK!}lX6pF!QE;^gdu*Fz(;`=Hk))IOc==*`!Z{G0c&08`>g{FQ<eezy=s^D%D$oFnD
zjYYWI`2R*9zwYAwPN6mxI)B~se{U1LhkD)J|Dm)a#mIiUW4GN2Js`w;zdgAhARN@9
zKvXFb#-6e3L@-e{GC-&g#anE$uSg{UVH&4V&<Qb;)rSIrTomeuXMO*LDveXlheCh)
zvC@+ntN;MxI8fI$)>-PR#yLmd$$At+qHUU@gq7k>)SDGP1`&S<#DLZo*VddqI>oJI
zY}<ET;Z-S2R0JPU@scJ+s^u|HPfsvSjWL>6Z*RH2-r|TDqNw|0z*ZWa6DDh%sd)70
z0nc7MC(jqW*=@<r^7O?s=F2(Cz<zzpMZ3gcI6r;F^OqkoZ(3?C&GPQz1GZm(!6xj`
zUEsmoG1H3U`J4x*CoGPS(6tjzwidp5;7(=QZ+CRt6~utc5ocjOn{m8caB@7S(uwG$
zaqGGc9}*!eNovqaxa?Y}92teu8E315*BjjVmWmyZaNH&X<*XG^bKqbCV(=^0y~Wxq
z5&S3-XdzNNR|Fca6if1armSVHbrFf=!k6bz%%@lit9wnxoyE*uEi<8cshC*9-I{e@
zFcD%X=%l(+#Fnuxc~ZSjg<ZO{b(xB~!)2*|VNLv1$G6Oi&38pXd4|)Sr9xhw1TGoK
zN*^Fk`C1)j%{pduR@BzfcY*a+ueqJgc~mttwS(IYO$O@}*+9|~0j9!8j#=i@Mib(l
zf#5{2Hd+^NJdHKAyl_4cvyd50fp8Uqz+?q#W)GaOX@gpKoyh&$M#6t{mhL77x?!LV
zk;*!TuE+NSP1BH4qTB6hmBLmLWyFuj2Z<)B>zXi(gcw*J9dUYg#_IZ(^=8A_*%_|9
zVzp72nRxU?KMK^GB6FR2(i)<N=H(g9(a4`&tXOq{eYfU|uSULkb;+aCGaj8DasKg=
z{m}8%%@wQ7n%!=Xk3nSCc_M6zMap1=N2(*(oT+j^kK*T+j4U|woG@eL?D&|R5^ZJV
z!rpJW^gXXGF8K1Z&#1Gat{SSk#TrN3wk+mzn%c_PB_(o_R1V2a-LPF<!nmWxQy~cK
zYEZ2E$Y$)>#K=0sUc+7?F61gRuMOXOe1^Zip-GuAcA^4RnX}V|Prmyh&8%j(TM<H_
zt=hu%5}&N95*wB@3`8&ak(H|saC!)#kgmnJQpJqZ658F=Ew!sqRn30z_+1BWF3~ed
z8k;OYwH)&O7&0jenssw^#m&_<(Ff|PVG5f@Te4CI-!Jb+_PZ_DS65tJUkiHdB@#%K
zUkYn2i^YN$AH1Ng=d?|W)}s7&)?ySGm9V-@Gl0r!S*TBrS+p~((~tvx6dQ)A6IEJK
zIZNf{<(ikmc&aP3R=BD<Y|ay}F6Atgh{^4~_(R2%33o$~UpU@h#uUFAk^7%PS@*AS
zykD&SRfl@--}~R6zL($q6AE_Q9ozMm-FEkm=;hAVs(3W!2q~abBo6~|=$Tiw_(ZCb
zFP<QzDZ(oq?mbVc&3B4AQ+P<lXKb>=DIBajJBYh6>||Ln@iC{slLKKLu`H^p9&3#F
zsZ}j|(;>=uPmI!lPA`D7b>WLfoYBmhswjF##vEwoEv8=34mJBt#eTOZcrQ(lAL;yn
z>U-+#nj_zF_V6jc_q~sJ`Ebh@U%uj-uP-oK%<Yl^RYBt%*&2M)@aW|;01`p%z7~%k
zvNujVMA6jqBV1L}Z?_Dqk%teSuw0(-@X=GAy!eo|s;~<5II{oX32z^rGS40L=y*1F
zEUe}D=$O;fhoq{;>r5!oWkVbpyOCkPC+s_xw&vN%F@%602S(+nD@W7TxY{rd9sBK;
zaUAd=pbAo^Ca!gqKzofKT|SP*Y9=52D$~7DxHQkI4E7XQjjBZ4VoR-2xk@;RF%!C;
zZnI_RdqVIub<KRfkQzX7XQEzWDYeq?STEfNoN`X2a1TL7>MuPNsrN3GiHGO?ctH3j
z)NOh$G0I0XYpL6sx{-#$I(OihPN_fguiW=E87=q3R8dXWz^`+BJH)Oud55LpFk+I6
zGpC8z%wZHOBKYZnP7hAz?6&xH<OIXziW~PC<I#fZ?3mU^DyQW38gwPzbUJICG7#hf
zlz?9$CCVG2P*yg^luGb=&M4)CzmaA1C?=~}OfyY7!2Z$|h?!;8HmonNXslu0%!t-9
z76ivcLmWn2+n_S|ab!N56~}3hvl5Nc?>kafI9rip!Wu&acKbbT-SF_yBZhs)YQ5t4
z<e1rfE@Bgv>G~q%P@3A+s07vrp3D}swxYSdV&^><WM;MIx{nOW6O-nfS6}h=HJr{{
z7RO6oK7CGl`i$*n%gySRo7-DPKMEOCi_v2+ve~CH_fHvelrGo+DvO!s(pjuFoSvOB
zYiGE&#Sa~?zIel?+f!Rji$<kD+N|ijr_UKbZ&@rCB|^sFgJQeiGZY1r_dV8RboPXy
zr|%N0+_0u%%oSdHqLFUODX4Pb#K8wAXXJfkvtCJ7Vi>4_$0v^8{^XqFYQ%5W^!pW%
zsGDMu69TqGY*}j$nUdKDa=gp&><cd}CGkYeqvz=Km{|!&H%6mk#3oPNZPBNmq{?D8
zwS$y@%SG#lf!$`u)x}%B`22IOE-pwR;GCnbYn*GaRf9GTZ7oV=#xc-$LkX^)kc>?5
zqma4_RPMd!!GrUnzEuY!OkEHMtujVQ1YgdQ#XFz194!|tXKh(@O2d|tjJKvjH5S#d
z5~X7<X%ip{>{QT?5CUD_vl~Ztea~vO;^yXttE($kn~ivay|cLdV!?6u^IxxI%lH2O
zH}<#(q#eHgwT`muytiS!`}==g$A35CR7hN8-}mhHJGMKCBa6#LL9gHc+leram=p+M
zB>I7-S==cAGbmLyu>(1rg5JeDGuWxHv*v)R-$&=}qZX-*auWh5#1ygF9FUgjm?lfE
ztlz;0QBu|tBWOy~-VXz_aXfIKoOLKwtdRUb%8_>FP+Ey4Oh3?vfHf8M%wVbo?Xo3?
zLcaH&!SC7kfnDFxy?R4GCgw+HoV5+jhcBp%Vzb>4l1E#Gaa!UFmBt&#?CcaZYY9o=
zwINx@H?J=xU1%I>GPH|1$0rYX@Z>4;lT+HtVRC|g&ulqowwPgu1xqT<D#fy{aI-n%
z+!8B^)l9(?#({CS!*4g(oSD}xr}H^FXS@<#o3%CC84@rIBfi@a#(|s@)@qE^WR=lY
zre{9IgSaKe$T)b&dno6EAhLHwlu`zxMBb(;pbOL3Us6hP81W&A@^!ak*muH*tZUkK
zb}$5vp%~t4$=Nm5yz`juiW#X`COe&6(5XW!G))E6)Zl7Krik${+P{B(#$vK>jG<B$
zv-ymsnNd|W#+m{TQl@5-8$KbH)8C(fr#or^hvwwhIli@zUC#M#bfUO>F(_|dhI~kF
zQwJn3W{H#$m@*+!4AEKZ5UCaH-&~MaH&k;=tumD{Bqvox2)>{t3GYW#1XG!VU?M7n
zz*tE|+z*2g6>|YyQ>3bE#+VrTUS3zNh&hP|TjBBf5DJ|tQ8x{?s_2IiXC$Z97s<TQ
znxX4)*3r0%^=8A-a*4`NRbqcqH7!+D5&VeCil(X=dvRB9nucK*XsU|GPoD7Q7hiCF
zb8`qA9{h-?egz4-*{=y9Vw}S^6$>mM9iK6qjZ{hTr$4)-Cb2h>b?ouhFvh^uZs2lr
zL-qPKr>AF}JvigxC!c_V)oR7tt4nTID|Uk{T)Fs<D<#F9Dg34A6V4i9O!z$V>h)_j
zs~g(3#izvL=$LvwlL)K`ORZ_NL4|<J;HC>yWkL%0FfatquJ5?rb$DgP&#9025aeVB
z_Bk=CfYdTlT)3P$b(#m(aONuPZeX+PsY#>|IJWTd!<z3sKjJ~_(fyWj=<z<%&gX~t
zn&5-vn5(<uE{wkLexig^TU)F_q(LA=(h%9NcZ?x0KT*uXK;vf2QY39}F>#MFikyq-
zc9w<SXo<fgXZ+Z+Uf=TO)oWgV{S9wkea+bKihWFjwKd8%xT>Y9t3tXpjAMjkI6Zp6
z7)Ey8Mi!v7r|$>hMvY!ZXuB=5*^HzUei$X{Dd$pnMd9XhN0*p7=B!GrP8ok>FIV^>
zG(Jw&6jIEEATd6ZZB5s8?7NPB=vnVNw)-91?UueD2;t761n!p+zpCg^_ip>Ej=$kW
z-E{o_>}lj*_{#f8U&d_XI1)p`XvwclrBN0;pH%1+J!u#qM5?BQotM8h<!esJw-k6n
zRu{_(WFs^B9vNNTn>5S)%EK%uMvBGeNjAq{`r73}A1aM|jFFIn7<AtokBvSOQaT_*
zD2O>ShCzUUm~hTYqZK4BDy2XwJVMDptx1V;J&e>;*!@V?dG@=J?aeI)Lpxt`G;4YK
z_zYV?=LeF`)UHJHrA$;BJ8Owr;e*gojByP8NJtqS1I}8kYp7=nX3Jx$W=7>SIt5ZR
z7-w*;L(N-S?WnB6H4a@@1gr2WqO;`l>J+gtQBTcYUC~xn1W!4VRmR%tz+H?X5xf-F
zTB}0a(4<nFYb6D6^pgA_q5z?kFr61#I+;W9q$kG>F)BWIrwP}BoFuX^je$6N#=a+|
zNL|&EW>LwcC8R8($(&Kj6uT3N+auqNP7nP_85>MgfkWMQkFPl$3TYxj@CS5nvRs4W
zwt$!++e$3YY~@hpJWFM|a!~f)Lz}0^HSy{Wygseq-&kRH@3?!y60h-YlUWiZ?u}rl
zcSnjsL8Ow>Xk?@n)5yb<Bc4AvX6~R;!tH@1J}!PB4IY&YIw{DC<V#VhiWt^rtg}$e
zSMwxI)|zZem8><cu2CisvJhmG5d+P`SoR)AV>X}Dt=DvYM^)EQl(yuIL6bw20@MrX
zT~HB4@DfAi>I!W{WjPK5<1msbj}!32NIPr!;DZmixVXSL%kub`oHJj1{v|h88@g^J
zM@g74R^w{J@_5eTc*a@d_{ZOQ!L#L@KmO?}K7G5V(w4o^>`)|Y2qA$VxqAH#*OzZt
z9vyM=;Ea>g6COYQ4cUOVJKo-0bGurz-Ry|b-)sJ}$kTn0lU*yEm7>1udVETZF_YIT
zc0*4XB1@%N8qHiQ8l`BBWFBTyyqHwxhRD^fXFEoA9=tY;8L~1YV<1OrJarz4N#QiK
zwPoq%xPD}oEcIwGInr81MdHa(!zYjCJU?|TO+foK_yK&v=!&*&h07R1fx@Nenh4d#
zPzpOy5f^$7-j7(MlXAvUyN0MW*?_f*Mnha(lU8qO!wZa4L|WqH;;upLR%E0;3>}-z
zEvwsW)~gl$en%Mhq!8%#hMX;9RAeNIsj3_5dWN$tSy@>Wl%cjQV>RG?AdV7+6i3hP
z^)=tPn%!oH(S~l{bMfYa^?J*$8}acj+Nt*=7jG{)S{$)kZwUl^js#zniy@UURry}<
zQqV<-VUpVj>N<Kc1NWmZiuC*5WcM+l`^W!OmCXN43cGK0<X`^zFMTQAo3sAH_Y;%p
zU{+J2O{DKf_Pae}F9M1*Wo;s(B%?$ykV4P6S`qdgCT40Q;xd`!%O^$&-!_%=B?dvy
zaunK%QW~R0^rN*on7K}4u6IX99L85dE4dq6l~zQWk(@<@RDcGb0%HuK3O5Q{IgEAq
zPzt-E;>|fr&h9uc`jHe9N?Dw(k*wCe_`s!1Bo;wN1t?i_jb=GN<>(}_-*#MH+_1h`
zv(1SddXDS^o<2H3*O_hKGel2S)r|c}7(HVak=Ga_{WvlX1C6WDsqC&<6LP{dHT7&k
zJzHXJjd3C-OgfW@nA*a;#+Vw6CRdiIH6cg*;DytdpjH`YExB!RMvKp9&WwF8BgdR*
z$|P(`l2aIR7D|hAm29z7J(gwf_MtzVVp6Ax5NaBaq9)rAp|E77p?nw(S;fNP%!EP2
zi(&LcABa9lz(8Hov@?vc_~7w{mz|Wt8H2JG<D7_8RAh2tRER0CN*NW!n3!Bt-yj};
z>fV)8_<$kENLhp)LB`23kh3fPof%zenx?_l4H&uaghpo$McriK#G$a0;#LY@GhIBw
z&9?t0N3qk(q0~Wbln+0C_w(u2`S4Nr5XE&o6*OpixQ`#6v7A|Q?#Z<iZzDfI3Z#BO
zg-FhZtSm|yd^ZZRguM8saw4usGCs^PiL$43YmiY+x|ryd{N0?%u0)z0lrK^Y_Crr9
zzFOPumX<8$igE)fI76hWDv7A_La;r!;uo%mwU&^|tY3)(HwJ8m>?JB{v0N4oQa+GG
zx9c@yKhW(5`XP~!5m+-*Y`Puutz~(-#56S@EE~@MosW3_*=zpvvlSQPNLx9s`#@c{
zgy_K-d<blAuUM^DeEqedozFQsKH}xemwfR2DZAaC7$euWw|w(j>`c0Sk2PYMqO2%7
zCp9S<oON`g$CJof;~eB6GDJaQjnSfdHCo2xDicxc6>LLh$ckaYZVIVZ#Vo<#WNHJm
z#t5&%C+N1E7{$ZdvM^Am$f8Po@c5MH4_f9nU{i<hJ2_`kCT7iSJ{OZ>qlqC3SEHye
zb@ApRYy3D$;56zGC)4jcYGaAs%X?ZiRLdIMRt%x1Qd{)c@%F1vS$zLPs3}>tni!?g
zNCmZ%u~fua8TVMNfQTZYbw(G&Ol3vNkr)g?MXb&^V{xt(`-mtu3r0C=TQTaP*o#Eo
zUR<$RZw3B0mKYM7?T*b58H=yvH00>Vz?WaYqT6*ct_mZb!Vw9vpmG86%q)Bmrt@^_
zHt!5X57Wd0f8`gJkiU9y`P<`L9>4BlKNUQ`3>}?*HqBn-772w<nQj=_ZTAfOo;U`w
ztIK*C3o+f3!@#h)<;`cGVfGuAO4AxkWhDtf&X1CBT0HR5C_Y~@f`thdk;DO8=mSOM
zR7k2quun`ux;d8&>I0f5#PfTbRr#L=BljOtVhjUA>T$ZEnayc>k1~#~^UT@>O;bZo
z?7Kaw_@bJsLTQImt{_qs%jJ^bW0^20DqG3B5(m1j=lW*L(fo*6yWsoZJ!8AMWxZb0
z4?Es|v*q~7L(Wf*IWtFeV_?_yWMxUQW4qb$<yYVEuwFm{<s&9Y0fEXGqp9XK^W!5H
zN5?E?M>uQHX&{X~zFYJ9)#qaPUuYJ29N7$>(CrBQmXo6eM~gX?Ht3Yu1&@yzt20(*
zZf|bb?Y5v0gW{B_wWTsb$3csugfk6I)w1ih_~?nb&{w=f%~>PrSZ$>M)YeE&>x4*^
zx5-$GQIp?YF4i%g>_?Ix3B$n9cOs-wn#$EImJ8~7hINjZGGj>iRN~Q$fy$vO2i8K)
zXzdQXe3TMyX988kmeYwN1_tWz7^lOR6d$c%d9=j)QCvQ9V)Q*@7(hd19re6LR~Ds>
zM4c%kpWow&s$9m-a2U@jEirp}LacvzNBmbF-&#CzKZIIMW2`*=e7cAx@csUynJUjL
za%!X)ODNEq*<!|v7f-QPT(6@pFnjQr5YS^NW_@Crl8r13#pwPJemrW8sUo(n<ibyi
zpmQ-n<N&)XmC;$Fj1r$7W26{FI2p2WMj5GAj$z+{67d5;X7r)(D5@j`+nh4iSe&u+
zUR<wrN@$A3L{az{qwz$#uEQ!#-P8o{Y1>x3cC1*X94%V1kA#?+AI({woN;=7%53iN
zVT&>XkFU3DcDEO}%JSP!Th14bKmTgQSG#~shVAGN{1w;K_!x+Yx$>^-xW2jOvoF5l
z^!S*^PoD7P>0?eFobux17i_j$uCH(Cx;@>#qXFuy@S~tI3Wau#RX?)#K@i2lk!Y2o
zaSlhqk+DSx;A0}B%+;9K_)N$$rqN_v67_Cnl(?HWOF{(S_ayIW6MRr9K00YQuQPKK
zc(iPJblhM~K<5o63HLIjL{yF;iAnWhb~lDK+AyEZ#lik?zGh4zZcj$3)=Fx|&U<vs
zMa3(~6~Uvj!&;41nX`GrK1E`G!_1v=vD(wjma>?r;s&1rT4|a_6mP%z8$V#y7=Hd~
z!<V1`jLmvY%n_6neNNX{H=}Om%xA~U=BHGy#W+XT@6bD0bLz^K24qu;yFl0V><j;d
zTt-b<BU*D)77Mv<cKt~I`U2UmImi?yF4f(da&MJ+_riZ~t$6qQ?^{;<!ttvnv46#V
z{O$2??#KmI5DO3S9~=e|C8a1d1<D>Uo!8^gvtO^6Cy$%87$Xhv^yZ3}<)jLnP>XyR
zMX5vMYf2$tIN0*sYZ|1Pm>%cd26+!v;C_sh>{a4zH;H=0nxn_f8?19QP0OrZ&~;*I
zqIE^(45=4|Y%YGAW>S2Y`>@6qlvCDeS2?j8sw>cjtLqJ$?V8Q%7H2if<(%^eXT+S?
z4+HytPtpooH#BukFoxjZWVysGmTYftncc1!v?2LHtYfCfVl=g@XzGgEX*7Y@5BP3H
z>Noh!4K0b&wt=C??{<uRfAD>)t>t7nqp51r=o$7qQu25EUrwynYcY(r8l57Ol9Ym?
z%Ht49)G&(Wi4R_U<8sDoEAP9r;@784`Miv6<qeXzO(`klLEJ1-B>Ry#41}>KjPiMm
z0j=flwevZ3LF#fAu|&+0%c_);Lb7sVxuSJ3oWD!ExI2f^yXO#x)Ak*+YlTd2lE|hR
z{eZKMx~@=KGmImNavoHbsBn!GbCVi2-9K;3^Pk4TQ*Fg004?tY3T!mL_VxR%3cEzc
zmJQmfd%?Ilp>mYGHpE6l{ADCi>`p$Rut^diw30l^s1oRr`Ml<6;ZS)*rA+Eaa){&t
zxh5}`B{4<N5?=!mr3x9bWO~NY%a~M2=a;o;=`<xF!)RiTq~cs|oxxan5f4U}$T&_b
zT@;eyoIdP(W=)IlM_l1%crPAERvXe7P>Y6a9kCyXqnAREbZ~1=V<!2Cj{`AyTv1;d
zYf1!Fl)@`UKKa2X?6w{K7{S!2$rs2$m0P^;>3v@k6XuX2`+ZMXZTR@4Wo|VulH)d7
zzSw%+ZZ_m>ggfS(Tv$nLFyc6H^X7uLZ!Y-ZUpSU$3(g-u;^Fx@FJHc(F%>$9bbK6S
zQ5*+Ph>4is_;Fz04+Z53*c@TlGZGneQLq9T%@{L2g2Pj(0Z7bdGmO=kn6NQYB_KuK
z?i8CbFwcQ=;Cpk$?>?RL8&A%7dNjx7O!gho??`DR#*x7ff~YD(&T=u-b<3ijQJb0+
z5~g&HDJF7Aq!bTble{h&qKImo7Im_Zfs{sUrCA)!F|EboNc%vidXlS{IYa;E8%|D7
z*xd%K{eT!P#?+{6O7^3mbtfmsRQ6+z7IW&V=62n4aoyoVEEWw4V?jAEm8G4}SROB_
z>$Xr25=qT5E)${!EuFF+#M4fqg&?Lxj9C!VB&stx&nL7;kh{oz>_(2ist%JM{)(Te
zyW8A%X#d;eZ_bhPon?bq>qxVjCO(nAlgLSyvhbgj&RC-lBdnyRGpy7=_t3eN?_&Ez
zlpzI}DFV3ZH}B4$Lq*4=icRkocKO~DIJl9Of>A4VxE}mCkY)*`49?ZG?Sjo}Pu~aH
zCLMz6m4cMSswHJ<_(W7x#(8DR6N?~38$)Fbi-jUY#ReGquH;c0Nx#ubkop)2eJ_bC
zex%zE?7M{57TdJ+!yrit#*t$Z%PJ#N4_hpboK=uK{q~m8?-|#Z#NCRxy=A<;MRj`^
z2XY7!6K0KwvQoxrE04jC_)(HubeV3&BwR#o%3jaiwVYxk6A#K(5IY;D*C>h+yS5r#
z-b?XtGl##Yv@UxB_lm4kY)1kqc;Yye!fqs#aky4m*6*qo@tPHPg;==EDW${IP2#}p
zp*YLjHJSGt`V0r2+dC`OgzDXEJX0!@mok~l2o+2f{-)6u>nhwmUfqF{pzbD@hrfUL
zygVkY?$TDu=Thta8{*hy6|qXw7>z5D7=l2`xKt$}JqqVhwhC2<wZgd5xTLA0BV))$
zqfEe<foD(7XzYlN;QJ9Br8`hD9VE+BRE$z3zf*JQ45YgZ<g7^{6S^ebm2+t8h+`lb
zg))kW#Dp1Vag}rwX>vv{|B=OHbze^6##LP3uE|FyD3pXjqfpuqe57#=(Fc^W;3||h
z#8H$+N+*<cWeg{T<H{Q0XcdpX(R-?<rnVJWi6Dz8s%48!sSw_KtQF2xZEHraQG>=+
zfnHlt<JOk56T@n?;dz}&d1S9NbJwy|p0Bq(y@Ig>!dF^TS&g#L=_!ypvcFyN^XnBq
z&!18m%e-!xH*-{Fsb&p}qa&6_$5?TW=3%?Uj{%ax$B1n#ihxp{y3!2&4&RUD;_<Xd
zijRh#O}}R|bo3=#uMd&j1yl(l9hIh0upsmOGs{2zu;m92mYh0`%YmUw#rC5s3TZ_M
z=0L<XI<sgTNAsG-R9Fl?1)93WT8Gb`q*75~A?1zMBAzH6Y3(YkRt(*ioIR09KWwR!
z!-W}|24<FsCJa69=8F94OWJ0EGb1V2<Xl0Fm>dA%wK!)vIX+<+JjNZd?hJ1?Fa}k|
z7anUOb{OemAfWLI+D7a~hOws~cEsdCC$t0Y6eel1a`-WpYeNh<r-_tU=lo*!V1~cO
zP-ME*uTiAkAHV)OMcv=legD=SQqUxP2n>EKDtb@Kfs7TuZYG<CC}=cUqLZ90=aa1a
z6T&OpP!7|P6ib>wN+hi%+R9G(tp+I?%lUfP{pFlW<@>o<MyLs12FQ1m7I`jr>hPFE
znL8nnR*Rwca(TixU%p|x>9NL9Hx+f=N@7nK#X@PyFpg1BpK^~WXW@2vUr;+o(^wwX
zE%RB+Zo6S~dqdyv8T!2#Fk6d392&;1qw}84ZeYDlBwJCRp5W>k{d!HPnzA0rM8{}_
zH40}9DtU(OE!X=Eetk`U^@gyyA#QHzuHN91L{wU1I6gh4vJRC41Op*4_z~at_;IAR
z71>B(<%2)$k(8nvkfJLi9J-WoZmI*MNS60I74DtE7$?P*(HJNHPIe#imflt5F$Zxa
z7)PS__;HZpu5i|s0^@3`rloFMs-^{His5xB>~e;};j?zih@GP64$ELUP4X{%+@Da~
z%^#<U1CS(BSSwLXi%GnP(lk_cgRANTr*A6In2<Pf*=+80zlZa5DgwEih?a}--<Xe`
zQJKnuGhj;>Ha*#t??1fvx~u6ymj@V9LMf@dGl`~leE7k0Dx>jz2mOf6xuh`(a-#B7
z^s1tkQNmeL3Zuo-CZ$Xa5#@?!P6{QWMju@60U?wci!%}cpO7kis@PMhl1)Lj#=d7>
zH+Vm?S+7|v7W93880)2paBFRm){9uek0YixIIA&MRJB=AVyd8R8vHOGl#)3n#x$am
zm?RtHn1bk)9|?WWei&(#SiUH2sp}avV2$vrjOo$F(Y6gaXR4}Zwmjvd&B%|w%A8zf
zes()@<t5o-A0RiaERc=EDMOnTRta&_1G^Zw+HUw}vlgT>LjbC}##+mKv0yP<P}!RK
zY=+hj?JbH#P6Hd2NIBqRq#qMQ2#h|k+upF>?up}w3PIlI=}d9LOPKkNn#A9IzTiLj
z{u91`T5*&-sqfhBckK6`5JplGZzNj`F~!`}P*slQa>-(`L?H`+F_yZn@5Db7B2YM=
z6G=OzOr>?nMs@5q!f&aZ!`2Q;=h)6>sH&nHdWezr+e=1Q6JNX_jy<RuDMwV2ga@rA
zV@$;uJ?q_`*O!-k_3E``NdhWHv@eTLr&#ayT;FbJnm5!{Q!E-HDF?L9C><I7o?#e>
zAs4|zDq}K8spK}RJ1djFJ^uFi>y8Y>$-^uTq9)CW14~m0iXX#Bx7{9$tevwmL7m1~
z(;qa&j;c^&ru7Z(cy<%FPAh%KDKE(HWb!@H0p#b2RI;lJ^ioE#6B;A?kQ9xU5Roz8
z*wb}Av!=q>isPeGtZmqBd%5;2$Jv7gRo&9G4Uiazp-gF{1)A);CY!6>ZijWYSf*H^
zNi-H~7R=k~AVwJdDDi;(fQr({8t<u6VrIb;iMtIqpL>q7r8+xjwm3r7vg;b7a7NQ_
zw|x5Ne~MA!J=&(g=GB+P%`K+e@o3R-a<pXL)U>su(h$5%i?a8^y)M%U4f3^;+~vOS
zsG3TqNjVk%n#WQKa3dzcuB4H~6!1mBrIp52P7;@_739h}OjR9n!8r(CCbTS(yeAgg
zP4pwaWUf*Qck3Lp#ge+6;i?90YhqFB#~6vl1!GDsF~%}+`0_8SP^R-`;tt<$Le;xk
z0?OVHV+NH_*3h<dp;q}Ip1LRzOgbw?oD#%(LUzsFy!GA#kTdnb`BM{0Hx+_+<hEb;
z_*R9T)sf0(oEF}gqKs2>ey_0m<;Mie2SnD5%H-rZKAJPDE5@!vk0T}tzph}DCE)jd
z=<n3-mGW7syP`lPlodV)6J*@?U<}zPqDo{7$_R|AVk9cisyI?7DL9p}paWG^iO-Oc
zG=h{QL1Gd}+z%^PN|kN_CZ$h`v3T_e>C%?Ckx0ZkCrYz&8?80odqIzCQC7(X(qO6v
z+ss5lZ5<(ax~`MrUqNka+Nx#V%;>s3tJN*r-HteBS_XdWyyoTE$Nc4+8y<dh$rmfn
zn@!-h&uCSVa-v#GPHj{QZ!*C^){4qNZ5#?s9|KrHTZ77sW5;mQ^Y-oQ(#0wa7M*KR
zXU70pcnuNo3OtIW5(ZBdGP49NNIV@9b=IlKJ_a7C%)kGgWB%c9e$3OEV>WbzT~D{&
z(2YGqiPdsie5z6s#Gq|wv~3GORn_G^(9~_q5bj0}(lC~8O_g2b9?uSBLvhX>-shMz
z{Www=6<Ev(LJ(m?jNGhOg!%^a<OyjUP*osB38hXs7Ia^ea%9){eEs?rfAQnL<l^?0
zs34XL>9BpJOdcZJVT)RC51q3D(1o99riH0AS@-zk2VN2rck0(ySN-kr|G$UQ!o^Bi
z#<4#5lJT4mNK4KoyE6;Q<Fv*pjlGX>6i)SolH3nak4XxF3AmX?y)sHuC7C3XwK}NF
zQ=X8FyZd-|<Gw@frp<U-pQrSalq6m;lniETG*#WOTpV$8bHjGM=h_;YM$_hssuJYN
z+9Gf(K9`5&qf(%b!$1fDmlDcM#bqKEf4|B)nx>&?We*vBV2qy82l{bjv+ddJJi9Tn
z@c~0%UR5lX3*6Z$yMjI`GCD@KV~<S%lLI;oT)p~+yuP8$k<-TV{AfYz3?>KEC|o(G
zWt8hvK!sBDq(Y&oEIDUFl*f~ElBhm4jbYQF0oGcJy0z?^DdqBdfVH;xwpGG4vkqH$
zZe?4SJu<-*Ar@nlk?YWtf+qzzH$X%c)>X{f8SP?;bs}Vufl)e4?xy{B3LDVo7kOz1
z&fMWNnXGckS@iyd;n0=H{@|=-;wF3V@m>tr6JXq3A$cg)%6%#=F>U#d0dr^HQ(jX!
z=e0B|S=7xF4xPhaKTnyCZ->|!#6XrL@?bf=NGM{~9Mny};z)>}UDA?~H*X!k@x4!I
zD?{3i;5{lSa+<tuRN088`E)u|+1gA-nMsNWtxGPoTu4zq#iEK(BNKAOY6mGD{AMKF
zS!0wryr~m!VlqDN`kuCJ@Lv3j0QUR6<k1zO#gq|ufPiBVeqabF=VXIc(&fb*$vKHH
zSQ2l&aU7}szwG_jlWbX*-iiITI>OyU+=wevS7w&(u5NTg+i(CI%orYsA>U{u$Hs>I
zr(i}tkw!Br&`6q5Kj48JU<SfkR0HU)>aMP;tW;Np@NjX&R(x3d9QT`*IbDgKo`%ib
zT@@Mk#`SP<&K7HZ>s#Mf*R)j)F%x3~XQWF=N!cQVMAbMdZ3TrXg1utWu8TuL<ox`M
zrmeWUTeG=&$@%34k5A4x`Q|%ZTwL-me)Kb*oLA`foKK(KGVDj1p=KBd?$XExPnRM&
zg3F3J$z<u3Vn~^Oh#0G>Yeygvt+aPrtwgjoWpo0@jEPEQ0kYi5WaRp`NYJ>fD^BW~
zN^5rgp6xKA#?0eW!+-Rx2mI~#pU@`Fcy-5q*U@cm>Bc?DNKl>=R9ZHZ#bN<~bB@!~
z)46m4aC&+Q#xQLCY%o5XHJ8&N<%Cvp{{pg6Xr-uJB~W_|ELKYvrz<g(HkyP;qqMHZ
z$3&wnoAsLWIHJQqwt<iW(y(yYG-AwLuh)F^*{A&U7r(%#_)CG^JdRNNoYTD6Ws>LF
z0~mF*aqtyS@Or!+zrv91yd*QFnP}>Sh5|a2zfTEBWrdGC738L*iF`okW;?NTD4!{c
zwNxpY9bRD0ljj*ci)+2IKnmw>mOhz|*28~<XLqRFh<uNTX&WVivWvxnyY+^?8@O9<
zS)Np+)RNi?aC5S!OUd}S3~Y@9#Vi4o%V$VYxQC-3aaKA*ZQ&7CH4Ni`)q`A<wk8G$
zDbx8t_8!0A({H!<s$y@&aY+LvCSpHe@(9_ZLPzd4)H$#;inGdcQdtaXMpaBsgpkM~
z%*`r3gkp;&6_0TgE@ey+NaPS_j-qfYH3$E<`SQ=4N2NrE&{=6D(RSk1O*gUJom?6u
zQz&+ykr+HN1agpblU#&;tD35AXch}xRim^8rAts(^fk-G(S`r7CNJtQ@Ys&$%<-RQ
z36kRnAJ0LGWUxMHprn-O`)<}mG|mBfp2JEV_=(e8oEkgi-{qRjz}@s>a~58%RT0kr
zRma!k*cp)uNu!sNoTlhh$@!AzH$7!VhfAcD5`kF+EsHVII712@IS$-j?P+zugd`m#
z^A*l>88w&R6@k=9ibZ!PiAZq{0oOJ~XD8!xk~U^#2_r<6FqNG*iRitcG)c`mcoF~#
zzt32Mwgy+%blV*nMb~w-ZOh5Y3C_8r&MhI$h7_fQYmH<c3E~(qbtx4@xh~ok`4y#T
z+m^~WHk%E8kj3w+n&hJpLzS8{lBxU)QmURd6NR+<4}C}F9I8kWoSmNH_Z`=FJ5Fsw
zdvC?J-gwAK+wyyV_$S=;FZs@|y+=1hp5NT@;`s~Chm6MJ$59B)IpfE`5E5Q%x}4Z+
zSbI+(bMoMVlk-zT9J#%@=Gl5juM~~dEE~&{v->QS!Ed+Zu}3FQV+{AIGfr1aR8<Mr
zZnq^8(F(qG|AfEs8z1l+Z>-3-FWFpe=<i1QT~F8V$p+lYQMH=3u31!4TD`oy4`kNs
zJF2Rl?W>w*!E(9cdVNET;ZP1JnNmurXdJ0oJ8OdfmDYtfW7usyAw;^}9)-f4I#D$*
zGCR&#R&~qO`U=0h!=3KI4JcJhc_|53B_s*budZ+T>CZpn<@K!uz2!a0%>QZ_JgYDH
z+5Dg5@x|TZ^>{sgwULSr=tO9bNlLJZ`P7nc$f2iFrgDy|aKOYyswCnis1xlmLD?r`
z_YeXK35b-DM4CJuq=KM-adN9)`h$!K@iR>5`}DYihD_xBpolR^7iTqMw8t99V!5EI
z9Nj*1x7kv?6hHHE)evKaHWVvDEmnl(;5FLNwqjKnLuBZCVoGSE@P5Dt;SlGhoZzL@
zRyj)uf{eu=O?j2oR7PP{Cha?JU%cRU^n}|R_LW0Bi&dHmMe75K0hLEg?pc_~NozSZ
znv=><BRXrsF|?AL58|n+vLK={cv1{#6wX?@BJm<+b}<JK<9gxOVGI+8E+vX}=#&^|
zw2_WgTfQi?li5XLE`N&ZMJU1C<l~+TT9=C+ZB>azUEQ?Q?E-BalcQBE%`6i@GVRy1
z24ZPim}ibs6HGLhHxK-}DZ$U>TzGY!DK(+ZB|ae&b7XV3Lt7&p`E~-J$}}#OxG9i5
zBrPSVo8MSQgyQqzJ<LI!1me+8R(F5)_<CKj*3qYgC1Y~JOm9fSIi?<@Up_wap>|w2
zaw^L2Ta_^+QW~h*O0;ca7X2Bm3%Vt>r3|Q~4qk5xB4STdX>kb<*jqSYkt|BrBy5oo
ziN&beX?dQ|_avn-t}Z1lIiQWvU_~f%v0TvY_xKRzlBsmN!e+CX7kLZ;k_c1k(kMYH
zn@x-;UAl9OSf!cr9(;&6V^}Q~>~=fW*SDPCyGP?1;coOjzHF$@2@I@rrq(6MQ%bC&
zjD{4&(nlF`%$%*x*y)P>zGv9%IWdmq3A}T;;^E);HGcHRf67Nc{~@+r@~wB@<GUYR
zvVD0&e|txxEvl5bLke8qu37I!?uN*fPpF`{|Ku@$?|<+gv%I{=)$KL^@?ZXQ#-IF{
z#iL73S4)<r=8d!aeE-dN*?sy6KmXzHu?Pc86n0GXn=McGJJu<%37KVO_^aQ1%-{XN
z_c<YP_0ylw@9r4-NWY78{YceRy!q}MR7=CIzoTs{&Q8wgyAERvbzQUBY}stK<eXV9
zmn;@b#&Kl3*-Dwd3}-FDXtIx`qv_SNBBfL^#!CESLAsxco?YMfjNYTPl!1-bESd(p
zhP#(fX)oVIF9J$MG<HS@Q;hUo&&|y(PoF-g>q80F)R)I!M;pSI{+y$Ke+8V&U-7`#
z<MsH}MlJ`o9HbMvG2v!PB!fx}RU1cDI?08JA#MYyaM1iP%-pFMWATucQdo!)n=G$1
z^&B6ze8qHE6H!NzkN{N0gi!)AOb0Sk$E~s`G$|q6!J)@!htZbPvzC+7mcHN7-*(*H
z?(m~$x2qWXhIVOZgLz{u&dPI1kUKdpmA76{q?~hcrl>@g%X?x<<O<Ll9|OL0r5a;}
zwvmGPhWmk==by8`y`^tk#?uvxlcgv}><8XEJ*UwgtwtOJbq=hIVrevuQ7D4wfm(~L
zYa*FAj^vEEDTO#UCL}U<cX#aiopd)(2uQIKOhD9ulr5!>Smwv%x&T&NOf7=J&Q+x&
z^&m7Y(Cp+DUYcgS7mscqNPZ-F@e{V%P_+x%#R^;37*~O^qO(!Z*%X8D1xrwek_HW3
zEJ6=SGOx@lHKTEpwcX*yP1>c(7?cq)Zn7ZKF-R#sjw8WKxA@bi*PLIR;+jSpHOq6L
zjl@Nz)H=;qU|ugX)h<cNK7)uPPt9%RIz{gp_x-QDX1`WoSD#<f-QKV=VpV^z9y+Q^
zPf_jkrhlniEt#DLa`Y^gmYh6(>~Z&&w5wJ;Cv~E#tC`C-soPJa$f^>4lOG05f`zT<
zw|im?g02`%(n^BEm`fYJ!c|s8i;WVbE=ad+^nMPKloqnNMr@{xbGWKvb8{!6jAa;O
zB<qZkrfFnDDZj7u#0C@zmf!=X5><6$4KYi08>@<O-;q%)niiup>&+dv*Eig|xWw57
z7+u7gM|OU|cYAacwQN^aC?jqd4l#PKoWmMNx9hN`1|84~L-Zqen>GFZj`Ooq+P3A}
z?>?lpk=yl#boDs_&4YXAtls_x0%B@x1nU!F-}CwBPx<uvit{1T^za+M^#gwEH+}<s
za>91#uyw=58;^PP&O;tNd6SeBfABAVkM_Yme(=qAv3JjT`G^08Dn?RJ1fOU%bRlzg
z*7DbX@BzQ^jkgIep7HdvPuOkNj4|SULL0~BlY2aR{E(_~bi)>v6HQfdetO0on*DwU
zNCV4yeaG3^87C*F)OF46&6=+3IJ<W~Bdt^0eac4$=ae~|B`7z>iZ<0`Ze7aEAtvG|
zr5w~C?iiKW+SHZf*~{lV`Ou@22Le%B=@3@{nxPxGdU?gw)s-xMttQCsXvItzz2d9I
zXusm$|EwVJ^>{u0jB$MCrJ_|L{>PJkfT5_(W3tBC3a2#|3u7dDDYvF5GD+ju<A<Jc
z=oz}6F!<7lAUq|V(+m|%^LOIVO|$$<frG*Y)Flv{OQh3tm7%oAUl`-SkAxT)LywIH
zhoxOv&h9N3h6UU0hH;E2BIFUR0x3H}9dJ%G%7jN&8n*1A>?SX!P$prLCK*R|nViZs
zREiKYnNY-)_1vW3<A@KVbopnnMq^AS<-l^a;NIn>a2WRkPfpKqyK8jUqxu~tjOfas
zj07z)M`9MLvnLa?AeA|0MxXGxXAGWL+@Qu`B<5I3%UWEP=B9$ev74@EW(HEG<G*fd
zY+V-(H_?xq&`Cj7qEMiUKXCLTaTtkyoOyMSF~;Fs&0=vvwOF8y73YS+&6HApu@_6B
zTyEh|8gr0TIW$gCDqGC&MdgFM*#uCjBV_Sa58I-JXbjF65{lpho4XAl=WyLL0t}U)
zQK>1gENkVJi96}oO$g@EzbGW=jpE1>$3PqheAoSIi<Yk$*gbmZJzkEU(Cu+56@R{2
zq3F;#G9eG;0~`+x=#;69Mx{twJFHQlvRI(VOePscbi!FnDU*~zS&7E2m2MlK6TvAQ
znm9zR);CnvQssnl2HQ4Ni<YdVOEyGLx8GA4H+SJkmzB6+qyl+rD?ynx(w$oSnzQpW
z`dvpD{EQe*yt=OIigCLuD0`V)S__vi1X74-mrLNQOM7<2RgO@sdZLaj+7_)fH@COk
zym-Ot{EYVel)5q~?Syxi5;=GXiQU){OO`r`G|uFDWw1tg4mOX}l_d^6DU4jd>{u*U
zG^<k<%NCTP$&S_8J#NO1r_Vmcv@MtSAJR4}8f9@Iae7v9_SEtsD5`bOd+)!^!#AJM
z+<(Ys7>RXDHkKF9KIP||k6Ar<z$R<Dre^Wv5syE3m+{%pS%i`6&cg`K!*?ED@Y}!n
zYn&}CFMjbep8ewIs4!4FN402KwGE3^%j$H2R)KEVP&byQUZ9k8el|_ZcDrN0>xdz-
z-*r5C_?Wg?5JO_O+mTX|j@W5-7BnEGMAn%UGop_;X~E?gwR2S^euw~GY;Lqxq!8(2
zkFph}P!X1G%dmTfN=5Rfj^ei%6WIs{W4GJ!^2G}_cRP#*pUSHK(!j2?$sZ+9hoz-o
zefSs0my~T^kJsZ@9dmG(nS;8FnuT&D>H~!q#HNVorVvP@1a&b+k`JWO6ULEo5QJ)K
zKA1{h9D@Wfm%#Gz$;=U7X$s_y%UnkchB*#7ddh>d#nkON5`rbgp4u%~oiy~_g6pe_
zejkVt{Fv$c5$g_JoRb6>D6kqsL3`9(w-XD&BB)PR(ouQDqsS^_tfOkAdCP}D8xlzw
zvi3BC#-6Out|Dm7$-Q$<PEXJ=6C*U%34b{R(tZyFvUWrb!77HJ3I|Z5V<cuRe$6Qt
zxf+fFUkNmHX>cHC@TF_E+?#S2=6f?sgrH`vI_D}8Bi04&q8zf6a(sy765Pdt)};`{
zRxW4Z=edfiYOqx;bR;dBhKWhPFiLQkBBWRn4KoY+PNm}mL?+k&7<eL2e9mKCxe3ja
zpH9Kw@rYqk8%PZOK;1a;^R{jdCZ~xzA(hASD#sJ}i+vi?&vFUsvJb>@B#t9#_zNn<
zKKbT%Nd3U}<Bv#VUywS<u%<jqAH#G<e^$AesHa@y9z>|sxSG0ca86;H1(Zm=YDKC-
zD=q$lJ}{1h1YHX0l0ws(785hO-3>`4E-vpede0aF&Kl~bVk|At3h9ikDu;5G9E2MY
zlT4=RfY(WGNC;wZ?CO@B;PU=`Zm+Hh(a-R+R^Z2x{eH*!`ME&C=Ad08Mi~-BPOZb!
z#EL1?)D5u-blW}MIH0t2bQ)uM`QjOOH&=Ayj%u~QwGCCt2(pom<dlQ7|3|sj!dpp%
zqG7h*?#VgO)Y5G;jtSQ`>~|fz-JWb~(3X0!WOuir$s>=?&uF#fZr{P(mL5$!S&<#w
zt4_Jch*6H)<&MQ^%i?Uw>iz|BKk~#jeD=xbJpbe~_WObL@Qn84jJwe@))lKK5Als-
zu{_1cfa^vsniIbFjStYn$VdO?$Gp6H!M)1`Z@l${RlUNQ8dM^rftUt%`wdPh&d<*Y
z!^qXunmc<-yI7))meR?7Pd|**ZG*OkuIuQAzKB*@Df?JE^CLs>vZ*Mrst7_V>G+<4
z{7JBstQfk!WLBAB=uyVeRE@|LX~m+g$u|Rjw<X#guDO)56A+?g#_RPRFJE5M4_-8}
z(l4!YzA8a_WheTx#%rYR_4xCSd~6M=4kad-11DV)Nuxj|2!kh%vWKP^$sv$K66cmt
zG~cEUSZx&Ae!*(#=;e=_C+4>_K?;-3ozjOlG(S%&fnKrcD@_2Vpms6(V$IhRf~9F{
zRwp&VpKx-rWbAh2qzMtm(KC)ox*M~UkcHM!g1BT4!Cm>XnmO>v=;HG#RUZ+0j#SQ4
zH8stmWeg*(U9#*-hiu%SE-z@Rikvd%_b+LeOG4MPUvJ1hphge<K<YY@N<^LUD$^CO
z*qo<^0td;=V!9bbtX^4EmH=BgamZO1Ln#a9`T8FRb#p;pO)_NCxoWFQ+!hLtc@CaR
z=XCOux2iAIEDtfDa>6L6s|HuMxT=x&qonjZHQ!Or-ZTewQGz;Eu6@zTo9mp%!QB^T
z(L{|oge;|d^$QQ05{Q%%!!StcIwiahoSdK1G&RN^@%jq#r<E4!+59}OCN5v>z@4SH
zf~=P`1$Kic{Y7!?PQLja-c<05p=bZ`FUTeE$r+-`s2tELlU1=<QD*j8(?#z{BSz0L
zc3^k7lN0R83Fl{TaeD6zYcsC31P#s@{HDWZ#it+rDL?+BAMx%7Z}R55Z(wRIU5O}C
z3T(Dp&Mq&g7A<}oNn;`oBPvEJXBd;f(ss1~XYop*>lzZ6lt@JoS7k^+;AL0b8OAt}
zT}5-c<b=R}yQLrY5EEm+=kneu>&*rq2CCf14bqCBL75_qI0UrIGDo$+nhK>f**a9a
zAdaOQ$a|D^w2Kw@E<Lw5R}7mqez#_^T2e1pU@J_$Bq@y*c5CsjQ#dP53c+~%Ff#78
z3~3;Afz#7dQdFFt-eZ5~v9`iQLzTpuENaE=^Ji2KGWXgBn>?rYR#YcTuJ>DRHg|Zp
z!kj(Av^A&on$KQ*!mo!7(RZxcQ$nn{_vlSn{sC$murXjU#C=EK_k4PF!}XYW{NRK~
zjpn3nxLmYspMT1Wk3XZfhQIck-(z)lkK4^1&tE*}(Sr+?iv?jfLeN~EJz~4LqixQ(
z8}7h3`f-nsL40g&#mApNXSF(Iae9hZf!(+x=8U#(mgY&w#+p+=5s5yE#+Io|$w5$D
zXEbC><!TY8EL@rt6Ww-;vzC*yn%0`4NDfQucvLO;)8G3&-u&$q?))J>YKS5m6Z(<e
zdd<u0TYR4Mu%@jB4%_Io5gnC4zOWViDzd__$Lk^Ecz%8LBVGxrP%{dZ)gh>p@~oJr
zlQKqWD&sI&(3dnuvKKy;N@-5+bp~C!3UxUMwV4|31WlUS1eH?9_`sj}(J~Tn>Gb96
z#pq&LW-E*uNlF?AyifE)qG>B!RdagQ^2S?_>Gx0RJK39)&3GT?W<IS=k*ku5iMEzn
zi^k-P)adD2D^gZytpGEFA|mLYQOZ*3iiKWsl410T)pp?d?jv$j6I?}|B7MK(;=z5!
zn7G;P*mQyF`UcnSa6<2%9S{aD=Z{vxSGT2{siNEirxn%}i>s8O_@XN<ds{BHS*D;^
zYJy4OfZl0^(^87o+RXg6<4)9+<ZR6;iR+Rd=dRb}M^qA-ENjG?t6rX9TqB%1Z6`qZ
zfGZY3?8yK@E1`BNtwaRb6j5SR8YN0#HNE}BRO2Y6(8BSYN0Iqq>aHzwa4N}$5ZJ8m
z*zNX=K5}+`M!RSWP*5snM=lexSPJZ1xN~1@jGF?!R7%u21ag#yi8zW2M-G2cf!&ig
zI5(Q{<x75XeN9;3N=-V4a&vMqct4bjj_xm0(bk8Etwc2VTi^T^-~8~~C?(xMG4zDd
z6KsZ%u-ejZ_gp={;^yj>hmY@(Vx%$-t1T&JKL7Mns;1`j>|7#xTQuuRRPU_An2JgR
z*2(Or%*~V_q`3=nYF*Dc<I9B8)`=cY@+?l4q!`(3H(Z>aaej8jcE3X_*lsr5KRuU?
z@i=3c$iJn8D}IQUspG3^YO1ke*!P4WE)>=nmZzs=ow>WcqT6jnAE_T`R;SnmrfM-&
z1w~AG+Bj7OeOEw317klj3?tW9H|UhvH+`|nF<5P=oul_7<2W!31D}2V8Q=WyL!7hp
z`yIA&y#4S#M9<5+Jvl(=c04>gBQ!PJPd`Q_1FHvA^%-ZU3s$Qo;}Gb_fYq9Dx8wHN
zbDn<u3E%zJH>vtP56@4jROagQ&oF-E```T#gXYD{Ykv08FIb+Q^5&b5xqoqvDJ5@L
zx2%>cHtP+ka15)e5<zKWIX`>A=TBd--6g*9!4q6nGx#mr-G&$=Rnwx28u(O(RZ}LF
zl1Ru*k_qZyg=e&s7_4>TiV+WlVWkQeb5c3i3Y!h9rsgM~e8TC6H`Gr?j9ZekGzSfR
z&*tuq-F{EZsgP&CY$(1usCzwLkAIu-XV0blh3`sbw3fQe>?NF~8Zo&G*&MMdqf#WL
zQN(ViXHof37cB^#tE>bOg{M7?UUbw_Iekm?C{;8$n$@H<@k?-Cft1p;j+0n$8Btu7
zhiS5&kUil*i4JRo5K*IN97mz?ILpPw8MpVg+_GUDJSsu<iEI*CMUo1HC`}<LL}_eW
zwkU0d#-S9+mbQ+hun`o7P-2Lz6v`NKiiK<;j!t7rXp>Mbv-cx+FJ5q`G`D2B&6@S+
zPic)~p%qRCtVQe?lffB8OR)H18^I9CS&?FiH?nxXYf}QvBARY=34n^D#C+?H5=@iT
ztShUOmH@ewAhi+|2{jqYPaLwTWEvBAj~R=UNsJIOD1){dQ#tIS#k4KPIFz<b-LG@t
zHR;V|L4{=$n+B~>w(tv&0<uHn!jTBh%)$KPdXRZGB`;IJsEejzA`?@l8#6=Dw96$;
zZG}#zv-sUA&{;zk%1uOEj7SBgE!TEBb6#0=Ng6(L$|b;yC8&#kaXNN(b;6l*y!Gwx
zvVHNK=RX;UyB*|&C6&6io_QC!K(&W9VYzVHIx>m3-g=k!-~W)xh#gmoS=PP*Z6pxt
z_C0r3S8O*M&dyfcJ3q%+OItPU`<`ddo|ClV{K0)Df6rt!*(d^;kfL<)8ZiXU*31h%
zPy2@!w0D+GQNkAyYJg;oDGC=t<;3aPIh*Snw!1xKF@`@oJ7=@o3D*tDM72^gojn)L
zCQ~NnBxWov@kZoLLd1uN_ny&MD&tt3oR!jtX5Z}@c6&ksGYnMi0^2Up&do?sSx>m8
zmZkupBEq;Q#z<YYc<-qy$1r#{+dIkPv_d(<vR>i)k^Rt@O%~QSHz);Xb&Ec)X;)`F
zeg2fq=bv!r6TkQ3CC$UPS-tr_?%qRGU9qfd_DzMdmKY*B1U~tbpRhvl`A>dMyt!g?
zS#!2%IH@a)QoMNjIo=0sUGt6i-{sL8PjFVDbEfO|bo)K4<znW%bzO%qO)u?YiL*=A
zw;i8;az*VHJbdsbF-G?LE#0sqBjr?G1S%&@t_hBx1gj@Q=qb<hA>g!<Y_hHiQLKnU
zib5{d(x?-d4ll8-YiqG0i+&)*fT}VvdsK$O_w2eY+uimnbmv}=*W)j8{8tBe8JJ}o
z<eWq+3DA<tb6}Tq7Gx)QOo|wjfJlg*kUSd8WF<D0PPNuWz*bxce9?W2Az<v}{ap^;
zS9o2o9@LbBb_(>4L4-powj7evlawHEvKW&DPb?xs5Jb-Eh}DM0!f}4y5PioEMU0RG
zWS@&@awM4~2ewwkFv=&?h{T3gBvGcq<WjE8#gjQ_k}6`q3gIhejTV+JAuGll7=TUR
zu}zsCu{KjmQ!kgCwF_FUP?bRkgHBq)ps6cZ8_|S4x>soNbw3C^%k|YtVU)OKr}V0X
zd7dttcsOB{kswYPX=<>dDVLRe<^zXq3fcptC<+T9pvC$_{+<DA(axdk3hD-|CF_HP
z%CsL(_mw;~HYjo~EJ3ZoXpvr-nl+AMjtPf8@HOZ9>H&nDW?DY_GZT-r(9Uw8A3fcW
z8N9|-U^Qr+WS(mYcnt;s7D4I0vMKj^Qua71!)E4r&MO}+m&8UAlN4j&ABJMu@T-ll
z71&WPu!{xvKl~PN-QLje_gw${=cpJl#YBA)@tcf`lwx*_09k;MOx4u<+V{WDvR)F0
zP(+DE*T_`hD#$U=Z#LZB-0=L_OP0%u6e3~pjD63ipM1*d?34!&-jR0t(K8MMx^|dY
zOR%FwT0~Z5Ie0~ksANfsYy{d$Nl9Q=kTmikGmM32khR7-hpQX<-JY|P6RdTm?w%%D
zbS7yRHE{!Snln=k^Mp$U$um~S`DqLUFDmaUS?a20ae9uj7H+QC?Y8uNhY10SD!5xL
z(M=;8hc+l}FtURZ$vB-kJv(E4vo4xZn!0Mx+7e0h<H#_2aWQZPlPuS_*IeAYkd4kb
z(%>1rC&z?qY93UU`)<k0+dF>tr$6RLKl}qu-~T4{!zXM5cxAb`yks8|Mj2KqardWx
z%=J%y#$_5<u4*otnv<rYb&h4zuxwjgRbi@1vQ{e#GUdd2y=E8&PF73NgUXp<7+EY@
zE-o(_$H?7!<kL@H@bbkS?|=9qi*`jnT(jM7@ncVIPN}MDUOY-47TRoVttHEvyPzm4
z^+ir+3JNoMCLufL<UrD;aY!aTjkC1QvR$va3?rss*%|O@q#O1O<8U~vzLaD4db}Qg
zc|#6$Rk+U52#|Cp%+${eA%Gt-DWHvXE9N-j<A}2*cuEu-GSMZq%E?(!<Rg5Ry(LdP
ziE{8s=ir36O==5=E=vJIN-{nnGg$(qoRV<NN~gJ+TpP;T*4CnfC=`eYGYn%Cf95Jv
zRf>D}8sb=?vt_$MbhJz+>qOSYF+)SnnixHhW=X58H9-pkgObBmr72hyUV6%^l%|2K
zNKrE;#XcrBAu(D<KLmyt$XT&)mWPW4_f|_z>WU?q&{#+gMFnFC1ag*ZfFo-|)_{(J
z1m;)*TIQru0Y=F=9aK8-!qk-d3ir$?i&Yk@E!tSI&N6nMD+jx`@{=q&dPM2+RKcS{
zL<c!5jn?35R9&H)nq1c;V@k}E=BmnE`Ym^TnkIs}QqFWE`*-2D%>m~8o6PfQYQmUN
z<SE8FZeCMrZjMPgqf#K{z;55O?Gj^9)J+Bzpd>J8Eo&tkR5qZJ$gbt0Qd&IbIeK3L
zMwh_vD5%RNsG9=2{1+A2W#>qhL!V!8_U`*!Jb%vm`kG<vXRe|`pkAdEy1=&j=z~Z`
zlexUS<ej(Q#*m4_h||@qspFi2abW2747)8iS66H|cRYUlE-?gjQruqMl4Ioj<a8FB
zRn}mWbi6iMI3G$Y`mQ7A4sA8sR<fY2=<DRN8B9XJNgaB(Q>l+hhoqJWJa`GttYH{O
zKKtx*mWu_8#S*==uWUMp8aoGq?-WSPkCO?cwRHR%Ysn!JLPW&~A&4LA>6ys9)D3s*
z8-gDh_8C*u+Qod^f^`@rg3HEPD&3NLs8yn>I%%DbQwg!4lx4$>QCu|?3a>Pm`jV!p
z+28G0R5d0gbo7`o;Czqod){0&*m#O_J3e{(5w*)K7x(D&ipm(MYvMTK#zdPFY`3gV
zDxRF5bN^mT+f>}YxWpMtijl#4Zq_$oHH&4#7&9qEhGCfHWxUTcO+!`5z25J7Zr6LB
zKfUE=KYd13wLE<I2r_u@8HSz^0;}2<_4MK}QL?jRTR6Ec)8;Q5gV80cMmn@Z6ojE4
zdKy1Uz+NnID7b{qV6)<Mv0zLQO(uqb(ng*mMS3q<UP>Ji(O-|(<KMNRv}kZrp>Rlx
zXW8phB#u3Czni&p$!C0@h*7|t3GGNzX|5cqv8?MbiP0)moCt*5^<^VZNXLwJ92U;J
zZWR&~L|q&eRM}%?znvU<tTk9|nYt`v%;NrIEKXZ$SF>6zi9TSpf-J(r(_X1VE&zcp
zK>g9)Ke-P?MGOKg=m<KAmTIO{gtH=L&6pLQ#281OkDlAnbK8&j)hVav_h~bXW1wvs
zF3;|9QrEN^njDGs0%bIA*}{HJ*xoS~T{sgm%2dVoTslj24&(x)Bx?_4OMQ?jo3W}1
z`I;`Wmh!PS7NZR)J@X?EjR1mTib-`ul}3Q<<%=Y?idciGDqOonH5ICMWMfMyxNtH#
zpvUSk0m}Rnmm{qxUbME9my3?<SB&EZow?_AMuHE2KlX0bYD(x8`!R608yUugHIXqy
zs=B~gd3tB+DDFAD=Xny(e^rN{P3704Zs9nN<S6>M>T4VlUoWszhU99}@`U9ZZ}ITk
z-(!7!#b?_)V)WRlE2#u?Q-`cT%@h<b;cVu;cc0Ky5QaU3$QZro_E^c<;(pKm_J-~H
zh8NGCvRJe{xV#is25b53^Oxf8u<!9oW9mvQOti*oO+XWdkzwDl+iw`hQMAaahGwzE
z)uOkftd$5qO{F>Jc5khU)JjeQ$0|)$rsP_(*?s)+FL>jPN5TnE(n)K|^B7}9urLH(
z*^hvk<0X->kUpbuJ93nIdWar^D(cx9*DhHpkx|)g?il+{ZoJw-3^YxPs%2VXOv#|Y
zI)?%lixqv`%@FxPSTZvXov6YGFAC}@v%b6Kba_h72|tc3o0i>v%dp*YzHHe?56QE#
zaKBdg47ppg-jCeIEx-TAKW3~g{cgu<KX7t=#l4}UIx$!k$<fp8JzQL(lx7SAzIX$Y
z6MerYMQLVPtyY|zoD?0nj5P-DJ=^V;{VwwS`G)JOJNka$$(xTkIjI@_p51;!-w!}(
z!#8F&Sk6gMx=;|nS*qZ8{gsmJ<Uyzqlwlahxof)bdX(Mbe8kl?+8V4;G{&-O7TiWK
zNG2(~i%7_ska7{){N<MV*W>m0cVWn(W-!`eRgK9OofSG|3;`8K($L}iJxvA`H92Z>
zl2YA-x=qPbDgoW-N5*k1fm{i6m=a)-e3;O=8Bi%dJU%85rPk@NrmUxokRv9PPNT#W
zyp&<p>?NvlC>D-|24g+ec$z9=tf6TeVhD`m2&!~rBJM{f>nZY(qg)j;wX7f(&+||O
zd2>SI++s@>71=n3lo?~94<lD&;F(WsQ)UFr#eEiw6Hta65=-YeuPPb^b<SKY8a`j&
zk$lEhj?;UWcxMRR9^d!$<AALMLA1&s(onI)76pr&9d~4dOJ!e@>sv0g)?%48fg;AD
ziVM;d08Z7p8F2~%OsE)1qbGW?3bR_SyS9$1X>rR2WgXdSLYV`{fzP}p9}dr=7g(Ay
z=g%O`4hJb3HSg<R9KReRsaL*z*~lnjhW!xP4Uut-tU?xQRGtF4GRMkqOo6CM6nbEu
z<wlPmlVdL3<AE4PQuGqgl|~Vj)0e&JuN+@1u+upcl_pdT`uvi6-*}(5H@EC=ueka7
z&%{$M$Qzx+_|*JA`I}LR(-X%x-hRliyGHq*#@39ZaFyz|A`yr~$FSS7zP@6+-th3@
zLo~2x8d8kx?lvsyhRw?x{4in{4OLUiV$GmJ#P2$WVb9p@+3)wLtf=dn)Q`BPrCPMu
zrj|(G)=aZ$GI&N6sS*V8^%=Bt1Ru%Dusk^hZF&0nGn{c;Twcyb$9k^wPufO@OaUkc
z+eHhILeex+=FPGA-^CGpMr%d3BK_i;mQ!bOwqkvAMK|opDdIzzc^oMvbXB2kB|Hl&
z{E5}_6o(<lNC<x3U^22<2}v^uMWk}<I#FG(ZO!&>EiN57(f2(&b@012I#Q`b69)Wx
z&8oGWz43&-Gc->g@bjPkDcy?~+*9J1G5U`E0L!HW>HRq3gK%6Dnv;_=lm%ln*=Rwz
z5p>ay1KtnxUB@`~Y&Kigx1QUZ$bRoxt}5Ps=K<|1v+LG$-3A{5btN6GubPLaSx0Lk
z(jK>;Pbl3aS1{#sAx4HGebDzk>+5U0544Lw(}q$`Hq_R!s1|I|<Ol{!Qd+KehVYdQ
z6R*eX@s~6XBqaG>&<~?6##F+x%Aj*ZWe;OV9QLH46LwfqsH9Ozqm`Wx*~w!y#fT4n
z7M#zMi_n)Wgt=&S1f)#w>j-Q4<=@LvHOg5^ZYqOt1Cz)KO=pu5?^qMX?|kyaoj89!
zcq&ujs*0v<SS(w7=o!bMc$3QK(q=l{B(R&K&jgH9pqfF6JeDemzElfD=zJeCcRq2Q
zU~3$`RYaw+^@8l`;=*8g^ZtEo-;=wJu-nn6p8a05Nv#$1%aI7iGH65U`Xb+vi78^z
zD_8b_fFkcVt2~SDhai5j5;%#lvnpScoQW#sTU9DsDrboK)l#DOvy~h<gEJ`SFja+V
zD%9N2sOAPkl}lA?exHYO=^W{4ik2cr%@W6n`Ipa-sl5BjXJ?vdUp+G7yfVbh5HmhO
z$eNJB&B5()`8WS^IeD6US^ix?ahH$)F2$mb5d%3DA?GYn;nzw3zgA#pkf?IV3RTrq
zmk)UO!8f^m`Wc&>8{*Y9cJ5jcg5xAfqLm^m$)etPbWU3(Hn%TWI*U&p5LM~kkJy~Z
z{T{#Hv%OuT6`Y@+5>ukCYIb)UbcWVghTA(*2-so3w>6*$!${coq<$o2k4aE#LBe#1
z#C;F3K$|&ZVnwcwf-zo&f2HFGtP_J|=|VMRuXyv#NBrVvACXgLyV+t_3-nQGPxeNc
zw3Dt<CYJ^N$`p`AGWm{bO-h3Vc|kgUvS7|At*P70Sq8UJ^!=VV4h%@PqpKPlGp^B4
zEK#(vwCw_`v><&&iw~U-8End;5+4F>+d^GKlv1R%R;+JgVhBOXo2j3Jqi#14wmZDv
zV(u-lQQG7mJbFM?8Lq~W?>9B$Pyd8(zW0F3`%BuYqOKZ@w#1~Vjf0#}I-^x8TnaTS
z*URe;blr|_zh@YFw%aY+O(uE~E`ImDH+b^qIf)Hjca0AnS{a&pRRo+10viQJo(>2R
zr;epmd&msSjy0bxscdvI7Y6#SXYY1Yt|6*HI^KqWs*Gs-RSS%A<d7jri6&*~#!I{|
z!M+~<4vwSXPWC^8F`_JBl+gQi3g|o#hYhj6BMmz&B7LHgLT5u(NhBNM!OPPJ={y~W
zLFguM;Fcbhzkn~4tG<9B<)ef9DDeBT!#tN?j8Qc6rt@{`b~nZllR>3GOcHnw+Axlu
zs%@xh2Q+v;(vMqW_M{PF7Pkyn1ns49u<Z3o&PwIWdsUFFBI$%F&PzHgLW03(b|G<_
zGFw*@>lP+)Ut2MZ;%UCDYo0uLgYLx(c3sbAyOqRpy+Nrc&PH{EGmt8Wa~5SSJ`_Ry
z6vgpUG5g(4b=L{;OevS3E|lOq%DJdD+S&4DO82RtWUnLwInNcR9Fpj;^&%;gQWQYL
zS#(`t>KfHlQVGiw`dXw>%KbS+_LQH$%9WctaF6W6UY*LX{_Km*3zN*$aoIWNFNRQu
zb0TKN7{O<SM-x%RtSG`YM?T|+&yyz&+az;+N&{4R4^t34>DHxDDq*>xc2n2%*MR}P
zUSOw+J%`aKU1N=+y}01X2OqM2{*32$8?qnETS(-r4`lKZ#yF#BDp=n<XIV#55?vte
zG|?ymo?+-nyDhiZSH$3X^yrea)03i)quK9vIBhvwp0Hlul7^9>6n+>O#*yR$dGO#P
zIR})})W(X0Oh|wiOBdbG(6dSu<!pVx7fYaL3e+B%xM?zpj4@nZUU2jBhF^UA8SlLF
z)&b%#H~7G%n*zH$mn3G59aXr9DgzEeB1Vr1Nbr%13N?nL1(mZ^g|%nYPO;w_cDs%c
zBB2}56Ioapl*_Xfic(6XLCC~l@L^z#<6+`uuu2QpF-DxRC?gZmRTd(|apFfPlJ|t1
z7-IkuwH9;a-F8cNeZ_9Orr+=Q<l~QMv!H~_(<Se|{~oKQMFYV{Qr1`uB!x2$m4gI#
zIx~iZRhozhLEo*{ct1#Kq#qauPfm^$GH<@|m=8XD7vmzE?w0NTM#QJJ#o1a)(nma;
zRJ{Mb$gP`(Qiq?3_b;<)xv`ddk>+ew=Zr}i6%r&boSo58S8YK_3&lV|&N45wKFUbH
za@D<#{9cd0q%k!kC@I5PqeWOAkIo4lw+!3o_}xnw*5U#q+zCM;#JN=(-U>=L_9J2N
z_~Bp?IgKgEGGCxXsPbNZDQGJwk<sQrmOl<!v^oMx%HA7_LsAm)Sm!LM*l8uCEIbAe
zC@R;})(ec$xVmO>vY;DVVhX|;946kNhSDW^e5eyLE*C<-Nn{raU=Sh7i8%Iwb?3Q3
zv#Dx0yDx|sh}1<&bo(927`)N^n;-rTb{L5JJ$|#n_>sHa9yI8xLb)1<s2s3O#aUf3
zjAFeN{fLP9rXx@x&w=io<dy($Dv|3V^;FfOEvK!M!0x!jIg$Qzn#-=id*U#VN=)Fa
z!?qRmVu7hEv~zL}XDJO&s!nqyNr`~6+<s|BIhGKZ{fdtSp*h;GzY4uOSa}_Fsg``=
zRbco4&dgw2R)nM(qas8I*^nkDmQt>sUpG%>?<D#_`G5%)d{U6?Bu*T$+@loan&l`w
zJ8~{v!e7lB{NngpfnCfALqwNupsW=6<b>+YcUXPvmhH`!%};)UBf&Tlb!OCtE^C6V
z&^}<&NFFy>E?Kn=b!*U#0V5yKgut-fGi<hm{T`<jPaZv_sx3YRFp2%JV|m)3Dph1k
zJoo}34|3?InB+1jX)(_^<D4%4Hi*>?xn!d?Sr?KmVkE7!5-(W=iHJgztj47Zl%lhN
z6f%__I6YhO^hWdIcE`gwkYin<e^dMwCBW6?W(zzEw(#KOTFRfY$waG3){=uoXYnP~
zIia#Bc4wu?Mq`{J+Y^-4RC<fqbT9;L^3u4X6-fucVN@fKywWHf+8JE5Bypj^;~2Fx
zCn%&n-{o9Pv!#t*YlTrk7N6JPH8yE%vh2fvCDW^nYc-V{`QGDme)KQ@8CRb_!;T|k
zihTdkC8Ms;R4j60n+L|CHDzL=N)i~4HqzNro{qu^?7KU<?JX)n3Ypv|(mrET<l&{^
zn;$;p{H$R=ZrE%F`XLjO2w$qwEv2-U0Lupnnx{@?Edru3r#Xw&MW|AtgoK^Mi_(vt
z95bh<r_|NKN-nD;I$1_DgdXJ+Rb#0=Xy4;gqEnfmA?b)S0c8s1LaCY8Qg#Ji2X(K<
zU)q?0x>-pdFd3!2bWG+5<A(n3Ik{g`=N_vp8b!<jkOOT_T`~k8%KjT?B6v=zlzPQ0
z6{F?@PdL|hMs2h%avlYwk+(UaHq7r=#-ZFO$c%L5YIR854<A*{HjsiYW`YtaqcM0Y
z>qLmRany@OoMy&R5_7~|NT|w^OiyJ)rJyoKEFTd$T{J>D2?;-Vp07RYY}r*cqpRqG
zR7_F{lt`Y|C>jGcjfAda-0xArV{B&V2l|qnR^33tkhMh_hqe+hh~1STI|DID8C^**
zn$_eEQNV^A1^m((bR=but7}|cVVsi+kT#{{EW;=bO73?G5kC$R)OiuswMr;Pwyto?
z7TwgM-^Zj1kq_4_7SvB6qI<w3C#=*_Aa-osR-&@$j(-_ScdULeSI+X}oH11tb$N|6
zI|$8<b0z9aIpb48oZ<Ar?o8?VeWDbqh#u!uN~mQ!&si#;6GAE7!eUGyU&THwdg&%E
z=;7o)`>T$x71+rEbTE~PDiNK7lNGBs-{#S`?ueT^Hcvl8sZvhUS@e05#v09{bv$@@
zkJZ_Vx^>uAlU*j5L>hu{<GPW}-I_3toSvR>a(2Sqe$VQ3g)y3b>{+cEeDY+SF-G)v
zR8H8UKArUxwaW!JT4742l#Am)^dr7EC}Xi&G;k)bxoM)K<%4G}nXAi2ZkSR;1y%*_
zb~~QmM2t<u0wt@7u&YI7e2&)3q&hBj$agiDBDAxZ<cQvtW&!zpnk2rrWMWRlq)0|%
z7Ax9pNy!q@PT*WMf`P0P+JRC6c_+kuOlc#?pRFK((IS{UWxg7PDT0|&#t4KMiD3|(
z$*A$8WaqKp(hVKDU2y;5BYL;QtQJ&R@f+{Hj~X+({hqqg+*F32Y<dP|d8}ZJFhq~m
zNGUi8EI$Il=+fC-vZ-PrlR_r+kvIfQ)~rq%p1gIzg9j^=&UD>C-;cy3+@QLuW*3o+
zG)$x%i=z4wQl~^ZfhlvFliG5xa#3VYX-4m|%d_dUPXV0`)>y{T6L*Q#3aS(t*0<Pu
ziEN>S&Nx;4Dvw_3^>{u08yUaiAu-{}IY(mhh~9^W9Pr(SezV5*J9PFKmCAaO<^r;4
zd&yHsiEiq8O;gunp*iTIHJFhS*%u39hmO<vUCEw#T%If6r}8VPg3cTYYSZ&h-zO@{
zgdR@3zz_nXA5lsy1@+_??Ht-#IS*qN-`1(&VEW?VYBlkxgwrJi{4qrKz2~+Y*xHuf
zDZDcHg!tNoNE}CME5TuH6fQ?n@{mSy3M3+dfRB+F6aC-`F;OWkD5f>UoTS-BiN+i#
zM8rwTmlD_&j&YVUI+K-R@Dg+zQEQl+P1HfaY?3u2mtforG8?^@0A6K`ak#3&H4UzA
z$kVke*Lu!j)8X2^60{wSNyS2|<G_x>$NSaCah3m-Gl1!3)zP`2=2@4L_=#ua44uT8
zB~UKEhnfR72`HypJS9*oW!ap{d3KOBnLKlIERHP{#9UkS*Z6>6PseV$5Yt6W5Q{hJ
z#e@5te)rp`amSB#Yy9n+%34xZgcRlQCUSPU<noP6s%4FCEK2O_$P~h67ujtG_T7N3
zYR)h2u{u5F<=vXPX^CaQIqO7Jv2^sLDCWXsx#_5qZBp#bQrR%do+Gr{Kn^3x_XHzl
z#H<8nmYZL0hEbw&UGQS95K&qbLsD$|z}04;YBghm<x#m0IGQkJ5~lo1nq@PAzx=MT
z7Mv3+o{~)*c&|s#LzYhA<&vZfqqFq8Eu$X^`;Ll?X$&zAn5x1$nT#TG5{b^CQ$|+~
zZ5_m-wU#A#B!~%gEFC*x?4?woD?;+@w_8*mIlZ{Vo!=)e&UyaXHQ)Q-4Q>X{8u;#a
ze}i2L*vfKy^^D(JZ~5b!HQ!sc+;1GGmBHqKN&&AGF)GGUk-ft>X`s;Bp_KR>k}@F&
zl!aBj;NHar7nc{<$}x-sUEk4<12IJ^S4#=NTC~=(VHdtfBqI|-?nO$e!2MHp$9zww
zYaWY7phgo4=Wo7e#>}!OT3h0vQQlxv=I-Vtmv8r2<%ro~sj>P5t1DplwU^=R@p}9>
zF*1@+#pLmE5V6^;$zfo>S+n2V()tc%HCAP0ql`0xBTte`S*i%G#zQm3amj2t<Y#Fp
z4u|X^eEX8q?Mn{zJOxw7pCn6hcz*Qk-F-ONm6AX(CF$Jl$FYC|P*ob|993OoyMc@}
zKB$;xTd|xc@!aWcAJ8E|ZxmyKE@ald=Vln#=}Hn=Ehv`GS~zLLK<#Q4)==q;B@z)d
zG-i*Ffk?!sK=6^iHw>d^*=Ui?%7p@wAx}*PCW(0N#3wtt_8LK?bS}dA+RTnRrZh|#
zQ_!h`av3BboSZ)VIO0pUv&vb5;kuTfuBowAO;m?Y>6DTLleq-IB`{*%a}VXmJZIa}
zzv~04H~(18d-ki1a9{ZQjDStgb&U2M;Fc-Cn_fSqaxNAG*;z@>kLS<vl$!%N%I8Z*
zYa&SimwcGzO(sW<93`<)Q()~JCV$D?|BCVT0=tqO2^3!r!<>jY(VVQnyutlp$NAGw
zc=ktsg!U0fr1Bn!XqS28?faZuwp6XAu7uwZvMlyE2F9+Z+wDnYn$?o3X^Ft-#Rqg4
zM#$phs-i%t(MzxwQxxc5t75*Zh3A(`CYTMGB3F`&va?de(Gar|TQo8jCt5wKto<l;
zf29*SPOagZq%9#SMjTy=Y{tYG3nBgs1kB@`o^nD@9aUP-|9%V^YptbQC>=y?bDpx5
zLgvp|c!UX}G3choS>Y;<n>)g858DnEz_yyX2w@J-V_Zc>W111y*YrcrFpOB!3%u{D
z16-{|jwQrM?>mxGtj<q3T`gIho#U$|pKd&#KEI~=(NAEp;PH3A%Lnhg%Rl{BKjdn+
z!9I9G3?82YKi{tD_ks7b=F~~YbTo>PHQ6h)hg2tWD<uxC9iSQgKrFtnXJ<?9KfK`N
z?2MG*ZntIE?TJyYt#cC8!67J1u^^_Zcmt<HP|bX;kFrOt3euYAvdy$;$8jv(eHmxn
zyd^kOqq2oD(b&w(^>f1J8RkS`)e@&1b+y9S>MJ*+*W>m0%N{97IkF#n`p`3mf$gRz
z?$+FX{0ZrLjWZd&tP64@MaM+pla=z71WS^9#~4XD&78UeCy5e&9rn(HCg{-vrkd7-
z_WaU$Cy&W_3eq^NtLeGt^_^#UUM#XKfj~|$hCn|I!kyC!i=%2=7BS-^tnc=WLobJY
zilwSCQaNcdNjZ^>h}z~X9^P2P5EPpL*D>?iz9ZKwTvgF7S0tZMq3B?uXsl&fX>=U%
zT}SNp;$)LzsoZ3Sab)N`IwjUOYtA2?p`GZi$~}}yLn?fr2}wLw3e;K&ClQ4EXf(!X
z+QmWwSyy3gC8`glq9g4cnOyW|{W#L?_mD#g3LRBli9?brrCdbpKb3~K$zJO?(2{Sz
zTr!`7`zfe9j=zo)gf9!~UU`-&fS-T=c#gd~iw@*Ei2+gykOCTmQ3_`bu3W2$QZOT!
zrRtZV_$klNItO*pdx9T`!&vxr!l@JN$Vv&js!^`O#J{M(PCh<Hm=1DCl7;V8HQ42X
zcI<fY?QgNUxuXB<1!fGGIMP($Ruxa)yujE%W<<1kyom34j~Yi*-;swNS2r}vCG~Pi
z$|C4$tYx>`N=Y$OjE+TzEKXdFV!%9+7<IAsFjF9|6@FCYG!nx|j6t{;F=G@Yqe)6(
zloB357H%P#BIS}$O57ckw#2N7ScatNqb8vb0ooT(IH}VgNSnWe(j9eVW*}Kpg33vk
ziFqT8QCvE7mJX~j1`*$O26wVzr8QO={BF;%-4h~IF;LN>oRe~O2~c(AFm=UZwPYAN
zx~?nj{F0rR*;`S98fz4dvn*Ug+canmQ5$r<<nH-PUat3C>~66t@Z&%HA@QR>CAAIP
z^$jQP0ZBWuYZ-0L)2E-&?(aA~uW9Q{HWEOZ>~P-UlLY-%iF|@mR!Sa5QMWa#dn=Y_
zE3k&%_iXn&`f)5Gq~arMtwABB(>VDuYPrWKP~kwf)%<dBrIaRTQFGVEP&r2m!Y7GI
zRN6;l*=@HpO@q~jm;x#*h$C&S$agpFu6{<<HW;h0CgCc{f?to<<KL-qH2-r39|HSv
zVB2-{-H0-Z)F;M$Ko1Jt7z{Nko6=ZtD0SsD$zBu=R*a}Dyey;i)O0>KswK@UoH?nc
zP66D3ulFmx{lZ(8bNy&t=OUn+vsi;2J>RTxsiaB~V-)Mal!!_6GwY_MZ6j^l((MLf
zPQu3<M}i-5wM7|)QsoRN)s7T1Sy_e%>oM@@X3NVM>27b)KC&7U&FY**v!FRSp*cUr
zhCS9Ok{=lN9Xi3XSrS8Jx8F1Lf#?&-M|>VwUtiOoI%?(Qa95)Ckg_If(QB3SCZoxv
zxsQV0$+eAXl<rw$shd_fdPM`&C^JPoaw%gX44yc8!Z-pMSJhOtAe*jjP{tv87?gk}
zi5o}Cr6Q#d%zz^VF(X{>xh$$DI#~HSz-+IUPUm@WDC@q`xOgnFa|HN(*+iNo_0sga
z#lcFEFe+oU!fJ(4Oa<=p_%!P)rksmq+*HmLUfooJ75&)4s|x{?TzhSuG)dMi+Ev(J
z-FV__3KUNy#}Z_uqWPn&24_g7qPo1#>3i>T@!}buueT)MGvx-CmuK9&e~-Emq$GH;
z08<Klx5sZc#BPI417^9TUM;arO-!&{ouaL!>w8>Pp|qiDYf>)e@Gzq@uhh%)<X=0L
zjTOo&hF;97bBct1z(j@58Cz~pB%_QioHm6vY9{m=dw^cins%`QJurrzaq9Fceu&A&
znIJ57UwV$_k0&1-m1?teB~2PqD(4w(GIKvo26Tb<BOn23B4Lf8U9G6Cqf>_MZp$!q
zgyczN+T{{gwfN(p(pA{1X0bY9zwdEXL)|WMQ>U89q@z<UD6Mg5Tgkhy8rv=yUCq-M
zH*AJTpF66g&@r+!HQ6}cI9;&Y?+Hm0ng!QHF4`3ze|kj|_dL36Skw)Lr6A4~I^(0q
z*@~1iqmPV1R*0=@R%a);wr2F6UB722`crk&NV~rhm_LRHTITE|O^^Xn5hhM*4(w)*
zq}EuaiV}VnE3=fStBS0X7>1Yo7el}g1B+tqrbs9iQ1ny^YSr=L;~(+(;y(7gC98p`
z2H~KVdvGkO`Fgw_e<|Y&b1!GvH@y!G<A{%mrE96%6}~yc>z5d%u}T8F;$0<(3X>yC
znzhkVOs^!fKw0nQU@i#S(ecdmw|o>(eBm1<XFh)C6Z?z9={-G29fC#XO5CCBrY5_1
zLE=IPgkT84VvUp~tE#4LR~Wlxa?$X9WE=<TRFx8EF;P>3P^ORwM7G_?&93LgcF%2y
z^Z}f9RNnLC-aXDPE?KrM3#(aw{t-HiBB@al+_qleb)egg<ft%8&?8c2zumIiENN7W
zZOk0hDg{~@k~%WJSBa_lWb$}OInB;Q#@J$)=;m%^IOx_%mvKpOi&(ocnyRjFb%m`Y
zFiF~oIPcUsTfU3E+R?o(LFSk4ergIxU%jP=lI@>8Ug?4@Pcox|Q$<!C_;GVptjx(-
zj-}YbX_c`?=Bzr3i*f;iMCY)SVyBlqPIN+Xmk6PB$EJcksUqNPF}A{3S2SsV^>g9h
z3b8vt>{18LT7xxcTceu=tH+Od`0Z~ptT%k}yFZe0lSy1WdPFo8`=n?VOZ<K$E5k5$
z>^l!ZV9A_iL0~LZ+j4!orfOSKR%~~B?wy@e*DZD3P%W0M3NK8n;^{WM*D~2f1e2oA
zV^Q9UoylMceTN?hR1CzvV<?}|&WMhvr>D~GG|h_%cC}j5G!@R8;&mHdLEXs5h+TP4
z#VGiTb@^l72}p1$|Ch3%d_IxZPzXVm!cUlZDbgYi$~me<vCp!GwBO=`XM49U_GJ~;
ziI{SV3Fm53Ra4b9%aavKD;A4IDZ9&O&N-LYOeAek)?tmtT8*u1Znk@_Uf!_TcXVOI
zsfHYS{Pu=gXC5xjcy`wje4tubOyvkU@iYX&^$xXoz~hCXP8otHSpz0w4a9WNRT{=f
z(^lNOyrgLt#GKi6d+xSd`f(6xhQg_v7H2LtKJ&dvNzj3r5A$lK$o$yCons2PrR=W~
zbBWiN1jSi9bL^B7;bw)UQVu<iyuAJtzxq=yzj=aBnr(kaKlcBXWrM@}cs*W^znqa$
z#)n8hjD!HLYB^o3u;-7teen#ZGe%Wt<s_L=38jkMNm&megi?l0zsyIP_Qj*<Y%03W
zK^<!T+hKn_euSJ%IUk+#nO9yzXI(5KnX5NSC4_^mW+j9v(#wcWoev)Eg-=yi4y_DT
zRkJ!drMkLd@B<+R#&KXAdSYy_S%WGb;pIAnAnp>kn~s-vTV8DUtc@WW%}Q&Uw&mUT
z-s9qpH(4$gR78HVy&<iy3lN~ufU`A3@gmj9QX7l12Au+Zw`a4yqcxe<Ez4&trAnPq
zrCEYWd2Oo8B49%-jToRLptG(7v!-y#N;s5DX)?rFqb_FwSL&vo?J<=#WTgu-noC_+
zs4(jItS&HI{=%G)@cghRPH67DBdS;L*WtPN@__FcojpSPCK&K|9_IO{<yR-PuK+Ao
zDXb|0e2%CTMT57T$1x?5NYe#8E@$*C+8jK|doczmKCoIJ9IK2i<!@Jk6)ya*I=-f*
z*bxewvxqQ)Ode~{%A%FSo}93H^9iT#y-oA!75%d-vhujal0W$4k7=8VckVwVZudO6
zIAPJ8l3ZlWJ)3<;Z7ua;fvqZ<Ma%B<XE;|0m%w|L%N0s%d`>89=GuLED>DQtL5Y|i
z$DAP=2o~im#yXsJcr6w%Y3xZM5EOVThQcXJfFzwmiKu1LCzH@tgnOM)G*)9(LZwIw
z<3VLT9kSChmCAqBJV{<5?H)aLLX4)mhQbI}6qz6?2|r=Va&?-xVu_T%=S(yj7B!33
zQFS%j&6eF}Pru*Gf^8P4s)B@gCc2uuSfSADcYAzDEEnyBEerC9a2v8x7*pY#pi?;|
zo_+q5+p8N0;sUW~Blj=Q*bhDX+cl3bAF~cnU7T^X*|WReuzv9wyW1OXb>zGY+zT2<
zB1cbZ9m+};*zY>JzQcz=%5ZXe#=UzFaCOD##bdS~2SSY4%FeO}6KPp|xb>WwWn`YH
z<0C|=8I^vuJ)U1GnG4Za%G6mKA%#zwYls+UNeV(p*kq|#Fs?gZetwI(cgN0DTy1XI
z_uVUj+v_0i_4s#f#1I+!f!(fSyC2!F6bsj~ym&&t`~^0OtB*$EoFl{xK9iH65y6l6
z!4pS8t&~2L-c+fr=AvabQ+%=<=~)*y2U)OWM7=;{s3_f#S}HSATcL=jmc3EWXhbBV
zFl2q`j#Y?LQAWyT6CFm)?;{WbXq6aECe{vTYMQ#?;<9GF-E)2Y5<m8g-9W$V2&W57
zWk_1yq)!Q-Gks3H*pGa+zT@d;&pKxI2399$y#2uk{QLj@-{o81`wgm-Q*sJ)w=c1E
z#isN4&SSD+x4mQ7c|y$W$C1ivmL^kKX?jRw$KCA}jZK^^YpPn*5Q@5qc-Q7!5YnlV
zsU+Yi0k~2M>zp+36`L-tMdk#AqOG!McIF%pfP~T1O^vN8P~y;&Mo%_&ju{{qo?1D}
zlt!$gvKZ;2z$z#el?;L`mT*mDUagxdCBM%hW)mZXL$JntrfRZELXlAviLvQ5=E->Y
zJtm1V=J%fp;%IcHcF<Ic5C(D&Hfvlx0SJYUM=XJpn$fx(JjshzogYR>vFO+utgSGv
z!qqLNs?kMaW+wdqisS2d#U76twwCV}gHh7WY0pkMd+SZ^zjwo@+a8q$?mc+K4tVm`
z`#k&TWB%2@{C!#nzy53A;Nf}A>DdG9-4@?%S+y<8lNA}wzUz3fTCv@3i*8tjF_s{y
z$RYbH*+iT-358aP6i4CYVZ<!iwIpLnx^O%a_#_JHIYfr8BczN?SqP#=_y;ja>Dg&R
zwh}bv=&7BeHi{e)!HeLphS@G>;&07}-3<B4jhe-bzW_pu=9;KBY9m%YOwca85-@5+
zV=ZSOCZftjqrh35u{f1jl@6Qzu4n8AP!_8VR*PMla}H@AmdtG!2CAyUS&K0u5}cE4
zti~9Ou_z_b^vh?@`26EfFgbDBG?-YULdE^_M|}GHGknq{EKlD1fN%fxzs=2N&&NOd
zH~gRe&p+a3=vj{)^+#}NYTjN{I2Q<JEJ&qgySrn%xnmeTRo(LV@f%ciLrjV7zGK(z
z@gc|tUCIlSXr(ke7*Nt>q)~?h<tr@s4#C|4{+|#(g=CJZkXg0X_+GpuhrY*IOJl9H
zsOttzLt{=^RPfd}en9i+J%0GnN31t{#xcAy7hd6o{J9qA>+xUj_%d<Oxqk5#<IlQ-
z>gbzf{$Dy7zlc0ckD1Yh@=~wVBPYD!i_fQyo<qHQNPfjfNJ$#o_M0u8&-B?anwI7H
zecID|m|Z014wZwr8cYrhnc_YmU(ri)t(C!0ykrT&Usb74BE%g=&<=>`5)UYhvS<<_
z+ayk~S!-0T(AdLeRMY3@sH`g;oEasQ3=L7sK|A}HCtZl~<yvaduZ+<%5>Y0w@HG|@
z>2*!T$%Du2V<xU|7`rX~W=DUzL|KPwEAW|qNbH8d&EWa8>$$Uz>fs~aKRu;+_?S1}
zeUG=_{RZFu&UZPxcu34}^Wt+p`sl~vsXYc#7r`f-HFzHwwI;7>F6%_wfj82X>vYEZ
zf!%)3?Pkk^_JN$yDU);}nlP6V#pP8OWtv=+C6er0M%+fEE7le5y6L)#E08L_Xc?h6
zSzS?8johCs;6oZc)*Vp3l+!DcZ&Og0Ri3L^*%*?Rdpk91CeqBuWGa_Znqmtgh|06J
z;57eKmOz#=hoqrIv`UrIWNi<XxGX$AmGT@{SZxbU2S|z#L`qGmgjF7AJ)2FBA2Zsv
zTn5A0d0i0A2xGvQvJO*5#Ypx&*$?CpFr{3NDlSn~P1P<j&Y_R2+Y(AVjeo`QwE{aq
zuhn#0X9!O#frXRGB&E<ejILRnUvT#B8@#xFN$VUBzVRV$v*q@7!*<_u|KSsC9vQmC
zPp?1Yjf)dvUO;6@%X^s9OT5<f!-zGSyWNJxqQO=U<*fJtzDmeouG5!;Ip$nMf)83r
zDT>5}wGPuRvBqJvq3`$7s_jRatSV!SLRSD{7M<{ftRPsnH(R>fTbx<2s^Koe7>X89
zlCou-Azm%&(MlB!IZ=w6x`C$bGb?FGp_E&vZZd;xMj-`CKL|N1Oo*shB)e5YB_Z;X
z3?^xa5$_Xm@K|kFE?2}bwVF%GP=dxx#-d|Fn?%zr@#DbI4cMjV7bTTtLnLFILR$sO
z;6vc)(`WSC8!lUE<T$GoTLnM;=|?0yZgIxVi)$_)J!EyU;_@3G@ZmS!Mz1RVH~-uJ
zHN(pnd^QID@Wo5&Hy?7Q5|z&+6Cpu&b4Rz?k+kK}+wb!5;aeEx8A8XtyJZ}Da`M#9
zQdO038m&Pa@n_X&Oj78qO6O80MpbwX22*;xMm^wFCTUH^kV<QRHsTW~-m+49NLl1|
zQiL27zK@J<#JiEYZpq5g)U_CYC&k(Qw>WwD4X#4P?Y8IU`UW2+VeDzsETx85&gJ8O
ze|4@)>Y0D(-@clw_r;U^tAdYT;l%&%<oM!s`LgTu^{>r*hmT{uR|3DIr=Dm7zw`w8
z)j6Y%p751N=cC{GvKRQm$ilPhJL#DBJ`$1~yjd&R7qTb~Os+htbXW>Sq$E&@LLE`$
ztmd+*5-6rj=Y!Im;OQl~nh>=@M^RaHOErSLO!|{^Rc-!gWg+D}SDbV?BTUw_#wdys
zOcc>loybPSQDV*{<aO6R5VY`rm8)?V_c_VpA-;be8G28*8&I_X5ZPyXAK4C`n|;qF
zB+P2X{reAj_`!!<y!|ehj~;Vz{{iQh_o<qeVHgQvWV^ZL)Y*dcNVP<S)+5?k>eB_M
zqe9<W#-xc^c+fdT#&KZ3@9<tY;nD~oRGk@kK+Td_DP;+swJ8ovrnrsR;;5kydBAk;
z<^#we0?!t0tlWQ5Y?k{T^8pc4nhA81KwN#{3zU-cF)KaumFI*y7mj(Jf<-}S53@!-
zSBX{*kIe<4JI;L$kmaESdo+Pwp$kYnOH8>So}g4lDdFAkb|WbVY_(*$IxkI)>436j
z;q!||T`2!{STmC*o}hFU+S*w&Z~_<ptR4Ayd_BZ2^U4R!A)}%gDQk?-6O6&>3ZE_7
zRn(Vf-2e7_yz%f6oSb5ZfpcfLr!Tp60qqBNH&-x@+<W(u-Zr?}a`NOc>wZUf^OB6h
z8AG?<@ZjMCFfgQ0ToI=CmP%<j&b9nWPQ==}c%X{Fqhu?>k;|?^84KDHwI#A4N8#LM
z-;oC;S%$G#Wl3sejFIiy^G84YV^*t?)mfr8fk<NTpp$|q<;_qQg}jC#7p?*mBBL}h
z9p<3u%Vh~FlV}<VPfr3>ZDet$(oR1o?dgzp=rZ|L27DoECV3CWYMLgp>-J<KlH-JO
zR)}D6&PqxvN=rvq5J?-Qz!N80A4C!7eSUqz&D|}Hi?o%-qPh3vgpcn(X87nS%Y{Ql
z#klFIOym!K_|N!{{)hiF%*6@+;n_dnJHPSkJp1Tp{Of=6|KsY9{x$#l#Y?_(af)3J
zEG}AZukTpDSmT4@?Co!G`u?{VnkDqNyx4BI9XjDXxJo>HW07XT3bEKYtaU^$;;6->
zuTTgoS$s)4pr*iIXN6UUF-ud8($duxV-lM+9Z_CGw&)~!gK325Gsz2Ltc^o!Ma~|d
zMyit)d3nly%&fN^>+7``X3q<3N&==snfZv5an#~}RN6pImS-h{o{->)dy=OGEAver
zGd;<dZ5LlW9lva(qxbdSrSYqN=3frBzx**%@RsMeY)1CK>^EM0_$!>ydAtGy9wiFL
z@2OBYWUl3TdgQ}{=O@VsYB%%)L*FxwgESNvu|Y~HGMSXi`&S1pR6g!Tgp`$d$rcF?
zqg81VD@iw*q>hg0Tm<nm7$m`sDPf{~K_;q5n%iqRd_O`Yvn+8G(j2F=!Xza;H8hdp
z&s_xIQ3+HeDJveU3bL{IsPI@?Q&Y7os<!6h^o-`@ggo~2H&^U-J=7I#-4e&de(-F%
zp4;7?!8WWeE_w37hkWm^{S6+x^B%Tou~m($YLs!PVUQpo1)6#R1xPXFoKi7ko0|6G
zoYT$_n+hE)5!t&YPu_jEm#PZ~Nu{aWJ<YB(al%TJZJnD{C3I;NQ8Tx0YO%=kd7De3
zp|rthi!x>ouvJ0OV#@RV&Dt>aj~-^(k^Sa`wiVz=Wd#MTn-I987eFOR5K=+?4)<Nn
z&mr?zxRrT&qUq~6s4J2$vi9;ke<`1N<{&0nw|YXz6%2!C7(B+VdGKgaxagF?E{kJS
zDtyEcioT#cr?{@z(ri+TW}P`STIRCezs~r#F2%m0^Hl^Kv&zC>C^reMS)H9@mEw*2
z0tp*3HhU_kaHlQn7aQ`h=lyry;QrZ?q0zLf1(#1A6Na8oyFG8c^(IP*CC$mnNm1*U
z$?{5)3bU{!9E)EbB5;sMUrL(tvx#Vr&L|b>c0qjgymWWk2)3%x29h>-pLzN6hNn-j
z_|A79fYLbYAd1~oi~%)y?WyuUj-*)7NM<-~;Hq?!WrfOiHV0N&VUQ0um30jXV!6I$
zK_->RTI^s%Z?ACeC~$Z#qRmKVU|ogrUM#etz|z`CdBbRpvVxEstx7N)@i|~^S#Twb
z$UxWkynOkRaU5tGCtXz8^61Scm~VWGpM3I+=43@0Yep1rK6!&-xgdo=-8k;uyHDk6
zK6v|GzVYZW|L7k+W&7+?e)8$(xHlf-4P1TtoU7{{7w><IzxAK|r}%2c`q>rFKmI8{
z-`)~iO-;b&OzSEZRgEngcCB&fES|KZg^;okR29;CpFGMWx%WBK2GD(=8UuAL*{{+{
z#79X4B_;3LI8sy$eIk1BlUz>WKSIP<fgw^s7lv+&?{@TE;O_Q@{jQr*;CTXTl?$3H
zmpV;N9~=_Oxr#O+&{IaQio`<MgeUY`)@$J`t6B5?ORIiID|}uzugBLNze-S7hAw<3
z09Eq4qx%NTWvB9x`4vj&B(Y3s$eQ<rX>XnbxueD}qycLR?kLDuD$n<&?>uLQzGvum
z_-;q&w`7y0*)58<a{)CD=o<&QhnNKw3Q@{xsT|*~NJeNQ>m+0_c`|1gSqYuNWQ8ST
zvjl{}N32g+4Wst>DWH>2Cs#%3!(KHdqX;cSWl<@Rs>&pu?1@!_Pl+)`x^ZMw0V+dw
z6=~Vv?meQue3O%V4>-MiKzq8tX3zHXPa!4#<e&YQ#JfFxhLsZQtrs_IZu*ffD{vKW
zzyBd0{Mrxr#y7u1&<3m|o|P3%(Hs+Js}<k>?)T_Ee^DeaYT8OuwM)jUTl8wdy@wCD
z7+U(LKPSgMDJf#o_+cc)K-YEb_B)I<MJ=M>blS{~3li`azM%p|LEfx6NQWIwMnx`s
z>oQ|hk*6{SZ6{vyt8+FgC6zoiDLPCVbtUjplg4TRB$7G+tw;0xF#D&+%8%6Xa+@Dg
z*3i7)O;JKBL3^H&yorODGek8h6C8N(B9T=@qode-=_!y^imGbSdLTy6cDolxCac6&
ztN=_?V`UsiV(?71qM*>$;_8~JZg6dl(Qb}aO8rh&(%0Z$|G7AJ>4;FSrtC-|!JQW3
ztErs0N32$aVZ>RD^$|rxWls)0=cg?XE>F=q(l&;iMv@;eWW4XWymybgt7kM#O<mXG
zK9S-y4d#pR#WwosXR!z&rqWI<<+KuPPmkA%s;N*cNh#3hObp_Npz$bWh*6QOBB{h~
z*8z&<Vnxc}OoamBKZG!6mLRVof{&)yxG1qGW8xN+gd&5^2AxYrqda~}&66paq<JN?
z7ZFZKDvE|)K~j<uFI39GkB@{D<nT8dSBd9sP7){@tqWo$X53jv;ndCblJa@9CPn#?
zwVIoo8@Ag!l#0|%MO9gHGE^r^9=`PkKdCF0_wLc%LF`AGy5g;Gyhj>FbX}1!G|dWC
zHGKGMzsaxv^}oeG{lEMlFfsA7XU{mhYq<INB}!L(>(~DZ?RS5j7q_0%w|~IffA=5o
zz2TOZ&p)Pn_BrEvP1x@l?lz3wo-mGdF(TV#B&Pwb0w#mXksL>C77hnFqI1S3E9hOZ
z4l}Btr7@sWnl-CbfU$?JhcIeF$kM%=5e|k{Xj@3$rAZ_8d-S-&Z+kY^*9`mq6<}dP
z6O2kCyHXnXl9Df`%~z>e|4ycQDYeW}F1)f1Caxex5l9n2IW?#K@_4ll{CfP?IgSFh
z$trL9k~XeadI0B0e!>kXsLOngnYo5}6!gi7n2O`UlsM#a2&!BH$Z2htT|*;Y<uewm
z=669(SxT`x3GC9af(i%&qySpVp}HjW%4&?A4p+f?Vup}2DrRDi7z!9bLDP&Pq=cJ-
zx)jk#P`wEqNmC$_LHH=3#35D6ex<o$mR!+R_RJ|fP@1SMNn5<KQfgI#)Fo|DRYg-(
zEUJq8ZHsGaRNG>j7P~rOd2%0TT5{Ixz2`P`oVSjAe#z?Xcc}jG59xPXhV`Bfg{~ZT
zea}8*M(wCpCp>xYeIC92F1BtZs4#-i6bKfpG?lSb&e3Oz+kphVDMUQ5sB4zzXXxuu
z#Pm(kZjBIwNHqB{h_+_kph~GU1=&Lq5OXdRoI@&d2%SohcU1OQMFUS7(qt1+s%Wm2
zN)j;Bd{Ro%K%z@CqMA=^GP*P|C={l2!<JxtivKvOSEw&=Twi^o0)k9s+G(AT=KXd4
zIxGkI^|Dvy65MI6iBlX?=GxR85>g^W(LfXf1Vf4nTLN>XM7hEHP@0KK;1~scB^JA|
zGzHQcYb;eQY80lbP+Au?j5N=ODX>dNpw*u>{#=ONG0}Sxq!HjrNvROAA6YC~X@k}p
zzg%MWk#PuYc0CczqsLFESEr0&AR`9tcQ-dwPSfxAq_*Pj_Lc{i4{**ggjibegLs-6
z!<Pkg(|fbl5<-A@=v19H%_$;JI#X8;SJ$+wRmo60$xdP*4g**v9bej!B7~Tk+#u?z
z#wsm01ZE{{Q*`yT*2TlNaJGuFw#1ZT!9zv}!jiVrn~^blaUrq7$IB^0Oz503N=%AP
z5{{f8b_Iz{hYgd$X9<*MkW$Z&S;(@)j7FiYlQMD6L^T;>Yw`X|(#8}qXjem448y?f
z?XCE(T7`8QV<BcueX`<>Z-0y5JwNBiAAQWd<s-5-Jb(Uzk8f`H5B~e#MJbCm3Zoq3
z(BayOAN<XKkN@}o@qbV1NAQ{#FRp2l;u{})Kv<q~r(6D)|JVN~FFv{B+i$<m8*e}2
z(OZ{1di%F{@S7K6`J*yzj;QUL{pN;lx5n=_sPzWbkHoHHzgaWvH>5a_Lr3s?avT_x
zMw?7_meFKtYcMII$ArmQ8f^+0d>9llRSY2$1LXh{85-v-byI`2gki*ok#T=Z-rq3Z
zZrEO5F$_`ypLxUxl2{wL*GkI<8kHy&X23bo(6Y{PvRblQtwi5%7zpEVn9pUN#~}J+
zDM~c1a#4Oy<vB}$_sZHn`tR#8jjwY(@>heMC5T5A2Y?cnX@ozfj~=Ke9m}KNef6#A
z`9Pf>7m8ek_#v01b~Ym~ubHJ3d8io3dq_FY$=&oa>gZ+t5E#2XzT1<APUgEG$uW>j
z5n|QM)rtAAFNb!{O6Hm>N3K*mnRs;~4vzdh=fjtLR0+ajVpNJz8-ljN`Bl<rA^hcv
ztR$Gr%Ajn8v6XPYv?d#ivKFiZCwwMlE0lHUrX@EEv2t_{Q#TlE2uc%`CSbVgEMvE4
zyX&~RyCqxCciwx<sjZ-0(wv?%4g;g_88$mqU9k_5KEhrp&QH%dy}Zxz^c;$CuQ5(4
zMg>CCz&H$Czj#UZz64<~#E9ST$;MDGmz-W)V0O1CtI<Z4Ul;UKVuAqoKnTCYFbs_2
zh@JL!n5%75;6G2N38Ya|_{CiDn)C%Jj!nW5E5BpZe4jG~>B@5z@@N*llCsRGu2zy(
z$ls-smCKwsBpD?LRL3h(l5V9mCY!mr_VAk%_k5bo`S4tm8nXb1<snC9S&mTisZ3J7
z6h&aWluD<wK4(a=eD)-PnMp;|I3*~KN2&y{T8pBGttyO@;20q0P=LmxB=4*Cldl+G
zuX*B#Hz86TUnqCx;-iYd>R3bFEI3&%@ZFx3wN#5G#yYBMfw49H-3@2wC&b#|m7=N)
z&Pa#vv(G+5XEB;jG14@ZAR$KMR3&Ayd_d}6d9PpiyObC$St)$-B_KXD5#$^RW5gOm
z)3g{RqKd<Qk5WlIQWGKx5~U^%nyw#NEL&3O(JIq6HOd?~10ZuK<{~~=G8m0?pyn)=
zGo^ikm?U6SC_Ei83(q5A$f<m$m{2j|v?6M0m{3Ik*(fKAn@kRc-;kx_Cg(&hc55m{
zvX}ODF^JX}>%@3i740@tP{J5W03a4@DR`>7VZC0{bvsfDG)+xYS7-w{YLeD0&d)i0
z^oV^N_}*{)HrswcY0b}m`V*2LAUKq(P}bt~iWmkSf8#^m_{N9)^k4iFOl`=+z=v<X
ziCVOL`tpi5qbAVs^6HB2#n1Wu-}xyW1NBME#rZjP?KoL2xPNcOYH4}&=!|=pE0&k{
zIC;EeQ8hHS;#?^zmC<<euw}P-$$q;g-d?l6eaUrq%ibmyVdNaJ%7~>)$mr~(TP`U=
zw2VccOFKi%h_|9O7-vvc<NHD+9(Sxi`x)Dtp8oDe0y$*UHs!#VVBJtv6^*MnIXU6%
z;vSFRdV{y$d51@j9&!J{10Fwq%v*20#p(HZX^`=3H*0pg9ev-k-|e`&yJNfEvR<#{
zUiCdMZ?E|2&ws|xe)bXT^_mc*Gz|v?PisYtuPx4A9bf-CsIS<Jr?W%rB1WnufpXf;
z#D1+rH_bX%0z5UB#q=yeVP+rCNi@Ch2Zmu_=tW!2kDd^G@kcHqsEN|~ci_TlDCN>P
zf7;_BmdaV^41GuFcf@`toHXAPqbFKvM6g=4)=Vko&D<&!Da+U7TdI;q6&gWC#CLN^
zJ|y^*qACiKRz#)nDKQwuXf0z^lMBL<v??6p7G)|h4y7!nYN?tQTY@+$OLCU%EV-$v
z7AH8@V5<hxE(ujjKp{J_V)P@sUB_<Uv$@-I{d~jvX3fp*ny1&#IepM_adC=URd{Ww
zPtP%%HJu;W@AhOL8MUU%ihYLD`wv;2Ur;Sh$i|8j#3Wo#F7jL=H2lfOA9KGEX)k5P
zDR#Zzu{b%WJw4^@@{;)cW0W-#T;&NIQ{=44DdWeHaU5|~HJ4*WDodnFCUjmYHP6L4
z3{z^Bx1t106wY;?g6<>rfkSY2v|6ShD#Jtzm?$Hv5XXczlIHi3bt>~OO9f<#0;iHG
zFl7KrXRI>wd{Jql6}*ByAN_ZR_{#Tm6nCn_e4m56Y5pE!EAgnGlBt4HmOys;MAKT+
z35cQ4ex!p)l{Ku5mHXl<i6^uc`cw=OWK!1p6kmOf@o$CLy($H9+&Pkv7D4Ig6m20*
z?P~Br+#%|g^ZSpa6gKn>+YPPLq}Y>^XR&NhIk3O}l)Kw2&X+A#D<DZ3HpH3auCy)^
zA9)V$j<w-l8FR_CY?7s(F->TI6ar&N(&FtL&0<l+1P3&kF!rDoF+khaXrmd%o_0|Y
z<A5g7HWkh}fm{(wF01Vv)lVwlIf69?rG(qz{Xh(noW%Ae`;19ovIXtPS$t`Yu@aO8
zPg0p!1)`C|Sm%UFBSsl2>u|=BQoyGqFn;;c)(}(1`#|&&r489+`LofY4XCx~;H7D?
ziS$a!36iJpGwby&qaSe2(KHQJRZGTh9okuJRrAKX@A1jAPx;f&K0_PFw%)T`Eg8q2
zYOy$!YP2D%NPF*+-}sw<o1gvezo2Rz$ttdQJJidUsCvcj_LlQ{$*FF!qvdY9LAi<?
zD)znNHthM7HLixL&HQKo^Zzlsam(Eh_%D9`6Mp!^e@$Z>+N$Nz{rg<rJLTd1GcGS5
za`EPe-20~E&4nSRJN(rzc>eqUg5mjdES@w5a!BN)h|v&&V(`#=MM@f*C73tP;3^9S
zl2V8hT_E>cHcx+n-z#Fbm+V|AD#e^oJbL($AN=6g`R(8SZNB%N@ACFLZ*y_~K8uqj
z?P4KbjaqE$lsSsXrmRvc5x?{z{W6o+lVs?d-G=qunm_r;pYTuq^MAtc{O<4a$3OmK
zcDpU(DEe5S$vMjAqg8P|d5z=!%0ta(g#7gjf1U$71y$`>EEb%dopO10LFKBlo@!1`
zPH5YfrfFC#7AzJ^>bk~Nj;g9yELSX7C-b3=Qf%(-xV^pQ>ctD5J$uG_y=J@Fu-R<5
zzPjf2`kH><5r$C`%~H}&Ig?`)L=Ps%tuKs>-)|W=w+stM?=@-c$SRPt71V69=1X$u
z8)If})f6yEbCwpt;H)S02}F4yLzD(ITME-jYl7D7fxUC|N#Rw6UEIShmY7C@x4K=j
zJh{X+E3|7sJ6yfME*6-ol1@;o>Eg)V53Ku+F&p~M)8BdapI@=w?77=^?7D${H?Z05
z*lf1!w|n|+k4;iAk0kbaK%cbuWa$IETyOANQJtRRa%6wCXV`amZRnL{s2p#9_)V7g
zE-_U@PzG1q+3lv7x-#}1o7-DH|LjvPKKKroj~;RF?1G#FcR#zMUae?PPiRh7Y-VlF
zDKInS<nRPS2#jH*A3CbKrfFKS<%+_=Rt0RBFgcWxbEE_<lTiLNsb`(fG}&b_Q+wvX
zPOJH>o$V6KWSS_CMZBZTs0>V=u_d63h||W5*va|xN@If(Z4Of^4_Y}iPP$Q{a98sT
z%w;tt6q-DfxJn4BDJ4&7C!G5;;OI3G3X?V1qiV4>ni#~bBWFY!<`~(Gk?g$$c5|u(
z+B&MHp<OJ|PIT%r1!<4cnXM5rc>ZE6^DB%$7sqZw?B;SoDO;+PG0LE_BF8L$&*1kv
z)|)$q-In_o_b?=^6<(Nil68(@Bo2GVZpWgnsjOzd-*K{9m5c+dv4m8h(J2<6%MQm*
z4_`nWCMZ>D)I@4SP!g3g!x#uD6KxWY(W=JQHJ$IsQ_FR7qyTlZz*dHS?0_YTMCBBV
zWg|B`MYPf6Y>vtjs#N_mCW~fXh^5&A2dN=?LyDH<4aryJ?1-izs+ztiz)S7kV|6Bk
zo?t|%(&>aHQkg^=J%%6_X(hm$xH%aF0uBk2i#}e+C|78$#nk#$B0@>LvdD96wp;G*
zZpb;(G>hWN>dG~@((11b?|t)|{O-T{eg5R5kGQzJPg`le{r&HVsACqS6lIyo&PnlC
z|K{J~|MS24UtmmT(bT;4;33aGzUAf3AM(LZ|Acd=QNxZnIE+qs?=fm28%Gd-;GysN
zdw=KeaoV2odw=jFlz|W5`8H2~am&B{;g2w?;!pqZSy|?RoCoTrqFuDye{#;lHyZx4
z|IvR)bN?yN{`lv->&&d36;dWdMTm-FRE$BPwLM7L6a)q>kM}7t`Vk!msx)x9I^*(W
zMeWu+c<_i{`@s+RD}VL3`Ky2Ruko9|`J0^GzrY$Rn*+#ZnJL+EQI4N(rVvA<lh^2C
zZ4?XdL%2oLW|(85SuMH$;34n4{Vsp!zxQ|X-t*I+{*-_I&;L39?4SLY{O<4m9v^@5
z*-Rdu#!=I)ETuos#_`{#Q4ZBZBa2+5xy*Dp>i|VvSJZXQa=GN)_ul1$4?p1X<0m|P
z^oaZSE~$%dq_vj17F)cD{O+6tbt2<qu*MZrc0DJpF?fbypzr%xS8?cj`o3qqzGHoJ
z&31jq-R+uBKly}DKmCmL?JakkI|%~!osh+I#wkezMVn5Elv1~6*sSR)*z6N=*nt!5
zIU;gMvcZU2g0j=$B<Fibg4*e%4!sR1(X`8A=19uq5Fv_FLPRra%TQG`_aCuZp5dAk
zn)VFcoRVBEdSqxxfCKoD8GPdUdc*G1r|gHGuG^8dVzFwd8dnZ)%d@9fJb!x4ryoDZ
z*p{IW1fL))V#*A|K=g?`ih_YL3T-THrCHWXthPiyvfXUSrSW1&k+Ia<v(eb9rarsi
z@!RiFw=0q^GEB-cl^+$F9K<&?r$ALXGGLpAq#Y@F>9##PV{vwdYg;fH>l~@F2aqEj
zRn}sp?|Z7Mqiq@#X6|Ax)AyD6%LJI10)SA;iYd>*3b8Z>k_%5g6{QCWcV^@JQij!;
z1JaBl6;b0%DkaxAmxWW#7z{b-nL>n?rauwsO>^l~kwwU!j8}ocAbX@al$6m5ZJa#E
zRCndzcdLr<?<@OR*)LP!AWq1(DF~Zflc|v+;X|P`p{QynY85%5Q;{}_zMzs4PHN+D
zRgJ3zsgqz<s1qWZf7n~*@ii+ae?Ea?4wFJ&JVB6pa$%5(S^5~LrRYzkGV80SbT`j=
z>+wTW^o+hI=aCpkQjVOiYPNlc)rz)h*j(?#t|>-b?L@F|>i#QTC&$FXsZ{(+<A^an
zITDbIUtNq4qhvG4!nM{=H8nBTBtK$w#Ai=R8ExR?<b<wUlXJkRL{k|St%&BvQXaKd
zO2-qNo2Qa&_Jm@Q6{99bi{gY7R`h$r;2k$z#oYj#A#vL$?)tzGB1UHv5liCU$%19Q
zq_T;#rQ-YqOuA)JW!f4H11blskpMhr2@YLd38FW6;vgKXFpgMVVTz8BwH6<Pkgg}R
zF-2kwY&N&_Lx(Z)I8$2VMRvnsjU#1v`1lEb@9+P8KKuC<-7xU$fAd57v7;Y4R@tMn
zlg%%QMVA@|9=!7|?|<Wcp8x2F<Qi(7`77Ulk3YTX`G5VR{~e!?E3RI8n);l<X$E6S
zbxW@@o`^A${X~=b{eS&K-gxUV>-80X_y@nkU-|yG`Pm=+gl^|iS-3bpCcr_~?01^a
zU*5ovule_W>$h;Xp3gqJ=FK-Bk+o%1iXjB>niwoTWMVRh%|wg+jwuOCh{6RO`VQak
zG5a;m`HH{ycmEFm;6MEif9pT^`@H$?yR23#v^7PbT#(a*ome47&q=b^X}P7sdlCMo
zDys2@X0c+jQ9GstU=2D&LiW_nf*Oe4^H+ZJxA?&iev|*n|MWlM7eD`q-}#;2<^TJi
z{}cY{Km9Lw_UtL-G+Xs0V4~Z+9$z>VDO3xXkuPI3&RLqK;o{<g2M-^V?+^Lln;-Jl
zn{RUO;)2uDGftK(oVDmVIk2S!duUKIMc+!#yyByFi;~DLCtucbu*VSaA<*|7-EK$U
z^>n*EH&<6YfBu52>npBaUh(wv&-wJTPr1FhWqo(Yezzq=DGkb5m!&EY11gLpMu=W`
zGbtTPQ6&+alm1ByvXo*|65@$SWyxWe)YP@jk*}y|kVZ7oIO@v>tls&6=$1rkx?Hp0
zXF5O7`#r<p>HA3E2l^q<^&{);j(+GF{J<DS9z43w<7R=$j<N4JKfA|j1$T8|v;G)T
zB6yhHcu?d~kptu)L5_8%2<Ix{!^TMPJ-f{v<IS4T?eQs5FB)P@1T>~@IK6+L)B6uF
zRXvv$CqCDdHK{@mLV>FsFR!l&cab;*d>Dxt7AGs()k-`>OJ}dvR)P>9TWPp3W^$A8
zjN`yC^i)+-lJk<Km3uV@5sJfkF3%=<vC>IEPKRI3ITx$2@}r4*ln-%;)a)jI+3DX^
zsUE2!on=TxQ!u84OAeJZIcb?E+6iJ;b|9^CDU>Hv6mfoO)>H?Lys0TM1FERWD*Xyw
z4kTGVhS879$%jQ=5Ye)hwT4N9E|ZBVkjgc6&PsW=bi&R-^i=vq7Y|rj#~G2(I&lB<
zEK@b#lTwNKI^)k7*uA=`NwCD+b)~a(da0NoBy<Fq5;=MBfz8WjR7zUbhj9-oVrvUg
z)9<!4&SAA-=mt(!%km+mt5q4PQRgIlGoAE20aLMsiZj`<M8SEwA$b-E1myP$vZ%8B
zJbEwCtx*_P;lqegiWo+;QD~DmKU?wi*-Lx?l&7(orm_e+l4p$_ICM2iwR$A&*Qjg>
z(J^{Qhzk<;SZ``R{%ptdn+-Shk_`s0YTV+SzJ^X0CF?X2$DY+XqxM?}9VZLL#VYgU
zqT&9D;jHaBX}}rqzN50XbU79Bs{x!s2SpAuJXZ*{`4G?@E%cNEeZObF-x6b>s@i#D
zG1_6QL91H)XRDfI2j2hIw|R8mv+ef0|E<S-`te7&suJKkN9+WbBU~A`X!!nb{U*=;
z<PX{IcMSWUlXJ&gmo*>$;BWAh3;y9h{6~ELVohh@{H-TM?EsB(rL>+r*WDH0eETgP
zz4MTFswdd`Pe?gH1HKQCH8IL&Vx477nLH?L-C)fLKl;(n`OS+ZyM0GUks&080KM1b
z!7v6v>tY5S5;+#`ldHis47z4WiZwM`cS`l-lyCgtx48Gg_i5kxF81;*lB;<3`KNs1
zM_f~bQPfRCyKF%ljI&g(rnZ%I&K0)D9JC?p0i+Bd_pX#xl$p!1s_;sbQX<V$hIUoC
z?TPcVOU{1%9^d-bclZbY;6LWaKmHN__#gjc{%`;H{|7(+*-zQ;J2VIG)vw;8_1~^>
zeD3KltPfo{tXeDT%CcCrELThJU0m?(Z+)BhKlp$*-h6|%-g=w2-+hO+X@o~-EY_4G
z@Q710V+lDEe3-d)LWD1mIeHHg{HekLo}Pn}%%<6jkR%1EejK^Iz2)VLmwfz-U-08U
z`Xhevi(l}`$Di=@v(MP>c7$;xdyk5q$`~pQmWU3ccw8omtWut4r&M!{E*v2}2QfJ-
zVk%`Z1y+NyMb}O$P^A;G&<B!Yj1Y9i$%7{>pS;V9P2_q#u)FPe`S~@scUv~Qj_tN%
zza824Bi-QXdQZrOYh}RJmdl54VC@2HD!Q?wtxsqloufj-ADW-B-3*Wf^`FbsUXz2C
zj>w!)R)Q#1I__iN(e-z1uCLi&U1OC;+01g)&<}ypXX<9b!^clp-MhfJ>QJFj(+w(j
zy9hl`_FT`Oy<}Vmd^a+TJ#MM^|C9HhL7HXheJA!i=DOFVt1MsLUG41{%=8${U}k_J
z3`u|_NCuab$mIty7}6^AP11@RQY@t)#npz^d{8hpLfT-2s0d+61`$gKfCLBv7>{w?
zGu<;i?aEhXrGMQu$LYg4_sy)vfIzPVVnN-BoUY2uN^{?H&v~Bz^EbosgyCqsQlm<h
znt=ow1O-|LoYPq8vDVPG4Ruo|geU<Tsp&g&LbCGE<n-t2TguS;vW3&pZ4csyR2Z#<
ziD^IYpNF_-kzW7kz~DkkGm3d4xDaqI;H^eF1zzG-;pd5eiAM_o!Xt6WkR$K}6$mP%
zpWPFLL&#+8??TL3m64MtlHzpb-F=1|vC<6m&vbegBVs3GQ7`3cgbXn{%y|a_@;r-3
zA`U`C>_i$dlbUjsX;c>d0i6$6o5rsMq=@{~{_}rVPRH+Yy>El!Riz&KhgYqTARh!i
zy+|QA@CNBEgS;Sj0V#6?I(B|QZXHz~MFe>{z~LZEDFN0}Xa$`G>(N4`*}g~R0w)|!
zX-Fe4<pU}aMy+$PXbFL6W8stKNj!?2vvG4v(6S2-Ekra|^w9>x2a6;lD>NZ5qA91d
z@!N%fmWIg?ZnTbNQ!pM1MiR!E#Np5)Lm+X%febOeMu4>f*LG;(=vqne6B=`lljR0)
z-fel%K$~4;KJ=K0XSB7()@Ys2eCE^Ky>r5^|N3trq~Nok{V~4#o!9x`OHcCP-UD8L
z^$qqL$L?&-+xt_tOUDPVZ*yr9$nyiLkwrOA9y~#KjP!&;MI_H_gtZuwXHW9y2_fdR
zN+WQvvH_`=OS-m21jv*oQfw2VAT&D5(L$n3pwk-fJj2m|I<9Iq2NRUa2omrXVLj3b
zWTtUJlX=afAOAS*zxeOzj$1mb8K?r)F6r8u$3OHO+n25}8G;F(pZ^d397Y(v{*7<)
z$hAi~IXS{~4IAqvo9iRIv*h`J2fGh=?AlWds)D2aInII<Ii_<cnMW@^ZKwvC#<cA1
z>@vLeC><?E8CqlM7J}tmgDq(s7!f-%t&%l>%IWfw-JrR9VmL8yVut9)K7t&r@x^6J
z`^sC$@7<tv7K5U7KAN>ENeCXT6=hMNv}QaWv$1}jb6XeK-agOPxh>A0Kc5PMeAUSL
zARtJBW9139qN5Ny(`a3Dx<QCD8+x2_1cH$0yyV=40pqnbKKjv*@)!T&U*uQ+)xYEy
z|M4&K-S2#d`C>`Lj8ie5=nXv&@3h^FyqSOBud}odts~!bJR0)MvrqBji!buXqmT0F
zqmOd^@y8jB$7ETS>YO;jUo}sGIFdYjxc2u{CIrdH#MA!#Ob{8pm+0yH`ct`Yx;@ca
zLq+$Gv=<JFlEHY5%hw*|r4N3H&;7(7V?LX6d~(d)TetYeZ-1Q|Z@<mGTQ`~Q?Xa~r
zVQpAYWdd3Qw!=w{v5CeKqFTQ{lSDFoMM@AVq4SXxFFe=~`E<Q#(2J<Z_cB_Q7#o;%
z9j)jXiXr!Qm;B~eU*kKk-Jn?r?83!pNAI!L0U9Bquq_bpXAqDp&Be`a9>08z<NXKh
zcMB$yA>V!FRUWze1Z7sTv9`hN=zc6zW5!S{wv(Bt4>pFr3lm3X-s0;Sciwsff4s}i
zjhj3=ih?H-3=gJ9E2MKU8u0j&Pg0d-BFT&N-(IPlvPFYK$UrqH*}ia|t_dtm9e*D!
zICpWv_Qgw#*EjG&M#XO`^ksTYr3`Uwn1yI0Z7p5bP}hpQ$dgsanKjlwuHy6^A)WK5
zfJpj#h>B2|)}v4gm1mK5M9&V95iv^(Hx}w?U<NpHTu?Z#FhOCx#sz^3GA>J_30B~p
z#08R>P5iksr4sEZpoJ&HphW=Rfi(zg5y~O8Ln@CHHuBN&2;osOtZJ6iLbU&_<Bd=%
z@(KDl(9pM60*Xvi<O<^wDu|E5Bs@aK`|G_$g=FbQv>ogJz0}c1R;ozdaW;A~dhZZ~
zmFI1r+_JjP&|rQ)*ZW579-c-&)NZU61E;qFhhJsjIsg@YdpzKTq_LJvi^R*2Dfz%h
zwWiiF`M~=qE%t$SSrb|!!M`XGDjE6;u}VY`LXim>Z5<rQxTnQtBZMdfNcgcAJ?BnF
zgd_xv>997&MoCdnWI3)F(D)@$r0XzUOIhX=Wy$gJl;NO23WGN;jwTRM2^($B!CMFl
zX%yHB*9}=N2DI)9@7&9{H!WFa7nqlD5*9r9fhW28_%<)R^aAsFOV`5nr!VuhZ&h5r
ze32jd_=gy*Rb09>VdF~0wP!E!;NFaPUw?-%tC{WZ^7VHXTwNb<{pxipo3U9g@!^QF
z1Q~MhL!!icfh-d5MVumbkt3wgN=0iF2i6#jF{?!{dus#Pn^VRIS|(nEPzr%YWd&=S
z=VEYmQ8BM;az8+d*wrZV)ks|n$;HPXr&?R1IX)oER76U>V=<pHC`%^m8|=;J49kM?
zaKsBAewv^7@lPXVRDG?lud}ne!~Wg_g3WpJjoZBV{ELi+YdrDzqj(F42XnO6blnW+
z9Xc0mY_4-Oonl?!?!6O&9^$pcgcfTXg7M_!cp-4X&;*CqV6~vl6w_|O_is*F-jRg0
zOHfTv!*%SkV=>#q3V{ulwq3-sO6h3Yx=Kokk!VGhljj-p>73pB2T1vL+N?9y)+U@g
zx6S3tmpOm_JnQT0tZi&C7>!a<=T_DRg0${kLZ_8_h`P}icR~cD>^C!wkVLx{=XmnT
zCwb)BRsQ(T{0#r{U;YaJ=pX(=zWK^`X*!cQSRvIPXA8T3|E`B7R%a$SQbZcT#fz7C
z_PJ+b5qJGDo_+2awl7>@G#XJ=6?vAeC=E&|P;x~X5kysItcW;0P$jFwM=Fuj?LCbI
zq|^@~C;bm9-uoJeL}c%G<|Moc^i~$pxV9RNDXIY*8(Unxe3j>){{Tma`|R)B=f-QV
z5Dspm>O*vWjBA(JuEh$EH96MWG);-9UG(mZwh%I*cD+$>2m$Njo)XFL0b^nVTo9uv
zJreH%^JUAzYSx#I`|~NgI}7&rI)WLXe1;Sq1d9vNeyVp^2?)G5NEImajO}ylShJ)v
zHChX<K5~T{H}0|QmMj;B#<v*d=t&157_?A0X>g&VQytbjIupPftXm@AIilHvHQ90P
z(k53189^AP%MROhm@c4X!FaO9#p{nHfq`79k^7g)6KO)uTRLM9O0scoi;a!*+<#|}
zt~TUF#)Zr4Y+t-US(U5iO(ni%pHAd`e3qq*_dX`(SlYJ1n3_!I5yU{PM*OF?Hz`R<
z$W_`w=zZy|joh)6Od+(2(VkkSVOUaC_Y;^9aCn>#gcwbL^_r#=Sg)`l$M}lrBG7b!
z&IYVgEL%@!9Gwj@5kz=GDk8M>RGFmAB&CK*XACmIAj_ySPp&ORnb^>wqcaPT4k;~C
zdW4J-l7!?bEBhBAtRC<v%1mB;>G$tBe_AVwToF1)U&`9fAVT!B%kzxjy97T+<WMUW
z*OXG|EKapzQF-Q`twPHKcon;z-ZgpkSpWVjyZ^7n6Q`4ckYIU#`aT)sVY~vX;@`Uv
z@Tph`MUG1dl9U0RNkpFGU5jxp5^;Tq-bHPTop;o28xNhTWH=h5ivqF?w8qO26`~>W
zPOQZ?Ey0=8p@_);OL=vqxp`E`+PLA8;ex}ok(;6rWKvOP8K#Y$srL@y9aS}EJRWg$
zG-I%}PB9op#3W9}aKa;Xz`GXL8G_Yd3v4r{v6nfV=KSEk<9?GfQzP_b1G%pF#E*QG
z8+YF3;?+kuI9M<}o^j{y1GJLNW>a={;TL}4AF;8qPL@3ZV#1}Xo1DK4Pd@uN$}7I`
ztN)t48}IVH`IP<Jffp}7PS@Pz;<$lsLFOIGC{$h$gu#j#o@hS+5e}9zB92PyNCXW5
zV+_{XM4A<=!frJRI;*A%tMV|$(`bRr1X+-DGf!s>!>mLQ5-%>o5FWw$_64>tUuAmh
zU6k-R-+(uC^BLo^U@#ux^pbho@s+RrHcveN2<1RCS*uts=Xe^dZF%kWH~7epd>j!<
zCTnXL)AG5`f102AnV({@XxV$P#}~f%MWhfs|NL{5S;as2NB@F(qj+%S@M=Of8q@j)
zW9!Hvbrx$4Xo(nPgd%4qJv*UcF<;<>L`_DFkyI+DUN$V}EnaABa9G>nZB$V!okqy1
zxfl%csJitccC-d`7B`eU&yz~B!a2vyn>V?2>sC}s77<;3^2w*z+}LD&eS<735Xmno
z0ygh4`Rt$L{$UoU9!uUil!`G$_>L?uxcbN={N=y$AM&YBeTrZD<zL|!{=0w38$Wn6
zVsgHh1@+haXfOW=TxUTIu{r=1qw$#a^>r>@xWEf9zR1(hKEtKUm$`W10+%jbVlW({
zv&5B4G>3?w#_^O)NZV?h+aqlKH>EGWW0kiW(58=t(@%ToTD@QN-ELSNwmzlOgv5mo
z$#TS6+e5I)tIWl0Zgq}fnKP1(W_N@;-oYO4(rV91*Wv`ULeU{HA<B`Ih(_lky{!yp
zeMrjxWLoUOi5N|Z3u)c=nVj+9v&Lbh!+XWi;f&?d;GH5k1rD4Pu>ckk@rhKRsKWHU
zFhN0HWjJA3+Z<C43I>w_<Mj~&K^Hn&-(rM`e60|T=&g|WM9lAO&2m|zoks?b){b-A
zm$^8ROhQe0Vi*>RrnTvQI*f&ER4^QmSwFXhNGM(4Z1Ej;$@Hoql}IH0A*10sZmBSh
zrK}1z&W|XnDk45TpEKpV3N#_O)ZF&`#$Ev7B4A;<m3>Y?oi6T&|L{8>-=5Z##;_u#
zh4iY?kT_>O;VzE-J)qL)RbqprbDBk~XpN#Zf_dXOUU=Fr&{=^;(~!}5Faf;KL~$Ll
zunK}wN;2=sI!_@Tl?n`{W2hX%JTRVUs!EaP3W7mMo1*IiLd8@dlI@tt^3#IpZ0C0N
z)#(8%5pB~l1)YeHa|i)zV-cp5BFhxQ$1yBteFA&CBBhd2H=LRupP`IJl68m$XE+7O
z;?MC1RoJCYay5!QGXq|^Q74gdNdND>#fd<cE2PiK!XPS3oxu<o<O-=E&m@r+5|B1f
z46685!l|u8@G(-xb}^St)|RZQP@@V41wM<T*+4|q+GS0*T&^Ha5}8=^;llkJB%VQt
z@gAW7A23~@>kJ}5nL?ert^owzd8GFYM<aH2PdJ&)8EmY@<OmnA!r_(25usq$7<?xX
zR?v29+&)_4or5{kyrkRK$dTmpKmVir`WL^#g=?FfyRgA4-}(V>eE(ghCo`P$I2*g6
z`P|_AE=7^^+h2Q?!LVYqHsrY%pXMVUeSz7u;lrQ%ED!EJ$+y1t9UdGm`R;AQ)0fZV
zj;CB08YZOx-=HM~9q?HmM;d9dDb(svkgLrhrY@`&<9lZQLiGMV(P=zp464{kCO(O2
zJBq~pnlI*zN0SuW73JbSal{mw$DVqQ`>%W*?^`-w)6@&v<3n<9+1OaaW&2p~`N13S
z@X!C*ukok;^v|UAVp$wbk;-%N;yI3wk64?mv8)%|y!kHU@q`y&e2ME%Tw!--z{{U_
zf#I;CX&Ua{x{oR(W}0y{3mh*4`EZS;*`sqE&O3t25rx7@!8BNox)z&hY6UZ)up*<<
zL!2ntJ(^*>q%n?V-O@HSwyV*?GbnR9>&VNJb6e-hvK;SVV||0iAAgd^u0O%WOP838
zCk%%JilU?_^2q&JJ>!ASSf<lMe)ZS>1+TpF4ZMddSFZBH3omi~`V(wzokOX913SBW
zh13wHPH`1T7)N`~`D9m-<E=voL0KuDdioi*&TaG2mp{he|NDQRU;5==X11v5I<xw^
z??v<fC|~^@>v`5nN_3_fjV4@s<Qk7&e~cGic!7_;{4$rXTxK*LQxwtuN3IGrpZ1;p
zi>5~ChZS^)hr!P?2;3=GM?SO{ufA44Uh2_4f2ub0$?wp5i**)jx<s#VSQ8PQU?VLc
zB$MMT&tvjK@N8^v;xdCB7j&bNDS<`P;cZ8&6iYHRiX26Tpiml;KbH_9T9x_!;Ez9|
z^I(LGBzz0O1+0q^eJ-Kc&P96-DK*E3hnUu(lDDUGE)~JCA+aGAT|&snuSy1IUU^im
za9FmtH<+xC$;yn;`iQm(1nIHC;H9PKb9wJ#ii`%K1i{2TOS?d>CF3IJiKm|6>Y5^7
z9I$MTsFa`$0cRbZ^_U<T4u_03)>&I$k0_Bi^Ewh%qx0U}Bh#9Y=Qvrhwy{aG4X$-m
zRY5h$B1$WxM_>r4u#*~rj}4qi!Sdo1i3={8m3Lhi#a&XOWIF!S<JvbYvA&3rmZ!6u
zGs`i3ZcNxYQxAkVYwcsSoOgi4`wZh1tyL^d#_>Y3Xe5gUj^~DH9THbrq01aw6c`<?
zadcj+IC%m=m1VeG(wPR=)yU3~<Eflu;4PCvuo*HYUNZEGOjRgh2{M54fQ{Bu(QGl=
zhpnh7Je<Yc`<|Ss`;-#oO3=zUh7AEa6Ms*w6}c9Xe;Im_Cad+?`%e;x%~e9)SNAU6
z+p`u`K_o9A!73*3_jA4fQS6zGWVL<w?@N-Is1}JnZiqluWDKj4YCL9ftm&G_m68%h
z+Z&88orikPa{oT_*_2@x3FO!gho>A4s750u<1wynab3r9v7ntVaK@m5jYh_U5|!ss
zI&Can+eD?OlB-CuIAZO&2+6NUMM}630xcy%X>^_we4uL@8rz^%wD8bbMrXU^Z`V<k
zB>^~|F4&r^#n`dd&}o5@0dEb;I=s=CwnU0e?j8wV+ihs7A;($EkNxZ?*m&du)5VlO
z^(X%r-~RG<c=hYw<7iqFq(X>18sJKa?F>>BbZv_gC8nJb?1F>CJr0g`xN-Y!);G`b
z<P%RbwT^`xqDDFU$9sr9&r{o1u}3#iTb5xl13M!xDuk$z-r%hTSSho-4=Yl1%v0>s
zEy6>ueH<l*6`@}!z49IkeZ&ERqk{v)a7geO^`eP|HW2{@kqE&B1jn_<p5m3k5Fad^
z?*M}~4a>s=&R^If=yy<F6P)0;zw&*$R`BspzQp;9W2%9q?Ur1=a*pGpIX7<IXFRId
z*xcmY_Bp2W6I2E!G)P@>JlkhH8Z%xSQI-XRp~hLm&VwU1&Rt@$v%_)SLRHbaj00P9
zU@S)hmO?RIv=}9k;{j3*nJ**<JNKC{YG(5#ZPVesqs(%&2*_YL2p&Bs*uHpyt5>e_
z<3Ij+UU>0CTz~vY%4&#AooYI>`UmjzAU}hQPS%E$Rl)9qd(_K@+qZA>)vtb;3+FHK
z;!7Xk`RAVJi6@?BP>qv!32DztNSg@BGDW1rq0fE{AVK8{nTi1+CFi!c`N(L*6Hh$O
zXFl^8{_p>f|CU!?{SH1POCEWabN5H+S`}RBujx&aeCP6<(P+elix>Iir#{8UKJjrL
zyMCPu7cWp$gILT-na%>4MxCTlWItm)dz~$)f5>&FM(!z+ag^ClOp?==IQ?mR&YAat
zx~^%PXn(R;%sD(dU^Y8pK0jf4a!k|IQJHT&#&prAGO#+~q)g}=fn0*}$(PGo3NdEz
z<VV1Cv~9!Pdd}2nsvxL@WQ`1T3ECo+g|IvmnHGurosS58<WdD_4P7QMosYr?=?FR?
zvS=kSo6czKB?L=RKy4k?MQ=%^G|qOQqQ!=F7NO#~q@<v#N@gb~jK?LzQHggNDFRA*
zvJ8d;&19%4t078hnx?}C!-WeM7?c(5yylhpH^`)-C^Y3*aq*EysO%wO8IaCV<Q0+}
z7c@%`ofKTUc!9~4E9mhEv?3@4(WNXookVmLJVFWzoul&#RKQD#Dhi~8ye#n2MKgJ#
z;6w$`GMdQ~BsQ(Ms6ve?I8sPFk!Rl2@hnqPqooHKm8}R?=y7~?#HyiiNpKTs8lZ9s
zxx{C2D&?(1xPWqze`j3EH<c3O6kRi*vpKcRI9N#DxwmB6LhThkAK;4uDv2&Kwys@*
zEMvB;X*<jI`3u~=yGz@)=q%&$XP)N5g-v#M?y__5F7@(++IdVHs5?Oo?CsYKjty&7
z;F0qi49k)bX815i2ZL00#eKA0hs^S1OVmd;MiG+?iMyKiNrWd?f^jAY1(8ZY!E6~Q
zw4llbg*0dc&cu;8GV%wNP>4jO5fYrWE1$bn^iOYHMLcWc^f~Su0`bG0K7N<$eHV7#
zpS2Qs7$)>9kU1oiz0_$5FkMS7R1_<C=&U0+L*2HFO~>N+h<dTWTZ2$K5<GR}lT@Pt
zGS^YIpSpqJa7ZR3b<@zy7vw^ai72U#Vvcy9Vlm=lo}Lh7N@K7|25wgnYe?LITx)Wr
zkUB%>vCFZ>&~zPTK8$0qu8CX{h;eCo!STT{#!EWuP}*U9jPMd6AVYz3Ip_=AKFE0E
z5T=6xdTWhmpM9Ka;~FPZ&&{{*@vE<Xh5dJTDP@Vya~54R09HyR340DG@@xn}lc^k&
z5*@q|9331Zl;?YI%$clh@e_aIB{Hj+-`(Z@8}D*!cED2`74PnRoyRUUl%m53Lq6C-
zh}6-fLan!1peNC$1=qjch*NlY)%A#&_b9U*-y1=SNa@i8riTxZCzA2yr#Wf6^tGd<
zf|LR;1=a>~nX|EdiQ#CBwTDdGIo26OQ*(U(4(BhN!^z0Klupw%im!h8JG}ndYrOQK
zC;5rbf0E%SWB2ej#u&09P*f#H$0y`PNm0fIpe#$;&Y{$R`P}l>o41)wmxxfJrQzP~
zTYTcNM|t(tFYw@S!C~h(XgiLCV=4kO1Pf#EUg8W0Z|P<y94~5|2^bqYGbscLL9iW-
zbCfcw2K69gW4y&*{-69+UjEo8$jbpzWrW^%w13QkU$x7>oB4KKFF8Ei2Sno7Np9VF
zhg-Mb;R|2<bsl~6Ixm0Zlf3xi2iVx$OoP3Uj)D|L)aMxX3y7^i00961Nkl<ZHDfv^
z5CNeSnj9gbS@obQkz#{C{ipvF*RNma7yiLN;2;0uf6r__$2o^ma;4b)qjH@ULiFfd
zEb0UsYZIP({yCn0_8C6-p%3wa4}OsKtu3mmN-FO-o{f}`p1Uhfi>`-8Jv|Q6zcY<t
zLVQJf6sNZ{=`*DXXTD7Ol6}KvtfgMoEEWsq(<z5XN1PlVvsf;umo-gYvs})xuETpn
z+tqmM;@H!Btmz^uke+GhL)?30<Wb@rI>lcJB*KI84cdF8x5(D9JDsDk6u~m~4TWj(
zwZjBQmnmB9Xq+cvyU~PLMh$8_;4B0aa5jPiHgwo1tHavB^khNXn%JZZLzYMSvhyAx
z9Km;lm0~#VXIYj}RRx2pVq;@KRSv)_Bst^JI<;!Kbg|^(pSZ+gzFZlAo6g{aXJc~%
zoZ#M#``rJ*UFyXWsRC71F&d5tNBh{eO<ADPv#ie_^-3|`-e!FMJfbM^A_<oIX$XOp
zvVeebHmPp~T4fmH$%+c0A|1oYE;eRTranQgq-(tuQO}tZG8qo{Q>io}bv~*vS9BU7
zS4wWtiv>c@z6`6*MrwfyfQ)>(5Of?*3I!N~izz22WKk4`rs+o1^*YN&a<cHe`(Q?+
zMs#^jP&xTzj2aeHqk{FVF)x1LSv-o9<0WlldHkvC{F`6<B1cCjWLd`5>sPpNaf9)>
z0o&KkA!XoTZ-@E*g89Lc-TOxj^#~{9er$IP8w1JuFk@5=Df5zG_s9V)WwfeFS>DN6
zcqLxw(;rq3TL8SH)RL}HI3cjc<6J<Oiab*gT%159)j3io>$GS*E`-Ee8-o^lg*vH`
zmI@(zv*PzM>Q1Mlzn|-Wka*%OYWL8`62#a?XW%_b#JRYs8yf2=m7*H2F{%po_jftk
z-)EQ$>iG%5Sqc@4t4yny#2^GgOR_<Q?^=BD%;!@IoiW-N<7R?(v7npJ5zeDrK$Rt=
zI3A-Vp|x=`p>!gKW*8wb)}~y|06s)@i?+zJB+qn=J{wffN?L1?){$o!MUi8dH6b{P
zvY;9axO?{)RJIE6wILeWDlhS^#&kobew}yclKrgWu37TOfApu=Jy~)xhcA5b8yxNJ
z<K}au_Si6Iw6Vtd^$j}Hv0N-!ES8kT5Gi4~FbGxRP&lO+uWc|K2R`{@FY=>5_937l
z%PWLX;4Jg~NBQo?d6xI5?B9N!jq(ZZ?0$zwuM}XTC}NmrI7t*o^jL%Fg%EPZ7wAVl
z{Spx3tY=m9D9ZAR-ojf6P{E<K#4S(QJw9S|=`ogz8Qw=<zL<2NQeI`;N!GS6Fc_`V
z9PYEQ9m-ikv*cv=K9@iA9MwqE?$szEaW)=K`+Ez%@wL}^u)ELC{@fp9G~D3rcfZf|
z#~uft({&!LOX|AAnuhAaCU3vF!;N=ubN~JxZ@l(4!AUY%(@l4I>-7VMdl}z<bC09x
zlA|ur6eassanN>5t*0>%T<o56VPY<&kA~Ti4j+qSCB-T($@_TYwl*&CXa3Zm=Y{7!
zfD$<bMF_p;;py{yHbwM@T|!FQre!vpk>@#io=3zYjaLp17aSbk=Z){b#-&SF_}Irj
z!N)%Kan5aRrvgOB{VNrOXGcW=(x@OR$93Y=;J{@Oo%`SiKfr|x7r6e|b^h*u^WX5`
z!Gm}}{SmzSdnSb-%QVBPVsmqY54`juANb%0x&FlCT)+MV=PzC)PbQ;MM)6gzP7U%s
zr1aGR@w<N4)mx9C#&@T~CVth`2*-OLy~&z}wyA0Bn&mR?5zFP0#eB|eHe-5nLfzD~
zO&6oZY)1$dh<=pON7E)gSTsuyE8jvvtD*u@5@IrijMgkt#l!^dW0HUl9_`ae$y$VI
zIBHMOrbD!z(0D8!n@MaCi9hY5sc~=q-Zv46ukF0S1rddEc@a^trKRb7G^iJbjg2*$
zwx(_yOxL2cXE>^;1|^wR<fUXhsn|Zh$=dpa!7!sUhQ+eQ`ABAOn+_11KYxL9+sS-4
z9k|9?w9?Fu=PVaXTr^fgYDrPX@6kDjvnB=@tdf^l@w$g=Ym6pq5sf*ExSySRB;?#D
z4^$}xQs;OCI?Gd&&8k`GtNgQ~hZPojrU6{>?D}zDY6AMAqF<vjy`B9!51sMi?3+ZO
zqX(%6e78!Za6VwXqVWZdFIk!ihjY!e_RQ))BP+ykjqzwgJ{}<lC0m!bC<htUFlR6s
zrqQb+FD+6j2IB#B+aQHxG%4w<Aun>awzn7z3v?za@+s&O_D|-hYD`l*y2fBSMQ5Nh
zg3bum297+lXcd*LsOVY|RpUJ{#fmx<`)P67hq;iZcd_{HOjJl~t;zE&KBMkzR9p0;
z)GGQ73Zgxgb2io;fzwE~6k>ps$@4G1&V8bXl>gQZ<e}^RizmXFnfiM^<b%iI<Ns2D
zC%waUEtv}_=c$S+DrE>L5g*{TS;p~{1lP0#=}#A;fV%5wgCol|g&d$}IrC+W9u}1A
zqxd<~1^Ai};`l?2q6|CFb1-RC7%X8WosL#5nM&#(=dr{Wsq4C^&eRf_Wn>b}($SfY
zTu0~l)^s$^g7=hFN$bGN3==%Y*0hKS!FCe6)JVO-jR&5a%L%uQW+|3@`P<*%M?d>x
z+`RuLhl_V$U>Qynz70J0>{I;J|Jk3Xt{ae^@4WgNi+RoY?aRFRgPXkh#tk6n!u3l$
z^2`;Udh!ZYF(J<!O5_~w8NT#|FL2}STNI_{$tND=^0nv52N!v3cb}o~n3EIs9~^S-
zyeG&h!sQfML6$}LY$Vt|J2Lbeo<HRcq$2;U!J_v8;~jZfC3H8Sf{&we*Wv0Z)16zC
zc8O_ftm(*#B8`At^v;ut;Gh_9P!7f{55c5@+1EA8qeC898!#B=%<t@>=<q^feIU~%
z)@a`T!7l&&pMHTq_EVo`RGp_@c(%^1F&ecjmMuak9_$|Q<u84Q*I#>wJGXZUUQy~1
zS*}p(gmYs<dvd~Wedh*d-qPd+d$QrEH9S~&POPVqG3!-(C{t(WOxiR^KIak$Vm0y*
zT4Fq0zIchh`d9xNfBw(@1qwYtsUrOgv1-m%UHI?yOb0kVo^o__!f;qIo{Z=ZZ|5x{
z8Dw_dlIi?_H{X7ZFMatp`1Geg%g0{+BwL&3Aw`L;(&_1Ik(3tDn{3J${p5j6NwkVd
z370Ni<Ujl?f0?bVP5z7j>c8gQcW<PKvOhXkUr<Xa&|0%LnQ-C4d7gddDSqOQeV%8Z
zeU7c|Z3e^9snU~JdZ{zsax$fb0R2h(p6~LW_njs8aHbIK3DhCroW+<9V=bNOm`$fF
zrVEyf1@qaA`Fu`Y*K{c#)LF}XKBw(EtTCzaW+lLjX2FsCZH$d*kumAeac6uJ<6H2K
z5L6OEWibsSaoIGUJj6ZKTd>a2G&M9evh&CevY^Pad_~U?CN|4H-4BUOjZgeu)*GBx
zNL`}x3P;AGi9~zr0!mrVU)rGU%9x3Kd`t+MbK4uNZ>%#K#XVv&8L_rDVzM@*8k8I!
z9#b!B+6D%LlKE^AEr!aBqAamC-m5&zX_|)lY{`5vXJ>bZrtJtoo@w%`Kx;)~ZKT|Y
zkc?_QLWJ}fDMphuhLZ`A|Cox{e(c9tgf<mkA>EBE3Q?rak|A<b?5<3ydldI~><wbo
zob;lsvtF~_Mc$(C(<fHh!-;#2c<AR@aW(}6nTkm?!sieI-gbo4Slb|JjHY%KOFLq2
z#_Y@+rnRGX0Xdp5+}>vM;(6A#w=g2`=o611v*?j(JXqsaCjSo}>?hy=2p;D<W(PBh
zB4aQr84ic6Z)}3dFi4911FF#$^SuT0dVvcC%{+Puww)%p0UKk5ra`y{scXQ1as=r(
zYfXZPGyxrC4|fuLNLk<>V*{P(2n4b`BP((srZ&lNdjFMFXsJ=jn!_dEI-gMeha%im
z+*4vS+ItE+@o#VL&tCt7Id;FZuuG=4$zTy5V>L1%u}wqiJ-%xRTBl(2I{Li`PcR*L
zL!mTAN^ILkzMIxK@2J~`>1@h)IH1%SD24T&PI#&;XE3agox^t)-_*EFBgz7$C3#U0
zOrUKVj4{zSO^6i#lydQi(PvV{?`KW){>k!;qRdfQfy!KbK_HUYjiqT~XIJDo21#cF
zopbomC54v2*Am%`sDt4C3GDfTxh%;G!GnYQ{Kl7lk#konp8Jsul~)J~GBpA(IBD)8
zLPMTqoV#Fo;_34=wc#^YUts%U!BbB?!-Hl)Ei>*O&dBAG{rkJzxp9-LmoD?skG;U!
zM$UY?;Fa&Z&M*Jk*Ku{px-D2Bd1>t<93Ov$a&W{XZ>a0Y5y`TQ&a_x>R}Xg2QS>Vw
zWxv*=SCZBmDWj6u27#!C(UU7K8ho&10zy0I*7v{56VH5z;f1Ss7Y#IJmaj%&h*(JE
zqY>k^HR`*P)>(vi2xmEdaG$3#$@aMo{C>^F%}q}1f?K!uIJdEZvx@m*$s4cVWN&w$
zE7vyp*e5>1aO5yu;GO$-m`xYF_T4u)*q<Uoi8DFgNrWmWRLA-#!yO;9d;2Egs79-T
zy~c4r1RgY&HpJ}8kZ5QYFbG^4Rp(hoAp<&%vwK%{>!EX=YB=D>fBL6*`4gYQbOBJb
zb%)F}9*-_ElwJ|*e$VI9TFafgce!`(K7&EYxpsrR&_tbuB?K3fP?SL{gAkh6zxOJ)
zZ{Oz2U-~ki|NKw$BOm@4lgWDGd7SaE0{Y@EVu?MkD~)bazG0qaD5?0lpZ!_ZH#Yb?
z|Ji@WcfRxb`y&1Rp<lh<TY$-U%x6CH89wsSAK~SfKg#7RSIF}s#_{Q%bS+ZHeQIHl
zhGCp?=lVkK-*Tlf>k0->Bd|d9-*nD$aCpG}Ud(?zIXU6z@Q@-e$#tHr7Gl%c=Tch}
z<HfqJqitK3%O&&qEb;D^bX|;}^F9!WWO^qZGD!M}M(YfH1Lv)cDrpf_u3GmgHPK|C
z)R0T!A*%~cq@oiM?bF1@Cl5B#U37FBP*x&sO#QLeL}w@$6K;I4wB8_!43!UAG=fDf
zV>WY;WJSh@KJpxswGlpe_V)HzES79;MIP;yE0<AH&~=etxVOJcUTCVp5Tyq=CpkRa
zrEZoCha~}r5EiMDkPCxzj@fL+-h+Ma+_{hK1u9BX6gi`^WMK_<xdhwdgdhZi5S~mc
z2BR_K%`K|QIz8vI=S=lacaNBPA?U1)jiOYM3kNCDLB#1xirIR2Wt~Om`sH*QAipZ|
z5J`;Ul9y}b!3&WDa*4AmWfEs4)H2d<;xli6O!T98g*R@+Cv#rVdd*2&P`e2WyTRRK
z&;8b8q(Bro>le3p>LV|4^|5PgoZI03&LOCrW!<uzJL*Nve15{o$uTF#$DEv;(6$}U
zMj_X4{Q6fZikz}47z_q%ZmqGlF(N2Qxjy8P=bm6tj1lD#J}=41IJ&i@X&tqdbge*y
zH9}QU<x2|P;arPQ7NkSy06OW<)v0PY7P|txL}dlE4(mE<)1g$dsLC_IBwE_(+@p7m
z2>rb`DMC&3*}Mv&qadbNI*T)jEoqVbo?FP->-`mWw!@1+NEvBDMFphy9_t-G7`*EU
zzC#3y@EwH?l*-{2Gqkld^BPr_sLV6o7?1@|SS}I9Vzoq9LmIoF#UZST<1Hk?dYZ<N
zR~dyaurkMY4wpw&p&ogf*#hGNw(S_SCh~!l#8x^63p<N&DYAneJ&CjJkTT|^i;LUZ
zXn7(DF$YyDoDvu($cq88kSN>Y1r&MYCv-~C=^W(>#zhuU4ZX%9zsj5MmOQXyYUNSo
z3G3T|i<d7lTQ0fr{X<&QQr9iPNHR6Rc9z+lLq7Almq0Dp+rLLyj<|H`NnA5Wm5zIR
zKj7_~#~d969((i=9=URj&GH=g+Fkro6B^6rxa892h_!6Qum0vYS#}vGbIYz5TpSgQ
zvQ18AhYW@aUF0|vxz6!jmk^1l%n3M5f{!DSmbCfDKYIs3#lwHv8El@T8cX4$I(jOh
zu%2w`5%>35zVRmOmmiB8jq&I#+8hWaa4vunR9bR=dxIk+b!`9=LG8ZKIzu%zVt&B<
z@D7hZa)tYE9r4K1FR`fiX=}@;Km9UV=A0a#@b=qpaqH&0+_iK5)xj|vm#*^8JMYpg
z8gK#IS%Q}+9ca4-Aq3~PGM>0la`f&ld$(`1oE@`Jo|Dk>ps_51z-f&YfQ_6WK^()1
z<maYRhv|gCdKgp%66l%+;~iO1@bb$a<wt+?lPD!I&a$j)l$OX&Vx1)~a*84+D{_o$
zsG9}z`HZ&hXxf&xu303b=*}3XCr7;d);k>T9V0f!>~>xBTq+8xK~!{0Dd@x!A#0S%
zkTSzh9k0DU<JRpP{N``{Yd-s#ALpYV{UpQDS`@vA=p`nlNCF;`wL~1JIhV#xsnS<c
zpsR|{fA%v}WzK)`U;Wp7`O9C61Z>WjB=<nj|H!Vg$XQ=o=aEMq;U|CUCwcLu7rA`-
z5}R9_l!HOsgjXEA<kgt0OZpD{6h%3E0((|F^X~?~uHGFH@WpJ-$;lD>dwcBMzsJsQ
z6qh;Y5<+^3=bnF&$=U`*Uc>{r|Jvd7P<GDYtxKWs2Gbe3wq-V-vH#!!hX;Ec9PG2d
z{{T0QjcwOiy0+%z<QV5Hgvc3ct&3C%p-@>yrXy;VX+@Dqv=nF+i-|5Dvg|CL7U>-&
zFc1ReEXsFO0mecw%tUnb4+;{3XM?9ThE6HuxPY?aXxh@4=z%6=KxLB2ddbC08w`dO
zPd#%j{@h)(^cW0cgSuEw*+1Ch{=GXWDNw3N95IQIu(md$8WuF|0%Pmw2`4qd2X=Q3
z`R=Q4@XnhzI6j)<f<Wp>^dF6fNbhm=5>wBSw#KRq>pKvRBF`9ZZ8IFNk>^F+H)W5}
zMQ=7ivc(`GX>HS>lF5H-3>Jwbi)+mLm2F1!T@??ZZJ`%a^-JKPhwSvv{ixZms1iR6
zQBKOu6|q0@cC}1)9Tr@JH*sxsPSV<pqgu0MjiY73!Q69XE4sX7yuHTx%jdat?IQVb
zz`QlMqba)ZeD^!w<>0{qvy(Y(-C{bsN(AaUi;~gC#u(4tI}Z|9Ee;i`s^r}E7VGEM
z*f=-gh3BvH`s+8CT&OsAX^dSi@oE5bunkO2;Lf3=8D(q^2CP>VI<$z$7eWLFLf~X-
zVud1VOp?`0kOZM<jiYW`S{oP)i|DhPkmiRT>4eykfRWCaButF=Gp0~v2$huB;Y<ye
zRI`0!eCj&zdtdLXunWQAy@=g^KaXEQyd+*hNHMzs8v>b9QEe9-0Y}%i;0(T04A%za
z1e(PHg`jIXj5lPJVp%U4%1AWKm5vf{Kq$p<G$u5L#<fwzkmu+kN1134?1IO3hB8YT
zbLl}0-r|kL851{JDFBa90V!e!E|owNMH|K#+NMDYL6PV9OvFtRg)S;`FR?*ISYE|O
zHptk?1na>|7E9P&2yUM^{A7%`OExYlE<c`g>*h_44g{GRQI_jWwgp!&USVyx%{y=1
zp_w0ZcrfFeul_n8{rCqMua6myDweh7;^i$iw?=&U<&SW*?|87g%hCQ#zVKULXM6Js
z7cX37xoA1ReIA^J?E!z{kN+fJ{L*XOo<G3piW_@JJhQdI-m&4r#S-fUt+C{VN2x3w
zRDSiW_tNap|9yqt#^%mBPup~KK_iM1ojOSq;)CLZ#5ab#v)F@UDy5?eSCP^}qM@`1
zDdeWe8LV$$ZNRudDL^|1?E<$v;OT1@__F;5Ya3&nm~rXa1+p?w6arn1kmY0S9NgyW
zW7n9hZ}OX8{RV?-#HT*}8E)UW$z(KUXJ?0lqa(DpY@gfUgU@es{LX9Kf9w0Svtv3H
zI1G-vjb-T}R}~0J=a!MT39Cr7KBY}daMn;KfuRW@7I4<#Lg1MvpWx%4_yo>bb|2i2
zV~1cUiVW{P$0tWf0egFUynEwa-nsQQ`-c&!Su7T-aa(`Q=W~vZj#w^hhQk4M(_(DD
z!7RvhMm4CS(5H$y$KWE*By+U1v~A7d;RD`$^LxDV%D4HMpZPf+d-U-r)(L_T^eP#H
z5Un6wLU4n%QH5ws5)VLC=6v!KALDQS&Ht3Y{kQ)Pzx7*RAP^g-aP~YQ`knR2@Cvp4
zL%ObAy~dMIKFNnZ{2@N_@<+IQ`7-0lI3i&^!Mq=D^+jEesH}de#MxO9QBM!e6$<*E
z;x7KqK1M&)OJ=hv_wV0jcV~yg!vl^E517tQ=(-lAB!fXkQ9i=@`a0|Dn~{gLswhuK
zXJ?<&Ub)^|C$(L}rHhxC&8N(!CmbB?aqsROj*bpEIym6=ts7WlY3gM(VfA1`Gyx7G
z9&m{0Dxj4j&oyPP$uo`4BxC|d7Skj!1nY1vCLm|&N0j}&NCf^Y(*za85{)s~gh&cq
zaJ;A^B|vGE29+twvY;q4@-pt(Ix7$|+A<l}#vF3t$udnf7|@w$Jr#lgp~-VeS!EQ3
zrfqA=azJA`OxI8^9dEsHlUq0Lv%hzOb0UtBr9fti!LS0`5ljo#L|z~=5sd;`DaLE-
zl+`f(ToFX@tM!n$=xM!pAL!bSQWdxmXl#Sg896y|r#Z8h`t`Fa=KHTHSSjbv@cGV8
zSmJ&i&hY*E&x!tstI_M3t1lpXR4;f-@KIc1y{9#TMVC>#0Vg)&q?ODJG)iMLg^>kD
zYO1vni>9TSo*-sN%uWuubL%$KqZ!SzB{+fY0)32_?F49r_kCk2aMrDU<*uo5_K3y2
zWpPrYQA~y-2IG>V&@85c!QfGj?>u0>yU+5-Agx3#<5aQCGy|C<h^Ze^_*C$^6;kJ7
zvH>yymmvs@^|Y<U_<+{KM7`+wb*WUA5~U;WHn=$Pb7^0Wk*K{jSZoYWLDGbjq(-<G
zv^nz!RM<K15KgQ%;MHR26=Xu;g+|cQPi%^^K<k1a8-kEH?{Hm5XB_3AL>C%Qpfw$X
zQAt&4W{YF$*__d6jBO1?mXYTLE)kZpszl`(wysf~A?T7k&+(N5VWMlbF=*#jf!SG6
zB7DI32H|}YB1IEP9|D;=EmEb7r0EbmUDr|9G0ZzJ$~1Zplx2ZOvTPmB8H!9YC}YQ(
zXELgeY>sE0dj~CQZIcB{2Iq1%FO7MyxW|RZFY}R4J;RktSJ_(M!h1tiM&kVkKluW;
z-~BchE{%9Gyg-&K>SoT#Y@e;|i$FkSfvpPz&R;FLbd>S-Teo>^XPvF}P3rEDwY3p@
zhx?r0I?wlh@O?h`(H~*^QpxeYqCIfT+5*ES)>a%GFW5dm#(78EwiH>$SwUksyAhny
zZqgd`t76Q#G*;Cb>qAugB;P<uz;uzsJ)53TNQrPULd*IX>n78P&lwZD?sFHIcLCEH
zN~suXMerR5H@?rKPk)O4>7V=*ZC2wqGPW-~fm8wG7VIA#(6$yU4eOUS7!AjaC-DJ(
z^WE2}$^rREbM^W~8b9UwQ;$<<xb^0%+<xWjEN90Ep*Uzdc5Gmo9!8I+H7&hzRHTPd
z1Sm2^k;T}j&RV>P*?|Z_ZEVan7J|FGyL|txx9~!+Y+6oc$K1N3IXXUKXXifOd;K-;
z-M`E6@iEn)Lg$K*P#hs7z4@ne4r5}tyK|N-Q|WV56c{Ge_j1`VUo4`iq7sb8Lk5EZ
z&Kit~M#P<H$@7B4qdk7@U;h%{{qC#$#83PrKmF5xl8yCqk$acWD3{EOg^0g<8=}gb
zG>UdUTA$`w!3)nl$N&C6`5XN0|MTDB*Z<9LMJN2Tq6!gnO+!jE`NJ;ko`3#1KJw9*
z`Q)cQ5p6C8gP7%-EMf$SzLC5d``;lTdp`SWPn6>H;9Q9+()a0);r_7o-qE%-$H#{p
z9PYEbyTiM0y~VwIcUdkM$%@2AK9AN28OU-?S(X%KiBhNezx|<3QbRseM22`Ur4Pe}
z3wR&Lp!4|&H{N-ho%{E=d;1oL`}-WQkL@h=q6Xt)L6=5~CPV=mM1^;jXH;cDk!R$&
zCLhFSA0P9x8xs-3OiBvb%W_f_WlH`T$s8*KCcv^ah$>L(0(8OrcmbrO1)auNO0Hs`
z+O9$C93LFsTbg!B+q58}q5WiS4QniQ-C(UlYRzO4Iki&7(Y4P5(~~9B<0S|CbH4Z5
zJ3QDuU@>ozXuJxb0xDAs1{JPtBbCT?5M0bZO$eOQn$cvPyc(jiJb8Qe;fwt~n#6`a
z7I)SdkQv@t+IEgFDza=qf43gKVuNwIruwELtQzq5K1-_{@H797%OX7V@qQi0<<mDa
z3L$i&TbU?s@qx~1YLl_(avCO_G@7|pbV?%zWt;>_$-L>XSf-0<6#ld|cW&HbcC?^t
zEY1WnS>kLo@<qf+n-{UL3tk{(jHnbzoYWb|iLIF**PJ|<;%dj^&ppDhs>lZlp)8Zj
zn<x=zo0|EeqxJ&2XtgyNC^A)|h|#0S<Si$BEZ}ibz~e$1#YXPDF^;ZtSV1x%Pd>Rm
zo`rp*B2$`7?*VHQ+J4HbQ>%T!r%1kd|HY~=>;V<~J{$(#S7GP9$KifxVF$5`BM5{S
zaZJ@&nov`0Y_be7ISr{0iEL_?OH{FzPz{5fH)JlucS{CYPN_3mGp86-kv~vqyb@R+
z2$_N!Olz>2LFXk|Q6wjGfeW!%%v2o5Dy8r$X4~Op?5>=Tv0qMr<{=f4l#)Epl4Ezv
zY~~mRDDphg+eM7{Y1<|BV#c7zSRW3M!Q!34H4fYwi*C%ZQCxW8NhX^WPrdvClM7?2
z4N2EZcJEA)PIGuTV{dPt?|uJuI%Bzd<uaQCgHoGZzy1uz(+7;#CNyowfBt{@&$)2%
zB7gmV@z)~3Ry9n{6?|y(S)O_RX_m*1Z-4y_X3Jw{iz9^0dE)8kxqNk-M;@J^g5<R?
z-A4|#IjoP_o;=Fk-LJ5*IYbG|tZh+i6NEV1<@H6K_#K6v=<kOVap7hfnddk_%lI%z
z1i{Amu&!;eZG)Z98E$MxV!sFk+=@UQLbNbgzjT?JoMzz}YLCitj9v2f*T2AW_k{Bo
zpQhTnMs?|Fva-UuXfd*=A7Qqrxqf9EgrsRYe)NM+@x3?R<cUYmvwwI@H2~W=UV8p9
zcJAHc_3ysM+Hw!OJi!Y%HkMr*m?P+b-~x(BmZk5|(mg0MMOl?dA?TVoc2s#JRy!9~
z_finpJ3Qo@uY4P+HFxjcW_RZvckbR{dUArX9YR1^MiEQfd1ec<;!X9@V>TVA&N-T<
zq3hzPx@~RrJG3U|H>OdFbB?xM5*D75lO=;e$z*NJa9Gi{T`CG=$b6pV>^|`P<6r!L
z^WE28;pcwt=lST5e0+scDIH_ml#aGUp$K%LiJ~dnqxlLSJb9k;(u*(hzx$8>W2994
z`fvPZbfrB#X`-i7I6YYYFt0!J=YF1No_Ut7txdF2aqOeyisY<RD(F&|9j_lgTJ0mB
z&Z-qdZ(EZXR^B_BwqZJ*^2YaH<JQf0+1=e?@4*hs<?_rSe2VN?6Zw1A+TVSe{lkj7
z{$~|u-yM;ITxm9og0;znXP<qR{k;dwr!)3;9<US(G#YCx-dn5(PgJXlUNmJab<@$#
z7SRA#NwQorta1ve(IMKB2vi)gg+QmE1DfDidPi*?jq;47z*vQxwNzzbO;@PA;^gQc
z9>7HM|8n6tI-D{ZkB~Y85##xS5Cj2j*D+r%an@lvgLMv}Bzct~3>ec;Hy-D6%Ccg8
zV}tz%$GrK*4Q}1I$E|mFm`xobC?pCYEW{QIp#`JyfObB`E*JQ=#W_P`9G#2OsZ3{#
z*EXY1sLJr-R9thq0w9e;Vv*-8nFv@@bF#NXzIC2~oFJ8s`($6RrSYlII!;T%Do5gB
z4&Ips(x21)$mMrDYpV}GT-cqi8<}F>8oa4-ro{)%@jPQ;3YKA=r8>{CwS;QOWPP3S
zrHhEska<04S<iXr=1r#4L)yBgSvDN)Pf?WMb8rIh3d9PbjE2DROpendA3_?9#q+^h
z7pHALVA`5@zxyT+?%!wY$~m^LZu7_!S6FtCkGI&mCXiGd-Pyx-89q2}A6S$s8I_Q$
z<nI~G3hi{s*HI*(D!Aym*jh(p3}{VRR%o4Z7J*etqO%OGGdw=U=OwX<lF6?%0}-v+
z`dTDT1mg=K!oyAOA6W9lN*Q^kl0W@3X=Eb^&Ldsq$Q8L}FdESu>|?A!dW)8UObWW$
zf~AsFRmHTK)0&t{FgrRRbR8}?n66{IzJ^l*sTDyb&QFn{q{P{%0L!viz=<r%@SU^q
zL3TdgbP?fXnHHxI3L*U}q{kRT9}}b!X;KJ5mSyQsY;oq8(P)HD2^WoVEao%X<&sII
z7-TwD{mv3>K#CE2hYAK4x%~Vy6pxMAo9}buyYH~RIpN&eqb&C<-+Sd%GVQo<X~MZl
z$=*TDtKawm${&#_#iu^=GQ-g}MVWEq#tq(mcbCT=dkMva#XOLe8Ce#8_Q6n4<%&=J
z#D}<ZW1nC8<?nIp<^zP?WZCU;?U8ew+q}ZT9m(`Y$No}setdzGea&oMBZfKFbPNY2
zd9I@@K17>^Re{&5xl@tXqZklkrmI%ys!T<+OT|SbU*kO9T4eAv$0v+0Tbwsor3f}|
z+B)%<0*=wS3xs?`XC6@37C#XTD@|(`-1_>z;j4T9j02}o<4vT_**bTQ%hw*`{H3dG
zY@O%1^)<G(&r=PC7}L`J<Y%a@;CR_^G+Xff@4t@5^VK)Ljc(_NcE)g+b7X2Bn2s5e
z1{t|A-X%vZVuzic3nhuR3C7}`#W^4IUqiCeS#e(ae;RoE-J2YpoM7q&&a`MP$n!jU
z$pw!O2GPVdXl%@ojXXDZMu8kRf@mD7DasMc<^*d!&Vn}%LF98pNI%vj);QXB!E)I$
z7!-^rLy979Oj)k6w!^vTbN1cWzQx_Ucli8Ie4n5CsXxiq*7nM=-ur-(lDsMbXzB(p
zA|DGIxiKNc9Mu<Jc!B@qZ~P7Frs2z9`AXtu#(Q(t>+TP`uzTvsC)wIOE$UR_;3C9I
z1^cj%;CuY0Vl@@_mx$-a_lMrN-6xK!XEvR2@BTgR-@D80TQ_<4#ycz*GrF#gv2$td
z^y3&IPElA07?TP+akj|hJwM;?yaez4cz>b`q{zs1U~Rm?kNn8Xy#DIzET=W~v?fvQ
z(lZ}WF07icevA}BD2KNmV+2hXuyhH{x2vL3$Uvr{k|MqdiIItK+6D#=7RInH#uWQg
zwA*KI=RRHAB4mz0<2uhbzy3PMCrd6|+Ga4SD9fBoD?-rB=Q9?II*$4s5y3Z>YM4`2
zIX*PhO-)ghgdn+p_kdTw^?h#N+TmdDgy~6xmV;FltOXbxN=wQrXKihSncu<HOT20E
z-q3kT=K?NBl+KuJo}(y-5orq1<)JqaCh<3-J$^`8yB!`w+blTQ+hJ=mMhlOQpOoB)
zdb4YZRHsEy?>j0}i2W(&;l0=CiakrIdYD>uTKK6n^%7EH0c&Ej>YZS*^i10ftG5Wn
zRi^cT%THfrw7t&c{DjL-U7}hmG2YU&G1Bb8?k=_qG)u?Vzw#<adksP;l+Li7Oj)X=
zTvtH42$lpJsZu(g&ys#T+C~AFb$I8fXPP_j?($%N#%vzAa^(`H(THq~>o1<C>0YHh
zK4Q9b2s>cXDP*RTO_fEf2r~E(@WLZbZ3YR%`oTJ|PLSt{vMi&6hKL2d7BM$gYmJaH
zD%69I#?o4$vOFrCQ`DYK9+cKadP{t!d%L7)9Id5_l%L<{^}Y+c_+a)2dAuS*z{=cG
zI7b#7#x`VLk|{?f3_)~+&=S0%QW7Pi2VFHRDaS*Cm1Hsy+BP0mN)bY!tST1XM9Hs`
z$P7r`vomqLXG$9bcJn9=SN2TdnF_REEG-Q_!fh+xA?IVbF}`=6j-7SPe$z-<u~>L)
zYdAVQBrkJRu4!HT9lN$=R4J4<kPfUOD==L_V}~@wkYD_@Ut#<C4az}@S>)`z-I9e}
ze(Yl(L_V^`-u(xR?U2?4c8;ek4i3>-;QOyXU~BV$AUf{e{~puD5idUfqrCNleg4+p
z{BJpb;R2T*+2)C-F7fyiml=;o3<pE<Oj1^nH}WGN{W1RWfA<CU_uuAE{h80PzBcB<
z)r&m1bqf_VOE<u=#_{o#@i69*&gTm@Hz%vnTc6k=LQ)W~x|rl|6?`O~22W8IDDUx7
z5=^qLaRC{kuqD?D6+CU-#0L^jXIsjwK!w<Sc|1-8^kj{Ew1Es6%XvdL@=TOuO?nh9
zE?Xj<#~duF>yDi_9rwQcMRZYMT;$OV#}fv_A%nFcgRKjUE?i@L=@BO9FR?wY_||WK
ziG$m3^W0-s2v)JPf5QERWd<}-61+G)pHH{M{;9xum@S)!@S8wH_|8(E;xp8bwJm#l
z`{a?2OmN`Z28aby@*qtBi$Y)?QkP1Rio6h#WrbtDI5@+n3OzXU&`;fSd%Yu$dm77P
z(J&g9j7EcKwwXDcbL2LqjCjvK`KSMwci(-NKlkVU1D<*2c|feVaY9P+s=^c1zt)&2
zC`kvH4>6MO`RAVHfBV1w-|&C>TYrmhefyPD#r5xwJN-Mqs%j7s=ft;_$*fd}Slo$J
zM1^o>mlZ!Go;yV3qeruQoW*%Z+tf^_C%pCM54e5%7JGXSIN0Cg@MsTb%}RZ~LRjcI
zJ1W}OIERoCy-nLcr|a?exWpM0C!D^s&#4rt>CJS`mCM(7{)LyKm4&f9ST@n^p#O(~
zl(U@*fb=E%H4f2%KYFx<Xf99c2W7Mf>QtbU5+eg4i$9<B0b?Rs;t!X|+dHUfpq(!e
z&Vxt=rPtiL`GC5y+`aRF!Kmcixed;rKgVEH#pcGxUpO`rk)N0897H7MYgHh~XqJwf
zH|}xs-MgF|FKHS}v}&|S;p1!Q4PF>BRZx_g;h@Cq*Z8(cR059;k%#DrNdwhjgiK@D
zkkGPK-TfYZo3#dKI=t&}T}?ATLAVYPV#Djh**xYrPFG43(x2l|cUrce?m0h1MsenK
z>sP?(ej!#IwjS*0sRZ7|obccz#%9bG9?LqNSi{H(gL4;{T)f0+Ys7G~!1xwBw<}Jc
zl!7Z)FC&GbZ7j3df}Oil7PF4!!f<ljAc>9|;0VG{4l=YVKxEYQf~wL~gNm{$DT)$Y
zpgx|lIGQp&n$fxjjKePB&fR<HEMrg>R7FWX9H0gz-HF6{P-4Jw>nO88COk?P1ZR?9
z&_{k~f7YdyFQkeicOU<(wMsQZw6Re_M!(!tzqqvSg%qevBX#5t_OHXrThR49<CHd$
z?!S;?<zx8!z1~-0XN)DNutK!XUP6fYAQJuQgm1}xO=dl2?<T%JAYuXr5iHWGsHBrQ
zow2meQf3*hUSR8vD$l8#hP*5iUqQvHIdv6LO=(s}u*O<+|G8vQVd5fLUHQ0?M9;sz
zyNwlGFD382qRt|cz4}Zk#b7X?HVc}zVYx6!qwzvutwCTT4^Rp+Aqh$$y~J8c-74x_
z<8r~>y*o4~4IBCz)5AMV@-aN~A-?fzU*wT%Bi1fvTz~O8n`>L#zqzC~3toTi9=C7)
zD$DkSAR9I|hva2WR|l4}m^AVB4-Ob)o9y2H9*e~>_wV25+U2XPZ%o))+eQY>=EgSl
z>@Lby+`Msz-TepDEYZqQTg_;Ajpg!NbX^-cF+S!>I%ig|Z(M+JgA`{_0wJTHk91(V
zj!X*3vlyKtdT;djaCmF5T}PobVmc#p7UK+-$TJc_Mx_}6Cqd*n!_k;S1CBu5S~_bP
z6a|Aa!w)q=v=A(v_qdD{PVG}y<uLPIx<kY9eZg%q>S97&u2ZX=wapFo?%(9`%jX#t
ziW~DK`<-JZU;*^P^)oB|y(^#9=A7R1&>{cBYL<lR5Wu>Cb0937+GC`qJbhb<Tjlun
z+^9Z@#3zNa^YLd5uzFwLq&#$m{&@uvX>)X)qp24x7agO~fTC0sWgdgt(_x{s;+0pv
z$%6-b{QS@Vd4BYhpGK8Kq)Y+bQjk{_LMoQCIbGcllw&ZgD2jsRa*0-&kACDM{IC9_
z|A_y?fBJvq_MO{5<mUY0UMf>7Kd(L;R*<Y9Ry?^MzQdlWlj5Q2!zxQNB5W;-dco1r
z0eA1-<{RJqDm(Y@u~^LMx(@3adWB>Nk-LiMbFC#ZfawgzT0#&jBKG^a!kM@CHX<Zr
zaDqh1jIFKneBc8g#<rHle8Iu~0k-Q9-lu}~%sNZ!#D{)p+#gy!*DHp)8^964BpIOd
zF;Ly85YsiXSlGGb7i*7dTJE7nHCC1=p$J4o4l?$3ryL#4(Yd57HP;@w%<_qj?ekk?
zxx{&a7fDd6;*=-LN}TI3oue3)bRA4j7QFk`U3Tst5`46(5(u2@l7|=s;qVxgDkySA
zRpr!mjqf^wwRjm#f*jz5K&XsjFhc2Uh4ilOeVT&E#AWf`M`L(r@m-5;7j(-R*yJ#Q
zLl6z1SA}!WPvw+!-=n9i5q@91{Sd0dL$%$RB`t*@$b>AWDHkGg?49ihUf`UjYXggg
zzz?_R#2U4)n4G&rvA%&ED2#CIJa|CUM*gkNH0zsVu03*@EY}!gc>0Cwte=~6d_3dm
zXhybkgk2hZXQJ4^dp5U9M&pVsQyd(6hQo@@txYyIHyDq`h#+`y?*aF3OBQB^6&9H*
z(3Yc<2MqReE}Xx>U|7&OiwS|H^R!l=^pK-=$v_7x;izPePiKwC<5eu|WH|GDNQFaS
zLj;<l@vtL=*ubbn{W`U|N=>$mX<EGi#Y3!7`uuetcNV22XEvV*qW#d#{CB_JUt!mE
z$f8Jc<KOvB0`bL{I4}*lZ;@@we7D6n%c!dJ9)Uvoh^z%GF&1n9tqa<AibOIPk1%yj
zF{mO;sldxX<}(B$!mcJ71g|=hsNjg^q5^EXu{}ej7fghBw-8~)bOA&hoh2Sw2%ccU
zx(KyqSq5bZAyC^k`T)rYwPYHGWrik(XCoqq&vb^)35<|*!m>80Fo%|-2Q^)jv3)7y
z=-^$Rd2+zU8Z`4AFc%5dpk+YVjNl}7tuVGCFEvdQSX*CbCg*so5kccz#}B@Dlb2q4
zj3=*O<mOv<`Hg@570zvK@tMzjg87Nz_-IZ=#XxNG;O<TC-oHa9Jyv!sTFXSPvskn-
z?L*1b<t7DB57DRh1s7HxNg*Mf)^tgl?HQE?Lgz6)Cxn!~0E9r>L}5mu72WZ160&%#
zu=wBzPJ#n3JXUyu2#m%PumLXuOJlJCs^HLg2DxAq1eugDD6q~(MQDi8b>0V@1)~C1
z268;hVMe<);)~yX1G#sPt+gS?^Alz#2i!Sca^eHaz7Zg1`Kh?ArkJ#|*AE$YoYs0j
z^jd`qa5~0%=;MNv*BU?8m(XV?E6zq0-Rkdw6bIBl|EG%OvsWsy)4_1M`NzmYL)R`C
z3^K-R0=&f9-ljt^9BNLcNBo0-@C)oeIN+y$`cE)fU&kfB9WuJ{_eid5TiUimN=2UK
zDfUV7+0TBK|M_44>-=B;%m0$&<73?ah!DFhrQQhQ=$h01C0_lD^1EG$k|G2kWh}N6
z8gu8)Exz}??{e$b4Q}0hm;Jq+Xaf^GKG{WqJM;A;`8tGH#2c3khg*Zz4bzi(Eb`v(
z{X0|9rKR#vRVt-MsX(4No_+Sk*cA)O-TU`=u(N~5uTl!6qz}i4SK<YQKtiwFYNX_(
zz=VJk{j&v~gEj;j<7lm?D`2byPAIg?F}0(%$EZbzJh#bsxK0DKE{=MGD_Kk%tZ$JT
zmdo4RxwFf7t>WUvb6kJo(HQxbDF)Rjp&xONDfBwc9B$md!~J`Ey!-ZjY?sif5D3BH
zXi!qdw*cVCa}|}indE3b!*^YxX~YHNh<tOM<&>ilQupK16(knX+*AY$?;?exi_?s-
zoYEa1P#+&)n<Y4dJQL;UgI}FpD$9vymtQ4#^f|#k%@0@mz?rc@?@Jn1jl^kl+Q*Fb
z0tN4*sH3wj4v*;+%Y|Y-&9PwvudY!0F~_aO+kovF?EaFvIpXBxfW@LkYt3Xl=A{q6
z$j<#^y!3b+#bm(qk6c9O@mV=KT5|u*yPWJFvpkt&%#!PmUt{y!I-_xka}CBBHqULc
zwl<-wJMyaF;NUzr-`(WMA5snn_~0=v<_OpR9<>cjmkqaXO?l((w{R_}kmF2r)2Ogi
zK`>GorZt4TBjk>d^^J)~`dBk~A#ov^7b~qJ+L_X;q!46zj?#)CBCz07u_v`oF{!<u
zVB*vz4ug~_+gGV%RQ&LS=nsrz=e;|1NakT^aFrAhEZ%hp+mK_yc65t5!dpxeiKU=G
z4Jk80P#PgLI?Kt6k~YA2G(lw&YZW30I%jFEBX~uoGjgHh2ru1Ctu#7|>e?7^9{G5g
zhzh0N*wF{SI$Wek1A^3<heUh~Jr^cKhiGd$ARtOmT2Yl1-au_ytm!bOC0CMRRiFhp
z1X5^vxwOv8Xc3@bFdQ%!9fAQy!i7uQT)4DBsV2y9j2D9Oy2js{W1ImIu*P9+htQG`
zB-XZ(bcuqX@z$e6O!s)^sV8Zg1)uw|ALrZOe2oj+7Z_I;xVLk}a$XaBV0L6Vd@!eN
zmvo_nJQBlmxyG!E_s6@K8`(4sRaMfPmG*Cu)1^|ek({>Em`F%TIUeJs#D$Oww>0MQ
zo~~Z9HW{Oppqb55NqwfPOp4_o0#3vOWHKJck)DG^-LY&thJ{26fsz853vwNQkHU*Y
z@{V6e61%5>pzQ?1Y{Y!D&Trq@AsZGHlL@zW?x9Rap#rVUn0i4c9Q3>v0-6ws!z+sN
zdptW<P}u3aQ_#c@z2e728d9x>x9LxPcbz(FA)<fMLp~w>XPw$?L<@(9pUL038JtmP
zMsEXFaYk0qI*+x{E@d<_42H$(^;R+?Qw4ce@>{?4n;ae-^K*aZPjm6oC9HRmdnbr0
zO$6O?8M|KZ$+9f+TfJd89P;yj_UE~`v%}x}dw-Ala`C&|rGNKVNSx5<(Rmuhbc%ey
z={WQ3_=?rX`|&FQ>bmCe=z!a|Z}PRTe2M!zx0y{(Sl06xST7<{?Y&8y%p^`Djeoni
z&si7f8pCWFy$rQ3I6gT}i4whm?C<xNO#c;;R?#0a!8#Ptp2}I98eze??F&5n{ENKt
z)*GBmr!404Xj`HLO7y;J;*2sdii!+kW$_bwWa>;&*B@NAM?(^w$H;(_60IavkodX<
z+ag7aR!|NK@?5h-v9yvv6zVwRXzS=>H$9ngxOc!}zDz3ciqUw;#^xqUN*0Sb@4j<~
zoA2CZHf?BXi)lS&QBf2*N(rnlST>H0^)aKWq7pgxj}N2hXf!~{!2I|CZ<^Ey1g!V?
z5aXfpqNExQqrf2vI(`R0mZG>wiup3nIC$qB9=!V&Cwse)vS))x$PcWFIw=$(FYq#o
z1)cYi-?$p>xfOc<-k1L@qW7>hO5D#?$}Uw}gLMRJur`hsOe<)coTeR8RM)vP&DnRB
zgRtPuSHFYHTk_m97)q{O-s15mFEO5s8I48=X?gR_H)yRvYQ=>sS2&o?Pk(+a#YA#_
z^8y(cxPSK!mmgU}X-{BEUIpey2h0}@Ld~$ok^2F=2m9=tJYenOm|~!CzKv$_Aw*$C
zx696hU1o=dywq5s@HV5iIg8M7(liW%XQNQ$GK)4gBF2ry7~^ElA2il^j0cBAYt3m|
zg+!_hofSxx#l77nF<BqOCsm9xwQfamI?d{p$zwO|A0kc?LyC(D;rIWHzpuhhIg3W1
zMFuIl3w#<Cxd<Zyxor{coYJ=xd5DBZspEX!8CutJ(k+;B$Y}E%xf)W*645zqvp`Bu
zZ~@yHl*$O5#dQrq3akS^9MV}!=K`73C|yLZgwPP8LI@!d8dOL$)iXs25<>C?>NoF@
zBDCV?By~#wXDqF6z<F|=BQr&YrfF+z+aYa0A;_~V{_@r*UYf@ViAADBNob)^8PVvs
zrYP2U=be3qWzPC!jkdF_ZJgtrxX8xUAs_wplzO_)bid}L(HyrO0fWeV<Ow!4%0>S^
zV;Ai1?DMs+ew*iCc#3K;;>8cV$l7F$qR3d(Ig%|pf0uc;q(jnrkIWsJmV{2m5ocp@
ztwWV1A(8ln40s>AN!xX?Fh(F9cone1<FRBx5Uj_S6~)>)2$I4WIv;4I1RR;`aI+<&
zYD{n)t~rViw9|16;|vIil^G6!@&<{bn2bRO!V>07&%E($j2t0@%;!`dvRvbpqVNk)
z7DGkn1Op||xuRi!@*@_VU}gl1wxKLE2mAL>($V=qUY67j+JrI$7m}5NkK>~_^yqiw
zGuz~QcFiCDk@sAuNb;#iRX?UV9g*}St>5+go%w1})$(usrmLv4{`*5L4Wpbse#}^~
zro(mtYXo7U5u!n$sj4By)(GKw{d=!4TORW#{^ZZ{<kQcjIyjErby=aLrfKE`BHAQH
zY>YZ%IDc-Nzxr4H3U9ypCcpA8|0P0ldT>ec5E}W1ajovSI4zy~4}@^$eL*CON928x
znnddi2M>0+efuW6JNLPB_ZByAzRhfYLf6$v9AVSO9eKX}H|>p_z3Xr`zAn}}tZ_6=
zOWSlv>Dhm<$E}<1(zOj)R;-%tXa}N^N}*O9!RR$7wML4_eheYttxu%=h{SuL2+ps>
zXWo01){G}(9)IF--g)aSocGkrWgJmS5&6n0dYoBf3F)8{p}&)T;m~8hePhwz(wMs7
zrKAZi7I?x#=okrw#vxrGuLE*AC08TzY{XzqvQMOPw5E$3HzNsM%)&KI;EudcQAVTh
zvMSiVaGqQ%+U1h_x9>7LKE@izq-MM^;{3%6F@{ojns&jx`wP}4YpjhYXmWPvbB4nq
z!!pNp3tY2+t_5fDItrbF7icXhifI3$68h3piQ<*0K)qRTx))ghdvuIBID$ouv@vtM
zBVb8JGl6KD>jFk8y7cVoR3P^v7T}@cExw7<Kd0EPvwKE-eR^ND*jz-4M|39lPGSsn
zR$(ZSa=_dO<~GpaY4Ld9FenSwM<cd2CTy&as0I~zrs-NoS38XHWTmE56`kqmnvO-i
zWU-jCG1jP2g)B6>AX<>%ajs)g&sfZl2_fbl3$K|?k9hl?U2fjqV>YkRIrtFmfwTsN
z2Cu+t@?5jFQL>y{nq`OaV4Ps#H76!#+6lJD6LbzjON8#?-ef(_M_VH&edOs2k+SFe
zsg^{F=(&iLDY`F$9#X{o*Wjag&?Qb?-vml2R&QrEg1ASg0u^t6-}~45D(nggxr|}9
zHbmn-B~tf5bh4H%kaZ2Ro}*2T%yK*$>*B*%NEywBOPRX-8fhF=?s2m@i`fCeJGRPo
zys=1?5u8V~Q4Q94FiKMlD!Rb{*I9y4WV%3VjZIw(LPS}*^Ju_?)XjttRkLZciJft;
zE)sx6;!;*>AV`I`mc}`35iKRO(hQ3V;RJ24Sn1Ju23jGVKnFok8J&ouhF~?WmW;{~
zlg$dh3@lG(n6BhEzw!<UM^=K&-r<w&7D9DA`N9Q0|ACkJ+An_@wf!Xj>bJkmy?e(j
z!<^a|C}}}j+Rmb6hG_-A^_$<~@NmwDKl~I*=Pa9s*>b@*zVQQQvjv~{#C3e`m^J|$
zO0;OnGH4b{tZ$>-+=O(Z#Zg2Ekyl~57TZN)w-yRa0(=ypJi>P5USsuuYPgB?g2EbH
zfTfaXI;3srPL>ES5g0ORa1Q5WBGj~mOk;G3^Ag<!@De?!aLOY+bVgDeN#`9_JBp%Y
zOi2(PCq0FfXqmBeIb}6L22W4{Cx+~w<m@aR2f<;3WHCG9`ju^#^CfrN6J}lD#29?)
zl9W;m$^kPo!_uEb?~m8?yOk#ozu*5`K6bX``(dB)ehZ-A`Sbs<QC{$p<<il03#^HV
zYo>GN^BLd}qX~JDv$J!DfAkN3fzN&JC;0fsKS@yxR)k=gWhkE|<RJ2IMF=SnTl46(
zNBAp$`7iT>H@?TsTeryi)DeEqY4#uBb?WaW(%*YbwI7`m`jJ&XQX@8tA$HcCX_(JX
z_`#d6@x9kyW9R;T4i5G>IXS|6iw_p(QYLHYN3kC1bv&4TL@!L+(it0zx@2!-T7&7L
z?Z&-3cX{>O-{Ri(ov6&q;@*^J8Cg+~=NVa+qjfCw%Ce*^EAp%)Q$^y<M!Okn9Ri6L
zGL7jR#>Pmo5Tb?A)vH&zbonxE+hR;~Ebz$0$%;xw8^Wq6kM}IzzrK*`i3~jtum8lD
z))Ysib?_J;)wG?K3}m1HWE)V^DIy<}S0e_S>zELzjb+)HxX!GMO*t`h_2^)V)CwUZ
zZ|vY`#;`2F8IJc4!3T1!DaxF!bL%|z_+_+GSm#(QryL*eG9Hf^Pu76~#BD}{0mCxK
z)HCRo(V*Bld?qlF?1~bCqK~9h30w#sgz8J()x@Ju@)04RohMA^gy|gB7^Df9^l>ks
z2q^@>M}1~cDi$X`ko%P59eQz(NHxLflhmVf{Yck)|4@S`d+xO8<IiOD8AL#`FlswN
z=QKtXxS&`%XdD<3r&?uEa&BXtYnQj#-W)L+SMlGvfHwk7f%OnNfeMl=DTzgAX-<~d
zEThZg1VCj1V+~qIdz0z(gt}Q`OviHGkn0I|ZXI&t2lu&i`-l(}QhTgVg<Y=5RD2GK
zTrwJI%5?$4;)B83fb*KhXUv@Eun}x+Oi-C8(*dC(;L~~^$FEZ3rR(c*^ciGdB*6?S
zPp@(07OE8z7f#V|=Mvo}O`VdDZlv*^qNlyjq4<L=>_Q*M75Z$rh{nZi#Ta?wO^0n;
zq-n{lfe`t4);W~-<SO!jbf&5D!72u&ZCh003gpR7B*^HM<dX#0T8mvqLgi>Y=F-(G
zbZsQJid?T?ei1do$Pi6KWk@-P)}Dc!Pdmy8bEn|p#5oWGgjOgtfv7t4#?W<^vMA9i
zW;3?NrO`s{gp^hU@6fWqnh@1LwIx)N<KsD3AHBlzaEJMPffoyGtC&nK(=>CGnzDYb
z<;^$W;fY5-MK@n?^IKoyv6l-r@`jK8^d~uyGp54vt-V*VQXx!*aFWg#@?7xxYj1G>
z{(UZ8y1?Gv1M0fr=xD*#`AuH>z=!$vH-5m~TQh_h;_L}|u5fNamOGjdb0K@<=Dw?l
zCb(_9fBmBC$Bap}6TBzS3uIX_7!DIp%*Bm7ctUXauBBcs@VTcbazvgZx;CAVA+4*Z
zMDl02BzaN9g4{=uK+_nyPC+K4Dk{rS!Jxy8E{yP^WE3>kW%w!vO*h8}+}ztE+q%Tv
zTen%ZEg$;uON@pUufP61iXvm)MJ1k)5zWy0eMc$(dvQs6aZI$-IGN5O#Ud3Up{F;X
z5GW205BZf}`4#HA<wrmHX(nsy2?g|Ic}~|fE8dRx9-Vk5RaNoX&wQ3Y|L6WJ|F^&U
zcj-D4quAaDm*oHF*O{m#p;T)ATpEALl}c@OXr`0_;Q&0-`3XBaclp5^uk)oZ{sspJ
zdo)cG{ohu(t3HikI#vl4L@|Wz;yv%Wj<#vAY09utnYxJVdGGk`uYQf!U;7T#V8C!V
zq%2E%icFuiJsb`h42O(HBi7f~*xKA?WBnXko7<e*I>-9@dKA97NQx-S3IJyvb<;AP
zp0HTVSuExZs)`q0dLi093Al6TZk&VwGEp)*V^cvV&@%LPkEca3krysS<b`@-vEhgb
zEtM+cRL4j9jS-HGH0sTLpqWl%HyJ$Zn`5p&dKIsr^M={10hy6ynzE=EkH=W&8Lv%{
znd125n7eoHbLrwX>*FDt+vhN;S*?Z@mmfJ#UPi8|z)=n~Ya3(AK}l8?WZ4+JXJay8
zb6n!uIl?qCNy#~!hc3i;!CXl4sz6mGQmfT#+*2QV0Yoh75)AAE(mG__qUsiH9HEO5
zZcd0~l@bR-3J@U*G@K9^AJ<LKpNnWwYE1&GJ-bJo`{Hs{KtR8tBT^Qu6cXnXV(h^N
z0pu)eLv0it6~~K?r7!6KF9JoG@!0i?JpR}vu3g>6`vv2T0gFY=blNf;Y;j>b<L1p9
z+`W4fYvw$9{R$hCA(t*(=Edh9W3&d)V!MXp;{~Q`5h_FHF+nF_+1uS?dUB7AwQDTq
z5Nt`6Z8Di`A+=*VKW1Lf2^Lx-Xq&)XG#pQ-JbHbT$+n^x6}Vd+)`1a<I^^7+E;zp-
zS9?L8lT;4z-UK1>Y5i(RG|Em6A6h81mLPTVr3>-rmJ&fo{-Y+M+94#xv0Nc_;tZ`L
zgt)H+=LpWP8nEB*_5L_^U5gBEwTMHIX{;IT6M}OD+eO<36C+llZG!}rB*0SQf{W2Y
z!VuE<z;u>DS<!VJRXK@DCK(MFh4-i|s%X4*G>c^%S53xbgEHE!_&A0#HcG2Sh{F;I
z1n23R7SnWb(>eP<2|*As{>%m+gYTV7DzCUB1s#K~jWO6RcFbiFH_@`HqOGoq#+y>c
zF>Ew?h)D#w&&bt)oA2J`(I0!6bC;If-#Os?r8S<o{uuY~9n<YBP*O3f*72rcV_P8V
zeWaP9-7z2i$deSqoKJu3gV<4toYj2m)thKha&hAv!(qww$FEWiGQe@|u?sx7e?*o^
zzWBv2k!7Cmzwvc$-|$F3!PuI~dWMn#Ej`XH3BkqgQi#O+39F*E>rCX~5P2{%1fW+C
z#)zM?8qyRc!|^zp6?Z22Zv}9kX0fELmt@0nG=G*7yR4(9QzG*^@9@GVRa{h!6;+kG
z#<<R!#?UkllWTOQ$Yenw1cQ)q<Z~9RF+d`-l0&)48@Fz8vVWgj$2HBop%I#oe)5xS
zY_9R@x4+5WgJV=);(ePQs<_cvYccO5#^v9aE0ET3+TW}V%x1Amj;RyTf!=$GHJ+*(
z;ywJzFaI(jz-K@EIf}fDn-#IZ=;N$HI)L(`Bm>6dF@NDN`~`mFH-Ce#fAiZ&8PSH`
z%=6#%)pKd2IPJuxhzR2;IV|oQQl<0Qdpg%~`_{XB=hbg>^TxZ}y?cxO{T;fliEG-&
zJ&)i|GdNQtV_nQ=Z5xMeW15Jy@$EF&J#yHB2c;#Mj8+QPTG~`NJLewaKT9d8>l*g=
zfs~$;6=XUm*9ApUvbD9v`uaK>8=G9bc#)0Gb5SUk3eEX^PTkZj>m~aK`y5ZFESs9n
zT7u{mt{y2uj8MuFG82-gmx^gTA%O9IRkVlrE5&^rAjE^qsmR?$MNaL~g7hArs=-QX
ze7(f%J-{i+MV&JqRb08Y&2+xRHiq@}4c6Ay7>_3$og61ok*6GFcnmHWbg9|Ca+SB=
zdYi1&<ii{*E##0%MJ^ORXdZjwTC^Vu9m1aQ)a4B>ua%6H!_JQou1f-hsOIggjcYzg
zbTvj6<%)1Af;cmZO=~B*s;=Jf4r|QntaCOt142kd@Tc5MA)`rjPqrr#_4OH8xlb%O
zThN_;?^AKb>D-nTXH<wtiEz%Q^&7{yZEKk~fsUMzk2skIi~x~=E}*D9m#&@z6}Wxp
z4tWvK1Bnej`3iT~uq3Mj!!?Bs$LMN{$>xxaa|1RuHwmF;KJzSm!(?rM&I-D=p{ZxM
zV2~Pmn}wp(R71g|PhR86=O#Sy<Ta|1rf!$a77O<GmK+>TQyk-zraNJOc0^fbWV&SS
z!i1@rA;=IR!z=}}##1OwrgF6F5GwNER8psA3i1q0mTC+s$h0Dpf+AB%sch0St5K4;
zwp~0&y?2}%o~Nj`GkZ`-&EHw?*!O$AufncAIKWRfkX42ddWF~!;tV=iL+Bb@SA%H?
zMB*pGML66ktcx!CE032nrje;bI6gXJb8{p9QX<W{y&|tlj5WBXMR@3zHL@&lnIcy?
zQblA>Wtw1ZB=x7SEv$<KM2OD#r>mQu3}QpHOzDilS`&-Z;899MbHL=Xj`Y+`OWPQV
zyo_0Z!!cP=Mnh!b2qGX|pe!`QQc`vxtzvC*f#c&to_P8)dx!f}MaAcS^3x2589R3$
z@U5?XpPO&*a{0m~MnlW){arS$p2y~vi<c(6{=M(=w0asaeD-rMbLG)TdFrWW5Z<u9
zK4E=x!ZeMtU9i-)=Eav@<n24Z&2RnsZ}9amzQVQbA7}I2N62N(=4OGJO&NMkv#6<x
z0-dS&drKM3nQgQ^Y3dGR9AO0&$Bt2=$aTiBC>adKn6ji8jSwoP9|S~<{R^J9UgDi&
zFsRT<(K<)lHqjo%`A88{?o3A;Rk2xC#NhEDa9&`Hr)vXUC(wmOD@QIR)@iI$pbI)R
zq&5X#e&sE8j*mb%4myV(Xg>Lw&oJKD;;Uc&5;t$$jQk>FV~3THS1IK~#<~ChboEDX
z@Bx)LAKpXPSq}D(2_CvGW+kpod@95oRW;zB{nLL!@Se|q{_|*^MT8;K2peOb`aDx>
z?J8R5i6@`rFa4#z#1GzlE24vcNJH5EcR%IJL{runbQeE*wxbpzrWec?Q}*^A@U3rt
zjkn(V0sDKq9G@J~G|MPBN`+mYaT&d>;@_<`be*AV4PEP^+Bvz@pS8D%6M@9bN{AgH
z7FNcXWT6nE^@8kGjWVkD`#C7k$Tk+OM3LFv{w_sPP?ZBVHa0l7eV*ZPL{$x_s)`T-
z##)-T<?#53z5RU-kB*rymXWd$VkDRl0<|Ixr6ML(3Nj^k2p?#zO$XNR9JL3Co}#Jv
z@giEJG(MqVMCy;S7S}8hhkFzc?qLU;h-}1UTrdxoYM4<CbA%3&j3#_a8fvHtNv1*R
zKvrbY1Xl%w^mx}Ig(b_g$VU{4LAf6Dpb?a49-S!GMj4`>a=dc~Z`*jriAaSqCL*v7
zO*R-Ik{wBJeUqs1aV@3AB<S<i<7?@xi7HudBYGSX9Y7?{!epD`32~|=d(-SRiAu%Y
zsaM?@h_1iU{j}=e@~HJqnD_nb42!zOh6*n$90N?q@KT|2K{l3bogcBbF{IE6+iAdn
zl=xsUu0#0_mBZ%tnDLfoFcduU_yyKBMhr$J<p8`@ESF%c;rMup5DF#1dy7dTj^S`b
zRgBr%+C<V3#2BfE9L%TaMnttag7cRybNSJWESEL&#hjy)2Q1qIytM=iVKhLt5;2d!
zPgM?>)kjQnO(>!YS71P%@<Vl^lW8U6NYy!H7K=2MJnf=Aj)=-<=ToB==@-$iN}yDa
z{`acoQw8oShdYhedlmZc_j+H2-TeMt{N_27&XF<)U91+W4<6UHNZVl+b3(mD4Puwk
z8&taB5edQZ&LvDf!U9q$I@6KmIb~6>?CKOK9z&Od6l6uwKmF0iD{VU7NA;wXGU8-P
zfr#)vA!e)@V{kss#-&IeP$8Kh6VW=q8bd~;CYmuuRZEfO6j>xiMv}0jTP$!vu(^Gn
zEHAKK7uBpjAZ&}GrP4LIX?XnVH9qk(FVOlu-g@ghy!7G6c<cLz{J;L6|Ac3ry~N8O
zdy2pIzx+84_vd`$YhUG`|MOqr;~#vSM}G9<oCL|}>Uo|Se2C?;!3NDB&-uyEe~M~U
z(sgsnL4i<^4N8=h+`qrWhhBajZ!M2Jc9lib;RKA<a`I}*NH55><mBLp#|D>ZcHUq-
z&e2(1nASzOxNW1ttZmzV<d1bOidIxqDQlfk76Wuqp^G8DtZ>fKHZ^%(fc2;V^XUon
zg<YnxfJ>Ra!8*K*W5B2kB&r0Ri=$m62!->Jyj!=Hx(x(bvp&qxSxzn16W_&h97<lh
zb;P^(b|^+Ahc-v9XMF54A7{L|$t$mZm$%<~8`D}ufVS(SlH%q49}Z&lKenqEG$;hl
z1{!BMJe*>TPtOoo>lls3Y0R4OFMjEt5%7HOb3aa%#XTY<I)(`8`3${hlR~PTpZ(dN
z<(Gc>m-$z}`s?)eMSmz)icOP=2N)00GwR~Bw}yl+AD<ksyL+E^-g=X7f9vZU9PUxq
zON?pKMxIo!eUq8S(J>0Hqib8b&d|0U*2D$akIv#oo^<0OpcKf2)P>+8A1XvyPx`51
zLDH*K&-}T+d{34S5k)wjMundDfjrAt-`HX@S!1%c#^&ZGgTWvs4miib;UPN@9&m7U
z6r;LU#aZ%K>uE3|j$)xNl#{QQBJQ*9)ZgwsPz5gSx2yY3pbL>slZ!wu16c@o+aZ=y
zj`!}v#3DzQ(E#H>9KZS|ZlEa&l{i@)K6GfQkkTVU$L89AwQ<R4kTa+<*4Kt)dCs6J
z84farm8K{%GMy|=WVBMsgrM|L3??+q0n?*hj&|;lyKbdmbl%f-smK$tu#2x3tMz`m
z&d$7rSllU~X_F^jU)Y^ZdOGv}tJ*A0ykapI*VkEls?{g=|1!My?LB^X{Y#OkFvdil
zv=881V7}<^tm8$62_>E!FF_ZQ!6ak*!i3SJpwI>0R4nQl##xN*u+CtEp&DkaZ*Nfz
z6~l?*(j(_6ikPeuL=xvIiLsX3cW%=)hK-FiCX?Z+=^9VgDDxqc(I)fR0YNH^_uRR&
z8}}g<X$X%!cAj$=2RQF(x|WTT5hh$<F`LmWJ7iZ-$&&fbA^vzyNy*~qgx1EabdRQY
z9~3H%RJD@inI=<eMezzjuaZAj1vVj3$RYyUoBV67l7Vs0I1Q(JUO1&{^sl{u#9{pd
zD(sGLyu);T^Z&!%pFUfbo#&m{Z%unR(;aiH%&Y-G6+n#QC{E%asU=Hl$zgTV7DM5%
z!yO8TBOLyBY)9B(>r-1+Q>2=5hb2*@2!aGa0K{B#)|7MRozAeQHTlC@``nvF07RJp
z*wMs}xOKDcaPB$#?BQMS`#isg-J(R#_pni=ktU=ujqM2ChGDZ}9$bRW=2dRUg4Rev
zV8p!Yh#~24io(z}9d%Wwi#>wE6lkN9Mo);uh|xN+;$kESi5o0FcUZ<~E0x8<TAGYM
zi4@LZy<_Nm5Q!5r7K0%sSCk;)oZOxSf>gwK>u~*muRM};(GNZqC7Y(D*ld!hGG9y!
zqmW3fZBZPts1|fmbMLJab}mKkyz_v|SDxpE7jE(9>$myJSH8~OyKnK!zw(G1xAytF
z|K%^>>@V@(|GR%fdwk4){lEENW9Cze*_6G@`xMg&V&0<a3MasO#qzX8N;o?^!xRNq
zuf4z@fA;fy_76YDAN|pva`(;?KK8NCAc93#E3RB$aPr^*#ub!7(6q-)F6->*wcFK2
z*LAGdo1`&}Nx+(-Z4_luQdV_JyD(FPuF+MA&^i@{DT@HSN5n|C-XJARrxToW2(9q_
zK=eLwpj=1}+~|{7HwNMuIImQW7}fMcx+!**Lh3|jH@cwHJNWi7!(h2T4E)*c_o!zz
z0|Ske{QS@V4X!?OgKxcjo3H-a*NFXqM&QQrQ0|8Oc!LXS4}fif5Tm1OEg`Id?EcO-
zYVSS97(xjA5C6lzLrTG~{KDU&o=ox9GMUa;FIRaCk%|l{6T$uZGuQd2|J8rRm%jXG
z9G{+^-}ryTn=%Sr6zI(AJipkcRrZJ@0nh2_F@O5SKjyVpU*`7h*Eu~sPNs<9^2L1E
z(&R(TI$a;}E~Or<SBXjIeENF2AYL5T<$a%#B7Hw>yE8%v%98A`Bcq7iQ*nrLd(#(N
z9c$`FCv6XqQ7S2FtLrzMoSo$9MvAyQIJ}f@qLV3YH}LLzcUY}f*zh!cFap7)#gJC1
zAPO1NAY>xM3n|gsBuRP9ox~B7m>*@-+>PS@i`9?t5hWaRqbZEUh(NbIqU(C%dPVR(
zFMRl8iPPYkv<UYdlX^;3lq|{xmkw@mczDQSXTfAL<<l=cpKj7dQ<Vh*pE6Kgf(19t
znr6LZeR|6MhY#o;-Dl{|aCS|wXB-BL6p@I)+DP9Agb_qtQSa`jD6Va1F1uibBxcPO
zd7>p^K!!lS-r${03kN$UB4v2(cswHV{Zk-~PEhUf*c7A2lJf#jTqp)F6m?sjU(AM~
zZO1Ps$T&;ewg{CpW4*PUtu3NjVB?hapzwME#!=1;dj}PJ`xTRkLQBm6tdBY#9luRk
z7ffdhib8PX`OEAdESN89FczsDV{HJE7!{_d5yJ5KKmHQ;?%v^(pZGXG`BR_a=B;Ph
zY&M9fP_jhnnmczMuxSN06g0iX7}(vJ1BSP6zroAj8i)vry5Rb)OT6&ni%6C3Ev^;V
zRY&u9Mf+q$ckh_{Up}F84NVu(WdY73%G8h&Qc-Bhq%_z<;eAirHB?2q?~lkdkw2XC
zbZv{vi8ARN$PsPhff3Jx5w`NDoGQe42E^~(U48%dfeO2$H(qCVwxC?>V6;Z4B&G~R
ztn2Z1K=wWTX3cJ1#U)1Ex!_*r18vKP%8xCjr0Y7$vPi|Ebs1=!RzPv9bxUFL-a4Xl
zxU7MaMM|5{Rh5FtNAbM?Q&;H1P}CJ&+cFFTwjXjN(|B^F4~f{$(<Lc^0g-IUeLp}1
zh00^WK-)L;4<0e=1}<H>f-wa$dStX{v8FaB=x$DbHsNsgaRjG)^!cCWPd<N}gKJBE
z`e$F__8WJ2<BfND^Ua4GT$%Ie_zo|=^l5(MpZ{~-d*e2jZeAxS!NwjFouF+k-SH7F
zI=rdyE_DT_vk5zk%e?vKZ9ezeukwHYzx_XW^yrxR&LPi#<l~e|aC!GK)9DV6A3Y*&
zN-mZA(4U~<9jdAzgmjXP8lTl_Mca0H_ap(Y5@3v`EDFl9qAqKcG&m`dra&u2n=ek$
z2cq;4J;z5+sLGPMt`fmrN?hL|W9qn$<L_})<%T{Z<0VpRd=M#`C#GV;`+!m!Q`VSr
zf|dp^4I3Z%#_O-q<1tEd+HCmQU;QPXef}2r?myt$-~1Ls(=j!QXe~aFu`TMaa0vK6
z)-Byst@Rn5kqQSpcur5xwilxmb>-OE*<lz2&JFy|@B9{6&#(XO{{p2Hw(F?t8aoVI
z8HUy=lhp^u-}yU#hu`|G-{Rl@$KU%ad$rmW+l)tmF-B2lXl;m|^?JpF`*-==XaA5l
z-gt$RlP5IIhJI-9&Za^skBqz@vth|5qyalPhThWmE*E~}@oKtOrqSLxWW|pi19Ie7
zzBv>{$<ViX-`lMwPhKeg#tz4JyctJw;y4PFS^t=c*o>oh&uZPX-fR#u1(rvlSguw{
zy6`>rq4Nn1A!9^|1e2G!TPdX47M)`uc=5q5P7IUCQzi~5AP6kILn%RlU|wiKaOlu8
zbZ0zzZ^QG?-efXaFqtiQ=Ee=?(+ShMqErHGj}hsfL<H|0XOA9p+HM&7mZn*=JUwQ;
zJf&H$Sf8D;TCE^>gm*~0VOIy{lOjR9CyJ_&I4_Ad$rHQ-X*8m!DR*{rVoBPsB4bTb
zLPgG!jy^{*rdUSXY_<zwvL=<xxg(F;<G636LTHUD3__-{AX%d*^YL?D$cb&oF2?Z)
zAV1#usCGPNqCdX}NFsv`w2h_h74>w%;A&Qbq!)&2XTtTHyFCA~efF<Nmdg`Pj#^HS
zTUMJD<#fYgwa;u(^O2XHWA|W+DN+Yl=p>jMuOmaBScIE1PuECx7FRfWw8WS>#w^fk
z!E`c3i2`RM@7#XCop&E|y0loYna*d-CW>BwvJNQ<+Eq_Iol#W<IuyM7`aPDLW140~
z7y`5Dl!>g+r6SZ7UeD;PrxT7vS*2-Fbla}07El>MXC%QoY}e8k3Mo9%=^Ry><P^3a
z==wHYLo$j;ZHxW5E$Tuz|Gd5CjMGf=xc!I4v0L7Kk2{ko)#WQJs*<WK(4|Tm5!Vxg
z!)S>N9<7q)J2}|L)j+yn$*r?QqO{IQ7KyrC)pcrtlh}7-`*}>22)X4S1>gcD5|PIy
zG8ca2i&+%8tr-JIfvyUsg=XR-eb>=<T?**6Ls}8V>^jP;Ap{95SfrD}`qZ&<F(C7b
z(uc^&(UMK)SnTbgjUn2BA_k0TsPu|??RYY@xQ(Z-O6HSYCdGogx9@R!^et}Oy2MZa
z?5DW?%pq5=-s0x-hv?EXi4_xBaD49p-bCza!_m<ZlfylxI#SgJD*~s-OWuC_Exz=n
zFLC?L+nk;*aV~Ojd6#FOyM>wTFx$Dr!Mx_ZS0D2B+xJnT#vMsMw7AUr{@1Ch`z+?q
z=M{fiN)N-pa=FUE<AUHLvJhw^D65jmWI{Qaq-2g*;DSV%MBR4IC6tM9pd;Q{9^Svt
z;o(8L@u<Yt8irvjKn*svwY_lZhL|sK5z-AoDnb-E2ZN30sR*V<7A2~03A*C$YQ^d5
zlq(09dHeArKK4_e<U=o9=Ird4uYc|9Jh=CeN@>Di5ivo&l`=?FqNT>ap4z%Uzdhah
zk5P<xrR~CLwOZxTqRXpN&+hIn+Gy5m&+q)s?@*Q#e&H8?F@;{|5#wmW$k#AJ2wb~%
zjo<i<-{A9K{1PW8r+*bMIES4}{(R*U@xgKQ<T3Z|-r<#3zQr4Fyu#7ZW7=+mwLR9R
zag)zT19w3$=e?(G2l~!t=(xwZxJ?roi@S6A)1MD669TyClFj4P7*pI-3=m}`$hdVi
zKRp^9$3&UanP%kMj$$%h<lND%M%fc_nNTM4-+Ow*_no#;lg0ShL4e3IoOnK7yzuJS
z7A@Nc85jS_IAw*PkI*=W5s`_8dZMY606d%ZhO@^HsGljR3q>J;p`&S5tcR9ibw=n`
z*tWy=19q^SogCA*E!K8e+tW8|`nJLL9d>B(&Y`71$$%6Crc{(JP=-h)N|Y2yQW7F|
z@F`bXYNRPKlWC5uWSglgkl8Sj4JO(BIQG<BKV(=j+r=(h=k)opaz0*-D5W!mJ-xT$
z;sF)n(`=dW^uZRxXtD^~<0TbY=@`~oBj{Y9>m5EAd@OOn&<BB4FsVwWiy1q+3#8bv
zUaffi=!D0QHymEtXF3&0yB!?$w8R(%2%JqxL_^=xY&tgUmXqTR%ab*$vlS;tC#=_w
z`TUR=3yvOdARIHD%uy;}d(V@{Yc?xOx9Rc0Qb>(1chKHooWdAMkdd8AGpjYjdc~8|
z6PkX*U|YQJv2NhZfo}r+{T8CZ#R4CCL^OHd<|a_|$S9H{Rzcr~Y>XPPgUC&t)Ikr<
zZym6sQDU4xl8=G7?bMxrW_v`4^F^AJABaKzfeJg{^gMX$EsDcKo|{e6NU$)tZot_{
z6pFH{DU_rD3Yok<$=evtqmlARPYQ|2PuzQtRFYuXj$wq7IcqHmkVRyxRf!~Iicw};
zBnX}7&@uxdL(DDf<Q|J*91ABiM?IZl$`adk_^!oTL9iA#wi0vkCh`d(d=m3VDR3?%
zay$|11<fE4s>so0fx>u*8$@@)zHyw`$TvR!Wq$D={|uL}?Xk0KST8T(mYUbU`G_ul
zm6tw#6&Fv5>gU*5%xKO!CUu?MHl;=!Ebt++99Deel~;N7%~yE&)i-(l)ps~PI>TGR
z-tHwn`qIbP-#b7X&2(DRTe$n+l!tHKVWResp=MY~E|>dEmF0B#1_%3rqDuKd&N<q)
z!}&ndG-<);lQu?2L0KrOsbRjDP|tQSMS*labx=cuD7P9BiAKgqY#TPmPk8RqVZzcN
zPyyV~62p+*XD2W&5~Rx)Es^tOU7DyXfv}MnWpZKXqCV*lG|KmQO70v#Ko*+EXAjwV
z<}xpQ>?Lg9@W#t;a_`;y;5?Rq7G#*GNC<(DR?R;y>;C*0RE$cKFhaf4D!XkR-mkYS
z&LosiDJxZ3vJNZ$&A<IEc6N98#K%8L1e8$NFr;EWCf%J@2J0NZ_N%|fPyggk@*n@x
z?=y;tf0Ub)$#KY7=s@luJ2uUl_uhSn*I#>;Z+zpcJbZYMVd!yw$Q`;-!jN$$HWhWw
zdWK=3+4KxHja|nH+qOo!SSx+Ux+xV!QKW-bc#j+G_Uofuc^m^t8S+SV>~@U{-VwKP
z-YpyN|B#D`?bmYg7*k%R4;KK~f6*J0i&Q`U?6w;DPIo79;q&_jyxEBy1UR4v8Xr=e
z;1Cn|R;8bjZs6$QeRg+taf<~fPaZ?x(w&{sFHdMs9^+QW^y@W4+v5jEvsowXOz<cf
zA$Vd)T{I~ol~EX@&=Qn(%#>gvJre?jILEp8kh<Cgi4+E9N>p8^%-DGTdgOR(iKFvz
z)JUcREM#QKIUCmJrbB9uj2f;;+=}y+PJrlC>?hLzTgZOg9$e>8?epCnMwj-5Vtb3@
z@h&9_g^=U}y$3PHQ{Yh82ws5DsbDe2;H{_Kbex?oIXm6({D+=rGHcLEChdA}S>tgo
z&<~Em4Q$pMPLJ0dJz4VJJC8Via>jbKrft?(XIaem7<$S5d&e9<={Y>OiqR#u_dI&A
zqHO|h$c3|ooodd!sHn8YNU*_D$v`C|>(z$GcaD<QQ+h(|u%V@II{MW>|Il(RPH-4}
zFvOS=I6~gnF-By}s1q6KavWgl(kGUk3^Ao2`NZcECN-^;(rBF~9;u6e-l01$?(+E|
zE}Ro$%&GqVuZACtV}~I&9gkjnmA(Bvc4kwANH$>=OJcP^>jAAKm5Wq}r2o(ckmzWU
zK4A0F9by8K3K20TwdkF9pd^Ec+nwNj(sczNvJh7gWh5#|FF-4W$;`2l%M&ux(`_^K
zNN;>hBaM-zZL~&LsoiRIkyg{8Llp%g#=Jrmcsn3kPcV+I)EKQP3&AjWysSB1X*PpE
zmjPXsIK4y~&%Thf_n&a1yos}ps+#fHZ+x2@x1OP2J?7}qKq0SD7{lrP75#I8FMs}P
zeEM&Eip!S|nM^C*edjIizw$1Ez;rg{KmDiQ=fC^czeCqr_I9495LdV|`y{jK5`80i
z@QUEbNPOHdo#&Nj@cfg1@{c&!zs_slc$Y8!hd<%!q@#WEH3aXmcX)s*3_*G}O~YV4
zO|!z;i0~pMZzxSIHIs>Eva6ZTDvX{bu&7)kLXVdab&2l=IwEL6H4K>b8r`j#U*E?`
zP_jraSl@zg2-FBB2vHEDB?^#U5}m~l9g{-PU=e6SAW_6kjaCK14PY(EUD{1U*F{!J
zbEZ7M_BTIGQ5fEN{g~UY+^1dkXpy*21LH(PqI{qXj+PIKApW1~_P&dV7<IC_a~Bm#
zB=Gpq2iA0qH|P<18VQ)v@!tKn`B(q?zhyC<^W4qnfrybZJ0Kx?fkB~BTspkWKluB9
zpFjDN&(m~me!L%Hfj1U*V`Tya?`WD0U;o;l@%cab9Cz-#%W`>|^q_vg*&)e0tiuj2
zGwM=Du4{)hb{#CXKX-#(yyxZ`@_a}Cx$RtEBXZa4LXaHew9kk9{Y=Imy8#i-^*k5u
z?b}C>;~1Fp0p0n_&mFw?|8o5P?=$bO*Z$l$FKP-e7I{xUQ%|55$z-5`l?zl|&u(2a
ztx67e=RCf57b!K59z5ju=mg^(ZgmRXnn~;_ZHw|A6~QRQEOW$EAt+RG`H!g##E29D
ztw3slEMTXS%u7XCWGs~TSf60d-ll?GDvc@%bXlf?fMotiowTu28gfo>9>aQv9tNbf
zsF-4~yz>xTD%7`oCr__N>!nl(l{9<S4JeUEo8g?t`u<}p29c+`PvcNR2w0cM^VYeX
zA`)`Z>ToVnmIruIVNqBSuvnU*<!pJz(IdsAcHk6Bl$2#;Z~uU{Z-_zQdicgyUQ6!O
zlM|Yz;rfl6^u42PJZGm3U;Xl%gdnyBjxhzD7WD0ghj*8(mMuoTL189nS<q}6x-LK{
zsf(Ix``7sCPk$Um!_c15E?3;X^B(PK!(y*vwtvWzJI8q8kp)O&Il6xbWi&>V7&F0J
zjk5|H(|$@*K#7nsH4|jSNP&?eMUw`XlUg8pn|Qu18AX-I2E#%pX<0tDo*KJy{$8A)
z0r`#F>hPX=Bmctf0~U4y9icxt<-zN(GvC=mnhHPJ!AOHBOH^MHrNoQW6i)bvyx!0H
z4y{rct{D57Qlw_Tx9GA+NR5m{AtvXlK!Cg;IE7>tmJ*}V2-Z1=>yt>)JBPP6=~Lvn
zU3k1Cim<&>A`tmRtT<k;IX+vmn9s6kF>$P#-ZHE_Lz$R&MhQx(5e9-NXa~d5%CfsV
zr>YcMCZ}gTtJzx!hW;4sEZ_dp*ZBwk=$A193h^zz{-yV58;Q3&gr?xVSJ(X1&s}Fx
z&FH!f|N8&^uUIU0`01biX)YZeGW3FbZ~p<xzrc;lSD6;qAWU)A(zQ?)sU>fUk~m0)
zUeopiKlh6-a`l-VuI%sd?LR(aw{R@XlKaQ6ab;1kv$w#Q3hO;h+Ymya?OJReQ%eO!
zA+tuVU^1;K%Mu|KKBg|;Fbs%{un5lKA`HWTaRaN9W9q_C76v6GF{VscJM`%!&hO1&
z9TuCke9jTQBZd^bZH=V0scRq*D49(%Dxq?0fZhfiu<1HJ`SYLQ;P8NZ@7>{**WcvH
z>1lfJNID6nBvDGDQkh}+LDocny|(v{&O)}~r)y1enQu0Ug*e_RB9Vzq#rK<UzRv&a
zU;VfIi~rsKp2ciu%c1h#C6Kr=C?Ozte(hI(jTfH(fAY%fuQ3dk(P;LgcyCgXlP<nW
zDTc1&;ll@f>ziNa3t#+Wj*lMEbqzvz2HP_XeS!%ODgJ91641NvYyxTb7H0#_CuGd_
z(mZy_qTK#}6o2t6+(Ku`eu#6vaoR*ggz<Y91lQtxEStuF7a#Pz*!-@4_x=C)p}ZOI
zFY3lWlt1K{K?^OgbOF`$sH$V86orxKLUQN5cj;~FXg+)SfQf-pXmqn?u02`?RE#JB
z#*8L9k#g6?c~?;ilF(fW&=5;OWdxNHn9*PmlcvzRe53XeB_u{0Oj+Pb%ukVxB&qw7
z#;`u0D=|iJ4!REPkfk9>4&a82>WX6QALn6G?&QcMe-TQh@qc#4reZu?*t6oi>z<2?
z@%sr#oFD(<`<jai?;XRC*r7tr7@S}T9xp9k_pIB2M~@tJ>G|YOJP-B+KM01-b8@s~
ze}BPdv*g~p@ABmFG0Wu{MOm`5x69Sb5S?Lt+VJS!l4wg(S(4CUQh17@!dZv&J;VuJ
zs~FmdAR&T;Fc@jjqDa(@(9^6Qu|9i%!LvUHQ+g^jAnd^UXw9bYK*3}oh*hMRCb~nr
z9BAEy-h&Nk1U=qw#tV^*xv*1`#GX#w(S#t1L>G`T?N_ZeWucK7>K#II8b9TB&d1Ug
z7M-~`35CMBM)rs0ij9B@AZ?&~@R<9rzKSd>sv9>DvpHJTNTmtN;Jm_1iHan!;uusd
zB0A@Hioy_8UU>k*2)y+S!EOb%g4B_7)@2Z;1R0aLA(HuL9_M<7uE$xM#~=yt7}LA5
zb--@%Dw*{(0+Ra>F@?c~$kFMV5J_A&K={be1)476V$$w;84x-_DT6Sav;%h@o$}m`
z1zp#nh!mn=usyR$%l+jWTw1)y-M8N4-ZwAN=_SAPOP}J&qZ8hI?HCbe*rw#|S08iq
z@Q--w&2RB{{^!4lkQ@Hy&;1HluifI|-4*}#-~2xJ-U&Q+?HAds9d0wls2brrA|6VO
zvmL%~6J$07KKiMT@-u(yC%CeAoo|2s74E(9GS4g=&Es#O!hNn^-(xbJBb8)mmo!aJ
z3`rjlVoVnrEii>-GO3x*r%WbOiXvsYMm&Raw0(~*3ZfA7T@OVGAw{}9e)5>bWI~}e
zTBiNh4+CA-kuqt?C<spw=N+1aim)kXPz!_z+iANL0wp!2DRUPV1|r9+Q*JzWi;sTn
zV;mhHarf?h-o5_-8xluMB2x62(pVy0gq}dhkGpk$(RR_v1#EAM0CsRRM3hPlL}Q8s
zl`RE@D){Qxzshg__HXgee)FFrkticst=8=A>=C?Ai#;K@_3SNv{^$QDue|nZ1{TNB
z)cHf-E}|d0w&m`f_ju>+w|L|A*IAyOZm~%t5^1nDbu+SVZpZ-hp?CD1OBs_slCSyW
zUWDyU^ukK~eOjvX?fjxH1lymFxf%=at9nycZ!7}yH}X_dY)63?{_KU(+|&D5T=;Lk
zx&1)L<ASGN|5JZ0W_n;e=Ne`aw2LqReH1h<aJFd}ydz3Q<0XTaSRWXM9`7yH#4syM
zgjdX>!bDIcMx)9LjB%fhZqiifCCULJ5@p1z^z)Jw=@nwc$B6YQ_A6wa_PF3wT5Zi>
z<l{=j^jMMbl#wN7%vKmITF3;v)iOI|^Ba`kv-4#WC?QeCpmfrsCrEXAvq!Dus2v{*
zw|HUv_x{};N%_HB+y=?r9P%+NO+^SPt}`l+qa#?@Yy`{Gf@med3ZfH4FX%f<VRmrF
zqu8)nrOfHYe1|KSuClYU$7XGjtZ~*OP{cUKnq@$^G(gJWjZqSuOk-#SF-W8yb%Bl;
zY|1|`6upYLe#LYeP=&`7o}x@<Ej48Ar}K!ABE6&*c-+=;JihjevIP)`7*G<7k!UT_
zbBlo(WSRu#T0_e;NtD@e7Gr8SIX}LqY#U3A-BnUSC(;cr8;&j@Y`)L!gJmW~iHecR
zgWNP6-+mK}<g$)TpLvE@qzIBAHC`LMREVe$VZ0zD0iv^gHcupVlOT2Ns_BH&lM_5i
zyeWyuT=k~32$2ppmxFYBi)(wjuBU5T`mUw#2fWRp);VKMz<I?h1xjm*vPf7JqtRsv
zy1)xXK;fKVxopw9iip7XJ$>)+==5gy9v=g~j06Q%B+m4cvnAH`lsY0rMequ|MTjTt
z&J63#8&u*szWv!R^K<{?S11a@FaPStIbHoJj~_Y+H7f4X_kC(LE6X$2cR4&<@Y=Uu
z=6C+PKj52Rzk`v7bQ_QG1$O8#Rf3HMU(vQ{9v?$Pq~W74KF3f0+>3NW&+FfQn?L)*
zuTpg#!oJ7py)W_1p=EzZgOm(|V|liwYkNYliNG(xXhBtJX0w|4e8F@wp)e-#0Q1Pg
zI!ikY%uRt4ilOT%Oo{NG=m+}siv7g`CDKaL_9K|JPe~yNgi4vQF&k8TpOa34L#c??
zBC))-g<a06i~>6Zyp%j%ouX$oANklz^j*uz@iDjGeveIW5iw#!BFk4oroul)dIBwx
zLGW}Z<Hy^d+eUHqcu(*lbs=O<7jV|I*>ni80+NQ#C{3`Ax}5O`fAk03c=i^*@+-fB
z56Nk3hk-&Dj0FF}6#V+H|2n_*+rQ0|qa)x)Q`AAeVaCMhJ34y8?Kj@wt+(Ff-o3jC
zUy_-0)(yG1bK9hdRLl(wgJ<X+ZU`f(Vp}|&r_PSh;|toS?>TDu4nMQ)o{glf^u?5#
zlfM<k1$v5jsxY|tIPt<`z5j1+pXd9Y-e35Zy#JxT=P!t-e(-z`T>x?9T^B+s?s^1$
zfMwUy*@1w-$^unPlHOVi2G^sMp*97|6->Ry2!R%Pgv~eti`$}3$a8F#l#&=dUO-g&
zrk!7X3?X&veEOUckH*4IZCR_MKJH>pYkX>_rv-)gDYjAxfyoG#6oVPI`mR(XUO<!x
zkqWyk$`9V3A5Zyb$Bxt3?;8ugi<pzK&=-iL6V1%EQE!?wv;rl|RCozMc^n=YBz>=F
z);+ch>9}_gy`bwXQWR9hoN3)tnF-~j<jUnMT)TFi`FugJn`EfW;Q?FJO-wIBh;uYg
z%E6WSm>)B7A?2Er@?n)gl!~fWY=(jPLZx`nLZWL!QI_aF-@AfEAdylbwIqm`FgZCh
z_o7iLX%__}%g$(tQEC6gF|h{Di#nyzD!(sz^ltOMP3P0tMLjRt#)j6#qV7CR=7;2p
zl~Pd>sf0ibmb1r?xi11@TA?b##FTg}(r9AT1fd8)Pza3=LlT*eXp4NZiU@-pn9gPx
zV>mlGg%IeP4V_6l24nD9R66uMLqE{B9d6Uptu}1d>ojX`+Z4m%{dQ-Q;l+aFc#8sb
zVK7yhBXA7WY)Ua*AiG3m-{0Tm<oJ}PX_*#fifYjkIZ-$raV~<(<CW6T35&(iH5>YN
z$?nb$k|OB`<ABivi-}^<3zRtI{^^Q8{k?DS3xBU9NXsw$+Q<3gpT5Z(uRTJG$j}TZ
zz2Nu%!x#AUXFkp!{r;cvyTARPIeP4ne#X#wf&~#hRT+@7!Kj+v`lM7+3TYHS`IDdI
z7k}v!Jah93ufKAK&;GmLhleYkF_HUkeVvIo;@0&`OlJ|L16^ZjnoZ_vd4xy@a8((m
z(~9|g%5*xTtg9@}&fsZh>4$+$+p=S-RNVWRGwufb(1Ek;&8OfTfyD6|h8~+Ecw)XN
zo;&W+$X*drERmLpU8Mp}kh;DqWfY63aR^S@4YwcO<@$#|%ycp1<n)xc-@MJyqho?k
zH;ck(3ZXDUG7RY$^bz_9ZoKFIH7oAM)!=u6GGDylY=M~~Yha#k-fcCpL+{woVswEi
zHA=48S?thsYuecI+yC~r`0z(Q#I>u}C{4-G4-}>(_<+ne#Ls-@Gd%yo^E`fXluZ^t
znl~lHz+eYfXG^~Pr+>;TFMpfk<0n|xXLz|yqt`y^>V`hUN3EmT^t5fyFnC;|2IU*E
zJnyz_E0puH?);UVcdNeJ_MP3N7rviAE>3mBQ#bbWk37s^yoo;j$l`tf7oRc##P9l(
z7ar&PD&)r2;du*je6Aw*&)<JaMsi^%Bwj1J;OU)b@R6uIrPNe)&CbCt%hf4OE!it-
zc6S#{`iRD&L_kTO7Q`WU1w}5oWj-fV{(?w+GEdB;hjBp({gCvN$rO{GpVAU-4BF^z
zAr`iap{J^R#%eXj96M!uw33uXK~<FaVMxfWvB-^L+x?3n;zgDb6=gD#<p?r(Ujew?
zuUY*Xvt!%&r!Z5H;Pm4?R0wiWmpgXCY>_oeM1t*+l*ABmL*VSBWqsDuH6A+zmd6|V
zHd0qpu3RpeRC7>~o!uFW-8tS`g17V;H{+epT2YyE#$$@Ei}8FBsT9E_?Q0A!4FaV=
zsKoTM^r#|G%@k5isp>h19RdsbP@%L+@MUK)9+c7)rofmIEgeEAgwhD~w#A*go#)d&
ztu=))7_C#pYR19{B~!CV=b|q21$_*e7dhZuNFBmF0w2$Zr|?8lA%x1B-w_Wt0*(LT
z?E@8dB1EJbc{l-_r#U*|;TyN9XLC$B$CN2rMrg(9*_!3Mk1=J<Y*sLzM2MbgnFPvi
za0KU=%%*J4R%mTl9iKASCBE(1#50UGSfv>HA?aEAfv#y;uU53nhHl-au1{VXkGcVs
zkp&_~Le@-#m`-5leG1nmEl@X2gDh4QRfQ=Ft{(0&tu#lEpWuzAtSSnvDT@l93!t#H
zSO#2a7}gOOn3WOHH;6D#`Ln@Os7MqYld9nI!tvIFx462u$7}CA;48oX1%C2ZKE<;)
z57-PXL3!@qeTds=4zJweo!bq6{Cju#(igu;q4(&|EJ{p~Qlmx6@e46hlm)GAFeS)x
zLsct&;-k;<)1P?>IRsw+@|%3|559_fc*cizraXN43l#2%4?Vxf&Fg#k=vZ%#XxEm_
zW`%PDN=1y;6h=|k6_e?d>1;+>)@V~8M4tt^0Xqy>Yw4{+ngZ)Wk{wu&P!ReKY)_>X
zdaw++<LYu}HH3h}Q%D$fAR%9n((1+`Wk4HHI<>Sfo>(C0ec-kC-o@93%g^1QwH+r%
zM?AjwBsr6!pwdb3>Yc+1i6h`h7sqYk@Yk%U1Gen5u+`j%Z2|ZGmE`#Kluwy-MdM95
z8KKgRply4ETw#iI4wgkhSyg!Fc=yiR{MNtuE&iMT?SDgMDv02$O&N>%@vmID!q5Kf
z&+?V8d?jZu|7hNt&H9W-5AN~pZ+(+jUw)bOdWrKL&bP@Z;Zq^kcLPJ0@?HDh(|49;
zJuujjvG^Sx{HcoP{e{i=@9**|zei#Foqp@j|Nq6~JAT)n`)-dA8L1h6{x|;e@5@Lj
zPW$3Qk1jp_*hxbqr(wQE*cfPoWTG@$D)2o<354%)t)uM?dsRj6Bg@XSm`^ARNufQ3
zvKY~$T!;0kQ1nDfX;F9(k)Sn>BqtFuprS?@gV2dZ=Yn8xg06?5C+Rqfh$s}YFlbd0
z{n&BRIH}MgAaDuL9DEwzMTkCPZHEkr5+EWJs>E+Pgtf@%aYA9GL<EpNAVq>x>vD=J
zrwCOde2*59C;~x5obX#`vIzP5DlT*g$K-^JRY~%$nB^`hRxh|7H+bv-qO-WlgONxb
z5iwo6hZwO))U+UIPm}^FCsfrTyZbvB4MX2k&#tguFA+*(U1S(MFMsPz?%#i$Qh<b@
z&^ebmdcx3Sqb5iTDt+ISnrc>1oAmxpcQnOxN;#P_o$ey4%dD50lcnS6NlUaHKEU8w
zPEQ}QT6tnjD7GL5j7Z5k!V5wVBAS$NB)}u^F`|7kz=S+X7vsbTFhV9`Mm`Z$DUnKm
zObbOf*vy(t5)Q~Rv#gESCIyfDT%{3GgHT&s&Nv$Wezy-+*pV>@GSBywLRin~od-Os
z7Dzq8T-!l7i_(h0dcO9RZ{Qu=eC`U*K7WbQD4H%+;K~@Lvl&kwzK6G#)1zYwr5UOY
z-}JOykJpAa2HK{jYg^Xq4b6H@*L8T8E(Atv>Y~ONLtzS(NWyxRfsW3*EzvoIKtJ?b
z5u<k=7kfl+@m))+Bt>1ZGn-N<!O6)nct>59lqJZjK+B3{Gtdkhx=?Y}b_fC2<~3^Q
z@q<Mgi_{URB`O+hZ>dbrm0ii*<F9jT`Uy^NAM=Mt@9;PN_OJ4>4?WAR8_)Ca|NZar
z&Rb_ZxO>W<{>g1hHODpkxKLr(-2yP%_Ka#`Fsh>O22|lWxHjd(FJ9%7pZp09_jVbU
zmXmvD{NBI)_tf3Mb5qHqH~urq-sYn(?(?A+rpzZYqbCL)KYW699YT&LwxBEvCX)%X
z>6}SDrO;!jwMeTpKhWDgA!%YHYK`-8+c6a)V!MVy3WW6xT}x*jP1DgdZCdSk$KZOT
zHbmhQ6Hz3gX|O%c4QM6s5|m6MkHRR7(G;d63c=B8#k)_QaOK615h}yU(J}Adew*fO
zO_Clerb?o+j%kRFfk;Q{*2VD-^4GAk8?E+Z5jTP{&tt99zsKS)WMct&<d<#|>4vMv
z@l{BU(9@&SGy*pok53VEQSirq{6~E9V;|>Ne)(5XO3}7$ijIq^c$7l&i@)%T{L6p&
zFIlbDKQfKoJ8!?u$?*|)-g}3pUE_S>W2H`Aj~zy|iKp*V3P;}!2|R2=($f98#^-l>
z;^!T^^MC(U9edy3-~GWpyYE(to)@`c`;~OxF-kK8yB&=MB0~shao%dHN<&$dNFnLF
z0p}&w8+s3$0c6ZPF)1lj8j~8CZmlkOv<PToGW1D7K&1{!j1WaOT#<CvAut56L8Mfi
zu~n)hLMx=uX@RQ*K}e#I$TXHny7Lh5A@k=#gkVv^rEYjk&(-$>!dpc2iBTw$fl-Et
z$_6H-3Y4ypDrt>nNGwBSWR$C=Ic4rVsXSU73p?*}qDoGO8ku(ED1IcTgv_@Ko`}l2
z=tzhXA0$H%*yz|fm~po0IbCg7oozTdK4UsFNGT`^143YJS|B~Rf6Tr6PgpNIVn|Om
zgoqT$kt~GA#||GmV8AHNq}E)$w8y2L856tYnHMfYJ)sGb_wJl>x^8*+@QizRj#-@z
z_#tEsY)jt@dTR;MA*4s>z+_sc20@fWpLKjuB*c#(n@du2KtLLk$`lDErAj(Sk&h{@
z(fRtB0MuDKnHw0yQRY8_!&NqRrov9jlu<kyD7GN!_?;J3-);Lqg`Gk`zQOt^2ndvr
z*v-J>+wbAEW@etpxB(vq7PAHOdd9;C$2`1uL_O79x_m$_1<t2}tKA&2I$N?jJ)>Q1
z_{ek5vwnO^7<vZV<Bg$fJJzceeb>=-JtD>=#GlNlCo?8l^OZ6IQxzRzN|8W<^_gLp
ziv>5NsFclWP1m(aEZPtB{RU|iog1)ygQ*G*5B4}aJ*DXylrR`sG1!_-r#Wd1kNcLR
zWn>|1u1;nYE?~P3trV)zgqWOiN*E+PI}7l!;pFs-93Fm*SMRrc;op9d>z`_wUt4hY
z=n=p4Ge3nAlArzQPxJ0O4{&zP&^eSe#4vz})KklJCh&vf#?5CKV#`1MU;SOWp<!MX
zJbdQ~uYUDq?!9rB>0o(r&v5^}FF^ZMKK9~*Ti0r;!qByjC?%WKhV^Di#8VU+B~zqZ
zT~C-yW>nP#t;?JM;qe4i2w0n<XPkHR&LNdfs1hLv-hmMKu15iQhZ_dY&Q{!c_?W)8
z%x~RF#Y4#BzDS~-n-pV{I8K40(0DBpen6(IKAklTAqwt3eu9|R%&s1CIy5|b@|gSg
z9-%!r1eGc%VrC&~MeiM*gwDgT9Yg)aqp<%eZWrBIThr2o(d@{X9RHnuC*vSwD)Psl
z_1@<Tp@f_ZIt&AC+n~zCyOKiDbuGX9yT8jvo_~oOw{GHdo--HpO`rJ0$GLIi1~0$-
zYBr?(@ZXoW-@47o@e#+zM>s#Em39svABLWO7#M~G86FF|z8g~{#&O#RSs?uNI@bQ;
z?K|%eArk8_MH}YgLx_YLAsZcppfsArv_@&oGNha8;5@yHG&Z85pwM8VqE;n@#2Aw?
zBD_W8k~>?FMD#)+qe|TslExS5Gha30e8S9xIK~m?Oi`6?&SMl*`t^j_5+fiw<N^zk
zBd8Kf&l7X<1UYiik@|@Np9X|TK!u1WnG2*UP`XHkLbTiB49O_1d}=&D>h*k<-K6Qy
zx!&+8b4B(&5@UMp!4XkgQvfJ@Q1nA&wb`)Pn=(6HBOab{dV0d$yON#Vn#rVQI*WM`
zG|+Y}@4Wqhlj9Y^kzX14RRY2xq(f+li4hefMPZoKimQh+ZXE959-r~t^&R|l&f|^e
z>)(EhWwYer(GhnZ9MNq&eh@h8@j13n=?Kyyl|vRVnNQGKGjx&QMP^J2f=tDokP0a-
zs6TUIH|pP{kZ6%Kb%oYB`6>k?Bv`r6=%I1_A-B1*G8b@T20gjR+iG$1dl1D3yHb3x
zj$N)^qsSx8n8d*vLwB;_{u^(zJC|JFmpJ37W;339_7-QyYx-_rv+gM;Jx`80*6S9d
z6+=6)JUzvTz=v)=gWdF$bw#j_<;f{Jk74_+r7SBBukACP&6rGQ$g)BgWzyi}aZSiE
zR(;=L^GznmG=rB)p@Fg}s3$c$JByS)(R8$DOEzu85YmOi+a82xHlMM-zr)$-8E2;r
zG?sP{oU|n;L&1B?4V^b!oz_fEO&nGP=g>0X!T?cH>WVl7CY7cQ9s7%(iC^)~J1d?&
z{98QsC2xK46}rk%RFM~NzQDEX`@DVkRet`LKhDvkHK!*neIF>xn$P|5AM#)R<9|qu
z1CwgOd+$Es-~Wft@+bfKo7}u{jn97VtB9ecKYherskpW%IC|xe=uW=M$6hpCyXq;*
z9%~_Z#j0tz|L`G7fhjUoTPr4Y&2%!OuBVhmg%pV)=kkR|hKTp6h)=i^pW=r^>XwF(
zolM^2t)=f;9zJ-$?RW2TwrtqHbb!jC=<dSo%6pGciE=+=Ez@Y2Q27i~T4s#`teTdS
zw&BXN&(cW2>U7CF@4idh^^_56Ss;aC=sTp0NNun&(#HfmP0jJNY7c)6Z?tl}$Vj#Q
z_m}}aeqV|-CY4*Zq(dOnm{fq)I%7iEhFd%5u{N?<HwY~$>VmR#S-&K?{nnfO-tYfM
z{@FkKO{5e!YtdRKXFYIuaL7w9y~Hc8yo&!3xmsR-?Nvh799f$%4}G5uT5Z$PwjEtJ
z(6$|$O+&NkvDW92^tqewyE|Ney#4>PrSX4?WQrM3FUU1VjKn;-F<P^`v%_pYXMcB(
zx8Hn+P1m6!EQ*psM4G{(m1HrSgHZ`a>@7-eka7;f1D*m;1w>(pnGdH$(t5j)wA|i?
zEpU2F@et}l(R0D{Fm@Sb9J_X-)-HX>7$~%2$dK<40<P-`&LwbiuCS8fkgWzzYlSHc
zrYI5Gq@3Hlf20t}5WrS>xBYL*gLT_naPI=age_++hM2lX+uul9>?j6=P$-!aBm^aH
z(5y}!uYBuGe(o1O%>F^m;&J5O{rC9#*X~eOhPtk~a`g({2To2-IX*gN=qw?meAqMr
z&<L6Sh8G=Db?ol%FspW`rD5W0PM$0&V}n}_Ji7e~H*U_Uuh(c<(yW(wPC%W3+F-pw
z$~npekOTd2ni>x>qO_o>3KqL_Hcd*uh)&>~&tw`&44|~h#-U`MxrlTeH>pMdaX}=>
zq+A$Hoay)&B`hL$`cf@5?%fL}jS$j;`J%?+11@YoSYf9EI)bn=G3tcEiOArS?OUF#
zIezm#yYe#A1B2Ia<GD+$&n|QOt$W;g^8vTtJ>s~HoSd{2C~7Tu@%epTeDNl;QlY$I
zb-LpC_%4M}Osks86x5e5GufF@O{b_TAvI(mMSI1FlxfHKklV_5tRE0QZXJ8UrwA~e
zu?j`0D2gdzQnIr<XT4shk7h$}2k1SmTT@mA2a7!(ixm&g22Ps5age;%T9!Ui>z=Y$
z(~1d`qDBXY$?;SIjnBY#?*^ny+BLbWxEy=#KK?zX)lEJ$zs0GqdGti^hrj)Yy!4sB
z!NGGo2ntMX_{onxPeAkN@nfz(IN+n7zQq?m|5fhYJK@pY4Wd)r`Nn&+Zyw`Lj=4O8
z%exhi?|z%t-~J+(iU)l1BL`f&y1*Dg^pU|0fS?;T)RPM1ZQ?QcNLeU$7Zavc#iXh!
zbb&-;fgwt83hP_ixME;~^$v|edw~iTVviL99TlP-2#w?RyLY+$##^k9TP{r|T$(OY
z91b#FTtZ+79v3{qCI_qY1-_3I1U4a}JYXE6)@WHEz2o6}#b&<4<kDsQvcum$<?*9a
z+N{}|E5n)hbWu|xnMQ;44pcgB9i-NG95J9jpd0e{fQ4_Z?P1hV{OBOO7q3Z5r4KP)
z>{MEX4&6X$)HdY>QmTXx9x1Z>orp1Jn>ryx!XShQ#uyM9=LB}+k;YQj1Clkf*_2@z
zD0Ip1eeVCu&;H`i@cgY0r`4+m92B~sDk^^Ymwt)={&#;jMKApbc2Lu7kWyfCOk&qI
z3`3W?b^XA4y{6f;Y&IQTGvuOf+^gHG^pCe6;5O>P`mCk?PAb1hTD)(XmgQ<qS=C&9
z<{5~ZlgE$QY&P6Ke!@^HnmEvGQd-T<?lq=5Qs@CSK-2;s9Rf#5M<u~xa1;atl8AT7
z4Q>aYczG_3;6+T1U8NJJHO81@Ek_V>N?;o045Kqv#F%uzgH74jI*IS4gtl#PLoV*N
zBhvJkfdCSeF_^MUhQOl4MN7ct7M&+uh;o;3Yyycq&h{y&-ud$@kPjh`4kgOuRuzF5
zEntztW%EgbJZ~D049RHMtQ;ptPnb>xAN}}chPY%tol#X4Rb8=nFhNSmd~Vp^o6@u$
zU7OHAeedu-vT-d+Im#-sw>RaNf90pxTg=claPsyMtDPEc9kN+bhL+Hrq39?}gHj{%
zL{pX(Cw_wv0b`(?1S%=$dJ712WmxRh%oht*>&LX40iz~Jm!nA`>1A_`AqbG2&Lgfw
z<}?YZ5{AtMM{psz=0l<`s9ekmAxJUvDN;4X0Qyu*#pKeCB9TW3VQV28N3m(%|NU(r
z40j@vGfL*0Q2<;_ox2E_0L_y#j^>itWFH@DX0wWqe)I+Q_x3qjc05{j3}+i$C(%K3
z^V$wCe&iZTH8jnJ*h_YI_t-o5Fh!x5O{Pdh;-g83T44zc-l9X|>LtffcJ-*7B!Eg{
zNau+5+|3ta#Ce<Es2s`*$tgSATTsuZOqzz(YQ=iJqVE%cw5%)YS<R(tpL?&Lac_Cb
zk@EzhDd|BDI6NLf({8BSIYm(e5=VMY=_Cg&BuK~pj-sw44<5hGll%8s?0$&Z;sz(p
zV{U)xbG-6ZK_47@hx>e0ew}hsGnr1gR$b%UpS{haH;%aT)?+p&1M^<6t1O4K99-9&
zJ^BXky!|z7dzTlk6+D0KD*JmoC?yHe<Dx@II9s054^0kYPpc*+6?IuNn=Y7?Q%X}J
zb9W|qm)rFMVFkf~_W}_L3Sm$b2$v#goK_4j&^=l6?kjI`@Ag}qoHktDnR4^$4QBHl
zTu~B<^u5K^DVuVH)()Mg?=3^yfgd17G$LJWR74^f?7)+Djo5#NhDg6@dHmjeY&YNn
zXj3wHPbUM@{mb}GL+|?3@|J=$RL+Ig5Bi4uzVmRKR245SzP`sF{UJ?;NAizS0;Ljj
zE9Et(ouH*b3x(DO81-|h2Vm`hbDqLz+?IzGE*LH1_9LxO3qo+T&A@880Rmd5zZq~m
zczmDV{r~(s{*V8M|5GXsNi54KB!$v^>JuMlKA&-Na`uDW3EyQK28)vExN6%bIbR1$
z*Yz|_OS5TdnvT95@>nqaUV2|MJoZQA;Qi~g#XMHMI3i3pyz_3D6cTR-)~hu}8)owb
zm#<x=C=9FPW16F5j@At`;c-?G&}{q;R~9upMS)fmya-7@hCpDG^HV4yBlI_E>r%2v
zF6zShc^zV+M<}IpS1jacw)D*?Mjcz$Ad`PLL-i5){*~R~&e`M~b$-jEix-YFX_Bs0
z8-p?hvM7kb5Mt6Jix81{0Q~;Lava5`!gSPtjUa2WU8_i$u9b*nauVhE`+=&|bX|TQ
zLSVCQc;)49QBFLvh%9y_FTQkxot+)3y22C%5R-hNvnXxoTgQ6UuwJ(egX64S()BBp
zw0z_vSGjiMkVRcXYgyfS#NPfCyIj(5dW3M;p+O4EWTv@%d5_XeP;!&A=LI`^6ZQ`l
zOy?!e54`-X*9Z~HTC;bsV7*zhUN!Ws#R#1M(@Ig~4zNgK!ZaQg`S_Cpty34$NU_x#
z4}&9ki}i{AAcREed~K8>PqfoiaooXiYfMX>{*)&iFT77N{rw+S$4-)8K(6{jrdP)h
zQAi4@=r#?H?sXjQYvQFDQ3-Z;7fdG$F7=*+<(lH%d%V}RMCVyd1c$p*s-nP1&=hIb
zu07E^EP!+LDcd8VRHVpgj5J0|Cs6t<MV@;X0vR&AGtaow&hfeUjYMw|8Dbn`pPZIO
z!E9&FWIknY*k`lZ&~-hLz;t)Ujm1^oy#ETD$43Y`%WrmyM#2v$RbXAjxq-=a1}I!_
zpQ;=Lkx~UpV4^hD<%*p;a_9I}9^ZR~$>A+7J@a82yT`JPoP6_5HeH9*25k(%dEQYP
ziDXBWd}x1<{oM-JJmKWgw|L{*Uqkpi?9Ljlf4Jc4<sIr`iqha*j|&bLEY>-OzGMGj
zpT6(tx-~Kys%pYwzR#qZrFMHvS{83TF|-i#MPP7v2MB`*0;3GXG*S&t6NMuB$n7`Y
z;TvE2CPi$ybajW9o_mg)dsm2+qMPp^g`(|NESzU>DV$rE1<s|YooIWu;hMoy8A(}4
zbYWPwJu4#BrF|ZC9mmThj~*WpZ9<zwsaU%XJ(*HZDw@8-w#k0pyM#JPtJrac^n=@^
z9GO>Ho{}X4+Y6=;3H<p3{j1-56QqL9m~^ZglR~bQO5M4nQBCFtrBv42rBo3q)410A
zfwC$QG3n>hYn9v27ttuG=!3o=*leJz3{BH9oz_?zFs9@SU-&$~`b)pYXFl~A`k|*Z
zRm$s>k{dT}a{c;sPS4K9L+gjSIco`0B&10<rb_g*O-Hk7Y1^K@%bmLSHFw7w?vJ-0
z&K5?w##ne<C~!ucNC=T;(;|grIy>P*FI_=vMPUqmvtf0#B&vk(aK57rOEeO_s3@xm
zfgu-+snaDyLYxG0lWW9;_>l6#{nMGhV__$iIxoP|e2th>(=T%EQqo0EqVTDB8%NGc
zDSV$$8o~RNs-WU_yc<N)6-uo$%g&&R5<;J#6*z=VT)W7alb(y>+_m!=dE#BzM$nBZ
zP2<&2DuGhzZ+UB>NO5(+d&;Un>p+YtU$^U8-gx6RRM|4!F?{4>ALHiphg`aRNIj`(
zn--~M`g<Oxvz>%VIa{$jThsRgr<(=K<uTrMyzt>$EOw^UO6Re0%VNGqYZZfz6m?0!
z;X@A$%w{#gffOq7R-Z9kxxUY(%Lhzm6VA?7y#B_UAYd{#EOw?mIeScS5hV-?l{=A=
z!XzFck(~5FlA(pn*EO2Dt`RY8O(@P<oKFl`WTKTI1yNp1{cw3fI2L>oIW8yU7<(Eo
z+&eC=9>2fsgPSMv%s#}JK9hy5QlT@KWqr2d?Bo<@uOPL?yG&>I0U0~0*ieKu#wes;
z(zj>q&38~DQs|PC(<R<!fo!IGYNaTOf?yp!#=PSs<ExMPc>o?k;l2nhK`EqGlv+~Q
z0<7DLyT^-z_eo0^Lfj%Z+F_t~4wZ$BT5ERp_t+IO6=S8u8^fi8IpurDY=TD!g;qsM
zQ?MNbh4+E3>zU3RbzS0uqqjrulBmqy2^3o5hK`Am)CU!dndU|ZkI&xXy)XS4$QvA7
zd5&ie_V5v`Pb-udK+AxZnr=Aa(OaM6!SWFztk81Jl^wWzSaWHAhe?%~JC3BKkWy_=
zz?0JxE?>T!x_BWWN|Z3nCUd6sH2tiU`J$6F10k-kvBfz_>;uiv)9aKc8x&wV(2mV!
zjhIXbQE>G5h|Q+u`Hvj%;g@c4d9jD^0*gQ>gHRgp(gnrYfuZjxjYbd&w#Bw*h~Sc;
zLrbbcqGV)<o`;(TxxdFoL{3*L9vvOg^%fJ;;oUnARmE&)m(%5vV22d0EhIK1@aTBb
zi1W+vUwRYc`4XWli)7Q+X%!zs$XSdw+ug?t?hn5(kTQ))%d(&-N{lf`DKSP9LZB#0
zw9*XakW36p()R;Ob!ogxL`sD=n%)jiMdSDjI?2&R>x0V$p~Tw2X5AvCpsp)oj95Fc
zvS<9q-}?`I<U=2#uBIrNsQZXWRaJcW!yo45mtWyWG2<MDo*1F;JBGosTCM5Ymagv;
zyu0gDhUbNCgrxp>`_XTLWVrG^(DglrfQ_EjGVty@@3MdSDhK;}Or{fz5q$egr!?LZ
zy+*2n(??HmgJtM<DQ{d#-Ithf7%oVH5-BAGurbL*Z0>S-AJfDln<r%Mib-`&WEk~x
zQH-XEv1OnCJx=dqLOY2#LirVi(S)YUuFAwoOZvQU!FeqaMk7ptEVAZ2R^SCbND|a3
zIb36I4{m=oCZM>>fNf`e;(bbmR52HNA#>XlMl1T>qG~~O0oQj_J2l1(cx0O7Im^Af
zcZhn${*@`8`MFP{YfCW=l#{@d<40V+a+#{GD9Qp|REVh9YYgqCW*9uyL%V6Peb3HL
z3Bl3y9nIMpXX{fAFCP#W6~XQ>H!omzYL1u3tWM83J$cN7JEzPSd%W}$FYuw4ZlFs+
z-+N9^H#~mhUCP=pnHB8qPcbFz?eA0Seb%+($(tLD3>fLr!XriE-3e(yj{7~Jq{OHq
z0oJ=78#2$Z?}!A1k|?1PI9evub_m%_5pow7vAxGCt<wAAQ+H5`G?pDNdGW$^_WRjB
zpn2l_0+z02f+&*vREd=9s+7T5&)M0AzPE%zqD1EGd4~u+ivlh$Qu@PmDmi+17Zp18
z_jge;wMV5X@WC_m12&ICdTX~`1f?`V<&~5asL|1<(~gt~6b4a-RM`2nV?szRbh!;Q
zzmRk4ax{rZ(A~ifTT8PPX%%pGdY@SZ*Y>6?pY#M2g)C5{m6gj>Y3Bw)u#}}qOgnE=
zJeN{igOL>}d^)xrM#EG@s#3Ez*Ie6QuxZvjIr<acyLSds(Yb(kX;elKxWQ4@mck6&
zIH;J<3U=lbW_8K5o}f|5(Io^ntkX?4dX~#2-OzLO>Q(AV&5-x2l7gZrSuEz%(<*h8
zLcn@QKlF5c$KcLrLyPStp%JY6o{jRzLEvM_#<%QBi?t0VDr{%zno}<A*WA2undz)x
z-Ja64a5jLsc9Z?yVuvBEWW1yAd&;szJCE=KzFTMBQi5p~O5%Ob$!5*b;1N6f9Ie-^
zmTOMV*7PAHN3@5Iz+|=nV`%&1d<<qtGI?6M)x!PY1n;96<JPTPeBu+I;Q08MM~@yc
z48!&W@4Aj|=;^J6e8T^*i<hmXUZ$uUtqo;aQkEswTBMXzMTHW2+ckg&5EMnR9XE^@
znNiB%oO`NU=Umbi8J!YJqR@EnX_}6DQqwdoRaF9!!FGK4%U|Hl+kb~oe*9B-Z&A@8
zfV!@D@x>1($Jlp-!~I~lw(STZ&~+_s+t4%(P1CYjH*{^kWi-Nh6Zz>2?vIPSANqE2
zOhr0tGhrzWMgn;2*{n9a^~Rg{=bz=$r30?ty20}72}k#iXf{1MLZJmm%N5o~COi9F
zo<K>;`i!z9m=YcH<iI=Hw$D*=j(&((F9|*ub{S+`6lKbZ9T8hH>D$IJtjGmluoh)Z
z(uDhDuay#{P7AG~Ebw-T>s!3F7{J8<A$R(N2QTqRqBbZ~qKXm~EKz{<AuYzNAUIEu
zNlzC7Nt>7N=Y4NcO@}cBbv?};!hjDkX>jxXRw>W0?x<D8#z(q-O|@GxTWE%^$8Id9
z3d|NW8avPpfqw8*Rmo&pF<(r%Wo{%BoDwJ%fB{ohC|yub5^_YU5<6JBrsMSJ5pGDb
zn`$!SLm&SM>e&?Pl29morRfWJc)a3d-E(kdpO-%Rad!81DJspnS@O-Vzrme%k9hKU
z&3e<a*qw85xn?>wtX3!7eD<TLOEcQNz#Av;P&`>us6Oc~#W?L7o!IgmAPgV^P1DeI
zEzS-pW7^1kEtGkYI3zf6@aLofGU7zVL<&G{Cpi(a2Jd~N_YZK2d|-}UDvq-GdR&c+
z`Vt`-f+L_gU2WJjJz-X%h@?EeK%oTl(%^PVbfJmS(QbO)ZCXx_PngUqE?vGvz1XLo
zOqmoFb{H_uVTS~6G)gDUno-#7AQ`z@7gs=OZzTRtirEUO6Q@R(OZqwG9A~Q~eV?Ga
zO6#rGKm-Y-II3nj&{>Ock!LT@IB6WGjin3~!bf`R=$)sk4Au@b?V4#nrJhVd!g9I7
zIfoDi8&lz-L_ib<v_Q##kr5LDWuZ9Oja)vgXgcVeV{j2`14N1I9ZD8dWq~OxCX<4?
zu965I3BeIXL7>F;9ZonRfuo}%kb-O1uBCASX>=g6j&U-bpsRu)1lD=lZeYD>=(?Wu
zdc&seSlfsl6slJ=R<OchyugOQb4QM=K@&nt@Cu^@(QWwf3)k7-n=uTVgtj?qxqs@p
z_TtB|c~szB+Rwqet=>xDY1d1<cUV7;zfye5lTCx0U1B9PO}k=svShv4FaTO;JP@>|
zSS*l2;d5qjB;rV>iHpq^@t5qFT?G3orI<}-{N``|Cja7J{0r9WH7~#XGGG1bSNY0U
z{*1?uA8~wq%%*MFv}yl_p-<1^2dL2+pA)1pX;Gw3U0Ie?RYh4=R8^V7s!NKpK$2bU
zWnNWl&GGRug(*P5Y&N6oyQEdje`5=pb}ko$a{ImUq!dU(fUfORw?20Yy>~o*{D3cf
z;g5OoLoZSmi7Kz8qOK?0ym^bFC~&s@f$f*?wheuc58>Pe+q86f{OVjBM+5o73{MLw
zFBWw_9?JeGwlPy7X`s^kGs?pVfukqK)RPHDYvwnua{202oQ<4uO0yoQih{lmtOm=y
zlO;Q4i6S5cX#uS&MCg$jP?$7n9_xIT+z3JxjK+whNlT8$5W*G)oF<>S&|}nuhiy7i
z^no}A3g>QjT1cf)X~I555b63q^5(|BA*4iRN4YcwQ3i|;7#D~Gcm>GwW(EO5O0G)T
z$JR3RiP4zM9ofy94Hrt10`Gm=7fL8pRM@_!+9?xQH+Y7&VZJkm)-w2(Z+!g~s!7Ss
zo6m6d+7)ixx=t~vaMsc`9Yql#Cf;B0am&Fhivo>6`hA!e#NcxZhd`AjvMh*7^7h?(
zEKfGnvt4doxxvAuOH8H(y<2knsN=zdBW}O(E+@wuy3SKo6^q4!=pE~2i?=<l^8_Do
zUBpO9sXa>cNLr*AP!fbnbd8X8mM*5zcxnKp<e88O0T<)Qobw5n<Sik`&MB?7*AU1>
z9$D&^`FJC+STcg+$5qSsw|&6EE^X=7m|Zj?J5vFVWblGPD9$zmt4)t}0-b}rr65_-
z=Q}9r>4!~<@z4`Co1WEb#nJI2o*X@5cmEn!u3l&V@Bm$w2xBl+l?%Jnh4L|`6E2_J
zk(gLmWOh)@t?N9?PFbX*14V9cNInF-b*#^pKpItf@3%tT#QF)yAaKFrLuB8~c=2Gu
z>+O=#r@Xwj3k)`3ObR4!+lKXKi7M1~^whQkE(oM9Q+H1!7nKtbL{3{6wGKdKG^Us`
zi-{cw5+x1Z`D7}Q5OD|*P)Z~nhmv?7X}Xgn5-kdX^>ls5-rgQ{JwfQi*b6>jhX82m
z=?<o<alz0m2UhEr)3Y_Frz<v1N8da8enMw!yb~A;RzM>xQxw9~_;!wnQ@kILuERP(
zsTC8gI5|Gy(b*$5%ZC2Ma@1>Hn9LXkiytp&LTsUq1sWxQ;ITf^+ekzrBy`TQ8Z6BB
zIqL%L$(rGKL)%(<vXD9oyi^qR6z!5|zaNrbGLfO4-u|+l*m3-*l%g;NFMRkT{L}yX
zH#od>h0E8j^Wm3X;_v-~f5^$n3E%$qxA~(#{49U+`Oop%8*lLV@#CaZ6Jp!-{XuR*
zq>+4GP0|?FlvL$}x~`c_CQK%iWD+V%>awP78mg+M>pFH8d(_nwDJ7fDhM{*T<<Lsw
zh2JJpq;oH8o<efidT)s_1#ib}$+Lr}Z93A;sX!tyrr=LL|2h6=fA1gg%(a_X+oOsJ
zS{ZKMyhUA?G@BoOVb>2m-Us@wqiqx3r0<3_rrIJ`#P@L9{4iWNKi<CEmd@AgERW|4
zHIeJR&3eP}(Ft{3arMd}yZd`|-GH}&ll3ElfFK232s}JF<GF)9jErarNR7}X()*k#
znhMF(anBzks)KPPFSd)B7y@yO!;7g+7RZas(d-vTow_IbF%vc~!i-79fpvr&skyyE
z<zg{JpNfeLMNVRU4YI(fh#4F~M^KR{Jr@^_F_6LrLMlLpVYpyW$wgxfj42I90in%W
z@(D(ibiGH4NKpW_VCXHA`4q1#ZWDO>o%={BxclA%Zr-{<SxvZh{VLW*y4I7xzc$II
zL!>AqLU>e69mUCHf|yo>;IX>L4*^+~6tN~qK`$jufXfGm+<NYLc6O(not>~cJ>}8E
z6W)0JJ`e6a!pA&~+L=%mB~4?o{lKKw1S{~pXV^GI3>azAbh)tWQ8I4XbtBSgOzp6J
zo4A%Pfu)r?ryQi8O{9)-;hVmxwq~<!vQO^dNg*$MeJjS!27>SZy!haw*qENQClUmy
zP|9DoBJF64#@ddf(>3ko1Bx=mXK1CNm>@-sDFw?FXk9R=_L$EM`?jTNkLiXr>$4>i
zMP1j-_IJ~@G@rd*WH#DZl(`EUx_BPEF*;GCdU{8WXogg@MSKisr7=nqiHWTillGwR
z`+UO^#841}C1R;f;QD07Ft1r|0>Q#iXlxKT?^4VefvPH*E~Xe!Qr9&x3QkWq2vMW}
z;T(EzbDLI5jUbIhMb_D;_cyObky)!Jb0L&iLsBMJqjQPq9YsK!fO7-g&{CKg2h&5!
zvSP4F58yovgQM*_2J3JLR_l(zHY}GdXUm4BbqPZ372YdEs0pzoIuBu>G@g2<IjE)_
z99-wd-V9B4B?OW7OY3>ktO<6>a#%6+f!GSl*_>ydxy8_@7Pfbei@_<W{h!J$9|c2@
zl)4}iX!?Oc?&8gavu2H7c7#=jwSi8i=&;^dR6XbJ{Rh-GFxY<U;Lg~oAK2t<Ol24g
zyV-os-~Ic4j~71jVPv5}q_h=X8ZKYE!olT3e(vXgjtBRClh1zkv;3d`um2am_O-9&
zaqJInyV$Ke@6x4?-DEOhI-OG2_4Y$+&F*5Cr8Zk*MF=UvZCvaWMZsiRBekSyHhAy1
z-#5l&Lq#rbM23Y+mAW0?)3!aPNRR7-M{7d}j{6Vp^R=&kjVqV0rAQ_35mIpN+Er$=
z8Ane}f1vy3J8f;#r03YQY&H#Dn~teb>^wG8y&ugX<8S_W1p8ODr7qaH12qE3bghgb
zusS>Au~8fz>~izw6|O&XoywGKPM54VEn3eAQS$KUh~vu!JpofmqRgfQStQn-wy6^r
z&+}8qVlEA>lCpNp_jDOb8%1J1N+q|ZiPY%L2Tv~S2noz9L_{i?6Ewi~9k%aNyC^S5
z$KuO*M=>%el}hN0EVVEpQ0Ri7a$%=@28y3AU}I#fp&o|BYfMba^w`=vl8GdswMTIZ
zc7v7^YCUIg9h+4{HK~~`CP*=(9UMXLAgYS4?|A#Id%XSDdwla7uV8KD=YQen*xB8o
zswb31&1RE|qu>LRNh&gB@Ts7V0#)hU)zTDYg_%#1E43>5$R|F5s7em@_nAy8j-TA;
zY<ZF<OY4r;UVVq?;Na3O2Zy`N7dt$<->^O#P_p1d&)#5ndcb;cG`1)94VnfmH_2H`
zm$()}Bov8C*YTn2=-QTcy<%uuV(>@-O6n9>>c{SE>bAv@i6R-KErmQcsa!Ctj5Q9K
z6+FhFA_P(Xu(D#`cR-AWh2%DcJ_M{(^ez=1kB^sZ+MbCjkR$_yE=!^SBO*wTwVq+{
z7+o=&?l4~%x_*KkG`gtpA@hr5hEe*ova!QJbUqhh=>*cFi!Y15rIcwUlj-<laVVn9
z9HL}uaLyrxU^<y>3q&PSW?i8*!!YErq)wf&Xc4YwrV6f1Yh2S29oV2bTc?zT;1f$|
zwOTS;%u&i9WP#F}X<errRT*g7COs)RLSgfcl@dY1f27ttnM|G9_Clnu$(Z1h!6^`i
z;3Y~KeDt(!Ls@CAUcE%XuwFN;*Bb`wX@`Nf?P;2ht{-SS%esfYPhH2sg0+%}0g60o
zkQfnArbpL-xpqwFuzPLBrK)B=FW8-&BE=el!1fVi6+;UrPa2}{>4ikek}H=larlvs
zaO3$Gc%x~y2e%L5{J@tok7!dcv<pxvL!hU%0kgZ{WN<W_fw1flT{4yg4Fi!N6~+{F
zZHwuXSwYJ5I6j{&xmx~9Z{uP?Ws}dXn>YFOU;o>fqD(6;(#=5S^F?V*DJ9Q7_Z)|Z
zhy3AZ|B#oz{W7b~`Y-=v-nR*$FosE8C!SPQJ=LkJsw!#gstV^E)>`KC9UwB9Oc@5t
zX0u_lSyNSY%7X10%A&y9A?xmjZDBVSd7~aS_*B>pgT;HFjTDYy0I336fDmaU`?=5k
z5x@GkejQy1taWIW@JIXm`@Hk^ogetVdH=R)HaP2Pn=~Fj4|f;DkvSN-P}Kdnp!=)c
z#5R;(Adt}~S7AuYHzSg`{rJf-Z@l(8^O<3HXU^rT`wZ=~y!!RmKm@c@RF&l6$r+`Z
zQkE4{AyFi00j5N13mJQo8VV%y?8OzL9Fqf*;N6Qr%rI}E6d`!j7Bo!erA5f&Pp#6J
zRZ4+TiN)47EzM@buJ=Tx^4K;N3&DGkBH>ZANvRc5p+$-n)26@_0pTLT1(a=(F_#ND
z&LC#<N#9%AwqZJ*<ptq+vWzi;x(X<Dg6*E5^d7Tnmy_-ct9qQZ^yP+~ohb*i8C|ff
z+m@4)Q-*nkwH?6^eBtw-=apB!#pTO~T)lpko40Oo{n|6^?Cdg~&j7H#Bf5cTZOZgI
z%SI08Qq!u)-Lt__>yr8YA=90jvQj8>fO8Q$Y^dtU_3Jm7O($qAu+H+@x8I~$l~iWN
z!Tuimi%Xn6StDFYsR~4}D6yeXEv4#^^hws_1tHs9Qx<l|>TJd8Y)PRKIDfofg%JZq
zO!|K%vzaDOPL$5~^3<J^<4`tclY|iIeH3W|BvmqU{_xBb#QaC)*3ctJwEko^2J68)
zg%KrZCmUAl4HNB{lp5)i{wu18Kw``UfyE<mp#vwGPHGm@34|2M5~E<)w5jkRwWVFY
zK@d|msWO^E8&r|Cb7IVKlaOcQ7voe!h(zbp<MfVT6I!9DQXv^bYHx?&Q94!WbbUIh
zUCNV7Hr0-~(Jab9<095gShs;nI%boK=sLR9n&#w;Ni{*02B9^@q@vK0VdxoRisKsh
zaBACUmYW%fBBeKoP@qLZkMw<$TuU)IcC<_Y#Br6?cLTao99}tOGOZZAWw|=$>@1Bc
zS8L09<7k_}rm^%xOhRA^LexM+$rRI*qHa=Fq1KX_ZmH)m*`0DQo3OKMD5sLCQv^9M
zgqBzYItC_A5ms5B=slXq&ZOen;U3RD`yAogGpPB5&7%jjL1MK;`I0FWl`wee7<9s)
z$TkT|qv;WYWL?ZK`*YTI17~Z7&4$+6)ai+mJ~(!E_KDFGyA23QDFuvfSmL?9Zd^E{
z^4kA<pFkIn`|a06x=5GGus_}5Qy=>TpZvtf(Iyo(AtwFEMVW*YGB0jAPL7X&aTGmT
zp+3mj`JJC<dTxS38|t#6E^8+BgvDZ!7)-`6sVB^4vuy@!QIt6Aasg(UOmv2PpF6Xx
z?Xlidmo*}$$R1TC?Q6<qPBMq-bef3v!69VKCZ|Z(bs!XoGUaD#gSS1ey!tKPyYntD
zyzpV18&E;9yO?u$aPahN`+!`h?|19E7CU&lzE2~tF@YnVck9wUFGF9S+S5PYe)L;f
zC<&QTXoMd-T0lr)?IAY;&Q8xbK0cz*P*)WPR}Pr%AJDHFoE?}Jl6BLv9x4{Oh?g<l
z8=_Q1rMJAYada4BmcGTLtrDZwFH4mKPaikIC$R81<^*!bE@4!*uPU1;rjr^wBv5v6
z4rH<AwuYD>!s8f^+)Wf2OQZ!9g@-aCJ>WcgP)MKuc1!?X1qAZFBOBLJv3DLLsI{Od
zpehE|_KfHT3<lv#!VvH=X~~3G6ST(EhTZ)g=DQ2Ru)+;3*7hl;&kY=%JYu;%;<Z=5
z$>HH84h|1EI6UC+(j|8Hcc=?PsSRaO<CVzt3YVsH>qMi7E})`7XYN}F0jUehs-`Fh
zOd%+%v{)P3K;H)1Re|dT)vRP^QBxKrYN|MU+_E`cg0)m8qJ>4!BZWnX)Ugjf`3TZ@
zJTX?g-XhDSQ4a!q$jLq1QQJ6m%N@C`xPNRiWxX!Bz{^dIq_rE<EKHJXeSm4s2Llw7
z9UE~JjHX+hC*Weh5^x?Q1{otmJEZKmsi!m*S{C>?(7Pdl6Qe}Q0$Gr>av~C?N@4JU
z%-3-YO}e>{t1>}~QBg&D!KN_iqDXN^A_4!#EVdxUHc(&$Jd!QtO3Kvb!yu-U>`qE;
zX<sBLWGe0i%BVydZv}o(2n8xgd>q)BO6&@(Q}j~fL!>BbN?9Oq3{6A7*-+^cuOvYW
zT0fvm!`|c&Yi$;3w`uH~N3WZ%NrsPTP$3}65f*p~nT!uY8C;0izNaWl76&_+s-&zc
zdN*+I=m}?MYferZj!#y!t)uHA!yxhA;JhNn0wHz2D1eL(87)RQv~-j@QWcWLwBle^
zGubtm#f-WvC`$#=p$35slE%V8Q74Q@+u}|)xS^rPVk*U@YrEXMw$F4&vYD3z5$O7c
zP|vVZV1i_-CdeqU%G2wR++uwy^n-NxzQWE9>82&^qZMM)q<xsjtkMUhDk$n?3>Y>`
zMDP?sV8a-0Ef`7cX_Ev@wp4s?javdB$ZWw#4<#gM36(MIP8a;lXMT#k-QBb*m+1tS
z@4NpC5kbO(`}go}K&u~)K1=6gUB;wwold4q>eQ(l$FWK&s;Z(W3i`f>m{21^OnCC-
z34PyTjHWCTV00KJ$PW6hPkFsQAeF)nj={Rz$(J;nHWgr*z^`>ephOQ+#SB*^Iml(r
zdb8rszWNoOefD{97C9se{H4RgToj)hRDLLhT|36(B?^8j;>K}Xc-n3A)C}Xt+h4_Y
zL5Y;QUl;gxK>l|mI7ho_I5|3DRwWtCY%yc!@*z+9`?z6)l!msqG((_s5Ru4`^i>F%
z@DpSBY7TM;7YcqM1VX~L4VHC&TTNUZ6$j#ZRBWpDh~8!M#JOfrO2MSA8SH@V+SK`r
z88sG@geZCtDt8HU_b;K*<TmeH6scPmtxb)JC{hP1CL=(Usc=GsEsz_s+<>HwmkM+t
zD9S+CEWvG1F;Iw-&JJm6hQQf?vjJ%$%1CzR6I3*KpYvhchRtS8+pIY~Uh??SL+bh+
zc6Jx+AMA7E)^%>&dWPMd1v|4jyYqdtX4`>HW|E{S7Y>Y81n247p24S)TtB4FW81Fi
z`weHumTsl6y+*_uNu)5EvepTUR+QNGl=f_e3Z6;@G!`KSq;w!tVdo)1rUB<8gN;}h
zi6#(a+Mj`(FfnR9x5ueqJUlP*Qn$x0GS?&tl_OP3FVUG#D5XTY_p8Fc`}TnfyAT6u
zy^*y7kRzD_E(Bck#30kEoTN27U7oTtnW2J3&30h>!gCyv4|Hvt1fSVS7ed6gLt;tU
z1kxRKJVjMf6a{5fp$n68`Euy8kTQ!L#WqkkJ!0HCr6lRRR4R?>B!cJ-(HOk9_zW71
zxpU`7%mxuVINGvdXcVp=7>0C!OlpnU^z^m|Wm9gSG3+cBh+t{D4JW5(T$;=QmEvZy
zb}1-DQB{;xg|l<|zQg5UX491C-6ZdlC0>HoCd)WPf=3nwWm%(iiSv?nfO~hBJbrS@
z<0r?QoVK*BrXLhB+p4z$O%>AtEG%UhFj}BQM9WBFB2^)nPiyA0nw{B%MJbS#g1SUU
zh{53l2p8yV&v10YhlF8T?c!IbNbB)^LkJyZJ!5vThn&wD%8FvKLky9@dd%FUm1~Ts
zjI8m_5{EPjvp(Q`0wpi)fT|`8KC)b`30=qFJU*X%Awrppl)@Oib;Rf}N@8OqWi7_@
z74wCY@B8(|M$WZ>2(}l|s;anr<qALZH-8pgq|+<y>+kR&B6SmP-@eT-4BL#r;D2z1
zf`BcEIGKgcyKcsqbS{+%B|n)=w&T;HD7ME)RaF@Q(xG*FO=VFa2uQ7X@ZbT)n8fDu
z7d1d}tH~{lrtjT${|W)l1%|;P+rWI5N4#yzSHALP{?R}D$83!xN^te+)vRNT+sFT5
zY+XCxe8AcyG?ycFe%nM1{P{oq<HGK*Y7+!haHyEH$B{I>OEpepq*})yaGm4Hy<=)s
zpoQVm^#xaNUSqRaaWZT;vmK?<G()6m10fW7TqzJSWO%U7dgd`EY-~29(V3R%XXAy8
z(5M4QpGJB>j4EsG(qn|s6aA12yVRISMq9yjS~2t;q1hmZfmlYI6x;i*aGt0%0g(%j
zq=`#R5lvZWRFJ3`AoQ3)QAC9Z37z5tDFs2MzuR|~R(HAcm@q~n$BHQ}scKDJo)GOL
zw0puqeTilTL`_i{p1FE}t^>y>Pk8XoT^4%_c6aIwV=oyRiD3sHJIc6bQO;?aGwz<Q
zdHapWeEExCXYX*r-u|4!gFSBDe3p-X>XRHE9<sAoFxjc8CsRaBlMrVtM`vq>Ge<~s
z7;eA*Chxud9{1jR$o=<D2>lL)JVeVKVvtN`Y5#UE#a?#H6`P||{Mm|qC8@o|$R4Fy
zFda^T4=QEa5<w|?FX@~l2$M`HA|!aRkVKi5figy<7l^13I#V{X+x0x%BYliTU8yum
z8)R7^b5S?G7vV2xI9%L5P+{k@(9*>;Z+0<bBZ9}pKnUXj&bAbF&azxD306ZSIqd2>
zb=`zy$lNG=8h`rW@k5~RTDqo9d6Y##J)KZg6{alEMsHv6Xpl(a!aO9)I|9aHZft{&
zE%T&Ti&`g7g78En!YAqeD8Nk`SI>y96<SkGCRAmCZ9AIHz|afopqSMP(K?KbsJwzI
ztBUz#hjLSLw%M>+E}8EYhzQE)<bD-wM|wg_>SC5|z}B-rFOX3Xu=qIOV;apSvr5vI
zpaq>9*z|#J)391O-n)Cu<Hsvb&pHMhaaKc|CLMqZfCUol*eIV|NWxPIM_nq)!cdlm
zs?yA+HM3bwRTWf5Qz?~-Qf&~<)Aj?)W`$Zs)}5lcdk^}+q}pNQ&rs1bEhP0svokL#
z>w=&)p{$tgA25Vec*GG$qXe>06m^+y{9Yo0X0QV5()&1sNVT)i(x%nr(D!uKvGqqw
zm44ScydCh?qLPp+F|Wp@^i$(Bh}%c_-rL2W$c-tKl$1p>Bz){AKE|!*pG)?4dG34~
zJ53iKsB2s9+_}SYxn!`G^H{DQ%q9>SagshJlL<vpY*SVeen&8y%_z%q%dS(IoZq(X
z_UA(gTjyNJ$eGbar?sYOTdJ}|X@$#+F=s78M4U?nu6F^gb7rOoJ2*fwpGutbh$y&w
z_b$uT8GCyNIPWM;#o^(l#I6kX2exm%(`KzpoUbre2l=8J&+$$d_wJ92y1&{@0Xhau
zB<rdLF^-6cv`NUR5s}qt%j061%xjYw^ZA_X&pm@G6z{$9HV3N0Ij{o+R}h7Q(1Ujn
zGVfCxq{`Qd0K|Y40V$GGT#oHLmqvrB+ZPGZp#nryj(yE|6K{!rAOZ1YZVE=Cj7%%T
zx*3pNhv-^-?QlxtM8p%&J^<!C^f}n1hmM9wk&+aYM;3+<H07WPE+V}`gphweZjX28
z66$A+H5=T3Gzy^vx+s|~b|}vrt$#?rdY$P($y`My2M1hv?mExDc!O#tdHb!md8K)k
zll!N*GffN<=LIrWgdlOw;M{~-98k|J&NmEhg9{sMbIf}ur@Z;f9saC(h2^T_<Dd8m
zZr*%`_cj~s;4!9RI-PU<nHwBlyM%R))pEtF-+B$-3xu!HVZqEy32}$$D?~78Be;HZ
zpT&MjIW^pSaK@pSQ;Rf%Igp^+HIs#BI@O2*q7o=yKrr|Ky^E}S$*R)`Dx}yXot;iK
zNGCvqfbs#|3xZ6|1ZgruRGagn4uSU|%0yDA%Lz&=qRu9qynRp**9R)>91$7QipWRs
zLf}KhM{pr-=g5%S;@W8HNliDba6`i|Sc<Yn7pYSaT^^m~U|o?%Y2G;`BHd^TLp7;U
zWw9N-hTL(I`KBr4xkcD#@nQPD9H*Lj@5HTkZqzvV;1fL1+Qg(82HY?tLxT4>XH#BZ
zcHqX4;KsqG&YqN-*`$DIcoH^LwWh91goMF5igK6zsv-!*a=E0er%a227?Pf9L`6WJ
z^2Cs4>?)1Uh@OBWs(_G+h$co&aEh)8Y=)L6rz;*mIpO$t#qsfmwu|^+AQ~WpXh{@p
zAst2ql=2wmDGMlyl!-SlDkfz?nLmoUq$<nQ5tS*1N(;ph0=Bmd%Z8KH5zBssX##Dl
z*gSkh7%b(iW+gqkgc~>a@oOjx#Y9VHMS=B#^5Bq-4;Wn~z-I()bSl7dx5q?{K+{Kw
z^%55avM|)M1?$ya+BR|agb(<vw^6#F&DfiM=+P=UkW5{n@Zc>%C`w}(e3Jv615Y!_
zzWcUiXOYIMl8Ba)swkMx=X~ljpJsP=FCUlZzxSOAytC6&?%%)1@$qr$NZB7;;UdT@
zZIy5%<EV8qnQWc0TI;P-7Saf3Hk)lj*C&$+T}FQ-?wnz>+0bkngordv!)!KV(=^m|
zjq`pBVb)sb!Y&nUWtkkaqr=p>m^-wQ<yphLo<WQp9X;XRy}Ru1A29SiWmU1Wv%8Hg
z`(bV|p1-%m3lU7;@xuD!qVBJH8wcTGRI~|+6nUwfET8$Fl-|MK4lGwIRH>O9&$v99
zFk8&A*74-Q6YR#}LP#C4tfAxL{p13Y;_1RZnc1|?=X^*2Vc|Vqj-$DV@TBb72qR!A
zb?(yrWXq+?*N4Iwq_70<7}^$gT)OYZhz~J=_D6l6QdsYBUZFGLfea+qT4RbNTk-~?
zOA8nua?BxF^zS3qDKu$nVh2lE)hMZuT2j_i%BrR7I{NMjRa{YPgTaBbOeO`hx#rU4
z9nMylSgtk<y-HomAhJdcKy2|&-bV;ES{SrhphJydieyROZCD;hf~%QLFEgDw*6W65
zv*GB;G57C1;H8g#l)9Rrg{IOIj+QIhvxsd9R8SB~gwQChsArO^S9dtPwnIHFX_|r4
zHy&Y|6*_jv&@!o@DpgA3Q5uv&NDW>Q9CX&wS?Ha>$h3czjA$WnQ4$<PA5#H6fK(o%
zB|_(sexA}ZHp;A{t|uunv?x;Nc{JeU#bl~uzW?jZ2kY384sstNHU<VC8A4!)fgsXh
z8oWmq3MrwgQ(+b?_!!u94JT(O)RPI?dxA?M<q!g<EI_NInKK4mr5zxZM3n_X%T!Q`
zctO{cE81NBh8VZ8O=8?BAtIH6%<54hrzu!V*JjId)6#B|#%+W~#^{raS4*_iXswY&
z!jO#7F6+}3ZIg-nQLsAm1Q*!b-(^}A^ld|D9h16ZXK#<rI&9bBs~VI_c4p}r^Wua|
zEn??=iYlQ)nN+L_V;DTNZD4uYb9UVF=;0|RXKPkXOWSo>v!T+(C`3#FF**v>Gf_RI
zNe7@&P?egxDyd3MX%t0asEhP9sdQ=$8=WA=B3(FQNHJcWA6TAlXpWYwyA>OH^d=xf
zNo-sCZo~Xw#^WlYqhfZj;IuoUnv@h$QTSwPneOj%)D4)T#72_VM5{z-7XlFlm6_l~
zK`RVhjMyQvbGU~uD!R_ncLU3&XDS5)k(iy(Ql!ynvph?wBS=nHYkDVuf`k3b{Nzu4
znm_o%|Hv>j#C#F$hwl448^m_eV5FidQ^57%;US;;^rsVqGH(~hh8H2mAqJj2dBTGS
z4`|yqNdSI;wL&DKRZ5m9iUOsRj!vm`p&tvoF;;9W+%BR^x~|(UK1SECF^1WE#(K46
zI-SyW9rMM4wrzRv@ImT^XQyn6D)T5M>AF63<=u9!N$}8j9wRlbNPy5`7<lyPQIaIY
zlxI1c&7Uglq?AuZnEk-^*;X&b*8Mu}%O7w5<J%(XzLd<VkrC=DWYW-%a*GiX?7U~w
zb*SZvlhZSHE*((R74yY{y@P$uHV@JSDi?R7Nl*%rv`_N=PR8+oA%pT|#*QU}1R{S<
z*_$b;B;VY}?tF+byKHmHfl8?cLV__GB?Qqs`nIFo-k&n$I6Gto$!YJrPXhjJ!IyKK
zl|pCfg*Rzp>wUm_k<-m`{;ZGLxh{x7<_>lQonCuU*3`9UJvdx{Ms%lCdKY}h(5#4l
zpeRfB4i{V-E_46E39I7-CMTvkK+2qnn)jO@L>haG5-oI!nPh^N3k0WZR*H5l39e-S
z@PMjpIN~wu)tWb6e~rtRFXLv8Fo116r;i)fr=FqJ;46qa<6|Vd`!)LqQ}%Z1lye&;
zcTY|b{hCsAh_IxZM2txWl!)0b7(_16+uqUjp56td76_qo$5tT1h*3-XSlWmT5m`i3
z0HVa_;NcJh&IkH_U=d@=jMX|F1LMIRE?9B=_t$uU52~@tYo9QlOfg<KsZzm{#*pcW
zlvRn34j*jdzqr71y<)LhQ<%N9C>D7v8B?)T6h*?7kYU0J4zpdAjjS-ObrNldbKCgn
zI!X(br1c9i;72j2b8Om{wrx2*J>~TDG?_AhSvg}dT`-+aC<>FH%0O^7bMibsI-Iw}
z;4w<F*jX^2OY9ICs>phyIaw~T!@%O|C6+|mexROAkWw?B%~`M4R7HuIOq2d^T%oB5
zA^|2x)=*FZF9c3$HcjCE!!z#RJK@3IW7elFzK_JH5lSOOiNGO6K&ybzj<SjvqnK5Z
znd&hryCbWLs?6AnszS*WAEiw};xvsKtOB7VeQy(I&j)(9X4S6QoV55=gAF}$BALt#
zGrdE<GsCWqsrN5a&8M_OOWfI|Ihr${?cmBO0~%FMk&_vl^%I2D6hh(#OKA#xZn1_C
z(Xzmql2ac!?LAVb{oWv0E!Q-wmca)~X&A4D&PSxwY}RY~wxv)CL7*&4y5KOP<~RNq
z|4aVvfAMR)^Uj;R{^rYxJZ^iw-!9DiNBt`P|JeJpXI-=Vz7PAX=^5T(PiLR8PoJL9
zK!YFwkP<*rawN)97@Jbba#b$mR9P7+Ro0z+ng75^<svsJyDC*qQc2`kw##xNiIO!?
z5+jKeXAl4ZG`i8#=`-!|9iM3px%jQ;-Mc|R>L$8D*~F?kXP>jD_j#XTt>60oz5^7-
z@ZyUv^1@3me!xufI#M^97^ke_di?k?=V#};mgdJf6ly24x7OmU+Z97~U89uR6?bDB
zcI4dwyQsYw<IdA={QBYH5q;NkesPX<j@5e2a<$rV(#F^End2nUW7oNcX;(aXFMne8
zlx*8A@4x?WckYH^;O6m7>bj<BK3S7uS!^UKQ4aIFO1kl&e>DD$7!i>=Y3XGkR_hWm
zX;361esVF5$X-V6Hf!9;DL3xiVr~p&Rq@g*FY(n!Cv-zk+jp|I#Rx_sGo5zV{KxWi
zlanAc!n1YG?D%~tjVtDrcb)j{r7Wt-^L$R17>0omVoqNW=3e2P_??D8yIE5sZN70X
z@j;5vq%_tR_#qISLPs$|R*Yu+7N-iTvLcn@?BGMdjNRZ}=taT@zf;z>ea|#RT+(P;
zP*j$^J=pXu>vlo2e2>|2P3zaJPXp&qo^kv6BaZJJQ|whVz2bL%^&PhBo)9B_-%wT6
zu2?ZRp+QECQid1=n9-`BG&5$0cUYcxeES<uac0J!_@8`%{rMqPRp1KEx4!iix^BVZ
z^pvy5OWywWDUaS=6Z#1$Oo2*bPc76l!z-V<!)(v8Szd7YtcSj!zdU6Y*VJx9YECE%
zp(lk9Nl}XBNXT>y!*bKHXgW6CK#}J)v22hVOpN~_c|{r$#wTo%2sWW@5;Zfj$!PnY
zwi`I~^0_#6tB`pn?wkk`vH$ShoR5gu<(r8N1tV?PG;-P$7;R8WtofX?^g~bZ1Eq7g
zvY;D!7Rx1dRb!Rjx$L{bT|=wXP6UCH8y}=tD~v#@TI=ynlpWM~03iNB!$8w_bZy6a
zy=J*w&~=@7()9zbEI2$o<i?E~OsXk*0ILj}&6@MGQ`XBRUDp!)kiFy}rU3|8XK~J=
zwAuwh$Ec}l#i6O#Zr6D4**iF3y;{-@9z|qQ*KAiSeBa}2figw{fk`N9AOy*}HJZT(
zJc?}}cy_knTW>z*;rr(dZ6NtXp%kT6=;W~mw2mkSTnVm7Oy?zaZLt<g6R6b$s}-&&
zsp}b4S!0YvqUt4MxA($vv(93aCIEiu8G;B{`hH;Btyr!X`0c<%TZ-w7uy5H{1qXJY
z&8#M_mZa$s2S>M9^@?+QihA%GH(&cSsvkHzz2N1SUuA0@L7Uz6QmSP(p9yUxh||AE
zan`gv-mF;;3a1n=-nz{yCYp6e-$jNfygQ5~MuiX3-dHWyL?49PU<?PhZqhtD#}yO)
zXFvZx<<8wZblVmp%w`j|+vfMjQGAG@5D&#kJ>fH-`3!pp`-BkDR$tRUNs)4$`{TXm
z;lqbqTwG-H-;b|mP>ghiE54D=!iejsE29`<MDfOto6Tm|ex1!`TwGi*<V;~}ElsoC
zc}A|+Yfy^){e3PMmuz$KCCWTxf!hg4)b0q=NMubg3}Red6_%tmPo6xI^b%XrbpuuQ
z;2VqI>*w*~9Ak@QT;#@9-v>|bkBYm0Q^ttcc?Eq+xZp92yy{Urr!sv=I!gfU;CXho
z;O=sXahBO^#=+qsN4IX_FE8n$#}i4~5!2{KaAhqb&k;Kntj-iStwqowq6qO`lTCs_
zDJcTR!Z7AQ&@qay?|Xa*s37WFqZLL$RTc!T=)0{z+*%6Bm>}e05Unt!qw_tz2NeP)
zMYIxzxK@f}3`J3qYDY>EvFK9|Dx#ER!}^fdPao)$M+XrU8C_sZ0T?Dz$6nj9ZCVzK
zH>sz4xbm2k;5)zjI_2Jky*pFP6mH(#=g$2@o;^8daj_DFE(sbKqhv^@s8C6|UH~yD
zP!5$W*n~-%NAH}lUM*3o=f}VJGJE@CSMu`9&oP;aKxAnhZ@>8vr|$q05)}#us%3sy
zbNk*N$G2zPzIDvv?1IztXT0~d@4)gB70*#>!K7S~Vo&tF6vozOOB*o@PD(6sn%>h#
z2-@IdBBX#O5xlg>!;XJxNGZ|{A_7y!fHh)rJa{(SUhV-D0dZ~cUJ^)1;(-{m9ZuKw
zANslY=!G3*skIU@1(}(Yc>yY`6h<yv2%f<Y^aGTWl7z};ryJ(eDQG3h16pB}nDA+9
zNZIUf6ho*Cg<ES;NXWC&*+L?TNFxlM=mVSen&o1_#rXwoBRhnn<74jLzJn_YS#Yey
zI*a$7Zqu?o+wl0&BTk;3&}_FTBGwp8CdO-{&_a46_`r6v$*f#ON@8AWwPP|Z3BhxE
zddkgPw=#khFdBT@QdrA&wZiEF;~YMCk*#ZmCy5#~MV8x!XG_mJk1lxlc)@ZDMYWe>
z*?Np=DO_M`6@~V=LNl2ZOeU7;)Zq$^_gj>XIOV9_5mi}Xivm{^`5_kLUolNIl8}HD
z>V5E;N6`|4>|Xl5W793^hnBi1IdFTV3fdY94297I3)|(2ST`K)-{tbm^ArE}U*c0g
z{UzoHbK+{jw|?oD7-mxjm3f-oAkOMaRH>5bD(wYjESr>A_(+E(X+wSEn1|~Q*-RTe
z(1!rl=6g{hs)nu=7E6!^@^Z1F8$8v-^6=fKeEn<R;Ntv(`E<h3%_H7_{|P>P$cJa7
z*Q89$4w3n6&gVY&Ih?g9`}_G|yS855j4UoMdH?<QSuB^kD7=qjjNH<cM0hl&M<_%I
z>zsr9@9XW&5k(BwcyhDZjJmD~K@f$-V(|fS#jq26z<RxAv)Pc4LU9-ds%dq_9m?-p
zWiP0)fFgk(BAt>Quri9LPoL5>4U<_xhykMw)9F-Bz4~OVPUIYn$Z8aV`&fvMA3nj?
z{?Yh@ji@qe7p21yC3XUZ%-w>n2?#O7DE@Oq($I5uawgA&v6Qyt`1UaugIIGcw;TLH
zAscT1MpfgLCz;M&Cw=YD74z@Ac4}7)gx$}}C$65m7y`j-6bX}Ul}hIrj3$oJVotJ1
zRg$hoNa72Nw&ETTrTwe3kKc9PyD=8$455tpSmI-Q)voptJ+@?{V4(@gPaefr+Bgaa
zlL}^&$a=G&T|U9i3W~{`<>i`3Z#|-oH#oSv$HCzqpMLE=RcUxT4V;~<2tkpOfv5@C
z-6WuPy}%r6780=+kw{Dfi}Q}({q3)!bju4b-RAIU!p)mExHvt*#5vAY#Aqdl9W8~F
z4pg;feo%0DJmu!C1HzD4oo!h>Suk8KnbA?HErniFx&~uBsGJ%S#qeJ&gN1Y7G+PEg
z5Xs;cNyIMMA2($#Kk2GNky0+;gHW%0yb5P1%7T={4Ph99wB)_;D&x?L?OKRz`%kE_
zOHqp&_SzfDw}jmC`CVuvKQ{W16-kM<X(?<$X$tzGW!r9;&1cwBkSp&4c!|dvVa=!#
z=(tOgfJpQ`+DCK%uQXkV_@QUJ*|Ny()vjwP%aU6+Z}R-hdsKBLA^;@|UhpD}*ftHT
z#gem=Gfv(=<?>=dSr*(lzJaxJQA%g9#)$GUb0?C5K6s3iEt)ols;nit41={Ij_Cb>
zwwk8hVih#o4b!S*wcXH!fYac8D|wngGk8{w=k#LDld~m{mYT&j0R=>jZ+dDQnUsNf
z4SQ3~;jF@_L{T~p_Gc7M;R?-myQJ@1rghD9vX3(dvRg}uA&P&St161jNdxGIhOTc3
zgJ<XkY4zSS^j$9QWa6o+n!{?Jsi|mUhfz^@2Bl+G*Qm{wu--DvXB^*ug-?C?XDN^N
z@Uf@7Gw0Pm`}2I~_kNFd2wYrVK(i$Uk9PSf&^n_$G6fE%V69-4dJfA8x~}M+oCu*8
zah}^Is1mpYs7PHET=<3|C7e<iA6Z;32+^=wU-EzaKm8{p0;}yBl*gw;h#wNJQ)5$S
zd@vMRI}Q(zc;(esMck8Qm%N+Bxk{&y;vn+$=~JFOd(3*fqz}C;AU^;YH2&=Mc$y*?
za`H8S_O+44$vW9cNt;xDzz0w2<N(G*il%Akx^9Q=c`rt$K1!(ecDrS@UXxN{Hk;88
zgV^=dbzZPiwpqzs&Gf<IC<ft|*wN-D22M_%(KK5olR4HnW|IjAd;7ch1)rpJ;d)fq
z)dl=fQTK1=7@x)wXh>sFY%Y#e-u>&8au#bA3i!a~#U-=Zl)9>zR8!`M`)n^SNlVXS
zyQU9&1Z3`qU1^uoaRi}D$Vb`yORlO?nRBkj5^pT_RQ}(U&~SV%LzMY722E57G7qh=
zLf}uq3j(MONr@?M%F$*TP)3{<MCcbVG3L1{jg_Qq@~qHoiZC$vLeiXa4H{C&oI7#3
zNC+Q&7^HKhjltMNRVii@MWI{z^(jgpP!>1Q$#VYq3<`~_N^aaZ;{N@D7zWyQOV<XP
zts;3yN$)UKImUouKr7)LrXWR~QVA8Ii=Kz?JYh1?q|o6X+@zY!c=*mECWRx$iloHX
zRq2Q}5#`|U)|8`LbM_B~W_A8##p2nL)#(b;HdLmg)D5L>C|rj&<N1&rU=1QXOb|j(
z-?xP5<(^aF>Kdo3=rD5qV64g}#A9)nP$5c3&WojzSW9UEPr6>{UwtQz4qZpce!5Nf
zNws2=n8Xf5x@-gzmH78aNu#lVCE-%SMJUi1>uCFi&322PXjJfAEYF$k?csgIDvK*>
zIU!yij)+8VMVoh6gQ%2LABasLbb*Ua&!+c0J2_#oSW;A$TX$~r{EM$}aJY}QhNuEb
zN4!>8Aoe0)=r&uHXBR9kE;)I2O4liF-MzyLFF)YmXrHFt(li^0(nghtv26PQB7t@;
zOTk585Mm^GAZTKU1m6*dMDl}>Mr~qXpbZ_dZz)iyu*LUV+IC>Mi99;*dH=NIvN5cg
z2n!(%Osr>W2KMX3(Y|AERxq7A3RmKNqN+;v4i4!1hHmH?qT=A_Hbv=BSbQ8H=$$%J
z+H|7MGS;wME?8Zj5u+!BNZU4y+>ntgVzgpb>@k_ns79*{Z>X@8)eJS;qp*(Q?2H>P
zewyEV``di}PyczU!vi*5OWUtG+?&!R&ATTTn6~BQ@k1)*iN2?(3*nhWMev?9^h}Il
z)hEtlpfQQt_l{|NPgtIdkDeb03bq<b6w{d4rk3Sp$2vun0UZ=mQ;07i17TS3orjC7
z2_?a5y}7Cw{&hpaXi}48n#X9%-8=WVdF!UMXSB-1>#?}IDiFqV`rgBLIlVYx+pqC?
zqDUhWxO<^%(BOw0BR|I&At@W9u-1TzI4A11q!ZoHQB5id0@sNlh$?(qPbtfC*Tx-r
zb?aQb^?jeI7h?-wEK{6wveQq{g-BVHY?>`7lkHI;#4G>65y8_6IWA-9!+;Kn<$A?(
zx#Z^Y9m3GFSzJ;V1%-8DB(4O}|G2YtuL8Kg+pkW2_cvZ4E?l9p->WJYzey>c$Hbf;
zjgNPXSdVb@a{)eHql^d{CC`B@l%jZDFV`$CFR99sz3CohRWaG0(S?q)^@2@*NL^Yg
zeTCK;Jz4^(T-eDo=A4M_Og0BLM(>U##gq%R@7k7(2yrgR2JhuwF$n@F6)3F(qv@k4
z`jfX}la&PtF-Xy&jG^~|u9vQtvIC{b2^g8TWn3Y)7D*A4$H!98bTlvs6D6b_Cn04^
zqY(^AZortpWTHr^U~vwo=a&rX1Y-|5I=aD=la=JE+L8y)-)BCtyzu-Trjsf2*%Q9@
zwRdQ?68~tlCI;b#8=ZB+2~<=>BojUbO^Oy3r5XO_H=lC$Y{|RN9r2}~{ydAbK;yOy
zy(b{GgVsIDf}L1i{oM21KHg(mS)M&Q<^6A;aQ@y2{rQRm)nmhg(ypmoOI7xy<cY};
zM>J7xB~%I!dQ==Bc_Ip*6v=3OKunaC!fFLh(yyFya%1Y0HyGrOh$qCvbW(BS=zzjG
z{NNe7o)kQx>+pTYkUM56CAPySWKf(hb&fQ-GJP;8aktOAlpV(*xA=@Hh-tvbNYl0$
zrRdgcX3I-vlPTUus@i1-=`|@9B$r5iVhjR^KGO69%WccE#U<yL3))pf3IqFxhkW+4
zuX5+!ZCqKRt+ez~k{n7R#J;EPMKa!Q8ZJ&wS)5<6S+A+92`|3%fO`+_Q_m(~V0t)X
z7<&4yMMY`LY9$YM$hIuX=$*_yAY@f^PPH%wYanP%G(fnrQ|X$P6fZH832i^HX%pwG
zz_ZJaM`u0jz98sY<nyMdPy<uvIhZQ$-rVEn(Tv%&l!ZGf>T1GdQq%P<mzU?P*9(l*
zlx2Z2j=Gvca%gQOCaRY_P-`7UQDU@V=z7+xC6W$cQDadTqePnIXcJe8@WN^(X1~dZ
zQffM<tZS0eG^+)B?UvQWITz~%M@msmCQN1}DtdykL~S{H{DkHCIl49sL$@;lHF-0j
zwW9RGAL)Do14lP+v$$L`bRE9y09-+%zA#Gew-6FR_STqeK%DufSHa-d?&1%O4_nY(
z|M3xAR5~k~H7~#N3T0Ud4I#yAyV3kIJ6K-$H*dfFHm7H&Y`0tCwtV09{-YNHJ2hN>
z4uu=%b{=yhQa2V?ZQEXnDV$~7ZKaj#9Nv4eIN5Fic3iiMg>VCBvl-4gnx@(D`t&v0
z#E@U_I>+z>$d`hy?TNut6b7vmj~+kf$<wF2{L-svt$FzHA?K%ORAtF#yMgcGD}G#K
zB&CnXr<CxjzSkAFE{gfUiT~~ksXpor_M`EUN4kdGA)-E?5?aDsWtLX5xer<Spn#zr
zSYNKGUCI9bKJNGi^P7iQ9XNmEgs111)Q59yRmh@kjbwis`KVM7v`&N838fPePlzJ4
zLyw+i?2hB=kP{LgDRE+w{Tc<e^U_1r)?hUZM3V1`!+_F-6bV_)8nQAZDMLgAgVJH~
zSwWkTI|HCF+L0U-ctWhQA-ZfteUKtJhDZn!6{gy)Yz`e%sidk5)7mkuN@Aypeqhxu
z*q*&ZHNDAUxzFZ&#k<RIQzgaSmv4*pnX!E4)Au<&ea6M*nzn7}n>Dr+kwFLw=So`N
zkdj6#gR#?W^c)Dir{A{h?ajG((y=%_<MP>8n9NF^dvL^jZ^mT)kgkWxtl;42fKPq;
zKJ&SzE)q5dp1uEs^S7ViFV~d)mYLmPyK|<Kp1p~}nQ?nyGJ+Y=<2)V{^c}TLI1_0D
z^idO?B`AqEQ$aJS9kw=DXK>0Ayv(5@;;^i@jc_OHlKsPdoU??k#pmRykWsso3u@Wk
zE5aw0JOSfE8M9g=7j{YtJEs(ewBs6xKu)2JrBoGk9v>pg6&R&hEH0SVRhIWUQXJ?8
zj~}vmZ`8DHVzctBwmlciHD`+@t9C<Di3fKNdG+N_bN~Kxlx2~fpM~gZtYMV<gr;S+
zT(VlN*seEhRx7rfE!G%r-@eWB4<0a??@?4GNd++mG#aBTs=`4?I|*{UhB(CW1f&E(
zp(22^X06r`4Z&#oP6QHI3n^l>W_@x-Uk_}e;n_Oy_%g6)ElsHKx*&!~omT8sFs&o=
z8V>d>`-h66&OVk}l)=;KjG}OOKQNh0sj7mM0)y{aEjKjV9&1XfYMKkXL6ACQcZ*X<
z5g&SD^cbsg&O&y{wpP#^XAAM{8Hv4Rg(^#2QHl^G$gZ?dhHi1r3-|95hE`(Av?c_P
zQ-;axkbBR+#OuHGd-%2`8pmd_M8~*mm&iGdvTM`XyHIOp(<x_XXQGJq5fv3#mxPaj
zs02HXY1r{EL;l()Go+28l;XvgUX*rSgb&tqGMVdqKq+Z1F3x!L_<b(UFByiw7=rQf
zfjMK*A$*&W<CbE|LJcWiM(&)}n)!T=bB+|#4t*O2X~!EQ9JkqQ#*-&cC`#e(b=j_H
z{N7RZYm8yBSjfgFrSDw?<q=cD8YzxLh&U&P)yvfi9|Bq#p1=P9v?yeaHViTFQIbMF
z>==_RuEzL`t6eJ{uJNP3>)-hk@{`v%l|O3XeuCo){1U@_5(AkM#1-zXQCJjF3x_R4
zv5@FC4eN5vYPqD|lTC`7PI1!-m+KYl+EH7BEA$mwr^gM9hp2XIo*AF<7>FTrwW5;N
zYD&TZ%p6!BgK)QE%(gz^T6AW@A-UZsll?Lw4n4_Lj5+FL<WZw?q!>H0e+ZJC6iH|^
zB2Su>Bdc60@I^u4W6okt3_b{#Pa&2h5n}ZC&@=d+y`rWpTsC4h+&D5c+Xj27xZJK-
zpS?+W{1UaPG12nyn{Ppi9Nj+V;Kl*vtl%ep;#Jn04eNEoJ8wT?b=lHyJ)sjdsmoNR
zloWpOSW`&bHX5w5_%17pwdV5Fv$$v>O_(1HRCPt7<JPSMo`2za9z1uO)NR?GEjc^C
z<nilokrpkL?<v!U*ga)hdkX7O45Sn>R*+Vm(NG{GhXLh16Q?PyW-}yOALyc^z@ia1
zkHS?Hg~k>Fyrv}7w2%$nwboR1K~a=gBPmuf+rPySiNheTm&E8=kvaax@L|VCE9`Ph
zNjL#%_vSTfSJX*ica@74iP($;7-i{(gind0aP-4~_dN`OuG92gPqS_CU8G+pR&C(P
zrsLUmVA=KzZAYCH_YbGMboZFM4{mUHbDy$uS=Hl(a4&sPQNgYo+RG&uXXmUp8-`v~
zC^rs|I66Aw_~uO}(}@rubwn9M%<#Ih78Nxig<Sw|h!IBgA*38WoXOTBmBFa&V<rk=
zrJz@mhntk5p6zjYy5;fXGq!rh({*AwSb{3V1Tsd9_8dCTt$oLAZmFxtbP8oLpcx<)
z7*$Z0HO5++u4B_Q#01mXK2@EK4pXGrwj!2@fxc-l#xR*osjFIg`$@vzwbl4`AjSe`
z9oC8xs&Eycu(rr%zmkiJks=;r8529Sl0s|w^eMN$@G{ly8z?P{n^96o8VeVzHH(W2
zN)mn3@bvxndHL1XFhF4}B$4B%Bw2=Nu&!ipZ;yxXzQf#=w2h#t6Ahb??RT_D##02R
z1*u{eW%V&4C%HQsQlw9wdOG3m-MhdCIc`_0PP*c|5P9eAw>f$CjEjo}ei->@SKrR{
zpZx=j5wTOb6*LxK{m@gU8enHa9Aji}Z%<mRDX}-(%iNu~D+0%)iuHO;2!YLJ!!Qg?
zr&G4uEsMp1&1SPh<*xhQS!**#C+>tCSI4P#9M-`{jF$GQP4MK&GuG=3tL2)L(=)#O
zr7x3!uIo6vIA^h3!tX1@ez38-4qOA;ejfn$`|qRg^||ZC+>Z*oPjXycsF<5PM&&=m
ztozRy;~<%@TCA}YFmwaWW-G}+-4S|HqsxMNf6n^+lx^G5RVA*ha#1I3w(G4}HKus!
zjPIt=zb<61R<<=!N;aS4MuE6{K82VsUaBfN+z61E0&Q|(=LhM*s&usrOwxWH&2L9v
zJd<-(Q;xS)dYlFIZr--Gz!e#t^8sTGDvTnEkwX{=VZaZAc=c(6$s6eD#Bp>e=}10x
zoNpdd&ZboLF{PbxdV0ahq()<@ii)zf+_`mxkNdP;ONyS8V#Vdznoa0XNnx}SVM68{
zCR5~=wZiBUKlFg)wfAiz5?EerKpFOCQ~WS6os=BRXUwY!7w1bBkIq><e$4vu6ZX)Q
zsijD3Y`UZ_BhGph(l$v+=MzuvTm@1TexnMJskKxlp;e$6BHKP;lY$^|ZKZM<8%pRn
z8p2%_@3xQt0B0>J2%kL;p5%l0){Qj=2rp6_jI$K(!>?Z-sjwS=Z6sxek-{h~ZP=u6
znw&E#--U|!n9wAgHkiUmB17=f)>9JU<NKb$2f87$SZ&#?Jf|m-N0%F(_C1Rr=-<7g
zidUb%&1Y{Na(7x2DosChtk#jLEO#(@-?yw*OU}>FX%`#9T5L|Na~vNZvv;u1{{8{=
zq!w~`+k#bSB|b$e=jO#22tG>IXg=W~+p{R8Na5=L<3lC{aB1CWg%pP*7{qTbMX=K;
z?(m4S^Cw(%8_qj}vQwOnSQ;ip&q1X*C@jbO1@%<n%0M--D3x8rT}f$cjJC9c<myJC
zs;5kHYb<5wL0i;pHfzpKF9<Pm^Tsig*%YN@KO1AhI73ks^ld}m_e>_U+^Vgy#$|M{
z$P?GNn-K3E*;GmhwI_+1XS-$STJGGrO(5~)$un->IKn7ROp2$cXRMbCOp$ncdBVGI
zzr|-xPqA9(U@w*3(*sc{e1J(=QI-{}<(g`y37toKsExygL0T{-`zi)WNzsXHxGOow
zAIrE_BSsQF1y<V)zx>bt8MZ2U?!oiy@9#@SYf<EI@+jvz1Y!!j^Ugb*o}6%bxnvm9
zj=cQ6U+klgQ7kbYXYc{<Jx$Xxna)07T@qv5i6fL!tk-LrZL=%RLI|waYm`!SUH1Vc
z?O6C3V|IxZP1Ed(xDX`3ZHE|s7kVYfs6`#_Bf1nKM~IPM{g?j|r3}CK^>6X5*I(yL
zU-~kC^{@U_wAQ@-#vA<S|JVPV^?LI$qjq9)t8(jXmrctR|L(eo;=2pF>&WB>Qf2-X
z9K0Wmk8g;kQo4>t%SYjHYDV*+LWu$JpaLc-;^1jFEteMy_L?o)X)34KyLp4{vnQ<E
zmZqziO>+W+79O27BBGF(M>2Ir{8K?T)hR@hjahgYi3FLHuJbdI=csL4x{C|yNllp4
zIHhn}p_8Pui=9T0!cL27JA=tWJJDJ;w^`^mrYc3Hc6H!L?pJ8Ewm4f5Y$O&^NVzcP
zJ?Ab1VnOEpK-V=4-9WI_uISMU_V)|nGP^+Y-ZO@!qcV~Cv_hqVrw^adY+6=}B{v`3
zq?}jm9nHCY>zL_upUcw?-~Q%1y#2;Q2#)n?Be|&K1dfrEjF7OVB!mHNBGxFZRUF<p
zrkWHKMsw@t9h&u;e!XS!?2H3F<?XM&$L7gn{N)+d&~R&C6ShwwoO5*GDP6$nfKi|@
z3_hZ)_!nve(Tf^6_8uJ~d#+?+HC7FDeb4#Qla$7znP4e`#)U}KkraD^_w+sx6S%@q
zO-eM1F!WN?`A8f>R{sd?OBo|`jkXkJNm+@x|A!qPv9MFQa?bFgQW>F9*aR*mIXNS$
z7Xu-Q$}L5O#$dI=*b-+9RzW`mLe#AKKr_I}I`ZzbHSb)&svXckq2Tts;tMa{<8v?F
z;nv(z=)iy!I-_|iK-V-}T%L1!ddhaaCiRhdJ>}l>_u1RqqnuQdIhhiD4ARS1P`VOb
zI6>@oC`Td%d642Tpb{fEPV|A0&5IQ(N46B$LJB+KIS8j>2$3NreA4&?!C0nuZ!=qa
z9>04gXyLG8S~zY@69<*&Xkw^LMO{~tOQ;P>muPJPht-a{EQv}nL<uhUAyL%@C_|eS
zG)iSFn3$N&=W@}DBAY0B%r1IHQ4|$MZy3C^28*J~$eqI)F^-mtpZOnJ%#`~eGfa@q
z04hWnJSWed@LceqH3x?WIO~XAk1Gmhvl&CzaIsqQ?EQB*IeW%pxxhL{(~s>1aYy$t
z(nYX^rD;3*uBUT7DoO!oouN>XE(Q9$<p?B+VvGT0Ek2E#k&kIyYyTwR^!$wf^gsL0
z_;3Hazsbw5yuznG^(kI>;RWvBe~#y#yT|d(8%$?2rjwesZF%#JZ}Z-J@6tByjs*R2
zjIp5G`Mrg>D>6pTjrU?@Gsb?^b-m-;0c@KshX;pv?^!Gs48wqPj?2qSR;$(SXZyaV
z$q2AgYNwVRO~hkvF<mc$uA_O^KNq7Ep((e>Lg)FtZ+w%c8F>9$-==Llo<4iVKmXN#
z!Oi2F92_07H{TO`m5=eVes^ry7}hNdVg6c5IY{72K(X@*RQY?<mH(y`h;ig$|ERe8
zSjWgf0y4L5xb_tT!n4klnKVQ&!ryYrVv^Wa41VBpv0#63DZZC=MLnsR?C;S<&#KvQ
z=%*;uuCN<hvN2szk94-_iaDkd7a++TD_yRfj*;(^l*xfYi-`1_E$j7)by?7F_qcWQ
z2BmRCWe8e?Ln>WulE$J{X)(9=G0G+>XfeN!o)C@P!!g@*=A?=-Zp;;h%wq|L?8Bz?
zfcBmth_eMs(EhWt3sSJu#RO*!(L}WKOlF>B13to|#}8>QdQ9lKd;bfB&a!UTy#JjQ
zmy0F$UwMVg)1IKtxOwL`58rvp@^V2vas1Rz{sjH7!4_il{Pfv5-+BFA%5utdI>-Bg
zsyd>o9g~@3|FEJg4aY}E?C%{AlcQO#IeT`&w;p|y^Z6y)C+Aedz#PYo$}*h%9@BEg
zq#l@5SqG47nGnF11wlsw9ziW4DS2s+4w2e8DywnXgY|6DldQ!Xv1zkSB&r_gdQ9k~
zy{#0sD46ff&`K=c{NP3S62}dQ(2Xir;3`S&GR{#{6-D9bKEkx|5jl3^nKOz9$TyYM
z8i&Fv5i5ikNGb<}<0QI7d92cCgcs(t0U})=Xa~i~vgOHT%j3(2_ZK~9$>Fr&o-;hC
zEuXr%$7@IX+@2PwGSV7_Nsb~F7^8_XiTBXO1<TbE6%scNZ?Ly_AO&5{!c4%Bl|C^=
zT;XtKiFIZ6&jVu4kj=^ze5BoO@m(){aVIE8$cvOK3d+KY{mHJ4qlii#*1-$6P-1FC
zMPseu?){sbueOBK4Js-2E5rU&Gb<BCop6&Gshr@)y<T!mi<+{mG1f7t$fyid%23x6
zs<OrpBB#DWNjzm$aCkU_1WnHVlwwZp{Ef<Diq<QF-{O5@Qco~OgcGt*%R?_*-K*j&
z<<R1Q!lwux9EPARl#(PdnN-k6v~?sOm`*F2c0+sqjOA)U-wmwRD{+ZOqT~`<eD0JW
zptJdm<!VLniFVV-E;J@gl31@U1lsJHo)|?u<c0W?3x$fvJ-^zlF()Sk?|Jn8`_#1*
z=Cj$1gM&Th^EvnLKgVP`;VZxM+nhXmhWC;0lEVJr#}gv~GCd*Yma4~l!;p(QV@#IW
ztKAO2$>iO#D0c-MNb7U6*#PWNz01WVN-3(U0<Kaj#$qR?h&7ge<P2Ryx!?-XlH(OV
z4JtACNC>SUA#u&yZ@<m9ZRy&crfXTQRy=<EnCI?4zq`Lb&LMT$2l;bjakn#O$p5~M
z)#+S3DAMj{bke)UTxXSZ%n?DK*v<Gya^y_{8gZ!0o9{GcnoA)Slg#~Oe>W>|jF`j5
zC<3y}6|=e~8H*|{)!_kQwWRTZcIcRx?60Xc%BUUHX1o`aO2Q2eNkZ>ULg#d$lp+zG
z*#m;UDHYKm9CahX@9TBTs_6;+fU%kbTVS-9^%9~KY$@cfP?A54ic*yozkx_kNQRIc
zU{J}3Z)w_f8uR*XNhT3pq%0gESX2u1c_Sh=E>g^f5b%S?n}9J<l%OVJ4AiCO_^_tm
zc5F6Fwin-Fh)~>kiG5cy;OJHjkG}OD$yiKTGPM()Jvw7?xkBrNaf<o0BH=+LKJ&Q;
zES6^+9Nyse?Yk7E!&!wQqI6<1br_u}^n|{NEY8nKTaRuA^tNT_pE3<S6Wvn9CH?X}
z_KSwuBv4NhN(Y>^VvCayP$uTbGo$vo=avH+0HrY|VvRV0G(NHdNa7M;`-E44V(>Tu
zrL)p{EgS~n2}kcUXq3dQ&>Fc1tYcDD*s8=jhbdhaFKUASgd)X8F2ER+t+Ya;vv5LV
zRlwxMM5)B!dz31KTM!aPi^H{zvJ3DrvEKHaEPLKPYk2Fl<<ct#N(v%}Mss)Sc<Hd_
z#p9a2!h`QnsRD|8_(C8Uyr<c=blt$Tno=L!U^bgkOe!cOKUK0gB@Yj4aHbHn&laEy
zdBD9F?%&Wcv<<##U=aCn?7f%=S|hp{Cs9mgEyZ3)fs_<li}<7II+{jWn?h3ey9dR@
zmgBkOQ}=EV*A2rEC{?7cC+r>WF_}4rY9C)r^7}MIWob1e6={7>X#=IP^qnlI^<<K5
zHZ)Z|+3o0x!tA_xgySX)JV_R0Nf0nNTVhO!_ge~A;ar(_k{<8Ht|B7vu(eKjDkO=t
zNr^#88^mdYQkJToK~WG$=nz;vdWzMG^5}qr{XN=l%X+h7GO1BUvEHoNc3ZrzB&N$6
zbP$W3q+q_c$K~Y$tsO%rB>I3xM~M*vnYeWt`7~pu<HyLY`#xg`k+4}42FY|R3WqBl
zbzM`IF1O`kUgSkQvAA4hRCg>BM*Qy&KE?uS<k*bG(=ZH_Wkn2QbWzxeC%Uc!uH~;r
z*lxB=CX<~#&Zva#x^DNm-phTx%m&M2K@)<seRnv@_Yo0XaqTiHmKWlfL}9qMzsILu
zeU-0#<!j;-si5}*+wGP&-+WWV8b8q4`d6OG4+vcKu6VfWqNVZQ>8e#rF6_oaPB=SR
zrJsL6<bP<Ooyz>s?;7cY-}0~g`X@NXE0gl|N?e)d$0UZuA;~k=CD};nlqj^>6@wmW
zhc>GXey=5!6{)b;`3&E-w0@u+dZx-|A3O2?RO;&KmB%+1Ub(O<^5#=epG3%;@tDNG
zU#Xbe#)1GD=h%is+lzOcs|$2Jqc$blhHQWv^Hv{p%F$sdieFmNIo2u=QjbUDcgJRO
z!a*+X5VLSJQrOu*;T$12u|f%Qo^%$9*qmYO!+;M1E=tLxwQ#EoW6`zaXdhJQxm=vF
zJ_}5Wieh$5S<JCPvpBuLD~G9SE@$U#&ers8kE;xA+f$hUI^dk)_PrwxZ_YV9+T+%9
zhwRN~C?Z4OF$}$g2E<6W^|Z^DXAhszUv4pNM>VvRp<|K;%Cy0TOYkRDd!DieS13~E
zMvfdlrN*S59DYFFv}IjEREnUqxL=rn)`6{ubq*)d8mvzQEr!@w#l#g9uD}{Y^a0-u
zBFf1AgT@$~wG_@$6(#j#f^`;cE!x>#ETsAfE&v~`utVhsMUlO(u?nnG*qtI)WduU_
zHAd?kPA`DDMq*M@2y|`W{9?=Fi!JY;dlrMHH#I3Fo=buIdnGU2nsE2HVt!aKsDaIP
z%e<ISL@P?8EPjna+Bx&toVu>5s}gIhaIUm0c0Nd9YOJW6thG`VYezpw0n>I{n#~&D
zH{vDJZi#)50h}%%1Y(Mq!ZE2Uin@YK_CzbRGT9zt%Qhn*{UG7)A%?6jH8>?^%Xju_
z`dbG)IbSl_pJVopSX;$=XA8b_{tm0&Gx&&eC6i*pUR`iJFFBf7jt?dr*OtQ7lupR)
zT8Vd0Orm@;+5(;wn~Z7>D5a_EIy+r^K`%qVSV!RsG@AKrE?I#?m*Y~v>KxE)9ZDyL
zp`&RV+7M7h39bU;(8jXp25wAdprpK7EjO$#FKL&T+)yuY`{uEv69muU(LNYO7rbl=
ztn8qTF*%4??xCWnST2{M*!5mGc&eAy!L?On<kKfc?poq{xAXU{Hy^_SBq)`5!?A&)
zx0u$jxLoYi;bU7okyx*q@8bQB|M`QXcO&sXU4w*%p{HpSs;(J^o}|Qou_y}Kw%x^u
zX{~d9tlqU(cf6mjqw6}h+bz~w`o5>{yItEi#z;{V#29JXhL|FSD};Nevxi-tltw$0
zD|h>=_rIf|DfzzR8^^nh<^BCVQjAPzGZxDQ&rY6kdVc-~I!_<w9O?|gjd6n7$nh$z
zbL&eeCRb5uW>*+1g(&M%LdE>P(p9t@5-)o7e*Z9vyz7P7zvA`B7U%c;mhZP;`aYtQ
zPimyxjulFYv>Ppg;3=gDNs};yq}XPp$|g`LP%1+t;#FWsiDtEBSTuy9CRUDE)%dEW
zZ6oWU;mA%U{z@5e3ZtWJ=9AErG*J*cpK?1CQN1GKE|3k4R!T%`$h~H@rZ5IqRCKPw
z50O>V@#y4|XlnL@VWuazcG#^!k=#8C11d_$yh-xTV~h;3!=!|bWVW^2DQC43Hw&Hb
zTV(}(jKKo~F%1k}W>sl1i!Yxu1>O(zLyNHrlQdRYl(N~TN^^W@Deac3*zn}!gv<AS
zn|gkS{TuhWfBVx+lVR0GwyO>A{O()CG+=Xub^7EIU3TmrO}Tmdkg6;>IzFVTOHh*8
znnI-S2ik4RcC{r8k#^zPUUZzl_k>}6N=eI9_Z&G7+oy!~loVGS?klRPLOUtathVyZ
zrm=G(*r$$Cw9CjAG|>~Zp^J*B2bAtm^mwINgh&%TUVBW05ERN<rnRN4D@=me2l~F-
zDQ$HY;Z$WwU6%q9l@&!L-9?qV05L14L+~H^nfOSBT?D%17?MG&NI_z%JrkX<QRh|)
zq*$VKHs*_JmwBg@!4#Hg18oE6>%_B_=JBTBWC*lM<3r1V*1T9dp4(U4KZx9z2TB*w
zszRxPu5U4_Ln#ZYASs2iFx{U?5OX1)8<1-yMPCrPdzNaC#Bqs5>hR4#yKQK<8`|xf
zKBpn1;DuCLIEvB{wOpurHpP^VWDLq0jJ5dS@%_Mdz2)-q0`CLPSxTFEYzPk`YUp%<
z9XvA^xqYMNGB%t!&2K$^#?xiPWbZaFe)jYH?B_qv-RB>$e{=(_q1&vvc=C|<zw<5L
zdjB1Y#}~YOaKI}!Ys}EIUn{ik@kXLmhUBoyqNC1k){-??6a{@6Xxnvevj*x?$f!kv
zX;GnEfvO6WmdT+C39l`|XfBs$oSr_R?OLkIjM^BKPBcwNm{!y;y};q8K8GGkiexah
z;yl4N1s)vCD7~XF1>3D>J2xadW77!AqhrDvhf3l{S2}|;iqaX*vl6zh3i`GerI?m%
zVxnXa3b8py45dUDJOdxdiul;a$fpV^vF%&7Jy>INdrpeAt3YRVq2yOj*2i=lBiTQ(
zL(ZZ##1v@TEtOXELrVyC_O8o|6q=$)QlR;WHqZ}J)Ge1w+P2NnfSnMsV}KASouRNA
z@5!5(KEKzzg;iJFw2^x^qKa4T*$<oyRDv5v$0%)h`t&LH?>|RX6#UGWzr@S0yh2Ek
z#d5*_^RNGPzWVjAGZuv@{ou944vkCVN9l~qJf>{=oC{&+48!0t)@By~`N0sxR!19+
zk0XtL<hq7zjvIIJLjR^~Dx^2_U7r}o_bcpvaNzkPj`aKf7t<K~JU(|CWh}G#oQNle
zp03$YD2LL5!lmeG*DdSI1=GS%&kA%|qw9(odsfYY!IkoAT7gbjs|iCwr*Z!A#4r$s
z0pAY{#-L0hIh9X~h%G`=^n<5Ry(mGAr7Uae`2jH+RJo#GTylQd68tHb%YnUh<hff9
zC=bMVJ!plALT*+`#D7lL*yJ%OWZx4D$qGk1X_67wRYoqZN+NvF8jGVSMHmqR483^l
zk&+a%XtcEuVembvkr+iXXrp8Y%VbhfSi^KuGMR39bhcvCtT=oB9TsPAa&YrCZqDwY
z%Oe)8=6oGk_nvOm@&50AhiYbN4<@vaH?)sChSfk5E;tD%Y}P9gF(@H)Z&w>MP-=ry
zFijoRG+>4X`URV3kEzv~{VB|+1+yxl&4AV8=1qpuIRzsr5tuM0X(jr8$eY<g6tg=k
z?2ucuT8xonWYKP@N$lOInayhs4mF|O;s;L}gqCHk=mLteU{X)0CsSNucU<L2${R|-
z`Gu+?$MIpuM=tF0zFTet6|gGhw>2UoVo@x;eyM=1wUT+3r-=|1>t^6&*>l!J&fCB?
zDw0Y#5_^SVRw^nR7`hc!JB(J;<rM8EH2p@hVhOkeRb64MBc>!1ZH`LXk+vb5vs%gV
zytdE`gOFUiW<#^t(r>qfu0zFywFaAe<K?8nx*17ntaB7qiLz#=1`Q$7G%efJnr6Kr
zhDccyyOf9PB<n7Ef>vmoFeb8pFvo0qzWL5G?DG%!5B|gdh(G=F{~otrdWmvA&ral;
z=p_Xsg^t)YET2B&^vyT;wSVy6^BZq`lh54Tqtl8yff^!2Pmy=1-O#bwu6MCb!!Srz
zY;MKPW>d;aZU^TC7LF>Rokd%X3ke%!fx29r^YGDoJbw0!Vpef5-(%2@pbBok@(MRT
z{Th$gTY@c6#)?%*S@6<ppFx=bDv_GRq?%IKGm2_TGA<*V2|I385M-n(ORQGZRYlu&
zbnCTb3ugp0sw|CX3|~$ZlEOpwC(S$7Pkyzz{tDmEJMV{QeIYGDQNuRq=|qmpd%C_8
z&PR^w8-{_+rlD;+G{PO}`yTHF(JhJs0Il2H0#R(%Ym$=Rx7luYsR|=fs+6D=*ZFn-
zItvh;dAs-TKhN^=lJ$Dcy}Nh$=`VeW&;R(3OZ!zTZr{Gm|KczI2mB}hhyOi`)#_s@
z>g0^&1vo2;wUK82_`a@=ef)0q`-;^u)y3+odv5$(Fyd7eVnhu?=0J{<_s8%CKA|B5
z`izQQFT(TUm)SX^Mj;mte;7x4)r$2|#1y2MxolF1Y31-LGAI~GnLjC%j^G2^wq?C-
znD;$JDygahVoB3Px)3o+A}*~FWo%+3W#u@j5Fm>jV)o`tQE<-?#f+GQL^f(KG?=2K
zs%oZFp_u5>5u9clPHEeoZdkDH6!pssj_-I}B^l+rxSAxZvsN0F#x@q<Rraw9kVfi)
ztXWYhl9uO9igZ!RT3b>SfwCMjKioqu?5wpo<FZGdsAVxivq~yTdsjP*)l|xIKtWI~
z%T>>6)6(<<s}oPVeo9f_qL|#_-u@whIemb2(+i)y@438O(L6epbs-n%UE4@^Nokzc
zl*Tft9JTg1)id}r`sFFguW_kmzZjSmiuqJAt(5rH>2!@wq4PNzcO!B?OZ!;_Cf9vH
zwQ$H7jYpMMq>y77l)?)WkLjf3@L<N#fnj?w5S1oba6nZ_?yhqMMd7ecoD(8RM42I8
zJD%LiM_R!jc6_wLPV~|rXu%p}GDa&A1|(_IrU*Vn2A?QPOVSbREJ{1tK61GpI9+#K
z_L@zCq%2M;<{FM%L0L%TgfbRaO-M=<Cn$}z1%2oUVW97Yw>F=P#fK=^uI3)uVY_Px
zfr2OqmG>dHXN8-UtfHLOxY<P5dzm<4oWVK=rot8_t`rr~NKB78QDePc(=`oBYxWNg
zcEs|l*Up8chf+t1QX?Kmn&)3A_}S;5=TH6RzryE#?$2ZPZ=;<+vN4l&lLb;h+ZwGD
zN6$TA|JF@j{HZVS-dDcDZ~xuD!?Tn3xw}814jo6vVEvL91IxZ8#6Ve9D6QG%g?K)n
zOUvAd8F571&Lfd9^!V6g%L1L1#H;0!r;i`=_B&5#b>#5ol%*+{%@6t1kAIo>&sO{w
z|Cj$84UzxdPyI=fg{==bCzGJ;ilE!JXk)QOg|-e?6fBmjU4)ggC16oOGn+|4T~sA)
z-_iRHlVFleK0jBj)(-#*N;nVdlkz71*Ni`G?=p$5$!g^osH&2_Z|S`lae6O1K<@+l
z`xa{h!_Z?j!UxO(2d(u^No$PR6?=7Ev)Zh$Ci&n$z)O=MMay*?kncV(*9*Ux63$9{
z=<NKAs;W3RIN)=i`*F_C&w2RpA)kKjGjv_Yg9i`z?2r8zzxpqKjgPHp&j{;IaX1!j
zA!g%cou3u8%dj@aptKPMb}sCUwL}%O^Z!`brHt4|boQzuh|iUrmyhQye4In={^WU8
zt4z1cf|5*gxo+pO<GSstydTC$SFe@potVV%NhMjnW9CUIqhwQVZBAN{;@xDlrtb!t
zrlsjRs!&r{1BGMIip~crEBtb!71oHjD`RnTBr%YJhv;|bCMDs$C7)aRS>zB9qg`Vi
zwkU9gWSS>y7z`|$hQ;}Yrs>%93s#F22ZP7vT2l5>dDEf7u6+mcUSdiF6$H^kk#x+8
zZs9>Ith5}nGDuRBY%V%VT-)uEDaa_6D;#PN>!z;jBqeGmCJ@n^JOhflG8`s@(;8<y
zi{+MW`<Q;%;F~42Z<!r<l$}s%hcl6_5}UC&B>Z|sO7dJ81!eG%q6jvPp)!fWdt7YL
zX+vmEQ0*x;b<{4hUpr={qAsAc5i7>KB$YRy`Hjm)E^i6cRr@(*I!4OY1+rNk4Y@K(
znT`Bq&lBZ+PYUL<n!TBWZcA{|>Q!3rJnW3MyL6zCsVTt=9ZBb1(hgHpY8OfNVa7+~
z*hyh$&^W-VB)(=!oSMb)8b&T&3eXRd9$>7+SVz|d7R!ONMaSAJ)=ASOO(0Om#J+LN
zQbeVWvMRZ8e3Po$6OIas=m(4|un`nJF+}>drE`wy++|2LpA;=^rWBDYm7jePa|GE5
zM7faFv=%rtMoi51IZ@gsHs@GIXEAk&HYR79W&*mG7OV!QQ_0z^>l$mV-1up<e^Ho}
zBzMp#jEf{wl0xKYuVQdDT|dzD9doNm84W~<C?Od!whVs26{6hgwPH}3o1gt6zi|Ik
zeC410Q@-)5|AgDt;bPBJMSQo#l@*8kH!#|;*{mtca+d)bV`Q^g)3!a@I&g;G_c#a2
zN0d=$>sW2KJUd_U<~z?gU3t>f&@4NC=F6XB@6HSSxBuWD@vFb_6%>k>UV0H*6?g)V
z&d#`b^O#wo(IJw2<o!3^<o@#y`1N1=RR$u9#fsVffuNyfNlcPtp|$4t<}qbeQ&c5`
z&#Kp?L?D&|QR!Wb(CD$~Lqv~s&&>V)H*nqjM2EB|qi`*a&6^pIacbvX*ECzovSPVh
z&@@}>dLqhZ#FSW%ZTr6efIy<}2imq_v)PEk8o9@fG2*iYj9j1VMbh_DvR>Jt$j{8@
zGqh3MxpRm4Y{nPA_(iNWoSmKV)?06J|Ahy*j39pD^PlH8f9tohrOyw3&gA{+-M`0T
zPFYKek+Q6Itz~T@g>zUZO8*!ogH&mSal$Q>l$W$S7h@5-I}g_i-uQ9BjvvcVJGJLn
z)Jf=l=42_cI@;OQ1W6<E+4Og0zWp%X&K0FErvz}-j!h~XmIK6y?>d|kd#s|a@vDs_
z-6(~{ppvKW2ewViX0xU0Ys$LDI)|z%)_pIgy((f%lEThNB`HpMkr;Y{Z;8HxR6`Q*
zz=uHW1+g_s<Tx>bG6t<JN?Xuc_}flFr5RK}#S{G5hSjEJvs@E~KnildbbUvlCYcB(
zkaXA?BP*Zi5u`GifWhy?0*V4ul?4})jDQ+J8jUrU(p7Xt&*1%4jA%xxoD;d);_{MM
z*HmRGsT*38k{<y~tbut|P&>okq++l3oLzc0+jE-j1!s$gTt2;laZ_9|Wxjul`v<qM
zwnQQ8kPj*wECWdjJZHsfrfs$?PM@*bJ|zt2l;yzD;!U1Q003LBNkl<ZtYCjyFt17~
z8z@agE00c5IusglUx;L&EkBc)A8WP67>!nn!G~S3HHu(j5R+)>F4%lv`3<K?T~$mb
z1=DHCY~nCspBOW!bZw-Ra@=Be63Je<qmQoAR~ySQqxi#)kBHbsRMJ=?7Q`7lHz;Ix
zl#!F92q;uC3?bo@C_jc6*fxR7MaOa-=wd||rFc;!Dh)SE$6=8$Cb73aXFfk7P*B(c
zZJh90lENs5u?mmec%vOdSy#JOtJ34-BmW+q^hXzbV3+xqv=}Sf1Ulv$Jqcma7z4&&
zs|vKH^AfYfSk&d;S6D|?S7J(=k3r0F^Nu{fo;G8VnZV{_1Q<N;y!i$<UwcG->mJ6H
zc~X*dD8Q%&V@OKnsZ%baFDye{^67u;=b7EQ#jpIWzk&BXNq5XE%lu$J7uGFhRblf4
zJ@maisMfJswG>r_tsI?isBMYqL?zkz#FNE_*B@T;?I#_L(zJu&-~LlyqDz*)@n8Kn
zy!YrCs_R6`-L=H^3ZoK-2U9Ai8M+O|Iv&6CE`RrL{Vo3NpZhsp`OK&I#b5k8Y?_wU
zrV(?@on3^qHV%(&P}dbjS?1&qh(?R~wdD9<j6*dNB&3bWgyod^AAfYmRq8}wyKP7)
zCeCI=PEfn;c1=}H<emnqvOocC+cGL)L3lWA+p^tmY1;-r$oGq2KvbYw3(w{X#Tip*
zuILEgZ;W4`&L*^N!_WTAPjlza9q!$`LrRH9j~;PydLsAjFkqeI-gEbGu8^XWAH0y*
zO`!RBl~yQigfAHrlj%%^CEgQ4koZ7rDO@2fE3OK{Ar81g7I3YNaB#iPnUA8V(^}d)
zS||7j<mgX&sEl0NEEW;bm(5Xf_5*?DV>WhC6s(p@nx>)ayC2e#j*Zvxz%!DS#LsRZ
zVM!=4y@b$XlED~9Rn6FRVm+Zk&iVwpVPMmCY_<*6rja(8g@frHm&;QW38N*8wJ_k7
z*nml5KqQ5p<XZ@xBqxL*wktt=Q!~;a=Jv_c$B6HGng3QxoK9d8$s`UAXY}nrh>FIq
z>Dn!Qx1kuO!kz6qLh1w^Rvlev==&{+fixrvr5TjLS&fMXO(YnFRRty&rwJ?VWo0rT
zqHs#W1VZ$9-^qr0hhUS<>kCLp1XelG$r_X6z~tGtw!rBEN201=Z=%_sH#BX}wi#G$
z2R7S>^aBj*#FNv&Gd%+e6b@H7k{)@I(rz{y!qAg~r!E|ms^GAK`T!31ORB<Bn26Io
zR!f?VpneHujG(kCulrdDCDv?$s=2~YI0p#Nc<`Pu3i}XXg_Oh#(MS>HQ$iD|3>?f$
zoC^4MOS{fumXvn(Xd#S9e)_;hl5wud9SgCTF}r%k_Meb=!UrKlJ8LO4IK+@x3Xgnq
zBOjyE=w#?U#3cT3Z8y+%iPbjHdqc#L0%CBiJhK93HnCr8_77|3vl$nc8=gLWkIAIw
zg%|G0n>Gc$^Ax4UT8$Zm3$WcvaJe!%TX<*`Jw{uZSaT5;a<y%&ynP{&O17vqPW*(D
zxa^0d1ST&FJfdohA>;~P3|M_i7$^L_qN;Y8Zez5Rpn}<@93TXYQbg+*q9>(*jREK3
z;_)Nief=BUeDOtQMGZ!FMpl88crlZ2v$<pva(6P&X@z%*8?XL2f9w|;e)+%ozoTXa
z&rK>+@KT6pX>}N~kB(CG*_LJKJyj*bkae@+;9w7{4M|xZpRf4p+mHC>drMBcl8p}h
z<WGNrn=gKbfA~v(pQn!>VSCSB^7KRE{Lv$R^Oye#w?FkN2M34fu%+wPy#DpK_{V?e
zAFy6s@^^mm@8P58^>4q)<#J6~S5hR%$%EXcxPAMU6y_-sl^_G%Aj-3eF_^MoRu-H!
z=dxSREm&jn9P0n*7I!hi5QVo<mR4f-l%ngTXq!yxT$st^Hw-;}*HSpSug3h>qA0M|
zvR<#*Y&JXet*R=9VPMFP^)a_^B_ip{V&{WaNgo_z`*uWsLh!uv&byqQobng`{GVq&
zpR>GN@c8j#VvY)&PN$$?Z*Pz3bVAd}G5H@5wi_qZ-R44x7{^&F?K&-yQdkR0Q8<^0
z(B*E!sFWC%tDttaFS}NBjIvN&6OxQY)F+XB`v)20v1l!PV!N|rD0Z7xV+?g&b8v9L
z;o%_%hld;-9`Lo_{VMOg{Wg!EK9S7J>(~4D_vt^T@x6biC{g9>I>{UyqlYj+&=_a3
zN2baaCb{)S!iPZ9w)9O;UpnCsyOLfTnjv8IK+d?4_H#zFLX0`7V2~n?F@lqf=t_Z-
za=Sl-fOm$*2hJ`QOfD~|3xzE;l#<poo0fF@6SjRI_$?}Vl*%zxu_qA`Px#1k+oRfn
zgeT}oWgI<VO~SYcRiHEqrxQ*q%0gPJHd?G5od|?lrLDmhHj5EQVV2sVwMHANiNq6n
zKhQNT8kLJ8xz-_oipA9xNr9@gSPNCDs3(fHQ*>RR>pa`7XBajF-w@)ssSy4>B#e=C
z4_zmWhH2%PO*B<yC<;aC6or*yPUrUcRohhnr+`qp<av|hlntG52#d1F=J;MZ0byX&
z6^sVzDf<B$G$M@*IGxy=6wE4%N}i!>XpKc_lS3uOmV8G47!mO+msN4-6XBh;7M%qZ
z!1r_<`hMf185FCma3xV|gH-{Y5<=<MM3tEa*$gs9MK=T@1tG?a%0${e5raW1OA(-q
z16FA!Hc=W+=`^!x$uLCDFB={|S>Vbw^MeTwp1%WlA+2U^n9cc<T1n-tg4~o^>&%`A
z3|S$#<2)!0VGzp{5<|a1rATiJh>50Yi74iW2h?RrP%!j<SIk{kwrMLndh6sueD^%#
zoFl}Lcj};(K?OxrIiOjAOMz*OaB<4JU;iEMzWgGu?%lwYH7NO>QB|pxfz030DdG~K
z4bfVBbo8m<=4U>~ANz&B#6SE0{@0YVJ<Kprr9|Zl2Jab$p29jv2|o-J#^R=Pl8(5#
zWau{ZVL<7MAsF6xc*5_z{e;IWO`|Gi2M7GbpZYWWgJ1b&PF5SVEvO=-KA=KkSg!cn
z|Hc1{$Injr)K7e#!yEgAq2s^&>;G?V-n`4d`{(~6PtVWzt>5__eD-ZBClk81lL9Wc
z){4S$@7{A1MTt_HgM&TRmx^YwmdBtl93J10+`VV#QfSAB%};?b29pvh{NXQ{e^^7h
zE91^E?W{aP@YmwMbZ*zmuC#D2t7c_VxxBpGWyy{y6UG?2exw>)q0?i_cHFhY_vY4(
zBc+ikA+(3Ot{8@%gZ(-8@84syU2}GJ%5t^j>tFvmuYK;v`26QT&ty8;6&SK{`@zYS
zD(~R0Cu2x)m&4!lyfU^>i=v<`ONz1-{^1}zLt`zNt0an%lXv}`TrKp8m^pl77WMAJ
z{g8=wvKh^-pB)-0Hb|4YW;&ZuRuy$!bNlvfo_p>&Zr#4c@$oSS2ZtyWecyArSTKZe
z4ZZo{j<FpZV8?xfNKAUy0=LegoFn>>Y4bXNzqIQ5uBY#MhQ4HQ7N7art#2t}*d@Ft
zGQ}s0U<9S};3c_07V~t))k+bN;}Fr8q4$C1dP~1p;%iM2gzz-66Ux#un^x>~o^>2h
zcobgv>}kkzo}%qxwehGn<XqawU?V*WV-m(f>LaBAXP`7t*NP$poO9yfP!<?v#K2f-
zjI-FnUA5Vj-l@QqQaD#Y4P67BZ)saWICb_2j3M(J^QJ}U8sg<@oxzq+)rz6_^qpc-
zLEmWx58g`;&{`wvU7avi2})*!V>hWRlS<>Ph*==w48j{E<u*H{I2Lh=fKZxLF79F$
zxf!EH%wRJj?>)iyyG&t}B8G?$xs97tF6=x?B`OR1wWD$x6$ggi)2bXLnl}w8Z#1r+
ziP10f>i=kkDm6MA6OV|V`3Sk#AK9STXq}y2B2KKXcO=bG^_nodNUf&n@xfuWNPXKO
z&<zTeN?bIQQ6@NNHMI#?>v3g4Svl70EvKhjx;|0QEX~l;4qHm&P?+om86*~|s_2IS
zGs2#O$0L>(YD5!45R}U~KnR+B-xG#`*f;phhY1mka~$2i!`}WO7%To`d7>I`7^Sj(
ziZvnw&q^b`yGC*&iF8^XZk0)w(&`=E!y}b~1Fg9~oAB<V_ju>4zs2#L=Q+Or0x1QS
z>jjsW7hGOmQr9zX-@eCWwkJ1Cif9YgiS0}u4AD(^@yEZ!qxYWj=2w4%o40DZK^EpP
z3{1*eke03|gh*ACL=T&8i76c+S+-qE=_)oM@y^MThfB|SP;@%-bARp^c=PNDO<i#O
z%FE0Ew{3X%d*7tpF4=PmwOaBI|HgmGFaB5mHMi~@^B4Z&U*NR|&+*TG{kQmYzwisZ
z^6F=p%=dWj{r6aJH*B|}M(g_ytqM#=S*Nop^LY&lZr#2`vuQ}ckP^u#4v%jzbe&A!
zBj-nDy8|t@73u>{c7OE781d(vV>+8q6eTeTkIq_Axs_$bY(AwbDpE|DJIAh|tEy@z
zoS00eXcF6#MtmcUv}v2B*|lx1zxIb&++zgibUv1nqZ>DvOeVbb)>{<T@v}ekv)q5~
zKHvDpH+cT}=V`l^)AKWm!hM`)3fR?cjNWKs^j+o+h7iX=QPmUbO6V6w@j;r0H3h*9
zJ2jh&k^||z-?ffaj&&N75t8@v^M6Ri-N;eaN=dt|EGb<<T~!<$9P;3W7r1@rHus*p
z$4~sk=Xvh_eP**6WmRIdp&vY#mkZv1^q6hiW;#T^{-4nMxZ>D-@Drou$#oo5DcS5*
zRmI-H5nbD}9R@L;R|qIUWQdWj8|b$!T~$z6OJ^Oftm%DA3<0OFBwGm)OA6m`6i*1E
zCq9ny{3&F;OgbmR<l^3kz_#t!oG%!rQ1%AvOHhf@OsMOUgSnznpi&^kA=92bDJq7b
z*={wLmmb|^nuSWVT2W@ySy_nAP?|_#1EmYpt<U$hqbwb>S)wcpQ4eb+;>Ds6&rKf$
zlu)9O9W#~dx)uvDZMeKRM=NQ|J6jM5_!RQSGIyK^pdwmBWetVXgu)W4$ZVQO(Gem!
z<tPbt=^A2ZG#Dj5e9l;`5!8#!4OZIh?uL}xm`SOO;A%n=WXvRm)q>(0BOSb=bfA(5
zGY4@15Za2|z7p9MiBX}V!bGr9TClUia5Qz4%9Dmb9}}BFL^_!9e5n!4%+G|(6(gEA
zejOzS{l>X`zGWZrVaG=*?5uOBNRAJa6C0xxELX9?a!?74kI?p>VGyU6z8}O0I9ZG`
z*k~wXlw!+TN)vFUqO48k%>|n6fYy$BGLwFO*MJKYsw4#{oJfa9bqR`$8uSc<M}+}P
zkhl<~uvxC0Vx%7in(dZu+YmF4NNK~~!67&A-evFjnCL8xm-a14+c+vj9i|k@Y6<K$
znV*o<i2vu~k$5Gg)Xup(r366?Dt+*zFrdA#3igcRW@$Nn``i55FaKi>@7||r)?8km
zbFsK&+ca2Hap&%RKJ(gVdH$7`v4zC1854z5U^F3Guyelfr9Z<j{?6}mdb!}BE_QIU
z){3U<7&<}jy!Q|l%4lp+Fbqb>yvnfc22K_Ymji@Kb9Cnhw_m=^qHlTa-WNIA+h?u}
z{_K+X-g^d@Yo<=qY&Ni!$d-$@HvB*S&;Kc*uykd~>#u*4FYF()e|X3jzVJnU<yU_t
zbK?^IFi<*CGKVC2ruUz}pUtH$MP-@p)%44T5F<rd(r?x~vh?We-UrDq#^hGmAHlj#
z(1MI6B)`jZ6;4!1Rn{0I)4<4+n$2evwqS2>@2XG(PEJlpDM>_MQLwnYAf+Tm+`;2_
z9&NH?9>owJG$Q`~#|gsG>AH^9YEAX&g4KG(Fa4u`L<*5vJ!dkVQr9*6`}_R;fA|lm
zCllJXBY0Un!bjNEey>ru5@WAO%GS7DUNUH$6@+e#R@=p;MX0J;{1>&NC<?lv+wl`e
zA=PMtnv!DB9;1;}Dk+Q!J08v-rlNW*ekKz^<?i0S$AbqC_`(-|lF$Cwk8$J14QBJ1
zIOJzK3sE>kpZ(mA@$lh89zJ@++2uL0zv^zp@X5*3zZccw3bD(NehR`PONkUj*wV#F
z$S7@0;Bx_xLSkq-{HkRrN;+GxDQoI-O1oKNLLezG0xh)Ak8)mj%nC~3Pv&-SE)H}S
zD5wlT$YL%WVOQ2XTQv+_#BCLJvt_$!*_#}om52!pLbR6@CB#rT8B8j%X+~xDiH)ZB
zJwp;7#TX@_DQQix39BN`3|QOYiWX<vU12w$_e>`hRaH=y4(kl)3R_svo`j<BW&Oly
zv=K$8wU+tb9+Sz0wrRvhYiOy<njtPpeiURGoE4GdXq{uV76V@krIqJn6k?)48R;~n
zB&i!(WpD&3cCK2K>8ky%uH6?}i<mR1{5uWW2vQbP$W#hPQCO_e^j%BeHMDI*+lhbY
z=y4Y!Qp0#b4s1+J2+T^$!NhP}CQNAY!$1QB?XG-}mDnfgjNXskfYI1ICc!rAs>+n2
zj4I~&S?NzGd7{31hxXZ1TpB2}6fuKRG}_QAgAE>=(vG;=CruL+Ln@$e@q?!8JU%*5
zhNvQ?iBvi<af*sS=`_Wpq>FI2=xG{FStpFyf{KJV$0v&`oSY;hS+rUk>ar$APwWK-
zjZH&8v}m0utmL%WB&~wJY4QDl?|M>@tkuchocZw)`?qdV?#<|x!pA^JB16`B2Y~`(
z6v`&FNhB3A^sCUx5t9L{83=eHDOdF_Lg^LJ2ez@L?bfWiEnDBy4}s1nTBR9W#d`6C
zue|Xlq28yS?NQWIFa=;Rx@U3z4cgV3?PkR*pZOGaGS5m|MVTx^3`7?>xc3~-{n+RE
z-CzFO{Pb%tQpJGPk<bok6!mnU!j+IBE_rOSG``2Eo>CVSDl&8}o4#j_BB;bGcW!Xw
z)-4_!9B_1e6K4!%Qk=f~9-VPCAz_lHG8P*Kf$@DH8Q8QfTifu~TW|6CKlyX$pgB4?
z<hfh7>4%Qsd$ds!@{>S22MX@ox{YkyB9m#&Y*O)P^FCS|R;xAJuFG|FF6s~^@emV+
zz%@&;KQd!%N9FwvXc055*IPnNOs7)tm8G=Sv_WP`g;+0?B|`{w!@zR6l#NJ?_!Lpz
zbGcr#ZM#fcxJvo>U@q*3JLx9VDfagE#P>2rX0ti#oDKWv(Gx!P$}7D3sZaAy|H=Q4
z_a45Fv4)$sZZUL&&=Z8f`U9P_QRpGZWW@>IiFk{=IUG53!sES~oQtBw7*W)fWl3tT
zY+kOXZS11gyhpe@xviB287U?L*HF73*2tV-tu<9$^V(~#@ui>sGB3XPBF{Z{pL@?g
z&(Yx#RnD;-H`5A8Mxiy#5BItM!UJA;^;3NH>tB^kb{ZJ-c7ad!eW`07({*V@Mq*=R
z=zE&Y7L=jQv;l35h~ReZ@JR3pI#22bLg#6R#9$m+9u$TcC>a;&EFMgHCu-Y?4fyW&
zEquI`(K>~o!SlTXPPJq39kGr01Tl7`euL4TKtZUdBo8U%hdOtGtX9kl%blYGIvr^`
zSPzNV=d9YK#aA|Hk^mh+;=VHYBz9dsKyMUn7jf-C(<SO^pehUMx}dHK5uW7dTN8=K
z?+`2@`j5!7rf_1{BC&;@m^>-<j2ty-&jvA7Rw}oBg@Z2*Kxy6XLKbss&)6!L4i~$^
zPQFiC>q#MTZXl>={=CY?^Ofl^5aoI)6>(K5_9ObLfE@h5Fm&{NCu_epNRbRa$w#8X
zB$yk+eqq=z6cq!?cRNJC*G7(uEZ)#6d+JImkLXoK1N5vm&#l+--Zgnnk^=X`jgM5=
z-Tv&WJQ>!ci!+ph5H&t&`lRT0UO+-{_b7&>Xa^XAfvEAl!Vj91WCx)&6c!4Tn3_m!
z5|dI4bhk~Q?F`_srbW{dd}O<A@WG*rf_YU#Q1nBOQd$V2T2oj{>jQ;0XjkBq7X_a-
zZ2K+S)tYv-5hzz%%A%w`JYxUWF|(UTsA)-0qK{V&`@|$MKv^}EtVY|+$4Mi(QBMBo
zE2MMFrmz-8B=|_zdp2Fmlhu-UpPle@vErf`SoH&KKVYq*E-iaUhddn;adybjo#)wi
zQ*Pe6!_m=AW|Jwa%L_KkCGWoR2Bp(H|H4byqRK|RV62e`*;-!x$)Dm^fAPQL?%9e@
zAC=hHQQCs4nxc!TTnhy{Y)QQ$MvXD0-2Aa;vtH74J#BtKZrr}bD=)u_b_M$fhr}UZ
zVk92l;^5Y8o_y;KhN$t$;G>`uQ7L*Kh(_YVzV(gQSvOlMSMcsT@9@eCFL3AH9TXC9
zjTYsmG6tpK#?cMBt@s3nFn}`n9FewIu4uY8cdfGG46#6oj1uyXa;N`~Ym6Y4N>G#~
z))iRmP|Dy#U}6nLnThd6;e+Sw;`{@=I%5oN-_vw0tJR9dV!>v!5f8iXDp;<}h~Yz5
z)42*Q%L(h%0;M$j`v-U*L?}U`>w9`HF7R#F@$dYZe~11315Tfvb9Q=;7*ze>sL<%&
zH#)>SD`_=pSJ-96nzdMKuJP|IHmlx^H9IHy@qAb-v43N+=UtJ1r@XE;;%y}@x*tBj
zZXANvP5AU@KEqG{%+GND`RBQD;|7!2TngG;q{TRH9R&SAY09eN`1URCKX|}wHlu6X
zUAyszHpWfdcNKS1I0XDKNX4KnK1H0!l4TRFRc_S@YDXVPT_E&<-Y2}VIBkhaggOP1
zWsS-$*ECWO#*Ui$uH(rCN3M)w6d{ExO7;&AnM~*S21&2z6ES*J-(j>IvbPgrkPcpo
zD!)#|>3}T~w+?F>RkCRYTo-6Ik<^J2Qzb=ER?b5-#N>z>g;GAz4@vB{nm|>0s_np}
zmNhA#BCJX{XT;vdMxqk|W6FKPQ6(;P1w%@nA@w^y$K7<05YAby&eZOU;^CIm_y=~X
z=bgxC2Y|_+ofpdP6zA&@^Z8G^4#Fr_+8GWTm3=5jE^y4YP%_W>u}z;r{2@rd!bp|W
zF)_7<nKMi+xYUb(VN5$dr`l0+lhg=^HI7!}AGK>AkL}vgkU6BBMXb1*C_cpaNQK?4
zmma|8lJnPlw(B*P?5~;Qx^(V#jj(e}iM|hPwmnmwXxp9?G?tN249X-y=}e?9G?UUo
zh_p>l@Srf+J1OFQ&*kMgi8VtwWO_FzP7IN5=;(WadRNN@tJQ+pbSmMyQM24ESa&^r
z+c9(<l`EJ`#B*-8zt8;mm}-9yGp*@W+y&dG7==%#6%<KYu-bwuNy>>!kU}ycNkn?Q
z>CpikhPf%|heQ(;=WWk-o;>E;@15|*`xiXBY}gK(J{kHT?H8*&$93dh@+@_rECch|
z9zXlDe}bR>w|<W4`~a;KtFsf{f9EaUeeZ4FefM3;dcxg%&tptM(23DAM{CXVFTKpI
z2QTxrH@?NaFTTjMEHGBQoMH-u=(D%qAO&!WlE<1Noll%C*0jTrubSr8t-H+k4tVtR
zgr*a_gF;LD{guzW#y5WLH(7Xzu!+eC|0E>(ywE#mc<;^c5Sy0eYR%Wb{xyE)%YU5t
z-i*VeLm{gxgHLjE>!jG5&$;L&hcv{<{=vSq-a=%tSg;QL?xDUu#!3C7D2o2aG)Cp+
zSf~{4S|(xkr@OegB!o!Yc3fOsaCUluF^=QoW9qtQyWR5W(IeLDHGSW+TrRn|xY#K>
zueT2Wpe@AFGQ}9pa=GH@=!n^DwrdZQ;QU;C?0g7(=R4owZ~o1{$zS;^|4aVzU;fK{
z^IPAfX?}2vjnTqnRDPO_&cqlg%1U<IMtDXp+Z&B}yVuWw&pbm{6r>ol*u+v+*^VbA
zLMcV)j)-0aeBMm@;D7k-*nAzGa~vKV^7$`*kuQAli_G`-U?jxnqHaWH<F0~{UtzT6
z#*JfMeCZ`_-n_+XwagK8SML2kw2|AY<i!z%TnjuXjrCsO5N${%;dAvBwIW5u$mdAi
zfbRxUALvSH%PV8?eafx9T-ZfHmW1{oN=G$Pc4TGVaqLiO*Zu`XRZqEn_c<1K?{WF`
z3EO5(=f$U$RwA^LRfIZ8VHYwbC@M4_t2?aQa{JiO7X{m{V%aAa=MDWj()ZAL=#>`X
zUeb6<QY?hNpu~cw?>$O+w#MUJN8yy1%Uez16qQpHMOH~?PMR3*8lnlrEHHA`Vyg;g
z73)=^**0|TfGG-0QS1amAp|fvqAN*Em(H{Y$q1KotVGR`sdi$6rFI4MxN9E+2(my@
zK9f<vgjNx)6V{j=b*w1dZVr=(KjFXjeNWeQ!od#-9U&6wdhk(a#7<hlv%)ejG!qjr
zv4<qyy-6uzlJ>dH1*9>f(p(UKJ*tp*w`^8tM>298(9wW4I}_;-Gd@ybR~+qg_cNa+
zb`8&7e}m9Rk`lE>k%F{xwCsw;0&nn%%SD5m2_aIWaG5=*wX_8b?J2dVatW)&<a6je
zDwSwuFv?(bK;^v4ZL<VDaO=iRTwP$CBlv-#?}&}(;^Km7U9oqthgF(pwZUnJ$5T}`
zd;13*9PBgA3&(tapHx*)7`)CgG+8;Qva;4Fg?C=edMQw;gjD34DiXCPB`+@DR^DeC
zG~J+hw%GEGM^E_O_a5@Cr>8u;=;;C^4CtDqOCk|MhfW)+q&eI6RJRW}zIlt)YR#iZ
zj|fqalegb^gSX#!o!iGp{Me8E7{B-WH+b~u5qtYLn9Ssc30^EJ3}7Zxo_p<c{JmfM
z6&hY98cWml*Z{TwNTiqpij!PFt3(8|YFbVf8`@q>0YZ>`K&>p3=^SMpMd5HLs`)<0
zcka<9!(|VZil`K^M2wP~dlM73kMx%-wx=ij{eS#Vm`<m>^vcUzEG}drO=GTT=EKB5
z;RGqTadd;ziwoxSJ+yVGq*!mZbd1KRc?QoPV{VN6Q7!B~SbV{1wW8}=ilU^>Vu+?~
zc<ZgVg{u}K!HfEBI-TzFl*dVKRF;p6)mXrMu>JZ68}j=k=58{nrJWh2;OqxaF^HuR
zDG`ZmHXD>u{INgw$M|3Wul_gu+OPjQzw+z<;s-hg=Ss?=#5s$xl2tsyEmc+JR;#5b
zO7`~lFd3a734dl(T$T!St9R$BD%qVwm4Ir@qZGw$rxaxfq$q`8@Sfp^iE`zkwHBUh
z-?3a?X4C`m(o!Qb$4=9slg`Rl5W=_O@bHl5A3WgIPrb^;$q6aMyl#s~;D<S`KqA3<
zjLK|-7)jCN`<}X<;0lLtWuvQdA()cnAKSKvZ3j(9>l8(4NwFes46~5icUm@_$%_&+
zgh0<gl@*)`v5nFyp_KTLdXn%keMeyx^IJE#|KbDoOUvr=ob%-co3<qm!W~T{*`@ZL
z)DJ@DPoAiKwt-qx=$3jGnNAF|RI)$Ow3|rNMAl7U({Ax?z<a|G9KFODNh?ZQS<jRp
ziESHbnuO93qa)Tpp#x=+RjAHTm6p186wXo<@_*+{F8CbURLuAGsp~0S+p}D3=r=2j
zHn?o%6@wIXq<~UU449Q5p()ADrj^ZhG$KNY8R65AIjWHub3EACp->96tlc399mtr(
zbUMN45|ox=UTc&VXg3Uj?P^7{+0eHg(ML#HfC-=2v@rNWjL1{Osldc(3N>JoN9B%z
z)jHG7G{z=T@mk^D8dvPlT%{!^c<cm>8>bk9Sdyg?kz~|ae5mo!>`o>X2QNHN=sPy+
zhUNQD(ZNe?En#?>ydDU|{BO%<HE>i{DD9*_o<s?1bfPd3OGj-3l}i{UoP!h*SLb}Q
z8EtS*6sw9spz3TYXbOXM!gDw~S&~#_x;Gc1J%+_*i`9nXo5$3X3H59$y?SfW)({Fy
zDhyad(n<=`YZ+Bz&Z7Y#sw#(024ceKM8u<1M^wGse7>afuxcXT_|8Lq^Bdpc^(QAh
z+IU){2+H7`MQhlG6+y2sg<)tV+cnK#9ec{QrSAq#p1sfE?MHO|z<RymYhU{s@4xdF
zo8@g@`|PXi?eBB3SaErINm)<jA<!mIFdmfRh0lJDv!>^4-Egy*vhOT~Eigt&i>qrP
z$<E9VfdGW4I9YBvT{QGj)JWQB>S_uoo_p?juuj^}Nz}(*{PLG@(>a^<IZh`^Jhj#o
zWx?!dj<*I?Xi{17OaGt0$=gq!@C*OmUu3qw$I1CQ8bhHhK7`y(_LwAfcJ!Y8y%}A%
z<>K;ECP$wLYUHeB@Ia4kA{o-0{i8cZv_vAtuFQr-O{%iz+>kR0$z)#7dcEGY(5@GA
z-^~^KxKJ2l41M3>oTaJ+QM|kmCAhRHWMZ_=k|Jz28-D4Reu+Q*r+$vV_SgOz-}%m4
z{JDSk&;CHikQ<A_bBUm>xbQn$;GDx)Ls3Y8cZdO<^sd-5M(6V-ygBFg1f|LrNy44U
zIkLuPevw8a#gR9?v~`tiCO^^feJ)u_Q9R0wV!gb?`yOeJjd`M(Q<u>vrO0z;V&rlb
zg=2qj&Pxwo<eR_yRhmsh=m&-&fB*M*eU)LqF+JkL4?%Y1*IkqF<<?iqEx++og&vFT
zh$2vEMd9)}x0dL`E_o=WL>~jr57?o{G%fvH(Uck&YZ7zf(2-Oi=_rWGpizTxym}qz
z)qufZ&=gU;5YZM-q$d(}5|ma+o_0BP*u#R#R5&~3=|y0ZJl;nFfg*ZhyM=a(HV&T#
zl9dQDuf#9SMUBg`WM<z`mx)Q|sXNbP6Y-mYzD+cJq8Y?^cb6E&m?D+qF43gidh|Zw
z33PqLYA_~I6rOVHaZXVbin_LxrKYSLb?u1BQ<hFBQj?OSvY_3z_+IQpG)g*RC{Pxa
za&Gc96j&07qVU?4mT0z=)wMc8Oeid*T$nPt)&WWrmBMDhlXDhrogixY>nXQp34t^W
z#BLz<*|Zu1wxH`Iesl;LA006bC=#WOR8CsqCIv~~$UJ2<Qq);?*}}$(a08Wdv2#~I
z=}|SSCGswF0zZIIeuOFEqa{x`i!MrL_wRFiwZ(_T^4-Un<gp2iar6m%?B%2xjZP)Q
z;Mq2TiL2<-7H0&(DYER^YL7FX!Wp#AFlx+5ixPVkViZHf($Wv|J%jJFFHoZ0Zdosv
zOsk5UhX)kal7i=aamix2;`s*;uw_Y5iOxsi%LUK0DjAG|%pnLVTm>~pY3+(5k48m<
zQ3X0GQZ}E|DT48!28P&S!<=?o@~y`w{Q4X3@QtVEynVJ}*;&fjA@^?F=4djf@R2?(
zc(Qz(n-Avv@jvlJzWMrFSXJ}nopVl}TyXj9lJoWJeC8*9lD+vpzxgkJgZnQ$;Ns$f
z^=8fGqU7~&f17$f#ruK9;*vY}?h9erXM@Zb2uU-$aSMtGr<W_z(Ogt{!yqQiL1MbJ
zEl?y1s~NhUrb}F`dzQU;6@o$~g>QS-mrG{*ha?5Pm*8Uq939`_{!0(|_LCC|4L^SW
zh}T}Y$8^32lL?!lr|q}A_5NeN@q6FoKm3pWx2#lT)3;3LbHWga!+<HBkVS*|6UI1B
z+G4x$?tAaaZaQcoZmX2B)9dZPwEO#yC}8)2aXpW8%u4=VV=VB-B#P@l|F8Q0A9Rex
zVNn!RRmF`PH#j;vLMg?1y=H%ZkCT&A`o5>GYf?-Uh2$u|_uhNF@x~j>W;5K!V0)r=
zb%8=>Vz)5{>++&t%&w&}7QJJUthCB#C%N_QWal07xlYoiGfLx?v?qfP1RJEi%T?P>
ztFeIkA&<O(C<WWiMo|0^BzZ$l%+-h8N^_lOCxrS*J>lNnyUeF^7R7?DZG^6)#n$En
z$jXj6HKyr&ly|L0+y+L$iTdEz$4@B4$sr+pY4JSMa$S`{8;kONZu7}pG9<AiN#2wC
zo*^1~Q-E_QT@!=YVW6a-S3#ko6m~(0mz*J?Q>Gv(lMAREb*7RC^ia~3C=G7nFsUR`
zv+P>7y`X_U1#Ixdt|Rpws+6Bs#*nlTDtHDYw35^iQy4KLE)1^JI2(y3(iMraQJ5~$
z4<Xx&<TfL@^@jXANgFQP$7n(fcooqqGK5Gs1guRs3vH(<Doxc|+OEI{nASr7sLGOZ
zQi0M8b|CZ~MzwSt=jz{I+^4a%om3X7jKPsnM0P4=Dp=kSsk{jt=@i!Bvbna+(SSQe
zvd(5WF+mXiWf%rAA67<k)3u{l(#aBjosjLdN)!$Xn=o3`x??g&%EBsRCCS2!tyn9+
za~Bh~i<8q=#ht#gK+<D@mp4)$c6_wLPH6^XC~q8d_p@Tw`1+=$dvcOPiPaTim&6Rx
z;D}z)wNOeT7dj1c%jig9BW01WR;qFzLsqL~x%h6P5$xY6!=##kk*~#JV7*>*_Usv(
z^_t_uL*bAq#cH+W^z@Yd{X?qh45cNAe%-Ya0&O(WXxXie{reC}UKE)Iao4tbte#7$
z;G;Nuqaw;Gf{HXg&_%=PlQrLZ<1t@<e8TI?74P?nyPy5z{Mn!U5=C;D)>HP0i}#+f
zXdV%C&*j-8e&@Hpfhi`eHZ6UqIXb?<dyig6mx;S~pX0gbp6BV)r)*o1lE3i63v`<m
z@4WLio<u#Jb8vVQ@4X<MDM#20g2W{2P_AUXZg&%__dbUbS0oZc)TlH-6d8Z-6K9)_
z%O()D#VMJX7iXt@<7;2#;P{y29NHQ2$Qye4X2TbM>L+>Ux4+80<ATqB`YyL;B4BxX
zdBVkN&9jqJ9zI#~^MC2z=L@fWhQIq8zrynEftO!+nYyfT+DftLhpZBk3EBGrqqP`U
zpSS1;r52S{7+bDfi3NT%{)aKf0&(PE&1N%Ref3p>54`#2oAiCY4>B=ZYenC8%w}`;
z_x3qCIbpS0)3!f2$8L=GyN(XZPG2%%M+NM4UZJ&uFa}9v2V}x_&Jm+QDW3%>;(^y2
zLu{I?#?-s`LGiW9MP)|r{731(d<<iZV3Zbk%Vx8tA9{+i`o2FS{IRQt&S)`>e(wH#
z4)zZ@KR;)?U5k-?W+JDZLgC{&@@o*!_do;3g;6Up&`<e27L~N#mDXZv-)Cx#kYl8L
zh>753(eHYLo7i0Jm-vvsC*?r-Ocx{bgVl)ADiIbW6e&cKN*F5z(=ZGKZ7_u)S&J$h
zHkQ;t-SpTdpnOmC67Lm<f!GJ=BdW4kBS!Om*AslURmp`;(i-Iy+K63QVGTG(-zDlc
z;TliVbaX?YAI2gtqLU^hEiGP*v_z9!+c^5xCWwQyC{-@>>=ulLvapn8WVVXTX9Kfo
zU^Y|i?^Tr5oT>(5B>0|Yv&Q$i1D6E(97PT(BYG+G@ls4ACFgH!Po_-s(OP9L-c><Y
z79~Yh?)af8#ax?=_eRdYjgkr5HVu9lu-afNM^c7n0Ij8X4<Qo0P>hU*(ixl+gj;yI
zQcz;V+twA1qAKO<Mmmp@CXrD(BSqX;)JgJ<q*X|UuxujLF8KVzjgJ;7HUx%JQSTox
zpY3B%oWB2<N0%4GL5d-XPYJ~22~nisF=;V-G*Nolu}AB$gDIU$SSJ@H0%(I$JtQGB
zrW}x~l}2exfhFmHPOx3AxjZ{#vszIX1qbsP$$M7KmcHwl)Dz|h2Mkf3IAtu=q{h`H
zz4tleFW`ca!p=tsvIjIa7j;?IA9`uK5<8#+4c(CNX2Quz@y+i%;#+TD@c1(DWZUw%
zO?>L7zRX|!kN;!-<e&PJoW1vyfBxV9ect@qH+ZzY;O+O{=d3yBvTN|~zRTcYZ&G1n
z#nIjov-uuR&(C=C&F`Sg84q55g)e^bCkaExZ~w-><l^Fjrzekj<E`)T>CgTc4_<f~
z?>)|rxwx6DlY8}ATe@yQCr#xFTww?)(#L@W3?X8YC;EZt6Wi9aY&_clNf{IYlN8V1
zf5;0jzQD!FGgMvDhroKhB*g(00?)tnfCsOBig5WhF*FQq&&6io;pv)Zi-z;_Eua1L
z9bUP2hs$Trm`<ns)8F}R4)<>`(-jPX`f!eq0j=`>EO<8SHO{(hGm+aRStg!75G9rP
zz_$Gl1^W2W_^{)8dv~+m@bKb-rfK<~|NDQ52M-?b$}2BZR~6rR``s)&ka_0by?Y!U
z9`fwjvt5eA4?02+bi`Q8WHOON4lN7$n2BmLY9}HT;p2%5zr>4?H=-jyag10C*Ro_s
zHSP8La%ZJtgsL$N8XZRC_aD;8ZK<K}*{nA#&M!H>ag*6}{;%dUeAhGP3dajCy}-Tu
z_c*&aWw~02YB(1)Q78CVYSw?<xyfd_NP}+VX2qBYp~JXB;w6m{qi$`bwVDqvCh#HB
zdk=k&Z=jr(l*(dUNi%qoig={Z8KT5!r2q_45IT#f9<#b|Y%>}y1$^*<=nO?!Lsb%5
zM~nptsHPKKxy7^{5~$jvLJ+gIpeXHx%GS_s*(_IdZAZa?aWO|L8c|>xjj_3~&;?bM
zsAf>Mp0+J$yPmci*fu@=kgZViJsXmsc`-udN<k_PUmBIIhC)jCK|z<;ZUrp4yzHs#
zj=E}?OdIz1JB|+L)OA5sTFSyPKRO_#K#Vze-Fqp1GpbfZu@{Ls%1eqgEya%#97`KX
zn=;N^wKy}<J_>H7HW8%A>{PIFkq7VTyOzyn3tC(PtaT)%@yj-=+#`wsLN9BfhQgXG
zjL5266D5+&375{5B}G~63RQ*5?O1saqbNyh0h**WYvpTWv1bKM)*r#j>7yAGD@|2R
zspk6>z`SXB?$uZDmuFl)e4o%}Gs-j?O~`{K#hv8YIiqk+NqBFOs9eP80V575#Eg;|
ztE4SE1cD!gv>hZ?N+t1(Donw8xuV@{vA}#Xp)Lx7_pB}#Kw>&S#1xi3c#O6XBa`U_
zAH3wd76t3g28GxSsFbi|Qe##VM&w41I*O)3%S{-7ZFIc*WX*4U^)c_A^qgY(*7*t5
zWRIWvnV;d${`@a+T<x=bbixp}eB<q}@%R44-{#@7XDm03K>b)!QaA-++cS3q%WcPW
zT2ea4SAOSr(dC?<`tr~4%B!Eo82H*(ev5bBeVgC>t=}Lh*gv`<ioXAUdvDew*_r0|
z{od^?vE^DDS;f7Z&86wKrX_o14=i~E7=|Snu)M(f0EXck>vQ<ZfPFDC1TQdPUl@?!
z3k@{mfjw>YxHUb~-P6q>+1$yhB3YGHm0Lud<z0R8o)ejzO{`{%J!-_CNEEXwBl4Vx
zH_mz9=lL&EW>T&9^nFLLJ>Ey!rolOfqRL|iABl3XA_))$#E_68K@9j1IA5(fU9RZ^
zI3hMUu3x*xvo~)dQ{<hu-r#t>Vr9ELvhkMH@{IeRJmQ5{zR5@5`H;5n`Rr`Tqm}2R
zSKNKj@Z#65@$xrc!5}z&^pM$n%FD05!ax1(-{L>`wO{A!&pb!lG@z@@V-z|0qiq|e
zyNj$qWVn~4Y;j{G_%QC<NB+d)tnMv-QMf#cEvt%@avtp2`5Cuv-NG2ndbQ@0Pd}q=
z+RVLl`C4;uaIi_laNd3CW04|vI^qzGCS@M^W($(IQNU`YsH$=prHnpy#pZcZ%BHQO
z>1axct{YXMY4aF0+<>xEzac13(s}$Rl^W)jUrk|`hax!=%Ua8Nz2^M<Y!f>t27~oK
z^RXkrj$A+qA(+f&Tz~Qj?tXHYqlfn~ttKRojQKj6E)|1c?hMnP{#Gy$zBB)B5HXT%
zF$RWJ$Z?9!xnY74hU>;S7S2OKA0yHS2%aKjtB<55s;;q!i{zz9_MRJluS{9pua!oT
zM}Q%QOB3u|#HW-<QX-8(l@&=Fl2RZgx=>W5!j}2>8ZGmnbm*9ARUy$h8(5#6)2<t|
z^eCT;ph60YVT}o?kY?a?hKP&_p#`Q8R8G_Pma<Uv*5iE0oH`%!|A&MNDSNmv@b5Ne
z+Y&b8+fhY6Bt3Kv=Pga=Sgs?>wPUsJnawI@(~`xaq^cxJr990m4P}w7wz4Sag5x_1
z353u{p>x+usNtndpwhrk8&;(F{PZRbYi=|L7HPPzhk&;h*INMUdO}$iNU5<tB1FF_
z%(Mh8GZV4Ug0fI(mFWjcNlYOrYm<e2N^f%GH>*^Li#0X~CK!u2sZdI93Reajlzgoa
z$oMa*W2cnC6eUubJT@r`c5mFkzw{E;d(Lj(Cxws+j6!ToL$gv>K-5TSP%@%qo{b9`
z2ja0I$?OCzaFjaIDTkMelnl;7?;@4<AO&SHMYu>@H7I8>5(=e=0i1^<HH*a@HL0)#
z*}Kb>q$F52Emd735kwc5*AqJ16QV~8jgkh^Kp2gQpdwxdOzaUpXOT8Gan`|4Zy)i!
z*Y9#3cR21epEM`z-F%i;zwrt$KKC_D^c;V7m;d#D{te##=%@VntsisRFKJ152}B7=
zAyJ5wvtK(4Ss?^U)>vox=>2zi{YQVmmBTsjzWp};*Khq6Z@=|s_V82Cty}7{M9Iui
z2*J^{4c>Ny5Lup`(6<drR@uZcCR}i&VWyvgMa8_Ub6U~+#91F%b)YDzlg;bz<Pksk
z?ce5wzw#@bM#=BL`38OPSl<()LrJL03Degm>_71=?;YJ`p%qP%+&x}$<K{l!e)VPc
zO+|BdLO56;&5T!Gc!^y-=R3dm&$x5<HviVI{oANsQ<@^*=zY(sTeDYOVNz7c04hRN
z<lT52qJR>YEowKcyT@7PTYSYBO<KLnwhr_8jP-iOg9i^VMzgcCpzT{;ef3pt-n_+P
zG2;*Z;0LT$D}M6&PuSbr%K{Mh<qNy9P>X{_iB`kzS;WDUYUIGk;b+HjNgflAa?!C<
zM}(xZg-9+GFGeoo@5-@m`AcbqisO!6eD!;i#OB=FwjD;?9U+7~^811iLQqx}SFc`U
zcXyYnsxaCNo};;kb`gEP`PD!BL62XbYu0b<YcHA%N-|QaL<kp_8zQ^^Cq}iR%#$Lr
zBGrWm-Xp!i`$*KF%8Fh}k`tS3^qex0P3L6H=b4=qBuPW`xnnf)_;M0P%;RC5IqY!=
zj#n}h?n|SnON~gNs~n-l5NvO>K_ZFP)2-HYO^Z<h=~LdVONBHFVY0i#$U8&IY|$li
zx62rrZB3M<?=8+{UZ}HRUBnHl(0((9jie|yo{K4M<TL}@dU#DzBs!TNZv&leu{N+;
zcTA@RU0*Yu8Hz$t6q2egkjmr^oXjP;7$wSiFpJ1h14we-JP6{zXGC3W_!-yNApFYw
zQ89>oE{=B0w;uDa>$=VvxCorLS?D0eAb=1UnN_Z3At{R7$&*s#V^+DlmR*~37n2Y!
z+1`+^5i#b!BOwQI#K6SZcrI=}{X-}JajpfAXLnMTIdVs+%qdU>#qK`yr=DV`S#y4T
zf<Ng&XjIIzZkarikY(OM531M(a0oew6Qn_;%;AXf9EHP(fWqUGgjiyGMbk8xx<?y{
zmXgqPq^3dmo?=o{R~1ocOkJ}xnX<Dt!|Q<8f}*HVq96q*eSQ^%lp7U`)(YF@=&-ub
zIcrp8rbWo<foR_ltYFnD?mV#k(R&a0;Rg?}(?d>CJPJL}y!aB&-g=U~odwIYW4vGS
zfBxqG!SDUxpL5nWbkS#l$sjUIF{1Dk2#he)dd|euR7Fi$C}Oe*_mCH#dxoP=-sivl
zZ@vpcaR1%|%A&}NH$@hc8P~5|MIq>&B?L<hmcCyTT;#KlKO(j*ld>30RP%NwiZm#E
z0!7H9=3XQodC&b-PumMjluVN2dM&uIBk63zqYrQMwO{*N{Nw-UyO^qEZ+FiA-jvzA
zW<Fa`O-jUkpP#<_F$c4P&NW1%c=`D!*uyhxTO1{ERv?XLCJO$IZ+?rLx1QiP|LJdX
z_|Ch$`YYci_z0!MCCBmdglCHv*k9~3m7gJ^L#KdD37s<0)F;m%Yq@lBj~A@?KM&j5
z?-vGvKr5YZutPVi>pDKY_Ziisq^M2KSr(vmR_un5z25TZGrxI9NB6(<v1pUkt+_)G
z2MeG`Sq?6aGLM+jrQ&Vm-~l<4Q|8S^7_2>{kVuo$DBPvuQ*L;8**8)fjdB_l1Inv`
z2%ftX=_^>E`NopZD}<1Zs87#m)@!_X=%V_ZVCHfj@MUC<prkAcZajH|gTn)+lk6!v
zhQvp4(dGMcoY;%9?UEWJ-+P2YB)w4-2r1E~Af<o^eu%@&hO{Ecjg5vHGEmdNv-2sD
zLcm*xbOqLWf&p1nIH`%l)FUN%QP?37DUU%(3`V>Xsl|rREe6h7PC<|aWl&X}i!qVc
zRi^}1DIprF(x4_&OsNnmWu9K45dwm<wC5}Ora`HIG>Pa1J|<EDu}qi>WTA#mP&V{W
zG9jfvD@kEAbx~k#kF(j--#L%73hM)Xh_nuT6j`X@!TE#>ft17`E=bv5ZwM?HdRqfk
zB6t6~evLj~VGKMxF4)_fP){l*lZu^%U|#7V16E@U2r-IN;#jriVt5c*<R0d@0>)3u
zpI%f#<c#*LuNg~ya{Q>xo`H2Wp`K0=Lg9VD#}3G<R$~;UQ8*Q;N=;R1>QW<RKXeIm
z!E1dMg(NZ5K*QuOr_?BPw(8BwY$=f{qGCdg*|?glFOX@l=lMLR_lw~1MzJb8t0M(b
zWYv$X4fWm*`%m6rb?-V4ti?8VP#7gNkphWku!c(lfnho=INVog8F1Y|AoUR+JxY(R
z?*iLpX|wnFtM46!nV^iOS+*=~L(+<RZx=nEqD)0KouZ6HX^^3(2r2szZMuVMFdEMC
zZSMoI&%zDsJDiF`(I+XqM7Eyb9mhU!f7$TvN00cU_m24BVaM#+bF7l0L-Ork{Z+1A
zIUx3HOr>d>Gk*HsAM@Ss{~qUkLzf~h4f%x$1o`4^grSfXg{&!bjaD^~+s$=7VLB`6
zTgRj0``FO&+zZd+oa4dKBkH=Ms3!c?U;XP`d-5hy8+_Z*Hf`Q@H!UJ0K6v}55aHUv
z0b<x5WWw)Y6_O$XwP}C1s}(1Y9?`d+onp$t{x16mlEq%d#}8NB{MOgGa_cI8?XUd`
zNYKLLZA*K;V)gKh_2d-y@R;Rtg`$G4;rW{jcD16=f-F_fsw6q^mq_r;GtcsO|Kq=l
z4w~QtXhq*y2#NDY#|RQTdwZ0ViuJjpje%MYyb8bsLBF{qZ^?E4DnJaO+jUuSc6LJ3
zw4{{y=%bI>+uP;Jm8-;<z4Y8LI#Wv0cOAPsJ42CYdE?DDzVxw-QDLSis3#M=b%-Q3
zDH3BhM=C`Kk*ZRg$6QvrX|3{zFI{A%j=Vd_Hwxz*VMv;Y!<bhqO;A2>c(p=!$(M~u
zdn}~PW%53-UadGeIp+NQoUZGrCi5@;Jj2NA`i<*cy?&kD{eA8~xQ8~Hn6k*kjU6&Z
zHSEuZF_JB0qtv0Otl8VWLf2Z_^%`$`LX51s7Ja!(LLS9QsVT~mew;YTOw9Kwb9rM-
z_|9Ui!S)_kCE}zc)l>SeC-~frwdAjarqty{VK*G-nBXAea14>$Efj)2M#||FD^%v_
zC4rHWLQ7(yPzEOZQ_5KZrNrt;wWvTuyzQ{fir_4rb>I>)40cdDQk9O%2CAJJWi(3U
zF{+X#cg6-&UksE+6RZ5X#2ALpoX5IA=L61N+>dRDI0w#qeDH%ZS8ft5ToAaFMI&Q6
zL>wODeHMJ2pE<N%QWT2Ce9q2n#<b3b-ePx>N3~kzx<Dj`=-7eCC<rMKg~)z(T!^t4
z3b^d3aIwC}orFjUV#N17VxWRdXERJu=Atg-&UI0hh@t@HY1a;;5_Kt=)S9x8*<(;d
z82GW$n0$C?6e&i=M4Y_VyWy^s$FYSeP+AX3HiojUhVS#WW$ZnU8^tez#~j6G)kjJJ
z7aTTNLh^VGW_QN^v$tq%PkZ-1Y3-?$LdQf&#4y^IC}J=KpNL5jlOrh)ArV7%)h1}P
zkyH~w*OCPkDWa6YTR*ryN6wlv+T_{Yy~1Sw3dc>y$De$};gvm}dHN=P-Q`YL%%>_%
zf=UoXKve|+&#GPFdQa(VirS!c!pZzW)m{<2f>oqhTYj=$^3UJ@n0M};a@4{8jc0i%
z3Z{F9{JsD1KVmkUbNi!rsdr00c>f*V`|us!eD7_}+f_EBgzWh=mUK}73Xuv#st~ax
z`GVFaVz8WxHPcG7w_h`<ud=s)g}2`MF^?XeFquqv^2w6PVvko|{T6@Y-~C%u)r_{!
z-gkX#v2Bk`(63kAe*10qD#gLh6eGppSC#wLgH$9-zeA#LdzQziv?rEp^O9FydX9s`
zlHNU_k(TSzeUj>Uc>D=nd&>O>_rW>ry2V+Gl!m)eFx7(62)f|8T4?ry<Z4+_=!7HE
zdz%U8WnS3E7?jFfrHYcqS$t4g1!V)>c}KFA$+Uv1pj+mhkCafvOn8<e^G~wcNnGsk
z{yd03@#nS<{K}E@6D8mN_P01aJL8Q%ehYxra?NVB2E+yVTjw!mRh1;tW`xo-Er0Y!
zKm5|iqGaxz0odMR2F2@iI;ALz&lPZaWE-+z$r#EF=WZO&MxsuL5b{PbgiT=wX%I-L
zjrp${6mTQ&E(<mU;VayqY;GbkV4Y>XTCrZQ>Dn%*@Qittf6?)bT}F{u?Ch|Au+QG!
zUKX85xv`}fCv0E#)J8q**g#ZLGoLSb`Q?}K7T$mNU5@U5MvN98F6VmZ>!7oa%mp9w
zt+KK-2|4iW@)*zih_@bR9o7fDlu+0BGlfeY5p%Sr7mBDfN|zT@KNv>O!-hJI8Lx6<
zt<uLpJ(&`uCJxppQX9|((Z_69T@^$v5JsRYO*JV9z9+aIBLHI{_-wV3V%`Kf^8GdC
z?p=yG`9Mj%@m3TACrzpR-b4|JL(ElSied8<sGYwkdP2lHkF_4>ayQNmqwNqAAtd@B
z=|VwB7XqmyMhAmTRS4OLfnXECb~I;xP}LgdJ2kVNipjK~o)%1}hRUd%B9Uz@1_R%m
zIx!}VY`B{#UbKYM@cmGArIZLRt8tYY(lU(B{-`11bMzIH>6G4ioNETxi7bdI3q@Hf
z5EW8om2*`U$YLzU1V}j)=@;uY4OD=bQdTjOH&WJmVrY=dHM$rqmdcW<swk^+NN)OK
zc+A31Atbn9v96`DEw*on-V#eged7vOQ}*OJ`RFrbA1PwO<X+IAU{N@yI6Yk>RAjLO
zN(PdOh$xXk0UBil)l^fg^TP4Y6TOBa&(^!PrSDs=T)Dz*|2hv(d)|ESWA1!%%*!tY
zX49HUSz==*TM81%I>Oo_wIqraMV&{I!IQ_wXtYf9wxwO3v9by83eHa*?|*!s@7_D%
z-Sdt$hQ-yJw3B_Fe*PQ$?SKF8Feysbr>E@iF8JN={x;wL(f3(3OZt#S77i}u5DG++
zNHm&?S}!Qo6uc&Sg^>kyot0PG2$rWy-g)zgThCnKg=hZ?yN46XO7YzD&+_cozs}3A
zeT&7;ZXPW(4eQmK=sdRR5I%74gAcj?(Z_u2<`t$ob2e~(P&bX^i=j0w6dJ`uE3PaA
z*Kh3e%5w(@(cnVKc@R`;!aJYc=7-<=2}jG0<I{5%Gs9v&;mVb#I6S<@`|p0h-OrBD
z!7~?;YXT-ou}}ttXXPzU&Y4dpv_r&9=6~euUb;y1fg}Y!T4K{6+aA9<V^SG(W$-%h
zNXrB&Z_g-{qVj>1azxO-uzUSEPR<s;2&9-O3&sB4oGXWWeEh*jESK4K;NGYA@UJ-X
zhFENrZj<R`*!gE4#!o)E!#nT3^W~3a6gv#0YBd-o8)G&jmCaa1etzMM|8`jz5n~!g
zKoRf8zZ*@AGskY^1m)+Ho8l32(JSK5&jk7lyaJoCkN2K+v!+?EX}fkPMm{f;`17A7
zqn=FI+1(uqyBxHxgcy7=e@@4Hc?3)Dfb8z<^5Tmxp+(8^^qiANkFeHd?jkXp<ddy5
zg7bhR$vo1FX?&_|-!*z;`jl`(Vdq@HBcZDBO68kwOmq<}5Ja}EV8flt$2b;uqgShx
z3I>s@i-~zzBXvO#Iv0VeAe9v%Cam+sq9hDNeQ7jBsfocMf}>IrQ<u5;6cAM;`Ru*s
z#`mnt9pFM27)|cjjiYNR@-ZtAgo`z(LC1uN0b_DuH;Rd@b66{IPT*V`_;}eeC4__y
z8aW6rLQDjgtwJ^|w<I9uW=9Bu&POagjq1@_uxuKp>zdi3Vz#K*byF7elG0?pn<iVM
zC9zommzG8vUjLN4bDK6={z@SdtR;nrDh$RH2r(Em$E=)|T2odfKIExZjH4NI7D{L%
zQK~>HP+Fi(_OpzM?AS0Wc!&Qsf(ZsePe_E6g9I|C(G$|pB@Lc_@S3gq689K^=Z}{>
zA#y;ta~*BFqUl$3?V8|h&O=+w**7{93T$9~`(AGG4$<@?Wf$v2LDRR0rKg%!XwxAS
z`9{tV$r2N~kQ5Vv^){cll@z6dn4@9#cNZKU9CFr4KK%F*?|pPaJ+E2pU7=|M$$8o+
zSrO>ko^IW-IzPwtJ<)rvUs>?N^G`9UC!ht<dQ5Wkk4`y1KH)*rb7U27-COeJr{^36
zO^^!?Z#+$`X1wt7Yy8)L|L^nE6W93gkKg8j>-k6j=(qU(AOAkBZE1oh2?)c*Aq*;_
zT+nGW1%;e5F?&p_T?mF0By~|Uoz2;)_t09gYEE!rMZcc#<jp5|?&<6N)qnF_y!hJJ
zDQknOD{SA>tX7<#p3yE>*!7y!btG?j^9Mh~^gXXVe~Wo(NZ#cgSV|-z2+<-^AcTbX
zE-yeIc=pK|Pd|M~DO*S#-Zr?Ca~_@Er|A@C?GbCo!Blhg`T_NHiYyCS+fh`8Vye;P
zg&nL%YD3%hRI}{7ZpwnPDhFby%>F?m$3=qChSoc@wTNa-?RwnF3H8-$%oa5~Pz3L>
zF?WQOmefWglpytvZrBN@A&KJh$9~(G`^91`UQ$Zz?bPfn3hvzgke9ykGEd#Q&b#k@
zgm;NY4<GQ}yYJ-eR3a&bfdiM5N%r=3DGJT&um6PAa`~l?tGE3|HD`pN7%WWks4rZI
z9R|AsrSxVLI~GJ^0XkwFwSK;6yOiVsgux7YV}meYG%fzh5hLDtR?8*l=jSx5^*}_w
zC{i}Z_-7tp{;9Gq*+1Cl>b0vFZ7@pX#6_;gWrp0(ey9VNQD#+hT~%B?yu!cs&2O=@
z*yrO9Kj!vFALCq0RU1U`$Pf@AAw!N0^Dzw>rMZ(LB5092R*{$m6fxqgr|TTO^;jWL
zb&XesJ_(YKNEMj-h>gQ^DhbqRv$1(;at3*DZt(WYg`JHc>j|Qq5Q~c5fh%g#6rwD!
zO-rn%L{lMj78z)z5nY1{foW-w^BKKObg@U^@zEg$Z&pVlGzpR;7C9;_4x)h3dL<_x
z3=vm@B73O&vdOZRL=y+=kcjsR>jd6Qyc2YN#Ce~KxCAZ=ib2rme8TmK-X>DW{PGZn
zsnuwR90scu7ifu`uGZ+&CFP`GGA}te*x~y1U3PaS%q9g%B}B?u^+M$Lf0V<fVZo*l
zhX}9>uhWzgDR{i?A;p0wTjXw9OoWs<a%EX&8=SDh*&dQ(GMP};nxd3wEzm{{Q`I~{
zh&By5ule&VF$4E@5LuOG=*Z{stJDS|vJk~NPusVw+BIF@(seE8XJ=@m@lx?c@puNs
zDMt_WT}#(2v0a1fyCKL|V~UFDbVjWwm>{^*S(<wfFvHG=n1~Ww0`Cpo(qSf%y@Lv)
zES?^zbH&`+4y{1y9LwPyw2jRY=RL{pVnHb?dh7Z0Uc<+qp0l)q!-E5SG2@d*$Jn!D
zPQB+&yJmUb6FP_JBSnOHAxNF$%5=fxi3*!KY}eAAoN@N(nB#TF0~FuC|A;?4YFR2t
zsTW+k@(k5%kNq1@@?ZWJ{}r!&^J^p*xpKJ2|MCz1JO1#8-{;J)uqoj4x*9HCxu6pm
zbSfw)sPv4=>@X?!DdZF-3yMiaRaO{NAjO2rOqm?+p>&|0g6-g~H}3H0(FuEd`#k&V
zb6DGPdUC?q@oDz-J3ApZE2QhW{oXsg@x!0+E6?BH#aq{yV3`(WCdMX#i#gUsMu;jw
z64YwS!QMVqT{D>!^zAuXM?_KxD>*1<6ubL8>hDwT?$OJHnO3wh;B7}I0@6rsJb4X;
zplUk&x}~W~8e<R<X7!9TUm&E!g?s^0)A@$iAwwFBDFb*1s};K>+|d#HR}a~pXH~OM
z2J0P_1fwKcK$(Tc$OvuBJK^#2ahY2C^SG&O@%R`Ex^blXH~;3p!`aCZ@4x>(AAImW
z*RCD%;K311(^1!k+qXZ$2hS5v-q`T#_V@RgOs2$;Xqq)Y`q7U!)(Bq)QV6_r`Nme&
zdF<o^m4LFU5yQWG@0d<!R8_SZul8L}RaFSXCYj=LngnpEE0-_QqncJ5gR}O6ggtQb
z<e(t^%2;-n##GKZmgh@OPfl}$-=*SYEMWiQ=gRrgyL)>)@#K?CXA{oP{)F21=P+iL
zGIJN}x}@(ceb;k-wq$v><i!_X;>Hs<*gH7jY;{agR_GK+&f?Zxw!8=-(|f{YMI}SG
zE|cv;%*}=v1i?j|5A-pzSS(0INsq+EgtwNmZP~RwUI^4sC<u`@{%??*7`C;@ve+a?
zI3bCXinzCrb&jTMSX{qOEd}*@jW~M9Dkkb`K|R}LvY3->0P9e}Q4pzT6@?3!WKcMQ
zvZyZe+H4GTeG5bCNs3UEHHvJ_;*-axVbd~n?_$a$$(TdLF=KI&vOJ>D5>uKDC84w>
z)^@{d8gUWoBne3nlg9f**9M%8cpI@U<!sm#v-rekkHa8xW_yf^-b24~oHw3(kJjA1
zf6T$&jKlpo*AC}QCk2I!Xpz;b0+B6gM#JzBH{Cle<v`C1IrYVbY(^}z5q?T!GwP5F
zyQ<2r6us?<(KDY{Xesc%VYO}%GTXFV_|%OSDtR*yV;V#}VzV*2P}VC<Q81k^n9gQY
zbxl>()O8Lh$f+bb+Htj7alBl8;m^cl7IxsBrEAyp%^KIX#3AoZNr^TkrmiWADcC@B
zeva)s+Ou<H5){f1g`kfK8#Vn}l8z(g&V;EJI6Ab5#As0jqVy;u(ItQ)_=s%-BIST`
z+xFZ$KH=`8$R}rx#>_dR<n`M}oZWxO`eDl|MB0?qB|D|!%HfpjI}4t^alnm(1rsA_
z&X4gT(ym&L?jL6>lc0F(c**O>Eh}y4v|%y5%5>)n^>ogE{-6FmUV7nqeB(K~_Zh$O
z5C41q$#?z<r*26{7$Rc^dRNL4ZG}NmU??e>Qmb7`xj>~Unkj{;smhA7EV8$i$}uR_
zq(Ey(vg5)1dz>|&@Pi-SVgJgE|LppILKj+&A3kJxdQP`o;+qw8EvrWl_}=e+hay=1
z`ZvGMR3vIC2sVeRN0mDfN=uT5C_Q<TDyi!U)ubY(fQy#8-o-^t6IQgMVE>6*Tu(Kh
zwGAP3n6emlX#(p#LMQCHr&%_LwqunN#aeObdY(yBCPCv8w0(ytrbMzjE5)=?i>Vl?
z3WM$f{iBCGH8<?gvyhR4X-#eNE-1!;Dg<R|NFgAj$K<#QoXZy%r4)Kp!I1{V)}QV^
zZ1FE2V^NS&;-!~f;v3)i2JgR<xO?Xl?%uh}WIACwsR`aumYS5Z9A@-cgG_XO?X}ma
z%8K9p-S2Sw&K-RC@`c^V*vfpMf$%(ZGhzrZkerJ$$DdX8P}n8tx-MrN`+zY6pJ?3K
zjn6K`rnt-B4~5Cl;TtDKz=p4f2p8SG$5hAttQa@(P1|sOe#Y79Nw#YlZk%5PLgrD$
zbTVas|A56}!TrxZy<mSZipaiPZ6}4q42IH@Os!chSG@b~dtANxByHPoM&M1`Q^|x3
zIl~vmUp`P#l+H?g5^1ndBIa>fwxP)D!^cGD1G{w%rXVSWPact=i@9Tm(wo;R*UM?6
zsLu907B5AX7D|ar39mJ2u}7gag*IIO=BrSYbk_64$r(o<d;lS$Oi5MEajgTJO?_nw
zXq6KqG!i2W%7GD*=mWw#EFCU5a9PZdAa{*M`aqDFk<72u;Jcx6w=0b)8*ww$l#HDy
zgD%LXyhTm$o)A3NSyXbHNH9vxyf!$Oa5m8QG3T(lEDi`h<;G7+xCp^#b#Wx~4pW5V
zQ%lpFaB{Nb^mNAI-i&#zn3RgT$coyrD=JbJzl@#7Al0x&0>N3F>nV$3WAP)ECd9NE
znMR;#8l3G(DPoGOvJTM?KbP%THlK*m_9f=`JBA@CBVG6%<{I;&&^tR_QC1c8q-H)}
zu-MzdObS$yIe$Ali`*rgeR1*9;}v!x(zPw^dWGvc(vYd9g+dEWp=yLE3BpnA?Qrdd
z=MX7!^wW3o=W7JnJl7GiQDftTv&M09o-n1sBM~BIV_>7R6^RiPWmXBLkYB#8>1o?F
zC(BdruN&^IHTPU%?Gr~Ie}?xx$_8|l>=&Bnb~M)ycRAc$aP?r1{rQwh0i})v*U}sx
zvsx`#_K~|y$7h}7&MNWlX-6X}QYhG)?sM(QTb!OX{MxVoI?p`)6syy7&QC4>@PGOZ
z{<nYpzjJEO=tabjysfc^9dl7vQ06aDQ>h(FHABnkU>aN^Qh}6)vaFd*Dhib)=}JX<
z+p$_d;*+BfIa%GNwfA}Z?bmtd?UxBM(yrFHw!^L0$i64FE57&teuuZ-c!z)QU;jG$
zbxA=W^c_ApOj#2K??B&Kyz8+(fY1a7)<^20UEH3ZQU}Sp7u;R<xW$AA%Oxj7I;{}8
z#D&~~*SZ2NNHXG2Ez8p-0$8TV>Ahn_kyyO^0`tW#x~yp+Crk{nZ6V~{q!el(L?;TF
z*eN7eZ(he20v#<^ccxt1UGU<wukhwieoXIHLo0q5d&eQwL^6y;#%8}7ENm|I>$W_*
zuMA_CN6Kvf@yaW&P!<)7*__#Aino^aatS1=QezZAR+r{NCy(AvPfz%r-}xQ(_Ya2M
zbU@~4;V*?eGR;0*qvgrxBF_Lh2z*8XMIK>YbRx&mD-0u_Q3E%2?f5*G?s7-(GBT(6
z65nn#H_cD{<xcc3z*szn5a|1!^?J>6xx_h_&-+hs3;(R{o!E586J=R4U(A`$=9_4z
zKh-YZRChnad5zcF0U<`#>op&|{{g!@hdelXfOXk?w(tAw+jfb6o5(lvv3r_=XUtt3
zIdNe)oAEqjOnC!4nUb_7422m3e=rQ{>5JDTZBBVKf%eYfr9_e~LI}f{dcL5V&Zrg(
z7Ee8gC`x7lX4hIoGMwGNpN(>LMc17pozKM{38`{ctsti>2(2(8cLY&_X2`Y;(c`^E
zA_*yM)}xNu{9O}C<P6hvd5vyPB4&@qp~EZ&-l;*MWI;3`(|VLd3_^sF%3ySnb7`HX
z?*m=uu{Pjcz<aPU(FeeT2SO0Ulx<}4<1*K=?=9YYZ132eYIY`?#k8O-%3MPT2KBGl
z6kOT-YKWoqAvXe~*mxKgQW0Vn&mi(z3Na#;qO8h+d+P^loR~YkVhCdxKI>}WD~ge}
zAaYtx6d;D@OEHh4eTZ1=Y1@uw-OzWIM52xp%1N#lOi`eeek@+<7b@(=3)|`HjO032
z=f~K-Ay}J3;gumq0jZ&Po)kSgMyi874&77O^ED^Uhb&u5)RJ|Gv|a-h)<!-)I)kD>
zgg_~vR1zNqdJq~oA#hTVjHXMTJ4X*`Qo|~)I9f|?FC|AYvBVSCJyO6-NDk+FJp05U
zuUsoR*xkVxMOA8qjP%jcv@NUUF%OUL<JOiF4|hBG!Trdm7Q8GV8B{df+`Ym3``i6a
z#jDIRzVaQl-<88&3-9}U_y6qw59|N1ueoEH+aX@IqyEvMr58jtrDUm1$#PqnJ|(R!
zESSsKeYMnD-qP5BFt$*K9^JLvDzh!D?VtW@-fd&Gx4HNDj8jEb-x~hEul~=IS#*3s
zn3SZ_t*CoH4%<K2egAK5T+Jt~?RobWysf^bsF2fa#<e|b!q*l7=C@g`M-;uCc`iiu
zi*j`@dtG&ISHq@)?-geQ1y-DFVNYK!vsSnN+<8l2Iq9)!(abwu{RfiL+n;TD{vxN9
zS!!y<^Lvu|XWk{QE`D=r<M%VV*Gd-|i2B}sRC9`Jk4Q$X0*8N=bWoSis!64vWaMoZ
z%#khHuo>8ExVIqI+@fesWagD;Y7>)gX*}KYJ)`KKnnw-S{Qa$eCYii*i{`U0*!FMJ
z!y>_gzrRX>2UXQRFVG9TZtUmh|Du}PCGE3h=-gs4f4}LLzDJ))2jBm$GhI9G$1QH)
z$&#NQyihzHy7{N9{pmu+FW1tht_y0JFm>Us{FlGyI84;#2zfSt?X)FRG*0WRE4cTt
zM(%Ro-F*UomIZ98_0RuZ1H3i!*z+}K?V`>-+AD2lCg&kN@mNL8&J#~Q=h`a%*eO=|
z*g}hK&32Qwvo3EI?vs-#{PW?Vo1=x;#-9TJdJdH+tX)3i#c#2Otf1$MDm8eF{ImbO
zSnADRv&UKPc<bx0v-Z92lsk7=GJoGstA07VCr`e#T;80l!1q3X-^o=bZo9bz1an2)
zPR%)F%;nL+{9G&7MI^IhLDp<G>$MwHtP<whqzZij-s_NYt1rNeW3fk2-Xc5irAG2q
zm2Gc%4rQbSF33H9MuICg;JWFnSBk4S7V$SIy`1&(s=9@J^uu3$+rH1Svq^LF&a(O?
zpy+x_;AQX<aZ7d|j@}DK#l06x_PM+C^D;fXq_#eS?~uyVJ8l0}x42~WPGg=P;<4z&
zN?@sQUE^1WLe3FaFDao}3yM-wJ!JRxM_gBKNzdbI3CvLKj1rilka@tlh*9>yO_rnw
zO2>3gY*4)2x>RFA;HBdSZmqS9E`Q^rx?A|@!9_`Ca|1G3FLs@pagbY2KW{CsLh<%<
z-z?q>8YZn(R$CDqX5at&@DpY;RhQ7GE2r>xJQ2`TaynsVr5N+-hO5LoW2M=58D_?*
z1?jy>;<@Rr?!v*CExYY)hwt%|yAq%3e9k+UVwY+)H^5g-V`)ZP_Al|B`#Qg{*1ZBA
zSA1tz>Fe&=Q`>GNb?eqO?kIlFXP$qrX3g=O^@Y(n8;|bi`k{1vyMD-xqY^*AT=t)N
z{`r21)@x#5DF#i|X{S~FZ9X1pJN(Uc<KAPr8;^?B&-`cs)&SywlmUs4)8+fOue}Sj
t0c_2-8;|sLx3%7dNpC!=&q+n@zxqpFNAlj~ACYDN0#8>zmvv4FO#n7J9M}K=

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_getting_started_003.png b/0.16/_images/sphx_glr_plot_transforms_getting_started_003.png
new file mode 100644
index 0000000000000000000000000000000000000000..d9b6b1d5ca515ce61c137f4c5328194d0e43661a
GIT binary patch
literal 284637
zcmY&<WmFtN*Cr5TaCdhP?(VL^U4y&31_*A!T_d==!{8cpg1ZjxFw6Vxo_BYD^y$;p
zx2n3jtFJuIz2DVU<xr7`kf5NTP!;5*HKCwjyrH0=MF8+0Em72L6CWo5PZ>SWFD_p_
zeazjhp;XL0T^(IK9qlYAysh0m>|C6=KJ$I%Vxh40^mO$QWMgysufS&)cN?~8ajBz^
zK@eT#^*x}V(9QpIK>rl0w1a|%f>Mx{(DKbYdjlpp83tTLzsY8qm&Xbfj<Ru>Tq|bw
zV43yRnaPI+!2fgGbb?T(8hX#LZ$)*Yg|a}mSOa*i#yH~xykufC_v8w|d!)=Y{0ji*
zc#eZS2kmo@X*#h1JPpfI??nTL8bVoL;EkvY)<(dLRDdn|d9g@S^!+T^n$(L+P#UDS
zC+vS;{r48g4!!T<ze4|MCRYf6^8SB12G~O6*FM$#x5xhsCO1zOy8i!7AeMR$b@nmO
z|CM`BX+ki2_>cI1+mgyepgRA*CkUZYfuu*|d35>T#SUA9VeF7U7yO?d5xEZqcm7jx
z_-x77gPuzWg#R<v3*_*g^7@<}x>L8-<Z6t*@Kwk8|199066CWP1O$?Yyh?2PCA<C4
zA|Sj?g;bltSapUhcu=pKkTgho!zg>j?s+Hk`zhpv?NRN$7J}aM2m&7)$T`*4{{_(D
z_^TEtVwmjai!*;qzkFhwd8~w-eeuk=$z|)=?tN<PEn`r6f9k#4+%9}g+dRB=jMAiN
zl10L8wh4QHt`C<6gkfwUiiH=MM?XaO0$GE+H#v?)7<b#dKY=l?kYLE!zPe5C<eh>K
z>!@-u3ZN<mgB+?cG=VqF>3;)k4JkGzFi3jNMF<>Rr?)?(j1?6sxSDhrs-q8AUQ*uP
z42TOU3+P1#Ut(fiy9W=I5r8&V?nhFzm`KSDz4*8x=rcPS07+oOa^Y^FXVt^|^yB%%
z!uy-R;j=>k_@Us&buzH+G5_kUmKIj=mxh=E#jV}r3bl2P6ihiROqevZ#nFmy@F5BB
zeRm^#<_L9=v(qDK(huJukspwGe!;`Kmcfq1b^o!>Wz*~HTrtbeccfySEyKXdIvvX{
zOI<>59>>=3#gQ+{(@&*hPlvA;U<gO$p_;m1BJkl&_#({{;)_1LL0I|MWh0==gJQl>
zAW!Y##C6&teo!dRZL538&j<YU?035U{zgIqYGQ@NK^7qunY!fj^FHr~koS;2PeBQ!
zu(M^XK|Li*(KM#l`v?@mTA>Xg-H0<u-*}!gZ{m}R3OgdQ{KA`~xi1GdaA~XY!ntNd
zgHwyRn}K(&uX2!WNQMu7NSL}q1(#Y%So0PsKf2<8XAOg*iVjm3`{ebZ@;LQ1l?}*b
z)*H3FfP-yDAY0Vh(74g=^Xj6mXN|RJP`|}HA-!Dh)>xz1&e+Y2fT*Z7uVL~pl4XI8
zkI9)%oZ~!L0mv&1Y|w{e6E07~<gYWLz9j#%vmaU)kwl*&3u*qdE0qS0hvVdItJ~Y#
zty|Zx-+XHIH*~D*RS$(|h>$~(m5czeOz#=KhZtmEZy4$6>3t8|+k7YXhiRP};ZPQi
z(@RF(+Q?vKP^4gdj|*N!pje?puU?{ywxav;F<>zH<~i!H$+}v7!F5abhhEd9c@vb<
zp}81PVVXpgG@ABhZRnog1E6-^9_*{n`EIWcw*|LtUf=6hM}2#F3#gr20{Sj4WdFMC
zl`h+{3|ukwAeqTAeb>S&9yss@aK4;E&7&WZh=p`ia;&u1nnrP;i*H}{pcgU<11ZH7
zgTTB0t^+(iG#Jzw#V@Qnv5>O(<y6EW#mL@}JnXiXr5)u~gulQUk@y5Tx_yhkBNUrC
z88!Ni#?E%XmGX>Jf1&*n_6?JwpnaVFVjGr%W+>l*mg7A*0q#8*TR6-s$md`@VAmAw
zdXN`jfJX%EMtQND_`Gop>~<#voje>|Ji1tZVkayN=^6=zatA{PKZM=&#xe2)*=VX}
zrz5VM*H+dQ46ZN{9@KT{ekWSm2wFz;536)CtUA9GWuUwn4j-O#gYsTr6wKXHO2R?{
zEd{=g$!SgL>gj&LiK2SQ`iv6sP{eITcr;nsEN4Gl+USvZki-~Eh_5vBHDF@yEU(h0
z@cQ-D#^_%laF~bRw79F1#p7!uqnw<Fs;Vl#sOZ-oehb0ulrfU3!97CblzJ!HrA1#9
z7=(=!{~iW<5>nKe^X|U(ok8IK<*-_f<9mU?C3b~uGQB8+aWKn^_<f(qI9cYw{tFxZ
zPKbHz_if%G*T0w9<RNuFo$u2R7k=#EwBN#k=AJHZHoIwM;Gs<g(IlR?7cd~OOEqLs
z=w;<7I{?n-k^Nj{ryKqLfx#M5Q2Q=2{aAWFjoszaLt8WYQim_sU@JT-Ir2mYXFxok
z55al5PYX4+pC+vGIP*5#=<*w!$gDjgio+7vO3@6vajOej=_&{MC(+koI~iI!O?c+q
zoIMeWL56{Q-zxJIBR#vlSxUW0V6F7D%q>a8XI}_9u3p3@Z~uwZfbZTZFif`G0yc>Y
zefaSMkOQG`$QVU7gkJk;#H^%74l}7Un2nVjg4jYy8^63*;)PY#NdSgb7VmO@4i48V
zQaA;QJwQ)i2COtSErpV-08=glTu$+ypeML2n$cc%FE`K#`>w!);Aj=z(1dq9|1csr
zBGG>VA=PBt!(g-r;u><^D%WmyH8+WB%z?W;XJK9+TqFgv*F93#zai1E!suBX<aklZ
zSd{&%IqUbSQwO+9)ff%5KZ(Ln8>e^k<(OqMBphI%Y?Xnze4AG%`^aQs!IM^kGW#;U
z2lH=8l=tF2F@5{4c_WI^r~U40udtH*@UD(2qW_j(`m(`?hsNd9Z#^zG@5VVeR1Bt=
zeW(xFHKf4k?fn<=^`INBF+^B?7rT#NgiC@F18JK!+}DYck&bBl(9`BH@ctO=>h_fB
z=!2{1ggHP#1K>*)GJ(A(k*VOOr%&P+30Bpgw401-aaX?KnocaZIyw_@GHldm(=f95
z8Ga^HY?X*bsAwh6)*s7|#G@!!)c}Cq__ny~_Q~>a@MTv=xA5{*3}Twti2)utRWjR^
z8vRM4P(J9wh^a|c-q-o_A2{sUTsr(8Vgo}LD|I~vP-bkdBWd$T$`5=pT2hydzzLV&
zo8?zA$PFa@LkHJNc<R~M{mD2?BKHV>k7;cDLM9SCSqnKp?s=YisLCMj`jo%V?;x&M
z)`EjQHny7nWHI2#Va;2ZVwbQ)RLv#cphu#y2WMaxf2=ooC@~!%Q9(N`Sk=hyG8Pxl
zG9eyQusYK~@{5LS#i`}5m&`J+r%LpazE%-TM9=CHfu)9Gstq9-(A`f-8*a(B`=6V?
z|9D{VkRvbsg5ab|^{k-O7)j)iU)@rDb&J+rVny}O9M(UJv?HDXL`6orG{VN?JmJC=
z>n{zCin@m4C5@>zGot$oYQYlvpOdK&w8}I|O!S05uS8NEDXduZiG9+$C8>g3T`9Pc
zFwYI{2|VVYzwIyNc?Z0N^d=H=+%iO}%Vz@;p!CX#3=an&-_3sg&WZEUeYJtKhV<7$
zc4&27S|2}49SYo}#W&h;iN16L1Mi<a-sK>>5D~S4_w)RKAk!BLGaJV^`ZE)m6t2RV
z*=(3?0Rw)~_x55hvM}m2RO=Qgq=bi}AAMO-a%w1Qg}n!*YMfmUc?D|iUHp{{L<OQj
zQ^(omEOZn>|8Av|oUbg@;s_#70Zj?KrQEaOZ>W@Z>fQG|OX;xvx41LOlXV*i2{`1+
zR_5;OjVZne19D$gM<mb;LfBc+E39O{uGCvJ_bSY8?gq(=K*x7`kV5GfX2kkv48{{X
z{pJ1gMKijGD6C526G;}aeBXDo)Kn}+El6uJ44a6hik3Vkkn;2R&$oG(SG%kBV2yGc
z#O1GG6HVP1KnJ|H%h1W=TRUX{B4#f$cG_{&_9ixt8`{5PAIyWDqh=_pnTybPlwYY#
zj&!9{!vRTNcn;yDQ~C<sFL_P3PB<%XkDO4hVip`giUC5~TvpH&oVHgOhGDWX&R2`S
zm0{M&FtBLl;tn(Ro!k3=T>P%)fO#VRvC|XX@AAC}X-#tc{iJn__k^~LS|o*i-U`_(
zZJ1`Mq4hqY+XT`hCLwI0FP(Ls4?MrL#1c7SB5&2kw|PY5f^>_8wK#zkq?iYFJ_R+W
z>=na;aT$V>X>A<7u#$@D<D(ukj=YCYaff-qUQrLN1j|L~gZfEo<FN?koK55ElDni$
zDoS5;0W?m!b6z})&QnuUxb1ljUEL+lEfQCa65q8oi?yN1G^5_t1-t0W#C@YZ!lt-W
zSd~|pSdOJd;&br%GtckFvfds-@iioUzu#bO<C5Xtld!p}%IfER&Uog}D@9lpm3Q9$
z#IM@Gag?J#c<jO)ZqB~3bJW&jaPzzfS8Lj1*<1xh{Javukva`HS!&$i^~{%INXX2o
z;9$u_o%Yu{$YzCIr5V&PGm>5BUvpL}IK+_|bIkAvzK<E+$)aK#ZHT{V{P1Oc50BRH
z7Q8+4B@7~f>NqkOVjI+vGN(oHKKX+LQeT9A8_O9t|Mg1n>7-Djit#o{{)LiP;$Yrv
zTyIc)+<|5zB57ZR+mtV}gqq)C4#D8;!1`e&m=r2A+pMljx)hK;o{~<0y+T_l!Td9F
zP&PA7*`8B-mQd<TbC=&+Vp{1TK&+gFvuk#{lPZ~F!>}t((T}K+Tzw1e$1iftN3i9~
z0$N%kHa|a<*I-$kuQVVWoHEc}w=9sk*Jr%6A(T|FzFcQO`1v!FPHU7sak>{6=hQGR
zne`?u>X6OzETV8j6tF@&Vk?+?d0a_;g%EPd{FzPXtfls8A*AO3@)*+dU5>~M!KnLC
zx%a&&H}F3(mj~CnWzVB7M5t?KliD-T3_gDl6RN%$8c@FPtPJ$`o3g*dTUe)Bp^6+1
z_@2}gOh!x)-*$yWm=06^Y0&rTf@d=Xg32cPL^$l3INcz^>o_yS?+;G)B16_?w9;ac
zu2mPG_bnwHK|KK9t`hu1WvMwABPY)lh0MyYyyPf<_53hS|MHjFlYL9?rM8@p%4XF@
z2;vcm3ug7vPX?kmuJsplN*Dx0L=GPxvy6Gnb$|U-);F@q7YXCRAY$a6Ch`WMD(*EI
z%PiEm_EgyMRq12`*j62DZ9%9piDm1s;)J1w7VPEe5(*OG9t_p$_GXUQSvTMHpP;XA
zP0WJTAnOh($EeWJp*a%Ykr^<-rAf>Y;eoa8lXb<0pn4P5_Vyl&MicXD%B*oX{bH&(
ztGb$E`z3(&n4$!S2EnOo=jTuv2{XbU@ldq4-|O+LBmSh$6E;5OJz4ZS3R(*0yLr3E
zKg>3x3iXydsIVgoMN`+QNXGX4l~Xsb9wv&g6N=~2<NDXAf<(~27md-syx2y;wbc68
z3of!-WCE$d2u{rB7|~fKrTH4hp<}cp>WL`B$W<&gyiZo~8@_WAEa1V`Z6wv&@3#IU
z{9)$(Lu=hrG<s-GBW*Dj<;P=-4o2`6lW9L?RfzVN(^QTiTE~Wu_Bg*Z7|iQ{tUav%
z@i9N=Q<?bMXx)gkd)5s&MP{!qUGuK0VgBueJ#z7B4FOt+kz3I|R`~EL&EtJ5DL8ZA
zebBvK@)$WX6tWyb`GgOVq)gt^m#DzqFEit1&1S2ZJfv+Ljuf837%<N-{%!9KB3clr
zSfIv>qrMi7!x5Zii~n@;7TVo8V|4S*iv{l^dJ`c@z{4Y(tJ%Bio{37s3ai#DaU1+~
zblX3mC3Ktcc_BU#JH>cUuJGJ7f}}TukHLHeWu8W*qVg^H@;G}zmdunH_vpzLYc*s6
zd!^2G-n{<S(f;`-m`SYTUyKtdq+uJ}t>S)(-VXNdKl0z%Dl2VayM43mc^dqZ)XP%`
z7m3R5G-)oSTvbGHF^CXSe;XRN06`Y#9!3jxj5edF@3pK;yAZNH;pse2wGW865lU~e
z1D=FQpH#9Zm4YH*w@DQxwViz5PG~bTU9w5@-HmX32pZl{b=T)&(ZQxMdve<;(z=mW
zVp>>~lA$t4bcABMCz)<Vv)vUZn^=aKBc5jn#B%MXoyXc@gJ@7v5FfSIiw9#0l|h20
zgyK_}PRyk^f{U=dO?phl6kGuyy-i+TLA}AuFMC305`a1NA7sil7O2i?PZpQjU#GEE
zvrQLB(sDLi-sh|5E&RQM&!{mKKHoGnOPVE_L_(hc7u-Zm!~O~Lf+AMWwU8SV2U+{z
zqok%N;lVB}gs*4CE-x3WFK1{NxBb@i3>=@>dp(FXMmlK_5RvBXy(X^jb6S1{pineh
z2{OBu49CIm(ts-9okkTI1jF1?Taa(0n8W6&!kZFxl&(1+<C*hQtl_LV%30_q50O>X
zW_)jtFpU??63V?;RTeasQN4`ezAh7W)ReQtL3NZ5c^YeBoE0(nCQaj5XGf^7OwTt*
zbvPd>%YmAkRHU@e+l-B=Bh55bvsCD<{3J5=tpurOxD~B2fF=~S4oO;A<+z&m4a&t_
zDmEx1L}eL!GMHBJ@tkwmFoE%<HN@#mRF`*I9Xt02Imfq=t7qJ8J~4eO0x4BiZp38f
zq1_UZ8?1Y-Yj*)^uPpkBT_DrUh$ODJ6Fda$3(>&2M)&WSc)JfDztprqM@#9NHj9hy
zhs&sZIPg2^ZNI8$Rls>u$H=(K7IZP~7yL5_qYmWEr1J;Wt9wv?CeY+M&NOIxco_Gl
zR0-;R7b1T}{`vx^X-k$by3?p8deAsydbu-9T_7xQ@aQ={_Yq;M>n+cE<F>Z82EI%=
z)O>F3^1bH@zMI(G-8<vYVux09e<fcXk04EkdX7rq6cCD5^?Hb4gAa2p%wUrN6v%&N
z$6SR0PsWH4xbW_}(6eI%qNTCPlr(%P1EZq~&u(VMuiX^SOq&b+Gs7+d)UqyRZ}d`R
z-A0KbKTu;Hbvr!rPXxv9++Jmu%Zr1yjV`~B-J}Y$4FoBLjeCYTOY6xOa8Scg`A*7N
zPd-2m2?!A_1bJ*y@~a#y58?*l$H()i--{9_5{?X!LM6OB3qESWhR{)30aCxeYbYu;
zr8Cf?AC!SrswDw*HxQ(9m-`DFlCazrj3i_aP7cU1VK<O}ux72!WL;gCIg-0O84(x*
z$T|))!I7hRt;Uf*ZEx$^y8ELm?0=0JH!Liw%l?1@C77p__<T|M*h;&!Q@2^J#ejRD
zW1!3EyVFxI*mdAC(TgU|fEiqpT^{ZJODbSCqvcg1CjiD4f2U^hSX+zsSzy-7X$2*X
zgnH7Fk+yw)PZB=#GX<L>T^e>`QUQA|S$)@!w{jq{B9wmt&abt0+$Nzae17jOTmWj5
zcmy`<5+6TkL?2!?G!ur~m-aIC$iO^>D#rMiRd2?wG8NnuYU)1yZENzUWbBt4i`SR8
zP5f@;(W9v?{QQG8+j#$;j!n6FjyX2G2+hQjbdo|9Gwq1Tz*LXG2JzFxz~30zr8PcA
zn?_$+Nk`xgyf{-dMiBhIfQhQO`0*zz78F!#JihKn4ttK2WpO6;ICf6K+X(4Q&^lrF
z&_@ohpq(16D5WQkx6w=>Uy?BzMgk=C3>+95vhMmC*Q-j&ZJ<J@1(a><)}lH<l;viE
z`5x|p<EzI4;d_wEu?CjMX9Y}SCYsYlv{?#T6WHN%+a5Dr#IqVHoBlwTctr8P1Wlhy
zXUbw6OZ&6&K)hS5|51fAGyEcc;h#$y<g<Fk<Es=;6H=_3$h*u35+oqD1vIP6H;fXw
zN4<~1gx;HD=%a)jqeDZtcMX&qhk82tF28a!7Ttj$-Bw3uhb|2(O&LFu*G6OH)7->A
ze63A)Via{R*MGo%PwTN2oRfHjOb+!uyRTOQGHzJplp9&x>E(N43m&5Hdjc~Bq;8&7
zSyRZdT`0&J@?{+9)?w4aDve*8dn$hW<hh5F*`1nGf8TI{(R|hJAd|N7HZnS+w3U!%
zy>~5Dt49AELo`-!pl~a3+AlrH+DDKL-2G>UrU>JL5r`<AsL>4~f9ZK*h<P2QcVS0g
z&kwYl_r>^}Vsk~r;RKJy9RB?y&M1+8H{BYritQ1D2DcZpb)Uke`QR<<RAur7-YsT+
zV9=zJeHI*t$Bp#D!S2(gHy#aviXcY11pygRnG+Z=oFYCyGlpWAmhW6tE-maj?COO@
zdNwL?x)Raj*22;2$L7Ja{)hjPv(VQC3!Snud-+p=D-Kyi*1ZCU<}C{#Q01oP>5jfB
zCv7S)|9EQBsZ>c8Iam~YQBR~e#EALIMH$9f!d3>Qe4CKDWx?G5t}~X>Hf>l7lyayU
zYwD+8#vsqDTX@F6>Vi9l9rtTOIHycA!}&WEyIkA5e<%i_y|@w>w!+I>lbbX&_#|ro
zTZA3mw(;SU^bjhymm&AWb6ZWMW-+8Y;cGVy#g}9)Bw?g}#<sq^^Tv)+i7Owxh0W>&
z+`C}v%Qs)N_&kEXy!65RTS)nWw2BcOL@KV4%u?9+P+hP}h<M70Kl(`AimXzw3qx-o
zvxMRcZT7fS2x3xQPp=#$Vuhr+I0FW6Xy9>*d>%dke}T=96wJXWgSj+jhU9Uznt%H^
z`kRR`h+2k3i5$w<%o@Qj@bWR}=HfH|rXe*#?EM=HsBUD_#UI^=1J)O{pHHEkV6#w}
z!o)WjPcE($nlS&NG(5vj1*B88Ox&c?9nt_O!3V)niQ{i)$D?9y^C#J{f~>#EYUE{1
zPQb;jl2mhduH!kNSew-8Z_aajCC@Oy>R1~@6|$5vjN8(oS6o%t{q#n4TDGX(w{??V
zKG0gsxbs&5aW|=8hh@;R#g2}^=rv`q;L%Uo$fXaXY<pY_)6Z~S8$}nfxC|I2adz;q
z8e`An3b@P(GGJ?l#x1KG%z0zYeG^Cg#qDQOwDS~M$8gNS-vXlTT#sGSI%-%c6c>d{
zTF=*m2^V{)JK726%z;O2E9W!HLoK~Q35mk}2WGSM<21OjV<TX8QOJm<pu!AUn5Z*%
zanB;)ZeIUe<?OwJ*M;0u9Mg0NNJh#I8{ad7e))bL5xTzl{w(x90eMB%X+=#VWKkuJ
zlNtMJ?fcB)R6r>>Bp|o$$&&d2-Sv!k^3W3xUmb_<rfs2AHO7}@?F0g^t~lr``!$+>
zx0tag+R$ttV*@eDLs2k_$bWFmDpo1a8d>*3ccK6DBgcOT$mu=Qwvd8{r()L~dQnz3
zv;e{P(LXz+9w>&8Q=>*w`tF&Srp<LHf!Rd@08{`;P+Z@wFC_&>6cc}o(CI+Meh}MG
z6^EWpvak1D)czE%u5PBIj*>Imrfa}mhVb$v{+9;j>mx1dH`Yv-+M^Qzxj3M*O;Vem
zPrtWrX#->0DhIGKDO)*dP}T`mpNymnj))DQ+M4nt$)9&lA&W>iVmCScl%+#=_pGP4
zx%e9WEsG$lO;wIq+6+0rd9L+qR~SuSKD;do!9t+Cs2PVm^lDPM%}Tm|vcGSy=Iukq
z#ezUI4%R;HQgr=*LYF3ht!`ts85cB#s~v>ev~5BnU)gu)ChN^%z-KZO%|l~++4tom
z=I6I;>j<1!+!rbcxea$0ltTK38;K(x5GQea<NHctk(EU`oPr?&67uE%xD3%ae#)GO
zJdOr`tM#|K&US}>1*&d#7H%4x_O~5DZ$OF9cM%o&!JvT!Fy8_RlSMy6FEbQA7tw*Q
zmUf-gU4^Nj0*X^Ok?o2im7xnYrprH*hH+yOoV=r1G*wa5v28cFk<r0Z$F9<F!Pp}b
zr-C7OYQpSJbM=l@c$=HGFS}w{HP~%L(7oh@Gg6z=FThRmE}_wgi0^S2;nPi8cNrbO
zN|bHXbpXNkDhsn-bGlQ4ocY@xDrwDI)M>1#UKF=~8qpTKY|9zq@)}r1xl};BdgRI;
z303;l8G!B566_Y6e7kh5FboYMFzFKPLG|yoqe3|Sj8g3=Tw@X&!cU<v04>dw)8x{5
zq39_K44{>2ikdo=6o7QzeXPz0Zir~FrjgS0t1%z9iX4cr3{P1a5Ie`DXW@wl6NUj(
zD?Wx&ZQ-g;Ly>A);E~Z%Z3oJ!akvLo>uwxLtao#jAXp5dkW6!m_Fsmi0;6w^dX`04
z{(%}_;P^VCUBzoXk>F|5+vu@1J$O`<njKcEnX5$7|MWuyCU9?d^GX{L1z!+Q3La15
zls3AI7MjCzbCDk2SQFuGxI><2e8evYvG?>`^jL}CuRo~47nhoB6)sg(4>FPbt3P6o
zd8Y5zHhBePlv+_Z0t?3+HZ}$89i_h_GBogHxF6)zsn5kq9@8J2Hevm5EMZJFRv!C}
z{P}GoOh=$%!`bC?T>lC0xCO&vq%rzzw%YO^78Bk9mFpg!esqCv{$Aup4w`?uOg7<q
z<imuG_TjgIl3UPnY&9%PSB?#D`PJ6uJIJ>gIDERixuAs$1j+?I&wUBLi%ZH<hB@{3
z_6}d7T$cs$dKynSw_19=;8xcYoZ1QkTkJ=#87pm)p7t7N{H}JsI9AU79x}k!6gfHq
zoJrkgB+*S5G%Nk`^pD<P*FljAW`h#V!+yvP&=jM?K3hS_9btp+>kqMkknDPF@`;WO
z1=yfTvisB(I@9qe+;+OgvVWo-m*aywJzir4EwWD4pfNICZaK-LB@XOjnqZQ4jeMbo
z0gcjOzg|1LxcJ>1p!VL}px*2*5J+zm{4JCyRY98G?ozD<R*e@E26g>HW=s)D!)mfP
zBpoO({E3?M{Pg}B3t>NxRt^{W?3dITNim&ILtJ(g^gRa3R63ccWYHMRp5*1rcq!^Q
zule)Rm~drc3qqI7fe|&g27U`k_O0Cs`srI3->2;#V>YgGu(ZctsN6JCe)SdBZ2s=G
z?)a&jxP2jdH~3^-Jk{dn(_3DFLsyDWiq)Mw;1>Ny75{F#AN~OxFK?iA<GQPd$CQ7!
zq=JHm)nAopy9jc_QOlF#_);h0bWN78K;;RPW1+Q*QZCdb&_+4c?e|aHSyOb%hXOK;
zR|Bq{zZoVNvA&<9co}o#XY92no{pz|;LdDW>0fd>P<z~+l$M*Ds%01%=xG~Q(`%~j
za?xBwWOU3t@Ts`NjBNPP^X+CezgWQydydl-rv762SAqmhGs0iiB)|VcbMiqcs|<U0
zo9ps5ImvO%*$PN0exP$z3mk8_Rv+*;N>_YaD3mYt&zlW%<+%EfcrxG}oeA!tX^)vN
zV^fSX;J$`krva!_{C<OJxB|weX1UsUItj9ZFNHaZzTrQ${hhJiVbQr}uA;oeR_gvu
z&N}=!W#<^^Q4r!SDKHh{Uf37S=^b*DTB9wmVIe5Du>V)(8waUMexL;`W0Iy#kyHbB
z)V}rvXqnlhCooREfIf-bEwG{Jk6=bCE}yPY2Ylox1|X+`GMhI|+>=8tD^58(eK54_
z+bUN^!G^|s`5exRff>K30(IsE*&O*qD6zjpTW=WcmkX#3P0`$-sEb7K8hX?A3Qf#N
z+QVh1*u9|b$$XMB$^5S4C%uz%R(yPy5295Hbf!01$*nTHxDV&mT<Snv(Z2Uo3heT|
z%2N}2iV<E;Peb&>3q-&ySa+6N*8lJa>il}kPwm$_Dt<eh{62VY*q%CN`+y{t17EKW
zUr%fx>BqOQyA(5(k6A=Of(W;{C*IM)lolF)-f-CKz?Ht>Nwz~#bv}5*@2rGL3qhj3
zR0Dsm{5B9yE^Z(l;CHJ$!-tIkO`|lfhS^X_-)J{hl}2k_R<{;9Y^-A~7ZaqBZpsoK
zIyC_d=k0>@WO)*mFCK1J8Mz>b=>lOH8|#($mo7``3@5Q4z?BEjfAn^wm6n>);3IMI
zRDdd@wHZRb)QE>5anmb~Ip4+R!yQCS`ZUb?EM_Af^6qlBsS8_+XvI)HtN(M{(s=d=
zSaA?WeuHvBa-7ox`QAYvk<HHoX9eIc_}qoZDLDS^##d;RhEdDP2OgvFAnp?jr=Ft!
zn_1}doALLT!$&o{fZTHOy;D7C^21S&jR;T7m_&V4@;5?kaa14TzMB~B<0Q13v)tO=
zo%K$i55?wL@xhsfPr;wE7=udEN1nE7tIBrfo&;>@ld@Ft7<jhVbCsNHGR}S@?HcCs
zWz|S9Wlg$YXG_}-*625^nat(n4G&Pu$S$|oGX6A>a$u2+w5PgR4X<g|%Ki9O73Y#f
z)U>#bW06#BZqtoDs~U?uk7$;HYZb%X8Ol~tpH(bV7&x_ytW+84&H4mCtx?N-yne~B
zC92R7VJV8MCG*bM-)oT`LUF79O?7*IKLb78Jy8TXgBiDEzvX2u8Ju)jcel&LB=b)N
zMB|57IIJugl7Xkn=0<{Mgu{Z$Un^U#vKQLX-dezK^CHU^VyYh8L}bri3^9)0YlE;V
zmio0V|2A0KyVqg6XspcsMnsJsm%^!uiEJN2_5{=l3)OMRVHZVFR=WK>#%My20G!6N
z-h_`RjFM<gk1M`{=@>D#Xz5gPOy9UZNROy}_ld8>+;PT>ms7efPuF~#+bNTWTnA~?
z@f^jb{?%mgKQ`_LSrOoB5bXQGjNj0JlsIR_PJm$gek~3f&GV@?w^Oum#mID&2dK^=
z`S@=$=x~DZ%qXuIQhb0TnyycvYfB=9y5q_Owrp1InPz}_zhQ?nFw^uMbu;jdI6bOm
zb#96H{eW##b6H4LEm1&BX!~{e!*B}j?j6OH?!rmZHrgsJ5RJQsKdA3vA2eFVmJ~U8
zXMF`O-q9{E^ZD9fZlu^?S&}Y;O|CB%9|E6rP2K#jBk>CZ#}5l3Fxfwfu6vTI-S~;e
zc~PM&#~1~}1?$<0(D@$G&1KQiyJt|yV7~+2O%N!qg0?II(Gj=KjD;9csh2nqtDr_G
zq<_Gd_X~ZSCEDaUb5sIV+2zn`HZdGyey-=kqudSr?wB93fy?Sunnb7bkwt@)v*hX(
zG&VJbTHz4ThvCfE!5i<0u-h)I+Vm<~{3m@4SyVBYO;T}>p^v`aTcKDph|_jiZejy2
z2I6||Dk{%$-7duM4DotoB}~_1r1dM{$X2%S^kZ)=&$B;PIt~drT!j}ud<}Xd38O#?
zgD+M<oDYJ1GiRZlX0FQQC)T^h$S0k_1XDhq$clNRp_;s<z1u>d`>O;_COC2_2}%gC
zABi4blDBL^63BIaC@6ou>{nb9;J`gDo}|cu_4!Zg)wy#7enHphD00FmPCkz<uHHgT
zT!X1XnI^PtJ8Nnp6<d}vJ-=}0nuacFi-%P^K%=N)|KOg8viMZ+`V|(;t+>yB$Z&&l
zsfURKgCnE*Mc4Q6XRF~_MO-@)k0w%H{aFX;H2l6CvW$apg_mbSxs%W}3%bUy&(B_+
zTQ|R~H6Mc*Sj<rJUFvI4$3yTnX~v%7g(qHy<_r*-Y<~VG-5M(sk4a_8O+FOR*SLS*
zFXmLz13Nj`ebBOZtXN4pSf`#%L#zB^z&_QB1UoT$*<N(e3yZtuH~Jqy9CpnUFw?e}
zOx~?oA_a0;I%defZgNW44TX-i2K{qAR=&U55m-dolLwOQg=)U9kC;h?d3F3b5i@OU
z<QwSC=rG=k&e2#x<C5kH(30T>rv}1Kz8L100~Xy<yH|fHzP9P{0+{XWp;a}OqwX5E
zPisGa5IH-@W%I&8-iq0{!XeMFfOq7jP7_1qdfim@F_^BTclt3LB&j2hqg4~we|M{w
z^_t=<@{*CINnbfI$BMGxItwjdVC+!N=D{>=X@tDD3&HF2<S=l9n<==rEf3nUf!?q#
zDg6Q%O^kv}bvZB&BpWRNVbFT)pFgc1K*px4-&NG&NlwY=inESyr==hxznDlRi@WIC
z5n|jhW}}@#fkd}4Wz)bPv-xaQ8!P>vg4>Q5TA*{hWx_5V-#?=bshqiHLeW?`fhMUF
z@gGcsv&W#n-{Aftvq|@B)FFA$c$y74myjsx^u~0G;=BM@5@xY!W4D6nUc9{(8bzwN
zS3Q4oraFDZ?&yY+T?*g^6l)N}Bh)&+Qa(|+vO@_Oj|$oP;)$M|u4z$v<J8N+$-gGJ
zJkRa$dmS|<YOex=v$Yc^6N`X|zJ6l|gFtLgGgC7hIZEfdIMU4B9nZu*Yv+1mY1)-6
zL8S9ZtyHn^Bg(^5bE1Bar>Yp%jwC!fCB2bnoyQid$psgGgx$YQIKL3!eb1F|;m+db
zL_kgcO7r0EH-O(@Y>7DIs&0B1rd$eCNZ!%I2Vo?%Y%vQLCTg)5*E66jOh75gsns<g
zl+awz@~y(FDFL+ts3DUqv9IE2(XlM?ex{N(7(i;b)nDoJx^dM1)_-*c?qBwjx5N_N
z{mdn2wiTUDe^s!n0zQXm5IX1N<mNhfcan~H@*@Z`XT_xc8%ebr*vg|U@3B9DU+6f3
z(J0cA_!7;8)?fbflgk{3S+$*gl><~XPsUT1bS$aCw>NHG82SiJ=W5Csub2X7LvN6Z
z-efj<+4sgQUK(=zp`z*|?3IZ4t3la6ksiG4cAW5cWk=}X@WR|WN^I%$R?aZ1Vv9=|
zc5FWFI1aiBVaz?55hDjW83hY9V-`k&aAx797Ej{@A3R#v7?{&>*Laed26NUjO5A_i
zW*2E((X4P}x14e)CsH#o<WQZiXK!1d8IeR>LKx`LII;)~d5RoJJDn7-zV_P$JY?a^
zpm2}b5(B3?S)D95UIHJQ(W`{b!_d@<GNg!Lp$Ray2PE+Hg-QE!dIsYN6;(=H7ycT~
z_%Ak?Xm<rP?tyAQ>$AZD<d%&%b8%i@8<UPsY;gQA>BZAs!qvk8Nw`!UpK0*u>yb*Q
z^83wSgTHYCI@Q;>?b;bNna~>WDK+Dvs78nvj7L=O5-Sl=;`zEf&)$75deH3}DvH8B
z?Sa!E0Fj?6hF%>N=^QDn1sxl%K~EXIk8vTx!^6y;@bq$ART}c!epfd$qfOK^xb4$f
zs`TcAO{^239IBbI>-aH?=|FGk0{nj6BE;#!CI9$(J?xV3run9P>&6<}95gWOk{<6J
zQNDFvVL|=6Hiy~*E1Rj4zSC#FyA$RD(U(Yk@*q`#>|HoL;t|6W_DyM=OLJk(!q>;w
zph66SMoWc3l89)jMt{z?i>Z<nasJz1uZ*G8b(lV3Y9Y1muqRYUK1;K_#NM66@vG2y
zzucQ9BWC${O!%egFnR!G`Hqrn_<`~6aW{NS*}tgdLEJ`0UDll-3**rmHUhYm$9|Fr
z(S@Pyfat=z+R|fOlHhc74)dItH88mG6UOU<QvlR^0SC!}OrXlY$G<%xdmBEoea`=y
zmMCEAkgZE%pd}C-f*|=<>m!?`MOp9pm&S*de{1LyE&th4Q}5&t7$)!$I$5qdlgMmv
zGTJAqGU3U=yF92Hg#`0uC|F8+Q$&E904FiyGV)<=K4%wce#hVqHuPsyE~lYiY~e{?
zQ&*{|#P36m1uN^LI)3SYc3HWQRAkG?oeWe+v3@k#GLAvx2@0~AaBm+XLN`hI!#Lp3
z0*dXhvsH<EAge}PG>x}uT)m?}5<kVGWv)WPV-g#yU%EpKEYF*7K~>S#&^5ef0eYe>
zM0M9XuqrAxRF!2Z*H=r}B`qOI$S%f^MZ);n96Lte#~Rr+Va5D>^~x_k|Gm*WBcPUY
zh~xKtsUN|GGdG2e&zEgQn{?4z&V;Ecbx)8wAeD=;5<tnmRJXqPmi2&?!{uo#waZGm
z!i>RWFm@$vqh`&Yr3y~_1O8I+uvGgH^ds%+`F;ca>Ad8dtt0B$3kLZ_JRLopo19KO
z?m7*#k(CN)q%NUvnvlHOP6Jw+qI`r1=xw|Zv^lTC3WAuU&VYq40ze-GG3l{|tL=|c
z?e91|;W_=0ge7jm{fwJ_5zVZM`wZ1?wkK<iUF#Y+t$}Zg4jbL;;l^9_j!O<7IN;8~
zb4EvvJ_#GlK1rN1LKBI7n_aOiqi{@|&+MK*aQJ}$ZG<(uQZs)x-J3#Xv>*Xf3}q6l
zj<`4E$boNze}rPIPG1P;{hLInf&fed%Gb2NsW%}RcJUldiy1y_y*o4SXQtP06KvVa
z+}NCXX~H4yBV$*HX~(btxUuH+zh3M=bT=#$C>L_%lE|qeQTA=}fI|INB^rt;))HJb
zr#a`|N)D{@z*pN5D_--~KW%MoeaI4^NOwNT==k%JMy<u1+&`?D_-uYpZ_9x%i<{RI
zrf-x<1wrA{^cpWs3_tGRspi;Xk|UVzG)_Q+-faRT!LlwnHZDdZA@rL+ADoN^zjAdw
zM5wl$k^A@-7Sydq7hyhsXOebxcg`X0Tu)8zFe7Rb<B*Y0nkUS6z)<93B`<C=YYzN&
zev)iccYe<m`B*mLTXouIyrF*X?P)BRjoYs~1mxL~`8%@7rVI?(T+c#i4JSj8Ht4b;
z+z}@46z{6-4PlwtRdcd5Ls0Mw3PF~@@F@&w%Y3lCx-!Tfl^oD5AFRM)jTOoLaRADS
zKQa7gfJXRJ!$Xv=1~+4AiQvzb2{k)Fq~AOBclct<?z)N$sN2FoZ(7Z8=`>H7ZI;PK
zGyT(b(lWekyh$qLDtmLmbmK)toS_@NM`jurxhPf!fb*ljn2d(jqNCNjo$F!b<Tp^Z
zP;KI<L%tCH{8y;jhY~R9>6vfVXsnTroj!f^<j=?$8(uQ1SIt}HZJ5(x?Zn)~GM8PE
zXJy5Z!$h;;VyL!2bWKO&*~-i5evP+Gls=9jUhq;txu9sVMyi{ss|Dhzs2RH<q2*Eh
zmgl`BULz$clqX}2MK8QWv{oTueoWOjv`}d~)(}6&jd1EVZRRLtW1r`w!H_=xQ+Ri-
z)P^My-?xYhPbAC-Unz35-(iV(#&MxoD`MK?l0AP?Tzo-`MMz3po5MOoifJ}`Xe4rY
z#B<Lt28wlJr@OQVp<{zj=ps$ZB1D94<ifR|#K0$2z$=}Dev(;0aOUWZJoU=5=&-8P
zWZ5x;xFwg#icK6Y6OpZc$s-^@Y268ft@wuZe97ZQZ2f(!k-X)@E8K`Su+{v^aYojZ
zW9B$T(cVz^<SP|R(Y&-{?TO?azFx7LTWjj(_7nx~?JQa|fW6s@D<{&K_}HoVAJHsX
zyN6Z9pzMx#?3T$SU{W5FJ280k^qZL)<Lgl)ZS_(TAD+di7nZ?Cd?jT@XK~T~LS>#d
z9`N|~fQ#~g00TJ6uMCh2$yUa$W7`sT0%<PehTPRc^28vS5JJ9XaPoZWajIkR&|Jyh
z)f1ziP7#;Nf>)I;aSI6bLsu<(hHl;N!D=XO^t5mJH(Rvmu@}xT#y0dWeQiwOfss<#
z-ZK-<R=qNr^TbUeE4;d&>SPaAtlOxb%O3FjAkvY!d;fagrRFM4)T(@-j3L0+TTC{S
zz{%rdfg*)=XzMC!!V-Z8zl6ehG-!D}aCi{?+22|zCC9-a)@(Ry?#f8mX&-%%(B>(1
z*TdAkG##&X;tgX&)hA1X6Ms2lr=2MM<X_aWJFtf=E#1Rwtn~D1Pp>v1%Ek&kuRFba
z=YL!-MYD3M30FF<uodBQr=q63rIwqOQ^gWb1EBh$DnbC02@=NsJtB=>f=lfiPi+Xi
z-;X;gK^IU%n=o}kj$*vd(GMPv-Q@NodU<?5kGNU2@k-1~$58OEUxx!3c9G|5?=V08
z(8D`qSrQanrpyaHqFeA5>Pq9z2I9m%SxJ5g;62SaJtf`j{jFDPc!|5<K6YMh&MP9T
zG+dR(o;_1C)=pmZSeL>|T_&$Ev|=JA2^MO{pCaK~-Fn{vTZ@9b$OK9Q;9Uh0&7>pK
zF=Mt2D5TfYwb3&ceDZugAX)EDL>)#<7`DjB3OmW5#Q`qHSasYIRBejz9Ton}PoOaD
zsH#yZ@%UC0L}6Ggl?tV$30C9fF!nQEA!UcrxI6{@I{l3i60iQv#nqhCKS~`HvPg0u
zg^yZ@Y?O-E28I;L?e5ZZZ@{MvG|B;!6zd$}8F^>{MY!lV;Tgub_!>KQuee55-@&rU
zu{ew{p?nr!&*$_&%`I92@QUN<)&e6e)3Wj*A)cCcGoaqOvznvg9G8j_cABuGr9TVJ
z0aa0nvsNltrdo2ki%QB4_ne)vlI36PCAhJR<YMY0bz`(5z{)2ikHlCTsykkZb9}RV
z8r@MyG~Ac!C)Y@Wfs0<P_g`%;1`@jyBwJ9%61+p)9fjlf&fMBJ1P}sDC)Bi{4ade-
zSyGB4mc{R=t#a&GK0^{oJrvXoME-=Eg7uo+Q?^ZtAJ(0A9Qi2OGu<*sGR-$eMm74*
zEln*_a=&@&c(IO2P4xUjR+DwpM6)ASdN(iEz)6R;Mv|~5C+)tHRP;1&u5~2SY~TJd
z<I~FK^z&n0g|bqs<J1*4J+k+%+kT2iGzv)2!$L+8GRiTcdLF02*b2H0tNVMGxb8fh
z+55y}8vH~&u4&m3Td_gWwSVv&n^yXfR^k`0lT$US!BjeljJH4tI{!%Lwj~W@u2^8)
z4~!c3pj#kpT+JIRD?61OadVIglC&gRI(0bdC_X20bg(=5@UCStXm7eJ>)@O~k@B?_
zvc7Q(Im-B+=cu~)4&z^^{w$l`tm}q+)SJu|Uas2oivNZgRlpn#H)BHY8>OQ@??1!s
zcD)(8V~%n=Meug5h%d6=B7ctjL7j6dnt^Ju4@)U;{w5idHRN4LrId3kdYX!Y5g~@X
zvHL9!ZI)B(Gdru(m$^mCVn+Lc+dZ4W+gG2}#nwQ`n<)$-dC)!Zbs^*)-0+80C17}e
zSSY9Ff{=jHSWCs48%{nl!d|z2{o=#3!?D$@NGJJ}g-1pANle|}0v|l}wo?5zvTI~y
zBi-ow8;@F2ftM|u)8<=}0Kb)dBC*u+;ByA`<;#PE<yYS7^X`nHeui)J#fV>DO;+G!
z(enjEq51?Dw)5f-a#+6qX4m~D*7nZeXLrp7%8iS#2*V|?#H@iUz7z)$RY8LmebY(<
z2c4JeOq%~~c>gmpT*^%bx|O491`UIQZ&;Ja@^2omEE8>{I4%95_@I6_wP3}$tyDcw
z{+Ng%z0!>G<dxRP<1T*mKWgP@AwRx#Q%NG-lKZqT^~4;FuK&_SkZ)PH%MZ5g{11by
z4GwoPN0GISA8gNxY-f;n;H51qRIXT2EwWPjqM!aZT_#a0d^$c*wbZd*cJ^;ZMZNs-
zm{r^{$}~|^WPXyUgJRSc3}|*qsOor@6NZU$RuJ}|6$cED)@6gQFib*vQS-?s)_H#|
zK`7APr^H6zLTVg_KNgpAuC4l$Y62#CR9xS13#JWEmgG!(>RfQGA3GZ}e%GuzE}eM`
zT?-FNZ=D7uM>{P?2^V9oFyoH@`60t#=`AKCT9w`DVL!PeEpTme+P`tu-ny<m6!pYC
z5+R9VLGf?R*ZYD)$?ccimW!}4sD8uomp87lrL2c#*79@|w-x2580jR!m%^2gJXl0K
z%Tz6dd02{!2u`L95=4V8j1==hbLlt?)C@F657+*=U_VO>pj{uemFvZWiJ9Zb@M6>-
z5UqA2ZJ{qL(l*chpA~4dY|g(xXnBgy`Ue5U*qB!Z3f{MsLzJt00r$~QostHC<dfMe
zQ5c>S*!NMeOguzY%pM#NM4n_C5EhsH{(>rnb_jixhuHN4ML}**8A<idxO4ktN2urR
zp06vwN2<VqET6nFiyWKE8dc@X&B=?4c;DAArP?E_tAv%6l_aF39EHO#>6D69EK1|U
zm^L;xwxsstWchjXUZkgYIs5Ko&)(l#qa>~WdQI53%PI_8h*qAWSPKjb&HPz*Y1`mY
zZb;SFpopwIx6gUbnD%6-wsUb9zb3F_-3+?JXQ{}!0g@Nr!ca#1S`8pnX`&lZl8X*V
zQC>9v<T-tNI)h1)Ts|9cAWv!CJK%qWf$x0eu^RD1$_B)-e*x{H0_o925-8zBlZXH-
zafNLfj&c;yH}Fq?SgT^BnB$`pq~&ILbbWzcbVx!-G~~VaMQr>2-tZ&yZ2^+=QIz3O
zAp9mG`h2VQQ)MvI^fkodwqD{VuMTdE$KPn-bOZk`Fx4)|ta?~K8fAhAEY#I1l1!lN
z<^%154vVFgu&gLrQrx*qAP|jBRX&Z1%j-(hLUt_igwcN(F&NM=?MpL?u7v1nN?beE
zq=C{Gaw~sT-MuU4;1fa$^pHDdr8;uuEt_>!lr8)TsQ%K_N(C=hE`L%M7-g#i#vS?%
zWruw|z3XUJtZWHGOq|UO@>_#PA+7lBx}Gsrv42u=@Z4hA9og{-%xbKAbF!$^^ZM*i
z(Ca%5Bo7+c^@v_(bax9;?tTzKyD6O+$zPJA5X4`$BQT+ao1aJ?5<D;JjuL&8M1^J=
z`;BFK$FgM{Vr9Yc+`YPrybE=ocgO8Dad$`i=PtFPMv9z)&|`V*CrjsC7b@rB(NQF4
zp{6eCVvEECNpbj&wIe>Yf@KtZ<<FL=^`%*Ex<J{urBJUu#o48duD^TP6{DOvYhhf_
zBXSwtqweEXG!t>Q8K~A>F-Sf0w+~sr)Q=hedZUUC<=aVNc^^vF^tfYPM+;6k3fj~I
ziY{#36e|?-Zat?dl%+CSt^5*Gkua&Ff1v2(bnyK9?4n<jF;|t7z6Mp0fM`9{(So{L
zLZh$4o<GeH>kd_9X<e>MbR<`&+w!8Nd(~JlMmEzyq4ePSTch(&Okz;(!;gv{84>;y
zP&cP+2e5$VGKk7|mvzbaxABC1bMYEa)-(piO3y$hAWEEKoSdcLvgbZ5mEFq`ig-HO
zNR$%OQqbz(oMpneENiYtWdhGXXlXzCwvLYNXZx|Qjg~zC{f_SM=oeqeAGZ^C__>%|
zu%o@mXHkD|6pYYybuI)&15Cz$YLl@$Mdw2Q!@Yi)zT4@>cI<7nu4+#XhjSd&N+O|x
zz2bLqrK8DgUNTdevJV<wHhn`4dCCkK`LkTRdX558iM|Zq9nMP<^`9=o0HFAhF$6FU
zPzebLAiiM*-9=(3SAMB#X3=HV(VqH153Tjt3)$3}9-sF8MxIs)ZGdtsEnodQhdNP{
z$Z*fRaXXFWdyezAJHfqDr+L66wN>Y3M_z}f4CP9$stJw?<kdTwD9r+3(#%i8{}AjF
z3xs51PwDO@>P&J-gpw;S@0pmwW#7m~&J%llpng(L6&ElZ9a<4#61dO|(xpDzv6s@D
zmT+(MBn2v?m_Y4n(8W`kqDRmN!sLjy2l;n-5rDqM2v}ZDf5F3kf82chsEmSSEmV5E
zUp_!$S~o*xEWAk~n|@_Yec-l<V+~u&THgZo83*$*dLW%n#NH{*KNN?d68pj<k!;Ag
zeh0i*Rsf}Dxsvr>G^T!qEL<^mD?A1^9Z-R=c|9IVe}x7sDs9=n`iJ5ViYQ7eBlz%{
z?i*V>Z4OVz6+sCt%=31dX)=2epzpdanf<JnOnX8cP;!Oe9whQnwQvfvl5<~{7aO5z
zY^rql1Y(r==}Z|PzNFbo`p0|fho_^xUJbFcZsJI|8|I!8I?+e5;P3-m?_rvn*#2r9
zezE9rTW?v+uaHz!5vNWB;wo}`-L^T!rVSbF-Z&;@Gq7}%KWqrsY9kRQA(vxOS>EYj
zpyHdk6cfQOnY>On9NFM6fVzCk*469v)=hP=j^4sfzV(8B(~@!}C8-Q3DzDmW(M&Mz
z7ird(|E}?6!VN(OT{?I`$UVQ&!JrC1OQ%j>*KrBn5&$(R__bVGaB#jlJ3UqTM?Y9o
zyn;ur5pF=i2pI?jl8};rQLVrkj$CzA&A<(7j(f|n7R<?qGuXpfrGdXeyk@cNZbzgY
z&7&^d!(he`aL881Ct+1(hz(b^%ZsRRz^EAUQ={a$;jA3(Tg|Ddp0y^Fofc>T@w845
zAAFU~_^=ug)Q<r|hU$DgO9pglX3(kb1IJeaajW6V_%w8aV_mns;IxN^B#x1|Gm&c`
z`RMy#kT{Y6e`h@J!HeEu@aIm9ysTKCXfOW9kl6Qva2O`4dUfrObgGOFUdkW%93D|h
zzA;A@+{#{35fJf!p$`YTjjBsr{+nj_bfo+LqUfx{ntr=5j)YRuf=GAA=tjCmcS?76
zBdx^9A01LdkRF`^(j_3}=o(0O!~1>z-mYuAuI>3g=bZb#KZmiR+vc}t>>IH3-#aAf
z2;qz47?IWSsAQcE$d4ZkaMfs7dFf0H26KZ#2ku|5pT4g~;u~dnOKq+ktT#{UUUg&@
z7K+|9Ega+jl_bJ%eVMDbJ=MnZH_z=WdgZUo6*T=l*;?dYkMFH19Id^h!mu_o_~z*U
zPRqBoW?h!B)foW`l;8*7`2B9F_1l3d$&Kkde7Pd@D>fPLg7ghP!;D)+c<g<+a65B~
zhzzToscz@<vuMVcO=0`S<6~aQ0J59bYB3}?H`jpm4ba;H*KtXS)#wQ73HelkWN(`n
zCl?nmM?xq72ODs3o0#^{vSGSr3X<pNoq^d_yYDi_{Pcs*m$(8Rx9O^y6iLq|(R>;1
z-Q~W}-aAkqAKjz=2_Jragi`e|zJ1daQirGqF(f8I3g>Y&|A-qL@M{S!hlwfF3f~Jw
z5;`GJ@kR?sXRnOvkJwiPgth~*p!McoF)HOwTRe-1tF=BbS<wgOq_VszNb;ddEU9d6
zG$s6;0j{u{d0w&GIt;^k7#=E5z~b}q_h%V8Q2h0W@y*mEI+y~rksP^=3ME6bLe4_$
zwP)<Y`_N;vht;iqnM6p@n*W<@0p`4fRajK2+kO>Jk`JYoiM3v@Vcj&(aTTtE<*5Od
z^y@;<i4FG~Lwn;XSGy%W<9|55i6st5^s^4b{mlmK{^z$WNre(pc0fM@fx3dU374e^
zg5=C+I^VPWxXGDVRACoCE-c+k8d?!1G2ry^e+*Yqb#{<QxJWJ21m8WJ-h4qiq07VG
zS4%jm17R9C^*V!SxOmfXMy=r9G>&sY34Y7VoS^FdC3kI`6uPp$0PbpYpB2k$87HY9
zBF(vRH8{WZl#-5qk|-Gxfy=!hNyOmMRI&f3>KPZ%w5*vuDJ28@yah_wnTU4i>wHs+
zRh-NIG~m`T!!T?>iJ{dvvjXU4bpi=n-V#_#*u@cg8QP;tOLm=ggXWs=Yd0nE(q9^~
zP{a1-Q*DT?f*(Ou`fQ!B^fZD4b{Au``L#Ft`JKN#I>ij3*^^M~fKY0?p^FNtC$j+D
zvdpl*foo|Mu_p}kZ3_g~JqIgKCsfB6L5C8(c59+*OtZccr7ioFF}83^GLmzQ(R3H;
zL*1Y``+EQ~5Q6Bw_Nz0+rYK0ogDPX~d!;=3$eQR>xH?oR{ab#1?Yp;toa$)T<{#AW
zs5|XLT+MzySprEhy{)R9Vk)^Cy_;@{nf!o-r}cKSpGiDR=%(oIAuCSC++BZorE5MU
zf*Wmw!Tm!hIp!v9)1L`W$5g3<;G1KLgPpgzM;vk~(|_LO`n-!0`$ps}$RW6(vO0+c
zZrRzN^-9NNQO<!%9cU0UPI)n=Vc_L|$+yx!s^ip&wa<x}b9{%w62Tbe<<(CT_JFn)
zddHy(clY!JXRst-JiiM_@e1~gnokwW+zwDO6+HCipWRQmbDkW_<N^(Pcs_pF5AL<?
z7!*-R!E$w5tq-{)IN`7T)U%EId|G%eD*5pAv>c63w8mR3`K4lru?5KgQoTVUM@|t}
zaE(!AdzFH!kpSvSO#zGDuwcGIi8Iso#_9B-@tpEDn2o02pnU!}^^ylG4;%KqNagCe
zC98W4Z}wINjKF_<qLX_ShQ1KR6+Uj6tjQ^(Zv63~fMV@a@Q~69Rez1kp}MCL!hnlj
z+$OhmpLG4|PJ)|#3hZHG3I*5SKQz`0E`;3VLxfR%W2feN7%f)6+hgKMfFN%WLW?r5
zHBKIGX2lxjU)%h_mDT&>SF%PML=9@S`waTs45R^<4A=>R2fhEF5m>)`%-h-`1QP!K
z-q%@o!M$fA4#WD6kzwgXo_uS#dfBZHdciq2tnxvJW;1>A2ad~yPo1qdV82>nZ)qMZ
z)_YT4N<!LL<K{UL?wHY%s8VLWG>34VBA!SWQ}?J?17cOn_Kf!9I#UD0JSr8R8=A{e
zMRUf-anT&c+gUV5p0<QIJUQcCz(tlsCBAcb8MW^{TAJq`7zp`TET>kWE+iQQ7S4{5
z{TUDrQEt9L{doee(yWnT0kX^)nHC0BWx8aY==ev{aqDz0V9919meQH7m(~j@dDNaW
z4X<Wq<UK&N6-{!8G+bz-=E{M}J7DzDG?gCw{f3SgAs9?<NA+W2*g)HqUT-q#eRFAY
z((cC-#(JjlkR0%di==Gp<yy$vuMaYBg@<1)W@<)8><;y~bADchy4wFM6{D~6_TO@V
zJW#b&+_d0c)T?w;PTP%$kbI_~Wr3eHeL(eH{!_JI*C4~TvhX`d@)X~IuC3j~BM4>-
zTLNXLhlhXF42azYebHso%DhZ4D<$bXS$e;gi=9d+9zp{3wZ}oer|)&(6$N|3*PaZJ
zD3GEW!^>ZZKL$Ux_!J@ebCpKp?!P*97G(QZ**CpW&z{M)+ST>V=|kyxUSWP6yqc~x
z+A|fl(r90u1;%XT1Pyw6dy^A>K+r>dy3Asu8Bxb}QlSob((c~gzpNz?+uL8n#*lh1
z?9?j`Fs;~u%W3XB#?Nid)5D{8VuHhMRICmuEp@+<evChDB&x=#wXLl$AXblbc0Xdo
zdh}}UZO`-7npTE(=vn+H&tUKI{Uwl(slw{qx=Ug@gPU2191Q;Y;@1*ZoFze|I8Shu
zQ+7jw3X6Nm&o4K~f*|MREl|*!Nf+!J&}uQPDU%4A3`y8XN~nL>=6*AH_^FV^lsWH+
zdoTh^?RYZU9RK7kwEFgbtjFSp>X~X$fTT0%g22uwXC-HDg5gKtm@iBko#crwGAuS;
z!bvMar`Ne$_f-5}PW#%pEeuT=D+g-X<cvS5@Wz>B$4xOJS76u7LIs!kb(jo3C9`WQ
z^2oYQi2c0UQC;YX0pw92w>7pNZsi*E(|w`E^)dgHwu=O&J7P{rBu%aRw{-F0nDWOO
z%JXT4U&Z^5M2x&Px*L>5!l#E3Jqrr&*MDE<U{+!_(Ao*)@=8kV+%XlN!3v)u6c9<m
zA`(U8Oz1Q$B_mv4erJjQEN_$ivJlGq@CwgwQSt$MZ#Zh84)R+2=IPsUx6{|>9w{0e
znG6Fe<TMtsdA~m{1CCx~@wHgzU{G!bm{C|&KUUFzik(4=)%w!UDf*WvsR@C}u<U2p
zB5@8YcfcKLRo$P-O$uvasg(ItT5Wb>!2U~|Cn>UZ0T2NWFVhEtdo6(>{9{xNFLO31
zd)#0=X*eE3*-5s_fYutxcQ7!g87KUmp*)l~Dv#Eh+b_WCXJkTCVKn+HY^^0Elj`2a
zSHa30K_mn;Lw92DY><6fZ^s0ENVaVgXpg}vaEz$N^x|*49C!qHZ+3a{v^no1U(bA!
zmwf6s@pbq~n3an1U37Sk{y64lEt|=Tz#)rSlF#l$)B@F7T;+q82&DJ|h|y<SoR%Lg
z>hk-{$;Sx3j%@q6ka&zLUwaew(s;cg;ft?|RrFhWlTRogO~n84w|Ft+FRce*Dx|(}
z{E2kyopMH12ETr0%b~BX`D$Qpl&q|JBIO)JU~t`9f<Rz($eX}VaO3wiztyb+_Rqy0
zM|1b>E|cl!8*=GZ(KP4IuR<qph(}`+$w>Id96Ne*asX4#-{pqCRi`+%FoIER8uJ!Y
ztM(YN1u!G4wR|Q7tkOPtz6n?`LoURb6c}KJy(1$S1JmgWNtBvI|Ca*h9MMpHYnU<r
zwL<}Wgl_cw2L!q5tj0pGTd|9A#ux)qVemPq=K?lmwKDCuyt;}!L}VjnsJ3|I__rCS
z=$6#P)O6bCq;9amfc<ti_aqw(ef)P5I$86#INiDf4)pxx<p`iQo-HAbgShoPM--|<
z@CbM)Mwf&l<g=m(Zz*q+q*hFoLwYD6(+|79Y3eSzrfjEr@IL-w#*;54kFrqnKe>Ki
z_ngBi3u7s>Pb{kOGjF&5=R-~2XYAZ!)6%wkzH+0p!eH9bm(7tohvxPvY@+h7yH5cA
zV!^Y>rL-Z&aOqHh1|vGztd9^GW{$cYuMV_sHQMq0e5?+ZNA*~%LwoDjK;-y!4CTx(
zAO5`;3hI0RHRy5c_yAGAtXMl_bpdyHfUSq>6$}qPW#zs1`YhXE-@}uboH7DO5k6af
zgYZL*u03ey-krbtDaOpR&(^<u7Ih<v;)A052%?;XP@Rn+diqZuAN3#8B%cXGpW_R?
z8ik<4%75ESD%qlvX^cF1Fl?r-?`i5<wvcO{<oy<L&kP6S>e6Lc814$5$<z0KB9}ec
zJl@g3GDYanC{q^Y?C5lfuc7N?wlZ?4v#C3ABeH2M;k+>uwkWJMpx~-Oe5BXX-*n|Q
zSxYay6Z;M2X5WJm$unsP<O<}^4y?(Tqtf*TV1Mt>KD)x6^(B}c&H3##of<@qB_1QK
z*jgg{8sHWsj(Vm~()IsW<a3%tm!taM#Yc#iNS!Myo%N3|Qi@vNE=Y68PgK9-9#|<8
z;7%Io<x<SKMrtIJkSfH>9cq(?M4+X40#~W_HsLw#A(M1?Mg+9lp(AFQHE-77K>l2V
zI~#nuI`fADO7U;Z_#i^70F+Qo0%Uo&+Roq>h!ZEPvC*J`uehAJ(ott%o)wI6Es2IU
zjI^k^6O+2SKdN%Nx<Uu9cQxZAI7gGXJWpR@8*!}<XZZUkFiGQLekhjpb-QDJX5(b0
z!ZZ;jA^xfr=q7c{W&cXTs$<E+!(+nLI5zzQkKCGkPnizIYPo9Ku$e5GAF8i~Mft=<
zsJY*sQCJT#^KtO;lg7GOC%#t7VaN%jj-WKEJ=V4&eCO1FN+n-T;h^Tneh!6dla)`<
zK(3&hfm>$!1;)h^vfEo_)&_9N4z`nT9y!(7<vz=TzSFMPMA{I-9!>HE0DT?$nVp_L
z0Md=$A0HLd7Hh-7>1li)QvgO1gvXZ+W**bPHXEe=ydArc4f!#~=IZK}EPZ;oZ&e@>
zpbYnQZXqiW4;u9h7z^yO?WhnE4f*99NF*qLC%t9D9Qz+aIDQ~+z;dwXc_D0K=ZYch
zR-q@JJx*gyLus`uKvkP8XRuhe*Kxt~{E`j|b1>hnim%E_pr&q+I!SpGWX19;-=q70
zl{?s%>TRLHQWqNO5MMCm%Fqg)MHWFARe|V4CYa>A$!}rc(~F!ef6LackVbt69i^@u
zr%tX~P7vD(nwSNTk%Wob8M?K$2BRPYP!27HwM8W9GIFr2)-OJ~`SniRwrVJEy53=>
z>F?&m#9r;dR$L{;z<(<mnc8@?CR;QS-N}Q)bLKr1_bQ3m9te|!nANrMFt*)AX7yAp
zpW)#Z1XbSPK?t<qm+aN^f2Dso@s~xU-JP_gaiuT`Xws(C+&ytCz~E#*SMLK*gTJhM
zv`7&#m~Kx*eO7ulTt|5J4!^uS$^ZV6^#59H32bQG)~j*U;DH~0ghP`mnGe2K3Wn76
z-fd5A9%`sx7l0_rVTxNUQF{w{dp3KfF~L2x{d2hg!9ohhhWX^m<bKhw6}kGH54_zX
zx3-;h7AbRJWZxyt&S*@;P&?^c?<{gLZvTtgEcaf-=G3gjQa1)F@W;noHs1q%BYQ#x
zdV$T->rF8W{0}TA_eU_M3dOg!)C&P|1z-@MT`b|lM)G(HHTELd$BGkl&`Gn}P(NJn
zS}Z4_98i8V<u0$xqNbi{M;6OyU)2(P=cB7#?PNrFaejy7UVRYZmJc&AUH#T5J0#wH
zfUQ-z4%Jp0iAP3Ae~yB!H(7VoAj24AekwbOcf<Vkr+r<}xgEp@t;Z26&@_sR?6>%R
z!{6zpD~;J-V+!D$n3gehc1sSMv-jU@`)D~o9DWFW@eVDO{StYFOS5OO&32fFb-{)l
z#)2xt*F??bxvM}Za6n;dS63rb=JX{TGhSN=vLwk}7f+iV9;5EP_ljAC5mUX8TZqPj
ztIyW=yT=IzG8dkG`=nQZ$AD*BX$@7HU1S?~md@<OP~tDFL-b#Oe_LryVV*O1gVOy5
zC8&?)2XFU{49t14?x<^REi6Rx+d)#I=|S;nb8}fui_QUqex_T%!){RJQHFDBNt6Ox
zueT8ZNpa$Ux9<REdG^c}aJewt#auW^-S%-4QOKK36nweKTOc0r>xcdt4DM1<83(YK
zBmtU7>r_qmmk)674~v~k!SV$T4eVAOE1p4K6u~FHqER<uLHAJumQvhp;mNSpXEh;l
z@%Y?a>$c8~yNe`X%!_$tqy4quD5mDf|M?LEwuE|s*3!Zzad%5cUr^)F*XfzDUuxds
zaC<Ho)?bqJ5;2ri$dk>EN4p*WVYFw!kz)D8Uz`&DAR%#CE{Yo9@C*^m5a00}|CW4S
z@|KG5T?&z*@9?t{{=KnH#PY5`m805gs;*Mg5AHZlEsYbHV;|gFggSP1cloPztKfJB
zk+`;ECtD_z^}YnZjjrQj{Wx<@n26-2BMII#0&|;DWYK@Y=eu#6y4V!>t5jLqx%Vkj
z0?RjEB-<a|CdCc(dlFMEx6W|C$sA*N;|u2b|5BH5>uKI%q)nY1t5Yfu**y+PSWTMl
znG|6MomE`;V-4$sy8T)mNE2prm!E0Ms6nwpvyUF-(J3go$OGGnvu%Z#m#31+R|;7r
z%CSz5%_h&fBb9F12?akt{BwLW$a(!U2>E4@2XxtM%J7)n^zJu;I*~?)ewPw-*+E25
z%_nEZT{vl2jZBzfgIV><8&sWy!a^jFL3n{Az;ie`-CPGZ@->b<kX{m~7L&{N3g#(c
zCNei<hmM!??!wS6?yBB&^h!?r9>$S_xEHl2X~l_+MX;l#gOM3liHhy|k)`kpSKi<q
zsIp??T_;cci<{fcfqvZ=sFG&=I_c@jp?QzqS$1HQ1BLRsoF+9>mZbuLs#)Q*5h9kz
zCBeC<IW3EXG8d7*oia`<nR#GL!&%?O$#rZ4ea#c;i9ZGvIYrATC|(m1-#^smPJ?ep
zMpyjRuDWZNOMWs%{5uUQdDwgR-Q!sYUn`%!ie6TcSv+4=ILLe4)p#qvd`YBJbxjm0
zGeR_KR^5eaa1rAjYPWrgbV+GnKy>;NGpoS18R?OIcm<sB4yCDJ&nNYW@t8aN4`2TJ
zx#!5-PF%ZV))c!LTJz)`5i^L|ZOy6kAxsAEOV{ZJ<^BZ<VeyH*g5rljq~U9+eZ?}0
z!q7{<;qm&~|E_;>>Snk{jn6{+ec!5f?Og9hwA6MZ&R3kQv1xG4e`*{F@aevz{r(wR
z(HoGSzHM}|c}MC$<z0DqHwDlz%{J=CcX<mW!@_ltXMLQTjq{p@?HO~_?7()~Ina$G
zc+2whQECHw!2Ko5YS2HCKgfA%<>2yJJpUphQ!>vQk^r^L7W--kV;Q)wza8){VJ4=F
z<d+J-XR=VlXgPbGyBZ8)7(>=xXrd@Dx3m9`#xUml$1;!5*b9o6^H0w`KxmKdc_QqB
zh()DTZ6x26=zK&Hwnnjsv2#XrswATfIhqa@LbkIwo^d1ELMjSnS1=;i+#+}S;)%K`
z{^~oev*gQ=O;VH=b3WSoy9rKHyTJKM1ag;*MRNDxoq6TO`^taB2fV)datV9+b<;5j
z9b^qw=0Y`Uxjn%XG(dLB25TBwIFAEG2y^`Q^;lehJf?c!3(|K8A_UoIKKi?`=zY+s
zz}KlvA|_NV2Rv8*wHM#drjnHPVK}|$TZy|&hcp;E*tWw&3=}o=pCh{t*KOs^Wxupa
zy_uiHAdw>b#RGj$@#sFQPxxa!;pclPII4UFZ`8SVR61QgKcA_Lc1x-Atq;3xA<=YY
z4fA;f`)o+V;aMjW3;g*RG93<Cvf7vt!7C8or3>k9Ns(U?AHM9kaO`Bk84p<YP9wvQ
z?EFtFbw8mY3uw3p#{>xwsH+&I8d)_ZUav_wv?X>lE{lpkYh^Th910Y`b2Hj87smOR
zG143+_m}V$6c`K{0U+HrYdYG+WP@2Xp7d`%inOm5dPaZylJ_;fN;wUb4v)YR*twXQ
zU=jwjw+HRX1Z)g?{SlcVvQP(ADsmtWW7hqcG53$)0NaQ}Od_mzdh^SO^Qg4|0~ZrI
z!XBPu+GN#b)L>#3i_~Qxi$p!;1%;8~889EPuZlLX_b(D#k*m7k6{AG7tqR9G7>iiZ
z&a&;b3+QZoL?vv{c@nykA>BDKrS@yp3%%Vs>r_^iSCKaVi}SQIKv_q29r+YO_r)jE
zoK*Qc<s`iv7@ZX54N36jJ|Q?kx%*#6>)@v4pZkr_?SatB(!#$I3H1Pr#7)@9I*(j^
zKPPXZvKNDdoUh-iq*Pjj-`>qbTPJ{7C<%sMl(LN~E}`=s9jAv>nmNJkqw>7dR7cA@
zc>c!XeDB#T26cQl4$FZa-si9Mcwf3r-cD|ICa!Zj@|Fx?63QRFs(MZtN_Z7nz^#__
zic%g8>sTuB%?bS)D}HKCoQ87wfygQAt_$+c+hPaqtC*F4aeyUl;{2Y#S-}b06s`&O
z6d21;5|;as01=>%ZOkxXp)4$z<%4VBt&)}2ISaO+V|)sEA`3ma2~Dc{{YO=G1F@Zk
z*5QJ#C-u`>izhq%d)3c-28E%aEY4DHrIdOjz%(1sZwte4j!PyIDaivzs7F)ll@Di6
zFaks%o@{VyduP!BI|6~&4j4RG0DP}JQOSI&*b&@7ZXrgs;Elqj-s!RDh-Vjex8W$S
zq}E|2+jH*Aeb2gI*o{7Uz8G$9j+Q;w8K<=`t}g4IR6n;cVKSMPy<#?4t|4ht&Dq18
zmZA@mZ>x<b<LRkoU-4R#&p_rUqo42E)Vqz2I<qJwE_{3GVywhJXlja$#J^dhFtb<v
zIw%06$K5|7tsTWR=?>|QJ&|aFoJnU5)9^#SzGiB6>79*x<pFXv+ALI5Aqx%6^mIkt
z;tNtkvj}&4hT$?=U|cRLLgzsQs`t7qNvmAj>wl;KXPlt<T24jd6*+#n6e897<xeVF
zS*y-qx$FDU2Aya%>$g5@E5ASDqwNX43Dw`%JEwTxld@ZSX0uC9DI-5s4P{eyQ!Xpz
z(Ma5W<TnKZ-tC;S3|PI3T`so~)Vm%iZm7*^-lku^@Pl@xF-B@C4BZy07E}as-@*f!
z)idpccS^TQNT|?0{~)q48{F$&hwoedeMf?LS$B3%vScF=c$>RJu}jC%-;a>AMxeVA
zpImuR@cOY(B&ZAG)$Ci&zpfyfWQo#AA4pdKhIGwu;d2BGG-;&L=tc{h_4XVijgaxi
zA=`o)ALo{gxzYNSnEKx$&d-LNLK6Ah#@5?};dn7{xJs1vcDpr5)Hbw{dP=3C6Z^b_
z_fHha7cc^J;QdZ<u-!=?QZLXTLJKg9%8Ss7e$sobQJRiql;Gn%XfD$4+EM1h`qx##
z5K~RPP#df~I+70NlIH%RV_z2di7OnS-VpPuW|Qbl#F#eO!pus1zPEaG3Q3KzZ&H1|
z$WQ{IK>bhl9h3F}=dp@1VmtVI3fUu=A7*%parh2l#>j*-b2We)gYjIj8@7FM!$(6G
zq>}u?I>IIUNAs9xLcGY^J)rm7Ef#WX^Yr<*|13Skw(NRtFwzW1&ZZ-3x5-w6{V1Mv
z2EGw}-C(lF!F8ZQJ%lR6OojPQ+7*WUDh>WMaFFBn7wg|(vj62edHyJ&QfEtuS>x6g
zT3Ya6py6y2^2xhz5pFSx65l9Ku+5Q_wEaxXmg`Jhg@SZa-&xFx@+GapX=A%TdG3n$
z%_c5pq^!5~4^wn5V0|yxZ#xk<?&kW{P5n=VkXDwi={UdqZw>bZ1fGj!*7KZkb)T`~
zsyd5`7^`bVu{T*+wM?)wQp;BT#GS3`@DFWL0$RxkRKn6x^nHNbxVx3~sSbX0yBW`Q
zdk9k(78Nb2Wall)no43Gqvfuiyfhps6c58$tv9P)bZw#8mKZbO%mn5^_3)#tDFir<
z#~Iaz6j>#Gdn7Fxmu-wUIkko4ggbF-&pv#T-gIHT3#goFqogqtQz(YwS?>KIc#xBx
zmz*8Pv0%`Stq&=IX$FIwp@y#0^R-*(MB&R)&t}ifN#{=^5~-e~1Ve-NUc(peH-Ddg
zFghRm9m&u{cEmCAP&M(uoeZAx+_y;23IaYvr})uW87ltTx7aR8d9;n9zZI&)Mu@8t
zzRzg%AVp}@$gssXQB*rfy}th8+G3Dq;S#UYZ8w4vgpX(-zm}}%gY9RTN^*@R5BB<m
zr002Y-ctCOf>U#r`C|BI0M7F>Wo~seUcI;4_I+6&);r|tJSI7!SgCXrHhNjG1~xr_
z8B!|7b+RhtVv#-cn&$QCdZvsH@~6O?-(dd`MzSGYQFW;m^L_CTRpgunw_%c9!Uc}X
zu<>5PS00y7>ni_^Dn>8lw}3b@KE99MC!d!UmqoWuD98f?oh8;TjAeEyeaU;WG%I<B
z)JRs2SMHU3=GVzZAqb02y22=*z;O-xWj`yKiRR`(&>;2I*oNDQ6*W?^KIP2UK0c-2
zJb!k<!m6_f>8KT!R}u`Orx7NIyu3U*o5{eo0bh;)y|f)ngp1C)*kly#H7nNBLKEY5
zeY}m;f7m`|O)*t?>9fhsi$*ebtj-4jF&F5d%Yz>uwF|=TBknBK3Np7@altf2g`$MZ
zZg~fIpNV<qqXmQ(Hk@OB8`e4QCF>ZWzYBqJtoraxK$@2<!B__8U~{lSi=!c@e&dm7
zC+^5h)8vS?8i-}2QGRUwO`)~Js0-q34-Cpy9lpq$^scvzlcpYtr$wFP7j5TUE>Ap0
zQR<z|{>F@^><EPc^fsB1PJL`zVc*y?xLxC5VcC;6U&ZzH%5=T!HqUw_a(cQJX(5-P
zo5|G7*XiuRxO}Ql;?McZ-F-31S|k1^8d~nbVhTb^+3LSV6aq))I^K1tDt@4d`x4zX
z+Qsw^p7j5y<I8o}4as9KRmh~XfK?Y!l`C1E+epLDywX_8Vl$g%Y;v3Ii|u`3yCFnp
zkk`c6i5qP(CRyf{6x-+p9`$rHfr<NScXzi<Ku3{_x}x~K%;)2p$nEw1i`A1lE?M(1
zuMa(i#pXi`AE?4U!L44M`MaT<y0akT5Fx{TK-m_tynrnr1hCaS#x$>J>i)<&E|&Rv
zx^mTSqd(UO7_1FXx2LV3z`S<X{^|>KiHC&;l7#sG=twebkmM5}JT~q4hmQNgb_By7
z7~%&awn})bJQ_`z7fl2K-;cYmZ{KvVT8hnkD+6}o5bkI%PLVlZzzM1M<J}mX;bI@i
zoPs#%XS%vfV^000z!6nhoVI`%<m#HaTr)oofwXp;YUztYVZyqz4MuA-P5h>IDXjfL
zw>u}#oK#`w5x{BBfUa(sA`R-!o*}}X?H;G8UX1e!C5PIF(?^m1xu?pwhb>4lJsn-2
zbKt}qAG&<ss$cVKc)>5*a(_-72lqzpevDaG3%R5(@;33T^qm~~4nqRvA#ur2VO6IX
z#*|@@c~-%|c|kPHeV^T@vG=TGNrsuzqO`_|v216Br3ViNmV;>OZ57>qH***vx=Oq)
zI|geoG0;~Js0REe-ZH==W<%5KN^XV@91dUn!>UJsqwJ-E+lQ_v4HA);<!=|niSIQ^
zHe%(<n&FO0%-riN{OpY^v(fX*UWBSRZD_nkb!#`~JT^8cGySXSWyrHV7qS5#u?QPF
zr#L|csvKp+YY2oG=zvD|qdbkYb_qv|v8%kQ>OYJpQRKg}N@ZEO($(&~p<_GjU8ZuG
zLE)yn+P_76Sa-e6eHIVCKe8~<UP$^_uXxk!w~2(+l_W|1o4G<bhf3R;x6^pW2UK3d
zcuXbZ2;t6<y8HDY@mUsAL;9P}?X)5KXfo|BzGDRo6lj>r^YRP5JbecrZ=;G6b{c~(
zT^fmJeHvX<7kq{6@K&wlnDMBz(xSF|{=R#PBI3Elx*mn%>VC;{apz2b<@`6J($OB!
zSoaNT*#1G$s8V+v|Kt5Mo}Iyv*060aAV)%=?)~ruQ|)_(mWG=0^5muqSbmNafSP@V
zdNnqE)g0=L{#LlgnUDn1*rr)1P<9S39CAJFe>xMuWAB&9^V~U7uL+&-r)I^Bd^t!q
z8%^s(`L#;)w@0mWD(IMbVz&2_&~OXwEI9>Q9)O@R($yq3Zf@b|gWi!_uYN~bQ+EY!
z;O93ltA*#cBlv;~WUJkw{T+TM_Hfr<GXrYW+YJXQq86vC+@I1rZ*TMNXv-apYq<XI
z9l}A`%4$F(3Pm$LXzRk!%$fX<(Lz>J<GSQR>EX$vurfP!<Ks9(-xgclL2^VD>|59$
z?6_rOd=nD7c6YxZ>-8-eL;#kJq8gu3?yFfkod7Pe9k@Q<L<Ss=1x>>)3h`b?vl8=}
zo&GQs+`K-$#1enU&hrYg<lToT42@`tp`GfLfqRp4K-x^r(x|BDqKlnmbPjyAr+s2-
zoMcq{Xcb11Bg<XLD%*=4q<im(KxBL(0EVcT83<*A%9+a7#Z+gd$s7{{#mu5gJAL4m
z?U07=d+o#r>{Er3h=2E~unPr%@QY(b4%;V3VEe3S@>9UFEd_En#Pjljg8%=PqtXgN
zW_}A1&CY<URIcaUTK`KoNh!AICe+XLL?a-dyn<c4?L-m>0-RVf73B9fy!9hGelNlP
zC!H#{Hjgde-uNiJ$HWhgU<4^SR4yEJe)EdUOSrp4jW6NpMdga7<sxZR)X7A<{)R;+
zJ8H*3dz!2a3@I_kEIode7luqupMh2h1<0*D6umQwnr!uT>x+NBA6~(A#C_<rJF@)e
zfiDP<cO8dSmXi8cLMbE9xBIVd9Hc-LS*=KW!PfLMO?38Tx^nN?Lq=-Z{-9Ib3vfHq
zJiV?DbwUk}koZr#Hd&rfjoF;4G2{Sy9(&6d{omtr<R#*oR!3<MyOD8sTAQGZVOD8m
zI=FY_3>062mUhQS|82t4Q#Wc)mSmq}gXUf@rdvHD`E<FcOwRMW?8yVyQ}W}p6xHq?
zo8UGKGgC}w9OZyu+o9d4Hkd&JLH-N#wU#mjqL!_$-a5+Z(9Zg~{rzW}Xe26wfd5o<
zpc=Fa7++g@Uzv@r>z7*(<Cgmn7cZ5`vN3y3SbB_>FsTbk2+L&>im-9(M*8tL^JYP@
zLhW5-v=_|9`;1GYz$#IwR(9Ud68D(C4xxJK?W&n@hVv*s&dFw>we|eooNwk_ZY|C~
z;wD^&##-E3yG*STWWEi11EL`QtWs~V$i{{Gr_=2vD{pB1<aKDdVFrjrWj9_})i|v|
z^N*@bfx7$PM%6bb6ytX@=h#t3NTU#2*Wq)F012r|>`uywa>@}{%Wce8PrvhDGqfxQ
zyEX6Qu;@0CQ-~mqdSWzK(PSSFSN9)O&8etl6J3NL^jlgx)u%?7i39*}i*6x@ozUAS
z`syvEioTDx@f*;mPd#;#cuOJ4^pBSp2<9FCtABA^VHXscBu;rOv`3<zXYq`B8bV~9
zf$*BXyNs>bgbM|`_u6NfD}IN~{IiC=OobAgB}r6iB&2R@%MEY<iT?V8{@@BS7Ed;R
z5C;_i#~qs4ZGnFOa~nuP{dwE<#C5&)QgM`A)HY=z8v24`MUqddHDjkVe0-E=)na;f
zq#xFQwPXCZM%r9h+|<0CRixLoc>n$1NjhCfD%MXeePbRDdncr^ku&@sHbBQ~A(HPI
z5;AapogwJ9`di>_n$n0R$PTtrXH=6l0kt-wvr+uUQfA(<QnvzFO0tzm!La%3>8jX(
z3Xxe&7a6UoN}LMEgr@K8pYh1}TUU-__DjFzfeqZ|L#6{*_~9-<LBZvrmTV&&U{<%}
z|F?NJvq!AFf{dPF(*9rb&K_S0ZyntE_Mqnx>;HDxs5BUpg_P3QRGw`YOJ+%LJCk5!
zA1h)mUAy<8gPyzYokK@vu$M34i)otT##=%5*^jztG|^ZRS-j~p(^F7voo*fUB27o&
zRPu)2T}bu{^=GarnsvEkO%jU*J$pI-xIOM5b9HN>HYFDgPp6w#ilQ%OB>}917k7w4
zn2-q0l|{)QOBD8Wolzw2bDE6AXvb6X{7x1~y@K;KvD<P_3Mpw<OKkX%Ek0<w?=ku6
zp_>uy$%wr6Oy(xRq7b8Tab-#_SPegt?(xeGSwyK;P+>8i1eUYmg0HtGizg>kSgE@U
z-|wo9u>mIo2~Tv(!7+({JzCb5RWEadIX*!trQ|<Z8VwI^k5x@w3)N=O_lmsj6x|P=
z(XTI>q`!Pyr;(!q`z?l!zE4~j4hivCV_aav3Xu%N0}ZW8sid?jS7fQPf9{!0BMdzo
z`}DhhcXz&|>CJ0-t#$kvsml<XZD+2}Ch-OhD>QPa@v|bAW$AGZ<3eWSMgI*Njn%Ar
zUQEQq$NxyibnJL5WyiRk%^sfzOXKZ)Tcc*wd$$y^;;I5-$)=bP78Z|p4iLEY#RxQK
zI`u#J?h6<=8Cn)p5~3>|0iMn2LfDnl{1}BabDYCSTKQC#4+BZ=yH9Qzz_Xkp?3^iV
zcz=m>$D3eLq-?}nK7}MNj~upIWf>qtxT_qbnqgxdfYafE`!6pZQ%#nFa;nC41m^zI
z#btai3AJn|`T6wwk+kgPu4vwv@Xw+h?{2Xwk0#E;?a$|dleF>XX@XR~bN8HNTl>70
zbo&7!+CRwinE5QfxEKQT>ar+`o^FC)=0x+(P(L}1JZW9U0zvBzu`wWxes_H!2v}Je
zu4-hz5-^A<mu>$3T0bjpkI!@K*mMNxyc7!+dviy0;57dO0J8@UA_{xVc{qSzTj%4m
zOX)Z9Rji!B)8nIe!0Cm8z;VB_68np9!vFYxJLdh55;N)YzMGW**dE2tHni4r(ixkp
z=1z<o7o9Nv(cIC}2S@vCWua$dZmhPV#W+qn4H4JAI_(31v2#{jBU@+!oQ3mY24I#k
z@)LGd<KEF!C%FawLxC0@T{*aSx?JzX75|Sp4nckWbq?4P(Fr!}4*RP`%L*`|U@Eus
z2kQNV<{F2;0=FUF9I!ib8q|&z9H)MeA>XX4K<@fWCeW=Al<_9O4lCp490!)}gmq25
zIyxDqc=Zm5J35OT907D&n`sTOH=qG-)S3C7&fTou>}2uC2%-+vY1-MoP52SdkhYnX
zS=(*7;LU8hUbT5X;ba#2zz<yV+ae_wzYWQ-;!bj3-wbhSu5Sg$Zru@O7%0<u6q5Ry
zh)dl+wFwqb#ZognD?+S=<G67F*mSh?mH_lJMR^Fd%#d#yv-G#3W?fd{#rD~dkl5JO
z+Y`<;_eTHmS*F7E1lz9X)VcH2<c*L|+V6NS;|b!#7R18N@VO-Jqn0eC%+d`OXgcE}
z)UTBwxzHjY$3u#$udICE-NYA4F-~+NypmKhkcP2z<8<;{{UrADbSxju>)$@D*SK`M
zmL|hppIsl@?Jj;LQ#zRYE}^hK<ei<?c+tG_w;M+)4ta@wbt-EsNa`L<XrNNqGl`#S
znrSm)&11upolL@Dgm$g}1!?au)J9vsv!iu$^E@YaPx)+QhzgYZZd-XhAyG4W!!SKv
zDCK;JBH=3k`QgkrQ&rur{iRjZ_W34nD2U30uoGAKz3CUE1*41jPtU1W#}v|yjm>>a
zubt>%Z<#Ab&N2F&az?FO5@!n7-<()~B2`jm#(ek68oiyl0r^D6C~eMhVqkz{t*hO<
zrP|VwaAE*1xNNpY8x7=zv!*1D{htp>VGiM`F*|`|;5}7Ug!Add7$Y<3OryUY?H(?>
z+icU}KMGyl=_U$QxF(VI|2f)A@rY&i+S~iwOrbXblYe$r`+2GGnMJi}r^Bt=vcpun
zg`;|(IXl^g6c`r<q<HxH4h{?e@E|B}xS~sL2K4BjNSk(dK<C4`2{6;-M@H}|tBy1n
zg4t_5a7vZ&yFE-Kkrr1{UeO8oE#q)_Y^{AgJD6Iit=|w6AB%lo-zW2-Swd{`jf#mz
zaPcycNIO<6fzzzvNSYl2oyI|R$hXBJz;t?!7+qI-718t3eJ%B~#)aaC?tokOZAR|F
z+rFt}o}u=@C}o>P>r(mFQ*PzJ1Q!53wAw;WrJglmJ8p-Tzb_;zX;GgASOYr)0|4X1
z!XKabbWC_OPDg9&#%|!dya5EpFTetFlVc=zD!w=c@(ruWk&%*k4mxDx6JU8Fg6>L$
zZ<X1R53@%hApm?RfXaI0AJhRe@$Sa6_)s7-f*0iVMyyN}3hc<KH~{QA-xUPxm<~{T
z-sS_lC7kk{>KW4ud0doC-Rhua?7TUFSl&Vlsrz_WF=LCa(hl4RY32V$N@eXZJ8d{2
z)2jX6B0JmL^!6v9qd{vP;9j3;dR^@`+VBfSWGZU{Xzp(yz;G*ziYg@&t+x9GGGRL~
zapcm2576TU4q~nx#{y@ybR*bd)@^n?3*`CZZJnL#d-jyb7}n;L&2UTxWAK~WPa#8@
zVGj$Bn<PEbBwpkMMMR54XYEuwe9qWXvuYjHNzxYP=QF`eR3|ce<!?w!ggpxhl6R$p
zVLfKD_ss7cuWdzSY=mcGD{*L8+~uKZP_hv5&hi@88cr?^qk5xN^Y-=0&$yu<=l#>i
zQ+Q;Y@UTBhDJRZg2D^WMIlNfg*oT#m+w#g5Y=|WJQJBRHFX+s$A$8Q)A>qPlPDvj+
zAQLSXZz~sXWTL~n8&lqc_qQg*!Y?0co{BhLDbXCiaeU~B#TT=_y1z-4l3UJaI{T_1
zEjrPq8BZ1mqj{4B61YWTXdX{IH4U6f>S2?9tEu9yLR~tpnGM<J^HzMMoOl|1UsX|8
zjkpe+XsKO2zdi!cWi9^XgMpNVR9>CTnti_zn4E$ZI4egj{m&h>UFx>Skqwa&${$A5
z?PC!a60I?p#1!}jgBP$1jb=8k#p(rglan8NRVBaC<jg+nd;8SMMhPoYam1f0azymB
zmg+~O{}DO>wEgzu);2SbfG%4B)4HrFz{6(J9i*VkZ>?Np9dqIC?b#P>G{aAp!22G)
zkxSf$rHc)n6*1rkD<}^Z5HTXewSKWOu|n0e)7c@!L}`1*!ujUT$!oyFnVlGGRIDWM
zF>6W8T0oJLN|~h->FBM<Se9ye{iM@4^?L83$30mrJ*qq_2MZ&q$QZZQW^ixGh26NV
zg0&$&X?3IJrqIK*!OPEJ8r8Q1e%D(K>t-4^1p#RgNu6JPQa2B;?)j#IO<puD(!2Uy
zk-7qEvZ<ybdRAPA<~mdonw;(TK9FfygnXO1m-;y=qB7g)?ZWy4dD2g44ar4})%rXG
z74b3rS386jB_FLQ8=(zJ0=_eEo)Zfb+FZ@n!;iP0UUpLt{9MMhGX^I0*fX6txoQpF
z1OQ7@QM!CZ$A|S0c9TW&<D9kEVF?{;FEoiyLs{W60v5qss}CPDgs9FnEb!ILJ7xTb
z{MJHMXAiT|a{yNoP#6O)#SRTwk#}bL-|dglcSmZxzGO3v$%K~HRP47i3Z2_Z<;okx
zNpFQflD&Zbwx*>@uLa;rZe1Y85nj6&q^1j?e6@&gUn(pCXTHAf30HDl3YpdY+n^S>
zg-HB7;B16!&1wUgQ=0!a^?Pm}!~&l4GPyo!5;4rV#uMVi+clY_?g;zL(fABHgUa}g
zD+f1ih5)p)7>QU=sIoA=iCNdL_PC?tjHS+R?G7IEw4K`%KlFpIcHFp4V~5$*&`WX4
ze&F!eZo5<QaeVzH4PMPPX!pK-Ze5d9zYp-mr3)5Pct6oypoFeNjn}soNo8t|aHrhI
z<`Gds1VHpeFePKZ*)|3De|olw*Gb*E=4hq(v18o*v|H=o)O$6@zpBDxTKwP8Ul(S>
zjA$P+Pi<w0y87MrtU8eKkSZNkBwf6vAGQV)m@M%7n4Z+{`F-Sk2*<Iy7E>B~`1t2G
zNQN~Bb^H1q`&oFx?~rQ2|AucO+h3)Zy&O+QhS-K>k?mg;J@5b4@=LgZm}^}ua&r4J
z^7t0)T@NB1Zj#7jDk1q!MvD{=hAYTNGk@04W+kKE9MrNh8uJD6zH_48&&L${3oTU{
zPC3m|ieL8&)>l{*VAy480W6<$lPl$^Ly^{j)pz@`axFK`RvMWG7f)@MbO^Gp%uuFd
z6Z8>FkMYtxnrR(4W;F#X)}JcSYw_F^8`4#!sJ#d+vK7nV@gLs-g-HWUq=gpNk)Adw
zI_282aqT;BpnrlF*jJ(+7R#hwz*9H{n0np{W~%&u&r8V=Fw`p$3mGgOZteE;-(S=t
z%k%o9+YjJ+1!_lN$hmJ_c_#ldM}A{hIi+-BS6$-CoyeNBC9)ZD=GnNlm$DO|-3RiE
zl+mAss)Z%PWHrUg*!c}9BTCY7dimz<tVRAjsa>$NF5RA5xwuokIwx;T-80cl2jVuL
z<$nklhFxPpAd^-iW#KG@^t{qYQj)Ko2A%BOf)&oMB%DuWctpT3^uFbvyG&_^2G<Eb
zkQDS26p$ijKDOcdtZo_J9+)@(O`BB*Bo|JW1066N`r8PangzB6u*Ltxd)lCSqWnQN
zv16QcWK_yCeM61G5q-(R+gU!$4{UFZQkQ)Avg8a<la1buPwg7$&o+ws!gKuGGCLoS
zGXhDpm1G0|2FNtQ0Wb*~kZhqWECl6GhU$F8-mQ6+wPKt*KkN5++SSn`LtlI#_}@bm
zFE+5p@-=ZT-DeSt5*=cD8o*7_dH)<lQ~+=&o8u*N<du|Cb{bt|veh+gJS``tl|6$0
zADgfg?ROTjaS;L199e2uf16*QIKX2@8qHhAeooa*x(5W{1K&rGIU8nF3$RJ0of2;#
zB_(umJVQY$rh@6S7_<9;HUmErNbk#@syGnIauYC)N`B8IBq>^QVDx{$@%bf7?o|$K
zmiOanE?1Aaz^6KsmCbx$VbOpQjBWl0CIS^^sHErLS;uE8Cp;DqbjDF|@xF;)$}Xdg
zvm&vm+L%7)&U^Ig3T@6ekrVp8_&qppv!+cv+azr_pH;84P$p*29YO)f1Aq5^?LFKy
zrR$%qtEPGb@z;8XvIs8nSM;bF4Aa>-&ot6_k5P{8uI~iBVw=dkZSDV&t#<k@s=bLl
z%xbs%4*!7Z9=rW5ybTb|5Yt;sU^_4Gg766Jw*oTcvc-Sjn^;v-%q6V4ZK;MhL*G&}
z)Gx4{QrDaB3#RnKarOQ=@m)*3?_n5J!|}3q1Cu8CeSp47r+gk}9xg+!Q259n;!7=h
z|KUGuGA;kLS>Ga8C%bmMgl#cJ52km0pJV%8TPKvei2I-<zB&*e9r%cCr%B7YX*Q!r
zAqaoBu)tQs>DJoWI*<k`RLz<w2Yy8}$2?wd)%aXn`xH-!%7U+_sTz}xwN^I|Ld#o8
zD4iitT97x2lV4%*k>Sa-|Iy0mhEQ4QsSx8p6wfDW<|TF~Jb}@u)0MQ6pQVNfhYo>+
z=r1a)@`3-bN55K1aQ#-U%bLiVvICE$I_hs#D<}<2p8*K{Au$P2gg?N8+W}k~zg)YD
zDuD2-;Nw5Pu#x?JmNcdC$@L>A8BJnuXj!H7(V?TUrY(c3qsKfpwE{X%nT*8L@MHXY
z>**$rL8Iq&+^|9b9)6=C!o(#leLK=f55Dm#y|m}&p457siMPeBY}3_K41?4h{fykV
zd~21v>xO>GmRFvqD1sP)k(M)Y-7ZZ%C*N%n<!7Q7o%wcJ4S`5X?oOYVJne9Tzc>vh
zwRtX=e?JTfbtkcKeT!MG75NZHPO*i@Oj?;U9Psr|(ruZbJwx{7>e)(WW@p!8i2G<7
zpblkc08GollNSph4YKv=<&JGNxMvW~61Vj%bYi&|MPkpAHRUPsd<&YAlqkviIO{5-
zGhrQ*8d0;PsrxoFbAjEh1>N(MLMtn|vX4ciq}9I#D2{=B1lu+!CBeSEp%l+J$e!o}
zB#|6ZI>@$3aY>baT6pH%xL;E?*Ih1QU&x9t{g`r3CAP`cgj<}e81dBLSa{jR9Y~gw
zIXgf~W#?VF0$QJF)Dbg}5}jB;ingF=B@`o4BL5oahkuqD)ZnC|Bg0Z_^Nc6}P`B|-
zjlVTQ9Yi`Kd1`@k&8}a+9B-k2^wHqF*jnGkJ||<qa`I||0br+-Q&FWMNTCO<-E9^U
zfVAKlH(!tZ^@>k*Xn%w>cjCjWZk20o!2Kah=82j1k$_mheZs*QR{P}X?IGwYaOiHD
z&i6+S@LELjpN@<gtXG1{Qc%=+;q*&?D`u#6b1uile){BxQ6;5WVPI|X+3|#N92isx
z?6{p|X6h|6#-#f99ztxFhhqVz?N9krBeT{fela1Q{zPLuONB^H8Zb9nEV)(?jd`01
z`CzG-^I<Ne^wQdZorZo@#a-DP)!<cuMs+V{EHUG)tMb=9`WiHD|7OLavp;7=9OlV!
z$@O|se>1egJtOY6vpmnyQ>HWm7P-^kc{8@o)4iRK#rrlXk~2ADs4yKQ|Atq?D837L
z{_le}KM;W)9)3b!6Ux=1tqoNTU2<964L02ya}Q8L_9ZTS3)_5<NViUu@XzzEsO#Q*
zXs9(yTSTIook!)(No>-8Gsr3j@^gX1d`8A$`dz2gk}$8{mKqlsOG^axm{&vbM-pM>
zH=mR#F4Fk_+xT~%(NeqSad(XQ-@D-ZeI{W4FOM<1^dCFuZ{)^PbZlkeQf!Rf|FAut
zZgMSSZ$ig)ZBCvd-wLMJV~@$lMP3noVtOgZ<B;I)*w1^)f}SbwTA_Z=TRO~g2mp8K
zW=+OCekLVi0Qiy-R0&1Arhs+(3kgvXrecOl+}Gkf+m}zP&lmd5)~Xl%q4@PC?MILw
zBIUs=?KKb0`3^WV$;H^HN*s=gc&L(^pP%Jkp@dU&)3UrO!5<*K26w<$7Se@7AH2t+
zreg&ytN&~B(w}R)X)bmwTo(1(;jvitUq$$A>gy4MD=}tAP~m8!Gn>Ik45RIO1?$EU
zkvEi*=Ahg!4WYrmn5e>}3!Gl%&$6hmhAG2c4c%P#+g~F%nw{!R0b&y>Q#dAzqX;Y2
zp#6D+IUszX1(YO0bHu*v%QGU^B;bEia`Vum<!{0S9mUd@Ur#Dr)IeSdp((RfNN<UY
zC`zJYjed4{(=&B&kTGK2yESxju~ZI3vbXHAs_2ixSvelF`fRJ#4~>zIBkoU%m>*vs
zbr^Jc?@Z?(Pr&P!(OKkJkYWrbej(Y9B4cp6a4?QJRPXqD0cGQ*=~4dY6-OWqpqp2~
zal4UJxRF;}(X(;%gHtauIrg``PBQwBZ}f5UWQk22sx4&>nM86@aJ?)rfrJ{A)D9OR
z=Ff-7<#HQta^{;X3(O$>0nhj`_AYlK<BoO+pPIgijda8YUtvNp%A#SIOf=G6@n=*w
zWXoS;Z-UIqV+YsyTr@YVR6p>q;A+2pZYD<6uio<`xW4xUv9!GF&-T{`VrdR?3r(Lb
zL6UuUkP#iey$R2WX+T91$(e63To8h9ijG+tg?kZ<rtT2kp(2otl)9<YfNU4-pN9#X
z(w!uY6!9uSGFup_e`{Cw&Q_8nu%p_}WR3mq{0CQk{?^`}P(&v$bT;5b$gLx0U>bCA
z6$gZ7?*@e7?Yq5^u_mT#fMQ!dG;|9r7yr+xGbHFMPeXFme-oBQs)igKxrybhmP_xb
z{2#87NuD5;BhvQH2zwbBIDgf=Ap3%#n4uy|)A%;tbqN!pAc1eFe$bXM<$?zpR>oO?
z3|^Q(*_)p6kiptQs{WBc8uA&&?>Z+!RmsjT-!BPEEh8}jw{;nPgVcO9s?PRVHu5XB
z15~-n^b`Z0vf?yY47QS9B{+I$U7D%eM}W!u6%{$>;<KTMQ;Bxio<~+)jGt*J3i4<K
zRqB6~^nYh)EiME4R+RvRLRnL3jnL169xv@DNe=@Y-t*X1^)>Ry=3XMvyjnrQs|$t*
zKCTo$t3j4BmAd-4pYJb^PCzu5L6+8sKbJl?ZGI|mv<>C6^9=@!&vrZ<*=pt60h||;
zSnwHvRPR6m$<3}I+<?bbD>XHBZn*^OWA<i;MOW9eBDAV;swP5`+iI6Eaf<8g_J_8t
zLKJ`X_vPyQQWQK~I!0P4+7#6j{nX+kdkJY<+LU?XmRgf-W}gTevZlrly3N{fDGy8v
z2N_f1jpD~N3&`STk`bgfni?4ly8J2Mt*JDOJePNtqFDcrN>kE3zuiE2Tw)0*0Fo|3
z!s4{?m=T=xw|#edo?D5*&Ga-dX0)OfQ^s%fAK9loD_Q3ViDiwpV;a~;;v(3&sN40U
zzFDEYUCJe*nd~AaF*)I~O%PMS>;x$WNY_SvJ@-(*GTZ8;iAsHwBmo4Qrf%0GfqtI&
zeHvjZ18sF$Ru4~j2N1AE79F3LIDO#sdMff=6Y23<21+sa*!;GPU7J*1nh=HUIsvWB
zhwa0t8n6Fxbe2JJv|A8H0|XBa0Rq7#xGiqM-QC^Yo!}0^-4on{2X_yPyX)dEcfMN%
zfA~QS#m>BQPIo^IfKb-E{OFsaHwrqjYM;jf?vdl-EwIhvSOC-?|8`$I-|arCum`UK
zt^D){2s@zrL_4gr=1pX%QYMSAWsE!ei`qMn(lHB;0^=J(Wg4Sf=idGqA!I;|A}igz
zN!!T{@J4nHw0^xLGSQe`*WmVD9TYKJv^)ZEq3z*Ns|Bgmf}C?CZQ-I2T|I+>#=>f@
z-MOUSG@5oTm>Aw9**Qir#wx&olU;46C0mx_6=r9{S-}V}-`Q_7MQlo23#aP_$w=&2
zmj`G6<#0c1D<tB4o>E|5o(7{<Bk#6uUG`Dio@6##rBB7B%Toc-cLTpi)cCxPp)yi{
z7^6|s+VN-m&YS>#Ci}>B&<Dc5cblbA32e~|H;>&TuP<H(Wb6SnY`vkE&lp8`ax#y@
zB!H#&|GPI&2}Y<t{D)e*#`&3;F@17#C!VxcR}a8L#)2~+eaZYt_{R1=lmJ_>K~rTX
zf2W=_1ApKdAo!O$&~Hi?zrmH?^AKG!>m=dN6~?urMNHH8N9cdV#?N<h;nJuuUWVLo
z{#w}If~FfIj{^zBztENR7ycqoF5wt7<t9~O@4uh@R=jDbjm|<K7Va&={>l>^UyW#w
z=s)fENbG$d+Am92;+-$(I%Xs|53$KT;c_f-f5QC`FtSk{OVFHHN<uxj$fe9l{1mW*
zdberpI@fOEXF)jXDb-bBDJKLaq_{n_yc7p^FP@dXlz-kpaA(^t7|zs`K|hr>_q@1>
znUjdu%+6NC&G@5@O-UOY%gmDWVb7E0V5A74<Ng7Y+Ct(`*;HG@gFP#A%+PAN@!QK`
z4_fS2_4sQuYRfyH)xSDQj}nYu<row&9+glrKH8~$)i8TI9i-`tN<%=uKp28Wyw}}Z
zI3KrP|Lp)rog>#r(bS)b8Of$IKue(o;-=m2w*p#cxkThFwN!XDE#;qBkrAp<9W=F*
ziATVrwAUD9R+sML%MJ;_rf_7W`hpgPf&Hx^@iz6=qA~L5R}!W4)l5mGIXR-5bfU!v
z4ov#ko036EEAU~Pk$9@5(Y337f?4KfrE2)qr_jA%yo#TnuzZ!UceTJ8YYA<0AFJ~1
zi*AK1Xw;Z9axVd{-VnOLGc76KGf<i(8Mo>0U(OvHOkqo6J{82%#MbG)?3aID3&WEA
z{(OG@O88kpi6iOgYla{i4dLRHD=t4M?*(W^-nNWG&n>O;x_)9byWMLMeAoA(HUwg~
zc<)hjbMR#r62P*xW8$z9n~Bu-bhfW#)&SnEbNi9&fCi<(YG9^gb4&E1XV(4d)jH|=
zVur=wMY3*<{bj!UG3;cFQFO*uPH#b1U*D-+cdj{=mC3ux^K%u5v*30z79n)>qJ6P%
z4(jZAD;7F)dqpez-<{w}4s(<+;%G$^BS5C9m$g36kiEQn!O_yM)nowLo%p6K+rt==
zf7Ar(3TrI5x{G$&)!NCGB{*~4;&SBWyBd16Q^3hBfZD)HsmdI^?!=y&suGo}(!K8?
zg(siQlrTFAmzo#BEIAX*>lgDiKoD%iB$v-WZ^FFKyR&eq+A1r~dron1)zx8YY6;8;
z2I(TwB)c@|vLL%uJ2+@I?A>vibd?9jJsOXmoe}0(sLo@r@K6o0k??!}l_2F8wHiC&
zVq*F31mId1-0)3ma|U>1ly(8Ja8661bLna!;57~KwY{-XT}LHcDc$}xJbYtbW)5!W
zEnOGqq?|wxL!5wIO-LTBIfO*ovkL}S=6GhYI*`GDe#F|6BnGk3fn8!_Tp?u@MH{-)
zPJNS#rgf_sWH?vm4mqb;Jg(U5{5aSI7lF(&Ryj9TtnR+RwzI0Qw;s<Oh3BP9PLw31
z?Gi-#-0{}T_+fj{Ys=|dDP+11!Y`Up>zGC%n=k7(lzkX~_N@D)c8nkI%j);njZ;0u
ziZUS*`rnykBIbo`4;CjC<#~4DFt8Q!ioX~196jm_Px0y*nL(O24aVS(dYlQCSO=U+
zUN0Az)YL<==9F#IucefKYYG{tU7NqZuySlGGS@b@qx&e3`Ywc3eNW&f0jb^?Vw4n3
zaQHKJ4JLpe!V=@sT9aR>l7$g2-9S9v!{Ji`2loM2L%y$j3ypS+I4-hb;LYrTr8q<4
z45e~eaJ5b6pLp!B!i3G>dTdU504OpCpq)c26EbVbgw$lYwLh}~;ZGT-3@P-lf9Ikh
z^wKb#QNcD$r=Lki(`j^P$%Pj<4?B+Mh^*nieNQJ!W-%ByiFMED6PEdCd(Z>W(VYb-
zF62Yvs~JARw-zD>aoP;0L{6n6|6wDDRab2-oC^X6M|#OqHGGXzAibOebftichJeAo
zaD{ZaIH*=Zg0{|q6fpkHmGIzAV9WzaioT???ae*0%Q<w3&HZPO4Wg94iG?^3;s{CE
zgr;rB)ZavN4lWrj#g)uI=gH3Ovs7foXH31{t&rN+{C8f?Go^LV*fe^HYvGQUN_bi<
z{^P8Nxg1QoizkDNEyQj<s94DU%;fpB{`I)3uRWGx)vh^Z4^CwA@)8rU2tuD!#B^75
z7h5q%Te0TM0oH3Tz<`ld?Z({BJ8Y7N|7UEIcs*3K8#9(G#t&>a9^;$!f%!w~Gm@`_
zMB;2t#6Hrx4on;d!`L1w-U8P+u7oTVpxIK>P{N+uLOY4tEwt=r>_0t)*{wMOK8+#-
z^pW2@(j&djdSerNCD=NHKpcfTnv;NA!7%|R1I~b(+WZ<^;4J4DB-ajTkQlmO=v}ZK
zAMb$t^>Ik5Rvc2)$_Y{G!oqmmzHB_6C#hz&J@6U^G)7tQ%hB`1HD{qRO$H1U)BXBQ
zTOP;~H7T9bH&6)FPXms<>&+elpw7>f$8->23oqHXgnl7=a<c*UM+2>K1I9h(Ds%1^
zCk6o{E7x3nS)Yp<P#`asEYU&=kK4108#zq(EFV2i&>`u@U+LZRkUY|1pIwX{1mGQo
zZ}TT4N+rdG3sazfZjJ16r%>e2XKuQ_bAi%k6{R^h6G!XV?nMz_&9mTUG|`W0J}*cZ
zTUXRC|4JhBWDZ*BTXq43qqpauD+a_r+mk}Y;k0eEi9=`q{TawTfA8(<*{?(eUH^x+
zq>(g{gE-`#g`iGwVb(V(yujiCd%87ju*&N7Qw=AanYdJ9!yQqcGG-$U^cjg$j%`)J
z8H`+6`DhFseflnz`9R95B)?(_&Y@6h&y4SA@$lt0M_+Fdt;qLe2T|8@4W*f&4w{jy
z&(X21>nZ<MP-2oV%6MamJOvZatD|2FBBaG5HYX1+@g=NTX;+S(ttQX2on#Z?jOK>$
z<S3bEELlhS4rco9AE|*Rx`(muCl8>N3I}Yb_@S*7^KRK4AHFqFIe`wY@=vwWerEkJ
zP5mdFSF;f1PmcMhW*h}{KhEf)k<2(K>78Avj$$d$^J(mqTcxA|ye$dCuSm|w;z#{1
zwqZ{?2i!TdMmaL}KWv8dEu&SE*O)8s30pD8mPZN%q0n+jG$PXMArXpT{HS%x7p0c|
zN{E9|p_8kW8diGo32=&t^`!E6xsCS?kzSqlkzO<iK2T@#d&bvi-_#2uXc>geQ?S;5
zIj3^^1JT*2zjI^ndnymSOD@l&cFx3TC|+CZhU5kq&fu>NX?o$msg@_e^a2F}u0k3A
z(k@#?RFhyX{js1>EU~a;vV>A?%;%64)_sJua1rueyYB3d<d+FJcmGV+cQsXBp(-P*
zV+~0E12%)h%%y)rMXt{Ck3ds<dzpK&WlMzi)fye1Z%GR^Oz8^oUKbZEUgnPKO1L-x
zjJLT{I4M{AZ*AL}qgzw)y7nqd%8y9OSkr_U7G5H;*+ac?E?3KBVh7tDNXeAXHRfqx
zyJ!>B&;jve_n2a_6<2zauYtn-#F&ommX<q_@ARK$Xh@i8%Q_n^eG02=R8&cm@AVVF
z7$jQP*KJs5=kw0bkh5UGHU7;I^(nTriiqHbq3jpexH)no3$_-OtnexIqkK>k=aWeD
z>&Aq;jFRl4eY0*(cR^*bY20iO>jL11wy!4q-72PDrO_gk&p;0_&=4g}ubS6uf$)0U
zpD#@GLLcQVzB@HkJ|0&bj9ajQ8&LHYpyF2ce*F6vEyQ5e>%~LHBqlQvTWa6TkS0?(
z=J|L9AHVS&A#3L3haQuY-u9_h1~bxQE~|LeT9kxC3Yty8eDxgG-DE)6hM!pw0OS^C
z(7;1e-bu(b#X%)SGeuyz=;<6&O}i43_TNiXNP~-%Iya9FT&6n%Z&+yQLj~*<uQKZ7
z&wC)_I*iaH(!K|?y1DXYo{$k)7HwWpF)RG2@Vi&$xCpuXf_v~Bv7C3KK3Y|SLj~+B
zQH$swpFDwO9?n&xm=EuRaL0dOk7rv`3h%R0TKK_knhVBySU<a<JrT}f%5VG;js4wb
zuuuGy<vG3|;TsXFEQCVSsf!Ea$TdU{z$6oNxUjpi72AI?TU8K7v_6cR(f2&)I%ob0
zJFIeAPuqz?>hT!SN%RXTBA5+i?t2dF*RVs*X0D9kSk`E)JmL9tN+9@jsbIOxk)}aR
z2G=J4sc9v>(X(=xhQc5#?k5+;s4h4<8k&`^6q~iqTU1SMB$~C9{nH28fj7m_g34TU
zI6eWi(wGuqYi*g$*l3<pvAED-*^mCn0-GIuAPE8(>CpnoI0cSja!L+x;rXXP%QpqB
zTR*t(1@}++bkbC6;%d0IE?9ni9Oc;^+{XSdp=DD7+dB$^2c7qM%4&0moGgUli;=&Y
zA%FQGli{M*v}eDp{3w~Za+7wva8$X04yA+B_ys91qP3|n&G+V7_5RarKAq08rwljw
z6fhJ9y?@Yuefh!~>t~MAAmzo}w9?Y#!BCwJn-oq}j&^`bGf_R}bv0jbA22QbkqH4}
zWh2pv#NeQ#e!>`V>EPeP6mS__&uv`ykU3Cj49cnPLQ5hVld>q+l&!1Q0d7LqoFVsA
z^3*{diW`+@&bcGMJmO`ce~0%h4IUehgF~hFNl-{WD!up<rphzyIvsX11(aiS_4t@g
z>y?MgaePe1<eNN`Jb)!yYR%pDB=!1Qo1KlcoJwVqgfG9!27sL@kaSKxKI(90;gr(+
z(OS#r7%xRmxx&sUu;-KOEMWLZ=<tKZQY=DM9I!>~9UtSaNU|;i`mUFYxLPFBpnJO!
zYt^5Q&Ylzk00XQF45!VD)dj=s4jcLjc<?K?%7JeKsY4LC*0!!5AMrVN>>o4;b_>pK
ze&Tu#iqgNWcYkc3o`%+E-x_v~yJR*O&>?@T`DS%Yc)+l7mbtQ`BWrx}V#zQ^wT7@P
zrnrTE(0{BYK#4XG)xMMYb`Bl2A~%#c$p$*HlrMQ!;wQjVAZcCG?BKUtZrgd8pzMCx
zbLj%U^*h_d!~aT^j5NiSDup0qx41To|N6|VM1$Hx(APn!&vmoOWb||0bu9~fHjL&r
zsH&-+j0B?#X8RmwwcW37WkD<TjDnS<Mbw7lk@EG9@#AH!4$qE%vtb#LgheE2-1zUK
z)S;%Y&t5Galk5%@mW6$8daIS+uYPw=o<GjLILXh-{u!=4V;dEirmnkP7rv-NN!^u`
z<rxLpIDzY`C^f87Bj3}CcV{s>io2o@1Opp}zHpvlUAAx^pt(Eo`NAEGd`!X*sC3td
z2*$QOFT=L0Uh9tH=Vs?#0*4(n0C6m6qVF)6ab<i6Jfl!*1S|7$wF(>yNMr=ItEx-;
zUhxgWxpM9nRQ`@)pdX1L8*{y3M8oK>QKKzp_CpNCC!UbxImN=1*rZ9E`IWa<>e_3%
zZel0ns8@Mnb{I^p;)4s7Qan2o=;8;Z;3%NYAC^xacx2k^%3AW9ZnQpsoMWASnx_ms
z_u^uU+V^4P!1rdP3|L2RWlNhSNn8gJGv8Iqo#k4&j&7~_Z-8%h+zu9r_sF}-dUIUP
z6u@M|CXIvmA#I`_9rrdun(h8Ma413KX-|ES*>$8vFG5{>WCYTtKSGG^2c;^Q4kHk(
zFP<Wfw$1c_Y4YH}p|z`RYa+`g(@ws9W>xo_(o=RJ)SgQlE_OyXwIyrvT(i!9mIg+y
zi*+Qj&|I_NcI4-T5!d(5SHBniEoK<!a9NS?YTmbwZUm0pu)dRROZilS81>yYA)-j_
z<;pbM1)Ja2BDs-}?`o-yaxyDg_Ubmxa?zvg$@LVLAsS&zGeFjvem~7vKI5nF>jZ#x
zhT>SUb!rtGR*#0au%PB<0&cUDbq~5R6yRcq>$8_!g;gA!s!JGRO;zq0v)q7TZ}<3j
zkxQIhC-gH?^@(N`*cGglLRqmw?~Bo0a8glHEcFXE*?Q8swbV{Lf|z&@n1X|mbq)09
zmQopz@xS&e+Aw<E(uwPvvg=9AvD<Re`yy&ujC>q8y9%pL4(;rznE8VV@0~f6h<Zs5
z|9rdn_1n`zziUj1&=r>qQT;uMaf0x&<dDB1Y>|iOc_>^M#t+E>k<)5Bo&cAg_ZNqE
z{`aZ(t?r%q>wx^TZ`>K)0-&s9L4iUB6uFf;?HWx7b<d5hPO+EBjK8txHc<8>Og0`1
zG50K^=6Tkypo1`MQG^sKuhgy2M@x5Fws(t}kh#Hgv|q2^d`~~9yWN9!rshrxbX|*J
zk-qioy4(;2DZ{GpRHHGgWBW8S!69WHYJ*l#t&}4+<H@9)$jP4Wi5Jl_2j|={6c1$!
zSc(0{w*Pp=+qKikfUbIz&2CVFTyPKgfJ<AlMMPB5-&Hnfu4$fzmEXt$2h24)I1zMn
zkrX|5-TVF3gBU|4%xc75fnW|X{;v;C+@+2%95B7TcAIxD7yiWEaBkH6!cb@m<N2cj
z+9F(4hnJO1k}`q;OR%HY)Rb0}d`!_Z*ah=KqJ*vV@8$ZIt%N!Ij}gb;$rOL}VNlt7
z2R}FrtgxK74vHoGf@5x*&JPhmNJNOe?++fF;d3}dT@{{l!`*QEB!=0qHi)1k53ok+
zxQ$!2m2p*YquJtchcpquL;{byWqhxj2z;IoG0Ah`G$kX(YgXi8iZAZhd!!XVFvQN$
z#`XlrJuRaW^K1`s*5R^`Xmk4E&<GWoI(uu=b50N^k~SIte>2en&Gl>`PXw$Q2vTvm
zrWXEoHXRPSHsqo}ly=Cy-59`_T-9%0T>(#CoSrEpdFzR%%|&|~XZTRQk0LWbM#IQ&
zb9*}~I$B2H;R0H`F3aRtk+*;Jtm`-NAVi;vtn)mmO;o#Q^8R0Si1+V6<~GsFv-JOP
z4#|g`EB#OiM!ChD<*F;SfP~<nS*1RUr@@snpR!a~%s>YOi|mHSkgn%c>hqCeEtJ*p
zivX@rERWji>MGQyHv1Tn1A=&x+}q6jCUy3^>aU(t)=m4z_2RUBrpn4{IMw5pD~!@Q
zoGX8??xFzV=Gh+nRFg4l1~B!>=u7@$-D6zP(7?c!v(u_#0Iyx0hTLnNq>fvzS{Avi
zV+M-_(C+;j{dmAvF;556PL~<^84t*+e-AZHHM0aRgF7A|pc}@|AZ=IX$`X*lb;z9d
zFy}RPjZIRyOjYNp9^%BZ)Sh5X#X;rz`e-a63gjd8iv+*M$4x=_D;+dm)~4R$-tW4B
zeajSZ<{vHNX!q<RO1GG5&!<=Rv}TS0p6C^uCu{4Te9J=pf=l~T`gGZ8qZ?n$^s`zo
zSj6k~IX6@{y=a-io{7|9>$B>N%S^S6!e6A?3ctGFl*VgC&7MQ<SznycRa)*D$TF%-
zlrH^ByAURkF5=$qCQWpVPAi*Wu(9wVHE9@!7>!>76^!ii8>V?`q3fz?F6VBH%FP%S
zZd~=k&$L-hF1%(Crc4KgpnDVw(W^13-$NOE2ANyGIBMvSfG+j-d>9PSZ=c_E|4dn5
z+5NTmefefn_BDvvmxK>{K9uzBp{!iisl97^FRQz61$Ti5bTVzK$-9e%BCkB?YA3PK
z{3&LSZylt&uqB5k4(BnbxDQjPdI3lDN*yAKIsA!6ECLg+pJAhs!){^oYv1ptlp9vs
zVR9;gv_`QwhRh#oH;gFj+qxm>XgGq%GjyqDAIw+@P$0Fx@H@eQgKmtI=v|{!GKt+-
z=GXA?qyhz*<F)cwG!GX$q!;URQyG);!_fo@$U0#9f7ank(n1eiZtvB08^)=FhtG}0
zH=|(~jKuf5qq6ffieIDU@)eNTI5*C1-;`uzj)E`pFD#kNJvQtuYnU$e#@ZV_VBLkB
z<v&Nnr0BA5DGC>WK!w(<s*)oJ{8ddIzJ5|`AfScj7Ccg18oR;rkbt%ArS~r;e77%X
zqICaiJ8a&0%ol)ITnS9{>Phm)iAV{><Sblz+~K{&mHu&AUW42WQKVAZqnQ|BexVX4
zF%bNSlf#lV>7Dv}?LVtbhOD`%mb_L(QQLhk(72#2Zs_49kEkyKG>47J(#PQ;M|A+<
z`_W2cZ}tMKmTZ3SULZx9IpIEF>NMcW8e5s5c&u$TY7Ikke3;y_&QMpqb8}N5hD6>9
zvr<Y1r;wP2O;t2Pgu(-&8I^75Y9*fwGH)-oL0M?Axv~}Mw6y}s?66n3eVMM)Zt^TM
zsoMnvu2EhRe!jiaT}hEkRNsrOX8;oqC@V5ByI-uhz!mE3N>!G}sS|gZHQi51StmBX
zkMVOxYR|+B#D#V`#kyyCJ<&@SjMh;MMTXb`JrTxi;x8DUl%SfqxaJUnuiij72o1K6
zJ^hz5-wNWL6{6(k$iO^Dz%2kk+jxDQ`jtg!!m4f4s^3?!syvbAYPx9yZXwD>PnX8B
zh2qO;vaDiWmhkAD7A#6HVC|s0po9HP$E`{OZ)=VQivG~JU<og`P`_3=K#)n6`0GX~
zg4TRoxWro1AcTt7vSoINEG)X1Y{kO~KdU=H1InQzZ?%k_#P|e_&U;3z!Nv*nS+&6f
zZ=2|JEuV&YIL+UmYF{%J+pp$AeM`3zlgM4S$omV|KSZ$KW;CaEppI%B#isO?P0%0@
zS4u)Jgu(&M=%yol*@M|?=dBelaE)iL1-6|xZ=kjDx|4nIG_I*P5b1Ls-)0E9;8t1p
zeRyTd!K&_tArRmkET6%?FU0WLhFu4>$q&bEhHH28_kF(!A*#y>=1!WjN>aNibzb=|
z>^Wy^Qj+$!I4N|5>gQF+A90Dt9aw$%#<(7)o*iF_TIdo65Yr4O`PR-^HC2Xlf753=
z_o;uVkwU6Es2G1#cG8t~=k$gQoiw|^HPVexbs!`h!lsj*aKP3$YT|6VQBz0a`3pM*
zNAfDrvXHc$^fTc*M5glQK(dUq=^1S7Jcd7&<)G8jzP}8;7~3i8Q|xL18nXHErP6N2
zaKWZ|A#FSA676hFwvN-9gMYa>-^G~j#TENY7ar9vje}xOKC>r74k}-O)5xuxgkDP{
zqiI_ojW{CJcQq;Hc~u+S;?PghlXGAIeU9p8w}%<b{VVF?@+&lbmaFmC*$-VdBE*ev
z#JENfYW^iOS5<_6cihZ7Z9h%0Ds5}}C}f2CObHq?D4wdga<$Y9r>3kjGk;g6%K{{c
zMSCs)KcqMIIIyDWMV^28Z{mXmzl8*{xtUZo>UXo05>wn7b7Vzkt!!r(mw~f0RI?1g
zVql|fyQXXkUxrSyT{#4bZz5hX4;d=Nam^{Q)28LwXRuqT*Iie3;QI6(!m6#~;_W@?
z-O$_2XqKwP#R$~<VSO%8u}zDcqydrs(Q{MPp#S~5*p4q*O1^Z!>iO;|sgUMk-6VJ`
zt44p?jd<)fY0`by3r+eHA?XCJuHLHT2Ui^C!R!4A8GXKh$!7&YES>_a2&5ad9^n!5
zL*?BKe9u#K7ilmpK|~|x_F^`twUcht<R~q*7WxV1T!{3qx^vrLY5ETVP238uf4!%y
z+qWLa@LrA=x<w8YRm0BI@&8QrxuG)EaM>Qm@io#cFf*c#66z8uE)VoiMKL6+l7K<k
z$q~C4>^(N036j9e1%5)|pTY+k2InL>aOhpuh5|EJO)-ico?;EfE@Fcg(>qn@zGopE
zX%RmgmN!)*^au>@_>9YG_jN}%i;l;d{{8l4-XqmNfT_rY!)KEm1**Aw@UT4ul6ZXL
z56aWkAAC;iMc`-MKoQHhC$uP%nnuw5Qqpp+6;T{|c$l1_%Xd{+OoMzyPEF6sidFFp
zr?`ANECm6{ma@p3tRWWDpN|f_`Y%Etutat+Iro0=7!U1`>w*O|sCJU+BTq5DvKctD
zaw;|wL9ap5*Gs{@`4fY!vo!Wwc)bIPX|vsSA_o!(zun5f#*YDyU)8p%FSGHLkBc6{
zVB;kIS*E%t$(7nQ+yDBXn}2-qI(AMrN+Bq??*WQMMtH-O4jF_TxmS!u5ooAbh?6EM
z#c^4r>Qdqoyc@LPgVB^(Q58u0cS_i`Q*jt@hg9bQvMZ@cPOmns_2|eFR+MubdAQ+p
z;D*#j)_21MENRNxY+ds6;9q`s1{cd7&xa+~;mpQMUAM>{0fOucrX;6yw$IlduI#UJ
z<rqJGivvW3yc7<#SDQ=Q_3AqQ{B3?ukI@^o!NF_0Mq~ECO;6`gX^cHjlQymg_^~$p
zVDFE5-{*??b~oGEXHChS7pVHuvrCUOYnuNxVGpFXm2T$5Q^l_h2v`oI9JF8CwwHvU
z`T0vGe4a%+PbYlYWaFTL^fl1JPkKca?%v$QXk-B<(vshWCT-uzcQ_f<c+7yPq;9q5
zHwPt}%)yZ0vt~2ppFrj#E)J;n_+0fvpDUx30l5*Znw&IZ+H33-3I<A@XQW+j?>?2V
zOAMh?Xjd@QHJdbP%{1$PR;1ls4oaq8^QG)tY*JfVd@p)3frwSl?#C1D^$tI|#D#J5
zp!wx`U1=%$6Btfz7QKhRdwU${q`8<K``aC4*#wi5rHy(rEAy4=HY{n|8L~1`wWs`U
zYYBTi(pFS+AAw=e-odIpP(7}B&Df59sLY>~%A_0LC?a3o+CyNK_@F0rDa60~L?Ml8
z<-uo1&dK%n<te5eKb5>(qHVpDWDg`dnjZMm9$q(|32HVvcIPWJ0K~TLg3>}z#R-lM
zd@_S0-wd|bO~7@16ZM4&?*ct&GO{!~`^ipLBwJk?@oTahKG<y3tYzx&Z6GQ%_R;5i
z7W!vsT-b7=y(%KNe_cn<)1a4FokHnzs~xD=h%NW?6M>7_78=1JTeGBS{*{gQv*$~Q
z;Qc4Wt=Dh7sjg8N{ScXPbLEG%vA39z0?BqfYG)Rt?-FmU;#}%UDsn`u#K=F$p&ffq
zBi|bvq7wF8!4Z+8=ux8{cYXdXt4F1YeDdmPf9yRpFFa4ve?TS-5GK7I9M<B+mD_y;
z7-_zku<`(b-GJe|chYzIx7+R~jh9=4R~*6aR~|V%nZIJIEqpUPqzZGA7-?tPAd}(P
z9&IVqug>hj3z`z_S|yEym%KeQ%c*vqKxY$B9!-C2M~~Tz<T&&moE*AIC;5?#lp?JX
z;>&(HlFTuoJ*&XM7=09D%vzR7!lo6txEcb*yI>SN<Sm1kgTYhxh^Wbt3E)*`W>~1k
zXh%OP$RkopPtlBi%_l$G66^g?;(d=-@*Nz>KPa=#%V*)<qeocrJn@Gv=+{j<$=|gc
ze%>>r%ia5lUpX%H-qEN=6vz@T4O1U7N_FCq;NC6yT#47r<^23c&MGu?yFBGusvw<r
z1$UEwzZD!{C{z6WH+Bt7nMJ9LIh8vbpNz4%q#zfP!QW39GE%kkc|+P@&kI`_Qu~~%
z&01FhcA5`V$~nCqIWy-}1ZV2sYya2Lx2|){@q4Sda0G3qaFaEuUo0rQo~-*+@kHX-
zcXhn*ji6u1hkW8nj_n=$`%kzKK*uhjIPKJ`TjTys&J4KY<t;8VN=gl!2i2rQqwj+N
z7K>JjQ^Rb^)V?RWxW1B{g=#V_O**Fz3$!bT-drI|iW(-{ws%{zIK3ug8c@sFM-717
zIp(YetebXkS2S!-7?))M?j;Uai%pEx;b+w!F<B>kTpA0X5V2qk{94VuR#<f;V|6d}
zLLT{w7-XWz0k}aUc=6Cse~*oXk6YMgYH#9W4UV10HK4OA6G~vZ%#3+rAV!$eA(_o}
zEasqlgoY9dg}caW&UHQ({@bi!6POaPaq~bMY{i67h`lQVPbd)JISaUv2ncQ>KA8Xt
z;E}tYj_JFeZfpK@`i9uFRGxe&D;S3@`8n}<_5mHm-Iy+kSyg(V_tG6va=Cq~pXv-&
zm6%>_Rg2!|_0JTT(vG)kh=sS=1qeV}`1nU3-EX)N<+-5Tgim$jAe?h@xD!u*-DLHS
zOMp?+HC45{VUvsMQ_K)y;2M#za)iDVkfg2qws|^?Dj#VRjr^%r*!OmFr*b7RLdxoT
zRBWFPdgLF$wqOot*`^6cXs%Po_2IqwH%YiW(1RQ~RRR=rie|{{v7EmGZ&vw}y`h(d
zJi?RM!1hb-)3}&RcFKtF)gji~k>Ch`2hQWmyGjGb%etzEhu`=;NsFw(F&p^fsBaq1
zemPdMxsi`LL2iY`&j~_aW7eI!`%jq_V=^5L*y3P`9y&_H_Tho4s^O2&m+M72lyIWc
z_2;OBcp0FMz9yY&u9lPrlVx@3AA8kbrzqA?*Nc1+>cZ--Q!#X>*h@RzII|LQ;HkzG
zK}8J&he)I2(^`=pVY|q`9cT1J(O>-(mTPlsU-TK{ctd;%;$4^=eAGGK2q9aOm%_d{
zu)k|4U-ax4^3XnsS-YBMnvng171o{YD6s(`A6%>1c`k3;JhzXiU9S=-xflrN(>BNb
zNX4pn1VOHIQth$~8yMe7BI4Tnhv|7gI9^g$HO|C?0j12bJzs0R8rU&3+WA-%-PAp_
zMU(Ac<!)pNg=2s^y<pB+F?QTZ|L``poT?1=eWx%#+Zs!nU(=RYX--K7aFmuS)e2`u
zRou!Jt>&%TCRMxX6(@0p339$puChvqiyTy<R-)>X0BbjksL~Bve83hI9Z4<`EgYHm
zbCMd4@+W`2NKwP5H8p%j^Tj1wx>k76zbj}KjLWJIo9XiLzglb>mH7O(y&eu}fwwmc
zQvXSyy6fTTq~`G{3$LT~&;Gvqh;u7txsV?3$Z9Upz6-7HxjpZwl(;!evTovm^msog
z&JOY4`fMv&Y=I33_+`G_2ks4KD~%ogugY*WliqV`HBc0>jFJf=mG84tk?jmRhd;EV
zv{w-5U-yZJTsLeN^G?v@Fn@&L%}GGegP~r4f`xEW|M6sM1)3;wZ~{t)?`?4R^W)TG
zR`=8WOW)LcSek*~K(m#W1FSeM4YGJ5eTws{RN1EkTSS--9SyGX5}f`XY#K@~^L5Ju
z_`|4z?Ka}#M&5|4sD3SDz&PZ(YG5MR#F9>Qe**bUIe7O5dk85Z9reNPRZ=+RGcY#0
zBj{#vbmRYWLBWtPD9ss)IQk435@I``q?-vl^cI19JbOM7bJ2u}>LjpAC+Z~XAwiTz
znGxcR84??Ow)a>yW4dd<!LadYQIO~k{g>vAuv!9H=c0CBC#7oP+Yee%`6a|3Rj+&T
zaLJZ33TPu0*DcxIesZ09ArP96qH1Q!v^3gD%I}~96Ml^r*z?&wADpoU+U++pt(pa(
zRj#$O)2vBb#=b!o-9?i2iVrW~xH7+-D)9XEvctvDRo1)GvAQpRxKXUETHVGr0*qpk
zo{uPw=SPFn)BIp)B33={U7VBFf8Mm$x9+bgPyW&8tM+e7-;;FEjY=>hkhO4K4pCXz
zZ^e5JUYT!g7~`yFmG2A<g5$ymYtxusJrO>To4Ll#_?72hJ`j5`r|N`flAa;Kj!|RL
zR}p*Nb+)4$t6l0zG-x%nwHkSOjT~=(>b5h%iH2;fC)Tkk>wH@*-=cB4K@OSF#l&Je
zo(=&`^NS{Ypy+3%mX@k!6laJn?ti3_Tc$vy2p$rRAsnT!ZMMkUeT%@dpOT%U_5drT
zqY*M)PO4mK9qkl=Ptu7Ox#mg0tk|NjT-#3mH!n^*(;0{{Un-U6*ekho)W>k_=IZqw
zF1TY7`K09Lk`Cb45srZw)LmEq8_RNpR2%YVQluSsZy-4kR^1S2+^__cfQ(GsqK@r-
ztYRz0Dwx8H{z{dcvHK+xdppABV+vT4X`rZSq-3)hL8bpvjYpcc@<hl?(?W^rs|t^I
z>Oh1(ZP-sd6{>4giHXcP1ayD_sFLe{5sb&A|6lSus;28HY`)p3;Zju%K$qznIGE~+
ze)ugiF#aI^tEZnGH7E2|LuKSlV;_D+O(vCq>KbXlb&!jxnn!>~{%cT+UN<dV+KVu(
z{*c|(-{7?xDe2|E_>B{O@87CaGy@iS%u%2ausX+eaM)@qxcA8rcNo3uVZ^^B4r2gl
zt?ZnfB(uyVUy`wIczcIbT?2imIo)Z^!;S|*j@Y8{ohKELJ$DEsHQjW8cG>&V0nVv~
zW-v0$$B-pGa>%@)P8C(ySFR7SxMV?3W!ds|^{eIF>p7cixX`-8oLB^P$@%$n5+8e=
zNFV>XiCYPRz2;et$HS!Fo-=ccbP1Hhz%AKe<*k2o=jw4lNW|QQJifk#C@J?$UpGyS
zipW(iB}EL3?vQ>#%*)##7g3%psu5HtdcAS^zJRqh<vv8Hd&p;nG}@?L9kv2*C;AQt
z;`iMR<s5A?vG%&&CFd<9$4@5x=y5x=kOnZr_+H=EOw&1fU1nCVFy|5MH$**l6w=9T
zr$nBZF<1ErG8^>xbR_W)^HRUWdbrw+O-yhsEG*2#x0cd;Oxi;+?O@9>s@FXp#HFP9
z#kaNnf|=ZY`l|=oBsB}g0Gr#wij#gzqtSP&o%4e!%>_wK@waL>VsO++Y(uJ$A!k~k
zU`mgBZt1n9UQo4So6sj5`VaKrP2O@L-6r1~5Gj3Uah%FmqK~|P9Xv=$usHi5yfO?J
zShMKY76JOF`UZ<zUAwE$iXM5v8Q5sVGRq|d#Y%}nU+P`Ow)PRh+nHhBUK3Z_*()sv
zw%@9YsW$H~5k5BoodO}HoZ#Kv=H!d8-G=<84}yoDMMKoy9|wD`zq%Vc)RK>V5l~|U
zop?SRMxYCN$0yI3IOSRxs?U0a$Fl1S&ZDGI6Cm}MOPyM+n+Vs{AdwuW-DpC)8Ey_y
zF@2~iLa5mU2qQx$Tn=8>`=3M74VS^j#wnS-)11V1Ho!6G2v$<Z0=1igCMdJK>@^wU
zYQcp+_is<QbPYBXu~mR7Wd(ptY3vl!6nASq!8K)|xbSHWl!FT>3JKH-0-J{qF%0xh
z!buMMKp_%=FPGxAKxTyM1wK}?BtkrADc5}{1~XL|4VCL;62cVlvTV1Km=Ro01GJ<;
zFN_h{+uPIrCRkX|v9||$nfu&=>`pS|*skiy$6V|kzV+zD;5w?R@_^tPN4Y`2nKO)7
z>9yt==cLAtjzSO><z!QGWGO$*XO7FlJoHVz)lZR>5VTE!&fa4`8G+1$Rn$`aVM*<1
zmHot2gA=d;5-TQEshy6LOM||WT5@?6-v5CB@Yh9OOQYenczWO`N-xI=Yk#KxJQuHh
z{9?Z5D)++{0~}S<bqiv>6f?7#b`6geyy<1Rm7`-x1=rr^PPRIjDcl9gA7{apdeF0@
zY&;Gzw~K7!COw+nB+s6T_78rgtP0q6VlM(56>61IN-5Hj1BDtfqOM4e{YjXp0{=WW
zqNd`w;OJ=`l+U>n6FS4y;K=E~uPBZBx+TS3XT$wjbl96a0c630UqY+ZUqfyLgufPZ
zwG{?}N_KhWK|-2qYyN69F{V?*^8W(xRrt%>M-1<#X5;5tpxXQmw=>_~ggLfLr>92V
z<ahiASkNu6tVff}6dny;9o||19(}@r!Fv){O;>R>lh?~;Oi>a1ylWiz!wG+t8_mv<
zt;9mppK2(FI=inY`qV<(PSr(rMCuik2M+oT`c;-EAAmZTwV*U-Q!-%q#hDxVe9)m9
zghW#HBjcnk?9BbE^EbLoa5MJIfj4VaI=|=N#}${hd5%dpQ&9d3T;2iA*Dl9za7M@3
zeeGxmppFy7NwTIg7h8m^<%Vkd-y^U{#mL*+MIx+OW9httXyme2C1rbxbYje(dt6Ho
zZQR1AVHpXKNpY$$?Ga(qm$UxxS96Q~SzBBAnIQGm)^S5u+O`!de6OqP2FgO*H=TmW
z_cJ8rbjdB$K6*p9&i|2Ptft%adBUdmQTYnzbq9tQi>s*s8@wnsFIbG{RMeEM|8lxr
zic4%;67nPG+7pE(O-B2>Fbii0Jl`_+B+2;0b~x!pA89~;pF~-T*;1C48$)ywCmca2
zMdqLYztKT1jxd}7K6%~Uge-#Hez?!k<VMJML0iux8ui;R`Qvm^9mT(hDw88jP6`^f
zHPS4inwJX6p;Ck*38=<KSjSvh``lA^ir}Lk6QQVclyE6Y{U4Al&~t$@+W!cuWo*oZ
z8(Nud=AHU3E<O|2HSa+N@5d0Yx7!8Zn@e=ht2cnml+I9e!b<xS%daWht{?8?bA~$i
zN`9bsyIE^Ti4~Fh398+kQI3g<!;-BA`&!uEaoF{>iwF^-Cy*R)bmcSiYRh*CoW*8B
z;Tnqr`=qXf&A0VDg;kGb$T}~#^&O&|5R2<?19j})Qfl%UlT@}Xz?6vd|D^P~#EGnL
za7SN7W5&mev<`(&6WMn|{{}zyF<ZC`4?eiv2ve}T2g!Uw<M&BpA1|Nx5kWz6=?o)X
zj=S|9b8a<mud+H0yY-%!pXU%)s@1Rg)7oX!q_51?L_s|tH7K(70aW1#p7xR%=Hm<<
zMif~<ikk)*${M5pK*brfqj2qR`k_x1WAh~-FEyBY)&p=@iQFi{T#4E)Knica*uUHt
z$bIPy>&q((`QUPz?jcFj8vebH#Yw+y^TH^iupcwoV$7)^ExG6sA8y!-cenOL>yM|V
zwA$8lf4P$I2JGbXRMo#gn{cioB-Q9`1}nvLg;fSG@b+swVk*hMUu#WnwYQ`FdR!U2
zF2?T_K2zy!4jOA3uMiWoa={|?ZQcr;L2Tcb5sOV7d4+x5u7JV$JsB#q@3}xUEdT8D
zH%})Vrw=C|oywdTtLzrRI`ymKq-x@SXE->rZlfwrfVEfTiEu&8An#P$dQ_GYH7+p(
zcBp0ObKYmREJlijUniEwnI{#<nRaq&slX4N+`XlzohoCtc2A+d{hgTSz{5LO1SFf&
zln#a&udqoJ5=Mc5^_{l3(x3egb_YD&n)SL?jkHRoh2`?or7*6p10tq2r^D<PeHUTl
zIhP`Yj6c~un&h9|mEUF)P?WBZh#~_C9ltAokT~HSS&P6!n<!vzOUz9mn8@p|E^DQX
zdT@MW$@!^|Rl#V(425PJy=H%CZ@<{#KlaFFQ0IT0>fRbPK)kNk+-X9>ORSgR?l-X8
z$MP)1#1vth0Vc#tf+F>&ATz;MQ9_iKAoAWU;ai52pTd1P1bmQl8_%I-qyRtz2TX4E
z7hTvoHLl4wL-0Avrfo2Avxyu$7y?1xK8O7BBO~c`)UU$oBb&m9;ik{h$i*8kWwWnQ
z|D*2vab-|s5hkjThJ~t#?gvb_-g=S+{CA?Z>C20)o%I9qv10E&V<%}f#%a2u6F$4S
zU}epfS$9UNIpE#Qk^Mc`T&H2lT~p`!X+UEfcW8XyrDG0TJZL6qOggEIQ`5ppx=u^R
zI(cOUFeP@B*b`xg{mxWUS8>f%lA~A5-FZr1avdBPbJ6ei8aUxvC?8jupo<yd2IV@t
zsyaJq4fT!&fxW;)GFqOc$+Tu#hQ^@m((^@kA{5Gx(86vyb6%cLTX=lF7ZDiP>qy9S
z{@(;L`47)@3NeoT$A}i#<*r5aVXal^-EH;H7KAcC6kVTA-*{_$iNuJCl6S;ttSy|x
zN~zTcK%X-$O4jq%RaPshZPiQvbigY(S>7SHBtQ2&4=9A~@Ge>bQ?^QysKSfBFzrG+
zj&wp;pEi%fiZo8rbHB79vv}ENq`p=Vq%*8o@H-*bP3wv`#`qW6`=uGt*LR{8ectzT
z*!jahqENYS9FNzbK~;i=B5_aR=n0e`m+ptcY6VR`nMg1>ydGM3w98~Ew1o#<6Zx<B
zwe{?N4*K_3@7B5LzJTOe{VwX%zMC$mS%t$EFL5*h4!B;Twfi7BJCsvbku9GEETP(|
zD9Gsj*6Q;Lq&8!Nhx(GoxErvS^ZDOk!_q(~2U1c`QG|n#v%11X9eWp-!kWh{+cSIB
zwyQB<1ZYGY+2!Qc{w+=>n(<76vM(*qjtWs+*0;8GwH+4Va1OIe?6vFpAPXu=a*E4}
zy988e%c?z=mz3ZzSqp}^xXa18CgPScO6c-ZtkIEBW93gNaZ0r$7JjC&O^#-zaBRW{
zD)qimlfQ?zDL*Lpbiom@6q3DGvHLxygtm5>p_WdaV_1IOmmkzzdyLcdskEG#y3J7$
zmX_xR4Q)O0sDs@y->na1onH<S$V=9n7H`AOZc$qt*6aOUq*~TL6*?2&JwtpNJwCV>
zsv-Gtr}GR!DZnjZyqYMQ(wJi1Y4Ryo!xIy<?K&b-Okr2WQXKUqnO$sd1?Ro3m%Tic
z<z2iru<c(vBW^>*NMV+VwwyidoOW~hvzEX*4V_F-NhRVGGfmNzg1PfYjCSoyd)D3%
zpL@*~hs9u(Xw1@?;()K#R$r+(t>!HELmWE*#XM08Z0pPc%rDE<=vu--()|>E-_laX
z-#vV}jDsKP>mXX=wMJARr=nGDf4L60AUBkL^89ee*1D-fzIyd`kuf)9&D!o1snE?9
ztimx|<a~H&({i||-sxGQ6&6m#!QP7*aV2aL%b?G(0qVtd0d`Va_k+*dKIDzRW}R;+
zEbizuqJyTq6o;yvtPdk)a9KMlgLe_kcr7Kr(rmksdfO~)CgQ>+H`IJyPBdkz?&y6L
z_$ZqsptBIk=8af)ZX}IInsfzr&<(USk(M2;zbI-X`1vel(w0{v$x?gvn_64M;48b;
z&(0*mFw_NvAL+(>ujEtZQAvEIZkNL2<W+V{w(4+vLqo{&3IO?{EPY(kz85M7XAlhq
zgzR}X9P`f;l_b03*_G$@NfFvcr>#0R5dvGjh{-1Wa%y^B!ozqaOfmqGeXq+#_6wNH
z>ca{lkQAtP+E8Gy7>8cmu;@T2N!owsg>v`6zlR8@#+nXpnV3!%Gf{A*67L87QwSlF
zx~f9)LAW))NSsYikKHtZ8b~nUjOxdS=50GqpE(TLK->JR>o7a@S13-%*7Ou@&RdiV
z(WuPWI^i-L<~U{6qss$f@iyQ)7Q?+M6LLX77cWKsVMk;!x9614sT|>G-KRt|9BOI>
zMPZd18`<HDe-G2MZ6D@1ix6Fyetrds&RAzsS+LeweobtlqsvP|s}!Y|Q&aQHBs<`<
zqhl<?(fU@Ig=&m>lwH)2TA!&b2XN+9CUKFq!nG64KK>kfIv>oBfisw}K*lycf7!2a
z0J=R(&KO+Kzgrl>iGlKD@1ih@N-9QJ25w>)@{GpeAgDd)C>=3J(AQ?Yj={O^1iWF7
zzEBRgetE+g(8XhTgQFW=Iw{g2l7dAL1bHQB`Q15Y1NR{PMID{P^@5^qdF$=(m^f6V
ze|2>A=GNEiTmS4n&D%5|EQGcRKdPozLJ}cs>!^$HU&ocoDS<b5B;k<r>3$kem68_w
zJm^O$#T;9u&0#ODqvvPk7|P9y8>MT3Z<&UQ@|Yb%ri==CxmOK>FELXB?}6w7zP+{}
z`24=RnMgf6nla<|I1iAUK?O-m?zy^2Kx|-?c(te?Y7&$@QFM}p`c2%SM?v>7L4haB
z7m^xw?F0O(U%%lSmM~A``mE1t<!>j`mZka&Sw8IX%HzV{miFj`)9yY2eW>B_lZwD@
zZU`Sx^2t1#TV%SPo_LtnJIp$(yxxu}y%LSuQwy2@OCJu+!8iQg?pZ>j9vaI3YgCs*
z_7E&2AHjjpVi$Wx*t~+Hr!vQhs*$GQYKK=r=2_qIw*nb=2>JICN{*GPc|U@}LHbfk
z_*Cvy>oH9!W!A(e)i^L^^b(SGUNX@S4`q~<79pD9Ya}l#N4OXD*|Sa5bvQoU3O=;M
z?EN`~07?6gBF^78rjR+8hp}l{zqNH%^!(zoN|k2$cdFX%V!ff{v$#`eBq3E8<~p8!
zjNTJjh+P=4@)uokpCGv~MOhG9krG$o+__%Y<l~QP)CCO%FGp|bT&pYcvLzbrGRx1J
zdp+EZk=~NV>qYe_kw=NG%S{}UoNyFNF3N?z<3anK6sQ`~;}inJtxaTfHoc0bqthDl
zD1A2ygjK>Rghpg$ebV(aJVrM^h*9f4{YYkbwuT}f_360stL=T?SPmMaEVWOwN94WF
z5?ae?lu4Yy;^ZmOiRi@Rs?NP6%Hw@I38DF#^hJGCm(sB;A^)TAmp`gxl(f~)f%Xsr
ziL44^N7Bghjuw&P4o?uP*FA@wJFoox!774*fch~3?ch~U=fV1`Key*$91}xdk+054
zUdEwfd@(u_bZIiysPCw+ckjmAo`GY7w~S4i=E8UCuL_D+JHD|^nWb~Kgnh+2`7z%S
ze*IUcQ&%NU-{;CX;D)9#jOYy`OmEwGPE$gA1yffnb(>H~m+=i3sJz?1=DA)HW56PN
z5*3w|2~JODORQ;l4E~*|n2M#}<xRW4W3NlaPFv0b(T6H=cJ0bf&kp=uJDHE@m$MZ#
z(S}@E0x2r7?<a|8H9#|A)0gqdr%ME8V7Zx>_64bYd&#Qe!G*;vlUI*>^TDo4h=J0k
zt!s_3g~1{}9!p(hN{mX+r)9>wyRi9_1>FGP|JnVP>HF2T%40L7_as!XOYjen^3+g;
zKG0FL>kxO^MA5Hb6g|CvFdUS8l{|26oGRAEzv=zG5G-OWG5Zr{l7nXLa{bxjWgYS!
zNcwt+weIsy7CnG6GO+)YBpQ-R^pqJBVyjksjPv}4zteqBuW-{>hX8XwB)Ig3K*?p*
z;$yJ57nC?`Ag2hT`f_LLHEGIu;NF@#t4Ai;V@Bkc;gXRb{TMu{wVvHQu*01@UZiC>
z2mot>gp^X?AXb?lEdDd@zXf+co#eT6>MBj(+Ae1vZ1{#WF(%h<_W>?wAHgSrr3^UT
zM!AU{>iDlV<O9PEzsbi?^iGUbt;ft8fN8I;O|QPA3m)5Hs-{*S*Hj;OwK~<*4Ej5p
z4g+w=0Z#f9s-gyB;GS=oQ)yIYSkSppREO89J_J#_87py<mY>AcbAR9yPjvFuh9ZQp
zr6i)7yB9CogGP^2Nc}IrvL-7z*eKRT+(vMc^7efsO1wo`Ash4vG`8~eaQRnqZy4Z2
zs>m<lKvUHtgZ)!UnW=@#3qf-pRj);uUQ{1@*9_c1`*AqNNLjvoG>T!$xCA%WEw1#;
z((Z3;_S}%x!cZVRi=Ezr@=&Ww&HFNb@{LV+kZ`mb|2bl-{+F^Q#1X#EjhdsO{EPBK
z74~9g(CR|BIC2QXWAn4rf^6I41Bv@N1%K&sjwpZn6_EY8Q7U}($&N?cahCDBMH0-k
zu!bPO4QmluYG1aJ++MAle4pks&L6b0Ij1ruu7QVM@1b^ab!}1lP{nUT<z3^l^DICj
z2s+>FDk-qg+P*$plR4x%BAXYR`Uc5?O!IZDg!Eaop9I`2$CrFJ5)xH6{Y;Gaa$^xN
z`nZRjT7j&7i*vi5IX0%TXg_CA;KJxuG0ZBuW^W?rcKNg7BvTgfQ?qA%181OGvi>%f
z){-o4RVOTBD^;Q8$pFC2xyg7qYCI(j)G5hPQfgB3QP_=Jv*zTT2UW%MN0lXKAX)v3
ziy5qr<QT;00Ze(X%CzM_IjF8mtqr(*1i2`#&~$yCq(2uN9qrCp`)`D`j~;|0rCdyQ
zb2&&NE?Ykg?tU>!J~vu$v(;>Q7I1%cveFLK8ykICHCf8jrBNNnFCZ^q>&iuA4R5kR
zn!xSuFuOgb)Brki&#v+Q>?N%}mM0*!Iq_e-zWoOrksu{48jrza;;YfuFH8{~fa&dg
zkbHFh@=+q>E~!e_i;d5=w)_i~_IKj1jUD0^sOY4m*U5SQqwxV(Jn?zitQo3);bs|_
zSZ?_iMjaksjS8NAtk`5YZQgOY{K@jYNjSPo06SUC@zU_!X-6c}@o>=^V_Fy$*=sP0
z=c?>&+x%ls8t2($(Gs4HT5y;`A~W1ZBsB57*n|c?>f}>EG{F+cXm_->;a;D350Ji(
z<Xw~n+>u@cJmbB?K!_Dc_;&-I-rsZtUN%Jq-d|j*CpTs}7}M4Z8@C^JL#;tXP{NL$
z7MD4Q(P!>s0|W32cxiC)V$G>4zeiH8qh;Q<nBJot^!dWaf#6{<Yi9j#&$}C1QURa%
zxq<Be0lz^&zPfJ6^OB+{C|XNnJ!O$&j4(l2mIE0<S&R_{K%A2e-o?})O4D^6MOmOt
z!EBoIqaXbdPu_d4ueB&z>)^Y!-^SYAw~k*NL}8xP9Rc=jAY-iUjlI{jQev(~p@aB~
zh7jnwj+^D0o7)vHU%n!Qz_X{%c>MSQRh7%}O2?Xw8Pmyx>2yLpo6&4)?0QSndYaba
zeRR-{AR+mnq$tgl|1s4BZJGva-N8+IY{pB^wGYmAe4$3kX_M)C+;8K*U5un1zaOpQ
z2iM>sv&%nb>j54IK;1p}Fb0Kcd|c@Nkv|i-9<y^jP)UE^J%4)Jdn78hH^r*KW2C>v
zK0_CVfL}Gp&BG6J_>CUmjW*=FeK3G{HTs(a$KMCo)9gC5&F#qJ`xwZB&$O$646ldp
zKkfJ_$d5EvbHn`P6mrG<<d~*ivR+A9$mP<aurWwaKznCn{n$Qts#vd__t>@*Wi&nu
zh*{JaO_t|mCL=Qi)9I9|no59OUXYh1SvJ9B1zDbB@*Jfk^42RSQP3`CHE<T!HPm%2
zpjc~)vSgO!eYAX0Rk+|#&JvWx1&dNca^2p?QmL4i4ph&&-Q6P-=uU&Eq9WP%X`INS
z8}3@%7z3rfF&~4mL-Jtv@7?I@Lo!_cNG#ReDRVDiyho`SFi1c<M`&BzW=(T*&GPLV
zwioAg%UiT_C?9$`;Njr)zpK3yu=}U~?1w@FtByCXFX((Xihzo&*!W)+%_-5uq_jo{
zjSCjp7(I%V**np&8l@bIW5X9eddd%e_yJGeJ7YR6(MCmWF(x}!K+O3h%%G2E(OrPH
z5A4PXh|_@I@sZj)#~~pxM(q3F*>88W#SqO-X|RPG=ka^b`1I2saPjs9i%G!yj?H#U
zQ(Jscbk_0u%{eB^$cq9sN~rw!@nfOmixO=NSxm37-EMjE{`-{E86SQ8F-J#794!|7
zV!h$#7eD9O)5m=N>8JehAO8`5@o)YozW(Y<PEL;~iUMb)9&8yCrdn;WZA;s1*={!6
z-rlfUt+>3t##ztDAAgGT3a4}a{+GYT1$c9@LT7=GKKYcN{ox<*_kaKQ{QU3!o?rdL
z-|@4b{)okFLfbSHc}bpk)NM^!=1eNX?X5-fHF&a2%u~zTC8sA(n9d3|t7~qzHx!eK
z`C@|B8A=ssUGlS^{Q*@e<!q8RHFn6~nIi`P`Mu*AH{N{#GG=wYPgA2``Z9b$;gu#M
z!+Fou^)>6p@$+B&n!o+qpHpT9UwrX7pMCNXPaZ#Dv8YJV=|+hkW>FNBWkFFExVolu
zK^VJOTQuc^OBe6=LZ4D`sF*vk5iQ@kyTT<5o9jLcezy+~R~#6=yKC0o&-bS#+Gudq
zH~^x4*P!mcXW3`#4js##``dlKvE&as9hp7~129R}fPRn|Ue)ef-lwu*jC}0$&Sp3s
zhOl$a?+1d$o3iZej1iCARN9@#O^y1H-NVy+>}}HH{u`a<4r@NXxp#InAA!2>GoRA+
z6ULT;Nkht)EUP#!9&yYWfsWhjIj>(oXS2CQ(kmf*p4T`R`n;=3fijvr&*+*C8-4AZ
zb0~z-%k!K}Wn>r%Q&8j;MOHGcrj$iRQB;&w*|TS8Ehc>%2-f2}kGGDxZRpyDw%M{?
zUbEh8$jg$&@iFtGqoE9_Qe;I5u9G^>0klcc%VH@rMoHQ!mE27t+eR91=^q&Fm$46h
zh)S#%1~(|*J0^!P2GOH^wpWP8b7Z_I?vCFbpSx5AL+yTkaNPT6*gbBX+jRKZl?7@2
zj@=Rz5Uj=5Te{VX?d1i_*RN=<ukhOqCZ5N6#9*6z4|ksPyW2YfyU)LP%G)<LeNGu4
zEJ_=x{We}|3Spp(!aL*`ouQmK_{ioGJ2L)Fu%t>1^F_fAfBZhb_xqpm;YUw6I+{km
zjFiJB4eC&uN9@~!)AhI`cZ4cKy~l6U=)P-{9<vWSf+%VvhlE6bPOV8%IccIRqw!wx
z_{lSV_J{w3|L$M@H(1{=olHR)tX*L{$F`9KlPW;nw45GK<tYQqW;2X2OePa16RDdD
zaCCIc+37=;>kV~ldHv=MpMUllMV7H#UK4D`q?)k2T}rOIr|(BI=V`1Bse1OPFd5!;
zSl6*$uUIaZtk-L*$%M1h2W;zx_nv;pPk#1CeDnMTc~S87^A~*f`R5!hj`;MGPdPh1
z<6r;l|2NOS{0IK-Z-35@e)I#XJVP5rnU}1>mON8TrUk~-INzalES01NlS`iG6lIBZ
zKn%g(Y{ZA<{r5iN;e#g<Rpy2S(f2X1{TSWV6Zbat(7O<PA4cF1U`$OD=O#(95PweX
zH9;wybzCo({OX%m{JX#TJ1#G-I6gk+7r*=!o7Ix#M^AbGy@yPuIll84lVMCYRB17W
zpcK{zS_{@GvTR_5Vvmg0i?}UFl3Pc;-O@CzB%jXoo&Fh0l<k@+$I4FdgEBUbF5ayN
zdti<1D_uTFda6;|*P4GP;7;vo++%S0T?*ix_v1%e59i!hG$FBbKANuOxxDlJ&j7?j
zG|vpC#)PfpJYa47afk6*N=p5G$*eoiHNt`eux9|7@rLd@f7+w&{PdrzKGN|HI~&=P
zu>x@S{NFFfp&EhFq<oCW=+5z^djg@9<Ruz^NdZ4WEAR%%;bthx1w}C@GZnUNP)aH^
zK!#FMN=zF~USwohhDLL=IAXKfuwB>u)nEP%ZPQZaB`3$n%;$?J-zQ9`b7oVE-8zb*
zq{vIAv-u#uf@E9RHe0E=ZCe^!vs|uNua?~2T=0v(|69sQ$)`W~f+z34Pc@MY2Fcv3
z0NM~;H>7%x6etLwJh=g_96oe0*V=gBOY%j=!;^f{Kzz$3{h8SH(_W}{&N1}<fiaEy
zt&XPa?=`0n9CS#A^~_=fm-mU5@mlTWRqD5aL>#bzb%f29?q<p6;)3fJ&$)f`5?`+{
zLGlw>7jQvg19VA_evd~Rw|DcgD>KgDUi13pCBYlC4mj5m#QCf5BppqRAr77iSp;?&
zG550k1RWVW4@yivi$%o`fA}H4_xqpn!H4g2d@Ol+P)W#}w5{|^prB_z5;S8w`7Xe^
z&&+nvdGb3z9qjt%$zAP!|Dd@lPLwFeBiJ(%T%h2GKmG%L@rz&b;_JUA)Q;7v;da?c
z;cmbMaJHkG6xg<7yWKFK&-+cs7$bfcCX=YD5O5x<`J61P@WAo$F`s_&3AU+mT|=f7
z=jZ2Wr8zn}VlkJ5#?Jc)%w=P9))Jh>Tgzs>#yLlx=bW6(na@v{SkKYP3AZ=beDvXm
z{C{4(<Y+#_D0p=CKpcOp<H>`E(5|V<9DKl-jKzG$bT(tX-9p!Lax~#}aZ6o$j8V}!
z8@jfmX=?K9lrAgSZaeJC)44!q49*3Nu2?KidqJG#2Imys?*fp0wkhRV+@Bl$^E=N_
z`*#@IalXSt?GVsUY*iYwAfW^VrExKcK`X`a(VU<B_zSw$Qk5mmw&Bg2H{4uba&hs9
zk3af=#cYa-1>I3Z^NPa<Xr1J<YdTMshn`#S&3^r69ill;)>6~9G;Jfvm173^;C<Y<
zhos<R0JdL_FKvAL>*o+dl9Z%<-?qfIGWy>BwSQOb+myNS_=>#W)AA1if(Rgb@S-3V
z<00itPiz2)=d$-+sz*4Bi!3uCg`lK}2RVPj842iF7k_pS>ziCFnd78JtaBcBPvX@G
zL+_kp$ki<W=^p&bbM6D+q0{bZYCwY$U9DnemcyqIWkBS9dRM7ERt0r%Lmbv0KqMs&
ztj{~&+u3A8WK%*6Mp6U~-dhwgF|Rjddcw4tQ<YPe^%5Nu@(Smbm^hKsv+10rT9fOH
zJS$Mz@a3<+<Zu4^=hW*hv+0aaKl+4|vs1d(QEytd(~hca$n%1ts3@|PB*HZrd68qi
zi<TitA$D?fN>!9BrV|#ElAFu7eDc|6JpJe+9z1veMNU=}z2(bU+ow7SNMMB48f~;N
zheaMUg189UvQdBTt`8H3*hladkbcuatQEnF><fYM1lt+Kp5az1W=Dy*BSzMa{Vc~*
zYxw>!kJ2rUEdg4q-c3c;Yf8H&<2@#HQj@4<ar041DoXXLj}KrNDvWduZo6W2b-~T6
zSDZh8iCt~bos*(lL0ESikbI*I?g#JE-U--UUoH9NFP^hrIkYOJBmFi#cHd*7xoAR)
z@{UfXQkg&~11%NnFz6^0r1HUsA3fwJzxOF0fBXT*$4BBzWlRt3#v1^8+V7AWU?iN!
zD($H4FW&o1t>5<f!8WFgV@p01gTIicuUZ-bo+J(pXr1x+y$|@)fBL`R&6{tyxOj!N
z9*-m@&W-A`P-iI76nvnnCR9}=T3-aZc_wB#9|9hY^MRtAQck7>FZFtDQ?tCjrrxY*
zZGhaM2^2-P!-#n2K|4@doNe2Nu4!o7R#?L6oG0(SPY6QaH%-kKUwp>Hho`)HCE3$w
zvl;8#Yh2S(XvJ(gWl~8^#Yr_`Hl3sD4Mm<&w_E0m33;yQIuW48r1VRQw(ZbbF{vix
z#<1Qvwp(GEvRoLuvY14$Bc+Uznenb4*zmqj1G3k49QKid-}~V2kXd&Bb>}<#%#hYn
z;9P4o89Hd5JUQlrPd=fjN^EDbUEuli7yNB-{PI`-K-X<qEEZ%YM+HS$6jWtFH7U?a
z<6BF;TuZXz&QrO>sKnGFA?AV#L|^q`p4+aYZ5p~RGIk*WKQSR$>-v7V57bHy%iVQ!
z*jJUE^_^JtK7zSFZkMI|UAMG}^uT}rcM@0|fSNEk!+RkyI4S|1FfbPiB%8hb1U;Z6
zk)x74Sc;FR9=QcAsh<Skrp{#ay=cC1@*{^G+7gogP8c@Hi0=0{!`<PG^Lw1Une=Q?
zU00*`?@WaO{O$B_WOalD8?B9aqP<(#xwtSvntt~gD3<;j`@(-1&*7bb-s9=jFt(u|
zg0Z0w5-q7beSMA)=tItf!3WT$kX)YSgxiqrTR?pvuAd;u{Ik5Ini(krhM}B!KKSTE
zHrp+i=NB}aEjDx%lal$-f~uNQ&5j5`lNFUzx=<S1beOVYc6^L=HYyPTt)VJ4`c|V%
zz=w_xKKX?AKl+%{hY!h%3a1U3$pxrJZdg)uM5RKBub$G1OcfYiqNTh5eS+z}uabpo
z++WhSJtL+=KY^7|!dg2<>}(p?v9KJEM{@HRlz?Fx?*T%FKpZvteNfPbd9yp8Qc%Zm
zf%p`?k6>?#4H1)wE(Spfpe0Dju(qaIUb4D)%ktu!?ad9L?x6DoClTu|sNGzYzk!&0
zCt&yd>({(~d5i1Hc-K;w6BK=3uSB(mv>C=_AFEGjW#T|&XfjgP={#3F{oo-#`RS*8
z{P7b`PEN@4f)E|a74&(~lp351Q}SaO4ACs4yp`%T*55>*KeSQ9E+D+KQHMi@yt^%#
zV}=7&tPkqQ@)E6be*gFX1eE1p{>%S{U;W}2xZu%7<E^F4GXZ$kl21x0p*5)_pIxM;
z^B5&n6a~r{vb@B4kG0TsQZWMO=(Zc2?Z{&J7=i$yvM{3QB_BSOo;}dD4TQjCIwf1)
z_UCK5j_G7US(I4UNe(^RF{?`2wgn{n{N?2(^?FWK$@!*b#o6g8d7k45oE#s~bQ`AA
zf;`ulOcqX-$Jj`PIQBb7Rh8uVgkUX2rdT(QXV2c_&;IN`#ONHYO>c@-dmBr7-SOv2
z`M5@7HJ%T>#<m9l#&cl&`?s=6qXpYDJ19w~U@|7<l#{c^oSYs}O>$fpXdBC{EV#M6
z<eP84jMta?y-F#ns^aO>r_2^Jy2kSI>z91}i(lY;N83591)Tw5NExzwOR5)wowaQ1
znyzcf%e=o`pDJSPMWr5Oy9OfWUNDw~;<%sp`*H_h{Ud?8bfI*uaoMUJ1)!(j0;-oV
zX?GewZ~}4z0HWt=M&K^iu8!Mi1aF$)H6b|pTzNcQzljh5c>+@bVe+jfM${7hchkKw
z-tqS2;&5koI^)jKyNFMXeg-e=!)Mwp(MaE)3E1tlq<9?W+(>=hcX#Mlq<lvyArtnL
zv>r^3-u7X5it+m@!Tf&BkNO})U>PKPfuxa<R49@yLX!F#lTl6Ql+|?DvkWnBs>5|%
zG;?Y&8I$>x&3Z!!po@ZMAALlg8{WKn$;;<2$ST8hG2!^Z35(-Xs`)8(V=2lhi{n#*
zhi0{5GM#h!_zBip934t|G6NYkwO!NLjLepN_}S+?eDqiX<cuVsH##FIsWjn(BW0ct
zMoa=I&^ehd(Uf9(PV+=XU)X+pB0w8~Iv;$Ww~|zKL~1GaE*rgapol4N(y*%?;Fy$}
zq@;w903;*%Hd>Ey&C2-)zqqvU1#GHVB4RZEBY;a_PsOTP3Sx=M&|@DR#C1fu1llFQ
z6WihH4cn`8Ru^wrU0l$v*5q}HcmnTZKwAo}i|4#2J><RZoq*kom**^(HIPGS30}p7
zs|lV;X9znbn{Y4iQfN4aA!uF2P;bSPN2mPcN1yZQr_VS&T~HK~j5P_BB+O8wiU4l#
z(;;>6zSg<(1U=;D`|b+|H<>$|*SABtaRq}OJfzdbn{>{TnG9>4lmeT~`Pm=+8J9Op
zoHsoG@>lqFO9&35AybfNQYI_UVx<XVFk0%$R#hcwLSi9eZ8SxegK~6TDl8tDR1+4H
z3Bfx7&`QcnbzRM7vyP6fD%QEvSla=fEXyeJoJm>HG~!^d0<lOnl)1s#PVy+(g1{85
zHO;n$(_@T=oAot{K$+*9ot>eSVpDG@s}iLYMV3+I8CrLE7le5WA?7%<bWOvwDk#ee
z=a#fx!03!0|M>U$@Wao@^ODY5l+pcjr13b!{3IdN4^5maB5cMwp?tc(=^yg>8aoW{
zPrC7*)BB!`&RdivWmQp@IR=l(HF(3(e9FUzr#yT10o9~rI-Q9>T@XLZk3ac@y4})j
zJ2EofK7UTL>}c8`=4g}Hnn4&5Bis5M%z>VnwYDRfLX&_SSEM+^!i}GcdyX6TQOW^1
zPD8WfL$!w&ewXIgJ1v1X67V>z91eO}xK{?G%o08EV2B%!8zl>lKFeK<-u9Vnj7fB=
z0LJ5Su_{Twv-KoVUva`J{JYi+SHs*N=5w6WJ;QO};nV&Z#v9M~zwq}N!yRxpKK#DR
zs8L_1KSqkhZX)R@50pN9x^I1yxEUo=aPZ)xuZSN&zEV=Vd$Bl1>m1t&@gPagp>119
zcVh!x>zN%-1ZZlFb(XxydGy{RPS1|`^wUqcxVRDpKLn=J8K(~)638efb0$Y;pexvJ
zAj@HTMAkLn)_@}f%ktuy%Zp2FTay<Vi_-^W)daM`1;|wNu~oj$Hj&E41sAZsi+S`)
zvaFI4kX_6-s39b_S?S(qw4aPg{)V{b<D3mvCb13%kB<Rxv0KN`EsVN8iZA!A6MZZ+
z!surhT-D2aMPDF#Z{F-qyqI1sW)P5+RFW`K1r_Uq?@klt19TmByQaCm<ofjsu3x=m
zeQ`lpw`d!rWMGIvPYmmM*caZdy%VrIIXR^$zNRP~ZS7IspnZk+7H`LHq|vMteGc#)
zihx!HN@WD<NPf3yubEG$eERV-K6v&X4<4M5mj<vv$}XPDH%U}2LE|DVOpx*|+RzKi
z#Nt5`9)Ujb;b<Ta>zzZp!)A;Q(_;as=wEfpY;_-lfoSG~+9>y7Fyd>V5=E^jrpNs3
zkN%9Rn(;rH|DNZ+`g<IX`NZIDLy<vY3UaNe@|;{7l!Hv=D6cUlr!*zX$fQ+hvMiH2
zv4!NwGg@J?2>x77@Q%&)T1tdjOO_clV7!9XQ8#Ox7jS3077Y|8BUnog%!^!7PzR4L
zGEfSa8EkYe%_bA(MS;q5CW{GO)1b7%23aWLxSNsZ6DDOxRb=F*rSni^0%k!>(ynW1
zw_9{Hi+Tsoo_)ao{(tv>;_UQ&N!ppW1QkKuo)8$bF#UVcrJ~mWfs~|ISO*#FvFY&O
zd6eqMSw=YsgEFdE5z!tn%o|aRgkdepDJV_T1wxkLfzCTrmhtG(Lw@>`FL2(mn9s;`
zz{T`KlYB;@OIp>in9ML)LF1tHC9QX4IC-XEg<lBbdnk9}K%3}4)Cyp3mpE6-T-91k
zCMNcbOA+a=|9!U~<QWswE}&7&I(ODr0*rC9Oze~>8>+88nM!jFwMECj|8{L(3hc8x
z<Kyl-fl`Ul8Ow)c`blP89odV}=d4N=2k)GG=0>VS7=hln-SvDz+z_0TN+Z@m%G25_
z8wYN6++#|mI~qTCGDY*fx^~2G{MKzp;krjX@BdEVczh>GNg<g6?*o#Nb3O3cjuI08
z2@z=TC~e;3N8b<L5|U4<@=U5Zv)M6uIiqQ=FnJ;1P8+PZv0iM6&P$5f46Vgi)+j@k
z8`>r&vk!`s<1<3g1RJ<~d%>(aCeNoRUqiToB05=i9@;Bx*HSlY+IB<Jths#s9PK@m
zqNJLZ<e3mHKKLP1Lj*3OT-jcG2*KbwO=kn!^@e);lCOWgp|L9FTJvZ$nU>T`S(Z_j
zB}EZ)?Bzv_*ssu8j@IHfq{)W*#Y%x0CFWM3LPpA_9mX||k$u=FI#1D8C8KBV?wR?6
zR)&fWuSzAy*QC;qL7?oRLIBzUggNOe6+)kA+bfS#9&}7k?LD^M&|Kee^WrrZU%q7f
z_L{J&DHHHZV3mMN`%k|&+^aX<+ukiI_6JY7zP#ao{MUbp$vTYoxB$5^7&UZ)WV08R
z2<xHGv@A}gUV#oye7ORgP8Pg(HsitZl(NV$TJ$s(11jRmiw~|zeheyR^zMa_+GPcX
z4VV7j6r%+@W?TGsOD`QW-V={Zo6g`55U;l*AP<kokhv!cA2l4WqBQkE@#x`u9R2B(
z9LHZybC%~Xs8?6CwI$CKMPWEvELhCu*v`u23eopY2UH%To0KNY3KRz0Np0Lf21>ko
zy3SHjU^_>>ZDR51s6B;%3PKlq=V;p&Yt!WRRE3mxYw9g+v&CeFvJ_AkP=s8em7$H_
zrL$Di2}NPps3q1r+L%1kI*+v;rG$OdN|76I9eC>~CZ*IfZ9AHJOJ=05X_hMrU+|y(
z=l?mMef9;$<T1rmPM||+*E`96+qlF5?AJ0oKqdWt7tGx`?;ewv%tiZcFhM4v6u=Xc
zj%HZrxV~A%>KjUeNPy#$W3=+LO@mg7wreE2Ju;WQ3u2p#LeB<htfF(OXCIXcyUb7^
zItWK?9gDR!ZL^_k8;mJpl3d>d-{JX2Wds@CzY8gL)mb?_DS^p(S@mmcILBR0TTJU|
z6-9aTTee*@;qHA4(CMKNhvY?C`$3IlOn+{A1wjrP&5MW}#--B+R+R`U(pED>>?QMF
z2m<E3OPT(^8(@X>#dM$jgR`^oeG82JM{D;oz~AfJ2~h9)VBZJq7>(-vHx$8l1dA!a
zD*1Y+dloPxUr0WzSyob23l{TJ-fpk4E|7WIe4TeRZA;!X<ZVOOv}AdKF@`+LXrc=f
zXh|bwW0bVBw4vta;)<$hY4a^jxuK}Gu^Na$nM|tHSch*r+SLZ-JY}9!78zw;^ogjG
z0v3Y!Z2Oq+Gx^e5Cq)Rl&a+uJT%6Zjo_B25Z|Q=GhC>Nn%aj)Dk;x=CX;H-F$%R;X
zCR55{LXlULrKD&vxs)hVnJAAi5GgtcSrpWu1oiq7quCj8LJ0TFpEOQ8Z$&JT=Kq+j
z?oX>A`q+sPHm0{BCK30DZ=DiroG0K_su*LjT}!jNWqWnW?W@<+=U1@aVp@xipb%2j
zFarbMoBBDpcYIrWw*<wXJjOZ8n>Vldo4@=e+O&c&yAB_0>V^1_nxJJu_MM0)pm7*-
zEG3qi3e2gQ*e#{+P@RP$7ZxtW)W<uin1;<FBwIT41Cth4mSsCy-?-iNc}hE*+Bs&w
zCBSw6d4BuW`++G?XiSkao1F23FMh~odCAT3l;>amg09Y4%qFb18y52;o;-en(S|I`
zaNhR)(6kMzQxs-Kri}odxLBR{gdln1ymORU7i+TCgdjCVWsP}!5YQ-Q1Y*#J6pq%~
zFsY<;jqN&&R`}M&6fTlfSsN&|AulVs%u!|;+pVX~GJ>T~YVE8QzrHk{K1fD;W$C(r
z(iut#sO+p~Ti0xxhAc1nU;ZcmYyPAE=s%{MOmH4amK~Dox*-_fZ;<zm@pv5WvQLK@
z*!0=XcSq;A|5_sul*~6Fpp2M{o%3v0Yqr}hUDsiZW;UBLn@q{`jOlbL!oRMuouzBr
zp;AQL*i|ew6oAflw5`WVAaW8{Bt2Mg>~QZ2g^OUk>pHZ~`}<BC!q{Y*=G+dTjSBWY
z=-cUq9h2nUK2N}S0KtcW<nLrD_N>}nU^m7$d;Adle<U~;PuFi1>2FjD2pOHTf3AR5
z>48Um76p95x;-ZB*#lwiQVzR2k2X?J(p*dTq!{M+*!aBb|Mu;hOFtC%=*Yv^4rVo3
zD`{ZA_4IqJ96RSppQjs8YB!qjK4v}j;T=}+&|DqOzp;<BVDe#|Ck9E8a=#T+)q>-b
z$Gm>?wWOoT139uT+P0->8j7}|t!qhJq%?V+4{lA$5Q3s<tVlBrIvAE$*Vv*X&kAx=
zFr92sWsc4=bXHK!Pso(U1ci2iSy@pQio6h3Flg1M<B_@NB7kyuZ?Ul|g|i-O9c^P-
z-EKL5bItQ_mfYTeH(89jMQV+sqpzOi-pjL`EYHaDoIEp3CNrvH%A}kzo6R{oI-;Cb
z6qACy%E`)1SXPyaw~0)V5JE8%_j6bUv>&l}A?CwNbK_oLg4B|AZ(!~i1NS(#z3<+L
zjv)cD(9<(0HS(2J7VkUS?TY&5n)UfPt5>hE*GqKU_Q{BK7}jf&$DyYlfkSigJK8&%
zV&{DF=?DDbAN`cew^w{!zozRXnd0Chlbk0tC^OP;JvxYqLF<gnIVL!cw4<nRXwIM0
zTs|QjFECYw4xtBk@dPk3x2PR~zUx$dD8JFTjgt$e0FNO=`TiE7Rz1Gv@3M`bLreH4
zAv&)<e(x!N@=yK)zW&wE*=|-mI{Sj{>W1RxhJa!=Tae{4oT<u!4^ATF(CE%%az$pu
z55(i?tRr|E&EgJgw|L*6l&8pZOcr{rO(DgnQ(%Ve9Kna^$EPUrJOaChJX2^T_hOWS
zm|DRYDYYU|Rw4T5)mUpK;?s7$6DolSC>-8NpEwK71_GiLZR&=m>u5U155D**|A#;S
zbB<1q#ZNISd*(x;wj5fU9Y**2_=O!Bqy2Oo<InCgR);@}?~aa*QD<?Bk$$&WD>WOf
zwLpe<@^@!#|9f56G)+z0w7oeLBCs<Wof(3H&UxBSYKe-e$dP<q5{3{*<%nw3u<^8Q
zOI}p{c@uj*23QH?r8%7Hk$zX7rrc`x^|#!kXPoGG|MhpZ7PU%|*YQbX0E7&cACTCH
z>ALT}ciZ5oJ(eB1XC38oB0r^)N_>kHnJ6djho(15`3UTsg|36n3R!z!3-LPECZ1*3
zSoWiv1Cc#Uc}E$~+TG_%gPrbq?^C(g9beahnN4iSxc}~7pX{vTJNMDw>349Akxfm0
zQNL+GPI@NOGeW5!6xPZ4i)z7>_dn$ye)+d_euJ`}%m=K8wzFhSL%wO)mRpR;$@7dn
z&&bLG(?n2IIO?qxu&Bh;oGC?BWMo>Sf~9RXOs7ZWxkhJ-qR^mnGA0xpq1BuaYFy}I
zp#X&sUci`13Yo+E4(F}xBRFhubhcx&-EjW)ir25Mc>QY0cB^n`yb1&z$dsZmIl0Nm
zvz&5LQ54DlTay<hn{7j;SLCLk%uDjTpqf@xvk4~;j#-=>F_}&xlb4gsva)WZLWiso
zk3_5_Ek}d*7^PzjuP~4U3zq<QDtjmCW*D1b$KZdYXa*%fGlU@liP*Z*dN!nOK`D<x
z`jBqR_UfG5m(RKR=4&=DU!dy-?R{kH`pFpOFF1x^jF>p*ojLP&0(KW~Z_#?n+36#m
zKK+<iFK$?NEt#31)h5pJC}mXau)dn2m{x_fRg`!pcoxc0H5;~XUa@}jn(n~^CX*?;
zn)IMfsfc0+@c?5)%dp|55-r~rk(mDOHU1DHD;N$Qms$b}`#^H11$wj(cQ^O@&OO#9
z`_FtJFLHwSOy|cGxxr{?>J3@w$eui5vt9H0&6_@tQJ*HlS)9_8g_PKm0^tf(WaK(+
zx)Sh!?{J-`vs=9HkSJw?QebU^w+<I9L5t&S3dqPzMpc%qHyf7ATV}HvffKxUs9-Qi
z9yaTpm?ykY>V!aUGFbphB8a?mbZsYjlbisA?RtxA9nOk|?1QIkJ;5lv?XcG?@*?BW
z<EQ**|HXgF2OoX{I_uY|@&Z6ok?FDd<Tn-KhXDOS=61v`jK}sqP`Qs2k_l3RQH1%-
z^PIBG#TQBiTqp0_2%P(A6W?=Q9Q3`CjlQu`-_bcS^M|OB4jZwS_aWw83OkIQi|I`c
zF>8kaS8swo@cFnu1=!plSTQh=VQ1{pnoq|XZTM}Qigi-c`QzOn+IhhBWebz(W;AG{
zXG*W>khHPv10x?9x&+>mDHa^w>*#am<JuXR%fKK4LlCyk?wMf&jFjQT%*4L#W#8aO
zchJvlxYO@YdsSq6lRxy`Gx?Oz+7Gf^MNp*<(0qqv=y$PtyLYsA2m5UByPonMCNiDd
zMt6zf*wC*tlT{p_Jm&FxA9H^GOWJl#o(CpbM&}%L*JA25<z_-|awv=F%OCyKvy8kd
zDO!i?EK0|u#uJUoV+p=sP&&}rE!H=9V6$CPlqE(hQC5_db*w_vV07<S=Yx~_<Ve(e
z*R@#JVx7ar3PkJmmbd2@{KGGQ&0qi3*R0nHhr(-5P!1CmS@fM#TK1pmbjqZfFr7|W
zERHxjI$}DRGOeb}s|BVon8uK;a;|PJna-!o=W`Z|1t%w`Os5k}7T0A;duGJnE$BFx
zy^3oDR1jZ3r98gRLzf0TvTl9AlK?gqL0un7o)jI0z&D*_WQb7>s%HbG-n;6@+yqCk
z9kg4z^)<JzUvTx}8<wwM<2PF}>&35*xK@=&2CpJx7X+iG*weB6_`dC(fZhM_-~T(#
zj-Rl+t=Vn@_!6V1;1$7VX@-Lbon=w|HQ>ajOL<2g9GUN!DOf1YOlhiM*<M_6@%1;H
zJ$S-&x)3JFR46klJ7y5dmw*mxa3UTF_|*0j2antS=AHz(-2Pz(K(hJoe*a)g+UK`0
z2xDM)2+*Z->&Ve+BhkcJwcyeFA5m}DwCgJ#o-DY%xe~T6%PGqe@1;USs65G8U8#VI
zw*0&(z!*%C;hKzgeT(ZX-ga!aE3{VFwjs-m0IgP{w{-C6Tuhrdxv$w|%67e>*=%V5
zP2l$GitO~1(ukHBlp-^Vprss|GUBZ3tf%c-5q=RwG<Dr)@HR@bUag`j!V`il=GM6w
z^{1(3Q%+7#`7i&g|C%p;^!w!H6z{<lS+9u`^k_reW$O+ByhG!^4<`0OePXBf2KhcR
z=iqmgj@huHZ)%ZCen0YpqO3$<7)_QLQ~>7<!C9QO;s?{UR8>V;mMoV`F{vOzLwr8K
zE2%zVt^DXb1^%aPHnP^eTnmvYjn}XiZBj}C2}TMbMGgBq^S3vh?Xe;IeW#)(EajDC
zt(lnqXMEf^>b=hiIqt*X0c$sGrs=wABNzm`<m}RMjQ2s{oinEYhri=bGJ6U;FpNS1
zW{m&t0dDBQT>t7LV|T~5E~3E1RH%WGkn@HfnC^e}8}(h!d=2+LZu{%arwy27j!k`e
zNbz}hMJ(M$I^RK{xH~xaH-$ubs^RZD*Xy;jguDSX!70FCvXaM7p0U}yWwowpyAJ0)
zPFZZ%QP(YXv85^sw1O-bMMY^9{m%?}m4Ox(2O}(*SDs*Eq9SE+YM7r6NlqWok-X?|
zN`k>KGUhvgKPJLX?7Q<8=PYg4(R2+;)0aD2tyZiy5|-$d$19HxyJ><lW5{oZz=^7=
zn9XL)XEPSF1rJWnSR5Tk*;Pei#^G8=x9zZv<>>gB=`1pMg^mPWBzMq~3oi{zvJysy
zUymc!dteuaiIGgN5zzH+2zv^)Z2oC{hun!nJ!aJ>FjmeIx`wdb(A`|Kxj1KYbxF5c
zqMYk5-z&W+ypBG02{`si?B1QNgLedW|F8eYUoxG32|;sxc|+UChO1RfJ`9n433wBF
zn$m}WcOAHnJhW7FoM^?u6qG2m17CYCzIn~VM=zPq=jc2mD>77wpiZT62of2KOm4?`
zP=^h7e<L2Z`!;VFQj^@>d<MJw{&#n;lj!0>kbRfgb|2^^bBKwNtYGj_$uPx~Cr>}7
zTP>KCiagJF@ZbT>wib5G+VPX(T#Mw=F(~iwZG)~V2%g{?*0(a@*2^2#^_KZ!fi`m8
zuJt%4#iv8iF{mM$L%q<?Aq4Tm=-QrnnM_KuJg2BivdrLOUl_zJ=zQ#pPJHTY^bHf%
zt|2%_UDtHBBhNK{)8T_S=GrdMbe73<!Y7~nnE&*j|8suw`+r29&j+?C1iXrl?kcK8
zQG+`KKKAFveeb2Q94DK;&-5z2e|Iw)&(Zzt{2}htT1(}SqLfHE9|F!=j1j<t4>5AM
zpsG>@ftKyIj7+FF^*Zk*T1^ch^j!qSN@x9cDvDv$fcwpP0K+~Ab0?_}(mfwuXZuRe
zAwBah+b83aW!VnfKJK$I!?WL)={pG+{x&W7PRUYEfG7oW#QRMTmYBT+NVM8n>-}rb
z^7|gwC5CKxp3%Gv2|&^_&0dh-4wy@OO>c^gjNQ=TyFD+gf?{|W`AJaoH`+n&X@_vw
zxQtvHH#O=*k*|mnr^e^n2fW|&9@4+zt}?#E#vVvWzTbA{@tv_t<CYE-xg$dvgZ+X+
z>zqfA-{<1;E81>J+tyg;FgUETsJ5lqZmEkhre!H)^tA$`q}oH34>i(NkPHOMqqUQ~
z$ON&h7=zaG<F%6@!w~SwN>H6q=uF&lk}E*(o)nybEAMdLVqHgP8(Q0t7gE{A1yAR!
z<V{3aEvI~sozkf4IMw>UwW26;rjrRrvpJ6+KH|}%_c&f0GhZxN_+wEoyMV2&)Xi-j
zt~*8@YpSf2AgQEqiPv3RuS(pEl3ziH(enKSPrz&#&BVI*YdU2o7_)pnjb+;3`)||Z
zKG|&|NzoDNEpD}>zPw<4{+8zE23yx;ah`;}>{OJ^s6LQ4-MjQ-QUKqry%VroEfw{q
z!TXlBu1U3BgKv@aB&jFFvq^47a)<>dk+BoyI0`LCs$`-xnJ-ZoGOcNrEpLAPocYl)
z`C>sepY>)m4-%Q_k<j9xM3AyuchVShN0Z-g`&xgb{rYo!o0;_6Ug0;z*SptB;33)n
zr5LawC{3neGCxAQ2549;j?fCW*_!owjqSP!rbWVAXT`_N8*rAkUgPUEZL`JJ8*b0v
zvbj2Ey}G5CFCa+j4JjA{CX+~?5WuJmlPNNjp_OK}UJ`sD)0$#3Wj>v9yS(M~o7V)T
zn9L?P8_-6hO@>j5tSH23R4H^5ahDgTUFRHaBWAy@ZBfxgOOTRcX;aIxl0W{FKjY8;
z)BiI+`tgs+@)<V73MNKL1Zj{0UE^^{?9_JwZP*2K`v7iVdAN_|J#>YAX14!M0#iN(
zARy&?jFy~(qx&L+=-29Y-nUi~MT`FlJs?Oz%Y~Q@!%AeV*D*y9L<fl|pn{k;d>G8G
zUDwfd9jI*RlMx6@V0RzzejD?$3$h2_#XC%02qeekLyFrzWBA*&l(Q>2zocsp0535T
zcfYs$MtaeS6eDd4yRDbx!+#-;n~y$sLGpx(Tnmx89DOqk5wYX<N88t^|L%gg9^QRx
zPOxMf`j)2pTm^SNN=m}c;R$T@eU;J(-97d919c;iJi=MK=Q$MccIe2%r<cCFd#soT
z(|nF=Ii5NJg5U#-#VH?r@Cn9reErQWb<>emxopHCP}enD-OzOnli38c1ga^eFohx4
zIZ8_n*wA@uzd?DhE|WS=T0Dx4O5hC?zIQQD<X{}y$X=P|rn6#2K`X57XuF2i)^uG*
z)7DbKLcwf4LtI0`P7b+n;8ke>eZ%;Ql4zTTZCi7Fd&So;zhP2MIXgb(@x#ab=<^?=
z-hT#DB!71hAH?l;%Vx9TQFzE?QHejG*2F9HV;~G<7G<Lo>6dhvVId9*4*P$n2Z(?#
zj1;{jC6fZ&kF^}*Lm)UyXg0K0mu${obNT#hZr;4X*IP{R(e-IpNerpd1aaP|-8HPW
zI=m>prM(ld^EN}7T>Q7ZM;njQ9nN<slcHh=!IL^p1alxaiaCmt05h#9QkgLfZD*;B
z;p+8UUi|9UsG~XAv|^HJ(JV5vGX|su+Jo@B0ThQ=5f~PCf4mPp%QkNN?122T`}ZAF
zYS=yhyR<Q=6LXY~S<$1fqE>kCFj+x4nW8KdCXi{xY%*msnegi6OS-nD$+B2C)sYn$
zMtR!p20BCCNjj5eb;Iq|IoGdWP~`>l*^DeRcrS^5%OZ<@BNA<Ud~$*@ntHn;&vLYe
z%Zqa$Izl7n)Yorb6STq<8OKK_6uOk!lp$aUvLH<i6%Vp;Niy747Krm?#xSWWv{uY!
zIjwVSx0dN*&L8~YAM;=TU;eK=`|uM?UeI<9W3tGQ`;nL)uO3E>!0z}T8uRfSI%Mwq
zUZC!t@jt|Dp@<H%LkX>3BaE%z9Q%7q@23R6DbOI+8TA6|QrdtJVy?5!(RLk1IROPb
zeU$RY(Vr`X(EHh`xY2}I0&o0GY2J>bWDm8}|6>+3-FrIMIO6pXV>ce3eYX9#X<Da9
z@bTJ~4oW7gp4G$m%4HhsF_6;6pj2WL5;75`mtY`hLQK}0;I;D>WfYlGcpXhXJu5P>
za;a;*S6DB_!dpLNxJh@@m-$k{yl2M5sC!{1YzSDut7v%%v4(e&t07q*_CVce2jAHf
z)$rba8`2F88*17drGLXnF&?dtbhwYldzq)BK=OF5Abs19&)s%NCacl0qm?de!3P<m
z^uRINKFNBmR1fNdkIX7!niFttNzVExFle4U{t&HTd9&u``YQ+l7a|DJ5Nb#38(P;;
z=0%_1O&hV$6nQ2c5q)4y7E?nRk<&Uil6J%hF_6-yD5dapl5<Z<g&RYT?K->@i-~uR
zrmJb4BqDCNj&|$Os$fx7sH~7@3%$n~3C`=V3;n~Mq7e4}@PJYQyV<a*8(v>r@Z~oz
z`TWz*_~?U=dG`JXJbL(mtaIGlSnB9={lNzxVkRZJP`C{7v(w^-s#PqDpkvNCuSlaJ
z(=5a_u6mYF0zKnxrG288H({8|L4#MqXgUa1vPp!PBUj<TH~94}+w(UpU%uq#n^*X&
z6{>Y88wd_zV4;u8N(`q+0EG(*8v_JW4A;B1cLH`Y$#swZgI3XTaFh@j(d0YlL2H6{
z<gTN1H8a(4Vgg61LdQI1DyE}wofm=91g^h&#q4Ck@uH$E3|X1swL%$#Q3mb7poZk0
zY5nU#dzW%3Atrj?-4w<EDuf_(^vHbLKkuDQd-VO$?@735rAMZCnK(L4u>QJHw2vLM
zM}krQ7UAU41eLBN5tCYHgixTXQ|h(Bc55CT&nPlYQ53v+{hF?Av3W+<WoTn)U5mrv
zThaWSwRBC*=Jtl`>kF3kma|8Xm=p!d2b!is6;{l8A<)?tl>xzm$|X|Od%T17dW()}
zZB|Puh0`@17jNI9LLmFd44Rb5Xd7XAgY#f5!CAVlqqPp#Q{GVIIS)=w$UawGT;KBY
z^$m|de!?d|_%VO}&;JWP`t&obmt>j7NP<gI-U3F6Cfb9qw0cIMdw-1ka!1g!1}ctS
zI@dlBld<3R4;oIKvXGD84{^gugFU(z75dVk`A$Ykg-RO;g8)byg$RSO<9*M+kfO(u
z;J^#7-dT^efsn^@j+ELSVt7x8WvwJHVu!OEf~jIY4G57U3f^P1_`dXIA!Rc6+Gj|8
zJ?f5VEFUpX@!r*41zM3TpXvFcPf-xh6j?^Ufdt%ZJtKFg^jy5-ex2^lh2i_@Co+7x
z7YK*j7`UgTtO?kwaA#WWTreavtoTv+fSBxJlwe3n<Ks1S&o+)g)y}i)eyysw@$X)B
z7m)R{rB_70f9rjhMt9+(3_bj`qct0!$tRbNks{P1y729a^^OZcdT+G;<+ITh$39jg
zd4A_!!^i=DG*1=794fL2lj@ko;sKWzU(t50=(-V%HQtJEoA)q@HHWpxl047F%o^na
z1TvFL#UU+`r#ds3EF;SbGE+$MV=YyMtPQ9Lzr-?=OG0HYebTv>&bE>oscUJP7Q7}e
zOD5%%ZP$rx=w+GQgN|||3`6prZ4Bz2b9ByQ?KS%AFIm?$SIb+zaG&$&!2`;yB-p^s
z<&t@I!^zn(6J1I4xE9lFP!KXD`#`*z-o-=)XLMm15EU~342g}$GM3~{M`Q#<u1krH
z-qI9{!U|j0*0i^`tS_%vUtZGQE>UfVvH>gvr=*%=l!d(lMUc4@=xfD_E;T3-@6z50
z*oj<qViQ(6Mj&Pc7o$>QuX*nXwgJGnhDm5S#&W7G6F`&2%`=&b97anrV-s@xvf=U<
zzvOf=VK$wTPb-u$1a0s-pj<3C9pa{!28n?+aDFIz)R*$wEB?MKv_)|3Gcsc^7*c^>
z2F4|!efc{AIlX(09nco9Hvl{NC(*e?+SMr$2P3hA@*!%11v)EHlY-S{L-*#I<9UVa
zwsdVrroh>bwyn`wj`fz@Xz3^Gu-k^sdc}*EFS%VV`N5BW$l~Z2qcrWdrmY*U+M27I
zTb%byW;2dYPOz@U2Sw9#tk)~vzInsz*Ka@ts;Z=I8=MP(LkG|5_J-$Qf63ElA5fMP
z@}k0GBXeYNT`TOb?S%gKj;3qz!BgY~4<9_FDh#(bOa7Dp_&?;m4?pGUCqLqcKmG}(
zDCin1rL?@?$(A_u*!XoM$PbKh|GrP4XyjL^2lM^_KKe<M{<*VSe9$|erMWPerH0k2
zjYd3Fv}l`($Uj8MvimT-zf@nfhjiYQwO#yVyi4Wd<iZZ<T)?>?iGaNlArt)*rCLU;
z{@@kC34;^7!@HK?jie$9N+eH^=xP(4Y1HV6wG!<#fps4pf_t`(K^qppz5Dxk-_#77
zh7U0qOG_kd&(7b!kAMHBl7hIqaqlYwV`WWV_fDgRSB2^Y`7mGPXQV~~VZ`1I*XV(A
z`h%g5y$^!3Bb$}w86lO0a?Nm`FffjNKQf%Smq}{x3h9|s=M11FZJ5J`I10u&un`?{
zXBhWaV#X5S-(AOJ9%ZB`?A%{}Ytm;S<qaINBkY^_)9>tkJ^iBkYliV4>^<1sw^YAr
zCuM#N_%n)P%6xH(QW|SpT7{Hp%%pz0^B(JjT`IFe_Aal;^CBj0m#~w}Xfh*J988pg
zCNr2UC(8?pyu@S$lB&a_x=!{<m8#ts@lB30$tFd+ZK&Ipw(Xb}Gs>c*C<<0h6Vssh
zUhznY)rYmO`u=~H7J$}uygq-+?Rv$<^(9I}mMM;BC#XWRS#G#IzY-8!81lTpj;c!a
zsZTr+l|^haUZA6i8+tiEdS9h+55s(5n2&P)h;w|brR!tBM7?EmbH)1VlFijMZnHsI
zhq9_K8@3}P8TOkr#Y9mbU09C}|DNsLOtHb@yv7GH;i#Y`I@PNo5orKt8<gFWhmNAV
zWv*Jza>bF(&^THj^D+gYPD2D`DZg3pf%U~TFaG{ll#64Ii=0Yln3*AD3fBi~>`t70
z6KE>W^KE9c-ZXHaSoHqu-~>Ged!eVflfb`Yo*kWY&=5lp>LB_HMiWU;YS5n3cSp~d
zn8vcKC?CGZ^6C{|y}p2MOKm+`TXfS1m77_N)^s7zH4QjVv)=OJ#S7M(HDCPbN1Q!;
zDCe{aT?S@V#r4e%FW$VOZd&Gx1-F|m)pUXiil%P3yt?50?OT?sTdeJ<CKY8-kQaHM
z96B=@>-C!Jt7}frGEhbw`klvi9j>j}HXB;o&~^<rYKoq~WKuD$D&|MWJp1T}REtxJ
z*)c!)=^rwg%;|iH09k?rzOQfN*bF*m@Ae)34;?>z&mUZAyPH3;(#p;OBSs+fn;0Oh
zcM#LNj(%QaW;lL*v?1;KYmHs)34>qYz~T+^PKw2cn1Y63XJ+v6N^&I$%Xqnb00961
zNkl<Z{NQ~L6s0l*{ahdS=XkBj?>cVUyM2AA-@mJk$KlZXzl$beLwDEs;nVN>`lZ)9
z%EMS@ph>=ZYIKg?e=ZEHb^2`l{s>CN+OT4tb<xZ@olY^iq&Mo^6<uR@K;d8tJ=_vw
zH+qMCIXixS_Qw7GHU?qC+8dkV6XxxS-TiS5!^_UwVfWlBX-}wF<>>y@SNn3{!0q<W
zn0n*8T0hcCL8gY9seqKbgd#6lEKbR?3TIo|t|hF+1*R${NIIiha)ZW%<T_L6+K^`@
z#^f=cSfh|2!$C9B@ke3EQNc(+qjR*4N1K4kG(l+rX-=ZvT@V+I&1TDbyJ5TCqD{up
z<0HzVjIJFW);2iT^`=(8W9p=@x}xILd8d`BBwBj2k?UVwUeeSXzWD45K78*(@;vAA
z@|?~$w7%iTkAKW{(}9Z0UQj3_QM#gN?&Tfr)q})}M|YP%(!z!bpax}i7+8rycD}>e
z7T484K)H_4Y^blVxP0-Fn>TN1Z*DND8Wck@v_pH<NP5TnilA8y>*8J7yP0CKgdl)Q
zMTDskCP?*z90+bp>9$P$hM8Y;YAmNYltIyX=ny|1?L*waBwMr(!ZjGB$gJY(|3}-O
ze%W@M=bhj$BKC0RoNnF;xBy53;3P_<BuciVWvk?=%63)v>aOZ9y;k=h*sI&M+$~qx
zU2T_SNv@J5ksv{W1Bl^HnR#!HXWCN?{UIXu-sfZ{fJ@SWl{k4%p0juC7~Xi^=bfIO
zFqnNpG9Hr-N617W)d-;!MhYKH5g7pD+=zB#8)LA+r(_$DK?jnA5Ix4RYuCWt*84XJ
z1R0;lbx!5|&&7$_0CfS}#sqA>PwKj`*}6AEAhHpMkG{;#!6Bkq@yX9>q%a_*F9oAL
zTbGK2rqV1fuQ)zF#u~#{-u)`Czx5UlkcmP&O=En2M>{)w^V{DcNmGoqTwX3XJ3nK+
zT2ic5lvN1{jAs*kF=Xu*yfX~boGj0oOeQ!XSu7T0`Oq_8VIZXvtm=a0YEDrsX_^A5
z{QF6kk`G4=$5SL3)+Weo#8<!Z9S-h1pbnreO_k4?$8{gcn+}@mtji|Aa6!@7-bLHr
zi<Loc`%HKz_($H*+zrw$9v*@=nyoX_UsotZZ@0T}y%vOp*O(J6)?9^F*$jSK9ng&5
zX^~!dGXlk!sM=VAbFwdj0O<R~^<Smz0O0N54z3T#cI$xbHtW&{vA>pf<Fq-9vAwas
zY0j@Jb2nYOh3gw6LJOt+k8r*;TnL`<!9^AwBAu-R8Pl8uU<W>{f0hq0&T)QmMGS`V
zN|3m{e;ezkzdk>EUi#zTXQ6Mr#Vvbsb3SRwmv$Z59(dkZtGD;5TP)nTS8r3k`^}H7
zX%(G#&a{!1v7`O3M*)XF8x<X^E&^C^NTkZhvk7T3z}W(0Ds)|eTVf}^Cbn>bvZ`pB
zhCClI91IbzCeJjA1QpmluRQq*6^Tj^N+Fd*$pmByBq5uGK&b}L$b+$trtzsvYF$$4
zg7eFBmdmBjLXc$KyMLeYaE$l=Y<zw^p%IBC$pb@cY!`$T^#BLkZkeCm;(Uf2xM7Zq
zs^GmJexJpB&hqUgzxmbQB$KdOEpaa`<HsZJJiLQJ(-`gDp$IMz&b8%7+VyF?h0ssm
zO>(&RD!T8tO%V2@PZ4J8np8SOy{1`SvOGEB>iCHI>I%vlWkH63CmDT1Z+-r-zGEas
zWI=xkQ^c#8Vns}hAX*#02!}(fLOD%h3dVN9o?9^!C3}frmZV@{5rSC!cSBZX0p7*b
zn~D+#iT0Y~*~d?r?9LdBhCT&Kn&Ff$n<a6r?@`ol+d6Y}n=*p8>**!pEN&i@@o(F}
zE`rEzJT~*^MKf@_vBh;kdd-9vKwT_AEV{J!L`t~d;<F@4g2Pu6ST&YBpLhyaX2`@M
z6JMV}V7=c;a4elK7mQ~U-hTUScK7#jK$4{(d|tCbmZ7zwX$)EB(_*AbF>w+jAW;%2
z9Z43dIADVbR*_{sF=du#j7Jmi<B_G{Jh*d0FdPjLLQvHOt71u2ml&rpwrTyuwDXSO
zMdNci%%*#M`@6r-!QK0m+JH=a;z*y+&X>m7_U+yl{37^3I@_kj*amFz0PR2$zk@e9
z>Tk0W*8%-a65wWb0HVI=e}oXhZ${yq+*mi8^M4b#>s@<8DuwrK|DIl6XyeVXI+(EA
zzq-!Y)t>F)wFQbM<k@H$2;Yf*Aj9TZMf3B`rZ6Eyh<yLz?K+$0I>7xawm6ENF)--T
zTcGvEJ>C0vfBrfQa9wGLclFnf|BUzRfUmRNTzdTEEhVuwsdHXAOR^l5r8sTcTSXGi
zD-7NG^2~6nU$*^d$H6(oP2Ht?jJ{H^F^iuuq1$`21#~U!?cZc`9Qr=hHz^YFdhItw
zI39cFz~_PQYki9kBqFF2?eE=n!d{MUIPt%@Ps4n+?4(G@vN0;jk-7^=EXoQYgg1Sr
z8BL?xF&236)>MtpP#{$(Kc+(2MTL?nLWuwdBDe_&c$Fkg8xR_Y#e|x$+WN@XRk>oh
zT;YUZFc>hIOc;+RBr3t!21h6@nLwIJ@__`GAf3Ym1rn(c!LQAlE=|&=?Ca(*a4_u{
z`rmyTwb&M`70-^3P%`1({ypyBxepqOqU7btDZ6*}yk8+Pp>UL|n`_QF=h?*mBEsJ<
z&REyOt&aBIwn5QywgGFfwubeR`s#}H`5E<Mi7P9l4z*z8*TpS5W4N}9!qzV!-=kl`
z3=6N!*aZ_*^fhxJoF$bYLfCPNW<tTtt=SbNd&)2okO$LGDkU;zUAERc-n+Pk9gQ=j
zBqUBEte~7PIr`}@7z~Fbc}AM&NR=Rxv~>nH!Iw@cg`pGheN78I(2JEBV7q*dXy5BE
z!rR!nn*d%EP%$spcH01ot-$c(>4YGW5-F2*1IwO+_Dve5095Lul(Z>9C~D)#Bn*Ti
zO%#a|WGW#QDOwwLR%`b5_Gm0D)&+T%BZNUEi3gNYVj9a}G$u<k2Kf*nB!g_oFdtG*
zCYa_PO;dYNW(>>a0^RtCR^2oxl~C6jK}}URj7DS9G-a`vgYa`+6l?0Hp{z>Q#mX;W
z;{z;qclVg>9)d`D{o&hu>8*Dd&-Q4XBh7QP_JT<WUz4!i6gQ0ZhN(wDG;!ai@qb=o
z>sC<TzyfTyM0IQF8Y6?-=!0w(57iz_QCpKvwz&ZNH}0aR<98xk?2C)HizLV~DLji<
z8cVG$#;K4F%7aiT@XT@Os~=}CoMoB>Cw&vb^@G+QK=O7Z>;|)Z!#Kv~l`(;ct{m2G
zFmIbo+;zol<1qh%hxKQ*))ehFbJyMUEH|-tH}@gV=eA@Qy`Ol<xekzbsJi#+!h8y_
zvS+3hs6>%w87fIa9q4{iJJD7Hh<)q_Q~c877yHzIt9RG_o7>~JttiDc+8>in>F$Hb
z4oq%dt2>ZA_U$Pka`QQ_KWpz$?ahY>`$x3$<N8nRTc@*k@>#^_%bv3Du~tf!4N0>h
zO`RZoI3P`}S-S$QEp`AdrZ2MI<2F$VRazreLZT9~wDT*E&)BBN2-W1hmDbODnkG1W
zDKl$)InY|y6jjM;wW6wPrn4#2*^Jq2$IE|beY7>9j<FOZsjoL{8Xw4^d`3N_%uqrF
zzfYeg$rCKm4>r&ZF8IK^uIymcB>lRL11o5>=KSIuXB@BHd&tgg$3JT=xV*TeSXPWC
zL!SpTuyNZgmveD%TP$jG-3GDhh~wsn@x3A>m+LPL7yOEyHDEPH7ns$Y>hgm1#W|)}
zBbx@Pt*?GWi~#1=)@}2sK!jWvzobliHDDJ73Iyj(5YB+D!8HI3T+N=W*ugM%n%o%@
zWC)v&k!B$cpmRM3)96s`v(H!ri4$Z_P@P@y@)sYG4l`7qk;{Z+Hucpl!hsm0VEv$M
z`u)IW;^FHcY10(DEdbhUd``CcPPp7I<l6&}2t0%K2e?>-IE1cC(V_j@4h$cx;kTAh
zo|?CKQ>+XQrg@f=h?HCi_M4i~c#O4{%Zm#ng6VL~a5$s}@_fiBA0vfel#j{soHQRY
z9*o(W?$XpHO<iJ)rYs5;%Q?f;2To}1^9@;JvCdG}nxZPXx|;Lw_1Ao=7Uel2Yc;E)
zKpX49tN`tNIh(h>^bUXT5C4S5N+bi6%4ovQ(8l_}otP5{0epcRW+AF=n|bTztsB$+
zm9|ZY{=IgJZ=BzD-E1+y+t=J~GJNOuXK`EK-CSpJ{dvLDW6yfe<~dyNV`r>4&vXZ3
zSii1YpW6qrbN%2FZXC~T;MV{1O^Q-H6k7&QT;GeI#g_g`z+|gA>pUxoXJlG7YKs+$
z)i;y^uPe%%<$>@#Kz~*BfzM4{_K?mWZZW*91)Pz&3*g6w2s(f~&lrt|Af&G)?BZ2u
z4~sY%Ub&PAsl0!u=*h>=9H5)}+5@3cOTT4oo!GR5bjpf<J`;e0<hsS=v&^p@>&LYX
z8UgI**Ld!swD2g`Z@IZP{p<Dp*!I8=bF0kz<3S;$SCppXJ*w3O4VBLsDkWuIQI<8U
z<%+=|BUK5FN~oI#htEML1>{KrBIF4b-o2^TC8^3hfXC73hD7>6M~_6ov^Ay+)>Xk`
zxnRDyVzHbvo{c#?yvuMnB26-sO2KJ_37%puB$aOrx-m2wzX3!DdNIZ)hZnx$P@+;G
zMJh>}Buu9>taB7)$#T8sYCcDan6I&&9lCSWui>gGIlnyRZ@&MZd2s(e_wL+BNkvmO
z96dSa!Rrs$J=kH@EP{!8lP!<(%vym8wbDD6gPy;BJC)&#i8*%SdI;-7IE!!wXKR|}
zobv1?tCuGf=Vz#<Mp^4EkEZ+HjS5f8Er(k~w}W<#!RATw#o8+}b~02P2>Pc0Aq>`5
z2t2SGQ*$S&897UCe4S2EIFW>;e!cuujy9Ep4L=5pMxlj7TI(mFuDN=C#9%sNw6lZC
zh8_$|Mu;?NH<-277d;Oej1JWd`UgheblPeA-6q}kiSvK&z?fk3P`buG+;GTT`)t_Y
zexf_glpt)S^1x2GpnYrs(aruloNY*j1qDhttkwvT`ms<6GR=@OA<Hrj)Lm4fsLP5`
zS@CE8%fILN=`+6i&R4nf+Urbrb{P*xWNF59ye3g8)5#7>`GYG@bF`@_7bWX;!Fsty
zAgR|ii^anGgy@>Gs;KLl$!x-FuRY}8U?2UmVK5j_mPLrFN^m40Yr3Cj=M)Eb9$-a|
zK_ZeAjl;>n%u4AK5Qp^>8I0Qw^>(~&0aMpLzxCalU$;KEHTRLd?I-Z;Ywe!Vz5e|+
zRf(VU9=Q7aek1y0&{J$|x5lH$us~yc+7l)F;H_KdQ400JU1ZZ_I~JQ`+D}j%$17%m
z-vn}-&g>%0PyB8_7hX4hA^<>f>;C(7m|}g#eXk@&uo?TjZASL}^`=lT*Gzq_<p2+W
z1Q659JJ&N)1+ZpAV3CwQr(1hp>zR`1gX}ATSX|?a%-c{|W;)wpG@g1;w_0`UFPt4(
zYu4*ES>_MgK|Wy1`1FNZyFR*y``in(V-O=>x8x>x)14F{SVRhx?ABR)3?bZ&u2H?7
z(Z^`rIVHk+yq=S?|NYjrK7iuuxXznf+tfR89riLINM8aXfQmlQbv_TA24MxM8Znvd
zaB-gDeX1qK`Sc~4hAKp-<{=<0RSDMC7~4=cCB}MxIXxJWCMilKB#A=nG5`~aphh+&
zt)&7@Q<s!Y$+{{yyS(J`@&bopI39BM?p;Qs5ozkvq@>vZQ&$L46RRaS<LAn1pBf9H
z+xp!8VRd_M3E}gxsw8JH81Tj$Z!noosGEk1t1I4n?+2`l5P%S<t<J5+weyGbnG{aX
zUh<<KzYkx7H(!5?x@tIodBJ2dVK5rdI7^atE-bgJ?7bGXpI*J!el6H|{|M@AuOf$T
zN)}f>8F6uTf?dtM3D&vRzOcd6M;x~SVrgY;+$XIiL*lI&uZZkkk+GAwkl|l=PBPSD
zGbrh2V<a@YQlp&n$^1w#G6kWqV#^PMIG(|Tguyz{&U#?y1evqsQqmM9r;nd9nC&8m
z6VkyLlcjC$IuCGcD3Q{c+jP&y-v@6S%pD>;+w}uuQvhv`$EMbI+t2mu36SAHLwI&A
zXwK0$!r>W0af9X>?`@q!U_jTPEU9p0X@V(zZaJyER;aBdA7;obWtb)28CM7%<oEgJ
z*T2D&pZ|iFN5|yDsTVj-ay~z0HJ*~lg!A+aiSV^u1!$|WR#VkAtJQ+@i*u@~V!d9m
zS}u9=>?uG0_!AyIdcgnT|M=hX=+PsbaO~{Nm`uhvEP0)ira6c;t@E!)$g_;`co#&D
zke=X|()nj8>z2Q0{j%K6+x0rr6V|M_UJ7h;Jzk%;Yk)lp^&9TD`Q6Pc_te4e{-XDN
zN7OR<XTcXcbY0)it*iCePa+(_n?s<DqtT8m?PNylgW@_Sre#u$L2FP_lE{S4RC(PT
zy3Me}ao7e%H@)`VDEimj26Z>h-xmSypF51s1Mi!jZ(HH(o7p4q?LWiDb;JLo2d4+j
zF&VXID{k~tZ2!LXH`~=%7AUeTXExhmI2_=L@&;wVe?BIfw$_A1%smD43!NUW)o}Yt
zhl@%;Smw7~sSobj`-+Zjixv#Oo;{G0QPFXszrm7m6F9u_o^Erl`uFXB*M)W8-*bJ1
z%x}%8c*VDS>r~2C;f#!t&#x9vX1}1oN$OfNpD&m%R@Ak|SVGCb1Yc>yLE_L&L)|pK
zY@?KDqy2u=K2x<60%J6JmXoF#sZCKybexnp15mZk+%y$MxumEHRz<<pYKah#4|68d
ziD&AvEC9RA=LnTeHwI4lR7u8#lsZ992m%oCvN<ZJLRbU|R>!*58s{uomNJ`7+1=fx
ztX_D)9dNRcG5tt4CjyVZ8r^Vo{ET~t_c=Vc$7GmMmo>}9iotw9KJzX|O7(N=_K6wk
zS!3s;(;dWoX>rE=oCRf57!d$#*rTpXC8e8&x-3~<UQ%3L(yW#UYeG2z(PEpfRM>{b
z(XCIqb=N+0yR=-?B3`Mz60noO?@b_)R-kZ5*C1TMh#G)gScYf}0)xb%uqYBC5W=CH
z^HH=A3g?EgBmgYHI7b7dDn#E|p8&DETJz%ZGt$|Nd^98-C{!X*Ns6-4YkX3AN78V3
z`fS=3k!}rUd#=imP_jEjV(VjmL8xFNA_5mJ%HCwuq@TWUK)0t_6cFw6`U1r+0z1|D
zNg%x`MhKrc*<aUR#+XjP_p(NLKL?dE7!E0#6*BSpe{6`1ZLE)^$_E2fl2R<!Bw5PW
zzVS_jgY%bXeEWBQ2cZ(~-haT8C(qcw{~FIf{uGg<U<HfC5|Md7ymeVomTS^9^~ozo
zBebsh`O!0;K6}Qy?|zAY@W=mvFTeXXYA`@s&Ef7tvNU16F2F&fYn-!fz=_tH!E~4D
z&VHyts7REgY1XKa2E=M#F3dSvQ>^RFRtF;J{af@t;e~Twm~X9&ZoUGD>w~=(*lm3;
z3iWu2b`7-a?dFKk*i?vwc5PX7(avrEleZmUE^3m-gp@KS2+Qry3!D=e=lyK7g`_pd
zb}k7m@Cc3()(Wf%j1u0*v6DX*;X~sQ(OIsi?e@D6|9$;+a}REhPych<rs3|u{?&mU
zEtu-6Z}cCagXUr7x%eIL@GT<azx{>qCD(g%qqid7Gd_?1gmbhB=UfZ?XnmNRgCtSp
z!y&tSdyGaSUw<@Iu8HMomGa4JZ5!DfBEUt@Y`D>G0!=+o)B|!iDeK!;jPiYBAh&MX
z&R{qcY`9NP++MraX8&$}hHlL4#{BiW(0{^>DeKAYey~LJxoe+Ibmwmbbx|VNnEYNy
zthQ98W-(uJe0;+B<$|K}34*2aazIIiNCGHu4&Brkr)h9Bp_E|LG?+#QKen3TVCd6d
z<vB75L3jvE(^ZeL#<DI}tk!eNs-moF43?d}9mbOhlj(#k%R|blj3mkEl2+3yAWd!k
zb%U(1CQO-gC`U&|1Q2Wy=tfXi73*S2k}8Iy5yL^w{Rj8Byu4x^1GAt#4{&T&&k1vR
zetFKt<vEMxg7IKR(=;p>OR`bMy_v6@p5+-?mbG)yDpmprDTks0+fQ5^pE0s@QAue}
zS{G_?n+DUA)Ww?B<t5d8j$5xuog?O#Z%G;_h|K=RQjKR2-P-Q2zgBO!#w>q+dnI6}
za3ofuLgwW}!EoqVx<l6Cf;JSoRah4?_gjZjkYX^-k|Jn?#0W<d(6AH&B?Q`8tnfwL
zEeR&(<RQV<hWU$AvQIu@G|t$W<Roc^v_6-QX}oi$lCn*p>MRIrQ6ge!)eTn59|vJ9
zqsyfA6ORX!ail|z`wihEm29(iw{sv=DA5xHr%x%=2@@f_Z<CN{YrIyf0O=1#sp3);
zA?-!yobEwMbd08394^5L#bCU{dUeLSoHHCu0uvYxjiBKxQnF--aUe${-umV@$eumr
z=<JBSJ9kL-Cfpf*nS6JjgYl3D_g_b;gr9!@dkm*zcJ_8y%rCh(eZlT{#QE`aTvKs+
zdCbnign#mlf55xn_&PfW2WTa!lq4I^7|v!?iv>7|sT-PljY_<qSJPOGaqR6M`f_PW
zMpHxT6sgEPIJb@j(6*PIe?ukcj;P4C&<bFQ?9~;FjKlUp;bMKkb}l0r_jB1D1~;5a
zH|)PYOALM<`?XD5ZPGb}_3VTS4!5m9xdv7@K~TqTdFE5eF0Uf(IQS+-$TexCuiRmM
zqEQtWrxgh0gCZ;%<4UZnQDO`vJ_Uy~IH%FZhVnvTzqkgR_Rr)xmd!tRbcv9?x#{&g
zjzM^CAL!A659d0+v$zNP3XdP-R~LZSS_`Cli8=egRK?{Wz-)m#XxSDgrFVRn;rV=I
zijP=~+I$c{9xz1F?hmr)w<SWZL8m=SW1U4iOXGZ^W~T(h$$<UCeI7h`=!48;!g{e{
ze`klfs8CWe7-a12&KM3dvP6+cA8e5b<yDCd?A8Z$(JmGYlfjotMDQzZThd7cmPpt@
zg7lm4Nx1zr<Gi=37Cst0!Mbi2OJW);f5`WP%x}VSQg(rHLWDJl&)evYIj&XGBJJjG
zyj~|+5aD~RvLV`a*LFulFc<&rx<5*Mf^R3lN!E)Mj~_qh<m4qM$1l-NQ<=c<+L|PT
zR0xbwXe&rliIa&361t`;OV&k!k_*nWE6y)3na-w+$73cT;0>h$K_F;qgVFw{u50GY
zIdxs45)ZEKKDfvJodbr$A%nqyBu!B&4^=CCMIR>(R`@)3+QMp4Qx+zi%i+8Z!{$u1
zXvT4NIA?GIq^>!?y1)p(FZT8h_`To%F6-5bXU~tAUo8TVa=tX0OlY*hn5I*>Bv|3N
zTF-fL@q)v{yJ+LMTFsGHlKZb8aPjhtot+)>-OQIGb@($-L@K1`e;?Po?d{)IiL!{q
zhExI|jYC?8(KU@;QP*>t#VPfRQ}p=?QD_M11Z5%+2{`9jQ>z6|<7Gy8#wcnmU=U~#
z_O$TU8)-o{z!oX>McS(+D7F?SYJ`F;cig>m$o<Ja>0-+6ehFO0b>3#XkIUT;X)I+D
zKGSM@@pB?*f+71dE<`qYU%gZ~s-oib$#Z7AJE+}#R5l`;rI0vGH1!5XL?Lk2qK#)V
zyb0DKglYku-+6)vvx5;dY|H?$TWzt|;z91(nj_sI9fW-}jSDMcm9SQ*h!CBtg2RTw
zrv<SMp+u8kA)bZwheJ1ZK?`yC*%tz9C5_RH$1_fy#8`vYmQ1E7>FIrMPBSDSSDw*6
zd1;bm?B6+Hb#cM@%QHq#pL6exFOelVVzx_~W{_mO@#sqoW+PN0Nrfae7KI^AQl1|@
zXD}S_M}PE3j3*;6T!cU+DS19%cXtn*rO_H3lvNd&1B)?G)dq{?!D}8wdS4Ue0bMXX
zxfn6oGpXDJv~Dt818uIw%}I<K+I5Sv|K1JP>A>Q46W4z0K;mcn_=y|-6lKoMU+Xaj
z8?3Oq5wON)gGTB7v>aC9tGW0qwm0;GFH|UPuS;tJ+^AKH>pf?<j_cNGyY6p$cH!78
zz1BTf3ts7ssq4nGKS#d~A3K8l@b|G>Z~J~qA#fqLLM+f8WsXDmKrtJEiSW5%LmHbl
zO<;G~AmDA!r>lK82t)s_#i6xDHxBJH{V*7gxcA@zZ@>FBzWwd*kfs^S<&rlZJ>s2r
zzJhHuLJH67<q6togpj0ZN}l-Ajv^$mYG=&#*W>11fX-6T&IO^AqYDbFq)XX!L%V4|
zfe5D8IKKqZ+(&b_Ur&g*Cg@nUmimawLA+j6rn>vxc-_sfclYYadg$++&2J0k69a>y
zDr+uIPx;`3pK)<<MNw2#b;Glh=cI!*!=Xk^e21h={2EFW4nvj-tW_ABpt6*vEYXdo
zn9sSmyhI&KKfkGuh}LbrY>9QD++RqIG8pB|rc-7+I~*MBvv+5o;c!Hjd4?;^a`H59
z1KOIprmAZiV`z-y`O}viADy$9m*8YvrR`9AfEQi0f)Xb10d-Z)VzoqNiX_b#jwal@
zd&ocir~ioee((c+`2PD`o?n6U!FSRrcE(dy<(j6ceYGOzSgzNcygXqv8uR5Zzl{<(
zsw%11Yn*XBd;Aphz%a;%Or~RxBV3R(+mjpDRm36@Z3UB%Q=XCZ3YOo$#?(~Biq+LQ
zm&Zp`r)P*&iP9DliO2%`8ps`6Ikq^3c3rjLPSBO2>-%PJDOz8&y&7XDCDs-UCOL0>
z`3_%w_a66m66%+G4991{r@l{8d8W%6Z%Pc>N)*nm(JE-V-r+Pl#d;gP6*1{y6gG|o
z1x`?26}<fTImvX7;b_WeFvKc}RTd)+E(m#r2RPn@WDqj(gO+suR`d=J5&2iw>mRV0
zB(2jjyE>uqtv>C`)^+_xiwuOp`Mg6Rf(Iv%)^9#x1h%o%&POhlMM+teltoEV6jXIZ
zRr<u0jq~}BQkjw`l0nj-M8P1<Fjlt3kRz~j&f<hcDv54_HsuV1JcE3M(2B=D|Co#O
z6$f|klOzLJTe37kH3qw^&~=GkEs&L^EY}=AIRXQ3y!Iw_U9w(R<l_Np>Pz7y%KK#1
zWyN~EX1-j|G>sQxDy7y9>#8A1huk^5??dsm_C9vTmqYHr#HP=apC)%ru-*iswnNAJ
zV9PV2@l^OVFhBo=TRh16KO6OqZ^e-Q8I85njUiW}&Hv;dJ^FNY6%V3Gs0|b_n=S!S
zoA&+cfjPPZZ`)__^Pg|ax3Oi!ySe$bB_!@Q*Yj!XD*`=g$NS9m{k6PV%0IjJc6NgH
zz1XA-=Jm)v#GiQazR5R8{1`Oh=L3v!KBC!q^XlII0k6IOh_Af;E^oc{C6rXGRx5U9
zGY;<@c=;;*y^U$Oyt<@m8k_@_B*As%y6O8fTNIDZ5#7Q3+d*mNHf8AMd&DuY&U*H+
z9pC<UBZTT5dK=e^tF1R@Qrx)y`VZmU&9XvtAGuvo><`99A0)ylTq7{ruvlDietyZb
zXU{o1J7=+6aW!A?^7JJi{^Cd6dpO|Uy%DqEe+(8Ol?Mm$QKUe*#Fr40lBTW^Qc@Kq
zwboQk&1$_yNFVtwr6fsGZ(7YW@;qlS7&4iR+1c4)Hk&cq*<p|mNRyOA`QVIHr9ocf
z8DitzFv>!6cDCTBAAG{)`HH$U#1fnBoJ3I&{_BHu;VfE&YX(=5x-L0Ad&z?b5BR$N
zDndHm|KX3BUo9}&VywnF!+12Hte~oF`~JFEGhZxd>IRu4n5LnwYwEJ5EK1JL&dIWz
z(P)TBI<^&j=7r5U+!B{z@&y5?#-IuuIEyn4O}VC8E?8e(Vv3T)#6<Z$MMFl*9Nk^w
zUJ>8|qudUm>rAmCeCXBVVSRfwA3Na~Pb6P?_b%W1oiFp~&3#6RW;JnS*+alniVgmu
zd>MP%dY7(;N@KbcPm{(2f!24!T8oty>mj1oezie4L5gH~e8K7XF{8<pEYFeqJLo|g
z!s#W65GW@xK@+yo|3!qTrQi#<rH%#AX}z7qcDl3Q{4tGnTqK>-uaqHevyg3u_b$as
zFp;THQ9^lO=bZ3<7uI1Li)}RP<t6j0D^|-Di{%B&)ttKWC1=XAL~D)Ky4@gJ7|Ocx
zInIqC6%}84W0yR+$2iYPNO4qHnIKZ1qCtE0z$bZ@3aJc|gls%T+61W4s}i%W$i@ST
z%PY<o3vAPn099GATCHfBl2x$;ZCNkZl*(`F!6+w5)4&uYj7B3&qggGN6vdj=YDH01
zs3f6mG-YGZPV(BrH<--!0p;_fxf*9Q*7ek#&^8>};TjgZ=@7YTi)=XLx1Z@gd6Pd}
zWUoH&e*b0Lb&$I44E|ZoTA!%cSb>$E4eDds2;^E;F78X`O%t}K>~y^6W`A(>#r`X_
zj{R#*wqg?`I@guf>jy9Vsx319F+gFv#pl`tc5w~)158SR70xsDy>%gk-|rg(wlR3F
zJ3j(bXJfcf0NPqRjPak$2SX0;-sAPx9`WWIZ*l+r18{=zXvHAU+27r14;XN))=R9l
ztc!J<(^JJ9XkU2!cB*XbQ><6pf;Vx?{@Z$Q!=Es&w~aZD1LUvWy|%R6#(r(T6E<?+
z)j{-Re0!L00O?!idi(d=hN*QH7eE45Vj9PKRdW3NCC4WxJbV6}qO7>My5hx)6F&Or
z7ku>5&p_U1JUYbPAGg3r2!~SA1B>8`?MoE;QjJNPqLLILyzg2Jia~%(y%{ylb0(81
z)7g~Kc*JNlVmul%n|jmjcrqqSeLlQIB|fTF`TVfj+Q1H5)~lMAFXw#x(GfrS;1lLo
zYnsNi^X-+DZYW}+o{q+IZDL?!Y)C1jsj7nG<LA8g`a>SQ@jCe+V?Mv)<By-PUKZ#^
z)0mp+c<Oz4ouRTdSQmoz)--iPDsqf8=tfhQHCh`kFD?;MaR1>w@+5E9L3HuBRVqb!
zA4nx%!@bo~{TkIZx+tiZE6Vu-TQ{CC5~5A`Ek)$PJ6zM!n_IeoSA^*HJ#=PcFH~+>
zbho!x19r;r@U=s}{>?Xd=j*RC*-g;Zil&?}%J+Rg^NdCcL|}zvIC%S|g}SKPXbg>1
zdK>^R(2NTJOL*ifg4UV~g{};jPmUN(Ck!Sdh8ZM-J)a^&ga{j{5Y~^Rb%Y8JY6B6+
zpACoYXhK#3UkDvPBkjqzsr5zRCPQCinY0*8;Xz#}|KfxX?og7tuBobua$R!q@{-Go
zD;8G^idEsMXIoJ>rFWDKjFmBls;Yt$w54$kwb4kIk^oj%j*rhcJQ%Sv8IY)yOeP4^
zkS0lpN^&Tb`q22M#^*cIB+jwBv&;T|&brp@@9r}>IKY%OS51Xcu)dh1mn$*}S)S8W
zYgR=`o@EquNs=a{d4|fo@EH$>IIXGcnsQxGSDxC})?lTeYBZIB+9=-o%GXh;_x*F$
z`xD`;54LGpqt6z`vE3w;CAWWnb2D$IM7VyR?r_)+kZ50foA<VUN59POeKsZJ*4=J9
zeEPn2@h3XSM_(7`>8@63G!fWo=PeA%D_bo~(d$${7{IxZlz4*}5v@RMot15Sx^}|y
zmKGC|N*Twx`-xjr(v{+g{)zq8Rekx@TFa1`O;)(;Yp7s=XiYcU>&V)u)akAFebm0~
zgCly(U3=a{P-$tQo%2;@s>Y%XBuUEQy$5{x?YDX3%`b6qaL6DZdFw=O7!Jn_^0ZZY
ze8vGyk|>l?I443HiU{mJe{5WDUZU@(agIh?n#TKCCQ61Nf}4Te#@Y|h9LuITCw!7y
zp;{&=o*BKXY+P^q)rmeE(;em;@_nn>8}?QF-sS^rjYni(ydx#i7Z58j2I|W2!3V$K
z$&)9{mvgk!oS$FvgZJO#)5o83bo3l&Yvz{|mWv5hWf)`<D}?t6SCSNgQQ>J^ia;Uq
zP;q2rnM^1C3@J*AVojQ)?Cc#d4pk$@<1yoLjCA#BQzny%uj`$pKG|@h0#M8lBJm~a
zv<G49RmqQj_#xka???R2-~5nY{QNn~RfBUrKqFBK9cbIAR|RR+4w`epZNxdsqM|6h
zQjZWcrsgL<`3Yb9+SmBbx4%Q0B>efG{{>H<Jm>uEf}&nBvIEAWA){f=%NH+0{=Ax^
zSW%V*!)%N-4&4~)qDCP(KR;)&n)9Wvyv2UHAL^HT+D1AN{N*=g^2R=<=Z_~v>py0N
z!xS~mVo7y5r<oVX+ITsYD1Uv?I(hkjTYM1H=jokMtAh@gv^fwz=et|&m4MxBH{+{c
zdy{v+`X+~WchRz-u0iBE`E&}v^v(fdrf4pfB+ifu8JtgpznqO(_64F%j_(f@)#loX
z0?!7Qrw}0(k|JB@o#^Xz!R6CujCUuvfnqXDN#q12edg;(XFDOorV}3ANeLkfds~x}
zo&b^7Z`MGg3h4r_^BPk($o{~TG6}*_cCLFOMzkt?aE54e0P4oDUaVLymb^GQVSX`Z
zy;@V16{ay>h}Sh`v-YOVqF}LDuv)EHEEX(ROV8p}6_u;72!vH6NHW>*;O>w&-`M5y
zDq}hxW1YcTXqp-&y>E@RK69dTHiW55?+7g%DwPamMpLeN@#&nZs=yjj<qs?69|<cg
zQo&%Dv#JV6B>8AaHW-9O>Dh_8tSHNp)pE&dy`rc}pO?-VDq~oc4X?fVE)O1k8EsWy
zg92+4>HQvPX-EG?(B<1U(-t+N?eA>+Y5DhVzFz-Y(dTg+*j+Od_t>PHKkGkd{X>54
z0oy%S*PlM);`fhh%jxym6M+l<UcumFK^uow0;S>=dx8bO53O~Gm_yJ-^L;OZGxT4?
z@o+bf;b%QlJjA<d2_3ML(ko_t!T2id<f}9j{BgxL0O{2$ZRggyR_OM@wIIkj>3w`-
z!d4N1YxuFYV+SLP(9iQpe8*WDW2ozfs<Fr<XJ@j@yI=bzkKXtSJG+Mnk)cf-crSZm
z^*28yq3$d(t&C{dI>hbZX1m3{x&Ypk$dII1Q{kKe=W`=|9?RbC3vapDAV*y^Pm3OF
zyT!VNgRiZB8^0SBiSYUM^_Gpb;c5^s0k?iWaNBr4vgnIV0Cc|3>!RS|e9ou8c*dts
zKA|xU%T>Ygi(~%!Z~k|VUmUSoEzw$|1YFJ6Twbmy3&Su^AW1+bDB-OvGP-eFMWXUx
z`RT?=_k3QblPQzwE`#BSJRguG30a<zBngR9D48%A4w1_HttS$nV#Nxo(s6Zh#pUHC
z^ZA_fvkN}>;3xdu-+jRG@j2(`E0#qQ>Jozp%1;tfcNyST(%gZwP*sMiC=vKX$JS2x
z<l~Ro*_m;8c*u8u@ApwE<tIP+fS>>TLq7S%V^+%r`MzR29aEKKR?E`6gES3VHzCcL
zzrNOnG*4Kru2`L|c=r4W!%0r=1}K%1CTUMCr#%>2c@ox7YsMGeca+vwP*WFk%F9ct
z%PT0o6-x<0C>+$LyOB}`-#mvf4&k>)XJ6kOo{jwx6<UPA-tMk?YkMVN_uA_ReCf-t
zad_u|ELAj3iLoG5N|p}+7~eTy=d}T6#iw92&V*#FE*yj{YbV=70N2%@mH1Q)Q9rcK
zVd8#vy_6C-q;V`>yyWG_Iq7u3U_4-$q)3rqodo3_KrN9S5pDGd|KkB)TR>h!J-la@
z6EdW75Quh0I>yeg0oA1?iIYpm)=8OqUlgENSIp;gF3!(6eR0aVSW#CsMOomiPffB~
zEqQtRl4r-yIXgdNu~<+PWn02)!we=&py?A56e$f(k8#dXHx<kEg2{M@limbv5{*kd
z6)pgkct19kNL=b@OoP=0kRYvW^Bze7wRRZSpmKj;7-L8WDVfYEMS~Ox*(gV52{Q4K
zS=Jg%V<=W@%5_0e6qw*8y<Qg-Wks5g`T94%4Knj7avXtmwGbk!Zz}b56Ue!>x$Zmh
z^*Ru9U*t*~o6s*Cal^GbHs&VZz&@x0y2QV~WMpS6NFuNb{&A5>_HPI2+U_mTfnBf)
zP{K1rTR_@5>my%7)b4gpwRrz{+O$3<!fVF#Kf@;Ij=x*m<(q7;6GljcpMLW2?RDS-
zgF1%nmu<H7t-H?D#aD^M29((^RkcysAOhl$*Q*7GvM1ZD-AG{)KjW+m*Z0hf)|#?v
zSgmUiDdWj5hj$LSckdyS$sU72s5vT=ps)oc)`}3r;P0=s-k7gGfVm0Gv@QzQKwr3q
zlq8{kccL@uy1@i_e-q%ow)bKh7Z-cbyL$V5(Sn$5VAtj&4Ze%f7n{JOimiNiHz@<x
zo^MO(Yd=HBT8kbdDPqNukkEFuTJ!wqB`;r`bMoSpvZ*+~IOpWWF&}>PA<vJWvo2O)
zt@?GaT9v$fIp^`?7dUswbdoa26d8%PbXk8)OMK}uB~z4AElX`fA61gDv$MlsFeOPn
zBNxX?Ye!uhnz}@vUilnsb%ky;x@l;tnzF38x>|B^e#vsVV!2%K{OE+^qYGB6hO+hn
zc|H)xkAoDnWgT&`N^6|Q0M4CPEC_*DO>I!JK}m@T{_pEm$#TBp;e&^~_10Th>yyy0
zR|RJ$r)ceSFYfN`fPt!L!Wyz0DZv~*Kch+nN;&F!MO_#89#kx}WjvjD;MUH4e~n4H
z)YKx>7xzrC_nS4?vO=%dn4&~#gRpM1WSyj~VK3sEa=q6Lb52-tz4Z<;KuNUV?sGN%
zt@cX5?wzl`#_NynF&&Q)(jlox<NzlPG4*!7GvVP^-=es@qN=KPqls2i6<vv!R_ON*
zASZ(Pl%To!5@UXl<L31nEfEsalq_Dn;Np`ZgHcYF<w&8hX^vG%TYf5LEf6aB)`cwE
zQ3#1HY&8sRh%Aymcb^ls1*hEvM5lGp^*^K*i5n_j*Ez>(RdRfE!pq|m?+egqOl>e)
zQ&$BqUY_vDryuj=*<&uQE?8F;+L#Ru!$+U>jX6MXJ2@c9JdHn{PM9w)$#aDhl0>B8
zplE!gERZNcOb~2^ps6*+Hb|Uj52a%`$<U_p^uIP98wf$I8wyjfYSxHUkmV`qAVCCg
zeCHfZqiM>9#pM<2<%+WMrN=5$Q|TIQE$@8wUG6=24U!yf0c<egn{eRuHq90wXc>|(
z&^{Z~eIa<*4tHceZer-#d$0-WZZgNl)*iI4%DLax#V|e_rIh+8$%R~K#^JLq$N)sc
zHf^gRxK{XWwjQ7n*Ufn%C{As%)|=+7m8OBY3K6*eZPy;A4iBbWNG^R-^?=V}WnQIq
z&l`Ph*jO0t{oAgKmi2NyUl=Fc#=LL*#HWkwSKPZ{?5%ZPu?fLnb<<Fk6-C*QWkV+8
z2@f7TVt@ZG<ME7qFeb}~BuR$R4Q&!`>%2P&`0M~$ci`DiVQ;RkkKJaUd(Res5rQQ3
z4*J#@bW>t&<JZ--&2(L{>8=}A{TG?OUcj3aUF}=Zo(<tX!YhZjv2WXqe!TrnG9b#B
zuChY7U(D5rH#ato;q>%^Cy$?UdV0ZXS+K5_JUe>Er;k73!;d~>v6yc-?*pvYHK(U@
zKKl3=ArlVvCDSpCN0KZDkw{c<@A26NGNkmbG%5+8PbrctCC$bdr!bYJC`*j?Iq&M)
zR|8rtSFBbm%A%kw3f4tMxn5JOz2EfJ<%0QqNmbS1J2l267$azmFGVPEfx+_S-#p^+
zYsiQQ)cd1K*xw%98Et7AgG#iIDo0Tk6<3!FtajWzyxS%+KEJr2TrF`}bklIKzfWCM
zIOAC|@QM6oOlKy1?#(2XWH!Nx21kQ7{%kuxyW&y)7HOWgXT9%c*eCsc4*`#Kr>ij<
zU6$ygz?3Dzm<<;=FZSYk3=1ko**ZLfcau2Xk~98P_Q2Um`Nhws&|ZzPd*dDlhr8s1
z0hKKgHpLk~sd1BJlL-fpUZXs};K|h-d$rhjinxs>Jwfc~fKb>tamz;t<G?!<22Ct}
zFH;His^I10r{t3{BF~v732wTB8)SY^Y(G{i9(1uNw%TMGqNs`QIgjTAF?5~YmFTf@
zA$O24?YzaPA>H_dlt)J=ym)y^Q`HFVD2jqcmwf!mFZllZ-{;x!Q<m$6j}B7;Z9`vL
zJQ*T-0`0uf4`@fLZVU+6KiK2^^abnHit%Jb<w`_IZ;+=F-87JS$5|-_HgO0g!Ag`!
zusS5VbzW1cZG{oq=Tx$m#%dbZkYrw{s6?Ss@55tF$Z}sBmh%<M#R6@0TPCF`8>+@|
zc=rLn@tfblp*(05-cP|FlKyaO{TJgXYyfhv9Q25uTU%rTZf6r5-3im7pv`M<GDWw6
zJBQn>*Lvf9UAvj3l-{57hQ4h!(R<z8SN+4c51#sF!oK&e!~5D9=YxwPu(QrH<!v;k
za|mnv!6IV16VDuPKA-QG2MXbEZTl0)i{Ac-?{|6r;<c`^dqFmMvR7<8XTKELui9cl
zb0u%~p^IQWf=d-_*iy-B{dO%%-kSfg)LanuT@?x29uhv8uBECgR)vowueIjhy@%XA
zyvti}yv4!6J$82Y+1c4+XSPe0579s(Vg@*abJ9DnI~&pkMQIu)$8`$oO>KKU_Tt7s
zNYZqO6v~&oG7YQZ(q|&rw#m6(6tnC0$Y-^i`qV1RL2*EcuGVc_UturDyoT{ApUKp5
z9qiq<7vrHFVS&|9l@(`a=ltNk_c%Mhq^xU<v%LSo5BbptAMo<@1x2}zkKZ=qz&4Jv
z^EH?Ag2zwJxO;fO;lTk9?%n0i;e_!lWss+gGN0*zBtfVYAyNpzg-zpu+{w!YmlsQ}
zE`2JLi}MQ>i#bJEur3OUb>W#?>wPzk4hnos!_)im9rf4O)>0XZwhkq2n;ZuT(ssV(
zPWZGa@fsV7s&j2QGTmq@0Yy=<GZU;91t-TZdG_pxH{N)i*B-vkV2~q)r7qSSKR=<a
zO7{16sa6%rBxGvfnM&y+g;hxIzb+Oyp~;5{`A{<38By1o%d0sbeDFTg*$zAVyJUm>
zGv``3kUHQG>u}a!OhvJnQ!bV?>k?tC_ty)1BC?n=FqPKCl7WOly<wk;D7%9~DFxDc
zGE4v_k>1bm71}ESyN8bs$fr3*Sp*UVB*Hs)S{ZyQ(v;-x0ekPh#ro=;^B?>eS89Zk
zl%~eX9AUhfO*j|)>Y^!Bf>2m2weZDy)4+JOQMA~$QiKyJ!b(yM`f|<Lho3SW4jE2n
z$SjB908n0Ats6{k$y0?Ci3bkCBYc4i_1HY?;57lid;PS9iS6r`Q2>NUTPwV&&<Y2M
zAQd^f1SJMki#1O^KIX;CW2|d%(owmFC#O&O-uM2NpZ@G8tO~D{w8yjV7Mk5qf49J;
zi|J!Rm^;CIZ5??!;czeI<;jt!zQ-wAH<ZN+tVM|g-8d4G0LBUunfUP*3F$!5)D^XM
zq-g?JG(LTaHX6}Lgz}NEGELCBMoLN3XsmV=O+gb(%Z=4&+fY^otM!tr<(kUn{KmKb
zfSuVPMdc8}Af3ZF?NhQyK_V1_=m*(~=vH_~Hp$^a-C@~dzZ}uc9b#zt<_6h3jA=71
zI=X{b22)G)O(KBKm|mncjSD(3*UnQ^8vO4YjL8kH4^}%6>JKxYZX{~SK~BW*BjIZ;
z>maa&6bLIwlp>L;J&PqGT*E(WBS1N5oFj2DD%@caG{(_Jk4xcMC;=9Owhg{Ms(&Vd
zgqW)>YV;U?=sJf9NwxcIQF{o7QzZsiIk!D8{-BVdm}H`4D3=yAT7vV$=YrzTZ_(_G
zxqkRZP#1rfH<YLA1L>9t>#>P8Fn%t$RCH{e+jw@cXlq!lRxFoGin3t3v&-GP_c^?C
zhr`4B9Nf9b!Qow|(_JRxDOsN4bQRcB?>5r<A8jZkf!z|a4fOw_lz<Onajg|4Q_ANN
zlL;1s6Ag8<zUH$R_fu4^zGzd*CwG=H$#cutMIY%dXw46Bu$F}NwDDklrJ-Gi&Rc+-
zxDG}`>*f<+GqE@pE-x=RK6=4VfAVu)ym*Playh@^!;gQ?_kZvM?-SiL@#8%)<|UMq
zUS4SJ&{rkP<q1!ooH0l~W-=YHzdvGccf_5&8ME1x;mD^b5mIusm~(b<#bU8yu~@TQ
zFDcgzwKg<HlZZ}XBg!SuT!0Xvf=OWA!gG0l@HkeA#u~IWG{&HN?gZda&O-=+KR2CJ
zK2@2;2OYM$Ovl{mCddurSX?dGo$a$;m8|A#no={!hvb9In^Pq?WBJh!f5g+rPr*8N
zW>Y5lm^>YkCaQJ8NUTC8P&EZbxu9xR{N!gp<lV2mjZ6}<Op*-}R2oZnZnk))wB^mL
zYjaII>lwrKYRT&IlDb%9b?vv$R$W;YTmxE>WUT+`C;YpiwJMkM+tSw|vLw?HGAV%O
z_AhU>R|}rln;{gfKx!HMAn*Z%RHTy;<9i1jzV$lQ(J{^QQ=EWADQX++$34N<*Q<o^
z0S-36ocI@iD7iMNvA<u4La`3hRIJZ0dHMJm$=)uL(GWGrA@vSnF4za8klsgzPCx@K
zl^dP)BaraSg8z(AeIFYCqht)o2(L=w1ZZtZ)R6VO<dct{aB}<-V>QM#EXyVD|MUa?
z)BpaTxV*SP8_gG%tZ_`DIf9_&I}hrl1f=sh?wW?_XhO5I&)MaRuHr%F$6=5sI45yJ
zktPFh+Rv*}<oS>+%c<+i`(o&tIw{ez@f#msR;3QU55`&Yd`MXqG<Ab;K4J2Dy=Gmm
zJ=24s48b5;Yj*Y@vAcH{twZVxz&P8{*a2LNE!ONNZMzF*h$cbTno8n~eU{m@Ke0YC
zw$rXQ4#gHIc5A2oTV(3`fU~{F7W3n_0wv-alO0>V^@Lr2T*q8=hmLIf6t1)hFxsm)
zl!SbcfwLj$XOKJATkhIJKTd&&>pe<2&j4+<xL@Mrxz;S<pFJoa)@_Z0WjhKvuG6;1
z&JhccZ^80zZ?*^Tuh>GJ-M&f2wX?p}_n65y-nV~$;e#i-9C1Bx&`YG=&O82aFwXhp
z?)W6$>vch0YotgxJh;o<J9pXNJ79OV&)&{1v&jyl!H7IfNu&&g9RtwyLNnC=jfZ=%
zmxwO!VVmZ@KMbD%zJ&AM#?A>}i9-q$k~|+U7>omMxSlTGMqqbM)&-^_R(Oc#LuA~-
z?;U{ajc2>=HvOc5VCk?Ob4rx!uFdZ6@817ge119wG47)vTj`DC)zyN>kDqe%{Fw8L
zOSCbZoWA7w$umCu=oc(jOLWutGc)4JzRnl)p!l{m-rUzThDy||tCGcP#fz6I&qikq
z2fpN29E+k{vo1=ix}mAP@@#Zh>)5E|7B<$e4<EE8I=~&r#04KwyrN@_MH?G{VC;JE
z1&#H(g+MAnqP(u=6bKgruB>Y<lDhG=TodJ}bj5nLVlo<|b%hXMO@m4md7d+!PI>s?
zJ{RYw6y=giuNaO;s8Px=nUE$~ka=C$>1w^ADAttKikHU?$4Ae}(jjH3nayUt!iG}q
zp6=6>t^aUf*!<S9IHz%@#*}NC)e2EnB-&?euu+$_zA}XjiZuvau!y*=Wgq3KbDgEu
zUxUuqIuW=$Wwg7CO6Nd%!#v*HUJ2MGN?}7)gw{DwbRZ};DHYjt!tAyC9De03UaU&0
z#TpgdKMgS}v;TazRc|>cqfpvz)}TT^QX$c0&Dqg0$=)t%w!?ThLXJhq>}+WA0mkUI
z>O=qD-Y+GNy&vhwviLJaY}Ny9pS6Fu3L2xyl8mx3?CwtZ!QcKBr!UWO&L>tZsx^Q1
z=l?5z``-7do4R)(eeuMt?bx)3f<vJ3jHB~qSZsxH4G6=*oqdXC#q*;h?jGLp4yQ69
z%e-1;jP}m)G6|7J8eD@A2~wmeX=$9s2QtLq1|OK>htl8ISj}p+_M6%omg@y&St68w
zwxX(7t=3d^jZi7G*&f5u1cyMWghqQ%r;veV2;l73abSmblW$%Bcjr1mf1AnHZGGqZ
zXX20Dy7snNcJsmFLDl#3y1i{3I=7%2&ua8e5FyC&guVTJv~E~lUi#dI#$t^0^Vu>A
zw0?Fn_Sw<y$<BW;_P^^F!tZ9=U#DCB?6a5KX5+Z8+VQemk&`!!|F6Tx&h><QUxC<m
zyzl%2X~A7oD2OsK40{WCd*5+=U!(vOz$cD2;ZR+L;38uzc{b!={|<Nd?y$SF&+crG
zo!JhP;g~GXNtN=T+(BffIuPY(0S-}_?lIc|ptg#7+pW%jwQo|B0fffG%7Y$Bnq&+H
zV~lk;2G_>c3bEmQ*WN3ZPisr71%)nJRoc&YCx!o77+KeW=N7FAATVJ*A8CKBhz|H}
zyxmuvV%H+56TW}S2c<XzO=CDYIpyio=e&4v%DVIk<exr&#^Yy?IeziNS3Qbstk~Mo
zE}D+R`=LKJ>@Ne<#;`7Hl=S}4O3L8)A1@IYW_(npR8hf?z+KSaCf%O#H?utkvjunI
zEQ)4(=N#7HYr?udumgnk>mlx21xa8b32qY+40@kl0WRjUvozYXVAX7jbsDLls!MX2
zktj)?<;-?w+&Mhpa(>Rq@ezaJF1Z{rOr}0Sr1wb17?#T=Ras(;rd%1W&M(OF8e=8n
z@t7>nkg6*T&;?8QHPOyTP)Mz_KA&n+p^Ft}wL+E+srG4JobY&G+E@<K1%*~%!jD6=
zKDcomw0DiScAlh?BH)TtAx0VF{XLTGm{)19n8&V*#PKyw`>a5_biCOmNl-g8cE9u{
z#l?b)4?n@Ks<3I4@4WDoZ@XQ?b)<*~0&L0dwncdwO|(gpBG44eg0rX380_tljYf=;
zghom%2CEx3_+eTR(F#5P93A`NCpv<QZUOfXDFLk$Z>AE;b;W~wukoM%>Tj9P=UD5R
zu+@6W|L@QKzx?3GKcLY*nWT_};>VwNkjLSTV2<9Q0e%<xkPl;hR&N)8U7DucfAEm1
zsd@48l%3fOtt~<%<mnJ0lpjCm=pvpHqzI0=GDXS?DO0jEL^4@lD^QT2q{K;BE(>&H
zu+Fhsu2`;@NFhi?O4-!RuNJJ!0;v>AWsIl0{yr)tvSO0Re)fgXHVASXzdLACPuP&R
z@HUr(n|x@tedF5Sw-51+dw2Qn`lj4%A$lV#xYhT~{`A4$XJOj#+#5jg=I8BLOuw(v
zEMs?X4_(((t0lTLSc5;117m1yD0UrkC#t{#`GVHl0JXhaN2owJv~C7pH-qZ#dR=nw
z&sKPDo}*vBxn88|cAsy&Ph^u+59oDQyMBWf2==|Gy*b?)sD;KlpI7o~KBp>c5|uHY
zPPlvL9)|~a+1uOaVD}EY)4fm|Hn0!Iwdoa{GodU~=#uL?;6a^tNr|RSDI=J=eH{JY
zV)-%IURQ_+BvOWh2@Zv^J2zawm&oc3U3=XQRJvXjl&vxBKSK<5Ybk*qxWkXJ6%J4c
z8NJ`*ehsk3xp2Bh#N}F`+P){EdxsB<6UqbHvTQgzx#DmC_Iu11OKPpL!ts-z|CAs9
z>;uj(&b_%hkRi6mhV`3|V%~hdn@$34p|%cR0)RHHmTLF6Kst$XgnjDA$=;HWF{o!7
zOwtCZcO5Hlt@6sc@nF#VtcM?ouYVj&<qi|7EBXEj35u{j!~Sk+jlg2G2IsiExMXi{
z2bDNf;<&s#V><OomD4muNX72nE)QS3&&A~_!*NRLa#ER*rnx67REku->dM90Io4@L
z;}N?%Q-*^P#tF1G2$3RH($*CgfRlX?3n<&>u_ORH(^TkUNqu=veL079i3+}n-aAJi
ztsrsn-@(-dT42`!mJU4jVRkIokU9`TBJ-SdcZWL<?vo8Zy1sZmqrDQaa{<f|7lYW^
zOtF(ffl83$38RM(xVx-qmL-c%pCfd)v7%W50uA3jcXyz7hwXKNBZWXZkNmw{atNzg
zzBuFZ$y0V_QwGBUZZN<KiLNWOZb-767?BXob*`uN_5EQ8{dDn{o^Qv-p%b*$^}6Kl
z!F@jZ@DrXtdycL>4Zkjy{P|!0ue|sEdp@Oza|q<-#%{-t8xQr!jw+=>qEU%84bB?U
z#D}+MiD73dQA*G>71nB$k_?j}Z#;UF<22>$?2Nkhk+HL>M1)$hO8Jsz&PlQ~N2wHF
zY+WN|LYjg|a_{(Bm%*vtp|$aO>Q?LE#Jgs`xT3CW2E!rMdd*_BWVKp>6l7USV+A|A
z2OyIc)G5`Cze8c2=~xh4_cs@#2K&HbGZ$5SZ*v@-Wk6Kj8ifUsknWZS>F$ya>5!1_
z?(UQh0qKzLmhJ{Yni+CvhVIULzWa0jp0m%MH`aRAv*Z(i<7azt|80WJrhnDwLRrM?
zR&8-6Iu}qdvS<&RD)N4J{p=C1$sC_DL^Pr6g!8MvM{TW{)E*-|^37QWI!8*9>@OpI
zT!nhrqvuUq3Cv?#cjK1cL-<Alo+d4U5X-{IS0#PPBYOp_H+NzqKbT)o@aSTj?OC5i
z_ZKVn87Rb)xYPHIW?x3(`^hw=<R~wyrlwjd?t{^&nUHzlSx`sB9UovCJ+(_)vWwVc
zyL+H}314~O82<5>3s6#h+Tk-K<`c;JN|7jEP=tb)o>i%Qts{8<Y^m57oB!_U$M--a
zsH9U-Y(T9I_RyNk%P5A7YxRn@?__utGVwa@i6>xG=p1R9sqSyr9_jW6HKVKU3!>{5
z)sDkYCL_p9Dc!%`1&j_2(cq1^H0>mh<Xu9Z)=j&0XZB0nX&xz4)2HlTMh;FK!#rWc
z@AYc`dJm%?S?QIMOG(#oOF8OSc6p%d!Sa3m%;@27Xm|e;3)C1p;fqp>_^|_HG?B!Y
zEz$ud*I17QxdKpw*`$tt7~dr^We(j;>{`>Wvc~A&ny%9Ns!%OE5vdm|)M!m4z4R`l
ze<bfY#J%3)Z8FyP?yQb8>+*UDGd{(SZI{t8Fr15UBW%k32zL9k2FkZOz7>qZfSh;M
z%rz|KTF85)mQ_%#X)Z~1hoc&@0FCP|?ll9)IhQYpu&DBIq-DvIQ0Rgh`DKb7kMY^&
zdyBi%WD79i`IJp9+%YBN!W2F2cXefOQKH81W>T!1WyD1xc(F%UO^?NzB@Kc?Y&XNg
z^as(nFfY7_&dhf*lK7<$xsm#!$r?8)7LI8AafjwAhj7UZF2YkKG<d2K-^&bU&;&~y
z(N0!p+93r3gRO|C^w}`v=!e2dfC9BQ$M13`WfA`$AQ8E$dTvp_H{tHP&@<Ho1}2XP
z3C~OuKiXjBiet&FI)1rMP+Iu$ru88=nk}ky8sEzAonD!Gn>dV&DGL1O1&qWZ{=%rO
z8DhsobeCLV@4pu@*hLC*bJ@DQh(?L%ej@EmG8Kae%Jc4mP?7Q?F?qS%ofPxnrXLkm
zb)W`4W}VtJ0KS73U13&#u8^Y^%`}F|A(i9=J%<rD*ESt(fW!Y%gTG}b$t3d*g!Jl#
zA$foUGBqP>t)aMEDiGNwN`8>0q0B^Y`3G6KgD5zhKxdjJdfl5Hh`Ck=;i7*}HYwa%
zvVEKB1lPM@6{iIL648_{z#?0vL<&KPy;)vx;-Vx&qyNTenAf7&_!O(BwFAK>A!R!n
zX6uyp!&fK2&V8VTzb5ipF*(}m5q=ll?X#w>7gO}oq><Z7=mi;EwtwGs;{JhD+v@Ff
z>#^Cs-v~cCmp87EC$r-AfW8QTbqpU*cr321nes~ZDA8iWhYF2upWfRQ{;Ecjx4y<4
zmJz$H!d<cWS61X$RSxOve2HDDuGOI5<+O2lCOsP|FS_&v4V5jFA>O|YK{rc$BR)T5
zr9h^mAdS}}-IhM6PnFDw>yMl8J!Of_1A+6U?a{HMN-DLe%4B30_ZRAuQ>;$fc;hr9
ztnm-`+lBxI=exH_%k5+3yUwiPR$usK@aGr3Y1)@hx37L`?L6sW5NHV3VVo&1{3j>(
zm*ce9BUu!gqjg4!-%^RlAOSPj8F>t_PASuqE62DRAn^p0ojtn1>Suf*DUDU>r#!IQ
z`<iN@FpT}j;d3-i%uqTI<8U}jt{;ul?a_&yNc}N*a}F^0h~#mu5K|Qufqtkase-$c
zhP{9bXJc27pi(v;X=U&CFN%%~t#*`lhR^uOc=-79Qeq}d3_<Ffok`%35P}Sg>btZP
z=EX@8vM;iHg)|T*R_rOiy=~0xE@9xi`Z+lqdA-np(}PvFGF>5Qm~JN?xcP-p&h!J%
zKRNW<F1cHqzDBawu%ZUBUP8DEMmd~r#zzGOAsmN!+7}q2@ShDyl6+O{8d_R<LcTwi
z2{}oMP&Cf>V~GrmFsW79u?G>qU|Ol;f8w8gR--jYzeU5awU^DJ)UHN8J%&I>&R)VE
zhJ+L<cJJ>admh(cLmF86n{D0OZUySI)iUX;+<~<u)h9@YG9~uj6=QsSykDGlp>^!3
zJqN3#o4V;A3T4xcu-h0&aIJC{c_`W1<|f5EnW+Rm5q@dJC=f8@h)oiyFSLTILCd(8
zV7GW@dZ>+uejea%u2ZLN;Oy<*x3EB?sXs@csHue`6*a(4q9{9uQjW9C_hXKJU^9h!
zOwOgumh2yXxI7jC88O`Tt?_zIQ*N)D>UVC)dJ*e&vtrPNu%SWu`%jpIBetF>n|s0d
zoRB-P7~F@SC(GUI!@@ovB?9Rs`&KUJYdFyj4c}gzf1f0)8CdDqmS+iEQ@o?vRLB2Z
zoUTx+P{N6qhOcev#Pf}DKe)a+g?YLQ+{2qcb)x7aylTO)-D9e@DPnsFPG5Sb*iBX-
ztk(?V^88v$wI+u@;Y>Puvr<AFz3AomBuE@fYV1G-2sm+ahiyI&Ko*8Mh~hG0Ehj8G
zj!%#OQ`VUY0dD#qJB@BHIDbrS6w>001y!hmI_<e<44@^JUB*R`)H*FwTLQ-gE>GyE
zC}MJHq$}QYpQEZBuC=MB$UUZ3JgP?)qda5>_aIU7Zlx`bLai7V<K6GGUfp3UNz+PQ
zb49!-uPkM7q~uJwE%8f9NQI&M_k>Lj4y5absH<&RCA311v>2{%t~_`vkKvx7%YwQq
zwvKw0GCcQxWO!rK_ItA$^#?l=9^DbGeV?cpIA1+~A;O1dudN9KQdDz+99ZCng+lQa
zXiedggou-%?*U=iz-FYFh#4P=I`hXcMR5Z(^opX(rl~v;Zw1w=P~YSt2d>b*^oHwV
zfojmEUqT<Y2#m9A%;q{~!F%`q;Dvn?MQ`_^!%J9oaG5L@U%OA;W?=Qw(w3{7v=QC;
z(g)b?-~Nn20-MzV96xvFkbtI%))P&oa7J3M7vH*ecrgmH)oBe4UoU!J4ns8^WyJRB
z=kXbv{}n}Jnk7zxvJ+oQiuU%wRJv90t<!#n@fxku@%s2MChBp%)9~Dk`@rLIEHD=E
zxbu{gF|D#@D?!>DtMaUqzA_BtMB)k4n(i&gs0!Q=geI!h8`NTUI9(ryRm_Scm6hLD
z_B<lj;HAmJ&>DD$dIwSegv@}nEgI?9A8sAl36=b?#+s%yAY7>R237d0kSZeyRdZYB
zs6;--t`oz@%tZ0emv%wp>b>}w>!N(WD{KsMvo8;}m0JVPVA>hV#(~G1NZU5$2?v2=
zg;8+wao3<w^QhNl%ll7~V&0wJsLkYGUbhSm`z8MqN)@I;Ask-Y#_>s+FwPT$GT(v0
z(6gCE9F2U50bN?V$|>NgZU6Y|7iR-?oWUe#<A*<X`oI4U?RgnmZnWbqwcop9Yg_p{
z^%K>zZ;eNv-n|Ridf@@t`<@-0-Px3di|BRugA?M;q&E-#m4I{sR01GHDk<mfHjO14
z*)-}sJ!OHi8mLUBSshIegP&tRX5k-y!tf%)IICJSi~#BWVHfMY3mOD@dA_W`S*`@G
zG=IGB8PBy^G!rk#xC75Li*5hgh7_+a!h5a6>g&-=zo+Sff~D1u;F;!sl~kb0N~P`|
zE8Y3^4)&1(9)pIKmMJSXN;xdu62>Y_nX<ET#qkOmX>x(Zti(cesQ7%cZJ%aDLXGGX
z&s_UO_j{<1yI+pwR&zC<Ggs^onZf8SBp9;KZqpTPD<;K-BqLQSh+T8TZ_@d#+wr8Z
zmABA=B!h;|zl{#;T8#*<pH%(Dz5dqsB;<daJ>-I<=H%==#wzldNH>*R#{ja|U!R1Y
z-QKnx9WURl?JYX#rxME4?lszG{Jzg7mYDWkBcTGvw@n%QkS<0|_jr}u)Y``0G^ws|
za-6D@QPXMmyYcMyg(8rUtOj@7)lF_>E1zARZ%SA+0(1u?kPos^Qjyt1(1nrsPhE1E
zHWXV_hP!V6A)hfocV_Mx7(uNl7jns&rRe>*!w>A;+8RA0)&*O?CY6X3repJ^ZH!a<
z5VDA4exQ3<Dy<B;yi8g+`1-AsmEwNlsT^bN5%Z6x3gYj{42PT>ZbI-=-aAAZ(rP&F
zi42lYY+arSn-Ca|Jowa$m=>)kH)Ue?lafg9{Vn28&kRR}l1#A7G&VfOkLRCi{msWN
zH2V&Z&F@=}c!>Nd7IPi-qmy4%L4l&=#>OwdE9%NK7oz71sa7{h;TuUC4&p8Isy^VR
zi*y=cF>|kg+kLZg?^{^T41D*s#{pmr7s<tpOr9E!dD+9E*W)!(p3uW0vKD5x)qv@J
z0~-WsJxs-=c%!-n0szNc+`Yx>4H|V>*xE02YP0?|$SH33fQ{kA=CuFZ>NOygyo$|o
zjl`k2A9Of*c||560gvhad;on35R$~xDwK98&~av8MQIgdl)?3y*o5-@XkhyMHza1<
zwWg+quczIt$v_S7!!UnrsL<3{1Xa<sR7tUdlemf|E<V22^G)C@are#1x`hBC(8*b6
zn#7@Df`i@ro2&`_E*?IO;$L`O?anT9Qwx9?L-y^HfhA;}b->X;wf${zpjFGnvSNuK
zxOzv#>!5{Tr=EE^=O~_KKXl$}GYcIV6J5##*1=G`f@!>YV(#_@Ar!{*Y;AYznuf0O
z7kWC|Zw)R?hfoD#wB>&nrlur3N-9=&+UTo<jeXShH+Le5i3P-k>9Zy)MKuMXkIdeq
z2E{{1Zb1RA8|{};-)-|tcW&~!(cPrD=-y90d}PzqQLmDZ^?;ml6S5JZ088!7(@lWJ
z=HCm+O$Z@Hr~91m^AQEn@;iA!XrMtMw=}V&i8IB!zx;U@>pwIYCMCy*QV$mO4J%jv
z4WSQD`MOYlrlC071NmT!11gowl3~8L2Q$;y4_O!#kTOq-nRpw=YVGAOAMr8uTP&X&
z7a>X=F!^?8tb`I0Lhsy%IeGxGW1j!044J?9+C7f4rv3#$Cz-lRhB9f5CyKfq{X9pp
zNTS~Hp0sIdUs!$9Vg7rr-Pn5K?D!_>iAX#t{nn+Z0!n+&Jol;Bn<TRJWcy6`uI}7Q
zf7UA>?ypWQQ>NA58v>m{>Z^Wi3?<nj>|^DT<6{K^G|s;+;jSuRRQENbxV^8AM43RA
z6rP5;R3SLUJbVUER>kHWt(1+U{eP~ZF5UCvefFze+@}b^{M?#%OeG9uAT7OIVJO1-
z8)NrPbuIhS5}hgp4AE{w+M!ybdWW*>GP!jfwi`oL5#tIEV(d}!_^O78U*#>FF-=Pg
z(L-PR{o?{Y_PpC&;FrSb31^pQ+QBBpgBt?A{6LqyZ&R6ea2`1^9TTLs<tHiEq{?<i
zYC0YbVLWDpC#E5&YLbOd*U%XBAqp!pJsl_SR(sPb41SZ4pGt_8>by$KaPd3IHv$ZJ
z3TB3#-$_cFe+oqO)kFkKx*L_dL<?tdCk0Y8UwhT?7(YxhASw+}lAqlNsXoOZ$(f%s
zVu+hGiIIC>j+u7fWcBnVtUnQI>K6b;009yIS0uSqBjlq(C`piHc~~ELX-$mE;1W0>
zU^@<cFB5o4C*=K{Ub)LZJ`^%O^kG@YraE*2%S%ychdn-PF*pTDvAC^Wk19pHgQyfm
zO&}%7Y>5pg4{wKRAopr!aw<<05UpUgTNA_$AD^Cbeo{_UOEO22qwG41<XMz%tdVD7
zVQJOXb8Z5Z)V0&YP&)Z5K!b<!Aq}@&P1ab<I(oT4Q=FLtaIvtnD%IYfI0nYZBzy|H
zK<lRXSj8@~=`u}jUgk%B9*~IH2@}k57Uns>AQ&Rt-G1A_EQqivyvb{26yspS5(G+R
zxBlupbQH$4>2dY@@X36QdcHVu@iUI38@to%hC|-qkH*Ro%I1hXl#A#FMR}%R4LRCv
z%*KPTA(JTV|G<c-!xJTfkdfW3?A-<acI_FAS*(pK>!n0(b%Bs%r`0Ds`cO_^ws0Vg
zsi;0Kfe>rA4RIp$jeHoKXCK&K++jM{6u%kiP(xM5QS#meL{#;#U^_G{Vc)^y>>G3o
zN;ORQ6SxlVNo)ktsiT}%_6XjcpH~E`>_sOlRCo!%k|QYSdwh1%*wHcW#$nDzK%pA&
ziWyKn{<}R);y?Fs$<4*bnAOJ0<@;z1nCuZNOX3|_J?R|&OAw|Fj4_<bC+utg-I`Dd
zdYVu_oGN5?vP|VK$QbI};eq#WBS*LC@d1tTfG6lFys1x7I!xpSZR$J&)WVWY^YZi%
zBekpV)oVuzSag%|>HMxQM&{k$tEnt-1Mhcx_uy-^OI&mN`!B*9ti{w*SG^4ZB39ky
zO~<eHKKgLR7fcYoc;C}c1%XKQ`BZoLSutRjzVC}O)jtjLbdhuSPeEvHF~2!RBc^#N
zjMzwmg4}h2LM4kKyvnk)jrt|qgnWG_^^)?z^)yGCa+;6NW&J*ZBeH`#-orB&ACz<b
zqTSH|t0MEz_YW~9@il3SK7G&gMn_+9B#M#{OV5$+B!k|9Pm#!%iT?RC8M(XnJsx7I
zKFxDdqS3Oj6?I)8?h+tj9-eIYvY#Y`mtpcpXhKHTX|XP2H(yx>x(GwZ2!5ejNmv`^
zcTt*xl{!R?OhYCOP6T1NKq|<nvp*Tc7_iB1g))-X1<meqS_lkt)cqdeR<2`ygx8($
zEEVAw;U!p^&R5;3<GcppZjz^<ZKCvB!(ko4C85;U(C6kjxqGHvLs-=Hw_>Fj4#e_Z
z2ncPG+X>tk*u~j*hLm-uF_7M&N2t-$jg0s`$0XdJtUscSS2D&{*UvgN>7H+nkTo#j
zrIRlfYZO^e0w<%swLcx6f_&=3-P!u?+iT!cj|+5CKj5+t0cRn)K(}5O-^$c)sniJe
zn)shb)^vL!mxf987JiS{UjQnyG)d7^p6I^os#pwkbLkBXqg9?HIrJ)Rl;f`Pz|}bP
zKu;#>Pr_w%c21)6Q%J54@c)_AU+Xfc{hkf;;{#q_ftumCN>te3M>V9+p&qEqx<pub
z_+Qb`PSvWAU!O%^X#nrAefL9LZ9@mI1h2X5AEvrtx33BD`7sDFy!pOH7{6{89|e5i
zJej}J3aJxm8O&cl(H2=@9rJ=4U@`9_t~fIZq6UKzMFhTow;Ip@J5P&O^vZsS*r@|)
zERu>3=ECpA-d7d)139^gql3`EOeI28e=awDUYsK=C`sr{Is!V`<$fG5M*mNm=*pNh
zz%5O1f&%gp6*8>`^Q~O+Y`(&-_wdkLr8Sg^s>AWD5MkS(+!fmCk9#;ihattjTg05$
zoVA}fi7UxD)y;~E5B$|^eRxx8nI&hZ$jZys=bzV)cl{mwDYZT%k_Uc`YrkYzcZ8gb
z?NeNaq9xExHgt51xN(^Bk^H2wmKpdnu#z(&vTiggXU+SOz4`Z)6q%_X<+;N@2NA8n
zF^AT*RwtDh2n>q=`OKv98RHtCJbWmH``T~&wagQ-KP%-idZ@m7ZobEjIMs**CR(N~
z0*{{yk|2!^6RxaQg==}A{@AWLc?1m4%wW>78W>5?6fXD-U2R)3rMgzI1@am<1CjQy
zl-dm4k>ziVnUU`(Zr5*e1)n?O=wjX#<LVR)MexR(k7L}|$4Gcpcc?BMya~QyXjgPc
zx~+|96m*&o^8;Sck7HJrv1$PM>{!7vc0@fJyGrYWNI^#d)89l|spenea3Rz?g&nP(
z8vwur9S1wyJ+-7<4uKHglCBDBx$h#j&EVz@;}b29jiKWNy$mMc5sA2}97_KIjKjHm
z3UrxND4d*pnIxZ{F>E;C-`eBGkTDpZg%z@~`$OBAq~an_?mR~=O=oA`6=O|E$33!w
z44Bu(@U;`|$Fh5?q^hMbva>9)Jq!YS*XCrZSXgJEF)I1gc1mK1VrNKmt+LGyWavp=
z^W!Qhi&<t8L1CdT-8;_@>;vGlLNc7z@{7#K6zv0}O6_a=SFKipg<b46c{)IsdFFrS
z`#Gadx2CKzv!n*&Bk)h=1sD(7t#!Hb%<1pWNL`$r<<?TD%Z>OHyw1->-Ub5S)~~+z
z+%Ui1^t`_P1}^sp$_dSCYa7&29&xsF?EqP(HCaCNQ<3L$q_dtk3Lw_v_vD9)na@j!
zWno;3b+o8WXIwg&6m%aidf&H>+5PpaHwthc8XEdgxj^H75+hD09^~G=@0NLg4Ic_K
zFcZ1y6)P`<gE<Rer(*5&&ibt19_VDAhYVj5YyGe8fScr1|B`FBRP=?gw6xR@y3#XN
z2qrU$m6nf6f;W9C)(}9fug5DO-?o5R$(!pA+=;6e-Vn;_?BZV;_%k&TmT;peDaYy>
z&_nhe|EfnW%88@y2Zc^^9Kxyn+Oa<wreD@G30j0i(qQ9<h=@C&0?@{uC{^Rgooo$!
zgV9L^guEj*0dn>KDe<G50U)>eL2siV58q0p!QM+8`@XPXO=am}&V%AnLd%IIm@Mtp
zm)HX0q)1T{6?EI!gT`ss`3)i6B+y4ePpC*Y8TkZqnlQrcBqIwpHifl`ao_^*DDI&+
z^m7McYhC)c=xD93(8SQt_kifyI;rIpksS*pcrobUBR$RO(QbY$3HMq6uI**pCYij}
z$AFW!D-o`@m%J_s+46aFRF!&O6Mb%S##Q4WQ3R|@AFbTBVZKm?a@a@FmlZ|!)^ILF
zkFx1!u|}leLEEp@Pu6(5rxCnfr@KMpk;?s+MJaj{(C24M(e2~}Q7JI$!DrxC%62kU
zu5h6A5{{MqbO{3C03=bKBkP|ZhPu8w&JcXz6GJt69q%J)eWew<$EY(M#tg~QTgakb
z;fe2s?)N*lk}jz?^c*$3bltP$y(vg@%6ElTW*{?ZIN4ct&V1U6cAL=f=@Fd%&qLPi
zuTPr=8^IO|&e2E1&YDY$<wS4FMfabwKwSX{hVfaI0Hpmb4a;2EjnCLp8#(_f6ngd+
znG97f`}XygY|S*_HRXcZvV#2I-E@E2ha2}@YlPXh#4itr(F+eBb{BB@|KQe{^b24P
z>)r2k21JIeSP6QC#0A?a$08zXh7h3=LOKSHAtFn5VWS*Q3~-jKE{LnM284_LHE-dr
zIrK(HU)0t5+318<cQ>kKn5L%qnMH=H8O+`>Ql|4#zO33%{EUZK75MLHf0X+eg@t)^
zubb~)u9Ks$mTS26_hZ#Xo$SZ#L%Nn&e;N|%tXT!w@XIz?3~?1+ZeX_Qb=GxvUnqJ{
zwTJ(8Dq8#9_|OACz+c{@x!37@*rQTXR@Pyv#h7I$+4DO&Iy+Z>9TVL{f7?fYm=k^B
z1?F&tOkeZbLux)1qdqVQpf?z-|Gr^N6mQT@eXb0=In&0+eOv2!uFY6o86uL={}wAP
z0St#`G9{@;6=p|&n6XOGAcbC}zf^pri34Tl`6i$HiCA*yscYc^U+9>dc&53+u&GAo
z?~t}W0aU9Z*Uyxkd$5hW=~MrgSX><RKo1%%pF;!B(Y?IhV5aiOID0a${wo5`y02Im
zzB+3J9mwULeCYO0bRetaZL&o>?K<<hvg6Y>tK#x~YHALaM@Gp;MHw!^=ps~je9}3}
znYK5Tjk9$v4XEFH$_vv|wa%mnGvy)z%(A_&kYFvG_v%SI?*WuGA&N`y#XrXvT2`g_
z`wMhrV9+1uF|Sx<d8n8}3CHG@(Y1>WZMTg3yJp{eO8dogu5}piUja*zgBA4f$I+XB
zZs%*980O2vzA3RB+j|cV<joJAF*+fnts60B;=2Glhk=mo)70XM=+5cpdo=V@7t6QA
zF|#yD^Pp=yQsXE~+R3Jf$f<MlGIz0ZEw1L?k87z1M{3^|@#dS}agTNfoJ{cLFe8Hk
zX(HXDrZSI62#D<v#7^ad-Ll-7!-J^55+uk3#LAQLnM6IYTQK*Ah7<7kD*i++RyBbS
zMKxaq$V~4#!Q;QKUs-BimU(QM-Bm9_&w8e|mE*OU<TZGD7k`W>i|)=TiyF7-7TA))
zzU#1+1$=(&S&A5ZfI(ODVc+ekvhMH77OR6`dww+TshRT9V(TCX5`(f)l5Lwu-5PEe
zhYCWl<_qrcV0+-y4il*F!j|_5%Gx6=1m3+FJ`^iVHbc?#VE{Os-4Dd;{knii2}i2W
zO#n?^KCU;&zEf&g_U|v4?Gq0M&EwOl`@6S`^~-}fI<83$A~a#=eBxAmEw1s9v$uZ<
za`t_jOuN<kQZG9ZI1W<Zy5t_u^)V0-v9PcV`~Gr^l-D_PRnk;8s#jEgwNp(Lh~TPe
z%$15!<>HN?u^bRVA87rVp8mrjwn&1mm;A!}=yc8Uk8cFix>$KW+4rBTHEZj$0VlC*
z&3n};m;6n!Fcfv=1%&1bpFUkaZZXbVy6)}P-?2v#E=hIb>7Q?YLl*dnpmPsRl_d6#
zy}i73H2+|ss+w(}VmRr1M_k3Wm>4mk-!Z264))BJI=uGekVI4UKlhNc-*xtd)r3ux
zG#piXWBkMJFXhN8YU_&Dw+#__zSJ0gxx>YkEZ|fIga%#%D893RyN>7k5z##vQJ=g9
zrt-g;<Kv7htoSL#w5s(&zDH{fGZOR{d6-DcM1OFz4QlQKUn>JgBFF;%zU{tUv)<>d
zKcLA<=@2VNK~4YQrkKkm`Cv3Aqb6psc9k@hRM?!Zb|io@hfMmG#@O)z&@i>JKu(ec
z{dDo$IjyOv_N+WsMnThUx2TdT4!5PH7T0j8(!+&vz~I3Wec<KDTcjr>AfmwTe!S;-
zy|q>Q9QbZ|3hNK270sdtOPrz^3(%)U9IGM2pfR>|_WCnL)$EHW-Y)cDdl%4Cz-;km
zE83~v)8Ufu=2u^S2?wMoCdc);EOCMC8~a?TnHw7NK*Gxv@5&}bT*-G>w|8fk#K64(
z7@(B{E~#v`%ZN^50+9V0tmwow_>8CAUYt&zYW!=_Pq&Wjw-<P4Abg^r2Z<8iANG<C
zM&$wey)|cK3j$Gh+IqcN-z4Bxe@q_yE)FPICoJGr=I>0%0O6UQ9|Wik*;yuZy>uL(
zk#c06uoFEAq#HzsOU#OOCf)^MA|BK!nc68XRLwMBNB?rNe<UXxT9ukz2ML{@O;oY4
zxs$dc`&JcyV*19*SLkf~Y-?Jlmbmwowqs@cP(??)06RsoY+U6phT{iJmLfNd%|QP)
zCO<<v#0ncl!vfK3kpXeIw`)qQZJMWJIVP<|lTkhyw2Fqk>z$NR<piBtMsgZCj$D$!
zzYjzF<2SRG?9(;Ep+)EAKQ4t#*+vO>=wpV602bvO>BsGuyy6yJTp!MZCo(<0txv7t
zV%;7eh}U%X$H4bX@y@5~oQEHHZdH9~2sZtlQS<j&ampltHxDfM0k%^CM~ZA)V_;gX
z-vOiFi)~MzS*L_PHy^Xy9E>&X=fAj3<tqJKhtcg7R?*QGn%gKvQIv8tF$**@Q+aA&
z-0g#uF+0grhD)vwP_Nq^{z~({7wi5$uijC-{VQdsWH|mYj3Ls785Cw0R{Y{?ev`Of
zZusw2<@jlf4PS`j`tz$ildqy%3=1K;TWP8F2GAK`hEjRo7SH7}`4lcl&sI)Jh)OU?
ztKi`wcZI}3NUlMoR#iTwhrrvUbye?55$ab*+9JZ<l_f29R3%kn%@|bpg)L~DGwC2K
z2heOh&r(GXg>3l+4?bPXF3M7OcC~JO#14+XdF4kd@7!*`8;aou<H)~{Gxm0o7dBK$
zv&FJ=Dm;(S|L&=BAA4MFL4=NVK;;vLF|R3=Bpa`sdy$MD5IM+BnBYfE1Y60xM1GNn
z4Qb;fvk>xgN+9cgqpD(GZEG0^zU7vOP|^GM-7nwQDZU~+pCgEF9lFG;MGu)M=^v5v
zb*h{U=OeHMNzUMLH-oe`>v}8fWE)`CpZXAld>_--Und<ei&?vd1|rbCulKn<JUj^#
zBB&qAP^o2q!Dgs}aS_F^lZtgMGfq>EPf6>t4I5ZM2HMh~exIB*oDXXERzmcyHq~rO
z1<v!sPEDV^1Ki{%g3WR^cXrv|&O~3pes`V$2|ahdfq7y)J1=*lMk0n6zUwcFx3_eX
zWDGsITdMf$9l6^ztfP-8Yl@gP7aazLr`8?fe|}fppdV$q`jDHiAdGW58T3US^%zzc
zU6Eb-PcN(R!kT^ntp2IqJtQ#~yVgGM8qpb^rm)c+48<gk74Q|ALNofTlv#`05K!w|
zJ$VQ&&S$bZ1$(wk0)}SBIREOZr{KoTP!~bgS+n28IqWxH(H4Ohd4?bs0c%FQp{x6$
zoOxC{cCq&O2Xs6^`p<fz>04>`q-96h`%;dicGn~AETa`t1S=bBJ!HFScl_5=A48bP
zl1jW~#r-g_FqFrxjESpoM@GYL=wbJrCn<}D7Bk!F=__#KZn+D-zl*j7JLDxEkWD*J
zK0lsKOic&3+Yf>HkJQEHaocOy-A`=MwQ!Z>FE6rvS4?B_z`)YO?R_z;DfC76>34i@
zSmbT>WCa#X&c1tjAkX!q!&X4ijbm8+!eNRj_^HBhe5ie6snK+Dl9he=Au<4gj&5WE
zUrR$X97HSmE~Z&fuA)<lx+K-6Q<Q?(auG^4n}<5;qjkXClU37&ipFbZvU5OLxpdl6
zXR!v4*TwS}ug4$-Ow<?NU>kNQTYh~iP@aIO)JPI+B9!?bR_K0NtEbNv{gXFgD3u&X
z0~fmcWEbJQTxgAaML`h*gIjSQ!)!46zM%I-HjuhoHG_iio)&v5Nghf1yFqSl62s7>
z*&u{E!X-*y(9HddnjytFGW@{PJY^vs7I6y7!56Dg=C2jIu;6)FOV#P|o(=!E@4Mtg
zTy}_GVg@zR@u1v09x6HKBK|CzX4@2nV_}<SA@T@_zbJ}WOKFSk?ym6dYeyID>)|f`
zRijJg6@FurK!$x@hok(eDu`WDyJmiUNpsZDukiy;PMtR^VdkW_n)clR6+-phQ4%6q
z@KS}d2<V&DS5Cv33H-&9+d?vr7Eif-g$Dop)lb=c7417+cL5V=fPPicw5;-jC%UJO
z-Y1n*#*%IQsv9?_H5~i2I}eMA+maM>G4-_L)3}GzGyjP7r>=D<fbIM1)-a$Pc<Wk)
zivmvt;I*3=*nu#Z%ZT)RT(=iy+Hh9mq)3bzM#Z`>Udbl*OywHc*#?)TZ&!ija)9F0
zSb&ssC8MFEWqi_Z*o(tLD8~sXNOJ@|7uVNK1#*6t(94OylK&ftrB(kiz^<jG#ibYu
zV5CtL*QV#Wf3t(~t0f}g#FS>Z^qBF4{9j|q-ma<Q4Gm!OdFVyNwoSO9tgmy1b=oZV
z5rNm}OBpAZV{~Noy87pL0dJmO_sbHqII$NN29Ic~k+!zUkjEz`qZZ~b=BxS4g4?zw
zk7_QUw6rCc`5yhJ)8`iltm)hXq{(BDNfnh`mk!G~f-EkBBcB7#<O<j(qy3&qZ7aLD
zf+4en;;U;|2rYG;CWAR!Hd~p^k3ytE=e|N2LyI<pIqmU^vNm1jHg!!iogW|Nx~XLM
z6Y^(Dgvlwj7cB%^##`()o?GE#d9#|(hznw$dTRpn3=vGhmQaVo7+d};h>)Zn{^CCo
z@+OMVAjGQj<Vl1Q0+~!|fJ69W014sAv(48jVrdfaUn~lyJJN*Qj`a(>GGK5uAvh0n
zbCv8I9!B5z?Bf2RM%}guO(?qguvLuW4~w7P*~~@y@&#w<`}WMb;ukikfTdMtv`tfy
zcC|$JKOq0~1B+bxY1c-MquVwu1|sxGzVxp#Rz%VD&-T>R>?@;UGHb$avR<8(VM_8Q
zo;i6|Cn^TAm6f8r%faw7J}^R%NQ<E+D7#Atz*)QoReT;rpbEob+H=hn_8m59`|Z#<
z<!r{atvfcGy<`{tPKk9gv0Rd%#@ywhz+S?;W0&*9$9;zS81}r!Uo?ETb><CkPk{Im
ztmBYuSMP1O|L5jCXMbrMQ9OL?utdYece{dhHpK4d-Ij+*&wz;a*IUPATKe%_qP6?k
z^iaQv1(PTTqF4|gS7jyck(5|{O>b~gl8~}9ALQwEUA4ylRO6R?N#V~s^Ffm)<L>mU
z+wbUJ{;cM7U?vuWtxxf?93$lR1UQyvh3e6(Aw-RT=n=!&@Km|E#%Yd*Xr#5O%iUd%
zeI@7+Jr*O^n|FlQh*IUlth9rPPzX+r*_T!<**;Mv!If(qvA$#^2Yv<k8eoClfMNZD
zAl$CNr}*%23;%K^I_zU|Q?Ik2(Av(;@cub$m+hmq8?k(<kFC}JAahCE6lCvtB(--y
zI@uaV%>P{eZC%IObXiz&A$7Y&!b!Ramo9od0Vt7&KHtERw4Ue0&=>tG{vF@IMl%7@
z^Q|GwC&Lbj0yDM<@4(Iu&r>!6Grxzgj+fxKhqKYE*t6GDPoQ^xmVo#jF#;u@6JQ!&
ze+8xhaRWqsHkttbhHwWvb693``F!x@S24AjT9WK8Y2GhH-IN=8li1>UZd%o%+(JoP
z8FL~$p#CXM<GUjWqf*pN7Yvo)808#JX73OLYDqF+z1v#|HiW*uNDL3eQWGYQO{B3k
zc9&Gnpn_zHs>@<c+nCxqJ2whP5*~&|$Xo-w$J!%<belIO_+qe<O6VUt&fZYd(xA`d
zfv>ED=p!Vgq!*TfxA@O}WY59?+y5U~RW-lWceWdlv~s%rz3k1aM_d$)o`MIHK!kJP
zB40r_`R_QJW%$S!kYT;QMfrM~6a0)-oSLbHps#|3V37{qA0cq^r@RFdHSKz5Wm`du
zaiaH`JZq6^NY<I`Stn|mzYx%t&ucg};`Vm#(}6AouaYLz_;mahE>nhFLU>jr-KKoP
zHpo6}>6p{UfA6&9BbBNum&F{;im<6!Z@|r7PUpgphi6(R%W5ekTr;8*NykUO@P+`@
zV$VO{&?w*rHWAC=V@Cv!Pfpo06bQ{in~CHDyZE5My4ytm&xDz}rj`1`m&%fjp#vmc
zt?r$T&+!SY?0Z$VSLKAJytfLrlg;q<#3>i@lybaz@EDs<koB2OB1N5VDZsfm2+xlC
z!6hej)w2MXizH_zj#aQ1hgPB<jII#t>uB29;m+Kg(KR}KBimx%BLEdzw6+zRL+~c%
zg(Z<uPPpG6NbtJY`XEkCU2dXT_3*$N@Yklxq{%>CQ&V<`LU&vx*Ru0s?+E9J+gvlr
zlD?!~W8P`*C4lrK)5@s}d~~dTM-Tp+3qf#b--xT$EK0GH(fFuRzBRc?At;#3#HP#H
zlU>+Eq38m6xYT-Zd_4-B{O;%<bzm29b>#{?=1Cw6TNli<mo~2%%=V)|w^6sI9F%*H
zsZ`&{DRq=N*QlN={Di|OoWVh`$22*qg9nJ9-qR*y8onJKJzsbRhU#nPe`Do;df-I6
zwN#L%&nO$KP+wh8mMy&_h(9%2KwDNUbC+w>trrn2HDJbQrcgX5_m63hWpQgDC2l5I
zu6{^D%cm~F5|mx!?;JQj)&Jr9lumsU`x|Fe(xYLchuRc8g%Y^pue)cqeg2VH`CiCD
zP1eRwT%?WD;_Y9={87lEO_Z=<LBSrTQ*009#<vNbxQXisqg-;i#;X(cy31+f&)(s4
z$gbaHUCpz92w7}XGjx<u;9o}zNRtwyKdCetK<OC1-9$WZL<Ej>I0l9eU55B2{#KOG
zv@UX)uuzA1UBmw};2?^xE7?4Ic|E%v34BKPemZF(RQ5<$u+TFAaDm!zc|8Ay9&qI?
zG&Hd%sH<~{+OZpeK*@~=1HL~A@mJl9apCe{ay7{$ilTn)xy&{aJnX;84}Kz(w$+oW
zDVYn8A!)2E^|w@i<qL~U5*dN*spskH=1qjB_eq~W+U*?xpwDL2uKkv+Z*Y0hXqqbm
zFvOVXkKjX#*?IUf7-BC<inQhbkp&ri3Anz9c<y-}c|&-+SAJzr_|)Aup2qrs{<b-F
z{}lLk=M7zBontSbLHGyLp!m}no7gieZ{_xBISRFzp8+X{hXibZpb<0$hv%;{XTV8f
zQBNcH*Q{Z{tyv-!UF_o@dncIdTNsI+o~*m5oTH3H(rURG9|nAMSk3RB9Jw-_Fy3W0
z#V>PpzD#x$=}6b>Iy-v<M)-&Kg6Tz_4+{=L-_3q3vme47qOx^-SZ>B{mR=u5m7eCW
zLKbQ2IvVVa%JRDl$Y@>zXX&QM%#f^DSU$oY8+fKiW9%d)S-uC{6V{n6SD?UGM1%1L
zvo<}Cci?moN(W0CQz~)_7R7#rC?qVJS^pQB?czNUbALq8F}Z42)feVXiUKc!$nvZ&
zA0@Be;1{Wa^kAhRFq$z5;zc3BSd=F8%UH${6>hw9*J0%`c!3~=w2fp+m((zV|F~1&
z|EjMZzuj~1`phpNuz9~j1L`+QF^&af5H23KN$vFFH<Z{ZEHarTDDdws8g<H!sxe_p
zlIdBMr+lEB?sylCy6Y#fJ}-N>ob0h*JkjoCRIgXR=-pYfhWg&cLJkYAppbr!lbaW!
zdhG+kDJJENe<ce46(T>U^sV9Ki-B9~3L)@Zrt@YvjUa(`UQ<yu#j08k%W+trrPk_@
z$e0-`n1y?xeRv)oD5*NF@1CE-ET=?@D?XYrEUo0!cW|1v>8fZ(yJ-H7f#sEJ^}3)R
zC^xYKo0S_`a+svkr`QF3P&8+sPxpx;TBT9FvsQ?50Hq9(tW^D+46Be0)4eP(Pyg;1
zTUWC6+V<mU3!0MfRwrK|O=W=)f#T^Q3on6^^-(n?{7s6mBOReUNB>RxJMrk;bv+R<
zo=!z&N3zCnVW1$4lS8BSTrblW3ha9@iJOa?GRGf{=Hd8*j)TeUYgE(=N5qFoVbmY&
z_K|S01}=&VPuOsF0@~W}g&+<a>Dyo*vI*^V+uY%EZ0GH=1jOe`F+L&B_3@IB?Dd*X
zPd|4%dviAiJ0+zrdy{+F_wJNRLt?2Y+c4Yd?3JbGs$+e#32b1>MZ!MnYJzE$ZOWbp
zC<$BndQ2Px)2<r7vI;rpw+jCk)tRN4g7X<!DI#R<+n?RM0{R|yJQ`+7zB74>-uq`E
zj=C7Wuj&Zvold=9=_$;r?=ss<*R+a_H9;g@ts4lUkNi5BiKXu~&aHvw)Ow(OCcnfk
z#Ca0g<bYSolu6!v!_L=;j~pqOUQITaR1~WGc9M6g{Cb~v0V-sz*RHw0db>{}u)SXf
zLr}%up^D2o{&i-+DLh5G#mf$F0yj9V*&=WC=TlNe;;lLSnr(V@5#LY`2+NKcPQL}b
zOPImduIx(=m4VmVlIKU%tF~Ng#zHwiEnc1OknAsDi_&<^SI(-AY>)a5(UmaKlqkN7
zQ5E;n{IOul>s=It#!H(9o?C~46lKjA0jpvGlWpoa?m*wWOPBy;!Kqw6$7RlDM%I!L
zeohgY@2;*f3x~HZo`^S<!Pmk1VN#SIH4(eJyK#QzJsG0g;y*Rozn+rCGGFZ<|1ep=
z-<uI?_WpcJs2~R<2X1^%shDWg`o6&pt8J)LFidm=<;f-m>13+7s6=yBsV{kyI+oy3
z)==TAm0llyy2Cv@-v{Yd{H0TcXT=9d)8P+_vni|mDe|K^${5+eM0ji$6f`W`3&{tg
zDX4V&ew=6mou#+J{J{h#uD=z$%%c7i`-3N$xD6}l<Q*_Hj&G@_rx(6+%wM^X^sp7Z
zV579A(fzDBy8XGBd>p8&FnxUmx!U2yM)3u%1H3M(a9GUBAGY>uR<^s&ZsgjW4n!ht
zmF2*$Q%0_eGP9D`!*1KH$g&U8=JX)~$V!-Ig=t-KNr3C(l8H`}el&J%Eqlo<;y)ya
zqyJG0uv4lZR;lp^g-i27(kA4iZ?H;33oJ<7RSPTOOG)ML^(v*>_5i(YaxxW0ykBBN
z3-9(RARom4NRN^RwM>cO&){KFRwv?@Mgz8^>G3~M)WXG0!<EJD{=$P(%9hCy__<Ky
z3{>#aT1&q0kshAGuPei_qpHC9IsbTM!Fe?m-c<@hrq#ZT^|SoDvk_z`BxXh^JxC;<
zMcrH3Y)Hufg2^gLR=R?%6UobV;wu$?kJB%u8^;iYF8hUHcn;=LY2c;hQ@wy-v7fV&
z4M%pAi0kVUPY`*8?OYZ)oWQ)jLXm`#Y<Aybn8uE*G3I1q8{Btpr}`2f=jdERqq|$D
z1A-V<HkM*!qX-p*B&Ut1eZF62DVqNd#$62py+)t6l2ipYmXa;`T>fyQ{CNLO5e89u
zoBn~4ju_S3*=KC!tg~<O6fXO!Z9AwCs;pfLS`}UetP25?<H^`XPW#JG1V!?Rw6mr4
zvN+;iGEaOU#3+R?#4^<qU4r?5OxJrMZi;nBZ#yjpgD?O2nLvU%Gb~R*l8u!RyJ_$<
zBDFXz0i9J9Z=QfHJwpt0YHRLy(|sLnu6&{|jH6Ton(jvooteF~{4hF6=Ud0wZwG-7
zp3gCXk5k?!n@1^{LwRoxhQ?LfEdf_Yh(t*mKDSe}t<A&b2a`c~^WGLybRzLpvey^M
zeDq!PbHmq*{v|Yf%ZG-Ja}%H8WEZYzT>CJ)La8@hbLkt=+}NU!0B#=ID8e=m;1#{s
zlngRPqnrlr{sGR#)ZN5}EY3659Y);6#DEcFt61sCK`7hYa0ex#T#ZXoec1bz-Kj>A
zvi<908m%iDarxU@54lEap_cx(-qtW<Tn@>xB`@s}|D6#^*fSN$uoJtFEe{hnOYgI7
zsV=LUFplPLZMI$;m^r?^P~Q4~YjA7_OFy}xcePy{kXdCkcZXoq8dudK*~cM!)40ys
z$tD%Gz@`F>nsE-N#IQeKzj-~WEEJhp%O_C_t2ZDp&XCwX0x(%8J7u-EWQ%T&_@q$)
zn}mc_QD6q7B#Ibr%<&-Ym?<Pdi@ebgFZo`LNz9TGQHQgpg04Qkhu8sna^0gi*|o_g
zi^hkxgzeA$!9;u1A(qgqRQQQKQILTrwuHkpkQz-sWiyaWr_3iPT&#yrZ9VLicy+y0
zn%&7)z>X@V1Iw~w(zp{EJLWbxc^bR+Li!(75)hVf5TJC%d_knX_Fr9`G%U0KOpx`A
zbiwB#%u`#aLWy@eUjOP|I{-lg-^j%gKf?7TOc;lWm4r38G-%eCWGEb98Tg*#{pVwB
z>3QaH?0&_Q(y3?p2zaQkV)&cN*`!q%%f?}Y2W%vCQkeD_-ON=!Yov0J#yw>9JWUA&
zJkmq)EfqY5kI$6be9wX+&`({@&Vwqv;__ba{c~AZ<iNO89cGBJjj;!7^c~xn0~!=z
z#i=6i=z+=hw{-J3LcUC0RZYvPJ-Wz}F9M-h7*yC)t;zkNp9!bnp^LKL){<0nlFU<j
z&FK&9CLjY53a4w~enEIh!y&PtN@AQ%)@1-`Fm%*$+k42f@~<TSWC(L9R;6aZ8)1uD
zj&Raq)%%1ZjgD>timb88TleQv&?t5BSuoMH@0X^C64Wf2kAx(o5<UH5#Tw#mEUiA?
ze_w}awx77I58uD<%nO6dcM|swqB0?9c^h--Gix5N&{o6Yj_@Qq4}2x-9(#KQ2aXR7
zWkRQN8#)BHx2}`8__8XgDTw2j*4H<WfRrAf$rVZVOKmi$ZEJ1)&BX9YsC8;=45@M<
z^S~|EfyiPGQ!X<MQD&-WaA1gxgaVen$lOUff~Y>r)#$xG8Vj{?r$ObkEzxZz`F_@q
zs(QYBE9Kq9p2*%&_ZUJE36|AmN;sERw$mTum9DR)fp?Dgj&DPOBRMbciPp}cfv-@r
zjGe4?KZI-~3%V%Ojjw-ul{V22(%Xxw8L$-4<-^;gj6R4hMWWZW1qh}NH6e_Z@R0uD
z$G?lRdZE(OPD4zL8srySm}mCF5a%F?lhXikB<f=`zsK&|Q51uPbQUj5=+l#(kkXb3
zlf%KTaV;`QI<2S>p@?OcD3s1QiM{1oQg0p~1CFh~balD0k(s(jM|(VqUR}W}MtDyp
zm!pS-^<$H38egwU+7@r;p4{)3hWs2i)?ghj-D7~`_4I;wR|F)CQsCu7>{PylA5T<B
zKkW>H1ZKr1$%#1>5cufWK@QX&33u{IVPj@nFPCBDnB7<kri`;qmbD|>^lJCLNEH5V
z0s5L;?r{!W(zdY&=@ndK1SkkQp&OzZh_gZz#6_Qrr*&PMTDO*FIF*!?zVteAQIqWe
zfy@nD2=vy6nI+bB`PmuPHfoI~Exa}EbLR)i%51}b%;y151;98taFb8^ybnD(oboiA
z7gVhNg}WAWmwz;mvfT5*{Q{nPZf$^Iz^{7?N)|9^QZGIl#zm&4@xc;Wnw{Haq&bP%
z&nPc03xxPYc%C^yo$|QkT3cHO`4^Ygq>zS^M-u95w7yNhsF_^c;_cc^jc}$4y$RtX
zVj>O_v#z`!&s*5aqk^%`DQ05$R>|#NPG}|p>^bk1ABMGh#$I1xLj6Y9U)Fs69uMe7
zC4Zv3MBDtc;~CTmPw5V5n8_US>wC6QVmJIrFl59<u80+IJBNNhlMwKwnz2D>_jS!5
z@;rMal5uA?IO+BmLG&53Mw^oZ*RU<?{3ZhX^wcso0*R6^Nm+W7X=tp@ph!}?u4b6q
z62~1?K&*y!CI(tGZzf-mXe7js74iKY$3=G=CYp?&+uS}sJ)3;e*#yB_*Eqzs_UGB@
zEMn9gp1KXZfHuiC>Z0zP;qNZp_T;kQJ1)lBa>LST(Q#<;KTfE!F%b!mYLTEi8pLOC
z*RO9JJwes0S)L%msrk;T;j8lweFgbG451p0_7sXd9D~{SwAoUIHXrZd%j4|On@qFL
zZBpD~OIIP&v@d{OO!V1*%e0TT=f?gOD*8g!#wS|*nGr21te^^*X@vDWR`%@Vy(R}v
zZY$>+#Q`obpo{*Qu`_ez=yZ23(wL(U34oH4be~_K2LeBr(6qDw@4+6>MV+ql8@1_q
z_f8A$n~hp-b31`;!r;#sX~L;hvFuF=S!Obv%@i0h=A85l8EUwPyg(C;!Z70B4WoLa
zbNP!!OIPqajD`Jv;-OF7ESIXQin};_|LuA4eESi2JM~I@`iW$Ap8M?rP*i&7UQDhS
z(#o01!Y@hbdn^s`zUV%dL|D}wDENzQ%#d!Z94(p(p85V^{=6MbXej(dIFU+$Kq+gp
z_AX?`3hSA)D{3vP(j<G`5)BH!xc@~B0WmC5w%DoWy~r!F*c{I=UF1oyq-B?!JqO|T
zsxj+B?UXg2p`Nq1cX(5((6~RF+;1D<x<h#lckw{qoJf{n@3XFekkzbia9+%7ZY%2k
zYlD3<^Xm1&s*?BYtK!z!RW)?Kv7DV-9_}~0%ntdTLM8|>>~>}twnBt%%^ZNRSN0Z2
zb-&8d#vKhw9NuXAFruQWCID~Y`8OPJThM}xgDLi)Di2Tic_wb}O?f5Ib1W(QAZKH0
zLN7di-utGDM)lpt0+>J4-}AGW%l4eoc3jZ!)Q^|W8X_tuH;=P*+&v;XIJ5I@Jb0sM
z+lebae#uukJe<$SA1KZ^A+tEe%-}_iLpJJE0(J3na`dmQt{!{m-FfHnDypZJbH+Dm
z@5S&F55XhSoz|pg9@tHq2^foH`4T5ad)#54cD56NnZ7<iI@a|w<i3MVpFq{$sZR#w
zxWjx!TG7HDlDidjbqq0)wO*LlrY`zqP{o;UzJePOV8wo>W#<6u5`~#%3^)pB5@#Qu
zQ9t&=&#rEo-q>DgF2`<yf0o(hk^u8cVGL%dSOR$>j`yTV+qoE>t-p0(d?CD(?eS`{
zXP3o<L5f|K!SXT8tD+4@5yYBV4_PVmPJ1O!uk#pXcAPTeT~gV<y<t504+QqWoINyD
zBH(`DE>V7l_?p1s_P{+yC%{_%Qi#g1Xo8LtZ;roTyWn%*Bcz@z@EOxPmm-A*h2}&_
zymq~K#dhwL+D(A8jKRnGi)ZD?5wG~E9rd+9G#lFQg`qGQ-TT`ZylE57nNQ*hedM3n
zPeRtj6!6*#*saC9meQ1kRCu$_6WYic^Ljl#2HOm|HS_MDoldV2P!=xgkgFbj3r{w9
zsHMGm!W#JsE7;>)?Pvp@8O*!wXhm{}lR}ER>}Cz0iTpz0TeG$ZF&VowllEZ03;N4*
zO<jPO(}A#54>xJHc)~tgXsP&bOi(!5<7XDTf|3$}sHn+XU;5-Q-d`;rxx8q%mwwoh
z5TsqaFHC@xj0d>!U(eCICxHpsTPo9d)OPxq*+j>gUxeu6K+jR7FO#{BarKw-$ZzBI
z>?<pRZUXuv%VxMB`CL)aosEqkK~D$Qr(4n_>dnFgk*)Ld-n)RR+BvBhbDc*>BHrc+
zLYVt69{CTbb(QG$dQ+<Pdok|%r`_-#YI@D+ZG-XeboCs}dj1D?L5aR%VlLh37vB4&
z4*@~o`o*4{AM@w`=|A#EfB45FQ9_m#{K=pEIZ4vx_U)TAX;$jZ%lO|9SYMu2I-CZH
zaY_HRcHZlY;yS3f%Gfo(xd!gsM@qloc@Q83rPf$0NLn3w{ejas`lHshT?KJhSu9x_
zezUe%H=Ea`Uu=Lx?GaZqyxM4rqG;a7>EWCPGm2vRec9*-t+kB_@N_n2I-S#QX+oiJ
z8Nrq4-f8UCPGNn7t}tHpmy+sIb?o}<t{aC>^~Sr}W#DHx>MjjVP~MnnG@rkAHO<u-
z<4Wr#wIAkcac{201CJV17e&cpzCaKYs1|Y9CP^;?JK^O4(wijvH+0V^0t}?Zg30uP
zi_ry>$rvG>d21MkD6d(qs*19#)&PF3v0Gzx9Ovr-0wF!KTajnEV;8MKSDN`^PMW5T
z6rd<eCetZL$0wYeoRFp&T9*trpI;i+s;W3TK3uIoQqqbO*Tc4{`CsQju56Gs^R@=c
z>>BiFo|&~Etp;<eJGQnCteu-RhCItTJ3S}MGOP_yks=5bqtOM+Y)PqIwnRx6JV$X%
zx7()EX(NOKhRfxW-Mt-*F)Y(1d6BQBfY-I6Kj+bqDr;c2S#s+FJDN4z+WN<<wW9ua
zy_K#`Os-bIpJL6nKB$w@1vTAnhq~Xk6i%a96qW1yYGHIS;^5!_p(I<y7Q>AmLb&y%
z$d}|<%6#UE@X9<V5Q-oO$(AW+XJ;HA9b=6lj1*BE(C@dM8FUg8#m-vAmO1@I(%Ect
zdpl&K6%d8K%Cz0SR8Se#I%WhyIUABhXd(@wW4XOIVEU|0I?AZDs{#fPwMC8xeA2Fh
zuUZt)1Q33oaqUtCVW&-ZbCdq|CP}Y@RRPuqUe@9PU44IE9@N200=xQ#39)h?mo*@|
z^6Bd$=(-ZiUG-{HltO@zt{sh%h*m<hF{Ho0$Hx942eUca=w@kco?YC%a1Xkx;%NY$
z6U#Zhk+yYbM4zs8n@&1gFn)f>{Op{;jhm<>K?&7RsfJotr@S!I5=x2G760Mi{A+&v
z-Vgc1-}@bQws$Ftl3u@0r`zVylZPB0K4UQGaeh9fswy^yTV&M*z21fsNCy!`ULsXQ
zyWQp0S6}DP|NPI`+uP^n&07>jiBgg`-+Y6Mi!+u>S1Rwy>s^}8Xy3b&%cAGzK3sVq
zjj<TN1{l2(QLp~IeDAb2m)^rz>-&s?!FbKyWq^38;%e<~{UqjgHLv8Qd19@Lh>g)y
zm5T-|U2$;ult26P|HvQy!5?FdpeQx}<Nx??`A7fgAF{nQxO5)Mm*I;hBwb(BJ)2iY
zUSbU(*-t;`8oXdHzh6CqP5j@4Z?d87t;I7nmLQDi^_&syrH0aLESCc<>rCDCDV63v
zu+G-4?q|&otw4~|w^o(*^T+`|;XUtG!B9ZIKSWA}t~6y?H19E+O_^0QT46*ODU7We
z$NQ`2Sg#B9eOXWK*WY(VmY3PF_3K-|&QJTgIxau!;alSRl-pXtj966;y7O+<>(zN+
zFJ1TQ-#!S{V6~+x4Oy0Ra(c?~@rc9r2?rZTv|25vQxrIWwZ_2KvVpoNLIA?RGMh1)
zUT}VK!gMx5N!Keg3In1@`D)t;rCc?$u1o{o)(;_^K2V|%#+`ef=M;H?5P~>~DYBID
zc*I4wPq*76iW8P;#`$Q>(`N@fdi0Q@D3D50nDa~TFQuZ_8xX}2N+<%Qh*~k-ZjUgC
zT)R5IiMg>Bp5=1mXYr&!jqI$Y_9Aslj{APi(lr~axCG5Fol|{3%Stn!FF1I1L{V8n
zRgaJ5VtURpoujJ~DI~fw^oAXml8$0_ws$x^J!3qcAcSILxW(boAr}`T3ay*Zai6um
zgT3To*ZveW+qS;eTz=c;<z{8@Qjb*w>ij%ItQaUf@lxAE3$8!g8Z%egSxYM+2$Pt*
zckW;<RE0&!fFdiHm>Fr1VU1?Cn6gY4-XPtuu{lIYL%LY7m`*6Ol<~#b^+vZ4ClOX@
z4v(Mn$p;^i&2yqy5{Dt3Zi}0D_K0Iw&?787diV)C8?(_iY!4!~dZ5sl3anD}l9tOn
zS%)j6v8W(KMgbylWD`~+N<-WY`TXY|kWOYi`RIUgHODH2$gOkW(AK|)w`jB0b)8U>
zMiFIQw#Gpni46i;+dFLBxWQm|583IEp)k5^T994dL&e(vfBNGkf!&(7lmDKOFK#C8
zdahrf{lA=WTDWM`7~gwNN`g+8?)DZNH+MNbIVKxTojryX1fp)zD?G|J&1E~c?3I2@
zNQbKHL1}=X1(6qw4-T0hpO8J65fA%_TAa|(tO6;lW9raIZ5SUu=iA@-QvzF&&8GxP
zFrG~4^}A?YQ5HGh{N`Wq_22m;PR~Y2sW3(psK^14!cY_?ql+<7(s4d>S<d<Sh_kbE
zdc7WT908UjN!Z`t<>8|_x~#C)P?j#*uCYbUJ}H4;`~w6o1t+v=G-Iw^6-_l{6AOsd
zhU)L82i}+OyY;`X4VUx<*HQfSzZ>T2nz?ZKbq>UYVaRYeWHcH94v;9NoKK$LKr3DG
z?D<o^^WAUrwXggNMOIQ27610%{5$^PKl~$Ft(H$Mxe8t(t}aWW*=Scq4K*lO&CU9C
z*MHUTbosq4tNY;&r`N25AZIQMVHnZrcIoy8#BuwDF>?3StbJC!0I%Qk%VT@pvsRmn
zy1%bLIHFGFR}_UaS&idXvl(J>&w{l?VM4bn*xD*R>t{UMR4|{-D3%$WZVPKG@}g)y
zr`9&EU(0o!Cz?4>Z&czkxVsAKu4?ao7E$rvbePg&D*s%mCVITwVWMF^UY!4ztAGXC
zYe2y<21Yx^eLkOadUnEWmLgC@apHnWcLSP8-I{t%Swzk5T31X_<~h^(m}NR=nJy@b
z47{hcl!9)j#T&1_&d$y*!@&kp1itFrwRv1WYM>yE1ml8WQ+Wnf2$C=aU9wzEI6prj
zi4wZ)F0H7|{`O5G+2ZEjEwm|7fn>Px6!<?h@84~A_?^G~J8o_A&!0;lTj#qMC>qp4
zgEeTNMa^~@fmkyLcZKH$>Bo(x^E!>4kio8jJF)gDE|6R<GR`li93GsaY=R0xthCIt
zG4p&(u2ZxyD5XfE7Tbd@qBtTea}<jCbViZo?CtF#37F01OlNa<|AaTIH3DI!XD5Ho
z0OYcp^ZF0m|C{;ef`M8tApj9v7J96nJ--(ELMtJi_O>q0U%mU)Pjkw56~*-W18&^B
ziK!~kif(Iz`}glNpQp?hb0)Jfvuwg(yTiujCbw_jBu*e-&RI;x%qAo9*@7Y~kwOs$
zAvTo!;KT26^7x49Xi6eNY*nx^>~s6>KJ8w}W`BUtlKI7)(ZO?~;)p?O!B)2-j3t=~
zp$rIxBD5j4D6wvuQ%Z>nLWj;Q2<<XN#()hV>cMBf_JH;<;`yUv4jx{(LQCtgL7<(5
zo2ra^Zlt7?60L-rm)aGRB7ry{+}NOZ`vJ+$EyDgLB8aeM*#K)HKnv@9XvG@){E}1T
zC5tCkk4OD!O-k>YHTJp<UEn5@R03<IBYu4=53r~(pf%{zzqLc}Xp`l9hAyo$vBd)h
zV^^DYv%jtbJpXTDeOgga;XsKLf*3`9F=ljl$j12@?cHru7}gi*z^VXYD*`l;FeqD)
zUmP+%c*M=ScX{&3kGXN{Hjke?W@~GkD01yT{^Gy<1;6+8KO%`^@}hF*D?@|`SS(V;
z;|W=8uq*)B+ui5QH{Rs^_kYBAJf_?2kfuw5Kr!4HkQ})p3!te~b-=D(0PWg2yG3io
zpw{5UEiSU&OuZ;zwZJvElV-uN>-SKsy^r<%LTljhqK3o1@XpsCD~69I)wj2|$MNy8
zH&R`t0<Co^<-ExF(T{$}#>NJ3z41Ag%baJ=pYx4x`~_eC`tLSXu|XJkE!XPGRtul?
z*KeD%ZGh++c)AKKt{&@vSWs(IR^PLp`K`3;3o3*nX*m<(cBkvou<HP=InVl<SD|<f
z>aLFURdMvXhN~Xas&XK%sw&5fSf`WBi@ZtGs<o7^HU^`;21|y{p1>H_8&mtVo;FNJ
z7tCfeqDVGcJlATf?zpfz=j*<jbGO#}lD#<A^)+3+AJ@&LUj$?4J`d1d=e7o_SL5!(
z#VaMfP7>ig*4Cz+^|Y|CK67COp^CV5^EOgOC>0TeQR8f1@7FE`uKh-12tsFNVzi+w
z3r3T3SJ4+rT1o6cqy%G~hRqmFv6hBKxO6INEhR3UR7H9N<wkO0HI;D|4@$YTWEcjd
zsY^MIrxQ}&m0>vSlO!?Q#U_<5L0DSt^3vEyB^eHf)G0}03^LCs%UZYg;vVusVA!lT
zt9!N1fz`!D^><_~E3H9It$(bsmDg3xCQ~lX&snA^q3AfHcWYQK78F(C=c;2+_ix@L
z3Iq3Sgo?#tPFa?OVZiqGHhEE?OZUM_sY|k$COyA)&M$YY2B=A~)<FRE{=B{>tc_p2
zBwN3C)N@qWCP1mru^GeK380zISI%6l{^iWCL$ubeIy+A2_Xq6n?{j)`Mp0x~s|f<d
z&09AZ4*LuTT})Lmn@pHaCoE<&bY9}*O94UzOvf|wEO)j-3ar)?#`&VEC}3lAo9)3K
zXNMO|&d#akQxaRyQigUU>9$*Leg_d@98=4Yq>xU!6a-ErX94Rf)$0m$socCBZ1#Bd
zjr(kE?y$Xa#={>!XE|C>WhJ_D>!k7fX2bIPN#?rc2}mM>ewX%*J-XY@-b6+bMg^Fn
z@XxqjTjN%G*vohRr#oI+V^<rH8t)l)UHWIW8DFyi`)WJsd&?nQaf48`;i*6b0r6&!
zjoW)HCsWGPsS`MAz!<ELYz1}|mes(nVP%{ph2LI1GvFe(2;b!S@QB^%1g%SVUZQ44
z1WH<zHAKP?>WpGOW_t33NT)pi<O6;@aP7BDRgz^HK@<`Ml9SV8zW?sK+`s=Be)Rr_
z1c762f*{0DG8s?Ny#TG9g-KDA?C;-TXJ-d%4UeBZ=HbJS(Yitf(Cv0uEEZ&Gfl`91
za!=^x)KGoxE@gTdB>4MgS0<$|f;s>9b<l9-ehO>xea-x9X%?1sfL8zPr#Y?~(M~24
z!Z76a?b|$i_N+PYdPS<VrK}1T>74I>|6TTWZ_w@Zu_oZ%ci&}aXPdX*ew!qTDXXFZ
z|Mj_EisUV<yIt9}3tMN#u49Van5@8{@OC_+)_Quo3}X~Vc-sr9h@zNwyGzn)BV^z;
zb*vUh`{L(cf2UTd%z7Nvn$v%yrLvb(pAB&JZ&#;ijWG`N3-|e=(R8_!3xbe-f8dHJ
z%7U_Vf9fLRVsu7T2@f0{t0#nB19mT7%XMRUdH$@x-=)G2F5j2+?_GV}&w5yk(-3>M
z%u1hr=fySdDkE(D7z!(1c>WxjWSjdCM=`f<-l5;$B#v9CAar&LO+#Fl5-ZEfb@<Sl
zD2W^kSXE?M%Hhd#qDYaZ3-1Lc5YnP`MN#HVW)qYO0jG=BXdf)q%pc9#xB^iwFf!Hw
zbwn)?7k&dXWO>GHK4qCMNm^}^Rzi}*7;P!a9E%}{x|gqCxT<U8*+x8IT(87`+56Ny
z+pLzW)jg;KS<zf){rBc|#*OoQHfJ)PP*x>j==7BuLz*tpUQ4KiLWGjt-CdT8ImT49
z+Z{%eF~%C=q{YT?gVU39q*TOFf)$3P$(*%NJ>WktTgP>4!}Wn){n}OGh9^e+B4Du~
z{o209l>3ogd-LlEfJ;BdT9=i`vVyoDqf~$lT<fo-6%z&ufhCD!+T9-APK#C?kf#e~
z(+SJDE4)ENsA3S3Qd?%T1!0)b?sPDjW?oFOp&;(I=naQ#Z0!)qHdS`aVl;ttP7)c~
zp`;T_I_(z7fGCQIlf+$_)lR#w{JJStnsTGHBc7z2^C&?q1LCMnueU)v?o(zJC*e7>
z@tk~_VQ{%aYwAKdgYp<e`YKq}X%TPq>F;fm4Eo3<#;U+s#MqU&JJ1AKHJJWc5)&^6
z?5xEa?Sma}hFYgk)|m=>^-R{;Bd;3)T&t`)EvP^RC;?W5gq;qXH}}X#GY&^{M~(Xb
zgx~B})gpH7fn`mKYpDIl@SiCypq*xJe0<8{VobJ})9MdBMJ<siZ+=>#L_v0b#NkKp
zGkW*|4|X>A>mNSk&;I1kSmYI(d;5IvJKte2=(}i5YyQo@`=9vV{cpcYyWJu$QZRx%
zD`|IQjMnI?Brm}dF&u8VG?uQockdow``Xv|fBZlH?|k^-hp5WXZg&_A`WQW7xh!0P
zfECorkw7|Q-=>>{b<B!w7Bl~lt_5w^9qU`vI#8>DyNh!4S`WO{#(j0-y)0w5&fc7#
zpR>2S%dJ~CdH(D<Yg@YWek%=CR52cp`1W_c#ozh;KcvVrqzd@6Kl@MY?d`LFWA73$
zxwH;8!T&0<x75K1ml8FX7^eDH{pRcbzJ87M;^6;|aP4-Cdmj}9wA*buosI+fb&pOz
zpRKWt;dag1m>O)c0@C&3yAFtzTG^aj1spGE-u%yu(PUZb3|CEsZ{<yWYgcJ<Hy&Xa
zF&GRTcr%u=$f?SVtjL(pM-)Y|I>)BiXRV!?|7zqHtQNAX!Npb3_BR2;zv=rgbu{<9
zj+iyzHEyh)hq}W2>NKx|s%BoZ+Gve-x<)A^opzUvVUPXYo3z>;lnNc7R;znxtwooP
zbv4?MBrzgTRAoWBoHKmhXLDmnnx|-85k=01s4PpC*^+~U=LAv_g%QSDvNZLKf&$Mb
z`?GfU+B!=kEi~FxP5(}d#-OPU>a#p$G8r?Oj2ZL?1d&3j0HX}nD%Lu`T^g4nckGwO
zpsjIzZ{=lR`8V7gAC#=EKmNO`Oakkxo+~P=pLw&gk*&XHjA1gKFu9nx9m8se|BPXo
z&YhOZO2R<WPP*LK+vD-$l-yJdhkc$rdrS~227^As!3M{N=X5)LqBur{mV?6+tS&I7
zg7qHeKj*Q&R@Zm&sz$Zey(+)|t&a=VrKo8FzE!|<N$wz7TNmmzr1@)eGDTJ}oy?d{
zXY_kR%Bo^En{a%5gp$hnTMA3R(`LBYBai}97M!0PF*-j(mtJxsLz1LLnwHEK3zlil
z{{C&U?vRV*go}$a23s54ef>Ufec^NL-MGWk_nt938&OTisC>e(E7)j947(BCZkI(?
z61Q5kJ8ddmqK!aBf$yM#6V+PlGY#7H=w6K!R%)aOh`TX2AMDd<wfN-2hdh1!oaaxU
zV@gY9B^c>g$<kYk37j513hC@@(z|_w&D%GSgAO(d(8^`noYj^Ou*GUGZ|(u&#Ruwd
zI9?9e71}F{4b~J5xN~m7H4t#7rPfMZpGtpf^#)@Eh`RNYB&-oD5XjKYm>@zWeUg68
z#{P`y!7<tKnOl&oB(MT&1B@^(9pZH?+8RpdHsewhMpI!ZWZ-R21c7l;o3=1tEXhvK
zFw+Hg&k_(IZ3H4lOARJRrW2}<e@OhJcM<P@pSL&rJWASpI9~FtZ~cJ(<v)InP887F
z8q(X^BwOYj9Ut+HZ~Pfw{Nh*n(U0CIlp%{rK@vrvHP{>@G+9;98+O^dag!pi2;(lj
zZij#RPyRXo@?ZX6c>L@sG7$8K+Z2_?h&fqSG&({loN2L?()a9x(rRZOe1$V`)-4L8
zv)8DfOox9wE3p>2tS`28N)#bnORz?tV_2t)>tlT){-n+J64>yYN*IQmo*Z-M)=lo-
zy20~<LqM!PM+nliAPPzjj-K<Q4}Qq!KJx|Uqa}H<<lp`KKjR<&)Bgo=D3QW|PvIM3
zz}D;)jd`%|j#P_vK)AWoq-iY(<-F(IZtvJ->CJz!D@_^FZ6r>(EwB>GD#t2_TM<bo
zCWs`)I3rzS>(8((XEQ?Ui+BVGDN&)qDv3Zk5wxv^{hpz<0<A4&sVU2fbud?B#P!sy
z)|x;{*S1ubo&lX<Y#Rs=B6Osz0IaviK!bJqt~lv17+NX~Wm!?`lG14RXJsk1p{(2@
zXzNNdA&^2f8rPa$u}wAFYA)8G$<52lmA}8JUa)`DQ3Hhrpi*1^T*FgbVeGDdzg5I3
zgnL#4r3ixnt2OiaoT@a0v1=<P5Gdt3QV1c?LOF&DXeTXH5Kxpk2u&-A>2<oqVTi>b
z1Ggy@Wlo?JXDV@NMPpekmYkiQ5e6ZH!H`a;>kDQsvz-=apj=j^tEknkqeWGgl%_-o
z&_zL7PB}R_VmKJk>-LEv7qsSOhBg(U$S#e8F@{AtZ(LWKAgN|+Kh0F_x*PPun0uhN
z7TjCmMEyqKpL@6OxVdco=Nj^5%6vX!v6v&Q#aL$wUw7mQgMc`Sh{J?i`?rash*mp+
z+@MW~HYJ-I+iY!aI$gF681^@ifknlFjm<66WzJ|c;^~uTm-_Vn&9mpH9FjWlbb1dr
ztU@%lIL7)S!kV$K*Bw7tS2HeGkD5j>Yu{*I?k!lfu2?P?oSmEz1qn%#5QiE6pa1Uv
z&DXyARjkp3O3)eh5Z1DsPMJ?eT$~(H6%|qj1Yu04(_=PUa(H~q>FGI23g5;uVgKeX
z2Pb`Q-QQ(vXOkpulTHfGo*XiI`kd_ai0x>>t&KkSZuHpRh{&^)AarKhDh$1(!L2b`
zI~`%2enkkPIHpdiR+Tf-R)KPzR&e`jZztxh-mBbxb)V;tZt~+Ff52>%k}oUAI0kP1
z86zmHB)PT6@WEXMw{FtCxkoD6R6?SSMG5!W_4=~19$MW$Dc35Mf6C*ffL&D<-YnQh
z+}5?WQ8IAeZ~i2tRE<r?wTFE%N?WfEls1GgBy6?mZ*9@v*`yp#Fqtc>Hr5dO2h}#N
zJKl-|M~l`~D)mDzpeaqpx(l~?&g}Gzaxz5~AX^F@fDi>?Rj?sS^w}|Lb^+;_lSdEH
z-Cf?ef16Lf|1lRA7ku)`Cv0vEna=0j+~4Qy<N}jb{N*?Qf-imHOALBl7K<6iYNi(x
zLTND=f=<hcov~~TH<(Q3UPC8&<-u$Gvw!x_`QQJGf5Fkwb3&iOGsZ>jMNzR_<_)t_
zTV_<1{?vtH#rUkm#8}rKvNl4Lb%ipkO?KUaq9#Y{<4VT%%IDc@|NfH>AzXT}okX0R
zoN)i%9rpM3I6Xb1_DTZKmG(9cCGWlWLvHTfp%eGX@|4rlGrs-T-{v=d{g-`p7t}3=
zR;^yV25~L&Y&PJA0lv)iH8ZIOtjLuqv*+%GT`iUZTvPSRr9;CoCJbY0ovJa8v9Q)H
ziccRsrV|ZlMKNI*61EZok$_Z<z<JgGb;T++V^^o+eM;miV`q$O^d3Z^V_>mpU9ecp
zXt%ov6%t6$+I1hWm$*_F{VSzNk~W*00&6uzl~PrO>(?j)%F-2C7^!_bwad0S>+@m#
zyk1}luB1HI=i$|#U)AmXB9EGVvvuIJ-p8>4B0t6UT+`atrFR5^swz1@zu@Tb1QkRi
ztrne5&sjG_F>xFdhOUxT_)1r$1c7zHuE;aWG6OWVwoH0&E?sHztfVRmf+)aR2kg(!
z&uO(<BuU}{&g%5m`7dJ(MOl!hDOsM9mpMgMID>Sz@?)xk(ddlHc*jYUY{&gxQDAL_
zRMj=Z;H5S`Faf`Qh<Y5ariFjXF|OISIy>Xv*a%#J;V3I_Az=|zNTCUY=3;ijayBAc
zF8#cOFbq&Q(jW{1Y$WJ)``oy3i!{rT(&-J$svrmyy?&2Qw?kRELb<F+5lYZ%b=ceA
zr`PRsc6vsZWgH)!tzAoPsdN2tX`XZSiNA9BcEwQEgJAIUO{~@jSUDF6QJebv=gLR9
z^*v}z-Vu!#=L_c;;LNS1d%w|mf|LQBPKR!%%f?`nG+hveiYN&P1Ho)MVKyEyACIUC
z*9I?+TSRdSYYaz6M@+_J$|9q{FdAPVMW3V<uzz!tH$U?RgWiDI#hmez3ns@0<g-(x
zp0YQHxU;*#up6L+p(soG{Y_L5QELm=GaOjAwn?QBt|S&yIq)roAXFh`RbeQRRBq=G
z1{FhZ7<2p8T|}fgczVj&@tE=P1Y7$3AqbFBo8JCDy}f-p+dD*^9%-%|b8fVkq}gWA
zsEf&1Yp*BjQu&uSUXHQz8Y_!HYvCwOrI5;jyXN>C7TEPPpL_XbF}0Ku83q_B2-|Ho
zZ|swvkC-2hh$N1abEPR)-np&-lq;zUUW>OPuW5ju(acBZ<kLAKGuWO*%Zx;5Hf>IF
za!7IfoUYVl8y%t~=Hb%=_TT;j|KJaPhyUrDf5neK{*Yh(>X$is_=wHTeV%>t6kBOV
zmEjxT_)q-mul^dJd@@5RNuFhlM-#L(^x_Cz8OEbA{q6=~7_eB(N#cmA*X2uJ{w4m|
zKl@+vfBZlF3r3@JjDg{BlQ0Z0#xfpF8IPxp@o5bZ?58~b^#)Y~B%xN_D*)$hc4dRC
z@5yD2+2wmBu8FkQQ?W0z*;>2OHbimE!NC!qdFw4^vpHtDpwf;_3&Vi<d_fpSEEaQq
z@WXfc-QWCu(riJIE&2Ahzr(Aq+~vmpju)j?E?779*!Wc0+W9qS=^ET#W;@q_RIO9H
zjBx9B<<i(`?FwF;*=dU)h??}5({p(CSZf}B@{oROgF&Z9(r(ck4(KXLpkjZ2h2I!g
zwk7U;4A$x<>Z|?gtOxzB#guHAyPBENT9alOXJ@B`VT(N12oWMhM4Tj8v<IYN1FY&J
zj$;Oc0Xx|)dAcOeQxrapr!`t>tX_@v)j72e?A95h7qxeynSbk`c0C<^o#yW6J1)PU
z)7xEF{DGhQFMj@Y?1yMJRkwNQs^a|ol*xF8MGz+r*tOd&k~pE2B!rPm&xT=0yWK`9
zh~vb8>0(Kx3oI^GYOO(679#{jmXj`*&Z@+dRe6zn({r)blJN3*T46D`bW~ZEESC!w
zi#b`5x)hf71rff-^K3d{G942}VPnnWUgPHEPa5A^9<w$;FO1t|fOi?(HDKfgz~|N^
zEASg+O$&m+5D1MhCBo*2a>-)+oOC{>$TO@B5!Mn05+f8sIv-3WB6{6E+gsaw^5kP`
z`9rC5q9~-->(XkqNarO=1_&#WNRp&Quh-|sjT^MvZJr+<FuIs{e^Zb5FJ0A(*Z#7{
zRRCCz(z<G(){-M+vu<47lbgRwXt3wnd(<Bk28yH=*KrHRSh~F)-A<QIyG^^@!Wd`B
z9w;b^oato3VlpLNEWzM3nsEyihKw#o%;!_eGDphV-!3J~Btk{RNx=Tz9!f>b&L@lx
zj#(T(N2g=DVactX4YoJh#IeEH41^$z6NFM2tZRw4Mp9Vs?OB_;TjPqqYVa$Ca8>SD
zAJhPTZ46>TKWKs6Lk5a44iUyMJzi3j0%S-y9Maw0rM0<D)Egk97NRVj?$UZqpIO<9
ziF!t_vnD^oRCqbN6I~&VL02HGiw>m-y!-cyZuviHvAx95`IJD7qNS9O3QB8{fugsw
z!`7WC#iBq@GGYsu%9XuenZ-?2&l&;agyFV%i0a5K6p&BHEG{m{=L_1~eIi>D=ToBT
z1;x{k2o_^Zxunxh*xMd5Tjsp^>K&3-K0}qj2Tz{zo$r2!XU|U9*<SGKy*oU6_MEaT
z`N0p~;j3Ty3ftQooSmIufpl4*q6`@?Xtg`cr&GqG3$}Omn9Zil7jrt@4&7dtzx}s=
zlZ%TH|MI{7Ka-{@Mhgbpn<Pn#!C--S{FK>b+JFPNEULfmsM8*G5llT$TI<H2?z($-
z-QxPGhh1a9Uh;5khGuVX$obhh&z?QyOJDvX-}~<O$g-^IS`bA6v)Pm+iFo$x2`48<
z^twHaE}2fIeCu1^<qv-UcL_uHz}D5nE6to|+l0IXhK=?5n-><=S~s<_iCNp$n}DZg
z0j+1{u#!N91S$elvq-ugt1B<JcB=)}Sz*XrI`iPNbUl-CT^-7D)l#m$KU!B!MQZ)?
zCO{Ik{w`PnD7yyi(v<mP350Ze8B#<9Dkg3v7*mlX3B%#gn>^R&?Vi&pO6YY53<jG_
zCL@H^Nbi+)c@A9<*scS6SHV%e$y^0>S3%$}65L%qulkU3z4xnI?_9yv<=0-nAq#Ic
zw9+|gQ*kjm;nCA4WLbuk0ddl%9k+<Wr~%Y9sN2}w#M-y$4+aPcMU?^-ps}`eExok6
zzticA$#~3ixkQF8Ajq=J6*ia-ZZ5-$@vjBiwwX&sku#Z$IXgQ=+p=k!<X+|0)9HN5
z`Pr#!3n?X$w`#(hEMFM{CsnxmJ+5c|*Y#ozz^mc0t-nlD(0B=S_*Wy=Svp?~M}ZPc
zY%yVe{*?2_@39!2qq7RU_EmYKcPS-N6c7g;wl=re-rDASKlm<jEC~WZn&q^6Ery!|
z;-Jm>$tg-IqA(#&0y^yuMP9I6F6nez{K~I<l_!s%Fdj`QiaMipDarZ57+k)jwehf@
zzQV3R@A^24CLVP~Hm<+AvRB4g2!x;K&HVRkV$E0A<E@mSl|&4O{br3C47T{YfAsh0
z_j(k0MxHJSB8iX&Q&pTCA9Heah|Y4PHY8!pW`6?}hAi`fCy$?^l%&-TsVdMVv|>q7
z%vokRv-y;eh<sLZet62s#~-mc{)8JH!|nYRZ@#ieud67vE9QzjZA2JhJewU#M+%v$
z!U_lC!YFn+XYC{iQo5>Vg!}wBh+WFoYVv&PNddQSSYhdJwCVQmbL-9x4tBQr-hcTa
z<9S6Dw&~v5XK?!tt*sqYr{nsMIYUc~HdNYR$_kA2Y~ba3g`f3!ImRwqBFh3}LtjGf
zOnlulTQ9<{KWgdA#IIXy*C*1arR<C~AO$*5guN~sH*c{R&pBT_qbe(6q1NL!_lKw%
z09VcCzGacZA#E4MIgOFBn#fv;`IN=zgfySBUk(VD7i1?7xfmZ1r&AOLTc-4sVs~?!
z_dj?>Hag;+|L_st|G~!`%yLBBrr+tX+3)e?w?EIbXV09+-S~{Z{L63fo4@&6Eao%P
zRHMt1<vgP^G{iwfo|jyloe+f~+gqC)pPbR_^_fm*Y;EoE2Y>KKESF3E-M{-c7^@ld
zdmt=v95J0v$(AW)UN(TPNo{?~Eo`DS&)B=J9DcloH+Eio?fSjB%H*xT<~p6mPn#aA
zQy|XvK%%u_u~_ozYp?LZ2OqLn%z5+8*ZJh(C$0}+QF{9jL7FWIRmA)6zsGO=)^C%g
zDLRgM|3@G2)|+o~@7^txa^7vyxA+3<iW=&%xN<JdoLXlL8Y=<6zKWHOt%-=6`Rt*h
z-^^X<{;JZcet4yqJqJp0@7_Ji)IG{dIS(fQt6c?QQMlBYkglS#j*#na58CKUpx)i%
zx@FX=5JJ>ZsBpTq`C^Gu=Y(O4AP`igCChV)GNaq=k|YU15WE1sq;zT7R?>DYe{|lM
z-r9QS<@)OQb#QhSL|r$>{-)z+WjC%n+}*Ry;@U8rl}o`mfahK-)mj?sQ#A=dlLlL{
zx&ng=EOG2*O2Qz60;>dtDae+FtjNF^;y7V4A2QsEvFa8{D?tkAb`y-rNwXOjqcfg8
zc}Ar*i^YQRc+6xpWqW5E87lI!U@>3NX?LA%P2HV^8YH>vsH%c2PnpdpoSvO<aCm@_
zmZTli>2wLC2dUPA)r>}GL}5Uv6q{RH1flRjw7q5={2^XIVJ~^Cv%&5$yZU(oP}U%<
zArJy3EP*XC^HUb1XJnH@^3hY|_!OCG&;j@^Dp*%^7eygSlCU+{W_Np=IF4D)=fp|R
z>0eXL_VyP2UJs)K%AzF84Xt*YC{`?&Im68^$H#|+VZdOx!LR(vFY}#mzr*qIDMnY<
z2Ef-9lZblU)`Oob!+P~-LUCKq>9wKrKNZ*Lx9TFSTIbs6hOrnY<0A+paT2k$y+OCr
zcENTS^E+SvBktb4hcSlvY=Wi08qIio!NvI*$HzwqZHZ+-FX?jk&I44t&Ee58&z?UA
z1VJbXMTo>{amN?OWcku#3dNI;pHR+nrpKq`=O^s6HJ^QDgU{R@V)dNkqY0V{83k<Y
z+;(ynA8}jb0#L180gn()(^gfbvxAZ#l^ZmHB$N>d1r%VNj@fl>aYNYXQ>7&72=2eO
z&wRS%>1oPRv>7~jK(f7yZ1q7#Ze_7ctk&47LR*U|omrr$L3eE$-4J%-r5~4<0(Qm4
z8T1CoR@?Qy6lz6lt}$@y=9eptmhVC#l&?N91iqCQI0m7%Z7@b7{h(<;#UY*jZFbX)
zbUY_N9;48%j?dOrwc=8xQ$7EszsL%&8WQW@H{|Jp`S}sc*&)@5MO+*pC(jAw0vS}~
zMkCR*tz>(%&;CxExR?{p&Io67?A&te#x1`3tAC5Ktk@X#dHuBqJbC(zEGzl`557y1
zMEug1zskdh3m_m}<jh7hdYvADRG6w_Ji4Ga*dR?)&d)E9DkO^9^!r2p;1B+o!^0Df
zjt)rE85So14a2~t9qP^9SI_BB(KfCkxwY~D>nr!9Xbg?lU%%emyxDAhdTjkY*N<Af
z|FVZM#+kwbv)P>8ogL23&v^LoV}ALUf0@a2j5BI>Esk1A%zQql*Y0w7^o+@LM5~oh
zmL)}=@$GMYmm4>BiDQKja_#zRAa?zEG)z-NGk77ZSZ(U-41t9;BUuL)jjbc#)ibc_
zwy?H73%8JNZ*NnUUgLuBO|lcD6clB}bT;Pn^o%G9eOtN?brrO%+-JsdOcX_HI@kJh
zt#3o;nJ&4iq*cm!m^pLj7^8h<rnF58wl<4izfVF4qNqg}CImrDnk}g+=aXh!pE1^V
z?{&7#tT6&rRedUR_i4=DbuGoLwGA8j^Bi_{8~yFMd=0MMn6H>!*K|>XyA_CQZZNA2
zcvWBRx}zjXf=~h>m1~O>#?FY?LRYtae^Nz~gt+vkM&Rrk!Z08V1GMi``reP;C(B%K
z*gVh4^MZMrA(U$&H=EAr_XjQ&Xf#Dxk{1PWSvas)X$qB-7E9*ojAgp;Y&YIuy}$+v
zB~}{MMY>=z9h1Z{-Ci4^k|vP2GO}JiBUf`n)Z^>Et_=OkTd_H0Q^Dw#et~tJJ3Wx2
z`5D3@r6v#tr7I{i(!~XflTToA2HAwDOp&?(1J?R|$y$o4LJ0_hkR)y~91a-_hA0{O
zXCd^N070wWLMnx+T<ljBC8FIzYr}jzr6?9?QzDgOI-6sRaV_71fU?wH9&#C?iF)RV
zE8nlL+rCJOyfBZ~=G%HPm-Nj|g{-Z;F8`K5_#&_xD!aK;-yA7G27*>AqSNiTKDpYk
zxv|NeyZ89aXWn$dLAE5%m)J^kdU4EbJZ3r>W2=fdO4#TP*&b{Vwi6ycf5zGQ1+(Q6
zV>Pi9U=3xJF`1l`W^*d7iK2uzUi%ypNtG=z*_2pheD1Yf?%ilH=qf^GkTP;=xF8@1
zLl4*_O1dB$T!p#?q9k(mK9z5|C|yy5RNh5HxH(spIaQURtpkj8g9Vf#%W|*hmIOh}
z?fW;7gPfU(Q9C`tc8CZAtj9h|f<R+!<)uO;!WVDU8ape*r|`Hx-SJYuZuayEJA)wt
z2%-ez`w?p6N)}*kv$HE5(5D?&7wqPH)|*Qjs4T=uOw?{OyscTBk2qeWm}Ty$%j@R0
zb&;GDPB&|<TMmV=4RXgC5($ManIAtxkDib-hl^wSszkLdm2wS#rBVb0Z0>BcySqs*
zf;Tt1JnO_<j57`&KIhhdbpsJdrt=wZfBr3WRWO?_IX*t&&;RsK=y!VDzJ1U8vq>&a
zFG%8qt({G@)s#ifY&xMTOHNKskuv1r$BzgC#p|!X!9V%O|0@m;pYzw>`b+-xzy4R;
zxVg(@GNvesOXK-zpbqLy#YF`eyY`*B2;g%3xX#MC2+ma#uRvY>y6ZmY<wJ5Qr7Px;
zg0gg#WUs&W8sGSfZ}Q~H6W)63E#7_i9op?SM@Oev>)5U7d`c3<eD}NG=8yj1@3CBF
z6nViX4<GUD*>fH|xaBvS^-rr)6YJyP^nH$y5X~GGUPQdg`P7^Bs(Kb78aB=`e3nWV
z%;z(f%LRFn(QdaIse$mRfx)Ih1VEsOg1BMxg3v8W0E_vQ(ddE?K6sxvj$Nvv6?-t|
zbWMSB8-1(QqTOy&RfX3@xk3u11ZCy^ErlWo+-s!@==b~ddOeb)hro3Ks47<h8AXx5
z*Dq?!0u|9}b!jDSvgMSj(o|ae?hCNe0bT}V*ZuqIXZ)n!tABP~-`Jn~@NZ%Ly{_*^
z&1$)&!#3I@yS6#57X>tyPxbpXb7u9-Dd~ES3MCOr(QYL~QL?f-uuh97gkx!yl0LQQ
zK2u560QkjX$>HHKi)H34X}nQ%npc-VbWv1jZCocAtz8;eDUcE+LxNH<n@<>x&sk<m
z!YFb9Kp-j0!d1K~h4L*7@-k<>oYR^oY~`CyXN0_-N~#CU4~w+28DD{a?pPTPtyf{z
z-<Q;kxW!60p31+WGVU2QLLj6?*%HbHRXV3ir=-(kii?MY#hgIrSZl~DLuG3_S!bzK
zmKvlW2qNM*q1)@ZDq`U@$$?UYp>rfilDJ7@hhYG|8raPR$z(G17Ah7d6Qod0pWVxc
z@YjXk;MaqS8l2UCUk^5}o5aGK3Oju$zQR6=1C$C-wNIshyecqWbGcg8+&^T^6&bfK
zTaWCbI3Q`oBuRo5&gXJ#XOCB2eVxIe&+*X_>0(ZvW#r2R=clJE=QD~dBMBomh66VG
zL)z^Qlf{(7vm<7+1w~mPtVS0aZ8Gv=!7`mwmKkB(qQ9}pt8cu)@_dFaW`uUm?nc0a
z8$Gsqf+#SKUlJiwhNviXg@vZ>8!xd!Qkj~ha6xRPON`M?z)?5$pl+R6=2Ur3sS77b
z@ODAe0NYqsk&Hzs%f@yO+iug<5k)&fM^Fi;hIK@W1z|8$RHkwPn=lPg=0Kgebe~`L
zcqw3a^x^w#ZEVs;!f?ZPYp@vW1hcRL0Y8bgS&zi8M(9$yZWJM|Rfn+#qa-4UX$|{q
z-MT?~cERk~5xO!2>y>K$u%4e^b&zll(6u`>fRX}hD?+PjRVn*=PLgY2NntADQW1)P
zkPsUfmfE7aEe3-&n^8btw)yIv9mew$o|%xxAAiDc|IzRB@bM=YH2eE|7;}g`6_i!M
zpZw{c@Q?nhf6V6QHfI-SRAt7|!68AYXm#48>4G4Nu~<I%@I&5t_XnJvp3>{~>2&(E
zTWy30c<Zgt@vU$DB~PC|CJ2;gR4(@x{j_nf$GWDUt+lQ}J+6J)<z_Q4-ZNm$?DEB*
zHDmuXPRLp$U%R|Ne*A>r`TB3OySvG=XV1BP`zCkp+~MTpn60e=<MEVEC!xqQjM1E&
z9D9@Fh&VRn>6~}I_dV|2ySX9?+!;4PaszCeYq|c~n{<ZHfvyyr{{8Bov6|mynR9Y-
z$jQkGN-0{cmS<Ta;y5A<BX;&u;zaY|Cm$1p9ilLz)9J9Wxj`JaU8VBOaB*?L$;k<E
z9J9B#$M*I%olXa(l<Onxi($er^qM%)1USanW)qLYgusTL9aV%Ou3@}VblPo{FV-+t
zH;k>6vSHR%V9V{(Vc@Q#Dl76lcds*H1BB{1@oB(ro%y|PO}GxgUB}q{q7ENdHg~mY
zC1!=kC9c(vu$EeHSvNqhZ8hpSwe~JewRz1-86gmXYbB-vw*e(li#Ul00_E*CTwo!E
zBn%>=AVCqho~~A6tj1a4;40<Qg2l2xo7H^I%WTDfDe2sne7ieiVKSRCTg*AXIA_!!
z5rl%tbj<nr8S~|ojs6~S)B<5SIe$iF3*s;$iDH5vq%tMTe8G6Gm@l_!by^4!UVl%7
zXYGV^OO18RkgNKWjlV%HAYYYuNZ*;nx*{EK=VaV%aX`=*gw_bFd@4;NZAyCfn8o>X
z^4S@tIAbVhw5+Szo|GY{OIT`&7Ik@xqNp?|NgTI`lagL<K)c&Tm+mgAz-#Z~xM=~2
z6m+_6W=Y}L^C)Ixd%y=D{fMH<sB}qD<}Bt5CX<=7tg(6>qWR2|AD4!)Td};>GsJFx
zTM?ft5bft`!}K`~x{3liy$&0j1A-u2y|=JjjL%ul78s*fSv8~se?Bj%kr?UZ9x7C{
z`>wKDMFER6=apCPaR0S8c=e6XuvpBPPcF!oDS4iAba23AG)5YT!;oQb!0QiQLt>ds
z#(eLG?=miZR~(~Bgkf@i;>OID$Ur~^8{0cPc;!uQKX{ezJ^c?<=@Ely!moe+Hn#>6
zMd4b7#Q_Fak({*JSPPLf&h`dNT4=1X#9>Sr1b|zoT{cgklt6?M8TjC;$f?o^x-75)
zg1Ai>w%4@H+CQsKTBDR8ZU=<jgn<p2BbdvAr7g(3{KpE75rzswX`R+j2#b`ixWfqd
z<rwMuUj38{m_J?$*ggC4`*eCkZiO-3I6<`%jB%+`>uaIwbn10Ffz`3v{MSL<s(3;m
zse8j(3T@C*kaSvX-Q0)yf``)?#c1Il0)Kc=V;auL)2TE7yopL2`=q(Kx6N0-^m$(G
z1Q2TMPM5OC3AI4wZV}0)AXk=9D)x4F(Wl33g>CK)27KXONV|$KWx;54!OreBi^YsP
zcW!cY<g^u1z<jpgPyh7K`2FAe13K*v^RyscE;)RD!0!GoG7PauTAcx#o147z&JTG1
z{SWxeo1byfOkS~AELkj5EP^c0*xc;XZnv0Cm#+2DPr1R>IxD>RTjE5Z*V)s!jdX3Z
z^=ZjkOLpn5)|>Iq$k<Uw?%wHtGM(`3*)v{${SChLt?%%`2Osk5zy50+A0M%|w+q&i
z<rzsFlcr0eFyVs_e$3}T_a(ArMi4|idi0djlQZ`Ax7QfPE9bp-C1wqnx;ZRYudyq$
zYwY?pu3HGKMc!D9DVa=1%;$6G{=cS$6GHHnug!prfBWzMJweo=)9LcsYp?OS&wiF}
zzfVz>T#QDXo}Lgz5pf)|SS(mBm-PG2KdfN@*Mh40-qnkERaMOAbMmY}1&-y(@{CFw
zX0s)r$a75?wA}p@>%Cs_z;2b=wZ6;7az6D<(5H0Tu}j4oyEaEI&yD);b&7ZWdzV39
z{d&9hdRG;^GCcMho?W-H`S}i)o?6K})>Et2)@f+>IjZgj+<d;ibPwv~uVKoCw~z1}
zvqYg}h!Co2+vV=1^4@>8*<`~YBoN*u->lxVHa4x(>@Pl*m#;01LZ7mC?C|sF&pA9i
z<m~LsRr&>jtjH*elHqWZFMWQKR@_AhSS&Km$0xLsm_e_{#&GCZpCV^6opF9~PQO1u
z3i;x_uwW{0sx7TACaJHZhP~_Z{2JuZ{AmFzLWB?mx*{wrl~Py*Wgs!3pruR57nGwj
zj-R~;vlA#62%DmU1%i@FK!N1E)SMg3!Wv2pVcmu;FF^)`VT)E$(&_YQwK~koxnuqf
z2rFqN2_XEl;R3f-J0}WTu7DzF^V<D4`Q-6OjM59LN;4Xb$x|onuUX;dj`_Lg{`9P5
z!|<%8V*MLJ;kq$3Z?Sr2@H((K3EAG+WOH+at<6osD5fk+x8F)hx7%SppR-&pSS}W1
zY3?F6>1>ZoWl$=h+iTJ3b_hc0I+z&Op=8kO@f*MTtK7PA8;Xk4<3rAlPB=R~=i+=s
zo-T=^h|S@U8@s!F@$E10{Mj=eJ$lHKCr=n>ONbRp0#(V;@ezy33=v2~s3>hkD-OA_
zwadL5dpvpn9iIRA`@Fsx@TCWPymDiREH4n!fP%8pbo&G1R*VV*D&wlf<0M3diZstD
zvVurOejO|UR}dytK%fFAJF#HO5<SnzmrJaYgi%Bix7-LD<I=g>5U3Cl2;ZGWQkvZN
zh_*yzM0$=9AVdrzps?1ld<Lu$7`vJStK00JWGNs}b&JrS{&*>1H-CP><M_M8gEnD5
zrqfXncp12GDH|ySfm8&-ReuOmPfX$2OXp}P5PmUqms3~zS?ey;DlpQsc-A-OmLctp
zP3(Qc>G_1S<wuCZK<U_r5{nH5S_nb`p^-?TD0Ss4cP&P02rOFW#2v*i|HkKe<IOu@
z3v4BkVN5FuiBM>plPyxT5}3--_1&b$CzFTRj>J?`?r%k$9Gp^|clh34f19uT>MxUo
z&Q$cR&%VON*#*U1k!6ObpFH84fBk2C?bp6aUIvu8WHFv{fu%d>qf1S@(`C5PrQ2&E
zRmCEmaC~}z!08d=*a_=jx&MIvpw0Hy4&VIdx2bI5Sd<s;+6rj6Ti^_)9T1RB|3qO7
z$~Qu;yGuCY(?t&2hsQN|thF~a$Z7b9b-Lpv4(t61jBpWgWi`?YKKS4xzW%#kr`zdq
zel}u!G3LR&2OJ$9a{I<DKK|sP(@~fbP07K@6W;#9+o&kS7?{l$eDdfiTf5s$w}2J=
z^z*HW{2fayj0Hb7mQ|X%UOX+VjYEC0Ya6+BKM2Bv8#nH<v9asHi8B%|vYb56D9e&0
z{#PJ!K#(pAtX`6(CAzc}dC6ol;`#HZOvV@Vx^3bprYZ{8Kd~|{x+FmQO>xzc0+$vo
zU4KR)oYla1a?W@%rYtpCo>Nr@rDAq>Zm_$vPn@(Hb`lS=>&WynQ|87iT`o8|IRsnL
zZin31+i`89yoX>72Hm(^25&3o9uE}P88hKQob_K$-{}5yU$(wae%y`ubjJB*k81<d
zD{ORq?)Bd<gSr*CTQ4xGwScP@LM{mit7q4NA5TS@dcGM4f2`ZI@yv1kdp`|xSqIM!
zM7j-ZKA&@Vc*vuNkEpbE%r*w8Bzaykx;Y}05pfh_jA1sNQsf~LODl<q;{>OB&3z^A
zf<XO*ZKYX=*2dO9b5?qPO+PPI47eY93zcx4OQcXpV+o9<l-d`6l~m;g`S~-7^Fze+
z8EuncZH}ctd!==SB-MuVR5MBomSsq#EunI!SCy61SjBG40<VcvQemy7C=0?;Id;&`
z2Pq{<D<+Bq0`KjrM8wY49$B6fsF0JhBLo4?=*+Ln%C6;{ZPU&Q&rsUB_o%E%lR8D`
zgJ4s)qjH|4_~3I*6@U<~&0LnHzOAHe_Fol5^!r0vNt<rFN4i{+XPNh*bq(x?!%YUm
zA>Ce&`C`gqF~b^56h*xK)>{nPJ*qV4;$q6d(`Ot!d&*)yrzlFoFy!XVo7~vjXLC5@
zlShwu^yo38iwla%X?Fz%ou%Z<Ir)47#!}{nGS?K=(r@=@$1M_RIX!&L&7R=FUWdE8
zT_RC{gvo49T9$0>Z4<W>RG|EH6C|;-<FVSHi;_So@6`#qs*uukPEkSNz3c>5SD3P-
zDl)9G1fiP)7)?>S?orlSf*^1Oh~8M(&7T6$NGd{65!=kM`;=H4u`tH#I^CLOj6fL2
zS|jSNT(<E={hyt&6M4?r<3~ipF73@eVJjg-p;Zgp_7hx4lvK_aP3y~xjc^-AlmBn*
z=<D8;o@uZaEi7)GwxBIm8)Oip+6nO%4DR1!d2vqm?3k96SY2U-MF&nKs*OMxcjLk!
zKv+X%N)Q?=E4FV8_}Xv0&CR<zn4mxeft$!hiI9rYY6t{2l<1;FVhKfvP$5ReWI&V%
z#OREB!<Y}oCsf^So_+F+*WSF#<D-Wl;l|B<K6>|nM77bX;=T9ZA?R2>``VW%^O7tr
zSWKprc}{DPGZ+qOwc5P$>K(LMa(*^qKA#XrZSLKBz~1gQpZUyNJbC<MHaB|&VM?dh
zWs#<C&1Z$J)`q=dEk!+}R>R$lIi_ARXf|hr@aEGlop)WeTl0BM1o{$ls`<YFFD@-B
zO&lvMn(1`LXf)+>pZgqt^;h5IgAYF7H-7Ec`Q*coxqIg>xA$|-F3wS5Kvpc7FUQR0
zV^pA6YJ-#^AAI;RZ-3#lv|Anjq3)#J(pvA1K5DA>&gSo4+hTpH#`~JBU%_?Ev@0%1
z*w_qdwGvv1kMy+b`&m^QYc$<%2Sm#6{?6C2wuO=b-ENmor$?6NoSd9+dUA}w5Jyfd
zucX3M8eKUZ6~+Zus~)fJ8gx}rRwb@GvSB)%^4^c%<HHX>psF-V1{mXJN1!5ly#d3)
zrtf9ziYknC#>R0Ruc}`Cm`SCu)==a*Q6yO|mz1SOhyZJ=OGe2r=x8q;tJYTQU#=yY
z%id}2eE+O<QLgL6A*uCm?*6<e%lKIiam_tl2ck8h<bdb8^kmhWw|TE7owj<P8r;<$
zV(Z}W(m2#0SX>^v=Jo5Z`Dxh7bu8uY>teBBGMR9Ge&M>82qzqml8|<*%XB)U*B&?%
zdf{w%tQ?(#Tjkx`#h)}`nkL(}ZQFK}YqBR}vT?FKxf3Vbwrz7V-tX`I2de8joxSh9
z?zKLvOgk7tXORL<w8&#7mkIxFPF8n`NmT24eRX6WYBjnLc_~d0F7P2wlO2Y46>rK0
zG&wzH8hOGtx&k5MCPGTDsSZfD7mt?uXvLN6pp;~pn$KB{kZNdJ)#^;>*bM~8$8>WQ
z)s#Rk+SvB63erdl75pG~d4FQE)hfo<w&w7{ek^%)z(Wt#XQVE7b_?`_Q{VRCJ@Zue
zcu$Zeh4Mh^#d3ugH~p=ij6tC&)FRlwShHrX6K_E?xI3lu>`6FnPc_X7stPLCs!f|R
z*w#VUHa+<cd;f)zz^?t}Yu?6Bm#alPTOF31EVdSGm#2FkEK5u6vn)LKrlysR)%@g!
z;oB^r8QIy(NipZZimqE*K1YTAXg><4i5B_qB*$t1IFbA`J{Sc&9d6S9-hs)*h4x|>
zB3@WUIrX)!fH;~gi21*rEaYgG{=lEY`Wm1HJ_C?tauqmyT85~|Rl*&6x>(YCjK=Y=
zBq__P$5wBGcC|r9x9o{qgP$EbVb(~bvZb~Sa#O!{OdZ55?vn9j5I9!(dgLjz>fJLo
zwVp2_@@Kl!^<Er~{RNt&FQ*3H3?|&(EUhN*1>9&OMg>@pAXs1b@SK)ZFE0K}J-Xi}
zxja$TCK|n(wj!H(c`Z=u$|<WDy;Mz?m6-jD4o*BS9dvO#G=qd8{e}g;sQ>a|l+ioC
zga5NaVovWcK5Xtr%;a4bGaS4|0D2I2EIg=y8ez0uShi<z5~==94OpvdV`3D_h!oJx
z&cMI|S*MBS&21NlPt>i?owo6Mm?&pSD%Z^E!9h$<O(!4c!OKY`iSz_4oRDhisGy<+
z3-8Y_nH!uA9nW8s*1~`lc^BUg?qip;#cxf`&D^|^;Kh|6m0I+EtKdmWgAKTKNmG=;
zW(?EDk@-&h#t*RHJCl0tX06z2W)N>&9&<D}dH;Stn#u3P4uH8_0VAx;989JjUeQS+
zlh4lAw^D~UOS5g=|B}sn29C3`C-{oW`=LpuatR!rod@6B47fgVPuDnSRQgomjkHwB
z`1ROG|8W~y*!j)wy!D)(sxhe$+lr5fZkIO%?_O^3L5t)HhY)AIzB}_A?Ct{lOmTDx
z=s+lUV{}ft-tzL>@i@D*w1qS;f6+7fvc^Sm<KFkznmU$sVnoyOQGskpy)W-19#@-|
z-j=!C_B*rk6y_QU{Z4(iPbe;5Ggv?pJ<lsBGSnL;r%?ZyGG@;E;r`b$sd)s$U@#$D
zd2(B=Iq^YitF)LNTM@?6L^9Srfu%Lj0qJsf<b1OJ2mWv4S)ENW`Xc9&R_UKAC8;Pn
zVc6<p^rhu_ub7U9vU`_2YJ~5-bc1^=Q3JE3Xb@-Y!IEu^sCO$t2gHRN?ilf^vJQs)
zmYr5;RPe^z2$Yy(Fmul!#nRuLOqg)eQ=ye;$cvUP!T%i)HLY436D;$bGfw{v*lG=f
zwV?$~i}RArnjRhzksZPIFTjAa(B^x}YC-DQg)_yyAOvIR?7tK2)9S@+8jHzo_9Rtv
zRyepi!stB6nMW|jDAOktuDk;D2o)3vG8IYjllmCes1!fYn;kX*4wtvtfaR~NccSa}
z$IYGR9X8257R|Lo(&8X@zWlWd)R*3}I{HjbHCpvxm2Hau>?aR4shAYl=xJ85Q)x>q
zc)xij6smGPXqd~G$mNM1EtHCe{$>Dyu+?UMQUPY6*@514eJmIiKBta=Msrisq>XHa
z)+(gF%>a?+splOOs<2XgB0@GEAu0rl4w1NvD!b0N5T=5vgORyXexJ8+S*>v5Zdn;?
z_Gc5BPVUlCtTa|C)kt}{?}B8qWY`a|t|k(4=I7sJI~_eTbC}bdBf?GUSchoE){#zJ
zy}#}02`MoWdu+OyC(Ny`O2VB)Nwx<Ub5b{9Hi#U-Xk?hmZrn#KF0$CE;B$!x{S}yX
zD|MtRg!0)W2#l20qDDYRSyT^hfUlHnmE5(@Ec@pg(Ek*v@3Z|==pgXp9x!2NFaAY8
z_|<a*+)GLK5xtlrF`402?VieZ7Z%5J`v3Z9-a}WddKPDOI2^U(mLU%cB3ox;h4Z0g
zFl2Wp2-m*blSP6`$B0Z*aIJw{nFKSn0!B~@MnA@n<u141MjwqefXvh;ZYNWXHCgZ_
zkbxmkz53Hg2L(=s!3eG|o|FlI=~Kf2FHU1lUaClwhh`7%T!K0@7=C^}%9D%h8Ca#K
z;;lE<+}GLEv{D$UsZYL=BQitJ1C7m>S<4wwM=xH{FA#kbc7s@@*aoU*Fn}S<gm&Vi
zDihILp!XHU($W-?X_+8Sir?!kyl8=v$r(uL1Kukel{YW1X@bNBi-yU(@I#2}Z{b?U
zA+6H?>2qvoZ2O*6q$jd>ecpyyW73&I<KaQMKQ^UjF+s?n53IMigbl?B9*)dOwST<e
zwcllX0JYhNlNW9LLcGL?WXvg1cP<Z2SN}1-a1J_OzXbpZ0?+N!Ih<aHkwtFhD1&r%
z@6)+&fkzKNyC2mIbD?@W9Z9D2F$$5PH)H$Fdk9?qK*}7S_(ct50Q7Re#LuFWlGra0
zUB*oLU-Y8^?H?b!8r{2J@Y>q}T`U=zbR@B+z=;c}#f(5gk}?uCI9@TgjNw8vb<N|_
z>F&^gHyc7!IK>=PY<jL~K#y<bf^it@>GN@qGIxRl*?a>#%OGFT@DXLZabS{^91@!e
zWrHa=$X+?Bv8_(j^@kLJ#*bHEj8?gh{Ttg&BZSRT<L1?(wa4OAddX_-rg}AI-1eHI
zjt$;t&p%5&n=}3v!GW~)FTtXJHD!i|hO}*PVkD_)JU`!w0P$goJ*MD8JDO9TN^Q0Z
zo&4PXs<yFZ!d7rFEpf$G5u4(^l|a`?-rRYx5G7kPb4L<jXObFJ&j_8(hG{m^JbC@M
z@|NQb^F@yOTQY}wZ(>v;Cyp^MX;!X@l2E>Ycjy9|Yc2Vli*rRhw6;+^MHMMB0vW|`
z;hDrdR}l?ugfi(wxI&m#c969QoxL%wSUMJUS@>QW{REuoz~v><hGJX+JE-_8KPTrG
zago4k@+j8`5j5yTVpa9QY3nxHTW(opQPpSE1Y+d|KKN$)4Kq{I-^9uwbBT<S2`};@
zS6{jJ&f`iC)~z}9LbH&2)l9FBa@jtgl*CeuF>XssZ&)&z@Br$cE}uOWKF?!Dy8xo<
z{L&UJm}Kt;Cyvms*Zb-{{=48)d=W-{M}6N%jPGhyI4#qk$ROMF+e?qFou?I)6I+q=
zRB_|(4e#%NJw6%R+5j`^$k5yIkdQ?>qrrC*wOZ{GbO{*^D^a}}l~CybL{W8gmUBcL
z+lvI(7k=ND=m|q<CfbuQ9`73anPiMoCB?KHg%%f^%!w*ay;p->a#BPL&lEIh#rks4
z?U{q33P@aKcbT6ltQ+s|81I%tcZu9M>%e}9-epHesPg8%;@|(8QvMLb6g#QH1I9eZ
z3yNabKqG8!L^XAu(Vx81tcWn7y|8qt2#PUZ-wXmwJ33O|;frIK2$`1X7DbhDR)f`t
zj<PyFQE0=L8!{#@#~I)>O+!k`bpH@d^7i$v=5l<Fo6m4<H-4LhSbYkS(=(DOps$-N
zaIwKE&vNbD-+d43#}##S1+%L|1f$%q!8`aIx@~^!t^pVub{tO_5qcf7ukyw4@bJIm
z4V&d!<#p&W4e`I0IO~egtRJF<lJr&CO~!q1@jpV6pcu)G#@3LNoXT~X-`YRFCS3Dq
z{<X*E;QZOh_x9t45cts=q8LCYYtP6EjTF!qd%M)xJNUqv6y-wwwQe9N3CJ$?&Teg#
zE?KJ`QnU`^WanOQefoF(&%M*;t;g@@GrV$wbHeH-VbSzuq!yBtX*a$i)3z_t>Y_Ut
z-Cpd0)NW_`s(Ai|_s?B!Qxl1<Uy8t@Zr|Wwyk4_C5HJVa^<Guc<@5AdG&P+Mv|+GH
zX;UTQD$>nc6bfvDpU=&}Hw`*rt`A68mL@xvNL9b+e2qmy0xza3;E)JtX+3&Z(d{mh
z7JgZuX)nnEKg@l3INkXg*+9c0apK8~8|Fj-lQ~G7>?Ug9OkRCtnNpmUo5R}4{doDZ
z7?qQ45VROuHCNKRtD55`wA`9jKr2U?1V3oS$*B(;LH=`2#>>{>a@xKzn<Lacrgo3q
zc=?R&`1|ocPGq>0oy0o&i1v#R<e)kWger4znyEq{D3LoF-e*V6SMbdCsI=Z~sZ_8g
zD9f#ze5r`=Y=Bhg%8B5Y{?b&(lUKkP!|bBS(Mwkdvf=LbYE2bn_<~yt_|FQ|@0mrn
zWfuEyl@LBqmzDoc2F{{KLYV%-!}kTyPiT*0;e)~er!*W%+IR^Z1d!}2uR}Lql>UbQ
z>l(2j(B4D1Yf0)Xi=7Q!L#9<9nO26FA|D}r!cx9e4KwOIqsOgBUCb-$A89hj8JZp}
z<-faQKM`GF8__ZqtH*ZqqqGx<h5a-uCvjV6*?jP<=b<&3Ph%AqwiH?AnI!s;cwJN1
z%QRjzWEcTxB*$XbLbRszXeu~zAgIQzm}Vdk?&aMYoQ>Lk_Y4IW6`R&In6P9f9v}qY
z*-k`_W6_%J+mzg@BJ5HZY~eAVe~&1{M#iRawDWz}zuhhbCxJ03APe$B6v!iyjZ@(e
zMe@Rv!tI;IQY;y}=e`(=x;H<zc<ym}*l!(dR@6AXJ}3YY)(yo>?g(t8Fhp3q^UF)X
zai)L`UG49N3_xkpqX(lWnf`L~9lad~JF6fM2kbZwzqfg;w1CAjN_{@VwvLXUf(yo{
zONg0VW{c}NSQZP4=A^&C?RI0_H@rY`J~$AoAG3Qk<=J5k7x#>lHZ{SFCDH8N@f0S3
znA7UPjWUVk<PinZE<=7g5si640p;b=Y^BzO_O`oPUQ3>2t~73`EUd~uBiY@~8gC3s
zMA>oqgSyMU1Ute|N~<G3kvcr`hjm(ub!<XO8^Z;Y)P|;v%M?*)rBnuq1|a8OxjXTA
z8%O8QRy$sp$EKQ|M@#64*b-OMpUj|y<@YzC_jzB6XhLfJ&4bp}qyJ*Vu1O7Ylz>t7
z68>QmX0t<A2a4hciBrp%XB+2Lm0JWLrc}Bnh|J^GMG$8mq1S^3_Kt#AiETrQ3n1;+
zTYI#+UO0ixtms-~d)9B(WBo#xA?T)EXbq8MEI3mGnr7Dy1B0lv_&*zx$D=27v<eP?
zd!FutZ>;*WHw<oBf4Bm?utLk=wYDrl&%uH7ARW$w3nsrDTfx!=gf#iVEPMjCTm|^P
zaKOLXwcqT2=fJkUZo)~JKNe}*xqUt2B1^Z@<(p~TEM`uBWJwiGh2?X9Bjt0IeU(>b
z(UOKXP+eAAxo)zD`E{tawM*{mv%wjo=CjitdyToI*KSfWs5W&4wOQwrbhc*l`yWND
z`Nz-Pk2(L|j}OAFm&Imf*6m*`#@x$htm)2=AA*^BVv+2{)T}QZg#Qc8oyZxPdsyb4
z+(#?JP+MC!nmQan5_9wZiX9njXeB_P1G73?6Xasmx#`Yh2bXbKixE&6n9yx-G2ZZM
zaXonXXpN82ZfM5kJk$({suUf|a(o2s&F}zcS^5+uxy(MN)){GU!LT%2YNES<xK?ZH
zi?XbW^~A{uP*#mzY-r4lUm&N@Dm^kAZQmA!uBcc|u5AF;AE%B2mH#9uE}&KesND4}
z)&*Pa3Nmw`16K1Eedb(6WGk=iw9AKSgXS{aAC4IR8Nq^n4*tn8J=?Z=zxtPExJ30p
zJ9k%)-&u61db)So^0)io_T9)}PU(w>o6v>^g0IF86`|j5fxkP^#d)R;Pa=h36<SAL
z5T#8$8r*k9;S$LKzQD{$15Vsvv?N)i$)E}NzpRw%xr=E<-Ticcb4DD@W)F0f+<O66
zI+DyRuxy__`!^g3lRTL@ddiIuQ5Y``v+p@*!ea>@Hp182qvg2JrCrP|`}|z>Sc0s0
znf#Hx5mAe=oK=_cqjYMfUQFC8w1KUN+G&smwe;V8-Xuc%-_mDs8pLH(N3ZGfVJj*O
zQR;@^I%JT<fDfmbj6~^N@q)ctj{0#3rJT!XL2q~~?JBXlFt!}*i>*kGbpboMGK5Z(
z8K+*nIZyHxmYz|UDum1|X{3n}%?Zb(rq|-)%WgU^#}&JgG4hN(&8iISmmu5Ya^1dZ
z>my)XlD7noOb{vEU$;F*hxP}|$IH#5%Z*pti;KCdC0i|Fd<j53g^bu+9?6Rgas~h3
z%yD$`8l2fd!LmGx2ErZ|Ut66r{Yr-%Yeky33M*wLlTfHispKuuqG@O<vI?|SQ_9Xh
z0wmx?s9_SKtwI=9F9`cCG8=%ls3@2d6G<*qTZ)vxQR}rxPKdE~Vap7Dj%X<!bNF6f
z9~BUd2sTIAF)uPhMdGMcFZZo+kX)B+@!N|TShp}Li<U_^Q{z8W=*hWizr*My18+KN
zj4y^{M{GuJJ4plo0{@L?%FW@5W+q6>P?A`^o%LG0Uw@C9!mi4>NaZn9G_^E<$v_Q4
z9>)5rNa!bz(>r+<yMeljj4MY@uG<P%ctSiw-)>{g&=%R9$sQmz(hG%X-J_ZDb0S)c
zHYD0Ks4Adpi8hkGP?wQfNhv(U;_uwvKc<eJvs_PMJ{7Zr=pqarz3&w$i^vtorr9bb
zbf(RQ__pDNa0Z$&bDlXz!0|#59UsfOxoP3$)xUX3_}%_eS;O)A;?eVE&m%%F@~h2L
zA<*jg*Z&Ul%J<Q`=mbcz3_5dCBByAcTWw$Y9Pzy^Zha_#l$S>nOTW%~0y=sr9L>rG
zz#SCDJPB&^u+JxzsaZP6^nXr$+kU;8wx)ol{Q=faT!l1faZ};2z8;H)t>o<Sx>Hd_
zr9O(%@+>h=j(yKmGx%4qF24Hjap}LQMzv9HJM7+DysjG?pPMM7ZJ?&Py6l0KO2Nf`
z<D4aTD!d>(lgr_He};-a#Ra&i!eED7`3wm8e}%*e>Y|1s#eef6fVL;&NVah0-5qK3
zAIQx0OmcVlt%^g7Q5e$yH0x*QT4>qY`mfT}pj^gfamgdux|24@=!(i~<>vsW(W1IK
zFV#QZp29*XpwgVVM0|L-*dVZc)=VdmC1oOpVM@1$*QVjr`e#SZ&mNN<;U6b@NAgr?
zz@lE3XHTbeFR)#kIXCt+a}w~xDCG)}O}Q-1^qtjP+_lsj=!{FJes%X~`t@4xK+^o>
zSMT!g-G7{xOCI!PUm+K9S}7-bULKjc4sWm4jF3&G`igXHn5NDST_ifNDkq@+Fjy|u
zc)T+YDr*9;>ZD?8Q*}M84!E6+$uN;)+WX*Zh0%@110W+}E=?0I9*o>*F~w4?RdpJI
z3}CeFH`{w2CNaZcr|>H9kGe2wS=N7eScU;Nw@C_suan1yy@xfh-)bGI%JyDg?i!ig
zWCYl1i;JA*$*RdXW9$@T&Hy^HH^7}PpWfy0j%RKqs!Y3GU6_4Oo9%<npL55dcePUI
z0HTrtJ+vEBd9wB>PABIrd|6S2Rg64&%-%{Enwm7)op@a~6*7gkA5P2ah}TaYZs*?3
znzwX`5rx&!FjloroBofi<!J{vofRLcc9Aug;V(B4xT7rsQIwvsqcm7tf+RsjoIyfx
zM7kbEmHL+6lr_DzqX_&`wuBoY-ojp6-M-=x!jV!=Yj8l9LS5O|6iFQ^gDQhQ2t@3@
zyhAu;<++8RE0(Sixu#DY6^-E=;0nH`Pu&6DgTR;=g#vQQng%-Ma(d4H)Nt7|%AG?j
zn;kX|wpY{?sigQno=Ai~wtXk3lxS`tZN61eTR3>+UQF1vZU}}^!DdA(>O&~v7l`)Q
z@>5NvfFxqkF+^K69b|>+$#_NNg-%$GuWJ2YF_0Mlx8pVy<k_qD;PTv)?9FEA(Huf9
zap5H`ARrH}Ghvybr^NB3B1MAOy9}&nfY;YFZYi3_#eohV#%Pw_QLFH~V$oGgrdTnF
z{lh*gbVMzKEQP&uKbYJIcQ9VIVfojSBHCRdC;6;pvodh;uPD*F;5l#pc!Y#BDFp;3
zg??1`Dx<{mi;m8j_I={;<w3bg$k>J64x&6bFO?#U$hZ4sFVI^-2lc0dRX+Nx8|}62
zZ<TVSoHq^cpN8#ekyRWps=vMhei+XawbsFv|Avv#Ld5yj5RF>oI6Jp}JbNT-CLG=l
zwOT$G1?o&h{$ybm-%FrN=uBHb9=m%L>BuBtNgsH*W9stJ!P|uwraimLCSV1N2$wFH
zBv{bT&NoiuAmL==1E#V_4hRo-*tgRoLdR>ZtD85etC?1}+$*2Gwe?Q_-+ch#)b}Y|
zuldsX^^kLno}#8<&I&#B%I}H*=s)&PHFg*Vz|o`%SoBzOgns`6ZaeQS0-4c*jab0k
zzx3<kA6Lugw#LW6Q&62YaK8e;hk-Pj76z>q+_UZ^b?2(Ho&#|8JJbR%we^1+$qG-n
zP4>QbdY>-uzfhKW3^ele%XjN@-x2+RB2+-&A-cWZA$xduz%h_-aq7ciRw`lJ<v#48
zd*tE$C7M%u&lP5$Pme&j?jN$ZSBjlfP(+M~uFI!pK!d?MmF9Ep8#{;cTH~8?c}Y;$
zt^5V0_&>U0|1>58KbK2|2Mt62Y%LBWcH5oP^hzVf)0<j4GhzW3oo6I+DygvX5=Fg}
za;6Q*+B7DD042bTV_&Or$f$9d0Kk=uVc5&dofWn_6h;A?a<Vw|s5Ngdah}jIj|DGi
ze*j;A2KXNM0Dhl0ps?zPU@p)!IQND5@!hPkdfI$okXS@|=&)4|Ep@r@?6CJ#>2cxs
zWU;|b>#9fNKaaBTZLI&!qZ?y#%mt*&j-D18m@Bt=Y#_=UnRGh_wad$YaArFJ-aud+
z8+*pb2!rl12RGY<9k`{*nlzZ=16XMhWMv$M<~d3kR3nb|$s!}dNBf#Fzzqc`+N2R;
zu}X^5nuTryc}sge@2>#n2zVim%m#TsUd<4>0t%ps!y~0$2f@obuvvq}vtp-Ffrjk6
zJAyScA7xb#;5$bAYdMBfICGvc(->e$^j5{%m|7A~%#U|8Gzk}ioZm)kqfANlL*DHs
z>!(jZx*(YjAzCxLvDM6Ip4dT^Bl|3E@QnIk^FOte0t;2ZT}J115x*l0mS&qoVoZ}X
zZ51X*Ed@8(cj=kY6yAV<(*v+%C#?hzDG}9sc&GL_LSaBZ4Pr53;3B;AajKyI@J*($
znw*Lqnd3D$EuWzy>ZeadqUN*E1R5C;0lm0PKhn?CbQthYuIcfuueXg%UIKK`0CH<I
z36#5mpuRpRMn*<p-cM7lZF5WdF_oUutZPSft_e*pp3jz4eG78pg|d!E_&!bLb6xoO
zv<XDOd*+}fl@xUtkKhh;ksrJT9y}hV%|~ulScrVzNQMg8%&0}QE32qyVN`KftI|aL
zy9WkZ<U)RreE?fbmAG1gJOo`p1_gyF2??)&HvGD=uD*M^C{3fX1^I~7V2JXW6l#Ib
z@rE`JU^EPW$Vy;~Pd*gWbh67+vh|V|EXeA&H4r<V*w);M<1OisPBBXaUc<<U%Jy<o
z%IhK>LqTAoJZYBqPME3z8)%aXRJD}-#bSGdEtjcUxKuWm5Wve)7Ox>L)kp(%u@4p4
z^Yk&N*~Zbu8ms@3%YmWSmTyBJ!kF#dx=>VzQrHLc-++k>kI@f`Z?7s$vk8cQD)=*E
zCIkkwx<d^Z;yLPzzLih|Yp$YHCK_dkI;f6J@9+Vr`P(cm2YaWx2hO{Jmb0x?EousB
z-Aj_lzIaxqVRJ=%O7W_2`b<wW5iFLIa?(o}apPehK`KxryAf${(Z&h)PBF;!)gU$&
zC}x#k$Gtx6Fz{=meyV>&q8|D>nUgzf%{TPKo_lw^6L$ZKcYc1!di-7bs~CC+dhubG
z&t1*a#MYCZ&oiCTp7+)>r-mTVesFu8ltn-F*Y`?L6hXHNb8jS#&*Spe<MS!*ndl?T
zA7-(xqcP&Y$tfvNv0BQ!HSdlZ|5^9sss__?28GT}tDD8XbI6w4kVfs@T#-$PW5KK*
z*W=U0)&~Qj;7jn{o+%_oe&ddmkXNqOm2a%2()YK)qXv^~dsrI=8M+?bF<aifj}BOX
zN5hhtiz3x2L>vG<mMx@JumsHE6FshquZui*k6T5JU<LGneT@*9p=|?RH%r=c%s@KZ
zab4wG8m+%g%|?CPRmF`I)u*&m&WcL00QbEIcW?cD3^J;iTCOCCCusydK6(g?-ory?
zm-FmBD5~K4q%(269y%;6Y~OS8c?1WrUr1sd0bTUyrijh<7k2|pehG@eZbk;Vpn#gr
zqTYp3a%lVr%RUJnL#6=O3b<1dD=@4=qwDMl8!&K|d;DEhJIdH;sa(*9LDp(h(br;W
z6U3@9z4puL!wG909c$IvU+6Y}Z@0-#yGa*2c|~dYv+<k$v%Z|2{t);JaguI8{7eza
zxAfH#vSJ!3=)E>jm3Krgz6=c3V6z0tVPUTcwIDiPoqx=RL@@B;fd$F^_q+t>my~On
zzLk?Bby<crVl^AZXdcqNn5vTG7r_%T-(U$j-H*u$bgMX>@X0{NF0~2?79gDgpE~g~
z)GE1@ldKWak|ly4!`k~=IDBW2&Ro9?<IF_nZ@*vF(0t9>_}#y}T-+5cxr%RInl>U8
zzQShaj<5ht;sR2ZKKn0wdLDz%UQwo2vMr|56_seOLj%otCB`^Md|VGXG-zcW0u4_a
z5{Imcl<AvEx)UdE{%(BHh9vFC^KsVERBnE6yq`mWYJP>~y|s=le8IhH90Hw-Y)Ahx
zK1r6ZO2vjGoD8x=!zzFTE)l|>{@Hlq-sak75N((qbal104bb3i?QH?w2}sL}tAr;q
z{!OAa>l~(*l~ciWO(u(1cx|$60cvZ+XnrFg<}2^>md;~HXw;4a9kEA1wAR_#r+W|R
zc-Y?u4e%Ea)9F?rUD&5ipKaw55vU&Gm?r=*Yd9?m2iFJ|jOw<~eEOaY$f}K2{166S
znbe&hncT-gk;9|WGb=1k9EX>R<Cz}3J{C%-neajAGz=o>#%XgnR`ww%tX5zTG7@_c
zb2&(qSHjateV#EFTk~+x*(;q{01DDPd@x-Got$lo>D_oEyBL{h{3y<&mP9NZx+twf
z@$TFnA@N#4|2|HM>x!Uh2SJ=;H8E>6imWk=y}WwFJdcykELNgAwJWv?3@)S?eau`?
z27>AOyZ6e%#X&%J+z*ZtnF`s$h!_c`_1{?vL-bRM)7j8yRB7FfSg4EtSTvj)NfvFu
z^&#4N>>=IMRJ&T-9<(;J0GCTv3|YMTWuX;h0!_{!Vc9&CUE`CY)#Rm}X*`zb$y>U1
zF=OxXr?RjKoVeopMR)d!cgPxtI&yYq3=?@Uo-wsS*Y1{`Y>KV53bU96sO@T4B$s8y
zwog<%Jh!zstu#qdcrISLQeFx0e|(R#deoy&#d}NooDe$ddhdE)Dw+fA<ZJ*U6d<zC
zUD&c_FrY+in>kPVKMnftH~T$nJ^eF!$MwHi+JdGZqdyp-iqob~S#R;rEly7?k?R${
z_4Yrya&&hZ<odYtFIiArUS7_ll^aY-rz?KsQ_RzEL#OfeCNswM%>@7QCog45N-ivM
zGk+uWl~H(rv~s~&Yo+^bYV&idK6K=^ETzI)mxf#g{q}|`Ha%<V2oYAV{S|OQ@5M}c
zSx}eG#n6leF{9sm89)cwH=h+ul$)5kBwb|N(RPskY)ofTmA$wu=FAY5zO5^$VsLlv
zapg63cOM@7&4si1cdtCUxCk9y6q<T(T=y%x?$1i-5K&Lxkh`RN3!2^AWo}9Xb`eB3
z#QL1jR@4ydLK?X-AS-!&95CV|iY$$+tPTS-edwbN*>Q4`(i*Vf=CWnRiD=)p6En%C
zSj%RZclN=U=>SkKbKcCm{Sydt!^9()QY^G0pYX)Qx;OgV&$l&l8Bt{DZ_=u&z_WRM
zdCR4>hS?S&yrEh;F}!`a%tVVrZ>n@Kr;ZTq>_{5w0pDB=55%=SSt*<ob@66nlQLjv
zPu67beVYXFpyWUA(qMdyd-@`gGHGCtEnA_%Zo{NK7p`+Qdu*NYKdU&STDdHmC$V)g
zR0F_f6B8w|!BpcqO2*((QTYqJV{<sdikAT9E9CcLausCcAryhNe4*pSlCMSR4QrFv
zyvmb>jBt5NXLCe|Cy$}7d;&{eG>H6rOveqpwn)v?^+7W$JUW7=35?Vgi3iR=iUBMw
zYhcLLI)AdfK;jso)vQD6`_iE}p&QkJA!?as5}Yn5W-&DUGYCwhnogNS9%DPJN$6sl
z?5N?tNZj5S{_lY&Fu)M*U3=Dnrqym>U?7-1MmccFNM<R7Sc)x^{ut3z>Ps;E(b509
zJm#*xM+BuQdTetVpGC)4{fpVG>Vb<<&d{VHA0sPr&1f;K8FDm@GOUrRS#z=I_ARx;
zM+yVl7)A>UhUFsbOt{y#4?h&XKn)dq`ADy%($k=DZ6io+no>Kg?_Jq}d*n|wAMjzJ
z3>OZ?#DJzxb(wV)X-#us;XSy&j=!5TRg8-RUU>fqjEe4NqD`QK3aEkq%N3tUmdSXn
zC;A*3mj*_#X50Iwc6ovD*XiV&4}ZJXY=zZq1@!8W&?;6?I+zQ+-HB|cO3$t-Dd;E9
z<3$tX7Gq7poSG{vic^I|8-~|KiMO*slpq(@ExQYY?C*H$@LwdOm57u~%o3Mh*AX?b
z&5(%;|I3G#OqJF-n}ayp7bq#8HV>3xD|d}3)0dnf7LkFyEV>n}eQJm%Q4WPGO;ts2
zFpCn!jdl(U574lf^?D3U4R&WL1@~bvS#2??$SNmOG$Fr)QMPfLznr7&#NgEu7(4n>
zKsj*#`3Z4GF}NqoM@}*=YZ6?6uMP<?r$$MB3_1J=7W<5tJRVClL!P~|O<0bkMN8A4
zmT4u%&@l`G^~)^r$W%o8fPbj`Y{i2zqZ9w~V+<jmLsAP}LJ=F<mCsJH5an)MH9b;U
zloU!{1(nLc{lT6MPf2<g`xaj#>n^iqi#Su^>9U@PG>oLi|KRTJW9u{Q?eVkb6LU1W
zXm0m<#f8*q5cn)*lfVsMW+VMSOFu)mo>yybJUTzTefBaxLyll!$5X$?<_J(3m8|<Z
z65&u!#v)!QMs0m`2<`9t^aF)OsUlrs{)J)jwP}nj$!wA6qr3#hRY&EBPWbs_iPSR7
z%h<CQh%{rf5J(VMM`VhMQfZ?)jPuDhi`CJVJp#&*Sr!u7l}+4t%dM?bI78tTX!EVA
zH-PbJ^D}`W5(=>NeJ2v!{kODf0x0FBadF4)y7ym5WO+H=xbbWC&BXI7$Y~>inva$B
z)k6~>$+iv4&haw*+aUxITKYJ-<taILI~i1*!99wwD?+Cm$J~jwnptnh?=$vl$tm5U
z8p+|>LH^se<_UvAkYoniPAB<jZtjhX+S1J<*k=u@gu@ZI+Z?i_k9T#nE!f#kf`Vde
ziOkZ{QPK29@fa|Hr?p)kzw79^ySVoY!-NJS6RWCxXW1?J3Dr`nS0@pcu3P6*RPpKx
zdR3K~Y|Rg_H-uYrtuI!!jP+eSjY&Wx8}7gr-6v5d_#fM#r*F%{StK*FVq5T)6ZM%>
zZKVe&0E2t>+gC88Yl(IJ?@XIYTcw}O*GVsZAEerCt*mI?>28KKw#9@JB*@N<e$QCd
z%-OKmPxFd~==4}$XU$~|K?;^=?KzTK8MC`VET#$h;%6GK?A8WK%LSELHI{`P;0j`6
zJ2?IA{l#fz8fo48SKm5%2^feA_e^HjvdQxD=~2`Hw=wXrHzlv~k$ogUlnUSZBHL}k
zVw|qPtXVIe?OwA%>mvuoQj=;*zng6DmvQppW3d5?7M`&*#ln26GCdRLp%Cy7O2eNu
zx%B$2Ye~=4p3tHFH3zy=0o0gQ_r*mqNgL+^x9)3ra#4~xVrxb=!^p`=uq$o+AZjQ}
zB6e^usFt2}l$1gf{#`Tohyq_wpoqW?&bB*mP6{Sbl<#3f_qT{z`s~S-=wII>;H}J-
zI{}QC6ET2q=%B{=a}(;pVLdOo0sH<-Ln5a!I?q_r|Dz}Qaq8u<=rPqGDd3=*Q*|lY
zd#cG5Aw%Dx^&0;~rny>EH`Yfd2KMSsm^cMu^0=@Kk)zb#v6WC!>GSh#Kh3CF?6K%%
zsHkI5B3e-qnABXz)dwZ*ksY2SRt_hq#*;&+P{V_Yhf{JW?#Yo9+YD2xQQOOj6-jlS
zazz|6seg(kK;jsuf$Q;Zp1pkyJ7#tkC4&uelCb0w-J_@x(x+1L^0qw2jZuLPWJ?8-
zVD&^|L;^8G4FIU3o(I>L#vGmz1h;cruv|xKR0N1=i!eVD0*RwVd&wbjxLX3ytI5JS
zv<>pdPz^xp){-sSsupt5)yg^F$l>%U?6AoyA7|@hNo?sN)Ps~ot2y4am1(=05Kep$
zs-Vf%O&{6%E0|!;E6bkYoUD5=CrZB6-?4-d37L_F1L>GQe!)!{vk0WdH-`qPNHhb*
z?D_xDrTYGh$lySQyOLx=6fYq!LkRMT9mfhM?OP4f$5QMfM!mL<uA_z*Fs}j?tz9yA
zU?*66D5`Kfc33B$;Z(%*fD|+5uUsM0qKEEZ1zI~83+r7@3l8nO;#filY!#cc;~8AW
zO$L{`Zea$u-jKGQ&bmsPdA8~N&$ylz7X{;pgvgs|z$(#$RPu)|_8kuhB1PKr&X_s_
zpI1hoN<ROzB;Ovc{IUVnzRl<A^@=rf>es_PkJM1<y1!$tZ~fFwy;#%3!?U;ln+G6B
z4L<E=QLEZIIIv*bDq~=jmRbh?Rj-7QKcMZZ#IUuqn<GM6UR@<cM2#%S+mHbsSbHDv
zw8~jI`pW3BJla<_ut>SEqj-yv7LV>;0-pjW<XDUwZI>Dr2B&5%@;m|pdjMm8UuWQ9
zQP83Nh6gE!2pCbBTg_Zdf6HpA^8!G)Xm{&%X2fhtSr}V;dlbR{`M-hSVla@TjX+K*
zW=s=jm<%bQq7SUftAh@aC=JU}ILs3CAHU2rvT!`Yz_c1=?w6I&meq>*13}XP9jLLO
z-Zxs=zx7S7nNyJw&pcy3zN#(Y$<k8L$kE+>)C4RXGRf=;P6KcVqqa@^m)Tt5dzh-S
zlT%aAPf0{~lYCdq__hfpFhHPjZ}6L;?>~)~cfl&nF}LV=$SS7n(^H`1xZ7+3@vp^-
zezCJOx@5KJ92AaY>o_xFCMj$&<mbl&%v6h8_rh@xqwkfo(C)w7DmJGjh-NvVSD63W
z2Irg<>WHiLbe16J{_XvI!QMhuun|jP)S*u;@;tOn3PNVb-VtO2vOEFo$~ZFG@U925
zi7I)=uHFs1!J6}=>c&u<*}kJuKeV>%=#7!5Wmgf`P>kMs*P@3!(iD&AOXP}}_gA6;
zmOYr%KoOXa8ZMtf>7;P_#xXO*kXDVRAbj4r4cCao$hms7lz|}rY&7Xw8{dG9?s=mo
zN^Tti@<F{yY+5pX;b2!xHF=s6*!{gBdZex?3;sUYn0!&~s9U?N>+#asL7Tt*x_RbV
zl<V5+$h?$O*Z`WEH={1i`o5Pr8Zpyg48UhOtC%449c~Jc79+C;$<~guiQ<7*Hg_*C
z4RVNt4tAuz>&PGF(E&+AH)fbckow??`n-z_%>bbK_4RZ=s&n>_V0v8fZx6<-tuJqo
z#S%o77o9QA$nv`C0@gGn8PuEm6*53oz>b?tmk$M;38)e(Y799RUrUnM4(nNTcuAC3
z4cgGPPs$VTY$sj4CU28)N{@#0v$7H=Fww^l^>l<u3d0q<l;$t;qd=~l=I*Q}>^b0Z
z1cZxjCv;d>FQ{H_;UCnIA!nbY`X1OMkggPeXqN;l2&m2-E@m9Q#U~eI=gwScbMj_%
zbnf2OtnWR9QZ{|H_3I@iNaz=%4|-;$m~w7%_t59%j@B;uTIqPEjy))ik*5$+jcA?0
zX2ck=7O}+UWgXtQT;2_$FmH7wV2Zwy4I+Uvf!v8CYjD1*0+r6X$|M(qtYar#%IV3J
z*j!H#_<mD6)9B~+ApA5nxcV=wtkztF3-J%|e6!ZYZtWGQNQR3ib<C@F?hjYlElzDp
zg>qjsROG|?>c-RJ_}#f>oZI^zKgo6Z0?}%yS|A}IF*34Owv?j%rPIA?slcsFITa=~
zfbJb@@izon{A%L<*VO`2+EiBw34DZd&)5bG<2HqLgm{U-24$~%!{kW`p$L<5ZYL{)
zgK!Did1p>4RydkQ3B74^37wRCZ{NW}7b&_U7@1?Qy+a8_{9xBt-&uY@<a#A>=(@pm
z!mb_0dIKV7gFX8MIh}U~oi8C%F5aJ%LSCOhNiYRxympk#5&!G;Q2R&76x#>_Vj^UZ
zk?#>-Ry5<&>*w}WeMj?{|J$6<!TmkRC;eu#O~}lLcQ<AV-j?F{LGSJ6>XYPi!vD^H
z;EOQv%G%b!VU-@BC1riz`b06;AW1F&p3Fdj)b}s_GNR~G#QGuj*diGg@|31aXwzRG
zLu#RAU*>HIdj2v|fw-s-*v(sSza>_=`3Vs&R0bchPc#^8n~mH>j{FiSEquz}`uOHb
z6W+^N0<0b*Kk|+awRRB0LX=}nJ5?u_o3k7AIblakz~su2qipfk8n?DUdAi>DKRwsJ
z0A*2q%zjDK|K|Ewrl`=$on>7=1&dFO8;Chkz?$mXvNX)j#o1_XCD#byWxtaaEZQK6
zImCK&Lu_@D_4TvZTCt7gC5jRyi9CFK<TEhWJBvr^qi@W{sdD*ufpiP7oDa~zwo?h*
zP4zn&<NtLXz#WUHw$nu;u~HPbq<z;~>2KakV>FF!2xI9LW8e)IHc1w&S|hLW-|~Hv
zeQYT|H(0rCcihwxlrn-r6Y5j3k&DsMs9%U$DCQ{Vfkp21um=~==<8sRth^Xxr_jQM
z_l8p=z<+$6zn{bv>$x9>y+7O|S%&i*?Zk4_a#ysi#9*%1Er>z|*)1c}KgZ2U;ibGU
zup;7w8p~9rESS~P)RhmLD0p<H<J{qF87qC%KJcKVyG2dN)!FqOHzOx=ktvC1{R-<?
z@>*o$mcfaj!JKC$$q6?38iH0Xs<s3GOk!h|z)jk!>>i$Jo|e4NlldS~yb!0*)@p7Z
z-jZtiiX;p!u4bRxF{QtqAZjq;@~Uwawq?{*rl^xVpY<QQD9*&Il412pM_2O5HBgAs
z!PLO|MjRPQbl0ula*Y01I_7+=FI28Nh(!~e=;TQr$7Ne1RNVH2^Yb+_jjZQVzLTU>
zxhYST{7}0eyX{PaQqC*&4;f)ROXR6*$6s@oGn}UGWA1Er#I+o^W{Vw+L3McxC^vzg
z*2v1D<H|GmI6@&P8*PVF`c$B!5zAjhGcsuoP5<H#EftF$=up!d<gUF})FbA^EYypO
z{lc)srNus>396BNTOdCmo_i!%yU3WZm~dZ{NizIp+t!c{V(a<-tIHf87VKD;!Te>=
z*sMht+-^|Y-co$U`Q#@@$c0$=$G%=QZBRugOa)<9_^*EbO7+Q0tld&l0h6i44q=?B
zMWb%rMP%{fho(9;MIlmhp3}Fa)+aN?l6~&!4MwNdy_pr1D*^jo&$WjK;DY?2M;jDF
z?5J5>B1jf<%6gs)v(QCMBYy6))&#3I=SkJX#}t=Ajo1=vaI3b=fVg5HI=>S>4yRtw
zBSqs-zBKpVe-smm#<`6O2>njP4+8mD@bALdu;C`#O#dZI>dBy~N;?1c(9}>y9Q!(?
zmPs#KG;M4rR-6<KS~0J;q-Q00X{z05ra#Xc7v175@=R80tqhO)mfH0M$t57V#PISX
zHnTC0O(p{W`GbVA%*p>ZEetQ!h#FmCjDd*MICV1f3>(fD$`-M2dc?vH4UH5O3aUEW
zjxeFKjf*kSLM8=7-n3dQvpZoBi=m_1t{TRTA{rRJ;Z{{Rp@A}MthwXb`i67-Uj4Sb
z=6d=)jc5r2MY+Z{Lls;#PuG1{KRS+{XF&chuSB<7A8%^6rx{XI-~VJnab_~GNbN`m
zC#-wC-7EQz*FSCzDOkuK4q<<**=j!&`keZ7`S|_0?LYdt>;KWUZ}eFqbg1-su7u;W
zduQY~xoa^2yY(E7btQ0gex3IDl>5%M{*7bqKZyVSlMxqD|2QA=r~F>#=K9-`HS=cw
zgiVEmS89V@zC>-1#nIVab}~I`OfFQBJ5=y~zg>;OqLM*Hg$83YwxAl15>1B9Dx)=6
zQp)E#TI|R9P;>lu@F?6m&-zhs0w}HV%Kr*j8*=$YJ#N0HZKEwM$sN|CF)~f26Dg)E
z>ZuX}qe&5V3l1iIDqX0t8E3XExnhb_0ZZ&b%rZJ0MXzaeK?Y!f8Ezt)=}F4!rRY<~
zk(V!GHwo+iXY@vwGeLlMmnb5``j;z1m=Yc0RFKdC_Uua1gHL@|E1qh&t$@0MwXTiM
zzA5t?7S@q%lT9>75xPz_CVfgX@H8y<3UcA_WwIu#SU|WElZintHVzM5_^B~VdWoof
z1<kN^uS|xY(aYy0m)Fe)qwD95{-UYkHWeYT_1f}i^jiJ6-Z@QYTK5V!zjiiO!v^bz
zcMx^sP!ri9#MSBl&RVekw(#i0Rl9-l<WT5-aAj5rZm=7(<xG_6r$_K@i6y;uRy3|;
z$c)}x(`!hg34@jL&gHY5$SHk0lbqwjE6bMJw`DE1{z<z8n&5JA^OafpCtQ-lK5l`Q
zHPdUSi8`Hg2cfIrm@I;7!a8VGVmG!j0^c{yaxXA8p!<}!iT1|jc<w$iy-|z1Dd=5%
zlAuk|Cw*Ftood9!*i{nK98K<&%{1Vo*c*{}=cx6o?u_cOp78k!#MaWkIY11f1mJxe
z*^AKldQJkfB)z&!8aq)x$k=c>#4;nEd~nBH-MeKJtt?MozE!vlX&H4<mW-eOP}I0h
zHcc<JwyTvi1(`X}6$77frJ5Q4@2x})v-^^D8{GQFeG|8OLk8@ZSZqn>*?WS@@pqqN
z{m8`xmS3SNm6nT{F~5}Tx#$^dwAa^>LkTgUswSza@LkN=C4Z|yprMU=@4EK^Fo$Pl
z(DOmdgeZ8F1bp$OA;DO`I6ykk3|&(XSBu4l3SXXbJOekIw4Wh))!9Ao*Jt3)J1mBh
zBE#KJvvphey@2oVXSAZD{iG>b@|J(g6LX2sR<-pt@*B|u%rv;mM8Ex9O)0Twmi$T}
zNdp(i9vhrbg(yHHb3qb9@@lEHSk(51<)-B>v0Y`BEAjgu`g;t6Uv9d1tcfRdn&J`J
zrLKqrY?#8oE7&O|=n3VB&!VSH%B$QPqTlasmV`QejY~ZYI<BcV52m@<-IVE>96ayp
zjZ_s%@O^*G#E5A)Cl!oIOI>jDf$42;-J@w2@yz*uFYcqWAjm7u^C-Ut!s}U-_oa*5
zp<@m)yOLrk1(HD1nonyEVqO$sP@Ni^&DA&#b6ev}nQ$?mGCU}wsfptKcG6_D2!EP!
z@p0{67zk<Cs#>w<RM%*HeB$Ni-@mzoy+32{oy^+OuW#|`^APfi1CItD@RY)`#eP5J
z_kD)2KK}dkmMXs*3O-XNxjvZk6S22E9*OSzO;bODZi5JDW_$iUSMQToH%cE=pTR=_
zTWsp{xoi9FEsM*j>+t0?qek#Edh0El=z23MHA{fs^g4%yveWNh#)9*}%niKT26Haa
zwO-dd&^z{#mAOWds<rg?G~};nFac+(6rGU&Sjk1FSs{VK)SsPA6f_?xrMjlb%@0b*
zjo&dhRbr7-DORU>gZ6!SIjd=DZ8gR(jQ)Z#ZlTusxN&t8_c0@MxH6Yo^rvNok$IJM
z5a|ffvUJF-f7Fz<Oa<q~Zu@lii}G$L5$WFcKpV9gxKbtqTcB@R1!p4Ph9~?>DypV+
z;$#teI2esF32D4xunKMC`zNUHO8EU4O0&_#DyH9PDA=6Jql+%>lK=CP(`)5d6oeh)
zRP<)PsK>R}Wcq+}&xPnT?*(G!Md-rM0e3pdIw~qld_sK4$jnT=7kj{PcTu8Or_$;7
zo{{^uPgM}mkqGJd=yC~-=s?7WZg*7CM2A{L9B=8IDwv1h)g<S;jAi>m0G98>v$^a%
z^6j_h;$7`{_^|OaNT{Pz?;PoVy6QdM|M31*f!!H4@7t&4*p=BCC%Gy008*V-m;mTr
zZW;iF_ODL$vc7xeqD|A<H(B%YQDu$f{tkjo8v81Y$e!xIAgA3++@gZ}c8|`#d&}1I
zf{(2wB+vb`>b6UIh8-^%_H6^&6K{@#wwxjf64`6te{G@kilLX{<14b5Lunz%UvLZZ
zl7<Q+r4N`4WOCJ9?}cCab9jnw8i}GuoQEsEf-B*opx>9ovre@cx`->gC_S*OI*P8K
z!0GXLG{P`D;REelmtXm+K*-ApK}^Y#ZcCNY<zdiZHtJbP+v;&2MRWQ^uO{Ok3q~mN
zaZ_v{-r_QnOxbv9Q*Lhlk{pGTP=<<ypt~`frj72I85Cf6DygPL7|_mwZy2x=F==E`
zk15W<F4r`zkk=#=3*R+866o8$DWr-vq~~e>HIMUcf0+AzT|gLF4K+}4$*E2M_H@Fg
zy1M3=6ED<0%4lqO8Am~lM|BR)f6^K!|2y8Y`yXc2A>kx65tu}k=uuX2CFwSDYGTm`
z95PCj;7a1~;~8r)zHCQl*YO$G-b#1GGZ=&B?O(f|Uggj4R|31QS42+ia}A8htr+j_
z6US&W3*-ya-wAC96!?r;1%$1|ViPLa-PjWL!${4dNKn+iRMlz>tZS%GV$R*q9uEm)
zdkuDy1*@zJj<WY^Haemt%ag{G4dbXv_p|K23Lirm!;$9_U#QX`j$+?5%#mWp(4d7W
zi=)*w1;-avD+i$bKEaj9ZGU1rl^3p?Rf{mw*m6f2YsOGICmgP0xfnrx^mqJ{*>fMb
zpmmg`Bs9?Y4R$gw+ckZC4cbg)lHz>%yoHkwT|8M)syIL<L061`pDnelFGctJ{N!$f
zWFZ_-qSa$5gTXB8)9wS8W@UY<OD|TUHoHx+0U3e|WuU#AHPFPt5^>A*Q(=_V&~ei7
zmrws0SB>Y}G0WX=s7e;wMVr<M%+-`Q!4IUde<?Lxt%qU$&pZ_u!$Q32Hyu=^KAkO$
ztVc+DQXe}+ZxBX;{0Dcr9~4y;C!7Qt8;)%EwnI67E)y`3b4hAhHcx#*Z$wYeO8%ol
z?{ObwAN8LX{-gf?{NHPy2AISdUJs6pZh<=U>zzlz-NGxs{+k;Xjg4Equ4lft^HQM$
zDWNx#I6{F#f8bYG0uzq^R}iJ-@`HfT^N7*y*4xtOH~-rqe{sYdr}6d8N}NG3wnl~k
zf@w6yd`-rW=sv5kMa|YDdYlp-b7zL&Uq|#B6sA3$MA)H$Yik^Q6h%Y=V+R=rCK8cG
zY`Fxr0O6>H3!3QH!yAj#&^MOn#LuHY6<>~MJKt_?aO@T^{P0^^Iu8c!9-m&uxZ*kb
z#SwBQyhPMI=Pmjg%E};8Ir5P;_tL^?nfDWDAF|XrPG4|8kAx`gcqxURAeMOI%hJ+t
zG1PQ)t^{z~P%LB}p8GGAlRF&u37t#F(s+yi9xc**qmke|!^oV7#3oA`38qrY0m{b4
zD{5+V))^l#MP<K|W{*DYot|b0ee4Mx)=vqVHtIV0`7y3kEe`jhj){3)JyDAERx58v
z$!_4y^{$0v#6MjbdVgLy<Qrb)`Mh?eLLO%t`UA1fUi-(E*MXht<>xiRM;QG!r-oS%
z{<bc!$__D}vB%YSCOY)crtBY8<^Vfr#eU^pAXO{1-Rs}~&e7;l_fw!w6gXCmV_*;$
z7uGkcxwERQPc_9MifM3$0%zi)H@ggwQ(Y*Qs8I{Cerpiv)GV%Vt~)yUi8NopiQtl&
zc4q{W#_x8!yNvxyE~$%bU6i+Mb6QaRTAAK}rWiL>JIaCtwKomc->BQL_di%apEzS(
zUx7Vne)uldn^0Ru_0P!h-wM&5gOJoeF2okBK)N4|lpKx9O^=8!A*vx;31}=N>e1-b
zJWMbQ=t?}!ptL%cq*nQJwjQrlKqd=3oA{P5C(i4m%^E8gd&Y1fPpN_<r~0J|G08$v
zP3Z=hLdf1ZuNN$h<{Ip1<_X+3y2X{XO$jEsJ>5;o`ytejkDQR5)+XzueSV;6;zhF+
zd&b`T{b<n!uT-8i6J$=LtPLbiw3ZS{zBv7utAX<qx8$h_OE%Wzm~SjBryYjBkUB>O
zk0kkqxdnv&!N4-zW{sxzwb;YZ%f${NwrWc$X%_&)%1*0)8wwi&Dx@5Fb4`4dj<wbq
zQ_~s_faB|K153*1&1dvA4k>McHTHguA@d-^FFcQ7RP$c&H~9YmL_xd0OlETq4i2c=
zTIe`gtZR1W&=Mh*2{9JqX|V{2zVHhXKiwQ7e%mp2TY~RVN>kMiWlGbnpdTZ3(_mZ;
zSjI8nyFo^;+mX<Z!kN<&)n{BK-gYGe+1P3?3`>A9PTo(>pgmX_^B5yd+>ueJQM1X3
zo{^{nzEIq<Rwxztb)KNV?)Y{F#k)szr}o=__73b$IbpP+MG*(j?H6A&Iha#V8*F9d
z1RIA^c@K>$W$qyccMIxWms6ZvLQicJ2gcC&4K@#$Dq)bSHl>7722&YSj2LUEtP_RV
zIF>3|-h)<xz9?&mqsK)>P?7aw$;=#~oMgi_jbhA@eIVw<@!2V#fAIx}hew<o9kE)i
zaN1y$W_7#Z@bHkjo^rK##r49IV&L}cb3T|)I2t`a`Pt9;habJ<zy9-2`NdEFk`F%m
zn4|d#U;gYDG<L!D;+l!g7z-hd<g8fUUNM^=^6<eKF=qbD|I7cGk3ae@fA9x?!22J3
z#2@|9A9H^3ilOWH^B?}0SFd03;YT0y(RaQ>TetlEfA8<}2mjy?B>erwdw{}+#4rqm
zka_vyL;jQh<T3xV|KLC5zx(I^E&uWl|9dVku9!`0zVn@rm`>-sdi9#?s~gsvEwlNI
zwrvV8p(ZEII259=F<2*jhMl9n))r?i4<9~e96jrG&p3Jn6&L%M#bU{3Jz$<%F0am6
zE*Bgg9&m7QKuU5R#{p8?uqukOtE(#>KYEHWhU=?qKKaEjd2n`0w;fQ);YUx5fz4(`
zV+|?{I4p!fc964?t`h^wVVwePFgfAWh}vGU{wmPfk;4zaN30GoN)xjo?f6&nK<p8u
z9n8KPEJ@w6D~9fh6;(PykWuKx#W^92IA^(D+;DSyLkK}Ogr?#6_=L8d0yjmtP#5y>
zh&6`O(^J|>&CSg<uV266&71Rm(Q|Zk$fHM(*laeudUJ^%Mg6sVo_86ucZF9ji)7d%
zb1@~1vUfH|_rwKyw}_UaE|)R&yC~YbT)O)_w0E1*Zz=-s|J?hi-~9~ZzwNP)*h4(&
zVn&)+efPe0*E7ANSbO&!2w%7qLFz6ec(>_nyDhgjw=8a#Y&TosrzVFnhS}^6`SYV^
zy;`%lS>mTM2F*lr9$DPpvRo|LZa3^&ioF`vNJA!(P+Qi^6(J-xn+@aWnM_)qK7Alf
zlT*wOW*p3>Oqzz;Ijj*yyZ1ePw`RL($f?E}S?@Gr^jcRH^ZA@7kDqdTdrgcU^2Ilt
z+?^lf{SoZl$y0z<8LK@yt?6&Sz;C`r$0f#iqKA|Ul>uZ=NM&lI)ID8ZLBn$tUU1aN
z6fp^9pwJg&qEKO+N0~q(vhCKyy2Uz8-Bc_$;vuP$r>0|!SEw{%>Itfzpqe>G3#Upe
z0T@E%gyW?o!6BuQag6l6r|SlsbtOH*u~=)G#VvJJQ8`H*30};JWeOq9bqbL<2Eype
z6h%^ChGAeFM}qfw9~g&$=yw^)a8NnwSxsd%wN<nP)X<}Ij~TZ(RnDUp6mt_Jo0RBN
zU=stkqi54+#+V5SF?-%!!Ey;TBBrXGWo8<V5002mCQO{huW#_<iq?*t9YwzP;+XG$
z@PH?04QB@x)5?%zj~_il?-_j*AVFnvl67h_Ii#JoRE+>GAtkiYrNNO0bC<y4y%%1h
z(sJF3EtQfN5o5#;omiLvuC~~!LK*R=^TkO)S%<bY2~*Nv3`RSY&6rxFb1uRphq09;
zSZEpBYGc?%QOfP8QA+9#S7(BjGpJR@7{#pc5v!zGA;}*oL(GY!z^QL|Uj7Zox3oLC
z&l^$qI5_vl6k1C`6`=Ns?d_V&&%U5OXmM>tTQ_%6K}o*OIkVqfjIpJ#OQ>8HD5Y^)
zVPc?8(8dwn^|(W!_l6h<NxavzF{Dvk-fd+e3kS|>@ZPf?I-(Egl&OuSnN8V@9>##R
zntt1p?2^gB9ODEwx7LV%j>;&bnapQgFK(DlC1X*gj8_98L+A(kL393U!8mBn4v%;^
zJtJLR^Wx};=DY9nlbguTPY?OU&;AlsHT?eX{(T%RuYUe>QV861SB&E*b9?IuW6$mN
z$T;?#J$S^Er;oW_T=T#CFaG!Z_7DC&{=q-^2Ym9`FZkKd|C04~#nt&G%f*T>KmUr~
z_`z@T_y6IPXCJ(WGM4L`8`^e?)()i&F?a?aQOffC*$4cu|5p$B!4H0mKmON$$T)VK
zoE}rTip63{2#Kny2{H2e^((Hfu1G16$#h*;xI&UG=&Zw4lG(Rg#Edn>aF+`TJ9s+q
z>eXu=J$lM=xnR3pF`LyKpG>h%tT9j!qWFOI{mAE^eZh9K<?QSX1zcTTqJZ=B3!JSP
zIw>+%D~Y_ar$;0ekh(6RZG~|)eu(6dP}!iZ!s?oVM6@1LY3y*#<}Vh+%Ug~=`fch5
z&*|KRA>ZQ(WrRoZ&T$`=QU#gVbD~mWr;3&LxkPJG@vh5g&KN^o*Bl-l^Z4;&j*gD_
z=%a>|5|19A^5*<0pMLroLmznd>;(shQ{KFJ&Ch=Jb5^TWc^oy*o;~BYe(Se5zqsJ~
zc7^xjzTM7Q3d{SW)fhu5z&8xTfKm#hwFJsH^}c}H30HR4a2HqA)OA6d%Bpb><@+Yl
zME-{J^-Vm!`~Ur}fa5-z_qRE`9~q6g`+n*!%UbC>e?}-ixcdqG+auf^nObp&&hBhL
zhH<3pI+n{No6Uy4?+GCmv{ro6j5fmiQD%>9Z?`=l?=2>e4;wD8uDHDwgzUbt_U`?Q
z|KT7paLZK@Dh*^3$H#|g8;RMYOr~vG=JPpi(-d)sm?wwe8TyXRW=)Ke!sE2XD2p}~
zIV+47QP%0nDa-W@+SR{uZg+wrrS@_ic3Acf5jA8?9wDspn+r_ZV6^1F1`1M?Wo#>z
zMJkAmm_qAab(eSlmm#OUz$6teDW&i+R$-mOxQsH1aU2NBkhP_48WwAh_c9tD)kw-~
zqRNmxIz<!)WgAqPTqr>g#DfsIIbuL!x_rV95v>Pwac59Ya$Suz<(j*aCoCV{k1_@f
zk>rEipM&r+y&vy!Y2|@7O3_-4G3CQ%oK;jN<8;P_L{=Wsh)y1-1;q~uhL~CSkxh(j
zV`S|k>yYqa@4zExwBR^0;@8wx)~QL`a8ym0HZ_$>Fm6eH#X%i-c&K^)pyhWzKIY}q
z3G<1gwg#OtK14<z@m{jIwT76yaNnGxoitQcEeS%!n{<ar3%#UJJc`m-DJ{k4Zfz*f
zXG#)F8oU><XJ^`9{0g&DqzKNsQs@~#Nv64W;)z&kvT}lSD#`fO#t9WgsnWRUJ?g*r
zxh#gr+9;9|17b}=r$|-KsUi__pwFX>W0fNA5$S)ch20&XzIWidn=r{qR0^FHDhiLu
zrp)d61@^O+_HafupP_1b=yQ@ABkA<kW}jk!onq1`vPqbbnCMKCM#lA$Y%@8{7<^e-
zOK5LZAze%I=u{I;MAtP+YpS}&I!Prk87ghWU$5)w7FVdGP?craZmDXKoGW9CL|zj~
z_%L#Odcvna|0R!3A22^SV!c=rL!fP1d_ORB8A4#*G#pK)95gM5lZw}`Uh(k3bJW=(
zH_`C=%^UvwfBX|tFq}+|Xy?at{gOJa8Ane*o~Ej@ynOF{e)@}FvfgZ{+BdWl8Rz`b
zzxp5e<da|U^1T=Qy?^iz_~g@H@a0!uG7g?sUwzHZ^({aB=}-B`|KuO@kN*9CpX0NK
z^!-5BkJ2T$mex2Dmf!=MO-J94JbU^AYc*ZJ<?`y9oDI*Pzu>+1Kj6`$M;N16EEc?a
z^@@v&bAI~MpYrwBU$I`V*rbi9gOtI0FA6y&o?M&FmYgz8(?CvyqN4lit1o!(<x7-~
zoShwVa4<tDv8s_0DA6<xAp{o7C7*oyDb_ijK6}RKN6s%VOL3nVeMbtKap<|ayg=p5
zgOg*X)t1TBp}L4qnJ{?bm{1|(l)_Sxqb4?%TxpVx)G;&Lbol;r)&mSL)=WSAKGDrc
zT1c*iguzY@k;>+u%4Q-wvVET`zGf*AQz)Hpq8~bxu34^@EEbD`NEzm{IUjxaF)v@d
z$I;<K&Q2|}*_74#J#KE#`Rw!0c=P&_NjqhJFfBf1J)eB?OM>^*O~d=|zsJj$FHlNR
z6_2OVjsd$Ut2=tdF!Zd}YZi-Jx~?N-Nx(2=+#uI9yfwldhMvpIOWLNvS;u@fD;dD|
zcxDA*lb?D|9sKq*?yoHDz6k|<cTD@YIhLC>K7@TiwvUig@~$QNN={%%xm>?r9b;>)
zaaJ~Oqh-@pC}Ik1yN>IdYtCPvOBQJeQe+rKRg2=asaxt=C<nF_4BapwDHD;=cdXaf
zeE!APyncPba<MK}M0f93YeVp{6p;&>x<R2CgU1-n)2B~384rnhBoVO2QCYE4an=^@
zQYOZcZMVkWDz=p<(4DbZUEy3KVgMys(<dj#th+O^eiyNmjBc$;$^;p)o`>BDRY0XP
zI&{S0mbkfQQVB0E`Ao=}P=r2e4=R+Amr{@=Z&mzGgwEr=@SkH0V*7y<mP%!eN@$g+
znoKi+s?X%!Gi*D8ZJ0H4Dm&17NlYMSPIMMuOGH^|Lqu0|l&is1_Yg?|B(gU9=Om_t
z&!N~x$$cF2uy5`}?QicyOevF+-z%o2c}V*TOD0benYe47XKl-ITj7eQsE<8kzafR5
zJoJqH7MnaqMU2Xfv1idGHa@fRiN%oVBKY!=vMTE@f?gw(F>N#kP30QqP0PV-#=&gH
zta5CZH}u;zgpQLo@!e;qeCOpMA3s0f{fCZ8y=x#N<Cqx7gdY<jB4f-fRwOZw934!U
zPA8aRYZYQDjtqMLe5N22pddVxVk#W*G3@WXoFdW7wT(VttmK2|{i<D%NTqO<Ory$;
zg;gRhK_yTDjEurlQ4%NA-5QpPwT@Ev)({aPp%f8?SPmy3pmM+@g^LiAqE0=vwv0v*
zwIXCONd6|(`(Jf@JMjdGpmrBJqszmk)!jF;NnxYH@1kOmx%TB(uQ)n7pglUEnYQvM
z2XNH{lQg<iu^|Rh9I+{qbfj;@<gxCy9FGIDaYY($iN_6NtY}q5@{u%p{LmA$Ci{Rj
zitcuWolG#ZDXMZ9Ckh&C6n+@!yMgQ@F-Ou|lSYqIiYjLO&4Na0f=ejp7y+v-I%j5+
zIkl-+-7J}!hMb{pCiLULm;>{f<F|j~gw3WSjw9W8&FR@=nr6b{%~za0$o!Kh2c!?r
z_`m(=Gd}&(KO%qsx0%e3n6>ZYwkLF-{3Ts~i&7pPA{Xbcu}Yrle!a#Ifp#(_rjE<&
zFS)pRMLV7I{N;Q6=I{PJUcY+9#hW+uUB~OsKjpvtKm39}`tSaq{MO(5dwlmde}|I?
z52<uRjbmaP=F>UX!<x&>ueiOrkj-8ja&}lZrD-S3jvp|Y9nwx_IA@t39P{Gk2ZRv$
z5C7@Nr=NbxkACzct}f51YVl#azP{$oo7aNo^c~7%4vywbClk8D4GH6hK}WReQCZ_`
zV(3;hZH-R}Wh;zUD5F^{m;C52{*0@eOCCRZ#_7WcoZnpY`udVChGK~m$tiMu^P0u=
z8{U8Ugj3h@u^E^Sed*jaX3WHXB&39GTTx!6j7tg|oe-R}##NaVR=D-&s8244tf^l9
z4r6<aH!YcnWyHpztOX5GIaDqR(PCCA#;GNTkW9DP2=N?9Ce!z8ZWdRviJFRo=^@V^
zy~hVHzRS_<glRKn42hGIV<xkf*});Mo-vtx$#%1)ty(5k&x4awF3(>PqvgSaGv0gg
zl;guWR~HS=2#rD)%7dnO%Sr!8Op0zCxLw_FySkzG1E5GgVXehyif@yQnf4z(ad~k?
zTTeKc9WrUA!s`Q4sq7Mkv<^hX+%WANiQhq+_G?alJ8j~A5%sR1VkiFi+Z;dp<xj=5
zx6G4wW$8DGI&!*O&~r){W5gPyuBq#mrmm!!ET}*xG5UecX2aF>729z`GjU+1kfEt-
z4i64EIy&a~@QnHNfTnJ-Ml3S45_=^-4)ohK7gtwYTwHRqSh5}ce!!`-SVU<o-UlJl
zua~&C!DvG!l5(J$Sd5zxVkINWB<VmYiQ3xPNUAJZiK%b59g3ZUmW^E-YE$8?#Ttt(
zo8RH#A+CA&&fZb(fgHtrRqZ?xm5i;j&}|HB7%pJ@nksG}CWg`Deb1Nz7zY-zmPt+8
zVIoRl4MMfanHU1Ywx{20*qJ2@y*o>;XBO%~<1(!YOikd(L23=xn?%<KQVzU)@{~_5
zF1cB+S%(NQa@cwf+JWQQ$Rq{uYseL;Zt<!{WrxBNaAX2Il$ESy(3X?j_bHZ5TgLsu
zTf5^3k^(w=jFxySWeru;QW;BAIVMfT!Sn!ISu#SM*ld<8Z!Xv_7i^YG^f*v!C2Cxw
z2~kopLe6w4aov0Rknst|#2%e1G47=D926lx7HBI=Th&bKmIw1w>Ux4g6GGtYuRh19
zj^j4-_{8u#?;Y_EfAEY)CoKo9LHUvGa960PoS`4ksYH%y%k5%|PYP2_nam$ytC|=y
zAxT4`tdh~Oif9xmj&iOP&^T13P=04rt;p%lOgN2OadoH~jB`X3V@QM?QQFW>CXh43
zcDuLL5_=j@QPyx$67fb#QluGCG3(vBwL=xlO{a|AqyG_U2yGTW(;S1Oo5TolB*zVv
zT{CS@QR)z%W(41VYjN1$0I`!UHRjzivZodAUa((al9H^&IF5{q6>q-yis_RFsG}M7
z;7H(24N2#+AjxB{l6<jI;xCrT%#GnlYs|KzsWda^&?%8a6z)b&_+gM6ydN+Kan=q4
z)_bZ+OKvJiQS1baW-JJt(lDJ**)BJPVZ>sHW1wGE)bp9-(P|?}1Q{kJ$8Wt_F&-Q-
znM{Z&(fbi?4Iz6z`sf8IMo}g%7Tnz4a(r^adh-SAtJj=8*8H>I`4FuVfB5ModUMIB
z!L@TH2M_qYzyFB;*T4LKu->e2I<bsPK+dOA3g0go@<`LRj3JS+EZ(g7`pp|=(*s^S
zd&YM@{D{TvHS5JKDF)tr_J-@P{_p(xfBz>ueg7pdzw<FqKYAZ~Fhz|PRwEa$zhc-f
zXxa(U2gYto<1|km9`Wqq2}koOP3_9!DcxiUf%Rt1IP^R?JLTz<N1UCVad>#hWYW_2
z9bbL*C2!7OvtF;bxVYd)Km0R3`{hpw)^Kz*=gE^NKwz_3h(MyQP*hmJ8H2Hge!Jnv
z|KpGN{1=}?)I5FkoXw)+_IATCBvP<w64`q;i%TwF{Q^hF$1jeV>5-Ghp_6Cy9!-T+
z8chO~MXqm3x1Exhqb605^+b#Vb_Az9*T4J=+Uk(@d4tb}Xo_N%Vgo=RD_zt~1*zL3
zchU&OnAvW&V%RLRf--{LqP1c+nepP;3%>jD_jvl`DOKGNg5S&K>#D{F&wPHsbUtOZ
zS~3o!6c=s7bUI}iw#;T?Zo0F`xIaqVKhyidNqBO@IMVe!T?r)LnX3-NKuW!g4)=Vz
zJ49_5Ms9C!S+6&op4|7pyGKM*cN_8UzjIM^{yL$>H;rB2Qsn*}@#_G7@|Qmq)!kd~
zh5fa=f2h1G?Bv`X9LzX7JLBl^2%{a|8gvw@f+@<(FnUsyBKhIj1E!P79%*QrhH29>
zX(vo4bJ|7{Aase$N-+q+Jq(QFKuQs#R2jS7(PYHOP42T4qgV~~1HSKxMhQ<Tdit?r
z9Co4ef#Bm_U2C+avW|AzFli@D+6k#{vBsc$K?G3@y2EHOe>K+ZH{Hp!`PGw~)jo#r
z-i&8|T9pElXYAMb%@Ru{8$(dyg{I4ER4F<rmMw{$3RV!n6lD@Hjzjsr<^JJL6&s^i
zGqsb3cABX-2J#l26>Z^RZ+y?}<e1}mL&%=nRmbZ=a~m?Veq=F39!?^YMp8OV+Q?{5
z)i_;~O#|AIv1DaQ+7h%TQJ4EGLAJ8*DNWW6;~L>5E74~)TA~e=HQ+3D(_poR5D0z8
z;^vy=dPU!N^uxdy2K>+w#(_AF)EX&%qcm`RjD(c%8TypjL>X`I(kD|<@D?Dwj9f*m
z;;d!bPB@$&(AG7TagY?t^@4GbhJ(E3`9sIk<BAt2HQ#;jh$m+i(>i1HKnNo~?9|I(
zltQe4G|Ir}C0WH*6_d#vSGB~DK`G&OI$M;`MYNT~yAM=xvM~2VT#?FJVkjfn;L%#)
zT!k+DJYy_r=w&o+9aY^5vK|64c!+_lf|!?N$gvpAb9Xc56y^Pt;_m+4E%bv}Gy}3;
zs1(a2Kt313uGm))#VE!4o+{T=xnjHBkiTWq`mGDQJD!j#u<UN3mZvi*i5!c?mR~7_
zGnSA%H|JMeefc%@$pO>4K~D@(C!)?|6b>;SRzZ`J!f8j71gbpL8aIs0jAdGDl=oOI
zhJvId-JSy950D}m7`l!zc&e1-0>uVWM7Is5s3}2WqD+beJF;2#f?{OYE>|K+R$!dN
zC_~l~4K<lexO{!idb^=&WVGy@qjH+%;)dJhiic-sJbdUNE3U6^$S4kvj=4O4&F1x|
zJbdvU|MBmC%=^!t@Uz>I&lU^j$0sbjVml-b56%dKm+a3NX<H`(5v|ZB<NFP(F!1uD
zkGWo~SZ%iqV_>yhu(*7~>Cq7{pFQRM4?p0l>sa3|h?^~!ug|$YKj+7P_9q;_c*J|Z
z`5hj=cu75IRLD#kgUTc0IH1x<+fHx{RBFJPNQ~p&vn<BQ?d>h+Z{Bczalzr?AunFM
z&(Yy&K|T_7HDx|K;^U9L4=VG^U;cviazmvP(`n1W!F->PAot9jk`F&HX&SsA`LjR%
z6aMngf6BOxoE$yi;n5Q=zPcu7M^1+1BU&f6iyO|r_ypD8ayEAynJsGD(`6%(NaX?Z
zMRBH#5hAU#Vl-PKz{q*;QB*lg3;|tdRJY>dC;ua-7@FtbW1A|HQjAz23W+_1yMkaB
zN}hzopb)yjY9%9l<6!6pR*MzU2im&j>EkDS`2Ghxee#&eq!I8wNpQGxmVOu+M}x|d
z6hvGy#0WXiwhhmoJf&@G9z1-&;lY8}F65%{R6Al=X~5FkNSM^38fzRW3AKSrV|0G2
zpxOy8WQ~*2#*RC--EOhY7S-b0#pJt)TOrE4_3!^CzTNK6x9wEE^=kGj$Lm*@qN4kz
z!tUOoV6WUK%L1KpU^<;Korc{d7jMNfR+CjuyuI4G<>dH;gV|h0_KH1$-dRUg6{HXL
zO+kv%Oi2!Bk{sCDQdhR*3g_alC`FxCclVu)`o!NW#6Uj`ELOK{w;M(;YS|FNUeIQ=
zrgAm2aY_skg%HK1ppqs=15xF|A4KjgtFV>bZ*1?JG42A*@2kak9~oj46WDR2Q3{M>
zRJz#Im#-}6UChDDy|L>bvd}R^AQ4M~gS;-qyCzDKCq@~4Pg>fk#?}G4Vke?9It2_P
z2aRD2j&(P%8V!9)bSX0?4+-WW&{UennZlPF(8i&1O;V0bMOF@F8cbF11L9GpSf3!{
zI^)pA99JRN&lDS}UCdrpTdY!q<Qd0~)#`@p>ua{%mSGr(2)ZiiHi?uqK2aL*L?07A
zix9$R`N;Ryg8kJhXtlAL#yT43m^3w$c1GPy_9%HbbadU4Fb=5fd3Xd*P7TjaDxRJ;
zJUVTdG#O`P?5K4r)HQ`FO@y{`pXZ$Mov6EARf)Zi7J=JtebHJMfGd~tlJMgoZVg7L
z4|mFZp=g9K7G8he$Bv2W{qFuT#oEbOyThw?iz6w>7&QT^m>+8;oZDS9vHKE%`R^zK
zf)hlP78q1i@@Y>s+ks?};_I3-CVFfbu(`)k5mu{zyM>)JJ_SC{cLdn`4S7<?n?wnE
z5oGM6lOpzk%g?@Ic0A{3HpjNL@as$xm^w+2b52USG;OGp#*wM~i1r=x#$g%p{Yb*2
zD=nKNFlkyg<5==#6@!tRFLga*k`0g;#U-W@zZ&B-z8_E#ai>Pe<VG9Dt^;72!8SEn
zE39>pgkahAorGU^UCC8+<fJ(`6t(Q-^)+XwXPiBFfFC^<S67^!oiU%!=x@)t`QlTK
zpTFeqy*T30am4-X6@U8UKjPe5&Yr%{ci#IBhtmT-`{E0_v7;a5h9AbBOq7nC$XBo5
zNFh7)^h3ZY1u?L=yx=F-mrM^2c=F*#JbnCtq3hVJmTXoFj1OFY`UTgYd`jK4oIHKO
z!>7+UIX$JBwrHbhrwxkC_0>7ucE$A7Yo-S$G({EL_dTz^{+i3HE2h&a4^EGnO<N|d
z*fMN8Q6R+>S+CbD7B{SxOJ2Tw$?<_>yIrFTF@Lvt3JN0|w$7Q&a>2j+!$0KDe)waC
zA@bt+2YlzF-+*ZGJ_~Ch1~ie?;)>T_{F3hWoJR+i_fHSF`uSfnn#h-prFr>+##Z7v
zRB8%wsh5So>{2CCB*!u)N^!sNjD3fVFkavC)iCnRHQ2Kk4Avt0?1aj(Y{)zH_zqPn
z8*LV!u+3&oN)eqi-FC~^4^+0|;n^cT{_ta7Jb%u7J`=?^K|HXGVhp0}+N>Ac++1>f
zb;0H3YkcRiR^y!IH-Gatc=qfW)9IAs<6}}n2=U**L)#%3<h(b-JvljJGMVlD@cO=!
zJkb>KVJwA!ETFmwrnHiAjPPCVE8Npw!FqRn@3s*oGtak*x825Z|KrycclZDN+a8OX
zO(_Ka&4;@8r!|VcpUQYJr~TTPi|7gsS{bwzFSE)yCRJM|9P(OmxWCUAOoTWVz*maH
zkfJ>M2HLh^KAW&wH#ChyuYp`RPzsDr;)btv85L<6!-YWKZ@Id-<of1@VH^lv-nTR|
zx>zyP48s9ACY&)$C$0F4s{0wY;uereg5oOS&(06}UpZHz7gvP_07NaJMQWGkp~&8o
z#y}2{+F7(~=u2FkDb=!0rTN;O=bXt|7q7Bh&=zSrL^&z<VN8*oVOJuTB%VMsnR0N_
zFkb~Om$xK58G|#1u)SsKENSXkZXJuL2`Mqg$Pfd47@2h=GpjhL42Nw^U5{8RW8Z)!
z(vVe+)-CPy5LZd`l`>)+EylJAr8B{A8TwJ4wcy!~TejORK4(<rc=qfG));)>vAnrq
zwOG(~YrG%HS&*MGXJVG=!RSLNE;1oz2C-h-Hy-*PTB?*nmG~-EVm_Ou=5#h=-nPt|
z3C21W+mY3#W7}=$#tn*|+GY+WnvY*jdH>Av<jC>(xZ-%GF?!si$ku3drO6?agG}eF
z%_UFPv*|Wi=V;oNcG60LR)k1Ko3ch*;moB(3W1>?s4C&-X{8y%05PG<vk=A+vK0Ay
zV{b%B8%7zM8Y}y^HilgRXmlbINIsTGMG1kBX{3lWOexA`(@+KN7vy^{5XwODw^T}(
zkd##FBuKjObrm?R#0*;XIBjv+FkD>{$8Sn<`Sr)QN9-t3P~`#NZH~L*PJZcKyt&AK
ziJ+4#YTeC>o6o-F?Cb${K0#F#7==<!zI#@Lq!_irYDJ~Nra&IIRFxvDh))5mVH`bR
zNKqc*s&W!7QaUnQ)lzZCi0^h8kZRw-HS-y+u81KrY_=F}nN8=Yf=iFXl3}wE612*w
zszGTXYR(VlJb(Fus%Z!*Vi7BdU0CpJK4<8~D6*+)&K^CY>j$oGZa6(YB>Dq3T~Bp&
z!Sp!u-ekt)`%gKpEB?igf613W`Bxmh_zsUAK4CUJ;&O4z=U;!xX6VqmBBc?n5-Gvz
z_L`g(AxI~t6-7~Ga9ZQm8~mUBn8l=}Iy+>3G-on<K;I8+Z&&Csvt6(G`r^;=Km2nJ
zkB@o&{(Bsr9CLbl%JJb5v&nQ{$o2gjLJa%OcNhkAPTX8v@ROnE$<td7k4|Oe<2^AI
z!f^Re$}^kQ^xKN5nhS42se6_;nYwP7PG;QP-0;W$>R<86r=KyI&H3JMe$3IqDMr_%
zEWUAu0BOt3%@yabzr?Q>5Il1`<H$_7{^F9Rh4^qnj**Z@2B0!dMoYV5E+;e+98Qcy
z@hHmSRO~g@s}<vR!$ccif3{_D`jE#5M+7}5C{03Pa~9#p4$aA_kY!6@ljXb&UAONn
z$6-J#&C&drk3aZ`_g}u``0!Au2uaBN+9>=GxW2yO)tfiGe)Edg=U?;rXP@%vC!a9(
z0}6Qj_z~~F_kx!%UUGbVj4_7G%PaAjGNR7PyDT&)F7i@1X|0(|rdTKWlU*m~&6k%l
z+Uk0VD@qwrvE_{0p-hK|RL(I!IFQJrR6G!Q8(DkzPo?&!hMn2)jvT#v&96i4-u>^t
z?I8{(Z{tn*-v7V*`CYFY$*eY8zJ7hqrW?3ftT;S4U^;E7T}9(8O;t&uQ5H<AO(9qB
z6u9i~le9;A-$o;iDb^B|qKO*sXDnA!8dqZ!pk;(-w9vwhDp}m6fXGS4NkczyeRa+G
z`8ne_?$^euwp6aBay42B3NF5E9h=pP!P;_5CrF+c2SSuF*Eo)(oT!@?)Hftg>`!Ak
zzY2C_E-@tz#9@mXdu)mrt5LOMI+>tzqVGD^+n(e-!H)!M_D%TS-!Mhl4`M(obuVr!
zl|MynV3Z-n#CpBN+QhV}c>l$W(RaK)A4qGDkDlCT#?GL%<>lESkz3YXVDO0u+$LCc
z;@#?UVyZ_bb!Or+(<b4XimU>ej+j<d>lJl*UYv7O6X9Owl!!jDTCVB4ftZAVpDQby
zzfz=L^JVBcJecE*p{Xs*YMM&Z4Fj9)nqjrUC-A8x%EaCCLgBVT+;Qd-RGGcmx>B0j
zTBef;lS#{DGGW@b=#&WKz;d}@vDmQn!h2UlY8#oinS+Vq>`?Rl$Cmfcrku@Mj;9SZ
zprc&plrv)rDC;oV;wpnd5*E7cmcHxpK5%?`Mmw2;Qg|<MZaWG|M%K#UWlAA-Hi;O5
zh%SLR%DtzxlI+`pLTe+oBgQ&1;<B*aZm3I|g{|EFfmY>tOum@(mC-Ru7rwU`5^HNk
zLCYPzL}vxuB{vD$9t$#PigH>C?4S~=Ac#Avfz=s3_A*Txwk$vYgwVa~3;9<c-?Fev
zfG%TL1<IgQyi@s>6Q8ocj&hc?oES7QD0B)ezrN)5@oOe0$260P#F<$6g)@pIZKL?a
z_=p}lDjhjGo}-2h<KR(R;RlaGp!2~;#xbC>#yU&zBerf(F^Ow;61n-<1uy|@05_tY
z#n_r`BzLcFMcF!bJ;rFfAK5M!IHRewJY+FMj1jY9tu%fZ#TZeQ1yng~CBI7yqJgQ}
z38xPq^7_?lZk8LSlM_^4vRG`fX2IkD-kTlMet61*iRMp!{sn*h^(U;Cw;Vrw%!|iQ
z(5~YA`kIUDOZw7byDZN1V~~g!K-6O?fw8#CGL4C62`rK2#Rc8%ExvB34v(mg&KP`P
zz3DL9j$s(Mczwy$?Goi2lj)4Jvj;qT@|5RKo^m*!(=-*kAEvU5eQGCYkjj~h^H*Hm
zoKx2|ZQD@S6-`@XOhrnGYRHVEW719-wJ!vE8B+Rj#9BvAia-6+AMxpDpQ5$l;~)GM
z^MhlYZP4066cV&bi4ay?Ui^~l>uZwt!Ywm~li2}pK0C+tpfIR}c*8kKj)*akqKpD^
z6g~#3AWBi~$CN5%3^4QqtHmvTv!<;pYARNr|Aey_?@>K`k5M-}JiCm~D0q9?kHoZA
zs4TpYZo4jVV40!onbs3Nc<%!~c<%$wPEMH=^*9QA3XFs2=4Qz+fB8$k`0^`mm$%$3
zuDHH@&2n+cX1QUr8Hi!z_kQ>HB+jQ2u9F~j8CymN;_sge68W|`K@cSwYYc;oF}B-|
zuG_F!+_2uP@jlQGVk&-oe9Vhy&v|fq#-wdIJUF1)jl8)VnZ7%My(`ERcLaC&n)~)E
zzs~aH{^$O#io4yI@K-<ktHZ^8$tQaU+f7f1S1gt*&Mz)Gn9rC_TAHe+t;Oo8ZX_bi
zT8p!8FGARjh?FW3SUUuzn8GXAD|B~e%(^rrO69#6GZ%C(rBuAVqIjTr;pe3&-e^Nt
zxRTyWS_Xx3V2kjhYN=g=#^J}rdesq2(-os2B{2@9*c0POh$G`Tf-+1dGv@POJyj~D
zGUnbHUhgKO8Axb7z_`Z50Y$)&<i7V}zZ1MC`6M630%Q!Hx)R?qrId_Yi$S(2aca3d
zgE`%y!Ny22I1XD(j#%TEwU(DpDxA&KH=Z}wk%gD0E~$i0il?n%L1t}X9WvXP@mk}P
zrq7xoX*PqW3W+LYTp!RzNccu+YLlp}qNy#jNrfJTM{EEG*hVwjT0|^Lk*sA*8J}Uy
zf>L&y4P8z|?~9Ri&oGV*ek6uSOq!GmUyggvS-FL!wG$MqahA5kLbgrIbUKl>B4ttt
ztZr|_?=*_)G^B_%3ZoL0iA-(ebOMh~9M2z4`0&924<;28V=-f<8$6>L+2x93G^UDJ
z1KA|9wxr<Mc3ZlkXEL3XgeYkYRT0CGg=Ubmm=KeN0w4<Md&cQWEMg>U*OI*O;`i%H
z5wWElOYsc0+8GF!`XvdjY_ugCPaKOcU<R}lXU0lVm$iAzz<hUqApjpS3i`a`F{cPA
z3WA@sfMD7QcOIk46t|>trHmbW#^p8X8?%>x{qZf6Cj?R3Jw!VmfYxb`*r|8EKo?(^
zlmanlY=Zt~%bQ<(!Sv{ex}8vIgK8`xXOhxnozX<<9BIapR>R?>p{@)`RRj$=D=M96
zT#eS6;75rJQL>3CTVY}p+ZE^W-ZKQrD-D|glNGfxG&3<VuG$I9{+2NoQ%ePPTZ78@
zzGJ&wqHKkBsf12zfxAgeW>X=LdM|b=QY~OBO^h8rW$LQp@c4vnH?UYP=zO4RXAG+a
zn@vZ~a9}Ktr&A8zf54-oh6i7)_!ocv319xjE9PfUc<}rsXD1(X+dbox&p+q<@-^O%
zcP`)sVvj%{6S?c@$B{vUY78fpVWJ!A5yB>*;)YIbY1&#cA`g$aU9Y&=Y}mxe5Jwi>
z4V%T5SD(M)hvtXOrY(m@GmegqIXOAz_~@9XZK!O8u1Z&>Bu|nQ@MBL-p5O<nO1gbt
zu2ogFkS*gl(ht2v;cPm#+bw<H6Jz4Z<M(iNF_1(|rGpo7jERZu%{AB8U$X8l2|lo1
zdz4Q+di;PGGh-hxBX|`ssz&7`8%s`i_i9$+4>k6<Vh19Xrd;1$j%o@XKS+8;7<yDy
zarKKIGC4iv+5CWZ@_+#eQQx`O?u)zKuAAf&f>8Q>K~ovWv&S#^=!1`WboQ90Ze<}$
z0cAxs9(>PgdBe^16+^#8<w#{UO=W2+%SK6aaeKSq&Fj}J77N;TLRGoD@#sAj>F)QB
zA=3AKF+3f~h{0yxcii0EvRFv|>}InSmGk*)*6S_9D9YGp&z|!9*>g@#PB=a~qN*yK
zv86ChMGgD*m=*38+uiwq9fJ4vf4>#crS6QNe>GF~Tdih)#qrlK5_hl68FGlkZO<@_
zY&ToBn=MUKQ#(iPEOk{=H<h3@)=|~fUNLK|EfzsiuxX|D_A)YtQpJ?Dj8zfy<IT3C
z?+1L?aXMwpC?05mVHoMQ9h>D!c%m_~*{oTu){NyDqoiwy@pVOIYn-jI%8`R&>;frl
z>9&a&J+X}2QyhSioINSY{iU=kKf`>(nUVr-e|GaOEktZ26dlR0QE`h&16CJOdNH3R
zt0Jt+67?7ZAx2!jcW)$trsO<YnHUv9)1Eq1iW+AveNGSuG#=v|M{`R^25n_36>k%{
z&*(zsv08zZaauwQWPK_q7#W`vDQAW#2G=%#(ix)yRx4^9Xq>`lgEE22gmOJaNT72D
zl?h7tZ=yQS3@J0@OiTfflHcEZlnV6yh!36+WlE4s)Zp&A=F*oarEpqfoY>h^&N8bT
z+NPndDrwHGC5$89j|_cJx7{!VFHQq0V>HxSQ|m-)BWDwM{-EZ`S<BP2DaW<Nr9|ik
zywAijkW)r$vGcK&rD`2*ThX)?IK?>hjA0}LlgR|@O75#Fc4#S!P*TaHL9&3Q4oUvb
zY+Wh#sT@ihbXJ7u_y4}n9Z6Eug%l+pwv57e6bu2C1-09miSMVAB@T6m*l}-t+M(;a
zgf6)jcM96wJ(z=VLbC=`mU>4U0UydZ+@?%rqBH_J?wP57-Q(LQPwe3Hyc;1Xv<Y{G
zT`uaS@-;ipyO_nl#iCImvAI}q`HQcZPG<s57a99NBB@B5GSe71NQooon5#sk6Uk1f
zCv#Aa3XRD-#9;@OW=K)^9x))2AL&->VIbrr8G_cLF~lg4=(L?M4kL@3TXbb<rY*#P
z(a^=n*!P&thRNZfbb(4y)ir(J<4UAeKlH`Lx>!nBL)A14U5C;NS2>PPPw=5<yS$}p
zTG~lN*Yzk9(d&*@J0@+*hmTJ<n9um|LCe4Vi%<F2pMK4kiz_A%pE5gp%y(bD$HX}<
z7Z<DtAwZ{sdMPs5D1s6=ev{yGOw7W9hj!%D)J(JHFk23kCT<7%#U<MWCx?eTd-f1t
z)hx%r^?F0s3)&fnj+Gx-ua>-dbxz7j3@YoI*=$-=uv1)9$%iw==v1r_1OXDHV7=ZD
zJ@~Px_D1RWcVj1`YaxXz3&~)NmBQ6~Tu!9Tg2nYE{mm6A^`w+pEeD1G4-XsKNyFvE
zCDZJQS^V5)vsv+B8yQj5PApA~)-*eGtQc29u>nbU_m(Wa8nhv%NNqCJu;lueKjPts
z->01%;mgET7n4kR9*azVSKy>nkeO1jpm_M;jPHK;`#gR8lu0u|Wn=`L0?|i?FtXY7
zY&T0{6bpo^vP@FV;lUjL@Qm5CCB=-<HC7we%QeepfvYU*)rz6pq7aeKFb-v*fqv+@
zy1L@}`npg>97<R8{g%rM$(!8;$%CTrdzQ;JK^kY?ygBFg_C``?j2P_hl$Np>zrBvU
zYfGY1zT-ZX&UPovxX-V9cbt1aaYFw6?`kCbH~T&U49WWq9{PS1dPXVM+>W5FjTBLp
zY^27>->oyUxfevqI=SzSm2>Z`!MgJG%3v_uEN;2JUa(#FB}HWi5GVS6WVu>lw1OPj
zY}cfm=(;WI^@?uWkz+zrOyHfv*$S&GEDnN99O4k@iG<KI_$@w+q!fS==MqXqcIt0!
zYfNt#`ggV#1xoK6(PV?o1)b13qST0C%h2DThb1m`SY0MTWh1rPp_QP9Vn|!~eJbzC
z{-oUqS`?2cMJVVc6oi+I(z0%KTA{O8a-lqpYj|+%m`)v&Swep`a<zzTI>Z(uNQNg$
z6AfZ&9|B{H1Z32kGmKdilSU_jQdLe^m8eu^T7%D;tUQelWfWSjolk_A7*nDTnbnwB
zM^Ug28S3h8GD3_PqxKo7Mk{>W!L~bIBYSggW8`<9Os7mIEwlNAgUJMIWGp?71FOZ7
z&2lL!>LC11W#oGs1C@$QO=O}I2M#`XSn(V0AM)UM!r`>S_<?Rc&~FF&?Z~hV<+;ol
zopH71;BdmpgF}oxKx@UO+lnx(u9?ne`vOWAHwrOpmS0O*X+R`tDl&#3XkOuH1}}TQ
zvgP+L4eXvyp-P}Z2&9~GbyIdsMN-25lwwPj4^1`-m366gg!`{{;8v*xcHC)c+-1*5
zNs#iiTPTtHof0>eV**tSwyhKOGtyMJ@*d_{b8;}n*>8uzeQWWADkIh%_$b}wJ2+h~
zAIMxiro_&YG6POkFl63*`UTCj#?=)wYhXSl8w)DYL{Gi#nD!&B_t-p;O(JSd)T;Q9
zYgF1T?)L)63&g4l5@4)Dqp7u}H5yk}WNV=^L?aJX$cegbX^L#Ssw+Y-1WoH?RM&5|
zU@X<7#WX_p?}r{Arh5neZMUH+CaoYvf-;8Ddz4a4W>b#(IeNHdv%KKo=mAYTWr!Iy
zCX~-a8*nP~bP_myKId%q9X@#eg#Y1ZulUL9&$+(1V0QM9=MT?#^!SYPn=8(*u2^?l
zDQI&f<VcK#M4iC&noZ7puC`p7juyqqxZv0rj-6p{4V7~A%PTg+8a+Q?Hk<MA*%M4P
zAtlXnvF7#d4L9o*z7#-_$ToCrHysz}mr@j!k6rAGF1f142#;=$Iv7Wa5<O<hh)`=?
z)Wt?}9y8R&z>tVxL`P4*zQHdp(ZiZa?C6G(+it)|Nx3*WI^|}uB(23V;bc}ZIdJ4G
zh59b^D@L&;!Dm9SR86%XZ;4t;_?f$zks_00pc_1O<FHzBGId}Aw_kn2qlfR&)JF_O
zp1)gQcLZ<B_sltBi=D)7eE<G??{ju~iqVF?>q*Ju{XmEVJ_g1x(Dk0}W{uIA$)q7h
zOI3+d?)2o4VeHYm!kLEI?1*t3xV^mvGRxJ1&1O@sgDH!d9DP6V*=L_|I6vU&lP5Io
zoVuP7lc(=xtf9!nx}lx4G);qZTYM?dw%d*`zW9RYPoAMsR8@s3#;hSnSp)9bquh@p
z10itR!o`b)((p~*bvf<M;qIKiy*`cp+l?!}`S^8X*T31Zx3ftw5>dk)_iW$ktDUm|
zz%Hp`H+p4{+U@^T8lx1Iw)DZ%5Aw0AYh~SqZVaT`6~ngU_WBxE8KEB(v|=3Pd2r6+
zTt(w1I921Qa8%UI400yLkr;YH9;Kn*pC?c%$wxuroYAyR!(>wZ%6ZK>l=GovVNxYA
zB8@-?dck(_71Qw+s|V@HRYEy~sU2~E6gDzeHX{*8B^OkcEaY4g0d|N=M)oXG`8viZ
zN>+=dsT-QfRB}Uu$0ty_M;p(qv9wLi>A{Sf+raJ8b9ob3Y+)N<J!Yydaxr)+t%ydE
z8zr7_#oRX(b?*IZ?K5mV1jNJfP<tACr|MNHqC-ZBz#Z=sF)6W6(u%<cJWv@!Qwb$T
zStFAw1xb4b6$>gUNfrkOGp5rC^XZHSr>B5$u)QC-y18a?bIs5XgfYm3ASVn;CTYmJ
ztyRJjIj%A%P3G~j<He(fkKaFJVnEXodnwSi!$?TTUXEe#Jt+pXip;dedqs?s{hd+S
z8e0p6YWF<s?h)+_y5Lp2$3?3<QI_zc)&670>(vPQsdC0?Bji-2i8;{?8==@}i3S_I
zr>Vs&b@$`7(clcKvfy`BqLyM)X_VFMGVJosC#J0T(TjK|dQa#FQZYQ=N8xEnu7HU6
z&`4vEtOISyuBJHf(Nv0u4^F7-&%f2W^Q{Uyv&R?`Rz+$ORRW_7a3}lSMUN=*UYU|}
zB&tNjk(I$_gI{)Be)bj3;S_Z+kx^Z(XmVyYTw_<)q;5-2L4*QCu~Mlb(OPs;REpwd
zQ^s1m?!ZbMScs8O_y|@RY*Q1uk)%hougFg0V_*ov1E{N-q3>C&)>P$YXeU!hioP2d
zmMfIX7%N;6SBY<3)jG^*Mb$NQC=J?LLQ1&W;>W;xvtc%yad3D{=(k*6tx&6u2M-=$
z#{eOb`a!zTLE=QLbG&zO$We31Y149`f6k9y-}36KUvRy+;lY!aeDvfaPTFHO-IA~0
zd`UNM8M7=Rv9xW1<S2rMZOqg<b2}tj6_^Y#uPaU~i^+<(9mxF^@y!)q`Cnj~hS|vp
zkDfi{#qWKegVQtKyt&}@tJiF|Th`kRA*PbVQ0!*{<f!q=5GjGpMq`!SDJjRIoZdNf
zixDJ17!n~3IAb99B)_I1Vf&7-TG0587<<-!!})5%t1)q_z#7AEJbl5TcHCTE@WfRd
z>%g(iIKRc_x|jw>l+`7^ClY;Ty;!0%OlLEcv!X!sUN#~!St;R{RE~frjRQ_OoYu^h
zg5@PFUgJ(IQ8ff*$wUf?Z6D;lQ}QWom>tXuf}=S)J|soY_4Rp4&lniI6m>qx_$#H%
zFph*Ah&d1ny}{UuiIt#iXDh6&Fvc>*f#q7p4$H-Y^=3;-NI@;wIJ=p_n~Q5c|MF|J
zGMt=bCR2&htE!sW>;Ufvv@smak2shg(|4ZCx-?UYo11ICeDx*984l+1du1rmN0M*3
z`*T;c?V6!o5d&|JV()6VdllSU#oXO!4Mh=C{T&>Cox1LC2oB!=G8x;j@8);Dczd{o
zA}k>mB(-qs5Kk?m6(LFE9Mi5KRb}lJ8HU_T9*Fq3OGS_m=R_q~*ELNmh<0UKYS&V`
z3A6S<C<9@{`wqPYbV7{^6-1FA<J}~usvM6VJ>tpZ=X~^G4gB<--+fo)-2GjQdOI>3
z$V+PLsmW9tvCLMxg4XOMoGC_v_W~@KTD<%U-)e`J?|#B=0-&_URTab?nhXX*T~F|1
zB6>wk5vwC6r(z5oF;$>FF+7-CUOcF{-DEZ%R(;}PG4iw50~cLp6*7I+1RRyoG&Z9N
zj788UdOR&6iJTdO6z}~Q8Btg(;wXhskto+N2T|r1V-Qj<)*P;)X<Mu*0E^KWXE?To
zF^eg#s~onfsVj#r&C<<c!D_h>E3jdtA3L#?D%K;;lrcY&A*)hi<}9t#OdXug63<Qy
z4-O3v4xu%jOzFzctKEdj6ftSe4&T9c-QmZc7<+0PIXG^a%_bNt&IHDp{n*nwCu-4@
zNZ#+8P+dl#`#|8*=mjtLl-8!8d&SdmCs?>QfyuBR?d}@O7y~3H_efbs_M>ZKFm;WR
z2o9+?OQO<lZkrY2wvq3whJp`*)Ok+|g4mgoS1W2+F$pI}v??gJ3MKhsR}f1fZ!5*o
zQHyi$QXBs2<J%#2x!i6lVN{?}38!RJ-lb&koVs=~9i=g22Tv9^MvPJrVRe4Rt4}{i
zAI&*!8=Nw@9I3ZU(&~!j1IiWALJFA5l46iiKo)Xw&KVV>l$NQ4kY_LcFi8S~b&^7?
zTWv7bqO2i?jxj}vdC5@-q>~BL>6FFIHDip->j_3>npPIxVH_E|Ew;9p$`&{Hp1N{W
zjgy6H^du)YZA?<YR85UiGRCT%V}A6I>EW8y&4R;&kxA1s^aE*>4R=Thsq?oqRmH(%
z%I`iuA;)Lbt>)7%@sl^V{Nm?p4v(Jl@Zl*ho<8N|c+U0hC0C0Zu9pjZOkjkl=~I#>
z0qHsi4I2$AL7f8^n<W#asI;avj><a7k;UybUtYfEXP^HJ(@sHIzVq>Sc>nzmcyRVW
z1SdJM>N+m2uUOvP;@3UB5A-1u&?K$J&^UPVDDLOt#pOiu9|I~zYOQIl!2$R!gcWJC
zAa1wlm~c6>>Ic4Ft$5XYRvE^!jz4*D#)r?I^XZ@d84s%&ZR}}Gf)Ht&38%-0__2cs
z$yyne6!X^@Bi&|;b&k5Ibj5@@?A1<UG$h@tCZe(>2T7{%+coR+ub3a4q3XIAc$ej5
zH!j?FZL)x!oSsUTsvt$r)%68o3{nJ@h9s5hCPxV{AN;^(yA`V)F9`v{o0E;fxv>!Q
zrMR=!ky2)}-LmbvQb_F-r!u5SITMRgx9>XouE&}R-PTmr;LL<rIh-}6`-IceGZxE*
z<j$g)&8DOjS*{jjgj1@G7Jkq$F!YjFdU1Kd)#VkN&8CP2R52XASMn&{%9;g1Jj^@h
z$~zGOyD`kpIP`CO?B2@X6>#@g6@hQntnU`E?|Spy{bfgYNx$+h<xqEj3a=<j!?i;J
zmBKj5>8@H!GpT5rT4KVC!#anpEJkU9$z&DKqm_m;3nC@CtxBRHwMJ{le16J<2TyqP
z@Eyc1gD$V%`5MZG)<IfBTtd9XD38&iVoV{SW5(nH8yG-k5>aS*5u9qoo@IxSm!equ
z{ktMoy!kAdM2dniiddw=nTlu<F(h(Uh0iBNjZq3~GmUl3Tf?k1j7hN>6Ngic&jBtY
zX`Si%L`VvQ5+qcq($I?LH*3AW4n`>|XEDx@YfZF@V0L$b@Fz{wG&QJXsI!*rS*38s
zP*;|wb`p)Xv$QFRAc=C3jV2<tE|T{<(5*IXSK>PteLxZS-j-ssLrDMuOQvy}vzg|g
zhIy4an!wo{jwXs}4X#Kztu@%nQCD?|(5ui?#39p8EWuZ#)L>L(c2H9_wd~7Ai|H_w
zB$C31ONEP>GOD1&O7CbLcax74BbhR@SGs^eB7}%(_xu#kx?D_pVn$n41axs<1ZDC*
z`&pq1IAbK)BM6;FSZD>Q%Si;YF$p)aAjVSA?PAAvBCxwAT4)U7t9K{riIB_aH+Y$U
z3kFHkiUIkz=_UULb|>YRE}Mxd4$gakKJOl=a<X;=n|+}zm2^b-O2K=szkY)|oH3bB
zn3;-(h~0Li0P;W$zfrgYV~ixC<QE$8i7{Cf2Y)$nI}U}|IUogJ5|`)@#aArFD8ygy
zQN{=|rWLlfBxA58sNU5TRny?NUD<SHQ#H=wl!O*<HyftaAx3M)VI(vGXB}RRY=@qZ
z6IBb^3FIGxWSwqSYrOY3=Qw-tfc0X@dcEQ3;1Ht?A>Vbiu2R%>MXfcN9;+hHAI@2%
zfqK>B$2I=SbG`hM%hgv*rgI)Ve8kz4mz+L$!g{&j^87Vdiv_FT_eHiUKR^}Bi=1Id
ziJJ^-1ujU!LT!u`(zSuQArzl2!oZLI^iTNnAO0y1&mM7fa?F#LFF80l=5RLUplN87
zWip*%>V^=N6xq?^`z>MUg+oQfIC@ccg^@ID37ZxE`j*fwQGUcEk4XuiB>v{Q8@U)f
zS1GeH6~S1lFz~d|eE8sy^G|=y%qkJiXmMIUoKAW2{`+VXSgmfUZCz|9L@M7@4c1y>
zv2HQWQrFEsZ9yBew*=4`T-hy?4@{;9v`tG^hUL`_v-cya5}Th~9$a}Mg&3ceaAKUT
zI664S8cWVuGIm!Uon`SCaseb=cppfSxm+wczc}Z1v5<Hv@5u;%#yLw>Rn$$x;lUyE
z*&%f;%T$b_j#N1dYN|4^m?c-vaxiT<n6@~piN2@rHQ)=mU!FB*4fENQsxmw{J0--(
zI10r;i1$v&&m|YNh=EZ`&=Nfi^!>o=H*a|J=7LO7w(Wqb`)ka*^bSPjD2vFw*C_Tv
zuD?s$llwpC|Fg$m?}w&u{Ony3vwO~!QF}X|-I}e9Y~Z#AQ)z6gajk4%8ZE?bm5AAc
z(u5rFIWp$R){k`kz&Ln<&xCYW*kx7Ws$&igA8~yAt6Pgskdh%KP0|@<6N(Kw-k|*r
zY%fr;Cuc#mQpo5rVMEC?)=89<L=iy6jMcg%$SBz)$vXt%?p&6?Yh$p=;m3eV3abo_
zam1>E6zI~3%93flYs{>UMIe|lMsqToVyq#MI6c->6OXM1)SJlKXVya^hfLBLZ4y{1
zlBq1ZvSh@B-)YUnIojC-Ra>G}WD`Yo=_(Wk8#PDCk+Y**lwyFa$Rt6VbjCVyR}-6;
zh>j6I2D)*>&1S`R5VwP27$A6@l7g(#ipEGWAv}0cBy>(#m1zw;K4^G(YB_2nGZ(21
z)F$H4D4a|OtTd3uIkXkNvMIuqXaiMa0gF~MjLsNW@_dnzx!a?yiIhSqobuiHk0juc
zDuq=}<SdppLX~l#3d)r-Rte8HrA*2SgDy((l0EK4AgPtY-$T>(d|fE(3K*-XDofOn
z97V*li&CXzlgo%wMyjc-vv)<EASwBtT}z5W8Bsd}@tn(aGoeg`!HY$X%!H{cC7I2C
zt(8HJZ<!UF?}`F>Yi}vocCXp@N$Q@<II^iJtqMYu(OFSjN8bgOU%uvcGUZ{_(9{lY
zS2P){Q)B~4O9Yq-qg345fYC(>l=ldWBr_E0u+mr~@h#a0V)pXys6EM+(gmoA-E+cr
z8`QGvSgu#h+Xih6DQD`YLB&8f474FqS;x@z_<o>i8gb*^Y-yFkRuN^SoAkxKy{>D*
zFcM;5esI9?$q9?=Tb8R8CnqPka>i<-Fcb0Ui++@DAw{M-^SCuc56#OHnuC!aoiA8)
zfo-?s^Djrfes#&i;}aepA8_`;obAx@<;@j~)soF-i;n?Q3}bg(rI0g&L42<w*qo_y
zK#$_*mW)Q{#55<&7&$pSB&S4M8KNH;wrehZ&#UuuzWDT0>MVJDRmr+++m_jMN>f{M
z$xzKnQaL0yb;Wvd3FC$uPlX_`i$SsMBdf7z6(h?GTMb(!sx+-K)7tQzM<@8}8=8~}
zV<#$5l{q<X_~_&JXeKqA&4LgDZPga8mqKePhEhtTVGv%#=&4-A&d4%^fVEa$6Cx^R
zoR$Tnsl}&ERW;ke<2N0&!rze5t5hv|zhBGeV~`{PkjCN7>#rHROkFiHRxn2R7E0|m
zaqmaAn>E*0S6p3R3wrED5hsg-_;$@^GhRG@L0eC0n-;A_`R%O5C@?Bvbs5tGmDL>1
zXB-_JGHWMTry&RYDC#`D<DO_i87eoG_nq~fn}03>fvOVFq;$h#yCD_@yBM+<ccJK+
zaBsbM_jngDQ~!pg?f)xv*#8e6clH)Cs?8~r6qqdNypBo`0A)x7Py;c*U==x8<jyu2
zEiU{a1jf*_+AP?19bpV&H<ASHl@XU>2p#7a8(usc@&28yrEJpKM*-}t6cLJn<X4QV
zE8^8F)X)>PnDNzBMGgU<dfE7O1|5stg~nAzV%MC|86=u**Al^QinO~kv{GnWQ8^i{
z#o(!&L(e8umB#8!-w&YTKEpV~JA_3W39WBhhpP>zb3;8pW_FZ#_-3U3`I;YoI?|7_
z*LMlhIKVi_f@LJm$~uRxD!g_ql`t9|R>U<<o?%51Ut!5Rqw|#L#vb2!LJW*?r0+Y%
zam0@!L)Y!qv3nb;oT;3l(Ta)I;3~3_wZfpqH+H-;nbtZptKnp7czo#i&a)ZsJ(|$k
zh)%tnvr2?NRb5fn4R!6xl)x5VpDoi4&)^5ilDD>~ah;?)>`=ETVi<B!kCLIJNa$ir
zP1-5yWkm?y6Ni!559E;1b+HR7+&ZO{n1mOwXB@|3XOii<4n-~^r{bbf)aQHN@xHPv
z081@`mFOj!Eagzt*%TX!B=KWOILA3fDLnIz8ld(GT}kC#ScM>TQRGCq%<GNI<vc*v
zilhYv)nW(y*FL@-Vz-+;@Be;R;qDicyzd*!Uw0eRZtNqRC1VV>_;ln1!{ve-_bKDS
zjOyf&)<-HQ7YGfy5^p-4HBK1_f&j%BP+D#@DBZi23!PF4Ue6h&oZReXaTAl(jCGD|
z(%#o7l^e)9N7XhgFRy5<VcJZH){2p=F_>tG!-#7eR0cneOeYg^G39mEVU(udcBHIu
zwj#%bHHHXmwp-e|;laa)4BL*ya>3!zA(QD;#3m}!^+m{`G_|Y6rP~J{PiC~XqPf0e
z<2~nOCbj0ej||BZljiGJUvlvpj;AfN!#OXXJR?1M%6hftW^v2S?Jc7pi75%h9V5ZW
zW}gtbyfh-hEs!ZlV)TKzvsi67J~?L6PH=6DA39!r{)SbzrM8+DjY@&ETF`k<pEG{i
zGMmpt*=GzsDAwC8gAcMe`5tRBI(x#<(|3tQZdg(=<_fPp(FjS^DX6jpcRxQmA#WqA
z<pOOKCJfZTqa(*}essoRHR4xG`t1Tp)J?I-31UH@lq{yPa5AzF<e2v3iN5f{QW8(Z
zJbDg}51GuSR81oW%Q0e;CvMj0V^311IPBjeZ4`1XDEu(6S*^trjKF8V{3T<baL!TJ
zHO@8IszDnkCYUNSj)A@#&=|-CCC{1QN5;|bRiWN{PEXJF@ld+>d1<W+Zz*GxnAhga
zK2_jgK4U&<rP(RZuh^pKyFxV<&WbW3cqnNTAPVf%^(B2G(Dgl=ab(l?EEWrHZf>}`
zx?-_f5psT~5c}0K?jG*{*Iu6fU3qtR?26|stQS;bpHZn4bVEX0giB^dtQkm26|0Og
zGA+}Pn4@GB$Ggnq9hVxA#&aC``t@rby*cCM_0hZUrYKkySrzGcj-)t}`i{I=lW&*I
zoI$(Y1So?NwPHe=0}&POP#h&{d0jY92s*q|pv#XHYECZC@{Y@8j7H;xTHqY0Ob8xh
ztQ1E+F#1s%7X<wYArS0_s0>xKRI#FI<~%-_P`N2}GsCKLJ~<z_*(QdpXaR?yJSd1!
z)TY~^W3ydxeY-+yi%|+^iXp$zf}AEmRYGmbiO7yyTP$F*pmZ7}o+~A6PB>G<1S(^)
z!YU;OMlqvfl<S{Cp~aR<DJr9>oZ@Jzcz)`5cGB?T!GsSV&zV#Z!bsn1LXcR-x^8Hi
zhNiAXgs|g5l2O`<a<^(E`*yeA8xdnro_(Rf1c(B%D6NP=ip!lcJSPzkWhljcDZlca
zTE1Y@`?W6Lzy1{Qk+DTd9nrcJei5{*_Nw%q|J?3=B^SkU&IKh7Qg4KOClHO4)SfbF
z+kN>?P!W^l((c|<p~A3xhe~1gYhf<1sT6i8sbVI*^8)_sAK%)?F7M{d_x^tWBnrPy
z?~6NewJ#ATYUg2<5GjV7NF5@zg6*4g^5TYSYN=JGGKS=&stCasv_uRDRRmL+y<j3L
z#2TQoSm)@60qdOPvLP3!s%yrW82VmB_gWEi5)Zb*<MSaDI#rUu<*K3|Mx2qHR-fXY
z@IG`s&N>=bv0SY<n9ores!Hrns-~r?DuN$TS<zHAV?U6Q!0%xgXsU`wk00~J=bv+Z
za|1<1JNOY%{R$Fvvt1HG#5jj-DrQ*TKRjVF8L5)u&whDHO=4>z%h=<sVT^&R&A{dA
zhU)cej*m|`IX&UQM<0QL#bUw5)g`x!1)E_Iuc9*7S4xUIQ}|2KCtPuQ_j%;i>({In
zH?(bwPl?&VA@y`35m*uCTGMEQ3IUhFDcP#C%7hf~VPFWJP2X|5>F~<Z$B6GEJ_KTv
zn|+%TqY6kZBR9CqnPaCpwT2T{VK)P-O;3#&ryg4P;6crIo*i)7dUU^L9D2NuwA1PS
zK@L8=6=@_MP7>0dQ$&~SK(Pjq27wSsLu9+$FowYNNHGlqjhiq@k+i<W#BGV@DW=<5
zLaB|G_&ajOk3Gw!Ac$E-zWm}dwyUn#$23@5qiln#TJbw`4y6s_7$F&s4^A1w$fjGB
zu69e`59ADC^laA~Hk&n*$%Ld6ei((wo680j;?Bo)PnSrg5X>r<jYBq%><bcKi25m(
zO(jb)7lJ2-JKLJB>)3W3{m`@AbgZ`<*6THWKM=y5Mah0~W5@UVD;A!&R+(RS{JWxV
zckJWaDC2P8ttq8KHq-!0MYM|S(o%9kBjmow`zg%bhFpFI_X<0Jl(}9kxVXCF>gHFk
z#lEjW&@e@W<jH;@^gRp%#!azWOq(;s7Dr6!Q0C58j9fN<ZHg;HF8tA*cclEXcl(pN
z<D!-VKq;}R2r(0rOlqugm{BwCcv>p#k>{AQ7|v%Y5TZ}CsiAg`le$7baEu=wVC<SN
zuSU*SfhuJhB~ejE8^#P>NQ^l!gp3b~m^C}rRU{Y6L78&dS~huA;;jhm98DCe*o(>U
z?EtM3mEO^9%Ij1tS-XjvjDvm5T&x9sbQRPtayEzWJfHLYtmWBh!=t&u`++V;f(2_U
z+P0>trMRoAx=hT3ubfrFXtB7Gy<U9ej22rd8Hon&Kl%FttxVihzT=ke@3<`Y^v<L>
z-?O1oJB9WBKD$%Yr<C@N3%h#|rS_TNl`d4WvIk`D^clM;nLHB(0W8;EXYnZ%PsJqO
zt?X{IJC%Xh<E1@SO{v|zS*TEZV{!RCb7@3&)@4bHIOE&WO#Vh3J9&e5gv%Y7aDVd3
zua)kNV0VkR6orCFrNWIcXkr?9aCE@4(?g~XDka<=C=jh5NP|ZugH8&vB6)de%lQSH
zg^a%Qu1R8?th6SZEUTQ>xVlD}K*&PSOh$>@_HHb@C!)b*I;C4K>H3bUuAwMx$ysLi
zAxZ)2h4e0{h$@jM5m#4eBPz>r7#N3<Ox{bo{V>u_TAn|D&iVN{##!cvhvc03{If5(
zxmwY6BRPtNfw3A_8|H^oW``3_8pl8W@C8rjQ~u=VulU8qmP%W;MzcYYtRaL1e&p))
z*IZw|VSaGH(dh|C$45MR^nKZY*BdTwuDM+-S+6$5JhxnH<W5UI$c?KN&QhT59-k6p
z%;e>Q&CnBu$XsdWMl;olMk!ija7L4jCXg9a=7z}Cre{4yHXeLg__8u2V<1OrJarz4
zN#QiKwPo(6xPD}kEcIwGInr81MdI;6!$%LNJUe#GO+fo4_yK&v=!&*&h07R1>58Q2
zvUV3DO)2a|MO@YtaWh98*`0CJt|4koHejuy(GVBcq{Rhoc!6<>NJ~tz7-<b+w<3$h
zFm$X|x8k*^6GOLU>^o8jbX!BtmN6<a!qKVfhPs~MY)e*F7CdFBZOd2<cpr$PY(jDL
z++JVvwX4~zHW+Q_wms)>&RH(kY`PI2cd>ua`%#h-W(RDRYXZWl3SNq}5E8y56a*hd
zr5nX>D}+#9BdF`>#SGk!z9`bmxGvu<<lnI6|4o;ne@oxz|9_6R+wOaR+5Pzz5Rl&f
zxBG{B>yNva-K*>raI;+T=HiO;i(f?;lmc||kt3pVK;@A<_T-^Qr$ntam9gSMm!iBQ
zrG&(;t34`|Gy4Wjq4sN))`pbB-cN2<*GX3CJC-p<%zC>%p$b~(YFE~ODO6Gt^X`;9
zN+rC995NwC#uSPDhN^+8$vkcw4!?iGbYeOBV$F|#xhD3B*4`yNgan-r3^6by@feL!
zNcg%;lybQeK9}?fK<O-eKV9--vW!u6&Zr<rU8SH_iB_kQBMw6r@Kq_<><_GEq9r=a
zxr#VzIXIfo)S23hoXp|3K6=38<C>GH!h{vq-Ii25UE9fowrQzcjdPWx3B`m~MzUND
zSSRT!N^5-B5p4zKhgg)W@?o?gMoG*_F`=#6iw$DB!{B$6p<IeBNe4=T7MgqUgWP#v
zrX=f=7_vt}){r7J78S6y?w*(oY2bHvdEwCIbkAfs2}&wWl7b|lQ7RCg6x(Tkzllyr
zrXIV|ZYsvnQu)bVvm>PByaU7JtH_y1v3B_u$G0EFQaS~t_9x_<)`7fV0@S|H+rRxn
zf=mSs+P&pR4^Eg*EIIe&+KEH7A0P!%KcGS+XG2yNr3}84aAEnLU@BMOTNyRvn8><B
zsOVx8=?umhqF3^Lb0)hIX;unVN-@|EC9h2>*6TGb8CO-p(NkJ3c8F9}B?=oK1cFv-
zp8=Ya1gnQ~GwL!vz(~Ge_7W8}o6ideDeuK6>tYnyZ3p@x6}6M0nJ8A>hUwZeKb~Wn
zn&<O|v)_N8XP>_2&pusnK901N<GK&jbt?|s#^6I>b$i8Px!|j>4DEEv!QlZfU%uq|
zvnOmeTVjk{-`?`|Yq2xwwmsH}Ws0(*=-jDE$>6M`8$F&x)*9y^50N1X5^IbW)vFP|
zDXlUQ#a2ly&yW?vu7lGedQnka&MleRz@#x)6v`*))*KndgW57PP^ZYON<4pb%(K&$
zsSViF;rmYRnUslHGnr1sq}XU;h*YlHt1orwV96Rkj-(g`Q!b;Me%nzSOY~l@Y1L58
zYiwIFgq}*R(PPKOm%m{4-S?oTWZ4QCTOr~dQ{aB3BF^4lP$H^uTy#OqR8|sw6AVE`
ztj;)Najr#aO^jZ~N6Jy#ict^6UL<mHe#L6JrmAX;vBZ#Atv9TO$XI+O3x-NR2EO>}
z72T%8S<4tkJi>*GAr@3F$YPT+yMXJR>M$25W_RE2QJuTB^4%rm*Do&r?)bNGd=nb_
z*ErwbP=$YEky!pa0{tLO!k1t5?|iQ!;(#)iioie)TYR^`Z<ib!M`f!$nY~Cf@nx(Z
zin>$04mDL(iy8b*)yWQUkg0>*qtd+Zl<m9yDrTWo93LDLvxwt@4`}UBI#5-LgZTkI
zc<R2FNVAYw4_o4pi9=@GM*MP3U3*-!p`KM7o=*5j?;Y{|ho}7Z<DP&0<Ih=d2V#ym
z?Kz##m{xNv5WQ#F_iRHHYbq?&^qAyoHo<ebx#eQLVGIeIq1BofPaiRxAEGk!+m6dG
zzea~ZV=@P|rq$3`s4P@Y46zHWB_BppH7&E*l%vBr)7CQfOSFo#wc`ihd&%iRh0}pN
zZn#}^i~*djXsQb9s^U{A>zFkPjKe75*QFG3#g|kS-dhU36febRHfB&_Y#qjde$$D6
zX=SOJrU+N=?%71KKq}TzF_b3OOTv?}?heV_-)^dys#~h6-W%YLAxPqc)51A+4sG3D
zkt@ZcMAoI4Adip|lxAO*l&u)VlX2&r7s~Huv?}H_#U`t$dsC@?Vv6LfB$K>|n(iyx
zLm9cILODn=6@>QRVqqsLHcI8G?y?KtdfM;CD*3IIp?pqd(!F9VAu#)7HsQsKCs+$s
zE27RN=g?z9K#!qBED4D=8%SfI7~K;JoIC20MORmXwj@Q+dGFk;cPd_$(OIL6qMA<R
z^97v|&m2vRQAVnjW7u|}MEpPycs3M2r7DTUJEx2_7H2HImyAuFQi00FmnLc{bUcx+
z>##~wH#Na~+O`#vbMf~%n6+dd2{AK0m@+>);rQ&B$<*P)8f7HrWw~Cmxjn~Kmfw8R
zayE1P_{#-fZUQzL)?<NkmBKYOJ_aIUuDt0wu5YgS^z$z{K0M^n<HtOH@`$6;V_tml
zg4KG>_4N&1x24;5G(epdeiT$jq0r8;=ts6b2%@OO0=QL*#yK1bN5+=!)5k<enX56e
z@|jRZD4L8*qTa2HqIMe7%n1>EFX7*9g6Eau{iBAnIx{tahx3+)hYi*QbY5YSa4$nj
zMCBNgm{iXucVk$i4b#aKwd1Ol``Z+Sr;J_-l4lJY@6j<A6|W#y1dqxNYc*D7PNogp
z6p8%}6L-S-VoNib%h*V%{TM<kP1967e0Yj;j;tIXfA}Gv{Nhtq%Ox>KP+Fp=bd7Zr
z>SoGxa>!(QOy$~wG;Git=>qG@l?G&0in~D9^=!ADEQ(n(`bgN>CSc0OEzixSAL(D8
zquGsw%2Ru{-o5jD=exPyH`ECEozK53vig<dubRaEnxFCSj=%2l4JYs2^Zo8i)L-*;
z?_hW_!e-mET)z9lJH%EgF=thIB=`-!UlYcTrk!C-;W?)OIWmsIGmfL*^PLnj)|AGq
z6oLDB^S!rKkYZKFfcK;9a?TRPHSt(0D)@2mbiD|#jF!=COo>T5qiw}JJ1@Yr6)IZ7
zG;worBMo^MNy(GbfNMMl)r{YLe#l|_9zXx$lGm3Tu5NpVxTL8yjg^VULF?%=1Y4n}
zhrIZW-{ab6Zn`Z;>m@IUVm#mV*fH|^zx}&Bef&g3d+Rm-`d|EWmao31&I2ct%u%B`
znoXEZ8>Z71WsDRR3bNMdx<XkA0U)hmSaUXW93RekaCXemp&^BiZ6}dpg9mF{F|n?M
zr&ehZK!_+yVj@jJ&Qg$T8$*^UQxxQ0s|ZPBQB+!@A$=I=Hk~M&-Oie+6yjPOE>g_<
zQL9`}QJwBQ3l&A&rKJ$c#m`aPG7`>JIA@DNa-qc}u-Xbp<8~R%g+7+EM1I9cik=)s
zcD%S<4=z!1A~uPop}q42Ex;FuHCCm?Ym(AT8!f1ugu=?bXec#_-mkYR6%SyP6v6u6
zv_QUne5=APk+Efiwo32w-13euv3sUekp!!qtKAMCP}n315L!tdWmF0D$aGqBFmtFp
zqEaUHqlD*<zL@<9Pavi!y7?%q1f@)Y1OzgTqnHI6C9r(f^3H@zD!xK7@1Rp9u{~B)
zw0oZ$iZ_kYMIq^M#xQJqCQXa)M_kEx^d(%|YC{?WYSxggBlfZw)pZRdNn%i%G>Y8X
z$AK6;uBb1KHFB16l)@`UKKkBAY}Ot97{SzHnCL9VL6uv)@9BMC5)-D7BHL|GSgiQq
zsAXz3FOuUnT0UQUE><gYmhfWZB!GM$5<U*xygBFM%{f2(bI1H-#@VBXJUBb!<;xc|
zra}jKu*PAOY|+7syLy7dM}ciW6qF}mbA(~bNMz7O!3tzFV-$CGho@2lkeEy+7^^Wc
zVPm8ccU3Mniq#mH=D->7ovGrtpG^7w;}f17OmR7reaHWsz5jT&Ez9%!u+MVa9dq6D
zFMU;JRkarY0wBl;GDtuXa+HP|XrO@@`X~55Xr?h`$PgukOlU%B04W$`1ceTv*#_w9
z>hkHHz3!N8meb(3_Br=uWmQ*#XfzslE91U=@1A>)*?aA^f9w1En&@|=G!f&(=tn_R
zl_6*0SJZXOvhJu&O$rH9S`jHGau9`-F4O@zE*YZmGn^K6vX6n3CTyiyt{0frVsWH>
zpjQLQRdmiUJpY2zvr~500c+nOMvE~uDx2blE@<89=~{Nw+VRPEe}`9>6E7}%e2C)l
zq!q@3a$qV;yXaV*tf=d@P!AGGEif(<q6O`Vc|?z5X(1kCl70|l7Q{3ubS@N$!(LI4
zyIE8C1`zYB3aVfAm^%K<d?S53{tCyhbziu1(!M(K*Zqv*jVCVa=QMiy{uiM>LC@qy
zMQOz*KahfSWt?@TRh39twik2uXNt4$S5X<3d|fr0&mK^oJinC#Z%c|vT~cM#K>-<x
zXsfELQUEx%+l{o!tSaa<G<79ntLUlrwnUp*a-|8uGxVYs4I$9)0?}uxCekhqPcK?h
zK9}cEH~8Hioj@mrHHNxMR7#^3HQj@Iy#Mis#1~h@eTzFk#Wsf9+iU!Ozy!s|pL~}m
zj~>%li{EXy{Mn!LV!FnK#Qmk_!OC&(?2K;N(X<WGXhJR~{W(d&q@~*wd?4bvIO#Y)
zt2kR*$OF^3XPN?&k0fnSMhZD+EIxU3iL{$NWX0Aap_48uS1BTsR^_}&D^jbBE}Prn
zC;T*$Q>Ln`axNwQP2b5#p4*nQe_bjZ-<%4Mt#sFBF2KyGOi5b1bv+kMGEY+!b9!w|
zDJG9|h>ReMGjHzjI#SadJcMKtA`S-%Z3<n95*=A|1>}sj5^XmlZmI}lQW3z+-411r
zpE;*2B8pPj<?N50onLc&GsI3stkN__<4TU3QjpUq3J}U?S3>BNhFg-Jb2+&#Y3k_6
z7_!kQ6EJ4v$)j@`JE0@^VG{KY85L6@dM7zSS`3nNnWS+hcq0=gXH5#3(2D`3Do|^j
zg2c}lMMPr4OxcAfQw(%d<PesY#bkA##doi96_?ky<n<{Eb#QuD+7NuCag7*0DhsYc
zX+xYmVHyR#wXVR|Qi)gA2uG`U^iAGVH8r)Zz)A#JL{Y6;Y~c!oFkp)>Qf+G{uTi7M
zRe?cUQRCK@^Hamk&4#CSCgq8}(kxudN_jrt4h#y(*i&LgTv?5>(CZ`uh8)>n-|&;m
z8-9|1L}e_Cx@FNUP?e?X8kXxdtMv(1oTItluJBWUr0_9f8;c^Kl&7vV<FLaI6FDhN
zR4kL?gK=Op?AeSxgP-u;GlWR)1FGcfO-j=!Sd#hV-17I{Z~4x}iZiEiIWYFA*nady
zA*~3ZwAHo3=*+TltQR$nsjwJ)3N&?#wGN*>NlDRJg2q5=5l@ueSM4gSR*e0YoIR1q
zFm9=n!-Wn_1Dz$J3FCmfdQJZFQ`%;UGZQJ-<XlO%V-A2v=Ok7OMw8JPx8k-pyxhPP
z6h<ek@mLeF<3t|=0gYGCHexq2P6NZZBPI_zp&e+aFiDe@!%wkX8(NCEvKFaAKK^Re
zHaE!rf>nTD3qHHse*JZdy0>-TznJkWkF!4hz+D6^-+lb<xC`s$5;vLgQ()-doJpBd
zuz}1~6(I$p@5wVtU%D!B&iPhCO_>X{xk%GWVU6VH9^$3)o2YP}MRQ+MfQsNk6H;It
z2CB+YIkD~tzSw7>7{^IAft{o(Fl8#IP)baFb)~tePubk==yyH+ZqF1VA?_Kbf!(m@
z<iWdq^z@8-XKPNnnxFjWb8HG|ok%88*U}woPFo&)@DBIhdxsz04CrKdczVieU9-7-
zMe>P7v*P3L{ucL6PpJ&V{VmTvd`dfhiS_Vssd;+7;NJbmxV9r2Eo*%ajDBPodl-6J
zgyWbM&~@Cux1#GDih=EB%Qy_gtUy^zHLG4NMma)JBa4_S;(fyVvCIO6p1EMgkc}y{
zc8xVi_HRyvsb?H|{N!mWN7FWP-b*Z(DSovA9OZ*mi2%99yDkgbHKT2FD;_1v+E{9a
zH|Ad!Wp(Ww);e5Ol~#4JO3PUsON!!_A|{)oNh!~twcPimQYe|kBq-Uigh_b#WSlZI
zt0hxTB&Ea=qWJfQ5D%Hyb7r*GiWD=+3nFQh2p9iS5j&$Yl?7)cmu_a)iHhS*NdC?L
zmt#^6Oh}S1sFmDsn%eRHJ5Q;M#t%IV6E^44QWxYzl_GiegrV#Pj8Yh_P_d|QLqs{L
zdVLCFal)Gf%S05StPzMl!NDP#oJl3V#1v#}8U_}1gZC4g+gp~)CBrZr+IlG^sr;=i
zrp+Rj@Y94D49;rI3^S{e9|^QggCD1ZQZmQHlqOUXlVoEYa}d4q6JZ$Gj}wg&%NM0B
zb=^?|)(F4Km;r4ZZQGD@rmAYX)fpdbCcgh!=Ja*u$JZ0Dy(C-g1LUR^E<odO%Ft#p
z??fv+u#17$+YMiAZUw2#5P+(#vDUI!E?IU<DqFMYI<$6ZZ-wWQMm8#ua=^#PFeSzi
zn0#Qjy<)%J6Q>CkN_g`;Q=HfsbiSu1@w-o#{5zjK;*+zAb@HU4XSd(6-+MxsNNGA;
zAX;1D^=U&@IaaF`%jF7%gk2kBsq6ZV5-J~B)k4IUn?XvMO6!u1>ey`pDF!O%u(boF
zG1k&{9jdD6#{pvG_T?)kR}<fTmpBceI#Q0PB$f|ic&ZORaw#RAzk0=IUp`|uda<U6
z5$#JY)+=szdoHgxG|hABO3af}j-(vWI-_)C@_WW{B!*lB52>_&MLM5yCaR;A$=mUE
z9LJkQ8gK;Ve&OfxSN?q*8llS+2b?=Y5+i2Sea?XtM#lXuD$C|iHaW89#~2A=6drR-
z<PZ;}bz>c_s;OM{#_MaXpin46h-6f;G!jkd@#o~aHOkNr6HODb#?jO*DM>WlGzFB>
z)O9U!T){K+TUM(I?F=C%w%Z#_U9)~rvrZi;ToA^@<Olk3;C8#`^^bqZdVP=8>Vm)d
z$$LCH>G=HRb9UpNY@uFSa@4dJr<^{0g0EW!({S1Myu5i%l_pNwhI{uO@a}sb^W=l?
zaI$EyOibH~`;X5t<3pO*@p!GdciQ2bHNiHdtcklF!{!!0_UPn!a&e9(<Nd&X+=K1N
zTGJ0B+b|HO0auA#NmgKMMV6)tF@PT<H{rG*SQK+)%6(5!DO>K(G8s%kvvQgk$AI7O
z7<YS&wscL!a<vq<2_FbCpsc~UimY_85joDES?S8%P3ZybInK+m7~tQHmYbEg&N<d6
zYr6H4x|3G*JkgIz+Hfc+$x>s?(@Z-}8aaCM%@!^tnlg+Yr8358LW*qrp6Yy!ttzt7
zQWPSoFFIxH9D}FLQBV;Q*-9rV#v}*v8?M>ksIap-QrV2tsSpH9J2ijRlj`e7Mo58y
zOkLckbMl<57j$*S)c5FV!W5V28D3QdIhligXVU)cN-b>0S#F$ARv=Fb+Lni*>>w0T
zC9(x&g!HIlBr4F#7WIr2oXS|xfvT$5Y&QfWJdBhiL1Gq2+zl&NMMhn)TjrDvUC6*b
z__D|-A`$Bx(=ZAbuW&%T_i|0tiV)<zxT?Wa4YuhdZ`C?N@brCuK!s~t(^f5urlaro
z+}vEV-R+1|re)+e?$x|^{vm(%{EGWuyyE9Kp68pu^^no3BIQJND>=1MDZI%916eC7
z1GRA|G(!wv1#Jx~Gfh3?)xgV_&*ZZzg~6h8jnT+zP~5j8;1zfjNhJ)PDrCBZ_&t?f
z*QlXak$nt2P?>-GqZ9tl_deut=jg_sup8*N8~SNrEU{Wn8=SSIl(E*(wjFKT0#H?T
zIR~1$Z5e|&f6wP-=Hley1kN3v9S9QOoI9N7m@~sPQM;-*wkHTdgbgurb$dgouP~1u
zk){z<1yb~-9Tf|zt)<<RCGzJ_|NIBMxV|PTLE+&Jn=NJX7}<_n)a@2k7A)a$6n>&9
zqJ~`j2a8o0hyN9tLB899eLMcO8Gqpdoxk$+b-(v@1>K$B$&j@~ImNF)zZpf0nyey5
zd0=4NU$T4kiba}ORTa)Eq0l4+3BnYZhLJE%l1@`PD%RL?Q<V)|I{JT<SAyAMNY-a*
z2hQTXn8m6GW$@86r3sr2+F9zhW!F#S99bF3J07QzoFn5jVQo!iJB&{BDqxhPLTIC@
ztD0aVj0#h=oSZ&jzv=1s1LN)`IqkT2v8G<#XXhvOlgB8{=6cUAD)tfVZh)9@bMHt2
zM&p{6_4*!Nv!tpuBu`QvuRYbtikb@D8hWE7<}Y}xajY9?f+M+%%8tfL%8d_lpRvlI
zf}Bqk3_6R_J}cpn6C-g7<m5p)bOu}ablOOanTTF>5t<dV-<gJ$0%02Q<0K%44>(iN
zG%an@k<pAk;FGvy)V4v}ij1UD=sP@ql`=Vqx%8}-K7z5cl6E2mFMy6##5fXyBE*TL
z5{r`s^<sf`H5m(tdFz}ckyAhwH7;7CR1M0Ay0x@=-BOV444{OUont}#Qp6d9aTPWt
zeBR@IpmwfM!^+!@iea3%xxB^diqm`dsGCk4L`pD04uah2A)xJ>xyygi_-2TmK@4O`
zVu*-%)+Llcx4*_nh@V~3;yS%(9pC-<TeOuS?I!RZl@vLJVn(RSjYxXDk}-<3G)<ZK
zIY|+rbqNzs#1y2hl)$2jP(x(uRy*0*Me)7QT6h;mnWHU3nKZNUaX$>SZG-oLVGzaF
ze!s^#C+JHl&OrDd)>?EdVZg-%GpE^_MGQY>B1&;M4nyI4)U;I%F-x`XW<G=P142kt
zjib_5P?%!aD<SQ=BqT&mPfuvtird>8Hdn7Wy*TI5>V(zD?{R*9!9V)`Pk4M<q1#hF
ze}2ulpJ>LKaT>Wz6B|5zisT3`E9xYZWl)MSWri_gtfsC7o{u)8D~;7kOlwm{Ct%8$
zs4&Pyk$`06__j#VxTq^ubxox;yJ63EoKREd(c172KDp1|`{*%k(oC;!+3$M#%{BeB
zCmAWqb3%+gt)=Ta0NI(X*9WK@V7*?0F^pS(fWBvy+)>#n(tT?!0l8SUQi2XBG1Tmq
z3%d1EoZF2iA=W6ZtMM_>D9h&NhSNBq!$`J)kOCrDINls%ltL_oz>h!uIX=Z-DC`d7
z&V6+DZ^}@5r}>ZpOgSFEIDQ3x>g{+tfWNrSK0o%=wLZTwyM!Y2iyXV06xn1lfgDHr
z?KR`=4Xf7BbS<VB>taMrt`)-njdxnIN`w>ExIzVzA|l;+-?{LeBu|`z?i`Cx;Lyek
zAu)vs7cGr-G~EI&w*;Rd+p?h(Gzma!htUp5Xl<!oBO7pSY3iEfAY?@z134)XH6eT|
zr6XF6EEbN6jy7cW`#{w|R+ih%mTuWY3>ZXYXq1)=*0!NrEU-qC(?ptfglU6Tk*2M1
zNfA_L2txI!DnnO0T*_z*xmb&ZX`t_S2XlU>wPY}707J%=&W3Qu34$CcGFfY^k<4gq
ztq2Z{bi9PasNe7u%+ZrW&ouV<Ni0pYa@0*tRo58nm_o#-EWW(PqMSo%dze$w-dD%-
zShC;&X(iu5({j#)Fp)_Zt*};ORPn&g31t;k-C*kqZ5?RIW6gIq!kU@SS2}z~gEo?9
zUAmOc=!1~QB%z(nmLyr=H;l7rz~zYd6E0R59NL&-*Akifk?U7CXe?cKO4A5EEE|K;
zj;u4qd!o#_Uw?cvj-3&!kTiK|$!UsD)g8%hKKj7SPWV9abSX%K5!FiLM1hmz$o1<z
ztqz!wO8<Jk2=k$hn#=EsL24vLQM=CmGC2fX+Z2zRjL%8j)|Dkp5LLod_AnQs_kz+S
zb@0KH0#NvU#u~IWxVom_?#L+mzNc+lR;v}xxkGUPqPPty3e0Z}VH`v$Jq1i%9Q&gj
z7j29Eic+*~OJy9J%?3YCIBRiLP4W?AofH&>OE|ZGL0Z+;xB_en$#figD(8elks~MT
zHGbc7dAnn68``rapFDZMs%`m$fBNU#4zKv`Z+$>NMqXUq^76$?PRESK;-^UjCs}wi
zV@P<d>2qSQ;l_IcnbrMsR;O!1oVdQa<oV5xK`9!mSu~c%C-+z=gWqn+(|}H%#u(13
z6V}TGs;Y!*x7!klXa%3#Tk&^(_gnn#lO_536`R)^hTDl@H_#7zvH`brRIR41Yr0BW
zs}~pdfXvO!Emc(?obsEdW3gCrd2>aK;b>YpTkfPJ7qC(X!^)Y*p|r+TPSmkm;YRek
zJqm?eJG3<in;mB?%ev+D&1?Mb7PsDm8&Rs3$eSc~4574(W*@!Rmp2q&SZ4WEzH=k`
zg=_I^_EGup$KCNYEw;Dg?fBa9>-JrK`6j;XkACUqLN2RF^aI<QE5_XhU9StGA;pKX
zhMectY(60B6e9gBoG``~^jeCxgY@mtQcE*_cNA01wkKMPU~NcDe#B0ex>?Y5YqYwS
z$h2T+>W<OxAceB7R|si7qm^aRF4^w~Kv5MtkjZDpVaM&wEsZg(PZqe+zP;SuV%jBj
z-O{uT)39T?Skhy7{o*BR{gfO$I!81KYYk0X({>GQTR|EbHm~vhD~9W5xIBuoGek+s
z$P;Z9SyYa*WzBH2At=e9)xgcohJNVFaa3X{rKM_#Dxs7k1@Iv<3?ng#fxkq8X(}gb
zQf+PNYLya9&6YEn<R@~Nh<?ww-{Pm9T$H8tVokePVsu5$3LgR?m14+RY*o#spC#M;
zPR!TbhCNQF1NZ|%0hmUyKofDCD4X|VPgY5|>1|ttgBGQ25!j_Na}-M3LNOCM7S2#$
z!y`JFEac*N5i=xD9P@!gnzaTiXmV!_H=7MEcuaCsrV?<=M}}R`&FgEXF>!i&pGCJq
z+qzskr9Y`mi4^-wEbNkX3@KsBn4Ai0VZNXWJL+rI8~J$R3r9{x`F*Q0hD1suRa>Eq
zmKZA|2Q-yHXl6g6q>fJ23L-&LCA6djWJIMYn%PW_A~m+fLL}ydN)e}v|I9Ful)||B
zptj0M0-q`2#@(W$-|z9o>`-f6)UL4EY|vU$Rh4k%QUX(?^;s(3S=ptuD1R^#X=2KI
z@Im0)W!JIW?YOzT=Jf20#x=rM83sJED|OC7PUxJebt&?cBnxVzAw@!%h{}*-=45%o
zPFL*r1LJ1T$~YD)c<*A#gTM1zeE$bO;%7hoLu|X?llMR1d*8ZX`|66}`j$ppRB3UC
z6u7*;;bu2+J4RmngbIp#k00^x{|Emg78hr{zP{x5fB*l>^us@+dw9WmxnN;xo}Ap{
zH=n-G?(@(1=^y?9T^L!Quv22#Y<afdagzd@kXclQzxDAW{@vgCO;#jc|M-UtyIaN~
zGVCJ#Fi|xXPv3t+wJ_|4TiUkbWOc&O_ZVZS>zd7G%Vx7B=geZUpz9V))5LbOl{S4D
z&RT-eWFJd9#GALR*}1%`D#?#U#0RsV-7pMH-lMd%fsNL5O@qCG+gHzMFP@^ifVjWg
z13H*uWawq#Oc{o0-ifNOzrX%+u;k65?)=OfmEPN;?(O&&FfzwvqOUxV5kZ8?k(>fP
z?5S$WM@3~kNivm7css>{BE}>hb0m}zp+i$QSR=k^If=2WP<x~sB5SThP#uc8^8OUk
zp>ZWjY#e(~iMnoBbZZvf8N)8~{PQaoT}QiEP}eo%xFbe!1xPs`^0s3XKh$v?X}bmY
zAKc^Y{G7fYdG-7Y`hLT1D~U!17;Y}<ePlCwUcVagYJuubaOZ~WFTNyA;EFh;s-Ue6
z#66d<e#F(6f6Q?8Dbwu>rkgL({Uys<(R2;#^#W@wl}Z?u8E&_1H#ZXHrIjT9q(IZv
zpoD{*Vx}L4Vk0BDvauXfjVNlZsjyXzsVYp>V4OX83{uLfT#mV)2Bx7W`jO}dreP@e
z%a*oXuv(vCO+|{i7@K=JE>3upb0+(&us+PGT)>=}HbZ%Bb=-u{bf~&%1rhTelxFgY
zs%da-gRR}6w#Z6LY98|(p|tdMu~3mOrRSuQ@_w}VosuG`KpqnzC&|6mV#1#^!WXP7
z7FyE}1B1Wi6h+l7urAPcEoT?!eDU*VY;SLAyOy<4=tYIf1_I<!1SF(ywtj!3!md8O
zpufIiX(X!t7(H}!mIn?Bl{<XK<6|iCXaREcbPG#P9zPAZvjy$4mCQ+<sOsv#Wt-jY
z=M1^35`L2(M@)jwRt(!cF$O_bjL5Cg!odi^OJYK0#kANcLF$62ZKL;xB1vf>n`_iD
zytt~^T-{>zk%ViE5vTtcX_`iISIh4!J+lGDf+hHXsWe6lniq4Xa*kM4O#7aUqH9`=
z(%furxxT#O?EC^}J21MKHBapPi0}94D3K_xs!&GKFdP!}UX^H^e%E774LYDZL-Z53
zn;VAxEvF}I+P398??0fmk?Wfc>GhWcH22R=S-$%r0%B@x1nU!FKk((3&-ncEHK${w
z8Q^#S#_#Yqe)o6Ls}<X^$JPz!Pag5`y$9TX{FIawfAsr*Kzsj;-}(4`?ClF){nLL!
z6(cDqf={#>`j9y}Y5Ci~^DTb&!*>ZUpY!aC&)98ln4+k!v~gTKKI7q|2ULxtAGfHS
zXsU|S^$E9V_WK<m0+ySbTTV_+SgqF7b<OqF4SnBpa&~$^TIcM@l#guhJkQ}QMY%B|
zi+4_pFez>35EF5dTvXI3=@?Z_RaMlL<N2!>JpQ&vCl3Uow#AhkKr{AZF|t0q=nUE2
zxnkx`lj18e0dHJYU;Fr9R0O^qZ^vKK_*zU)MNo;H0x?cZL(j@tDkmvJS<AjyX+Wz(
zs)VElNc^8J8~Pb>DwSG^7gI!4>^2YqhQ<i+pg2%ra=A&)D#VZi;{-!Lu<p)TFHadZ
zkuN^`oEOiox%Y6va$U1NIVHu&SOjlUbcSN!9ogj_tJNB19NIutYfjFVsMMi~wZ%9E
z2JacXryVAY*WC7*{kUWI>IDxTe2a_IQ-X2$1eHm|vFGyH&l#>hCl6PQuRg{34JR(}
z=)7go80s3F3DUAoQ%S3!q<4rh?l=vEBz7N4Wg?N7Co1Ex+F-Pn`{tNMnVB+G-BPzT
z&ed2qw=riSVnIby60>wa5%zoh)Dy!b-f_m#b_?2eLDMWSt|3MU#osQZj48Gw#uSC>
zS0+*}ZTMV-guGE}%wUo_tfMLA(xz8TJ_!G><j13wByp62EzQQn#k%A$cjrQGR@)vn
z1|?c7MTskovE<onw6h}mbPi=5%2`ZRvFtkPaby|?uCK4TXEf0l6MO}`-JZ?O4Q<y@
zcNJYza>gShU)tc`kQV)o3cH8zeZZ^96Z$=_@EUTSCFOTchdTVhSMMeo&?!?HjY{Hn
zZH?@<k``V8){u-MI^nFPlu62<tS*zlQ8hI_CxTNrG;xf)zPX~ZmMSNdGuXDF>RNK~
zh6&Nr@Ap*36;72zugobJDJ60UGI0@~Ay-$_zUJiggkjecMt?vIXI@?3_a(Sp+M>ER
zZ)+_pRSba?BHEQjNojGEjuBTm!Z3=%zm9Zmi`JT}>uauFzGQiNLVLQVuB7iCqsXvx
z@DLKasV5>_J@L!oNUm4HMX}am^F&=);y94P#O14=Zn30UuIUynC`FSU%ab#%rk-cd
zKgYB!7xx~}HcJ|1aUrojsW^FNc^MSd&A<mAy~~5Ak7@2bU^7m{x+NRS%jci-)6J(W
z@84&WHGNalJ$}feZ@tg-{HJtb;<EQJf%EX)2j~3F-}^02I?KzS|Agm1|0ybr)Xq_L
zEz7o{Ted9M9a;tYaYNl$nz}=&;y2Z_Y_~i1yPg;l`(4k&2ajl*ju;ZV-Hw!!C}QW?
zIV+)4B5PUD#2<0?f;;97x~jq&BO6sO7q-@l6e2?$P`1J#ku{69W!ybSrDAzgN6F%i
zNz_ppcDr41(HAO*@I3B9Zu9k{xDzUPyv7cfmVWcczdC+B`s>^AcKqdy!?Z3<uPp0W
z?W~Z^m7p@h#VavRS%8BOqj22j?8rRdV&qBX@A>c=>dw7Kmic^mzM20eRH%Ylg-i}!
zL=uf4cTL@rB3!<@!MF@6u{Q0oBc~*%HpRr8Y7wKfL+4QUga^_=WvZ2;X)@I`;>Hn8
zCM+{CdZymvcN?bNhNvrot?@`?SRD4GJfixM+`oqXOPuzs8p~<jQrU#n5kq8%9+kv`
zJS91n+DaSPdoSkaF&<2t#V=6kNwd%20lh*|xmqH^oOsP?s}Cs+A|?xB2JgKXN&A5q
zCgGqdP1AIA-IA*AuvH^!*Br%LGYNNI8{z9|Q(B&fHEy0X^J}F$0MfAoI74Z(akZjY
zo}`Emk$TaUDz`WSsN%nSSQqa+S6OZk|E45J<wA{_jf+#1uP9;;lu)<GqT1HZ3J<Zg
ztQC-BVCy~OzL$b+W(ENJe#drm%j$GN>l=A*$sYK4cZ=CyXMCf=?(xUpB@H9nPk%<5
zhJw_U=#~RG3Wu8e4rHft2~kfaCr*n|t8q1T+u)qSHXRs;b2U{fRux+5;^h*Urm^f6
zD6(NqYg$aq>~>cql{mjRWAdIU1e`V0O~q6^jTEActtzRea}ZCXn6m5^%1vA+L2^RL
zuXA-vPH=JW9@nog3DK8HCWWD&CieRsr>Cc4XjpazXhkX+faG*sl*%q3rc6^e#3s;h
z_w>_<(xT`z#`5arb8fF*(@#69Wru4UDq{pXHlj#QIf#-b%CU|y5<?;s53~JtPtJj+
zmI%0MO1QRRzw6oU_GDXww$$B%-R*`ZPdqw3q1Be#eGj)=1~hTCBs(~()|@e6l;e7_
zqg%IhCkvMM&WZbp$F|{%&%WfvXJ4@2kKByUX;&xQPM)c*SU!G$ZyeoXjgJA>Pn<U^
ze&fS$p~s1z{gXfA)$5m>U35Ho=P}EAi8D2*L`WksjqLUtoKl>go)X51*ROB5wb!)W
z0&O&Y^6d6|hH;{98?-g_eNR6QW%p(YB_DRC4uqnlG779x<@AZ-doJ>;s=~P%vSRFq
z!|s0^2b6I%RU^C*t?1gCd^O^CTcX|JnhW7p03l|kapLCYmKfrp*#G*A_v<%E$AA7B
zRoh#n?(O&&G%`|SeeIN+d8>u_W|U->Rz_n`#n$BL`X53dM30Yw$$PR<C|famnjI^-
z;|r=g{{e@MS&Fmx;`q)4<ul|g9%{od5xR-WS-P&FswysDUy_x_dvP!5mLg;m#9s^Z
z##m62m@y0ku8N=`rob@vXw@AM|ISs|!3sa!PplV;en@PGku>bMe)XJwHxR0ZY%6M&
z>2koR0d05a>563&II9g0S2atQK?TWNSBB(!thNQY5;Nbq5c6Ro<vBV{%dxbn+{^Ug
z{}K(Tid|HVwRI`#gw8aBB9a0zF-@cRvP~1w59Aop+TdKxa(P1AEzzcu?AVy_rOOqw
z@XmEH`#*TK&3t+|{3#uJsUPN+tGZLzt5Td<BcFZP_w4(gq@?vZXV)v_oF911IUhC&
z3ckW!E`EqPLn-8v_k?LEkz#?A0!~ZOYqUYv22(k7<)o-9(QAi*gq+#m?xbMW2CblN
zDu%x2_V$|d{RzIW#V@gN8d9A2ON|s;ef(XXD){+0u>bVu#al?YLGuC8Dxp=9Zzv;0
zhf-)IsRbG_6pXP4yTh$k*wu>DlXqC3onURowU(g48H3;S*sS>cXFuZ4{`C92|E;Gy
zeg6rj)}l&8ky2o@-EwkqPSv&eZ6r;JI8LY-shp9(<{Yuh4xGg+g|2HzU{WF#LtK>+
z;%pLSr8b%=j$~KStQV{Z?6+HnaSt&u4SO!m*4%71_%KpsF<Z=95i}?xJD@ZMw96<9
zXoEEsN@=onsJ0_cUf_N2QP$CROU^Dl*H^C@H#hj*4c&4<y;y>+F!h3@G*)Dl#kWV{
zivLY89zRY@yDeiH34LI_UX!9?y*^`q>#?@NL_?J%nk;I?^^50J_cLd0gH4|G*^+9t
z;Bvp^YIBQsOU%hbOk1<AZ}{TnXZ&{95Pi?GT@zx>*~6!>_#@OdU}L~yi2I&l82J44
z6_-=u(ft(<8_lY1xaeB8FFxnxr(aN8!@u!+zrph4jO)!UFJ8Xj;r(+KT}RkW5HuGj
z57};RX`2&n$6GLtVcO$kkQ`fE@#&W@ST5Ie>or~lcGFH466@yrT3CyP@1uxBA0^Ay
z)XWB_DJ7iIkgZ&d)}ayZPl}0tyTw_{>ZGQ%(#}A^!a5#S9Y6YmKj7)#T;fh2;G>2B
z2|@P0D7G<SnAqLiNQ=o9%ap?sHZN#*(YrhTC2tHt7~hV!<6oBXmHy40QB13{#w3kH
z(Yc0NJ4}*H*f2^F<^9Oi_YD0&C^3G~gLeK6<JlRNQ(yOjrE`&sKq-}i1Emf{owYT3
zG=vZs@}B*;p{<szPAv}}t+~E>&DE<N{cdEp-Er}#qHP<hs>azG>kL(0%f{@`Ml8h#
znYL|FS>b&|fEQJAOGl=$7jah_P;mku`1Hp=;L<M9>oZR7KY-nicUK*@ny~(wh1#;N
zU{xy`moPeDT_tgqDH2CdveLGkyeDMO^>vSro|psHYC<msUj(9apq#YGOWRwcsBA@T
z#FE6i24fA%lwv5#nwe%RlYTEnUGU@(X{wf{?P$6Mu3n<7aEoI~#NdfxBqIR>GtEUC
zeM}KKoWHrKyJJ#Kv8gipkWewVYg5Xk2;18&uU=oX>jy3#o)_VP03oUr^7>dg$|`$j
zQ=Ll$SgEQYdXo3V;PK-~zuVx)QS!5Mq+2X#>W;c<sM{7>)nH5Um@;|V6ULEgw<qie
zhADD!aY0>Kx~AdY#Tl<&y=HrJ!_D<;n#Q1)3Traxnvfz@{$^|ZH!JKOKjGA9rdO}{
z`Q;_y=DI+)#j9*SO2_Tfk#LA0c^OgM*kVZe$;Y4Y@wdN&QijTEVvq|t*bE_IwPo1u
zdHv!wSFf*m@aT*bW8no@Qp|k$`R7zk;Xpv~Su0A?6pQCkMWq4jr0+$UCn-Tlzd*pu
zIpdS~vS}+aWs;{`El4r4*={&rpKy9|!gjwyE7)!}+*_Z@ZbRSs@bjcB?EtNciE$B?
zR837aHH`a#5QOw;jbX7~lXd3y`ZfJ-O9%lsj5NzNHi4;HOjQXV#o7ZZswx=z0vZ~a
zhKX^UxO{zuPMLi(;9M=OT5YJEWAL)rjN`}`Uw*;I-~KkvS%&=%TRGl+a1WyA)$N`f
zAoM#PoSYDvn(gPGqLP8-eX9C|lXb^(xnLRt!z2cw({9K0^A|k(^fSKq$;VW~o(HFE
zDwTQt<rf$~@tfcKHU`bhSC{<cXFq4LUi0+nBkrA_VoJ-~)h)}#lFf}o|F}v*sZ}wC
zK0Ue5m(O0Z-6cN!)?-{%Gx{yt-G&$=Rns08_e11Rrf@itSjf!Ygz6ZDC)tq|);dYW
zi0QD4QmSw<%L0W~*lbufH9!39GuGd}qJBJK+(LH7F*6PW+s!R8CX6j@ldrw%X71cu
z*zN83Ydo@oWYEfpnP~#69JOgM3Gx_-L&TSs-ZYNv`i*R&_Y-~sQ<bJiX{pZrAf@JR
z(VK<9taeOU*4mU);kwPwA-UbwRFe0a0^@W`RYA3MJbL$x=g*$A+f0moW^?1Q%|P&;
zy7surmm;sh7Uk%{XjV&MmmH)h@qP-4w=}M%SwJ?L-L7X9V2p;v6u8%&l1+y>IpzG}
z1Jd;s4;C$5x`pi?qQpW5;~0EGDT5kEz>>0I^j_?>m?PX~s}+UQ;(sa?CPnSZfu5F1
zE8IGRHWuqDs;Uu}ieib93)edZPYRwe?uk=RjscxBu4?EOE9z!}Yg&{p8|iW{4sj&q
zi5U?zr42jNhw|~HmX`Odip@t>${kCUuQ2k+F*8jQnMhS>7K^U5LuFl_g^d~o#cm2!
z&_K!#VOIa-NRAWHPsDK~dYPB0A4q<JBn}SvNMvM+5;EWlfG~$B_+cRQ1AZ9MUXEY%
z0gWkQKZntZ;3sysH-z&gF$lt$lp?C*z3A5+-^`$Rx#Gk*-uce=*uH$hiyw}}-A=e_
z`FInYd)1lDY_h0Idkdsc$|T--=Y2l<=-X6A;<!>QBtZk(NFmhk2X0@#X1m#Na<b&?
z^b}_;ZPl<J2A)5ELDGuT`}Yp{p2=#mQ3Ns}MN#n@g|?P#4YTV~5v_`8pyg<rqCio0
z0tNC-F=MPl<;42rl+EQ8+ua_r1jC=4oU+;NgzE+ePoD!0lf*^Z6lixgGma^tjS+j3
zAe;d2J(ICi#?h@#r1gqs-|rcBdqM&;j#O=jZ5L?g4oDG&qTm{df=NI{glSKVk-BQ}
z-cwbMarA7qx5cMNp`2k+FY&{~ejKosB`3JKx<V;9say1EO}jkd*^6gvzWj_^pZJ59
zuV@~;%kt?*xU&bSx?)k+>>CN3jv=B$;IlvfAxjir{_v;7tJiEUYEHU_Rb63};^nh1
z@jhVdnh!sEpNCH#<E-#Z`hHKp-?LnF2hLmH_xK`uX}bl^F1Wew`TVoj)UM;f{innj
z+3&aX<Bp89Q+0v-=ZiUWY}u$19(}a!5d&eT6?I)v*EJzZP<TjDOgu|2VxY9#k=VMn
zmMF65M^X%^DigCuWf=XycE2U1c*hg}?RY!>`i?sw$k#ufrF)?Ym(*&5(-xH^(Iriu
z90J)#atNf7!<}=IC_F7|ri8uE&srLGg=3fBc)jDbedK7}eY_w>#$mL_hmk4n@ySwG
zEz6UN^LwYnu%X{i_#rX&9y2FoF!Rz3T8qKGwWbKNO6S=qQY^vm88Q)TKs$r29Ik25
zwj-II&9sGX!Hyc$)pT9MaH|pjkr6eHs63G?OAIi21<In;Y|1RlPoCX5v>Xtni$5D$
zs6uyYm^n2XCEQ@Gv=nvLVO(|C1kZ(4Dtx+Bio58gP&7&vYnBB~(_vgKUTo-6)RoYI
zd^bB*io1C;d;mh!%&a>K7v}$te7NIWo7d_2UM%)A&@?qoT^CgE*a4XtVk%R-=Z>#Y
zD&n%ykL2iy(?pm?;xv)Qi5vndm5>Id$srIDBrjc`kkg$uY!-v<ek6xLr44<S4q?_a
zciT}n6>3WOp=TQQ__iTyOV+03(*7lK?5I0z*KzOLpYYE06~lhd<xhVq@kfV)OlgqL
z;mZd{7P**|*pO6B&2RnYZ?dQt#4!|8;z$e;Q-P}>$H1`JaC>#di|4OcEGklngwZn%
z1D}8PIm?qZ_wT<a)+Li?8b@^PFtL_mM~k%xC0~v#er1%@0gWOXfwodA&}l)^xYJ^e
zDHm>^7GuY{Vc6|CS*@_vF^;3i>y!NM%mJ7qszg033Ste1A~zKz&sZVnr%4twA3Q2q
z>Z+z&pQ5aVtJmyyTZW;>gaA1aQp9x&bkiIz6r~NO?8qV+r!(u56K<|<2qB=*)K!Dl
zmPlflCdSE&*O4=rWVyV)<oxVh3N7PEqZgh_Ot_}zer37m7QDK?<tIP-Grs?af5iHu
zkEtI#W*fjO%lX9x`;aiouuO^DAN_#KAOD1lG_qLMoHsSArlNI@MbofoTU=FPs){Kk
zw3P{+a^mLZhH)HOEf;w2MPW8hbY08&#Rbz8xxJbA{Ii$5dU?x7-~KjTyJQ$I*>1P^
zX`nW1s;W9Dtdu^gxn~QKtc#QS>~#gj<E|ib`iKgIAhrO`IU&x@VbwC}X`H2XmhH_A
z7h%E>4>ts#Ci-#zM!LmYRPOEg>oMjMF8{)Je&uf^*+}aoodstMRx1cmk|oOeqLRe#
z2@sdjR$;Y5Td}{A0QLF)QdExlNa@b=P5REeo7Z47BR>F26<<&>9rkKu410!PFiz2R
zCp>z3pK%Y-ZwPU~kD1ANv=+{?a0Uh8YE_4}rr2XhyEg~OT73CJ&SVTqS(FwxfVEaE
zmJ#JPT<$W#)VL^CB3ea~pBRG2_*?2}#Mp?+ij)+S7aq6OB0f>&j;4(+hW+4u6pu~H
z`c~pOXBMKGm7t|PZH>iRG4M6U0qRa+oTVi>O~tFulVZT2aL!UUHQjQBF&$YM@wd$-
z^J7+%X0c+Cx2g|D#~5`0QF1x&$8%XG*u2@Eqqh#?37LE;CHj6~yXk2bHH+nvrmo7p
zevXcqr^K;~He0D3?gdHEMLAxcIE@8mo5(((lBiiF&P)q>nSvZ}?NQTM#1jT<3<Nok
zIzio3Y<peYG?LiURF2iEqwg~@3=I1%Roh5WSBWXK^p1c1@y!g1og-BaeR|Hx`yX-s
z;srOCmyFYJ5a!MOYNZr9i)DnqbIFS{bLQgWg7@Bi7egkF6HZIf7QM$g1=Gmb?-_Sn
zu3o=pySe4jqxXp+pp)YI^))$0PFL%L*{re#n?&*2WO40QS~2uJIrnI*(Y89ORdnG1
z!odhQsiT~FhNfqia<U-?FU6TPjMKyyUwp};>*%@#dSTxnbqeRAOwgjtnx8u#evTvO
z!xb;SAm+6yMhHRjSl1_F^HMk5-dqv<#JJCxl*me<k(^Yl!zkGu8fU3=OB$e7iK^;}
zlgRVS%gTEyt-cr~RYRfhN^_wvXquY+?T)UhF)5*=$Al5*2mCPbbkShr8s~O=_UvcW
zF0<&)==GAy7^rLFG~uR1n-greELRnePfxjb*3vc=_s%bH#)_4I_gvjvfz@=2hAC!J
zh>YWSu$S>Z(=-iLCFgp-8@RsN^WxbxKl$-<s;cF|gNKm8dw<9dT-K5{AspN!^D^^B
zESwydQk3i2^?TkSib6ZYfcKtZ9BBM>xY@)K{G7=Wy^|H|u4773+{r^gX(P{(B7+}f
zj`6mndprI*jbA=n5BJTllz9o*r%X+zN*<F&8tu`_6VpJ911KY0FD3T~Wi*S$f~u<U
z-qQ~~F^5u|6nes)0#8sX4)^a2$NDpgr)k!STv~0yi_<7&K{+ypY4Ys0BP5H{4UZn5
zk|$4F2VTA0!T^{|B9OCi0m%_<6Ipd=*OCm9$*dHWQmCkj5q!=BEg*-(P$>nf0uvb7
zbJ<V);N=gwjv7_fY@Yp$Po6wsZ330gn5;2bQIV;DT4_`;WUWX#V|B#pbl|K(rpj89
zKvKldXQk9aMro|FSmm&$l1;S{FFCDs@oS?fGd(GJRPeZIB6%t5w9=S%fo*F{+d|!u
z%4-Tj7i2c%Vh;jH0*?djkPNx`JjkOn=phANl)@|D4MNbeF-KzyK$CL9hrs34#ICPc
zo-{0%6~bB4811mik+LPG;x{{!txKzTp1kssNw|bQ5{8L5jifL^noty<TcwGal8O=w
zwQGW?(UTN*s&Yu0Yy_(@wZ=7$h{tDzwGCRs+1Z+|F<e~^4BIVT*HBjm*IBg7X!DoY
zip|Q9TuoZ6SUh=$2jBS(ZZ2Q*#rBq{_wqC4K+P%;5Tt8{eH963GatPFn5Kd-?jb~`
z=!r3)tOS?G{hs~x72BIDUOs<D*R|ZgxR6u^Yx&~KSCa0qAMi?J>I!EaR?Eo_Xu>!#
z?t6Co4bwD<GO20`sk6q`wGca%Dxy_6voj%6t75g1MNfm$;#5sSh?tg7Km9pRo;<`D
zhf<=XH3$BNL0fn5Ji5dGJaW9G5=I$~i3aPeOsEhE(L*Q>*jnS-1xqCt-EMO$Mv&7)
zRy&A+rfE@i*|`}buc^Q~hXT58$q;u^%;a?3MUjQb>jN=(+O{R9%+2jJ>&2R!6MmXl
zG%dURmT|k~bkVYp9+GEi;a;uq8FGKa&3@uKZuxJ1@Mlc5W!UXl?nhRauQ?lgs+GYi
z*$wr359b#srJ2HrpF<{dVi@+MC}Ni7a>;77Dn4)-YYg6dw%aZHUF5}!4VSNP8HSO^
zPam;b)l7cRZogp|MnE`nrl_x!5(}IZBcULGdQO~}=X=>Psu@`=3t|kC8#$|{hkiiW
zEzU<=U6-AjqA`|b({UZapk))0H91E@&V-WfN`5n?<?VPo{^cLPqPUwkw0FOkA!KNB
zB26QVd!dMgh#v!C3|J!@>zE@ki>bW#`q1?d|G8paa!^O|3$HU5bqPs35Gs(MGa>lW
zhS3K;oOK!(4L&-2h=h<CN6%P|G+oEpy#}pUpnA5~w~!LaXQIh+U8mA6F73}!M+(g*
zfR0fD6H+Fkkj9QCDY)^G=R40ATj<@A#rc{%2AaCz(ftb^trxV)V|)U8Mw<<J+Te!?
ze8f0VHe;<64q8P*aOk8<$3c|>HoYOS%B2c+x{_A3b;1iXmYGy#D5z<Ytk@hT7$*o5
zDrZr3RyC$upz0dsTuC^L$LXOnx*({65|&s+Bj5_w9=d0T1@<t#^1-Zl?f~Wk32$Dj
zC50*H#N-2)HzQY9p5<E3XRAvon1%P8F-$5jCz_}`ZT%UP$~hgZOp>2Sek28d5ZP%}
z6s}rBGGH>9cxd;}prQu3UmZO0GBHPtQ^Xul$*^2FuCF3>Wzo8296O>P8TWf>g*P2k
z?i+9XzfobQb0#WHs2cR?1!o_A#JihocGs^-ZU^i+?4t5~u<nY_6-Fi2E60cLK49Ek
zqWnN(Yo<xKN_AV22*j~x+-<qJe9iV|!-EG8&_LHTq!`)VZs_WU&8sW?IAObns;Q-z
z%Ai8T?|Q~@&(!bP@As&zsOy?EOt_|{>RN15qnt(CdY(;pyoric39`HM8MJc*AIZwF
zSgk=@o_+ZoXB_7j7g%FUAo39=o?WZoK!vnM<>CcMnFL1Kyg5qRMvN192{X=CtY2Ky
zvUV0{D{ih{(~o;{iue#Pc`im1x~kB&5?>#yFiNvrtZ^7}jD+A1<o`^8o`Y~)G@vWT
zuHWG&Pi<?qw>Of~krTr(uv5ppTcabDN;F}_-`uclEhkSNvv-E(@qK>!;~&w#e94)T
z#Ei-J>_=EEq(~p82_G^cWD=Uy>I7wl*qn_9TQ=S);V0Q84}H%x4Qw`BZmvDoSCReR
zvshNV``&%pWoFmk(DxgB2-KA*TIcpS$Sy6$csMVza2zk88KujGVLLA=Ax6f@Gfi?V
zZZ0qJKG1f7rVXW?Y^bfHt2(x5as-1VDJ{o4L->ZKlW)h{@h|oGYL#{OA;h~*XHWJ$
zaokgr$w4S&DMHrAmf!3X8bc(ecuX7m+EyzE{?+jtcW4&OUAI&aBz@rI9b1qZs|_Aa
zj0x`p<1~@GNZmS?>yEKs5~hKaJRqjO>Oizd+I^BqS*RU}5R+IW=hDfF1|Kv-O!$!b
zavylvN3OgjcIPa*hBm^oZFqEkN~a^OGNd#yVrW;Vq&}0ULHKTy$0<>J3eP<dg2Y?u
zETV)=J`}=~5=+m&W9;!u8)+{W`jkotg((IIp6or*Pv|VsmeyJkNYo8DOHz^qa=iC*
zI5y1g7+L@arOh$Ak-3gL#*yAVhlhLS9ai1lR;+Bi!N<&I=h^Hc`!V8UVm3PCPC~`q
zuX#=#;kuby7o^1+1)0oA60QznjVzmVnM>t9p^h9*RX%)9sl3mbZX=WeKL+87TWOO^
zEM-mGcI>B-X$qLZ<HiW>n*k8N&iF=!ok1q+m=&t3sV?sG;9DPa{p<^J_YAl?UPYzK
zo-IkVQe>4;1fD!RrL7X1>z6E?#U~FKl2QzQ!sbLC_W1pt?ez^>!RhImm=bkWv%B4(
zGqlDsUf+^Jz>XuntpP<CC&GRp4HGGQOoCbq5~f2W?gxlTB+fZwVnwd*6l3#go|QRT
z0;#RDOw&|UsD|tnPoF;I=Rf%wIc2t+Eq2+V@3i)EE(qX(t01MER=hEZW$vqE?x@zJ
zG)j>d1QyR)fszZNZZjtt+(t1Bd*U=QA}wEC)!3MEjRs@o_!~>xc37na=__7*=+ZXV
z3X`&=euO~Vwoum)DS^rs9|Jxn#t@{vnTA7g)bB>Zc8B*{%vpzx38Uct!~0Z~;q^4}
zn@!F1<3H!)5AJhuZ$VpC)K!DgmY6iPagY;AXS5Q>?Ua?grx?rYj`aPGe!piN2e#WS
z+f62VF)n`pgC{(GdP-tL-(TWGk5-1JUKRuAg1|<Bk>^{4m{S)%$`MvQkarFpog_Rt
znF}LBKd^T@D%TLzD8<?sP?ZsnzpBG1M-ExkfM`<IWTD=^ZNa`B|C)@9!J~6em^S2L
zBch?8QPDsWib9Nq){r98I5JHWK7?XsJ0o&;TOab=^PdMr?EJ`G4sFVmph=W2-4fIx
z&B$7d@?tdX@eIR=_Y>AQ+O89m`f<m8p9oPhPW1tcGH9&a@IxktO!S#a2dqv9#K{E3
z-oxz(ulA9jZZfyg;8hE&Q<kn{*|eOr4HsR*^y*6tnGh8FKGHn8Kv$YDY#4T1DFAXr
zgei(lBqWUnOa@hojW}}?iz%?0gX)Wst}#YHfEkS{{Ilo-L@DahBoScFI#lJbO^a(6
zsLDyY`l0i1=gyXo0>U|}D=URE=4fwHs9SH$mQuVQk|quX({ZDHP_HUMNPWm`_L1#A
z(GQu?LrfZ@j>1BUF?Nya9-_MDNHuA}2BP<po{>gDE3<Ig1qoKDvX+Z|4~I>HbmEF2
zqr^q#xmeb6Y^|~MX%cgKEkYb^ELJPpZppCQ;zLA_6K?P*|EB2Y*BRfeuq)EzY&1&O
zSfgmq&w2c<Z!_s113x{U7>em(sjTOd?Tn(S;O6QDi&`Y++Gw=XM573J#<3^uwp?Gn
zCI-*LhZmfzR|i@2ez(JE%gJKJ&CNAwoCr$c$B}WGNIsB94?dD}Ksilqtneg50z6qM
zf*uO#GNTd)_v-_`SPDHORbKJ7nYkH4yuG+M=jzoJKmYU#-h1yIDUb?BK`C{IOZOEL
zY8JEF1D`8{RvB;*B%;Lxr1;1J3ug>T3o2);3TscOonpT=>~=jNL_$BJODvUA2IaC`
z#3e$jswxzj7)()t74yTwH?T?z*D*$%u_z-8lB+C4h?68GV3n9P=EM{Okf^mmk=bpx
z^p~&MZEqO%J3jmLXS7*R!o_;Q`yYM4a$(Ux@R5`?7O_Ea#-VbM;!bC#kg!S<5hLi^
zn;S6;4S{JGnMO}ejubLapFHAQ-+mwCBAfo2?fy#4r?oAfQ|@@Y%XyC>{sOmd9`i11
z-qn+(4I6?4g4dn!^n?<WF)5=&9!1KLI63O7EhuTB7$|8c<U1U`Up?NIes9NL|Di}&
zwq*aDFn$TsOX^CaBj{u>Ng@xYX%JU{abmyQ(+|Vp3Ye9iM%}?~=8gCPfw@Bm`U<t;
zu60IEiI6xbM|H6?a>3$5BIZOt_H1@L*4+uNu2^53FoX@5z<!_blb~)58m%o#89^SW
zj0)hACK*lE39T|F3k7Sl3%nQ<&!&ofcSbA)Va}PG&4!DUQx;u=A#wToIceO{?>FrC
z8@~0?dvt4qS+*=!OZNLc)8qk(wlYQ)hXip%nAQFv2m(0_UhW6hQPoY!er=>JJs+Ex
ziXm?xO(VW95n<V*w83>9Ro!Ar%1}(a$st!9Rjw)@i;_F2oigT)%<jAAaz39$YauFV
z=0g1N0|(SrKBK}1&9+wzLuT|E9}NkERY$v)L@1f=pi+6O@+?w_g+TBFVVVejB!`~p
z2RJmSEeeOyPPmy_m-H$Hb4)`nLeMPy^%w==%vNJ6p=hNHIVQZ1Xk#VnO&RL0rRqBI
zZk#+}-;-C}p#c4L$2ThMVon$$T1#X}R*Jk@Q9XSRrQQR;r-U-VG!b=X(uO{3g00X#
zVA4dMHdroLwheV_(2W5jlgWg@xZN{uwuJp2rxcGLKA@^CJ_Rs|{kUVXZcvpfHYFZ>
zf$m4S>8F@VMzPps<eYI%m&ZoZByCWo1p>`FGMDv<;?|v1v0REMG|6gQsz51v8%QBj
z>5=uxl4n<%m)AQU#E~59QoS1qvqlPBT_z+&c5e$0PL8GgEt^cVnq)0GSag<LQk@ej
zi!DG_iflB-DY9Lmtftai%%+DiV3QZ&idG~Y0EbbH@IRDLceOLPXi0k@fX6Xu5ho~Q
zx94)kXd@yNBtFZiAng#Z!E0>N*ksv<5ld!J8P{qmH}M;fPWk@t|JS_!@;P>zm{R07
zA6_u&3Qa|q6Wcs889}d1OjJn<<H<&PTiVl6IDtcdOTWEFB_sxDNThwnrpSW}!^hu#
z!0AcDe%i3vj0|HYCWSSU@tO-SBBv}nQ8h1Oxu~bb`*7CMYpu&pPW*^S5>oCbPmY=O
zdQDv&qvWzmWG!UfF$^f5s2WS{LHhxp61~a<4M|5T6OJl;tq$oe%nQxiqVDbZS7oS#
z(YFl!=cMU5j%ym%p~e7A<SZ?`9L3$hmvuZBgnUjFFvU(JWmyBXD|8>NQ0nN!pvneY
z+F*AJyaG<7Ov(YB^kHMCjVKT$#!S*K`f;RkqdZ^b=+;a6v1gb(X__D+b|y&!hFnBh
zBHYV(6G1Ct%1mSA_0`0St>$%X*w~gK8e+_azZGb0!Vi~R4T;PBhMY&n(6j4%ZZ;GB
zm|4^gYZ9gkbV^e-(mGVhV|0>&0}%$M90gI-pmic<=&G8!X>fImwl&JwgQbQ6au%+2
zoJOX-L{Ms=wl&pqfo)opb7XCVCwy>xmu|?M)2e0*t{6+#Lmj}91K3kG&36m*V+;K&
zg0_-UkqQs{t|B&PhzX_uF(}9ur7FmVgZB(sBEWRvXiCFMI*GYF7AMj)5~qPQO;RhQ
z04TIFC}T0U!5D|u7F6l*s8T6XQ0FX>X(^-wM_-V;Vs4dZyR%GHH|74JsVYa^bZE5~
z&&99<-wIdZ8;oyM*a;9c6B#p6C2_7^Evc(>IoM6jPzdf?Nn1^2jZVU2=vv492WKo#
zmej4owwmlR!6eccgd5jSY;JD|)5Lmx!s=wj?S9X4y~G&JFbyo14L*6Y7KO^p;jmhv
zjI=Ff!bl(7XoXP`;*2JWQJ^2>WLd4H=+WioUr;r)d~opw(zq-aZ&r$^z^cIQZpVwO
zh_TW_AkUE}!edAUN<Qp}=9m(Wn-Eoa14=2hvzX+F-W9RH>_U_z_m)h|iI^10XiT@H
z&6bobA?-*$kkkYQ*)2L`l9ydvRKju4N?~mU0gM(C&pf|}(p+#QYAwb<h>;jZVgesE
zeu|I+aoEz2J-Y3<_uwIeTVR$QRaX4&`yZjE%x=G@t~6Jb;fI@nQCS`-m?DhPqczeB
zPGUDQ0m109KvERE5mF?DOc)|@44AB0t{NV{bI$$yOO(#^{m3v(0>f)dT~&t^ky-Yi
zax4}BcaS<&_FQUi#g~&apQ|kV%Us-qV~AuYyD@DvJ_U3(SYw$cPuwMzOQ=$0yt&4n
zC9;K5I%Cx_((G*s_jdes8eeUl{o><a`9@9zf6MmfIc~V7&OO#Tk_tdT(o<y9n_`s6
zrWg)wI;8}OI(Ve!nVrmmD)H5;Xm0f#-%Hs%m$f}-NuN2;hnVwYH7YA&ko1OO8Zp|?
z)HQWeQ#B2CKa$c!4*AZSAe%9*Ie5j9A!S8~&<~O4H$9hQ%|^FOwjw1>4jz>wl}Rib
zjmZNsjD+zPohN*p=)GsZ_w4o~7i)?5%FHwyLyj7v#%M8u%?FS|qf!PHN}jm*rn#z;
zl%8Tfd~guSqTUbFh#v<+^g=_jma49)+YaR_G0vTZ2LcTiZk>Ab>+V8Y?8nND?u~hL
zw`Kn3Nz89_zGiROL(5p*u@p;5Y{}-BGHswFUCm?~5k}-=d$*kXBd<;ru-Q)sex0=I
z4ca)2tud~~n5uXe7JC+b+=!cEYnDr-TS`eb4bI}KnvkUUjR`_X7*~;_<dfT~LOX{J
znjAejiZOEj%S@ht`J9vsDT(q&8N+h90$_FjKCDLk^$nG^GO1IPn>Ue@iv<@?E~pkY
zy0IuH!jg=%|3V+xZAbR~h^=Z)&(Bz{*SxyDp>A4Yft8(glHgEw2Pr1W-OL;aph|1h
z%zH?bo4UMj!AAl)Oe8-LjI<H6awscRb)F}tDMZpN9d@k{Q(6>bQf!96>&-~jYNiB>
zJMBI=PTs>Lo(+EPj01sxUc{^g=TIq?i7iFhfrvUk4_TDLiv>v;CTE3v;U~hrry^q-
zLyRM)s<6&rtfPR$MCZ^cmux!+v3RXzDIN)4)LbzYvhIK~7Og8n^6a-;RGwI$U*Jyf
z5xY}fd~wNdeCr8Uqvr<r-uHf&T?*LBa{c-_e{i$q2Uj=z#<JyJ<5*V)n*%BZyjH}h
zm?lN`4&y{Xp|z7%N7)&M98eaPb;sHHIp-JW*vc_ZBmK}bOd~NxDp!kI+KPc_O0pYI
zDT=FHDmN4g&rTmRTcxd&@9Z#<Xrh$RXlZYZ<MEssbFe+p+7d^N@&=nSw^y&Ycz3`m
zN6ZdOjWsJ2Z?+@ej<@4qh4Iz<b4rnZd(HK$XRNv5q)AwvM4_mnY_<!*Kc#XnC<zL)
z3Q^xF>SlBCJ3!o<;|*HHn`qEo{_}x6e<;cX5YS0}XY?L7j2LBb&f@BdrtPSkk(9QA
zFc#`c&f$nuk;xHal&--rDX#XB&#w1eV;E|r1)DS~CY*|NMsd<wRM?aAK-LloHpPk2
zPwa<@&1OS-D889WIRXk5b3ubN&QQ8U9O^1VMtKp1RIb8RHO|%ISEntetU^YCN{Kj4
z_;F;KMid$AYFtxOwJok`N#%Z<QY`hjDm*iV#-Iu+^@fS`yoQ!ty1MiA%Y*@Qd3gBy
z%y}=*H3N1a9i4e?HqKz3!MQ?-nyc2r*~?n)zv@ua3%@!APaJ1!6EAAuoW$qQRv24j
zU5l++De|-t+Y=}StGvc>pHS{2DIMrg#u}JLtgXmCG6ji*^SN;MU4?cQMkUD@NxXUA
zjM4jx$2TkNlpsq7a^-wLbHdgY0LzE>IsL%!{7?TB?IVsz<%I*FUFONV_gG!DRIR42
zEQv(OVjB^sz|;@)yFH0avs_R$EfJW!<baOj1X*%iRg^Ya^iu4_6c2E(E-u;<ZmbGi
zD{^N+>=HR9*?CNcm=&t{jaeMZNI<Xj!sX;EoyhSBGbe2cNipH*Q)DwGrdTTYI|N?l
zNm&+}RFXPUN<~<~VS*pp0$QW26RrdE1eD*`WrF9Ng-4hm8iQ_XoE5I(w7Dhh_OR_y
z0c@)|um{cYd5o*bXiPKVhMHj<7{>`~26E0+t}1Av>{_g|gcupT1S~I4SFD!{x|36U
zwczuO=kphrRNwz0bRCbr_dUM#-uwJlfBc8M-fgh=A4}$54*Ya`!>|v0lr?K7is@(+
zA#1W%Xb-7Q<W~5ClKH2Z{75XhuqP)A?makXb#g+=aJ$>G>-WSc$JRM1>fl(Er8sjz
zN;dGFU}+A=Iv?&%!oxZKj4pf)vc%Vf5X7k1y2C=_OpVGGrbJ^iuWnutHqS9Dg;fij
zax`vPc9q8h@|&uk-j289ufqsF@a)r<?7qCFTSly_AsT5{74$8oa8$y^C>wAsZOwTD
zFH7K^Do$}&Z&5SX>v-)R#sQ>B9FyLUDA8S|t<qXJVAkM+Aa>eV>QK|Rj@9af7!;RR
zJNCCdlMjrOv`j1OKu3tumW)HdW7!OfFK#10yxMaeLA8#h+ECS}thyzQ%{122nt}ax
zM-C&6we0$daY*$0z_=fI@#S+qc+{|}TWVv`N)x6`A|Ci&91MU9r6fZr@mY)IimqEw
z)eX*7=r@$Si7@nx{XhsFtu>2PN8NT}hoT*^bSh#<Mb>{;@i`yXEX7`N;@6l%CkSRC
zgE}CpUmY|0c32BbL3l@o0XZLS=nCSNqL>F;>!>SBRT)VTy>q`50f`iKB(YyhDHH?Z
zkuVH|sh93h24^d(W>L6^HOdHLr;SD%RU%<kX|W#P;a&HIyPu0BK7(~rlbG1=_ne%Z
z6QVe6RJCNydh77Pv!6yPFKy;;GQKflCru>z04Wu_4x`bPJXL*m#{KU+VsrJH;ft4;
zDPZD6Q-NDnJbrqPv4PAa{psK#C6xE5X+jM>c^q(cL$g>=FBYUMhOWk1cDt>#6tmo@
zM=w5Pu@p(BsBLr+3up{vQ=Fm!t?-i~r->LQVhkh_I%bT5WF()>C{gariXt+oAguzW
zOp$9_V%9_~V^R!JlhDTk?JFqU-8|a6Uw09oyAu6mO(`m8A0Q4Qh<O3jS^U_h7$i``
z8Qf~gQfsU-_}!jyyC+1bVxU6!5y~7nXS#Bjx}sYy7{{Kz@5_QKsGTXUpL0=TjiPau
z&NZ}+7%WC@(Dj1b7q57Av*&zwja7j^`=@_M{MnC4ZNv8Fij})h(vIv}CR_9D*%!3?
zTh^yFZJo)cWC&)5^A4Y+=(kFvl|?BlyLF?e+nVLslEujqtYPp2+x?DVngsf`k|S$v
zv4||z2?DsvIYxmBN1lh~7mF*UG&xIKNE<`toNy;XlzFEVld<f!Tbic9YC}vxtPtWv
zTPyPI6}#6zp=uk9Rald-m3?zTd^_Hbzoz3a+K^aN@$ljUUcUG_jqY(-W2n)}7Yc-^
z1pP4KCy)1moW$~I?tW-hQYO^=nqTui%9k)I)EnO~eUrm=o(e5SF?UzqM3l7Fge+E4
z<2bVK9j@vywx(OGxH!Mhe)AdA<mvmKez#+J+JP?OIGqUIvl%_luX}#**(E=|+_O)K
zwOa9Tal&_h`)_b|vL^dIH_v~<^!n$-ps+Z)s-e#j6)hGcLb=_B%a<={%!-S%4uuf|
z^{h!rctc931J5w#0(wND^fYx%UDr5UVU;UhYmj3k2G2N-!bh(ws>af^Ev9Ke8)8!6
zW!)nQL?TD6xIQrVa}Nib+30;fFG8&|zlgB@(z&D#f9Co5lsfp^&67Yj?2=huSBA<V
zYUC73@kVi&NF*ca8#?7eFA_dooJNu#k#r2X|6A8mH67YEXjh|LEgwQ>vd#pROG{YV
zzA2YkH#=p|bMnx$7LjIMw}b$@+Z`uo;A4{Grww(tz>fnyWMb4PcLw_NUw**8W_&Y)
z;yYeAsvJxo0?KNzQhnRyiuDKYbN=!<U*2p<cN8TT7bl$EJELv{DG6R+My0^_d;E4o
z>^ImnVipVP<pSH(!~~1w8f`88FyN{Rr43bElX3}%hXXou7adc1?y=8pV}-Jcamc6?
z$te<s5fc?YXRH#-!$?LMTR3eBZB%Kqg#v?@*U*}_TY?^$!oU>5!L{33OR|w%icVd;
zdxUrKi=2YHe1nv-M6MJ9{{d#6qfy9Evi~wsCY2K5hc$+Fxumv^UKzH#E#ue|k|&XA
z7Ykg~;_nufuEJIl3A5i1xT>LU7dS2UD^lELbgE@1Y8={D@-D2#wjGnJdG_*(%@`SS
zPn8rpMi!<f8^@D%$8NtTBu!{KE{UACOFsSlHBH>}@S>rs8wzVek}Gt^M~|}=DQ6}h
znS$&{ZC$fGS>f86$$NIgp3!@J@InrD_Q1)BA%d1UD+{^E%t^(VI2{-K14mM8tWwf)
zNLixHQlhRZvQA9XbU?0S2>5ZNE75L>gi-;;K&7BoJug4~K9A1tVNYAK8i{HY1cAAD
zOm9)Ux8twDP+wD2&1=#x`3-HT+j}e*_o>t+Dv#({O-djqC5m$&2va2b3=y=Oi>`#&
z6R87FDx*kwj+zjfjF=;fB?SmqPR*4@y2HsT?>7}lt2EHiSvng!m#}!9NIBxe#4z<N
zf}=8)+O?dlFL=@35aYl!P7M3K7$BES+ZYmiAGzEHUT#KS_YrS9y2S+#-+7OZzxO>p
z{Ny`SPO-m!$?mFRKSts0<;1|ml%TSjMXjLCXhwFs8+Kd8dRb9b_Ha)@i@k;v7Ugp$
zh#{3C*%p&_cgVIC@oFyC91=OEjkKMsO5(T<H54E`HSu#(1$jHRSP@IEQY|Je&wVJG
zp&)lz)?)G<WJ?{3@dKjDq0m0QhCZC#!%t;MF_WU4!>U$jod|Jazf&yjs@zA@F|Rhy
z=&l!DotI>ZlmZeFXRy}c>JIA~l#!aF>@dn)EU(RMT$K;q!MV#N);xEvk@7jL6CqGa
zu|$|9@d!>DWm~dY5H<K@@S!T=oL_(ZGa+_|4^$xZtTmKgv#u(Zj~?;hJ0CONZ20US
ze;;Ejlu4XFd`L7E`=n^P1%5w~m0_HE_PvMTMPi)ASP)~W+Lp`f8>+S?WyN;4=j`N^
zx^AiKhN@ezEW9vXD)~cSUP&)Cr{d)kBOzFnw_pusw1lC@Pa`S@;?Oge&u9;bjuF_m
zs9)v<O>2!QDHlys;jAfHx8V)cjeHleEAL4km0EAKXiFb`ZqLRtDMad>%jXko4TZEP
zNb1fcO`^WFpq!)XN_>_zr2Q5jJlorw5?@wfoh%9|CY-BDRZUgbELKaDR&-rg+V1k1
zbI#>86G<DCbz<ykHMXv~+U|M%>Wa<2rw<cOHRLehw^!6U^PoH7`E5_|fvU5Z$`Nwn
zSqOy79jd#}qs~yL48fDE0VAQ=F-eS92$6A$G;PJ%#RW~<5p!nO@44M>OWVX8ICV1w
zKYI(kkq4+@c9MHTmBsP%LgCJFgz`fOsKlYg>&phkS$p8vDJ90u3QMINdYX83`8j_1
zBQ8E(;ge=N+!9m59cRneB6x2Ly?>p?8@H@qGXCQEaGXb9Z*Kb4&e{As^7p&1|Epe0
z9q$KsFtM-yJ%e(mEY=Tk;|O`eC@D6S(xm{A!Y)pUGzk(^Fm&YhnLrgDYf2<U96+AM
zW%8<XoR>}pg*mKW^XFFFvEfomV#elDP$hIQ=#;S9$YU`RqGy;ggRil)Sktmxbu7*{
zY_4A5C(p3mF%A>DvJxRSC9eCxv(3oMJ~Ful=a0|1|MWvX{^U3K8-M5Ta`xaoZf~A3
z>|e_!cJCo(LJ(Tgm<?U4sgkBnnZ^Z%e#3Ta=!Yd$-HK0c&Lo{tDPaz0qHxcn+%KFf
zIj~v^yV;*rlgTR0oFiOi#bn)9Vh-(Nsj12lF6vJyawx6e;wE53jhheFapI;>4!E2I
zolEUj#Civ8RSD;C?wVvl9<z$#8lmzVrp8LP2I7(s5rr<9x=|?YnMR45wABh#JB-TY
z7$`<Ch?mMp4&y=nDhWgm>l&)6!L=RQRiF)7=^|p#hbcZ|mSS)}M_3e&Wtq1}_>#kg
zXRN_k4U|mUl!z)b`b;(pa+S!|5%Z2|vZVCQ*x27NQtVjNiISw?4JGnQ-n_xCRxF=B
zX8pmtG@rj_c>bEKJg!^tM?d(KwyAjU-UH%x&;9ciU9%>+$dm^*`<~ib>aN396;0Q&
z`|>%?RgBZb=sk<Y5~VdhCzLf9=ML{tt5W1pAkv&rITCY*XdqaWvl#1e*5S2KN7FQj
zhm8Vn4QL_iE9c0{0Ewt2xHp-Ewqo4tjH0m`t0cTys(v}aGf&{FB;LC!p0AQf?+Pu{
zOytkV0;S88O1V_}^NvR>QG_cdu(QvZXf$*+UF)d&n(b!GZnJ0D?=jY4n+{b~kicq7
zb~U+MqR{MjdwfVNy0-L%rEt`RBaxNDm<s0voysZk{L5!tzrKPXDG*&7xp#5GejM0e
z-|*<-5jO#<^Alcg_UtY<+`RmP-SrjMI&#_v&Vt5~$kCHphq6hO(7T>~=<y+tGOX4o
zoSogr)fJPMjM-rt2{B5G^N#stP9pPH<-HdWujCCLnm2L|(wlDQcS+WFTjm<s*cheB
zS|s=>NP9emh;f#rMD=8orJ`fH>3Q|#HR|k^ovC=exhBMLBysb$h<iJJo$;&Dv0wP<
zU%aopfw;|x%G@gZ>J<E{a`6|Q6AF@ceEe%DZcIUetH6t&7{o$um)2bLfjDK72fP#s
zDH@cK!a_SiWfEBwnUsw#@?|lYlrT9H$YfnM#rh~@le+bI%big}gKYL>ZPw&$$xIk$
z@haj|V(&Bk&A?fkX`2SSIOD;0&TwD+h~erPH&<75FE`Yy6)9_OZ%2Og;+B`AMV&t6
zZ~Ys;&2RnPe~+_!4_U6xS*^~=s^YUx|Ag)41zoLpF(h$lFq)V(?Y(<ENCvg_Y^&FV
zoQP>+v)}Ui>WZdoOKUO_l_&!xeF$<AYbr@YQfO<jw#8~!%=lq8Y)(0fPpj2MTv4Hw
z73YOm3dZ8pp-d%3&EdW_yGazkPi0F1Ed`V+DGjpzDu|M8n^XZJl9A$Wo-=p57E*0f
zP9~^?Dhg_?j)mQ!%}S=Sk(@@hyFIVI?1{l}et*q!*+_~8l0cGE5TtJE$x)7#))v=v
zxO$1J8jP!?PAm8BkP$CjE!Ty%4z2IduaGWaDv@W$SXKFd&Jc1gAqN^~7157^-l4g=
z*>U>t31{^gi^a&by`-`&&N{w{F7}(NV(*T5$1~q80j&%;hdo(y^3GH4eQ?F++X0nE
z&h9^C2Rwe~BcA{4Q~vn({~KBdzx`Vu^5C>)eR3aryT$igmTk*owIrk2_dWNQOSao>
z@eQjm#u8>y<&1ij>OYo1-h@J{M2eGe@(@!*=US4nB&`pd4PS`6IYh?3C!~x`MNn>p
zv>Su8o}D&iD@9|Dp4usDqsSo<ycqs!C`jQkgY<ydDGr4k9E~*dR~BpWT$77K8!d|J
zbcf`7r(o2G$C}J7ss5u;;4IEqoJuUU#u&?fH!uw&D2vqwt0gYYIY(j93V(4Nsj3QR
zEyjpRa86<iV~i=gZD~`!dj6a*Km81o6YHkI#2OVU?wvm5^A}&>lO|z#{K2>Q&forf
zTy6Gz`u%^x|M~y&eXhoVo2jS%8C;l}ce@Jb0>MlLskCf&w`@1JjFYFTTOK`nLRB}!
zl-TZjcKse7Bm;2fbttXNu2ds_MnwwGyr>lu%&)me9xv24Zwhntn9@qpE6VxvLtvUF
z#$mu&OJi+mcQj}k8ndRW;GGYDhvwl2{Nc}j#?59g^Z@nd>(1x)ZHf0Ue*D9K7+Eg=
zD&N6hgx1|1ziftl^*QqKyT7z1{3i45h|Yce(`B0G&F{3!=lrdI@E@VV9+gH+4mhW=
z3EGsnD60DykLp<9plDPDv4MzE*J!iBRnHDI8<1PjIz^T#77o^(>ho6|*Eev+oDYiI
z!=|p>kh79v!sySAhQv8bmnn*h($zmgX~WAOH2m3rcEZ)xW7dy(_};hqn}7G;<>7~)
zV5<e%*5bgB6ZbF9snd?zIH0X%xjH51iS2&J>hWXFy(PJUG*rYC7=6UYiQC&7&d$yx
zBYYNg!C|8~cOQ%?#cOc~Fm}!^Ek9pcwbGriXyXunHbpL8t_n@c#Xbb-IL&3f@YrQN
zH2E%nZr(JP4i(a^D)rf6qan|i3t*6eAYOA6hvKYC@q{3P#TPJ@`;>~hWc(&b6t$|w
zXhR4+w>LZ5MMYB^2$F!3Qy@l97)SAj%^9VPWXUfUST}pssdA4fcijVQrVj<^-8K8o
z%OID{(wkRF&H_3)=U6N{MnAAvoM5YttQr>UQ&h}UTEo<0b%nM1FR!qZ;#qMBNGTAW
zR<mGQs(6jV=$h{IoRjyT@bdB%t#jP}@Y}e}mh0;c+x@`32amCNV(b$?zWjnG=PP3F
zpt7XJ8D@Qf*P3CRutsyc+t76lwsI(E#Z*w>tUGT~m0hPC@R;*K{yqENh{0jTT8C*D
z5`wG^!?4Fs6Uk2k&8v(t3S9vhF$l&JvVvgQUTx{GuW_bhS;K9FDP)NoNz%5AW%4Vf
z$p@2M$>dWK(G05WZItBAC8R=AC9ozHcs*$dDNytv<g73uqT&pVCsay=7$F%<(hwuw
zC*tU_+OSwGiD6a;Ny|`*#!SYdV?vum(=71Q$k>nAg>Vj1;Y~y`#woOwR<#d-XV0E9
zY_GUzMaXedDYgoJ{NtaI@VM@TtCyEtJbb|Ne96U!-{RXJzKdQ~{O|tL{|)1-mwYh=
z{^`qC)K4F9q7s$QBqQp${_2)~vm<HC!*}23!Gm`&$}@(ZeSgg~4&>yi3pYe-jkc17
zn99!1B!$k!4~tBUN`yNYO!0)8^oUoPq%|2sDjbh&5JgzAHApFvve@mU2stSH5SiSB
zcN2Bpl9i*WYb=nH;^f{ttR8&G>ripM9k{-{5*PHt)pfjK=6vT{hY9`*P>MX3^DjL1
z%ei`Af7Abqf9|W}U(K@f_kIu7>%Yo(@XzD8{<A#CH~pP&_VdPg)@MKc`p3S}>y#Tm
z0T_Sbr~k#62mRvf9>)Mc{ycKIt=SH6@#K_?hwt#<$p^IU37Cqk9AF@oZ0>TcXEakm
zo}$v6UfiSI_V}U3IfKcP(G7{okHRl^Vq22sCPK`ygxPbTD9uh0M(ctiA?rpeeC{ko
zD@KbDAeRPs%EB{OidiH<@rG4%dsW@}egP<CMLGgL^F~_D=vCI3g2ElztmU`zf&0U;
zFj4|^pf#!Zy3QNzlnRm?#T;5IgV7ePqzyWb9zO*_i0GKnh-Y1nLE5015>()Db%k|R
z@jMk?{#>7mU{pTLe0-0aieJ87zR~@m@VpD>zHM8su5K|#W36Ef5v{~LJLibCMacP0
z_IrF&#4g|Y02y+~Ql%S>&T@Hbt#P`-XNz_f^~DMIzViW39z2BA8as}hI>VX1;MxVW
zAK6{KhH2vL{R;-$;A+e2@gr`A9sSiSG74u5{eHv!2lv6im_kWKnBQ9}t>F?8m5z2I
z*48BhHIkA?Wy!ELT9aK>oVP7eTOu2B6wY1tJ$V$ETN#U0mZT=87}?%<{`3!jz;ZdU
zJW13h5J`+4@j!~w-V7zqL|(&?XXC^fkx^Qt%;nk;?M#-UG6{)SczU_0G6jW7O3VA2
zTZ^gi|8n`A8QS*J8q7%-*KAGGM0Wk2jQB@EIJ9G|t;kuSvMdBSihFh}ujPp|tq)36
z%8Sb@u5PbsT%@fu7R}k?6`$UF#Q3vkEINmZifJ=YnaCgg;eX9P_#gkzG3P7({>lH4
z@BZ#@^ZaK&;ZOd{|AN<l`X~I!%U68&e2u*s>CRiOFK@Ydd4ms%lXpL4{n00k%>ssN
zUT!yBk3Bhw(W9}39E&vz7Ng{Xwbl`R6!Prs9;y|}R^n-sL_R+k{yHnHGE7-SHA*XK
z9>pXvYdWI5hHTME{07qm(PxsE#X`?JzU=X7qFOD<i#7WxbF=Naxw@hFS>NF{9nohB
zT5!C%={IY<xu8>X<k=Bv09nuvb>Cr?Ec}o8dtY0?sbBK_yCdCsUvJ0T@pjxDJ|?bq
z6Wh`9-q{1rAH7R=`VdS*)`E!5%h1^O^d%Zh8^?a!5wj<0&FTGzxO_>xwP+`-GM6Mj
z+sBFBZb#d8R8E%|RB1A+yVv_2x{sOdwQNMW31TQFytAdTL0f;M2Du>J2~%3)rF2#Y
z^Q@AxF}GQZQSaR9o<$W0)L3N+3rH4|trScul9jB>%1B%Iu(sulPWh1Qu0kZ)38HYO
zl|$${%K^La(w^12fK38E#ZbHkgV@kyN`h02MXjo^roz=?K5d-Dj|$zXtfO}~+%xNJ
zMy=JG|Gzulr1a!N+Jq`yxTb4~FE5$%)23-6`%xmz!YEEO>YJ~$|IDq}H+;U-oBO0(
z$g610^5hh&6i@Em2c<A%Z1z-6;npoTFE`|I&qwb);oixDvC*{4j*G{S2;;!#{hoK;
zd5Tivg|u3&(Atp5g_QYbJ0RaN@;gSAeB}h7N=s9|HxupA8KolqPJC_fqVBX2Y*nKT
zByI3M^Xk<V&z`;JyWhJHO5?19C~;FU22?6uJgU5pJ60^h_f3bEtOzv3Pe^53E_mT4
z805pvZCyixSdMRj+>;X1#gt{DDqXmC<bo^}(`JOvVqJytULv%jkY2SB!)2p2%F6X=
zv?|4L#OHvucXw=3nD)cKt5>g>rje#`IOj0g^6=?n%!i-w!_S`6td^vyW<v4w@e{^H
zM+$+uah#pqqjELhdiQ-keE5j})BpMz+vlJ2!_U9OJ$ZyT@cN4vOx>0z-+s#9_z(Vj
z1h?So*-Ku0_L|S~fLS^gNebD@IT}|{X-(}MjgeTkz`$O6Y11o((MnP*(lMZ0=S&we
zI%eASE%hR!QpReCBwXH8KMm+8irp|NrXiwzh9RR>rd>wr)kJkV(40=_l>#D-?y>eI
zw)#0$iZu1-+`GT0s=-?L=%dUufmZoWd+Eqym7RCyKoF#~gAvtk&WFu$X>rUgkokHo
zp4ciKin?D;=lf*^-@n}B#S7SO|Ek}`+wpe5<Rh2216x*HJo%6Z?|ci~5^$&kMk#!+
zFFyN}>zAKWlX?2VN4&iGoc*{#8@Tu63H#lrVo~L!+vBoDP11?kZFj6zYiYGA5pQLR
zYPA$y#u&`pB`TM7`^o1l{;faxD>^#={KpA!@&nEP@Bgzr*mFs%Hse&u;*E(MJRuE`
zBckqASXWVZofLJp769@<4Zluy0ftCoYj(G2o4JHaAN>`7)p*0+49fY?g`=+}S~Nw1
zAMk@`cYB4h6<J%dvrKW;0ss2rzZJ(W-65182(}8cF7U0^n950d#Bxa(C!Ez-A5lb9
z_T(^dx^B6Du}14i+Zb}1NPfhS@qXaq?2Oyj&uJR*;E5>_QY;|Xo!7qee(oOB7)59x
zH}SchM!6Zzwk3+HsZcCPDKKP-R#7RS@hD}8QIV`7sl;yA1BykrBxP`>LIK_frVzw3
zCCyeAQaBl7L_U5{4GHY5lyDAm2AxeoKNVSdyed1dWJ=rF9Ew?KITy%W31M?C1Co*~
zRFv4X;3FXgf#r?HRg!s|lN5@k<S!~E9EGf-aO&oYL6y&=HA$lXuvT+*b;Wjji&Bxg
z5&3;ihHACo!8=d*VO_B}JEOmaI7~Ek#XBE<KpH1>U6C*}%@S2LeEYY4kKg{=|0e&{
z|Mfq^#KcdYKj-AO;r8WA{-^)>p6>n+QR;sM3hw93CkI69&7r>bU;hl~l2UYc{`oLx
zKECs;hiAO|)M`Sp0bafF@@E--4~0_yZ!+*d`Jc&!J`<Av_>b?t@UL>C%)jLCf6>qW
zy1(;Hf9_wN@gM$&@E`xj|9qaS@8R$NeZw^U^Lf5+I_}O|S)K}=qx65278R&*V#3`t
zH5d2p@$NhC^3FR?dH=n4dFQ?NI9;#Bx6C-KvFLJB(OROY(4{5#wKooFd!`f*DJ$FC
z4g2km-Db<n=g;}<i_dxX{5j8Gyx_|(zU1ci7Vkapy%+f3{-;M)(Cw?1|KflDJ5<wi
zPMXZw(sA$nl&Wqe>7d`UySnC!FE-p>dv?2sAHh0JRe`HBRh`gPrfq9ZRx93l=Ur5{
zrmtFJcS3#tBb2*<)DWX#m=wG1z}N?dA+p;~Y_}8JzGs*Qd>nc7^dVhWQ`HV<YJT+n
zpY!~)muzo(a?oh3Y@TDpPXi_-Dr-?r5j*%d|Nd{&saIS)_&(qHyZ_HSCs2VloLxNP
z!Q=NiIe!Go0<Iiuag<Pn%<p5sPadMiqfvE3x9sRvYo@x!IEQl;)+BV+C_Fh&OkrT(
zZ&6xNR}CQrjLL$i>EoBy66vWZ9fEWxwc}rrp%iGOD;HyN!HAK{S+s_fJvs(b@<^#9
zZQH6A1*|dEA;IF<K`Hl`J3D7nin?EBunb3}?#=h7ih=YUQ+hIW>rmpDOw(Aln`=(*
zJ*8fql2wBry_gn%gJbdE3bFfI(x4K=E-6hy0(iTw#W_c<HGZ+c>?6|{*z5)(num`b
zQ!m#{VI(5~?YCD~R8BMO_oTMs_WGLp7x!__F@{J=86SdVni})91>OAKthIy?B(VA5
zXsr%KPfU@dGj-)~bxpfmLe7jsFUVdD#Bl_x2r+|pBCn3cw+5)I8mlyOj>;^{HpN#@
zYh5yI3lBlfBvjBUf)Z~Xqhu1I6oZ+VvX{QT6=*r93^Ac|#uVhqB;m*jVpovJf=H&)
zp3NBoMGT@PRxD;w5)ow~5QUG}yF^tu8Ajua0W?|!DB9JK72`N^eSIytt5#v1##o41
zQ?Hgh`OYW&<I_|A>}Q{Hws=U^h8Hhh^6AwT|K9)LdnjemMq!j=8hcz@@jHL_-{wF6
zKmH#`!vtRQ^5rE9Tdsw)qTl{o{P7Qd%I)=z*4D%nakXW+Ua(%RsH=*$a<om2)8bjE
zG!jHT1@J<c9rt^tX_RvtMnaHmz?m&-D=RiGTBB2f7@5YA>zkY6il0S!mW5*{ZGS~m
z8=Bfuw>7qHF<KErU^|R#LS}jIG1p_@^A|69b#=pk_rDKy1ONB`oA5U@9NHyiBA2#?
z%x4j;2sdM~SkSgDP%Kxbkrd-0;wZ!*xsOFok1D3D<ip0Myu??o?SIx);jd=A_a6K^
z|ISgE|IcHj1b_I4FphsUcRzLKFUP{Jcq>{73qL(s@!;WoPEO9bfB!!3z4t!vJbj0U
z4<B&v{zI0Fht%~#P;@O65Ur2CV8WBOFt^g)7&8SRA%ca3F`l~VA&+R?les~;S9HtQ
ztX8jR+^2l{*{57yz9xoO!1oW&mV##UWsB_%x&^ncSj7dpTaf%j+-atlBio;kT)l$Z
z8^y2(=QMSbshdpGf^IW@VX&tau{mL?SB%RGQhE$0zs-;SY|pq=_(8GX2X-5nM$I%T
z_C4$eML$7O70sezQGZBZ%i7hr1-ZY+>(76{i_c$^CNT;YFd#6DJ5&nPMx!c7%go2W
zaSuCe$zJiDube?;3=bbaVRd$os$Ibmia0hav!|*p+8Q>uxA;wFH%-)C#o5IJY%RI4
z+DHbqF$rs>FiSDvr@%1w2NdXV=H{LML9F+cQ~R&rPztOAMIvWOqRKhpDvQ#Pb0lX;
z!ICqe(Z*t3h1lqbVR6c#@Y<vRQ+GB!$9S^i_-Y;W)!OK8f`&8@HA>|}hq|gN`Ad%C
z{!v=vS}QOXlszOA{kr3ui6`{&ji;jKJGMLL_b#o^lrvaE-E^!L9lqbQw3ez{V63Am
z4&KA<6(^@FVr}qBQB?+K<nuiL;tO<^p!pOdO;gG4!)TnU$SI0zcFuei4*lxCrNwB;
zO5u~2mNrKbLC%pdMXWJ2O^Z>6oM7DVQ7TDBYC<eQqLwUIH2pBqEm~3-q!riHm{p*<
zKcL3(a6>g!FQN}C#com-*)_-{(<R>O4mknMfX$Iy9Ieyb?~fvqI7rsB*pOH)=?tJs
zw(D_{siJnZZjhZy5EZZw9)&1$OjVJ*#Xnn#cbh3FVX@>$Ng~)%@Kklf&CLyczaynU
z)6_I|g*K3*CTUG~ddm9YL-ujxH-7hTvK>a0*8JqhKNN@W;83nYS&P$4Vi<Y!;kSA6
z;kWtmKl(2*wIPoq_dk4$HZSnQLtgy&KjJ_6PyQeL`5%7HBGr%+JtOt1<^1%Nx^}FV
z9rw<bEEkrC4^KF|ShA=cE4QGl8mz5wN>Qnd#$(4VVe^V%dqcdwWPkmNzQ3lii7rf>
z#)$1lE?>UnkN)YO5T-yHBg3Q!p`!1_zp!Zx_s(jbJX&+{@SNqN$JDi9yS?SJSJ(XL
zdf@as{~f-6HSquXNB@{V`tw`<;qPZI9>D+de*x17R;hykq;VCi)rynzGakM3gm>S2
zkB1K*a_|0qnn#Z~UBAQn^t1?MJlo9;yWNiD2Znyf?d>hw?UtLH8#!0Qz^ng1d;b}1
z>y_SjV!svMw8P2g+z8-a-~wo*2D*XnM(1X?L{bzf4MkZ)OQcvXyHw+;@`ou;kW{uM
zyCiEoHJX_ZW0hRCN>e3Q4M$XoHCA8{DT?ge>;#~Z!_DF3v(uYa@`v^Aea^i=vm4~t
zqJ~_kgM05ud+&FzcdcjrpMN;o=bbB;xqSI5qtOU!k5~D5j(JQr`FAJc>}P%^7Iy!0
z8S_0E=6iF8xM)pY<n#vxt1C-9c<~~SJ^le6c;EppT)4=C4?f89(h6CgMMR#21>yZ&
zW4_pYKR)IB^RiQSLy{EqdQ0#69c|N4O=s-f-sX)rU+2d4YY6cwCks14Adv<_i}zri
z!gk3;2;u>6j)I6NTyj2uiC>cnl0*<(z<F>%;Jsv4T3ntYSR%M(zVXdB7;QJS6GvTE
zR8>i^67L1p#UutVJw-p~zKds2A;p;%Zw=D6oZdLimE<PD9^l*VxrwNZ#Ca?R;RDio
zXdTsf4|{ZjUVrCZmrqH>x%(bqu(Xa!3XY{6^K0)Qv=qIZyq9x$5Exe_1{n4RY@Rt6
zGfJh7VgQ+-q>I8i4NiO7wx+5}q?Dvdf=XhcezK^$>pk{ypAq*S=8PgV3CEn=`SU#a
z$DIBnI@GTdSNQm~xmaV4b_GGC7m2WV23>QExxulNAXN7}=K=~bFN)^R{hsH#P)rA8
z?-)`S0+Li9MD)S+A-W_aX#^0W2c1O-7aP?=AQO$$34!SOXN`prLkN_L*YRZk<YZY9
zL6k0)>cm=kx9!jI-^KhFEDF20FK21mC9QapRML9i3G^&TgH+iMZ1>(feqV^)-Bu8D
zH}Clu<L)4kG4%mSmSkXTOozw{RyIx{f+I9F&2&PMNW81@-q7pkh+ruXZZbaFXQf}D
zWrVJ?EQ>y2^A@U<oyEs-G}lRQxVy%p<r>=%qML=sUNhjWrL{53O(h9=FLJ@WX^;eL
zTZ0rh3q=w0;+nRm=%qN<An_D=O7HO_*8cLhe?OP@-}-2;FLUSbei|iZ^ug=g+Cyi(
z;)4fosO!LJbVO^Kt`(G%rdeE5bb?GYI?Xuyzy)q>-{kFEw~#7fnw9hiecHAr?e)47
zj37uw5RPJXou{Aw1ed@44bmdPYr$bz;=H4omCTNgSjqY<$pYPK#?uKRNeMQktOZB5
z<R&AM4ALU-3;(O1W!6p^H<qtld5afb_#U~=Dbj*d8yl>zF0r+-%=-EkYv(Sp`f$Rz
zp2GPtX8#&H-~9&7&JG%bZ*6A{AaG7$t)MYbo4Cp75V?;kQ6#C3qBkL;_mQ`RddlR^
zHB2dRRmogdC4fw8@}%I@))pW9;8Q&N?6W-nfycP-{`*+l*q}G;Q}lYBCsl08iFp8Z
zcX)`se1SIxXKaVebss*u*-vH@M&l80zV#+w`r?=P*0;aS8?V2?Y&NBBO+=)LMx2p|
zEV|T_ZqfR;H8KB)3~}<KqqAx)$+DdD=g;$@XP)5$A9$3<AAgL;AAg*!txeLb2coko
zI)S|X;H}^fiblqtPtu%ZEoW(Mol8$VPG}9d?*aet_&WvUDO9eZZJ}-yt@C6)L~j78
zx)?}BK*ximHF%=;y;9J$0klUuFhStG;OL;n27_s9#&>S=-LGHAO%jyXNY!?##yF=V
zH`^m+pjRZEKYyCCIbvmFi8M>8r-pGer3?*?wzSGI4J}d$oV1j#!kaOos$+6S+3?Pr
zmUVRi=gW6p86_1DJ@Od+r6rJxWBu>^^loWPOVRIf{(<|s@4g4P(!b4MIcIHSh0|wG
z5klm6X_b)b6oDhCc*s|YLn<(~rfwk3Bt6wf>5MtGJU|zJsiGAb++`;-AMG3GslNPF
z>`GZjaB<DRJ3?1PJ1;S9%50X?Oajf3<7n4#IEBV2jMI!;*+pw=g3tsnQ7TX<=<C2B
zaSRj3Flia)mf?zEaNiQL=OC+*$pn>JwC-$L^c)1~E}OHswo74Jov`+^U%z{f6$+_i
z?yz+Zr6N5@3P<oQ2;0>T3aJy)JS9kpcOljmoq=%_)XMml&E>Bh6gsDLEl}?hk3YXB
z<_EI*SO8$%JdiRVLhSrRJFb9~kfoZkX$ii;xmoPIv||SL_flfNd;Eboc5}pT0q+w@
z{=x^8P>2vuSi%QX08`EwO~y2{DI066C_Gw6Uf4WhqqQxrDQT-2y&@yklCqpJ91Oa+
zuxPkw7eUX*RdVc&-_85GYXZ-eHA2RocK0<U1i}ZJ)?$6YX&*C2(+r(uRHh=(nGO2{
zZz0Qj=u}a+70^fENfSY@pChG12q<4!qrK6O*GQ?jLUnJx3<4iw4(i-PCxnj2=VZJg
zL>fPb#0DAdHhkz#JkgFts6Y@M>W4=Pi<SX_L*hCzzF2JXbAPuGLL`&As6Y}jF#O!@
zxS{xf>q3whZ?x$6lfeOk$#lwid`Jk6JnwaxtVwsxV{}cBiVF`v%(q|qE^l7F%G&w{
zMJjppi4R1@Pe))y=y-#OSV%ti{KxsV|L*^RQh{EcasK=!N=}$oGw!_eCM$_VG&5Y2
zprpqbgA%=1RNCmP)YLU!`0N*0Dwg@qi?1ORT)O{J?p!<KdoR3(5-D%LvW+Alv4qf&
z<tat4VB^dRTW52A;V=9N@{K!ezj1{J5`{-#30NPnE+PO;D`>5Vb_4STL&g?1GJe1D
zp4PNT*O2;#^}#ah!vT3ZLJC8_-{U|1PyQpGe&%C5^UO0WZ>*t}hAu6GxkyOH-_M-$
z5m)Vw_vnyH;;h4%7=*2L<W4QjF`XwL^x4?l;{N*{;Ip6p9L5;ldFLIz{`If(wXc1Z
zZ-4tcT)%OvJCz^bJ10c9@mBhSEF6D4-g88Jtn+(UkYyQX&Yt1I#S1+C*b{u>`RBRt
z;6+x~))@|$;u@s&af@eB)cr68+xsi(;sL?NOw_ufs!OVB79+rn!CHf}F0LmYdc7-r
z=%dF!FO!Vo%wu@n$C-+$3~ZNu+9W5J@z9UXV=4kM^*p#}SJk$PfFy)Cm(xtpnt%;~
z*0x9_&I)ec*+F@YAb}vM+Yo)v{2ZI~1nJN+;I&62P^=WJZmp2#1>?NrU}s7rYg*ml
z3Zk*Sh-R{DYs*}FeH+m><SDGIuJDOZevGj9JvLUxcU>=)rWETNY~6o9Nx$FOfXz8u
zoktrEohXKbKEr-WmiO_KKEsuqyq}`86lBslIB2xa5WekF3Os`7(P>&ogQ2WS(j-O6
zIi;p+;h*4&9k+q+d4C0X<zp;YjM)Jnxqr^3Sd(DtjM27Z`_`1hv0zjx##K($YP9Z?
zWkW(bK*$t9^uhDSv$nQEq5{Fy+#K&fJ0<fay->4q;Mo|}EG-8HE0Sz;iL^JurwuAG
z;6SKBEKX1G6GV^*e|LLQN*H7Tr98$&R5wW!c?PNSD54-jsK_ssI<~-rw~Hb+63CBv
zhyja3cVbY1@UiBJBFzqdWKL`7xT``$BvM3vW9ZtiVU9@qd0`)gY)Z=b_;^&1WLglc
z!M80c%doy#P`rNF@dq;~hIgZ5K2C^^K^DW(TT9kqw5bS;_U}*~?r{F}7Qz`?Qxigq
zb1gnNmIfKqx<X4qk>^YfW-+eGIg%_vbe?JmqRqry!P1R+tN2}`du}XHj1FB4AR^lm
z@8a~AN~5(R%`=?K@TNschY1Gn15&|oIHamZgkVv^lc$PakrP6c=9S+$#pHL+q7sct
zV&X-R@itQ~21Fa{a8BY}hG2-d1L{)In1sVBW!%7|@f_8jacyX<Lv@9l#<Mz1>1Tpe
zdzSlxl_4m9L@#YAGK-=?1dCQNMn|Yv2qcLH>+y}jH7!9J`XAd}VChoC){k+v%lGRh
zsdqSMnM{s4Zb$q+rQ%1l-AK?X!3WqneTFam*pG4R%05-o@YM5{sN0IVsTc%<2uU=p
z>>|2E)3AB}13Y-~L3UnyfsjEWV%yo&6~FRJ|2?<c0sDJ~JX@hjB#qMete_SFgF`7n
zz~fZlyWe|(v*%AU8twDSi{IiyPdv)ySKp$V8AOQWJ>xudn?X5~+}=Bc*AMvAhd)em
zWVm(ffOF?g5u~OSg2q}f5@&S`YW6B-5DS5jI`Un*1B-Q$3)<8brY=$Ch<s(h7e4n{
ze)8u(z`1k(cb1kM|NH;=uOf937g+BQsyp>|i?Hla@xZatHToTPr$HrR)R|D^y#ezW
zH6g^YL%dc<@36s;<vlXM8N-L3`7j^+;4}O$|KeZb+Lfz(>s#OEx4-x$zWn8{u)Tc;
zf?q`Sd7yjPKOXOE%yGCAt(oICM{L~*&_dEHa+X(?c=+K<Jn`gHJap+HE?s(<OP3yD
zd3BX6%g{PG@jL&g98m@DY&0mZs+y*0XquY3E}70IR8>h;&zO}HjOn8OOyqM$k5Iwd
z+Rj~topFL2(**0+5j1GqGM#viOiN!I@=&8HOXa|ZXo~9u*w{TIbgfOP1Fi+zdTekQ
zS~3~ifQJX;xbDli_6HZaq^>i0S29vdl$NAfLSCdKi9)NW<}ZqZqG!18{<91Qy-2|C
z(C{?T42mB0&LPccOp*$QD@&}cuTf63d7|cGs3fDewn4G7ib_&A;T$n9>gHV6R7d+0
zmE;VURw?=#oyCJ;^ktJ#-6}P~+t>~3IMC7UAjVQz6CD{kL%VoTm>c+akGyE3-g5^>
zUVK;4%sIqP&>BrMO=+hI<w$V*4&2_=%$l4w4Ds0#*&t<Ub(Ph%4FWkTEvOsIY+7-6
zFyhRE50d8!>uR!t6=t)2{B%k+n_=sk`k-dHaV(dPmDaGdS>Su1ih#~y@gsHIbA^cD
zgmiOTuj3&F3<`lrVp@jtfh1F;%Ex4=5IO)sMHR7BNFlJ!MjyO+Qk05xsnBtf{TvDJ
z_Qr*dLd^f1b5!Ss_;fU>t{o%8f?Gc~*^c&X(V|V#S(EwL-%ez#Yvx|KTHUS6{b9x*
zu&@*FF6_EONzz$nAfkb;^+*Sa_XKaiSSEYhq{5>mG;ImOp)(CmQqQL3NrIM&rfL`t
z`rYr3Y8Rm*@iBOha1QBZXBrr~XsvkKBg}i*bO^XVF6!p0)esiQqKL!+oH6l)s01oW
zF}6hsfwOJL`S7eP54f|vhp~V#<T{Y2`j{gSI+%;-MFWwK;9ZL^M~I-YE}=CE*7fkL
zGMZ#uzcpj$aKd5MXM)0rjHI_hok1lnN(t~SuB{o20-~IPtr+$MYlFa<wVaKiV!5ap
z=8z~brXtlk=HGS^U??CFNGrNl+cEkX2*(B8iTBrgOI??g<rL>EX<Br|Y>ko$N-Lzu
zV*YHJ;dR4<k37n$4a2l5dGL|b+`N93BuxoK#7=knT>!ikB)yy`KKu;ZZ@$8`oY9mu
zAq28C<xl;IAK?yb{Jp>TOWfWaQ7KqCe+DNLKq8W^weAfE)jkj3cb-%CZ*hNm2A#c$
z4;FGTwS^#YF5rAbj9VWFjX-BPS`B&awJSWc)@N2$Snp`8r?F5QNoW+UjcA<<Agw33
z?#50sNOBsP(Rjg#jA^n&dS;1>AN(+@4?RI~|6}O&^YjP*9oFvi)~hd}vmq!!mgf}x
z0;EDEnl#DCbQ)FF9rmy&+KBf9fYM3#-o+J13U!Rs5*HOAM6~AVhD_4#R`#qcue1Er
zDvvz!0e<o)e~Q;%e~sV#&EMpo{qujpmCNr?mQ{>Wo6m(m*y8a=Xq;@Y!byJHf<q@D
zNi<9S9$Q<RJow;6zVL<5^V~<D=iIq-tgWvj<eY2sKSeQje6Jrz5qi9Du%@M{8%CoM
z2L}g?MkB`KG2_vQ$!tW^R<vzRSx+(6V~oXGgLe++V<h^yv!lBnt+Afh?o^~QCm}(H
z7G*}H<BCGm@xzr=jidC?3W-r24=(^NkQo^dR1H|;V+&STQYG*~a&$C{k%Br#{iRaV
zIulLEg`g-Bf)oq}Im=5!hQmG~@dPLER*>nGLBGdik6nu1oz};7vFj4$D(B`4?{Mfc
zMB-UmSs_g`DtwIZIx%#XG1xdw(qE1dTz9=^5wY{cE|ZoCL3pf}<i!A`eI)x!y24I$
z1zJ};bm*Vz42J~~U3VRC+WUY|9E<kmd^_fxulG2X1wrr{5!y~rCNNf0mnqZzoaw$~
zwClOHFE}(y2)#zu+hS>JgWgKc*4YiV&Ti5+kR}C_(UhI-Jtn5)>_g}14|5!b^~o0d
z`#X$wj+ktZDG$cX>M_%{W-@7LCnKcqldlL8Bakbx6&%;DKoB0?U5k5d)|3<sdWu#^
zYU8jWury3i%F(5L6NKsx-|<5bDOldd`(P2}DPqEsU(mcdKI(}}H@|nII#zWBql)#$
zoc}IjG1@(P)E@Yj;9_SpxMKrkh^V22F2a+TTdPG^!2j^$eH#>W7pIK~Y08tQkuF6c
zIPeDPExj}&bpa_;1Ue!@Kx!RD8o429-oxP_=~4o$CDSS@8LdYP)m6A2l?t42IHhBI
zQwo$1s1PAd>)df~JUHZB!rT>d&UMgh@G)a9IJAf=CcGoK79T8<ge231G^1@Bth4dq
zLO@HyQXj4~j<U`e_5}k8!$fu!KB$nOLk2S7Wc0MNR^XZzEgWql34V#%Y%nTUdGlt&
zP7O_RnrYu-mOO*?W!48PeC9Ks;^vJJzx7+cjgW%Re)iLR@1<9G`l(Co+}h!l7hh+u
zc5F|kyuCMOGj}|Fag#Gkfi&Hx7+93^q`?z}$4F1e6k187*QmDb@PuQ3zYt<ht28m4
zk4}kIl_hP{AOa*x$Cf~c&?*sCMp{Ue3A9?{ou@zOQ3(MmVXe1>Qb`Q2_7-7%Jb@(|
z7c`02Jn*rP(frH*k@l#ewOy7Vfwrl5@R`ThJaZpQeK5iE)Bpa@V1(hD-}*N9-}eBc
z(IKX-SzXClTN&V;B~5$mZ0~U4zK7`*8HamQoCPUTOzTiGjb41(Q1mplY1qB9P5-_J
zXlXFY&=^ZQ6O>a8Hm7!AL_B?IC1cU7Qra|UJ7{i>3?l<0(??%;7CBhv%Vk6J!du8!
zuh2NlfAAme(CgRyga7y6CxnC$JX$OAP7OC44q08<Vq^Ujo10s#Z>+PmwH31}lN32e
z@47t3;3C=FclT!p21T@W&K2nvd2xK?L5hwDEhIY4**Mi>xV+4B&ppS_{_M~48~^Ip
z`Q?B5%Y5&pmzd5<B4*qb;~l+W&cnMKo%-YQZ(hXRyP-y1Ufj6TG~vGU=lJxep69t|
zKg@F<`4|s9^bk7BI(6O;_RISHJTGHPrqf9*+9qR;j*gg3OUkmOnpIRyMOn@bblYg0
z8!XOSy!F&ggE1|}v@~r?YYYe&4(~P;G|q8-_mE6Uv<xKDQ*>%|Q_nEXA-*|6O#`V>
zBtoNm8Pawc+#!JS5h(B;Z(EG7uoBRHyjx<lcZa%mtgR-jtuM2>vBc%em+?L@81#AQ
z;fLt;bC#C-3|IQBZY<-xrJS`?lNnkm8q;#&f%{omS;jg`q7#~?<?vv{+pk{cXgsAQ
z&?|D5R+hoF;9H`7`SHl|9$V)gfONpz6#VY_5CMm#HS|=9P+(<L(<7VCu$N<QxZoHt
zKy-AEj+ZDqDw21!ZA+f{uItoAYbn{afjf}lglcm>gi8>05AR#74K%f6Hc2?VoUlDA
z*d7IT#ty63IDg>@w$7br{p=aCl_5)O1B47%Ymud)oYgemKq5hUT2~|L`0y7VdVtL{
zo7}p+!>#3Qs>7P<W=WVD#?y-1hi`B+ny|U7SzjwqmwF__F(PXrwV(}NBnY9pV(Y|f
zkb*4BX^o(@79TuIO9Nm|E%IoYkR};vl156GF_^}5D4dMlwYfV)ET20LXlL{+kCn~L
z+qK?A07xpNir3i@*Iy-oAdtSxrnQ!)nnmEqdnAfPC7p4;jH1U7g^C1-nf2i=F5eF`
z-Zx@*H-!6Mi`D$Q;@=#xpb3Il94LuUNd%*Xq_&nsca^Y`UGjmClSJ#7eBgbQa{EA2
zR)hwGR;VmTsLs$=h=oJ55Q;>|m?!5VGB3K;b_fv`4IM|x&kbRPj0i~x7}H{H^ly?P
zBTG_T)}!_%QKaiIZ9|@?WO>ff(U^WOLkfd8(cK#85-W(%rMD0i(kQS6uI*FKdNl4F
z@7zkbHO?uMQ%rLhg&CJVc!~25Zt}!aPcWS}v<+N*<SgI#PQlr;r}^;5o}sr~aOTVs
ztLF;t`@k7?ZcVuS$~%Nf#bj@rZ@x3*{7R3D=Py#&gtffHhePrlWJtmHiK?X;GQ0a!
zH%~l}QlXWK)+mm6BN)Rx19!2&%%%2S%WfX-EtEo_QAx(~+6HCYl4b?dsv`BhNNRN!
zMY?FIYau!P;DZ#)%hX5vBsxx0iJ+d2>E$^~E352Ir}XoTVSm6AA9{q(|H!A1QsTX1
zWo3ms+jrR8-67bNH($TblaD{ipufyR4?cjmaIil`YfaluaNeO)!Rp!yhvPBU1#aFN
z5p*A~B_=djTN8{YCB+MY3x+y4yauZUO`;gLGhVworo15u%V(fiLiJa$Wy@@`ixmPJ
zEKM^*5Ek=nclS;-D3(zXn<Xh}nlK$t*}lDxly9SyCP@;OmzUVs*yQZlvutf`v9hwl
z^6DzR!JsSZ+(I5N=;poa&}pUaS|aAfoe%*jyL1y7?E$hp1LwGO=@R$fcb-4><3G-?
z|N5`-Pk!l_`1T9mqizj8fD$pa;{+-5{VpqigvK#qqJ*R<Qcj)P;*m!#@l!wflRW+O
zlbktohSjxolC%e1a?SDhgI(LEnK7PDI6OSy_O0t29v)z<L2J#W2Os9t>9fb^KnU2*
z!_F8(T{o2FEVg2+lJRK7-u^D*aV!e!riI{W>Y8#ox$AM;)^Pdib+R-gO*B~|DN;yO
zh{%x1sY*>!SqcxC_Y~ezNa$<OGWv&l9L@)<x3s>&QD9p^Q^RyBXq;rQHsqnl?&HHB
z`yf^GET*+cAz4{jp+D#`olTfb$87K3Vs&MmROjFwv)P2Kuc&Rwq=|`2-gu6VCS1RI
zi*Nnz3rwqqfMAfNEG_lH)%bRbH`QIQH_Zz+Pn`#$PxjnTOal+zVp@+c<l=x9p|%d0
z?;^1xN*7%Z%*prv?nYT>*I_%}v|k7V1)ZNA;H_Vn-QV?etj9MNW1*}y)3IW-r?|DZ
z%Fc9|S`-wUeNJDvpYs<VWbM=jgOvdX2V>rT>l%~MjB-{{%}S=TlD27STSHwneDzD;
zUU2A_R|c$au9Eh1HqM^LoeJz02gt@D>z3x7DYxw{$|J?BZt<;S?U58eaLBBIv_)qg
zL5h&~9vTHgrc_mlH4c?1vMd9d#eN^l42AIEyv5Y94J(z5N?4r`WaPCkqSEM6qC$v?
z72Vr1FSsK@>QUN}<T>b$Di#SrNMi}D!5f2ZYnpNvg8}HQ+On?D?i9EIBL6Q*KxgP8
zjiA%}7=K{##DYv67JoVZoD4BvRDo6T-)#u^u2=|Jit7+0DFZr@h&08!2ICxF0zTlJ
zr)?VSw54jAxakx*{lO5OWsoGGHC_gsqBC$5U@f+e0nKs_M|<C8;>8yrmFyI<F*-$&
z;DW<6(ZWI@NTec95=^5p-9{|DqbP<9hXW1|C-l}=$a;N3XGsU(rN?WJcMYyJ1gpVj
z*m_89&vGzMdE>U@cAYX&1N71=az*j{ho9rhjkh^{{(koNXN->~+_<@eR+7nN%$;rc
z#b5j<tgfz*BoBdD;>`Iqw$8$(4?KwSiZA`fzh?K!WnP_**}EQi^6Z1O^;J#}YiMUA
z-l2>_rCDc$>=&F5?|p2kN{I_D`Y8D*Ofkk_tz~YDF)!>EsUCANO*a?1@(`&sN(*En
zNP?uDcv@rVCpm)9@#5wbXTi$mDK^iZXMF84Qu@VwYo`;2c}8zI!0D1{)AF@%{4NhY
zem{9nv$R}LmQy@6);7HS%IiG);g2Cg&eHNS#x#8H3!magfBZ+8%^G%hcKOnmzl;=u
z#~*);JSq6c|KwjXtra_m4zHF-216QOV{8>kKQ6YO&=S#02wBQRdhUdp*>r{z616m7
zh@?;{RasL`8@$li;G$=qjVkJREdzC3W38gsOQY)63!GON13D3r@0q3)MM0hyIOn)}
z^(xn{U5f^^SwxpFU3!?c)iqXDR!Nc!A)*>Ec)qWprkGz`F>l#MF@{n)W*z&MB+WQ~
z|NZ>j&;MmU`N>c6E5G_{{Nn%cOT7NZn-P=ilw!x*#UHcqzTXfc77eQ_%RKV%Lp=KE
zBRuoW(|q)!AK~oT)Aag7k}QWVzxWSY%(bSWu1k&%kJ#S1#s0y5{Mg%}s!Q@BXLDnd
zjm=YJY0*V8MWo1w<2J1^mbPsdA9YnR8XqzqA2AvoaQn_p?%cV_ojbR&X8*3|;H=|d
zdoRWsD%FwNCB0lnGxQ*6+l<se<2<Qz)OZ>dA~GyME9eV}6&@2jWn&O(1#2Zu<1tpC
zR6>@g43>Iqon6D&VH9=vxOrz;LTw~Mf%A^CD#7V!@1s3Qmf*ahu1g4#W;)@<)orf4
zeVseE_OJnj7RXFT9$q~o*c!s!_OXdhSz2E2inMuwFfZz3e3W+<Yc1ABI#3&+5IyF&
zT$mRZ@<bapQtILeoIEGQwt5GIyvwGjLw&xVeUZoFd`;WH<VZ3)(i|RU?3Fn+tN3Jq
z?hjZxb%s!+Oiaz>pyJlGJKVXxO*N@wR(9)X+hfj}bAqFTvI7AGlcS31sG+~yXRzF3
zd3~9Ub8C?T)ikK=0d}sIOe%q|BzmtRTlW+}phBb)dF2Tr+LGKo1TF}ws>K0mu23B>
zT!gtjTT}<fY}t5d5uLbEcCnv+&Oz)13n3gwFwXxTTio$pb|J*pX$Z(fA@sZqitn*#
z8bZ?$niktO2p^M+q>SzxGELAS?vp+cxs|%hXZNA=vitvKE4D*~kG;{3QJ&*<JbJs#
z|L?uUi9nJnq)$mgFDgr&!4T-B3aKDXBw7O^AZ;M)74f&6uyqJNpaa;}&@={FS(2ha
z4GQRG_$0Pw0})kg%8Iru<1Nz3`n+>-vpb=jjkjE9S*SHh$iN3o8)#dD2#{wPNs`jG
z5mWZgBfY0T7;tBM#Aq_5x4PUVM|3xf0D~b|gR2d`6$mS6+GVaEF7wX*lyREVZffK}
z@`a!N1i$s=uW{<WH8xJI^1^rC;PuxoGagNF&d01f0n@3$`E9Z+<#)gFBE5dWV7bp@
zPd>u4&pp9pT=SvNeU_b@m-x;%USj8<<a^f*kDT4Y9gR8FH!S4>e2tb6binI54<z1@
zikN3hI416U@1ud^vEkX>(?f{P)m}j-v7bZcq00<3O-t5szoxS(!@*L#@nj^)&S7VT
z=EB2|ar=dD5_~kS7QxdT9gup<>gqBs*~5Cz8?V2^zxe0B$)EYNKi<t3%j|HBRG!nP
zH#j;vWO-?svYK)A>Scz*C7yipDK0*Aj_o@=KJxq%^!o*MU32r=ZB!;P<AlRW;HV6w
z{bfqCOY7QbdZ|)Grf^a)4wl2V!6uqY!9*ymNGNn4CvvtACm0_&A7$0h)D^a^(8ANp
z(?wy|HZ37!cn7O1t33GNB`#cih|_1zuryqv-|vxSIa!uQh4qR1tu>bM_<-N|&40-Y
zFMJE{;oQ0NJn_U+T)g-Y>+2gRm3HUOySZNT1uG&1$B3Kg+5$=`q)74BA%q|=6c0c0
zDC--WJok~0@{j(}KjK$@^;en9D%#d8e(!ss`G1th{8kY{kR+OZui(<92l@Pue2!<H
zd77u5dWx;nTb-Rv^vLA*UDU;O(^`iyEtAO+`v*I`bNL<Ke&;Rr_xBl%N3?B&Qi83m
zO}4f+IeqGEL<oP#$BYNPnzEcS9v^Y-+GSpS^`%Y}cI&S1_0BQgJHT}9ekmncrWo`K
zbk^mdN{wqRjqxbws0B2M$9OQ^!9=PQNuaWUQQ05|1;z@h%HlmFI_X-Jf;3N%7D9-v
zuBxu68b|Be*iDg=vMO;lqFY&!qEm@;hH`4CW`^zSBi?%b8aJ<PGukiX8>uBa(+md%
zX1Wv6x_gQYoumwx)(B`ira8yqfSwSd2+cV|+d6{qG|tmYSLd{BKoCSE@}JCno%3e9
zw|epC9JPzya^bj`V{WA1alB4&?vCF}gYh$FRm$i<b9j_<FdZ-pL$cu}6a(Z^k8EYg
z%r@-rA2FRyxc>HaMmu9d3({*sNDw5)f(>Ue4a_MSv#DV^YUvdtmX{0kCBf!}EmR7D
zjHQi?Nwv>xS2CI?!o;w$Ye7P;QiN(CaRlk!b<IP7)(OhWkYq6*JZ5#rdPIrizyN0*
z)<h%sBuS$1PsKgYch*qH=KjaFQPDvnx>n2u621ecf(U4rAbnKA2Ji8vB{Vgmt_e*`
zFpebGBxxF>J(HZEW0-^Wv1Srdb#u6LJJ9bRe|!I}*c1Kv#qzyq-Oaxz=GBF6vq-Xp
zevwlQhs=&NZC%nfRg@=ht};Bc1=W;t?>5uPn0^uo<k%L6C-3(t1_PFcLtN9~+Lm%Q
zqnXBDwF)*G8TayNvKu_XSlXt>wrx~|c9CMS^%W=Cyp<?bq=XA0&{86lMyDyk2im%(
zwl!Ks3lE(nw6^X1?OKXFCjdv|8S6{S(XPbABbCF^*cN47H2!UJq*&wTq2T51n!4z7
zlr(($C!c5a{!@%+WB$ya{u6xnt1t25H(%v&ToI%~h!p2MQc7%VkRqdP8kEQ}&4gfQ
z>>up1e|U#0*WYGkZG%e>J;K;JX0neOr0gB-B6d9wZ=S~<UPY~2`q>2RgfuG<NpX)z
z!Y^QE-yv__d%Sat_xog#AKR#NYZVYmM<tLH1lQel2oCr65&b^FCseaK7TQDvG(yuF
zsWJ$T`z}1p3%!06JajjeHZ|qJK3k{O3HlwB*90f{-LJhy+Xz1P@u%22J*4PKnzrQZ
zxebmEr(C&qo8h2fb#0A}%?-xW5h?)_YNReWn(Q$g3>hvD$n%U|U*oLd&dwpL8)ul^
zxx-P_KvB@Rgne7FZ!CubN}(9f8jO<2VGk+$OlOk)JGYt6DkjsCrf%`xktZoy1Z1%6
z2T$90Am#ko`}pLizQ7YtKEuTaFOe60q|hC;;KYIL#Qi5uAfrpmeexn>d*>EaS#$mR
zRlff9uX1YZ3{O7wK^}kXaUS!o3D5um002ouK~#F^5qib2^Dd#|0(VH82%Tk0R45Cf
zkh<drNl>XmI*T<eLP|C^H+gn2;Gu^e<TIc74FAvn*FWHe7hl4MfU&M?iT`n72i#Sk
zWm&?R)2DgrsVDilzx1<w@Pi*@ZFQAAOA%=nImq1){ICVqyNAcQhU+)3^U}*NaQ((r
zZr!@c-rjDP9ql`%X#6wpr4hM$pN|)d)STtzoaN;W?z`_2Pd)KCE5l{perpW8y7=DU
zJ*JFaT;2z4fF{6Hjia(YQ4D$sxfGou&?A*+lENejj1){-&s2GqoM2ga%#AVke_+UL
zS~EJ_0~$auJ{q%qd!PL~BeH%5BBt^f=b21KSQ9OQoCBpL-nTg4VyYQwu4rq`;ogLo
zzWq9{y?B|n5(FpFB%m6+^5{&mx>PXR*}>Km@OLA2N~iSt>j)7GyJIVwxy?mHbCk4H
zvjePaX@h2K^DK^ln~N|J&=rSshmDgc-TW{X=5*HJ<L4p_-+f=lp6>2PDUQ8c=fgP5
zv@E%OM=>33F>cqW^L0M(@Dr>)coxyiXjNcVPZ%8?P);jm<C5v&4B8B3(@5_Ts(aY@
zgoHTeY>RFl1mkI&lEeKm2RF7^z3&uT51eM}+$ok<HaY*`2N@T;jPKlGd-pZ+TZf!?
zl2bvUlOd!fLEF1>;UQ8^ng*sb%W76o<VEB?2agnSZ%NY(Pr!D0u(I>4l`269i8YSK
z%tb)+3a7gp=fH&!(IPEG;hGPAVNobVhz=Ve5Uj;F4ee}(Xf29B(T_H2p6GEZr9fzn
zE@b46F8GtNJ@13Zm>T6B?`OR4!p{4XRw5^hxf4eRA|xCoz0r5gv<<0HQLNyhwT|En
zRnssuEwiITs@V*04MOQC<<>E$uNd@@sm580>3CTEexF20s=B70&PatI5dsn7g$beK
z)p?J19-*SNIZ+yeUBIw&LU{<Wa7wj~FsV+^X;d0nW2oDfJnhFeSX)PBF2uOBG~;Oh
z2;(KKbrjw8@$=UoGMr06pW^y{!s`bx?)A{?%Y5Ji4^ph&$7t-i`t~h;<HfJBcX^v!
z=IAtK);8UVBEqstrGO$$`VcgUN-<rcgExZ1{Ue0(y!!f-rR8-#{}WG<SjF__Hn(5D
z%(clr53d$nzVl5koT<r0ixGw_SzS!xMY70TvK@}8oAdYn__}pP#=I3P1lD_GlH%tP
zIwERZG=cHK4ss+Jo_&N-({{f*O3|ehcqy<pkjj+R%`@}|LoEAoR(Xf0D~@j8VC&Qd
zPDbvfbegtSeEq90@yg3D^VBn!`1}_>PJfWFeQ+IP3`rKqik!ov5owl_<?#TJ=Q&O5
zP^!mtYI*C;>rBQaBIIakxOM#+&tG_e7hn7mI|nlkTE~9Vawr^Q5ttyD850dW4G3>(
zCnJt#70v{VjaDI22o!=~TWaUXWi;R<K#>>x++Y05JpG}Mljc37O1d^o_m0FD&s}8d
zlcW1wRh1kZ>;Ymy{=IhN9j;w}hcA8kw|L-zi+tqSkMrb{A7piHts_*=-K|7d)OAJs
zyk?AFiEUb?(4+_veVlqljufl>*+2VdxOnj*zxa>;G5_?J|3@a%Db6{Rk_*M|AL-G7
zBt?<&?1!J>BOiH=&wci@Jo)4kEDZ;wX^PYen8$m;`z^BKyVkJ1z0Hjq*SUQ8ZC-!#
zb@mSKP}U_)TgNQYaE#gt1a)0g&PwX4frZoB4?pgWaU2vYeDH}6bN18}_?5fAmwA^b
z(vHKtNa@t4tqT}J$D#9h0V;tafnFjQXOgCX@yv15I!0wl)n1|T6ZXr3s`cn>lf(Tn
zufBMd!~I`j<IEbHTkEW?udueULOY8GQ*WaZwpSE*A6@>vgSHgB`P$oDyL^kAS9X|;
zDhMii$ti)iH9E=32MNo|1v_u=6Pg*p-Cfv8smY2#ypK;@)d+aRHgMBacwaLc@8Vp8
z?hOzk8stV-g4lkQDnUrSXweFBoc0us?O5-6AAl|+_GD=im7_|l4$L?nbtT!~O)2*W
zH2OhS&pyn>(+{z9-!j=&ib!PV)457kJcKch@nOk#mtJAC(=Zz~RMVEw2ApqE3REH(
z7AaLdCCw6g{T{>N62b+_gE5nX5!0$-XMdlgrs3A^5f49di6UE~w<&Q#aA$XiTh=1R
zB|<fvD<IVdpG8&aJ+qNOLTf<kG;)NyL?^AI#g|ZurnV7{)0!krqPvC=I3FV2ps~lO
z)jZH(k+&PgD02snxk;i@D%yT^5tOcN32lq7D}39MXho8wq(x5EGz2NpQXsnn9ZfUG
zE{-WqQX*x**!L@)_<f2ePNH`2`uf~Tq_YiCLZE}dR5i8r<VsNtml+fpdwbg)?(NY}
z1=Vy!a50B3)tW@B$c+$!pe0GKz_$%Pc&5`anNAq24sjDfGn>&)rwHd!E}-%pv_>Zh
zMoU6t2@rW+N+Lyq5dvdlo}BlwM{b?PYm3Zt(oP+d_6leujkQSYNRxytOR;4|2#zez
zD0)3^-aG=8%-xBAc~L3wjmETn#(ss%Q^{UZaMP6hsZabU+oO`v6u$K3Z*jQ0hnr53
z+GE3%!Rj(wE334orJR+_W+i#nM@lGX2BC5s3a1pq<yHE_z{fxRB%k>7GeAv}7Ew{>
zEYrOQ_}=Ok<?S(h*I!{Ze~25~FY&;+3@ljFGB<ZtT|0ZR2zSPc^Bmd*5w)8FLL8%?
z5t@c1ExMIkpaiJk&|2ck5!*+H49;AjoK5gP`trr3gQx)VArgj{H&4+Utk6t%7o3Ms
zmyEV=bM~3XC<dBlw?YYtv(dI>Z+FJGzVQk>+k5=vPyGo7{Z-z+{2CW8JP1CeZ9Q7&
zR8@;LHN~kl-hT5AS1w=Y_U&C>fB9{KlO(dD9dGm2EBo|!6JC3Bm&0+%VH>EkoIR`9
zZ(7FIQyT~lRDe{N$kp(jmBIWvhJcb{5$EGQ(eM&Mp67hv(I-$Mg%FKB<8_8P+u_|=
zo_7x+B~9HhnM_F2lr&AT&MtT$`v)`j4{r1NYcF%=%sD>#(dYT-M?c2K`ewILMCm_W
zLAcmN=I7S#$q1#;Ld1LAflDGf_w>^r<kYECT)c3RzxVh5K07-*Sog<@Scr>$UgSLa
z_y_pxXFtO;Pe0A0k37uE@-jN<Hd%SHoO^%Ec-z*@$_Y1aUE!^_-{9u08{E2egYDgI
z%6fuzCgw={E)l5n+zKu*olY4S6Q<Ky_uGE|d_Dmvk+HFMntpNdJ-;p_sLs9uqRwpI
zLTpStEZYzfJ=6V6<DfEu*2GP;Ndv7FU<9~bTxzJ~kWA+wEw1%U4jZa+o9*o*);AAW
z+gM`j)H=OEMpmRTmQrUVDx;+#SW8p2T)BCht8d?7`_=)~EP9p(+RndAA%Y{xBv~G_
zq+L@HY)iNYQ4vxibPjjjx5pei?_-->VW?;OSX<LiW$a#b#QS;OAY{k=@}7AbY}|+a
zVy~IkAwGmIWhUIkgb?%Rv}nWLgV@a^<#tMCwlHcFtyXAt#$i+9ZXXgv6jiNnuA;LP
zp)~n0=b<MbqNy#jX~T3>G20tc&n74>D0&I&Tg%MqF=?LA9}ZYqUPc7L-mQIZUr)Jl
zYa29pT~Ud^-t-o$y=BU3!qH?x=@Nw4<fu9zmzIsg28wuLMEmab#UpV+E;5ym(Yk0I
z6#_BbpvyrQT6Zc`7p<?Hk8AW?0jp%!1wRHk7UGjg#zV=u@S@|Io0v&X+d^y5Qjnwx
zI!$ph))GmUBXsn(og;|G`<M>0a4qoQ9ZJZKM)Sju_usq?CpmWS_2uZ$<2wUB=^d_Z
zNL)ZUPmvWkJdS`8f#4lYUB>p51Xni%>5mmB0;;y9363PyWU_~vq)f{S-OtEZ2Jw5w
zGw>B5$jH$eC{&unfL+tI3WFt}7MDmWMWSTq85nZ_gAcTA8`YUwB9nwff+-!XX-RcT
zmSr@irFIs)Coggu2VN$a;AvgVud~+DmIhmDq+aF9j^}E*#C4-7Mafsc`z=24*-vxz
z_M05c-hrN_zohU@;IR)p%rE?}|2$PygY>-g;>*mY6<eEUdGn2{y!rYSAm!A>Gu;2^
zIUc@rjv`wkO&m(39PApt@})0v<?U-^x#rSC4{-Lr$4Psqd24%*zVMjQh`pTyHnu!8
zlh~FMf+UG(77}c?V9imyIH_4ICo%Ekn6T@8z<5WR7Yn|J3O=^VU5l&6jPG0{w<V^o
zu%;!+viNwxMejVR2oAF0DtT{6JM9V`AMjO0d3eD6%RPGil<AFK6fIsztPdnQ#~RJs
zZ*21~{@IuK6F>SX2E`WD%(K3+%wW(kD;tDR>}>Dz)vvt7D=)vpjqBS4ugLX)BvmLi
zVq<7%MkBuX(iP0Kp-wY)WzAt@*qM1o)>F&4IcX1h*Zy$EE-Y^(5v?Q$!rW9_ON@uJ
zr_Ye)`yB-!L#Yf-pqy}Q2>b(`E&?1KjX69V(eD=wmj*z`<MDRk!_>AV<LN$czWp*^
z`RZ@;sZV{DkACFitgme`j}nW~CUFZ{EHObK=Uz@yAwpyfB`Tg`wbq<DbDF>W^FPP>
z`Wk=tzy5EyeECWjQT9i72r0;moKvT^`1Geg$>%=%SuS3@!0O5}X%bcA;aC#+L;v1*
zuSI){(RiOb+c)^#Z+)FtUVoXx!#&D!iuDfXqv?AHo~VFV2qeLSX)NRMm^3LEjm9`<
zPgqyY=b$)7j^FF~g|6+kXxK;wy(N12=DWXDoN(puPAs8A>wJLL#XZG}s1IlW9JCH<
zLq}TxV+AgWw3&K*fLe0Ux8&(6PC0@B+XTj=89O_N>~0@+F<k>LJaj)Rt4kEU9IX`I
z2}~2HCzhjwlFM)0WOwI?YGz223_+mvB~k{mOk=Gg@1<n91}wqUG5YLYLwun?W%pdO
zE(^6=>%6mQUsH~Du(qMMG>nMeUEL7TaV{0kdqjwPYZNB<6Wp&kTV=jZ9~1rO=-=Gv
z*u4iu!d04jcR-n)!P+HE1=V<jYo@ez1_Y8MLnkSXaS>eBf^(P7Are8=v`oh{MmKjU
z4>Pou42C)9FPvi99HBE!(JxpU3;{{Lp=ol%J|nyexM1;N!eM;_WW#7QX8+)TGANR;
z%A_$QenNR@5zEA!SabI^55%;IfI`TKu!azjTB181o$HETsa4EvmkQfPu3Z$Kgbpc_
z2qh7^s~1AoiJaSzbiS`b$PNM3QHUuZ7)xkdglUn^q4R_^%Mgjic#m;DQodqt8pgJ?
z*5GU)P3MJOEFc9It<oR7{@*ix-yFN&tFQ|lS|@@h5fWS1<lf`khM;xa*y;-JV`4)v
zEqFtwG)78n)5OhAYn*pfP0e^RX4vnM>jXOE%2s%aB&F9ckgdbF7GGDmL?iO7a~RJE
zCeYM1#+c}vCPW7-N61qWqtB#@k7rF(D<x@ESg9mKCDD!DLnN^qOI^nsC`(felGX-V
z=b|YlkuYp3gRBQsL2!EnyFOzobJ9$(vwxf4{>m@2ajxL84=1R!K#-BB0eHcvzKIAm
zX_ByU%JR@7Thx`|Gv}XR^K{0;4?oIIJ)@EdHxDMHvSjb}HaD(Z<@}kmJonKjSYAz;
zj%U2^02M*%zS1lF>TiA%SLLkOj2V)rmY?P5=xgM?Lzeq>Y>Ns>l4i7~!FszW*ybEX
z-?^HH6AOv87Ogc>MkTQg0#WoA*@T3E50*qgXr^3y?Rz}*=ri<BosYrG)*zFVMFgCP
zg+w|SFkD_{e|JGJLpaOP&TSq}B%2$n_`QnLYio?`jBD3-*;rl0S;cf#^7<=R+1=jb
z+<j|&^!aD$4;-cqymR{olktp~zxO)(dt*e%aVEt(iBK7tYFQa1xT7PsuU{n`R%n&6
zTRU!tz)o#xf~17##TUQ=2GMbsh{&!C=mo(xB8o9P+*9;>{K${~D64C~PTa;cP1Pb3
zjmM+21f`E(_YWG@T5jCD$*o(r>Gg6pnpM(F6Ll7r5L`UPDuY%AAvCYN`Xbk_U+1e|
z`6^%d!jJLc4}FxSrIi>_#tBnzU|!q_7G$2qsWx<rVwxl<srad%{7F_;SNS{t`G3w!
zFTE1cgFm8U(ChQ)qnG%?=ReCY{K7A=v9=aH3RRa^ecbcm{S|heqtPxezVuz*c=L7M
zc=I)G-MK;CmeFG_IGi)lA8VoN1Jin}iEEr`42Oqfya{Y?++ug{I{n^KR1Qi(mZk`m
zbWObXh%4)!elRh$M<5qD#V)*i(=|Rs#h>cX4FsrYU*a5?)^<++ce@|@4!x5|_NRfW
z3QV=fW{#c;=<xxGnNh5GmRC>Vk{*@Rl#R!=o@!FlRyF%~_PBlHHoZZgUcX>zdBFMm
z?`QYUZFcV5;m-9P@S5Rpg^hD-tgNn3wIhzk`$z@nPoH7u&H=rx9!tZVwmrf(V}flL
zX7`IB1SI!vuW`c`cocDMFb8`aTz(T|VC&R*1YvQ%BqER?Xk>~Liq^#FNN)qqMvhqM
zjBMvqX|YcG=;OBN7K(tE#HdH-PLOuH_eZVZl80#bx4125Om+@wF1Ku+&3WYUv#f3`
zGh80hD|%dc=LQbw4f;rtv$Ma)Xga2DOE#Ahwk{0NsYEA|Hl0vacW|X;(kQBWg`=aE
z%Wv*-<?US>+tP*^NY`b!SG0ABw}PZEP_<%K34-mR{0g`BMm%_?N7j$ArFY*GP+Jz_
zoOnp_Xx&HYSlC%-5lVKc6VV&jnigl{KBtvJ>rRC%L=;6pmysU2juMORmQGW2mLU@z
zd5g~B+vtMft&f%~y#kRX(IU&aj;|L{bMHM>Sw~-3AtTkQ<JJiRO6W!B>-~)PSJ>GW
zF9M;9EfJ~<R$grQkp~rgiwG9sTQVKUmBY;@Xltpb6)MkBiD$UlBMF{RmI!09TB3_S
zwVl!65Z1)@7Ls5+b!|wCgiL2xnc`cAOC4#RqX(XPGQ+rlZCZMbiF_a>v4zIFXiO|z
z=!m0p^d!1P2N_c)q>B^Y+Gu$q2{8v%Dx4A+CrGm%GLtCV;00u9?8CQ8(CQQ?qO*T5
zNm*XrWR{-i&C5AEc1Wc>Dj%`38906RG?TLA%4-KSrlG1Df{`SuhixsB8wY&mb5DVq
zvAcJRJRfl8%q3hsMdglLyKnIJ)gulM0~a2+pZm|<$6CI@t!5ivYC>&U8|Iu@8?c-V
z_>JHBHf5VInp(EK;PfD4kgPG99ME5lxuOVx&?&xcqb*G5_8o8-XBUdAc~R%O*dFhQ
z;nYDeZVfg~QMD!Ols;n#ne`;4N8H|{eEm&U&R&QoPvg;vz<Cf#;9LMDD70j2bCt>4
z$A*x;t`O6GrUy58@cwh$e(R9?A9;#dwMSD~KJ}@Opk>PFXvEuZzs0qym$_-D{Hy&V
zR?nR0op&x%&uVZ1+uE4Ps{>72BZOdMJ>j8KIfs|G*}Z<9a&p8>c}AgOr?$+3z-f&Y
zfc0Rz7F3AD+34#Q^$AW0tcPBaA%V88G2W4686Ww`b9~|xA16(Iix4bTT_I&awi4?s
zX_k^@DM^-MTuoKam`*1&O-tQ0G*!i{Gm37FVLUqI@>}n4uzQ498?xQDXdQi=ie6N7
zODSkYiI5daB}kdz$Bvg@nQ-m;6@KT7|C-N!=0|w$xsTHyEJx9chymGBicY{oOxF<t
zFNro0MAPnhyIB_nU-;~2DDsrQ`(OVzeD$l}h-zX^m?Y0@ntzwae62m%^6JnxDPcGm
z@`cZTj!%5zc|P~KPjlw<2Fcu+-ud`VKH5A7*|lDNK+E0Qmc9KQu3UYaFMaj5xpwUe
zqtOwwa!S)y7-z6PMoWd@!Nr!Vw~n@HXqp!5T=dZj;Ed&9|A6m&=R5qL{>EQtd1-|-
z%SaL(1!HN}&7p$jr6o4jHdtO-WofX&%F;4vl5~22Xbx+QrLG&in6h^AQVD@h(#~fs
z25Lpb(ax!(v4azcd7MOlTX)>?!U)i4X2wysp0Q~-HG~x>7}Os2w&|%p%Ug=&wKZ%4
z+jlA)8Z8rgX^s{tI88aV930%{=AAuyi9+~5f0)th4LEb=6e}w$q(wnlCM5e2VO(23
z#qREujnySKmJ>{Q2Vo{SZ#w4OJ#$0OC;EJzbS=a#bwB}ja!6KA30gzrn79zBD*{qN
z^jnjXCJ4&Lk!ar~g@}&r6xSRdICf(QouM<~m?m?g1#gT+-Gii8PO{r1OoGG!y<wj-
zr#5-|=?7Rnoswh`-3kq?7HjMu9I~@_hhdSk-YZxi<alApR}BJ(X>A9DC>$^?qtD%)
z_K^0l;P7fqdqdFcuVNc$oxyoe9RgC96a!7MqFEYvrq?Q{HO8)TWpcpU&Xl!G)2F!W
zH3T3i&`OYKNupJ>>PjR*M(ZXQ0=bU4v%z~>+X=#uT}xG<wCD^Wd_16d56*Q;<8E=6
z0+fP8A-ervdI;9j8W#aQB~h71rWrwsSSMLak>(U>j?zg)PlF(lI-+teCaTaSu7n_w
zjYYmsTFd(?>_TvOFXHL{gk8vd&O)Rw#(<3&(wrz2)po%Vh>Hu(;B!TPxkpN%p3P7Q
z+PcMfLsBTps-!O^-da+nNpz<WREqv!NT?09YhneQ#=vA{qCv0=9@`r7B#FvQEzwy*
z@D^_@&X{<Nkpl1t6_6szprsOMG;`0krmm49o?d+tD{hZMX9cO3*q}%gP1^ZNhR&E5
zAUMHn2HP{i^^wCb4e_>Q^|a#bgDKapUgdCKkf;H9zQWR`;QZ-xEcZ8g=dBym(<2V{
zCw%+G-{QHCeURbGkp7^ctSqO`uCum2;6opImcu>A&h|Ekdsq3=7r)8o+Br_2I!!ri
z*xK9zXJNC)PyDGr$(O(KGS{a&I9+gM_mD@|SJ^!>oVhOpSRhMEC7sIEFYcN-w;;^_
z9gg9CAHX?JQ%7}|$nxmIB?Kl2oD?`A@wFjsEq4EiLTS9Upy<*<qM@`1T?|~7(py=@
zolvz4=b)M4%6%TW?-XCP-(q=nh!Yde+;@s34`i7@7XxH|fjj%xIe+0kmR8pIov(k3
zUNPX4pZW~fuUutmFyzjiJM14GqP=BvW0j{L-{k0>m%07cYc!K1S`|15j+?cm^pL6o
zgrs$4<ZZ$t5^bK+rX@IQs5;c%gn&{q<^qSnqn94yW6wX|*&Rgtn9=BfB<bV5=V)|@
z6tKIy%jGMVdFR^O>>X?~8jYCEW>i&0+qR2qGM!F2JUpZ<EBgH&Ro!6hobxjy(FsMb
zh(eztW-fw5D~nGY4Gm3Gad5E1n{U3#3om?^AOG>6;=%(D#uH=^gb?|u^Om0XQ90&=
zCs>ON(Y#uzsERA{l#f6EQU2!N{2To3zx{Xk;upUZdGgjB=UI1!k?IbI^HcEm<M004
z9w(ohlrejG5z%zwDJ!L-P~zEV`1EH!$um!XkTYjaqq|`CIZs*K#Z~>iPvzcw#*-ti
zTz!kH*Dmw+<u`fljh8t(I;3vum=hXY<O%yYudFq=Cg4m&>Ds0~MtQn662t@VWIExc
z@4dv*a4GsCc7iRf6Ots_GZcBwa4=-h8_>%K40;2WmWTB6KK<T+^|f`hPO;XJqzBx8
z&ua*s=a=`MsDTe1)gUY=3tGejo|~^T9G?Sx2puOc9w3bkWOZji?;XOmjE;8Ey*`<e
z43^iZOo0f3EX`P5UO^~HkRleqjio7ThFcrtX-;h`vO$LKDO4dy2NIFwc)3If0hN0;
z*E3G7=dAQRwmtyYF4n-iUTgPy-!<k17xA^usw#G@h!$M|(M37NoL7Q~`-zO`j_dqH
z=SUqKug7<<$>I2uaq`B_fscprdll=>Y1$^oAP_kWR&%z_u96pN$NTCKECM=}$i(AS
zgOf9c8|xIqf;3MMY02pD2${sfGk8OgF+R-)hqnf;hpca`Gwh#Ybz_B@i3irEvD8h=
zte)UyjZ+OG3#2OvauRT*r7}6A$%I~|a5lZ`HN-BTB#BME$fHa&Nup6Q@;arA2|<xl
z7xQ%qF|9}GSfI_X>Rd12yVR=KrIZ~TC8EwciE3Iw2XxqoI@9^P>LfvGg<VAUdD0|{
z9V#7Dw_@8@MP$7@aGACxh!$P=6I6&Fe!Q>3&UuG$VzB_j@kE+G@<OEFb!@UcL+gwn
zYl4tC?{IBPYaDqmM`w{d*qD~yAg3rali3l~WXfPL#5RU3Nl4QS*WLX|k)zTCTUDsm
z5OhwOruf2vFto<e8iRJOGagittmp&A*HKpNBiw0iJl!PfxJZ>UdQFAkY1@{niVDUw
z%e&KMAkQ;AlCp^{k1Ub&@~CvtT?;L^1kVb$_8Zjl8Z%0I8!2mNhV0C4aq7XdJp0K<
zId|qf>nrPcZz%Feynp)RPjLP6cR6)tz?0z#l2lREQ$~|L);CWB0hI*SPxaV3pL6Cg
z;qAAsbK%Yk>nm$i?E%Zn19lJg*jnG>)i++_>E}Mo=9!$MJw>zcm^2xNHLNYzI~sQf
zN>9@?WJ$tFL1PHvIEv=vnh}Vq<E|KUu4}7mjrAegW^}%Rl7MMjtTjx=BQhxwE=Fis
zA7kBQ*Wz==Af=?Yaf+##`!7kPRs`R&f8{kEc;u7(r9b^inxw+7CTyO12&n?b&DcBK
zr)ex!8dlD%GUyK(E~N-8Z(e?dBJYt7H0Lj#ruJhlKKvk=hHG!W$n_V#NjW(}2*rNg
za>oWrK&TiO*O<;%PtFO;kqe$BiY(Ct?_#8lfHX}&2r3ih`9cV8Zg2D2TW>|5Lu;_s
z@ZFcb&Hny`J9lpL>MJjE>-J5Kj*cjL1v*ucx$1pXq%Jr%4r45JUE!Q1NmTbe(Hdc{
z`Yy|w>1-B76@_3h?9=P@aMoZ<G$L+ALz-qB9PaX)|N2+>-uGVQ^Pm5d{Me8EX;xP@
zBKJ;24vFi`i-qVc2EuXpyYn8`p%H14@x)_~@qhU1e~rKWfBifB*1!28O3J%M6`i+K
z=+aF7Fc)_7F;}h4eKt=HDFj8wsr%gLKEw0Re~kOipQBgwy5j%1{VGC;dC~|39b$gZ
z5*|XpIzv@eT)p--U;oCJdHeF4T)%mh@njT>H|G}qk}<E<VXdR7E1Jq+yY{OwCT5Fv
z5LtA+iH~QkVP|KT*=$CZM#N6Y*s_{iRrsz|rKCbjjaCV3>#MA+tg*Jb!TGai87?iO
zq+)sHHcy{?9YoT)4I#E|eLU>=dG`d+Qlh(vpy2#*3%~o8-~*0sqw7?R;~JEYIo~>Q
znDHU9$QejUvN~jWE$`68g7vL6jCELNsM|o&3WSHjN{?QVQ<*8frJSUYs2sAMB<p45
z!<A?>>m29LWNa-9a#d3v?Gl1vk<WbhEIIMYPr}pP!2p34g0`t}#w;ie^W5N*oJP?Z
z)XsU8$G_^`Yw?0t7w(>$a}dVb`CPes-UgT8%>btb*^**;v&Y7nWl$1h0!}(a5>aNE
zL$Q+JdO6t&49}!2uMa2&J$#!m9v|Xc6Wf+93Qv@lBx%}FjS7~AD<sJZD;q;>Xt5^H
zww_s8vU{-0sM@1yj}XGq6Mg(aLpTh`dcdUGV|gaA_1)!*5RfJctwGATA0??KO%p^%
z5!0O$f$Jzh&Ki_VkV-D%z!nx>A#~ikNN1Sy{-ZiJrc@|>oGcRWRfi{_^>N`Rr6id6
zd%X9kEQv*()=?ZW=kdAD!9xhFX+btnxF~3QpCr5YRoKOomV3{_?)YH^!i(6ZYAtoB
z$X3@VgNQEnQX#CxTZ1V}RJPoq8U{P9NnC<&OL|F4t`izFCF?~ubdhPi5?CJyiGm7D
zW3Y)qr#VTM#nWDQGHgssqGCH%DTP;2?u?TJAKMWwc!U#?3@VO8lckiTY0{~e8YUCR
zAV8L;kqskamESZa)oem9OIYdmkRi6I0i4U2)kBVq;?xtDSXwK1_#;oSbZSVkDrp<Z
z_Kh*pX$}r1?C$RI>T9pi8q4`}XIbkRlv?BB#YZ_B?=W0mqHbFLi~rky!Ku@y`78h1
z|4oFMRn5{y#xrXl;L*n)p*(VY_nWUXDUX=U4iPftp+_F)?D<XZe_#m}BrkvEHnO+L
zL3PCD(gWPu`AT=1^GwDQ)bbLY2<YxpU=Z(GOy;c=DMbw7PBb!2BXdkf1&l-xY>W?U
zn;P5H*y)u1>SiSNi$K7|mO+PNIPY0GbC!y<Q-}&wqA|AQ?QedGa(l$q=|?Em@1r>L
z2uWUGU9=dPRrfQQRa`u`2|`lWEuVP$VP1XxO&)q+i@k#*iXPb3@zmoNxO3|ouYB)i
zmdjmiIl>D#GL~%{m?CI_-~x&^o{?p|IP-lvPonZs3PD@PlebD0wY9Nb*<DLPVE5pF
zZ@=(eWTI)>7N!&ai~sWPu)jCP*cKrm&m(Hmw4TY#EEH|?Q=hfAn`4f;u4!9?)(K4$
z3Gmh$q>?A{ksVD_5@w#!sHE4+Sy~>_?-w*p+ZBb;g*{DDws$=L^q2p8zW4GA{M1kV
zG|zqbWAS7vRp<THC6Wk9$O3JsqiD*`(R_suo-|E)>d7bhtN+=5hLnom`t9F=lZ9P(
z%3XM4{vj^vPCmCe_2gctrpR+1dH5mz<e&UJf8l@e7g$|gAxm^@uy)*;ct(l$gwP2>
zBpjzq-1C?;+`e;-7he1>zwxjC6?b-SQdU!pX{b$ux7|9{T~kxHRCPmH*3@N#w|B#%
zt`kVbe@9-NRMAgsF0yEnh9uD)*RE?(b@ML<YITJkf$O)gcHW?YRHqDwORTJ{^7vzd
zpZGJ!ufzL5*|aBs&8{UWyETHioFr+23?RJ6R!t0P=$^jlo_|!T#?Q}e{EPDe89YN}
z7>%}>HWS?b5i6%3W@G&fgW&*`I%->zrV^#V8CcueWOH*sQ50-!Jjn8}$8gZ2*URbk
zQqru8_i_ebRoMNC$>eSJZ(rwV_YKyBA$WoJcel`^wM#drPcD`+E^Y#k_6}1{Fs6=C
zfX;Ra3G&2y6s=RF(hJ{8F)#e)7t;~j;U@~<;KSWS#SjbgvMghG&OMw!Nr4g@hzw3&
zT;|k;RW{EKS<N>%oa|9H6WY3_Nc*gvTIPYRhgm)~AX`o#Ieven$l2cA<Hhg1$o1<t
z`23H2nqEJpmoL#vU}NPR*WcJ@GMzCyEcxA+_UQL}oZ8%Cb90le)0?cFStIE`jxGe&
zmiQwW-8^7?bBE^E8nbr}nU1Hl`*+`aN(zSklsu6H*V42Vf;{f$-SziC)l{@?gR_xu
zpJZy0Brz|X=j*JoU3{j97E($@#jTJDQl%&v&6u4F(c9J;q*iFHx_vMq_;_%#wnKN7
zWZrKVTA^K3>-sK>J<YRNQ+Z49Eyf0vzc*#$_a5)F6&vn~D|xTebqElgN4m(7%Ti5m
zFrePw!&rm#whMt4w38X7k`zV3xSrCOm`X4?+$XdxE;X38Ww^48Q!!>EsE+fKB`7Iz
zHpbf|Ni5(*qLB!kvsmXkL@4P_c~KeUkx?Bc{ldA)7}HsOMAzr}>r0ZP+Y}p|Ibtvv
zpu2>N+Bjy@2~AnDR496hMusRFiKAfmKtb;`XCHr*?81QE=^j_U_YNy-OKdDZK)Gvq
z^@SHnwByv7B{r6FcK0h@{MH*Ne@LPfpZv^6=npo@@`Ni_u5kJCHWw~Dg<^@>G?3&8
zNfMuXuP-Q4#V0@i3^%Uq@hiXjD%Y;=AnY1tyUTs|Z?Lg;j{O^w@s*anQnEEX#dudQ
zT5Dxh(eLG?sm6EEf83C|0?*Am3Zg6QL`PH&INQdFUF2O6?YiP3PJCSC8OY$Nk46m7
zTAVjnr3ls`@#qlSL;=TO;}jtsEW{dK2>OMlu`{lH^WX6G-G9!$Q>ft@Qm3qMY;g9z
z3v8V^&+7UXkF6}TzPUxw>tjqq^Cv$;Wd%oN&EaImYp=b6#`E>pzl(0Bh-N~+pK@p_
zc1+6zNlk3u;2mzB=i`rCMM{EB5*lOi&c@TA>sk-~7|ou4zYl@8FJI+wG~%z0j#yiB
zT)Tc5LK?l~g2xAgsN)<o*2l$Nibd;oelNPtFj`7YmJcZF5!QN~1#evBZFTo)@bU0x
zjiYI1lx0J&moZ%GlV!0jkfa)GTbwgAZN>Lq{th>9-rx(Le~lmg(Lc@l`X)-oWn-=&
zON$(Ux~lP_Gn2L!)%i(kt$Ff^C;01s?XOYQHDCSO*E()yyf!Dj?*1?rcJmSMf#bvN
zJc~2W5}tqlqkQHwpW-L})Q_{awu)8}iJS9L<9vvyjLI?)X+%QrDeAm;930)@t6%>z
zufP5(Z@%?9J9~GSH8YH9Fs8*A6N&ULb{?v-p`Nvv)?%!~+i=|W?oMFyN0}42rSv2+
zMM{k~6;0jHwq}7!%)Kn-Tqxr^1KjTS&B^*9K-;#ArxR}8x{0-S7Y^MsIIe@cgAT6y
z9mM;>5*6zKP+H+ycZ^06i=<;0RgFbZx{l)=40$i3uRMt~ct2+U)*CGMHO0n&m9?BF
zAHTpb*JLUneM^?c)||68X2@2DRNL1%ET>Gz2OREiQ&uIWt-w|+sR>I1NxvuP_X9~H
znYCc-_n)_O!BOb!D+mUB3pPef1!p5S-8q8u1S#TrBLv8JNLFdm9S-e0Wh}mzT|jzA
z-vci?ATPR8#FjfCV$o376~5aaPYx#%j2^Il-++scZqi$oG-k$&?`$)xCy3P0TTOZF
zsYh5oJ0R&N$V`InX08R_1j<nhO^<3UIJh&z%oJ;<?kCe3!3wI0<@J}Zaj;)utfn{Y
zp>oCSV3+o2jB5iM=g&|qXNcUhH`xZ`$&-}j3!CU-g}3j#LtS2Hw!e4RYmidV&jon~
z-nG=TDYIDup+$*!f!fAU)g|B;hgjKhA!A_`aLz8~bm;71V!Ks!_l@d`JJHd(;=1g-
zk8zSpg6?wCg}^!+TeLRDnTk#Tit64(DPtmr^R&(oT#9L1aF7qG$m{3b_tWn+-gjX)
zhry3!@!a)yLgYau!C`Do;w6c4B*G9xO9&0Y8wwR~sAe{!=;!3aKEX;78Dg$l2uP&}
zA&?gZGjBRYl|&{idf3jy_MXXYlr*PF^sTWc3eT=U3&zq=NBCbVBt+!S$CDC1-URUm
z@?EB(M#_TO%wrqF;lTlEo}yAs<Kk;<n}$K5P~L!x2t>RVDYflWXMKM8H-C-I$5+XF
zIcAn}=k10hZ1d@lK8?J8o!#3z3~isr1nwM-ne88-lfY}Q?6AJJPY^9PZ@<cTcF2>D
ze}cE(*yC^g&40kwsZ*T2f0KtEKEs0#on<&2(C_t06G>iZnx^8z&wZMI`agV$y}h^j
zGymS_SXmx&>ilVTu3bX~P3d}gu|f!<;|CQ?r!&^pmKLqId18kMor1V)KgPC+>k8T!
zKAq(m%6q(&1k)h|E+9h`wxn92f~TqKsHnu#+J-#IP(dSnT$G#$=%r=S!3q%3s@8j!
zlw?_Y6b&vZk<MfGORB2n&YO-~U;Q#V%P>xW(Da8(^!j~z%YAz5rx={NkKvj7S=u_o
z=CI&9zxx&TufNS>7tRr^;?CZP+cV1qsHG%$u{iynXl^Vfyz?+A>l2%1w+Xp$e2QM&
zL-V$_Wp{TE(@hm)9oE)BESO}sc*o6HAuxBTOJ%39@*yrr);gxM{o^)*Ke>^gz)9Tk
zpWfN%K{cxx4D*<M>0(REIY(;ed>+sL=b!ykE?>UP&-~24&!dk%4u}OePDn{w6nLWg
z*P0kr9*s4kuVJ2LJpR}R_)q`i|Aha`-}+m8=esW)E3Wzeo8lklG4EW+PQica(j`9e
ziI4N_v(Iwo>=_c>DdobPhZ<XUQc6^kfJi!?+ucQ7Rn0g!+~eEd{W{-y;al9ib)DV)
zZD!Rp8Vy?$W4nyQwieS`+SXE+Eo~G1xV#IC4gcP;m=nUs_#@(BBo=KRAtLW*;q2eF
zV&h?JURrg(*dMq55H?yyxIkIGcVYKl{|F(1P62SnFS<^56@uL}_lqAC-M<AEK@$WM
z7Z^8|ERD#1o_UU{F|*Mw(|(U(FQ?oO)UHOj3Snmm8(X2aZK>*#x|v~ZgKaCy*_d)V
z!nu}!MJWrtmLxAI$S70>G9J84`)(s{F+Yt8qSX-|>nu7}vr&X-O+@KJTt{Q`Vjct*
zTjjA$82!|Oaj~FSbWkEcF|Ids3fbc#C3N;2zBAul%)@&oh69X}tS$}c^>dUo%%&AL
zuH9n&^a?udjOMkD7Bo|X8AVR#Y*sNjC>iZd8ShWIe*HS5-Ih!)v2$|*?Kb^<NFp*O
zhZP6gGbW=NLCV%}g^g@Tq9oorCbN><INF0LleS{qlo(&3bmT9G9gA_j_+gTJ=5;`;
zK%xU+F}9&KHLA7|vGcxDUb|yMVI_|Ni_V@arnAhgVnoL=R4kNZ%uS-zB4KASE4wH#
zsT4v*zNGEixiJJG9+qNJ*ZI?l$n6XBL?B{af^&r6&`Gv9ME<bjeHC`bSb_=*MC<M`
zC)q`RI^i2qUy)dk*}aOd_KBE)K?IAmDk|w@N^2~Qv*byFt7h1$rASk%x{k0U0i|O&
z?=cC~M-fD41Z%8C&%c)>s`gwYt1Ay7@=#&1kj4rwbdY!GR>6~qq;sx&^A){bkIKxb
zo0@WFkVfN$z*>XAMjoIPBqH)s0IY?oQB<kMrGlHgH>gK7tNK312RB$shw$h#eCs#A
z%>DNbSU#O_@yUy<Ew6L?YDr~gyz=rbu3!BP%4S56HLGiV(mbWD0_7woO}zcaKD}g(
z?dz{Hn;mic_HFJvd!CimCDxZWkwLSzx=A&;iLwP(uiRjJZ-<H!tsIrrEcfpNzJasR
z8w(#fIj*yAS+oINRAimpkc14hrbW6)`b?yNG$C}^dr^}W3qfz=BAO|U7*9x?jYMYR
z4JAV7+)a$t5ot<)Fa+Y_R5g~?T6$SVFHi7&jSvk4OY1!@i7j%Z$GUh^ndvs|f#K-3
z;5rFawnUY$P^pyVwN-X+U**BGTMRPAm1)Ud>zD|L?oIb972k_XGjF+Y*PrkH$H`wI
z7CTo6P`A*wCknSN9uhzjL#*c;s1PiYNPMSIj-1<0n(vRl?_KLb_YgvW=v)ul)=^h8
z%2~@`&?CzgS)LNQ)4veWt^S1<zRk|gE<gR#f1Xc#{8OmBkDNQ=OF>!`2&pJ1Q`)K~
zC`YefkYyQVS)!HZxo4l{fA=5$hx{M^#{Ze?H?F_O#rbz@2tX^vU@+j>XFtsIAA6oh
zKkx|sUOyUA6A`J1*3C)R8lA)hnY+1IP}eoPd)vJA_G^6c%fG|T+t--QCNynL)6_I=
z6NBT8rEMHd-O$vrK)0=<b2sS__4j_x^MWZ5kTUXO5g<iCbv$|R=k^XKe=2V7kqEzl
z(2gH(aiZZae(qv?AJE}sM=`kX_l@(`WLW$?uG#`G1V#u-4@Z?FPXuY|SQ!SCFib}~
z>`6hU6{DM$c6NZNjtKPup*qAjE#4ZOkHPXu60b#YCUU&AK&OHv6)YEll{{sTcv2N;
z37vh+dpzgv;Jc8k;xW!3W#{W@4Z_=40CtMW*up*W<3yxXb_odcHFDksSzMVD?|IAz
zm>=-w=zd(!;Y6$U#Bd_eN{>i74w~__Wy{|7A?GiiBkh3~C6n=#sxGnC(KZ(6J-gdS
z+`P8SjVn8h_h-zSnA6?M*ST~3h`k#X{d|dDI;0%eO!f_?4&;eoEgP~?43NUmgqF%q
zDS~CXU$VD*Kp6~;8q-qJ+8Vp(IA=0k=waR6;t3#j19cSrS#M}<9S_FAL#GbcsqSn|
zqW4_n(RGH;f#W|HSGQY?Mb}^t@$;gyRdMsvvg~{sl|tyKob|-u1Ls`?j%4WSkj?<P
zI~d30rFrql@lD$K;N1nD{Sf2*6?Scl>=aY?R*=W1EDty^HK}iqO~Z7%!PjN1j=e{q
zkUk=7!Agt;8$j!frWqrV^o9dWRgv`yjI~$=UIr2$eU-d*coPkR7l{ZW_N}E9XaQD;
z?v&g?{heSUL<S@x%(v`FP*}*sLl+!u!8%2r=Ojr2c@7~^*(TZvND0A_Xz1q&8Xr|H
zoxe`o8d@hXLedJ$@=}30upI7Gv~|MfnS{gr%RG9i$LccF(>q{J6RbhYfUpU{NvcX=
zY(bi9>N>Ezvcg18@m3>(#<`X^UcJgwPhH^B#nW7U>n6YbufN8|`Z}Na%=1h~hNHtN
z1qD5^#?H;F+`N5*R(h;#nKhQRY=wD&8#-g8d5i3%>$Y<)B6b}QBy>opHEmQ71<xSQ
zqGd{ln9kEL34x}L!i-ES+M}aR$l|dgo}_}4;J^!y?ShpD!zCap<w|3*0gB+zczUT|
z5Cn;o(95tcwxI|yy3YH6vtU%f%0P;z>?btK1HSyd*O9xoSYPgQG#xP+?Q`R(WaI<o
z{2+h>@99h`@Z5WYy~l!nT<g8(=uG$TaxU|Zh5A<ry1dr-Yv$i~(i+Tl#;%Kh2V$#i
zaW5V#mQRkZ#O{#nWA3DL;Ju-3X7qXq!({<pb}hBIk@owV(fE*m{EvT;y`6o2?8knB
zrIi(2$G1aDvOHfPxvsO<kW!H*X*5_BlFxqjv-}5t<*)Ex{#XAMM@L8Rwu<<7Xb50w
zX^BT4eU!iS7ylv;U%JF_IDh~$(cnDJ`c9=RNP9U_C%m_Bo#%Tme}@-edV%k}@NIVX
zZqb;Mrmd*zin?wvt)XcQP1RCWhPrGqtzRIq-LJVD5&J!!zxQH(#*K5Qn|Fw4!E$mA
ziLU*6lD^f|B#Ub?ALjj0f_}gZIOdz(>!&BWc4PEhfU$$5v><DGL&NHF!FW1h_pKvd
zdhvTK^-`9zz(BY3IAU2gBq0JQX(5T8v7#g2#}}qmKqesaz+g$T+LsiW#7UT0Ph&jh
zUL)=ZjddTsyPgi?94QWATVypO6(GAL3U4D}FdCD{?$t@q2|_76fwrxY$gYU%-e?gP
zyt{aS=y>qm;V~92bV_zd)7k|`cRsAJ%$Rch_I1`yd4|K3)D)<YBAiAAfvW}o=2yPN
z-u?knDTc!V7cM-+<iN0drDFGT1tB5OLlq#M<Y>2HdN{_n2S}I1z12e|dJMBA?mu-3
zk=>%H_L$5L*xo(h^xA1AJ0&~Uw%HyXqSq|8sPU<zmn;)}fi)>@`|cVfm5^sLQbLH-
zm}bsh41rW?@-#zd8lhyYExOdD`P&v23$0wBbY1A;ad9_quzMCY1|5}4>X=v&RrAr?
zH5SvpYsntxt@{&;x{%Qq&k?$$CunC4z2f8n@P`}kQ#|2L82^cT(0~r$4<7GYgsn-j
zU|ZVR6yYtVcC^j|3RIsw5d@_XBA#H=ET;)D94w&{iM0w51g*0)))BlS(Fv*0v4z*&
zI9h3R7tIl+UXhQN2pPo@LO~FV@axF6i}B~7<GTcWgvZ0Yat}D`qB@0u$Rp>Y$P2uI
z$~0KhVoXD-B>f^oi;j*w=OBfEPI7Dr1O>f*kEv)8^e_@mo!R8nnN@PNgbYV`AsDV`
z{Iw~@84%rR(6$II2|;3QgH%bphCxRYL=n?H9)0){bv@&ApZ*cP`|X!GwRwtRaf(}a
z4k@P<!3QRXmV=!sO;gf_7Sd?Om?p~$(yn*Ww6U&hiX!heez*7t-RWY1MtFymD*6OU
zNj@CnrNo8M6>hOu7T(iVCCf`gv=Y>lNmo+O@#pU3+8Js(5#wN%hJ%jS3bU%EY+Cx6
zL<@nE0+|ZZ&Qvn<0&8qXevO5UQi_0}X$Adcz;v*}@7}mW($C12mbiB37Rt0_D$vM;
zu@|($#XZEwfGthHBk}XJ=f&1`H{x=#AG^rnevcu-2{bK0@RGR*X%6fkBOxN5((m#K
znSYm9Y;hroyB_ep6{ho0BIs5g;++8A3L59JHrk~O28LcgTfE;&CL}5&O>(~Y#oysz
z|A?RZ_x>!W&z!+}7rA!=vOJIWMP(TU72cC1N#wVBL%-kWr~lkfbL-9>{^39ThfK>E
z|1J&ztgWr_;SYU?pZV!O$0H9v!l2)auT{&~rtA(GiB=><5e<LuX}>yWsp}cv`1;rQ
z=68OV+jnkJmJ^I^yMnHwsw(QLp{`q+%FtAXwsz52bxxiKI+MZf`N>7w^zL@|ajL}0
zIY@AeA}vn4`FxE3n<Kz;X2KjhJBi#aKIz{*^MAm3=Uunb-CsDqvF1G%g^i3?A)-;Z
zfGWVq2>jGxZtS3v1WSew8XqLu)|^=%ur*BR`57zHkg5);6j5<IN6UqRB#9a0K?bb!
zBwC<FK!jMZH`dX#9%Jr)(gAS6BNwLL-QE}?*UmXZQ_T?GQKTtWYfRh3{L>R{R)>!~
zyCfq?HI=c^1_c)F!;?1V!Ohcw=9V_`x>;vw>IUz{LY;YXSX!oGL~}G_yt_>1mXR{S
zwSsDF>1P90vURQ;-X<aE)aDi!E<V8W@+#lE{WhkuooAv(={&}0DJUBQE_w$BmvnR|
zMIc~(1+?@Ffp85boAAJe1O{W0nPREG#NmNKH5IZ5xXi=?RZE-^??QzI(fhDhXj*R(
z!r{C@2#J=GL}_xBk)#PhM74R-)Qf9z(!OoZEfqo|wORlfv5trZ*_;{^1xRr}lrlOu
z%nwI+alBWC&d#K3k4H*Qv?U5Y&MB2DP>NipgrMWktBkod(GNS`S7E1|MI+E6fiAj>
zn9jBk45$!DZG&j0<h~(GL-dA|8t*K@85-9xYG;f&V6e79s`?Z%N3;%G&ydm+T)?&l
zr4mAGacxbI0_(u{`?S{5x<Dc|N@r-5A%u=6ln_8jga#ElkC_ui2{N9GAR4bOkS9s^
zJ+a7(khG=oH8@YIQ)Hq@(9}(ZZCa!aC`2^sRDkt}?l9zqz#>s1Cp3_$glP0zk!8!g
z^UfaqJY{8RnWnWYuWqm*PP2Nx&vTy|Q;qi+?^TRy%~8`5Fo?uQo?u;}T=efVcE<Ld
zJ-+ev@ACK)4^#98Jo&*VSzcNu%MxZ)ie#PE-(=dBv`8B7k*OomlF&-3CL#(_MDA4R
zNc=*E=!@01*tU&@F#_qptAG_Ak0l9RJWyVcEpI@OWX90?KqDpKNL-7Xlnja?!L_*h
z5ElZ@X@Up{XFy1-OmGO4H%JuO(r^KRPfO3V_N)#ZA%VoF6dsaP<CP-wGf);oLF)uP
zCD5s&ribzaX02di1hb|l&oujcw@}j2`aqiJR1TUD-NA#4zDEJyae!iTq<eR}Kkq%M
zzvo}yW6U`m^MCopDR9wNVUEb3eBAf`t&vLLe$az1qRvik_M$`Z<~Agz%MP|}z#2hV
z(g;x_&=f@<V=IL4yz=S`Ov)pE;!po1mmYbvtAiujpz{JHHFZ5D5HQXnMLZa_#;~=q
z$uIo;&-3=1Z}Mxu{_6<IBI-y;@h&@tKWsw~(?uS8>@l8u_BlTI^wSLb{aDn^Ek(SK
z2S*jXtc23>)PJI=i~fQ8T)TFKmtT2_o!vW3XA>IJ(wdg0o_|=X(ooJCOcTx1oe9Uq
z|GeB;6m|EObMJk`IEluEaFWAz{I$hz=nf7+gp(r7<HBzK7~wsB`1gDk_e|TnD!`M~
zp5Wfk9v^YyXH0;`3$ztvh8V^!141S-Dp3e*a0ICs^al)-AWbYvI;3!&MN_mgQYs*I
zY{kXb8*{Xk5*_#0Q#BUT1iX2VbzuGh&;bW=NP-VoQ$w(%NrF%sZ(1adfL>TA#Y$Tt
zRDwzh2*=){A#~)6<3(hltc=B*?dCv4i(+<kvS`zu99~!q4LE3OM?JGq$YsE{0!c={
zm+`>)ix_WMURz?&AL6ZNI+@Z`RcsIYZk~kLvUi<$E2c^a66sNiCruQ+r3{s7WSS$=
z0-5#b=NXbb7N@>L$s8ylky?X0=DhiPk|#u0lqZS8na=sbbg4*6p_Go+O5L@v-Y(kH
zcY#%%sz0_yRR?%<9V+H_C@eU5qO*aDA17xfUAtDt!YqhRoZ%yX&I)YXV4HT12X^mc
zM|TUVi)4KNYvKDU>@o?dj6Oy-L?0m~m`6&-3*-YyTO+F}+EmCS#iOy#AzdJoGMWwN
zvODEhNaH9{kDE-HP4)@iv7WEsjYX=2;5?#<mI$o}qcmBspzZZ=ttALWqBEq{Snnh8
zPzZb{sn>uB@$)`JsM~c%%!N8krQMrhk+?1^H4vo2TTAU6Hj9=LT50-OfpCH*SgiEu
zGy$!UE*3oy+smzkU^T9i4Dta>YX!axl%olz&H0_Ly#vCL<RFuG_;|C9P%W39IK>w}
z_z}MGt6xQJUgBT<?svI$>xeQ;seFc#7Nn(VElMVsM)1Yo`3?sMQ$F;ehfzAEtZOFa
zjBkDG4JMNr&p&?=pE|~Mz=j+x8j=L+S&7y4F?>Haz6`E|l3PsMVB469sD%R40X_;)
z9${Njud%vE(O*M)L1qmuK&d2}7HMnRQHk&pfgzy+=WtHO9*}JaiN@$0=Owxg;3c|O
zEEfCLN-861y~AopmgNk|3Bu!~CzBE_6H1qo7fZ<C2`b=3pS@Aaozk%%95zU1lS3|^
z+oYV9+-ycn+Q7(|*xd2)=IZ4=CT2pc)dO(PG4ThC?{`avd;T6yeD>a-dG}|(-`{_j
z@$TpUUgCw|C1vSo+ZonG#5K_=)9D0oh`|zRmU8FL4gSe5{UV?H+~@h&$39M$^%mS7
znItHmED!=uH06z6XqpG^yPu!`xu4^W*I(u8wQD5v)DeCNbm#xiL(Kozp|Da)mY0_J
z(9_TG?6c2u@xq101}!AarQ=dckh;qaJsvTN&v0<K&y}lh^U_N%a{0<Tl+BdZ)U>9F
zNfC8RQ(2nQQq9^<^%_;rouwOaH#6%baCQR!z31`oetltatl&MqdwvbV0-5{%$BQ|h
z_WKy$_lMr=3nyQd=nf~buxW|rzip7TO^Zw{Ng@zR6NI3#hQ>;CIv~j;S=x}O7DY?z
zU@jm@lp`pC3ku;SGB|W+cM$|w>!~VF+XUQuwVnfj>@uPQgeX3X9JzqA4TP37(F7Es
zZIQy_iQF+6Kz4~OQl{vnhX@iC=6ct#IH=8q%H9*X<((4qBx37>Cy}z_;NA07g~v;S
zC&QRPV_f8>yNHev6j|Qq;)Mrk+6tM*hVkfN%;<1TRaKqhH`V|lcx-SuEJA6t6l7A-
zQ-NX_DI5La5YmL!XVkW!wgZ$XaHXc2w$zR54!m*Akcq<95*aiI{jS0e;-N6n3S%Sp
zEm#-RrG!992@<R`k=Jc)bZAhzL!{%}l_;dpCpd4%Em-DPbIwx?bgJ4pN*6!G{68V0
zB?_7lT$l43@XjIHfUO#QWAIXSb~cf}y08EW7QEs65>&s>cz>$cc^p@#x^X_DafoC1
z-#d$K8l-7Rtbvepl3$1No>avWP$!xq?M0p%&@>H-*e*>wJCPux&x`l*9BHk^mQhwb
z7!Eme{v2)7f)t2UN9FgtYL)^SVkIp@wAygi9>d&pHv=KI{W>Ro7mg+3LMs%SKvW%i
zV`y7Tp5<tjkft#ZIdrRj2%$3@jKANyfE0?RvV=l%bTsAs1Lr6Y?l7Is@M4B-6iZ8I
zsp~09jak`fc=OG7c<6yo(oSbw{mxgo@R5wwwC1@V`#2*xVJsZq*?kc!6~Yt<Cuxl#
zO$D#K{5rR9-{#DjQ|#{UP*pXDhcni<)_CfJAL6^;dV`zSCJ52T*%4`~aBfE0pZ9aZ
z!nk=JjBL7?AlGp~<|f7QWk;)r;5})YA@hP>zu)o1Ts(yYA18I&P?a-$>dCSck*0{Y
zp_3sq7q55`7I3!+!~o@PqB{q5ZD?Bsi6qN%NK#ZV=rExT1H8x?1dVkGzF5bIKz+2&
z)!ki^^)uYOcAc_mc;-V-G3Xb(^2)1ZS;Djpv2`zFTU6`!9i{v~iy@hdW1^C7G@jst
zUHF`ZIlTdeKyh$zz_0z<uTfPEpZNHvSXy4`Hd;@TrnGh4sYj!GzSf!$1VvHs+0T5I
zKmRj-j{o-W{k_g+qeI;P5RL8-Pq(Zf0Iek-eePMl@FV|U_WtzAvg|zX#eQqr!x?g}
zthsvbMq?mA5Clj9BmfRqyb>ijLK4GNC>(ZJp>WvYxDoc>;*WON3V*nkbs>@>MGYh&
z62wRlXbe4}yQ{max~7~n&pCToqd%;DPE}WfBwB0$w1|aBRCQ)n<~eKcz1F*)_ZdFR
zhd=ybrsFyrj%KJXZ|YhO|NU^LWxmgE-FbsAf8|TuxcMfBhx_!QqqPg#zM*X!+P0@%
zSQfLM`K+aFI#Nia2R7#W<<0jg&i~qX_RjX<U-fUk|AFKCeb@I_I{&v_)ps6ZmdC5*
z5yhx1aI@Bu61b39#DEe|E5&qNP}c?_621#Ged4T<?CTP}R-ns{s^~FvsHBPXNYPW7
z7AHhrN4!QwD2zbpgiitM6N^UTdcngsW4V@-Bw-NB0Ra)a+@cO*kfqO2%A%4$_(WK=
zXgXS1k%XW?QY1%60xfHbatlL&NiALkP?0#$MMPLOsR!DOw|L(ZgCj}3D)5MqY|fi~
z+mL`sxS;4zh;hMWO<}5<)039T#)7q-f}$#kQP8#>^Lq=v^Np7|+&{zhX=wc?RC1(b
z>9APlR7J^!tu-#6KgZ5S&*M++;MIs(UvhLWaP+cc?+zU9OYEW{xz|@$I?0|?d8%ne
zH7+^2+ai6551RhrcE5n6b8S(2P4rMnL#YdlRtS;kLyNO+P^qTXx}-$Tj?HaVGpJe>
zi0m<!;{7`@0U>!*a)@NIfJ`iDUHLVs!I)o=&7BhkQ6w-CWi-ioVjqd~HVY_<A}3do
zi@qcrAzJ2(j)@%)#oZ4+-W|uz_Q(_lcXLQd4sA_VEdXL5^c^^tH+DI|oGSafNGWsl
zjn6MW4V*n^J)^3|T1#E6WgdqV2(1w@Pe%4_3UrGmAAr`Tm{FAtS%=et8zMbKx*uJg
zLd14GuCsZO9=HihE0iR%Bf3v92j4Gc=L1P9X@wNo>&{z;8;Uy+RCS$gb(JE-oTcfr
zv<oeB5<w9Sg&OhZ>$iC17k`$Wiw$?~9<qCJohKf9lsmUiuy+?IshHFoMAxx-P9WL?
zq&q{06F&AcPg0HxKK1br<0lnz*7D77zljzV7dCeok1HN~{0jBR0D-HIo@ejQF~&%~
z_{A?_Oys4Pzs}7!A~LSwe9PK~LCJ)c5%1f~rDCZ%4ugz47q!;9Y>7bT4$72(Eccc|
zqcl|Yn6AtxF)0P!4%`8eAVj)FL*F)-@id!1ONnpV>?Jt(VTBlo!sp!B7?C8(y3R(p
z`Szmg99<W1g+?2NktL-NjFRCv7A#n2ghZN(Be})PH*RoxaEBWwE#17M6Pl0x{Lizw
zwa&Mn`v&`aC#a$##y&r|Lt)|leW?EL(n#bs(~{QVePT9iaLx{?6AB?TF-p9P)b*Gc
z;XnPy|42&k=}&)#vZ!+Vg=_^92d-rrY!`~M!T{6hl;8Z#-{epJ^iTNuH=fI~|0Gt1
zyMLR;616MKf^$3P_^1EupYZ<ozmKi0O^{~DejT8+RbNH(c2zaZ7c=hMd5h2g$scm(
z&TZz4Iep);Xl5+ta~6w+`Fzgl@r<*R8Ew<kcWx*i?qhU+yUO1m-&k^^wAS~PpMdj`
zNvSCd!}|J^ix)3&_trkPX*oQdGc7ghxIt-yHfyMQmm+jX(IS;4sR^~TNHi%Ke5f(n
zgY2`#Odqf|+sizBF%o1kS;pcJBH#O&lbDffBNJtr5#Hb({bE6kHg`%g13Yv71Q@NU
zM<d#zB846pIKX0nkq0C-s{HIa=W{Nw_c=N+4*r)a{w4Q~KqQ6qBlg}prPhMUbdAeT
zj=1s0bq>D$8n3=tv9Y;9X$t1E1xH6GEKYm89q?f(aH=7d3r1U-?aS*ta`7oRvLqW2
z?ht$T26pRF)Y>CxF=p@NRjl+h@r*^+VZ@Y<Goo-H141XZwrlEXjnWF+%yGS^2x$c-
zz4Lw;1P>yo+8AR<B&>Jz-VOrJeBVf&lc{pMJ)^!dCZrfXWF&<FfnE`#b89>L>^rDT
zR%WLaWS1VGg(`}=MCy#vDlPFT(sv!+W-m&$<U)(AnkNK=)C0XOW>ww85bIwv-c@1O
z9v%|cHj%YK2)#mVNMzq6AbC%)Ex3+^8{qqpIVM5jLw3<$dekiqS0&jz_~iJQt*y;W
zG8G`SL_|eVS2*toU5|)hn-*D>1fwWafm9jUQ$~{p70U8w3m;bU?&!mPrpB_TAo2;!
zd5?8D;wcSWJEa6#W+!P8BgjbG<%PB^DyHijRO2b8tg<1p2!w$HQC9`yN>W*nUa`J*
zo|BU!o_P8a`$q@VWzA=P?Nf}$hP$`+_~zGM;?39Yaq0X;CS%XNJNMYUvWqV~7cQ>x
z!i(SKY4tSV`1EIfmdlSk!c$K@gNTlejWss5);Jql*&%sa-}1o^eUR7Q`g6YU`9I<7
zU;GwV&wZAyou8qQEn8b9Zg$2vYPwm&{b`9PDYH4V&ocS0wKx|tNtwY;BW2jS(1vkY
zF&a&ARYf_OAXH92NC>0`ks^KD5JO-zs?kc(hd|$V*&Zci5_nPr{GRgbNFq@b<%+l(
zqQJS_{;*b{OOIB8LP~tl_@F?SST&|~C13v5tK2;~1`#;40X@?E{4f0y)6H$Z`sFY2
z<{NKjUZwL{>@gs(Qp&d(=l)$9tJ88!Ib)G*)nTpY@Zf|L!P-dQd)C+DP>2QUdc^Pj
z``;zS$glj$ub{O-0Lo~D&zzrSyTki|`ywSzJozNQ^;^Hi_g;B5ql14-OO_>vW_^8~
z_kZwNKKOwTvc9p2(k45}4_+fnQ`2Qb=0R@Q+3bW{w{G&n3(s@=)@^3<GrF#$w;fH>
z(li~5MZ;{i;Oulp+jO{P@c^q0{BK{g`o}kx-079FarnO_wZvLa*IGh|T)KRT+0iM!
zS+IY4hE)xSlHMjZ*4DT%EzwFNgeBt9YEB`6r0GO4Aj&;HM63;XH~7VUpVoFrktDL@
ziWKkjTUUzN9H*#tc2}@{mzA+GB4lQ$5pzRK8BJM@&_zXZ?Eqecfp3?)Of0!yF=ZvT
z&HjW--?|i2<{Q3^3zmnHf<kLt2d<4AAD$qq6FPZHx#^iqrtF^IVLX{&jA6a0S>G8E
zo#5=Gq3`qFdEA{*R0Wrw+~JXDFLPygo0IEz=uT$XSp#Qhg#HvQTXdb@r`9U(5hhTL
z1sjtMns&~7F~f(*+UA5L1I@gJpirnBKc~b)bDpPGDvc>j=Hn@q&AJCgu#YYq2#fp?
z5xAACI~S-5ff8~_a)h`n>;y|iTndpWJ;5*C`Ue{nk!y-563FaFiq_ezJ!WBx?>bzc
zZHrQjIekWo9I!CdG0T9bW!v`M8n?fz!ft-&7IA9_r3<7iK$q`0KZ+D^i#e%nP$QZB
zw}#V4ND(n0I5C8L6d78vQYoynn4+L6OB&k_f#PE2#BIuQfD^KLa7-y9bTJN*9}<Ze
z{#};8lX8BWb1r+yh{4=-xFAvu3Z1kJDvm3pCI(MRIU=bv1*IvoO^TE`RC}=?B*E6X
zT})Bt3Ou+~+fWHdty>D$@%WXi{M@g9fIjZ?>Z{-3LqGi}ufBA||NCG6F3-I0B0u}_
zr}&rur+>)N!JNPN+E@Ag-~Ug1;=_+~?H4}5X_5faKrO#auI%#6=p!^uhfkVOQSfWO
z@=5ARg|&04QJGa-qY5P@ckbNfBR~6oqW4^T^a_j45(G@v3yS)TiC$1>$?4%SkB%<V
z-FazP7_vdA51G^0_m;EM8GYa9#WH4ZI3siW!)SxnhN>K)%NkvdiB(Mqp1x}-iV}Q8
zC77R`!m_Y48V`ga(>M7*l(`L<<@#j%nGhI_#>1%!LX<cMZSQIOM3OBV;{t68TCqWL
zJp|85s(Aj!F|XggOF5}H@&$6k@bO>z1k<f8zV+?z^4hDf;d+lqWD9`!fHWfh?L&-y
z9AhbHPzZugbis3UbcS<rc!t3Hz+^HV+N_5E@CUy~BJ!Efe3sheH6jgk3^ABsGbn(S
zQb<+s8^7@z{K0?xkNlTE{5(thqQ6}u7g9!R&Ye5QM?d=0T)lRUsvZq8N^#%!16VST
zSM>grV}<VCz0GT{y~2wxzQD=JDb@}}UE8o|8WxL|#k^rYTd<fnxWSul*(c2f%zq<?
z?Z4BpjDLD?teU|I`pz+z4X3APJo&`a6uM+~a?I(yx3FPB44$@?EE>ytv5u4iB`q;o
zBo?ExfWb$dTVy`sT%flBXR|?hd>B1S+(*b#1Tjz;7-%Okl0xR}8Lg4h;oCmv#x8A4
zVpxH_N^6R`Mi&)`GRGSRJGeWr1nq`mAoKsc3xo0W>NEi4^qGfk(Ug)Dg~ATG&&Nln
zob(5%b91(KE^zgcEuMJ%F{GAAEr~u7`iK-I`*#nS&s!FYmeEY2b<bl@U*hQxKF-Fd
zB%T~n2#u4P=3XSFo+vvo9ZG?wVlu7qyPoy!5nJab?ClHO-3H+V)6Ef$ZSZFSBMOXA
z6iQO)hu%cW-6Ug7ZVCqnBy68M4Iu>(5vi7(x%;T<3SyHIBLpdCLq{>RuYvoBog930
zeIz)a=@D|c-z1Qhwqcp(B^c~a2smf4eRkA{agdq~dG<m|P%7tUizVut-8p`!h26;;
z*E!qRVxuV0(@k7b`4j*vlRj@d+Qp1%^x15dK|7o7pDe_n7F$lNDGG|h&@~-(Rp-SX
zL17BC(V3$+^skN9Su&gwQAoUXgaKh$Mq4SJ%?I_eJ(3cI_qgD(eGekXc`S>;n7~Ml
zAmM$!S(fK)AF{H>_m)tFOsrHY`;IxsqG>4>3!==1#Zsw(ObH^jC=OYx*65_>_Up&2
zpG&;;=3RErKgQ#aU*ok`uJes=e4AUhUgsA-y~mYnTm0HT`~|%G6#xGB{~hho5&!Ie
z`@g|V#}tz>o4Z>S;}K%oqUs7SAOywPNsE+ldU}c}3NBoHoG*RhD}3QkzsR5c*`IUk
ztplEY_9KYs(AA9dm)1DGdl%yi$|$hy5ljAD2yO@$_jFyye7?vW!!%?J#$1S(MM+uJ
zIqkxX5xPcKB|__580IX33@xWbH(wwnjK?Fq_Xw>By(NW^TlhX^1#Sx2fFtF_Pb$4S
zK!B#Vfv&T3l|pKZNCBe@I=xP4_p#P-$6CI5{U-IK#$sut<mZ3>r@3(X3eUfIo#(#!
zEmCjMh#X8#E<XO2(Yt<(<31T5SY;@u<mp;RjB|h?r*z=lhJZ1K7$bl12fvS$f=~a#
zFHw)i1m_ryC(O@ggHkMSeA(a=E?vIFKl!Ksl&}BAH#s^vd7GR2-;R+t@U5*aKKQ}+
z^FRLjuQMKxvQkgT0V>d0xV3uaO9C{+=(%&}O}_ZWFYv}2Z*X{Yi1Qhx>$;A%U2uGI
z!r9pw^ZA1Le2%qY72kIsh2nvq|C0gtsS^*QP$_|Ro)jXB=y~({4JP9)E?>FA_PHHy
z6>#%=FA<VpzUaAq^9^)(glkvUSzoVE>J0BSQIYL3WT5d5+k0%6mC4=;Vz{qHUj1$c
zVM7QY2P>@*bNg^9<O?1l1x0Dly2rVmFeFh(A}L9H@F)dRYji!rR5i+skzQbXXjtlA
z2RTK|X60T)thHF%Q&%+zk=ud9!Kd#b#7;w5Cl=+Llwj}Z7Pnvf4&B;a{^39QM{GY*
zapCC+r}I-Lqmr^3<&Sd8qs<PsvDnsOz2NxpHuZW%T^qV~M$?^PYRP!AgPiT6wl<ji
z1t<4jr*AA9lPMqi$kVK!o1$yY{@pWL-*eiWvb|nVuNBmcm=@a{?KhY%QRzhec0E#_
z_b4(o#1y!`V=+Ia>)I@ckpfc{B%y}Hp3Eg*#{7vP+AAf2R{2n-mK9dMs0_u-paICm
zx^p&b1*Afv5ekG<gF$irham=ha9NG(28t2_sq$+QM#B){x3m(<sIOnyp}d>%t_r)u
zS6^XcvPQYKj?o&S1W;sI^FZ$Ny`^8w*&y=>vPax~!o3{Y6k>2xT)tc>C0*B1mPIZW
zogWO12k3Lfb<1J$!FiI;lLX|jT_Mm_mHAjp^1T34SLnh})D>OZVy(puReW5Yjt26+
zWFS*YiA3j9lgK^!zPFG-p@z0#r0pB}yL(JJ%enLCF{XH6S#PP$F}j=5pN`m>d;q}-
z?|bZHeEBQa*}izj$3FfH*I#{$S6{usYp>m7`}~x>qqlhanGf@u|KeY8^VRE|yLySJ
z1PgbR^Y&UtcXWtPo?t41&!&{)$%yr}U0!?bI$!+4bNoO4@BdHs_KujY@9@}DAD~o%
z-Hlzw<8}7;_DGA8bLAHF$H@2=LJVe~Lxh<3p4n_h+jhhB2%^YcCu1~aQBal@by=gN
z!AprW1zIUu?=tipB1r~Ffuq9%s<Nc6D<TdhCBE;FDI>1S_IFCdgJAoWGxzd@1&Wk%
zF%d#UDUB&>OgTbJLy(3=NPOp&@6i)5N^;UH`1of&#Uqbh<My4qyzt%Uu}#O=<kQh|
zA&jV>Y?<#E8F6wh<RJJVRmIuB$;m0p_@WRzA$Zo;*RfXMz2$d)=X2l!zx*rz14=1e
z*HPCs&e|0XLu-{={?YTRzxu0u?sK2xKmXw${dnK<Uor}#_~_64G{5i*zrfQ^J)JWj
zm&L@;NkC-Vpa%=%uA8%eaF_q^pZ<XB*RM01ondWH+qN`KOWQWg7juq}Pck~*S**Kn
zU6V4KWf*TSxPLP4j|WkWL^kb?A!Vc`g0qfSzxNtGBraV#&*SfVn$9k0_GZjl$C{D6
z@V(mvG*{w??e#zq6-gM7pk+hTb#$GjZ7p4!a9)$5<eewNrEPK?*0jNX#0!#$r05}d
zq=d34F}fr825);z@VOX_8D&aQWCL<tQ<NjL&wimLP#6$e;DaN$%za;4v2?9vcDkS{
zN2Db1&Z10_x$f%iXjk%<Xy>f$PKl>y6h5%dI`j5`J8vEE*7@6P?{0DN@<pnV$(^m>
zDMe2;Hk6|>3p2D}XD6HN29*d}bK$9{@)jEcZQt_7!5+yNs>wQ^__YsFk4JQE&+*|2
z-~Q8A*uOJJYQc1E%yd%`T+g(gvc0vz<`vIk@e+y?O4ac2SngBSIoh^m(X=22|6L^s
zG2yyC$MS`cIgm0dU^8l#^IexYuvtbR2LLOrY?;KMaP~1!4qQH=vaLxXTZE;MlUwe4
z<L1J?D09#Of*e7n^@{5!q{`l`07K$c_+htV&u-o1t<jij_dIJ=NmUl;Qemwl>7l?9
zS)#2KC?PT@VFirm%2*0X6yySGAUBF3imR$?l$0p4_uZ=1D{@5<Q_4l20DPn*_obCw
z8VN27$1F2XDJxx(0$ml13(Y7b`mUqzI;^z>XR~qQz<E?lX|n*KG_uG{85c5Q<p<+Q
z$d)h1hi5E0&)ViD+GKkc04;hdJ!4vX4s45G1nR0}I@(}Vta0o5ZB7oK=gPHneC*>t
z#ih$TT)1$JtB>uV%fKjAjAX&l?Yjh%a3>9ihlh-IHW}+gRU4d$oE)9;#v8Bm^{;=O
z>#tqs<m3$R6WhBRJo4x@%xIm-`Z>0zH8)?r#~W|lL5UiFD0yOSm-(G<Q<ZlztS<yg
zWo4hWmb0^2<^l{JJhBjIBPgqq(P%_D8s%h;RN$k`3x`(O**pY-@SqdHJMP`N!_Llj
zMt@WeQMNV<TBH<2mzB#w4B3?f()WNjH!%sk2kX-PR0LBaixSnhL|t)fHsfS=!ujo8
z-q_#c*`N6kPdvWM>FE*Q{?@m-d;1=h(nRYJ`S68O28qfZHR<mrx9*+eVFX011fR=;
z(`+`&ZHbUimN5o4Ha5^kGoJ^3=XZXGvK;XXpZsLD4;YkaOA*Gfz7b>O;>C;n=5PKc
zU-{bCIX*u5alGNbbO-?(8|!@F10Ue&r=Ov$$E?&Pt8Ypy6WW)qRLx?>ox5-G{EN?X
z<HnoJW&@|L?^!Gwnr4wXb@LgE`2yQpybnYBN30ISKN;_CtlE+T?`*JBqnR%_*xRQl
z6k9uwaQWIb?uOee4o<Mn(zcO%hi8~FFdaqe!ho!hDj`x&N`p8d1e_zIb>W>z|MK91
z%qFxMr5o(nhc@gGTYxbNZL)GT*d84M=*(RuMPeKjms)2(I|_^plp;_T0Zb(MIPh2@
zN`}K$9PTwcDAJeSc9Ms0suUpEh#q_52uh(DsWDTv&Fx!9g!o<d_Kx`QPk)G~-}fX{
zot+A-2m}`pBKyQ?r75H&DoHeoXiC_IBqSk5=BFoAD8_3WY;5gN7|EhL<L;dU-g;w?
zyLa|5MMYKC7;RV_bxh7r8OafKs_E=8NuD9Z8B#7Dn)6(A^?gUzG;~diwGM4Wwl2#(
z^NfN8KZF&COt)M8j}`DF24lSCy_C{`>d73jXbMA7l}Mvk``Pk!#DPy2LVR0HsZ0z7
z9J$r2mV04t#V)g_m#J7r>s;8q%SGT_6?UQNx%>L-6gxXSIvMAdU}5l`Me4l$8e?+*
z8z_baE)EXpE5^yt(vw1BhBp&JKq^UetlBU_Nn%=Bkz}Rn(i1C@Wz9yJ)ualoSh*C6
z;pd3l@AiYBA|t6;%u$cWn6kum9ieM+P7s~LFQLS=BtSDr$}^ciNrCr~;BzX43xdW<
zger1$*|Ho}Jt7{nWjx0&@tv>y1)uy!Kg;gLP1ZLI^RsjKGtDdC-J?s-@yrJ<;L{07
zeS-D13C(H8sID<mWcAec8X+dm?2PZc^fE8M_7X3?{2H&k{3b_-rvxY1+&IVko_Ut7
z&26;NjK?*-gIjk`xcB;7jMOGF*4UY3x7=c^94BY5a_+(dD#+kz+YTQhP16kCNI54~
zCqY>#s<C0ZHlm)aV~PUlLq<?zf+UH-6QW0=ktvbdhQ-kVkDlAfwloM-1aDhXwD~%F
zf$@nbvv+_%4EeHtI1y(QI3+n)Cgq|&$t+7(p(jxC*3n&Lq1ivZ$NJ@69)I>3T;K5O
zi?4C}#vSkhM?_~aZhkF*iUFD8yL77i?{U0y8Cm-0r?@or&Zv0XdP0~FxwV<UB1BGE
zsmhXhobm5|_j9aotn<MSe27H00&vmhVm@WwomK|tJ)ix|XZhGiKgu8e@#iz=^T+V+
zzu%Bja_Q0~o_*i@c;wMXhgh&>E8%{xM#zB|oHB}jaBz?7*I(t^-}x2?2m3hZ=zBY$
zbqnV61&c*)y*7&$@8Y1?AByVl<2m+|@q>@NA?Lryl${Muj*clx&E+dQoI8IGyXc7C
z(H>YLaCkPStUQ;`jhId{pIqrg8mzlQN_d~hItRg^UwG*E4%yctZ|uvg=T%`B2Ly*S
z2q-crVY?0!JZf14rnur%N@YOgtYMK1m8d8SODY1<1%zESo=C%*-E}>E-wzf;kr0;R
z!H1r#AP~JG#(*jnr811mDcki0&Sqzvo!#K(job9CLki8-_9j(bQjcl~5~B@eSt3XX
z=MbXEV2enUqY*+I0#I%2ux&<c?QFBPy~)AeEoKL2+`YTcts8f-on?J%4J~y}y*NtL
zmnMwl7-J1?(IDh0QZJB2|BmY&Vq!L%(=`ih-x9;p3pY<nZf}NF)^2unSQc_}RnRSq
zyaX#=oglZfmC~4^xL?@m%*RX1dCez^%-;)P_;XN14``_vzzyVsMO~(osnu$m`N;l<
zjo2j&sp;7J-pg!mZL&TYBZSFbE~EUTin>NCN#zq2IrbwA*@!whWQaI9C}~qRBo!i`
zTC(4r2PLs0-|&!6NDMcs5ZU=V$wX3;UgQvNGkA`r;VF)OXeSP8q%=I?Vo*xO;oM@h
zMpqRAiPJ@n@QNL(C=e-SlgA_o&LY}CG@h>17_BJ_k>&2P=4hr_Sb;7hLgm+@%D~oC
z(%w1XN_iFUJXJN}3*UKxE7va5&-OXovlQ|og)yAmnbAKQ`HQc7ix2<wPqDkZ!)RP_
z<IUH(^U@6jk?~~AAOG>^`CtC+@6ff5&GpA9#Cgt7KE$LtN8d>9z9cv_l8_dRryb6=
zgc$h8|M)l9-nzv1zH@`G{g*FuVbswcd<(%%Hg~oW;t(MOi?+eKK-0|dE+IlztrkjC
zOU-Dc8Et5$lM16JL@7be5wRyoNV+6+mW~8kP+5za&(YnS>7^~a1SN}nN(?Q8hDeQI
zgb)QOW%8&DlH^@hSr2!hAV-A}HHE3ssz7)P&Vk&ZT{LuEVx}~wD)8B#`!Gdec=MGb
zuD^7L_N+$>fg6lWg#@W2Dn!cYX?eF6#Q(jHhZYjC<jD@mT~a8KjmK>tnbWaE`0~Lk
z^L|X}xq0Ue{>{JrpIIAEdGzXIK*C5#h(3$zgTSEBD9-Kd^7nuJ*ZK07ze3Zs!_WKi
z7I;Dk%Cg{7zxYYs|G^Kid*K2rG;>fJ62d@|UJ|DrXQ#(}^{Zdv#g|{?jhk=coX1)>
z@K$Hc&d!+4<}_{gsI%5XP;V<1?<a6Phz|a3DR=+&j%6M=fEdJkBIOOQTQuz5xyScv
z`1sF#h>KUxQPqZ*W({TnZR|N-L=I*nws%x^Mo^NF#z<+1A$JYDgXj`6ytDYubP|UY
zfhYoU$eflUD_~Q~PJBXxjEbT#RAtHO@iFDD#guVyx(F~NUFbq*doZC<6hsMCRYQ?T
zE@Hh!y0DD5CIsmFj<#tis|u|%jb+(xd~hgVpxX(jhYjTzs%g#og*A5c1&(JYw9SmZ
zneqG!FY?;!uduPb#<>eST)ljS$DVkSt<6m)a)R$$WY<C{N#0}I9)U$y6{;v%AC1^O
zcb;mkMvn|zVwZNN*xK~8=fQ7!=Cg*AqZ8~5E^J(4v)W-SrvwM|9i^O6uX(C%{f;?z
z&a=OFgzI}qfx41IL7Sf?Av0f6=zIWGDkn?`DYIy1Na<y%Bqnzlw9#l|hI!0MGRyXE
zzTX(`wIu>L=nG^dMmczZX6QH$#bie8)JiBp%Ku)8oR%e|7-%YK2`~M?<6RbZ0-c~g
zKH=^wuP|NTM4E~)T1Sc@FkUK>l-Yxkr6c%Ifu{U`X{9r)tR?77GW10H;Lv4}Ej0#9
zl%;Z2km1xAm()UrKDflFeA@Be<9jzKRy@JEoG>9-RprYm63Yd1sLoYUpv#J*`JAKE
zvz$)~+0sKOn%-e&0b6FzJEH`p)CdDn6tvcGICE@lOfkh!*r|f)q-JwXVEZGqcf9cR
zZ}azm;}<ak3h_MO{`yVYMiSgQu_?Iu@|>Uf#3k0M30=3~-~NC94Qp%beC%T%<J`^;
zwin!f<G-T9<6POjz__>wag2A4u7$G5DqmBSBrCDKrtK}C_~g@ExV+B!t#w}b(kUB-
zXU&{(=jeN!Un^MOm>}gbeFy_e7iqf|H?*mxf})TEN3LKruJZ{^s^J9db6a(=u!!F0
z<PB>v-ZDErqAm<&VNgO2oDYw){h%7l*Ty=J!(}d?_oR?LnC?G1;r;_42a$}*Ryi(0
z?;;*pbR8f1`H!%@v(4?BZ}HMAuW@j4lCK@0MOK0(DM?BV9)>^JGXJvijtHU{mnwfC
ztIQXRX4tTn)ZfHN=HmOc*IwcO^KbrV{?-5f|G?U0eGrUhZk|jMVGK$Lh=I?3=CeHh
z*#C={Uilu@x(BVgexzf$`BhcPrAwFi)nENJcF$i#DZ{F;6GQ7hc#<g)F$*P{X2u`>
z;UDtNZ#~EU;U3mndfR79lSRXPHsko@jAqebdxv-a0Z!e6kx%b#LI26fLk<UbHF&*6
z5-x%ei4Z+q+j8gTJ>J}1=kn$AY;JFI`N^la@xqIgMEV%HbJ+95)hQ&wGAl>Q9LT;b
zoR*7*c=$36mL^GN?rDr6+oFg>ijEX)c817~`V>{c=H?n_$NOyCZs6pJ>~p3Sa_EEv
z9}xv;r9mjBV}%ZY3iPy?c_8>fpgnAr^H~<Z6=g+fBvq9O{tu2y6l{t$W^+#v5vy~H
z*PF!d);WsZa|k?b)6(}H^ZA^YzVSL={^PGvPfOO;r)-_u<guqN^T?$uY@a*N#?~&A
zszwS8M54>qR<7%_Lq;Q+&l+yN^#(U@y~XWY`}BJX?KTi`OkEhJTNRrd>vRW}!&`^+
z%>g@VK~*-$s(bsClfk0-A`yK=t3(+CeaziIM79c-X_*Inp9{MT2N8f487R`mpiD_o
zR&P_TCY6(`a=d0v#egN^C$me6%<Xd}6;Vc#iGzAO<ao-#d|yUEQev>K8Nfj$SthB#
z-~;?Fmw|U#*nx^5U8K9a&z+ZFMwS)Tl`DwYE?)&v8GKL#DN%`mXgJ(yBB6);PEi<=
z$}lH@FoNI$HoDwvmbr4(O5t7he33$eOj+D85SCL2c;92Y9`9Tx{|{!yDOUk21?<XR
zMP{zw(jd8y2~!wcOdOufi3#emL<1ruY!_*|gio2f7i2_e@G%jE;kdQDwRgg!SJv>}
z=J|>m+xJXHEqBgd<=ooS+<N^rZh!Y2oj&7JpZX~d_Ktb&dq;>k!8Ijsyu8ohy+7mi
z*PiEB|5u+x$OS+5iBEIk;x+Ewn(@2;?(^J!GxF%gPqLVM{9=qzH6nB*0+br>Izr!O
zli3(~-%q`dpZ%rxbAIy@FMQ=CZom2>m)AVa{`07Khf9|>8BN9rk-giRrYFUKciHS&
zNl;o~3dv|xGo6kZjm8v3&U8%#toO8ik1mQF+tc+>l-a8Z*gx23Z8W0L8ZGm@_14mL
z-EfKovE+e>`v^^55Ho*MkaH53-S!+4qbW@}ps-*`9L-L+^5`|*_w2JA9vyP))*Wu#
zxr>YW;gN_OJ*G5{M3<l^((x0t?ynoGlm3!3BJL~fob@y$l*%56#uOk0btx!R!E@h!
zj^F<6&+%LT{9hoEC?lE8=4`HS5<|#`Jt4UE$Tfcc=YNiuzW4HASe#bjBtOcrbYq%M
zCp`Y>V?6rEV~pxC0+|sSvFwu$a7|1kJZEReym8|?-+l4B>>cgXwAtLbZ95i=+;Uwk
znvBx*4)3!1RR}du16T!&!%{T#lYZ>G8F_<_xveG0#oe+`38WBd7ae=Iju}@I);A5?
z=eN0cXAM7_)A&Tw1blBuUV}-H?7#)rIgu!2e^inEZQn8Suk!;MqexK?V5}vjyvC9g
zQDo!V$+%$AoFRRWhyjPh5trL02%Xn>5e6PuL1`3J9U*jd-ca=(>(a_JIYsDu55i^Y
zN>UW%_u*`b9wAPdB*XT(bFA%-DAyFP-FzLp=kRi93$``MmB_ZlEGkB}A{=*|EfVeV
z3CH`Vyfpb9yBE&$$YYQ4vmgB#Lh!@uRg85_QI>?5XwO>iy|u^j-BX&wo^UFW?FbcW
z%DQAcEf`P7bS_~#2Yp8sX4Gj;tvkv(JT%{e-U=3tMhS(|5vlXF4@2gf@F^=;2QI7}
zgcD-<IkFo^S(IpF5T-yGGl)Gi_^`Z>(m)l;=iBVXiyT}r1R&IqMmEHpMzOk<OFz$*
zIF#YJS7M-#ErkT>FY{skE62ObOiB`!5>)`XXgIq58W_p0POM*@ttK~0P0)rQ6(T7_
z6vISh6{>grAf8A=M>(T*)p*3o@p0A?#gI>3LZY=E(ju~dO!6^Dg!S3`r|WvUwx#b{
z`rZ<p&xFGmht?~r3xmuGmZB^&3~MyHEI}6pp@=BF7o457=nX|e5c;0J_XKpluzet;
zNGKCgffES_9Gsrve80lwV^oCT5aNK1iDABYl}bFy3t#vPKJkw~O;H$r@iQObWcKIm
z-}4Y_RNA2LdkQl`E63$a8|>_?@x2#b<ahp;|H^m2{T4><&@BQY6gb;qs%5}%Mcd{Z
zG{uHQ!~33ol#hMlX*%2U$_sDs%|CsPs_PK$CMUPQ&gC7))_S&VSQ37_wkJkMOgTxz
zXhBtJCX<@!bdB+7L}5(!1B}^v#(76;EmKqAg~E0{g((pskb<S3&)8a9LkUPR;(CX*
z4%_$HY!fk@ss<Cs6bbz@>BO(lA*JqbxKd<DJ0+3xGo@sIc7mSNJoW4|^j*vG(Gl0*
zxXGe-+3j6`k%~&lT==I%PoyPb2hQD3#$P^!xF3QIgZ+(=LjeZf1s01AA!cN;OTZ{i
zbe_7L@L&JzzjEb~Ykd0CpC-htv~|`}=pyrpfWj2~@-P1~pZo3K=HT!U_|X@2%lE34
zU}JraXP$bB-JRX+%eW$+N0xphB9nLNx&{09?(o8k&-2EO*J)jc_ZIIgP19t-#G;{D
z6?HND)Wsp1k7WyJIev0OzT2_HVdT)CSrX!xuy5u}CtM#m*gIvqwuUY$)^~PT+uh;F
zdAg>fw-MKCqSugehPadxB@7amVX>43I>$pVV&GZD=yU5WNmA6HJTbYPFcAVNN+J=d
z6r;*8?-ww5ae5RkMbuy?B%~q|a$65Y5}HCOs0=O{N^hv=4z$N6R>hsO0fEalL*#^{
zhYPzz1MD#t7r4B2j`LS`nQYe_ADnRatv$TW$WiPyB57o(8HF_}bB@-Z(YFo#V$RXt
zUFPnP?cKYa9xeF5)9)t{^Y2X17lCu-h~OjLtmojZBWCwn>=DRCP2tzkvSvCNv3YL9
z#?~6vMr_-Wx;a&vF`}h5p0axTl*b57CukZ)RZ2!epmo9zF@-5cq)kM@iqZuNq*`$?
zy-yUT%7vZQV6tV}{bDX<z8<U!t`OqNO#_+0<ULpjx#G^Hu%hfE<NX72BISZ_Ron@Z
zR4aRvG@#Azb`^M6g`Jd&l0+p0!a7d(_qih?Vq9_VsiD^#BhqM6$|1cm3JReKG3WS&
zrM1QIVnqV$?8?#l^!Nl~q-z#*CaY{PhTt-R-u6AVxAbj?U-Wdd1&jHdX3@~KEpf;=
zWw?+G=EZ`nMo9v6VK7yh6)wh5O~w@CHDuQ#g=A}MgX5zUnx<u3l=u+PS|UdZuQPEL
zAMqg(OG76d4o}xC=-V@vG|CtSF?o!(tc?_Fy+DZ_?wri{^FR6yzwq};qICSiXFtH#
z{`@sw{oWp0By3|*dX3Njm#^~Sk9>eX`}{Zgz2E*54);AWOt4)bIuI#Pl@Tc$jH+{R
zGkPdk6(9ZRhxp{DKFH;(=XvF&xA?;Ee;)45xNH)4UjH^Damckx=NL~C$_yumlntF@
z2ndlfU$Qcc#}(7*nDKZ*Sywq~Wq_iCcl6e>Xj|4zMTmnNdx*qn3AO|8*qn~Rdm<5Q
zJ=XR(?^&iji1a|qFXfiKA`YHH867eTyda~#N+Oa(B#lRK+%C9&?-rMye3J3ngyWMF
z-gxaghkHjW@@io;g-{qFv3BraOwgzNbMG@||L%+>Vfg!*GA|dP<=2<_in)EBzh`;L
zmYi7Idls}9U0_O$l5^JA)@iyqZEE@L-~Aj<KJ^3_FI=QFCAPN|rX+@l)C#2(ANj~f
zc<k}V*grTNL>52#i;@ylWzFSFm-)G$`#HvIYbf>Ll9?UtMM{8#=(zdTO`iMCxA@YR
zzDV13xM;D?(zPA)*_^Yp8Ew<jHa)?|)!Ze91O@)i-G1-+7)tJ(Co8}xg_Z(ECc|Z+
zPv%@5pPXTC-NlhGlQEAx_9&sX%-ue{bJ&piStN<k3N1BC4mqGH+hJtivWE(#<vI$3
z{Z^*+<gW!na5!r*fRGZS1*00myvOwo&UOT)vPe%Pv`8qOZGoI0#AP~j#EPn*7A1YC
zakE759SJ`eyl0QuG!Xw0f=PU5QU>7^+#8sxM1_WNExCAYlaJ~T@#W8do8x;eZjqI(
z%l5WZ5<)={HYvn5YYL|7JdHd;w1LI#3T?O8y}W@99S2AIeCdl{<o(Y+%i3ritfJrd
z%-)KG;}NAbNKrGI7F@Wr#+4_xx%B83XLsh@Y7RL)z0MBznC>XX>lH=u_Eu~NiKAo3
z{4`)TBw89u?Fe2DrqY7g=5AX}&bd!g5M08?+*%F6a`keJ(gi{d?NA?9_tg?PAIdJ4
z;Y7>SqNp^|Wd5B}3T<-x-UvmMf+*wQERfevAyocT>J?&$)LBVA+#A9DkPABz^JEBM
zh#;UjJm%i3*SYqQwft7SLkSRCadJB6?8ZK(teH#-rlZV98JAf_;;koo&uB7caXLe5
z!|dpUsV)g^&mx^-w81Hb?X%gU?JZr?GM~+8&l<XUn^B)ZI-Z8jNy!{yo9!|N1TAG7
zM;}5~nvkpLMbjXQ8AVlL%7P0!n~W>X;r;=^Xv(Uh(3-NS2s+X=j(Fy9SbS-)^8{EX
zWkU2#uE5D;bD<JRbd0Kk-8IkacVFki<|f~J^Df``{8#ztr+<n^u5PoiEl~yT+`5Nf
zXm-wD<IU?0U;3k4eEn<RrO=!7rw%2?NU706WbQ^v6lFo{8cYeYTu{}D_rLEEKK7Ak
zkT&wlU%bZG{_Au2d#60HKIYzwU#0MeJn`5jS1)Z6l4r3vTycB6w<wie_6wt^>x$8M
z%y=@PENiqW5Teh@E+THoyz8AungZuz4o!9ep&<4hxSmQYv~}14arFbT8e_y0C<fX2
zvK5uc;uVqy5<&`;&in+eHH9&QXhzV7$oFpEAk>E4N3YPjj^o2a_HQ3R$eu%$78ofA
z-s6PC6Y(U1c%X3jJGiULfsU{&?3Sq9O7cH^PD;7@d`<$11GzuLsaXh>H_W!}5psqp
z@-tW#1!YwcyywPSZ}7Q)_c{LM|L$K>nF<nk=LU1o{PWMBKhMWM{&BwXjc;U10Y8pQ
zli1nW;qk{F=Y8*c-^#;p_!80i{ETNBFMEgkeDTYF#tSdM!13uZ&h>OIs}j3K$85Hs
znKyKu#r7^22g70F;rF(@20t19Z6gAG1QQZUMhpU@1WE`30#Blk3F`xU2d8)hlZ`d5
zTziCzS1xg;4X1D3z{ZFp5tBfpk-`j{w#W$%MMi}lvW3eOyATON5=Hh$U0&=sBq4RS
zqf!AQ6;er(F&OCxwkOygS&x8(wH~Dtg)Z<RVN*T~YaLM{Q%VvwlaL70oJE>sBoeSL
zZ`C0LZ10Fs(6ttC-`>Xb{UPn?cbJT(+`CyL%ZREpYnRt}^21MZ^OZery}Cy;>(R;N
zE?d$NBr1{**a(UwFnS7j#60l_pZfy8_$xod`A0T5cYd2EpSgmm19$J;WbgV3uYP5Z
zINv~q5t0&2;M}z_SD)DC;-yW__71uA{2QFzdj(<NK=CHU?g&vOf_vyagvj3Ef_dW@
zjT0eAQlFp?M4xaj5o|8@g^48XG2{fKwskn4&G3B^sE{dGnUiU=QhngaEi<o|0*m_s
zCxMU(T`cX_G{zK^l^KEy6k6xvFXe9^>{bRb$I`EL$Yl>AbI)ZV-r=nD*Nt~q*bM~1
zWqgKG3gH4LZ{6kOW;P2<&LVs!d|DUy);GRG@No6f^E~p{IZC5wx?JHZV;D~+9NfD}
zaE_D1BMPOlRYz!g+O8*PLz^OP)6%sq^Z9~iKBwzCg3lX@(VDubv(<_zPy=aE$=ovW
z!LI`CV~q4R7lO_?f)DuABYH>ZT3RJ3>WcNrm_iATkB=aD>awIPK~@D?R-84K#xCe$
z#cA6i1YDffDBBag$@8TmvXH1`aJ{24J?A$hw~oHewebfzxqif-9=^#>|H@~0_K8Qh
zcI7et^MC#$-hBO(ySGmG^DkegR8xGjg^v}64T1-0G;MoIH8L1g(RUVA1hy}ZdGhHC
zeCR{(XJ>N*d*(R4eaavG?tiB4ERT*Qd$0Zp%3bGuPjB(W<71{HiHjDK)QSKshCD$j
zvl?(T8ZjA98P#J7U9A?V6cavJde>)5nv_Ud<3m~@ra~lK*HB1-Z~@!3bk5T>*>yjL
zKnNb|d!#lb5wN}|i6JV;_4r_i`!6rjF(nG4Fh*0Dk|YF&vl%xI4mkhxv&719e0ao-
z>u=DU&dJn6#aKzyU@>5m$CBvCXkF&P{2herU|5{us)&;V4|Q2&E&skO{$h+nEEog#
zEqi8x$hHkK8+vxF<>c@rx4?}i5OOOs3BL5DKjTBset=K^;%87w(Y7t4@feYE@hFAl
zlfUpu{`J59*UV<~9}CA$2*HI57r1)$D%;!JD_+dfKO=;=djFm4_|Eg+=K34gIXK+U
zxz^s}dQ0E-G>iMn*R1Yewq2K8tbr5t01N*o#od4VNPq`yOcWs!hC_;y*;^GMa71iO
z^wx1UTd;R;pY82!%Ccg*u}*t9=Co~St;9!-j)Uce5Sb35vi47XpNOtR_IX?i!&o+j
zSI<kxS*|(^uSueWU{o21j^KJy5rAx&5d>fqKIU9=Ek(9Ok@-GRO654tDlzgFM7U3V
z@{nRS8IQ@+wLR9lx4&TH5Y-ueb_35EhqtH5V#MzF3cI^o$e20puf6;h(MnQK5EUW}
z#Q(vUCu@mRE{HnM-8c96=GVT(wUcvPcx;pH?M+Tjj<8)%a)M^zDPlI}#R$fmCA(KQ
zSl=9>5X|nJaeD7fnB8V1j~MMF^r*`%?T0!HF-GQ1gH4$l<x@t~qRTnt3<(oS<Y(8r
zfKUOK5`FJ*J~0}XIG;_%MT(^0aJF69tYuMV=KC)NGGS=(CvuS{Rrd5-dfnCah@vde
z#^hMkBCLcMLdq;L7`lonMTnl5e2zL33SAVdHTee}@2ap<2vC9`A*3uKS3=?zmf7A>
zp4`3%Zwb+|Hd$j@Pq=sYh<mpWsmGdgyW7-K;6pCB+QlKW(=%o#r?j&LPd)k=^ZgT|
z?Xj*W7&EjnX7pW0*Y${$G7*0?p&m^b4V*762MID6@sS9S;KJZxHxvur4zAR*IbGKd
zxp$VnUm%U5^A^`Pn5tlBdy~_X6Pm6;34@Up*3~RJ&2eMc?^_Pf5^J*N!e~O_BVl-G
zq$Z!5Y#ewQJ?m=_Qp54d*Vx&4mY46eeD!y~#-*QXnO<DubZ?JO{p`=6gyiEN`!F}&
zyo+~pZ0AwZkfH^VsK<`+L=dd!%GJx*)bdaM*S|(*8>Usky*Ced`MDRl{pu~o*75YF
z;m*ykLi-%gKE1}Zi#1hY=vq&blBR1`b5xWHB{GS$u1AbU6RK*2)+JKvT*MJkG2&dp
z`y7?hdyiC^AS{I-4y4)8^(e@^5o<X;o$=Paefr)py>=~cm;>)1Ch*=Pjn3wdOC!%=
zND<LeQVeZ+tu!$SZtWi+#x;`*JDk{ty@P%3+}=Y6@CYhZP^7^NQ7ii3=_GUk?6R<1
zNss<+ZKe-KGOa{Q542{NK9kG8=YNwIHaRHEmfssf82p%IZhM7<wU)MRP^Cm0jTDNm
zYx%w3`#qj|>=~|HyP7vT=U}<XX~hRW_yMk5xx$MtzC2K*ezey`DaEBr7rAovGLz|a
zIaw@?!g8*%z<I~X=?UL?{@dKXcbnN_Mo0nY-9Q`ZX`7C&?FlZd;)Gynp!=7<uOGCy
z`$-Y_&Jh6@vrsDwP}9JJ5+ngC4Tn=7Xxom1{Uc_lXKZe4QBTH<Hr6;<%xSs;=Ou+2
z#6kn7RLP~e`S;=2asS#b1$@hqQX+<kciE0h>%3<vA#-6DJkj-t=yS&@1frBA6d~pP
zObN1)vB(sY%tbeprJ@QQ5l=`CKeYQp%v`XPJYCo4XYGNZi%7X<ad&T_^hFkjQzA9y
z!i39@UBX=4LZ-yg(J5#99^Xk)lEWjJ%V;46-<J@Fc6efb)bYLVzQ(*crs*#5#QQI?
zIBD>$M2iX^Bw9*T0cI@OzA|O!{2GN(bZ0H6w@<Jqcc|k&<9g2cJgBl64*qYy-(!k&
zw#AD?5Ddyy1`j*3vG8!X7lHwp2)H2@5hR5va5e)zDR`pou>B%m10M)3bHf?dKf*H1
zZE&8GStPQGHZ3%LJ)x*dimId<jZk%gE(}U*GzKvwqa*>u?2j9R&yBw*D9UnWe*A-u
zcZb*wq{AeJmS)PHVj4qebHBTt&4~R5-gv6XghwB_#_7?VzOyXmJ>{t9;ILypZ!ub7
zTg%zW2}VSoxOy44=qc-p=sah~C+MLK+jlKxS+R3*i}7T_Xgood6}l)h2WM#4#39D2
z?>pSE$;3h0ER{k7Wl>O%YS!1+a{5Hm(Vm`V|F)P<Il=WH0+Z>4t*v!VPfj^KX`pem
zR&d;w9NU7MXA8PuxG=66o0?>21Cj_ax*R#8E0T?jDoq<Zw$^$^VaA(p&Uj?!m)H*_
zuYc_&y2?{jiKnkV&c#bxym9Mge*PCfz~SDUljD}YkCbK27r*qU{KJ3r4@k)}s@Ax9
z<ADGCU%tSX|Kz(|y>gK+eCs)cZRt<;*en$n*9s0_`ZK!Y=XmyM!^I1MqU;Cfbj7S`
z*gLs5Y|8l*s<mQN*NjIK>UvCBR7jz7z%)Z#Tug+Ji}-AJ5)d+P+-g8~Qp{0d!8!V_
z<=)-9T)%OP)3b)HbK5A9zt4yJuzCmqp(Mfkp#^VOo4OhlM@nal69LScmgBbJ{3DOh
zNWtvnj5lxGpzV6f1hp)XLSg$3DHBqgd`e5ulTY@7Y-texjumylU^pyttj_TFWd`)}
z>r&)4sa$zWdIWNXT4}9^76mJBC+`E!B^L7rp(RCKP?kOykV<m>_1E~L&;KF6^;`cO
zDFxm+wAM(aiNMbG4$nOE3@^R(GU3OdY8j1co_gv@9)J9Cs_H&dJH-Irq6hnlljCEa
zfAM*~`K@oVXy*7BKtS7ORJv(ex~9YS_Aj?F8Nfgxa$ol0Vte(wm#WT(zWyKw@+V^$
zdB4mV*C`RwlBAxCA)`RW3^Lfxb8>LP!QKI7Q86CXY+c%A?~Ug<oEtXBf|1EqO<JnV
z|J4$y61os*^ZgN(6jRP%T>kffj>j-$hmtb-CB+J{OTiLdhYlWmE=-bw5G5f=dh2pN
zwy_|K)d556f~t~uI*{2+KZXSf4{m9w9O(KE=lt7V=Xfl+a8B_0Yqu$?dq^?GG3D0R
z-(b65W3n~ok*nvaixGeEl@~cWSkSkgWD~^fji3M>AtVVw0+QsR+wXaEe#G7zCmh|J
zQ`H4Z7Fa8AN}x*1XtQGb;+PM7^byvL;pEm4``7QW_xc-5+yRr(jLCY-`i4W9o){Gm
zW)i{B(GeY?P<SnH0^HD5D2=2vilWkp0Ev*TXGqxAQ5GdtU9+*VMzdHDyeB$K=v!>R
z$j?<A{BfOIDjjk`mhNlTGcRv=1_#Q@S$R4b8W&XKnrb?s8dVfkL0Qz8(qy5A5QG@A
z02H!$aRgEFuCa!9SJ<hDP9U7j9(6+DMZ$)J8-DTpa7J_D7;hVbhAWSrV}82J_1ACn
z)@ygUe&dj%HgSC1QlO}{;OWP<c>3w9OiG0chS|xCqoZ3CMlr5xDpOGJ?lM}RP>si^
zDqCvENRIYO2`PsF`<Ok05&;)1BBYgKFNPcerUzSvqEr;cm^dm~-<UF=&+|vKpm!Gf
zK<np}Rl)Y!Ci`N>y;I9^6FG{Ko2}z4Bx>DL7IRuLVpP<`lmpUbkeHzJX%PtCB4y^T
z$qmJB>bbT5M~thhJTbk-NvPR75d7(H|0&OW<fqwwbR9u~sSO`}-(y5Hd;9xby1UK$
zK75U@edRfB-#%vV)&kKf-uli>+SiWo$4Bf=V0WWp|JDn<^2XOVSKQ@8Pi=GY!Wza1
zQb<^D0YPUMRCR@vGo*|dlPC+t#@dK+RWYh+3SA)4IKU<eUg1JZn`T%RxZu$kbP%ZM
zAoVyQ&`BX$OKd#XZ`|VgtFJRZYB@I=ac-Q+=n9GVju0a@27C+%mldo&6!>n?3nfqi
zFdk8Bv@DRpb8kLlF<obLZkKS@5$>F@zjs1AaQ3Fka2f(#(v(QX$>4$qm9hsPWa4UC
zsvX~>&GDVdJ@=ob2ljQr_jOD7p^XPi3N^Hna_c=?l}IVEouxEtmGS~{$XX0Ykuw(~
z31iGEbwUV1Bob3jph0N77q~@08b@7QBy%Q{G1giNUGhg?{7-!RlRwL2*PhI1N&q|*
zx}Yj5e(_VE;@|(??+pf@Kf*Z@g7X)4dHjjTxp?_9E0Rm_q>$UJQoww6#;dQs${&CJ
zk2#&65>igdan8{$<}A);EEXM2V~KG%7!QTpgMM|(F7MFQd0RpEV6pbFnB^zquO5-S
zx%UbBl+c9Lv!;Zk){>q?hzZ|$_U_!J9Mz!DG2K{4OvW568Yah)&G9x`8H&PC6q>>a
zl#VD3QayBYrp&`~DPTnQLKPxYS5oj9De42Av}l<S(I5+rkp(vNMAt#Iq+~KzFUTyy
zB|m%K2dsAlZ9&DnbfM6ypsvP9nTyw8BR;l+m?9B8Sli<K+wYx?^)XLBeHmeU?!Nju
z_N>E*p81_Q-~Gde-D_94^u#qTpW8zJ!uvQpIpyH+gqyG4r8)8>D@ZX%fM!=Wg@}^q
zDntw<>*yDO>$4Mzv1Iey7NhkE^|a)XM=$WmBUiX|VF&G7UVi>LPH)|z-9MnPCzQOw
zbR)8UQKRGnRRB>U&D#%0O2N*W;_k7h^M-|2^r4^-imEac6Gc@)=smp$>jU09ChIkI
zl_ShtyP%t$X2*mOh~5#Szuzt&+N(+`mY#KD5J8B6s~uBZeS{FPF5p~CS~Q5+Ddxmb
zR~2PdQIAGc^@y@6QNy9RcP^)82!&5Qd$$gF<oyj|{Da^3@6PT-rj@B4<cG$m0l6E#
zrJXG}ynDc8v_*(DlS#$<-uF0Ln_C>7b?nVL?CApENp#d)y|~WPPhCW*oST?>$;QSe
zo7+!P6pG1cj6`IAGzm$o%#8>R9kZ`qhP;Q<sv436$o>E5J;~iy^Tm|#!I5R^S-ApP
zDLdI*qn?f#H4U@bjQM;<-}hK+DeH=QQgg1_;`S@2+&(+u&<CQ>l=L7io<KlgZH_w8
zLZOnxlV4NLt_3Y5NYB=~qOK%&_ut^)&K=e^o?x<eh2v(Q>tFvOFFhydqi1txi|6FG
zDMvM<@tBL%MPB&Ab@pC8<gM5DSsYuYy<kH*c4*nYq&eOD4sX8kEnItx$1fH<cJTsR
zo9if*U9f%f2nnZWC-k<V80*!>qNJiOYbN70M&+2&l*j>{$<`x-hmR9P4<QIdEGR_I
zF%&)$e8g*o^^xvi&W)E|=l1p2IX-E)us-JMg)2;^>-eH1kvR^&&ML#DX|3%-=03FD
zedQ$rWdx{%L}H!gpq(SOF4K_c7cKiY@8CL%kDyJ74S`NZ##_6DMMLlV%xRE<+*D@e
zhVY(_hxT2u%nBd6c@GwM-}hJlkUGOl=O3jqH%G7Xn)4)RY0yHUwPDF6TM8tc%UPy{
z(fGJ5PT~WFNLqblkwrm_p0=^f&K5u%B9t@{&)xky{N8`~eg4n?$Nwo8hh#6yBqW8>
z{L}|Oz;rs{`1tg_orJ$;NGZ5<=`y>!yR1#8s~aT6FsQvVr|``iH+c2cSGj%XR_5S{
zfD4Yk&v~%UX8WZS62yU@ye#Sl537g4yXAOT#PT5L@!|0kr0)BTe9c12#hp!HVn*up
zz>_OP_P-Q>NRj#JoYRvNin^p2m5eqw=-LC$nwEYr3(n8X5CbNC7EwH`iiIpL6A**U
ztyDQ6b}1XWH_d`*=OjNSjVqL>Q56saDf)qLwZejkWVbXQBZ?uPIe27@_gjHNp_L(;
zin5H9rK5LQ9UG&Ap@sSI{i3yEJg&HKO<?Vu`N<)65g8SR?o4oa3t~u|a~G&a1-siD
zOr~Sj>ou;wa1%QZ;03+wh#{{T(I=#mxu_kufKotRR*Z^@Nl|n0+)m~SwvKSpbA0m-
zN7rA6**(Vgh(aB)zUvunCQO+IiVx&M<n8Z43CTuPAwxscB#w1r>LpWYkd?;dfO$g7
z95zIvC=I62h!pTP5NyEPY>g!oXekidV2nW<O;Hq>LFu}*(#hhpp|D$W`N_|B2tF5f
z16|2`M@$aabveG&_PDOY)fJ{FvbU<YC|!Xt_#inuI>lRuc^7y7yR$nXn>{WQZ<ax~
zDdq_W=$e+j+jrR6(xh_}k`ioetT7&~ajp++pUo+5+~#K2l6+uoB-q&)Qxye9f~G*A
zLFXQ~1Mr?c#>~Glnj{Bfz?9o)iDl;J;B6))vJl7^b8F2H*>FOTQdYx;m`K4P1{33y
za-LpM6in8qjHY8Y?G}s0g0AaHM8+EvuB=_)wL34d*gr(bNe1;rjz%I_lqyymw$c)9
z6ux(Fs~iN0QbkH&q%_rT#d@81>*!_nZ@<K7=NjiOKS|>@IcpQg-+hfm*CDk*8$<Mg
zH<d;rSyv@bY;Cf&QQ?~dj`yDD)fc{n2yd}5X}I)c!G+y*>SB!2;C+ve9v>afdu-pa
zwY`<$`;-xdx}35$-C|TtP%=jxxex|^Pzxy&NwJ>b0bvkPV3dJDrX9T|2}KHt>#x4a
zcfS5ziqvxM!aC19`Y2a7&yy-eH(f^xMcd6-^8xEE&gKLX?*lS;lJD<xn_{3clCqHK
z!f@91%t%z{w%F@Bj?T{5+dm|^Y;BUHV(vTiXiPn-X!;J<W*@@dXRDJ;XpSr1(|a|f
z6f1NfC03My<z}vg$oz^Q;Sl74&OES4DU~;?7~{ZE&Ekg~w<8fEr;13K+qEHB%Bn=9
z%%98eRj&Sguq~JdK1<(Q77HjVL(_DO$2HDHj4Ao*SHHq%KJ{5X@>3t7w>_n)NHHL!
z<jR$+T)K3Llhf1Xt@WdsCn33V<qBI{Thw(u)Cg}^#`S&A>(^iB_1CX+a(bLq+$rY5
z&h>P?#aU1Eap2X-l}KXsA(jG^2Y<JCpe>tE0L!8|e`!g=&jO6UgY1DHXe`q{V<M;E
zr0g@SQntB}WKS{mz;oX<9jB+Km~q8;Ys%)@lwgl(T~FINrc;d+5+!AB#mXT&`C-U?
zxc6fsCl91F>{D?ip6L6I?(_r^94U;^QlQJc?-4>iM8rYVCI)j{1cZ=%K4UKIbj<v)
zM9N7?QlnKtVIpN|DO~P`gy=ygT;`s<eXfMS7|GVIqHzjT!^!Ohh2O#p!Qwb!Z6J!s
z)}>v{q-L#}Fu6Qs@6AKJ^c33Ac|}Tzwz0GePoc}4G^HZKM9QgPYkR`R#*}(ALIB5y
zr!@N~oX!FVuim2Hzss0CCgzB0a>UyCf@+k|I>+|K7)dITLwfHIB?Qy5M5dmujhv*w
z)GK7EkflIX38gGPSYpa+Raq9<kJaZugXnXlNLCvwOj!)Rw`Fdn)+NRiS@=MH{)AMh
zAxkz>QS!CQHB!L)ye2qj>HEyt^Uh`gN8i(VPx20578q0FgToYqM{AJy){?@zL+9VU
zctQxG$fu4d(tvgh=6vLq()@JB>G28Pokwa-@DArZK15{ds8T}_=NO}q;f%gLWplcY
z5{W{W9G{#K+@L5LWW`!3ilQJoPZ;3N>{2^CN5kt7I<r}&R+L&&xB}dL1x*~dTp?sG
zU5sgExzSon?>%Zzt7xrR-`ZqD$Xtw-l3)zywx^Ugk66SooREr~rr<h=idBF&P!t*;
zJ-xF7N}@<HXTB6#;%&#sNb2p1wTb3R2m7b5bMx!p1bKz+^N(_QdvhSo_6Xrg5wwhG
zsp;$?d#`_yyR$t+n4#sI^Xss?Q*&->olzywTI0#51t}FG1-h=~_~e+~-CdNDnI9*z
z;%hRRGOoufNA3(Mdwlf7G{dD9?<J{^G`6SLIZrk!z;vKJi^Uu<8UWTnDZdeu;Bfzt
z#iHf0r?z?WnQQE>Z6boeArQ(SlqUFmIPlJ5`;O9R1c~TcY<tRbQ!BKjDkMrKYzo|4
zG{~(@79w#no3VFzNY^_|%C~p#15_20^$kwW&WO$;Nkk!WaZsB<MsSDS_Py9A#qxPc
zP?kkrIJI7#<em2eS0ke~*=zBKpNT4EZj+W}K~a>$=~H5iCdNon<b~LlcCdAj^u0x?
zF1J@nS+;L9y|ZuQb%h6bl1mkBh<+#tCC)__^A;%ubzPCO9W66=${+sGf8nVoo}jMA
zD5)@7f=E<V#gk7y$%`+(#E-Iqky7x?Gf%U*zKPKVAYt&x48wu(*4<lt`@7%b)$6Yi
zk|UB&=+1fCu4l1m@%=zxU#`m;w#jKcddbI0tHWSU+mS2u93=-v*wSeIKG#_+d8{ej
zZ_ECq$Q#D}V_HaAMA0WG0u%~V%KW$B0wVW!NC|qII669mLb0*4!OqSYGns%B9G*3d
z$CE6eP!d&S`z0yfNu|gR47nQ+qb3dLQiu^b0JS8TAD_~kopN?I!ug)<YC=>VG3-f+
zFeL0`M~WB)(MP1R#MBX@!Us)CIp#`5kV2zHu43zIj1M9IJ`w1xMJUh@!9QtO<ExS9
z;#DXsiLM+cFa8C^rXwgv*9Y#r+;en$MlqUFj7HQOV@{7x+1Xj=@h7iRtV3Ou+_|&I
zom=}{x%4P?U7@r<>O@%!M&p7?jZjYW)|+=Y_})$SZyqxX0^jek5l^Vi0d;-A`KQ24
zTo%g-P4YR=Af~YLdRtv_A*oGCiDK3Unl_>r5;2PTz!d}CxW#BFtAe5|DN0S-w1Ze8
zp@l$~InAgnOX~5MqN>p*t6Y~}fWzg(hx@*EDJ6sw=+X>{7R#iorBEsk1!PO#cJy6G
zvsf^n&FPv3+xKYQ5@KXB))=E8YSyPywAp`GcmKN-Pdq>Z&38&x8VwTg{27UNmebP(
zeeZ~cM2V~#62T*4&sqVyYdQU4JeC~ZyM>A!TU#3_soC6KLz;pR1Gcv~Yq8d4C|oVg
z|1?nzF#13|6++}#A}Nsw6b4boT-b%>0vIw9v<ftTAm`NQg*J+8x@(=yP_k6{;T4M0
zlRHc*xVSmy?4T#2C}e>m5WL5GznZ6w-*}U?&)gjf$oTd-OlkFSGc43TL~_ERpzV5`
zzd=f0CdP<1IiFI9`&Ja0+vI>aJopK%B}zhB7^L8S3z{L*Q07o%KP()^Gy`H8Fl&MS
zuw-p*N<FSJ>>VR6czWB@^&Qrq(#96oOJXCK_dN?0kX8^<$s)9DNQZL`CMjI!=$aGG
zZPi@8vdef<FmF$2S~#^}E?#A;&t{z|dSdYOeNS1IXde*461&B40)`X<g_4BObG(>y
zXaiz>i^KVx+1Z@q(>Z<2PT&OSh>Rv{U<_@4G~9!ONisvXbj5{RwjBTFV=2bCcI_G;
z{NM*UIyz!+FK33v7*{8DrA~p2PQ91E|1W)AVz{4&D4ndN%Zr$E4k;y7QK5uhA$iaM
zf}$u$Df?<Ii@7D0!F&HUw9fk+bZ&G`C`m#OadjQ_sHSOJsw%UBt?T%UzxXPzUH?@+
z^nsru1cyon0n~NH(@#H{=O_H=AUh!>&p!POYm;?EQn@+^`RR-#de`x%U->+*-h73F
zvjalT%b{y~+P0%xSo+2jofuFT0V;dnrJ<N#mL54{GH=*QlaYv&&_=B~aoIy~`2HcV
zJfujB!<v<CMf2S=fZIQ=8t<>@F5mwKSgn1lzrW`r7sY{Wq@KGv0_P+l34+i3aIJFo
zcY&ZMVHZ96a6xzXgfK2C>J4mR*lP~B=rkhbC__;Ql*lc+cjP`LNFf&#9)&}qgv5JK
z1yowIR+Xq~#KFmoJM$9`&N`lmV@BuKDUAhdaY~?5M&b;TA{uNIMCs|pjFw3B6NJd2
z074{G%64B;8d6<>@F*1#$`PVP7hsCFzuXWbE?BgZjMN0P8JU>CU46=(o6lnpZy@C^
zYv-S(IX<C3-XgHcv5>??Pb)k2ZXY5~9Uu7VXV|@R4m~n#ZLKq%Oc7*s>h$E4(0Go|
zX0!(#cW%6nKfO+wW*Fv-?M=#!mdSQxZCg=}B6*1!z#Jqv5K5uK+sl_E5}g3GmO-0&
zlW0zRtd(pl&8Cv<mIafkq0|x?p_?@%pA&F&p{a_J$z)7bR@pkOPzbe@JZG>exMjv0
ztNS+VB?a0nW8VNFkwOnt3UYR=DF{MirV}P1uwgBB(a`rj-se5}^z4kLX;3Osm=Q&}
zN?rYd$Gi5iTP`AjbU$Vz-G84hL?W7#*@8vW6DJjl{0%}x3MH792ESgS3r$L%cF}XA
zX*oVRW;Cfdw|kCyZHsy|W>i$!q1$_$wTJ{p>A}!Z;bP3v^TAh(Ro%2~d4zmg5;FU$
zNr_el@xb>f?>U{F(f8SOS81IKFCh?70=ehjoLM^O5I*t9?u6sUbJ94<SRq29cb?t{
zWYF}75+OZuBg47~R!YO=JU%e7xmUk@{V7pCH1`S1!^@?3=sWUQCIZ}7;4QJBG+Zws
zC8ZR@M&VX%0(Yhe3xlo-q7XPAXgkY%(a?20^Z9~B+c9?uXBDbfG)^$XV}iiN$fJjz
z3sDneOAHF51j#RW^6^V-ZBDRuk*#eGTkf0$E<XJ%Zg_p(yFuRXS9~i$pq-y($a}Cu
zVIZa+EE@de95bP5+8MLsGv<p07SKWyfT%Ua+8R=*96X(}INg&)6A!M(e^bQlK^~M+
zipglgKmX_doPYJN{uT52oEKkwk>{R!j&FS9o9yrJaddQqlrw~g7^CUBe2w2rzSM&@
z7*a}#A|rKWSyELMWm!>GIX<T>N{X^TBD1Q!tg`=-)*KxjQJ4Y*OePb$zRO(2;diWz
zroA5uLb>|d@=ncd6zJMMqxAz)7=mYi|1Mws>X&%>iKi(Gh{0#BXFcNT)oT<*fp_hD
zHZMQOyq!CDj-n`YVnATvJ|()YXa8W27hZmmlldtwX1<R1p1!wqt)(Bt5-icTW$$%)
zzb#v>tN#=zoeRLcCuHN<DDPWrEY}b*sQlGnr4a|jP7am~`M{T!iB~_Q33}k~-{*IH
z-+k!ePrkRzhj)zS>W3JVv2jSGh&i4T@>-Xo#V3!>&^a+A`nG3&d_vc4peGus46PKj
z)}m8ZJ1Q;Ea)`!!Xrh-0SP~EiL5=qc=R6extqr4kM7_C%4RAbfI6djO`Nmz=?FQ>|
zla1{_C>X4wGW|$Vpi+T|J;}=)KN>-ZL=urCeEyS+m>X><iz@rs#X!?`#1!5!^-CIo
zjM-1m7|nF5h?gfI8_rH!7K_^~W)|Xkq`8PHFCx;IH7U^Zj`qeOi~@Jf&3tg~dyWn)
zhf4~BcQlIyGRI93+7{p3rtIIMOs5#VV6xq^eh$WKP>mx>5BK}<3<;IlK9X#;_ReeW
z5h`;VeFX2}*jg$t*pLNl8|xTjNinSAFpILJEDDOUpe}35s>l>1DG4co_xCyc45B5m
z+{cGCRtSX39YdjIzV0FeCSt|L6GSms?MOAWa3w}4%Bs!>{unV;3CWP6B1TD67AfCd
zTJ*at?DELB&3%6!`GB!&NCGP)rwhw$(c`>84?*5ifYBN=T}R14Zx<L-qV<TyqGvXn
zad@=H!QmboTNgQh;SyUr+vu`H7=x**p|HyfU`PoW1Ivmokp_RBrSWG%4i{r-v49Yn
z7&)K|%hpP8p84q+kXuzDgcZp-gplKoqQFN-h>0yT;py!Wue4{BAt0lq?IPAi?){nJ
z?w2;H$2D79TdZwtP?Qx-W9ix+A0-iu_YNf^QbeT4ennEqA?G#?Ij9VtJb7+mKr4Mj
zrtGDZ@CcDlHc>2%WD{-Mk&>V&3+&*4KdMJmb&XIex3oi|>w9`@iFnGg#z;eRW|_@f
zPEO~XoXl7>9ewZV`w^Y130`0vI021tj8O<v6WS>vjS0abeTVacQY%JUaeQ>l-sv8T
zvxffIaoB4fA5E~<<p(GY<k*x8gaVzN0^$(qlE`Xo37vP$tb^$mr(L8yo@0*|w5_8j
zb1g#>1gR+MG1_O+es42h(v#=(!SOfM#4g*9N+}9c@c5HY@lXEQZ?bdlJi8Y!@#Hhl
z@b~`yKj8TIm=|7nfj|4xFR-@$e*+GO6L!8oz4w3Z?|Qc*+aYM`YLwfsrlcxI)OF2h
zG-5OwWsy)>QkOMt(@<44UDvU`wmB&6B#XrY+k2GqXr&1vtP&}f-dIaCSt&^fj+7L}
znA}WE31<Us(~(cW1rmWV1z-Nk7x^Fm-rwi)#j7~iqlyt)8LnQvMqQUQiyv)ax3Rvy
zYCoq)-V8!yzF2VM#!YVAyh+nG8A(VfcNKbv9X?{r0-x+)5?4sw^5?((7lKwQ+Z({D
zE1?E1kQ&lK!obm#YRHfrin|aKDtNriX)YneK{WOwo3g)l$bmaZx_`j^;YP)!)&DPT
zPjZ2m=S>NTE-RcHeB}mz!59PHdS<6*ESeTmPf$kTwV|<=u_`i{V6rt-&OCp(u*>Ut
zl9{3q2n3PWHDtCXsp~Q2#wJ}zI3#E7343>sY11B|-evqqBn}&TG6TIsNMtgIsIzbl
zh$<mvAW2ISkq~kbA%#E*h0#SmbQPKNX5IZf^X1TEM}2XD)*2<$&~FgbK2TqcoSya^
zpPbSjA5n2kQ5~a4$Mk89)Q^Z>Wkj@0q;wNyBom!mm0jBml!1sc4k06|a+K>m<E;}!
za}S;7l%`|tLc!*aq9_8&1f<TCD>2Z(MAAf`&`MBj_YYm}d}J3gcRpguD0UM9K_-gI
zu(iE`aA(;7Is}x^Lot^RyQL{tmL<zhn@9}L_)x?RT*ehBAOt9Z7${(wJ~5!?F=lVO
z6=!UqugQcWCsjz3JB?}hnkoxY(nK$Cy~B$PUj6V2ySzf>g@hcYyCU`u_jQmGCp6A=
z9G=W+ceg3ZoU5mmhGK*iHKr7t%|PpdQMJi*V%TymO?yOV=gd#fkSOZ9X0o-hLbXUa
z0xE_3e3#%K;Lt5woXgic_}4w`_q0SZLdYi-tu#hy68V9TDd1gh3+4?=kYYiKj)bE&
zkxQcqc3N|`h(rgr(6}gYE@1l}ZzH9u7}ZmZl2m#`RFY=4KnzhZ4E8KOh81~TB$Hdq
z!<Ed3^{hx1$RWmMNvuvNp@)Abh~8nNBZ-I-0v{~xyr(b|HYaP8WrcN)u5*L{)_U5`
zVx7k$h_T1Imb0^#)3b)A_1T0tC_+$(SQAr8@&TfyG=X}e*{;THZ(rid<|K!-galFK
zd1(U&&7A1YII}ZsA4#pCoJ_fV`C9h(8m^<@{`FHzqtt!Ec><&^NJN_6V&w+Gj5uxP
zgtLw~>u@g8$s8Tldxxs0+`4m@+C{AEbKDNYV)@thjDGO3Ol4RWc9ZFpU;FjH$Ky{u
zi7c|>IHZIw4Z9c5v%R~+Cw~4DRQ2aLJ6rI~GtcmiZ$6(-k?-Mn5Usn9(&a^LG#W7;
zkE!c=^`W(9V{L=8K?$}LRcNhOhn=D*7>#SBmNd<R5W?!~#+X4+F)RpTFbkK8Fhu#a
zZI3C`a6%1eZHUox=k6W8_3dwQe)l3GK@kE%3NBu}z+^Jv@ZjV<otwXE2w*&!46PM1
zVclC?bnNZl<Evl$3a4jhSZ{|G5p-=v>$=Wit%n40wff)R_}_-qJz!s=wazg{!#!k_
z&YR(oDxkH_R1cXsUV*scHRiP}^ZWcj2}#KjUHtL%WFIPAr4Urb{q1JS4Gb|s%4Jb_
z&x^dY`ZNFhY*s811KJr-DIts@Dd@W<morI_Twpd^aC9<5j|-I1l=Xy@ZVyvygo3hw
z(qts<`?bnKjJTK}SfmCcb*2H;236FQ^^{4BT&xPpwGpb)y!P@vZr!|tug`J*S&J|U
zPb3rpqzC0QN*e@XFo$=hMQV%Ck{}htnuJE7kt&NRj42r<7*~A;DKb>BdaauuJZ1BP
zr$6Al-UGh-H@h@Hgps>a4lQy@1~DV4NeSw4!PfSOOBc6rt;(Fbm{3ZuGOht!NW?yi
zO%l0)6v%<oDiB(&s0Lynyb380QX{lMnj&`xQ;srCWSxKzOYJofmNgWR;;NR3DXl1O
z5Ku}nnpU7ov@yJ!<Nmuw?3ReeveUVIbITEhq7R^yWdG=lMcXq{1(Ga)&}B&yU_=5L
zauH+$MpsP6>rB@ST|dHEjV`Leu&-a8Tw=&}1J2sq8y9j=8s!tO9E1-`KQ5u<;My!!
z#i2+!%-Gp1YVh78g<w1ytqMdX1clZVT4Svp+LAgWW62?W&qNiRAJ_P%BYAL9b2`r{
z3o$q%p4se-$=Vd93_@mw&A6^}j;f3_Z8IFvm!`1G3PxsjNDLXtk_=gL*??Jy{5hFi
z&W!a6gdqlrQihNMZQD>*nhO`s5i!i?4fFW|>jJH{v~5q*bacI??HuzS`aYwM)`4@9
zgaHajl7z&Fh%!C8j!d;@JcW&m6V6pN(`muR=maU|5CyJJ7^ko;93M2K(9;Wvk|pPN
z&$08=Q(SrMab9hjRnZ+{e8Am3E4@PSt2<CAL!_s55wo$zvGp_yOFZikT^5!^4VFZd
z3S$bow#D>WtRQ8E=mW4)D!&KC-SV)Y29eLTt5^BuU;Y)$0|bc}6KZ%~l-87ksGuxL
zo_XdOzWd!*m@Vdi^B42b5I|uJqq?T5Dyph_8&X$QRp!=J72bQCb4;h}Kw>l+W36Mc
zSg=^ksj51o!CgaH6gX!G-kn_)cFQ7f$%l<07k1V<LI{J9!ecE+7108O$Sv6~e(}%v
z%rE^ix)3<$(JI>?ZEbDw<{NLl=X3K{O<GYFdF@Dn6nYYgdwX|y<+Yc1{>5)$UAIEw
zg3l3AecR)#8}8@q%eQQmF5{n8ze42eFU1f=Bo&3uX)4B;+<qRoi`r<as>+AV;Xbp=
z49j87=xtAyBuP>fC>01{$g=zq6?c+c#HojVp;C&?$p%8^L>n_GvTxtMMbmYwVEgxO
zEF~xRUxx?!p_Yjxh#u*Bs&S2$1{*x|Hn$2%G@a-8_zXQU2s5UdjyRc1jIbE#5CjU7
zh>mykyqAI&>n%u~Ev{0;*??66TFde2oY0*Sb)cRYE*0nK7LFTx5oa5^{){4wh|%Fv
zAQXuvH7smHa0_seRw(MSM;8)RgVPF4K+6)N3ksPJl}bxWSu$OlV!cJF06Y{c_W$oA
zC4zM*sj|XY1iXOhM6)*4j4MHK4Mm{{N)dd1HhSw;G^J%NBL<XQnQTVAtaGwiyF{v7
zX9z(q>f{hVl3Qrm4It*CjkGM{@_i+9>uJ?h41=M$7^28@VFsqXKp_axzw0gPT^4qU
zBm@Du6!T^1nzXt_LQDj&F{0%3c)@JGV5B{xQX>P|N<n4mzcC{OE{8J54!mSMs#zP4
zAgZDA71%|a3m;OBVHq|EQqCq-MpI~mDhBS{F!566%$F$da{vKEOeCNEYC`XcE|b@b
zDi@L|<^+Zq14_$G-t2m^leQ!UF-WaCrbe??MjD@Re#E?uRMImURV3fh&E_=6r;Mr*
zsx%0#DMl5AmRQ?kQ}(x6&T!@}4qk31L+dGKpExbD^+?}0NR{RE&O5YJ2(?Ub@bsNU
zSBjnUJB-E^HaN~^N1UGKR^@E&m@hnS6InEl-X?mRa1_L(ft2G{WNurFl%3ACmP~X@
zJ%!Q6nC;1k^$kNgmW;h3N{fvxsfct~MqUwT13xhYG>P?5#l@XX9)09d;>F9T>4?SN
zUD_ydTB1V9n2Jgmg7jEz^JbPH$w)Fi!b;}F1hX||?iO%5$1WDM&Sj)0N&4tnU*967
zK<XACGOv)O8kWSvU`k}^X8u?2f)C#JtIvtNp_R(8HD2eZp8X&n`rrr9CKom_CA3y=
z12W`5ns?rDa(n<RThVVP=Ki;i2a)SS8|t#6E^9{hh_$sf>N@8ukLnSV$z+uQTNEYU
z`Jn)Fj7EAe`M$4!b*{&SKwZ{|B=9bvs?2@OUQU_jFdmNyJ`$tP?FW%XK3&&=P$0^@
zk!nM5JukidJU8FE!Q+oViT4&26&q_)cDA?Q{@%oU%SHTtvsakBz@`{8Pbzut+_}Zg
zn{RS-d^o5ca~scDpR+$hR=_TKH3{x-`f2r}SMNb7l?yu|GT+7+j5aGPnkAA~RW(&r
z()R<eQ7KeXc~cl#Mau#&gB6hwG+w5+UAG_R5D&Z$WduBcF%(6?Q%^m`g$oxscW#IE
z^$m)`@E`u*4|wy<H#s;y`fKCW-u*~#`&&%;P$$Ju*rkMwo=QrDQY2-F$)bRi1TH2P
zO~ZIGq8=H#QGpc|3)fHxOC%sgCeA+`jFM75$oY`DdnslWrw<+%6MgHrd+&(OA7CaD
zRcchB**Uk)Njs+~A}RJH(UWlasBjLBPb{-Ti|Z{(CngdqBalTzR*_LDD7B<CE+h1z
z%!axuBTGeDP}U{7=-$J5{@XDUx%Df8M5ja{fJ(5vUNEUa2}?7dQ5Z!`BFBv-hCqV{
zeo6u<6-FA2DNv?ZabGhHb6M1h!Rlrx>ZHtEJAfoXCZx*tFCs1(pE<E9tv=$w*IfD|
zX7OkA9`7w(-?ErDn88-+2OaOupqS+YB65pi=w*e0vjn+yFEPlJu&rgbSg<w@l%_(<
zf{-k|w}_aKNup$dEXZ6rkw{YIF!%_;XQioa?EO>lKtt48p@zIQQ<xl-tFwiETEe43
z5T#fJ3Wy~*lhVqINGpZJAjYG7fkP&}hGOuwi2<br$|#%-^sOLRg;1cPBqYoFSmI{j
zyrP$y5EDgFQ_2E?$2JZ9VnL-#f|5inXoE$UhRxAVcByY#y0#ro<bX&lx+eGNlR?D+
zKF@t23YpKuLK%EaxW1<-OV+m6F;z)fRrKC+`|yC%(>cc{4M)c_+Sb!`32P-G7<^Eq
zR3N0z8Oai4^2p>c!lR|9)QPH)tc@$SCpDuDgISwUmjz|1AbFG(xF~5HY!`L55oufe
z$pUX1dK{)woV&Qe)r(t<*CmT-NfeQ;Z;15-Cj}--#%hF25~l*ajwByfXmOOD&{w$0
z4&At<-J2m6O`gM{%_>7gs)C}<!T`HCL&QKK@?yOtvJN}&FeDK}mQ?)J=z(DHdc)_-
zk?TdYgvuB;#%uiSM}CIQjg5gXMfR6^m{uk86SWMN*6Kc2^}QT=xYuRM9M|z^oVk`&
zwQ9#IrKqZkq9}6lkh0Z?5F-u_4(R(1V>D%1010bH$PW6h&m{kl7bs^v)(y&VA!!!v
zz%9!}Vy&|UN($MnUu%Un**vwBHS@)cZ$9@89(m+3@D6En;d5?hXDEvA3sipC#fsKC
zw@p&aWN>G>efuqL-+7C+U98;Yy$@LD@Xiy5TuQ-x+REDtjQcsUawvk9>zW+s8O9VV
zR7=XtPaKU#ROK=ZJR^J_om)z2DP-e55)dOnN0d*b<X7!t;Ju%>2MSdw1lkyiQUUil
zk)zRwPkiF%c;d+?xp?scTU%RbrMYqACX2S^bUq_Gx8f1K7q9W*r)kxW9sVxzuAYGa
zZ4jy;ro~`=lJF7Qwx?Tkd9AS{tQyhUhLAkU44$3wow=p?^Vx+V4Ne4UP{aEuXnW7m
z@eF5=sJ)_^lvLv}lS##R+#?G}sUJQ(F_t*1n4d&Wj{<!gh!U)b6e=JLP)5chK~*+X
zm8Y&fbsZUxBSoRng+$1Fa7%r~>}J0@TrQQH%ejAhD#|jTo|WCs+h+MeKzQk``0y1F
z1D$Af_Yc=#`E3vWxU{5p4PCZnr%f)^;rC<Ng7>Qq;^LOT`G|-LouHCXmWr))&8Uo!
zdKNQ_D>O+A=~+C?UCULv8U&h38?-4h`ksr2`+UFUWMnlH=|Q2GT)t05N@SF7;E>Cq
zpbIe}Vn7Ur*Gm@y=N!H5>AMAs*(tNLhArp#K5@i<>3CO#T}+WcBo1bXOJ0O0uJ{p&
zr6-rb2hYjb3G1T?Dmv6;9gxF>3~@vu(zR_?q7F*B7!$4?%=}z7(p~a+imIe23d*WN
z7bX)HRX%YEnFF44&pQ**2Xm0s%^}J4s8YF2ClRDzNX8I`sGt!1U=EpYoG_ReJL_r7
z2HPlnZ?QH}7|p2Gm_<+TdQc8iWPZ~6+8QD{nr^}I$tmYXQ$S^+<iK5uN>L2)BHm5u
z`z~9Ngn((v`)E?|8P=7cwaNA|LLiku6$NEkqjZT6lGeiQTW9PaoUngz#PLZ>+iH5N
zNT}gq8E#CK5W*sbql^}#vxHyEL}3zDA()P9rjwfW$%wU5AS(rRiB6EBCqxiF(zzac
zc+3;RFs?QTvlFBXguWrhj<TLG+1^A>r`WQhSX(E>gmnQkH8AjQ)NpDE-jQs`d9%Yk
z=tAJkSyVN`hQ!%yPV73Y5807|e0na2BBd}UJO3w-Q4*KZ0B~m{<H7N6oaB~ylfyqm
z4K1dss@Oe$o}c}>kD~`p(SwWT!#`!Y6nO9Q>1|Nuy;noP+t8m7?xQ!$c3)X$4fbd>
zTD4D$qFCJ{RaFsVr0cp|?2z-81rezgckkZC7?ZvE9^ic4=SJq|r0@M|euV(<Bi4Fk
z8<|dq7H`|~jc@z~zwr<L5i22y5?r`&VI>M${`?>I5UR+h>f{MAFgrWp)mOjA&6{tM
z$S7ov3-Z{`;)Z)wkdgJ=f*s0;2Lyb}7Od9UQf6skq;-MP2BY)yJRXmSXI!F<p)N~`
zqJlw4VS^($4+Eb^qL4`=(MS{|MT&wxk$Jc8IU@GJdqS*49^>(t^|b;xS$%zbdx!t&
z|MGujZF4<m38zHziC_H8XE{7O<KF&0;p}u}!}K?QJ@@}yzT-q*AC{k!5{#5t&}T~8
z(i6o==4r<)4C@+8H*0C@lCz0tZ8Syr6RH>}wFi-7D*eMstYNO>N}8MorEDZ6MTpRL
zk;UB;$~jaM!+1JpXL5nUXoOLu=<u;egh=#|Y>hwG^oN3GF6d*XKu8ZLMXEH#x}qBQ
z6jeu2N+y#LYilFM<C5{Xq8@9c9C7@-;ow{A7*$f0#c;rq2$2ZA$9bFAqy$EntAmP=
zU^Ka~8LoQ_`HY;ccNBiPen~O3%0fo#B^cy`(Yg6*EaTV)PJHx=6f^ZlDv44Nt>OPi
z-k&yGmZkT7*l$gHIKv%tsCn$_ZlJ+#5ELd;L6H&#%OEK${KA$TVfv-{5&R8gha>Da
z4k^MBvK_W56Ouwym?9;UAV9*T0D%MrkZklk)RdLe9nQFiHTlJ}_PIB!8;#0BHDsEr
zJ1aA9-h0kIdsypP|K~qcm1Z_;iOI9yZrN_uTwh)hLneeo*C+aZKqpN_W??hO=M68M
zDq1%XhQK<4o-!0CD#VnLUFQFk60UM`pBjU9Rq>a6w&On4Kp74Q-1MMp;UpaFpvW?p
zTCEPE$SHnIzF(e+7$cWYFS)*c%6_+^+i%Iy5q$XZ6?U;e{vnrWlU!^gk~sNKANNqW
zDp|1B+cn8s$cVzOZ3T5xnz8HU<}U&CF-78-82XNW-$@>2Rng98R851cD=`Z!3)Li$
zK#7|;l=z*S+otGy%C($!9a@vEl>`eF$!Jt8?%qYUC9UwtA+E(*nzD2DUC(|uG7gG1
z85XTUbpfZdz}`wzHx0|#G4-xywcD{-uUVc}Lf&;&2%1w{08m!bRts5xedKIep|c|4
ziFqVuY0ZjQr9i+|(T9=Ukmz@NHk-ivA71k0$%gAy&*(G3TgY?CyE6&lR*2cLGZCdC
zMnxJGXlp}VIqKTcG?vA@WwB^!nu^9*8e^m=wGI^`-7vD=Z!nw8wm0lQd;mjY)*Q2o
zD@=~eD@{AI94~9?wjx<esv8z(=Zr}Tk38AY7=^A3Ra?u#A2lk;EuAy;BaA829G|iF
z;%_((1HBI%^wHYL&-y_0Bhh<|5V^$eVA?C@$vbYkbEjAM(c|`aOdB>+xN@`EjGy?)
zpWwAO-yp?wK#t}3e|RKR%8B)Q!{|LXv0R_cP^e;YBEM#{8C4<TPaH0#6pO`zx~>nq
zMpIe{UDqAHKc&Pt3McgjtrA~0Yc2b(qp2nHFqF+$@E)ZkfLtj>3<+zs6h;yJ7y!d^
zt_dNcvf{%JKV-dGae8`Ah=JNQ+_`gC)ZpollNS_-K}#%9@O<k#-{Sob-{<oBQjWh+
zCWM!m7>AJ*glw(FL~^<aGAG51wgo*krubuuP{KOPVz#8VHMX4px@mEToxG)O8giam
zH<oR;J*f6$jHoGf#}p1w;jE^diN%WM$34!@&iK=R`cLyqzw|Qjzdw9k8^h7r@w4~8
zRs1VI|FeAe-S6}M2OskFZ~Q)K7(TNC^zbCzf`}44g9?seWOaRwu^od_Xk&{JdtywP
zFeLJJAT=HRVqohQ=v<+bg(0IU?gANZ&m~M8%ftH@%a52a)XbVCckaHxM<0Bb-OAJL
zN4BfLP#ZRbqj8ZiW=vFqAgRo(hIih6krQ>z>e_L=-w`fG!a5N9K<X2I=kW`Lsw92s
zvL2{vPtzD?Gsp2sLse<%9I8z1I>MNQ0^khJX+evfcuaafkdl{Q1Pm1^Ik5mji!cMo
zDwDLVn?@H?Z8ag*g?4~gw}`EmDgbjL#i>L}@=SoN=~<CNL>t*O*NsCNh<>0ScI@^W
zx_%F6e9#PD;e#fRfw_X^%y9pX;l%ARDUkY1hqOqQkz!LDLsQo%YY)$(QeexUQ<7_|
zQHA&ZEMj@Hfi24U0+~-yya9>%m}-Kgg<mao3$g|k)G7Kzzw5caykzJ^$a8XXLZeTq
zt9N-GPULeHb^$SCoG50o>{i6PbLt_>5J=wA&RY6$Lm2mrqo=A{Y$ZrR4yAQwjO>z1
zD=mhAM#(}`Iht9EsjEZlHK}RWTOee811C$dqz=2{($$=<!ht(C$**G)-b3(SOo_*l
zFpeTfh!PSUf<Gv5a}smo<OSK&+On8ckR4Ccj;6J=ZH>|}`ao5maMm;=WmvD*)a{&k
zRUPb1CTJN-XFDgkxlB>^WNCC~lTg}_u|*1P==X`;*zxpw!;`01TwZRtyxh_CnV1}8
z2NWY(B3dPs4mguACgMz_t|T49S<A9&nAH_^`Bk+wO<fnFsCXM$Wf)V!51w(o=W27o
zdf4FhiLNv39z7<Fo_f);(Ggq2E3cjrw@_Dxnbj<+3Lh2q`5ksK;cP?7O0xN!l>)4w
zJuX`mV$Pg1?7c!)j&^axcJm=!Cq8>BCSrlejjia4?aeR_SR({=*ESd;M2}L2+BwF!
zFFUQ<Q~r-0lUW0(;yRzPTGLb&%jJ?U|MXWlIXOL$Yp3x81zvfX7$a9#mx4(7&#rKp
zqNfC%c86B$Y&JV6V}-xf98f#0HH*cf7$^_SW;6P}r|Y`nJLlN#cI<b1l*;V)dlriY
zyZxTFZ3!_ROqi{;rLdEtt*%8eo62e%Lb08&ne}SVvRy#VTwFZm!GjMuJ3D6_2I{8a
z`1s@yTlU8_aumyo5F^2leEVD9;(B$(=m%1k{6|R;iN%naTPU#Hb7(^@6HwBr*xIPW
zoMen3cg{K~_>{sJDY}g|G|i04NtB>-mV;%Ak_{113TP>?Q@HUcOEJ+@u@n?HZL81z
zL>>OaY&PTd*I(m{Kk?2p-=Y-HUOy#;@7bNZy#B@;y!F<by!-BVWus8$@IP{K_whZH
zy5ZVQt-X{I!FzI6L}M`45xp1-%k-PcK9G9PxF6Zi6RomT#xZ)Rneu9W@U690*m6y$
zGZeGrh&HnsFTL^_@xk}mY%dvn$G8t{x`8e!beBkDCVP2xr2^J<oZpGm-g0tUaoh!V
z3&sB0F?28tkcWUDA^A+|BDu;07a6O_zKLwNfyyeZQ#hAtYiOF0s&Z6~W!B23-dTgS
zvW8phNO8pb5kDk~4hW@6n_oph1bs?bDaa-~uN1c`-{LMS08P%iSVznMlmb#B;ju;p
z9}0`EWFrI+V_@ie`fe}Or4S%%hC$PHn&1s81{$kq8pCmuaVnxqB5=-<>|%_<x{SM}
z{-1dH;wdRk6Cp-a@Q1F<^!!fOQ<XlntmikLl9ahXY;Uq`_{#mKO+eeC0Z{s2Ws-{P
zz^rMhoI<HYRa<nJV~zdssbVK3VN4mH6JyMbDKX~s4DK-{X*8`AX30a$iCw?vYIQ|B
zn_-KYW<Lx9ebyCN*@-&mu#E^bw9%Nl5-GQWsPZjdQvoLnteSE@#5M_RD*bY845Um5
z5u-|)g7@@YPv3QP`;KlW9Jh%Jk#iIkuhm#>vDTt1$%NC0vSoX{q1%Up)yQTgS#zgn
zC(N6Qq1)5@z^rXJK0T%P0pIt;rUhe!c&%fp-lcfT3PKboY)_AIQmiWH7$bCDVtqZZ
zy6kxT=$fn5md(DS>-&<vCVoPsjH@8$KxGDIW}tQkqchGxQ(M}mrl~EpGgOtMt>kOk
zlsrUdh5V^NDaj8VbEJ<W>+2o+i#6MR!;S&FOXyUSyN;pXu{>Y!q{-N9Sezem-Cxkm
zYAS80qKI1NXJ=gWBd)6P88LvhhO8vlS7k-xW(4I5j|kRd=J?Jjv1;gh&(M#o_XBgS
z7|G<Kgw{%SdHeNBQb*8SZ?+6U0Tt(G_xPDV^%ef!Z~Q9bxF;jZ(P8|gq+)g0G&pT&
zY6)?^bLS3U{>oP{w<Pqp7pU7N#wm-KKum9Yp#Xe_E3e4)C0U}XDvT8#-9*xz3cD#*
zY%1JtTbJ~Gf7pCXWPj%zi{*mtX3cy)r|)}~M@MvB$HPYtOS*})i6p8lVvMHm2SMdS
zo(_-{Vd$e+K~x#1HRCw)`0-<D%Vh}^Uo4i-6b9ny_Tx>W$T5<VBE~>J^t}7t_t<RL
z_~;2!Vdn*{3%>YX-d^UFD&%P~K2YTxOZ=iLi~kKmI&Jg>O|7xkP**i|T@?bnIlRZ@
zAvz&;LCLhA!y(e16a-O>v6#Cn5kq|@w^;tczbZ1$S&oj5I6FK2L@zU&&w1&kmw5B_
z*ZD`Y8GYB0O5EeEdJ%ZGt@_E1Q60|HROF?6<5wx%u`I%d6hMJ>6=MwL?^1k59|?me
z_9Ol185@hsHO3n$?9?qG*5P0UcyP*KB%6thB4o*Vt?G)id-oY%yoY(>*=#bqeaF5J
z3`FWNlEy&xk!%xLN0b|A=Yi^6F`HXvzGBlTHb=1ECHA|7Uq$*Igis8seI`d3ePkRG
z-8Nv9M2b}vGz~OOpsp?T%rY!0+Qw4VMmFNhn%Y(vQz7@1hhnugZDg|YC?y<u;Vm0T
zVt}vBRAR`Q8r5{C2=^>A#gg7!WK|F@R~fXjV7270<}7wzeb3krgwabZT*wT4rtcDA
zbRwXu6th;*x`>@j`<0MijMY>QTqSFSDQKN4g`E^{lcGNnLo9#KC@X$-hZ+M!V5l-#
zi%3HJ8I!b=bD0+o_CC^a)6N`Fdt+pyH|4?3W-~HQ+SkSqdsCb}KKXb)j$L^=q@5+Z
zkI4-umG>1CO)gm7)WjT!$zu$}kXUaw9PPGL?o>{+Ds9Ou#ZpyCz^<W~32SvQ{7a?(
zJxzYr*pg7dP2W1k!UYfq?UWO76ci+ez^?1)x{mAXYp$=aMU(+7>IFyhBj)oNRprD~
z8A!nwpFB~;h3LsC;*93#_=x3F<HyX{WVSoY)q0H|M~+^&%bH9#jI^^ET3eQjCEM+m
zrmAtXxj@dP9ji<RWlB4xoM)4WN)e1@w@-ZZXvIelu6X$2CEM$cIAn6RC}UBoMiI~|
zVNF8WK;2}VGc20S!VD6dV;ybN(A35DqG>SNmUfVkL?>0Ia}~;HhQZ^Lm^2Sz%ck41
zyXuIWJw6TSnP#?dEbKAE@dCfOq&>Swvz*h79r^f#{l${y;+Rm+8L^mpj-D;pZJ(mF
zrBa$OdTLjdhhMmsR#!w@lTpMB9dh3HC^qXY`%T9f6Sa1SZFq<otu4FlmZ9sYj6sp8
z>zY0VoND=>{!jjA{F}e@=Xm$s@9?efd`%2W{lF((IzHa&HyJ8pc;k&Xc>T>cet=U2
zH?aL}0<bPF@u865Kg*#Iaz1cCg{EStZCjL5hvIH(!%n<A;1IPZ(R1#Ze?L7vV;p)e
zuP(99vDt1}Z#E?%K^@*Flsre_n%w3Z=4ss|1o@SUr)1yn`S8OJ4(HB$&)qwBY1@{r
z`{Oeyj>(g<W*B>}t}prSd*5ZfTQT^avGD7L!Q+R>IQeL3neI~(HkY-QoHaR?4V;n*
z-DoM&jCQhdRPwc|s%V;;rY(r7Gew!NODp!KwWW-xmO}`HI$*Y13~8NkDsvFy^`J+*
z)}r=2oc+&sOc+Mak*?pe*{=TZpHT{D;FT9&<R?G>1@0Z6vt4h>d|UAITOf~~I8dM9
zcru!VLDVm|MN-qo3anB%B_bYl!V+oQIU$VX=rITiJBGk8^tAguA#B+hO-eKBe!(*D
z$R?5P$Bs&6lupc4g{4B_7*Px{(8rO+Ky!A)ORt@>nD4o~ga?l@Po50;90_qCbQ^R!
z#iIy1kqm?ss0~yzOJf>NmJQo&pzB6<+kyTfv0hd51N37ejK!-pLhQtE)khB9D^r1O
zGgT9*W}argV>VYbGebKw91S%mrwds#nwn-dqu=+0F-o*n&X5!eaZ^Z$_he3r6bt>J
zI9fy%NsNN*A)Oe-4LGBWfC*JyOBA88WuvP=OJSJ($k^?OV-HDuVt0LJ?4jE!hP6=j
z+ZiktiOPD6mh=h?;wNYouF_a%uv5KJiaBYMipWG8E4NLWGV2TKtn8tcoR!>>DwU$n
zNM5W-g-4nUS_i^d_{T96x*26N?FhHJkD#dA`9WBhb7bsypYX1~8PAE>$-)#+rKpoO
zY*sh6vRdCbtvhEK{YVU++BsZZG5C?ydQIE3*pi(G;3{_utx^XO1d0Q}P-#tNN*He;
z8&Z%2eTz(yQzCj#Hx3N_z;?T3y<Ra4gLu-79#>bKo}O}ke$K3!qdiz<*zLAlUOZ*H
zUNa26g!UdR1w0V3&f=U!X(c{U0aDU5t>V-)?Dtzj2ppZ9u-&W~d_a+ywJrP2hA@se
zTS@vu2^3e>K#Y=gYcxJY0*d_@d3>?r8{hee4?eua_mLbjl~UAJp>q)TbDdClTn(<u
z%$GH7Yq1t;6KT~9s}-)QY1;)&-C~SIqUy!@dJMvGv(93aSXc%>;-j>j#?iAMHmtWR
z!rn8}mTJBr9$Su8#fd#;w`fV5HF<u<$=N+NqvFy&MZNY#?tby}sL^xr^oqCMdWXGr
zL~Rb&OR1j4av_EzQKFFoipyb;T+9sHsCeVveKsl6Z3o67;gj(0FqV`QA<6@>S#O1-
z;RMP*xp$ZD!6mMm@jv{ff0Gwpe1T!#LxRO(#(v-ZU>wCK7z*)FoV7E)@P#jMbaG6L
z32pT=-Jr=Fdm16geUUQ<wC=O2?GzJTK_A{z7h%Nh)RjpLF`;<V@4MaZ(0*Mk7F=Cj
z;Y(JtwU(~iA3P(s+bt-?@$oU&t84b9_>$eP?CNg>BsW|0JP}zlcrV7qO=Zbi^YGyV
zNiVTA!{BL(2j5iKd=>=zA3ZYOM`Frswi~|lo$v7E$>Wk~oJ24&h{Z|piIi>xTEZLx
zl0MjzOkD**P!d={Csj%qrD-amJ+y60+qBeGjkB^?Y9;*5cGe!|OYydIgrdgv-iv5$
zTTrGF?KW+CwKmw2HuGZ>P{=CCX1(Ic<42$P=kDIQ!{^?5n>Sy3jVDhY5rd@Qgi!pw
z@(*@ZKhg0NY)))o@&RIzR0a97b%MB|GEL=}&0730ko-svfuO*rj1LJv^z8Oq+;WCi
z4x1N@eqGwDKl})smO>#bRqSdN<1i3Spl(}+#?YQEIJTKCDW0w#6MIh>JJNnf-g`{F
zKvR>vqU$0yh#i&H2Gd%stywHuPESVSNy}yfyM1K48`x||`X0il@WJ56vR=hZj)FUh
z0ezqwV6#(blQGs~tztG09379$7d1ym4fpPzvzVVSKa%cGj2+|Hk-Sis6iN7Za=R0Q
z$A?>}sgX{c(N>T$V-HC^uBvdAEp#c0ZIWE5lrzLYx7{)H15Q^Q&1-zr^utIdkV1qp
zQ7;X%Mlox_IFGAEWT=hBRWs_DY?!Rk(w)+-2r!?`(bn2xO@%UYtP+!`Z+N1LqJ%ZV
z4cAC9r$T{+Cj_&2>jp!L4nGcJEIoQc7=(H@nUYtQ$~lZJ;tiz_QG?Gjp1-g|)EqGj
zf-*6arov9<Zgt?Y#Tf9x3m;)7$cvbr?pV&}pp_&KXoXS2A<@=|M7#i=xBZc<l`1-V
z*%_1#vhd)%T-@Du%X+oq>hg-d>u8#WvpaWq@%{_NLO^4z#X50vA9g+aiya?5c)*j#
zPw4i06iG<Tg%~dm-qL1?A+q1?CZYj3%MQhA$825`W8mr2r`)}FuOLVfqapM?m9^|Q
z8=S5%&Jkh|K8IF>(yoa)vEFw)UI*TNaK#57tyu4&YK}@Q+lVnem5a=+qS66ZX=byE
z*~~JZJ6xp+VUN-YryQ+2qp2HgB@d7&xuPgql@i;RqT<X-F~&gO_Y8e6{G6B=$AR6j
zX7oL6RdeEw$PM%@R2V9wi5B+j4XN!oJARSti@=}w*Z&<p_wzr?@?=TctoX*S{ThBT
z$E(8A96Wv)$}T9EOxGffH<n$_hskr)hW7jpA8ZFw0;3E87-Ix$Wr9@_QRRnTp0Ak5
zN^`y1F!(?-vwZO0Bfk32zsA+&70daIv%6<}_~AoB{Dgi%6Aqs_!05}xk}rMfOE_y$
z_J`ycPF%{e8ChLlvsm^*whf<5xbaEGje<*rM^k!)f<lsZQ2u<oy*Z(X@fJ^Ru~^Wy
zEip#C_pDZ{XT%l$AozgocFS(JBf~+q-^?4~tCm*8WZZjD%$9Q)d}2^u&{d;&^ym>?
z*D+gE#1t{wFrUxm)PENIt4}b5XKUH-_B{OPBl@n#`%ogOl%N(N-BRju>v!SYO!<-r
zGvZsEtI46Ba=xswOtEqxeAA|GsGIuGB6rRmxRcge#xggioM7Olty>lfV`WjY)?%HL
z=j6w5?uwoZ?>(FKn(M2pPyDmZtmW+P9bSLq4ZiyI-)BE|j5(l_AZE8cNb}*r`@_ef
zWqtDnrV2yW+*Bw$BHoLkP$8y!l+vBb;*U0ro-_=2-;kUJ+tTg82go1W5Go{%z*xx2
zlmb9(x^j+W47P2_t|9vxU5(W9hPp9?m@q1$a)hL)bc@B|d(UpY!NeYCMkcknH4bYa
zS&PyRTP2!BWY#;{jfnh)Au$Xvj)~wCA!LGwkc^zua&D4`)C>9F=Y-b-`+1;V1lps(
z>Jm=R<}8*Ci-n_|8BAScYj?voj8mJIgJM?F8;WxUND*Z;+NvAG%t*^o_IvqI912e@
z5XX_O>#^2hD~nc^UFWejVsu1nK_IJIQPmn($um?{3R@YBbJ(gWbD}PObO!63@SzXS
zjKgSmDDG52^0b-66%$&=RPdmt-Yfz@O|5FO76ea~_^?SJ=q<qqlo`n+a+Iv;lp@JT
zswPW8XFeWQ_sPa{6?QpkQNvQux&uIXi&sK%AD<L0NDsR2dMaB{n~Kr*?E4*y<pNs^
zauq^^kkO?-PllufBA=2ZAd$j|4hbD0D9sQP!H?{Bdsd~rIt)E^U32g5U0!|bh^B2s
z1fXQWi$Qi5`>tcNT66K_f+rt7<@#zxU00mnImcQlPHQJSU8TVkBr+=)W575uB@|^%
zD>o(vYegI}29LIyuHR!7bo(9ire?F>(Zz_<5JE3`nn33Rn=bJ5YRkimH4oOB)jk6i
zBu(f>TAP^Fk!1@<bIs|Z!Kh4CJ5G)lR8C;#{eI0j_RQOs`Ro{HPN*tDJbVc;c1=T7
zIB5W*?-<6O=!KILgS2`4IF#Z}Y*?D6<+M3wZW_8YU{pdWgRUKmwngps#O)rxTyW==
zxB1-9{UY_*5h0D#FD!ZI-}oiI`}@C7?;}^&SJ3UrF`(U{pI=ZOd0K)gBypgxD|Fj1
zJbWUP0pdKj&(LIW8BmF~skjOqK4+X#n2=aquZYR8*<SO%`#=2qWFnjW7E~bQ;sE|h
zhMK1Q>A_HF?KnL><L!6e5phozaSXQ}_$i%23X{a6$B!_!Ba`uQ6y#QY`gM2u*+Y;#
z!W$H;lQKc(bWlv%LdHKx@)d*_$b%fflt|Ha9m6mj&^-ye9|$SZG!6Uxp3PPinv2DP
z(R;D$X-lwm%*Ez`Qq(<Ln5CFu+O%m+Opzx~9@BMuX0s*MI2N-RCr8J(=~AD0JMo#3
zDXLfKx;>AdJeIhwC@5ZzQjDgM53!pQ3UqVPbNS31uKOWrqWt+8bq!PTf5T_2Dp$l4
z_RvnBF0{TCw--y!^kj58(N&yvvJl!bk16$;pTEQF<;>_uHk%Ert3SLII}KM=ERK(O
z^|jZTFBh!#Yit-WN(n8a)N(h8te>*51F?Yt4qgN$5n_roOhrXp8r}zPZ%B!L^yI+{
zl4~r=HT0vx$4m?#LsZK3HC7Sx<W-n)rg+*Zjjk=&24x)DTB_PHn;CL~rcqcU*VdT|
zgT;3}yY(8IcGyZH>r~^YDk(yY(^zFtF5%ipJxWp2wwj?2^h0Fqb4eBwQ~KVQ=P^p_
zG!2k~CXb2erGwV*6T5A~Z6do>V7(bQI&L{WX*j)8(YB?Ka#|3|G!<pCu@OMjqOcg{
zAZxPD5>()(d+63(IXyl~fE(etl{F~_+PcBH8il4CQxPC3tW~%w)3%zr6n3U~#98Ms
zl|@-A>x*EErM*ABzgVnD0cuL3#1wX*Zan|EK_sWbP7Xbn0{*t!f+{pCRn)?%6p>;|
ztV5x$jDZwWi5Cq8t&@5ve+pjgbK0Fq8iXe?6eEzNgg9u_^eBKO<8sC&s1TFCe(c!o
z_vo2M#lY42lEu*xAtbD_xT=-)GN_^wk;tuNWk;NRmYlyalDbG7B3HYS-57ZM<O!?Q
znyRtfd*ME>zVS9Er^je(NGg(bB4}X&q)`M6!*0*|;)>PPHBTNtWf&CqUVMSq-+GOc
zvtzn`PuJ}rNxMuU#)|C&kd4APhjvEDluEJza{%H*x*SM8lY=Mbgti%vr;h_^?5R+w
zxF?Kz`rfnNB_3RkeE4+Wx-)E<kukXJnc2YHc#hl5*|FnjQ88aSDpwOirfF)9PEHud
zj=_)kq&PXdPgOe<mXJI|eNacL(w@l~j5Vy+D>l~`q$KY7ecu6-QDrJ9x?)isF`F-F
zCaa8KXt31H0<}1zvJQW7!TB4X=l8$=Ex!D({RNuS6Lv#SKW;fanloh0drz(~eb1AR
zKA=&76h^AH!h26lN_NkFWM&N8QQELkL-o=fx-b&gm*S%rJW;`3i$ZOlGW*=Kz8=`-
zgfgI`Vs0w&CBzf`itm0fEj%)jY__{k-tQ?bCs~4v%1T>aeBmYT-n%RO2(3!@XDaS)
zD!l2OzW@Gv+`TJBy-(pLzc$}~qg&A6CmfR}leo=`{f@N;RKhufwlc8}exR8(GRf+U
zkD>~nw{z;cKD2QsUfs47Z{s+YMR983i)D&)?w|-8Vxq2UcHJJ7DfTE+0DC%|2UR9-
zSIER;^yrvbZ#S&hYwq58f#^qet83b-qOwj*UsZ{l`!Tj;Q;dv0u-UG8^!TA{Y*HqM
z#L#<U$iz^J=?VB$Kq*{rj!c$~i61=q(Amo2tVJuuIF2YKorILCLz`QbjiW7z6t<wO
z(n=3F>k4-dv0KTKK2!OMrfnGd4j)E*s1EOyeq4xWhH>awtyWxIR-gRm6x2=4J74$$
zXJ_ZE*4OlX$EbC=){yh<XW<hb%#E=LpoD*@Yw-gm9t>zzlsPXYL8YaDPdH=91Y(9U
zW^(6AooDC*!8On>*arwfF@8+4IB^1%F@@|RGe%{N_|hp$t{RlCsceOE7T;)2PAv0z
zO|xj308Bx%zRe9;MNC21`^}ck<rO)vu`ZBjir8kBGfh?3m_}nNL!C7?X{yRFYaHG;
z_|Y>Cp1vE!{C0@+Lu5B9!k7s@<3|{*B`Q_06>&$5Arp6zFeJL|$mOcTHHvnoIX!PU
zJ)LoMJY%_>akglvYm0RXZ8KVjVrQq(NsHGamZTCv#~a1?bU&ti*aQg|4^+zdVPxn!
zoO2u<pP-bOq3664kx^}#wW1<EUN*F|OjSv^07j9NbR+WU8B3~;pxPE2i_MBQ(k;va
zBpFjg96BF1Tsezh<kY2@$Tmqyn_cYFL>wjmFBo&_SdcYFC$!V#>~OKBt{T!92wuE<
zbGQ*&<aF!YJjeJH$rJey{W2|8f>V`NtSVkLrTrkDOcg16LP~Ufk5P(YyJfMyW-*%+
zLZWFMDMW^DOHLKpW%8Jrf&fxTbfagz?|Hns=JI+)zv;-)b9{QrPyEC?yztU}TwSAW
zkx*yJp(H^XNBS`^b{+k$<Lb#%R+m@owp-d}#v5<G#!IigLc5rOf%)kI??=X=M<r>?
zit(~ONa2;y2=!gkH&jj%BQkkY8Hiew3=nVZREDl6r)$h?Mn8IXedcl#d3-(a;9_Ju
zRz%&RHCQuJDbL&mPUeai?;dgYY{6n)7d4HdZD!17EyLJzeSOJxyOM0cy22Po+sq-m
z!X3~Of8;&ZI;yJ1XoDX{wwpDQ4q#DJQI`_dx)PS`aJ2|8tX5+7n+;Xn($1IEZA(^~
zZnNU3-?O>8<Z8R(Oevb#jM<__r9d>6q%9X8J!E}(iEa(v52&2P<F9N6v{uwXI0`}m
zfU~>zSzWL3!$24YOu64;%tRt!FvWnFf?%Z|*ztVqL3v&R&n*;Aiyb;jYu<YMZR)xa
z8bVG?_SlEFIc)62Zs+^&exINI>4Z`ceq@L2`5$bhCVL5WK!2=t2amZ4shf(czVC0u
z6wb0A_R`9Aju3)aob2}i2d>-I>WUa6i^T%x99`Ev({h|>6TZCPZI0nHg@uBlA4xG#
zRiZw7@X<#+eDsL7-h2nGH6MKN0hdoNXzH5XekU!oA7fFMp_nocY&ILNF0X}?loZKl
z#$J5J#Jfsbaav6Y1UHM*B!rsOa1%$avX-`Ov0Bn^oD;i~x~`@5Iu&erx}NHQ#u;ld
zS`&gsQjx{Bp>FCj?+IOG97kMLQH>R&l|LI>{4+np!tMZ4W*El6YQ5(2>iMi`s=DUQ
zx8CHHmtW-a`YGG}R%~TrB1oU)IiG^bN=gbt=NzXecX;#VxB2k>54gN|Oz%7BV=+@w
zL}H1)lMRbXlW%Kgh%lx=>IQb(fq6BD;|06i68*^VF^|AQE7DA5y#&&mBrN`<3_)4y
zszLV_tu3{4RJFs-YMP^#`lMl;X{sZOHW|O#qSC-P42)wSdPVH6@%u<KYpItlYSA2u
zr^-S$nWnDEL0l-paU{i1(hUM#9~j0cK8$0e+ed~Wil8ZG!T{Z98AnC%nQS11$hd>h
zEB5Qi!}mLyYD3%994%YU&K4|=8WzVDM`so7%-{@^ow)S+lpwH^t(D4UPAFb*6B4a+
zCWW39MA)`GI;N^yhTb!b5p4{|XD5u~if%hHZxhGIma5KZ2iiJj%T|OK1*wZA9VTbt
zyY}Nij6$!_+K}1Gdgnw`*3KL1y1~^obzQoOg-56-g>05}Xd>F?+wFchY(xbBx>-Q9
z*0{O`XQ`s2t{V}6Dh_JvoPN0f>5m%EU)W_1=A;uJ)hKn~jTEqg63Z!*o3YfYfkAdc
z%2gPpSgo#@w@vx*j+{J$4+LK(pQP#g%x)9d>_@KFTP{{>HvNvQGOxXO$~$j;o>yLZ
znYymTBvFab!B|6Kr4zfJ&3esdvths8vD<9e@Ag<@xPSjXufFygv*i(0Q<GIZg!vlX
z&{R%11LXpxTZltC5OOm_g-UV;v@uw%AsM34jDrXyuoiN{XwCM?1!L>kCBx%w<fH4z
zs<(8pA?S(}6K&pb)WEz=EL%7^vK*f(suq%_kUK5&`GTr)gy5OY<}^)3&JiC*HtQYT
ze#DxZrkR(*P8`L}6mE>z(S$gXQovY^a~2_rYOEFX##xCtGHCJisT)*X<EmPOATfi^
zRLU@{E_wZx7m2<<M8U)uaLO=SobuAEZ}QDw`+Y*+lZ<1xTBB1!XPE%yoF?%((qe9H
zv|=%zb8&GYis+C~Npw{T2q}`3Y%@~!2mWO&|Mue<(%w``@y45PNSigmvo)P8Nm@_s
zInCA81rHv4NK6rXORo6Yt+x|8n54Z_7W16M9%EAIPTV=IHOu8v)RZY7(6>o-X^e2(
z7K;TBA3mh2g}XNtyQ1lPC)KYphSh2%3v#}hn0{D6k#feCQ0*8K&WU05db1(Kh*pMI
zUwI9*C}fQ`_!N1bq>xX02FG!fR`X_Eyevgwm<x~S_V{7f?qE-nZ+OBKE~6htbX}1P
zGTO|=Pq1=wZjGLdh9@`sX~QT5-htCo3P36Nrs8iZ+KXjWAz*86TJ@jxnVUA#!!R5+
z0iSdjV>mfG<NgaTa`fK!c{K5vc>aBtj8v{iLMz3*o$=0_pX1N}Yrn)l{0ILHU;V1*
zYBdNllSSRBQ)cLTVvuS*rb55T#i*8y>PGs0VDOP(DuSu#LtywJ_LC_kk`@||@Wi6b
z0ZH1kK5MGk62GkpR75i;)yH@Y#?FYYLD#~Ub~%vJ!11C5*AsH4SCKe+hO`&7Q3bS)
zRHnsfL1(ow(*9M3s3STiC_b`H9htS_)98Z~dtEou_akEnj3E%lOgF$VYWhJFV!{uS
zGC@-0ArbpTPr|E_?&*%z(>3+HqFL6Q-d}KjXU1|_vshHL6=*eC^4z+%6X&)hb*KXe
zH{$z&oW#1xxr&S?7USUAY{n2WN+p!?G;>Egw^TJ4C#eyY6=7yBe0XPsr+Tm%Q_~6}
zMN0^h5E4^j-*t4u9^)kHZ#HY0&F0v`d9}_KCx9EHVKGW9&E?X$nLew~a7;nr1m!b|
zTVA}HoXlZ^mdGgod^xf3yoFsUO{Z+ajHpqIjUZ6E)N10Dh(62~B)}-k;4>j-s>(^e
zVi>_khCwq9Bi+6u42f}@+4PZzyMf1h&$=J+{Xm-)ubj?#^Tj*7_}V$AcaO!9S|wr%
za>1l578UHaqrYBrb#cjdx5JMkC^$bo<LvB=J9qCgo6kfTq7%v(l9owIS&K@Vm?NPS
zc5%A#lte$NoDap`ROcz9ijSEQI9b7{Bo8+$MY}lS`stpJKDuD97d+Z#*4`3Tg|?bH
zC5#T7y1>0-$6{$|o5Xw$b>-1`$Q4Fav~`QImTnl>bsZ_gd~r<ER)va_==QycB~oPU
zI*c*QW^>x6MJYibjgb~z=shV_g_|S(MU`s+g|$^N`;}Z&j9hDsJ?y@;l0s|!=n?n7
z@)phg^TWhulr6{uTy3_juCA!bj9tg04?pCscfN=LDq$QGrUIl8YMZ9v=;(+K-g}p&
ztLZyIQ)e1>vDoiuq}7zbO_45#D63DmMErrzLY9Jf%z<`3<HZ+W1fJ!%JyWr%sim5D
z_xs=H$>YZaFRJE4QIJ3WCZBFhh@F~vLZ$c`{Yaf#fP)EfN{OSRBWbbb%+cbAoNj~^
zQ?kU=j*T(0+wJh)GoR1d@As@$D|Wlx0hPP$duOf1drwN~Amq3?PIch0`jA9vml8I^
z!-tRAZg*_fTb?|<;OBn!=g7b?3|w4YvRbeCf%NsyXiRs?I1Y5(j;`B@UtCN@Fu|eI
zcetz6jh)NX9u?%_hZVJ|a8?shCA3b5#bPo`<#7JY4G;3xIm?AckdXp?YSB&w;IvRo
z2rArKK=|wnI}mhXEF|~m8Cq**^BMQ<-)Ffz5&_aJJ|KJ=61ij%1JthK{Pdik`O44m
z;Qa@D@cw(O)=vZ}jh-9=IwgjFAfn6l%@T2i2nAyThQQDV1|J!*q^e~|!w(d8xgbI*
zX9?iUnG`}PSQHV(kQGO>6FNI5xQMQgsdLM?9|+ZgXy?>TD}{U*(P`v(*%GUcA!bq<
z(R&dVjAJHr0qZik(pa6b#$weJvuO$lBa{qDFG*A-HA;;7t?=;%7(5Ii;m1TfWO@(X
zh_q|>ir__P5&JCkqYw!x;`_*M=us<!X&l$Pp7mzQ>DipKvzD{din>w<(|A460!n^r
zj)_Cvm&Ic+j3cVdL0X%V!7KB5)ztLs9jJ(LS&V&6L>VQ3f+^2gp6V8`DkJGI8Z-x&
z4P7iwDk~ez5CeT22x-KZ#GM!e!3VKKn~FPYF-E#t;)I|JvUM{LNQ<@rU|POMrooBp
zsY=bHDKOok!2LAO>2no!(?gaDG%maajaB*pP-q2C4B9{?LXsQXX@jYpBr?PxZ9RqX
z8O9MGB7;w?HhXrPz|$v*2iH3ujU%fl1@TK~4ez{qpP#sQ%8T=s*l0#Su-PV>x-Ju-
zW*mDqn>Cl0m-MS0aVs__);aFnxx>-PF~`Rzw6j*o;e8KQp_MC2871=cB&xUQL&<^@
z5~o}TMUFRrpH^lefXgJL6;d3M!HeHqPGIMA-02w?mk+rbc3chyW#>4Zuyo9-k&{Mq
zQd#aCSG045t0T?KqLeA>xSHC^&a5xJ`2;lWoLOm&<>EYOtCrnv%f-_xVoKaSzr$=X
zFGzqQrHpanp*8j$<2W*#ElR7l#Ts{zej6=Ki<}dZih)4sMkM_KnP}2}j~{woc;P;g
z%)^I|xqp6!QJRz$kDgwzU9T`z=F#;N-uwRd_`=htSglL2m%<uLCX&j-Q8xmKZ`NCy
zg(eOG9iTN17rh92j39J5N=k~(6vJJ~IsQz>tr{_r2|2Ra@A!>>_z$p6&C9R7%JK2B
zWVBY5Jj`Xb67$u_yYIfs(<e^|-a}K6@E>}3pKke`#1hkS#t;c1&~-ht`QjPtl9bXx
z9HEqAyWP_5yF+mnV`RJCqLgA7hG&$tQ{iWfIV4teU3VzrVvL8j=EuXuVLC=F>hO@z
zwGcUCO8nM;``@CJ;rGA#4ZiWsZ}PK0`*Zx2zw%emTJz0sf1Cg4|NI}Z-R?g1Ir>o}
zBiCf#?bvQ5x+tW`I0R{f#Z<U-(}X<z+|99QE&ma8L%x>EId#rrD|g7&b=H(vu$xwP
zN(oau24_Wx5`xFtq99CBICzEBZOrXJcXd-F-YSr$;*YY(eAeyk6f;9Vu<xF)6|1y@
z%JAx|uX1*FCLGT{;yz4kWYg{?s;=L6+&#a;i!Z#$-Me@B;G^%+v@N-p!dN(hgLDYc
zvKW_^TFej<WDkBwV)mO1YIaPYHuV1E_h3G3{4^q%2;LKifU<^wrca8~(-#@9R;2b0
z$9G?(Li7H+@8X9*ut%IP&aqes9u+%Ik6VJD(?=`Co{>1QF-yE7`$USG%0((y6SIik
z#1GMcR)jPm$Rh60Q3^d*JK9-;_YI>T8GTO}NBWTP5yqfN$ujH{W1r}|#JE$0K@+1Q
zCh$oy1Q>lHta`3jBOiRYVR2M*cHZ*Fo9EoWJ7eA$no3h!G4Re&ve`-L&;is6qW8ow
zVx6IKhFky&0mUS0Z@=GziZpYvVA+o=`cMF-DXO)IxI&0R94pUmD$!_@Dfn!$<ox_D
z%h>`~SLntN3GDVe_WO?AcF)*fAFguBiP|-^vxcT^XiKV(D11#3V%*eDDvK$-F^;T~
zaK6hNDc=AnaoLzQ<ORHZuJN3OohtCC{F*9mgTj<usZktc{$lDYDz}^z8iUmeV{4o-
zSS60wF=;kqr1S7(n|SZ>mUpjU(|a^fDY(CE_{tkE@ufFk;NH?w>4--Poyj~EVCXuo
zt}l7|^eOx8mOLhw?VOiheTAc=BkEZrnUgs)rbr5&te|!^x^jZp9Z-%eMB9`|-lH<b
zhEaG1v6vStWD<2%g{`EpOZg@P&&Pz%nUFOhLo}B8i}zV<10TJ2A!woBFs~fv^UO&T
zIGY(7)6ljJ+DLm?=^Cvq;ILXk!IM(>G!kMW#7xsxpbUN4LMm0PU{YqWT*^hSDv3Uc
z0b{HzMoLjt4My+qA(9hRRdb`NGu1)HE%|nn!J=Y}qOvWqhNzgp2cA5B$jdQ+)|{N4
z;H)DJ!u44!7Wkp#YP06?hwt*_;xViB3hNx*IOX#pJEuZqk|KOy97YB=qLLJF))^|5
z7;<DRTaJh*7judzYYBPUjC@8T-~5(=r<WJ}2mj%J$bbG{{B_=X`)xk=xzF+X>#y_5
zD=+i%%P(=~?m6?tg88ha?|Z)U?Qik^`|shW2(ip(QlCr(-NElIrbBx)Mu}?*L9A@1
z*sr#24}3d-eYfZI<dhHstJMncJ<d6<udmr`Hiw@TWTrcyLz7x|5(yj>hUIlXj^^F|
zS}IC95kyE3`R{-2>vWyxo8R~reLwK%@ninsZ~Zsiy>pk7vonsCN9=aHKl(}g14hb`
z^=ifH`ih|+gkLuycAlb!ovQpPqdJzBk|52x{E9g<l~*AI_WM0KX6ll-b90{LJ^bjo
zbLW_*5~Jd#X-m6%EHQ`UL4-3szaVDsI!GH+>zfVt4f<y>K1hb8niSff-E)x$^kYxA
z|8xc8t8cu)3-@1OIiK^0Y0bI)aD4J(%}QgPhLY^^^{;)Mzxr2yg?Hb3m!}t3j6QOG
zy`d(crwB1JRk6SDgM0=FO391~mW3c6Vq!lGw98{g^N8*T&Y_wfMiRwBs2|wxHyBru
z$Y@uwZ%#S>;?GkrPno^)7BJ(r&6@S&hdh4wJIoqO=sME2Cl3R4EnaF?YbtH2<_>D;
zgC~YPgYQvU<Nd&39L7nz&<1H^+RBODh81OIikYYjQIrD0)GbvlK!zVZL)tKgJ!2GO
z;`)3>PU4BzZxg%S$T&v)nAnf7*=4$3llK~b9qG58-Q~!W57#{S;EH?q8t$Aq&QBfZ
zCo`JL;f$8rKtVzfXANSvmAvFi`#@qKW=+bFv|_vOPzKt0OVwCNfsi~gB|Hh8fKptW
z>3+ys43v|k;P?>fI?p!lc>M5!55D&y$0x@epB{01|CIUhg86*T(Q*m*4tEzZCWda{
z>gtO1wYY0!pq<T_FXpsuOWm|ob&WEUO^N^`)3c|wF5)1e$xL&AgcS(cBz+2tqUYq;
ziJ#ac9*|w1++_9^$4;e4RwW=DM~n%tLP7R~7v{7962lnjz2eEb=iznFN7o%6u0}4h
z!)e1y&hT1m`P|(jzIb-b{dt9|6TMNG?5MJM(xjB}eq?=h#d@<w#mxEXIY&n)QqYww
z%#8T+OhM+Za=5z2y1Mx10jWUK!$24Z2#J2bCk&%R{Ww86CK9r%D(cFL{mHFXOj2_3
zeGqP8NQsyg*=Vday!gspE;oDP(;X@+jvK@AT(hV%Rhx0M1-YIPBr2Jh9rLQCu3L<C
z_>x*6#B4*`&S>hE;1fQ?TPR7OZYoYs7m%SV*`HF(sY5<xUCq&YLkxRD$jsW=4aBVF
z98cT^5z!PMMr+8{5K&Tu1#lRmwoprw#BA2Un9$adLu5X0==vS~<zv>H731L9Y&V1?
z=C0a^^gJqQTjX+1*P9J7WcpoK5^ORiORU!rBYg}9#MG4gk^E`M*PnbOAg0J>x5Yep
zLJWZiAAU&NN@2cOEI2tiV!2%M$}2B3o6q>Y-~I1+^7yfAdh%z<trK)CLZK(5(ozkC
zVDP1=6Dy7CfXGcqoA;i&st*MmNb7Sdpe9uBdUcIbil%9Ro0N(PElN3Ijb$AD;cFj9
z>L%nb6$Ak#MzAsV(xORQzW@F2v+sL`exw_E)|(9<ee@A8zw+wg{{FFzl-cigY<D}x
zFp5Ud7_6lnr@VKbK2O>G9}-4Vu~bkRXJu0?oG00gh8S^Kyen;0kpLgOY!a0#v0Aqg
zKvA`$G@}nUD%*)3G!e9s7NW8C*1w5|nz4e?s31=NsaPL9|HJX@59I<%hLjk>_(Kq^
zPc}|ZPdPa`WjR}-ZXunr-lXSxP*O^4x4jrbXNC8PU;l@{%5Jx1y<HOt_%UORoaGT1
zREFenS>aTm%9)T90T>c21;az7h#z*eRn1^6gVNk~H_r&Nq^(1vidwm+@2_FBjA2Fu
z=FGW!=PuRB9kg9CjtRFoVVM+Z@C@tsak<CDo}4_Us?o7Q=Zr~!6-7D8l7gli#TPS1
zUpgv=l#M+5-r_=CL^u|!E9!cVHR?tzmy26%lD?dF8Jj9F2^B#RivOIYt}Jb%G1Z>2
z*uXR~Qn!))4t6fFNuc)SxgIj>rz5^gY_2p{Cx+Geo;#-v^O>P-G<7A`Q94(EVF(eW
z#AGdIMGBdiwLo!2)p~StPRe4yhA!(~ilV5_iR5FkT)9!Fr<6-vA^4aD#X}P#tghDV
z_j^V^(uIyY;hfogCW#nTg)ZR)%4M?EEDf~d5n&V)|8C#&^vOyDlEyM$ENEvfbyd+g
zM_q}R9Y&FiD3DXmq7Y7?l9st7DOA>|JaGU&`FK7by8|>RRi)AjjkIFz6!t6IRg}v3
zFrri?v0gD_bV8|wP2z7CQf9j!d9ogP@1o;-PkXL|!c$X`I5nCV=Z-f|Ti&?Sa#RHf
z11dK_RSsWz(mn*bea|p>=FOb;<ebG~K{acjlKfP0_m@y!tic(H-Z8cUT?v;W1mXVq
z0pE9ot^*&4!$2AXF-V?iLUzY>T~oJG?8O|5vPQ%o-7wH~(%KY~x`c3RqcExG&eHL@
zm(EGs4(}tiO0?~aqthd13x{ux3DvB8PD4_bUPD%iK8(~hQX9)S$b#C=W`(S&Y1-KV
z8L29RHpT3-sOuy=Sr%kne07|yF{UPjJ(Ux$AR*WXLJ+%(gcxsH>p+)QWVt825)rf0
z2Bj=byMU@9lF>1;dGH9U74_K(C&x$h!=CMK!)(@~jAFanvLE&Y-AGKAHRxzcPfL#d
z`g(=d4nGKqKBCbHO85&tCv<s$E#gz8p7WvjG{eV4+--?IFmydt<)pRIw$!yN3O!k<
zwHEQj>Uwo+otlpEk5;Hm1yoE4lhdJi^4=d5SKj+W3w7ez!L9t&1pD2d*=%;O$C;F{
z!!R7aHU_z`*X#AcsCX(ALqW3PhFkXWHtbCsP$~~qT3#uG)|sQ@BR==eJN(}7{d4h&
zR4@k5e!u5C-}#P+HU3Cv>qn2A6WiU6&34NeJRwdS+5(Urz@&nX9zJLJRk%V*t3&*u
zE!HL0T2l71K<Jy0^F!XO5>LSi^#qtKicm$ZnPTzQ6!jd18%+80^xCSV_-G@dCsj~H
zH5G-=4X&wMzdw5gLM{<p;nQ82#bVC!(GiQqQn=FPgiq-Za*{sTYsvjO9?)=4<XhkW
z`T^-o#6xsh*5Ju_z~q3<;Bv&G@EJVdGs2n9nH&RQ=wNn4aE@SpSnK(gQE!gKI56}Z
z5{AKhLNY9D#hIy5t_3V94N%p%*@DIC8P^ZLj&nUGjpR6DY=v<RHYchiQD)jmXc<w8
z{^&>Hr%}2_IiYkQ22air6D&qsvdy&h^59RIb0VfG>8Vf*5LXLwMn#BmC@CwlA(_n^
za)z-Mo0y2FsUvNjXq-eQq7r?t7`<W`GV3ui_K|My8GG?dJvlZUEevU9u~jNvKg%$V
zI2V!IUe@{y(&jW8tu0kkfton-!f7XoH(^SNGzPIzDUU{sL9R#6q@;*w;g4%ejw1Hl
zt~U6wXLUW|EQE2OozG|&3o6y1tq3f&Gqj7j_yC5?*pH;V!uu`5&=Y*memhDWTU{}0
zYUVSIt&QZxR)%c#1el_dLQNapj7*3W`;_Mx&sW$XJ?SKS7Bpa$WbRF>SW##xKsX~t
z>k>{cu(~cPkQ^gJAGy5R^U>9w4=)2NuNh5Cj+vKp<dvhE*YC}E@lM0?w8AUTZog+)
z&8U(U?o}b4i@9HBxmcE1o*HYcaIUm0b|L1%TM?)0$&KA;$LMnj``**-wuG@G4g;az
zlg1GPI9)*rs5X^j);3gaQwX-QOV-L1dx$;zf`E)(9L-~jMO|ueO3apDIBFU1o$&B-
z&FpxIIXYu&74KiH`0nMqY{q~O3Fm5N)r_OI;?A<>Y+<=`GUHBbsa#9#gxszb*qJ3|
ziAyos0)d>mf@*q{(zI<$idi_0#jx2}N98Isn&o1N)`s2>#4Lrl)fQzO%1S`7_XAzu
z(Z`6YYH$r0$@p6hp7YrPl$2MS^^VQ;HU0XUbM-p+@7|Ggf*3eGI|d^;iZLaWRZ^Ig
zkmN~Xnt}CtD~jC^goCF>V!E*|g&RJ7K1{5+q@!?nosZqvf8-%3m3YIkgKD&x-mtn}
z9n|4dTRfB5Zn|eXOgFFesShaUL-*(K8s3j|okF!O-j8G@{)<&r(f9og4@>KkAFB`T
z)fl6sj|{^>>1eHG97o1+IJA9JN>o)vN{O!TNI6ltO1O8rc-WO9?$CZHKX>cs4@GV+
zq~km1cbGDkkB^VYDKTFxSglt)e)5>7mzSUFJblvPW90Jkii^uD`cCK$KI9^vxUpuR
zCf%H#t&mdAvbj)VY2r)z2c^KRtBUG&GKkS)NuzI4Jz|On;mb6%5l&D@k<m{r*_(oJ
zQq<*~(5~2{9K;lbzcleAKlQ!;gZJrF%=#b}8lPs&XDz3vr<|Uip-L1JG2Xn}b3Pu=
zpu|K;-rxXa!ZdRvDM}-fn3S1dgDl)JWR2B2k&~FV58V!R_Y754P&r^Yyq5x*yCS8)
z3dA_F@3y1}=$?HSXx)rEou~1MN@?151}LI2sH*1j;t}&Z;9@VyMXshIiZCFHvALup
zSX?ERM&r;Cf|!KI9EHCY0WIMIJ`O0Yan;CTeu}bAIDpX;V-UfKEx~I#V{|R3Uv!Kk
z?7N<vC6m0a=A4|J5JF-cJRx<YBN&HBw+l40J<26EJ5YN?+-Cfk8M;ImpdTXr&~bid
zI6t+Voj7JQg>xB&r`zvI4QM4+J<b_W7PO>n7+a&XDJc*cZ3Q^d+!C;5$#In>T9I<e
zYt4zA6~;D_>|!iL#qse87f&v@x_HXf)nhImKW4w%GG8n>@}9-<F}AKs{HLhbv(gx?
zsa(Z;v0xlV`mX2d;))O6eZbR4Pf1bYNasg07K=HHS;KPK($r2eYK?4QvXM%O1j$PZ
z)ANq!D(n(4WQWfNtr8WPxem;9#ugYlgiKB~O1tuuD6z&+h3r^al8y8oTy8UuH=2)j
z6;FJmSDFxePPFEY*75SO;+2!c`7%<wgjNkoRSaW~Q3Fa@P!(Azl!f{60;3hKlCOD6
zi=dcKY81AT3pnvjWC9%sou}V-^!pwCe#=<W5ON9ab(MGwC2gtOhITQ>)Q)Tn%F5&!
zVjzs3{dUjw^%WsRoU_!n@YpaYrAZnFU15D-;S%@HTdq^bh12|#j~?@A-7!15&l^AS
zWq$F?U*^SEU*q`f9Iav4ZMk~*0Uv(%8+`A>cd0(Q;;oYt-oD#n{K#>u(0U{oi_Z$5
z9adR%(!_Bf`UqN!P$GN!ep}kCk+v51?kdB)YEZ61H5E!{Y=S0cg0@7Xxn5uJ^yx$T
zp{JQGXpKSXOxF#>c|-f=>zsb>OK4wHWrMK|ml^gs^4iIQIyfp*vEK*wOGCB`cAcO+
zx=7nq7C*Y$iJ~ipcu>N&O~u%c;_<CzcS=&^BoCN1pi?5|%n$>fPXSCk&$Cf1aAMBv
z$DaL&Ocbi{gviArXhL`7=WZ`MpUH9Dinp4Ovt$h^NBVwGqZOm?iLou-b(7p*Q&m|C
zw2;sSMlVI(dcCIa`w|T}6eKA^jMUChSxpF(P0Uz6Yb^A$8%5^SI-F3&oAzve=Hsgj
z=Vy0N+VJSnBVKvsWtytu7k=(%dF$=Bi8--aulT?J)xXL=`|4LQ5)?fFyPwg>1Rq#!
zH*EJivGehhO$?K0?MJ^to=Ew^laT}vbtu}Jra9Qe05nyDDeczzd`?|eC{#fnVlh$G
zhxTh#*9?7+_oCXnEl_!;7?jnG8#e<E9LLXmB&1#TDG@td#qrrGckbOSMe4*=gOA@Q
zpVg4O^L(>0x^>#7HW>(i++;+-c}PQ$4Q5?Tgr_0aGB*H-&Y)6a-0vZ4A<geS^=n^Z
zzBr|w9dR3#5&Fcyd|qSY2IYHnx%aF#=&HtQh+{Q1hsi*SHKsNM8;Qx2QXm8`=dJ|S
z`#6Z(yOJaXZf(%UVI%}~NL%p1<4ZbHh@Kb{P1^{Prv!Ok%$MW@L+x2_HIYPRJ*rlm
zR+dyL+D&A;&1`3y?FNQT#_tlL)5Jlu-6$?D6?f0z-dSW`N8!<j9aX)QM2F%TYOKS!
z8e=N7cCu+N_cl4nXVdED9+&owoL#Lksy^h8palgNA)%a~*|Nq}cR4#-vbjFt;lmHv
z4+GuB1?$a*<?#_s;m@_R8S~{FZK^}WC@G0Wjh(Svw7mGzT`r$qbMf?w4?cXvd*6D5
z)0VcXIG!&!J6$lJ)ijNvnQL4N+LRkTNsCtUl|K1+?!vC@yTyQ1MXbtb;-RX{1+B|j
zi5jdZPLiK!?TVNb+s^Z3J#x_{F8jznNww<8994!zt!Qk-4;!p@7_DgQIoi$W#+@jg
zh`0<*+hD9C<t&+QB`WDa+J<7zYDGqWYoYNz5WHvTc67Tv<9<&Z22>KMysax-<)~*3
z)-4Jl*HJY!%9?{3G{!{N_3SrWy6uh>ll0U}%ENS%rIZ*_AZl^uHYRa=vc&90zW(lG
z?5nTw@BJ_SSNv<g^grhQn{QGtm(r#&TIBL1avVrQ$NJF&o_^=s{Py4ZfAf#O{dK-@
z_lQ9&+6>AkNP$Yp=O6sQZofUmHhJ$Q9;vkA7K=G`Be#QdmdaU@%4jDNZWl9FIG<dv
zF8ScW`+W5HG1a2sWO;<wj;Jc`zx_7npZ_8sZTCbg0m^8wb;X-s`~u2EnP9riteMlc
z3#w*LHm)F>x$rgR-ciN06(taQ-wzDitz-)q1T-m$14~HE9Vt^;Fb3JFmKO4lWmWjt
zxcv@)kaymXl^6ERkd~XMVLSAECdU;5!#Ln7cZllq-m}|v^!<QFS}kLd_=gavDoGO1
zS_%oR6}#=0tmO4}`~9JanUGRhB-1nex*t%)er)K%+kNGgS6N?Qv)yia>BSfM`Jeq+
zzWh@^CGA(OxPSjX|MP$M-{ar^-~MmNc~f5RvpaJ+!=_~44#Oz?(wGiLrb_GLUUS2O
z`#6_yLdk$b0XHdMr>~2JgBCubAMxR)O+IB<9tuQ|jLV{g^}!z&WM4RrF+_5jHV;Wu
z*d;L|=i=mFROZjy-uc)W`oX`d>D=YR2IG0gbWYmYoTKAov=K_n!Fm73@YqatN=k`^
zlhW!?K&9#0l{a=;c)KY?f)Ava>AD{09CgzISI@k*(HctZAt^ET9qX%0nsyGl#z#rK
z$;qIV*d!l*)@Yix;?DUI`T8kg=nfHuf?P^_-&B;2n3CVCOo|qxBQaT04q_3K133o~
zAmk`YY0cq1wHDrIOu`c*#?ooX#Tv!Z8ks+G5&$O_F1Pn0!gJQ)Y)y(oK`bLplQ=$s
zs)Dvr%qvB$65FiUXCt;Lqo&(J7$d1bqw^-vR)NI=vE8z@bc1Y#t!hl^3~6nkJe%Z9
z4vM6+7#+*ae(QcOUBlwXrj#TGVahX`Qz4CKjMLPr!rD`mfz5Wqe!r*hdN!*S!+uXy
z)%f|G90IOsF|~*%4^$1Q6{#!B@$sCA`C!r&t5wHlJM#4Uigz#e%-V{1V>n(I&W<fj
zt+7tBZ(Z%4TfP2#g<bjZw_C6ViZ{#6_hrOKY-(N9wI%C>bpoIEW8!-2dAc3A9yPlR
zSy`M?EH#|Din@}>31uv<nMshcmR*Lm6=NKT(KC+1TU#y{g;@NoN0TWYGdBo<f+Pr4
z2(h$hg`1VFqMo<7#Y|xMvePwAcrlm;Th+L_lJ@u1!%vc^vE6PNx(=l^$0sKTV)@N`
zmqIc?trJzP$yt8A`g+ALzWOSE=HLCxeEHA*1<dh%v=hOH@Y*!l0y&~>i&l!WmtW)f
z-d*1KnXmBv@BJSC-GB8rdHm!<UOb-B#(^_quwhL~k@eUUQ>1Phl-BIaLcClqrDa}>
zP!mWL8;R&gLK+2jE=uCfdd;JcKH~fDKBU)))4Oxlred)?<#Rvvb9{KQ;XnRA{a@%v
z{I7rJPm?X|L&P~<NV*y_4Er8!EVgRU*5Rs()q2AeVWn&hSX9(37YmdU9%VlcjA6iJ
zm}OJW&rPfK_S~<Opg8Ks<xTvDj6c2t?uU;UBjZ@q$|=$`HRISb1~09bAV^UNk>g{F
zwGr<}tQNk_ByiALAC$Dlm_xDGwk?~T_}aw~2{GJi1y3UvPR<SA?&Ig>b}^K4#uZ<w
zi^~g|rs3q|gfD&Rr?|Yl<bw}B;PYSn0>d!y+H0@z6F>QrpnjXr8qLW$^ZxrEaCvz}
z*Y!6o{Zy{2);AFFAMkk&1zh>eAx1IbF6VSA;FJ>HTZrKxWC^7JoREkqXE>#dPD#vG
z3qm*f4NkwNIVlze^z^mEc`pmzr@uX?D(;6Dcf|?m6P;t=aQwl0x~^lrTGRJ~aG=V4
zcxc0ZhT~?w5Urw_JZ4L(K-QXAuH9tgWwIg%A>`+n7`h%~EmaZcO+!)WNlK%Y@V$m^
z&(-BqPEIq{NoMzEyOkt@k_mdajkU%$4fA=;?&(MzWy2*M#X@RQ8f8pLZBU2fPD)zF
zIvR3;s$!DFg~NSdxG}j_C8b7LT&{}0Y#{kWRa>m9ip5qD(Z$R-1_nlANQ@!iO01%@
zm1J(Kk*qRxm0_-MMpGMTwLyn0tZ}W_?-A#WP^eNlU~OcW1&-#4<?$TtYN7uaiz;4o
zx(ND|R;ZjOQMk-CxfFHe!Y94S!j{!yvLI0hde~uerj<O`Dx-|T)iuYbr?{%3scY7H
z&A#u1OC2J)r0TfYK-D%h^9JW8i>ql~(^zMi%^L2UAE8Gu+7PZfF0MPSpKfu+&^p8E
z-0;G&<7i=N=ZbovnCFHV)5mg*e)M=wj-3>C28~0E{){ee$rI0{z``jjauh$CgebI(
zbqqsfwf0=B2DU-5&6+N2B9S&_j*VlH6DkkXP0jh8yEM%aWKBYmg2%WDn?Ny=Vq)xj
z2IrVBWruqR=`F6(!Uf=^36B5(R~kt~K~$B3LWrzIe(uHo0R_#x#Z?xQ5=N52pzvF4
z$+1k%Vp{R2nfNa`XJU{RtOjYJ)V0I~S<5gCx7s5LlQWH0Vq)wP+0^8iI6G?au4NcK
z-8itcnq1I8l!y|t5&xYSJg#cV#xiP!SDL#&@zebBE1%={{=whpYrpmPxNjXUjm%Xd
z40~MNaC&@>(T3e_OI_E84A7JkyWN()AJNu<GmK%xIZz=LdAwt@-}Cr#!*||&%+pOE
z&kfyr;1_=GOB}uMI{*3K`MdnqKmI)wiZ|bU1KU&tA`dPuxO?{wi%O$oB8S9>-~JA-
zy!smd=(m3hPhz#&usA*uG_)2@M~qTfoEJV`Dhgwzr9c;=ZzL&A(L^0Q7DG(viSDT=
zo>c#G4Jk&1Ty2ako0))d>fl}1b$jZ%VZC0_b$i-&Cdy{=0Zpo^ig6sD5lD=qr|&y<
zyPYVkORLWqbIXwLS$mVo0_Ed{-HjcJ{LFH>KpVviFTB8VvEZkF`lqqhaB*?L_rCW%
zUU~gBTtN`O^5rjcz5X3|&Q`@AHe!tY`mg<czWwd*a(%r*2>hlLNjGiZAB8|3OoO?p
zNhZ@~S5*?&J{51b&P$4tO5sOwoVUi1s=A!_fe?#fv6&Rj=}`PhyIhb~U#?jQLMMTn
zj;36%^n=CHCw@qN$TyZsDFHtWto~u=DyKx(b?mk~E-o(k?zg_hKl|O^<M)2|cj<;+
zio54Nwog7H2OnaL5`hyrlawOpQY__+A=4NmovBjTg~7A$I*b*6%^&!@(o04uh{3a7
zuXym_17eKK7AGjXV7=MkON10a8WTYw>q9N6Ovw*qKcJ%s*mRx%0fk{QNs|)A8Y4W~
zl!yhXgG`7DNQ6+3UtPe85)+k+>06~6hA=RMfqv*PcEM~h6)RFCYNLTb&W0}bj3JQG
zIHRddgR+THd33Tw48}MrlV~)|oyDSPXOY=a;&KK1UeR|j4v9_gVGU6!s@CBar?Sb{
z4wP`mQAWCMT2$Z@GNvTp_7Ea4p>?KIIU=e+IBZ$6XcJi?*lq_$sC+=XbC4vQ;}gf?
zXvy*EDcjYW-DXSIb#(iVuJ2KeqiSoGM@O8UolACaF%kFPGlqdgpq*J>fAxgr*^HyB
zp76mHAAI8}+tmmWS__XBhI?lfC#RN^bImCl9}_j7d^{&&mrz+_Nm#%sA(zVYJnell
zsW@w^Z16D?vg{6gitM|{^=e?fjSQ(_NK(8gGL44w+HqQCjL94wFIg_nh*VUzLK`Qr
zZ&n!PFjf&5@g>U=pl+K(t5p@BHC2kdX=gaO;Da2OF%pVywZz%!gvgDv*hI)AYOqZM
zS~CQhESZYB@_LnZG;Jf2=yD8VjvEiV&V#FbK>)Dj7y*3X-S2#xyI*`jd+#NT6Q8<_
zoI|CM_>%JQ=EJOP36;gSHJ|@ie~HDtd;I3#_!aTG(gVxJvOGC1g>_HeG{V!+`mlI7
z>)331s;0p<jv;ikw#E#ik{m+j;cCY>Ke*;w4+lD>>Am4!|1&?!kS)LRpZsUM|KKrd
z7)T+Ihn}?EU{vPxWKQEWe%N8G<D+-q<G=bFe}jMH&;ME8{=(<@TYu|svg>*_yG{&S
zCz5xT-PhUK`QagvhZd62Vt%ca#J?Gb>O51VTTUg$Eq~@4{ePjx4c{uV-*;pbGiM7T
zC#c<izoltrhaA(Uu24YV_e@GykX?P>_tKK+JA#+*j|X&V+WD(n$lPS$dyDh(2aM_8
z=ZhJA-|>sT@bkRz!VA3g(hKCAdGO!?Po6%J`__A`bG-EOOYHj!2>c;0|6C)*$Z!7U
zZ}H*7r*wTM2_s&Vbkl+@$6UBMw_44&Kj(zBnOiN{DXUl2^`WpE#}PkxhH4bfk5+7U
zJC2klxWM2CS|gmEl*L0(BbyzC6n}lWR+Hf6z_FZ!9@D*$)2-|CW2lxA6J&B2F$Pwv
z$6Q}svtDmltyVn#=n)SdJmTu=imR(@9zTA>2k(Et<>dvd)rx)J|5$mkAfJAs80nlP
z4;U%<#>`<Jv070ng-QuhmR$2;VB8qy$CV&GA0Gv^QCgF7Vz=M$^yw3FiY$-z%#QE#
zO6AZ_igW;7Ybwe*8YAoL6}cD7EuS(P3SV7{6H|Z-IciZ)S|_PTIU9TwzLxS3q9^zy
zRwt$s*Mo8naxCqBC56n<kzu)9l5%1g#1_UF>2FRM)W+0+C&y9@>4eoHNKCpFu4Po@
z<VCQc0w~RFsi@{7ix6p+k;@gVH?Y}e;+V-6nxh3r_m6SM$0U^qNg_K{5^qyfI^@xV
z-2Z4n0moq^ra+1lajZJbOB8;qBrREKlj|~1m5!jX6<QgR4tP962!}bWSuAK5bIwZ(
zy6g5_TwT)j9b-Rm=`Zj;aC~%3+lobsHX4`3gEkXT^O2KfLv>Wq%xZF<`Tm177wduT
z2)}<Bd9NEdU3p&KSze6|pC9s%udoZno!ePUrNNO8a{G9T6drOWDUHqsQ5uUsTt9gF
zA+y;>#$ZS|as-3Is=%Uxh0Pqdn&Z=!<zm6r^^Ql6-e)#zdHwa5itBns7=pmWR%5)!
zPo|$a7a6oJ=-nu}dAhWayqFhiReYW(?I9)EqSiQ^k&xeE7>HS6ti#q70dCYrlbkwB
zM4EBJ->aJDU|n#C?nw!X8Z86Fh*64U9X<tej@T4&0j@rJz<b~P8h77#gGJSX5hD+)
zKoxRlR*9G;0>$M*$Tl!&MR1w(cYcaL^~)W<@t^%)QHzR~XALR_Da4JzSu1H8rHC4f
zEsGxmO(VjPZMWm(<Or(`Sy?{1-0;u7|A4Q*zvk&sv(u43`SV}l?i*j=SAXqq^XQ`o
z*fDUF1EbGeK6t?I{Knto{^#D|<m40`_YA|9Z+`WA{QBSgJ8U=C{LR1hUlUT`o8S5l
z*Xu2H+encxO^S%5{{H=YB_UxXDOs2XpGYAyGX_&vEb5Aj?ouYH(t<Tc+CqeXd5gQ0
zz$Za<>e@=|o>B~h6m7FvTM9Gru=0Ln9C|TnRqBxcT2<1H*%sc}guXRRgZCaE1EECs
zO#c?bjm6H#QK{QwYTr&~#W4onefK>so;>9*{)NB5a=B!Ez2c*fJ|dN<!1;U*3XYDB
zn9pax{=gUftcJ8Q(;;ahrlQ=<H-!2_Gw*;pszVEV`ucP~=OmGZ)7PgBpH>=Egk0jm
zb<@T@6ku79yzO>V1TTty><K|QbRm|7to%#}lIU~b*xkwpmK2>V-~VHQv47C?c(q#b
zE5GvBxW2lk>js9tXSdmMeZ6M4-Lv2CS+6%-Ts&pF+tGFXLFn<vH55`brGjXNa=uhq
z`z;*Ql{rE1iO9h0h?{ob;l&iWSb)Wp7`<oT^%7Gx_LzPTZi91WV?R}x69Hc-6b4(>
zL~Th(VuVupu$E|~Q<0SvCX->d%8^NHV3olXSBOzLV$2dzrYo$jQA!uyXcQX|BDVM}
zY9prRVa%imDGK_WxJfsuo2DrJ<#Q^N2xJ0=sAdG^7_ukmNUk%gR%kzB{0=)NW(fwb
z2|f~i#OTENY0KGJ%kfEz(n%aYQYME)RuMINYzCVZ&Y6-4E|x#qKvfw+kohNuNc2hg
zk~z!Ya}+lW=L#Rc68kQzuv)qlqGz|DG+xVdl%aGb6gZ2uhI#GKbw%HI3`5T_jO38%
z_5(R-z>4BkYw<Hg5sh4*riSx{<;{C7%8%3rK3Ye%+rYZdj42V4XAB#Dz6+c@$G!Y~
z2F0o<T*<WBU{yrtOt{sf&gF(q8B$UVK9Z=2DHT*E(T|xF4O&^M2zByUrJ31GZ34B^
zEao-dCoZo$K6<#q)mxS)GhTc31t4IJA{TC$Ep`D~2^UC3g`V=?5T|EfcCUxYO=*a}
z5M(oc+@W$}w8BR1x}Jn$d3r)y*F*(=42NQF;<1Dn(bh;VpR-cbeQY)oQmWi!09qMT
zR3sGy3RdEoG*1bxp7P#Tf0q~EdV_b4&M|chO1@{>Y3q{rms3LLgv)?7Bx?!DG3JW9
zU-%M#>X-i={=xt2Kc!wAVZ5iwnZ{Km@Oh-N4ss@VPh~7_z9j2}Yis<lV~if98+<f;
z`-3O^?)M+^(MHp$hQ-MVf8x*lIsVRX{svDrJG8B6667(WV#cpG{Fndn|AUVnKjm|O
z;>(<#9~1q+U;C^7Pww7*k^j+O__uj<dBH#V-QOj|Oh0<+*^Hqd1O^YDn4qd0FTM1#
zD3}>IIXPl`t>{);K%pwb>78?&wdBW_QfQ}yEuL4#7);Kn_%B-{_GSpzP#w@Et))2_
zW4IOnrAxaup-@!L71gYi#Mjr?hb-ACWx^Q4Fiup18+3YV*-nTZ{BUmFG;*Gp5<+`u
z+XnAPPL7wn^2$r>_FFD4p0eJo`RZ4{$``-%lYIHhUuHI+mE2<?b;su<82@1>t!bK)
zXrl-rKU*^1^fbBsz(1?@z4_M(u@`#VP0@CUWV>ZQa?`fGd4Yp!aU9WF3c=BjH+;Gv
zxwAP+fO!Z3U(mWZ{rf>R%ON5Wq=}b5Gmhd9-$L!STYml5ewE!;iqRM&(Fgi|#E%|7
zc=};ryWP?cMS1;M9L^syZey0JaOtM6rJRaojnrE@aU(d$iX1az3WrU}$6ptv3NJn4
zV_+yb<-)i0;wVs8C^B&zNrOjK4pX^;;2Er|iIW%+0Z|kgld}|ZdP0p$J2hnz@!;78
z9K<=+I8st5WzkxU=W`bRZJPLuWixGzl^8N(&?*xHq<GkYmD*Fx`EPPq6Q0PB5<`SB
z7&@=m_a*n6Sc=#rt{JYOY9n=*Y5GhUBDGbVpVXWjRm^7wqa!()13yq&z~Iuv-E}y7
zD@R$2@Ifw=CMQXVQd#cz=mRlFS^J|!MX`FxX)=7)!sROnp3#ty4O5!BK&T=t(OBb{
zSB|==1t8f8TApLZj{#o=v7<ptQ4UfxX7QkGY~s#|rSEED&h%;Ix=(Z=v+pv=XHx9x
zhaaPOVw!;DlqJc8Ejz%Q9jqc_4v|`2Hxi=5YT-xqJ~DWP$~7(->LfRovzpdKtPQxj
zqHY}9?VhJk_Y7mEU06Eb)B8QOaVP|WPAN((QqwSck3k-$=mP<<yigOG5aU7p1Cj$f
zjw8`~(%2D3K__AY;~Z!AU*PEY6pWQ!Nt_gCxezvseTo%W7o|u_p%14*CIBJQ$}WEj
z6_Hl2>;kk(Gzw0%=9R^a_a3~@yZ`K;aOZ_rx%0~F<Q!RVS6p9TaeaMF+b+0&|0QON
zBPpbFLR+vdW0Pi#hU8|v@l!v`gZCftoqzU^xqGi=D4ti*duDYjNXsx1W1^{QQh?pC
z#?+3OE&HLTb`86jdH2bh57vRps2FtO&;I#e<~tV;>Dr1rZ@<L?aQluAe*f$A`!z>S
zp*CxN^;iBHf9pT_Pr3KP9sc6K{TKP-YcKN;{?R|-&;Rl-^Y%MmV75Hs{SV(~yWg?j
z_iXna<2Z<KQ9)U!1zao5z5Dm*b{!e;Sv-GE@0{a@;bt;TiKL1TycR@cnil_M9a0#G
z<;Hw5qpE6B%q7Iy5o4sT8y3qsP1TT7F5Ec|1zpoL2jRqQHb;}$Kk39b(ny=O>pI~<
z-dgB>R8e<(qy*0Cax67x=jY63Grsq|?@?LDFaE+W^2*Dv@U^dfjaOfNm44`XdU;Wv
z=g)R1;Oz8_$4@tO{XjoVnb3#-aO(>{fUrFyo?t@wZq9?&8f~=9S;pWh`3y=j8O!;&
z@uR!lKJ}$&Lo0F;6k_xvV;H5SThz>B2nXf+fr65YYFXt26*8;O>bdxXaq+I}_?>_J
zJ6KaOA$qA4p*hN*M<3{iUJB7qaKHSa<43&2RAfuhEx!si`|$Nx&^h7gW^^k4aY~nB
z3P$hCdHli8J{4jyMSS#(VZ>TXU5j_2_nxEU1v&(V>nr-J3(WG2`JFSddr}7LTDokA
zNs-Zn97!>tQoy7@Wi`&!7%e%lF@$0@bd%^V{-Z`RObh4Dl>G180v@d`+E6xrf=U`|
zaJGVM#LPE|GA0-PxyeWdYYc>(@Hw#Yo_(L$_i(+5Y`Z}32Zqp-2`pN}ajRL@OI)q6
z^+3}oRLnS?dGXGi<0HqcN*D`aoP-~uepSlj+=z3Am>TBo3~P&!g2}(mN+zp@ek=Zb
zIVM!fs5BDdKpqv@ID+>W=OyaY1#&Hgtv2EX>vV-OB4~;!%6&mOPg<e0p;oPA#@2K8
z`;KvxXD#?ZWeSP2C|gXd$YCIiBctzVmv9~&$z}qXzDqpa1h!*hGsyWJo-@hixe7b$
z94diBVHMI&OfqR5>Pgv>%1xM{9|PV;G?8)iVg#HmMj31}R4GZZWi7RdxLQ%Srts#(
zomy*0J6k}`^g{<OQmL98p>ksKIraTflu2yF`+$lbJMCbU!b;=?xxnMye$TM)NQFnF
zwBhLFl)Ep!$kClUBxmV@KcE2yi2%vPCr8<$OezW^l~r+zE+<1toQzTmYs8N&=L}I1
zb0Wk*jvgJLLUCjicWcYj?|+Nm{*7Pf^u<@`x-Hk&mt3u`*>@e*G`#TQD}3RLKf$YS
zzlE(Ne$AMK)q=jnWWg@^%Fq5e{?_mQQ=VS0IB6^Kc(WF*72Po42SM&a2ohmpB;Lq-
zgVGk1VLy1DtU9he#71-W!a4Wfy3cCt`Ql4I&Dqg0OJxWb*S!D!W4PWjcbaaugMGwV
z%hmUG{NMhM{{gYG40X*nzxj2(a(sv5(^J0km7nG}f9p33H!d@JPwhm>9CN~!YpY63
zQq@@IM=j&JBc?=E*NnTZD8|&p^Mo;i%Nfk1PJFIs<Nq&4&J^@q6lYVOt3-;`G<Az9
z?BXeAY_VKW*@~m1qnknvc=F^4IcJIJt14F4SLB?<h&u*p3#NoH4<@DBCkvXN)^TeL
z!!WSfY-v7UvDt3;wZHp!$uY5Lm(1pp!FYUp%-{aizX04oBfmpCn{nq}d3Ha`^9Ed9
zU(?hL%f*uYz9akq>icX__i@zb*{>DuRtmxo%HXu6vWCi9oYAD{Nq!_+OK^s<-?LRn
zQQ$}B^O>}D3ZKy^i6!$s9t@<rzGLtMK8_4yPd{{cKj4$(%LYHl;va4$C;u@#7a=At
zF4r)%@d_Vo0tF`I@iw~o;~vv{gbc(iDOJcCtPpWlo^lh7B3oUoy$aQ%vgA(k!qdmp
zBJv3hCESk`q3;ukmLZ7it2UYQvs0RRBj;zk<GejB&w?Op)yXUDnlpB*D^$P3;xQ_a
zV}~Dhq&U!2GiJ?#`Qn(WZqeBhNrXHVmr7cfSXyVPoxxbhMg#)lieRaXrLJU?mt*mr
zRF<R+2}@R*YzCZ}(l?;*A*hziA@F29@WFaZuPm-TrkS5JzyAVl)ew{LKUbF*eDvYF
zeDl4h4C^Z@HE@1Zv(%R5QN_Kv;?9Z1H5xQTWrQ<W=mUpq7sYIP><5w$64}*ATpCKF
zR03tN%F@gYV?T)3V;qSYd>CczMdAAnZ7Phda7{~UeTffqR81vGOO-|^bE7gZ8?6{p
zu}_grnK6!;vsAt)+>>MltFoaY2XgkMO)SS)HGz|Jg=-`w%6Ls<43D;ohh3!4H@>Hz
zbUasK*S`1y{o_Zt?5VUAF<vP;Z5Z@K(3QHz17p^7QQ>n1V^8p!VF-leKpB!s)F#pB
z%*-hoBDK?0vzj5n#cHJMG<BOXW)CV6(~^)au5zWKmMmJW4Q<_$QXq{UKL%3QG5Q{@
zGnIA1<;zN(^ShofdcrVDC9e##qb18bXB^+VOMSFpREkg>s$&S~37J%2j6&IrHd*@q
zh3BBNBV_|t<B0?kIe|?Hm!J-c6e9c7(+^uV!=8N@8GU33nO<qUYuK(H@_XO@4zWF^
zT^v!hb1}8SVD!l9@@w>)E&JVux4-Z?>}*+-wu(Aie2OHOIC<%1UjE51^G|={zvSn?
z_y$diSe=NyN26%x$5gI{oNzf{v!x3oMvc_EqEZPz^z6owEsCfzZ@+NP`MrC*c5=em
zox3<=sI%hfd+#$CM;9|DYZ_y*(M$YMh-3r1zGrVczW2TF@#R1LXVFn}c5=$g_wF<L
zffz=#Q6wbtOgl#^HM>><<Im<Ti&?{i-G^vx*lf1!hoRKbrKm%6l0Ib&vG6dS@8S3t
zeH_|M1&Kh5m}$G+6H{hBmx8aZ8``!-8<=`HwyJP-jgOJRd)Di<Y(!Eb<b(==>+P0(
zKNQ-+P0Gi!t;kPy(x>!=qobq3r%WssOSUB&_Q8XPeD3YHdFOMV=kNc${}1nf@FB(;
z?%un{(Xj}CNT1zB&N)kJh?&teElu69+jqB{uAlgI<hyS*{gOoNh2V+N5u?YClHZvN
zk2a+wWbPv18b(lBm3&`$AcD^L(Gw<nij)t62@#0N0u)n}qOK$={P-BBKZs=BIuUqZ
z&ViyJ=+fFd6mx&foVw{KC-Fq4sGSSXE~lG+OJvCrSDLKKdG32T&oMqzbjXmLPhr7H
zfjEv4ZEIu;Us2hbs%p@#re4gr|5i<1&oGr`!VXQdz%1`jhXHqV#P0HahH*uVnwT=X
z%?6YAEM`6DIdgt|R=OJkG-Lz`(>bG(p)v-oC9csqOV<hiHWwk99;QuJDO^&<mCWIc
zA|SD3IcI{;tULH<<r(x0ef|Pp`tqC1@4P@gKgG?Cam_Jku?I^r;roH@)f3iFA9L~O
zL%#Ek-{GV0e3dZ_=sL4JUQ*WvZ8cdNlGYeyB;S}D+Q39j2tLqvgVY@{QO{b5(9{N{
z4CqugS4L_DQ{%G5dq)VPaLqx22nv@~A?Xw;*I-gfg^*gqX^V+T>WqU4xME5GlVleG
zy;RiFF@)lQ8HiydWlv5Lht@W8h7fQzV_m?SjztScGsP|_+^85o#dGmoh28z1c!!66
zOTM}g33t+jtQoT+99&kCs8~?=tmr-XXdr3As0dz@v#=JlhRQ-^GINt?O=eaL(Qw~I
z`oRDWYkD+2F{H9{cIc{N*|ZQ9qaRDZTUy1Hwe%rUX@hnZAqP?LX~TZpv)^p#H#>|{
zIBlt`n)dXJ<9l~l+&x3hYes}WF*)p$<iXBLCfIB>+7^LEQGlr|h*vJvxmJo4BbBu%
z5-}u(F|ZqY9&XmW_xK5qRvWH5&t~-WqsLlBTU(CKPI=@rX>rQg3$JqQ=G?va0%vD;
zS<L2auCLgw*Sz=bZ&N$XtFOO_t(s!QD-e203asUwKlwBK*5CRsc=2Mx=g(?v8mMhW
z)6CJ;T&{%z9rxr>k&+h0Q_7??vfHldhLJuCui*UtJ>Gun9ki=BJ~<`%h)Icb=N>2b
z?(^^)-^M3R$cB(amY<Yj43T66&i}^OzR9-R)3}QF-hG$1Uw@q!UU~sVBIgKNl$**J
zg!~{9{h{k5;940%iAY<mH*`Z^Nci%g#WJNPgj=CL(^P_gvBtFKs4Qs(;!7p1o)9B5
zYpCi<)Z9k=)Gn?rpW)RRW9Y|`Zs^%;MAf+4?POP)Z+zCCH6w;kTxgXwzpiI&H!GCV
z93P($LKL9{nPD6mgJ&E^`eEST`g8vl$HymB^;6(+DR@3R)55On#Qa<dy1DQA^0|KS
zGu>9je*9|ysAs=^+hZ<-Krlfpl?tyb-zo(URwgB(LaB*E6o_FmS(n!ty`+PT;~+w>
zLW&RGAF{Il`|wvhv^F`gE`Q+P|N9)%`|C2tlM3?BEzTYJkjEQiBn(n4|8P-|I&h(5
z692)`2ULmw(#GOUg_y7#T+`y3_HZoIt0iT*-jD3IBds#%>V!}>SVP{k83L*A8ODfF
zhGjd)H8sj;%+x$8RjSq#m$R7jT8mOQdD1aPJYvUlP*BGaolVJkSA-lGV#Y@pJZxh{
zoX=^F?lHgj8ZUj}PjGzqRjSz$rka7SK$SwZ6i%t%vs+(t`Q#(o<A$+cldnI-svWb%
zOwciF$kq__ltnMmT@zY89iK)^tJ`~7n~SxR(nVCEC8kf67Id_poD|ksLU4pZvTsF4
zR62gi6N0p_brb=GQTpJCr^!U6Q7WQT*(~L0(^4EADBR|(NFf~(Ktk}KB0B3*vlx6d
zh2xx2I#N}Nw$ijt)0iY1q8}$>ckj*DV0X>sH%Io{Ef%SHND&p%%}tq8X#vSDWw#%h
z>rCH|<fyTw2viD`$%4|EL|bWQwS}1IyO9_`VM>%vB7~9a>q|0Qd^~0TVo99viNOzy
zqohS_)+;uf6^r?t%2n8;S?^YChfyHeVW4ppvzd6#Esl>_-nm0_e1w^|j4B<1?Q;?s
zRcQrPaVggpR83Zngq9mBO?&N;U`BKVhhb?dMxW`T;<6w4?!%Ax*85NR_J>zIzV6t2
z&6o{ilnK|Wz@0YnQVy(jq^={&#Sy>wi~j?D{$KsG%$Fx<rPy3N;lp>o$9wO8pZDH-
zkGh@l;!Ce$Ohwe<jt|1+dG*b=xcAyy{PS;rgO`5#4d!(vu`EWSreX@k+s{h@oRj3S
z=EM*(7ppD3_vNZ-?%jKl<<SWb9zCHOB3fxGE$#QWzwkxA_S?V1Di~DOqKF1U%#3BB
zch2zscfL#Nde)mQU;XMo=NEqNUtxK);PmWNRD#Ngx;hIXoSbpa6$B!2d~%G|lK8P&
zt=Pt~w7H+XXQv(cXGSCX7i>%sURrCcHB_#qu50SLIr!6EU0oAnqVET;uCBOvdWA8L
zJ9qBTwk`Ypo(B&eu-$GM$C34V&DGV_AtLN{vG8e&hRHI;7|nXU;q2^;#bR-250l~Y
zQhe+}jC}XI-{r6W^}o(v{>%SM{?cFiOTh2&+2TE7H!<`BXhmJuf8eYCFu}yfUuzml
z0mUaPM49ZPF$5cE_3R)lm=wvkb6zJ~l&L(*g~ac@6ubNVj@@p}zLS=0KXejSAH=g$
z(q?kuD}9Ei>N&^%FFuMzh|1(o%dNZVrm!n0o|06Veu%^jG2cqZC_`(7a~7px$O9=Q
z`k`kk<czVHs>1g}Q77kvT4s7J1$3k5cYgQxc<{Y%vS=E9;pe|1DHr>ieYazuD~8=b
zzuS`f$Wc4To}W-@Deed(DF_;+ay%p@iE*-QSSF9l7y}^&tQ86l8HTaP<ydmZ4KZo1
zgJLr(w!vZ-_jvP9{vv0uzQyw1OVFH<sX;m5W87CIg_EAmS<dDx?%w5<myfu!3_Shz
z?=n9A9<zm}atZA;I8z)m4A!_J0Fi`^+uYUtZbyicyrPtn;<b{vGerq0a2iq;<a$&X
zXQ+}R)-{yWj&!Trh>`sFR8lr%bDI%SyDA($BV7(Msl>NU|0?;hl%R&dI1G&ANXkhl
zSxUsCQcQ}|hcPuu@6jeuTgBXIj;&!FpU<2ap0lv4&W?HU3!f(q9gn~HZQ_{73QWql
zIPH`(d4iv_Bu`wgI^4n$6KE8!RPS0#Td>lBS_c}Jv0CD9{18yNMk^ywtd6LXce(G@
zpgs4_@8a4D;~X(~{5X)hz}3|i^S0sW<Or)Y-DZc=jzFMkT8@rSI5|0HUKWn!@iDn+
zpfUuV3+cW9zeQPVlp?qwX1!FXTtlu3S0s_NM7Bf~C8HQc_L{*f9<TO%?ZHF->H8n>
zjYm)U;A&)ukTIZJvaUsW69;tO(PYiVex$j7!kxSK*le~uc<_Lj1UdQsx4+Hzzx_?_
z-#OzafAS~!{cnDa2M-=_bbQWiA#&{)#G=9gW;W;LFMf&t`nP|Rj<-n0(hVavg00Fm
zOObILN<2-(DiOhKx}GPi9sMY&hcPCWizQlFX7eS=I;zUyP&CV9?!534eKuT=(5QsU
z2}{B#utu><89OG%>ka#-Px#xv{`Z*A=e+s$TU@QKgan)mze5|ynH1%L(MEHAcFxnQ
zE0)V6v~{Sg*zWfHf7^Sr7)`VDzVCOIcl);5ySjRvp6=O)!y!d#4IRt!0y~y$#=Z&=
zBNs_*<R-cckm~>eaupy5kOaPn5g<U2ScZ*25fxjC94ZniE)qH9%=Ap}Ywh)I?{>Ca
zocF8lA&1hm*pw*h`wt2=-PKj~E$@5IbDrnFG`Ob*OsBd+JY9^^kK=CgXMm3uUof3c
zY1*1J%_urCL{(SZx^+9c{ky236hhMP_fPYbn{Ht`^<e8xrnVDG{PBg&Cqw*sv9Rg&
zidanfSa7z+lA35~M2NhV`FxHLqU-3v_L;N5!{_j-Mf-|Rf9BI1pEQ(JiJDH(htZ(;
z<DTWyb|;IE1-DRiUYbwapmcj2t&Iv|gdk07v{pEmbqd&qBuQcn;zAU(;Gq#WqPmI^
zZ>?>q+mfb@iqa4|)&6OWnd~;hKc^U$zW@+`^<D86K02m#-RCZK_4Gi$7$1M!H6d6i
z#S;*cwzW*BB}yd>2CM9xy+}2g;i@_G`jAGavZ~`hCNiKjcxO2{JmmfT18&~G#oM2H
zi3?}f*f{?xj~^d$GM_`$GCnTJvm=7gtgH%p{T$&<Ea-^&&0Y&sq9a|UHDqb;RJh}V
z$C~bdr4>pmY@6ZhDK^v?uQ+K69##o@bdJq)FLUj!pCen{LKHoW&Pd1+q<oTbe+YrH
z6Ymwh9>dK|(xfKU4%JP=Jdu~aNNW*FqD71alTt?EQKGv<r|zI+EKbBk3aJ#XEB1u$
z#35LugWy#+af`gX(DCO}Awb0fT!fgk;hn`vkJcWk)Tu~g@lFdQO2$S>d~!f1o~Wuh
z)>uRkXqk|x3=w1;ALCFu>$<xsn;V+>c#kx3^ps~!OD2LJZDB_*XZ7+$T-z|8RZJf2
zBArDNow>W`NDvN(@8-^$`P8sp<OrEWVrqy=P$dJY@@N_g?Z}gW647oUc%svFClyvw
zktDHYDlm8#Nvi>mN)<YZa^l1N2|;-JqhTc4qZrRiv{Y<vZBq1l6oY=`Q)rFU3YTg^
zsz57(6p=&lR1ZFOJX8!04l(>#hDhjCGT^ZY(cncJz3Oa6V_{l(-o1IBufKbf>yP(&
zIJeYF;DqYpS)g=NoSdRkg{fn<YZ$=Hw`6rq+Zguu9<a9kkft@vW^>+o=N%s0xy5|4
z&8x4y#As!O<MEV}lM}L{j}+0j&%1zg7KGsPYj1E^wH(eWw$eT;i6+%4N{Kk)YHA1)
ztvgR4Vhqjxq~u^+(Rxra7FR{y2L$KOUj&`Rv2%#(<F|kM^CbNt^Vt!J3}jdeDM_=G
z!TJ!Z6(W^{EaSKS<-g$e<H!7+zxOK)Rz~a}9U)PqLSvol#>p17;BR`*%4k5-lpLR&
z5Ja@0a3XY)^^Tb$#YeI#J2uZ>TAW&9xLCXcU7|$Hr;MI+CK`WsUvthfo6SCAZu@ir
z_i>}w9~Mffu9(r}dF&{j97iQM#NVH$DLT;v5A*q)-}<fJ;+KEr3%qa@zWBv2@?ZVe
z|NV1$EJ6s*pFhWReh)J>T}%>j)WQ$$Og?Q;d@7s>#7;%G84=>4S1e-5x;4fcL)#jf
zrs?vV2QiAOscGBxRQ<acMOp7@+Zy9cB!hPe6W+Uyk`PY2dDhr&WcdAE?Z50;C}ANM
z+@Z5W6e4mQg%FX?tz(jdXW`W<u-@}=qS!?-65s5254I*oI_{jmzzdhJaP`VHZru0)
zoo1ZBQnGX95|EOq1+~XfO;2dc39gzGnv!a4xpm`1s@VqT&uy`~^&)!@T4v>#iF2g)
zDoEy7V$zeM;|;o~%oVyDO$v!=9f^)>r<8Gzl~Q$FF$h8+Q;iELRcn~Fj{CEgZ2KkF
z&%Mf-3$G)xExa6{WR6lPpu2nIPg-#;&XpYOJ!Eoph^%We84EVib*QE2xPrtrU8+v`
zD!PV?iiKH{CY_~=qCs?He;<=5gy<~T98yL>mr9bDTH!20$R{di5jo_M!n5$Nh&*IE
zuBx}OiyjM2g+$<ji}az6MilFZZoO@qSkw_cWR=p0&<Tf#D&G(!O&gnFO6BwhIlXy9
z+XR|6q1U)>5B%ge?z1INBpR7!3@%(?ds<>$U~=~nD%kFcEf$<^T%aT}WEf*9D@QNM
zX+ue(Vh1Nxo>V)uv?R)sCJHH^m}tgzG}^1+JxVI_Ow*c$?7xj(Od(L0C9}zdex9?n
zwo0ls!C8*RCyXakE?&BX&N7?`G}cGHoU`=v45MO6EG4)Q&4+!=t3vD84wguSS16Gp
zy&%|lc9X$_vLFn`H)uDcE;HVLw9g-Wc!zg)kGOp}WzuM}!5Zf_wps5DNv)?1V;)a#
zvvq06=YIZe-n)JaEeamrIb#3uF(-Q`9L=ut%BMcVXt=`Hzw$LMT)xEd@iDXcjFWN3
z^$$Lv81}K&Fdm<9_S}WYm9PeH4S0tOlEKCqNP8TdObP2lLN~&R4)rmnOX}#Ulxm4-
zTB;^+JZqV>(W@{Lf(5p2nVn1+tgI0PwAK+~|B?004K7@{#0QV}NhQ2_VVzempQAq<
zL9fT$#LmL42akC7yYKNI{^S3csqoC(n*MNzGY)SIDof%h!9_D^x@!TV?Z(}E_hR~h
zlLRG@BCxQh{a)h!4?0o*Wy7Lt2e8OW{x~cOyhRel)8GHc{rOLeMRAy>DS4i=v9ZDW
z`Z_`gX0sV9D<k&z4`|z#q9_PHkapdTd-v{f@7_H=`+4HX_BkSss-t=Llfn*8ucjaH
z^go`1_rykmPIbB9;|M7_KCVAavWeWrg~yohC_K?)?3APDe3vg9KO$#ukvAJRg3}CH
z8x4%ZpW_JjF9H^wR8DJh0a7BhCMz<i=QtmtS^Z*s`f-B$(;pPag|0K~wmybfNtKYK
zIaRsK!~1ve!Xh)r`o<1IB(b2cOCH_5!R+`5SC=HtBCKb8bO=OF=EbuYki{lOdM1+z
za^H~lXGpCnik#J+L`sJvV7!f<t2)Mmc^@s1f>)iXdv{P1i-uDyUX9eug5%Hx>SV~~
z*(<D^y+UtohoE{OQiMpNHyxj(r~_wPntH<F{zIzCF@>p6I@-wuA-WE8?9zF7QqT$6
zjm;Mw=-6}-Iu>XP-^LI+PNxV+@#JtSmFy-#iw6>CjEnbN$#~Ee3(j2V4r2a^n8Z8R
zC03~@@CcSgl1~V(dtKw?!?tZVoqVcrjzZ~J*agpgRx_KGc<)GaO_rw=xuq>UO(V%u
zg_6$}nfPplosbx%$Tl|Fc}=5K!u5Gg^LQVrJE!$-6dQR55=DZyf~JNnCUPNzLE=%;
zlWI?v2DA>5N9SD}pNW8!0lBz$;w%Ft6urC$Dt^ry!)!L=aBq+KY{ur=TI7%k!E`#|
z;NXCjl{NDI03l-#{j8}GS|OCgt4NHDic7$Gf{ltyg@~DYUH+^H843m75Qw$~LJOSm
zRMydW#lhnl?|=A+?>yS)`ee$3Hn8*B=lG@1e4aEUsM?aXf#Z97jH`z@*>ZCDkU#w9
zyQs9sbY9cCg7wV}?mfJY%mO<*=Q)4=BD=f0F@V@PE?>S(GoNzj&TTA#qCaGHZ3}B{
z>~x06Tk<A$VuDT(Nycne)1`97F`ATQSxz8eyhH>8-XaCGHgGs^IH??7Y7!xk62^xI
zy!+0#S>4<uBry`FF)hKgwAGxqKl>T(eDm9!+e~@u<sG&MQNXf$vd{5!#@_w`_aD#r
zn}7d{eEQW__`kmPdrWQ{UcCGwMOKhV9g9V4y23u*g|cg@8{;`Ts*xTtDWhb6F>>YE
zrQu8b`C?HBElB8t!GM=udI@J8H*VaZZM*A0sZ*<-wrv;;hOCTMx{e;F-N@(ks2GpO
z93CDpo{Y)z4B2&p|D<~M>9K75Nulc_sMD`W*%fp;=1>dqBuCe}XkTJXduo#;rNEdr
zdJg&+ydH%Xwo|=2j9p~IKH>ZL5aZ+)HZ<KiczMuy3X63^DiK`{y7Nv<MvL8{G)q}o
zUt_*o<D89F8wiqK(M4E({~;*&ZfdpYm<P|%@r0W<?_k=15SlCJFLQkOn4`l7Jh=ZZ
zU--peLg|dAYB<^7;}8Drzh?jL9o%eA4&*X&p{LV1cOUFCTD`#V%nMA*lzR`ZQ`QzE
zPcR{n=P4@}RuRf|>$^oJnk3C4zsrNO7N^vc{O#aQ6PdgZ)YdX}k^`gJw+U;PUuFC1
z8x;K=tjNezN}`fjTy?MHN6oIgc6IkcFV016J!5kCh`oC^aI-^{tVp#-34fYj8~sN`
z$BKGV#32@ke0ka(6e<>fsR$V3x~W1(##z*7l2eY`A}PfB7TZ*(rr$~{TqimT2!!yl
z@I9S+c$@$)B~D6|R)nYj^9#{QXN%%vh(e;$I7Rc3i61@pJhrtwc(}*W!4!dDYkP%W
zUo+@yTx+N+N0BeMe4iZ87Aa;NCKD7Zs|*J#C<F%&9`W$x7^m9qZR}?G!Qy;;!+nr6
zjiFT@Ap`gpDIH1&w1y-J=p-(59zZIDXdyU4umm6F<Wfk4)TC&FbjSeZbjHcyA@k{!
zB28Hx4hYsVtxDRap;z=6uC8KyY?TV7$$JG!k<oUf(%>D5Q?anK9$c5nr@NxAQxds0
zj$QE$(13)-1gz?@KNY-p^C9owKIYMh<8fK@s198F>`(Jge)U)Rg}?F(9NydIcfa(T
z+<50*9+t=4e(-?9>WGu3!rr}$u`uf8XrHq_T4yjEv3qpLjT<+S*?>zgUghn#KZ7$3
z-~8HFI6gjRcmEL|-nz-luf5Kt%P(TBCDHQ9MQAO?*sf!sX&QqJk~~RCQWY~WeT;-L
z&ZC0GTZ6ZOvbIbrOX(m8g}|YLVDG_wF28V@<NZBEk<mKGY&IcygK&<ES1xhsrE9p8
z+jv)D>XzfV;r_vly>Z3SQORpBpXKVgvz+Yh(eL;9oj?30YbzTJWDdqrtPQc=A%*X}
zpyG|2=%kB_i!nc;^TZ17=~l#3i%GYjEO)XWUpzhDozLdnKR%|aYX0FD|2~&4UE=E1
z7b)_bo44;qufuK>dG6df*4EZIJU9TJUk7pVn0C(ZaCYZ0K?a(pB~6pgLH|$wJs(xX
z(skeBHJWrLwW&_xI@2*|2E#!YTNb);xQqP0&U+G|=+un4Ry4N3)DB}BnzjZ4X=j)Y
zfvT=L+XoppOBd*-i_81rdmtB{gIz)3JW=gUTw~I_;LP?}ZXX_CTN}3t5!hT?VKC~G
z=Jogd#5m$~&QUiFb=A@~jy&)4!VA}U^Nlz8%x6E#u*mrGZ{J{YvQKDN_{x|572o~N
zdmQZUQB9}#*)DcACRec-SlL)djYjw+$B2~ckB>ONeUY94`S_4Wj~-F&dAwBgiyoIY
z*CB-HGwCf!kt1Y`&Pvh@hdqt{Q;8r*Fijh|g#wKA?2lm*R!}S3Tz%`WkPOd&?!{QL
zL`GBX*cAa3;IOX7nu@xfQWSl3k|9+FWC%NDx_^hqAAXz3{p)PVhEd`vRHSD?EHXV&
z5xXEs3q{O>RkG{ebuaa4i!~Y%YNhZt7FDsv45yq^83Z;&V|J-Dd9TOuY)VzmaBYiP
zc+!b(B&_lnJ=QgMAP}6V?e1^Y6@-y<8O2O-dKptpgwk>16hd@!h!-2kC|WeGVO~{~
zWlKDZP1!k@p;JLN5TtWUFN;RY&w^(&C>D}D?~@N#NP%HhbN;2P*pov<dMjSeA#~O5
z!T?it#hu4_NR%K+1Vx(AO9Of0QPQBg+!o(;$&?mxENfyUh_#mBVz!+Kk~GaoRmyBK
zrJk2)VA$(Xq$$o?rYB<{&>ya$QcdeDN^0<)UcZNRHs-sgDYN-J5?z%>gqRX1WE4y+
zTEWhHy!4%f+Tp#$0i{p4`*_CJzWs=M`z;4(-apzW?~VBE&;BgG^f&(woB0Zphx-^;
z^6u?-_>KSX|K$GO9+P>62_A<g1VJLel`X@>Few}Q{fsP0_@h7k7BU<1*`NMduD<j#
zO2IpS^bPLZz0KFZ@im-)mGupra~)Bh&WN_gwGGyKs<MnKJ90!IO0Z>TMdIR3>|H>L
z06w~GPiAvYrgItx22bk@XScVxa^V6pc&@+u9ggNRW^EmX7q(?K9kcu3kZW&zirc?)
zi>hgOG@f!evm7^y`}-xYytK{hpL&A?!O7tPgJGZ7-+Y7L_{}fz?|tEyc=?4Fsmc;`
z4nB}*f686Tix~5?DBTumD2W7nIxXuW?*1|p>;3}a=}~N!=Y$aRV8@d&7cX8UNi_4>
zoCgmdQI%EXUYb}4ZftCj7e_!OgFn~D)p?pZ>nO|Wsm1-1;c32I^w|oL?%_L<dZ(Tf
zB34>y7S}g@%Hfd|y?#u#>Ha>?vo6aLL`_urIYr`QcBK$>YTM4*#Jet1><Pba$+7z=
z7Vl_WH)Bz^5KllC3npSNYAn*bE?%lpNsQ)FAJbBa<3QWRyP0UzuQX@R?r`q>1%|^B
z)$D}*gI#v--{SCr;6{Ms{bQ<mi8BoYZAhiVC18SPb^8(<7cU`ueQG4L%A)du+!(el
z@36c7m`NKro;i-j4Xp{J$&-<?Qu>K0bnqY|&&zjdA8DS($K6?~)W$NIOXTVr+vi^+
z?`<Mv8aZxSqIEzd9#s69wzVAWJ?7;2fbsZ<vKrIx_bGaPdc}ZVF~rm}jvwFPc<(kn
z)iP3oo`yukR0|&zo@iGhx)Xf#krPicm=`O3cTx$lrUE(_070ZoxhJuCB6Rhq?09oh
zpjE<fWd$V!6;rDD6xW(q+)0T@D$=YVNt4b-C5kd|K6c*(SS#^J5UPtIQyp(Fc7c^r
zosXT1#hnD}46QTKzfWm|Nbt^5l`Tyzkva-zbQbO4<TLJN&s5kcmBco#O5)fcO<CR9
z!M^qy`T7hvibUY9MJhYVbX1}U@Lfio^l@e^0z&$DW2mHSl{y55OnYi&uu>61f@z^K
zo?OQ$oh<DmjK|MPlxayMq)Oo(7z;saMx!CBm(!+1@2(^x1i`ed$crL|@*2mm=ux)~
z&Rew5D49TrsT$S?obp)dNPL5^k+fX4fpHD*-QDB&zI~sGU*V|MJSdM@yKsd!UwwmD
zUVMqfTaF&x=YRR1{|DZ`{T_ex?jLc|OsNQ133v%XAyJ4BvtR2L=z^S;END&3?Hkwm
z_Ba24t<51Hy#GG`_m{rJ`|o}?diW`*=M_bk#lUA`yS}bUY+K`;XF5KnsY;a0!AC<%
z;|xK>j%08x%1e}%7_De*VBC0SH7GKQ(8ey=@gZOT&EMqOXMc(lFZs%MzeD3Jt!?n$
zpd{p3kN#Ud*3Vtx!@c{Aw4w}>`$toDF0AvpH(zHx$tlOjxQ!7q8Suun*I3m<e&=`p
zU+&$%%P)N4-$6B+EJ+FB$(SQeGkR%`bkW`}PrG1hAz38(Jz<WWvhJSku2|yF6pMUX
zYb;U-hQk5#`HcPjeUe18vNED-D&BnaO)gxx$Y?a+5B}im%w{vb_3dx*>9=E6thLX#
zusa<Qhk!BB20};~J)WMW3;fi-i=DgB@f!m!x<aFU)Ndy`=1qrADXSArnxrTt$@79N
zOHYe8Wz}f`D5Z?nOr6hH<luGA_dcACV3#z7@5f@#T#Tw+_d10I&#o(i{DO@o1X2sq
zUWU>VrQ{R-<VO%rN8!;fLsDcpYinz)tgfPxjOlE~(a|CMyN@}sEy}dG#u2QG`@uk>
zH3T7XD0-`BID6?T*=U7ENXD}{!C0h_tZ#4NwWO{Ev&J(k+bEvv_T-omtE0HC<F5rF
zx<*XobBf@*@uSCB&$KdBw#Q1b#`@MJbkZZpML0lA&p>*7aG17ba&*FjJ2%-s*yH&4
zh`O1P7b!)aQ{*{$UeK0P+VKO-$$e5^;xQNuB3Kk2<uzV)=_FD{v)@HW6;8cc1)}SA
zcgHQ!H6|9LTQBh<zL}@42wk_|$9SArOXR&SH#z2}yVgeI;$RWRA-o5d5|RX^qo5<W
z_!_z3yB$V>RHxP|Qi>-9o)n$gb?4C_e0P6ZoR3W{sbZJ1sRC8)$@@_(q7y-)f5e4d
znq{O(h7u~8XsVQSb)Dg*OB8F5fcN9pDm2Oq5E2<I5{DE4;dG~pEdhg&oj5@zh>*uO
z%}-Dm65&u-Y>?nH+D1{9B}vhs6N#1*SJs5GMA(M3S5V|RUTBh{V5Qe*Wo>}f4yy%e
zTA)NqaFE$>N<>zjOjm0~TgT|IBGoZ#RCwYb(Bm;^TjE;5tWw<DZ~4Oy_xZ-necJve
zCnyeG!wavx#+8fbSz8$~9Uo!sjDPv-|2x0?_1|Y)mDJwS2C!X+Ie3r4k|Ib%LZOHB
zl7c)f$Wnz5Ey5h|%8M_s_uvNq?SJ!S5Q5#u`($a#qUAs84LG}X28Ez8EzZUGl%|>E
zjOWpv+xV)Ym!*+ophPqv76Bg&0!QlNsJRh=Lu=WcHB^lt@seIJoGk=9E0Ve`IlOh3
zm%i}V`4_+aWs*E&ZFR`{TA#tNU^p0&_cFwAo%cSt!^R+`HYGk$yngi@Ygh(Vg(1Ta
zQe>hTh?JlC)Z1LRc#dEHjbCT;`UkxEQ=f~mD``Ca_~6m0W_`3yUp_*3gANWE0y;$U
ztPQraBv|yW7ZSHWi5T^4H_j5z4+4Q!F}D)p7_zQw9zK3V-pfdfB<3s&&|01f1gx_p
zNx~w~@Zj*^xjruMqDQ0b+^zFGrz~q+Cpz(;Af`Tq@X^k3h>N`KjDw|!>UD&otZSsw
zG4?P7T!`^{QYmzrkQGrW?1RO$HWnW833<P64NWJ!s5?cy4^O8jEHCqqMMq>eE_A+h
z@pbZm5TW~;V^&u23lfqnVYEI(C2<O+#YcDE$fR;0;{PRDvNG%=JTa0lFK`GBkB&G#
zI^_89h@>?P6{rxqOA=5)qJp3eV1l3+Zm_-c0xMf*a7s~|hS_AyVQ-J7DiNuot_`K}
zl(oUzwxg^mv`h(PWNA*K5|mT~G}3nyC};7m>G*cg*ud_w!6rlU(HYjZcA~RI=dHL%
zUjl<^Yo@a?w?6ogn;+g_Jf2ckH4a0J(Xs?@FwG2Go-mYiR+EyUXmFt=k&Z+K@<fuS
zDZR8Mq=rmE=F{#mNvr_d6>hOuRmec(VF!;?aaxFqd%SW2Zv(-FQyUw6XHg@-NYE-F
z9gNU<j!tu?@_=eOqpd1Jm+BI9gH4inA7;hrWYRc?^)8l}GC>89Izda-sgFe`Wa?D3
z0xTBm1=huwN2OboicTC&C7Db-I!lmRlI4ypF|lBM20Ys+R%x_~VZ&YoobbpZp;%j?
zzrF^5lh6tu+lw}!@uWyJoi$t#2on1Jl+AU8mJU-#S-!I#>n%!0Gu0#!v~`s3TN^)`
z#*ijGRH7-T71OpPXhpHMiXQe+NlxDHqY@b_Vb_ql0A&NG-9gnE4F@cdc<h@f+-Pl$
zQNAnsB&n6i%Hpiy$U1hXB_G^A<PSgG<K{ulVEaX8K~W?5+|T_S+gls>W=@i8%5uzm
zAHKtvzxumOnvyzrOo$z&c(90a=|m)?GAES<sV>l}0OHB2D0=h<8BJw4JldsoH7{Pf
ziZO=$y+ev3C(V0&{^x$4?eiCqI>A;oRar%kva&+Bz|HsH0}tC98we3E`csGdPSGQX
zS*;;>s@aU=!$X?NvXb`MSYKs*Lo!;+xpOe%!rL#ib@2>8{nI}cc@4s1+lp#3V|FlR
z-aEk@95J2FP~=dTT)i-2RVz|02)M|7l>({a$%X(gyl{no@DKk1+UdA~YDH7GPYCG*
zR@T<YdO7onq4JJGDx`>xf{7C}gl^MalI#8=K#0gcDYBgL_?WV+2qAF$_8r#NR@vG*
zgYU8?jWLl@pd?LQv%0c!n#ged`uorIaVe!K@*ZiL;jBfeMVuGUc3D0BZ%?~GU52lS
zVy#fuJlfr5I4tS+dh`bavbIHPjSGRgX_(Gt<V6ujv#KkIlLYM?#yXs@Iwhe->lB?>
zq|^wyVEE;|@O@aU5DW9+cqn{Qq{TWQax9Myk7yecN9rMVS?aFX``A&e@EteIv%WFn
z^FQ@z?%#VrQ&)_~V~XL3n>TJTJ(^-Giwe-#=%co90*GTiXpy+2$L85{j7IAWdn1C<
z41i%iq^ApfW6?#yY;35Er)(O^YR397Lu!d6!J)b-jX-OK4JnC=f=5D&cQdTD)Qw;~
z54?YO%xgdY1=e>ik@WinDV}sFJEQHUEIB^jXK(i*`wt&bqy=l2w<!7p8f!T|+UMx-
z5eNGZ*n4=7cCyDzG|Z%-o=z!COYl*V+RHSfUc%}iV|`R`@r-3>ZH1!9$a00s6iO_p
zHu2EpqBl)^Od<C17K)gl0tiPC(dhoP3rno>oCl-gy{>hSVN#GK1ywPonoO8aj%k{T
zwr*q7N-2s#|A}P^(OXubkx0C8$h`BNJ3XA)_+X?$NyV(JD9eVXsj1tNAj{YR6_QXx
zIT!R-1%-w@OX#QeN6WE`u@{jXXq=&SEiPEB23bV=fn?Z2_7lR~k}HMwfecTwn2HO9
zcL~mhI7<(Ppe(r1836)LqZ3Kq6Vw&4+7BM35^QUuYr6N0%Q023tgdd+Ti@cSthw{x
zHk(^(ym0vfc3#I$n2%3YCINDR6AqQ9I4rYjhG{IBEl7(5tpi3jIBU>W;jDt0r<}KZ
zYd+=oZ`|Sf{S)?TSl_wAfk^4EZSo)dy?>v<V94FuA5g4j+`Ms}4{u%PyC1&Kq?%y~
zG!T9A7A2jJgijGtM92|7BiNLx4S3fw5p(*vW^KKoSDayOeT#Rm{}G1=$MkwV&Y#cd
zjn;VM&A0ihf9<c6=L4!Hp46MFrL7udfM!19?)&evmMb<^`Xo|x-bIT%!ARj2QV?9A
zsT!t7CsfBR+rx}EUVD*^&5XwEQ_7aJ{dIz>IXHSiU7fJIzYE6D&MQpYB4xsTFX(GQ
zmI!KRIg@JEoa9WFlj?xM)7a?Em1HSOYmy{EDH&}q$S6(Z4+&@DMrbXeZRz!M$W!WR
z<ikh_sgDF;DTuj<bm<$Bi1&jpYVkdvmwVrzIpQWG`P}E;=43qPJMX*;z-&5aHk$)+
zETMG#kW!N8S?r`3L(|4l{k?nlp6lb1LZ15RIqOfw4?HD=_|xEN?j3L%7Z&qgBfqe!
zYs#{W{G_6YS-w)TySEoT2fZcBvo41@8nd=-dn&S6cyCInPSwPx2FFX1-A54U3hj^l
z?V{w1Vis!+lgSk89f&C6DvBOeHAX3gPCjNU6_J9J>XhMN$oAQ@cn9N?G3ld6tgfwb
zw7ZLK8bmiPTqrmZ;Dw+u67K{qNl3CjgV7pZNUFwSg2jbs0<Uz+aJWLEQv?F%ELClo
z&MMmVluXCzlqYtgZC5;oML2HP&lSY9pt7+XO$_r!vA%ttm5t5L$1ld&Ek>ixJI0eS
zC&$Opm11*)UVnxDU^S*Br70&zyWG7!M1+>Z$G0#xFl`-8YdP7sjOQL_0(e0x16kq8
zb3?yZv9(cg>Fk)ZTPtj@k2te7Vpt?3I*x)vXmBdvg+wBeE+cjfp1g+v3o6zbAy|Y1
z;p6Btc<?fINFC8jGwDUcbDe6ETrr<cXiZCN40YWwo6S)=CR-#~N-r%CO5p>T)-WsQ
z<RPGwv=ftd7Ar#FgJ&{3VK$x7wl#tpAv%vxBslOk3RQguI)N+`akBUfc*ep`AtV^r
z(wd60t!SGPZ(CM6Q^n!d2J07k9N&I~Y&@wC#H6+^b_azqij$K$LU~3jprj)xkMI)d
z6rfRwAn$9^xutDm)~mM~QjJ7WR~1cDv9-0uVErryCk@~I@DBGL9P#>VjzPbmmu0kG
zAcT$^eQR*@7O5p(%t(tkl600hK1QQuplK_r>6lp?uqI`4Vz_Z<moGm)=7UMioP^Pt
z3sk*zE?<3>zx%g;nO>SPKRIE2b;SSk<=^D1-~1}Ga!TW($ifhdIu{m1PKiXL$tm=R
zO!dKPyj3JJjVBbL(TQMsGUfVr_qh1N7T2zPmetK3S+02T>J_fM{4%e<^){oG)yNSk
zOXjmV-dNhQMp(z=o444#eTTO%Y|+<}EKy*hcim|v)3th~P-qlAt=Jj~&hD)9#)}&W
zQDR)i#0hfM<NBk!eB<}N#on~$=w!lZkT4qd*xI_p=H@mxKDf#KM|)^z84Ay~fL>6H
zR06^>vn@sr8TNWqX&N`Xj@+B28PpzcU3|^FiOQ~~!Oq6?@&uhHSRD<Dvp_V+7LjL}
z+b*P<KiQh%dj+&h{MjS;K$a@j*M@9uu5st)ZKl%-YYdMcKE}Rb5Iz<wi%qoG@5NAY
z(-3@o1D~(mN#s(cX-b+ze?0Hp4_)|woEPBtMzUEO<B_VWqAW}DJSWd{q?AmjQ`(LX
zoELc{crV-}1Tl86D-@$grC*9A{un%6)CHahCRlv#ykVTRlvRc9R9T`^Dn~mSjaJ|9
z2`Rc6xssA3N$K?moY~&y&YcGw?(eeyXq9?;g7TJ3Ns<sqLp1xB#3i}4U=awF&@Tpz
zR@X_h9zrDuAHX@JQY1Pf&w9~%L_|Sc+j=ImhNcZ9X%qy~U2EPutn0Y%LLu;{9VQnf
zmGMkUgI5KEm34~#pcCdjDe3@>vrMNk#x4+8U0orFBqlT|$;QSeT6zxl9(FHo6dANu
z(3FmodEjIgFfGI(PC!*2ojB60<?*!TVBWAZZaKf%a^;NU!sd|SAd5C2X#nL3BH$u7
z0TE(i#1q-9?3__zXL4c7BD^@wD_V?Zy%gv)A)!EOiP8d;AZc5YwgolEbcVx(=**C4
zjnql>3UoF`$4N;Vd`!ELohn@foVS$KjQM<qvn@#y8G};9ge&O-!7uV<K_`MV?M$Yh
z0ne5^A!0zdF*Q{+qikl>)f{J9n$9<AxUt60OXs0=%<n$#=GneeEK7nH1X9YTLQE}r
zKSw7uLJ>D|VuXkD0i8<Ho}jgn1ZrDJnkn!CC18DZ#OB5(<63g-&LJP(KBgEJjMla&
zD@QPv%1dTA>Z+lh*UTmpOw-`4<?PmoYgaGPD|(;>zO^L5&>WsHIXY&)Y}jiR?>?UL
z-G>wQoW{u!n>&~B`G9M$zr}z4tG~*nbKBf{=Y94~!>|3?m-y;CU!iI%N*8%BPIR#}
z-GvtmI*leJl|y>T8vT3~TtaY?A}#0-hO888Xswu)#~3%GnfEw<;T$hsKFjBS_HABy
z>t(VcK@~Y|(@@T4Oisp;A26R2>Y89%zWeoWkTeZ%UA@RKO9(c`4A~F}Lg0Lh2oC4s
z$y6Y~I<A}_aQX5knXDkx*tWzRPdGf;rK}an+#=?NjlSm0*$s+*ADN|8ZB3ddq<xJh
z+AX-&B6UJlHRJ=0#gk+yS)QGqR;)4T5W2u+NOVGF4Z*cQz|;-q_?Y6%HiJ<C`wC|*
zt&bgHr6on85lRpmL%mqAJO8ndP@SIbK3U>9W07GSLSSvBU}cnY@9r&Ld-ZiLT|CPN
zAKu2Az~R9@AAWE>#$<KQ?n=q{Su57oRw;^TH6c~BT6jK)8(@+qq-pe^T5#)x_~B0i
zAJ;WoM6@jmJK!`Y_UQ<>7%SEt#8ebqq-jEyM=J|w9i}xHYvN?2Q{??HUB;he1fDMH
zo<MZxUg3#(<aHgDmYwLp1y7PF(nOQ!@44>cQV+H}$VZtaXV33&_x?i;_IL4h3H5}5
zhCEMD1ajf<Z5$hVz&J-~AS*Pzm36kxo@2PWfzT<|2duLQ@2TaKWZ@|%V)kj>1P;dy
z^U9ItvABySeUT$?osE%`Iv#*SXL{WR#k}#H%v+M-d9uL@NnUjEn)0+T3IUp?W-=aQ
zjip!g==b}4aN{=4O9rD=hNCr_#&CQ*=F!7loDC#8r=q5^o>mB2A+SnBmxkE4LV8dw
zOj^VK)Nt!z#YX>xYum?s`lTH%?5uKTbHHdt)A=F;9+U$w5kbVLu(05uhR$o)A=E-_
z(RFD%OS~WjPWn#KnxVBqYl%oSzEheH>Y8$1Vp<dV@k)|vjm~qt5IFCdl_#AjB~T<W
z-AYP{4<2J1s&Yo%)CBKH)2Q^7i;klY;9@F9@E$24OFUBk2sw7K?Qdx6in^TA)+MH?
zJ0f$uu=CM~^VhDCIKjQTrF^_k(z#jV173m&U~NJ@H6%UH+D1;IS}YAx$2YfXYqSEX
z<ArApRAq}03b7_x9gWCDPSaW*J}$ZQaKf||Y;J5|(*X|-k7&n7oLI}fYR+`h;A(?t
zJgJ9aDhRb<t3RT5E~gDOZCz0xj~O2xaWt>lNAcC&L%z0GF;$XGkJ#RNfqbyW`p$X&
z$$$Ev@z$qaA{fus<{JO(fBf(G+Bd$+*v@E!i<XsPv1r7CPLQBON=izu2js~Ly=<LS
z_E9n=?d9ZIPLiZZ(IZd#^fuQ}+EMhOt>NA8+~e@@n6<TauDtmoZCi74e9ZXhBzpRd
zkMZRUX&UZ+c%AQj<6HdH)g4~BxJ{3iewty8#RL&eD-lQ!UPUQbq55pBt&<l8y<SRF
zP0-rMlTa(!$OfdV>l`+_WUFg5GLZChD(|puO)VTUk?fq`Mj^<{8auBj^Ncb{5FQ4_
zfG`{(r0hJF45~lGcczZg1w;r0=McVAcKLwW+hhIACac4!Y8EP?wT4_mq9j^C7KO%0
z50wYg6?F=6x*`6c#e9h$Jr)JsVx;<uzxeMmKHlTTjT_v&d4uikP4@TqD9f6nNVt3V
zHr83ro!>d-*R8Lw(d+f`E|6uB26E~w`#cb!E-RE06upf8pa8_F<Ggdv@pr*Tx_0nn
z)cb>qvBh=leH`1=b&Hsmr;q*VpS2vl{_*(!$iXLVeDT!Rybm}Rg)e1QvAeg=>dFw6
zi0>0;ErgkPkRdums3b*b&C4%c;ql$O+<N!Btg0#3Uf-lh&Z2}yTSq;eFrQbPOiFI;
z))+4s6n!pSzQX#MErLvdn4w-MiEkTRUE_^Mt0bzbt)OvmIB7YWG$g5HAazt)%9!d9
z_wyQv2NxkFLME8jF_{|<CWf_Z=TP|&kH#AtJ!Z8&<z`POGv@OWfkR2n?K`)5^zb3p
z33}r>=gvLmV1JL3qX})}I6OS2JUPJDHQsvijUJnulC((33Qh2yqX%OWpOK3K(LfnI
zb=`7z-ty?+kPjc1TsphT3+LDQ)YUV*urs396X?|8)eNOlq)(A90jE0K9}$Hj6wtXW
z1OXz-RXd(#KuCcVBHHm0P0>kHTpPt3Ns+Lc6?hX9V(PMHI+@ZFkmMP8k+B*^I2R|B
z-n*`G82O0K#gwi@$LUy-MC+zb3W5Se=wcc}RQYBJKU!6+4`5A8UCn9AIi{-cHWmh|
zBNeZoJ>u1$d(6da6&GGpoILQT){qHFDg{9VY6s3)jB^Z!fz7RqRD-QOGAtz57NrEv
z2%6gCjDX<LI&gNU=FE0Quq8()HG}hx^Oq_%MauShPgXp_93JCNDst=D34v`NH~e8H
z7-bndn=5Rt4(X*1=Wl}+v{i#Wo}i6q=q0ntuya!J1u1EDLQg2pZhDN{=db;Z&+w^F
zzk^WkQCD-`yZ76OwSUE5{gpZOLO$Mg@)7aWCOZz83Z6s+a_vcUjaDaUS)*h^nxvf{
zl_Zf-!IWlkG%b~*byLdvkg`7F!Tv56?B{7+gAWr_I%l+6QKTiltr4c-!Q-#<vwvgC
z-+Ft8v#XCNghd69FqTAWJTY=b_|6)ki#y3mN0K%;H^;b^Ua!Ek4t?lof}q&WSUFr_
zVl8u9qjN#)15If0E+Cx4I*(}#)HQ<;*cb`=mvUC}coMQX$gPxfIwNr@tAin=inOSZ
zR$yFUH8@f;MJmJUI{eHpo-lmj5&ah`E?x21)}cJ~5<#8{g6|@gg1`cm3$#8u{i`54
z!3f<cPMo?eeguC;`=Ai8x(Y=BU;HANAJVNuEUvd7@aG=~EEYKcDLp$otNhI81&<yY
zZr^qobCZ6*&r7dJ%6WWkl$zt5V}0X<GiUB|_1ZC2WqI+%b)--XN0F2FJR4Fvar~qI
z=r?Gsr>-j|lPQ7N4O?*4{#^0Bx*+jA;@dyx_kB40=$|iLf`9(cpKmezqsGFpxO2ro
zcc%D30Lq|}mNc0$=+9VR|LAkei|=s$!hZqYckz9eD@}va3ez++RmFTVL5Jp)|0a-#
z5OCJSuB&y-8%OH|Y4MPijeo<+_6_JPK2ArC!82Ghtel<mUvA%}E-Qq0q!QNqGggQ9
z$g)nwDmxppQ~p_q-mr*M=Yf%f&C2jrX-NjZ#o2QokYw5E?-N3ucCB?*Gg_M?2r-$(
zI<CD@<AXrygtfK*jrB7P=PpnA+%KF^%_lh1#``D4?+HS5sUI;e&r}9Mi=u>6`x)Qa
z`FOI#Qsgzg{E(IYU3z(fQpE5B;ZD77pE9?5(zSc~I3KzL(xQFWnI4~dRYq?j67R_u
z+|)qF-;S1H&UIEcTAx1tk?a0#=Yy-1?tWZg)mv7}>U4snl#SM7wZXOmUk6AfI`ilx
zpcFV8SikCNpXJ&OpWrw*nM_z&S>gZqr~eEk9h30^<!nM-mG}T!Xs*0+pI`n*|LZgS
z>Lr#~Vu|O6fBc`o|MX8`JRUO|jh^Xy7i-*j4A1zjODwU(5<hZGCa|*d$=0_|)UjLC
zekZdr!PLwqN3>0eYugwKuM!^LAMvk$?e8MBhR$0|d5fLS89#i)^wA!&wIC#A2pB6t
zXrLw0CF|=N5e$idM1q!4zATV9Be6z+7X<G>7>d5a;Fz|8-LYV=5!3*}AtmHeFic0h
zus-7aP*LPr^s7?~r+167HBB{VKARBQK<Q!Mz^yS%14xzNoJ65Hv%bwiIpzGN%e?XW
z8{|pK>dJ`U|LX7Yt?zu3+E>`cLEs}NX@Ll$<I`p2DnrX0rE;{&(K18p94RxDR3xb)
zO&30MZI|;HD9e)Z^oV(NOppyav0S}&fvufwq>iyAnNlb~I>*iT-{;P)d%Se<3@`4S
zVO3aiqGz5Ig2sAOn&E=OG!;q)RFdGmplvMHd5T_!v6k6vO3`2C;P8aKrbex;F>W2>
zy5&o;0#2^DtV%f!JVRWN<Iv(c2{jfEGr}G>_#BqxUB8|`a-x*6#D;G-x39P?XB%p8
zqMC*En3RyovC2IkPW)_{9se`-*TID5^jXWc{1tuozHev!^FWhSF~+;S$%Xs6|1^qa
z#Jn-fc6om5eyhpnkjszN9YnQ{H0?gtZ<-mm(&E9bd-tAA4Ub!TLIil>kc_NsQAveI
z=u{8Q6Q91ANSDQhtyw=mF*mpOWXh?3|LjWde)x2H{emfvn)&lS<kzeBDg8HDeEjHl
zAGPzJXX*hHg}05o&b4L!oA%W%>bv{*_xGI@A0P4mJpf8hs&!X(l)sk~*NdrG_C9i5
zAt({Wv(!xb`c_!9>|V8;UCoICiyFPI4UojwE5LO5<&@&*=jNt))dlHX-vJbUbX@X0
z$mm0PonS>^LcZQ+@!PvX;4}$InCl|M%8<B<ROJ5n&*-GU#`{@b+lv7RJYD@<);T3K
F0RXe?-cJAk

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_getting_started_004.png b/0.16/_images/sphx_glr_plot_transforms_getting_started_004.png
new file mode 100644
index 0000000000000000000000000000000000000000..e51e8653745d99c90c7e55a6d1b599c8cd00eb92
GIT binary patch
literal 264577
zcmY&<WmFtdvn}rKFt`K>?(Xgo+}&LUcS~>xfn;zE!JXjlFt|=|cXxYy_r81Adp~;3
z>gt~A^W#+Qs=e!cQ&*KkM<GFhf`UR<keAkkf`S1;K|zZkA$(Y(Y1qa-egwQ^^t`^f
z+IabxdssuMn0vW7xq3O-TTlY6Jv{AQUAS2JSh<)f?Yz9)JO$a<od5R)R#y*OHn6zV
z!G{xMH+g+eC@4(x|6I_eV!!R7prN1?q$RX`vrb;^tDO}<`^c~3P7?uwv)R>{*yPv(
z(mFN^;yQeER0frj6Nutl=%&!$lAyQH*|2fgu#tEy?2udYWKg|z301VMt+i~lY$9Wl
zd_=c%7R_MU#+^<Z?*$fDU2Tf;z6|}l7YWddU;(^OzW<8!1iPD_jRm!^-S2k(%Xs!M
zeFK>8yjOpA`~R)MLc8QktN(faKQF(KG2VBd{r7+VukBpE1rhi^e*ViAA~z~h)Aql<
zKWvXenrC6F|Ihc%#@16jFi-x+?SHQ!F)hH%=YK!-VY~7vq7~brbY!VE>9FAZMH%@3
zsha(7FM)3E;94VAzo#2U&ETKo{#}Frwa&NqmcZesyNJ-$a^32OgTkcX&p-a}O1dVl
zPMg=SULLP81^0s!{>%Epm;aJ{dx{9%Hev<S{eGSMYK!L=yFli;ZhYnXcAI-W`8NBW
zH5u|6Ue!3aha#1bN2@lt;}GkH3`W5zPR0>^Jwy(9!4&(Xz!iM`xcCAQd#3Z#xS)T3
z=o(mm4(J-X96>4;r$yE*r4&KUy8en!D~S{d6%&Ho9(mfe-a(Q605KlwKQBUC1QKLP
zMG7Gq{r%Nr5YgCFJ!vk>OGQeq{OnKvO>&sl$TZ)oE~W{7s{^fPa}gZIOXA3~u+n*^
zsW?MRr+~+#0GzD7PYnu6K%T^^!p+?bl8)+H-}D@e=&R>%ag?INMoF%#jzbCYncPxG
z3hb*}%<0n=UJ&ppf4_s+cLgu3ysd#-I`)N6xb|Ix%f+4pki#?`rc`k*oR30)^29^Q
z;*4S9lp3K+JWhW*8wqe%)<_Z(Ua>>?2H(+#6iK%d$3to<DjSJr&7Fp91ol^t*?b>w
zb8Ah4zp10Tu~dB_`kt3<T|e*SpjQna_k)H_?J+{^2{(BE{w;ZXHl~!hZ}SOA-lR|=
z1`gwCeqwCO^u(}b`TF2QaBN}BvOv<%#P2voko8^9_PpIh%q4rZda|ch1#;E(VER`5
ze0wCND7^pvmisD3lrXeuh{`(7LG_A+D^9e(aj+77>pw9+K*wg%A@0MXkj|N{?Dw?8
zK|e&VOwUy^OH(picy0~5US)cnVfq4i8|Zqoefj<V_#VAmGANjgOoZqR>wIjwAfD4#
z8a{$&gDWn%__Q4v>Oif46PDU37LPnhNu_|V?;fw{QKZ~$zjBALI<M#KMv@in?%jVl
z%@uqv_x>|C@G8mI_lnVW);3F-BE?ofGf+&(A<Mj&aPAZ(27ws^we1fye^#{qrI>Bj
z`?=^}%=3?+X)_`~eK?>Ap8C&*Ha3dS5f!0Pz^STsU8-1Yf4Jkh&Sd#RGTVZo5DxuP
z1Rp7$*K7K%o?z3AyNX{>uvdTR@zIbK%x0I#llP}AR(P^zi5w0dO$yGpR`m=clmNQ>
z;#o@l!gxa!DGUj4h_!@_2qw;|C$fAxHgLqOvc~T1Q@IEg{{u%wtNBUZo$vt12dcmA
z*dhLq?`HG-P4tcDB^dF_s<aw;5h?SOyrvBu^pp^IA#fKZW)h`Fej-noqDY-~z}MZD
zBLy{cS|t$pruUm_sioW)7<3laYu<sGJ0dnjB_8X14$K@gi<jx%3nP=(fVmq=O4~Hy
z4A=YS=5y}}w>U^tadsE_bkX7yw4JQ+mne;Dn6I_iF}-5n5p)W_Fs`5-qlJ;xO`VKm
zv_Fg+mQL_psTBhR92g@sy}X#lO;l55xaL5M#M~#<JcElsimYw(W6$z=<>=kaLUum7
z8$7${;5EYxeRWHOCf`|uemR3V4k6Qa5dO`KQdYqKpk1I9o;Esp%FYVa1VNWIoZ+Vb
zK`beeZk495ps%H6J>YBP&P~RZquSN%qGBN+5IlZ4Ej8+{suWX;z}Tb5j*xcC23Lw_
zg`z2ude4u%U#_5Ikct<1vufS;AP@7)w#b?wM`Z?+gK7$gnUa+D59K7hb&BRL{CAIj
zC+<x<L7K5T%7Z{34sbFkQYB++iXSZ<!;M@l=8t`&rKYtqg7Jn??6bL{HbXcCH_V$3
z_t_Sr=~E1y{k`XeUc^ExJ$`79ApRNs6PL@px$B|FtMxN1V%`2J_1!wgq9EhWhSj?o
zt>w!tFkr~oay@G0P6T*NMEJJ;etb*{Uv*)qM{@@Tx>QR<(^#H^YB9pNR{#<gQ@^+J
zv{!@c!~4+XlQ(lXgy*9uVvcRv<=fZJl;vkJ4cA4jn#?K(^~P-K_a-v=+5;bBt2h`^
zpfMQn^$1DEqa-%i!J8K+(vStBSxs%?q};5FWwo5|yNLy<NcOO>)G091li~eQiW~nh
z;5w4>0m1L9%9E$tzzJeZn9axxR>Xbu>bACpxBew)l$bP(edaSLEYdovzT+|IG5qr%
zv?JM2z+u(Xdwr;!-i?Z5K5d9=0RiBNDZIfv=nN+7?c9Y!Efp?V=&shb)i*>yPa!&Q
zKiRzPxaIEIF?aRbq=>K>C-c`HxI_9(RNUe7Xd$3ejmK7RKV)_2-si+pOtsde0=xIF
z+hvKb86M@E?aK~}+gz>`Tn^&+&ud&Md1Br((8N>t{1N^jD^}e=97zs3*8-d1H_al(
zhLsfBL|LD9zv~ZF6W#5JKupt)8odez4?Lkz-HCT1ZJgKSMoY$=_IUhDhLeQT3Wq|Q
zdGznV%xVb^Tq6+NT5lUXa{Vpe>6!@yn^zRu11dOZ6^@>@Q7X8I3Ts6Qp3bP{u{dR{
ziJ25MW?VTW!y74Od~Y4M)W`GI8_fBfR?U`%M0Dk0l<0c}szMCtdKY2TgV8g3Z&|P<
z`eQb1!>aFByZUQkaRSO4tB>B1<L*MR$Z6|#Okg>U#3FAzC*BG5JP)J5db-D?oZ}<I
z+-Hdxp2Tse?=okr@1Ij>OD`UUa?*^G{@waLcmxEDfpe!f?A)CGo~>O~hg`nziM@%v
zxjyc{zvRv~Bu)?zC`}puRN35N-t6g%Kh8v7d9X>r5t}p=14>>`bJui+4G2t0zG2nD
zXu_A{+mR&>pknx8==kP^c);YoK1BXjoAiBp;ShV8{;m9ps|E&Q?^P}!AX3cW8Hv>%
z{2bl7mTb+qTWY~cTyMaZqF7L5(PFZdRmd!>HUE?OT93M^ngcZNpknO0ykJNWhhILQ
zb!gSQe8@N{=C5jN;6mV8<x=H)dqn3wi(sx>Z&0aMxyb(_%JVC6Bo)mO-5FA2#H!Cq
zu%oZ4pB69srKBFW%t108Uk}k9lw$p+O(3A3zGW1{L1wJ`<@jUs>j*83p;w!B$=HR>
zf8f$bc$6?Gj+U8mgAbKIZY&xN+=ILa7|QuY1tl2=JQ@qucxmmU#b+8pBTl354<R?<
zlR1@ZTzf7Vm@vSVhKEDvk?Z(NU+GjPti;3qs~0X6M=}^CCP1uL1RgS2q5#f-FH2-j
z*#8P-OHs_f&o5f}mtTx-5<o8EKXLgje^@D*)ydz2UZM4MzQ}YiCeOKscr_IP7!&ZP
zO^j{Jxl3`BJot331`J<2<A+ZSzFl1Z3=Ve-`*l4NpK&9j&T_6bf;Lwp)ZYb>-gd7|
z^idRsHajssh8k|YAYIVMJcG%*KKF@>O;uoPMEga|_g7boMs7qy5uZR|zKc+EV@F-_
zGOp@?kh5!gc&Zt@;qMf9q19vt0!?;zG=nN&e!o&10IM@}nuKn83+<bnGSU#BX=c}8
z0@Jp7=Dr9CA_zTM)@-LeBvoXZG((w&H6^5VD@5my<4N_uX1s~}wG7J=P6p)zOW-*?
zYJl!sZ_^=Y+a@CSm1h`C;W&_|0uL5s<2TnpN$s!ItM+|70_2DWfJc^TQY)4mlOBA2
zrvi~a|Aprr5QSxlLV2Z8SM-;tUPO$}fNN-hM~k84Y<LQQ{%0Cf+*W^*Put5Sjo%iM
zjzry^AKudA!W@yd2G^H_+YFeEFl@iS(>&Wt)0BQ3fvttw!7xoVyhPt}ZOeRo8g!Ir
zRQPc<`AK;FZ2ny6f<Mbr6Qu}89zT+RS@NWQN36lz*huIZ(EZjgYDbk4$^Z;k>*^fo
z*vECBKrc{_w3MRa3RjBhZlV17Rb=>xD!X1pHgWNyO%kRQ3SnJ$XDz@T^&c4u>htzL
zu>_gC8h%zb-N4%k$KcoUYW*rw89Ii~F->kAQI9B*oL2gMO?Tr}X?b`~j!Q1DJhONu
zxcmWvg#zCHNEgSe4J*P+B6dfs$C!q@Q+~P%ni~J;wxVVz&nl>3h>hiD(__tOXq4bf
z6lT#V+uQS(T5b5pyuZP`D{q!<)25$njf@df%c8Q)FDg*X@Hv&c*hQ!b65Zr%2TA{w
zr&-C9&CAb8F8Y>sX(z3*)83-TY5GsS8C~VK#$&uPH@CNnW7Aw^9uZlJHL-nacxdY}
zJBc8Ry^xH-_;6f!1%rlu)TjJuiE{?p75$?o*+0>(&Wu;ICtL)_+zGcKkP^#UJ$LVV
zX+BWZ^jv1NbG@}zV|**Y=TH1-E67%<U$nR?-0ePNGto^9@n>kH<0}>uS@v1!G!G8@
z{P>8q6pk`kI*O2&`1SYKTzB_Qv-;o(v8V5J?=kPk-u#-dfQ>5=-KTQQvYn6iBqSs5
zjQrv;_wC%BwGBQE=wJc8U~wtlM}!%<MNr^9<PkhzaY?YcN~uQ^*(2}`wJ8__mbGGq
zKk^39qWEGH^mfL>W)e6tn-r{UlAti-m4ps}8ObNvO!g)ZuP9~0{%Vp3g&?j=eWLnG
z76J%wBa3r<T8|kTTo;?ey{qydte=G1KodJB7Zff}GbueLUV*ZJ0UVz3-`>3*N+_}A
zp`US3?wr}2U{v8d?3wmb_UBQgh}2x4TyOXROFxCmO>^~`n0T*0GdCaC<Ia%YFepj;
z%OkO|b?o;FQ9ZndyYN$zHXeU$hb~m=5~$;fm{5x$ak-p-6h|uV((BmfZ|&y-_efFv
zPpcb?J%Q%*;%C}ocnN!_ZG7$ikEwBSpX0;0!<Y#v$&Tp`2wK6@+$FY0TUvI8>uU-y
z*2wkxJ$*4;ky=?IC;Vcj^+pR7I%J=@^@fGt4z=G^eF2m0p#`CJ@{F;O+$m1HO1$Ox
z6N6?3Bh%8g+C0vX+GQ8JR)Qb$UXFHYwlhZBa6*<+pc00|k=&BgB`4mBMZi&OiM@kC
z<w&a8Xc11Tuq|hHN@=7Hz=Z&HA(LmtA)YH}QcoC*kSU+xa~-pkqQatm#c?=sMwd;F
z^^S}x$mXlGdA1ZF0@rv-(jk5t$()0?a8^OhUXiW_rGclR#2n&e=iLWvD|Qxa)v2%K
zLo!buP31wM5{atP9PkzH<p$9zD148kRZ-BA)YmCeK#-iVCyN}VFcC#sQb@@wa{Y>k
zsrL9P_CU6Gjeq?J^COD}?&bPj^NQ`gSdwyNCW~7G&l+6HM5M+vxE(%SL_0P!r4Bl6
zvZ=3{&Mq!1QU3I$j`=oKKMT#}?{d^Px9+va#?2U3z5?5`vyC4Ak7}%{Fr)qTwa*6i
zpP=F2)kljBb;|aerS)pt=sSA9kY7FS)XbfM=k;VR$$jUpLUjGx_w(fz=l(ZI=L39f
zEl-n*jH$G%f8-G+sdmgX65{mRbzlCtGV%Rg4RwZu)U7^uzze>3l$MdPR=o}eBKEB`
z&$|`6MO^M-u!#tl+tPjmM0`CT$Pth*#oV8lxP+!3dglwxi(Or=C3BHZc_j<a)^n}o
zQ8BL9pIol0u62n>wQ-FAMgXt2m+qXB+z9B*+l()t=9O4k*Em7sTw~XUXv?C2io0Mj
zxUe_ZIbuQD9y|$oEP^y?q;oUuY=!7ae2Db{gV*n}r{;l$&9DP!TPU%;O3z@lfkvvD
zjWzV{loCd;g4Pz{rpSKLJ!Z%2vuCNoSEt5xJDRvUzE6lrUH6SQx{+|0+oaLm=_ubQ
z*&{eh`5L<;9U*yn>eA}JlmAFEJW4gpo-<mwcEq0pLGYJr0{$8HLz{n3#O^~}grEjG
z4;EwARL3FEvPgT+0Sw~&v@mQ2-_B3Kw8Hxsqo>uTwJy*6D<P?`)15TDE`5ZiK_iQ%
zA$|TG(EG%j%(z8=EH(6wYVd*Ch-RE8M+O0Ax_Ip5ebXsBJ$;smYBn8ZmoGh8i`DDy
zLG@;}dUdT$R=>p7iS$^Yk@XR@;rS;uKQp@PTyH_x>@x{wQHmx;=Hb*EjFpU=hqb6>
z!lx5$AQssY$6!bFBCETdpH1@>&FBedrJ11CwV5WCY~CCi6`WDw`wD~f!kPF9n?8pY
zi%z3d0vvkULA!-xbmPToZxTy*1(Rgj7DUJFlG6r0c^PV4iPYV+ZEQt4O4@`*{#CzB
zt9`PI3J$N<9swx`9c^Q?oVKERyhmLF(sObl54x&>!R!<qW=T_J$1i;gg(j3gqza(a
z62?MzcmXS>kH$BZJ70-whM8W0E0w#?P6`x=rS;5i)@<TO=40$yGsr<s__AZ0>k%A0
z;_DCH9dZ?-@iO`LTR@XGi!AqYL~inNXZE*nkpJLEcCi`4pWTiX)+K^W0S#STTpsq0
z7mZ9R{GJ|$UQT{@jZKe?m$O65alfoP|7DH+bP~TFw&n+QaCp;}2O)oUQ!lyz<7%LP
zf-&5vAg@8}o3QZ(!ojZXu`8sm?a3<62RxCp@;4u09nlp_Gd9|~`sfP$Go#77$VDGK
z3NEZ=CGkDu)u5scw>l?hm8WC%b#J{Gud}78Xnsx!XTJWfs)($aH&fKb+gtu_<eRgW
zIcGSIh`e?S)o;LcFdO8(PQ{;awAhz@%}YeZuJt@bR#v^;atCGiZ+<~CGPtLF>4#$Y
zifjRn0yUlDNCd~BzjirRxW`Mi#>lNnP&D~p4d>mOU}aeZI?+N;yYgixU1q|?IS9O@
zLjmbV5;ipZENn+9Y3zd`WAdPTusirzx0*lTA%4V)H0LuEz9|*;(vOIyXR14ADRE9G
zA4Yt|$6EY-ZLgfa<<%JHkIlBG&7V`2W6ezN%JgK)aAJjR1+%xZ=q5d53#6HLO*fY)
zfVrE(KGRQ@zaJ@}synFtaeBCrC6xY1%TlMBk!<vUD5P+*#PgGT#8&x{RsAEF&~;+m
zZScR&C0)b1-7T_WcdtKf#+U{$oWxy>FBq&xWBmlgHE8@GSv>1m*??oM``KgEk1l>&
z8L(ADba%Q19J-fe?i2MnsoG~pj5lttg2iJBoUv=(ye)Tqj1znJ?0mZV?B?bLXZ=e(
zEyuiWg3i5+0bT{w8h@#&QMW6dPU}YjlP6wpM8Q14sz9Ab|HufkpNZxUyo`qo#2&OV
zJ-?R7#z;9Ld`u7LRkHW8#j~%=X1{_CxlNbWiI25RBl=#cnRN^NZJ(2BP5kM{!Q6Qm
z-?d{Mv$YmJGr{7pG>0imi^=%%96FSrjyfK^_3nPk_i!)?XddM3sqmX;L9nh4X18V{
zx@z8Qu<oDJM+iHj>*rWPTap@Nnw_)Os#=D%lceC&G&Fii&`fP5S4j~i%p=TxuyHU1
zDi{e8#{^Vn7N1CMe>42Z(+pFTrmPgt#Hd6+XJ!7bpkiwvIP2jhOf&W+wO?)K&`CDk
zrZm)NW;-kYrU-@jWGTe=F|V-e;R>Fu%XrF(r%)kQ(J=m)KgdlnhXNg;Oh#TsL!Mkz
zaQ9J8&G*H3)T@Lj!SkvD$pQ7-9H?MeK~rv)S)aLjX#tQZKus)v9880};2OTx&%1^h
z&LPC>M6hPxbH07GAWNZiZbVtg5(zNkTgtmN>xD=^I<h|iBz^<ryW6_3B#I9B{t+o|
z-aTQX$wdK{b_N!W0e1i9NRa>|^=?U6ys&ZDWU8U~F#E!8>YC6Cu4ldQjQDQA#dL6e
z8Tk^4Q=7Fw*WMUg*T#Ig>+42Nyx+|O6()I(2hd*;q}`BRP%*fpAEoS(g<DEHL=vWs
zQq=p6cM4_ga1fRN&`!~@g7jIK-Cc-yu~mX%U;<dp68*~6V0j~CWhM1mh|&Jn$PUr=
z35zx;S2`ePX^z#2d?5^}$)o>G&Hv7~Sq_cBZ_^*SnoTj1YV%G@l0TfK2&#e4Ap^(h
zLKa9?;<&Ux$gE>Jd-Qom8dqMx4Y@bUnKjuO9NGY5P4+FKUxG>Eb7>}g^#>x@=Dy-R
zZ=|)8ns4QK)Sj8EA|SLP?vXp+0YkB!ZhSdQt&M-`N;0m$f%6K?F-=ZSOs6l0g?^if
z9Cu6}VF(Vaf|NsEXH1#E=F9a{u<^TMs1l$eJee@9aN#kAF&M^+xW;4&^dR3;$N;Aq
z%k%N`UPi<w&SDDE)i&1gE>CSH#Tu~+{Oz2}bc%%@c9;~E)e$`Gp9q?)Zf_W4R8VE%
zn!}1>VdH&;E=5Dc37mqM%gNdCQbxeIU!cx1*L-9xkRdIgd#@+K^qO><rT4nu{FJD4
zpvN_L5WcYt|FyFL7x-*)0CA?rIscAeanZI4>zz`L2aO>cAYuv}!Xc7~a3?ct+nvu`
z{9sU`$KnLaiz&7gsz$EUgq8ROvi)6c9cy*I6DMe;S6$!LEBfuZG}d_;Pt4!aNh+q{
z6#cT|6fJQP(isCiPDD>g)*A@45-1<{bl{jkc@e}%A0GoSjVcOH(Ps3^sHw17#&*&5
zH6q_UetM5>|5g+C|50$pTcNA;ue6}TV}^Uz-+dym#jH`^yk^G_dUU2GeOLBApwu(5
z)gD_i)mp%QPr`6<uE9BumxNISscz7>(HnFJ<4?*pXcy1h(`j_mw4jJwCvC1w7-^vb
zap_rs8e;Kolc1hzrO`nCHac8Co+Oe_lr(s-qN$B<JX>qBUW0+b=xpm=g>jrAfiiOA
zJLoG&qcS6ovvFFTu=v;4Y3s5(CJhsm-u3p8+N$5;CiVYur_uZ*HT?8Fd|vYUz}@Hu
zPm-acaZY6@uyb$X__LR;XM^T=BK>$GX<<t9=Eyb5v7k*ovD}bhcg<>*1VAel$KI7t
zAiDG@w&9^NAuH#{Wj?Z1mt8yNKLJ9<i?+D7?JrhRKs4#JtsoJaS%T#zvW>X$1Tin)
z#m->!A8YmX#O$O2jqcAp@ss2ULOq+2s3tBj)Rv{~=50I+N)_O-0!_vE1)}m4e%{;I
z$6LCL48jTmPZ!oTH_yj(W~GlRcSg$tm_@mYPUy;Zc*HR~0$Vmsym$`QcXiM*&`7iy
zPoz-S;p)-=sDS9!79-*4#U0VZ9wqfE4!cKsY{29~`;vOUmX4&x1C~sJVgD)23u*I3
zp!e2C1#3(Ng-9o@WE=2>j*zntUSc_S7qrUW1syJND2d36R8;>H(Chb?MUWKu=H6Ld
zgYz4&7dd{`>oooLDAhR#-dABpKUpVitoQ3lc(cpj+-pyH1J|(Ma7c$W2A+PG>0@y|
z!hUyY1te5ay-0ybc^xB^Pwv8j7`1?(qWvjYA`l^NjtkSJP+4HN3*tp!(CZ(J!*Hqa
z+YK@9CzGJJl3lmXM;>}G1X3nR2-Z+c*(FveXrEBcL4;O1$+?ZeYhL^r;EGv2^B%b9
zdIA~}aFyHZJLXTuV5IIx`b3SC^@BfiqoUW~+cZL@ISrbUwr5bwR;z(#8NPL^xn?7+
ze$-P>5}%%sRB=Fh3d9kzR$p_T#x(~*n{4D=8XI_(@Ta>uHJCD`2W4O_3IFLe!G*}4
z7>wq+-dFL!c~QS~UY{nov7+!a_v8ZtIbTx<w$T>(7RL$sDi#bu1gQw16-`?*zqM3r
z<b@4ODH8Qx-9HH`U-c$9C#^`fVayz-!1b;&RDYUdRaWavd@9mGF56_R6_`9r_bH8e
z3HPPO%p_epdU{pnM&YTT68oh&M`gYDKvQ3JtBPwT%~`W71aoYZ_*tXL>R;7_bH-Ad
zMcgEUkbFTlS|5q8?1MA*D*9T%m0F{arba`!bXpblUj_D#tr+Z7?HN2B5Bq&V_JXUo
zy}#9A@Y%&v8A!>%B=Pq4N9dYJsdGXkJStKk8~Go}dji`dk+dUID|vXzgI`uE&A+=Q
zdqkX(LeLv{eqzM-hF5&>Z*?bFnwsgwT_raO9m`JavV{ql_&-W_(rK3O|3e*^1_$?K
zW+=7lxCM0#bpGv`G4t;IpbpSCPA}ky2ki;2qfOr)KlGfZSdAYZURmyUcKwsV?_YwS
zW+^-ag=I?d7&*%v21mYG)}PoNMK{?qTnWqnz?l<&!ep8udT3aO^S|8|=j>N{cc0+5
zYJZaX$D!8{5lP7BzVELvgAQep26g8g2V4}^Dh$p=xl^CNG42}jcovkJzrgU7&#0+)
z?d75cEBTq(gIJDlNf_y@b7pxB!ISG3w~MA%Pq(O`)AVJh(V<E5%lcg5>-fmDtSr(L
z+O^VFXFzL}b^UV9-^WxPKtnBewsnWIe0}Wpcf`P(k&jCIiKFBTZP`U?m-<)Q{zK2T
z&H;FQ+sTyybwP^|eXGS?ADE3k4tK!K=@WOES&YNdQV2Gbpjzf?>!2v*-&HYpu2vp|
z9>|khb67<`D*Zwk<*`y4stwBEbfe=!8~~(}fWwG_2!7buVCc%`$N$M&+h+gp+1l*;
zrRLVe1iI^5`=(rOCDCMJ1a8!N77&8-+D@|4vT9Ak7x<RH(%{OriGE!i$FBmHK1S^n
z_=xVg1oj!yeaZX**&{YT8(p&VnKUd;<LlNb!%WDlyz9C@?!_#V&~rT4cVJ#v^}g=}
z<+HZUzC6^}acd%Z{XR693{|xC7iUr5fLmaSS3MAvnMDGza68J?m;NDKEIPw)7&>T$
z`@+eTSXwp`_Vn;d2b@+d#TQ|??G0TD3b7qKnfvo5xQ0ek=Xl9YZ1XKnHyv)lUoT^o
z=wGeCb9ipN@~Jru<`Phh8tAA&n$mVy9A097A9pNaVwxG2M<`W+6dCat2zID1aB*lg
z<7!@pDHv!;+&*N6d%0Wu9Gf+<P<^P`Yj=X2TXLDNyyt1xKtZqX&@!?=N0sLdf%@b-
zhk%Ge24xI+45;SKOQ5+0z8V>cKOxO%)$0IeoD%TYIC}N^oCGBx2<0r!hL-rH%BeeR
z3r-!zS9`0D3%4Y{NxwUddWhV!(BqbQ^jmb~I|reO6>Ca^^1Ki<ry5yS8sZ1(c1y*P
zMjQkXD=?Q#j*Xo(#AD>)vhNZ}8LCBp`0le7$uNa?tf$}lD}X*J*KsH^VltOw+k-fW
z)%XK$ux5!XPKWE#0@DfsV_P2I#4XV09BWt(NJ`{#k{QMlOY9E?RrK+8T5JUpc$}II
zK~2Rj9~6z1eK_w^(CiWT$~=Cx+GM5?g%QQe*1k2qb-BP>?u%R|5w;$I%<p_D&PC9p
z^%!Ik)ub~pPCYRO6a14*<)De|LuXyg0JI`R%47Rlk!okc)mTZz*OV?K&!{2r>G417
zDI~7h^bX!Wm`z^{$+`0>)<E<1SJSwBrFQ$bf*?-mf)E^b8NW*<<ue966&jrywE6oB
z%^BHe_gOT;4d)P;`6C&_@rAi+kuNK^Yxm{lgQ+&>tg-{KWDxGpLC}pM>ei|BwwS(8
zH^>wrH}T9$45O*cDt+Nc#qm{K65(D*%_YYUE3Gyk#Rdda9v975Ew+DK-nI33t@gN_
z^!~F<P||R?4uw#1+pcn&$}XefK_zG^MARb44s}y<|KI9D%&EWcIF{3ZI%8jO>llJs
z&6nbTx6c<ZZ6{r4W$WPJ<HXh>o-V$dJ1WA+dOOGrw<9k^U}3IElDaafA?l(c%Jv8$
zd}OTzXO|!h2j^R{y~9WT@Cfo880nLvwciw>E_6<%3=Ll`@wE#Obo$Ppw9PSH=;KDY
zk#GeMEiV*PLqJ?*pRmX>#2?>x7i9YU?yabWdrrzAoVO~a=aaqP?jA&OCFiSMI8A2Y
z7=H;=L)1GJ8EB3nt&q@gv_ED2^sDPUQKQye(04v7m%w@T#<A4pgJ@nHrsW~pi+7St
zk5Kdw*YHUkm|^c)FzSrlHJq9tIf)MW(y@0LG5NTx7Mzs(zDA+>{+!-k@l(YXZ`oXZ
zU&vsgCEIO}^gPA#jQ_jHm-tQo`cF_o#6Qi6x8~rRSO!Ica)%ncpPB5jBT?9_R*ieI
zE$$bTkGdexLi_M@IXrl<0HJA9>-WI0CQ$3LN09e77m^B$qv@HX!q!V>qa5dmz64CP
zT!dBMEJAx2nfzf-xdP_QABgtGh$^S`#CTDYCRZ0|IGcTi|6$|1oh9g}=%#)o6ES>t
zHK;KOSU5XEXlg8}Sl}K^2IDUfTTwIqChmHB(xp|91by-<*?l<p>HPb^#h~+T4`9;a
zA66n&6n?`rKs{WaMa^cfl29*t(V<AEz;viQVKJKziWD9j;T!?BS=IhEYc#oK3jT$n
zjO1L<x5QOKGsc;!m0ALiNiLcPN_WbPZmnA8MyD<{I(CV`1ywL%{efZ}k;BK#6}AR?
zjeBu!-#nz9dPJRa<4C0F&RwTV7a`G)Xvyj{G?1Bz`%D>0Fp-RUDFv*gS`qJB?Sxs*
z))=t4^Cy*qK$)4*CH9@3iKOWt$mzOfq;8W0B>~!LFh*fTZ=$`_#E^@@fx(^&)3lN=
zu4%RIFyP3o)~|G2DY2|~es~&d4!q2Cb(bI9#@_abU08{zFe!`4mZ!+34nOO%bqqJ}
zx<$)8A+$;YQ{R#tGhX(X!?%f-b&&YbpKnH9P!XR&0o{9uIFZ+P57@`qFO1QQ-^WE%
zx(b`S${%DzT>>?n|M<6jL4pNOc>_8hwsZqu=Uvs28z!QgOfJh!JBzFMNGT@xMFh5P
z4spR?Req4_FYqTM{P_YFm2$>doPEkk#lb>izYl-~c-3!S9fg;dJ7)19*^X#_<4rI3
zVe`8`f;W9>V-pbdAD&GT-J4Yl30Fo{*Iyr$HLSpvrw0NfsBI{a+w^V>$QvN{VF&k(
z&zt!^e|lb{S=u`m$;Q|3Ys;`VPF96b&|w$chiQFr#{6h}^4)Z(QsEYPoRpOy%=RDw
z%?5@;tR53=b1cpkj;Kr6jwUwVLR2OlIyglXkdCLHchBuDPpcy$GdI>!U%o1kKRqTa
zJx2MUhN1W_E2-Vt-0TzqLo}r^4(7*Egal;G@Qzpf1;Vh-@^u-@w)||xta;o25YlS*
z#jdyY=JgiMVpJ5$t9i&4&YPpZq|qeGCO+g_4X@&+vyHb|K)X7c_3YL-s4ms2rJrcX
zrcj?5ErZbQmE;DW|MYW>p2N{GAv3FkNmr}naQyZxuITfP9NzDB#1lzl%%5csM!o2w
z6b<*_Hh%UmR8y_>HuM)!j|XYJ@X)f6jhIes?Pzgn&$&^*tCy$C&D(GDJm|;ov2pxs
zT5#KOH~qsU1`oq6*C9`=2-D{R(TOFHI)kdeEiFBQN~wye{q=5&vbRh|#*OhP5;=4`
zOoj7g28OL**oM~Vx(0^buO7b|$zOzE_3UKSC-&B+3+xdZ<5E0|d_GN}zN?o<ZLhXm
zhSD(`tMZegE27k)jcW8ZnbaFp%g;nIkLgttmP$_nc~%62uXmgI7b$Vp@5>u)KG=Of
z$yBJ`<P5rRiW*FG3S6z=86cUn;q07a5mcdJ#XJKYLtGspg0wb5SZ@+E9lU)yPB!40
z6BWY+hRWD2m7NH(l|}Vgb8wqOlQB(EKu~md^ZU?v34MGShqVE_>KOas1!WqXEu5-Q
z1+Vf@4FQ#CUf3zg-e__OIz9bof-$-e*45g+2G^^!A9Ca1c;E#|YNjywP$!)E0BK9=
zJ1B)BJPgHG3F60yrBnRo?BK0FtOZJ0lUT3N6z)c-@>0hMkF|^WjQy9xt{2ijIOHOt
zCGws*!v08lGqeM6tz>dYNhg1t9lHfK(Z>pz@SeUtm=B^A{qBRGk5YyXrnwx9V=Wz!
zri>7lnu~dtJaIOVG9!E^FEbAkI^Ab^WmT^Updi4p^*5w@N1_gGE@fTv>1tj`rB%@D
zhL<(Wvd89M$b`?xaDBWQGj(@=iyqjD6;{))Cg88nXYy1{Qe(V|Ps$as;{3%~hEj)j
zbPJQhkaaaKt(_qMCD*%`<e)Y2)<5Bt8*G=QQDiTXlKvB1AT7zuI&3u>*lAJUYV$!O
zZC&A4n*>jMx23qgxb&2*x1=@S#$Y_GsOPVz@hNB>4YNgGUCnbrVs>&fO>*YeYLnT(
zA>%6WcebNgrclx|j-mOZJ1`^~h^Z&`5QxNyITV(LgUh@Xkj@dW5;?GvUZP*OY7}AT
z&jlYdfDCxm3?Iwu=Z}!dTU8aHErtU~I#VY1x_f&M1P1rn)SLTr!%Yg=XnYOQ@jyDj
z(5~;>r>G0O1XVTH_LZw*>0wWy5M*6qWJXjF2+jX(L%(_h6*s%%xp!a%1o`W_`x4YS
z8XDEarHsO&h+!ebuitBHpb@F<i|ED#f-3`r*1HzYpt2X%hDCB{qH;Nnk2^jU5*F8D
z{Fo!kOm_S>A3oo1<5#I~EFpw=w3@(^m&wu^BnKf5zC0j&dD-t8yL1|K5~NeRKKJ`e
zX1FgcT`X-{5UPIeKZvwL{h8E*b_nrX5D%P{By|*eKm_J(onruY|7-T@$rLl?TD|R*
z5$3voPs`%c-?t3Wnb3=7X%y_NkQe@?Kd!DZO3RgZYrdWK0D?FH!u|853d2^HsYy{u
ztc70ScCs8sF*C8~gjvIDBD2e8`>I(wT~@Y&x+r_id=INwSREc-lsLlqjp()fN7iSj
z`<7=<n)$=BgR%n97ISc;3ZJ#(CSp1RG)z^U_Cn8`h<5KNndVU`Y%y^9r5b@77v{Xp
zFa+~M{#+yTNjJ5K;TJV7#U~Y{uv9_*xg{)rrR076>IFQ-)Zcz;2S}Tq3#g2*!uZMf
z@Ty2Nv{aLt@n9_TRSnIQ66E^B4~fLeZ)7<_E=$C>*N(e*SP>^*K0$2FLCO#?Lra5<
zjnmMCCu24n@o%HTRw^SdG4nAi^D&%Tg)x@Ykvn@Km=xYt$&$whd3ar0l>IsqT96c7
z6T6oZNjzfe6$-z1<KW_K50FKauSNVxRk8!fOvo}eknZEU7kIr3Kh^?eeh5O9wFcYf
z3Ky1Be76-~Vr?+58KIOQrfR?JNMKgA!bivy5{kHNBh<!x33Qk>bCjs5_<CaO_O00$
zi{sx<7cL-Er0Y^x_+~U=(yG}0-F=Or5pag`^<3li1b+4J100Irzb~U$!vu21)qh)k
z$8z5vKXjV6SlXz<L!?4{P5zwVJ7jx{5l)%(A}xuT+7j#FFW+s6?b9_?2jk^F^vod=
z*G${^<JFObyTP$<igrRD3RJPZU=}`+v);Ukpn!;|dB?KmBD!}_-<Yl=K{={P$5X0Y
zlGxT^y&B^vhpO+7h(0PQ?hx!O;=qu8-p9Ag0K%mu*U6vz-h`=ptV0>Av3x`_u;RXU
zUmm5PVz6MhPhf9L&gY;@QS;VGR3H^gPF2(XutY_p@>bFH2$hamHD9$Tjm{K=4$*UE
zmcCjz@MDkA?(-Ej+4GBqjQQ6O>ct_U;=6O*n6NeHyw+5n``;Nt@s*nNY$04?`8-#&
z)kmDe^n}dfd{Mc*oH)2dlsvieI;*4u4~aX1qE-E&C5uHv%fEK3mKr+^oilJSk^Amv
zh%RhDi(5kh1$~);E9J|ryuwK+Xi1@9|EWc9a?%Ep3OhRziF89{*hNA|mo@!`^U!-M
zc8AR9a-Z{u*^o!J$C!{8IbmP7!NZ6z<l~)~OX%bl!&C5$!A=K#kB5s?aa_<#@G8k%
zO<fhJmOUh}l2KtNKqR^nY1BBI_K1^xhBZpDcs@cAM2tY4pny!{gG%z3=`Upl4~Czy
zX{dZX@iA%dXEPqlpJKjpwR5FU0Q&~!;nxG7c?dHc(?2UgLPx3WJ61DmYucmBE|dW$
zy>K#ER2R$;yPb$N(cr9<uD|9TXH(gjmZ>GlpJJsMc{6Qn{qnwk<8|wgCkP)%FmHsa
zhjvJMWqBa_`hbSdAn(yP@dT<v%GV74b>5_J4j6No)pX>Kb-(7eLzHAp75+`Ci)SB(
z^(U#2Xq@z_{lHMkk)pE4b}EI|{%H28m{DJYooR@F{{xR{6_Q7D3)?Up(Iw@Zyc3ec
zX;m0B&zDUI4dXfD2WdXd(u}r$K*r@uq1wnbwt_)&Dcm7k)tbLtVO8G|sP1F_h4jZK
zQD;jOZN0HnIGo-6eUE8(clVLUT>(u=?ew(~h2|b3_*AV^WE`=EK&%=iEdniubZKka
zGVxdvi3!!H!&79^wa{6?1P&HO>r&dDRQ1Kg4~km2YqOlq^!n_SBwnqt)_~RT`YH9e
z{9}f}hBK4W`X)5N9It&ln0k&fGRF4&PBNp7AtNZHus@1x54jO7K_7TFgZ<dI+RY5p
zws{}Z1e9Xuhl7q8AWHi8EZAyi$PZOFobAAwX7`d?f^6WpatiWc>)M%oJ2Ac7+9gc0
zmQ;vsOr^(E;|Gyx=+I4VX3?djw%(OpReBOte#|h`BVuAn&UhQEe?uDOlwgzrJ<aD;
zIW8RYRB}+VY#)C0(5hQ@5)cy`LY9Cu%y}os!}<Ne%Knh=J%U>HwqEGiOalk_Oy5Ac
zZzua%B5z-5Ka1!M<Y_5trOL?;D2)i=`#3@RIIX^&*~}I=<a7iRclv1DUy~nMBa-{9
zla;Nm>tdj{T2>2UXUiTSmSLGG%gs71Iqii~DKNP)N@^h;sw*hk0wNVlW4W}b^~O!j
zTUUmNo!WxBHx9Hd27&*^tCw8@yC<n9I-pxoJ^LSpOV3@$UhiX^+4{j1c^LyRENc+6
zDEBvG_klBzWytq}G)X##p?^`W0#q)LApvoNiYdF<TKffl$LS`T&-nT}eo`LL*?2^P
zIH~udacpFHuS*N?f6o!hd7%UWORc=2tw<-S{{jpI_lcWTqcg&QrwN;J4WHB;P7nw(
z)J<Wp&t2#%dDeq(@!#(z-{R?$$S+SeammNQozJvLVa`$^v2ar-vQGiA?s;UaHss&r
z0SET`d0_Viv*JF5Z|K0-6r&7LJn;~HMm(m3yVAEdvdp~;mS)qU8z}m<`bkbIpf<;X
z1r!Y)GETll^fKj7@|%$O1&6-Y8Hc{zTlB7{`D)+GEkOF`bN;Xkv-ZxF5$!^z8u1>g
zZe*!N{f4`Z$xVqa87i&lQIteg#VgIJCYDOnQHSznZct`1AzmtJ8C?2c`F%$>VtMiI
z6%Gh~BSOz?Q61oVF9%adTw!2W9Yj{bCC??V`7=JR7JY>TEm|&V<H2jR>6l$*Y+Cwq
z<IpqFM((3fl~k;&pqxaxVy2(`5EE|_u)DoXtZ1oBiYDzg+gEr;hR0I)F0t0Oy}AEj
zbz3xk^sD)8i3QFh+LAd%Rtj5RzB&Wq8$TvIHF!2pZm4Fbv5SfPQmOJxX}FzJrW$xC
z6`14%s)#7m{@Bn+??nisZbD}xJ3GSi=b8doC)fQTgBrWgEy-TS$FKqbIc@Ersq2pl
zm>DJI%Y)TEqh<*M@;9dGL9sM8`{Y#a>4-U^i0@Rjb@)y?WTqjvvDa~YAJh3+rz@01
z0fx-;H9xCc_{D5HOgwGWQvRsvoSq&^#TP*TOrciKzR?3`<d{DOo7qNWrvO}-|3cj$
zm~B8Uk->0bTTHI*A_I2b4EHHq6osnUrgejCX9=NYp;~4*xW^N`y}{E>EZ!Zv>#qZB
z4+C9KA?`u97>wxln#qa3mFeLzRg*r3g=kgmS`&9I2K+lL+SaC`Xv2o&<i1K?-AxGS
z#=#ekL|*!yY@<g7RO&cVRKewbSrVkcYN=q#`og`MUXMuM9&`=9&&6TpQ%cp!ZyThK
z1$*NBnws~lVnYil26P5rvTSd2<%kA>FE1IBv9?JR@YbwWKRQ0D-6oQ`<`%4&%)>WH
zQvI1WMJ>R<zyhLM=fujIpsQIkmlg!~=h9likmUGqrMHr?YiV<Nx7ALO!-uHs*Zg*e
zV7Wj_CbBh4@lJKCb2YI4j_+z3l;CMH?{t$`2@uWKiiBJF+*(zf)&o6s$08m-dMg?*
zO@-^9k!>DgB)niy-RR{$fuXz_;I_&V`d|nJeDx+4hRrAG{IV@!R){MPcg<Uh+;JF*
z?gT;)B7!Xmxu580PWB`y@2G8b7sqM0Fh&x=#%UDV*AG<<aFru7#s2=YhKc_@+nfBS
z*I<A^6QGnu9@XVkp9m5h0$?mAQX68{{k+y5G&Yt|?e}(KSyuL-y2o9WzzX%-FkQ-r
zJ2~M<*OJJ8z@R>)CO()IJYDWx$k%ZJ?Wvti9f>7Gzc?k{WE;;?@Vhb^@-?Ke=s?$i
z?HRIagbu@}G0>l=ov@Op9uX`r7bb$jk1<qYa{YArxvF~cfR&Q_*m8_Y;#2ajn-`k>
znf}&kQ=d>>kIiX^S>vtc!Bd$`8K5XN6hl>>>Y`=^89`=~c(pa!1{H`)PJch?AA~+Z
zI*y_?4{a5iw2n%}?=kyy%+QFMnq{4gk8%v>Cu$+cLZnuL#Aw%GZbLRy|3y`8l1Q7T
ztj<C!;m;_x%wdml0nQ99x@B*Dliyw+nhcvwq&OX;ed?D|jz36gd2aG_)7(X61;goH
z#zRGA47je;JCXTZ(N)K8s_abk@$+9bI6q~`P*w_OI<n7!lZWF~?INe-^M}i`+Q&yc
z&JCO@T;sUWmxYj=zx$qVmzOKeXqji*2>$v=_QtMu5QM@GpUZV7%YDL6a<=Fz_7z?|
z-&Jr#zDm30`Y@o0YBZGesE4c6fNPh%0&fQZMVuA7EUn-5wiED+)t&efujA;h=Rpfz
zw7rndM|cV9>;dHM3D@4x4Axe{DOZ@<PHF$3CEf^;?tqPfAM8fS?2kndGG?E}q$AK@
zWTaY%U3A!2pEyaVSeUyA0*b+AfKw<{sH5vSl?2?Me%Vk)VXS2qX4a)e;v@f$PHiJ@
z{RQ7N@DdP9E#YGb@{e+hJ}y^EEcrq%8n{8PoRhEhq4Fu0VISXus%B}wQIC0+l@$;r
zsq4$?ClDnJrddlYdei<<(_=0cKaSqLifwhVYeSeN0e$%Hpi?P)T1&~yYH}`Z=5hW)
zlhyMH**<l)anUG;jS)mwZ4xrjZ2Be?{EqqdRZa8}S#wuh_C}n*SGxsKHaeTcS`V0y
zLu#%<*cJ?_EE6(h>#Y^Sw?7$&P`+Tb|1|4h?Wd%bqJV1?6S{%8d|#XQ*4=0LC1@U@
z^6iG#xV6~1$;<n+x=^vKkmCAe<h<sO>j#izh>zB*y(Npi#k{AfINPl#v4Yp9y}Xw~
zvV3X|q=ewi2YP|^)*WlUfC)+g1>*LYQw(L)FieDC%A3efYk=o`{fM$*R7dIy+BF?c
zep%$))Cpv^apKKzmyzF?u^ecxFX^xs&wR|`K~1c}uqm~tUAV74B4$aSuvz`j&gZ<>
z>CjGAMXR)XaQT<4?tykdcAWyr&8b1K^lZQH*alUAZ^dc1b2qnd0sC7=AxIKsv>=aR
zOtgM*`toC@KAy*<2C~3o{kVFxgtYa{!3rmZSjtZkJpjA6A@2S+(bLz-YPvUzEHK2*
zQ|LvwrT|a_>$5HocgKssmt<~d5i(@C)j2y04~|hJ$La~VPaIO%oM7fX#&@<9mih}x
zS4Xyj*P_AvSroh7BP4&YAujckb3qcKC%D~FFn=~k8#Rdz_*{r%xA0c1>#M_x>EwnA
zfnQ<?&bko?k+rlK3cB}ld4E?J@l8$6h~-uqzK)%?8~5e35&kvPKt`aDx+fg`kalUM
z_>0iwRUgc7mBl;}$f}z!Up%>gG>i4*W%%tz;x8StqD>Je-4B6ZS;@(pQAiOI_15sJ
zDdjMh+;Y2Viukmg5^ag#4>_sGzy=?mPC&*C5}^{jq5DI932SEXd^fg^nK(gKOIrrS
zloZx*9`kAEY~G*Vs!?Uap%KaC!o>NXH;eJ&3@;1D>=5*8{%%BQcJe1Ug+|IoL<?eu
z(=XOx8Cy|EXaE8hgGf7eFZo|#x6yakj@33^tk9&A{@cO(a4SVwrd5C2f7LrLSQSg<
zVX;aGejFb-zJ)}1DPZ@@2zJ=67nA|{H4t4z?7OP(PB1-%x(Qg%_~^Xtd?`XfOUw2*
zxO!;oLm4aUcoqOmF`+Q@l(zsVDxXE@gNnO$?coqYT@OdYR^Y(yV{_p3gsboU*$QAX
z2mCYM)>lnmWp=LJ|B?quuhIeMhlhtp@8u=O>|xp^!`v6ax_@$uY@fcUs;ah6gr_Md
zB_4iYt;a~9sI4E@5V1;?=25FPhv~<W6HfA3Z%{$0+rs71Eue}Q{ey=a^^A0DFjo1p
zK6QxkgSLT>M}|X+r0*j{1->(F-t=&3aZ1QnI{;M?@GCJ0IC?frq_rF6ka+jxdkw)z
zwkg!hNyvcJrLTElcyYpK?K`vGDmTFEeJEG!y9r|ysgr?}&alrdSaDf~M~Wz$ntccy
zRY9ub|Kt1F!4XT7V+zKMMgQdcTS#}veeN4(OUq;Q`_B7I*VgezhQg$HWt|j#qj6Zh
z_&{Fo3(*Cn`KK1<ULtQw8A)m~uy@#{bD)`_Wr_kE5*hdx6gxur(1qU<`W(86`-z-f
z+yM{K)38Cs1t-S_mlK?p1xt_HNFi=l%CEKANg0xDS3Q@A)?uRPH<EG7E*w<Xq(wyf
zq)3gE1I+IY9j^t5Km(t(VC{z?a{sQgFhKKjCuWQ!-291i6I_x!B+M}kJ@{JG>M*id
z?7Jp>-}6V6O!0Bdk0vou?L)XxF5&vL&GCH%&JxWVN#piyt3VkME^B5I@cztRF>=P;
zd>u~~$P#ssO?bC*eo^IuSauRdqlwqQ{fvZY8MLN*wiWGDPgx~Fy`&6!6TmXh?z}!B
zl7VrHE?GW{gk`Iz#>b<Jmu3GYnR5Chikwpie%!7TE*1Z7_PzO5nOvnb2Nnerr?70!
zKrc5UAx{<N4<a>$$IBDAAtoD1yk)j4RYA*G^k?Hh@9N-^xPF^MTRSf`a{D{iN7bwR
z3Aadb*~#&$-GWnK>u`2|M_(ajH`o;GOO|dK;_Yu*D&$J*X|FJytHVo9MNUOc4^0$W
z8sTyKT5fJ9d3pJ|cz0fSW=mN-D+bz1yXi8i&k?uD30dLs%th<p_0Y)3Fq9;rfUt%q
z8ay_F*&Nxbe@00e`G{bRVNv0&&mg$4w+q#l<@2#s&dU|LSy|a|YR`$MK=>0$;n%6F
zu+QdVCsa<HO#CP}*X5>oGHRdOlBq{;^|7};u{X3WhcVcGP_Bzc#$f$Q-gtTeeb9y%
zkq9fRUK{^fkLUtTu7rgQ_Aw+QnEs%v5&Rq6I&eu^3bZmB#{AjTL5K}^Qf_pga10-F
zyq?W#K8!WWDTo?r<TIQdLkZsRKt;e8HCCLk@Eg%JeTy}HyPEX9ed5VXu<1C$Qf1GK
z_q#hmwr0gTNLCk=h|MY^arg4tfzUCRNFtS$^%C_3)|%B;wzReuFnA1ZBP;p|bulXm
zhzM+2@eVY69HA8eSMdvtEju+i@fwzubsJ1HvvtxbaFPr2+jN-J8L%keC}=0J^GY3q
z5%2IA6$L&X<h$^ghP(=>{Pw%Ju#DBr*NVoQVOt7%9)lv^vN?V}b^F!%C5R~6#_Gq`
z&v+4-;1`U(h5n|M`UKPp^!~RyWe2UoxElLG>_Y?qIRCoeWi~Fe^*YQ}h0uwE8!Q;x
zzJvMxCbo4l;+aU1@c4NC3hipzZpI1X$R_=M#8Oy`S@majZm1aIPqQjoRPOfr+~m9j
zx1dvIm%B4hEZX-=3iE0$IG4$5ok07>nN+G2t><!Tj3=58J1qNbAC&eLYml-|u<I5_
z!npnY0#QQ#Yt<7o*BT|H8XspfL<x^mgQ{xp9>2@aFO%{AaCFXbk-lvdZ`-(S-0U{D
z+2&^3ZN_HXHaAYTZQI^llWn_wpWpj`^Qk^F&)oMp*E!!yc8RM20!zj!XkB#gh;6}v
z@`@m!i=<=d<>;{urn5J&jSc){qY;RI6TfqPY~*uyEDv0Mxu&!E5A$RB^Khd*MecqU
zIsvr=shHSU1sZ~if3Jnr;)2xe8cDFjts;MM59fNcQwg?`!#b}IgDo$gss7q|C%fuR
z$2M|4Je<Dp|Gq+zxmfxE+Jsd^KEAC0+FR>94fSh{!9Rg`n)~emhVS*b@3_OU0?r7U
z6ViNQ&tKJm-?Vm_Cbf4bY9t-`LF}K@Ui=`l3fVpYf96cKo3)AmhcKQ%^_#0Vsd1Nw
z3j#Y7BL%b}w=xF*4;1*Qn5}K2-s|hs-Cfcw-@729_ao(^c^p+$#TYSOubU)A)p!c(
z@<~iH0#z<kovBp&pznpml$6B<1tR$)5)vYX9mB1KMv>JjV#2~9#K?c-`hJ`e{e%L`
z#xa^k^t<&EeX}Blrg?&KYH(s@ZD4iMsDKS}`#uI^Op97j01lF%A;MCmKzEl^jTyN`
zGiuT}Jz|}(CFZ-=U(?W_&*5?@2p5fviC&0mf)*BBY66mlN;{oK*WUY|{Wqqlgd~EM
z0^}?iNPb>&yF}VT*{~Ntpta;ob#&$;aj=k(t;r7W?}P4tN3spK1Z8UFPj9xBYl6-Q
z*SLNSw}{jujGsJ?WWwQ5wsQOVy4onYGFol!qqg{#VZXh9>~AzH{eAb?qSlSlxF8Q_
zRpd0Y$Ox%AXB?VgXB*!K*>*n%s8@F#ER?HPBwMZy+4|XTR?TiK!O;plpOH?<74Nn7
z9g6=cD4o_k+^9y#JG6}|%HD%!mW-EDv~BJHl;rI*h)i<i7BoMvT$Bher@gv3!%DEU
zv_y;P802^FG3d~0%$=#2L_VcODBvE}SgfO?^Ym9{C27)h97y}5s8>te0;V|Id_Ev-
zBARWxws$tjiM(%=33<Ijm|&=tJtTQO=wZX^jOYh4Wr_T`QI>PO{hU+M->T}6AMh!H
z3had%jV7id+h8r@GjRt5BA1R<7_r3cjL$0|ynf?XwnL!zI)&JD-cQ7Q+Hm8`LRO$d
zS**TcF0ig=cB>mE!7<w6A_^V<J?_$8bdHPMVEmQu+X)fkjtlEuCA-5m^{Xz!x<XG@
zin;Nb4Id9rkL%Hl1Kgeu9=nU@Evp*G8zyv}-o}ue!kT+8G88vXw&ug>4l=dR<Xukd
zDSqZvlN8z|mBEBuChiKpWzWzVuMmU+P0$dW26YI}EMf~T@(EJS)n3ESb)vY}&Ft{o
z$F$FsyzdJSukQnhQa<bIYSt(S9ouMj_2mteNg>;5G33O}7q-CeMHR5IEhig&C!=HR
zna6#-$omZ}vLb~b^nR?Vw2}dLCSzCt>pTbBxbY2Y8H+Y#qCNW(k}@jR_Ju&G)HZtB
zPBsbG{41nef<4TWResrKs7ujjraNhLzUYf1EuwV-)DAUATc=+!9iIn1cFlRjvw|$X
z9oVe}UvI;ZrC;?4=H*D_lm`%)a=1}N!#eR7uUx1GT?+ec_#`F?!aT#{6TuhNN96BP
zBWGs<=nfIN=@>44uI=6xoOxi7K2pV49#e7?9_#r7DzWx5ML0Pr?>ptNw!9|_IrS(c
zQS)gcen4`iadbcNHBL2eO=R2twjQ30i=A;;PoBqQjL72@3JsJ>y<0Qkt`_F1Wt?3y
ztz=rkH>G$eH^K?^pL3RzK&nC<V}2!SW%jFOwItf6U0d*d$X52&;km;xi)#TjFe03{
zlAMPk@k&muij95lAO9udT5FVHnDClvLGl5noWw+NOy)i;L_@`_6cOZDmk_Gkz@FOF
ztJknP<`&8O>d>=`{O`W`zB^o^btI>&&c<%hk1ZyLexTRhRU$SNZ!p(ejQ5~w7bz{}
z-65sG<CW($ukuCi<@Q6iOV`=5Bm>d9UDR^iwDDl0-IexmsbZ01&s|E{R+FwdsD_UU
z`c!rNegV|Hi@#Cnes_dk=oOYwWYNKz_2idea+Pn3+uds0gco(MXD>Tnqd^-$8r)pj
z+SPS%`^$HV<CmAnJUc$Yq=a(tB0%A&<27<&C6^@AJmAYZX)~xVEiFUcolQ+)_?6eV
zZ|)+-mIdTEm=+Vv0y0irD}Hqiv<59z|GIe<_y-&fyPIpVy4V!!1bi=<sU9t9zS)$g
zGoQBXayQi}pRYNMpT7&v&O&x|Z*Bwx4EyD6K2<=Q(Pmy8Ifc$<K7Ke}UHV)j$PvDS
zfG+kgua6&$xbsaUiVw+cV2`DbpShlFDh;vq-ruOV)55Dxh0P(PX_N~R`0X0%^_)H9
z6nE_KiI|sMy|Gpv4OT42dk<bkpvpH*!<(D}ZDEK)t=_G*S28=7?LK^5?NPt*)C}b`
z#xiUbs8{3Q5ilR6?e|-u=l;%WwC8DWGxsz>^dYrHDl5?E&bZpUR_1@5m4AMo{zoYA
zJ}U2ivj>k%_>`KABwmatyH=}1YmotjPpQWfc%?VRmp}D?Ww^UFfIF+XS>%MP#82g1
zL4fC3ba~=96DP6T{J?kcoW@-^v57O4heL%6^GR7I5I|oimA%Spdw}$}<4eTF;f)mK
zB*^|fbLf*gS~D(5GRk7>m5rqkom<jT+o}0#uODTmKJNs3@WRNhNPz5(;!8q|+-gJu
z2NP7w2xW_kY(o6Sj(}$-1%j&ta;|pk%zmoAm}P3p6qK2HwfFojOTZJ*zf6fMNyo-g
zdO$x>EO}3+4ks<lDVs8kEFvk2_R6N#+8Ft$PN?v^=usw&Nj27kjE+1JSnG)NI*Hh0
z>oC*a3muFUnCslwtjmtrSf!akK7J5VWLYDYPbL#dZK9P>`Zet-3Oi(r)`WISIN^zl
zDNP3D6G$4Ky&!me4h2%oEk`{mnYGb4t<Yjs=uMkJ;=Phv-p!MII%GR2)|Q;tFuJe)
zUxf>T%ktl-3&uK9MM;H?3l`E)S7t2-_u8!kkI~<axwPJAxL6%GNDD?=*7V=E@XBgj
z49|K(`bAIHe)ZJ~pd&2^2@;WU+L(q4*}SeIXC0#u4<PbNU50ieZ{k_n*}w1zavH~-
zEtW&yyF?cD8>CKSb$EKpEwM`M0}&s?dQBKR1ktowy+*R|JX*(u0zu1uCvJpK(e-+5
zy7Vbj`>%HV+4HQ3!f-|N)cbA|v$ILG_ldisyW)TEF)I<r_Y(%+(p_ygfxg<*{XL>>
zL{!-r^>{!TjYNG=9qn-&&_$3YQq?U~SqQ4DXld&>sQx%vLKhvB_qkCvoyQrbWI1U=
z>L?#d!=*+5cf2H+)|;|tF!8xdWu^44kPd2ge_GQ&B`fSf!Y2hItPj|b;ykRp4xPM4
z)m1jaVOakVdHZ3>_hZW@^x$atH}+?d2L}E?lrNUt*_VuN@LH#tqM-RBmvQx9ZxNIK
zjs;9NK(`l@(iJuq?9!M^ShOc>rwfoIRyDDAk>*jO<m3|jaY$(7HAAy(Y!m~o*wGX~
zwuMz2^A^(}Gu8yhDjNjHE^mN!4BhnPXJ}?i_4e8Smc}vsIQv<ll3?861H8?B)2tfu
zo8=#^&F~Bg^jc<mYRxeoR&=|O{sx3_H<s^e);v?D7~T<Rn76khE)2_v)eqc6&m8_q
zTjOwv<Fv!0>U7xzKkVz|g%C+#{t7%YkXPOtb>x3J`u1O`zaJLa^Y$wVRCt|4u0|XA
z5NFa2M#5$I2ntC8JH;{xT8Nl?D~NA!NYK2QCs!Vsi&Y*kz=E1wIG7*hOtdrN&rz=i
zXjd*6qVEnC8M;v;E!Dw_GU~3}K^|>j6E>BII-e1Z;SwH+po$t6`CD$(K>0&tp0SHl
z6v{6qT&^OPk|8u}f>qA8xcNp}uS6#jS&ibyK!s41Eg^*~7gyn@%?t&);uW*K9n(}H
z_D}iHRW2Vv_<vOrVxe|RcZLPH_lE_ZQO$~rLHI=1d&fpuygo8qxbc|Kn#s!OM$Bo6
z39zrttbXkO256R>pC9d!&R>kXzmf`$K+NfzT{VyUIcw~2V1_#49%@ybYt$7LT{F}g
z=vJ4UrWci3G_BQiJDcav>dK~Pvk*p8MJ9Aiq(%)>Q8Q`HH-bNwKi`5@oI;cWYGfXA
z!<_UL=O;e5v-XQuNNCTg#FO!phM+K4c)GF<x+Wnv*E-I=|B8nTm+hgTmLh*=>Rd8o
zcwW{gI_S^(dVJkZI`kunR%61Fp221rim&6yzr|G&xsHyH8Ly<#!&y;-Q-^^L+=$xZ
zv-2qb@)_GN*LxwRH{R7j^z~5F@a}9IhTWK{!jT7a>^U0Ga;D9sUC+Qtwo|uWW1~ic
z&JI&7AMy(Wo9&{s(-Ht1Fe&wNaI=v7Cb@(Id;)sbI2bt7F{8w|#Kpz;<|FFsS#EgE
z8ntRQXd$Mj_TA-lm;PFho3@inqE$4qk_=}t0c3)-=|c#wyIDuz^pY0=&$_O!uS2lh
z^m+cZW@oe3PBfvW4;~+vZo425!=p%q2Q*<oUB%PNrO%LQyBPJTM|}~P9iJ$>Zo>&*
zUYv@%>-ofy|3@(#yv0sGorYIxp!p_D73_zIf@_H>-`RaXzsDvvH<U|N^GcNZ-L4ay
zTnERkjSQ#RBX-S47N*LMb&FPI=_%q7eTheBl`LoA@uHQQx=x2@Z&Q1YC6-2~X(Q+1
z`DwL=mAZ}|$SCUidLZwLR8V4iu5iwCk@`GEeZyJ9NI^<5TC7CH02GT6Y%h@;a77{v
z-p6*F6R3qB9W6u&Ly<Ct0}*t^4iG%pyEfk1ZG>fqdf%?m3wy9jPzR6AWeQS}FXALj
ztxY}0L1n$i3nz?Vo4Zr0eB``BH}!H!m-D2r&V91L7BlS`!WK;3px9oQlRd>BVXfiZ
z*4XX%)LecYa{bCdF^*Q2XQ&gMs)C`asfZ3Od9w7dY^DVawhj1yG5?wvy$z5orWQA%
z$e(~s5}=y$x9`oK!3D1Al@7gen;S|9!{39?R9*YI&V0Xl6Nw@!F|K%3XE)%NnYosn
zHW8y~RDJlod-yWa?>>DK@zweD7G~GpNNg4E0}h|r^^q$<Ii;U;ED{&z^?U&B<_i>#
z80cVtvL-b+*qE)48Xo);c$PeM#Cp`02uGyh*R~jng4Ewl=C`;yKe-NC7pG|`DHuIH
zB?cT%oEk-2k48knA`>bkW4LQF{}=uQy0UHqZYqkHTr}{#6M=ytWI_ofpW$x#YkU?Y
zBo)2U12g5gAUkJz0=o%#)Ym20#3(Ln_F=hyv&MuZlV7@x4xZ4Z+&a>OJ4NtN=89>5
z+=51pCycqbduKlnjfSCOY?)iM*mLM?>eO#ol{Vwq*w8=!e#O|m4lARPO#EuZ-B3&R
zA4DOxghq1SZEbCI*(_6GEYz3UK|s7&TR^S=NH8hIvC$}U8lisQW$*mh*#C^vzTDMS
z8+6D2ebx@Uh0h>w@7q_?%6q(crBQHSCvpK)OVa)C{^w_m4Cix^-y=ufcxTCZ*I|(v
z6#kvW%k~gc_n+vDI&$0nHgYZAJuMyrH}es-d^>J)B_5{Q8)x7R1vf+*|JvBw`;a&x
zg%6%2SCt>9BLWeTor44pTX2oeH1!bq?ktp<Q#K5FJZTizjy2=vXqHJmMS5sAC=^*v
z65Z=zrTFbhT;Bb8$1-^w(Ro8@iOwRO8E~>r)34pu19R$VslQ}@)q|n*GY?%ljv0@X
zuClF)7ruI${Y=Q`^>fXoV(AB=fX@IBn*-Ej=c<E)*HB41OgkCM_g<|_#T1jYY>h-z
z-)k)2Q6RV?@Go#v=zvQ~0*&1L@qkUX(gIf`&jB+dIbVFbMC^~*K@?&|>Q`do8llVH
z6HjmIfl8k)YnqzZ^Q%Ul;n?zC;v?u#{wz(nO$bF<OIux@m=xAg#k8FIkbMWyC8*+X
z0-^vEwzkar6`h$hyeK`R$%x2X+o!iBql5pCkcqEau}sjc@_6C?%+(9N|Jr60Gs+|#
zcO3g2B!!-FK2#pm2p{u#w2Baok%Y%@qgI990}pnNCVlB#2Ju0f2yEDsj5OctZhPb{
zbY?3}fS_n|_h9gS(H5oK%JuIiy3Whth1U7P*&^P$_UVGv>WeH-P}JS3G~C%8(qHI+
z2K9|06{CiyC!_mbYl4=c8LQum<7|<Gpz`t`KtE8JCMvj&s`K%*$+QtXb5e!OGwYl_
zSwu0dROckHtJgqvYee0UaYa(Tb8yM_$5q)!Ns@?F3&@(xLl<)xkB|~9Z09{@;xQZc
zGdI+(Sq9XobK%XdZ^(q>#X;B*ml()|z3h9Wey_ZaM-1{KHWn)y`xaMrnDxCwIstUO
z7gj7UV5lP5q}_in8#<QUusGAi4`)Bmb%H{sK|TFag8s8+_^f_)`;}DH&YjQ@W6>?@
zE-S&5F-uY|c339%M>Ts=a>WErteM0gegEgTkozWzKM^w2F)?0@n{{O)Hvz2WBXtpZ
zRHKmrGW$7DQj%zeR1s0k3=)Y^DhD!Y3d~<>rpk>}1*$`swBhM0uA>P{Mz>^8MG{g`
z&JD*yOkDfRGq6CL9XTxgKc>&z1_Bd{qxMn%U@MPC{EU`RiBY0f+RF{4jfugroT5NT
zSCqB3Tr?frzj4-w=v3BTb2ea*=Hll!U{}`BbI~~EiwlP9^Ej%vN3$ZNt7wkfhq3P1
z!{}_8+%#=Tl<*a6?5mVPerGb+;e3e`?34|@%2>%A=w$c=4+XD#*uvRu%%82iggi2r
z*)~(I(!V{x$^yv(Vb4yF?H_Py)VmIS(PbphY3v;0ETA&}g1ENs{_%0x{z=!@%3RBm
z4Jncd$QYQmgZ^tYHrwje=vGUPe<jBcM^jOiD^rSMsyIz{y#zC5Zz`zboHm+qnTeD7
z+^iHQnE<jP`|4>M2S9h~*K5{q*gOw1`1D$_Ba~`80EZ=n>i$Z{tlQ6IiUNx<kIsXG
zm$5alH9Op)+uM;r&5x_Q`ndcWKSFyiFEn<uQ7<5DOF7-r(Q)@|A|jLb>>;}3!s~t$
zH(N5m**iHI5St|+vU8L^6#K)nWDGP}whc^LBkB)*rvhPPsm9@jC899T=DMN#UDq=a
z?rUr>U!G^OE>QLcuX$JxHtW%`{dhyylvf6_dG(i9NDxoSgKcg9d|^)92IX6v=;vdp
zHtAV$n98ovGSb?cMzdK~RsWjD9jv>LVdD(M;JLRez=2#5N*K|*vRj0??~MI*aLQ0C
zh$lQ(^0z*pePWa_FG5%3>C$zgH2i$T#=7Oy84kR&X-;FeEbp5!$mM#-OahsRj)i*Y
zVtt)Es1-SjTgfJdcZp8&kVzcUQ@MGQr*jPR;IdfnK1(sJ(XmR0M&OAI+vM2bECO)L
zY}nt^Yc>}5+)*L>A%zpj!Z{|Xq*T%s_MqL9?{Y<u)+RaELTSS-=0Q$;C600(uJ8}+
zPVpv@o*|FYSFgS2Z1=CK;O6lxUE@8{3k}Bgi=fQoUT`c4W<CZ#*nAn*$ei9b^iWvA
zo9JNev9Os$MY691*U{s_ua1_p8$5@LPzR4`qPMk#o1RX%JotOSB903A=DX1$C{}(L
z2K;T~hfYtuWMZv8?-=zB6owx`Umbu@BO5x+pe__H{7yzEp+HguLZMl9`K`4<;-vrD
zxWo#FZ;?-R6UBOuL#{Jx5KMv!k0!ae)_?A|G1W?5#g1mnaA^G%A0|*V-$F-LdVH}`
zvz@$BC($@SC&`$`KueZ0gV&I(7~Sm<5eJ{KBb}qngAPgbJAOe7C?Ozu9V3T3eTpY-
z!5P6#XY2fo#<ZR(G8owdL=~5&o`G=uT3LZEyW2=SBsj07)d3$9y&`gFK<VxJ%s3=R
z)x8yoNsX4IqEYNc5dr?eE+rKhOfR#)kwE0HkTWHUvZm7#5LdvQdklqz<w(zf`FAbk
zdSolf?6e{>2JQ27FXQNXujiE@17w?s$m*5drT2x8gSMUFuV~kCOsbmD=2e?3u=qjC
z+AA~eXC4GO4cm-mHrD&}j8mXK>Crpbz9iu_#)u5Vx;7aB+*PF=$YxmCG(r;+XeHfV
z_MkxdWoZ1bXWtgbp^>r?E=LoKCCF!Vhktht;Rb%y^HF8X!&POK?O$uQZGFB6<x#oa
z9idKfovcQF01mgRDqLXB(JG7|@Ay$&GMS{gw(Ob<rzP++@Z#b|C)NPd$fQ+gt_e|g
z{zY`E*A2)nXO*6=#_aBXL;LeREx(am5q{CSrfT=sd%17#eDU-X?7uB+Y6N4nK%UT_
ze9Wt65xN5K4~xdA!Et_0UL27=w_$-pu^awD0(K}TuMRD}gR-19xB>j)CLLw<L#GSh
z`*4j3P^EGc{(i2G%Wge&n%274$*HGeom<dYA_4X0uGrWz(wofV1g*hy5W+nlsz54?
zu~?`P2itNSLRq9MC-YZEv$s7$vJJsFGkQ+XA@E+me$L^>e?<L@W9gMO(EEB6FxQ^<
zvlrf91s<;b75x6-ztPp}^+xZ77Fdxj43mR{EqchORrzWMBab)jBO0q*>D`iu-CwcU
z(SJi@(h$8~YDC)7?|yH|D$S%w`AWFGiGAt6)EMpS98F}lZ4xC!PVJtRXo53;j0%NO
zqLe_pVq30Zh0?*N@pvt-J3$--%RdK`dpMOo*l7QhrcXKsv3%}xx{K*8zOphq+Ix5f
zCFi#W?L(ARPdT$eP8$M;^4)FwRR*)~Gf77)G>J1wpJyqL+2{EqzD4nTCS*chL)P@G
z9qVzI*;+lkD8Gig2@tyfG5ogV32wq%b7%sIxE-#3Jp8b?ILb*ES=^t16*2>r0|2?%
zI#a93X0?V%iI&iy@8JS)1-6Y{29BME`V(RCi3k$+$!SJwKOvA=9Ne}Tt~7zStIvP4
zXC(hgC~%ikR1N#9k0yJfq!h29c<%@n`DGiA#W(4?h17A=iD-ZDm5x9flXl~~vB*j4
zBn=@*MnZB#l4^7$jo9jo25z%xc++DYIj6@{*9}PwAFI5?XTj`bmZot1a<a@O{ti^g
z6pb6+!@awx=1bsLPI*!uwl_E_^%<upveU`tA8Tc&PhuZB9=h1;`5Y>Al_Ln`a7re-
z6I{(!<XJMSMZ)}`{VvDsbw*)<VNVp<5H#l1=bjxge7e!!AP%>w&%abC+zwM<-0IZf
z*>2J~J0TM-GaqNki1>VC__s3bsQ;(<6Mo{<=5(9>=iMtL=)mEd&p*#UReZ#UMhS7M
zF)sgvR9!_t`KSmC;ZeC0Tcbf8kPB6bs5<R=rdotlkRwfp(XC#l1%kUfKMvHf^to5{
zOVUzs%H5?kRrssW=|s0hg3Y9^x1N%UG+%?R<Vb@JI@^RKo^HxFuiCp`+Y8)<H~#L~
z$f(IJS9I2YfgG#Lp+!(hH(CcBTZLSB!6hEyWRSrFWoLs%Eq8WNi5s#)D(SJh0eE4T
zh^X0;y`!hpI@;EbmZH<I#l^+Onq#L+4AkT3BxJJRK*Ub}F(?0+$P9}+Rc>tzf}R~N
ztm~7EcvY>MQr^~lZ-3qJbl<WRiOciaC`SxPf)+ILJM|&?%$E_T;PzZdSoeBNDLw4P
zo0eJPF%8}*E?j4i22i4O708s%o03Z$AQ@+QeVVTkmY$!#zlD64u<OOvpx{!INcbsO
zj3PCM14N>7&$R^fDuv6w9&4FnQPJuaD}2wL|DlpX+74x)g#wey?CV<?K$jrMRZUnY
zoSzgsOQ9;15G_#;8APraAVZ@>%~X{?k0wnB$pg#7UHXE^RJ47<!i3T@99N5lL`k6L
z5$HiFA~NU%KNaYZ|LM<n1<p!uZED-at&02hW!u`b(mngLGk2HWA?b|()_YUvDXj{2
zjqIHIOILionJ2=tQfHBQ8^B{<g+diMQ&|cm`AokoTNro9PgQXtEW>hanE@Zxg@)Zh
z^T0Rfo6uv<yL0puRv&7{>*Y3dd<;;g#V1U~H#1asItO3=$}a}6GtGpqXew}ZBaB>K
z(oLWWo?Un%CwYWMZ2Jp;75=>utsi<;^pD4Qm7s4>gUIJAva$v@NWPNN<?EeZFrtP*
zxczYm+gbYz7eFu*P=gT~Dbt7zs!P4vnl+jt<Ubh4`hm$oaF4&~YIKR10hK6!N^1YK
zs`E-SN){wO`bP1~QllEYM)pb`Y6W7>ijhzfHRD+|mEN#QW#A}b-9Iw!+oha-?i6?;
z<$t}FcXx6b3d|=p9gIpqjbM})fhlVNJ7YIhnCps|bryG_#y{9M6q!yzw~md9`WKqu
z1$+Spq3C2J_+wKBeX+$)bAPQ?>I}vs$SX^|`>DGLHCgf0>uU&fmvWs>SR5PJ;TX8w
zlM~6Pg~^lLMN4TM;uGUZ`dKVkVdVA6B*k{V);vs^PI<0VNcQE2#y&k{AS6d(z)jDI
znnkFIB&4X&h|bh$*=pK$99m6qq7fMJ7XZQf@}hKkU#~Grpvyu##p$bi>+C7F@$=m!
zXGjDE|5DWQrNz8{C5RcX7PRD4*~7KTG=*#BrXSS5Qo<0&>O4I7%y#=6kyszV`j`RE
z^|+hbaYYGfa=y%8%&#2CmAa$9>}ba)bdj0XU1)jlM0qWjUUGj;gtFy4uP|!Y%gI3G
zSUT+Rc?axWpfZ$Dv|<=dKJG9#5jZqBiKwLwSRN{Th4`wh_fVFEH5<t)sv#)+VOIdE
z-u>f^8t52J3;Y8C>Acj$#BYGQk)Rsdr526%EL!l8+|cOXrIhc{Opb&G;(@f|67xww
z*?)hZ9C+2Wn@k#(1fWs+c#^3``$Lf>=aN`&eKu{UH*@jEow!LP$#s{uKqj4fq(Q<F
zP;@+|i4n%^VsPV;SihKlRp7|eRAFIZ8&O3Dbd0Rnq`bYT`wlXEoXJ05zQM{f4Wad;
zxeZNdH@O+T_hXy8N<R>0tn0xS-t|SQXS)s2!xTA-JoF%m4|on)tTS4LIbaxldvZfG
zn3a#Cf%i8jS9;b7mz|&aYpoCq-_S_47Ecw(L{MoX<nd5$vxJ>#C=;ntWj1Q0U`V`7
z6WpozY$mNT=H&4+8A~E<%W4U)O#9R*%~@AG{XocsU}Pq@P3w$D{3}2@3+p+8s`^C(
zy7Kq&o;ff-+=EPPHM;yM;VrhjRrpf#bHz0|rz_!8O*nQr^6u^vY6Dtsbv|9*)d<}(
z)ypKWBKYU}>Q*#vY5gT}LdCuJ(lHDH`6tvp@ihi4Z@%v#aZeMU*%#qmc9|07m<C1(
zdgS0F>Vj5glj>p3cs<?#@No9^R&OB66jgiS%g6R$_M!DVuhG)HQ;YsDN|;c>ie@EZ
zG_Dpj*ovnby{dn|HwhzoPP?@;Ts;F39aj-SejbP@Q=aWG1{?QtTAZ8Q0}zvs_&H{H
zFP<KFB9ZZP#rY1H{)nd$B1H+;#zT}u|B?j$?Mnw~x`$%<!yekF9R<Vui-1Z-A-qXS
zdyJAzg7xrt2g|KCD#K`|DXX#s$;4wTZKFsfU;xpmX-*!`f&AfVw~r?r%H3~6D@igj
zaRufEvkiP2ESym+PEGkXzireUD{3?hmaUQ{_6g!ZcINZ(RQr#$hCx4Q;pyPvDWLEk
z;D`~{b<q!gn%jPBX``#uo^4~xwC1g{L)tJ;x9)dnG^}4ys%}ngGiqF}(R;?Cnj=Hd
zf)`iY158lnG#RQ0P~!ZzA#wgp(|BWzCa0p_BTQ<NX$Pk@!(dXyi28XW(|XYJIxL)R
z21<(3GiWlSjY64v1j;DarZZ6)&Sz$A%17dvSpV_zJu&<0)%S`ze>QAir>4cCrixsa
z&v7Tib=b<yXc5YcPBb3Kgi%x0D}y^*clIMs8L8ffOhD7&$tGZ-JZo|d+YWDM=l(ZS
z(KSL;vtidyZQ3yBpYgIDv%VE=<;qSgSffkd1rUHO-QILfVxb4NLD!Li81JTKD5+Jq
zG}Idg&i&=;zi#iBx^t~LtePWHF7ia+eJ>nV>Q)vTDos`7+}s}g%0{RoqJC!1vRaPu
zny};5>s7nm>@j8gesGl!rmwgSc{Cnybbkbl8uwXw1k5HkS<bA=_vMXnad8=P6DAf@
zYgtkxN?ytsRJvvXxk3QCXjrT7!(yfrJu>&z+M9SIBZEsxvVs^&NKGXX^>iWNG=qyH
zl05%7sXT-0-bI8Bx87O{a~!X&2{`(d3OQi~-ZMA10ZN-YQ1M9j2&^Z>w#|t7^|mwp
z6e#dE49eo~UG7PeLO!WU5V7+;x)T&YI@9YIEI|q|7=n6u@sTHf%3|`cd_(-p?Q_{o
zta=$wZ8DKk(b1WClZ)#WMT$II0YT}F7fB{mRGLB++!M?@9#ngXxWWd9r=r7e6&z>K
zjgfVYFH7-Xy(8o`c-d2m0?x~;zsWk#v+Ly4_s9v&f6E}_t_by=ccswkTi@EGm_oC%
z@8!Y#%!lGTre7w?onEr=#oy}|xPW}MzRwucc{75@t9?{BQJDnF&)$RB1b=f0`pt=3
z{Lk0@8WCD-)qNkjU7-}PE!wdccVb|t=i&|>cujdT8MB@y$77qjyVBRz1l#$~BD1BT
z_C~><!%OmbOV;FO_&NW1KYm@>fOU)!QSnZRyU?k){a_j3*22QS>3jW1_3IMN`F;5v
zDhP%1xX;U(|H=!(6*9E8a(at|+=Oi>QQmiA_N@CrciZtr3Z4eJY#BRFYX-LBp~8NI
z63?&@M{uy>n*cG(A1RVabc=LUT|Ska;2YX$E*|fZ<6|63@nLk$aA5-;>3WOfMQhu<
z+{)Zzt*Ck@wfWSLNf+s2o{B4d+fL2+LoS*>BR`>r=_IwA_+L@@i7bnlE9&^$KEs1;
z*&oa_mQV(~YV>eUmpCb9F1P9dQ2C@f0B35-NS`$9e>;g7zeTXAvPqp@I#SC~+3Aml
znBT?O9Zl_^@zR<aHUshVZr_~+9mZxU{7IWWU2Wt%<#k6*^&}w&DkKx8q-8WFjb|Oc
z(P8d}20ru5Wp6*_m3x2hdwpp+`eeI><g?8)4y;xAch&tE>vm-~cE1qOT)la9vB9(2
zgjeHqhe(%4sKbemd&s&g**kw&97TyEyXN2wHjJV)AD04vgaBc!ZPDIwzmv6ccEVOo
zM-N=1nQ8CDDi+><?B0G?A(+&F#D2H-Elaz5!-SD&oN;kcw|mpM^Mgf*fXY~d_@{!b
zJGG+ZGV717kI7&5*AE!7^#=zBpfVZM@L2exXe9#Y`Ptv&Zmip9d#R&tK`MHPOl4C1
z?@y#rvwNB_NB~lTJI$V@`+LS@obYTTxZJsOb<Wwx2q1ACrfv2_<yzx4-6t1a!yo1k
zS2*%BR-y#(PP<?AtZC>I$j%F&I!1Hj%RS|EIIrck$9&(MbEX@@&Ej?M+h2gDEH>+K
ze@CjM%l{qKnBI1=%ubLWZ^(+fjSs@lo?}G>EVF`Abt+Y*KM@?&dN%;bjlk?5SOl%G
zLpS~|q$4I%c8RfK#BRbZjP;H70n=+Ud!5^Lc`3foWAoBArakOe$R#viNhIzs5;6<M
zvB|oM=_Sj|ocf5UBzxze-ns=nXoleLKPVa`R8k(cdNpJxw6;G*i_rK>kttzQBvKRk
z-cZl4d_DBSXRbu#H*l3oAhT*(iyO9s+q-~i47CQ=yYz;LLKO`(C$0__p1p3n@E{Ah
zC%j=g3=a7<`*%fs0%an)#Q$XL6y#iL-gw0-`1IxVQ;%FTo%Q?-JDu32U#4U4v;U`M
zo|XbU$Zmzg)Zl?3;`&RbbS%NFXM0t(k3U$@lFVQEeTU~;*ZR(Lhf6sikxv|ixyAN*
z3yu3g2({aWctnfVn^n)RYpCmBuW0A!rK@b)dS^Ymc0`EBDl1nm%SX`D_sWQ3&5Q47
zDtQ6Q!5kzga>Tk2E^3BRB;%bpii&D)XJ*~~iNoR@-mei;&c^WQqQ-x{|9xq_O_gKm
zmXkR4*G+OpL$H8dBg&Z@XO>X5<@nvv`7K<EdMxUAf~QxOQn@S-3c-4#ZmvetMuTCo
zV#zYSDfbvR`4nRnTWd|}G}&60B~4E7O9Rti>q&E+c;f`tl&E3S0+hdNbdxTTEXl00
zr4e7$1u5!!76-#1w+y0eEoDP<`(fr&N1mFkhdt^KHlBG7mZY7`j<xztRd06&e{{XS
zSc1FU#~1c7H^#uVB=YFl?^E(_w-1P`zZ<O6@PUWz=@YQL0OsEo(beefp`PPxlZKJR
z?bj&R0JNU#V<uqy3H%N08F+P8#d_9FV42bk7xdN~TBxNKerV7w@_pc9jKOoxpQ~N@
z?3|?ig>vbxr^#j=5JN5Z<AbHk(I%nu<`+w^r1PdVL*4N+&i%VHsLJNIy4U2%wj$hp
zue)oU=_~=_ab<8<t3iEN4){0d4{Zt_{ZzFkeKTug(Zswz$$8cKzJaIt+p2G~uc7Y;
zGF|A`N!_`Xt5@wPU6c%?8ME+B+KNW2MTD;B8+9Gsj>7XMxUa**-<+C7Pm|m9pcj}L
zGDXV1l2iKSf~zFI32A@EjxkjRNNNkeMO9UGbak1#Tektyb%U&kCZ-I9k=iLn<a5Am
z$#>ZfSM7wSt52u*5ddOJHJWOyuiP~HxKlF}p*tx-&edv(&H9wmVaeItNTC##WC%B^
z){1>e#Y9dqnO1$A|F|Ow{?vJL91Bb1hs1mo?YL$ozP0x4e0J8{U&I8+@^2?iCLW+U
zi{%Qnw|$0xp#Pvk_D(l5(tnj+AQIFbo$<>cLwAJ-2v0usGxok(4_(xBhwjTYcH?nx
zew~UfBWPwJiG3)8B<h1Us~rmcaszV)Etx=4C>#b=I3!~h`+RKX8`PA}(3RZbv8E7B
z!RvE_Q>5zkdd|kdVHwAMmnk^*N&CAXIocoiBJpCcm*Zj}5T8x7C@LoFVY^KB=@6?(
zmv?OK9OZD|E<e)W#FZ9Y<E^VP0&{-xP|e{*bosudj)C;r`xqG&mHEY+1o}6FW(TK!
z0usI{aS$+{_35hqE8Fj)t;`XPOoqR#d?~Y^QfJUs0*+#U_?NYhFdcAP#<{W8NwTSo
zMJ)>93~nG9_ezd$?}#<BpDIsCE>B(<Gb`fClBJRs%|eVZN)ar#Rmah=z`$3~u@gfz
z)K`$$Y-;Cv>K_Y*=-IwK>!9ua7sV;(dYskAU>awAtdGc;c~04df>vsX+jb<8s>C5w
zm}2_&G%T=_{g(KT-GTxrLcVkG_si5F7%0PH%2N=Y4N~A%X|&`-y!!{aXbNh6nfvRR
zL17`BT)gK#T7nBK6BzQSPh+KS)!?v7*0!lS@Os~G_IlmXRy$?WUD^)HKj;3vF{3G~
zX3XpT5d~cOZ?<|xZVHT9HfK`lLjEJ;jb2AhdQ28Ng1X&@DT#c)D0hpisfOo?%xn{2
zBXt5LU+W43Z4TNL=_!V(*0wIgS}j<ixljvkPrISPwIyQcj6eI6)!Uo@iivWa$4xs8
z*nZt^{)_z}9h!%~-ADYl^Y4}caD#r;uT+(CWax^m(u$MQq*9V|T1z+uNI@lz2Em#Y
z>tqo*dOL~~<yd&-7-DI(->SzI#pHh*c7iIp|Gman4@RaUf;y{XHRA8y&Zq`oopWIQ
z&448Tl4M7PfoMr;lya9NDq&l`(1cgl2dEnCJ<_OhO?ow6?}t=*0dFbIeOfF$y)L&?
zRn*08P>u!bC0MUc$Hp92HcfoU3Z?ZGLSlnFnn3EQncdmm6*IiQyC3wY|Ms%&V!LQJ
zdJz7Bt18<qATCqH_W1$uwNCcc|DLkBXT_OBw1H=w*)r}ECQXA`_}#1&qtj$)BEI))
z0<MdJ!3dft04-7jZrIwZMfoF+Rn=7*v^FhW9Z+W2(KMh1pqNjVKDnyAD?UugX$C0b
zrYeh7Of#Ty#Ect@DqI7=7>9@wVc5=pH{Lw>T1X~il#q#P_nWqx)N0r{c#JhwwWe(X
z>tP~IBGZ1&vjPth$NB^oP-0lAK%A0nz|<(#py#k+r>oD8u(sw*&$M&rFP<>#@VY0@
zD7w|t5VOBNU}NZh9`iIKr_KzV!nUn;$-!N}?8a8~!M?QP%z9J85!u-&3?VR;kAsyN
zvh%_ZAimVme^pVO4G<?ETg!nv77^pTL@iU1N+=K##dE-j0F6E+g(7De*-xY_b+o(n
zJKgw+BMbDM1nu7-VBR8lp{vV#M?zo+-JRzJeIG~5;`JVyHzQ(7ml?T%G~x-mb&mf;
z%L6FUHaTT<pTG!eJ1|E|NOouV&!o=BKN0Os5j4h2)14wbosUc<L|QM1O=htji@o(y
z4ojO#0f-@&oAIAr1arL2?4dScHs+iVGi_c|%lWkNI`#7TFo@Ty7IbO`?NQH5uQDI-
zptWuArV%8Rg*2bgC73FrrIb|k1Cd`RxfoX+yp;ji{8ER{eaNjV7OU*=FcIJG@%%8a
z$DQA(V(=eula%)E=PK@f=l3r`N?G?`0}@<2+CEE!wB9VCc3c`xhTL-dfK&dto^2qv
zEy>~EUFLl%x@SLl=X=Z0eGfapszX*xFdaMm)f{J>kVpIczWML-;whKyrq85{W1vbE
z>_U8NWS#uHu&Piz!i|h!rG6`81xL(&gdX0NF8L1(3qnJp?H9K~(X`J1M9~h4PRMnt
z`B&%O%?8_2ilf@N^W}mzI`q;lK>0esOXi?!s~rh1s6xCwg-(gBdhQo^Aq4^j0TLa$
zV1WE}aW2|o3t5kIczDRKULIUl68Bx?3r14ytmVRIURWn=Ts)<KN{+vim?Xs0j$>+$
z9~TKgKrNPt_aN5i<}oc=3^3T4aNaR&!EdKqLYAm=mK0M{Q%`5&0-rwZmk>WQ^l<jO
zv6}J{8by;xs)Q}1kOucqB(u&MYsSFH#l&*)@azwkr?9TE!^MKjHM#yZUu<>U6L&^t
zjT6dwGq$`E>n}AW4;qf!^8W++rWk$Eam@Y89?kwFBEsuX2JaL(1tJ*$a%fj)+)9lZ
zHyS(EfqTQ=!DGOf*|hVw`~3}&&!{Rn#iYh4x=nm}drLnp59Yum63;dBodcKtMvT1j
ziK-3MuZKk#CLv|y9j1H%mnQkZDLbCBxJ?Tn`IR>6=ymKyPX9xzY*yJ1yDrG9>g*4?
zSnZ+VYWR=lxZ>1TLIaRjOJx*<f${PY+MsFYz$*JdKl4#jPL(?Lh7vlpY$ptNg|*~2
z(Rd7f<;5S_xD#4LN`<rt2`c_9QG-zh7A0j7`T3+ENAKn6I$B`cUDX2MAS$AB5|1j0
z?J`}0J9p2Yd4}q#HREqj2AeC?5UOl#UK;pzuY(rq3|cknbsDXj16bJ#((>^sWnzbP
zbU-YbF`cCTDrAy<9-X42m>a#P>YF!%)&CN*fh_}M&)AVa2cnM7c++hWJ}-J&hb~Mb
z?Ai3}Ar5Q{;Xt<Y#W7)%y25|5rww9weT6QFRpK|ZO)n^672(ca=!J!r3-qs!z1<6l
zsRG-yc5sT5M-0B<Zn!(bfciBfXOYNR3S_)z6rC8nB+_!wpr>qavc3NM$oUMIvkDti
zH#u^yZ5KRxMN@b3)|`64T-BVF&wfv>_uOo);epQ;UK8(?y;~7649m%_1fvveI$t8w
zIZSuiTz!W#hv*GSh2Tbx-DD6zvEmj##J66XZRGb>GzR0J89IQSxv*68&NAo!Vg9Xk
zEpsB1AeTTBN~b803{ZksdakJ%dS(sTazy8l9d}u))3bJee@VCPdJ9o!0HobHk6Jm~
z?Ks|nI)YjxqP&2+gqEbOFP#jWIfK0-A~m0~xEuRQ^SwU=#HGdBrR7|IC_&L7-qdUR
z%KhVc5h~sW`+RG3x=Yz=U=3CLLvt5%`!&Iki`ydWq}$aNNi3nj0Wi$SIL*-0Uf{t;
zgH(rQvrHC_myY-@YT2vhX4Zj?xz7Vwm<k1H%@%AyP*9S>2?ZMXs~S4<3LFGW>cA<w
zv{}=kInSQ1$h19A^bWYJMsu{g3?CZ|rD7PTt-01;#m#_t;P9TiKY9d__X}wqZ{aK+
z`{Wphq?vgJ%!W%v%8#*%s|Rn*p_Y||gF@uykiHznBXH9c(uZRoutg^Cs+3|)E^$9=
z3mbT<Cu7`MJqP;5W8HwBQoiEobs$KsnF`o3KrUWylx<IWeLCmFyib|^4i#C5iva?G
zz88KufYw{SpUQ)ua+6J*C)mAC9W@rtDx$UJCCxDP8CcdAF^jqsf%kUPuO;HTfc<X^
z*br`dQK6+Q92?}Kv&K@*xa^MOhGfULAVEVb@hstlmte+mV8R_SY8=!^Cz@8j;RLYi
zuNQ3TTRfIRI87{4CnBD`Mh6Em#up;;R~k5Mcxhd0?TA~X-xuZ2c%}r5SM*`wx?*uS
zjlYFB+fTXI4RBv##tLcpyI^k9Zu;B;sydezjUo{hs18nggdcl3Y^A#C=0YlOEhOri
zp^}QHlXF@grKeW`1J^c{oB}9=kz1%k<{<p~%I3fK>4Q5!wen``4j%YdfjwqR&MK<M
z_3S|kOtSNFR?VHgHH~0C9m@^d<%~&~4V(lrv&L0M9{z)~BdAhzf!B@A-R(rm|F#SA
zGA!FHYHU}MVBT_BBrDqT7$-2Nt2B5_HZgIoF!2CajeueK7;35B!F}ndjXLl;BkrM9
zb*FcrT>>DnulKRKg<J`~vuTjTO%+$|=%dEMJo<b}miQX9)d}o13%z`N-PkQ4GIM3@
z3pEPrJ&M$ydHwPf&8kSVJnQa)bQ-7jx|Kq6-Casn0f{LzY4`y!o$EwtM!nuK&q4;@
z=r+CHfz?=BCpTXX$pAnHRVb6CtySiW`>Bz#l<Dv=Aex4OIgy)ZbDRzzqc;*{B9&KJ
zYa7^T*4AI=fq>w_TIoH+Ek%%x*A^>+0<zWnlk~_Vi<v+}SPi3R4TWbhMx6IpLzZaL
znSWdVf@`tT$H$J})YFsm?ELcJdS;-}oAYD0<F;&d9>JSUXwC(_?ebMZA8b+br{(7w
zl$>a!K5nPX{hQyT!1&`5tb23LQSJ74LXdI@h_4MhU?bg|NAH_yuVHOY;lbT@Qr(z@
zWAFs8-D9o7HMsEllTKJZu?<U5*xKpMBU6LxbTg;ga>jgUN_ckn_(YM}8SE0;cA%MR
z=(hKTKH@9OL-qoUrwK_%T(QcS8~5(hUQ~X*@<Akn=Yv|ILUPxo%;CJ!ej)5vclyzK
zTE<*DY&!umV=*Mr_=LhTtODb7fEr38>zT93hIo3PEbuR~d+)v(OM)^jih|ecE>++;
z(RX5H1xL$LughnUl1N~{DMoGD{mp62EZaLoCE(h8Lf~!JcVu!Wq&cP9qCMT7Emwv)
zd7P^pK-oErIx!oxZ^RXZKh8|i6i7?_3@8$b585Kt4P!A2)Tb>KBPUe^QbF_SQ|5AV
zv4B@=%A7@<SwJJbc`$Z;Ene)TrHn~FKZ-h?%~wqIMJ<C|c1^!=+tv5bb+kN+c6vlf
zl$lWz1*-iBpoi6Dmp3PwjJoks$g-rH%Cg<KdwzqyPu|dJ*2jBkX#P%`O6R2k_}%Jz
zStKvdL~rzwr$Ld;Nb`*nh2AQ^cMgi;=#@@lc%|Jw-bet?CTi9^2puPU3@1B^sj0|<
z!)g(DFQ?p}&k^q_f)Z+w3ehU&tvK+&U0jDRzo!D{flY;Xoe-AI%NKqtnXML`PtFJ<
z1RwWVVNIJV7fQ-KNXr}G9^qQ1Ma7yJr+}Yu$>RV;qC;@o9B3TzkWfIVDJ2x+Sg*Os
z7krn3!3kHIC>l-@4L)YgFEcFViHQLy*Py1#Uar-q+zJQw4D@w4BQ`9oL-2)-PjmuK
z<bJ@46<{oJnxz5VK+zCKegR{}KPn4wr})5l3-H9<c#En4tgA_*7HQB=?cU>2c-<uj
zN5_8P>r{g-1@Orr0ye#_x$>`ro_Ao(=v3Lhw3=ekVrf@kM2SILT-f!^IB7>H4z1nV
z+JTfXpiLAL3PS9a6f9-6UwaaE%U=G~J!VhVk!z+CKC~>;48*JNV^B*wtE-P655?7x
z_<f4%6GZWgCl>HGeIwsf|G-DX@9Sa0GZe(_#rfK1rlqRxtZJfas^{%gLZ@>T*nZhb
zDhJANFRz=BV0me5IRu3fXt#L1>soe3Mn*PUsUbKq31JNC@1QqNp6^Nu6mYIhEwzoK
zoIljw;Zl+fhmN%BZq}Wjwyh!fu0D$o$%*!wQ70m!p{hta6$M^u8w=(Xo!VUK{haCi
zX%<q{wYMN0dd@1Vo9HAP(orcyttTalViKH~sC=>E+hfcva)?7UGNaEjAV*df-OIMP
zsyKCY&a6i^Y^rsb@XcwtAgCX$oY_>}2KGmx3r+x|kK>SBDRI+}yi&~wh~G&GObi$1
zl9*-4Hj?MZnXu3_%GV!x8kNgaa`xqJ(`8`0dYxLW>f=@VL5zY6yA0%rc6TD7Od0*)
zk*k%Ot@e;L>VY{EYsPXdqkJ5_Az4Se1j~~UlV-zC8%}}l?dQm5pljr4?=v{L|7AND
zgubpmY(IYlEX=`5v2}kn6;lgPaCFiW^TW1M5dLC35h0jl1+V?IR-{o2dlt>1;u<}<
zvD=A!P+cuu#{Xg7k=DFe_d2NX2na(*10+S}m;Mlls&mgsyv$h&l2#6g_!_w6I2q}p
z{h-~zZ*rI3g7Z)ihhEr2Stv68aH!Rp{`1PHc^dU-M!!$EO-r7*^(p<iWDcN3b5Z*g
zF~}(?hh>VsIb{J&o+5-UOW>4~k&?S4Vy-+`=W!n~%k~-SJw|;l@Kdw%x<rqcjB0lB
z>1FV}N6Y-d4-DIg_}}C5M^TAn__umz`Sf&%Zgz?O(Ji{i$jfti+l}2abmv*{{4EkI
zj~bi|=noX>pe6CJeJLPGKVdsqUr@^b5=;>&5@6}CBojfcsaNwm<?ZQanA76<sa~Fw
zlXpwJ`yVX+q<pGe;A^-b45KCnzG*0kpA*`4#4&gZUFrvjgNa)&PgP7VW+6n3dbJ6W
z;)PS4=8r*Tj<vVWQvOO~{$nwLc#Y?0J@Kta!u?J^-6C>iqRcBE0V_?qrMJmnU0Vz4
z2pyR@oM*AO^dWL?{>cF=tcS<*uU=PoL3sg1x`2P8$|fo)v{bQTg*!Dx-_oT;GScfR
zVGy7atn4Iq{V<s@9?9lBAD@C4b$6zeB`7iL(v@xA?}h8KrZlEkKSeo%arT3cCu^ea
z`0}F|@%I}IeV7AsFhj?s^2Qb)IHrS$wC0MFY!(lf_@rF-SC+jaK${Zszd|zLiIACt
zlh@n(JDDy!pYsQx^XXTHTNLUm$<~0CUFk$scdpUhN4I-F>EfZWaL9(bDjWd&0ps?C
ziifrOJ?dc$Mk5pMO(o|WCJtHX>8Ck)dAF{v;&`VSb%zJ{$?I2uQ^}|y21lZz*n&GI
zR!)N+E7hidase8dh;Q#~Qkwmg9Kd1k?3~f*nzi^kWP8*e+-(t_W~3k5W(CImJx#7c
z_{1S4R0Orim2~H3L`CsUZ5*S1G-wKUtgPS||4|MwZBN)li&7^YVq%Mv{s_ZGqBF6L
zkEmd^i<o~SDhm7QefBaHjTPg4@Tkq7bj+EO$LqN}spgBS^gw}yM^7+Csxl*!C^V`h
zC66hK2p5^MmDB@Q1Ia6-*YhP5lj6xnD*TA7Dmc#6EEpYEo`8sZf=~(djI)m<2sy<m
zK+{=5b3<fLS;fYFF{dJjMM#yvqJ}>R!g+DlDH(teP^rWqqhVRlJ~##v6|O}XF@|BY
z7%F>rn@3ZZ&~>pgTNpeI)xOZB8U|Xh`Eir836uT$(C3!nm~D2=b|*O)$Lx31!kY`w
zGzz58zI<-f9{g-w-|K`(|DrMa@eE6Fw<mB&ihUAnF?<+tiHB^BXexVE-JVZY`n8h0
zZA4Mi?V_nR^*&<IaP)ZvTHtks!R=#*&e{@%3gCRWy*yFh9w+gRGwOyxPDDoKO_Ux$
z57JUHU?`31wTVk?itEGv#Jv#lcwlxJMfY^gu#PSFhG1CelrFto*NyrWF@UL!&D*M0
zpwk<OCvSMNgO8S`hZ@KTayIBsN_9_Z2-@*LKNXeo_-08s=`*y7!mB1zVhSsB5Xws6
zG`_`wlNZr@aHrMz46h<OSOAJU4Zy1}u<&M__d)ceCBx6^F)o`;8$cK~=HK7C3S04t
z7-LZ@M~uwCQz@b<OYlU#WYM%90dC}8!d&GVi_)O4c(vlOM0LlD>%LC0hDyZFLuQ-L
zNoLboH?yHcbxoGZrb4H`%MAPGDTzWpNO*O%>swlcPl)fpO^A{A5&Nd1^imNOg@r;1
z>^l8L(&1!=XQLyjn29s=O0$Fi1EfJ)zWytP!h46*iiaOO;Nb@k(W0PI6Eanxqq#TN
z1-T3xj*gbbuMtT^U3ioq0(IOd8<r4)OoyDfT5GaAXEG^~G6)cNCTP})ERO4x%Q<Hl
z)G1TQZ&U7MxvH5>OO8+GeD4QOuvXC2U-9-;7qdZxYL!TY59y7)5NH*c1!G#8ZbQ>9
z$n%`GEm0bDCaETxlY28x@6Y(^SJz}REHu_epyoBpo0iLqAS^hWXIMP=6vz+~zX$Wo
zu!9O{IgXgo0^zwm0fYrPvLYihQ37{EykaPB&mzCI_=~~x*Zb$hi6JiYAbez79B<#g
zqH8wbHJJiq8;o^Slaj6rs+9BD0oHX?lL>9xuwJY<IG$l`N7EYCs~T%sTxXb1O0YJB
zvgnLFtFYE1MNkD-8s@X3phb6m!L+J4JUBo^_OM>BnNFwR;O6Rvk3RZ{hmW6db-mzM
zU;GNwSq=^kBgi{pGMV*`^QLQ<9~`n;EtyScoSmPsscTM8?{Rr`P208{A0Fbmmf0j+
z<L%pvklRoQj0xFDCgp_X!mwUD%3MaranIYg=U5w(!I#xM<R7(`);F|W!*X@aVs*{K
z`3d+Sw2R-3o{WQV=n=Qtco)=-%XmTAWe~gzKp?SGyR1UrKZcX_i-b@9o+4&X0g`iY
zP^1m3x{ohplHCYEF)TvS*HvWknS$()5H3Dpk^@NJjT&|mGzzVAgeq|;tnY$voyqW>
z$9ap5Nw0xm;xMKUf(UD%7zIU)>XQJ<ymNgvioGOOPfv|LPuM$J0!U*@${>LJj>Z3*
zmXrk`0gTWNBR|PqU>DnJ>>rj`#N?`A<FgJK4E7Z7>6)NlXN(>4TF2*3pso*$@M%s;
z_1&K)0Ks^y>7vxh9=1kjf?O-K7F1P1S(fB^9s#Pr8b{Yz)~lN9>jke~zvk@hj4Z1-
zIy&Le<A;3qoliJCm{OG)&iN?N9Q=)h2B8T4#M1Ra6Jh6@^b-RzO^n@M04Hw0zYFAc
z$LH2K+x<?Q%lDgW!)`BaJP^A9Ry((r#ylzTDh*|clp)wj_@Te{*^qjcf7h3jo~cf5
z07_{llNk^0KjP?U#@Xcx+a_PA=sKK%$b)${FLJUhM@kvf!TG@czAGv=>?fH{gO8og
zg5*b86%=I=q;K+^5X{(MZHUmH&8H!VOJwLmBXWsx9i0o3SqH}lJZK+qbP^<!s>viq
z#NOxry?Zz;P8yU@Oyrc)llutiLg3#D@;vVaqm4CDAJ5SFjt<kCUQ^_?@8|P47E{|q
zzqhh1(b9%?K?G(mQ)sQoR7R<1%sC(zMNAqdP{HvDaMaZ3*wX3jDLtIS(?xS|GO{Ok
z4y84DmNBVHv=($-lr(`LB9fAwYyk{Xo`&Bpu>0%3{5gk*Q}QB1sgnEm@AK?KN4<W7
zm!p|OKyZRqI4fB$8eV+$mf3VhRTfmaq^hP^w<c2=Md_IxWPJR|DXT?GU2nL4yTZ9(
z{s@bhWVu{&d2!8pwZYnh;5WtqoPaH}8SI}Z#*@G>0nmSc-*?|Nv5ZT){@6^r?;iHO
zb9r@%kb=5ilT8ZBJZHJtL_p{`Jvl)s!Ma|N>5Q&znNKDZnZ~p=>*a>4t81EiMJ^?i
zOpwc94)8dn(zK1CsYA{>XB@fCIh-A#q@t}GZ0E@`#e;hf+0->{(=eMJFsUjw>y|Hm
z^%V~uKH`&)KIPS`*Ib-ma&x^P%Q9xO8I#GBM~@!!`pqj=n+<uMb9uACI(Yu#B{$0r
zS&_52xuMG7=xD~9vyA%>PPw?ez&MZAs@EjjY#I&@=1ivW(MP8&mRH=|Tp^`oHk*e!
z_m0J4jSz5fFr}(W&d$&I)vtcV<0s$2C4hZLUl_A=x7J=Cb-3Kv@p{_?w6~3=M7ROe
z?PL3HIf?hcAx=yYz=ZKFuxZi~QTI!FluRbR_<czjFFs5vN@=W1A(HedzS0Kc*1{5j
zjfxJzo?wFV&f;BkMX*EggG}kepm$G5%%l*a2l_*`?Dtc>%b#htj2+f@b@9Fpf#L+}
zWXvowfIAqW&Cv~ocjdVKJ$RdP;11v}`CRRm1kl>1p=~>iF~N^@d;hTo>LC2?BAb_3
zy2S1|9|TTqNARs0`PT^%ByFY#IR}%1T!#g3J`WjZbf!T{ycZbbX<EZ()38{qSZ_9*
zR5Ol`kNN16kNEUEA2FNd<nPGl)|i&owsfuqNm$GUeI#gd4wQ-g+8q}Rj7X}l@N@sB
zYwW*zyE|=<*)a(ay0oXK5QU2y;bDC3-GhukJ~;aB!VvA+UoUQIU|96&-62B2Ldaqx
zM(dknHap;>4?p3fj~=jm`J8rHr)eT85QpEhJm?W-nZ{XYn~-nVL0Ik6{R{+Gs$_%?
z5w%63qK=(pGAYUP44viVMS)U^rfWl+_;BA@R^ojy|EfHk$4igUf`gL-%2`34<x%{4
z8UhI>GpceDg27ZsBPqN<XBozItXDT|)&USX(*|w6C>S5a!Yn50)%DTEQ3tlqLCkQh
zkHJrQmZJ#3F2L`MEDKp?ln8!%Q>K(uWMm|wD$K*d5l-S!2vnR8LY57Wjp!@U1CkV6
z73SeQFPKaw6h+M29P7LXZwZaL(JQIBd=1;WTiy%U{g?mbzvcLN&hg0sv)LSHG@S{_
z%LgC3)WK8gubjGG^8Bka1d8dT<Y1O_?;bdBXuI(JvdXx3|B&x~U$9v>{LP<##nr_o
zM#P2l)pgCq#U(e36>V!MVpvI<<o7ver@!}20>keDA%;m{*AGwZ4D5HYIH~M@w!QF=
zC$I3+n3kKxHCjuYw<s-md-e*Q<sq6T4%W77&^kJ60-1oWuIbF0dedOqnofY!imu)u
zW)(tu7B>sLmvpVAYb`|)vPPILgptq71D1A)QX20CU1urs67K}frp2^2L<c#=7r%VY
zgNF}z_}~N9>ou#@l4i4_Z93|8LlL#<-kzPYsauS(Snm-k<9e}V>StWtT(e@s&5dR@
zFM*EPe8SDm3Ylrja)RkB)!~%;_fPro!zT!7c>U%jN(pXmZbF_vrDB3|L!M_G9Uamc
zN4?(gw}1QB{Qe*OL#kOB2EFq=qa{YwKVB%jz1^nH`Y62<|7M8L{5mJ@bl%;P9Hh7U
zhYw>X-e#P{h=5LDBAZ8K;FO^E0|R=!G!b>Nqz}G3LTCg!raMz42?ln?$@p0U3l%pf
zBWkGs=LSDgt%FO%wl5;1v9*6w)C0-1_ua#Q@7CFPJg4uEeO`ZCMJ6_8CykZZ_k4t;
z_dc5Zob@<kf)?SP8QOVhjAy;s&@@q*|E@ljcJCu_mwxy0Tt2cm))IW{TDErlvxAEJ
z503cwqfa=VSHXu<X^!sS<M8N+qA0`0Jc?l#d0_J<lR2}d<KXaslhY%fJ$u4@uD(_u
zEms+t(yX1t_Vi+s21yo{*l+f%VxeE&KJVS*&foey0Jn6FG*H5_eLJ`zA*RspL*VMD
zoyYTf`WXR7V(I{J!C!9Z*YW7ldt!$j-U4=MqaqXshbR2t2S4QB{zv~V&tLu%7VRti
zHS{S{JqQEyIaHGZr8HgLvfen_m`FZco1W?03Z_vY3Yb)o<$}D_l#?p>X-{WN=QAd=
zDOxDX$%Mttg5}MEX59o6r_!h_=idEOPESvIOOBKhW4iu%3ta?VQk|nBWD2HK5$9DA
zEK%zqbnFTJIhZjg92^806``P`qoJW<LWsAHL7S0f(`Y(6B$A8?0?9->jJBBA)9?Z*
z62lg<6`yfVPDVyfK?y`F1|ewikU63UdxkFFyN%kk3Pp?T^8owkjv|#}GMiCVg4V=r
zAX)T(OzjA;Bmg=IO!y|!inq#p0lUwCvEW5fBh@R4B4;`|z*)t*HjxzwWuTQ6-b(~U
zV0*rJ&C%hU$B*t)Z!FHPgWztnp(rZw8Ha}leDv{CR+pNe*1w`@3_?h>%xF5exVYx(
zVnNqhnBHbe#!l>To59;Z8Tfu78ZQ{TtlN0!_AZCrF|gbf@EsjRYAx@B8O%9@$Kt(V
zvAE{^>YC~NfU3$lIz9?JjzVKi4sk(kS2tu@a&>jhrf#^py5#=pJ#;w>J98<pUCZ^w
zMTmAo(KH=cgH$?*Y1)t{vMMK7YlGRmwTMjOt!Leas2*=Uy2uc~#oJ4&D&z+(bcXVR
zo2zRUS656AW=sy}IOka`Z<tIbLB_#2nzp5BJ2vYzFU~%vDhtYT0>Yz|pfe8VE%SNK
zbW(6IJLL5Agv-kdbOu@}X0sWa%?9s-#7f;7PEL-owqrh@vtHM<9sKf_zvAuL8=h2$
zNUekVeGd$X(MAJfm)dK)zua!)aO~~;!&|1<xPaYd!X5g+ZlBocPadonJ!o^$<dc|<
zumEcuIvj_<M@MXLW&bxa>vz}zDh}5i_x<F)5bl$KtrF?zWG)h<9MH4bjIt=VvSb82
zn853PAE<N==eI%KeiPYHce?;^yrKI|i|gAq)z!fMjN5D1#Ao9@I`2Y~_~BC^FrGa?
z-o2)Cn9k6)8#>csZ5NL3FBV_Xr)u0DP6pYF{?v)Riz$ixZVE90^Mg50o;~9`pM8hJ
zSw)tGAf3r<#=-o6Jj-JshDRw$HJNaHa>}Qle!{)QJx)%JdGhoL_wUa`pvl)%dulT+
zHP*#Ifyi<AQJysC6H&3#{=B<S`1tU>OGRR%?pO@RVBF#QMOsS7cAQh_`@u0j_E%yb
z!EZmeA2WAB`_M@jd?iP^r!M;3rmyaTI!M72;SXzrKya3qOr}Tt*}wIV`03BSghuei
zU%ecnopTTw<XYf`!#G228`|iDDJ1v|dPUtG78ebo41q9G2d1pZ3o@M{g{06q^D6k+
zNhu@CTVRbtdVy3KSy}Sn(FaVT7GjYXl&a|K3p`2$Nv?o%4SAFp4&7U!<qVs5ER>}4
zCT6t>89TBv4-!x+21cN8SUTDeAQ55Wm=_P}#X_qTB4fr3MFFTl7>S^SkO(P3m_vtn
zWf?)8>Vc@}ZKjTp4LXDPt+<5-A+W-RK+-IW!7LeNnUj?nS?y3=bbqO$yF_ZsqQCiY
znT+w&^Nq^;$(^)QjL)#XWp!g%W@|z&SV@ALVdt!oK15b2pJQx?N78kUi;ESDYe%zD
zTwg7D@%$yPU!OCZ9CCW^0hv}<BPjBc>2yZhUK2!5gSY7NX3gd0f_h`Hhfw6IZ+!3l
zmQs3}?31s{-F{NHnY=APmnd=A0(bjzUyjc^oPV34=%aUfQ<CfR>Arb+gUua_ZjJL2
z+cx0rAa?6Qq>>PdDlb_W!&k3=$$Gu!?fDg}b;JEr$#gPhJ`bkKJTH()+PVpit}~i?
z&2&0PDn+h!U>IUOh417TW1=r)+yC8cHspE5X1(F^><vX(prk}dPmxJBn>E*$4dpC2
ziLRD6fTt)*E-udTgvjJgy=J{yb9sJFR+K=9sMAVty{P%(i|5RyQ&!V8UDp7Xo5gv^
zZyJG^);fqI7Zq7vB7sav>Sn`YvA|f(%a<>B@aRLd&=GJ9jGcGUM~xKZ5V^1&Q%}-X
z8+6oy1kCuDak&fFjW-TSQX>sm9~m*hzzU@x10h4U4i7P}l*brLT^s7g({vK&^JvZ~
z!+2Cf1a04+z3dkt=UlwB?`4A$OJHoc2e;L4WtbpE@SRJ2c`}*coI}Sb%5W_?>fhS{
zD7_E#erU;*IUY~rag)l}S8;coy(<s&`xG$(aw!wXzBzT=rfGgjQLx@`YZ(S6mOBBE
z^OnwRuy%v@Eg=+cPZ(UfwH_64`{q2o3CCN8G-9bJ&}GJCK4X4($ZR$V0Z>{|6lF+J
zCqxL|!lAO9vMM-ma~^&0fa8-x4h{}^aQ_}fuD|hmBZNSCg%lzEQ}Bk81m`+WazooY
zn|eRc(c=dlWA5?q-{4-tX$RBZuu&TMN2lW>yC`DVq7aF790B|-_C0|>8BCI~6OG}*
z?!AnEm&B-bz)OXcLH4UCG#`KT8GrbPf5_@;&Fkl{X)fGw{g4PI_`Bk0Z5yg1Ezc5&
z5wv4cvX|B<1bJDab%9nv>oQ3o$zVfAI1y4BmXomddyf_hrE;>OWIml!mH`lIsnJqz
z)lKkM8#b^5_Kw2fULsV;qiehkRt7B*I>-w-7i^P3)Y@a|sHr2Y%lhs=3>AUJMxDDr
zeHBbl6iAW4pp9*t;96P>!aFLmfFWdZchjNbb%VQ$r^Cn1f)Em+Wa!&cAhbm3;J>Ic
zfzloo_T@2vF9Pty+}RK{Ta^l(y`MeyJ(DYD5G%xs4(kG4zwL59>^@QynG~VgymPd*
zrCxg$H!aK6hJW+tKj#-ee?gHSFrR<Pd|on{RIILi(8jW1>lGGPL0z|8UoGjnu*=Oe
z)#uB6msvFZyldxox1UCQDlrVnT~p9q=NliJ-V5guqxuR(DM?k8G|dKQTBL8$x(v}t
zI9BT=UE7gGUxg5;V0rucC3V}esawv@YMi$i=P9cRC#R>hbxl=OJiPw_XJ>C&ua~Gi
z=X!aKlrWvl5yFvcMcp`>dQFk%G%iHAbxkn0mAR&>CR|?LV45IC+pO0tt}iLe9IXSW
zt0pBEH`k~_b8>P*nQ2Un_RF>8?b~y*JZH071fYMu;lZN`msi&~CurJ^JP(_Lwrjb(
zx}>hx2obVsWVztv<cOk}_F%opGMcubEOIU`FKFr-YYm&tikq98;HP2Tpw{2-7=hl$
zCV~ogcaGNt421~0v)y)0fA;{zZp(3`^Nh|q>Qm`{M9;d!lSgh>1Se@PY3q)&Hy8Zk
z7ti_CFP`)2)eX{nj!H=;Wi*Ydu)aG(XIg~F2=<l{W3k0L5wGie@Fy7F6A_t=S-h~=
z6hX(3{n9@H#N7tCx9eiNuC>-%LdN5=Z>*-s)Zes(=L?{2>zcPlZ0Iv7Jzj`>O)PH6
zrePx(ks@&k`?5HAtm|mZ5@Xlk6B4uAg2ZXX>Ti_W4%dD-&O3(uuw#s<KxPVEWau(S
zmt_znmXXyJBIK`j4uM4%8mT2ZljuAwZPUq&gZccOzWBH56@eBKB{ICX2%nscVnBf$
z`3mmVwDWj-yC3vNrA5*2d?*GVHMq^Prnxx5%w$Flt_<-wH*SF8-0AZ{&^&%5hCACk
z9(HKN-D70@ClXtzHA03c6p=|DJ$lBU{pmkswO;X)&;N$6t}lDDombFWPu(_XCunUK
zrjrEReE8`@_Zom*)Js#jKo=ERUZGThK!*$uA~cR^yD$eL$+MD77kE5nQJ_^prVA$7
zgfa`hcv1|8(SdYMfe1bsLwiu2*fuLH#+#6*RtA|8p+Y+u+hLu>d4ogIF}SH@#DL9F
zYbj)}Z)aj?186i8G!+8PFi27#OBRjczD7MK9y{h46=^qR@mShuDG>-2+z$fxC6HR7
z^9)_IWSv8|A@f8MPmq${XHp?3wBq5>jG_q11ioo`FJR|ruzuL>qzNoc+~aLwFcC>1
zg1IX?et2(LZ<Z_;3(n5oa&>vln>SZ9n~ZiXxxQE+<QaJm!ew+Vv2zK84;nFTyW!&E
zisfPrkB-8`Xos2W>wwn2_jcFuJ54$xAh*vXwC}pRlSw+y?JcHfelSC8g_43y=13*U
zvyeezvszJ>6`9V+w5BWy+NP!L4A;vQo2Er*NmUehsc=FuJvgG8w3Irh%4Zb1;LERm
ziIxGVnP!8{LWFENpL23@gmnS5b#2}IPHC+;J9~rH8Pmy}rruy}OHnA4)Hv6&UNx*%
z3ob7^I%`?iYcL;UyN=~@MQ1E|QLwzZ#CebH3|E(nAd^yis&d9+v7&8*q`<*J#%wmn
zThG<y6`g6(+M`s+Xd$Hxx_YG`%Qa2c(KHR~^%Cbj#u{`kxLI7&bwRM#J27fE;sl-g
zab$FzfKmARmTycyVF!`k@noO;%@UAvE;u?3i~W$br0)y<Zr`C4wdV;NA`v~}ZQFA8
z=A3{17k|b7<G=Z@dG-8fwB0#J6UT=qGqMK<6uF{lJC>^@tMwA+I-CoJg0SNqgOCBZ
zg=?nA9MD3j7uTo%>cNb)gT&CDO;dQ5;T-ANc3JGZshYdX*xezOZ^y@OSQY{xi0Ytm
zz8mf{9^(e!zV-XaZ!A>uU^Upfnh#o_?Pi5B8;tF6)&^hCBwD(cG8|4n26k~jflbm0
zJ0iy<7MBp++}V)2Pf3YZ@m5F2>r9eW1yW8qa))F(BiEYj*96fL5hFQ0Uiqy@89&mv
z46-@<%I@Y`>}N6=z}Rj#m64nXb+f~&#aN>;bGx+}8~y%GJOR=Mas2&W)H_}|31J5t
zaAG-wJcr9le*gFXh=YSUhX)7zAO83MN8FX`yXj^H>-7emdFoBaW)lNwe6OB_G5vnF
zciI;!TrLDnX%GRBG)+g>8oIWl&03VT=&TI6qO%}lGoQ`K^NLKDAtR2+BivEsXfpCx
zXAalX$HTg3@N^hD8Z5BfEU4QJ))`cup>qvF^y|Daj#h^>8W|aif)qLt45>uF-2o=s
z2mvJ-iZzamGTvX>NQDwNz<o=Lo!^s_YubQy{nLb}^R#`mE>W{F$lf^X$%={xA3Q?H
z*T9mxHB1%M61;^>L6O11B$&RBXAk(^CqLr;{r?y6;_EwDEbj&EgleLBCOUX~6-_-M
z&^J{;5|j*yvn8#wINu?JWwTkbSX^^)k#liz#m&tU?`C)#49m`2Hk+DE06s&AAu3U7
zjd7my^GnXpFL?Uwlsq5%D7+hF?PCpme4a7mcPEpOe(r#}eLg^2-xJ#o>T;xq74HyZ
zIwq4zOmYnPCI&bJ<|*U`4cRt~F(E<mavAF4@|vQYFr60^c}d;0AQf3rF*`m+dx>ux
zRav5x;QZn>>R51neGS;CFC~zf=Kj5tUdY)PgOD<);5)~1x#IGoMF`c)BiKUIwi|>r
ztT&#;%_WoBoSTbto=ZVhRa{-&kmVs&nQ(@y%S-TrrZy~>24in{^7I2%vPH)oPO4w)
zEZ&C+M+iq*lvwLnE|(G5l{9sObC&b7GqR`xuC+!g!TI?a>-C!HEQvjBb&Qax-3`Fb
z1!1_Y{v_pXb~f$I$G|PUl36YRyJ4{(0ABd^_GZJo5OFmm1r0A5pWiuH-7NX(Pkzcj
z{ipwofBkR%g05X61*{LCunQ(-jnWO))?8d&vD#eGXH)YEK@;~C5@*N0f@#6=qvso#
zyn)RdK$vqumY}Zd-Yj!#%!srwZjYJqm=%71wCG*qDb?4vZH9#bpju~9G6b50DczCW
z1%~@_kqJ%!CA6WB2uq0|XAvT8_s=OIH*Cl@VNLV~irFmUNpA;Bq@sq;lUtv6;Oq$4
z?QU8%NS(AehvI$MPz4F0j<&67x|Y&*WLZpYr!w>%b8YH6U69p83dr`XE+Tp*dD6HT
z#*a(SvfJKwY#P$0pl8xko%P^#xa#)U-M69IB?MA`AK3JLEhtC9G~d+6a~lwD*Qbgm
zzi?3MHzLjn!Ms=LijSUr#J~TK|1qC`{uBPo|KgwX`t>>30PN~bM{dEi^qT`gV1ty5
zsmD@!xoC~_f*^^);1lQp>w>?Y?E|g|0Ut{CBIzp2$g-S5m1HsOKnWGY4Uz#mjByjq
zqOryZz|Qs&{~ZmT?GRE@lm&Q8(>8t3jqwf>*>LMTnT$+${I-ekDnKk7ajV2cR@TsA
zsIe4SG#-tOJD^bG79Xkb&=M)!JJI9G(N)Gi=2%nrf-61(SBJ;?uny=f3ocJ(&gtn1
z53V25Y#KIegUXhi9)&U6HKF|;KRo2o$s-QuCmhV~@xved9`pHM^V<V<@s=4e9-&u2
z65I9h0RvLjNHoHS9gz@@))-#CzMyU`tHp+TV-eorY5T|)=d)0`KISTn6C?zXw_ev=
zoL{qCY?w_aWYM8>8(|N3J+Lkbh_Ur^n@ShA7N1nM<^r+(o&K1_vczp(5B}2Uqa9oL
z83wJCn(1WPvsCTIaC37*RRx`(A`g!0&N&u~1uGGO-NhN+J2I{NaCt2Shx0jkmXl=#
zg)W&KR5a~|llu?(>Z@O}-Ylu+mrSZjWQAZ-Rl!lVtN=kBeezc81;%z9AK$~;V6yPu
z(KZd<JMyd`SDGekP-I9iD70jCeTkNyX0xDc6jwJZGF?$GTAI$#Sjln~2IO+t;%tz&
z5yGRTL^#X+d&is{9bjz`(|!8v1J>&mZ{NNp0dfek8>?j-9C@Xnsaw{YE57{VS6p9R
zQI>~Dop#=%J`w?qb0PY*=foLIuS35Xz=RYsILeD4pez~KI`1OG<y{{g**AD(zr0|}
z$=>$}Dmm<rV4n;VNYl2QUtaMyfBU!m^zVN{-P9<hX^r7(1+OwgIa|=7!PTs8)>!9J
zrNo=CYIepTr9ep+-SviuJrSotCvu$71aqK8_#!MUI!N0~K)R5Tq3K#S>lH;&g*~zu
zdb<=pO3bLeo2p@ag$Z7y_ZaFp?ZZ=(?YcOO&EMY!z)JupILQxwHv*67vt$5X#I|(<
zyBaqN_&xfDcP#qG7`m<v5+_?ehUtbzKX+NW9g3^R-Hu>x+XL*R4C1Xl(Coo{0(KT-
z4V~%e+Lo?s(JDj9j7;C26NifvzaQ&RCh3^nM^Aj=xcwp^PN42Cl>cj%+g6<2d)l7M
z!<b0&C)?+_1t!ICj&Ba^?pkjkd2*$7A+UK-5(PnqMhe5Eta$qHG5_e_{vY!Ei<kWE
zkAKRWH|I1p)N7A5QNyrzKk13*!IC(7^psE7SThKSL*J*J57Iq2P{jEw0=wW(7r>l|
zfJpWx*C20`aDj^>I-VULz2AZfGnh-QHy}Z08fSx>O@Fjcff$iMir_KKkS3lbfX5_-
zqN9VM8$n4&ho{5QL>AG-d-btW6atNk8#<8`pb_!dq*&l+N$StzCkzrR)>=|NX{|G6
z^Ep*nk!3kqo-vzd%#I3*ti&0~;=1GHV9wJg&v<zMDW``I`Q*b-nN*W+)VET3KU1uc
zQHV2{uMEhHD21Pn0s@LwK0`Q#@Im#(Vd#wK`HKs#t~%<vrENjW4kZNMD})zFA4FAC
zu2~432ocWH)|Tt*HFa&mBX|W~AXTt+4@_EY05AGU&uuMc(8JYH3b1{z^kiGmao>dP
zLn^6p-pa8FHY~5h0+)b>(wgZs2>sS;$F$5bXJ<4`O<9yw)dc4(o6Q<2C9Q2aJAWH9
zPk1Dbx>;k5<K$q@v?$1=LJ3V#Ovv(zaykbA)%=+2>q~BKt|^KPsU%&~V(OM$=NzBj
z!y1Q)yZ24A#$jT<NJHDSA>~-iZMj(7P;b^qa<H1#)Epn5uv#xMZA+mwb<@&rHZ<Pi
zJHcjQ>6D;zo||PydAX!43);p}RT)|;oVDCsT?COz=TKhp)fX@5x{k$aL1#L$ECUe)
zjSr@CF3+#A&Jv<*9dF*e<i+#neDL_wu#=Yw#Q3loi7)R2&ibf3HC~i=n>nP23hY_{
zl(=DG521Yl+zr}gqlwH%|2!8jy|ao5830yqq6ED380T1S);xdtg3o{XIWmL1RBSeZ
zf$JPx)$sCa&ART$e2cIRlR_bk!nPhEZ4`bMXdN`VjImf7B9gtAcqg%FY&2j<=aIgM
z3AR@_nvhM!gud=#KuxAJjM8TbAx1t^+dypBM^%bIv84D!A;pj;Obk(Oqd1H$24@WJ
ze$z4lv@ze4UW3brCz=R%0IQ40st-J2JK`}Mm(CD!Nrs4Vvwi>LZjo5KU9gv!yz~;$
zPt6i6MEii9h`=s7!}sSC^tyW{=6`P6mbPh8GD8YQrZrk`GkxO{M3G~51|OCLxl5Aw
zz?gmm+_T1RfPG(A#+&b94o{R*dTamYJHB)0`O*e0q#qlOkx@n{q_=}!`2Nmo+<ouc
z_aFP0jN<+TfQOLrXM9IMj>h;7Efq(T1OD`n{*>omet}kwfB7$$ELIKcCFm?joY;=R
z?CQgNe!wvUPO;#Or|Ie;P>V>^xWMFwpbQzzv3P_E{=B)$(K18!o6_waM}k`8v|s7m
z;gj&sVlafLaC~s{2<v{FCzKl2Oc4awRg?>M5gNKEWRlJrf`D~2paxDZD1vr4S}Y=D
z^kLZWCukRD;solF2%sQj@(K5EA}~#!eF;>0T*OGO7d_9ij3Up_N~3i~S(e<>f`g+Y
zjt-8IXqGo?l+XG2!%zA6>34Yi;2953p9Ns|&0Xx?ld(&>R&gS>E_nIvcGO_jj8P#n
zk+H)-uIrkvbr|EwWf4u2o)8)`-inU4ryiuzD3#GP9joPrx(++tkk=HPbs^OU+k<8>
zlQEp{*0T-rCi@2HF38(wh8-U_F3CKTcB63_=<h`cu-1|11;@uHyjq{rwk>5D9n&Qk
z+hW!&Wies1S+iIySuAc42%5%$fX$|*Dhnpnl!JrApa?F5s<zUSq6li_WmQmB74!K)
z2wl%ZNc-j63xtpy9G+5^6SAVfn2zOo*-J45{~s6X+=U2H({*e%D;765Jbk+2<*Tny
znZVeVH*a5)XEoEQXSp&g79pjTu`%k&f^nXUi!1KkJHb22`B{TMQJkDiXuFoSt645r
z!CZ=?niPG`$GUFNSrE9aCMBjbT;JT#b{;7#ZM(skHXNVoedvs%)Ivi9b?D;TayQ%U
zJS;|IbKux4%V@DrWxRP9ood{B!qg)J^Xj8Nn{zH|JFPiCzvSTHfZzM!k9hs&C9ht+
zVzF3ayN<f{TwX2N2&lB@u#)6iG%v<zvIHJXXZwwgwM2#@01^{Lh|@^vckM0#7T@>l
z-ALB)EBOJZbL@_b!PLDSU@?HYk#601LGXU-wQn5$ro|J?P=Q5^o;7|ff0CG+<bybR
zjiM&w&Bevh;#`P?187@Av#F^!8|tRUMy<P~rRb7B_r57{AGjl!U|n(<+LJ~~I*D-L
zZP<9&&IG4!DachC4{?8jqw5SxHb@~Uivpqao$RAl0$ruLxw~fjTk4tY+xNLuZuOfq
zPj8>5KUu0XwZFfG?+jns&H0`BZ}1@<`fl$Fw*%<L-+5QLeT{hi{=A8+8w{BJS{7dt
z=m_k#78y@J_>lkAfAYWPJD+^U{gXrfi~s9CWxeni3vCx!I|D)ncMJ!E-(>=I3D~!u
z%jJ@XO+#J;p=xU_oi(&whiqFkn#@La89+;v(7o<k!i+5j69JQn{%sJ1eTV(Yu<3P}
zXcG0b&4%T2!D_t@3Ixtl6k*NEv*3FkOud0YGyTl&BS;G%F|cMq&o0jJ_-M9ubaZU+
zQ5{%C$zdIVoQOdm0U{;+k-qg9R&mLM5{HRif+hw}MD<Y-qG7Wv!#hur1@(g2WXhzd
zxOe|Pli3`tbG#F@8_T>n<@=xhkf$Gf$isV2IGUbPF=6xD1a^!lbmuJi4j~4!MWT<B
zkg7~{vL8HJy^F}&2kKYLkPCEsgl=y|18@uJu&ixaEH<o`s~!NQ`$~%ljLkD)P$m*>
zH-BY+Uwamx@Z8&p<9ET`&e7Y{0>GA}Mgn)yVLbxbOl4c9zf_$8v|75ZiQq5iW^u*!
z)eX)WZWfCm*n7QXv$mAEXF8oSnM_HErGrRjOh_(NH2Zj}YDSS4F`7_PJelDf<XK5M
znNn605HM*r(dVY7Yg#0VqRdH7F<sjt1uPc}x~}Ek!^hP1ipiv8xm@u2?JG9*2B|8Z
zKYzvb%?hO&ItyhU+Rb*3>&vS?(e@X=TCuokna)FwTp!ImvzewS3i3RoZJS;Iv+Y{S
zY|82B5mGuXu9r9|XuF!XXRiop+tQ9L^e-Q$0=x~8fa6(i)E{nv1~GsR8`+KB{t|r0
z#=#ofewMe-o30uAM!XXr{Y#B;7-N}Er~Kg`{xQaMtkz2|FVFeni!b=%m!ETU@dn#&
zSk(<yLXmk2LyXQ$?LRQsFqoFoJ$o0DkM{vdJNKEEncam<#!MBr#>!4T?y_~^Z%E^D
zcyEW@L-d=YG5Gto!CeAr{mq9#N(iGEL~g^zVw-*N_!v<c&G<_70MR)YHQ$DIHpXC#
z#l*}MVMjqo6?aQCmpcJa@~eXg>{1G`ebwRIARQ8?T<07J3nC}abBa8tD1u}~mgN*h
zNm*7DdC`N9-7<8AcY_;6h87)8g|X;d5Qhj(+w}Xc+vguZ&h~pS?mMJ@vh)3MKS{^l
zV$CAhN{1<>MjO2mvGFx!m!aJ)y>=aGD;GC73CxK!PX*VA!8{xOD#!_$6dVN=_l_U(
zqaXeeMPAah3;xBw{A=ExEg}mCYzRiJilJYOv8f;=rV+4QSemxO`H-<FWld>K3yr7L
zQ#x<SA|nPw5#WZvSS6w0L<A-aUJ7U!Up72rM@x&LaSh9Q#d5P|6$OfG)1jqAml>+i
zR0mUJrty-<$_*^9i=3T}JerT-Dk|h5BeT~L(sOzoEj5~wj0}yyV^QP~UxSa`xUWf~
zcj#zE8K7Xg2je?B)8U;*D-BubQ<KS@4?YNE?BT<Q9G#rdtUKy;N4@HJ`tVaedG=lI
z9X(=J&XGjH@b@V11?++Yw;!UveB9NFv7qL77n=q{>ZD=NIq!lSqLLv~H9IfCt@or1
zNr|OYST9-JELkp>v~7E9p>#3%y0syCFTCfr*oECmVhjd*=k^^2VqES5cl*|sZJM9a
zLKMu}!<RuRjWL$CZ7HgPqR466u8*iQCZrtFnPjnCVNB4NtLu(D4~zKx;E+5IU{@)P
zkRim{3%I(z#u!VUmrN(q;2SksY=cCAMrIXWgv=f~mq=Y8yAmaX8Fo6I1Z_d@ai+r>
zgH$EXTPBl3>dgwN1P>lM?%jXPVzuDOv(I?-`W0V(`6XX|^@2C&S6tsLSS&Vp;i+qj
zQjlee*(@YVwl^I`Ric%r$^=;!a#$J}+tzz(XXtE)MA0=JQVH(en{l(=uxVOezkZ2x
z79n*%pY<DnlqSrFAO&PHeb57l?e_P5gy>jCVav3b{QhE_?wSqzbL~$4A_4^G@iHz%
z@ffXx+|09QAMx<vL-IUF>)?<0^3@A|`jemV<G=YUe)f|e^YV)?*fcGxjVIThA{XSD
zKnTk%IhELFz4ut-an{gT8QFLupc?wI=z{^|7LXfWJC5pMx3U-94RyE8ekSI0Twi{m
zx^FGg>Al}pNl<^E`8O=kqvvJ&?_`P(&ygacBfd)gD85Dyc=sN6_x<>_zIUd!y~`?{
zK8t>!Hs&@;phabxyvQlag0d(n%aW=rsfq%X5YBH<6tV9hduM$i7;+*DxCP+Y5A5!4
zbMa1mnauy=c|H{>WDi-w+uIxZZ^F#Uoj$Sm^Ypy^abnbWMF02p9i;n65J@0Vlv&07
zlZUA9N&fwR@83sI^0S}+lIJhp@cPXvWSvOC68%lYpQ&@lxPZ<;(=>E8m>ZpUSZgpw
zA*H}s8`o9S3%<MlACVLD{q{L=WGOlo5`Isx%LfCjYuI#aF0RhGS=_K(uUK!^)Llze
zmchI)C0U*$q{0a>I5Zv=;af2F(t|D!bTObIkDx2DdKQoH4J{oxC55OcC@Cl?k-=BI
z50=<1j$td8t{i~mm_m2B-r^E0XQ9RJKX`y}K`QF#=$Mn!d(^9zdezZR9giP8<!FA&
zq?n;)i9p{bi@v73Cu65j3cL@YPeO_i(dGx+dC&q<gv8{7^JiiR`eyYmejc>8L>%zJ
zPOzO+8$z^fU|oW2iVk73J^K{yk2q27m}#P}6Fy>V5$`ts=hJ(moqr!0G-3yKO||<b
z77@GhxP*l(J%8N&$*>a(6ThWt8dmEK%f*tmX<4n-IBeWCIhv*=%W|Ca)SH^7?YOvD
zpri^W>CSU_5E!@Pqhq90be#bN>-C27^9!!8u30RWNU51rGY$?8na>ZXswrhzQdBuw
z1z~9+CC0fh39AerZJ8>M?JI)jm9ZTNjj;xyG+JcHsg4o7Vd!T0gle|tpjmNzddj26
zk9qv`1utH_=I6iooG-rkishzZy>^skMwS&^U9CgA`XGreq(^JW3kfPLilz&y;^ic0
z`xRNvX4#^(qMB9YlPO=lxZ?2WFtr5Eba5r2*T#}~oQy2lo=JN*Cn+OyWZEq3`gchw
z(J^(<_?R&O98C0~-wGe}4Ml8o?*rp)W8`I?g-B%Z%;$4XPL9!9GM&w0`Zv$b^@6L5
zORk%98Uvk^7%ONROOeI#Cu9Twt{*$WBpaNhjSK#b5Zo>DvCyy|fK{5bw$0Ytphk>x
zV~)qyHhVG}KX+`P(*5o-bo)Tww_g$vALyLK%E$zEdpDJEsE_E`k7ma4yy2p0)<y}L
zv7crFgXy{B35SI_{dA+QxO?0ledGQ+arJZ7;-Yi645=o9fNGwR6?rs&1z=a^1>3Fr
z>(pX&H6W}DeKL+7;K_90_V>hgy13+4^KM*sH@Hht%1QTd0FLpm#2)r*)zIcrM)>s(
zaJNsdaQpVQ;oR_kdd9)J)r%s>PX@RB;aghd%qm4yR{Z0C=YK*`PWX#|{TKYJfA!}q
zR$s8XF}T(Twyt2iUhHr=uYy}f(>AnS8$g}6y<C)1vIlk6*zj_RHyfa7HVOdoICKPc
zNy9IAxefeU=NoFXX1Q5%cJ`K=#ez*;Q#UnD(}bYG*ajhov&0FaCtI>CCx|8DD+OK=
zgD8>}5jqSlEf&ktEm3GDV#Y*FsYHc`$l>LPzJkUj`p`wsKw>aPc)4LSoaRYm`}@W@
z9zS_PuJXW$=Q$_Gr?jP|n>t*hcyRQXS$Tj~CBkRH&+k1qRPPDw#HhiBkrgT|P2RET
z(vHY&?>>m%K%}jvhZH$D+N~wOCnC!bOdl;$cig*wz_Sk@addnboaY&smGs?VA&uAH
zN)MEL{`$RmzqpK9gE5eMm-)4S&xg9W?aMJXdByhvn(QfgCnUyLUcY(En>S}D4AxoN
z)}WM%xnUJ;)6uqu#my3HEJaz-nGKUkhEkd=Q_SZF7-Nt^^7{1~uCA`h^9hs5gvU=l
zWIjK{7|Uw8;;ZK`sOu2jI+^60-n+-chfg^;JVEOM>s`#%myv0O>pF;j4?!ZtA$&43
zDew-dR1js1?Vwafo>z3%ka@?!!69wip;a*ZJbZA<)#94#>l^ZN%GK3|>zkUo?y#L_
z(>kUTO;(iXEaT?-0-Qm~0FLr9Lwb)CnuCK9r4`k*U|F|Js+Qy9(+E&Qj8lqQB}TRS
zxDW)uzs;8HvPY?h?y^q1e-jt|F%z7!qU<i-;oPnd$na?clYk(Bg>Kt6$lR=!lx1LA
zvn*#)O_|N-%nlBi%;)H$!g|TNaZDR9r3@q8ffs=>H^%nY6okdz7#rmcWMtx_|4#(M
z^drdk4BB|iB?sX&hMjYC)^63!U1Mk01e?J0mfzy&JHLG}w|k?pk0k67<w6pJDMn0X
zZ!%6yT-*c?Rs#`Pd4r8+Z)+Xar%38(71;;ab{{)V!)YeNGnUud_ZZ+k*0`W-A0uRY
z;NU{vu*QbyU^#sLn{jpHTOQ@oO>NtwyW4ZD6e32U4;y&K<FuzMM}La`z*{9T&*QI-
zPb4EF7M?%cnuzxh>S0@VF8w~7<u}z+90-u{gN^5laJ|G;1?El*#RvC3;XnBI4*2+!
zAMo_)_mKMk#9!1u=jME~oq+@rwYA0Cc~1aUbse<p+LldS<74K8!dk3%7-yplN}8iY
z%%vI(n<@ELN_dg#w(C=T1x~A_;i|df`sR|Wt4q!<&dIZ!qR0_S(wQKA60}K!+d_^+
zMuj|q>Rko{)DjTyglMu3S5l;0J2w7nUB_a*;-<N#tSYKv!c<Q==9GKX9I&Ba3LG$K
z3k5%iAkq?oJtSRZ0e3_rBZla@4t&V#ofiei_f9FZGFm`G)*hJ)DxHzdCmc^7kTXG&
zG2|2bEi-muybP(ZcKsZHxLAlFbV%|kh>MyVCce1R;BX-gu<)b=FL=S&4CZl2WjQ&W
z^4;$}<$FK)kbC!!DT<<>Fl9{9loo`rAP>8T-32(^v%n+d;M0@LO1r;*Gt|6$OuFvy
zS<+4G0iS?Em*iPNQ#TkJB94uN&IG@@Obgm(gNOhy1WHU%Izwy2q?%$n!{Nc4`RpJ7
zCmHhnJ$v>sC#MgYOs33cGbWQMD2eL~O<l8Et+>9q;`x_9<CmZRJ<q@Vk`F)n4i7(g
zf)F8ZVAtsor4)U&tZ9M?)>^a_SkoYYdb44%xS?qqma8>w(_yS-U2j-#mNea(#qyf7
zvom}fJR?p|jyO1+usEL6SwKkg7mDjeN9!DmO~Z24U|d7jWmHv0TX!g_G0u`_1=bm^
zuNIWWgvqpGvuU|mESXGZJh=Y>Qficx0hC4`rXl@Pl<<iw<aq5IZy1I?vkfxFKSv*I
zEmGqR!q_b!Z7~uP=I}8TFMurTEWYhntu`zcOE#ODrfyiRR$N|Qk>@$9)spM0YnH1;
zaKZ<?6f_3v*3%h*HHyv|ymlDxBO?f1Ob{-mN2$af1~xP>ZUam8?&AA7Ax-7q6nv%N
ziS(TW>=K~fCqRDN1>&~{^tPwf!Q3u{3|SkZrdKlY5ue6HW^NF7j@pnSfTyOZgMOVg
z5$yOr03(R<vK<YZPCPD=fs3xGm=@e_8nJQdyKksxV=O9eeiH+mqWRLsY&=_j8{jVJ
z|Jh#0Zj8wG>HSF=#WNzqoIEaj-MZcLhvnnWn@yJpaz#U59VF?3f2<M#5{%EIP+NUZ
zjZAez(y6=qV#>BK{5|~K{nqW<50V}-dUMHm?cQD{f-!JjJ>UnQ&UtkILmof=kpIi$
z|B^rdi@)N>fBW<81oH~0=owzoLY5PvPl499bjFa!e%o~&#)gfE^KR=|(jDbw=*Zc5
z{OD(bsxE>cL*pB++H>B#e#QCaIhWU0EZ0l!o!%!?Ioen@n@yNURdf?53+7@@PC-Ua
z9=WxUeZsM3jiG}c&?TrB(xN#!i1{!w36n|1b$8D7`hx5AC6%7QVTgz}v`m;&u#TqL
z02plylt@X+Qs9IkjAKJ^4eD$V`?od(e@v>1gZTlao<wPwNRkPO_ZsPPN;wbpCltb|
zNQ`N?qp#mp-c!d;Mi-=QUom2*N+pqP$^4k8blXRj4X+Ru937A2&M*M1gpff9hg6>9
zlR4l2!83mMhadC!={=^Cz*-~)eD6cf((x`o;HllkHC<z12Bm<Q`T9LT+}$;`yNyh{
zzOX1Hh9g@2J-s+-mgP(*Q*JJo*f!Ya+s1(30D?R(P)bu4LG$iraYLqalvH%S<KDdo
zWLZX@<v}-RHfKIRrYz@3q3JqD+q57A#<a9ui_(&llOqmh-=*2qtT!FT)VzN21=IN<
z)7cDXC3&8sWT4}nb(ppZ16>H#s~~N1b92qx^S9J(gFsQ1Q;Kqe%rxbs1mQTW51CG;
zoWDHd^_y3`eshJ;j&d5(8yy@@I5;Rczo-$h&$QODS{sVoF{vgLMc(TsHFd|PTcaot
z1Q1lWHJyXU&%VpO)BAWYa5k`@&W7aU&PLbXsLSC2?`1z-jdPl&ei459m=b9KBD;OY
zr*1Gh&$dZtJ6Uzu?8J?Zx8P-9MXke&z<M1W9Y^2BkgHZoN!zy6b<Miou-YtH*K3S(
z2%X`qr8CfV9+xS+msk-rsCvdp*xs!$1Zh}|3FmFwZis%{Q#^wyIt5P*?DT$id+%rw
zgHC0tqa^a&gJ1@x?zWHK{`7v|W)ZhQ`XIu-#q6gqB#@TCQEz$;?h&1_7}EyUqo1!5
z`xY$@j28QSjSx+--bJ6i5X*Ku+r+Ih0?{LX&V4V}M~#1rC7GxrB{^72`u)&$iGbM=
zOvWxE%ze!*7=Ybf0BfKBo7nw*Y?>W?ON{=7cjc{5+;+<ikKIGk?QuW+e%IaY-g^?)
z2%4f}rdSX(2d8C5o)`S_@BIVXrel6I2j%$X-+fNAjKMY%n1a?IbI=u(6;xFbgZ82>
zV_ffz3HpX<1K|ANT{om<8*}<DM!WjHmICnUSerGA_J+5Yuemrs=j!H?<!TwFb^__)
zy&J|B0wolctf;8aWRV>#sh9#HiXS^ReOQB!Ei|a=S-NPN6)2%8WkFGtY?d49P0fZi
zldMAK5~~eX2D5EWjz*)%`UGVm@-~Peck$g2Z`;zkmezEb&LD(fR?aw_9CDZ)k&6N=
z!o6A_grTJ@(6XcyRWGp;tc-#4e0Q(7UEUkm1!w>0JlLB#x9La3NCv?Np}!PD>|$`T
z^gj6Er3H<i2oJ2VRCsIg7_^4ddvkvGhoA6+-~E^mo}4h76<aBMlu8k&!`^4hiTj<4
zjR-O+VZ|GE-F|ot18M+<(T2x69UH|q<B?3WW5x0iM4_0?517plxxRQw(-}7FcBmDp
z$g+&}x}zuxrqdZ^UU6x!Lnt*Kr4%P8Ct*m%Xv!?hgXyj7@ZQt34R6oRSS)XF&W7}0
zuA?k6a%stR^s$SRdcD3uE2yRuoD)cqg@`9(0x)bFT>QCSt#P*F^yC;XK<k{mm{Lt9
z_}*OQ>1@kpv*P^aIluV%-}B?2{uQs@JjeKs$@G9M&nYJ*N?)T<cp)*S#+Wc6ZR(a;
zl?5}w%COor2p!Z7nx<n@chrren&uoIoidvS(Pv}3;PmS~-n!n8EZMTcR4@W(A-zg)
z+$Hqu(4Ir!L~7IBbXJ16*-BrD*BQ)7$yaAMM>0t{F9@dC@P3|Y%Df<pvM$aClSFG;
z*6Sr6$Km0e<D(-clL<<PR6m`E&IHB|6jFPfv}hF^cLUQ?_i{P?_z)grf;g><j!=oY
z*)kw|A2qLn4qoWf(J0~}>3xl?Z2d33uWx#Dcw#U;$;UNrkrK=4lf>v~97j-x|K>pQ
zo0k53ai7+MjtKI4KVT_hN~VFWOF-Qx(<*}_x&t~Ja`=Y4mti6DPHfFNqp2)a>vr+O
z59`1ihCHBmJyZOG578SWU!g6qm(JS*yI;Es@4O4)MR2EvkPb^kSFb(DE{r{ejPM~O
z?hzv~>@lC}N5F1>1j*P(E@XK~?~jd!eedkQm!ekXXrUF6{TY92A8U6Tz@_)nh#G+6
zKn!eGM7X*ho#!^krwFb|Qb=;8_~h|tOy?60kLEaEvu@sS@!HU=qMzIpiUY_?C@RVH
zP%)jB6h#)ssThbHYl4=YHFoe1^{zK<rbyJJ>*gYtl_FyU==IdD=BB;oZ1I*?uU~O-
zdC79Uq;4BfDrB0lK})Z9$%qVgP9A;VlBrk6a0iLOuwhNz1K-3XikL&zM-dhiL7E~b
zXI9Nvu5ReshI-R7$)+#^TN!-rDCsDopi4$Z+k5u~CNjV}gC&Rtb~LoMq3s&FwnGTP
zyqa@5xyJ!VNHjGyjc>88LpV()C*-oC5R;x66sX>}{e8>(0Xwe(bL8s~@ar<XmmoVN
zDIJbP=xD070!e|EQ;cos{Du?^Y>{Y!`=pd{7`eFMd4;e7?;J|EJbrx4kN)6OKK<+|
zPoF(xItePw9i}%Obe*Rt@>?B!YbUnD+>J|EoQIuwd^dEU=+vIhmzc<KhzQRt#_|Wd
zGn42Up?EPM9xo(PW+*8s%Q@5O0jqVxa%C}AL{ouB+7Pj*ltQ?Uqj^rfZg9q;@^JmC
zs<?Om1c4yWGnA4v%?7+<bF<)!FQ4<`&0AKRn&+=xlI0rHwdA?t(S!ROPUcL@lGBqT
zjt&n&5NPL4;a!d=3$(HDY}y*rHEcFZx~8UW8ZIv`>Dn%4LvZLU$BA?1hjYs56saXb
ztKfqtGt_j#y(f?POmyUvg7ta<B16i8`Sb`W&&ZUB4&Kqk<a4sDA}drFUYj7dgHt#a
zOch21m)6coKKbN(WGcryi!ojF4ZNKOEcKRvyJP&8a6KUNu^oi~FA_tv-KWM*{b9U}
z`(bSB@d7%U(7m@|qvWH18$(|zGEJ374KWXh0)r#789I}wTwq<p)#YPOkB`Z;B+E0J
zwV<_v#)1(Jlto5a0}}Ht=>5im4}a*ZU0Y*KgC~eRr%hB!Zton?{1}8=oe$@hDlV?x
z<Gn=-H9&_RFhoK`1bufQRAeE8dCmI}oNR5_^^2GnDP*FsAt|{Sv8B%8gzKa3ev=}C
z4tjKw503Bw$a#|DKzQa@hsGn64?d=uzz{YLIeFaGyI}g40<Yqp-dZ}-U|o%MhLA8k
z!#a&~3h#omlOHp0$s{`_k|g~CM*2C<Wxl6$*3nrL#)3CE-v&cP@|?F0pS1K`q!Utb
z+HYLDp-3;#R?zt{!wNrS7gAD$=-W~8sZsh|1_MKNy8wPcfxg7Fe0cUpG~RZFzkP=7
zGH`L}>Pa|X>^)w`Wp!Y(T>p39ry%-djjyu{>QX)IzvF?m8?$}m?}cos_h<L{;~)MB
zvq^={J#Q{v@Z!x^{OYSOc>W3|hj8y<$^A!jCX+d{az>^@^r<zDu5J;zKxPHDwRBBK
zoz=)%GGi9})o59>iTBaOy9uw9M$E<I-qvq8yL`*rv$vdGU(j|foeQ<05G+?~@;plK
zlof5;a(%VHbTB_U!OjJ_C{SoZ&Z?GR;_YZ?Xjx)nAcKzaxGjN6U9+O5p`xH5M+i9L
z06#HgM)Gp;lJhsO@g^{F<c>LpjvO6z?viv%*nW$i!E>=D3@y%gSZC;L6Wd)zCN9Gv
zXz;8%OS6DUc1%?rlFKrh_7%Nvop3lA)|~e#?*;7g(qml<q9s`NRWt<z6?=%2cU<t4
z3Uscu8PWw5nk?T*g?iXQ`q9kcv9>|Tjz=Gy@WbEx4&VR&$2|Jr9`pH>pv7kUltVC>
z4McBFjf=;4K#l<2ZcxN-<_Tigyk%nBso!7%9y{{x?6i8iV)R~Q`J|9|A)^3s$SGG%
zXWT4SA$6DcIB#(#nAk-Y7PI+mPG=2o&fem%99A=g5KJdij*pM=c=Ei!J4e$rSY!A%
zfBtXyo4@@DSBoX(bk4f1dH&T`7}HV}CC{Ed<cHt?5&9$qT(nKigNF|yOX2BEhb~LB
zR&=e6QK>EK^@`PM$?|5w;_3zvygj?bIPgM{6*)J{1-i&VNv88T<)q}~<b*<LI@3{A
z6(`5XNFi9PHsGKrb95%?x{$L|Ah6C;*BwQE2srXwb9|f=^sqES1vB^bK(JYN6y-7B
z`|kJAN>guY5Qg4=Xiqe2V9|8Xu<c`fxA_6KOoO+8-PQtin~B(B8l(glBEwZoz}=g?
zJuK^nu3Lu;0C`SU=zbCxfL4mCs3?mPo$099E-^OfQnkio#lT1QT5Y2)Ixt0!u4}1d
z-?`Hiw?)(sZVo+z7O9inO;k8;#8M;%Y$x(@JRTJx+AV2|!C(fyUDv;U3AtS(s1sut
zad){ZGIw0ZNrJ(8gYzb6dHTVGg$Rqp7E8zI_rY~>>xP_<d4OFIqwP(xUvs}hiYOl$
z*)$IP@PA~o$G7rr)N)`s6-p?OVQ!N$yeE~s)s{gXCFqHbx9a<WU5JSspw7PH@AxkL
zcl>DG*k<UqO|Q4l3@J)<`)aQLJpG=SzTs{LhsxXMzPskPxj@$kF7%mPM8*BXN0i^s
zsQsEB|NUQbar2gCeaX*0|D3by7nIYC(~}bpk4}+3X4A0_*BFcnsTzfcu9lcdi?dl+
z_q&ePH3$h#1l}%){Su!M8d@wJ#<iR;FL?gyOO~r8byJgNin<N$C1SN1*Yz1~jB%`1
z>zF&X#B?2(kDhRNaL8;rV=|dhQbZQlQTq*-i%YH-S1c9_7Rx0FI6gV%`1pvEqoZJP
ziZHCAKnTm@$wMCGr<~4D_{+chTb@6E&fB+ddGP2Vr$;Bu${Cd|DbSHiOW;1Vb_jg3
zIA_>&Elu6hwH?MfHtU)<R~NJ$%!(S*Ic(>V<V^At4$6BJstQ2WhgM6B^RG*mzZbCk
z@S{V7n)2d><JHRxgs=#y7+fQXq5mOB*tzI?<5URub<y$Ed+-L~W4ewpb0R!K7^H5v
z_u!D<`Q1<W!S8&=gNMgdRoPGUTB{JU@99|>=WH~OKmuj`erp8ib{BKluRcEh3EY~z
z_eSUKS;TG?9|vx7ZW3W(S2D0sH_Mhh%P6zpYt%F*&an=qbG#J1etU-Z9u-?qDaGmO
zX;kOV&??ZlbzSr3@+z2LS4;lrkN=oI`LlnE|MNfjCpc$$c>f;%{lEGPo;`iWXP<q>
z<=I=hrscvJgcKYe9Z^oE10e7gXAQ0k{uZ0fhV%1tuFfy;ju6>eOem`fXXh9E?63cZ
zuU<R{tvEbBLZ~o6pFDodvnP)@p3kW3hAb<>F0C~*#!=)MN(ClJ>kQC%VNfb8f?1|f
zTH$Pz1`#3pC@*rP%9&0wzW@C{<l&=7k;V^xY2FXMTf3zPZ=*K78!-6DJc(Q2;$0wP
z>=QN~wo-^-u0|kzOz-7AUDt8FSaE%`<mJm(c<*`g_z4dloKluqI9^DUQfQ@_R27p+
zMLnI;tZU3_L(@8%*5F*wqVkb>3-=;O+PF#o5oGLl4aV5rE}fqKRz2;4vwr*u(VJ9K
z{f__M{oKY#+VT6*D!z9Nj_dT>D%)+ji?$L|l0=i)N<lU{sK-Xhl-+0W7&FIdX;1aT
z9p=xw@O6I;zSh4kSYUVj|M0^-_Af^B@;fT(*C?USj^}B%=IFi0kAPMGEPJWR0WO8(
zZ<nzz*uIhg?7+|sz)tQ%*LQq>NTD^f>+nq5&$7$C{W>L56r+0`)b4Usa4<RMyB~j_
zb-kjzthicE5SirO;t-)cRT;ERO=s!aj;;xrhO7$`C(6p8Y1g%wPSdrTrl~2~Jk*d3
z>%T(xT7D{;NITb3cN?ykSJd^U*LIWAMiG65*CDvl8ABTbWxBSbuImsPYb<AP-y#ug
zHY;YcIkWkkLgvv8!qfSV)n>tZwW6svv|WP`0_%b@a~7GfB)=k%l$1ybS(b56-Q$A?
zk9hg!HOtkK?yIk8+m@5VW9HS2QspQFnG9>J5P(gAQX#s%?K-;FU|Nec4riciEF0M&
zY=aURg_=<68MFL|LX{|?Lko^X?ROgXy~=w5yFdMh-wTe7qT|i$b2`_J#1?n`6T)I)
zt?x}tLQ15UA;~Xc5fLe5yPuULLRjX9itm2^G2i*#QyxA#VKOO@N<=eA6gw6`OluXG
zK^M)U+W>7B*o}oL$C0q(vU`rbJMs}*v-_Rha*MS{W~bO8gRF$)!J{X9^2v8NfBS-Y
z<#DcKz1h&z2Iqs*_3JlhD6NB7<492X!Gi}VrO5LFsZ{KEg3V^b!w){7m`wTbqmMW^
zIN)GD=jW?6KRN#iPaZ$uvrj(ZkN)Tn`8R+5-}B`czvAflfIQE!F=u4q9M%|&G?=!f
zZ8mJyYi@3?SuU4cTwP&}<D-v0!8(DJ8GrwaUt&GHIbR~R=fjUb;YWY)`~3ai|2;qX
z>EHAD&wk2}e)s##rxk6}kY@#1)={@LMUhdJikq95q((Tj4i2@8n+3;551CAJ*2^ny
zHrM1;$!uOBrACMxDGPq|qu-}2%FtILD=~J+-|8cKS*1JL)5mFe2T;7P6TcY!(x>6`
z0w*LIjdhO8t1DKG<tIP?C4cvKKcUb$-~H}qeERW+JbZA<d|r~G)0GH*n0X%L0`elq
z)-|2=fwA*3o1=H5{w(^1J2~MCA)<C-L)Wz-uxIy?Mr6)iE)HTeH{8Yg4d8D$&#vB|
zjG7a@tHvGx{cZc#_2-PI^*f-_d53gt1Ei6=#n#>QjOf6Wh2Ri-%N=0&$~yqn9_q{!
z+VAU3v5A#+W6<VDH`xP`-WEdN*xynW_w71joPn`?+wM2Ez+JjaiQIQ^5A5EtCF5NH
zst=wTfw=J<d_DAvAb#(421c|Ge%Sit>nr(gNBVHqRI5I6mw|eb<g()Ny$`u~a*0<0
z;~iO*^T0pC*uau$8B%_3sPV?pb~ef`MSoFi=-QUHk?34fua=Z$9$3{Z%F;x)h~Rjl
zu-?#^mQ}N6xm@CXkZIBxtk#B1Yj7c-L)QeQ1#1kQ2>{dEa6V@Y>(!ETDQPwvmXigu
zgE?hZ^ik@;U)PbLndVb+m4g%<lyfG!qNIp`+V^c*kYO<hVL6ZoeERrfOlN88hV%1_
zkoBiEv<EGdsv;A?=Q$_jG^dMW+W8^>c4u+M;SETaA$^W=Iik&&lym0Q5tDpQB@ZB)
zXZt`-#Axir_I}T?_g=v6v+q9U?VIb~q1Jg5=yMTrwvBpTV4#%1S;8?=BW!T+35!dd
zoa5hkL(0T3o9BG*2Osdm-}{tjA3o&ZV1iVUAWw4s(1W}njS&5CqQBedR<;%mpckfm
zSG3pHEz)E5VGG#B$$hUGFP&44_EIALrX<b@9z1-)kN)5v@L&J4{|e)pko;FDj9Fqj
z%cilxEbF0eT8<AV;VC^#r&E+tR8>V)g}kW%2M33o9N%NHT2r@%*KgkN*{7e9Yt3SD
zg*P2lS+TfT1g*QrA4D_fF-oc4+}w2?xzae>VQj~0wPdkauv)DstBRB3Q#N(OqsPzq
zoge)nUp;?8mgju=`~{zW_8ABB13vlq6Hbm#_!s};|IPDX{fwXf-B0-b_rF7#X`~Vq
zS;5k8$h4rE<S12RT^FL^oW~0K^mSR5krxHV0&Pp%8P@9>V<aCu`jC64579d7r9B1~
zce@R@+6aGZ;@mdT`(bbW4XBKLfH9R|CW(wha@z6&=_FnVtg&1z7JUBIEB^AwKjq^5
zlEcG8e*TNkSuYnXK77mvkM2=TGF<0SN(cFkl%+)}ybu`YX$=@F(0W+hV~vc)OPJ_<
z6tAr}8=9sK;?r8*s-Gc6*|ynq%iquqo3U|p@peAgJ!52d)8%}~N;oR}B4dBk5;W7I
zZ|mrEVZMvfH$H64T#D$AS#DDH2(i`915mQYBD{^-bq4E96rUcRXDl(YKejTKWSSj=
zxD+_?HNdVHN47(BV*;JS-4e~T10y)Hp8rNAnPT^yYuiUGy_fvv2HN0Vw;>r6750^}
zX>|Yh$Lv}Hba%t4{pWu-c%=W-jn4daPc|HN*C%Nf3w&LU+WE$md;I9*Kjrl3117T*
z{`Tj81+F1eE=Zk-8tF5HlVBafL#8EK36uoyTbefbHalrinIz9L6p~VfXxdb4^gUa(
zYZl8JZWcEf>o}MnB6J7>>ueW9_C?Lc)O5}xy+;dyREnYsl3hf?MhS^GhULu->*b2g
za>;yp$fTMwnNB%AIpO~4DWwcXQ;Vmg?&qx(>LQ~lH6<I%2>$s_z+?TC*=KV;dh(2)
zefc>*`RU)WTCA8(XFPcDfJgW4A$0~exS+YV8KjAbjAf<*ptg>((tLFPQyv^VX2Kyk
z6F}n;*b#T@2}DooZ9a0o;=OubDDUTES7^@OUh(?n1>Py7^jO>CgY#GDLTMrvL7zoR
zf$};6I~_Y*e0doeI|m||eCG3#?|ttXKm5H<c>3%E4iD!c^_WPLyy`n|Vgdu(F&MVC
z7`AsXZ|^d*?tVTlTRZoilvCrg?7Lptsfxh5_8?G56@m){eD4Rp&(DAU3toKrxA@w!
zTsGV+I&2h^wH~bLD61UPc5F6lX0ut(+$p8dIzuZ>8Pl;?2jy&r)+G)&JUrx+k3YsV
zHMVQeQgC*5h7^*6g9GNXAZTozi@-b#ENcwjnz(yfVXY<0GLDXB%w|Va#&K|T#Le{;
zA3l4=zkKzQgV_`%;Qq-ebQX-|;pshSSCmBt&ZCrOJ`00$wb?+|a&%DfI=`W=9ZChN
zUkd2jj;5)}^f8^z*=#z@($QIuRtjr9N|wy$$FW;_bf&OY;M_I<*=3uO&f;$0=%3#@
zhTOkH+m7`P_m#WN{GxBOlp@;;iH$)FQV0$YX8g_%zDw5{%A%mzG`xB9hU=>f&d=}j
z(T7i&PbUZ=Lry{gbzKLa4-c&kdUj3c&{@ihvK@gLYAFJWC@<2qElt}5ab+LJv<nfm
z$v<%yfbFK^8$64)$Im`?FKv=;XBmhl;?ehZul?I9Qb_jZZ5=;P^hfdu0d0LYuzk*j
zK#-tq6{2NRYOgWd*Mq8kmmO1r5&gx}KyZEj+LD-oz1HKH1*6X=JMt6!))+g&0v$3G
z4cR*o;+9U__V<Pe62Xwk;P!EMvvrXA`L^89xa>FaeiPC%@o=fv?*n%O=a+n_($lCO
z!rpzA!6d!^7!{Xv-}7~Oe{5QzDsjYz2NUjn=P@Uz_xZb@{vB7BZ`mxb@xG(24W@PY
z4tyJQKb;gf6*p?u<Bh<Xj&`$Vy~>elAb2vJAqX3Ybz5_FeTA_Gm1W$YJ;a%?S?nwv
z9vrc{St1IF_KHktysLs>w{;k6sfvnBXHnoi&M!)#wW6#lrjtWv)r?6sV?H@xD&`a-
zN1<ZSMjn3#)3}b}C((Y<=aB^;;#-BKV#32BWA@RUdq*d{di|Q^a!I{jfw+&7AqY?c
zD$nsY>P~tAF03;O$fUz`0*pqwf=tZ;g|nfTN;Medlek9eA>+gk?_1sr*j-&N_{Gnk
zvszlDC=kNq+@|l&>8_-dcq#EIk2o>tSSav;G6>x`$_5I7lp*uMvk&j_J3st{k3M?J
z;o(8>rBbQ~cH;#Av2A8eW|1vs#{2y{!=1_FE>r8fKHpp7eT_@XKRfsfIr^-n5#S}o
zc_}py9zEqx{`BAG&6}?{KYxWW4o46s&XnjUTa5@hXU<cW6=hkXweEp#rh{3|d5=S4
zohL6Q6x9UpLcU(p)GV&9sMkwc;~`T>JbA9S7!hY3NDD#+n!Ii3nufM*14}rW@$k_H
zcrVa-LDSTH_q(5R@7^)5UcF|uTr!<bS>0S=o0eP(rjrR(nUQHtRaQ(VGeo^6&op(r
zVK%SGGC|jQ5FVw1TcT8gw(XEoP?Z&#QmodN%_cBSItz?lkyjDy1ZUJ_X1whOHoWgr
zgY1+ohkYzU+y;O9%(A;bx4tuGmLroZ31dg1kzVrf;UP~y{+PThFrC44p6AbB@OR$w
zi_d>X*KL^3=V+B7yr9T)$|9$%f_cQXhI+9IV#S@Kw7v&sB7!@Ys3E!C&rR3SHVs`D
z8N0+JMaIV%+t<ropmv<DYFtm-#|*4ps;k7RyAkNwb$j^6WAfK1+d(mj`5#!_{`0ha
zd%xvp7y3!$hDI&`_>mM8K+JZMJ_|BD@z`;7rage)`8{Dt$L)Ne?|i`iPJ_?oDDQEk
zk;A1@=810+I|~u~NTrlP@+zHQB$<$%{?WfTX6%xm+%{|>avk5iBKF)L<CyfFbz;x)
z+xL8TV8&fwO!WV@BW=YU-?^K;5j)4oozKEqC5l2-=rZHelkXuv)|?%_=Hl`-*Viv;
z)f!v17+dq^%?psWXR4(LdZ4Bgw6#QMimq+LT9DUhl@WFu!B@6kuW6bFsbx%w>1f&@
zT5p{T+KWh(%8+F=r%Kt+GtPUeqM|4Y@+=SQs8UgwUy~OFRW_r_CREvkNj~F19FoWE
z7a>HVqoeHwggq7&u}4ZbmZFyf@fN=2ghOSDgV_<zcqYY!x@(w}Q>tu2F0-&b&@E0T
zi81gd+H}0fS%K{UCxY4s!Iw^m?S8x4qN$JH_4(J8_X2j$zkJQ>mp9n12wPD?Ue2)l
z@9hcUCVW~9LkANv`-BuKPHH-Y|B9fEmt}&-Pw(+NzxxRvee{r{qa(5`$45tU0lf}d
z$m^Og)WDRa2~sK9FBZS9^<dvZ4Y!$>TZ{WQAWcOCq{D$C<_ERttU!v4-}}8k24VPT
z|Lniw^Pm45>m5=_oG}!c4gk*>vZ@Fvw5l@b+4&&blIg7Xw-QRBvjXEB#z5BvKL)I&
z+pMvsLn{>~bRU3Fk*nYXn0)v|dUj9OHsC$gWP)DY^yh23j!8AA$P0|^!az11)3Ts#
zTL6?)TwGjGuV$2GINziwIXONKIRWt;9UjnhYbKMNER!f57DAoH*hmo)RU2z5%YrPc
z@Wzm9!K$%5dGd%q`?LQ6N@hr@dQ<Ee)Qth%mOocYp*3o&aewGdj12wZF6MU^)ZOl6
z+kGN!R3s9iG*vO-=;Q%M#|M;EhV49UW0)2>*B2Lj_0_N9^#l7MgrF=-9zTA}bUvkP
z3@^WY$(KL>Io5Twoy8cC8t{W;$o4G_i)&{Lo4Tg!TCyVRZ`Y-a7&}p^dtDdAzzE(2
z#>Ni%TXj5UyLYbf+XQtG+qVx=^nKrOBla$&3>zrVaZcRce=dT+1nywS=#VNj&0C4p
zy9Mmh8hzKq;u-ww-lJMXVqSg9gyV0uW1369ZBBHs?D#mhz`JJZF^C&@?!ljI4EaR=
z$!_0{ef4Xe=58)9eV)!YKJ(bTyB}DJJ-?;viLECYo@X3jA+|n$S6xA}jNRTH3`L2I
zN99u{A5XZv_W^ItU-IJll&)Ljg~7U-^?F6qEkQ&-UZpTT2=96tbdk|E9bM7VH7%WP
zP%63Z80z&p5IK=)TP|1BbscI)1n}oQNUaf(l*sbvGbh8kUzQW9vZ5?1v{vL<fz~-%
zW#}xUpduFqIVBYnrc9$wortimre=d6tU)a`flm`y3_)GjVthyo*13+3CdjFIM<#Rb
z9o^?(ddTJVCG+xtiJHVf42Mzc-mSq}9|Yn>7~jqWaEBj*clKsY+4qimM?1f(yce*0
z@$!ttq6RYXE#8T^OLg>)BE3l_EW$%4m$5PxiCz>03SLRU!~4hl&i6m#lTV&-d^{)5
zGlu-GVwecUC?kL!?J^lNcKxl~vETaUkzRWjnOlq3yDNX3!Jr2Z{gzO~4(`xOV~h={
zz^WNP`ollt;(CE~is!%j9M^8}-l8OE0htbIvN9brO(=zuA+Ky%hFo;Pp){mC%5?_9
z(sdSZLr8H|R?Mpk?`#0jBBYn<x-h7HGBx{<SJIe}u0?B2o@G=;Nz=5^e~17i1r(XW
znvU7D!kHY;1SuuWriSA~l!WWm6#`F@Wt^OxAcSCDuPMp`Aq2VB<e5gw4rhH}-n@?*
zXNIn6n3Oq1QDW_aw(}^d`N0o<k7v(5CCdsrV-QOA&ym_=AM+zdOa%;!NAM3DvC{F=
z-9>Pp&)2~0z;34-Gj=|0h(_nli-MvIVxbC$$|N|&!ED04d&fL^@|3bFm`tX@pUwwA
z%a1<(n7ZB2Y&tZWx6hx`EIOLj2XnNFtU}x!g*8?7?JOb~OU$e>U2h5vUmLPPq<g;$
zXzc>syMS-ffZPH&<H+Tmb^5nyehrt~F%`!#oEE5k_b>z`xFKmd(bO2|(7?*MLEzd)
zQ086eU444-IIZ3)yVab)T*?poP3AQ}_ys!W5h+U64Zy_rbNX)sKu2&Q0|1jEZI;I1
z<*sji{h7BB@!jm*UDpY&i(miTx1r-Md-wI_UGQud)X8sD1Mgqv95bID^WpS6@ZmoM
zz_Mb|yx?ki#V^17Db_bAtr37y8A3^<(sZ3?z1pCpCeu07=>#PK5rV<$w&w8onAK*(
zn>TN`xVQibd0tQyVV)OC_7<(I%$ZE4<k6K#WtyU>D2j?ItC%xq${gM6%4HN3WRz%v
z^c9q`o*arbK`~bX3X7va)6r7XQnNu4^jbSwR&K@mW{o$FTxZ-nx<{#kHlZyYRZ(&(
zPss_9z(K!G(aIJhIy!F%sojFZtG6LXY^JivX1=@Czpe6K!0zbim^}ZIJh!y9LpX(W
zCC(X~8T<-7(X1rFIRmak;E^In2#u$Xo&+t@NoJD?pM3O$r%xVndU`}wD8ft?)Af+j
zH-)HLybNmABFu_HdZZdA*2Ln$z%qpyaJ+DjhxPV8FqK<u#^|si8aScH@H1%$-N|4O
zCyiwCi1!h4AqMeZ4<H5k<d7fz;h#~KQ~o>k&w2j&-(#`NDupu*xrSWjWKvLO8JScF
z3t9x{SEVuvRUniI3pX-2w<?LtgGQc`0;MDPvl-r5)|;!4BFq@Hj(!AAKx?U+RSX>u
zI(8DsmBt%G2F&s-NKtx+%ryvs)d~}RkaSfs%X37QQOzs5ra?%7iHnWQG(u{!tfDG9
z%3PDFmd-)01DJ(ilD1t-yV)S4S=3p0^5iN1)Bo}RoRi}ZLKuD4;^S^}%U`JP8#^p)
z^t<-a#ucXNAkr&{!nACg4)>l%h`yaeEBg=uJ@(h~{eOMG2^*upu;#@Cgrw;_UTYlC
zIg8Mm`}gnhyT9{Yth3B#Q?&HhC_hwXQ*v3*iiUYLMd_TzLhA}zXVFn19&bY2nh!o9
z;Z7__6$CLw8vt{=z}h18RVfilAXKawn<CO}|NC}5gl9}ldjO45)opitCBW!itm70N
z8}%Yx8a8R1r$u!9`)^i4sx{Gr{M43$2<`^@F^rj%QANegTVfx5*vuxtEM+th#%HwF
zjnZ{RZ>uN<?9*m>+)wsBT2G-8rpxx5Re@)pJWHYl)9<%-#ot;<(`absz;mT8GXx&&
zerNc%`}Z!}wLQ0w0~!3*`QCMhN!oZD;O%bt_DCjiJL|;P7oRwbc#eT-+xDs3jtt&k
zlDqvkFN5{SL(E*Es3_zK5AHwVi|3yQbE@^|GDoYR=pbYYJqSUO&c@mj!D6JOMk&d`
zY)(~G=uBfe!{zlASL<s`lr<{KGWe^h0Dx-aD2tq;EScp8XsM`8N9R10K`1}RQ&3Pa
zK{LT4X(K#Sy%=)Cifare2p9W?jo$>{z_y`jHY_(w&TrmwetE{(#Tn0EJm=|yC;agH
zKj4#RA5(}TW~D6<2xj?|IWq)7L$|@>DR3feQk)Vv(S=rZ4r7BgOv)jn6<e+S+bQoC
z75kmXTwPrA-~EffM(GYEL$0$-DO9hF?4z1}V2LonMj-GZa|IG7KzbW|xjY<Ka~_>c
zIX#?E<QgTz@DMSeBGiM33#}wc=L=k1w9_KH%?b<)F8#d;Mu2V1wxs+|?<nI)IS>7!
zlLkhAy)k6GgIO5dfD=0~9H0D6oEO}`_lSc(nUG=m>q*Aq>;?7mlD0NvT9D_8gZZ5K
zbcX3nAobHAlOB-;&?baL>l}f?bSW*5juWe+>kK71rnA(WhPG`2(2_bxvv>zF4OiQ?
z7-OP;mZQu=dbg(D&^8;CRun}5bsmAw1X3v4_+2_fIjP8V#ab*d&eC=v8M?6!V;n*R
z_E8E!roeXKj3uv%kk7R3XzC4Gg}kOZ6XY)EKlxApQ$GFlyC{{#bT}EF4k2wXm+`K4
zi48c6y)B@Ue!mUoZk>0B$r~2??J}4k<30rsju8B!HO5-5u9q?Uh6q6<z~Rv$QaIYC
zK?*_JHTak*!#jaXNeF#-5aXdSg3gMbeH0@4DUH@WFCydQEDnP)HEpw|Ya5ixqu-0`
zf$#8qJHBJ2akg8}+Y4nlJZ%C40CG<BV{15v-}-(U^C68{m=C0SmjUgYm%X4aS|WU8
zDtqzbeH%DGzQ^#E^TBKz%&ajQ)edqeeRZaOGJf25RUw2u>(`_b+xq&by`1x?t@9=G
z@%~HwcFVi3whz<|pAG-B^ZPKIlLhGAC9FQ7q%0Q0|GW1ycE1^*+1G))?UiB-PUX(u
zx1al6Wp}jjUNCQ5XBs6+4(2Chx}xdo@V*E^lU@WDDG2(jTICVU2{NTAWMJcFvl&{e
zkk6bP>pZLVikGim^VRd0ygl16on;&z%{VxmF`dqsR24;8P?Z(4*@m*3QWQBwUeI=)
zjc(8)CzBaBaz(DHUI5$^wCWr-06DwFdP{2>TGLRQ4N6K@t0k-DlAD{5+U%EK{DN1n
zU$S1T`Q+Kh99D;%6erA>_EED^$O!auK=k=y+ui~h2p=UiLIi(4=M73rlt}Toh7b{Z
zn_l`h%liowfBXPz4R7AO=Er~i3#4lCcx)Ha(*WQJc>smP%K+q32^@yRqR2257^cEA
zqoy(&3fCbz19=u$I3J~tM;qnwOcw$u;<-n@H_io9;k$f#wiouavWcx@c1r?Wcc14s
zFN5z%n21!KF|AJc&Ue4ZdU3(^;e_YE`Z-;lF`rf}H*4mz10Ftjh>{AeHP)HF9-6j6
zbb?$>(NYDV6CK|}^6{V(=d7jBUChZ^<9)~}8W<{v^8qv}70k#XXv5L99a1W)vcg%5
z={l4YxYk4ommsPv6%<mD6(yav6k4;{I9jcP){h9htTiV1^`-W79w`)37`o0Q(xCCs
z8ONrs*)$D0EBJT*-G9tK`bYmGimJjof>?G;2PaT?6`Xqanb-HV!)^9yKLeXS+xqV4
z9Cu%9H2u1Stb9yqC9&4AS+3Y@HgsKwQj+O(!nB%@Wtz!kLZ&rMU1PeSFV$y?h}b|R
z@lT*L9c}9{ArKislGO11M#KBz-UR|1!FbnoNSXEb9sFIzmdSJ+DVzRI1m!O1+p2|;
z|Kh$nPr!KB4{!IJ-^Nh91=LBv4Me}O8{OlWu?-)^dmI+;v1!>F8*^=%XbMg2*%)2Z
zN7HVRO=HVwny!*>-TMSikifeefO^s?i)Ps%gYy3M_f1OjH4;+n%(~qAYxkYoOx?SW
z-u`XOn7r%ZL=Wx;8F8Ozf&Pf`x&79A-(<PoW*^fQzh~}ykT?7_{^q_iaGk#?T!_Sp
z%sH5!P!tn3?UJBe5&a?&;3YVL6u}Ie=Q&C%a+y)e674j{2gfL>$YjYx<)}h&es)IJ
zwtV^J3;yyipOcl2hYyeW;PHL#Jvimy;DAXrp(-a#r)!F$Vm6;Mn@(BREsJs$f&^8#
zuFA509+OH@%3wt_uA}Q3w3oC^AS>FgVV=*Z>z2i4&2qhFv03rv{ERneZ#kYH@$AV{
zJ~)2NX?{vcj=-bIVz`BhK^LA5fuhAn-VTUj@v=_^>bxb(G+HKLX9=)toBjQ^%X^w)
zXMFt0Q~ux&f0v85mwZ{jrt3P4H6tIHFqsP^2sOHWhx9?UMM{mfmI}*(wB+?Q&DnFB
zi--8bIjSs?-uK`xo&ZMX7O~^=whQduRg#a@aBLbx@1DtgzomZnBLC}_@pFWTF@^WR
z>*m3u$Ncd>__z4-^PjL;FS&p6T{g>W^6P6ng6VXQ&SH+jm}S9vi;Yo7vU8|RpjFTe
zaX30-@y<lExW$+a&NT?(f;N%%(QIT9ba*iYW|+?6osWKef;`J2uxrS480<jE`xqTd
zmC6Db@L>uH{&_XVm=N)4y1^OU69I+A*-$6efHfW;2iCf7Xu6K3vwY{fzsvvZKl~3l
zI64e|in{2N#!D0)i<D}M(fuZVVf)%>SB_)+*=@#Z|7Y>t(XmMgDNM?t-VMgMFvd*S
z={Xzz?u_Ywuj`toscD<GH-|!8Y?MT5g%=@DXxn*=jk|3Tj_v*bJ#i-<sU{KbqitK-
zwk6BU{=A939s{fdUQ!=+Zd;@6u95)7UG<$D{KrDayFY&$K6hJ->0l{myuXVo9Y`7F
zB_cD|Gn#&gdNnq&(uUTWpn>J$G2Vl>p)PRSrqVIE8y8PlC%X9hro{t&WOEl}gp{N+
z24gKIf=(YMkKaxaBI?b7*jnVaeaphTGO8%t1?C1Yv|YwGzYF-@1@6S?s3aFVd}~)I
zdH1}#_rLFcZ-ce;e*{%`z{IbIY-!C+NA9D1zey2Dne*tuV`kGii}eMqX>pD~5D=$n
z@8f3DMXg!LD3~&3Ca27c1Kx@=K#_@z2kHTT@}oayHb3Ct=!mBNzwwiwUhw%ZE_w0l
zh9CT3$#=i^U8a*6c^=$9*6TIi2~6vdVnv?klto2ZR%CgB2aI=kA(+l)$WmfW7yZHE
zgZVS;#L!Y>>G&R>asJ?(E3VN<e)8gPnNBK}%LRY*2Y<kyKKTb6%R@?Xz)~?qlQF~y
z)C6zX;Fs6Y60$%Cxfzv#&Lt|7F|e^BQw4Ga=zQO&z8x#WZh0?YcmDPUDJPs9-{<k;
zk9hUsnnl;5RfQDma8Stc1wzHyFj33Fw5p;HK!KyeF&CDyS+jZbiq)Ifbf>3O)dX2q
zJ*W#-*s<?L@~<22EiJglm@Vb{eXLsgyWjD9A6Y@a_qbFNP<Ur_o{xLG5o@vUe0SZ`
z7GQLqEDt&PCbL8GOraz+^&0Iv^uvd&H!EJhc|(@v<as&BL|B6rk|GZ&wwxzu5apUo
zrbRadJm5NP=jhA^=Q=`^a!8kAOoKBP>kVFp1l}niLn}>L6s*>37K<CE(<z=KoU;fY
z1sa9GI2#2VhcNI=X@qbgp}UKT(c2Ci0gH2v&1!>fE!G6i&3Q-HI=l)RO6F=wmTT@m
zc+7wDzx>a5`s`znI;OS(;Q{~=-uL}yV1RzT6>A@$-^<+kp|_6P_AXF4@(vh%E@CiB
zU_P@fqbRblEC`S7WCRR00;<?;;(ONFAax^xRkkr6V+a8hK8Rrsm?11#E;4p0;&aT{
zSu;w5we!sT0M~A`>%Qk6%&B(=RtyZ}c4TsI9*xH;xj*Uq`xNI59vjDbJfHB<=Ma#Q
zHWLx_hz^0F*NPAN4M;Dror}!2#rqELG}eK05?lym@~*|(4qS`38@y}4#+=5{EqFu}
z#mZp-b?<409m;qv3c+s4^&!0CrFz=%+bjJUMQ~gHCfvm~Pqz)^w!zN64f<Vx@7-6V
z=jg&!)7q8Jk<OCDWuuXvAw~Rt_Hm?K#<oQ(J&R=bsfIfrL%xAaOlKJJ%>DI!f7kB=
z@B*AZ&F*nDzsJSR8JoodYdQ)UQgGoxDvxzFb*ov`8Iz+!oCtb+l8j?<h$N$?p{2tM
z&{px}^dZF`{Rt0_Px!09`w74J<>$P5^Ag*xD74_g$sr#;{fK9eKji9s!SU>v!|4(0
zP0MDr;^F)Q?#~}!v0S<vZdNNWk{`}~L@DQxIjrBX;EYr82|_eD)^x0yazx9T#$B`M
zuE{O@=#$@JQcU>h-bWnE6G~B{qrYK*Ndee-Y}ly9e-TL_un6JMUa;D%Fq;NiOQlNY
z(<3rHMNmM$nfdK7b*a1;u={WR`CoE!_>jd-&1UVv6(~6YC-5$w%KMNWMC+K{S%D2c
zUBX#1Z_%z}Dqt=oQz0q6VRL@T`IldDa{7?TWFD9xRU*_V?U(`|{eq+yq)dF{2|krw
z;ox!EUEIf^$9FLRLCt^r`+G~Agx|#=jDg`kK$p%fr3V-2Wl-pH&ixNQq~5G(SC`y7
znsak~8Q40VQ4|Hvg$xnC3?bLSgcKs*g!g254oabNjcqjT>IT~xoaxwXmPjcuZG+Y-
z0IfDig-Gv0$`~(tC+}%BVY6D(tT!|Sf0UccOZ4$Eh0^FO!wZ2{0xv^4G@(LNv^9>l
zYauZ{&ePO&SO~pG3CU`?jL}jK?*&3hT5E9zoR^f-2}j4r{Ad50|D5lB|M$p>3C@Aa
zL-dvu!Ca7hhIZMyeE@G?`|pB@T~MFcshviCr)hBaJ3>a$gXo)@XF=bOEGI8Y@E)Zk
zS}TMH>lEIan8m<ix|XsmDT;!{ViDvvAn5lIz?~Bz`-Cy!B{R`4U9rc-&O0m_fhmpG
zFeV6U_L*Kn9i*%hyZzTEaO>}8FGF2Qybr+_q2CAd@tA!~X8B=k$Xo3Ex5wHA)<yJd
z<7k&H2KFZWmzY*h3J6J$+DFMc7npZ%htK_JyeE6A4*@OS8@7UMQc3pm;c@}g{aU~-
zK4}DQyEP}GgJ2{be`{rD^xd)Gjp51|%<Zypznya1mu?KDw&CBHb=$qC1l;ag)Q8XZ
zYuoJdd;8BZl<j-;-{|{W=MsWKOgKHh&zq~~oL|2VEVGSkD*z7X3~kp^w@qN0z3jWE
zBxA}UiVcc24K313a$RsRoAcqbr>I==@X;}E&(Bybm&~dvZ1#=iXm-FvOgPC;Im(Y|
zwPTa5IW8V@nBB*E$JB3FPu5`JP@SMD5yIoB$jB&}Ajt9L<m4C{*7XfH%_SSRVOmU>
z7E=z&Lyof}GP1}fOXRKgh#Ge>etql_8Ah_D>l#egg@{;DP?d9(%JB@YY5!jZ?Eats
z_rGQ``4xD{)x|Y!6Bb-4)UdmC9%nsHd0b?-ocB1}f$hkAOG(F(6wFnKHIWuv?KuDH
zHTUknWHOr}GmXwQ!beaS=9y@ojnV^%(SDDu1$TEL9+$foZy2OWZZAG!i;3IExDAf=
zvOi~H6W#~tC0#xxk`|G|1(F7pPk8wFBf90BX(7lm&FSeW&880Qm@(rg#o88p%-G>9
zu5FNI8B9jDVRdssw_LDVT(hb-%;s~X3fJvghqa)j#(Noq8lpMW1xCYrk1>X>YkTIU
zstR<Lk(UKptDwl>eGsG83hQEBbit?2MBlK$+BJA<sq31~bYz*ttvj3#StU&8X*xqS
zsrdNgAMhXj$Nw?E^Lu|tmdysX$$OlLUNXYd2iu7K;A6LM-1S~+%dyz>U8dJi`?pub
zaUb2yPwe9=DrLwVk{1OsD?;WFV^Asp9-NPn!#QOc^^;7?W>Z8aR0I)W=dj*Ka2Lq*
zE`nkqHOa+0-~<`KuSgjw1~BZr?{%6)f6x2J+3u!imuyP)v_1AhyJ)Sq*!FRq_1;4}
z@AoF(c#q#s5k0U;I&1MxLqHD6+%^Vv&~s5f3O|oF6CpE&8#Wox{|ulYnqi~s#omph
zAH9-qU3&d`0`;+t_j7VdN9?y?<qp?o`?}xsGIuYEV<znD&+@)yn<v}%{usI?K(KqP
z=#ScejWOeQciA_$+y=?vj0uva*xUE1XHChy(+AAY4#5Y%Ut=5yNiYt%xGC#s+cs_}
zLzK6PyN#R~A(<B#Z17VPA<M*kHs|B~84n-bXSLbz_U&74u9pa)-fX}L=F>Tc;*cYG
zf+!Si(NS_j!4U|U32<dY5FV~ZHJLNvn9xER@4?craF<-x7hJ8cL0FFF#~c+$%(FQq
z#g>Yp?DYWC2F%4Q775q|mjZ{4W7kIOhQqs{-J9zQkEF9<b2sc22PrHu>=bViKEo&9
zyu25%TP_6ky1}`YwysIJUA=1w+s1SYfi4%ZYP@rJ8zN&XVL9+E2cn=76732E3N0ng
zqUFsmpEEl+B%9AEXVcz{79!(aw893)PRKA!#(5`|Qnz;UyJdIRKdRZiAnaBp-|-WE
zb0j_ncS;6xwngBD!V8I(P|XgIwgDRE^8=)Sjb5=@tuS2|!E{(~j5Wc>%qg&jwqD`t
z6>YP@)N5|e-m<<tW3{{?pUnfkPRM+q0HsAts{~4Dlon{EkwUUuE%45xr6jK=%qA0V
z7B{?p^BOM%)wIGIkBpi}N(gkGBV&Z8Y$EP*LD1D&OWOpqU)Q#XXrjf3lwxTy)Vko0
z{`k-M5C5b81>gU{575~Z<6{OBB|-$L_gmtQ(HVd=n7)qP*>&jO$2fhRpVwaQVTYOR
zzmq_K^SxPBsleWOI}*i>D{=5^wOj8S)Acz^L%KTOv)V+WwKE=LLS(FykyRpzB!uN4
zu}mq4sP};d?z#?y9_nNS!V=is1-y4LqT~8XX6FQSZ!vW~8lc1;wq1x@jN!Li;(R6t
z?alGBD`Gu+=e98(9wJb)F0h?J5+)`Lw?lA30FfbEjEjlHts9{&G5R?Mckmm5y3w8X
zb?RbU^r@TrsN?^CN#(AH*D<en`?Imn+<w2RfsOo4?rrye{BL+h%?A%3^X2RND3Q}{
z>L`JglW9#BwFtfSA<NBbO+DEJ-8K{*Eg2OVIf9Bc8*-J?(b77N?K;}7!#gO7k|$4|
z@#vAKsave^eE$4P4rhmy(>ceLLQ?`cl5FcP1^FFMnQEk=;fkNTzvE(c#>TBsRY6td
z92Q5EqC_CESfGutT%ySO{IX$|w9yC~c2qLLUlBoH5N=mE8(m&HPu+B!bys+^qN@YT
zhY3N8QmK$>NQD$-N-44|kGgjXr4>3Fh2O8$-lI695h@G*Th1YsL&y&6I)qA5u|Aqi
zAlWj)#O%qGV20qx!&HV8#ewNlw4I?;ip$q;dGYx#5eGB$q@>aksU_$kX|xdGIwH)f
z@x5)vAnoFJS&?vy+Y0O6`F<iBm%To5V*Bq~rc}Ru{`V<kP#4Tm!RIOYZ6F2CS(MHx
zstLkCt~^=_rqzV1s(AJCC0*OnXdUyWT6C^a!qINl&?)LJL<lv@Yi=&jxO)A9GRv7w
zr)aH$1cKBQxsF;xA=>uv=m@1G^=3(?Go*xz^D`joj1kPKuiw1JOM%KYhX+UGvcTEk
z@1^iT3{Ht&$IE*c#c*2?A>vjmMOBtaDVR<(T5H*C43qhc-~WR@;y?f2{_lA5>|<1x
z({>i6f^Qw(ji$2LITDe=Afd5q3g2z#aUa@e?)yfd?vD1~$7~^pfg|I*p<>GnmKZep
zdrR*pfnU)t&L)mgA6X_u_`bE!b(Xg4P$H<lyRAA(`f&qW=e_U!?1YRF(SF;(f81}!
zQL_8WuK%1E%@gTd<A~RNjNQ0>cG>pdq$D3Rh#MLiSzpqEoPJLzYcSLFY;n(m`H{ql
z!&ry2K3H6VF%s-qA3vCFhfwA8Ap=Pm)}OGS{tba$fA4)AN7)AP^~}-#Z<c*L^81x-
zpYP$UyC6!Ed;)*-P5(E!s{zP~ezPXx{cijAeRJOKxH#p5<EK1%_>3=~|2=Kpp`;{F
z*&;*~AG3~|<&rYL#!o$>gp!J>5NL8FIVB}4vFd%ai=t%R)fj7HAY4aTR7k0D(s8l4
z!bydfUve%NOo}-IMahJUDJ2sMipZZi1d=QUkXS4YYZ^8z>uYZ6OU_qsu&N`kG(|C`
z%nK$|<T2_sJh6(K;$U<Q1c4hdb`C>_rHipkT{zN%HPE%5^{V0OV$F-MZuoEin;&Dl
z4BIJ8>rsy2S1Ul}1!xuAY?Kb!qN*YQJFU@~W>QxCJAd|)!^t=A3GW5$0$_`&K4kEl
z@<K*2;?#M=?zOWHq{KT*W;+U7GZhU-%5xw}WYj4WQI5iP4&f!D@mziJis{ju!+A-O
zD|Df8QXo{IaHRu9H1Wns6vFN@EiBf#2qbPVzGDF8y$30JU8;SiwPDd7eSh>j@gAgz
zet3~05hAgIjIJAZ{c`kuF<^kSdk+Is;>F<05LgY3_c^jWrd}ybx8nZclw3>lJm<~p
z*K}=*$uwQ3kxJ3p5Vhmlz-m}y=$e}K%{5n7=Pc?CC-?7D<vGH6nx;eKCYbfSr!y@=
z1KxnhLZoWU^S4@ULNae8%f%Y+p=&zM-@Zk7kN!~gW-X~SZ4;PYZygx(|6%Vxzb#A7
z^G@&=5!)PpecH-$s46%D1W1CEC@zYkq*lvxYt~3TU#8d0KQ|v{tywebRu4&$YKj6#
zfTDq^0#HDqOlDR3%xjO`E+YCvL~L{JxmlU0EP!Ama^1Y=?z2to*s&v?c;5GYUaYPt
z%aWqfXcNqTki;>2+uKB+5geUPx&QEtOD|pK<<~yPzy7!Xju&5fmC8uU%A!={7)lLb
zQ6fZbdGHmu?-uA@f5yT5?bi(MIr(Q7&b1E2+_P^?9Nm}20R-F0JL^tX5Af}Lv%-3j
zONb>cP$&>KWG`r@B;ig}vu(x(i3&Hk*DFjXTIc>YK0!fcs49z%{W)!$2=WY`F<2{|
zIl&5yF45Hzt9mXt%HRV9-21MibOU?<k*+g)=CA=h?Pl`E@&>nry8HJ*boTY<CfuJ-
zerApPETD<5`~;=Fc%=QTnrU0C?Hu|{d#VpCr1f<S5yK(>We{I(S2=+8u-GBoZo7mA
zGi$#;L1`G+ZvU22)MM7qu6&Z42}t<Uyv9YKPIUgNncfyx_w*5v?Kk~d>(=})>Nlb>
z#XdW@l=S3?z42wPUww%O5AI^hf~qi7u|_A7L`H6ms5JSiAj>k+C?!=Ht9a$vI@!t+
z5u;Fb)}%;ANGdRbvM?B(H@9vqt2{@jDFvD#1xhBwaz!HM^yQE~{W{KAA{jFD;>D@E
z4uYkcVN*e03_)s=I3|e|J?ZjSG{*BLG{<@f=j<Yl?}MT8|4L^C(S_UB7|KGkTIHOb
z%{e@ra_8=2jvnNgGRA7<{O*hZQz0ZosRMS>`&LIPB1&DBibyF`9MMmEeC=13pB}Jt
z<EnPJRY)I!7&-ru;5UWOlUU^dD4o-{1zS|?h>AWSy}jExI9e<kC7qbDvN2|sbNv1X
z?2P*i2Ls|k51|xRDvY!U?W4A=pY%LC=m3)gnc6gIkLpxjYsudmpcn4%d;mim3|iZ~
zsEvt8)|0%o9>~e=H9DZpU#|gn?ms6nL9L&lMbKYs+Eo~f6#_$oj8a5DVSb!bK00M<
z)I*mm%AzC^psSLi$dFM?Wh!DNT{~5pYL&BG%(-{}KGWHZ*FN_K<H;6ENQzZPk>#8e
z87F5`v@!IDL$<cJsdRy{f;=x-Eap6Z^oWNKAAzv+dMQPbqqSQrNo$x-&$xU0L#|za
zfi&$CCq0*l*}9Zqx-1;~>qFBEqshwxV+~21u)nuYFHua-X8hjo{zI<Zc!g^(f0j2s
z_jy#3Q0A4RvW@A)mNhNb{@py!P0zUAZn~+xEEXXf^L_(9>P{2>JNLF2+jW8teW7OF
zS_8LK(#KdytPj+&1YIk;elyNQ)&fpmHNydhU@|b;iQsDEA}6eMD@<*@)P``rlGb{F
z8nUAJ><>m@wM1CAKr_0)8s(&jtib2O!Wx&s1luqK_^xba@Y!=xNLcE3CorhtS*(9<
z0}2Us@RG4U81rP3P5=P_07*naR7<)@*qWVx4ge%9v^v{bpBp|T=$r+B-LPFYEbRP|
z7My9@=Pb72d3JD}@1+g_cbM8P`lajI_Z+aS^$UY$1=_lI(H?`If3Y=CcNex#aZ~3_
zp$&xL+}q#X?`dnb`D%7;uT9{t{lkXeA6Jh}`)|JflglPn&UXG6=X&<$`o;9w86EKA
z)mM4@?H{nnPSClbj4F~;A!0WO8Eq)@f+8=-lbkHhSw<@o`bgoxTue+#LM~#IP()G@
z%gAf2y5s}dsz3;ViWE`_@-l~9Vx)jX5k(b(0x4G{WDM!!WP>9ll9Yr21N!t4C^96=
zctIZL^!h1ah`@`N*v^~U9kJcZradk0qUE(`@*K3&ShRLIWrKMCa*=U*I^)r!6Yk%C
z%;{qTIszLZj36?A)kr0X(*%5=sgTkwkYnY<o>hcWiX=(M^86W|{j-{4ZG|?{ISE=2
zR=Vg^BN`!63%be?)r#1bB;}NmDA<h!6B!}V6s8$Sq!6B!YV^&lu`G^Gx%ci(((x8s
zN$imIP+=p1u7fq!UbHn6Xj|Pqd=G5~E_@cvtkK!t{T%>Z7wlP-QZ3Wc+A(X_3an6I
zCW5(I_y>#dES*n9R3F>A$G|kEX^(XO3bT_3eDv@L$`x5<kg`JNIR=Z0DwLFz)>7s<
zXhXhOaqr$e7Rv>%fA+KN?(e&ERsxw?hP@uAXJ_1d^nffc7>&nFmn(XMKEev}Ea&*-
zh{MN^na!tERY|YkBTW+GB(BAwBNee&EI2(mWoI`6p=$r4sw~k(#wuSrRq!&WstU@g
z#9-<7dklI#Mw2bBzxW2d@eavwi_d@ISLydhlqRH_azO%5JNdNDpl$YU-Qj=J{>^b_
z<G9;p0`ZwvIy=<tPFVNXt9j|t<r(YP+#mqlzAqt~KIedEbLuc5I?&Mg1vV^Rg9)^S
z%6JX6n(=9l52j#`F{pj+JWzC*A*lO$yFJ_Qn)@s}=I`*`rgs0h(tZw`wm<J;tZ~4)
zzFcUPz`ZouYElR+VFB5^{~>?^aQ?3{@@oOnFm!FERr<c*>|0%C4!Wk;pEg!RP|G_R
zqcN34m(9tXpPzA2jq=%DN<OZHo~9{qQW1gOTxY&E9eW=539rQIj=iW+0t40tUF{lO
zRLk)0w8sZ=Zd>~*U4Vyw&uQJ?I_Ubi&(E)KTc28U6BKUm)h&{p@g-hY@aDnyc>mUd
z$`qIg%0-EeV<P3k40KkoTCIqqm{`RqF(c~run3GWL?}{Hda_4B&XPGv9FrPDWdv1e
zLFdkjCSA59i$z3|s24LF4v`}EIUh7DRwO8f^yt?N9ll3E0Azi7zzz=7NlG$GGBhhn
z%G#ICwNOdqnZ=kGO;InFo%z*CBDue{Ta=ehu4_J>GoP<Ge00M7yAQd0`#$+1#*`7J
zl295X25H<JA`*#+W1=KRMG=vTNs^dhzfY2;L`h7MXXs8)<j0q1GsRM2tpilTEmIw*
zU_k_dVk}m#NcD<=Su!*Wc2vbq45<~A21=YCk2FESUSfqC+zi$^*;j&-I}hniZV?TK
z#Qgy>Qb;vGD20~71yckDz<0;S*Ud-rDb<+S*1%c-NzTvieD^*F?$)+%nq*|SAKN&U
zxAzOct^?{kxD5hqu1;#>fZCZ5g+L?&4sN{4*4_c4TyX0r1yX2`(xu+299tKQh%&F3
zou2XF!2^ueeCCbM^1{n6V*wc{DqB%%SKq<b7GM0zSBc^nZ49TUGmegrST5$Q77MaG
zg9wI`5iX&4g%JjAHT^gxNmE9n5mrcMvl&U+cg&Y>NT~#iV#RzhWwn}9mMf%k+oL2V
z?GNY=$4C;40N+3$zlo5^fX{y6tL*JxrSPCGj+N8Q<GhdLrip&t1iKC}tY2uX4|VIu
zh0Guq?c=w2|Hutx%>);=1yiMz%0)4I_GsN7uwj8)eZjUsm<^M52!Lq()hvzFq+_7N
zaiPsTk_e!~e-1*`+GuB@U7KLrK5^~eA-(E4DC>TP^9{0I8X&tKSkVTtpR_oT=A-U?
zD!vBLt+N{KHg)Z8taU_Yh;sEgbX~-*LRd|pSwp0;@ob)Vv20AVo6M{=I_KF$iQ6?7
zk`M(Q+*Xti#1VM2zt3&yG7WqhjGaIbVJT_|71Ynpmk>Z_Y9JlVp{@sqh21(^wyA7l
zvNr8=-rV}rmG<?5)5a&Q-^Z2nguwbW8ZTbC1~@mJM-kIwJH5m|{_TIpe0s*=@m*H)
z6;j1iMS+T(tCYo3WEm?FA&mqhNs>NRIT(}@d(XA->?Dzl>Bp`=xD?h-hZ75Nr0Der
zq-jE;1dt=FBocTaCJ|H=%#jpG1Sga{A|@fBN1ucgLq+8Iw&n$L{YLq4L06mRtu|B?
zK0qRP+?C{Htf<JStP55%TA+;~&vPC>eoUEHJb3VscYgRT-~QA0u-f}d+g7W#QFVJf
z_v5{k5*f!t!vW**i2wQD9`f2tPu~Nc%@iwwFoKT^v$*tO5JqRP6_H*sG&6SWf{Do3
zi3F1<1`RX6pmN80=Td=}_;l-u$U;;J5Q3wdcNlF?=neW#0wqqc%B9Vc*xL69&ol%%
z21^rN7r@(jd5NZV+Y@E@ZXMVK5ZSaxS3fV9fx~$=Iq07=;RVLhgF4|ePK15d2Z3-e
zQ50cu*#s7)Ax%fl*GeYH$RHz^pFv>UL}grCP3JR)lM%1I_8Qx}yI3GfVh~Q(td}HI
zm8L8;N#bNNVx<^ai59q(nY2WS&*FgbCRjz1L`W$~(uCn)L=+|N+}43RD+K*PA0Y&J
zv0|~BlNTA<R%lb!eqxoi6b4qM0V|k{clgTJev7?JSI8<2(mCuJCum2cqI0re=YXB?
zK9JVbvKX6u?O?N`{<@Fdc|gBDA&GM=aEH-qfdwIii;6;Ft?cxhb>E%y;I4J<dbDV<
zW~O~jFMXz+Ikxg9?D|`4>vepMJ#Ap_Yz;ta`){iq+JOw+whHFwhVcx?)eO6RwkdR-
zF|R@Sa|oUY{-B|8!~eE!?C!+d?A|~N*Iv!^*#K@s?WfTJ;+uauShC$f=Siz)xWA^p
z;h9Qf>gOH$oeX2qr%*!cNqoeAXbgOtiSEq>SQq3&HnC{k^RM%3>nvILya_e4Vf(X|
zbuO$8=-M8+oBPx!Db0AX{;7R_TmKWSdIg4S9U!}J6p>81y7K}rUV9l+WgMQ|rOZq6
zOd%ubv}=W9e{<<frBWrN{Ust*$XL~91UL6b!sStoge4J{UYt-RKC=!k4}BU(B#|PP
z5^1f^OW$t0!po}|6q1S(<Ha4F$<+ZfONFKMajhi+O@tuwet0$e?!zU546!68Fm~4a
z*lFW_Dy`94kY^dI)q*_BiKCRNELbj<WV2^8p>~Rr$i;%}Y)X-3&vfWBGIrhs6@1Ms
z2y2KX2v32>RE$_Lu}ii^#*WgA1f<^d6HAHo?4C8oIo{jwnpsM#iHV4;L>NIfopb-4
z_v!WfL}@~trbrbbqPUJ=)83a(DTSsP;DbpiOaxl%Ix_<_AjuU`Z);CX7qN4j0ABF3
z3%XqEr2`a2%hlUw0f?kT$_Rl|f|soUZj*Ue04h$fLQv@pgrd-vL_$v};z$uGL82mJ
z5mQx~t;Ld^ogGR8v*n63NfAP$qR0VCDbb~&HyDz{E_I6#l3voMpZ3W{BXoJ0vMd}Z
z)0+8wMpe3q)~YN~DxxTyA4Fc13<g8uIA%7Rf^g4#wOUe?C0U-aTrC_~sa=4@_Vx~w
z?E?@oFI;<#S6+Ud;be!>TH-XNsw$k>%lRZ=&kyqFpPTR#nxqXd*2~8UTi1Z<V;MUK
zg1bPc`_rz2tQQ`t0}@=mTUl!XwlCa7PlsayTMRl#q4@dBT6Ya&1(`M!l|ftO(>^*7
zDg_R}_`14wcFeboBd?9;qBTS0YyipYk+4rPwXOqdAw|&r?c|;f-#34sUzn}?Nj{Gv
z*3Ty#vk@k}=Iyhc6?WavI6qA{(ApAo?m{1^@540gY8`PL@XSRR<XP`cw131}-)wej
zTt+^f(gq#j>J9AamL^!?<5{#WT&tJ-YmC%7sCypF+@`a2YqAdac*<{|zKFHI+O6Dr
z=Q_Z6eU+f8RT%20613J7lioHj-}nq!Hf6pzrYfi8SwR#l;v_*Tg|0NEEWOxzM!a0O
zU`Z7rrEr0d{#{a1qJ+c<NhB3%6r+;_CEdIpAv`<h{o#y3DMMRZt_cwVi_dt3BQ{*H
zLt{#{lvtmxMdHpMdtet5xvAE<t(P`wKD)oQt<c(K<EbjhmP@kb0)Zh-Q?M3YKAZXU
zT+uF&WR>OUXJqG}4cJ)-QGQbRbj4PKDZsib_iVwA%-KRSv=yn<L^ucRT6I}?kpOGG
zl0eH(2YfRMA}dI&AU`_c@%tYT_Y+i_kjjW?GWIJ3(F9A_W}xj3T3zAc^B}2fitR8a
zEf65RS>ha%SGP}lh}t~a*ApHAcnUvpfW_y+l1|e`ibgGb^Q7e&Ln#p`go+SCI*lE#
zjT0q`G)aj>OezGsWx-%L#2CZr$q5p{xId)d?^6J2+GmgskwP#?ha_oAoc0;^hU|>D
zDT|D<$k4hXTdkPQr}SeNI8jv<tJTVzl{7_BvC1>f&Zb;@;kuJ*QEtF6red*LQEB7A
ztboe8SLfwdUgsbDqu-}A5=jrG5=!GbPNj_t+_9$aJMa`5W<Feboq21XThpc;j!bk-
zh;Cc`h&Mjpy5F=1(>1nl6X22ln2Yblt{vO$XQBVukgqg=hSt3~4%hkEY2)-D>w!J=
z-xf=U?bhAec5n$B+w;81`}}w7rr0*Cx4vjvZ?E<C*~E<hw54+nkAMUryQh1%+H8xH
zi*!L8O1VgXT?6oLeT;25r}M=<H)9)#nc8C1#`C82^HUTlL3tH`=Y)elQ3+63h)r$M
z6ti~IfnDjQ<Z5$up<lZYq_#inW)NjB{N~w;^&lYhnl7a3g3We4pF`iR&(ayeT1%0z
z-HpMjv!C&^Zs>4fKU&916bWDc!k35>O;OGG;KR45@|+@(L}~0&L{ho^t;Gt3k)W+c
z3ytlMNl84v=A!DA1go45X#^<KBNKw6DyhnnY&yjhJs9**SfYVI5fRL=3Q3G45$-Kb
zNouGDB^4E+cPO8##W8u3wAQI}_ZRsH+0fr4s%8MSE&bU_DqT=k1yxb8T9qtjr<6s(
z&X#i@Qc`rI%zwt^85uk2GY+&F5bYAjdxwu91^ZFKz#3BR@^pg2ipb}tZ0S>3YAFZf
ze;m=NsDwlsgOUPM6rA0=Pj5V6u(gFsoX(9H4G?kEF$W3N2&Gm&n?ZZRXq!%()qXRe
zulvNg?;9{C=uS%K*oO_1%(;DqXeQ6rs>)TQ5i1m4rFtERcI$7krX&^y6ewXaRfQ0-
zYYP=2;{+)qk|beIT|z~QBFh<MIbZ+Z{|gW9+~u>cf0q60FEHNPX4oH)#0lf!l1Rmj
zM_VZ67>P7asdP>@%UCW~EayuEl44mfo6X4boU$s&@|>b57)?f8zkZFqy<MuuCB0sc
zEL%Bw5)omEkWxF1zM{Zd#oqo^j7Xh#LljeCv3><-rE~(szW)Rp(nXVc-Cj>%={ozb
z9lQDW+Ky|_KCn01w82PocE{+RKfZRLcMrV<uAX9m5bZW-Eo@BF$NeV^P--W8qGTJq
zHT}FMms;R1uxYYxi*B2?1&Tv^8PoLDbx_wf#fAl1FsZKl%LO)k9cZ_o`|~ix;$Fde
zl3Y*jbWx%x=TB#Se%=6R4Ytx)KMplVN-(1KvD4ZM76&O5V?AJO{e-e<+q2xB?<cZu
zzjYrv8&>d7*Y7@2aeb~4m@xI$?q26_*li0es(bEjAH4Hu_zA1A&N<s9od0QxpxbMj
zb9R@C>wb2h1k|+_Wfy+Gd3*c%wsnsU;Ov%g6$A8`@Y;>fGZ>|0`IN)cyA-PmWX2$h
zNQ9^*(Xg6qwZ!O(B3n{UR*ZUmCzoc8OA+Zs?uBw$m$8;Oim0ri)Xt9$VWBKAxuUcd
z?av=ciHu^bbYM&oA#1O>0Ij3eR*RwVjGpkp7B18xJeP`!f)axvA@LEg1tkSJ3s%{R
zsw~jjnPR8YDFzshM(j@ZIXPVL^wKt;PI*SgPGY^ne?a6?SP&ZJmDvZPVp~=yYrzN%
z5n9F|6h@q<(goB`y4H*QRn|CQX9bBhq*79@G7dkwLvON;?2m|hLv#|?+I0@#m^zv!
zOaLK4dV4ZB7fgZh_jZ25=mNL(_UN(*7yX<+pK#&LVKSA@^jg0RzLQDe+%_))k;b}I
zL{tSRLo6&w9HBF(Ehm*TpH;??_7h|h(@!GGvLqFPtLYWK`1vnz``!0>eE$Jye~d7K
z*h-G4hb)F;A{lWUA0ZJgx2u54R2Wl{7X^#OjN_AI@;qm`T(FqWxqbHz@7}z{jT=|_
zxBr*_fg3k&V1;FCYr<$W#9~N`lsHa7EOC*O5-B3mBw;w*29YA9LSvB9x_ikAFOOpz
zzbq>@4n*gf9^Yrh`BY%t{&@bmodfIw)(6zh&%Ar)HXl%5U$l<5bXX$%Wu9t!wQWDQ
zmTuY_+vfU~pwgDIvLtac{9Eh&n=Ew9U)K6uv7n?Rl96wdE@QWT|LuiS_wR0$ar<}M
zjiP<dbx^nI`8^GA|M=n%m5{b(%-xe)AGe&Ve<yt)0X>Jd@q(sCqbp~ca+Zt6hpuUR
zcUG$fMuX9|!4&O&6ry2&%3GB!mHS4!(ch)T3n*JB`(r;XlWZ8{3=L?rop0ME1Wr)b
zRJ6{~YQvg!tC!&3@jV!P*LCjNM{fdJ?in}dZP&=1y*V<}r_0o@@&6ZYd%Ds+uFFm~
ztr*%!bn0=w7R}!B+?g7mw7`5dV}c@L7;kfV>m^=)?aR#GT(HQ_sBDIa++3cT3QXc)
zdu%DK0bMy=SzVzbg^&WRE3C07=>)M=<RnqNf(nKnMOiu`Knh~zvQ#PGf0c}gM1W^5
zOM~*E5DKTA7ZKM$&q<c3)&fmni3wm_W4y7}^|^u^gQl<<rOl~K>GH8xhN7q_tBR~L
z3<o2|!x8=Agp<R}FV3EWa`;bKo(b4V@AoE<NaOUGr7aP5#ef2UR2ceHUhzd@Q4U87
zN4G0$T@)=i2G75vp)*=QTT2PV%D+^`y<XYbl1CriC7w)32Yuq6LPZi4#V8{&+O77b
zaOA4BE+v)G2C<ftqnTG_?HGNIX>)tlpF>=P0|=sKGP;bKk23NTbUmvD#-_esi4=Y^
z@Zy9r06XbkEVr@_Ak>{k>&8hv3?VkGK2%Jv-)B`WkP$?Y6OF8@iqcdlBGO(D6~(OP
zOQIy^b6@x(!ou<6Bfjz*zm8B5m#$pp_U*guUb)V_n;#;g7>r;xn<EnEhqugfvTR8l
z$HZw$e=wk`3f{ecmpgaw^2Qsl@VmeFyS)0wYp7n2%2XU|Un7Ylmdh1bD60Z%O&xGj
zRTaJQHsh^bDgzotq$FjzM0ptyQ@M0u*0a4~aIas}8br|gTeS87AzA?i=O)M&;PN1@
z4fbkaw{~2}s~JvF_kp^<ZH@>HMfpf*>j$NVcj&I5F9I))QS%X)K7!S_(ATv+oo1r8
zmdZe7Agax=jZ1<99;O-Mj8m4^+8}X0j^Wf6gK(koh~O+|o9EFr#nLjpwtsiu!Sm);
zI^eE--y8kJe##;+2w@SzYvl<KZae1={YlpEd4Pbny^D3}@Io%l$i_98v(_z6YT#%A
zq>os3`L@rIIhIeHLQMEHby6un)I*f8fN`MCdcDCg_MD@apL_YZ=hn1;1MNiT+wh(3
zj0Wu{TyMxwHiI8qgPCyk?zauY1C0k?13Nceb<<s~!@TucpRZYWt6i)AdD`SjWgTdS
zg`dUx$khN=Md$32NcNc|2Ylw0FYwMgKViOHk}qe7Si3pQRA7t~ix;>|I7NZ*3(h!o
zvRI)S)&~_uQA8}GI_ZIxlGvw81c>4Y6~ud`LdwWPI%h^rNKnK`A|x>hDG7=18>W^n
zY2H?g^&u1hn!=B6?wf)VOGRN<l%~M?akVTfvb-cK3PhAL7;P~eO^8(L^xMM%>E~3Q
z3D_wtk&!6t9M&TR{k~)A4p{mV8X{}KDvb5Y{l=mcF3{N;Vg#j-XkjTOfb}VC1eG-y
z;gYx;BD9sbNU{i1Xr_-2Np5|>V3@EqN{QkGX(WOOU0S4fs1E|F*5Dj2>-0D(%(?`5
z5%6rC3}dFL2Xq-`X@cUCAG|#xl}gqU1>v-go%H4oA@I73%^FDv=i4MCmC<ezQGj$4
zqf`On3NL%nIHx;M;ss=(o;9tFutL!rZn0b(vCO9Qdn3;T`W2XQzE_HvB<iCr$iaY@
zzxV}`yLY&Mbf2C5eWINa`~6o*w|Cha_PKiH1(b?-=X>9!KOVBRv&C$B%E{p)wub|b
zAKb&1IfthY*xDQM&%f}yyzzz4v$eNJr6jqMB*O{)$%K4111r%*Nm(pWk@NE^OM}*y
zo!tYME-gwZ3y7^E7O4a0#u5RQ2Q1AH-n3U6;01=t%}Xu@uP!1yv+m31fJ}>N3P7%&
zz%LlG*T>>;nHEb7?m4z|n}$KvS|>m)1t`;){DKc=>rtNrK@Gd*7$hYdT}A44a77BQ
zHL0b`++hp|iwd372!!%d98_qVVQhgCLm<K$fsq<(D=KZAV7B$|i!H$h{~FN&Y{mqa
z2-$izt$K$x@b~QkJ#76Bxo72B8~+Zp7d)<wKREz#3oPf<-G<kqMpK(iRf6@vj!^3O
zL#ta(b~PgQssih&-+e16y0UDa(7z5qsbJAquV)K@*VK>n&*pgpJ9h!$Z*YD3+cX_^
zebTyrZ)Q^)_AV@yx~$ZbuGX&Q`8`bM{M|5v=8O$b*j=Et!Kc7Wg8RSL&vbEV1Inhp
zU1srFo?7rRYmN_nKLk^RU<8Hrex^z(t{mLp(!q-?@;OImbC%1M8!C-KT4yS?R%5Ko
zPhI9YMde&_($vl0NKZryiRcRk{k{Wr{*6zPK|Qv)tQCoWrp|;au^z@LQu;)sC=w!K
zpY<a3ty7V+M574l!336w6o*_aj#3J+&$)(Wc21=WbXig4IYn8rT4gL(8M9@^Yp;Kf
zQQuj&syw5}GOD8R?iW9w@@xXd#$ZK(P>`gSOZx|08SN6!#%%9qz$p+Gt4_IIw7PQ`
z9P3F-v{x;Dg0^me3HtdiO3(S~#ln)Wat?3bW3s)4+TKMa1CmJ$kwphnuV+LQ0&9#9
zN!ETqGYFw-KnLE(#HVo*;Z+sj!~fjm$l42Dzp>5BSWnU(4A!0<MuaS3^$J+{=v*Zv
zNP)u1f|wGYqDg@eo`s|~`N;Z8KC(_4x1W)ex?(t-aA>7ZJZ?y2jFJ&S=A)`Kk=M%8
zRfPqjBw=@dkHyIe$B&N~+_}f)7hfTXQp9ANI8Go+c=5(7^d<vTB#DJ2)&_+pj$-cJ
zzelg%<9B}NcNmTa;3G9v6qBYswzqe{8p^5y3t66fO*DhnkQW+*<mz<?BDJ+dN;yDR
z`yH8@g{m1h&I7GYrfZ<qTI@c=0E}(ZDcjpNoTmYc>*f6FzX6F)_VL@~J8u1nHXqj1
zq4C}HSKI(tLowEI5ytxL6AG8jg@z3Y?Q5;4hwAg<a-W8*E;T-BF)G5J$2R@6?)Mkn
zyRcY${`)q2*O@q5%w78#Jr^IlR%!Dz9-i0Wvi84P2S_$(1lmseFt;uX&qj3H&bw6v
zExZdVgI93)c!KK*#aRl3Z?C5T5aQfDe$rCoMgr5>)(YITzg_!mYVg)J6P|bW@1A4x
zZTvKixxN?-*F68%#oyDZzh6~x(E{OTSGvY)yXI%u^&OK8nXtcmg~PK4oX#JiDai8z
zi=c7}*CgPSEQF#gORCwNA}>gLDZ}9i6(=rrr7fu{O%nT5qDoQd3MCS6N|jznIH5bj
zdcuYXXF85UI0b?P2>Bx_av!wejtG8tioo5unAc_nWo$)J6ck0~a<JzGj~_m!C`*z!
z<;DvyvOU~FNr~1KWtC%1fz}23SwmT$V|g~lPD+ef(Ho__`075Ned97$wjzqhJM@P~
zz=s%%i|WLJ(T>iy4xEL-xi(t)Wug!sdt3jQti^c2;y?onYh99FE6C1PJid94XuLyz
zFlNx}W0b@wgO(cW2YH19IL?Hm5i)YC=BV+v;+$`krPd#-SFr6Ra%-r!YpX@iz|=4D
zy3ds#R1w0-vIvFpSsJhcX<SIS&;nB$3R{w8E3#EamS<$EjMZvIUgYFiPFa?e)+Kw7
zWlS1LdQpiID|&H)Ha<O-r3QA^8mur#C8<g`d0VULr3s`1M5XxX-J6^oFW9?ug(&L5
z(vZXvs??ZyPE}-7iy1OEWZ9Ajx9@|77q7oWQDiKaIq9%R94AgaTsfbtBFkAWmrUm~
z%CZC@kSeCAN|t#^6!+ObxI$%c0U%PLwF^scz(m`O+N{dY4dfeUNVyL5T42jDqBhWU
zANwx+yo(Us{uRvjK<&S^xww1O#!!@+R5{+nXD9;^7$yfytY^(^8=&Z}O90^q8Wu7T
zodI^Y!1hxGv4}O_#+ngtp3U<RB)+IP0PG^LyFH?92K0B>=`MbXu-1!1TR(iX=r9l9
z8flYu{|>1TZ8vRLnD;MgIsyHDLbfWO*uHgE6WM((Kijgt@VFS%HQ%noqrfYf7ImOI
z%+qG3v7r(+FMhgDyJlS1>a%`Ef#`5=fu(ECD(YBS!6zSMy5wrO?(-;)Ic%tk;D6U#
zvH9M?snKwYL4QOP#UR~cyvPcSaboLzD@ap86gx7bEDD#eymbC*z5W2DBuWU3G2}%-
z6gg9EtRi=%U|FM`Wh6Z9km1!u%Hloj!cVBRgKAFfB;>7f+l}`qtSBjbW{DDG3NpQN
zdUUHLS(dR_EXnhNet*E$WSjlT9*K;wA^<||GAQ{>N<XjiY(93vG8{=h^Ts8<^y{y3
z<E32&QN?0pNs?;-Xk2`(kghAcX7&!l0`Je}<Dh~Au?!Oep0V?+fOPVJZY?ITl}1^I
z$L9}DI2?Y&U^FI4Q{?UzRWHU0XYPrFKv{{lK5*Fzm(*SO{F~kvu68o<pWQHSoI{;y
zjjOh`A`Fg!t(CWwUbb1tIt@=FvEof+%4d{N4%k^MutK_(eQ7Xd#d3bi^z4kqe8Fsf
z!hA8M$P0?nC0DMhimIy6m9zC%LX#Ca5{1y3SmeC&;x=h?nPHj|5o5_QGD5@>V>Fd<
zKvJ8^2U%z&5y^0jG!alxEi&{nCmHrwot|+#n_<e52*|S)i^YPn%vh}EP#Ko<C7E)A
zOmC19#V)WViXsMs0lKVM%;&6DOBRa-t2{?V5m{N0m72;*u3vkR(PS4;&Zoi_SY2W4
zc}lwt?gX40)izC7n+Cp)Lw>zaYhd5t4;R?0kGtNdTh4>j^`M0(6=R%Gu{HuD-GJA|
zv~^TxV3xvrX$=^cZYz{c)3RHvcoOuP@2z3~YLjhX0%@LETXQmu8K1Otd6s5+4*jC>
z=ycz0o}_Kler?&A2LHDDuc@;?w85P<rXfn&!0v*gq+l@YQI?vjyx0UJyzuZQKgLZh
z@tj;2J0*aTi&*80j#+2H!{_dDx*I9GzCUyU-#Yke^J;btQg-XpJ+8aB>>hI-V0IS=
zYj5&~tF2$VxyJd&(+0rZT73K&!+wE!X_)ErE!a4Vj0R)+y&;iGutb!)q|};HXSggi
zF3Fb@;KzYFpQ%A>gSM7lk`PBRV9BzAIPMY00wrS)m>OVbXlIL%b!2N32OEH00D2IV
z@Phd^+(!U*F8gay{}@|PYB#^l=04lfYQ<`mISbTehrOMBCjBjpb6RojWGX6MqIHSU
zpHRo{6O?BIc1m;Y`T?K+;)}fg`4<>%M^yQOG8-{ScL5M-LMes!6?00_0y?RMnyA_a
zUw52%t*1n-2Uy-GO;{&#oC-yiYff+9XD}Yo8x80u5cPJPM20XPFe-#`rW0fF86H$K
zNYoeYGHbO7SqWSsbZ};u2GF?@CYk`~JOk*{iiKbb2kM;0oD+goNF^zXf;`X3mKi6H
zPdPm~V|F%UwOCP>1={2kWky++^}w&SCeL%1s>m2hTT<u>X=5V52*ZO1M;z=8*c$a*
z?l~DDbV(dVE(*z7=YSe1s<J>}DXWUeTDG^g+1*W978SeOyNves&{@G*nWGgfPo`A!
z1&M?tO)2vwi&aLNB&>>zD2|BJ1eHWYvCDa9D@SK%%auzIT@+MCW27K2D{>8mR=oU~
z&!b}J`)7@I;K3RfY_n#1gC_%+P7qA+bK~OU9S{j-GSk_9{yME;xpkcmlh_(ZFRFL&
z%WSrX|2AK3)BSeLl`W8?gHIdbJ3LZy1Go;ruCf>bMtLJkSmU%RRfA_=J?+!|gd%<0
z)pa3gIEjGHo(Ljbdppx$<9p}V)BY5a#>*}EC$>wI_2rWlDV<ZcD*{nZa&6!sTAwd2
zj(H1_pRP|MyHjzUPOQ?{5cHyT1MI?Z)7G?GD@sAqQ;dgW@>R~tIDgemMF?tB>_sJ9
z(t*;(=k)oNVDzf!Qj*k*fs1g6b9`?EWL;^e#R?$6{cbaJ&*m$*4wu?svTMN!pswrV
zR@<7+ow>XH{Cx5>pe(RY=j#~SrwxTSGnbqH4lYXPdFt-DLf_X+Cfke#E^weU3K=<}
z`ZBM`^C?;z%Ce+CC`o$>)<RL0lvT-c<>tC_IHETg5XUj|<qCwt3dv|N#;GPa!)X9+
zk_bgaOwW7Vg&Moq*bqPwAdd3KX)4<3!@|-{hl~;wR5l|oSFBbGPL2+p>ceUU)-o85
zc<H5=*&gjONJm6OKH#Fnm<(My@<mCAqo1Ey{h5H>WIN%rpL>ZnKKl{}m$s?oilP9K
zrljLB0OS2V#6(e^%!#Zf7B2cz2=6av@GAQPQH$fduBbv*0Q`#18t?Mt26f>@k&Sgu
z^u=<;>7Ba_w@28XV$_d`<On4d#yDUn6<Fa>y3g?7(soH7dDG-%tp_`4oGgKlco5Qh
zT<0Rmn*Q!4OerJpzb6~lJrN>WeM(|0Y7M}u)GTKU=Ce7E9zJAxGG)0~l4m)()L852
z>a1Kk)8=Z$Y&K)DSTLK-m@k}<LXqd>Hg_H;#wqwDvgGQeJ}<qv&FNXfcsRsZ?HpIj
z0wrV8sE09bmax`%(~QJQgB2DPOL{V)%$7X*a7v!%U^KCE;*-iRz>F|R1-*XCB40ry
zNe6wBUe5zu7i3dpxudb?a~8`5t2}eMbXJpV%_1wg{?Z#<z40oQQJxL*gRyPcqIH1u
zB(vT+(-y4k7e!UJpZ}&)cCYLDVyDlnovQ=hJe#z6U$>v}bk4^Sf6aY0^=W5Naqo|<
z)9E>E71rMtD?Rf@VC^iG5-8POw}*;YoJm#$4M7{s_pKCSzWyS#huz$UPr4`97X!%a
z0y`<)!c4U0(`R8PpHB&+l>6MCd<DVGe;#ZGJ$rC0r?K8#XaiGRH#Mb~elUhc1F_B2
zU2v;Ok}w{+NpTa0(=Jhz5XT8gl5}(fgHC0TLka1`+T-K%Di_w!KAFjD@{5|a!<t;N
zO&<>CIq9EV<DJ|t?IW64$cC{C`35%Cc)hN&?jx1}_@Ko!*Y!BBhWM-{+*!M3P203{
z=RR#&2bk?+!$QY${+tQp@h+3`F2nwW)#8{)IhQGnATM&xre`cyOM2edI_dXZ^nRr&
zO<5-xD9e&w-!Xq>S(2?*AOyYMfJ7({=h7Be*LdEAR<$_E5@$D*L?l20SR@q#K$F$&
zR8soq?98yGt;mWMXJ;o&&rUf$Iwa3BcDA?K-PvJp|A4LDE`8Z2A_A<>Q0oYRs;rRG
z<uQ-r=sZUM8I@-OcGq9n<CRyhbFjZh5-Z9wLmLn(CQ17M4EOigy58ex^&#jAtGyVj
z4`&uNYbWbT0NdoBl?bog-b@%eun4v#DS<_5%k0r(9^Xuf$31$(9{ngriU?yRD3{T|
z5ZDn>XOD0{-d-OP$cu(lZ&*F)r5pr;W+fkfTZ?1h0z)SO<%f`($%}nra9AyKrqd}W
z$HyE#I%K(8P~-)xY=t#Wnq;w<^Z4*FcOTs2==g})Y{qJp)hWC>W-y`MJ=n;vQ%Y<#
z#9B*H=FFEfM#DZ<x)(u56*jUMT_FG!IX^ZPNo;H>b&07~fJ@?BmN~0zjucQ-7HvyZ
z>fT$eHE}N{ktvxdks=})q{t*fdYvL;G`iHR7E7|_iq&d`Hje#VE?2CwoH!lw`7eG2
zWa1=pES_~WV4Gy+>*k|PAm_po=E*g{)dgZt$4%JQd0e*<8_pFLCw8N6;JQ3VE&Tga
zMs|itxdv9=1Q?iPx7iYuZ{52El@VwSMghu60yF}<jY-j1Q%CJ~KwTRYx4>oa0W^pj
z<!zFC37?H=^d{@`&`c!)I^yB${lEnVeR9)BcxGKNv(8u8&Zl}}6<r2Zc%@@20+?<+
zcN=sETMHNJKf0>*;c^#r&Nz-4jYcF%?ERT9zQ1*oeW<4tjj+3v!vFj%7>71Ey13#E
z+<H=gM8n`U@3I@u4PDi_Hr%rH|FW)4O)gvrLjy=`iw$;zeA;`pLER?P?E2k4skpAt
zen%Q`8qCtq3hO}BZLnUi^*+De8T-+g$#9p+WRFMFkBFpj8BP>L%E_iwm0?~k(Atpp
z`%d}6Q?o0a-fUSqH=`(0^}8hFh{UJp62j$m4YhD^tN<hlk#ApW!>W@X4p}FfXBF<$
znP5wDa!PVan^T&C(~}cUPL7z*W>jUx{{B7(yZdZSwi)#%^kk1%xa<}#5JR@=1lIfX
zYokBzQGROWnSkBvpS{itH!d?C4xNA}Ig#vPr3Md<(ybBKKKnAO(=+ls_aeO3`_oyR
zDboMcgMMoQvcj8BaRVG7Al~J&6bOV5T8M;1ml?B1k2ty2r#DDRk`yTvCQdObs?$$-
zX(oj7zIA>841PSK30w8u)~onPr`>0Tslll${&fwe<xqbwEkb*6$4&f;MaF~s4|)9H
zAw^bDl@+?s=&GW~S3G+BkXs+#<o4Z<I5|6EndekGI0U;QrF2f&ogUK+J{E`)iBOX9
zc*JyeN}4)aI})+;%d1KkDGNl(k<vm)j1ZJXg*GJ;>)1nS>5mdBT{^Q-r5!d9F6GNg
zuUM2zL@Y?sn79`qyf?nJma?oUvy$2A8O!;CEO)8La$S&D1(h+p{@FLUeDyj+sf(n-
zcoV+%^PK2Rn67KIgl(U42zxT9dxm$=GYf6@ZzEGj7t}q;j9mMee!`V}`~0DA))RCa
zC0S~{B`u+z)WU0OtY?oxz0XfZ?Vycy2@T;FPGKRp@wtTqJS`h?QM=97Ps#3+L^|f(
zc75sEOy%=nmcnzp=s5(ByJoU$9asM;>gP%8UYb78_<*(j@33q8Jw=C<QjjDGgTcT7
zJDv>e@a^Y>+qGllptS&KziaRun)ITI@Jw%=ZNQzSTd%emwBES|7@enf69mxpiE9Hq
zH`}<s!vC8=-N#h5zD}rL@cU}c*?y-sy<ZIvw_b-f47O`q+feMuWz#-SI__dHn#w+-
z{uWzXdvIGIh>$`ODX@_XX3UG6yeKHkk|<k|#1U~46GbtVuBa+aRhCp$>GbXV%Oqn(
zua}~u2rDC;3|Y|Hv*3Z9Qvh{hwQlL^o~=tQ=D-*J`5G@nT=`rQE9R7@B(HLo`HYj}
zW2Vz7MUm4>6E0u6%<g2DL2pP*LPSL5Gt4@Aq=Z&fzc$Xbsj7ZK<W8=?c$vL}ZPH$k
z+++w7W3{7P!=RdsM(o|VPIi34?XxN7Y(~2igke$A8i@UyT1V8?502VH#ex$wwovC4
zw2-KXVzJ`!M|Vg^LqwV~jw0-M3)@RvRH$hQhY}__&v1BI5eSoO7&!bk>+qaFInmHo
z2?L@V6dgl_(DiF-y+&14R?Mag?%#jNqsNDod4Z@btJR9K%D8#!eZKdD?{W9R9p=j!
z#yTy#%J{m}c+#5%B%MIrW_9LTt0*-H*xlRV`0x?S#e(5zKyEXHa1NwtEUC&8l8DGB
zXEu>VC<#WQM1-ll&yRIFWQEDm!Y4mAhSF4&wj@dvQ4*sfg^FFyOrs4&QBW3|`E=n{
zKc@0#UnsMZywn_Ay2`J7`Kwrz1C7E>G*&bdVC}ycTA>5T-Qu7{^jui1KX{Y*uI;41
z&J}gfeV#GA2-I~Q=QmxqnY^Wx&Y$yRO^fYSbN9V!Pj2U;O6z)>`Z$a8wR4FEL=EhW
zrIWMNSc@>)Gf+Y9L^n~NZEzCaYvJ0mu5JCiS_D-692=d#aISOgo*x?=*(*Bt*)}PD
z(sJ%O)xgZH_`4v$f)8{Ek*qbtX$+OFyg<6bDvQ-lqqE_5z}mn}`v}=Ot(d3y+m`7K
zr9Vg*j|Ys0W51g-TVGVls$jKRv6wCBB{50TL#deXu%2WJeg36uQf;2=MWEV^fDL#F
zi|_zyL}x#Xc0px?ky9f5|Bb0qbgmGx5`<Q8`^n~cciuTuzjxg;HGt0j4rbv{zwQGK
zchLoU?Ylk?`ognr13Ou_!={DF`ZIlgh+Ux=_s8t-9-u_z+E7TRPp6z<v0JxTvTVib
zZ0XpyIAS;&k){b%SyJSZEXy!PyZy2yMU^9jr5fhwehxSt)*ue<owGdoQvnL&?0#q}
z&)kXHug>jTP}!1_oCOOCRxU_lHs$#7LynG)P)ac#jk$jH3YYgTlc?19B}wA{1vm-0
zdluG#cc&9-(fZSi+W)xnOu+8ijXlzFidN1+Qvs0>Xaz=kbQ;G*m-g6s<7Jj-#~i=)
zHa4peN|NaUD^rAa1DUYa`_%<gs05)f7z!a!QV@H_tB#_@)R`izh!I8-qp41p9KH7;
z{eGYRXo5^q==T8SwCsu!of^_uAw}eXgAgFR?67gy7Xno9ml1BYC-4kL`(+P<kdL%A
z!kG$<un-Aiky2%#M2~#7<o3-6JbL^9V@s^G<hJDY;T^vF-EZ^GPkzi|HK*<4LDjs7
z#;m{SO%XQ4Fp1i2)>xO|J&i{k?8H2NcpoK8)G(&1O0w0$tAa;VrFCgol)~s0k#ynZ
zQbfc(MOow&m396C7%H56Nu?`9>GJzUQG|>ms;WRrNm*8y%Cag~l-{&lnu^MlWciB4
za?aU&Np4eq<x9WK*5rUyZV^Hwtwr0)NwP>mBz)L<$A_sgX$3aPVtwA>kS57(ZKO&4
zAh?G9x*#G<HL@Ovupv0Wvs2Q87ECQKQk%vF4Vde!xWi>dV{W{lw83fvLLKw#0E2aP
zuwhZ_%{#*7TCR)>-4{|Ij382qD6lekzZ_#j&^M(3Wudf|$OfZ=n?6fzsiViGzzXYJ
zXV6rp1hiWa7~k`3@b$Ae*%a;kSgl2SQSJKPMFaMNc{l`E1=D03+*$u@!4N6w)1^6?
z78|Cxj|IgaU#x#VMi@WBh0>0+Y&a8bEtNJtFLelVvUT>4<|o2N-o)zJ{~#IT-dE$<
zQ7<IE4(cwrwY}Y4_V)G|jYrI9^V%|TQOUBL*=)vXzegM^wzd>WlGYcn-VFR{CNS@U
z$B(~`=e{rat|^<KObDR}&n`4+Gd{Q@-LY)|x2d$opL$PIZ!xtoumNyg=bm=W!p~2$
z7+Hsq0X&8~4%Y~Mb`v;!enkU$S+QDXfTSw5Gqp;El!<#sC79SCgquIgsz6to)00!8
z*vWoHQG|$<^DAF0SS;q`MMk9yMxznK;gHp6Nfal<Dj^aHDLoVp(mC)J)iFl4)hdlh
zwIO3kVU`q?1CM9NQ)XvV7V`x~wq!UMva_?z&h{32`@4+f&<A8h#H92XBk&e4a==?Z
z5H4$jkj_u`7bWw=WM_g<)TY3is`mpvfRKuKG+=mnkAs(AAiw{B^4=j<-~<Pa>*TfR
zL^SIf@Jxdc-f5Nkn|j?8R+R$qt1)RTy3AP~pYr&lyF@$Nj0SyFFNN4Sgt;g)mzlz?
z+nN^=JRm^S&iVmJIA+1Ggj73(-u);U0y6w>C9wjk$`GkO%W1}~4{r1D!DF<k(6(fr
z&H2GQKjJU{>MuDxIib=aI_&97Z3e^mNMi3QECFeq#$8!59*iirb~!qI)MPwJTs!pA
z2x}!)DB`FGwsOx^Dblo0lB5(xPEj~bo+8SqWQn!z1<|^ss7mKYXANoEC(BlpMTxLZ
zFnPILvdor_=|PkEV34Y+*xJ3p_Rb}$%1c)O+B%=ix=#rA6xyuWrl5%?m?4-1t@Af(
zJ+UX5P1_IFt-hMUuq|!cNOyJ_vT&aIUjUr-HM-Xc>{V-ntG{g6>a{y;>SM$6Z=23S
zq+pA`STkR{trjIA?ImChT32Z6m^))!Ab9W(tsj91{oU47TRU>NT|Ys1cz%44qBf`F
zL1nic4C_(Ip`X?@w$>eQ*QclET)}ksTuL}6o52F@?biQSGnNf|W9xQmv6_DQ6+Q=c
z)b+a>)z+uI2xkZZ!%3gLy<N7pw&)M~EEbCmHSQK;oIhewmUV!Q5N;fk(z~UHaXGlB
zJa<qKpjw!I8)(I*tJ$!iZto$eGm&hpU)ZP3Upek3G-)GU)0Ch!2h^Zoy;kj!VG~am
zuGHLV;0wd?rnOJIM_n`rtZ(nmnbdb$_q?@Z<=WrR=Xn}qI9n|E=-yrKJbsV65ARV`
z7Go+RA3P8#1s?K7IG=YZKq^UD7S5{UO|XR&SSc`7L7r!BE-7=$s$e-EF&vG^i=6&o
zND`+caY7m8L_!kzj5`wLvRD}Fm}|di!jHd|E}1P(S!Q#xY{_c5WPUcsR2qq3I2>}Y
zf57(En8|R&C>r@}7tVt{p@$&x{m9fy1~=qcoQ#;EtTfdx3S$>3h4EPtWMD-^1A;=4
zQjv^DOs-$y;4?4tXpxc6mOh`GZ~}qC`OHBxbcnSMTu+o8|CLrT8s+N|3rUp~9Nm9F
zw6l$xY%%N)kV9vhFhWzNJ+!XstPkyLyX5fox)th)S9mZfy7@rsd+qwAl)55`60%aW
zy*=WsZ-0x!$H!P}(57OQFZud6{wv>p^LrF!(VCE+UbwYxn|cDW2rADwTFOeJO^&uD
z2+iL9E~|3Ez5DmMbg+*#1}P(w#DLXkT{-7?8KKM4IicGUAtI!RQPNP_3Kz%_f*agR
z>YCDBSDT8(V(BeQn)z}@mSqU#?roLlPT;X95GrOe*`YreVG$@5QC1GrDIAS&12}t*
z^U5aQy7p%qCfK#_JIAjbbAF%jW7f{SZkFxdSa3yXz|WKVLwNtrgSXCJ-Wk3g@P!~r
zBX)Losj8Cs>8aB;)CR33D!2jRb`O4bGSu0(Pw;hXznh^S3;&0O+MHY8U#p&N_Hx~9
z9Nw$Cz07$N<0e1FPg;zBzScSv*|0AD8qL+Z2YLW3!I@h{7(aYxAOP3^ksH3J-r6z6
z*1MiB0%RoF-P>aSV4v-+ZK5dV>~u<;#Hd_uGy@6ga&1Z_>w81^KH`(A3$MW`d}RNp
z1MJp;NVi;Y%Z+<X2UcBFH#g;WzZad~=bN{F^ILni<7T_t)&Xf#w{y!6y5b}kfvYEh
zyY3m=chmhGpvHQSecE$>e9^`@|LRr7orjP3(Yx<*czT=D#VMx5=aU921uAxtyUq$0
zdA#6d!4Oz&9eZcAbN#_!tBO3!StTp-EF;eg=EE7o(TG)^vo)E}>-9*|geXl(q$HLR
z$_FtT1X>#^Z86$VnacZ?SLFGM)01PCi-iM@xl8@kPkM}pBX)Ln*xTD<JQy%cdh|u#
z`_nmK=hBa<x<zYX=kkhzfwI)BGDDvKf&jbl@=F<VoOT^;UGZj@C_-&b*nZ_DRwpw~
z-n)fa<Q^0#huVeHyQ+r|e;$|5($*8?rXs+bA#qtLq8LHNYQEy=&Ru#tJ0ybvgD9eu
z(m704B~FxXeI3>?H10kc_5<PB20v-hx#wox6huE06_G7-u3o;*-+b%aOs7+faZK1^
zIp<Hm{-1g4?YAha(hsU&^v5{Tv`dRkK(vQlgJZ><hFMW$SP$&tIOfXLYvg6YqsNDA
zO(s;8L5PSn?jwY9?Psm?6A*Y%=emcKF;eD88I#0r(r^(!j`~(oVkOMyE2>gstYtA@
zFkjA*LJ*6XtSp$G%~)nDq*5rAFdT2Y>!`RMN+p5)Y){<l17UYXYH<L>MOS)Fv<t@j
z^=QiV-)&IW1|;XN-RRx5O}QQN?*b6HUiWT&x;|?cPpJ(6+3dg9+;78Ty7wwh61I1C
zsEUGoF{jEjM!Qw6^^9R{PV^>4<+%@QyLx%u>xu>oKx1lMtIv9StOH+YeC{@}vz=Dh
z_>6w~Qkja<R4D26E}ipr_y^RVnrU51fmALMP)ZBN0Xr!)(wpstM`}X2_tg2(yS$rL
zICEC14MkZxf#D~l4wDh|CW@C{e36%5dWqq1g0+I8&|JFo2(X-<PCLg43B8eGZ*QCN
zXvlEXBT5u1j))RPoJOcPa*{Tp?Y8h~fI5S~<qUT|QEg`H{B27iy|hGA+8%J{!Ceri
zjr{c*V6JJNoUA`N&=a7St<QhK!mVc9x@OeoOpQNXJ-^V%wKP6{=Oa*`+}@!+?dae2
z`(HmVYv28J&(pEV?(?i%areOkKKS4k@4t7GEXxt9AeIAUo*>a+N=jYO-%3Ea06?J>
zR1$3}=OQ9fdZJHR6qGurs0y;uWm?i^i4~SOj_LPK=neYx20hYVkG=gphNB_9L60a+
zNJQjZdachmB@|^<QI-`&Q6YsO&ofrpis^LDYBk4bmlbO;9CH2gW%l>>*xA`(e}A9B
zU`VQZq|%vs1x_}_fx4b=j|y;EkrV~O6i^jdW0_4e4v#91Pk*YWelF#ifSvVV&iz3R
za?Udk3RHv~ju>3K%B6WuInS7Vcn?ugo7#dI*9~~K0cs5=(wc0~cStD&(mGV`B4#bZ
zRLmb8aeDg>Taz)peh=I0VT7bAau;=zq=blsV6F?1mu_<MP^q3YXa+hF^zUSM(h-i)
zTP`y$?Oox6_il0T?meoa#8|^}HRl`O{I9(EgEyU|h;<$SPEW2|f{WGq9S3$)DMb`1
zl#&=-VvQz_oN{*(X|~1^r37V}V@!ool77_Z#Tzg2AdWdYI-)3CWb9-t5k8l!QZA*L
zwUQ)GQ7XnITd$BZB91{sDOFjK7nygew^Ws`)2)=HVU;bJ&dw-`f?mH*zFaa}%$=!7
z3X&wIG=i<|J&;ii>Xd5Q-=du6S71HDKrTX1Kx5+at<z!C!nSVviTZbNwht!122^wb
z(>kzl*+TFjr+FsLF3o~u9k8?+D7U?-oAn#4j|l!a_BNRL+Vu|i);3u;!o3an1ZRp>
zN+MN+vAH+J3XFANqn+NWlNWXktag}1|5CRZyAY5R0vs;#F<b}0>p<;kxQ6Ee?kcO%
zR=ahQ3X2U{JK=}AboI|{cg{SA@FCERuFsfeGumgzanruFU^G-0rJ0m)>Czsb`~0i?
z-tYcXUU~U54EjTifrk$t^6IOva&mgYY%%o~0Z>X3DMhcJ0tRC$m#R-jq;ZdaZ$J{K
zE!Iwg4EeO5Kd=k(vXz&j2<e?%y7{2N)-LDkaFGFa6R7EOSLeCc=1B#n>&d+>XcNjd
z0pO2OHBfEp+0-}G`jfWn^M58ER}8SKN`CyoJN)QJ?{Iv4%4(SrM=4cNayC2S^z=R_
zXZN`F!i1z}u)0KRgHQsYBD~`ev5i54&=yHa3>XxK9(mcPC~~qgCoghVS<Z64@*0qC
zq1hV^NRos&jz}UWb7D%DIy6M-8}0TFE?`0kq*C<KK9li~$#~3Uvc=`AS9tO2HAcfB
z{eF*ruTQE5q@*N1@-<<EAjQEY1(aAy3^_SVa#n7fGZC|sGydu?zr*93M=*V2lYXk@
znSfnjrP`WevNrWJMG8r-EjCJsx3(DGxX$j`oYm=^@^tRZu*Q0yIZqLLGmkT2_~7I)
zaY+Aqti5k;ay3Me%T23rdN!1X<-^BJw<aW`0UYe3dOfGBXH3m>c~ug3Pg2$slMToU
z>iZOc70RXHT9bBwptB}PqaI~$xc%W>s>=DWXjAb0AH2zT-})|jmDlS!OBH7H=AMKb
zQ>_)XU~?2X6S(WlF>xFrrE<P)RfaVsO7<L;T~!E?I1M@PuZDmqP8p08H(q>^w3l-C
z_8lHPdWe)Uahy_AB?9QBJ*P*Pq(~LH;EAHBpSi9oFOwiiqm(RLVT>lva*Q#|7ju@&
zIa!{8g;kz0n>nUVDMb>c2<cS8_YbZjRD#i(B#E3dKfxqLCt^7Ou=U8f5dI!OUl6~h
zwZd%Lwqc9fjBD^U3X6zfK3q3@uV1?>Ztb3<>o#4`xrveL-b>R51g`JZO!gBPD?yt*
z4jE3A`z>9b<VZQEWGS5wofV*sTV3@B1JZt<)neiD^M*-Bv>7=stn3*?gXw^;&SEIw
zJ1M|++invBdJep;v322Z2L^+$UTYd&kEDGLQL)+>s#aeM0V8hA=$haH%T&+n%`oQc
z<?CCoV!a9519)wz&F<EN!2s;^1xg8fmnOXOnHTxeuY8%WeEBO}z4`)4(nD**WU|G0
zJZ8R_vCNk~WuWT|N(mw*SuEWGYq^~J#a&FA_K2dmHpPa_6ylRN=LIG3We(B>?k#v@
zooB+%vtwa<D0c0$W7C@ma76pBz|gfr3q<QIYYW2qds=%p>Sd;LOc!v<R=vYfAJ5Kx
z9I*V^0=v`IoTHOt?mv3Kt&cw9@#90XEJs&{e3|jV2k&#|_D$~J`-no!=#SoDcfUuP
zIze-V1#3V^loZ~P*S#0svmN6jOjQ&S#}RQE6UQ+^D)KxdFA7X$sC3EvY+ARFl0GP+
z(pYVYqlhSqU7wF4(q2lO_UQL|^m={9qcP*ji1B#B*4Bhe2bb8|+NPJLByminVizVM
zln<6j5JbLrD&K1=3>6g>8iOH1YsF$-aP#I}-unJ6PL3B(w;KJF$}<7GW-=A^%G~&9
zu68FC7L=BhMquKUcyHI4WY1>Y{r-=Ma^r&(G%hsSO&o#16Ky6YObQ!T<AJyj6Pz-V
zNLY;4<Y#A`+`Wt18q)8@^pt{@BXeL8>NT2Y<LZ}Oc=kv5D+qy<kps7_3B(vv4~UE9
ziv8^?y#3~nnVw9sKKl2;gZq5zzx@SSv2tzJiCl6eq*IC$g)x?wU%tV$%U6hG#B_Sb
z{rmS_j#yneh5DK)bfBavU6fH8hro#nYzR6~2o)i{AKk&h0RZ>!-RHr>2M8glD$RH>
zL`i{C7Euapffrh=us&a_)`iOg;DL-b6<M}oHb3L^?1ZzkQ>Ry?E4SiWWmqiz!H76c
zkV;W$OMg73-yai2DNzbVk-Lf7-&f=rA0yn*zdqU5eXRw29PqH9omnjO%5{)NYnxK4
z13UtV59e+e!WM`^)cJth{UESWf$dtqUgsLE1w*?l1lF<UscXjid}(hLg1SyhIUg2d
z3`JQG+zhOBc&jpwogIzGjK&jYrzbv*l|&mHm<jb=lWc)zTl?DhI@cFz)o`7u41Whv
z;WlgP`k3xy)2Myy3~P01+8!F9_eqNO;-ZfImo)+8Cijp%r0v7hTATR~VS&);tG*vw
z&)!+Q%#3mQWIY<eYU`vDHfSv;y@=~CT;{dcUgh(j`vR}O{yMw6mx-d7vaE>XgfvZ9
z<x5IkU@<5ap`>SVt>)<Hh?A3JPEI41%Q=FWI7&UR^Fr%V`b-+1Qn7yN(^@EnH$k!q
zxNPR<8er8hX6_M$_V?_c5KjA4bOBm8>dAh3>)$s46rEK-R9hH@L8ZI9JEXfoLb{~8
zLAskkI;A_M8|en=PGRV7>1N1#?)yBQ*)wPV>tE|zM9KmVoD;T=I-48k%F0Nml2#`-
z%&?2RX_Yjds28r!=k_8mmsC?qQ}l?68p(S)Q^CFukl9tyt->s_%gZ<60Vt7oCM^Pc
z1|N>%DPH@}@WI#>&>mj?dfSCsp;?tXeAl0(W7IGz=_%C+GA80vklLLxy`~0bWkVZ(
zszjx@qR1sB8`chc+4h0Wqi3xLtoe+&h}rRf0wS3Lt(ga)b|AwJs47rXp&jX?huu`l
z!z<*F<v5iOa}na<^yq5{i3sf;Om+aZf1(Hd52Q(x0Ri5o+d2e-vp}QwYnkAPEK>c4
z#{8&HUE6`yp3ztGi|IllrgyVn-o>b;1r08hl!kon2sH-`wwAh65h_ilNf9emaTQeS
zg#>mU-~5|BTO#4U(8=V>JLW3rV%f62S940lwtcQ@49=($-WF`gHEq+UY~Frb58w0N
zC*~W^cJx1C<;%*Ul-k{$wIC&rqD}z^><b@M8bLIqUon=y54yDFS1^v%SKSP@k4^}X
z3GZ^q;+sW;3=l@jH&i;&T1sW!{w|m2FXBEuKTG)EXi9p$Ez{EK0%6qXE*RmMD>|gV
zuW$*@n+>?D$>Hp3-V;&tXijI(QQ{X6F&D_<$eJ{$_}$m#r^Ybn7KlDJu1=4hy6rS!
z+!~6GA(he5*)TW;F>I#1X>?jKVqu)Kt!RV2#VU&9kmRb8odkQAL(ke@+#u_#D<JWG
zay^4Qkh7@;>|ap*CVu?weh}tUuk*2RvT54UabM-eozD|eLYiY{^FvOQo;3}a&&KMH
zhzpx+hXZPd@9%kr<46+}55En`dPeA0PWcE6?yju2r?enMp7@I*2epWo6q<;3oUQW=
zI&d6^U4W}Uock^l60YRe&(`^p(gZ5ncBx0G(c7F<$@NugEj$2)7Jtn{-K!iI2QIF%
z^<s2ev<LXe57Bdd<Dv_SM!2PBtJ9*WSv%W@YXczie*5D_zz^var^c{r$g49tc}xaZ
z^DB)|*<Y5)W~&q!TmxKf=9HQd#I#_6M0PnENoPsSwwGBoS-Ra@hQ|QCs<d$KNmCK#
zT!ZL5!IH_JP16mifk78p<6L{`$}R9WHJo?Ie*-zw68gZ4&fH7rUTQ`?!lWV7C|!hm
z8a&WF4sz`dll{U??;HBLpZ=YO*-sl*xs#eVRP3z+OCPfL$KmYUT$MfkSGS1IOsGIz
zbeHGmuJZe>@~lW4V_;72j-71tKh7#jN~$t-84TRNUQ9jiZ(g|Q7+*utmCX6L9fa^>
z<)aFy$2sbSN}1(TmxCaFQ{USPxH3(8rhoIAFOk)EHT65O0WuW_ahq`j`NmlQ=3L{n
zmyjtJAqS}<l$C^*h)^E_sZ6VWgTKBUhf_Hw2V)}{8Kn!-MVDuIE$B7tWx>cF=-oTe
z!+y|7vwq1)vobps5Ro%cpZSGEB!;g7Tf}?uO|5{IYG{$g^_A0=G`m`xSoppJZVT6t
zMeD-|+NVI@SiPL4E`zO!rlBsmv5!qIbz7Hj{-?bnU8A_8UgLGjLxJ^om4_iAG`$YX
zJhd=vdRW(=*Smb?*u$3nhVn(Xg3(0Go5uX6w8nD;TiV>8X~BdsPFUuxD8<QE|B!Ww
z=y-o^?WY+PEuOe~`1Aql9DClcO3Z4eiuw*rdU;R%8zrQ%ITZ~hrg3O>Pb%kZQ@Bh@
zR}G+yUThRViIHAy_+Hcrld_Y>;K#KP#L;S>^vxBcm%;zZM^Y>f4%~}M`fD~dHN__)
zXi=A)RIO=kJBvHVMRSH0uZA(mt|qM%Dm2Z{&%ayye&7h1(+Y^aDwEMU7;ui6O^Iv0
zufl2uN4t(5>XG*t{(u9e{ekDkIu$U3Ga!dvKbrU{ZaHbb#Ip&WSfR~hU%lOPceSIZ
z7%}sQZ)r9;$wyJ^kK{#Ls?qKKTx1(vQ87|K0coouv5fh(c;b9~O{0WLcH>lcsfM5b
zE!z2C;Dj_!{(6Ah$}fKz6tX#JPu=H-XQ5U-tNS8==fQJfz(@PL2HH4auE760Cxhgn
zpW2u%*J^k|<4RfEdh?rJ1HFBDGy&X3QxnU2!NR#7QgeNBTgeEQ26*0vEsDRjeNgme
z#qlO|Iv~<k%VFjOg`KufS5-b;n^}$TOyZDgF~)f+EAELOoQk`UZdB}6P?zvcG^KrI
zIAx|by1rlrMwK~dL+8~z4<$RiD>i(7`-R8)1L5#ah*TTqnT$o(@L~!4++?u{&MF66
zGY@@-YfZ}<E%`l{=iLnxS{jo>z%9nb%cIO%r~BWNV8+C<>8nDj?wfT-*OS+WwukNT
zBOy`Gm+e`E-Sv>(*5TGPV@6Sl=m>3wA~o$@4mEg2><>(oRZgxU-&j7T+(pZ+wxrh$
z;w$Q;Omqo~=i*2@4{S_{2<olc#kMSMY|zMJ0!aN|{~n&Gh2j2DUc{x9F!aC08)g@L
z*CxxYEl|#jAg#AN`SZi`5C$=||An1Q<Rzl#1-yCr{swx5bLep$IPS+8jLH)%3$}`l
zatF(adU+dhJ0{vv)z5uLB>szOu3GSh?(Bqc^?h*XDCBvZ^VUVGD4R9>u2Ng(hpwFJ
z52u)HT#CsCGnAF!u3pm+*k2fP`JEp+8xi#wqWbxl*Ws|AJFq*j#u$eQqdCSDwa1K$
z{t&`!eA=$6e;d35QArAa{d!#J9Xjxh#bESIkL;?L^&Qc%2ZBXcp=}{qx8LvE5|Ies
zz~KPZ1cg~l>C=^5pv|wU^HKE^YqwVp=;^s<`V6=@+XYPc@$9;JNLZjqX>eEiD?4l1
zXPb$S0W2myK?BO<f_fNDo1#W~8i4epq46t(oW$CrD%zvl^MVxcL6tJf!v<5Vu67`Y
zQxzps%SvL)@_K(GRM%v_iAhSrnzFRYp0s36MPjU0mBTC*1{NvAgYe0|8KKY`b&SX<
zQS_31aAbZ1h5eiXXG}&rL$Q_JUH<03`lMJOR(92RWp_y_cmCvJ<WzQN`5RU5ME#M~
zpw%aUkN2@hkvv2Szn_)|ZEc`T*a@lMFMVs>n_Bbf=c?{#_>I8;i!25EhSoyVmZ`Bn
zmqjbq?hhZ^>oRfkBz(aLOnGw6#9605=0c`CS!OUX(p0cr6iHXONuwdhqwF^gW)%xT
ziR8^TImkbGjERvhxVri~9OLKUgk|F^NRcl&Mo?CdLa*(h4mLpOXV=qU(YN`W;}$~$
z?r8z2)8?XZcEX0y07%SW<PBjAx}nk39b7dx$9I<nxmvS*+hb^3vnbvba!OUsEHz$I
zL-!6Z&Z-4ibqxUGk&<@QXZ%`y3l}Ay=IMRbG<^wZ5YE3r8NJ3a=us7OM`LY4pU|mN
zT%rdA5ViR7EJZ1Fru*7G+>m&SsX1KUGPM6yQRTsoIN4spQ<z58i|NowI(D_Ss%ETS
z+JX7nJr3Y(Njf<L{7}w0`yNRCd{&6R3!0hFuPV<f6AiV8CTt;SJwLSEyo*8GZjIJ8
zB~d|~Dh#@7UUF-`4{}sr^%pAR5-aB>+W?Jp@QUtj1V01B#u**Cn0-5*{j0+8&byq8
z>&p#;lHA_aZ=__o^<ARo?{lcZ1>tN%vf2otG()<1S=#U+98)d2e0lN|)UwH~gXNbu
zRmh*B2DK;iMT;`|Jd56(M$W-3yW0TcXuzCLS}FlcNR!<rGrov{@j8lBVkTN$ZGkG)
zxl*Rhsv`a?=7p@5$`Ve)em2B(%9P!TM{xIQ7K3c9YxKEAWUmAC%5u>K2^^FaPs!-P
z%16!3sQcI#eb|(#Tj<;+Z^IQ@CZ=mj6JZe&pD%4THx6gl^TyHNV5Uj7F_e5(WOBwu
zF(2I4Th~VM9W2X?z}t%`iXj*Mo6v;m6Q-HLHCi3B_PYPba<To+_-afgawUqIEH65J
zbgBq%Im(kd3e}z!=kyv(4+`4lZ+aYHC;8E#`yIOHzS77dz$1(YM*8~{F6TxEg!E<s
z*&YS|M{*RO5<%S1gr$|W0p+Kn@~yEH`4`c5vUkt>2iB`yveyWWZsQbIZDuW&mO`9Q
zzqH^&UjNWS?ZPoIu}fdQJ$v$r<ghOb<j?Tek6@^^UtYb%GaDt~4~r;ZmwihpVtBaN
z*qyzi;JxtMvm}R>tH{KWEyQo6Wpc5+Y0H8t5*x$-*44F}GJ^<eL-gBakJ=yP)Rz;-
z@Z~5i>0IycS?Y9a9*6QkNz>g|ghnrYR?8b1Dk&Lwy14l3Q|n7?ZoSLP88akoVvV$l
z0+Mnl=igHm_(~dcr8{jJN#N&4N*kw+<28<myW@DOjEplwA5L_J!|lP-U))*@7=z53
z;B`o=$|nOPLNa|0Vw`GGP%w}+6Jow5B?CHD{zbtrfRx-c@f$oxgxuzHMhFNt7!=I1
zMSOzZ6YFsz1@(kx*<`sqLm9g<Al|+^n3@36t<dKqrKocgZ8Lq{YPrSo7J&n8vv$!p
zXjmw_s{<z+VKZ=}*@zE@cG*qAC~$*=)01w-mp&;nN=Bo07U{{gLC&Ab%sg{FM-;(Q
z<ZHGl@i@4LmHb}K5FKQfBIu;mTDC@?czPP!=5fZz+H*_r|1QJ4Ua^PQ$>vX|b%UPP
zb7PoDB6lnEyd+COYK`;pnBAGYmJCFG|5?My+e<g~_*gEGPfl7Atb#j*Hp6ITfpBHp
zn9Hfiw3yefw|e0CZXbPxqUAt_h3mEb@Z@R01{X`ZA%%X{*?s)v&iXy!+{gjd?)z?l
zBg@$t8QguYzMOj97?P5e>qfx)#|_YfKS05=$iR|{YWS;uzS%VnchNPTwfV~OP0BOh
zfXPZm9+gaSc|9kwgk4sYvhHUiJCBPT_Si{}?%8ax#3S<;IU13zadwfhXj`!I3acpq
zY7)x$Exhq74OMoRG`KN523ta)cZd3CBxV{%EvX};l&ZB$HA6)o+%$ytFRlRrqCva{
z3{uzL>@0=MaA|ffQ06GGy|M1Q)(N3#0mPf|i=4igC*hY4wOqyXJ%$%;g>fxfNiA8{
zQ>NmO(THli8kgfs*yz9N6}uT`buVJ(FxsWp%jp(`;12aZNlZg=hP8oHg?bUOJ_!JR
z<#z%7idUmzU|=#_!>Pl-KDRR$#r?5Y>P<@<&PD&0;wzJOwI(o3Zc2@g6jP5PjZw<b
z^YQa<Uu~O3)mJID%VN6jUqjQ;l>&hdY^0~xh|0y9OVQV7&~)wFzqH3Vpxjis80or`
zd>rYzpnzF_E*YKTt9?NJ&_bW`HCPU~Yx0Ob_WJ5~_@q$cl+{a&CVp)zqbfA3a^4w`
z%_f&YbY6gNoi(3u<1zE`7#bZ#h#q8++Wyu@wXNx-#!$dVsf78|Mf%zs_~sN5MU0Ix
zS5W01Q{^bPn{$7$-ac*Qck=9a9^Erq`?}&7Kg52~d4q+Dh6eek)jbi5SL4kO8<|LG
zzp{ds1-CSWZNvMM|IBLTK5k805{r8Fp2J$e;IDGu1))dNL~4Rv&0@{)%IF#$0|pNp
ziXrnV*TmStcaIDhz{oDnAl0);7nQxdy|u*=+>2h=1^Ffywu_)uqfaACdGFz=t)Q&)
z`5>E^OpQQZr@hflEax2241I2{idNTtrrq;{ZBwddC6T_VpnaV~*jc*wM^{>&xSZ*6
zWTg3^iN36uF9{NS!lK0jP*{Jozi4QF6DccarwLCIc;`2;sj6=h3?D40jOvpYm?&|b
zQaY$gB~U}WfAD!3JM0-5X9b0&8NMJV&9Ve{BMA_q(gDHrodFoPP_a*NBP4$b7<5W=
zc%_pL`E^oNacRH;r8iJ?F}059jUIs)kPjUym^i;e@;Nf~t(wrRFa?KvyIP-`wDsTl
z#Xi`5yvS|(M7GZ6!Kb7=i_81=sOe>Rc1xZA;7kM6R$426R4TI^^Wd<)11IX0rFXGO
zqo(BjYxBg}<q^0XMvESTeJ`$$&T~oyF8-lAXrS>6@*f}>1;m~#*0N_(-@+vpkx`ib
z`0O8`#e~5qUV;D1lo=ypzMQ)Z9)|(X?XaOButBj%7>~G4c-fR=gEEz2433wv_}^gE
zZ4Dye!ncBnhQvUCgq(;KPoA@%-^B4e%*(#_l&Q$!<xxZ-;7nGY<`Wj)?+xa!Y%A7<
zX4;};DCqgJpYdaT_(uo&8JLY|(u0QPR(&Vf{@$>=_vJCYe?jq-QpV@^&bRXZqJ_ub
zb}l@QSo;(>iP)GlWwF$l;t3i64|nvZa-9yUw4gRMj3x}0en-t?WYJxs_HYG#sZ<bp
zPyur*=t9J?8$#Q2v)a=?k2dy`j}15*>Zze-hDFp-l!iXXu;Jn169-oCHpdU@A`g&7
z$l?1--c?+{A=woi=rVk*2b9J<t`B#2YJB$BT#qTuW8KL|LwwkZ=l@oWZDs8mm$zZi
z%_e7iXY6b$07yc9Dw0Jr(2%5*piNXul@vr^j5<OIKRp0+_y&Czei0Aa?PIqwQvr9S
z#$f2es|Y?>e3bCArP)oWA)hO0+no0hd$BV1RNym(GT}meFRW{AzG<7U>zn&4DV1v-
z4}r(a$+a*4w7hrkcvWO$Xu`LQJDhP=s-{+r&tzX4$1~mI+z)hmzQO;xRZeFE|BySY
zwTgv<GPpP9Om|!494c?jQp2((yLp}8ZLIA1%j*{W3q7^h_|L`$X@{Wi6ym>(Li)41
zh6efC!|RZMD|w@X+5YFw*H@edNOpZacJcl3M-n!4kwEBZyatCKb*aZJ%;zj0JhGqw
za4rhXCpEtE0&v5M)0?uP5K<9JvjvYM+yYl>G+p*!%MbLaY3;B3Q`;I4n6UaX`s<(t
zmq9y$-P@ydKMl5{J1@u})5};|_a7ohzl3qE8fPb`(Dij7b$$;FMrMwld9$rr@`;id
z+$X1vS6p)fGH%<G+8P>dWYzwp@@U6)SOetd#+O3DFUQ7$Mez!A6b)<fC)pLuIgW1i
z4o)!B!|re2Kk|QZ3bl5<Rn`LU%b43LK_F?0t{WxSx2O6Y)@EMsR|aqY9N$xYJujSH
z{66UBV-EWHOaym$LUu*p%u3?O)D4Y2y2`4i$n#Q&H(n6F--`IA(doB!Ps_7*kGw5>
zgbhyTO8J_RPl-OZPofn$@D1@>Obpgss|+^H0ORAKtfQs-@t7Mzsf=sXY4QT&;q5H|
z9Ky0x=QScBCB1$;V*h)b46*A41a4CYdc-dpZ7@|$u<AcPf97dvinDA$Z<@Tk_l!vK
zRr5ZNS$`VSW7y|vYJ{a=t1BR;j;SHJ>^tSemQgQR{;05zvm9FfT-L<Z8*WRZvUesf
zi`hvWM9^mw$V4^MkiD>tFnEmZTRZ?Cdxc?kzb3`vr5clzT;>J?POQ~jd%FBcy4Ex7
z#AaTn{q&<xDh~JwXOq__4%F?8EiloWWK5Sx)>&9t_2B`ajkjabXaDZ|rrN9fRmYot
zvbR9y@sWnUT6{}$9jPJ-9}m@@^D{!Z$W<B1Z_?@nz1zn@?-!s)Ea0v5<<0SZ5j0u*
z{-h<&STE6$q9Rbl5yU(mVai)%Pa|x;qgLC7%?$ILR;!``bm7JnhNJQEi%3Xe>|UiW
z{aq9P-k%X;U?xiyX?{wF9ornWP_cD%6r(11oZJF>`GOjU3VGdpF713<IjO2obPTxZ
zadzR@W$Jmtf&H?OpY;bjl7$Uk*i={zBD+u7tn@U5oZyyKz}J73)f{Y<Rps2B=hi^)
zR=chs$zoOJRNhrZ%gxu*bN*DO-VQbT$|Vq8P#!JwHg~ehU)gj?APGh3;mLZ%Wc0*1
zz8?iBZQ<A}iD@_~yv{_GIrSY)Z@TMJS=BAP%$7uJp__P3TC0*BR}lOYeHLANZJ9mI
z1=mj=xe(N<v#@US>cLM5CaWlfHV={YF5NwhP-<(G3&!YRHLoyZo*Wksk1?0Ff<+=W
z;+PDDq0XnV6a6-qoUYf8fO5UO&Ts7Ln>Pv$xiSyb`^bB*$zV!1NUe{H@k?Gpvb?hJ
zDE)gA54Eas={kSryXZQm8``ro8AbjKuDx7Qv$l<Zq=vmq6G{=%>In~smTj^5CS)r%
z9!;XT``J5HvCyx>&h%{>i!4j-$3AdQyTTtJ_Ya$U8(qFtsZ2dkWx(ds|9tH!B1_ae
z?XOH6ahY#a8>q`i`pd5z`dt&}q3fcp%YK5_t3bE1qgb4djxQef(znAO({u_p3mJdW
zb9K$-%l|;PIICBw;=sMZ8DB_T0L1G$bM|rVDW=q$M7zu+3vYW9#OKiLM!@=>vJtn`
z=rG<G@J5<!byL@1D86yRvLM1<EzXDB2*x)jUi|2E89EoJh*fYEQEi*X@3jwGxg4`_
zg&cWvoG^odx$(H8yCxZ2^=ew=^7CX0Z(86Oxm*QD<@*5|8bb+Zl$~tIRY0RdJ=Tw6
zjJ_N=ke&(URbvf$5ao0v<!ErYC*&%wJ8po35OpR+KF5^t%UtwYfJA%=f{p0*%EGwc
zJDli%_2k>WeZa15&yeH$z2i-a(JQ0u+U~RCYkb3B1~~@H1_mkBUrd`|ij==A_2N(o
z*Z)@F$z~6q4k?);6G=DyPh{4*ZZLZ87g%q`m{?uUR=6G3TBpGGJpCql^d$2!l2zCv
zAq6Ai@MC>l$U}rfdIYP2z7wUT%v$59&L**D9=G2^v0<4gixe0bLz@#3<$rNoHN^Nn
zpdr}Q#4HH;y#&6Wp}kQVJ>PBak-Z#<7Ny-ndeH&~+}gKzd+sefJY+tRF)rqPFe7Nr
zS!}gc9<dK|P!F-?14D2j<P;Z;Mkj|bm)it;_v|NmU$C3l9e!XufdutrNo+_ItW!$p
zVOHSh#Pg_S*SiWZxc^!T{&8RLBo@maCoh<3kEBIKy>G84r*_VSq9`q(I7Vsp*rnM>
z&voVy-hJ!uFqax+9A|LfzoCKUY$>i9LlvyeX&&GQ0=Nf@s2#_{u{?b06&RqJ-bCwM
zd!nV#SDt+t|N9--oOsTrZo|rpM)9z6|Bs6P7mZ7+4kG%VTyL$E(>g4;6IPsfkebg)
zS0Ptk6UD*f`W+s@U0wbj&7Mmu8<tj9aXDusVTAyJ!KLOvJbX$N-bm(nb@F2el(rB&
zUT>rU7YkPN-FvA{2s$-mukK-`&cC_4Qf~E@CULo;XNH=EGgFlo0{o_JeY|;Y{q;?f
zp>v)4SA5*%Ci2o70y5#&VZn>;{y5OB((XOwC#zDK&i5oT*PHc+w%6Y9ys~@ua`LPc
z^M=ITXLa;^*dRAz=i@mZxU?36vlx84yJHkzz<})2CDJtynaCK6--oHRn7VeG8oeU;
zdpuvaR!XO!jK_={=4V6k{P^!B@0SpjS<4V+NUW_OOkSZSc@by2_9x0nRXspJXJz+#
zNw-KXbX5!~Vjgoy>vr0vggT~#i@iBXgva}#!+N4$Rb7U>?<UTlV>0%kG~{gR?B$jy
z$!_(ynRKmgPiiR3v3s(zwo_8)dz9Z$aj?B*sjvBR(3KOB8FFwyUV;9C#>R3c+cH5%
zQNPsC{ZxGI>wy+za`WZ0A?URMAVzlHBzdw+1Ili8H7Ma1FbTQvM%oO6n{-K?$R!bE
zv0<%`iN%9u>t@2V8A(ZVui@U0rz3XKsej_fg)1}zBpy4!dB{eLLhw`O^z5vzr)T>7
z9Ch~!FqlNnK|?tUn6nZ7y#Ow~f`!cED-UJNa$^NhjE!~QhA1UN$SpKd0V!~;%`>fK
z6A}m)Hw~4Sg&p6=p{D?e?s-A7ydii88o=50lI$JUp8^MWLO_>#JyT65ixFl3717(k
zBV^`v%8Dj`;XU`$>5B0LBvj6trX2S0<y>u)z+5=877}RB9NV)=doZjsrB7siL4-3X
z9M+E@yYhuG&%57)ph|HOD=4fS>4B04Oz2Yko;<)H(Ux5_waH$xNH#l;^c0K{yd6cv
zYiQA}uSVE!s^Qd}-jH`lExS`l+a;QH*8Ho+?F`HM_(RK2CTJ9u6Z027xR{Q?<ALL;
z20ggg>EOD`Vo~CKBamh_SI{}BuvjH>`+V}ofXzv71-qH9ZO$bdVW|3UZADb5au;vj
z^hiVEmGK`i7`3CrSHIgYR6ecdcSFnc)D(-aZ(3_3C~~$)8M_<>=K+y!1LupJEHzQu
z`;0}^MUom{2L3O}oSXs%@PqT@uyN?2+FY58IS!L*A`a8x!&<x4$%&cn1G<Pk<9TkL
zIj`Q~QReg{!COt5qI+lShKZ7)%H?{?<{q7>cbp!tBR>Omc{b1S_t8yY8+5hHyJ^{`
z9ROJ6f(*4&qZ9SeNL{0cl~^3$ui#;A_JUXv)2Wq4Qu+TjWJ7GAOY;ny-bDV7?wXJX
zV9FXir#8O9{u_neRkP3`F!4cn9rr!;Kafftr?K)RP5iD>9m(v(*_%AWLZO|Xv1Fr}
z2ks>|L(ybp3v=l<_wbVJ)}LF%!3g^E6m6ZvW!bXU?w!=K@q$|0?GxLzZXYAE)d3Q;
z5<0&+n)UY(ZBZ^N;tl5qb;>_K-})fbc3)$HUU@eu6oCjKk<q^A+pOqLahhoV`4nk*
zIGx|KV~@OsZVWCnE)#AvO_!Ubj=-Ve?nSGxHo$XBmQ?!{BEh)*<t;WTVm7RbaV7!F
zkV{NM$HjnST)7S871ack(hSsHPb--v+yd}*>ydjf5Ae+q6Q4TohT^rl><5~NQbhSs
zb2u|h#Ec(TRALs=SQL*mS%~~UlWhkBQu}KbFTaSeMH^zEKvB&Hdlb2^V;sAJo6mYm
z%5f0h_kWvgr)X`UcM+raC!@>No>#P1=UstRT1>EKWvph9J6U`59|OI>r8{m`k2O=a
z?Ba1uFw%!?o0%0?s5*lc;pR!^rW4Gs+4U7F-$%@C!)MK-`DBT&B+@W?T0_2lFK|^f
zBeWd1M!{)Xi%`RG&SuN7OFC@3(WvBd)K)aq6!N<a{QH;FGMc|DKn|OMRqZt0!Js-z
z)Rysv*jUesXEnw!YhfkXgjzvP*Hf`>Qdpq*!W(<`Nj0FZ$b6>eHl`cTV(weheDw9o
zkD<6PEJ4Dslx!bzIy$S1f7CON*5k;84+GUlqezO0_ib(vMBisEVBRpXvI-whtNq;r
zip<Fo_V9d2x4UC?qxhp6RdLBdr$nZlHUy(NLf&4<F2KwqNkSd?O;NI{2fAf~+a&g;
z98Y$Dd}UbEHTlTB8%vt9@k3?59+|_FpWP4W&QU|TXw2Sgsy~qq=zUTv?7w^Pjn7M2
zGs^jN6(Gu>m*Gy(4et0LqAhb5wMDn~WpBsDHQ2Mef49T4tRn9VI0PqN+#6Ya8}~wj
z>sBu(XFqMf{D0R@Gslb6h4^H)8!hX*N=TVH2wRK>#>IBzTz-*0mFc>H|G?{Kee5Rj
z9b^hHsrP+wHDjhs{rZ<Wj(1~^hIy%C=kzfI;wINKvhH3b7|lK;6BhaO7DV>7H74ql
z1pb^hJL$vYNl!prOZPj;hSwR=!~tDVSA<;Jo_FZC)KT+^XkJ6=+5_aLk1;F|xSl&A
zz_$2mX6Spy9ai{Rx)7n14oOMLa-+0*XYUc22J5=>_JBln(~+(0>#<}o6$XVp(_jX-
zwYo{wFD@W`8-b5O17AKPy=ci2E&lvf#N>-g#B<jeBknjBsi?qUI@-uJA`MR0H+DkQ
zrOnO5P}{UtAUo&V(SdKDu587OhxAiPxy}0l+Vh-*Gt0_c>UjR3DT}^H#riCWQG$`%
z%)dmf{Bi5R_vQU9?H%p?^x>U=x29`kcQ_8xAo?uueo6L*TibKRh>}F;H^WSB8yTZ$
zw<Z<faI(s@Iw^*Uix48B^Uwub2BrK7*Iq?EPd!2js67~rk=Bc$)1$T_E6V4f0S9l~
z<;N>x`Nf8=(3(f{`uZC1<h?1&X_O8V0U6Z{a<CnP+o)}Ak9amUP#2=46nA`wxkFEx
zs`~o#<Y&h0)6ue}>LTns30nv3%E})Wp`^PBtQFmuJCUYppc5lfvD;VFfPealrxeBi
z1DZqdx4POMgl9fdcq)A!O#4|rCIDeyrOAH-$v$3=shE>g?K7S+v+~S)1*QbRbicg3
z4~S@RaA`mW9y$D&|KW!sx2I14A(29+NmVnrgUw65$myE4PX-dcsn%|-T~2s&I^DO*
z9Mja*8kf9oGx|hd@O8lY<b%)m=;$=KsuPB?$s@H^R+`UgFm}RG9(hefU*1GgZMwI>
z>w<poXDtw5r9P^e(O@Tq0T7Qa{GV6Xy4|C9!_lS@w%!$@>2!!h?$f@WmwukMj<PT2
zyQ~xuz;1r-(^VXi-hQFW#7Vt`LjF?`Va-OplBDm$Zsl=-G2B4Y%xGZkD$D{InfXZ|
zIe+>i^+T)}QbY(^`fu{;BqA=c5XtXfxO3E%N|)?HXK*wn@waI)L*f#y-5qQ<g14UT
zY&&1(<b}MP;`)a@y)SaefMtG6QzqabDJZ%P?!?mU^n)Nuq?lVwT#bHVF<L%Y1npJ=
z^y(aOkQl3CZ{`X$Z<bQVq-8d7UR^`S<kf6BXSqoihmfRd$e>rMSU&R77*Z6Wsyy7M
zgZz0|8kD~FKs1JszyD^7#~CSdysMcz9dq-xa>R;GNG0S7Ml!_lGAt8t)<AB302M5a
z`sdeOQn{cSWR2ocrpfv$Kc(f7@>5?738s+BB)LBV)s1We(jsh9h@<Op50Am3Zy(<v
z7q3W$-B*YWkRZnw2=A=)P1|(Wb}*D&Nc;wZCcgn2Gev6wv{m=T0<->22ds0zTr#v|
zH0+S9jV*|u!YPr=&{fseAdmM^1!1~znY>xAR#canvNd3*mK)!Zome>{X&##BWRRuW
z>D~jJ-tAg4z7m{*PD1K78T6i({DEluRkj+B?I*e#J3cbWUrn2eirb{pt?mA2bDdtl
z?)=RYap1Mkr~3)9%1w>5M3%LrbChE({I>yF%}*U~6!MmaAB{f@*scYjwpRaW5ASH7
z<-OjRoUIIE)~AJh?v+DNmwYJ5;Q5Uljk^1<Y!-g^&!J03oqkRFB<`N~&t#VU4QsBe
zi8U5M!H(0@(>j1y2nO>27_~$@wjQv%9W&98z%R?fCcIb&a)rM?H+mFSp7>#Is^be_
zJ?}W52@-T9;CJXPNsM*!d=v-AMB(YwFweX1(<|}z>W3OAV_<As>Uhj%LyD?F&(V5S
zP39U?XEG07hp%~;!DKJ-$IIE&KQZ9Pgbwdz$gY!GHV^hL(HnAy&(rqtnqRP0>vD%2
zGL-(!ctOPgzy@4freP?^*J`$_?OwRn7pHz_v^FD`{e2?lpQxjxBv>b(Dubu91vk9V
z&Fh*v5hPR)c*&*aC}$)!3l$}q)ry=fBL}3@J42FaAcbElju9-a;E2%JTR-ip>7se`
zRiKG(3#!LNVb*tUU0wl26@tb-E@R~9&JXtn%nT{Bw@<>j`-tesVP<nWavVoDN0-Tq
zhM<J%GYsvM#@)c@2ZsiIW+00-7~Zn6f7KuU)ucQM(?COMi$76mh*AxHoA~UD9u+l5
zkepuB7-UiBS9S@mN{Z6L`U<Ag))L}=ZXEbifoyAdB>EU9#kmge)>lgNkmjH{zMFFH
zFbxN`O8(}3@R99-V5|{zSj`vX*33(YQ}(@GB2K&{M%>Szkiz#@<;TVM^`4!lCOiNf
z@94LyXKe;qU<Hwi%VNt9zh%Jd9r7*Qh0F&E@Qw&>n=NW7p6$}r8TKGy%_pv$CJrEI
zwM;K^b0R{kHTjM9nQw|=_T-t3#e~W|D`2W|WS%D6hc%hUG;ow>+B;9@n-B!I74cjl
z=G7I#{zg=pVm{jUWX@F6g2&+JkCPMB`a~hCA;*>c+lug2Xg4SdHICGkCx@!Y+{!A8
zFY?8oNkU0uJ4@qhDK=}2Rn2!Nm0_NsX-$lvt;OMVMG;quz+QI>Rv4^%1n=+ThwSX!
zNj^};ELahUIJY@}KB}B;q>%q!3qWXGfXBoB`s=fW>Thjk?HWzyRDL;zurI?aj*gD5
zkZTK6R2D#;9X~%OeSU_!ea4XUM^2j8wN2BI(&0BQ{WhO4W&a$8xUCi^CAaWlPmPiC
z!V#~a5=85<6vx1F*xZ2swRr7!xY+Xp+b2N;oJ4;Ze7vQ{`{hii$h+$Y1TYNj>(rJ1
z+vy2tC`a%?dxF!)CI~YK2n%-s8fgFfW6!UKs+6X~wEc@O-n7PehVirl>hq#mieFB0
zXnqqEDA{AlV5nmM%~Mzc0RMl((K3{*%2M<)<aePWVoX;UGGCrZRql$8H77-yenJVU
zqYG54VJy20F7JNvP*x%^WQZ`s!{4H`_Fuu;{`R{X-B9GuD|NXs*Ro|@W`i!y|5$^b
z=e+I_qu09@etsR7i*u6Z7P@TWTufgpR~#woSBj1fJ-;`kJS$(@tN@T$Vi9wiypodA
zh$g>Qt>N}|e>gL){Ln6ih7Jb-TExzX44(ik4d0;Pp>xX|vWS?bmWYhq0|EXCE$MHi
zEZfOjoRx+%xaRX4>^W+kZBD+97YhgU+n8+oxERqvJ`D2s!6%Q!>|Bx?8PIk33)=e=
z$Q<MuSI-C;;_Vhs{M!9ExY+w|rO{mIF>$UzJ(s;EDKNZ>axP;r`Fa}n2gZDPYI}Z-
z&9IM2!vQZ?`*ii1dWydb8g$;s@;QZ^6$!1I*!mnTOp>vjV5G#$H9ej?eGQwM+<G9g
z*v+Qk+vokwYdA@j6dKwqjJ2Us*e`rE4_8ZhGxn}A)WjshmOOfV&?mEjztt-J*X^8I
zjW{D~*rV;$y^W(AkLq1dN|vK@2ix;DZY~@2JUOp49}gUZJaIRW^-UVn^y)TB7u&bs
z)d})YU=@mWCJuq;##oIZ!Jcti-__L*cr1N=fo-Bk@awDp^wQbfh3NXF9W!=HF|$kK
z6*TeZ!`9}}m4&~0)E}7W`gd?zF44#8j6|`hPKWVqVN!6H^#96Jr73@d9gn(N@x}Yt
z{71zh$Zpc5U%z#l)s$Z++xg~A=4CYRg`I9#eXM{IGf10(RHSt}4s`jY`$h{CI|Uk@
zd-Z3YruD?s-1_VqqF2r<BKAvY^spG5;o87PgNwzmQ_b$ny1C^huzTf;roz`OXPw5`
z&=zEOdqym^59cH1Q~qf;T|Di_RC%ZTagp2<DH*${fTd#_HKBt^@ul$}x~?<~_~V=K
z_R&oQGNkUu;@Zo+w=K~2-ebH*Y6G4Wz&w(PBfYM6^iO&}&wFPQX?G_o<nA)e69+{o
zQ1cpkCCI~RrQ6xxten)E{V1S^yy=M=b1d(i+KnwQ|L;A1dn3Jp^kqHhW(akj=d!os
zs?2MIN<ZC-C4QA<#F}%8^iiob|C)wl&Nqb?%=4Jw{mwdk{|PH&T+&!_ytM39VNtpH
zm#B~;LejI#@!q%O`0Q-b>*4=TVl?QAwXM-pOb7b~0(c_06un-Xv?3`ZXC20>YjNgy
zbXGIcZLun?6$)b`$Fg*7UT(n8b^RXTjBwr6U<RH?a~SZ1u4D5pVFWot1tvN&EY-!u
zk;3Vcag@;XOjXN;4|&)}Srlzd5l8E!%PuA59qtLPlo|o@S;A;FjKW!TTQptcKQ@F)
zKn@@`*Cd8QQ4LRFr(74VN<z1(WODieYZgAerMcRQe+fTk?!){^!Azdv^1S1sxE0=`
zUY>=7o}z_5Y;*;^PL2L|F$OB1E@(YHz0>pa`i_qA-d-9O0~fn>Yc6WVE}{W-Hp?M=
z#%hM~3fNV(g=G_x{<e~b!<X;<a$V5B!LaPCJrNLsOjB0I1hoOxQ}qI!L)U-m(ikqO
za9NyU<0`A4WVrNm?EM$<>-wE5N|*yw#|s;%95&wKTLNxB0!E>B1#!~HToSS?@I&|T
znO+Nh0K&hwjZ~|9+-iSAfj4TQg{&=R4%`@D{v2gOZ0SO(P~r|Zvoz5gj{Z?}D=-WV
z^IW@MO`viDeWy3wh_Y8+YCPvG2yH%a<U=3+6shpe?fJ*+B=s`HWHxRyx{wdW<Sx)~
zk_;Tzr&-^$?EwqGK9kX_dp)q&cbVhbKRr$UUw4q>zXN?rwg_$dP>lKXN+I#}{s9di
zRR$T(OuGMbRmxhIZ(iQZo`J)f*ALLOazI+G-#!ztA0RN}($MfKQ8KH>v-w{HD`Mo9
z=Kc4a8n0EW7o`5oU((;;eIiQQVP`4bSd()=#Ayyiux-`w_k|M=T9ukj^XBM?V)OAv
z)GT2#A%S`v@ogLgfo-`E>cdAf|86-|sfIASM|@lS*HceH?e#CaK~CI({=1D<^T%dl
zX7|(ZZ}d16=8_i7U__r+irvxF8|Pb^-@+#rIfLIf<E*<thX^iuoH`~FXvnoZnFaLU
zQ<>7&i?FF>$AApa+XghS<9Dwo67&UH8!F^Da|8|;rd>*PbU58(vE>XxJ&>B7#5*Q6
zK4Wj(mPW7m8eH=Eq4MjBzM4+6@AL+k2)9-Xbr@nh_-T9hwAWXuFcvJHa1zEYudG;F
z4UcA>o{{|SXc;!=04%P*=KuEaI$T)-0U@8~)p&#fi1_!jAf}jEThl6(XZ=-^x1af0
z^ietA&@j(JE$SXUMI@pk;;rUixh7VV_+|9o9R>9Q*rS_Ly_<J}SIMrD9ySmA(L^c7
z5fAfX^G7%w{4Yg#H+|#!W#T{^3$GRs>~T(L!DW}7`!5-u|BTS`j|9ZO^hY(q*&LoS
zvP%@CgapKyTV;)=jGyW+4-e(feu6u=-(lbDK|SF1RN98~Jf;&R^9uYJG$gRu>C(J)
zd+vV_E_1Yi_(L4wI|w`2i61G{B@j)XnFb?F^mSKsFQ)|rzvA>86j?w|piGM_^U=ZI
zE$R8;;&;>jgpP(rT|-xRBKz(Ys%$o<41$zl0XOgmAaawj$S;`X0=dTdf74gj^8dp+
zXvp&q=2C=rS@O<R*PyY`@nN&-^j1c1EzRUN$l8uY$Uw<9O1nSRE3STzA!7Rs42vuR
z&kN6U(C%mS$BM{vCG;bcuW}v%dbW`T@D!gk_Cw;B44g0C?7QD$MZeI8lSiuo@77y7
zVv?Mh$5d`YhqJtUYKtx#4yYxMu9I#b3bXl1E_6AhdbbHZCAmIG;&-!hMtN+ynBmnJ
z>0Q*XmHx#7VpAT|Ozm@w-1r5PVD2Vo+LsEL)7V8-*XSS~mHImoa*F};`pMom3l<W-
z&D!;KAV>ANLX9(PQoB0!f3`;I#{6`fm0<(T=l4I&k4{-Oog+DgU9GL~m%^hGm$Gt!
zM{0A>i-(%^xKR~lH+41FG!4tZ#7$uJm!ScI>9qUj{9MG>_gTN)tsHg=kWiOaH_h0-
zM(&o9rG7^FSSf6Kn9d)JimdN-zOrBP`IWDr+9948ESgn@+73M$R`bTmYBO&Yukf`1
z+w4<2$I+ZI_Xa7}ua`J#3C!nB3UvozTeqp_XyhF*aN@?SH#$7C-k)N7wkCN&VhOe(
z$+vT)?aXbShp6wzac}WS!!k?%E@Y!ZjZ$OxGZgCVZ6Y%Nr5@5@@rh=5zI)8pu~4r|
z6@5RMB14ghxqnz{^N0jS;lvz;9HB5A$-eSK!H%>)RwkePblFxP;}9Y_w|5k6q2YDW
zy?c+lo+rpeD19xY+l8GgJB>>kTB~sT3VzdbsSzn%Z0@fD*<i_L$}Hu`<Sl@xd}SU2
zkjoQpHs4lT9!uZ%jIJLqKQZ78SyQGM$L7-`OR2iJ_{q%?3FNdfA08cr^!HN@?bgA$
zrUGxy6sELRZ`+?NBur|lEPk7D3wHyxpYHORooM-fDV}lSiWx+NRiR+_L29|Cl^$jD
zpVx#(qwZ~L{+R%2z$W=s;<j>Dw+GiG58`1v;9BkC@<D}8e^En9#lFmKFsGvhGUeGb
zxft-;LSLU)Q1+N4`ZoQt3VLQeet6y(o{+i>vz`Ux_!|(CbkZ_59A}yIOp0zRth<Iw
z$u~@XM((5hY8%S~Kr;a-!mL*`7fH5lp)sm$4UXUG%5*ke&<r6yI8SflyZu=0zeYiu
z70>DQ2GzdOfATw|g$c%vwYPsL*oI(kmSapa*op7znkiYMo&<VZ>ik)=IF#r8jbrx>
zHm&?AX3s^ZJG6HvPv^iG)$pCH1xWRnbMs85wmE{509k!}0y2|ANI`ipj%@JlE5@um
zT#U3?+J*aSog1&i+Sc`crJel+TdqPx0AdqLG*UX?KQs3h*WkC@{LC!zt&!88sM26Y
zJ!|%+yFr(akXds6-QcEJOO2htp<yKLTf19`IL<9T5oeVzQ<j*BfMLR+j?i{-BxrX0
z*k!C5t_5QBz9?$SlhfghH4Zi&Sw3dJ=)OS(zHG;L3F_bP7LdLDjIc!VIUT%JQvse%
zqAXgs?f0ZHd5*0BB-YgOosc(h$mre2AuZrcipfXgmRRbtHA6GFKV|Uzxd9=4CtTSP
zi^LB4Mt{Alu5`X><>U5MrxBCbw^<4iy_vEehJmKsAc1W2@OI;KGv=raS!e%({Bw2=
z;3q_o#iT0Q(E~v~hYZMO1j6yLOGpCR`~My#AWqTyDkJ3MHnK0T^>|8@2ZK}!Y*BHM
z056;O<(9~*pdQfwp5ozH2nH+HOf3#HzH|neiu4w(@h}9Xi6SVyI0UuWqD}T<&O+%2
zg{4{Nlp?vTB)OyT<hp!bvwRMEpWYu&gFbWx7Zo7=nTy9G5Ph0O^S%hq1GZQ3a?Xe6
z2HlWJM|TBxRbuWpw;Vw`=&+TZ68qeJ3FA88wqc(&G2K_^u>!AOVJ=mBAsGSJpMV7a
z{bOUQs&$utN8VYRK!{o2Yk0`TwyP>OOCtDhGMMn1^5W=E=rFEdMhrBK8Euk1VAyqr
zTx--eG*26|6;%+sYRw;BS?c_~os6C+-(H3*kBtuemn63BW8^~`v+hIE-(x&gH96Th
zkPvHYX3r(1^LtDaP*{0u;AIm_fJ4>X(sB~9*wytB=)~+)UJs;3{V+1EX=Em@m03?e
zW3k+rKxdxqm{Y>YWW`WMEXKQm_PYWgX}6K1J5HTTP6zDZ`is!C@)kMERbdvX;5(+e
z6TlkWM}(Q#!zplyQp9Q}VS)4V?&@~%O0z9F9X6ihdmvMrE<QeC;4u0qAO;<!F(fGZ
z#D4L<Cp`623ACXByi);gD{^i=E;+X$r(F`{Fq_P<Qk%@4uGnk|{igyRBYZBRBzD62
z&t>mzj`0SD{aps@9}9u(+S~b=f9A?MeaO`sI9RYZ=n!uv0`<me#By<nH$Z=ij|p_s
zY78lVQ+=TMJVuCV65%e*E^6^~M1|S4a}FoEo%1;@VEh{Zy~m(oSEcs8eWcg~oqRS7
zfVjRK4PRhN{cMXY!2dG3e2gvpb7n<09b0NNg-MH_kMECee>j={$Yj<<p?heQF&AOI
zHuH!k5Nxg(<8&}J;9HO{{ujo5EW~?~^BB4%G=G)13T3}eHm2)m$l2ovRgEHMJ9Xpc
zZ(|T~e|j3j_tSeD*!{R3Dj8h-My+iuu0q7E<llUM)6?&EOPO^F;>K|7y2FKod-gk)
zH!4_^XQ(W&Je4r8o!MLb1ikds^>3^tqfSvF!8Xb1ePfT2nChR)6Jdx7ej!`mp<F6Z
zV^4rAef#t=B&Ji7_|DJt407TapcWhb#d^5e(Sn58#GbBVIpg>>gSjaZjDOL?xONxZ
zV_K$<J*Ysd#}ZM%oZVK-JR>!9HA<mDBjn_kFAi@2w{F*-rf9!>CTf7bgmAe8b<T74
z5Gj}C)if!LUo~z`mBoW9PSx(B#vyf{7_awHVu;$N&zQ}eBjYg!6M6~otnFTR)5=Z^
zGi2&(Yg~|S>vOrOLo|(C#5Gs4Hv3yoKQcCoe~ncu(V++LT6Mv`bf|`qYdXK4${X-y
zFj8d?*$6k=e6I|drvCdvWBhR|m?h~UTBWK9yJT;wP_WbUL|DY@U3jdXGBQa(;O+A2
zi-W&Y65Io+ie(RS>Z7Q<ck}juZk6-TmLVVZ`9CULLpsBm?s8eem|UtQ^#RMtPN)Yg
z+?onP#Pz;s9M6k2V$|$?irryDk``T&3w0iol9=F3%0NMA_?Cbq+@j_IcqK9sOJ_g1
z=5(xsrg*jyLZlLjjO^1RXiw0iu#&@Dh!(0xy0FOh$@~7v%b_S_nw|o(0j!CD@b{Wv
z;rkWnO<94K3O2Z#$S9V4T5gBly%$bH)^IOb<Nxm3XSg&zzSe=o?t8tLXVAliH-P>g
zy-5@T42)rR_KIIIdGlqD{Yd)FNu>ZihkKd2%AKFdp0g1W2BM4uL;KpssGL>Ou59w2
z`PXCt=7L4oWEmM5pwIH|WbylGs7l9saM$}={b_51Z?Cc>v9JGA)mS|}e%~yYcyrQ{
zU7<&P?V_74m;6m!EZ9dhCRf@KoT+m%D)iaLyj|~e^O1NU6g+m|_3@35tDWQon)B3T
zR_KRF%3jNcEOA3~&4WoGJXdE1dhm2K`^B97)sKttI_DxFiw*I2;5Smx+aaiTbd<qc
zRju^1Uy;7nw66!hS;utUT55El38W09t~z!8ypT}~>wb3uL*J%vO+Q^YYdvGfLo!lZ
zQ`r7uqX<LzPv-SaUiymISZK<J!QH{trF$~2t_)9G1CJ!F1i?GOL_u~R`Ea8@`#p!<
z;cD+0dfg$ey4nCRfHl<B$(Ak_&e|nt(5&xU0Z@{AcBsAd-ppTm>n(8?H*}=IWPMAo
zzw_Q(3mrby4Qc9hAI1_seJFE{u}-y^z(YE?fssx8_FOgo^yDqM1T5Y=I?n6r>cp|%
z9`mRCrRB>fIVzh;@8)V)O6FC~xQ#2PaVjU(Y`r%y^w!OU&@?ny-vnZwl(u}KP0+V}
ze{_r!dB)+$$L3VL9lCjLjP<OquTS#h@ptFd&PV#N2pQZ$qHAAaHlM&Pz!3Q;hv`My
zCsCh%_5TBvc_8H=QU-y^Ab7jAPD1O;*|x*}tC+SfF`s{1s~nCCSY46VY3{f%{<2Ib
zHY0~dvo#pbuT4IIO4@>mu+TAyNs$yx!P<5RaxJdxnPV5^R|y(dhhrd8LI1+&%dh?=
zrn+TP+=oZbepI1wOa&i9{yS(UD1-H@@-$l#KUiBx=mXJKBzn=sSmtQ@b$<&2=u%*3
zebL*!t6y`TJO@DSt~$=zG+M)wW)TUm5S?)O8vOhj_qek7mYv00W7^c)ZzVFt-sbL5
z%+ntMW5eGc=kLK{U}<*35wK+Hzl#Z^Uf=i^Yc}$^MCPK8Z+UDybR{SMgo@J#rx5;P
z9!vXbq(S~6N>@hR;IV||65)r`k6Y(DlMi+p5>-%wLONxJrA*A@^EhS6E-BnXWZYj6
zMg8iljW%=eU<>7wv9oiDcTL)wJtJqZLUISvJ}9?Ki6`#uN%i*5iiozp@~k&;7g;Cn
zU*^4!=OrEZ*bqjupG&ixdM~v&pVrQDm-RvD)6O5^3(Jp89w5oKK~4<7k_hsZRjdq(
z!arU_^N}3*2NsjoCNNGZVuR5X$A^=gL}!i1PwF@s#vb?YPFG?mah(tfr54AAk;4zE
zPr->OTf(adP5PLqs~1&J->wbq>TcoQ#HHu?2N5APNR5?#jA5uGP(Hd&y6v0l$-rIM
z%7X%iv|@XgI}a~wo$oQ@iJoAUy#vP3=cqX~>o2PRkn(0z%5l`btw-Hi)v9x+&PCox
zNp=~R_kh$&ACztJ)dgk@bXw*yG>7JMigC*R`Y^~l?N0>_+KJ~{!k-Wpf)^LHjaH3a
z&9davSX}qV-0$xJx5;i`-NXD{V%*Pqu00#2Oj>M<ZlTh3-R49qm4-}QP43wsM!l69
z>wguft`t{$M#^^?f0oH%#P^z~S1N33&ZGVHd9}2-{$0=<s&7u?<yVG7A9xSFUpc;i
zzE}^)TXvJ)lc8s@<}y8KK&S;BuD)zdw#ec}mj?JV6gL<@4L-IJE4yo^N&Q?%sbeT<
z-<D6lO@#!wW5PsO@E=h1N6{44zmRh?6(#>Ag@^>K<u}xcTiKG}JfRNA;$_h=9bCMS
zDB!pnwBhUMLoW6C!+wL=J-`e3_NnZ`!iD(CDs2r?v!wV5yJ-j~&0;wug(V-qN@uTi
z$n)Ueyu>om7vm$tuC=*_-!+kr+70cm8wgbMR7+3rcE6>;>&@A>tW~tmD8QOo7Bc)D
z{lvECji?Rwv6WQSFBblq+Xi8NKh1k}^;t)`ZX9^fN2p!4n>lH+Y2LYMC8tPDC9o)m
z@ZmQ}sN@h`=^oZ;{JNk?s-KS)#vNhV`f&EpKgN|8V&d%wzdaUgvgPdJBGG_R*U_L3
zQ<WDiWgvj|A4O*w7F8ETVMUPc4hd=LluiNZ?rxBhX6Wwj1_@ERy96D2=s~&}Iwgnr
z?)R4;Ja_J#Ip^%X-?dh2BdZUrpp6tO3eh~eB^1Cidar8Ev{>%1I!l8U;MSEfF6H?>
z1L4}SmDZ@v9irFL(TAG@^9j{LKQf}!&_pH!I11hlXF%wCskY#o8WmsSYe-nzCa?jU
zj=S^d%|+7L>1+{sPU`~_HwSjkihQbss{mb96GEJz{D!K4@qS|mBE_HXdsobHRw^b|
z=`x$gisK)`+_7*Xl0T`3U0-AVzx_7(#UvyEYnHF<pHao`j7n)>0%MkiE;)T*w)6RR
z8Kvr>#d9-BSGk<c1f=$rz2O_jBcZsl8MRCm1ob_#b#!|^;p$s{{WIr=j)sYw8+tFV
z+WL-`v4<sw0xyTf`-iklLC+jD0`yq_5N4hYiVa0bNWkiy22~#Vx5JXRp_d}A$dH<(
zz{aGzpGlRkwjWH(-P}Vw&6cf$SP&_lna^lZ0FS!6;Gs&sQ5m+i$sqwj(Or(dYfnlM
z0lG;f4gT})Y?Z2sjepr)RDu2Qv%V<yf0|Sh;#!}t2e0aaxSoBJDhzA2$XMbe976bO
zuN0W6TF6!&X`Sn-ye(^k7|TCBP|Bh`%7$GlcJV^}(xtTC{=Qrp-yhH;6xBV`cDT@y
z;hCQdpB|44XdJ;v>C7w0WG>8ZbY}cLbc>pJYcH?d4W)~gN~EN$@0#AEdRlX7Y=G|d
zV6P9M`vaS+{+Wlot$U-#udhB};QBy(eCqr@96qj}%R`o!qhfT|#?Du|iNcK3d&X5X
z`FS>dhD7Wo!gQ<Xsp*CML?w%Eg!8R$k0<awx$ce<0s%8dAfn8@T9xi+F6&A{eh66?
z!paJ!vb?8xsFC$7yGkx`(tH}j_xY`*S*2EJCwVJDC{rvTqTd?L5QQDl|G8{}_YOJo
zGb;1!U6H!<JTXk@#8cf7fFRzzSHM}2h?CLlgest;_qoA;a;Oy~!~0U^bY~|Gz860k
zb${3IhRLDv5N;k1LrP@(Q?5#KLPTxgaJtqp76!i+6cX6^_b>gxN-Aad?`V4!5#Xj(
z89#RjUwU)qFT85&Y3&R81w~x(V<$p~Kve$6C>{b-s?K0nNC0y|C!hK8cHSzA6XU?E
z%hAcXsiWZ!7|h^GQRMd_myp_=J+$_L!f3Uv3gaSTwG-hGqHl4Jt9I86Tg0ZCzc4QR
z3oC~7Om>*eTho$z*jLI_d#u(?V68%g9C|SMKeql|S}}Kb_x|HA;PauVlQH2zk61=Q
zops7OmJi8HkD*!%7IY$h&|4?`U027KYSGi#?`XKE2>Rn}G?At?T~z>C%z5rTzAEUm
zIz8#Z7juVQ%MFNOru`9vXoJ7VK~PZ7kQM*#=BAj-(*|;UBu}+E6XEtf7lMGuU)W}t
zUp{4#2;Yw@|CW4JF(>;$4zTH}k5CuB@%seHENVx3X~IIov*;v?-&9@})eqkC%neX>
zmlX|#&xjE%pH-*VmMpdp!~F|OhnT-1DaF)BqPvOtVVPy_zWWirN$3z3hmEi2|9a8t
zw?8L!-h{oQ7mA}~wlWICo69JwjS0DI3rS#CYFf1H9QN6SYM0oGNt6CMLB_s6JzC$L
zF2!%<hplu5X2>Po0I^7IYrbI%y>~MF9~T6Kg?}N8yq7~_=FZ3em^;pZFFW`;&rn#Y
z0w16}Cv?(uy8f&`b(xxTl3cz#X1%b;r<yevcXg-`{PE~=|NRuDG?Yd!5FA(Xz=-KR
z*fz$H?JH+rQ>C=XsG5Gd4qK|B5;~Uoddn+K&7?R1QAv8%FZzE#bkE_Rc-HeQI>T#x
z*c{q#DO7+X^E!!AbYo{H+N&yI8VNv<U4r1AfoEq=M7^e@M7o8{#My#TvHB-i{80T1
ze}V=m_3aA7o_~i~7CPl8QQg?scj-Q(Lz=IPGgo8TEw5MLHMm1&-<Z&>7z#jN)!WCd
z@-wTLAhf%O+oe?kEHz`hXS%|k#>}(A$Z9tff9d5198VwT4NeG{5kv^_z_{S_PwJoQ
z7P{0@f7Z=XN87b^dUPGd4@rm5f8I4V+h18=eA7B4xrl9fqLZ_&9k6+5Yw+}dpO7)=
z>{x)~z;dG@WP_r}EI}N}?Y&rSZIAL8&4gjsChX|!To23&fmX`jl@)9ED?IWZEEDQv
zRsARUzD)<!4hU?)MXLx)w~lQgWA?8|sv-$+`Y^cjLDl>E!8r3cYe9OI(65)46^340
z<iWZQvJM-4FlS58{jSUC{J6h>FSWtEuKl5TR>Gjp83%r(FZ(Pn8xYv%4&z^c`Gwu?
z{es)c`T5(5Cpq6fmDmPAmhSy}g6~{aIAkVkOc}&wW8wYd@IMW#9hxPao<(XL=7M^!
zS+tkEtuXlEGQdvN!gQ6ymQ>9xrrIefO@EVtaKilP71uK6H!jEXa0S#l@=S9dDz<u~
z-PvL?JnyI8N3hHa$tQjKtfqsnrzn=kNmZC^h3^%|{%j%P>z4IvKbSM8nDg)ZZ#4wV
zM|s_TaNL9Q=#S4w$o?-i*n-(M(e;uQi&j0ed+2q&1Eq3F`9rJw@c{GS01R*Xe);p$
z%ky?J>5XgTic0Sz0Wi(Rad`jj;))*ZZYSQWYtV|@+R`#^9o{2u)}RMmDX)SC^px9X
z-<qQi29H=tb$(`Q+zLe++z$OW-(a=)#Y^c!xV64c)}^c0;p5+GmEx-M#-X5_vc7xs
z!KWHazKCQyr|e1M72g|D=k>s%dw(ErQe>#0%NzLP4oH153MJJeqz+*%;PzeM_Qzg@
zT)KVEomHxpy+6bz097TL?zZOC7<-3s<l(7mMxmP{uhWX@Y9iD`crI)zo}cxnriRo!
zlTWl<r0x7?q3K8Rkyy^#^sR^MBLewys>=82uZI4j4qpMgD2_(E0EfD2CoGe^s(9_4
zWkX4X$3n!$O0JB<gk!J7PqALAZBS@qeVM+jNm;W2TM5&^Rd-S_2Q#7i9CacYqGb3y
zsYZ4BX;nmj8YD#f>x}@D-;Cf*DOh!I5-Yg!Y8@?3hL)x7<7*OgW@T$;&35*26PE-5
z=b3}*h9>u%o*tIt6JzrLr!eQbFsf%Cs==-JK_u<z-v;sQKs0K8K7gynXLHz4V;I1h
zRB6h1h@lZ)^j^jSsBy_YhmWnMO+4Qf#;3jm=~6o=kHcGInn{SCo#buY8d(7!I=ykC
zvN(dv)Sd)mekgIXq3T76_G~INw`A_FP~xc>5>vFv33a<Y#(lY3Uf){lK`yg5xDRRt
zh9wTXLHFV>sle2x3+Y8O-lj{kYI$pC=l4il5Lcj$v@iQ(rC7w(zXQ|lpTWjr0%2an
zq#uoLtO<$Og6FM&7~DU1`_*yZ6MXsUSviZ;Z=CMg@rNX#S^FMY?VGB*ciIUZJ?X@7
zq8t-18e?zmwMI+1v2iSduER}m=UJTzxTA=?(F#%>^xMMsGsprr?9Gl3l`15~MBauM
zPkB&*e?j=znGfZP^!Xn~kkBB6)i3MD=S24zOGdpBdUU?h^bvG^j%#ml{?ul|7>f{i
zfL`Z%drEQsd5?5<Gu^8xfsyb#D=8*&bC(01yi}jK#D^kkyQzbJcRlNHX)J720U;r3
z`R{rdg&Y?F*u)rpsnlxpvLg2R4!cG}UyX$b+P|Y*g`>8!2h1xieu=TY=F=ULVaZe#
zi|2wk5@f5`I=VQyLdCaFSF!Iev--w;xCco2vPc}FOOqFYypZXw^L^aykz|2Hz*ba5
z@gCA<3Y#f+8kiz&DReQ#{9X7cn5m@6)?6#IU*i1Mpf}u4Ol!*t*(Ur4#y5wMPQi}{
zo^V_10fNMorInR3rhKk?32IRp5*haOcBUL9IbbA?ix^0#xKMr5B$;_CI><?f5?T&v
zfjIMlh4|}iI(I;fH5zElR@SE!7R{eN#3Xy!%W3Ax)KN0hAWGtF!*L6VQ_JEiS$}eA
zu!XE1iW#$Yu?x44T_R-r8D(D3*|>NH)jlgv$Oe)Pl5{EMef>zJPoJz~T)y@Al%W+a
z`KfC4+oIj<QB6~8D`aWu?TJcMJMJd@T53P4&qcqH^&<DnYxOV6nZzW<W}Y!EWR_@e
z2Q>kPMvv7d-8z!kH{85iq)o|~sIJrd*PnNtPMyQ2hGUn#)d?mStxdJ-Cyc-4vGNV}
zge(gg=|EDMHgs2laf_kvC)9K_in&xrZqR=w&#39F#a)LU!HAb?RwydfOL7jvUHp&8
zkHvj|JMv3^jyQuOhN$MU3JCf94vV^#Cqh+qXZBd=WHL=Ib9ZdGyy+tevX%znM2em0
zpo^>NSu%ZK;j=_5RKt1t2CVnhf17VwJ4Gv;P~OL%m6l16FEqYlNh_!sBc>9k>@(*G
zvSDjH?B-w#SN}X^H`P^=a~nl5i9i0{9W92jARL;3Zv?6Cn&KM>DYmW|$B`fQ{B?YX
z>w6a*Pl}Ov@B?AKK3$<0_w#P<-a;Gy;+jdtyMA;BHcCF}x25TWUGU?lG({9u?i^*%
zbq4XYHN>NF<L)l}<=OTnL8WOTgJppwh<{|mLxMoo(Xpnq=KC<u_cShR^{;MjvFT#o
z;@wd&=0d+y(DP2xOC1H}=@nq59`-+vjG4vm!b6!feDyf+{3xJ?4BJ(wkvTc+V5MKG
z+-rPTsH)DN6%^dtpqQJV^qYZ;JUMAA#7sjr`<S>R2Wcin;jE_cvAU~`UvEtOFJ#vI
zuQ16H-pa{U=pFs78F7>%2=9J=@V*OMha<BRF@rMQVW*;$qD!#7i5^Cea~)$hFHdoS
zfT9}3nt4H1E2OX^Jne~{f<eonjfR0G)P;P`JY{{>Ju8qxDXw^?E&8hGworkn%tnMF
z|KyeE0zKBaEq>(S>o!2_KoZGBAG2MLV2MzJF0vaONC9~2VP<hv$63=xXoQw{ykrh|
z7JyyjwI~ZzqP)7430ew>Ml{o{LvU~qB|WRD7ED7c5VqQ7DEK}T<1Z)`SJ_~XT|NX+
zB#R!bC-n_s!@C~5tCqSM?f(VNsK4<)ycBL}qf$)B{c(zIPJ%ZCd~Ajqoj9*9>8@4!
zM9a{Ax#wu`+CwR<gF#WTkynl!M?u3*IkSBK{h%F8ws<VRDX?hst3+lwig}E7cVoX9
z{Y!^)_|I28-hsUw3(PGlb4wYI2Q#GFEJsAs;+vcj<_eeG6~peajm4~0)fVL_J;Br%
zMV;tN@@yikx$a}%`t)f>G_azEx^LI~;u$sLxex;-a0C)I9r*$OiLUra>CuwM-y5v4
z(gp<#pcJlm%D+0#gUR+^EFS0Mnb7t<$Q-NdzKO`B01_Yt{Y;-i_a~p+3$6as{U`55
zJ;RES@HMj0+t+aoD$VG`<g{X^2!7q<&YltM<qq1nxQQ)IZ%gUkqs92W<G*1~$;ZqV
z6Ozh$o0PPMNFiXPoLv7|-E&OmmXRI_nt7*#pM~1xE1Pa_Ch4p&Ahn5(O+|G?O0-mB
zz_scecHLh}OWy6}88>|#E(eA^M|s~JQay^EuJwgUHg$D{aYYSlN~-TYwu!@C9;e0c
zEujk^1<CW(iJ0Y(qkMYbJM(`|6*R%IXjln+30U*FVuPrt+P4Yu=Dj0xeM2JxtWxXV
zrxTa+^UbH-!{u_sxzPJ0j-omEl4qY_L1r~(DT~h5HW+Uw07Kf_eElUmEjZ4p@|dV>
z<M|2sanB_v<K)Ei{}nu%F=o``n*o?s^7D7b%i`xwM2jeL$pT|vqZ|t4@NYDVHr*FJ
zXRc2V)@4Sk92U&PMmgWBoIsbY!bHWbHb2qbs#uQVcI_S5WKw{Z$S~lJp}@Q)0KV-x
zyh6GQ?1!pIk=U>GUC|V-)1AM`onUY{atJpclsQ$#3=3iDiZCCBK(O-i@_6%<(>1L0
z_w@K_&I9^pe#h)0RGjuJ&!$f$omZPSl%lO8uivUGOgr+beW?Cx*kk2bo~~$fLg5&~
z2hkkAeKT58eHpJ<eX;udfcwg6UhA&3|6=>~N*YO*s!M*>dW~m9^p(RG>kEUYp@VLb
z0ZzKu0tj<#E^nR_SpvT~&bK+nt_g!TaUigV5a81wxN1V8^RAuV`XIP~ou|<ey!q?(
zwmg*srX&u|pXo~q&yKZuR1=Q=SGype8zW1DrixR6ICnGM_GYx2DlCKRi-qQz9ZXo)
z`3NRTr~)w~Vee&12wk#RtGT+=OuQxk0zaztznGhaOV4tic5l!I?CkoZwVuPlEwcaZ
z^+$of;sftd2ckJqVoGpo+#f1{U&E>Jj{IRYjo2?Wg+_M}`uf%IY`n#{Rg2oqZ;H|>
zk0lInU2eqcYhOpFn(31obpm?%#vg7o^a@BQDek=(@ag||{?gmE^2O5zSXA=>3ZWdy
zwnBx@95@q3vrbhPAXxOME4d1~3Qz0pbktL<Eg&|Yi(hQ-MsE`IX4JggTuGs45gs<k
zTyttUu`-rVRy49@xsi7V*`_Usy`Z(_Qr?ku&6f=G__L|y;OE(DBs0fx5DG6Z@6THD
zKfYb)G8Xe&tx$iP{oYo9qQ0eP#9wUpaDI&9-McL~rAEJ_5Nz`Kc=Fz_@5+ytT-iom
zrBUx|I}qb>%0F}4@OjH7>-ySr&1ahpxKi^Mfi`)RD%>xDqCYkcF^+md1G*s0OU5-j
zGSD#^G=N$Q)u+i1Y5HC6CNfg^{j|(brC3|P5t?c&Q0Ya;@FdU%t^90+IzN=g1r(ZF
z9k4>!^7=D7u=!2`1}$}N{{Fuq5Pm(y-11N6pHo?Zf^E5NeLEvdT($G_MR%~@Ev1{U
z8yIrBhc<Zk7iO}E*My*y^%L*f{kUV1kNA}%<{@Mv<mhTVJM_Y8*pa^~-g&a+s}u^r
zNHQm1zQ8$NkT~pS99w;E7HvWI<ym%~LO>dV+5g>?aB38k!DPTN(xRbw_5cTQQ}9iD
z<B!TDV>V47M-I4sW$rOeX4ucNmd`~Vc)Vt6RND4y5Dv89JIX-&Tae}Dl+oCbtqcg)
z+^M^8J2k2sSYX3{+)JMWTgF<S7iFe&#n)>^*11$+--ig74MG-F-)xXNWEwPI`K2#o
z*YX}goGtaZ4VK!m?9{>~B`7{dZ!;vw;)|>KYao13H1GYKJplvWyRQ9@T0zT|UX9);
zsGz<McfrKxnYo6RkR&w=5Ibu~4>TPlS-6+3z0jSGM^A|4`?IZCi%g06!ass)PE#6H
zoAF=>Gj1#ea+S*TPYgQs_l>VfSSzt58B9EDUO4>G$9F`5esd4!XP|HHSkbhPfXG(Y
zSwN9E{46N30YRH|#;?%g?2`WZpXv*XOW#w#Ls*(B#^ILiXP2J;bB@lsJ@f7zLtxA%
z{<}oC*jRc>X+@{9PZq30O9m1yH8NGJU~Gi&^MZ&u>puzARsv+R&OO-iViyi#g!OqR
zRA0sY3+k8q&%$~0(7F}gvKx)a>acLD`k3(zXnE0QtPL}-t{^(=2-P2|h&v#G(W((>
zS%@YLl<XxRc%jgSj~EqCT()u8I2|Da4cJITN9fHscL@HpDN^eDjfX2OBtL&O+2!E!
z6Pi9>>-s!LRGGOuGN%N7rnp07YM%HELp4BEV&;@!`QmQtPrRI-?KY5*5S$jKc^Q~?
z7H}l&f1%r}w8Ik5*L^p`g8_NF!y}VRM!5#+U+BAr`5)+-(z1xeDsfXOlbG3i3hztJ
z+O<}OQP66Sh>*Jb`v&(P;0{iq9u;-Crm3|3+5-~-nniD>nz#iJLXj*c?+OhkwGtBI
zj)n+ZyrldsLRCj&i%TBA%A{VH-0dx^qzdK%WO{ZsU%&QW3?a$u(X@0O+M11<6YRT_
z^#><|3I3-hn$lm<c?&LX9zMevihlk+F!sVT&0GZKU%y7)m^>X<{pcSe$%gE$blfH<
zoJ_?2!5N!8)I+~5gfK2N{2tKzaWm2A2ZjDK;cMWymF{VPC6HiUp^Aua&6GP-*Es2J
z{3b<(*~?~@6UYGbK5`sV>d!=iIvk(V5RCOMtIb3x8~a|8FYF|`=BY8HMnxDEd6evN
zdW||y1_j^HBG$4R*ommpgddeqB*5drw6RiYHl)_}Y9Flpj_{TZyTa8iM^rRIj)hcj
z-fQmN=r(2pryxpe+4Olwb5qN>HsM_Vd1*P+psqt?(sA|M$YtTzLKE|!6OdKi+xlzQ
z%9}2(B0vp*$;ve^S}A1vC9_qmuV%I&Q&#iP;3Lz;Q#5rnYG}Jc22RnV0nUaWr8<cY
zTy8V!w6`+B{XlCV8Ev>Joe7PLLF$7f&K)A*)!|K4jl`Ghc@luX{pZ%Ky*QDvYpg%X
zh-Epz`5pzKTOsJ)Rbs@zkPX8|={rv<opnHuT-SUVuBp{3;;9LqnmozxKkM3Gpi~Yd
zPE2tR6!UsjcX|Db`m9Kq*r#i|r#<j8`N#^XGLMI4m9B}6<8(P}s;^$Yp|#w9g62>x
z|18Ny6%;s~u6L>Cen&wvpRH)?1)Rl^_aBLk8-lcHClyBr?xQ~J0!D`tb#1bBZNQ#L
zIl1zH(bK{c_@!UPiq!V@4ELV_FI4!>H9SeGie&jKaVmi`-;Xj8Y+}=132jIv16G$u
zL0y6_%lBUf&GnFE67(gD7D->4N?crLiCeHIQJC04sig&T!4EyXX2yCvSc9!Z7JV;_
z1`t9bp(JF`=Vf*JA1s*eUbe8sXVF$pfpUmZelk_RRW%EUfuPe7ncku7T55epLrjst
zm~3A|?+_LG=IC^}51}m#*d)76It>j6JR9#fCf>ZCm#9khUJH0rXAe49oYb^JvQOoq
zQe|q4Z_k@PxsKj`JGs&s%a8tGD{a%nWj+P;aq^BGT`qGbyNAQzXAQk;;4A#x6QV0R
z*K$2EgV-9E+|#!Pt8-NNpU97Jj04hY7;VFUj8{}4#twbkM$0t(GF7;jZ}6-uF|VhY
zp3AAF$D1~izWH@vjH7S-W8b}}^Lj5}{c>363VXut@VX|`_2PlN4M;yr*+aL9!<}73
z*<~hLqAs62xfS1;pX~ia>#XBqa8#-jTMv8PI3HNu*EU~g1R5b*pzr8JQq8?Ha+mzQ
z)Tv4#2cX_--??Mh6chMzb*fO`ZvtURfN)DzK1ruarfHEunHwl*1w;j9%~0N8uivU^
zm@6c(S8?b3Jbw4PBc!3_bcJISAuG`_TZy6jYE8KN<=JUeqqI!-HE~P@!qN$Q&#7;Q
z^$~-oWkG{0NhHj;ZeA=+!MuiMk~>raAa|J1PX4L18Xdw{VE1;&$M-)fJp48{yXU@m
zjazusJmH?2TmLzS4#5~ze{h038Pal}s-KQ<J~wFuwv~KQ`0+YI1})+Y8E9QmeZsOD
zQ9X_OPVmV*_*E#U1ar*4?pPX+78i5HvU<Q|L3X9zfUz1&s!!qQhrX<4xm7yLS}2**
zaijb|gSW+E#JkAN#@9o){$;65_k&M2sdVM*R9}JU5D%fjY!)WweBE`+UUr1QYFaH0
z-Y-v{#`k=H-yglbPBM%>f*sXg#ti8OwsgAFoy%vS)c&#|CMP#^+N(t`^;`W@)8a;$
zcaX%(arnJs?2|bWeQ5cszBcH$WruAL<x|sZYt#gt-%GN@uQUy+pLusN`$)tc2|Ah@
zM;kGObGVDx+Wqf8WU(7>E?tu!n{+DF-DdT#sy^nb+Iqhy+<m{h(Sv>9bi9T94yyfT
z8izd+wO)g|`@`1gsI^6M_~?72cFvr_+5IXgDHwPh2j<aGDF@i)8M78`Nmmn6TdjLO
z!%hXh#la?d7%S%I1j10M&)s?{yXYTD;Zp~EcPkEc=ZC~5)L&TQ$G;-j50TjDN1ol^
zWn5U3MEWSCzyxJ7Y0J$l9pda?(i5&Ri9jP?^YIT-tpqkUAOR*m7a)_s>#RF~X5?0c
zjmI%7uPaMAIFMV}k6et9E*l!|$S7o1>~zwNxR<p5+B>C(8e*FGZ-OZKu|}HGRjL&0
zmd#zx@bdD=Dc(is9XWD}#r*ZjQ1&&-+7Uuy8_2fCmw19NEu37?#j5Jon>0x;$&7&@
zhLBrnM`j=@aNE5oD5GutSqzZT(&aSuHpr&3-W}VDkD2<P1W}73LDdKZ!_13nuB+%O
z`+daVNm&1rjo_9{;g>Hsx+9x1e?}K8Yh3`VFAKge1CWy;RTG=cM*edhCT)3w^GPRl
zB$Xg~f`wY+KUYFNqcV&4y+amUz*K~j4D(&s7Ld<{{LIZz%{o@u?2x5y1cNtUG6Eh{
zE4Lw3uG?TXU>JQD*scWTT$(xg*Y0)lHV-&=%rQ_#8;h9~OE`^e@O1sgxpKk(TaY)=
zzBk?4YF|MpW~X9Lx*XcyEVQb+!ev{H?Ku6y40la9RTcVDeB<+gB&8-Sj`lT9)>uDA
z>+N5Kj@mgUXS_&d$MU0Q7N2wZQg-^cG#;Pg`D>fzulh%0CF^(-taD1!sVDK5R<WeM
zurBc}`-p91D;e9xHqt5cWGNaG>}zP+h0UwT*;cW+^@fX0*n|mqEwN3yEA!51D%Q6P
z9R{qO-Gc7UUt?2>0S7N_I6a2H!34tx-^RYNKYrL$i-KeRXAjw0CjOlsUaYMau2&^y
zB4jCLF^2u{*CBN%0TE{}B4G1H^mWqGj;9S#=XFrg_N-<)F|}wCQ@lMt*>CmXqjfkt
z@KwE^Xc!rKBmA>JX+<#LYR~gkVxkHYOR9^dQvu9*b!X!)x$)6ze4PVa!E;H`2;Au?
zxXDC|XIHmc76Nk?djE|b8~1>rd-VJtUkE!Nck1r7Z2O??c)NCxi{&<Z*>?!y44odN
z4JohZM-}wic&kQ@)EZ>x-#PQmpg}@bxwAVz%=#$M;EbsVi$vVem2;pjEIY1shYsUY
zvab!Y^3K?a9CR0|<<20pDWWL$*LQix0I8M49T&%Zp^<d6?%{4l5F#Sg56#R|GBJ#F
zoewL$XKA(&Z{~A<{<hX^EI6X%@aZ1W(*ccP$k1w&WyuW@)k<JLN5C$6_v8H}kD^=R
z%?mce6V0TEdmOR&3!VXrifYv7$<nso5xcLk(<;dB=NmNN=Z;T}zb`m%ucHVElW&4j
zsn1CgB3HVCl9-&6cS&t+0~oePKoQaNm&N3GI?m~^YWbeJqY;h5(k$z=R|1<iLox#T
zM+&*bJ~31dp?r#>d0e4@>G-m6UOXIK0uPY}kO=NR_W-2tIpw72t%!^Nt#WHCe~0fb
z8C(weF7OqYd;fNEu^Ua#<6u-38IEk!=hlNzdm8ei{k;Rlk3WTZyt;E0%SeQ6x~ft|
zU*QxuwPeJN1oYbU+F!ls8b9Qq8r?76686G69EQ^YHV<&uke>}_a@=-P679E6%e3xA
zBy8DpCEU@YIR7Zyg({4(sJFf|n&EJ4+$)9i35BLWOw*9eHI7BM4p4&f<xa3mN#uSf
zh_F0{XtGk+xtGfiiJXde4_UcIN@x%u6?)`CuNt4_7CYl67+^#~rxbb(=Ug^9W4`8a
zO2d4Avbgx%MLZ35cbzQR5x|rnA`WNlGM{%skB)iv9>MxeK5^Mq3JV%jT%~Z|5|tLU
z3KV11vLf|TJaS%R-m31W!kqQV&o{DWEiy-;BDyNrG4vdYSyQR)mU{LmnL-2eNcriI
zVT1d%MqIg9YeSrgP=i!fSalFs|0RCYY}i9+bN{^FBHz$xX*F<5%%n{4N@moG_dN^u
zWp(i|dZUQ6!$ebCVtb?a^W(>L|FGNjmzT~zp7V>wIt2qVIvdEZfgWB@&sg&k*Mw}e
z=JC^GVrXaz<z+?Cv+chqX<b3d=`XEztp)&Xwc6$F^Yuln=?}q^v`qk^fEkSW4+}@%
zXMzZdcZIGm5Smw6H_*k7W?7tvl|VS9>`Ms-Tvzb%Rb*)|?HBG}RF9i=?pJ$kKoe!8
z#EiS3p+QHC@XeD}e{J(dM&w(uUDu)}Z*d9G>m!}nl|lrw8ZcS_mo4cKl#Q5f4DK=!
zXhaYY)|CS-M}ke*OhMHwn9^nVEpz>r)okdRkixvr@V;ar+Hs=TGl{m%3ME9t#IGpo
z0#CJ<dKhxks}VWU$7cn<&)wbK6<(zPop1mx5O5R1)fB2JCUCyI2q;<uGdyieI~(;V
zm?$(NOGoa8Za$026?j6h>Y6}KVZgsb_*mF8YIx=C^UTwBo-qkz8|)mXm!{G7K9+Md
zwsws8OhvtCECA93{tn@1LT=bkY{K5NOo71lK?8j+ZKjq};+B5401yFKD2dYQ(?Ozt
z71ljZh78NO4sWL&rxH%HqW}tV9oE+>M{kUnYU)Be4_9^iSMUAAC+eZ~tx!vsGwwph
z#l^+)cNOUoHe9Oh*T-yF*p!@kDkum=Iu$ahW!;yfRTxpy-Tn6-88Rw7i2}3I3s2QZ
zK%-F4Ff^HjhNV_T=xVWZ++fyyPI$JU<l6m^2O7AN7~rpkDYHb7-Cdl>k~)81TV}RO
zQr4JVB9TVZ@R<hqc|9L6mr7FU%ag>GNT7UlbuiA^&c%yk0YO3aEy2C4Xy9!wupJa?
z6#9vP*G|@nf1%CpHzu)D#XB|B(rA+|>9*PORTF7yy*u>S5^9uCYfUY6o(L#+(PTY?
zoPzv*luzT$V_j(#WD$~Ndn9IX3U9Byf2cA_@vY{M^Ut;PywH3y13emtXQW}$$%B1F
z{_KRxZj;2x{Y2V!t(f_tpZ-J0aX0qt``2<l=E=qL-N`>+?0SBX<fk6^jN8ot{*cj6
zPsnBW2%H&6i7ao>nI-i`8}+<C95t3zi=4!ZxrjCN_PI@(Eh;rE3H+mDnA0<CfT>tf
z4RLMc=79gtQw}aDI4N&!TPtePADzBHjf#>JX%8y8w@@kE-I0+r+51dm$H#%tHCXy4
z@cdM`cBQv52!kEIOq!<&;)YtJ7FC#2!L!U8vaj}cf(OH-8>?RtNco~5n&WU5#NVAm
z-lcjZswgU5gu9zcG;I^bDln6|j-Q^g{}0-2r#0<@0uC<O12>vp>|ThAPi;AKm9c4%
zWHwK8?6-*G;$&i>cur<6ZN;g}fh89n@dXV`EGU@a<@AC=TT@wo^`^T{k5e-G`Pi94
zphR)M$Px(A=#4Z((g~^SJ+aEV8bF05OwC<t8D-7jMYnXA-MpE6TxkC-$`<7j%8b6O
zm~!^|`p|+(lm#<$h@xMX&q<4Jhp#8LlVYRA+tFg41PBE_Nfd5nIQJMXSZA>;mDQ6~
zaRZK+@K!2|03@mYWvPcxwt^98$=5o&q!CgM{t^XL|C=t(?hhNdK#G~Y(B}MPkCadz
z?hxhnMRZ!JrOhmAFV%U1i+@-h5MDE5xNasyRuY%$!4EtCB^Hr!Y|<fBH($n^`Mii)
zdJ&8*_85HDb79;(&$`DGs@5^%Y0o^?DNK6!UD}q$GHbm~mam2k%ucGg$zV$>`8UpS
zgxmH~lSJc3<?R4#(u??0MbH3Xb+-O|)22SN?IZSm!Pr)$@aqYf{fJ!jl8$CllaZ1V
z<L(h=YU$^YHL`<Y3SzR4Avd?)MW7ZC`a4ucyQP!1qEL-7rKhwyr<o;_Aerr6kh;6I
zW8*-iYGlP-nBLz&_N0Aut#kF!^V*+v2NcRaKV$lp_t@R%JbBSWi2zBZ&c;~Ivao{g
zCyV??V2)FOLEPW05v$>#qN<vs#K3fo(PM&bY?lg>$}MXqW3z2mj7c@~9I+59AI2Z$
zmQ|23p#uaq-+2~_wd7I&G{qvNUT-`{8H~bvV)NOd3D&xtu6-49kMgSacKU<ak1n$#
zcSl0Cnn2UDS#8q(!!xl2^Qr<+3LyatCQa>OaSQ#V+g_7kSd&<qt5l(VWEj`e?CTSi
zx+N9cww-ECb}tUapC1@QuKdiOx;iJSGv)TVEk%xbND5uOe1A`0ytd8}+}KAg;iN{P
z_3mM58&)sWHMT`?GR9xx=J~tDo9lJ`xJwcLgp+JlqsYWOitv-pl@@$O2rA?%P$t2E
zkHg$^6SC*qM_5^Ml=kuqu0hQ~!xmztOlY>zj+p8=$+pC$_34-AVXst??s{RlLS>&m
zV5=%@$%X-guCDoCw_&O$B6s?jVdkCL7aN<ex2G{^q(yc+o{9+&1}(U@_rFrrxFa7@
z9z|v-Y#_zf<0iw95@g4rV1`#=o6!eWi8!G;U6Ng^H1c?fNh(2a4dHFxK812!vu5r*
zB~y`B>cw&KGwsPC8V-)RDP<}O7n*Opu%bio5U7Ut+*M8kKW`2~-4j>wmNv*0sp$;R
zemD>0&VP!{XoMm?K|9ns*pU^CDoWV+c4JBFw9rT+eaI=vCo^~C6*{st^9ym%lwGPO
zfa;_WCgz?dTlP7QS)D?_5W5N&it~ZWQgUf@a76D?WQa_f1aZH_WAsG9uVQQ<goZ|4
zQo_YOZR7w~cv*O*yIV$o;FtH>?xvp$9ym6H@9`Ja6Roq++IGQ0pIoX^c47tVcUlqi
z7-`+-yHoGJF_*{Hy5jJG%uLptv~fHvO0iF$M{Xckk{dj#dq!QIQD}50UU79n+vMHX
zPao~TxP_Mh#6rg}mv#Eq*ie;i;h6>hpwcAq@UPG4{Z<CyteZ*V({y<)+1?x!b*!lF
zRUu=c2DpX+H`Dax+H}{w2+oR4B}ScSnJOcjYISVv%#){#8X7#7#c{E*_HHkZbk#X(
zFwD|mg6N}**dQ_$>v;h3aOcVQS&)4gGvDAW4ffh21Bd?}64?nKnl+PlG!Y&xl7@Wq
zbTmC@m%z7zqs}?BJU4an&XwogR-t5_<thP*rcSEkOLJxUpFF7wj6NP7{RaoA5Kob>
z*1E;4eD=AP@zw!Or4}_m)?oP^l}@^sBDM|LJVnAd`%31D4BNmJ3BR09W|QP24b2#A
zoMG^x80h;A00QJ;Zf|$jx;8VRJi+CR`*bTVz&ImJ*VGrcW4gv8UaQ!LZ|`^{h8|YX
z07H*9IP!XiCyBwf)|XWy8r&U?{_vVi{gsdRDp*1ovLTMJ5z2D~jA?<Swst=l&4i7^
zV))FLmQ}7Ho!1PF2+mraeBSKy4`V_alWbl(M5yyvY?nYcsNYV+O5k|*tSFQ%1y4kE
z<35%;_FUAqtFfnR>;X(vW0NQFg;jK|*$aTp@apCWXH8eJbsQ~!54|eOZBd#@*KkNb
zo3{HHoSl|N3!cT@LIK8%J~uhG=K(K|WVmSI$OCdeqd5`8J>uPQ?mUE@gbE-Lw_Tn`
zAhMcuhG;1Z4hayNzBToNZY2TU`mH^$kWb-@J9u&as&M%%hfR<i8+41@Ccl>}J17UD
z>Ake$e;?})vGS36v7@r)lUh#F%fC;N^IIiNu8@(pbxMF01{DJK(Ol%^=^cNr2qwjX
z-p~hp-+s*~SK7CA^1Lj%bn>>Y`;x-J0TY}iI5<r(nOA5?e@`Qy%0q!QGLx=zmDYDN
z>-h(*0@O4>3b2kGXjtOy69-SVek$r44hb&baAxQUwH(9LM|bTw&K^r<^8K3$6pg>*
zWFSTG?^pj&=mhTyC{W?wAsXiWE9{UJBAYVB&~x~BcU{*FP9UyJ+jIBd?L3UFjCs%D
zmYcv2ykI)E;U@o|lEW=tI(SPsZ)uuF!bQd|M3ZbNcvLm^<fRBAn|(zi8IdvGR_JZJ
zdUlP~b^E7<x%3V`W3SPyb^Y@0yLnf;(*Aq1^bO@&&(PW$U#-!aRSJVu`S*Fi%P(@F
z@Ahrq_56Ct!l~<77&b8Wjk^y8=w^vbdZVRfTC_K03CXa+fK1yWzYy||E*CE(m`s;Q
zCtgu$X`1ml$%|XvqsBF(<*6(`k%@0TPpE#drx#Rm+%(AFp*pU9_+j|eDI|t=oL@Vt
zi59+Do+UPT_VRq@_VR3e3Oo&<pBelGv_SJ;U-@v;V@NxTxrdd8AzSH7+4c<ih|r9=
zJFodfZf{4Cq!pL--xBy9ESFE|kKWzf3aUgNi^1_~PO8cZznvqXz1C5fi#X}NTg;l*
zz27<UH@f<h$|hnvlj&~QBYq4z(A62Y22h)Pig(-v0#4<#T7t<uMQjwosv2Hi><#Kx
zwOS^xJFebFxqJO!RRprU>Kruy^uv{4LmTU=7dCy+`i|#BvC%)0l6V}@IRd>=LzkXE
zz<AAQs^COLXLsa_F>sY=aSl&Tobb$Jz)=;@)tQXAsyl6%o2{`sq7;9d^rn3%k1cR!
zQ%HUy7TYv{z<9@8TUBP`H+<y1Etjf3D_$0Yr7aMq=pzEwEM&=8p0^2}mkImE<UzPv
ze$icQVY=Ed*ZGNDS)IsAe*lO?<We=3^JELFPr@(367VhD*#k^3{4bi0kBy|N1vl&x
z^!QhO3X@t~Td-Nao5mcPd}BYf(Tzk>aNyDtS!bLI%wYN%GZ0+Dl%p~v;{Pvr?SaUl
zfZugzKQiH(grofz^pxJj_kO-y1{Jyi%G9a&V{m&!ymkhiAYbXeD9=!(J$&Ja;!1`Q
zO{V=QvK)YvkWPX{g#N++&QO+_D*O|<w0dT^dH?jU*S14Z<BV@?(lD;FyCFxu|5|;I
z0q2z);j{XV2FZ%qZ&UOxfiI#SHWH!h=b-zH$qVSRi#tEbqa**F%Yho&z;NmE(bPe+
zTHt3Kemh<$aAgvogo(c(@$<I<6KlZ#eo?b&N)_qYD&zIfhMmrSl#u#dD*-DX0c7eT
z4kjD-YO9gJxsM0MGbpUOD=gNDab|<;q+P!Ab}helkd_b%t9KO9(xV)u;Fnkyf%q-^
zR|C74fSa5Z{~0Y^x_*pCC=(V_3&X#FSXRa~(r;9_zhB!6{n4V<bf}O=U8W2<1MCRS
z%ioU>q0rd#=0S876buT8<mBK?xOiY=llI;T-K%`@AXu;kFU;{J-G)|bGsTMMbTyzz
z#Wtbl#Hq7@UEUcewYsMsC+mXJ>K-4~H_=TW@z&v|-b_tgO$Xs!-rO^+yQA0SD9q8@
z+0#D1=c^X(;6Zm$k3}!AzSj-_XDavV={dRNhwxgZhJ`gn5883H=wxV(eyY1s+i1Lq
z;CO(i2$q+ae1{Io0!foMD5@^-`eVD_8P?Fykk1DRhe7PXbN05{zVXs1<oTVGW{U|O
ze1{tMrMAoitI>p`k3g@(Oh}j(!++uYwHa&Y+z*Y}{KO3jcg~#Yzei18+elk2%AwSB
zEQ-3Pcy1$(bAD3fPq^iN6J`EZl5Vc(x(us~oGL8VPN;Sc!!v-pnk7J!)PpxbV%90b
zT>k2uq?05`jKNmCWc7d77|_V8_mE&96VHjx>R&)&?&c<qe&)2eprrr_C&fR|zjGQe
z$<2+|{R&J>t_gSgi!6$IDlz@6ebKM;>_x|5=-YK&ARfBHkykYJcS(X=;U3o9;;6#V
zTyV>ht?FM<;+9<CxEz{gDxBkj`|yA(<^^7}S|hVeHTY7sF{d6<{=k%V@HcH}?FC<P
zboWo>O->*DLwPe)toXAMF=_s!b`q(rMn6-xqnd2J%T8Y-lODKVs@f=lrEhMjI4KgD
z&=zpEKTv<Sc=gkSqbe6ip-fHt6cB?f9c0va8h@9iJ+btnliFgg`v7PDQn9J+gRY$#
z)1X&+@dn${KfHc_24@Ht$_@lD0KpIS>x@|bTTFVCV>ZXQ4|uDnI*CAtS>f5HAF+c9
zq#mGpp3%)wLz7JPo*ywrp5WN+?zimQBjHR2hK(J)73TqefUq(VN{w_uZH<+C1+8B^
zZH3=f?K{8?D{dg~EE}G^f5*S?XZ80zmXD3QJ`sE*!H84>&lYX+znqX4rut}T*qkn9
zX`j}ek=B~_flqJ93Y^hv6om?i3U57RB~_7+E4lN)K(l$qPRerHov}ojk(S7oao;8!
zl{Ox>6%p?FF>I-RHYDcIlj=&LQOc~6pF6_6qVanf6Hhg-kI6=-X$=eWTL}V|+-0ZQ
zzzh1qMLSEUST;19k$9ad<<jZ0X6th16eUK@sZ<q3nskSg$v6r_&p1@e&qU~3?UI>g
zR{9DoiOlil{CTA<3>vA<E(KO1mJR{<mV~n&1015>88g0*u(KeEQv4>p>I=q~+aO5a
z^ZE<@h*iIjKUgHk_2dxKV8SzU7U?}+emwkfq3(&f?_$?<Zy|73@o~4U7fX-GejrTv
z7Pyrf^|*#b<RsIf`>S3G_rbgJF(G$LFLzYG6QWyM_(n_2cngI%-fw^6Da2#i<rR4w
z$HVN!n$jSEZ<r5O%7h|C>q_q23^*D4BQ|%$Iu(p={3Yvn-(B0Kq?^ZfxGqnFico}o
zCSc>npErHve7;BSa`AY@p*^l9+eD~AAlS=c*Jg-<SfhjPV7zd;^dOXHn3wJ=iSsdg
z0^jx|_{ig5I30@u*^<T38-_Hl*(PbIbFakNjP`;xD}6*X&VV#&LV9xPw1y=;6Tzr}
zsW9!4#(d@C7i+s?m95?Pk-nIgY!-xbHCD_HrOaq!Rzd)18+_py50l49&JtX*yHvuD
z#Dqc-6pXnku5%!=)U5iMX}d^qbU-8L!zRKEucZJ$TMJF7q!3!^Lk`U1)4kO-Gzd5@
zMp^3cWN-^%HR&*<34yJd)5JU@|DZ*QGEaE|BrA|CX=m-?gTKL&CKO&~k#UT7tY56c
ztN+@2k7B2R6?@70kFU02+vtjD%EgQ8&`|L2O0(0}NwfudN28qu;4LJ`X6ZDILBag$
zBpd;+bjh7Q1q5m(pksyG5?IIZNA_Q#hAPl96tgN2jl>^#eCi95s+hIl+)uM61*gyR
z9pQ16c}RTi%?8SZ><Js*Ehq>L2kVZo9G!Tf9IOKssOoacn6FdWPQtL9B-K`xSD<HB
zv}HQEFLZF^B27C8pZ+S*Hv4`4%a(3h+L=7CFfL{%lcYOPzi4dRCICIBJfGY?Lg09f
z4hxfpovN|Lmk}^cw$LjBvX&<MmJK5i{_C0+BHh|!nsl?5X)$$@#$M>4!Nl_-bH>*+
zuZlnOEV@SW0fnjBowDm$?4yUX`ixH&{^Rum`V)X*=X4^)_ls9~G!fJ&1g->A(X?a<
zDA%)yUK_c@&GjUk79CplsBCu`-^_&stS|Q|pUV~NNzJ}?q!tp|y&3J8?Wl4Y@ev`4
zqC+&0kN;W_rtoS&O~V4|^N@>KxTW>`f|(Q~^@WPr73xNM>koiHg0w~A!%X!nX4Fz5
zK>nMNIsqF;S9Yk0iw`J6Ze_wXuxvJDgBQwGE&d-btH@T)d_n(dW^f0|e9poFh6cz7
znJ{R^)IddbBHl3`LVMS8F0_7PlK~fm&uik3Lyz#6$-c<<^k2XfVoQuTw$*1Q-#MAl
z9jKY_If8DA?j|^%KJ>wXte}P7XYoaG0khU@_kVUpjwxSH_AtvT1cbSHbGb~P3f5s|
z;yV)`O)8e{XY?It>Bbq+mF!UmUwz9bwRb94&y3{6n39pik*Th^w&U|7klB0{6)YJv
zJaT}e**Svo#p`Uvhp1o2TJBrE+RB$Wo+Qs)n%l=tMvsi%DQYC!C(X5|5PB4#QegK6
zV(~g`O|5|TP1J3Y9iQcrF>I^OQ@(D&noAYcpNbg?RERH2b3~qpA#2M^_Ra<<?`eXn
zap(rz`hr)h&6bHig_!die7)Pl4eRQvHtmvu?|FVImDJZ0LATbYmO7|rp!+-XJ^=|y
zwliWVV*IZGxi)db2U;G(oV8fA;TW<FHW1A%yo8kmV{$(UGS%9D`#x2Keqyf-lvt5_
zd}>s}<n|CIE!GaF)FE(mJWiz}EuD@oVWPOh#bcq1XeC%m-lNaV_hn#G7~1!pc~WsW
zs$~g_vu@l_EYc~Jsq5M!8T<0uCf?ORi=g;iG)znEZ3ynxN+0Hea<>9xA^6uWxhbLt
zgyJtV+Yes{-KXOp&rhxxs};YN7IV3bUdMVee&-@fV1ZjlfDv3vl}S8N!WjIrY8y2D
zUM*dT*htQ2mwnQ9tDC`7P*%5sKt@RXIld+s7B&rxT!kJuZgiTdChpgGlO{zHbl^`b
z)R`oQ^ylP5Z15ou75VvvYJ2pqD@I_*u8d0VUV3@jbogc^{cu`vgYon2drMm;PsyKg
z*6&txXXvT4C_TV>-)B*6Z5cf}jSWt)dLWw7;e~Emv*5|EBJyC@z#;f2s)<mr+wG|+
z7IsoJxw0<zEhb7FBfU%xPYy=T1S1KC35s>i&l|#zMO~0xmMjWGA+q`aP=QL@pW-0!
z#YGBStFX8Wy8UoXK5m0*AWWpHAlXLFLq<Qw&}|xD1Q~W8^vIk!14>u`bY1$?n*^mj
zUAYg|H3h|(xV)RZjov!1FpO>azDN5zg1l8vWr_17N#OCE-N6I^;nWLWIq^jhA`Y-0
zJk6_j6>*G(68>qh3jQ=fCdlZst<iv@B&oIMvj{vWE-1D|Fb8rW62tg@iY%^t82z!|
z`zMLQ%opAqjL5R3fbachosI^(+?Fj|*#q`zLK8%<wx^?2DwL~MEqtIAz2&V!@8Y?}
zD|ieNqCHZ~_cR0HaYfeuuFo!8Tms^5e6u*73ru0l<X!)#Z;dmA%W_c^a28Jd^}8YM
zmf)hY9iL4Nx}|!it$Vsmu6u^MT=guN!u-~Q(heL)OQYZv?~2^E_b|OC*oTs^7XL2p
z0u!8d*!~7?sbmqg8lXy>I;mCPqpzs2y=&K~eyybcK77F77og27Fv&_t!1U*MKj}_3
zfwF!pc$!}<^5<YASj5|a<SKi)^2c9^t{J*u9F;oVr)oFTue)p7b=)jCWJGb{fE2pc
zs6#3<)6x0(5;#-c>w`X(PXFEyy>8)CFxEIs=fvPz%5V)>AJU7{qC7exzq<|CY6oOp
za(a^LL@4I^`}wXov&6Zr^**iPGh4ev1r%=lg|3}JMJL3)61Rsu<I@DaNf~H;$v#W&
z1-2SGc!+JvRqYVvE!GxyVmfcdp3)<{A+3IAnIXQs2_WmjTCAwlKZc9`>OOw<7xU~v
zsDo}1BZg&J`agJpSoamBOZUawp3}GTslYJ>U>IA_8bc4EcUuBpF}NDVd`W|cySB9_
z91wj@Z>KI5>!N<_%$#yh!~%_454Deib5dS$PtmZI4;z)rksQy*7L*r<XOy?9{{BXH
zs}-+~<7-HDZw;>b*__}<GmVIw_n$B$KH=VRxzYl48?#8b7O&wZQvW6rq_ozCBO#*N
zPMez8Ulj+5mCphk`CB=H*q6jSyIgZiz+@Do3}Kf!-9yP=p9t`luTHj#V3=HOv2abm
zD&0)J7x`sS@#h}`wylXV%x`;qBW~cnL|f@#yA=AQZ<&e72P#Y>rtf6MLGV}Q0$$5W
z(Ue^q3OQ~MWTNQq?opYNXl!s=T6QGdC~KJd_ueT+x&|KjRL{m*(%eX^n7Dm3wgSPI
zT=ahz`{ZP6X0wj_am-Z5UuGia>ZBy%x5xkI@JJDS6c8k}h$vY+X@EMoTYINH4ZZ<y
zqc$yW(LVH;#;dwm>4Se=;3nv#{g?)hardVO&*v?U=TN#y(8Vr*NZu3cS!u!y?+La6
z6uX6ZOa-iC)PuPBd-c!z-QzVehA)k)$1^C#A}J}Usi`XpbC!*H4>rA4C)F$wLQ#Rl
zxHUD35}Z2J@e*GY(<khsQ;YSc@aVsBFxisMWZosbH9w}gqOkwQq(~HpWW8zhuavqB
zgSHVQ^^N<t2u*asM9e>bG^c4I4|QN>T;LToU5^dcun|l-^Ptu*#Pt=cQHi2cykEOr
z+na|K(f<?OTg>mfnrWhlU*Fgyue(~ZHQgLNwFYKKQDyQ{cGQhTY2!Zsp8Z5_7j;Q0
z1mDUhZv3C>$vF5L(6Bs7^Lw}6&x*r&fZ@Oh5%XF1J%8ye&14^IeR=VG%>7{BE&Q=6
z=(^8bv_I|m*x4QAk|c+H3wl|lQcI`TU^b}XpU{a@X;UPTK{S6QrQM3a&R{1($yKEK
zGXgH9CqiE}Q?2+ed43Re7gIsdMmkF+Xa_>v@QDw0L#SmutgQ~w1K<gf+*!vDfmc<1
ziR_3W>;38I{-!|hW%WKIOIxnULd;Yx)0!}5XFA%V+BN>vzqemAw$Xw*t#M<z*=zxP
z>|tTh8^uAx>Lx{d5j?UP{dAGtn2od2F8OVVgbhVm;s+sprW9Jd$O0s$Q{#>#`X@1C
zl^5xl`)YtDjr7OZMl;TR9tsSu?{sR@8dTo?{z|DkMvJpn_M>(pTaW_3mEI?@*#nB|
z0lT<Mg1&!>rW<<`d1F35Iu|#iOcp%veIoeU2xr@mRcc1VsF0hL<~_~@N^jbUYN4;i
zoV`xgo{ernYm0u)V^=AYPk-Ta#?3`woO~)4Wazs%+v+qopHW(jH?bl#6FkV(niKLX
zj2ISD;$@3p1pTnq=_3)<Ghha!+F1sgne%=qF=zLRUQ;@4I!=7$|50>SQEhcw6#h$b
zcXxL$F2xBhEydm4rC4zbPH}>}6?b>H0>$0k9qzdgyk?9foSn7Tob#KLdijwv^ad0K
zg68%d3T8HW<apSv)J#a%wUR^+M`LR0yDmz~O8mknRD}ie)-&g>YwzEkO9zCfk{jfM
z_K{wQfq4i*%Q}-cUM>Rz&(uv9^7UuJk?|nt16mYO%fP^-a_vd>@4wpdiP}6*CqK=n
zR&B4>69JJW3mvI(Rv5{Q1aGu;Ahw`(j2RK8f*I`O&J&F;A5rpO$HB{EVF+yfij8iY
zUzU7Iwd2D&9CcJMJ^Yg9$*ia#jn;TPW*`rMKTT_}kL6Yb%@R{cB+;oblE>)S+S$#W
z@Y`)5P5FlklC5ve<bMW!n*9U>PaTe?m)SG+g_=LY-whS0?9q9(nfDots0Ix!8;x;~
zz}Lk{r#WRI`uJ64$9{)80i;v>FP9X4bz#URrp3`4kz(3lc6z5Pm`qTa{vZXyr&+>+
z%GU}$I9JOkbl5^9LM`b&+Hl)XNrp(<(|!067`S>zzSvM3H{pH51LRf`>BT@MNJ%{2
zj9ndiR08%B<Xbp7Ig1{hZn7EocB*^+z@zQg`Z|g7CDZuk`RZG}L@6q}tq_s0p)jGF
zaTiaip5OY38}GebuMcSFM#Z1q7n<PnQE#50?7l!QsK$shXjlDq?C$MXyB+E+>lalp
z{}0W<``9+PB<i~Q{@%0k5$5x-vauD-yFt!`jCL+$C<G9hTXb|g{6Z$q3H>kIy2dgb
zx*~9(_MWJ<#Pa92XZJcH7ajn~T)GYfNvUwKH>}(PAdk#WgnfAmHI3(Tu36@T-#&12
zlE{(u&P0(V^5}Kl?{Un;AisC5&S#k!aTh92#|jMy{>f1t883+3^vbv2ZZqk2y)&B`
z;Um-5>zrg}L8i}ntawbo6P7$H_>~0a0L+d-E4l5Fl7Yk}Op+mc#8FAhl2<GPst{T%
zAEdF*JWH&Fi*K+=B&AND#Dkx>V;0(j);XcSQk+{Aj{pOONW)N*^!@WOCu#LFsZJ38
z<d`~MDqZqZeiczW>yE&GHK|kl!oo(KskuTS9O)yjKv2|}lN2i$#YdD{QEwUWuoTjd
zGJff`G6rs}SxV*g;oE_*6K7X+dVkMhP0IgOA)z1&!A_qJs2#$ne7A_Vj`kJbT`Uu+
zde5Ml<|^t{JOI(-U9c%HObM?m<ch+$%xlLs)_Si_m46Us`bK=4^qA39Okhq`#4Wgy
zNWw#Z#)xhPwJp(t#RI2*Ywk#&x`I1ne`Hk;@<jw6c)>3NOn`;cw1r(MX$YOESKvaS
z;6w5eJuXF|iE4--k5zKKIf0j|VYy8svs`4a+c5+J0W67w+M{`{&z19UHYalr?Z&kN
zV2&yxK9Fo;FtdD?(w||5sCMJ$Rf`)TP;xPt&ZL*GNR)6IObNBIH@py>W1~^p2p9J+
z_E$b#KwFo-gwR%7ELlu78<UY52sAma<mKp@vSci+n&gRJ+=dBgJA9}$%b&RSK<(W_
zOf34FfO&Cpef_((9d8r0l{h{uHxUP%mK>yz9_CDAd~hGQdMhNG7icM=x5CtLiQ@vn
z86kridk;X!H^^&Rknkc6>M<-Gco}#c32RWn4c&4SPuSxoL|6kl1SOJLHOGqqjKLz2
zVfOj@{cX3sRSrXn`1U+wcq}zDOky3z(8UIYCHK+VZCrdTypYFS%6E{=-*gf^4z}|h
zdQ~llzYsY!<7jCFl?C+Wgg?WUe{yIreazs0clXhaVWu!y%GK6Vqa$in2Jc9NuZ2dL
zEg(u)PJ0DHQ!^lGRy<F=MQzsK;!mWFqx8(zferex??<IC6>7DqQwV{EM_2Pl?n@oD
zQUW@0+G8V*-LI#q2M&64Xv}-N!O==l4N4hfi}uE~!E<mPmi-uj+if&K=E6TbT4z`~
z%Dxy>KGUxr`SI}aUj6iHAv*Qr{rtzjwe>Bw_iY;>l<w%`*q<Y6_|fC5|9&d3+x0Vg
z;!LqNk9;vSsE?{Bk)fl(t(QG-?fi1Iyt>w753z41xbFQ|(}rIdyQS(rgSJg=y&+FF
zXI4!4K8)VV(^jf>iW%V$O6!`^p1?(_aHXEmcP&1R!B3(?A|%Ab(-&wXE@yH<P}-HW
zY(gJUUDr%qgX?=X_$&mhgY-x7C(s)|Z0#$Ioc8jisU$UJ@o3)WD98`%1rOB{Enwul
zQfCFm2&rb4wWOfsX75Y{v;o&H9ksMJm|<r5NCC`j1&R>mqTcWiI(I=oUo|B}tkpDf
zjDkXQGvagwN0pdo-`u0aePb;4(b40&&buGi`;mNQI&OLTPvD4YKZ+mdR2?uz{l6}F
z<lg!VSRBOD5=j3@=k~7J=NABYVjdfAB!7J2{`+zJqFkY|z^tv6JT~>&jbTAAQQL-{
zobkRw$yXDq@%L`<cl74vt=7?h^U{5HE<dtwoM;x8FRnqbgV`F#!J2EK)|${7%<3U}
zlBwvjd)(P`N2zdV_xzr&l4<-z@#f_c#q+e0#HlviCcq@)R0qO`euNx!>%tswvjZ&>
zcoaPFeek8!p!8Zz1x@B6)1$oF_L<C;4Zh?7LitiXDPztI&WtG#mj(VsZE%B$f3bRK
zK84ukzt!18#zJEpl@+9y?JH|rBVD0&TOJcmQefiERQkhAKUM*CciwvG3ra8_FTxU{
z;pgWsSUj2i4aEQjC;jOxX@$vl4J+oO*xA*;-3epWpRgKQiKDYyMm3JJf$IK_%3kzm
z?3SyfhGoI5{S3}8GS7M)juG2{VrU?LFM(zUt@_v8+au+K*J1PJQ{PPL6D36?Rp2z;
zlzJp*-viCcN=8NH!{xVbOHLyZ<)ivGWTXj4N9lxu{)ssEwvqjPSe+G})6;WG=>Io;
z1qJ0ROI%$>)o|A1_8TQCwMmrEQzDDs-?RST=bX+G2l3n%z6y&jwt=A6?&CWeKWjrb
zT^HdBAu+O~oo74Fwuh0)xcY?L>V@GlCH`t-Xltfsm{+q=uOnCTtUxc4Bv7QMU`EpJ
zq+!Mj=A^n6_F;&Jm)i?0M><i1NpkE|l+wzpm%$F5<Sba~Hw}?*`bvJCP>|U?3nI>o
zK6#y2GqwfAA3S#E0T=PZ$MCf*z2z)ah#Jr)neYU)ovJe>GGyAgY3Semu;Jb98h^tC
z;6Iz|dxD<+E9L5?6vN@~dfj8YvGfjtYucc?mjG-r=+ek_Nm7u6Nw`ePf0`TJ4^#jH
zw)&>$Y~}U-;uRGDxPPPq8{l^;GQY!CGcsY%{`YqYh}RZw6QpZkd6`fQ|7LV7d+%+f
z$?N20!@|NM<ztNO1A(pM6%jsS;g7635GWLWf1P^IT<p4Gan*`DNS<+&PCiH~6q9Q(
zHA=cBKEdNw8I=K%F$f2C4{4_uSw|0qDxqJztUFVJ=D^Hj_|fKnv*M)m+0({T!NvH#
zK={Z=h!l{d7@`t7nA&a^DC-$Ai~4U6g5WpMJZ}&Go4SGT{Y5!NfGVxq;hYsDGh|3(
z8N`6JU&o=ULcZ^kt||-vsG-K_@qV<LBLy;dfy$~S4G!=5wVl<(aT7c!&&KhGCizz=
zBF6G%v(`dwu+=zB^>pitH$%R@(YfjOA|7eB1~+km_;(*3;F_D)P&8xdcr;71u&}@<
zgwIYIouU^ZE+D;3LMqlJx+%Evg!*u?VIh#^?CiW<t0NrJkgWCt?`YIhO+p;yn+<uR
zc2P*w*GeUh=1tM2-g6cQl-L@dG5=~;JCXO-`2z@1S{?I}ev*?<@5{efj${JbdGFNY
zTt!g&1OYgv<?WmZIn1QKF<5L<uGT0RP$i?MImYC^bC^u3@WT6S^A#=X>=gdc-`<bT
zxD^o2<Lkwra227RDEb<smb{vjL~CHZl60+0TU`SVrHVmA11LpifP5S~Ui+U-D%(DF
z6pWU}cIkf<)U`h=0?k|U-27%`>ym(;znLu&U^_@Ng;n6#=_NP+Z26#DO09;UU@E6-
z8thRn*R#OWp%hCFb~GD)PF?K2cBpmivKL~i!OnE-{^Y^0P3(BFDRLq5EONvAAElff
z#_Tu#Kgr~X^e!6&sxPEGh*!R`Sy|;m53e`vkK0kiq-QQZ&XTpMH61lFkb?rNz%Trj
z$x{{vz9DP!DfqE?;6W2q)YulECjq&i@XSuDh8Esf6ovw6SCtNaiFkl1;)njv__IpI
z_kpfQ1AbuwX1stdd}V&<M~)<dDi0O8_f0gvSkcEm(cR<uas^r}z{wnSE!@l3oti2A
zSG=^itwlIX%;|1^Va;zW8{%?5NSO4kAKq1i{io;G{5e`%Y0m)XluGALnMQ3dq+AXB
zAQxY>UJhtEM7)UaMZt2@s*xqBJ4?to_;j*Wv;;KRj4JSzsu0RcA9mcd10~dwRg1XZ
z8n#~b!}jjQKs8JA9R!wniAU1Wn{Y}o`yO1@YjZo7AUHI#*|U4ii@Q{%;zyvF(NutN
z?#vybkM`|~BFHIVh9*A<%f6B5_BzmG<Kve>r(3kB_W>!{x?gNY6Qpoa|6t~Y4AaX5
zcIOkyRGtvMVoXEeNQrX!nwhr^Rwc{F`NoG1q~lJ=;QczK#pwnYKyrEjbU12x%=M@D
z6=G77>+wp9mb+D#exq@lu3JKb_t_M^$Cvq&Upe%Ozp1Q%3>CrqVXQk)uKLLGK4kFf
z(Rnq|wR>8xTuyMk>e2+v&BUan9EWIW6L|H@EZcnI3?T<0np7iPGI;WZmp*vWdztq%
zYCj(#L%@OSrO%#GxZ8rIl~Pz2OAt#pZb5KdJEZjyJYuTQw30gq2rw(LUa+(=08yo&
z)TF@OgEc?Nc$&=dp-&J1p2ab92468a(?MlV6-z?IlL$melV}#;Y<>fT$bic$pHFJJ
z6+L+T#xoX;1__@0YZm%n;YgC=EX&!Q3*S~DL&`1nNjPh9dN_~SK{yXwaBB5D06Ik_
znI9m_;iD16geIAR(3@wUmX`&bR-r?`?}&$Az%eHfD>Lx$@Prdl0M^(1e3O`1<>Z{=
z#gp2)x;`I~gUd%S<rIGy<M$L3t*ZanbOO4aN5-v8RC)!Q_!ICx1?;UF{?jl!9@45G
zYH*cD{}J#yl>Nf@N5^;L`+~!lzQ^J(tNQWsS_So0sr%_c%#hj~?ok*1`B^4nG&8y2
z$&tFiQl%7vF0-R_tYDyUWPDjWR+<?;iDG5BXz{FhAi)V&bV5V2l&gr#!iGynflr8!
za;oka>;E=nGGJHxZa7S-^{1gz@q3{eM>bd11iecT>bRPKk?lOU`k5%rF>GU$t=`)k
zaPDyv|3AmoC-cB?-Oi~LYx%9l_+YJ{To^%G{Sf?q2u|c$j0?fO6$+u-!(?75`^w;T
zUGu7yV(8E9Vv&96Mq`H%R77qx3C8gC?*Dv4F|;Hh4ayE6V2{TFA8tnsvy^<Y`7){1
zo#W%-UpXff9KN*H(Z4t5T~Y0D_13gbtyMstxnUK*_nSreYuyr%QV<4wk0YKyELwq%
zs}_ZB5h4HAmFr-`$&akxs>!uoU9sJN8XJ|chhEPw-cdj9x?UNIi{A{A%(Bg8FMU9D
zIKQ}==_3MRr`B3W0QU|V012AHlEN@kZRJ<S2ZjkSR8bwmfI8GsalN3HxZn>tZEHsj
zZbCJIuiwQr@p1EkBqkZAyYVX@Icok*3Hi?Z<iotPv+8L1uPiamA!~YytHQ5G3DaLI
zTuc+TJ+ii(O6BEt+7}DdQdiZI`Smx&3IlGaCsE1efdmV8vWk~vCGAc7ulD{M>Lbbl
zPb|VLp3w)DSu<8^ldQ_H1ny-P*d4&(mMD!)1ujG<C=B78kmVW~*-8~kv~~Bk<>lpg
z`}D(GIPbvW@?7PnM>+*<!ygSp(r3%1dJD_tu?H@-)-?5Y&04pW(>bje49o8YscmWK
zD;#S>9{l{i2k6J1cZCMq@&?bC9vA+@FJ$dQwj722&V!~xdow|R3FpX6|7DNXo41-O
zX~>X|IwnK*h^jz51x5?t>h6rLE5472uJ0G7=x+&bywQ3={s6-h0P-8c8l6RlrYKwh
zs04sgz){q13`!`N^}&f*yY+b`|GLR}<+CB<g{TKP2n@`qAFJKHybRCE8dIP(!Hr_5
z)EHj=i8bZ1fDaFtRX}R9+y^~xVa)qTJiNRZKoAi56L9m`_NLI<{yOM-Oj+F?O)8zl
zr)v?X|La;JQo%$%=nvu%f)GK1(X@U5BDFq`;d>Crvu*!!<qw3Ws<=lKPMvb!vf3T2
zTfOp%BuyUmKSEt6Rhkd{bh1aSo)&ckQ3J$<*hcuXXO%T=ZSvKNI}4TYCdh(qM8w&;
z`|BMqK>iN6ObV>g4h;c)+V5}*#dF8bNkAv`{}p23aTzt_E~LLM)Ev<ONDi}Q3uZ9g
z2c}w~=uk8?0QEw83$uGZn(2-sa|)=tLdu&q(_y_JWMbY(Os74?x?fCm^(o&FUGJ_L
z-S=JI@%f|^;G&ARO47GmYpADhgD>Ea*@{7loK;$J0Aw=_QKT(%9C_$M29}AZn!rQ8
zv~v{E=CZ^L1>moiLQ`qw2U4;LEYppqJlT#9?|OrM2-<Cv3j;orTU!`)Ppd4Mh<mlM
z0n8(*Q)($VI_YIIVg~fFZMBu9r(VbB=E+fyvzGfe{D~u}SC;*Nn=6T?75uBrC9(S{
zL9XU6r5p`CVxKLK89RP+5$Oh<Jx`q2Zu+idN3kg4F)M`To6L>{+>KxT<WyBA*?Aoy
zcy5NH$UzA|RXt{cigX!`AcG6ncyu*8(VIYpl1cQtIvJ7)eS^M$>|?-z^LQUE(8dJF
zC^NPVS@(_1;o`B<`Ko%0>D>>9Zf$^ukn&ZH<ROn<b`DG*dF0xfNrD1cKIGL`ffl)Z
z8ML&i16y1B+P1cdg?rp6>Nl7_F9i~Tc<_NL+i(f(?Dg6lDei1nOkLMz`kq@{{Z8%N
z$qzngCM5DYN&FUXbH8C$>4=oiOnzin$-`CMAY!Sk8CHI^EH=3Yo4GnidWh1VH^MKo
zNbY5uMst;Xnbl1_jb_a~PyGf}j$FpgT|mRh|706@$GlKK94WU_fYK|D{uN?osZOpK
zu0co;3eSL0Mg&lG=#c`xO&YL)K!d))7~$PiI?V9B=rlCqUi4Ek@DYP5N&GfMC#x@;
zsDeZ~^@75{w9v{Nk)KLAF^q|11`>=s{F3gPmugS7d3YIjR>>4T!75S7Cm^z)b<q*S
zfO8T}>f{|worQi(;H9r(+@l%Ty#vHxSKlvD>z*>nMQH8o)7Wr*PF4&K07hAh`x~Rd
z>ud^Orjj5%6Q!I&ME!r=lk{q*Z-&9kXf^O*y`KP7=ES13lTggxib2;Lo=H+8K9IN~
ze(#j(ZLb7sJ;wzy!%Bde3RIUmC1z!jBsnAZ0H&wnV!HpoN>Grmc^!ibmkdJ$_9%=`
zir4q&FXIPD6*ZEo(2bbx2mCKDXf1wcgyPmbyudfi5<ERGuxQL~FQuLgro%d4Zwotk
z(OqAmVsloSoMd!#7><Hgm%C;z^mmW^lU#ZCCA}w5dQOHqy;pOz;_Csds<scsCHx=l
zJSsFIS#U8zL*W-?`dPSf&f%&_OOYFb`i)-MfYO<GhSC)Zg?VCAaK>;}Oa&{KR*6ZS
zf>z})id8bX>9DB)a0y{s8(yj8X@Ynq+x)1WI{_%GTQk;Q(k838h_f7l?MF=PWTs}<
zT*2H?ToD5>F6S~M_R?<;MxOxPbvP8z0T*ly>TX{DwBj<ym};5L1Tr{CV`USdP4u8b
zuXWQVN#?G_5qQVcCMPGu^WiscteUjH!(KjN`<$P+i`=PNowAnj`@#S5#zcDbOuMHe
zX-IZ`%1wR-l~CxBEE`t3-+$?)apBJ3z!ckjk}b64CW00Vzjzs2tQPyGpS)`o8i(4f
zC09Q(Lpv0S$;1r>cbbGZZP`DLs}8^gQQ+EdCRc=hI!Z%u&qL%{zj4IUS|G#uT9oK%
zk6PP1Mo`;RXn2SmXbn<o0dGzWJ;!aA&=@R+5b!6NT8}*~h9G(fn4?GYBoPC6LfeNy
zYG~}9o#KoUpW4a0)e-dw6P^Ho^PE|Nc=17ZP`M}kq<jsYnC|?bx|4zO5y}Gt^5RqH
zby`2;##_ObIY|8omL+Z_kiq1>rLSX~wRjKXZ9BEpBlOBeE-i77f7ydZyHq}<j|qBE
zcC=SYFBH-i+DG319O&LQ26UeMTVJ98cu%<J?}V@vS?(F!X@+20_|FTx3zwg>XZcPT
zOEb~FXHl)4bF(j84nAxc+PEZ^|9AjEs_*Y2bj5V_v8}poKeN`>NQ&n8gWfNT+jc%e
zk&h%+Z-KQ&{K5D^Iv%8JGzAb$Ui1BLW3_ZnW-ie1l?TVovDi_;7FvUzBHQP0XiSxN
zbY#klLIJthdFhrcwZ^9feEt+u@Iox~$w{4q))Pzb6dn3u&5IP)gXh6)J!pjVJN5Wb
zily@4$t&=ntRi6x614(u#U4SWPvnTqVMDksTt8(uHen(jww4?NO(XqP5;=k56(P#E
z@(6gJq>C*$R;s#T&ROa{Ttd4qMJya%0c_9#jo5u0@o`C=Vvi#Hn;<kRmP+Xi6&Qsu
ztaQd+ZTNgxD@goGg8SIPJl<m36qYL_arZM1k4Hov?9PraPb0HI%ikWMx)J}MCn68)
zkGS_-)w7llmIJ((cqh~z8&n_}cvW)wiK7t=h~xT?BU|bC-~G|F%hZ`Dq)q}t)b@=n
z&SwH_{^wD|Bmn2fM<Jb>Sh#tp>&4%9xaoZ;`+5|wYxsuEKl!fG4vGL{d+w5Qd4ea%
zc5rD)B!8<k<Fv3>n{3Jfa{xu#Mu%HlZ#%Tk$<kw?`R5re02i;^m~8R5ApL*6$Xgv&
z;PlfU6xh!(v9ZJ%iL~%$N4Lw|L?{VQVApt+eHYHscuFbFhzo*RQ#tG~>_PJO|7c*u
zL4%110CiaLxsS`;<Nn{j^=%+^Z|NwVH#Uknq}Uj=g|XAa!CR$g+!yPmVa8qew}w_a
z&pArTMb6}M8{h6fldI_Nxz0Iqtq3wr1^$6-;jgR0Qe0Hc_gdGU(c%M|+@$=FN#a2)
z4DCmF9k+1p6i$i?Hye*i49$cmVm+xe0yEGYL7!8$6lb6Mr!4c7j;xc__Gh7v1%ZhY
zkUk+mthcN;*3M$}`<Dm!WQ{w?QH=@3Mi-u8E$G55&uSVPM()Z}v`!k2vZWtKN9avy
zyyCvCJIf+t+ChUL9O<Y8ieUv(u<!~-mqM|<;>iVv;_Au!ud4WyN-zCJK>!?p%345=
z%rLo?`y}||*PtR}kvp}17Vi&WC|awJ4eA&oViZAHi{-#Hu^x*rE3n*+{4n2JE~RO#
z*F=E|VUuW`K;@nkGsl(=j=u@M^&#81$<o*4V|S=5%{$dKQStE2D#6&15`jkC;_su=
z=?N!~0gz8>Iz@>v4;Q4lST7PZR!iIY$eQssxOJ-p5qRfDh!O(IX~;u|?FF*VwpI)R
zrSe~E(q|mI`@8{|ijOR$z_!W`gE!r?^M*Zhs(9Q2p_7idxqP%%JScW7MObio2$!Kz
z)~c!Oh9jnCc7ptiXB*$@_4$x|!6UbfqRroEZY2)5bFA_`E>2_uets<4OlNTeGMCh~
z1HO)_exd|u1T9_}O!5jzy5wtF^?UiiS(uWeT57OKMGbXSl9>N^!<wRaHHf$&UJY=h
zjzpL9{##oXlN+D}(MJN3+M&B3_EIoAq7@J2f;UO=B3Dj9!JPf<Nn?C(0xiOk&@Pg}
zBjv^0CQD3g%;RRz%IT?XBCQLU6=RzM?LyR02yo{JkxW18@I%9sX)BWDqeRsg3~zEx
zZEVMW9L7G)8@zpoc<tkot-t;1oBK>VhoGgp37630IllB~Dkvx{6BQP0QCIx)ge<7W
z#Rkvr<o7%$RHt9B^Mo0+EX9nMk_g24)iHa&kVUdOWbufiAojOmfw!X7+8!nN#e^oa
zD|KL2BNCESA|;M3tKAvP4Xa~aB3=;1+-J-%hYXV>6_8@)DJ3=?UZeV-biNRpGUz%o
z01Gqyj5cZ{lXL7!<*=PP3pV&Hc`X0;C=G@t5O3A4#H~9@YM*mSYT_u6in+xuWNCIg
zVU)I6OyeWvB97M7F|gqA9|rKQ`&GLhqA$p5_?=!C$o!&IDVsxfgjp;iAIqp8H$)%5
zmwY1GP_TqQt-jxe<jB&Q^3X+^T+bR_;ge1A@re}8pE|lZ#2!Fe20Xb9c}R_1N3PyX
zU|&AcJl?O}037OI!vSRPx{}B<i<a`iFx};6y>cTy)&PpDY62Y+rGc|m+tB*GK!<xG
z{jJ_em37kp5n(>kkV0U);P^EAf2N*mn;v0wD!1al4gdbD5sbQ}q`5iu_2>7D)h?He
zS@qkWcC31>&`IM5CX3`4u<%eq-9ILane@iJUc4QyHv8amXi_|?*vaU?(k`{*hH_{T
zJ-9G8R|FPeWZQ|Cr}JO5IW2^*<BQ1I5UD>w;$$<++rBBC>Ty~3Y8GHjrf3k>PhL7-
zO7(0`qJ;;sUi}X5Wvnj~Fn8Kxl8C^;G@2T7I~dQ>fw!{L8poUb)50jxf_VzX08=R)
zRqp-6>wO8Y+2@E;3|}=Frq#9eFFQPdF#))gb!(;qS#cIX@$)uc-jZ#C84~b~O~}<>
zcv}<Pxo@TGL(vV<nt3TD6j594k2{7ieKgaTCvhvLSRs^ywUhV2(p2Vk{6r_^?;cE8
zZUdY12wYHqij1#Bie|P!W{Y%MB|ZZA$78y_ASCg{J%gHhNvCAqbALHTA2wY9Sb7G8
zAjgq~r%vU~Y{6k2n;Ofc3qp0FjW4Pt<X`F0g8RuCZ4yVxRGMiC@^hJ4it3fibl(qI
zu&dOz3A4)%xlFXiahc6s7E8;8me)u-^x>32^$dK1!kc|)BBtDs_=EI@JC@cpKt%1v
zS4=Jd4+hbU;qw^h%~JjrC%-Ht9|3A58Fp&sg6@9``zUWQ9HmN8Yv`_)?D+(W<!Q*(
z;h95}-7=@_%Mq_zEf)1o%%mJLo3L9gc9~PB4~9uy9Gp`8oX0DO6W9wZ73ll0clNOJ
zBn{EIBF>3Mse(v;jf>&`nGI~z-0FVxNv8a*jL#F%ORJ{&Gm8`m9I+#hT-S0U{rLVf
zLNKP%+{H?<Fsfp<QG!^eUO+K8{L&Y#5=E-21xSwpU&+Jy+SctZSc6NNNIroU85ZfF
zOVj5kB;Fq<e;zn))*4dHyuMRldqR};sSBbqndf5nf<H_pWEV{=qH|?YpkmJ^SpUD-
zXKR1#Etgb{0zjjQ?-YF_zogRD2<1N>!F_Wkp*t8mN2Uy60OIp+_qi_Jt_HKSSH8Q0
zZEzIv+_tXw0aNTOHnQ^bEg1=Vm4{_0Ms4o&Y)+w0;xpflk!tJrIyPTQ^*nG66OCW;
zoEqtb3S}ge@-sYmjCUXp8ry2WM1fo$^SYLt9o98s9s@rbO!aME5rLIfjzIBf?|i$5
z!z5}1JE<Qp$A^ptTcg#oL^{7`X$GNgAd}p=(9Gi|83-jvCo-&LWY{{q7{UbgphxAb
z=<J;ChCj^+H&*K(pmkm{^WL*wz0)7Odn$O_WjRr3nugk2Byuw8-WM<Q*IW=v7JOxt
zkDp9SCRnr=9y1g&Yt@X~qnQKlwo4pD9i)r*tS&og8MleSOO{{=8=s(nk)w0(ibLn*
z@;pgZ*DG<`k7^8aOjT{dc=3OkLR09pbTLGarqGmqKL{&t3N6wW(q}CDN3J;Nii?5s
zqDsF*thz1WqD^Uu-E!9c&hvi3;;Sa$l#E+4RU${3FbB2<`n1%|E`T?i0S22OnjY}D
zuD?#6v_DMRU5A)=G!3|4U>3i8uDGj_HbRU}3LWqq9_ecfckMS{DK5qhmy$Eaal&R1
z6p5;JJWW%R)rG6B!Bv&|Wu)iPm{5hJm_ox@A8tB&)V5*rS;}<}2A8%QUMwbz%Yof+
z^u`=P{pbXY#Q~qWYAE1F<FhA-^4C>J3XjGK-^J?zb85ADWDn@|AjH76IKqIPJmr~i
zzSgCv7)+{ap`IhO1rf`f>a{8mhuPXW%$&Kxf5vNOJ=Wbn7ODArRl&oB??~G&GRSB3
zIY4qG3gYV42lLHP43t1^#F+-v&Fk8iY_{9Cw1?=Lu97|&J@HkpF@{%k>DQ}!9k(qd
z+B=Z>W-W)L<BWhhZmb>Ys7G&-vx?LhQ$9+37F)`Mgr=&m8*L2fFysP~e#h5bV3P0h
z16-*jl=7EOsCvA_H<2s!UYf|iztkBP;kmP?2#}wFxb!SD6_~O$i$@LwW;g^869Piw
zMB&@GYv)BLo@juu6tBjaF-ac@qo$oQY?5_QSm*x9hFi-lg_969&29odA6*naPEL)N
zxH7LnJ7z1g!8}Cn@d(f4q)Z6Mqg5LbHg5h-0p^w#nSbnjhIfcKl8O`uEe%`3S$H6v
zUv6LX`>$D6<7vzB)jSzk;py-Y6P0*YcNLpx6-+(CS#A#m8o#fbVesWgt^xq0*k^-o
zLf6XI3flo0av`|S2d^rpy34=NDnH5Lg;IfXB}4{5)zZvuY)FEs$B?8iz{59~c5XC<
zUILYi24NVXzzL&Jy+|2Ji3>WuErj<Gl{^9d>3WPI2`Q7sGl#`}YNH<7vxp)5&+@<2
zRQAtL&)fxl?p?C?=BIpP8t&X8)wt482o$7|i%p7d{trw6ill<BrayJmy4DsJ3oT|>
zF*OSGq-2%<`<>lp6g<|(#h2o)-dY>W4V9IgxwcW<uu!yUy)%$Ix5)5$A%NeCdI4UJ
z0z(ahFd~+Q-q|#MI>txHIg)!wh5YjotxOKa!1zu3D9}JY`rA?=A%TeZQ(gmCH(0!&
zSf#T-mMa*x`q=gFvfKSW2<q!xAYO>(5xNMh&lDX9D{>a5`lU0B>`{~I$W^cDbjh%L
ze~>mqM+e)h$Es^;f1ehJWV^cdUTz65RO#+S{qR1mumaplUOQ(FPHLXE`kg&4i`n@r
zr;Z<65($7^jl~Z=4xquE*`iLLl933KXA~A_MX*61CR{BsT%b-`x(n+5+{aDvXc|rk
zesb;g5s=2A3GzlGSnZI(AE+?s9KGKYUf|^kPv=lW*ftOD<~+=fV=m%Gi?wuLh$yT2
z$F0aO#J5?oQM_bZ+fXB2PB0g4lH^?YDIckR21sKL?oiB)1lufD)ht|TF`Ey|r5Y&z
ziwSYUYOGIwG6lMso)p(#-&ai2_K$yO2zi`HO_NF)VOUnM##<Kch{&K_#DSIm9eJ+)
zJDmR56m~B3!uFC}titpP$WUQYpeUckUuaFJmeAK^$*di+Ts*R7@k-yc8iu2~ugeq`
z?Okb?+xKdIw+;Gp{xnvhX{}KT<+-2+Oe#;~R(cgcj{rc*?MC{yzPhz!0ognL%~Zd_
zvUsH!kw7YkaIL0pig2@(x^)Q_`$(#*qlzr(z9E+NeA>fXT3N{l-fm-KC<i+9B>uyd
zsz<W19*EtZTgzahV8)W=?Q)homwAj)iLscuo=c`mq2&7jY-`Nab`7dW-Kn9lXy<{4
zVn7MC9bTJ(4b$)W_J4p2XuewXX!?8cA)_Ta(}N&|)WG!cv7-PvB~!AJ++I4Q%(s^C
zH@%~Pgon^4#uF?pEdkKctnTrwn;vKNSCU4J41&dhXOq@I`a<}?X|L0e8MP1IEG5r}
zwFgfC#U(Q|glz_@+dBWeU~OK9Ar%jpfdH_KiNy<Jn`TNt)aR0*E#Q&*)c^zR=QJs-
znoyip9cx^mB85yi30!`<4-PKysGiC#<T@+jD=+*zqA9@oPCKKFwJ={@`9yDW5YtKX
z6$$~DzS3N*cjaEthutg6&#!k?nLY@LqzFIJSTV8&CMS!OIASE1*y>a(LS!sHP3P81
z84Fq^9ZDI-G9NUX#4Ug2BO`a1XKH5`ZJsZ-2?wuAu#Xh-C7F5ZlBtbk&r6dzN=iG8
zRDRoBp>3^*p1Sj3GK`s5bE~e?D?+-i4*)U0^)MJRW-n!n8ZMp;R8%$`xvy{&vTRf5
z(sM5$zLHa3!ptoIK}6~FB4k>V+J3eD-;LKYtJ(!g+=3KF1&noW2ZK2cQYmrc%vt*V
zTvK{G{!jV-y?Zv^rdE@doZBV--6M|uS4RxR7R9F*WX?PSE~I&*Bo;D1U$N_q*Glqo
zf=jmZR(CzuYtJ>?thJ@#5eN%ZWt9_<i;*uxja^)js@J|kf1(dSKi{XU$l(suMHFIy
zr}ixB-$x4kY8X<)PU9IV#P(mvObkCf)JA2bQ(^>Ouu@6Y7b)X1P)WMvAIURZKl~hd
zMIM8k0wByIZ8=oCyA*6IGW38grzwd}QE+2z<FTyU;{KaFl0JH{PfM&hgS&p&u9n*j
z(rlSBkwucgoi9oZG*e08k?e!yrbP=|%v;tiXCBAR7J7eGcld_M0QH6I^9TQyIwi}O
zc5GiiEG|U-ZO3N0luw(+8;3%=csfcc%ZKw?Ekf6Vf7vT~nEYo%jeJXXoWY(A!%dFN
zZEeh*itF7^tHqow?_u}IXa8b$yR~-E-=U@TJ*|S0=94a7yc3tbByruM{)yhyODfHj
zen-wRVLh%tQ6&^bx4D-*81aBo=`|anCyH6$Cc$pta?yXRj9FG0cO2C`X-<8)F<;+)
z+P7fc-plO{E41qgA*xRT()mFel`jG8O!#$)*;(jb#H;MssISdeecGxdhPv10aYWCw
zNrm-m`S=FIei6a?;8)@dq7}jx&4C-0(&)vY-Enc%&?Rdt4Aqoc+S+#f-;_Th*g9V$
zodtHhMITdxo;ai}g6TV*VBvE2&(A~0Zt##-aGC?svYZX&?6VCL@iwRA@*bN1&3!5b
z{TT$)<DwVN=NHiNmyt>F6H9wXpP5S-{771MeE&Xi{0I)a`J4&a8oW=7Rt+W-gcpVt
zOCG|Zl=L!aEg}%!p!8kL{k?DKK0HL`)BkeOIksg?@hx~$C7j}Xn%~~x!qeBO=V`!Z
zlFi`iw4CkfI*Khg98JjQ@fTp93G>+SO6ik`fNkyU6r(yb4@q!eM9uNeFDz{+Vy62?
zEoCFN*^Z`bAo$#wZ)Nk`NhOI7`;=z78(SC!fKAymrf5?5z@*{r(G|XLbw58tEv2D!
z9TnbZsI*moym^p#f8OJLU0WyS=zl3;<2_Y9>EUljt1TVO!ryzpET8wCtg6r1I}zcd
zj(`rt`_>=Ei?muJn8)ombQlSyzudU^a*RPS9tF;$zkjOh+w^>YB{1;41D~9%T$P^d
ze+W~ywzipjt^Co<<jf#(@uW7|gSK3OoU90z{gj&LIi&vmIlu!zO#`#nWlW<@b8EAq
z6Ca&oG(41&+L6IqXoB5C!UxBfnzrYq_Pgd!hXJ89O}fEIeoJqeUG2~E=|0ifDD0}x
zD19neda(q=5!{Z`^tRmf8N>H`$;Wk}XUa?^P@jbbh*W6A54b{0zD&Xa2&1n=wh5fb
zpRF{&0~x9MbXJi!F=vh2_vSz$aX`+SDn?+lU@ARGp^kHJ>j)ub5m+AB019>P38k6V
z^*fyM<Hfr&98eeg;lDROW>Jw+?%(h6_Q0DCyo3^_aip<)f~G=3WQzybmKUXm;o4`^
z)%`2K7CUl`-(hKb<U~;d<;BUG1q@?|=*!$kU{DZ0#ZUe0uI;@Oyj$NK8{sNAz22_(
zrjK;dx4DnVxneX=_osw;^FHg2?d{{tuICYho8tkb#uXO&0!L-6qB#Z&3$(}TkGqc;
znb`NK?z1MpiH({+SlcD_MmY3AYWj$6S#q|!1U0<u3*nS1J6YmzL(lVH#=;JgWWsjM
zA{t@MP^@1&#GVZ-3uR$r&k5De`1vw(TO&J(RJ1u<KO=@|fO8j7alY`>H0KVP^N+&m
zif#rDp#*d~84_#4fnt@I2hAn9&;#SXhc)IqXoF{vC6qh#<M9>CD$%>LF7a{0;4BLX
zT1k2L4G^8<4b{C^1c!N9<YJ~K`TTt+*IJF$NOQN-=gS7uarTJUGf35ZJjbm{pT-PC
z4*Y%ZU{>xD`s%{kZ!R$eKrQVZd&l?b?vBCuvU!w}hkB-T_^uq=;{Ta|)W-VZAkcq2
z7l6jgF&c^FL{?N;G^In>p+;+VH;F|z-7Eu_I^Fz)U}w&~-Q6U*Skjx@857((0zh&0
zX8+<mSj#~R8W^76v#~#C9jOK94Ug0&wGXq?;wTrg&ORWRsm`7lQBTV!Iv;vhD29t&
z{qVmt>3ltwujlc34HGrx5v8u-*VfsBJbuD~nzU=~e9l~1(dlJ}ZyLvsDRN=$Y^Ame
z*{llWyDE;jRBrcxMq4J#dni~Kg(d0@UjSvOg@k`<Z8WLV=KQ7;$xK6Kc5_;s3FMQG
zmb`BTHtHOM)ra7}2KzE;*r?*S=H{0+nGkCh*l5#gjlf;_cK`oxWQJp}gjA*no!FPx
z(W7bqW=iznDJq3|QT|+mcU3>TPG3*GlBuk$uMHylG?Q)EK%?x^GShGgclOU63n`S3
zGTtEy3*1-`L66s_w0!QiHqDJwK2NeDM;W`Uq%eR6biW^rs4K-P(?<mk_$YPSc)4tF
zzWqeH?<80E4t9Uwa#*f@6&Oq8?Ye(qJkKMGXQ#4R;@uJ#OgdkWTZ%0)Ud8ikJHK8j
z$@g3zVQXpHU&HY-TI^k_8+K2oa;P`*jf+KpUk_PZH}$eDe(+V^K>O=9;ojyuHn^R!
zzKEW4kwyQ+V>8>HLyt6_&Wr`wn!c;xAzC<TAoj%I<{KN0pTo0UN#C15iCW4$avgL9
zO4aHNN|*oY_%BlFKg4pa_k-41kA>nPF<%a@kU9xFH*by)ZtHL)YJ2EV>!nT1s#PKA
z-l6D@Swl=P0=ahF5_sKAoEcmhS_?YTZAdM2K9}@x9f5bPz3IW+6IRyz_=6aF9E0px
z3gM{7Wo9kf06HstCYh26SaO+yK*)$SyEQ_dj)Tq(xmDiBMO!me7=u=c({7N&x}n#m
zk2C+FwX-k0i+DbF^LRr3?>O`B@*5w0OIfGuCnQq5(C_Q7yfu+dxB2K5C(MHa^#r@*
zJcNG$3Fm4CL29N}zgk660)oek3s>LZS^d08(9Q>CX_A_R<%D>+yYcPsxCNZ4l}z}o
z(aSp8`!8Ux>#cz6QuRZgm9^T-&8uf%o`y}oV{`tA>A8xer!)z8D)M7R_W{1-L1L_^
z5wLeFIP&;8vB3ctGr#=xwGk^1Uoh&@EUA9`71tm|ab(gDYk6VKD&jr4yZNhv9@HSV
zhl@ugW|bFXDWr9~>#Lk=q`GffN>PH%3o%wF8I}>RCHiR`M>%KBA}4n-;7$|&>v$5R
zUs^7p@Xd5IPI;i8k40|bEMeUnq3uT;4nfKAG|%6YFQY|=lGhe9gk%Gu`I$<GLz)~J
z;hDsCHKl_up%md*Z`+sQtw)-bKe&UaSc+UTaK6u@Q;kROZJGAIzNwQE?`Dc@Tz<Tc
z;I%)tEsiUZIz8X&DaDAGF9~f$Sxhyzi&)_Qd~F_NqDkENT#43SRCwksfWpX2To|f$
zU@F`^n(HH#Z+%T^BESN|g=20aIxmYH^#-0s>+=$gFV)!{`wrt9>{-~%s71uY#BwDM
zJ-L#L|FE-S`W<NoE1`#`oBG)|Lu0rb`r-RLD`)F}Sq_kN*jS|}msywC)aYG`Tpn(D
zQ!mNV%P17Ck#FPge7;*_X=-Bo$FtbnQ?rIzLUFmd#jq~(ykPVbKdG+-8Rs*{GlmqA
z%s1qqM`+r7lJL`ZXI~HgvK|!VtOASXDyc#S-t0G&y(@r{;SzbBgi2Le&!KH)uo`Wr
zsg<=Iu7dg!Nc#mrnY6P{hRkO?C^2kTi{fw9ri#Y59ec~Aii_s9j$9j?Z?^>+iN(o6
z4d3KY@y7Ndqi$R@%5kJr6N*NBy>C%)jyl(Nsh!I%yj!d2jKeLL8@JBwN33a{>b`%E
z8hyIXJZmgEks^569WjB3c#Y1?V1&T|jcszqr<(X-O^-(1ot2lggtsT{=j)Oa>&)<7
z4AKg3r0y@mA1sp751Fh38Kp#Emb;g28yPZq)mL5*yTc}=Hg7du;+2}-0RoeQBtrf%
z2g=`qBI{KpL}Vu(;VLdl71)g#EFK;nf87)p-7>R*)6TyQW#y&Yy1}Szjgu@9;JN^n
zlOWaVFTNXdf4^86b4VxR@h=pes)=V3{i|54_`$>jXVdy_YKt`FK!3g)yZ&)T=C|5p
z=pk{gH}$Ky&|>0CwI@dBb~mJu<-bq={Rfg1`}So@GfkeJNa#yYvV8&tjWwKVWLvmt
z>Wh;fUwld4TvENlWn%F6?^`~x50ux4BAcIzbINQ0I!_|?HK`!ondn!730od1CdCam
z+eRJ3loMeio-89*=?&Ci^6qg{<1__CSUji6{NDr1*y!G6+!J0OL9t%9hZyC1SU`$V
zn@PXRZ(t(2qNU5uk_tXov0xQmWHV?Zs?q)74pNnKD;pnwjwjbH6V}7kd?9iX_NjYk
z?U#H{Y7(_7K9V*}MUAuJG=dpFVBiLKo?P{)ClgpUx3!tI=+^2+!TjAUj7lhOSxv!6
zpx(K?1@9!9=;D86inQt}-T)}FLHyqrFR!1|EvED@pSJ_cS8S^6t4dbVnQ;X3gMvPb
zykAm@JYIeV!f>k6H#Y{anf_N{I;dj~AL(7**TLe`%_W14`Qk(#2LA@K@O1}8e$p`t
zVMIx{e06Ksey&SVpjEG>rr<!(JVgIIMJY+U&Vd#XL~T?ARf-C&lkl5M1%ZJO%Okw6
zS+x|tC(3$C@RBmqF^K9!KsSj9q!l%|*>MR)Ck5KLt2j?S{F|VRc!`>ydj;vV<M=Ms
zq6u#x%P`BFZ>N*MF<W+~ZJVCwo{YH=x6iLPms3v(!a|LW#n+YZ(2t>8%Pg`|9n5b+
z2~h%m_Rz?Hkfq|&W&Nh;`&!CtSC`K~)I7@Gl1ULA+SBdY1ASuf$vn<V6=v^2_)iAP
z8T!lwm}W%BHMk=CfwVJz9)iRYEy*!mE1$KQs;bh@#4NZ%Ue1GY<`v#UXN`GQFlV=V
zeddQt_6xf0oz(rmG@RwKmFV&j-Nh4%ujBEpV(D;gJmBL>7@JlQMZ=eE@M8Zyy!ml)
z2~f$iXlTKkPfa28eZ~Bot6kIDQPj~!5uA)pnc*gLuxT5wp2WZSV^LxISM2ryPnMDl
zStEp0Dw*2b{#&QSkoYPdsdvINUlaHCcr4H3=7<9I*}|&dyUC=;tB90WH@mxT111^U
z(okPXT(tLOn%aY9oH+1n*IS(Jmpz0=roK<ZNXfFfc^OpfBK&<`?T1Y}89U+ZU$s#1
zfX|?#d#|e+x=;ThayiP%U9wvzC5UBZBDAVhHGH`Eq)nKk>rC6;-oCk`<M)du4Sm*Z
z<Cn!B3A{`#HM3yrmE^mhQA7xXcz7wyDIN&|CdiUv(per<W;<tYyC;p*jQt(8j$x-~
z!2}N1C!On)zFbbmi|?<#VEp{dPOtnN$ulWf2P)a#fSVncTN9(=d{Zm9YIr%%{z<`a
zo-Y&dlKiiXUz~5we~$^;Z=CbDIAJ|dY<BgazpTV`UTCtWwEH}%Q9@PbS`ohAU%V%0
ztmr&D7zv$77&hqEY1{J&ob#)sAGwYKJA|gDlnRX|S6`VGo&@P!94pG1K*8MpP2!B2
z=FCZp%<pIbXUMWs|KG@3@kcyT+uJ|A&oG-uRwbHFV>W(+v0kL2CRVIYbu6Rq{kelP
zNEW&l_`!+AN%YGoi1p^v>2bdxMbt>)NS6(R1A@Ki_@Sso>MGKtD^iVa6Q<0k3JKG<
zByZ%&ik5xGPT>v7W$S;mD?bgGa+>f68VY3PvqS&b(W91*UWnh>kw~hoSYTP_7nhZ;
z=id(f1<@?313)$mfWL$u$856x%gNqzQ>U+%ecF_TLb3XP%SDA3f}d!lajHqMiomLt
ziQsR@9R@OOjrn$j3oFSNKZ}mE54B@wAr*&dNrZ*!^`SX9S{ivt6k~{^&XZ&xS7c6(
zj@^I_4@eM3@0VLlJsvNmyq@Cu58dtI*sX9P3){z4DMN6GNywa?-4#?#C2?k?;}u=L
zCDc~hROx2}$A{mfMYjLsnziCegYllmE_NaUAY%Aj_Q1Wi`9C<YOS|~G4P<{Pe*}E&
z`M+mBRewC;Tx9ybEWTz9V71L1p6_GUUMxDc4zDAUOKROp$GTI-m0-6H52Q?u&=G)=
zIU8WjvGQtipjya=k3&H#>d2pY+$FItwRajY<q^0UPjMf3p@xF+T%9DLz>3qBoOTRg
z+7Y$9hB(645^`sbI2;(SMUGn%tK@BMQN}3fD!Xz*FbxZm0_)pTqYc5T=|#5zFV2zW
zETS4Uy~$zNr$X#K*IqF}2@m<QJDeM!^~YHJfT|3wgkrmrN;Q6{aOd#|oOlFGoVOSs
zV{H02b2k>N`g$Bd=fTku0vajA(aWO`m~>a93l1rM>fU;?`3Pcl*QYZ|LrT+a=2UX{
zj(-pn_8A$cYDtt~F5jViZ<2})=(kWNHJVL8=RTQFzS$%3SV8I2{Lf3M!KS1<v|OIT
zm($NOKSW4mzl<<{kjU#BYI^V{%}?7VE5r{v3Pogl{BR6TlN=<^#r(IgZ5<|>diiZE
zg$|aj5VJWBPf!a`M!s*~Kx@0GNhvdz<P1U2$3AYGX1f?j4yM@7q?wRee2oucAgUvy
z>l`mWBvIsz+e6S`ilIfoAG+Sd6XNrZ6%x#Gx$4Bj^MGE}GolO6olkdFpaXnR@7@<K
z&OYFo&_*i+qc}&m4GvG43xRa^oT^5A5{5D23l^&~5ZBn%OkkD6i~tP>5OnMD`y4j>
zvo5c!>`DOtTuXntkp$RE-~cy+p;*#*JfP<-6793yN-W)S_>^;`n)q2_!TQkZU$?%S
zgDJ1)!-&+IA_nPLE6$lQ!pl(M(F`Y3avmA?-R-CN*-OQBH;&WwJ_p(_4}BNwgv-Lp
zl@0Y~MpKC^`AEYJcSdnv4x*!JVZ}_ajbfjDv9wzZ%Hwb=HQ}PWWIF<sI<F}A9)>fK
zw=OQ0@Stp)P(HRqU-^N!k|vY%HXh#MMr$F7Nx_kzy}bj!fCvDGk4xOaTK9TJz1xpz
z)pzP>9h70CU8FFMHsfR}C{LiZMDaDK(_eG;O>=VnNhuy_OzvHNa~n&M@`yq;{2u5e
z$*6w?!{#~Q0&$r<a5ay%_=<I8g&^_S=s%_``e=+gKbsnD4EzF|a&C@T_W8%s;t7G)
z&C)iMPF~ZcVCVFFW5#Xl!XtqpXFZN=T%XICAz-Po(n`xl^lSvsQVqG}=f1T2`A#0&
zOK9vWSa6MwU@xL*3Z{p{e3qE%gDW04M>;>ZF0%%$EF{QeEYIYAZ7#6W+0h$Yvj!F#
zMs@423r$K^Km`qe6^kqCMfxU(pf_sfJUmaBFUehyL<^ehI2`s?-te}C8idJt^EMd7
z&#9y|GrtTn4$o2Cy*vbqKCY8phxtDOcJb^tB)`ibWl%zse7>w{hyG3W$G+&z29OTL
z6L~pJvE{L$rNeqZHFzbs=z3tSdrE4x)3iIs56nLi1m;`Omsc*4muq>}>T0b;V`Nfs
zY$r?py^)=;*D?R6;@258zv1F+pWf_`r;nfyASNEqCh`IXJd7hznLQ{QM-6_(y_2U7
zmA}42z+VZw>4Q|{H}>xsDFx2Ig^^Z6L4aCZ9>P$!zPlf6BY0SU+mkf(iiS;0m)DGW
zn}UmIVd=y_wQH6ZkUFpk(5um)(Mx^(*VGFB&BuNq%nJL;Zuq-rpP=dpp;ZQrfnx@F
zeWr!9a)7MDMn@xqr8GXHl%5V9+iq^AlDC9IgkD-XA<{uln>uwJGb~{G3q8uRwk|w%
zn%N=AN|~`h7N-I6_Hz1328#8Z<Im^+Rs<abrriNXLsl=-A5gFN4Gi2<;7+??mH$ZW
zR7$(w6G2nCpb}Zfm`tq5b_rjB^+J2Ar`X)$Ek-E3-zgn@LcSxRHQ9)(=@Ja$9=%|y
z8N*QNfMiIups{yQSM%21B?0YJ8Y#)u=7`DqQ{n}esL0m#cI3_CLGM^8Qzn`TW}l_#
z4*AcC9AwgHPPIcV8ZH`1QO05qJJz5tlsQg%-61jk_~9%@{}QnN*afXr42Qv9=rl?U
z>=kXhk!dpi8N7sx-7w0<I<KnsOe!F@9?FfTmWYmCvLrlJ)nFzIe0Zd&)vKJrpq*o2
ztl6I8Mc(F;p=P9^Oyln1K!1*rq@mX4Qb$}^{3AfeU+ejMCy8d80A#C19y459!h|yS
zglr?_>f+c;{ur>bGR7gCpHEa!LbS>bDWKBZ*vadY2_rf}87xX#P*bv?<^e`^6XuQ!
zO$T8S&xLZFl`?76`w`1LfuDy1!vp^VK0(30)h*9nyh1C((NUzFNR(chW=yAZhM_|n
z!+dtYe0E6N3`DkLfugv(yXD2}7nsB_pULZrz%V$WL5x%d@xqg$ZFd;bZ*p|sXvN;w
zZf_JDZ`B<mXHTl`QB3u_jW_LF-u&9WRV4l`(wYyFx$3;+B;WhoasR!Fv9mEAa(1aC
z1UZ0(fQ|}lqpVe9K5vL`pp)^whz%oR=Y9voZa2J-+}zxaXOQp4Q&dR0hI7&l2_cea
zC5Hz`93CBUusEP|?O4bvhGAg6*^u8#dQ3T;ke3sNp=VoFn423GuU~OEpEI*Pq;EeK
z|6eKWBBJ~o@O1QSc*ZaJif3GM$ql!3Yzb=MJ~cyOwIxJwXowbqhM>fsRcYao!?+)M
zVqyr^;Z@HN2b}NdJ6LU2q~$TV9v>n>Cy**On=M*L%1O@a^VfXw`DgUQz<hQ<QI<?5
zQ&N@DEgcsZ7d(FY2<JUj-H;mbt8>mn2t@BONy4N|QE2Mwmi1;s2tf?SqeJV+Fbu3#
zORDXf&3enw592}W*AHtA4<4MdSRC#PC8Yz_q-bSH>`sjk_Y?#{qh+2e6$n8|+Zcl}
zsYF`39g6?W9+r8YF`Z7yvV^LxSuHnjeP3QodHT+Ky!YM*{P7?E3E%tP_j&Z-A!ny2
zJbZLSnx>=m*4tw~oAW1s^d~G9bAI{DUvPVSLy{z%o}S@?qit)dw&HHR=KSil<SD;=
zC8Eb|Ep2mSh+1)Ubi(QB86SN30pI)dBl0}s#mndX`Jexf{Nm?7W4qnb4;{0EDM|&t
zYpBk5L+m2R!WA&ek*YwVWK-VJOm|LQyT}!-cHEcI!a@t0VUz+7o68$sfB6i3Fypw$
zNt7Xpo~&LIHaCR9p^{Ow5F9481n=a;8pS^<#)$G>O3N^YkjDW(qJ`RMbdW-<*;E*7
zQP$vHLmxatbcE>1^K85^Camsm>Ah!`l^7Mt3qdQpzNc?$Y-Ta3_;huBOO__2c_IsP
zKM)eJ)9^tGm^2qsaoe_JNy6;ljLBll=5EDe-c#lUUE2|Q*>JlMkS=~ro~F#p2|s#z
zMD$O{3eA^I;HMW$KKtdC#ld6F&W?HK@nen-XWTBYxmn$DyIwK)07ll9Ap{UGliES2
zq0*oNWYKX|ttpiv)tbU2q&9)*SuJmQaedA&p8Wz-Oh8#a`S=swe_wb2Sy3>=z@}-q
zy1ik2w`AD1w1cB{5r-ydEvSSWh`l3@^T8PLPtx9@yeHF|!Wt5Qp$4}hR4aU4qkSNW
zkxkq2YO~>WJFtn+kCXn<=?Nb`dCHf6_1B!GQwrabn*h#J<R!<43x>V{55Zb|kk~VY
zAc;*?W0QnzRO!YT@uNZSp1`OTS_L#76)n+8QpHej*<QY6Ha|gS*=XPmdu|<qSdRov
zY?5$vd@Q^Z1;G#8++N{&Cq>YBAPHl26GvW3H*{2WEmk{&lLcDsC=rSz=_Q+YEbgpL
z2q99{HFdMIL(uyogpniWN2PAtG_*}Ci*J#UT0>$=e43D0LvRiS93P*sTCXH`7R7Wr
zAq3BQvmzp#Qf)NOMY^t|Ydh+?=IZ*2n``k()8qTB^gSvj3k`2&&4Lt4=4<B4JFXVv
zrr^Tw-!%1`L>e&)PiPoC{m?VGL9PRW$Vp4}J{3dEp_8$QH}{-5cV7&|D23gQk|5~;
z$;iQcLo_m~{Qb9lt!bK$0;Wvj_*UzERyXh~d^%$_X_AWl5YW`E6s}{Tn`H@UmQxfZ
zM@MHI9vm`XESOE_l;s4i`jLNoMAdFaN{pbkP1~^EtkFiHts$S3A`I%qwD|V+hV^>Q
z@oa#}*RBI^idlZ0tmXGRq^#>`)rvYUS%wQf4?pEbU$fC095sp_5DOCkjm79h4#+47
znlM^<#h}JHZ%}HS1JTO?KnEfNKJ<WRaGvdE&El{lVu@O#w3ZYQC)OwH^@h5sxwyO_
zFG{j3BhPYX(<ynHvtF-weR09z(E(l8Q`Zf;m`Jv|^APpejF8r9S)`P?WxElsZP&^5
zo=i$Hq2F$~yu4()t?2vTj2J5=oq@?@iZOO{Fc40z_i@Z39%&CFUk&bgrT4r$J9cyu
zl0=c`GBG5U&S31ASTY`<td+L!VlgMrGghk=uU=g+^nw&lXLEk|!$0BUPrt{9AAQ0f
z{qTo8e)NQCF`>v(id<4u#@q6(VXfux;E-bS0}d7k+}+*M4?W$`^ZMd7m$%ove*Kc?
zFJJKT{3Tb{S6p0Ov0SZ)QGlMTEIB$p;lZOPoSmNW=-~tUrsd}9lI!a$zWCxZZf~zg
zzHKCEaJAca55DdAd)S?fXQy#9*%@lyUl_4Z*o|E*&+bi%BYNQxzLFaTZeLxHET)u`
zlBr3_@z}Z{^pdsLdr!cVSUaBbJtnHr!G8?0-ua#xVe6<-c@;;#oo2KXa1J*NC}RW}
z(+Znef-zzcY7I%2lIA%>-He;AY^o-)Bq}mEPhC|^(gj9q`mV?2j>IMms;BN+TnwZI
z=){s|Io?UuX|>tPqMan1oSw2>t=VoX4(1DtHn>q!trR4wBFj=Tt%<Z)<#~EGV--5G
zP0P@48Eyt{*Dtu<yri7WI6Zs7$)k5UK7GV?z2f@voSW5(jT`nwwi+Kmjg}WNLKgye
z5w;2vCkYFgF;Ylp2C^JyWkJGq{P?f_lArv|UvYNwfP<q$9=-c6^P@u+(+TrDCs&qo
zGR0&$&MPUh{lL)HxK4~0BdI8jhJoOELRaIe4a4mc*Q`-Pj|l@N1cnf}Z3k|fo~v%)
zCPX$S#TiTLIv(em4^J0de)&tLR*7&%M_TV$OeQ>f|9!M^Y&J_WD`uTaX@U>rsTipG
z(YnPXmMqKnX$#tjidJ%vAwh|@VsMm`x$uINVSRJQ^nH&?g&!jA!kv0i3v#VwflF-4
z!Tb<wEHOsO*xd~1sKxGr@sPClO_L&Xy;^g5b;)wIqU!|(izti{3yU;OM`PIqv+05?
zlNHK)PjDU;qoAg0=L`mEVwq10=97ZNYW&dBwi+0)#>hsjAhCwobV8aMPEU@<LPQo!
zV+=`>P?jZDkF8!21Mk}_y6b4$j`NEPE-tQ!<4Ij?Wxlq?e4XBb#30Ec^2Yb<&@uSN
z)BEo_-d6999KaAIZ?$V_+J?4m=)0bXsE)0*d!))(Lz1LqX-1ysd*O=K8Xr6svtJvB
zVW6#B>a-!tauOq^zG@^l3-8H~wm{#muv;wVJa}-%+1V*oRrBK?|1DKj|Jv^_%bdf*
zLk<=P1Qah{o>Ny9L*Jpb;^^dnljAdvPtG_!eaw@mAMnmQ@A35M6Alj#sMEmN=^52_
z$!5ExYHBHdhk>^1SZ_Ayajl&n&M{gMf>=@9+}?0^cgN%7x7%%d?>-psg@25qP9Cvh
z!v!z6jL&)Qf60}5K}<c0K!P7BK|LNvM4?I1+M&==j0Q!F3Kb0zX)js@9xvU4=&%@Z
zf<SoTN{eE@s+JsVH%z7pIxz%ga9-$7P1DgeaviU4Z<$P{92_jj^PGdjLngBccNcd&
zd;Wrlj~+0aPASuzR42H>;hl%5FeZhOE1f2Wqr)ki%?jrNeLt|>Y}icK#3*^t=jZ46
zq5Z9)^iqt8F?T<RP-2&A<3~q_5CjSAM-*QvH5z7%*v1$m`(KvrnJq;zCCf5q^M$1D
zOlQ&=)N&{(%Ywth1Ey1nG+eJXeE#K^)K!BqmJdJv9{={={5O2~@h6;}Jz!EKxE|@2
z4bat4P9g5{;of>p@;OT~9zA%>Y&zrB#VgLAU-Inb3qJqybAI;ApYrox{G7Yh9bMOr
zKc{3*DN2rxPWY1_{WDG<J|s<3s@0l5{~!KG*6TIa8qUtnI6e{@P2JSg>n;6d*ylWd
z+wmQ<V&i>LAW!)%1>5d>;`p9KrT5Je`F}N5)duY&Iw~@o&^C_si*uIcgtIg!&k|T}
z$RpT95e)<_5n#$m#-Y{*jK+<H-EIMvWClHY<7uqP<P<PCd=yTMLC_Z^EO)UdDD9pT
zw%efAO~ZP-VOHd5V+b*l<vGebny#a8p429EP0P@B<atgA5_VlEg-yjZ+$hE1gn5%?
z8LsPb-Z7icIXpUIb-QG}*>H4pL^8f(nNgThJo^05qhcfm&qPNa76v~+{_YX^yywT4
zD^`u8Zq__|(evu{HD`xMoE^?N`CvxfHN3dHVYOLPRW*Zmn9(qH$5nDM(it!cv<Dj_
zX>{mb{M>@k=opy9fayI)iv=+RiqznTj;`KvJ+!>OyyW?pUy?;dmgb~cfi)7BG@VSy
zGfNyZRAUfRC&^9Suw7k4Uy(78A_(lNQ><Ihrf=DJ&pJY_p;lwWkcvz)!zT}p7;f*#
zL&Wuss6bWZ=rHG_kKZFNGpcHZbB-b{My{7aYbl082!yV~4+E|rM#E2;x6V1NwemgA
zqkJUM3WFxk3bG_cr5SZMFjNf`QPKlqkb_MWpjw7}-MdH_Phr6aF3w-kH<2vOg+Ol%
zXe%*N<4+DlPhD-fy}9A$_Eym25iyEUFbr!g)9I9Vp1wnoO(^mLtws5rSc_3$RKV)7
zO%J42vzW=oFfB@Kq9HnlKH?PnwuBJiEomaB<`AVY9_voDRwBzEPux4cT8O(gqaWLs
zqtGBk+#Bn~*T>g4Z^dseW_}-K-HzMy*5i(xzc=kcnyTXJ;)2&N&$+$1rP^-j2RCxV
z?w|Qi4I4&op-m-@53NYOr+1>hbIws$HFx$7AN|O^^TKCLt+cjBLi<iY@omP#hYxx0
zy?1&1_z@ul?(S}R{rZB<=I!iBr4*BC&L954A98wn#%wmn`M}Z9DQ(k0^gMWQ%zGbv
z!13`ZhX=<j4o_JuPB}d}<M{Z9ywFtJDMhZCOcH$P`0B-1;~}JnC^^F$vq5PnCONa|
zT$KEw=l1G~*RNmj{-d{_o0M1q{g0Rx+tP5(3!d{iFZq(I_>yb?oFpHZV35Y8_)}^$
zJ(hvMfFjtG5R}kUMygN@9vc+S$$o|ci=t2od6MEXMaVQ^nsRhJr*9(H7uQ@~zve_I
zFg4i3GB{C0o6(ITwFyH%uv%}pIKRLEPaZwy>cs^we({3pvgfNWUJAdloKO}eX_}+8
zV|WX#mAyU7H1EInn6@3Lo0hBF<#-THDDs?(*XKCb(sjQ%D|Y8^y4h^SyU<A|qU!}<
zFj@}QAVM7BT0@>E<ar@Jrjsd;9zEvKqem1)L0OhOdFLJ8efpHe;UVQ@3K4E^Z@9j>
z;`;iEmoHz@H7&u{6h$UhUGKa{j0&v`<zz-tP6$EK_l`VEa2~AHSPOYNHdRJpSd=Cg
z)=G*5K+`r{US06w<qQ7m$N!ms|9}4<c=qxISJ#(Z-(2B`{(do1aCCOUlc!HPJ$u0M
z$ti;i{PZV3;r#qH&p!VGe^-+)GV)o@<<&Vix0kq8AhZ!2aLA77_N~WvL+nO)a=&ct
zi@7&W%W(&rj|CsQjcM2Rk=BAS20Qw6!~os(io4`X`uUXfXhAV}(nQJkYtX59)9I*5
zlmRCQkoOLy4M{p094m#gvN47*2Cv76Qi*u3MQJ^X$!f$V3DJc8f-Y#IWH+Wo&ieX>
z+!`i%iMLjaWR1aigYSBhB1c6S`d;E$MpNFzCK#n@>xK{&iA{+<V2!~evh^a%IXye0
zs~cA96^nxf<zylgT}7I96mlp{mSm^^c5pl_rxZ3NzrCRv1}=$|ndY|jbYZ{;&8ych
zxH^Z!Nx^h6<K0J32#+3%#ldRH-Ev7k^!N}YkJ5Xbk<C63+0hl_c*7Y$#XvteW{JgW
z!{N~(Wl@q81w+^H`q>4WrY5tR0*wleuvyUz18s~9lY;4NicJK09h_oY*K~s$+fOal
zM06Z*T}#^pRxxKyN*_}OJ>ZSpafyO7It~(edT>OnJ)89kZ4}0JWWa-igdco#!XoV%
zHf!2?1q8Buw8(J|JGRWMwSxWzIRr!>_U(XnJW+-qo`|s@m>(`ErxVgV7lUQ*v0=bh
zTlC>TP)1S%;&{~LUQ7Xsq3ftNTf7eh9AAC@Iei;Ql0=k)Nsdi(QHWU~G55WrZ8{-R
z#}Oru5jThr*-jNY3<Jl<C!_jS+JXCzu^2W=BKTsIEZ#|3FrQ7Cm4zI3c5K=Zgu1>f
zRQ;$LRz?I5W7-6W0y{MAm_FfX+Lo&CsoIv+YQ^2%9XB^OtTr25j9)9nezlBy1NZ;d
z%hT^%-bICqBSxz3f1TDM1nUMzT{o<jYc4LYIKQ}Hy<F2Zt>j4$0fiUuHs#Sy<Fu!@
zJ^heK&j=;0sNhC(btzc8x<cc@i-+0H7gAATbV7`ph#>|gIkDejlx4x<V9xQ$5lLeB
z;KO(5`tDZ>J7W!xpFHOH?2N<XQ)W{+xJkaPl9;+Dk9qp^T^5T2Cet~S*@Cj1GoR0e
z9G_{DG@?yJS;M-jX}u>X!{8(R;ON~z*luusAkR~@k~XzcnzkRfetk*(HAHcT*hx_2
zt~(UjkqO{!4DVObyf<hOZ~fHQzboSDxZ@R9yy6Rf!3|ezxTDc)V$x%=XeE_d96y9y
zA>MX=hl4AFHHzMep}h|fm4lc{VvCBB1*;;I0euX~dntAknYk&ON)ex5v%ET&=uGcP
zH6+SNM(t?e-!?538<J?qqUPx6fYYNB&dyFK4|3{d%j>T$uu(C|Qs$EhX;NY^pi&|R
z6*NReP@WJ6T<D1=kQSQblL@JH#OMgI##7TaTb%ptLZuiZeb=C1<QPY?#ExrRPL|~q
zd4_YI7&X?W93LO^?z`{t@X;e4J$%H|r|)obe8Oy6P);&2V@~oh1w+#`9WjU(U|)N-
zla8gk9q+#krAmy+_Xnc&ddvCiS3=J+ke6qq(*!j1G1B;fP<4c1AUIEOjx@3KLrdLm
zxw*UGC!hTtKmFxT`TWZ-_}MQ$<NW-Jw(02m9^W~%Ntw;3%%(FAjt)3DI6xaswchaT
zt1nrt*K}=5*LL`}0UPP79#=c+SdXa~NXtecM%DkIh24&1tM>cTSkS%Est{mwU)+V!
zhQP;2P&-waIDo1#mPb2Jrl7vKB(CmACzeb_Qey~-R29w*BU%y|h92d?q-HOe@Ct!&
zQ^{OxyAGQqlE;U%ztSwD_kpf$<vr61AA@+XjXb`=xsgs4$nqSUrnFs8VkD<^2!2m^
z@0ymx%86^e*)X5YBq}Bq`;#;;NYfNI^r)!F(~Q3Dh(Mm_bX`ZDraXA~kmt{yaeH?M
zql&g0dPMarNYGuq#yKIn+B{{7<^9DGW!aMk#b1AZO-7(Lp0#fotfBXgo2uh_b4Pl9
z&f(z^N5@B;e)JJ2SglrE-CVO=t*E+=5WPGmV#g5!m>sEg2qe}>P;~5heSXenbw^PY
z3?VR`FUTgPL|}QCX-%#T$~lq<$;jo1D&m4;a2=f+sM>~Q)i5Ya>perebMcprzm9=k
zIWdeKi;E;i4in9>H5?@=w(8haEg6B}919B{oMwFTWX^FhpxZ5d*D?&AVlvtDGu+@n
z-}5=FwYYu|)8OFI)?%z3twH2~fb)dTQ?Dy}=a?KQCaxn-O3L5~+a<==D5HpRG~JHU
zhA>*<?}(A17ZXh998q~*JpYP%(~R~pIo4(<o0FskX_m@d)`q_K5DbU&V|v$9H5=Kc
zLQUIt#0ah*sJ9hWwWTaef({H_PjF7qGv~%Qu-&2TjUrA-Dn%GQ*+PuOVccBD#Zi2l
zgd-zF>;doYZEKpQp>7)5u4P>{Z0l-_>*>Zt{Ocv#uitF<3lHC>Wc%GQI~5p}|M$=6
z&EM+1r|X39f45w*T&}1#714)1VjsM;xV;Z(y+iprnoa`jwbG-Nj@a3RQSjxxL)CFV
z7u#Y&o@OLTCf6#eam`e}mAJkeHnn6$N--%Y^OTdbV;1M<B-v<;JHAkxXy%IrML8kO
z3XJ4gF`drDboyYylP6Dk@Zd4C*_@)BkmV&=mXqh{h$?EbykHU*IPZD*<XvvoOF6g=
zg4j6M3ola%{a{i~P<e?>Qi2O?R~v?3v0K@dc5z=$?SKMlc<&0mug{@fssE-YgXNnD
zRJ+)+H8-5|CFgv}6<@MtLqp9#L(~IW63uzDu#5XY_Z=%xSfv?u-H8}*+Ayg5`$|wD
zPe>fhz&^037oM^%nH+ZH^NjYgVY9v!4qTpMvQl1g&_rzn6sdc*s={JOtRYQOX0xfN
zG?y*Sw&UvZmao2g!NUg+I6cZptYD7foFG+$m>oy_$T<&0jERKUiTO4iX|AbO_ngbG
zk6rxL&fQ=-oidrsBvUv`nNDZS=X3JBAk8wYF|_S~_YtivM@J`o^zp|$ee#sYk00^i
z>;aR>l(I}oGbvYm*ipJ7)}(AF70tAzZwH#Xr|Wy+E-EwLH*&cnk$&h{u5QRDDOqlq
z&Qj)PB0hoM)Ak*q?HRg`p=)WIn!aziU0!p2d&RR?U-08!{4HNT|BBZa=UiOg;JYAg
zZDlDZ6Uxbq*<?yl6l7^ZwUM~Ux~{ldUb3yW;N$%?)DZeW7y|pIgneU*I==JL_V?o0
z!JFj9F<R`6v77MWjTY=~@s^@c5UHTVo5-N?q385q&XeN>lLS&F+#V3+P8>QyH=u$+
z2PG$rVZ4>pD2BChjJA=5RS*kMT8lSQKmxb33}qZHim_ubN}Tz3ZP`5$4a&)cX1%6q
z8`3O;QE5wz7&Os&DL}`COHdJo8zhD*$x^ftm1W;`_gtJ&B{XzBMOpCl=~FH*FENQ_
zwpb8j<k?rxxx3lWG(FKv!i2G!Bs0tw6Q+xjqdeiyK75DAvk8Cs%h!B%Rg-E<Z8Q~%
zU=7X%7<z8bUvYbV!E8R~;P{Ax!vh{X_`Ynw+ls5ZTb8Re+f6l^=Z@E!C}zEb+pGK%
zD>qluv<xB8`$$}`sJa%{d1gv8Gn$E3<VsN(Lt-@1XdIDFMegw2R4v=yQw=a^Lmwe3
zLof!SCu1OsJw7NB4Mk>|B@>dirwo>?H<;)ttRf}waGvwg*@P#D2{Yr+!y1MThJev2
zMNtTs(K)&HBlp8vi!}yg7z;a55i=fQhM~uzAw&|BkR>@@YoY;b6}bk#x+ScxDBL@k
zMByp$A&R%7X0%(0V-R}7X0sHpMIGpxn!ara&e7C{7%jb5L=r)dvz#m|No+w>lKY4<
zWVWDBJBGpGdr?&Seqecf%c~@#+Ef^AXzG^Bi%Zt)EmhMq_+9KDv_sG3)iv|!oNB$r
z;TWRF4N|O$IjC43xWS7`*NZY!qNs-Z1cO1$z=wVq73ufs39#?ler^1|xN*OS-?Klb
zju(DUa+m-0KI{>%yXA`YW=q$!SY^obabq8gUKOMTEhU2>PEz0-lu`_%*WV6ljnR(?
zsUp#qZr0G36H@00BjJ3P=<{tY`%I!FS}?WbWzON@0f)y^W{ZT|OCjzU3u$i1OUZMM
zAgbFzT5!|pl*7Yg4i8S5PUlP}GxB1x=R!plMc5VuiO=$q*#SIy_dVA2R{S+1bX`N&
zHMpV2yMcPUWm~Mp3Pj5xsHv(^kn{D6GOpF^jJ4lJ=+ww+d$Yv5M{qs&JLcQLVDcSu
zg<NcS#TR_Z&$v?0sJLNZprfH@%#>2axWDQBPdXHWvC6e33bdh>5^9U`Vsx&+8(CXo
zh$4uJJx+rz4au2>Sx<hLusEG?{AkY0%Z6=r3nq}Kmh|WWiOJ>o9kdAgx`_8JT`%32
zNs*G}C5OkyTs*&q2zPfIn!o#;M~}{!O%F-4oFp-*K45h+#y@(ZibNHJ58V&Ak>-I?
zL>z|)6RNv4-5Yj9yJ4;6=;(-|5KE-8EIBwh<nZu_)6>(D7Vtg~A3oX_fOTDSeY2#g
zTZTan5>KAK!`bNxM~6pDr={%6S~0jdQb|Brh>1p#F`v)y&Z9@lMAOyOO@#}RHj#Rb
zu^MdxD95(C<MtMeb>z7s4o71w;6T?kc!Ib+RBNtpFZul0=Y09>OP;@a&SzhJ#`(=T
zL+{4+u?S+cw&G=Xc+Bb184AVF58Pbe@YR=J5GyBBaTgQk#%Y-Tnkl-QrUBw+{QmDS
zzI!W{H^dX8y6&6Sn7F@rzJ9V{oIDUlG-$WMJUBgKR$5|giCKaV9zS#t9HH${&J&{{
zDvMHvq3K6;k*s$nO+;J~q<t8DAnFmkA0w4iVlavL*boH4h{+gfw)55+i@~;Q38Syp
zcDtoOB0`LOgA|<3i#%E>{NSX0GEOi??=5RBZj2q$N(qc?z($CJL`6+!vyp=o_u`Z7
zdQIPUG<8SY3Cjd!$V<hhshDgnv%?uC&v-h^Ir-nb&yz3D`RgxNT=qRhnsD1XvaG=Q
z0gMq-(B%!Q^@^7-4aH=_d@<+Uci-jdlSfoljrX40<&szDVrSCSE!K!-in5~U+^I>4
zkl2K#9~cNkttAh!^K@R2SYx!PUX7MExr%rcwSujSbWzdm&beAdFUm)Z21_I}Q04}U
zLJa|$EeA$%npvg>vf!Dffu|1+d2(DZu?`y=hPIJ86Cyrp%E?4bijBrQPnx8A^`+M0
ztw+=heNXUSFy&E=*47P~vG_q`{x;1?XBoCg>0C>yw&=d$>cwYFKYb4}MkSayxbNB#
zMhi-<lqa$Gd8kAoWNRHo4K*6whUoFm;FQPeNMbBWQlPZPds%pmO2}+Vue;G+#B+6d
z!)CoDO*4$Kco*1gD>j{{AAKc943)NbJb(F`rfNv6rFT68!iDqBkEooJSf4QFpt75%
z<KiCgMN@k`WcSkESRj6VJ@Wf1?tWiGo{t}eSrR4Wsj8awYQx>l4a?<{q3;B}*b#6-
zAm|`6bR=ma2Dy$PZ^=$v5ktiF!aa;3qKsfcS~=3br)zuqUasdpVl7AlMf{d_Ct7KY
zHQ3aWBq_7`jN_9N&K^8q+dLyWDDs@c#ezIb#l{Jba~-a4nUp0qF=SanQKqD$!SGHU
zYpi5|M-}N~PXgpwN}45PrA3>-bXGE(PccR;UN);0UDr|74eI7j4y|5MmJ@o{|4O2W
z?1XxR3H*q_jp*f`cPF#%SFyNz<?-9r`S)X>W6P4~{E{DU{v+%BHAA{Z6UEv{j7!K>
zF>>qV@xmK_q;U5rB8d{AL_`qVs3dl72%gZ6799e2<A(vCh@pMd5jz_&eZ|~+K2%4%
z`tpLf_H1u&_^eyv@Ek85kS9|HW${DAS~2RcT+gO`MLCh~(xVSQ<flLTIkbWfwwr<f
z+n@hW+*TEze)=h&eE2c*#f-M<aJ|?-MV#zw!4cv>2!7vo4G}*4=zY#kPgpLOT;JSq
z@%k1b<XO(i(Fy<d-~R9USO4l?@%XWbX^gStMIpSoyrjqrd<e2m^^UG<IeoBYd3Vdr
z%{AB87yRsJOD->uI6FJz@#Du#r!%rVB~2}1<mH)YFjiqwm>(9HY#^OFe6nHF=cr{v
zRrmD1rfO@_A{Po!Aw_H5uerV%=$j3jbV1_95PsRN_}dqs@ypkr^X&2^&n{ncQ^|ZC
z`T^HFlv5<CAjuNSBIn@rfIKfzN>Nu0fAiD-jJxfL?HC&P=F+wS7C<&m%?c4!1MJL2
zEceL{03G*5-?tp!Sv)aPMD{iXN~zHXWjF5ID!cR3{nw?a3vU!-0R>u9PD|c-=MmO|
z)e5hpkdEDeaSq+PF=9z9Q=);;J4T~>+<1cMwZ^0#o27E-4hp9u$xar%Q}L>Zjv8eY
z>7<mekB4-v)wpqbX{n|OUEP2Z@dHkJ;Ldy8Fp!2!Yz#&d&cs;S(N$ZAFrMDDw0^Y|
zIs;<dX_Y3+Gu$vx6a_&E-vtHpX+a!3&POKm3A2ME4o?m#CkcbwqKurd*4s7J@{%OA
z{NPc+$u!|_U#xghIczX&`}<Q@l4lIw;SqD?s%f~ryXDJgFF0H*c<}Hc4<9|?;P{Yt
zK6r=CcFXPU9Zgfy)D1b1MLCTO{UFP0lCWxf>cI)3sATaglp;?O5&{VkI|=~?A8;XZ
z(+4&~#6>yXY9dJxa&2l9S)wud4Ci5JTf#6<1bCV%-ap7W$s!ZuIGYumEpn`J=(xcI
z;a<8B@G7ATVp2UV@7u6OOU|F#an;88ZAOKs_=Sx~o;6g%fc9}z@d|Q<8&EM3;;o7t
zO>*ks@$DUDa>V7TCNF1lf>-yvDy=kmo^p0}Op+u-mGJS0AM(X#U$R-R@zI0QBsL+@
z88#`&@(Gi2K{+`jO$u?HZ!2^qYOO3y#skP^Ebbgl(^A)s6pvBU4;~k_*f8wWn4=d|
z)%LXKm+Z=s-8!PiTloGK-`(FMKM20DCW~MD`qxEPzcPMPC-&Do#y`l2?Ax3n3Uy%j
z;Qi21Hx<=(%VxW!scI?O)y^GYmoFR$p&tcZR_^Cr`)Cw+DMal*c(G@)#!^m85}WQV
zN|H3kCh1#+VAy|ryTUHSh&F~i&zUZ!Jbd(kNMJgh^8N=OvRtkN?Jx$L8V1)BThOY-
zwLMlt(^S+=McXxkPigV-%!))zY>fsbb6foU!01S-42MStINzgnr0Xl1u0b2kYI%ou
z15MYkuD6l|lvpyAeZw5_MDA%KyP|IF=djD^xW9yP|F^e`x?K^t10y8u^@8WT?7!sa
z>%ZnkeMQV#6dp~Wrz7A+<%#gyq!%T0BR3Xy?2k&4aUKW^aYW~$>}7;Vw2W$JQOz4G
z2ehWCscyGab&FeF;#-Bw6y`BF3)B!=&+FG;GPuBecEYqgLaU4z6@DaydJo;EMd?U1
zIcM*^%iXJUhN=SxZXNl_U;muVwx%6Ae)PvbLghji>D!imj571y2^~c<de8~rq0NLW
zwan%d9y~bV<Bxj!-lKKObaKGcci-di=!C_=A@jun*5+b3pQa>9I?{|Kt*9t-h$UrJ
zQWP0wnPN=j^74|~+iSL)73<X<XAjRfSR62&PDw{fiYVbD&O6rClJ)kE<>r=)>kBsB
zEkkTTJAxP6Gr4Chj0a~=$F_4+?TRmMf6RKdV!6EK)zxdBU7xdTH&lH~>m6P1#HuDL
zrl#azx?qx(WJ!ie6<6&I*EiSn+a7=C8Jdph!mqAb3b1f|Hlytu`pU5fr*~%##jh4|
z|In@2K*Ww4v{fUaTFIn%Q?aI$K+#d{T<vy~USWeIKxid-lwJkUJ(Eese43zQj|!2{
z_C)82{UGw_K`cvr@Swv;FqS_0j%gS}!1sf+Nfl@<W`G_!?`fq_M&qN9S5eZYFjiEw
zG2Scmf)@$6C?pdSW9aIZGA|g~p2UbiV;IDs(`rNL9cr2rZGvw*d_PDbNRT83r3rl`
z44%PvcsG!Y>Pusdyh~9+hQ0TE^u3R$whe9X!DJ|7Fp0$^kj8>xXlaLbOiY+S@YHpS
zTW$E@pkQJ(?}UV<w>;YpTx~YQXoNeKBxs|@;@RW-j=PIXt}ZV5o1Y}ij;5SEIOFu>
zgm>S4huoxS7m3l+cfDkbb_31_e1OFRN8NTK%Hy!nLs!$|>Bh`W!Av#1j|?6X22#}l
zfwC+yR%3j?dQTdF;JK<4o8B>rjuYUMiQ<nQP5A!9BOc8sBry_)20v7U(BpkiH*|ui
zDnpEdq-0saG%LwWMsNXR1;GeD5S=IxcbI^@FA=<mY7#B#WC?%nv8iS{pI{1$B_Y&~
zMzw?_rA!R%#VZaD52%(7YoFr1#h45g4LU|t6twQ(U@k83I^omr|1sCMJ(sr)gYzT2
zFL{P40h3yaNy%(6Bg=}BdJqU|f=MFITQO3N9l8@wyBNiCL`*-UAg1F2CG&h&9E`D8
zywQmq$8U5H`Az@!20`uXmLdQDF!r8x`#yv4>%8;0Ac-`)0!`x9tRVF<3gQv(6E-A<
zN8_EsiCa6-81c5!a?r^0f}^7YvNV_0dzwhxVkZB;OK*v9a38<T*hRr*X-YXMIpvfj
zO_|N-Jbbj~;_`y^YC~PsIM=gU-;Hak!nGa2_hWlCW4$?~oKECEO*Cnql4N>Rmjg=X
zS{w_0WfXatb8xUA1jn{ov0B~Xy%W#Fwb*KOLr>ec^sbkj?)Y}mx|2@amu`ChRrU>7
zMh@OC<W0V^o1|)A-0kML{LUrUJa0bd%lhYhwfzNQzQO_~3comTkF<udsNAh(JFM=W
z!}oQTpFEOQ0v3pJeP=u!Mn{K$A4Cy-eY<448R)j2P-#@3FnMIyI**UQ9klC;rM(iv
zVl-uXAPU)F#&)Ur9BLiWMss}n2pynau4&daL({Ro-SSeu#wf#NHs$pA7_GE)xAyaO
zeAm?IKruRD7-LASA<qNemm}X+VQtQ&IAAuN;=H428<ILD%kKBpoiGul(Ww~)i<V>v
zWLZjHWMp|th@PshSbHxA4dv<kmZqJPrRi96h)H<gcWj##+h)zSS+m(L@nOIiNE7J}
zDJ9(!hs63-O+(jrBWf&yi1l{Gdb{TK?uK=>rT2dHBGn`&#U`mxaEv@hrD33_sRuT9
zYpPYnPz{2>{iYGYw}IG0?6_Y;cb`3PA9BCGAb-p8-4HwFu}YI0O=2usk0@8@Al&<}
zDJ8d?G~5Cz-k)-kpvin9P_@x0<1nV<@xvo>+oL@UZIAW>Ypc;ETZr9)dMP#LcWUrE
zf;YPpqb9hBYl3iDk_2rNeD4TGp^U;KF<~aLB;$!QjJ{I)u(T*9tM%x+mnSK=%QbO+
zfH-8!Bvsnr22Y;k_`#u+g(O93gYO4i-=TCs*<@_Pi842}MmSodM_)e-q<KbWQ?L?2
z=24`x0z29)IM-sWaIP|&(GMEcX_C~@YAb5o%yM*KSgkfZ$s!^4)Jii+3TA5H<+i0&
z(2s%esn(=cqbxK!h=3t_>g9?rZdZH}KO;4kNmek)C#cktmO0b;oY{PV73XM9iy1@j
zASed!vAIRzP--AcHC<aVv^_B>j8{y9;{C2=)7EUdhIZ$5<vg)*s4<ANSDIYGl*p$?
zmOp#1;Csh24iilh9bFSf+mB{cNGqHhPjp&gbYz+*%qJPSNwFA);K;KAYZHbz5LDQo
z=Oryd%V{;rhL9yGRx7$@ON;{^Pup$DLPFw7G&z(OkH&Q^$=wa{^>d1RN@99K%!n}s
z?=jH<13CuMal?*iOfqBL7_K(ZJB85!YX+?G*siDX4u@t?P>d#>UEk7n6+R5013Cen
zC``~ql`!;vyf?ILKI7h|#zpb#Rojm0@mp2_zST<eX8ZNq6e0gVQuoDp|9tm_-MG+(
z5J0IB-AYK(lo;dym?T1~e>2xskEl$DYHZ6(8{a2#5AW<@ilXFTamb^G4=IXV4Aixz
z$TRY?9NX!Fu*T@ek8g+A*-?d<rYY0Ol*Meu;n5Kf9z39`Y7So>^6KR)UcP+Ai<d9u
zVCcnu!gU?h_LRQwh$>J{Gf>bCg6u>ciDph-WF*Fp@oCZxP(aX;G)tMzXD9=MZ>gGU
zUn;l5K(noI-q8&mecxiob@<I8M1F@+_bOD7gUQ#+xO=Xhs8r<}?+ZGBpdud65L=$d
z&-mH$ulQ>7b1r#KX#y6FjfMn?q7=8Eh{KU0t!mM-?+QCVJ{LyBd*sw1UYc5qx%RzN
zgtXS%u0e;$&<|{@itC#d*S9ci!EYfp4G*8@!lCN|+yX6ZZ?7eDd+0ekiDYSk(L#@j
z(TbBn02|===pl>goaOB`7cXD3esxXok>&Y{&GnYfwfyjp|B%z;6Q-pgVuKaxhS$3Y
zz#T<NL_IO2I4@lRuN`?lVN%XnOb<y@!)kepb3L2whUs*{q%8J`o$OyJA!LM*pp6}E
zP+B_IGWd?xHB{|Z{9(PUZFN&InNCQOWY3HC-bse+P~&`u3q5_`VGu)Ol*MUF*R>2y
z&)^)}y5i;eORjHjxVgLI^7eu|mUu0;UMfPOG)jYUn&=|)GN+i7BRAjBg@IMQq~6w4
zw-u~_p5MNB3dGCu`8QO5zv*^1im$arf9v>ekzy*6T1X67^#-6Fg->)pzFvmPkVAmN
zjd3Ab$%d0>3GY39LTWTa+d$i6<48}9f}E%*P61kt*|b^-XGtlHRwzHJZ=FXa()JpH
z6N?k{-sK*I<Z6s1F%keD5aZ8Xd{5A0Q7|G~ecLk0a)zO2vtBcuPHEeA-`11(6)Wuo
zEgZTSB17L}T0>$r#>#|@<Ea*e+cI?hUMU%UpbtGNh)J?B$-eEPh91|p)Ll=m#PUUH
zOO};nfHlIeGNwhFgrdlaF_NYk<?N96H$6Xn5jnVte7@|t86?@Fb`bMIlv{a1q6|e;
zSY<#f2B7keo9%{Io3$X7QPANu%dplmnNDR<uo;uGMC$}SSR^mAW1|8gItK4)`#|SJ
zC0A|lsJAt~?@`Xlb>7XC#G<sJ92znLKYB9dUw!(JPY+Y(VIZ^(Rb5fn1J3n?&=X@E
z+g3vA*4mJ!3A5RZ>2!v|5PZNGJDOhJ=W*^@)ruXuqzECBYCUG7CR7`-r%00on<bz$
z##)N9M5QTB*Mj$~udeBn4FAqM_`U^Iib^X;!h=?0&h8rp!^QPAFJ7Nhb%Pw*yhjgX
zE!HU3Rn6^kL!MudrDC2Oq9=@zKT3Q0p{DCPyo;mYA&l)iB$_8iyt=nC`KR%#W1nfu
z4z&|>$48IWTKH)51@WOEyj|1c2ZtX#bzRf7?MOOzGS{pGo0~*hbXk^>Wd&JYkQW7I
znF>$mV9w*mPngc8C~XK%_-IL<?5P?1=M{we_-%$7i;|!M`K07vu|R9h;2g`lJ9_81
zUEWc36@T@&e@(xRZ@JuGU!QZgx?;V(qw2PtUtMx=c+BGPkh6zpJbC(rlaoUZ4(613
z4oXPkVOMVHKvEcH$rPKuLyVEhbi(5B0JP@ii|6!RPu(<Rd4?zaN-pLQJY5_pZNj}Q
z!V@vD$C-D%tlc9AX&WQTQ6h4mD9LK=sJY{UpRWEBKVSc6F8k+%ql#X8+}2|-OtU%j
z$wI6_0FcC-Fe+m`o?&0ay-8NsW!q}uO}TL%Ddey^bYc?h2i&mBtPZFIHUoGM44`UQ
z&IatECaV<PxtFNa28M3S-Kt`_dd=g9kC@FC6!{dz0)-&1;t>j#gyZ=shs7bo)0(TR
zbC$QatZrBQ>@Pm!`R6Zr^7J8}eEc!r|LA*^#S~)(s2iH57K3q{kY@K$TiN(prSLP9
zdBJ3wF_{`BQ$v<`RH&(|j=Ey+!5Bf%#t`@CN;mXW&6dq}&2qJ*YZ^(}NE1{FRkx*I
zb}YAdXsv`gpm(O|DItP28HveAC|KSu*={zhS8LACFE~Gc#b&!<TW`5p-EeocBDnFI
zCBPiW8I`EAe^1MjNm-ER8Jm8~GA?QRo+gZoasnKTp98{!8<6YsyQlr#4-5H+N0`Y5
zsUE-ek9~*7(4N#r5*@}<(b!J?6;G;f9uWviZjUT6qBsi!i}{2yOX-^i-S?O<auUYA
zzKZuZ%RUTRuw6h-(fgu61(Zd2A0yfp+ZK!=8iiMZXr)aMLO^+sSD=->cZF9PWh`h%
znx<^F8=R3ef)K>k;7!xgUE5~YijiLVXqfC}Lr}or22tOl@OW&J(6^m%@kWXGFbsko
zWeVp;G$73}X^zcH$y>Dv&J8q8BMX*-%w`m6!6Ywfnwr&W$+oKSeWak{4^J}QJ^Fy3
zUEFc{>Y8V(fs2h}*+#TV2{DkaC8yRa#b6@NK-7xVKxSkyYrO+2Xsc6^zHjL6TCT3n
z<+Un>!J=b^(MZcYN{*Qa28Eb-s(@i2b&)ba0TNG#M4dItQ+vmmiu`XsT=1uV@Bxp?
zgtBXJRZFwo(DW_c7^{_NLt-r<NIXtaloUk)K$>RbHPB>5K{t|tch`ko+f7Mf&l@~D
zVDE^&yUspF+P)`CQo#jc0Ov&5;5~Qi6)wBOJbXy#J5(xM4<e%sA4$E4f=Uj~@pr%Y
zIhV^NUWpPO-e9wZNbEe@Zi`xP_r{(IM$U*roBPEwME08*`%fdyAimjx{ipHu5xk%&
zyH*stXfU)Kk|yIE*CI&h>XxQ%*sM3SZO!s-$#S`5v)zbKrcorRA<fd!{$WB{PMMUm
z@qlN@^I?pK69$)6a$uCmC$UJ0euo6cabx^;*KqeTlwzL}#3%>~K<9ck)t2SmihdJ_
zU#l1EZeU$+D7H70%L#$Nw%KyGxn)x=ajv23AL0BNM@I{iMCdMBOLs(Gp+=horKcR7
zoJh;m7}i_i?%mzq(zP8;*U&qOXnJ$(>_YC*;$0I~^CpN8N6bu=v(gE%NMv7B;KnFG
zM?=jGFZl&uEq}(HJEztwA}#cuz8=tC6FnqZN@g;ww)>(|a+AjFa>84b2I0P2JdO_O
zwD{n}gqpZtueA~6*9~G;?}8WgFyl)K5FKnUJV|1SC2)kGcJih|9*NtIt5>hsPHveT
zOgK9H0ILeDvV<r?Edv;3h00^$^ymTea!z|J>e1FU++MBt%k|%I@$!a;j~_CbOvut4
zrIRs&E}(Q6)vWhXe&R!DD9VDo%qWVK$s{99Q;F6a^PojdFSI4!5A-7sco+tpcl2H;
z4oRMjdCX%xpAlM#8e1w#YmAkvJ$O&ut=X)$)OAZ!%l&up>VoxZ$!5Lc_V$*#uJ+eq
zaPIz4WPl88nv&)jNtR#|EdnAFSW$CX-Ql7qSWifzs91N()?vI>2~ya-DR}r@$1KNA
zGs;=c?CA+<(en>h*cpQu$P!Q?jxf0RZ~M0pyI(&7;%66hAbQ87NcjFIA5o-+Q1vhj
zsGx|!$)bv4r-IsX2;#UyX){i+7^H~MdJGc}ax=+av8Yj~5pmIDbpm66cO(*bA>GQ{
z+cJ#m*q!ll)3y{v&M-LIRw8fex+X~yL0`t=48~TCwP^2zALtw^No0do;*saacFK;c
z)3q&q-;-t;MVf(+cpo4!7-jBx0xpo|38}V%!i<K!64I{6gaj|~GDXUIy<&5B&B5^z
z4`&NzpS;V_(J}w8pMJrkgA`pH@bY3wSNG&yM%Q<&L(hhRCU{xgq9O}2Sz4v&LZt0H
z#%i)G!4dE_qEn64N=$1*L<gXc7_Tsl<Z(dgAc_Jh8YfxGEXzo>rfO@pU61M`59fw|
z_vtBr_TeLnpy_Yc)Kx>XS<>`1!AMaa1EgBihGkg-kR%E7`CM#G#%nsC&%qeF?I2!+
z#_Uzy_sY(IR&xCuLUz|mktV|J@LmGU%lQ<8mNq_OjZ!2TgAe4&vRSP-@IBgfMC)+D
zA%cbbBart}h&ku@{P{D6;D4pC+mAOmYr<(Aw=_R~@r_#Ko4a&x{DOZPzjo{nYpC7Y
za{q7ENIalHW3?s85}K-Evs%-4J$=`6d%NWJ=8n3liGa~kP$j9A<bc89oE(a6nqW;w
z)3j7o!?vnP^O7`W<RBwfMPqWo&RV7)hF?Loe(mdtVT>UYUYqld=wavwnzmuP-LhJ(
zSglsXZ?II*CJLu{G_M_8N81~yBibrvcO}Inr<`PzlLT#3tg@iTgZmB%S7Q1;nayw^
z5F)JB8<wj(LX3hQ4g;>O>4$H~Pm1xa*6Y16Z->t98lhqQ|FCcQip7>3+&mpMOK#jN
zKHvT=FIJziRyTA>h3mykcW4K6G^9z6i-VXFkAKGWX4J^Fdy~h<{uz0E^f-nwD|~!D
zV@D_|z5mV-{QkKr^Ts_r{#xIHQUPTY_CVv(0QLCeJrDwYYzbk&(Syk-@&htkklT!O
zG&7G;jp;XvEG@);H807Eg1W8fhK|8IZkKCJn$k2qc~%ImYCPD6s8HIHi~#^zn>{ti
zm>tK|5u(Ef1xoxtV~AtxRt_+O>t!xG$1wB^K?HPBF;WzaaGWFiLv-wVVien@rfJ9D
zCHc3j)rP8SskSxiyCuupJDO@sTUXR|h4=1WJ7YmjM~+`+uv3e&8k1O5VhA=8tj8HA
zRyVGdYmv&&l8o1<2O3~KejfPFmLy*v%4lqvlTUK=JV#FwlETodX3zEfw&S~T?2K52
zgnp2goM!Z?x-V3B(MG`Dp9y4g#4)Nz8HpRp6GLzf(RVCwY6|TzE{N)7Jl#Otw^8Hx
zvqISP1TSjW7{<lVIg&z9wdf;53}QH{EUpKy0w%TWZ0h_l2uc&w-Um+#K;id^l;4nK
z8O^pLqG+0iq9~ZnW@E0M*|&JLF}UERg=-D2YY72-?=acu*ze_i(RQ?7QHr7{NKL|K
zBj%imwIpdq7(B)%Qc#Rs!d?3pq*ZNAG6I_sr4?P<h;j(w;w<KKhPvT)T`{*g#o?4s
zAD=NR3jXpZzhK>7^Zg&bPt$oW@77#hUUASxG?t<7MQ{?MNYh<l(3&O&Y7MJlz!909
z9x*$Z<9yHZ?v{&HMXMCK)l72Bqs0jmWf-<wV&9^}KyD0&>4N!mf=W~2+ErUT9<AWh
zlNo>dqmTH}<0)}@&E{r9yY6YLmZq(V29jw)T4;(aqfDi>dVG8WL{_UcX`1chtMa^L
zGMREalDEyyo?>UY69VkiRy)%Bh{7tZNzz2rv0DjzZ>k!FBAF*>YxXufiLp$xf}7P1
zL$xNE*N}85l}S`skf=Ks{GL1Nz31k3b${U7DM<JCz&G5B-*AqK`~SQ-)El4wsi^yH
z!v`;1TeZujReR{vSc4iZa`G&rm=tWYEzt+=u5Vbcw_IP|v084v{#${5Z#t@!5o6!J
zq%|GYwq}{FxxTqWo0PJgkR++B7ej!N%Fwh8bzO@EQQQ3X7o)&=DaQJtqw88sV(~uE
zw%wRObI0A?9m~~fH|_R63s97Xa*|LK8ChnrsS$J-M?2JP>J8g!O<hmux`MQj0z$`#
z59||E6-LzHlM=L0Q-&c>Ron6X4P0Mdv0g9fyY5$B2P%%6+dZdlXM}$rG&VwAZ=ia+
zE|HXi9#6wEUhus81wVfE7kst)Icn-5@w9D=8Vq3wBr4+kAZER?DvfPeDe%-i5$TO5
z8b)THW1z>yVKjY@BY!LG%gqqOsN8q^qRz-tFQ}>C|4Ljdyub9o;^w`eL%Y^rE^mOt
z<9to!K4*4t%4~YT+$=zq((Q^06~w{9Xoc1psg-n_Vmik~c`idgP;DFPszoVHnutO>
z&vQyAZU%XllO~EJvqWnEi}%oX9aXhOn_cFsMk^`Oq@Z*7F)_+{5s%1wATQTNKq-P2
z0gZExzV7Jyp0;bL>v~MDxu&janx>(uD(a@E>l}UO7}^#U#jzj;C*^*kr69|Ii9*jT
z$s{8!GbXc=zH8{cqYE7!J;6FimH3ebq+r`kGKUm%+wtRO{QSEPt*~WA`RJIF$0z6v
zy!JG{;r#9^y5ZX$<aXn`sbXzH8v>SyiD6_O?9N3aJ4$`Sj>G+_O^qBm<>lwCiWmYR
zbfiU!GFr%=#=v-ieU#o(1XNJ__NTP>M3$XuNePIAaX}FlM)X2;wIff($AAi+M9c4S
zZA(xJlVoE{OEzd@G+2wZ3FV}uscQ!3_AOIUyTWF(L2FH#roxpA0gQZ)^u<Qrj`t{S
z1!gu9m1f50FgQ<Q4AZitsw!5uOAZbX$&*|n%G#D;JohDug&5E=l8uTwl;jp_qha(N
z^2!i>WHDV(>6E%|={7YplQ5aVyT?<`{`80Z^yi=P)t7&TEhc>W-uwK)N5^ch?`W56
za&1v#i`xaq?Q+Gc>RET5n<1c_;^fgo{>}gK|G?z<kelT#fAJUpk^YxIqkM48d^%xb
zG9E8Z_`#F+s9wI{%fI?dO4l(#Vf#S4*>YZ2tb$|XB9qkcUwrbAfBnZlU`F8P^Iy_d
zYr59cR-UHqN%NE^?>#1+7^-$nQKT$p3)-d`gQGJxn=PBoRyaMAiFjo7J=@Jz+VtZ{
zti>5k9KFJfej6zZL6os+no52wB0kvp*|lv;KMW`>ZD6A{Wu9YKu)aR0IDUdI9ZGpL
zc8?AQ?`fM(5ZefCKO)gPn!z{RLcrTk?#-|B#=pM>&V4iA^`FK!kHHW0Uuy*hC0n{p
z5_al99mfO-7i3=570;hP-=q28IGi9o)@sru5rrR-wiC;tyqFTBVltVM=OsGP1lJQq
z&&Tp^$?Mmzxi~-P=Jxhmy`FYx*)$c)?V6j_9aSfDu5CNkt2F@nKn1^T-;?Ai$7csL
z-5n6oS`g^OS{93A76*qc4j1IpoV*kQys>ilcD|!(H!L?9`P$+c<Pelt>}cqY(!$d+
zIZA7?qD0#ir4k~JqRhyOlx<V->f-sYv|^QtW7x73l<KWXz9SjP-`MX(BjU-9w;NG7
zwyb#Je#vL;U-NwXGkm@wou;T%qgss)2{DR*N=WN(KTa4qb!x|<lV`Yp^xbzF423F)
z8PuJ4gNE@thC%$$`mQH9KeqW3X0x2)prySXSw<t*jwN^kZQ8>bqP+AnpJ>Owmk6rl
zVN+eP9d5a-QjUu==JtTxlw>v^531=t7A<&9#EO`UK$<$d8%92-A_S=Go~9YNTW+w%
zl4hwyf*OlWQ?k@xw8CmA@|4MBB4UX&CCl>hapVB&T%h+3Hw+BD_%!+$X@_0}bly={
z6?M~4S2f%1mZoXOmaT}vc1P3@0&eIfk9SwpX91OF7@c4fO*u)K9hM{$i!L;w2voje
z-LKhnH>|c-QXKEzznwX77~lH}xI+r*25E)w+Tp*iF@wqJG3DWc>0-vRyJfSwrtcc)
z!v4B`m+_qnyX@eYW_icdBqPMViynI0fm)4=-`86iAPbucqcFk`loLw~14G}E98M^v
zg=9|ZK$@j{F5AxCUi$8`(FS1XI!u7lrnK7{@0~0xMiWG5E($L<NMb^2Mb>SUAazdE
zwtg7)MUv7&HrI#>$RwhG+uW_O`aZPH7&EGKB&^$ykxXjj0_^OUeLvs^he<U?3!3L+
zBux^0n$p(|5k;977^PWl)+}%DI6OKgu_YKinl<-SL&wn6Xdh8Q43m|SG>ilZdQX!C
zYotwNO$OSbOM@SJ)|(Y=z2;ysrzi@(_ud(W^(<E#!p&<On$yDrrtf@!fS4NV!43ge
zx4eFR&db{y4mwZX!jJyhAM?+C^dt0a#<pv)S<cbphdg-qjMGO?2to1JfAN<Tr-%IU
zC+}g`mt6nkZ%Mr;IE5Pmg@(pO7K?)a>W@F-M<2X{ySm`~)eEZ4O8mZtfHn!oj}Ce8
z@QgH1Xu2&b2J$rJV7_3DrmibM2(YWwn#E$lY&Iv$GM0BMnx<iKc(6xW)sAoUMjtBf
z=5Umv+!zUAPm)A9N@JVZ`GD_*8-?m59U~QfNtPyDTwn6&;{hE6r}x^5LR`pkbWJxJ
zS??co2id)G$LuyLA%=hR_%;9j+ls({8ox5GudXS}0)-ky3u+8|P$XF@ytt9b=z7Q1
z)fKN_z2f;-&$&9kqwjuwKSrmJ7a4g`k{5+=KaCXyADXsp*=#F(fVLCfRo_)O*V4O|
zi}Oo9|HUtP`Ro<dUHxkx=grYJE!)k8)$)$3%S*Cs0?2w%jZ~7;(^HI&JbZM5cfHVH
zq9;u(MKL2QN_oy@DXnOeNGr}5lnM+(!*;vo=GM@64M}P-iIuG5v2ACxC9%15ATZ<k
z*Dzlk^5DTEA<(Z^tZpvJGf|GeIU*4k9cJXr$=2f7X<)}55pndllb=!C8=?dTdM@~i
zi~c!lcTIKx6Kh3m7a`LreIqm;5+k-)d*C9lQ<v{ru{)mm*KLsvdfXd(x}jz8Qk3{8
zUUq)8IZ;YF)*=Csc6*jUKY`F7g~SS|A$CHsFca`IWTRh?59p8r#IY9wL@O-6+xncw
zG!$k^nan8l3>5=<%)B*;v}uC}?>un`Bqkv-@}30m8Js*{;gLINx{)4$s7LMSFBtsj
z0lT{*3TqSL{6^slhZtoq^U|?!-s6LalYA5#AvfB6dFS?GrVw`tHtcChF?f{Ll(T|-
zR*+9~X7h<egqny_o}e6!-_q02#g3}3q_B*hU;@mz=gdW<P~OUWPCdT=cifwM4*Xt<
zIz3HDXE~Gk6mp50tA;h3SLe9e;q3rrfcN3-hdJ3<|D7KAI~8^h-hH3ze!w*~i5g+$
zxP!*u*frGtKYV>R(SQz-)M!-j6j_2D<-9?Qc`?=yj1os~M4cFv+4sV&O-S>MAqHHc
zNYMDsbF;c5wU#snRAR72PFfa3t;Rn8KvUPGCK>q-0X-6H!&pE@=g?M)GHXPgnhhC?
zg9UBX;JRUt818s=P1B6Q?a9cu)}u#{7A~K2axzQe*ox8n6HJmOxV96Af9)xY0<AT7
z%O!VLS4<BU6bEy%)Qm*d03i_F04`AV5`p4_s6A!kk1cOyuqF|c-Pn_*7T>i5*K>Q_
zP)?@g(>dj&0Hw%d!gO)SUEgqi@e)%M9G{#~<Wq8GNnBvQNLid)uACxWwY>k~JDfdv
zM1FF{rt9%pK{S@Di<f-4dCv6olugt$c}Dr@0S`ZVkN)CIO4oDS4A4U|!1vFN_~(D}
zLl&jw>e&}uJo^&mI<h1oEeobaPB|@@&P%j%G~I?Qx8zxgQlk9K^MdWRqOKae3)EG^
zgR_Sed5L#{s;US<2*CT<xl=-iK-7{yDE^2$FSxiz?UFRb8YAcQ!OH@v6~TF0-=b`a
zA)=IKQWSL61uBe|H<_2Vj1NSki4m%*!h1g&y6yxIZ=(_Sh26OF><&<G9h&sp-~RRS
z?dY%nG=S@CDeR24Bx$lYD_2I7^vRwR8=R+WTCOfHxV$*$;^Km??SH-RpMty;4rp0S
z$cvIR6>Ae?60A)Dg>!Dq)D?fXx?a<DHFdq=<?~m(c=3wc+a-SZc8h%1^;A{OX0>6t
zT#{!!N(+~`>%<^^HYu_CgoDKtH*^vw7#vy~k|dLS>d`3KXg#*_r1KJ^n7G$<jW(LT
zZ7|kgtP#=IuCOzfBoWhPrLCmAcnO^>%YwthLmoUh<Ke@Hlx4-Q_P55ajoR_-cI8xL
zmq770Zni_z!icmD^s(YPyk^ZEooz5vh4M(d)r(`-(1?wTvU+rFh<lmD=n}X8?U9J#
z-YglU8S4o2VW9R6?a+!ZX$*UA?9LJlwexy~sO7Z<4KW8MNMW~2CU6L9)PtEN^1Qs0
ze~;Xo*shV~AP17bU^=4NFxY`!_Y7trQ8`v6SW;2^y0Pyzl5<^@L#>L65EL=$z0vY+
z-wWQ4wlrb%=o5v0R1kyEq(TtnQkYVsIrsQDfw1SAiw;N{=etCkH~k}(5iwgjIvr$L
zMxG{nR8VCKU1<4i6kXR4gChn(A8TLH)6vt>gjPC$n)}$U?fCxn<KI_Ef!A?g;Qg+}
zoCOk%wh3vLk<JTjnPO5+Ka6%PLx&5sAaqBNJ~Yg~U-0gKc}Q_|Jx+*k|8U=WJo@DO
zgtlY*{3}A=jurIo#u+aH?AZrzLUt;SA?hKHb`N6IYLbks$Vn1~%}X!|Ns^HkVpXA)
z#wdyF>ibTLE(KA`4ON%|<0DmdM^J&I<3svkpm&bM8nQg4k2{AWQO2fef=VpW$*IwY
zC^26<|EVC0xpSW214&j8100{6u)Mj&`C&Z4NMR@oTgAb_fy7~r$RApuY&{l{yTHjE
zqUlGzg3lezwx;Pjlomy&F_!DA3)XixG<`)nElG-;)EHURj3|=jW>ScEMjTH%0XKS>
z)!UjF9eE}ZaD5+0ik!M?sH&Q1Gtic-oKUSd<gw@B!Gc0t)^!8xEiD>9n-UW^Oy?ZZ
zV^qR&Qc=ze%Eg4~$q~Npd1P~5y?D*#i&xZj$Ev%am@QcM1AUe<eRRf<CzO*pgLfoN
z&rv?(4?p+_-SvF+w?E_h=8D7PlE+UUG0mnVCIc04p~Hucs@{+&#lgV=uIstES+TZD
zigJQBT6iLLP22Tkg`fsa)6jG+N^7*WxG^7gM;&l(Kt(aOwj*&~D^Yy!ihS{6${>o(
zg)Cx`u4_>yAy4zM^Q0(?jCj{ER9n2QNb=*oe$M$w-}S6kYf)&8#r`)}1;0l05WoKK
zzhV3nQunRL#l;2Xq$EuQc}=XvSdC4r7+1$Y@PVNp*sNAuou6}ge$Lg^<v0GmoqEt1
z&EjCfd_HG3oiizC6h%1}O!ASYDMV@2^}VC*8aCC2?RL#-dCRMpuX+9Qg6-;Cl6oSv
zZO3*~v0Sd0OztSl7HtIe^22~4P~<6jG9!di$n0(ykrH9B3ExzSpiAv42=OQNgQ06X
zVvJ%3Wvx(cv=U{c6>*t(|D~jPE`)n+(MFASRWO}SI6FJ#gZJNKvA6=>en0P7r=A$!
zv>}coXL^U)i5<NZeR^zT4|p0H?znI-+2|ERKFC_Jbo9;<T)_9xRt;&HVO2h+6-7L5
z&$spaHtg;ZcV*qUrW0smN9}7??S{VZ_UA4&w(qspI4wr|Ar5$D2--su=@rC+`(n-^
z{4gV5^EIT$IZPrhX$W8V`=Y}GuaE*|F@DUz<qgiXv}#MEH%#n|Oy#5@MfDaPH98t`
zhOi0UFh<-(DUD4uqc<&Lwc~}u9>>Bicu@*_JaLzh;v?QgaYgV-khma8Nnt0*F%h&E
zwJTCc+OdOzq&or;n_99wBg^ETTpTPoK04xHu^`Vg?&>?LVZ)|fb9sHq%Zrx`T}K#3
zPFBXf_+dxNd^25c$=fL1Z!;PGE@Rve)KsI(l%$+7J3JuI3Nh(k-LSsCgi1gk7Qkb$
zpD5;^<b3?kPWkXhkC`4+(ES4ycC$~u&l3gDx|aI+GvZkI>^*t_tpZwwk#uaN=uir+
zB(*>zd<*XzuocN{hMmngSUhEZxWL*-QdpdZ#2AK6gN=%pUwy{U{_dx|_t6ucy!RNB
zX;CGj2*I)0Zdn{3k(LF+))V@G?|PK?q=}&q0`=Ny35jJ;3Y}#Tz>Ftwq6D@FCp^7m
zG?49mM@&-k`Ggscdb_3VYVd)+tvNoNv)XJJTt^ynKnhQ%<+KqagwCOph_Zk-Sd*fZ
zCfWp5l=yzYJ4YM_luangDTl`c%ex!8&5EH~QBEghlPTB~lT8RpV?|aeZEB@RtRyKo
zGca^LeYK?v9j<ZA=W~Kr%;$&H>j7(1j5nm?4d=CDd3iy48aXU-Y#5jyPDy7IZtE>~
zn>9l+#Vj6Rii~--;?>m){>W|cL&LO~<9x>9gC{WgYt+_Zy~ASgbwk^>yu7*Nwhuf!
zo$(;o%!-`jvS54plB?&h$gJVN{F6Uqx;SLHS#x=H$%E4)CS{4MdT^TK#RImRHATK)
z-L1hSw0+Isov5m8%JbKkOs8|o`J6#Hs=mU<h~BjZkkfUHek9p?{NN?a)@0-9RZ!f-
zXo!|H$*@*TOCyp_(rmXR)-qdU6xN6dwStLFc#xKS_LqOjlYc%XIXGkR8ob!XxVEQS
ztwgb9N6QqMIrl5VTU?^I)pz$}N5=W5@%tF-r4-wVZ7HV(sZBA~Vy*1XZnPyChK{?t
zCCk;CVHlWAN|pyp`qmTtc&@WBnG}?hlF4kw>G282C#THkGiI|HlgW%E$<W%OjUAIJ
z#AtmO0&Oc64s~6zZMHORL)VFT=i83H>1e8&&1OSYZHU1Me;D!n^kEngmLPT%ZWztb
z2cg8kTb^Uq=+WQPVpGecEu>qKrPv}BfzZf9wxcj<G$2J8GZxAQo5*IJSTX;OLAZ7w
zy#E0Qi(kC;8niO#-B$H>*K<Ga6U*PEE_k`GX=zz=!%MFE*EHFdU>&^=G&}xt2%sc=
z1tnH3b`)6o7)A|^!o7jmC3}CnT}kO_=~%lBbzjpC!nc$hYK8W45BnhAo6dDKO-+)>
zwVxkMI6U5P-9WFAf^NorF_Ho2(AOKL^8+6L@Dsk={!}<$yUuk$YT`?z7<>dg1QwAv
z9cb-}mKDl_9uy@7lX5~vK^hA}FQ9@7M(hhZ=-YufNHM3Q@C>7t_hF~F_Bh0v8t|GJ
z<j|`T8jDeoDCA&ZrKnR0m|T-sO>Qy@n=_wGIXGI7PYSX!Bb($*7E_WeMQKG-HC$g^
z@bcwL{_dCmnOCn~F!UY<h<%g-O(E^o$@p1u?=jc%cG27-`K`b2!>>Y$$pe_59WtFS
zNYVl@+}y6Yy}ZJ$Yq;y=IzN_crQd<<J$Ua&2R#1%G4KBHF<B<Y`h0^}{MU`|R@gmy
z%z@GL*VjC|y~VATqE?NgSK00b?Gd}TxTpIw*5Eug71Q8PKlzkTKK>p`8B(k9uEq5O
z&PH$nt1a!e=H~K-yPG9v4-X06lNveo1s{3+@+E1WF<%_WZD&W%x>4$%*aTxzQVrNd
z`d*9xNeJM=UIn@Hql+;z1o5oWRyZYufpRt>c+Y0L;b^|#V6k9ZS7-&>&4!ctfh;Zh
zjTb+*`xsledNe_d0^gD58EK!>)h*6R3WYU>$$U=Kk@fP1rrP41BWXMG>0IuOw7`fr
zg1FU>sHjSzZP;@-`nIR*dTwv-&>>RiElH9|t5zGbB%vL|wY%#&UcG+BCm(-Il33ch
z!lnuDoSlFlxL(&p2d=3&TP$#S#`fiNSqP@5q}hVSykt6^(07iucUZ0ItBU2tCFjpy
z@CToMLfY1x9n48p<mUA&%+T|LKlm7f=KsUqe?D7x<@tTsXQdrZxcP~Akb%xU-4nza
zF=!|bO_4S&S;Z=wwiWCutL(Qbhw=;m3ogGh%YNZ1x@g&y$das*#AJq|G(!((I6c#W
z(~WL4y3qh0Jo(;pPuO84zxb_v?gbjnY4k)|Sg3>NhI3Ead$09fzuz}Je)5d(z5N5`
z2M0X&+-=TIjxeR=?dq1re8GCPCKZlhRmnwfjOFO?2Jb(3%x0T-`PDnPs%G#THrq8Z
zMyjUWN90UxSh25@G=_9doHMma6yjSpIj5@-Ax_>}Qh~Kr#5>w(Q8;D8qN#cF-FG>7
z?VS3~h;egiYsAdZ_iWae#F#KvPQ#C$TPE(@CuO%kh@rH!4MjvHh4vg1p~Ni39FTMp
zMLnxI3<p$|V{y3T<mQ=dbd#J<wz9NsLseB+=V;rOcD|r#8v!-7D~f-}IpCw5lOY7g
zF|g|d9E9Z28k$+fV$qO-r{Df~v-}tW!zigP!#HA%bR%*W1NC9(<<|G?x(+{1Rup2Z
zGPRzk{9IEBIJ3^tGz~|Ghs>8t=F2$?mQ+oR!AK*md=AAc7qpW1TMXdc*2?;44bB*x
zGn}5C?)@jHp^INy$_b-{-9FXqt{8MvVNCCDxri7Nu**I7^ZV?4Pp-hZn)z%_(=-hI
zPQaT6(;%B~btTG}TolMt%QA~MpO*Ysxk4j8dpdf?=!rR9b1*2Cu2dSxrHTWJew6BL
z?HseV<%K&Zu)E}<3+$e7Wq7NQ)%p<FK-X<}xO#_!7iMh7o_Lvlcz+iWfOmlaoCDwt
zXdzb+U1q0ydPa0!p|f~1I%+CgQ>likZE>o_;;`9bvutctF+WyW+MlRIQW2#hMpqbR
zu*y=qhRV&P3TK3mS5+GA<ozTq+&r5ME*p%}A{tT&pF9_v3%>Kt8=S8$SYNCe9}M`d
zXY>PO81X*HQ3_lMN^JSJ<4VOk(Ke?Tuj%%CIkg)3oacuNX7jmR3(n9FJ(rIz@ZDqb
zR=R0>p=AZ!gW0PvdqJ~!(a}Dyx%ZhvZalwaR%rqqS^YlRo!G^aL+7~n>Cdux{D?<y
z4#e$NxN7-o5fqRl7lC|V+K+{Mr>!HCxOeYGUU}s;DkE`RDavGkDgH?*(d~LJpFU-?
zUUPW3;P~hWXDw~ju<LssK72&dilZB6`+U!2HQ6WvnUJEWc#T3^k!SCtU2-|DS}x|j
zZHkftI|_19@}`(E)}eCZ;P8m`vvW4v9b^fHKRi5Qz1<4ejjLVSz7A7L*hXf%y_s=L
z32ltnn*<@MdhZ#Hr817$^04qpG`nucu-y?7m|>u5XV`X*c5YuOLhvcBsrM$N6%j^J
zX4O@T_nxY941;IAxfGiZtx(P|uNU}!WH<EK%90bT&d*T_4(k?uRMRdFdGP1~>-XQ~
z(kFiN@e`Vx&$D>$72NSnR9!KzYj#bAvR0hJL*U)-yvYK^`)_`qc>a|2Y0crRVOdug
zrFi_{eY_9Yy5{9qUgXx@J2<P*In#AJy4{Y&e75JjO=>A)4DD==vvXD#9q+ySl-kX>
zdE+@^jO=zBx?xL3+NqO!x?~ql#=UuOSiL6-7b2Sv0jCvpT~XIHA$C&PNJ9Qai%=&N
z9uP>&rmdwK5&b}l0aayU_NXkeOq<<?lp<GhoKMCd_L#C;brC<owHm6Ds*_R*qcu$_
z2UJZ>8|EAw9hMD(l!T45)K!Hk;LSRRtz>hcjd*_LSl~mJ7Jbf{5F<W_O-UxA3w>lZ
zZ<!yo5~K8E0=o%R<^9}HI8G_yqi5Ug*sL$vtk+zuRt&?4_o80(rA0YKtD(SF8tV*|
zvovjk51x<{%4q7Q!BlQfZ%BnTFFtaG&zfNJavZfLMcMdO)-s#5)Qf{3a#j%0!qi?h
zg+n|wW3MXP)7Mv4_Hr!-5`7-I7}sD0BpPL@OpU85hOs9UYE8C~w0Pd_-#6?ztvO9G
zb9w(v(jZQZcmZ`K63tj6#^x!Lf>36r=B-v&{P7TkmQ=aOtZ6wpTEMo)pZEAnC>|<u
zE2_;%pdTZqt>Cz{ATB=Q1eknn2d;QybHow8h1evbOMA|l(F&UfhU9R$!sQyLYpklM
zRKv_Js7!;=B~lbgZ&EsgHW(E#wid;#c2ru{rdTqLy()RaLm+gX-Ppkp(NT=K4-b|!
z^Hy5gq37wv6Yf8HkMoOj*5@0-eJ}6hdSO;^<^2a&K68;P<KXG@`(1j?{m@XLmXeZT
zYloXT>ZU<!E25CiK)>$kyB)sw(50&iYy+Ia>{Bp%9_Dux2X`z7&ovyJ)znJ^+LIah
zxLui#1?;G2*x8J;*FMF)i*x$jj%VNhzQiA0z0bYN&qNfRE;A@p4Kh_z^M%iUo_Red
z4xyM5OUSmVz*UfApkJ@KJU{2r!zaw=6)8l*;2Ha#_uhSv#o+-rZoDAU=g~6`1G;vY
zSPR(EVl6_9lQc)lSS58pqsT@|TPZ04cBPWWwH9+sxp4b5);V0=&~JAfE|*yA7>0oq
z3y<!awaXMym9B+%YuJO_R4REwHm*pcxO4m9QOQzQHM4^wl(lgFl<jsy-*=c0APaB7
z&F1K)k)5qJQgz6H)<iN+XATYzS)H#4L6l^5RZcz<{Wvm=UQ!I4!6eJGi)Wl1p9s)0
zjx@*)KF5S>YHn1PGdJhS#U<bS?i+mTt6$^bmDj0n-eD8KE6d60DZ3!M>_tjke)rov
zd+U3grh)mQ=A@}vHWjUN%$tUJ+v2JUQ&o&9p{>lODJNE|6~i#FT+H#_7xMSWY}Rse
zddfIPE>|P(z5AFak1u)Uwbz)n3;N+1o6QD4_SEKps;c&juhLg)?#Y5AYh45=QrR$2
zJnl+GPG7NoLJ%&Bb0zjkUnwZj&^Sx$ESuGe(=cL6t0DtFjda8A1L+o@ROS8%j8e51
z0(SxbS@=dqX_S$a51k88b52^AF?{%8&{|6yUu9B`MTxCTwKy@1J(DOV=ZueXcBCZc
z<e7l6nMzrdX;{Ps%k`4&x<mR$zr)a=twLKZUW?WU|Cg!ynA!F_o~<r;`s6839zAB)
zc|s`D<x~YJkA($WNz=nREj&`hzq6`q<_B}CSP?*4n>p=i3j&nHk59~cWiIr?AjK$S
z=!ZE@DZKx%A{I$eU=l>g*B`!G@;;YH7k*vo5aH=k&${1I%`1Y5BpekrRo%dDBe+0i
zl97L7PDSvL_vgUGVMZZfru-L3j0AcDA&&cW8Dk8!tM_L4Im)I!r@Sw8C_tw2h&2PP
zUT}JJgb6#kZs6i+Chxc+_b0gON@uodK-~nTAx}b!j~c$*E)hV{A^`6ISd^dD5bI2g
zqonir?Op|{u~fR|aB;@GIiPY4uF#20md&>zV|2pg5tA({8gieBJ~9jgeLvuXXBayA
zUQ#bst7p<ajZjsNS6+LG+t1xbJHwa)PcEKte))_(^eC61mTG7A`Kn4cO{&wl?gRMy
zF*N)29RvD~qg^g&n;A1V$L|8`XIr|}hVf$k!+{YRsni{V`4k*~3EUAh$BOo*;pons
z`p}{(Y3@fxK5kmj$1^B)j#N4H(FunyzQW0)N35PblL=}fh*E?Ss+c0?E4>nw`iRb)
zo}Ti;^Uq_*#9_o~0d3KHoKr9k4Bd`lyW#xlQ#R{MZr^^97y>#eE}mYHW8`RgurGHj
zYp_WauT9P}5o<-?b>!TY39dq0i?)(Qn#=BRG6GI2?Mdt@3%VACq=sz7<0ogdH4Nj(
z{rm4TpUs%f=IFV-I^d-bJ(<krls^!>p7-}dIEY#lzs6c}5GlBd5kinW)`LS)2GtFh
zt8;=M8Fm?y5?RSKW~~M5FiH%Ujk8p`CG}9NL{)XfF^IC3f?d;Oiz}Ihecz){c%?bj
zr!-B??sChlsxc{{qsN2+=X-qL^W40_#si$&^6rDTsa<A1JEqeMDr2CoiQ|YH6Kzhg
z*|1nv+&Mbp?6{?ED$Y($amJEjWbmHz)j3$rY~C=&EDFS7*xSo^pJ|$gs*+>9+xA?n
zc077;!S~*JNL97mym<=>5q%hXLI^BsNt+N2Pm=R8^MP16`CJMp*RsPldYXjG4!gLh
z_d`$P$KvjvO7xB_Cpv@8ii6pVF^Rot3_)^0lT?y>-^)41Cy?%s%J|N=zRU7($!s=b
zxmZwD4pSL)W2oB(Yea;kv?%m)v3t;_k_S+QAV0--<<jp_Vni!bzyuhig?CluE7zq~
z*ph{rlo17Cs;RAI*lt;0tYu;Nof3u&7$;Su%DQ6PQ|iM-1JUn#Ru?NCKYGIXyT5}8
zYCz0ngIN7o8yKUf^8?#{N4wn+Q<P)^rSY-Q8;Y(pWoe<s<cZM>MPwX>i={NPx?z3i
zIezL>s}DSFOsOOWS&}qCnZycZriT631xS`YTfo!NvB_Juaa$^`27{%cWielpb!N7l
z6O-rZlMBL-=%c6iBds5aDT+cii@C0KQuUW43h+cm0wYqr1W^rJYcYs-X|{k55Q8Kh
z<s{yiwW04KeJ|pZzIsIKj#%7qygWPME%TJkcOg9Dssc4FryBA@L~OAH^*P92{T<E;
zxZ;|fUp>w~Dr>U;m;@W;xF;XzeV{j6?7<9MNkpEt5|=0TA`%=Rn`E&A+H5z3l_z)k
zs*3jEJ#49@jo$Odo8RI7lXqDjEtxGB;2^7r(iz=q+LtTZrlQ-7bQj~5SmY046yBUW
ztynIEQlT0{s3P0G=h@RIJbdzwm==E&4x4zA$G{QPci`-cFnbLSZZ#}UXB^%<pgO6^
zbH!*RvJO7}`tb39oiZd>ljck2ckglY)1PDY>?!v*mqabTN@Zgsd!|x_P>PcGDr^X6
zGcUb(ho*us>>xxbVaI^7!sm#)9lMKjHmh?UKYYM!)^g+YR8kqN<^KCmB;8@x<CVtL
z70x=W)>y3xXu>cu>^io)HRCvn7g^QN%;vbd#@02;TC!3_v|6Ty($>_fSgmAHLZLKS
z8QF27QQGp(J3rv=-CG#rBvBwlOnDq*4B9#oMQ}y61Fkq;f`m~<W1^6XbBqG+qK6;=
z1FdoGoQ0A$(RO{w*muNnB&#jNK-09Sx=eYdWK5P8kOF421%2EKn3-A+C4+GsIteom
zp0;htDYLq~;9!10&Ivz`%$t_&Zo{zIa5Qh(MGwiduy9r@e1_buSnWnG;)dV)_8W|~
zrQdE@>;{(4o^m{NR7-<ZksLkU&cn$mN@>P0;A7cp<wW1_#D*Xw7K;VT<+Av|WvnrH
z@7ZiN?6#3dkJdbUdP(08+<ERc%Vo{zcWifS`hEa}BPV&80^CHs6$zyRsHen<>3lET
zJ2h2S%MLq+$S{t3)pXzWD7(S=h^uR~HCUr)jAhZxxQLUtl|tfuA|YqOL`u&eOlkRK
z{Nasnee(^hvs9Jk=;(;1sj1qEW>M43W>mFQ=_*%U<z^OdO@(X#C$CMdb=hglEyG?B
zFGk1uioaFTu$0nQv=w8?I?KG9GTBE_>ed!pL-?H*w9<N=pw&KNt5o`TeMdKR?D~OS
z_d9?(4Qd8;qp+25?CMzq$}x^2Udv=$<w`dq<D*zTh3E-sB*a15o@pS2k+JU?$B3xT
zEsJ)E4}sO@gVo5G5;055HfB#QDHfne<l<T}RnjFRR;)=p13NaXSgRFd<*_P}R46da
zlj5W*Rnoyr5uZf;JA}XxMk*Yj@FAdL5(}8phRnqZCFDpbb|1EgA*_)ovZP|E)X6+*
zN;OX@P17{2RnO=np@*a*qgu&ka*?_Dy23S?%`tTE0`DTL{M23rX~7!EHSi)*%-td2
z?Z+93t0appSBlXO-%9&1Y<r>}@J30J!$P6NT!C!z9#|&@sfy%fmI~oi)Yj}HJgwF!
z4>~AJ)EH+ll_9B2KX&Ds5Z~Dp62oP{zZFQswTj*E$Cx;cGng;u;@nX<u|Y-r=vi&n
z_=f{|SK7-leYDSA%f2E?3mk%Z9@>``Cts|Xzv5_5=QIZmcG(acO;i%ws*S>8_BPu;
z?)X^1PUlQinou?9qf?Gweud}P7i=${lH8Vkh+QG1%ja_?3<rf#iG!u%<>zlQY@eZg
zPh)GwQMgKVTagIFp<~!?SUr2nX0_(#&6{Xo)-<FT*<P+?hxg<hKaAK}BV<Km&>2*S
z_-)59>=-+V;Za#p*EOjhaZO7#Yq3p@au#iC_WQqB#1yJnl^_$Y&!C+n_()cU`SJj?
z<-z+8amI0SdWto+?B><gW9W$>eXS~_^|gqZaxEEY^X4dN8!?XHv-kvAv3_w)%Ym~v
zTd_KSN;mAtNdlHlp1^2AR~6by5VNw<Mq11da2Rrogy8q&|BQ%CQt;yA2k6SN?Y8*Q
zQ`?%&<w{aIvZ(d8>MHNn=t!j!O&IX26^qt#c=rxFXK3!+;QMcVm+tXnj+G>4jJ{(x
zz<e$sy&p$>5Uxo=vs@mcEEuE7MpHaV%fvX~{XpM!jAPGwy<v6Xxj2vPcAoj7;`tYD
z&@M9DZbjFv@gY!GqG+Ak;~)#1n1vS64D64O9CP-q^y?LZ5F^9r8AtgnR?nW{eW0BM
znl_YnvZ1z)Sv6yW1d1ChNtI}#EY*)s#viS*-O2{l7)9TWIA?K{#kFEmZYxoPR#nBU
zZHr>o?lYr}H5em{y0y0KX!G8?Ocy?@*1{=5t1GZalHl~R!I)BNaw+_U5}F=E{1JJ;
z9~x<@SdVnWAo;aPBCbY1vfXV(pwSQb@keYd6w+?0plW4)w$5R!Seb;F>H1Ny!64Qi
zl7Ssc(7V4foXs8|JuwAR41~c8zf6O*hJNOm%@_0^ic3qmc&4f1z)*gE6)iTE45kL{
zWWY5N85r2nsSTs<vE`lu2?Kuc41G`M2RxzVmV%ZVBOxbrN~jFM`y$&YRoTL?+gBu2
zDxgKw@p_7-V`?W$W#8Chj-JJG$yx#OK1k3&$N{Us8gK`S`EB8VYYV#o`56M<JXhJJ
z838);@`MIiR3`Do?=&(W-NL*M>U4QdC7YC3y86*dD_E81%=+5j)Cj1F*KUbIc}OG=
z%0uoId4y~r)QPbVMPLHs;E77?RMIw+*7?e}@8jW}A3X}kP*<ASd?p^2m4+IsrUEU6
zpK%-syU5rN_`wr){)g@5$~7a=T3jh~7r+tBUx39+6^AcW9KBpq->%SeN2(0F>=}1b
zp-u%KSTz0N?_;ExA(LWa$O=`}RHrw%`ReOjJh)GuJpj%pmaKp~*;~m@R;fZb4cxtT
zL|Y}+7mt}ci%%XfB&F#6h|P)I@9?`Fn~N1%!O_tHF(vA%W_!6tXK0OOxVR*RfE@;W
zTLX$PjD%fJ>PJ%cm;|+!N|+9jxa-9nv%F3dD{^%Wj8P_!iR`L1y6nWpaV#oSL-va2
zp1Z{lzV|jcWj5;#b}>U=Ywb;38wx;}eE+n%+GS49r<A(N9o3qY1_60NgVtoNK$#ND
zlR3=b){4I05ywFUKPjWD8XGgN(O_(;Bv|pyQeq++ix(fdw9U1`6mCxnL6m8A4Y3ev
zt;NTHk4bj2Lr+Nk9vpSsfw0-){RVS9!^VVBaO2hus><+m9Qk}xGrsj5UVrHZr)P88
zs-muhh#QlpHV$%<vbz%V(v+3FrWnif4s_j?Znt9?dN!L4n{_67F)n`brMujD?uf*i
zu6u?L9a<ThdQl9VO9eJc8F}w3pAtF5(yHB;Rm;MB1?nWUdpZ{e`mSf^wp6YmszJco
z5Kxs7kH2b$QH~t4gjS+Sc_M@VFi+@{@ke9Cm|3qs!Y|H%vI^H&=1nW9G|o}GN^+&G
zv^%YJlG$t|iAE|EBH+?SV~oO0)nhFiFRc+_g%)tAi~?uCNIF1wx#e=bX1DE$KaO@H
zK;I2)He1eDmsE8{Th{{Yg6G-AITx!-cH5EkqpE)jsm9C{&CF5Na}kkgNvt2oNJs;_
zen4s2Fvj5V!PEDHc*Mn``YwjBQ<Zwp2h(6&B!&$K7v};HuZ@(%mSw-gqim83VoBv%
zQc9Ymde4?033POM%`RWks|_vne!Yc67)H8n$IkZzorx%t62V8(sU#w3h?(F$%Gl!X
zCjSXmNkk=_K}xG$E2-!u(tb_MdsM3S@4*^NRVkc<+`%XhLDVu`SeI!}p=zO7z^EX-
z1NkxJLL1+=<wrR`S(KkS;7R$Mf6|F&I#v@nyndco_AT68m(#_9asu*{7RMBlf{4t4
zr45|MOzI=yGKwMa76C{aX$Kc&GQK=##Z|P{1n3k$9%%ijag8UZltNXYYK1+poZLEL
zHg9pxkz&RN&*h~EjJD@nsgO>uV;a{1ra>xm6R0g*MtuzCdBxe66*pg)vAogJoK>hp
z!;lg~kPg;%+cEfn_kl6U!P|T$e=M~2@qnEw`yr!Iy2ctsdvd~^S6^e)Zvo%m-&Y8-
z`ie|mV2m@0rh?V^Bj$A^r6BDKr-?=p5OQVOZn${%lo&j>Zk=*?uq+Wan%#Db)0V^e
zlGW;hG>inL@Ph;fCm+az2Or5fpq!>QRwkDqtca`>LHD4f4W<$$Cg(~GH?{0cxfrK@
z_e6j;#&CLi!ugYPe(=tHUU=c203>s@>%QXBscT?wa^b#qa%Ipe0}etWMvn;y_{dS(
zvlx<=s+_3n4ym1Dw=rzDokX;B1G=;plrkunuUbV)OhXl!7>tC2$58-YQSw=(h3gn2
z&RCR@N|viEM2MphgRzR_g;O%d03>QH@jBbhhVI!@wwo3GZp*vxyiJ>B(w`s9dGVE(
zSj;UN2tJas7N4G^aK@o>5OAk6V@Oyf6R00YE>|nOABZ6^_5<VK$;pvI=DEAKdG)my
zF)p(1F4*kO#e7;@oUNrLy}y{{xW^EGh+8*}D>b(&*}u|;4IyBxrJe~-PogX{CS_De
zkR<#&j*hx&OO>=x3>4&C&I_$RFduyM_yqd>ks1+l%#4F)Tz$M<^nOg$3<K6lcC2+4
zXLJ$b2p7~D4V6V%1I93RBah#I#JKf8wz_Zouw%D=kNZ#F=Wuy|HCkG+(X-iZ@fVQ)
zMWVaOogYY8V%DUnuDDV_>PJEv>Bp@kH%vtR>wscX0aoxsCw&&)-ll`q_VS0aVsrMS
z)FUh9l-J7Dd{wSrtBS8FBbgCR;F1TtKYW*W#&7e)zen5L!m3CjU`QOaGn7)ylE+(6
zyEU83EjbF8HpC!nMo9R<(=4ie)nB6tQ!AE8Mp3Djs+2S2GfR8gi$>OziTVMVx~f<%
zD>)`PvQrR+l1Y$eg;ek6aM;3Z3E?&9E@1T#Hs3@zT_aa5+K7OpR6g?w=l)(#B>{Mh
zv~dq1A3>f#*;FwESz9K6)*~Qp;`!~k5`|Fyl~l+No+mw5z;#NEfS)YlYX;TQF*|5D
zIXPyrm|-ecl7=D|s}&C)y~pNqi{Iv}^pEKp`KUJa^$O(z0o4@(>Yj($r{VY|!*j3B
zIl6a1Hjy!pJlH<t;@QAv3%d@sI|29~I9?itd%)3;{}w+Mu!}iih-e+9f}s?7xukmT
z1(bRL0H2bSWXF-HGov<iSrcr9_5qVd^0>xw%A#$kTOlT6#Lvcrz_8gdtT%++4yP1%
zZr!A+Ej|S>iQTYee$b#QRcuN;_>#IX$fBR3Tu3NU?BtwrPM6;eq*2P8;#*ad^)yo{
z5=qLyNeNGYD7!DKaj62O=xiW`Or-}74i`K)*F3)1ax)I(SeG`eA&G$#S(Wp=D06cb
zCD^q2$V$G-CKIhDSxXKUoh6r4=Y-1A!pKUIjm9`dwo8=NRC<G1cQ6EO@*-T(iloK5
z+^9xKyh@{RXlHQIl8WNW;~2Gw6BJ?!?sCRx@fK1@bc|6!y6s+r*Vv@7$+8OrmQ1fQ
zuGLg-<a4)=_|~ugDo@{kh#g196#4wEQ$}5(shH)&CJ&6Jq%Oupl>``%HqzPBo{rMS
z=(|h0%>^n!3Ypv|(k^3D<mRd2_1A84bl9*P*R0nA{g8=CVU1+G=Av}XDHDAV&^#%b
zwUX*!PE!w7!1=_ni&A#b8T{zUF>`QmKwVu$$z_$qxYKBc(4%~!YAm$}?R$JmbSe`x
zBps<tnD#vK{2!Qkr>*iQ<Nxo*)b>gcLME!ns7l<F9_8POXE*h5l`2$(@Z+^%uRmQ6
z!6#PdYbKKrvG<Xx&yQRfOT|G~3gaBcHdOUYGGL7$v>1dxr-DQ^g%HWatA*}rTD@A7
zrv-a@UIW^}askIDM>NeP*X5nlN0128N|x@n%$>&UIdVx0><f8^Vq~TgD<1NA_a0A&
zhoq$^seutsq9+m18FT^_<=P(y&*&2>!C1WfbX4S!NIu|<#VJb(7A#HJt4+Eo|6K)~
zm5xYJ4jZir<FFTOnG%sTpCx?g7&`EyaHC@ZFgZ!EP}E>rY0)m`(4N5V23$OX;T+-x
zSAq2-Qu&%ARl7B!KK=ceQnm<^Xc15|0k3IOu=o5zz~J=r={orER2R59XR7kJ6IFEr
zPt!T1_|aU#^7qwJ)6QGwi#f-qC(IVDe6EPa)i89NKY7MxwPk#g2wpbtDdwx^y>_l*
z`Dh-ZLsZ6-+V~jsi}3sx;P_rcd+ac6=AwTh+*u7M>E2ldwFNVSSVP|z4()U-Kjt<6
zv4EXiKNF)N6ICKQxgzW8L_W)=rY{6{t)#7{vPLJ3HJVxLxN-BC#o>awb=X#uT_%`B
z8iH`+x{>weiZG5G92~MdTynYFu{c;@jHVxZ7K;X-JXvRHTaZyXVYNaTX<JIwOvnL7
zD~yV=<3<t(q95_SK^cqHT0oD!GVn|}?@7&BOQp_bBkfeBhzhI<TyD2KI*%BegbR`<
zIlC4WWGTTb*mDae6~)#0DPTe=g?1K`9MQWX7LfO+Ns@a@k>e*tG8!{m&}K_YmXNk2
zA4qDHlUFCS1EnfZrAeVcX(N?CTR{M$#V|QfuK|TY<4V+8j6njM!$3^nqZY^S6o~zX
zZs^eMjI*1!=-nK%m{DcLmtK4YHD<QE9d)HSuMBUldj@5>tze8WL@{7BDHD^}jf_Ar
zx+t3|3SJ3xOd%8cNE`wtYZl9fJNHhwabtnfnXVh?`;nM5MvIr&lp-P-i;~-F6xS9}
z`6iVEm9HP1=~!io0d3k>@3PBMp*0$xf|z_7%Q$-CHnCVhl_JCH0(&ezSp_;{)m5a~
zClKzxg`+HJ6Dv%{?*kjWEKI@wn2$YKo6NvAR<dGkMbpe^XG^MPCYi5F6T*NG13uJ}
zXrUs090+3&^V#cMsA(~u-fEII?`$c_Eq87m^YV+Ya&+`vuG_gNt<bq}f%kK#%&%fi
zRXiuNVvGY(4H!L;fgR^~)1j+OTR8!p1&YZWH?>ZS+F2=b7M^k_i9RX<A(LaoT8B}l
z*tkfP*5ubJ&Djr@Bd6o4l}1^IGnxTW$WK;Bl0$82XAakUqL;m9Qedh*ZY37KRi!~y
z;96*BkdDD`Ag;kbhPZ_^K<Xh~Kw2Q+?g`RH{JrM(PQG!IXWaDo*FpRQ_NHpyhk@OP
z$Mv68*PaLV8w91nB#j=Fbleb=`r1|EBWn$|wNwjH?$%9BJ8$tZGOP!N&J%j!SZ=l(
zx=lw|edxKbZ%C&1a2-@St|H4CMDcnI)lG0Wp?U!pFF|#jP<3EPiI_)>5`MX>D{Ngc
zuOUVS%963{y2Q{&d>0so5nFw{e)`Aru`6$8Pxeis22sYaSS$fp-naqF0e`WgvX+z;
zAw^ldi5#BJIlX&IHLuZ)MLFriXQcfXy2y4ju<HhFRdaN5%;MmHCzmVgrWNUX2srB`
z!9h!lF~uagn^`U*6e3W-%JwG}gkE;26L~g=k>q=Vkv3vhPFf!bDVGcwIdLZI0<92J
zS`<T4toy*z^+45X#su?g?LJ_?c<&eSl;z63HuePm^0&rXa1NDHY3b^H`+@6^LzWX{
zJ|`)|=q&wq!{|rCuA?Gj8bgc&rmC>cV5}?GUnDw*P8nS}v~>`R*IE|vNRU}Zcy56(
z_9$b~x*{acZnHt<k%N;{+|e0vcEqFm&-mP{cR3$CE8w%A{Sw<0u$ATF=|g^Vwc*?6
zD?YbqIcppTmBHqKN&&AGF)GGUk-ft>5m0FDP|C>1mNFp+l!Zk-<M`x+lhYGy<rv0+
zuJ7o_ftVtdt3@qsMX8ojBE%rP9Wru4$gxN%k@prA6IC=9&?ld{WN3;%S16pn{V_9U
zZ+oJ(B@P<p4K`&i&!2Gme2-O*m>rfHYnG^ezv7ee$@n8OEa(>M14C0cSW{zcLpxhA
zUo2_b1y$2xj3Yz=cH`LLtU>9>u2W>S+uKEbfT#5#P+=Q5z1ec_-Wi{I?X$f4@~60Q
zV~xl^W$3~ImTgzs&!dxj!Sn<Q7{)ExZOPi>b7sp0w2>xPPzmR4l38f<KD*Uma9Uy{
zl^TdK6N1NkFTgHJ=R|UD-JU;}3fd4rH!=6FUXRrBL;$C?!%{)Yc%=w2l7di?T;*sp
z)W#Er;%&zjR}CdgMpU4jd@d@>aXLILZirE_dy*N~iqJvm6vNYuKg*=|A<eilD*nCR
zKTK9EQ}RRN3edv`tG3rFx8mTS_L${rtt|huK%M1pa=E_)?G(l;YG<**prgWMMN_pL
zEEaq5Qr$L^@tnl?yW8!!|L`%ZCmZ_J@WaA~>tm`qUvDj!LQjb#mTKY&SK^36P|rd0
z94v0a>?XJq2zJCL@E+<)(NvbUX=&#zSW7Gi+qo948+JpW>qmCGj@52OGvDx|O$vW_
zd^})R4un=pga=5AEarr*1y(O^-QegY!^3ZW6YV38NadwLtzG8s^JgqiTTxh5mPAp2
z#6%neW7pGdcO){+VouexvLW!213C;N<Z=N<*(pUYU@s<-&MQ6b!nE-GBuavu4Pv8M
ztUh8QDMcI&F)LK@8?!i+G3atJ2}w1fbRx&1XwsUbEg>mJ99@d6$HW*_AqQXW!Qom}
zsC;I+D5OMK0aJ|G#P`x#c6g*KQ`=nJlJ!L9&slhc38FFRrp8&}Dvs+*!gdFn4i&(*
znmv2aTs@9)6&a0bMqFRh554RVOi#|4%2lPBDCWo3Swf8Tz9T8c;%LdiV$SUF2w%;4
zZ|!;S(KD)Vy$Q1!w?F$?UVY(3e&y?5<>_{fy>Um#)j9C}&5C{(cqMBNoG7NFQG~3?
zUZFjtI+0u93u^5EEnZk97xwUQ&e_cqmWPL=442yt+ipjU^4W^llAPufC`-jC6jic;
z(-o*@|5@j&N}X144R~v%r#p#CDg-eqwr)R<Ia8yug)!0C%#+n4!ula*sjzB}Q;x<h
zB;JfG0C`=N`^orZ{DBV#)f{F^&B4(j)-`BdVN69+x6B$xRa>e`43D9~k1(i29xJ+R
zFd>V)bH~+UKP~jvhmv;Ut>fFg_R?qh(wBaQS6+UZvm4+1;Qe$4B@ikbL(UqNH7aSL
zwWK1HaDmnEHc@ZEjy%TmAiT$*))+1Qy^)cTfq*9y>1Uc;IcyYYHcvGh)d{^PdB$-^
z4$@+sHI4MlN*5@Le*=_?o!Rs1a=|ncnw8&MG!+%mMq*47RG%}tX|Z*Ut9usxj;teH
z5B+e}eN4hsA{I|;V@f)vk?A~XP%d-wYC~ljtg@J-87@1{-`laeAGvrN#uMNTBwzMN
zftqQr`Qx%druOKh$W5RIP(f-cC4o<>uA#ivxxuuS>ehnF)D+GWbzK+nktsz8#pnl=
z0%H}fcDTx6taJdat8mVOmbr86268vjZFXE<oU?uVlJln<#%=x=3ywa_-Mh{WwOlEB
zRe7&7q&oL9a0k>J5ceP2V+b{@H0;(ey8(5rutuXXQfM$v@~3mAwvs$$t>m01G;OPh
z$>LK@)<+!J@4;sEV;l?_AG=~Fpj-AlDHXd8qtTT-RDFERjZfcZeg2gG{$tD-Fma@*
zz%457Ja>Yzfy@ZXkb{SmP~M})5!H9(p~uw?&3sNhpOdl}x*BWQZa30W%(74qUVO-6
z0ZFD9gX$s{&=|-@uuS1H`B9P6NDLz}1`-J!vs@rXla#_Ji4V%T1hi*RLHKb>NxFm5
zLMDz_hNS4DXtMSI?L$?#>v^=-Z`Uh6`y>P)SyMn|W(v=lL=Z7bN{7zk$2P_wff~->
zmJ1eIW0k>gcMO{yAwm@c6)nm+X;&A3sw;=7D`ty1!_d)non%Cos-2l4d<v+sM$tIS
z%r&%4gVqqWLDzFGA3fp8YRAd;0;>XV{PtIgZ-1B6Hf&bsEZq%~c4XHw+L{Ls?$hor
zIXJ3m>r6HRAWe2S@9-s{+$xP$7NzW7eO9+Mi{l0J!v$DF?|U}8E&Vv6jmBEZk+oK)
zTnRmwMyT)wOZz*hz1T(b!{ACOP0m5Q?u?;wjueuJOj1HC#b_+s&4#9Fu-Xt)KxGAS
zq^%YC@|^9{?@_f4#wx5y*y^ek`^orZ{GpFhp-><j_uz@y5LG6qC>v<+aaD(N4c1D9
zdK`QDVasZD!Lz4N@!OHqrH|%MT^|`X=dim}_`%}rjCOV)^pI;K=}gQSR3sxU9-R$j
zi$U^9l~P#cG4;UCUb0hbF7uj|x}b|YyozWn4o4`^%d5<eKp%Ute@dA!bRtwsBIxjW
zfUGc8Lpz`E(-vSV+=x0?k&0ol@UgE~;&rJsNs^7FHZ7@wAqIwVD3&({t18;VhGaX`
zHj%fow+y)eyK>X!>T?~EfTjuPr7%#{naWveXRwEnL#>z{C{E5YPir_|LU<E+i;o6(
zKTfOG02S1WntImaW)*H$p(~9sxlplk0b~N$>Z&0Hg&#6GYE-tUESvR^gsWIpLO(Lb
zV6AYww|*_}Q5jUoY}Z@q%5)ulKS(#Nbl37Ok%PE*q!`7gG~|!wrA|fGmMazRRC$~*
z`Na2|Kwk^o0nR|3LiHTfFG6w!uo^H9sv7DR7E7p{!)c2%l0)4#Ev|$S7-J<j+E!>S
zs_zjGguQ4|p$i%I#~&Zhpg2txB>@>JNuHpx8myeicCqB(r58DQ{E+uo8<M{=(m6dn
z<oN8Ey2*QApm7j1?z<g+vnF<HY#K20IrU<WZE9kI`QiX=EnVN^stTnIRa=vC35SP$
zb>@0?Ol4g-YE#=-p{!!)Gb%-LiiCc^M1{{8tK|4aGRoM(X_KI1X|shZA0jcPqMa>3
z4~(H_3_+@JT4Svx*+?#+Q#`Ds+A}G+KIogK_CHH0gVuWj|5R-<d*VU{G&!Zh8xcU7
zNLXWN7Yl0Z=#*iz-7pLtA$by6Ca9{#Uk8=0!d5l2#gg5w$5jn=JI9smOaXTpoocDl
z8rfZBKwFJ%XN<1q!Q*q*L!{3gRZ{2}nVXty9Cr_9Y<D|C(u8KlGa@JLf_L6~N)vb7
zI&GNMjZA`MNv_ZtA3e@i;v?iEV-Oykt!oyCOI%wsde64sF?g|BsGA1o3UDfXIJBIz
zYR}IBQZXh@Y7gx897(OQN|92O_KEl8l&GtUtP|rn?km?Z1pF{CE75L>gi--TPo<z%
z9gpAn7Pn8%utzOf2^VOxY!L96KB?M$GX8K!ggC%JhA@bztpM;q55M%FG!|>u80*kR
zs0}%ag5Ud*uJ0JP!<BjLk1^5~Hs>7|PcP{<J$?)yY1W;UM4(;$pj3deBpzfHCI2B0
zbYa6qKIcJx!lhs1RVZ{Hg~MWTBqpn&D=~@6SuAPF<HrC#Wl0;+n!2uVS~zwSnP;lb
z?O)V~Q0!!<By|PpnHX(wT5NKLFv?#kP{vR-wQ$GGNX9da0_Ng}JmzwipM_BHh>`ET
z7d~{X5-BMfld&_XjAdRcs%U6*Vmk}$Y@~maD+TSx8HFF0#zat}(oS6}=}Af|iN%XF
z870-mkjY7rgA&p@!lzD2xalcJ#uV^nUR6lOZolp14AdG_rXPCxp=TUMu^)@qUelxw
zo`$)0o5uC(vHsBaAI_7Vz~AIWcLLSR(A+h&Hyz7+4en-SOd~!C#YX|IQq+xRJ~OmU
zLsLs-aRPSM6-y%}oNTRSt;*zLaTZA_q7^t}Kj4Y@<BXrEVwa(Slj3HsH5hqfT~#b@
z-{$70UuRgYdG{;d!q^IB5+}EA5lzJ|DVo_FzZ=QQFpQlzp9fFygBTYA7)#Z*JiAy?
zwJj+tHrpM?hey<POI<fqvpI{x3)4yx0`_fvCB4*~Qn?T!Ay|~RU=3!lgucU%11bh$
z-|Z7y?7pI7AZL;?dtW*zzEmaUqG>9eH6`mde4y$^zFx5_uSpkS#q~-a?8&rKdp42_
zxjGm1j6{X$oRRQlu?Qivk1Wedj4YLNRI?JFWesV!!3WRga#iBXDy$PT<y73%Q&mgm
z<8p!0irH*d+V1k6Clwbdk+eZsC*^Rfv31S)X2;Vf=d5=fT^MnyA%`BnIj7c{o3le6
zUUmc@sAd*ZIYLf62!Zfyi<;fw_RLTxsTL$_z(lNpn2OnHhzw(-X)BITPifkjWL|bV
zE;k$caX=fh=hRItYw-`f0@67p*<nr7`!!HKUE`B6u`x!j{hartYL(cxczxNRIBWMD
zJEg?9Sz)P^Lysd*p1p@(e3#SLm-wXF^q0hxa93x`M^^AY0eYW|Kai1;*suU$2Rl82
z89&I6o2nP;+RuJBqle4Km%RV(1D-s7%6hfI-}oSxXtIk+h2uOqHcy+{tRk6!&jUm3
z*~AUc)HCj<`&^DIOk)eLxOh-v3&;}lSF`uQQ&?@JTZ4QK2{4Vq8Y8J6I1Hw|_3N-r
z?^W&7|F0-C)4f#bcxkM*I9H+jZqGv#@Yqn-7V8W-d(=(<+XUkL2$Oy!mI@xo*TEN%
zH%3Agi5_U6sSR#ssq30$)o?TqoS*hQI_mM;NPK{RyN^>8O28-%0e<3sDNxR0YlCqb
zTNzwsP)=iPFRl9|MGu8VD?wZ-LQa940x<`6+m0AK(R;!$3McR<nF1A8;)d&EvPj7Y
z;G4jpE?-YP!3hv9fiuu&qH;ZW-EjJR!_l2Nr_UdgoupfJ{f==IGi&24&RLqOrk%|N
zBRGpTVs5_Ajb`F!N&!g@tcD=nMr}2%ZA)S0#~&XvQtS%U$%R$w4JGnQK1_pME?GQx
zhl7`%r+M!w{llkZ<#Dq)U;Fkuv`xheXE%wP9XC#v%$fs|i;THvz3Zs0rJhM1T{COh
zzW)&CDu!`n@Sgd6fzleE6UrKlbNkn*RRKBV0!k7pN7?N~1Hq!4T<Fd^e2IohV^0c!
zpuk%LT8R3}IkGZ9B5L7AB$LoqjC-9?G*)9(LZwIw;~t<*lJ={`x>ONMyDt@C#B{rU
z->G86UQuhr=oBDjG0L67q!D7TaK#cSfzO#}G|Xydt)uE{HtP-B^^ShG!&rxHW~fSJ
zS5{lHtI4wk3e9f2!-vFt)|QD~CQ@B`HCZW)sc=rJQ#mCbzW;!Wr{@qP1!C4l&Q1^6
z4L!Sy6}L}svkFk19P)I%WBY8)>hXQH7w25)$Wa$K4jM-yM^9=U$|g}lZ#%la!-qi1
zusk^A`1l5{t{A;!%=Y6zh*4Uc<$TO#0f<?4%!Mu}r*N7qg0Dc;RF&QXzyjt!+%nh5
z#>Oa3)<(+yLE7UfM2xc}1tBDCvQ*3%R~=8@zd#*dvNaV?*B9cGmHEK)eKI~7e<a37
z0k=}A`Mn(H4_Bm|c;kg{a_8nvcDHwM>-s&Wg-=LYs%&IaDX%EyNy-bbV_+9LR$;@#
z{t2V*S<IHSuEtUA+ptr}8URtOOrnqU>#nrFWa>=_;IhtQD|hWRi6VAV$I2sqxJ;R@
zD)JP5dI5kF$a9s$)=r8|!Q<nA(~in!>bgOx5fp@&8HP;XLEjZ|MQOvPse0$7ikU{K
zy7~de5t><~bue?9#d5~7S~5R9<i?AUcW>Y4?OzKFZ)AA$M;su>l?dWrB+kr`$ABNh
z$DsND9*v1=HR0@OOrCzz?<b?<X>(!`FgN{v0)p2nblijH7A)^-p8H(Q-7nmvKB+LR
zChN%dV#II76Lw6QS#6okXZyVAx^1q+dr@32(6X5jWUi^Gsi<(up(_NK7tOUK8(%;7
ze&_M=Rk7E{<gYZXmLMqHIqcy9hxeZ2?4@(w+w`b3aD3wyTj0*US9tjLJAD1ue~Z?^
z7r*c_H;-x#4sT#DH~4PDqHUQk7i2WMuH(jH!Dh25zF`%{Sb`EF>vg*dg(^{3kWgrq
zNO6>tAA>R)=US4nB&{V_Ac0R3dy_+C=sH5m*p#s-lu^<O2-13X+K{b)#vDDhQ`APh
zj)EWeW|X-AyM08^RIyW6VhP~N%=klBjCigURh7}`31IzDbw@N>(wYEbBC1R@3Y;z8
zONoWn7-QLOd&YhMWwBcF`h1pFw(N+d^*aoM>{6}87%>UXNo-+E;R`4wZOSJPAM*Y?
z?_zS|plL9%Mum#Aqg%Z9=srGa5|%qJy~?Nm*dOP7z2lv4{d4~2fBG%Xho05gQNInR
zrsnxsg>!*m#!{)YY_^wd)|U*Ur>a|S-@Z##H^h|K>^ipH4j&|FeTrjJx`eeW4cUlT
zLKhK<0?7N_?}x_~xVus}PrGl02r5)m`{U>Pz&MT!eUG!2#!7iz>jq6jV-A>AaPQ@x
zqPg`FUw!*+R_mS61JnnfcRIG8K)g@Je=A0Se%14>fBr3Af9+MapZQ?o$F%0k*Hd_y
zfE}w7N=1_PgxnMC$S&+yhc(-AOMP5R>>Cv|En;6B3VlVWEtRb#`J)1>4C8RlICS)0
z8c|vc;k~jVB)eX%vltADP-GO6B0gQYA8gc$y^2M`7}9>;Z&V}enNFnW$;rz+W>iT)
zs0bml*=&VZZ3^g20^EjPgcVvr=mB%px?K-oci^f>TlX9`nU;pSt*GueUOT(T;p-PX
z`PPm{zd5k|Cx|5q2*{b0+hi^s5TG>hL51z_MfLXMj7gz7839*FbuDpaCLGFb5<obv
zy!NK|)0F#6UU3UJy1fAB2`12+$jwhXo_qb28!sMk^g@F<%tSLX(zEF=IbUr^K2kZ&
z!9l}fw!~Ext`fmfTebT_Lr@W|Z1Fl2do_!pqM}j`M@7Qp=}UoJVQ?SdZ2maoV*xwS
z9;m`kQ2P?z6rTc8$<a8Bu9+R3aQNa~9zT0R>l`;;ehs(YaB;C_v+Fs#c?X+EhA#2e
zv-{jVSrYRMDodIlV-8O7TGJ0B)@UxbYi6?sTRD`os3I!(075SSYFe0M&c!6Ss7X~Z
zn<a}UQ9H*PhtY<<-{HrR<VQqSpD{+ED*z(~!FWPe5G<SX4c)~B&dgZUa2a6?8I@!k
zWwa@tWrfqVH_4R{Yc(YgDw!;@Qe;(jlBL}`nJ5~hiXBp*=t0O?VM0X3Qdd?9l@cLF
zNCuNM#EAEaIC!i!%;yVYkjG3(T807|Wt!44p-rM`=J;`7=mzZEpp67GCoL^v;}qIT
ztJ;Uag9i`kH|LzTBIG!%6dMI^z4bN;kDDEG{`eWEw{EgHS#bLDtGxE|^XNszU;BUl
z-x;1f=KdJ??Z;23pS#JSN>o0RjD&P|=a+QrElFE$J^vy%Z{EWw&k#Cx-38;&lar@*
zQtF<3jI@z_R*l9ah0Y42r419K?3h&96Dbs<9`GuYv?gOn#Y8U~h^3vMvv4U=7QRf1
zkb}bak<pEKH&WLvSvi`z#sWDh4$tnfy!kRuL&e3W=i=EpF@$SZ*VT42=W9ROpWq*Y
zdZk5GWY{12)sN!peYn;8qu}H3B<lDCc78H`k4A<tL>~U&DG%Ozz;?6!;IGU1eNETT
zl+v12AQOm;jJYSMfz5Eqs=Fi*Su_>aROsS8CktgFp)qA$NHLN8Wb>4m%$uVZAGxb6
zV6(5Z$;&SOd8M&C4R!5xPaQt>q5n*=T8SZ&s)SR9nQj=}jBeZr-?z9B)J=mK1HBn=
zE;6rWev2`&T1l(iT1Wu~@m(jC)~t^(_Co0I>%`~>W;W5-k-2l2*_`F;4tua<ezxGD
z={SG0C#=&|l-G=_Ao~KIQ~^$ThXB3t+AV$PgAjW5l^Ee))LM1rcD>y{0kw(lIf1#F
zEAfNlgJ@;~nv;*>1ne5FqP`l?3#eU9V=F3W(KAJGJ%NOCisezut6#pstrw0sxG|$X
zG<0^u5Iwz@E>g}3=LG!E<_&GzNUpRMx)WO3stU0!Gl)<|mk&mC8L|dhEP=c~qdr`K
z{Bg&}tJvjhRiX?zWK>ibqeZ2JR4<&a@Y$kWMSXh6*{5IP?#)}UJirbEN6v7pPq}ab
z?FY8!PhlK6e({vvHn`fdymOmXzok2WLPl~GyWN@_H*bJ}A%&8PFuk_4FVE+)otf-J
ztgTB1Y9u8uBw2(bmtBQ2lCT`LC9)<*;oN24kq0H&qQ+vCC8?1yMm8(YH^2IA7K@R^
zVWKvHNMi7yOE#X>mY}4UF3(}e#)!qmRFP4dIGwPH{ccdHT!4r!JiT00nNo#HO3Uk-
zT8pXh|8n^o3OCM6YcMBy4aRDkCbI2z!cQq@zZ_OEEY4X;9-<O*bP8f=qZD}JMC*en
z=Dd%dopXM9LE|EArLky^?<{%e>^8&O517v!Dk{cxPh}!s`|7Xq-}$%xXPA>EfA;V%
z@|iDvk%w=8k8k|)f6dcxeuHm3e!^!?4zR0%*-6XAvrATwSNNbfeEww)UilP5Gl%|y
z$D1`5Lq|>t<0{GUjb&2AVwAxchqaFAg^(&AOct)A6<;Y*mezX#{yIxY`<O*kqqHIb
zF(!#w(-Gx0WQ$JXH<(6<KFcDJK^uqGikv+@ja178d49ld%&ayYtMe7b&$?6%r%iy`
zgSsoG4(uh1<@(zLIyFU}$@lw8aCw|O@h7sJZ*rgMK|X5QP(R}5*T+ZsS9~%)89&*G
zK5_p3mizC#&vyM`U?=|?rLRs+jR==iBq<S1^s&R|j@9mhi|sjS33by7u)`LRr3$Cc
zNNp`g{5UcWB841>0Ux87um`9ra87`oQ9o=z3=?Q8ndtj#;wpz4MfvR&6-Sl$Gw(fP
z8mUr6t>-N21tAR#VcZMutg&PxmBY#@j*jLyYw^J|^druyy&Z~8c@OA9l6Gwzz?TTG
z-Y2&8KvNHFQFzmHe7xY~WWlYMTAITZ{-2$*eS0K6h@_qm0;xHuIpU>f6?x2(^*RB(
zORntJRB2;ZR@&zt;$<K|O{A%%dwmKq0<_$bsB5)W<fsUnK->mmp9EOWxJuQSgbkLf
z?Aq(e6oofBsalmQr;k&ZR#P1r76)?<kLS!5HP#s3e*e2DGR>mmrPq)7%ul~edmt^+
z$Vg0qZtU4~9YYtWYfEivDk~<)uCi2hg|!Z)G?BvPR3c}#7#vmUvz4Tla`qSuxwKzn
zvDC>w!Cd<jw_-ow^YtNOpH!ZNcegk^!Yakxvm2lkhK$Xg$|>AI%j)r(JnVSog}a;`
z&KVj_yO?o$=Qd&Jd9T}X@7{AL1;GcF%OzSHDS2zfRIuA$PU)&1Je9il+%|P}VM$9<
zem3>L(HW&8-8K;Y2wv2kHiE5cw1K1z-e;aXIp@KHr+ntKH$Z8ebr6M28e>2usfy)t
za_{q6wGid9PWzUu2sAWgRV3gDn9EAUJo&1vYe*2w=Ucd0NhPw*q#Q+AsSDSRTq;Y&
zv>8dlu&%;*FA-WX2{15XxNNjWS>b9JtqK^9_#DJ{2gT%=nSrkFdGh25<2cYXqVzJ^
za_hM}n3q4ro9{lPSuRLp&4}W;J9io8Gg1iDjpO+EjLOx#`uvN$eCsxU<F7nm^YA_1
zeD8hS-P?ErPwzisoNc)K+H?HOzxfvkZqE6G$2@xXDevYUvvACl0NKhp8dp(iP3;_w
zkyy4s&rW-3(<_D13Q{a8S!u88oSB7;j+yr0l6sy|DPy$+eY?D)elDP+D0aiB82gC!
z8G4cTw~I)<9I1|anxhfDR6wNB9o9a^RzILhk*0o+vl~0AT6Wv7ypkD5ibcS+K%#VY
zS!<QA2!fP$FrwPcdC!KBwNHJhjV#y|RN8~OA5G``N%x6=xZ}|y*ld31`+YM0h>i^J
zzIUI?_4<Q9&XNM8CuQR_rK=3ObWwy|$HuQX-#+28yP&>%M57uU;x8GCcSS1yc>?_~
zup2vuaX{x{#EvBJqztx#x~^!d23H!dSBlyXkFsehCMdCx5tS1EGFz70J><(;a&fmC
zXu!=H=FAzYo{&YPHI5@W@9<v4V@=a=usFmRL*I8S7CY*u-dkX;t1ZJIFGX1xv*z)$
zM0Y8P6k&u-1G^eFeP-A8Tns(4dc~q%aQn}^%*~&_$HVVE=IyV&!;@bP@IC_Cw!i>6
zKwbdL49_E8Zd2=XL|VB6n5?;Z<AhJ&`80K`@s}e%_{MiRfA0~yi%5S7xkrGY1~XS2
zf7bD{Kl{hHaq||7gC$ujzWeT5yz$T9<el%lCxIkO;1uzZtC9Gw3D6ybTFR0-6+JZa
z+Ef4&;h@r-%nmu4A8>ehz~k`=tKli{4?FrDoW7!X{-x)*bN41EH;$OOnt*4_1DE3^
zPuG|1c7vD=n;FZ41x-~8CC3LsX3vPrWGXBb6&AbaS5{awx`4M*gpeE*G0Vo0Nem2l
zJn3Whxqnw2yYyj?HcEF^XR!&<8dEt*k60`S!-%sQ>m!PY%AOp0jt*LGoF1Tcq-_j2
zjU+!{$avp#dVI{~(}y%oO<mWLK9S;8iTj$O@xwvxx$wnu0a}Wwokm#<C)*N5)l?|v
zq!j40#J;E$(0G(G#HdJCkyK*46-MTKwjgE1-w5zNFoqzODQU8@5a48t5z?em5>ZAb
z7q1^xxL!IJk0umZdHfWLm`rIqn?0D7mUFSeR8c(i3Jgf9)Z~@ewBREl1!+ecjjJT{
zHYWj!M(a|s0;3IC%e0gy#GuOi(V7(HN7ic2&(GOxE>S8{Hx*T7$;nVH=iIz^mpAK*
z`SGz#Ed5ARSKNE~CDJgW>xzV-X%?ue;k7S(nJ@maf0bYP&;DafOnmR*Lk=$+E+0SU
zFaMn#vm0-s)c*+-+{l?v?JHUz9O@&#`w6N`N-?|k?X8@7{o14MAMyG_s}aQ-$?v&(
z`r`~=MxoR{A_M<}e@7~7rYu%p|N8YO{uf=u`A2;HJKguk{hN<_zdt(TKll&eum06P
z>{s#0_}z^6-+!OY=0~YB#Gg!+jsiK6bR?(|&p=WmkI&y{-CYm~%-R{1YKp)i5%8rW
zk_ZI~_CwF$M||{HYltcC=M=4>c2HHKV0DG-BtnUaWTQWF(VaSvLhq7p&%~?(X}wj5
zJy$mlBn7UmX*g!C7pO|^_oRQq<!U1)=sB}kE{eN@r|Wy#w&CdL7;7!TdxmXCB%!pX
zZYoZfM=TFKn|06SYRl6Lsp6#p1}`exZ3oT;>iUvRc!T5FO%7jg`I)<)<?{6%Z~Xc@
ztiL-F-u;L-IC9nIoHA6G(41(_4iETqKle*Kzj%So`x_p9_W|7O`T0+MiOM*P)-;vl
zH-76k*^C2b*6_;bUgeMd{4a2N^Njg&ft^*n@ZPtX-D$};0`Gq7p#ZrX2)J8-y#<RS
z%fVsI;qf6&RY`FqWmJ&0&mg|8m38=WU>pX<amD@cm<=A^2C4%%J<z=P@;%OuPFc=o
zI2U;qAJKIk{V*`}p0OX$QPZfIta;Yq>IRge_ggAwY4w6eHAE5?gWYe|1={5z-U$SH
zMhrcskb*LSA&ydnAfb&Vt3M!McU?L70gDDD6}zM~3CTm5&03su)LP@`bIdL>4uSQy
zC!)D^`wsQufH4fR)A51J^K&Yv>32I)TXA`D!Hv@!#rr3cQpSfMrFCOI63|Vr&00$c
zL40>g<*Ss-t~_*frmh^Wu4xww$eE$<q_P(SaTvfVLd>8YDGGUaN`t7Y8mq(!9T4Ka
zl4vkhT4j=9TR2;#6fc-E3TYvOQC0-TO3|B<F?$it*wUj<8Dc^g&n2Uk<mKh5Vpl4W
zr6QS1dp2hX6fuaFSh1KBF(ir0L7}biNXVI}q!nw7#uo!<v<OhNt060fVc_E8LULEF
z!a9wy5VNLUF1Y*YPw|zbBi?xX9ggR>$lCDe(PQ2@Kj+W<_dbhK7Ht$pImV&GwG}`0
z3xATo^Pl{6Qa^&%JbwI)xfMK1TGDO)1YiI5_qn{-(%PDsBCfV94(1#zm(*27TRGaM
z#%b{^R2pFuj{&?V3<JY%$2bmhY{NhZBPn_&wy3SF*tlqoE=E%0FmSP25o0XMWXM@K
zcGC7&G_|3rEp=OC+ZLl0F$6aKz&d0WXLq<50`EO~%#-sKf9}5vbp!wT|GSuv0!Ck}
zfS2PrDfmRRVrbiz`Fu{>wm`958HZxXS>pL(AO^|!%}FY=sqj*-72JQQa`!KrF!zNQ
z;1_@KN|^u2_?LI=E<FB+5&{H5WeU9}OLacwz@$b^c(P7hY|in?W1C_tQL?CH5uJ2~
z@^vB>4t8mH6%KJ>vRoPeT7xxa&w9G<UN>>(<hxup)_PB)*e^AxeQQ=a$qJ)1))=~>
zqbIZLxA^7&s~nAK@m0XuOD@MA?<2-YZ=oMYQi_b;(=;tj)rwMIX<VgcjZ$EAM(4o1
zQ8-i4G@wjR*G0USc&q4Ph~S6J&_`m}(a)c8(kwV_7M#7hq>s0`IN5P|w&U6RJ2@ni
z9mvFGn}A&@RUItnym<ExUwHL19EM}wdvc$r<71X~$>(1G993lrab!1cc<z<E1T?e5
zBR=z)&+@rXeU8K9W16<bRu!~~C;dYn?%wCU)kE}&fH`vpdI^UwEIBz@aCop_UbWO%
zbU->Og<l?mmwaAE%q(MDO-h<!jJTL^N?~im$!W{!*^;`^L_IKg&-wNl<KPKCk^&ed
z_BVUFk9HVS$y%<>RYtv%c{`HGc*q%#CubPdfI<;eWFVAIvBnmnKvY*740DZJ`s0p|
zC!Ww(m9bRRd{^zxa-Zv-L)K6?GnVrizT2^|mTES~SVvXOFt(<@Jm>IeNvsWCDXPlg
zin-y#`}fgVg630<G)*OOVMgOrMNUzA_aCqnx&At?hO`(hSt)$-qK4#3M38eNj1g-L
zP19nOAtxAiJCsV2k(v-okf<fg6;0QV%;qgA^wNrJYRocF-RM!nuwPJ3)rsiC3fN6b
z0`(LTMRDn#I0wZ^Thn86B%?`LcobSGh$52Ey0qG|*pOH)=?tJsw(AuDS4Hh=T`%4I
zASz%VJPIW()T$zRi+`{b?>1AagvF90C5d25!Bf=@tJR9G+mceCX=<9f62(^3B(0ep
z9dU5$7P~m`xi9@Jn|?rP&G+7VljH{o4&^G8wK%;XhJo8JzsB8{U*oM``(;dR$iu*m
zm+zp>BYc00M{oTf_>cah|AOy)^*!dPhMedasFy7#M@Q7PW4V}dcD!IQx7@mQ$noid
zdF@!bIkT$4+6t!>mC9&5cGwWsPv|!*;>9y|7f<NA3mTi4g^{BevE9J4$B+5iZ+}Aq
z&WA`pDnh8}I`J=T8pGLf&E4AvoZdQNar+K+ZP;usdH2Z$-@WKL`t(opt@ED0_iJC_
zYu~x#AO3pg^d|f-{}(V0V3pbnKpI!ETrN30Ip+4gyFCBG3*5SOi?bUyXl~!;=-?g)
zM@K~<<JqiNY`0sI@9DcOmzS4pHXBx}l^m<S=gGxW-unLc_}=&4X0=)o!WD{3>8gqy
z<G(x+XMgE039$QQ{L4SWR_r>ik4d1Aaw?P*T{f)&R3v6kEWmEHJtyf%-PTmDLb3}c
zJCaBc>*exY7W7sY;E;vVl1mFuK%K@~BbnrTa3&M=1nOi6zgx0sC)493k(BDH{rW+{
zg;02MO!yesbvycIWJXJ+8XA+R>V^<E3?n#~%Z5z=ZVZX#L4(ybTEWPGv6{L~g+iPN
z!xm#Li^j2>&8c+cVh!6avD*o_%L%g2#1Ss~%=UamH|vOpJMNvl$cvx7&B^VK$1j|-
z{Vh*;6l8-~?ccg=yxV<qhm+GoUVZK*UcL1aJk~rNpO7vSb!xbIatG}c7n^e)UOwgc
z)-4u`V{YEK%jZA)IX?CLr)cL3tgSHFGBi70x_p6mFW%-i?|+T@j-t5*^%<B$#q%$p
za(Z}!WxYiEj2<&)%qX2mDlo*3%`h?yJ=?Iw9o5*m!?=d%1I?(=IZ-*s(a|Ao<A`dc
z3j^C>%j)u6Xi^1wHC2OA;)N&_0gF)rrHqjX(~!jkam1I+g`6QH{gq9bRz(w{_Q0;f
zQDJ0lFXxFXzKb7c{6rPI4_ZN#+y0lbIw(q>ADY_L;DZQW>XxIkTc{k#L(i~T(>hIx
zJt=u+^9GdzyJzoldGVB^d5bLp&~;slifC$~YQ1+KpTOO9Z@3SPealrynQHQfQ<8lP
zkV0S#32QCJIhvVp!IB@)WWv~kR>S~pTceF;7<<}TMT`TQMB7x%K6k;%zxS1&<fHvV
zALBcRJorEUDU8<QgI5w6a<N`XIYILDeP*?~VDtmdIhv-Ss>C&2S%-EOTh-is@g?4U
z_#WTAe;;ien|jB5F=rfms@bdvI~7?QvWm3Fr+n!b{y5+J%CAwij${?*yB#q_hV72+
z#RW(8oCDorN6Y1AjdB$^RP1`iMcDBkD_jj#oB8+u=YN^)xZ!dL{FCp$$ydMn4I10f
zRxP*A&Nw|j;O5yOr>8eLdG0liUw1q=GsJX>fBFL+{noEBJbHx1lg2;}iJTNM8bVME
z9(u1xNn^8cAB{7(%7THU6p}<2$o+=(gCF2`irDQ6ijs-2wwA`V+`4&_pZcjU^0Pnt
zvwZF|pXK=%p6BH3jM;KdJDXvRc=Y9*QN<LTvU!kRz9M-~^nuAyKTn>MIkDcZSzWI9
z&YR!im;akz<~P3b6~6t(x7lttjH8#`ej$}hK)NcaCsR%0lkv&;gB^R*Z=EGsNo7!(
zq$6a{7<>BM;j5m<t4HXwio=5?4HZ|*ql}S2!0#(}S}P)nF?w<+{-arZN~;E;|JVvQ
zE#&(q8l|yn_JGbK6+~4E20r*=axZk255$b2uvlh`8JC+&`mv|$ckI|u<8U};w9IF7
zQaGpELZ@KfbSNXjKj%bspLC=fIxa4skxk(Eu%#|yk7v(TbeoQdqH>1YM+clOAw&^^
z4L&jiY3=&avwa3np1{M$BM;Na+sEfT|J-fPPH*tqORsY4Q@42I>)+zp`#ZMJWDT_z
z>I$kBTm}6l+@78A`ps83tIpUBJDRSh?rOH1j(_|Qe-*DigBr0XjvKdbar@2-yngR>
ze)jXfz|Hz5MiptSf}L4$=kO+X@0{}D^D}OI{($OCp%1~GDxO>1rWxi`>zc~6BqH0m
zWjk(It)KDe`~e|HOk=t6;tdYZ4hRIgp(8x)aLpW*!5KrVYq8H7F4>J+V)mpQXqyIS
zEzu{O&a|~dVYuA(7-NWqsx*eoZdkLkopfIE)M?Rsj;K`o6gf-s#MC94g6ech9iA$?
z5zpj@_~VY>6~}I>*y++<1R?pWlqJ_KD{>S9y2;>oTUP5!hV6#4lVc1CYlX95oyAzg
z7)Ii-W9+uf+M3F0cDpUh#iGQ8fi;$pre0E6a3(I@hj@P<s5q4VlcYXe8JHQRP$@Hv
zfsiuMCdnABYHVH8`A*93Iir#z1*n@Dwlef%2h33<DyNvu8?=rn1-suqVSINkpOG<Q
zilhL<b|YoUL6z5R$hq+Nq9!MeQI-4*O~_j8Hc~E|OtE88CX-dE`jOBoV0A_jg_^1I
z-Xkv;b>{>Oa!lkXvP`8j%Rl|ToKPtfQ@mQmCv~NQEHf+X&4$a%b8?O}&8%dyy7HL|
zCnamcORvAqSHAvReCO@AIXOL}tu&wh{AUPZAZMwRqAcQtC_wV(e&Ju`@BY>QE5>AI
zP0hV~H!yn5rrYx1t?zK;G-}uq2ZzxK?>$D%1XLEs?xFAbrC<Cd4%$P0^K0Ki8F=l5
zPxIgh7kuNZ-@>Sh@Ba2fG#O1G=YhJZXlE^FcaFGux8dLaAO2aIvj;r<_V;<w328cz
z2`LkzB1FY7D#oDD+Fn`NXlY@i<^Db;Mn9tCK$Qkg7l)iK7c|uhtvvJjjQ`#L_CMrH
zU;cT%{N*omcy@v{7D`$MlSoL*;}_LT(KzVBdyHr!_6goctaVr`Tus^J$nkC#bIxwu
z<b~&7<QM<cpT>L7TW`I^KmDiwlwbYT|1V$p%5U<{yZ42=dF|S{Cc2eN>31`6d@?>6
zA7g|h6)l%d5qp)Ji9{qq?CJQJhtJ+(qk5L}8TGO)e6~p(rS>TkV~Y4ZFb0Vs?8lMO
z2V%OKqOB1cOWV$Irp6TSu@7=~_5N!`M(51PC>*w!2q8+Sd%p6`qnNEkk~rT*kZ4Fn
z=mY&`OKcZ7D+~TXHK(l%7ZKKLxOJAPtl85w#=a-Vfv)S=Twa1sC@i&|(byT=#xwLg
zCIWRo^k^equ(Mgsp*D2=z&MU1A9?l8A=^H1vD)zD*@{PxGjHC1!sGQNXGiaG>-Z+0
z|MClLU)XW+@RIe@HMOyrT7!ejiyfhZYA`gRMtgYr_!*n+232e3hf8RbP>IOI!BgQl
zY)`p)`~oNS?UD&T$;TqC(m0=}Lgx13lpC`XT&*!q+S#+QC3TT*x29KHlF1~O(TAC8
z=2$Kdz=CTmvx6DQL^kUU>&r{J-G;@<jXkZo3!XY^LYDO3F(7FmizUt&)|YEs>re(_
z42<-I{*pWj=np9fnAf0;!YGTKJYNxAf-DUJ>L_A~$%<!U-%$iUA$zuL8R$ROHS@bO
zD25-dj-?DK?O$dJi8Kc44CG9uGOMQ#=*}N;@Agep^o+hI=aCpkQjQ!fYBqg`)rz)h
zSU=lJTvLp=+M$Z48Va$QD7R@$t>PaV<+ZV|VwCD)232xfQj}u3HWq6oyD!F?<VTE-
z`0PnZ6k5yWlCE2kbHJ!XQyFG$L(U@2+x_M(*8lZg$<K3=c#(C9m5CCl?L#C+O^h{)
zB`GZEcZR_`&bx}s0oFs}qEB4*o*_g`0dAJW@v@Q(K$|$6D~^_6(gm|>q^$#n0hI&R
z=#ob$K)^W*A&~}897eMC%zyr|%vU+1m1N!e7)onGO2;`xVhpU;7xY7iF>*g+<R-Qo
z4r?4K!_C`w_@#g2-{AiDpVAEjU;KsF=*NzJ=vZVgA<$ZpOGK9%25!9YBCovs3Xi_^
zRdNkZN!!`?9e?|8{J(gAT=4XXr>Tz^oMteVRJZgh<C(Z9iD)vv^^LD`_ug$*tEc?-
z*M5Va{`{x;-Z$T*+j>+M@}7@{B$73|t>*nF=kUfe{^Za64DP~n|Nb+cd+rulTSlcA
zLIAIc!Qw+ECL<ZdN};qB9$eY5geY9lq3`hh4zpX)94+{zKmCjR#eerRJonuHnS+DK
z|N4LPzen3v)OabPj2JraiQJ_sp1D%Gir-<OLl`AdXUfpb7ECc}YEq~JSc6WHkUe!X
zqXwe){PdT9hM)SWFY`bCpZrhx!S~<hH-6(Q{Nw-Tm-&@n`6oPl_&`YN#h5uk;SYT>
zJ{f;7Bj!X%39G8YIh`yZA|dsx`wJdjKHxh~-o#f4GZVXsX}Oddh}aPuBzH3>@X<2{
zFBR#%m`*mZN(;DiuEJ<rmS4lPx5->jt(jW0a?OQQTE^vdW;Jo|a%l|1bOkBI3sXGp
z0z=>9>z)dOQI1x%EZdf6uIFjz>9&#CtfFozj8S+$%4TjDL?o;&T2&Y=En1YrIEU4i
zS=-X>22esJsvBun=M2$D#&Kj5cbKYTdDvoGL$gTq{SL**b{KhfsX1*I%nuE1b;!v<
zLmY%f7GowF=*Eui`HsgA9y8n<*zC6SaU@v*TrqpH%0eA7j+t#ZoF8#=c!tx}HRv44
zbaY&>>7UVW*W5fl<#c%hPBL-BHnPb(i5^Rd<>8XHu1PNAHH^VvwLx2js~x6MT%2EU
zxn9%lI^m_OjIt${y|!p)(Nw4uNaIM=%yF(k0X{?qKT3BaL<ZlXyh2%v)-vxqZE@CO
zjl(GsiRcofkW)e{vC`57*j@27GYQYYKw>0hPoD=ievNk@;wStL<97+{)DHu@0+ckx
zG6N+Bx*?$>xRl7rgAc5qJfu<yt6>;-pdz-m5H<aFL*pD)8-{LRxtLE!TgviAi&`Wl
zRE+4P#jrT%60IfX#L^Zn`LP)&;)$HDg7#F-`^uuq^7H7u__-N{aTPv{7^R3|L>q-R
ziKD{>4<0_j2S9lmn`tV`6s{`Dt<vT#FF<y4BWbrnWlM;T(K|w%kvL|xu6gJFmPhAn
z&g(gA3|`f^*%5sWoemhIAdSSaXR*qt-3CI(a;7+0WbT|aoGlH9ZO^g+XTbZ8%39e?
zAQKM;a0(q%X|-KdKQs9X&|N#eDFym|$8NVF#z0lIGBN2wBDDssYRR9iYLXpz<x`*L
z)|qG1?Re!=w|VcKw{cY^7jBN&A~DTC(h4_g`25d&nTOx`Z8p0t!>%XiOkGv{g+KeR
z^MDin%3t{#y#IJbXW;1G9io=fy>g|so;=UGr@a3BJ#M{llNYKx*!nxd%W1&(0kW3K
zC`rY5Oqo0=Y~5hZl5c(M`+WIi&UV)kQe+5;Awcgnc`%GYs&z4g4v8E~(uJ$RH4M6D
zNQxCT8+Sl;=YW@g>Ss89_4BkZd=`6pkNM)S62ep7{N~rO^%9JtZW`Ko3)*0urE<~^
zGNx4PP*<Q0xgG$Fb>-LzNk}VmRjEZ<LP{yI<|#wFLX=U7qr+1Uzj(~2KJ^*?;$Qsl
z^Tr$B;vf8jf53n7U;bBo|9fw-+jUHt*<Abglkv&;{TwkRd`eViV(S-Zl>;GmTyCH7
z{?q%s@y>T}cfmE4M0J%~h_tm787Gz{%7RkFBrV8PJV0~le@!+S#we<)z6R`cNt?X_
zb`ooLwatlH?4eSLD-^@|No;Xtlrp_0h4|DNV+yeA80K3lYjE1q(sHy|@}wC^e#QDS
zadxw&ZRb=@bGd#-NFy-@Y4KT$F%?=>C>$v(tgD!{a}Ey<8NyR?QYd3^m19=V@IKJ*
zdPW~Oe|paRU`e}ZIXXV#<YwY>mFc!UyPfCh>KQuq9JdD?o*rQjkJwyn=sQop3(PCc
zrHZUqYku&9ce(t`icP;|NF&iEya~iOVywipwN=fsnQ?S*$kEX;#gd<xBaxB7GoGgV
zoc9mejcd-%j#$o?jM~$OfzkJDhD)LvaE<2X(FwDrC8$7`5_$w3gqN=!WR-aI_%U54
z{P)^wd=5BUp(=xE9J&(25gjYgma3kibwvoCVGO9$5q%(rNKgqAEY1oZshq<ZM`bFB
zhNCLSyU^8)Ksi@r5)@UnROcdrz%FkY=@`>M7dv*t4zE96j^D>NDCPs4HqgfGL+j-(
zB}E}eNFJR6v#O?w8LcZ6Rw_b36(VgVF;-PG6W<?KQUXGtwnmhUAz_t~T~JDxN)c5g
z8e1}76eeX%*5s5WdBITHRS}(2u{@EbD8?Y_(wxXqP9m%1WG6*(97#E#ak$!&tD12f
z2r<ah<&4#ygE@SEi0t~B<y^7Quyj}Qcat49*Pt^|i>F-(ia3l|6&VLjP6zb<jMeU#
z@4PqgsE5JbW;0LtgT&(WkkiEx|JuLyCwcGP760fT{bQ6?{NgYEDZcUb-{wnSe2qu<
zAMx8?dxIx^<l%b5cb{BxqltX!<r~~N$W+x++C{*`M3oa+CA?1L+F*@FCyN<JCe2rd
z(o&V2R_%r4NfNt(ZntABJ49z}p#hg^N?EL{uu5ZmX0(<R6Z6H4PAM>sli2~rI8mRa
zfC>p671~*1wzy<@@#lY@;cxxtj28oAm>l0T<IwTSmp{vmJI`}42cHxF_W$TF;g#o~
z{l-^#;rSO?t<Ldd&+$>i$<cxo164KS(Zff)^!)40+M4qx8)5*hD*PBRx)Luw>uG0}
z-VZ!}@R0fQFETRVjc4$IajV#EEQE$Wf>#l8!Wu20*;I^G!^3QOZ{=BeSot~jrJqGF
z4*8Yc!0^>K(ck<&Lk#?{{=-Mi<~{$n|HuD@oE<qQtTi;XL?JDgOOB6ja&~%)8#iup
zdUndqn>T5j2J0&HRJ{|QA#wz5RIwtG>s29b+?=Ox)PAxlTOP@rOqr}SwrV)LHDh^r
z$j|-U&+(W4@?Ylf|9}1-|Ji@~pYe^af1S;Cht!M<7%%jOiHA2WTi298pNvn&Pdq{h
z_;CQ!fR@NERoY%r1vcH9r{_<2{PYPgzjg=bY+=9YGBf4<xjt3#bV*H0Nvi6>%h@^t
zRtaycGKQvYs7#G12FC&`KUlAxl4k>nSQP1!Lrv)r!8Ef*uEY}xg|&{_R*ZUJ%su_C
z=hB{|W2Ro#SZdBr?{K+;_wKKF{0z>vnZ<#nogK*M<2(G=(T`p6OO<|dB#P?+8m$ag
zTdJz2-*%7^m904BkY?F%G&|<zje#c*SG@cFdpvmjjBbSEQ#iY^<i@Qd+NPC!?<Xss
ztRHi^S#jK)Vr$E8v*pR<1+_WgwJ$w~-8wG)CEs}K4K7z(Hsh8|rfwW5iD#^>G_!fl
z>5XF!Pg?4ECNi?+oO2%WjE8JF=i86J$z^lN!OL^j>5}g~eUoT1rnVfP978+DG%d4v
z#Zh}mN6&eG!D_g~6R0Wy&M4?MJ03q6scfQYDBZh+(F&`@^uO!HAUftm-Oj}3Wb6sv
z6GLP>bObMnLG64-Q#Dl83|E!(iV_j#iVy<W^D8H&rP`l4fsRpt90OzO*^X;=!<I3O
zk~9V6`r}<QAG>1r!IOHLys!U$x|e(v5kQkwp%WO5GENHVO4EmcvvQGPbV)uCb4G2g
zBp;+CoPv}Yb{%;DWeuioP^K9ADz#5UP|DzxCZ;SIda2O2CQw>9W@Jc{A*|9;A;}r<
zM?#R;D6ML0R}t$OecB-+-H0Ctnx>*|8ZIs_na^sp@+2Q6dkRWo1tpm|KsIP^Ahg7B
z&Tcznh|lrXeaHRFhMl{`rh!%7^4d?m#=Tc=@cA!(p3P=p9N^_o-R0MQv*qsH+x*PW
zf0@}~%bhz196#6c{Acd)=>D4T{q|erb;tV2L;l%YTkajrc=_JTw83%G>`3{XrU9KR
zNOMv9s4coyS3JplrTh}-b0gL$5osTL&y<0yi}CUV?x!qOF=aL=V^CO(t2sP5V>gae
zb<3vfsL~9rA~^;O)xM=}mFD&<uh1SI(qBBqSvgHLihgs+tZ6tnI_B|a!@Q|k&KG?C
zr$5D?`PcuX1TCk?(a{kP9zNj7<45FB@trr`;R~Pp0*m<}ufFmkDZsO*8?3dA<C>Tv
zwo)9Q9C3bmNr;*E?ytypPO=)G2SVt{K2cGTlp<zNpCidau!_MMF2^n3di#>yyPAA>
z2igP7{D`m{*{&ZGlqjZ#VJnMMInN$Hejo<LS`@LhtEei+=5oWs_n)HmcQHmB@DC3U
zI6J$+-Me?WdGjVmM@JkUA2VAl3aE=>TT`yklp)7boi=9Q>X6@;Wlv?aF6k!PqBPV^
z4KecCYp?Ob^Y{1*fBw(&_x|4B<!}Ctzrk0&`VIQglQI}3sU6oynICsr`DA=De&on8
zl7f)fb%<i_ZX(8Ldf(HJJ3{KITg~FA#a8ATr|qgGIw@HVMr@^&E?l@wS*OL=Rtv$s
zb~UxFu}b0wMO9l&hp(OE6C-M}W0^eXEM~vL=JHC+2UbvtNpn?&^A&2)<i4Zp)>xyc
z+8Hxb(U>`>%Tw;$S@Fg<o^b!2r;L8Yr_Ax9qOKgVsi}sVF$F@1?AXcs%MoiUbeY2e
zhH)U}jN7bP9xl;JGc%3MtIr*9a&pLOd&$Lig%3NrzGK&SkQH%EL?6i`<RNhX;Yfc8
z{U$MNB5Jkbc(GwY!y!j}{jIOD^e4j4HImBYstV)8n#Hw_A@BIX!#BAYpD~7!!EJcl
zz0bq#dpsH6=l<>;IyR)rvpk*QP7TY&g86(-)6RIh-C-(2ZVk^^u^YE+{FX6!v@=4@
zNeMrCE-zM`9x0ryv05y<w9>?s$bN*Z_5vBD6>Z&M4TKm`I%1oOaVwc<T4{{3v{fsy
zq*xq|hE`gw<t({MD-paj`PW4PK6`etqxU<8Ffhh}A&9qPN)fFUv)K%zr%>J>Z~U&s
z6Vq9`Ct6RWZjdwtl}QGIk?+TxNd+wAx+0belGYjPG^(mdQDR+^29mgekHbLNjC8{w
zi%#1xUo5e84bFkJBrQFDoy$iW0-+zs(U*87CHz07rJ`4igGC`5P+9_aV~+SiEG!HP
zrwxsB_`%}KLab7xZI>*U3(n8i%ubJ}XLE9~qytDgk!&KxfjD|{un=lOzoZX$d3IUx
z?e`<^_Z4fiz#bf<j|{)?Ge5`o-~BGP@4dj&r&}&B*1Y@PBdpP^*Oxqa2!H*r|4ojM
zk8ti)PzT((cf!rP@Y-iy!6d^k|NVc+<L`fuZ*DGm@=oRpcVA)b-{$tRhjEKb5#tS}
zs*4eFiM1}fSYrjO#T+r#qLT<yy!U0|V(-s4`3M5iikvFAc@Gao5^fdRDO}c!>%{0i
zbJw8Ag%>An#)RVN#w~8#y~pJbzK7Onf4+^IHOr=Ewp<YHj?FOgtH1W^y!yEpXl9m!
z!<OA{L!u{yfq(wnZ}77}^Yf_OaBz5t_XB_Wm;NMw?$7@@w%dWnj~?^Ozw#?+rTE<E
zK1<_T{*%A?x7hTCN9U1b4seSlL+bIND{`_x2p+6P%^bO|SnI@t+_T+miArM*7A(=U
zrlQ;R>^1{QSt+-NkrYHlJuL!#-xGpiHmgK+n-nn_ya(%)@I9-Fwry#emKY;%zx_5p
z_`wgT>l#-}b@{c|UgzZagrlQloU6t4UW;{)x>i~Lz;!%*xN2(IiWtKfD>kfY#8ox-
zUU-3j=imMJ_!EEPPw=<@&fn#)|HprWH@^KHsmT@d@9T@!C*zaxdooIAMI0Ao*;Pse
zU;VJ78@7ybz_kX~IE=kg!d~aWiNH*iNGBu7Iw=uDCW<Bur}lT3I4or`#dGw^>g1|N
zQHT#r4A@LAwk(<SA!mjO+)eouS;^m8W2DMxEJI3+qo*GRy1v8B15VADw@c1$+~m#r
z37b_w=`F??Dtp4BwOCW*T1!j<KNE=%gH+KC<eZ36k~d<?q+wvQ+0ZXLT0Pryz?w?4
z&^6h#C>KaHbYo9UiBt+DzW1at5&MX;k&pvcK~OOI$T~%Q$oQZ+c277uxk+2Ms8*A9
z!k4yATB8F!k1iguiEFBhib3~WgeN@M+~?u$0c(4SJ|3}4sFxOVs?fDUx1gGgt`fPA
zgllO_qY8u^i7Ajn#%PBhJ$={F?RGeWc$}idRU<~CDqPI6P8luk1I{_L%3?7SA;wIK
z8Rs08tC_Ve&Qv&4VU43!k}+{b0=O!YP+Iblh)*M98W_?*7kBhy#}G!wFc4GRS92@p
zIH-?s`YFGEz^+&`TzR8iRe7%F@hbyIiL?qtm2s6prwW&6qAZObPi9sbw1LWLtOZm?
zhfF<d<x#HLI^>kG8N%op29NFnu5B@k7G^ccNozJERkdN)G46J9K^IwHicuE3Ys%S)
zl%C0GYjL@Tl<{L`96c&SQ;U*l9D6{KVnV0He6irc!xgLbhS~99Nsf?DA}Pd>FnD6`
zNuxpq#V{W7&iNs4J>76wHH<ebdZGEHfBTQ~kACG>x%K=BXSa^|>TiCVH@@{fE>~+}
zlrldGHXBb&52@>lU;nkQF`Kt64(ELK3!ma=f9~_FFMEFaPyZs1-g}MT{I#$1=-G~M
zyyN-Q-J8UVOK#0Q2aST%V>M)(Nw#9YqewzC7AEl(abMaN62rA;8ecy>lq6VaZ9p5T
z*eQcj(&9sbA4clJ{n~6dEEfmjN2n$0Diw~H!Sd4UpXL3p{xfnC<7$-?!^JbI6gWOU
zB)TVrl=$`=Z}DIKm;Zo&^WXaO<$MWj&o9v?ar^ce7Z>Lo9vrgkw!HoJ_gF3u_`(;y
z$jh%j$HNCR{*_<&Jo9-=-}k)tgZDAD#$P(l*O`l5rkWqJ^N$(hD9-ezLe++-6_+`1
zJ`MzD=?tuuAt*;{=S0=;@O+I=n%+lt-N4Xygt5b_#H^|Iz-}Bya;`}cj*pIc<(1cX
z>E&0sedi7b%LC@~nW(dB*?nKX-bNp|ynM#r{|A4IuYUD6NC}>M?jE22{1<uo<ySd9
zJre`XvM60w_T_!0E$_LZ!kv|naWOuvNFmC^-x^;3)TcQ;yTQ->D}Ri?_SgOzfBWzJ
z9oE~9arFDg{itgGC*zaxlZ>2%6MYqASel|LF^oI9v6H;jS%s~OLF}H@bWO@l5i6ht
ztSKcaMnPJ!s@PGK=x8AC6^r`Xb#vuISJsSu{jTt-%j-?%8edu7$+czvN>E{p5f!_0
zbUEXFWavkB-Hvt^aA=xZ)a=dSx7j@ntinjjPnpds6fKLkX69yCt?1HD+$BP|_L{`&
zQoaEn0z0;Jp`&tFwlD^X)ig%<;`MyNK|Z+R2}bC4J$@9U_kP>a9%Odw#O5+{S;6=W
z{NNe;4VUdD?x5m8A5irb@iK_3gHvQ3*==?_xOgA75nU%@J@C%z2Rzt4<UBkhpD6CV
za?Ik^g85NPJvYLmw*hZQoYn+25>rAYiNy>m2~{gca>(E;<1nz@Y#F)^XM`R!9qbgm
z97koaNIC^p;y;Zw1@CxbOoTC#lN6R_^EvZoj<c?SJI4g;^tH1h78@LK){xYbG4>3x
zr}H~@X~*bCX}!jPQX1zfs;XvIw;a_cIJJh4p^W|bt=PSX-L>gEU!A&gAMsnxWwCIz
zW8OCZKlc7J*w!q)@56p8y<vxw?z#DP_wC#>Jw4Nd>FJrC91ti3f)HuAD1($;VzFe|
zDt)m{QDv8C7ir2YiQ)$pOI0>a+hvz2h#UkG20#ED0F!f1_jHaohm+3^Z&=A6*4q1=
zo@oHnkN~oXyKe1!!ro_xH?3#=pMRhpkD2Xj`gTs=HAo4glk1Gno`Pn|eESB|$st3N
z2;{gPPoNqOs7E7K#$$Zf;rpKXY(_Vo;jKkQmyC=DRWjL)foL6l*W&s<DME9kSZaNx
zow{fxN|j~tBK6sY(&(ZfhDhJGw5~;~WZ|KWp?7`ux9h2^iU{l<&N#8Mn(Ruf&pXu9
zxgO<wGXCu<q}brvp5VD#Ep0ty-*kNZr$5a4-KRO69rEXY>PLC{E6?)GH(y}yupvr?
z5Cz@`q?EYcB1K8xbtq9`y9v?F*xB7;XYVF&UU`kRjZGf7{~->YXC{ZJQNi~97Gf*#
z;K>WPy-TPQj$t_gH=!tNgsHzzB$2XgLx{Q9$_B;WFZcKHB0uC6-Gb0BlujudQY7!G
z?0dMkvx688iNVm!+Ei$h5Ks_FS*uEjp1UqS$ae?BBzVX|7;RhTyE~jZeS)Z8M+Hsv
zf^U5FMfy(gp{E|>)R{5$K+^Sd&Yj<6e{ag0mv1m0)vRx9uz7Nm!|4IafQ>Cu*X&QW
z8IQ({S4UJ;$zZ7Q&T{ki9_yQDnccj}e$xQcKrFvOUDJ8Pj%(Pljy(Z$p*Wm%SS69;
z0a6Z`&LlfGZ!n!TOr~?XHbLu^DbON0!tcaD-vyxH!nwQn=*K?Gqwjf~d+&RIsvaVR
z&eVd`9?B=z^86FWkkOUZAyr*+>-Ke;dCQe6m-yP(zQXBKXL-+K@8#X^dN=pq{}6+E
zoV`oZ$(ww)Ac~kRQ<6ei2!+%MTR?&;6w*7K?GaM4dGaJ57>&69{`>gECqBXd_5b#d
z`0g{$5@N(UpIhR03cEYwe}oYN_<#!JcoCoHm_QIo1g1;Pcx8+z7mNF&OxQ(3Hswy(
z)KHCy3)gjlwuP=sg<X*Y<dr%)&lY^RMH}}xa&p`y@964zG#@(W@Wv5vKymC5EQJ_S
zDYQ;B3nDlNEe<`KF&H<P!lG2g%3#d-bEmnyzeTgxv%4E{;X0GEHS22wR#!)iMis=|
z3f6|vXq@@vB3YV5Fojtf2~Vdp<Tx26D<Lr$+P+~rIpA<|NNrbHSsPQSDkYj~!--Pj
zu(V;$?c297wV@hWs<EXgEw_5G^Hh*en6f|Lf#4__L)n`AIU`m&x@fUwhc|P)ZfL^+
zH{B)ts%2xnU~pH-$~^<>wGv%R4!c8)Fi5R1$}x!(8j`=PmWF_5+Dz%X9%5v5Wr&iJ
z1Lx?PIl)2GCG%hb7!BSf*N9|4lJeE1U+mDdhz{Q2v(U!qf~p!&S3^u$Q|p>irYzVL
zU$@x93xv+%=u_*Jq{MFfj?Q(=?Tj|ev3Xuaq?pq2aFwARGGN4*F@_uO=#+WKg<S~8
ztwfF&bMe^k63LznNlO8??<j;yVub*`^F(iHx{k5!neFe<%w_~<5i0Q?q}Ie3sYe53
zq4CaPGY@Mx9Abo|X<ORqj6w*ENEw1s$^<CMMKKZ(N+E^7D2>JC=q0x#l*gC~r$TEA
zrI6a7i<B4ZtflRHs$!VhV11jExsc-0ijw`EeQc2Q&QpgYqstf~GM0E>fIiKY9m7kz
za5xyCPptCDBll6S-^IaU;L>Z?`R!-E%Jv(#sAPpM3TAyW09Hy7LPV(oP!wheQDanr
z&4~`d3ifvP5i0P)OH)=>Pw<(ad<^3h(`&c5@zNVyp6u}8dd(X*zsbe3EtTl8!cv;`
z#gV)$)P&56*cg`=t;N1S+HSeXShQk=z=eP`1&cMd5J}^ri5%|UMjlAU=N{r9pL~LV
zh{-H8NP%;aLKdu_Jj-x2#<7!*s^AfA!~Tt{oI1UUmx+5Ry{2y!U;E0lJpbHtJofkl
zeCD%HF&r6g?OwrJizy>zU9q=+Kv7ndRayX4RYli(lo~LdI$nL{3X{V*B35W=xqjs`
zAHH}G&ph)*Ztu?6?L9kP&z|rcipT`P%vwT_cncyp`pE(Nvj%S?)+MVDDFg~ZbUm&2
zR5F=X5}~eYe&OHytGw_1Pf=6@q%x^tlP|vc!!Lyoe~|fh)6Ci3-3CPF*hwy5eVxl!
zUgwKn`aJjCb1xtKz*D^EJ?~|GW8+w>Ah#t{imZ#oWg(gFlc`Os6q*7dl3Deju8?A#
zpZ(cC&%O8F%Rl+WU*vPY^earJQ@l?csD)zp&bTxF*du`Oi?FVRY&G$$oq(tWvM^Mm
zY~W2g_8ki@Q*M7@nKP&NJ&kK<x(**BK9D$e4*C|PCU+ymF%2%HWl-R#+<W^W#p2wR
zOLJu}JWe^p(vEgPq)$98qmqb18-?{A>m0M$jKOSzQNzSGl7`b~Hn??bpSJCZdoa14
zyzF*Xd)C$tIeDU{C?!T~j5cU(65U9o>#z=zYNK2($d#1zy(PGGE%ma+x^6ii0(ct$
zN01Snxrriz502@4LU3u{C?l{DL;*r4YmnDpxy|(T-=&@#im7IWF_kH?(qXlyIzOU3
ztzj@Hh#8D}^kzg2L9YwUx<Zcx!gzuJRSJgGRH8se*lP}HLr0|sXj9+`v@Y4ENd$dw
zX__g`d`b)+trFkA?^6LC0byGT6)7|rG>gJKIb`J4dL$!hhCwl;s)v+ij&sW#%*8%E
zex&%EN$jxL(9^k&wr>+c=X-kJWs5F_)&*rzvsSM$lw&GUQ<4hmKha1r79TktU(I~q
z{Y!tiWE-M{Ku3XXT3Q#Vl%gK5GOA0qw{Nkxz0I%?G}8m3ca$m>S4Jzm^F$$tT4Dw@
zq3;MWFr6M!YQt!KjGqX)*^GWVMfia75mi;7HQE@gmc-5xA@RJFL<)ly0_$ALiOtx%
z^Nye$vZ^SIPSIzB8ahem5^=XMhO#Vh^M)8bWmQrS23)(g56WbmGK(idAb3IOG`1gd
z7}j`WD%mzQ*X*2s=OaJPt%ErSQ~2VSzQx|wHhwxq>VS(=M(eAbT3e^LJ@eU|*=$Z#
z4v`Y(GmB6a9)(wm@#;Fmapb9wzlV=}{BfYg6m?pKc*k`69=^AJiusL0wy!+TdUZcn
zZ#~OB=Sy(lY|p~nS><-N$dyGBLS$3pypM?)goC#jRSO~Vs<39^AD{%N=+Ro@=Lg){
z-(z(4BJ<gV5RxxnN;*&}w=pKl>FUYT3`T2o)2*~kV<5J34sPAx+~e<}9%;I*1|?+T
z$%Mf6){Jj`{dsQR+UBQ!=0_O~*Lm%Y7rFQ1eGm%zKA?3))ATsoQlH-7wO4NP<{MYI
zabt^@o_mexB}O*%hqrk3`5lH^h8JJiV()OyULR@8ifyOZ>3R-bptTS^s0f9yNToA3
z)T0;MF-DXW3)B4Mw3B%rRaNoG!;hjw0Z}D$V3r_fp2z<1NEsm|UE494Oel(iq9`(~
zpNj3B-5EQ(H+bpA=Qw-zJWoFPVV->QLu{TnnZ-lNK$Bb$=5n1S^D(~}p%gm%G2y|R
zgwDP1eedP;>C@bM@m~JnfBWBZ`}S?zqTO+4+!_BvMqE$^7AJ$u5Oxk^lnQAKMO~vr
z2|{I-Sx=W!CS3INp=0i5G`6AbTBfrpeIK!Xiq-<Ked5_Id>fAw7XokXO#NYF$*#(a
zWFG@QdyjIgu`BYzSBTX1)26`n9l-~B>u8%9RW(6MfssQ_u8r8-tnu@X$<z_%(8@@#
zi5DI`j7E#hYC}<!sm+^F!Q`>2jyUHrr9{aEJxK<%QB-w}R2oqf>00k2J_f9`L<ug7
zG!||0;FFJC+p&yM1L&M%cX~iM5e()9>uY0%!<u<$u>`7dMY(R6$thEDh)@=%EV7P>
zN+64fRLQe2X6rIUz>(0i)^@ny=v`0jQ_H#UyPT&Eldh*}=GeYZS5PI>u@S-LGDU}k
z(h+?ME?A&#G1+_h5SQ&yqYR^RluY?`g_b7qHFLv$Q55?8FJW|+KBOY9v2$A6;GD&U
z#7~r3QxyYhQ!~)x)V3BCi)a4h^gZ6Wcp@I>*uCA|S^S-I{bca?zQg#4@`1XnbNq5d
ziAW5duAQg$lmy>)M48&6h{RKC`kpR&Ora^|05vI?&KvZwq*@!L#~jWeG^89sff^}P
zVJM0MY;F}sM?AKnq*54_Rv-9uu*47(WF1l}hBO8v!OlIs?J0CYS(bFRr}Yj(psFi6
z4?!Ai4D>#k2|DNK=N32DNWIRRw*!~vD_pUfxtQ~nr@zHVKKXGj-FSuF+3PTH3|ADP
zi@fWR2l@B^gTF-6v>*e|KJy&2X~U_L=XmAiOT6;Zn?S+od(U$B!{>SMf%DYm3Wf0~
zQLwvZ`SKUP$eXWSrmQp%+<y<}?s_-H;0&+c+GZ#M_TYf++q-O@3eZhbTTTd!NoW=l
zTu6w^;sTYj8jn`6De)trs0$B48$3l-XLL8BVo0rW-{YG@4sTwja&v6k;B1d6%k+BD
zXLC?R^iYo1sRm>EX)bg^Bs2~4y<P5J9WWRcOs{UC=m|pNLd56_XEm?Ae2ZWE=U?PU
zfBa*N>Qgkcz=_ROMx&0|yh8}Z?OQv1<;&0V{By5!^~x<`P*i$^DHKW_usODL2M2uN
z**CG%j<zV-k}Z3k<@PLa-~z2oB4{0;N-6}rbxX?|NkS`$g19i%))E`w+?lf!)lQ}W
zlqgl^zc_9}oX)xL2kGAKyxiYEWN+_);jm`BG6IsBv=0uEq~rAcoWto3ue|mgU;fJP
z@v)D6k|#g-6el)Lv4|2&B$x%Nkp2%*AoFt|6(Xipx=|9XG+JxUo;|}~{VTt~i4z<A
zSO4`t;Egxl%n@aG#+~uUA2A9D`Ky#5lHd+f8V^b#ivn3mlqiV^ES@$Ob}l+xv~<2@
z-p-h{Gn%GhKA+RI3B9p49qZ0P(78C_?dLeL{3_m3=p8@OQ&AU|U5yY!4ufBK+8r?f
zA)RhYDU2!UW#)6aWDeYQhZrp=OUa2*&6(3f`le;JKP4POoI~fKw-BPm_AnTLtvias
zQI-WoSx_OC-JeAo0+8F10x6QKnve=@O6r=F#FQ`iBFKb)O7J-gwbUgx_R9n#f<UQA
z(BKW2fh5{g=xL=my?K(8XHRp`P1$pYtesp#pVrK}9VYWRLbmiGP-ubH0)!_9mx2S5
z87#f-=&^Vg>8wq>@4lt;T{3I#dt{z_nx@V4CD1A@2wX_+gO2=`;{vP*c}bi0U2;ov
z>Dnh1WvMVqGb{#_x~5P^UWrFyny}EjVxz@j>2vH|=h|gq7eZz&3PEKmhV_`C88K#s
znjtB7<hc0aosB;>$L<d+>|#diL<|@qacxT#0-^7STBnJvZ3rPHHbmP)u#`$;rNnhz
zn(Va32T#+r98L}y4+m7*fKs>+=tZD51%qLY>^-4(gtoyOji^eLmK0@4v@tn)Tbq2-
zvKzmUsckDs(PvVn*K^k4oWm3*IeeQEWqiu1B}a@{N86@WYgrUnB)yCD-qZWE{0rV=
z8;fj5G*NKl0JcKOR8|zF;P%c9e(%e_#OC>$cm0S#6*Ynqqec(}2kkXPY$=Rk^R(ms
zhfdKpmQP%Gl#^#l9(?d&ZnrZUX}GpKp^$U7Z`|VQo0qt7_8d<<`6#RF1=GVB-+lIZ
ze&ct(iEk>_T*(Z{W2+xvfB&mggFRM<ZCbqvi786@yb@XzYzvNJh}p?0E@*GbvO{Z)
zlu1eKqCnKc?5v&6o)8^IK<uVme(`(U|M24sPhZG(0~TqDqX;;W3W;JgV!XP_&epN$
z9`D${eS-&$<mBc$VY}hX#s&v&#^ozpY_6~4onkth^V0K|*t)gN`MWlF^1~lsIP%y&
z^7@UdOb%x}_q~_c**-+X3U3R7mk3o-s-Crx!SC;L>&hkKUV~O8Tdn6tjNEP=U6jlr
zdGW<`k+38(Hjt2A8PU0&9b!ThU4Y&P>fwM-|M-uyzVR<foE=Tq^hl!#1axUo`sj22
zQNuaM)oa(de*Ff6LB(dbPEl%-&cYF+M@WfQ7OgBoXr6!J8LnKp!dJfh6+ZjfpWsK{
z|0F9bYbl~^F{v$II_(pYSF^APvX;q?!lw{zB^5vO(?8AH`Z|C2zxXeB_SxqNe*n6F
zXWSV-%t*+Igt(Y?MMAY>c9v2aqbkZ#V!?GdS}Zd<91De{$GV=G>zTDJ^QL7o3(V)K
z4V&vj3Y`kK;7KUd_qAS+iZSs6g&m~pr1w5kbp*QjL0<aeITziXMWLspqO3>soh8JG
zvmV>eY3CB7JVhB9vc|m^Hc>j#*eP4j=O@j@)8Qn&M{B{lKvB%8`i`<Iu$5ytNMblC
z1WJKIqlC)w_NlmYBp#EL8QnUTj*t|$4phN#P&2xFNZ-zJ*3ox8YxMy;TXVJw&qPV?
zy|_Ygv%u^YthOsW{P05wY?%p~w!_ItxmwU4dUmICY@cfPx{vtA!DI^3B6aQ}M9`)p
zMwL{mHWC$SURM$&HKt6dO1O}?@wG<DwC8%)W7~AS_7a34r!oo_rl4XmzlBu9N}(`i
zfmWK*l++BC2Ez;UsKtKu1Uh<pT5PoRzQ_8G)-`m#qjL>Gcrc)prYZ-lRGZWcsi-L!
zpx(m8OtwAmw77d`g`MjOA`-oW7*hI!RL3U#yfH$Lhz=2YN*$?`$Im8c=V+%5s;W>X
zFkT;EVj#}v2<vcKqU#~8o6+GB&ZhPjlIQ|$Ybk0&sY{$J2))M_o}#MIqd+^EVSU7P
zJ%i3BK9G{QT4$-UL-@pZN=uN$(~wf7s4pduK24_1C1_g`Qx2+BcqOo2Bq!=pqFhH1
zP!=gHy;p)>7kH7J{RgICb@e2(;sUR{QE}UiX;eT}2dtfpoH=)f$$ZY6FYeOWF1cY_
ziBSVw@0eWO<rANJ4AhLR?dw$4h_h!Oz_(LW<+;A~GOu0QXKycZ@t(W6`}|#ORGVDy
zZV~31*g7`G6=yd_teO$O{rlf$-Wv|4j$1)+W>hjV8yrk_8LsCXL?IBmAoP9aiX`{+
zh{tC6d|WuS#{|E}m@b4-rOE%Wx45D}wT?2RHR_>+(gn;sAZ~0kf9Vz0&RtAx3LDTy
z-~$LH@IHbP)LL@t<T{hrvSuqnXdA?Ihw1KB?z{UuH(uT2?uQ;@)@;)?j*orpgJ@ZB
zuz$d7uf59UOK)(^P5G^zeb&!j;Puzvpq;hgBd&KTlUGN&zC{SZ<_W|7rz`f}xW(3$
zE6gYR%v9hYcHC|qvncRd;{^HQkuIbvg2;YuDgz5%2wZ?cT_Q<b7#lpMEcxIEpWq`O
zd5Xe(o){fX+aP5`_7dkEMOjdm1*R;pzNKkrOs5mNu1|)UO~Wi3MfcWnc(BJCufERi
z);?lm%&ophYeiX>)PtnxmQv7*IYKrlWsuSk4n5C3KjHF~H~IZ9{3|~BiBI#y6HhT5
zt>(aWN$@!=-G{_omk@;@$Rz(4B-8FiyII#YpZ(+~sH=ki>c9R6eB~=&PikV0nIs<(
zm+p)^;}0L%2sH$eW@IIjFKS4ok<yw%m6%%Kg{7fqPDhj5ut`9XWD`N8h_65@3L~kQ
zD|B+rk}Cb)`H1s2Er^a-sK;1$f5=!M9vS?IB)+u^E=OS=A$E&DkW!$GqADsnQ{!zQ
z_y!*wO*26Vi<FLHBp9la3nw-aj~bqC-(+&j<2%3tE(<~AvG;+;`z`?%)+M^`Kv5b=
zl{vFAe>#mwR0~p0hCvc@MsBwEPcM>r(L-4w>J?-KVkEe*+Ov12<BjqLSL|(uVn{J7
zP>n=cm>eGPz=IF4)9-Pxy^r6U;@@;^%N=?JRwY40(5YM3H_*vGb<d>Afd~dfY9|ML
zvMDnXZ4`x+cq>yHhwzvpp=mKhd~jH2=>vqsI3a_<%Qm}@>C>i|MupS`N^7(*XkjQs
zu`KEqs^~?wh$kkGQkv{Z*VFl)-Zxm^<D)|;K~+>#rl3?c1F_1G4Ybf1{ag5tCiE|D
z@$il>mG7vqi_sH;NC*2dyO4Yg=QosmdK3{CBSxvDwu_0v+IJmzOQ;mX)d2;Ob~ZyH
z=-VC}ET&e>n>j-%3C>Zdl)V-Kp%lZ>nAloc-=&pNQJ~8LWs^a$j{(<PDr1r|Q%iJd
z5~OP#-rCe$mm(dYDlUVEr4nfL(zC5=Tci+_ML{qk<qD$EWla$zE-DJ6DO6Hb#ca$A
z5WQeFgIhDfl><*$857)`^)rfd_Z3{ebcwwkfl(u>YK@hXf(vKPvpPJ<>#ts=o$j-{
zGvV9Me4Zyh^j^kmV}_%edE+>9?gSerM!f%nA7F3WbNkjU_O>ta#V>r5lN;wbbNUSP
zS;wi9r@%WnIp8P%ogd>%Uw)1&)7yAm^XAqb51&|PYu|GAu4HY!K$es;S>>8}Qjy$-
zjmwqO5qg{8Z0`eI+tWvlD62&I6apIsUJAUBgw|5@4!5&Utu(<o$Oe$8<dY;amqL^U
zgSB<MJ30o1_t4Go^Bo?#>oi|+-(q!rj29Em-gO#NMaoj3>k+cL$jzNAT)21_D{CA4
z{@1?6pdRtjk9~qGZ(d?$H0I{bo9yiEp@ZY(<~r|t_eu6&e~ufkzDPIOr&p2P=(*N9
z<^c*-gOK!oo_L$Ej6_?cv}p<6S(<D*Z(~F$nR0<+<lzVI=R+U<aB7og`<R1+U5ptL
zLSX-34=G@4Yl}DDe1q37zsC0NEe;M2n9Wifsqc@n1E<p|dwYA#=MBT*fTr!RZXsHj
zVYHzh)Jf=5r<`N(XypjT)6vm&4ZFLydF7QC`0jU~=Fk1PpW)&?_a(7T6vUYLsWFB;
z$@=7N6$8;ZWK8DO3y1f*DtPL{Px80^)_>05`8$7?FMQ#PM993H#m-D$R9+k|IG%qH
zfA`M#vpGbfRV7qP=If43idZLlqHzeF!qpR}sm0Rg>2T3nTN-d>-mt)EsA~vCnhs(F
z+a>BmkVj3^<45CBIkw0U_&y$KOboi1`D<{j6&pN3E##JRIbAPA6p81SfCed3D=zp*
z-+83y(Q3}1YB5?<6`GT4Yn;2d&SkgH?7$K8CBb?SF5OIGxFmj@i}b!l>6E=rU174A
zBVnDO^9-I?ZRvfA{BycTBGEv|3w|&{Eq&XhE^xl2)_Y(Bw|Ax#w-ss^%tYa5fxWQD
zTr`M*#|#x+m=R=;)S0hsGLmM<5#E9z(uy)hNeHAQ9YCnGlGaLM^de4D5S1W?BWvL3
zlZef;q<$xQi-aH(C_&Cq7b4M1lrGS^NUdo!TIeO8F74{{&tdb@A!4y~^mM|~OG_7f
z`rwE%p`cnS21Ug{)zor8&6tuh^Ya!*S6CLBEOp&~f{2|D9^u9E0=zhAm#c6V05uE#
zA!SvfbxD*hQAm97_`avNo@!8`OHCls*`C3uqOLWQ**?u=%4jskb(Yd35~$CFB~w?Z
z!r+<))mx&jD2jqmdk~i1dU|Wo-lqy*DNI=*Ld1p^5dtCPRh09|g<BmJsZt7z(Zm?&
z`<|vr3dW*LLBK*ps;VR)nRg!VEv1nRDv6c?om*(p8v<)w-|0}R8_bw9*euvMJLdN6
zI;ZbD#|J+8Fz3%+;KbSqLa@|TB6z&-sYki;#?zcWJK{a@QB0v|+9?N<ZBCp#14NXG
zoH#w;)P;((dxqCuy~4$tYn)iypy_v6T^+HtyUnQ+r+DGz7kS?kKf=kg75m$YZpSm}
zN-P^VSF^o;nBnI@*L9S}a9q%sE>3YNoH%|9J}iqd?{iyKYg~xQHX{>rB@x^AIA@t0
z9#Be&@F_yeg%s;1bBoVgi<FYV=4qxjy*5H3wIYU|oi|_Po`*il|LmteN@p6vy5Z#M
z`;jVQ{fzCs9lFlpq-E{wI-}v3@k)Wf@yZ*|Q&$6uk><j^XK2GA_da+ZrH0F|Jj0dm
zev|oRA0ZSwZO=^?nFB(lxVX+{Up<-IHhB?V8bxU|F?f3C2m*@2fDkm+rCehnxOVFn
zFTVO}@;P)C=PXY@`)zi1CfvMvgBPBEj_WtBvA@4hJy<yDCmy2^skmCG?W}dQZG-m?
zV^n^OA_SJI@A<rCI-4a?MJ*VOhYSV-yt7!FjEH;DQ4}S+dt3a@zxs8)_q}KM%x8X#
zpZJNNVts8haqomg2(sc;C%^9qT}&!-ax2=0wCXKP$)oRj7ytYJ>~HdS{?C7x&;RZh
zP*T2CR3QXPYhq3_xl`EP8Q(udj7T4meV&Ag1Rd}qp(8<CM3JZlo(>mU96bn==c2@)
zT5zYN5Ev~fOOdZx;4E}4bPW_H@x#I~G$&CL7D!v3ClVuV*`i&%mWT`W?$P`hViF%X
zd@2kUMFy!gx+s~F>HDIvOEd=;pu?kNM>B3g7ARG-S`9dVVUzu7gOi^AFogm52)ze9
zl!5g5LXt#vwx=v+3@Ss6<GlDz=#v5rsi1R9Y*Mm_QHwEIBdLsG21IHn3k_&QOni0a
zBC?bCX=Llt0mUxJ35Ypt#Vu~QYiyifXZ8FB<%XeCkwQYL1W`ql29Z&_SU@nT*oqlq
zlRh9bQWup+N|17);!Wn&QYF7aR9b|_n8Ff{YtY1^Y>}kZxKW9(sZ=t2)+L23h_FP{
z^wLW}WS`xDqtAt%#|lS}rIS6qu-Pm*AeBT5Nhu44YRFIwsHmwKAah*ZQDgoHqgx<?
zf1-$82<gqgU14_|Ru(~|BdK?^v7ubwU>=i#38@gy5uC-&=csZuqZ$@BZ86>u`Z)tr
zP-#PFr<8*_L8GN6D1i%+XcRQb2+<jfE-Fk};zMeI`Iru3qf$FoDMe5^<-vMM3_h9u
z#(?kwG|MD~#n+-R+1RmTGVzQeltqzhYmqWJyKYW1n=mL1Yr_FDIu?+G_a(D-%)V8e
ze)IuWHfkRH;G?XZ9#gMN`c87|>LJo=c6TRiZEf?yi_g<r$A$Cf*ce!p+Th-MA7=mX
zHsjS5+OFsC{a^kbr_Y?>um3Oq8q!FjYFXJVd3@s$9)9;j%=bM{fAb|K^L-|>J%lW{
z|Dktt?!rm#zGnp$CC`2N26Awc-DaPYEBA2y_Lm7c<zsR<L9MPVL{o3!RgyCnkKe>%
zdq{!v9&I$TC=zo_rWKS#5M7E7>$?`$wYce&;rhu$>=)^PmZXbAj4Awk?d&-kimVV7
zD5J4%&THTNBJ*1ZoI3Lm^@+Qv&pw2yYMf6NBeUjiCbNcn&z}S#Y1^KUyzfC?c<B}H
zzvmR&yZh7waJ}cTcVFb@^~*f}z2{h+Z{g+#1OfZjamz)f2zns;h@wv`DJe2{Fl~d%
zq+CfU1bv&@j;c_!&ZTx`-j;&M*6uFf{_fLAqv`q{rW5|&fBARW**?U&9wDHrN{lgd
zePA-Ph=t-f9K{x;W6aaGEq!m%+R$|_abuiCDtRm)+0%7%;w*4*FlR8RSXoU9wXW-P
zQRom-Qka5Uw*#O1rT;JAd+xjZ%+LHBPyEP-l6qDuluog2N+;`sSVsETrnRM8p!tdr
z14U8r*n8f?fBHB622v_M|9ig=#|yidEQOXFq&r33o$-A`1Q3=K<SHTR)SjBtuy6P1
ziViVIe58gcO@Bz|Ev6iyRf6L}2*f@i%Z1h`Ay87#s*b*k#E=T2)}_u%YSS(-z6EM>
zG!|{kgk)vtheTJ1*=0egoN=A+jtfO5c7c0j{72<3S3;hO(%^l8w*}VriK`8?ZD2N;
zAxz0Y8A@4kc5O`ez?hqBft$A)4tGgvb)W-g0g9077A+EG#QB~OdIUbbCW}^VD*lc>
zt5^yh7x<+l+k}WzWFsUNK|~Rv#?Cy`D;@5pqd1h5(|~fROY95SI9u?*kKE7j#1J)7
zFpLPjf_6QH@0qC?tw<cfZU%*bUO5UG(Im=;Yn!C()gX*RM~Mq%@;VeGw*e!QSb~rQ
zF)|r1i=NbN%GRh-CWKHG23=GXWsQ~vg{~-NL5aadpD0QhSp}A5m&fB{hsWWBr^Pa-
zrxiU`IGjv@2!*VuMa2pO){<ch29ulneu>b<A1LCEPqsh&c!#anxNMmKZ@aoqN|J!|
zi6d7Qn!#v9yR(gT78x8`MvN5nlNoa*sq31<cA8cbLNVFfA@)7Ku-Lw5ytay0BB3=Z
zbAC#be0N+=-}e}k3OHd>P3^r)`EEX-waL(Hg1ZCBj8MqH0<5*xW~&dC+DZBI8DsLK
z*x~IyqtOVR6E0fonN25j^EoTEVqi2fCecV5HCww124^_;?uRKaj@X)R^XB(nXKiDJ
z&DDFDZ#iE0?lTzeIem78&6SF+orY(=^)f2#VU*&dpZFld(Mih6@aCIu^2QstxOnj~
z6f4Z85mOnAN$-6y6x4;{qn~-4t8Z@e>%Z{=moME$xDDq07I)pf$>zp+cCJbe-|X3*
zOHPeXbGW6Kt+jd6FdS4Ag+5yC#BfyLWdYQp$P<XXT8enrBeg=;RW71^E-qqfyZS(I
z4jBXO{sH524j(K|DWda80y>L0BA(IaX<{*2iZy}|3~Np2W?cT}@A9>+f5DDdsPP6;
z7o6DK<lJ2sId%2|>nBd}uC-N8oIFK67-DTl_hX-+af1DM%id(hi!VNpCh)bFo<?_5
zL^okLEZDORw{6b^NsA;(0U=4)0SHHQRwY3jLuXwwvh*ofJj86JBaTkd^rw;6-nhix
z!2$p2!9E)sp37I>fLJ6ixflr1BHDBeS{IN~XP%osra(@p-6G$$tVYb+1Dp$Z2f_L@
z_2lz3hQuLq*3)$}=JSrhpk%x<q$~{(Fonjs9`7w(-|)TXzQeU^SNZH`UgXDr{HHi^
z;v`C@%SH&v2dk(n0NSP{h<wF#4wd~Rwbs1n(MS2u{^sALX<NSX)vso5X4*E#z3%Q5
zc6Y`P45>gFf)qsE6ST$2mc!XD2i-odbO^2RF*5Oy#z|b%sJK9y;&MS0d=gSP@96rT
zrin{cYzRp}k<e~%?nuFUG>-J{M+vlxGL3ZZ9y^{>!cor5@Sa}uh>#F35TK7xKqVDM
z7Zk?g3y-fHG5F-&>H-~>*|b3wbL7yV%Tz3!Tt9&-G-b8NRW)NePbi%T;JTcZt|7ZH
zK}?O;bTdRmX3j;(cDk0B8;r-J)KbByMCxBgmwYO{ozw3g(B0_huJyG077jG$<J~;+
z@CKEtnM`KP=5x+oJjM9T7_*|$V@-9U;`ZzYW;SQwdd{Ca$FAMt(9W5<p00!a8A&{%
z1EqmMRU-3T<ob>tFwSDMNxp!3(It|ZvG#q3lRVnbQh*i;BMn9t7+Fy00wWCz)*E7B
zPm|_2n=!S8orx138!WxI%tAva98pA+5EQDSR5e3tYEdy_$dGawI2K4r6vxMLk<Oxk
ze5d0b7j~&YT1@1N4Rs9PjEM(jqQ|-x6Oy9G2ul<_F?PgYsg*>D<Uv;tE2{C3=p;r)
zV%Md~N~JiDs;-#@3o>Q9N|W}EQYx9i_C2<8Xk$=?Ns1qLtnkbQTC|RiHf8w9v`UI0
zWy&qg2#@T2LYALvq^z0E0<N>{?e0=k1**_=K7Gc%>loDv6)gCeqg;Sk(7GXQIpmjq
z=Qla|?scj`g`E}Le67R8TYUV<_aW~-!Pbr2jNOpVMsDsOGTYfjo5+jL-{!=|4pH=6
zyYT{tvpwGP?vL>5%iH|zzx9teb^0{t?mo%=51!?|`_C~Rj~EVy6h>0jnyzd3ktaUR
z=l;7dvc3HpfBr9ginY}-r!So0_T|f{sG0i#L97vDDh}(qW;&g*v9Ypjy)6<uM9d1}
z1qV1pqW8I=^$-JPRiZ*5NJ+HWy2i(}+Uj}=txz%0HEmK<66jq=WlB`kh>%u|UPScD
zD#d6GNN80b0xL?gDg%n-ST23Q?#yYLo|~_Du7BlA=(5Cm0YWnzuP_)68LSQ&oH)(s
z^j(b4-p$IXvz#2)eCHcqX6MRlyzAn5qEp=5KH$d8F#%dBlg$2zYM958y@3xfnYYK-
zH~FOWarwjI?-mGXh>oqT?Pd75bspzZWK0M#w-?frr3<iqi@H>bT;yeo8RtCH+0Iex
zCl{1QnyoZ~Kl&QHqiY?rS<7f#F&Yi<KB7%h^%X9sjD*0i{`1fA#v5<&7ysh_l!qUF
zH{^*Z8Cgq;x+Wm9n$Ip<Hj?CPSd}I3e%B-X@BTOcTmDaf`)~7|r@wooxc*~YxbBQU
zLqp^Ry>tZG<5f$?jKg-9);C1y6N*g0MM;zf8B?+37T2zf5(l5MSi8>AHi5oRMNse@
zaX%M}$b49)iFToa{eBCIJpD!%CwIDMu?U9cITT2u4-w?Du#?_rBluJ_ESws&QfQ-b
zTH~xstz4U|xVpAOo0g)eC>T<ynsM#%YmV8hLr72C#)MS)l;JALDo{wu3ZW1}v+%48
zr0$g^;Iijl5Sb4sa%m?bfwX<`k<bOY1BX4Z*xMd=%Oh?p^jt7<6-pMYRI5m-S*h1i
zrRJVT?qMpY%$d>49y6%v4<oZ#Pt*0BJwIS$v`!T(27b<@oim*!!H0J-wa0xRsI-F?
zJioNa4T(SJleJXHVz$Nn<!r$Fw7+9W^Ou%jq@j?8GRJ}`nN3V`L4G_=E?YGxwKWba
zdK{K6n-6!c!wXPG=j_;u0Ye7Vl$2Cdq;hjn9EN3s|LAC2u<RCB?&AxKA8x#(!p>Tk
zdf@8q>fi;*F8R}m&{2d2;{tZ;5~0~4VFDHr9nz_!q>}}`b#&fQ8G~<TxTdEr3Yxa1
zs7jLZNu?7l3pSFP(k_i)t#jzbV=+m!=M!07g-p5?a3mK>6<o{?<1wGj#}P^I{nFHN
zFc{F-8Ew}xpIM~U1R-$FB5;WZC<R7HR2EM-2TiAF3XLxW*S4<G9<;3MyExpv%1SYY
zhacx#zw;&TzH7wl8N<ErxtERA6Wq8or?E4hf9^V0F8wz1?tmy;);ER}RYBiG=982(
z@!HEf49o_%uDrl(w$F_lH@NHE1=iMAII(&X88sX0Cut_vP_E|En^(EDeVc|kT6r3$
zSsmU5d>!w6BIf68L7%N#mTf>EiScL^3Ls>pw>{E>?R$(AP#B_0d9w@aMiC)6i|czz
zr4fe{jCWXXQ<`4zh$xXE!T(-@C<=z7F^~$LrgQYpF(^w0l_3l@LUa%veF%7yTI9%p
z^GTFpr?=>LE&DeFS1>f?3Qe^}qY73x*4esoiTlo-VpJ;LoX*+qJre;lVBvTF1IPJW
zetzq^>Mdo;+tn;0|Cc9&7@_T<?~fI33lCOEe3B(%nid|5B$AL7%094!@fW{aEK=Sw
z7L%0#k!W-Dy{Bzv%x68L(SWj2lvP0tB0rBnX~lQH`)zLD-s0ze?l19?r#^<NhDe$C
zAxJ?{*9fVYPp0%uOH`i0u%;|a=JPpPX`cAN2l$WwqyLEi!+-vN<jU16Z?kdkj6374
zBTA94-!>&9h#5^dV7IwNM}tzyI6KkG6hbbX`UTF~WJK>%L(y7C)AY2hr)zWKL`>+-
za@};aW?K%i{9o|>C-s6Km3~;1e&X07E+C@JRF-T9mY>UKqMZ~Hku5o-kcsDI3|c1(
zu;7!gnzN3!ZIOjWS0!546v8mn1L80+U!M{~gA$&3lYRMID(-YX-;+RKf!K+R8YLvE
zVc~EQ5F%v6>p0UZ`N0GqakIeuW{11w5W5n!S5Qv|cpFi&pwJ~c7<%Vf-(2U^U8lMD
z;6>ipzeID`G4Bqkz2k5?XMbl-(=>3uVr6xmK`~%NO*xAw+0wTG?;;|^B!uvQ%vLAM
z{jhLo2+7V&q(ygdNe%3JkF`ll7=7xr7zKqcC{#(23p_0hQe=+fv88b;fK!@@#bIL-
zO4!)bhL+y<IBRp>yrB{m0|ty3Qc_|_DByxcm&@m4jr^lF?-5e^76#`JJ>FSi*Z0UW
zaeux))@bo#7LtQ)DME+rI;OWeLOV~Yx)2a3WJt(bbQ0^pMbNsW>kg4f2ICR7X($IZ
z);XL?eR2~_IvAZN*klmATp0-gDoMw+0GIv4Vt!W(!9+x)gF7y!>~x}qm=KVp0t`e4
z&MT^_O8ItG1u@dN4)0x3+(wVlFsuxkKun$;0Q9}3_W~;<y>P6q)Yx6e-tC6IHJm(a
z*xPx7haVWQz6$O1CfGAXXVEeuoFRHi(<rQ~DJo6dMpoC>n8+!?X++d`-}CYdmw4>4
zi#%}e87{qgjo<rMUuE;e2|n?O4>LWm?C(tzRa0zm``RV0-MC6G15WnLI>$!2#-hND
zSt7P*kuBBF`FGK$i%&vC6VmBypH`(YFsjOAnUWDxy*Mmmq-&EfqtuFie?JRZ0!}zW
z^h7VglNE2tnqV|u0l>$|+&Wx@IwlwQK_M7LfsqmhCC<B~2#qPaE`*48U{%D)NI_se
zG<2&YzVyA9kXzR|u{vabdcfphhpYQ@4nkzUSP0-Dgkyx^?@xeun{<UAy!OBK?tI^I
zM}pv4s(*#ZE6QB^Eglz-y_9$1Em^ESl04;>=i-rK`S@6<uM(1d#GUjWLa_ARjKRP#
zUQHw#=Z+?!;ZSpMxW_O4;y+>g_6|Ss6F<qy+8RFd?U0hPs+LHu?>f4!M@mIuiWJl>
zB%l1`C;1Qm`d{b2{QG~O{r&xAX6Bu7XT04IDu8kHW`-;Fn8!VK+pApJ`5s~I5LJ`}
zS<n&*LK0-^IXdrX`yOkPV9PnjWHw_qhqg`eVUeYM830-*yCS7iyAF;fb(yDNv205o
zHDLoRdB8H6M<#EyW0;G`o~eYRqR)rq0l6sVB7jO<;(`Lol3*=D#DqTDNHd)iLqtSL
zxmi<Cqvep&l*AJj8E;@z(zH%z>ADhPYNLt)*)l{UNpq<q(i7-Jq>l<IK_k&f3XwR0
zE?)xzhCW8RnWejBu~#DF!xcvDh=CYUloeJaFd&M^&fYF9J);w2%0a>2!4A{;K0A9`
zT;03M8{4n5bFj_!jVZQ?v=0YXH&>CxfT6BY3f9MoN8C0o?Yzf@JlA;82Bc12d~(^~
zB%ygZ($jX9zISw8pqYbh)8CmwFdP_0#gKuisdPoH60Jn#RrLZ@TX5hUKB-=@$*s<{
z%=!aDOmn3wG{a)VP!FlafFVO_asV-81#LX`oT;$WS&Fkj^RhkaA43nDkK=dp;1K*W
zIts-1p#cl(s|l0<0RR9=L_t(BMIiVd;aUnDxSoDCMFfX!J-ttqQPhyi2%^#mq0y$G
zC@Z=M<IxJrNSsrMDCoVTbDkI!MjHyDQwuMXRJGD*lhm~-;5_m1jF3qjAyPnGlAMRk
zwTl@VPo|8KlrIwViYms4cOGXG+fr1Zw4$zRf`!I*INM`wN1-Icx<relD3S{fQjBO*
z;bK}@42A=yqDL^mN;rM?B&W}=Q>hhX+$RXZcuf;7PqE&DNG6re^$4A$-p+MMWzsf`
zI%WKcl<x8HgAdTQGd}h4PxJJ*pX2n&(~RrWT)(-;eA=W?<~_&m?I~S1r;j}pNpVz|
z)f_D)l4-E*Y1@{%uJXk1m$$^}Qn}bj7bhuI@(GlZYCI-LiH|WCZmC!nA<#E-R#(Po
zC1@v;rHnowJdhR3QAE5*3bvK;C@C-l%$lBg*E1|7IuoE}At-eE9Ayxxb&;P(l3Ml=
zLDvh0X2f)~#y76s#0*Qyl@%`EypFOxrHXXYa2N!=@GRU6BoR$&Lxd%3<NG{2mQdKy
zua~N;w;3WHL(?L}AeSPf#p-{-D@2IA;y?bHaXDoy(1aMpTVC+(6{g3?zft7YQM$MU
zC+K{@xn!3z8d(Oza=E{iG#FJ<n2Ila;rH3y+2?2e!q0N%>{(p!iF+rYtg2*RG@mEh
zc?cL|62CQAhQlF0_wW83*KgkBAOGWj%yd5c1Gdtgac6)K4rD_RfnMyh*IwoJ>~(hL
zS5T)FA~go^F{BbDrmRpQB1nOYf!11@c|+4Bl+K3)CV0=1_MnQR!p;=w{7HtqsbO|x
zkDNJinVYs4frwH%5%}}TD}_vX)-ox|Q||KO?!~n+l}s`uNG*vIK}$&}3QUm<Yo$mn
zWZQd$>`|tr?He#Eal_E86+_%;OmtwomTt}w9c0;VUPv_*p171pfI<rmI3zB4+TpNB
z0+v2{dK`OKTiTnRa8t0lyUw6nLANEgh2Ho0;D|D%^R_gokz(!-dE@#^+<N5}lQ<<5
zo}!5CPY$v3j&R`dhp@9f;rs)qQPB~D#ZVxGWmpZVivf0M>0OJ*5`-s8hmZ;>QYk8O
zmRah&<@76|&@$o`X#+$k4U7ho;b26eOA1+1h!R6V(kmY|;1@o1jzAYXW`0KJa@K3?
za4{M57rJDm28_j+nu<ac6eMN#LRsyONpTk3#U#Pe*&PpL-ZyEV_#Y7N{4nDk6?V$!
zoh=N+1xy*1ts(#wBZcb_-IOYHltq#rYpDs|5xu4J9S8l4Lv|T$Y*MHpwX6`m$F(!0
z3`8Gsy+tWQ>>a*uiBjM^gyE3hIeH&4QloT<R%PN0#Dwf1WFm~lEKfdGlpv34VwsZx
zVvf*Cj|s~Iz*|QbTJV8F7f7QpXxgs9bv@EW6oSH-WMbw5A|o$B2pke6Dq;tvG9;tl
zhO%7c_1Cu<Rt0M-t8~3%b$ydfafbB^L!S89A<f}7huaMYt!BULiCBaQi6_`LD4+cM
ztebJ`<~Cpd+S9!I(FdsqBi{4g_prLMN?96aO@ZVDeYnQ7pVK4hLO>QCqb0GIG+jEF
zrAXYVm`VIX#*~-W_qe`Kg)su@A*hHG3CK555M3ZtHRb9iL`i8aeTZ~Y0v_Xg{AA9k
z9us|!Z}(CNx7S1w5#EB3IBD<*RIo@C<;r;3+L+D*(>Aa^@<anB6x0Dsp$SS+h8ZY_
zrKb0SffDFK(K0}V5wl(}v4UCGQk9yW?dvG%=|iNbDjE-6Os?_KN2HYb(zht37AM&;
zLeB9K-}c+L8QJ7{`O`5&g+)QfqFa0Xb>IG3<MPV=hrZ~si^LI6NMr<WVMAhbcChOs
z&I;m+Mu--Hrmlxr*C0gT`4_&+WWLW&{?t$Nz(Wt`Iykl7bycIJrfsLG1?C-6q=iv$
zEvGh5^6&qZzrt&;yuxq(%YTWG;4=a#q<9M&d1w6D9zuA8vP5l}^}B2zT;=xUGIPJj
z$cz!P$ZaggB4s5K@G<eiAlZ#r7c)I0-TV<4omBlg=`>{iIfRG{o*)w;Ymqduh>uzr
zpDb~?Ecz08)fQvo%#rCn#Kn1<zt7yeL|sp4U|i%;<SvxVG>8}!F%x8i2iFH|*U>cT
zyhKNgGL*tFEC+P8Wmfh`l8|ZvJC-h?{V^Q5q@~wfyaREp=*Y|Nm|xR+4i9D=+-PvO
z1!||H=qi{>yv<0Yj!frsgbwHus=7oC1#;-HYK9XtT$rVfTlC<1V%rnvo^UAG+djbA
zfJR|dK`kp<m#wx`7B^~%ixw~3QDKKd#2mvFb72><-=Xv=<2LjN(V-+1Mp2a|g()d?
zN|jLA^mq|Fno!ytSb(F8mbq(~_cNSNS+7xeOlhc8%|O+R<(LrzlH;Asg<;HFc~RgU
zzxtB5`r=6X{*DmK<LsS`cU0Jw5(=qMQsJUVepQ*6xrhyyA!7O#*-X*4L7IYq#(9tQ
zky6TJHeAWP%5RX?Qx^e0nKGN~5QFDLwMMWGsSMEvM3*cP`T$mG%0W#(7~p$H6bhqD
zq}I3)(g`3^r7e6w12(4LhZvJ;Hl0~%qX~&rMc%LuiO*T7kth|xIa=>=rOkoLnqgTZ
zyr7E?Cj+`LpcT@mq6gB7u=fz1#y65tHDYC>Cd?!Ag9)~;`2DZG4#Hz9kmhxs>P{e3
z&jXL1=CkkpAYcEDub@soz;AuyX|7-2XC9|Cp+rds($Vz}B@MO{eBt-M!|v{s_rL!^
zlrEUJEtC0-Z++`!CX*Q-{_wqo!gJV0T&&Qd!x(61bDVB7VkB}65z_@Sdc5ngeTVC9
z%Fq>R8DSF@paR166hY(kfO@!r41&^Ge1y4@XnLe;=?~_JAQ4y$4S0|Dl7)mT8jaNz
zK1g&QAxQL~US2(WCuyvt4<4sIWmz$%B8q^Qfl^AeG|YWLRj(jpAgYKLL$(hJZq7YB
z(c_|IHreCe^Cy{4=UnR!nDmhYo1pTT?Ft6ffQg-C-BMaXy_4|=l_&B0zK_R#`~83O
zgTMWQzW-sy55E7m6EDOdna@3aKf~EH85vzLolXFc7_Cs01vhVA<)8ld|AbF{>N9-k
zLr+nbgQW^l8iNXkkd?fFh=?)eR7%Y~ciqij`GsHL<(FRI(&fvTMd}EDWH<ZH__H*G
zC(_aThMmc6Zth&+=JX2lV#>(W2$h`IpoeIIB-;-QszLteLt3<0mkPT;BF{^i3c3aI
zmzfNS9ud-95XEBNNE7YRRJ`CI<@rRgq^-P_<Qj7aA@l8EDSXPX{Icz~EbKCZFQr7P
z1@f18bKZISuBVwbOGg4J1qGVAs2SEhqhX6tEx~1hNLKrLo7MEo<!V||%4NBcuKSKa
zfgl9%eZ)>ZhgWADUYWzLVKCJcvVu+~@reYCU>4?RU{IC}PYj5Kg~AiGL)8J*Ih^mY
zokawXYaC%7@H5!iK43oYDAy{CE~rI?$Ki7U9HdMl%Bb)-qRyV0LF6}ZjJuqRI}AwC
z;$1|kj@02cltq=$I#ZyP$q|@GQE`h-NFdM$OB?3Q`zdYLurRt-Qcx8G1|oUWjTlo?
zE}!FwNq2Oee=GjBIF^#(%#m?i`ir+NAAb1p&Q!6BIIes^`jF5##FCIi@DA5?NZVnY
zg_!tw&U;h{6e=Cm+Gy%xka%i9*L5gTyR^u5B2fy2&aU(%!&Ns=j`ySSn6nqo(|3LH
zttm7?q_-i2$U8s5-!dkv4eyc_!_uxLEwfVFFFWZcI6t4DI$07%lIk$n)T^(`3at!9
zQP5eRKYCiBDZPMDQ!I@Ximq|QTC%@C<-$GZneX0YI-L>34A&`ER?gA3Q<OSnZL{N*
zS6=7-dp=4(opI?qU*_TmOV*2)Cw}574&;PG;rY(iGdQUbwnlhKZ!JY3c>cMUxN+kK
zXV0EyYwI>m)3UcW<HV^A9((WmdHP!~bM5j3A%=K&Kv5{XpHU3`@?csRH~X9cnoMxJ
ztoB{_*)22xDRVc5Kv9&)s%9`8W}cW&>yH?S(G&WPW<DbnfwC+RMS<u$jv3~q0zjk{
zS|S}oQI?rloGwspYw3FhMpBj)m;x0oI!@^0h#)FPQRBQJ)F-ea((dnYX=@8};w;xL
zUt!*LJpTU27!7NlfBpr^(lG6liKvjN9isI+j#B=&Vn~iude%A)4kv_|Igl~6ID}9v
z+Bv(syZq*F{w7V+@sX!K#>(niMgaq+DCpZ1*Opqb0j)JL3hKJ%lb`q`f9Ws&yZkr*
z@E_9mHbt@Bu{^mmjt$%Q93C97ed8+6Z#~PU$t&!{8;G+#o9d_%i&zL2LP|GDg_BFg
zSxAaVYgsB`5s62a<Hu7mmk&LaIcdweB2TO;x5&~29hVa}i?3k;(1|6Nc6ncZJh@0&
zQY~`vDYED($mJdBwez_nm7**R;P4^RH$BsX!}MMPC?zOVO{Hqqh7zSp_IBUoZ~~c8
zgtM9Z?K*7i5Mq!y(2-<I90~MU9^eoZ1go%z(B8CkmwR@eom0+B26{++s%AV|q3;4s
zpG@V0icm?+s$u2yhy%Mz<EG5p35VU3KIC?7@R%r>>_wVG2y?i7%X00;P0F>?Y_5%o
zwA59N^Pb=%*7+rc3ZW1t$A4v_;}H`kBam6N60ApxmO%mKxL`1-sH-}~dsQQpEOKJh
zqR2`fbsarz?3vgp({4i3OpywVN%Q}ptQd+_N>NczQd1z&*$TzeV{uDF^r%1+93@R0
zxd@77@wc#6If}FU6D;gP-yx&VUO*6|%xz5)zXy`nSJ#5IX=0bz4n;|nN?ZdUeTo(m
zmYDemws#Dwn!fL;tCb{cmI9$QB7|g2Yc0Wf+Sxp{t5(LCL6r<yLux}=m!#n$9@$1l
zAL!c-+xBVG#+dU?Gh{AO{Fz(i`FWo?P^o%}S|LU9x^un7=Hd<nRb3}rU8V3LWof!h
zq|`Dc5fs5tr~#MWxW+vn{~()Z=iIot!>O~Y+<)&yZd~7|zd1um#i(8**p~H^0?}+E
z?IGIl^TY=ppd1!_?8*1xMip|>@SSHap+&`+^-YGuihJ+7Ks`v_;dfm;&Fvd|7$f=8
zm%fZKffrx;CRg4J$gqNS4J&H~B_mn}oNE%7iiPSpaTpgxZQt7@CnsGnWlWq5DJ4>9
zl!mGv(v}s&@i>_k_coDyMFbycXLGt{jv0=V`LmR`d6Nmn*_SPNf^aD}HUuP*vaXX6
zZaTbZTT9z|Y@yLcVPr`u1cPYU3k5S)86c6SVpneP(&ft>Y~SGWenUHLX@%yAr=DVc
zW0hx~{x(~;_fbVf2pw>AF?Y^kt$jx))V~cQlH%bOs$AzHlgS)w`<yzFtP(<yI2)+z
zAtAtT{>E<*BYg6cpQ0?PL~}r(wa%Q!blzF(D9Q>0jK^bs{^x(5-~0XF<D1`p8YvT3
z(EIR*y~R7@&(LUF%XHS!*vMhy*>61)9a*i1$Z|+f1tFtp%e(;kKn1_3TzpyYlQP~1
zY=GWoshmox$mDvP{zPWPPG)~IMdqn4pACFAQHLd~{OGo{?vcmd!SQjNdyyx2L3#rj
za0ncscrxb1o&5AEAVFY77DI$o^!81UQ3h8O_}ZZ;sAYjDD>gUx**};hljANGwN2}3
z+K#53K{Zas#rd<P^3Kx79_JlW6@*UE9>C;=<HmOm@RtLVgN9ji3%~I+7dG!<G+3jq
zN34`<oPKzdHcsipjPcr-+t+V!IG@tEhGJZ^Hr`}9nX$jO&vZVao29}O1?<E1tJfKv
zg7O5ItUFLzA(cQ#MRYQwa{(!iw)Uc3E~QBGf94!_T}y90-bc!6!20ScWm#cNfi6mv
z)(DwWM0!#js?9}N8|QSiw64K9i_w~u;gCw#)UskEh8W6aVdjbSbW5K*n@hY!G+WLW
zSsp)$%$7Y6k@;|3H2aT^n?KRQuG!fktgItzgAjU&*d*9GuSuN4_YK&Vm_4b99;p?=
zE1XX*`U{V`h2iQVD{p^qkByD>41a^r5)l+dU16QYw;dutKW~s#i8qQu6-bqkJ!Le}
zxkT#E&n=u!kSru;{N?t_Cyj`>m@HFzn<AcK=GrMG&{CnbLWBS^(6k+0YblG0@#-4Y
zaEvLdWQZ(0KIf~}Rl%^5R6R(iSY0{I{{AlaKXi_*-EHc!=2JiRF@{6K&1<*$&evb$
z(rdRkcls=&q2tz#TdZF=g)1Cq&aUwM3*X}*^$=kB<flH!`Frl+!3Q5k1k2jm3TqoH
z9OhQGkB-JQyyvm^@Y>aH@P*I+9^d@Zce(52r`g#20EKMW*eJ1+Lxw@qPUc64Q4A<4
zlR2|XwkK`VW35l;6WLB93M7g`8-``YU@*p373FAzP$~T&B9cRR40O$$;5~yujaG`z
zd%CVo_9#9iikM3Lm<72wibzqExxFFqL11m5?;?FK(4|8wPa!4FYn)f0OL{e=u_a&m
z?yKC~-vi;<={<U&dFm6NV7z{UuYKjqTzd0TYD-v`#2y)Wl~TUNIQP%eSfC3rL{wJx
zg#dl;*xBAE2I%`h*Ev>KLN3HSbv@u;{PSNWhQMb(`&qP3#9?JL!ugbJk@-f>CBm_k
zlKUTcfWQ2g|1vMX@@hf{?=)fWjAJ8(rxctxb%8T-o1Lvg=B`5+i4ZkVkS4zZDJ>#E
z#NlJIZt8kVpRF!>o4Hf+sE}Jw;DpTIhvW2!w+?|=o*xUk!1A{EgQZ18!d@2N$79by
zE=ZzE0rebBMfv^+If76wY`GT5UAFQdr1{MDJzcAq8%?1VrB0SFg%qr>ZLoR56I{b?
z?{Gb|4a}xJ)7c@mTBAhg&yx+!afINo&LM2ie5Ysk>LIsY*=O(NK-_{bi%9Eebj!KN
z&vVz=i=13L&FRx;sm6+HdzZPq^9BY%Z#~v0iB=Gjq8cDthmDHHLNF037El$thYdGx
z?@+7;PMu#Ts?6LK86e8@KqLV$hzuT(4AWCyELqr>d<%mQD5)t6O<C5IMS;-<lZ(3L
zL4e5CFCH5$eN38w5L2h2GBqPJOsN#2pr%AquwW!w7RJR3HSIAX=VIT#4M+2698pmg
z*daggm_K^Fqrz@_;~HUO6Qv8JEI^mbmt7LEqsPvs#Ac2f$mGA3U0Hk#h>YL_@6)m<
zw_>GI^tQ(o1yxxx@0%PbF0zl2z?5Z%6OwswBE%*Iq|Bzt5|RHd<`rC$iCb&&!KV{S
z2)s)?j3hBnXkA#gArn#)d?MlsAt+5jY06}iA|=jx`q>O01sf+%VTuyh_gHV2Z9^q2
zwQeYE%Y7H_;=_OLQM$0jtFJ!GWADGnt1s^I|NejdDi1$$mJdGpAb;(D^)u{lPx;o@
zzs9fq+Hdlq_ua?cANddmQ8K!4iiZb}GoQD(s2LOmKla&=QjaS7eo8ec5em$pLP^Pu
z8#j6UgYPCd$K4k%Fl&3ffYE9}Q6DnWGYTy^*xBRa;4JNp7g=~L2?5S0PGi^g93D*Q
zx-MOOLh^<)GWq)%ZP40Kl>>BHqst+os`1{@wGBm4f(xh!)58N;6m~}A0G~5`qw@rr
z$-BX)mPAH}2ZLd<5CD9TSPM<(Xu3$04Qs;!Z3-H(MzkGx$9}AM?(!aQT)#;<s@Qb}
za?SAMCqBe@V}tKL^F3aB^)+ng5D~iG;leR#M7V<(!I7XrA@DBJddKeWA=ZX`4uNx?
z(P*68tcL&Jum1}oflqzv)6^ziBVwjwh-`vQ<_{^QkgDLPfBL8S_22jne(SeCza);|
z8FvPNcLEVhPHddyLOJJl++jMMgK9yPq)ga|!l6Wp$|58ehpz8wJ4f3^tWEqnA&+hg
z6L7?`u2>doKC<MwEvDgQO7T`SZo$b*cO=Gnx+7JK$GKH0LpDE~fDkjb7h@72ij4il
z+%1wa(Ur2XrI3kh=Ve4@PdC?hv`Qefq%0&#6)CJ<DAvXsoIH_Y)25Rt&gQ~y-ZPm^
zXiqelSm(mXrFR*8%Bbxfu4~!7IpOxD18%)~$m|Np1JEwbiJPmNy!YMj<-vR3#kms~
zIJ0?{s9FYB40C(PzTL&Tm<lnau-;<>SSPS9g_=<d7{yeLlrU*Lx3~9E<DSuGMOkUI
zR0%1S5+zfvZCHBDguJ<PfB{K-6=FzWO|IsR*3?y<kh(l?8J(@$#1Tez+<@4{T;S#!
zU|c7&;Y317?@1x^Vszm?cf_m<Sx&a-9v(lI1dABA<9WkJ&Up*vH5}t|{*dDx6?Xe?
zzRuy=25Uu$9<O7gTB>V&R_wUGrI}3_2baud^BY~59wsh9G7yYF42g$R6oz)5`sqp~
zJxO5-w9$#9msc)E>*N&|lV_3ZEj}YGi)bsQ6<TYwG}-<nQaFe8j=t-bIQ*g*3=xdf
z2qMlM!R&%m#CYp)U5~GP+KoyjVy(5z<_*PcMv%#{IC%u+oe>)pJFHYIv{G~ZjXhRR
zMy_7E$*I%#a^HP-^V%z~^X+dx!?kN~@bORH=EB_@{MgTY1ZO|SKmWB~qS@W$fAT;7
ztC;bSVl-s^)CR?HfEYKZy21(YUU9hJASE0e9AJuqGiUGP%U}2^U-<nm@x?EGgKJl}
zdDpuhM+A$mCY(OE!rslB7*|jRLEr2yFEHLG)YAK8hB=+i62~y6{8j>)?@>}#bxON1
zLxir;Rf*6#6^3bz2HqoLq@B)?5{AP8&gB@Ot|w;e_vFo%R@woPxOv_ql|Hs^(R971
zZF|~EA$5<49-|9dy^3$P=zGhJzUMoyU!fk=^z_W7<f*6L&zbWVc<zPQdHOrwC3Za;
zfy+znD8vu=zWv!9iwjBGAd3v;7#(e631JGPmaX@hbL&0E7(xjA`mg^QQVKr#kxx($
zhInfk3`a~4CrPX%Q}HaN#0NNc{v3btFa0IH`K|A;yT6}l0C&cn0d^kf_JQe;)8l)%
z>*OwN*>fYjnziJhkQ5XsWf0aTp@8>vy`}9UO%L-fM{S8*dM(g8a*+^n!I!`=e24@W
z5L%$5M3EMw;@AYPv&pZBAY4ENh0p~kmC>ZP66OOWO45Soupr0<lBg_z@IG<Ikf1bD
zg|xu(9%M3J5(<zG>r?hCT|fv+Q(DSOA>;@_&5GFM{AP&?H;HY>r7M<M3wrJt%x4^g
zDU~#+C^8pJ(f1zL1i~cJZqK>>{1$t!v~<@&^)L*uHmJD!uJioIf9-$Gk3IS0td342
z83J&tdkzsTMoF%3-KGf<rzDoh;qE>MlNLYq#F=Adb;xZu!IeM@48Tg*J@kYd4V$NS
zID7t7-u@B~7?G=rxTq`Arb?h@cBYLU>nv^4;BE5R)LK&(Nv$uXMk$Sv3$!<T*)IQ2
z;*v$yc2!hVx?sc*g(74Y;*6H$d?)gLP9>es*j;*S3w`XO$?6gSg~*|03#;e_8oY=Y
zOY_|k`|gJy@2IfbdG!_6Mk|yns|mtX0w|J{HzY2S>w3D`lr@q9E_uWq5$<J%ltq>w
zE#9t_lD2Ir%OVwv)@4KEymDD`-BOsmw~pu>E_0w{k<un~RV6;wf_yK))D^lg6m>=0
zH1vIs?fM+al<BVv-yy`2_YgXfx<vB%>bf3cBG9I`V4&&dbT@A^YI{zeK8-QOG0454
zHhbuHOm{F~bMz>JeIB{@3BK~x*Ew<akSCsenAcyu%B!zl=C#*uapLrt+q+kJ=;8PA
z^S|&5TzU0%PTqBnpae6!OLT&!v9!B8xajbvN>MApGaL<AT{*>Tuf5KfzVI~v_kZ;N
z;P&lZ#;cp$`{1LLN^olJ6vN>vTeoi$XC)`g4e0ie;VMF;!h0FP;vAF7gr;fp<`D#u
zeV&Y_EDFl9qAqKcG&m`dra&u2<7`qT`9PE&qGxw!o2o3S>xzIyNr~%PWK4+bqW!%v
z?sR>Z+7THlUtmWureebTkWS9B#*_oJG<a#4`N*@cyiAA3D9QeO#*?4=828+JH`j06
z<oWMCM?Y^F8b!1=d0L4;$TIgk<M>F3g0&tcWhNMd?LGVZ2Q1>tymxr-SY2JE?*-2F
z{K~I<4y@-hpZ#%^QrNbou4`=HFKHNBtJLxjjvxKeALVnO`yBuBw}0nf??=8f{*4;0
zjr4~RG1Ux5Yn-S~vg1y0WBwAJhCoE;NOdE0HhE24ZtF=m2YrIf5~WR%It1AXR*)J<
zg-ZN25pzO8529Gu5+@E-dj2CLi$EYmNAxKlG(Gu|!cvMOpdgZy6G*1S0w82gm=M&3
zPN0QPdA8oCB2I(|%G389K1vW8qZ7|8;0e8_>phM2Xj@W`74>L98ON-N0mWd%%Bc<H
z<a^xgTDmT>br86|y^R_f#>D`mOLm(%?bI^cYB;<$Wp>%3u1YSdGn{_(JWu}IQ#^e3
zAx^KKW^?@vcb$GOby*{hrLunEOTWcT9bi@@)0-WQjM$9)59*S(byT5XrCj0U!|P1<
zcG;OWv`yYddA@I^j;oh$Gc0T5s-&*d^*VTu2p*M@#i#)zK_}nZ=(9~tR<aaTmC!k*
zDRoN3FxsG{W+9Y#t5!7&38N=^N1>CQNll5NUMzJ9cse{0i(KMyi6d8_me`wImTkuy
zu`(z1%ItM_)Rc|KRJ4onPUod}RM;I}yTaAMkm}TFR;rS!EYPK*?~}(6<nC(lXa!1$
zq;g|HDve8?KoBWLMQNRrEK(F#RoA&X7kR@2h-Iooh;g~&ae)$v$o2^`7k;U5lthtQ
zi#{ectgZ@%g=XL*UE9*NEq&kPZ4V(X4P}+cfT#eWG_pvW-TH)Bxfqa{r0YXuZ|9I%
z>sVP|N1J5N0-!}lr6-JQ$9CV~W}dn#84uPN6f0bN{W|+Q&vD`IlRWX{dpUQ0lQU=T
z=B|4;(WPe)D+aP)_xeq|iP-%)J3BiJHrE;INL3rG2<-13^5&aw@Xc?2lh<E+o&Eho
zoQs?|wZ=Uc@5T&P8Lgh=#JJ|lOSgFQ%^N6D<8~zXubg6f;~A>*Mlx*{0;Lksr0;tU
z4<|V1Qc>q4vJhw^D65jeU_dz-q-2g*;DSV%0;LtsC5^oo4s^sj%dHzX*xWpk&>xi|
zboxFCTBH;No0Q95qzhWiNZ-*ZCkRpC9P~CGr6QObS(K=*A?S*0lL`BieNLY^#hY8V
zdDjOX<No_jad5E9Gv9rNo7ZnqDNX1tA|?{GQYKxnUE+k^8Sn$_k0_4#M|t5inM|_c
z(BYgV1kc*q8ro>4Q_rvb%CAtC13vQ6k0NDm!3wbyVdRbzAq38zJ<HGk{Lk~%uYZ%h
zz5Rb3ANbDrH*j=)$Gn-*^*z!#tnF~NONjo`B`imAkvUdGLi7oB3mFtknPWr&!GRSi
zVM5ESj?IFK)JY=G?+Zt_6a+=?LoEI!UX`a84r_b@27*Jl05KE!^XE8jkg~vMlAtKv
zFJ;P-4Jx-^wMO}fkcrbCqo#~yYVt*o3l5x37C}BH(Z|%_XOg&L(3?X>!wpvS3VI;%
zr+V%+ckzwwXSp)F#m@dATibK0ZVx~7jOsD%tmokNl*1cy4zD$cJs5;B51fCHCm(o<
zk3I0?oLo7>upUvDLn=K!mO_>0_^K{YgArjxXHL7s=n|u#_knKKGB!ib4^Q*J1NZRU
ztIyEM29?0GqofQEhlifs-Tl;o(>hz2q>i14AhU=>WRjzIAw`OXG&dMiAcagJ21;X8
ziPi;5D%7#|>%tP}=xEB_MMN@Y7X~egjM{p#ER&-$>`2|J(7DiEH1`*ry5rnBxh(LG
zvYE0D;Fyn|Cy@-^&fo428}F#F^Yf0IZ@fXVxyi-RFtr2=gKK-FP9M$~L!l%EP{{N_
zleRG|YucDwdQwPCZt-~UkxC}DdTzrAB~$ou@c593k_%6)NU}^Z$}Ecnp%=2I<nA3p
z%q`!Tij%Nt3nwv0Jse`n65F=;w!vCKusJ>@!a`w{R#hN;BJ)Qna4z6o(vVv(nD-K)
z3KF89%Td)K!X6vOv1cRCe)U^?^v{2gQ)ky%T{BD%PvQ<WuYB(|ZG4)CA3cMM`$Y92
zR#!&M4_XFwjgf+~)QA%+_z*elCp`P&OT6^ji@fmCYrOK(C3bfX@K&(Cc9KUPeis|-
zC(uSS9M*IeuHD?{)*DwDsC8th=_is?<px7#**|=hlV^^pAibk$TAUBe=kt6~OgXVS
z3d%xJ4GrU!0rhAVQxr(&6M`BdL`m=t9~=^mjFH&PneA?K@#JQ*r9q$qxV|BVKJ8~G
zFfI~gf}$5Kj*x9?$S81(**7T{^-(Gi8s$4YC0BQEA`8vd!7WzLpW?oEJ&f(<y!ygx
zT)%t+Tms6qNEeWpPeT<vGA;o0Kg;6>+n|d!K@1C1?}Umsjl=tC&aDkdl`Mt^!vR%U
zG7S@c=~q9;>e?#rdGs+NpoGGPJ{9vZ@$R%TSm*fkr#{USKk_5|_P_f4TL|ZWTH}#N
zj-sUQjDJ1j#EG~5T$Z~Ghi_691}QuvahVhP7We3uO%wwH21`USq+=?3X4*3Gp7jEn
zkh*;HNSg7oXPs71XfOthPTwmDRzY-7NVrRNoMg@jiJ{G<9W1Erjvm95p`6ne9R`W&
zEh6LzJev_Oup1%dQ9-e|t_E^!V@kI1nNugEM0$ua5|qG6MbHx8XL^tjSR%ou7f@bM
zb_O+%45VYNcAPAF&Kb?99#GVZZhgS&;0Dj$I^>mGhn%!M>&*cw!l9Kk)0X|46Ao@R
z%(i>R^OCiCgNN^al#hSlM|u3jM-amBhZP!U&zz;N8)k(ixPU}c6p~?8G4Eg=EGy-R
z3!7(o;KF^Z+^m?`2~wt2<HC~0d6>^&Z+}8vSIJPg$bPd?5W|8T8JA7B1t5?xB!Myt
zE!0tACl_YKk|V!5k<0d&Ge#BxVj!i$<osBbk+&mZ&4Lwo{D?TRuU-^&9Q*zV?Ol}i
z62B}j;i$Mf>iZ?b<Uhf}E@FuDmfJ7C#QMfMtD_-87(_J4L3*m9uF*<TxkyC{Ircgo
z2s&D%4_LYAk0(P?AtJ`8?04rtNqP~{Iuqu7Se^tzNR$#p8Hq~L3D8PmvgcUL1we4Q
zofvZ<NUrEbRz8Njsf^a>szM;Kx=0aTp+yx1B1U{htGw+IjVBmKTWXBfl!c)0Jzm!A
zPBgP#pv!<z>Ak4ZvoV%5H@3M@-i5P{sv7ZyXP@W7-RJ2hTkPEKDdbrSW7xkjp}QFP
z)>psF``-UvPMz9hFs!(I=?!kYco{)pI2!V={?+ICXaDq9XdBD=>b(@=G^a<8F{)0|
z%_TQq6zmv@k28kj7TY)Y;Q85~{b^2Yoa5zZFZ1=^`U+<TEzS0K5nN$?^8`ZdAcSDn
z%;~LXKA+%hM0k;sH<YH9n!!LbSksJ06-JM;Y2*|UI=qCaOMKhY5<v^9zQ;_b=yuBZ
z+y+j9l0{O%`Ud=*K#gF45Ct(>q5$b7(b=T3&PSjiWIl|hFg02g2-l~pfn1}R&1u`n
zL}?C`=hGj4A4Orf^vW);zj%Y@utN)h4UwKqpjHy)17&bDyi*I}e_O}*l~m%0&l0mU
zwGzpAyzc^2S{Bn+T6m>ZuPGf@ZoJ7a{?mWL%5coZyY2-dMoN5eNmTCz28Bj(a`O~F
z`RD!|U-`;cnYT@TzCXPJ?`vO!m6boG7rQflsPU<fe2Gte<V!#NAC&nv0#5u9kNqRZ
zwS4;{#xczV=6~pGe&6%{v7YZ5@Z8(p^XW6dnd84{_-D5QfB%gZ+jwqYnql_?vJucr
z-t*{VJo&&UdH?2T^5g#?B7Z7i-h1)A?6J*j(<?X^kXmtKWt}r8$L#OlqS<ns7;SK2
z^E_wO8OCNvUF-op66jTag)aH-?eAOaYD!raY@RrU%0}isL=X}wV_KjmwM$09!VzGG
zSn}zUH?3SYU>De1rdS>mJfw~xg}|WExm<HgSw!kXN(3o&H+kN&Y)LN;2$gwz3s&9n
z`}6hkEk7OIdwiSziHav80v(~-+vnyhuP|O+N16&hSVf8)7%vr3O1xkRS;v^`+?YtG
zT4|)#2rbjHNu~|fd5bQKEG&{aOgSl61p;KuYQ<b3XH~6HsRiqt!*wYg!8?byHYH3*
zvP67g_m?>fuv93iqCl4wyVEJV2Zt%26q2QfP|Q0^Kk@Ws^1L%jP)dz35JW-K8+Im^
zwY4#($c3FM7>{b!R|I{xi*}ahzxfP5`O_cA2q?sJJoC*fG;@i!tAu&Mm6xV`;6vwF
zsYbNzjDPyy{~{|Zt32_<6P(=Kr0)dR-~3&azmE&2&M+*_LKxz#rEQ=rlFHW<C9#+E
zou=t}KJ?LtICFlL(;KTi|K$VL3df2$<i_sHoL(tdT^k|g9-a3Yr3*A|gUxMfsh}ui
z=ExNchBakbBBa8{goyTikH{7l!MT*Y(f1hFGuhjvE(~R1P(l)8%5=4TN6c!lh@iI)
zi%ncU=ZM}B!qGZ8^-{<Xc^1iJGO`WO*?<FPZOda%J<f^E6I{P?l^0)mjqUyYwC^B!
zH|3T`R4RKIzH=_apOx|bnX@6AtS&B`-g#!TIWaEwRVwWUG8Nyiz4i+K-7o$V{`!CL
zAF?u9%~!I-&680gj7j{Y;Q92YKFxji{y%x~m6z#zOYW@Q**AB_o$)6$4z@d9eb!=P
z#O_7>0dNQdB%gfz$9T^@kN?om^`%$7h^<>}k%9|~sM)@{#qHN$r)eh`D(-pcZr=aM
z<J2gIvSy$)k$#D7<!&o|4_yQMdsCG1jD|zBHi?oIlg*kBj;InbG8}1-GYTqFh=P(i
zdQ4_^TN-g`O_B_S9hODvM9OYeXbL0=oLXGjLPqsi7Hr4cy6OMLYb1;=Ebm>U3?~dP
z<fiV?;`JzrVxgzaBE)wv-eF+}DuA?s_U0BhUU~^xR#X=*AVNk}q$DVV^9fQ;+dJee
z+0+czIlogBhNwg$<^jS8y!G_KrX3>F$I?pStjmGE5@hzPNpjpI&2}Aq+u^KD<o|4D
z95W8SsMweGDl&2X76!>(h?v4)Lu6-vN{CRG$>B%%NZ$tLZN$aI-SaXaG`J86!m!u$
zT)n-|#S1Gq*Qf0jG=0}G8Z_KEe3g?c4{_~{*SP+@leGGfkA3XDY~SAFwU>7hVT7HR
zy!p}=JGZ{b8?QabkN$;^BIJw@f9R8(IeRy^u1)yWU-~@PF9j~1{V21k!_9^mRU>?x
zI3T6Q*%se5iFg|VkG%I0KKO}ub9((8&wuqruD|*M=T{u_t>;kT2ItPLGZ+mKB6+vX
z=N&P4oK0rWN`le?Q%DAbn(=tZU@)X8Oy&T1dgo}m4qX&PA?Vr;iV{MKblckAVr4L(
z&>Ah%cI*0{wr!KeicE^85PXUP%Ew#?p5POIQxFj%EW7O~*H%-Sk_F$TC$c-)=fcIi
zdE{O1VrO@UYu9dY`NmCb@L3TjQuLV8SR!qNjzG&DwC;}@OFm65^wN2ierJ2foQP7%
z1JRfQq@XSZg(`UZnWy>XU;Z3_`QQ76ba^$B$z;m<>N>&uwAd4ZyYIQ1r=I#SFTVUz
zHY|?IaFRRY&bTxF#~*zIhqn?vB~Bsifj(56fA}8m9zM*vIn5u8)GbB?dR~70CCs{J
z<J=im>Jgi#)|eh_vv<3vX%AUl9dmkkinGHrOvNrnXta<lLaelbB1fPtZe16c&n@$L
z!+30#^?~ya@6v;!Oa)h(AJY9=DWp_Ms3OEtpL$euN%}=m7n8k7vYg2x%3R#3x7eD%
zajs;@W!geVkTkp5KzX@c7F+13_)Cw7%c3sk)$1aSSPGfT!S5G${L$l`WhO<5ijm5L
zoXy#N{WUO>Q#!JG*Cah83WC&lZSXP&^9GUmGD(H%Y?s9oi|RF@cGYme{@xy*r1F#`
zLaf$0r$vZ_6u6Kg!a9p<I@-3QZ5q0^q3e3Qb;)!#gxq?~6H<;EQ(9A$MY2^f8eNv4
z3%pPS6wV0_4;%DaQnvW6qw5?VotlB2$H#y#BSC=`5esY|9AaI^f(;vj!h4Gl+pLWY
z)7h(3;v&y~;ahy@XFo|%7(V`~N7<izgRNT*LXC=RbX`Yb254nDe{PM<%@tmL{sn&J
zpZzZ1d*&)eZqm*?!WY=S#Z+l!6nsU~B)_8==0xT^^3X+|_|QYNea9=$zsYxg|7ohW
zMc6CsU;ifOH!T~hI{i&nDz;5W2$m32l7`WOs?v-`HRJIL!@+>UnB)hTmsZwUn!aaj
z3Y<{%ZAW2Bg!e?>(@iIAtgN5}#1OEZP0==8my$vd2$eEpV-~3RZjp52mgtaDNA`yc
zv(6X=whwqI*_!O5M>P+=>tVXKVQ+Vr*WbLtth0z1F#?QKR6?f0KSnwN4Ut|PMGD;+
zZ$FadNn93LK7`y3Nmen=dS<g0AtofTi^nKUu#UPM@w;FAT`t^nH=q3EC-EUEZEfFE
z=preifx;Ag<};t+bHDt{Z13y<fBHpT2=M&#3v=9m!$)2lS5?VyIHW8~iZUzzw92aT
zrAG5l>}@Q(CzC=tB|h|hpKi9N@6y-c!cs8scH`VR*xWq&`PQ!BV5TSr1+oAaEFOn6
z0#oXoej{_-S4vRuG2N_nIG2PI3-QEJi;lNappNL-K$<S9T4SU^33XHy#FSNPq{ie3
zsWj~`6uHN(EAqA>U0VVvr`5aMRq{S}BvLU_7_)qXLP&b+n9t_)))7L?i{C{oW}4Gt
z_8(pPKjzOOatjjgsj8CIl_BGC&2F=UkyBQyBU5_UL3@xK4`?7Bf}Cj{y7ygdmUl5=
z^^e~F1Uxft+`P^xj95E$lC{wqr_NnuuPBL=Db4OaCq^gO9GqmW*kHy!TBszLlq~al
zV=m|x8MoF#(<ILKU@%I@G-)nb#2m($;z5N7>4`{P8Ku<n7+DzpM!?`X%bX62#U6`M
zlgob^P0BUb*={A!(`O#u0<W`6W*xcNDvx1q9MiZiMWn|Lza#0!LT{@$^0!Oxx1|3g
zxBv0+jtV;|6(x~M2t;o=*xKTT2#8_D$p?=%v@}V<6NDlJK_N6g^oiWKu(rq_LqzCp
z&u}!t7{kHdK7>F!pV68`c*GdIO$7SB>*%|lu4!?zj&?F*Hk~q`&1sv4kTXt+DFm2o
zUMxu7Yf*qM45liR*`qO3qanp`1=)5;A=%hiV{dn#`MhCRlsNCvS|SGurz6fqZ~^Bd
zp)|B29Tx3;M%Nr>z!pLfgy=B3XJw#R=>$q_a$|qOH-6_?KJpVKL0UfY=|}ncH(uk_
zmv5s*q@VXFy~5{z>ubF4@kjaM=fA_h_~qYYXUie|h`#j%3nF@|G9cv~qiQ<q6H`ek
zq*46Hk37alKlUEZ-*uW-UcAZ|e(m#cYr=UGx$(v`48#t1pF7EL6j3H-9%{_Fd>$cU
z;!9SB;jm&n9x@z`DC-I-QvoJOir231nKccorb<dQpJn}_$M-Eb%ldc-&JhUoy`%5D
zWd2A-W#X}xUra4~ML?^JvI>Dw-~<WvRT2>;B5E9hy=KPix2|#Sfd?3_jM&@X=grq%
zXXo}V!KVVhFq%RrjF9ww_F#<A#q_+(1=63Pv1nPmohkDIp<CP^v#K?<&(mWTyKKRU
z?K{Ve2BQm1sZnyu>dGqfc1jZ)e)(5F#{&=E&)G9)DNRY=^%SNg_{7IiO7ZyPk8|&R
z_p-IUlSLMPx;v$5;E5-o>;6<<_K!8BN(IP!-}4|J{n*EN<dH{s=%I%=bM7oBPoJc$
zs<fD;B&9U`NfmdIS-ZpD!5+K&d)(gJ;kDOa=Z!aB=gl|Y<ki<+;mtQMG25MEr|u6t
z&;G$bfdAw_Ir{k@{<8u9^h;|z_@TS0F2b$;3KO;&p4Xgv;1nyvRYqpSkP#l(XODxp
z&vw7Z)oa^q-S$j&pxFTqID+L1V35BWK*4y-GE&Tx@YwrLaba+t4YR>GjwmT8tzpcV
zv&DJNa~~`5BtzC2ae@KsNNQr{Qbm$0nFU%Rv(U1;yU+g4K6|%z*xBEqnH)k44C;z|
z?>$Gb9wjxpsJL|HHedSEH@JM|76+4tqF7}*Yq2h|!;~p$yDnc#JAw>2oy>~0){J$5
z-#_5}@426!`}3dVCx7aL{C~dp-*934Tl~d$edTB)J75m70O8ZN7>r9k^Wh(5JQ)8_
zdp{63WCv47j8sT3IJI$_a~o$U`#@K3v(F7Kp1GG(tLG_M!(b?ML17d{vh5?pyl|K4
z`=&{S-F)WR+1+J49-(zXj3gi>ry+?Dvol8SB;@i+%ESjwOgWi9Zs073{0+W&^Z^#v
z|9_q14|)B=ckndtsIU_uro1*FM-VXI+2hu$uXFd~iE$jAO^~Wq>>o@yyu5`eYeu7j
z@gPF<3`+$<;(AALj=^Zi>|lb{hRN<eV_o8#j#)gwXoFQMt@ip<)V1@5>10B4IH#R9
z3H1pfNz?P>q-4^qC%cS{pyez+msgVu;p=QZM-~%`s=|~7XExUvR+^oyZM@NxRYjpS
zWl`aEpq*R7p~ceUN<%+Q5cH^wh;9xclOkh4s7MqogR0=visOx&Z*XRPotH1&<lCSB
z8b9*M_j1o&Cz$mOL3wUmyM>!+Hc#KprPt?t`FF1I&98rtLa);uSd<tdrA7;Z_o-D{
zlm(5QV@i<ajH*_=`;mKi;_-)(ec+XEy~fvn_i5a%1MXiPa_fb!QMeuMzjvLx&aLC4
zV>a8#E5C?yJxWE4))Yoj*A;`|kl|=VS=MM%AVil|O9X7+W38pL4rvN(UXe%Z5eh=r
zg6*iZLid(FBd*RaR;&Ssr;tbL%aC&yy>m&TMnp(~(hzgJOJPhF%?LUlc=^g@d~G;&
z@dAx)+1uG+>-skMfQW)h3yc(a=deO%hw+FP#|no(V?`Y#4@oKIvanmAa*Kj*@jEHy
z^7knTB;~&@5QZd#N@9nm=@4>)DbhJu76oNh;hp31)i?RvFMW=`_P_XRRHlLm&RUe#
z%jZ9R`ZP~I`6S=|_O}zIfIpMF<_AVS)|@`Q$vt;p<gxcW%H!{UKlj{oH|NivXMN=?
zWm#g3LF=@@Q%Wv*#D5yI)~H%%P_A-z^KKpqfsefJV@prW#cg|cn;W;T^TJCn@QrVN
zgO^`^nah{1uycLqZFjP)ByvR%>Xt^dbYezNhw%a1dqNe!9O)D_3L^}KF<Akb4QP`$
zNh)F%6&D;a=ZnPY(qs!H2fg#S7_+mPpmULn>K-1HkMbxVMH1MWw`|WkUOu?RwHq(6
zvvtViaK`j-%H*(N-g=tWGM~0g4jZEN>H60@G=Y`TfP3$|hmU^rNlvca!_}Kt=zL^j
zb&a3?(MLgzabNikFMshH2tC6G3DO`<fhh|*A?YM^tp(rXd_-wYRVaGnkwcJUiCC3f
z<tne-d<A1~zGWK^YB+T&ZH9giMjJH`Zd~BU?)`Db`iCp*AmZpbb#jwiJGc4jm%hq&
z8~FM+pTRX#=;w@vt32|+yEwIZ0jSaFgfP1TZ8b`zd0EJaklD&afbF60BYV3&gTXeV
zQ9`B@w3k2$Aq6rfN{?7XuW{5gU5FzV-~ZcuJ$Yw^U51CFOh|`P3Sm9_S8uX^CBeGU
z_6TQDTG3n2cfb8C-oafLPjk<`Cn=3$-hz)2Wemg7i0xZf@Yb@wvrC~gebwUU9ZlQe
zwV{cDX5P>?4b$n2`E*L#ws@B&6r(kDQ75YvQ=mj5)~oD3>AYJ8+J_M6`cw#7Yw^zG
zVu$D~zHMlfq^K)aM?(rF*xTC$@2Ja?vIJQbXjySM@0s^A+E8)Ov<Lxb$2F?&2rkK(
z#Q&eYKYO+;yYD-(&l>i0#ybx=XI55L6;PN#5G24^BqdU!)b4Q0tya6OmaXV;hr<v4
z3;fg%e(3OnyZfd6AUhm>aC9qLY)hg<N(2d#00^Kk)?AfUIo-VDnfI_JKdgPuy;%i-
zNT2{xl2^rvn{}tN?;h6rt?%zUD344fDkPZNkS8^}Ymx`2-{e~NMMiI(@`aOk_{6XM
zGA}%HooiQa@_WDc`@Hk^n1cr+zVfGUk*Pj*v4QnD32QhDGSQUfn7osasGPcLP^n{c
zzsv1Adwk?0A7X2Ljb>svdpPFz|EJ%hs2Xl`C5LbPA<DeP^LI9Q=2n+}M`C<K64c6t
zUC6NFDP&nnr_-U=?b9i`q&jDrs~3F0x`x`+(UK+vg4WnpofN!Bh=8dUq*5S^qp3<N
zV_7UJ7K;+^V(X@{HBu)8;b?3f{hYkVn>sqH(+a<(kHu`FND@t&WCS5NnNE54_?X>0
zFW~cpvy)Taed{iZ@r+14RCJX@wT9G9urvfJB3c(9R{8%<KknaqM3Z5^D&pj_&CwQF
z%YQG6Kkt1T3)Z$xLNtp8f!49@5qzL38%8Ii*aA;94i_Wlg5Xd7<d6Bt3m@i}e&Lr<
zN>P?2oo*KqV(}=2<a0m&IsWlK{>MzGvwv?ncEj~Pn;Yxw?d@^r&a+&*d70gvJ$AP*
zvAes&-u^yoYisoTeNx@I;3@XgA8{iP*+?t#U-tg*+uYdX(*7l`UAfMSFMNcP;}cHL
zPB}U~<*hg0<hc)h33zq&{dqT|w^kr>hY?k5nzeC~wuy$d5kdB?4lzdHNf{F(R#BjE
zPFGsKZ2OA)6r7D+B)M!0`lZ*Hq{gv0T42^SNiihz8~nka{Tp6+^C6E<=8PAPv+07W
zE~%=T#e6|oI-HkS8?cR|sZ12^g+TC*ZYsHX=N5ZcFQbZ#FMa)M42B(q6pZe^O`hhg
zZSAtXdkKWb7$5DTLNr?uGU_{*=>^spbS6>40Szu#yet{lC%paebq>GuO}_Ki%WSLT
z?|U6xfVH)Cp1b`leZ7McKfEkgBe=A?AAMs-Q%2Pk<rPw9<eh|@H!pL2c!OcK#?tdm
zi3BBM5iX)$5~Vy^dy*u!(47NkpsrvxTcBiX79!f-xja{oE^V@CKXcxGjg}^Y^XBKj
ztK+E(JB0uxa1vaIq(LPlX5KJ8JO#klHP{B{8wR}r{i4U?gHs+qJfY}nws$ruq`<mZ
zaFz2DrsE0I(U@{N=h+)KnH`PrO-*BJT#`_g6|?D-x~izE8WBPy;&*x!ogST*^QC3m
zOD7^e5&;s7Yd!4RV!<|svMiZTXH->*w;oqF)b$*hC@R}v>IF%jvbDL+cr;>BEl?sM
zkvWYim{*#!MZ!^Caxw`FWWipiM`}H;Y0a87cpOdRQ6fQ7GaP^m3(iJgW^3yOUVT*Z
zrT^*6T>5B9e}BOE@Q}~{%%@O7@|jP6jCbETz?vCNWl=IA_y$Cv=o-2`fom*Ru3n}I
zCI8|7>es271^qnb@jJ)7`qIlheB%M##&Bmn;nBS>LHQCd+!=6fzaUQ&s?rjqWKk_v
z=P1n-O2kOFqUg}+^vLrLT4zY9V-bf(d5<vxYYo;}YHN{7N2?Mc@GcTjT~)_<-dSwZ
zFdk32|M-ZyHuSGui;Gd~<>UicYmte@IUh5HFSH1y@MtMX+cv#c8Xp7?j*b!Cg5KU1
zqh`V3@ez+69-<vs1i4B{L+gd86}7We5-JDHN^yB{_W$>1EZg5J6<O;oc@e>L&aM6(
ze<pG(WjmoRzt=ezcg#o#(fhDz8p?8k$|O3`NTH~zlHdK^-{skx&vE72Ra~3rj69D8
zeDTE(bLGkvUVi!2RvG*6{&mT+l)NbDb~^O>eXd=<#MNt8xc$sCeD>!)%d^kkW@CMe
zZn5^4eW*Y3=oB3~MTgz3U0!_tMF6aKoRw3)_VqvK&h-%&3%m7=0ehFX(A|n!SsYQp
zZj8s&7H1=OJG2Qw5>P@SLz^Ei#8RIZ$FjIvqIT~5<xFhFhPJSiK_G%61VQO7M~7n;
znZsm?+QT3I*-O0i)w>)VE+}mhO^;I-v8%eld5M%MLa6qcXx}%4*lsKmu0C^(OE)j$
zwdb|BU*~5&`BAji92}kS#=S@M2fOrpePn=OqDTOI<dXZC?xd_k8HcUwxUmuf6M!Tj
z?2NOS=Iw_$i|KtHj~`K*l8eKKXu!YZ0&Z?>@u6owjG}u%YVd=HK%f!q?d*|dIcwbo
zrGG%yv#4MhWQrG`eU3}r%k)%_Xab^<GC?Vg3L4S2Uqf(68PGc5mi~f3V_-3FNE3-t
zkfm8WXQ}rc=Q40Wv~i^89&}5mx+h1C<T3fh^6<ina9#+mgbvI9Y5)4jWfH9x;QZeN
z6=N4&Xg{_hia_k{dCu>P6~Cn|KU*a7`o<WZf@k-1h@D`S=->&_BhG^+!Ic#N<#dX#
z*RY8t@AbHG{Tk!b8Fke#pVefYn&Xp-*{medil%Isj7B8F^UT%Dn0ZZB<alG5oQ=?J
z8@8@WvMgt7e}itXN2l9EW;r^|A_u2!*Z4NZs;(<cyU2Li!n7(n&17jx(J2@X2QhtO
zQBjU3(Z9{dNy3>LgrnE*v9U2^JQ_0|Ens0N8^Kwban_{Vo6M<P!d|zan-qj*+9HYI
zy}^1%rgK8$>ExQyS8NPwI&R84?@YPA^@|+2jJLo13RP}N^T3^}x7fe5!QBV1@^ioN
zVNMQbjLu5x+LL7kU;L9V@V9>Fw+W%4lMlG}?lHgjFaL-?{ljl_^~yef^o^GgO-VgE
zWIa>t4^mEE`D3cHmw4e$!v3Bk&1#GR?-bL;g2U0{c2SNFG+HY<MM1aIqbRy$X^s>+
zW-!vm#rc49v51d$Ck`Rw!mV0#C-@i@=8U1PN**5^@YcHz7*7^#Y;U4O{66c>^{CD{
zgpxRG+ZKGYTGUmW&!e<lMe<A+C1>S=-Rsv`NWpY8;hlHirL1bQ0EJAELebO}QU;_>
zFd<L}sG-e&j5|C3?^#g?w1&fRa$Np<nE}20x)iZZDp%f;k!0*#RH13DS1$Up<wRQR
zFeWgcEf89g7AaX41GR;cy!G~*{QkfGSNzBS@!vzr7H!ZPsWcwg+T7x~=bq!0S6+>t
zRsSB0o40T9(T{zEpZmGb@*98c*SUH9S&E|bmwi}2-uNMmrj?Lvb%y-L=Y9?Nr%!zC
zg%3Z^Z0|$7HhYPt8AB;C2C51c^NOZtuzA3yOKS;*76}PaAg5$o*ew@N$13lhnkNJl
z(d0J9jiDeIfe#WBptcPUA02Z#C^;)C9_eFVd*v|)j~$~~j-kLyk2VX0aNrU^w8fWS
zwZ*+{kiwIug0(@1OZyuP2Pu=YV?O$!+kE^(&yi&X``f#`bN4Z&Ii)%|#gFIc5b;Rg
z=GdajHfFPdYKAm(W@if;Dy*nMH^|J9=2LdYf_yOJnHTrCa#H~3?O5YrR>$*`r*QlF
z4Sw}Azsl2$K)~X-w71Xgz2`{WfSsGKFm4>fZpQAQ%N4axAqGftEIwj(NkU>XQkjFW
zQILg=wS_h~4O)Y2jf<yKP^v;ogP=o_W-Jp&mg1mQStg^XMzxkC?!vJyDO$^Mj}HH5
zK1xnTif+B>6ahzr!L}wRvG%b`l|Zz03ktl(+mvRJ&@57HnP96N*9@4RIi@4S=(u4v
z^#rR4M!`bi=Iwpx1S%x$^=;FLWgzKF{@?$*f5I=^c=Lz87Ef2$DUS{yOqAgYp|B#*
zxDa;-!0cqoVr1xUCO8dOZfr9f@9@^!54r#50dKu~!f6>eJ1a?16k2fS<_34}T&0&O
zluMY7rktKWAWanAydY0fik%%g!yb9Ji^`*=hV*eq6?Z>vfW42tX90(C4Z?+$V$Zwy
zkm=S|A<YzN+QoM=hHHIhvsrvC=G3Nv+ELmWS)Q^vSm#Jgc|2}7TX;^r<X&l*xIm$6
zvUEl%I&{(kA7VhdbdtnrS|e<nZICi@*W{XFC)7MR`hB|jRi5czW8?}Bj|E@&Ctu*X
zkAH&A8$$#sNs;i$=WpWC93CBU>0p!RKX#2T|Jh4Ce0av;gE^v7-2c`+$~RB3XQ%A+
zU}r7o=)re+{qC39P7nCVvzzSi4M-9}aDm1)fS_vT<VB8@(fvC~0$Hk98+7RAIi0*9
z)lvOrfF?+=3gb%3Fr{IRaTZO2b^_%Mgc>6RIw(Zh;1`y+-hIGZZ@kUyv}C)}VY@qs
zJ2fQM8l3ku-eJ8%n5ba2F2z=HMiv2-0}_iUG+L%eXL&rEGVc%RZ13PE74Ff9qr(wp
z%h~Izgt2o}L6afrh6Lj*s1QB)AnNSHc}&a)GM2yYdp%1RHkV73;iok&?rW8buh@Ez
zRwYtOnyMj7)GFl#VycAm4k=_5$Ph@9WR*H0goyT@AtuluG}a2t+#wS~Q8Y+q^m<*I
zrXkfCzyHO5!Dl}AGu*s(JGNEYMK{$cd7ASJpZ`4n^ml(3AKbr_fL(WxbK}N!KKZFn
z^E<!uxA^!+KF0dm25DB1XBk>QopHfm&EW{a&=`ZUhPtk4nufZr@B*xFV;>2mX~x!Y
z#9HtCr*}GgydXZpu-oHteS>SsLr!PM%>A5XBFSVIX?m#jK1nVJa)AR1YI5Bp(FUpP
zxiD(^@<7RXv1foZwjvt^wT5Qy@Iy;TETJ)YW9g(_1e&+*-{)j9#e{%LQV<E#a*C@R
zTEr+S?_!2oyCRWjTI;V^!1S{W<t?sU;Ho*l{41ZQlV@0CSt~Loy$+MnlxlQ}E5`^C
z=RGY{>_(LpB2o16l&#@9Z{0t`oE&0>fo=*N#qH}i_-nuRv;3XE`zw5R`bEBS^hIQT
z^u!hN71%S-Ib7e}<c0n%p65ePb(JNGH8yFmDAtj3i|4mI$Dv^$GqyyBR44?Ri=~H3
z6sb!|UCa>2*+>s@9wAM%!ctKr1W+|Fn}d`dt!ji;h>)-h-+~sUQ%e`Pc(LOgyZny#
zPbPw>p~(m+9C0=?lsH&&zX^COK15}#B*VK7&Zd}3u{aZGv(53Nn$eMAcHA(2e2A-3
zTqQAML$DgHGg6t6`<(K)rbu!Uogqa^2#Tg@P)f13He`2wlVSRYPjxMxZq|@(KHE|p
z0NBtXckQ>7(>W&x$MiZIIA74~<vjoVEjHFSIGI!&PAi)699v1W*IeBna_8B7lv=P@
z%n7w*ZEc<P&D*4@qSxt0cjq)37YD5{cp7KWKKknU3vzE%n<RijIDxk|!q+Wd$cKP+
z211Cc*fu{%rwP6F0Y$${XR%;9oidwESCddy<P^Pv?R<lWua9{+8F6AAUT89EkPQ}x
zL(nu)N~eWF1&JjdQ_QXfEhI?G#!yk@l7pkW96x%*VC@-tgDadZj(F?OzsM^u32JXy
z-`e0M`AxD;L8segKi}uOfAkiIZ=7)d?IY%A4gFfMrVLw@Y+llg55L7bcfWxtA8>0w
z<>vk#8|y=qk}+p6ScHV}WJKL8NV|G9@hPb&vVvZBKqu>xC2>a-Tcfs>)eD3Xcni)6
zgilFDf+9s&kFy@D6^->&$1~o2<!v6m^)_dt1$)CTSNE>a>kqMMMj%kv23thw|I)O!
zsT_4}FlBYFyhNZ90V=xvG^XLWoFO(Yvmj8<OOEb6!c+~`gHAFU=cuHoyRn0tFQ{!D
z3FT6>F5;|Aa6i!T9+9b#ER)e<M(TwM>wCWSPpC7zbpBCFpj0$@j(N@TBxsq$7#<z1
z7b1Ty?oN$qu-1_#8r#|~gy1g-EzVcYIJu^HZz&fI)5#q0XdU0vV>vi_#P9xp|7ZRW
z|NH-AEDniYmO)5TrTOTKAEw{$adtNTU{Atd))3GeblJbO$IpEF)4cGZ=ec_28dtAe
z=gyrwY;SFor9VmK>IXlndcmTaGoQ~{%;%J4NmW-E6NPQo)L7eKoox%dhNdxCV+c~<
zeZaLDplO!z^FMo+wHMBR|Kqm~dE@Q3nb`>?ip*}X=5sotqN2neImUTI)gPgIg3xWS
zof?xl#8e<_g|w~h#PYN*Q9DH}=$1pouQ>%(3Uy=f-hww^8iaKuLE%l{==g}~Y>xA3
z%$in#qzNr=J3ykaCR%q1DM++zI}-*e92)2kx)j|GNg63Gku#p5geEI2TiY8Trd+?d
z&oBPcXP7T)dfh%rk|BiT-aB`>bZL*B-5oZzHu&;4zrmk=>5H69&Ty$jNXO3RC9Z5=
z<I2WmzBy{J(&FWNCMOS~hQz_0ooBeZev^c?r@9X!=$VqxMUa6|6sll@J{>ga%9vNN
zBsE&1&_g8=9gCohAP5OkL=K_!NEH>cp|wIW2C6zx8%yFHDzwejrAA27uJ7k1&a#Di
z{_)9)RU$~+<6QFc{Hl~&8k~rT`gsjj&@?HHS;LwxO_?w~ay)*0!Spy#j}_)D!Ov21
z(L)Cf#?kL~dH(t5*xTMA37T(y<u$HdyGF0uV=|fZ(n~KBGR^wNIxoENJTH8Bi`}iK
zyB1GpcY=r{<`%N{?ZX&+XoHubT9h0<e8kp<CT#ZzO0c#zpxYU+T{|`>Gtze-a<3{0
z)-mV^w${4jX-XnNlg5f(JA$+EnX{JKM-946G(n2i8sl8x<XfJU6cKTdsX+R8S+LR5
zE_(h%)s6Fk;0&TQF%F@QO3O5*H|*2tcUf;Xn9t`_RZYOtUF&gWu*aK^USWQ8f{?w~
zD;F^u3D=-hx>~T6mgq!bYx6|qAP8j2lK~y2$#-&wMd1GFs~kOih0fMBwlCjiVb+<H
zfwOPF$-JtNIzcB1-a6h<QMx@;8P9C2v$2+A7ss3(eup={`wfJ<&suN6rQ0ccJ41@J
zi_&0ijrA7m4aQoUx?*E<154GC;Uz`ZXVBlEllM?kK@b?{;)3f-2vNyOV{sM`3Bn5!
zl|YDKx3`)g6u|}FdgC3w_2=Iv4JF%qL!P^FgRASigj`Yehe)9)t0@EPXl#RNVgiY^
z4(Tkx*70IT*p)nyWT`}_36rvBN+93f;IOJVolH19Iw6>7Z4#tnW-D~3OVP<$)G;>4
z1y60`?mV_&{143~`JtZ|dF~n(0xQbEa`I6^AjAh<P5$5^hy`7e#CttSl2$z#(YeS`
zjpB!;|C10gRV12zDx7P`@(d9oe=h#5a`o||HA*b{pspL{bI9_9#iF9yEilHDBpF}&
z(x36mpZ^s;{?U(9H#J$36TCx6$(1Wtxpe6gqw#ooYyDd}gpBjxaD9y{S1<A5k9>&V
z{;l8U)1Uk_n;Y9vl=N40#Q8C%gjjp-QMdGAo6g3Jrl*X~&KQlxOlMQ(^C@*bp`6dD
z>x#zIc<X4ap)s~KkZ-Y+5F)?V2UMEy%(aKSc%iU6c>91?zw{bHNJu0iRj8y-o)nlM
zslps1D<;_-*>QwnL+BYy&rmf6Hw}cTVC6t=FT5_+H@f}0k851xX#L|%EbN*F6(AwS
zH*k1##%x~Utwc(X6b;>uVqrxLqxaEz%R7rs6ltCkoJR^pk|?^}gsq)ThHC@5os{{!
z$827bCmobhBw0>gcm`{MXP&*z+Ip9>(Tx6JNS@_LA$je!O>W-0!S4P(!;LMfO!4ZY
z2h0yLEDq&sHaB+J7;Mqe9j4_JLxUIJqp-7}E!^IFhW(AJzyKgY-@en_lgoKbp;`}_
z03_rL=%9%y3(N00N5G<kg;EOPCHU6=ul1Z2QXtx{f>iD1xb?R)1{&kB&ZCz*f6;P<
zr9>zh*RS^+;YCZ}Cx>Vou*n6v#j?e0a3KH+=M}O^@pVGI$SLRRRAq;1Dw!S{-uccH
z<~gb<NUAP9-(!2QNiQo%hJnqsE}#GSr@3+cY6McIk`I66d4_9!4iArc^W`^?QgU@?
zlTUr*4j=x=I_rZ@ynlbl@$}*eA@CwZ>ktuIv?IDCL);L}##6>;Bdpm)YK^ld8bvyf
z3>A4;kop-(qL6MvU5;7r4^bkJ>Ws6|1ZSe@V9TULD@B^7cw=!P#N0|LS39uK9*58p
zltOAnrX{IK!PphK+qYaU=OUNR`>?XyD4T}bT2vdrthHvivCf)^%A+ikxFlhFvrl&K
zl(}~Zq0lOgX$qzSABnchk)|5!EwyP{ltdA{1@B3<#5NV3L{e<#40@U?6&#J<=H8!w
z9pn`@cW-cca~&7JxEdiW!Go3_Ej3kh!r|Lr<Y0P;a8tCLu{(sFt%B{1A)Q>HwZ>wx
z)<lwNa8y;v+31X&ogI{tkssHROnaR^-J%=6SIWpabQbF^K1?y8#9B$HJ&UHM)-g}k
zD<G*rTjui_qSL_#!O77H^Lfe5XE(Y1+%<Lv>j)<>2!u)yO5<$Y99YxP)D>Bx5d^#`
zY05DX0fJOZ@>HT^pb3u0^96Eaow*2%rc(}2PN-_rx*uq2>mbkR4c8b=CV0~z33wqf
z-r-iLShzr~{qcubJ})5>;)RIn?EulB*~Y~!LU(QC-3RpEKWIo9+oV~RlBUrbLrO`K
zXuS8NX@*vsCTpUIK}qVmL8&UXR|!a|(21rt&8mO{Rsst+SJ66eS1$0zc;>SbDFsE5
z6GFh4hN&6zuYUhu^6WFuP~=^dR3uu02;_Or?c29``Q=ymw^r05A?YOijlc8P_}jnp
zxB2)-KFOV1AO6ces_!}88yyD_a9Ce+JUZd@^pvyHGfvM=IXgLHG8sjug0jS#=pW*}
zBLs&K4b~g1wb2f#tf}id(k@K=Tvb;DB3))lk(n>%PrU!>;VJiDzek>SP>CieKudI?
z2!&fE7nQzZ=EltIDGR?~0oVdc87WLFyA#8zfLqZemOHSt{n-A0V@jN_2rk7hD*9=k
zj$FepGVZ;7%5-dy7_=^tuI9$gYn*0*>10Z|n4y$Mia@{H<MQPVtXa@0a)yI0dwUyv
z?Bg$R^Trj1!#>w8U*hdI-{$0G!C*LKV{?OpgL`alZgA~GyIk7(G!XxmW!n%wM-^aR
zJC080Or{f@1x11snmp0$Ztc(?_Gm)Qw44$WhkWnKA86MV_|UEAxw?PlN4b64(WjmL
z$k}8CgRY>_rHL~axN-3sUQ~4pA=*u+L+Qx<RuXJm7FY*O9TRWTG&)Z!sgOF}OS+8^
zJO2}xsN46OM<js^OQ$_%kEg~ki^kgoD(jGoP0XZ&o%f;MVSLcz(Y-NehjYqPgPL!#
zo@}r_SZDY04w*}M@%cMky|lwxC&M*Uu3Xun*Uza9eDdQja{We^!64<z-afy#Il$V$
z_GXVe&#WUFexRZE_Zm+ro;W}0v33yz5!(2R_QD{sreQpuQ`ZKc#;6dfKsbx=HG>p(
z1~L7i+m)O=et_~78yjmVsaf9~Ad?j59ZlU}nh4$3#;i~Tr8HheKN=xLi++e0OC)9M
z2%R7@9}7F@TU!Sgk)Tze`GuTQ8~wYyh^D)ZX^={`kt866Vmx|8FNgj0E|cRLk0O;R
zinxQd)~?Re+TVPK!7m@Q1*CuS9JYxVExL;k66-vnW%8C)jj`_%!k^;3M<<bpE5x}K
zMdUUaz(@JF2xuK0yt6bxwpJ8N?h_%x(;-B$g7=;zS+qNHM>25KCmDl5pQ4*bWvllX
zXQ`W-s;+44n9`S+TH+UiSzR+%4%rA?$e6p5HEA$rK@t?EGE|EZ+ZzQ}uk6t6rOe6^
zixS2SB>Pv{s12rR2;SnIrLJqTEJIs|a1FkiM*?<$;2f!vI9GEvpK;PS#BhU?*^KFA
z#@TpA?R|vO9aMNay#XW%WqsP-gRMz2?S_5Dh5O*#jh14JYuB#v;)^eGdV0#?;UP^E
ziy`M6bzPy<7;vqV`cK|nNr?7-mO2(_Iw8w4vMj?GgOrjy%~3+HkUUrbf;3GDA<#5-
zRRAoh4A$Bw&^l}5;-2W3P!fc0<LWAkPQhYPlIIx^XiUXdzxpNKeCx0Akq>_q=L{;u
zU2{?7+_`f*o*(yb8DuBmmwx5*eCD&C<`;hP^L*|zzf6%AZGrq(a|9ZWjvn*)@PMZE
znRLthwP{*wkwBJaY;SK<bPD>5XbxYN3(7K5S4>m4?be8_*p{;4y&!~`j<XcWc;{N7
z%~FK*q$~7%I_LP{m{ezIl^~QtD2dPz5{t}1W`QKh&`pA#D|980WfVwN5*kf3BhEQD
zBwkXzVx19dkEZ=xOPg@kU~Ppl38pk`b+6IY8#I%Y`)^NZ<~dRXbT_cQ*5yC>yT3)E
z1e5WIx8J_Y<42E?D&f-QOMLdTpW)J_U9^@Yi6l=o^XZ6G3W6~V`hC{ccCcPyY{KbD
z$xGk*7N7k1M_C`93uw{{=Oz%3b96T5JKuhtFaPOR8J|uFCJLGo2|L?ctgQ{PSVoHz
z%%H}L_dH7qbPC&plsmmU?4_5U`p!A8lslYP%HqPui=TUsf>lIo8Y|D&$mPe#`}Iqg
zE!rgz@1lWtV+@%+2XCZOk-M#;$7vK!#BF;+Xz4f0BJaX6AduuZI#@i08iS=Kc6hu~
zgn7Z_QIAmVV`oFA4^tj~>oN7DLDaD33a)M4<oTP=@!YMOeC88389x>zQqd_Cn;Y=*
zs~yfp8E0e1VqWs*8+Uo>>l3P~qG=q5hetRY@Xqls{`r?#TN`rub2IjTSUW*}1RuLt
zP!MCk+V)rotbAjP^LR9)>74nZ#`p3V^wLIsq)O1w66`QTr<xEf<-F$IMakLe8J%9v
z_RcoNV1uI5rIY5ErXjHw(=><xiPDj?mMDz((XXiW)mo{AL))7TF3Q7&jJ|48#>wHr
z_bF=`PbbuMG~HEN$HGg92oDqEs1}ok${2(VT;J(&wy=yAhRo*(7yJFSbx5bFPb5NG
z<eEdZUOP*r;qp9QII;1zUmkx5@3VgjG`#O&c+Yplu}lPL&TTAOEGV?sOGrs3MT;z&
zRTPbxDBL_j=P6zYjCGV%!)(5ws%mDlIrFk&W&);Bs9Lcwf+?2735@sLII-+`jrS$q
zDH0_JcFyfvm)KbE(KPdDZF5rcXyn+x^8%(lK5OH7_Rg;ORsu&kn-ExXB!S@Z!E-!c
zV0+t4g=SGsnVwCU&F67$7CLUqwI&@5kV4_w%;E^GN72LuaP`MV%q|-HDy8Ukdi=e=
z_xJcm|L7kvo6UIn<(GNsrI+~H*S^lt(IKa&r${+P2v3q|sw!UN?}eKDxW@#sO`4_=
zsmrpAJkQCpoIH;t{VdH$vlNMps`e}=%~G`1oSvSNCTZk|^?Fow6}gJ-d#sG6tvxr!
zWjQ&YZ~DB4s;nbg-y((1S&ohl_|ljDggejNAxptK8@ZlEhpShwk)|p3Jr)grRmb|q
zI=7#_#n1ix&+*yMewJr$-Qw!CtE{hW{H6NLT^u2JtgU$P;2x97gxPG)WHLf2Fs1<k
zFTVH@q$+@@l6EUI>g9a}#x_`EvDQV~BWqVSMa~(jdck}#XEvR2a(cw!;Uh-l5!30E
z`D{XPOT+lMp_@)d%;)bx>`o`V{Z4}>-Y+pKO_QV=l}l7EP^m!bK#&4iXXw(RCS`2j
zFX62ovF6Y$S@Q97U7lPCmtqPLJFu=c*wA2W$8L8MM8<sXI69aUY=#dOSxAQc4$t4Y
z%6ca!n1)Y2cb(aMju47YzsL6e9$Wohd|n7t&T#kb$4CL)ZkNr?Ep(C~T)PRHPg!3d
z(&=PK1*NHDM3#s>NgtrH@YXvI`Sy2T=Zk;(=RA7&n5HzyAkb1!q!}BVYvhBJd6+RV
zBT6&H+9!)=t)bgfT)TXOVY*Kyo}!8X&U@S`Pi{~(=Y`yoTX!x)UV7t7DI<vDBfIT;
zJ|-79$B_!w#*W!Grfp@(BwNn!lew_*N0(M7-^=g>qCr_ql4?RjgTeWPYGE;Rg`4KM
z@g~#bg2joVe5m2DPv0l3b^F}Dew~~9*V$UzVt;*|VbWo+1CJlgc<^}2w_iHqjknJ*
z&S9*jtZK&N3A5RZrf%>qFqzhEwC53v#WxwQt??Tln*1aSyXBvLA=^x}%GCBGxgcnS
zWIS(}&TEW~H&`?sg+yzT{tzV{bu*8<EZt!~ubED#oSYtVd~(Rz#y-1ym)O|aL}wWy
zNl5a%E$mc$I6^@B^YFhwh+bH<=uQa8(3VnSnMEq1RY!|1tahZvG8@OxjwDH9)@)dE
zOI(aQ@*?IG`oKog<IZM>*UJf+b4V{KD^FuQkN-H~;7`^mx&<2>8w}RgNVA;9qM<5l
zte1E+)*6)bND<+B?^?y7b1U9r2;f9(vlQa4M=6cB9vK8mLck(~M=9`PX(Ssc%Zd;L
zX_nHM*aqno9rB_;C>2}UE>P8V+>K+&vVue=EG7-pS;=TTV>FsFUsTk!rLH?vrodT2
zV!#Mk2tyZzND5rpM}#iUHAq`wtRT~hj#iwVo^d!nWIkC?pBYYS&8<$4rfK;LA^y!i
z!~!8j$Lzv3OEHA#{w^d`)-Y`h^fwq+p7Lx)b2_IiO>{625*GwcD$=5hwvn`7H+4&f
zJx>-|jvrSOyKFxyrAU*MTeqL(Km1?*M{I5Hva^4Q+s{46Z~oSAb9Q#dcfb2x{`d=j
z#9;V;15RQKp><NV*8YX>^|V8X*q$izPHe*_8F|*BC<;2A4xLUXiiEO^A}c5t3-Y|6
zsw#$qb&9-;l#=;;PE%WyvS_7o!mSc1;<I;-CMzXz&JcnkNs`D%4FS_Q%0&fuLP(Ja
zk|gC%|LlwWjo<t&F7IE(m>QLK&?@2T)oT<*#$x_gyRhqY3N|)3xO3-OKL3lK<3IQh
z{tmZq-J#d*^WJUPPkY3<%u!VpbzM`}CDZ9CufFmfj*pKy8;v+VdQ6sQ^tv5(clX%Y
z-X%?k$O}BBr+v@5DVfb?jK?D$A3oy#{dYJ%Ip*x_l+)8g4h|kNozIxgr;Ns9PEOCL
zswewSlWEDpL5U2IB$8Ce;x0`S(mX*X5~*Whmv=NG2tuO~QxJx@an=xOV9q&OmvcS|
zNMcQL@zL|(YOtlnI)iUuZLo>+ipjL$?4&}53<7AW=@vPc_BPn*cF2_k&fA$L1S&i|
z;xk@Z%e*nnXA^wz7~7C&jWG&B^jfX!IhXc!=yfxM@GR<zPN%?Fhw~2O0!L>HzW&WO
z`0Cfb#aF)aEym+H!L>13PSDE>);9(uUB%QNGj}r@JIDDaJ<+s=ZdY;b(k(jaCW_)I
zuC#Aw)CNJ9K-uDBiYRx3bHwgqIB#sl#*~Oy+=+HwUfw@X-cQSaT%cTsWu~l;0-UxJ
zwk+pX(WNYlI=}kd`)B%C(OVX)<XGb^-Xv5LK{e7a?cz`SOdcqvj|1UYp{Hx~(=NN|
zHXpk3A}`#&fvY{*1!gBDS~MIyJmk$cAM?uVcR4sZiL@_cs2WS6a+<ov*+_>%D6F+i
zz2W3^PPaE^Hha3c@9A5yA)X*dh{;(?HC%g+oRk=$F{a{VG^5<vB+Vouh?x7Dc95bV
z$pn)rXr0o@*Xj2XHcZK)Jf&)8%*GQWilQj!ZLF<OEkcZd3T=h%d<;`wap;yU&gJp8
z{&h>$-m+D+L^7NUxZu%BlPFCfL~pg=uqN&z<ANm!J|%cVz)&QfOPwA~zhE-=kz1Q;
zj29T=q8hC6WGbgq^huN?*B!i)ET(gWknw!9_AJ)B6?t8RB(|2@`7Hu|p*38T;3Ueo
z3KkJiLbpFB@Yc|HLl7P%1lBc_vzj#NvECbyWjT#8RF%OwXc|jdHIdMa!26oUluRZi
z<H>?WX`>0TQ#hv(zQBi!U>*2|EO8V)&1T+ZbMq2c)_Z7LRe~4syp)dP#SCvIOqwZ8
z?FpqI>-D*O`5H~#5Zdb~IKO@oPM0h6H2_kl1U!qnp^<C2q{DbI!%Zsuw89uqC1Z40
zZ4IjE^Wf0|h4D0|#)hT%y+!Ljp!If{%CIc#di_4X{?~t#ThHD`rcrV1LO^E;JNvtA
z?rib1Kliib#s9!$GUvJHp5tp@|4!@S^udjbXx%wV7ZJNor$e{fr6`KkOKZ*AV2z1R
zR>BJJV}#psvy-MNoo<2DlEq?<b8hwZBuQF9#W|VThtP^HBI&*?YmzknUgsTJCwOmp
zbnu98eDfRZ?(8E1q%Nun_xJbc^?IBfk3P_I^Fa=!6feAZhrj>#{~o{a`Cs5upZFQh
zdwf6r@%}~8Q&lrwdF5r^e)~<{e)~<{dh7KlVsLG=UvTs~UBd1*yE{8%+4_%#FbW~a
zvkrOIVQX`TXKvkDea+a0)A0e{_{LXw<IUH3<<(dC*Z<~_#zhX*6UT1BH?0eU@su>Y
z&#YAd8oGVxW|}OCX1O{9guPJIDWD*tbp?_3RRS|2Ieo2twRKH1gKA<ixdoH6y>*Fd
zYB@ffF`L$8S%*X?bh`;zny{9oB=UQlv*7Tq2HjajTPgxKuU}(2pVQP$o4_%Ro31>k
z-|MikzDJfSjHzhql5V%p(dmfEtY$K6_|rf87JvLFU*+!I`#e6HfX>hXWC&<4*%+*`
zvAKpQ0*}iBj5auB2=d8ot=7=*YHr_no-7@*WbplPqrp;QX(;Gn!U++&$UwZ;gK<e{
zj96Mq7L!Waax8N7mz7)miKRbgizCHC?Lu+qT#O!Drc1GM$?za#>wECMq!RDFP60wi
z!vaIV2C$7_G0K=ePB5bm9eatxHv^ML20M}zRYKu==q4r1QWlSE_CJ;J)*Fv`a5(0{
zn}K&9+~uvi_c%B@;pA*e<pLr}2q@|riZsER$YqyOMpd<oHA|{Bi$$AH{}kh?5xXU#
zu{`f$>7CFHFBG+leet8y3G=e1qf#VM0HL#tARrL|q>DvR<4AN)uREkaNT})#rimfQ
z&O7S5#!FUwjNoEEX|S#>!hEz5S*rM2(zk4tF+whz^X=^uB+h$G6GaWyTBHzkJDpX5
zsM?d7Y8^Qv)=^eEB4fcIY)wz4>~;(6q9Rx@UNfG>lm+h$9?Nt(p*QHGRDzHxN^81B
z5pz_fXHhQN&CJpic3Htl1kwIu=~6DF1Q89Gg=n42gNd23R)I+HPNGzT3y!i}kmZ`a
zy=^=Rv)O{#Y))ewWz$fWHH$?>RX3ECVOB$3N7S(~V2mV?04anRtCI+iN@{fB>1#{3
z4{Q59w)291KV_{mLW&u9fvE$DQ8Xo-9WMy3rWO(<Gj?~j*?RU_uH3xE8;ixN=yu+>
zPVw>j1)>^n^(d+YPfck&$=ZOk#<G|<_(_GRqOip4$ffX7kt8WqS(4OItRP$Rtpip{
z<qu%(U2YasEAqK^^(w#mtG`B)W-D^H_W{*DFG_2&R#cFs8P7fU9N+%->rCgfAOB$9
zGX#((37w)K&vWuTe*&q?^E`6v@_a>f@Arp5pwsEnG=}+n&U`*2&x?o#R|~Q<#h9k$
z-8HMiZdv3l`LN!_!meoy&bd}dVQCtW%A*Ac5nHlf{Nf+;%fI-m=u}{gMXP9kw6U?l
zJ9qDY;OFMY9lLuweB`4q@;83#H~F2v{Wsa(*r!u;Irq!@%Ny2JeB&Em<^H|9JbZYM
zH{X1d$z+UaqGgS-aYIGB2>Rfe&&LSm+1}ayi5{C)30v!z`K8b8@VU=g9zMEH-s$rH
z{D1vFP-@0g<K_V{usDNK#Q8!9@HtUzAvm`;DOrL^=dIWRH~|L4!mc0xJ!#;;OmY3d
zX0pb$wcEUXcg(vF4-l#*&7&ButQsCY3cPXeF}E*oGbo-^luDsUbx!Fl)7hN5iJNKH
z1QwG80qA!7n7X2=XP9b1$%2FOl<n<py1fBu(c$agdYwP~!k2jUwRd>??g8)IJH=Rq
z^Wex)szwlyHd2f>`du~#LmGF&YmZ*XsFKDtPZ+`jWQk&Z*yHxqJLKv2$btKTBcgS_
zEw*2oe2Y{{wz5lXC6|_n4Daoc#KJC(0DzAue+%lz;&-JO2ulqmOAKjscZ68f#f{AQ
zj^N48OK3Oq%RXa_KYZ_S7-k$%**VrLs!DP8Sg|<h&~dj&%56^G7O10yTis2zcec2(
zb&XBEL8uz)`GU#lm@j|v8yp@!W;7mA))n_39x`1t%xjMc8BSzas|iA*lN67NK7HO3
zyd_ODq;YM0V?av1;@1DL<0%WfKoA0lT#ES+M7RE>-o^QVvzkO?jL+swXLCB*(#bT^
zMd&rCfIyNY9Rvo4!1@ZTq}wSNbUWZ9k01m=GcRM|Lx_Bic7Y(oY*LkIQk|gEmOCff
ze!Y~S+T@9glr|p-*2UjbTfB+n^)!!#WbiS8!Fz|&F}j2*T9gMX2~I>r%2f0d%^>qE
zY{1wKv(l4GORtj?Y(+Jlu{ayk$vdblL1<0d$w{@OX=<7f<G7Y*IPw--FSn#2h?w3W
zd}~S>1a-YYs<<mL)}Uo16UPJxOI<bST(PygMW>t7IKyOm%6J@GmD8DFHn)@u&wOF1
zn?T(J3@JWnqPj>#K}HCOU=YEfwWOy@iavDKx@`734A&B}uB2-fUN$toB&42-hK^PE
zY0FP^4ozU#$=Tmp=f?FL`2EYMeuw$t0i~B1Em1C`OHM8loU}B$iHK={AUz?e5shS)
z_DDAR%*-6dGn)CF(wK<!1WD~J!{J7>x2fhJBCn988kTU*uUqa6QSC$jgBO*H_x<W~
zt+F*!2^-xZAAR9PKJwy+(Mc?9d<bZ*F7AP1a)uBXYZ#p!1It!)c;dKz)Nv8HPIW?&
z<rG;#r|2*k3@D07<nI(6dcEE%12#>gL24|(44sZ{O}@_+u*TFF=P0rQ5d_vaR35pn
zapxRq4&83IwSKS&8CDiARaFtgI5VUOpc9;_dF9pbaPR)R+`4reYa5hTtPT2XZEimK
zw}}syi}<udO3BAR@lk&1SALOCeBu*)>Jy*l%BAb1I{%SdmmkD9K7Pp2(PIt{9&vE+
zh<EP3!Dw{MWHRDxG>RMI_Pm7^XD*x<=iV`$c3CVIKhfh9LZE~~Wr_^g-QDNcfAeqf
z^>2NP^^KQ-@1JG*gT@lYS&fSG?%#4fzxY}ZTe1Za(W~0m6qrCghl9DJ9M;&i8Sfq+
z@W#8xyz}6Id38unZ6gS{0FznG|NPJYh>za6#s2OVYr{TimN6W3$uf;lg3)xrWIDrH
z7dtjWQ&o|pnCP6%jdjjW4Q07tJTBSV*~16T@yV2jkMHv3FMpeFe)CoCKRn{`(U^G&
zS``SXKnaA3#6T$msZQD28Zt<_Jgy&cFnT~EW7p!nK`x0->34_hZ(SzQDNi#zCDxai
zU_e^#AK#^$XLL9E-v{_R7gT66L{Nwj`EU`KX`?~g%<1z+YJ0aXH5#;OldCDaWz;RZ
za~E2^=OtZ05Cs+bJ>^gEfyNq~7c6Fu`8Y5?%IMhZsKq7Ze3QLo!2az$_SZJJx-sNq
zdqV~?C$!?9vMf=ZjL~d^Sy@vxhClhU*LeNShtwxClz>e4a1F6#t~E*poQq6zl}JRe
z1Yr;&5{5FBBHwFI^B0V#wlxoc4+0;=s+Z;34m<5hmn6s#Xv&7^e9oZj$dVi_Q(S1M
zZG-Ru86--k$dt&H6M-OA41@RJY*d;yi|B2(yjX%5d8N_>r8P;Kpwl$Q9SK2b&1Z!W
zcqvwa0?X{U$mNnKDI>~9g6MXlOy5Oo50@Ieb!~AcP>I4cj=B`MCRW(WOI&Ceb|q#C
z#wu#5aXyfy1({3{SenIxdOj!D8BR&O7L=|*X9??_EsQagi;}7=V-XW!MdsBa3O+)D
z@;(Nl1OP`WRYXixg7pDY*Q8m-U~@>4XJmOEBXCZR8INb2jTW4qO({!DRRx+x;#`7t
ziV#wS)c6qFzA{*3FeJjFr6tpWJe3T(Ih(zL&RRk;=uu=TS*9RZR3k86vM{ij7STqe
zEU}|GwpmbPNOHyY{u)>JH|P!}^L~aGo~mBpiylS_5-;hh4l+oLa@5)rtY4wUURqq8
zV|rUu-Hh^ZikL6rIc(dk(s`sxNsA~9Xyy}yccemKd^msLgFl~Rka!Zp%HZnz4?*ko
zhF#7l2pO#c@+4ucJK$$N{wdbi)*=`yL4^0@RYI%oim46GTC_Uns(z3|xA(dXk>lF!
zbR*X?&sXhOr4)Idlcp(k9eo4yJV%HQ$H&Libw!eBvMdDxP18YEP*-)#>vbNf6sEB>
zrd5Uu$zoo{LPA82m)6k&B{+~O7P~r$@q3vpn9Zkr{iUyQ{rXL?2H7a25NvO4wMFr{
zK;<v1u<LXRcK3Gqjo<t={`znI7SBEVJbOD&t0MfMMxbeGs%pV>GUe{wm%01So4obr
zo4otZJ4_}gSX<-0Yg=03yr}VQ7ehq4LsgYD#xS4Fn9qN*h248cr_<q+pZE-)`}{9+
z>GG5xwXl2NSmwZfzflnpyamxmq-j6Sfm!6(jpu=SY>@LQ-+b-6+`Bj8@bnB5#spa)
zLV^#PvJQOVkG{^~!H6sSyX@?2(Cc@(dU=<vjRAvR#&}XPU(|>IS*GwoU`#Xt4_?sk
z4VjEjBYD0yBx#4)!tvnYAz%CYxA@|pe1n67Q^wPh>AXhB9D+frm<$szINy*U$da6`
zt#yhd!P<t?#R+w&aczg|;?O#wllR%`?V-ifskV(TF=2tl0~Lp74;c1)Twc4%`F!?X
z_Htaw&kMT<#VyzEr2yi+<9qFSmwft_;8DodbN7Oc%IaMsxTMkO_n_(_MD@0@3G<0%
zcA8*kYh>aIEOyA6Ev{@_<JrAyT-xY!X^`^Fz#@%fsmM{fPbhLcA|{yFfHM)T9XvQ=
zx`4_`M&pL6aYzY>1{-41MVcr&c?v?~e1mC9@>GyMWu^91-|mCQ;qk~70Wn50`PRIz
z?IN}()H=&(GGf^2p}ax$hCqA4L>ousJXKjnC2Fgr^FCn8rZu*VM!HKLPnu_>X-byo
z=rl^-rIJwsFPCoBqHS9)4Pv7Q+LFE*MC8OINhEFumk^Q!=S*w5X$dWI=WGm_4g^eN
zDYFI5LSgHMrU|5prc-E=c};C<P=+Lp{G{PvfbfPzHRo(JX1f#peWccemb>JYBF%HM
zJja?obzMa(66Z)3**TizEWtUD@gbCo*4ba6(v&PKP&&gpN!h@|2NRBtM;slWayBX{
zOHJJ<0;-+NTW+KZ2+<&gA@dE17AO(WGLR;LJQei21^r&Zu-9Rb31qIIh+K}~EzX0m
zp32lTCuck(61w>sZaP95hpQKOUy&6(dYkLWexD}GNe9EYOEQk6pFm4~Rjp->vxd;P
zm^a(rgT^@~ra|Q$8W)&MXZWh3u?`nnd#V6gTco5(f-{y7EQyks5D=({KEKBt=xLnf
z!UZlMB4*g8NlKpQ?CkFHGoSqox_uhsx&Ge2Wx2*$YY45v+q)00U=gq~B93CAbJQ(K
zlK3oTvGvmFbXM)tG)-6cNS^0-@2RS46>*iNDFTnwii3j#k|c@Vd~PXpS*oz(xk?jF
zUE9_96#}gFG>t`;o_?=w@s=fD``TCeo4@@#tb`;=u(!9j5(O=P|6j(?TJpk&pW{FK
z2mfpS;s5p@lIt!~{B#kzi+_6vo};7ty!`Ul`NF^X*WA1J4)ghx#x%IbVC@1zyJ=pg
z^g!F%i8-AJh4(O<Euz&>nlo;NWIwH;Wx__W&42k{{-5aTL*S*K-kUGI?PkCj*dn%K
z3*v+_hw997_uz#8&%^(bq2EG>oMPP(<|CZ%BItnc@ZDDqc<t2(asJebVZY$T4?oA9
z=Wg+#58YvHt<zdlfpvA{<*I}%FIZpS1gV%;mc~lD-EER|gRg(%HNO7!Z}TUA@+A%)
zogsA&B&12e+8Sdjq>my+V;V%&kf@Y2>vH|-J|Q(M$_aBjXW<tR{8Ep0e*RTTHy^UY
z9{Q>HbueqDSXCmm!1I{X`6EQKLG5P5N;z%@pH$0+wy?_)LF1S0*!Lh)e%U1mJmKZG
z6iz@(=Wb&Sxi~Mk?<3p!ep%E-E3}wS_2iIv-{tJgae4^Nc*th*3GCw?`qe&{J6HMq
zi=Utm^mDL_1^?{Z?{IiL#ydz<!rEHFM?d;JceW&(9Uy~Cf8zyy^|LQ<G6{@EmT!IQ
zRUSS#=J<5X$#llisA>ZiJd%c=|M^d|zR|~;k_Qj&qLpTS?dc56pQ^BPA)aU(1RfdV
z#0XgL+N{5LB3BLvA}&~y*#vJi1fs&OC?e`6Bu!bx3(^MAJMVCfr>;t>MH%xb)0Cpy
zA<c7=EJY_8)h<*^fdoPjS}vUs=Y>GLm@P&lTGUQNzFN>ak}!oMAQ3KJ{^jHsRhVtW
z2_Up4?{vts6jN3#<_%3PD7>OqC`4&VWI(C7Ysm7PerHHF&lu0=OeYii>uJn9O>|W8
z2v#jUp(RDy%HfS;qn{#!z+-Ts!G+kGjbarM;dd>lZ0z4x^99qX<=%r+j*g~`#ubeT
zSfe3yV;Httx^zaf<3u@xa46x(g{8<8S(=b#33;yRbqjjEf;>;j6HTsEEJ}5Pu#U2B
zm@KBKX<$|<77rdm?djx0=5CA%j&3R`I+|fWBP&w8*7z)^x3Nj%W8o2&b~H*LQ$<>2
zap89o;o}1sLd*$ld>|igFfq~Ju&HY*V_E5=rHVgmEY37IV^EREg(q6|M8t0C9k;x5
zm%rf$k0<t>En25qu3V?n;e`*qz_pt<+N{H6)D|nG>Ak}T&tx*CF^2P4t`BAiL~C&p
zU!6{eG)-42D@)(SUav=%Wh?JGr4%8=LTmZ`-g}xRa#DTE$+9+j+G(v>lqGqVp|m<r
zC@=;g0@lU?*IAF&l0aNsOcRMu{jS7XhX{fP4<0a?j#*#d#5&s|58F`<?#*BLH~%Dt
zl#&m9_yzvc|MWlM^Pl?#KK}7fQ0Qm{^3xv9)y(IoeBlrOH7~vNH6A{=$I;OvX44Up
z$d_~8;%tNS^{Ry)>%nE3h$09@SuQZvQ`LrpgJa4vFdA3%`kQ~zVf;kKX4XgWbgjS-
zbd<mYVl6Txg7+K5B5wv~<6|arL?H^c20IM9o7gg?DO1XY!1)1XImb2?&Mes5OZeiK
zzssM$^fKMf?{WL)Wp3ZT$)!ts?Cx%p=NbLMW{i=OIo5%v$Ju1g%dft}KmF(5;M?DQ
zgU5%bOlLJhbdV}Ri9jv_lWIh_(BvIW*2_q?Hklty$epCuO?lx%&v3v4?$`G)%FsB6
zbZras!pO3mwZSF>_K;6m*dZW65rX3n%HQMllkZ~H4QBR?xYi`l&b)Cf+p7ZTkc13j
zBP9jV`oWs>loMkjk8b&+ttKM=Z)7gvT=ZsjKF&+$jaP}sBakR08byL?<3FD~3C0Ev
zj}1qU2GCp~Q&*w9&Eczt*<!)mJ>>YSXRK!eA?7GmlNFLde+WR*?Xt7ALnZ}Zd+n6*
z>4bV-vc0jz{_c?d^^^}^PWbufp2yk1qVXI|1BY|N(djXZat1hl_Ge!p(UP3NFnoqL
zCnsFn_{Z<l3;m$+RE3=-AVX`v5x@x%Idy>*Vz->;q!8#tQ*;WdW{PbVG>sw63UnHg
zg3w0FNTs5hwQZ$2Ymtb!(4+}@r$A-ts`cty#7Qn;TPasXUM!YcvtlWiUCq^9u;k7~
z$NZ?uvn~?roi%acYMLlWa7%9{vr^y&AI*)uiO8OmnqDV`kZ|nh<b|dvGK7qAG-<ZR
zMxNt^VltVK6<xY%y0SA_T17zH*$zH(uT*TE5n`h|D36eefF=Zuw~A`vnKvcJqbWzn
zXPll+IX#_IRsrV|2vOHBr9$J7!Xu<5Q68lniE?CVOvgyHrl00?vXrcSrA0=bWf2jT
zF@{PDMdLlDHZ+q3XVViV^^|1cDNDut@gcr3WW9o^bm$DOT;0IUAWIb;E$O8x#tE{`
zE#}UX=o}veGJsBWEWp}bS`svZXwDo0EQ~;=2}N(fZ2Ew*j6Qq9c^pe&LQ2`%-qcNv
zR?*!&DRLAJoIwahmLxQ8(egN!WR1UggbNds6w$9-OG%!l^!t52`tgskwzl5hm*+qC
zz5=iPHO@KC&Q2pDWj?sVMYL+6h)yS~R%@rzSt(<+)+?nf!~&+*>$L{THJwg}s;Vf<
zvh|%y+Dy;|LIf6z1-)L6`C>s)6j<k0Cd^vvwy=vuTb4z|tZ)0*)`k|L4NS%h`b7^y
z;N;|(hYugHv9U>0*JOFlaJaULE&FMY;c&n+w{P*E{crw0pa0x1apU?;I-TxM^%z3%
zEXoOQzx@`E9^T{ry}Nwp+b{9x(S0V95#?frvo%5lytlFY;OvT9L<o2UKDdb7`G9fI
z)VAFOdB$Uhb)a=jrKy_<Y1;d#A(HRqMg64>Ph3RDEOh}<RFTAI-)cjd1WwW;(!9@L
zbDfS$NzNpr6H8OZE{`#e*`gvbu#le7*fF0^IXWG&v9ZSba7dOV3<f<m*EiVT-=}Kk
zeB;~S<lcjaJbrx2(eaeBcF07su{C68caKhyp(Hq4@y2V8elKBdeZc0<7B<g#VBSGB
zhQa!PoxKgdIe(qw*&|9n!+1!)$77VHDZ_prxw0qu;YX@djA)qH314~VD?FS%Vq<fQ
zMlLtX{=)0x##I0*$|yn;FxDXxAB7V>wx^?jp!KW^t01cQ^5+<!^I^5FKY1y`1zTi+
zMw5^M=6#5rcd(d&t@?DtKHBYZ@{VP81kGd-J?D()kt!LkDK@qW)_1nq-5VgKWH{(@
z{YIb1kIs1X=$P4Lfeex>S37v2I2#-8-J4L(7L@aOv{K0w!%OR|^&w9+jd5&r6O3!{
zwq{}$oE+cfT6aP56c^^{5Icajm@~oCxIp7QP4G|P9#TeLQC`Ht%o}jQGp`n$jn61L
z9kg?J>!|7oM<-bdT1C!WlA!Y_(2z=^vJ|0ZEGR{|z-wwROsh2XA*^DX1Z~U02c%Nq
z1J*iJYj<Ldp(?BBn7$|}=aJ(ULYsdZTvYLDiIy6zH8NEZ^&<M?%|=tog<V;VOvesy
zJ?k55bn}$DTu>PsBV^Xssf@)`6)rD8spx(!Tkjefi>DyKI){=HLxoCWu_}^;#z9$n
zCZn41Y02T^5ohBW(?v;HwQz1wapClVB!v(xsjBIynk-Q$8Auf5nWo4y@=TK@iZo3q
z()c;4bmSl=8mSav$;0t6#;bA-lhK^T$%I)oWloKrd!)|@Wl3Gl>2LNp$^$wmdYc1A
z)d_hgBbAEOMNvz4V}p~bAxSe#h)-6~iXdXHuLy!X>0m_?z0m?pn!s>t9hc@*m7%U0
zCX1S`6f^`vtAv&!wni6|aZDXSGMdh)ZJbOtH+J~sr#{BN`NF@VX`(7EO%m$nhYjQt
zt4%{96?vAEq$yilTYU6mA4BO28HE>GXBSP3eTZU$;G>wpIr~2D%a1bx0j*-PM4F~3
z9eH$0sc1;t7Iw>6v1Q?Q(YmCns@3LWsrpTlgkHbLY&xOa?NU`0{lS2;EID|5(59P2
zGJcFIb0{ULsyd?bOX+@S-Qz1qA~iM*BvR5e4Tpz^k&+PN1mElRpD664lutyM{iMd$
z)+V=~eTHB9<<Ilmzx~@>+Pg|o{Pc>trm3i^lIe89!Q*%M<~LsA?YG|K?%g*zK752R
z71lOb8<Sf?h#OAlOzb=e+47<T)(5QhSR1g$(=?8{w$U!awfZn%Q8k>Lobl2(zD&Qj
zjn--8Xv!yP9Lvq~65(CmOIqvp-x(`H`wuZfSn3tS6MWPxPbs>Ra{d%ynLqA`bg{^{
zlkLZhmH(YhEfYVXlbq1+?ouQJib4YhTY8$(;$1`CNCFBeJU&<&Q?jTVibscZ@|-+N
z>2*46ZESG$$|+u3zVpg!92`DoGMRC5GG%-B66=E@{a&A&H*eDEWN0ODreZRFK(8ND
zKrUaoN*xkL_s)>IVq<HKwPBaZ=@Do1$JESe0;HZN#Doy36@$UjZ}I6$JFOBbJR?8m
zwUakEwj<`$H1Yz>_C3200#GO`@eLSj(Rtfe4a;Pub0gx51>G{Q{@lyXuL{8@fB%#3
zr%));wh#AYh3g%ZlK@R0DTai3hw;N2J64!cO;ZM(@YrO=*4_^5+v{xaZ*ukOCcz7e
zBBRsSOy4pr8pon>*ruTgp2`Jm<9TpUF`FzHpPk`D&H7Hp#^sz|N04a6bm|e#;cdgB
znzNXlFdiLZF4YuI0c1ZN$F4n5OHkE?u;Qlr_8iKVKM)9Hd4>xX=M73lhv>;{%3wYt
zP1a*KOtfvu5R0WWO`}~&Yvd@T0M$-TEzw-IIj&miHlcu(nXh-QC1J`~#08IQA`0TH
zWnPw)WyxqXVl)~>Q3lY<dJMV)y4?<GnnY7&z+2P$<T+f3MwP)k5+xZ72lV?A(*&A4
zFq>=6CKF84FxcB>LZGZ0icSY9HT_<n*=$ChW+a_%ysg@HtWW`jip0;QJ%;i)A+Sm_
zUw9rp9`oqo83zwenT<+Z9SA`qltzdQfkg_BRvw`(Ssq9dMK2HZR868<b!48CXRYl;
zo}*-pkJ3qs5OSr;Op+9#Bz0{tKAJSwcE+@vF+VGD(*?%Y$d07bOX%q#^{|JTo>FY=
zkoUWkO-UH8u{i0|?+vk8mj;c>y2wtC`Ro`WHK~x;#*if`uAOzf4^eQVGeS#7Es5uS
zfnYkBv6z-L&XdWcos%TiMTz))Hlr>}Ql$`hvMi(WmP8c%NB@hz%dh{&ukg-0Z}R$^
zFXNrZnwlTB4f@_zzY35l#f=*`xPJ2n7w3$NNZr!JIGXrIbLyD7;XZKQm>|aVKpQWl
zlT|TP6a_+vRdKg$!!CJuz$$7lM$aWre0_a=gQl)HJv&7w3Den($#fchsDxO(j}7r-
zj2^qlHH_%_lCW$o_|{XhsunzWaDR2~j4^C)ZBrBlW%<)HDV9?3TfhCA{LR1hTl~)7
z{+py~@MHP9{jg(&;D^A;$pc<}_1k>qD}TmUzw%`!(+PE5VwxIf8z3s^t!-%PI;z0i
zpZ9<z!FmzftyM``g*IQ4@2e52O3!PrKj1(ANB@Am?L7wl9^Fotywh3Dy-A`;la!+9
z(ChaY4En5ZY_PYx$L8iH>+4(W?OsKYJyF>GMdz>dmQ)1N3*WYXZI4?obotJozx-pI
z)gN&L;E1T@NPtUEe(@l@OFGQR*Wvav&v9w5!?(>GQ&yDo2?8)iAeAJ|5{h0Q>pk<z
zVN8t_9p1QqpYK#}p!HYz^v{2UMO9Js3Wgg)W{X$2d25#|moBrlvBA#H4$Jth5FBs3
zF`(b?vAMa$`q~zV0bes;Bg<2+-oC^jQ=ClibG&#!CFhg@dLG^@vh3tJ+v^)Y!pQ-F
zW<kTDIOQ-sWtJNjsil%j8~FDYi%;GIx=9FS?5_Ac!Mb)65+H;NX;&^5U*|L_z{ggs
zchTTKcz3}iPqZ=#6{%Ly*$`N21Hu!BVol)ev1LQA(M%Nek-?mRSb#4b63OoV7Q1_w
z_{CrSB)9KeCWr+uTpw_JG-ot1+`V5h+}=fGYq*Dx`RZ5x91_WM&)sBuYn|O^ce$Qk
zW_&h61jD8M4*fwu2+4e2@lXa%r*o7FSmT&YXKbvk$JC#v7*C1V#f8Zt+M+JDVS_lg
zm6dJ#DG+0l8q?scAxjdHETuLL<H>{~FVI4+T=tVRxqw!Ql?VdC%FjkfNvbqTw=Vq`
zlrVA$<3TjLHHNaOsjHgVY{q0VrmmMchK3|dSzljgb90kU-bEVFLNTAuI6XOLHknY@
z70#O0OAdU9>Crlg6Pu6%=K|h>_maF2tgD>GVurPr!P*+L>4e%?1fEV&u$WG<O+%v7
znEv3S7q-v}oQqkvN@A?TB3Lwz!;>+uzWIpz4^A<aBiKMH1eq4dV9|=0rH3KOAV~w=
zenwGfw1!MM3eiDJL6T+^MUOly+AO`8qbIedv617Z6O9rwhQyc#<D!72X$*^c%49ai
zE({&1NxMDVP%{)MYkJ7MSKy};LU)6;jUA?q;8Y(Yu07B8^LG%9;pF&?TeqHNp%a``
ztLr61MX%qB@VbjCK8xVAj!G^cVCDojc6OQiKsl>vYLD@ecZZ53T<f9&b2^#fZERym
zrC8h9rhIrxl6Lsb-}rU*_V=h46?o|NIxH6D4_459pCO1Iik+gvbI(1;U~PzV9;xM%
z(^-oaFNd}9zKAB9_Wf23KgJNSLR2nhC@zt?W$MaO46#I9mLKQy`KtZe>-9K0JHxbD
z&01^9a<TG^oXut+1jFHw(Rjq7Exx3Txy2#)3j&gJeOg$~&;g8z#>IK62~u)!@Gz#A
z=#08H<gEwas%ZIujhx>*(lq72{=fcL{PHjT7T2%eAXPs--d9!6IX!#KzxZdryK1+d
z93L{9PFIMWF_GKoU9>1!QcHZh0dv+<H8879^rUQ}cp`j1@Z>$iG>+H4^N2UZWAcum
z=qkFM?&|W1cGI2c3tVSdp>}DS(e3mZZf!GM8?v#!&h?u&xpL(S>+2h=t#5MW%2l#<
z!E(;CsA^76PuX1`u{AtDmVU3t=Gxi^dy=1a1Y%3}5fRDj6aDW>xL?1;^yG|kI^iQ%
zewGh?_%eHI>%9HueS(0Z)1^Ncu)VXxjhi<)K0GEk$FSE!3C-8PR&nrXMm?LNn!s~c
zZ{cxFCKK){&GnsaF7IrTCJEcaPTS1%7#sN1$6p{#A{gTB3{ub9WzF_Zmm445XCe-%
z)P#yDb2CN?;(GO-d9Fx1Z1sLL@4}P>gW-%b8XB4qsI8+iH5Vj{JgH4RADy&71&OH|
zd~Mt0CBHiF!Lu?bhI7KqvL~{fMq<lqnJw|;ZLW|l%T7|$Q2PcG>=PA`cY<=H(QBKW
zJoZfPO$f6N1C^pO!ypk{yENq2|HjXOHavQ80HS7pqf45183xDkNkz8<b>hhrODC;)
z`MZBaHy_bSZm_?*$#eVb=u5+x!_ivtoSG8d%h(^@1}Uk0%~aNunL?|aM9ZgIo1VU~
zBf|PFL<HsBMkZdYICqN`m(4kgv2k+AIuUt^W~X!d-7ZKGlLw?g2^|p)tq83-?NSV}
z+@)w8IRMT?S9;mD?5#C8WAWB8pUs$z$DEy>QI#cmp0lyF#s2OdNt(t5M{9JVvDQ-0
zD;6hn9zA@>(cuy0Vu8S;l|r?|c&P+3lHPFEv6#<Wf25eX8oWTugibfZIm_|!G21&k
zErR4w5^PmPC+X>wM5d@D!8#k|=TcyaIj$~vCX158iRGP#XWV}@X0m`ZAGEP-4N6s{
ziKD9psk9`iq|-_1bTr*=LXt|XT_B_<5eY@IL7wI4G$lz>oDVJ8G)~f5shE&JEYurk
zWBgE6wYiU;rm30N6B<)dq#0|;fRIB~K#C$&5~pD?nc|C*jp06{6U#?__3v=|XMcwN
zS|2|h^Xeb`0jAf*h?b{W8%$ltb4h43U88DBD9t>C7=H#dQc-Mfaer3hJt(0f4aPap
zF;pB8BdSbY#VN&kf{={HQ)*+$JDU6NKIS{$ewnk=Gy2^Q8`~Q^cyNGqKO{qSNfimP
ztr483-|O??4}X|MYlMD3k>w&1BZOc)8qw=l5F%ptL5U`o+|m%D@MxJHA&5%YWeEMU
z1$*)Md>i<_MCy9I9z{{KBxb{SJbpr4VOD|<n9XL)=W_zFP&CGnck{?sUB>;ib_6S>
zTYC<z@zlaZbX5r+KYmPEmUMb4&O4M;bi1(y`qQvFQA+V^zw+1E-@ihhJ)I)$M;Y<l
zSA6F?-{S7wH+k#r*Z9^qzRu&v517xURMi60m=$*}ZVDtm*i{=UUbjHg1nN2(aT^z+
zf$NX3VR1IPAh<cyjiss_I#OVfa@mH}7mD^cUrVJ>QX-Tld;9>YG+HI}h7n!rb$bj4
zL#|!B&erA@{lSoKw?~#2n5H3r{_Kgu?&R=<2k*|m&r$xAM?>_uTb|DpeB+ok6YhpL
z$bE-=zJ*d<Zr<GG<m{B9lQA3?Y^?Xt!J@pu*oJx$+P3+OG<6h34x(XeV;5^1re{YK
zi2_^ER8yQ0I6ENAQk0hDx#IRS`*BTF5)zGC>$3Lj=P7!yb-BY=?*B25XYaDG6FinQ
zfl$5wekgRoX7QsGc5$ArUBlQ+@ZMq!G{$4B<-82Lpr~B^e;kQUqo*Gtf+%9eu3ByA
zmrT9)7Hq4cE`$|#G^}<n=g9RkHP!`38ET9_SLiQ?6e+#@60RCxOO2}piO)y~>}>Vf
zU+;4D@__4CI?QX!WG1Le!&)jhIh!$^HYkx`>XJk{b~d~G?5AJA)VsX+%yq8pZqe=J
z6sq+wz3{g*<$`HdG6@o?6;eT6&uHqB%GcOU8{;*9$njK#UGOrhVQJAi!iBJo^OGcr
z`LW)%uxJQWWksq}GL_PpibXZ2*YBaTh+J9gz=lZF^^sWaMF8nnNfO}kwn18tbYO*~
zb{=aQ7V`z;w!K={6<L<Cv%SsLTLbbUk0OBfWI7v#5sR{9I-YQHbi&btV@79VvMgnD
zYZI;GWR)cftyEM#wn$_U&^U`q6v&7M<yjG<%TQ>YuFMXlmXy^3EudV?>E;>J#hlVP
z5((B;F;5dHjb&O|j?ZQsoJ@E)k&G9CID1HJ*-&UtCv)@*7<47;y&NS1X_l}y?2#sd
zB$X@{6Pl)?TNHFVLlU({nnuLK_~_r3<T+`}Ndst1N!?U9V=+xcTCKI1rf!S7Xv31{
z1?%~cuF5HWjS?Oq6f#Tb6$N6xz|9tzevhpy&v5%QKS#DP!1{)4ug|l;_8YwQ-S1Kv
z$Jyu%$_2q$WU}h#w<u3sC~S+CIH1Z>WRX)J9L0Q9qVwD$Kpr3ofbbM~%9$-OA&>}x
zvYzp1jPr`=Y{dWNzy7ZXIHrpk2#XB?_ug&n_Zwn4%`6`bfs_gB>l-}t?6Xna6QVfg
z;zo6uP7w<e&*Q_#C|wc=kt}<mJ^7=I<<CYEdP_}i3%U5U=;kg}MC^1DH&U@p8h^k#
zOQ_=kEE6fpvZSu-6}o3_jM}r_k>@#!#e(T<MhJmkuSa7{wCgE~w&-Ymx<bp<3{Mtj
z-UV20+N8vJ$I;OtWx1f!>Bpj=(_w8ed}3+%NtSj}3U1%NLv*kGNsj0rH<^q$IeEaB
zzVwH@`pS2C>#a99JwBwW=2+Xte)(uf?pn2A2(dtRF02F)O#^lN`k;ol(71s2h9sfo
zP>UA*_2(Px3uqohe^W6T)6|y6IF=TAsn!%l!S2o;J3BjUY-}=ITW2^NlI00I>yLQm
z#`*i7os4+=@GnO=^S%*?%Gsq!Zc9ruq2c88h_}S649$c!z0KBOkAY8EU+>fHc1e>M
zJ{x>V@C`0F9G0RdNQypdYXi3T`|Mrb#5qG_rv&*7`&YIY4*N(2{k4>=NYL%#KU8U?
zc*vNzWBZo<%RRhWFb<D+^R++aV0MqvPYEiJNMc0WdkZ_&VN-4WNGHVOsa(m_ObEfz
zn1C_9O$>_oS=-P1p1&d0Y3v>e@F9i;2q}<WJW<TOx43)a6_)ecg&g}QE?WG0Lqlmx
z!uPONQ8J;QU!pn5@nwp%iX<fD%5Z6Kja!%3*xT*0wV_DnI`V@TbL!cQ@#z`UxuZYa
zB6!O%7wq)9+}gj1VvnnvYpiwNpTE1D%NDaa_Z~gq-u(x32R*J{xkRT(NRo`IGF&W#
zf0*%<b|*0n!r-Y}BM>hFLIjjp_S-=dNJ1d-kRnk!p=wIziv_YH5zcZpIi)ukV68_B
zO_COIy|hv7>WPtCUdLDsZ(`qFG!6hcLB_scI$Z5In>WlG%i+-x<MD(v*X-==a`nbD
ztgR1`TH%EwNRO4VyY8DPV5sK{CMRc%&qf>_9#hwXo&7zo-@3-y#*ngFP?mG>v0WA;
z##-A4ASi)K5@ezft-!!}Pq2WKZ9AmKn?SGz7d%o242H_p_@*L7Alw4mET}5OWbS!*
z+VJ4GW>hL>bO<PtVCZN|R~d#yU}Kmt=%sY~32BmHZ6MDx25W0HO-XGUj2EnJ?2={)
z0*&<soLs3RW1CJ90w}GROvX${C#|9`@^NTe?;%9AM@d019nk6a$d^_bR*|F0@*bi$
zAk_)x<b=%|clhqz*ZJr#{Wkgf8uPlMYG$kty3|4P?$H^lsyKRdpIlhHZAgn0W1{o7
zvzA~QI!ZBX;;zj}NH1+s+6Ff{jXrwT-~=orWC*%Gum}~CQOzv)7}eqgU6n>(LJZD~
zdF%eTy{qGPa5|g+@XwBbrAbZXZb+2W?C)J-duKaJ@nvi^3YNv)d4;z;r}yr?%l39G
z>W#PO`L*HV8(n|~f5;(a+?^^VV%mceXstkal0+f3B*;K*YVuA_kbxir#zj?lx9E~(
z*{Y4Z<kijE;;m_#R=~7u;YZ7qBuQe6G(hb<S(-5~7a&w?j}p~uOaH(nf+q#>xRj!C
z2I&Hm*_6p-!uHl4&NR%&BZ@R7)k!oG7j4|!Pq-!flN#sft>fOkxA@{0{}sRgul@x`
zM~|q=dF0T!hNg<9<<>=yK;NPg&P7x%gjmE?6*NECSNq32V6b(3tv<12Shww@l?BCe
z^_7t3ExFybEwv^li_Xrb96g)@uRrlN>$@2@u2_ElXU{)ZmNl~vK56eqAD&p;mBhb&
zjVP=Tfk(&pIGLZ)2|d<|4L<R)&+wt=uOrdC_2yl+HanzKi|)YYc8{!=vUXL|=@&e6
z`zBW|U8dI`kR~ZVr$0rSE4sZdX_BI(z*AAXImXpAE*2@%aiqr_O^>*H|4km1_nFxd
zr5#fq-N$ZLEV>mUjSIeQz`f^Q>Cq?U>W^?T92E<uX2N(e!izxNM88olVyve83+ApO
zD@e5lq6<aRdT7QCC@iPY<ydxIeH#tA5>T{UyeBF?5z#sbI4m<?G9Hi7I>DwUUbl8=
zkuglJ^Ui@HOuG~|Cl7(GzT^34F7e#0O|sO`>p>8by3!mO%^&{hcL=Vb+v&2kxlNiR
zJhPRu)mt`;*S_!HzWlq6PsaT5AAgzu?!WtQxOw{)|M@@s2mI{MeukU3t}>aNkr!RU
zQ)q{Ngye~~Dqb305?NYm*@_jh_q}v})=HsNiVp_sJ!Ms)grJ_y=uJlSI$f;w<VAwF
zj=G!?d`d_Hp$RO50KD~-jbXB=I2?~S9gV4`CBYen>+8Jm!n5pM+9k;{q>e4U5MmA`
z9(>bKHI}9<spchTN5_m$&zR3<6nTdmH?MK&+7*gk2NZPIdl=Ku)D^<Vwrn(Bj;+!F
zF~+Oik2pLcc#v(Xg;Hn*PD;E2+_{}fT~-7?LUlS+jbUB|PN$B;QO(1XhFO#1WPy~R
zRYNKa-Ndri73^;h*xu;T>t=D`4uT@@(CHM^O~q(*%4{}9X&E7Vl~Cke2nkZ^7!%dR
zJW#D87e+~isT*d~xZ{fL?)DmckJbrFX|zsAvM9XJQbe=gph&ZVqT44c3WAW7(=mf;
z!Sw8uv)PypA;>!&I=u|xElz2?)SNszU~+njEELAnh~T5gU%MHQQjpnbTC{waH@0^f
zk0zM9#@02ez2BS*I2^RPC(#mPLpV2OegEUsj;H0rUqIockgcLwa_gCA$g(`r5Q2Z7
z>T1aw4<T}*?!I%EkAK{^W~Dy}se5{CDz?j78gZ*;)jC;u%q@|+WpPzi)w!4=(Jbmk
zY~>~i*4k)s(#A|Jxo&6UGn{kudOeaPp)AW4uTNf}O_=umE^-W?Xr&6Ms|N2ZX{wMi
z@bJ+i4jw<|*3D;;QgZ+PeNK;0$g_<3Vh-=&EB@OyoU?rMn}5m|{_qcY{q<LQaQ_`9
z;}fcSfo&>GUDMb)s#dMXx_FQH(1xB{SS(_@b=8jg%Nk2M)O!JM56x2+WTV5H>Zcoe
z@#BpH;slft!Cec`P)^U7y?e^;wQY9iTkLG@^W4X8^J~BKYb3cQ%~Lx44vm=OLW30z
zR@MXsRZ}r4PnnF*c=+G}v&D?XV#Z`X;^g!(M`wqOre~auPMOSR7}Kgqg~w-(u;D2-
z0>iDq-awPDrxYEDRf6#(#7)3^5Z#0>3A<1E$H(G+LBq_I%$xa25a~j|IEQWXe8dWG
zdp~N|FZyIj7P-d3qJwU&q9n-gjna8>Zbl#4Ly4%J$2-45?m&=`-~r<eWwnSlXVzlG
zlh#tfE6lt{eWDQ4jNEh>q>^9#r5E|+N7lKtn_!!gFMsJR-n~EJ@My;9bj&l?6_SQ-
z4tH+#Sx4|am^j~eOo7U0NWISH)-8ISJ@RamBpIS~kFE7CsVJy_qJ>>xWlkDeXk7^r
z+fKA?SkZb0%a*{cG?~buwpiy8Ns1DJ@%W5xk+($2gy0Reu~^fZ_j*ZH1?E%BbkT4&
znQ=0nFs<eU5xBO$&a=1faOKKnvMeQtfEQ8dptPc8rQ^zq>14umI%P4NGoMab%ok{-
z*xlXb>a}Zh`UBEDBM1jlfs!bhlcx!|up)E%3y6bXo`4X*iB@S8TeDh9@Cv6SO&tlH
zXbr)mq-1t<LQ@##UU4{cJQ_L1m8Nt#R;GC8DZ-RN4&B1jFJNt;8LkV`0=xt%qcXVL
z?UAMl)*3pUE_t33LJawyPUe)02CXvkyxSIbCL(n$KL#R~!MX<TElNw0L=#g&Bw9!G
zCebNUs#thQfy#13mXV}c6oR+_GLQ;IJwE07m3^G4VoaHoIA=+OqSIUF($$;1_6Og^
zRux_)%*PX?_s9?sOBI2O){;R+b8969y>6G2lar{3wjSXHtBh~&@FH$Ad@w8irE9<T
zlNqtSDTLt0%^R`J`o8%quDY^4pb#WyXD2*-_yFe}TCKk216-h&=-|=|sx_(gp>5-Z
zwv89Lb5ctB{XR*O5PVpnZ^p!Syi$?l*6Z~+I5;59B6qKD?TVJ)yHx!ur5KOLapM!t
zjhVj}H_nHER<SsC&XXk3uzE6`Vx2<@#nmg<Kt_eEk_zJ;Pm>h#S7nstn1h1{yz<H`
zeEG|N$lGte&gtnf)9Hk&oJB63iC&t<M$VkIv8Xd9qH;@5FYA7klK-#v_#g#DKa~-H
zv-YC~dZ2p(G}OlM=4-FAwy_2&?C)ITmAl_XB+$?>Z^oR~kD0j<rJGT@8IyVug>#Le
zp4ZH0bBt{;-e7!9S(a2)N!`@crose^Pl?5)29ZKG0EQ5TAi4t4^~fY3y}+74xq!NU
z^3M=UAD5>d7K@?5U@&-L5h||3u?z7RXKxoOCzT0;*5V|zUa{dqF(+DvRt!uz$BUOX
zD72P7&SJeod57{Iq2dhDcth!`6`&DZc;Cj*Tc}0?Q7U8!(gxB(a{cO%ZYRYD(3$3~
zckc1smme{od7im(lRMAu@ZiB)G{&HnB(9w2w}%I62W8`!OdB3PIOg!^jPYd7Y+f<0
zOCB9O=Fy{v^afWLuI=&q8%IoLUnlQ$*xlLZGtX9Zp2AM@=?c5HrF7oQj%w5>B#wWI
zBrJm<B65L8w@82zn%V?x2&8F3V+__dU>tQVY3hb@QDSRPGYd>B$HBbjaABBK4W_Co
zg5b(}mz(=r>|fhteLM0pgotkIf{0?_d{n{CN~+O>vy)S1^EswzK)~ku1{)h2Y;AAT
z>2{(pM0$i$co~sNp*6xwTyWU-CUC8xju0{WNwj`D;X<J+lxTg-6hcJaR3l;@ZV-Z^
zH(+$U;L)QK7P80VSzux`PNYaJ$-GBN%X(tj87B1lnj-gfyO5;@iGh%!L`snrD6J{$
znt55`19W>s@*-_{Ri1KDMX`i;G-ZiWicY6XkrxON5lE$Ci_TUC?^6<;ME@$4k|sGI
z&^m3+eq$~ws`Z;vnq^{#6tU2nJbujX$8M4DZX#q{(3OaVq|$IUoiRQ;BO}n1C66CG
z;MTLxqkvRt2m-q-Kw9gWJkJ>n2Hb!59r{T|RYo+mBVq1Z`yC0sEv^z7kt$YER!<Af
z{RPiLhy`&IEJe4&{{BAjB**QEicOp|Mc%o4m!rc&tcj}T)nXRj`zAlySR!_UrCmTQ
zz8cezg~;_-nh<;M84L!o#To*G-XL;NS|P<USz_6ab<Q!L&oRc(?Z&|P@p#O9K3}17
z7k%%v))-^(-miol=f^2l99CmJO2+o84sdXA$ZR%eI+<~Fe8Ojb<}(DKu4_)tP8m-o
z=ea_E#fSHv>2%Ef`)}~tYv1M1zw%|i^PR6VpG7XLY4cvIs>GNW3U8W-tSr&GWnuSk
zRpImR!&nd>_5#R3cLgdFSg$$0`<S~A-hwQ49o~8HHl#4OGma(?II)LJ-6>_5QHBNM
z#XQn6qE}6<UL?yX((_6|#`hv^KWY#OBoeYs$TmQ2fXoCUYv&ycu7Rck_dXwV*}lQk
z4ogjq5e-iH6$jl0k8zeI3;%q)e-18W9+NA)$N9JkSIc>e_oH<}UM%L?k1Mac^SRCk
z&+<Rx4UKDXHhN~d@Ltcm`1AENAZmeVJknZ3D$z1fHwINfx34%l9r5tdF~Oy5ZEmx@
zv%~4xW4x_68_lU&k6vpnP2EuRdelKt8NqmJdFi{4dG+;oIXFCJI<1*4Ce+Q05E`;<
zgCyVJ?mH*kefJC-1eY(5xwSdv+EZv-o~p20UXu3m0Rkl?TF6y}Dn-1xAe!Y^ACLqR
zsZePW6B(S1Z9PGp#M<3yZ9L=Yg89^PeB^mJn)A467`u2eFKy&Jdv%u=cGlVN7Pwr}
zn40O#lV=%PE0mHnO~rIN;q>&BYCOlyqRoj;61KLs7_1E$4%a9;MI?t;6=;DJNfcN}
zK@d@)<E)QaqduZ`uC-?oLJ-{f|1W3TNC1z8h!n))kN_t7%>@s-+a+1w;N<jxvwF^H
ztq{6PB0ZXtPTH`ROV(1&)-a{$3X;r`cQisQjX*OpU7)0<GBH=z19{P<)3(M!>pZB_
zg86*L$?+M^d$u>X==8b~>LzCNsYH>cDNR+<G!30juWi*9XqB|+VB~oSTtqEH8_E)5
zTRMoS?Dfv#7Yj^XvA4I2BXDqV$nNF_N=SSVJU%{QHW{PRz~j*o@7{fz=Z=rjQntZf
z0<AOwhZh1w%zw?Yoatmn-jlf6A}tguA#o-OdQ{81a4{uC23o^i|Go&BA9-A;5j_DL
z9Mi>|FMRQj(Rs$@YgZW#hcTlyO(U8uqV=kG4j&xvyz>snM@Lv=Aa9ZI?|FDX+A6*j
zODvDmwG&NQR&=_(C#*}n_bYLP5Q5okM!6_g#hG)C*=&Xog1WAsP|_|7Kc&<vv7#)?
zRT1Z$YbRd6LJ6PDIt{IIJhmKV7Kt3rd;aW8UqT4Qcfa#0ufFygKl3x6;lKDV{tKj(
zy!OT${P+L&|DM@w{v)5GpT=1Jt=1Y|dg;&jm;dsg@#dRvaCG<>tqj&wnAU`NK3}v5
zX}~rxUqEfnTd@EC@mF*Nq6Nw!pg|1;#UQ0yba-%b%IkMu;ojr>e5?B{9zJ?VccV*4
z91C{_+wo^eiS{XqSYPQ29~CP^4$Xyr<oWkY1zJH5$hINhfP6D{uDlo6I$#=)T@X{g
zK9H#};${=Zl)6G>5;PbG#(AtaZJn^JL|ZcEdq#&2iPJb^ThWh0%J|!@J?CyN=iQ3}
zGfzbAmL}W4sypC=$KsG!tX~>;JG`lIE@1SNM8n`b_0%96i!u$tLN%}X&UYSh`C7)#
zUcvR@Hnmj*k&@;;Hnw(|Eh>ucfJLeKldrzXcrs%;o^p0N;^4tUKK-drQFQtg#eiOa
zi&x${;hQf%=J;rW6w&qChZLnEg{3MrN5>PUvjwA3$(O(UE`R-JPJWVwow(4;j;6p7
zp%JJMPo@$em#Ch1xDZut!N;^VEd@$vBub$LG{)h)WZF1N14lE@yN5I0IfH3skU%P6
zx1aN|8<+U-%{_Mdnp8RrvCvtXrvlVv$=T?X<Ktr%vl*fB^ouT+u3lj<7?5@Hm^m2&
zjdyr!2m-PsL#9bY>{cj8Ah;N;=Z!&x^XJN2hiesT0uhU?G)1Seu=By=UF`E4=P@B*
zgTw}KO4Hrnr8l!YdiNxvg=R`OP1x)P)^f{6N0F<XqR5ddwugnxkTO2;S|$`(h8F_k
z8?5tK7s!hggrcfVdy}_TFh0=h_v1xR)5yti7NvAVvxFc`bCjH8Y>XXB)4WCQ610k5
zN%7*#NJ^E_Ou2DUW$Q#-fQ0j4EJue2Ty_?uWNm$oL?`&VAxTquy&k46Ih#&6eDDrO
zCx?tDV|0>GHZf&F(q;gLNF<6ycui9`)JcQzv4GQwA{CxGIGT1UaRh>c;2lD1Y*=na
zKA;iKKLp_T^o0NBzx{9dAO6Yz$gO9d;r8v@T)%#uD_1Ua`SK;Uwm0eadUQJlRaNok
z8?STk-n*D(gjnDMsZW*#-OBIH`Bi(=IZN;{-?gf$RqR($6f3?Rz@l8RzP66FmhpIu
zF@_{b7>!0ur_<HXHcdm>BEUk3m0EUbA|7|m=g0D1H1Fc)d@GrZf{1{t|9|cMSFmN-
zncs>1mf7yO>(iU6%<@$z6wok@Zg#VqO>#I*Hp$j7LYxT=^)ezUDHOveg(9R!MU12;
zMaYCSF(Gj@!`9FQsTs1HY{T{-JPL*KS(TMp>GRrSw_8pRYoD7p698EW6q-Qe{wr=|
z*3E17*=O&)zV&^7V|nZCcc?1Em3QBxt{d)eZ}Y`3{~NY8w^&_YXEYqKSp10CV0;hw
z(DMHKSNQ$k`yIai^{=p)PpGSCKw4E5<$O-vR5VRZS+*F%;m4x-Q7&1J@enMLEOiaW
zRn26ji5oTx!C1=igz8`ljc4IodKU}MUpm9h*_{|y)(3jT?pxg3JSggxHBmQ;lcG~5
zYe+OGwREC@!a&jkU4Y0S5F@#Q3xo!$1^9;V%dhjh!?VjC6DD|R>GgZ4bb|FQWnJ>H
zLRB33cz{(#^h6IK6~-BS4PyDamr-II+1zyXp$}jK7~hR#yJA(zm^>u9iqI2dLydKj
zie#gGmlF@4gx~^pQ6pQ8G8WZptbwn8{p+~oIff@MBKy#g<3vVTJKlTmLoQr=hW<z}
zp55Y)e&<WvzI&IZZWs*uy!hgaJpaj0F`br7CMEA*zstAZy2g0!aI#Np;~T4`B2}I&
zQ{=@edj~aDDM1WrjK+t@i}ia<j$QN<TZ#t|0x2X~O7szqo$!tzx`1#JobwoAyUt$Z
zg(Xq}o~CuwMzAxjxi_x4JFd7rY1s=2iBz0P6c-B3Q(GgRTVLU1KSgAoS_xE0NV`hc
z`+zYm(}M%1vnj#_Hr6&6jaFkp*JWV_z;xl|K6sKeA;~gyl6C%hfbTM7n+DrPL#Vne
zu}w>wMsm6LKJoxV^a{&T9qmt!NdCQuHV?)|ZlU!a7X(2`v{sxxx5fUf#O*E+L9mi5
zR{D}b7D$Ug(i;%69=2;IyO7XN3$m<0=>*fI7C0a6Mv9_Go)uW*G1eXFl321lWo>N$
z0je(hGZu4QajrC3+DFP6&X!mk=oLMbio%JwP{&0#w!a?~UqLH^Ryaf~!fXtMc2YwY
zlO%e*99oam3Bfx0{hX>^Q15RuolR&P!)!jo`pC+VN)nLK=dPpP3L%)zW;h$D7gh9+
z^8poNtXJb4wX=sE)8(^)uBiKAEs}B9=$I|$sGS|0wcNRVo1%z?`Cu?$b#=sWION>9
zv-J8sUi;Ro>}+qxP0z>7EArk)zjqO@(Ku4ISgSByQKwYw&@MZ2%TAjyhAhnv3pj|Q
z&t(C%?COmtV}uaodF16RQ!18rE<Sj)(zLBPV)62Dr*7H#i&O>yB4Dh?xjH7v_&L|F
zU#F~Vn!2TGYNoRpckkZi?78!Y*Y}SioO4{iag8fi-s1YTt6Y8m3RN|yscYJ%p=}!K
zs-&tb>bjw>EoBwCPqqtb|IvJ9$M~KIKnr0X!Yn+9B3n|Jl5TD|cQCU^>DaoE)Am|y
z?r{akJa*rNjQi)Ko!X_PjFJyJbb3iJ4j>$82TFtJio#`c!nB^IieeW3v32YK1hX(>
z5f<2B8IA@F`Vt{6Ra;{psvy7n?zNytFX@qI8Q2=@AJnphE^%UcKZqk<0o|0tyRK{J
zL+6><@p=h}05({vrlKq>Qk``Zgg|f)p9Il4U)azfY>RIdUMK92E!%rDP8?LsY8cO3
z%Elp0;9xvsQ44fBVrAnj7oRSfw}K)|IC*lDPk;K83<d?GQKVFjCp(7<cjsfHN1`Hi
zO)i5N61)po7w||3njfm&$#Q-jAEwcHToOmIx~q-_0Xh#dV_SqsW2~18DCrTxqr2D-
z>jU$$WoO!QW3S@cZq3*V3>hh&H6=OSPk3^z;EBzGQEI_9h_ECVbTq{vFxFC)HBDpa
z=Y5LR4F-b&X)lLVb!xJuTYCVlNK_iTeL4l1;)ymJtua{BVCo86RbVWxY4B|fcMe*`
zE_f27n6e@kdoDOa5J(xtA5{~{u5H_PBy}4{v`V0S&E_!SsWTh+d4(~KOn8c-$7pRt
zZ;)W}6>QpLxpnhGQ%eZKQ`?qIJ2Itdo4BABy<TUtA<2v0VP_;w6;ioFk29s?0#Aqw
za!e3VBsxQ>3~Ni$BqvF-j$j{y!sE_KAY@d?M#WOVcGrXvapRgug%Fy&7(g0x`J{8q
z?%YR9LAJii>dJ_^DVZ;3^m+wC3FeDAWm95h9%H(+LORvibH`&Jk0(f(V46sxcSxk?
zFuK|Vj|`%--4Tbzls{}3=W&a994AyYX_}B^2}MzmWl0w!7RyU1qj+L6o*XIYJ|1_r
zWdY^9M+JXaJQ-uivK-%uC6*)9CC?6y<ga=t7bU%3@6aA+sf2Bs=J0c^jo0;bIz2Qh
zUKWbB>tw@&apa?8*asY!IA0NgXnGz%%D`x4#8Xc{&1<i{5q%;BwAN6TC0DOrjbe=-
z=xqHk5OaSI#s|Fh_UnA>)o-wU|1R^{q`Qv{#zb{|SuTq@Lsfg~I$D<;k6nL^5qMyQ
za&*#{5Ib^d{Jg6i&7x-Y$$~QTw7J3eK;;3cyTXo6Ak%r5FOv#7Z68|2&Fhj5g9Kq>
zVW(95I()#{fN4B!NgTx<^$q@kAt?Qv$}OTNUy)Pv1wvYCQ{!0j?7p{5G$kk3J@Tvv
zc7p4!|3epnN0KL&IkAVX2i+Jpj$~c8S&qA90a!e(H8ib>iBBTtbGt`;?3VfBh(#1j
zx>!Y4m1jP0n9dBd1<V&Ut??+QC~JogG0|ydP4UE~r)k<iua|N1#0D3ixPX?Bqyk+k
z=H(pgUFTs4h$I%RqU&vXhwxG4wlr~<QXNh>J}MrsutVU3h@;BR%s~s#M^v#wbb4nZ
zzeY*fh117Tu0%wmU*~9Q$NoXd-Gh?b`<983w5lMuz}et9H_EtpqQ~jYoZ(uE5r#!s
zGE94<UPtbgTE4k)oHHB@DT;z3&(KOo&XtUdo$Zt`N=uZC{34wsD4EcjSisaxNwt_`
z+X~k-*t*2GZKN{D6r95ck4h7IMNV4eAVnPWNr4oqvxg`tyADWeY$x(`ow`(!h-kKa
zYE;mkSmoaSl-|k^HCkt`1vmF5e6ar^v({p)CrL7TX^&BnvN_CHA80mLdu$e(G%3iE
zNN$$`bWh`hL@JGv8n6UkbzMz^5R#%O@Gf*j?P%DnbV8b>NF>9-5GfV4X>cJH;#z8i
zN)S3h$bd0T>^L}wNHa)sPzh3LCXHdEHvkdKtJ!qHY&@o(j@b|wIk~kN(+QkqZG8ol
zpy^Clg%+_eQ7Q(HcMpXCOsDgx*tIrt@I;IAOF4bSa$$^?H#|hbh3`^2eJmr;C?07{
z0}Du7jj9!s@#Ih)z8s4O0`plFA5P!#yTy0=<VQj*+?U<I_%n=YsVade3XEw9BKj|;
zX-Zw!hcRMO$}T@v9*$R?b1{9SX=s{;vMkYB)3z;b+dMEm_MS9N@!nI_6+U><BpRnG
z*?HJ?McmR3W%+$aj{dO74I+Ns=Emk>#`4O_h~Pc_!GOth!uHNKyZig!*LnJ3;6d@~
zxqJ6EU;5G?@%2}~#(XxVtt+gF`K~dAr=e*orn5P1YiXK5RYCnRnhGD|M+^p-0_#8y
zjWulFnsV}`3wU8kG$g%%%1tc(RIDqk0bA28@E}uS*G_;AAeY&xKuBVdlO&)!Aq64e
zTZbtF^$b|>F>>r8s%9F^ltqiu33*=-jAd>rEG;pvL-5c%`3Mv_xf;<=S5f6%nr;j&
z6A;THw6i^PA*y42mr^FW{8;OviFRva1ZLvX2hS5?p+s1ku?svc);&Cm6+)8g1gXG^
zxMoYAg3Z7xTqBw6K~pw(E0H3@h90AlqSqIMkddXQId}FnWm%$xCs9$=zJ0gh+Vx$&
z{ra0!O@r;ICuypvDu)psg~7B9v<(u65*`;UMgAZ^{-fft3Of%pA;AQN6rPkoUs`%H
zpu3i<wSnL>giJc484;~91PGx}sm5zZUBUi5usxI9U8L+7M=d1Q)vQX%6Gg(=6~Vbx
z&&JS^B_1hqgh*-H8YLQpi2Cjz1VTf9Wq^``B#oal#6=<&eJ-kZ(*&&^WMalxqy}3V
z>awCP7u4mPwo5|@&f>fyNfXj6!Anh6<P?KGDoY4TA+$ni9b2rep`4eD#|K#JND@t^
zJ02SfAtYWxBU7}o3=+@Dje@bS*h?heyt~c)X+>{zk|&;jnJ>KjGN;d9U}b#+DWO@+
zIk<O=+aJ8kwc8((-aX*b>MEDF3Y2MCDFjlsSfw#RU_ycx8tEmjZE(hclu;-V40S#4
z#;lGa6KE|+1N74zk)((`MaY2mkh_4D8mA=V=^neg_o$niyf>gw3Lyhk)!_O$#gi9V
zd+J4`$w-4j>74xlWpG?r9gx|CRHc-qr5q}P9<Zolmq+@5RtX|RKe{Z5E{V>%LnUmH
zr?hn&X&5rZU7;5QFJtta0_i<21R7^Q?q}I|-Wwgk!-YWE)|4$VCP#GKR6=JF)N$XA
zaP%Ji+YfXcepz5}*g5lB;e(^DOL8e_O^tI!=Upce9ib&nLoCp&M=EGdEb6AyDRo^R
z7A4l%82#fMS)xd_#9F#dOxwNIZp9Ju$jH?n=F~2`$%pYjM_#@Nu(7_0kc#{F?{n_l
zS@JaHr$6^uE?vHi3!cez!hiD*{sFJQ`6em`MK2q>|6bs|=i0Tm`NlWC!p)o4+1|cS
zvncWH0%uB^dO<m#F<Zou&}_D#svXmLz_i~>ICG32OFU{6y+KsW2H<}8XAT&+f33m~
z6>?ZWqv)L$q*;a6EseHh8IVOQ4(CDiqxw|>+JYz_E_RX((q42e&<1ENwt{8>&77#<
z{a8D~j)pNVnW0q6qFoRY#oV-PHwV<r@zgvp`}=NOv4noQ&Svj4YW8+?rAXsAS=(5^
z9*WBX7!}ubDL}dC<=9%#CeYZH##E$Eb-c73?^=Sei0(PzBs3vVe}^j8(#$6e3r($C
zOz`+d;DkoCDLtRCrUXCpQ_u6|fAax%?>21jTHbi$KF>b0&e}>3=^@XdoLE}tF$TWz
z>Q!EQ?Om>4zs<e-`?#PFIsA=lNs0utA*dRy4Q{cAtM>?@A}RXx2RTXdcn^rjF6_E}
zcQhas4lM!_7rn<Ia%E72An-m`=~`>NS<>1C&I{(1VQ1R1S9$hpN9hGY1QG(HR58c|
zxptUlhL#CR3W}_cOnOx9B4)+nNCM<Vj?xJ}1eDguE-LAev<;m(tByIIM;01mEY=vB
zYC*LqY0DDVGzcHiTA{O)Bu&VAIXW2-ghVF^X`UgpI#h!?=c%fiayF-$FYwNjrRibH
z!=X~gdm3+XQXq9eDbLF45VdG|=fiFE`3wBV|LNc1&;G@~#>pq2BpVKqD)!#}GBn;b
z_@-id{|>uX-{-5p{hNIK{dahFYeXXjMF3$ucuU%G5*pL6DCdW<P0`KX9ggA#gFaav
zFNY-2q>09hfJ`(}OA;5*E_Q^*lYMU8xyjw_ZPG!`>TraS2~MP(ynLCBXP)Ekyu|4g
zp)^R)S;~{oJ&RB-@=dBhFYi+n1Jb-tP)XO>3>{xHUOPg{$d43plwQ{j&3qoSg}V-@
z7cmaZ6Jzf1fwUV(EXNrB2k3-74vzkY@8zBMC&vqmDK1eBTOs?scwE-fv<*p`97gpS
zV^}QWXiyT<WZJgH+St)e(-`X}WsNZ{QVJIHd9*gLmPJ{{F>VNlol+qnbTMj2$Fcvm
z^B_Ck?zwa4nU2TI=X1`SKF#Mp`&nLo>7_V+m4cHePx7Dq&HtGH;{W)6AcR@>^M1&u
z^LyanV2^9puJGDx-{j7n8_Z@C+BSCO+P0=?Y8HzH)7gyqqM|A*7K@7M+#eTpf7&4+
z89;9{Wo0ztpg*OY8eA3d2Z8zCf+EQXNno*GllBF9k&$HusR%e32+HDu!MFyUhByJB
z8&wV><s@{{SqbReBNK<f;~bc_`)GLB?e_;nAYd`b6cVpUokF=dy|T8Yq@>00z_;al
zUFd-<P3iXrWJyM%TAW;RxsT{+A9;iyivGKz&$>uuFl~#mF|8$FBL&9C=8o}>5Ta1*
z(Nhq-r>PgLr>itl;(S1o0z+=ynWB3ivbHXG=JHu?Ufbqif6UJP_xa)%?{MqJIx8!E
zd=Ol|bdkw)!E`d?t6w|d+O>P!-#%b6EiqQaIlG8nu)+s0bEI4#DDkf5!ZYVcM~)Il
zI$CACXi4+O+qFJkVHYBQ?&t_s5g*1!N9cj*BW9h)*g%$Pg7oM_BV<C|dd73Z?z~~#
zN)`cv&?G`Il(3$pWNGZZ3Z+T%9ziIAj3-3vl-4ydPp^%<wc#*Yd?3U_OSsY+rII5Z
zg1EKu-eIll#<P)|6|^Ah7bJsTWbbuXjY<?cNkHZ3G$YB<sJ!$K%wMazW<H<OR24!>
zR#sOJiRBM|cUMSS$fPIDBr)$%a{gk<7tWvO=l|y4;^klX%czx;$Rr9rBCkzG4rOpi
zT_A;E{p<x+PHgeSr$5Qf*Iwh*fBEZd@7(6}N{_-dtSg1KQ@nRf+ZyK`S)L=Lr0f>r
z;cys7<|>*Idx%kNc${gmzC~v#vQrYzrc>_Uz036v?@>$7+E$;bN*N5-c<QCkaeHsZ
zul<+*Zz??hn@|5NK|^UBNg~lI4&w>XlywXTPjjSBNYa$abaohFC3KX53nv*41_%*(
zlyw^&`Bi{krxfyp@A|wFoq9}s_xkFGj(^|8{um11sg=DW&okP#rnQC;Jk~}<uC<Pp
z6^+&oV_LMN^JNT&DH8UG=weyy6-B{p5q<5Pjmop5Rs<1oXL{72_|fxnwD9vGkmxuX
z-rL_J&vRB+S9$Tpm)PIm=hm%TJoDVMG)=>W3m16)g%?14l}EdNe^S`r-{r=QYrOHs
ztGxT}J5<#I=h|))QqeRObyZPSHH)%jHY=&hma;OB<s^TMA76+7X`)%(=#v!*`?nS>
zW=-T*?$jjxSeiA922}`p{S+-T(yTxf5~oY7D3Kv%y^0X6#DenBi=-&wHNwU9LL~-d
zT%Vln{14kW*8Mog7J+EAER;k`MIuy;E+pERm9C<p#&_xR;;=Q;%~21XJ_BizGAIUQ
zIwRGFmM$Kwvp4A+uMU6jQ5wrqeCGY~wHmC6v>s0<BJ?i0ba>wp!oE|E5Q4|L3YD%S
zl>`y6PJz$Z-LDWgce!x8$LD_PJm=01IM_@1;Dc@M-Bkp?U^MKbwPa&$jjC>#PG?-Z
zev7^RDb9J)G{Fan^9mg#!3Agxf|5iv^s>NKk@4~;p5VOQWRAn;8QbGIz2W1TM?WUV
zE*5qQ87+21h*hVMgu^#U9CGo3gVx0K0HrlbCp3*?GBxZ?8s=6o50WZK9FD>VR#d_u
zctmK(@{EnmE%JN>LE;g3Yfwpw_8@5S&ePU4O_I<bCNYYnlg`9g5*3M;9UC<SIx&Zf
z9U>pCI`V!&l4_LqsF(}}K3){M%dzx{Mim)SsV>vBBcR(jf|WqOA9FT~qCjg+(=>;R
znfC$}0=X2?#5nN;l@Xk0eUxL8f~GZ8ZNpGYLe~wX6D0(t5JBLqAxR5@(zH@wgk<aa
zSNP>~Px0Cx{{e4*`S&@g6B6Ij7arS`Bw5bd$_7d*7K=GqmK|n*dhb~*=G1kI)CnYt
z*0v-G2<s6_Aa%m5EZN?larMJ(c4wB*S5(u6pZ?s7j80wTpZ@m0=F4Ay4T0dvC!av)
zDHg|_y*;+JHW{Q6={&)DZohw(bLTJcwXc2|gJ&|CF<4oR-Owz<hnQp`rDUVy<M~cu
zEMzRuod+KrUdZTS#{-W=>pXHv_l$zeA3>u14;S%)=Ofor>25P)QHeP8uB)n&EX$cr
zCsb8QQS_p+nP^HZmt*_3eMlhDwx;u5TSSF5;n2_SK?~|(dy}OF%A<wd13Q%XJ;UJu
zsRXA^onkl`@X9N%ptWLeZ;xx&u5s?-1(L3V_{o=FW<31{-#P#K;o-0d^`5VM^-H|{
z*6ZB3ajoMX))>>!wl!5*vY5}h0<UI1t2o#%DJ$3cbsgg;0qs6ismGjjy!7HH*gmnx
zox9syzjh1zfn%;FgkiMqShPS`u(qDEK3rw}!iY-lqLP$ErD&6YZ3v=)tVFC@Kq1-;
zs(x&0$g(rwH4k!x>n`VaX?Xm7Jq#b(HWToaL{oAt5)zTdh)Sg}wqb5(Oc>)1i7>u9
zfJ0qt_ot`Ga@N8cd0J4U(LvdmXzqWQMZEm`QL=QmX%C_6*wxT97Gqqr8w<<HoWmM>
z$gvmV(ZI)8-+0lIq=Gcl)LPKg4%<K@1L~mR+J|?z{G{aRr&d@QUcj3jdcBOTt%BjO
zhtirs5$NS9MLHr;XPKN_#~4Rhwe0OqDQkx{4pTSOc21tO3|8UvsZ~Dx{1g1m|M8z?
zp(S@0Gv2v-i?kSVa$`MmBR?V@vt#EGL89@|KT*V_iO`*t!;=%DMTH0o;{rAi-)!$F
zE5~@!Fr7OZpVRnQya)ohgpDj=Ee$9Y7_AH$4%cy{q&h{a1Y;~J2$V=rT3~50U6v(4
zR^*4HR?+#aiLS_7I_WQ6@F5<TQZW==%Z`xhK?VUK+C(U&Kq+*dgOoHj#w@Wc>bjqo
z>V&+=qbY574AC6d9kzazQgJuXk^0fyF#;INhgaWc>$y7=C(fXf=u;Pn=TL|)n^Y-+
z5QoJyXl$x6MaDCK<}Wfhae^=X^M8o7EkQO6bIovdr7Ns!vOGsi)y0d&6(-RMvsq1=
z=jc44u@!~RP)$@xHa2i?vf#?CG4I`LsDz|8ia+=BpQQ<!fB2972{-R-Bbo+pEupFL
z^BGD6)>ixEiNrJuluo$&;SK)fKmX_a(x3kYE<gJezwsNt&Z4TBEviUp?TUvOF9z%D
z8_{QtF7!QsS2CJk%c%cXDnV3+Wu#m1U5s1!&Yjs$5*|3QJIb;mAm}B!>&V5fU0Ke_
z^Ip8BfjmnQKwZ~|N?3@h*}ASN%b1X0P5gT2T-SB$@@cyc+@X42z)?EF_rmhu`-2{J
zUGasV{ye8no#M=yQ-lz>bLS2_yF2l^H3pp|oH=`jvW`;2#}tlz->|l2GTY<V{=u(t
z<HHY`%_bOAV~oX^mbxyP&u7dQGhEwJSJ9?rXKz8<gyYfcPZ9*MfcdSO_sYBAHP1Z#
z92d?%$tPa=1Yi2*7a8B1V{f=9nrPtuUdwj2i%Mra^{LBjJvSg*?-9}k!UVi62}<I%
zLn+WQL7>4Yym5FRur|=tP%ff(U0cSV-MPR20K43WJ8d&&n8;>FqsY>XUcX?jTk6`d
z#{tnge(6rIEbf+w&K3iPY>F*1)njGgNd1DQtxy{31=yuYvwh@HLi~}=MH`o@f~q$3
zTXVS1$K2}JoQaL8fYuWAs5hK|pe3!T=npdzm&Nrlu>>WUPUrl=AAFrZ`n}ibr;ZaF
zYy8wFKg%ayd5)iX={ZhqZL-}2F7!2xfZ9W8Jv|jEmCgsY+rS6!@5j!av6KfB-gy00
z)=&3&_VOA2!e?J0(V7(^AKkfq)<>(PiTL3acGg8gbfPt>goF?ku0(%3Sega~ypYJC
zVzyz3{&00;sGGp7bhK8*cwh$#ffklQ3IiQjDI_av1;fFBgYkm<_ixhc6<oY{rh9`@
zY-33?9p4~hFxFC*F}PfG4$iTA*P^6G%5H>Y49@%L-KOKq4<Qgl%of!uAyJW?*)$C<
z2$W9HSsH~o*5IA%vTUOPs|^8_ME+iy=kWr9!)U446^zQ^0N@--2)s@(-V%aCdq-m7
z;O-r6TzQ+VC!S!C7NDY$hZb><6eC+aE=Z(QBq525s)9xeY!cXb`Xzqqmn;78pZtBq
zAmwZ?M>rb`aivIfbomZl5!H6KET*;Oc@&1ss|Bm8BeYZmp}D(1<Mr!zc<1Jn-6mrp
z9iRIAC)s-9S$^kte~bHf@1R@DC|Ft(*uQg!Z~Wo!bMmRDSzTR2x{{`ubLGuz{NAts
zHuKq-U;mANh4q#z?_Fg)os$)LEE3|$0|G)yPM$o`B_y;l=sI{B<MB4oQwo)(46>BH
zYCqoV-3Ye3kHP-r7I)r*@zIbo%XEz06N09RMO&{|bcI<UggA<Ao0>F<*VQurHBD2r
z*39Q~7K_DU_btzJj4_zbo5^?MTkk#AKCsw%*edB^EXTJ?vts8gAAWd)y`5eD%3uD=
z42MIe;|X`~-o<xOf&G3T1dK)_`u!eI9`Xf$m{=^P+`4^(Z+`RZy!YM}nz}md&^1j%
zUB@iXvMQO)<}BuOs<L6Ru*XH+Pa>AZM+<iBxb@C9g&uPH+$lCUxA@GbKF!;2z0J<O
zJr-jFdrO6@!~2@s-@1Y44C_yfSUt6db90~}5EFb<n<BDktRGU3Zw%HLjCF@b*R289
zsY(wUX5Tj~%jmH{z|eHFlGt=e(PUYQ(h3(mGc&~}(bx8{3X4o}0s+|>RVAeK>657p
zQb$amun-JJN9z{3B-<Uihl1~>F);{<#z0*Ms@hPr%@I2<dFV{wgGWjZE!qwt1YT<9
z^9Ac#(H~?C`i6t%fL=btE6<%fd-z*#F-a}SrHh>WrC;RS$sw!5jA5RTBvAVqnaU}2
z$|#9J461v^Wm#bLsrA^{@d4Y8xb$QXmju>^J@PD0B4Q4%$F(g*B2hm~yAwpGa3xSk
zg%%DO0``d6L=b`?qQ!~#g2p&JDbD+@E90r#fOiThG^vBk8?=!0bRbifERhWQ8OC__
z#}#+)O-Qmi!_^)a&YuD-S_wkO4bxrzq!f`0B%DBYJe@e0b(l^eco=IWBsdc-QwW%L
zfe4;fr{DusRpSv1*H$UA3@5;}_OO^+a_g*fNUhLHqZ8fj3LZXfNs{1v<RJk_p%6~s
zg~b!|R+Hf9dk+V@+<5a_oWAq~PmeZGSpgz`%~Dk<g^C;t?~%ci1VAdh)>xm=hLo*m
zU*xBL`LFZG|HD5f8;nrKkcU8?q!=4J3#m>Z1gtTnN|W@51nEhN4AU%Vok7SP;}q}T
z+TmN*?{Rk~sYK3Tb(K&2{4es`U;0CKW(%ZF$vuSDAzZ*rXZ(v_`)|3sy~|Udc$u}0
z6`X1K@BYF6m#wYS{D*(}uX2BXk8ghKTUZyUTSM0C(bP?}$FK(HAx#s`oH-j6%rva7
zj+l=H)npC`M5<WZ+#u1Ku)QA(ZST<{z6nYxR0s(7eQ!~J>=C(!ODFcO%Zb1_dnEo#
zcH>%)K#(R$r<#oql<|0cm?gVRnNUj6v`ebNgYNWlWV`Ix!FT7@EfK<!DG_N8MUi7n
z%j(LIbLY;mDCg|$?J}KBdGpOTdG5s*c=_d*qszTg^m<Vg;(mx#QwV|ie9Fz6AMnn*
zZ}aM_-(-J(7ukt=nx;M+P1kirS;xX|I-OHjk?Z&q#|1dXgZJD6HO%gm-0j^%1VOJD
zaQ@7B%4&j?aJSi`nwcmb&Eq#sZ%j!01&V}WHo_M*UN61t6w2nPOoI}*)Zx617CNn`
zZ663HY|F>wU+0KrgXju7gYSN*M3A5<vgqsDhQO?v;*<lmEb1J~F|NlWJ9cz}CKp94
z>=I891|coZ1(MWa+o*t5M9y?@9D3L-H-@1TP*gRPm7%B_%u@IjVxGI`_`^a9lzg;f
zCsA3!Vz*?JW)!JnZB0_{&7+u58l;FmrR`$DeD8pAHYbsxqkchOz(M1q2KZ4TpMZf9
z3|6xvmq*UV=^+c#KBN~8L|28iV#HAAhf$j!6_0IDtfcHzD*n(XC-@MB72VMX)M2Dn
zRkc``pk?Go)yB~nfe0CiS7bgCyAv%bltXJvlBHyM!hBw`yIaz<fnuPkOigV{GL;~R
z+&S-Ej77?GT4PW|KQU)4mS}k)mfZyBVwWpPfS3c@wk^&Wd|P4L*qv}5RFbfM@)V<$
zHBdVGk2x3pa0rNQfzxRcyNFuHieCui;i-ruFM;R^vn2yjbfeywqbY?a7qBWN=LS7)
z+_}kzuYZ%xQ|H+{caabr)A@w)_<-?vOi>ItdGZXs!6+8e!6P;3B%r;dbqb&Kc;cnc
za_8oKuD<?twoVi@CN9j*7<yR|J1tF%bDliS@D>)$6qO~opedW0EXi58z=u0iZcQ!w
zPS8ltFZ}sm=IY)(sv>3c@+Ag9QdZo0>mBNH$|w<t*_7Y;hyNYF@sIv7Cr)khSN`f>
z;kgTE`QxvBlRy8<zs%*QpQSe(ar5>~=H-I2teKY;ZQCG4it4(o%LQBt$%&IEs1_9g
zFd-0>U~O{)(=_oJx{Rn2UBh37nE(5blig1qmJxqRlF%RYNYf1OB9Bh%1m_%CmNOXk
z$<v(RL&u%tu%OHH{7^X2>-CWY%AG3uMyfcbt*R>WAdl=Y|7}IxQFsqYBD-VBSl`&7
z*XwcZ+BH(0@P(iL0_V=2<L$TK=KT5d)J@Iq{$6)K|M0Lh>ArgP4Sw(UewV9PKVWxv
zA8^#BrEOc9rll-P%5qLwE|@K5jK>S=@>unH3?3}-CER^yhuMC~qAvNff9_{_;hE>y
zTwg&Yuk-$^dxX6JtH36(2fJ_FC(H!{y+Q9%!1gStfC>eLUquUWy1|JV#v97gM#JkC
zsxn%gxQb{e_ix3+9J=rPw?Lc+riY)YX%92FW3E(S*zc2NQx?8rcf5=1`RF%$I0u%*
z6jzHz0}hR%5CuirM`#e*qY{C&g3z1CSWf|28@qBLKrm4dB95eQcn6CG=%hs_bB3c)
zq=A_>j#Wdz6a96ULXRa>qUmj}Gy7nVz7-S&oWHn9?G9+g0X8^Jtc|#R`U3X;O{Vu}
z{P7=tjTfJLmNO?#G8hi%3C&uz@;zxMkDg&cOq1&;=|f+a02_l5pL+U}j7UjHc%1OW
z$13b}k{~=Jh_1aTVx&iB0Q=x*1p$fmP`8oDC~>r{iAKOdql7|xMe2PlwzMWwjwBOg
zh3a^7j;b_BnNaiw5CU~mLE=b7MsScOX*A}QONJ#~*Pz81i*N>wLv*7a9V0IY-qRXG
zRhBemh3|MoLMle9Yiyl9&1iEIpJ*y;4!Z#gffo>D<O~Td5lSJH>PY&6L+k90@##af
ze9$WTv4s%830&||lI;xALW*Fd1Y4PA_xgK$^$&lKwbSRQsyXBFJ_nOAWmTb7&Z*Pq
zc=oyHIe+;QI*svbN_n)5-52LI=pmo{>@V^g-}*MY;|Z%p+Fj5ZDFs#2V4B#uv(|za
zQQ?%P7^4tUgHV)>VP{e?HsEr}`l$_0UOLI7t$FUuE3A)J7z%~m8*}sKHjL-=6G^pL
zK<P-d=HS|b|Ih#R|Bg#FO_p)x$~$~=Ws{Y)H9q;tSNPJGztnN#0<AG*NmMes;L+WC
zD`Xs>r@5v-Drlz_&U?}<qg~9SJYFn$p3pi-LIBnAyM6-ex+oTk{&qgt!#vj*DVC9E
z1xm%oz%piRFdUHTl+kGPpil#Lc6JCM#E8B$Wimb>gb<CmoyB%Bc9D4y4#g1uBUCoa
zx=$-b(=^OxbMj|WX7d@p`>+2s!FdM7kY2w}Q53AKtngdE^V<}?9-Ak+d-sQP3fbG=
z=B>BC#c%%0f5E+b_qu}DGn>sj!9>i=T`U&N7IPN!irK8j91A3laTLqRUkmel3$9kz
zIeYFD&pdyT3uiB|G8&MjZ*uLeJ1nMk^u8Mc_W}ot1HRq9&C6@+1mB>%M~jr6TtzC2
z)dH_;glwYKn2TbGy6LiEmmNFk=6sJH+_G@{UU$*5+vbmaN1&vN%2V87eAsAZM8AK)
zfmm=b-p8CYgv1fB#F8TlDdJ{3wz3tPgp5>YIB7t85-IU0R3#BqI2tk;i=N=YcEy}K
zqQ?-#(eq*sS~{kK32B<*oWpxdRhB57gOCLEu;5B4Nw(H!nh#jio^e&uPsUujyvfe~
zg#GaZ;pV*X^z$5yHkj-k@WnrRo!8&^-*NujX`XoEJfHs5C;4Ch&;K)uJmUvNie!`Y
z`+BZDR$*71K1IEKpClMk8H*Sr1eH`YUCvnuv92*d8zhwzn2<tSV~wO~EY>F=6kd2T
z<;i8Bmk4qkSt3b$8I6a%NlR5pvLc{V3Bu$25F0c}njn=#s+dJ9rJ~3Rytnw)U|Nf>
zDq2$`Wgyi_<njd(AI4RUZ4I_*2`*->_C`a7o9nEc*diMZXocwFKRnJ_<gzm<Kq-OH
z0jUB(M2->&WJsbhqLvsO7Ekb?y(jUIS%J5X(%00@oLN&++LqQh8XKsE#3VWM$vs|s
z|0=Fnp%{!vi$17mNv>qeWdCjI*_?7Q<MOjlp?kwlX)DNr#(0NMJgaBUa`uIn`Su_F
z3qJqc6Xf2ZrN`9<iJ<7OkR}-fPZBITXe!&HL`x=9QsFU8&7y6YBXA;c`P2p*Cr)r-
zb(QtaEfS^3f?)T?O&XO@xqu3iTxqm3_|WmK6fEkRQdeBNc8!;R_7{*&vc9^;*%K#e
zO@p&7QVBdU@=PX4bQu@XR!0O+uU|0e<=k1^MoPtOHm7Wwu8!`CI^uK8_<+Ji>zbdi
z!tRhGy|fCDQOq=-mpC8j_hZ4AWjRGr#0Vl0uWy~EBw2=Wj>Z_K(`np@c#jPpVJ+kN
zoU(2@+QNgBkB4(%KiWxG9i3t{8g+b1&tNcQ-etqyxpR-FE??&9r=H;te*cfSdFwVx
zE4EIYV6@UzQ2yh8li&N`E#7_iO}_c{uX6kLEf$M8)-KK14P~|HI&~%U`GVQ3iO;yl
zAUYr8Cj`qDO~?DSQ^)JCy#vy4=G-Q0>jOUhsTT+#aQn_S<Gn?UA<N_aU_KbLeIqcu
zAm}NL%oC(0HZ~HE^M*7rXkEo$<0H2&EbFnOx8!%+CY}fE{Dkk`vGb37CeY9T%S@?=
zT;~*dMw-F)-VU{^F&GjQk#OO<kB*AeeZLZlL?zT-pioGe;B}lNn9P$|Pbv&qXKmuT
z+g^09ClCuhFQUCmS$P(V1%u&`(DAj~wkB8UVb1u&?@z#ML$<aB2Q5`)*xoByaXGzS
zPSHEb=0+b|?vYr@Fd5OTw9K1b_V#AX=kIa*_BN}-5vNX_U~_AO3+K;s@zfT&n{+;k
z<q<5;{L#MM4-$s?ScTom=bz@DnG+875P~kmR?-H6Jv?lW7aohi1VL@UI0ar}TY)u_
z5TX)BN=P-ND$rM+LIrwRG|(+8N8KnOL8}@`jkBJztgtRYrYXa`04Hcoix3j6;;1;)
zn%X*2sgOyE4K^zHq@rv~%GsQHwm=C%A~ji>QLL@Aa$=Lg);gk}(GqC=1BZQlAjDYO
zXt$vyQg@`x5Kq4dvEvmwlU6AN-Z@gO5qO;SG_7UP)ZCj*xv{;&{mG1j$}npUb!*UC
zP-L3X`Wp95zz^0~KXsm!q|ergQ>?FVG3fP~jSpB%g}q~RWX~J@`;9xAWMWNh+qP{_
zoQZAQwr$&XCOxrj+q(Vx-*xZ0?{2^AK7Fcs^{T3Ms`j%#yKKY#+ow^~@oO*Xg!MYw
zUC0t0v9RUnRKX{c;d}HuZr+O}R}1*BmuQ~KV@sm+^F<#JUrX8yeoZEAx&yJ*Mth`E
zSz7XgFxX`O0G8gs6DOm%y1I5ZM#Lj-n|I(WB5A|x*(W+}CBI&O^j9%*^U|(bG`!dI
zCCzkIBCk}(?QWQEuX}=uxxARRjXhzJqY5Qd<$0>Kn2H)F=KFPPlw_8yxvJcz1069J
z$5VLGqts1;&TUfU;k^u`WD`P8u|*DH@F2H@0?iz4%pY5_`k6+z`fubqc`C8ZR#myb
z*6ha28TFd9g<*PQTzb?71nEDAB<Pzv!<<qd(}CmwA}uYg5pB#66v)ihk`^ZO7F(RJ
zIfgGXb#-;ACOyNhePmzW(vU5ftIfW@=XTwrI(9vyDKyuLDe%z|K*(coHT1<gkdnmD
zP=Udwc3SV}I72h`;vm}g4Ir=g@kz6yp^}vGvky=yTjKINg~X7#&n?Mhs|R<FqtW?c
zD>XOkUzk=X4lATxJSk}sdC#Y;RWLl~-*;}tr@cZKw@ZX~`(w9D2^Swg5bIl8Le?1{
zJ|s-42K(8azYOHUBB-~YE!RwFq`lbqod!%Qj3~(#micumBQ#<HdkN5zPxW#<p9rRa
zzr}arF0~o|*Tbv8LHD0kK?TIU*i?C~!^{{EH|Qi}2IX=$X(y=66+Us7+g4pE6RI~i
zpd1?k{!X}w8IN$T-;;f?hG%W5QQ%x<2_)eec{3hudq19ATi1g8{xc;q3Xbm!<UOop
zQ_Rqx{4_K)2Ojd7#iZO^yj>EzvEEl*rY;{{H=xNBLBY^(H2sy5>c}5A?*QW-R{wSL
z9|*jKH``ictkn8onqR9Em}HS!a7CyZuE%{yr7XoTzr*XYxJL^x(l1ZOx6&Ei_W^f`
z)+zGql*wky9c68{k&Q~1qW}rwpsqp8gdUU5bVfVvwax)nM32O|b1`7Yz$DzzU%9)K
zpOpsx;QX+x3A*FFxv2;d9Uy7g!lMaMc$qQ0_?<HsF{-qSgCIu~OPa|QxdwbU?*@9h
zy4{?@p6v+g`D!U+iV-81Lg-xv8nxVKAg5Hh8>gA@qF)9veM5oY83y?Rx?*qx)^=g*
z1p51<fB&bL1XZ6nyiI!~OBd;QXjn^AaeW_<$?LWc<6ieoWoYE)=ao?A4AV=62d|%8
zV#fS}<v(hGRsfwqC2_SdZFZ4!Iqb0x%~HEvT`RKDR~DGy^%?j+aczB_ainX80GO!@
z!Op>L+Wt{Q>>Vv4-KhO_W<UPRvAqK-&cO~X0<$`|lKkmr$L`FeJ0k)hl|li{WRT_@
z&rI$MLd+f2UXE@c88n5(73J02ZlX*9>Pij-<d_?2Nv7^Z{5pezm}WfAI(K(={ChSb
zlBHKRHi{|XCL2M;2Rzfq4iB6;@?0Q0aWe|%WWl?yTt-?dm9$jpJofqdIRbRh&@Cg_
z({oW=j!P^O|BJuh6YvY$@7(TF;WfqJ6@3TSyDaVF8R;0W@3GD=r_TE@YopH%W_f8d
zoy(T|tkcNcEp63XUW=U;BlKy-{sn6re#-35ajjz`=%WA8?OL%bvYNu(%spW$O<v^y
zd!~QUZiDo}=3)cU@J?uAQX5I&UpXqa_`(6Ra+dNk2`W6KQn$q~PSg$yBv^{n?B+20
z=9uTrtD^oQeUbn{8HnH^=kWhXE!0Cv*x7NPw$_ZduN@LqOwj&4p5m3>w?##228b3)
zngl)*29f@xuKW#;?bNTx8q*)nwKdWpo8|mV%INp8Y&DIyE_1rPA!l}x3oo}ZVx`Ii
z_2-T4_FL-b$S?UH^emW}tqz>~x5mJ)fghG<oKBAYG77n-C0K<@)8_En9<ihJ<ync2
zOy=C1`!RMs5_$}ZD5dwXRaI5l+#Zpbxqbmazb~h30l(auN4eI}^Bg3@4K%NxJTBtl
zCwejmJAAgsR}FAo&ORaYK>i>*>ejmMI4eFO|GR(#Mn#T)P&^~WP`FqM)QLiY_cDjF
z$qka&V}TXTCUc1tV01-Dah9~iCXxk;!Ue@5&K_x|Y*3b*4Gm(-#URMtDpvkE&D6jE
z*e9`hGq*6j6q`=`IAjzV5MG?7s=BuJ-!(?;?j#%$kVs8)HKL8@F#jaEMH*mjRHVf~
zMQnn+2L+4IKu4sRZ4dZmV9uC03TUMuv)k?%YqQ(h+kEfZ;I^>THNhZ+51cM+9A_J^
zCqzo@X(SWqm^@m7HEZH>yV@9dUE%8Z`g+&(`-Y?7SoYWr!E`^LT;J$$3Y(pT?h#^5
zabee1r4)R8%l5ev`&!iK6CEMzo>P^Ja?PK=;=7M+wINoZ<7^HmPS;XvV$o7;o88^u
z0@^c*!pYSMh@J$q%1I(04570t<P`jfbb#-;3E!E0!P2gQgeUk{Q}zo;>E`xQv;FP^
z^3^No_YMyn5K^ZMGHa-cHC9C-p;eILp5tEN<1ymJ#&Zin4tOig?ZoPS?A<6RIKRJ3
z(Ouxv_ZqTi%{GHd$d#Op!YgarT`##^Hg~eew8D@Ref#{<%Qjg5{vKX&+~`yt<6o<I
z_x<UoAKh{O2nU*yFPiqPqluXomT0j59HxnbPq;np?}n*%X2$H7jDA0__dPOb(Q0S2
zz06QkdA_hMs71qO$^JPzbARxR_JhGcaPU1AxE##QzL9$*uY@_4&)c6~!oEveTG`w(
z@74Hig+vj+m%hpIz|MPa6Pm_e2@@EMtb)$HzG(69Ym?PDu9B^9KWV!rT2sA9rZ$H}
zQHAB2QUb{>U3`e2LLRquyB*kLAJ>#^oWeG|D?8z46Q8=(=nT8_C+@l}2N)_|FrgfR
zS31r+#Lir)ViQ%(_fSZ6JL}NYgoiP6kMi^+Gsikg(^_M0vpKpld2@riH=X_Zw9MJ_
z3JYxxy&%lotB&ezs2Z)QHKDn#yZg4uSGg)jny6AH-{|W&%&7Hy=hwZ5nB65TJCM7w
z(!~6o|GD(K!N+G1_qzM9-S<};*2kycu*HJ8lTV1uv4Yd->3w$J17sO6_i*o&!u@No
z$wc@jsQv*TcvYO%<ax_mYoj?$Qwj>r$q_?QoN7}OL~fYHiMYrdnXcDA0Avt~>6(<>
zfVl7}67z}{61kAc;jZulL1;L|v6xk=q*AKYa0px2*)5Q~)M_W9#G;0F=F-TyUlv#q
zPn6_QD0JvqI$5B-?npjC3zs%Y%V0@nJT-^}4@J}t!y!u18LvE&YN1G(Ko|s0pvb(_
z0GQno!|ge<7B~K_?y>c?gR$-Y4pF<YzB@ChOcQ__NjfVDTavYVfRI8h`>RQB=)a?0
zvMQ^!(fTHztd=A<3+*BA-klhZt(~W^WldRyyN!DV7HzFvLuO-M{3LXr;_zOdEipC%
zVf))x%+)IGrz3P7%e)lPq%*V<Q9K7WHpqx26f521B9fJi)C+lPZ6xW@y~^?Z826_O
zkAG5lJYnVhzK2eEWT!}4)6Y^SghR@bQkf+POJSNd$Q7g}pxyrVH%sky*BHZhOW&lS
z>F~@KKZ+PN>r#hpwzZPXSLt(gG&HN+#DKdjGPo6lyeO5|&E~!o+{e$a;Ub9OJXYEo
z52N7~h^AfE7~RYhE_3@Cg0C#6)K_m`N^ef>>5ywC&rhL(<U$+VMA0!Nu}o(EaB2JS
zJ^?uu=R#Nz0|Kh=y$eLe{#eZ0kC?it0wvpiTz^M>ua7y>BB#f=If&z%NiVxZL(x=4
zjue{kT3Y-Gdv`DG8+(xo{!g3H82Dl~)s^KI|FEjNYGPwEeys1CvE2KJ;RS<wL1ze6
z$qWa7F5XBip;S&utIn=z{Xi{8%6*HpW-;|h%ifD0`Cp8t>qYPKj>1nL%~UEBXw999
z{~h6rYKh>Dl}2t6lISH|yuHhAU=_)hWiRTzk0~jQu;GP}i?v%6?sW|+%!xz>#}AiJ
z=o{6u3t>fB^DX($60TjvUj%{AqxyW9sZieeFM(MV)Zm&x3kx{zhi3lk9`n7Y4%$j0
z&4L)#WTaj4xUHDtb-885oOyDP`@7tGM-4KQm_M)aEtM8=Qr$CVEdv&ZUV~fJrjJqb
zo=;6fbM5Re@pR%c*i-+kH*)m;T{iGaDokU<LF9Bgq6MAL-h--+cELSzL7A0EK1L;;
z)l+A;i4-U?vMkWYGY*dGMtX*x`}EYJNqXV%{*N}i9n!=i7*6}t5)2cTB2tA$FcfOJ
zsyxljq?3u{a8x{bT-z?%q7vK>_+;oFrIvCZIL7N1U%22dTFiO!r^j46R#nB?sM)4|
z%x26v4qt;%&Z)bi`3?NG_Sq$Q=2&Ec;t-;kg|GE-9%>dfih4zw?ja~tbMS#Nw(HF9
ziWtEIFVL5v3#Ox8phbM`o#5{2)2H_OZ#sLXb^T~YOj7l8Xz|i~DC~3T@@gmJ0gJ#r
z+_i2;C<fFs|AzVT7TEqJ)B8OQ?|Xy-6He_-Or3X*6D9-53g7R|WX?O$ez)a?HpIV#
zL!QJ`Hj^DyX=rt`<nPu_h;5zi-#m8ZF0M&{BFvg+lL>unB%8KC6)0FqR%%(1*+0;2
zx+Y#Hj2uo_?=t1hYg*qZj{HwsBK%WgiGzdl`2cFGt4)M@x5Vk3*X<VJa~I&rhf!K$
zD^t7KI|ULfhy32$zF5o#n=Gu}hKOc3yO}@%wUn^*nNst5@W?hCwj15p^Dx<@lmTeF
z_5!bz4&PmlA%5Tu7i(Z)S`*udQ{X^;VT^ZuB|`zq)REtrr3W4}qobfw3I|aTu0l%-
zyP>H`X1x_nOz0+nHRgR4lxE;~e<a8MxhhHg-Hr^t(mH?}azBe<sr8R;GIF)Mly<Wy
z>&qB#|8}=uL-CsHYLnBjr%wSzS-$32m)kGkvTHqkt+B7kA8ci5mOZx>eWwXTKftKd
zZFW;;eslZteFJd!=yEk`E(-s~>y%$@tZMx*b2!LKy<m_5AhI`W?sop%Qx^QGC=J3c
zjxd@UjZROa6=^#si`8GOXjyCYOQE2c#JA5hZ-<!@C>5y+!jGFsGbr46MA6XRvuA#H
z&OG!Jc+?C?@FaI$yo}7w#J1W>^6Y<KyORv_-=Fc_6-qEE7t;h5r!_d3MMWv-9fVAY
z$fCCQdG%kJ%q%|Fs&sP0eRZ2C31!k`Ls&tLN}6yJ&JqR=w7?_7xEsvBC7SgDO|CIo
zLHGspX2+oK9fnV~r8fEaLoHUSC#2kUkY=Q@0vn+eX>FV|541{_Zf`8Nyt*Xp@^H~y
z+8Y3Qc@lPN@+*j<B6P<M<K2@Vir11#8ZuMojD0k|qx_S>usNeii6N5}AMWtWU+lK9
zj_$d<*4IG-H#;baGfWm>)<o6S)df0FO?ZX@U<aKX%f<YID<IXEN{-LjL$S{_G+wO>
zClBqM%0nj-Dd@)q#|W0N2tdu9<i@F6>ESrue7?Rr+Wb{3YrbI%QI_|O(DzUQObmro
z(`jV4w|B~ta)F`H8TE2|S*o%tn!|4rlRJk44b^s!V>o8`Y<{2ACUc+@2-!8+Imk6O
z5`}4r!NUJ>!<{NPOVHTaik{ON5=&=^8AW=`D?|}aYRq4Es@$=MB`4iTOM%R&PaFtl
zmjJGfyQ0GyO%s-uDGDeKp4g(d>$~Ny%E}5R6(XnhJjh6G)xMc56Br4r>z&E?o1B)%
z^>;!Dt$u^x+MY2Bga?_b@pAgwGx%2Hf8A*Mn5y%OICU8+iAtkv3=}=nq}85Bf4Z$q
z70xe?9$kEAn-d{7l?B&|)lK7gX7aVRXZ2zW?l)xIIW5Z@)f!u@(y&~=;#NHSS6OC*
z3mrs6Q<PF$H1?b0ZeLJ$(%JmThVq=#GY14P#-zJ{y%lqfjtei^Y$ys@n*Fv5V3HLm
zZ%+X&9}|+Ces6d{fp>(&<e*!gU~c3!K?}a#AMM=3Qer6gW2nC_t3gR<i8!g~)3bwD
z39)*l94l8-Pz@qw?uL>dH(+75q#zBD{E|C__-<0{7!_2!HG395VLi>slRqirS*(eG
z5d~jMbB&-3P-b>c8=}u45!ws9^bBW)#u-nJM}O%a!JzZNwdk}hhBfpaqe4Kop5CtU
z;mEwwchXuq(E73aGa($(Q&B;n-;5%@2P~)tVOcfVHx?a!CI`=v@n`p41jR(Lxg+Dz
zR;=h5XM*Y^v1uTHU%4?@n9zQ7uX;EgFCmvm1&54oP(ksOTaBwgKN>r3Y-4><a^1K`
z)4Zu~cH|vnnqY|k^p<!eddMWu$TVfw5Gus*`AP388p*!$l0F_eXf<f$l{6MIsMr#r
z<_>_}-zIsg$eqtdDkVrUTQ&^=vUSoyN&r%KikRxrP>9&5G6{b%3vy+#Y^@tJWhoU?
zRSkg+Xq%}?t+PHv;twsVMaPM_7qEKN5mGU0TR9NjO`((-{Wj-oU-?w(q&edzU|SyL
z6K93~BPeVPuKu`^0lW3O^W$1yVCQPZp{}|^2PXIJ;rOM?@A~?-N#MKbVzV2F>384O
z{x-V4L50-r`_AsQ$iH7YxYZuez<73Ok=MkpZO(}2022N|8la#V1ifzOG)<j1oH>sE
z=bHv-zr6D`qqo)f1P0{&2FIk=AJmp9OzUW$EgivHKXm_-ISSfgDkvwij`lWhA^*3}
zORd?>=8j8gHPzgh{3Gzhc7<@I*9WvR=^hxG<0_dngRnvhuv>)L`yWY|?Kcl9?YA!7
zRg5lE1B3QfZZl8CMt*$<ujcX&uhUec>+7H(u&*5*5Ou3@K=8;AZ)6Sm{Fk!3ci9q(
za{NhgRCvUn@^lv3qW(SPDLu2WJi)3Y0t&oe*Yrp5w!iQc)!6*@HNLqq>2-JR4rS0h
zW?+`>MC*o-qXt;aXX;y9N4Ku%Z*E<XiTKAdTg~*@XT95PGiHvX%R)=JdedWXv5<#T
z*mnma&%=v`iZ*27t%q#-!ZN(zG71At&WB+e2mHN@#?pE}<Ti>2<h6xSk@a`7MZBac
zJ)13q=5Q(we>3f;g={pA%e=%!Se-y3Ha3xZ2m*8yA<|OvuYPd98L2;JD=)g6P6P|b
z<6{_Apjc<G>=-tV2{lNxwB`e;Ywx>Ra@bubv$K<hA10jRI;#^V3s8mVf58!`HA*To
ztJ?Wr6vE@K?x_V~H1z9mGG6TtwUMvojoVUl<Y3(m8+jsWv~j0Z8(cZv)ZktKQ4pNy
z`Cl!`2{!gWyTFMstZNZpnt%u^p0{WUx!U`K5fm?epKAn-Jsf59xW#A}3a+P*aLwxN
zon+U~4j9(R%|b=0NMXdBVQzf?poZv1-l!CA$i5$z4NR9zKBkXz4>9?@m_p-u$?<{t
znv!?j3jDf6P@!snSIZcT$GpNHaF7XV`DtUw_i*u|c;fEkDTNXNk*o!2(09W{#X@2*
zXx__YD%!&h)ef9i#Fp!(0HU8liKW;T(isZo3sESD2wr@VtRq@e*1TZgpSw8Ji_J5I
z#n9IvoP?&iaj#V-vjOd1PZ=vQBx??~Z5jE)a62)B&w}c>h3kp`9K}XE$MJdH_TiEI
z6n&=Kv)i~A3IFH3z+Z;X5d$+8gWsO!F#FP5y<gDZXWJj}wl;PwdaCI=4i>`5q~N-%
z%(4jcjRo7l@x)-zleCQq`@wwPyCuEACxY|<E~o?m$xinecl(1|zJKx!dZr_mQo&qQ
zXMYXaicd|;C|W3i1l;}nRvAB2%M)x1su-~IYl*8AMEhP191R$l=N_s^qVTxe;4-<<
z^>~+i!gM;a`4xiMYRaEw|NVro(DMRY`#r!`S-N{=GP<?l_>a=|GMV@NK}!E+IEoZ7
z>eYo1>Z9*|3OYrw`y}^!J@$<-&ute~OU-w>2@ri3iPg|Qpe*?z(`JezO<V+5jf*vA
zbxo0>`D`3qRJ1}1ge5u}kS|4@$rDgWv;aZK+pafCsH?_%SR0TxR#pfZp1>^D<1_{g
zqYJ7tIRk5`NB2`(AIJ8WLEEuZ&mgk4OG@CS5)BH{Z-jg2fxjs_2=5^4l0<5WuJ<0e
zhONH2HN}e7lo7K?7-C@4te}CIfI5+CwqQki!&%zjwjzSNSrOgLI93yP$V`}=&LX5a
z7{BKMm1R1)Ib`LqWLO-h7lb8c<!Cr-*cfXq#fGg#sNVVdFl5be6ggpG`9c&~)A)*!
z49KdYer9JAA);{WZqhPJ!ySZCSXB!pJ&61>w1M8JUnI^XT7svP&FPqkrjLy1WKyJA
z6!7F(@Y4@JqAE;zXin^Ojr>}HU2c#?8WPo&X)Aa~l69O7{5$Y@ez9#^9RoG%>v(w6
z$V5AInhXu>WworM($)6~EVZK1SS}dz+;_za=f(+Hr(h|!-ltY|PS0C(wws;P44)rm
zTx_1a#Mj2?uF_<rmZ7Rrcv+y8zf8e6&o9^8=f{Oa16U%KCjok_vRDeGK8a~yB0d>i
zxPLr;$(|;M<jnvPqsf{IO;AcCGKwsG%<aKiAexCK!Eat){X$t8Vj?xT>wZmuG7Ao(
zQcpHJnWq~WHW9$gR7@XD5^{?(R4lI3#(t|XM0^5LSaz|Ve<yiNT4E13I7X}36gPe0
zW+uj__u-AM)Hmc)+o_A!@Ny%m`(Y!B*Xz_0#cs9FE9(2``dh&L`HiRh>B%ZO3mYAb
zCStuJKb}%Fe6`L69ukwmNb6r_>bXYU&0Cd$v6a)f*LCld-S?5*-PO9C7?J?gV`1Sq
zb8<A7_YC$~<#6llkU2Y7euZ@M^nh0&5|%lD*p)QMN{L^(KSrIEozar_9tRCu%)&VB
z!Z6P6W6ADzq0BF`A<g@+hQl(uqy=#cH#wO1eAcgH>@9=B!|xq~q=qu+VCgG*{&?wi
zT85!M_aHe5Vq+ALf#7~yajdl{^0gD?H~d$J<7GAPouynniHwSfcgM5DQk>y$0{z<h
zq2+;#I}y*|r4mJx$kL|qCfe9b(g<_j5>6CaASnFNh{@dC9AxUNu8uCxj7-z_i~yCU
zf&TB8wdLLcZ~o);I=?K1UX&)I%Q0FxoAuTxsj3VHoXj-=_T@vYv6tA{^G2Og_k1zV
zPTRJBvIt#m6c@HvN)PYR>KfRH(AivWnMbxUHHh<@u(hoKRDf3ao+c5=vfOY*s3MYR
zQyA2t8@j?U#3Pb4eeb)I(?nrZAe%9<8WCmO9C(Z<e25@XFyVPgyjOOtFl^W<ijl}O
z0c-OMvSP7CK|_of>#XM#Y*?D_D{fbzQlW`(rf>p9^Ju{HxctKj9so6o%KCW!xRV(8
z+J2B)Qs7x!uOiK;J&Ts5xahPNa%|`tVsZ$%5cz2~ic8?#U$Kh=%-uQ|E*}F?GtK#3
za#<14I(Bm0Xf_~kz>mqcXk@D#2F4~xeuz4k{}7c<li1IRIu;Z=Z_;sdO_zPvOjd!e
z;`9AF<Qf{2l=tTCN}?r^V@_6lK?YEh>FJrf?}Cr4>FsQ`H^ar;fd`91M-3B+336^$
z7|HY#no!8c!x9gN7Coz=g^F^bVc?R!M7Js>n@TCOo`{YrZZwi7lMS*L_kf58x57zi
zP&7UKG*%GRu9|d`^WbF4fkD&IfY}Ib7iPU0bemsALy@Xu)k?cvIniUr|5}Gz0zFFF
zJLd0do<X60Fg(qla-)<_x<jZyeqhouu(6fdh<$ZzjqCf<?!&nEC5C~eLi&%LC4}ax
zOLn)!Z>gW8#0CFEJf(zkF-$53!!UlkdOXdcT5tDGe!qS{j1fXUUth+T%-%3T0?Rpe
z7M5Vq0!N!Um-qB!=4tImot}rzL80eqB^D*|(xMPXKW_#I?ThJSGjV^eG^M`ZhgSY1
z)t`e8RSDU7h%C01d8*yqIn8^&|7hBNH%{a=ws9GM)6N^HH{`@hcxE29&vDy&kB<?O
zwvqNObl<(hUaCi^gcAuRC>|GOL8i+49;V6W@eAF4(dZjEN_loaE#nVr!dJ%bi3pS^
z%Y2;W_VK{nMdao*0h|yf2pa`mLPJeaXlQJFc74dlVqW;!Ap@G{I58tul}9&Ld7iHl
zpPA=~Be<r(n`0yOj`5bX5=nn!TeNqzy*K;<RcJO=msdtsT5C?nMs*s0_-n=r$Hj`L
zK@Uol<@x8pkS}Z3)@|lvVMx-QPO0g)f6%@!{&+tBb!D(_*6Ek9TS-0vMI6zzOo-+j
zlB0{^FcnC@+ZAiHzdKyXBF~%Z6juZuSrm?6WSeo9+&*(zI3*4fA6S@I0FB=$h>fvO
zD4vMtdWi>ij3^IimXtJ+*5MgPkXttj899g1XDVoFQC3c%6v~B<eg@ELr>9e$7u1q)
z2&|YE8(r)~&}S6VBC5TPb15?74Mwok>kV|*XkPS9=({Gwq@ast-EBQ)m8&!QomuL8
z9Pt{wRH$V4Bs(2#P-D_(4|)x)UvHFEvel&ck9sZg@#rofQU>J<Nk$PGR!c<q-vo?u
zA|p!o540cPZhIyL-RkAO7;GMm$Pj!gdvz;S&z;N=jz#&S;%kwjS<yT!FZ-X6tiZ*o
z;{1ldQHT_s*BT0;jJkSOVP8hq=9c{>Cy?`~X{W@USwS9=U-OH|y*h?K&R8-DpL#aA
zDN;)ZrMXa3M-1t`oRs4FSt&x)GQO6`1#A-fM^{iK3(gOw$Q5pH)DYa78m37%u{_ci
z6fLc6whNuHPpt{B<>!~B_jT`j9TrZH2oJ>tN&fo~275;<h;3R)M>#Ed_v25Fm;E~5
z1WOfVm^j#5{@4jQ>bJLuVAZ4IM#4oXZFGNeI60+<6C~!$$@(_Gyro55&okyxEZ5d&
z%E)Zi>)Ta^4+IKeuE-gr&Z4R*I^X0Z+%7ml_xlfb&W?tk`#|An37xBM;TwRFx)mVz
z5C5^DU0i8nb+Hfol7`lHeDsnd|3;3}Rf_ovw_hO8=c-IF5CdafU3H%8eiJro3#+LM
zXw5E>Afo5^x)H8cBiRzpgNg0oq)b&DkGvB>k$WKUhWq_<%`)31A6;$}|E>%&091t5
zs6S)$r@Xr;-r<2TU>G*j_Si$V_Wjl$8SEHAER`h#Pcapp2(s4TJHDkwt+iEt$LC>=
zWmHr;Ad%JO@7W~W=&<}INP73R9zz4akbBCY<+VYLh2k1}8Q2Mc#?Q2G-8@#=D>^G5
zaPj{)xQTR(pwS8|?5O=mT|KPZe%CVcGP{b0!~J`)t%J9{pAt7;B?cL;1YN3h+({@G
z${LQI>#hqPf2KCD)~%H`*QfxkO5m>>hRalFSW!!AX$|Ze33_qH)rG=dC|y|`!yoNU
zM$>3SI^p51EQ+1(G0m{~Sk+>L2jq3Q=|VCpLMJXM=5Sp%GNet+;sw%(VhR=HKvigF
zx_DV+Vsh<87rZGP+@8zSBCYHEO2b>p-(|^$?ZoXb4><BbCJ&MOg5RB=hzFf!E-&C&
zz})?}mS3fHX}a61B{ROUX?Zw;xi`a2<}Z2KuOQFE6{7F6Ti*bT#x=2M+-Ngv>j;!W
zDxk8J_}Y|^rMeC*;1P_06!3Qtpz0hd7M|Vp&Uf|%GJN7Yq0uN&zjb4qTj)i>V^j%Y
z%=!{yKgE9r*Vi9N7x)kkCDE>z3-9U$Hh~eK@uy>ro`Y{ebwgx}QzUj%l+zUY6T2r%
z*Vl}A^?Q*Y22*RQYqXuz*fkg|dAq-JXX#k+I^5#9Dvimf$1$jpgE<DE{zxg~q`m(W
zYe}0yCm97<10l}NY4uNiw>q$+Ot;zFi)nvKL*M{p4_jxe&ht&wOPDiQKM;7e`X&J#
zy>Qg_Ni<~M4lL4i`(47>^?cLM|A{3NjkCpe6M>7k+5ACO93D^NEnZKhhI><MN{#X0
z^molusE%`B?G3}uz#|5vV}qcH=0LLtajWi^Ihv0_(2{@f>R%bL<tDc{r;f-h*p%+H
zzv-qmg^o+(gF*!;VEGSY>MAu-$&&t`Dq_<o)1uJS#P884pn;JwOA!-Ka^s<@UrzoX
z*JnEw4dbip4H#n!O!QKl1m@jT%!d|28OWBVY)LjNc}L<YDHmtzKh`pdq@*BgNG7S2
zqO|xrJhSw?QI0Sts-2`2tc^@vo=C1T(8x+pTCyh!{3jT*=&CQkNNV_f3-hrkH+u~J
zC66;0&4TWA+FbuPZ^q3Lt{ntp9$h*R(?1;J2e+>z)x^EUJ!oHHxnKTu1ghdccldi0
zTNw>Mz!ZB6LBt(Ayl1;A#8aufIBZnq^WrrK>DuZx+fkhNOxc+IcK|1*CcW8fqpQ<!
zKZ9#Ra&{`}1SXC2M6?JsL><E+dA;m47mp`Mh(P%<Eo7^)q&>PSJTlHd0!lZn=qhFi
zONfq)_~)!8c*rOy=R;Cr|4;Ovq9(`&Nhcy-JhS>FLLUw}$62rWOblnd>%jBO81Mk$
zieOeM<V}-*T2U>gnCGjehpr#lf@zYw;S04Aq7TZjkwQXDc*xFqU-^5kt%jaH1BJdw
zc65J43fe;PWXrmntn9zZ?eD+1a(lc39#hzoQN8z6%7l7B<g6Sc=34jXNtlZ`naQ;s
zU0tW1_!mNpo|9_JL`rbcJs$+xo2?yPVIYG27t>Kauj}tJWRz$tOF6Ms^=7$<HaZ&W
z+UTJg#3kS(XS*-vs=G3Qq!9yUVzCld5mp=~;1IjUKQT6+cTlD{stYm&aMw$&r@6*%
z*5g@4v?z(@q6YGqg94ZlmJ}LzY~9a4kvcrQ@-TDV5Z<7AXp0`fiJWy$bb@r6CFyRB
z<h6$rikpUSnQ-VJ<aCt&l25Puvr)}A?4E@^zwUd%baT{b#WQj{#8$S=j7$M<HNJ)n
z3ewDIM~#;>BtE_fzA+4Z9;>FOuw$CSz{?4t=>KVz(bnKN!SNBR_|)Urp-0G4MMle7
zI1{JGnCp62W!)=S-!3HOnYlDcq|(sq`6g22d4yu>`-aR;{;}prBa3<kx1+VuBB|cr
zf?00&`xMdVY7;Y%X%r2J*m4O|z-(wnQ?fwgt69N;mCw;X9+ANXC>iOFF2Fi2?oV#w
zdf&@zn|Y}>3FI>4kVc1sT+sCO9Uh5Q*FOC<UF#O+*l<88u%3LZ3dmqnTb*b;=8MPj
zIo3#$V$q89O_|Hvr4gqi#9)6*oC|MZ28Dkffs3>GobaMS);~Q`J6vWJ65@8g?>9>k
z#$HgxZUSi4s_PI}*K|P+%UYH!HBIAow`2S#v7j601Y@`y3(rsdEstnR0Do@)LD{C%
zhX<3>UHgi_=vbhjE+93g{?OvdVtBdnKV5w%#OY_NACXirK@jS9M6ZOVMI3t%Yut7~
zTgKJg1RKgZ^cYa?_&#4@i!j+!0M4_L%f-*(!5LTB^*u&%Mu4Bi>dznMW38wXqk~no
z)X^MR+k3i#2vFKoteFT=@@N5q=eW^-e(H7Il!qI=?x~GUJ=5=0Ur!I{2v;f+356fJ
z4QI3>!__E@CXxh>tY|W{8{P~UXJN4%p^wv>S!Yn4<4-ta2<L+&Hsz%yHes`7`CVL*
zmL>@;m{HT2Ki`(Y(;3ULZCCmwm6+~y9(&e=B|Cc+f$$31hW~&KrDRixJY6hLV`S^^
zWb6CP@>8O1zfGw)5dj4tGMLZSm{YF>>ym)}%IiLvLO2$(t*o5dU#cuFmzMsF)X}jf
z)(y@5hk5e?7h2m52!hTf`fx&6ePl_hEI6j7gHODi1%R0Dc*3pRBUp~~TyA~~-Hets
zjT8%%em-P%UF^!niixsJtxx;d`!W(`C{$4k-Q?uY=#slcuGr2$<(O>ur)cx6?p9na
zY<m?`m>?kLKk#ksL_>?GQQ2pt6)N0Tossotvu+pj$<}#JE~EPgF>n5z3QYYYfp0Ir
zICyh;$IBSERVKvmhrBOx1D|~wRU5HW+5#4Z2Y>imqiCoc@ub8_q8^hN-(EO+B2K9U
zj0YGI^+>j6_30^y=&X5K&ul)oP#H{x+tobZ5OV|H&}vCVG5K}Y<a@69+u%R!WqPy#
z7rA;mJD>e5Qn{h2G$6|~N+6p9#e$yhOj|f|(BB%i*|Bl%Twb1?y%9NPEur!Tn+UFS
zkaVh`MUxmJMvH*!^EW~T7wwauS*BtwlT|_T9VbQ@5Mf293^MVH@c+}~l}b%K)RJWs
zG>Q=l^FMnbijXg_CThn3q<0SB;VEWwP8h28#|~=AoEuAMT1q8Y1fv9xDu&5(&yp$O
z`K11;Tm1HIn65`8s&7@UXO`~p*vA11JcEP$czZq1b;fdrc5Y$M+g*V!n?@*j(i1Ob
z>eV$h%4iOuWQ%id=lcNy^k=cR4nrsswXjU%zqvvYH@DdskA-Nf{9QRf2RzSrUe%}k
zSw=xhTl;fM9#>mkBQrI6ga4Sg7mKh+Pe_TRP*pSY=hGVn2i>S0<NCuaFshl{iS|NQ
zys`sXNdZ4~qIVTXak<Lm)*B7tsMI26SZRjPDS0borDhL^OU6SD!DFl}EELgZGYX2P
zC5gJwa21vJ^{iVtQSEcOf`BG%?CtdLUa!1q78!&tp=eC83tC9YfNjyp6p#2{H0Y4G
zJKu4{koEFAK_X<ZRp(UPa#F+DT_f42o46O3)<r;V`E0+~V0<2pvTj$O$!<2?PL`w7
zsN>UT;9kkW@aazlgEf%qL+`*?3tFt>7<?b;p89UdmESZM9$B}<4}0#xa~>Roc69}%
zg>o@fT~O74aRFm2ls}j}%PVNk9L06g(o3(xr*8u?<HWOl&Ztl*+O^t}pIs$`37uye
z^u<YptWeJ~h)zA0*d0+8jVz7UF@zSEETEKw=D@7-sn;ow^^>Y&#Z8FIEa}{?QoP;G
zC#%t|ZLJOG!P=gmEGgcjYYxp?4B2h$TZCA17ls(ogV*X831)ec=0h`EYs-<yWuKl_
z4Hho)U(P=HdVBSlR1#(5I6EF8Ukzg(=gmP;tnYpUQ+e*0hP*miRv*hC*8sf1_mM`%
z3YSgzt4DB2FC~U^AmlG)Wo1V`LIw2`)AQoSxI`*T1`42KBB@v;zQ<fyTSw?zY$1{j
zqFtXyQcn-W($bR0U5QMy&EJ<D|A{Y7&o{XrH-~tZIJVt>Asl_LHoj+fmbrX>iTHG%
ztV1|au;BZ2C;<;UYW{?js(CRjE_a3b%QBXgRN4{@M6=TD>=A&x|NaB?N{I@oHiDg=
zaeMvsX<TH(!IpgEN^x6hEyeheAt|577a<&<#^$P|bcAgRG7BL@g8k+cmW*(~I7h4$
zh(05{j;ALSCgcpenPt>|@n79C`y3{(8Y=&TL0!-isw-S*axDL&V7)*MUshwGVx_Ib
zOQ*N&>KsczmlnSBQ1TJe?-t0+wQ_W}Smk{)peCQYVPbBc)mTjx8H~$OYn*d9y*eNE
z(fRnu=Gnu)1=UzBqlPfXGH?8WWvJCWBH9N7;KfRTHo8p#XPa)rt$pVMnM0H-PU_I3
z)jfG#j==5I+%l{=F6;;ti<zJVsY8?_MT4Y;NN+R7m8sq>#^~Lv?HVf2@j^g&Sz(H1
zWZtzScp}iD;O#*4rkVJD!T2+$9lPcxP6IQJQaK(Soc6L}7F^ayJ7Yk<ORo_1?5q$(
z1PW|x2)lnRM6bD}j);X$J>i`DTX)=f;Br7Y%B-aF+`{~M05T4hv~W7Auwd0nfE#<I
zQH+P7rnmbKZ;25N6e45ELHqwO=?BXxRyG-RecUFWuI!Yfx~y~EBaXkOkFh|x#qTq6
zwwqt^99!D^lPbX5jS}-!U9Cfz?Ts=js51FS?Aqh$Nh-`4yPH*@PiV>_*fM|j;K;m=
zYRhjguRM-z&pa`7O}%YV3Tzuarjxn`W=~#@t})t+{>;=)?SIMevTD0Fv<S)J=9Zre
z{6L>w{Nubz&gX6;@AV>W5pvQBG{~Rjzh;L<xI%~ekJQC02bPo?X#*v~XRb#8nQQye
z{yVeW_Giw2Ba+YuQZM4Hpv`}oGpJfeCMTKmX*pmvx%uO294t9;L(0Fk?;x=ZP)>37
zbPOJy#p|Ehe{ov~>9DN?QLVX$>ND3Nl#QVaP>|KgpTLe($i+GEe022PQ1RbRa3G)k
zI{P!g!{LY2@FZ{p@8=@)OV^qu%2tDL??ufOj}e_W0&uXFbb?beYM~_o>y}vPa^XnE
zOd{hnM;lvO;@E5}u63=n^uDpx0pxnu(q7&fLQA2?s!>O?jIgr`wruyg#{GCONWUfM
zcFa_%a-#|sgNkEJ75&2tM#VeGlYPw@3*I=FP>eW3$5>mkOv6|sa?O=VNGWeteNn^M
zk{VJo+mZhmK;G<*i9GW=XN(p4EK>+z9S!*j&qdP_nbFKOUX#qiP2xaR6+~1V*<2p4
zZm4U>#;MFALOpcG?fr#0hLG1T^BoZhui$BYu88j$L8t&VATd%5pYWyAA$Tq%9ROE|
z8vV}!exzW1Z%CnftTU|F4ggkFrHBS9w!$G`7dQd(bi(^QeEXpxV<y{PFz}Tr)ZM)T
z;jXrrbo;Itk0Ab6R4G71g?YO{+0xH7_BNW1N@45pS^-fx>3BNFtxOqOG(O=qEqB?6
zt_A)5Lg3deaF0cR$k6j}^Zk9<hv>J<@bS)-N`Mm^%0t`Gtr6!MnHDRD0@C|`ZGmQs
zVuUa$qxrxzuF3Dynkz-LrWKg`RM!P%JHdD2_R#Td6+t+KYJP9dju+$P51<k$CGjEL
zPnMmL-(3oX+%q|&jR^}MIc3|2cDl5ddSk;$0fhoY6VfOhGNYF-EU03IDoZaSXqI@F
zk^UM9g((R|@gGLMS@(H4zoE((lE}z}o5w|oo4pyOQVUi7W01i@nnP6%Q^D8Z+=v{?
z4VNn@*_7gBj@9!J6?9Z@`LGM)=sKB%0aBOIxgJ=|aA5!;Eo7XZ*tI;fe81GD#lk9R
z2cS@L7%z~pA7P5QM|foH18FtVokB4=TpMN@T3f{vK}b049^#=c!TDy{GN)<^cGu!z
zp17sNWBT52_cV<cfa~S_X>v*m`qzZtqd`%zG52aown9Y}3V|T}-z4DHp2pWw9p4gz
zz$jC!Lv$Rx-?1O`_n{wD(nK)G@Ta-m<vl7#ev`iK84W7-!4hzmEvs+5o6WT&s4#A;
z{v5rB9axwydk1Z#tk$vM>QJsWac+6OAL<BQnq5<g^1DZ7Fl4VbOkLe@SGAE7nIWTN
z=JYM3oG0Eo#UVwA2@~>51SiLP9L0@d{$l{}cOqf<gbXlQK|NChUW$vhwA)L<cAz25
z|5T`s<K$6A^T7B#OBQv84j5fxT?`!YmbQQ>3Ht@HOwI^Ts21p?wC-f|N!oQ7-GoVh
zw+GLVNI_LaL(==P$Z&hfVNmjf&Q1$lPDrAFBo2>NNK2?fpmR!|oTZkSLa1>no-X82
zsFa*90qRCQCvbom%t@>KQD^>yxXwl$9}8Tq@_C|vV$cF*T|7ySUv@U#UT@aqa{2uN
zWy%!lscE@pSrp)ejQObl+7JBBgd>BXmb%gZ$BuE%E5<OSIpxh?bW+#Y#bMS}p=4zx
zt36h1iw!)gu8EaYlkEmjW``H+96~1bP`%&dz?k63Te$!B8IN2@AMkHc=#$NPI;jX0
z+9IK3T#C<Un%NJLpfB~lNnw~ng2v@-VlBC>vp@d|)*;x6DV-+H?{c7>88hMh$x#s#
zwP_|Nivl~m5Ln`)>YbO0`7Y5x;>$`joY*Bk4fFNWhfaE=VP*oPpgyf0Q!<ad80LL`
z@;yC2l~<xrL22dUn+?%VMm|2Gf}S3ka%NUmR|{%!b3Kl<l9fU94K+|IU!mW?Gb&x{
zNABsvzIVwOCuk7e-WtAIzfl6);uadZT+T6$fqCBnNEcZDyXd>p?r~3kJd@*9ea+=N
z`fDGg@i_ELc=Y-d56*ajAQx^}Kf=N#w9@|XmPq`O!JAg2K{W@21g7HOJoX>aJpb#V
zD0=Mw+;a7Q8U+p8|1&h-+`DZq|2K&L_wN7nX5yf`^1nX*_p?ABe^3oL{C4|)FUQ`N
yL5_uN{(tVsL&3vz{eQj<YQ#_4{{OFmZ$FaZ(+%_&5%T{)FDWs3(OO}{!2b))M89VM

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_getting_started_thumb.png b/0.16/_images/sphx_glr_plot_transforms_getting_started_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..e419c5e460d4658e6d8a6f7bf749aca7b1600efc
GIT binary patch
literal 127385
zcmeF2Wm6qN*RF#*Y~0<QjYDwPjmsvu6WoHkySoMV;O_1o+#y&9?tXZl_x%N@&d2j%
zs=9inXR4;VR<Cv6S4Sv=q)`wF5Wjr+f+8y;q59<uMEvK42M_z%<D~o3;|s)>FR~J%
z>h9V9-c}oIzvXVj`3&PVe0y8l<a@b2y4O3L5~^h2s8&z;sg6++jglGy0i^`(t4s$8
z=?*D)tn>vL9y7fc)f5iZl%8reS+4k3GFT%6oWgT4wVvno?EIrgi>G7ml5IHK`phSD
zQUG8?ROibxy}7C59K&~r*K_#t^6o3nd!#*~vs9njRCS<*{bX?e<-ct+=8whXCJ6BV
z-0F<4LK*$vheAwkMgI2@Rw`?#{|q!^w)P+g6UP6y1)e|}Tl4=<-{)X`nI%}$|F)K{
z#p=1q{?qq47&r0%zxw~t50Qm#{iuhl`=42Os>!XQ{J)$AiJ9kt?bj^7q`Z&UBRcxJ
zX_ZuLsngVYYJmUcHj;7zB#{nba`K>Kw50)szPmr<Rdh=She57$K+4s-vExy<^)~I`
z--GGyr7rk{vwuE<)-FAN&fc6Zyw-S!RvE)f-^p^HeZyXcpj^%;1@P7Ov#%(T?cGqT
zpy;VovdN17ywVE3luSpP-{n{NzR1g}q`t=VT=6`PYjs~wE-RF(y{{R(SmT-GdNCgn
zEfHixkweSL2FJsb1NO#B&8za4s5SJSt`OIr`kqFpOkb>b=x6i+zs?K}G;`CiBIzPk
z*u@a<eghxlzHdGJDMDZfrkm+}{^&Uvx|`PNgon6k+ueS1`>^o-c-i*byO`l56Mk8p
zdF1hZaKlG|d%c7c-W7iC`5+c*^C6ueHw4%uv3<cJE`L0??;4*T0)E`0`T46x`rY+B
zebw#gMw0t!3JsqPbp)-7$kcC-IEa}gH{L*~2p^C<|FJ~?J^kT~YHb}|=R3ep`0J?O
z9J7p0Y7DQKB3b>6<{KBC7>i&1bbiyYP81-JE>b2jm@aZUdVr5@*syUJFUZ+W688ng
zrV2%_H^foadzMy3j@%?b8Ic}ZuKYe{o{O+eMFPkAO_2X+?a7Ly*7<%z9M?jBJKPz^
zqRSXtj4VWiJC6~XY(OwjUH^oh9GwX}mJ8MvS!x+iuf>F8_>s@~eKi;jxn}YAU0~RQ
z@cshYyZ_KX%AzlNqAGo+uwk_R|E{gKUu?HLPfrKI1piWy(wwwSdNUmicJ0M>s`~h1
z>Tparb$C9Qh6C%Sx4qKQ2oD=oQm6SSiXY$no=s3u(ej$8zm%P3cDP+!_dFWBvAzw|
z<B(&i%EkNo9JlZ780O6gzeIOxI~ZAnrRjS{$m6=Lw$;jGQ7h>_oX^?5pI>~yy^i=@
zd`RefnAFjc<pnjd5%o^P9Gk`;y|+Dt{-m-zLvh#$0J%l0qPO!GGkteKs0$)YGuU75
zxvsfclHcB43t2w73susn8OEns1g2z|%=zD)&JNxdeE2&)*bDj?`+~nwOn2XwqpmJ&
z-!uAD_0A+DTlhU}%z*1AXI?soyM-KchA)K=JZCyjO1rkJqjz7uw6;Hz2|d=js(m*q
zwjG9dJ%`g1@!s7&yjp!;wmtrC=UIu=GwmNvO~6x0hmEA`03%s|i(Hx63iqQ0<aptf
z@;%%8Id1yWr;#rYq-<Mw4kz>2MHn*Wy)aD|{W8V3uoB@*^42t!KoV8C9zZ`syee5x
zAcnRBj)kkWh%Or$d>CMAIJWaj8OT8A7kKquXqgaZ&NF=2GTZgtUXG$z3&8&RZ^Wi6
zksZexd5JGMxld-6;c{{0Sa1$6;CeZth{vS>KaD4MKA>TXj@No_c&7>_YPsoVfzI_}
z@hg0!O7M&0%3)HMiq$Y?HzK{t_{@mY0hjYdOePiveP<u0(fs%7IgOAe*UEH7yg~q~
z#T#ma(Z40j!zAf{!UFm084D4ry~<k3De*(7#Y7_ukt1mKH;XmCF-Tv(uGH=nU#iqL
zz)nw9HlWfHO?wpDH%bwT>n;CoD&ZEws2zT`4(4O8gdNgJPDBnHSc(^x4+h-A`LLXw
zhjLK?#5~|A@!~5O0Afz9eln@%x}CKX=fzRFTV12)=Nd(kd9A;QigY`!*w;IJaGsFi
zsF1=j?L}ar@a#lbgE*n`e+pk4oURFXH*)+;LQ#=3{H_RApmUa2n4F&Wm&a{s^L%1o
zW;G~Gy36y3Q{yO_<zPxN5QnV-;Zc1>X7?8gJNU?6>kI}gUF5*V#Lr)FbOuqx8|8NH
z?|)Fl%h>nUi5fij&|?MO%=CWcF)Hbwsr<+cFP{s<#vXaSA$=4281sAn*x4~0w?Qol
zM8d_&VPQfHO7`_xZ2I^fvvb?4augpe%1bBTqo)qdaBpD0aeRHvsO5gJn1+7~cPy4f
zL*L~aw{Q|Hkheg38=iz`f}*p``*WKRLlQXcQEFYMU8_Brp-?Iz=kzpBT|l*W`6qnD
zD~>KQv7RY?>hOz;mmWp=XjJytpJZ)GWLT5XLgxC$C}ni3#Mu*5D%oJ*XLL0rO7<9f
z--l8#N}s`nC&xqfg15?0P|C;!;oXJF6l+g9Cr6hx{b7AGpB_RQp}TWYyHbFHjBRI;
zO^nXTQS|I$xjGp64C=WHgz6I4U#LNK(~zN?6VkeE&-6Wa>(n=^*}In)Ks@9iBW)FB
z(GzArEG3>S&d?&T$a|E{*4dZ9u!{`_LkMyL0Qn)VC<1qAAw-ET9*FQQJ4!)Q52Egw
zi^<1A&${(>op8S&|F{ImOV;Hi?OwRzt<JIJU{G+!PXKe&gG=#v9|LJ?W2YP}FGIgk
z>M+44<n>j_QDAUfelVjgtv{|MX1;0&+;`Gv&o20el6_YjnuHNB6b?veJ%pmZIYF0(
z%QHTMP+dCK$*&&f3yBBNhpA_Nbraf+`j%Jgv5UbV_i8@CNcvIhT_oOW9~{S<qwF6%
z>X+-h6W<qg|F8@B?{`((jnsQppI^7O#Q>FDvdjBw&9teBx+6S2+nVws_8MMdDINf~
zK(t)_l$u+=k*8n^FU_w3L{=>VFHZ$i7yY`1_}r+{#4eGt#Zm6~%P#L0b01!%0tmZa
zl?S*Au^MSl*Y!)Oe`i+(+p}ecuh1u}$&h`&*K-!u4DMRN0|dA>>p82+V4<U5VNxdT
z+DUk68Fex(J2^{ZjO_lQe81{>6nx+E%Pqvv{!{BmDrGWp$gSSQ1M6{4*?OhzaBxi{
zN$>IK(0@XU9hE_;6M&Kv+y{@?P8B3yw69m;#kELb<lGuF;K*5?z;41z5MKgRm2E6v
zU0Evt=pEj`B#i|>8OL%!MV}=MYOSo`Ia@0^xI2w@;w9puKN<JU%>jq+PwqTd)a1dM
ze~=`>61Sd$o(&M23lLUo50-(>F8shvP@prt<}F>!;iF~$<_uGm?3i~ZCjD4Ql+Qd!
z4jeaH3f$zSoZwG3ljJ#7IY|#%@ZxoQjn=!=h1%;1)UO%f>AqueyC8DB*b?5Ud0&U|
z`?z34MPu4mnLJzyKRZJ&j3IF($}b?WHNHnfj1HKy1SA*su2Gwk71{9E?ECB!fwDEe
zS5r7oL4x&u-&aA}?+f7?QeGIzlRK)0^hEu@<wjC^W;;UWY{Xh7pqIdG)6Q=4y@Hdx
zNcqK&sn0DH3OJy{4v|sVi~iU-+_DL00+9Vv&EQ*hzn$_4srm8rpkcOik!6@XEiEE%
z;w0tKyxYoy(+I0oQ1pd59<&$y@K}u*&HX{%zGVh{69QSHor1$KxqJB`QDk7A3ici^
zfj6t-fRH?>Wf7NQLdBDCECq88q4!5|ecydOmUI2#wc`1-dPjK8bJ53Bc#2?T^s*#?
zY%tr(B0c=Y)#(T<-8B)@qx=dWAa1(z3N^{+5dYDBfr*-3D9}~0;0V6}a3*A>|J_Gf
z);bJ~I&*q}M-Ls#UnPQFFns9?HSHfRB7x4JIm<F<VTeN>Hc!AXHjiNQ1_+a&)QA~$
zpdtOMICAYqPVsQct1B<3>##($1qkFRh=rGSHIIfAm00PKZ;*jX<<E3wjpt>;&fD_M
zL)-dLgU*bpGj>)J)5D~fuJXnz<@$|nW80s?DELdiOQw`#=dzgZf%M<i<1Vnhq7D{B
zOJ?%1CaMLH#(p?|&t5eZWf-BWoNX^?&g7J*!{Hd;AP@dro~l}WBT1Ar3kZ@bd)b8x
zCv3+LWBkP5^&Fkt<;>)xtLY~AJTfRLN9S1*Fp^WKBdNhS3uhKoR-w+y>yV|)VqYs5
zEL7-9I3nU>-=#?Lhwyj?BC_lGOXjJ~tkNf$pcbj8RAF<_GbUw{u!6}z>_qg1eZYsj
z2Had*k*fXuxO0jX%1GjxPT!a+ub>w+p##60p4e8eni032C&lhJYlFSv7{z9mXxaJj
z`b>w5cih+4x38V=<I9ipX<O8h6w!CNjn4h{(K{(fyWa3jL22iyKZ|)t>h*ekSKW^h
ztsSIHY(v~&iJhznA+vvm@2ATQmC+%}IZ8@H7c&=&m6uI93ncKsCxG0)u4&>~vk_#Y
z2qi$17XU<g0n__e8j@}4*O$9cAf;G5WH{aS#)-ov8uq`5MS96{qC@cL<VA>I^ECSi
z^gLUq$DHwmAez736$SAyX*2vLBPqc&`p1u_UxSx@mt`Omfi*7fD~B!(lqJfi=uwZJ
z!y`^IeXzWDhk9|}Hx~W7120n>TafMK;1X5APoe6M4UhX%3VQx21P{Uy`r*Q_uMp0y
z)%^;RAyWLenWN~*#Fr6-;IL;%n?s9!MrLwijPQLr0qNz#^308_o+~!%)E_Lq;Hyw$
z{CwFu&Ur^H4w4x3{<+h;&OU_T<WjlQzM5dXzEk^QqEs8CQ%QD7gaBt)3oCLmR`h=Q
z>}Fmp%@I82T5~Sh_kUT~VmrYG!7&LoH+GyVx?l4DIe7<{8WIK_<7%D?Ad&<6`Yure
zz8e|XD-(r={@lq+Ky=XB7kMJK05XD-1_+I!%zi07j^(|FD+#rSV8v5K&N+zi=o>~C
zFA?N3-GoraiqeZFd^_LpiB^4aX&0|&B9}LbJ>t5dpR|}IeaT+10Uiy+2~)8m&8FtW
zRPqwTQSYp*Z1id?=FUVI%a*1A*Zl^A-bS!o6-(EPlU&Ob6b18?Nr?A|1|vpjCgjwt
z1}J76r_m|ttb-g;O_(%g>Op9<g5UE)M6+NtQcM&>%KE05RYf-BkMA(&nerFT*i=S{
zNO75PV#i5-ZpP>2FhB3xazAI$l><}_nY9TM%#;pKH4idi2YE;QuF(7jyI65tp1Hud
z%`M2PboJ`_!A3$^4j<=GRf7Gk^19tlA$BbT_n)CPhegsOngc1ZF>2VT=nK{oLiB<>
zQ~{AH?7;=rQf`bwFhIX<FMc|Z!}p0drU+WLgZE7<A{nRx?TD;OP$2W=5Mfjd9zF4&
z1Ihcn+=x5WJd>e?$de+u3f;1$K@2yas!SsbAOG6^R?fFc8rtJMEQ{}+Z^_{(G|~@1
zvpA~NpH)O_QZBGrHfOqKjddZ>Cbj6p(BkCC&A$@p6#S7xZdZC-ws5TlPq~Hn3Y!+a
zBjCqNnqiiXHJ^~^JEf01z5aGx5ztcxbHb<i&qB~}kOf*ZQnM!t<puSEk9Lt3Y>x3P
zzT-5AgOtv{Q1=JcmdCBkZawk`)CX$w?u`P!5jnpsm3BVee{~7NALf&A7}L8H=p+AK
zLD8_qCTpKwl+0Tg0ly`TL*IwTCS{ldjXaEhA(I^>mCGm^@M0trtZsX|qAOM&uHPeF
zM73#$+Xm?<s6il(!|4bGjwr_5hF-Raj^KjTBfMaP5Tm;&TE((3ana*2z(XaO_Hej}
zuRaHU3(5vrm`I-Fk;;$-2ftC}hm@tN%}OxH7lejs3r4T<(<p*!+J;g_3B*9acuomx
zfF=Nj*F?R)H1wkqyJ2ruB$%2NccaTKZJ>}|5+z4gF%{?bMenbUNlpnjwPg8;%g+j;
zmJ24j3K%fW@q>9bh!GTrd;nP$x`U*WTHDh@2EtLWU?Y*^AB*1q2Tv}r2VDEBk`~&T
z36$~Tweq-pB#~~w+yh0jFhqKSG`@}3!%`6vW|>0%<)@8>c@vF$GM3WNNY)5Ijy2Fo
zv8+BpZXcB0lf^2bGMEHUTA1b}>nzLUAK$|wOPgc|%{c@a7ap2k1V8-?ehU*qjN3!V
zu^Vohz~7%@F7jWs85YP>T@IA1sXxz6wjhXa=k~sBfJ=88+?U2Ixq;exS49PpNk;yN
zoDteQ>dVOnnY+JK;y!Nr?YFFK;Wwm<W1vCEW*i~6>lOSo75MSbVc1w45(G=VV+5$<
zq%qn>#K(daqPmq6_rFDIHd0_tx?qsNCND}$yh)I3*m)Zkjw|?OS8~`iQB7$ZEV5^t
zX~fTGY=0q?v_M(ppa4SU6Idzyh4@RaYts6nQ`~R1Fv&GuwRRLY*{tvOHhy6&d4DYW
zgo*a-_Olvjaeo@_C|Zr@JyGEcThe!wW3&u}g;!tsIhY%~y@3yf?7N_|zsqByI6;>L
zq1rn}i;UXG%Qg9H@pR14KNC7DEQv^yE;*SFzfiHy##2NnFFqc?2Su(LFOxdx@S#gd
z4{Y0=&RHiSC7mO+W#OU|3bf6Y=g2+WDt5tZfK-#k??vmZYk;*vR{aJj=|VG6P&Tw6
zdAcc%FkU-0#5TO{i7HIzjEUGEl};Sq%XT+T;li~^cSI>#5z#J)kOKR1t*J0MIm8<8
zic$55ltCnnmwg|ai#V9WW%E$J$$t%3dT%_;G7#Kz^Cj&5=z<t{zl>06{j#icnh}j+
z69uqX(n2cXo~w1OzS&KdBp0J%tUJ50_CFB8L1$Nv)IidFda;{~@r!Ma0Ya-jGQvZb
za8W_j|3IaN_Q-X1X;lI*C38*U#=Zr?e9iJ*ECxrjSbur;`PjJG0P+Qx?8h_E%Klsq
z0YOH=VC`-f_{ZYlA<>q`>U@j-S~8nk*WA=Fv^jze?|n;-A=GvUB=vnI*I60_2dV<!
z-oFG(n<q97oa`k_Hj!xtj%$xojI<)|GyIJoYgwvUs7hida-Dlz)@V0O%`?_j<N>KL
z9PWio=(f9rOwGX_2++o;{*BmeZVeJqoo0$c$R^ThUnROpG~y&7FD@!WycRezWj``?
zde$8|%7B%yp4!@{WU;f)w}rD%a|j7TDg#K6XtE(BB?Z1bfYKEZ3*7tA-Y<^?ylZDF
zYc~EgWtnL+ZK-ijd&boVWMw~q;f2a=y;EwlLgIa)VD{%8^B^y(!Q?@G;Jp=!+E7jG
zV-LAgk4y(~EgM|Z(-co24O?`!WIH@W@<H;tf%l{|X_s_{5_mF}?ixRfkW#<hF^ExX
z%TR!b95X~Yl`2Ih10YF%OFUo~Cm~qCf;ft0Cb%#CasBYLGjid0@K?U`b!|Pm*)4&g
zxB_KbCJ){=*V5#1WAz0*mB&=bKG$y$jIzO2S(Lm(H9fYZoI3-<wG3}>=ZM{xD0+DL
z+YFzcf)<@Y>Il1G)uf7#8V3h~_rRJs`^vmF;=<u)un5A+y4T>=mdURzcYJcnObU2%
zcezIDX}l!!PSgtvcbN%8+wijB`^ZieUvk<9<h0kcc-rb%U}ynT?KE^c<1p;~RD(o)
zwCdW1^i9`YdY6N;CJ75xn-p*Q)ay3R6Ql{~<eq9lb|?rg6E08WZ8TuQ<Mrgl`~97y
zloqC=Bkk#LTDJm`Aa(xs;ovJ-C;3---r8Z~#w#Wn;&7Bc{}sA^3YBq)Iedod4<WqM
z)TR}xg}3e-E^fw4yi^%}dxs+N2wsV=i~vnG{$gfcfsLcb#}#Rxc?xK(5EFZ03+H4&
zX=FS1pSBO58nn$=s#s>JplVU`Tz)+*Wh|6I1eHH116$quk1HyRhPqQOz6v#P6_k9r
z#Ga2l3BQ|nqSoUwra^!@gHv3VSO-%VetOw?(Td;#Ci+8;8cJ7LB1ZnrXggM<({*(E
z2z8Rw3snBeD83uX2yQ+x0;vJ=F8#b+%cL}Y-=L`&q+8R#mD$#DM}xE7@1i(gL%#ur
z*!W<nasH7K0kc?-T~vdi@ev{`imD=ye!W^8i4c*fQA(HB0woVV$QX%WG>f+O9V08$
z6_glp7R0w+u6#&@`OSTGu9qrV9pC<%C9CEt28kE}HH~AC)6d!O(~KoWpNkQ9YjTTs
z!89zv$zn1D-!H#I*JmzRCXYYKm?50H!%Ji}kZ_9yw<B~6+K2S2BiZ@A14+FvO)9=e
zHgJJcYhBHv^e@=EKxv><6=rh<ATswzkrl3W*ItFSI9UBBiYv1+yJ$ovQ)IXldc#xl
zjL-`L8mZs(&?cd7#}j|nx3mO?p}kez_gn%f5AJbGo_1}5Bx*%%0D9BhYGUr`?$9B@
zl=aN=A5L65fgg2<q<fG+5sGXHJb8_&bP$P&UTic*FeaXpZdLElN(KB4VQyZKjND-Z
zJsRm&=`ZVfP+;i<OQEFoSw`9EtIx9qv2GGhsx!$}0xG$}%X32$R}U{VdN>Z_y$Tn6
zTVh9VL*wC7b^#VV0+_a6_D+G)EbD4zR9uab)frWRjciz?`?5zZAuqPz+NMC#to$t4
zZp6VezTqPyf2(v=j28;d>*Y1c@Ak#}Xt{f;i7{<;bJfR~DVFA8mpQG$o!|-aX%H`{
zTrk-xdKiCQOcTZXU_pgZr4|VkhFB#Oz*Eb#T-DS)T}EE|sGG(^mEa(eAVC+V3v7XK
zfg-x8|8t-7ZKK^a?PSB4dJ<PC4GZEemy(Ml)^>R1Z0Mgp(d-#3uMmAaM&jgYW&tN4
zK&EqA6y98IVl20i9|qYHiQ$J0#Ns|?;viwBzHo`qBwzaE7(e8hH+1S<JF*niUYR&n
z>7Z<ws6hz)B7k0$h_NX0_|~&XHZ&4T7NgKL=9l&N>WIB`{WZzc$chR)dz<>Yx~atp
zp4E+Z84S1hVZ-r2W0l#{yND#LC=<3@tSgGJB$I)h4zyO66P6&~yozy%ut<XY<A#G(
zSPAzK3I{5cFqt1$-=ryIzU0*co2eq(lacx?@-{5Evz%J(9q+<l9^%^k9)vD@pKKGE
zAostV&B;@<iS5X88#(fw0Pjn~(aEZguJ@VxJ$f`M4j^q4+erE&b=*r<7v&=ET`O}d
zno7)G++m3HNy{o(UiAsZ!C0@arNVzk{9fUb7lSL)Se<T$xth6K9Ptis=s)v=pf2%t
z@VEnKzl^NXH~t(FvES+%XLVCf0dhoHxZJfbvHOLXkp+X}1g1l~yTB1?v{tWEN_~$v
zy=In8Q|A{*t2LzBSme1dUqk^n&2pBMqU%}0U|ki%)K%;+!ZJ`6OqjDr$$8HDb>Y~&
z;=%36r@*D5*W{r+Jo2ybbJFARryx9Jp0<^a{+t=V@y<-LMfQ5EX`?Fhj~$>#!opdc
zIP~{gPz$K<Z+a@CW!sLA*fuS9mUn<P(|9E3TzayS*13f}-=C=hYCBe#vuu87G5+X<
zKSPiv*8E-2w9l60vlJq6M*$EjPfu0Gnr&3m9tT>>xt@P@FT;r?z`X|nkfTaFe|}Op
zVjqxW!OhmEYyFuU_^+k&htc325x1Fz2^>G=4J}oWglJ)dq?O}w-ZZqdb_AU>kOAx5
z+$|&nR;t#76#cjA?H?`Jx$3W8;0Vxi5>ka;Vfg2$M1amq4*(SeBYv?=LPk7QXE49v
zyXg8Lak~&IsOXjB6}WwxlE>}UQPju7L6YyjZ06#Fv1O)6b@7B`^v+o=xV(sC2RH+8
z{*F63!4C;3i&ForAR~cNQ%0ItLcOJ^tuL94-K>PFw+r(C14-<i&&IhzgV5jJ<9%zC
zS{B61;VqbCF_H!ywWSl8h~udRljDnWf+YoycHBuVkb{cE38b+!LV#@N_D)8f6J#V1
zzNk@_!`W)lA-jj+edc)t+P~Rkylv(6d;8w$TkQR&Lvo0dRhe)&vg-z~lF<KF2+%@a
zWr3jN{G%y~!XW}25R}yl%k=^p@uMWC@?PggertsMqC%C;kU!s8w`fnFCoYybLhzU-
zh;|+A^+p&~$rAW>_cN;we&Oc<cW{s5AUamQ&LPBFfpA+i+uAahWPpW$h9C2(&1IxO
zBYB8;_^3%GV&kus^^S>)w@+#A_L00$ba*$QcZsib#EXPMG;t6JA7*n6C6-X6=NS}`
zr<L$obkH#}g;rESDALM@Cxr(U26_vPWFQ((8K>Ud<mCW--i6Xwws$HGUR>MXn4yWY
zZ)T&MxqcO^L)wK%2F1~2M%<_`!>Z##OkaJyLf$ZAMvWOD*GrN|oEhV4f<XD)+bxKf
zp85r=xt@&WQfAOAW+v?4P*pH)8Xn$(_B}5?Ti!2UU*CFeF0T6W-q2+XUJ+1Oe(~{L
zB};#y*2@yM9$qC7ZsAOyYDk`Ma1+>{A*8175))Ywp-$Oa*%csq)@^i?%$$M$M>ZF1
z%;fxondv)ovl$upzTY@?2y1ecwftE_pY3m>>W}Vgwy&<2b-{OtwH|t1{(omp?voo6
zq$&5W!spy4@9()CiEbXRggjK{Pk-d%#U*1z!OOlKWb+u?h70IAbY{M22V}h^Ix%`F
zc;d~3SZeT^W?fhw_PL4l>O^0WIEaTl#Epb@ofN9fh&xr=Ie?tKy(MzL`h6^E9mfrK
zhYs*bgL$tv2GlfrcSWZ*y=c?dW~_b(P~AwCj+1Z0eT4xAu2SJ0Zqz_|qY@o;23%0r
zc5Rw;Cd{9>u5N6YJ6jXllFN1I_WVe?a+qxG8T<K_Oi|T1g)HH>)QcwxHhMxr1u8KY
z8M?TLTB5Tt>dCM;KZ6kmzERzulHOfBDe-4yWjf^0NDO9*<ZxwnV_pJwg1B%BSrzwG
z%_480z1}j>KVd0Dd}KJRr@&PkQ9X^BO~jc9kzRx-$E^^RI?JSDls`q1F+Zpy8nqdD
z`Kku&n&7S2X!f>(#P88iSKHY{&?#g3`J<c^ce4h@=%{eJc-Cp=#Hv#VDRjHoxj&xR
zTNzam3oJ93(rjnQdw^Lj?w+S$87cY@XGGJ>lDgazYQa-c?HbGK@|E?Tk-Yauw5NX!
z-_$*1KNf|$&he;;Z&vAfhuk}d9N5Sr1r(B9Rg2Uut?9?dz3{B$u;}My*adTxnZM8T
z(hyTm9w9x0;SQKwKVx+}OW7$RSWGCU?QT`c{UoJkmQc4p0!x|fjNP+g+{-Ne=Smhz
zYi5xxMO4GswMN-Atu)G-W!%ZI<I2DB{e|GnRDdWy?vA82aLLF;GGnGZDzMKXspnPe
zz4SIrFZR6!t<C99Pv7rl`EeKRorU3Dt%yNXhYFRN5<2tAMhcv4#LQ3Xx>hWP_)H&e
zz`+h+$;JfiD3Vjb?#EPPXq%vqlO^k^#%*G#eT3}kU~SLj==K+h`Ku?=MS8=90w87D
z{uy81p3&3}KOQ&UyUb~f2p@x@uiQJ~0l%Ph&tk%|AZcfrpfY-w4q<tIbxlrX@p#-V
zYJG0U^QS1I^U=Ew;Y&(Uj|1IUzMJkxe>)+rBE3^imaB-{>StaTy;Zk%3#HM~_4joo
zb|*{oc<Ld8FHhIocbE{uC=ogKV-k83SW7lK+;FP}2h+^iwnDE$MRZ17{`%Q7NtBny
zd_(#H8)be@v&E6-F9g%_#h4A|4|OZz=;a#6-BaRK$xgpr*y%Fsb+Ma)g|l^1bq?_l
zH`YRR3B7OJQn;<WtqNMM_>rPydLuI4=r)QG`84DdB}x*wvbZw-7*oNqf9n>yB%;YE
z`wYtpBNGEdnb!u*VP(a{s*H=&k)Opscj~cvWIa}1dhakOTSM-UCmRyWXKb-vZ|?kG
z2Js<_kipQ1wwJ$x{=y1S|JFfCxRbrB)lSWlcWM<rm>Uq+_j>a*WL}MrzsZ|3JQ71x
zx*A+Y<FTx4AemM^s;mfV6s*rpT<6WB^ZOCYiAUnXrk|*d*m%dbU@T<ecY*#H^Cw>}
zKBANMv)dcJU$1<mf>7<d9aDB|gRmm|3@0r)uYW6^U?ZpmHide%^cZqy&7HX5p@#iw
zfJs_V!}(17pMneMM0yX0Ki2|k!z!|WJslrkj)S+8{|_!u|7S+AY~MK)leQ&V0U8aZ
z%TD24wdtolii`*7O^<<p=x`Xs0PCiiD0@WWsFiEMDh%>vQStD1KIM_)YM~+6t!w_+
zbc_)a3c+z$`@H#@T#)<N0c>;-lDQ3|N96tR3Uxo-0T!p%L;;%Xc;h10uBdT-^`_rs
zpKgakdy1a4yK4gU0y!3X|0<rQ(M8T0aUh;NB=ExFGTZmk%1x-MVfmVT$4t)B69X@W
zZjg$MMqt4l^RX3kk#wnqwQPCh2tVXp+~gR1kl1%W;e5qqQMQ^g&Q(jn4%!U|9qHbp
z8_KSqFI;9<Td$v1O$Uf%2PBw<UrYX>-x5NjUh{zL;m(H6w&EJU-WNnKl#OJhheDe0
zR5IXY!U+<6uBhdn$AQObpbvwCgNGnopC$JHk!yyHL}v{!=WAhM%HV`ue&EzA>|=Tr
zM%MRp3qZ^sE=A-x*2B<kTRXUJLh8J_m+!p2cKxN_F?>Ib`nDBc-3K#Yw*M8jHN2ZB
zF(E%(=H%yS|B$grLS3)wPniL@VMga9baK2zGPeAw=m8bVsv5<wOJPjo_z3_TQxl97
ze`sox?05;;{-t~FPtl?r$5x0r)(9B3IvH;kT9BSZ1!VctlG_5zwGxrg{yojjYftxx
zp5LOO3?BYowJ6iZ#0wtUDou*?a(e(XAr1#M-i5A5I^qAt(^V$Pzas8SvK5feQz#s|
zCZB!NT;Z9Z+GhVFG|{L-112=vt7$OxlQr_WC|Qjz$OdL9t$H<%c=J=sQfU&Rt&_1y
zIk<2h+`7n>VkQ+Mc;qHgbr;%bXqbYf+<2*ERiy?Xz?sU+TzTHF=ocTi;)gf*Z@yt5
zS~D-NAUe%S+)|nHW0&@9BBjrW{I&CSoYCgoMs*SUW|bT+N|0o$2X6ZM`qW^Eun&IQ
z-i8lN>5o*=j0u;s$qao6Y<u92S?Dvu5wy;-L_$B<;M?v`GaTj&yPfRYF(Ey*Wo%UO
zoSDa<o{tCR^1YoapYap=FfTOlWa(hULp%|eDt{6CAVzk1d1NA8>=^%NPL43V1dnYb
zwpC%I+%%@sw$?JyaMW0*gx7xPTb%p}IgLWS*4Ib>5bUHiOA7kdUo^ayD)d%WRD`K-
zwVaaAB+yu($d=eIuI@?gV>~cl!|*Yq@hAcRenGfS*m<**_7wBEV7H&|TiQCkGU#H<
zn~dV1qBBBh@-3qBn1q%sat1h23+}2IHFq4yy$k1lkPza5pW5+eaCG@8C(4${i%U7a
z)p5$Ha;jusCdgV#c*l;(#?MnEvr;rpt9%W@=oV6dmz9vfphXTd;n!Q)Oexaq8yln~
z1n(WylzCfYnusowR77IAbO<e*GAY+hl2XUp1eU@4aw*F%=qPu;yR+?nJ-6Oy^^M`t
z_lrU5=(vRy_v}I^o-3XBA{h*b)TSC7%%@YGRsZKDSaG@dg3yhiFbUolb<dG7DLbX-
z5H$Pigb1`ynW7)h9x@hF`J>KHtBCQOBebb6l{l;}8NwCo>(^_^OEp1)NpCNbd>uQw
zn?O0isiN+ANZEaP>v6rINn;C5L9w@jE=dj-Sa3&M)wi}bfX2>`6!eqLb+D_$O96Bg
z4L7ipAY-F8vXe^w%a!b(Q(+Tf7eYChp$+y2STUFXNML<}nLO4!zzZB=lX?fm%{HD>
z33LSseLb-JK=%rZtA(tmGO@On5pqL@BbM5Sjy7DmaC#dyGjHJYibMGnf?7TAC^R%2
zJw5-@^2X9fLE;kJBkZPwK;X%v&Zk3bZEN>e_qZo-Iem2($zh7TL@N5yQs;_#bN-y`
zxiHe3wQsEGvO_jPl6aB{N8XoPqz!FeAyY9?wn9bugn4ensW)x6WLnOmN92(uU*eRX
zP65ANS6-+6Lb+fW#m~q2i^50e(<Q;Gsb(^sgf3k>_ea7*@bVf&SI6(tzwz!LNIaT#
z9b`<QBOC^?dB3SGb9KG@_gRj*a2L;0TM>E089cvhD*EPEGsE-2Gm5NGvo(3fRnCPW
zhe*$CGt9()0T0PaRE{d3Q2bVf$FCULWWTtLM#3)~n^>bA(@=Kk3#(vI?*0d5Vo9H$
z0-@)h)hD;kTF+aUb~ieYTuc!SWyB<@@8s=0zPn4lrkyu>Msw!-$t4oTG!rm-2PrU3
zP<h1DMTSd8Vtp?^);K<*mi(eN&BuC{&03fSJKPMN7=-i(F7xj1bR&<`bTc#*;eEYR
zcaWyUq+w$p<EQFLEBy9Z8H?E`8=8T@-^T@x@pfavxwmd&sZAro$|li;dg`vdHAwY+
z3TKvnz|kUqTmN<&CSLCwcjb;r?rpm~HWpmKp4?WH)vK^bM=x6#Y_~y;j-b~rG5A=z
z+<Q$9#%w2jkt~2%_*;_6=M~klk>fMI`v*7;-+l%AX4-36+joBG`n(Q>6`S#O*RXwr
zGq2XI7OsFsp8an-)l~bZ(<+|nyH|M`Vi8@OQ5tqZY#Gc{)>urZw{Ct2zDDvg;@Wji
z^2c$vkNSeoYZ1hAOl!2b02R>yL?b0}aB%1?-n7L?36;B1b}<+281{cG(h<iOI3-@t
zZ`pjz6<=${u>V%T!;BT!sL?3ajIx``L~?P4UpeR6cl*a*`k{MgE@)jFA!~iTNLGT1
zx|EjMaAoC~jM)2Nv*mGqy(XyK{c?vSEKFLVlw+Q=c>7}E#D%RLHVZ(1f?csxBs(e3
zQ?qb-S5_Gr69a5mB{F$+haG@;d3nK-l+{SJu|ws{ndDs&-g|DE_$}EtH-?~b@VJ`y
zmVQ-C_(@^*1Ub}K*WNq3-g|K8?tE8tSs1L#h+ACxXEg3@-Y8oVYU&DQjc1K+K4ut|
zlng43%AC<=!s&+jit{hKU7%MkY=PGMYRGHD<28Oi$){jPeWw$SR5n2MeN9dlHVDFM
zsSiOGygVoi*?**~(6Q9NrVXL8Iljg=T-CT~z<#BP|NW@4+7H{2`Pc8?04{`g|C^6M
z`EMp=<B{$2cn#4)jdn|wAKgGQkh`T6VJ)>qN^XdE`_AKKd{@`VzY9{p5dGVhf5nnZ
zyWS6ViI!zb=<jEj(H!YIFUR7K7PK}=G7l8NB2tLdSaK{!JrRSZ{tvn4B5O}h-89Ij
zn1_`$uDch^rZghF+r0}RHSh*J4WGJ(xmF9$&e|QJ|DW@>#H}hRlX7cmV3<tkuXabn
z>8D1#Apy2ZP1Wk1<qvnN)>aO~&P7Q&sBz$o#biZb!F+NQj3btcMk?=FQjv4}n-GI4
zo>n2&Z99I~dHlpC<L8_Qr$>{Jc*i7|KKd{^B=-^pl@yttn=AZ7h>PlA8dqzE6K9F3
zsY}r>ktUVI%}c%tEuVocyrLOu-Enw9A9)hTiGq>Th^L|dipzcCf@NkD?&T|Z=b6iv
z-(=Q2`=`yk`x5K;2tVRTnru3uggR&|{Am#P^2=^CyupA`9eF^dBa0Mb2zLH@X;S#8
zlS|{y`zGd=d!d9RG&Qcn=FjNh8%gu-nH2ldElUy6Uf4k=bZ(8selBE6^vpUgd51Rb
zvkZqp4hb?kVQN9!#_^2obH~#Lv{f$$Tx%3Cj|ot4=nu7QwkoSrJ@4*y2nrw+30Dg5
zTYVd!IWLKFD)z!^wv1eR7t5wy*nSe``Bu-x)1vDpLnB=zui&R4s+QegY9*FTg&Zb+
z*nofY6nNqyr$33V`U$B*S#k9vJa;LQ2nv<VW3OSqUZMHswQPH4$!l%jR+m@b!Dvg(
zv6t91U`ig_bx;vMmQ8<&NknNhhp1D}Z86L8D7G!P&*@vkFtP4Q&K4V{X9ocou3>8#
zJ}CuE89^ijw+WYLz+sy(mIQ-=Fd1~hB?^&AF%Xnk<Wp}q;(VhaQO#I25zmfDY*Q#T
zVhHtnwJ`~98?vxjpPU~*-2}DjG9(r;B8k7YG<SRaZF0#nF{RphzL!(pU79|mDOD${
zHaBhjzB5^4jKV;9ahTShums!pWCTcMC1S%^HgI1{)2pQ<I$R)=P(rd~<(WzCIOs5o
z#)prF#@ywFOC}uDOi*NNbO8Iu)-4SXAh|U=kVWxZgaM!=<wD}9|APx47E-3bVq~4l
z)Y}|U=FmF(Bp3YAyYSBum62JKpa|enFSK0KBT@SG@aCIzxEJi=gSxKT3xz0pK?mJ^
z8=9Ip*9c+UO%TYiGll(_7Eg^m50rig^R`P<({;mQ;CrN}-|mtjPndQ>0|h|P^jDAU
z$DaI)8R-=2h>aw;Tjz*8=%0H$S4?_MuBW<2-~-K-85syus=!wtW|_r7<hHW<@9D^h
z6YfS@U@4VOE}{^X;qR;pJ`gcV0%I=fMY}0r35VVsK;PWlj2|QYPf&<L)`wXpTLO8Q
zengsYq!d<fzy-FnwDv9A=aVjTx><zixyo5~G#|`+9}5y`!jeH3^{-PWs&jTkGE)P^
zqD)isR*Di7-E!eb+O@eo_>b9u<`c+NYvs)@mUFb2rZX<>7b($g2byoRn;fP)Os2+v
zOc0I!?MJDRg3pWQ=sQIR8CC@DcfD+MzX84CF6y-vuFeXrOG;_!QXKaGqHvE;{xWC_
zTp~2JLSFLZ{SMs!o5F}JMd;E_c>#kIrJfj-#@e8#^~uy1@S3w{IJmoqH^l0fjyTLp
zOGVn%NUx1~gU0&YJNr)lqz@GOK%<Mv)N5bwY!Ne@r5seH&l0-uZFCHvy&ni4^xWBZ
zN2&*ipL>UNTh!-lVo!Oyhr<y3Es94pp<aXb*Ms*H`5k%nMJyP;A3wVs*xzBRIy_1z
zSuwhQHq4gAD)vjlSTBI9JKxaET$!<og<KY(9p6n}lDzd*oyC}UjUtxAfm+2fMlTs=
z({7vN*c6j|UvPDis_L@HKHpm&vp<8{)*9<1B{^PYwD8plb`YT^IQctSFPj#*M@`Yl
zZ<t(-!_%1}t$P&Lzq%5t%P=0mn_|^^@W>RwgWlgHUMk<F9s}yv8U7=it}y?pYIGih
z?lJ3{uIa_fG@rfOw6PmbDG6%{Qy(?C(~DyKgf9}uPG^J;k;PYJRoJ3P^1t3~@tHl-
z*J$ep@--B>I^bO?^0`olrQ2euiRg=Z!P~<!VV!BJkoj0?zjXJX@3Oq_=g>V53pv%E
z$ng}*1fdp@W@CJ>Jvb@82KEhy)1ChW1E}e>cz(1#T@!jgZQ5TxGuqd6)G4x_1g2P8
zl;`BL!3^0**?woJhT_y_%bH1d0j-kQsg%KFj9SWJ8heUfSQ26LVV0c7&x0JM+4IsA
zK(s`8+3do(PMvO(b92x~RI=Fx1?f3CNH)<tt?nljS?h$i9SCdBke77Ve6=xG;S*`Y
z*+itf(L6;_<MYml<R&F*;tAq>56~#}7OZKXfCO*EW97&6L!0Lv;>UHKw`es|(f1Xs
z?u$}PM0%p-UwWq2onuPAE>U9<O2Mc+YVA4E3;U5zRfq3yf}l7H8+(+rufCEst27rz
zwq7E?zn??SWm~#y;tN$*4|#y+5m2TP*tWzWmDIMpT*8!uUg=+69{dyASyaRv+<Axa
zT0Ng&yY$2CXhkDV{Uw*NZ=Rx8U$NA_3BCMr_UwYtA#5x^F{Mvqz1WKuv!1f#q)=4*
z;pW!%G_Qzr1Jgc#yoQ@Sb%+Ey%5}w-MDS3VC5;WgM@?5a#AD>uR-liJyc&k(G4Ms$
zmji)7{^<{X_iEiYn=iWE&xlRVqDIU~>l2uWf_L-laCKr{dIXsq8qbW*8?V(vC0Ru3
zk8Sj$P@+WCrdCn|D+}-v*$JljpTIDkD3)tJZ_zKL=*oooWv|iE2k~_x`-X;w)BNnU
zp9OeCe<VRQ%hGb#WMd6&SwEtGU1<+MeESuU39TsUg&8vB7#;Z=^Yqk;BJ+X%<RSRx
zR8}@15pwf~Bv_?&%6qlcx}t)K?wZ2IU7tNTv!MUp$;dtOG84n#rjFf)8YZaEKc<>_
zIE&DwV^u9?1oMXi6Xtt&Ga|u#j^&i7dpUBkx3P!sScY{=?b5svUqfELp0nc5-N41$
zqcYB8XkKf;@s}{q1UcgrG58>L3^Q!8uj<)#P0k6dekaVnw}rwcB>3%6_Lq*`seD+2
zn6&X!zMOUb*Yt5hvSd>j01Rr(W@&MjGyTKA2bH(;BzK*AyT);2WAqI}(W}Owaz^vJ
z>`&gAiJBFuKs(We((?^<6C|a96%9ujnuG3`{8J{vK_pmQO)a0k?zeVn#uToA_icP(
zklNuwpjXTJ<;{cd@Y`27ytAC4sE)5THulOLnbn)0viuFr_EVF6(lF3fFVR&sdkGoV
zhsB_~7uR4zDlN1=aKR*qi=%$>MavcC_ZHYAS76<LFtTOaBO~k~?s4ahmF@<5F0%kW
zT##c#eYExP%+1Y_$^E12t8RP!K$g9ph0x4Xp`T;cllt}J6?z;AvD4iei&bZ2UrlZ3
zKk(9Q{Z9w#)hA?daZzrKnPnFJp14s2T22VtKjH3yFyRli@uy&>lw>o-Zr&0WMj{|#
z7H&AD_M}odl5D07PE=xh1@FqbiTy{NeVD+pvy*ppVxnD*IFWp8Eb7>NFsMdaLxTwi
zsm3rw%ViBOvVwy3dt?O_2ond%Pe(gbDpy6tnkESb&7u?@-D`*y#fqIKrbK}TZZIc&
z;BDmfwJYxv;YOh-)^vD0xN4q#4uGit4hlzxmCWTLo8oW~rEi%r#nz!RT*OfS6C<rh
z7_98#<x}4n1OSSmwmIJDadZE8+!tmZ^Tfw(G1x6cqs=C|2one<4T>a^#kpELwRbpJ
z(;y9?2&b4KV2H%9CYHr$?2p{F?v^9>wkDvx)Q++xhs<PwMTsaFLwQ5w1E6R6a}T9_
zhA}HA`bu?!WfAL{T1C`!*27N`oM12iF3tj>4IL3toJ3-$XPHC5q-+x|-V9iH75@pr
zCDOdp9bwO{)%mgQ?hZih(H;IxAr2it)zo&etdCG1oo(xDF;EMDI2d}!_|AVJL|>Uh
zGQD*&?=HZMWqm_R&Pu=(Zm|Jg_$6p9f;3luUW5#*Bwn+A)Z%-?^R#zA<8jfI2cSl+
z&ZiTU9SEW0i3qWmEXX_KbZHN?8~Juwl>{V;KKb!tS;)u7zqeKWveEMvKO@vN<P9uc
z;)=VE17lGaB8hLg-(cHqw2Ce%&X&S}d1zuO?nwj=5ARM_^yJ+(C=0~rRlNjlR}JYZ
z0tX5yNrXH{_q%*5<{cbw*!L=SlF)>WE=0$smbw4)k;9m**P)(U+^4y`Hw59wBtKAH
zdIKNP<$O$55m=6Z_PD-e{nY7gjZ0|jIX&o&5(kM6Nm~I0JR(g6nfjV>G}>U*DVBJ|
zb@6tqMk1UIaNHdD;tnq#0Ub4%nJpk}<oxxZW7{t(&0zg-J_QM$;(hGm1OLVYf)1o~
zxY5xu(qZ76qg{L^rlV&XGtyB}^-6=qEOE)Hb6#P@x9wSY)g1y@*};!=;GObz>JJls
z0S+0<)nqw3d5R^>&^Z|dv^BW~@ZO)TS}ds9=&53HG%m?NmzIVxnU@MQ^9}!dOdXRP
zyKOWDrj*Gfx^i#_^f|~RQsJ2>tLD#{g45<KlZuj5HL;~fMn-ve=Yl#my1gVcQq-Hw
zxf%I*t7@0xOJdJ|ezIIL3C7<)>q;T9!On*y;oar#H@#=Vj>nsdTm>A=ibWqIEhvPQ
z_$<q5X{*?K6fUw#saDyaVQqG!n7-mPRE~kzLb{;6XkJh!k31Ge31T*tyeBdo;h<xh
zjF}ABztUT!Eb<ubx$~`<wZVCpLM2f(3j^KL4O>!FLxU>9K>Fwu91XPncAElA|43q7
z@dh3@qdhVR8D5`OS${KJ<(d3IdMZ9F(^?+3R_e-ijP0bO`%B@zMN)`XW+y{YBS9m2
z!H;rJ<V{U^`~$Onlf^2NJo!6V!PPcH(fNFP{b<Jb8iLg45jfqy+;fk;^)#M^?8aVI
z7O@9Ui7XZ}?n$_oRfV_GY<S5>V4lLfN;Fri!`S#6V#1p9<hevr0h5iQUv#POk>`i(
zn%Tz}a^j`$#?-A>^YY$rrw02sGr~Wa5Y?e`M6{T#j~}25jTAHxK_Izti{BH)o8i1D
zuv8N=^TG8Myg8t@yCEpw_XVZ&9Hv&!16YH%@#VIg9g<J{E&VMFDbX%&;Ei(2<7J}r
zdQCoi5-md^E<HCpz7l_vlQPoyd6QfvF(DBn$*_D5k3mZ<3o(F9QLUZ|M(yTsibvK!
zo>#zqvie(GnTx3X?9tz*%Fg${)@9oMzi(lQ+jCgtWck6Ns*#}5AQq(NmgbeUjh(px
z$~`<k8%LBRZy2J)&Q=I35>KLHD%n65dEK9$bok_E(Y1+<S!@J%G7Bsn9k`!%+Qe_y
zp*Y`17}6d;5e@$2$ntSMH@oel$3MZGQ~ltL-9DaB(a3v|>`;wGrdAE8`ZbA**JUY!
zns0Kex%WreH?Oy|uJ?U!9YmyOKoT_is7MCZy5WRjgotZFE9g`zRjLHz)MYAJ38J96
z(79IW%>Jvf|M)KB#Zk-n$gpKHb7w!46SR~yG*qx)tTHKre^xLY-tftlNI3ReHukwY
z2YGbwEa4ik<tg%Je2O7{@zHX|^|6J&L$~Pig%gSLgnn{kgQeNWs7ro*8ice&r2iI(
z4KBac=42hJ3tXtmSZR_~55u=|Tv#zyZmOO&C7+~F^dIIy)^pyth$7dkQaOrcni_Gj
zJnlv1IURw-y4@A3<I+i#(z6x9_gUn$vnL|d1Ct}GV{AXoCKUcn5*F3!JuQZU3WiTJ
zSo<g^hBjXT-4G}5<N9dn8+1u=6r_aW*(fU|g^Q3fBK2F_Re}ybUpLO4Zn7=m_m`wR
zmE*`U7pVl?;At`oMa7j_MfCie<I0h{8Mn+@vJj+r-9YhLI`7@n!Vy~U`x9cJms@VH
z2a<*!w_X?aVr{0;Q$Bm&12gXCXNe4GY!4_3;(mWqk4rTUxc&%QikKUNtc!J@$u^nE
z>fa6_f(%ePvo5heEDXnQniR>KSYCIHRiEzJgWkifm+g#g^1RRx_kJIJt-7l=Y;K0`
zvJ0JqRZCHtrt}Xyrz4kzR;fCUlPcOo9p;UVReIbRh;#4s8$3YLdWyO5O>5cqjq-a6
zBE59~@L%(mTasN_TiYeP@D0+k>+YYq=nSh31tqd&5OL%v^6?893vB)D*eorHD;(bW
zMuhO1+g^M2lCrqw{wSCb4$q?BKI+!}!j-JBf@gnoxNZ+*J^DP~=zGfJ7iGsv9gFQ)
z^?4MSb39x*`6<w@MZ9LYv_c?83z02==KSvu9r5>@)$0wr&V7}<i?_BKHoo$w1z{lx
z3^rTZ2mmj_a*f`fMV?=Qv%E-~Tj8s><Y?!l4o<orlr+Rr(gL5g!PY)zb2CrBG!N43
zMR;>r8DH1bT5pw(IhG{v1;MLg)V<eSDh<*Rd>CY?bFHN1vaNTnBzz-qqsI$L3%iwH
z6j3&@WYRe-@-tMSYJUAp(SIuF)H9h_JNpCPqhl!^m<sj%dUHDR)YP5#<cx4aE2i<|
zq>-g5_jpT~<l8q%z^TBT$?(G9dM1bFiLF97EHhBcdSRaH>L()cNv#Ksd?p!6V>f|z
ztmJcsuCm>{2QCz{d%^m`W(z18O6IpOh|(cgI2Ib|+wnU#wU$RbdM4RZh`;_Fxs_uT
z-M+CiDK?A^PuMHJl{~L;WJ(ZW(uXw=&@!{6;!%4iWYYsZV^eU9h!$Uib#N?`lFEL!
zY@z~E0kLz2NppwT>I{9~DR5s-;mcykSsC+kQz>q<*qMpUvfyR8vx(RHJ*IbO4yDMl
z{sclK3sjQjQL6e!QWTUJDz}>XDnCj^sqjK9?0yTh)H87zxDjTQRAI|foGZ3_&=0~2
zb?#Verv5(w+(0A05F(TcswrpbyDTG=Bu$8ugt|8DJv!v}?K}MBC+~A~G-c2q^VM&>
zORv*G*ELBJGu|4|>13f8(BGeo0z@T2x_IgIg*77GStAKcV{u*_;VhbS%HKKXJ2%Va
zSs+q~r2!3ios!L&A->*bRaeN6i}IZxl#fE<jP0GvM9F~JzOPId5<~(?AF{MVr_<-)
z@R)qTIg9Nj9>91zMn!_*c*v#8mx<DZ<D(PidCB(nCf!b--HX@BMjd9=f~`y2y!^&Z
z(q2q&H0Enx|0+8dw@I>ytCz2O;)QbTJ-*NB=`s0y!SUpn&CwQhTvAigoD#2uB2fuS
zMaV6QjAN=oQ@8@_BEJ6Zuk)>M{Tjn$gEZ<85tGFU_M)L=j&J_#51n%XN+q!*aYPiw
zbh}*ylKI>cnnavkPdT^(=sh_%jXe825M#OX#Un0X-R0}we3j!vfsiL{c&S{3MNyTS
zlhaeCvqOf1K3P}N?I~`)vdQJ^Jsx~<08SxMoE#TCdUV9AuU~1~Of*%)wqZrBEJVBF
z%?LpVgI6lgSyr^$mCEvF5qkwlh>Mq}2T18CtDNJ*J)Et5dwB_I8Z#b_u}*sqcTpj6
z93Adsjb&pnVBAkg(v(h?;*3Bz#dxrV-qq;3X4o5GjG?Y;gb_p{W<1_xGM#aBa7?e)
zr_=2)7z~-tCcgT(EZG0_Gy203Wmz)pciGw6<g?Fiq3hZ+q_89^Cd+!HvE=aJh&WDA
zD!^h&QdA2Ljvn)hlHi*BfM~Wa4W+l;o1w3iuMTbu`7*KWy~5>tmYctU6}>@kYJ(bM
zZ^NmDPFjR=ZR22s=k6<+;#`fuB88_9dl<|YqbVgx;tZiOw5vgDY-KUF#yLkFsHP&&
zM~#D~<vsxyH=%%H6-J`L-32YgC8P{a&e^Epc_M^w=e~rhLXK!93Ti8>LsD34jW!13
zZ2MFismsbI(2eoN-n3~NUqWNgMjzIMis_uTZb1m>^dbhGgpJJ+{eGV)j{G?x^PC6w
zAM@E~pV94(_}bTB;XA+nEr!F?Lsa<ZW3g1WLQ~*m-ExVA1>xkGQA^RZZ&~IYSO2X_
zLugobMra_daDt)J;rc6xLU<+tT%|=An<#N*k1N#_kb=O5pCfiIT;}DMuJhA>ehZsJ
zQzE6t(Cc-GqlhxEeX-`UZCNX67ly-#D3<gFz8PxT>7b3Ju6?ONw^x((QjDxAY|eOR
z#8<xYF5mj~Z?e6!!<9>y8IA|kqGHe=5vK`7K4p4(#B4S}8;#bQ(w0<|-dV$iV;e%(
z2pmxq`BH$`_fm<KB1wAmqY*polF^}QabJ>CkPs1yq=liPrl3IJb#4^L3<f>2Ofg^h
zw)deR-fu^PCuh8961}c!{%`-+|AJRuz0RON!Xc26Z^iEO5)sbE+`o6spZxGcw#EZ4
zUf3lsCO8Z)y|T;qzOVU%g<<dE6e)d+;0N~~kuTmL8)pGKUKVmat=&Wu`U>G%cGhYu
z?`FZBucTfrL|noK%dcfsl9xG&&`hRBL`jG7c*MqdKvfmgTI1q^Vm9IQ_!Mg_DF!p{
zF)OB&m7yvPQIw!-%^(|5sS;-dS=uF123<Syd5JN{^an%YEMsyq;pFI)!JtpnNjaIE
zVyz=DYo^74KZB(_MkvW}m~k?h(C_p}lMWYmE^_J87N^q#zPSCl$M2=zkUJMPsVc{N
zKm9Yl_Kn{}L<(oT_uFnh=Z$dVy*XzwI@CrqQQ#8_V1<#jlB+r+Elv%UbJW4CD<D>0
z(OM^b?WvRqr4Y_HffS+)*OX@zkPPRFxS%=P!<dB6wJ5QwaL%&t&aRE8L!83@5@=3U
zT(<cCr9_BUmw6AGU{Hg>YI885x0G?1y2h;bBf%Sm`C>ty&wcI1`9X%Hz)C*ZSEv`@
zavBCPyH~dO(mQW*VRJ;9b+~f%DoN@oHF2EM?e^K&*kXHUhqvB-jonMbr+{gYx`c%l
z)YkfF5!tpBI#cX-4%&P&>Ua{u^7GR8BP)9xU+sSW<WND#!L(=GpY31P<nW|YAx8t&
z8V}bB!S3!ge&;X$H~5o3{+K&|_OOjMyBT!*3C37v3*!w)6o%Xci4xL2bbFHiAY)@=
zOn)$7V{4P)Xhg5qXLdRRC&>B(E?>XM_Vy0FPOl~IWJyLR$%sS*PPUzmC#MHgbqzw$
z?Q|Io`lv`!QlY7dh<)u<FbNh*hc4ZCfQ}s|u>O2Al#~Sv42Fn=1Vutjhb<%yO-W6K
zqz=s4Cg-Q<^#}9^5tCVsilI*Mn>G*sLZ8;>$ukXzwea~TC5L;Dh@!_x(c{%O2;r+j
z$r$M}N?maG&K~#fobvKZhR;9wg8ls?wzqca_eNa3dX<x-&ne3aUBkl%hs>rWTboOR
z(L^DumBUZcb(eU(Y<eG^OMb4>O;=lVG(R^9daqy{)>h2&V@{3>#+y4_ys(Li6rI$!
zTshu<#N(4A9zNV-JQ_3T_lT9Gn3qgWe7`C!Vv@9jlrcv5&b6lY986=aZvmm<Xzvhf
zH7b(ms^sD1h>eRo<cl0*9d&J(=RQfl|L7B<D8^WS&c|f3U^I-;T66E-U5aw*j|+`_
z!QybVK^iF@KHBHA&p%`TaF3mf*OxR%ya9GW3H$ywjbZkYhwm$8o**F5Mk(jpfGnN*
z#&LHJG_w6%YyF9B+9o(s9EmR)4Al`zCWxjLMo_+I0P@Dc5CMWh5Jxc^8)LGx(>_-z
zy|>mRXO}RXz?^eV8fl9B8imyKb^2w*HP{kuAjYb5vbv5}>C*%X^=b&BjixB4)Ovw4
zy1nl<o2<Y^4Fs`-wM>Qg5CNeCJC`o<*4tm=!sZ5160@<j?TbL2CF!PIxpoz#U}Iy0
zE0?#QYCOOv%Oja!gzvgwUFeW)-LjN0%(Jsi3l6vZU4y+XZFg8_fe{!A1|>4oAvF$w
zuL1H%^zw*>=Dkv?a85^<;}s}ywS7%f1usyfS;jZN`LFQXfBC2U?Fau)rjK(_pb{`n
zQ`Q#koL7<^d?EKd+)@njAR^5=q-lpN$w*a7FYVGvx<oRf+N_8~M0YS^=i)BiEF+7u
z77C5<u$#gcIZI-Ma#-o8Z0)-?X9+TrL_z_E1q)L8q0_aWLQRb#Vk|c3vVjzVJkoTW
z2v0_$h<x|0@Xn?VNfM(FC@NVaq6isBq@9$s6BDO7s%ih<<bwPKiO%w!o{(SA$|M6f
zpsIYEc2D_i*d_|;{5joPYfg?z_8w08qaXg9_kVhies6=(IHR6}(H4O}`}pwSl+)7%
z);gj{wF~9E2YH_UcRnf+tc1Jg5~t^>yv^@gHyMcI2o(vArpIK7LP^Ku^cbZgE?(GW
zFc|Xq!95O-CmbB+^aq=4Y;DsUjp+48*vc_qOvp=rj&Yjyv5}yd7Fb;&g&;{1oT*t%
zPpRvg`F!GAQkasWDp-_LM&m7J(;3z|jt`G1tAfJ=-{`Ka4OOK<NQz>?ty`a?O+lIo
zE?n^Sp=p}pgk(0KvRKTp;)vO7%GS28rvXAIl7<SotVIx_X-ne}&S9NCGtVXz-Z`^8
zinTHJ%}h3W&V~O;2gV4hyyEBYf671kXFuWby#p=|1l=?yQj$uSl;wgXUTr+mJbPLX
zUNlOn{S1GAm`|QByBP?vWSKQa0ci6g&AmoZ^i(IKiE4k@2&6Hf%hd@>-Rj!oUj0$8
zCKRm(6wjTFhN?OzFOPAyz|*Q|p{sUU8VIm~v^3a)<z`ocD2s`^DYDZ=%E%W*T8j{d
z&LHD*c7dIXn+$q=`rT(&h6x}R5w(jk&U?xJ=;+mU_bxOQZLkjiPO&l%VPoUmSqgZY
z7m_aIyJ(G1)5;^9Uff)wy}mjVMK-6IE~8LiohCs_gh~)1<<g~>_&5K`U*+-RNBrIY
z%kP1N%EEL$$67-jq7NB^t3p&KfeeyPOp*-{BJ~l3u@v)yC`ssdhKRnP-|dm51G=37
zsY-~2@5v=_sFnFL0hX$%DNKnG4%P9DMP1iuTTw~L0)-`M-z&p|FDc2X=#r2!4&<nk
zB}5$zar>DXsF9>7Vl;x9%EQb`VsOYv(Hr;aO=e6hOBv=O>Or<48Fse0eqB%<J$~kT
zdG2{2R?aal^X?B8<C=;kiBx`qoYrVvaCo%G7q@P4|K4NDMM^n09PB?PmIC22pEr>)
zr>8R>J>27sH?9T<OQD+6+@6Z1I+`;bggARDUSmYg8<6w$Agei%kTF3B7>$O^Pcq_2
zpmoXd(IK0gTUcu;@|?%}hfEeZ!*LIkBFcJPxO|nwNCw>@ckh43$?*Z!R!C{s+!)d8
zWo^$TZ8Y=gDb6XPIAYN6Qq^;2vk6DXha4Rpknc=b6eY&NWHMtiopE}SlVp;?u!pl#
zRIL11r*n*v3<pC%P!t8do@HJvn4O-25L9)=!NDF^ufIlfvSO16*IN6os%{AdoNtBS
zc+G(C%Bpox7B(=Uz@H2lX3~ZJWudZI;#-}3@%b10hyT%k%pd>opA)Hyx2~l0(ru!+
z@(tc(MV3_HN|GeOniX%IWn|<XTHB6GYfV`eL{SpP(ySfYXWx7k4G)y%-&P@#CzgDp
z&3=LS{G18tq0qRqJyuy_jGt%gd{G`Yc-Qjmy7cR8K~*ehPQ+~c>ap^*fv=kUM-zQC
zLsiRb8)JQI9%IqErqUHM5{&yJ(kLO4E9b5gYD=u|5!KjFpEDBP;SipbV60dA?HpL^
zTLG~w<n_*FFdXghnqpNySIsB5#=UlDJ0ie&;~`nu@=uk<@GFFeHl_021w@1sim!e7
z>-_b<{@2-iyvHB^;5~|&V{*Dc2uB^}QH5rT&fvA8@*`0J4lApgs;;PYg^VI!SCOR*
zMq|3kkdzLINQj9-$6Z53(W-Wfp~5h$Csd{)>t>WiPF+{j#!woKQZ<<XhYrWM83c|9
zNsT6F&YUR`88H$^iEnJ{PiPcvci)Hxi>oOq(P%UVgK;K6f|CAl$c-B}m`xVUs}lxe
zK{1DoL7$f{U*r0vclpX!PWYqW|A)_<`{&2))`1O&x7Z|{tKie55fR!Y2n0r0?mal5
zEF%u~=Quhj5dn#(FNReiRYZ}`c=+Iue4$B`Zp))@@#Ho!*C>UhpW7(ArfH&G{j2b8
z4E#b3;dSE$h3{L;7GIRMy?ufC=?0703G?}!@yPdko=zw1=LH7``zRq%3S!|HcKdX@
z1BRUuI~T8U;qo;eJi3S0B~mJcaSZxHq>3qul6gMGI*r!;;G()LSmaYMf?{6bq-B13
z?0MyrIe7ucrv;P3B4#;J)J2MjL@5-O3p-<+8!{aBIh~%MwV~EEx-v|TOaHyleEP{J
zeCgfaKt<ue9p`YyVT{9Six9T;+E)P>B_z$kJjNKD_Pw!;ZJ@1Y(w9&dERDJYgo4@W
zoIm-KAMw-o{*+l&A%HLTE3)1mz1U)koNiB{t>?KrE72k8?VQ577$f{9kT$_cK;}s6
zyYJ2>r}X>7fCD?<0!4=2Y;AtW;i(6%y*KB4>v!jzhlK3e*7Zw<*)NtRALFAj*J@vZ
zpRcATm$A(WdQV0;+Gz43r><+-rQR&;RpYVjBI5VOYU_4<OO;qPKC91F)Y?#%1^Hq@
zk>|uJA&ZwsE;QHH+M%_DSQ599mA{5{%MKB~>+&yST*Egj`*|6GtiG}_=FOSt%T_Sn
zz{)ePd(yCrC3+!6`;0|f9q+RUiIX^?c=wHW_#gcb|1FAq$_GEc#mS)~j=ez^!bgTQ
zJ;jv&T2}@g*oWST8niY<&VmxeX-pPp#3DxFI|54-HI-M^;lWvKP~qVOR|%agr7R17
z`e^{GRJe|(@oF?R7DE!Q>t~_BvOuG-Xqw_!Ng5)bCXr#V2z+Y3#D&hj29GkRNHH3X
z84br|os0|HeYUr^s4Gb-I(+M^-{sAhf1PVre!ve`L+o?%3z{kVJcBij=eh!65ruye
zgJ3e5GoL*|*oaP);HZ(Ia<73(q==}E<oIYto>z=Uf^ZnziU;gJ!?`Cn^7$x3G?wR_
zXX#loVF8An*|NIavgm7bV~fXW7gH_J#-B$rpU*JHaCmr3Rn}<jyQE&dbcNAiNF-z8
zG^N|=GTaz5+}OdGnj}eBoX$wnF5~egNXcY&N?H1%M5R2WP~;0Hr^h^g_>k%1n7o*A
zc(jkT`%F&vsR~0?IgF_|I+{?GmU7{Uqni1AL7ta9eE677FTpzFtJ`JFa5!N9@DOb(
z4h|l9S4u>FXP5??rkx|-$0;}j!7{i|l+_XtY;$5?fV2V>BZO&<y9n#S7bjY3Q*(HH
z#B9Fc>)-r3ckX<_!-tQUmvcV5v(GSw3nS=cV0Dc)hN?1XZP6N%#Gwt;IvjUug+JZU
zX|TZ%YTxgpJ+|0MOfY7vu#{_}WV@084mP4;1U-*E^qh_1J?AiQz9Hdq?KRi(l;V|6
zv$l!UyfV!f1@rlwvaSPE=x|MreU<GS8XEhmX?**VmAVD#1j2yT7E>Eg0nT$iiY$wg
zBrfS^9pBPpMdC;hMax5<+P${&8)u&~FK6~)IO@1vUtyA;2ps|}vr@<9okfic`?85V
zTgPVUSw1<wkn?CD)wVZR!2A901eJqs)a7@7=U=0rW&C^p$NvLA`}rr-wXeW-7HZ3~
z{vjBX#k^v&m@^oUF;-Jn6^kNA#4(w($T%VvF)?vSA9!cUpmEqX|J*PWYf4I8FrA(<
zolP;;(ChVxlf;uah(q&sji$m-<Iu#UWcb{nrJ`htB0&-1k7Y|j^pWBdIStNYQQ?(E
zXpLf=k8R>4VK5kyrX8+dz0S^s3k-)Fl(}J$?(nU5zssxFzs64g&Xdo6MV@=0l<zg<
zEw9m4dpcE9i{TH`iSTXG4H8c#_J^H@2%y~!-ckx;j*d>5oKCoSaSLeb5o`<n`1K-L
zJ$&c<24D7FYSz~|Pp2_J!oRk`^AUc*H1D^@d(ho(m%1)+&iF%{=BE_pf~xd~{0s*}
zI^7;o6ca_tH}#uWs3>MU-Xe+;Hpbvkbh-oLB%{AkkQZ}wRfm(`EViyGi=3BUnKC&!
zLP*EK;T{hjKj7m}KH<ZUZ*l+e0h8&1$;lk0pezMtH6`hQ6qc+T5y!r1Wmyy;3Ob!G
zn;Sij=L_}^9#Yo@QJVPh*%_SGNNa-+AfA{{=QU#6n%{eN66stZ61vu7m#a_)KItjL
z`7Op+l#KZL*T2EnzV<b0UGm|FAMhtX_(MK^|EC;JCv*})uV;y(K*YBGK{X1ft~7P+
zcXQ*Jd)nwSS*lcoZ(JzbGwjc-U(q-stA`n_yk><Q@U->yJZbVV&HA#=wFOtg9}4bA
zF8m$=;cJO(b7sBo1W?y?po2Em)pli4&wgN2h2C_OJ>w7>7w*TQwQnO6CyDO}-RaWp
zcIkDyMCz>dSVsP5m9s<<N#&uK#-X!kTfH<?=GB2T5rGT$+YUd>)oPm;LbT)Qyb-I+
zyQhhIN`6Yp<q4^RCSTAPkcP)3NQ6JdZ=<)vU-<s-ATa!`fB)a-XYYSOep-4sFl8Bi
z;J=pIF`X4H1F)=0WT{BH9e-Ymv(%zSQ6Y$M$l&)AT<az~A*iUCnkoCodmJ7gax$5a
zrYWPXF`YpVrv%!QZgE&NmI{rBqI@O;ECmZHN+c0o1_=6ObP@Dep7<52_S@MY#Nr0+
zYP8WLQH*htix)4mvvrYc*KUv|T~79A>~7rTjhkO$*tv)xed_aQ<he)MO6iNCe6gOv
zhR#~v<7g^VWpKzEIK!W<A*ILTWE&RyKu$_alE`q9p?_lDa5M~|5LOcUj<nadZFPF1
z3|5b`UETa1R9JFX{hwspzX*Ky)6>%_rzfXqozs`T>7kSnbyaY3I>i`YBhl$({@@rS
z^Lfr>G9%9moRxIDeX?{)9H;aLUB(;RAQjqxjD2fmWJ#1{SgU<LB!S^ZO20S47|m$o
zB0HC_ar2eec>S&0eDJ|Xy#K+6JU%!^MKQB!#bj3Dq7K7BM5n7L$_iUo#EGIi?0|J_
zY>tV0U6LdYM+E!Cxi!HENaxRXIOomz=wx|_*D6jQ#SzklXA}&FFIMxBiLbx$XP}4a
zCd)dE#~X~tBQ9Rt;q}*FXEYi!Ii2v?2R~z0IEvB{D;Jy(*CzaR?a;3DuuemSa7&1+
zsUPsvTi->O)dq_2gdtcm@0Je6`N-o%Pb6$s90AH|U>LFDWw#b!Wh};c17bpcsX24J
zMGIDMVTJ63gXS!@<pad$-|OWmjY>)4G$re1^m{$JY4%Hka{<l_YG+BKmn(Kgw2)jA
zO`W&B-bkGNSt;2T1hxz-=?$KBwr!WvQlVEwoe!N@g0IW3-i-L4nzLc-%HJj9{uY1n
zw|<vizsLXUKl{)4JOAk)aB{o!db&v=c%6+Eu*h@Ds-)5td65$(F}3ywooS=7c&c+6
zKrDl9K_&la92GTlGhzRDj|UI#GhNJC<O_uK(3oeKNi>c+fUZ9Kp|An8^87|YO-XJG
z=GBaTJYqv_vW;dCV6GN_qsE4&jFu9+QqW<vZz`*zh}YhDoh;7ic6%hM&(6@X)4NVT
z+eEt1S@AjYTnI}^e;lK6Hk?dl0}SFlKpol}h|p8pa}AX5o~gVsae>2+v`A5M?Zy^g
z|JIvqZjM(xz=rnd5HxKoy!cJkFshcluU3=4tBcgXN1XXy$|jofpCzOuic+T2Ifuum
zNLNx;HEG&G*A-=1QJ0o1?GZ(mSjB9NH>qol(UvqDaP9gvwzqc>QZhZAaC&mW?K_`x
z=l17ZyYV`sjV+=$_EoaZU}}FhxU(8(4U74d>2%6`kyF>2$zsB6e!}VLDZMn~D_?$t
zSw3Z6Xx@MS4v!9VPVza6BH~h%q9_s8k|Z&vwn(LX+w)ShyL*)^?I5MVYU>*Sfv?a9
zj6F+6SZ=swV!dTEg|o$1U$5)X_QU$9(H}DHyQLOI#dJDnI+>xA@{wKX!)B4h)Iu>`
z)KrziD2r1TYbvaEzB!<<)IqtfCRW|b|2<r??EdTU@Y;Eia#?ZTC_=64Gf?LfH7ze7
z4OAk8^bt|yPwbPvvaL0c4ZT<EnyND7dEqP6LqghE-_-Jqu?V-_B)d(`nfwO?{Jw#}
zw8KWWO`=;Hj{N-<Z__&CNfe=a@l29i_|`Ah%8;{K<~du4aP^=OSb2YwR`$PBM80!E
z1Baqo0}@J9gF|>;&PKt@>1qs$|4ox8P$*pBL=QSU{MPq=mzQsTnca)8@Sptczsu(z
z-u19xQ%vW)Eiu+mlqE$`l9UB;mC!Sq+V}`WI6)#JI%Gsdp_1J1w~`8j=2RbX`_XOg
z-n+wWv7p}{`rJxbACXGvGa?HO6)ljHhp5Io&w?oAFuaL5E|0iBzfV6M@=|ZcWsx&r
zi-;}~iKawTg?}Xme<qH$I`p`6j7B5I{Y?^;_}S7iqqxszdVz#4*qC2b2=aUyZ|EW9
zyX4{zug7~EKA#}N*&q$iK48477^d{}Hzh3BZ(QJC`To1S`ObB+EDavN0I~FZR~bPr
zEZo*Z6e~TNe&LP~ELZd@WwS+M>2$gzNy_oT5zG`y<|q}BBq<x?ZRU#wMOo47^%)LF
z2m~7&n`~@eqSG1ToS-TzoN)ApT{@kMRC&qt<Pl{tXS}gVr<3{SiFHX;<QQEspH4VA
zIcAp6iJ}hO{*btvGTt1~>vg%MKj))Q-={D$y8R)W+kNDnBaBv@PHTEy!_HQU)0V|T
zvnWd<p?Uc55F;di;lF+jC1Z>(FvcHZ)s!-O9CamU<OoSfOWTNPb-uzhuPS1x0_Gj^
z2u5q1^p)09I<~g98IMP3qp53;8|USMO4pzwin^vO1vXJQVUS7%$I(~!n%YtpB~@KL
z5lu9as&A{YQg`sw6(CT`LuA%@W7n*jr#T-lY)r&bK?<pS|EMS!W0pB38(`Y9E~%=T
zqRP=GP^Uw#$*vfUR&2oOo9c7(<PZRy^$C2{<X&8;-|(gS=aH&kAg-lw3km1!foD&m
z+J0i<oUyRtY~i7wCb9DW@EajQv0QjwLEVGzi`O)3wC8Txr=DGFCYCM`XKr*E$%M}d
z>o!$GUcd2W{wM$AG20im`CI?zzs09Ndx))o4aj;h*nq*%w|6o4ykA*WRCVRaB|)Y8
zdJ7Rosi~=h60bR`4!ON|i#uQ3V(;jHC{5{x&Jn(oaKtDYF=R+W#+WTYvd=zqX5reB
z2t~mRhod7q47vjj9`A8{bVRZn<GO||mD8n1jPFNh!nq<PC5D>9Ey$}JU28T58(bgV
zpey^7loVEjOzEjH5h+qcjplsrDtRu1<qz)5Js)33EmR<se+ZJ4SoA6F1QUiVg~W-_
zi;5_8@+^7%^<BRA{jc)&+pm(OG1_{Z+xLb!n=o6Yn656F)#Mva;;d}FnRd*gO=pH*
z`RtxuQ8Lc?2xMdPB2{HMooEmmNlhF@^m>whrqJ4;BEiPym~O8daw(dVgGc=M)*o~C
z(PIvdkEqI=;h@X)OS@dy+G2BKjKETr5osqz$%wKjS<ENCQR2}N`JzBON0dlpl9Tqj
z#L0lmqY-I$h>b0ufBqp3MZZ5JO(KjD)EZ13p`wi1<>d1M6-CrCXPRq9qb;spzJaM~
zs;cyq(x6cMZj;bB1_Zrg1wz=C#(BO$UgKRhpa6m4(k9zQV7)mQ^g5(*ijb1p85ZRn
z8Ch;#zrlDkB#ISP0eOH7WGsl2$hXukwRgS+)KyMZEoh)i&t?h(5iay=0x5g}WQ3s6
zla@e=k}?#)+18uIie-*W_#R5uS(K1SemM;*0HOTSlFnC23+21VN+poa`E?x$w9^<g
z+8K=Sr{~pXL0K2ny1*bXF2)+=2_;R!yh$5|p<4XGNbV`^w5)T~x(=;gYH*%jWX;kD
z{Hh|YBr<-kr1OS&nTHS{Bcx1xKQTn>T>8KI0ydsvK+dOJT!=1YQ1DGPM|1PdIeif^
zw*7Be&EFUjf7zyBrePO~6$3;&o+h>d!;PJr{MCQ$uM)?W|Mx%rhupgJ1@{h4*t!H)
zuWz!wxkHksp5#&1XcbfDzKca=N`K@mi`IlWY|Fwru8z3<_zUjbzr*q5n6j=hPH-xw
zs3=006%QZpb8+JUH&kS*#%C-X1ruiGgj!c9C;gGRk%ywXbl8YSRQ+A<Ozv{~?q}4y
zCH1&sAP1x*G<}jSj*1FHNog09bxx%VV!2I6^bt0p&=tDqFv@o6CIb{Pf;c>a=dYdT
zv~!2;ofP9Nd2YbPfz3jT??zEVWHXeBF`~qw5ky{LTc7ZSf}1bx@aKQ?o4oPX4f?$<
zI#jhfX95HwTV@pQEZuQ6$MLi-ybW^ECh%PNVbu#+F)9+_dE!FQ=?<7o%eD`c)fOW?
zPx{fp0SLoz)Z@a13v@aik|ZI|7kv8hXZ+D0{fO(YyvFT2w>j8*Oeag({>|UyQTdo+
zF=zMkC5+K6{HE)Q>3qV2`*%6oKV&w`DK#8VrrdvYAC;wa2Lmo&zQnal7a0tOT)1$F
z!=n?@G$D#4N;*XBapyQm$udE=8)2OE1$HBN<>fDN<?=Q1#R6li)_}C^E9oD(TZ#0-
z)I5D1v~%rUP!}Rk0>&*v6GH2J0(UYmI6Qnnl4R`eZlhF8nkJ0;Lq>yv=f6+p%=3!M
zD$)S)_+-3auhv3c`;NP3h!l-N5)G5Cp?!Mn+sWpn#0JI;MUo+B=0bi#x#jxu)%Gj*
z_!O|@X_F#DK25B?r@2JqeXm4me(wl04(%$m@r)S14`kb_r2W_2c0dq}^|?X&l;dkX
zMCPN9aO|OvHUn$zsVu5rAW}%5cMwr<@}jn$0|+$z6d@W){i*>mh&K1I6!#Qm6f#=D
zcr?J(S+CdIQrBK%Nqi;LT(rk^adsg;%OS7<(()SN%Eop6+yClcVRW(2t@|JH>Ftl0
zl!sinuuU5GsPYP1YARKrgrY18>N2M+a~5U>$}4FpJlVFQVqQ=A{K2Q}?eC#Yjgy{c
zY-_``o?wh+K3gE;J<7b~=JlIwZ|~60`X~hRddA+-WA+dBDC?RlS1)t@@>L=k<8X9E
zhgZk1FiJ-J@y~zE&pv#Q<12?;xp0MH(j!3;OP^HN)?$qINI+gvlr?)t#|Zm?SagU*
zm(A=V{pdoQZ+u3O`-D9AK&k)Q?|hA0pWkQi@lQOQ)g-8z1eq27_evqH2--ZbNu&nh
zxO}b0cfbEGufKVNtdoF%IErw>x9>2<prR)Y#(6lQVO*Ur7HrXRx2ypW=jt08nx8jL
z!sD|DMjJa^zV-@x_cTfDm`>-Ewd3$~f|QCuKSe3Ug`Ep@vWzrKDe{7DugAMz{VH#M
z>0R=wBA?B8_2nCU^BdpfqYvKqoN?_>8`N6+97{gu_~?j-j~+3fF4)?-z(#k-XMgwu
ze)hqqT)uvVuibc=_uhM-k3aYsufKeaNJb<{OcGb5nPfh%h!b$m<M<n!o17f&`{Prk
z0#d&D&2N!q8C7X8&VT+Y3#O4L;2K9z*2~{NE3w?LUz!|*5EdaLj*gGH_1?$)@sIwL
z&5cdI{f)2i(#^|A>4;TCr_-h1?=ep&<W)tbJx|~F=aOFU;o(xFjX$+f>tz%jlHUyo
zB7-9!Wbiwi8V}c;xF}o3SObqX#iSP3Zk97^`d1K7^jM<p-Vpe_K~q10Sb``x9}yVs
zFuD%iX>F@mR|vmLS;>buhfnaHdy%#H8XsfZx&pfb?l^vR2&<hd8^(WbDF_5n@bn$4
zP|?!&f6{=6b3YTJ4}YD;h&;{QdEb6nh}g`QJM*5W-{6xJWg7r5)qei&=muYZ`!^7A
zMBMFAy8<OFx+*wz6N<$=uo-KrV$OV~n4M;HyBSf}cP{oVV<ovMIGh}@SojXxLQ2Z2
z0wc)h1$F5mOpHSa&EftYrt}BYjfP{=Bq2>xx+>$+#wCo?Z1gwiiykU`N=AafajCn@
zyDxu<pMCxTpM7?V*<{Y{&P7Ju0o|;FaspEujP`6&S2?7K$<hIt8#0Wy7<M=5s4)mn
zhgYl?f<9lKX+Tu0xO`=c4}NwlI5ZUvC2h5sPs#*IjBuVUBWX@4%eioIhwuHlcX{_K
zuhAcLJoFF}X-Y|~RX8kXp{qcB6lay(D#PZyVmrgYYN8O_6AIIxjZMHFB*XC*-}<#b
z&p+6EfN^swBgl&YiNwp|vaFMlWf`5ULmb7t{^lE0+TldR_rCi*zVz0cq!P-a<idp=
z`n`<dXo!#^kONAJe8KVY5dfF2T&33^ap~$y+`e;{hmRie^39ic^VQcu*L?cX&zPQ0
z+1c8_4F$7B&S;o&bd>wR+xljgS=zxEI60nE8%vfB*xtGn?$sg8GJ0$CN}OF8ZO<@g
z5Wx*@mB&+(&HigB=#-KuiV(t~onSJZvVX8g*3Gca@WUVe5sx44^6IPCks>0FV`@|5
z1QZqIC3HGXo(mvE6Cv3=o6o7LiYSX)hEfxeoF`Kl?VCn6OtSFQ+rqSl7|<NK+hF&`
zn5EOgFSls~;_Mo=ODLr=V8(>#+MVr&V2m$Lt8499JR!f>6ds;Q=QQ8j09}P+M$e0P
zo=lslZwhIBbmbcE%daHO88WtEl?fRr$;})r!z6jmT=<`!OupJDOysh^R&(_6)5Wy}
z*X5mv_WhE^NIdEAG$T7ZzJ{sQWXuH-DQ{o?Hk<vc{NU4n!R@=BQdcKnArS>ylqlgX
zcv;Pm3&-hEL?=s0WkS-8aYC@r6^AEB<W<gaI0oT}^AsFJc7`;LR4H^~NYf5+9JlRV
z>SBS(x}=>Umo_hPaj-*76ppv`4?YTnfY{RyFj!vMc!lk+?C`<epL6HIUG7&8$X<Sh
z1lJyQ;G7^1M=l^cT<l-t)$K2lGW2AK;6F*>pQodVLZ0t+JoiAU&p*4*{ks#C%&1(B
zwK~8SKJxGeAx4N8;gXQ*iIKvxv6=E4-+7y_eDxiF$dZhFj~n4PxeSNVG_@ZdzH&<#
zfTtS}pk)ZHQZvsE-O*(A8oF=u`^M9B0-T6==iP5}|Nb5R*+2cynanL1h!800u(c*h
z60*#nDILc#{eB;$!lH<h4p)T0<T;b0eHMAaa5$vbO#`$eLdyq3RTk)~B26<c?p~rl
z+(Zb=cYfpBT)%q7(_!n9VU}^@`c+2#9vd55r1^rw;{!&cj3~xe#mByzu!<B(oH02$
zWuC*AzjTF-@deLrQ3SS}Iv3Rie&WhpdI-q`Z~bik>AZ1n>O&gk87UQ>Yr^HrTfFhk
z+Xx{!IymL`fBzrz)AxSHVA!V{$MkzWHn+wI1IG^@F)cKMTHy2HA~1eJJqVEJIYm*@
zO9Rs?v_*ds&h)Lkqex&3p)LAZ7*hsgr?vJT^EoK=g=8h+?Txb!6TAYag^sKZ?0)B=
z1mu!^=xjq@w`b1$ZigPMnzJBko{lXD0^7m2Sh3D5VcZsG{MBwv2tg#hg7u{=XU(fv
z0gM_2zKlG?Ic2{(Wh=Qz;QW!e?%Zo#8-gugCszKy{GErs+__f)?a0pl+`OhST#ks0
ztHYQ1{@YzX9$n#!yFcgX@E(I~#KT8lpeg+GHJWm<V3DQFXEU;N(v}y@rZeXA1xYtW
z*A+!k;jBQ(lzv~*Njr$Z2<>EDdc8hL5);KKX_nDXhxC&X8_5<Mj8G^~-DQCZeXg4(
zjb6$7?6Xo4H?O`#uiNKzddi?PWEc$)N@G-oHV#udUouqJ*jgZ5iVMZY^4u(~c0PNa
zJkx*}ZTale`xFZWE=AxQ#rPHxfpgHn0<978o4VIY`Px_C<f~tMo$+XhXK959MftF^
zDdwxS-$Wv6jv9p}EcKKHxDtt+XKgjRWHnk}RS*sILYqJq^!g)y`!D}@$ZNw7{@EXp
zPmV~GqoV|I6w&Q;NTP&DDLPq)DE5`;$cUn})MbTmj`3)N;Q*v?Xq{7*MW~DgtxF;$
z>2|vmC%%1qC(RJTl8gony02i3r7BAz;TVlFI%&%0#s+(b`}BGly>3bz`^_200&8ka
zU6XZs7$b;f#P`1UTWpNC+sNr@H8SAaN;lA6LzDJ%+EfEC>ui<<b8VQ=vbUPw3n59<
zl%0zgxwyN96oRe6kUOuugms!M>maP*%GE1;xAR>d-h0d+{-b|Po=>T@BTWcZ@NE*o
zGhT|EvRu%ynU6}W4u?gpL`LWOU7cMgzTTo`K{XI(h;;vdLE&}IfovkV)pfgsu}niB
zhmZ>iV{xJ_f;85kn^U7!(=p9IgKuH-l$>VUbkcdcur)sa-~$HNq^o{K5$#zjQqW)z
z=U&lV*U{3zS4>GW_+^s&X~5RA#kK#f^m2J(1f2WKCUM?mRAl>_C&w<HxE8=pw#)eC
z4c^)PGG#SKbHd;I=YN}zZ~xSvLK&R1qNtcnX7oCJ2K_!NQj}FemZh9dChQ*`czqGa
zh;YbQXV9nD=@F+1-EN;=XTV0ZK~D}yMMjq%DP1Bm{Q2mbiYX2mx>{AtT|s3FYFlzz
zo>CMA-7I6+9ddK~CiB6Zt#q4Aq-aVinN!M=qW0(3TH}aBgc8e|oF`}DNk-e#<e3IU
zO2LJT8$7x<V>Yo!lOUR2SW+OB!kGm)jSy-0sG7)GZeH5u-FIGNYkSld3JCzG5H9j<
znT2k<aXJWiHZ<Nw@ZG}jm#hjnYgk-kH(H)yQ!Kdjq+6qdqdB-iA`s5g2YbU!e)})|
z>vWQoAODNr!<DB<Q_@Wp{Z1E!KNw09k*SWaf>tr9l$fOJF?b2eCY1S<y@M&vS>IN<
z)D*=6Ykh%HEELo=StM}MGoq3-#aK&}X6(i>!`XnkF3~uU0u^=WrCqYfxAo|y5ei4S
zm}9j;Ysc5V`c3})pZkkMQIE<M2w^<db#8&Q^X5XXO5kayD|2=x2PH+zFbQvLJ?duT
zN+HRzE;5QaotEBcI>)Q8zQkzMB}-C{P7cV@9y{ZUn3+MTlzFkBw9pNhv~pp{!r_F&
z*cnA}jL`$U;R`fzyKI*p$&d^czD57iTV2vitu^6aQ#2^TS=&v0!_r~@Wu<{~JxL-Q
zM&*3X#p-bbraekfwh*F^q|P2I>{bp-dqP(>ctXKb4M?M=t!<cgzv9h2_5POQaN&p5
z=wYe4mkNBT$QwRe^Pa|lJ~<{8RFFFZZ=G$|zKm#|Ez9+|<a4nylC$^rWV-#l#3E(f
z+kxH=P;%wcD}40D2Z6yEgYxa!F%oAzB;QH8bh4aYe?XEX9G{+W`}SRqPv>-d8G~V;
zey>lzUojeu5os4&6~wAWIIkb9NXW4mW(eji+?1J}kedZ_Goz|<PEU_{cyOPOZ+*nQ
z`}g>buYQ~Ff8%>>$}uS^uDeZ27K#u(L(VGDvc$R)=L~VA{=ZxyUY=t>B;WmWU&d(5
z@!@02B2WbJ8$ziFD?C~z9T*W2yNUr7J8{8Cc|N*u4kemAM}zS;8_yf6rDu31Te@Ua
z35!26>nV`R3e?gjLR)ss@^#^={%3-pB1AYtS+Ki%i9i3Fe}Rsw`TXZU#+EZgB)N29
zmsmv}Y6yL?A}Q&nDb`AAQ)6_EbD9iCk<YQFCW-{kY6vY<j5g@nw-9i~Q0EJhj;Fgu
zQX%7nL|Gywkxnz685YGNw9+T!90a<yK4I=O^F_{dmUH8!Yy9{A2mgJ>o7>bzBZWc;
zzonklqGOqJSi)B4=W7|&ooyAeRD5`rk8~&%F`wr=dhCx69SsKzh65Cq*=$B#mA>j)
zhsaS%N^L0$FkOklg%kHgIKI^zs(dl0t}2|fZBcAvJem{yd=j1(7HZbRYKe%C5bR2V
zd&P(n+Gt4eOC~`?yVjaRj=ZDdtEs&aX^0!)^F?ULXAI~Xtt;Oh*_ZVPXk<CwR)02{
z?iu-y<ufsD-o+>KeNMza<=Ou-;tk|#cH_KJYD~ql+>iEk%h#_~XFQz}+BC{p=13Yt
z;8>D*pJhPA3i#3(kERdSa^z=Kr44Q48@KWoM@11E<6Y9E2b2gIY9y>8jxw?|BT+FU
zH6$8F=-4t$H?YR>>8GFbkACkDnamcv^vY#kdF?to+dB*gBYOQI<MEh-!D*;T>yRcL
zjPQl1u~3wCj&g#kD$#XGWEH221xM2fAAj*FS)B0Z%{RH)xk`_Wn3yP>{!vm>u)t$J
zia1ITVvbObBuYZ{O~Yt=u}%!nG$777wzf9;_P4*rCm-DA^!UEt*w!lz@IA16(-4gk
zk`9h-S)vZ_GQWEZcV(AI#VBP#sFpV&Law8s?l$nklgV1z%`R7ZW3>h$Bn!{fhxj!?
zL!WJ4cODaO;1B;^g_MHbE7$qnpZ{%oofuKgI6gk0+v}5N9jvoxQxioBqaE7%S`sRM
z#)hzl+EzGMkfjM0jjl^@l1M3k0F+eJ&i}sF6^T};D8gFL$WcO)Duqo`W>Z03R=(|k
zwZ3ko_6k;4U}M4F;R#Wc@o)Wi|83rS=bdGNU8oLTP3o@7>X{iWy?3RP@#-3I;Rx1M
zhEtfq{||fr^(0x6rFViq)ys@K=N>#GBWYEV%Hq|ex|#rJ04+1X%wjMP^ROa2FOdC<
zR#tW%AhNqNv)Bc?8v`_obkXEIGa`dWI49=D?J`x}2i0rDgJfiORdu5Q&GhgHce87k
z9`c><d}jm=Nn^<ZLRpHCl%6$++K?;7FdmY`iKCi2JuY98<c1^#r6tly#M_#$)ABKz
zco>tU$<%#O+%MLNIEt=jYX6Gcp~yNOc27O5zVk%OSXf26``^xT*9a%=E~P>WjX+`C
z3>K{nTIb|>ipq1+-2Tj{9HUc=&c?ku?jEK<fpR{7yr==K3dw%wJe0yr=*xYPWxzH?
zjyO3!<Ki>!=bpP#(^Zhuj^^>Peqp{RwWDCn2Q2zT!iT(bS_IPWhj%LLn1s`cKfO~w
z|Ga#}V`K3Q_1Y}W-VPNZ@O)|E>MS8gA!R^~1}otjO-{&xr<O1A$KN;jUcg`d=nd}Q
z-=bU*oZZ-9b8Q`IeCA5?=tK~ih{bS$ei9SM39GXkRQ)P39cH3Af{FlAB0{!T?y<Pg
zV6IW;=`&C9^5V;shzRf~QAQxq3@{U-jGh2#$GuUl)Tl%a#@bn5F35n#j<n7B%m3f&
z)GNzisz@rx;to~@DLjxqpz(A@MQ6-QSdbo~)n)h7&sf_$Pw7I4@C$7K15!I}Et~om
zli?nwUj&oJ|G{Oa)ji}FVtiXp8cX|dPtU)wFV{#BApHtU8yES?D={W+5f3_KNkZTU
zu6Ua}{=Yqxs8YuBJ%;UsxHF*N?~L%;EVaUtp0RL`hexGU;b{MmEJ>(WD^#M8z>BPM
zP4S?+NKz|B8YkpQiq18|!O+H7QjlbdEXi4|FY=H7?7!tZzxxA>^hZo0VLZ#saSDHr
zJHOL)foUb#;$#-5e|)|K1-34wK>C(o5LT*`D^;Yl^&^$1<XJ*I98#&2@qHhqa+EPB
zVFT(UljOOF=NH6NVMl+u{Y~?XEX@d`ut-d&?uV&tXv+CnSah0|HNKyUvA&ba22Y;S
z+jv>vvKCQ@R2DE=P9dRdT!x{Pb}}}(W4n7;S$sPk%IMf;tnU6D*)=O`G?p5F#46rm
z(cB5rgeX{vUv7=f>w_Dagp8&$k1xI^!LUNVs(`MFtYT7DqsV?H!Kgw~@!_?Nv1cxR
zKRF;vOuhEOHCI|<d2yY?_6}Mpf<PcePLlQM4f_}u&qq{Qrb34nZQWv_KF{}m=X;zx
zzs}7&x9N5}RKkGs>l>`kY_jaHp)x~AjrnMuf$lRX4_J;i@I{0Uh7?ZG&|xpTO{=#@
z2z>qNS6HuX5>Y}Dy7OVB+zToAf?6O*84lX$B%@YZrWDOkIFN@Er=M&+(t+f;=0|_^
z33H7-+O6IQ%D`nIxNgy0<p`CrV1}&t9#zVC+UNM*Av>RZMrCf6MzxH_Ns9|@Cm9Nf
zauef(un}w!1XIRg%+9j^o-7KF9}uP-f-UOX+i$YaetL~wD{Y1#gonaI`W2Qp&ZD{q
z_$-jdF}+U5a`rK(JjEk`G$HACNRuJ0{cW0CcTqynn4ib@Lxi?96JdO$hfN<5pfy@$
z?CfmWk&=}XLVEa~kMw-9EQg+*pPgqJNfMI`ho~&WwkZqZIA>{fjsN~%{7-!M_r8zx
z$|yH3)y7ut@sbYuaxx^N^GZP(9iN*?*3}eq&Aq-@MaEVrGFwR-K@bF#N+mL-TvTqa
zRTzep%Vh%3M`B0J=0>B1L1{=*(8iK1G|ue0-jE{8NRkxK_of`%=?Q{kB7hnFzEI-z
zJa5GIo4jXEGtIu3p`9RZjEdETZkX%DD4;20Y0@fJPW#H%Mr<7hS7=w`8pZ^YJ31++
z6?h-9w7Vw8F%b(m?aQpfdDx?U)LA&a3NpdNcJ}f6;Xv-Ge~)V_lf7MqyE&a}2&SFc
z<gtvmI}VM;j-!Ci9<?gr4Cgj4@X3w$81_>OF-5`#dcz@c7BdWDGBV1{Qb#io141t#
z3PLv4)|p>gpxf;e4^vKB9p=j^Wxq-#nn6$|q)sU7%$ADQNrSLtTyisHJHN-h!7cha
zVWBa{LbO2WG_rgQc9^ZfkdLI`ZQn>ht}`oCnuGEL%|VxDnqhL=p(><AcoNUIO5cGW
z;QJ9(`6WBi0dy9Urf&M0%s?hs*$O+@-5`)Erz#a|LNF&KJ`q}bs91CF<7-rx7l`Ur
z{K&_6(n*|H?zY0pG%G1Bv|*-UC&2ZK2&5Pr!D);h#_4F0Rh)7%OgNZ{+shgetLkd7
z1#KfGK~zKJF`<cBm|Ni3_vv;zB=OM2LO#P@i>%jV(Cct;_cp!b12!*QLK;CRLwj>+
zn3<hLdV+p0q}4ojB6}Xk$H%riHSn05Td>{nzHcYXX>DPSB(?I~2xjV4=4N}GbYj-l
z&+<S0pZ{mR@$KJ3yU0ffL7`=3aK=!Kop50+Wx+sm@*tv!_Jx>&2*=sRsYp!-r0_@Q
z-Wq!A*xdg%8_;?zAgDFDo6uKa^axv=O;baX<OEWWy0Hz#c0nsm91jQr>7Xh#x-WdM
z7zLmv_K?wjyMq~L9iw|<JSTnRSeTY8f9w*)917RLD8?M6Nh4NdMcP5FO66pEj#1K`
zGd<eZqR>B*Fe#izj?F%vv^AKq`Ie%PLH}|N<jbwULAo>A!X$HR`p>as``_f4rf4D9
zY_d4%58rMeBQCC7VRd1X?SosCB1Jh2@r6e^%s4teCXx}QsDvj10wT(yOyC8CL5Na@
zc$i~EK)Djp>UX(yw8db-(5TE4$tra*OO=|dSx6!x6q<qQa1d{^*Sn7nGHQ(lW`l-n
zCUo^eatwhxb8dJQ37N^rP~<rb`-=VDCO^M=lYaYW<cW{S3_|C4r3euOL}5UwQYH#Z
z_(7fTf9ri7_gH++Kq-R}660C8swhrrm6Iexd^KcI$E=E!x)+WfD5DKhd1QxucHe!U
z#f1j5)d*3qV?^K{F^$yYZsshv&;}DS?NCI#uFzY^dXFMWmqJfwAd}AGR82)nMsm@n
zfFGnq25rI6s%vUl!IanNws*m#WD3$GCXNSou7gUE#-Njo!@WHYj}F;9yGf;7A|1w@
zbh@;<U8=Pjh#YMsC(R~z?`_fV5Al3So+-Mm7WZ!70&S?(X2=pN@2!%UG))+0DM^+Q
z1OZQ-KT9d}sm`tP^0$AVZ+`c8P}ojkIWqj7qqzdY!2lDml<)iY{1^k$Lkfu&#q4xV
zA%->i7}<e84mxlSLaV?PiveF?Jc*G4VLZESV-32Hb}od%2?c6}K{;&RdWtkRB$=if
zNjr+w2(-``qcBQ>5~ys5&OAFp);SSRxB{)fem1U;L@PV@*(nZNs=0AaV<O8iP9<7c
z9V_3%-DjV(PI;)}ek*=C4pE#D6i~3Dv?Eg#_i(W<MXhQC8&1Mkg`assyF6xNjf1Vp
z2EZ@2T+MQ{Ll5r6WM#pgjiP!n`P#=B`(g{@?~@Ls*lLl5JXmj;To3!4ho88j)>2%t
z0tyokys!9)$4cW2PoH~%ot-U`UQQM%q9`CwbNbyLy=tGf-=XZ+s8gp*nMg!73-JRa
z9`PW?81Q_b?y!&O9y0V5LD1z;p3smpR4G#;B4U{tW(kEyJ8H5RuHXsqr5$ry1meY0
zYn6I6hK$^aC?+~3%QMp04ye0xXP1vY-lBIR5h}n_1}QW_IYfk^?JaY?Z$eag=EDA|
z^Dw<0pMmJiHV+yPX?)8?H<0HsUT>E<al)#fvfz278WQ)gN>Mb0LF?WQci#Shr)FlT
zoewc(pIrE(gEtQDz*zYY>Dd{?FwQgzNrllvTQ~t>!`PyebIgV*AQ>@A-^E11bCjt_
z4+_d<af*xq;o$`y<+)XQodbF)WLhv7CS-X|;CW=jnC8JDC(RR{dgcm?>uc!5M&fCf
zFvt_Kf#B%QEr$I8&E^rU)(I-l2*bd2w?}BDIXpPTlOcGLJj>_}I`jt}(rid>a>8<n
zrR8N-)}LZ-Ws|Rb^LOyWGTKe%d$ipf!-FF_^Y}E5;GKdiGkK4UY9vNLp&jMyf(<0Z
zn26wwJHqXI$Ddzh0LD3yLjO!Vim?JRWzbq!{U#SrSqR8p4y`h)E2hU$=~O}A6!(O^
z_lz-U=R6*HUMP&~I5Yhs7}g{{@7yX-n7JYYaZ)D2j4~0G=OeaP(Isrh7&u6+u)PHn
zS@HPdd9jL_=vh8;QbjQFWwQ`#jNC}qhdPg9N5|)6Ou0Y3CJp}Nw#Ia`$$pWCLs?_P
zKm8fv)c5foKKUy8O(s8WdK<NHfmg14muuIrv43!jBuprm0|MV?*c;GpwFzX2uu(xf
zBOl2UzVs*sW$G136HT0?_?}PT`}mPZsT6>)vp!<_)bKF~Ivk<N0L_eSP;-4DL%P^`
z=mlh@@CdQ|vW%RVl$eYdjY6rMlam&Q`^S9v0Omj$zwTAu{^<==;$w1Qi66!we2EtZ
zNY5t>d^-RyYVhrswjX=vJl262k&k`9u_~hmgP-@9&xV{0I;=&OpK+uCZSXW8Y>T}X
zf*=tbT)jhmX^E&>rL+;DeD{b+$v6@KmFI*(@BqGD41tGn_90wx&=i>V#9<a)p)g$}
zo$x>wGLjy`^PO{H^t$3b%CAvhtpJ9TxR3O$p^!$9B{^}L@$~c0v$VQ`7r2g2U(txl
zEUcd)SDL;3eTKsUrE<h_X@w*iV9<m?NTpJzUax`hXttWv8g=qKquo1UFzldoYIiw^
z(0+}jmGeCR%GasZ=gEx0ldgY2jx)2OiOm|vX%c~*%{>YAG2v8PM#eORC6UgqIxJsn
z?D#!Pf{@a4U(?<0h0x*1WH@}Ps0S2qS?;osTsu=@=~N>L_K_?|o--+|sHWIOt{0;S
zDqwWqs)_eZEtjp1(vqT{@dV4BD@Wf&fGCQ&Mbz)U5Ti_|fbW!22;+gH9cwjalo=;G
zGP1Rk8OY=IN7!dXJJwrq%}3ACmqdFCRa|YXIHGZ+411)C(d5yI^HSsoV^=WgG$u2X
z@%j3^9*@)Sx3AWWowqsl{qxosZY=&?e1<~tpek4KIUXDH7x>!q-{42B0|vbo<(iKd
z`lvLc*XiNQh$yNeN|K6*C<ru>7g7r#FXwm$gzr<Ylqm;3o>6$hCq%e*!8SRP7Mc#n
z)M>b*OVFf96aoH7g3?2hlj7M<5}SRPA*qVVa?NlU(>^&NOH;P)ZgJ;>J5O+q01<KL
z_h4=5(Jy(dl!?WL-A<YuL5?(<Ir@~vfE5{2$3trmUr2;eNNwMfp~>8dp%=6D?uS%o
z>a14Ec=L<Mz_;=pE=wqQ@~3IMr|TFGV-ubJDh%oAk+<Mj&kj=XJs&9pJkRnMPJgx$
zK1OPk@X3w9^Ft~npK4SliYk<6X7D_ZqoXD>wK^h<kX}T&98s$^@coea#yrD~4dQq}
zuiK^FIU$aRq=}uM+-djlgMbK)AA6)}La*N^$x@_%s8r!AuYQlYrA?L>&f=GABuas@
z*8(BOMZpQDV6&33^C*5^_?y%4-n4(6{J9e!bVg1}Tc0R?ZzlteJ6eSlF1zr?H6}TB
z=!M}gE@NgSai+<&HxawTxU1UEVJL<h6;RJ<3hu+$WZD?#0L{p#jvdc5_1ikYCxm)j
zvL@W1jKXjhM$H&ILC*#<(;^Yt*zzc5?C=XY6Q!+`&gkXS0sbSGl;BB5x?f{QHfg}W
z+!7Q{dD_g|?8J^-wYxj+2RYrQHyzCv2ln~ReB<{Pnpt|}lEzMCGP8N|0ATDs#KeB<
z(bkQN?CooO3hJ*%$XA|ug`?&@-uw6`^t(NRa)j`5;^C0C*G7gRVqRcMGk~(JL@6<T
zz)ZP<XAG&zG4T*W526r1w0Dq4W@}djR0&*(yzlliMUYZP5IGm(J4ws2)V-ZYlADw)
zi;3fmPV)qbqFO0o@-I?MOUQ?hf63!B5G2MFU7b0Ym;l9wKcE$8d@0fRC}9T!2`s&}
zD6pm&QUnzw>CrJ;@4ZiDeva9o46}8FwDUD=#45>^8X5j+Np}mQ6oz7oS@nRUv3p(^
zgDD19k+l@2N_a@m$Mdjh+{89)P^$4zguX3o3(vy~BK&fNa(#xX%E^<2*MIVJ?tXTI
zXTI_>ODn6C!xD{Jms+KcCncWt&;w1<O-Q>L?N*0&r_J8MK8=|=ue|yarD};F3P`gA
zWjs7D1f5ZeX1Mh9S1C7^F&J{CM?tA{7|*2PvyGYVygY&N89QDcc>W6EK{a8V?YXD0
z@?(P~##&Ghe#kW3G$}8x5u^&TT;UrisG7!I1cRLlr&M8C$NN3W%9~`|jmQ=g$2T=)
z(Jf=9fB!JH-XqVO({+208)W}(-1bNNrW88QEhLNpEk{nnemAx*G+9CzH`G6N6_Xby
zgDo()&deu1rU#dU)*b>&S-?jjP$M7@Bzk0o$HxC*u-3FYD&*K1y<<9~5T{;UY~hQz
z$I0x6i9MX$f}H3^nN0o0`=&80R9E=^xBrABYjN%Ndvx19W`d9)49Sw1X7k9-vzcEY
z3PZ9y0iYBG_yR)TBUL$Ro?()heCQ!{jwd}r-%c_05rl-U+Xj4_#{d!^i9edZYvto}
zlxrRwGSprA!vSfU&~Eq0@|?A`HR8_h7XtMrPShjUV+Ts1B~p7x4WUUWGb9j-d6^MQ
zA4P<cfyLpK1!|439n)w*W;CgUl8^*SaB_E-`yYNvG&4u39O4HNDBLV$iBh&(x#$@x
zpsP`0Jc(CN1|3C&q!~2_Qovr$I2O=~CqN5{^gX06?VN8R@#NTf6zm~k>;qB?LxRA=
z3v4={8%?<q;ZsJ35ih;`3Q4EWOs&S!;vz|==_QK6phd6SMWIO2A;V#xLAS$6v&q|U
zzeT1MfBI+th~@PSyg=gnlEGl;_SpxcsMTi)q6Vq<5F)iw^vceBDCS_;b5mS5Z$cCi
zFmkFEP@FCy$D{*Uu!)M-73U0F1eZ?!-u_zHdIi`n`l9}Uc%-o^ps!+P#pL)|gCJ9o
z<OPr4b<!C75NoZ-Gez!CD%WaT?~u0Uq4ebOKN$T^(wHavTX3IKWZO@QncL#M#&bPE
z_EX1jhZIE(#5przFeXRkDOyGN_-L(c&|<I@W(6uYV|xp>_VFYR#FKb_V4*}I5C(K^
z(76-I`y0Sm57(hGF4`T_#>ZfknK12Q>bH{)VI1k(E(PrzQE?{6&cRkoCz}{?e=A-+
z?M5cA|D=N`T-5k8<f&~=-!%eUmSq`EAh3OJ0*}S&S-$%G_qcucE}iDF9WZ9}sG!|R
z6HtmutvZ^X83>;s3J6LeLP}CM&NvK10x!h(0|H@t$s^ZQ9#Ep>#?&ejY)-8)1PDsx
zq@=_st9nJOdt`Y`mMc2Fm|l`IJGad2+%j74+Y#5l#`V|?L~4VV8=}Z#bG6E)x*hW#
zppZJWB67wMpooCJkYv(~T7Gg2xe(;|2rbcZ#`dSzs4vd2F<U2^T|_Hu?4^;QHAX3$
z#f<m`w&T6f<3WIBOSuwWF=n@zvgcqLoD-2kjQWqHbP<Lzc)m4PLK>vN^@5D!60J2>
z-QUM6&*7I!Bw3$wSa*dfMHED22xeE-c<m2<pZ4(qL8Hv-xu?jqq+Xxrcz2gD@)-_$
z#O)^CqkTSk{~ca>^?5EneVLUrXEBvBwYfRccAGrS$nzndZ^-(Jg{4)zs7$UTf$>l(
zM=IN(W{<oOUoJ8(MvkMHS0N@cg{gr&?!|?a^TT3kn9lAdqjh1dewD8a;fV*W)O1ik
zku6LZrA$duCGdTXZ$N0_sNxp$g|aoUI<<8X3<w1#L*QF#xDbvSTVzT$lrNxBDJ)uF
z{Ci}kTz2GQjn7@qhdS8R-cN;Y)7V^{ifl~*AG*%;g4AXv!4yYiD4a(@;e7(1W|@rB
zKv9tbP0Nvd(KQK4onF&5j7eascu#@<o6**$_HgvmN!M+i$@m?U=6|}7E+#%#yVpLB
zEFdP{KmK~$tBZZJudxNGLSRMUb9VJIYb#H2?Ov01>%^XK1d&!4BT18l>aa$&TEh<m
zvN)qwuTm?Q(E>{43Q~H6lw1+q9#sjSGBrXfBVk32B1SmCP~%euB2tud@rI!pFid;I
zgCQp;E&BZ-bBimiENy^Nr11cirjIz{FSi~~<JB3JM#hV;T;%Ds=cx%FA-6F`AhZ*l
zFfb_xY}~^t5KyG((z12}e1<0jwm!H<WoeD_^HuVwgc3QP5O`XU=Q&Diq%cT3DQT2R
zNQ`wP8cWxlyfsI~M{DegsM{dG*sf%xzzB~(SW2B#9=d1f<pbhjpLiJ4@ArvgD<ht2
zt3wq@Ltr|rEd^*TK^yWcBk&YLg3>vaY8gGKvvX^UVd^ovuu72m)Vz`%>XI0;#Bg$W
z!c2XZ#f5oF)iOd#%B2bz=yW^u`vVZRu&lJl%Gw!Avo)GrTMF!?2;Wq!HVNN|Y5L}L
zcexq;Yz*6o2aJ93XTflLV4bonMw~JY6V7I&R(twAfYOFEGep8zX?KbyLNzk*N~wuD
zh<3)YnD>FDo15`*+0QpFOS5ZIjzzd*8fBYP5a^SwV!%;RI376;G4jaHaTJY$1=Oa|
zTDYI<5i89q5x5O19_an>eLK0ZV6cskt6;*3(?|_P=Ke+3)U)-_V;}S7ihEE13QaMZ
zc5LLw=6ZT6;kaXROyiy22MdVVsDGS8SR){qRNDG{Pg^*+Cp`l;jP;zG@<#71X0}(n
z8J3ndxU*j&RzuqD9tMvYBvDi*9u7&8gd|R=H5zyhUaXXA1Yv}6+ZO~byD3^D11xA2
z;SnM5P)?LYp$SJ>TSi7gPKGiu{Xv&j`-D!bO|Ls(X>pm&rA>TK5@!R_bU>PZPEq3v
zuE%O&l}hl+*U$02KY5W#pfSBYg#RB&T}BYNo5jfSp<}Rfq3uTOgX3g2q!yI5NB8a_
zTW^0zb*6!TZVg?k;AsyJ&o+-?**zdIZh~OJlqzl)Rd76!6h(1ka6*EfEsVRusOR~1
z9!cQaq%_s^S}jhFI-DFf>GoRmdM$?WkT^+5l9(*bh*e6iG``YoEC*CxyGSjnfG6++
zpFGPb<}e5ejar>ZR#E+wXx<}jwiui=Njg0S!#@4E%b?XIQ8`+BRAzljr4nHf&}(;S
zwc7OK0a|!;<Cw<cI&(|w<athRa)c8PoWlHJ>KQMx3Y)(@=qRjXeV9=g7q(9`v_^k_
z)V=b^SZ?gtY=&S9+oxZdD@am9#TRG+M!I91m=99aL8ha}Q!)Q3lEs*j#>BhK2%ekV
zw=cRR6V7>RUrtX}7!wc*Sry~51GKi+HqR|hSxqz&p2|LifNHIb$`$>7JcSloHr(kN
zjx*TWP>c-zB1dVBcL860-nC~``<Tvl<Y+6?1@`gzFO2$hKBGtPAKzVtv!1B;jWUj@
zeEETW)Qp||IIJA!HcV`D`pB%~5Tm->q(}GhWyHel5>ZegP1+<$%F$61r3Evy1~2eQ
z(gFR!fHcdf*6SD{>GX%p1!bh`tt&8_fQXR5sg4D1HhD-P+TyvcbQemKk&-eXF@5^O
zHhcSfoE#mYwPtN?lg*X0)Pf2zJ+z9+ayyda*SQ|c!a9F(iEn@RD{P)y;G}s>6wDG;
zDg;Y&q+XY(pppvfG(6>C0tB9r2qQ2?*~Tz@q_LDUGR5)rTkI~+vR)5~N<I)G{Qxbk
ztOcUz(QtIz33v(NBTYe3^@<+?63+{ep0IMu(nE;AX(CCIe#&vP$=>!3Cr2mr`vcNA
zqun{7)j8&*)uPkx(CM~`<Ag-TXyf5)IJa8n()kASvo*ZXkfa`g7vaeOVcY95LeQA4
zBV-9R=r9}%z+@Pm5CsBdJc7_C$)xMf5Ag$^Y#7ryY0>F+$h2aZWh7eg(sM84g>|dL
zC5(eYruC~%*BuH6A_Sv35il8*k6BbOruj;ju}nIT!iZ01FC!LI5rOKlLnw^+xaQ&_
zwt^5LaC71r7ygBD^VA{BA@H0-c6BW~P0fiWs!6$8Gol?oZW*JT?v?%WJOM(EXtYIf
z{fjIy5!Dw4a}2qT&&7iT4-2yytuV$LnN8)y4{fH&L$i-C449c|&~0}fC^h+>PZ&m{
zfT@^MUli6S>C{iU?4FG!)-CzOOxb5U4REraWn8pu151yhYkabXF@~XFTuT`{tMRrh
z3raB7yc)w-@{~3fZtLT{7gIkifL{A=dy*dxd6}tqO>Wm_3B|SrQkGa=+@w~Wqn|bL
zeUIT_$o~Ewy>6G;`2{<=J5R~-oH$9Dot+~L%Shj+UYnsTBL`vGIfDTafxD{|+PN2-
ziBti_5aXCu840P0Iq4j+zq>>8_=uTCgN^k~*5)>;$O_t3PO>aPU<ktpCq@2E)}u$-
zdXkr3d5R0?&mct(CcqC$1mzN|&ppl7hgUJ3mW?U2yG^8LVFwJJlt^t2jxvT!8av3y
zv8nP-mz|G3rn*pPrCdhWXVHGZXhO7<7*X7Kt|%^CtYU=oFVn?3ka*I!4b`4;{Ruus
zOOB6QY~8z0>-dEJphv6SWM_Aa+jnnqe`kwkyG5F$He;HKxP(hGXF)=@-@*?Yc%DZd
z=3oTUv-6=<Su)6mcv@qIkmf_Q%F%(HVV>nVlmz8+nNqEU2&|A~uifS3xJkd?C(Uy@
z{VuB;PjlwNGbG9->Ly2P=~!w-k@hj^xEVWvN#p;hcN7NVbOb+wVAMnhwU#4iCPvQt
z(fcwcKZudT8dI$8KIIDX+@M4Icuy+sYHbkGjP~Esh2?0Qo+m7U;uI~{7)`EntGOnN
zVy>Z(!!Q8u+-Rex9P8tjp@_Jrv*IyK<Bpp#lqeia&3=xin80YX{aHcfbsK-+be2jH
zGmQrQPVZqS*nYavNa5p2&(W)=Ya?G|jjI^b!NBMo7~8f6+M;&uzd|bAM)_v+eXKTB
zTz5Un3LeaIreUectid{_A}g456yg!vI=#jj(8JHnq~j3}`OW0-CG)ja=4V&gYu+XB
zd~|3?RmRCloBm)xREh|rkUUpbU0darW$JsB%4N#3jPK5gA|plP5hC%(NRcGY0!c^L
zg`A85T?VF4XV7N*{uU=EO%@mCxODCkGnF|iqDn|aN-|RLP`S-`f1T@*4#f98o_X;C
zjhQ<AtdBHKxzYDoeeP-2U(g)C^`@mK3W0E^!8U_g?J68!U=;s};>HsAGN6BO%+|Xf
zQmWS}U%rg4&036CTeV-CeOQLj$V~``rF-_-FCt)pL^|F*o^ddUgxq6yXP?`*?=Tz=
z81(vFzj2i}-+i6?+jkkJ$wVIoqlck@Z=`Y47ctc8b;2N`-|gcIpP|>M6qNC$B=93-
z$-@gY()g%c;d>r&l8|K?S(cHI5LN>G$hR`?O4I3d=(YQFdL43=(H-;&%QL+E$~TcR
zATw4LTnKF?;XOBiXTs2HV@CKYj7gJI-P*-_giYBCM_gd`_R5Tllh#4F=x}oI(I~4i
zqp12}t6*#u2DN}Hgs`*Jy;1kSHJZ7GR9mzg1Ipk82d#k>8b|YXM%Ij@`{LS-0*r|-
zDJ906++QqvN=i(ot?Y@EE(3c~V{7XeMHV$SaO2+<j5DnbxwayTT18gYU1~7n{uHCf
z-$%}4A}R8G$>QQ1GmROx+2{01(}dPZi@DhbjruG_4d$@{#^<i-py0ut67zr~oh)=~
z_tY1zLsD?5if!0RNimH=jK~PZmW+FMaR-Ts{~WySPS&)J)x(Z?T!|^RH(4G(?YzFo
zs+H=@%`75}?RN@7gVa7+Yx=!D-A;$F6j3gh=!XG=;eaF_G8p#JCMTa8qDu4lBElE8
z-y|VLxB@{!f=>qzi9wMvU}$=@k`tOI#~kc!GwAhLTv*`g3zwKJ&siNVssx1O7^^0M
z=W30=B^H+FNfy^;(Y`_C22To%XS0ZKae<Amy+FEq2X$``<6MVie4`m5FhY_FjVkD(
zM%(FFLLi9nQ^U#4E%p|!u|7YKS1MtQN1i$rN#Uj!V&yudv9%Bjdk701Nmou20^xZW
zft@`iJ$zN-V5h~6t9Ka0Z4O%p{N$HE<D(BhBu*cU<i^iip{!aI2W^V2lwomZh26a^
zx}uLnBFiAc8c|S2iWH3^3abbN!?=g^1gQ*&v?eS^D5EgK&il?&MK|fu?RCgaPN(0Z
z)#-5Ixo<PSutBCY(r60_2nWd(-qaZ?Yp3m)S}CnsK+)88`cbm*k~PxfUI}5Hq%<P~
zJeD4r!0Ze2XxxyZQ_tMkvAF2o6i|<}LYTsJg}Z`!;Xou1Ms`+AqBXgK(8<kcH*V72
zGX|}#tiz}g5J{oqB7K3*u@q;FgX-+xZ4@pXz1hm9*g4e#WvoGwMPTwI%it6MChIF=
z^k*?RN;`*V^tcOMD~<LwV$=gtI1iTrJtX^>ooleUxj__`9-1YkX-2Evrqk(Bt2P{o
zW#Zy|L8EAoNK@279yZl)?5GQeF)pMROOKu&`(-=6^r%WSDI78meTr?5?l|{%F|j|2
zqP|juJ`XyN(;3YqR#C7NPCJn=cAomfA)np5#s2;wS!NY&J*Np+iZq^-#L19&7&933
z?bw4(NWa&i+if$9d$bo9Sy*18TCGzFszhE$C_Dr{3OmExIQn{)C$xG;bbBW>51aHm
zeM&*V=EgefE6dE6=P6Mpq)NmLmIXHiV=zLZjP1Ajb*@L(K!m{e?XRsNuzv#(o}{+1
z%KA$$aKF`s_7KlS!LA%I`g#~Nf}?+WNU|hm``UHFwN+MYbxc$u6CPSAr0-8<9)((~
zaMV*=|A-Bu9aJETC^Mkn8F2O5C&Y2g?%odn<}d#Zcen4_EeZ*blDS|E_Y}mrOy$IB
zpCFPfEiZHD-YwE3K`wbnuSBkLe9t4#G`=V7l(ImRWs1O)HtG}}O6SB`%%C5WBq^P4
z8zc<lA+1ga;rYxjtRiH9&@OV5V+P2>ZgXQ?Henue3=>(%D1$h4J&K=eGi?-%%LX6Z
z#so)46x7^@ol7x>uO@c7=x_(y!?3inY$dTfEwWtW2VjgC^^O!7Pcc87(H`3Ej9FRZ
zjASxvnrs#{&iNIO(dmx%&s#-&temvI3yr4s$!-fL-f1%<DKr=>CJ8o>GQyEVjUiV#
z<{{2Q1aSVsCTGu{Cy7(4)e5b4d+Ie&DIkb^>(G2j5ZciI?!n=~JLXB2HT#}}`W^-{
z*_qm7m}}Z;`J+ssi227`EpmD$-gheF8Nq{v^BB7m<><7hA*DyJamV`j)diIH#m|qz
zuy=IG$Ddr|`rYfa`W-~#p>;-B^ALd{6oFMK5Vp3F!~>Efp*v_Z==5kFH)$Q8Ff%vD
z;_3=Zi;GknRiYpwbTFjSkR%ydmXoA0t@a7cqeGH7p%jEHFD!EQ%o*0#)|riJlqnNZ
zBB0`qFT=wib&k-Qem|$(`7QQ=*ay;4Xh#KBXJE=@=AL_o;X#YNH$OyV?tw8*P!ZQ;
zWo)6<@qI_z^)yx!MtXFQ+U$M&Df5eqi1ih6Bq~qwqtH1dJLa@tn&=7mKu8Z+bgmmK
z%&0X;74YdtACSaD_K$Y?|Nh0l;qK0zQ3O9(R35R9a5CY(j~{p#m08BO^r%)nv`WxA
zWo~Yk+?->3>pojMJ8Z7|gno&@D-lIy!k~ob1qk601QEUnh?5wp4VhN-dolg~kds!E
zT;*sC&DM!+Dyr0|Hx@v6NU7cQMkm!F?Bi3wctR9ef@RxGkHQ{ndW~5?<E-WMR6<jv
z5zgq3;VMQ@%ZTMPVM*+^i{6UqWz2-mlNlRm=`ecS-BfQhNP>Vtt3{SW=!T3HQM)lN
z9ya6Su-AUVLBYsLOcuq9eV^1<dvHG;y!C{3L>o<_GXh5dv;|O~sSVh&9YR>WC?Sk%
z9CY}URCdaxP~^stWtu#h-m@U`dG5ta{MkSGr(C@7H2vO?mDM%wZ{KGS5AdWQiULaI
z5My$DKcH5t5tT}o0`GZNF!W0}5Q!kbDoq$5Ct#8!U^BLYz<hf9<K|zBAAmqR%BmRs
zs~`iPz9Q18JQcr_cw?A(%B@TmgdcT`r?ZmD@OE;2{y9=wbKGt7?nm!)^Y%TutsbqD
zCi@3l2!Fubl0=shyvRd(fo)hOpd=!+jL1yNAQ>_k4(K*JoE)5>(RhJJwNa;BE)fV1
zosAi7xwcbArH}+cNTXh5ae0|5PhDnXb%mLkS!zLpiYOCN#ixcK!jPdDlF>(C=(l6u
zdE*+boyW{Y{u1kR9Ede4Qiv(Wi!FwFcngcHzVZ^?gBGn@TX;r|ZX1m4<q*bL7K}Ge
zHr?g0STsH*t?A$2;o$10%vD0lVT5r#07Bb>sd6g4Mu5?7?zE6d&$j*-UO;P2;6>cK
zu}$lwO`691=qG>0ot@im&y>qz?D6@5WOaFkFbL_i+oK}a2trXfb<+ZsHF$o=!qN&s
z=yT`REw*=e2!ay6=TR$1_<={Ra<V)_2vA1h`#uIut23b0ZgJE+qBrP~rYXZDMo6D1
zDiM|H%q^@Ty$~ram(wVNMxsR#QIDOUbW!Z|GvSDn&frfvl0qYD^vV&V%g$yWNAPCE
z_!`qsCn9>VSJOEdqYM1|bc7>DMRU)yI5CVz=xU>po@91zj)R>Ya_zW1MSZ|nl7<$_
zWf@|`G%N1W(`>X+L#dv2IN%(L0;S|sb~4IP9```FlddtOnJY&6V5D*65GeW;3cENa
zGT7@=FxL=@ynwBYMr$}--#mA5jUW8MAMg+U<R7xMv_ZGqV|`<tlTMRdyP-zDq}T1R
zySqcX)54PxVOSaqTz<)6QZAetnQGsOX%x$rbVpUs(X%}&y_+#WWkkWG(vGD$$!;n@
zr6R*F8hFJ72W(PlON@RpZS+R~;zN#Pgom7Z&y%k>OS!&vn~y&EnEUs4hzAMB2gkhm
z`Y-tO(+{}v)fK+=y^D6npHUzszHb?#7QFHBMTt_mM5SCI8uV#*+I0Ipdc7WphezbO
z9aBJVC`A$FQkiPC%EICzvvV^n&M&aIw8ZNADrYx0saI-5UPvSYBCIlobsh#yN=}YK
z(CaJSd-pSb`LmD7hQFmEI=F|^H7AXF4gjgaO9jdg2shVQ{^~O%$1UP!)25b1gy=%p
z5o9s}5hyG!=K;PEr2PTA*RP{j7g?K~L+K2o&B&OK#D@?tL7N?y+3bGiDkqB*?%v);
zE6qnAf5@kweM|v803)T?JlCwPt?}J&eTz=3%?IzlPpds3iCw3+FeqhQ<DlmxJUsj$
z!t($#!<lpExO@8^pIyJk`tm9ZvkR1>s*~%m#c!1<o4)ot?CtHcy}Qjx`-C`6$+adb
zRfwW8Mo5C7g!Cf3AhZo*P9w(%$fcgh25fZe?oRh}Ga48*D(y}hUB(Xl0psZyZv_^w
zl(G6u4`F3Zh5@5%`v@4Txa7uVH%ckm?G{NA1E7_aY+77gVy9LoZMVn*O;FSaiXysk
z;);cKl(MtNjWkovZIrPT4SMeTxTz7#W}7&+FRFuuhAb~e1PYAr$N`QY1i0G9xX^5C
zR<6Wh?C>m|=a3tWf@F}9499`N%H|xu|9ju#d*A+Do_gvT>h)P#?G^^jYPRAg5m+<3
zx3@!_3`w&LUxq|sNErC`*?GdU&gqqFJ6%FL6rxbXb5|8H0Wz7X`$5|}o6f;Kzm`xl
zNyIfHCo)L_FAC&igI{2JV}0Do*D~>`#b-Zwz~Ui)HxK#U!<RB}*zNG<yKiyr+Es>w
zA^lFD&p!Q>pa1ly9PjU;a>dbZNVhj6YLqd#03QNhBBX5smcGEq5GeyHl^TskgZa5R
z27>{E;eckVNw3?dQmL>oJI~U>GP5(Y%+1ZQu((LITBcGi(`d}HurN=xR6z=l(8-;6
zR1s9M*<XSIhC1V&Pp|U7{lEX5zx;PU<KTMdOWzNVT90=i(iv;4VJgmdK|d6Ls8m>d
z_G$9HCdaS82}x$56^brS`}io~I2O*3XkpnXzOWMTfsu6gns_&EGqbrtn#ELY%@yq#
zCs#3+rxrzXVKM(1fzB2EPM2;wq1)+@W&=LE^(mQ?MVCn10jg8dOi`<s_{KM0V=(Bl
zySGJu*dtA|5p0vY$T5(DG|?onldzUnh5$4(#&G7W;qKl0y!YXIL}AF{!V0BQnWz*X
zg+Zq|-CmcygMIc7_i1(77{HH8l*@IZD8LIO!(7wqCH(B?uk+%IFR;0>2(B=$wT%Z0
z`)#oBorkeBJ!3|Xaxp3u+#@)GftZ4OnD{;#gK8|h&2Fur3+t)ZGO7=`EN4WoE*!~(
z<1vh`k0&KkK%S+9K|m?Ag~C!;Vr_k$=HUUiJ~<)HHIdfN2x<EzT*rV>E(5b)qWBC0
zI#-xN!B;!@PiyR)5!+`mhN^@Uo*Wx(t@RVdr`p)TP@a(FZsdXI+w4uD9IVKgUFEWH
z<)A$6sJkcyT9GF?@gO4|*z>%!I>Yb&?ss|Zm2WXWw`?Io1sFSP-pJ5GXF?!GQz-3m
zc)o`pc=&-oh8(}(GS~nj%{5*Co~MyQb2>jOAf}=UqS20~3|d>GuZ=+%SBe(}|F3W&
z<Ft9~lE&q3;i$e7!xko-3p|)roc{Zxk8pZD{HhPLoR4na=H2(-XKU*|I?vhN-sa7p
z|BUNbKP4UvMz1|-w%Ol1VRoU0@PwU8=tW3RBSUSKGen3Nl<YM$Fy}J0B+cmcduVM~
zU0!8nX_dvfC8AP^QdFT*s}Mz@ML0?|s(xj(-#$JiDu7SSfKI<duY1D1!);#w@NNEI
z|I1(S{?D$X`j0sse}nb8HIM>pFC2*gGLw^tjFRUO`XOd%nWa~rr?$66^X5H9QG@Uh
zBDXgrwz<~F!Y=lCPzb5<OpX!(GBNaS+@<-{1`8K2B9%l7f%0>Nak6Wc>SaX}h08sJ
zE$Hp+@3XhRizh;oe8@p_k2_m;EX~#PXf$R?;+Qy7%*>Sdm;dtr#Ny&IKl;&Maoj$l
zKTODtW$0+*x}IGGHXLX^y?Tf5edlYWD#2F~<#H9RHT5~o`M?%0Z{NDjt*yJn!`Ma1
znm|g@Bqh!gyfC0rt5UAk@PmkefY38&k&-CGz3pw<=`8J5hbvbu@xt?0C`AF7yrA5V
zVn4foW@OMQenyeD;~s36l{jZB-K=nJY(D}NPDeP1M%f6tc&NoG6<h5dX(185hx9C2
zTi{?zF<FBUU@T_tOOI-$%9)K#E?szvG|5ofu)4m%Q<tA&d2x=`(JsxyLqcVhhmCR>
ztuP?8`zqK^0*vheQF)He@{!a_;Y5tG`fB!mw#O$tY0n!r8?{Y*xobFlf@MJG8YSH<
zX{C@dciXpDKpRDtXI7#_+3S-REG$oBjKnBSnx^zRJ$kJkDpkzQS9s~GFY?tFUS(-%
zjWDcIh-XP*rG0V@(zExy)`m1qtwVEl9_iWFO-t20`4R|1FC<kNNgCq^ioo~q#rQ58
z|EEWPf<c*rd^>#)7QxNrJE3R@96OKc{V19d$3J=4-cG*0*t-YUWY+Qd=j5o@=G~94
z^6vYeFdX#Bvy6}5f1kI1`AfF%-hSZq?W34$AK#}|FSE9}Ooa#|O5p04nhYUAJQ-0c
zRq#A-q`zkXU-~rabyk+vs8?pGlq>i_KoA6kVTjg-EKL}8GMfDZI=wbYmXQq;I^8bK
zZj0^1eYUoaY3`k{b!UrPx3)OE+k3KsWM6W9&P2ioA1Gt65B3BE;)Ma#bL*UY>3Ndl
z7W%kn`!_tmj60J@HF$T!*qO~>j6@pC;u@TEIlg_L3orH1iyAy1BjpGhFxntS%~8TO
z4hl(@WW4#-o2)FZkY_16&pA3gpwsQ(Nx|acyoFEFJ}YZ;{PTbIXFT`(^Zeys{2!!g
zN}A?mS&q?iLTt~SBxz;sfAr(G*jQiZ&;IeBP>vu7eX8Xud6pq6b<SP5K)G6@d2+%@
zv&q510VmBS&E^qd5Hd46Lw%+}sa&?YTx0~sDY61_obr<&{|qnsm_PXb4~XL-hX+Tz
z_O-9Gv@nY{{fVw>ijz6kgEEXF&WEv&CK*)1G1$ft>R9Md8{7P;l|rKNB-4gyB$aOq
zsku@pXCzNQo5kC#;9o$wT%lSm^VCyMGdnX!o-2Z=%=(!#tgI}tbN?<s{15-0VUm;O
z6A%;{^t5!lF~&)p<fC4j$w5;OW1)d70L$^QPc}(D*>Vkm&KYP|xVOsA_7&5_g)y9^
zv?0qf1U9JXwmR(Z9?)wKSY27*wXeO*D_{E-7tcS#^5O=oD;w17HT*#0`5wyYXGzae
zrL!z!7{|6jOiaMEUZIcWJ&C=1V0qG`7S`w_nk*e6$uYrX*ZH`IeB?@uVcg5{Bx~H$
z@$h$i(H$2Pe?DqGz7C>|=IZ@B{Nk<Gxq1B#Xu+V@<*$GELq2`~1Crt3q3=@(+<5Pp
zqkHeNxH-?63+tS}c%H^wnNlS2qX<6;kba1uOgXGjDwPR5-w}5LN>PMP1CH8VTJ3FG
ztv0Pzi-W@hn$0H1M@?EM1A6_KUVlgyXC(cUL9fqXIAqY*kbJ>Y_BULgGY{n9o?JUQ
z3QxL7SP1f3nU&`*GuS@h?i=shsLy!}GnRWDEn(~oY|r&D2m_t~FB1&54jCREkzdFV
zzLjVwZ03|5L*PrFJWJWxzRxEgeaM*~oT1<80?=)Dz-Sg0W|?VJxpVg}7cXt_&;R*<
z%lE$bJ^uB-{?}}6-J{d#(`gSz(eh-1?0Le~AUyV)J^pY1kAK7U>$mwQfBHw9KX;a;
zrA5M01L>EEYEW)0usrOOBq{y=fNrPF&i%U#;{l~o$+6iWNn?8sO=#-?T5^1RM7uMf
zIc(AV-%t4c-}ydi@*?flU-OM`yvEtH3n(X&R1~%+ABfXpieuK3JY`Jf0}QPKY8wB$
zpv<Zwe;L;b91Nt56+tutw6IO1V!~j9b87Cp@A>$ik0HbJBvBOMdj>BoVYDC&tGx2+
zx7gd;=lu`fraw%{jYb>gq-zReJTB}{vb82zTaz`B!a<BCO*+bKaej*5KWQg=ac|%x
zLPib>JJC)VC$~?rzlto+3EjD{eIJT$r^C^~0nOuMW@hHNa`{=l_O)+t{^C=d**wSD
z&GRfS&*KM<^6ZoqG@48)gz%<dN2Lr=FbN5L3C}UHJSnM%HF}xt=+{z>*smImG73GR
zx$@`{!gdwoehZnC@%+Ny#5%oI^N_z!zoxG;KE3`K|LL!P%<=INo`yTOZu8bJe#Ncp
zH>{M-W0uL`<RIqcaEJTXj`*N{lhRzs(prr(=a-pTo?&KghJGWXQOodUpLm$iX|_2!
zY;t^j#Nqx4t)qlNGqpmDNy1<_AWbv!j0Yrc{<hcS)p*U=C>c}Cn=T$WD-~TlJlYfZ
zb2BVGcZvP&9Xi{Gc&-3FmZ5Mrw}DYBxp8w9Oi?xQNShr72Zxr*i@-CUZ9H?4BhEM~
zmGjxH+pMiFFz6pst@ucxkU|i80a|Gq)e7rtE9@WabLQ+S|LmXr8Nc_tzt7J09w)~g
zk~F8=>60bO70HX@oporD=ix~!+mLG*=7t~r>`mTy>pf=YW?5dEXK`tfnVAOFMveLT
zdFJQmsW%#wOA#}RYs}0q(mrXD#tB(wCzoXzWZEOw5hv|Fcen2I)_YeO#2FrN(rNPV
z{_@|_Y&QAMH^0k|fBa)!edPt7d;S?fpmc^1V@&<D<9L_@ahG%?eUB?l3sI>2swv}d
zUo)MFPBNa9Qe?TNF*i%8T(uFfo{sjdud#g~LR#q2Xh@R8NqE?u^8JvoRAXUzjjw&<
zyL1M9PC9+M{T`KS6@ze%d1I$xvn8gA<wev#MFDmc<dMV3tz6Ay&xkCj(oba0grfyp
zhM+Iy1TEQSCVB20moXNu0<<!y+|Ip7lZ4~u5y!{J1YXE9SDxj<g^OIgbeWCKbDTYU
zftBS&f@pH8W+5Z*np{V<AQym1o26K(n}Xo*H?ugt=Tnyr`f1<xXci7c+bney#E6Oe
z#<<Q6F=@2NG9+W2tly05_&liVVXK$M{QTWFc;n4?XdX4uMsxq>b$<Sze$3(i;TNkV
z8HoESaX*DU*tv1Q)i>|s&3Ke&e5w`8u*%Y$*3po(nUSS7gU)mFAd&I!@Oq3I?>I7>
zJo&c3rd=VoxPb(|^2j`&(&h^5uRh11`v&=8&#{8s!zst~Jz=DUBs550BD711rH2{}
zY3}Y35Bt>T7Llrefh|r`u!g#W9@nmY#M765fHp&v>NAKt_<o2|hDNQ%+09MnW*hW+
zZT|T`|8Mxt_kM?Rsm2E%d`gx<yVbYwl^W9yLEw4Tc?7~oNI|AEQllVCpcIl0a!v+^
z+}=4v>yO=7QG1-e?^7vPs8+2*SzTUWX>pE~#Rcls8iQV+lam(r_K&!?)#U!pKD&np
zq`4cx>S|A^O8Dt7eoVJ_!tZ_e_xZ&ydJKm#Uw!F$JkPV&!5PBStg1=<smXo`ZG@Gm
zlhzR!p)BjrU0r+a&6L9tV2i+}_$)PYT-E9f3yaJ6QHV0?6zgidih{ckjD!PKt|12z
z3H-3cLZwQQTIrWz8guR1RrZe#Xv{RIL?tj|`Smf2YJy(O*x@Oq$n$*UJfxH(HkQ&F
zrNu-bqQ-LbPqYe(ff&^tilV%9HKSs`v=JC1LHn3Ylj|W#nsL%<vA=gnu3+=b1uk5?
z#QORsmo8mqef<neOLO@Cv`VZpRvrbDTW2plv{6ppT_B|o(zp6=rjV@Hd}$`HTb}T#
z`ZFj~D-Iu#dq^WuPK~;FbS&s#Y;9AH9e2@=Xv4_ybM|*(?X{cy@aI3}%{Sk{XhW8y
zy!HC)eDKB_Q-$&`yL4t4W+`-`w*`lEe+_&6cVm6dO!P4=ZYm`<4u>}y*DYw7LV&E-
z*?8s|`uF#_fAS$-RuJ%9&#GhW2xG{l?ei{U2n=}G&htn~;=N<ylP2}$b(9D&*dn3l
zxh)w(ty*PeWr^*3cX{`%*E!tZ;pF6q*_n0HG+}jhmE5#x4-WXwcfQ6eFTYHqR!6Cv
zJj*yZIG}yfu@Ssd1fGxYTRko534Ax$NGm%<FV|o)Mc~=-k<v{X6Mp0(O%TGM6m)wr
zCntSw+}d(K7nDjN)p8ZBG)a=u9>yfOle-px6ewe57VLYhA%5?}_ZSX`{NeBY5pTTt
z4r!M1jc>d{5O{WKsq#^%Q50C9Om;qmyKX`tJPRBcGwu;7_!0JTcG+angCoW@5nSS7
zn>uauS*p~Slgorr$-c%lq)as^7EZ^7RZ{v0N9z>ApdCBg11hB|>#LiXOp&@eLG$E@
zqvK<i7Z;qoyT$(PWJ7z+q~{^ELby{<h{v67fE@3ObpgWIUIU%y;CVc}oBs2baS55A
zn9MbDa3eaLL@Ps@Wn@WAyVK|BxJi~NW@hGDU*F`+nR9Gzp5@G$^UTfFA8;PHCoxw!
zd6MA;VnlLr0HYhvY`=Ubrf8LUzr<mkTAqh%mUJ*&t}XKnk6hdPLjax_&m|E;Gg$^7
z^P|Sa_`hu@vfDb~KmF{ldHb#R?fsDFy!oqN^4{xjkj3%e>c0KwupSR#O<?W9&Djv(
zA}q&<aUI~JJ_J9azO>BQ=dN&g{~qyfAE}DrG8QKlLgNWRjz{ihsryJg5g<)QXgrer
zCcW+Z%wM>KiF{jba}w=H8*O=kM-Z0T+dE{oF~{=K3it1CvAA#sV-%%Qg~m*cQa$C_
z=bxrru8?OL+uJ)d>Ls@B-9_iNad38~>Wa~Nq@h&w!K9f%E87i^vUE{<4($0TPy*`+
zth7d<7#jousgWePVmQb+>@*)<St<$z6O;*KY`;txgnWAQI>re8_y>Q&+wZ&w!tkB%
zykaE}u)>wb<nF;WuGa;GCtVYpL`!STq?s(97kX0TLb8RMoZ67UvC`bbE+t87X~mUl
z1K)2TW#-1G`q<u=>0-AKxW0vg^x%)w338pI@(h$gYDukJV{_w-4K}pl=FRII9iJei
zq*kj~C18sEpOiJ3h=QdpA`2xDdJI!d$>3u~FpO3TltioiiR@Y7kPWWOy_kyWo)?uX
zRGv8|n3X9R#xX+!alcQi)h5d|^YcrruC8<O;w3IzxWvZBCiC;N<3#<`{gEq0mZvDK
z?M!R~+A+uwcB-az9R9d*Q+WJZJjm3}K$dHSHiV&%@H{jcQ7B+fo+)9!PrHvX3+r#o
zeP|3@hdccFfBg4+cI^fdk6x$EuYUG(K6vLn3$Opz^||YF9LPADv@pN~ly$0DYE6?B
zP|Q52ZLYETmFMZV-a`&@JfrQ{+7S<bT>KNB#1ocm>In}a483-ngPV8Q{A!P=RJR7m
z^T7)+!biCWWPRftzxw%KA^nKCg$3St{XL$!@*+X#q0@kJxyHuQ9OuqinBe&6n2$fY
z%FfO%)oP86jSY^Ek8D<w=dKPSEmoapNH<qO85`jm<7C-=;TTb3WRPsvw>v-Da{qH>
zr#(Z#L^7jdtzr6nKJ*F*L7JvW0oQMQN*H+j;QN2X+wZ(fxg7G^Yp)@Ed$Vg@uwW+O
ztOBlbnS;$N#KiS8QvpD6F^f{HCw&`O8_V2JRZbX|2*S`tps9T9zE3j2th~5~^u(xM
zA%Hx$UHe63TCP+$v#~jXp>EyEIX-TZXE}lNU4MnEcND7L6Kl-2vP!$-!uuGan`}tb
zT0`KEYCB(cjZCANXoA#=EXjy<MxJK$yFIczr(U09eSMR&XU}oz(q%R`&(LVp9_l>6
z23~QJ*^$PspGF%)mRsz6a!R2wllPN(BB%I8mJks55lITl6L^6?nj~2eXT+5Llp`BV
zohdtI@Na9~*}2b;e(~3Q`pIX6zE3h7^2?w8luzFO@V`2bFR&gz6WuusZEQq48Uh4U
z5hj3eg-9hNiP4DJ2CL6q;qc}T$2aaF+(0QK1zHF^;Up^*7+<>Dc+p8M(MihwXSe9>
z?onM_#FHfu65|C38Q@8Xe2)*`eVajl$nxSGo`1q|^N8EGKVy0E9K$%KT&h#8M${^G
z0xx1_W)ZDnFc|W^?|lc~_xbR{54d~x7FlMYqdd2TQ0br>-}kK;o;&VbgLETF3kNcp
zVHL23HBfqLxSTbp1(Y&oLY-cP;T_jD>^46A>=R~Y8+`j4-{qITdWR^g@zt+9hm@{{
zBb=mt0Y@24(GX}Mx9KkIk#Tj*ChQSm9Hx)4#07~V%@jfg%rC4^3QHCWG6mjmKfkbB
zlZ-HX8s+HzQA&^M698GBlcm{|K3BP1<ILtcYk<>~JNItUY@SewB79$r&zUh3=a!MP
z(i2qIsaYl?XQ?0f`Z7&-#pjx$r$*S%)Y$%v;c$o{WjM&l;tVMR78aISUS8$Gg$tZL
zdyz9|&(f$@rvkLotDyg8nLTgX*alOjHCdLDX31DU5o2LJ*J%61rJW}}UP-moz~~Zr
zK0pZYLw~gYj<Ae)1f{nq{qxY@{^r;H!#)1|ul|aA_wM5R0Vd0M{pUa9(+@s!vPJ)O
zJ?VNxykm`s6-+BSm*w;gKJl<whLss%pazNdLTalktXw&Vsz_3k+30vwR<S9pG18V)
zgcubjwXmM8yS2mNt=r_u&=mqL&g`Km$qf2Oy!(^?z~FcfUmBz*2}7SZ-+F_3y@D-<
zmk<Q>dP8=1_BlD}QYuwAX|`y!PUv)6EG;is{i-}8NmBAmjifxJ6!?BHn!^EB-as2r
z+R39kiFA!bX+~e#*g;97jAII8pY8}hy^JwbDi!MWx^p(ds<vv)M<0K{^_!nzfS>&2
zr`)@Dm%vF^6b0R48_vp48BfEsF*PG2Jcb!YkGPNs;iI(GW}Tf~U~Ya9>HF@z6S5tq
zXz(kbuxZrW7~^_I)W{GTV+~^*4^esU3iMWTy;82TxqgmIm!4*0{R~ouG@C8D-41D*
zjw1c3UW<uZhjiLrr)x5W0~uY3qE7UL%fP6{VU78;6Eemc_2FPZtKFt~d_uF?BuR5-
zX686^<~)}#Kf^Q6yuhWWo?@m^d#v*SNVCii97{8s9XrOG%5t(icf<tc#^b69lFJhx
z))!j7^r=_sgh2(R1S$iJ=L8e&P*!_C6}=+kT;#!)!Qaj@z}=%Q{y%^5Z~5%%4NPvV
z^LYK2y!+<6<k@eq<L>Y2dZYuvMZCp>QBW3(b(-@ngdIZVNl73Br10>=5_1>US-7x{
z3T>Z&QAN)|Q9yGeY%xgTc1Q>UArZX+J2!7K>~}5aUP?R(Avsarr?+#R{A7zy-+rBw
zgFX7aE>S7sv+GwmJUXOOse-Wvu+{3Yw|~gN{t;`d>wN25-{$$}o~PIA6UPH)W@<=p
z+>nQxJu=xtQ7l`;EgI)01|}7iZ(+qAg{#*<VPtVImmFmj4?XqH=naQMq9|f!X2!u&
zwhl8G4tV?BH#uoF$?}XJ{_saMn<w~w@PM(@ZhnU??CPoSlXZdO8cIh=9e?>4BuYyv
zl{&L?i+Dja1%VZ8ywhKWW|tkUYBK@lG^ED;88(}Ulf=SF?))29>@QbpoH=ujix-|^
zV|^3f_i4A=BuV0^viS&xnrz$y+zgemu;S^ykjZPR-20!@uoVuZa9qk|o5k$*EX_!=
zgl?}#r`yK&1Lo%!SYO}Z%$c*CJ9nP7wKZzh^5>Q)43KAvG)@@tp0#zxX_}I!YIJN_
zmfM~UmsJXBK6m?Hc1a24piJO}R=z|P)l7*UUOoOQ=)0$3t>5O_?;P-xU;Z^WZ`>g8
z1EMhG+9#jz;XCh>r@vm6+W%74qaBE)e4tf|Hkp&BF0_$6j49v>2esfM3GjvQCiDg1
zg_IYT*?9UXf?7mMW~B~{hZYi~!eGddXyM@tjgbOR2tq9gb3^O?4uhj(gb@UUND&~(
z$?jd_=*NG-MPGCK<Ing{|KZ0BI&Et8Dyi1|>h)hRJ5xs(Njfl~3|ZPI&igE_tn%FR
zU*%8#^uOT`{@{<vGN{$(nVp-p>=w`1;V+?&7X)@6UGpU1Ac_JWaF2vDrjtgj=oLVW
zyTvEhqqK@f(2v%%+byc4i20cYp0GwSSBB<ElXpLO2QT#5+dtu_KYN`_Spz5?lp}11
zpLN>WSlgpC!dM2KF$N<fMoNVA5uT5bK1K#0Li^<fC=7n+)0k}#gg$wmV9=;6M-3BX
zk|MMMQAov)9cIy6BSp~>kE?w-gDpYpj5zL)tD%E~9JJ^5v0k6y?72%^x^#urjZI__
z62}?2Dl!n~bR02I2%FUw_h)ggi$=)P@N03yKXJ5fiaLj@qi990?3c=w9bcNJbh|ye
zoh~3)T3X`FnX^3g)YClu^fO$zaFIr%^2nOUsgEO0GKR?ztz6~I1zTEM>REAYD$lKa
zio5@O-*?k0fBj1Uf#*{W%lOi_`(`}+k#?_@xwl_(^xw3e)o*9@)0n^bPye1bUVk05
zrc#Qycjp$r_~}pSoOB-h-Y+b+`Crg_%<XGXX$nT8bPht`dx3ip{Rsy$u_@t=lx{uW
zM+PCXTw`v1gSm}ma<33NkZ!!K?fuA%!|1Wh90~XuA_Q^snBmR=UM?}dAQUMx-DAS-
zo8)&tC2Tf%_QHAY-`^om6B_jjYin!V`0P^-54LGkYs9^n)<KhgzfF?%DObzP%+8bL
z@DKmNKj9C5@DDNCr&?>Uw7kIFLIW=lR#QkyayKV^lxcuH{^Im$8xfxF1TP}mq6gx%
zv6!qwJn9-4t>|?-EY8m|)2LHWZL=(AYx^EIZeFKSo#B&DuJP%|*X+Zo?Zms$h*Vdb
zFv@PPfQj6WVvrc=VLTu0dEf;IKLkHQ3y)N3!l-N|STMGe+raMj4!5pe<7j(_q}Ro0
z3jvH_v~eafu0<HzOlYhQ6j+J<ZtsLnx5;qWCy7(iEG5^uQ{xph8grb#aG6V&udueZ
ziAInlRs%|nvRapY>1m>X5W;0aVg&O|7Wk(K1y5wyjFa&&&S46QnJBqZWSJt%H0dy>
z9939b-sJ4&1uk8@!c&)@<ILt+2eCfc(n=8z+a$v-#@I+%V$Y|ATh$1zEPUq3=oLTv
z4K7c5gkFGj*{#hw0t6v~2naER4%+f2;H}@zij#z&zWxi|e&=J1DbuLWa&)lEuYU1!
z+Q;q3zSj@rh)(}s*MsY^8HjX}4@ya-2=TpO#Ih*x@8UiXV%k%Co=4#OR_v-2vGUX<
zN(-|X&mc@8k0HP@RV<a<*p77L9EmXKVUPCiE@qe!(xuV8MRfBWs>3F8OSAYU<)8lG
z+r0Mj)7-p$lN(pBGaE(ZS<G8+{*u|*8d6GH%?>BWZL%bz)oc=mK5eIp`!D{7|B-Ke
z^V{T^VR3PVXP<qZGiNvPd|Pm~59rjX8NoL0=<Ux>bzv9hXjEaCh+>Rm%6y*u0|rV_
z#9%OBb7P$VszFu0O1V68PTg*otJgju&n;j5m%scKopujn#CRO-L`GD=Ed^{<9EXrL
zn<}Qo*$m6oT@)B4h{6h<7uoDlSpzT}^x50p<6!@QUb}-ybF|9IvXoq94qnO}!z~+i
zdi%cZx3Q3w#qRg^_qcug7I*I4VS8tXy}dm;Zt`BSzm3KW8*67+Tic{uu1-NlqXxr?
zugPw7n1<)3227cW`}v7hK`0pqQTFo;S*}RqoFvT%{D`H+HO_3D<<iBcx%||#oY~l<
zUN5^byHB>_xX++};tJN<LakWxK|w|sXRXDtXm=ls@4qKqg)7y@kZMgY&*>-nky0-L
zTN4pRlBYJfkOW9VB*6rHH96Jew*z7s;JvFK@$=W;Anq$lrCIXK@a~&$aPQ`w$G$fz
z+gU>oBf0$7W!IxwSVlvygG^G3VhYRO7YKY;P&P)7=y}HAMs`e7H;pl*o}jk5&gzvm
z_uC(#)09F{7BL|u4){WoW9gT^0WTSHvU``TyT{ycK(g~8#K#0#&f3fj%QIz$+gCZf
zc7vlKoH=)aOP4Ma$3yPkyTSGApKxaLEQbd}I>$X`=N1@tJM_yHw3h7dAMnhTXZWB0
zFaL@($+>&)7PVR#grL)G(eDo*s1?}V76tvr9=CfK<ijG{2jG&))BK2^8-w)hXXHw;
zys}7|#$0~-5;v~jU^tA)^OU2b1Mc3v#WPo)WpD404?g&qZ-4t$7m>Sjt10vk?S5!5
zUQsqLm@}^K0G7~T6GVGV(kBQ)R~NF%#~4^$U4{4pQX~YCZKh0<nB&$7Q5aGwS21QR
zJ0e|2yd5fKJHw^R&IbJf2M0%lVS{+6$W_kl+zge9RW2>23fAj$RI4+nv^!PXC<qFt
zSyq$Ie3Fu0IN|AH{})?DR+nqr7bLLK<ymU`V&>-;DVOS8x^#srPd~$%Gv{c`lpb1>
zFs6{Cv7Pc+Ax*C{cK5c}+TLO?>{6}PkYzg?Jxx;5JhOF2Kd`JoA<#<O!n)4g`4C?y
z;LnVqmB;jwKAm=(_Hm06b)G)gV{Yc*?->be<~h02g#O=FseF2Uwtbtw_^bcG&dxsd
z`YiQoiMN0Cb3S<gebP96=zD|^RO=Ol58adgZ$5?lyRrlXzDF4ONNMQxauAPqo{w}O
zS}A&WZo&E{LAi$K35*ywVogSE##mhTfrk`<rwq9=lxFIzJ%5Gn?jij<yIA2;If7iY
zHh2O8KnZKiJY$GdMsMd9!@KX2r#&jPF%?U5nyc41x4B3^i8-rs_PdJT`Sy1xRm;Tb
zkdHsU$`AkQFZts?`cr0S7ihMQINCjAWqpNiyM-13N5^fB4^Nnzo8!O#@BT;LeeW&)
z!+-p9dfg6LZkaxLu2828y*3&te2VyP@=+O6Z^g)1jK{j3x?+!820$|$4w#vlVQYJv
zh0+4&&Y$7_{T;F_B~JR>xOJ7wPhBPoOT7R7M?CY)C1z%-E@Bie(R2H0iEK7Au~-KM
zZFdYi+2~bBfhMEVX_2N0GqVi>KR}2wrD{r8Dlxmfh$mCRD8vr}jt};D|NVDZm|x(<
z7hj}YDG?_Fq9~$Tt<mjvNs^(Zo~EuTE(oYM>dZ7|SYBSn^Gax=ohra|)>bans8wfZ
z4%?(@PAO1RS=Q-+QpLG10)Wv}#)+bOK@#~Qo#^(yxA(0O27~Qi5>7-h3L|D`=c(7`
zSXo(PVPT0%C48jw81!>`y&+0Nt_{OsAC(Qs)sU^NI~*MB;QNA7sXRK4Tw4ix-wy~Q
zMV4haJy49+WO-(xK&A1Yq}bwCqI24Dmu|01zuzOtQYuvs89X}Pv<O7t7-`~xXX3ZF
z#-H3x2mIoVU$C{cjqit4D`k$45BTuI_lf(DmaVNds#NP04iAnf8X^BnfUrs>pViex
z>h&6fexD>cMtTo-vc-Dj9ij4!n{WJ*jm@)^E?&k&C5*|(P}M`;Gx=ajDQ#rr`VMm6
zV|I0kb1z)sMze)#5Anprd~&0$t`Ls6<btM>fJ=>tGx;IDuMpZ$N_=t*Nm;V8x`20b
z!XI8*r>onfgP3zqpQqFBaCmsY@!@rT`qQ8C5B~U{P>HJSZSNvI!OTKVr9Q{)+jn{E
z&3AbIxfl53Kl~HUY+j%gRk+(aaZ?M`*g>8;R)LMCtz^SP0u|%By3KfM+qmiQ^Dd`@
zmxM$q&C$^@>+7rR>}+%8$`!IK<LKxJt#g{qW45=q*j&Fr^XP!ju3hKrUw_H5po$^G
zmP}c&-0b<4BZy77{sp_nH3FkG_qXqG`}S>oub4ky;h+7p9h%1#-hcnE*xX#?D_?n$
zQn|)a^N8)OZL&1u(xpq(>ve)4!1q0(C>r%>D5dH5`}p1og&tKg<*!sKS>tRp!{KlQ
z(`XHOrpYr!zds~RQgk_T4rJ^oP75ne5*tPVffJ9+UAH(q!8qGf*GB82anvx9pK-Ma
z-w!C2%hc;L%*`!QuQ%}ghd0it9GcB8hlhtK16~l2=P5=f^g2!M-Mh<R*kyLMG0N<5
zj7Ftwze^AVC~c|d6YObJW?@AWgbzHk+7zt7>0H4Vl15miR1aCG&f~2>y;P?bJ-%)u
zJ#Tcs{cYzSnaX+Ny|;Put#?6s)M|Bn4?cMRZMN>*dF1;l)e<XfOC0PT5&9uv!=u|C
zVzR&A4n#^=UY_Ud*)w=vNH&wAO`UpuA9!?~=+R<|+HmX5*I7S%hAZ<m{3yhP*02P=
zZ!ynNM>PnFht%T3HcJuUxq@=8k<|)o&pgA)ev|!oK0;=ebrlDKKnki_po}2MHMz)8
zLCWRlHu>G}eTBN0Ln$C2ME4VtIK~7X)l!Ya{w|9Z!%J%cTem)Cabtyr`5Nahtg(0h
zklxXNpZ)BIeDiy+A+=!d?tRk1fR%GT%ggf^k#W@A<Hqe9l*=V%XXdzc;R2n-9&f$%
zfz`koof+4!A^@SSLG=PK2G7+>Y~hztUx+cr&7g2KpMo}Iy~`t0WTUOrz3~iNJKJn*
zo<XZ!hW!Dn%PaW8<M7~+OeNggxy`xrmoOsW>Swok?!{+`qR7o?P_|b^VjIvDmSQXl
zXwA;8F8Vlx80#lV&tr4r9P@KaD3z1M3H@G17)Eq^eeT}d!S~_mr*pFOh}*YrP_0Dx
zQqu4BVaZNsW5NX|X@b%@{eFjc-+PneW|KI{@Vye7n-`dyTN<TmlNf*DL=Fyjsg*;{
zZLXqC9j&Gztg*vzWCut17>(*xkmvThQuc~UDKYx<&7O1rR)s(oEHg8@FVKcG&u!DC
z@TipQRI3e2rH5x9;M5odgkgm6JUrh=nGk70@^pyr`$S=Ar_%Ypdynn;$g<Q*nYe0s
z5cq^)NU2n!TqzR-ei1KBt-&zi;P4Qw4Qs2*luOa*cO~f&MpaBy5bZ}*zDJ%H1LS!&
zQXu}@UU&Al`S}~K({A^enVDr~rq2C4xB2+v_Z^k|!Bwt?oISrz^SDK))8+hAn`B8&
z`}j9B&hPKo@_m8l2lP4v=H?gq?ss41#TP#1&h4K(`aO^Og%pD1q{X#g{FIsXRo46(
zeqir(&-ZO4Swz)<Z5Xonh;86<nTO-5<Y|VkNy7PgmY;uy!Tv7E-2;3rL0Y9(jl_r?
ztyAU}YrOQWS9$iObCjx{9auDsk$!*(B%bn6X@>6w=%7TolCu&e>>Tg1_32H%@rU0g
z))}w={6i#~<CA@E-@n88^{42x4>;J~K?Ca-F7n;)yvE+{A;V#pmtXoSM@QRy_)!nx
z1$bdVnoXy>uAkdRzWzj?i0gZ>G8Q1Dk3^4VWK4HEPlSmNSq=!a(4KF_13R5Q7cX97
zcjrDA&!1;yaS0!v<IXXy&JlxQmmrMT+S=prpvl=YXV6Aj+Ae@{u#2(deYLPq6@Za?
z^vRRWfS%{Gyt2kj!)6j_gHa(1%l{hRzr^C)e@CeVrKrTso7ZSIkEm6vL{W&!a`IGx
zwoDx<tze;LUj_7fJ+9xnVf$UeGAi%U>kYVg;WBdz%LoKno>7XTQ6s1D1f|khjEda=
zura&qG&Oe=4GUq38g6Dro@-QY|6WL~8&!15f4=cE+K}h@s0eSfH8&f_O&QFx49^Rx
z)@G?xtrX9r97q_#!a{>aql)KSr3Rx6d6t67n4e!@F!ZTZZ9T|nO`2xJNkW#{&r-SV
z`3by`O4&A`R!SB8Fc?YA7fTz%&dv_+fAE3Dy<dBcGwbVPLJT}0GIAPz%^Y1>sRXUG
z>v!qldmiE13Q<(~TgXDJBTse4J0HKty?a}PVToF`Mihp;|Nc94Pg)Pm)DRwcKAbzh
zK^TS{AD%Ef*I;pJo?F-M{ufaL(Hi=_gzfwH3BnqG@{RBC>es)|t6%%K+`9P_9(58P
zoq+%tkNEx}w|@8wqF|o2x_!X3G!WR1@ey+FF!rK&E2TtfWwoTFBvTnmLK;Zwn=4%W
z##gvGcms3LrkopuuZcs2@iH!++u)nu`#NVXt&)oYnTnA@qKzO+QheVh3<COt7$F2R
z^Yf$~!TBYRdpBF`-M-85;mdsGmDl*-!_Rp0C-3mJm%oGZeD3Y;u(7sGuYJt!y)7iL
ze(oHlu!JuK-ENEhgFV1!cV~|@u^3>{32(uUiL>M&Amzf@QM09aP(nj%i;s^I{Ld-g
z8Whs=nV+9$X>kEh1gx)bu(N#!>Df;8#b=gz=lyrldCs8Mr4-E49mZ^LZFBa_S(LW)
zYJo;7M@N<pQX4Ud9=#71GBFH>1N_wX_}B@Jnkb5xpKlO_hW?<-*4ACpG=XXbFl1Ru
zk_=shJl>uilrzNhBu`zq!r8OuXw1x#B?>8hp1$%t=gwUQ1aUkhic0vtZ-oQ3B1>cL
z-T4fi4_yP;M6;wBLxnaAmJc)$P75nevap^Exi>fm;uYeef@1Jro@ES&G3`!|;n1=R
zvs}~eWJs*4t$jX(8*gJBw&x4{AaEI}fDpP_OqQk2`5Ce_rQh!{jE4;4A+1)+QkmWS
zk)rk%h5@B=gdg~LLgJt5avuzb+~2y-_SP0!<*Y0%u{b|Zxm+6EudPmx-Q9goPMQpd
zF@u5P(#5Mh|D2`8{>T6L7wqnSh&C}<o{_`}<x-i;PhVzkzCo?}<TT}9=Q3z+?ce0*
zum7AR%V|_+sa9*;y?ckxu3sZd(kWZ$2Lj>4*|TfR&(HGlhu4vw$Hhw<40|yr&BxPx
z|Gq3Cp+D$wbll<}{lkCCSHAKp^K&cse)QP)J=TF}4b@CAym614m7h{x-C*-&jV}z+
z7a**DlJ90r*gX*=x};JH;UuL^ZlRml<cL~{)fb;-*y^+Wv$yd2eGDnVY?W8P{vyvk
z|1@Dapcgk$WC#R-@d!L0Ed*JT5O_XO;FDw-mBtK%lVi@WEYlfy{N*Qix%J*h1a+V1
zUwn>UvqiO1=IYHmbozayXE?jD&hh>pySMLR46K|zhcS|O-+rC<-hG>s*0C+VdBQ%^
z#>tQCu}0Wr*Y~K4*^JNM6dULv9(-)DrhjJ}6b+sf{NsQ84~e3HgZ(Dy07|7YOUnyn
zX+pJJVts9sy}f;sq(c-+e9u-<cem~$^;ZzmcXIgF)fY1}ti!h~FM;jE784P5v7{7k
zXj6vDG7@7DWXvw+s4QXF>p(@YyK|rYoh_uW(-wmyU@+|2>`hG8#ypOXo1B~+aqiqX
zd>K%VDlE+{5)TtV5K2M06e5I28pot@OsqA%eveM4OY7taoefx8ur&d#%+&cUYH$V3
zcG^Lhsbd^vg<p_~Rb1c6>nMO+8R9tQ=-`NZclYV{GLDW~G-hTI7<|`|ThNGozmTAD
zUGsntR^7IFxX0bQH@JQ0I!dKPC67wAiXT`Sv~<+#++_pBvD$sI&6~nW`cJO@U`U+A
zG-etM27Pw-b~tLD<IKhe0BNeYcXyXJ-gtwp`}eHF&n2F`%k$6K0{!m(9*0L|%4JK7
zE|)5Hs9Lp#CqGA{_1C_}S$&)j`Si2v9PXbGg>?%H8pVwp*NFRlOgg?#f>2VP2?%_j
z_4ReOw|8l`daSN3QLa?Dee<)w3uxSLd`T%;SgcX0`5Ybe$d#toA7FO(SzA4R?4$8H
zSK1?x5-*Lp|G|4yFYfT%KhBA?p)4es$x&&J=LMFr<BH9r4s3*#AQ1wo4S8&#pfC!F
z7MEFn?IrrnCfyI-<JqUr@XGJK#KL?PAri8Dh!KKPSVk#h#r%?lzz^_*N0z1thuIs?
zqts}S4*PuN!YVr_Lw@|rAJgof@U`!Mm(Z8&Y~4VpjA}`;x3kUs%q-{5UE=<&o7}s3
zhuj#>oIOV=^!V1dzR4ht`QQH6|0lIdM5oiE(;XtTol-d2^HR`3rIZA|Z)a0Gdo<$Y
zKg4<1?){>xc$(!QjEzpe^PO+<+H0?J@BV#SCqweY&dIoN;XFHAcPR?DSDt=`>(@W9
zMpp^6hP@ltXudzo%H<cys|yUJhnALhI+a5?HrB(Ohy4K~GJp}5=xbGoZBt^B3>d}(
zZeG7euh*trD>EGS>9ji-BiL9!Ll8uxp;l5#(lq7h=$Pf@Wd^+=`@1_-D`m9Sgki{F
z*y8A5i(aqKyYIfk;n6Wl<@EY}I_(zyUWezNxx(_dzG=I$5pYVlujo0^ZjO4P6E)ov
zGlmq4X@^gWyJ)SZ)R?t{>5PVUyUX1>cX|DnZ}U#@HuLjKL{Wq>+ELx5V}04+!KxGo
zz*t)=Nd_Dr9nm~FrqgL5(D=R}2t&d!;NnwfdHU&RnVFdb1bLPtq(@a3Ovu7fPP4BJ
zHiJM4X675nz;JwW$hFU|GB-O<wOYpWB(?esYa8b&mFsAuC`C0cT{^JOZy51MfA|kj
z`emX(B7M&iVgOItx%a<a_^PPT7!Es4-h1~WvRG0MLn@_^PV0c(dpFTZHm0UWl7)?V
z2Ezd>>+2W;+xHHUzT~N=FL88q!b$V*&)}!Wt!k~rE3dwY?@R91_c?R+EE}6=XtnjI
z7wb{$u@1zOKDhv?bF$um-K$OVp@-^p@bWTB+95eU(iy?xfg0n<LfVdQp+OB3JMbq6
zQC^AC+8SrR_B@F{;2SSrqB<LrWdp)c+Ky>0@WVQR?~|o{jMfbLeJZ6gagyQ-kV2xg
zA*$6lI@lo;F|R$d&R%+tPp;qLv-xHE?H>2;-(qQLg@uJ>j3{yU-WGE+OFZ}dOMLw4
z``q~K8iPTf^|NPLTU}v!ZIdkS^V7fnA(e6&t@arVlF>-fapZW)7#m|pm%>y;SgfKe
zU5K$N(CMZ`jp6LZ1~0tu93Oo60mU#-fgy<#E?ivY=wKT^2<Z3vT)upn!^3-co*+wY
z#b0*X4B!0~heyZME`OEU#sxA@lIp_Xf->5TcJ-~%nV4KS?o@X&EN*^>(HgC7VOt6T
zI%9u#o1N|ZNCA0nXE`4oHc3;(l`GFMH@iq-9CO!HczoRC=FOY5k6YZhevNBauX6V6
z2H*PDH>p+vdi^7A-@U`fpS(q{H=t4}Bc(v;9zw=I21a8P2rVsNeWG?#6w$R76RfN0
zLi;2;tAJrYcL`fVP+HsZ1@1a~o@bvI*BL)MKhLXQ`x^E798pkmk}ZW|vHi>}%NWK(
z1cE4v5CYOPrPFJ3bhOLC;XY}aQi>u=5f%^+V@^&^D3yIyS2n0rYe*z<YJ;odxP8wj
z2m*vR?LcyEP)1QIM}R`967g`r_RbbNyL()`a2DV9ICplL)s+R;<g4+#5~b)PYZpAp
z`Lkz%jR#K8=Zp>ijV`4XAARyMw{PDi3?q;b1Ri&Hw`nzx$dmCsec{S^Dz!2@_jXuV
znB(rPEs{9p?8Qyy7M6Jb{g40l4&=AIN}*5a2UKbm78ez-yz(+Hzx*P1?>48%6%SjF
zbRb2eTrM=8gg|S&K|&S_+IMe~`Vsj|-HE!{3VspUdY)rpNs012d?E4Cr1{WEepJyI
zs)1qY>>P`8uTUqY*KH%C03kwxAfVUjW0GTLX6GUF(K<&Z76UJZA;aOoKA3q*AOy4X
zE9AqNK&1Tfi)TsSyUW3yt9<W|{)oT)^Z%X0{T;MY)G7<8a*bbp^akH}=2^b}^2@yQ
z@rT^`>?WN~hcg$>gHinc4}QS@(Geeh@GgtXa~$p+ktK>K^zZ}e<ScTGG(=H}T6Gqs
zG+CC>Y_`dioqugc5i%fb;~~NzwTF<>MY4|F<GN$AEa9DZ-$p4-;FnlmIZwBJLKs^3
zqS2^Pj>=?N#_HM{i%ZLBJmNG(V~7o8gP7pf#~dBq$3OcFbLXxQte!#X5<+L>z9RDk
zk<xeul$2H*%i-|aSmQkEcW_^px@7luFzj*X-c7oLHc>fb(CeYJ$Kuik!??p|H$Udg
z<^_(9jyXDMk|{-!=IrhrvAuhUR@!yr-{d=Af0^I;o!=n}0<<<np2z0eItuW7tCrmB
z^%xEZgkflPUS**qBS&#46m!yzQ_OYEl(>3`9y^dxgQ)L6VMC!~(kYE8UN3Ctf2K0x
zG^125^X&64u=G0{oZUE&?}s+aGmNHKs$7$#2}u$o@NC1RFBuL792{+P_3DS*zIBr%
zN$>-K9Z;xgcUo+3-$w{Z5ZbI|n8qY=idPvw&ILqv8X&cGpd8<`?oUeaeM!6BVrTn4
z%gam5Hfs2uz_0kHOvL2*d3eu^lwTMP{G4SbKJD=;T4}C$@BI%L#wkG%;d>H{=IC&b
zI2|H<Fe%I~&+zQ?&vNI^ZI+hi2?LL#gCnHxaplU>7;WhHh6I77w&(fZ?@sjLFy-jz
zgmd)<p&t<s2b4m|%uF5MD?b*-du#?`wAHREzyV)K@+9W|haVxAXQ@>#kx4;@wu7J;
zQzAz<uW;H&LfC8ugHVc6-lLkfnIoYbSo!HntwFAHk~pJWtx&7f=o}u?1ES?Qq9}x1
zk;W;tQkgVK5E3eWNH&aV%+Ax%C-hn^=0d}tzIdMh`<tI~a(|Oozxpcw?$7_6y~8Fl
zK4j_q8G^YPe)@}_(Y*W|ue|memCvs7;U^z)uy@FXix&uk5`XkZf5Pte76<#=oISS*
zT5@!Fh&DM!8$3UtG1FjaWtm1}7Rv_G+_-Uzt^2#=`6SgD$Hpllp<?(G2x*O-bdb@(
z!2wETtgdeG+)FQ@b3v`1&}h_2)0k?lhRJQ#SF1Oeot;N(L6TaS=16DUQj+CR5GIFo
zKJAm-yH9lS3iHpuN?a{tL_j686AHCI(fC-@d|XchHu<oIQ!6J3YkzTke9Yd#9!Nu{
z)8gLtT~=2vvAVj*r>}p+&wlaO)N6AjX~y%<J;#L$7b%r1ARx;WljdL~j;H}6iGq-7
zwMOb{X+n5Jk)`qq>5qiMm6@dN8YgSwXuJ9NcVm{-xON2-B#<X;BsADYQb7UxS?8~f
z!3Zc<>y*nLE6Z!l%`Fi|CCcRpDFs0&34B|JN^(P%+CGe0*=nW5sbYI?7U63eGYxC(
zeUBiJ2q`$&+aoBI>2*8!K|rV7VSoR?(s4I7DD=LpGH>=kR+Z%`X_}BGDM=EOBneUo
zvNYk&{oBkhEz)@AnW@Zr^8Wq2pZ$8B#}`|>Cp&EI?;w0TWz!Elga_SW4?GCU0?Hw;
zeEn7C=4ZKnBV}WAnd9aOgF(vWr_Qr^W|O;j_J}GqHrE!x<QyCvA%xFxFeFJH-GKO?
z>*^0;Zr`~_xl*HAo8iYl`7vR+!n4o6#+mi(%hf=N2g0_t7W7<+(1zBX9pas<T$r6h
zE-s*?LgBdquQtjljFEJyzU_$+1i2u{GG=>+l+zAnH3Xew0!#BODW5D&Nry3k^oaa`
zVYiPgMO3O4L{uWqHEEVnE|uw=G^x*(F`kDunwh0}?%cV<VkKu~sm`A~f0j4yeag!7
zuc4^W>2#Sz(Y$k;x%@1p#sYVC_vruPXT130SE$t+yz#~xy!)$nX&v`?@#U}Z2jBZ$
ze*DuPfhSp5T%}yAadLD-R4FqvJ4dxzLt)5Lg;E)D(g)ONNEQN0QAo8~wvn^8lH183
zb;Wr}lBA=qW{sd;Yw+~t7kJ_MmzkYeW_M?gN+qIJuOLKB;FmZ&ZZkJirCciE3!gO3
zQCW^Kih)t=B?;{yAoMgdDkkdeaq{6IN4i7fYu`cmi})ralWww{D*@QXIs>NYN*~V>
zFP6YgO5EPQPaF?G8ukzN*xue@Y4JSGqb4`5-{8ZKJ|N4$lY+U~SuS6`Oc;g_7$|`0
zhQC75SC5@p;mv?tk9DA63|mb?RYevuxgIc#MPH6nv3-(N1WFhiU`b;I9|eXm43NIy
z_~?Ya-6o~zIy3b-W*f7Vt7U5SDz#b_-w#G|_SH%mT`nPjOe^AKK$6CGNRyWMz9fun
zhLozDVUm(-L!K$x?G{(BUc>WzW@hI`T3-dsCW;-#`uun}bWE;3!+1y>#~7XC2Oh_*
zBW~Qe$?Dn)OLOx@XA8<{;n+z(?pmEXAHTU}xBjp`dRtpmF2B#XvGob<Zi`$AJl{|%
zhlG(t`W7P0Du#u{4W4=K1@;fN@I8-OrNo`vEut!1dg?5`=QB)Fg0PC`l~`J=v$46s
zOE105d+&b0U;pH1fA6%hjDdsw6MDV3n46!cQfV-^Fh{92%kt9jvG0A178aP%6ggY)
z)5gLpQDV4%{RZWQIX1ukRb)8;<0FKXdAF0(GK8^wa$B5~NMnd}MlJ0#ONW}siF`w<
zB}tx<sf4nSc)mvvhNQ#TD&ppvwAmq&9$w(%2R_*#Ch`M}*2MA9(R{4|IdkzW{Z^M7
zckfZFMttkyJX?96jiqI(moC%Zx=q?Xp}V!iQMp4{uhVMobNkjUUU=>WzW43#@!|U)
z@$ow!a{u-no`30CzW=+wkM=y;y*}|E=G@uy{NC^X0UMiV*gH7lm%n<Q>sLRa-|upK
za!9*<LYmrs5GkQtF0-_>%<OCfFl1?F3!lQHS}k*Qbj+PQ_ehc)DdF<vr}_Tx{}I*7
zEPH!L2q9Qno+k(bs+BS(H(bAZkIl79<Y|WQM|68Jd76_XJxWqAE6ezKh8Jq2NKx5<
zs+9P5J|U11!LzT?TUbUJkHC1grU90jWk9(ALOA^sJKsd(#FOcEIvgDA0oo2uY93Rq
z)>vFz;$ZKXcB@MPY-(-z=%bH_qKMnKZ<9S_c#l<=%|;GOdbUSHDK(->7hUG#SsZqH
z<fK%|$f>vv^#}Vy#yV}qdx{6?Ntew!tP?3FK5Ajd0BDo5vvZ$!-+G_+NtZ@_hS|nE
zwR(-&`C01q22mIigaM_f#M!fFxct;pRH`+4{Wk5+A!$0mkfBtDHUdOQmS*f9?sL*Q
z!IP3CNocj(?Cl>=DOaqvm^bca8L<T2KIdA|YPZ<g-lBPYh{`jRwzO8yhhBHU{d>2#
zwZ2Lzil~*VC}B`2mu=~hA^&wdjVB{ePrpN;$Qdx8#ddawU)}s`j@xZQX?xa6m5^#B
zA_yc=CBl!YJon-YY;11w-iL2nCYuNGu*=%o0_QKBr9VjN4^n0t3k>=LYUkIeH)@>O
z++clejaKV~_ul>R*L?JUdrMf94=H_~d-g@nUpNO+vc9oOn)Z1VR9URYirYv86cYno
z7Gi)vdied6y$`NZo1bMV3NW6J@$Jo<=ZYjrTyvd`uzgRW)quJd%%($1Vo0gvArXXu
zj}Sf~22VrcCewLgfJ&{Q(fydd*QT;CiywrTAV=kja=ps_{vl^K&yXi6{a&A$dV?ej
zc<<+LGB+FY@^XXb;cYfNMJy$kzWF_NKfA%<_7-{4qnC9UNS}IbmiOO#n}hv5u3UMJ
zdSjN4KmLfH{O~7ioZsa6uYQFWo_UV1KK~Uq&z$3{FTH}&f^L7nQ<q=llMmnLCqMon
z&9{ER`E!@4R%`6;?33n-a=C=>MQCk@D+QrP7}_40D2iBLE7R@v+1c4gYeT2q<?h}4
zH0nKecXyeYtua4eN2>&*ebOYS**u|EtFp7bOFT#z_G470>37@I!XD*3An*lJ2IK}1
z3h6;ZXP7%z>6sMog_kKc)-k4RXSBeG878H)Ad#^JcB9Pb=Tb_Jj*sXMdZc;G@zDWb
z*jQg@X>pnD`$q&pi04@$#zv#T&dv_o+dE`g#>|WrO5EAoN9Fb!Ppaoy$&q5Zp23qI
zN-B42TEW3Q&nMK9Cn>cqn{l`DX-848=B94<(H~!Y846;8g{HJoq*;&M-CJyL?GOeb
zrLaV)Ql(O^V2mb9bA%MsYBj$4)vwa~?swVP*ueKC#tf;}eBxBm>a^J1*|jq9X~st%
zf6Tr6_gI`;0<AbcI%3d|iGtGj900zPub(`oVLV`K>mDC}^a1VG35`aTYOUfZ#1bJj
z&DIfDuYXFVR%LVJ4CQhetM#QZr#kR{6Hese+cgjTR*}%B&5pUl)%+dqwr=p@@dv~z
zB#;d|f&xug3W!PpwOWNnW0hB5{Te|SvA4I!(tLv?PLNXb%rlobbLK4XymOVK!xNTQ
zHqcq0T5X0}rNTS!yu;I%FY%B5(VsBr54myU)_>lah=5YX#|t2f;k)1d4*%+3{a=}x
znc>=vPswzO^bEOv{GhGRq4AFAd|Q6Hy&<J1sR)nG!3lTY|A@*$okq1v<|LZ~-%>R_
zPhymn3->%hsB)@VjGrZlFeFYD<QmVDD4pR)Au@^xL!Tr`NR^^oEurIt{J6ug+ee2U
z)kednq#`2_W#T+zI2=-oN+bsIenNAvO*#-PZ!A#n4S1;%p)RfQzr1w~5tP}y@B+k|
z_`q`P5dx!Py1g860QYvcSz9{8m1nN7ceum7t$XZj?{VkGJ-+|PKVV^Tk#@IBmRa8X
zOl^+ueEa*Xub$!Z(^rtvkfbS6_^hn1G8_)r-QDHz=zy*J_vp0SXsz88$06lPnJhCS
z$K7ss*xlX5^L)w`kJ;G>q(Uo0oFv@3b%!j=sn_c4?e5Y%Zi0#!4u`Z(nk<|x;Y9}R
z2?QQ0H~1>V7e28MK{-ZleM)kCLizb`A<teRmUehn0bxn(7+s@_0aResb#1iX?{&$t
zlxFLMZm&zd(ctN)uW<hSdCH|Z7MB`)di6t2PP(kGt#aembw2#yL$p?W^{ZcDX=#Z+
z|BL^FUeD4cb5#_!4gG$flU56@H9;6su9Pu@@obNNuWK`&a%DUy%H5+=D##P8r-xB^
zC&fSImjv*O9vf{y2!#*|DK$~qV{W!e6a+-25|v5?-w(*Mj96-haZ0bBaCEfA@$n9`
zGbL(`8l}i%cCJou(5BTs;wL}*DTjMaj5eG!TMYU`PO2UDXZGlI1|-9rq?}lBN~K5>
z`*)))v{o41POr=U{vLaKyA1k$l*&LrRI;+=Uf@BRvA4I&oqKmE*UGF#t4@e<Iz1cd
zsxi?R`x{)w4Hy8QoQz$w#f|I}Zswn|Ep9M~TJ+03GEujx7D|(6RtF=BLMoLi8yg#3
zy!aG5J3AyvOrySmQZc1+#M=5ALFl8Dvgbw{+Y5604rkA8((kr8X&!Ly{6(I7{tEjC
zhqPOdr?~!|T>?r~NvSMJQ^oTyUg3ZEm;Zw0<rNMN5AdYG2uZVj<PZk;riZLYv#^Xo
zp-D!uhrz=jjlkCuPk2b7IJkSC*-sZJEid5LX9<L*92%LDYmL;3G}P!YV|fr$?jNA)
zkms|g*pLl}NTopKRupb_2ES6F8kJBoN5=`qgEX|*e%|V!r4N-drWD|3f=X1Od(x)5
zSRx1`hJ!v!a{=Fcd4<T+L_t8@Y4hsL3WLi_{IBo)iu~e>)Ymszt@zy8`iwXmA}A5}
z`gA)z!cv89`-J=36=oKf**tfFPOHQ9YuDJib&WTF_9MRfjc>DY?gI0TMH=M>op{LZ
z&OI9S8JmS=ea>FE!1K>N$Lwr_EX(Ni`rN#EgZJKj7p+qenuCLVn$2UYb{l$hPfkdO
z9n$24Fj%1!`3R}eBA`_9+27sbgOA>4bK^XTR_q)ckZ3U3fP?+J2$gYpZjCixq0opp
z$0Qj9fz{VCc5Zc98gy%i?86@sq;0~dzRobHA<PgzACMEGh3_cURuV-hMV6#wX^Ik>
zcE82Z@jku`xOnz5FFyM+D~l`CstYVG&vEX;MXq1JX6H}mn!W8Uj8Q!I>}7l}B$yBw
z)XJcYN0w_2n}-}8ACTr5k!PDn<6-QY<}|%dhugRBG25_{BykQzJ43F8K^tYeVJFsP
zSNik~|Ky6RR&w33zD5wAd&HDL=LTscGqnbh3<-mPYPCwaQXvW>Jl{uaMVe>iO0l%O
zL{JJKln7rS@hE#0f-t1nZ1dr}S4q0lkzhON({7G6nuddzEK%sp5DyZL4~{??bZ*Gv
zA())cKKqpWcWzS(%goI#T1VXNk`DU>rGTJRA&`RJNt2sbKcQOksaHzW>NSLrQ}QNu
zq!vum62APhl5h@66k_Bq3MtT>0UcWGsa>|y+uY7RV#m8rH|&EHsBT7fk|A<~HaWsC
z(P{Tc<9X`kS>l<L%a@+w!nyO@xpN1VD`saJ4Eh;!b2H2|YGi3br5sVMl^FDTEHA7x
zFfnhx{WgIw$+MUnH*SHjb=u$N8a0}ZZqffgd;j%iS$5}lVxMh}y)JI}h_scJl~o0x
zfCA9n0NXc7aVT<WIGUjrHIyEtrN!a%`~~_S^sM(W;*b*8%ur&JY;Vxru!Sm=$x5H$
zBd$Mo8$Ik3ky$`hp$*B!CVQ_H85uX?Mx1-@IeYKl_xClv6VUz{hX`Z>tp%%j%h$j9
zRsNg*<{$Bux4*&%_uuE_^pwWd9G#u;-u*|ckDvTGg@obmz~aCWYKtqAB!Z2Jl9;04
z+4~=}_R4j(@1>9@7$uM%q;*IuNR%X#uwgCh*3y#-(HeyFxTdCFF2NYurX?vW^6`jt
z*e6P2Y8*ON^oD(wXH(j$r4Wi-X+#_ZOlzYN$4{QKwZ2K5rqre;>qmU$E3a~KcFuIM
zB#I-7)5rYnJNt~1yZrIlf~C#b+`mg<J@-HS0T=a@G>wU{m{mp6AJBNkd+)!C6q;+>
zyX<YO(*i&I!TWsly${&h+v4V1_qg%a8}vx1mQ$pIwkla4<m_(_$Wx6qE#4a1y2Lxn
z+i!n`8#ivVHX3!g9;aMP&-vHC{xA8#cmI&}^$~l!>r~|;cu-l*Fdj0WP56V~`h6DX
zE4II|&+~^T6f2AHirMLNCI|2H#k+fKs+!n@Y!zW7-da?a1d6qFNF(r4AY_ZG&RKl;
z9klMDUwNArFy3QnL)GCdxZqwZ@sveDRn<7>xVX5$8pEr%@9;Bk{|vi3yEyNNW5u99
zU^H4I%TlIiQ_?iy+WsC{maw(C$#PMkUV@UY9CwYSC`;<Pp*049q^a90XI}4vpVisv
zIkV}E{$PO6`bD!?2!v7{i_CqZ>Ht0gcU`#%h@Xrc{`Ap3*9rV3N7i~&M~1+8gcC@i
zNs}H*Bw+!`bCN72%W||-p{L_5UczW?Oq8aeLe#Uz<AlI@NmE-){nMcmfh1BfkqWTy
z$M+xb&JW%p%`(<E##mEv|Nf6SJDae#{|bA%w{X_+@%?u=nH`b$QbrqFByoyxhROLU
z&!0Wz+VyLsS&Y(wHr=6gm(F)^_4&_Nhg^bFA^AC+PzTr$8Vv{vmK?Mn@V<SY`|}?$
zYYwr42~AqjC<m(q^Tgqwdt?JX@p$VgmKAN|$@3AzUd`V29=%>4(;Acvz1BR-8LzFO
z1k_bYo~NvDjBzf(8qb|qUPnpAyYGIVI1-$mO(=`@C&(H9E=Lq8wzo(0`#EJ*fe%Kg
z^XZ)Bs=i`+e&!IN*CD%xq!OOGu<Y&Z^AG>Q|C!hBeu0DI1D+l|=3+MI>EUx8J$u60
zMM2ZFpBG#E`4Cpf>X3NZA-o+Ss};c{5eJcoSe{Nf{K1dd*xW$%bBvCNtU|U5BNUm4
zNUbGpOGa@7b%PTP0zxBPg}y*j)wHu2g=xSx$h8r18q;_Kl|qUzDHM|#0ztYqL_`sB
z6yxzMi-NqD)9dvq1gy%M&FvkY*H0;`8YMl&!9#xK_FeAm$Na1Jj`+^Ij~Q<6a_`kI
z@$~RHRZ}2D3*NG-RtOJORbpGqv(=JCma?_G%g%V6#x$Iq9di7yp7PG`{0@VyUAC{^
zU~@ZAH?^`HJ%7kzxgg1MymLHz{+#7v!M%I0)623zr)(;sC?<&#?%usio&-}y-6`9p
z3@YcNr%(9p-}zmplNE1&`J1GAL@}xG&M}=Fa`50?wo}ijoN@HtF}pW+*;yOoh3;-p
zhZZ8RhHQiIVOMF}7PorN!SDYIc6El_f0YW2^ahFPaQzpotkrT!(^S+|OSvrA+u7r1
z-u@aluI+`tAstdkCd(No$Ip27^btqT4~Qeh&wTxB+`D&=&CN~TfB&OTz(zr%VThuL
zjg4*6w8!!BG0Wuw(<L1?UE*CFM|kh3stRXq=PlKNDqa8MD)j5}dKTgnepv8ARQPGE
zu+O?HJpRh(?eadc)}a*e5_3#R1yQhe*E)KUgm@W{)!ooiz7V8d+N}fvKoIf$X||bh
zoY3$0>GgUTV|eGCcleWU{|U4CoGcAd49n$|G|Tz<pO@UY`5KYdtcrpUKl+F)i@0;w
zv$3^_QkuG{cz*DlX9v&d4+o4!eO^8g7iv91V_kjjB=h-aGyGNqS9ygJh-opDT=+TP
zTKz5`6yIZ(%qi0)RxXHCiIA4WE2@*0>cawi+F+X&5{Jdo7{g+@B1&S?G-qRTlPpcI
zV9T<q=;s+xkpy9~tjP16jg58c3ZghcsGPUn{w8Hr^X|Jp2w`FW9v2UQyce;y)?+jr
z(HaXv6KTz2*kiF=Fqtl?Doazh{5TUm%@TI@)^WlxpHHZ3L$!jv{Vo0v|M0)!-dk^Q
zJUQjVNAL5_`|tDQ>0^#gPw;pK!yPs@Hkdp+_{?MbDLs(j{UnfA25^^ENw8s%2<M=+
zoPG3!i`!4w+ZdywD2TQIt0k%R#7#l3r*KBlT7f7Vv<!I@N+;kfX%4n-sFxK;4N{Y>
zYov}bzNR-AVv34(K4*1tPNXF=@1d1q_xd&H8me&|k@p5vWkKsao4flwdGr`BQ@n%4
z;Q@E{_WAo?xyFWk$nQP6Pcq!&jk{mq$=M0VXD6)c8NwOb<rHH<dhpsKg(W|DPH%F7
z8sw;64-;jSRm1)FAM%5DKcc^}&UkB+8`p1gbN@P{Rf|+1$<Z|p(pjE7{FsZ$DRCSV
z$C@b8IPVybGs@*gpr~3$+pZ939zJ}+#}7V+tj8B_zri3K;2Vc)7M!0w<mB*xb~$H#
zxJ`4gqK+HPP8}8p)zMeIyE^B-Y4K8mb*xrPR+Do)hs=KT`|NCNP({5?;OM}K@IHHo
z@Q!M=plurJvZkN)`P!Gi#=X0D$)X7B492!Rdv?eV-+Pb8&mQsNkAA?z2ak|Ka_5!X
z+`D@ZB^CGOeU$p~3dEt6W@CMewY3e3qU7Y{gvsQL`Ep5JxAb~_Ze72@#(16acnpY6
zKVm*nU0g05;>X~rAFoFIWuNCyIlS{YYe8|fh=k+!I1`lituYvDgL|>}R}sVfnDX7Z
zE|%AI<N;nBw|8MD{ZoiA0BEg<lOVvY>ze1!pYyQmlYMd&y!-BZ+<Em2Y;JDi5lkji
zj*bhGyazZox3*DIGMilR_~FNl*VgFu)`_E+7UQdb|Kc5xpA-igJSz_U2mI0MTYRth
zE~{*c=qsGJxTZk57T1P;TyfMgf3!rF65F<b4U3X<oKL1G6(e<ojuUholjj+xwVa-u
z(Z97%6s06d4=FXhJmLEO4i6qYCyEl{WI%r~=5PJ&|2_Nrw|MgCK93(iWImt$g_`Js
z7YwMMwj8x>#rfons;Y_O3>{}wWrOz-*RH+7c&%VIpHVdxt7S>snl2?%aO?IpUb}OL
z$?SqB-m{V=+uI}lv%mYd_`+A;;@yWI@X^Bu{P6vEx&QG;v`xd>c*L8pzsXnM{%d^k
zi{IyWp1#9p9p&dQAR>sb1QID7LbX>fK75uqFEP$iT@*b3-Up1gwi(|`v9X{Q0*zx}
z8aA|KBhB$HLK;s(fS;^s33HnvX<J;~P?Z&}aTKc-o7D7rJ)$&cSxr$|1;3@mg4JX~
zUuw__8AWI%v2}|v77=Npv`1T4q-n<X-W~@B&xxXxqN&iQXKe58@n3%9E_ZLt_+P*K
zfcuZ0vA1)J&5bYd;Mo&S&JSp9+m+N%Sc|VpTqJ0PWV95F3WvyY+DNf67*ODu8_V(I
z=Nvpb;`{phj7B}i<2A<PF@wQ?e!m~+woOTr#-v$7nkHClSj=ZQE2&D$YPn*$Tu_uH
zrnPKu-6Br&pg*Vz+-%15<Oyf<L#DHe-O&&ntfm#)8yYPI)|gH)t?<_1Y=}U`daN@D
z=V+^v<#LG-p3@)vHfuNU(7W?B0&;S>S#aIe?gIVO7$T*4>y0n*)|+3ZpZ3AE2oWlj
z=g%H-ae54<A&n(jA}ET6)1wnE&L{MHIZacC-M0unYE9FySS*5(n}}E}=bWCNvs$e<
zJUryV!-p)EHRJI*ufO&hufF;!X_5xW$C?0#ywn)K#O7c2EQ6n<@m@Je|EXs}fp=jS
z?qx?gmZ~$|?3D2x-5BqQc;Urm?2ARt;t(oSPU@u!#Eb6dTiY?ze)^8LEGte=PI>(J
zG3V#!Jbn7?C;L3zxHzA3el}sezQK6B&Tu&5;Bdy|;(|1ZNwbVze}J==)8iA4kB`{B
z_6m9QNhOd(pf01S{{kaiond$^H3g?U=6mjs_*nb^JG!71bDWnUz)c0RfU%CoNsJaG
zqdtRm!D3lpOoh@t+O}oBSh6wB5L$OG4KNrEktoj3r`)}Bi}Bh9o10rSrNtV>`ud3D
z<8z$z#7Pg+IyN>o`P={7{~eQyQ~u~%zstA2^?Mv2AAjmw6DOKpzX#q^*L63?h4(lZ
z&C08av#M%}vi)56w60+`ZK&%7dFF{?OQaLDiWm+z_`(;yLL5i*2N}-~4*2l?2b`Uq
zVQfpRBI??(yS>ki>(}wpv2|^ioom;5_oEN_?hk&*$B!Oza&bZwN9^zI@yoyXOWeEn
z8ixlLy!ZZ(UV4{)a(wOr5@3&TfhZ-f;0EDeaQm(ESOiMM%$}X`<l8?a>-UJR?c$_j
zSXQjB3bvgiYYnZAP)R~ez~zP1T{w?M>4YeD=%fvhL=uOv1SC<KBBKbTL5Y~SiD`<8
zB3_cN4MF(Ow3mXSD#=ui(h6%eRa?^^4OuTHOpXr7vLQtaa$eHki22!DBkruf$!~pd
z!mq#oh^bAvKHg$`eT{?j6V9g->b9o#p4Pz1!fCx?s46y<VyB51Ye+<3du)tGn2j}r
z?ol_MX<cx1aKz)z`8P>ok~BrdF*=Td?Y8Rtzr4eHg|javKBW>8ElGTYsvNE=FtZcv
z#UagN%A~A8#^ia<#l;!Bx}`rzh_uAn7Ar;94+u(kDFoJ8oN;&&s;kC%(pb<pbBZVL
zv3dO-R^}ax$A@OLcQ|K>I@^!kogL)sZxAWXY&xN73aYkbu_~zR75!dH6nm6@mHoXf
zgx73tZgP5j%tUK0&d<7=DlBf(*_<D|^A31HufI-J1&fX>%Sh4;Yb|wcI6FJ#!Gi}}
z-@8Vy*9&&$F6{U(`x%!`2wHU2iT`BW?F)@a_#8z6F8pbe;brAutwnlyl`WyP57s@M
z$>Nnx#RhMzP_?+1`Yo4jXrSw#NFlmr{^bJHRWW~xv09b<)^GhbAAIm3i^YQH&ksHm
zYMC#V93CBW>-KGK+`Pq|J9k+YGpf4e?Cgv*O^IWTj5PE4lxI&Kv3LCyhG|0*U7iQY
zW%l5|6%LkjAS{JA;{)F3J-*K~`95_%2e-sma|{X*=OjX);|MP;-butr;5Qs5Em@Qo
z2$dqE0YWNPWkr^0#v7YVW^?ANC3&w$ltvsL9?{fa<l41827?g`yP|fEJWIHD?+$0D
zCDV&DiX<Y6B+d>oPOvuKK<nsJUxwC-{rzo*!vS^EcALK@_~&&l<xw0Xgbe+K#gej`
zeI|iJ2!S=8(pc)cBGSQOcyI3(Z-4n~{Mujt6@K<-ze<uM!TzaO@}2K}hkx=<{s*2s
zdBpzSHNN)tSLqLXtf~d8y5Q*Sn1A_i{xu&ze!|JwIo5jCHrM&tpZz*N|IKew6ifc+
z|Mj2qy+3(};Xl4$_j8>2KYsxUL{3uCk~+c-)3IFem+DZ9^#(68R0Buve@wPBVt<$u
z_XqTsN66Ddgm*Y$v0Bi?4bn@bb76<<0%q<-P~u`8N^9^f-gzn?dJ;*RGM`RS){^%I
z2)Ck|%}|j<q!F#R<atKhmbj`w=NVFi^^&G-8E@<`J)2NfhV{{D+zS8z@l;7fK~%LN
zwrZ%R3rYcb6!CYyc$@uUpMUzD_xa)R3^CZ}?%o}4Zx<Y%KI3FIp=iU7u=Ik$8O~b6
zqo!t{1>2D&Eo-950t3`i(%;@>dv~8K-y(?<9G+#<FkdXFS0!~-vGBnuS2>4PfYvCL
z5XUL8P$b5qT!m^*NXi1aTCrG8IathjR<;yM@%HswB$F9y1~wDuYlp((ZG(t4-oa`<
zBT|xl&<|D@O%rwr5{z%CTgx~eq9!N!^QY+TS1FMo0v7L`1swfBpDfQ94*NW-SA6{O
zIYnJin<nhCjbT-mlx;y>H$-v9MwWMpdI{&#Q>?L^ou3E5Mrb^SRbBJZgGUT|Blh+^
zNgC22voxXFsaY)x^3xNnwVa&?)%4cpCQ>S@x?#0kGM!GDPN#V9kxE?Ij0Bpld{LEL
zf-s#_LTk<QqXVS;^w`-?j!wP)lAEspgc4X|Y1*3Bv|aVnr8~;t=cTnqD;;_|opZBk
zFSE-H@k?;pmDc4mjv|VJp=Z;yIOjM&pKx(8x!TQM+Qh#MV=1lZ<vGqegivg4Zgca-
z4USHpP?QCA9g+&8ShKOcjmL9%c)*iKkJ#T>GFYQ~z5h1JyAC53EIHvl{~bQ$BaX!p
zjaVTF3$7Ls&MB}OtVGEm+%&>rVlYWV>6Vn$0%Oie``bZt<u%j!9Ic1kzyA?iJNxK3
zVr?|$@Z;wk9vrc=yG`8$MqF!JgjZ~=5AaU$;NwH8y2hA-y58dC<e2Bro>CUoPgEa5
zz}EH_>+4$tJ)9?tbF5AR{M&72n#Lo9!&-?_DN$5~jfD)Gh+ugWvAMZHlEfSw98wk)
z{oa7LzWg)%wZHa@{K~KVDzDzU!~XsTN%mqIAB{%rZtwEi-Mies{~<b#shX0f&mQsd
zlSh1T|3lvW;C)U`PH{>y+}PyKoqK%w%U@!BJm3f4{~q7|_8&2yPbr!jr9S^%{(K0l
z+p%itiIoccU=P;6#Qqj98h^@xS(H5f!Ta<!Hn?@?HtBSVtV)_lcAPwgY+4Yq^HuWQ
zPP^Nog@`%IvO+426&@c6R2~thIa!`l&Ssz_Q5@sFr(7+`k`c}rym!QrrkKug0^%%3
zs)(kn@m{jIbB(7D9w9{t-tpGrwZI8O;+A~jT91XhPVdnPzyI_JAI**#j@Q}VyUFg(
zH74h$98J$z)-}R`6N1uv%EmLPOEMsniX_(5IG#UWq7R;uMG4zm+g!hKi|f~~u^i^u
zRYjKfaVj`oy0)aPYn<0OV`<AJc5#L%XUKYmtt)0_#aY?%q-r=q;g#mC%>hH-KwS}g
z&{@Xz^=p`Bg-{N%kYpzjltoFLrXldl+Ro({A_2l!k|-s1p4I6I+18zxfLehR3g-mF
z!5Vqq<7u(t*|SGTS}fsQ2my6f^Yq{;2Zx8OiW2WVI@Uprzq!T6#uh>-nz}{$aD5pd
zi8Z5s4~e5(&B3>5ElJ~uy`62=Mt#bv##&1hMWk_pX$?{a|0{&#Vlw4B-~JA>#o|iI
z{^ACIv1lkE1NHV&+av_DRrQIU%;$|C@2@!Tai*mx10KCtEU?a!L?KO4DMcJ7<XMj>
z(jnHj3=4D9w#;U8di{QvJ0HZm))?Bh{ZyDFiXvWp^%eH^cKOkdKH&KHxT~nb_Xoos
z*RNkEiV~hZKcp^KjMqopymgzKH}CSs8*lT*8*j3;(PR1QT~1G*@%Z^;ilSsTpAjXR
zY>+b;3|XxfJbn6@H(#Bxv(`-wxL|Vmg0ArA9t&nX7Vq(q`7Y1=hoBo0Y8(wM9w}0k
z4&hwRNVL-kC5Z$`-=IZ+{e*9k(s6NdN`C+nLG8Y@&#)h0?W(pMAD@z@8Bd=)Vt0Fo
zy}do&=RNM<f5^4#`>YK{c;hi`+hqVb)&?o-YbmGaCDZ8zNgQ*2e#X-$&$?92AG;=m
zMC*vEHq;&MJVXy$sxsUMaU7#{u)H6QMo6i+b?a3&Ha8isZ*~iQjn<0w^)=EY;`zZb
zv)O{X_ul01{GGqY>#x7Tu-8XA0ajdHA3+BdHV*ghzRqwk<h$>Fk3ahEclh>q|Ck@V
z`+d&O&uNVz>-X5&-RBEm`ZA;S4W54dh~NC^J#?(`zC{6MUQ^92Y=8E9_xWaueDbEd
z3`TN<$-;GkHol{YDI^gwtCK0m-~9o1wIXX)#7f{IXoN?!7UczoU{9gCh&B-gRitnZ
z=Nmi*TQ?osBSPsc1lQy_#e7Cl74+g11lYF0H3o&p)D6Nq>arxsGTNp=sldgrRx8ph
zWwbV8I-7Cp_AR^+SmlDHgR{im^V(VueQS)Dj^BSgVRBY8nJgIg`fT?4tdGa6nud$%
zl=*x{-3A}4rG^5Cy+t<`q(Gw>2uGIph`gg%OnLCZoaaXexK&9aV<dupmN8lzpp>L(
zpsZ?Eiy5w&qisc9J5Jh|si|3DXaq#ku{Z3oAtcTDDM?+@A81~`yF(-mi{+B-jbNo7
zMG;yBfo5CP#Ce8wF2Fx^i)tK$UXOmSPpl)BizU8k5J`e}5*Zo{A@jQm3h?8jLq2%#
z9R`DlB<-Q%6sZ+eRWm)Ga(aBqqF7KA1;!YnDCP@qev!@fZQ>{*jv^xAh^40idP&UQ
z&K4Wv5kh%vTjFi#5u}kOPuBqbLU}F(uCXku74xcOGMV!H@Q7tuQneP^KfAa1pT?0_
zh=9dk-6tYgi5IJa<HJMlfA}#6&kktnhDaryTE45c@xhMAJ5;T@B*`VmhX+8=t&Bz^
zgj84;oM1nV?Vltu_wK#Mn_qZ?IMMv>Z_im3!M7+$W4`#MFLLkRn`jkr`}P@W67l92
z-r$v2?y|A9&(_v8_I7trk)hX@I9l32v^;+DlyWxX?Cgvv(cHYDky7*g@e@u@j=6U`
z{C$h<ys`Y1UZXoZpP?iV-+Or&Dvo)=dwidVe82$@h;+?RL_|0of$m%jhl^62HMDJj
zUX_S2c$Amq3D`|Xxu3Inyv@PkL!Li-h`Je}<p^&qt?+mQ2Tz}qNX@Ia@9@3xpK$o-
zoFD$+N30J=Y;EiiDGxr-v8D8EZ4c=UQ<kd&DO-#wdG`Dfruj*aDCZoD#RBI%Wm#ga
zBTZBCyhj`*2%*{A+vn!3TfBPbHD0^>8XFrUcJ}tz*w{wvP;GpC|06#9=tGKPNtO-y
zxo^CNN07uRQmSBX8;1>fBP?~%;+hVA6_&+n%DW%^kU#p)AM@+K@o#wVgAchlKgT=A
zU@+kNjZJp<_K4GzgJ)0p&L4e?lP3q@9KC*zrmkogHc;bIK6L|gIesb&>+<4#0nr6}
za!+V{T7Q{r7itpaBjVc79vqR(&gkbV2&aUlX&Y1n7=ch8q(H?YOlBR-prk-037C)_
z;k+kO3aK@%b$H=P^NiwRN~|r`hx9p9*2GytRTN}V97=2-f`SA#On@R7!)<JBa&mCU
z`Nai;wKbN@lGVJTE^DmyL|M#@B<FA5)btH}=h>7~yJWd2c(m|DGGc8w;@0ji`qoXB
z#gem&3FoT?MO9%<$fl6p1Yne7Hl5(whKt#RD9>@$(o0hY2x8|*ZHbsQG&s(xilf<*
zrGd&hS{ptp(PE_}(uy?%`%1Ct1odP|CR*<9Bz)y&pRH)h>SBqHGom=TVmGN+1typ=
zp>d6;s!Dugpte|TX>7|tSgO?wstF>F@LmM?$^*_Yn_f_rCC5jHeE&Na^m+;%^-xhx
zlIQ3sp{^Z+d>yUgkeOZ0shX1M#Rc=}lr+t$ih`yMq=zfVgO0x1nuf+$tZ_JN1FOkc
zjP*24OH<dBWyx~2WU*Q?U9OnV=YiJi{1rX)({1=ZYy5Opjqh#-;R4~*i%V^UqbMtm
zPEL99>^T?b7vyP%H5MZzqOp{+MlKf}9Oy7@g|QajHq0+9krWKGK3SR(B|QY{$90fb
zLrO_+n6t4p=9O1(^6rm*$m(&J)cV67o7=nO{UKUL+|GLJUEAl4H@?8m&OW`tki6F?
z>nB7yVSD#F_rCZx$CC+VV=2mlqFix)cE)HhU^G~xY!=KXFDtJd7K7+&5ZPf8p9BCs
zp+g!J<ewat=X}V!{2q__kU4XlS`v{Vy5%Eq0NquuC_>VhhQ?b;*|Fh_#YGl%QxV@u
zSwA1r`}l~-!E>@SA=9@JDnSZG-839MJtY<q!~THn{Vk3UPPzZyBQ`hI_{DE-kfs@?
z@kkV{YcQ_DQWM7x>*q*aA!s`|Hw+;JX_^v6nziAO&8;0aHrBaz?K<o08$?mWVo_qP
z;P&lTdHe0JaQE)3tdG~p@+fT1J($)}mxjgWiuq*4!SMmkX{_;}Jd@c44<7!ABp0OF
z9Z-_8Drw8AQ_7$6@af0A^Zxhw{qOz(A3uD+#pw*JAkPMDZ|yJ`48c2&9vw27Pna$)
z!a|{e7FI`{I5i@y_dn&9_$d$;zzgrA&@l`PMo@}M95OJ%g~JMua)FI<ZF`-~L4=At
zxE5EG__D@XMKI<!cmb|$gSbuWPS_?w!dVg#DM4u(e4>$Bb+Xsa6;=us%OzQ!ql5z6
zMeDW}uR<+vts%`aob#B*pruA8lGc=LZf&txEHKV;esadgA3bBWGN59zUc_)MX0$cr
z*IwP?<}l`4AD{EEY?ul~>sn@K$4n=ur2RgF@rYM$UE^MyP!uJT>6Fv6bBdw}8U+<Z
zOJ&{Qs)lm5p!Jrtb!-U1mXg7GS}Ru4b7(9lMML4BMd5@X(lC%MiIPNKFm$koVr(qy
zYl>HQ2i#nfBz{G`tngki91gq8?aL}ski|)$pn8u`in^{5O@Oj=n&BgXb~W|MW8~}>
zVvyl%+%<+B!b97H*X7!Vrfd<;VcdeoXmp&@>yOFvb+qmgYsiy=CaNill4nnzv0AK%
zqm=1v!P#U^WrA*FzAE_c_XFj(axIN%X=@W`iN?|z+f_{_?C7n*Soe7wq@R8V{CUSE
zY~Ve{JL<M&HeYabbi~EQ1x?+MB`MB3jCBap;wy&}fry}#qG?;|x&?ymd@^5AtP1L?
zMX3;$<pN&(6GI4rjx|Y^u)VXztylIqpG<JZF&+;{(ir1{JM(BbV7+5p4F*Gc{Q*(h
zP14@e>koMKwb!U!O`OIUTl4hsqfk*D9izM^O(UAB`ms|G#HQiKlvk?sPf7v}KVEq>
z)Ev48{BiS}948-9iv=}H6h4@*!qqd|ZqA8NAU%Z)H1RS-(%_ml^o;Ttx!Ewdsk!ym
z8s9sZaCq=BLRzkE-@&OqjZ@U~hT{G+hC|K9wd=?qen7FR`NKc>9y(3=`q#eB`o;!z
zQR7wy>sw3{Z10`3WSQoRZ@tFJ@d>A=6RdGu+q=%+{k#7;U--foSR0QSudTDSz02m-
zHd0FJvf||IjN_vtmdhp9nqZ~h)`-TVr9x>zoJclzMmV`c?PnBj!g4uhSsSumhKO5c
z%Of5<beQ=$i4|PT&w2OxN4)>!Bi?`VG0%<;I6FC|Dk@Tyu)VWG*6*Q&W_iA3c5uOR
zv7jxRz_J|%T_RmZ8>;(|-PZ-M_Sv_==R;Vc>#@p9SKiBu;qn6UFVQ5!g<}AnC~n-?
zCW#zI7+hPyszfyw=Pg!xgllnl+BUT6<2*-b6?WK>LdF`QK*fTnpCfh0`H+&PuDb=I
zo17h25P|ZcSt_N%v?la%r0Vh`9CcNpl|;o6{ow#4j7EK~?Qc@qh~fGM{ZWebD{5P@
zTF!ZG4e~oZHjhdkuLQ?6%w$k<*TsU_{DS9C6nSsJ=FT>|*LS)8>J3cY(v~&a8rrg^
zsahK6Y3B>1_r#GwIfGIG0(4SR8_%?|)IyP_5j{L94y|KWPUGtpn|OY1r_URkJ=S9w
z=b%J|GX<@M%ElylfTNUB<Y^Brb!aSiOdo44v5sil7FSky=h0DTvu<hzO3^M3@#Q9l
z5lZNeiRIZ?AJFS%+_-U*&CQ1M^XCXHL+(XBWMh4s@puoRBBt|Gq;@DJ5Eo}uWy4WN
ze{7AXwqX;3@U+J9@c9vvBmV5*(@*L>!h@f7VgEUP`H%h5z~6U46NRU&3Z~Nu7v~q$
zWrga*fv#hQIj4w;P^}C(9nR61K#}hE2lNMhdU=m1&QLl8McGR#@Tb0n0RQ#}eR{o=
zU;L$C;o9|EoE)D3lGapAXJ@#^qJ(9+n$xr;NK2Aydi@Nf$J!PtJ^eW4&a1bP(o)t7
zjBPnOJYc$*6D1j&>lxPDm;U-CoNbWk4wbmnrTeR&3p0_rdLtTY?wfb`qv>yQ6#WRD
zSS$?=OC;iM@o*Sya6%yDuIMg88}G2T#S2HI;_%bXa6XxFHkneFo_52dZi9K=@a*XW
zS_|7-uaN06(n49(Osf?neYWr1<oxk7idoHX{l*`#C<}h=Z~Q7r64MrSa6@+C97!qZ
z>y%gT-QwDf9g0;!)57ldEx!5lzs#4v{AJcg8-c}^q=d~ks7=XWZA`zP@!oqs<b#ji
z<>256ckbR{Z+DM8&%?Fr9Mi=)C({EiR?qQKfl6AeZ?VoI6javm@cE2)-uXjLj}CeC
z>=E~mo-iqwRK}u2Os+HbcXsI|JxX72dVWfET2alapYY)yj56w~rk-`*xGW(r-|C+<
z&HlvroIDT^STf~_bkc1LDj;BkO4^RbXOUiznHV7>tZW&N1)C#@EmwiWY%Oiwc63D#
zHqdSHLHuQ!8tFYMitura@je`xf?k@U6OEHyv&eSNti~d&r(V@WT8EpubwsF8Nb49|
zHJG+SrNL2DG#1&hXRNctX#`3!+*+r<zDCowwAK-)5-$f7l|pGnRg?@O!<%cGyCcnk
z$@uQUgoC=lN=08qM0m<#&cpi?#FLmb$;q=Gy(CA)nj|048)x)|W=qC_rYk*?nphdC
zy2Lad(Zb$xK~gMfY>TZLd}S$HD6HYtOz?N_4*ACZh@niWih{D7(7Lc2mr9exn(=Tz
zE)xWrBuTrfqf60~9$_7+j?j4K^@2z%A{=Squ@QLbaR*QFc@7(Qv2q(_K>3m^jfs<(
zwe=lZx#H>cfJix_s3JErRMscz4UkgN+8U`PNlz1(8YiHsZ79oSmlfk)ggN<_uH8TD
z_*5$7zh%*|=ziy=cEtHWg$}GHNt7gHS%#JxB{glgfGDN#n9$$zjwngcS~DCBxOwAd
zaB<L@e!oZF>xHJgGdTC@cd(F>I1V0LTU#4!?{0JN-s`;o{`>s!-FJEL;682Z@U3Nk
zZx5*;@2AwZp`~Fb+9B_y2w}0#;Dx6*>~a0}4Zc!*g}N@W-f?<-%&J<EmrGjn!jH@N
z281E_HM`F2J=71O9<hMs%KIOW6`lk4kl#Q3b>3V35n^XaVj>b92Zj;`ON$I0&<-N=
zC@y=Z4yg#qLqcGkp=l~y-Eww*&i%tF%TrLLWN=S2TN&IF!^yKp)OEx5?k)0kfR-_B
z16slDSKeUMU*qWEW0uo7fAU-3rK&34`pOsC8f}my2~FMNJY9CZA{7~FlF{!ENz)#?
zn>T3cg2#^@FkPF{@ApZP1g&GzEG3Q;R;v|7wV-Ybri%+^iwRD&OlRjLX+Q;xF%;F3
z+N`LG6|vHYC}Fu=ayDNwnN4}}{3*vLCmbvfSosxg+n_8cuNlWV*|3L-Vk%#9u{>jS
zQqW8qs6XXiaIiY8Kk0VRy=_-tN8(eS_+orc1>%XdC5t?|i#WBPxIth1U(hkLgun~S
z{>}!yMA0l)s5Ug)u%XAHRac}2>(Y?1Z7fc^pwhM=tivh?i3&}~%bON|Wq)3F4gI>R
zNJk?~C%DpD2eFY<H1&$4YJ#4GKm<sWglf48y%8A<VxlBN2!-+vQB~+D!r{phMH1!Y
zX-gdE<k^%+M|^Fx&KrAue(U2yesoqbt-*?z6ibT6)snWUxtJ9kUv`nShDZ^`39*jJ
zk`ygdq<1)1W4xubj?x<1)dE>oq&A4qYNcqM<E>G|fAO`ud~G8pT27fvj#w5I!dsM*
zL{UnXW#oCzsNWBUmr<uqw&AJ>I)*R-Mp_fcap+BDDaKo*Qenej8fMRr5WNBM=sHf@
zkYp=-SCPWu@d+P2nXxb)VH#B3aIq>loXkldJnSY-hf*Hro6zrQ34M)DT4MhUCWik(
zjE;3^Y^cV4_jMFSY^;yD_u6Z$4M&vAg7ec;jt`GnEarIcNK#Fl#0&-lhNCfSYwK*R
zr|j<TvVZ+X2hTd9L{pY!m#AnzwTXVIWLAl0dv7l+Mx!GRPmVY~IpK%jegEnfJvl$5
zGD{ZqlAR|HxOVdfU-;r1yz<H|*4Fx+uuGs6tZlCG=9j*R)QTug_`^T=0~WIx=Zgue
zD&XVZdur2?Xu@;3JVxIwB7%T_cyPhDNl43rbH2a)9lmq%n@scP&^E9XYz=o9=A%w$
zCa^eK3?>*RT?rNu9xp;q#QV_yXiYHoY+cK&@su@)xg|>#(YB&_X0aCy7sVr%s}r`i
zw-~QqBbDnotB6#@&c;65y$#OKk2yU$=7VqF=h?&OyngRB?!5A9aLO<q?<G;35k-ou
zdl%9qWqq^HaHNT%hNeE_ygH+84NdpawwAgnnXeX1XH%T(Bnjn=`XnGCm%(yP(@<9p
z(~BvGPoD=0^^+&e<`X8fDaE2|-j6}A#YAaF){9B92t&(-xS*5`J|Q%~D_}*aBtB&?
zV%r~scrP!tG(r#h^sY&~J^#$x;d3(4A=Vz<RT`EE>pz{L;sGN)R$B5zabtIrcC|uv
zP}Akpq)<SIrHeq7Q$7^G&ReVpZG%sf(Vj@99gWeUh!7o~uVWn?IIRmR_DeWR%Bvg*
z6=@6t+ZuFO-2x?`vx_ioi!p{?G{iLy5l2LEf|Y@8YTK5E2ImZMln`l+)`G=+N>Wex
z#aDN^vA@f2z4w^oQnIQQ&1AuBxnM>^;~g7bpdA!au`-4Vi*XIDa5PItZ8h3>G9=1+
zke*z@NWogaN2VmJV#RrFc<t7h|LTjoY)MV^Y{JRpghn=?B)uqNFc_e<W;7aw2`!Gv
z^8uALf$2tPwIK*0U0q)!juX6hNF_tVT!u-p7pGjz=9H5Q(x$+P8qlGy=D~Zx!>3O<
zTV2WFLgnDWH!Z8?lkTxsKAfMJ3&oE=@RwqE??UpUQkQC5(j;X(UMGnY#^W)^2ge*e
zKjgv3kD1Ow^ErwXNfxoSy}_>WNEwmm3oa%zM&nK5I1BO5jSWPBrlD?{&)5KV@xs<z
zX-cNc1&7B+oF1Ql;yG5d)V5$*Oc3YL_=+@Fq`4-^u95X(gj69AOGx^o5jSt&p>7Rl
z7w3HV{`<^kbLOjHyMMV0@m*EmugKuGTShKfT7fC#g{5W2<Mti?<mfjzX&zH23+%jw
zMv$dBX`T@iV=)~xX1Q|bMc}YBuEE#<={R=@$8uFcApiruhN^L>fkci3ZXQlXb5U^I
zJZ06a7_VPvl&&FsjIsi}U_9Jp&>OKT<}8W@4<8+}Dh%suW1=V`j$^!6^!hnb9FfEc
zc@~=IT6<LFkji0=!@8Qbt*GlJ2x5&5_tj_+`dZG>n3kfhST2_=77JF(6=!D?CKn6N
zj!&4Jp0J!vsp>KyfeFaXh-8!y^&&(h@i8=U%R<g+Z5^85N%tl!fF;oIr!MPd5;NEt
z6L0pYB8iXA5TEnh&#6EJHquxEJH%Mug-433T`b~-ErgU|L5OAWsxgTROdm;@ln4wc
zfx~)>@WC`K5)vVNXqbA7b`s$vN(FX;scU>2U=AHcpaeRKN%D+xv4S=nLy{y{+-&0=
zI*O>hqpfRV>rgVnh!<8CX`HZHt|;dV^1&K7Pun)(b`U}H5T$A0aoZN@EjG!B)+EJp
zg`2IowzI?E``I^m@BSmEs|I^}m*rxJov)D2fpyFlOU_mehm~ced;HAb_!U$Z^Lu~v
zhn$<5_0fn|$GhwZ&B^=k(lZT1nJ@>Qua=B8{GBhp#@BAF<0j9U9L-qFYO+Db`gn{=
zJi}g}EDa4?YaLo^`u!nQ)!;-h3e;NDwl!YLD?uGXkY*`DN=$1gmn-r(0}&|M!z5!e
zIip`pk;yJXYVd-lHdterFN4iXr`G%APWf`Lan)?*vVHc_<o1_u@qhqZ)uu)m8K}g<
zA+<tBn$c*86q56kvylICGU@VFpeZe@MMIXQn2oT3nM@`;dh!e%WekUFs7T^{Ls6|b
zI5^<+?DR8_VOcG>m`^!9Kch94qvK<yvl$|Cq@w`Q=t#4@yUzN?h<uO|B@$yRri%;C
zC&!F7QY<}m65~`bELA#YytT#aI;N^?bR_xh-}yrpix(^`A%kHd9i7I#kY)mxlRgOH
z5LnLqQ@%6(4Ssm>`_%nappQ4<I<wy5I>^d(>4%qj^AHvz;~SI+BpdI-;?=YvxXyY@
zl4;VOpu`9Jwq+3Ou?d_In!z^A4O4Z-(k-|cjMz+f=!+q-O3)%gCmH=TCtcgX*aq9w
z9G_0mN|7cB(`g{?q)AGeXJkpt`Pms~XJ_<!IZ2Y@eW)aw)=<|?$Tl*TqHU>7OR-up
zpUuOfQ4}nef$Cls1y$2xjRkK}QV|Um8^cYak;V-?WmI8=p_L6LG8kp>+JyyTPN+5x
zyVqQHulrANn+YY*YZ=4+ZIZQ^(#$BQD^{x|8=s@C{p<=vDb#RGV-2woS7qa+sy8ru
zUL2Y9AiW{g0%uF2IH#8mu+pL92;1$FlvLzNLTf{1;8r!JsZntf=(|D&mW>EIa^YPl
zv!p_F&3vU4d9TM@PpFq==P#pJEEZ@TUXs)*Oq8vmZ5m`2A$3rCS|Pz1#={Y_`HaF&
z8EkHVQq-nlIc-p_Lnw(8zWbXREhAJC(V$s170uye)<;9W^2#P3KYEH~h2HC9Z;r^)
zoW4w8Rq^D}Bi=i?K)&(?{@y?M2be5l>-se&i*wfY_BcH_;a6UJjR#`I{M{c=*CoDo
z+!&4c=2u^1R~e=s{fM$GKt*id++clkjBzDO3$`|Pn9ZhG>sT(9T)%N0ts`dhQ$`!>
zR}QcNyW;9Sbj}4c$0$ZRk8ccZRZ~<YLIzKyc6x?g&54H%M#Ko9saxvi()ly&-sq-{
z7YPwBU$sB>*pDqPf0>35g1T;a{_Htgb?iVvn1H><TgUmy86SP{A&(wByIMRhhl&Ng
zL7%~3NS5^wQn6gDI6gULwW?`sLsiZ=yEx^$fBYQ|pB#Paaae1aOeP#0JVS`UUz$#*
zjK^bs_8VViKA+%iO{5jQ{u;gEkp5su9H(fdv9{v;;spJ~ljj*pmZ8!Fts~O7hg1nt
zN8Gx7o2Cv)nQ8W%fQei(rA$ZPg;0G+#8uQ(SU?nxh6NwncQ|nOiMLvap{+FSvcbg!
zvG7ZEN1y{MWCw>yg3!}jOY2&U3r3^X7@W6A85YZ=2Vw+%0<6H!JW_+$6!=3QP_;Rb
zf)O>d=7h3cFo-r;i#JK-03{PN3THLOS(MI*MQ|8cEE}A!;e3I&fvT@G;6i|yK%i9A
z<;A!wCxN<cX|2UvG7H;KT{;&i7*Z;<QYa}IZH~w{a(cr&2pXlK_A3gpr0^?BTVXH(
zp3;N`;uM$y$KBU|+KNMhj3SctoNU~~Xis5gOrNc=S%ZFEv;J0sf3N<`U-6j>h}N1L
zUwWJ7{)e>166XW<7%D~oayKzFz@0}cM-=ODF=jnfFDJ^PkZ0j6Wl^9bjZPwzcgWgs
zKAn(9p^22IePotkoTF}PL=+$s?Ig|z+i@u(q!h$i&id9C=jOPxjqsSZp{gW<evT9p
zp=1zmb(RsAlV+M{EbBSSd&c8+CKnSfjt<DyhopJhIjDBYZm!GYZXKqvL~%@{JaMjY
zP0RWDDd}L1(fSx$HjIWj=c@v5YW(JiXuZe1kt8|aVzGUly;p9cHg;JN@mv4mpY!zL
zF(%fS^)cgDZ_r*mL3(&Y_qccMI_2b)=ZiDeH~Xw-JH$x@s>ReZ*4Fw|tD35=@Ca7R
zg0iaV^#@c{MbkFuNME_Y8q<cxr&h!=WXMN4Kz7Ztz?L<pZ>Wlrv>)EL1gPg1=&c&3
zGprM=7AvZ<L|yGn#E*UErN!Z;vg0q|g7M=+2uV?_xPSjXdDbII6Qa(~%sNX|lpH*J
z#z!B0$a1mx<a4E9H14srxyfj4gW+(5jzUsl+Ztx`1&h_3$>fB`j~?>=`yWv(t4}=+
zV=a@58Ak_4D4|h0qHSvWdBW`*H)xt|nx^bvUX4@=>4H#KC`a4WOeYhXrb25?6vxDI
zjMADo>ye}xv5pYlvA#a$%{N}>`u+pLOcmw;@52O0$HCG<zxW=CkfvTT_eVT2A5-Ne
z!V0Rg#8gl(YkK1$N_Gq@;XC25z@dpdXspJtuq)2X3u@B@h`O_fX@tQGgI5--J$?vm
z5hhU&ek#!08AO)$K^-R6v+!Itpjj`u$PZcRIeoQ8?sK9xLAeMeBdk?mg6;>e5jMcz
zZChhqaQ<z(7QbyQz6EbW{=5+uuN_|6pb#g}^^;VQ4tgY6Lf*?svy`>bkbbX6<tmnD
z!Rf^b7pG^z-B}uJY%t!2TS*cYjam0GC%mpdo$n?n(1Qp)>XDDeL`lqQc20RR!%73M
zczQn*@!EfOlbb)+a{j|7pZTjk8^ThWTfgwjETV|RKmJy*3Bh%+l>>YT?hzn@H7Q0G
z3^#Y!fAvkq!vRt3aIqkAuspfooo~Iz?R(eQyS+yuLfTr=@6k&$=F=J87~&|$BgqCC
zQt8m&GY-*Fazn#C!b?pu8ZxpCWw}C&oGgi`ssd|<5XD#}@D_X%7+X?Dc%e{<hPq`@
zRIEiQo7<b5o*XcJ{+O)aCmW7Y`H)!V2-618KeE)6)Kx*zu27MrpJX^R(r7X%nKTQg
zr^oDVZeyh+PXs56DKoE0_xGt|MYTL(ytT*9{w?zLHBKKtArgY=`32`iNwT@k-K`;+
zGqmSN^t5BNvx^a)(Re@{Yl_u^IMt+SpQ@U&Eavp`HL9~YNpFPC6V4CM1LP!R5Mp_Y
zb1lYs$b!5ym<Wcf1XENP(^9PpMx!<ISR=J$Gs*bqy?42}_a+R6U^VlLDOFkWN#4}u
zLUH*RFa7_|%d7eiZwLWx({g?`p&2xt$~ZzuPum15lgV^S5=UISy+gSQySFq?7_AN1
z*xX`gcaM$rbq2#Vl#UTX;e1O`l+5Q-CbJ3i<&3(lKa*hKOb8UZxHu=zQ?!!7Q>&?h
z$g-+vnu@Avg2|)}<cBB;?$6`V8vW6bY|smvI;lcrff7iqiQ>@kPm_d5N%r^mxOwv?
z@E$;bgg=Q#;t9(1%j-aN)yFgUF(>9ZE_2vY(wG+SgC<5Pfs(R=!#t+jq^3e4x`z7F
zuQ;!#l%@{VbmxfReM@U9s#QsvB<${GEKVw_#KZ8#F}b>7czevUn9v@YP>DPzKt0>w
zcMa9Xj4GcY1Y{~K>T%LT`<#drZ4zXwkgX)KF_BZ9`Kh6mZ4fRyg(f1_d9eDAJTe8D
zLZl_ROat6J$Plrl5iL#IvN)b|ba=?S?|hs2d=d0QD-SW@b>sy4ZAn-d4WYO3XSuB;
zM0rHgOVMLZlxGMRv79ZLJYL{e74j9xzbqJjJ?727evO-7yFo8{&S$sKe~J!d<CWLg
zSBm-bXG|VGK{#`z`vBh=#R&u+>mkegeB~=&Wq0!$u4!>~fft6RF`OJ6QWXWUiinkF
zGMi%s47bLZy2V9`)*EsGd6wfN9ilD0$4MK^)4H<J3XjmiIHflnQdm#ZwulbvH?1K~
zI=?4R$HfmEX;08^q<N05Tgs|L5BqHF?sI-}$ZR&Fs9VxOO)?mvk|Y%5kwz*BO0b%j
z6h%oQ!miD<ZI{p}sGEkyHb^w>a>06>5GzU1l+^Pn-~OlnJ^i~sBHq18FVzf(J)9Su
zJ$yucd`5eC!nB%VBgG&J!7)z4^}Rg?y_EB_Q<MmBUR^gNX-ZGWRE1(XY5CGu?ow7I
zi`9%Y9-@@Ga-;RmVN5s|ml+jWgjBXdOhb>vhlSQkiItG0G0nvVRk0u+Ih;4l78B~`
z)5_XEg>>~lf)OaRN$Asc!d5Q?Xi1!<Y;J9m_j2}bT<a{|1xXSUca=(<X2fZP6BcJH
zob?#%sGEkOUV?J;hdJY|A;q%5G(Sn7@pxKqG0tI3i*<scC^$bq<?Q5y^NR^p-BQ;<
zOz81QEl4s=ua~pGe~oK5uCtRw2ql9jo%1*g-qg@|#7cm7#7Z$5Y>+0wvn>R5DZJN#
zmDlwvLhT^9G<pY1h2f|@!Aeh}GkW=e$>NN<Y9SFg8GOVpNfv<)-FcPasi|3+vZLL{
zXsKxH2JdWearT1Mss$(58n03BM?5}WP&YnIv<cY8Fe}fAuE_xLT()u?a73s`#=tuG
z5fsCUBCZh81rZU6%}8`c67|USfML8wqyo!PD~VJFtt4@jk?IsBV_IiG1Y4$}USd`a
z5>GaWQK{r?e#Va;{D{MYBdP}_zVa{jU@y~XXN00Js4D;LReTKbT1q;|Nc%a$OU$xn
zc6vcGuc7r2fw~K$Z@~JSn!T?L*}pqPM=g$@v}}2Ke14?uV2!OWz0J)hkC~p$fK|uK
zzKpmHyqwDj%A2=e;nuaA*rLI=E{KD)Wig%b?D3P3!scOqF{QCB`!}!CnwC-&#Dg5I
z6m8ppx5zj~L=i#+M^WMkti7V3dI8Q0ybjIo+37in(r6WNnM5bL%m?c%&IT3)9gL+S
zO`N8*%Yw3L$@)Du_HQ6h6Q<L1%;XGX4cXcnNt_^cj8=+R2>Qa(o(DB_o(0Nl-Bef%
zRyt;@Ipfhf?Xo0_ExlMXl#UTC@4x?jzWe^WB)9I;q+_<WwgbHO^eN?&C)~~iTWQ4B
zc+4Qrao$sVOPWVCP0f5hBTW*t1TPhX{*d{+=Kc2`Ayv-WXoJb@Ax%><8f=9HxC=9K
z)&=uOFT%ot01;I3&O55Qre|&VS_GT(SS#{K(H1N428*O<R<x%1+^ppPScedTrfxYs
zK4opZMw{nEsV2%|q>|_)p&x0Ie*VHJ^fK>5Tr!IU&RNQO)iDU&l_6uCf=n`^c!T9^
z#>v6ir!F8Of+&fIlQ@t|2-o?^*%|jgddT_F68kczUjPZH6ngy`I*#a#`t;ZOI3Mx=
zz4KUM1B1xc7!$I2a4;Tj@y4wwbid9xK&bBP%euo865SOcP|mY-E2iy&I1jwiemW!@
z4Dd>_5*3XzSA78?1g$m1Iu05l90j2-gzzL$=-=SLxL`Y^b%3{L^A?vCZ0`@Tt>f`(
zi75$-gzktB&cTiDoGf2DnT7xfofCQsInWD>Nt3rUqGe&{(10eUKOB?w2EofM(kNvS
zp?sq$8tPd^vutRamZqsG$`!0auW0X;Z8ml{aoTgSIK!xh=$eAY)128)9#_+S{8@*B
zXc&`c8H^>aFOij?I4vk=1-7e-!yFHC57vGawtq%3zS(2*^$}_W4JGcUC<#Be0&zOV
zjRx$0{Tm!Teuk;f@YEqAU5FQ;g>%G-;I%u~khY?!TAcMnnWLG{I6r*O<m`m~{XJqG
zF+I5;$y3^*LMAbJFDLSXYPsTkazx!UWW7H5U>FqYaRir2Y8hs&kgFiEQsJehstnIh
z4!L>b8i@#`LM0_msmoSP7&7=5Nfc2M<LV~#P@0-FNf>YM62&pIi&Lt_9Oo=zFeJ`;
zh$u!S5xE#(s$k$~y9^I&Tbim2!AMe3RRzjRA_Q$yk>?4KbiBIWqn*uJT^z8C4EIk?
zST5)Eo<3)PHK)HeB8w#beohjlEQ%GaZK%p6&cNDe?FxTxn~KMepY!y|l(UmLU;nwU
z<6OmjJ_9sy6n6^OfZICnIt_<-0W@^<sHSNs%8I<Km`$hT&J$-T35rqL3v_=3UI=Py
zvCjWQeCB@&LrTePI^{S1^&>{ZA#39i`6#6~8j$8GS(=e1NwDLQL5LbfQ3$Xi>`FsO
z8we2^@WO|9+ek^0IKs<_Ce{oFeWF}ZE<WWzIxIX?KVef<lqJ*IjLAiT{jt`GP!MH;
zBo7S6#=`kz(lv@hC|pxFG`6B{3#w*CQx%j&gOrl>(H>v@LN}RskFhO6IgIz{ZX<EI
zoHz`EhMFZ8;*6zU;N_(+mLN$|jBrer1;#cZ;H}Z2`tXnsGX#PbL**NsgjglSD!$@_
zw{3%jp!$v?PNyxES<>s(3|@h~7|zB}zQ>E^@-&<?c7Y2*@8IS6(RDvp5ULVIfRSYP
zA;8FnX?n~oIz{#+YF#2zuqtc<Ok>0IJJSx4LDGwmLxHL#t8$4sZHe+2T!8S5armkK
z^DP!KxbKQs5Dg=;L5_+P7DrPxw5KKJiN#iJs9+$>`hCLUaSaA<D{lU3!rGfX`rAYN
zAZF%jj8W7-#VzskGts@H)(Pp&JKXz)U*g$E5Apto;2pyGAXGylFl=pPY>pI*^QR;-
z#wrb7Viz+m4iAvtGaU3$LQpMMh_dFwdD683QYqT`ifXo?X;#!liC<N?q9z@Ui2FTc
zoFXONu34Z$#x7QAyj1i@Lk^!Eu$V3Bjn_M9jjIXMx~_ktWFQ21f?zOCXzC?x9n#^n
zP8f{0(NZ(LIHFlDC_6XRsMkkEDM^&DHr}LOEHJj|`0y6rgyb<DYl^xclL;%c3>qF&
z;apACG{nV%W?@*rb&YC$NY+b8rWb6>AsgFk4A+JT9b=>>?T_fIpx}2dut8SKIg9z6
z#eB-?$%?ZR*xB3U#_a*i>VT>)=w$)Ex(X+9E(8p9*84)d)U)uuV~jZF(-~QI293w(
zIsJajAQ`Z-66qCqf$`!mqkQAv+6Zj770bnnlatdcuQQn_vR+Cr%SqCNI8FkoBo1Ou
zAqkkj)@Ut5Fqe{ODZ1Qrjf^BBRy4I?He1kEpAK}p)-j*WIXOQ=>OiDDJU!-OGQ-qA
zwz6x`L+GvL#7T}+nxbquIzH$4^b}(n%BrNQ8jN*V5u6&laX1swrRQ_YWEupy9#88U
zA{{W5&|~Xr5a9^5t--Kh=1*CQDGH0kVw#%OY{}AAG#<1FqQTZ0Ox+SI-7&C&`rKea
z2_hLGgbckCAFgGkgM1)OHA1!&CbYDZ2>RDRCvciU@h-ThgdReWGlfQbL|7<3tCwKA
z=e#^5umCR{cI>bTh#1JiA}t5t*95-t;kgyGD~p(bm<vQCDDt4~agE2vUB6@Y6My})
zhYZ10GLb~1i1GHAI7#rrQxpaB=>l`!VrmnP<8nlV3ab}hzq$px-;CIQYry!m3^&qD
zOvCc|oT_P9wuWc7+fFq3>Em;Sgu;8Q1T`G6{@Ojr-Ujtvm*QryLSdysT0t+iwB-e&
zFHxC5Xpd<tmKUd#iv`#Ax5=`Ex~WLCBGsB^zMySt>bk;|4YURs3GyfgYq6^pjqFg>
z%%gN*WeF)kb=*A!QYq3rVP|`jqoWz?`!%V^gH%^sMulFQ0S0=n5FlfTN@9eO2p`y5
z4h>S0kJiwMWOjZ=Q&bo$kY3>PKz)nj7&++CHkz_tp}a&PaW-gi<2WLg3N19YHq@=B
zF@~nDQEdyg!!9N`ZCPBL5qrbVwLQM@#k(kPDUHW_MM}U}jdVB@e6%KKr&MJ@u~<<r
z8^*(R-um)ucv&);odS|1&O7R^!`tAg^uk*?Y~%tA8GNzY){x~XaU3CSAT79}q!EU6
zFr>B)V?2!yrk3JTxb=zgpI%AQw&umoZVAitC0EJ~aoI1pR8PJ1xCEgEI+DbZB8p<P
zR%opt);JFr#~0L#Pv4f>+Hvsl3FjA6dRd0|mes1DnFS`$)#wChB34M9fy!}0Qq_jp
zYRRgYVODLJSWWlyIWJ}g57xqBRZ%uCW?E+}yl=x4??EAO{?ci}AgNgJoR3%E<Ip}}
zI~f2qiINPn0n(?zEVOK?O+!%@*s3KRtaZyrNSU$15zCZBgap8vy34!>o?-$R4l_2&
zhU1l`XalSt3y3%2`p-dg2Nus^_AWFh;eDbcOsIx1dH)p8rx!2$rT=$}j}LMfeAJyk
z+bte5@H6mJ7xWpIH{YBW7Gal8FKPF?Kke<-AtpENWu$8f*(jyo&#~UnmWJ7U&gsz!
zX5nz9!(0D?t#x@!3BeR`2>K4J|BPbyt%UKd3?Dm=>r>9|w@l}rY6-RmDmYVc^K+o;
z&-TLdV1+}<0MWSNi0qZS$SB4r0W>%jW4%T@&9E0S=%@6@DLw;lTKuwPd3H_~X*M?3
zQL)6>1{sUMQfM8%uAmNf95RwfsnH6&YcZyxSxKZyh*b2V>aZrfk4lmFlvY@bav&{F
z&n(wh1*5GsgwnVUr;R#Gb1lSGG;et0B4m>wOpEjmL4)&xMoL`2Pp%@8;|okt;>8jx
z9g+b=F+S2Ltw=l=*<QKcYA;Bo4*o#4#@hzdNZQI^%aTYb)^@iUZS+WlWHMVavyNv{
z@pS3f)q%Pwy+L@16;Q{LL@K;*sEavv)nZqceyn-*^({6xat<$!D68OIrsEi-=nN$-
z!g+)Y{U0X+rBsAQ62dx6(-NteUayCWHHM&IUzAItJVmY+SYxPEgpwhi^OME|{|T2Q
zpEWMUl|L@Bgs{79!IYMYmP>ub)%YaA>C=CG<Ef5oDn5g!FG2Pp?nfkPPTpUmH&{nU
zDQ#O))(eoH6?3dI__q6lmu@5uWFi=@N8Gw~gZ({#kHe2BfuqIQCMY$eyz(`x{fbnU
ztT^J_9757!MUAIM0}&dD!&*<>TB^E6w%{5=GLDEu5HMRDAkiJtacSNfM=`o5sLLWi
z*>TKhl(48vriI5ykP&zZaRSL6BpZ<4g5?Rc2hf~?U4lIX9stkavrgk@zX1fb_X-#h
z;I7LVRd*krpKN|#uEV339`DY%t_0FTq+`-JK_wB{Moe#*(;N0_TSGCgn4c|}TudlV
zYh3GJNU$U?U^mhIKLgf5+<^2BjK2ogUYDrcv$Rtp7tzZS9GXF@aNC;2(leXX#7WKP
zZ47@3!o^i?gu#*IIq9HJvcG}-;E24lv5r#5jeC0xwg%`(QY%Yaw#+6gR%Ok2bCW1d
z5lV)%Eft(fz4zS&Cx}$kDV>8gyK^$wz_%5qT!NBlkt3o&P4$T2ujPZ_&j`u1vOK$}
zm{o?GQheeyTxvjq=Zv^A!;?~?lN4m9Dt<{Xf|sOej@Z$xrWY*cGt|_g`x<(|HP}h;
zX@bsk)FPp+Ry1uJ3?92wIcq8!t0?Q5EX!Ei+2?5is+L^a-DU09?(yLKgo|pyuTKkJ
zmyXvm&AK!+(owe(>I9!Fq>S-i1dT^5*xTM?_u39+Q!-mFaNd(7DNz(*tqs}_XRqQ{
zgK=8N%JLqh=yEtb7DG`K2qB1~2xSz)N2o^8PS1#obK-oLL=L;-_-R+qUyVQa_=}!f
z0lE*v?ToxPAddPBM;i>*Hc8SP>spFpMkGDnHZ0pEz9LL+m%CsO{juWhw_oFze(7)V
z#V<YJasiPFT<4)BE`@{M;QWetdqkbqocL#me#1bg!4?RERT7m&q%OEYl=F(VE+~aX
zdxef;q|g`~9)~6vrP?kX1Uy;=k38X_s8*mA*|1N)XerB<wgFVfd^O?LRvDxN=-mR}
zNPH3FE5+)0!|c75)gJ;sAS?zyb{>D~`*G<#Z(e*Y2vZp;=uwRJF~W-wX{;48lK5V5
z*wv9nr-~#E#*)$rBCpV11ZsBGQcqf@2M1i7&#30jCz<n?i-qQsl}!(L1=s=BAi4$V
zK4jOy_gjoNB(bKKrSt}U_I46<lp<9|(OAw`CD%sB7(e;Um;Sjdte^;$U2v2mN)$TF
z*!$X7c=p2<b8<mq8#cGs*}XnO>aKHCx47jBvzQZQ3BB={7LS&awlzel#yTjg;Lxk1
zsLO8-8OE|ZF}7Y}o0`fyV(*D0#3Dy(jj`YxgX^r@y^LvzSy)YJI&-qimm)hDtE)mp
zmv0i-Tml^hmssm;mr1O!#(*)j%A;aMvM~m!C?``^lL@J_B>g@z)<_+pd%Zv@F2e#6
z$1z$)=qN%82a%#T8j)rxez3)}rw=jhiVg3OZil8R@kx)izs94*jO6^B;hIGDB+5mh
z$=EuijPPF3IKyZ(Was)eGLE@8Kc{MHk|ZTfg1=SMG~xWp(EOG%fCIi$qzfUg)agka
zV_QqPTH&oDQW`BJk&Mu;MU_XCC+`v8%1M$4t)iFiu)kVK{(~DD;uumbQC{GTp<2#E
zo=cM9tYNvDF*`k{JSlL6|5PEGT`2Nsm)p1AU}LkR2nle?(Mks{Lb)hZE3|kyXUS8Z
zamq*TLrOQt`%tYGz+5|usKvA<n3q~Y)`m2UhZ{su0!R!7heLNdB7aqZ1ln*c5?t&l
zn;NG<ClSM~8Y36X<_>EE^k(V+7j^~}=m_G})7uebsb=#P%j&LSdKZrWf6#o(<BwlD
z0iShgYlwOg=_n^2CTJw8@d)FIwIa(itd`hZqIQF9-1^YF&`J_%K~*fnzEwd{tXNG8
z7PbibmzBX*hPtY#iU$8O#3l$;MZyc%>$0j*Kn~&-$X5jUmj!Hkobp6TQ0phTB#jf&
zI3i1XBuPvZCuDh!)-f#>p;|<&SxXbnKL0ZNjM33$1wlX1I!hWUgw|}l@di6zZFusV
zzlUfWcJ{YP`UAX*Fm*#+)|B%JzM2!KIq{&6t=cg0J4=?Yk;ExF>s?VCFPpl;(FM;Z
zP)eHB98=e{Zb2i00#&3LzVmZ(!Xl#xB_(N=Bb27@D1ox8bd*$AhJu&p|5BhOguqFO
zj3S(s0YCH-;RCJFdq?9PS}TSd8$rV|xnMDy;Jv4}v5u1p9YNOblP>$jQB1El42uyU
zT#Atz8OOwF%mT@i!wdSOn5;Y_xiIXH);U}h)H-98^!f1lQ?zNhlWmcBPozCogaxo}
zERBKT#yZ)s$9%P5zFOkD%0a-Di>v7R9G9%LOV~+t=f#>9V@;?E9ki{Ztt#5OhJGLC
zYcz_)E_wLjci36mqxB`mH!l|Gm-xJYshoBHV;%<T8Jw3Ts0O@1c!`X3u=wc$xtgZI
zweB+~%Id=K?BheuPNy_48Dx4XFy04to&)JTD1)VDWoJCD9`mGmN)s7$oRE@|A}Lsf
zD%e_dw|S{6kO`z&N~%%>@+xQL61IX)6zGG!yA+Z{CB$(`Q!nvSko9vmBahTe7E4Fl
z?Pr~Zwhar4Y@jMVby<<;HN#vm+Kbr8Gd43g*jRD&znI`9^?AdBEiJVvX=g1ujffLT
zlqy6j5s3wNncnKKw!ylFm1(G}3fDBW%LdWGn@t;Nzn{>6z>EIKr9tB5BHx{NS_1TT
z$r#%O@d}K8R<ZlVAscsl6n@64Sb=k7dBk9lGZ^$q(j=r`MmiudGQ<%2*5X=Aq%_e_
z0jYj!pwy-ABJ><Wt(2^<v-h>HaQ^ri)dvp{X`gqWxAbwya>d?ujuR=RbM*6!G|QP5
zAuUZR9VYrXrr+ybUKzpDMgX><wUF2dohHOlLOH*{6f5clAQce;5vN!Y63Y_piIkw1
zC=%&$O?4IHdi5qCuAnQ|xdB6%{De|?7qTgY@SWXkSGhQks7%;#B^oszvffXapI=a}
zR_HiI_kyS>OM7grZw8x>E`uEF5KiEwB1$xok{mxhV0wCiinqw43|H4A-gD!|O;#%B
z{{8p(;o$-Eo@RY*gN>-h^*t(*6vdpPEJ^bLTig2>2h+uZvaX3T{o*y+kjXAOw_@S_
zRk3gdUv*WQ=+1+d3NIaP+tStzrfCsrfb|leu{<eBP8Xcc78G^WUFR>9$A2~c0~#J|
zduiAG=O;JV%JR|oKjyo?|80KZi&db&0zqf)kexHG#o|Rn$%=(t@O*j1fjMEcIU*6!
zl~1oiqP?KCAY7;`S_8dIk)|PfHq4D4h3>+peCMVXAgoKTEP+64MJ95jR;((A5}szz
zlB9;#coqvou>@m6Po!-_gScIJ>XoPH!=UFG_hIv^1J-T~S-Ugk$-i1reA~UyWn8-4
zO0Eb8uIZ|LnkyES=p2ISHf^5}OI<?8?sGO-mx+M>rH{+%A|cRkFN=Qz^gw3{*LR2;
zp7C2TH-0wf`q#(A>x!x=F^2{#AWwS?MnlG<HPS5WjxiuO(m^Z|GE#&VG`gl1ik$j$
z@rBQ)@m`(~-{G7hbb5qJ=<V#X{iQc3>YAO`-b7~uo_zFx`Qa(^a=><;;<7D9Yg_mv
zqi&Az%8+GQXvAOIpmdYFH3I7c?inn{H9ASqjL<?-AgSw$axuf<i3bAjBrPGCt(>pG
zmh>WM8iTjN+6@A96DHF@L-oFM^$Zq~pgMk%@F7UaTAUC_<HCn@IO}MwM@9)K#b9j>
zDV8C9E=|x;jE`cn-T>#ItQv;>2%$6vkBkCMbFrNB?BU03CXRuQDQkfimTJ{<>*rqO
z^>6+%Kl<*U@c%q{$`77B=K8Y}zP`VP>Isw)TpS(Kw4PV*-lLcHIGdlcTrQ9jBBg`<
zi`EFQ(c$t4{SPaUPGW-<8_ox?)C(WrKo<h1q>#8)P%KKytRzZuB9#-hmc9K~NjG2P
z;Jg2ZvM68d4E>K^H9x+E;-$wv>-&7p<Dd5ZzZ(C(MmeeZ{<nU>$=@Bqm&0Si`;IH+
zz#AL}FDq)lV&P_-7H9arAdWH;f^|-?cn5_>Mv%t|!mTiE$Wuw;v}=~TuK%m4J2G#G
zVhG|^Ap{y7^jspiOAn(F(hFwA87k8B`aNvxux3h@NxX;Y)F29ITCfc?1|~~5PaJ2%
z1>>}3Z6oF8e?DaGD>2XizcmN{VuhXhz)TwxCU6I$0)J_jUJ@z+@e*0W`M`$z@k>_m
z$pzrD`F)w+k;4^lwj)|xD(Ww_8w^4B6g`y@4J5YLU{u5In?t_%^ViwBGoa}$nKtK4
zF3xbyF&_6>8;wb_oIL3ftE7{(Dk2f0GE!oC<aiOHy)?2jwEXPxQx=eJ6?Gu2KzW4_
zphXZyjc#7!&9y#j*KQzY6YgE#U=h_6=O^^X*NCsLvwmxr+2oWciZRuK(P#rBgCI~4
z+)I6z5*UPnG2R=TX^}pdI;!CsLMOQC32nK=6!VaZ80UB$F>5N`{r-onjWtPRsoQ|(
zN@QoM7zOaN%Wv>VETO!#Sd^0}-%(z@#VZ9mIAJ0TPAhQ1V!<~CV+@H($a|8-YKij}
zAq`R}k~qibV451+gzW29I;1p6CwO*rj5AB_ZAnIVw>V03^n8IU47Q#TrJBF-_x=m+
zzV#*_{P7?1C;#NXBR)Fg%X%B%7R+WV?)<{fv+>I793Ma8c|C{NlSCO&l#nI~Ssa6q
zG|u2fc%>cQhjgH5iIt~S9-|yq8DwHm-qThwQfjR8IM*OziB7?}nq+-IAth%QC$vq2
z)`5+qwYZAp^juO`yYf}MxRx%rRF~DCccBOM0`?N0xcWZriu`v)6?QJ{QvYgv%JA^?
z<6{mFiWd(#P+dL`GEj|LN(wjUyg253bw+PzOsvvwjugS=);WrDg{d3Lx~A=V6=|A~
zL~-Zm8^#M;luNZq=YZQ$^Cc4Rpr{J^(ST7rqB2X$y6PghA&C|18|zqW5lrxFl4@Zn
z=032mYbfSWE*<kh!P%f-XOyt@=72Y@Z!>ti;n}xNnZH|KK}BMl;8B$j8aoSuNU?@E
ziP&7*gu*fT=#28jzF>DL7`~RWb90Y4NjN{AaPsgpK#1GGI$@DWdEtR|dCB`g2I$3#
z(fWXFBSz+-QmcH$_O%_}`r2*QcZXDB$=PznWU)jNG1|<D<AgX%NYezQK%$9-MxcmE
za99)?O$>M<YFaw~!_OH%m4yXCh=veXITB8S6M}p+WPfvm1c{Ii6?sZmvA;QFcK{AS
z@xF5a!Z8~3u;rX+IAB^Y$<nOz2kI=EyU9a#jc(^$$eZ>aXM@L<5Q4auG7xLbU5jaI
zSQI3=#K(fks$j7w+1c44Rw1FRX&S6?q1crgt#p@|7$!Ao6~YB^oEHu!Y~XTt$D;u6
zMCd!v;#Ej$_TJ#F!#IzU5mA)l+933kS_SPzoKQI%v=;)rkl-cr#e%c5Q_@V4#R+?_
z+`#3VtnNRewVK`SEpi>9Wz4PH_qe@(gT>=Vy!+4pN3<~{(VSoWf2`QL_lVX$W;LI&
z+m<9s5p^G0$k2rI7UN8q)Sbn<a1}}~LUTO5j~BsiqF)r)w&{8&p?VVn;`kh$EQw-G
zZY(?cV_f#HxbyZ2*RR8yZ^7UE8y>AYFw28=VKV0GeeISFe>o|@%iA&`Lnzo!j(>mE
zhLj<B`5*t|zgk2-YpiB-4h~*WlwQD~p;~aVr4=QGn{s&Zn1+g-bcjaN(hx|$Evsh5
zvJ8o!4Iv0BiXhJuGL<1wv>3wn^2O*tGGd{KM2za}Q?$?&Rl%7)=URWC^>h<s9cSmK
zR8>Qg1Q@(>CEMF;4ElX8&M!DST@WP^GVODIT61<%P!v!%uw2y~W;N^Mf@^<kkF9Uq
z=Hy#*9{uJC)gy~v5vqhO;5uP~UqI#*zx0)F@RhB%dHTJ_oL@VluXBVEeDv|hG?C{U
zfAbss>^Hx`+U7d%JpLj7?0^3M@X-$+!nN?bqJ6<=H(|UpAW5RYG;J-~3L>Ep)*zf`
zWlLsmLE4ji?F+ZLzPFEz;H)@gv05@;tWY*(ILzpedNig*NKGbs2qcM!x%3hVu4Rsr
z3fE=4RJNoNF31x;=eqb)Eg;^9-FNp7UJ9&GL}^YhO~{-@j>c3?OWYr^xizBE3Ly-O
z>4fF<0#Oz8vWzH+DVr8=464iVczMXKA7L%QWX*N930e?mDWmZQv&jX%HQ1u))Y%T}
z1b9WHbO@fwEIMlu)}XQ)l@V-BghYzEYi<jqSG3ZElh-M&F%Ip6W<)#D<t0h14T^4`
zg_EjuMACX%QzB)Jjx$`F;!J2z*QUia2CrhIhV%1tjIHVS6O@wV>uc;?zst8jddRi?
zYdnAYjEz@LS>L{a)e;?NeBqnF%y0d_{(Ihie8|lkf}O8M93A}){)d110KYN_?HO(i
z8LzFezSd`Lkg&GiqZdbHLZKvhTZ1hrmlK-hgmy8<yNbT>q;<o|!E?^eE^w`<ZZ&Of
z$x_MAR*#LnZIV3Yd^+XWYUb$%zy8O^{MY|Cxc?CTr+)&Oh9nB{o`dm-t?g|#H#gba
z-(!1cA0;Hq<$~2}NwF%JFM^W3C<-p-Q=T6jvRH%$ZO5kWuDd_Y)%h16<1zfV|1E56
z{QLWxf7yn&meuqn$#$m-_r@T-$H^8$O>Iji%M0SIm{bMx#fAn;i^F2P4Otn&;X3U@
zq#=nlT57_c`bBh|=Thk|y4|q8#6^&ZU~l3*^W~IrHm0uz46~L)d%<#Lh~k9Bs-S5a
zHaE8D_xcpYiX>m4BqZ9CU&|N`B}XSK7E4cE!K{SkykK(vgxl-uy!s2ftnCkY_$MbE
zy*~{zry(pJJwU;Y^&S2<-~1)ESNoh-CtQ2w%iMe84#yV<46gSWj(7N#U;Z^-fAdZ9
ze8Bkn8jqGA^H4q_x$TK>!S1~cZd~7Iq<h4*K$ebXS+ZI#C~Sf2DQeL`SusYlv$w|X
z^$|4-E+%Izmn*P}RK_S35vhbI%5YwwI(Kf1qx2P)7V9iZC>j}P3o(iq4Z?@$jK2_s
zCAx~h(@{X_;E+(1#!*V9W1_00Ki+14dynSEEn;C1q9KkYv!e&tdPOQcy2~b4S|i*G
zJwo_fT~nLj?Sc-`wwD)*YipWT5+xaVf5>WjfoV&e4E7^JO6s=4+m>F|XYFr4CVA^Q
zu96{$N(NZPLuek0PT<wilE%Q5DO~qY7lPXzm-xi&av92W$<#s!ROXoezrTq6_7+Mv
z2or2mtZQk@n%Vq<B#G$t`r&mYF*~<!Lw~@+d&=pArY^x5q>3;|ZoT;yySMMK`0$7L
zUP861IlBL4-u=C8&hDR4n})PMLdP*yHr#vj8o&J4zeZWF_~^r@OwZ<Qjkej|7%(0s
zY>X0yJs89bra$-=gJ;h;dS}k#r*rB;GG8gm%CNqsk*|-)-q@!d^f`Gh`RKglFzfL@
z{x9&&UxSZ7>hs-i-QgSGc#AK+{WHAw+H34w+h=`aonEg`lBOhC3LT-LZ5!IUqHRrR
z)He;2`Gg;R@IJru+rP_qzVjU(Jb1wQ<l^!QygJ+eF2{Hb|L`CFJN*;?ag3MN4{IF8
z8-#UuVX-unX34ZVXJO{7t?d(a-C&QWprG|3Iitl=)xm{WMUX{`EY8sAEB2K}W4e^W
zmKN{9v@NmLUC%O%NQP+H>0-*fo-)Y$4D^sB&Y8~^DB0kh3rhb;vs^7G>Lr_7Ia!=C
zxmaLI$5uaKy*Hw1plTge-O!en*@fW4<5LcMb8cSS=SzS8HtQc8@%+Pc%0)|@36eEv
zXP$f8w|H~&9_6tiStOjU=1dpwvQ$g<Zr<dZZ~X$l`jx-QS~^CcX-5UG-?_ybvvt<K
zm=Irw{q225?FNyJXhhAz&A2GfD5)6j59zJvOqWyi$`YgLr+sG2IkW15sxHWq6p0T=
zhKQ)Vr*)p%T2^&Q*;Gg&P)b7AN(ksr=y-&_NMOc>CBOaiH_=}nN=bxNfze|f3WaYS
zDgs{=oE3B2XuvSh__oElhNiLDx+G6yTv?FBS)kPlfyV^968kAeG@dRuUE;k7vD1yE
zY-?-^QJSIC7+bCIwn9Y_aT-xKC0ber{R}-`Qa>KB_>G%HSp=~P%4^fov<By5a6_sh
zW>qH4YR#l{EE*S9U*FP8Ax~h`gMM0)MJ17}!!jd5XdMh2s}@@~Z2n(AKxPfNt|uTx
zV3@gv$;Aa#RWKY5Na7R_Xce=4?K*egdW(Z6=iIopOOz<Asc}(^bAs{K4qtlvYkcd&
zcNmXyoL%vQ-}{$TXBpPEv`V3TjHz21Q?a?RPPtmJSWf7t8GrPn@A1RJcZS;yagw3?
zn*Y;({Y`Gj3}4l_1RCYALec;(j!>B*&NWd6Mj2|eLe%H%Y;N-R|K8u?&h2mUi@)?=
z@{j&6|93X8Z+zx;<DCPc@D8FlK@QSxF%L)dino02Yj5!n|Brv4i}Q2deeYd<=XZXW
z-~6rL;oIN-E|b|D_;<UMB*`bqGJiEbe|Ya{tid`E)*_U__=4l*bKZUa9!<}a4|+j!
z#HI3FQrn6}xuC2Y>L#!*wSqiLNTM7~gt*ds1n^(rFuukZgR@|D+%?%<*BWprsqt1&
zES9usNg^}Wvp%jlWihp^k8|QEqFBzTtAe_$84WfV_10Mx2Jae#RA>~#jR9ImOs8|&
z!f<0ZXSymlI4*c_`k1Y?jO%Y~vwtmRez3r|4&xmMhjW-X+`=$9pWtJGsT@_g!p4SS
zyv8egZ?K;3T*0{7)u_7S^_`ontq*aLLCv99U67;+I**99Byk=S>zMwi$6`KbGP|G`
zXT(X0m4e1YZ9wRh{@R#wxxy<;Wekh9g;fm*A{lzSkxGbEm)t5q%LqyEAoF+(6*V;z
zS1^%he{l#4x*o-4AJlo2_ekf_#uIsos|$*wDM>#=_!i?TvR;C(OH!p-wGP<}mz2~J
zbc8b!Z9~tawKhOj-iFG=Ih+ZLjPw>^19#liH8P2avy8H>@YWD}!yr#ttyW}_Wt0aG
z0ejr1et(O0sBl_^e3IH@+I3ddHYd|DkB{ML<yj;-<K3H_%^Ft4iYy7X?bX>SF(rG0
zhS#=RZf>;1@q|P<l#Y;z0_$<}61R*31(+8mjsSICFrQ5jfL0Mo#o@D1=p^H>{qnCe
zy(k&)$9PhV54I{H1xEAc&wZWW{h$6jR;v<;=as!JHn%nJzW+V8Ub~GN3!19Jw+$;l
zW8>N#wnU!?9~_YMB+H@(rx7ZqE(D_Xy!(U4j0O#&t|?oX7ZO`)DkG2{6bha7C{)Ht
zEjdm$5qGY$`}&u8<<8G?^OXZ;^94^HzJvbQkoGd-G-WhgXKlDfuh&CIQI}yI!cB!F
z^r75kG;(0`1l6i!V{4mV{?&bc;TL|5fAo(|`Szdu3IFV${Zszszx;KcJwNPrN8KcL
zb!GnhsZ#!Gd~(#L1tE!q04$AJ@Z|Ub?>u^!(VH1r9HL?AbPzES2ti{Sj0t9@QUpbN
zn#ROY8kkjFGMNG+)8nz$U`z;Ya^2%E(nlnbiqTS2mMfOk1!=y^)?kzI@SKC=DN^=X
z8x2^kF7VD0M+sVIND{oz4AyeuC}ln`h?9g-Ho_Y?T%0pIpQ4hKTethc=DxI?Ee_d8
zN8Gr(MKP<HoGn=&tTDN`;NvF`DVL5`w-{}Z%G1ko_BOWI+X;(D#gdXU9yRY%FkxeR
z4etbI@&9M<zoTv6uKG^!v)20M9Zn86hxgunIqE4Q1R@9|VW4bdzyXZQp&YC2;o<I4
zf7IyF$+$;%b#Pa?o!VV4m$5Mz3<xBH0Of$*dwMzD9L_!Y?D$J7^&e}Wd!K~fRRrTM
z%{%rzd57Qr{nlJ_&Tnu`{RojXtgOr-QbpBHiG@N%lDe(Ad}RaS4OyBXQi&=OL?n>$
zOl;OE%33q2rx;@qc|w|}<XK7@W#mytMtW=h5g}%5e`wKYYDV^esTr`>PMCD>tnYH?
z1mbTGm8G~X10HgSL3v9WMF<fQiwG%#l9`GmGA%+YN}I44+Hh;8Q%GY}n$3pJQILL*
zlTeMAHi#O|2nrss-V?<M(mAxQFlf>=MJENWE>X6n*HK*F@XYsO<|28kf>DXYMlIGU
zoRQS^65E60ytcK-)h4CvMx1!tQQrE{Ip%sxy!!Gbwzqe9^sxs>63gz+E|*^3;M#>N
zY+t^~?lj`ev7F^>i=HTPqKnK@V%s9+?3DK&sZ=n(@t(RYY1<~8xGD-Iq*P%?fJCGP
z{gpm(+Tx7FIgK&~DPuIAqYplWT3tcAA;LqRMx4Fx0bY21o9}$`UA*&q--<RmXnaYB
zvl49#M~<v>=FD+SyTNqQQa3FNa~X|p*}QoGiwU);XiAN(5-J-}3QHWruJn9ve~-#p
zs$vD(IZB;&m|WVVeBs~mH}1O3>lgO<g`eN2C?r{$P~-)@?h^gEMOIdpId<$gM~<ws
zxVS`rafv)DLiG_qcfl`1N)DTs9x)SwldP<*^ZpNhfcJj)ck#t9KEu!d+|Tigzx*G#
zc60j{^xOMG4CLWi`R(|_2=6_*ZHVF-AwoDBTeCmfq0$qM%`JuGFlz#b21!JO!aGmf
zI$8rHkSBspry!f5wd^g8j*v(!F0?$2CYBMxf2nu66oM?tNRotdw9o$jCRtR_Nmn^@
z;yBl@Pq=zzNLqB5?=NF?e#?htqNjN0Q7R^iVvN?*b<J95jbls4Sm~{Cb9aXuyStR-
zgd|s_9YxueT-)DdqZv{S1P7ZHJztQ-BR==)7g(x~;H9VN<V3Nc=*L8b<MQ51Tphm5
z4Sj)2<tv;YzQllSMBz}0Vt!$fNOp*d9GiI#svU;Sgg6cx;_<<d$a~UQg-tdt_z-D_
zREjv3lueCsmb$LN3l{qG6ls?%PAODDOgx(+&F0xSG!3<H80itE88d0dOzSCH2vk?y
z(G%orsz7E`5Dull58;_wI&=gg7Pu@6eZ5HHr6b8BL<ClVbsl3a<#0?{mBeXGzdw)a
zbud|sizL#4nKsy_Le6%r0)dQFV1;{w5*em6)K!B*lBXGMQ=_$}Q}i%qADbj7l?0}!
zav0HICl$U<sjE|5*k0oKUCE%kNblhjoVsV8Zl18SJ>l7>FR;72M^#r$stN0Bt1PZ8
z@{S*P7s|bhXFl;co_YQywyw<a$co@t?-D%^D$B4@70_wU1&FegK_aF0G)+63BOQVU
z0!v;<8Q_LO5hpn!mb9%V(wa04ycA;`y`@!-o;r;hJ`H&Ec*xRxAD?GD`@+k-^*xXA
z@FVx&e9M(9H@JRvm$lV3R8sQ(55AWl|H(yOfBgb=U9m7X$3OoUzryt`i+*esAvL-!
zgK%R4O|00BC9n8~$$m*M$tZ+jI^AW`Y|%7ROcShI>ndS59FoLy^jB6`Uq3>>KToIA
zrQ7X-5NvJlu(yA}{@^pDS<3O_CpmTMEWO@5w+$`Aav=a@SQrkq%HlNR?Qef44?prS
zf9ywog#Y(H{bzjW*MEa)b+|M7cH;O?5ncx$a)d@mM^#Q3S3}ZXMA`{ZP<d;i_d(4-
z#Na);^%w_f5@w5uN@k2_KO+qunnVCI{)aIJ8QsS4y|GG(qlh%im<&CW>6BqPBu!RW
zTRzJC(grs!??be~0}mWwzUa}*do-@5ZtI`}=N-mcTCLeX*k`G~$Xqf{C&@UrzD(8^
z)TT#m!q7QQQkqFiIjCvs7Ttl(oe9HfMXIlG^yC>9@+JJ#5ha3JRJ?lQ6|RgfFjfb=
zzW)l>+H2%zVg5vi`PD92k&<;Jx-^tRqrIc_HO6WZl~7Knl;aV-JcxV=V?z>`F;Nx=
z8{#;ks>@&`CM5lCmu}G|k8(1Zk&~fjbE$x~vb0`P+9_i_WPF%7oF-B+ouW@q!2TV?
zOz!MR$n7Ee?HylWjUU3jk|L!@mY6DN>Ivz5kA<~)>b*@&H6|@`>|{jRDd?sNSrpMs
zr%X1tiJCrgaSoNG*w#`_Mqq*-l`$sB76=)vWThvTk|fLV#xWgF>GrzBNkUalkda5l
zF?+)))&@aFi={CQz809e$7`E8&kbYp2afQr?|cF;bFRH|gQuT=iS4~X@XHVasU*AG
zJG}nV4OFZ-cK0f0?>Wojxij3DIBs6wLG3pfGv$=qr?0kAc^XWwoWEru3&Bp;7&9Y5
z%?9ppi=X0^f|+&i;9wsqGU}?NR|Jm{q#{Z?oIUptdrv(T5Nd;cdJBrUpR=<w;$QvC
z5Al8Pdk5zpID%A$yb$c{-5`o!b#;Z#oa8NUznA`8pV44K)-4$97)q0Xh#BdYX`<Mq
z<%aj{S5U<Y?|ba*v`j9%&Ujn}9WxxRwG8()3-f3B<3IXiy#1X|a^%Pfy4^lW8WTn8
z77c_MRQp$c_22OCe&H8L;+&&LPw?oYkMrmwk8|wUX`(oqd0dA*XZW*lYKlnWogqmw
z-u3QxbMC(T_@4K^mw)&_{3Bj^>6P2Wt#8L4UU-kU7Ue)h;Dtr|79$Lu#SSVBs(y!x
zx@JZObl%ZwOIZcDbP|JBhl&Vt_AjK6+Z>Rv5Csnqhjx0FmS)!%q1w;V3?U^|*)SXp
z>2+(0e2L>HkFa%R$o1zYID3r~$NQ`<_laVMl9D_xzzdAgByq5(9+e}ck|Z+BRj1<t
zDo<HiS_*ZPi)mfO_U=uJ#R=WzmW{|VI&ayo4%n-9uysS)CRhO@rVQ;aPK?>L8*G>x
zs1wkCaE0EH97~IpF{PL=_TymX?L3pFLQ05~U^*Golug)?d&mk<Mu*C7Cfg*0#2Sm%
zmMkfln_Hw?^eLj8M5ME{901Xp8cjo~YbMQvQ9GbC71lXAd6&iR5?!@IpFQ@zp#;LL
zgtYDrUxkwg?{L-xfS0p)t;v%dqg$p`Nw-R<rV~;uaTH@CMb_^S$1%EXNsFAQ)r^KC
zQath893qOsLfUFn6p^Ti)@WMYVj2rBh?I4*EX+TgrP7)-%W3N=2<UXX>}-x{ZD@~J
z?{HdyndjQ%2u}|q@_X;%_@j@ov03xN7e2$m#cM=aPN$!vjX}2tB^Ay}@Hx&mw6$en
zF=cTj;lH^5UHGx))la>~?#AnI@H?DTJ9MH&f?Ue13K8N<PX0qKgqSG<dq6}9o&dFt
zyyNP%3(POABl8I1EeeMd0#wTS@w4nbtAgX<{tnrqA@3)&!<I|0Z}ZPD|2v+1`#HYv
zeea+#BUBV)jH0Yswzm)H_U5>J=>jjmc!9}qjMpU_S3z8x^2*MHO&fFF8Md{f(gc3Y
zaKxzI!I*FkL?W~#j0auAd%y3!Wbt1j=_SAa%$bcg5P|Qb+m`+P9g-wtcYl)?UVNGl
ze`JZr9(gP8eAm0V|Ni@lq730ha66nGKj9rHMI?eC;M(#sfA-J(86J7$5&p0LxBol8
z{2zV=dnaj#Z^s`<2y;m62`EP+1tN{<%=J+!ow>%=v}kHyGd3f}^_a45Xqv#gNn*%4
zaJXA_Z?+wH19{LGLmO!_X;8CfOe7R(oRB67<7vZqG-fon%X~lO*!lt&k7Vq;JmvaJ
zHG}<z^`irpmlHZ!LYn2Ftt=HQ%lYBMQMENzwHVh1y*%sDtpO!TBZU=`TGu$Qkulhr
zeEXFvJN(+qA7Xx3aH4Y*8+*iBLg(I;T5eKz_el#+=X6HCtROJ~jw~X|X3E4&$<r>S
zt|*&|R4VqT6Y6qGq@XcAs0~Lg%1BZlKn3^!GVmU<G^1b4Q51dhxF9A9el2(mhL(oP
zmy~+Sq#9E-6}78Tv7+ByWGP=^PR<k2MST^gtFJl(IfUB2(GbV^Ku(j^;Hw7FXu3s#
z(Iw;YfNEe#a70F9XC5A<(d6A6)kzudPMBNhQw%(2FvdGcv@}mzWSF`Q<#7~@pHviu
z#jG`0ZK!-rl0_uFE>&5P3PF))jI<@oG8!M#;xS%hw5J*xwC?l8oj$Ldgym|?T)oRH
zS6;++rmQ{I<^AvfE>4{}%H>O!sHU31LBrK6J1m_z!pU>TS(@+B&0_|8W1fBH1*Cxd
z_)+#o9>xb`+t08bmsqD~#y}B{_LdW(pJ{7xn2^24ScgjUAgkgXDJ|p8OGLUxb^ADH
zKn4|3WJG`E2*erQLo+y_gJxl|V0wKE6Dy2WeEu`9GahOlf9pLQJH9}grc5V>%a<<k
z>dRMn>BY;`6Hie<KiTDrpLhm+bAwF;SB&OH<r(9^LV-ukc6yl<ND&6$#(MJ30{ulz
zzc-I*5}Xchj_y!Aawb#r#yB__vazwr+WHDgd8})>c4LQ)o0s_1XFkdkPrQxq{oeO;
z-?@i~;uPx-iDY7iYLl~hnI=+#M<02ZpZ)v)C)U^3`2YNi|B<Txdd(cZ9p6kiC-6R;
zn_f7q7l<S!?eq~Olr&6eX?#PiD+c2OhNA(4K|@u64r13>$Cssz+xl!@@_61OX4jJS
zb|#B*Xrc>JNz$xfs-`rpW<1^{&l6U9^W1Z8iRnQ}x#<~id34)S*P8jcl1`_lQ{;4t
z9z~WBM@i@ocu52#QA`pS7#FZ#!b=+0Vy&l03($_X)TD`}%^{wX_{x!`G54H3%Y=qz
z#SxuU+#J8gR8EKzNQK6>6|HEILNSz{y3*8bL+cIBI;?3Z8;h=LA^|dqC|g70A(|mo
zrLeOWT#$Q-<Ah?dpr7<9lEdT?^u<~l8eh}+n$k=e*CQs)80#%6iJ9*%vk)&cFBXW1
zF|!%go#M`(=+0nmV9N*@bEuB#yhd0_A}q3<LRDjEkwR1CF-%J$JWdI;a40Egs)}h6
z^zH=`Ej`72kIGo;vO;un(kwv~DbwK)*EB)RR?X%HvBX*xhDJ`2=Q&XvV^AcWdE~eu
zi5ylcN)N5`K?_REb78;3rD=lcMOZoJ`IkP06*XQ6D%<6SmtSIWMUxhLEG#c@^6q1t
zKG$b=Vwp^*WGdyUPd~@mlSeswdX<f9+x+$?K1Q`2@o;>EZn8&b^gMWvO46C;Rv=(n
zt8ZCwIv0Am#<+lqk5Xa_jdwHx+MAo~Ub)HYxyR70Ay#5GSQN}JF3`!t4#`Z$w1Z8K
z9$6u260|Ex;y&GCkqZ}I=fl7JTRiyKLwFUlbz_UIo4ZU#4Q1n5oSWnHl4NrEI@hnC
zXVeO=S<jw_sq;aHOhDs33doZv7)ZqtRo&vG;(OlrKJLEzHQLtFmMuazq*;QB6?I!O
zosOyNdUk8jM)TBDPjThqRqAp|UZiAM9EO9crmUv?#)oh6^wUrA&Ue0t@B5w~<gR<}
z4~%*nxH^Hu>akMbO>kK}wz|rH```RGEX>dI|M-9X_l(EWZ-cV_5F#`oB;G09Og-Cp
zi1Q5D>(O8sFl9=O@rKs6Om)qusu`DIGnJ<hNwCIaT@ZyjY+M0v*s#vLYOKAbHD$fS
zd5;n9R@EY<BrkfT*_5i9Qcm|7jV$@1<G$0YXrp=d(>G~LsK>AudyMt8x}>ffn!3d=
z2<CeUxyX<rA~_s1&*YChheD4nLR#imJGe!ST{oOv9q~fGg`AIB==WIKTjbP<6WlDf
zId3*FTC+pR#Cv)U)TBXe8KPLB;sh0&Abu$U<w)Wv5H77F&Vv;1w5oz~dra^enF*yz
z5rEH89FasRSyWJv-{RrSd?P9Sl&P6g>5@`UXl#W{p-8&SB`eISWioOi(iw<)CLeQ$
zxTBoAGY#J9V1Fua`(R7&klvC>g}`EM&15)5^imcU`k2k_;IW;<v?@uprifxZf;21W
z<XxJkA@6jFdtLOjrmY%^G(l!DkxtRm5~mxmq16*b@l0tpqG@Vs?<qPRJOY`d<eiMP
z9+M>C;|QS?%Et_v9v7w=buL*xZHRJ<H_I&b7Fe9`(n}ZE-`(J)7tTX6<m}m;&YWVc
z*W=!WHHHV8XP<frTbCqJ##}!mJDDQo0>ATH&#|$qSrUD+p@docMaUqT4Pfk{7dbfV
z1D4(Rz*3K55(h>~Dg<KOVE6V({Hz7%9QaUt7V~|gG!MggqtSbpdFZb1;F9m+Vog!x
z6nPht5!bF7_J$)~J%5EW$4_zW_;L1j_n6dE?mLyDFFwQGrAzGDmJ8ZZxd^8ejt1p}
zw2=~mL<YEKEflebS{qKBIL>_!+=u5?27^7WUw(<L?MrNKZg6vRgX^1D10!}irLJm>
zF${+Tc6N3F!QSow?;(yOy4?<WPm!c4aTIgy`bBR1;#EHP*-!AkANXT@|Mz}?<>hs}
z6*z22!hmrWw7~-1BICdQTYsJ2T%W)H_x}Nd;h5P96ntW48f)K<Z(RsLI`B;}mGZH}
z$Of5tL;@vKDyCsd>Rp&w2u~DAirmf;2eb{u3d~`P_e&DU8}w~v=VQR}c#Jb>WdL;x
zv!z5ttYV@lLsC)KhT*}GA}#3k*0|^NarTB4mtP&=%uKSR4!-FIOWjQILeb3%vUCnf
zG;@QMXao+$A+=bMBCt4-kj4ormUPl(wl7qi-`S<yFzoJ(XeKrJT8>-w3@7`z#3Evl
zIruoBON9*}8Wlwd<xmmiSxh7)#!7UoKm}td?PqSHD#%IsNZ^%3sfZ*gh@*^1#H6G%
zsTYqyV`f?a6Cd80k`f~fDv9Z59TuW_`YaQZ&zKrPHYQBDZ_DX_rMP1S639CisY37l
za9byw0PSYHh7?1M(;7RR(9aXJw+P=rYZ0x6)&`-pEQK^eW(j&U!KIF@Q(zpl*3z^Z
znMR~(j&C&D2Kb(KHHmd7B~eO{Br#Q0Ap~@~Jt8GY5=C22>E|(>L?H#VreJf>W$y0N
zyz_xmEZ@68R799bL8el+w{J2SPFOlI&*?KquyTk{lHdHT5A*nwkMZz39_EDbtewi)
zxn6PO$^n~u2T0rDu5(A&+TG&(C-xcHbq>mcvTA9{Y0wUnGrog!9Ga>5nMRZpp%j;C
z8rnWWqMS#!HF`8cH4RF7ys#**@Ya&_`q+FP;~-8WZoc*`i$#w|SI;8X*ND@c)_6Mk
z8P1$uLx_khYDw~xG|gF9iP+h^#`RY}OLO^E_O<3p>lq?IIJ~j59ZisvDe@GjHO?9=
zo~G3};KuegAN}~p`H>$R@WQhn<{$m{Pca^EQq@zUC_<{x6L-$d91#uE=@jop@KkS2
z5UlhygTaI-mGt`sbMrltBq5F!7calUzx<asc>3wj^T+@APw=)U-x)N<oIuJXz`<LK
zH5Ta&Ns{tk{>;zdo#$u&&;J(&gFE+aza8IvNaGPzSje>34BP=WZy?f4?SRJA#BmW4
zxG^XZ2!q}^s?stXL0JaN*+a+ELkO>Y!%Z+lWV#^S35PZq!PeS&ff7O7J(7wvOK7Wv
zwyCM6j=^wDl=qlR7rFnwHEdk7ajizTP+CwHasvuSUQKYWB1RIUZsEf8Y%#U7i&!3B
z(o)c%P+C!LR`?B%-mujBCAM~SIuU2jo~D1SAWAgG7L4KvEsn`-UXeOSqz#4=nYf@!
z<?`_A<G{4H)&#Vh3NKy+ZZ0xQiL!zu?$C)l#6Iw7;+bZM#i3n;rlgcDV^LA68Z9kJ
zk~1f|tjc+Mbcu<vw`vZ5E1`c=Vv5eW84~Jt^#S0mL$-z_cI0`&aIZn3k=9WU$HaL;
zQ<m825Z|;o=SVvpd=lYeg-By?j;gNc_WDFcimqF<F{D|HiV}oU2oWJ<(4T5dJ+p%h
zw=|8JjK}CklNA}2wM>QwEEExmaCm3XCTG7{K^$2kSru$<?6TBf$4{nQz4{7q9`o93
zSJ>a%BKe_r^U%Zhv$cDj7oWd|cS|Jc%d8(?;K-3>?tAzMM~_$h(l7ifo7cBl9c=Ng
zcR$4N`W1FwzeZoI&}fIzI#gGQ#JZsS<LzPcIGig75?!2Uc&P#c$rxO#@w%ZMjmW)5
zTTSW}!aL$lhj?K<2o*+(!i>4}(SOI){hC_!NwR|FBPW<&TBAR|!rbBlC+<H2-DMh;
z@%rm8@xp6gAf4=B&|Ei~JrqVr@S)qu?85K@rga+-WZnns(%Ji8H7%d`^e4G?-E;ne
z;R{b+q|+6|Q50-WU4SMB^q#{Si*+8QGV0a_vBuf1#(5BprJN2K98BrY7xa1=X&Pf}
z%NL&dBv-Cp<OkpPL;T2({1{71t5|DD@*GcK9vkOKj3<d={^DQw3s~d$KmP1Lpe)Pa
ziSq6E0|)^!#tDyz8m9JuopKA)53G-|b{H#&B9C_#XY34(sA(EU-GmARKlA5!0|+84
zSn}3xAHdmwz>1=1W-sArObbpUgd&L&q);SDOrGWFqQMxAZVjVh(7j6LyR7#YG55Aa
z=?ykET3S6bvjS2zYn7sq#CRk%hSs-1QeJS2`>usT>KI*ls@;<M^@i$FOR?YQJC2<o
zIiAxx!;$;fkqeNlCA3pVe>~*q(Is}wF8l42gMCLVTY7~-2#1uK)>^F9#41Kb!4(vI
z5M^an{Q-$m<WWW*7o<^6Dw117Oo5=8)g7&Ds7=YZol>EZk)$8>nP-kU`pi+xAk%vG
z`VMQrudQmoLjv(=9gx^!mOu`f)e1{$D-t`!4>ge=B77S}xEhJcBuSnzJ=h5grcy+4
zgcf1g<?&!GZQEjS$TT4qGE_VcWTZmBQAmW0@V>=+8w3_zkVub{n8JA`!!grpf{P^9
z)^w9Dk#k57S}Vr3%gdKu!R<_u!$tb>25-CX1li~Y=T1C;8J^~&7xwv&zy1=R{MdQ4
zACPs9@tIFt=d+*O<MFqg<-P~j*xlWtNHelHr^qCmSGPH~(l9m?Dq%2U4Jiw}_f%Cy
zoW~&nTZgj&+b%qKYia8epJu2e#>%i;kj`V<me_m5c!D!6TD9aPc<XQ~qF7kMdnj8?
zA>nimWF_TrLcM3HH(uwO6=*AhwRM)$E}f$8uP|*YqGpH`o~z@EUC%6e_}ROCyHIVl
z{qj@cOV$o=gYUHU&>D}QdW33c<gr<@B=n`-Oc>L;D)f)N|B?Z_@L@60+Ay7t7<DS<
z<_o$#O{e48+uP)q|HChE?b>yI@+W_abNAej_nss#@JJ@ZA-dHRNsLGme&)~rG@F}S
z{G)&Tk1^j)Ab&6+z$=N3YFsp6x4y<eZ(|aR5fUp>q!MUtXxav24b!q_JZdSYes&#m
zYhHC5uqWTZ@t93q-_ZGRjxZN+4uJ(1V5!;@%-cMnos*>`Wf(ZOjbS_<k>&-xe1)Za
zk>*H+HQVg(JDLV`191vMDx^pV=Dd#5TWaSJ!Vuw+!eRt?p|Fi*dZET%b|g1)vV%D!
zF;Sc#l9<yE-NoFQIiBBqj$u`@wKHI6Yr>IZXXqU%SkSv{?+>X;4aQ@23(7$h+J4qL
zyf;`OaN5#TEp-)4!jdR>sT6UCJjzLAf<WO8iPtz9S}JNvH)U!j)J;XCV!Bb6c^2r?
zBO%2jX=nAEnn9-z37v;r|2xH-CJ^sEM(Y^^eU>~9ndq)XnG)G<Q?{G4VLL>!M<f&`
zO^M?a<rPt<1J=-5gKiyZoZ_^_YJ-ZCpcggEv_i_D7CGCs3Z;-z5DAa-2ImaUdYZ<P
zx|AX*An!10r?gFtl8#=UpnNzPw6?TfA@ZPevwuUgcrsyeuAv{*bh8oNG{tMh&US-P
zEk)mv7X^}xX6m_lWsi3}`49s;;rTC|=N<2O8-u+KWL2<rX~K=|i*zD~7d@2Bu-4Et
zE!GC+zH_ry3ku=ESwmIU#Cd_ta*PkMFy1=6H574#9**z?u{v*pB)b-tUN49+j;fY9
zZRje&TvsxeL<s9?wZYm@Ef@o1;kY3yo_Y0oO7F>ig=sW5ourK%I74Iw4vG5`M)x5^
zbN1m42}JlHA|dfO8_ow~JYu$1OCe@*E}@(1XJ{%f4rO)&Pv$Tr!R$xOT<u22P}gIs
zN~0IT?<$q>{0q-A{U`sJKlvwrf+ybkR)mVkiUR2c^>~CcmMBh`>-G6tfAeo}_1bm*
z!>|6z9Z&vm$2T4#ZfW9urv5S)_dd@^T_sO@G=xF0ci^;UT9-^GQwF1!$r$Q7oCZQd
znnRL-6#iD@<WN5=46DPbB&8yb6O@!;lQOfx_FiDE3-Tz^Af!Y_k}OTJMS*vrEj~UN
zqf|n6v`bfZI5Bqw*Op+%?Cf}48-~(SMMx2hmKrRLh;T}hAxP0^3>wdPS~0rR5WSYM
zy1P!N>VcG4G{eaNdM%S?z>N!6*qB^nr`h7h);8l!!-Ee-967!~l1VyQPSZA&(+ce?
zh%GWvD3XAjnym|_G3dt9l%Y$|?JDMqKHa!SM-{VqtH8_#<}ED~H(_E*CVGU!kYyQ5
z(IS1-r%M+#<LR`7O}xhuA!ll20p0c$T^7D9?%2YLaKtLc`x$$BW{n!R71To#QHO_X
zD|Fj3GGkIRrN{!|#<`YA&d^6yO*1iMPGX$JI7t!bL{S>Vn^j<M3lEJA<y81O=<Rs#
z(YnQB@m5ogC-}^gBq^znsH=+D3-T-pHndKJ>`j95nl#IZV!^?v<%J6kN)<G{9^E5d
z-uJ${7;I0tyjycn4yjy%&_Qr)us`JdYuC8vp1U}H^ekCgP)>G9#gv1xrSY1%!lCN`
zYg8f--@JDi-O|=I&UqpyW^a@x(vGCFKon=FR?`ZBv^A<WBnsMkfVB!?EIzVm5hINu
z@ApB#aAJ{0)9VR3ae+xBF*P0I&^{xLB(3b>nhw?2;)KC_&EBNKs+iJ;Iacs@d7}xl
z<4ekouTaGv)+@Jniox>su*-F5DfRpQ&VOa!ZFlRQ!U)wlPhEL*Thde(-GUSeMb=?+
z=PLj9=l>^m_qTb^cfXe+?*<&dRJ3i0^A0H^jvqbF-}}3Nm-DZ^#*44M5>n%r7*~Be
zzPa#-sK%%PHx6Fm%FgrX<3qYK0W97(SZ4!j&O6G+P&WkgC?8OMNgQ_4Q3S>Ylu)Qb
zW-gkun%Lul68MaT9;9!?8$>YzG+sIn!r^?dtB&KCG)>Ujq8mp!DH%=2^a@F)Vpe+d
zjJq}aJ1tdv8xiI`GoqqH65($>D}i$`otA8zA5pz3IKF!tHB70Ur?eBSsEHOJS%B^8
z3L85&sAUDN#E%-<P0QZ)nBxyDktYSlx5O$TPjebuqh*6o0!0iy*lAm9iKRh$#aNc4
zG4#6~I$4)gr9>j09jBla))Lm?VRb-jT8JIptjCgCWI@c65Psjxbaw45b%%Eqm`$GK
zj5+cRS0H#u2?MIy%--0UNxG;jR5?QC7VE<RxQJuOBV=Mx?FdstU61J&S-3yraMOyB
zvE<z@i${;n<{4?QQ=OHo(ps$1p-i1EOu_Xm0%?OwVI0#omg#gtlQg7pPABb9S2pNt
z&AiNw2csosA~<qtj=|i3-Mt%p?DzI@I38>+aq`S6Klt7UxcZsjV|{&rvCJ5Vn)7?R
zn5s)vdOq>7r`g`#WqEmtt*sq)_6B_ScRt9|&m3S(PFzee<roM?S=QN5iPRNMU57T2
z5s(WIhLng-f0;xjM8;4`h;>OboD$DV5?Y)QAS^brXps^dNU}VbAGLx>V_-32uD2AN
z1HHieK2G#;Ntf+|5tqlFYkMQsRu_pA$t$m4W2-XQSwA_7B5D(`7I$7Q?({wXC13k5
z`{`GY+ij*VVFNhHWUSG8fOpIW^ny4}`0$5+js5*SKKQ{OWnq2^8OJCeV@+tRD;4qh
zqmS`7|N39!@BMH7*G$WbFJ~KkJHEMSn<<w!U*`)qKF>vSfu%EZ2xalE2G;~N24`^I
zVyvfb0`f`9FgTxeoPraiyS?xq^3QKAYJRqaA9{@=-XOXvvDV|P!wVNS35403lmewx
zfWz9BYEm)SKL8<E?DUz-=Q+A+xp{Mg(ZJIHO&tI`;80j%3?4)4n}EhD1(VASTc0~X
zY^0=}4CNCV@2Q2xr-sFo^R&r?vEIie6}GMLx~1B8w7ZV0mv=Z<t<&pwaH^#>meyFj
zR4A385{a-uJ8I^A2iD`wfKFbKr77J`k0S39$Js5ofyGjz8MY%1rh8b|lBNmWe2%^f
zEvu9?AkJt5AOLYozFp4#4+1nH8A;Uadv6|hNFWYdS9qhbP7%#YMm*@M#5EPNP$(5l
z*~JX-<vr9*gK%LX^-AJnMWhs>uBgUSaGu`EQt+^tVV#}x2=8#Mp{;6yQTJ?zA)s}J
zMmJbv!iX&ne%_O5NkvPVWc243h~fk*9ClVc<#9qg@i+(Tr+UbAz}w&cAeS$0Qe-iS
z(g?H9bms#1Jn|%q_uNJI>^lEG{Y^~U<=E;P$Br+fB1e=+&P=+vPkoHfef~ePb^SC+
zq(~zHuEKj8+I%z4M_tvljlLzYC{RRELLPVN&My!rG1jz5>8PhubW>9lO9&HuKE0Ay
z8~moNGbCB?J<-}T8P{l?lNC7ylFCL@av5BQ=Qbu>xOS6kjbr_Rhv_dZ@Oz*AY#3G^
zN~}0X)BMpT-vuVSr*1Txo4XjR@m^w#M9+CX{_&4fRSiG>r+$)tuSXmQR&r>62LFdY
z@hAQS|M53J!oUC358X0G{dRow;k{=(t~nSBZVoFBGFZ;~SQ0Evuubre)@YhmQ@2pJ
zfCLrK_JS(3NWvm{TRbs*Japojov&d4%Nw^TJ`B`Ca*?5J8YGqo3eWz~a7v=3pw*7)
zXhIUFbdr=#Qn1!fId-&UG#pUZp7F#mnv9@WB1I6*jIc^uVq8YKZQ1z3nB7m**v*EQ
z7hdCu2j*B?KF0k=?_=n9@e36f_cz(EMs$|ukaf=B#s<@Acy-%X2VA&%nRAOrP;o@6
zBBXLSEfF$CiGYffDva}m7u2T0Xh*NtWpRFyJS#}j9HAog?6{OPOx%!hJ3`=D=`GTa
zd(4S>6fq8{(KE$eJ%cJJ0)8ztt1O4`h4B4k_VG^frU}GZ%k<h+<cYJWq=S%gXl;9k
znU3)F6zMymY!tyG1nIHHprSZ<5YJ{NG76b2De}<QbB?lUhzJNTAp+CcInY{TjKv*R
zAhXj(I~zC&RfRPz0#BBtc&RAMnyS_;t*nvd9a^jL!jmPQUQ%)Q@w420w$IkiMMNYS
zmlM9{2OngvJJ0ni7kTBzCAtsZ#nQQ>#GRBZiTH^h{{Zt#i$t*}ilJ$z)Kx=W_xQ+f
zZ}W*y{~qVg^jKP!xJikq!L&M{NJIc242NTM8|FC)kf?-C-ldl<68GnEQef&Dl?uj_
z2}*d1BFDBZ)@YoR_}F57khM?J45bvt2-?apn$#pwg)9oRH8j59x$6zjp1;Bn!SP4#
zWBv3QKK{EOV}Eym779DN2A%tZ@fQD9g(ncmgaD&GyW2yIfwnOi?U<jR=To2hINm#c
z?8p8T3%xmn45&dlQ%_i2T;wnR<-f!yKlN$0_YeM4zoh@a3GY1CM4UKrKX;`!8I3E1
z>H#^Cuqe^sTuD_ojHfMC<!Rd6l?P%r2O-Qj?5xT-+y}o=(j?eh&*r0qR-~Kh<VY#P
zg6jg*SGurTNFU~>UMieYL`sG>S<^6?j7jr|B#Mb-&WR&O84ffTFHRT@9lQGnRLd*$
zWRAchyuoP0bTH-G^Shk?)E<*(K-8Ya^+kT*dwziX&pgK3>LO2H`V3EPe4es#7%$LB
zw2P>;#C3p0sGQ}sOB)nN;K=bMl!)<BOzb7zY4A48eIto*;e4;E8YvW=PKP`z$g^yQ
zNDH%%22JA|D%~K3WUjZseAFc+CL@NKO|Y4b%PnqptPd-`kZi*c-Op}<XjX&VDc*dD
zZY?*z_!M`Yy$f~h3{EMm3xn)(bU<NRBt@uHyhG09O`I1bX@KQsX%<>eNsJVNq>~dz
zaqvTKTB^DsiX!49xP>VhA&Z=jHB2W{j5gs65;MQ7!;{E)nz};QEm4#pgru$-wsv+|
z5;0k~8>BD%m~OO58=asdmzeHfBuzT(?%w1>zjlES{>a;T`+FZ`XLrbCG~ncukKh|k
z(d|M~W0j(=YLte_MDy%(FY|N%^Dl7m(iUgV-N(|Bhe)G#V&Zy)vlZ4_blWl>*NjGE
zoby=giK2*Zr^DRb0$J2U<(-*lxd11rh9i<BC~A7$;<d&riL*Ks)22blpy2DgqN+60
z%Cg*dB>6nvHeBBx@zj+A8Z<}lzl-}GeTd)s=x5lzev^oBq5kKCOPcLUy~lfnF`k3{
z5zab{36Rjaxp_YR@!!E&!=L%_KTFZ+qST?7WvGb1<K6G({U3Ng|Kq><*Z+xM&wu`L
zI70Fs%3bK>>)dzrt=!zYfp*&n2qZR{U|mVsw2UjyxDK-qGsB>Iz*=Vh6=8!QY;Y>I
zsM#2Oh5#}S<syhOm~A#;rkf?jZP12O#7-cc!+DF1z(+P%S|<^)fYv&y>6n3NNFzz7
zo3hq911AgI;1#YOjJdWy<@j`ulbtoZfXTF_+^pGqdCKsq5hpmxUGF)|nU!Pwi68m1
zeCLzjizMgz-b?6q#IT++agG+xaB~Yeia63+WN`n4!9fEt>>RXQcx{vUe$ISv0h}Xt
zfs5t5L&Rb3Vx0}j8^)6t9XeS-66YjQMl2)(JDVICoY&yPyklPViBPm?Ce%d4voU#K
z&;$<<aSJ-5@T5e<w^B%8gB&&!?-XyIK%C|3(_i4qJ@;_5*C$?zf@Q4p&^E-vqvY(3
z6an>w2z2@=3QX)MNs!JWy}$`@N<t)q5TkL#QlX?m&6FfP4iPKz`94V}r>Pq1vO+f%
z!hk}A11A+SNpa58YK?Fb>l}Mqd(=j8_qi_KLnJ5k(;arl1EThB*5=Rf$<I8`@_IzK
z*QIHpGv8r(ZH>JfdziMQ9#_=5fojb5XvoF$mwEBU7unk#%nWC8o_yy!iM?X?T1mAX
zbC+GgH@{9CwX}fNE#rDjUAI^VQVEJq!TkI@-A)g0B2*G%tiiPqM-lrwTl9KeRHQJ*
z1Z=C11D>50r`uaQFXtt#24fsD>yV@grgXfvxrKHm#nBZWdEy)|yznBI&tD=U_*ctA
z|M@@S)yM6aEfzLdZ(8FS4##-!&{|`yVPRpOkA3X7iABtx`thF(693j<twpMcqA2+3
zpZ+O+{lgz&XLs*g@~r-sK8ttGGZ~gd^M)vkQC^^ZgQG=>2qhvs9&H`k1l=oKKnAuN
zvN-slM-og6t_>|IJ6pD9+;UnhjnatNp{zl884|`DcFqva58(wWiSU8YfzT=5M?@|n
zu^MSwglVy*#}6dcfuVCeWo_vU_x2y*v4<Y#x37JiFYce`^(#BXYoS}Pb8X1x%lovi
zS<ZFN@~8gXPw_nue2}HN6*|c>hghZk!4^KM5wXBZNuK1`Y0G>(&tnhYOIh#Xx0)bV
zVqj;xX5-op>FEq5B^Zq|E)b6+g*FCjEJ7-ZPKP8)$l{zd&WMCM#5e|p1}xrNVii%y
z4hi`!R9l0?VP-9<Bv37GO*{@is+nTeA$8Kt(o%%`DsqW;kaL$zcJ_JwGoPlnzQ$6N
zW4jg=MdWEhhZ>2(5Tkqxezwz&!>mLqi8l_XLgiqr#+sSipi;;<Ax=WF3A2+RFlc5%
zVgip$B08#zO%iMw;aWo@8loup3eaf0x3m!`VbD&nH<g_7LAxeOYL=ogyTbvK-3j@8
zho#;Twl7Q(`E`1|7g$=I<LsH!#A(dt_3I>w;>D+5<k$Ygud%b!Fg{Sk%(1d~jHShs
z<Xy>VXUg$2$5}j?acS!+xw(N-H%N2E&}xR`ma-Y+v`2bD(T$i}$ylCWB2F_}J;4bT
z#36xx>o9ve%+GakLIno6o?^`sMhd)Y5jH3lhnn7DjYP>fKo^~&suQ**nmYG9{`N=N
z+1TfWr(VXj1}#FLIQE{zS@buze*B@tmoQ*OAe$+nu*P$6Fbx`6vrDKy*W))n`XT!B
zeSYKvKZ-(NoWXh!M)9^M-^TZR_jmJef8iHz%gld+p*1}85d4FG5O&c2O9^Ab>jp^T
zhIhT~S4lcg(WVpb>RrY6YBF{KNuye-qM}lcdJfu_!v=d$o}MKSA;4)UvXVvhDk+z_
z%|{Z{RMbg@h#k_1Fk6}ZESO1!a^@B+Jd!vS;PWB@msxzzVU5F?0H?=-G7iN7S@s(9
za-V+OC0dvK&|JxdeNPluY}%Xr^5s4^&kxzZShCpkdC%Lvpa0^WKh0dS_WO>{nPX>2
zY52^I*T_`P@wF8;Hm-6cTIcaIkMPBf&v8)g<1I`lLB{3g%>n&>mzCvZ@PRWWWYB-J
z-Uh<9kf=B!i4&4IC6S>Ac<9$?W(h<}Ng^{6k#NXD_Z+sS<V?AECZIL@C}yBtF`IlC
zZq@y>S2dFuzC+ycrxL~F+I6mf{_|wnJjJ;>!06%_M1%8THynG96p}d05otI(3>Fp7
z`fN&~e1uTog-1kTL5q~aTSr^B7~P`V7AHZ;Lmz<9i;I;a>&>w@nlPADbn^^|C?`{<
z+9G<2MHOT;HU^S=wLnG^5lfc4j=h5&wys>}fd`+Y`&}J=>o;Danx5q5t0fnfud_7Q
z=fziF<bnGhB#RdL#BcpNFF&hT=|9Rs(j$o!z7g!)48m{sJh;w#zw0}A^>df0lNT}j
zpJ#QgN0LmLmL+8gRa4@<BTEzJdwmub`{Y@H_MWz_gAk-P)DF^VgQ}-=^AwMwt!pw{
zgNV>3cp%8uVRSP)CKll!jxrLZ5MEI>hKaG<{q{!?O7gj<USK#XNkot)agjos+mOBg
zoZ_|*O^~#ZQczASwl-&e&BBv+3V!LAf032BWxnH`@4@>P-8!PUpwsE`qd)w^eCRiR
zgUPh~rpD>Uz_DZSQ$O`BdU5|5Mig1@IrAFu8UXiv`7cNaLqDMRwcMiycE0kD4hxbm
zcMSMyp=P%?{ks|%12=AcXElen91EZS)|DyGf4V~NJKF1T@A@%*<Xu1U6_W?R#=!<x
z?N#uC)TeA;+@>DYJo1)@cwqTH<~BQYvl&IofuF)~VA!~^MYq#IsU)=dw8J}%6*F;3
zDUebkl|srem9Q`y?*~IDCmd3wL_|meEkfU_Bx1rcuV$-hNFcXGVFcn9l>bl%K+I+z
z?i+7mK^STkjbrzP7g0rzd)8LM;7i5mIL6-+3O3<kW$-(fu|VZ9ZQYP~Nu(4>6k)W)
znUGn1)i9k_OvhuYvP3s+Al!)=87)%8NkW{&L|IClb*QXoYj*^T0%a|e(o$PN>@_a&
z7;?78B?sdP!LuYm$uYg;H8yVuE`2WNE$@3fyL&@E`>E>;b_JjP*sH7@6P!5KA&nFM
z)xY>PuAeuo_8x~eLdclb0PsK$zbQmyF>b`&=Z^B7?|Yn$8#`S5y{Cys&yjF}BS(*d
z*kLlMC`*Btjx-hY=W-U8mg#kSL{d;Y&9tfM#0gE^BBer4CnP|gBsguE)>Aqr>?*7a
zdQ!qzoDH>ulpdu*qK##Y^OF5)!u*jHjvqh97oU5ME1TO$0hx?Qkc^FE>@9zEaaVT^
zG1C?cUkI%Al+%jc-2<X1Vo4}`d%(Z``G3v&(W9I=bpqWQd}Q%n@%Fd9oqNxn<Edx9
z_)R~TLxTHY0N1a7(|32ra4tNawGhX!uu#zN&ynXjaT0vS<yS$<-0}XsL+h4Sx0Gec
zq?|IH)U-`NNJT0jri6f{B{*^<NU-17R`?`^*kWi&@>%N|71~!`2S1<IV>P$Y+F|Z{
zs~Q1_V;MZuMHCiICrD&MN={6QB!!v3dI$kH3>w{PTHRocC5j|UD!dm=CP8Hv<ji$2
z{GD^bqQKAQDZ+cwIAU#ifksZr)DDR<OeU~<$w3A7Cfxh@W1Q+f@TSM{!nK#V+FoE~
zZkc>(f%7jujrT40EZxmQyg)2eP}i0omxqmru`nDpOeRz2=954G)*7!YP69Hln{gbW
zq#~M)yL+_cOsHvSeaoRIN@!ArM9|O>gda6An!~e1aeH%q2vs<gS33Nj^b#cmmWUqX
zJDxvZg-EC<5Hx%HT>AXySU7i25H?d0T1JdVHE{;sTOuV8GfgLyBu!GP@szr5QPz^E
z2;H_+WyN?fVzM(}bTD8#nP6HSt`rwUlcf-NrAYFWD2YkB9kRIvq!;viJ*Lx=B916J
zDTxx;vcXtK6MJeip`PpoNU#%lIcB*xWwe`d?ZstYo?GIP$4@fxhA({XJU;2Lz0oor
z82-(t{|%R38>3hN?NG8oDMxK3aV&ZGkyHHO4?amXZTQr${2uw%ChgWUJoJzt&jaRN
zYmL@W^c4MG#?tZvbMuQtkpO4WO~a(BSW!vXZ3u^%j_IW_v9(O<hT3|pv$w7_Ka*Bh
zGvrDVLmUf|I6{h;K~-~QxW~!+A7XcBo7Z1@9o;&5i6WH>j6+*T?SowVAJrm6N;w;v
z-+qFJLsR?=uIU}lg+&KoIG6+=Myyy|T;kHz^ZeY;|Ihr5zxKDt;uK>nLK==AKhAf2
z$Gg5I3FI(-`Iq6Z{FQIn-2t~ztIh?)(%gK9<HwG1<meI3oIb%j-tkTzeDDE|pE$wX
z!W@x`3ES*%te<9G%V@gK?qHwGSFdpK;zcfAIM0o%8*FXtaOLV1wl+5@$CkhSx8cA0
z?_e|){GER^Pjl=P{`fw7(G5;KyvotFqs+?%YFch`lk<Dmxb%9-!Bud(gbi;e+(ZUN
z7nFyir!t<r=YEbxN11bdPPvmjD4%4GBhYkdM=90BQP&MsW0+{e!KCD1?|_4?UB;sU
zD@*e{{`mcPZzwuF&cA+*U;5DRa${?ZF$yPxEYh@`(C8^fd6FU}w~j|nAK~x)y+6w{
zlYh@U*Zv3YTOFW0B(ucb@29-~9Uq`j-LH5I9Sug6O!i0E&Rp;&X%fz^p5e^e{iNjN
zNlKnch%E7p12Z1OWbD}9-X%!_%GJ9d9~B81+E+-V3;-j6qygT_CO^(=%&^M0LfrcQ
zo4ml^z#Wvp=kM^VcO()@ArUh49;=;wHlBY4*EUG20l^nO`wZ53?t9=UbNwDvmMBYT
z>W1Oo0lh3k8I6%6ymO3(BZm72K?JdN#3~9qsa}s*1p_JTEZ#Yc3yX1GRg_hYIv5e<
zDXR-TrlmkvHQl*{tdlX`>oA&@>}e=sK_87X29!fYo{W~&&JN}96*fOzVaj`W>yuBQ
z6kNG@fy=MeoPXs8;{o8-aZ2J`O`OEUk>||)T^@e)E>0ai#?~uay!i1i&>3vu_CL=f
z=la}px`Vb2%t%ZslH^6k{K5kB^NXZe2YijuEp=7V8jF<?m>?qJs*=<@TvIa{jhP|>
zDNYA35#tOvhc<OUq(zQ2iO91IrD9&Wb`!VIg>K4)&puB%m{LTFLQ0I&w0J6J2&_M%
z$>T5%6UJdOIsE$Yb1`F-8xz!-y%!7)#w3v-P9pM7htGWeQ#^G4LwxWDeu%bdkTWf|
z_q^x3_$UALpHqK}_%FuH`upF?5Q6pfRqnh0ULJk?LGHcheja@ALC&2!$I9|LO5V9A
z{q;tyQdSmvtSp@3{xgrmcZKg)O~dt#Ydrh>vs^fTk(Xb&#C;Eb40tIpz2|dAc|++Z
zjErVnwY1h_WGF9$i11O^`!>eoE0{^z&R%DbJV_vs483CN!4Ey}S$G@Edr$C@Pkx5;
z*AKX{JLX_GV6Z=8T3VXgQH~qD2`H<(P9EX=zyE%=c0EEG@<wy_{cmOExqUwO;>Qq4
zgp6Y1G@-##)>DWqHnTWk8TyLPU4EKRT>ljJzdP)>qX_yvD6Qb(2kzm?6Yt<FDJ%eZ
zEbFVQT-kh$%NMV6V{bqv8@~H3-^tO%(^w8jlbj?=@UlTl=s1TlP)=cIXG)gs+>$C1
z;?UkVobPHz<_n6&VFD7pX=~{Z_X59j3rh%06m;oj>nU#zFtx*$6AD$(%e%bz;%jvJ
zG5Uz2X~r0B*x%lv6GaqjYqaf%wkfHrin^>wl7yqHs}ysK<V8WORIo8MCXfbkGtQ8w
zty?DJ5!GnQbU36Q)|e(D>*l0UPB|W9tYK94xw0Q~p)9$pJI5ShK#dG|?AS-l_bX1i
z=O_kyT>rFV>$+qZPkGxz50Pb(m8C_tZ&nzs7z}neaeAFBF1WdUz~hfRNCZ6d@#i?W
zaFylh7S--YIelut-KW=ZI%YTuqJD8A$nu1Kug6?}ku2?imuPE2TN+(aX$>k!&U>Tr
zZG*6y?aeJ-xo{b^c7)T;B4;uU#u|KUF|*Z5$zV8>#R;vk>{J!KyN+{d`zBW}UnlXQ
z1)yw;QG&5^Ot{_g{G-ds_6QLS#H0*5N8)hmE#*uO#ozL#399)Dqb-y1gyAqH$zz;#
z{K~KVGLJs^7^~~+Xx$RUibo!Pm=h<C^ZNM<-;!th<*tk0OspSW<B2C8=R3di-Q0Wb
zUhcW)ZtlA4F7oVa8X|nlBTrNAIdeDnoVgo-ansV2|9jvi;w0uRZ-1Dl+W{Br>r^#V
z(>h2E#dbCViAZEbtOQcbHZO<hjW*mY;jL9#AiQE?G`m-K7<b^=i<kLl|NO(;*!Gx6
zq2daoHI4`+lc1U9Eotnz`@u6DKfA*2_EjFb?=JQZrfhBRQK$$})`-^QT!oKglx~pY
zAv!7<PYmOz<Z^P8k3X}^Uhf7f4sWV;Fs{SA?cHyEHw&HhH$7Gk$C1@lzVB_{&ynlf
zTpu5BY@x?H&pb(A%rWpI;y5FTbDWR}Z;0cTEcMhCOeUU#gE5hEq*+c(c8gpod{~F0
zTV__b;UyZuM>xumF6oT&r)M&Vw}jdKOr+Aqm?p;538u-YCW`TaW-_u^Euoh9T7VHq
z8RH!^Uht!Dzk=Y+ua`TZus}&qWE8a*XdK$Nv{lV;b3~#F&fR~C$?$;9YX`h?dC2Zw
zO)rjk_?{)!SLQHv%XB;@OBIX#d5Yu5=`744;y4UxoX6RKj?mW7v~}pmi7?1XQ$?07
z;urcf)dGXTkb}vHc2bh233*;{W6*MCV0eAv7#U5P8jL8g)}U-a;aKHS(y%yZ@#Y%s
zu;t*&CULaPu1`7f&?)Y^>nIOA)WOM^>(?%E`qWWgdF~pMoiWe+#&gU|Phq#{3|>YL
zpW*)FaP;^x)*9-%W;`B)_r$Shu9vaAut>ktCnlk?0^=pt45;k@j3jmnho*5FHEr0w
ze1lh>eUV(liPJ$OP#cZ)9-~{(P?tf*H1dKd6T~uNcQl}0T127Pc=a-Sb;F`en0mp~
zr^L=-2)J}F)NF(Db+ADW7fJtRpYx?6i{DHfzUOfil1CIpVNNE4a&sV@d7P6$WOK&8
z_1=RF%-PAr5gkm)vYc+W;PS?Ke&r*-#9#QSzl^XkXu;a@8c#g_*taZ!d@~Wn5sQoS
zoV)KHKJbAL^8MfYeVjabiq(}jU#Py-5ma9TT-14ovlizX8r{;=HC0uioy9ndkQwXC
zp}i7EF>g8hR@9(j@nDe+Zor<#M_tTnM%3}7afe)XiK8)yc1Hf-a9kQxe-eU($k=1G
zrKP5vSSoFK{mMo5h8p4p+NMQT5+yUdo%x0+z&9iv#mZcPtw+4&p?f)Z*J-vk_qlj+
z8#~ygur+xc;ibS<Bj)FNOrtqoJ%5Qx)MP6;Pd@f;zUv3><MX3!((VnwU>Yd*;J)K4
zeE;e1A%Bzh6@VgUHC^Y}%6+UYY4-e-W2#3%4x}K0X<BZ(Xs7TrC~Z-ZAqobr>>pU-
zc*0^VQG^63g`7d}4~0$-rxf9HgXA-;vB@$yL2Iv>0eH1<gZQi~FuG4Uh?(3h*uG*p
zxEfRLCRF<#Uq|%2%cx0&p9u1_OI2Eowk)n4=Ph@A6d^9~W|hqyD-a<;I)Qf*ZA1BF
zv|;1&b+Y9SM^DW09q)LIjlGhWuWd0I++by4j(hGpP9kdJB4u&yOlU6%g@_}3BxzkR
zs`7r;>sJDm#(2}>+9nK3g%3NGLnJ~Lvvg#YZadF-G-5iPlIAgImX7lD=2bSeBUKV)
zjTIfNvt&d_8(}fv0^D__>nX%OH+F}J(Gpd;$E6!r7`G9vcSKo6o~3Mm>>8upDJSEE
zoL&6rRrL4^EEK!kb<Zk`b90#1(CC`+bV3rPK{Zq<7Uq_i>(3F%7^@A=wpiO@T8r0;
zRCb7*plUrvOD?{6nHN6(Mdtev58ii<&gxO5kW8u?Aw@u^)u5{i?>sS(X99#|r>-zZ
z*4WuPVE5`40!b^tWht@DX~uh4!7X>OH!H*6^c)ChZGZ%d8H$)U>9l?W?N=$nJR?n0
zk~nxENFf7+QA(^0##CB|y_Jv>WArVlc<_Ox>=mP7NtO+XB8gOzkN)m&@t*H^FAv@K
zFh(0XdBJ0kJ<8Aj;xFEkod3fPsRV0B*LcS}-^QQ$(|?*rAAX!8>qqGnZ@xf%tHN3n
zW^~5Twl%v4+iY)dvA4I+c(PA98BkSI+P0yrYMQFSdWW@vdH#Lh_Z&a;elV@!rH{VC
z&8;P}WR>IbJX0}bdVa*TyN6uT*yV~=S;SBxG~B}49AZUh=ZXU+mBE;X)Cm%s;%m5a
z{U)t$NHT-AhN^54LZRE1B#DA-5G`rZXJu`PqLT+Mnv3Z4daSNl-tn$SSzhUpclt;n
z8SL-z;3E&SJDKti|Jgs|^5zXzmODIt-&;Ap+T#<GCV1aKa6%ozBX7B%lXLgp@wiGP
zDRX!-Ix-_+j$l6Rfd~skNkn8K+D3wu2qlpbs065>X<%A9x>^(GF2J0n2-kx=#6o%O
ztPWvz2)iiyNMal%nwmBsshmWV8SQk5$!L{>8=lucvroO5AP)+b(^Yy=&dF2DeDC+X
zo9RJAIo7PLukg#i^ihoQeDIHbAMbt3bw2X!o4@uuCJ=?d$=e>pN(4or!R`SA4)8~N
zBx%I_T*~@TFzz(Wr4z6fi!1X;A}XV4jHYaBA}x_o=-&C+L|90Pj3bngbdm^~HW<p!
zWYDb(Eu5xpLqAYTdP@uRmlsJQfmI1B^BHnDB~lsEJDh0{2-0H>^G^<`O@&kf;Vdzp
zc_UeM5eL(KcE>L>7^md@6R0@Gc!BT|B_hnpDbvXouH7R)J?6-2#L|3<iYq)V)>+(m
z#7YzqB{6Z|(km8-<`z*Z3eQ_tG<Ki5wp3Hi$hJ5iVdILrPB8|qUc5q@NFIOUKKgN&
zMn+gKL&@$ft!YV&z*Ymmkw{6D3WiP1FwGcw%hipW)Kx>Q5_B9BcXP^v03~kxtmg3k
z>uz8P51lNhC<>ak0T0v4AVq`t7L*`LN%A}=&kM3FLn#>s!!wN}tu<{MR-me`XpIID
zWKIs}B-Xk>YPFuxu%sw*I-Q)I-7S9O!yn?V(`U(3fl3v3-+edTZkO?B@=ZVYuP5Rp
z;?(JreBg(FkU#Y&e~gD8c$|K3?ho`a>!zZrN}RLQx?*>Km;Jqc_V@PL-Q8jT;DB;E
zrD+<BnbNc+P18`7RrsE6sp=Z%4BjU^aQ_AnHj%G<;T5i3T_G)EWTx<iK_{L{OxV{q
z8R&h+3WhRh+2P=p|EvT85+4M#Mx*r@+ZyJRb=<VWrPueF3|o3)%3B}1mvTDc=EhB`
z5>B5!#mSRvbUG;~Ppt6xV-Ij-eU;4{TfBZ{hdj?2j3#{dd)~^Q{Gsp0h`@kw-ZL?R
zU-;1P;$%ve&at+z%=-Eoqw*%TD-mxPN-GC%Ir9i>^T)o*k@i3jK}<?Ka};m5-82?q
z?UeHNj893TSP;janf5Dn4Rx(4IyN9IlbA@wx8y|3m((B3?z=#<K&an#(X=$JqTO4<
z4AvQ5UF7O3B?s3_I`wfLIdYD>9y-a1<>Q<?vxf4XqsLa*JE#~9hxGa}D+@VwZCPI`
zg8S?p;!P9CA>Y$^KjSp|*~<`U?bz8HVT=pze>BX+p3`d?QS7;K<r33DpT(6$7FX7Z
zvw51TrfHfWX%fYlC<x*xbqFg1nIsS;cx!LDAqt>vTlRLh(OS<GI|(LLc<Q?0=wi(J
zj-+h`!Yb-kqxWLe+jm+0FNZ)Nq6uzGdP@PvAdvE`!?~BZWmG`uyL<0$`N2yy)9IN(
zNv#WSvXF=BF%E<v5q(UaGHqIhqcOwbkb`ncSxR*28JH3&G-pbUFQ(X5lc<Wj?>hma
z=Cw;V7<WcI*l3*8IB#g0hR8<fa&(&n=QLYo$#`ys?SmoPJ9`)eIFP*_GD#@)9(Y5A
z$IBox_w^JBLI@%u`I$feGko9!ALN;*pXRBjpXS9E&y%L-5whkGQu+17;UHWHK@tbc
z-J<Bw?REpAO2r5SNt{sC71PNSse)IGnT|-}gtl#Oy_enx*#(56X$;dzMI1-q9iRQ|
zC-|Ws`~df!dl2s(r%#_|eqsI(ErAFjxbOaR{P2(dFdzKj`+3`w?<7sKKaj^Vre<^V
zCWC_ms(Q*~I%H{ig`$|l7)vB#R+raUU0$a?Sy9zBbyd+c8r@b*t1*MYfSWh3b8~Bh
z;lTk~PiS?`a5UiHU>^W$ESD}^<K-83>2}g!MO-MNo<wC1sT0&d(Tp^;0^bMb2uY($
zn4BNBv=lh)Y3mYQ306ACC`U0jE;Y21h|X%vPygu;a&m2vYgeu^nU)+sb&7NM+(nv1
z3??J4UAx3|JfhQ`rz}02Tl<X1!$2#DCE0A{8diqSe(sBW=vP0&!PbD-E0*WxSy^A?
zeEU4(c7ImcfKafqmh!;L!xUov4o7|FS4{#lmUp-&BHmbT2@!a2Qi)ssN>U1h41#~&
z1(Ishn4m-<LiM4fnn|mNF@l|$Lcn1!A)+Bd`-&PzRSMeu4&}9DjIOV-_pHTkWjwTU
zh9{nQoW~xxkH;TaXYZzFe}9ix&TsS7Q&%_`4%xV|!E{=&w?DvI&%gSgK0^N|Ucck4
z@2f&O+)QHNaN+hIDk8QHrc5f!LMn))WWK*ZA~eReL~4%Fc*Mo?ud%(e&zZZ=(O+65
zwqd}gguEp;;k-v1Gus`{kY8~aFe}2rHr`QA!ysNsHQSZ8xS5o5Pf8wLNw~P7F|D9%
zJ+J-ql*>QY;rPj8^yZf+Ck^9iP;`}2gEuqGq(J&W&ufLpI)Ri?(B=^yDIG!v$pCFN
zt=34PkV=tdIZlLxIw?z9V^AVzJlbP>XUN{cgmS8>n>lo&khZ}l5leFmC^JWEOH5ml
zN=Y@UxW0LX{Tmev53Yg;k|s)ej5R13<E@!JWl3W-6^el<m~7l+GAyY@@Khm+Q97_x
zEgtiR_w#FQV1*N0a`(A=`OAOxukqj`5Awbbyr0o%#Is*~mAlXVO`dxCh2ZZ_^9^}z
z#BoBe*Q3+vFxQ`Fetw=LNrHx&kT~n;v@LUe&1f{j7>jj=RtGRa+qQutBLv!75*6GV
zC*ztpQ4~eO_RdW{^SMuP*V%h8CSiU32rJ7=T)XlIOyB^PmKXTo2Y-;C_=%t7J@5H$
zI>jH%K>XlfhbvbubK%0vT)udL-Q69uX*quEC_ny_e}N-MPJd1JV@%6vJZ5|QCR^KE
zZ0~IF{EJWV%4;uk@sq#D{vae>=R7xV9B}dC0ck2pqL?^|QHeyQ(CrGOQ$(dkI<Nzv
z0CHfFFibxr2j<{fXofA$WX#X4v3I#<v~NkHF8xl2`_3NWo}+6#de2D!YAo?gU-`f`
zl+!WBv<RQ#T}&hmXHOr+*qTbGSQ{~#wtW24&+&`D^kJU);%i7FiG*Nnd6{e}VP|KP
z$_{aU2BVGO^ocdr7tg^Rcs60`pl4OW80aHz!KwpyMInREpOiE6r$dc5Ztdp04{NP4
zKFGU!|0R5lLz>^=B;<B9H0+{9LtP05H*)Ou5(lppY`?0J*XLMrM|tFdhk5U#PcRuw
zx$yiAUU}sbpMUCAF5lRsGzP6D*l3o#B-&UuHZO4L?VI!iTlXuz{yWb=M4-&uSp!@k
z@IufKF&x$G9gOL%bV;Iwq??nblByaZeV2ZJ5$lK4bzqt|I`rnOF_@;snZSNLMBOMA
zK8i@)(*AeWQP(w1tI3lz$gDYs^d4t4b?Y(O6Eo#_DrvG9+j@qTCmGbpv7;0B$kaTF
zg2dF|ELCHIu$Y%f5fcfI_dz{aDTTMe#N2z0^#V_XHbKuT%@T@kpSpHT_FG2NiqWWI
zG$|QZp0e4etPFJ{@Gb+NlSt@v4GUe*%8@?DmX}%W)m&{2wuZ?-vv+<6BQ$e!D?IY#
z+X7k6S(K20B&1~UP4OV3m`ZEvw8Q?S#q5l+jiXkQB#O{=LphxW8(_c*!IwxIe+?mo
zpvZGR@PQ9-_T1fA=W*WCUzp=P?|mC^no}pobc$WZ)B0<>(^+^QxIW!ZhhDEwzu#wm
zZk}GhN0wzoDk6=u8C=&e8jWe1maJ`YSjLkvT5IAse&aJn3Pr0mP3x)anj`_d;xnK5
zG#~uIA7(j@>2<rTt*rxhXzYHIvAnX(cYXJF^5=i%&++bey@&a^rEl4N9~!F-hI_p9
z(z9H;_&Vp$zs~N?HiN+d&YIb|9EgWPeC;H|C`wpdSYmNuiTmyix+U-X{tt5X+GYOR
z|HJ<lp+3Z|YIF9XX=b(n)8-BL14;>q!Km5q9<suB38eES;0Uk+bqP1(3F+cIb8D+S
zd*K4s*W{guvaHzH*x}^LQrN*uf%1aNXeN^>wl#FKE_K_Y+aXOgrBmd1w}>$X=P%vl
zw|@I`JoU`$Jomyi_J<`##3Yd+(T<aAYgpBCd3*^i+FMI>9KpHMXIWdiQ~S!}DY1CA
zt4$iW$?^Pc#PYDj0uUsm2&Xu-1^2Vro%g|J;&5Z`T_6{RF30TW-XnyY)j5ts<F)YU
zj>_#)+5_sTVE@V@V(&hhYiq<CJ>I%{jQh^rMW$mg4VxD)aO1`%H#WC;`TP~O4@yd#
zVr51o@_^$KVOlCMcpqc!H`u~DJU|u+P9dB?Q)4|i8`DfR+q(y>&Pj?W!C*lt(zJ`w
z9xn|2g*E1c#8}UGXFs&VOjw*%9Ft~Q@VHP4p~7IvMItPc9-P%UZSc;McRM5|#psqK
zjYzYIvTUf!2G<CbYgp~T<~U+FR?MfKUaXiN98fPTkn|Tx=Q4!1$W~I<4Ncu(T|?s=
ztkX!B5|QGZMyfD}5eQ@~NV_Rnr;EX|wLf6v=01Z#!)WB0mV!pB8Nvc0;Yp+=Q8nFM
zvAW>suXR~ooTHlxgmD<vV5G$j4rt1XskPjFa*2EII>XZ4C)sFA>LR9)89m`>V?*u2
z&f7~(ZF=nY3+l~1s_}H@r{c-yJ6LaN(WFQk9ETD@-XH33n&0<M%Uj|GA<@KAa$@-?
z?|bj}k!LB!y09qt*(oeMTGz-owy(a{<OD=gQKTvTPLGwv75e><Nb({lj+0;upcG|U
zVT{I@ei#CqhBQuTWsq>tGyira6~?wm6;NyAu_cNec^dKhg_n8l{Hx3_Es<t9N7jQJ
zMQEjc9Z&pgh?6Ie^Vk2@U*$)C_>XhXJ?FSB`S;C7+csRe@*2-Q_cYHx|3$7}yU1vC
zFq8Y9iE~SVF_yBLvb!@M6s^CBnCmU_@B@qdxj+9G8SY*HUizw9V#b7>G=T_yI|Q&#
zNFuwyFnlySaB;82oEY(3@+!~0vPEqVDAWR@spMb&+;4E@@-`=quXE(c3P;x$8BHtp
z_lHC(qN)tjX~VH2i)`x&<MDunrBz<Lu)(kW`p5a$$DiWr^*yvrNs^qX1Ku^{Ny72f
zRZLT}F}jA1>@9Fb94St$onpRNyu&MXv~Gl#5gYrL7`V%<AAOv#t+@?YiAfPI5CNqM
z-3w_Bnf0DSn;!4M+rSDJLIysFkY7@V#6-liW!rmbOv`i<ad1uGcaGB;J;YjcmWSVd
zj>k?e^XO8B(2llkXk$fVj&b429{N`$j(tXT%y8<71mJ=#cp3|AVkt{a9$D@rUvb9-
zV!gv0mR8Qw`j9!LP~bHCgMIb~pyn13GU(ICQGzuMB9cMdrK*`uOSCaaCCQ3{qT3-(
zQ)CpO;=|;jg1i7e$eB2Zlp;$rA|Y{Eif$W>Zm>pUnwGp{Da(>(qNqzl-$oov9acNy
zIAvvd9^o4X2Ll$f9wJF_D2jfcqUUMq25mID(LwmpS-dsKC?SqO#0nL~2$eCZ;Pq<<
zT)no%?p{k--Ns{k?-A0GsfILyH1_mz$>Ku6%6ymkUV`do2;ni+lxkGLs3z+zGa7GT
zs}V^@vT|~rWO)u*<e{x6!o)F+Qzlm9Oeke*V`yZDQV7Pw0adHffKm~8k<(Nq+8Lyj
z2p54fvkUofg87vqoTh>xU6H3L_uYFhk3RM&07`!OPQ(XvnES>A;<Gd*QlZBi$1z1w
z(Cv2VcDtl$O53(G6qXHbtt_P|3XBPo8-lP<dLeE>!KDhjG9f%|Yne<N;z%+ck9p?l
zr+MOuCs8_LeRb{D`}VDhEX#P}iAVW6fA{b3!4Lc~vNZn|-Lq$Zf0wIQ&-2+&{|?W7
z@eAzj?=Tq;F}e)XHy_AnLPkgxQ&lZHJNqPYPV+5dI{v_Szn`Xk3i#iBeSi3s#$#h(
zN=P7kzyR72T-@B}-(L84WWzq`d>5t(+RpRQ-+h@+e)46ydCa}%PV&SP4|3PpW6bqC
zbUHagX3WhOWSx0hvq5VW&%d<Ahd=UhKJ}@msauD45^pCsTZOi>t?2YuIB{y3q1oa<
zAJA~yb~%ny){0Yf=)dVpv)(e*Ba&2dWAp+Ow}(ACC6re)^++HhN4XShg3(-PSGn-{
z?D)()ux4#6dkek}fyWW1D=|J8TZHcmK~<-0-sofZj?fuAO0sj32aY|&UEO(_y(u4l
z=^{<zSX`LniAV3^)V$+JC+FO;qYNgF%a=B|a&?DmyF<1I2N+ZG@WXerw$f*Jd!I)h
zew|l81K$vgS36JZJz5A_@6hfL2<n5{E{?6;ZA{y8;@By&B%!TpTv>vbv|6KQmahHz
zK53^Q%F<i5vqA=?-mvQpgQoa!pf1CIDvAP5-Dq0lXd8ny2Hk3`4OWnGo?}fz*T<~(
zJ&kEe(wI0&Sy@_PdvBNNXvo~k8eWE0ln{cXlaYvm$S9QcXi?*&0TE#u$*8H=+}>w%
z^ClZN#~d7lUZL<Av694*CW>mZM3W?*egboylukFN*XvMpa<U{Ljv{K~866yOFxaNt
zx7cepS$N_M6X`j!QsB#qZYM?L9mL#1I6htwrNJ>QE^Cwx=%i5%rs&huj{V`7+E}~@
zc9eCiX{!n;6<$U-V-P634|cEqD_+se{3HcZdJ?7Rb~?QE9dBoOZS_kQ+S}}K!EiL9
zX}|siA`ql$8qiYRE_t5Ml0-&P6eLMPQ53gkT}{(amSvcEXdSpBNsRFh>4LF@bsFzO
zMIn^HXirsY@+2Yhf(sYVQ<hWWEM<9l<?DXVcZz<$$DjY1pW?6nSAUKB?t62+q^~bL
z8#k|U@%+pD)^C1@7hiamy}eDmYw<q7UL7VtVSU&bXah}a84hc<w<{*4r7FKQV1Z5&
z(@Adc#=q&@>j(*C3=Dt@+?-?c#tou$izEG$EG`_Q9jEMU!1m1{c+J-Cm`{K1S&ko>
z=b?x0<>ZND%+L2YdgLfowZ|7<yv9o}UF6mC8|>~)$a~AI&-bx<%FgCRR@dg}A6;hW
z<}Pd9d6rN0_}%gEGWFH1u_!>AWvq0MAnq{#5Hmo~Msu0pyzv{H>L112LSMb{y2wx_
zp{+DV1%oQ14JHUo;4OzD+c@jPtk2rn)fxm1B@sC(j`&s!%Q?^PHjMYq($`P0_v#!6
z7lsT!`Dwn8G%PL&&fdLFzu#wJxyQ!lh)-X<K`$$K=-!i@Tym@>i#&8ZXBx$n!g4U&
z=dnl5GT%#ShEuZSBfR?QH-8)Nm_UpO(6PojYHQK6sgD$pq%ld6(YPVIgFX6l3nWX+
z=vL1#&QVw-k_;gwQI_D8z(|2m0-;oZLd_s<ayF!wJ{%a+woE1y4)*scCndHuC?QcQ
zprVo}g(#qll+Q?2P1^{hux#y=EG;htBASeuUtDH59AT;o8OL}jVAe82f)f@WdyG)*
zj|>~v_t?BPVt0SSXc`tSiw}^~BBqxN=*E^bR`fb4{q8(Po={{NSsJ5MOq3>pHR?Q*
zx?+E4z`^z|Q&$qz3Ns#(^<!jTF@J1<a&tl^B65-9J3V~e5Y32k)_WomAyk3$F!B;T
zx4>w8z`<yWa0IO%6$p5?ZNYh_re$GnjqiBRck<i6`x}fW;~=5^6?4ceaU)2hn5D%<
z-u8~SBctE1Pjy(F*Hwl7Dm~J#_((GFKC&z)%QEsjr~74z<c&xzAp}X1kmotm=_ve;
zb4;d_0K{-^#!f#plL`aMw)Rw|rJKcUZEZ3fjp)wJ)9=r{K_cbw6@C4oH7u_j=P&)`
zpWwgxD}R-fC(eFD_w(hkxp|f6o_mVVeeP4d^70F8-Mog@6>VGLY=bcx=RH;^w6#>V
zp{_h_ZE0GIZb7%OyF23NfAL>&;nI1MBqfUDAj%gj(j=qPEtu=iF+V>~ue->d^t29H
z(2BriSzMU|F%GNWHxrIfg=nA;%!BC~wz@ayoml16sdYLX$t&ZQ>A0k<EK)iyT&vkW
z7}6^`^ooL|`32UGFZ0sH7r42(P1}0zf9PJ$o<7U+@)C79;;E-Jr%s(<{n%Z6Zv1(U
zo>(9+;M&bsn7E39+oR$l&zSRfhFJ>)Qbk-GU*@^N%gik-;=~~;>y77BgsLLisirL)
z#(IqPK_t_K+hGgKI?yKY<ihGGIrIz+1j9pN$*`H5?1L)TX}3EZT+)c^E!IQ#WXe5{
zJisFlAElQj6x}YHo0>uG8BT|+mzvF?V`sBuJlN&LU0v2r^oeAG*A3%w&Bo?+jxQYC
z@ff_hp%kEfxS3fi%L)<tauT9QV!c6oOQa;b`}_1dU5pF9#7cmS6;j0rB?&(90m>Ff
z(X3s?%#t!BjH)b|PNvj#75v1tCNFXp7MI9H(AXLe_Nk{eR?nE0UK7O<B?Zny6gxUa
zP#9fYSfOfK@;oJ0igr99@Ak3Y;gvupF(Oh-on<i5ymonq^A~p+?@f`~kcgN#l4L1l
zg`nG4<e8_J2#P48)9KSGdPGq|q++~`5Gt5tS?6e)mbz)!A8%9cO^AhHezDJz>d{e-
z*c;NNRmzo<^iQ3`#tHP7XtEqvwV1lab{y6@oR^dcw1qm(F#S0WE?lLo4G9rWD2#*2
zXpAz3qLWg!@WkV9<1ha7U*xl&{TN!;G|ij0G!7F(gy7W4liYLfx!V_uxDA1}7Gz{F
z*3OocuRX&2BjDkqI1V492&F<RE6XynEW^yCZ2~tW{H)U{7#u`2O%n{GL}+or41Ilz
zp>}(R)-;w_Xm)pZ7!40tKYEN_e~vUwsq3$A`2Xec@Wc1<5C7rc=SP3^ePn6zP2HEZ
zhMnCFe)o4i!WVw;Q#}9NGYp3Z7}Mgc!C8$qIwTL{C|j5`lD2lV+TlWh`6W@7hM)b}
z|0n(aKV^QQpwlUWjY1rer#V@k1q9sOJWGqKEUv7xxUk5HQ>R#8UuSLoDD!jk2n6HO
z5{W%l=L3uW%sr?1`0u_PJT|@+F`C^bz$&!5;i7w%C^0;6|2s&{DwkisNt8w`Ew6C=
z)Jf`TMK3GZ+1lo_pZqKjJ#rtXP9G<e5mjCC_Q&q#*wG_&y9HI9p?HXHw?~o}EFbT2
z|2t08MmucRS16g1%?wZhx><+i=+5AWLm~7VVv~coWK4_kGi>r3&O?D9bt&znp%OL5
z+F+t+!|f1(<(X9^#)Tiu{yKP<$4C^Vn>~gKzdPMaFgtVXzg)t&AQzT%cOBsaf9&nd
zucRz4B`o%PY;V`}=H^*gp7Qz6e2laxdGhfGc<|f_Vt<q(lQg1bb*@7s6G}ZHmY#k$
zzvK4m&NC1~1)^OTySiENbf`~cw8op7yc<*1Q%2)4d6HtZLs*MP;FQD*k5nOfBtn95
zSP?SqU^rno2&%kzPtnO)TwG#xWi_;M1Pu2EoPX^+yW2a6S;9(UMHB^{9o=}G1fe8R
zma=wy86iPiixeKJ3cRB(OQI-2+ZKzXah~C{;?mZH*RN06**8d;(2XSX^PZ)`lBY1g
z(4*f=aK1!f=;mt_*&Hg4XMIb7ixghES$nObnoiM8L*3L&>Jj};K|k%#6qdfqiM*lN
zFX=9v;cY+pU7URM0ovV7wqJdP9p@=06QmII7Z+yyE5XzYjE6QWuxW;A9o9=KALOay
zNTIDkcu&(9S_|7dJN%n}^KTiCMszv_tu|qs^ZR0!K0M$A_uYFxD{HI2|Dq9sa#|zh
z2zzL5@HNIE6#US&66b@J@l3u%3JD_c44S4vN=dKRV=|dAnM?vVXm-<WAaWj_BT5O{
zR?jf@67ND4qFalN9pmwc-Mt;|cAidNkmva~oj{&^@-4jWZ9hUBfBjY18)Gs#;1i$t
zD4+V&@9^{&KFj|8Hf_^jOpDg-%of?AwWF#n(`j(_GtPb0;z>7-y`3os2UGGQDEz9!
zL9P&iLnFh$RwcndyVLD4*Y9)U_zCX0`(D<M9HHne^VmbLvif9re{Nj95d=TK)iLB&
z2dFM9UM`=*p7yMz_i*mq3f2jZ967?$!aTc96GxUvl`Ji%EG?znf7dCbE~#oo5j)bT
zWqz*5(o&D3M;`_yaGjVF@BR+DbB0e{`%U(nn@DK`V}7=7cat7IUj8aiR$$N^=>6Gy
z6D+?`hqK1R1Oi!-fd}jKY_k()S8uFW4vS&$0&ndQyc;-plGbY`w#JItrU^+A9j6^E
z(hL+mr#N?}&l8Ux;qix$aASLfuQatEU>c8-o>S`uE8q7RXV!Y0T`%Yr0x=u=Pe&yi
zn;S@}D7p#GgqHE0&h9(6u)Lx{(qM4Hp=M7!!V@`2MIhEq%Mu@3_D6dxFRvlv6iI~8
z0x)4=wI1hcs!CH1D|Uw!2g4yA%fj*^E9=MT^gBd}B$5)>w3LGpJ2y7jxPBAoBJO$c
zUix!g%4&k}nnbGLu`WC$F*1%2k-}LIMq=6)*VZ(;!qGBn8+0Ax&6sIfaeb%e^&1s8
zCn?&;6w0%h)*M@jSzYQ;<S9v#(d*?@b&!Sb_UA~W9x}?1Dh{wh6=R)cZ+C;XscD)z
z^e|^AuBEwUdPNtfTQZdp_ZLXC!47s|X@QfEyoFX;(zT-uE?wj5m5U6vuF&iCShmNo
zUJ?mUOBD3zmgcD^L!=Igu=I*04^>+-!lHzyFdpp<UwrYieDTH4;<1!%_2z?P>1X6g
zNgSn|J9i&hkp&o`pW#4!V1MuK?hwWM7zbaPKn^u~zVc%PGL<Awg8aHvpduM|Q^Mhu
z$2c?Ny(rQorPVFY&E!ms!OH-+aNbiC1+uCU0_wJzCA2_n^ga+@l~i~EgYl5<?JcBt
z*tVgQX0x^G8_-{wMVe-Oz0xpmjOldDi!VIQumAcl^2{@zV}EyxsvL*bpfMQh(ArVg
z&>E;JKZALG{g34gl;cpT2`mT~7P8<Qga>3)VJ!Ie06aXSUL;L_ha^deqKyC5->G@W
zlL218y}8S`61{dSJYg<kfuo>(%d1zPV`F-Q6N_hf_q)EAEXnAkkaha>7yF!kWS+Em
zjC=39ho$8e9{=8RARR@KgKVkHh^n13D#r{*Lv9>wuwkz<bz6)_=NT^Sk)&Y9x4m9^
zVxC-lHC-~p&<q(=`?SW>XpN=ijR>p|f=*T-WJn-!l+0$o?$*^}{TpuXb}JSixN#20
zSWg&EDomjv0x1<8af))YN8S`1?MJ-lou_%=+&o35iG}2~SFaL_5ywxQ;lxUh<MTO5
zv_PV6`^z83a8&V+|H=QC{exZp_TTxhIeq4S=F`=0yaMrFV4cHy7tCd5#6AT?LXboe
zGRmpyhO*VT;Q?8Z(@VPu?<hxm$Vg%|lq19bV9M2<T@I!rR{8~Z-*uX!$B&Q{X;?_`
z*rsMM7&6=+u(!QUJ)W|<vdYP`chgzuBNB;m%h<M^HBSW2I}Z24##yXzs92FEF}e|$
z+EJO7YBZvq7$!z=X{X}Ku3=bpKzQbghQ&g2d^O|5kwuE4Ll!5@_vaZ;hK$Aoq>OPc
zaATyJ+4MG+Za2m|$M9euya_W9sfd!CB+1CKoHSP0NRgyDY0(St%JBhhZ;F?mEYA_9
zMI;?=T)6_(9;pQBJV;3rhZY$g@_wJ2*9V{tirG$71Q@RsVKuN4gctZX4D;?NV)BL`
zrFq8P=kA`J12f#Y;MVZ&=FRKO%?;mVkb37uCBPHI{isOIW-{&;=R<2vQ50ao81w&Y
z?@fa&JJ0*R-}|2ZE`59No&~c37%UuY1PFp4A%eCnIZD}<<4ROzTXrQ@l1f}jCFOic
zIbWQr<WrK5iCn6vY)f`UvMoy@C5jR^fW!`B$IJkjeWqu6z56}uyX3=ryJr9rAkhMC
z*gU_gnV#w8+|&1-d!F}wmj8e6PZUNZNdm%2y$3-+p67U;H}yw%vJp3@)1CK;Qi?n`
z6w+}0#=6rMD+<b@*wYu<Kl`nnWSl?$E?@n_-{GYfpXK`X%cPS5WuB221x0QtN*6RL
z<s>f5CvdOcZ4>>lbJripZW?XL(wr>Kfg#)Lw;t=K5Ik-@wFz!sqy@TDF{w6r@7j<f
zTL*dI!O!#H{r51lFheOa`g((j?sG}M#ar7MgKm$F%?(D`fNp<_jjbE(47M3&6N~^8
z7=i;Ze*|tn7!uBT1chKSvg~#}hcpTKaXXNTff+NZ##mt~t)jG31Nf=8-!3yrXKGiW
z2!uafG)@0V_kOLMB@)KE;bEmUf$yxGOI<R_Cun7;D&?m7t)S|+sn#QsNy4LFI?0n?
zzLPjg`O`mpnX~WgaO_ZrPJ6=AoZz0-XzzRsFewZ>J0tpo5#y}j^p$IDZh<i~T)md!
zM<ug|njiTLd_(}*?NOfUU`>tJT@Ks1;u#Ea6d|QgURpOHoo9qt3T+t;Q^sA*m442p
zVL_JX9BM?|cXXAbcOGD2X#pt(*pjNyOok)I8#`=oZlScI*=lk8Ge?<USb)Go3)dk<
z5Fmu>@-oV}HpZ#4uM^mFx<`H(;CX_~8a!<X;s)o>Z*x75S<ig5jY&eotUuw%GR)0+
z%yb|K6at?(ZljH%QksRORT_=RsqYpAgV7KvB%RJIUf{dbk#w!CNzxz+n*?Ek?}dcE
zhbJw<51ba*ZY^Or;OL_()2<;y>?f2~FzoIivmVtX<Gx2gc)l~p5`rKUG?JLnU<Xww
z5(EP*N`kSHa5`wU8&p3Q)_D&2VMw#taz<8r#@}M<Cth*(%o$GIH~A3~2S5H1#WWBZ
zV^EcHiUP(^mL*x{<o>O(d-V;a6k!<Rc^+j^?iJS4G~KI(SZmz?==5v5H#g6dSOJx?
zM3Kj}>o<7w&9}I=cAY1lc!I5+9bS9mEsW`Z+-GXv^!qzJ`|P*)_P76(x8M2!lktFb
zGGsCtQxuAHV#x~^Jj&`5=yv)95g%S-|LHIQu7XS?t-~>?D0%6P=cq!%nTw~XWXbyY
zGTYo>LZ6`?Q&|Igx_S@6`WCCLn}VMOV(>#~tw7@#_#KJ$EoubW*zMFDk{>rFXQ)WU
z1m!D?u#`rl-`CcD^9|CH4WzM@%Hnw?%G_M^BHVMUnFbJpyY{t4qfsbQol#X4)_Qxb
z^VUMvGk6n=%qjw*8K)W6vz$G5nf3LGU--(S6vbtB`YC7SHiQ1q-J?1SlqTlcXW!xZ
z7vEy1J7%1Au|{&};32;D$3GxfXZf{XIlF%;6M;Zmw{G>N>q3@5;JJ-gW!#%L3Sy)u
z2z@8F-J9f$GsC4Dea>9X*%%iH0S}y5;)w^2bGQ>yNspq;i6htgyS{#%oy~0`<FmN9
z#O(YWjaC!kNlK%q)7TzHIPVZus=aO_i$#}ir5boXVH{4oa-1Z%@qCOI5Y8`9ZC9*q
z?NHd5#inAVm9yFmn44)3HVttc5c(0Vq=hE~#`zc{Jd#G+)mWq@u#&tO5yTNe6ivIj
z1aZ<JOD9O_lO#=|Fv0Uf1Qu%)Qd+c-7~yW9<#6cQYJCq|8t?>pQ8MUm)4lo*QKQXY
z0<4fAoFs`KILpq_U<7$ZghwDWxtQ)gw8i(NYq|b7+E@1GhmD;r{?Y&Xk9h2hk8{sG
z_j2^;5mpbZFgxGDXvOQVzs|8^qaTsH`S3?_7vVIgYj<}X6?svRWf?&bkR%COYs#ub
z;<~?f1IuVUqAbh3fa7^ivA`IY!`}@gyMJ6D5h$baBv?G2dG<NZpS!}jbLYAH-g`ND
z@+AGin2pWXe(E=BYYpcwyvx_V`g^?i!qaSRZIGuu#-ky_(TFrv&Kn|Sni{OZ0$nig
z8FCcD6Krf;=bL}<N9d8lI~EWg6=aRv1&0~jw0i^u(?3%?Ue5zRa5bRBKywk26(AI-
zv7s1R^n{PKsZ<nf%1umc@j{=x%qh)o3hP!2UYKJBp(K-Hcg|z_ZrI(ZOzR<~af6sb
z2v?)RQre2LDyNIr3gu4&h_O_|94{Lo^OQGUJ<akV$-|G`Ms8!25#*(0X7(V@zI>Xq
z@BA5gmhs^I_jB^j29rt3_g}im>9Yg0l6bN~s~My83~LuQxq4%p$M3&RXTMS=*1GhS
z6;AIe6w^U7YcQrlo4{#9NlD^IgkC^VXf_8GXRnWVZzpG?fYmVM(Zwc@9b4hhVgnU}
zZ(@AmlV=4xTifVDbK9Y#%r49m#gW^_Rn8hdZZ-(w239(~o?<dVXBow4;xzLEx2g}q
zfF$t|z7sjDlqSn_(!8LkG&b}(e0+uV-h_THWqGE-;pLF_YzGm~xq-_dAWUM6@1b$p
zQ;kLwYn@P^wH7}JIB@VFRi&6@DM1h+gdmRE7~MtS(`>d}?$R2RaRq)nfffSgNt7{2
z4GMt>0{lv0%ZhfZLASff*475CSxsIR1W^+qz3Ke8U^tqPPf8+X@vS5jkQ#**&LULW
zsaelY{$6+=lRV?wPkooSPruFL;sUdCvn(vkv#>Bnr!&L%pZ+et@CzTm0PTW?#uztH
zrIez~OVTW(*=#!1*gVH*gD?GQTdu2rXss!yLvccgy+C5ISmQd?<Sn}oU~e~SjbV}&
zY;JW)Q$@Eo;CtWu9xuK8GJX&fM*FXW{3MLVBc6T!X<m5l8Qy*8&8Z`2!EoS$$8e+=
zPuO>T<3DH&*lkF&X?{(Cmd;;9j1==r9m=DIyj5abZg*k5o3#unoUeNjLg3+Q3K;|u
zh|moKmboVDgUa}r0YtN@H?Y28exZRHrR3G_*!KGm)RD6^lP1*wsNArw)=Y;$-44V|
z1r;l0_qyAJ1!XHnlQGr^w7KPLLJ$avuLg+RvVLRCxhtz2J#~=YMAIMTtZz?PK754E
z;skkdn>&u)&SPJCoW;3>(`VP%=?_pylnC*pkLPPVFQ%#lLN*X0`pD<xqr0%CJ<37|
zLLmsJ9WTP(>@qXcLXVHdR*J2il(()8IkTe~*^tAY<^Gj~Q^#g#H#J3?6UrHo5+f`{
zQPF5LSzTTwN@8><D73;Tg%|q7QAFT7+4F2XVKVAdPDWJe*rla{1TPF|%(RJG$#kL}
zilSgJ96D1R<wP69P_PhN?ms-wds-29X6Sm7cXvuA-jFbC(vBlKSwSb9u$&~!M-3W{
z7~hjvuZiacNKv6xiI9dcil{1u(S{(5@nuLeX@e)pRq0IZgBamivV20OHN;J{_3^!k
zs6B_V(7S#EolcmYX;D=LVG>gqMOx;xvG{&~@O%(38o3TFPkK&vMS#Xm>r^*;%75rG
z{t4e42FQz?jm=Glg8?$|@I8q!PEc|%%v{Z3Z*TkK4{WDz6k}aSQBf3>QAC=jBw2$j
z%a}}3q;xB{LKXPFPg#~!Rpo+#08dJ)s>1W+UcMS@k<+vo@IJ2)gm42|z{0{JbDepT
zC}DQ4Lz-p0d+s8$q5VYN1#kY&hMk>Fo_gv#eE->}=x%RN7AZxUGairV_D4)6it!k#
zPpBLAKUj7xalu{W4bYWlvXirXPm8QICR)_UM56_i1=tE?0A4WNjOb}~LO>8h<b%yM
z*-(@3fS>cR0*JAiVY!8o1+AHYD$Pjs1ZQNl+o9ly#SEQf2}bXrd<ebi0HK-|&Ubeq
z3W^Hys-RLOu}xfhtSSlp7$m6O=!hT=0%BR=n+l!#6l2R^2Z~HnR3XavbY=riow|pE
z2M^(UFgM#`eQU&9r!O)bk1)a_WlSXvl_@cKiq$1jNP<A_SIWdvA<@JHLeWAI372{>
zIs_9Stn*<ILLvf5uU~O#OLKkzLj#G_+}#ejV@|U$tBK+mV?2s{gdaqNQB2ZK2!a_m
zagIeBgC`_Wqk-=`d6Y6QNCzWEy*}A!0J=g-OVDf)MlIr2lc?E120mIqIv&&C?oyTo
zVH6Yku1&2J25U<W&NsL|Y4Flu!u)}Uc=+djo;x0RfcfQRBrxu-vwr?8o9~=q`^Giy
zkOhk!A0xrq5Kp-lLyJWE9z{OE7()_!#GXepNwCcZQZ#62HzRfV#ue7LuG5-dBKDg!
zmJVQ#-%UJsh)7ANvlV$Hfgtc=jBHY9fz^gUNQCu?f&gnQlko&=jFV@V(|J`b2|a-?
z6}h?D3$ib)n^#ZEf~+hNa#xAj?$vK@{rh7a*0q0YMOhW(c}ft6<V8lBrbJOp5X5f!
z*DBJ{n9*oNQRXOJIvJa)#1q!}q)a=s+zwk#<Ar#iclvH5Z8ux^en?*AeCdl{;E6AO
zktm6I=J^-->L2|E)>iDg=YA5T>4eu_d!FZ>dzx$4t}y8LD6#>A-ZtG{pPg=>ot|PU
z)KK$LSoVU4oQL>8fZ8^cnP$9R5?qY&mm0Jp%j}Gyj4L-kRcef{kumrR$UzGUEkV?@
zgw~)lO*VjX0DkpRC-K8*N-D395Cz(^jMRj_?zsmd;nuWI;5Aul9VF-r^2p*@@GS(>
zU5Jw4NteFLOPJ(2MLD6_Xke|gaJ0hB<rUL~lC=hvPiQ3bsBMeR6DHdRnN~b{-<^DY
z?F_HHc7ert#mr2LBx%s^jkt2@I^X-=8@%)06|w@X^$<dVmtjng$Tpd6mjs;#?T$fh
ze)x0nkvkBBMPdo1Ah6cWR9LV`l+q|I@MM7J`4pAldN<?hNYU344_FQZ+L1we)~(LN
z2<b=UI;AKlAO!818Kmz|1^)!07dih8>58I7QBqclT$gyE$K2c;p0N0VlQ8k47C{uE
ztuwxwj3*5GeWd5Hw6yHx&2gy{p$dfWAqiML8t`*R@8*~PyZ?p7llSi}7AzKFHGKIi
zq}_F{yzRYJP#j$wEer`R!JXg^!JWY+=nynWuwX;b06~JgI}9$t-GjTk3~z9^fdn6%
z0nU8&U;ek}@;ucS`=Yz5tEKkdYdveN7c8<~n5~0@*+HOLy*YMcVL>8T6st_6XY2<$
zg^(Um%W@-Kq<muVG?j?g(}~Z1!>A)GX;l2?@ML-`d3EpDI6~o_2~u?Pj_)%zfBLHz
z%-ipU><zYvWqxN_jcYgC%(uv^q7$ByYR)mKO|fKe9lJ)^8NaT{@?lR~fffSptlJBk
z&jQS1yv^sGrNH^=mDluq{hZi1wsK>>R^!&!%vn{bypRf*v@lYWZL>T>i^BuXycIP5
z&<H3Jlb*Fq@XM3!oz3YT+e?s~>4fGDWVCvUa0NIavU_Hw5#2oi7_r2L#sq^UKLsD<
zyW`uR4o)7&H`t9Qv#$oy_86FIRgOljt?eF85&ef5<9~3>Fgdg!pdX+~J%zPIW(fmV
zU;73X%=h@>a+MeqU~+!QK(aMZ`(^L3D6Nnw(vR>xL$P^|26+X66Z)0EqUo{9;i{Wp
zN}?)h2B9w>$zHuk6Iy3<m$%axEO<gUFhZMo45Pa1p{EI~S{9o0F(?9}F_X7;IJZXj
z0Jr_z=6XqZ-UGwhYW|#=f9|q0E^&PGF-Pi|JKI<uur<20a5mwqKpmUm%;EpSd1oO(
zZ{NN@Htv?Q6^PfF4$nX;jH;$5B{jo{O<d>~Q?cgquIPg`$Tc>ZxD*-BFCn9X#n?rU
zo`z&+cCYA@N8XI#TXOx{DrlrwVfrz=P-;(-k_WYU7wcA7#Q~*YkW^_}7_etXF%oV6
zfx{ntk&NPqt7rJD5=)P`Ia`Brnnf00-54N*A-h1P?5LGz(=HYwA|Y?bQC$z>oLf-(
zY{wNuOMm!T^uC?d5a2H2><>*4yMu^eS}9R0y4xIn(i>MNKmDEj?n$zXsgNJ_S2D(F
zgAkY3!!@PtB9^`sLcws1A8_4m`_Rwne7S8)J93ALO44&(pLST2Zg%K~c-@})FQmrL
z^cb@EVddKOZS_YTNtIY1P|}b%`&4fsf}H}v(i_z;`FJ%Zdop@@#ES8>XzDDMyP`^*
z@=Ysr7hgKhov57k^sKyrU|TE0@ZUD#_@m9Xn<rw#S$M>Rq4)|~eNjZC59@;obKhFL
zURI!9M@bQercU2yH}1?&*ZT5b#f-NM_Uz4`<Trhv(GiGXG=rt+2_N}21ztq8^X>A=
zc)8T!w^lR}tJR2_JC<s|_edT%{PatQOByVFCdE|11^?=l9(z(FUeY!3FbKuronm;D
z&)6_wow_+hW`)2yNr1+XMgB{)gAwBc$O$bB)AIAvH6x`;h_+)xiLjRLz0Vp86q^tu
zKWzkAeEQwuzh8}5r(0JUU=!va9H@H@^cvW@C6~TILZp_05ZEQ4-IGO0S0SozM#13+
z$sdEuZd+4x*w2I^BESp6jpky-=Xh$i2_Pb0n|==u;Z!a;(p2R(p-L_z+4ybG_7m`M
zzX8ALy=La$x>{cq0=}1;XfKn7uiPuM6NS(xgB1cJR_t;0JjAWa1&x$P499_Wbkl&?
zc#I(6r-t>wv|&<JpRPwl5{yLSz~8Gk12Tlnse1iZg5SZ#J=yU$C29+))=>@NeRzEs
zD5ybDMwC<|A{I<jYMo=O5@I8}R0v8^Hra)5Kqc2diy=YA3KV^#YUfL}!HX;FP5)zx
z!-Ngs+|jm%=a)qy1I|n9eU=Mj&h-bf7fXK0!?@2q2*d1BJ6wPK!0)B{)~10Zf)WDz
z;zi8aIQ-Lhn>|BKWFMC^t$`a)Y?_j<SoTovdaDl-P@W68NY4mk@4cG3!?;NA4KdkP
zJ5rB%PMTHN<f{bckIBD(E3>z6*M)_fz?u3X;rQ1bjEO*-5rH?C>@yX{lkVTv2}Qo9
zR<cdz6e?f((Mst1*o|(SRU+n6a0ybVg>pq|wq~%ManXI^_D<=4F$|WDr<4H)>Hbmu
zE`1L$XxrX?ZZ(bQ)b3SbHcS3l{apxOR@G;os|d3vKU41q>r@Wq1<a!P#Fv86!$s1N
z&CvPsb<T@LUv;hmx^5$v!xfnCEz}5bH74I|Bfsf?W73FCvRC~TekK$hba=}eKWA#E
zm4(~<?jkh;$Irc%?IW|H(}r$9^qQc8^x-9Psqk;QF@!z%8(*V1G}G+`&L>?TWCe|0
ziJ?*3cyF8Cca6WsLXcB>SngJ+s2Ub}O!eBGx^~*krDj2`5>DUasPZqKk5O`vn3=+6
z3=CA2VaP26E(eg#t_RlF;7;Ji%qtx4mA?0)q%*_0MdgRmJ~IaRc6X6Zgxs1Uj@1FN
z(yQhsjG(s@R;tVyNjTnkxp*HriG!~!dlIBeX}Pgr5vo9;48tm6WHB7!)hi$(JXmOn
z1{Q!=qHP!tM{cB+M9`Q7=qEWp0Cp^!>9FL=a!b~M@<?eEMUz;ihd{+MCRttKk#Fg`
z*(Ro??Yx>5LY3w9I#0$Fk0=(A7v0$lzw;t^!eEn$13%KVAcWz1#^0aQpaz}i$njfu
zyx!tgwzk;x05JGnhj^TAg|7-J^Exc;HPW`zq_!saz+mAMCLV1Nw^}vdkVSA(T1o-0
zi=V_^=AF}pX{p!lI8Q?<u3pvPspr=8VJ3_?gy!`m;C1W%dF$~5jn@S+l-uNRhxO$R
zy5{=$WL@t_CzOXbub55G;`iwK06-R>w|BXJOv6>r#oarqx|(-`exeOfkxE$c;Zdg(
z{X4lOrd+;$XB6?K`;s9<nQ)-51l?ry+v>A0$<fqRPH|G3VPfN|+UHE#q;E~z*ZIB)
zD>FE%pB>I)4Zj%euD82!an%%ZC~lu~ke(+jl$jm;Gkay+^!@jt(Oun;6emLd@tIqq
zo7*^>@d`ewmhc}O#em#I!~15<**u>ZaJ~IJ)UbbGJDcdtZoupCVxBuEEOvRIcpWVJ
zHzV9hw&?2#E=7V`IFXuoxNGdfx5ubAV@}t2>@H}wG|8E5D&&eMu^ntxGHhI5D_u`V
zXpI^bC9-4%NrD9DD78t$U5u_p{xpYumeZ1H{XBmh8@b2QQHjqlqy7V<|4-hte)d`M
z>^EcA`%go_4vkLaB2akEWOk1C9#_7^4F#pFGGnY1FD+hEgT$h*eYpG-T`3ueF}m%p
zk8jnt`9psMIzIh6t}Z3R4pu+&K#6*2aU=M9?KqY^IBh>or$#fD>B7%QRu*#0-fZEs
z_PaxMmY-dl(w_(~pAI9cxIml44vgskgn&^77;)aZAHt;|JLE;gDxK34NsR9LVD|4R
zlMqqGO<tGNu7MPcf5mfg&1^Ujtgo}^MVaNnUrLTt31SjK%`=_>=dIp5!_e`oSf>D_
zMd*ygIl>~Sw{I6hXkmqss=x|ce|j+6PYZYyhMD#|P+Q=nrm^72h4S5-0=g-gjYH@^
zvyyCnO|it=oPqy(3lM&?{_jxu_3z)h1<0EEqKIgKT^wg!*VHmBk_;?*vz&gv4y8Fj
zFa~98YN?fd;+;nHh?Gqmk6#)qHHkeD9AC0S8xz?ty6yO1b$&KAs*@<E{bC~-0a0qO
z?wz62)Of^%wWCH47*@we$Sds;AaLrnHRED$pkcd;iRFp=A#{+I2S+ZtkWS?*FyBAs
zv;PIhM?6|<OShDBsT<<eF9&t--_QTK0~)*b|19P0+`UepyF5UcE4AC*5PQ;_pAMer
zIFA&MWpq?G0+Y>_2m|B+&MoiUhcr?=Ntb9P$acH!rg^Fjc~$%4e|lCDmd9F(lQAIl
z$s>>gjTPuL2fvALdE;H(UDVN8X@OHT!<l1HDJ3~Oh0}ktb56fOtwbI_jrTdqD0B^S
zP@;<E%%c?Kc18GP6^WJM-a-0|;o3>Yo(Vz=gV?wsLDu$eL+YJT5E2eGf?w#D?&%*;
zcjU4G7c1i|)LAhYl+j~&89o=-4yLAiw6D^=lb?HUHE4cr6TdEOz1E4km}iEq7{Y0<
z?kqG{-qL}QKkk~JH_?tnpyBgiWb{6vY_5`o&J!=P$_M`jM$W0Sq%m>&UFm+(lT!LQ
z6vX_F%k5W5QYel<KNIfp6e+oYm8jFs^zV!)qDoaFROBRR3MolrWqGWeu_7Ok;toyO
z5o0iPC2cK3+JiixWmtjJ-LU<Alyg$g5T4}Hn{OM~wPmO_5_<UhX<w^EwsWaO+L{-N
zvcdbR^WTPXW^yb6O;eXXH2Y(xcl--yyvxUzkFeRxxuS%*Gp;t0N-)=Z>1ZRsKp2&+
zxA`#PSfIWS=?<!Lu9i)ppqCLJ_3bg|<kbLm8BnsXpK&xYI8oOAy_mqq_60=)+ATT0
zij3vxrm|UnZfsq{t0AbgW+#GTz@RWM3l8bvP|4!^(^s@a%b)i%)SgZe9|Y5}Q+qCU
zX<xIu%eyX5Eh#6K1x4=S$COG_Cqq#U^Y~^2SQKYwZr@34yl>nsCF|g93>ZRU3}&(`
z&{b^u!!n6g`=wj8Y2^_95R=n!PEX@^?Xi0#6iX3M&6W&h*Yta}dpV&(1XE&@l~*Pg
zmLJ`C9^L-D_E3r))yr*sEGB4s$_l!C47Kfj;zywUwrzI;qbhn6RXjbT3S<VoR<J3S
z=_0qODTT2)LNpD^7n~rGKFYXC7d{KmMrXW14K>Ac@GOOgb2cg};V~C}SLDMC&NHVZ
z+MhsReWYShX?HbpvP|gxz-W_{f1*s2#wLIBNu)Ye^|`-@<wmTso=mS9A9=^WMr~#-
zcb&vBqmV1kuJ*l?FDs$$Qzu_yLaHW*)DXm`Z;n$y_8BPdvzcGYE9~`Ulv>gw^65-$
ze5b_w&;}~%b%dY0h!bpr?h(;(VE4i%B^aoz>3aR^DFX$(^AH0Z=*$9sG7|ZmWD~tJ
zcP#5IEN7Gx2P(*nqE7{%DFk6+0OIsfmzHW1NhkzZ&-yVYMe{$|P}qro{UxB526b?D
z)ZS-SA?A~on4E$GY7VRwIO#tY&0%2k6K5y~s@Jy!$$p&&Wm+i87QYF-5^G&K2!oM%
z?R5XEKHc9JXIXkPCc0joAx$`!aU<q&g7Uhu1z)qWn~94#pFefH+Uz`c`TG)a*@e$e
z|LwQv&%g9NuOYVWJ|!AlC>SC0sCr37zZo!mS-gjTdIqA2H^I2LYI%Wb6G+LoTvema
z(72}b$eJNgNXwb_Mn~7kus~V6?*V^;^rfN=)TS)GcKBPF!ihN87tUKOP0rhnz+o6}
zZRDhCFz6Gtcb|~A-^l3cuanm_HMA=lrtW(W4+N(yY;JC{Xy}aWA@cOPHEZ`rEo%NT
zRB;TipX5}rQXojLxE-&d|8W;vaAQr^`wK^}&CG-wpByO|TJQ5(uSx)vLkY1H={KK)
zK0dNM`zs~+MVo~N<gaXCh}P%}YkGQ1{fujSi?<#?nEeE}JF(9Y)95In=&3Sn{~+f>
zLV%g%2>ek|`e+tSyXkNunfToIx*<GV$fj(v!~#Z2kVn^V_w~-T|4WJ9jvMqb%m*o-
zJgNzejFgaqsD*hs5fD7_&cbSZ!cKllcf^J_bb6X#x^&1Wq4*STfE25D{SkTzrC~N%
z(v^^ih0%X}Cv0e%!FYzU?!Uxm<rO{eHFmp5%JaOf@pANU%UtlNsabUP?cLVo_*#P7
zQiCNmz%?=L;BEC(oj4Ek3pMf(QcY2WA&8|*a*D0)$7Px1ojzM^B8Zv%9f`&pk#a_o
zq&k(Z92*e6?(&AdSp|llhN-Ul$lKIfj6HmWC1fM!w0??GlWc(C_d(N$b%zMrFohbH
zQUc2_q*y^Rk29Wws(7;_%6s(1vXj*GxRufE^+G#Ob@|YW*4ZX~5C#LhEsB)ct?5H0
zF^3&RDv3ujLI!H>(IL6%dn(jYYjU`$<Jqur-mq}m*yfJJ_s7<#eUE94<QiMC^O*66
zq8<sk<E2a1m+X+(8~<?BWIBeZtm^9Ouhq<#OrGnW(pyg<(nZJY@g84SVej3kKiwyn
z`M=(YK8y`cadFW#D^T(xlka^)&h>~9T$5X?_|mTD(}ch1p8?%2Lzj5Z(!7WO*i)vb
z9S7)SG_2P>EeBUuS2yp@{kwkG+_B;^j2I;|mCh?KudWy50I?%v`@8_Ve?FTYe13IG
zK|kw7zu{w=5D{^5*>ion^*MHw=x2!3##y2hPZ7tz?Mt1#5Hc^cY+I!+-T?3@t-b5t
zvrcj+JTce|s;<Scb9!+K68#z*v8wKU&_{CrZLjmsw3#o^MTW)*9X`RnjA#N9G>pW)
z6Niw0Fc&?uU1GdBhM~da<YfHaY$P3aL5g6sq^<rbA%jS4^|XOv*B3R4PwlzN73Mx`
z<Lf9vAKh3!3YK9j*^fo=B#Qu(s15rkg!8*!rKDcBwSHzVdH<D|MFwXABUBcPAx*>(
z-pf|@LK?Exr!LqZPmFGn3>8-+jp?8ePB>y|umyEX@QnwwPI-pYK4|;y)BTEO&<Ab~
zM>X%CIs@Wt2K_d{3lJ>@@dm*xsUBSg2n0t+MVA2ejpa>qwhFN1H}k0y(Le8_N~;_j
zhnhNpB`3rIS=*FjAMTg`_z9wI5SK<c>$#z5Wba;*sE>~(MMygvieX}r8E9LsFjIi;
z+`9%%nC%+vY~Q&EG26Ku0SHW#wf+eEw~}3*ms6%IC1FqUPo#x6jt?7q5^(oyn-QD)
z-;l9$H)hZ5RnVG&fx&mjGQFrJkj_NpeDTl-rnMm*0N*swStzWsZGLij+~mnE&E`w5
zZ`i}1x2Xf#S<!o*n?PF&u#YW{S<f9ChqA%HJ8$SNrDv`v(9$|jMbCvUl3=7wSTAR+
zfcD<O8Shid)DIQ)lF6I@_^!6njniSbNU5b<{%v(PA@B7rQqbc}Lh(_8W`D_}M-^r`
zwZ@#I-v<$at^sYX2Uvz+agyx>H(z_Lur=@7C&?Di$LpV{(zw~NS?s&u1fO>5dcU??
zrSyeq7yzF6<h9@bMDC%(?Yvo_PBM%Vi`9^)gi`B0h{q(CbsWffj!g$-`pEcwpmG6B
zed}sVtjkeyougB!Ko}l=>o72~kG40HXdukTminPVjwoi?`{pPo92YY*;oPTLk!T}+
zOoD_lF1t7%Y4ucLME-8<yTjN5^`U3XQNaP};5D<jn%U5T8yB_VJ}hBFMM`YSTy$5n
z_km-n6DuU*+Bf|)QB6T0=H42JZ>_Aw3#@ZF9rzy-VMpV?&yeqPkL{j?+;e|46P$2@
zU|Ey7>Rd<3R<COv<=Wo{wqQ954G6QVsZoGK{S}WmlZWOaaUtoToKfMH0Z<rkFt75(
zKq3dvaz{fDhudzuq0$DoB6uOwxO9NdH(x2IKr#ptF2l~d_(vfaY4(nHMBXl@pzkLM
z5y{o<a4^}-`u3km6G-u39elO!vg-9=>)_A1<Y4bh8KSQS@fZ6E?u4mHzS$}w60{qp
zUt9A-P}XC9cma9nw7!YXe$RN`XBD;=DWMWT|B=i`HjO;%ee2^xB~|&8VI6kpL~aa6
z`9yG>>_qv1$rtPKkw0q9hZ3<9Ib84g){nnGRrr8@%$mn=lhu{v^sGW#*7V!&#^^W4
z1!lf<3~2+@7~{3_hK{g8Ik7k`bB|+LoeYy!X#L%uV;MsWDv8Nv%=XVD&h~3lJhe!f
z|4}(L@QbweIh4tRor>Hq8sZ&eZ3Jm_QW=FixxM}yzRKqh&JK>+UUhc+^7Kz{+R!a(
z2o2z}^$cG(gTD#5?LKd6-%!Nma5zSD804kLkn2JUMPIc1tJQCc)BqA>4mMxSY|h?t
zC1fU@W}@PY9~BIV#K@NkMbOO+Vg|MaK2zh?svN3<MB0Qaa-f89kh(x5W|*vM!jLlq
z6j~n3#5mj<{VSZ`O`BD32XGnNaNQyHCwWaAlRZO{P}|fm%0j^0;(*??`x-F1rG9@f
zo4^10b;OLxD#+RQ&QWPERxU285Z{h@^=Gkl=!HaJL=;)3w5!N~jW2Xq3RYEjl@H&w
z6>DtU;^v~mBHGnQheulCNy)5s2Oa#b5DS9+C6P1#9YL#HxQssx{O{8}4VoG`d2#tC
z65<m2TU?b4b5=1MfQ?OQ7>}=#w<_ct!zw+?6+Q&LINJwF*K{`~un)UIzat(siD=vX
zUc^H+Ppo@n5~{`>tG4@R5b&@}wC;9K-V4iq<=MJFqlI1L^~}=o%#X;gCYI}x2jB>P
zGJ-uK?sDG@j<wP@>e&l(1m=6kFpehdsYKz2b7`uKGKN3XSZD3b4}s!@fxo>ts7K0m
z9dKQn!Gdg<Xp-Drw=7$aFW6sJyMlg`EV7-6y95*9ZuV^Mxr*3&*yVlH+>qZqpKpIA
z|ESRC#v?Twa<?@bByq0rfE81c#H3&<V~3D<f|O-TKK%T<ru8u)8v`YXmY3bt$)Osd
zjvb^1xF;$#X;8B|oZ}qfJ%Wlp(JtiaaU$$TYy>MpMtg-k{w>o0w<kt_^JqAVUQv-i
zHO(ySyp!Zx-|y_(o)zDnsa-F1F1!ArSi_7;l*(gGlp+uKnG<-&uC1%9tja`x-dgcP
zsEvz9Suy@PSEzNUNGwGLnUSxwP)-91WwAnGP-0Utwbrd6d@sJFA4P{n>2AXmQxai|
zZjQrPdfJUjoJqyoTxiAreW`*KRW0J{zHfr*xp#2byGpEwB3rX!Myi|+$4tbYU}M`5
zVQcZD)?~KaA()>)$qU4IS*&<mkwgx>wQEX;8NdET1k1I|yx!l{o`0S(!5_uUBm1Nc
z0k}ur_Rj|1u~0_w$oUcn;VTpE8fs9d2^!NthIA=o$%(wtG*xxecxlQ)r-S>*_PdcG
z`ps(};7ZwH?P7)D3AeaC`E=C@Fm2b#<I4Z6R0+ayS$CX-7Ok#XDQtP-?`Np8yF}iR
zhrWNe?!)T&UP6vq^`$r9A=~%4!nskexOPQOydptaneqjo@*fKZVIiAs?FvG~4BIMz
zkSr|;6X@4>G*9&Zd0SptIK^6g9B5BA^J@WfdY=~<rD|D$AxNCJGs0xqls`|9EJ6C!
z2xmS-D(u~{$49Oy7KQI}6`b5SDCy`~qQt(JLmV+Jh4$ER(cSsCp01WRNZH25Txn=2
zqTL~FN%hNlmiD|%G7J4>G8#8^rRPYOj2!gRG2e&f3=heR+P#h$KE}T<gha2bP@p~2
zEY)t9BSJZ$VjHzO;1`wgn4DZwMev{gy)BN;^cG{wj>aJ;&f4IbyOA72zXV<t(eD?$
zBH{McxxM{|CAeT)2|B+}_ZU-YNV&=JU1&GbZn9n{vZct^)7?#&^$uXB8zZ>SGeY?*
z9N4FPJ+91$L5x<z$*|HVKwqRaqV6NN``JrJO<D6%Fqz`;5a6ptKTS?=&L2Dws9%6h
zPC4xbFge07_>xQwS07uoHxcgSsqgR#tSWl{GAxpR$RiD>>xIYU?>KV><fZ*l?72Bz
z@`mO*rnk>wX>t@-N6N)1wj-5r347vd_rBgBBA)l{7m%*_4%Wj0c`pbC6945uTAQ;S
zyVzu!83D79L@-JSYhM=r)?F0QDZEl3PA`}(KGJe$UqtX|qM+GBY9}Vf%75s9GhOWa
zlZAufP8sQ`p%t|}@uyGxnZ=>vBjMq{&UVv+>C&oCaULGSCc7>+Ne#QEPWsm#N?2bB
z<Cv_TM^N+pFU3r~pY(nzhLfIsy{H~yWlD8?BN4*EfEHG=p?opS)iHG`-|zVG9baet
zKmbC{mi=&`=h5P+Jj%|&({_Km)-0LfC}h5*vx*gpqQ~}=YK^IMe||Q%STcKojCoj9
zy~s8SFz98Qoy70(mumSkbdU`=go(ZvIl(y(!5<-z!}?z&{S|~zQ7gF&)yNKD37Yk6
zYevg!U>d;3xTuyrWp<ao555w}T>U#GEb4J0yy*)Qq@v2kd1PF|u@IEQDkf&DK@f|-
ze_vaoCJNRa;YqIwbR3rT>5_kU&RHETX)bj=5ss)jcU;IrOt_u_;ZF%UOx=R}>yw{B
z)Kaqc^enLqyWyJTdi%)SA4r?XSevU%ePxoZJh8gnaXXB(tk_MhFs%bGagx*%31X)L
z2PTp8CqSVs80?si>S=b3f<=)6!Jiny`iW?M<<B{Vp~zXp4J7U%Gh!>AX&Z`%xNFK!
zLkvMK=ZpeF(^BcCn^yK0I8|<Dhpw8iP9DiYeIbPY;q34iPu_luB@E&J*xKge9|9v=
zB0nHOrOBf$Dp3q2jL~!=jJ4}P3su=;{D>*-iq23{T+Od^9Jpx}sEjqBG!^7?X7z8w
z{=TRImLNh?^3HIJn&GV)<M&5xr3Wty83=fPO-zrwUQZ^CWa@`a(j~(zeEW@I+Jb|S
z9jgXE<2WpqOGB&U>-UQ3fpxAVTKqKBy;F_FCknDwV$J-G?dhcO75Vk&zCC#N>s4>*
z2j3o8jr8GN?+f~u&5n-N&L6wMQc?sB8Ab1_>2wuNgpQgIFC-0m#Se$Z<q?WA3xsnr
z^4Y`5GZ(FBUxBpM+<R>tH%P7&ng0bP-uo}b$3CEv9Mw8L@(=aQf~bI>cl>vod}{g*
z^x@I8T`%I!rZoJ*B$JacE?TjHTmAgWB6^q{es<sxhPDaaU?lH1^J;}?BuzAO>+&ls
zJ+SUG&9sNyp|?Rv61m(Kt5y~t?XJQ-CVIRv^PaR{n{fwmunptxkz<5GZ%_YCoFjK$
z@6_K3{%QTU(ipkwFM`ZkdbFAv4LPbb0pIe&H8YH%hA+AH-lr1@C!5!PU+^<j6QE0+
z+&QMUQd!VBNAxzoW45|4J|UA(;dZw}VJm9JB77<vAJ6_P@QShJ2WGI0a#<_4U|v@{
zb^`{aaQ6|E%^PE2D%5P7HHDFoSw5_gP7DH2eWJg_#AoPR4eSc+nu~>{*6$Fg#`O*E
z?mf_JS%OQot<rRe^Q}mAfB~}=IwkVvU4d2};d^CSAinL1DzbJ6r)b*+2`~En5|8Al
zQNR@AUEc0~zhck*ehA>jYdEuss_S;qI?v}6p!(z`rE4u~jcODDIxSM~Yr_Dk2kg-f
ze_WJFoi9$IFWPyRGM;GD*@x{1^grW{o(_#$GGm>KUM6#jthqYUi}n&5z!vDjQvZbi
z;a;$M<ZHd@5Tu{<S70m8S-mi&8$2(>Ax-SO#_>DzBsq=%ysMDND@S{ht6ARF5VEms
zWAB!|FqkaeE@eIqGZgbj=NvuT?|r8Ak7LdAjdXtfS9e*q^$2ya{T#LCWc#<chf(gU
zk!!d|qnyAOhHFYhi(6cqJg40rYV=`_^<4fZFhv!*4Y~T_a@+UZ@1D4ZIBmX%sH^VY
z@ct)|FmVTjyy`+d>SjX$ez}&eOo{0w>8nGiFE)B;dxSM+iI+4)>^d-2xIL3el9dw6
zfsrAI69){;xTAQq1k8FuP+jH74Sd5fs-hY-;;X66z4>I~aHo-Wm3fO$im+HZ`MKAr
zYk!~pd{Q{`7?B~Dqn2;Nq_v!XamHmwMPNL5HA9y6u=h^AerW8Eslz`!JkX}5x5n2I
zE$u#dF4v&gb2Crrd%nE<c0*C`s}!w^=Vr?f)S}xjr`|53@3|x45>FeRJQplF0WRxG
z7J%mtBmz9lDMeC712KGp3YL|E$*=QF@u&ihD$O_&jJ%SzZfiL$w?+~JtF85$)aHDo
z0Gm=gaiuxS@E??e8dpJz$J|N1V_zDFs+$G>vj%L^)AbibK8MO&i{ZzL9S#<9+5IE<
zwKx>SR9~A7nZS6XHXlvCCj52_f{n#u`H8t$jQZQ@bM#bYG)Xp5%l19{_@nyF@PacP
zi^)5h2eH*rU_I<?$$Ka4TnC}8^Q7*H>q`f`7$_5KDpR6*<^a^eH&JlmbRIKXKl>CV
zd$8)mx?7>u(@8`x&VawN+5XnUpWZ<pz%43elDUst9G4z7U>#cbO4h3;W=eOOrru%V
zV7iS=x%7cqQof3vfNkhMH)Xm{Sh0;LzecCj-`0>wghXrEDH{>ARF=DjkvbiaKLzU(
z&Z{z1>yZnjsn7GW;l6dOH&qt+)=*zR_KRrr*M!*JrbP8dThNWupXt=GimFZ?d0!s6
ziHf&Nx;H`>Z;DyxS{NqL1&xtgw|=~F6tp%qpCHsbbQckc;-fwiBv@3jYx7LXoQOPc
z$rqRQt~Rb%F7GcWFQ4<#=IxY=?clOgPDlfWgr8fi2^=lr1mT#pW-gL5;tMmYZ(8i-
zXnxDC_C(h$VYo)wbeEFbbKu)J|8Sf2Q+iylk<VqzkC(E4IOZ(;%5LyZ@?CZ9S9itM
zHN~G!;U87gSDb#Y0euB9?3ie#7g+)xc<f&iUjI?#!LRmcf3D}|rUqQ#oo(mCQ>c)x
zpYmsr9#t%BjJec(jFcAwMv<-3@!q{|Xg@Xuk}1n=$$JH3g@K)pW@}1fIbEqlx>zMa
z=kGaDR@bZjPL!%Q0;mKn`LDXMq9P(%J?kz@NAX+X)e3wJtfSd8{^!8JEu(@hlPir}
z|I!b`B4ok1PPK57JU5}-y!7Eu7JNqAK73NYn&yPSyf~PcrIOZftT|JK75pP*83}Y(
zRZzqvaFv6Dx9;O&mG-tST3;4<Ugs8Ge@P_nfW~NjpA#;_STsisaN|l3N*1tDEDp3k
zE5^(#g#$^j%1e1n$w9kg#~VHa?1)cHztcjLhg{q6$c9(fTp5#TN`xuGhM}-8|M9Ke
zM5Sh)kuIhrDpfmY=#vcbw_S=s#|TE=C9i)|#yA2{AK!oP7Y6ZF<+r_o(~zcx1T;jf
zEY9LSvEn!g^6@y@<0F5s!S0h(5R^XF<`g1cL;j9u3jdh?O^mELGw9&URgOl<f^8*-
zqCVw_47Zw61wBr#Oo^;G#RG(wJftCf*rhLs9PS;~x=-{lqLE8^C(1H=Y=$&D$xgUT
zK^5+Y57|}29{Dh!N|v1yfPJ==zyA)!X%hfXSRdR}z_`W9_Ft{wp)XiSqBbD0vQzA>
z8`HM+{Odzm^J`;<ttxrvQ$ySGSGR?Wi7IpV2g~?c@4Jmuv)3PFqQ!8*vMVXW_1&@O
zr-e4q{d^k*N83t48)sJgXR02>zu(Tg<uFft_mh~<@d_5dtxMEr(O5PvAvs9j3+RNM
z(vBunni-eI)d%iB=Z5$|8M&5m)LQPu>{-u!hPo{Swo_WAa-f4EJ;(t!buS>}0lbYM
zz|DU;)h4|?Wg`1Pq}U$ocA;+0NIY(N|ADUh`@V7U(N5auIG@XhXX)*)wci_7Qa17O
zCpXzwy4}_!JF^3};1qlxTAKd#cuqc?Wa;>YUPB__@n?Fg&-X7&YA4;GPO`e+hleDt
z{~n??++6a(($}FUervM;;OYjDXMW@8=A|y?7-oCdqj|ZXVrKK>$Ble~F&NeC`R|g-
zg630KfW|ay5E4h8*REFldx_gYv*&swZJqxWObd=<<!t{AQT~+wAs3e-%Es*fKUgo6
wHvIoCNYOc$`QPc^ZWZ7Ezrp|ghcNt#=dh(JeMA~!f`lj)MW8~hoO$s70d+3|ga7~l

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_001.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..3bed0e90f94f6a70bdbf62107c2da5740f38e490
GIT binary patch
literal 120009
zcmafaWl&sU(<RP8kilIBcMt9m+#$F-8G^gJYjB5PA%Wl!g4^JOg&}y*fe_q%dB5G-
zdTak|)%AM1tGcS6?$f9H#%pOPVq=nHA|N1OD=W$CARr)lzdnZp&|a@qU^S1|n~2vt
zLoZ!ddoN!r4?6^PD=#-^S1)HLYg!*W4^Jmo7k(}wE`APLM=vioPf>2}kN<Z8m#c>Z
zH&j|~|5XWwo05?y0s^kpe@?_NQq@ifhzJPE@-ljUh5w#>8|;jH*5jY9*GEi4OG3H%
zEq;zqp0`bS84xWGBBR8YQA{e4rc=f!<j3a%P>ITBq%fmdoCpbc<vd1nXOdVVC=*E2
z5MmLm01cbuAHQ~fn<$u6aV-{FEbxR1-rJOTo&+zfF(hPRqlCAMHZ_a!Zv?`NgQ4pi
z_ptUJ#zN1OZQ=hqZf`_)Xl_s`N8JDG7)iC!^8K&(GiKTpnYI63Q4%JM=KbF?p-5Y#
zzo`D}%OghI|6d8<ndfEHcw%pl=_-H<G<5TjpRz3(?U`a;KQflVoPJ)ud9(fY<%U0H
z@ln%)vi`25N7(#AD)=nc>_Uq0%gXBN&gH@U&CyoMcB<g<diTWr{k`IS{(nTo{$CMS
zlmCc(TRcQ~Tzk28z3_d3O?zHkd#7n(=;wm7Z}~%?MYt?F1`<o|2_2v%@06oGkzx@*
zz&maP*4@h-pb`guB0rfiMWH8PGff8=g=YmU*9$AX4G7P~hf~5tGe4nN#pz(Oq8e4r
z8pU|J>}?n!L}{^*(G4Pyv1MigKw3Dpo}@NB5VanCKzM7w-eLbH5b#Kg6Pt$5$Vdy6
zy+2IHWF*p2X@ONJNANC(x`O*eYxyh2a#t36hga&K=$Mvl1n~d0atu1B(>&Ca><;B!
z05(Ij`N?xEVq4L`Z}T+BQVBl&D!l*;!K`chjV2t$_gr6kUwSjH9$2)4rlek?Lm`1D
zrkf{EYoSBt;0%=(1*K?*VzFpEo4x!ukAHA`)l5?mdykdErPYqD8rO$k^pO~dt=<}C
z@7^qe7^AG@fkP-VafBY{s{llZaSIGZM!H6i4uDjlX9#{Znm2~-?D7yGtAv6{gf_}@
zY7r^DRYC(kvJDQt-gbj(97ndoHbUWUuu#n!z2`@?FVa93nNp;R)5TBrtVW+oTNS|A
z3J>79e0RRB>ji_mKi#1*%4jX3`D^;oBvKC~e*5v7v{?fpv9rtd{?E;IQ!`AA%J`q;
zGsuMe^^;o-9U390+#yfxw}0-FQ(O1KUc^;2)-h=4LUE~dSR<r&Zyq)Zj}o_fQNtqh
z$OK)kzeV4E->A8TZ|#TfhTgf7#nCfKJq1Gk-AJ9UNj`SJU~fzW-)@KrJ()wT)bi^q
zU)rrm!_yR>pJIE<3p#{-cT}EUyrAh$5_jdPoqHo^C;!_;Z`<oOXfS^fp&!N*OcaXw
zCWui8@xZidL^Vq4F8V6pL&awpR0*d7B5AQ2VJEtRFZW26R3bZLxV{bzqD7R{s9)X;
z;#&bV5fH<4*rX<ReND@~$yupn5NNgePGilVgDbJ(2Jo#YK@pU;fbdaV-@5A9{Y#!J
z^wwUN*fd2cO+-K;GAm(t0!Bob@VWFnY4Vr=j5uuevv&+Ee^zVeqZy;#0|2$GNsmtP
zn{VF&?)qLhIt>#*7B}HSZp9)8oo?B2&STuIFs-Tv9S@Rr;EbeTzj#tEfA;q9P&Mf8
zrYB`e<#v0LE(^q^S{!;t=Hq)^v+z~#>-#SG3zl5&o<%%MX3BA+!i6CP8boiE=qRhP
z$b_Ke%1o9@CInw3oGf%AV+UKz!Yf+WpIk^FLMtGiJcuq=nTc|vaqrnU55eul^KqoO
zJ61MM`BQir3LAq~0#7K{yuh&Sf!Zg_2FR<Yc;k!KwPWl8nXxOC@V|mix~?#BCGUl^
zmSvwsIMrY)Rz(m}^-mQ^8Bq*ZVY8%i*&A{;SSn+w-FuZf@h|f)n=g|CvPBm6!xp>X
zjaxMB8CrbL+M+MBrg}r#$uxIon!4YI03dgb`?onrj0?8O2~jS@lVXPi++a>X?57CY
zxcC85B9zfQa67Cl84bPCBc25-T?GWpn6}_}hv?xR;-JkY{VstP>#6lOQsD|E-`&lI
zNC7kWu6UP&3;ki<*Mcru78OxN2FR8PaKrJ+POro;+S>qU*gXfgW^+t9(zu*GQ}p(I
z>6Q5gcJw|Bx=#$v%W&Wh0}@C}qMi}k=<pS#Ghry(&qB6p2SsBCt&=L}kIpqi&MeaG
zmqyGlwr~GW*nC`po#3MWP4GKR7Bdal=cw<#L+kDbX-as;YldF)>Ko{jCSCiVtQCm6
z9pmrnU~r<qFx+rJ94g_S9!#JsbOe!Y`=WlizQ>a&S=MOG3I|9o5@8+<xpWWcVal`<
z>*G@uSJa?kG7Mu43MzW~_;XDT2Gch*5|aVKFt*StRR};t5eTfDFvs3Tlf+^FxGi3l
zkX(|e$iW?HJtYkNKNWJc7)bGH&&#!*qtdj6veB3h94u~hmOMWO`}>Jd&`c3D2oTa6
zOrQRIH2?cZI{~hCgVGf`Al0>!f)OL#fKd^@)0<U8=1F9C7w^s<EFm^i&C?QNi@n!q
zv94=$Tt>wq=U62{2ii~zQ<+iTZu6YGXw*Ui((OBxSO@LID@l*hP9(qzJqHURf<}nJ
z3^mI9NUN}0-MnLwx^!+Rq|kvs88pGiw#z1C_%4CPLUDTZ4z8w#P=r1}44&GVHhsnl
z8#R)Ngo$|WqUgi`2xXw_h|l*J+TjD_iQdnmg}@dQF)y*jFsxK3j&KiW*R`F!WU%<f
zYT}(|mGt-+JQK>FB9EyhtdJ3CRan$C@j?vyhf5$+dRs>0pV!&m{(D0sEz@12)K-a=
z>+Bp?yng}rd-D&Tk=39K3^{fbS!mNYT^{aA`OP%HQ)>^}zXZa{q!1Byso?g+7c&Ms
zD^h$EL^cQhd5&SnqI%{GhcV=zK%%Hotv!w&c~Li!+ajhkx)<XjW-PU0>{#@0q;O0^
zRNb0P1^^P$flL=~jF)(Rm>L+{Fp@XhcD1_uK{VV^JB71~WB=-Y^~bd!uGGD-cE=Wh
z&UP;Z*)Oq`%qTmZ0b`qieqNH5CHK~-&;Ng*g!Ygkl6UXr#fEhu)u$xuB>k@EznpC+
zROd90lM}VYj&$!P{djD9W-kjGy&b{n&q3#mMg^|dNKx!KYH3KdZUi&LF?d#_baqf^
zAHyNHbaufnW=fn$w*+a?fNr_Ab{Af9pZ%8WG)HRTcIdqvyAmsDn`=Y6?`TM1-`rnO
z5^BGg02`viKEq9Ze=B!Gi?11|51Zo6%<>^W9<W5xFr&^ZC9T4{Y!$UXU+s#VPDb)S
zH+!v6*ly#5V_%f)gd4M<q8A~%taU-g@0}_Lgl(An^!H*Na6~ZZ6A7Trlx)a4Iyuyw
zCy!zfsmBM4&nq9xZBZ-7zew{5Ey`ockY=(P(1{iyBSAMSWN9Tv`{*i>L2fXMSauk2
zb3zMBH`LO?Y$p$lLJ2<y8zacFo+I}NcVX`w+RTuGfOOIh^hsp1htv%n?r#b%@hUj#
zCw1$Ec-v5{lCuclHH*YyfJ2Vo5vXHi6uiMO25J<Q&5}e8s)x5}n1ur($Xuw8Zz&Gz
zRZ$|OQU08ABoP|2|6c2kfQkYr=w_7A(~&oIxO#8UG=sPN2Xx44A@K+xD$>a6p0f>-
zIG*Pvd`l`}b~nPCbDNlp8$XvG=t1+h)gOUmaGek86M3H4)Iql>7pjOxf=mhe=eBZ|
z!QBslnOr*~ceKLXhScVhCPF0E)TS~786Yrb^K!0mkThAaIPV%i$xSDvkf%cy9|e=_
z3oCs|j(lRMF3-<mvI&`@NwPc2M|MUg4o7XMFPAKW%urM#ZaIs?UbVBharpyAr<f~J
zH|3Dnz>LTC{fe{869NgB#9sp9N;i?2s}ea-CHDcnjuL)oZwo{ecP~mD&Fg>AsemN@
zmu8K$B+9f+Qa*C*-(4gga2TlF<=<JIYuAL1O)~!>v-TW~+c+UbDhi^;CuEW7T9(HW
zBL<y&<o=vL1N3%_POKwV&VqwVx98D)>6d{)Zg<{-$7aL%c$K5>*51)s!$jn&Z@(^J
z{zy}Lht-L^e4Unjc;a^P<ZC41U+kGvT)HH_nJ1PP$!roPTmW`i#bc6D$>UdgM&R!2
zo7K5c8`CF6A_^bdZ*ip`OH-kfO;jf&J|eO(_B^^r7r!}oz_3FZS34%V)zR%YHNwNg
zj7}qul|B4SSS<-{M(k_vbx)jSDO(dksCI0Rd3?_K?g76hei^f{_;ZgQtEc+5uu+~d
z+=tl8V^3-4pLPELMxRl|D6Tpy&v%yT8p;_|{u|ePgde7df>twy$Q*MjN+YX1-qB+1
z^*sNMbnAv{OH6lvMQ8jv#70Rta~y?qudJ@E?me|5YkZfS>_|){SR`&>GSJ0%GeB2c
znEpn(f13DT-1wF452VqD`gRc$<FIf#G9DFR4T1#qtYnw^KFSV8(o}c=ErYMiM(LV}
zZEFA%w^!%bwC6MEH3U$C7%MQf^3c-%tp47}c<5UdrabBKjGXE;gBFS@@3x9M!Qe}9
zn%&_Xnu?`D8v&Q*ZqL-_>S;RLFst!G_}v>Q6hR#H=y?<*a?ip}vty?&a(Ab-@4B$_
zRlMLNbh;2CYB4SZu~b*6Zq~eU%YpUds7|~>f!7%NR@*)DLon~+9vZ-jxe|&2m~IU)
zhnr;D_mit<SKh%-vA${{RWrz<2zM20vxR4dJ8ah>GrG~0F05B9c1OG2(eF{>mCj3^
zbYF;)&;+57MFFba>EwkijJ;vZer$yDchu;_hYWKpq0z?==zqs6`BW1pqROmoY!U_L
zS;)#OXcJInfQ?;EY4uB!-h^{eTitl`y+r=OD>=PS`3XU(xod|Pz30>66E`8OOyIsG
zQ>5Huj2+Xjg;q(uIeQ-0hZiqW>O&Vo=A%wo3ght=jla<obR_Q<Mc^||vc2u$FjnQi
zEC$3*lbm-FQx4R+-FBeLv5YqbZ@3(!Qa!Ve)+9bLJoy?9Z{N<7(qR0;V8w{w(~%Ej
zXb!f{@4CAG7_f%dxk7f*8<?!wbIMSr!*ZZCz)$#nbSAnWTet3v;%hTO*p8}e8OgXN
zDW3=`5A0R};YCO(h_y;;|GYkA7rwOKHCg$l+DNKHA-}T4B-WCXE+uY@PoWx1!A$Sp
z7d_P^1^Q1dQ&~4JVJku#)=#V3g`Jne6W<)d@#B(Yg{AQ1-dLohvlU0YWoJVF8mi5y
zq9!DMI4s%&LRQSjIcV`^S%l(^rY%IHiXI6XtOv~VNAz7^GdX?I`EJly^?NkMMMs4-
z3HaNZep~>sIsABo`0FaxY^yXTjhK~l`P!|*%Q)M2<L89y3*<d<j;wjWCc&@2yyHd@
z2uasJnR;1CHl{Wdvo_SH6QVK(#Q>k<&6`-5HChvmo)?f<e^q#OHoIrI20y7rSwS96
zBku^_B7p!(?t;T2)6TY{^a<7<b}`p9QgOd`4J?2+?4K^>#N7_az9#!uB)G9mr=;^s
z-L(=g;*YpL*Rqu)6MjVCXB;uA!bM<BQ~Kmyu<4dV`=L}_w$ju=cpbScUAlVFDljl3
z;}F;Hl(vfQ*2!sh<6S&L4UYz`ouJjwPB4PEu}R2Lj>YzMD|PoKlM#M~ZE(-^F7A(D
zn~ajDLn&E4{R*4NF@dZPYW8PI)t<%CBzy)nM3(D9(&5Z$(FINC+B=J8lvcaFIX*)#
zNnv5`%W&u#xCKX$EHywl!(;PNw3<dtnKU{nHgL83fomm@^RFknY0w`64vr!S9j`)I
zvv4|pF`-&n4mV+K880iu@iWVLEg2@Lt71soQD`kn{TjUMHl72+y>X@UO7K7=GnZp2
zqRn*MUM9Maq`pXGw-k@_C9XIxje4X;6Q-l&1Ee{j`C-@nz*T#^Nl>H-&V2d%avciK
zs_z}&xHAbYNpaGBqshDY<!n{;(D~!hMNa76^jzULdihz}x2O)1A_f^7-Nkd#IG^fT
z0S5p*Y0~Vt0pUdzjm0(o#%!G$0soW2Narfn4`1_$!_Gy=g>%!8_!hOvZ9JS3M*Ay^
z7-U?~D#*ZTpDlCVug5V1B7yOX2F%-yi&`I&$8<(}C6yIcf8KL)_1>i_D>SrLr=MX;
zJ<o?`-9J9=h$d7C!8w*0Y>mbp#hH{jjp6E={&{BLsh3O57ZG0}!CqHzFlA`d7JaxV
zrvMukKpsFxYZv8CCR9(eb)ExXzM>L+!cp~11h7{vC65YmIyB-$&HMX^t|3Is9NGoJ
z9o>ahEMi8zy<n)8Sg<<_V=$x~(Y5EyI|S|<QZn4Z$DAGL+?m<Y!Kj~|!(zY|rJW>5
zjFLHF2Z@v%Ap4pHC|6a=`K+EuK@*g<=@vUu{3tNNjtjC4^JImNDf??+h!^DK5LH*w
zgr1UvXA}q&w7>73aYUgulztp{{6$Z8dA9=(iP;;#6rBBcY`(p`dz0o7g^&1To_Z!+
zCIVNDdS`#QBnuY`Epy}Q!@ySZvczK>{*c&>ThP-=8v(3TVPRwNQa(!4=p1r=S_XR=
z_jLN&cor;VBr36|-SxYS<*^v-^g?rpSbwUc^5uT-5S^pM(+KPGH0Q7om!dX(BU&U8
zazOIJyEYgo?iLdLNo6!5PB;NWZ`84HE*H1!c{T-)tz1vX)=9p7!}{08u8MDBr|&v^
zj(+-lhMv1Lf=M`;aiaeBi#X%yqGFi=VTKO1bXBvd;=bw(RKv&&%$YZ~?~Ti&a6UN^
znzhotzICVh1nfQ6eBoNV=4AMjWaaBS9Z9v@T7?=Yu@InKh(M15y$y}UJSC5P82n{d
zv)AoguN`kNn|B(I9Sj*uE2;|<6T<a7T42XQ3oZ`Wao@{-ONjQ7s>c=Nnww{uU~z&V
z9hQa9JV;*~gcxVvnq$ce<0tg*ZQ{=Qn@M4?zPUHm&byT6@h=zr%P7CZ!Kz2TlDNzk
zi}-6LCgakU+XZoN=(gyEDzk9tfpWG!)PJhz62rm02C8f6f4moNPUz1X2SiPi!rXFL
z@&lo07$Ol_K_q@2t@MaV0+sM=6m2NcqAR?AOhci>i4)9tqhzcDghOKRjcjS9C_Z84
z*}M+d|Lj8+okHQa)9c5Gu%QrF0&)-mb-<2*>NmSgk4%+Hj4$lh=SzXQ2x1$Qy9DdB
z01x7z@c<HRwPtlKcS;)N<EYbR#X|wq+nCQ_GayQtYi%N-&MvmL$9OGKXPLCxD$e8L
z!=fTUx$oco{RafDA)|1tM{g0ZwhoXyGMT=1*1!58tmw$q#+k+j+Z{mBx!B=xhg}?c
zaKS=!%h&myK#SyO7;W4uqK?Z{yF1AuOP<IWd!<BE@n19CW<@(ndPwBFV83Xx2k#$?
zsG~Zr6_Z#F0eG;F#`o$$Q1Y;FhJzrQ4A<rRzbr2dRa;lGCU5bY;s4?qV$MqCzO_zY
z@1ZT$n^$@NW+qGWKMjAcQDpd8zi!S^KuZM=^^<w^uqDqQXkDlCM`wH#EmysGRZ!&|
zWkG>h<~Z-#HB)`iu0e;Jd%;NYoq!5zPv%aLf}yiGsx{}hSG2;2VWgh4efk`K_OQJJ
zC+QI>GrgCf{UPil&dqXdeKCW&g1mMVkUpX>&E95C0L1p5)SkRTn&cYapBTg3)uDDU
zp-b15POw+1Qn;Ej{G6B`M{y6*^YFb0G3D&{dH;f#_}yhBN)!Ohlr-DoYz%nj$5pS8
zAEhNJ`i3zfFyW!Dx`Y&sbQnJG_iNiSG*!srg}lD|T5#M5T4F?>fRvvslAV(izm@w%
z&pOmMN6kSAz3vI8$fbZo-N@KEEr~v65HWIHQ0O%_`>n#uVl6(y3}s8*WG$(}uKrbL
z_Ow9sdf!v_EZaJkd&Kma!2O=G28|T6?Xb|zkjPHsHJ?eBZ40Gt7+Uwl*?vT3J{Nm0
zf1Dl~W4!>I;lMM`?HHLG254stchf|^H0NT7T{ps3MVfxe7y(g;GFdjYY}+)m7w*=$
zyN98B2#5GC8$Ba=o?s<$dIy&L_kGq5JvHr1c)}IuNnH-&kzG6HhU;H240#pcyV~$P
zWfX{Q*V+Bz-O<LaLv0=GhubrU43LlfdPxZFeFEzuxo{4j+E3vuotoh!h|OH>@KFJU
z^IBl)ZU*l8>GpJUz!sf`yyd*Ij{QZ2yvlARu1JI1ykL#;O^2*{@hno5348|1L+*WW
zvr=0alDF9~HIFMb_|4)$-693mE4!(=dum&gjhol4CvZt9cIT%Rp*|pKY$BgcGpbUS
z#oI#q(f19QDi%Gs*zd2EM(ABZ*F!%O^WtI9@r9=6S-&nv8!`^AuZaJLwL{0FmQt>)
z#mR)cZk@@0$j_1_H*-?xx?z7e>OXtvil-JYG}(#{lNK=F;yYj)<?~QpBJ|JA(P?X;
zl0{;%;+HweB^3!Y|1d4sXL3#E?gba^a&${4gbgT{eIZA&3j1U|@WrWUv!Y^Ut$O~s
zFM|kXV8Rg7zHv{iDfvLx8o;mVT#c<Yuoo{oRuGt;gc$2i3|>p>dZq89fSFtPY+6Y%
z^hvD%4>S|uEfzYg$P>TkgijB%QUws_rLHvbpqpuURpsN!bvq5)x5O@uXV;&)^Pcx|
zQVx6HdRAJ0;ATxe8QcY%IYf2^0IUp|0Od>Eg^wth9_497ZkM8k%~*=tC%2|ATxYY=
zRD9E4%O9<1tkhIk*zY#p!~{<yANu<5lwCacw?l+zcQTvbGToSX5-`aD>dgvntwdhB
zF)g>EC-7?siD#DLB8xg#d;H0WQ1!-%K4hh3EwDLg)HSxXDRi#UW2qW~YZedXD|5A!
z3!%AS*cSuthUQLHQhA1L-bk5PpU&ie=!M1><-tFuzSTbpP8>RjcyE2$P^!D4>Q1B4
zQmj*FiufRY%P%-+Saj&>p(##=ugDlP2!b-Mc099b#CWwCEFB(~*@gjuBRCmO5j?go
z?s+h)0v>={m>@X_#Y)>{g43uf58kqUY7=`A;vAlkHWvsc%h*CHkU|d2GJUX2oOM$x
zCl63hL@vD6bU7|i{7#dHVCnly@bholD$A5Hs}Hqu)yUVrLc)$m=}8bvg3PF}uvr#6
zaTTLn1FTrNCJy=}n=K{Yf4&_dM`(3}g36+{PEN%F6bOg~1%VuO99GSU1{*I-Yx)J6
zFMmpyQ$83X;=@%{?=H+TcdO_g9(}O@irL8i5qVRFD2@^g$_lSkOYi*~sS!uNoP}rt
zhA{w$h;L>Hv7#K*8W1T!WuseP!alBAypTqUKF360>2I;v39#ZN)pJ=3Pb<!Aj|pt=
zBnaN##dWzD%PFDFcJF~8vG=LrxW9CIbsWW9o}5Kx8T8}vaC?R-zKElw@$<tRKh5SQ
z2MXOT_Fhq#hMaIE|7C;>a<zVlBmjdmRT#&!*al|-<=9q0OWlpUubEb5MmD4}`6<7<
z9*CjXGY4@Qhnjw&(p@++B1i23tU+h<8U;S~WG$1n^8S%ggd!`s8Dho`Sc($P*=NHg
zkB6%rPt>X6mmKXbrhCUzVoIBN<@(=th5U`YP<QKW268OB#dO0aWV-ax$mD-d^6VBL
zhhF}adw%G>pm26sx0x|evW`KX>Rv$?zueDN>3}{TQ_E6B{AT76b+y!)6AJLNMCb2n
z%UOwwsKKqO^yQV96c^*HR>m`rbSi>iU>N>9lK5mhdVFoB5GgoiJhW<>^~6D|j`GxK
zv%7;_DDIEJIw1^lW7=SjKAZe{@vmgEzYvI2N6^GE_Qa9Z>ypF!t;EYv?+ZZw&kYx2
z1Ix8<z4aXRoC1I^({o6+p*=9?C|OMWa){p_`BzJgM%*K%3~;1J{G{Ly1GI|tUo4SO
zC<CvM(8$Zh%l%8~!iTf=7uZYephn^czlhcaG<Ig2?3~5he<a1sSoBps3Lc+B3tQjn
z=#=sp7O5jNw^Jv^m(LBgeT)12&zviV`6OqGU|Y7>%Ht!i#yy`uCB8!0P4N`LfpuuM
zd8R}SaO!w%IPYL%6M;x+KnwT<5KcS6H)pGsZNJY^<mf&xVCDM2bQ1IcC%wX5T7An=
z6qKzQdNz$elrFWH1X1sj=YKW3bLs6e;W2j4&IS$EUF^$+l#MWDaQp79(pkFn+OkTX
zl@M)O*~Wf2E&d-5c}{E?rsR*VF?jDxwdq&2+}8#Nu2+a|SA<U$Mn8aM2j=s$A0z6-
zB@@l+t)S{_=?i^02`LaD)2Uf#SYS%e6u+Nw(x|U*Y-&?#^@y2eVPuOYwY0H8w!vC+
z$&qGb%fe*VtkvN%IJo}lB}zkE!HC1)NXpDw^o(fnbY6c!60+BO>O*P_QhU~x|4uZH
zYr^WiUYZtQR!U%LYCf(!Lka|0RmLP(@=#jW%2Xh8AX51Hl2)vHswye7;xp*aIdW(v
z(DP-K$)bh}srv}KZbZ;GlOJK}3m0a_(bC0HvMp@b5lI=}LdEL?wdb0>slR}OXsVD2
zZ;q)OdN)3s*WFG{|6C2}zeiuW<=b<vVyNxgxu$)#yP9Ho?r;NGrHu>yP8yLoUiC}u
zZ2zwFJ?oXZ!PD>z2u``m00(UyD3W3S{6|Bn^KIeF;%iZft0mLtC-B-kOuo_ux2%|D
zVv;7KJ3Ft+v8g_srp-IRME<q=JtP|l`#*!ILm%ycT#J!=zke-vRg|AkZ<0BD#3~kG
zB~^}vu@-k~>4Dz{luK^gM!l6Y4$M)Wr+Te>L;FiNWN0yvWhn0+_}d%$o}a1n&MMj1
zJwSqf3ZwqnN^nMUkmpOwgQNSVNiIWpf(MNEG~TFG;`jJlHKFgvGbh;b`R>qhcx;0w
z0KQmcYGaJhZmr9S|2f_}F7%Fh+N^)5qMZ5x-Q?96Cz|6~r}j}D#cvCWhm7O9=e9KI
zx54%^hp*(Ny5Pyjtr-r^DG+1kR@_mYM7BZBzIU77E_*LzTl30qr-npZJ|ob~G;+v<
zsGo3#$hj*mg~QX3UE3pn*W?YX&rx8=*2er5oY?ZL+H|z3<)?_Jy|?)+%|m|4`ku|!
zC}JLJ@@I!SQ3W+_=7fHkDFZBO{Q3EL=o$uM>rl_PZCR0{%<o>4>p3GDyI&%LC=Wnk
zlA2QISgpo+lVlZVe-`=g|HyN7^!be?01X)%KhG?`Q*4><tH3%eT-3-AQt&&GRt)sb
z02^%lu{`0LOBV6APL7=}spt^a>@a0k$BOM-aXwzNy|@nSB7u57_cI%b_=nT7L|)Ei
zImx=DYAL6XasMi;WIw)WxtrqGUw_d0gA~7jyj)ihC2N_5#@Oh$R-&x)-Mn)^PBR3F
zk$(D%nCDOrdS61W5-|>154n!v+b3;BQWWur)$+#)i|3DO#`%tR<x1zsCBgYHtJu2^
zku4p6q`wIll#=4MePxB{-b_N)Bm?*{yxNB6)Bew`&7TS)`m^Q&b?ozUh6qo7YThHe
zMiT}$d$BT(`_CBorQMf|{8CURhEM2G`{?!3;ImoCG{8QkD0X0fg6#D&mm~Uqe`ZR+
zDQo&Q;dXPq;w8rIempQ3QZf;XO(<Dz?2t5K6SJKuFf{|2Xc6}`{jwz$62zc&gIv-%
zdh!gZqg+q`uy)+eI9>ACg3AX7JSIzBe|S2i4(jJ*fU^pJ`-rH*`CVuRa+1!cg17kX
z8THkKd~i(!1~6;Ye*_+48mvDMcOP>pP>3Go-o;YG=aAus7G|YV`whK;eWiTzV__V+
zwcojP6M_|fI}4Is$J#n6<l8a)jBHKOsgylAW-=4A^$g6>t=248A9y(F>4$Zj6iU1R
zGyq9y1R1)x=8x6dDmzFH2(d-}WOcu&|0u^}(#dQ2K{51z1ZZ|KP82dRycle!*i2J%
zly6kO5t?ZkW$SP)b^%w^nJo=bbWn?__{pQ>qw%y&?|<WtUT}XWs9{|9^OmL!y0J9H
zF*3`8yDk19C5Vd9VBXxT`v}wI1_7A$T7}f1Rr5SUzvO?DztD<p*BPx=gk`t_=OVl-
z$`OUWcl9SzoTM?@=!p$+RdGyDUp&rh0!I?e>j&rN@KqGTDvj$uH%Fa0%6G0YktNr>
z2WL+7aIk050*I79X(don+#HutG5nmMd0ekQ`$Jzsi}?+Z%uV%kpcI1KHK58pkCSw5
z^dyg8={}b(60P?>;!6@1-F;-4!;=@yNBo~f*v^7T2LNPcN|C=;s4Bad;h&N|fs*}F
zr2SMjjOFZ;9E4GUw@e^rWO-n@RuADrBmK$w(8CBT(Ss9Ng<}Doo!YEvL-&H6Ltsg?
zJwJLs@yAt^$(1fK9>qidG3CiyOkz9HQ}q2ze*w~fA_*6B{{nr7)Zt;f@kVR+9<_r_
zBJ+R)qbOC!cCm`QswGF-Oa`v38KZ&t1xAhUoXzdNVZaCms>aSv)$HV@l@<6-{CgY|
zv7oDomxK&BSK|*i@~9#Nf7cXK$L0hK{yQs~&F=V;DEc?SIp+f}KZe`{`Z{<`d@Y#y
zC3?;SgK=)!@nrE(fFohe92{KD&7H|@Hc@({8OW+|3TKY{xb|Y5{`B*m8+wKE%rieV
zLVeb%YRAQqroMvu4ZG0xky0tk!wZwz@0_$n^l6q!D9m|PpkfdnEHa3(+T%^7J;m10
z4*#7;wooD0Wgg5?^2Z}n?@OhHwRa_?gTLCnZOHF;r8g<TA{&a#kej%ChK}nmQtS~*
z%5~d|i5w$i@u?+rss0XhYR0&~{<^7o;`z}bQ!8*3|1A#IsIqr<G6|m<3Yi1$7E?wf
z(4LV0dZwoBQ!{5n&-R2~5Lk6kQUY&WY?2uhQ~><y<o4BnZawPcg@Jrm2#mK-d^pWJ
zVag+o5_GP@?a4bv(UXsE_ok8DL1(Q0AvfAxoFLnHQ+t$<h(=UeXr8?gst?_w42k*I
zVF&IC;^+`}eNBw{ra*hFJPprU;GaV2f+Sl326f2fG`Cg?IiO2$Hk5mMU@KQ`l@%e|
z!7li{44c%|*ghTTOnSXwBhcs_;qGJs_AC?q6oaxuYWG@uI0{$>BUYoc>k1+y61r)n
z2c*p$GKzkSW^$X9O_;KJ!Kcc+zy{iR(o#jF!TYuL!zb02v{W90)!I#~j>7zYlX+>)
z<+X!&du0(4`Lssud-AX8XBXcXBr(9F(U1nFPPL#p`D?AR4<p&*(^24J$60?tZtZkw
zucoMmp^?O(;|vqoE#JiYI!>faYSsc#CT83=uqesU*bJ2LxJeoX(l%vkjv3S_C^}Nd
z!NnDJ#QwcRO^}16QS?168QBN7-T&&)V(+md)=foS!Qc*EPuEL3k1|!QIqtwg@6PH^
zwXHlUN2@VGrOKobv6Fy-<IN2z_Oh3lSw$2rrr&8gvyqrLxoQlitU@{7!iPvB$#oHQ
z1zB@87N!rkYPZ|2_fk(J7CR)*o8JI*6>)e`D2PFS=Slotjsk!EO+-;S+XMS{1nlB)
zcbx*<2{pb?L6gicCchOpGC71!`S=F#lz`O3X63kgA2<b1%*!@bUNg1M_epna|L6~_
zJHO%yDF2-_ws))YQPM}18gtq-i7o^rdvvIEm@<(uW1-UrM1q{!4;9Ch$E+Nrdsh@x
z#^$HRp~vP0EA1{G7m+u9k2RWBo}H*AS9g)6ei*}ebrnUm<5A%r+0yst(v&%@!$d$k
z*t8^M?^&HKi>ckUCuOiyR#k-M>^FfXNKXkU&(I;1cXKr8>ulYy++5D*S_~XeueCfL
zk>N(EE&g&gGsxU=ACj1I$!JKAdXe}R(hal`;0x>L*<Nnt<T4eFXBb!Z2@NyJ*t0bM
zbN*o1aE`Q@Atbi8Q*F7PBlS>6^&ivQ1YPjuLE9lEty%B6ilh=V$w6;lTU7E^s9@~f
z5Ag!(`&R*8qEwJ_-elt<<2u<6<I0W0=Jxi;#{5yc7MvBV47|aLVRnm$#mAkJ2QUje
zp6F@muf6p=ovlI(^g|S`E7@Q>BGl%6=Lpc?XzmX_McTKK{X8eeC)Ss7-QC~zD{u$!
z&=Q2b^uDT(FR;A^*LmJ76^WC4Zl+?=q-N612F%(lVALo;@A538?c@&YHj{`%7>S7f
zozm7%#{&1OC$kbm>yDr+K~3=s+5`kE^`7sg4&x^S_L5c;ohp$t&9vs#dx+f*y<ILJ
zm$7m*Rh0R!eNN5~d!M;~`0uhUz`2@w15@uh7>P+pzAStLl#y0DX$XhC#n|--gYkZf
zg7v@uxv?%&9sAn1d8e}LPYKBYbTDRf6U5wgPS|22eDX?a%jdyL<C9j-ny8?BozTNt
zHbj<s>V57#g;MlRZ)<k-p`Y|d|K4B(CTLX4<=baY@IrSo;?I4jR;<Ib5CO@aNO-ow
zS=P-DI}&$z^b=SwxA2ghg2ncIZ<L5QnGLs-W5iSm?{MLS18=d~65-wO@2C25I{wHZ
zU<b7krlj04)f1#jYwlglolJH1-_!JVz8k=JE!%@!jljsw(!t|vD)NOeh3=$Zyyzwo
zQ0e9dkL;_fSgmma71$nubyU$R7$m*R=w`#$<n?GaepH5T3Z>1Y1xV=W7)M|ov<Q)6
z`=klVf7Zsae$B?cn)Ms@?sikYEpR^>!%z!`h%Y<m46-|RKK)Omgj)ZZ0GYC<B5D(V
zS7WWhXKhzER-S6U6VFWba$eG_0_W*Bp4h!{Ilr|b@!XW6l{Z&;!zB4`xK?jtGheqo
zLr#)=#1L_A3(K_gQSf&rhdQ2yR-Jx)Ijw_^K&40D))u9#{cD|U@8m=T8qZWgET2td
zNpQpY#9-&r16j&oKDd1F68-qIyE92Q84JWFCi=t~G1Hvb{=Hi>|F!KHz;&)KtoD4m
zjN!;Ce)u?AauCNHmE#|)qFs|oR*?C_R6(V`|Ctj9*GK7NtgSOfBz^>*yXn>$b4lfF
z?fl_Qy3X9;jgg^gF;A|pR-rF|eY10?Lcvjr>0n!-H>aU<?FP(9pP2;wFj+H?ZGQU;
zuAGr%aewD(-TG&FA<%0%I6GV1ueen(H>{1!IHXPYSPYWSI=%IAwF7#*hg~?Oq=;Or
z!8)A4LDtZ`FtGp1BbAungq?kvcwD9Xja5D(Oqpy?fbfw+l^twqYHH~jn7Fv%mo-)Q
zG2kh)s2lI%4Wq&~9z@v3r_JvutNwI<dz#KJYP?zTq`r#Triw|I5)(rPlr;6W3`c64
zKSec4`8Aro11R@y$W@bZwJB-KCV;{t_8u{(#UI4oYa*~@=Q!xqbyXPajX0TD{p=^e
z+>u?CmxDJK+wzpLj}9-psqPock~tfar(lhwG~{Bxdyb=S-lyld-36wT?$%*+gX7bs
zX^D3oJ7nTL6uBgd%)S6n8qC_kE0g*kxFXuzH^+DVRx~hB=)m`SR>AG>j)+RxPX8M?
z*=68s9mq(PqQ%CLfJ96T{{jp0)N5X{6(HyRm%5J|+wy>1JNt!Yt6|MSz?mTCl`a{L
z{Abn+DX0%koDR9J&b1KV6tXtsu^|QrJ9OkC0K=PgB{XIk!JO2V-+=Ut?xv9VxQ=0d
zta(YLF~>u49Ce`aINMg4BDLIT7%hM-b>!FKR1p4zR~t91zWt<VjD@3kfahK^dv8cv
zq7L8=wKi4xGyYs6+>dS`NS2KMp>p8wu<FOCDOF9_R^jz<z}q4$y`jsz(c!o>Fe<8`
zs#Ckq(#D3J?*nzgji6&Iq(w{)T5MEiT&Gi`*VWj$(1L6H@83VS?Z^Vh3?3m1<^;mG
zMmEj-rACdW-REo+z9(k~+|T}BTm5VClyXeK7_P<=1<u6N-x`-ArK`31qGnRn#dSVt
zDt0|38|=6E?iY!Uj?BvdN!9oUL=}NKqfJZkv~wk_vevD60t4#e<S(R#qG7Xcli%u1
z`}5=U!3d5%Z8!eAxF{qv;)CBdx+aI;RQqjhDgtC<n*?2t#|m!F6SDd~p9f8`N$EmL
zGG-a@I3??W_=o3r`-Jhm!xgMT_1z;!CMTa5Kawxx@*h?kiZsgB@y0GhKsOOhZd@Au
zANWpsuirw={R#=HX&4f|-ZC95X>5J&q;d<fz#inuA)sQ+O@R8mcXX+w9hh_4kQI3|
zE(5e2OOu^BZ294Pf-?a=6wxa1G1P>L(z3D312;*nlBUcEo=L~+bEqZzp)pk}$M`_t
zc$gY9mcmD361buHH!ZF=F%tdB;FU}CVDhGH8U-J^DuH(r;^?j_qCvw8{xV>$m`P<S
z%Iq`0q0M!s<N)bb9oZU2H?#vRiUMEZ=C%RTI%QT4f>=9QBAtY;QbRngl&@G1819tw
zjsj$<il#jts5MB$1?o-iF>hn+%ZDBd7i~a<l{!lK-eaE--D}cmBn3%ecVG?#M?}jC
zU~X5?HUc{eX1vyV;+t<zML9T!I)seQe^u)NQsMMigjWzgoa6}(G}TY`w1hYtsByrM
zw(sY8X)G#68lRE0Hb>CCzkPetb$U+r{72PSn~shU@vu7kiypGRK~9IKLuBL=2!G;O
zKPT|U$5G&$WX6=-4q>3xA91HJyq+@I-=CXDrAT)EtOG`h3TXZ`;pqrR@Kot>X(z~j
zvQ=v~u39prRmZ;Ye6!l^OI4KIC|ExG1sDS|`WmmwSTVyemgb|XRi|fUV!XYaRgjdy
zo2(#BN_u^Lf;SxTfzeu69N`-t-}OD&`~ByLcyr`Od%|~k`V?whr=0g}3=y;maY*4<
z(hx)#BPI1JNI7`1PSB{wNofzBZD=SVLmPBQKOf_sul@CwY=Dr_@_>NhkfFF7wW1w}
zcN1Qdq0-dAGvaC<f?R~EME2Ph!-^`|zaRHyG|{pAZu}%T3`!|f@9OS1vpW>e>Z#Wd
z^KR)UPj1m)=ai;aS-<qJX_l?KPwl;9e!pa@<rD_FvQ(&KAkW84PIL6?_@YM=KdDCM
zw%A198)lSVAzi$A<L@-V+xug3=}x50r_KNHXmM$MxoVB9u{o!q%al<Sd)o?feEfDj
z^l+u?@~-4`cPfa1atEufi27Vd#p^gwKS=)^-vg+mQpp#uT%%9HLgvTY{wA)RFPx43
z9Ze<2n>VWBH;>HmDt3CLyws_lx@R@8aZz;0R`-Q8Ydv_*IMjkuy(?viImXKHT3W$x
zqes?;C2h5He#`PDZTW!D+-FNHyDks<mFxZ~A?C-{+SvxPrdnf<)o3CCS?bT#jx%bL
zf_6vfv~CKKY9bLaT;uFgH#MCx*tky67)+jwrnE`W!}6M#X{=mbX@Cp|X-|<70o`2g
zB1{PaLMPqI-Y(J8LC@1==R>P|l3$61W~YcNed?N5JLh&GGdO966s3AflsRNL=!Wy~
zzn#6Qjhv!IRL<)&`fMWj0<8f>I;|SjYxG(@lPR7ote!l18)9I*rhUxkZPwloAIWeC
zzAli5W8%><cm~gTg>T%|wg2#JmaBie#T??S)U79H1b@;wnw{$`q00!I)fhis&-j&r
zV}0(xwTSs}qq3KbiP8H|X|XxpX|dg(c*wy$Y12~W-Jn@%UTIpFUcqW3QL10hLy5KI
zx@Hdt8|0Qq4R^#A?`}nEmPh3KYMz?wKHo>R_iIOMtnKa-Fl|dogl~zh;PsSC&D)69
zeI(WliV&diIM_k$Ij^*ki9<=+T!)wy2ZE&ISeJA&i;FIbll5=!wnrIUsXJ+!$Y^K|
z#GVFTPE1&|Q4h5E=*Za2@0K6L@AkJ>dphJg^!BzY=esT-oNetmBI5h}i_U6)cbH0s
zjb>JcA$pFilBO|{LG;p})yExItprHjp2HJ4I0a^$#;<U{&m5(Sbc$=jEpz%U<x8#i
z#xEE|8MQt+R~)p*7&`3R$BrslW!h#s{_N(edoocvXpX$=@Xw#ZUo(%uq|-N`xV)!w
zS^WXs*7Of9Nb4wFBK*FEwCxoFpyULp9;j;4EfU({v>+Eb@ex>_g*bC1rq3Z^5Uk0#
z6zIe$fUoZ5Lv<#F->dkm7kTl^?e8W@?raagT;le~*9j^bo=eN(dBS0cC%vIX1kCzh
z^oY2F@9AFdEqVqWf5Tx@>eL~(g(tG(OGf;I#y=^4pvq{7bOtHiIgd#=(ACX)iJ|Y$
zf|8?Ap~WI{*(qc`%d1Pvdyn_5)_J_-c1rAqCgQv5C0PRur6Yebt!6|j^f>2_ugQ+6
zKkX;EyOyxv54lfnJ@T=$tGBheMrsL=zxTNjbZW;nGW@zTY0|R}6MKI6y45ntRe7pQ
z^N2pGI)zlhFt{@1Cx);)756^LP0+!76fwM=iT4gmJIf?^)HLufCYT7+$VcLr*yS8P
z$-j}>y%Sbab-p!`+7Nv6It_<khE*c~uKhmM*z|MAP1y26gTjilE)ucb<zv&2U*br6
zc3c8nDh=<@{jBt86TG>{&2;`$Sp7a~YIiSC;Z(p>u3M+YDPL8PZEkDA0_0CI`JR|L
z2>iu=$Gh-Bo-l?%S*6kxHRqF^4Wt}&;dxQZ_1)tr(uBcCrTjY}BB^_@Q|~a{HR*#Z
zz3DObm##nLUBqAZN7?a(Kq&f$L2dO?16_8ON|e$29Pfrf8l*V5E@iE|ZoQiMA8cLp
zx_Ejg$H(ST?mRPW^BERUtVA4Q5-3>H$C0~aZFYyAl$U!9qQYO%t^puYWIeb!5+6kC
zIBAe>r1XhYZ-4(9vihTAYV69cfz5%KGZ*VyaNhE6Q7z1PM)J&SV1-8pF~Lj{e>yz{
z{S~~E3w`fS2J-B;i8MArY;etK0UWZ%8pV>>8hBCWQmY@R_{fv@Tg3Dxh%7wn^$xBA
zoKKuV6*j+4(*DCV=C}rLxBG47kr<hpDmvm87I|lQbY_}+eoG#-C+nBzKXjv;sP2wJ
z(Z+tS({nZ@T=+4=mMUft4_X0WWqF4rujc8J&H^B<rvK227M{ektX_wm$8UJxrt(KE
z3`-S7e>F$`s5}NiicObj_a<!UdPZ=2uT`$r+u8Nz_wV)1bOr1z0!rBksOkmOIBMd&
z_3FNlhr3&qSInsf!5fnqsdkEo$!6xn^J!V^@!Q5N_9&!Kq(fQ<N2Jjw!MIL=GH2iG
zXBa(2HP*dZ#jl=;V3alJYqh;5;~>Fh8j!BiDl^-Bu!b2_9zCI4b#S*6tc_f-%NL$T
zd#Z2`BURXz9CPb)2)V6q==P6HB%#?}*|?S1Ip$IU7^BA;ZZcsp{<#XAqcE~TrX&Z;
zpy6p$-3;rt#ih2ex0qz}D0%X1KK=MvM19ij87tHqoG8Oh3nT_JMutGHVTkbQl0t-J
zWsyxekO!KpagS)gXvSf}s~&|GlR!O{Lll|Oy<<cxJ-e%4WFKUP>I!IhD6eZ^d+EyS
z8nq*rCg_eCe57srmz@G6Ib=zYf5%i!?V({<vMsz`om4nDZHDG9z!*e`?PUN}Tify3
zX?f3O_$a)`m{-KBalRv2bb=>udM<*8Bd&T3QPA<`HrnF$eb3VsyZA*PUF+Yp=Y}kk
zzTff1a+6v!Ix1U^f6;{(K8je>lUUEy@l$cUoblhaukYi30TpNKv%I(FnKA8dQf2)y
zU{b$UWn5Ek8`qq$((W^}j6X8(1k}|s*k`Ry=4n>v8(w~x=5m)UKgxZI#2(K|Q}E6k
z-hJES{v;~aX;$q2=>F9`9@;Na>#jHM9U$lW=#=?6^tZhLxuvu02mY^~7niX%*El%8
z*E%MI8<P}lmOrbvcPiFV>(;%OVY&8sb($_YPnW42K=fI}qH;W9612;FP^1<{`jIKN
zr{GQ;BHPRV=Srx)xz$0qlRW(|Xqj-isH@OCtmtuFs36(2M{8B}1r8ID$Jv*96kknc
zMRxjh<{A3ZZfImwu^g9<NeJq0+p80*KRu0vW6+OPO9qEO)p4!4j3<L{{O%7_zp$XY
z_{Vz?Pd<RJ`sSCM7Qts^D9;iHxS~l4R24~vX6BIJB6C(HYg1xF+LNY1F%0u*{eK^o
z`HJoxqzH0<HTT9o5m$Psh4NU#OVz)=-PSJP2@_5+GB&hwZybPKn?+svkP5k)4et};
zp@qOlti7%LOik-UdxKJ#o@NN)>{#pNs-ICwt?<U{V+nA`<N*}@D6fP2Jm*u^dC+do
zteg)@s4z8K_Ka?Kv>|l5TrbQJsjb!M_s<zGmv0`+U(Ucvc;)g*w0|y_{59%|tv$<!
z&s8FYdc0%VPd4_VNvJ`X)kwZJQw(EQ^So6L_jnZLK}SV10AwbL({-^8+ij?w=n?ju
zo~QK@Wtm5zRP!tNuy*jw&<>5uObF;6c>R&)L#s-w@zwj>&D?cMX@>^I{3LhTrq+3T
zfnSBf9V)H*@Z&uk*Pz40$9>I5yV^M50JDy8WNV&-!n^d6yBdP%pXw^nMCNcJ6p$>N
zN+sJUKDe6N8h%RcuD1!fT1+3uM@=B8=AJP%q*F(~698_k1yxL@v_sCMbOV7qs}&;K
zsgi`UC@3;XncBy(`DJ=ZKpt|#@4;@VvE0QsiyTzYT~SZn%=M}CK*L&zckeWQV93be
zJdQ--6ch*@+yr{z_;w8^WiV`GF<zwn5~$pI4EX5c;sM_?dqr)@_HUVmIf`Imtsk;(
zHtsReI|GCEVxN)1zgpYzj)*K?z?+ESg|7KL5Y7BqG_dz~ZSJ2o-E9&RF+oqNy$PN+
znPWF77}~Z<BgkK`&<4X7r#~Ept`7w*lSUP1FG;PUKDlp~PldGFB;gry=&%qX4%!YG
z^j;H}f0xex;37+InfwPf#cA%hsXQ>_8|W7am6hVP%M2?-EXZ5T(XAl1(8n5#%V5@Q
z^zW48qn<nT{=KxsJfTB=#iXhVE;JXLTDrh(>}-!lP|-$ZMgEEGD*SCe^Qrse@oIPf
zmwD^Zld!rv>dvC1y+T>zca?;o@iHL*b+VE!>7{~ccR=2B(Cp`mp%04L#G>h4uZ-`L
zOx0VX{B*KpLURF1K)S>Bv(}HGUF4qgD}x`t)u|eM8Um$83twh-aq9+u^yv->4v8i!
z^Tj-IdDszEq9~}*$g?`H_iYLV9b!hRKdh0Qm)CYslz)smRh6Yh?sL02mWYv<f^bH<
zaax!YBU0S>9QefQvF(bB2^k(TQFSWri3fy#Dyh<HXwlJtSYm(-O$)1ifpbS~U*3al
zl_Be|Q#>g~rXpj*^Kc`RS4|Mk&n$a-ewc@pk=isx##;>vhi-lSFtnpVvw$Z&u?LCk
z6`Q}w<YkyFIkMa#`uIgo0o~Mt{c)YXE*CR{)1b<B^RYQX-h;UbZ)=Ek3rL-y<+_qT
zN9xV16M3HPZane5Y)?PY*)1qa5nEqTSCVFWINAhVvk?BaHqE3_Bt+%qYyX-T77W7M
zDe~)Rys6dP9Ko*RbnTv{!-VuKDJVtILKqToxm?eo1qDG51qLiR<R~&&E<;BqVih%v
zDb;34PWjNVyMt!Ku+O-W{xazdGS!{2P}%$`O!mYH=j4&qpWXeh8$hibX6XRVn6;E{
zof|@8Qcv$|ShEwSE}zPkXV!zm$&&D2!|c*p&!Af^St|bWk>J`fS8G80hwhXytoD=D
ziUkqd5H49IHx2kd4!~wL!hToa26^HbCxM;|=Fwo}tykgMdZ188tB8GYPYON#TbDU?
z3=uyP>xeD*-kt}j$JK*v?|Hd!Do@Q}6!Ec^=^#Y0hA`0Q%rb6KGDnNA)YSWNzr>%|
zhA-s8qDMXZXMDDgJFmFTXC^6*yc%k?>qpao?a>hhHrK!=_Xp9ym3E$`I4|)_>i{Q5
zs(+SEpzX7`vxZmfYail*v8hvgC8*PLC{k`(=Xt%wc#VsDyVu?Wy<V5BgT_7g@$?E*
zRmh1Jan%QJvj;sVN-3LuEkDOmv~UsPr*?RrY308r9o|yL$nqfKq+Iww#meyHLQ=%U
zf`|WwPc|t|1#i0hl+N=_;=dC+$B#<$nI8Usr}skJHsrD*cnbe&y$Ext|1SU~LE65+
z`=D0KG1;J;Gqrd*P~Hq>o#OKQclf~{91%xxu071V-nG{rBc{!e7y-hcn-*Vb5P8SD
zQ2yWayZc;z=bN-!d(15?pwOH@v&j0XRUW!@jx0&&_68sod%HV0=SasR&Ye9)lCB`6
zqTlPVyL-bkeyJuMjd=5oud;sO5gvZzMTBvlRdt0;mCOOQA<GQfWCVfm;M6)ymXP5P
zBBIuuV>BLPO+sh?3fc_$=HWf8@eF*qBnYHpVPT1WzwamisG`#1ORm~lGNUoh=5|ye
zDhTgenf)VzI!k6#?_ebqN`zPgd;2XSQKO_v;CxmlM8NJ&izLa2OVJF-5HQo}G4WiQ
z0dxWZxyGT$Vg!m(Tp^4?v`INQ++}BXhv8t%x%C>=a*ZGgK?Yp7_$1|Wl`tq#sm(D=
z`ebQJC<UWIkAuA(di@sHZ(ZY?uf9(5-Zk2XJ&X|yhaIAlV{R@c2q26i>WvbIt!=E$
z2$b+y^o-KdIdIlYjFpA-+7atKN|(xm72hePsn+WJjlc1K<m%P;`O>fdGIzIb5}#I-
za9AlZ29m^h{f6_eKv9bcaw`qTjF3}El=J}D6?&E;Bl@`Egl8`Z5fTF7+s1eHTD0~Z
zv1$^yD1Uh+-A+R5u+Q4+3Vz-%liv&d%n?v0gm=bwOe`Y251!9ZmzUPqzFh(goz4O6
zc8}3;#9TdQVQzu>rDfJmp5gqZ$GmZ~Q`8&t^hRBh(GUrAJBQr6bBk7Im#a6fvAuPT
z;lPk2eXMZ^H^jP(YAqx#1q8}_vfa6J11BvA<$d;yMW`H{ThEqZ^QHi6u{ct#u|g0_
z&$w7?8H|Rc+VGjrf0oPdzQya8H#ybRtS$wl(xEdy@+j3lfiVt*v83ZMRO<vPACctp
z#uH<)vY4>|reF(anoT}d2!JuhN*oI73@R7ux7vq%?Q3sv>*@eZJ8#ZcY;PaZYW1ks
ztKSpMPd4|0p~)9~lDX$hWw~6UIoBW^b?A25RLfQFZES)RRH_ZmUwo9ar!R5q))lOE
z-X=mShNA&#l2Qt*bUGupb~b6Zc0F6S*JH0erdF>J1_7lwXS$q4XBmeF9mc6(XM2lg
za|L1Z8Kapwl1&Q)#%R(sB}+Z)77$M42F^Joj;ohH;P3tSKS$@_4lh1-f>`y5!VIN6
zXtp_s7x_GBXFQ0nR%(8_MDdOQBA*na@-;68z<)m@NfJt>m`}cZmf!yM>jWaA9EL0{
z*9k+Y)vAogiPzF@ce%H*L%mV<8p6EkKqznLFny?IG_+z`xhS4kD;)Q3Uxjo?7zWfE
za|F2tc|0DlzPf}%(yY(1xpfPiCQG4GTErNK(T*rAp>@j6_7*Y-SzTP=y|=F*gk`P~
zlj#B7e(JsTvMxFuQYtrmQm!1%`Z8OV%S40_gkgvka-x|v`DalwAdI|DRzwOXG~MnN
zX*#0c-37eKaTJBDoqT{m`OG-8hH<JXR|Aap6L!2d+S`A4F@+)!{@woL0CiCu@Y8?!
z6Lby)U-`W@fsBw+aq+>EwD%y>2~nuHbL)V^c8AsFWlzjF=l$GtnlHP8ycz@KM2Ba=
z{pZSopLO)Q`@Hk|SJ>LThB1cmu+OCjF7N}Ne1Rkx&^m0<Xfy!`!;moaNkf{ZBuOz(
zSP(=pNWow@;`;R)%q=W1PP%;M_kWWIAAT-xFv=^3*5RZ-XQdMPBq)P$3QVD4c4#a?
z6z3i1-pVReg47C3Ht<$t3QDnHdG!QQse}-aqzRQu4d<rbdB!<JpuDY)1R2JJrSg4S
z=a0zW3#`>h;oX;o^qwxx862AY(F6?Tis1B_6@dKE(-=ekb$Ny+rrm?HRJ`{&@(6y?
z38pi4clJ3v=uj?~5F%qVObEk>rKMHI<CMMqF7xwCBpMDkw<wiLtOX^)xZ=IG@Goj}
z-lD=X7lH~=rOw>^B6m0Mac85&2R9FxTZkzo8A0Y5!iD*H=H})JmBi`{tU=-?%@885
z8x>hq7=(WMg)I(Sht4!oYqU0$;tG#F_5=?;c!@i=Z}a<i?l2e#0^x8nLuZcRV1&tq
zwz<M#(p~G-<g+DAK`xia*r{vw{Rk-oveb{R@ob>?s<Sr1N{zF6W*m<t2qa08OsH&6
zJ^OJx9~Q>TBL|oFC=|ly?j9@~27?iU!H~EV(jW9WI5<FSL!&WA7)0zJv=DSDm+J%p
zG@5nFjb%!6%jh&Akb<T0kW;5FG8*=I^od;_dh~t1`t@&c^~P;F-IVQ}l!YZf8`zB!
zs<nU<Czb&1J(jFRIFpZZ%0ENSrWIwRD2&jy@MR+(t20exKJm=6eDRB4<h9qn$=~~L
z{%a1}x0tINf>7pK9Ls1p&fhX78Ez1Sxo#3T(rr0o0c7tZpv%LrvBqfsd@C8C2pNnz
zy!+m5Vo@i^4T^<Q4ErhjyB$uipZqt1c`2txrJkY9!F=%<tRs#qEX*&lm-a~0AsZWa
z*xTPD8;@C<U!Yc>V{Uzcwe@pIsc_oSXf83%#ta5sln@Mh9q!$|MXSBT)f?B@-MP$Y
zILLuKI5WVZsntV_HH0dloAlY<zQ?6Up7fOhQ<RN!3nS+{oC|<&a?!QTGNRl78X(ar
ztxlIuf9^B9^VaLUck4PUr7@>g6+$_T@tW8q1DsF(u$C+tBk_q_QLY3$h!@g(6H4#d
zHU;tbZOM<st($xNH~+6+rCcuKgrAaPvWRbd{W4EI@i>7JJpR;K?q2I~aInv<8=E}x
z@VT6PR~W$M?3ULl>3oD<Zr3uyDhnYH&fnu+zeV?8gH!8gSYABCsNdtp^&2#64OZ$A
zoo*jNz*1v{%w$NZNCq0|G{a8IkF6}n^hXK3!GP}JU6g`2E^+$QI@S&74|_BlHPURv
z?b|nb^zoHkV&PeLmv?e{AJU`j_e3g`zn>YPb8~I$0`D2ABBY(w%X%MCKV`rW$0e^T
zlM*W><RyLSnCyqeeTt&UH`o0UN3~kwQ!hQn|M1`aE`l;j_-@}5r^|f)ix&|{Uis>E
zZeHEy_N{G}7ngu3P@jb0JNEh`2}v&1Fyj&Lyzx~!?JfGf9gMcjHyU&fcSw@~2m5;r
zhFy}yWx_Bd8Fg7&UZz~BP^nZgkwZlx2kjR3wl-*WdR)J?PnN;6pI9Rq!n<$1!RGC2
zoWJx$uIZC`x!89#TRUYHjn+&-mY3g@f)FVJYPBX|SY~m4h5oopTq@CNA0ic0DkWZc
z@g;(|LJ&rbM*Gxi&HSVKq)%ImAn>NlDhH^+$Jf6X@eyag`U8Y8z#&k=3n(qxd&%HS
zL$#XVi$C%lpL*$WoU>%w*VzTi&ny%o*QSu`1&UwA2lv_(oXfjKt#=;}%k1oQfG*`y
zg?g=t6l0c`SBRsSdToyV!w&7vAyEL$N{LOKvV7tsaa5w)YtcE_=iu-Fp<+t)8fVU5
zq}T1!8y=84xP1MP{evORMj4Z3BnLe*-Q%&Re+Z{7S(fBfyuc!{MO|J@MbyqWZZjDJ
z;(UpXG(!Lv&Obn^Q|@ivMn#HD1Pn%&dd1K8w9aBu&1g8J5N=wRdmH7g;!-dfxfKT9
zR|{t`f}FX#KVzem3`yKD2V35AP0G+`G0uAkc~EO<%*FiBk3P-Glgnh8?-;YzA%#K&
zzN{>yof!8PlKDbK=lzNBeqcEmN)yBWeg_;xQ3)wysufsSUIP@3#uAO@B3YKu>$TY4
zyw0HCV}5Dbi;~A<hW#$PyE`PQ#!AWD!V>4toF*<gZf~}^^R2riR&(Z5h!WuJ7$-7b
zSUZD+-0k0Ibq47{p2$1)@->%$bTVgA{DGU9j4aIvW6Q$)q930VhPWuEGZLgDuM^eA
zA9BOd=xDPI&YN0GftWGA_uyW5ji1eVgPFwW<A-<q*IHw2Iw`FS8R9fpn_-2<tD}7q
z7sZM!9|5GbFEKeQCRP};G5cJKyy=I0f(U@lEbUeoooPmlNz*=~(HNyd=H?dY^#%+E
zDWy`CKt+^FCBjORa<ojfnGh;THXgFFdX~|!&x4Qbap~d9eC_Ms<m$~^92}0=-pyEA
ziAgh`j8-c()=n(?q}3U$_MP>X86ck<g*xXO7GgDJAgrA+12>NKQ)l?!{b&C<TU+<|
zZ~yE6ijB*!ve>jlvB*aRxe18(WvHp6zAw4@eFHuO;`7SI{RZ(Ljx;swY-b3e!5Wkb
zQ8MK4@Q^oOyUCm1zK7KT+0fu@!0nsc=*$vDa-x$huF+Q;+!PSyKcnc3&OapJ2tvnb
z)S=fspjkh|z{1wfA*-wFtgk;nxN(>L{T-^+CgUtOryUJQ2P2Yfz+r2ba-%_#S$6gh
z*xNfGGrnSS=k6UURY|Q@A&dg{_FCM%cZ*98KZ`FC<w;DgA9gc#RXWWuGmU*k5|4HC
zT0MT@pZ*3n-r3<353Vy#I)I^Q3M*V3imxe^;v>G2KCw|4J5AhYb>)9N<yVN~h_mMx
z*}gj@kR__+5+_d0^ZX|t<KEqEw(fS(spb85Zu0nJ7xKouX;VbOz)w5g^R+J$IP&w|
zS$4PX(b@OYUji9WsfGkfpmj>KQ6UP~*x5C-TD#QiO-f~zdxz%L9cNf*EYX;oXMcB(
z!&VQ3C5WNZ+aib{YY*AkyvYNXo<OR=OJ$t59Vr^2iUvbBu{84Up>z<JN|fS|uoNSt
zV0-I|-{Vb+_gPdl>P;Sc<VoT%L`X@0Fl1@j-%nRmriAi3WyeG_8-xLWES4tK3iR6G
z0uV|Nsv1f}XgBg3w{=u1jvxM!Pw?r_K0{n8k(vBu2=Yap4DOze1`&u{fL@elX38|q
zk3m9~h)PY25oBqGvk6(|SYDZ9b#(>j9AEp&@6s8jy!PfBJbd8+9y))H(`QaHx3qv#
zirw8Uu3o;(V32ZiV~h5%&+4g@tjyP0T|LPl*(QzznU;iMnQ|jUTSru?^Vp+LG8zs&
zsIxwE^`h4N#Z<+ZqXMYrov+UMOu$O4v9vlp4h}k;J$IVL#d%Jxog^xk>5e+AO5r;I
z9T@8v47!t2q-W|xUh6L=Z;+R2KpOCaqNg(#3lr0i6Xg<&2Dv3c;1Bmm))|`?CO2Oe
zPB7n$_^ChpNj~x1V+aLlni59&jKkcUNGXwLu5xmiJ6AA`zWjz@Vu1mQT62k3Cq*X-
zah%a?Gzo%$6DQXQl;Y;KE4=r?4Q}4O%R;ltGmk#X`g%sC8l!3r4i2`td+P=V`yKje
zh7A<W#W~JD_z=UaPpj`Jg`1eX>(IFosWl)2&Yii)craaak%JqTXO~hI?dqOel)28)
z2=9XK$AuZfAY^%I4H?1d^;68xH`%%&8Cypj$Q(HO@>`OOD7ogu#Me(MaRm8d3j|7d
z(5rD+Be1@c!I!q~(*Ynw2>6a}Z(=P`BJ}U7Gg!1YKemo~Q}M$;_8jNVog~XL->GR`
z{*LE*Iq7B+jhvkY96nPKfC~IUs#M5ekg#!gk0^|Zq7v0|gE%fzt<))(YRt{AVVtGk
z@6c*(G8*<-Tv?-1t}q^t8T2~r?e2K*$tYrBd4=<5PNTw%8+TjW{`NgIn$st|em2X7
zI2m$s?KIA4beei?kdt7YpBF0p2_kbHs6<UX7%iTe7HNvsn#F}BjME;(#1Z{eG9Cp;
znPHqjXO_`u<R|Rp`eCon%io6ygH2xgl@k&7MBpt1->>uDW02JApPk>c0}(Mu#{9uo
z-XtAFC|gG18^ty@4(Rt&DwQg;`e|242J-T?o7Gt5&$u5W>7U)TwUZodzR!3xpw&9y
z=FQuHVXo0+{mexkyz~S+yIWu_QCLPQMYrEZXPRNZ%k7(2Y4!H$?(MR5YoDD?O0`x-
zDMhs!6DY}WI3m=JL3haYYghPeu9eoQ^<B$KF)?!|pZ}SqejK2+ghCMJtgLm0%a<?n
z>g#W@vpFCg-(tBkrcsv&CH&|}8+gsE3vgC58Vm?zKp_1AZ*9)bX?vu^KHFgSM;(K~
zm|yt$FH<g6iOSw6+Kd%fKe)y7&pt+_rg-#;6TJG>>s-6K$#6KLUZ2i>oiNbOC;wsS
zB_zeT$Z12Ix4AxS?J(*e(p;G1^yxFC<01Ro_ed%=8tZEaty!!uQ4Y$vTZBNEfI-L4
zN>#Dq#;se7)66I3#!#zOIdx)<!RUZYTbj)ZckW!{MQ8JU$4;sP(?%#a(Xmc|wU{~?
z7_9eDRbDeKMToQBicCn)x&=|mH$OR(gSH&tlK+1tCPhsTfc-#(`-JH}8mOz(N_5&?
zgzy>`DMLSHz>bj45ycsw`TUc-{OPBu)=CHwVDX?tF?EanUJO!E$`Vo?VIL$CrzDR(
z{UYza{Vn!)H|g|7lq-(Yr4sY=^VI8g9)Idd3?Zwlt1LEa#LCZj)mk$i^w{3oL_0Wl
z=^-``yF{6$5|_FB-n*PRwML^>rc#mghu+VoQmL`MyU+amY1Y<GO>}DCEkiEKu+tfG
zVxoNm3DS{Rc>BHgxOr!b_b$KBqYs~@Qmzn}WBP}VR7>K#vT7Z4dp)vD=R$ZlvpIS&
zG#$0%*%E6lflnSMCC-l<pZntD{PO?u`{*plTm2PEmOMar7Kf%<b^P#;Kh3kxJxaM8
zqpk0f0-4j?Y6{?IeoX|ow!rdoy6z5;f|a$iEG?g)*WRJmPe@yB)=w?5y1GWWT&COY
z^WZ}dG8m6oTU;QHO6V*l%Tm(Oh^?(n?%dm8Zefl0-+7JQ)&XbFpCuc#SeR>2tMrLt
z&>Dguq*@Iaj>nuneUbI`GhREDd3}gdA(H8R^~{YpvgY8$zuulh&>sx=gIE7G$vEYO
zXCEU}0nNE4A`Iz|Q|fhrvjS`U!0P^BKqJ2|;63AlKurYpeuR~uH)?G@8#ZUjjWPFS
zUl2f}R-)VMlZ^xf9>j*Kfyg{F@5_;bW+ULIe(IAv|H9)4B}lW3D0gd@lLT&><lv;&
zEy1KKG9MC}-$O-Qs^dh7ZZ8F!&}>LlC^>UxnQFC4yLG_t{Lb&Qxp%;z-{$4#pJ)BV
z3YBV1sT|X4?{V|m2kh^*X!VDTtY%?liIeBg(@(nWcMNeTaoQ1;D_A|kO2^#%GD{09
z3<m>0am2Yiam!0bULG_7@(S-vZ-pSUzDrhXi!~0VBIf57NV5T}%gekg0*0BP8c3uS
z*i@4wGax?i2vdOv@uBr4NsB-U@9im^pPlNQ%B4!*zphBXej!OV@D&!Ja@JMQYITTJ
z6D9It#X>S1r1W|NmX_zR+Ow4c?^!neo--!dc)*AAqIrMldHW26K%11k{T=oX_DRPX
ziwjF2B;(Amw0wqItwE*Qpi-S<m<+t<sZQy2TJ$;xY;51-o3Fjf*S_&O+j}jN#L?|*
znoUKiB#<&?exXLMe+VwakL#GeKVH@__1cqC`JdrzKI_>haM~Camsa?jfAeqh=9{na
zi~sl^AliM(u|x&|#yG~~3}ej{oH<9Jl$RFeB(^6C1j>1Rv2{~}V)&y+b!lq&##grq
zf+(+)g<wOjT;AZ7uU_S?*Y5g`Xc4lz)1lKD`r!n5N4}raIz1N!<X&)Ya?T&=d>4kp
z!%a$M!Q$LHL*Vt>@9_L5o~BW4&^^43vrvg*k}Sbyg3jInX)<Cs?h;fhWEOV!57^z=
zU_4GpGQq~?2GyFk1Jl}Z_1YDNqXBVL#XB14yb&rNPggLrnfI+#K`?28@)~)|`|n=o
zjraCAGausIh)n00AeWjTyveqdl18(cH`Ph3_03JjO&f5Ey(A3pU(?e^0(HHXZ&@}*
z6H4zku4D<TOGH7ybI(1%=f3a)wOTFLh4@iA*6Ey$naF8!+SDBZACnJNP)$sMos)u7
zXD;#SFa8C7@fUs$%mIOQ9u&nfb93JO*buOO@)T*m>-Ap3ARJ`r2x|<BOUtaRtni~h
z^aXm|F70-kiqtGGEigz1G@CK0PSH9e3QHXBX@20Pr-<Wn&X|QvtRwO))LMhXOgbWq
zm}b;6w<Kxsqd)ir*1)g)>Tj^Sd&vC!JWoCIByW86>kP&NqCk1g4=@@I7>`GumC@N`
zQ<OsDg4wc=JNonPqXwhb8)3D;nhYU)2b_{6Br(cKDrLdvzVJAodhuy0)iO!~R$<aC
zXIqbM=HFpgGre0A)<FnATFoiJ%E@y)|MHLXYrpbO7>&nNB151Q)oO)iv&quRDpo{1
z{P3d;+6UMyqf+$~Ez<FTey_{Q+6hjcI#0P;=ip!;V+@;H8LLYxbO${e^?*(<L+gw{
z)!4Xqz>|-kCXQo_by!<~QaPc>tuY|XcW6S3F*HI+M4*VmDxdrG3+!wk@YSz=oyQ-2
zkcIhq=9d=fb~jnlK5H{pptYgf>9CgjRux&JFZcMCgy~RgH#u0O1!3=>pgM{afA%ka
zg5UVnuhZ}C0dWrIW0VZBHU$iovgL<<>?xjq;W5e;f3LJjp<rgEI`VE$!MtDp$+)%R
z-sKsw3rbvi_!)LKZ_sMB*>885o2#<Cyh5YVz+{GpAAOjmR+od_El#YiA*J`dOvfX3
zcedHw-ezI(1Xu6va(m-04`hbU{st>cbJS~X;s`8;Qe1)<hDpj3Pdr7bR7Pi+XB=fQ
z3a*fa*qnj<(3&GbS)5d4&hpi7ev4~2@A3FU7g#?zPovQwD3$1rdMsE!7RVUL(u`ie
zM`LAe0)D(Pt0K4Lr<g-4jJJdc0yjz8<oC5KCvux2>r3NElnfF0X*$+wMCgG}6bqhx
z{y}QBGDhd)y_`g;2<g1mR!VW?zWFWQYp12mCsAPNcKhsX9}q=h{__Ir^)mHZlVogI
zT3I6uLt3qU4i0u%UOh{tR`q9!&4;aLj#^`dM;?0us~wyBo4o$^J#KCGP?6U!3L%K1
z3QLQt1X5zM6!O`i#^ldtKH)-2NhGCblAXW*Mr%@$lBJqbT;=rXD#jXqZElIf+k31u
zBr?+I%wb%D&h+F=o4nV7!lYAev<F1_`*EK-+5g?c;YcvX=I5LwkO8gsF2DD?uhH34
zNGDMuU^vLQcXyA|r%&dDgPpis2$An!IZI2<Fne1jKQKx-n46!cR<APXb?Da`+`YSj
zb(XpLB_4VFInF=uFuS{31S+HySLl!WB;zqznsD#V4fgkT==6Hrxp&BJC!<t}iDN}6
z3XwvPj1zj@F-D}cS}j)ADp+kulQf?g6d>`}m8BWuB=PT?a|A*ll%FZvX}9?&KmX78
z#w#~?;%tM(@`!2`q!Ql0)a4Ea&S5jdcsTHkuacAVt&m3vmou_C;J;st>_;oVRuzMe
z#JLQqN(f<)L<lTvb;ZjsJVv7)BZc*ckd#QJF<CJy6TcxiL!qOa0J}VE7qYN5m>Rhj
z77<bfJpa-cz&XD1ul^}EX|p&tM<haGRi<33B4vz@46-rLsN12_-9l<ZG9D2~iPVO2
zAgRtbn4fF1xU|UpLY={=PaIdMS7LgdKFVp*@tCt`FYx0(_8$<0C5%mv9Gt!?YV!?w
zw!6=>tHu@`KMJ8F<w}*z_`~ayFFZ@9eaNtXz+(?S#`~|mLq8ePtZJkc80kp$kl~<(
z3Zo-7B|`ba6-<XxTkBi#{iq)N@An~*@AVR-3vp(M!*>_?!#5%jn#UhL!^<x{L8Tmf
zU652rVG2oPUP>z1&Ah~#H<&O1drZLJ<o@Odd=SLE{J9@zoNB)EuYQTZ^qH&GDTOiB
zQjJQzjuRm|)hsoe=y5{3vyHNjEFBXC0mcaGl@j#_PIK<eDZ0Hb%S&~@vD4b6Q7e%q
zi8l*HFjrsY3!nc<WDw$vCKAzf&227{$%Qf#pz1$Du8H%^mQn%Dr4<&ImRMO>BufSa
z3LbgnG4wd)%m4b<NVOrtA*=@t{lOtxXBam#W($G$_VYj2XS*g(Cftml7Wak1_eSBG
z_JPJa&ss~RP%=V_62b|}vEnnIeT0`@{)7kf%6G6Cd-Nh^f!uf32~9y$kpNjGdG=FZ
z<mT<0Y<%s@D4ijUrB?G}q2`)%tg9Gn1>;VeWHclSeRpU!9x)ycn4e$d^x1PPF0HV?
zzt3niWJ?)VR~BjaJJjj{!;wa7O`t+{w+5V9f7s8WM9o~|3@d9)K3QPOk`+Kj32(F=
zsVbj&`64eo`wYMS>%WPThNZ<t&YVBP-u3s$GH*(1jUQdt?Q}VjPcAqL;=l8L&02zp
z`?c8LBcxD7p`>s9bOWbkKGH76kovvGNh&4D5B>0yeCG4d5{7=Xoj=2c-}mwovDeV%
zEN$^jFeA6g_t0u>frG;#I@Q!`b;^|zi;D}?>veRN@dsc2GPgH3sFgz=yLbWN3@6S!
zfHj8gdpFs<x5+3qh&ZA)zr=-yA7=1&kA=l1x9;^AkDy)+=?@c3KjVw1E|84I=q$;b
zH-cRE=NZ%4>Wj?vw7S@fb1%}<r%odR=nvY|8#T5vfwoW%6r^5)F&K37;qCq%R)Lpq
zsA3wB6rR}=2CTtIiBk$A@89^Sj}`(+=*a}<^<qjyq^V(JYoCgmM^f_Vvz6>@AL1;&
z3)y_9%f+n2g>#^0-7Rq5*&+zBHeqjni|w5)vNU6PVTmv<5ye$j*3MIkE7Tf`Sm79t
zhA14p_8|xR+uYl_!`Hw0Ena!;9op>yZ6k=Qie@t)js<ZXF-Q`6{T3^%dB3<By{vNv
zYwSeJZyji(kv0Gc##qwK@H79>f5h`IJj<{B($ACbwK)@bY+?+I$Hu?A;J>%Egn`0|
zX=9s`(why--09&yGqN8E)IIUk1N_d`J~q{e$iJ`(OGP-=vVFwPJ%S5oQE`#o@gdg8
z&1~>fn~nkqP2#&e!-YRs6vVvz*&n1?7VO=9pM`~a0J1D4%%EJal4d=0(nn_lFSbla
z2nRB^=P=^|v8v)^fGU^CDpgvo159Sni8q)Xr;btP_$z<?f6MaPnjcnd9ah>So9)Ca
zPEH3C7c4A!Ms2J`3NM=NcKd{3K&@J#+dd#mGcT-{l78Q!=M={Iu{G8jdi^f-<^nKn
ziFPwq5ydqWhhp)NfTz1ZZhYYfpXU1e*Vx+dbabGCyg@)Q=enE;MpT4FVGyWXa5>cz
z`s<oblE6(LxyiXS3+D3Y!B^z`Azv<6`H>&{DdtLW=gO-ruPhPDz_S7xO0}BTQRp#N
zrzqjDS?V3_jYT>E*_cwf45boL7$TfuYj+c8e8M1{XUh765&!Ye{#DMNzlasaXO%N0
zcPEScJ}oO2Yi<jD5Fo9>N&h`>-`S*AEpzVdY3|;+1y&G-5n~Ob)DecBk+C?&qb|vK
z1j6MEt!KvVq!V-MdM*TcQsMtx=79U-hSxL(M3F)o>8*T%5P?Aw14AyZFM2RviSqGu
zfuA`na>hw|lkREv=yX#PISLi>XgR(>rV>G|KF^=~OMeYv1-IXQg?crjUanFKN>t-I
zVY!0N9L-7%Qp^6qZIpv-JS30;X&tpFq*+^JWqE<#V93%^lW{ttTCG#BL|`*)HU?4V
z>8C!)(@(vaPfeMf6FFrT<?OZQ>uvLWOyD(Cm0E*ZtwF8kc<IIG84dfKJaLNWo_UU6
zxqgFDHl`MMK;)cdFxWo=;-{T~1%S7cCJbx+{q+sR_a|xlUa_=N;qyQA2*3Q#w=u?-
z3YCfpWC?3VfFTStpZ?sVeEy5iQLR<I_m<I<F>BvVc2QJl{8G=dv|h5Xw0ef-+%nzv
z9_{{s)EXXn^kEu}2BYDSQaMHnK^#UTNk%drk!2~Pevc2XT_sB`TYDY8{q|d2c<>_9
z44H2(P_G>lg${vZacQ1jzeBCD%%hJy<{6(X@#mS$_k-L#8#$YBcwLono_RD@^3LTC
zc<;(hKJnxuR4XBii;EZ;Fi17A^}0f19KBxGdmiQ#fSSlEieymah6auxP?IJZE%kRj
z55NE6o29ZW@yUE{jO^bBpX>=41DX@3nw&eg=ATjLuqKx+sDh<cN8TI%yHA`vJLwpY
zQnt4bkirnC2%`m!dXsXw#&D=vTw1~y!}jhLjU;1!afMR3ipx@>xI`4!SUY)v7hVjB
z;}Z9FHu=VDx7j_&h)ND;1(j+<voS}lQp4#KjMs!0_g%QDH(?w_-li%q7iU@OoBu?_
zgAYE!=`*MK<~LvA#_OA`4h1sOWT|B^7*5{dg49qd@QfVO<SYmRKQ7OF$p7#@#76>k
z*RC9bD<Q1)BVmw4X3UzjG!wY<jW;-b?gGtA=Wvllx~Znd<T`|@9%+V|naR{=7k}2Q
zw)hoDMOc~V$xnU`ed!$Ou$v>@6rnW&gBkVMyK|kmQpPzyXHtoPNQG?fZm?4=@xTKY
z@tSQv?q@U@G9C`SK-xjMy2^k2v;TV@eeyXFUh5$9&eK^iRUG&QSS`BY+|0q?tY@7{
zr4mtX9w|U)3AJjCD2m98K?_Hc2(<BHOyeZh7{;S9mD<rcli!)a&*&4FQu+6z9QW!s
zZlPoy>;O^+QG{@Yx=UGCkX*Y)`^t41&ov+p5z-<HvFS|fy2!y1N3?!3*Y_R3Zo1}5
zO5*wgFMZ)BdGx{aq{FtiO)x1~OAtncSwd@P$Iqs;_zV+497JqwZ_%A^QY)XrIztFZ
z=NJu!B;zsZIQ53HI_7gf^wYfjxgSD>0m=$&F)`q~{E|{mv#cD1PPKymg^uGArD{YN
zNUTmLquVr=v2~0SMJ4dwSJrt=+IT!Z!qk|&>0$ydq?~9hr|-#8+klT3j*FL0@XWIh
z@LOA(*epYaWzSj&K~-p0(_O-TkH9SX2@E;oCUeu=2?>MC?Oi^!-rr;09lpUQ2m)4?
zPw<z1=4bil?=>0s_L*-kV04B~HF2%v8%#!BTsk0-4yzM<O6|+hdOV;M)(KTeP^)9p
z5qCB_eoBOPSObGmMsx85|C_(|pHi*W(JnnQm$taSAJ(WAiQDAgIS))6g1!Ai;y7e!
zae;fop7*MhlGMR)?5I?Ix!vLzjlIM|2qqw2>!Tokmd*6c{0wmZKE>R8EeDgn+qszk
zs$>bNBced^{0ry!!5{t<^+wI_K|*1SJ+hC_PQEGb<&=$glXs>7b!+Pn@Z8Hk&aeN<
z|IB#UryLq;wHmcrol_^*IWJ?LdG<MmtwWsg4N*pC40{~_E?jz;GY?o!oLI#<$JW+O
zmX{W|x4lhV0*59HBlh>kEG#V1Xv}%GDNi;9)8xZggLKo$JQIDWkSM2cqQtqy1y)yA
z_{x{R%K39ASzKITZfTM3Xp3fDVvW~G_j*0jEX_v@+9PH5*)zS^H<fjhd-=x^0zn)F
zj7Q!y7b#G(f)rt{sr5}HXU{hIi~r&0SwFi*lBPtVZ*WtpXoQ?$X`P(zk;v&D;T)Cf
zJl$SKw{=LhQlna{u)ORww1Zxk*WY-Dds`b^I=jxf)9WZ@Xfzku+quiVTes-+N7x{s
z+?eOWLk}_31DbO+_73;4CPJ%>cCXJ`?F7|QmGN+Z$$Z67<hod?W^f38AF|GoS%WqP
ztqltcO%@jy+1kBLbFRrKlcWZ!fy5=2BuN>Mhq=X#PrOv%@^SIP1G_N5$bykogv#X&
zGJg!3x^{aA(sux9Z3uM6x=1;p6v{=k?;LXdYv16Bl{tcF1tmO%8GPt~_lLE}bwc?=
zly}C;yj!x+FnKM8KOm)2$RNZ@MO<AZaE67sKKomD7z}#|2i^8An|E$fQj$0bF;);&
z%cww+r3RB3Zrr#|6i2L{Tt~<NlctPDBXEwhr<Pe-3;CJ9`d{$K6VH<8Q4~5zIa3Dl
zWyIVhn4&YzmlcYxF92<|pL3Wu8jx?yPUP&H&9KffPAy659qf(t8)}-2NykH+F(@hg
zn33X;VOn-8etEI2&ob40(AItHYquB;hX|1&Y=Co;K&PCP8Oy3f)^}X}%Bw7_pP*Vj
z2hwX3q>$b<oGB2UYQyq1&pW03x*w%#i$hE*MHqxQ8BuPoP?DOG8_?R@CP_x5!yfG{
zV}E;t-tH!gt0xJ95<yrZj6wprgVmaA*RB#*O3W=RA;SQZX-2~lI!$TTVw?>5i~p;?
z#^-<dCsAQEV_PzvQ8)|SowYfm#iSv|AN2lncx{m*iXu=Uf@x{ZS&Oy??O-@^G|LX-
zWNx5p7!3!L(rXw5USM2U0OZ$D0LtP(bi#4J>}Zp5#y|KUejA$xV9UA5vhS*{nh|Fd
z1bc1nee-n|S5}ba7^L-Vcz#{qD<;~+4<&HL-yKQdRKViN3w-8>{~V^fiAfW*$#5<M
zV+f?=aA$)o9ic*RgdPNvQdnZ6eV3i>4eGTAk<!y4&KO4h0pr0ClVunK^NZ*B*}wK*
za^lQ+1cJ~8ICT_=6u^CUKDD3sI-9upgRm6GM74R86r^K6Sw$&DYD^C5B)S3y&(35y
zh_~PI^N#9dy{uw-ukQC~p$~_%-dZLgATi^dv2sWeBN3clsq=-;KTD%g@r=4uC@gw1
zGRbS?tn-r+3dx1=nZbmmot4=LAqiB>XTJDTq^agxzx&ISL&IFPPB|*msLm0@WwO+=
z+*}~(_vjAxAj_~OB@~Kk98sxPSzTVB*B`L5)L@hhX*L>EONuabWaET#+~6mE>@N|;
zRgB4wB$|_?O-%L4PJqc}4-P*ypj581xOM_11=)Dy?Q<?YOndKuZ+~S6ZHzz69gLG6
z<53U$Ax73S+p~<UZ$g^x>3{T15&)%gz>oaI<NVA2<;x5QAOl69N+=m&%owa@ZN0>w
z|BIjD!o_ujP-G^-g_1yh50=(?vcs=^o&@KtXnk&pl@sSl#sfP2A=){fetMO9y+Nnd
zVR31ZdmDGD*F0mJr3uMsz^z+1*xNlIF^b>)%2&Dg&_!(8W4^IKqrO2DI;2v}FI4D{
z#=P*trx3zx>}=iuB6AYNS?hZm^A2PoB@zeDC-<#(n=gI!HCC4vxp;n+rNu>D7%@o4
zl&$iKT4v~VyGIhiAW%4)m)Q-)d<JZwFcW>Oyl;%`N1G@CNEcviO3>ftjM-s5a*Rx_
zm(!N**WY9D)EcKh`!c3h&tG`yo%Zu?y+UsRYRYmks~4H-MkXaCsSql_NtFYD5LH{F
z-`{0y1l?|zalgmG)-IKLjro%&$i^A1R)_t5hlBlnwzoIYI;Bys&^kCEjAFtlrrq9S
zFzykR%Y5?1&vO2ePjK<67ce$Q*v<nK|Kd+Ga{&Y@@TJum00Y8XA>g$*Sc`T!C`C@R
zY1X-1E2n+E-awi;LdWza&kQ;p`v=ZByx*VK58=Gh)J=4BIrz2t<ca(6kT~lF-~x#?
z89{%Ci}YCvls7}w(7m<I&ENkj56?GIORGrX4FNQujUhY&glFc_&RE<`Kz*T6lEQ;N
zrLZc<KVJaVlMH%Wj59~4+a^thjM_cAoi=CAp9f>v*t~;^N+e0j&70R5jRus<5qEFh
z1O!rr^m{F@qb!&C^h+;s^1|ah_t_r<86vC$A+gRLdC&Zd=kjhu?_pGw5nN%IjqhAb
zg@eHRGJ2u0!wG@04&yXhXE+PP)L?Yp(drCV3(|2Pm-$RCce(NRBmYnF4EkgT@*x+X
zrp+cFZ_Q!z&e($?rIFU_Nh0IH{8|u!fxE9=X7&6!r(Svip#mg$lU%0kwcpg_(XzXo
zS$C%Tu;P6$z`RsGfvYc`hEfEh7UN-$VX2EwQ#!3f4!7>Hx_T04B~pb{N>!W%t1Y*0
z-6D!3PM$uCQ~^3k7!HR>1ZPeyqAE-LmH+&2aO(7hd=2w9Y{wK_<_z=HUSisyU?<+1
zUO#KBMNBktmV!IC#-OdnI>&Hq(b}`J##)TFj0QbyrhzGl$9i*PDFg_IlfJn^O8;&+
z>+Z|@{XIiCUi^W_dG(byxqG`0h8(y#D#CI~A5b3l2&}<36sQ?C5sD3RDzhorx!Ip_
zl=&2Zl8UHW=g0rdpJ$~RakzDjrG+J|*67Sos#UQzLyvnHGxlv8N#d^$NQZsqYGnvQ
z!bTI5jcDI(VYT1OtrfI7eSZ9>{sK=w`&@2K#8HXG_mI?BTyF8=*AF3SwL1h#vAVL%
z?OQ$H`5c7wQ%9n`uF5#jI;G$1`tqwg%E(SMy4VT3U9hzGkA44W5yb({I?_a7Y)U8t
zl#EcK49>E!5c6k$`eiOXc#bIYBQ3Sf_n#smm(z0lk!C*UrvD~SO5;kMFaGF%$UpiA
ze~0w&9@W@TtJi5X8k{<H8Y?26c<eF8y)J?D)-TDhPcj~H_WT8^^(M>9i=;`y_Rbxe
zjVde4H;H10wHhTG9PAD_bNV5a5?JfsQ_swLJ#)d#La|4AZD}BU(_6i|$WQ;o1HAIe
zt6aNwo8_ftPMuk&yLXlOIyeJPINF`|L@G6DfKhUCFRUfkQ>vViy}xVxkwBev66Fjk
zZL=;8c~EwUg<{j?hq$x2q2czIzd>_#iTS6VLWTjh(51|jhzir;4}-c7Q}<q9<+Thl
zhzP=PqHS?PAfuDGo!f+QMl~v7*L>IAXq*w1Ydo-Y5}aUheut&S1zN3rKL<9m^twHw
z@|c6p0mf?1TzrfR4?f4rsSBPa#p?-NZYSYIm4(@;os>TcAUZ80%Y0s=!I->#8O6lI
z$L1hL&J4V_j?4;%siUp{=jnH23_2@xbVbcy<n{U~aMUv!(yWO4<Ay_{tUwq;V7i<)
zTRadZ46UG_mjJak?7V)3xsxY2{rOKLs?|xRV~in=qDj|oAp|Nme=z~?Q&3klPdFhF
z%Fo3s)|@axmChkL2RPF~I6<NftEbNpRcbgSXBU>qGQ-x!9ac}S((846lT2=&8kbA#
zw|2<PnEBOHoIUp>7ax5V8OIo%Quq}WYsq8`mzmgfD5)lzCePI6{(|`&S0xlOca1Mt
zTc7p2sYc4dD1*#9+<L#Hywlm2-)6?-%}5la@Fn^Wi|FsCHpAHjC!;)x%ZPd#Trj(=
zhJumx2VvS9bK`fuMq_P-=0gwr&zoPP)|xnuC+jMhg8A8Z-J5nEP2hx52o-|zesChJ
zVXKP4a7;GdN3{0I1|z!dHs>x~M2UcQyGNo8SMS_lYvUfnasU5M+@A$WmR;w$;J5b~
ze2#gJjL3P;tQ;z{X4HhC3IqX?qC`>DmDDD+)YfhN(9iwY)(_p*58bz|2eVe+dnMhL
zNQo4|0R%w+1r(})nk(mdh<T3iaQE=}oU`}thkcH_CkRSbLW108X;~E+nc?p1?7i3e
z*T4Rsd@e(IcNagEA(c*3u543p*U0u2c;k(?Ir;3Dm^ysS3eFn)x^$eh^?sI)bS(0m
zOoGNYmrdl8!Y$r=ap8urh=huh^(3Y(V~8S~lxd^UN)Wc~WInNgxJ7jZ5GTkO(6Iwg
zT=3vXp5~PM&#%S#Eq?#s{)|RdBDD_@fP=4EOiE3r++gv;&lsJWM2`Cik%;zorFME(
zmrwQN8Q-N2bPGS%aq#;_dExbM5S8W$n{`_RsSp!I=&-@g;zPW23S%U=f|Q#kYPDHk
zU1DIMKqhCw0~0A4)jG|39gIQi6t8{xxB2?Feiz605XK;7*RwWBKze02dM&Q(_t^L3
zdmj0L0fZ2Qtp?eQwR#GSrWrvyaPUlm!Yi812CnCI-m}(E(svSCMUw3H%D?<=FM*(M
zAWzzNXg4E^t_qFB#}Rv6KDmz*$M@mNxMeL(Oi!|BgCVksuSt?7)3pMXU9>(yBOKvy
z<is=l-aq}{@K=BI|Anx^V7`yElctdA!|}6(nGiWJ%I?k*Qfq7dDI9zu7%F5LJavM4
zv&G0rKRcyu`tt=cp2QW3FbMh5D_`N*v2)<4PHdBS8e?VGI?p8rr41(b*mhl)d|?1D
zo#EnzXV~0Wqf*-7%*nI7w=zelRqW<zXf%q1LA!&>+OC>S+KJC?i^!@D8v8W)v0pH$
zb4bx-f(;IO5uR~vvPFc0AP~?tklrct_{Trv%*Yto+3C(hqLo6XQZY0eqd!mZu9Dwl
zl%|u^NsulgHokRZkV<KdCp4}jas3pT!2$Zl##o<U;@XFwbNtLRv{RvNA+!~(YD9T+
zkK1={bMEqa4jn&2v)w=$O9DQ8;#npSoU{jV5R!~?Cl@L@;W>4=Ya*`Zlc01G0j=W8
zoS5(*CnZTG5dZ3^mSczzGy+XFZ4(3?kLZVC7!z!*ZJHe$Pxw_fA$E=1ZEw&;8h+Vj
zjKl%broth%nQ~jCrKD*jQb$NBX=5<Wh^6;Gq%c0t$c3}$_;XRJ*E`c*?wt#>=WtIx
zyWT|J2#eM@o{t;HK*t-z>bIpS4D{1KHqQ3yIv>3A9!E}|VW2QbHkD(%Zvq)Wv(#XI
zVUDq>aZXQ+p>#w$)JU1){N*>8K6D0Uq?NJ2%I9>rKU&8LTRhPezg|10-{rUwzgJ`L
z-_8YR(>OW_e^M1D&wWclJJfh_furLJ58ZtYb{e%;2T^ypH;GnZM+EsPRD4DljId;W
zM=Shbmjj~65vPV04sB)8%Ft-qYbEA?_&%ox2T4y(bvQFgyRwta_e{bWx^aG@fQ#GM
zNC)YqV$WOq=gLpv2V;b_Z8Dh@DK|}~&_})2Bv%+@Y-)x~I!~okWMpuVo#GbFMuRBS
z)M`}%)uvc3VvOR*nM)ix@+<?R`-qG}$acKnbWH!Sv9U><v?!?%IBpk>62iV#>mp$b
ztxn+<%Q_@QTv8Aipcz4+jI~IMlY5KUbe$sNlc6WBU>mki@+UhA{l9`0wqbpxPCJYc
zGD1fYQnlGH+Kf9c(n!{>KW6pdHD=#<1vlg2V5v4LrY0m*6@lAHgc6iiCRCePcs!xx
zNJk=F52Qo7Fou>O8e7z>yQF=OX05{J>MEWz^oc>5?TEgC6k(ufHQQA7%DA4xk>e+D
z+!R5xNu$}Ikj-)C!~xE|_*LHijo(E&*8RZ5o3oQt3fVc6b*yY)ttt~KVymkcst_Zs
z|EsmhwEJA?h)737%Ft*j3Z6yyl!8{HK{}nrXocrExKa?sT78IkIIIVmN5{Ya9UygK
z6j8|Mn3@`+T&zL_LOQr;hDAtra}IXPIQcxCp-4wQaR(WrV;#OOcm5YXu<>)^dLHL4
zK2Kjpu)T1Lk)ctdC?JY7=|YYusL-nIBGIUJ1R5lQ`d$e+I)vk;anflTsWk0Ilb{t~
zl%UyyqZ9l2-GB1WNN2NX-A2hS^8WM4Y;x8U^Zm3oacnhFvrRUe#d95d-b-Oo*)Yxz
zw4l{)(rPs!rdNewXrHT2p=i$wr35Bc$atFk*e{gS4dgtI%}(Ond59kvjB)U>EdnJ4
zLJ0x{TKN<ot+MdpN1W(?8<ES~$<rW2WCX;%u|n8d6r<2q))hf=p-YUEc9RPi>9}@7
zMG@^t(GJ?QgEm5I6b{O?P)3pU(@e}9z;QkNOo8FqWB5**C}@ybU!oCIdG@(;96EEF
zzKJR7^N*}wpqr+Da2#!9JQ<TDWyvpFWr9UijL{eoYtzZZ@`}zw+eOhNQd`@w-mrBZ
zeCSc$7BoXim@-z@L<eXS5r&cFXNU+XB1=~7a4}4LKw9Kb%S0eq^E93lM@a_qDF#uD
zSL$F=fW#-HDy@`5Xfz=%t?dSn-uoHZi7~QclNiS(h}srqa2<s7V)$Kyj%>0dfY=t^
zNXHtP$A_gwXA>CQCJIp)f>C&>EGFDTNkcwgU}`3fQBW8dqHkh9!g#1=m5k6h#Z7Yg
zG)GULWOQaf#kFM&8rMtHH#CLD!)QBnp-uerVpPpShRH^c2LK$|u{*Lcw+B{Nl-6Jr
zxIHIW(mL&Mvb5!CMn=$#tbBsjR?{jBEn=gMM#u<9XlsBj64JbFzbl#e7wY)yCV2{f
z$t5t_!4VD_(O`xR_IVly&)z@A9zuapL|U+W_Yun<e8!=--b7@xIKm-_f=)te3%EF$
zH-^MH#C~+$0!~VaM%Y%aYi)eG)5nUG=g@Q{PCCu-<Pc%A%_l#4pNXj%28N&^g4h-=
zqFStT<Hj|PpFPRZ(<f=Q>S!EXH^bo*ml+zLCDIxbL?n8XjuUqUdJg)|{!e(obP-%_
zEQ)W8KxiuxoKQ46D=t25;Rb;gG($~BB+o)zv}qe}#}T-WBvQI_mnYM5Hc{4b`{wED
zQU3L%(YkF>7$p#Dm;Gv;Sx>R2?0Fb91LlAH5&e@B3_Ww!qLf0~ghy+n=k=i72?COE
zrR85kX)WEt1ILXK@f4zOil$PuN;_2cO4N5Nc%H}b^bBF3KuA`$H&|O;rV%#rT#1Sz
zwl+4&<Z|qkHfRQQh9{?a=F*or^ZaYbR2rj{ZKDbm@5==F?zTJ)-JuVg)WxlM9ifHw
z`nAI`F`WjRysIcgX*=1Tb3h9#NF9X%k>hmk4=|mHTF@Qa$K<@ji3^I}=jm^0HEIF>
zzyI+An)RwZQzF0!26dBzj!SE$!or82b7piHC*>n!N|1U2q!FDY-u{2LOZQ|UZ<9LD
zaS^W1k<-s{c;7ftqf8LAQ0)fIN|CMQ#{~5%`Mx0>ks_UPafQS7#ya_YmVLAPt@A|?
zQm@w0K|soPIDPhc{`tT9-_bujW|3LkC)(P*Xlw^Lwd_B8{A-)E=5u*)T(k<wX0u2k
z3ACcEpw$+5fHA__8#Wr9g4>bu2}6MHpxcu`#?x9~D5?9}Z=B)mi9zoDVUwkweuk@5
zoS4Oroq@<i7+k!zVeO;a3{KB5aQQ6K@zKhlk@muNyEb7u3a|tW5j#~lRu{|ha2$_N
zYj(FP?5q{p-r1nIyG^-NqFS%g4nrmf9De;P&ydMBNvE<P4E=otf;~;NTSY|`>ct|J
z(iS=}^bd`Y&E;q{TI}rZ(pHB4u>*LiEMd^@h)`-{aC*@M;samA{+>qk9R9@bA^xq0
zM_`e?9*C|dxEPDp5hpmUwkAjmQeIr$8|$U1wXz(lo+PuI`6ripl8ijPR@~r_<hOq3
zW&E9S?*FGhq*bgVv07Ug`+BT)lOdxW%J)}T`q`%({Mzd<P(Ug>((hn7kwQvDtgmLL
zVeR{t(ups3d!Y-*A!=(XD<yV|MM}F{lq(gg)hey9O`orL>Ea~$eC+rj1;e8w)OQ<%
zxi%S3;?zQ@1-QcH@Udg`4Gy3U6ia0St;r2ckj~_7qNtPMhwen4<=@#f$&%xv__Izz
z8E?c6@Ywx(7LR&ziB1yMeKsNu&9)((NeJ>0T4|IDx~QbUl8cS#Ts%O`x3kD%Vn+Qm
zayo%cv&V+goH%)e@UwYD1B8+IQh~HFGujX!Nwr|%qtEG|nP&9tGvIreP+^p@`#eeJ
zb%cqD=5{$4A<#lv?x8Djg=@KVk*3uSXt!E4n@!qb8xgf33NUd~GL<QC{-qb$+bz*Q
zyPrbeD5~9}Z)c4_HF@^A^UNMUMqzA%>fA%Lj_}h3t8=A=HJn$L$e*;<gySGhcNobK
z^$`1?M8<k!DUIb?Nz+Z~l6{IH=3?2Y*JdCs*GFlTZlfZDQlU-aMT8U@87J|{*&>t6
zL5S|xB~K;Zzi4%&H;f<^mDw*#9Fi3>(xD#N*At=%cWTW2;K$@fM##?W>%5!}nkbXj
zdyG8O?ZEV2Y6$D4?Ycgmm+GLS3h;9MG~E=1zC8W?!&EC39^QS##Pohf2F4j5nqstn
zh_$t4Ha9mY?o=rB57KJa*{juXy%f*C`5TN)9z{#vI*16<{p@0R-5C31IiicWb|^Jc
z#`<d(iFX_)7L>IKUy`VpcxqK6s0W(Clx?pizQD;KgpI2t8Gw;ecKBTO^P}4l_&aI9
zLeNsExUN+}gZ{9{eplfN7p)A-A6;i`|0EMHK9BHII7qAEr*%9mVNu#{1(#549G>7$
zJC;L>8yE7jfGW(QgNVKA4(Nzx<sR);lgWum_R99oR?4ulx=Ii<SzesS_kA+CKI+vH
z^;Uy}2M%!X=o#MpjeksGa15c7ewXRoqmHzikWO5r5yw>bw2dPjU>rLon>hO@V=O8g
z1w@gd*)q5uC|98(Xg8~vD6)5@wC93~QCSnWT&09q(IswUKb`f3lDg+#JcDps9C_n4
zifikH4;QdX(@w|wBGQV;1=qNky%uvnd7m@meWYg(VH_VL0183%z8pLBAl;5$l|;l3
zbnH^$ITqP59^0EG7UmXNTAb(c+(RDDJ!Erxi(0Kl5CvcahbISl@zOMXeLk5~7T@ua
zzC*gtC2Cb@x9WsZ6PySypCa9t!)VRc>N=Hbm6nEQPG2U}AP72v>&>1MZbSD%wx@N{
zkMCslF%qM6*U;MX?|L{zJ*@~t*CGazVF_?OI^3%$vXU0ki4z052cm6tMOa4*gNr2M
zN#W98X1)6Q87{tbg3@w^=K2Z`{_@8-F}KVS+LB1I5*-E)rwywgU8Qe&n9)lY5Kan>
z1CHBG>SAOQVJ}pHP2i;`kuqhqrq<S2oL^vdWrc;M1(sKr*xlWw-l!8Q#Yis2%=m9H
zIFKftN|SQZIG&_`pbsZ%QLXM0RBJ>*1K~ON`5dm}va_*8X|F;((wskWiO2v^`w4I*
zSyI}u7L!b>Av#lvy>1A$0MH!-t|N2u1Ypxmy1SEa_HbwghA1sNiLQ!A=sWGyp2EUD
zlNu8znD)%Km2hG}{WMlz!R0r9?=t(Q9PX!TJpS<q_>pC!%a}tej3G1%Aw0}Zi^uPL
zKx%Y|^!|NF9F!5bB96V32qV#9(zUjSq!TCcj^iPuOTE!zYjusy)h)KR*Vx(FX1BCU
ztyx28!>gC38GPj&sZ4{c@8J0k1KB=wsZO)5sPF7hsqE6Kw#XLx$>wr2Yjw7ZMH)&o
zF?$T8Ll_Sqb(7?TuV~Ms?v!hfa56pZZ9|OI^`LrBeD=v@-AxdQ<K1W*a$cM`#RTmz
zwk1mnitheS_OXsH0ul50ezjaHAp}yn6g<H|dXEFTiqyU>Fi?mXr3jIfAFVO}-Ul52
z=GT$gxGNJy$W#XF|LerWC56Vs4}h@Gt#lBME#{4IkgkJtC2qoDR1qo1!*K<;lK$Zl
zGTA=%ihFpeG=ZBYY=$)PP{v_pX_=YB2bel^ka}eogdme0U}*d>(odnnC@z4y16uZ$
zjge$wI>ncaGxEeH3dgn-B@#y)9Ar1*@!79BqTqsNC<wH@yp^^tCDvEeAf?8Y5Oyha
zI5D44D`h1%WQTJ0cd!ha0BIyL3g|Z_PP0kgg`LngqJvtC$KU@cg^6+cjveiS4KYgF
zNsNCErA?C9&IGp*;CM(cWktND#7T{!R6x+IAOMA6U~~*p83rZ?aa^CxwG}3&Ca6{`
z_?}O>RHhvTZ0&A<V>o`{6^@>Gj_l9~Dr$GmSL;oi+|>$4+Mx-DM2*3=+7o(<77oI7
zkxt@O9w&bBpDP3Pz>sr+IH?bUu6#|BFLdb?PXYr^iz=0@FO<~v75Zs3HQx9rN51kh
zk2W_jwH9vNIz>n#qwKgomUvuS+G6qD4><6xLB!w)B1Y=0pmMyKT|uPc%EfWP5nw!4
z7q+?g@GjSHUgy1^{+NyJReRZWJ`iwlXoUf}T!HFd4c|%8mm4IL%Aq}l#3hoFK-&Xz
zFgt)YijB>6N~K*&)iSe(&M`PRYNub4)*q(B4IxhG=t-1}vD0t1BW4fZ<ic06Btb%i
z?Oe3UZ-)2*x9lAugq@;|aiW%jw1!B<#CaXXD3MhckHx4lMq3*rV>R%N6i6K>n_4{;
zA=(#SIYZjZ<EC>=y?BY;g?ZX{9@<2V?!ZfgAriJG*VQHq?|wjLco_HKVN}Q5z3T^x
zD?sUZlekFFCi(5U;_lu1EG#VW;m1Ga`t2)}_sT>{#Rn%iNC+~xzR%uX72i!Evl2g*
zM!U+kcB6pMw{~xRqXnerP%Q3HEEcKM_L!JH%-G~CN^2a^B}<#0Q@S_i$)6dlOpg?T
zWT4{b6M3B^{k<Q*i4i<uFJ;GZVrLFJB^N2|WUR9Pj#6R#7iG-X5;0_NIx&~T2%dH@
zx>uIrms>&@F1>JoXD^+mQrqCb>#wlAvO;u!4uiBqCkPyDbsGwSYh3o`ms$Dg&p2H8
zDyA@Or$ZgLC#j2P&2;>};sQ>(4l0r?Jzi#UZh`v`?{NF>P3D&u*{kjm1OeJ;JY1%R
zzr@A!v!v55zVvZCiOIPL-KJKr5STjJ4e<MXGKCCL6tT0tN~OAo@zNYPalxK7CU*W1
z;>kSt=i0A8oJ1uxzFkQYuR5r{w$Eh}?wFL5kglQo%(O#8q@}I)I|z=7f4t~$bVWxN
z$=VPKX>eSt!`FFMe`TFHL4QH;)~`K}v*5D)gLlvkTSUkh5f;V}7|4bWYoFX?Xm*yN
z%jXe}g9;S}X{X5@#|pbf$pnA=ISV_!ZQJc0S_!H<4OYu*l*>Dm%N1&k8t90F<BGw)
zEMXYp`aUy<4$#;O7#J;(PYHYkd(Jv1PMu=nz%+g)&)U)=fl>^NP2;#eQK&G9ON$WQ
z3pK7(5;We>Ax+y;Alc4IVHFoN-QAX;ZnE?Ch~wOfAmiDj&KN6VYbWP5GDd3>V4#;@
zDox_|ZDaAby$k}hgJ(2DW`~oqM85!pj3uiCEn{e{@9_Bh?{a2n7<XvU4sperv=)(b
zI|-D%ySj9S1pPKbB3u{8^YG*3Mrnobxdfg|W?+D!$!X?q-)C)MgM9~%ka7fRr+_c2
z?5;0TZB%*l?Jt8=L>lsiQ7*jn7EUTdJBVV0*mROiqr06h`+7v@xwaBfw!>tM%~NC#
zC?n!?yp!}HsE3+JV_kJ(9(?a$qwSu>vPgiVotSTK|13ePpT_z^Nu6-wc9bI-IC7AQ
zOV6<O-sd>N#57*ZiLumjn-IC7VDsh!vIq7va`6IDet=Xq7gy`?5QLC1N)x|1DcRku
za_{ba-g*D~y!*lXw1XB2Sr}qyT4MQ2BEoSXlgY8ZxsDw7NP9UPTr!z_+!k?hyfn3D
zjm^ys%6nyM%^KB4i*r*);-DoPi_U>*;!V|Cg(t09-IJ@?<iC5tcYYb20DrM&lc2+}
zipkbO%8tVL==cb&(iTd`x=2QMrf4upi_#`_S`sHl7U>g0Vl7F#NYF34hQ|7B3)%sH
zV1WIvJ;&Vo26Cs46Sv+Gl8Cs)EDS<B)E75d`0x`B3=R@X4`*P+{+Xe(k#Um^Wk<f1
z65PG@fSY%2@a^w@o8`4-v<|y3CQha#ro(4U`&l-(HgU#0QhhmG>5<9iW4?=vn@ZCR
zn(P#Js8q^Sn^j8HDkm<VvZQVsJrceo9WI6aIZx<^#SnVZViht`4o?!gq%E%#Ah%Pc
z+28H6-ka3r^I2SJXh$vbMs#AXbfQ-kBMQ<cbC!7=n}0i&n-PUEvcuEi)Oo(k%Ws}1
zo6AzGyLcl*9Qn%2Jl<HNz1OnIJOZRc5Z`%57*G!DpWY%rIl;gS&!fT&p7iWNX{<be
zaclvAu-rN03Y2nrczc1n4{!0#d+%`V##LJF`V;wz<&6uBl%$+2ySpV^Jo<CPq%%IQ
zlR~xyS_F7bis9@Kt#+IB^>xa7C5n3`j-Gp+Y&MTF5l)N}Ch;=FR$o7da%&TF=Q=#A
zq;OMzy>OFB0%u!t2^y8O>nsu@P(l(YD{T;IgEGptc#N?vDrxIltyP@BYg{4FT9Q0k
z+IGQzW%s?~7+!tjBrkvMEW3|#)Hl}H{rslo_X%6XD+J0YjBpWq5s%+_pUmU{*_nfA
z$H&Mxp?6)&qXiso_km65Z8B$!%l3MS`MEh3<`$S+c+AS`61$~c8m%Tm3;y0)=h!z<
zAe~N;P37@YE(853RJBCCUZu5HBMjSk*$jR<O?jutc4?QU5u7=8fiMc`$+&=?7IBQm
zC*2sJgXknkd^hI3=!;!(j%Y9J9)vzr&<enhJ#v+b5q8s)AjW8H(s2QW<5)nXyMASv
z``@?(;0ZW$W|Gm$2EpweOyuA>I&L2dLSs<cXZ!XX%O8Epfj3@9q<oYPVk(u7kPgBq
zj0)}kjqg6onYRU9;_=(6wb*MmsMl*WTTM*ZBq!P^qj0?x`;Q%`xV}Mlbd3JNNup*0
z<4MYuZB9&_U})bA+bfGGtr!@cM7SArs3D4C@E}RbEzQT?;}J$AJj$3}A-eIj3AQFN
zYqRx8w`-T^Nwxw(TNzT?pmdukv=i@!2uE0v<AevF=pQ;V87%oboPJHX9Y%ui(~Q6L
z0@alr>JJu(RNNxK4iXTf5Uv4L4_W`{GqU3&6oy6-W&nwUabgv3A#fcJ84K%1EzQGw
zkGXpN6W;ylk7-8%2y34*IX-TkA0lg+W3(cb5-1%qIyQlDBrEGHcy5}}!7-bh32W6M
z5EOTctgNlDSE~|4ifrEyeFNk8z88~~i4SHIw?2{#uX_mewp@UiBjLmcK-bpiiH)yy
z$3n{f9sA!-hbZh|1PqOiuv#uzlrD;FC7`-!pDiFI&>hR}jND6!Ry{HSPLC$sf8&x)
zTv5e>Y1#b;n0)DZ=Dzn+9MwW%kRpk>eSwZKxRGG{#ytl2&){50WAr4B@i5Y{m!F}-
zo3xO<V0B@Y`GrS(`_KO~OY3%+K}g7DvzGIqRV?Hfk<Gej-C|^P6jOwi^%Z<C%TV92
zH4+ydzLzFan(E#TYwPP&sudco7O8B3!oWDL=iww1&2f9n#@v3=$p)5E#_{a7+XP*H
zqm6f$1DWjp&f(w7-4R{5T}Vkfmm!}{(<-hLDJuoidvOSomTgzq_en9_Zae>J5pEsF
zVR(4JR<DjNvI>W08Grs+mf!gpS4Q^d#qC7<K^Z#`R128@;8RXdP2ui$VQ2)AaOYz5
zQhE~U#03hH<@t5)Ke)>u{>gvj;r!zmm5#w2L_9EI@TEZM7JdDBc1ydgZ*AaB<><>6
z@Vu=3IS~lqvsd4vxVuTIyi2uFqZulOMrW)%NdmoKQAtaqi2tAN720jV+V9a8x&?S*
zk82$dvY2kGK3NIRF^PFTpCg}5)2{8HwTP(|J@Jl_I*pz1*75UT<09j-HZIzJg@Y0!
zqj_Hc+C?&c8h>DbLtp(8t(9feb{SXLgiZ=c2r)u0sIP9b_`~-(_*(<Wp)sVeI*meD
z$qJjm+s|G)9>NtE<udnhiF^0&^3xCA<I2s?sO*)WAl!SnG=q}_?SQ?#GESdMI+el1
zsJsdT!gNS_AO&ukYOPASRHn37X5WF6^bd`rbc8D`V%+VX{30}7#6^1Cg6|~pnnY;6
zTbw@;=}zaxwiWGHTOs5+T^H+np0x3;gimaIpP=uKz2Jp#F(!N}-Io8>b?C?-|L|YF
zL1TFzpZyOdT5~J5H5#MzkqJ?N--M-iKc%p5f`QW~!8mB);=}__!dO`|X#gWF)y8#k
zT<hPwvAM(Y+%n5c%Pg%du(iF(UagAIn%{o=8HW0^v9G&ipwNeGhBRAi)OU(hOWSy<
z47t8O!e)!j?QI%LaO~h2!m!;{tcZ(u+g7(Iv)0;*0@MAT1i4R;_|855b5thjB)oO|
z^>$=vhv22`JrF5H7=`wy%B~qfhf0BtId$<6$y1Em7f$LB79kQL$bljDz4`*BjctPR
zp4|{Jij|;K#u0c3nrj=Z+_=fWk)sF=!jT}HuA@LaCD&>;*jV0SbA6Maef&P53apC)
ziUa#+IC=ai*RNiuSgggZDnl#p6Get}Cdb6oG)if1-@8d6lc%pRi0cW$R*SWbH6AZK
zX0Kku_0puXIRY)&*s5@((1-8YsyZ^t>V#nl=yr3*2fj8>9Qp}rV@%@nXMc5hG)aaZ
zPt2Nbf<}zi1?@Iq@I9Z=v2j|JJsQO|d*CaJ+9WFEoe4s%iNeSxesS_=Q5^?kx<uq(
zc6GFcL`*%wXXxTN%5$rfZ{D#!h$Lu7oa72)k&s%G#ZNyYeP9-?B79OP>DZPFiNLIj
zR+?(H&cdT*KKb+mRyUSG+F^`LHp43~y};_yGK-5#w3`7+8&c8`MFu~eWoq9HD|(T;
z4{kFyFiN2?fRu(PXtTAm#r;S3+1lAg3rQwdz{oVMKrraKw#{LTjk%Q}1D(iQ#ND{~
z;qPse*bi5tRg%#4a;xm;m2j>)aS%yTE@;&2F<(ZqZ+4ak#cd*`P?75B6WaEk?)+S(
z6e<=ijiMy6l4$LAD$A#}ddciY8iY(U`od)@OPlQ7d1%pLjMb*>6dVQ*m)7bQtDk<#
zf#G3{R=CDPJFcB9#THoB4^h+D3z&Pjz`O7LkcaaR(ec!F|3E*-4jp1)et{s?e6sQ~
z3fFa+*?)+YrFj-t7C1P40D(&ow$L$od3}AA#l?B{8a0IDk?$WN-#3b%@<16>e28lk
z3yhPr*W)fgEE&@++ziH8q}2|xbe$%;Q*>P?2>S*5m{>^~)oJOCjE%8gi3>Lsb=o*j
zv?hDo`cV|~uq25`XlKCSR|M4!4)rrSF+|vIf+NUHPqFWnODz4xySP!T3?QY|$21n9
zcOuF5%?D%;-eC0dCCtDuG93^2#O+q$iI}ry&w=H+4X$6i&VTyz|H$0p+!M&WMKr(>
zmaN|2H^9#B7HEfwksPk$lg{PrGbMZ+KgG`O7TY^JR2x;w)f#8^9mkc@c8`SUIj`*#
zn{;40GzvrKmvHGAz9;)pC&N@ibQ1VpRMucRQ`<wsLqzooN|_ju)d*vgLlN`0q_JFS
z9of6VUh;yd+W~%Rw2CjXJYVv)Z#>Jy#3Wm)!QrpH#)JRxN0>%{D=b>uHV_JfkBchR
znE(EdIXN_doR|R;-h79gg;w^wb6p?d34(^^?w$MGzIBVAe)KMn<{z?GuXl=LUwFL!
z!VJ}Fg-puo)wy{eqas?Z8f~wR$!hv?eJC_5i;EP?yNpa7U~p^(r6Wrc67fJ{&-?4;
z#+xMG@ky2>bWvQ$Yi&B2Qc@0fx@@uJnNo&UD99LyRAMNgIvqJ1>tvu*Bc#T00pVKo
z_9?C}l+<-#Neyii;d?GaCyz7v{37)qeTWV%gc-$_MGgWZBV3`7kznV>eJU@lFg|@8
z<!c;bbiuKL!9<3~cOTKHwOHO*V7s(MKA%Uako^Z|dHb!e^1+86u)9~G9VUqc)T$xZ
zZ$IYL$uso#_cJy*2~u+Z-hI|@U8P(q5j2}11-ZTgnZ7(JnL<d#jYsp$Z+@FsUV4e=
zE}bWvP1#Ad_>Ic=u-DKzaJ8|mCs%f-OrtoVRq?~qt2-lvHFwuW$8w@^rELVh<1sce
z%0K#t{{SgH_8&UT{W~}K{eSnbX@wCwQV`3hBq)-OG*K%c3Icn`BwP`pI>Qv8QBlY*
zyF`y$h1ME{CNnh5?3*vKw!TWVQ$q@u&?xNOvp@@jbRt}1sLgM(dH)Wxr!Jz?*;wo|
zma36fsC8?7ligy8?b0Uq9^b)n9WuU;<7&S7t*=uo7TMg{q}h&9EDS+lcsRGlD=!E1
z_YX2Ly$>w}H}BkLceg~NQAf1{T0slP_ekgSWW5ZIV|cW(PVvva%gZl5&zV!la2$ct
z;j4*Gd}7uW2J94=<KP(CNdO{k2Pc%ahk7TeGI0zX0Y)K+y?c#7;*j=J9GIPD<{!<H
z&SW`w_z-_k-DPWYi71ThBrXaeV@C@OI*5qc0V=WtJQGItfRHkt*aGVd`gF{~WMbtL
z-&ZEmxc!41c>QIz*47Ei^?0DbGQPYx?!wV{O0jX{9{p!eFnRbS_#o(_b24@)sqR)-
zU)f@9eTCciZz3ImCp}zG@X9MMu)Dn#Yd(c2t#E{-Qf{Fz<O_XF&(5+izrgz+zmMno
zc%BPNu~Xb;eQT2lO(vbkPkHRt+kE=jO)g)0hW<hpX>9T=j0DHT#ac|M5TI536!f&@
zjDbiAl*XO`bT3>PAC4xj!aGyaM&bwuVGR3c_8}aHix)5QlbsD(Q4<x}y=%E#F$!eB
zD2)n2OYDzg?LuOzf^aPtrw1u`+RN_Igzg~_j>FJ}3se_3+57w!T4@|(6P4sOI2c4z
zvHrm)<R-?5#;3@bTt})1u!;t*bt|Zqn>@V#knjA}pYnMAAqFeDyl;AncB^f*zR(D5
zrG3)bG*j7GR+pA>oitOUQ>aLFx)O~>gQevawu{@eRYb0Dh{C``+=Ve6Yd0Y^M#Wcn
z{P8()mq;t!`9zcASa(|Oy3>|T67J6GN#K*`qA<`;Lq~*>?JOy!Y*!=x0YX@Ap8Z+f
zj*AdBc?UYuNq^fblSy&n>_Lk<x{{$&N11r`H0wXT+SLpd5@o=u?n<hUR$2PsWA=Tu
zfGG?hbJB7Tg#_ux9UpsMMM`k{`aOR1{tx-?JAX-KuS5rd7O`ZH&=_Gz`B^qL*Kx;v
z`f~$#sWhon#t!SHd@8jHn;Tn{YZZ!>GA}&)Hf}62BZQ5I?jW=FGw%+(SxSw>Qj8Md
zCF)7yV`cJqyV-r7Bz3Oq;0Qr}pr2MywV;l+ZZJxRouoXW&xn{l<2Vw>v9Czk0sE&G
z=P#Dj0nkMZq@O}&Jf>f~$nK*>w&zw!piHEF!O}RmPV5)ETciANp6FSH>-e1uLm)7b
zrntSy+@pu|=SL|OcgbaZQrQ$U2M+R2|LK3ncfb2xwl=qEw_-0<z;`5}g8%W~eTU-q
z9<RUl5)%`nIH^3xpSj5XV=c<%GHc5Vb}C0og0KPFVXxNU#*O<tUjB&Xl~uO4%6#dS
z7Z@Gw1EMD&S7W;RMM)bh(c4NKAv&!wWiRApx<>;+Bv8l>oDcTByRJ(<pXbn_Lktg%
zQ5YCzaBztG_wG<C7YS5|r(Ao;$DEERQncIcPOGkSU`UD9AtomMhT$)t)b$RAL}6sx
ztBynd@GOU4e3plQ`JP2kbQhf!M#g@n0ud;RcODUZt%f%|jMr%yS#E<00<L`aDWgM^
z)b{qM)%F-09YCS@z2Ex>v|3GW+_+Ay*0RJ^VozJ+@E`x^9US5F_Sar#czA%p;c?Df
z>ZejEv$eIw>f!=@jwF>z5$OP-CCyfwD_3u^u=X*lt7~k%@&+%waFJ}*#inuQ2_6k}
z+nOqoo*_|!Cq|1kXoYa%@0B*zDiWOvu+u&lkWOcqo}T9P>1Xh~G=n3fly{5#$KU@q
zG=k8gH>%rajZ{Rt-6m+aJBUGo)EQ}wvH?3#pr1}sXUXDm?3E!f5jihI>i8k{zjTqg
zzxo+Mo8I&V5IE9SK&sSW{l+bhJ@+O~p@5sb|Cmdn0>#a%H<0)&uPxGO*BKnJ`|j`k
z_BUxZo7}s%MWYp<tj;K|E4g-Ujvu{$g>Qa+n0#S?eKWIosSJxtODwD`u(`H|=ei6G
z4U$WFc9=seKD&C8cRyNSZS8GdedQHqX2ubOSjiZTkx9`KH|65X)DE-6mVWVmP_gre
zD@3<_A{-(_d@l9ITaZqrIeO$UZ@=|*d_Tk3<Rre-{MC2=gdpq+4;yPsfr=DCt0&${
z+2Rh64iJmHMp5`Hqgy5IK@lVLMjJ#r$G+EIVrOZM%Epd8w~awz)%dkAq$D&qwpqS@
zozc^0=o=Wd2wglZBc+2B4tupd7Ux!Y@bDfF=kDRS4xpKtnd0=R<6OUbol3Jsq%{&j
ztr~!o3=fT<wPya&V;(-dPcECsbp-WlmHFibisfA#FHJU^p&l3>&8=|e^ik5DYvndn
zY~l`(bYMgxS!6mWsfder+kLUW_jYFNpX-?8njqD&c(Kxqjtnz7F^L=<<?+qW2or&6
zVKo<%>3AI{?@<^M1z{KY6<*wm|GQw+*>lWGCRn9F_$em7^fJ|@Z7OriU`&U9iU4U0
zeh8bN-k@)GAALik<V-(E2OSR?xUPqk4q6*l9<6fa+9&+^_x{vQwhP11&;Xv}(rh+}
z=TCuvp`js470L4Y3jR!nl%Izv!gYOgv&s7UCd+F}G}>*_xqkWvC-74qM#b%OjFmI#
zB-33vqNJOmb;6sKF}1=Xs>IQK619&%qgeg~DHs|ZW^XejQkpRAcGfy}m+^kGojlda
zCz9m-?`-;&1htCWLP-MW4Gl2;r3;jHia5pUlWhboafHAT8X`$~Zi%4Xw8Ii^Oq#^-
z=>Bc88G%-Hq!=NSN^|(YEa|+@KmS+%f^xaZ=0=IqZVeULgV%K=Y0pEU*sj+3Pv3cm
zZ~x_w$YfJwvuU!~EQ5ms3=a)6Haf^)UmsFy);Fs>UfSXA!#S4McL*ZI4}SDl?3RjD
zDn;IW<5gy7Cy`R4RF95OPf};Fl?ni1hzb+IV*B1Z_8_rpcs4&yF5gc=w6?1|4P*sK
z4X&GFYUU7MdHXkb_a{H3SlT7!k3Z3pOMWS(2!f#FF=vrNTlH&W2*Y0<)=l_=U<jkI
z+e-F4M$eyQ<MBN8Tl4Yv+Cxzq2cZ>iJkZfx+o4w6rZ96Lo_@8+()#K$*|eZh--f|a
zB!cP5afGM%`ZwR^!uiYm&;RqkXKQPhplxLWq!46W7o=vhQsaO6gFoXx{pt4@7%b4=
zSD?^eV0d_#k>L@B21d#F9wt(hx3^ec-{toGN8EX^LZcn=M}P8%6nA$im3DaZjW01U
zlt&01Q{P0V4VV14izWDKV~D~K)sr;aN?Uhg9%@5ja018od!#e$o^ax^cC8e`lcckK
zJagd%);2aMRco}`O^7cbrIVJd>bMUCLC|w3Xu?o;o+15|kUFiBp(H~Tw+Mw0n3T`d
z<!9J_I7j2*3aEsyV&g7jAyy}X+T&H4)e5Qcu^u9_puD?F)NWARUSoJ*hOpUUYJ3DE
zHGlsf{w}jK2l%)D?%z_}u2|J|G@cZsT!)>#I{&x-*T3h7Kl&L*jvQosVuV6pABDbt
zj-Hv|@QF5!T9qgW*{hamG$R(5i`=*~&-(TrYa1IBixpn~(o38^eH3~S@ZRaej#LeT
z#O_W-QHSGZv_WAf(xtRh|B*@I`00-KZ4z_Wb#PseY&J_S(@(xI#0#%{nZ=bAO4T}{
z3VLF!J24K!u<H@3ZIKx15EI$`pKx}56)VOqEqaY1j3T5XDeRwN|Lf0j=i5IZXeqP+
zouCW?T*u%j*tv0^>YM8fO&&rfjssHQIu21g;_i(*G^%YL&fmvz1?g0V>3w6o_12g9
z@S~5|tyE}7_A~Sy$@Lrae0udRUw-WshDJs)LNGVKz=NeZ)>c<()@t+(^phLv?;xXB
zZr<U;>s$QVTW|BFmtUfPAR7;tXb?I^;1jxpJ*>J`U~wx}+aw-iI{atib3n!Xs1Cw}
zAeYJV+~sHat>656<ogC`wd(v||Icf-%MdA~Yp`jsJukxE<UQ=}GvPWQgeBWQr9J&c
zSJ$o)PzGmUh}kc{!q(;=6V(Fjmo-K>)_+_I#9obsk3VDh>=}ker!cP5lhC^eWB32u
z;{`tW_<fq~2A=1T%V&80`Aa;yf1kawRXdPEP^$$vzR%Rm49g1(e0Kd4(q0BX<)MOr
zm9-UCHdZmhA)C$8hy?SC8=O9Mgp}_R#y|*8H!)9&Gaa|!leF0wNn}-YdhTdCIg0g$
zP6p~MPgE<#=*Tep_RUbwX4zW4-*JdhI<kmf%r}pElJ`WI7~wzl7W)@V>Wp<Sut}TM
zMzgzFGJ5<3GtVqjeg7&#$A}afg~SMfk2S_^uWZxU+9XpL!jT%^wS9NjbC`Q@o6U`7
z2K%QO8yzEHjLR=vVtQ(d<;BlgTV5jyBC;8`tKp*{h!la+xK4!U2pa8(-Etivb`h))
zDMdRnolB)1TAQdihX)`~A)j6UjApY*5QKd7tFLkN=uCWZ+JoQP*f^L(K+!msTciyl
zCem7w!NeWFpp_t#8)SHV5+S|#-%7N~EGgEBCvl8XNMLMeoM)eXkw_^XKe|skQ?V_$
zc+#~y4i1PeY~L%vpp+&Gn!oe|VmCx=d>o4g8#K}woPmChedz_tOB)2GmaRO6o!g3#
zq_Go^y4I#Nzrfh33kYE+sD$wFe3!MAN35;PF*3A|bSlO1<A)g<&G6c*Ut(*s%*xU#
z^{Q2hPdiqu(=mcDQnVt4(+X*aic+<S5L=zg#Ug0n`98kuQE#@W)&s&=%EDl%w;KHL
zC*P-DuhD9^c<Ze<7#r(HDQWFn1UOF8^0ETC8fjx~i1{%#fi^K`Ceom_L*Kv%gTtdX
z`3HNDn?$S5%D|Yo($A){oH%}lAZYXP$3J6Zdz-=jK6G5w5m$K#d)X?T49~<diV>4A
z2%i|n`EOpv+5{zsmrY8Qz2vpP>F?*j>(5hJ-=SG*km!4XgVqAiv4>Q+TcxtT%HaMZ
zXwTp{_Wc)kHi+6K%H?&21|}I98sXr9eVl&g1YiBiSNQ(--(!2TM6(g#NGs`*a;(I8
zONH#z3?E;Az^At!A~4ul4vFKq<g!_YhX(1-7uek`F~3%1r(7Y7EbZa(!bASUAO1V4
zdnM|%SGoNBIXetuyW>cipsjX~>-hX4vd@i8wt+-|Jc=y&Ub>8qO_9mwI~9KXD0U9)
zWY{K`EpY10MZW%xD%Y=n&ThFxCY9>Q*;vk2w<Q+@LHF-t2TrWJO2DrQ$1yz|QDY2Y
z7?SpVa9u_(oM!mmBO2Eq*rYygSqtOf3WXySs@$M5zlgf{5^gN!ixBLU%QR~hcDB|?
zIsJsK7E=?W436~kFaG7f;N5rMV`X`rcFS7sIRa9y!^Td9|KtDVU-RAXz02Vv2N)b4
zq|jGjXzCE-vqx#x%IHW@t?p54G+18U;mY-gtZaSG%Gw&+TV=ld<}2*mH;x@Lqm#(@
z>V#RbU1>Q{QG_vZ60huDwf2AzM3C{bogzIcXg$v(l}a-{Iz}cpfSaD-tKax7Zr;2~
zGiZ{=>kPj|VMq`Jox;2q!BMIOLfQiEY3%9$rj;c15U2>pbs0K+f_=}QW9i*b5n9^x
zG(L14$J&DJJzSzRKhOBV(>SrqjdZM@?e_W>bNA=i+FWC0W0`^e0_m*BZ~fLcc=+%U
zTg4*n(4wX(*WqU$-r<ceJ>Z4sE;2E_58-+|o||Xo@iN;R>!j0Z@&kRijzk*4r#J8L
z{*`rp^Vh%0mtKB}LLr4AZg0jNtVAyliFKF2^5|5Y#7Ej#EzV@nP)M|}=d}>s1U{Wg
zbK%@Me&d_JN#DRQ8|y3lKmYsx6RmcHj=f#M?xiRS37V~L@*XF3!nFWV7<-OCjrE0+
zIs*vl*!EYa#SjNAn`P$mdDb4S5U%c8Xj@wCEv+p&;UH<2Ds11q$LQf>xNaJ}jMXV}
z(gaFSZ!{=Zc1Wet9N2%5bLY-ulwfgTfti_oY;A2&tyXOX8$=x<qSl%qjKDl;L6RUS
zp|<1q$-&qiYV9UMVH9!u-c5Ypwbf(#YwVw$L8%afiZ6JJ(3zwy6H6~8Z7(SS$3aV*
z^?Ru_qZ8A(u4m;&5S`bTm{&WF8+#2Zf_8vFFflg4<!4_+MT&*RM`V4UOv<r|V`7dj
zlFQ6_;!sEs9rYC2ugS8SL-BN<FrYO;dh{QdW#-ap9{>0n(j;@khR_IdF`-sRnvGj`
zIr7F=@iS?Jn*!G--#^T)Yo9SVG{XG+BMP}OCdMZ?fB87WgCl(Svrq9nm$A`d_V)H{
z%wv&O;%V3*h!9?nBY=oGa7qzqLABjPsagyki@z50b{O*Er$58<9HfMA{`%Vt3=E)j
zh~++52_Jj8%h;3E{$<@s(g4>*YX@Z@lj~=AYzoJ5i6YZE2$GmPj*AqI6&a2!Zz^Bt
z=j`cos7P_;>c?#F?Bcr;D|28vz_$>@t!+aTMV;0`C;mMN#eeIv7gc=0h**8iSe~DZ
z1!Kn!A7%RdaT@R6v|_VyyCgu85ds$%T8f=J_n5i(0@C+v#qM|%`iHo4?H=`7jb_Ut
zEQc~Py!6rwH0uq@yW0fqHu-ESCM~NNowBX4wo$kkq)J+!019IQw#yA3EN{o34gCCl
zs4#5oZ1E@G{ZspyI`GnqmoT1{l;~)|3CB8J;KqEZICDWDowy|vOE@Vc!^8W?_Yb1A
z)#B51k$Nj<;$e)f-cg3p!7-k{{4z!x9?#vUzmSgybRv>L8~ahmXO8N&+!G2$)NcLK
zvHqKuHkRxiKM+=82N#S+W-{!5{dsnm*J*85Y*J$<Y5;*N1TuosqXkr}g?3Z1mZqRo
z+NM%mrd(QMe5Am^{rl;!!QcCx-)3-NgvI$)Dy13%LCTkS(wglxqmYeqjURt>gP(kI
z!zSN=6QiJgg#yEagESjWmNtrPm-c9fc7NQseVtmZLbY1r?YF<e(W5hV*aQ<7Zrvo_
ziRD^|tdoO9&cFy0bEQBFN&nzDef`576s@;lOXN?akW5U@^7@;<h8B{|?QJGTM^N#A
zOwY(2N<{=gS8Gzmr23%UdIB~2+gi4GOb};F<?&MX+@5*qS+*Z7P?=x1T^s`vi8K;d
z0?mlhgE`b28c#SRc6K{k>#Qz6rrm541rZ|yeYhFHKmMoxm`bI_<Hz$f>H*q-8zb{`
zD_i{U{}2B?CywrC-^?WagZ&Kj_cJhYh>@uS)@`C*XJ@y>!tyqsT)EHMc9pfYb+$Hl
zdF!jMGd?l&L<c5G;$7FZY8SD#o(e;Z348dBvGRe@I9?Xvbgd3S;JPlp?=w6!Ord{-
zf#EUUc<Wn~OGSdPjfyl9I)z!1yeC5AQc6^0lgu!9dQ^%p%&BwYimSJZlL8_fQj-(R
zJbRLd+wVhYc~1aZOI&F!w^SIh^7%E6zxf7!z7HuSY1&+Q_j?qE2QWf&|NbqG9C`+s
z9wKOK1d<?7%*;$9rJ`8eCTQChNXql#j5;J#_Hj(~Xp(ku6ltPPM>x^R5l;wyrqgNL
z(g*@>-oDP*=qM>a&2N0`ZH7k%<77a0$hIUrjNXZ;<llE(gto%~;}iQxrBX3Rs!NLP
zg<F-f967DEH6f2899J?nHqFHg&(mzSn16haOvajJBf6^8NsBOP(foXi_Lp5rvgBCt
zqz<=07=t4nR4UEBOXpd=^9ZxChcPY)jn)Ez)3FmP-hag2@)Cv7QH0~+JA$Q0cPOr{
z;J6M~uU_HxFa0`shDfLKM3JOcZF2D70m|h9SFc{7)e6uNc#buHCa@faj#z6?%xxmy
z6LVoTl9HGMq|<5p*@qz?e)<8!!^5P~8NT)FZ;{FPoz%}tR#^9su4a`^O8xkt@X&zo
zr5K->eqws^Nhsg0Fy7p*jA%BSWb*|wnF41{U!dJ;bLGm%)WXISB9^`JPFlT5TekZg
zKAqJmJe+PT5^OtO3PF_enR($78xNK+8)YzVY@cV*Sw~p4{MGBXIsJ_y+5SF+o5pb+
zw%1l^w?o1xWMSa}1O1as^v^IpK25dOW^b>`OE101wQHZVy<H><H7Jee`<C0N6hUN@
zX(1EMD68uhsJLY+M9lROPwa!lLv?3&iy!>>FUjRHWV3lLoIhiA0DsN~F5%PsHQ5pA
zAcTifn*M=728V{@&&PDWr?;J}W3)>tMXS{$U+8CKWSmRSzDT<rF!%UA*_4BdlA)UJ
zbL%<|fqIg5(Ge{EmGxhfqCI}E#-g<do9qnin`QrV&rtuXkC8EVNnwbHyD|v&=9Xz}
zZIa6OAzcsQ`sDkDczo{$rQKbI2AYhHj&a~bKgW-q;Of;IG#fPv`7FLC?eu>XVYEPL
zK@dc^=uVO#gut{Fk!rEBQ$;;mvHJq?#GX?|Gq=3RAAjfDAi($E;_#tEabl%#U0bM_
zIEgp0mY)3?F^>@$YljMQg%JivClFY*gl>@@OJ@lOZ4^-ykdhw9kDf$D5g&f^9-BKw
zJSk%V;-uR`hlZ1!;|5SsL>L5rH|<wr;z7F%7OjskcwhwH&>%<Oc#+cjpQBm{Z%v|%
z#Fthae(&)zL3Iz$&mdjf;j315SzUU7);Tii6emv};p8(%xOm|L|K@-C5A5#jS$a*{
z#f>+0BMe#Jtx~#j2XXZdT@@dJbY1efJOaV??jDtTlPEryZ{52=tzkLZ-~8rV?B72T
zt5VoP-A-RSIMT60PjP1=>B?zg2<=1^Zo0te#5B_JP*K=Z%-bSH>j)_X`(_UD^{@Xu
ze)8TA*xK19n@(HF8lyU$H66RhKn#h+;6T`JJ~gE73$gQ*3atRKbm-YLIF5_&cz9lx
zv6Dy1&x{gb$KGw5B0gXR_y+do7O8A);Rpv0MfCV8?fakc!4H2(*luz2_6=L{L=HP!
zMM{+lufF;QFTU_HsdSorUmm2fsG;k0reH1iG_lc1CfO_}Ci&Vg?p+(gFd(1H;<+9{
z81dQFPgz=9=A9q?grFVk1sQ`;7K%?=p`D9VI7r7sI==O}l@5bLBcw7Jf*?c%ft^sL
zD^<{e{9|Oz7(=sVMKo<C=`b-d&9l!w&*<1BwOXAp2<^eIyYd-bNd@yn0`N4Fx~|c<
z9S!fqT?#zM#dA_*M@E^ta2(}>R`E9&+a_{^mEE9RW$oTQFa}o${Msg!YwvRXy>}=V
zx7pa*pj_ES;87{nSYBS`)akRl_~OfWo=2fCA1BiS$8|azK$%1xF*e-pKw8d!G3H5f
zP{`+Tq(h_G<fBhMV199d_uv1Zb6A_WwQAxPjOp-6v4?`=#+q9`I6lK8<2asg`Bz~W
zd!2SC<r2+1tzwi;D;kZOMWc)$lgV@T>;;Y=KSMhRX*OFOX`6&xo!A!jN-gvjSx;k0
zVTDwk*ps*S-*p`qSNh~e$Cx;O6y?W6daUe+F;*B{2tj*olg-DEY}^E-TRSX${CyT~
zUBei~wd+?%rP2tIVs&MMdOhIi(Gz4cY5M!~7-Mbn3_bQd#(L&j>4{_-)gos~cc1&-
zy^+K`8BEd!R@c|~<jTi<`0>XqE-rQWlbxX_)7|5}@w494E=IbfvwaMYOm-+0i5x?3
z0+=AFLCmFT)a$Vmg=A!Sf=kc7z{L18rE-~e(6$mq2^R=Rtj2nawx=2EU$l~92$D8b
zZxJt0p2y^+bM);Sq9Y0~3|e&^Q0m)7HgDa=69(b<xUS3Uqr0>k4eHGrYio<7QaMJ4
zr|`Wj<#LU~hYxY+(0-yQpxq7#0@XoQBBco;O%NGE1wmv8B26n)ogybOhwojfR0@E_
zl?8tE-VgcVJMXi;v4!ipxUK~rjCoS3syF6#y4FKDY0|lV#>S_*!*V|t10m?}Alt1r
zK@i}&9><TK=GhA`5Codt-Q5_i=`9juXXwVZ)-9h>Me0`?=f7zs9T%|lAVHkeqY=_&
z`1nz#pE-^nCtpTrLL+c-aD_vCZHw~!f-St<G{VcEa1f5qqeu5irF;$?IKYc9yvSa8
zk4C-9^yD~wg)}M8f)@7tw#j@DMg*av6)4)Frr8cDR+`MMZt-AgjoqrXk4)mIfraHo
z{`fopneYDP_u1Sm;d?1u*R_+Uu2tQjBMW3*CJ<I)M!Ft2K6n{MC->v|DH6LgNX`-~
z!=|(%3<5mqapL$XUU}&aT7jlms>Jttd|$-LyG#b1!jLd%({45hgI0F}<Ck1tFsX}>
z(#3N#U>p*C9w|H=De+P%(nF(6pE`z4$0p7~6WXi?Aq<`Ysx_+f%Xnz|%gcE8K4z@8
zi)l4!*7opR$)g9i5K6GKwn?+K$JF!;m!JO<-}uJgXLw|s*#rB?_hms^zgdGl*p0FL
z81ZwLTvwyWuHKE3RQW1sGch)d=ebn&_PBQA3TtaSeE#`$bfn_TJx(4Y6<^p`L0#9w
zxE{D^wDduG<njfgFk=4EV;<eSO>MV`(NSl*CW<x5AQ?VU>{WMJ-&mrwTc+NufspLm
zcaV#hUZ8Jagmyb5ifn~$dRo5;g46NS<LM-IItZ+Ar|081DeONf02fEvL($9T89#cA
z%=kb|oEB&lfssg|ae;KCDLs6IXe+X+PUi77oV%aV53H}Q(BD_!`qfWJNsrQIiE4F+
ziK%JMpMQ?O|M&k1BcoGH?VF%5ki&7^&IV2{b_|H0yX^kUF^^5uqk!?zVH_#h+S%sT
z-J9HdFwcVrk4drxqYPRZ`wW;k=8kJ)?)V7T9$0=lLpqbTlbg5hu>N?Su+@k!iNpdd
z)+$VfV@$->&N8J^(I)LtNWL(@`3ui+<k%^~$PfnhnKAKapKxq?<Nfm!yT8m5fN6(B
zs*P|2u9J$BH2{e#Jv`UN%jB3ib&Blx0Fg14mVhQO4pL}51b#DM=l%m6CCKOiar;AZ
z53h6g^Xs%4HEN9-J6mh0CX{z}Pz1d2(yRQVfAqg#-^>w)$41EaXYtY=TExU`Yyt-|
z)~-tcE760}|Kga3K^srN&_F*%8}8k|%hJj+@BZXNYSp@xn9yL9!9*sum9mp@I3Bn*
z-nbc*kPHkBg3xSiY_PSkL|Cg~(DA;EweRAVO&D4lM|p3PX0v9KJfj((nC8-RuP`<_
zO{*0WMvCrWrRcP3Ivk&5Z|Yvn%cr{}!an=NJTkfko$I*xj!$M_nCVNW&}k7DbH?sP
zVNokTRIFaVNfcOZGrzJ*>C+!_^}TnAqJZl+uaff9w40iRg+*p(5AnhauaM2<$mTNW
z*ipp}rkPknH0Bf<ThJL{FXb-p?8yQyc@_%!JjMv_Kf2G2+c)^pkA6bE(Toe#q-e89
zO?MJjBAwWm*!2-^ijmPtgyRqeA*P4^V2n+^lR=j#if9KdTI~i}gX^X^e*84&&Rs%j
zO{3YgQ<8D=n9zuN(f6eF`FD}jL3bk4ab(ZgVh2XZ<k<h>1)P3Ac~Gs|L0r5=^#&`S
zUqKp$bX?M=&CU1zim2UWacPcr&?J|!owz6z3=R%4GdqnCnpUe#6n51kLe)*=BP|Fb
zLpw4Z*%2YS6Q)Vbv)L>{O6Hdqc=xA2=KDW*m%Y6`T-WUsTFJib#!z?LvrY<}G-DGp
zq|=!mq~3luHkY+~Its&>iyz=PKF5!p<-+;rXg1nZDwWRbOcMA+KBq&IQtGMP)t$I~
zk#813i0h>AT(48TCly`Nqee;!2WJ^Rb^sGA&nKJ4#a?Ef3Mf9hO={sP;?7?&K$XK|
zd8TEZ>Q<36M-TJp(Jfky8s%b*dZ|jaTx55*$eVAz#kaosI~+fDmdU9J`@zJCQJg!*
zWO?z#WtT+5#O>3+R$;|eeKwn6c6J&e42w(i?3T8;cI_s+yJh<T#HdZz9>@}Fjd0Ld
zQ)+8>>{@|R1tWt)<oo&%LeOe7Sl?K+35AX<M=Fs%QJQ+A&f?+%8=D(!Zf<p^WF{wO
z7#bY40=25^I?#In^iJqLtqa~%$_Op-oHVZE^%N7`i^lUj$Yt60>^Y=$H3nh<Ut=XB
z90~Q61@!u3{M$dk-@3*Akpkz&Qxun$m>nIZy0^{d<}%G%KxMl^wN#{9-Q)a)OZ?97
z{8LVzILD#G2Qio@Jb;t<Aw;JIZjy}mzY=ptN(zM>)6)}Z9r5ttJ!-WIAARsC^?Iu(
z=G_DTiI@uqBV9`|a8eejFjo282+(1~UZqTBuZ-3aO4&(e&r5iehLvS2th~IuM6Fgs
zYfWEYKl}F|v7VJt_}BQ7y$3``W9unh@<wZ-QV~Lp>!k7A6m<BAiFmBU^HVta98>2{
zBeO}|ZEIQ@i`=;wD)Wz#r46!=e?npBAyYYr%SXmoe*BoJ;X!WP_ykwL=E^qpas?fQ
zWOD`n<-hvha`ea<&YwR|KA*FQWqjtCgie9ziF5bh?7Wv>um_|R3=a)3FwjS%+2F>_
zYm~|*KK<lsk33Nq8SAv3V|fNG9gJ`R56{h@l|;8A%DcPNt5w2wi(0Kp6oo{Qm6=J}
zzFKP*7w38K-~sas^VDiJKrlHq&5^?=K{!M`E%)Bnkesoy2dVv)bzhZ2w^}HzaJ@8A
z#Cu;zr0{SZ2S1Z#<nR&lGh=80MnGf?#t5V^q%_nQmr&&z*|1FZ!6y_}?jxHG8haJ?
zDkVxg8#GEyirbqE3=MGp{AK>(@BA|+re>Kvu#apZgC-t6qg%iwm$Bv7(skDmzo2*v
z15&PJWO#rujJSH^3d^e-eE7kq=qTzb+;p6%#Yw#DV_b{YC_GY`9G>q}E|s`<?FOaw
zb+isU$^@z>$%w*`c2H+^ZJuIrms+Ea=chSy<^reATtq1d!^kG1Ms$kv-Wd1N9{#qM
z00<RIhd8bWxG{($@f@F&m!fZa9}{Pf5(<@gNui9ih`bi;+__IwZ{lO%-bX|?KIEex
z{{&+c4<Ft|8jtObZG<cNjo<tSoIih=;n7i286VeoK_<3|!WMCHtFjkGHm3K#yXSov
zv>EIxAf@Eay*u2#cbgx-`vFm?tn8dJ=t!d@WzTg0$3wa)q?<uYk9=Q0>1>utxlD7n
zj8O^S&cqZbdnbfZM7v$5(X0?f8tJ&4I(d$B=Pwg!XtzVlozuyQ)y+$i`$X%fx0k<A
zQWtJ*B2hT8Vs`I`*g13z=xmk)FI>bM%-TPXfm}Z(|E8p7sJ?=~_$lee9@2F<x_^{c
z3oge72Dx<U9Me-ny#JFQq1&*wu*%NnI*W@-{15-bzv0ZOi|n60LaWv6Bsj!B>F4D~
zI?yV;LI1^veLHBA$yhm*nVBh+3V8VFE@7a!apQK(*D#$V0n^0@h&^l(miKn!<`usC
z-9O_8KlmP>Uc18L_69POqF5^O*=L^+ge@|e6uDd;=}20w7NycIQ5Yf}hyMOPGMO}$
zN`>9sBF#oXyRB%o+Js@)kvj;2phxaPb%>Zr75K}p^-n&)30eqga6Pwkn9}nZI<ddy
z;K?J5ojPbe?qbfN8@Ju1fLyywYW@RKn~!mfV&Cip=QD~EuFLb6FLLheF+Ta|eQM=8
z+p9b5Y^<}jxxw%M{vR?nypQKDeTjO#9?MJg3<L-f8#2otCxj<AI)6jVTWuSM!NGne
zC&p;DYRu0+q*AHz@WG>=nCnh5kt~CaH^QOSZt>B_KjVk*e2))4{3-J*OVm_ADxc-y
z!+YGldka71l1}^Nayhip)M|Aa^*WAY?W2Z<hA>7`tyZa4>o_t+v(X}q!Z<;QI&tsD
zTM>p)M;qyBoD3=mSpC^gFr{6bPMe{pP>2swN7}d#oIJ|N$^B@do`}064Xy;Svy8a&
zBizLsxU@NTV4UOaT@L2@c;mH~$fOLPfBF$&)3CF;$?o<Bw{PEJaej^e?)Uzeq*8ed
zI_4X7nq)DTvMUSJi%|cUikct{$Yj&Z?w_Vm$g;J)O0l@jo!bv6mnxm)u(y&YK~iwQ
z%BHivzRH!WpK<&44Jy?={JuQ7!G21mT|WE#6A%jD_sHdPAO!7Ji!cmvT$fxf$3Xu8
zLI|rbs34ykBxu`MCvi@G|9L5+<ZHjG)vlBY*H>|r#CKEH0VJV(#C#&x^)M-)ea~IM
z>GQj1qlpW6DbVE|8jr4$y8d1K(gTi8^mE9mqxb4ObK(fMZ+%MG4%k`SWp8JPQfZgN
zM^5m2zx&TQed;`egZ+e2*z;aR9E&H5_P-vlo+ZHZJVr-`7#<#=R4y_%cc1(BAF;l+
z)k)%$wtGTrF+$p+O}J<yh$2I)8Q=+*;gL}s-y>)?Sz2B|=@4VAjiKwfaoiNED=QR>
z_I)X(@clG1Gl!7Uv$9A%QbkX;_>-jJSKDU)o0d@uwO7F?@V#tDp&{W|I*x;v%`tlN
z7^$%Vl3PA9*diW7xV=fRzD8l~bNW`Ub2KN(i#9TfSYCWU*le)5yh)>4#rIQu^ILzP
zFTL^=&YnBR@aRzN=#UiTNhrH_xCL&0@qKQzCY|z_93LeJ0<K-Z!rc4<_wGKh$$Jv-
z7>P#+jN_uEi*bCEamZx)Fbc}sMOGJ=XxHmBn+?LSO&A6!rQ>0^h<c;S-MhC~SXiL4
zSHbgBoH~7usp<V4+J!Nerjh_Pw*BvRUN286&R=Ly;?es*;KKMQzMH~#{WwiF7I88X
zAyRfDYeWCOeN0`x!1|9qAX2u<at+87oOt#SFFb#Qv>V~#&}g;Erm{>G9A+Cu+NBC-
zpE=6Hz4y3r`wAydoM8U$J*JNwC5&n;EiJNt_5deNoM3&Uh;GG}NyHD7XSKnk(nQ)|
zMBJjn#3bau{^6O}9iY5drN7X}<mfoQ@YvqnpdGcicYlG4&z>ipNn=p4q=mHEiqUo&
zHX$~<9><QKL<&jJYEdfh5k-PVE8xL{dscR()}WB@?+iIK8a1w7|BQ{z4O}<Jp@Sy?
zPeHK0C=~iF>LINlVyn5w<mmIHJ+FfrnYc1ed?bUmCI~fJ*#VVbc3uC@pD=WAhWybp
zr2RBbtj%YHmGf|2zdNnxr8xNF^VC+iFq>uCjv;3xzR?U!<#^?5&oGfyc&<y>XyUmZ
z11XQ^$0aKZ^PD_+h<mpl^Z93=aOuToS-k&{sY8l(dyj{Y?=v+u&5^@L*x4zuS8oIM
zh?i&_*SFQBQt|U<+y7R<{TCmM##lk&TD?goon~fgnk~m;b#sZ6C(m*7&O?r!IELez
zc$(4j9J?K!ZsCwhXE=QLC|YS+%_ha&3Xa!gXSc-a>Jt5Z14Mzskye1S*=kYQEAjmw
zewSu5WMt%h_RSo6q5^F;>r9UiGCMV4pE095hlAGEM>7l+L0b_<nuO2xOE0Z?_`~lr
zF+IWTn_t8CQh3rO<~3Miu<St-{1k^@dY+rB+ptq5a3F^yG@9I?$15)%XTq!DIS%c5
zKwn>h6N711gEpSgoIZVsAAj!(g}wq4!5CrCCOb0D^_zEj<>fb+7@uTndYa9xa_1!q
z?3Ah$0-=<hvX7AjDH2ZYFOGRTum>^*4(~s}`uYZoD~}i+9OvHsIWAo~4??3z5?J?%
zJx*-=ol0kDHtH0&ci5}e(MIxOp^ry%4{6k^v>Giu*T;1|!XRL~S>uz>KVomM&iMFM
zCZ}eBgo|gWly^CNV1n_HAxuYTJfUkCqDT`4iXgDv39X(&tpA2p-`eEvcfQT(fAmif
zY3tU}Zd+*;ZLILD>o^D_C>)q&--VN`eQ>=)3GoF4!Z4PXOjRF}_Ob|VnA$hSYdb|+
zexB*`r`RcOapjZGdFJ9XEYHm`JRS1g@4dtGmtSCFe1<!Bu3JuJ&!ujs$4%TGwuuJB
ziP$f=7mX3r>orD3hnSt2W^Hwi`NfBvICh%r*KadA`iezijg=3IH@Yp-jYR>2FyP9S
z&*H*KAf(I1ix&~?IyY}zV`TVE(&;prOqQ+f4YszoVr{=PgF{m^TMdfEZK6oAv%Q51
z%1n;fA}(py_9A{k5O$J*zl#+%8m#^7T?W4KTez;YE+8iC@a$}xRpJZ|u>YltEPn5&
z5ZZ#u$IzC7RI7@)_Y=HQ3Xe9YX9jq<@_>WqUuUaU<&%%!<MO4K*;-yh3Xl8u=XvGj
zuOfuY>hdyS7<Kk=M~J!8Vtq1!|M|EZ436W_KhTHkNp^NO*xX#_^Utnx=-@P-pNgG>
zL<fOJSXB%J)M_PGR+h<RGWd>9E}z3or}+HSk2rAf0O?eUTrLaS(5Tl*r8Br%lgZdt
zZKKgZ;Np8Zf_57jcUgLi@+VrbPq9GzLP_1?JMS|&d4R&jvk2c-NRGBVmxTL(5LT)o
z<uP&TG~xCJd)Mw`M99EwAJ4sVk^LtpF(Lq^iL}OZ93tOmU~rIAjWQqKy3eWK{x!xA
z++uKOfR)W{CdUR@x_^%ohmYW;1UtJGs<j4D6m}+JI)}N|7FIVg?~M5CiMH3zorMB$
zJ%<Aa_mj`|F)=<uv$o6eW5;+r|Cn~Wj%q22+eIe#?X&Gw9igQ)1h!45u8FgBB>8+E
z&vEbu1{fO~#rLx8mWpg_tkPE~&}ufwq_dV-8%FHx>`=(}bLP}}9LML-!Q&h}bdpx9
zh3mRBTTOPh)^J4^o$MvqCWP`x#lE%id;ewE-qITP|MPb^`Hymt_rS9kT4IPSV|fiB
zC85^jrpGw^`Xv^>^AlvVjp=hZd+Gp}Up-HHKp>PM3If}5)|&j_5GUFZiw_^tfBqte
zP9309Y_Phq#pKv9iw_<!bNCP<l|e<;RifQ)6Aw4Wu{WI!f;KU?(gs86WCZpX#N2h5
zncYWUVUUsGexjhxiQ^|&SXv+qnryCxRQIa%4Gh`lf>vm$W6q$#$XK@#@KZkN^bnro
z;!4SZNRv)wxqI&x%|?yBLLbdWqZ6}gb&n_t*gt!S>8V*fFU`egU&i+`G1_IQ*Z0`n
zTCx*>bh$Nl*h1S0-%t@n+O}|>lAe_T8s$Ci{>fjEADX~DydU9NK9@9hn#&O`wiPP~
zL&e~}Ne;aFEX#lSK2E!h$_Nf0n&9QHT&8c#Lnw_36|Q56)P;e5j_vJm<Mu<&y!9%d
zeR`KrM^tJp!m!Pd(VRVb1mA`J{(d_MlUUY4Jd|OzAW?}wu0_*2Ez@5Rb15Ywqa&pJ
z9J!oN#&am-j!~(WY+LBT92d`@M>r1BSZzBawQVn1h|0bOA()t$Amyj<q_kqGGDW>n
zWp#OxzP<vjW)sizky6lXwP-Y|xUSF169@3T3>TicfS=0XNITTiXjG|{H>@?7QBUk|
zr67s|!pO$k`fUH|k~$NF-234V=-aoCiC5mh_kDt(fe;3xB8+tHB`5*oF#W=1YAb8B
z7S>T1Qm)~^(J5Z|>S>0CQ`R2QfG=Ez3Mpnv+XST=XV0DDoxk{qt>Pv@J5Lyf%<dbe
zQYmuu$Wb269pLf&s(tnVfkry^Sr^6-X+XGEGtwl}-#>@i82dWY={!cYIC}6f>zf-?
zszn~pdAwK;$mX(OVrw`lFh*fymzSH(=6T`8SMVK&FbJtN+T^l*T)FZADc_;hXd&Y}
zuwJjTwzkXC>JpxtWovVX$*KJqV{l!UFbqh$nlOqw$-CX#u{2kpXty;g0;MGU3gi4Y
zEfYoD`m=8{GBeG<GnYvFDVnViFr8vv3W*XhdFB+=`Bkd79uW#d-U!^R!}*twa$;1F
zav^LQ1`7pVn$Gh<agEdG&-38+9QW?u;;=eM7_?~$mvXtx?Ce2iW@lMi*#r`>#>8Mv
zT+o<!A7bKj@8@GKDesjT85v=6Y@A)kV|i_kiP346me)CW@Bl`r*ipodJ26%u9Uz^_
z(r=7Kk#~2gHUgZK&(`J|!$ZU5Ckwc)ORLr5-h(wBJ${Jm<~Vxv45@TB1`?oND>2-6
z0Vn41*%rO^hYtfqs~r)A8ub(lv@ev@Ii(sm|LjjVkv62qM`&sV+Uvma)+f$E7zo;J
zvIG4be(QA}l(soOk>SFN$H^9av}z)S^=ZV|2S`duy04ED3Ks6Iaqau>bNuK*9zC3+
z8A2hKVYGjM#Rm^LaQqli*rdO2km12TiaTYZNZ~lWt`Anz4H6;RSiYnK-}kf+rOWfr
zU*^o2Q#`!4!r)*(Mpen@@*F#Slw8`Qu@|y-^*v_$-yrHAM_DndZY#Fu0FVNduy!G#
z))*DiGIj3Ux=yQErQNPmSna2;uMcA&3_~70dc^3+2+dlZ;gJlE3@Mk^`1G^Ss8*{K
zi#vD{zV^mzU0ccc_4l@9dRyX8Mbbn`$~SLwr|>7tzV#-)1RX_qj*E`@fEtZ-BvCsc
zL^FB$BCTQxedii4efb$?_KzamHYV~B+8U^<Fv5|7bYCB%jXD?lo4ofIf64x{r?|d#
zm7UEBQWy@;?q})2V`h#XX17+M&^JInn~rUiLLAwhicDlIh!ed3ddx3z`m~Mt=x9H?
zJ1vF=hrwvFX_v}Qovj-`WB9q33A6q7Z0$h(-3rq33qh-hb`;Ro8lx1oN{I*eZqg1K
ztgbCFJF`EQ<$$fNZQAVyeYrk7De22+KsQ-lUg7bhM^tMyiaVQ}JAIPz!9yK;x!#!f
zwrqMOH-6b=fKW-y>NYq3^B-~ScYcFZ#zSdkO}ephPy9i3q-YDx^z#=9in|22uJPP!
zCpmgz7AIv}98OI4(@}(!f^7c)XQu@VcNUoYe1vo7Pq4JG&U~rJ*;8kzY!uPTkoHs5
z>s1={I(@ke^;$FL5+=T!x<dugIyT|%j7>awP)h0W?6c=NbLJHD57(KP%(GY8VtV`_
zE31aI@36DBMsxKJ`I%#cju#_OF^?}vC_B%Zrq!ww7=u<3Qn=i|cb9s-LKwE#-rith
zVhVtAxy<Iq8iNBvjE{_x&1C8C%TTG*SXx@5QmL@Bv&l0jkAm*&X=Pd@zQdi>U0pP-
z%~NT&|DqKEQKiM)pZp~<n?=lwTd`JUy^*_9zz#vXjTr9d*ss0D!{7fi^0L9Rubkre
z=|jj2v3RurZ9OmhhK6}A^!SVG*O_?pHIAK`VRLf}t!>g?P@3$(5V!B&V{L5%&k-19
zhc{%e@2cs_7j<Q&#1qN96p|Asj&SJEL82DG9{u?Na=8M+L^P^(imUTvj-IfsR--^`
zt4kSEihwwgv{RT8Lm}r=+1+AeZJA6qO{G$zQ0N0-eSIA%;KJEw86TTuVqzZ$4jyN7
zbBji!!PeFW8Q&@?=pG(uGLfI~IxQF7be`wG&7}=gw@TdmgKu*>lSkw|8~?bdlTxDM
zNpY3SaOf*9^0-(;Z*4Ggc#s!fy~y~%K?E8dTHqy>N;5q+#N^{`f|lU;xg+eA+N^Eu
z5UG%%(P0KgM%gazQmxf(ULmXiHCRh89iz61+ehb)>xsEQFg`xYzUe7OhDK@DcQ|$8
zB-=Y%7^9efG{=Gc`!T`+6QPB+y0xM^Ii1aA$mg?oLZXevb2D7M_9<c1qSdSsg#khc
zN~K*|ttQW$xj=vaFvpLbWn^rcFbHTgn-sU#tQh|j?R5vOBf=;`MOFqX8G?Hn>kB1y
z0TJQyB6t7dyX=4aP1NuxAqwerFID46T%cXAlFMi5pPAy7Z@$C<w~6pHQfR_PNIut3
z%Y>MAXe(_rZpx>vLtZ{K$afykQTC5eF0GSI7ueifCtt|ZpUrdo+Eu0w9^`9Z{R&~&
z<o3-Q6t{LQY23AzC7Em*j9{<Uz$9w$m;_}7oQ}An=X<y=JbpZn=jQQ!kKy3~FmUqZ
zY1+*uQP`k%|2BK!F5aojr1l>}`vbIOtoSdiNUcQ(WSm=QTh;1_o$XB)7a!4%nuMX?
z!Na@s_4iRKmuLk6_a8js)~&m|Cx6D*-}*Agj~rlYV~(|zhad$}RLAkWxPteIjdx-w
z*H{v{HXV`Dr**+29m1f&!UsP?7E<KSouQq|;iMdlw3n+g#$N92HmcRa85rc$m(DYO
zBujq*q>NCZ#y39N6|~z;YXEQk-Vm<j=vbPKl|3HJEzxS%(CsGmN`|d+Kl{f<d3fhG
zqtny8{K5+e6Y<eUAF#E)ZQI?R3&M~|r>!1Ry@?5*j5z{iPt5(4kMFxIEG*)=*_c;n
zbrVk<KZ(R43L~o5KBijTBz5L_yvc)TCrju=2&^xnt?Co{Ydn#tLbQr_{P+>OrENkL
z5Qc(#4{kF#IZdtJpjxYQ<;oRQ81ng*EBvqjSO0>bS!Vw64OUlI5dvE63Xy8Xtp!^}
zXsuC^CQ4)+^pn$aPviU-fFpe(6S95pHs*UiQ?I{5lq;Zc!4aMJE?ft~kg(arnV96n
z8<!bAlx2Lx!*wH6pva{9iA+S;Y9Td%#!2~T-Qtx){ru^j>qImA+1{FGU|@{p)m6@(
zI7PX*!(Oe*u_K3>oEqobfBZ)*FD?_bLww&uIN-Q0p5tO#ZQ4;ll5ThSc66mD(y25<
zLw#&)Y~i{o`uhv$s75B6<<P;yfFao1qx|FV;GR5B`uKTNVFZmPkis68PW&*5ZbGkB
z1S({=QsnWYyR^avrE&$wy+bOUqS0v3Xtua|?FKqhT)TFafA&v)k5+w$n>VhoxV!{F
zy<R0avX4Yn0OEpFMe(G(id(T=QvR=MSszo6`sNzf|M-tM{@dR|^!E{o2;s)XbL_CA
zTWz8sWcb8Mj=#OhsZ@iJ$vo0&;HoS_NWyjtsVutY<#P0;N<5Pb`RV%~6X*u*>Mnb(
zU~FWZ)?S_Y2X{GmU_XWaZ}6Qz|2E5utEfofd9IaAkuF+kf*^{0d99O;jk$=e+8~?F
zaq84@w3bB8i0PS0s-+sIPMjueH)vPeZ2#mhm~5BukDev;bJ*KRBZb?cLr7u0ZLJS)
zKomyYx_OmGV~=8K2j9<d`t(^M6;UkivcA5?-8=U<|I8VtCWg6v>qCC>-g{`R*sE4}
z`MDRE9vi~8{}KUl^TagduUw1pzxE6mkJ{WE*Z$}aIP|TrgYRLKP3~ORA&MjtXg3-R
z&CYP>*Ip&};0hO?J48C`Ae2HPt+Z2YGnX#pxiqQx&+q?;{GnqkY~9Dt4KOk|M7>xg
z2p7nY40HC(Gwg0}v5}nduI+?5u7eJ(#8XnV%UDAcY_XHe=6LP3SFDz8WXR_R>CX+3
z&t~vl!N#pCsOb#ek#j^|DyC{^L_8U5ATH8F+L6Mjke$i~^N;S*idx*ee}|FLF^n;6
zZ||_ZwZZOok+Ww{gQ;`l+E2N0;|7gJ9nTSb>+P?D?#k8ZSUBF&3KGBgr#I#Ng_1gr
z!Pkb`!^h0$vrN4CCeENwTEysBY@s5Apsf{d*kZU*VRXnN6>B)@2pkRda*51PAMGHt
znQW^~N=UrXF`5g@{Mx}j{%m)Z*{S_}cI5{Bxe~2bmBagwGBr8P{cE?_S>NWx%P&*h
z-eO{WlAUdv@%Il7GB`Yp6tJ?o!u;GSM%xyd>k2!GqHI-Yc}Z<nSGRfY(klclMJ|^j
zaB{S2Fg`xX{Nn{89r38$<U(ng_QyLk9^WDN(woSk8Kibmz1m0yi{fdG(bjHeZs8Go
z^)d@f^Bg>ImLI+QWB$V*{5}T`9OBfeQ#6_ZmAxtgs8$;YB%b3jGCV?LG_LDe1j^{H
z4Vh9lp$n}9LZW4#RQUQS@G%l2bEgro^xjVy&*jJ;JB~`Ha7aX0ZRH*rO{7Bdtz9O(
zCVgXB+on|tPx>@UyJSWN@jMUH4hULpe8<7-A7ZDm$tx2XsvD~~hYs-<KYWkeW|>C2
zMr&`DqsNZ%=+R>~R@Qmune(izEHgAXNVQU>*=&&S?`L>)1lM)h*jVS$!$nK-^*tO{
z+QT@e%qV3Df;KBF8(e<&6@peoCYwf>JSK#R$!S)XHVAdZqh^C+TMya$$p&)YF|sed
zj*}ThV3S%%28Ci?M~u=n8x7{>A5q>bvaz+z%)aCNKmOnUcT^;J?X}l%T#xOYlJ${Q
zI#wSS6mkVdM@OwqS3G<Zp*v*mB!RcGHgO};bP3*1O{mluA`DKX*tv2ImCrEr{EI|>
z3QxElD=i_dv3V4QWP=J*={5uV`fNK)X<Xq`FBZv;4xk;0(291mg)b!oBjW_iYyA5D
zJb&3JF*-cPgZpy~4OeJ3_BeO?9L=E3XYYN$=?fQl_QJDZ6llrr?hbxB&B*vDxm<y2
zwao2X579vsd#%RP`H=;-l{U147LOk<ap~EY3A~VW+9#jO;kqtkV>9e-*Qx1<mC7E4
zyYJChctHBx3pmH0!PhyA(-mH~n<KVRQ`YupVc{{QYLVsDW%kdW;?KVGr~LkZ`~yy%
zI*sFctZx;GB1NTICkz!5VfniglLTRi=lNC?9b3KI!(S6c_W8ib&LxmsM8B#f5LTYZ
zM6~7>c>II!Gy2wBh<=~67uzICtM#RHh*oX#ropk3V-(uu*ipfwUEU)#*k`9xBSjFj
z@P$AQ3~?+d(I{8AGrW(v`8jI4HJV|IlZTGuJ85ox`WeU1p5@tRpTp=630e`2MioDu
zW^`hll%HZ@afyxfB2tK$V`=O@H!@E8THL++fHS9_rO+3V%Xzfx0aMe{tgfuE7gbp;
z?Qrz-?@^t*OZME$$k`*fq)DJ6i_}`QkSM|^D}cYeI?wv{I)j7LeE*&A@mD|mJ_imQ
zpdBb~+`5l3hLoS;*WP{`SHRTdIFVAMGiik5Sc@&It8L>PbD=RZo|Nwn3jA%aL`@*l
zA+_r_kOiNCFTI8<eLUZ_+KE!&NI|>Zz@eYh$Ho{RI|({K2#0X5MykI+q$8x#s3;_p
zPNQ>8-kK78zqCmv>$A1GL}WB44jsePVSf7_r_Y?Gw7Y{8*J(9cL^>p$%`!4NN~_i8
z@#7^3ZTCewCf*w&?#e~nyLX>M2TqaB2>J>Rj`5k9nPIQkBG@al_tE!|3-`#LehInn
z5JLJ^i&iF+7ZQ}U0n|~*gNJw6DQ(dT6<gaS7MB-EXR_3rZGQIA=NN4@5}$kaB06l)
z*O$Rh+4CX7;2IGR+gQ4l-OowUrk?6v{(_DY)>Xg_pm_Bj-tYv|FI@!RBN5YcVx-DZ
z0Yi;lCRL4;R5XpYlXQ-Qjx_B`4c~R~(>`I?L`TZ<wH!%qu#ctXBF_!xQ4NpnnM1U;
z*V&z4qP$yXWNDP+M^CW1y3NYmG8a#uBR4e2*5)?PJ^v#4{z2Y*@26b5`U#av71wh~
zIWE)trs(U-69xfNNLsBnj~>mD%jNm<8($}#>ZevK($|-!E+T5>J&b~OJ*3^NBXghw
zMVcCB;}*^*5zgheX^rj2#U<u$qA(rV2Q4Jk>K<!r%M^=Sw3;oVNHagT$nwexyStUn
zM5pDa^|aND6(&~6L9TUA#oqnCzb=j{pZMA@v#jo)Kxm2BZLs*`k2y9lgwCaL-4sD!
zjh&?d5M+ZE6V(dCP8%g5)CSLS>>;5H?VU2&p#j=}Cmo`8fGZ{WfdcC*JDkjGa!#As
z$yuUmnd<y9cZz$gZLDzo=t-LOfcO65ha5e8n7+{=c8VoVo;=0WzWsdk@h5!#=|_}y
zE4ZGA>q(~fO)@Z0APhnTf_A&jqet`P^98>A#;@Uf1se5T`U+`u(4y5gpasoZo1j%k
zq^vASN|i86pW=EN_3T#&`^Rt)*v`Av&XPi4q(lK58*7xxJFIW4;-~U#Zfx`T(E`n8
zfRqkl7~r}tK`d(!MqvkO>qTO^_Fr9uMnyj-j2gF}d*wHNshDH680~)7xJ|{{&#vJO
zO^`W%5<ENI=Lv~p>=bY+Y%{*M%81iKIrgG*agYW~7!vMQN%iGv2MweFp$*b?N%!R_
zt(CddpQozwR44Y)+*)U=-sZhhnc10H#zrRi_?@3JJTb&;ue?s6Etm1~b1(An@d6)x
z_&(cP#aLcPFflpC$jA_37=X6)hR2T=$mjd`%A4P?F>e$p6f#uHWgO`ujYFf_APgJU
z+AV66t}Mc}3ciHav5O$x`2C95Bb0cL3mT0&ODhZP?ru}7)o3)@+`apNt<7!9r3!;X
zgE%ry(2eQZ-Tk$AhktE<Te0yT4Xqvl>tEHftv@hY;wVk|=415Wb!J|F5tH^%D#Fzs
zjsa~X=~j!$N{N2cBy_CVd4gWIcgp03251u@ltt9MOd9D4_IVMwHEs@!Q{COA{CJK}
zN>!$(r<oj|;<F!r#NhZ4ufF^$%^)I`N^|VQDQ@1n&7IrV?KH6?$rbWUPfq}vu&rY0
zm}UC=`grY2ZzD-#Ooc+hVQpy#-%As<G(o#T%~j~2MSAlA(I?wv#jn#o@Qg+MV(3>G
zD}|$D&iKN@V@l;B>+5S|QlotM;U}!GZ*ckYWg3loZ0%-<I_eCuY)QO7dl9+bA&K5r
z?9<%SUpW48VPTBG2@IPb-9Qgckb3Sso^Y{tl^}#fM<FA78%)Rw(gTcd+qBxy-mQ@7
zFA#)HbP(b@F6n%Zd}WIZsfc=iKbtGNR31O#O1Z|q*;z)0M)~B2KVx!gl%ZE&rCg~o
zHabS%;4n9D-ehs<A+BqS%E_s5QYnw16<`dkuC9~M=XvhZ>tKA+=_XP%X;%#yKgUL`
zgbD*{O&DIgM^IiR^U|*)_g}zBVb80$LQjS(YqcuNEAuR`E|Sd+bLY-|KK=9)&YgS4
zx<zR#;}irTTAR*vu#^r;+h-tBPKP2DMLMP$bY*LvP;>Yt*B3+TV#fyI2vTj$$_Jm*
zH#LHL{5TjZ%VCtpbtKuKN`Ji-+i(~>VQ?~OYvbX&I8svE*(EnIg6I3VnH06XDg`f%
z@1$5-s&K-0crooTFEzPS&#`>}2J6LDHkuV`)iQ&_llX3$n^$i#F+I)O-~4qhzw{-v
z@i=_o3?Cgk!~F-hP})!^^i!?wvA(gw_VyN12nvNPDL+jZL@X{YGB!5G$Y>s=6>Dpo
zWV`}P3rp1Y>Xdi4$>zWjSt2cQ9fMcjqW#%lkUsGW95_eh`UtJ#<jx-0S}W!k<|vhm
z6nA$xcI*TvPMqe|SKr|NgZn5I^6|$X^UO2Pu(G<&`s!+DLZ-4;K^sjfm7>vT(hk~`
zE0vByx{ZmQ^6O3RdN2B4ep$tSgv1e&Xl0A#4?ZO~)Q{{Rq}^`Eg0GMg5yQ=0GM>RW
zwh5B*9E|G_1rcsGMQghRu0tkgQM}#qE@ORzI9jq@3=lftT%p3ftfV%2n$q$zi^X|%
znq?ZbJ%&am$z}VwfBPPNLxX(z?YDUK&9C5kISw5>%gJNUaP9ghXaj{pKec+5jg2+7
zHa8Fgd_PAjwGqdBiP6y!Mn?-oTCuXa&iL>?mKT?)?bWI5Zj&2K;gBNI0$)hn;vC^e
ze@^D?8>q>nC?Sz?`^Us_4}vz2=N__K+93=>o_qcxDonGpyH2fIXK8tX((W!Vzx*OU
z`{1M4Ldx2O)f;tO$BoDJ1(iyLR;!KYx}AZa4iejwxF@z`Je8OO$e2%N70|tEi^cao
zqHlZ%XWtCM#9oC4e5o03?UMI2cmioOj^|(;+roC!DXa73N_^i(Wz$r4iVUQ4q}>ej
ziwm4MDtRT7Wh3R2J8_xE53W+$n5SJYu|2%S;NUo$>su_XE_3Pm7y0^ce2ekPSx%pM
zp0lT(<((gW7lR>{$`VEaTU#4!tgiuvY*vy^Z4w3{E6Xbk4G%FkULcATD=Qle_fNC7
zx<Rc{qq(<BZoChHPox!D9U|`k3|TMZpMI4vH;6E?r(^v7jWH}Q+k|dsXPXnpPjcws
zDPDQ`RhE~RsZ>f_zkZz;UU;5QKD~m`R)VJ4YIWkB+}q`Hh4y4%1zvelo~2g@Epd4H
zRW5_&6CosC1REb-qknpg)Y&tH@pL&rT11RAisU`pwsJ8zX`fJQoU~7Cvq-~{q;pwR
zCe2=Pmx7nV_i`-Gt#CT;@%*6A!@grwH?~;VSm4p#7Ta4Jj0{h)zOqInBA$EcW#0bg
zw-^|j;JM3h^4?G0;o+mZXk*1@%jI2G*OrNb5Ge)Q+q;Be#P-$}ef@bxNAt9rHEPv5
zTN|6~Rq9lV+vIcLkRj5BOxQ%;{4q{_5BJ22M5(;BNTWMpS=p<ww7kIP);hB@vz$0_
zisvrBz{=_>8|&*<Tl4%mu3oto%MFFpYE{5kO;2ENZ;y7nZ8aw8w&jwxYj2Eyo*VkN
zvtnI=c!0xeDOTS7i2kWjyn_d6HCr8ST_M<Gs2-5YSOp2;xhO!UeOenudw!%+;7Y<E
z#6?o7YI0$nmj+z6CnwQ?<l*ui?v^(g8l7fjWSYf?^LV)wue|X(U-|ks5Hic$qdC6w
z7k|w8b1hP-3_;jtV`J4yLrF=cQbQ}l-rgRCLY8bU&Bp2$9vL>)HmU5E=reny=|gLQ
zA2#rA{si5q;huPzD3yyn%wrie42w&1Y;LU+g#pi;JrBmG-ELFet1`F!m@{WjbMwY6
z!Z0ET0~(D+r}%0HA+>4^Z@{(pv`KVs^^<#9KD`)!p`@;j!3WY<zqH0~g~uO##@UH6
zoPjZ-sEvpsoRng+Si#o~0+FS?w})vpaSH`<V?&lypO&a*i`G_=?AR!&Y>x8!HUmR>
zq%^#AX$IA9k=J#8cb~_1R~MN$c!KJJX0NtOv)-V%y@{9a!|}4LFR$_M|3Ckl58io)
zQx`5WedrL<1nirfrhjOJi_c!daUAC7=UHE0C!f!;y0*;N*aUZPUPX!qDKEftpkA$U
z^ZG5Gzx*1j8@udnuOXXT9L`Llq5v6!lh2`zAl0Z4T>CyI3gGZ_pqpqsTScpoMzzY^
z(i|Jx>x>MH@TC{uV0d&N!^6XT`uWGKEG*EhmwEp31?sgr>#M6ok)hEFSle8u9khv*
zqTLRt)fzm0G{^AZ2#)WfC1`0dQI|a3Ya{eDR)nE#AT0zb<KP&>@~wLe9h+t9ne%Ar
zVWJ3J7#3{?_qNcfEcI4MSgj%iWXDHI6><a);2YY-qDAjgKCbi#TM>@ua^lD!?Rp&%
zwfG17JpOEc0mk>yaA0+Pj-a(psklR`&<AdsDAN4tfB1bqd+&WtT{zEy<Hu1!mD%YT
za)o}Lz4RQO@3FA3$ol#kg+iXSwPhwIrnqzS3XW)y@<Ut~%Ec1*?>=Ps`588<dz3d8
z$mt5R{wPrp+9NKN2P5#y8>r8|gSqf^%+xVZO^op{Dg+g>xw*me<_g8)7N<{~=i)P$
zK&Ba;7~#t2pOVX@s8)A5d+H>geEca!8=?rdcXugON=WI@Y6pa2#Ny&Iv(vMTPK?{?
zS;&|#)J^Q3sOFziQE$++5cog}hrq!0>IMs+U*-7FAWq&R3R(!E>30G~Dq9#oO}!fR
z#C?oZK1<*VjH_sE@8FJ)<E1l1?Ff-}kdEQSXZNAoEwZ}7Z;lCmytBi;BWKxrc$LOp
zg=%Gw&Fm^7l_j0d^YKsK<H61AoI3jq$IhQ4Y)2fNokchvmo7g~p|79S)fE;N{{P(l
zXV7Iydfx~BGV}QB@9W?4uDja1>7E{MfWZJ5a6y0sK#%}P09;9AaJ5=;SNtHYb|W?v
zT#2G6ip%9LgC9Tv<dP5x5+FeM!I<tb?M-!8SC_APy}ADQlbQSB+^X&o2~iEqOwaI%
zsHlo4ckg*lW<L3t-~R^fc8i^z&AUFAA|~)!xJt0KvBl=bHYbjx%qB7W8<#l0B<ZON
zo!00Kti&-D)s+{i@)#cfA*`$JPM6!NAWsr5-?+lg-ZqQ9dX0DcYxEcUJonsJxpDOh
zQJ{F@u}66A)z`_=l2Tg^-sSH!%PEVBYgcb@{K#?Iy)H(Ix`~@VY<+)gGdLPcZY>cK
zVHM>p;q7O>PV?w0jm34UB11~cg2?Dk_o&<k*(^pUF;1&Vb9oW#2t**s4i9mZ!VLnd
zyh8a6d|&YNW5=*n&i5|})b<|NmY10d&F<bU()cR-dt1n$h3`c?^E<!GrB`3!zK0&>
z^aJ<P@Fk~Dp5pk))7*Rhe#YZ5Z@u*v?N*D!!9K@ddY%ifzJOpx;D<;7FTeN-zSH8y
z&0U7OTR6!M$Gl~#JV!|dR^jNJ>f*Chd4hfTGaxFAbuhZZmIc>uUSn%-6HhsO@+05F
z+R-yCEH3cXuRY7<x89(vQWh6idG+O2(b|&cCEI&D6h%I7{ixUV$zvz!EiTlMaw)Os
zZ&=fJVB+sb`L*5wESa@LqV|7sFyzX!U+1BvMP$c=BFFO#N0I@KEQu3AF&-m5m*(0s
z(h-P2knA7OT3I28nhX!O=ro(0I@TtS9jYotcj1|7ll3F_aP8JxY~Q-d-tHDMY~p(n
zzx6Bsk_*qjz<rNB!s5yr=T4nOxjs)k`7}}~E?&Hd?|IDPDKEYF9IwCf9Kucs{eU9P
zc<q(fdGP#`%*F!_Z(rn*b4`RUFh(N%7D^Y$i_cPK2|V(ddMegB*s36lW8S{@HV21$
zJn`t$JofNooH>1-mDOe5dgB7q*(ys*tGx008&s8HGD|oZ9-=Evnq<tTG1so$;L%4P
zg?Z)9a%X^3e?wL7n`Q4QsMA`216o>2VG&xe`_@g4y>$uyLmh;PusAGWSrQdNS)s|0
z^AvNkyN4+<+Uv`>p^u(;<dYc!2)kW8->1+OQLBaHDylr-V0V{OD&;?Vc%9$4yieR(
zK#G99(H><vr7CigxL_#?Ik~vR)}@PFzw`#KAJSS`W98@(JQv=+_!=PC+1cU5$<uT?
zJ(d<1=yyA;FL&pYk%ohv9bSF)0_)3X$VU~k(KdV6U+1ANjBmu0Cyx?1E{+5_7XjAE
zjKP<Gm6qGUuAfHO9CSVUKN$}>+}p(!4xj$mCs|)wpfDMSds|$;a*>UjH~FC-{2b@b
zo#UBjz79a8HLl}v?AR%;T)E7_;Xcz@TsH?BQwst+@*7HU@Xn@x=aB!VlwYZtXu`1+
z!k{aGi7Re=^(B^09z!%ER3-_f!=ldVOE9ynR+6xmY!*`$1?}}^yjFzqe6rb;$<_{?
zm1UwZB1<!xQA7hnv)N>KYnyg4<YymT=GomDUpSQL#Zzn#c1hDIGhH&B74&?cQwx0t
zWzO&2yn=E(8jH)U96!d<^`pFf@l~W$?C$Pz<k)d~y~SEdpwnV)sY8uI(d=$)@cQd-
za%}w`;=`2j!4~7&7dh4!j5iWi*Vphp4_AUH>nuQeu*vWbewqIF|13qjhqU>;pRVFy
zZ->#rA;(sa@$nCRm}cNGnoQWbb&WSKyoB#K{FxvB2?oPaUA~)ZeRey2*4K}7?b=mF
zqX9)_X!s#{UIV6?H}SqZ1N-jM-JJ=*KduP?VbNM5lt9Z`s9ltTjh8NR^!!=W15HHb
z;!2mUF6b+;aaNb9j3ti~s(gdakrkXqfaeC}@s!!t4y~m{!Z0MSN}^T^<tWN5;Bar3
z2Lr>Y`&RkYi~A%C$JiQ=Ih^d1Oot@HAxgvXPMda-^8Dc*&wuF;a9UkfP8{RNk)vF>
z^g6yDvbVp_($X@Ei%ax-9Xjm>YfGJZFs!+G^(t?@b(NDx&NJQ_vcGj54zJ@Cl38<w
zR=1BENgON`1ym)I3tz=lis*yikHH{}08<gq#vJTy;|hoG`S_<;UF?zO2|F7%c<aqq
zn2yK%#E<<rj^obNKk8nZPN&b2BPY0W<!uJT0hP9No2@xe+`My~zYDPYjsE^t9`^Q~
zweMOAW55UoSGPHM^$o&L_Yo$<lZs`N)0ToN&JeX2GwI$Twkm0@E#dhAW|A@8+NQIz
zLez*TjHc0U;{`sm(HLiNz>l10axuyIjgdkuoMdyb$7C`g8%<FLPPUu0vzV`KZ}RN#
z{SI!Y$I9_zU@Ma8m@LbXQqpWJ(`p_g3=~Vf8k3S|8LwQqN|7rTqaOR$udsjPA`dnV
z<68;U@gullh>OMM1-8r?eEqlZRfKcjlk+{@psRx2%?*k;<5M5|6eo|Y5oZ%_T)D_=
zFMkb;<tKmg&+zgquieG#IF8H7ljpc_;Wf6lZ<A#OYs;(e>Zf}5B>gu|wtqAEWq*`+
zz!-^e&D{Tz?bokz;(^O}Pqe{G7BDP(7V-*<L`g+HnK7v<+DBLL{IK@@<~FURCA=^I
zrD*lKgpCG!o15GpmYiNx{HsejdhsN;#{&k_L(<_GUr5fiTa4lfU$}aiLb^oDOB_3O
z8dH_{zE55htRFqf%2J;XKJ^5>c0iV9jE1AS8nbwk<Z#UP%}aE%0r_Rfqf_`%16NpV
zo@2{`!>|4uLZ^Xq-v<yzgVrR|F?*YvoIQSqk39J#ikzEQ-{Q5GzsBzN4nOy^Kg)$T
zFCv7Xsw%>;!SQ3KdHLlR*x1^n-Rcm95!IY;SPxQ)e=GZbXXxXbVehf0PS!%*($v$J
zR#8j~Za(`eOUI8Rj~qn_!<s4R_Zk@O<5`JyLj0gbm8Rr*K|INb+AX9X;FcBHaDY+{
z%}%=(O&0<;YNA|Cm=?^22Q=dWKlRWm&)ypHyH_DxJI#%QT}IPGN^KbDDF+8ztj?PB
zqBfONF&NFbdF>XjhHZMiF8zLwR;$H$Fd%A&G#epFnxJIBWHMzooslFN%d4mGD$V%z
zo7~*GLR&c$`v%*Q=pw5jq*f!U3SE_?lQHqZ9_815jg`;+MJnah93@kcCo@I|`+Vf-
zPjL45DQuatxv|NcuRYIT?>4*Jx3JaIIHHE83Q>bHXK{?cBaTzvzI2Hz*RJ!}qmT26
zk9-sa<XLw2_!jRX&A+=0`_0(&#r!o<szMtJD2%oYukNz@-0PhF-Y%6J(lsSVLy712
zkVYVl0l$eCW)x{kI>`}X6X``bVM#HWkRA^3S}lYWRL0;nn&`ZwyS&Wa)&_Dk<g+K%
zIN5RdwKsO~myd9JGG%+XL!m8`GG}+N$B}eKH)w-jGM(P!)s1Z~1Q%(y+qKWzZ4USM
ziCQ6zW=NVQU|lAY8S!jJmKLn8oxv?~MmJvN=FSb4{eXP8z%B>qGOx?YRzq20FiZ~*
zN#g6cul^pb5B)I4&ZWsr#bkJhV=SNi@F(dtn`H5bw=TTF&8rugj&_LgaHWf>${Idh
z6buJL%F<F6hJ%9xUV8Zj$_jq!Cx4QDzfYd$?`;0_6#l#Wa_;#5n|0$8qAt5AB@nX0
zqOewx9wglQ(u+K@yn^X2(6S}RJxkc@A&o>@13?SFky9io>9iyYnn*vq>-TKv5;hw3
zXSQ~})&qhpCmtTq%g6lu2UhvQ)qOs%J%Z>oHxIVSiZKGq?re;wGZwujK@^e9GOi7V
zT)uFXC~DB@b!c}wY`1O`wn7@sh%8H~O37q0BZ*_mLbJYp8YiBycl`pJ2iu%!wkY;v
z?3of><zQT-F_=8VR0ZRm4NUx-w9cG|-jNzZU^Hc(GCJ7jLm&JYr;Z$_N~hesev7M@
zUMHI!FddD-Fy}MYm$8F`1G2OvO-l|A4tV~B=Md84=YHmAh@u8*n%-@mmIlnbbRysJ
zx&D^raKxSYdQq?a8vf`|qy@L0eT|h9#|clJq+e#7@Fi}rfMXTt3frjAc~&dlWZ?P{
zBFqre3Gu-Izulbs9+V%_>^me$Og@|O_)?SOy=DIO8{0fDme`0JY>y5|b<XYC7+vP9
z`VpNlAWt)Hji%hbcoQe^h#C!=trl-IFAy~Ys;Xo>j7gG|BFj;<SZao(w=Z&i>pBg`
zp&V5x9a0rJ)^(6pV{%QEXH0iD$zS*l){dX0^qVzKMpsOShdglZJZDdwA(;%hxv|Oi
z#x+9EP?eC*60#&gDam*;;r8uK{GiEXGUJUm-e7-!$j|)LpTqN9#^cF5Q=F#OS1-$Q
z-bbnDI>7J8phVrtp{QGp3T?4Qkxfc&eEtP`t84fRJ&xxI{YHe99h5fMuq0>{=(0o?
z3gv}}Fons4bU4Co1XSAMxGqvix=YI(Y~G}+Q~u(UYdmvppI^V$;F=$?Gae9UQ{JA8
zSjke31OfdZATM*S9*nvA+FK~s!w&<z(8rHLu$m+(DDrxsrSC7mbinSd%N!&#`ksqZ
zN=%ufOSdK!6ox9#nH_9nU;a&&PMo9g8nsS)RpKkjQ;$AL7S9-L-Dd07Wjaj{x!Pwu
z95Ee_kV>(=yUW!pH>k8_I-T+AE3b0o$T2?r)YA-yL$V^T5lnEGQlWK)HHKN7eA9;9
z-cwNLVDP1=hsWnK8>O+t8$+&r;cJ}zq3>Zqr!0+z_@bonJe>KhRNx|gh3{D;9#vW)
zJ&(dsxSmfk7~(B-u&P#J@O>X6q0{e@jP|Ltz|S&1aiYV$i!J`ii`OU`0gJtR*&ppP
zn2oramfTK<^rs_Me2=CVu^2Y%J7=I6?d0rlZ;_WJ+6c5%C^SO3IF65xps5YsWS8V{
zhmHLmQsuC&TtYjh8Me?>j+T-#O>rGXSrk~K$;*_+{&m#BP4w~^z+h}e7zW&P?hFgP
z4p*+eO`OGy#$%x3<jLclJb4@;G?(7GNS+lS6xVOw=IdX7p7r$;_+CJ-*T-`=+1}pb
z=IxubT5V=Y!r<_*?#qyO#Z%^!JGH-xs`_KUAW8^)QQ=BKrRR;^#$YSS?XSH~Yki63
z^JiJu-=Li*XjkIMx>Md7f#U?I(8X~abXs2qjHB?vfMOPd>tMWEuv=;auEdM#fIFXM
z6tj%;-H>A+I>E2KzRB;#CGDeU8O0;^$NOyL6+8KawwkiyJ9M0oPT0ii3gt`UL(Syy
zfV?cJv_MOVL?e`gqkMdzSr)j%LnepY+&nzM_#t5hO;ytHseN7usw_i>0cBBOj7FCU
za^oW2>?5R&MXaqi;)Q;n`|m$bktJMs^8!hhaBy&d=Sm)a=ph0>WOM5lw{CABr9)n6
zzW)4+_<j@Db?NsO=ym(-?jLaV+Es$Ur>r!?;fTR-FyC<J9=G)YJWt^<o{s<6J_m3R
zxEKOysjQm!hZ-sigG-y-`05Ls`sCB}4{p<rXW)TUE~2a_brmjwFHzD#&q@#ytt5UJ
zl1?W0j=;DIofr77hf&b&_eh3^Xf5!vjL)3vanE9lUwGjvy3=E&zr_A{kI`(x8<Uhk
zrYuY)i*A7LN9bUN>nverf@zVlw^i1GvXDqyA*6$&Jba^R6)|degT4K2u8+s`yFI)l
zrQZWbR2W@QBrzrm(O7g@Qs{yx9bs=>!0D_|2!SybVGwZt{pV?f0dHQsK$0g6M<c2#
z<M{CtoIUF?9PV>C7$T5V#`5CJuh4Gx2*QxXg=JP()))*&T)B1?V=Z}6u)Dj*U~o8R
z;>{Us?_%&JY5FZw91zsP?4c--!a>hP+bgTd_hT-7{#hRR;fU^N1GkuBJq1EytEwK5
z&;rjo*hx;|354h3_yJXtf+r#HAkA@n58)`f3k&Su-oj)hiy{1l$B(jjaf4sIUJ|Vx
zB`s49Mu%KUGHxa*?Wv+CT|Cc&W`p*^GR<ay$!83P1NNm0y1*cj+7M+kL_B0V9CCRy
zLU=AGgil|UG<`58r_2;ZmVuJ!s-)66op=Dd*Wma=bs1d1efQtbWIW`x*Iy#53Pz(5
z`v*IOVT0yUlUp0t+1c4aN{8Wi!fUU-#;H?hS?DeB;C&CU)!$)vZ<jC(Nb`a#SFW(J
zu~|=J++m$^X95)thl6j4kN<t<0D&w~@AALF5Z^xF$``-F(;t3>V6=&L097EgsatxD
zK@<jkFvT=nl<y(@fGSOpo<O+{MI7Tcn)p$K7c>z@pz@4QpX$*K1i$e5CZgNH${xeX
zkc}i}GaD1h32iw5&!N%ja&-L|okoM%Xhfc5NDNYg8&6TQ0qJDK?eUnMvf}<;mu?cX
z+V7&P6ih%C$9PeQE(@$Kh|(GMa1++ffweT7E$+Sd9<E%zKxrV&GcH}e#BeksYD8>p
z-(r9N03{WZ>69DSuClzm!twQ^eEO50q^b-XTboFwxN-9qFTC&~<sFol5cRO#Jad<(
z**DE^|J(ntH26|6GobH$7*RjO-3wP(II_g4<p!plQaC_WQO(nha^A>YCK+}bV*`iA
zQk%kYkd8)~k}6HXcM+x{OG^Ako61<)3o9gp5uTJ7t2x?)AA58Sb@3Lze|t!)vA{~F
z$1t8SoKDy;3wFy4Ut~m+F~*TJcYM~DRylU|EQ4W+D;<2_#Z@jvVM)jP*vT$w9CI^H
zxs~VS%H`2cKy#X~y4)fPT#PPpoF-+KVZwm2ENd9AF|-_ya(4@{d=_g3!YW$LF1=2N
z3$ML^bUaWF`v(WS{^lD@rW0Jx<-YR|(CPH+Ktx&s-+`(inGWz>$-QUJapLIl8gWoo
z42MGo;|b3{|2$W&T)V>}o=@w%^Yb^u1Qt&kBp!vSgEBiG3A5vjYhU_0$9#t{KOlDv
z(g-vKV$QC!ww@kMrwL{v@E7OVl{Cc*e2OGN$^a!L$z)2{?V^;T*;yc+Y!bLGU}-y=
zAH07N;pP0xw{N0UhqG%ZN%EM%biyP{8O9}!D2Qe=7)ly@8>}rXbL`9+hIvZEDgr0O
za~<-+5)ZbqlU>Gf%1)B9S(Hf6<B`CjHH|rXvWt=yY%Oh;Cn>Jy)l-$a!n4psQXcLj
z8cX$*FpA}+RjyuplUKL5QC@%*g3DL0a`pOk(mbWt?eXX%Pmm=EuCO=?nvqYDOpswf
zx7FlhANnvwS>ZSilktREnsM#=4ZilZuaVB>n(hFV=V)aRP?UfCKpyIBpi%}`2vQA(
zI<T<9q6)*t^RIJau|w0#$lVGj2W!hYOR83z(3Pf~CKy)|F1PWch&;>jJeM*}5qJn)
zk(C90qm7n^c7Ks#8sj+*MjB4F1V8@pGQa%h2G8%6^cst-w7QJ4gvo5mc9FAL#DsBz
zbQIBOkJV0}qo+<|k{G{)t{)KiE?J&2-QOac?y(;iY-bt!x}qONESZv?C^>#)g|f&H
zqK&H@vLwL?e9EH0)rO`Qz;Fmv1<FI9@Z1p3ae3**XYsrU>G|yL?s4Jbn~bMpf*|0j
zC!eM)4UVYwaC@x=rc9`$rrn76$kQL9EHz3gMuQ=P>6D8XFY?-J7pUgz?oRgP-K)F$
zmQBD~gsriB)<R{Nt7_DT#_a74$`@Xuy%te;6)LGH0mpNYUd=KrCkbVo;IFg^8V#nC
z37+p$<^@7nbdlGbP(MOB9$}+_oCD<ZIOanuEui36-r6ED5l1=;SfxnvjNxp;VOii7
zIc^rS8qW|#LJp|3pyjwkQNS=QI2a8n(<uizOl(2ZQJj^61wG-ulWmO75Tb+cdsJzL
zis~$&S6Nzqh&?<&95r(`jv$*QJp0Ap!%;3?)FDn%E?>LO{^2fR5c08)e+pBUIKt3~
zB09~GG#+5FtS&FJ&|jn|O0HhH#?HYZZ(qK`&hGA9=5D_JzQGgk6R3_mDwwsVt3hG}
zI7Tpg<0`FV9fFpvA@CJgSK&8ml_aWKE-`2aNY7z98sWGOHY=&Lrphvu7a)Lkw@Wgf
z;uzSywNI}y{KT0K|N8BH-W(Y^e!$Y|Iu({dGGlKx!zLNsau25}+1}m6$qE*P#1B0-
zla!rlOjZ=cM&K#K8Q*86obmW^hqjX-L=#6j^L7DCGK=v_h+GE_wqf~XO>0O}UVY(t
zu3vc*H|XL8Ee4|*+xr`&X@Vy`lIav-Dgw{J=$x(Xn-oP!S=CzYFTMB*2csd|yL$|W
zBmVHj>-)f|y9>lL7K5a2pAbk1ZelQRzfG@w4%fAHy1Nn>*b>;j007h~Nkl<Z34S9)
zC{+UlwIPiYloTK(ZV-^pW;jY>%c`b?8x00@St5+Z4+Bcyr_h?&AZM^YMkDylT9+o4
zFWuZ>PfJ`k;PlEW(sLOlGX|3hlO#h_6|Odn2m4H4pHUVSiwHVWQh176Ny64NBh{K*
zTeK2%l;k1Lp_jxww5oV$RZ}HHobEA%5O~VPl$x@t$m0Z8OFTzl2fNsE0t8rNm<$KJ
z@U^dU{qh?)ZiE|lP;P`R9kMuOXKR~=*I;eEB1>YJ6Y`5PXJ>ztBCj<7i?Slka^iT#
zXguLyJYq7L*1WPmG=%#m;4Ligj9t(Dt*t;Qi-UvN9?;y~pw;)NsvMKgxptm|+w^Ld
zkq~5MEp;Vbg%ApD=krC@BJzq#*1?}CYaL;!6zyJ@B+oFVWNYJqs<J%Z3JA~k`MoO#
zTt66Ml*@Xji?_Z;T;vQU6Ovhs%`05384nMbU5F{GiX|&>l%#MKH<E<SSx!3F2S!2P
zQJirUi$%<nCtc3=QgmJ-RTn8G?mPpT<vC@X;aUe@X!3&%IC283CAQLR-M-CNzw`wr
z!vmz-B#62o0u(;8(H@h*m|nY2niLdy4i?Iy;9#&zRutrOpJzo$nxssoQ-;$i<H>{~
zFYe}QzEwVd$Lc^)8;eF#*&2-}rNo)$v<|juAMH?;Ic26%Rv<iy+iqc%zzYMiN@Jx(
zI0`8oN~3GMp0<#cHB8&oJSb}{VWWZ03i7gG<HkONgBZWz^CKq~Sd{~w-P+>7NPIWs
z)Y>XIifNuP9*#(p6gMxJjz_$8@eN9CShf&2ikU09oFr_;CD~k7$Z-TmJ(n{APWUAs
zIOWr0A4m29YS68&pJyD$h)g4-#G4*c&9<=3RcvL6CqrI;`6b?X_3I!!9KS;t^$DC7
zqiV$7)(&sI@g_=n#Iv~8m>!S0cJnG}UXW!4c~O!V1!<Bpjb}{bgfz|mke;e<-ytt$
zt$%9f*d8<I*SS`)?rMBlQjMonWt~OTfk)VGAzX#q2q;T|^gWQ4%GQ0xj!@XV#+%eN
zM}v}*W~WV&q)hi_T)nbG&}#Bv-=%kdkI!E@<l<~bYy?5G!RaHbNY7^$$7FGg#Im!u
z&#WluYeUzO#E#-(nsHdvOj1|Dx{y544mjJAoL%&3h$(^7tNR_L!W9aOAeqgOS|X%I
z>|cfKUW{mAs*+?jX8ZOv28Tm%x-2f8px0U>Pcse<_W080|A1Li5N8E>rTP6Yeu)Z0
zo|lxRrks~p%d)K1Fzybuz2n#SFN}M~YW-Aa{wq089tez-EOjHgp++YOCeAUadYv1K
z6Z*)|Lj@8UdiAt-RUw4Lv7#QqFd(Tn6y-QX%?3pP5kc<0RWXZG+(wJzq2wp;>+qFp
z1701bY!0SWj$*OZ#$Q>)8jGt9XqaRLo(3UHlDK3ePKk}dk&3o7d}1NssfCE;D90BC
zuJVb(2-kJ-WG#P~CP|&0ro!yMjfsz=RS#EbP8>bL^I!ckX)$1F`4kIFN6|*Jv3ZSW
zpLvEfw<M+J_VynC^RNCol{IKxb44w4|Nbu>`cBgQn{MwZsJk<!ThGOayFZK&pd_c(
zdni><&1RUSz~W%h=&~eiH4&bR^gWaUr|DyrKwCpmmPpq_A+XwlkhMOZ5LjwnOtW2M
zC(0rxP74~nE`eY1(W4RTZpE`33D=U8@j*<bBvGTmx#LF>o<o`DxLRWh%{a}e3PWZp
z+RBm^g3Y90W-X2ogr1^dN<P%{d8{9?+Nfx{8M@4A^%iTbN7uzw4#F6utTzDP@eslw
z#v4?V8|dgLXhG;p&Yn8O<u_kpFx<lnD%RHfEcNShU0LK@y7U&8-nz^vEh&uP_VymX
z^6S4je>uz?CG&?1)7K!ScYgWJG$E)Lw%}c3-t*vIN>&y^7Tb!dN->ie*3`j%p=uVH
z3Vo#Gpqe4p0qNDfWZGzSQ4x85J-w@HSm>PBj}Q=q0aaCz<ptJS!cGUTVR$g8SX52E
za&yY1S;gLPkMUUG`2p(-eY_J#Fr~(=G^R96vw{jiVM>fBNwbRGtRj~J4}s?>y0+wd
zRze=^g)D{zp2#Rm$wGe#-w$vcRg2o2npB@9G3|B>X>!Ei8g_O9<1T|XtSoi8=gc|&
z;16z6nUZ$bXZ^@AJimp}Ioms%Jpc9QC`!wusQb#k_Tno%_wuXt!8g~hWlqKacRS{r
zJMH?vG$Smok~NSMeYf;iLun<Aw!_*&Kvl%(Sq$@Ro=xkG-Hif-=c1Yc(pLyqVRebF
zDzF0IacTg#R1`&qY}Og5(i$4ACPqNJ9ihCCs2fq3m`|#RB~@@?Xt_9v8Sc%n2>eEq
zqeoY99S_$62C~AC<pr1mEi|dNOss?O1+A*0vXYZ-$)7yg=47M@Rf07cMnrU)U7{$e
zN%nI9be^Rcqia3qd`!7}k;)AaHl!K4eE6vka_Qm)_Vz|Zt$_YwmzL{cD5<K9x8Hu7
zvb0Q#I_vT3o0oX);-y-<*4|lD|7*_t_8#+AfTJ{?<5O1UJVRImlcja&^d0(<r5umZ
zS$-G2p{oLAEiw!dz9Q^IDBr>A3XGvl3sex+{d&r)+gO5Nj)F4yQGnlW(p>CerJ$lD
zoQ&~J#p9{u+3gvZ#$%?F38j*FQIlra#PxkbKcpjl91Js?kz^%uIHV;LC^Ric^3&&=
zJlyms(=kSuxM72Sdx1vSKuUq6=7&{92C~8}!ORAfw_n4oKZRo~k394sU;Oe}UU~a#
z1VMt^N(cbQ_fUaiZ!jP$Y8~0UEV*)f^V{mXZ;n|!rzCVdva&*odiPhhUJoaZG%@25
z%qs&z7|@n7p4Augz`^f^NF}kzS~@f<5WbJAJS>tTOYz)reySLvc9VM_=}}sZHil->
zCU87Hq6N=v&G`CeOg5M@+aFR&iE&(rLVQ1<^1IyDIZ09C$(*pPFamzy>>{6D6qM63
zqB4ZdCJT$pG=nC}QMGg(qE@;R6;35d5B4dxE+JQ+B5)jnR-fyK6=_zm6JI6T-No9X
zzL2FARmI(;$1KbLw?5u)nZ2i=E^-72g)mhu!-K^^B8@|%3(IXuH5g&5vW~=SO(3c?
zr39oOAOZ(BbZ`T=PUWgP`;nJ9LEs~iNY^FFXJ}m@twLcbt;Y2nR*$c<aAc*{RF#$_
zNpPgcLNH-S$nRd=<I1?ij%Mr+Cm7{o9S6tvaT^Vyg(X}SAcSVmWX!S>PYh`#3A(6o
zh2?YiEb@J8E@qZe<Ov2zv(aXu-6QnF`O8pGvxKfkUc~r~25up!oQ(ACm*IgvLN{qe
zAtz4V!_L%U?{JFm4~Xq`f*_)-VoG7ijioFKs`){#&3kw`?%22JCGnjKf4;FxEK+C$
z2FJKq2aX?ULOekibs1VJRUc24rA*>l9wzXyz9ecy^GuIGT0xQKXypOmIu6Fp7rGyz
zfJ#>=N78O~X)g50%M#%z(k#XI8$^!F$5g}@ZXEK`UP6_W%;HT-q0mwxJP#*|2*QYF
zt4}2?n>u5bmnc1?m84YK(DWofe*Y>@cP-f@rphz$TzZ{0y+)fTsHKju8muDnJhsai
zDH=GHL=O+iZ@md8`Z$%K-|ylz7Px+B&^jiJH#tb^l2c_8GHc1pin6*>Sg{mZ!+WWu
zeT&a{rAiv^T#n9|`oe;)H+2L@*IKB2R{LGoWy89i#!_V|7>)FNY$yqQA4|>B6PEgt
z)o|(ygX>Y|Q>v;!p-@6l>Jk<BEFW8+uX90BMnq8y*9m#xnC5l@zw!1SyEB0vB}_*#
z1r{w8Ht=zR0N)K6seqx*iFE}=(<@4Hqq(Oa@@F4dXI)R2O~+^rQPiT>>(LAv1VM-`
zDi9Whz>$!P0>=oXl}s<afm}Ea&QV+?!S%Vmm$5(0a3(u!j%GNHi=`mTGm2cJ^&Qz}
zdzW$dZFA^*G$AaZuW_U%F?a3|mO4{oSzm6U(;20%A+&d5QmLURYqXUp9el5et!f1l
z<v3*JtbVq}B0Yz4Qq1K7=hc9`z;F9ByRCY{zo>}(CXpX;d^X|9dyew?s{?Kw6bwfb
zX6cM-GUm`g*6UEB$qK`Q^3myx;}R5xsOR&!dly(tQ>LRsg21EOU8L1&(rLBvd>>ub
zpF;|P<0^7(>uQp=#J8>!EIo)NV0-_N=ij`<Ag>Vll<modySmeNpv`xl)_1E(N8xuO
zL{U-NJB&UHQs630%eQD-fj?KY5uh|wrlc$l5(kkBq_DWYha33JW(i>!qK!c+MVV$e
zZcsnFQlsMxw;SL(9-)E2@$i&nHl6Ylt4ln2e8$FMN}LtQLUU_>#B8}o=gcWCZQkHS
z<0#_tC8PsI81mGKKFigNYG7#ensnQ>HgCJtxhoD}b%hXu!1u}G38m2lNCvyN36~$D
zC@f$4+H;Ju9Ahl$EM_o^>xc9$%(ne;>^%i_o+>d`&V!HIYQ~OXWuXm4N>Np)IsDel
zA)V4l3SA)L3Q;LMr$JRnl;_owQ9_`r5(w()5GCt6eo^44x+kfY>(F>kfETs!90`>n
za3UH(L=-rjwQ&E*NBGj)d+d)jhr<!GB7-tzn8u9K7`4)8d-pnya9L6Temv&9qiQt^
z&*fA1EptX%W}AoLS9F>!8jS`~6xA<-TWfwAt#O5TM=wr-m4ZMz>~3xmo*84EfU43w
z^TLa442Bfi5GNUf+qdrQkM%SB4tz20nD1MZC`rsapHomv*}1M|q3Kbi2`;s&m7$)%
zl(rVQPjiDR3moB7W&%ZsQWDqqh;{W2OqW{MugY>9FT~C_`7+4}m4_2G5kZ726}}S^
z`5}R;FnPwgk3Y!sHx9Xdm@^p7h|?HbY7UYKhq|D@*yPsE4IJsR?ly>q``n`xh40Yl
z1bq75MHbVP@#X~8gkG;t7>0ylNTbn0InG@2%_4=s^E@oJmidvAzz7E08$|0ftY|PA
z&-lXEpJP^*7z;_9a4@_1&clQGX7HW4D+nR*n*q|4XyerZ$er+02!z!1nhv_ms085F
zE!1F928^jl3yrJ_JU1jyVw^C<Q4%NcNG3CyvW3<fCF{#!l@|y%tRIQClyQnu0n+#I
zJdeN$sJsAANhaeFs|yiljx6)WjRQuhCZ5HJvSN1>bK8`he&jJO-n<G*a4J}&xPFC(
zrx+{C>6JyETJ_O;LsDgFc3O1XZKM>mS{)q6se34k@@_+O5Clv{hC)|3QX-}^%w!+c
zU!klF&p!Vmd*d;s6%>^w8NX{_<*u8*GyVPVvSxn)e|ZUWcvxo>=b&RL1P*oC)s0--
zpi%Ghb6Bz+Cs=6^z6;7=ssf$VEi_TP#bo~w$8pHA6kmp*O%14aeH^8bXo~R^PfD~F
z$jTtCCGtbCg2Lo{?!G>^_eN;dLpUBzVR`MwCinit&$17^b>lKen;pt8{~nXAJ)(t0
z&M$bV>9CfTTU@2tY?7uKu3PWXp;DN#x~u3Ag%Rm^L>ecw!!~Ao2*m{LxLmq)g}s9z
zmU_VJ9gV>E0}}kLF)wUz`s1Hr@TJeA)ATOPURg`vYMQ+^?PH5bC728jY4zKrha<>J
z!hVZN38Yu+^Xf{|=yb?u@w_@!;P?S~GD8S~>-*@eL~28NxQpA2$WfG9V+)IN9CSA3
zWZPrG@sNIpptrzuGNU;67$?8yNBI~3>Yw9!4(&#P-P>U$Y~Y!M&O#5@GkA`|7>8M$
zVqji@u#{ziCmKXigW=9L@np(Evx6S&QROoxQ^S=j*XMz{1n%6%#CPn${horlN=Y1T
z?)F(heaJbsUgVB1aJy}!s+ldW=RsMarwKw@@EtS;4^5dQSio}>VXMh-wuk3=#Bq!x
zBBT)YjlL71P&m@TRyEkvRf4JnRn=EWS1MXjhgp#kn3x~DugA^p0o-#89FItN+_-j=
z!Rl!~{OA4xzw^u=kYhO0XfgQBf5mivglKnp_*f5bvPWfLd36yl5Hy=@X44st<Ds?2
z6&^||DqSEvfe?}~2q=@BBu!`qO+=Ak;zLAdfjCaNaN%M-Ir{ER0P2YNJA9LO7gy1v
zVLkmM#N8=N1q4zfMM2o_q2|&-cQirGBte%IP9wn7^czfBU=WDFCu%hr?j4|%BF{36
zH7LiSoF&+{MJS1*99#wYc#1D2+7;N+)G}<=5_%Ddbg0rHA37d!JSySTD%$aA`3<&j
z-DanAgpd9FU*`9}@?{3elr!xPqkr=&jCT)dEG%(uF+fcZk&eUK@nt+u5C%=g<1s-u
zBFi$`jSyqO=mO^A(!T3a#3@OV5H!0~lS3$`XfI%UdzV|cH*1MC>|CJz&h`D>KL5`2
zz2kYDc<Q5MFF%ioi+O8ZePAiTqr!77;X)U$5!5AB436hg9F8%0NziGcl>kSAF;sbp
z3|yL>4zt0yZd-^moMwcSbujM*4IEsIQaA>((Ez90q$m<}si~~NR2DyIlIw&HmdB0;
z<OP&&k5*@ad^qAt<?_k@!(Zd;uV3Je%NIGdv`G59Q)avStSlX+r*hmpp}pE7iX1#g
zA%w?pG-hSB!DKe0)$i5h_xUpk!w{)!?qbqO2_41oaGRiWj%1c{>8&esX*Dtbap!MJ
z?xt?OL;d~lvT$jI`+n$8a`B)3B03%;r2|4Ct-)2ABd1T$J+_V$z}DN>X!TtDz$G^t
zDovx`Li-L<IoP5?R~jdX=&mkNrWsX{<6*E$VDb{>H;|1cNjV`9g30a%UZVj~8*MFC
z*Mk*8l$aD28X=~th&nAy6!F+<kHuqsP9J-S+dDhxz^8d+iBqc)?Xp4&SUP$JU8Q(R
zP?VY^GjtZCdfyfTtxK#nbwwv?a5(pOe;NENV|K8C(`#|%@-@o4vQFRm>-mplj?3xq
z|KA}8hfKfnEDA9%v0ILvIKt`EkD%O~!`%(`H}`3E8u;B7Wt<=#L8IG(dEdD;1`&p|
zmilA|!y3_|#9aE)qI@3_X+&P)8OvyQ2fy1RXti-YkMuCXl;FAoqjLfyY3K~u3K?u(
z;>t6>N#seg>6COh=3K{RArJ)3HC#tfW(lq<84e19Xn{#Gp&zbbOhs8_fF%gR8aq@L
zC{H2B1M=A+!=dG1Fyfso%y+)8-;>UayR=4OG>EFMp$m}4phU^iktI6oXXqS1L8HII
zB+iKXO`KMM)9?`zn!P6BQWx23;sha98;rI@-5$+;o3hNPiVR~kI0{pg2rE#Ij|c;#
z7vNx-?%alWj1`v37^<Se<P}soQdj7#V4)eZ<QUc)F8%d3=N>=L1CN~I?EUxh#8V%q
z$Su0rpmBVa2S0FzkDTe#v^k5%);M|oIIW&fvl&w9idkCXMGdBDf=~);HAS9cF$f`P
zG+LOvB%j47VQDMLWaBzjRdTR*$mY)8yQ2E<uG`!_pZOe>8=!545*5h$ekzTj*>PAq
zeIM<kr)aGnWjZa<0>XBL^%YL&;`t7Z#SY#=7u9OjQ_z)0dI1aTE0j71TQ4p#XLwN=
zl=5(!O$w=q8WFSoP0GPO%(G+G0+mK*3Cb9%w4fV?G;Gd#sOa>4*3O^g%)RR@FZX%i
z;SW%x6`2(HYl}SmfipaIB&4YftEZ21^vn^Ot$-l#sf;1c3{Dh~7CDGoU^_3;x*Vqz
zji^bP)(k$+mH27O>|m3!)ZDspi)ozlZ*K(foxB;`HK{oHna{C!|05JuBSnc21}W>Q
z*hbss+yjr%Sv^bCUuCd6C7vcYu0n(^Zo@|@XfCt~mioAz4$5^f+9HFH{`xXSnIY!=
zw$_pJ0U3ev0%$a_jzi#kOt&}4CL@IF;`%|2KPfV*EJJ98FI-GMW7!pSJ%iV8v%c10
zwsW1;)irFT8IQ-9FyQ<>Yn*6ldd-mIXHT=R*e8k_D96QEg_R+hF0i7mGE`MgRTk8n
zeuR<^S)AZXK`ZbWU%yOMWef*Hws-g6rSkA@AHSdQ*gv{iFW~qmzL&KpKZH^ON7Q&D
zDFkar*ExIq9-Jzmn8a*ezm3U@x({k)0oe%9ra%dBTn8tt`zwvB2-<DR`S@S0Kql);
zfe;AKL;60(acT9JnT;n*2Kz{9i8@WfR*R}IILbxqf+U|onGqE^>!M`m-~2OP_?3T1
z<U+Z>PqDd4&*rQxG+A5g6NUi-Sr2Me4r%Pv_&rg}H|eUPnwR4|&&Oy@S(O+Gp(Du0
zhZI%D?&03MdYius*7G05guq=|<LnRr7}44qiaF-Uf%9jNP))~_;}M&0Um|o}Cc6i?
zjzbWJOb!p&d;2Qs!Jt0N??BW2fJvN@r6qFSgK3PxDv1{cweWqTMQ>@HVmu}t9wMZm
z(_h5*0>F^xF>y8{8;|kh2`jTb#$WuGB)6`!bL|q1WW<qPo5iIiqR2tYy5h6Fb;!*d
z`-G8O_wY+ulPl~U4OU5-WsFB-;^CN%@)4CLDRYXde4m~1-t0XEb>}|yy%6}g!XQPR
z!jwYL>Gyc(fk%nl1*B2zZtO5Q7@_mJ2P#<VLEs{!C5(K$Fu;idl<#4b1mC0C@71MS
zr7&982fVID>`@B0(V(;zx6vYMv>ER2Gac?DC4{YrsMD%R?WIKtg|!uNKEtskVV=@;
z1<~dezVHwJJBBykWU<#`wy{aEyHAwQ=!b%(r8aBpYglb@l#lEB80}K#0Z2uu3xp6g
z6tXPtF6=0bY9Vf85RN496!F0}))owgwHici|H}>kJAXqs@zGDQ`oxn6RVQ5}^)xF|
z^6<kCv)EcimLA!7%I5Xkh*H<tt<^<jE5b;Nlor=fbt7@;!+e8xn+=5PqD|e0B9y$l
zxe6tbo`>{Yw2*X`*2t2K;ocUy%JD*vsN2EyeagEVeMwp-7@gxK34Lkl?cU%I{@#Dh
z=4;Qg&~1`zZc=UU(#T?#+5yYU9hR3Csfr3I9Xu}}o@tcmkmh-f@xWq?rYi3&Zb1-)
z<XK)%he`;QARQf`P0rrI?))bI-MR5wA@J9abN+{Z1b<}}SJpfwIY+46ci(;Vf;KiW
zWP?LCZ{9-9`%Sv5OSs(@T0%CRV6waxzV}`59E_t-VMJ-6EbGbbx}Cvbtij59gVz>B
z*r2<-PCA=0IoQFP0>2s3==BgdILblmoYEA;*^Fv3rDqC`Cj&11|NcG~fBT;iIFM}I
zM(ysfAT&o;7Fk$q5r!UBRn1-FaO3(uo%RaD(E!hLYdCJD(Wchnbd<tcOO|I;w!YB1
zy2R?7Njf1;llR$c-&|8dAi9g3`Jo@7b?Ow-Quh`L%h?k>%Gn-yHe~<$6_lx{W*OPu
zkSZ&ntT?!Ni`|QFGuzssOfzhy>uFcV!77*iy&+iU3LCa=w@?B<3aCtpa6OjRPvB@x
zyuU}8r9_P;-Od7FsEU-T%!rdI*?5eN$Mnh(&RZ{%AMCMn^KDil#p==m%}$>vtcRVH
zRP65V@y6>nSzK-~nI7VM0oEF-s@5~UgM>=6oWa2%<Ly1dO5$4=V+C1RewVH2_ilo^
z;@-UXJSRW%{Rq#+N@)5HEk6No#B_KYS!G0l2U*7O`b~DPT_-&pp@$=;8=GW<300h8
z$^t1J8to>#`;!`bB5TZ<kP^r9Z~_-y=D1<V!rB@lD=7{RQ9{t_^>92Fsq67eK}8bJ
z7>vgRvkBcYL6!+e?>R$nZJn^$rq%7@`7Vb$I~;BuaPHhXalVi5MOb6eWreW@Aq8Tt
z?oj19+2H`Es1SvrygLvI-z^uk_Y~BvJpFN2o_G@Jx~Tagmsrl7UB|65irIvGaLC4$
z%e119(ar%%TfD$02m%hbx7m673i0k9x|p-J=G6-8xa5`P@L+@#YA$p<m%5S?;V9g|
zCo6M;PLIXaqv$kccCbsCr#OK}ufK%jcnD!B@`Sudh?6PBc!Cp8ShFeN&DV%ue~yEl
zO>SLynU2X=>$O-~S|W-9bd}@#KGs_H_6A&d?Hav)oAG3b>$o7Gsw%9l#dyVhR5OlK
zCgTay;TX4qNVs4vNm*;Rez$G{qQAuHANXOKCr;J@9d%&WZh0JAX_Ac(DANi1*RMiV
zVzQESe}t`cE#WeqvH8~946fZIAB{10u{9Qy3y#Om&S9O!5q0zG9iLkz@uQH^R9NM(
zw0;8L@rd_#$)^*95_Gx?gkb|PRAow;r^M-$Y%;-4$MkGUe|VGX(yJWqZgT6**XTPH
zEB!W|UY{Te(7K?}Xi!xZH*ak5_FEe)t~42shj?C4_x6?L-F}-ous2NOnBl=8@!^oB
z@M`rG1o51|_1$U$a2$v5zK1yX*&jdzA<|lukc6J#=(0;T-KHo89NfH)V=JV#Bs+)9
z_6Af*M!a{x-j!<%u3RTQIHXE*Y*8V!0mnhQKKuJaOl9W6<>GExM!F7O6i^f~%8yuH
zJArEq@%|opJj3$?78jNg0*ooCvXr9C$dVb!ctn*==~ZLG+t(;&V}`r8II`GeajB2j
zXwz!7a2!P#MvTW}-n?)PqakeOlx2nMdUNH5>h2*Af}$)Lk0u;$Y*QUh>AFpXmK567
zK<@Y3ao<D})ETVs<Z(`X<_8c_h>;f0b?7!KoJK~a4~Y-<SYGL4ObL0)@a9bho156I
zLZk`t?jHGQLX~Dzd4@pZhXK>fkY+hZg`KOhD5+4sUr)&E65&X?YwLuLPjxWF<S9|B
zMbK>GNf&7>z9YewOlJx4WJ<$mmQ0Q^741`}Y4n%z!ho>Z#EX1J2fK`ShnzUFNV6j-
zsuI`psq(T`>A=?F>ZbOoqZDNvQ_g01ML}+vPjte+eM$ch$>u{`u4L_lPqXyF58w&S
zkt0nOk94q(WpX&6zqo)^^<YJr=O_!V8=|C3F&mQ&4lsE}Rg_2#rBqzr%j-T}S@+K0
zY3XsonpbG8!77jT@+uN24+khK2%8=Js6ka$brPja5mM1wJ5I|8&Yr!WSEFSvj4G-k
zrbsm)$&!TQ$5v={JhHsR^ZmNz=UwNrotOCYEF;ZQY*J$Ouj2`ca=iD+2Y)nsPeGks
zT;!hb{~Z46dM#-o1)dTtG%Bic4=qQGcW>j^ie?l-RxsJzVei^a4sP8hAC9SpV`iJ%
zqz6OFI7O#9R5js0Y&J;Jl6*cQFQvL0NlT^hq7cxOWsVm%Sz0}cV=UR;AytwP_#y3X
z52Ym1Sd=zMVe7129CI+8AZBBN;ee*72t2`k58lh_(POAai-m<H8m%T$!r|^7TQ_&;
zcYE}fAkC)uexsK6DDOySfe;dFbS-`y$D{{Cs?n6D^w5>0LVi~-dn*KubLTkyy`P0J
z!deZ|q2G`wGeY<ivTTpZ-VV!4J#<+h^PJ($+w5Gq%4F{Vm6zoE`^3BZ6w{b0NijuH
z2M?Z4*a$f|oYps{`R&aDQYoCkM>z^z=Aab4l{H$8CgtIXay&*!NxR#_@%{Rep$)E-
z2x&;ul*7pkJDU*BCNy=1=UC1@ct6WWj^Z?0^t*kcc9W{gIoRH1ush_;=_5FPLZvG_
zFThk9GuLPo^Tv5yX{3@AX-Y91W0RCXLt!Myci-;5TfJ`~@>zZQ!}K4193wPZNt%H}
zGpKOdB~>+Kw!h2j>H@|VcnadZT?V&rk<DiKNUF(*<lvBUmQZCmRgoi+c!5u@C8KFv
zuOXo@M%RJ8Qm80E3W?Pv!c}zEju1K_=5T~gX81us&~D>90m=$|DRC8;GG{zVF-b;K
z!D3O%@A$_~5_A?(zDp2>xQ&oxHe$ScfM?;shmJFv9oEx$Qc~pw##)T92q_WPQWhm)
z5MauJY&OBKO3JKOm_h!Bb<o!+DT|e``q*QvfAXXF(y+edvvzs`;VI(b7(a^eqYy)#
zZPTU3_X9kyfs`I)JRuz%V6(IyL?|n)a@iOg#%awi6H;L25_dvMyf8v4w+3BUQ4<TI
zkm6v7O*1?{B<%DsR@^<{tu1Nx7NEUE%Sg`L{|K)Hi@Z22DYF?xT7g!Sg{IXGIdOWG
z;be^OM;K$t^L+lj=ilGfXr5#ilO!=ZHHd>-sBB2!HS7N3|A4K2i!DYQoPL*+Kk)rT
zYl|!%U1l#8jD`tTm-yWtT1Xn*KHc@B^p75=x3)^Oyowh!$p;7Ig8?SX(N&4#`V_Vf
z=A=|eDeh{NOUFS30Z0cb&>9;31zgW18;x*;OJ`vjKWq{<+qAojEUm3`I4jxM-oQ@>
z1kFCrO$sg@PLbB2s}hY%|Hv{2(-Ds6VJbs8FZoL)=NC$I7ciepr%0WW+<cvQyool|
zH|TADr`vl9>ZAhi*ip`W_6HG7A1N)ZrpL;9o1h<}MaKBxfR&Xcq$`jvU`nd#1ayg#
z9u9&so|2D8=rqNYC1qLQScz~IJCl;bX&r$ZWL<u>qOQSvE=~|35g1!yeV5Ma5kkkO
z91gH)isN|%%{IOtVJ#?I%RH6&jC7LHXtd~bJKP(!X{;UNb=hPq1ykjeRe=_gswzk)
z6FO1I{r9gknof|?tL07dyguNC#m;H|rZN~^XB;Y1)KgW-j4GYrC=a2W@7@hxSa2l^
z4?oJ0kA4K{f+sY`PWI?5w;`0whC>41r_pX;gho1oDotzXv+LnV4_lSw;~~X#iY?2!
zmr84ta+#KfYr8Q5ffcnRl{E%QUFLP7sBXBn6<R>lU#8XVQ%<Jn$rL3eLDVFOT63_i
zrYdsEVn&waw0lcL%?|f>msq;z0iL75<w=Fj6S6!*Yl*2ec|2onsmJjXeFozRo)>}?
z^IKhgi?`H$Wv0?#Y^`ru6%<K{7#(1X7~gGR?`{I$W&WO<#UcgJ2s!^Jeu(hc8V!?k
z@>G}JdK;8vHkcpkjYbWVR*Et&aioLqH&7}93)y%`J{m%i*Vu@vKG=6=hJ#sE*Lmi>
zFUHg?Jn1;NVFN4bV98oRV{r+m5mFwG&~bw6xCHGk()Z^C_Y$phkkCJRg2ei)`b|ze
z^f)j33%oEasgjsHFR-})E77Ln-1!q&smbz^FbFBil5$>2xPx-AwpLikl9btON?BAm
zNk+B*HlcKI9RGdv);Hho3_NRXkli-NKK*GLCy&xwTVy|T*&h_7lQE5jKGyY6UO;Pk
zjm6_9=p8vmcWsHr@+w}VMKKzX4Gz&+KG(lhI9@<e)`z8<17PoDw}tB>JRgJuTA)iq
z*zVBibx0>8L}7@U9a`N!VWUN(-KE!Grn|Vrjjclt4z~!>A+2VgS7s%z?2hI*qLMVP
zSX^IcTBhVxiK7%nQPgF?d4PE*6Yn|>RbCKJrnTzCXq&^EZ<3U=`g4949rS+`6Ht>3
zq(^i0D91njaeDVY#^oKsja`kqb`RPM5mMkZBD`i383quzI8i`nVU0$+Pcj-%jE7{i
z8Kb>Dd{JUbgOmy}SD!ILU}Q~`z)=X_K}yip;<a1&L4Yj_n$0$iRu?H<@B%7hnCx7q
z(+DBeJpHMU^Yowo2`)J;wx%Nv4h~tl?@?w#k!geHhe+2&N`*y%aIjWTArPVhEs2LQ
zLbyckGP}c)I8EO*96HzUnOEs*z4p6L>+bKrpS|<%4L%GQDOmr|(`;RTi{y8o;n;~~
z+RG8Ts>lw9C{;JZYAmkfp^ZR!4uRt$a4=Ot6%R;<2er0fw}&wfRtc=~7*`&X+_Ew!
z*8}wpel8#51OcWjK?qE#5lFn{1sDuT4h{+XeVj%_)ax^vjj7u>T!H|Cb{Cp`o`3Zg
zi(Z3nx5q9nFC-&A%#?n&g_8*cE~T-YKD$mO4RM;%ZuO|js&*uGF9>#?s<#$LDyE|`
zzU$&E4?i8Uzwsizv;0nZ>ASEbSF^_qPPfJJ@B3ahue?ol?Iy?0u3#19lNrT$L~Cgg
zJddi-xKRTm973lSZN@+vOR{(j2Ya{zL>Qv2MM{rKg^csMUrkDJw*+ggt@&22hh|O=
z&>Eu+qSc_|D8`2aj5hea9$}}4I~mp*pPrB7x-|L$ChYLm3s+fo9GVNu3~BR9JjRcQ
z99it*lon$pT1eJTtux6IO08)|9z|ZDWWAu3st=z#JvS1;bTp<BgaodK8s8w>c^&0&
zw@mxp=yTxil(v?F=9x1b|IDWt{nEdnx4z6Em3UP~HW||B_rMP@MM|UJBkcC)Q*vM`
zO*->QNBd-ldpJ@cn@uV+&%*i<Sy3}Dm3(K7+%0qab+a<YLS?9|A?(!j|H))TQ#p8k
zM7O_)E(^j&6QM0l&qKRDmtTI1Bc7tQvPS9mxiA_dM?;=CvW(}{L7W%0i27|d_xEsI
z7c3M-fvW^+j&Z27@fKG(R3;}LPv|z=L?-6&(hDR<K0x)Z+x&N{&;Ko(5AjOr5iP88
z^b;SZRaR_GGpfCPPV|oAEOeOeZ`12EQGpLappC9|pQWUuR!NmdJRT7UaH1B|xR!7%
z^Acfwgrn*Kw7FV^RT5!Utt=!QWL}`PMus6xp(yeUztyDIU#84bq;fHi!*DdhiTCi9
zmk^odLx1X%4C1GF<JbNf{n3bys0fyh;xDeSwS67O3vs;wC7e1dDct!5Le6p98fhi*
zD8a4@0@cPiD{So?yz7vE*W<bq1<lQV!=(GW|Lk|m`%$+<2&uqvIC}qmER_W%IgQF;
zaj8i<-o+0Cv@s~|fa7J&u)`8`I#@LMa0HoVXZsLGNXkm%IS$hEuu32d2(2ivSR9m)
zSkJ*)jV^K=KO(ORTqMn?NfOV{z}~?wou*~!sfW0=TM)X6r@r@3b8>l^-}o>8Q_d`}
z;XU=kj7KTm`@t%Wuo*U)A@eaZ8B#eO!c0kX&170%W|i+yWmC3xX4tACauuDZL2IGM
z^70Dpc8e^DNz#PLbjtSrK9k9;R*`t0sTS`ksI%6hCHS2_$3OXTwlCelU+D41j-pv*
zgjt2xY9fMwGD+$7dN@Ha2TK}+E-6ZbGYd!#cSr|^AS9~QrgR046f{GRvdDqvyA&+m
zeT%EPa6-=uT$RRR2p0N?a7Hqn61hHM*rdO>jHeW#=VP@%h8~-P3A=j(`fH1HT1|Q<
z&hf^Dmzi$w@xz`^rvtOu1TSpS?Dg2%-=9ky3AC<IaxOkE?vz{Rw>qULN=-IPsM-Zh
zqdB~Go4tl5ztdwSzH`|&f0$Wpoh@};8Y@RR_Q{Wu=M@KbkuP2tvfdIn(xcn$F`Xp@
zK}gu^5zQN*F`B9>QOyu<=#dN$m>nJv^cN9<kLM~PPuAkwcd|692He)LeT#4;LbwQB
zt1rk(*F8I7Kzn_O*<gZBa{@nPacLD@mIPsh)dCf`+}b-}ayVk;$SO^3IQGO7JpcSx
z`K`<Q{Au53p>G(EhjjWYs32glyN%;I^?Pd!c0Q>kgq(k-I{0vvB2Np_S%U6#a0|&~
zFJ^0hr|z+`@ZH2Fd$-jYNw7kqoQS3SA7uII8$?Hs^2*J9!XjoZT%ohDz;t*>Bk++y
zQ1`+bgEdvHsN(eS9gpeZ9<#v#Az=KFywn7;K-a^i2w5NO*ttrLG~h_2s7pXti7ZNV
zsZp&4ZKcT44A=MRbQhRS#wg`tmBjHqHg4UfQN>^!WN!KRpZq?q-Mq=GzxgXH4aOW@
z@EK2Ij-I}caXO)>N*ZBQtDhj}55U8S`t#_yvW60pY*vt^IeocA(_iQHK}MY1xy{4B
zeM^70Uc?`E(8G7ydm<1uL|#a&qgXj}hDD>1o7X9Iht;(%P9P}92e`h6nGcoJARmR%
zRgFICbg)8`jmMbMa(in~L#B(m1m<`iII6CE=z8C%=KqG07VEhftLr9u$4A-%DoqqL
zNGeTUl^6uuyPK@91hhZ!5Ib8_BG2K8Py9*lJ${0J{@4E#){NowPyQv|oJbny3L;@p
zIz{F)$YziZ5oug2BpFRsYNj(w=vtKLlIw!u_HD+4Axc;xN8$M%jdq)bg#|p%B})_X
zJR?pL29pVCTD*g?1HRLI{J&;*Vpb&pYYDvu4Qpwgy_dkm*kXX$PiTaG9hA?@sMgl!
zqcIj!WH>=Ym1ktx6a@H=#r54<JxJAn=`4fUIAWMJm=qFMX0)S{g{Z`r71j}mQbMkA
zT!oPkr&A`A0iAA_)iY-pL>XC;BBbQRLmxo(mlzf~H(z**S1wHP8$BL)^dt{Fa+Vh^
zzJ;nf^t<PH`nbo!i@!x2O9nGRr|%;Imw0QNK7D@vr~d*U`}8My_>qTMUR|ZpY~nf&
zWm!;_C3&84d*c>gdG;&(_V4^2U;p~^3<iTa-}BpV?<uGgbCG%}T^b9kocQF2SZ}ql
zqaj<HdmLX_qPNy1n;as9Mh0H32#v+e>D)+55H;}BQ6`7Gq@x2u4AyV3vv-I~LYXV1
zkx1o$;{b9lCTu`Rqy!^0Lb_0Ztt{GFlo!xwD5|o+YlO5r3*@sIP&3Z7HEiF$#=T1(
zT6uVd;L(ph#S`E6X}<C={~52m@f;7(U_2{0{^Tb)jJJr5p%FH5y{OJ6V6aAjsy94~
zq_h?xN`zL#qa3|#aivdp@jSP$Y?2oh%%=zM`dNI#Pt3!YcgWrURtoq1nY@t*tOkh-
zE_6?x=Kjw)EO;JCmXl7096RwKqUB*H*YKm}yr<08(~R@)C552TSwJ~1)4?wJ@Bp{j
zVLKTxIozuiIx34)4#KV98^b$+vqWO0M*0qj3Zo?&i!qXBt3y_nDCN;?c1dSbjJ9Y6
zg)Z6MzRAPKB8qkkWfbQgx`z*b_A|Wj+y9yuf8&?AFL$ZD4#z+CN%pocQAH_D&#Q+d
zuoyH#*f~hM_IY76P@xM&GRcs~9E9?4T1U9O@yvW<t227WzB5oG<_G*8cBQqvOZ)NN
zK7W5p^}^QgqBIJWK!pv?f9z=%jKSCpJ&0L9dX%6OlF!EYp^wEN40Q%u$T~pN6;2S+
zUS46mzfC?k#BFxCc6%RTDvG?$f+^Rn8{}2(ZL|wD-nqu;xE3umh7wcN`i+iX|NDU(
zKx8QM45J{;5)KDDoIlmYoIg!ED(Q8a{Mb+b9JfF41i$uw{Xen4yUWoJeh=BX50FiE
zXl65_qQET@bUvdhVklzV%AjgUA%&?(GeeQo*=ylQtkrDX+N3H=l(2ZRp5T^_!uJED
zs=?{jTC}bxw5gRuzG0nzXPcZ~M#G><yUJ;vJd1y14OeE2Z@-D>NL<fF>u)5KD_Ns}
zOj$Dst30QO$MxM-S45$QP-+fvSLmXk%mw+(Wm*QLrNC7=VQ6W&5(kBmwl0}v6;8uJ
zxeiI0F&j_NrliyFv3}ou?BrLOOoj-lSUdkP{^~JCTQ_*^@&=#(>KMEx&3451{m=(U
zRLsqdEf!kmSv=zM@JfpN)!!va1jCs^_C36Y&p3_Ac3<cGsngv5#FKpNV;|+-^XFM!
zTczD@<0wU*CFFTdk|bQdc9}1I@k@N+3t!@mH!m@ZWB#2|zTcr{=RjqFfXd(+gPRo;
z$sV2$wu<pgomde8NXJE27F`xtUDcE0#!wbHUKmhJ#{{lKgf&|h0cln+K8$#C_cUAX
z0^Kub+1wekv=9;RZWC=@<DsPkj&=ulQH0J?5DrQzCX)fy8ls>{v)SWgKllV*DAA@M
zXtg=^*rRxFyh_^k>90v1dir4&7uxie`uLI0n-?|^HzhWY_)wfNm^h5JU=PW&<BHw(
zDY}d6blVGTZeJ(K$E>fN;Ka$3EHAInY_(~IO`5>s(h840`XoR1^Zzk#U3`;Y`k($0
z|MZvtIlKD@bIHeV-(LQnGj#$93asOB^qzApV!<Rd7Fw*W^@utlHX9>FP?N~#+#e|f
z774bhn_`uswYb1!e~)4`hRCwBJ;0SN)qK=gxh~Rmu?oPVDhDkzg{V{HiaR6n25six
zNI#$|3aAXB7hwHbDBCy=<Iw@uW}G;6g!G<!Nd_6du-x;&;~ZZ*%0K*1f0oH>SGo7;
zPjTg*k6{Z<CouTd;8ZhIHiI<Aq$5a%R6^EidtH%ahS>~!N1;3iEiBn^hwU5J5!T>I
z2ptE<b7^+k^cNOrL=jn<P~;g&k}yqEhT{oUbw}s5{@(Am-6?}=3@VH`e(Ds>*#y7W
zrnlUr*>uq56yJ03gMdnFl$`5^3WXK5hON~GC-iA{`^3W`7{O>#V6B6)=B|8<@?1~~
zghuK*=t*r=`&`$&G=y8jVXZ+q5}^$?ukd{z?fGO`g27=pI0R9$d~|{8f%};3%+P7V
z>e_L>=TH3<fB*M?o3CAblSjV)NBQk?4cR@y(wPQ$f-thUd4fn2oN|JV2P70Ub&e@3
zrm<l%leoUb4L#y2XL9o*qupH`2@TI>q0^<mvc&58I!BKlMF@ze6K2yXqw$!{y&Z0E
zY?0^XJarG>2|gEM-q$GS0A-=*br<PZIo|1eaE`1JdYb9pbsS*{<`@$0PKDPc)rzui
zWJlBLEigUUC!bDg7`FxASM_we6bMyc;B*zEbEry7j*qDfjx2E;iz_6;D6}vL$D>Sh
z{DzC?x@37uQI?D*LtIbN>a}P*^dOgCy~e@rHupF_NAG(acl{LCU;8?)YVf<if1P+#
z(pz5S=#dVOJ#m7=tpjAP=`}2EH^*fk=k=E;QW#BZr2{YU(T*bCy2AKILbuuE)ag?^
z^~4jLzyE$toIb_c`Z{6YGn-ABO(u-SV{Y8O$rrx(MP7aNb#`|TC@YOI@4vtQoi=YT
z5f+3tsLJAHCFytrKagmZ<LdgtDgfoXb(y>}7+qntMr)0zDvGki3j@kBrRhs1wxk`o
zDCwc4C68xJMjbBi9c9C`aaNAAJ)Y8UIuu*mEVIKyYhxCh6THyJBpFtldF4cNc({${
zdNjf|i!1AV@=rWS)blZ>zzrIl|KJDN`K>KHo03&iPTYSI&-aLWO@srNu5B?Lmw2v6
z@!>A_#yP{O!#EeTq{poaHdT+-xpjW-vk&97yM)bvIFCuwjApZi6Z!;E^DdbG^x1QK
z=6ipD>2$`+FTcny{ZGHhumAdQa4;OrFLK|xS$y%X%Z9NQV=7P<uiHU88n5e89^_ad
zC}#<JFv3~vA%l>78e?^Jw{$C%LT4F{a`8QwO*OjIlvE7%mw93G0QP}LDB_$~uIw@y
z&)A)W96Py!d+%AEdG#$GFJ9){(irL1J=ao6vNT0WSx-$n9$wU<sAg0}h8KmLe)tJ)
zzWNLcO`mC!va-0qv6Fo!qnPjgfe-N~el+6h<qe+u{jW2!B_k@{3gOF>Db<DBOj6Lr
ziyS|Bj3526Kg<31Kgz<wB2g4EosRjZzx)fVtsUb7Pdv?$BggCB?l9!h$DiPV2kzs?
zfBeV!8~^R!<hOq3_bBerKE4h1o`X6i=Ngzu8exlOnbSIT56+Pzcrqs&-o}<0Q5d|F
zl6@BnLP=ToM^z=pR5Ut0W`}!J$)rxj<|VCmG(Xg-OEN}MRtZ(6C{u^T_~ey9k>duM
z$N~S(V1;rpra)yCDspffhqTBj@|45FeOm1XQKv=kfd_cw`8OF34r*DL#dS_R`V^OL
zUFP;+#Ko6x@b>Hb_>B%>%j2_u;wcIz<;v~bEHuusu;y@d&Efdrn~+~3O%<b9(d{?{
zkxzjkzWoMHobiE29_51{dYTV^_(PmMcaG)NRa)&fp662J8F`kG=NY5%h?ib|i7$NN
zi+uHKU+31XE%Ktk8vDL4eJMl@4Hhm91-|REa{q(4I!4%-!L@CCN8&gR+L*f&O?Uh*
zrNCAOoo7hL#|uO9v{vdU%Zx_o*1>RDt4W}136~=CNTvakJYZZ%lr0FNf=;OEc=aTu
zkqRLl$Th;TIF7;yNb?zk;U4XFi>TSA_t3+<IDM6(%&=Iz&LZcZ`Y><YzQoO&H@I3T
zu3w+wg*{rmfY1HNhu9k*^4be~w8MygcY%i=>~hc63+Mvk#9^E&mXwDRx}=i{`tWTY
zJ%64*{pX(MqaXhm=kLFd-ogS=6rmgk<`Y@Q7__b!jfY%#{dK<Z#V_*vzyAd;zIBNt
zNqPUz`MH9GQczVWEATYf=?rX2=t&ScjxAA2-t8MTm9D>=u`ut8#29p%VU>f|2uR~8
zQRr|G$1E&$>y6*3hZc045GP%>4tm_4MT|t7X`0bg1#LCsRDZ^i?iAlEa6AuL8EmPM
zjzlO)o=+GIc4@cUG#VXFoqqu9_E|f=j<F@mk9h2(ALF^pFOhHGAl`4X8!Lu|248Wn
zak;*i@|7=MKv;nv1pJxLtZ_2dOi~yp4!)<5VaU)xiRFhr`)U6CU;hLr?>R?zp-0$g
z(u^9lzv1rw2VhOfU;ayfg`J%pzWn89_~l>zpLyn)XP8Xm?<fkEcvn@yni?%-rA9>o
z6fH!gP%^7CkVVO4bDN;qME2SsC1qSw4Zwoyfv(U+iR%W0x}-7|U(~^QH7VKIKgmne
zla#0L$7y*ChB0Gi1kDc1=T2esg5Q4WWuEH1&Dq5m*LN{SqZ|jLHF=&B1`$fRNXMhS
zxQdkpx-4*<h~tkw!T#_5OL~Ezv$)Db58T7kA9;w)&0T_ikN^E&`Z0#X32(i1i_OCw
z)~k%PvTQft^)iMl+hk6k>D&9L=U(FYiPJpz;NzS;WzeM|NoE*Rv9+_oxpVh(;`k|i
zKSU}^yVK$0pZpY0fAE9++|T}b{^o!GxB2bg`F*nd+jr3aQLIiZn3~~d3OrpBdLgOn
zp{pEMfQ)?n$Ri()DE1EtdVQ>}FnLL7HKx+IQq=HpM-Y^PX<0JOW6HG67q*X*oIOf1
z-on?0M?bj6;2>s}!qTZ_jvndKTnM@HVh3;U*Ez9X;;1@vtaZ(~tkEkzj;*+T`4yse
z2S4hueEc5XtWu(U1ft}Z{?V`T_y_Lg{)bKyrv+OZSGe%<RrLN_{Py;3o{KZC<r;Y?
zu{UQ3pwVpdQ~&v2<i~&Nr;&b*OS9H8nGAXMD_<hZru?g4{UsiM>}h`R2Y-aqr|(5N
z9<CqpfhV8jAN>8l$N&5P^AGr2|HI!Qj=w!D%l97CDFxafH8`ckNi(L?UHYLzDKaP$
zJRxwTLb%Rdg@8&MtkHKH-KEx4SzbFk3aFBpX6SGbPwBQiq)_12;@&!*GM+7SeQ$|d
zSxDJhWp|X&^ubAo^wo$5R;H|UN<w7_S`m7h&NB@Pq+l{SAkP!p?Pa2{&B^;7q|-k}
zZ>@ti1;T6b$VWcH+yCaDvM5VP4vEbazuUuj;57qOA-Hy9hZpuUB#N~oO@8V=gDEV-
znPQqNI?^L?JoYOW(_iH$KlTwiXCESJwTM~~%}$rF(If~X9H(ydZguJn{|K4~9(<TT
z`;$M--rhc6{_>ak$G`MTeEFGYn8wNba?o4qfKEtRGsTR-t1NarA|pk(7NK(dyA7hI
z2F>bP(y7u_t^A-hwyGtBN}c0|K1C7Jl$PDfve<GER-%+kp)1NbVK7?Z(*7D-wn=#G
z1bcZ-BXW?*nBwLp=NgCHyEdX5Cb&_6jng{YH_!Y`r$fpjquc2c1x?mZo}s&Rim2;i
zOo4C$?)%Wwy#A}d#PNp9RYwx`e2(9@P8fQ0mfBdUnI<u}ZVz#!%fWC&cPVF_S_U(h
zL_?}vE@!amEK!|U<oiB;FNAG&4>w6)o3XsK&f3~Ki;IhdQH1hBocS$1YBX72KgRcb
z&kyj|{_2pgeB~Mb(J%Z0pa1+9n0$lJzunHa?__8zjWrr2z!egkW*A@K_!3oV%Az6}
z4)Iz|lpkP>z~nhb*Ti;fK{yUYn&5dZN;<SW*wvO^;2?3(wxF0Ows+Tg`QQ|z{t_oY
za*9^v5bx}=d~%g|H06c2-{$&_TRgI|$x^4l3Bp>PLMR;Nki;=sSGbOc?*{l$i}Q~i
z!Syv-7bvH}@y8$M>;E`lwc&B>)M?heF5mOHj{@}pcF*~fq^V|m>wqHH+}N11P$tZ#
ziZO!9g@3hsKqWd%g5xANZ(?r!B2u~Z`-`kBt+T#<jFYEMv9z>AyW6GLTflWZr1J1X
zpA#pW96NEEANi3V<@x8n#^3vY{XKr`_dZXYWZa!lct35Psl|e^btMEeVbrA1T~q@^
zoZ<K$ey@X`Oen@<C`+71gJM=svlmr~F`B?pSZd~%qbhQ(Nb-u(OPLN=`NIAx-KQUC
zZ+n+~Gsbv|yfR$BF(9)APkiVeTI-);>%wu=!N2C@n!yngsH*@*&o7ZuA{-xuW_#lb
zUbBfCc33`o7VAsOgH59DJ^YJ*_Itd2`5I3?eGk6p)9LgG8WE$>jC^OCm-a_IUu9g(
zHTg`D?cAowuOor`?!TA+^MCeVaQ@*(QI2yrr*-T0RsQ?G^Is#C;?&90eCm^*<pWPV
zMHsa}Is{RZr$6*z{^9@kUViZ({}cYkfAcrlKRA4!9Q^Myiz#!g3NY4y&hbovp2qk-
z$XQbRwyLl}0G><ex)gDW-wc?gF;oSfa;d5kAq8&GAhH2?1;z+A#!GBAdU)Q5Fl^HF
zTP&=_oPTtcu+?X9sF@z5G#V~1CIY#CieC2u3q2oW3{_PJb!k>&rH{51<5ynB-v1QF
z7+MPpWJb~$ZSau~ev(qS93GB&=F2aWWtQn^Mxe(`uRh0@5BJ$KP+ChMEK1hoh5H|W
zjHQ)Tp84Vz*xKD>WBVq9!6AEldwl&X&!H4FS|K-XUE|Bo`~jc()Mxq8ANg~vF0W&y
zpxKD{KmO(agoVXE|JDEf|G{`NsV_|5!ruD#6x6|dvZbnOUYW8;$3-azgTxOS=&{9?
z70GCVNMgcH7wfsG%2Ff+=$giEltOwgc@h(NK846=`hs0!X}NdGFge96<ks#PUKt%H
zT|df^r&cjWad7!MD<_Z8Xa-c1gfG4L9A{?-Jh*t3b|*ymwZel^3MB+_JVVKP<98ho
z#}8RKzJPQ~v@UQhEZuW2et&_$7MwnFimmZJKldN~IGtXDovl6GMuQ*tp-*t@<}OvG
zdHvP5nQDbiVzN>(j08J2=Y{K=*#0TT)+PJm?<1f20#VdrZS@E%D@RyfUgOA-BlP+`
zl%rT%J3_C!fKm=fa6DK$a+DwYiJ#yHfAITx=1X7VZ~qVfBVYc?SLgkC@NGNf=VgK#
zH=*nPGT%W(4XjbfNFu5!)*5tCGTzuGXf$wo9jxo1(-b?$mk1@n_pwPumDOVJWu9V@
zM81bLmWq<{&~kJ49IvKFuxHNG><3K8DdTF26&}lHPIK}@Pjd6MH`%;-jSntg;z$pa
z=VG)*xe6sEX_C~bd8u%m`Uc<VHnFOffAU<*;;FN=78kMUm?Ov5xwf~%4}A9HEG@UV
zd2@$ux69A}$3Mp5!H9U4@_WDi9O;@znr0-*a=RsXJszX3?@*{7<NN?#c!h4S&*?Mg
zx#!${2<xD$iph9HUS<qO1Ma!!K8_waf#(Kw|B-9xFD~-AANmt~>XV<~5B}f}`0xJK
z-v9tX|GwsH&%HodnQx!JzrpqO|7moEp#o15E-a9_5G}W;COa@!3K3~WI^0Gqc5#|b
z((weUYJQupr<#@LQzS_}ai}x$%AhMrW-@eQ*xX&?rTiSdPdrY#KjTYZdW&hCVvXgk
ziyN$-XmQ`ukFtAtm6!J3<dO0MD@zr2J|!=txXT+8Le%}4j*CzZli3)jmEw49dTYn&
zFE0=eH|ZUHg3o{LDhq2%JpSY<Mx!ZS=<uh0@_X?;k6-)MFSGOZ4F;9wKvcZ#X|_s1
zx+6(%T_lbRlqooM?lgbpKl^Ju{OD8k`-=pDhm?}T!5)9-@BR(itsYN4`C%S^>?yk4
zE>QFDqE?%af8vuo^6(@4>aYD8|Ifeqx43-e+WS9GWi3XRSn3uBtuwqTM@%bxsnL@O
zt|_s?l9~$F_i$Z7HBOPPkLSCj(<xGbBP2zh<G2bd1VPwDc?C1AX{H{pY%g=g@o8^e
zz{ra8Pn_i5`&+n9he-;;X_F2UPz7(sg7N-?G`(k8Uu$A?PNgf1F_fhyFAY+8Se>!|
z;#XKbeh;>52s%AP7?B)oaq{d3u||^3Dt`YrpJ(mZI(s_<u3p(@Apmpx%e=6&&E-ln
zF&fWk0;O;rg&#%y%+LQNR+o=4J{*$dF@xbgv)PPGmoD<^%dgUG`E1<2!V51w%YFAg
z!k_%%pWulPd<4(&K{>SAO@97oeww34j__Ci^S{o<=GJ#)6}GGjLVQ3VOG98ZRh8qE
z1zMF9(-EGKIL)wz!Ph#1ASG@P5RWs;GRAXVs#4cOE!xlx!RCrR=OkyIK1;W=Og5Zx
z_0|^0vm-qJ!X?h0zn4cIKhO9e<;AaF;*qCMars|Q**<ucMwr*tpsJo!wWdM_J{ZUk
z2awFLy25L;nCb?Njf?pE18lg+$<8uvzeDP^xxJro>DjMw;oyLc3M{&2CQ3o3>n!6}
zUw)aZfAueurqj7Jszhr|o);8lEkpnVg2-iI@sO?UTfFeXSNQWk^OyL@M?PL(I*W>*
z`ZGTXNdAle^1ou1#Q&Zr|HrfU6x7`*sTotDZG~0PSm+XqieMo^WJ4U^M+ZK>wdA8Q
z(^-nUvVb*^jV1`7(sf$ikpjo{$<qwi^Knc`k{ilGkm?kZ32yB!@N#jIrB6N1&i0sR
z|KJTKX@)g$?aBZ@)I9p(vn+q&BkWw>KnyMS*-Nx~KDMm;J?tDDdZ(l$l~b!3Xc!F-
zQO!1<-z91<vVQyo@y-=mF6<7+{L(M{2H*Pwj}bK-(oA#w%nFY^bw8uwj8|X3&S7ks
zPG($FIhTxPJax#fPnf>;0%cL*q4~gvKgeJCzx-9s-tz#BW)nv#gn%qh`EUN){~N=>
zi1X(k;G-Y;6!+eHAD&ketc7%FwtIZ$v!CU$#~<exfAJslxBth#%kJL(w{6om*3JRW
zbzg-x6;9;Q=q-?EWBi^+br>T(7r)oROlK66DOr&d_Iu<>f+#AqodX}Sgsx9jRduO^
z6vy?LB$~8H&|0#!zrxp(HI_g8B)2wpxboF&l&&JrO7{0NZf_j&;FAd_Kk@+i?g_5E
z*@eT;v${H;mY!#as(FJtQ1@#o7g3cQ?A^p|H1WeOVS9n)BS)ETT%_e24o4II?%)0=
ze9vbeW&Qup+<Qh@c3t<KzjJPWxpL?nfJP%T00fvtiliu#3YJY#lte{a!BLVWd+g;k
zUJf(kIQBTmmL+Rs1uNPVDKUzfM2_76I>*ZC<(p5K54X?&NPrZqhB*66SJ$epzVDuU
z&)#S6f7sihR4&s~)9gRIho?_oAWajltdBWAh&f>jE{_c=^@!K5kY(qQ*3xW5eBVF(
zVLtewAEnl4Bjt10nP*Nt!N32H{}SI1ICA(XZ+`PTc-5V+rdF$Cg@bTB8m$iRe)qe%
z^Nu_C#sB<Y_{IPHOALmiKc5BP4QoN0>OIv~V^oPKOG3v-yB_JNPpFDr{Afi8a2+Ui
z`uIlRI1-gq5<sO<m{Q?;4k8S|>gsDZVpM2IWAecwpWRx78)w<xxXO5Yhj+a1RSd^N
zdOHQB2sn7d0t<63njMdwi<6vsTI0HpGBxF2;~N;QE4YHS7~vqKhoe(A&pbw>GfAyE
zgJ0`VpPplM;RF*m3an1~_#b_N3+FHMn!9hH-JayZ2cO}oCoi#dX_aPrmQSu+;fYbk
zxm=?Zc-whJdJ#~oH~7&X{Au3%?!Q?1oL*D^v^b8(GtWE)Li6yWpJ#q<FK_#nZ{u6u
z`YzfNvz7j)>(iN>;CtWy7rFJ;Tlw36=kM~RFW>h~mPWCZW&RukQX6~=O68yvYMv#}
z5_Fc~`#ydck#x5S+6}UzsB9q9m@J8Lo$6p!WGP`-C#W@u{4Sp7Fpek4_8#T6uilIA
zwo$o8=VP=n^am@v_MW?Ny^yPC`|PZ1Zh6%{POrwS^lEe_N&;V?mBQ$XC0%HZbOW@7
zY-t6VWmseIeV@{6qgKyxh*cJDKY&+nv(<~~_J>SOHz+cX@!1DBv$DkbQc+T2OMTxX
z*NQ}0TEGL3KEOkd-p9)7RfeMhX`Yg0Im6+IEXy$~2G?_FPt@4n+Tr9=kMZU=ek<Sq
z!5?SuzWo)X2zbZazJ>qz@BS@+^>6$POUuizcn9a@gte9;%aP4CT7WEL{N8!c84$um
z2Wc(Bb*fIOQYhcUYgA&+Ryau4M-?S$n&3DNx-1xshjjb~LOZM`k_!(%#j(l596tzG
zGnfDI+fNa;H4l98aaP`z<H(XM@lZ+3wkYWjU6Le0n-Wu6qyuH1BEtx@#w`l;Xn-m+
ztaNeQCMNDOm7Zd_R8SU{k#RXCr#U$+NXC6utpX&L3NDtK>L4nBVSmVIFsg`hvLcW{
zrLtt6T`w!hbH!-TWoK(lJkA*o{{`!7>%8wf{{m67jY<-J@Pi*@WqFl<{crxwZVmLx
z#Q#!Jr;Wzw5=Rv{nWPZ}#JwGwGRJiUi6<*x*EGh(V%B%?XQn8H#FUkrnl^?!Q+U1u
zp^uBf(+0edvC^0{A@48pg{?)By>$kgm+AMmc>8y~iZBuk`UP7%8FPpBurS*q8z*eY
zd9Ivz9l_?18-i7u^%}~Sy9J5al~w^GT!eIR@|cZN57U~Og_cJ65uJs7bT2+iO^<1|
z9Ijlx%<ukYk5}Jyl%qEu<jk2B&YpgNv(Ky`%Pqe0*%Pd<k2sl?td|<!l(=NLyA06w
z!XkhDr+<N0-Tpe9(1%^0Ib#jRSk9e2$6&P0m8EC+!k0eA@tf}8eeeB?+<NP)@d6it
z#PLFArl<MIpZIZ(-+Y{(`-Q*5xpNm@kq6>y*SzY4SS-pENL?UeLnIsuQ&RTExLScD
z$h5%?JRB#2QHt%43B!;y?eE$E0SdA-!w&+qR)n=WL0B+KEJ>bHjHkGNeU1wa!EpT<
zMn#|7zvU>m-LXiP8aB65yv{T=Kft6p+az38bJ*Ukgu&y?PB+l@S<IS7>*~)E2Z@7K
zF&j_Z&-9*y*tS7<A+6a3w$46GWMby$8^lS;{a-xI!k#9FZaB<ik6h%`Q&)%wIa6|p
zPkri9&c`|D3q@)zq0#t`#CIf02;TAS-^qL5{g-fM^%?WLAj>m~qTs^$3!FZ6igu^Q
z>e@LTe)NkRyYV*O_ujw6t6y_(C7Mm8%{MnU!_WWh-{Qc5{rrP}_>b8Ba~#!Jonv)~
zkfMSo=p0{XsB}mW3fwHKMj%?Dijqbck*b2IQ6nEGgcB{sX^cu@q$BY?m!ilkXBa;q
zY(xZMNhT!|#^YpfnvEzTWD`%-n4Fs-DYjWW1h?IF0O8iz-ptup-eGzo<aE5opf}5e
zY0~TzguaV42Bj2Q8_GhHD2s<nrW5*)K8m>GjTLLvb7@UZvvMM#BL>|4y4w(*LozOS
z;LA^t8;Q;g+uJ?%*E817KgO9spA92Ot;GY7cGt=fUi-#3u>a6u&Yd~U_Rc1&n=7oX
zuP_)6SY27+$;VFO2M+am#OCG(=P#V)lb`w+-~HVm;LUG*2T@eRA`lXeA34k~{l|aD
z&;M_KpO1a~la-(JHzC;@Yfxp0=T+KmqAZEBE&_!t3bc&SWdhpZg+4+`%Dljzs3BzK
z*DYKhWKLFOxKiS{4x@gL0)uA~LM&VH6w6VDeP*3wGlqr_pL_T+>nqa?({&#B%8<9d
z{SI8I+1wf6XqU3nVPj{4y~9gPw!>-+rvzw?wF(jVSfwy=j2RD5WsZwPHs;CGlQ@rj
zn5*kCh#GF#B5F?1pFYHJ`4XG!m$;x5nUz50axpH-Yw}9ly*|JG-~T(VV{v5FfyP$r
zqckY<T%PRrGd9)-OwDu|41S+;=TGw^Km1p@_ue<*H(MlwA#Zu(oA{Ui?4R+|fBkQ;
zv-AI-QLHr@r^v9zK^TiW?&9V>q%JUWj7)M&nSs`1>)W_NL|Ch#9f>RzWm$r7aNH23
zVu~b22tgnm{4l^;m$D3K%o<K#-C!qga_`*-IXtJ>@GWOfj<|AZom*}_%=+poqk%v>
zDS1+&%7Qc-qYH(MB#~bubVF2`;S~isP0*?YD{+F5oz;Sy_P6OYOA1rx^3XEwgX<cW
zlL9LoEMHq>UBi@Ci%&J-w=eb+?5cOs@8#s#7RESy<d1%jZf}PVeBeX0>kXVz@wfiQ
z-{jGUALjF4`r<1ldGhiwbw+^%V@rh22~3HOV{oD7SV&8<vOs8qLgIP>t}n?|f=&zk
zMnK$+(M5slc?eHZ<QbmlA%lp>sSyMzx#Q5bK2Hy3*seuHtubXOnVa9s_D+|D13Aa;
zIKpk=B3T}C?x|(=F3i$yL_9W1k;4N__E)GkT!gI>kJbhd6on$O0^dr;S}}b5NyJ^>
zT46FA!NlwwSJRTlV2Rhg`qkvt5(IUgdFB*nFI>fHm+@eXD{`X#8BSa|%Z1!9q^jN}
zRQJfOw65OpEpNs3eV%;$G1fO$SXp0Ubz_C$V8G7K4iDb{AfO4ufJQrFb8CagpLm43
z?|K98|4Sd@(7_ukD-~IpY`^vO_wrwV=@<EHfAeQKaq{V1!H`#O)7NE&A67O%!fITj
zaEu`tkEyqP80O@rz!W9Y5rmB<<K8x5yGfGd1ho)1@+yrgYw$fEfh9{)YV|sv-y-bq
zkQjqweb&=i#?vz#xTQ|$&#=cojM0jX%_~gJO>_LtDW+%l@Z`f62)qV|4=l2|nsZ@g
zmf5~XdjdSCvYe{0HWpPHG6TMF7#c%*_5#kg?yAU3XiU#?F)3(tFY=bxz5yLYC?hz3
z{xT0f^axSVBnT~a+oN~#aZdGi*f5d|Z~(`Z6xL8!K~$?TvpCP^zw{?;ZEvu&bcOY;
zH3q{zaU8R;u|cobV|5wAD4^Mi*xBiF^5hfTeb*cK!1w+LM~~iw6&0$=6Os>n&tIU~
zYVx;#{_nEBxw$JBVqeijuXW{bTnUUg$U+ggE=8H(r~=p3<k<k5=7eF0Q;*PtE@4zd
zP@=RZsMjgelswPS(!q6IiXuZ;iyKC`Hm4p)l5vB0>L_>LbDYM+1nEw}%KAD=JJjkE
z-2dRieE0jmiyLokvb~)0*u&3o{PsOOu?frTbIc9SFgID>E#z&*<Fcj_HFaHVZpnAL
zh%~J_%|hTz?qL`!!e_q7LhA@(ashvC0<mW=gCr++BA$72osAP;;Z%RXvI2v~3UE9R
zZ8gbmC-~tLPx24`@!zB0+a}B6Ywj3jS(2tXMQN^$Y5c%pa;inQ*X7dXGu(II=lP)@
z{3{$fe6(^|NK58s=J?nD;-6zJeC*?&_@<0vjjAlYgdZU&aEc+`@G{0^s8Zm#b4X)x
zeUEIM;CYqVz4Uy1Phhk`TX4Jpfu_te@+=|n0+KkR5lCDs$h1pblnj^pEV=>By)6#y
zo#yhD9m+tH_Q#xTE^+X{96D{#-5jzpUSY8e$&!qsCMdETVMJw7C|!i~uqq`m1~nd`
z${d7Gz1d`KSa9ov#|?8Xlw*u7*h(|L^u#JxdIPqToLpL@(HNfX$~>nqh^<)JsszNY
zR_d<UtVQfL4k~FPqoFJ{NitwOEHK*Pzy9ifVrOTEcfRY}X|y`1BIDcM`A&ZRXMc|W
z{U84C6lMAUi%D#mW^|u^ipj}Ev??if){(l{O>l-hOE6fxC_oei20^*Yl-5e2OI5k_
zn36$KF;b-wNaInP2(cJuBA@o5X?l|*ZalJw#)M*Vex7f+?I!X}(r87*@qlwzzsRL4
zDT}UyCrY$Zbh`tTHnf8#Q}g=}jzGm@Twx*4QALKq#qmP62L-}Pg4!G>RvaGR+Qi?t
zm%*b?P-=^|*AKsHfh;SzbYT@$TC!9#9w+?xkNh}|T9e6%Nq+w4e~yP9eUwrB%8X%O
zHc<k0Rl^ERASB9I?08Jj@X1wzipTiE!L5gwETha~f>wjHDDmA0FNi4foHR@D92eJh
z$g-rG_jo=Y1$9r5j_SmdN4V?Wn`ul<l5Q8QZ*CB0DKpbYc;d;Yx#`whdDY#Ax#8F%
zPd|R1+$ipT^HI)x)Uw+1>9of*>;A5s#_qN6p45>pR%`NJ51A%tt8leLFtdm5NYlLZ
z1nu^Svb{jvo8>LHEuyU?5dj<9+dTc1PjKn#X-=nxv8~k1gapTTDN0odYhwBI7r(%h
zPd&k4utTZx$}M4cG?-->rDE4Ceu!L!R=dXJ)PS9x4NjbRgzxza-_N(d`@5*sYZyPE
z$j97z%dPy!fB*mR6F>FWc>0-B*ZrY*epF!8Duf0v&j^A#N@bW~0n$(mdkCconoWe~
zp|=zSAx;!h8iN;B-5ZrBXcgmmK2pM{-^YtWvT={8<|I0oZ0dc?-MvWEn#4;b<MD`R
zE?i+|_Ew&LYKg2^;alF-=CyCQndM6xJaYeeZoPAn2QHSZ4%*BPHks_yP+DUZV5(x&
z30$mnpi~sY0X802j4taE%<Q8-G)!K4oZeF>h}{mYsRfR;XSijtNzrO@=E52mPd&u>
zL6;>h(YDf16H-uEC{3k9mJCPy=l}XEIFwkca6JiBT5L)yN@Xd_UFKFV!P)H7Z1tF&
zYS8O%^UUcJy#FtLh<AV6cMyg(w6d5i<2~>CHjK9X%+LKiJH75JjKQ%lz@}+q2#OJ2
z(nl02u5~eu8dg>KPRIAabuda(8jaVeQR<2*UBMg_MVX;>S+UW_Ba%TE$E4JxCC+P{
zQyKmHALU5r2>aRzld8jKK7N6R9`rFX<K$z1$UDC64)Sz>umR(pE-IAlm?`@GlvY$>
zT})}K1YhJxS700$N`=TX%xFNVG8_V;>3K#Y#o_S={R>Ys9B33S+PcK`NtPRX=yuoG
z8gyAP8jY-ywuPi9UvRgUmseR^U1iq-c30PJ_x;hJZKcc)kfw^!s7q~g$n@-pH2nn2
z%S-(55B(&sfBl;Y+ijBGfccp@{@MTWf3mf?&4Uj={7NqNi-t;5WR{|XCMXMo9o#ZS
zN{j0Od6zs+aHPP*1@SncH8+Rj`xvdDC{aZT+MsbLOohqK#zP!j>a{vTI#_&M6VaN_
zxH8HZZor#fJ;AL94EDebk3P6Yz3JlD4dbmP*0x7zBhY4yQHCPVh{t2B2FDe2Cg<_2
zCCy@-vV<bXsDje*@WP0}P+^6`;^Z`$@mLxKoLbqUIWvtn9y8XJU6^K<x%<4ZU;UMR
z_N(VQHqTYy&nXrt1>K#DvRpz+mk)pV_vr3)dH?%ANEG<U((t1{{KI_ybAQ68Kl9md
znrQ9w!<IQ`|LDWa?mqyNlbCpb<C|U0&0Qj->*BaBezT5F3rt?(g#m;C&Dklkaf}}7
zs?aMGS*gjhgiO}hH}5g$t&%vmb7ku^YmGJD^v-wSNQp8fODpF%_32Bjt?sbXOPQT-
z6JGAqh<Xf11>Kzip69YKzev=cq);VUzfUdlF?otAatd1_q(_oe&tOWOCpK4@J${G>
z9)5~EDPECn)L220TdrPNrzlGDJjJMzpZM{grWQ4L!yDhox4!*t{Lhbk^cDEu{}+f7
zP@P=-a*UsDB2<B=Ejn)^9f2(sj_+2Y!?HR>`P0oxh*C&!T!iam3r$gENGb6=k5Ru*
zhQ=2;p)kZn#JM7;_rSv}h6g!1U2>o?$A>?2o^z-A$S~&1pTD0w?%ZPkfd)ldu)dNp
zKQ+ycH1v}W<M=9pSDhS;0#{b^cPDT$LLy3y%yZ0WNKvM^A|jZaC5a&{#^e{iNPj11
zpd`vT<W9iW)FD=KP1#%Fe3mj|*ZpAEr*@pZaKpcHWtrurr5D}swYF$0cXg^NLPDCv
zbaxUaCTk=~%K!UozsSXl7y02I`YX)L&Es1^F^;+YmRtE(|NLL@V}JRtu)MnRXL}01
zPN=f1IzvKWOi7e(Ve%_b6o|5gZP$@T?5e)oU5NvQs$1Mfgi;!-Emk^6&qwD8S)SlX
zi7Oojy&a5-X}VJg<Fc7LoIUjv{$Q53-q~hxs?EtqCBO3<k5TI=v<kR#_Dj6sjYrwp
z+#$~bcG8s1a+|&71mp2#>Y>CawF{&b<Y|TZ(?WvMcv@EpzRd852s;y`i6u}u3wp@7
zzeKtEB#DBd4LBiZSWZ%OI^?P<(UlA_?7|4Nd0y+@T6*2VuE21G!;sGtmDanC3DyEu
zAPPyIWej>b6H_U9uK4xe_+{4CHu%5?evpaD$*L2e74Lh`dstmv=O6x~e@vEVuh^Pf
zYq5ETBdYB|SWTF0L$-!3QZnmN^XCw%!1Y}6QCywOt-$p{2z-RL<Z*%(63-7SOl6Ug
zrzt@YAPP<0ckpm1jYLyz#%rrX0%@4u+u+U{XE?jMgYhBWQCvQ=$<ZVG7!EX}Ue4<F
z8V4j~v8Ko~N}X5OSqrAf@N4agg<a-EfQm;rWr1>hYV`(jqA3c++(eC;c8M;*%8;|E
z!`0<;Yz>BNWF=Y(gwd7G=2ykF?u{0pcZH>2kSYW(5G1k!Z7joqB2D_Fu|kS~U;m9?
zVP|KDcfIR9G&>WNdBVZ{2l!|I$N$Oq|KJa>w!Z%7%7-0qZm{*}L(IS7?UY4EwzGjW
z1=gt~i_le<QhE+SvyK?WWMzR1xX-%D$lY$0(G1g=JQ?Fk7n56BlQl}KX$50$oIAj%
z(Iv}DUVHa`yhwBG&I3#?&hwVH1{kT>*jVSpnJ;m9WtEx6JYF}$aV&%3h<IF$cPBeD
zl_i`iC`Kbf-@)W5y2vrEgY*KD4En=_na&)Sx(=T`w@LHZKDMu{GwhAg#-d+=J8g_*
zJjyxs%ti7{lO{Punezi5_yIgOq}^)qGe7gU_{swhG8(_KK>9yTSS#pXUE{(Ze}seY
z`>v`3uM?y-xIqA(gHaOUI7r7uMiQ$u<DDKwkyn1O(xKUCv$DF5)egf!hUZEy?`ZZP
z6zt8`xYX}*<Njkjb>bY4-*=jQM<;pc{)>G4!%yJLX~M9<@!Jlw;Dzj*{Z*O^F6(Qn
zn6hNwzB#6*rYX>*MNZM(q#lLj$q=uEQf2s#1}pI0h^^G2esB-xF0C-`4iJt+l<FJb
zjF*4+DvjO}*4MTHgSMJhy~PiF{|^yG5kK^UAL8Sm{NyVlYxA;-I!CxjBv^x=b`hf$
ztjb7Bh~x~eQz?rpRZ{E>u!TZ80n+pF9J|}C&<MxH_e09OQkyRG3S*UyQksE_XXUkx
zo>H+rh?$54^V1%$zHNbv%Y8DXDHES7XE!*scadi3alUt%m1IOBx@3h%nU$1fPLXHW
z(qMdv$uok+L}k!kWJE$x#zP!cqMVR0s-cu%6c<c30&1ZHWq}m|UDe=xcZ;Rofc_|E
z$AFZ!Dq6lu8UEVetVQX{)@=71yBbittkP>6{&OvRomw{5V&ZX0SqL6{=qsdY#`k{T
z4|42=<9NbBtAcy)xrcx7_y0aW`?r6VB+XutQ-^0P);dUA;^YZYx{Y-TM3$0!F^zD7
zqDT-}^6n5F`Z!^YAgosmXUC8Y#~8qIe1vBx^OP(}aU6-~IE2C>@LY_Rlsc%SZB}CL
zdrENVSje}2>rs}^uW%*r)7{W?H;1gRjF_L>!!u7UlLvxaB;-n?a!pz0WJ!!EHFyF?
zDQc~mN?Ik435`Lg)k(heYJ_2pTv-N5!A!G3qgG<IML7-54?NQC6lF55j&lksM-%-L
zKv&Ci9qpIKVLboLZ@^e|ZrNOqNfQm$<11hJ3~@Z-{qO&N4(vOO$`c%8`LQ4V5uSYF
z34ZSnejoh`=@(jKQ2j1;PsL=jS%w&HBLjmg1j4u2tVfoN@uUE47;klHOi$xR5lSm0
znlj5UN+Ylo)>0UQwVG^Hb!!^+UH1{`VH1zpn_PP9LspYEO)=*6H^S|A9wBW9JoV@{
zGYbv;MoHRTVY9bJnmPy)tX3#3iAQ5psc@x3;QMr@<}2abWQ1chqAajwNg*_@?^Be5
zL7a1FVH!cnSWoig>X0*|oW;G1Y>&od+8}ndHD63-<g2?kU%MV({hjB2t}J15V}w$d
zaiq&fKl(rEb$fj8_kNJb_fe(h?z`^dr+)IU@Q?q=KP6YM9Ju;xt~!o+`j0+LeeYgO
zdjhA-aGhOw8Y^%d7tb}gQlO+oMgeGB$&5&Whyq%(Q{-`q7+O5xqJ>MI=9J2^b!DAf
zTWvbw67|s--B{rAm2<2g+~SV+yb-MnR@at!_`%0GdwPv5HVDV%hMSI)oEp<~uA)uO
z*5)RvG|bF)n4MX`2_mvQCF}0cs7IJAMU{$DrvPZ>P?X??ElzK)5$vC4oFuGVS>}}#
z1Ga$Hki;eD&tIl2i`}lV;Dhh~5cu%=dtcAn-uhNP@<)ICO&`Hpz%`PchaW~xwm7t>
z&gMp+c0&?5l~!#Qk3l-5qcJ;cn+&#ksJuYz;@6$P!*8}I3WLr{0^cXi6x+oBRV>mB
zm+;Pij``PqCxfGVIr+p2r%qnw$%p&+<t$M%;-;GyI5gd4`4fN0jZ;_1ib|+|cDBRf
z;y#2ijEWLJD{zt#vQ&&mV>-rC76pNCsD%R0@mbGujN@_f(h_w?(p84*c{G03yZV~%
zHx{j7bEA*%U*%(;{v!?@Jj7ey{5I~s`)&>#IKY{+=dR-wRPpl0uNsUNkmrcOHjb#g
z{oFtx)QCJTF&G>n$a+J9T0~=Rw$ctlL0O=R0;4LdPH915F-nnS8LsfD)x0W^NDq^G
zOdoEsaiz<p-Xyxc%^Pp4b8L^mKXx-;`rLV1(Ija(#PcM^UFPDII*u#QI>u^4Sr&|k
zBaE>)LJ$QZt<DTmD~foCYc#SfFr}gp2FG#HR?-{g?3tRPBQ1%ku^I=Q+*pCuEYsGo
zK8UfVl2I^xt>na4|INPW-(Gsd*9OL;lFOIYt8uq<iGMlf1K<0Dyz0(ZgD?mUf9ZSv
zBA@-tr}%>p{~_uX5VA9)A*MAA2GppJO-G1Ufa6GTEJZq^R5`xqARS5E8`7AbgfPG;
z1<FvQDW=p2Yss`CDKan$TN-R>Xf#^jI2h{?3Xh3)3vqbB-o;tGKr-1*nY_W_?RUSG
zbC-7roHny_O)QFYSFdpLkpkZrpkt)4D5V$<$7pSE1D{5tO>1%nR6#x(R$`cCfmJ1i
zEGxxx<Io%A?44@Sb{)EUf)kgAoEdeQIdFuvlP4+kF01mof}EcFUB<v@m~#Hy)m`?k
zU@#c*BR~A(96opmo0l}3Eq?yzevYr)_hpt=)~@4qJQunsSo-o8*}r%QF)>9x=plw(
z#7tN{>k4porAk}}w^^&$a%G7U;JMFjD^^rpio$3H@rW`{@uY{zGn$={(i%cH;m}kI
ze@_c3*4eYz;P{=3sIkxS<2TXhw0YaN1Q=1WwY9-h=kMp-l~c@39l=@22&SMcb9Q!m
zl*-U()|r}KAefj!6&b~FNG+^7fx67mG6(6Ptw*<?a%j$A{Yf6as(E@lVg8OAdFa6>
zN#YD+U%G(*vqqKZE&X21Q%{{E%M@vvp;W;KzxM|!pHu1ZLm&DOzx#W?&xxm>c}1?{
z#ll$0=;}HbKlV}fzxO)`YIVqm2n)XNBLlAz_H`UQK>8M2YSQ6|WL#+lI*!D3J))?|
zFj+<_!LVE6hmy^)WUOGOu4#9lqJ7Kz=uaNviO;XHFnf~e`AI(hktccNzBPPj0?&sd
zH%)WL;YH3q@oUV^7zU#;gTaV)J!0?PIXo|*#FFG0roT-ThGfZ*Pz#DO!?7i<WANPw
zYdjLKhCkWn(a+tF59H6bU0;!~*Y@=+Rb0KYiV*O}pZqA(of+Qo&UX=o5#RrPALJ9C
z`V{$Bjrjk(Lr_W4l%?U^r@z9)TW(~yy}|i3MtBi*$1qn5*eS+<rPgRMdGs(CL#axz
zhAc@L^#%mGz=kfP%<^>CVY=x+9^-^{4%eRM!dFt7v-eUzw3oP0+;;0Uz9%SC&9o}m
z{?s3EWAqq%XKP$qUZ&L!*}J$8*RL~9yJT*JElbLNpQWcy)0sO!R1{=ciYpY2nk3hj
ze&}-Qu`_r|QcBHW7c+dl*4%LfuH!u?1!bzbKq<?{+769o#DD+Y-{7VjZ)Rc7Uhci;
zo^O6p$EXLj8pU{s9&95!0wXNK>{6PQ!gVU$s&H55B`tGY$Hft{!a3PzJwXj=nKK>@
zF{;Fq9y-gJm|38(n!w4~KhdDJw}T_snVo5I+pPyk#v9yl+s#buo#(E*UWMa3Xp`~G
z(`Wd|$DV+6kqOtqD+{CpX__(|#Asvagf%8-7VuhaRFRR721HSS$udltqnrXO1c-qC
zU__Q>)F$Rx>w0|V)GFToIjr<pU0S11yUG0(q$V3v0WB|GT%{-sc~<hP|L=dne&|QI
z^Y&L^t)X78^EdzI-{do&{{q|H-gUo*^@r}`hIh|WV9C0hh#~<zv^7|*D_o@G;<yfO
z(?tkLkyYAJeoz@bdr?SHme-^QvZ5eQ#+4hc62uv}%?`fEn6D2oGq-cWoTEEf<E?MI
zo$0wIQN70G><ovF9wpa`UVn#&9(<Cg&YmW$&rs7LK{3X0EaUNrL4Sm{mUg4f<jg!y
zqk+mZ^5KA5EkI`(qNozQ0T2O$;fS&*2<v-z@_fQmJ0teJ<`y1)^chC|G0MKOEJRCH
z*rkc){MjXnT$5!v|M{2x13&R&Kgs_62eHO*{O03){|7(FKl_*e`W3kzVJ$0P`7&1y
z9%j#*-ij(xJY66Ncl*l%*Kskfha)NykBkCrp>VCJaHB#HO?7Z<A$gpj205N^u{h*e
zL0K9$FRyWPYl6tzpp$Oka0C6dP0DP8d){)K`J=B!i<~Q0E^^|;Gn{_rDru~+I2^m>
zHU{Sv^<tGMD(Usc4EiHH*I|BcmR5Tb>HB1PO48k?-l$=+6kRGbStThTT++0_@gv4I
z;>^w^_TGL2XU<+=xYOTtG=C*~CRK-DDZy}%a^}nhilRWN9AV*u-}@n42WDqy`LQ4S
zF@E;vf8iA?Fu(T71c8xkKKuxDnjG5KWNocSvo2}2nn(d@lAwwbX`#EhMQ?qFY+Ny7
zg)vwu5T1kIY++3$5?gCV6iG=><>c`Y69g=JPjLFdDQ>;@ULHCf^Yr8EY^|T=;+YIz
zH>uTKZoFlYWBaGL^r_FXSDd8XGF&*<C5&A5?q8(Unjpy%N^NoC6faLOSxzz<;~QIf
z^GZ!Uf+z@BAC3tYCwbz;DO6cf6M|T2uJ<tRStpU_e{HR$C^Sn~HwXfs|NaNR!NG$E
zIeybE+;h)8?Af!2i<d56=PP-+VR0Ob(JtvQ<I0H(yz9+JSz6lWiKi~|rq>>3q8{Nm
zO}xOzNU=-a0uyUYX$b29UNlX!C`q!EGMAh_wZR)ZP-Ypms7Wm-xT$lR+S(dx>xYox
z933}6rzP%i1K&Qw(dj;&md7|95&1q-oeAn;gLFKgKOR7|NS@|AdH+c!m7rB?5hpRp
zD8?iSwNSEUG^=sK*47Xa)+w`;47EF9``UNhbtJy$qqVLW_1|EhtwjONW&_{zD!sg2
z#e=*ocGugoy0VGqIehGsALU1X;IHudd+*`be&aW;;cc&1ysV;b<LSo<IxUKHOg&GL
z)UMHeUDp9^@vN=<J8L1NIU-ZY$VK=Pw;s`UX2{}%EFR;fZgm>a)#*wXhMh~x9KZD@
zB6ovMzDX!<Vz{|UI$q~h_Z;H*9rt3Tp(sl(T{*`yr!TQ|WsU9a0TZ(au%iHf)lzq=
zZpYTv4r!X>`yO*MbJUw1gy&J@DRH+;qfx`;Syhx48KtVgm1%4+!o>+&oLpVPOgCuH
zc6jtN4`E6J!h!3^nt8E!Hp%CiWoc=hvMedef?xiX|HM!J<-f+wH{FUcnmg~jlka%n
z`}oyg`)}7h1)chnPceP;DDIvExL7LQlE6WLsO+JPQ;Gd5YruDLwZZc$nFD1sZX=@U
zdz5L092jEZVjZ76&B(1~)KBp`b1Y26%(%;J^)$YjVQp!ND_7Pye*00hD%jjy;|rhr
z0;f-{(cda*HCybNJw{D-h)-_O@GfJ8WovVrJS}N7LKYV038OkzI;2HP*4?JjsAKX3
zt*Q~cDiu;X<hek&5mrQ;+gWA))<Y~`S!R2E6Q2FduV_?BUTaI1DK1}Lp)4zL?Emy%
z{sVvOul^j<Q**e!<OhH72l;Qm@ms8{UH@*K1+sC$rB8o~=JX7~Ob6d}vDU@43O5WY
zw-^_!>*83zvk1q-YE9f7P~;ix+QS8&kFB*?T3yH04x?UzA35|=mwpcWCIowuB{l|q
zUVYmwJb2$3zVOkfxchAfsMja>wO{%im(C;v&IF$CFg+LY#<$+V$qOH+Q(MA1lCA9>
zjMmIgO))n&jj<MuASqJl?+}F{c`_o@lEP-#vLN&f5*Od`84ew4`>K)a)r%{5Qjq<*
zcym5`IF!WWoJ*IMKv+Kh=|5t+Gs8RI`E5w)^BwPbFaP(y`8Qm-y7cEVniEu(kY%Yk
z_n8Oiy!i&A;ehj}H}UFiT&q}Y){rX3r~*F>S=h5c;Q6SsL>mRhk|hblZXd5m@j{n0
zS6mv^80H00S%NHSMT&#NPjT^2`pnMX$XxpXp-5@hQ^LR{8I{Zd%b)lo_T*pU*pYy4
ze*?!Yn44W-YGMwhVHEdqjKGuyWxvnWr%utC-b14($ny-V3!)IBFkrbXSj$SftJ`?E
zj3`N70Y6qLMU@+$MP%$+Zy96BGsWuiI$_P{w|@KA`GueRdrVDEara$!ebbA&XZ_ks
zYmq{-u{9veOXem6Zr>k}4t9v$HZu!*u~=M3UIQo8!Vs%8#%Q+I)=07hV+@|kIMjBa
zu*AujnfY04>0txSk@+5b@*cgoAWkffHMC}-9tBujN|jTq)hM;X$cU#dtgx{iGa8jF
zz9uD#ySOGHQwgg(L)uA9p2p<y0Aouo8HaP1S5Q_Wqkus?sB{Q1U$fd$O5XEr@8QRO
z^vAgGzOV2{ANymTIdl4&RLxh1?|Dp3O*20?Ppwu%SixvCVsmSg;dnrnWf*IS#|6Fa
zfKPtvqrB&x@8Q^u#|V6%EX%L!+P$oz?)+yy#q`1+{AQiVbFsKMw#0Ef9M3~3jY00t
zlqGH$Rs;#9Ne3fJT~^E)$D!3|QKUJg9$>XYJjw|I$<|o1PYGt5n!V{dTm2C?-*6L;
zJaCr#KXr!M`vM$Ka^msx{Pu5t2G1@~3!3;rom=l%U~;y}%7x!&YQ_*}8Qa@E90YUI
z9cE@`P}X1&Bt?ek?+`^1MLI$#K}m+rO9Ic38Hw+P=+b9LOXhAqz$0IJxUx2L9GpKF
ze9Laj-WtdS>uWpJhF}%H^eex}&;Rwm%iNxM0x#f0-~R*r{vZ4gM&t2y-So4~9WH+2
zV;p?ncjDI@AYCvLM+lrK#JUpgRwFmhTBH?FYKkNy8I39Otn&PI9X!_~h?;aqR|rHk
ziousC>#@`=S)6Y&ANAQ>dWu(n%Uc<D4JW?1!@}%IX7+YCePV}?|NaJ^oyGNBbZmJ2
zTW+A8TdriEVs74J91rOC2l$?3ac+i*PKP+p@DY@Gh8*?rJ)bllQ8%)Z+bD9Pz+y^C
z=mjK&gEuoxCg9ZRON0nADbV_r0M&kV*lWdnk!x00HxUvHhF|^FU*@NO>aSB1bq*gm
z#5>;pcK+LM{>Cd}A183ID{DOSu@7_neeWT6e1_dIjiAy21C25o#yNrO5)X#-*S6?w
z?vN!ZD2){01U`PfhSzM7W*S!&IIf3DO8Q#SA9Pu4?qw=E&FWK4X5Rc>ZoFeZ7cZS>
z`qU~XAA5|IGZJ4-ARM3B`4(?}(=jf5`2^wm7dUuK)9dxg^Ngv<27C7ICGdSV`-94M
zE=dVg4y9r|9urWJf2=8~MS@T`2t3yEnAXe`pZ)w-kOBfLSkwAN)T^%3{e2$(rus(J
z<;c>4t5??WeUIP%gWqKTf&JWk(=E);&-3Oty@_A{&EI;3uIQ^T(BcTll}nl@Ts?7t
zcfIun>OsK$kDue#qw`En)RCe_;CYC!8ar8IK?_V=V$k??pQ&huW?2%)F+tnone$sj
zuA(S&Tn$>lY;#OF{4y6Vo?)<d10tNH<N1_HLDb)Z@@Z~tt+2SiPLjtUER*dvi!<|J
zEaPNMoFq&dhh&`c<d>gh&bUl8IwWaAlEsvxF_B{_F>FQwSDrjeCLDxcBTWYs&m*yY
zb;w;<frU!0#e8is@YNB89*su5l5w(vvMkB-oLrTaZ6Yuj=4`BO@vyj`haP!=Z+*+V
zc>TTi@H@Zv``>gSVu_$YP+G|-6|8LZIksnleFyfV3XrZP&x`7Sm*+U1i|czhk&h69
zERK<mgA{@^P00o^cQys<1D{@R%w%ti&g?$&EGNwrj%S%_OD3r>+yaSpKnX){yI_5@
z&(1)Rq<u_jY1ds2FElv3w@sF&?5quGD#O0Hg<ZQyL6$3G>#~v<2Ae}lKR}O16j-cy
zR!sPX_qp9{@e@DwQ@rC{@8aFx@g83P#@F+`-}il_+1DcQ1TfK=VE^6&?Ay1GW}}HU
zu(`Qe>7%Lwr8P!dw6gSi1D5<tJowQ4-0|AGX*3$&+@h|$c$G7s`Y1=g?Ol{!MRX99
zIGzg*h`^zUQyNZ$v4&(YVz9A8e`kjx%@IZ;9EYGD5jI;S+R*JM1g3J4a7#tE&~yhw
zrove!!!xWt-QvKveh0mq_i^Uw3mm-RJb5np^heL*=SOj!0IgjPAD!m#fxVpj@TZtE
z57Ft=xq4+CtxKk-nk+2t!FOG{gC2nqkfoLFjW&$OBLb?IPOE}iB(NU%Zop2OF?Dbc
z7cO08W2Z;QaTysy{0g3zpWX1E{TeF9vDSvI%^sfT<KXaLfBl#F8-ML@;Wiy^zx{S@
zzWF#0K6HKM4+10UKKUfH>m0u07{g4HtnDz-a_|ZttV5Qj*f>U7LBHFjySl}=H$W97
z!c<yL!gX-NfS}z$;E*OUVLe3Uim|nb;WoCMA@DYt={>~e!_(Y-?~Q!%p_DUEZ}H^A
zS2_7u54V`Wb)nu2xclxyw58zk=Rd_B?-cE}&xP|l2x*y{pJsM`5eW?An9vAhk`S2^
zo#n*iF*RGsj961rj|62YaHY@LVUNz<S-y1N!x&=-q@brX*Z0+XVSF{n7OGRdDh+EZ
z+XTMP1NJ@+eEcBa{T<(fSMzx9d*8!<|J~ntMesM?F4J@CnG5*!FLL{S$-@sk&f>vo
z!rG8LNoYAPp3TUUm^{nygF4eQGZp*M7>v~vdBJEfV6e1AGjPzZL*P5CY}MFFQ`)-3
zik#4gL*8Rt`b@^;&2PimyTHapMLC*aZn~h^2xw0?I50oS^5-7Lz4S?LJ(^Ju9G1u1
z)EglS3wx+HCMb)PQ8K26L}V#4jaj~Q1?rQu%930as4}A#Lf!F6O3k3|a`xf{^e88C
z1VgJxU%{Lzr9eved5PPCHE0d-C}(wf4L|Vs&ENcQ{GGr3_vuVc@{V`Dli&KC-+6(|
z&?|O*yU(G*;s}SWtv+d*Gc_4-+r9u-^%=*4>4imv<5c7?S0IH%0$x<B80f~Zv$;W9
z<lwous^q|gWHfZivJ9nDqI!!ghjvSGM}3uHyv)w9AT6jmHZ4u35#l;EvNFaGJTx^L
z^#;8pXMJOhonFQ;&bjsGgyv|QP!?EGvep~XP7|^;p%`^5PuNm%xj$xmX9Tq-nNp0j
z!HO3j#t9)fu>Sym@9+FwPM<!_7ryufo;vjmS^m;Hg6q0;+7ryq&Qhz@aHU5a#|(!<
zcKX|7SymNxR?_Q^iE06V_>te|p4Z&N@tclUE*7r<GxARz7D{6%jLTSSK(eyF&O%4f
zuD7r@#SMMp;RwGLQES%m!Vu{>6^7neJlCf+F^TKB1YtlJg#<fe;#3j_bsQ^Ex+Ll}
z32g&al$b)HZAqzpx}%)Sm&PnFkI1qT*D0w7uz%WTVYbE0RD)9IB>f${(&9S>M~)xC
zpX!i19-ixwr-A`B*86?ZLgBYs6v+@FJjAu}>$46L0>^<{Z@P)Q@4g$y5m+nP+FGw{
zQC}on$D!S9F*h^E-o?FiIvuhsqbPEEy(ZmW7b#MdDFDrQZ0H3;?tkFRy!LhX&~CS{
zyQuujD(a-stUUT8L9@ltDaC^iJjue|4pTGJ=#4JPU_{H43{*jXr%OCeaHUTa1VpU>
zyDR^Y#wmlH9@cf}G(yTeW475~eM2)+CABiaijsPmu`hd)D}N$6eAiph%yRBT!TD3$
zlyRMjrX*_kyy~7KG_B<FAAO$2_7{20u@-UOB~6F4S~V6I_fV_1Df5hRnlfP}GEVTz
zoZjX(X?=pWDk+UdtDGnV8#;Jyz{P%_tU1M%)0e182aRTA4C)o!@SjcWFTi6^wd~8%
zvaz;J6nQ-S!~=ZxPyU2&fA@QdY7uXL``dZw;YY6P3D0pHv@%@&@?+FRLEslW{@7C-
zxN$GqY+=%vxn={C56ROM0o0pqW~Qc*h{`Y4YE+>Z4F-(2Ht`*csClTOq@Of7-yQS%
zeJ-YqaXg>J)*5Q*4_P0)h5fU0Of{xh8FV?Yr$tnUA~W1@Y#(ZSn~f9qvDbc?TW)C6
z8*ZUZ!o)<I`PoHWuSTAYh_eKPgUwQi$6P#no}e*Frz|OSNs(uSfyIwpv=!{QaA|#&
zczZx391@`zas5X#&xR1-IL}#}0lRuvwjvqihV|7AYK?%8{LzQG<CZ(Q;n+>Q<~6Tj
zZf=gNOUu{)nj8%}B-5IuN1x$v>kzlye1vB%ta0}85_jFQ!0eQZv<{t_S%P*0AthEB
zv{EGT7^^LQ5YcQ!lzBmKFd)rRM#AO6?F_}PEJmq7223``Nc|`mAKSof-Ad=sQD$zO
z#8^vjeaQIqGFwl6fkyEN$M!11I`oGlYLR57Gs9GSsuKLn6Y{*k)q*rhx$x*I%H?&A
zzd9t#bMhi18)bMHqR3^l$caUhOP7|hj>|}T#Cl9sD!-7e{MGArE$pA}$&dnqz#|BJ
z98pPsXsszr#Wjmx0ljWYqq)uF_F=ws-{*PnyS|Ip-0><poi^Ls-9O{%e&cxdqbWdH
z$#y1K*%)$sUx$MS4^X5zRv6;pnD+cEN-0dCkd8&VE>0BT3PGNxmE~Mfk|Z(XQN~S?
z;nFamJ1nVfu5sj82V+C>JV6^pvo5IDJq!|Q6apVDEHY4t%w=mcVRI*DkQj!;0$m6i
zk;~$IlZ94|UYyWfA2KRO+}56^RI2Jal?Az#?BKI`WrvYAU?I;_a$&E%;a?rT=kbB>
z`ye0su^&YU!RpE~-}iwJaQ}l3zT~}Yx7sW&?q%<uy|h{_T<J0#j#yq^qBLbSe$8`)
zkff<%&>!;jGf!~l>?!8w=cw1~-_)Yc0x1w{Luw2S50p?WZ?92{>R1VW6j5up5x$G#
zdpJ^JwIPpVvNUBp8e=j=k(Q{!;7Es7r@=m5(4J^eZ%t6d39jRU=OTQMUQ%-EY?rO&
zK53TX`j)9C%q)6LPE65k)Cq$Sjioo(p|`n1ecq+j3Rrm6aoUqJl=U`JSa3Wx3ZLz8
zg5|X}avbuk!mHkT+iU2Lx4Cff!nH|<MG&|yZ+zpM=uC8~L$y*o@x&7^d64p4k9Mm?
zr_-U+=`b-d!EiVviXsBvuSB^`#UZf9VhjuiB^NJVq*RJlt9{+q?&Yyx@yHXy`7b`q
z;#&@}XMUOs=U2G@u@z<-0k<5OBvCn@W0{$nV*lZT71K={vNWMA3nT&olv%-G&}T3n
z;&?8bBgI3Tl2Te+145U8$6R{{vGXM^e0H1q)G_wH>Q-(zc#t^Fa0HxwY>Cn1ON_Rk
zU^01_V@FG*fOtHj(QsIto1@d5qAUejHYCprq=h_7*t&3))hD0f=-$H=Wl3HZq;ZZX
z3{m8emL*B6$+<HZ$uk9xhqbUP&;eg>IP+}&7nR5HpSBc`5`5pO7=(sv#FA&9DQl?;
zyY(HSddMGq^mlmGt6t5)14nrC8{fpg^#ApB-3%N+Yma!8bM?{F-1xSm+;-Cup1ioh
z#`=J_-hF`3mAHO`+UyK&qgGk!XpPaDWIRSFjTZ{0!U-m(+Kk3yM&l8(5pnWL#&W6=
zj>Y6T!in%*&4I~P+?_w>^5#LZ{_V_kCYYY{5yr5#x=ru%Um}Pf;$Ur={Re8~MM7ET
zv>O3?=jUmJO|q;c$;PO{;%Lio7<2mmlLX@tH$@H7JR>V}vQdho6+z_CEi&S8g3Aw`
zr4$Z@SEDROWPE+5ln}cKVfRPupK{F|DF{RVg-KrNib-lJVi5x3ShK#ijTD04`@`Sn
zZ~u+|jhW62x7~K@D^k?CyCT`fTF4YDmsh#zbvJU8RhaQMolw)BsNn<wswh#~Vub}y
z;)b<~ZJOmMRicz4FLUaXZ5p(hK;U0k#^xSnnWD>rs2Nh0g6Wo`>0e=Uyu{X{b%r@;
z?a-iLHW)D5Dw$af&{C78DYaUND%(s=&5#xaz2Pc@K}tVUEb5fOc20L=lUon(#mIm(
zNr>~^u8l4cu4dJBId^51GEvBOhcwPejOFVI#z-N!_4sk3Fk)q8g>Jt`nipU9-vESR
zz|_<vwOSp|^}txtG-Vve#A!@fRtaAV%hpz(FbepeANd1bf7ctBnVx3fzI}ggMV+=5
zZCqlb(Z*qUZJqto4z(acSa8Bhc4cREgRmA6HEO%v8W$@C(pXep(4Lwihyv<KjaqHY
zxHn`x7G#+uawEq50ri<Vf>ub8=M-s<DHULlu19Gky@BH5*=;tr67sxqU8s35-*(tH
z*I;_KjznQ~f{jB;m2vx>H_}*`W-KJ3=aXlKL`gP<%V4LEkPd+-sbs_*sC;vHR=P$a
zKw9RerueS+z7N-R5d!M95UpN%k9DL&r_p3#c7c6+_Axm*iPjp|-6W33jK@R9&X__d
zjMg9&{k|gb;Qj}{%y)m!2WYq3bYB7D+slc}Kw=G6fF}h;7mSjO?Ln7j)I>-JEe#^i
z8Ey|5#R-lhs5L^OAi%Yj^^F}o-^HPF6C#rAnVP_NBQjl5WMlHOpf^r9aW&@BdQQhT
z9GY{Pne5;P1<iVcdTkQvdZcMgFYb{h3Dz2BW+!ORv~fhpy}$6UP;sA!pL~?2g&?SN
zzSM|j#5gHYR*)4r4u+NW6_RxL!Yy4`Jn8bf*S!uQD;?6|aKOo@PhegIKQE=g4}4tD
z#q&K}-^Ex%7>1P=r<A)M=Pt$=N?l;BpvVoYYiq=#5$#6f1(Uh!5iipwOi&t&7^N&<
zSmJH>+(@HQCw=%N3zHG^^Btsguv#G<2P*|R8sUmc_8<nMDhna)L4B%CG|@zrifG2=
zBNwjFaiLHdlnNR6SZmock<f6T<m&c$x*va(#MH3TLmI(C*(0)7S!i!Df25AZ66bMs
zW8EpHCZ^F!5vMzh#$!t5VpK_YYsksRPxIO%hiJE_Ns^SL%qT`fWSLV71)Hj59nI#|
z4NBKTgmnNVLE65uQGa&~_w_gYcC%3xU`0V*mZ+BiSw24;N6>0EXfzsluD6TBD9H1I
zG|wo?%G%0W%V?CdwYh_^@W&tj2!Hv9e~KH99ASE9nw90%m$R&2H-r!<W06S4J41Tg
zJ#N@DNt(wrbdNw5G&*gZAVd`fRvD}-!S`_8ka`$k@*JfUN@>#35T_O}*@%!*FuAh9
za_pg1MM3XOO_64rz!w~v9x|Oh#pe1cx))uNqQdL!cSp1vISY#cVJ#pl$5>&BqKMFU
zsnwhGMnk%zEe8F9v2oZ}#`L#F6v>D~`)0|E#H1N<o=}c@m56revV!Eo);94V!PJ@*
z22yQyS#dADhOK&&x4!AkEUzqANq#V(P~UhgTB}7&PEJy<*YF&#ax6$v;&{v`8Sf_f
zO3A(79n)w;+;{(%IQ5SAa_60Qar^DJ@y8$g#C2QmXTuSK(o}_PZXK4lM&yO28TlMP
zJc~AP^~x2d<`<dSv$(6tr_sh>j6wPit{>n5MV7I#u~A`)vWz@a+&%%9)+FOZ(d%zB
zF*AWhk{g8#3+4mETxURPG{(rv8`>8bT(nULA*j`wjN%k68l1Y=Wp#bXU;w6oT1#>~
z6BKEWP{N6)Pjle#tC=ox%Cf)|30Y|{Mb1+0vbMBEE*v~BAXa_S-D3L14rD?IrlzL(
zcmLrRx#i|tICbh7e&=_8hhO>C|4N#sFL_Tr*JW~|0#r{<P7#F>Mi~af0UMj^SYe6d
zc-LVDvQ*RW4Y_dPG><+05N~<gJ2`y#5RX6p<aJ-~uMPvG){<yVsx_el$||m`FLV6x
zF^m95O2WveQ45iwkL!6zDXR{qHWXRTXfz@nrRcI+M?Z{cPf7M;$V*LoavH}Gn7qIV
z1A<yaRI8B{B};1qme2NC-Rv@oa{^b=2`v*d9+T~miDsQzqlRz<{oa7h&2>5xE7V`J
zmwiWOXiv>x8y#Fn;s}pihiuv!+vAdMT3|##k!2Vmx%sBsID7Hw7aWKUf>26sz2z2e
zy7@RlREO-VSFf_Z{?Y^`-}7j<+t)fblarGuwTtNt{7Nan^)R+7OsutJsbMrMIDP6A
z7E7zuzV6BX<%Twb!Wu#;38f|UJhWA84R-JxN&iBhrLCO)Fk#<pgJZ`J5!M3YevjqL
zOE^Ms^w=SysZQm?U8-xI+^*xG(wOmhz?pL^JiTejGRwiak{b@Tn48{%=Lkxj;yNy+
zHPk(y;l>iCENIo+)M|Bn2b{2mO$Ib)=h@kIIsf!YZa8*~#=bU*@8gsuV%O5o*a|1x
z+qn9onv&A>n4O<PD~01qu3o*w`R8@>o*zO8T(^4Ol~H7QM!jCy6$&A$wtV%_V2s7s
z>iN;eu({La%$YN^n=RU{7X88Cx~<#GD(cj=(U8OHMxUZkgudqZo*LSWSh`X$y)e(z
z>>TRZg`?UWt+BX%fYlmnAs!7`-*rMTMw4YF$ERI}15K8f#L0lgsWx#^SuizPCC8gX
zvb;wx&d5rT1{zbAdc#NB7Rr<;TWNWj(Fm{JWW6_LZEc6GUP&@G5c%v)Q}TF=#hE%Q
zTg&v<wwTW|vOK}0L-IVwb1h5wTwYpdoGEBcP#Dd~D71KS)(lh&Jv}weFa4+g48Z3<
z_jx|~=}&U@!uhW&M<9ft7S)-Ym|}WnhA511g-enojN>tz+Z%{XTw~Hm>oMph)S`gT
zeD)K(@7>?Q^!x&Q_wKo_MIEq<JdljE!_wxEBMTGEPPC|rDQsc6dgU_ndk@l?n}OYN
zlGYkyZ6!SI?I!mkXLEC7H?ijwS;>LGF!mkd#4s8Um?|f!)tijRCDtvd)f9&!!yc1U
zSdd2GNsw+>F`+Q{et@!qajeL-&y|y_Y;2B+V-KMX6SEPj0F@4KjOK}xXE=Q1E~bkd
zRTh<icA-&@Va<fBUtT7*0>=-?#zW%W(`t2g``U0_mv4XX`}j}4_=}9kV;*_r5&qr3
z`+xZC=Rfz7j46O12$-ChW?_DTiHQlKAj0S>$!~6MpfMy#HA<{T0nqD>a3%c7XFkr&
zH{Zsw8;?~IB7cfQ!=E+;)o0bVVzea&%D!TAr^muXlZoj$imW84`3UJU>J2JjvJeDO
zh!^-s&#i=oOGP=3X|>wqS%ND-7cp~AMizu@51<nzjCwoNrxtO&k~A|Ed5Iy%#YISk
zL?Mia>-#8Ou(H-?V=JTEQzV4|qX}KXo{q!9o*J_gHJbI3?X@v$YXfdNu!p@j9%Wz*
zkqj6cMQH?y(_p(hBpap>)+v$^!dkR|=LE%f@vXaJ;jSZj^Sy82HLraQVHENDH@$&+
zBjmS#`?p@Qs1rb=R%c;uf&F{;Gd(?x?+0k5@izmCJSWdH%0gjGbuiRg(H{&6e9LD(
z_i65V!<#v<fB$t`=dTyWT2fneYLuuBoSKqxp0U;4p&mr|u8Z5KAzg_p1S&5{#xZFe
zBScl)2n1zZGU)cOy1;P^!iBKuQj-zZbrHgl^#?eCPog!Kw-ZiY7!dbkW+TmP+vkRT
z0gakuqR}CU>PW{$8BLldjCw<gIALP4$xP%RruT5$U;l@MD&wlMki>{uljRaNYc0B?
zoUsDqfX;I8JPsc`%F@btilTU)*KUO;yz8}hQQtkJTWfjx>8BYEhcCX)Lg4uxzV8!;
zVTB_L0<5*v>ve)4tP=e0`zj<@Eigt9$BL!pC5pVD({A%kG5vjh?Ed!+;SkCKWfkop
z!VxY{UfSfo^L?BWzWd%I96dCH>j<{jR*+V(xPLER5a35OG?pZet9va3K^P(&Ng*Vr
z)@1CDv39^6H-vPiW~jBBctH(oHH~@`YZVrYkpj;TvBG6ya=LOll7<X{Ula((!6?g_
zr=O(Js55u)7}oO{?d;%}P+$<&zKE*qbDwt@kNBCN`5ErL_a1IMc9ioM&at`Cd(nSj
zcX94%nlc;?Nz;@#j?twiNfNR&+l5w?=x3)cfHKe>jQOYk*FQ&<hF0^NS=0dt$=FE7
zL&bJ4X09EwXWs$xG(k!il@}C+Lii3zmV>o8Zlz-;!?2=B7X^A~@uWjh<Rs&mB+j`h
zf{Vj|ouMXhx0#%pAc`gs<m7pdG2nZGnNEmBBCM(wNlWkweAguzDpuBa*y^T?Qp-42
zm{L&lC390Q2M)|qb2S^IjHOFEG$YBuJqIYt5(kMEhO876QnIdGMoV210n!-CBz_)i
z=Ed)k=eoS(9q;5F?|wJF@AK~Wy_awM%(wBE-v53^<MB%`gyS+ZJ;UPSo@>--S9&B#
z!s_bkwJx33l{t8Y#V8pJM|g6D2OqkR_kPFsaA5y_9(>^XktHkwV<iQGSi$Prkobm#
zS`^{RIdo|W>$6zlFzog)+8`ZC7)5x#hxFab<XUOUIHuWb5qK_{D;XI}s~B^*9x@y`
zBrQdMbCdZ)Z5pkJJcl&RFs8&9g@=a})tU=Jydb0>$1JaGv$az)92qhNj@8tBha+>a
zxVObryGa=26umL)>mzPHvY)-jj?y=pNCxy!lnOFYV=GQcx?{BKB6SI>KnbzSPLw>q
zTPJo~ITOt`AN;@vs5culT20>b-f!o`laKMa&wuVE@2db>^#+Uci|pOIm+7f#Jl8{O
zjpz9kyE;c@p-|c&cJYbh(U>r_-1p#@_~7^b5Q_`*_`Y}D8C1&HYFwcWnb!CYtZb~%
z2o~rx+Ne@tg(OcigzFHs8w62=>wA^`j+E78!$1^9FfoA~yA?w&2uKGb=9FNZYNn=U
zFv_4+fmg3pUXofvZqO!2V@V6eRxjq_`E6ELW6E6MxH+M(S?s`MCt_}<#bmoqX*B)q
zF0!BC7K+Y<%fb!&kgXO-LFk3(62_UOa9eDw?T{IZ<N268K}rY5Iw<oj+xfYdT?#>~
zQRmy={T`xPbgi3m@xpn^@});|QV1ILI_+kQ$%#o8<`)QpfH;mx(}XCh;d_<XFhW$)
zO2Uw(mf<kr<Wnc;cDI?IoBMO=;8-gttRS}<PYMD7fso`y&U$x){qy@!X-1h9497W}
zn_XN-GPgL(#LNW3YKGf8?5wU+Z*;in)wiN`jssXJuu@hMpJhot>ax@8G0ZI|uZ&sh
z!bDAT`;i)Z=O<{?BgS!;M!ilLv~V1sJWtu)T&XO>y@1KZMO;S`Nr|#>=#B4WYi)_M
z|Km5ApPwbHH|Zfs>LKO!HmS9Es-Ta8O!LuC{C*{@{ruZ1EK)k$cH12YSrzzNmpt<L
zL&U|iOvo2PG8`%KeGk`jzyWPFg)VR$=Q*tOE`B!fe9}Bao97CImGw17{T_kbm5I8J
zhqad*%7P;#4F{x_7!A_GRyX6*r~04;-+4=my%PaOgLPnfZUHCoKnPsdB};Pp%j@WL
zM6K4QF*Skp9QxfJSzaKd;O6}kjEp6{^YzT!c^7z+jpg%9?5xrZ1&)=JMM9S6_<o4z
zOGGW8$O~3`TlljFnVy)(sLIl?G?qb{A{-At4Cy8*j;)$f63D*JyQ7TZk;fk6<kL?f
zor*kJ6fdrNV2q_Gaz>*8-QEt9{VAkyD2vM5Ua11371}85^VYHI<OS4g4eoyJU0hzi
z%JS;+u3hM#JziE(r%h$w6&sD8J1noQ(uwBrJeR<P;7UTbPJep`=}4k_ouC%sg#p5?
z#<9YJqYP24j+~pN$P(&}i1BDl(i<~f7o>@xv8O@YAK=wHM4dXJt&ycEMV2E~Wwz|N
z4oHVgdn~W@xO{b!{xG8~Em%R|S!U{jg_$~wb2Y-Ik4g>N8q`i7;~H*$)iD|i(`*V$
z%dIgs3RPJ0u*p{ZEKa~9@?Da|>|!r878jl;RVk@X4V^}Vzwo~AA`B~9w<8_$bc}eZ
zDv)Yhs09J-MvM8Gc@`HIX|-AyZP?k_Ax#s~bj)ZpM2G^5M(Yx!W-u5Nc<|+~e1UI$
z=eMzEaq+rt`h_u+#v#!PZ3Jmnu(ZC+p*?#E!-z!3XdsVcoWLh)H3*^*$NQ?}ZUs>o
zAvzNzv0H(u{eUcRF&N5iL!;5AS#OezN7N>#sMk6~9fK_lRw<M%&;|x+#@a^0#mjxV
zI|^$Jo|_W}hJ}v9<V20xnHKG4L{{W<x4KNl3MbE*nDUuFx)<4Mf)qqv#EudqnV^_#
zvC-|36bdH@k=Ecyi8KzHg0HoL6S%J6rW<eMuDk9=?r!`=S#aU}*{`c==SqiWy+Nzl
zVzM*I{M<ag?~^16X_^v6A%W*3rGr4Cc9VQjXa<9n)zwwbpFhjw+&p0zUe}_|0vVN_
zR-z3?Sp2{xH94!>>(pzFYWx;>beau>?}6u3<2NA@w!%{7X~t+cBu!Fe3A)ti5~PsS
zTMZV3p*Kj08V#C_h;$rR-k`37@LiN|xw27k{>&!b^)YFY;&_Hm!(nPxGu3I*ZZ@dZ
zqN?kKWfTwDUfDodO(z*rU+ge<^9|G{7f|&Ut_56)7Il_$58tb^zH^yUxRixLD1{7a
zeD^!w!$<!3cVECPwty=Hb5qmY`^MK-ojYqWT65ybC&|AK8&WzBje3J-t4X`vrrB%~
zMN!4B4g#VmyhfJMS|fIyI4YNslC7;Bwzf9tv?u@E#&X6~4k~pVBB>D)qNt9t&>QwR
z+f6xfVMJyuZ@Xm=2M;Y!Yee*SHW+XA@I8<H2M$ncw~+OaQb&|&j?o%ZYW%>%34BUj
zkcu&1IkiD%d~Tf49N6PAHC?0Ann9bA*}44~t4Y(0c2tKlMX7@6sYPm03s-`01X^oq
zQh}5h>v8$aGl(){@4@3J>0%J5afXXU32+hQ#=Q94#X_MqPo8{|14j?Dckd#l%6a&q
zM^L4C?uVWqyRH;PQIMu7Ns?4{cm~R{!j~FjD2jr-$X~FIr8aE$2K@4`{u-X|bMx_=
zIq~#UuhjB>!_d}JlLA**3hhxUL!l*4uf%NU4zKG-W+xjsUW6*j%0Z+aAsr8p6nTlZ
zC7pH?r_;o1w$a8g9*?Ot8Z=vyyeLpn#4QVlx%(IX2@?lz#9EDx`z$~75KABbE#`;o
zxIsV|N>rZX=mKL64?KO5QKvz7X`3S#uW-;8Xj9UeS|BSl^{CF&^c3e$o<K^A20ZbN
z8=%@6k{6LLd9f&pg3)Ne_VxzNMw7BAk;0N?DM=iYBr%rf@VU>fcfDTc_|3=JyJwM`
zZ#~Za4?aMD@Z#Oymm4pusCyP;QCdUI6O8kirHvJCJahwjp{~)vnyn^oqlOcBIF5^y
z0wE>Rg3+3?EQr&LJSotsDwM){14lUQ1x9f}v)!iEs*%SrQKy4+9crBxEhA7y(H&+i
zt&F&`w!y~ckSrHOj-VL`+LMxrmd8}9Mx)ih4}6qXY^<y^9K;+RCp0@g`}cQfb|$fn
zHomN!FkBnb&0Ne(oAs?ev9egtp~y3g5KK<alc*8nQT&4U*#f?FIC$UycinUMHRh@|
zniD5ZP<{jQLcLz6({5K-jCRGO(OQ#b8NTlW5CkE_aW$j22J~}Sj0+boG8puko|<_%
zOZp8X0}^9Mjm4LiAoA!X0~)=M#i@Ca5-$v>HJS*&Vlg?MTVZP~m_m_dRdSEVF`@*k
z6k1h<aJ^B-6$WEs+MO0*vt0>=rWxAv5Wb5DBqIf9FArI{&}A@8DRqV)N*1RK6Lp8F
zPMc<<MlFi)e4o;22BSVZtD9IMn90U8_DnEy>=3oi9J<!RwF2u&5H*%_hahURwsejR
ziMC*~3@seq`K|Bf;m5wj_Rh`=uoV`FJfC~-xreFgscSE6uh-+;xpOZHu6;Hf*QL>D
z&}=@J<ijw;7=!QoL}7$<B(CeAv;tA#GBAR8q!{}dmo8sqe*ZzjFhYpobxriS0c|V=
z7GDY+QJqT1nWj7J5xOBDv7U={z|)2-Nf-==<Y|WMRn7zTFv4UdtE+2Z6^<}?o&Xp8
zmdE16G<YsqSx{yPv4vb^EO#?bT#eb-%$N){dnYB+6Afw&OXv$)jTt;Y1nH0$DT8!K
z633)@OuJntXoNJ%jM||$ar`@eh<t06XI56QX%A;Eq;CXQrDPt5ajwuflx2Zq4BA?*
zT)u+iIKbE9A`cxn#J&UjuD$+o9CPBS6KM0|@CqRWzUSe(4xaB<plY`QPe)Ni5ClkB
zIqn!!9V9A|S^-v26qey=#QNGQ&3YZrb;*lYmKiA!Xk!sl(r_GnQBrD!v<{);apF>!
zFJFpjNXd8JI>*t&vv^@hJW6nJ*n8{<LRRAB(vL`zn0RL!n-%!A8d0l(a0J6~%rMP~
zCY$V=9}$-&^@TRw<`jfAB$jp%;RXRh3c`AwA}<k+hbwC|nk}SsF#>X{D2>4vE?Q|!
zk+ZULg{V<OOM!7c(zJkF<06Tz#<9Evkfw@UmHglT_TTcmzxTV$%uEr4K2M!^25VlN
zG-ZrIm5MY;7z_qXO-(Tv4snF6wA!-t8m`Rj;!mG_Z%b_%3{$@N<*#70CCOhJpZSW%
zF7vD9Sd=vs)@76nY(Hf<^q6!c2PQ2`D?_|mgZaq`OsN<Th6ro1yN#DJi}78b*8CjG
z7)qrGn{|A*a%*5$8?cd$IeY(SI987s$780a=9s?uwWPwZ^x<FTfULx<UC&30oHJK;
z_{3_-EjJ!v=k!x-{lV{2<T;sz**yp74SPrld-m>yz{is=Vwhmd94Y=3mEE5*w6SD)
zj<uTMs819{U@S?JptQQyp@ruPy2|F=&}!8=fBqatZoH9tz5dN9>T+vItwl<K<4H;s
z!*oQq-^UjYzT*+Lo0V>nD=Vrppt6E^G$Kn<EEQ1D!$qft@nDSAC5RH~2!a|2SKvh<
z!tpS1jCK;V1mietjWbT3@3MTgi&tdSeTVtBpwYBU)*af7CQ+?{>-m+Zr_zk#m~L+a
zks0RZI@DB3ap(pb`wtT$7zTAzmZO4*9FOy<VXDz#`{FVg5@iG?Ns)oax4h*YJbn66
z&RjT6Q9QSaSEqQ%?YG^=?EKuW#6iVud+O<@UUI{SiY?{)K7J4oMG>`HjVOvpl7w2V
zMi_?pzF(bijlp7fPx1n51o2q0xw%QN*QMR=T-TF)W+Al(y$g&MXtFY+JM7bHv=9h{
zbg{kzuD~;fVw5o)jYyLiDFw}DgU}Bk(5$YkQsfyDjpu@JLHI87``b97N3IlBm&E-c
zIF><Ha&~RZsVgyMs#s{1?4R?Qnrh?u1<hK6Mr{Je^RPmYml@-ve=WgJbQ(mhI#HJ3
z-Sm2H{_Y>8*j(Yv_BJ~0;e{b%E4io)`~8qORpeBH*;<!q$73`cQI-nn2+Z?4@z1iQ
z?s*;3b*ksU8ZKYH#OnI$OQs4b1-|d$xDI~swMjk<L!={jv5C7$9#Gl<60Iz0T5$F1
zW!@wt%|?Tb_4Vtr-p__MlH3{umQYFpDxh?dr>u21*+08THX1S>4B6@BtZ#J*eaXQ?
z`<R)ZLrNHRci36oz?O=mHyxoq)k2l|F8RP>1Sn;Y28wZ?I8W$p_PMZ?b9T)jrQyJ=
z=J=s0rY9!xU5DX#hgv<t@j?(TuIsY8e4+BHbOYw576?5DKdj*-2?Q?feTP{$q7qM^
zXfZLnfO0&NMvY`NtcqoADNRNRq{)zn9{=*!-V|X0$<d=X5=POr!Q1NEGH1@8c}eeO
zDFm+P;JOY%RwKDw<<z{|^PY*St1YFh##XuB1y&oR6^xUFD_5^FH@lZw5LN`5SF~Ga
zDXl{!4UsSy?N?scLb5f;dGK;dR#@J)-(mkkgf0zMq|{pt+@SLBu8Lw9_4>ry+elkd
zYj&u$J7_8B_j+ihsf8hqF1h*e439b!-1Z|s!=7U|fy(HfJ;~tVFT!|(h7sgNCFF{L
zMpVP8*N{ST=HhwsT8O)_NZ6i3N{>dniR(9rOM~n9H0pKoJg+nqcSo%T^9=>2$7#a3
zix)Y6>C!j!em2IE<r%|KpPlX&olXZwx>%zbjfWIvjw&*gE~=dHEKL2`4o<Df?943R
z^}hF!7CHa?U;Qh_Npd|F^ksz`g)H256Xwh*Mx}#7&>b7fOyk**gOR}xBs*IL({Y!{
z$u<qIMi!4r#{*2Bqtl#vdxFLddr?kBp!DlOMWQqUZ7scV&f^!au+$%sU;Ph!?$JlN
z@wU5}nVRI*SG|g<14mg}JWl_@15Ac>qI!cY(VSmjrhd~A-v0jY<I<y#@yUPxi~Qlg
z{g3#E_VHam`PT@xU6fVq*}I4KWQ);84`mIh{wgV=KVukMVdU&CL<eaR(jm?=#!12`
zNh(u@=j{zr2pq@d>eZ`UxqOu?D=UP4bltA$Wq0cI?#bP=7O4$Mp0Uy0U}0tv&wwd+
zli^q~+!@m!r+B`@#PkG>$qtUS^wzgnx^k69bAqYGS%j(ZaK;r#-v=CohBS}KikP@p
zGET>wx*BtO!+<gzn35bk(qyvTz;!i<oK|xN->X2hS)MZJZC1nrt!XrxL>-q}qlGsZ
zFnjM?*?Z3$*uHv!C%*hS0vEzon_M_7IZzj%wIR3Qs)Dgqh}7`WPy7$^GR2s$nzIXo
z6drfp`D$F>tJJNmWn*KVGZ#)*7FsV76|mN=a2URea4ST9p$c5r#rG=)u%&8;2fmL0
zvOL3@>cC-)WiT4Dxv@^*IQY9K`RlRem$qU+c>+QS0RmSz4B`=+{cYwt(`3UTst>(E
z#>VCjjw9H+e}SplSv*gY3<qqjZIBK|EbKkN#QYSc%8|RaJjM|fZcKwp$0T{l&Q_Oh
zYIy2u!p2B4QB&M@$mhVGMJC#9vTTet8MS%~#|>~Ck33J=*;uXasT<IlS;Uiqu-?KM
z4e**xCJx=m8V={b{6!|(btYyP(2mEj7LaBMR%^5dt20tzu%+co4}AJ1?xiIRYTR(_
zrWbtg$&*j8)9+q`*MIeiUP_7Qxp=MvyKWJ=%87#Lc|YSI90w^Y55g?Zt|fU#Kvoo7
zx^kH!P4Q**bNV`yC6qPz!l5C+k-EZ41`RBN?Lm*_jhH7c3@NPTU3VPd#sf1n+6|I%
zkL}fUTqBsCnxfTi<2E7)Jgk5+%Q4zwl&Xr_z(t!nPNT)S^GlpwQ_M95N2VQS=6%|&
z8ueP6yiA#$KY+EGI3Ck%M%D9U4UJlpc59|m??+)hk1|b|j6941V_mMEJ4-y#;)dJr
zK}!cWj2P|=i463mxrUQ{Nm&vL%4(iIbBaIwpZ}9%HyvYkZidGmeVCP%jhA)}+)W~-
zDoK)rIF4zxTKKL{zRL`+PW#o^uGl5UJR8OeveNLePkf5Nuk(gCyn*{3e2{MMm6J9x
z09I3X1Xe(yC21ikwB+1M!cO6GG<2AqtWy}Dns70Nst#;URnV11<(%O9l9^o+l~b>y
zj3rN!iskS76h%Q)6a;?6&2vY&`3<ipoSH>g&D_x&>EH4?PW|z3lAroAll2C|TCyZX
zgf5gRr`BUWf2PO$tB>*6r#{Q~9KDl0ci+RsGtV%yxSu4;2*ZH+#RU|G!fLEvU9Y>l
zuHW!}R4j=<)x9fA#W)@_81&fOT*qjG?^R;U<MEgzu57a$N22v}zgGwdqkz}G_BCv6
zZE@pG$5>cg;N1BO|EtURzZ}AKIryz_VkuAI^c6G(-JvFk1QV`is$L+hgF`?(?$H}{
zaGHXm$cVfEEnMmgi!>MJD18Tt5;ur&iBKYAYn1ZoC$4he<t?^GF*X#`b~d=<z?gme
z57XaTVR`8)i;Med9l3?2$3D+w5TdlARF*571E!9>hWW*VNG-{o8aGTqVHIoVPILRc
zuR~yI)*>9&VK|CWWr?Elw!WT`7mBM_SGjs+g`#*E*Y&FRxF}wtxl%!q05&!^8OJF<
z`4d0QKl#7@DWmb|&wM>EtEekX)wZq)fiEHOJ%m*BlbBXIpiIXsU0LPQR$9@VCu`h#
z`w=>mEwoXL);8H*-@uWQqel<ZoSA~qt4y|c>CR@C9qYRec`QhhnDgh?IJG0#8B5x}
z=FmaIzCCp&+Y2=7b;j`k&u?K&r9;zb1QbO^5)Y`?+Js?~C<ySR#R==^(SUmB;QArf
zYp{IbJQIzO14nN~cs|r>B;y!aYLqcZQE6~lV=8T%mk5JkanByKR=BQ<6qeJcpJ8QX
z{p;Gq%3VzyV=D9DG)swU(KYcy&-1SBJF2XtL7vq?t2&>SaguT7+<6=kaKn+KoH>7v
zqWrTv4OCQYOGCOu(jufG3<Hc2<YmEDzsu6vh^H@)$*kpVH}BzwLvyqyT4ec%wM*-m
zq@dYuvG@2Qe!YgM`55h?cJVbjQ%FY=gg%ZV$cln)KIFc$L#*_<VOnu;!DVtLpw;N$
z`2oU_)atWjSw>pqbn10-m7#UP<is5HS_jW{afHEmKIM4KOe4S;(0D9gJkPihaqRYc
zu#SUQt26A4@KvP|W_R&6+1IfHswCVB9=QL0gd>?>n8WiN9=Pv5;yB+;7B9Vllx0bl
zWh6;Lv)ROVs|06^xkijBih?Z9U)UWGlDyQMK6`;*`sH8Y$dRK+&*kLPPyd<M`?(JX
z+Csx|sYwGyQEE%g4RDO$k+VBIu$0i04(~la&5Z{p@f&sWyrf79_Uu1^R|~GO-jN>T
zL7%kSg;L>1)d)_yl5vtS7{wG)a_G=B!(@e-Jq?mpo3*G-;rd`T%~~DDtHy9swOVyw
zlp*jNL}3#@^sqvraLBP}ZRxl^D$7BYY;3Mjs<aZ+l@3NovT;UWCArnR$yIzE;hJhS
zzwqTRbL#YIrl%)qw=327@#v*TWyVzGraaFXj)vC?JFPTE8L}*^WLAvX73+TCSWX#(
z&MX%$T}9&HxRr~`D;d&t@jOXG7)om>t;Z;}?2J;zBZmp=v!|7kWs1`mR(S2JB0@LB
zmYPwo52~z4Y)X^IBf?sZ$$k4!+Mu+;t4EbipWRJD8Ypwb@&!(O_79mlb}OS%%;6))
z2s@MPd;52CYG)n4f0Ymc2E2Mil576Q6PF2Zxs7smj;Y2fAO8>kmPbGLS)$ol4&Hb(
zk39A;sw`RDvq+=UVq<HER;$C#>IR;CA^GWgL?sem;&~25nPE^Myld~a<2pEwOHpRm
z7`4D|@vq79gpIWU!jZVU-Og`DxKq>I^4;IfL;vMh8E<vT3lHrprfOp*npJ;OVhG&;
zQEE0<mYHs~u+pWouz;uqC?ODKrBEQWBrS$KaACloobGe$yT6A!4)5^CANvD_WuIE%
zapm$l-JKo2`1#NAs=Hsu<m4=w6^v7Z&;nziKa5!$mdviNbNb{7y8RsvAKS-mw##Y)
zM`t^1cYDk&rl>L_%S!S>BTe;wU+@05Qk8i0bMf`@c+XDKYV{hElM@t$syg)E%PPlS
zGG5kiG1d}Fhq@5BLZhsu<wtm~&(1LA{CdnI7l#A}-gEb1ZaX@S=Lcl*n2q%{0>@+j
z!Gm~VP_g6!k5X&0BthpDm0U_eSc|YiprnJXHOQMO-K(4IpLRJoSEt#DXfzu{QG={3
zsWm5%QlNEFF)dw*7X%n1X}2bc!bYV7EDfdOP$nr8fm?M83|u~cip|jHRd4-vl#s|E
zq~9IlmDQNDz~Wg!`t>R^3u&42kN@dE;Y(lnGB@9HoC62<@yU;WjDBzQbq;wIAYE%s
z9LJ1CBRZW<1ujRDWf^fCU#t3+QoDt(eE}9D%@x1;Yrjq!7tAfp(`+``>HgVI^43<G
zJ2mMNi2|()N(D{7hAu4kUD)E0<&>5qdGGOAZn|L(KdO^w1>?bp>DgJLW&^Cnbv<NI
zBh6xxUJn%~cz%Fi3-N;*rBMvyn6WZU?wMkCeS{wvW*3@_!#2XN1Y0E#)tXr0P-_I#
zFx89Uct{b^sJHP0AA>{*O)f>1KwTG8lnAX^UAfG-)uP0rqyvu2xF6$di9u5ugCkx_
zxaPU&4hQ_bzyJ5xw|6gd^E0&DP40W(0luD1tT9z$%k%tsRQP0<0a}Ub=DVF6t#vgn
zdR~&(yST5b%j*n>pRG=X>iYW9R*oqyO{tMqU{r{pKuU+@?Tm+)GD-z^@0A?d6JWI<
zPh)EJ22+dkm3XHVc%FwUO9opTWc?nFFw~n9MC~?s4*mXsEJ+DH566+rO<3j+&BER9
z;C0{e0Vd{VDMveOJ@pv96Av(HM%0W$k!7%};~0h^wR#=VtW(A*CoZ0+x&JVxJ;_3Q
z9>?*hw;Q-#okR&-=~Sd0DpCu=qBZ!k!jFEVO~2HN)s1!5w>Gb>=l|E<n+I8T-S>T;
zbN9Eu-n(admYxl>00UqMfB*r4Bt$79VNtHwk`>3al}K@^99ynfRgx_$E=7tRmz*jl
zWy)5`i71m&n24f8kOn{i1V}6du@EyD%+|B?zD#$&e*3%oz31eQ+W-L&1TdT#5dz<*
z>i*GHeP5rx=iPgM=ls6E@9*u-$qd6|y4^PIc9BM-PCj2Gwqt_8!w+2IC?t(T;w0XZ
zYKS*{fMJ+8c8-tVcOOS@x{<Fv{16X4^2l{_y(o^4F|lJe-s#i0iGf7X@D!FVL0U}O
zis6z@y(KAE5;i+cCPybRl)&$Hh(nLq^$@XSV0@g))J_r<GL12FHkx5>d3GtfVZtjj
z%e=g}Mk|QWFD)^7_s6;Aj(h2BJj0P&ZXsW;FnH5lEPnMY2M0<R#S*C}aoar>dLe)J
zzI(a#w%d5jhZBGKHC}%7Tj)C`_}fQ6z@Y>CNl6$U9idz<uxXoge2-Xxo4$2xu&W`<
zs$`X`ahl)-9&sXR^}1{}n`|~)*@EOv>3Ur=&~=m5)m5ga&vS8hmf^t>G;Id?Hu9b8
zMJ`vOI<k!&AGwSA*B-(0L01-)N`jrUh-HkBI$<PGL5KzD`6Bsh4$XEDrcM%uAas&A
zW?`ew!zY`}i|vf=zlCpq=Sd!a=~>!-Kv9$#$yZokY4gCJeS!If1wQ=IpTO<5S(?3w
z&6cJ(PFZZU8L=z|I;;HY=f1#tbDdGm!rg4}$=eTM=PhyrgEX2=+MOold;uFpQE1=u
zn7#gWMn4c;*XVTGblo0Pdv-H@?gAUNb)svelfAN_4nbbm$Y}~Cv-BiVA`;0f3vHfR
z^cXY@K6a?e-XR;IS%5~8C?+R&VdQO6CDC;q!?1C?9-YQIaj%PQIOMBka)lC!QZ$<_
zJU<haPEC~QHM?{Rd6tI8DUI!*WE$kd1{0=3CGQ{%8=ZtG3W%Z@Aq*TNk6kFD1V~M%
z<Mv2Z4w1Lf3su}^8<Ggu_ef-trAQ4E%`gbNE{4iPoGA(&^^V#LsT8NqoMvur4#%<3
zbV0A@631`95{%=7An@t+y7YQoy4@a{rlTqFeUBuGQA&~|uNu*Q?**R)cM!%3p&5*g
z4KPrxUNa$RBy}teidqU%M<y1Uh!H|Cv+DE0qKl?9?%rXrXS;)pC9xJ@I}Rh;w{4kF
zgJGDU6zxWxZfygSfLy*rzEH$0=JBJDR=oiTY|}vKiktQh^L%JC`Ct7UQ#T*OHg&oy
zm)Jc0GGjrFf@z|qk|^>s7k()zI616h5n&mGp2y6}0)>fPXhS3Hs!SjgMXq3@84g~O
z3BoPYLDzMnFhq!KXHpY_RK2s_Mk+<U)uh$!aArD71*YlS&v(}PE27)$&~7(rHk(;h
zVH9swI=dusgiIn*xm6Os^^zAtWFA)vO2sn!_Uz>+?)w-kmsj~upZlY0)(19GlAO@6
zbwQj$n&`w*#|_}Z<%nitaD%NF9mpY!Vy1sh5+dKvtZXzLp$q~yLaCH|$)S>)03=4C
zNUFg1JS@YcGF~Q5lFaJJaoAfLX6WPharmAOBlWCR)V3o>Y25l@&VB6*3@)6YQY?W|
zn5IeS^^h_|h89nsoMlyM6kN&i=f2L4pZNtU`}T2pVS(|<-7KsuV%rWwL&I1(n?@XB
z8d(}lZfRBD%>!S-6{Wjh4c5EeHnq(SOw*!N%oaOg7~*;^uIpy$bzOVyJM`*(Zrc`<
zJGQfL?_So|*SY=n+j;Eq$BE)=8_d6o$l2`q$&Yfj-9fv!Oe|tNH>IKrN?J-e7g4k%
zZYzf&5^A-}j17<BxjvSup(i$l;Uf8Jnf&N5u_3Y@S}Tu^gA@sF81Vf04IVwa!df@R
zRWYV*V6Ar;kF!c!x4FUM;sU#;4luZD4<|gu{;EVq9w^DuW{)72j0}xa9p1*CW4H6&
z2Oj3wjse`9&gJuGIC9sWXedlW$F?kjIK}m{-IVw`yxkQLgdy{DOTZFQ6r*YSYdf|-
zNbf9!pin3xrKH{N&}w!0t$+EiSl`^_!o~B~e4KBG*wTSE{U+w#ZS4HeE!3Ymfsh&<
z-yl~C(UgXjPl-GYF$D7s#d5MvxmBX)U1qRiGdVg!r&;5fnFfz6MBMzzU*M+>9OmEu
z;qP&7@e-lZDT-kRoheH8D7MVwXjT5jZ~Zd+4<Ba!>}gg{yu?Qbizp$p9eYjkb9Ya&
z<aao`IE!hST<&-jrJ(9)$b!T0!~}jA6Rd4e&KYzyK|up2NeN$*S^U+bX&T!nhOtb8
z*-JA_P3<C=w|V*GalC7!oxHN%ufm|9Nl=0`u|b7sn!#!_<m7rnEHw_4H7192>|CBO
z@R4>(X`qq`$u%8QH$W*o&&BIBFjGmPS|L{+#Bdy<G@;dMVrUuz<qB~c6GtIC_fHbt
z^$G6#2mge^;2<iE=&dcX@XQmey!;F!fr}Q$NL?UPiEipRj*V^^XqLsy+&q_>HO6ng
zlgplDw^gJ%SVf2g+bQG4A(n1X$Q5yHi^keI4vK_qr%IvT(pCD+Br;o(@q;bt-hfis
z%fWSfbbD<!H)|9MrA(AewtE-FAyMEHg#n3tD}l5kpDQ4Q&fmEIezr}Fvvb!D{_j8j
z<99XYw;!J*uWM);B#9u5G|()XUCH!DL^m<GDX%dyki&FJ2+br4Gd-Z4bI^1hp_!y{
zmRi;=!N}+^LS-*^&9n%DkkAhp92!Chk+sN*Bg~z7?%DqV?)@MCDMo1krBd`Hq`kh%
z+!K$o{Op5Fr~t#sp&JH?uH)Gzqycf_@Z2ji)NP%7VsP^Dr<wY@zrsY(#9Qq$Jhp@7
zwG|xOp<1nyE97VdK8}Dyd~cuR-CVw`h~dgs(n*r_!t8XK)N1P(hDjlx2bHB>{lKHw
z>t$9ZnugkPDSy?p)X;UR)hc&<@OA>v;{(Tz@n63FF!KwGS2T_vh$biknvz+uP-@_Y
z&<%U|u1N)$EQbgoIC*A~+i$nY84f}!+)g)3a%}asOOu2o=waJ=D%&QAx4afL+r+VK
zG&0|^S7l4ZIArdrM;JSN6H?PiQpL#dIQg+1?EK`<aqdeEOm726=)`f1lP?e}pP8i=
zU!84o>=PfO)>+4Pdwl-4e~bNh+<`A)Z2L~mUb;jQM+^-QQ5meVq-hKe55ihKdv2id
zwpxiF5~VWx15MAo&OE<I&dI#CvgYm*!!WjrB*O1Y%V~P%USHc>XMJshZ9672Kf~)3
z)F}nl#CDGS?5B9~kG?>!(Iiey!bmZc^B5|p2wg`P*-mcJDbn4nllVT>VgV~(Bwwu}
z@)n6wh*)KMN2!x0F4OBC4;^2_8=m4*zxogP@)!Puh4arroRSwN3-hbo5rjPc#N!<O
zz-?^XwhK3i=!KxAIueD9V`kR7pfu*soFNv9^~MGV_D`_i@LBUE2e*x~UaK)NzKt~U
zk!ecT4Y=wD<2#V164!lwiu|4DNr=oBtz0fMG&DrFn`Lfn+q!-O{Q*d+Sl_JSw$?}!
z$=I#8;?){-&d;;5Vc_LcDn^K~4g9Vm7eqXKp^K+UhV>dZY#+k0bdH}o%~Nx2o^ELD
zI&u&9-1~7p^V#3!<ibUy(kW>}jO6w*P@H0Hc!WaU=9QOEGj-$)jb@$2<1Zm<s~9^=
zG{TTD^vDkk5XhLCr;#t0NQEMhidR|@CpLQQa($9kizG^MIt?a_guzM~<##C~XeQZa
zO?-{h?(c?bwaU#m-N-91zeFliy4?znV{_)r>1$Tpctt^7UPvrWBT6Bb20^0ZMT)uA
zh<0o-W(l@eO%n11i6WnKNW%!<^{^ZpL$?Ui%&SK?pfXTGPfBQ}4o;3JiV4GzO1X?>
z+r)8997RaeV#F%Y$(iH^N{F0;P$re#2iR90=kkt&%zouRGG!<n)51v|+-?soXF=k#
z(CqQxiB(3A?&r*j)2y$yxbqkO0cIt~iEn?K?Ys8!^68f`Op|K0LaAD1Wp)nT)IelA
z5Z5N#{njX{2%`wkb?J0lY}PkOk{HV{h~qdD*Lq%d-L^C!Qt~}lasj4guz&wv@_C2Z
zxmot@+lOUY1XtMyRzp$LG7FUmgE-dkla!5`OW@g*b&XwBiIf^=FRpO&jeD^Ti#QB%
zl|t8bRFZBPE`p4GBpUhZ0Lm~3qL4(T6sr|X!yrv1N(eMGVin-eUu5yzadzEu2fpu;
zvkT-ZV{E(o6I_r!m%sZBwizNb#|$KPF;9}lyg1k7;RT<YK6VESm(JraUgj&m{}~Q{
z=tC^bFEcng!SviLGEEs6s8SiIGOr0r<pPEu5eET=&=Bc6r{jJYvWDIP3W*=Mbh=H1
z%7UHDt1ie2Dh!i!%jI6ZF4)iwjU<gImy4V}d79zzF$M+(u1Qd*n<nE&k07Vd5~YF&
z2&JIuD)3Ttm1ojQ87%448VQD(&~`gGgZXUk1D|fY3298~hsYqNG&(|g+XShuA>$as
z(K9bY^vpg+X+&Ydx#d+Bnr*zfWu9MIVf^4R28YJkyLXDiHy))hFwF4L53~62e`a4f
zhv^j244v3lENrY1j|_76&wP?<uE6Ji=XY5@d4}niULn7Kiof@3zrn5@lf-FAp<H5Q
zXprJSktof)E29*QxABv_8YB=)NK%zq)mS=CK8Mf^n(a1W64U9q1Yx}85BI%uGe5-4
zCg{P39>lV8)LZSWt=x4&ASe{8h<$k`@421E*Pp-)1e!D`6f_DYjTDJ(gm|98ur7#N
zEs6sJ*wq2-l7nVBXogN2Mu3h?QySe4PoAproehWF<Us}p^8D-H`(HS_GDn)~RP_p4
zq0I8^GGG4UA93;G40qjqFTU4fWpNf2c%UjMnczkt>rIavD+ZTed5Nz)@g$mv8Bd|p
zYV*n44`bK@YkY$B%}s*9#WV~YO-EM}4@I1rXkNGLcqQnXMzdMR4LtVj+0E(G=cqR}
z>Aj6<{q>^V?eNe4*WW>HNdYt=GOy4uLMqTNNOTP#0VQC?BS^CEdCoL2J3gGto@agv
zQfbVb`X<l(>N9kEovmkTLRZ%C{3eUR%V-S&D(jK@;=la{hGCFI5vEXl=6QHEQU2<J
zCz6U-z1E3~BuSZm>Kx{a^XNu)ZzN$rnk4v1rkPdh?JaDsH|=(tM;?EgC=5Zs<=UC-
ze!qrey0V~7&sOd+8j;lL`VyIj_?}KBwHPnNlnOfMrZ02&CPB^_#WpQeoZyC;SXpfK
zlMLK0GLET^j!_&NB@Qz~L2FAuD*%0Kvcgnhs{sGpan64GF$!bbXtrAH-gAITVSw_~
zeunP*TP%J3PuM|>mZWGx!z$zm63OGISD0}sjO;teSaXBU{L_5)pZp`n_V2?P9cTXs
zKFA9vjuZGk#bSYx(P3r^IV^PSMB+DEn40zjTZ3H-(j*0;uq=Z(4$ySabsgXLNz(MS
z#=$`xZJCmOZzKxPG|+XpG;<NR=Mlu&ZW&h@)D;GbIDU_w)JYM95wzWub}z+|7TZlp
zr3`BuK83PIFX~av50E$WgkeCdvx%-!L@EioF1l$lGPMgWpCb+<G%ag#uIUCS9Z7<r
z>GT4Zm1Ye$g2z7hf1*#_2!_M<?K`>i?hj$*a_suRM|tVw2`c^?IiX`&IZ}}#<A{^X
zO~mdAj{fw=agzJ^i+}gO^WvX<nWz5ZD^!mh;^%(lS2=h7EJ5Vs6!MIXjZzvcArgTK
zV}dlR^1fb31yPdS%BE$!Rv{jEAzt8Pm?l9K5yeRs=)UR2W11$WY4YS#PvW})?l-@Q
z=UyZ016^n5y&t4518QexAyq^wX0w@4QG%kTC^-=&Te05OsM@f4d69Cd0w~;Wn^G=M
zr_sd97Z}_%On!Wf$P~yRLgY<!t(ciA<I(BHeEakw-<Y0fvF;HHouZMjEwZ@lr~eKU
z+jru1Ho1Ixg`uHwCT=*)OJ7p#DT7XuS-(iZQmsX0Vv5Phy;!D2b?PAN(@P9m7D=be
zspp^LjyrD0&@|9A%EcnOWl>+PX9YWmHw4bt8ik<K>5(L>%+4(k$0?5O;JR7rHBHmE
zT&*jW$z4jNGOcz8-``+xXoUan7yo-c`v;#Tz6J$RZ${HKbVJbSb;#YYlWm)aX+D1z
z6M+|4q(*>@B!#j?ual5QDL4gA%r}@4I(jF-4{DSvHlx)+0=L1Lr8W<otrLtN=5PP<
zFY~Q$KE$6s_-EAIE?`sEM=6`z$r*#Fz@lId@$3KM?{V{Ox3fHRiG>p{v$JR+gajn1
zA92ghA|qotUTAF+x7y^&MJ{c2xLpXUjvy*H42_Kwr70Irtzt(3an1tQr=UY8RX>9L
zn(s%~G$tp9iNlz=OSA0QIf>(#%+JrfZJ*ieg;JR};txdfy84ngerLOJ@)q`vk~H)Q
z-sHU}Nw@TZA+OiEc)<_11MMafy6z8k?q3CIn&SKE>wdoWJMC8#)Y-Pej|HI;L<*W7
zB!NUqhcQbqUes_Sh$4X(v~be_s#Xq5fZy&Ag&r!3h<hH6oo8g<6xPT<R;8&aOv^$y
z43rX?|4S-qbUaSXE;6&(B193+FR|^|Jsi04Ha3?pbM)9P6iP!(9Jq;#j}0;$tWmZL
zSj9Z4Cm{%!Z*<sx=N<gS{hva)J~J<#p|`xj%KQq}?wuUD=VRP-=n#?w%PBBEIzp*h
zAlU2@L<ylHD?PtXC?HXSL}s=nN`O=;iIN0KOxJa(HClAKE>fxQ6WD4(Llaq*)Z<S)
zfnnO*aKk|gPLbYKzUX(~a~riMpC;}pl#sYdO1&vbL!F|jsOC}}Q_%1<O68EH)g=o1
zD-cI`y$+6H(rI*2Y0BWhAOpL0qRRP9zh(<GGmj=5Bq1`5X|zM0o}S~G=_OX%327R!
zedh-IMTx)vEB}nm)g>-XpJo4{qu9kFxji?s{GAyNjSPa&5Q$Elz*?)rwxh?`x9<pA
zA~A<230KaM1R=G#CC;BY&D|fmGi&y384Ok`*m;NM(k4L|ks#jC+T~iKV9Pi+q0wm3
zY;~AEe+k=iXti2wZf@fH{u>Hn-iSt_P^8=KGB>w?v$D$Hxc^gZY}RRB3#*HPD*Nug
zpKkL@c+2Y~gt$GEN<N`%xeOJv)<-%frqFOqgDCKE>rJY~0-^6SI<XCMHhLjPYHFE<
ziDjc>BGMR7#hjXJ@!+W%)tf)a-oi1WSH45f*&qolR@b^LEv^zqDJzQ?IhZOcm0@Bu
zI?5(V1iH-L9<8p&ayvp+10vT$RCIP9z5#RkyBwJq;NoV7sexja1dCIIp)fS4H`>He
zirBiRuICFrO;hUi7WI1Tip~>4WZs3EhU3`8aYC(D<H+G7Sf+Jdf;y?l^nYthD4}!U
zBlmK#>l0m=W2t437A=PJ9<~{g3P>W2l_=!Nd5`a|FW@M})JT;hJF3jpF7V=Po&T~L
zB1R7L&wlBDVs-5zfBD!~XhuGU%2CnB8Oa}DsJxTWp<(iQi;L&Za@Tz!m(HK#^6ArT
zoP3FU_Et%SBvuioV>7j5fVStOE0gh&D!!>>Mh?@>4$E%BsHSCY6{W)Jwb?0Mwv7zX
zN)oDRLaZQG8tIl@*N=)FJ9luy-d#*jpP}2?z;#Ph2P#ZYpCfpe-3|Kmfh!E^%7<>C
z^~yywDM^TEv|{YMpdb`^J7h2~iGmbUvk1H{Ya1(6_LOP1TNt_^Z`g!NP~AOBWn>tw
zS|XGQLTYG^iB188Bq5;I4EV<JS)Q6*WwRHNSURSP8C5NI8V&~z9Aa&8hK2b#4j!pu
zIS$UQ8(2HJK*i1xM;<6ey&F>R`b-T^FfcTMHYmw&o5EkNv#VMrN@Gqv`!qM-c5CK3
zEfnQK0o^cgy?`)|Q3#Ujyq;4)x9hUBxXRqzB4HSzYuWS!LGZQ$TL6ScK3|~Q>yacK
zMn=Z?^<V!De(g7Y?W)E#dfTm}Zh-gH3rLj}lXN<QvVffBF<1!DG0;^3Q%g`uOcZ+f
zVS=M;kfw}|jH4BESj9YIt2d`%7?@fPO=&11TA{}a=h{5J&?L4;(8_rV#W~FUARG07
zk!FvJ=Py#byhyI-Fg`hjWtikAc5tB^aiE%@h|!R&HeBkxn88?5Dpb%+ogGJxP}x}I
zh8<%p_dFc8O=ZsxbeZ)Fv`w9^>*0s-mK(bEjztnzOB6*cEv_Jh_F5|a-C`I9gM$OC
zt*)Y!qSb1$d-rZ`z4;cN_|_9wJvLo4898_Z`|n%l<u8Agv>l;K8we?u0t!V9r3nl@
zA__BKK;?#{y&fW0$b9KaISk7|H+7^RVQD7PfGGA@ue&__N|P7YQ}*3*2ite<;s1T$
zi=0@vK$-}KOp}3Pj<csukt9A3{N+~|nV8JH2<saJo`){0C_pA5tF0cMFDbS+`SU;i
zJQp|Dv4mn*uE6PqW%f_)VK{HG6a>u7&N8uUk}wJ|gn=d^EClgZFW(<!W!zgK5ClO$
zy|KwnM{i>0(i}=j+MU+hb}hLYW)|l7&42T+iK75b=;*pm>UjiilQ;!^J|$<RS%r8a
z&^Ffy71Vu&-WH4%bL7)DT;9a<JT?MFAPsVxv;6k|@h@3eo}(9L3(k~?9yP+iWi39#
zg^os=Cb)i+lmGLR<nuYgUJoq_c<5Z_&@WZGm9~N)PH9Vsq=KzV_rQ<1`1mPGr`9nG
zWdNdX2ftmX5h(&c0u-_24Q-r$RMhJYUO3*P-EN_4f@b$TQ5<Kj=zgRl>+{!*D+=oN
z{^U<F-3jQ<E)gdIDzr$<ki6+LRETjbh+`cc1IM&6g-&y`M&fx4mWoKBGrDb(0R;{E
z^3*m+b9QEzAr+S8ph**yPb-Oe@m!rp&o|h2-~D{(K+5rFzr}hwizbS!)k7Nf4h!=O
z%$`3Brq1rYhcew!p+eJ*Q5BWt-(|u=tw%2wbZa$&UYD4Z-G>gar!&uFwZMgq4u`jm
z(GCNM6O@yHOlfyqf+&4WP<y>^ETt5l7vOnUzA6(!kaHZ;G^J85)9Lo;bUKWWjbHV+
z%*qglfBL8CuB@_tX@OKp!oWh$2b2p5g@QtAplLBOQRJNxQOCn;x5*6+Aq5!a0(vfo
zunf{D#z3H322w(tc${B%dFXhZg~Z@f|Ls4b)vfbKpZmX9-0Wb`VYHCr!ufgb*93F3
zv&=5caP-(6Bw9q*^@&4?fk`A&lGtOp-o;PATU+JPuRg$fuSwhOFg9GESzn`48e-4r
zAm`WDs1_;=Y#SpAe9~BDt!N-CeZSUT2XDcfwQ}zkDJ5%b>l{9Oh&WDYwK_DLO_FP5
zZIW~HC|#$te~RH_2kAU}l0uxION(3{>|8<+CFFAu_y%d3;CWrFLXob}7^@TjMJgmw
z5<?6LE@@rz?U@xGyU?Mq_fBrScMq2?z099J^hIjEi%<qdeGo$oA(X}C`E{E0RSw^H
z6Sv-aJH1Ya)!7*guZONxQKYCiWv$lW!}~_CM-&TggUgGHSejt1-r-go_7CUrZNbq)
zH{vHTffwL)+El6)8tonhP2kJ-Mk?;j*tS8TU~~Tb1&$rNm2$br+FFg7nM>Eq2I0+k
zp2zu_w=unqw~j5h%r_tv?T!oIXX%*`^cplTzHuixCBaX-_~}-<?{yiv<+V3v8s^^p
zzraE-gx=D-_{|@E*wpQIc|*1{_^a2^^?84AMM2%b)IJW}_YscQ{{p$$LSQ0gLZJ{+
zECYstVa9}sMv_3zu<_P67#JKRm5>`LqjEN;WuXDePcRLW#E{6O%k1S251wpuuB&tS
z1NV?~@_gx=U*hEKG-;w!)gwkqc}^Zb$<)3*Jo5FgGr4;gKl@*&wA&3DwGGNC2&o}u
zOq|3lZ1!*zg!41}!Ds$GOO16(md)-|r#`>R%{LsNYzulq=iK>e_TPAzp5Gy7Id}r{
z2m&%!_4kUP_RR<cy<U$_x5MEZ53?}8L>woy+pVkq>|DNxRnpn@k-MqYYM6B&3xy#p
z$`!D5jaY@)hQf0V21QEL>5$Jm=*2RHij85}2*U(T5d|Sa)6tTcdZ*2|POb6$WfN=X
zA@+<9@aJFt6P`Nx0-ZR*z@};plB8u~PveW9`vYp5HTEAoLLpb6w!XsR#dGBJt*THN
z6U8x0wGKB86{*CV9Dm`voLiiyWLvD%BPK=4-@I{@NKd)-z|FMWE`=bX(`z$St`cvw
zC<_f=Wfnf~$(<V^Fb$2xrA4Op>|%0qJN0^lW~)J?QNI>{0ZrFvcQ)AYnvC6k6dA^}
zzk3QPHCAe16cic?t03_FkTkF<j2C%+x`SN}$*s28==wNG!m-0UsTT4qE-muFOp}wX
z93T9|&-3ZO|7rfW-~CM<ef}HxQX<Gv(MK3A?4{(4Fj5)CN)?9>4D*Zs`0tR)g!$8_
zxb&UpxOu#Srb!Z+37~J7s8AUwvD&C}dS;rE0amF>-FHzU>rZPH@@TqF(KOgx+rV63
zqf*VWxBwMPqZFrfx3*{BE5V&%80_6MN#3!Ton7Se$`a*Dm1?!bx1W7F>z~=DPoF+p
zT$xXuCdlvJ&ggANXg&2^a(;rL3@p<jUrY#s1jhsyiSJ92Bt$bc8nI-2w1lQ>nOiyu
zNtmE1K^%4oS7RQ%u*fruF{6j?<(6BwGkxJW|M{z*r{;TrPC=BhwIO7ZgD}rGANe}N
z!$W-X{+|UY=rrmqojuLH#T-f~G$fuE61grvd+T<>3Y_*kv^Lgp(OKJU@WG<a)S!cF
z2@dT$K!halLOS&Zg98Js&99KxG&-q#i?-hH9g0N<%hH)Xe}NkhAEHpmad~Znv*%7<
z^=E4uG`%M2aEV<Xy`9C!ULfatB#A}Gj~T2+5NQ-DI&L>6jdExOlbL#q?k+PqHcT&U
z5=Rje;}r%A6}s(J&aQO$+Sxkmx$XQn|HD5fY`6JOfBO5(tS;ayFjbzaH9<)qBX16(
z#U?qu%%?y7xA^c+-A5e8bk;XnJ$Ztk94>>9fI>zodxuPR?iypR+vM!rS%@Mq2bg!e
z94t9Fri4OfSX-!+>9o4UYa5gX@|*|Cx*(6l{VUp~y;4d=hXea}aPG_{R+m;77^slT
z*{rNCv9K_I)t|4FWM%y_GV0=WyOc|VRBk*>V{MK4a*d{E(A`uF<wNos_`1e&D`HbL
zcy!6bT}c_!4Q|>w#E!{P);lepKR3;{Rzl9YHsgEtaQg@E;txOfA9(or$MI!~u58Ns
z099v_oHatFIz+B$(e*m~`IjG{G_FxBI@B**Wbw=?Zapx}O4lP+F;N&2It7+{5$C&t
zpjbg95xpeFOH;nl?lHO8V0*1WuCR_SG`bsWjHr-$&S9WjhNwqARroTiypB|Q%>uit
z>BT%SFu+Ymj`G6u&yaI8TCEzEZ4rb%3kxgPXoP+G^m)H<ML}JsTcg!$FnIGJ_C*P`
zZ@+|4u-?=#%L+$JF?ACw9}~I)+c4-x4lkTp<6w}|@DjRhmyx{5tvBw$ND|J@FY%W%
zE#|^Oe)8vkk-zz=pW%0Y``3Bmr6&ob1RP3YFk8J_!wgr(v6WzAD9<na(kIw^cn@>a
z=c&)kp{*^k{m6J$QKJ$J1RpvwO1WY)z1E<++N4ylF`PUNzlSglqy}~&ho)&1ESq+t
zgTA)LK+(X}G;9q9G(jVgWMRX5>m~0vHdDLC(S%@eah3J8%M?pR2C8MAefBMyl3yX7
z+oe%oB~BA|-f|4XPg#5VIf582UN$*U6gUJ}HmIILhJqlkc=|#cqhg{?*C-WC4vr5o
zWE)&szRdCY1`n<V_~j`+{qz5jqepMzSO4iRac+5z6b+j)Rbzso+&%_M+ZY)fqEsx>
zSYP1$xg~DCv(4J<9G9N^4#Po*?d73tUeXxD(mAkaoJ6Liq0ZPq0gVJ74ZY-WI*z$B
zgo<e*mBLV<$06H-HgPe>g>nfi_85pINt~79|ENXOug5e^?)%tBSem;)rEIajHczp1
zkl~R*E?ziK&%OG}uOJNhSHJgfU(F38wmM*|G({%al@JRGO@I(tQ*xyO5Nv6bz|=Gz
zoc0mZ%Uf-)6p;joSh)BcC;s{It$wyy=#&uBNH5d!))D@Th-UUTvrZZ*zwtX~GA&h_
zB9!9ur!Kx>%!*9HkSay0c&%%<+A1vi&8N?!KR1iejm!c;N@N_7B8ZdpwM_d~m*uOz
zhHn_#RqXKB>vg{Tz=K&@T`3SQGYgw<Q$W+FPoE!W-c?Zd>ceAoX#p99RLX;lAG-yw
z-oTrlqZt}>FH3feMHICNl{4t{6t$km^K&jQY^1zUo9FOQnL|5A*)^7Dd3laAOAQ`d
zjaf`f95`~6d+)xN&;H?Oc<j61!j}>qn~FBXP;M7xdy>I{5e6y+6e&+V^*kcq;zJ+3
z1Ao273lDvbyGBaHp-*I(gpo%e1Xig)&5c-29rVExrbr1$@Y9q>H#_WEZLz%<;J6-E
zXb^i{cIXKh2HS@QiMmblNs1@3Dk+J&E-vIhEPJN*aPWqGJpY|%DCBiIof=NgB9#eq
z3z?qehk5wFdd|Q5kN<%*%378gmJMM@H*|@;kXS<DLX2Zd43UsV8d|GHsw);-kTfO3
z`5fD9n-d!y(&aVUJ(opS;iV=yZG->s(f^0v|KewuS)Ae3eX*1fmrbHDpe`@6<Qqtt
z(C%*X><m2o+#_hIWOI3$@q)=?bBidUk;)jg6;y<AN<B(x%gjkw2|?fooC;#*9y>|-
zLKCa<2to+bFr+=VMpp>hUO<|}AR$(uex$BO-;Yv?(`PQwXw(RN55v&uojQl>dg%K1
zdIVq1oCEJ%araiMo7epuN=f*ZTU`p7CM0iB%CE@$;@;|lhi}r3h3^=5z41drskZEW
z-dcn9UYg_AllA|7-Gx*?IOx3EIq&n=kN%wh`0<YOr``oiOG~dl!u!Wx!((Y_>D^3o
zf6m`WM}N+L{Eq&d|M(sKIsfrH-f7N-QtuqLODS1hU8Phiy_SmU^WH}(MYr2!+qP|J
zns!C!_UHVsC;d79@uNTIKYsM*{Kt=XJLm5zsO!_GPoF-0`t<4Z-sZ|e{62m9^y$;5
zPoF;TN%}!upFVy1^y$;5&-;LWP}iqVpFVy1^y%|HpdZxr>C>lApFVy1ybtIHb$$Bu
z>C>lApFZyc`axZvK7IQ1>C>mr`+$B>*QZaPK7IQ1>GM9|{|AHy)NWtMFF61J002ov
JPDHLkV1j9w&)NV0

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_002.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_002.png
new file mode 100644
index 0000000000000000000000000000000000000000..951e1aa6a3f7750e94ec6ed4f5bd046376449013
GIT binary patch
literal 79983
zcmce-^;cWrx9v?Sw76@r;_lMmB?Jg5QmnXpahD>+3WVSe0gAh8(L(V;C{Tj5xVr?3
z!{vM4bKZ0Rf;+~}lZ<4K{X@pH_MU6b&x(Gpri7343I`1h4PO}y(nLc;_j>vqiuK}Y
zef=o&_vuaCQ$f#D%f-gi$K2f-P1W4f)zQV%(ca><x3#;6y^FIj-y1$*p4WDso~|Ae
z{QOS;{|S69?za4hxAHqr7r}M~>wBP~5t{$!^{h;`)E?~_8k#alPTLoLaBp8_t)I3W
zdH-+8whfiNhA*h37^jn&D(bDH>*b~E)u`OU6BEvfqbUEve6$X4?#60(84v3V{@iCL
zFuK+EQ0w^wD<U#cJ(9^HyM}D(4_>oIshO!-R)be-<Hrcwg)rHJKt$IYXo>`vycpHR
zhnexV?Hoz?r+-Th)m_No`*c@?INt{F|2p#Q6CWSDKjVH&_}}At!+rI)@o<FE{~S^;
z|KEH;a_?tiRc##{oEOv+Pqfyv@6euL%YqPROUBl%F_IfCMJ@9W-sP!&qe7WV{qJ&(
zio4GHS^j5vdHd#Kf8t<`Z{xA`Y^C>kjd906?c=8tgIhK&#)3~b{&(qd8hbQSGwxY_
zxw3ry=F#aPCwSlG^t4Ozbw24~T{m*g^op{lzvH-TNH_R8{o84uYKk_1-4%-notPd?
zb96u+i-jV3AQPt=A0Zy@3g(Pj3kN_nxPb3_OkWrGi$^vS6tig17N!6pays^&I;(I)
z{~XM>SP@BrLU9T8K;>v6^v&-Qk%BKQRZu337|6Zi%dj7eL{~>22^d)+khibH&u5v*
zAzVc%cGa9Q=$1{7jt<`?s)SP%5M7eXtL)2<>H-6T`fLE1*0D<7hi^S+Z`uOZF~d)m
zH(`Sj`*ldO>XIo|Ka=+<FX(C(^<wH@*z)S<qFf$39)pB~)R`|nJ!}T=x+L{uqb*-I
z9*b^2?)6;7rq1@9jh5=hMHr}50}OI(Ms+GpZGLXqmkR5NlMkJax)r6#IyC<<&b8$p
zLSzkW|C`Od`imT`uQ^`}RBPn=7rj1k)Uvo&1=pxL3TV7;@_RsDK31lB1mArMw%xpn
zHGRk)M>)JPF%nO=MWwDcuKJ+@gbd-_(9p-}$Iaz?kmW=0)$-l(^~=W<!Zwd>`gU}O
zm9tcQ=fx#+%fRpBlr_^ITj%EHSfETsZ{4EmY{H4gZ%!X2$Ozy+_Hf|6Z`eIKilf%3
zILX~oUYA`AthMh=XhZS~hHH20)T*0<0>hQ0#UKi007bAe*{6GreIf^^#<V@J1EWBr
z_+~>1^<Sn;JAAd0dw$0IVHG?-Bxdz2?@$$1B7e6Zl$On9#H}BD2}e+{DW`Q%db-K2
zkgiKm7DpU7Nu<gv%!9(=K5L>9y&B(IUqk2BI*$QORimCI2Q(=d^>}j}SRnYgd{6Ue
z_<#()y8L##@*ExuE~KBWF<H~f>jHN0&j0Bc<F4(7YHAez**`U*FEnVyWqdE!YG`Z$
z!*8ccA>kk=R^ZeidnEPhkWDM&_$V_$6^oPzh(jByaMl$RGMGBVVj<YJTWaM&X4!P!
zPq?<(>`;oK^&M*@fff3>fLI~OtO_0qZ{{6iOME{u>&K;d-~E@5oSSzE^LTLxy2D;4
zncL#!l|kHEBKGn;OXTO~AQOvvMXg<ECr24Xz8HV-#iDbk^e8Gccz^l6C%8bcr*3<q
zRQl0F-0K3(q`Pyx+EDk~;|?=S)U!Qq|IO{+Ais2)*4Z+g0fXSTWJ7wm2tN-!{(t_0
zVz>+mdf~$6tJx=kK_vZdF#t$91?3V}c*Mhos6lDs)T&oCNZAE#Zz5HQDcZ2#J~S$H
zk{I3m9#l^GaQ4F>qFKzSRepG8*^*`*O_PHul|+HqevQ3E_u!SoKDNzdJJ3N@j1o2>
z`}zYr7G`~eC&|$DMhfB5BX+B)-#E84BavLl?fx0gTN>4QzRYb*YyE;|MH>0y%E39_
z_~mKveJ+a{IEm%&?SQ9!@$TrpWRA}E)-Jq#;QEL(k2LbQX0vCcT+98!hfmUL(b?z1
z<FUhiFSl(L(pEfM^N_}35a?9d+*X?7eH-m!)b-^<ciRJmCO1W0XiJARsvt$gY1;W#
z)O~Z0jlU-#!n<SK>=%=&VHN&Ozgs9PmEZ9t#!!{_Ug(+k3;2&rA&$m{?@CX<q$GhO
zzVU59bM=h!g6CLE2UA#)617*=gA>@1+St+VQ7fOZ%fV|Dn!`3uA_R=yt7Di$mTtE^
z&HO8~z5-@kOxUx$;cqn9I6gYyBlSjwm?vyDJnC*R91u|;1s?qQ02ON-sH2Mh7<gLr
zCX5~+>*CP0S;bu4>C7`4T5%R<tbZ+PVvi3DS`+N?_ll=^a|r-^i?894lr%QE{%tLT
z$}~Km)nr?TJD19LEcWV!c&%WJyVS}$=(pnXOnroX0Bj!@&K@QK)S?QJdGA`8jecbl
zDFg+9^XrhJ{^ac7wWy{=MOJJeq$mMvyR+~@I2(-y-oEmdiwJ<_ffSEy<ZJUor-ly#
zAmC&wPF6*SDtpT?d7GSGJ5`-jEM0*-nYYneOq<}8LOMys?RarXzp-W-{$IyVqV0>K
zdBmQdbS9m3Hmv3J{XgY5msI3L+kXASe{hYmpZL$SfLHcmHP9sJf^GS6k}z`&adoh5
zx~1E5D$yY+Io3eB<acZ7yzHOpT^T`vz7KkzL{a=*e4YbLgta^IXJwZUR*EIb9|66K
z<37mmLo3^g6t{J|$}TjHAXreuimZQCjXPK<tjPWnHe`Encv#oylgV*JTKwD~M>Jd;
zi>c_mgN4%jGsX6JY8gHhYjiZe9h+1ajQwGyNd9#+Q*^?^LY2ob_%$3jfMd(U;mTwo
z)-kcW`w0uL4GYeOPGWw&a-Vn7Jn7QV<wL&B)8}VxQ$3)EOBEi*K~Yo_CjhAOz{>CQ
zGWp$WG(7z-S6A<yKzC5<!V$~)OS(0HA~eD+P4ii}@m(MFM<jMAyG}xItPgsz(c(3K
zk3MAA3&e{MRh-bP)oZS)L%!Fc9<;+iN|M^CY!C_cNxT>88^!2r_e2~|IqALO)}U>T
z`ujWHdI;gNN152Y@LHKu4kw)PV2>56POb1KXM0w}am!A{*D4-`#KdjtL$%qS<p>wg
z2gw{)4w`4^#VYG>!OUG9ryE_j;+vAe{x5CGHGTu%*LP3a+S-Oad{ydU@CcGA;=#S?
zk^zEYC`5BgO|*r7sVdKT(es!e`(8Ajc2S-cSTl|vL)U+Uu32Js%HP-wQo}p|>}3_`
z4tU8U;;3CL<FP2F9dAj`4r#03gCi*q3j~V@G%E~7iTyrW1CUjivuSra!rXJylsTYt
z0w)28h7;wcM?^ra%pwE`!Ko%2G=N$Rp4{>2;j1wTfIe3X5il%o-yK-<9Ud?;N4uii
zvLXz^@lkI`Kv+q|v6atZC@u9U^_u_QaCxKUx9X2bpW#YjD9?c10Acs1^sq?VyXNHw
za&P`xMg@pWcB7ykYSn%J7m|q_98I{oAdr9_j(S7=7N@RHG>mpA`ps+e*{05&J-Ko%
zfQd9sxE5$iFca=}a>X|v(7$s>6}|PZ+QI0P*Go=J)T$CKL!O%u+|B;-75~zM)O{}>
zGNitbKWJ5cRfvN|_F2v{Hr6B+*=H~p5f_Mx0y?Y_>8U_!-hf<%J@T%vGFJqEtkzFd
zgUl>B0dzU;m+OS{-@NH|F%WY$jNnx*26HG0JCgg616YB>!D&2yup%2BaH)QrpQ0Xu
z*oqO^U6fSfuKk3Hohj3YR|N0bLM}HPC^br10+|+XKq<8q6tgEwlEE!Ey)(O`t233K
z#!){ar&B%BmPSugvuoc>NaoC(wA@i=e-pq$)+k;iAwOc&Et-PPM$ZB^jc98XkLds2
ze)W5pa&W1ElkP>|)-`?SZanK_=l@r}$u8SDG1(h_t<7AdfHgF^8vgPUp*)r0%w9OV
z-_TqypTY)UbRCz*Q8%gvgq;y+ykf^h@J=;qkP$1$<P;YfsUGtvh`W7Nji_-y3qYLQ
zQ5ABl?9vs3Ly~^&Hp%xlejP2*PBj<fM^lEMCvX)k>N7(P`<0>4JF4G<Qdilr>$`t}
z%U3--Owlpm`dLGupKU+Q8zM}qWJ}eM<|XB*H8KZQK<o^8<4DwWvb;exvQLhH^Hn(3
zBoh(jMxUJ(93fv2+iMV+B1=;qXmGuHc_jRbfYU=@(Dll9?DkN8%Db<cvPE15?tVQ5
zTYM)e({An%*4Xw2F{3!Oti%Zn5C;S<MeBkza!d@13=|IaXTlZ%(~L<Ud316;yw5Tk
z8;5q><4Dq++0B1BP3|^DS%x%{ve`z$@!3K4@zvurjwbEEl%`(g&JX`?rc5oVgg=(@
z0??3G;?zt-GMKn#a(ZyoEcbxL`YIfgIni1!F)w*SU1O<V>-qB{4|pD(hrj<=bp?y^
zt;_{fUX1TcE0jb@n;okdsI60%f0POEqWpJv-L$Q~`CFfzL+GV8V9RPAb{{Etd7SHZ
zS!mgfDs}O&uN#VW(0xB;UZs%fUwmU+nS7CgAM@hzO^l=WA!NWDou{u!HA@}<t4{m(
zh`t|yUY4Qe`?$z(va|Y9FL@F|o9^Q?H@{%+TCUXk=NV;kS@^Z@;IG`%9fV=P-E`}G
z*BN@N$pfB)L;5{>(yGlL|6jaJDy^Hp>QjUO^cn(I4;PjByTwGc0wAtuT7s{Z<TV*n
zLH!hl96Srma!p%5B#%+Wrq?g(Ro-T9<%Z~XHvh0x0P>IlKhok<2e!+Vt2?joZ&0cq
z{S~*8LJZ?R*ZyqoE3>`h3d|bZ=)%zoX+%yqaq5+M5cA9>{g8J(GOA8btFt%9$6och
z!Bt?uv!-iDhm~{ugkL+3&5y{Gj81N1gyjEGR$hxzb4Uh|P&t)xsme?MBxNoq+sh)M
zBw~EV#WP1?RIgKUUbvs;;4%cQY9l3YCC=KSHhB9>vlLKjjC9DfrToTMaD{Rza%_xO
zwof8p&`OGh-<!sEQk04zgp`0<YJ-D-fB`*^b;{2E{h4yx_vxS4_JdXiX0XC3)*mQA
z!sEQDT{g0BtEDasbzR*<Klzt{${}t9mOu3P2do_m547Btc0E421*bxz3)<W^(ofUT
zY6cBoaJ$HBdw2%*72YxkW#_D^d}+k22CyADPwji{UKn8I4{uScO^n#uj>TV!%leO*
zS*DM_)7>R(vG?*X<G{&|WdDbc{}1|Ny11xMIGESeQko$Eqtu&(2R-_``S4aLK#N$x
z7*xdJ8Al|p`Q1^%+&4%u&HDW1eh@szo+;JKzdp`Qu}|`hA_TkLxS~{O@e%axzr<{C
zMJ8hr3v>aTh?X>}rmVj{B-=J*1$Xm&e>&YNyTNT5l&r^@;>p2ShdlPp@i31h>#Ies
z)M43X(^RB@AU!9P^vy<hX8Ur5&Ce1!;2tWrmJbiX*R6oL;PiCUPDC^Jrxcx$lRLaB
zg(7U_g~6R9eKE!2IJH)VG;6ZWgv+3t$iKH&Lq;7JFIr84!VVX+-{OL1!uReK`)cjT
zIg22aVGfjr9~erS%glK0(A}TC-xUCix+Cm-!5uy^;K2iDXQ5XFe#b+yXOdVeD`t5O
z9QKzTdRnDXx!)5xBZy^oS<p1>iO|c*&icwb67uAm{t-kF;aI+dab~J)(Mr)Lmk^^&
z<a8v?cuE)lWxOk9LCok?3g~8EzcT8s^QeuXFVgO<uN3rT#GkHWex#NlqsN<27nfPE
z6{B;%yW#`tA^nNnSTuz2eYx{)CQAx<a4Up{c4#f19{0D)RyYN10}F$Eiw`gCE8lfU
zW0kigD1I@!3L`MMk5|Y3Vy8xD4C0{<dJ^V3<XGi9wNHB(v3rg)TCf)2884M*0b~2Q
zwNqhmrRZfa94VjRJ2Wj3D!)xM^*RMNu|u7{*=_aV9iy9F6K}@2JN*)kyT8!(>b<bD
zQDA@V9Yo<3JIBixoz5<_H^sDGuF6FxvpdgD84ueJhH)p&y!Ox2JP@TUr<erW+w7h9
z>&IuOk+N%3Hrds&&+MzBQZcu4taUry_A#Y?O$AvA|JWS~jEg_^@DGTu&M`bN=^lC7
zq%4P+vG$!#eJ|Ai?yE5>kPd_>WYyFAY|8oeKD#_d3BHX%ia#lBpE{+|MFm#Qy;GUO
z$ssV8r@lZYEMvkpW~T~wvd4dE*5^GIJVB<<iWaCt&(9(I&H1U-l1l$~R}Pop72%4D
z0^u<gu>5F)#vjA4MM69Re76%OJ^PC<ctMTH&$*?#`<7o5sWHO`c{#5+{^2g^X8ktT
zk517jJgJWPPE~~>tQPeXTUV03qv*Nq=$SzJXaD>|N?<f{bHnQ0!asTB8I?@t#Ditm
z`e&n&vyZwx2cysstXnWH&d~KZG`1bt`LI>n(1qdxDgt!Z;OBaHw2+mqV49TjN<o33
z1}cYNilprHU|!?!ifML=Cf<<->;cD7-R|o@RZd4{xZ{(PkpWjnlHkmIP`{@T;{+}e
z$t$wrPV7xdP~N%f#n*?85B9VouIT?56ZnGh2Etu+k-U5EICXcg{;(33IvcpYpJyoE
z9pGhqN^SkkU2U*3Sy-6@6CF~$B~dy(ld`GAa98H{P<VXTJNj)Rz`t1~Sj;3~Vz+la
z)gwL+&>ehyxOH87cjHnE-@o|4?VdheuajB}m#?A_nJX4`IQ;n0ug=YH6Y!Ig(}KE|
zq5z~`q@&e%wbItr_5y`BRhud%L(RejYt}BY@|7vhP$;tRH~UpE%!|@jtj(T+lGmT0
z{)DGw;<FL7g7Nv7UUyg2*KimfeAani6twhF%RHxe3R6el)$=ZcaOqR_wP*kJWZ;zc
zq~9ry@^3cHI6%^C$M19G8`toH;mJ*SMaA$Lo&)c;ZTwqsOCzP;a%bPd4PW&K!URSG
zFt``0v#w9`r*0{BlF>@*c*hc*2VvS%!vPhK%7{IrtRGxQi(#@2euVE@De9!Q&8IZ*
zWMykxCR@3>645VPp1v^AkA#NO>+DVBq+Pp^u)i(G;3T0%4`bsNA|SlWpgL@OTnV3d
zgm5o@?dE1G(sKf<&mdCk+Bu67@;_>GUI-K9q-d=Fo_}$<CAZT)HCDoWw0NK35_}t<
znaeQccQ0|@!V>*L=B7~F^(#I?hrUp6taF`S+yc{2aQ*7^B}`F?>{?v+ymW4^U9Mx%
zROqjaBYV-?uK{iL*AL<Kvx=7W{Ti^cLf<JGvuz$l2Pjh^k9$jP0Ce#u|D>8>|H!4U
z=P@zx=1H36CR*z!SJcw^VG)}$Mq|)BMEdL$3q42{_eqESw^b3`PF0iK9j-TsmH(^d
zyF{$}5*jumz*lN(<&-+xfd*f3fLkI`gkpW~%qi@<Zaxu(4Kq+AW_X>8B4)ImXtec9
zfY{pIU3l2N(%<V9W5h%WStDH6`Plv9jMaaUFRT}>m`s@^6NbnfxB1(1`*|q-Uo_JW
zVS7<^#s`O-`+IqV!<19kJeSAG<;{U~0^fU7&)(0g>bC&-9L1B4ID?T$5&5!QLF8-l
z_xe4OsWDg)&1Ox^{6e0osH-DsL|mlNpDv)j6^n8x5w(UDM<u+02W&vMmHLA77o$l0
z2{FORwVx&3hwut|o#*PhbVX05u55h2M))ML7>9|j<VU0hte8hcK_k%`)6e9WeQCej
zZ!?fl8NdoV;pijG!0BocT^Q4ACUf8un8KFMPo}s6au%Sdzp;VBR-{}JajS}2AGXXo
zQA_YZ=rW`|ZDGw+>+d6NY{mw<mHICmN5p{~MKC)rWYl_!TB&+0s=QdR6&M~hy0)+{
z${XH~6YbFuy(GRLfN8i{yZAx^+Q9Q_ER76z^d)tfP?ccJj&7R^5YS^@#+<1d?%E8z
z+NNN9F9aL`fH3hFKBnlyW>$MM3fDcv-XLlI!ikXUtmze3t}j+qo*c8^Hw9RI@tBO`
zM$1a<Wa#RjlC4SE)`^FdUjF@tq9;6({;u=NSU-uSWO-kRs%r9MD>1-kofzde=qa?h
zv6I1*O*{luV`qZ4!xSw&1y{`;pW8HIr`N`P^Si~D$O(BlB`Fu1>V)$lt%5D2L_$aJ
z(J)o>Ye337tez0zLCrLAoy_EZ&OTNajcfoUY7MKLRuj;~$5Dwxtw3el+cr4zHhj3G
zkH%x0+a$1E*KbHo>iJ@j`Gg<+;k6dq7ONX=B%Lva6_pmr&sTVS8S1<Yi+<lrO#+{>
zN~2Ax|9gMrq9Skqe!Ed&*lSiV)F69CwqI9Ixkyv*tuA$GBw^d-X-64L!SvCu{r-ak
z?{98i@m*cgUw&Y_{p~Y&*R#xSU)&E_^BlM_h!46{Ac&8PDhy|u$Q)P1i<I9GS=(!t
z{$l$w;PTI7)S<(hZ~oWTJ<sdKc!J{DNvhI-f7(6iW%Xgz0w2~>bZzMI=#$xaw@~5;
z^1nSOV>pbO?!aA2MsWm-@wF;l$rQ+K>;@^eZ7aXV@o`E5h-FbM)i&F{)nT40x$~4f
z*cUHCT*WYHBmRvd*0z0gKH3nlCBspTg33(0UVn=0r?e?09Ha@Slg`*y9EvW?E)2~5
zB7chEkO2FC*df&DlgHV|wk@ImX~Xce5buG)J3h)uSmb_T+9*wO{V#m)&UkFavLi(`
zB)w#CQf!4QAX;RY>_fL>Hd;K;2eV{nAELIoJ~aum$UTU-rP235KOEfnR%`+o${$ir
zRn7zxOi>tglvq1Id8U}I3YpxYEq}`y#R5uXf@r}pop8SUDa}PI*y$MvAWQv<K^zg<
zlfu2a%K1cYP)+5wyj)3K>(fN7Ue78&-^YX}y(d>B_4u06+F)PS_T!>6jivs638t9x
z3)7k4FY4^K!%CUz;?n-ZyStWL#ugRO4^RCE?}9R<NA~M5_gN>}px*WllGdudV0ar;
z-#31LNv41Z$aVAfR}6OgoFYuX&LBjm{@$&RyEbrg-$+-<OkOb#lTKYVesw!Z0W45*
zPL0PdFBprrE2ylWMs^*(9(n9>vh(bdWQUYYcYLD^De(XWC!QPG<RLy4!NmOy35f$9
zSed8GlBLMREg(>#%@Hagvsz|4huZ#?E1N1d_s3qCK?x+Ji2aD<AYMVIP<T~E-uZYW
z)AK;1wNOnUx=e@&ospZrHGUP_>OefWPSbgW|K+ur5sZjl-a@p6OMI^9hx+mZy|rNJ
z^uBqA1B>R_u5PWz{Q|w@CqqxkH^U7{MdX&BmE!5mVdhfwq{<4V)2pMK#tVmh>mwPH
zJAvU%MTr>rei}kqxo1PcCz5v;%MY^&OmcP+0;{DW?Ja{ZchT&4q~pG0n+EgIBFOFV
zqB7d1<yi%?C#2a*;pWl4=PL>N_IFN+NgtRvs;mU&ic;CMOP*{sriqSC*Y3YGO4gza
z)K#tkPGSvU3PAgM<k@e^vl#Y86NnDEw$5&9bP|g)NK*9`JTWDKl3af)+S=g&KGZw^
zf>1YXG`-$#{KuE!WR$%Eqg1M#4uzeuP+nZdxFXp#tYNIaoh1$f0p`JB9SB|i$wM0%
zfm)F@fsof3EWJ(FQP4;MQKSfs*^;s)Gis!WAdIR%-90KiFRf$_S{(}KD3P_7Uj-Yp
z)!k)F@0JrFO8@QZtX6g;i#TIt@bZ~8e|=FnyNroG3)*E7l^$;VuIuFDm4U}?!=;E5
zL5f}+!1YTXn~LkO&BHdYHo!}<bsBRUc#(JU*WZADPy|4Sud5H<T^P(q`RGJpgv){D
zbP#@?kAX+af$;?y$tAo=fZ`kulXkDn*x0Ak!jw_fqDkyEGqiz4RvMppk=B-{OD)@^
zRX(XmrGykQA^R2*S&q}BZRLlU-+2y$wYlFbAd%jQNR4R&14CT}|Mybp;q?IB>wSXX
z^xcY$-zmjeK=!!y(zSXz<QeW!)sUmWyC{7Vvm8dvB#}@NWjmktrVctpq%GB7%}D~g
z?Kf|lo!{tH5k+w*Do|WtNUd)t;(mGvSU{D!?{z<nr(bpT_{=`qtzy%g`d>)+)1$5?
zqs9+Y=6}0BJ_y=wTh<(Sg@$_cR?g~anP!}jOC!a|Wr4n!77F86JfY>ku(<15R(cmN
zeUEdF1*jR!o6JR2=6x^lO}YbP96GX9Y?;;K)H1!cIH)yrpo@!62KGQQg{tam!8}Cy
z=qR_3bf1Az<+S;kSzdikW3X%g%>L|T$x5lqW5N*Ou{$b4eKq_$)_j7jgd?f(d&m{M
z4dVoMDQdR>Ib;t{DdUg>i7LguRp4~{ZDT?}tWBl$m5C5+>`>k4XFt3A+_t<T$-85l
zyxdU0#t^tS_-rsau>!!Tth*~3MDKTz>VSXoC5j-sr{`|!XmnrZ?qqWG6mevL`kSTG
zV9u>uE;ZU|mpfrw`I}kx=1A5^oIXd!G1qj_q?naDr+V#J;&9;j`22(EC)a=$cd5IR
zhcyqEAn(n;l-=hJm9s7sRK4uc(4PSpfJ2@mc`=%5_|K9`GM-l1+nKFzzM)4Tsp!-g
zisfm8zB}<|=$~7^jhWe+CwDKc;#boY1rua`>6Pon8J~`IrYg=tb$H`6G19Kas2ddg
z>$LohL!c5yA<YS@AZz+K$|+10my9W96r-YH<-jUWD*JS&_mc|ycJFr}$$*i;alV(b
z*c+_X(<~%aN)lC}vLPps1&b4$gt1SXSf8R3O|O5}uy)|s=O|V>^D(&yyV$Sa%EbL*
z!Kr51SKO?pV1;(!#k+0(8C0?cDTO4uHPBV<`IV6Zwx(ZI5>Yo=FWQsmRwqz~^^}G<
zq<iZbnzGwBPo}a>2-EiI+B3aM(qb<fnoy2-PmCj%0@UWtSAXpi^nkPc_|Gy0roqJK
z{A1~eE&E=YfIfDIkZH?f{QeU=o)W2nePuK)VB4$19{CMz4;_ucD*<UyuH}Qf7*~cC
zg4hl??=}4Uc*Q$d$lTO&?(cBmc<gp~#%CBNrZ2X7u+i@^smw5A*X-Z#2I^Qmo2%ea
z(#_IzfH(*s3f3}CdDb@^RkDhT4pT-Ao;=Znrv@YDsoV)cuwIeGMTzmF`eORyjnZ;M
z3lVp{3Zne8Nj}*l>_Hr>-%QH4R-06eLhxy@wep{&@oIdr)|PAz2MvmI|L0a=vm#Xc
zyDC^S*;ug&1mx_X5q6(FzW+*3{$pnE{D>q&VX+#36GOwCmrt5|&e@b+%HP~S>p_{`
zB_tPO6CN?tcXTzbFuDK4$bDzNV6)Ai?eP=z6^H}>#lbG_>l+!r^hjKiC6t}}ehX8w
z<IS**JTx<ZOgJu{Fms&k;Na-1HYjUwhiQH4NNt*$5o(##q-u?CTgVvs@)n+x<L)oB
ze)5{#Nx`UVDSwnX4nc5r)cWw5(zy$&Wg`A&lztJ!s*-_sp=N#k<_5|bW;>%M;2Tk_
z9TVG25!OuiW!kb~KH=03mO{izTpO^b`me@hNHf16gXiat_e-9ybBtZqEZuUcg*Gey
z%w}TF_amgbJpG?Nl2*);K|e^E*C7jd@Jbp$iA5Y-Tc}XTe(@0=u1u+7T3&nmz_ex&
zff9bBa0%0djTH*pPmFb<1eC8SrjZ0l!eTE+e=+z4oZ#}Oaff2HURDUCjRE{*)>NO?
zaF+45oGF|nO6s{N(@&d_s^4FlJHF4qP6HeESLDqCO1&^G(ZXjZn--{k_skpoDe_Ea
zDOq&B#jz+CnBqCgwX+CcvajmtL}ur=fux>i4D>#TTQeqJSzvQ=xA+(aTyXQhCr^V&
z1|9bVMHc>Gyp30zGg@X*b;$j8)6G++msSp;(0+|O_%Pry31;^_Iqi@NI%k}}gYBOG
z_JP@zYF~#Zj&PvF*H5D+uV5F=UQ-`pnBH;5h&${%6!9KgmDtC8kOU48m;Ugem-Qb$
z-4%~ivQ#WgioJLyDd_}jrlphKs%8j?oCrKhxA#G2k)?a&I`?f`sD_bOcYt~LOmFr~
z91<Y~1sNcSK{cQKlw6aTo?(xQrf07L-UJ&5>yFQDr|YfQtjnU}04I<O%2c{+YB)w7
za2hSZx8wD`Vz5w-Ech(tHbOdB=1FId!<w6AxMY;Do`kc3kh!f(ta|Vbs|>1Sz2|14
z`{z#8cY~iD#{pwBwLN5?LqDJBypI|_b_{86%_728%YJp@==8l;@7dHE-jjEs{@ov~
z$u-ou^<5SyXjNRblCHkFIWB?x_V%*P)1H!Xp~-+)UyNOu1Rhqvp})N2h8iVToaQW?
zOgP<m=2xC5$l^3)cay>KuOqR8;UT4A3_R6s+6h?|z@qR?%v$GwGQ|tq8GQC#sYJ6V
z-tk=6C}DAGP1JPtcTaDemWN0_nGJ5gfeDEvyi2~Q7(C>SN{Loc%G2_nnx&htAdY}C
zu||c1z_GvFi>qg!V4ghP@ROqNZg4XzbRipu%Av#b)=X6hhI~^CJWp-d26K7@9#Hs>
zY$i1pBeH1Z)e9AauhD1>lZIR`PMqAi&aP8X$$<^mR9e=cDaK!nGt6!hKs|Yq>*ms*
z=ol%Gl_ojtnWChRXdT+0%d*zu3Z_yPzE+SF{Q#JrO48*ekKY@-OQO&y0>*xRIs9pA
z)~qQU3+{&dgEcgmIh6^*RM0O;iM5QGUxib_rQ~pEsh?=!6d2a^Q&-<``K0S(>}jB8
z2uGC@WAU-bYw|un+T2*V7o`Bfk!1O+-wi(2Ibko^yI<)Hp8Gp_dn4Q2JR+S-f93`5
z2bL6~TSAV4&NC=oOqDoavG>!sF_gRU?6XcRvb}m~R=!A#UW%ua|D<%}9mR)U$&;{y
zAjC>uIO>8sK2eqtDPJb9;Ln~H)dAAKHnX+R_aDJsDCdp|!G=z8gz*_FLXc|WQ40mN
z{2}_vC>vkzw=y)-yETy&j=Rh?YW{Z_#;^{}-1+6b&y)jlby^x^$!s#!6aztx0y#PF
z$uJe3%39j5N|V?j4BwmI)N+4H45$EL%-D{obxp-x+x<-UwqjVxd7rG5tZXfL|KEQL
zoGfbo9`DQ_N?E6_Dj*OKGncWz&o7Qn)s0=V`HK`|?4j{8Qf3m_LJ88r(|c*$nif?q
z)6752xOdS>!yV>+JM2J-WQKFLscnaD@otD0)o_dXLWIG#zD>b-G@m4^E|GQvPE&2F
zRTc)yq1AZMDDNq28D8*oO0QY0)P(P4*JT2ig@aQpS8d&3otRyH{nGl~0a<v#43x7{
zP)Cs23I7g6F2=HUFN!2>SzRXSYn{YSUc?CjGHGwTo2&lkV<J(-t<h9hOgQ78%NoV3
z5L_24zs?oYtC>QM6(N769eLu34)i06!$LD%W#Q_hNqMp{N>g5q@f8Ebqur94$OZKr
zo`s5B@VeSG@|*)M0&$AZJ4V>LwurYyzSpgw@nD(C<MoF$z96jwdxazP52w9UpMB}o
zkNu8W5-ewp4~Sn(T0qU~lR^d3+H3<tLhaFx2ir^JB*0bU0W%gl7%OMM5sB@P<qPYW
zm0zNY4!e@CI`xcqIiHJG3g+vU5u}E8)5vAD?EJf7ss5lf;*dT3MUU!|4M;f^l3(|=
z1pB(5iT1GO_$VKwqO$=T(qfwy3~MGx_ilG|@)Dop_w{{@ns-#Jq$A~Qc3n${8`YA^
z|3;`+uo*AHqS0W?{5G%&^(bwHo<I51arFk4P`HMUT&<F=(_0?au%j%OQnCRsJMo8A
z*w-;Upk6o`S<A+0@3T}f@M?CA62#5EQoWQk40W$xs8^smczrEs*zx8Eoe%zl(0BRl
zDDB1NrWXy80T|7@Snqw_z5jgP^{#qy3A>3L2navVv6^a%)zqkq5c`e@pFMU?oJs+j
zDF9fd<7uL7Ny<C^Cn7m*eKcK`i{=AICSea4ogT_wbGiiG;2#ABfIDWkiqmUR$S0Y5
zas_Lg=G=D(OBmFkvMj5Bw%6t({UbVdZ2M5ZB?F?xSoLL?vd4W_|GhI>p#AzYN%FX7
z$T!Qw<5Sq=E^(v2Vv#2D4-Hv6i)Iqnw93S`o-LKG?J&qZg@s$>sU1WmcU*mt42`|O
zM<NYi?3i4TxkWmG%($zaLdnk4`Nj8%8+C(Wja~J`RNJYdRbSx6RdOq6p`(68RW%~1
z(RIGwUb@9#`)5zFU3e4pk*AofD@gVr>Rzqkp#z^>K{PI3jk*7neMncjO%}w-fSoZx
zQ6708$<AH-Zj)<pz0&kn8x@dQiBHjSYw&jp7wXvJ?(F{0Y_rWkBG|ILIsNhL%!G9`
z{qf|FPu!*S#|*XIANaNw?>`+^!*33@gRM8`_}u#S8zoHhKyP%e*O+A5TA}&bs=B9P
z*@l0{QS4eDCrSSXxiwJ5APPw5C;tA)Q<Cw044>pEok}ZcB1>I8IgRpJYFA3sS0lvr
zwB|%5iNt1(<1|#AYd;@Jaw`>!`bqj$<of4kPy305ExwL;A(suuW>gW=PYATLh>pD}
ziR@Wae{GC+(wZwsV~zRcf+j*}nbt5}+d!EAT|?)}##IUr_pelfhuj{>KHeO(VVEj5
z62Nl*AOXD6h8wzrjhucx@A}DfuUD#k`ZvhHLnh1ha*1&oMXDC*0Yebi^oy&v;V_=D
zKYuGjb)048`E5@XYq?}ZBGOZab*U(<;3*Vcz8=ADjDDnp^lxOYngtK%5dIUF90f>a
z-28J;bMpjP360Fn8R7aeg$y7MUBl!z==Q<oUqG4c7oU5c>s4k>L}IzOw6^NgR0(IR
z0i-Uk%3@98u`;ZM!`;rO_)^QihjNO1%US=6=fn-i$n3XL81a?ZJg|~|KkOUh1XF~_
z(+;r+6;!y>OZ$Jv<qt-*;*kbvt@tf`9CDuYXj9AkwCyA(O8h>J%zD68WcA8XQG=@>
z4fMz6b_!Nm#w`FJ`r7>GFhT8drU@LKLb1=%#|yyS|8~OT`9G?|K5X?HYuxC`X19TZ
zRMrW?@bY@{@<=Ef1nRS>wi2rb_y&yP@*COnRS`JIaEi)+c%GH%ny`nfM|`bkv-Ay8
zDfNowfX>TGu&T8mm~_eUL)vanZ^yQ^+Pm<rR(mFP;JeqH3n^n;+prQ#WJ^n?wYbud
zEo~m0sgTME?@60YPY*e|5Js(IS#q%@6UQ^iZZxaw%3Xhvacdz+<*LIqO03~h+5DMO
z2&i$=3scCzXTLk_x%atVYF%1KwYKhfZL4f==Ue?1lavp!!-J+IiRhoaoQW9}kHp#y
z;efvS%H=xAb^r|G^E+Rmt)v@#t79-a<sKxqxwI&+!aB@O;vf|_{z@rPhs;l*bg^K|
zuCf@xB1&Dn1uJ8mz2MHagV*3RA)*RXz??RNniQBwi8X*6_9O=%-7N2fbYss9#UZ4@
z0wI;1n9llR7goSQjX#uP9Z?^He2ef?Nx`1Xeo80EitU`D3L+RyyA}VEL@4Dz$Z%&7
zT4q#i)8a3cY{!Ny6{ks56i8CB4lHtFHot7)F5o<7;szv{&Z666s^PG>Z{D$|-ro~L
zZu&?#nLUpH(sZXs>I`22=Dr^}wq+{BX%W=9s^9!)m<+vAYn_-pFRqYpZ2SsW_mnkc
z3H2Ds<z%uoD5P%|`K&dBa>l{Q6i-zmDcRA_Wh0S8G{6h1bW<a#Qqavs^G+Kls<%i>
zr_Cb}7nd%?G5RF)k&U<>>)wE_aUQ00OZar31S<ffb_FkNBYzre>+S!#mnWKOs~3og
z6{Dr2M=c(A4fq84rdNw>Pi1|b5)Ln$_P^-T&kFvWRJlG6aP{_56NzA-FdO^-*tWcy
z%gS-V!Q;`Q_Z&=H3WE%fBk?1H==Sa1?{cH+6kRXZhXSsDx{wUXSRKH(3(jzZS3`jB
zqs%R4JKDc1A7`%({)fJPqD7;a@`N*C04b^b3doClQ9k2^^_wLx=k60OrJ;D@x7@q0
zt^0k&&Ot@iBO|BJE*SzqViMa=w~(QDcG_4jB$9lJOQITMGEO?*EKMU#AloY7q_^X4
z!L*KpvwlG%^%-ShKtvW9y`-kSs?G;(#W(q=G7Yq#x66UYfe$-|%Ui4Wob6?8QlFb^
z6bA@cnXA00D2I)l4o28xpx2(Tzw-*5A;cA#7@BivSLclW`<>{1?&+S4lxN$-q)g<h
zr(qG-of}uaQNqM6D#U76^k3@>EAXt#CkOcoXw-NVcGKSe9)Cq>+b2N@6z#9LxR#pu
zeyy#lpT<$XbBP46WvEq7gOt&V&WgUV6}@TvrcQ@w5C`Ty`HJGwIT{i%d|vW5yW~|0
zTnfeSmhr!i1@wj+1H76|*Qy-m{$=+iBv_{*MBz?@R0VAc!G|OzT*DPuhN(N6%#-<g
zKr<NXuTM(}6pOGv>bIx{I%A>`C}HpMo)oq>A-dK6aVk=i10!_Tb8v{^IPB;>BI(yN
zoy2j`Qg+MHsn$6Is&AXxn|m2=*n!%%Zg_6A(+uaFjihH!EC;QJcM!tyC}bl$`;SqV
zclJBObm}-Y=|q>5wa=Zy0eQ;}W>PZi=jR!Ig0==KnebR~$&vRO;iO-S<#0>^>qBY@
zwpr_iLn%6|e$!3Z14OwkrCSv!d-32JoK6d{+(s&&o|4W7cpm<@$CZRak8vZiu~HBJ
zY8Zf6Z@q5ta;*A~qyDz*;MYo$=-PI&f*K|r@ahQk9s`BgEO@<{Jo4Y2nS@u`XqxZv
zXJ9my!^+hL9CU71PLcnE9|)qQ%he(yx98SxFB6aq<00h|!^p+LO7335$0Dn4l85)(
zX}SA*_peSeqVy5f4$|$1^co~#FyGVSH}<U^{TeZ9E;)XJ#a!<!IEQUW)t<Zrm5Fa2
zm+}t>9kO42x8=XslER3`wwaUCECZ!kp|90QZo(*Rw_oZR5_3L7UKxYuj5|JKS42}q
zoKw;6(;CneH4vqE6pJpJ<OScGKAN8PzLec$ct8aAh7B<!o9H&V>TwFbfHI+#IFz93
zzA%%MNsDxdbFx5gFxUIy8sb<2G&5~kAvgSaFpBd2dEe%7B!A$;H1|-u0Y?tFy2NJR
z)7gWEa#{c?!Mz<9htQ3a99e5lukCJdcXkML*aWCjE;Daj>LR*sdwWMul?ac-CQy#u
z?tFQ*CCa*5@XFB(Wn=CQCaa}C*JGpKBwYe+H+^Lc-**hN%e+z3J<b%2`BYJ<!1Y1;
zIYOu!e@;T;83Lyy89vFO1Wf^*M_YkPbH0+k)42JlS?y3-BWCbki8W}jT%N}{$E*8~
z=86&Nv&gjZ5<BB0TXJjr_gpqkPBBld*4AK)ivpRSEuZqA+!Nh&s>46;V!znp4avh}
zs1z;fN43IdQv4J&W;P1Y$(0L+H7UgPWr}s01WV33-Am%IuH9O_S3l2x`G@Y>?J9?(
zF2@98R9mQWF=qRTV`CkPz~wSzDo3{tbwN5gMY+yMdeLH(`VEvPli1U#ec2gGZu_GC
z8u@B3%{e!e)gsqn)B?oATTX?RID}=;YbBwiE3>&5e(b;DSp6qG%02z`^UA08b_xG=
ztpZI@(-gt$C#q_?(GoE+!4VTG0@=B^X=)FAv7bJ-&;=!depq0(_u(A<R;pFrnJ6VI
zxpjV*5qTS*F~PcB_|i&2Szmv&fW--SDvd?e%FdiM6QM|@6>2AHX{e<y5&%VpR)?I)
zwbGDoRrw&ejEPG*YdddVDU+Nc2_I4-U;4B)cg*cjF_&@02s($T*L6b;Aq@4E3Xs}m
zooa31t4sae|LOWLEPe07@M27cv6c_!R(^>%1qn1shVH)!cvN@58~g}k?5wrPGcm*X
zD8I`ddhEj<jvL!Yr>Of>K(5{r1&hv%AO$i0sP*u7>E0u#r1fM$8k%$t5%{5O$;`)k
z%I2tP0!T5qb5j3LU0K_Bj7DV9v~8ee%*$KCC!!yIDJx?gC|s<g4N6_saCCO8>+Y7T
zV)<~p6i71rnf=}f3r8PoZTOO<ikS{65x+k9sv5t#>BoL;@E2VZzs%dqqtn~VW8rLU
zrd8Jlpl;kX&yU}deF<e1m2Oy^kpoO5zdPTv-QDXS{ud*ca(2$J1b7#~ud*Devxe=x
zuGi2#S-k8reb@vBnt~kmhjn%UZ*vrl+_%zbr78QofLC^OV9(8`8HKcTZR=-hB;{0`
z%;5=k>luh`j`AqxWevFqQC;qL4uO|16Z(~7J{UH(h`r?+s1hJsc=5xC#`Q$!vmn=r
z({;g@h2hx<Vz1KZ%@pj|c!OFW2Ri$#hWh#uM<Evbi|-#YR2r(Xt4|kkDD1N^?cKXt
ztnK)Y>Q(Zy>R#-~KF4Fd@ao`#!SE~@vvT3!=Nv^%cu0e$KlDG#VowJ!2jlbDHH~*l
zwnDIo8X{z#V<Y_8-Gpq#em-flYTo(Iw<8(e9Wk2N)Aov4SlS!7blPh>EKW&kgdsod
zRcl+-YI}OOiYpI{f<j^se>Qb2Z!Y~%lKRUa1gG$X1F@E;xKvqloS!(J@0g26)9bzT
ztA*d1o#P9zfD(V@UMJTuHzqCSNCzFxc8{b7?Y;xpUM)WE(Kmim<P6oa<b*;irg*d-
zf9F*fIpSD{msOgFO`1Y3WoiC{#T7~ik1o}k`%p!aI3(Wu;#uGi=(ISvzzAOD4~{l{
zSfBxH%l!bvm?fH=+7sq!e!h9;)lA(Nc@XFotxz7$;q2iHKp$)}q`T=7EgGI*T4Gey
zyy@~%Lnz`Y*|HE<c^Mq+CmsIOeNs*9BjSbAq-qmt#9YNQ$8PHp=^3z6m95egf~78x
z!I52fvmA|<4p}KFm-Onz#JvK%MO%j$vVcjktKS{*_R9yQZm9)EdA}wv0#a|U6chXI
zt3D@)@tNx*J(q?ylSP{T4IN=p3&H}TDwg*KXF$RwV2`edEMy2AZjuvJ2_wc(e=Ja)
zi?|C&8YKQ!-s!<crpdN4YW9~`17p#7EQbH6q5eM{dQP)kIzThO1GJ(}+H8JF42pju
zpY+}lOP$PdWqyts7W(07@Fal?hlP<y<c^nQzZ;b3X^KXifsQf!2_9o{3x_iKLkePR
z{kbclekK-GMboBAY`o5Yc`^|tFdOp5u_xTPC07Li4LqJ!tCZ)qH&<2Ar!b-XrXb;F
z9?n=nGZ~5VY@gz@gC*^g&}{u9?HbpM<kMo!P$LwumU|^>VaxeWHIZ^@xTF2D%qKLJ
zgkzK068#xO46{onaQ}~jbaI-8*DG3LCNDuCNN-z`Z6zQ*qSs(-w);V%w&#J#p5(Ty
zsE2u&ZS4rSC$!^5DPayU!^WLkiP453U3~@4EwfsPnQVR?3sd3Q74s&w#fpVaPjdF9
zjg`ue1|SxY6p7ko^RUws*rY4?!$+$mcYE7-O6Nsj8T5&(l9C0gkG}0#Eey3T_FdiN
zwIDD2iLPb8ECCrF{Qf4;x%h2<Xh>}gdq9u5u?~xqT+f@=$|PgT3YCi`EEc)c?R0s-
znL&;4?*5h#ZeQsd^!aLm-h9&hUm&b-Z@u06&r_pjZ)A5f{*zlzKIPV0vxncUu=vpP
zG3xj7L80E6x?tqwF5qFx^Z~u+CUzpZ#X!5H&n^~fHB9<*C03EU1K^++z8BU|+xrU0
zmB5>X%px>RCr40dr&@lp%~@QegL~YKFx=A@`i)!$pUazkIu)z!c8%G8Q&SgYPg+c-
zq!fR>vgngJ4v~4X9#;cKk7l?J9M9($yZWjCuT?Qp)B4@A|Gic;RGQjD-IOk~WKZ0)
z*qso#kve-<9PukRuDn7p+DbLH)WhT1;JI;ae)|%JhBQWt9l$APVT9rVB!1eJ0OmDj
zV>yw!A;<AFfLK5@+R|X}v}QF2HjHkr=)Eeltv`7pj@8yc8eh^xmWCs}VqR+|Iln2M
zb$FVkuna6zT7bOQ!2d^hWZ=4$=5rWs@ns>pGTN)0*)JrSTkTK3{mX3@gWHEn4Kh(e
zb7|tr&)+|>1N(<CUVk$F#+EgarkTiwZSaH;foM5-x@!xGaEdLoQ+(r^I0z)YXr>gs
zxTwFD?3GQQn5;&&k$T0$*g`3p{N0jB1~>_#FQ{8fo^`5Q&wTZm{o5fya=9T7RY<7|
zI%H4%Ky#M)oRxTrbc!qStJ7AY7-I5JIp}^?_Q%uSWII%&!6L30);%6oY~`nY^GuIB
zZ7U?)CikjmKsmCgSyBIn!p-_w^#|am)}LEjY)Au{yA|UjX{nX4_@k4)i^@*-eKyUv
zrL%I(04d-8haWGW{#^i_Dj*X!zqFXYc-YY0J@8ewIMvsZnw_({rq4jt=hu31x=A5d
zfP;kWiz%hAdu7T3-4_?3+cHgRUj_AIYQ5>zh=fWPo$I^15lc&a%BAB?!)+{`VH@z{
zP)#yw9Ub6KFc(W1I2`GN&e<7eUM>HMNX^1&QPoOR^%gby-x65dE5{^e(iKL;8?Fte
zW>$dKQBR2{HLcAC-Kz#|>dGp$&Ynq1-^IWG9dn;|<c56xo^dA0{3O*xP1)q}udd%M
z`||3BUEU;1;i8+7OY_r|<k#h6)S&HhKT%^x$4{g|y!LgjtAWGWe0bQ<t=-EfxH$Fx
z3A-uYEpl24+;(y{TD1J-Fue;dLE&uk<fz4m6_SdH)ik#HT-b6g1^rvF4>A-Mzw<=f
z<>0y7$nm+ldazvbyDSa3Y0maspqug@Js;EfsLVxk^<sUb2z&aEHM$DpbD2Wtq(k`)
zq(Olh&&P;I<tHFqo=m{?N253DlE1UQ@vPnD;l@@sZj<EJ>m0<{;-L<E)WofFcCp<2
z9aCF96%D#B(3wx|s;ta2OywS0M`1<`AN^Yz{WfcAJg`oe!@KaM>uhxUgKVoajy^+U
zTkTmiw1{2M$-TV!=gdBT75r9s)<j%}f$t>jAGUkShxT^uYW=^{1x8Y{H_2Lbj?UiY
z5LFF8I4nvVcOM6qfOkV@@-(zqJ?`~1&PJ(cxIewy9Ay!q8IuxYD{JbtV-_M#*2WrP
zs#I*?+5X(Swi{I0h?w6c63!)6;LPz(52~HG8I%AhFTWWAVW2hkn^z<`ikatH(}z(i
zfcM=@Nt9a`<#2ipdgL`IHX;hkc<dLR=0yr;|Cl)i#;BD>67~615{0-TL)D4p*L_3c
z(?)J6l?nSO(JguXa)mR;M`EBs!JntaONH_qh!X`lo>}V)<xOxv(dk?vKpbuGcv*sG
zi}whCPr;|^;`OCESB@G)X5uXfTpp0@lldA3LeGg?-|zXy)pJ(qBBZzfEEJ+^&NHCN
zVm=>YNbi&9NY*Ec&^SzxVD!7V;6Gixb*X<mp<maJBMohuu?Vybi;cy9>iz%KEN%=|
zXP-}Hc+)2L3nH(uhEMKb5E?N)r{s>PLKJaJgVbE`aPjyvny~NR()8E#68o`HV@Yh%
z&P#1wK%IuFAH7PAFb~xWbLoGYT;WLo?%|(FL#0o)Ra|t@%9LWd_d?0cu0UR%)`vK~
z5Yeyi<kPG-6x0@uuFBGhILHTScCjl4ES|4^=J<ZH7lmM_M?GaDM*jtWux%4;iwxRX
z?D_~pTI$*rNE8Bg-?d(ExdddI)G8IRU`00I?0ae{G;P#Pr=0L@FNX|x2MH3x$&9*A
zg?er%A0!?}mLGyc#aSB=%l#tOfs4tkub66#zd^+5)~jBa6A@Gh{dn~QS^ShVd0$oc
zT+yfuT7LP#NLN1J-852_QQwlL`Q-to{CDj@4|P!+m?q-%l2EdQv)Cu3aO8&Wkx8M^
z|4>NfPK5fzu;-e(D&+C(_(A%q)q5)0mBzf_WQK>+){ITzBrx%n_Q;m-nG`M=wvKRp
zQ9_%Y+apC2j+gRXWAD7tFn<Y4xk1(Jil;IQ1ZxT`v{Q_C!+Ud2&}D*4%`c(syLl+P
zJreA|<No<;MDj&Dq5T3!$u!+ljOFB|JSC=_l*AGp+>hN&S#XZ_ufmbZn@>ABvjTma
z&as$jjRUwMQ2)|*HnAP@u80_<0qj_0b-F%=uKT*FURXnKDs=Hb>)k~S>epKk$5vb~
zto2{+djie7a9=Tby>L1&=-KbvH->sX9NZguLvshVc=PmONHF!w?j|Z$O{T;thvuul
zwl1d^ruVe>L3tcRDm3iG>Pzu9RsV;nvkGV{+S+YPaavr8yIXOW;_mJgcXuuB?k>UI
zibJtbEI1S|PO#uEH|M|S%0piF)7~p<&N;p@H^%!XI=V*ePImA+9(M>|=9b@m-nO@%
zgUpmsWdMNG@*94`rR60XFJE4Kx7TA2_X1OoXsiP8mIfS*bi?xD9LS=6qhr9XVcn8H
z+nHU)rX5Qu;_$`~X?0I#Ti^DFz^P;Z<kM4z6~t-M^bGL8i4Tf6-A*yAo{bbE+s%3d
z(y;d2Fz6b&9)b)}(MUPI)?Ri_2W61HY)AFvk*Oax6%1@oH}k3HsDSww!8VKr&JjAm
z^=(Q?#u3^=mH+Y(EU&AYlTw(63;<j5u<t2kji<Lb{a5?T_&4f+#(x^tE+uT~x2Afs
zDwYxakWB?x{Azpid={ITiGnu5t-ZU<zB|HyL_s(A?`HOF=5_)362pLM#qLGPaqRrB
zdnTdwf<HI|;}#+)rlSB7w&DUQ?mY1JwGB-kHnH(j!Y%f4Ii5#eYBZ(v<2K`hhX+%5
zkC0clpeZ-X6<3R-0PIEh-zS)&8s1S=NEjsaFA4P<U^RTF;J&MUR(65{w!(s|C&q^B
zc}V|AYgWtBq737=In`sn>lHo*C9UzC@M^rC+afK5CEGj>Mb2VN*8U_EA3d8gdO2i?
zN`q7xJ+&Dv=PF_sx0Z<~%ScB?nrO<(+<YGSq1l_sl1(3YBvsV*(<CN7`Z~`@LLTFC
zI`%Y<I8ee-S&Ao&KB2z5zF`S3d5i$b7RP-)Em3j4dZmU!X;vUD9QjIbE3~(!1WL$O
zs<>Ho0Lq3f6-?Or`6X%_`=y@cPt?{qL}<w<*ItjH>AWbZ<{EBMC_ouwtqR4oJDGW0
z8^)29n7S!>6HW?Svk@Y*FPJB}Nq8PSvCYvW-r|o`7SZ}X3$uw<AM*{jtdkWjd98Zt
zaW-9|!~dzRT>@kM>_77t7*PgqhHeqwa0ly3u)}0C4DS9uUXF6^IsJ!3Zd2yacd>6Q
zt}~(B|Ni{fb<j6RRKNyf?5`J&jDSXRglFR8Lt07Rb~UoTO=E%#q8hceS9k4gmw7JX
zO&9%PDaRZYbv$K?%ubwC3~sw)?;oE2$4qN4_^<kP8UzgqkFI+l>2ZTfmHKX38T`^V
z%9tPZ#@z}8DCb+#9fAA&B63}m<MefZ=RfJ%(bw?#2p2o}@@B+Z^cXdeZqA~i25npg
zTVeg52_8emIY(jd6hv5fAcG*;`*Zx@97aO#F|5l&($Pq6<w9uU$;OBn(Zq)p-&F0b
z4yiXJEUkay>9T28j9U+iORFKf2qG44^A#u6)l~IG8xS=uEO$0ZZ4L_R8H<DTdaQyg
zvclwQ7A2`l&)n+y{g)SPb$oZzM)<3_WX**BN0HaUrr!2Q4|p45tyLV0HRlN?#;_-z
zu#MOz?i>2BOK&+M>6<QM;E;NC^Hq^PeYO&_uc2w)2!Fp>mcr=GGWXBMnq_|Hs5&EY
zp^qnl;TQJKc>jSk<*I|#Q4&7EP?KfMXoUpI&&uN@Uv<l9M;>Oxy-<*}+Y^tBi^R^T
z2hc^8WDTs+bTO=KbW62wP`Kicr?=`m?9kC9sLy#KSTT=l=rC8gZx4(t`H=lW2Q)Rr
zD9+(pXeL{5#4S-lue@21^q3>hXM@#EGiaAYYxS<j(s6*M|HX;<mcVPO>Am;XNoKy4
zZ{DWerfn5qlRv2TH>YCKcSqff6Sy?XPe_J*9zB}+F$ZpOoHjlF)&rF^?%V3mt6fr4
zo6?nPhT&~Wi97!4`SSB?nleP@Hr!ds2DB*(AQ<=FLU)>Hxd;1EhtU3Fgf%Gt?3TOb
zI)EGAFopNh#+kN<oEv|{K8&kOIh!9cgS1Omlj16DHe(YSlLS2t99%TCNUW-ur9zux
z@+Wv6R~z2od7S!Xh(=|GFzcH<j|;xKX)o8X#CPe(2#2SccW>M+4xol<=+KX`&p0Fw
zwIEhkmoT}y&nFMm9o<{ll(fm)+*=j@f2rn+UJmEFMnWxKs_3zm)QmQm^LH>>XW)tC
zl2Ko{d@W>(7WZ^sj`{b(M69V#MPJ-#<xt7QnJYzAlmKDRW4cp`ai1g~Y2=1O^bEnz
z0duSb$QJoos|!dwiXJ1mGV8Ten#fasz3(+;Ya99JVvapGZQ=B+MYP+@`S}&=d;B|J
zVc@kq8SHpfBc)OAqvFKZrhnT=4SMKo5yx#>HIj2{qR2DQY5E$aEGC25%%CbZi~|*X
ztP2!;_Q{wBFaDFuz@6Rc_~@)w(LX%i`$ol1%Hn;@vIdv9x<7Bf@9~3QSLmFz(Ofwq
zaRf_0!J_<>S9iblIezOYQ!8QWVt3@(zReYst$C`?O~5ma@T&RNoiAUL=NWX?@Sd1^
z=oEO)$aqL`I~-uZez8=?VDdcdq`zh0KMh<u_EMTy+~*--JvY_dV?31TZ+GqK=?mDs
zE)00P9={xjqnJlTY2uR6o^8%a;BvR*u)HTJE_bVD^pk)wuL+ORaxp{+a;REuTj`92
z*XT!XM2nz~iQ38PwPRIrfRK-4XmE7(vPdQu*Q78^5AZ1@KV}%3GHXdnW@l!}sUP6m
z)J&O(|4@+B#67AmNe6uP6=FFvq_n;5cpQ{p_us}@JwtUVvaHIlesWw2N1He+Akb!f
zXb_{HzQ=jEKCaBIV7_oG`}T$req>Y@a35VgchvkDxNDWqM=$yNmYGDD_};$Scv{H~
z%@Kjr$92Xdhl?o`Il_zNThDW_yF;)71*uIk33@;O`7O5zJzTqlRZI0+W{e1cD&fZx
zt09eQp|o-Xna4q_el?8xLE1_j_z`J7_0ybIjC%@Qf@~>vG*#gguYgebyv;0eEruj=
zT-hJj>Jka0BloVJhpQ9+eY>r_Hl9Awnt3D(7KVSKnLs}zC(Yd=+rwU^-%E$3e+90@
z5L=xAvN7104xb66bwjvSu<nj$I69f+kE^XPuCh=zj6$!6dczKkgi!ty-Wd8LY<2en
zUJd(lNcM%Hbs*Y(giF;`mp)DY>!!#qsPKipa7C9^&QnFqgjnL|ZW*pf(O{%|=mL&s
zU)62OU?z&CQx!IQOp{u&XM0qhsawGK*zWV>0ajS(<wex%GwD2(e2!t$N^)PTlu=EG
zSH~BAIzpArY4qDC5R=MkrkY@PPRP}6?`*}pUV9(!2UZAQk2vd#gZVhqt<>ZMgZYw>
zeR(#~WMudjLyPKwI(Scy^p(%DG5_k;kWYwwz<v(=C{j<{iWbT02X1+6bYS`QTy&lq
zxjrqJailztN)-Ez@E0O>!9sN$U`1;!^{<=+j?a12Q?6H`SeRzBE%@c>WT{`t7qDF7
zi~bUd_x+h?Hd80~)TzN&Rq~Y{E>}SNx9;f+Y-VEgMgAvFmCnv~s#REVLIN}GzNfb8
zxS$4(8~EOcq9xli0X27id*eRf)14-ZaqqzS{s|q}B{Xn_<ug4KpNAtxjf-G%Io2E=
zU*&33Y~vCqDCASChX$}X+^Xb?m|E<azOL%~LoAvnIZk(!BN9gSo<v=u@FPEI;&qag
zXOTQVO@;hO7bZaIn-G!9#cpA<r$^@9A|oPBB<I<n32WY$kBV#ybAMShRhQk0SQ?Y$
z$kPsOP8^IPln8|c25_*N?ISQ@c=*lSJ;j(rk-n?y?hL0`xLLN~r;dd+Gg!B&YY>rZ
zLUUQ>K2n&Y>m8oqARvj8|78s{B?+w|x2~kx1{$N`62=#2Q*%sKi}&Pq2%W8KzK;Qu
z5hA%foU|jG;LV6Fz%v9AM_YDrlG-?3lP*9C=WU~?pcu#dzv4K53JTY)-UMAurwf(+
z-rR|)KY%IXn4`-P*1ts>$P;BFJ$g9sJq!rg8*c1A@dU!npZF;xq@{i2K$k4mFrX7<
z?)j0CtIe9D!y}*y#_LUGpiB9>xMs#LM8Oks>8&7~@BH)-rPMzPOrOy38F_q+N(Dnz
z1X!r3ce7pQGZ5!MU%5QMu<W?A_aswcQtO)ke$CduaOHrXnZnN&Ta<GlZkQ7=J;#G_
z=-%}V^tL@iKPqfP)vn97anA;+fT?aVNnID69`S4$bN^}pc|nD3pva_&_0Ylmq3cgR
z7DuTNnB~2Vp&v~@tDXic{u4^hYmz7zUCryuop*7=)SiQ(aG(RKAIP!!xCc+UKJids
z;=s|R7rLv#?-IX3$E<JUd+fv^$U*8qm#AJCMgM)1#oJ}dB2u41cWi>MYEm?H@Y!Sn
zcMiEnk5e`aira}-F)}s!-SQK5H#Yg@4;-+B-hL^eL%yMlovxjiZ*;EAY0X0fZY&CV
zgqYOj08f96*?1I__iz_n=%A=Lez#E*@H9eTv1pd8dp|~Am;4!Kf3r(VdM&z)P!1^L
zU|GSiYo<U>*2>q+>$%6sbTr(*GWo*w^`Usrit-DWsYmHrsEe^mJ`?!fm!<M=ML@VH
zjbUDz$>**=V=;U>;%xb<;F!%N>b!Jb)3KV6F?%tsy{}f#Kk`d!iJC=n$F?kmNNfwx
z>H%)%oukX`T}7wwv*yzYG<~t_EG8DJt-o$Y?{8kml60Y(zWOKB1D=*h?a}8j*5hLL
zOT1(IlA5)26*+FnHpfD`RA%_}H{LB-3DK`kP(e<wxu9|piqxAf@74cRGRV@Zf{Mn;
zmxu8i{$Nz8oC>HQwzhP1Zf}yl(A#xq{(U&LN_Gh=Gy3|&hK)5ieIoVWSW=HRvXSXH
z9eK=dME@AlPlft_s1S?_oDD$uMk-y^pQ{+a#_nAi<3Ct^!bpSF)xDRO;rsX+rqs(Y
zK)xG4|A@|1e5TRih<4vFXc6y{Njn_DWO*ABjQ0`Ev(9Zj9tIDk%#yG{q}7*P12JMm
z@7i^d+{lVJ9cNNr+IU-C{`>YxHK==;`rgQgDMwKMMh#Jy#c-K69j_g0=Z|{MjdbY&
zWHEG6OONy&1gRtHzgr7x^C`Gl(gq314v>{fqA=}hP3M!KX(z;9QZTP<=)MNjd2@~A
zP`w^Jy7%Rf^u5j5PSocXz8&xQdW#jvT(dy3hOrpy*lRSXRC^vAl*?^LS8UI1xs3s>
z9-6gzu@sNWLP|R$LNz<eN=~M2G;Dl>puf#n+}YbRfxVhJr~aSQEg!aps;4mOZasFC
z2=QbQ1H|z(fAnz-q$(K_x7lF#e<m`RgKU;*D5kpQ4U8dX7M4yR4e9~fKCbTPLum*=
z9MmJR#D7)$fjoQK(%O!$8k3P|+1NW=%PDjK$O3)G2a{2{ulWfMOBLj8sBvv9LYloE
zHah#Um94c-LM8FOG_b~9E9ECH@M-AmF8viXsfjLG{nJSusT6z*++>k<3`40K7Vi9s
zr&Lr+&W7c=p2YY=N}k|AONB4eBPm~*@8M1P$avH2Y=7+!r^U}wH+d-oaxM4gRQ`bC
z6ATzyYW5ga2EC&@I^a<Z1L<z_y1<00f`Egnwfx-i45aPPA-+onG<V*^wuqxyHP-z3
z3`+PR##%st69%1k44EujRA6`U@%a-^MnO|gESIKV_yXkEeY;~6JEtDqpX|g5^q2$D
zzSF`YGyuVxxN>J=R!CdQ9Y77{-74G2fE6D!%7wfjE%5Sx8QSKjIeImAN7vtH-bzfA
zN#i^eK*qem?;h-XAjm7*W9jHmzzF2XKS%(K=k#w+_w$RGX^ijvFTiOs21SgjsshpT
zJu(??hKy_pRatKnXmPFGMd@P5DIjrnmXG$i;xj24eR+g`2|h`}V~fZe$-EEf?wngT
z$0SdpDh<#0nIh9E2s!2NGuW}IrnbSOxKYwAUyfb_kkc9)bao757<N14^f}SjMSzH4
zpy%eHh(N+2uDqJz7=BZ(js=$#m@$=L<MYwbJ*^|DP}+ARZR-1A*>^G4bgIE^LPQ@P
zpl)Op=ZLjjAa}{v1bS&R1*3@an~yG3dvpA_3kswju+`DWGAj}yaZP)=1OD7roCGd`
zo}iVLr0Z+)7N3Y;xU{x0Nw8Id4LNa`8s5V$azwOmTrP1c<GUMa%45`LEOknSbh1JT
zP9sSAt_rwmkl#DL7Dsz3(r(9TRXR_kOnkv%QAuv5JZQj(Nd22D|K@(PIsqNen2fmB
zRMlZ+il($14Unb_(1OoG5g(IWOtCcatIem>?q?!@h%`2>Rfim<*l0Wf>*A3JwfG@C
zZdYQ_TZTJr8%G07XKRf0kqdyfRLe%r%OO6WaM=a%JjlJ!U+<sXcY*Vk&HiCMuTM5O
zyDe;~V{u9ivFIxJ;whd!V0GgiAfwZa?v~geUI1+2!dbNoXYO-dNN)-<HI6DEih@)z
zeVZWNAijjNt9QWqy9He59r(TrW@&yUYeu}|KvF+uMrP$je{IxvJ=5inNP_`%yfj}R
zq6u*9CA=PwpQ!W_CZi_?(PN5X+0mm*sZcc1Y~Kt?ufHy}`#7Anqi+;6XwWlC=!+h4
z(89_v@nUu4_K3Md$E5jbGW(BifEa&n13P2~y77<_CiJBrDqjzWAo+li&0GDx<GKy2
z6Cfj1bvVmmE9sV%k<!CXrpUw@kC9cQSNO^(>5qIVN#-SP&nBfEcO)RE&&`O5Z(dkU
z6E21G%0gASue<JDR{)md^i5lm1@&QP5)lZgSx&7Cl=*!P^Q09)l`s^YfSyxr_^&Ab
zb{>ip5W-noi-Z%fmhm6<I=v=y`7bwLmFM43w(|wGoj>*;&;+EjGvm%h%8x#S?VU^~
zQtlQfT2rX3aiA`dsvW&GF1PhXjBa41^gkj!fw<Yp79>5p@>DixVjS1aJ=@KOY-R02
zs&9joj416I(+W;@RkAuvtNXVG_h|{a9M%peoLTf7DyF29L*yF}Rc&M{r@qN<sv5ia
zkfMBGiL>hi##{r5)#0VXX7p{=z(}wCl24^G2X1Tl+#zHJ_@h>ekhr7f7_r!iSi|g)
z31z0o73L(J79^BgIklu0IRWK_A~USdqmsS~j@jdvWQ4mNi8)^^J-auYHf6R(EiDhB
z-6pGluUmryX>f(m?|Fu%GHK476}M$+2sf){=s3{D&UvxWS$3$+Uzy73<+BKA*YD;5
zTKc5y+@>pPno%8UY~n%qM<2C~dq#kp7D{OMO`zH1&(`<c80?q1s3)adD7DMYj%Zap
z+d1Vot~QgeSrJX5#c`$ZUmsH6;5_tJd0g&yjWj}jRp)z$vf=32DHQFg&S^WWm77H;
zf*9x<)qJ}taXIk{HU3`)OFroKi}3X-Jmf7u?m6hjkiu+GGzNaNBp--@ftT>=)ISAW
zJgI%3vr1wZAc&MKF^82es9Y@T_fvwkwA(UPuKHT`gTanZfWW0WX#8^QV};e|=jrL}
zNg8a;k$sPmpm&}vz{Z$$88Lc7+xJ3O`&lSEMZzW^WZ@nd#!PINn|b8ky?33L+QNan
ztFzX*?~^nRBlc&VA1`=*K>w~2J`Yi`djm$9;us%2rc`SJa3b;L7PtPg&t<cl_B7&p
zQ0*72&1ud|?-I^ZPNRT<!Z_TN*oJ>El+~6$ggR>rl(<=Lb$R$?p-ib&npyY~S>4Ez
zV<M#ryhhqCy-$T?Pg6t{d?j6pY@1x?H*dv#GDVF^KEchp+APpPn#G7UZwZq;lz^2~
zHK`^ll?Q;EXTeyE;P0=4yAa#;*(-tzvQtEIxK;1Qt~g9UoWHNOWK<d5dh@_trtS%6
zKg)*Ll*psHwPVzSld;=WjE4!?5yrB~d5+wli!Jm@6nM+;rWE3GOIP+39zolMLHQ?k
z4KDAbgJb%T0s92xgfeI0A66F;<9Yk1LC?~F(2sV7zDTpSyY^u&Fx0)2tm4RXIn}bM
z%V(<qH5q(zfEeNEgZ<#jfRK_BDxZ*mB_{#flqIjai$(Voo^iiI|2&PNd{_#U#VK-&
z|HEKV*0hV{%SoHew$`1gyY_cjXX4RW#*wL8+Yx<)UMjcjC>ud_tZ?=W(~Qyf)0%|Y
zU{56_6H3Y7MH>~f7F!*ux=7{%tx~y*_}zS`FIKJJBYza`|5-|X>PB_$lxh!heT5dh
z|0zB+ls2H{8$89h4|utM0Q_cFjdo5E#wyQEeqU4fT^hX;kq7j6bpM{j<B8DfxDKp>
zrPcTilK65bN|9u`Ess(BBRmRwaD<4o!0o=&Z@vGX%}e+x7!CRx+NCU?r!Qj}Kr-s-
z8PM)`PahyF&i9nj2vNYU`yK{@dP3sltco~YSj)JMc53&JX3eBhkvhY9jNhplP(s*)
z0ov&n`9E|&*BkG%07!^pe>N@sfazBV=vf!8CMnD*n!PW$_R~YF52L$7-r8k}gM6$M
zliNY6O>85W4nFhx>njWPS9{ES#|sK~95q`Ee0b+|e0$p@4RqS~NPIu7OW%#pbmX31
zej7K|<-=oc1Hoh|;qrFAUx+*<YJpz23t!&GFYi`EuJ+zv`rov*K^yD3#(YWwu^GAu
z(lyxR%Z^({d`6s<hgKh|&XM43GxWTpX}`H2ASFcZC;HAk*fy=s*%?6#6JYJmLkX>x
zEjDvKz0u(8$Ske$g#Z3YVR@PKU;n{m_Plhrb<R?vsSl)8Lm~8E%9#23-3lLPk(B7o
zbI7>cWqj@Wg34oh1X=;;KTQ^>uYCsGb6k8~$Duys38?F&G!X*DR=+Bbrj=EcsKE3+
zWaC*&H?*_~o6MXfNwCT0c)w)SId{QmKgIOvB>lm5;5sbDsEpEj$5fr8QPPvQSqzT?
zMq~04e&f02r{z`gFb6TGwIb1T(+mC$U^A3R1=-PrTqMg8uLf$~67aw~Fi@P7x-V4m
z(xa0opHXtoh2Lbsnk6d<Vx?TN`ajOo@kH7yq)%l9USM!H_)9k4ZdK(EAV%exKV4*v
zZoLXu<IHvk5H4SDu&X2oEnuLM!=SZ!?U|pdS4*KNDFdPJP_Tg4=HbTu<~GzO<`x(T
zWyPV=S{a#5|9yCY6Js;4Zs`<Y>keu)PASFr^g{JLd`Y%Zl_Gu&7e*l8#Mmw+NJ?C0
z^>pKnlP<i;m+f=oRS&Ll_>rRLZU<7%pWHUw*jpVLZGr8!edD+$M0)jxHEJbZ<S*%W
z^!Hu#J*WXeCw|mrUPS>KV>^aHp-IBF$RZ~fuRZ@Q*P&lbQU8J8;hE$0oYNaJDGABH
zaz0hWHVT08!$M!Vx$k?F-}{4d$E`LuI$<XIKnva+1TjpbL#8Nx%LfJL>#^ZvY4>Q)
z^MLZ>ta9Km>B}s{#rXx%Ff?W88y&Wf=G@s`c3cG&gT>57QNCwW(|63`GC*r!lcW1m
z6Uf>5A?x~@14JwCxc<gONVrz8Z|FP21#59;6>R*V?0tfZkDZ}C%!N_W)HgQ*|INH*
z^(2Vsi+9JSjr91IWF#5SOmk8EJCj2j?Fv=!47KUDe%K8sEV8p2nL}3Ik|5SwHOk{i
z>g|-U>i$eN*_l_gfP)6uuh!Q-CMe$x)rx76NnA&sD_?c$aCk%bZ_aFu5*mxwWL<n}
z5d&8XIB(l}czYWI8jBo7NX(wHg1Q67tEse7WaNOU@@bl~WkRYA9?-e=Wf)y1N^uH*
z$&cEyTDdSR0MyOU8sOnT3{xZit>!Z0pkcFvl`vEfybNC?T<>C`%&-Y>dUz}2%`5RC
zZ(ic~c)G5?;tuX)Scy3PW*U#T%?hLH<lZl>)AVBSi>Jsl!4F|me=kyeWKJr~Gp6z-
zU7*X`n@~z1Ugoc(Ek~c#@Gg3OGV?tsbc4wBB1RuG=-Lwm>V~|fL0%vHUcg>`Q?p}G
zg$LKY+J|V3v~pPTO5`bP_XGlo)_K(7V63mIC916&D8+EA)=EmsUzLG*@e%9ZXA?Ki
zC$UYH>W<%+mp<+j;2{=giYt!o2=V9Dl=}A#Udq0hnXL>(MStrf>8BRg!>QJ5AdGDi
zzICicO^B>`Mh|vg@^Vk!B$A^Q)I>Ik51{$K<&Eoc7-k<%LP~A_4UC_$Z|bYa1Bo2(
zpwXw;-|*@C!IhXqZ+ij!?T)TsltStIC0T(D^pyKf&#)VT4ua2g)K-yLo3CKMmr{`z
zgqK~BNAVbT1YA{ga#%TEPX<~}IjE><|8m=G&ryN#yd=uH-R~dloMdIk2xqRVxA$*n
zLGv8N>`6Cazx^_mGHu3*Mk&Lyy=A|4Uv#|SLtlFf2bls`VF7(7Y}^F|(^-^l8&*l3
z2UWn|&0dMXz}63{2=&ZnpF2Jrqo)3r`@Qh!`R)T`Dv`z$hPERGn_X<a<f6oA%=NoW
zRg|8er79j4ebLnQ(!Yq2nP#U$rnx8aE`QJD%&%h>nb#){Qz^z#a?pUYP*v5$xhECe
zIcwiMxZM6I@-l|@CpNG7&$lHNt#DdC9Cko}&|&USa%T9}S=S%!KIG^qim)rhyGOx8
zdv7U=&$^n)o+dRMMH_~c3F>KSiZNrIlAQH-xyOs=g0QhUE=8U&8JB8pE?$FNQ|;Oz
z-2>vOYM7&JvUy}>*OgqwGeBSa{8=5sx<17n)?j`Hb3$q5h@0(}BF;+_=bv)wOU6Bt
zKAuu+1*yadNrP~uEm7u3!MbBBexz7mo~U|BZ04kHLa#}Hx(mTKD(Iz*6X3C^_N~B3
zk(VXNb;Z{Ee=>%^jBOs2tv6ZGT39t*fU7%nfVRsvI0DGz2e`z_P;Y)H6_vnNAF}ob
zJv#q^Ca{(x;^mj4O8=UX_*_ZxIj3~#i(%!r#LC^SQfCg(FbTMG?->zR1cJ?maxsc$
z$Q7}kdbxk)mBb^4vvd0w=*BEC5&7b+O8Hn#OT#$))f9^T0?j<34c+FgTvxaqhhVy3
zO3eMZx9gwW+eT`+hm^!Ct1K+KY!Z4(pib8AT(<YiTis*6PC10>OP4q^*urzFHI0t)
zMuM10d;B6vUyi09MuTqru)AQlqbNr9Hw>q->$t?c(YO`ifLivDQ}VRE?cGQy;(r#O
zUlaw$0Qjk-tmM=2o^<IhR`n%+fF|_TFog=$!OHgb+-n20$Pq&0)c0F`!KiP952q3R
z7g7sLq<nrG<C*cX8<#_98VO-*XF#;{kQ-#tZtG|*hZ>a{F=CeUZ*0!3o?J9n$w0^-
zD;07N@_z3*+cFJ+wK|j*YHjDv&W~4B4Raw2(~V!!B(N(YQcS9gIzMOmRqgCDNCBs-
zjq7XXWG|~pAhnB>cz4goEBFuk>1=rmv@)?vv6evnVvd<^Me*N}L%BCEB!9j$bckr^
zJ#!YcVXspB2u)&RK*niM46aw|hK&d3yfYEA67jdYzqI%Ui>5XbbyVf+V%+3W#83iS
z5i3Mx5}TUaOkEhi7lu=KZS!lF#!TDp(&RMDf9KS(y4aD&9^$(mTicuWx!x7RoidJf
zg>mDyyxEop=%2Lmv_pF#wJC9xC_gkRt+|%iNv1m9FF$SiK99cq8{M+t#NWv)UORK%
zzaklG;fN_2)+5$K_A^{M`;w$OcyaTY{>Lw#U(p(8h=k4dps_KwXN1JB$`xP9t#6^U
z22a$kx5cDbokUb46VV0y>3IgfEHC7NPoh%C(18{nw&)cLW0jvme+`uy8u&Kx)HAKM
zK4PxDyc3fpa00CnFKhha<|Nl*=|C?0j<fz>WKO^X5uzG{y!S-29C((=C&F7!lu#8-
z;P9LL121(HhWRss;LD}5P7v54>^7CC823w1dPBxU3m*cS60g7iAJTU)sap123yWEY
z$2EVU@2y4PO(?00FaH;Ed+AO2nblcaua$0B95W`T{%5SGGm_9ZcvYs#tGw6Hfb;d<
zA#ZN#82wg$Lper1UV=-tAW=*|VS=;IOC60#Lr%z_{++^@E`G2PXBgA_1{BGWRXaSd
zQfHFiUi-I^Dfy8%h1b;2D$}3SGO!Ec1}WyRcYHeu8F5}p)h4LR9orc=jOn5{*QI-D
z%lCkq<vwT!y`D^M-9y%meewhSyIgzT6W=R+0#9=Ln_O8|&l-q=Al+Mz1@C)M1w!wK
z&KPC^UJ{<^t;Zv%05LK#Q7a)a5jFP^4OOcFBH7#G`ZDEo$~C=+)xk-tof&OtD_vf*
zc;a_ez{(gB7}V%`-tdO@Hj-|p&eHK-U2VXpjZ@R(G2;KrUFZk+@%Z*dkOHtZ%TFGK
z5U^^QxyfYa><jc4qzv*u^-A+VM%lWOOQbcgr&!yx40H|AZ+Bl}ea!bG`^vp4O%riQ
z8DT|vM=G>4TZ%B*Om7#;f+!8G3W+Lct;}^3fDLtE&ci+XIP{oaq419&8u7d~Y@cYh
zoE3MmjH$7!IIzE*1<~~&JF{q@d!drkWL+)~{^uo?WQj3RY!t;-`1e>bDFq6_$WIHG
zG5^rN>W^AUxNUMh@GVQ60@7ON(%J?H8hY;-u2<GW$Lh#sLx#}PI=PD9Yua=iz71;Q
z6Pz^h8f~M9a)XPlel}GTMhJYUvOcqKxf^$&p|w)jdjD$mB15QcHu>WBE$8$i>pBs_
z=DiK~tA0I<*AyE4kPYWbc%<6BISTW;KX*CYy~j_8q=PE|!<LQwvUOyEL)Y+L=2Uy(
zG!u*^yq3LdWpg;A6_G?FBxK>kx?Af3yacY?{8l+W2pS_uskJ-1-OoK&a*^K~zWkG|
zb@)pZcG&_;M9B3DaosQd4R5B^j%(Dn)z>u-)j$O(OoUgaW4^UR9HBvpJ8$p^Ugvvp
zQC7YxemvJIo&xW#9sQ7d^QuYP-Mbcnx>pZ6J1^lhF#CV6a0p_ksNdw1LkYi*Z{iWT
zWytJzO<aKlqUb*&(=xv(sw3ttNDPv-{o5ObusyAcL&3)wS*94#n~Ik`X@T>X8}Wnr
zJOl(UK4FoP<r7ndktsaK3;m!+)+^erCoB+<GG;uIo%mb6#>>%}fd7-aQ{%5+K+ygK
z88xK$nIv)R>EqC~+c?uF`Xkh*Rs<onYzn~kC~4s2XyiMe@%!rX!&RoRbVVw~&_D73
z(_ho=^QWhESY$ZK1{!77+VesWnan&}IAYs?-~6$M&8XWNv@#l#pfZQR=b#5xa*N5S
zHP)+50g(fzSA=(-t8rF)vCg(Fo8U#<{LG`~8>~`xUNe7L&kx2SBS-($G!yKTi+Tu9
z4bBq8gHP>E%Opq&xL@?!^l3kGOn=i<auF{18XuLSf{yjM$5>#~WSWLUDG_RN<J*3?
zL5wV0vn!^MY*-uM8@1+)=p6dmpnTvn=FC|-fBp}K{D>-7HN7uZR23VV;k{zIo;+Vt
z)5@@MOb!<p5iXe@>vXq?RH`^sCmMgBNs2RsJ@$W_fD~jhV62BA_C;56MfcMDJ*>nB
ze*`2^xhS-+Z3Ax!H8i8^<>J^TX~eJWSCew}4k<sd0Q~Zrbj(7U(UkHp=*0_J_Eva3
zZXswL!<F-v&aVvm4f!1W{FZ_IaAI)w#1=_?ynsAQ1ybT~n_HtjPc+r@s1FRY03+#L
zaw}+>n3UMf`zC9*MSPBvLW#mCaG!g*-tb^IQJ97id3;WQJsC3(&8p{q;L~|u;P10F
z9{k$6)I+Fu&q|>}y?mZ+#a&J}<0}(IiUb4jBk5MqaA@k5MZv!Rak}>3zefmiUqmA?
zkbzzJ%oq&ZnFr!n+bA8?T|e039G&6u?nk;KNwm3+?jMg?gW$?A>o+?<%1gd4H+2xR
zTI(39yz6AALn0}{>XI9rlvVA~C2#;aP8j!(r6Xm*1H8Mu!?^;HeD%nFEE+oeY``DS
z$;F1ipDkz*mzP<}9OCMm?yq3%WaQU<Q2A$(hp(@a;(!#(3NmzF2R<R;y;3w0le+e)
zM=<_|r`W12lw*N=18zN#UJCSI3qZrT>U+ia1W9zV7XP;1(cV}pM$P&j;~SdV+vDZ@
zdY%sOIv(Xf&b;RD1BM2b5)3tbIwRivhod5ai2`VSQ=yTGyut)mHOB82zuFOkjmC=~
z;M%P{C|;+~1a5M^ca3!uh9>>88MOPUhVmJ6Vh*GyiY<BLE9AJ<H??$t`g+UE&CTrt
z+3?~eEl|se`76QOQu%v!RRvqpgtEQhgbw|?p8mGEwWASqLW*~Fv*T4}ipTK6A_(>Z
z8?KDV$zC0yr)H`grDMY`g?JwMNCgU0INs-Y@_O*w2otLB@wI@nC~`~FOx)Z=Q-HM&
zH*8DJzMfJAc$-Ch1fTqTxp&U)g%AEc<befg)l%gABz179m%g^v=LlD8mW{>4%$r6?
zVhH6V|LWp}_3Q37=<OQ4D`QSmi>=<K;m1Qw7vmQzEA?8+Yx?pw?ictLuLdk8l*0C0
zfxjT~a>1T=Y+Kpv+e1O7P(Ji`zlqvjL9*!%UE;5`*}c>PnxIc@#oq<ppEksT;HMq0
zarq5+r2*JrAi)Ximig|sc=H&7$xX`cj*CPxdO^y8J@gK0Hcml{dqSRaXhvX>47E1@
zzVIG)=cJit|J~fy3GIw#J{~Xk4N8*+46i&UQ7l_`M`gwX-TIz~?!lgZJ&!=PS0jXz
z&n;@H_!H}*(PF<=TPebuA}E$Wk78~5-cu?`d3{wbSD=M+@_)kl)y@;(ONuVvU#^tN
zK##E`T`t(iqEf_EHtK>}W^2+T8ueHEXM~?ltQy9vQKH>)eN&TX^z9k(9F=Lolz&)L
ztsu0)r~_878{8J4g3ULQb`L}Ed!XaJo>prAPpW0AK%`6^Pf*@2)b&z?ZqwlKxFN9M
zI#qoE41j0!hWvdRC#7bx9=j-gPLn(sNLQ^i7`oi{bDFn7sm&%Ip<cC##dF#yT9kj#
zep`G;8ZrfL5AYpYsmIZrhEdh=0!?@eCLa>t&qa1b-u?RJyHn@Dtw&Hq!@jnqt#br)
z2TvIKaZZnfRH?)5J**Jgw?QK*RIwd;KlFOH1A&zB<CPxcjo-N(*E{0UaLsUuI44qn
zlQ-U?_Emq=RCGQH@(T@5f;ofD-(=a?ByjY>4BMzhA8b+3m&PtL=nTx=da3Il7kPUP
z`U8H%w`uPFZNl`EJDx4=`$x`4Z>vTnv;5L(;YUZ18N!PapsjcI<u%+X&|=@?{w)=H
zc`+a^;F^~9(w+y@>`^%TH_Gh4p}fBs8}Z$Hz0JSd?>`Kv#oTT^h`jkR9zKze%o#jm
z3hk{}ce%2>vG(`6zVs8Zg5GcYf5{EZ*){|^8nK+)RY4ziwz)d)g3WrQf6VF0-2aHD
z6=e5I!iHTipu@HPPjmY2in&X#Pg6(4BUZz+55V)d7W&nJpMu+t_*v8dq+8O;hgr5H
zzh%k0teA8{vq4$J`6=#CC<Gr48-d@5(Xy3*3Q~uY6<9y0|E{f+^e-Jgp)6-j9pw3+
z#67Ox+c{K|!x0hXvk@sCdFVk8+EHlkraLvl@|dI@2Lb+}nQ1IG<dSoDb^zbP;*Y#M
zHtpmAzV?0ng}m_-%$ZiMrvP=dfFhs{I2=3*<`eYmYbcmkJYHdO8AsYCZl`mEL+ye4
zW-KS00;rUMimbpm@2fN}c~{$!h9v+=BH8h3iu#fn*v<B&L7z|OkCX1-gx~b*+_FSh
zpQA^6X3uB<X|NOtHioM^a}~xrFWcm)k)zvf<ln)9#G#lf7H})G^pEBWUia)=vMp>8
zvln44yVB<B&=CSPl0YW+QC-^ulv+ubU+aDufV{JnZZBLTA!Fp^%g0pjls+f&k-oBQ
zJ++WFc6LfNY?aNQ&79m>x9LmFRm`3laAYg+@sVEtyF?=Of61EVgxX(IEuge=*mS@h
zhe2nvVhT`xmK*m;2whNs{8NV9ZYyfoh;WL{$y5S5FFy2el5iP^NrsGCb-H~|0kp0O
z89_Ued5KbjzTSx?wRGbpRhGAPd$=1k)zx}~3$jG;^40=(3VB`7IkF%$Z{T*EXC-hK
zdfKVcvzcPw|3t97<^OvREhxBgAavwM5-+)oiiDV&prB=-yG(a~bV>%*mHmQh?ZeS)
zJD;HUZ{Vt?=Pxh1mF>*HM|p!@*KmaUzR|bP0QV|zYiW6Y%r_NVaT})Oxna3)tejRu
z_+okF-cj5UV$d8WtYjH|wS@wDTKP`5$P<e$)Tw|IbjcbRnrYle(qlo&P$#Qs_!|+a
z!Wl$SCk~t83_e~GDTw_;@V&cxVCmc<EN;jux|~x_Mui<YVu9>u{*+<g!$?C({5B{O
zGHaVCIBQ)QXS2$JB+3+={$&Ph%TYG7aQ(OOAJ!sH(VdY(2|(0b8epxwrAexa;!5@>
zhp_E}IqwI3+v>42TkcUU9PR|wkDP{kP~E+puDOQhcOq$fd&ikG4`^JlS`KTp&ms+Z
z>iA^h+Y;3Kq=p!=ORoyZ!VBjqP^$wxGt|*fKy#7o`G?kK^o-KxOwyyHIwybqTE+X9
zEl%cW>Nvlg`DkFZUGk985uE&7qBuRUi%b2GNU9b?rxd-zDRDd;Ge6^Af|w$KBs*sr
z;1Jr(R5hMgl7%2m0Wd0*0S}a%X`Im!KHHn%vKW*m&iLL1BuUiahU)XIye@nkDs>Ap
z!p2)*sgodM5clA=^mYDoP?y=rK!HhO#+jeMD-fViz?Wy!)F=FJAJ6#T_>@E6L-_Rd
zl0L*GkbhVsg7+#yD?FBcNsS<qcq)6+b@(;F_>ms>2`ZPA%^?e?VwH<X<-6_~)(Lt?
zRYrXbb$S){`K=fDm?=CxmAfpHGGi(h_9s{4Cy|Ozw@C)2XF14sH?Fi9mz<LSO!##G
zUUePWOijQvcIvw#Lw;o#W2VzZI4dvIk~=D$h_75u)g%^!c*0Z-mf7LaLJ97(iLs7r
zJ31b#yPW1Yf|5Nx|Fdj`&);Z^htq3HoGd9HbvX#^b{h1MbUcgC8<|d`U|)x3$4IsE
z7`0p2H*pu+fG)`#wT6g8t{FRrXnD;yG>c@jy0bKe2gQFxA~{0*ydbWxPh1HJs96gB
zgye>{$7(<9m@Udos~kWjIm2qYKat9Yxe!;mJ`lWb$=PiN-qTXahQ+|vEgA0|#Gias
z-~ML0?U%R-M{TIBy4Zn0|FNu5t%P~lCibkiMnd&;B2|~pkDfYNjSk1BVdD{(4`yH!
zMACmV((t2o`@9zoOoeo&@fZ!H(A0pypCmaL3YE424FJZhdr%YdQw<BH3N&)->(1if
z0^Zl(s}NTL0X3xjkP*pT`<mLbr?MfWbLS1khuM%$f)O3N?6}B0Vp0wg<!meb4Wry%
zMpjH}3|WB@-J@3hy}C52=*sa%VC}U{pPYIcP7Bk08p)S!tG}w<MgAKlB^tT>IY%BN
z0+3$uoN}n10y>YVF!qawLf$D<<gz=GiE^;q?makj6&uI-ymP5X6tE07#TpVd92djD
zZK1|WUutX>9lJ)YIXV3s-++bNHG7D`!R4Fy`Q|ISgFm(u{T9D`Ww^TjwfSN=Y}+hr
zj8%(Is#ti2&gLw6>!BRNCzhZCE>z(j=|IS^fcxMNu^MJ-QG!Y-b<ny0WZQHQ(uxgb
zNt%vKb1e*pi>Ie3z}MZ+kU#Xq|M+$dDrk*FNCe;eL3a;4`kx2sTc7OZ{U(r)mf~G;
zx4l4ZmLin4L2<IORq5BKKJtJ8NkdH<2tg4Zn?;OO-j^eID#;(1)V42eVLw|menFWm
zxo+i_K}Q0K&7C*>>J(4;<)Wm<6Nw<QtnoWzB_~nI4tf*O=W(PA-!rW8emElWIgeWM
z<F)gj=;4pYKLDj-5Z7^adE?Hm{P(!6kJTC|Gh+F4_lkrA-WRG5x>U~^hNo%@N^I-<
z_zM$8+lC|%5}EZ#)ry|nO84+4^7h<OJ&ZKr2kDi>q;(g_O$#5fmenPwH|`(b+~l^%
zPPL-ew0Kcc^l=>+=aoCF{A|rw85d+^ODI}Jl1nE3rwx}J_RKSvN_J-HDz!am7%lMS
z)7z}b(<dcm<=lfl6|C^Ci*8($x=l(%a>3X>g=Cl9jouT0job#Lq0eff+q)qPrMAty
zxQCE@Wu<c*Pxu|j)2rUa)fec_jj}G4P@bQp!|7!N8h!uFckmxeqrO-O^Q0=mmtX^G
zB%22L)svFBulRm6kNng+yEY0&CNVJj*c}wo*!U$`>Qgk>^K2vjQmT7jOH~zHZnAil
zj_#jso<GfTQpfDhd=&<KuT$o;HLGkb`~0`FdM*v>nAXHDE3f6$v>q`3q*QCfqp7Ki
z4Yzxlm71y!iWH|wrPs|sIQ%A<aiJQvwY;9muST2hrKedmg{YeS#G#cJYr?divZ<mf
z(Y;u_cs}Fk6}7uSSVjHX#m+vpg-7Ibj$h|i=D+9C`5?An^LN6}z~@?1hTPp9tpSA?
z7FGiqT!i#03kzPFRbx|-y4~lKCQp~=6_B`jdQQG)_P!^yF;84SH%=o9<oQoe-S2B_
zLNw3o>p`Zz<&i|Fd>mBN5yf;95us#+5%!<hnTEcZyqkWF`0}yUVu;;(sF7R^kVU0V
zK|j?PDMW19RZ-L3)xA?wojEjhSkN<%{APCMCVm_DCx^vz8hZ<dWpmT1&~KVgpmxjY
zHE=69(y~+<keem<BknMow!XQ!E6lBY>051RbqaL-V5k1%+4tf0+8TZoA~L4cN-LWX
zSxmi~(#t5)-b?Ym+q12GzlOJ@ZZFaW1w|iZN~BosJs5Q_f0nRHLp972z{v42jm>Ih
zL+o>touq?^W(KrWY8DhlPU#p3zs~fz1LLh2h)mnFTk&TT1tnr3#bROI!}Xotyu1|~
zT#sJ9P~uICsFZ<LtGP|JD%c{wNB&_nSTKKGv9s^L<UUvuo;!I0;-W_h43hxc&)HYk
zN_$<R5o0?a8Ajr@y&X_z=P;dmd}2?Od&78QLZJtf;dEyI3y###$M+GD)zS?I4qhU1
z6}mb)#oyVy2u95QC%8I!Mc`e#;+Z1(>o=zZsBzk;ePY39ym}{HyMLn74e#OD>6QC^
zweVdtMPl3FH3`L~ImW=mAWV$my90Lgz&xZOiIV{8>1q_Ny-E`-h*K761i~uTGZm_7
z!ijOC({b}mUymsVW`NCy-T6xx7*)tSd?3@c*a@_c>O5d&{ulqi(|dYl-O{}0zt>hq
zUN^5ipwsT$_QdvU(znfQv-sIHt2+cQCt{}tr@-yWrUw80Owj3jZ~wEG(xw}yu$w)*
zz*-93)Yy1uvlXi^hfu8YgYx^^me6M4IrzjXq4!~K6m(h`T=ISRtOcT*|Anq}d-n1d
zo7I-vyUG3J+aj%z;HKw6ZJ4ZXfnnK-`tq+oEn8mCXhO9p<#TE+7e$y9wPLASv73)>
z!|^nw`u8D~7bY<x;Rx=MKM7H*;6IHI1!%C9vcc)%Ogi>YnLJ}imzf)cGsxC#6jxCY
z*kILB76jbEK}R%_x*@@jOae4Yus%P3{}uRNx2T3y@n$HvJ6e@}-$n{9sXO&~I3iEI
z{0b4;@XWlu)m)uA8J2Z72(QUyD87;oIMWQf5Jp6lmz=YvEGuDPsMZS<KdxL-5x8{s
z0GU>Q=qd!2^RilVuC?khFKO|7nr|GZYuR=QA0O{JyX%v0^<Kgwdn<JJwmL>Gt3Ldd
zG((SHTVTMT8+1qD2Q3qOIuox<tKB@N6ktl&_tyM~HoB)*Q>QUkQpVP{Od)<-dn!O8
z1vmF~>S+BuCwn%cxbHOn$LfV<PEL^nvmgEZp3@eYTD!|W?*%_S3y@y;S@qQ=jefjD
zBLL~;#HUfP@)3OTIqda1Y89Yz5$m4+;bykan|xa#OUFg1{ztU-_r&iUkC~GO^fG!$
zBQk&Qc)VDjPRD6Kmy{?zY3@Q5O+G#8VVCiPB{+X#Kln&A=++J9(%sJ2z4QA1(W>8X
zrr!!kN3Rnt1>Z%}`R%@_*-~tLXs}%m7mY!rrdCZ!Vrg)B>pbb;{1D&ktg+2R9!7ih
z+kAZQE`WllcRQe@y?^M*J$Ra<8Jr~}zv*}A5r#t~xJo@SnkNvHtQ>TA_^@dFDqmac
z=nNi*Tiu{pcyk}c>Ow1XyX;Fpe`aG5=?nSwx7oxkAe?S44rms5Q3|;}n*F%-you(*
zB|p~vWXXYR+~1Ii>>jp^afD54mV{B~iJ*?}*zc8kZ{!yW{)Y~kWOD>A)J%@RwSd?Q
zKL6e99Gx}!!;~A6WYFQ9o+)z6*Ck(rEg8<n1^1>=<(<6E`jww_kH%Y&ByoNy`!{xt
z$6TU*Cc0h%zv65>_J+c2JZ$lzY=q`Cu&=s^>GF}Na60J`J6oBoP>g#Bhi@q@t9mdE
z=NJ4(`6_69&7EIJuMT(0GL<eS@O0(2BBed`Y{VQf)^_4~^zv9AalF}_KNq^cDimeF
zT+X68$HQ;V7iC~`nZUhRtQ}o7qRT>BC|g@|N+X~USMzlMDsHo*o1oBOP@KtE=ciC4
z99D&CYg;M__YbMNtmxYsdr*PtX`;g^)(p`>#6gb6E{-7>VT*r5CR7%Qwx=dX-(kD$
z)DHGUWy|I0b3-$<aKCp<T7MD7lZ+-tYq9`690k8_NMm2emA+D1wy-hQ*!e%r8}C7#
zH`6YU$La<>{y&ZSBULc<X|-Yjj!i8xH@H9F_RJUR9_F{EbSbEHjgFIsF%TjV)_Vd{
z?;SS|5E}ZV3ih`P+PKyhySjFe|NOH$@d(pKQUW64%v=AejEO;Lhg@A8IM9TCEN4pi
z^c1qt+#w9oM}-#<Tn+qv{zj4t#;(G)J2~`z3x_(M0t-qw2g>$%K^?mcGd~=SHC%-A
z0Ou}(?{vUxG#X(8ts~1w8HDvQBo0NQ7MY|J(4u^;!ItUM!G<&CLs2*2BQ0z)GoL@?
zONu}yIzmI!a2W0K+PW|2L40o>UcJzNZl0j<>s*cWqLtilvS`ps%J^l(*pfHrqLefs
zQQ*>Vz1M@)#8Rg<`uH{;eH80K?4Og%rE<ewV!wVqIZy5v7DN8&Pz1D2KL~fI)&0W&
zn$hl)dM5Z@V0$=L&s@!0bZHWx2PrSlL$QrTGimFeG*ChpOPW4BMMKrqH5%XxMbA?y
zt25&+bh$q#URn}#NrCP&G)ZxAFfw;~CIPN;-~>zR1~%F9m0P_F*X54xUOwRAUnAT{
zXFlHsbn4tMPIstD*u>bigN^v5=Nveu7#Zx)%f)yTufDf1FaU@B+GP%q&gusRhI~XM
z?Zg3;faNH5Gsr9@e=nI`356o1L$Svoma^cK8gz~AR`2(WEoqV(2y3G)?MK50e(+`S
zlUK@7e!-~_BD5%_l8syX_$W*S^*#DQRB_?(Sn^6Y9A}Nsp#{V;Y641HF`j9H+mea~
z&L*Ly4+e~Kop?0`aflPK+jz4HIfQN6k{9v~w}j5Ls&LlvzUf!>r68tlTs#;2Pmv`C
z{SGvyl+}4aOrd*JiCD##))+(zcG|H8u~K<!#piKJU$ndIUO&=t$-(2~LGoV2#lL<$
zDv%cv_I_#OOu<e-M`uvfslAGOu7)Krw7b2-d!N0u_hX$@WOb|H^XX{Tv}@qa(|yO~
zo&MzEBYjEm!>b6-+obv1GsthV6abjFN#@3c=9D7;XJR?BdWGx-SwF0@%RKP|k_LI%
zss-aAA5t0d_k=P{<0qEWQQQ2|i|UFO#!t+`<CD#y07zG(XW^C1U9aR_dVgNHv!rMx
z$jr#l)zhoP2Q7~EIG#cTGJ|wxa-cVqQhewQMNRwcj3U#Qj`%~{+d_0W0o~C6SFXY1
z`o#L*%~4VdCSHT4U@=h&;)xc7VXuC92Dv;fgYF(Fwrq7Ub;0w(X!aDX|89FwEbIGv
z;lJ7ZNV=+Tu78|ffXfe^tO41x&cka?{ht0}l}@3~91%$G=IoUt<R_ZbHj{tAs%dcX
zQX!*mmpgo@XsAB-z3n<@4cDcly#twjH~2l3PH&Q$KTQ8I?_Kto8IIUByapdX<G5gB
zNg|zjM&j{k@evVSj5GF+YWH2?5X6UR)x>M02u}0y6PUzF{=(@s!V5FuBrvTbpO%Xm
zOgV(pgN2_Cx$En|&%16D)|S*{Oco(Y97P}=@s?Pyq0}6Q$>{qUJFGN{A=~iDyICq^
z7I|ASzV2RTd$wN8O1$+m0HLQ$T(X`>9QI3brM)Eg7lkFki6}#;--irP-o79$Gcpv3
zmL$jWR9^Aq)E`tNBK&41G;txICreRxyoVlBV!|u<qiWegIJYz%d6B>)CaB9B(xrD;
zi&%8|gnzS%1uG0Ymt!WFronz>u;^s0Y4MHi9xIJyM)ec@?ylR($@EtS532xL<=q0M
zn4q^&2KAsUwyi$t(tt0$&MP_+<Ezbbvf*N6`po}}sJDu0gX^MpUtEhrakn5Xt^ta>
z6STNnaR}1V;u_qoxNC5CcM8R|SaE_|Prf7nxy%^3%g)++%{A9Er@x+(QDe)ccWK(K
zX|1UO)ETZmt0=D)W-)eu8qS=U;1u6fUOsP5zyG~Lp|<?(2Cv$pQxIYAS9)ES$LP|5
z@RRbr9p5=};&&rtkm+7E(RL6Ng;GyZ+WKxJ_}PNUIx;@-ON%t6y^6Jv(6Y4<<@I?-
zOB?u@QY5Ob>36S`MtXZJc}?7)@WA=G!2QJrLaz>_c$468I?s|%#KP5caCjK?^^HI=
zkv7CkJkR5BlL{_RKz)a<l@wP=>8#BtW)6j)(U3&mDKbFZ&Wl!C6jtA-{H)+1wTvUo
z%d-MP4r!*qZ$A>-@4K(E0xT?J^;rf*#J?ToRAwmBBal4}bYZ<0x<6L0do}6_E+^h;
zLd)po%?k*E>x%#_=ULVRUrAnj_nN|1K8Yf@4`6qs#_jNb2T!nj208;A_@CeCMf-Qz
zWHNjr3=ilf&h<SYYxco$^WA7iC~_@h#pw@OGS43Q-JKLfcr|*Ddl7WL;0nzDecaI0
zEXhHChbBtIe0WCT1Aa{Um>{49lq3Ry4MhL7_vV%dHbPq8ZlDTu09bb64yh$FtP@V=
z*~H;HJI+??@R{_?m?`ai8qix_*^oRTQUClk433zo_P?F>y>^;hTD4UyEVG&8WPUOW
z{9i6yP*gV*VfHBrlJ##z;wOIi&xN;smbiZs4*WBn=PxsXe<#1Ki}dMtHM)_|EY&^R
z-kCaNn@xRt2-cgSTx)&$lkr+zUA=d<a$0qlO=R0s>ruLFccJ_r9Dp``TFY?{X$ib}
zh34(7?qk{B9oJ+#{r>N9Z-y@?d!;AvpG>feh*&g2z<$S;Fx2!93vRW4-kY)Z$F7Se
zmP5g;yzCxe%S(~v^YfPN0@-egb+#bCn+Lv>v}=+G{ka#u;hb=f<Z1-|0K*?o729&f
zz$R6n;t4+q`befo?CR<!IhlDy0DEW)7f&vd<0#w#(%~=TCRJ)mmm-A_%f}RHS-@IQ
zp%O2Ifh*#gxWq|)^1YmpWd<x}ghdx|(zD>?Wyeo3n7}-rXu~VkoF%d>(VI*>@k(tN
z$~MJ)AouHn6%hvzH2{RhJ6;JOXRbI&rP>xLN)@gYRY@*@Ks=<_tP??UP-!BjK~1Hi
z0$Ocz>`hR51+5l=HuPVW*T|Sr*|cv38}Q!}_*DJdnD4{*#hl?^!YJOTP?_M`>ROqp
z@f)ex4%giM(pvZAo?wbN1a$)Cr&kia+|AvwlM}`dqUWP63(4%t7zW%7x`-_elbf6P
z&(ETKB7h}5Pd&Dnu{pzrhKBKZtvuG1VQw)sa{LdC(p{e(zB^aSdaC(d^<1DMh?uQd
zkh`gXboKSPvnlJL{=Hzn;+}O0H4CJP#XKDcnFj$SfUfRusg5P#>w<z6yQF=T;6eoG
zN=U_N@O##Bw`uMS6(OKVFX@Dx5W|F0d)NM|NSx&hNMX4`TQQ(@D;voXSmLe&y3D&h
z$JL(=_<kjn8_2i!h^F-S{M#*sle2L?>^eF^$m(t!MM8|up#*{_Da0GO{PGK?^DG(@
z=!*1_E%k%f<7Ar2!8OzftQq7*dKi)1Kgg~83fE1;0}gK~fcJ3sb}W1Cf~m=F-?_Na
z$=CA@aF%UuKol!JCAOSOiAPL13CzEhI(pHWbNx6j&&QPF_}UUH0O3f%ZU3+J>ePoU
zib|^m1a;G%>8UkAKoprk3*|XYgVzosb)+jQi*c*(e{muvlgdo?Eo<<s2u@Bo_e+kF
zpNx{K0Sb%%QYxK0WnZKk^3#N$de;6PywjGZ!FCW?sqS$bX*&j^tmZe!b*sBa=WN(p
zoa9zkLKXR{-?u34bL$fbf34$+pk8z<U#z3jYqjb~>kvw-aW{tO%2YE;DXtsl3F&^q
z5Vy529(q(}Z)|^;x`tVdJRe%3`*Ch?DqXW=6qD6)YqDaJjEV|>NU;oWun!NYN?*WQ
z;dSz-Ktq=$GkSa@&T{HSvEAn-qmY#zLQfux8fz0RN02a_sUQn>DIX8Ia*ZHB)CLJr
z>=pM;%tDa8)^QTG!tSc>y{sqDOpQ+9p2;0oJVxc@f(j*)@hiA;1e~c`A{+sn>8A17
zquzZnK*WiZl%W$s|HUnR!B~!!LSbusruvwq_t)FhHzj3?bGPz+uv#~|P)R18>w1*O
zCc~<#ZmvJ3U}pTr2Rp@qVcj^Go}kdbKd8#=szp9tknPB%<R+8ofv)VF;}AL9(AZdF
z9GtLIn$zG527JCXDwPxzCvD>tO&(-h00j21TKtEoxuM44`q|UI7!LZcsjIJ7DqUoV
zq+P=Duv7%uzyZZLKF!DXA_x+P>V|c1AqF0wi_h=c?db3~wRKSed~2e(!sL|O*-R4$
zBfGsIP<>C2Iw<Ald4exX)h|b2;|rRtn5==>=h@Pvkc#sHDGIqV8H1MA0~VpWJkS4g
zorrkx?KcHJ<iS+_Dhle3Wuv^O8;tCZn#J77=fKQV7R!CvH{H%D{F=MzbQw7#p$aA3
za|$DB_Z~-aQTZw5(*d;HnO=)~=yM4OF0D1*@A7d;0L|L->z{DC-RJnUUEg@Aq|BY%
z865iv14%7nWXl)W=Ki*7Svo31T>A#1PKvTA1`#EmSJYwsSTk4WYq&@E{$SDk4F`|c
zF~`RFh{x95xk}SRtqb<_O;qptPd8rNhPnOp=ym-l<tT@j&Gh%Yn*p!ke^x>Y(&v1a
z3)jwezk)Bf{4d0oboItccy)CtSPi&=euF-yzD!j&IJgmvFXgo_FH*O#2Uwn+UIs7I
z&4^6o$Q~~5_+0wchJP}Q%)x`WturLTLWe*HCi}Ja<N5QQ()*Pt1uf~uD*eBL$DD!+
z3qm?e4%j%;`1X;}{5k#gL3FTqp)#kTMpih=f-y33k2Hf346x(ZTGG=abD(my!CJ@s
zMfj6tVq&}Uh}Nu=q$qNGt}K~eRePQ<f8%`KzD3bekw9^ml_{2jF`<O?-Pl*zHbN^S
z8C$XhK3XNo-+v<7m1s{q@HXc)58y9)D<C<SkcgQ;9uIXKylBta4O;P7`1=bodJQi)
zHR=DIauZ;gv4OYo$;n0a!yV{{<ltI}n`d;<xu^0SHvS$9DeO6}_21c+bu2;8s{A1L
z=HfLi&txgqI%psJ3}$q>o2-?_Q5mGHUe)0HCA2YzHWXXLqmLSa*xs<3`Oql!@iZcz
zxN#v}nk_8&Yo@&tIg^60tqAbbHav^@rro(=g@a4wJ!!4TQiuluR~Qyni)><GOa~%8
zmB-_?tiY0B9r0iSgBCR>!+cSwE`c@)_LuHhBYUYps{CDzeB;l?nvTU7Qg9)!su~fA
zTv;FsUW+0M`^2R%ad5~vZv7-z3pji)x9O(jf{V`yCb~X2Rw}a=3sGVViIDST&2U){
z7snQh*bB#I0TX!Z&y!=O!d<8SKXpaXI5NLIFs%hHkRM#<y~&=AlM|DWjqMR`r^p};
zIZ0J+6!~0Hnm$2Ix1*-<5~MNZ`D^W3{Qh)fYN0RfxE(Q;xDuNN6wCzr!bk=l0AccL
zd<!@o9czuE&VEw@E-%}lY2&W(yr*IbMUZvp)oPs3It6lbgf4i;o=^KS=*kmPOJ=iu
z@*Vjnga!@;nI-?S*{NEzHt04uK287z->n=18nCseGB&}%Zv3=<=L?K!wR^iv?)K86
za_BOuE+(u~{uh@^+d1$cinI=&twJs`5EJ-tqvhzXo$_5A98pJc(K&Kb+cWfgzvY3q
z=Y<9QRL*{Sw!S+T>?4U8laXIgrppm$uv~_=E{kX5#qV*ltac;~zP$K$>sZ9E;U}Bj
z_&EJ<{rybm7e`Jp=)c{M5<Z<9mlsKU;5^p@77CWLl@^y==c!FNWgi-7?EwQnuO`_V
zK?lZGyiYgZ`mXM7zB<pb)MUSIuzbjo*qrvJ$rToOuI~A2(*yGed;=TRW;;(oAa`&f
zbknEDxAW^>j*%^7qs!xY@@?tq>B?I(JH^)EEH5Qqc;lOxGY!mKf4~||9{bBW?dq4g
z=9yU)dMiSO#LH83O~(b__@HySys%SfV~xEn=e+*B6EY_|#`ae{T`x&Dlak!GfUrwp
zUKTr%z9PODTNJh^MfhJ-Rnd^ss_3}M$n}lVtq2hP0+wYD`Sh1|L}L+zsA$*4q4K*6
z*I2Zgs9Kj-|K07AB-+QjLGt-lNnU3{VR`RH_2hQ|awfM9#eg01hK2A{WU7y}VhExv
zvUb53akr8U^jR}Pz;#O>pN^$7OsOJ}0XKDAtN*>Yfq*=%KAW=c^H--ua}lz-o)&3>
z5jZTa)6wjVKrM&SeYlku)V>h8p)TT-0lhz<XbmfY_4CS{oX|2RpeslMf&+qY%&rj^
zzd)~kp=8De5trPg5qL66eY(et9?4p@Q2s<DU)9LKEIU01&O=%m)ZvZt9ovyqYT4*h
zEg=FaI@(8`{I1!_8On(2+nM8*l#|UAQoEey&Ydj(3^Llu#j^$O>YILuq8Hg9(IjF3
zdWw((up`wBj$PuXY?m7>Tw$tFy0V_P27}im1uFxUN8k^*foa;W-@>7&Mk%M~Kr5en
zI{5*FCI188A20~V1nYRB|2?$^h!~d!SoD|@Qa#0gU@%<0<|9N&sRbf7RL<@A*2L`i
zYf<=wYdEic^G&+zAa3tbo;Nq}XVLUdWRC6Ax$2jeET^~Kj|ETM_Fahy{Cmpik~<vj
z;|3XV`bH21hk9|_Pr~W%;_Ntlg3oZx3lu^a4UJejJn~H2R`+ResNIj(txX=^;*Odi
zqc9A0lkVZRYS{Wi-wCDD1?*8hFk#?TOhB+1k$PD7=IQGbJG%k`F17gfU3Iu50x)vU
z&`kAbEtX`gIsp9Onl}^hS2?+@cu|QJ6diy`K>^1+`J(M%Ul5b!3AL+V<scov7f^3Q
zO=Uiq%80~@F@1|*v3i}}ab&tO^4YWmQn(*(I@g%^CTu;eO2C&^)&is1^>Vk{4s$JY
zQ(}e*k-0J^@fGNa>)RWmu5hXuF6qFOYpMDaU%rnMoT3l7TOyQr)1WRNTs`Q05xS@|
zzyU|?kh+19aSCh*HP>8s#QH^(TZ%U4`jaZYS4Vj&97%NyF3m`9pjHRc0~8n=i)<Bz
z@F3EMm&06nDy*=eadI6Z^C--np9y^W#s;$T_)|7mG71*4{0hX}dEqk4(h7`SwXa0%
z^wmhC1|x%tQi+T*<MhR%sh6%zaTWNeqEx#95Sp_eoYEZdgNRsPvBi-6@gnHhM<9<o
zBW4s<h1#B`!yG|nYXA={X8IVlr|YQs3=G2zkx-+|mdL4F{IQ;=Q!?F{@ftqh-XiqJ
zYJA7BR$Dd-5@q*g@JQizN>;lTicT{l4;e+Ixo@SkAqKL5*+y02q^(SjO+14eU0ZY^
z>2C$$EiZlZ(I{c4Ng53h9PVyn_IXTOFSnqKEo=b)brCe}l2bT(*l5Ig-8pC=4PAh9
zJIz#wx}I^%apXx63@dwW?ywHusT=6(8@ch3E?nf6E++FCVY*<=w<@-Dgg|r0JclKP
z>3Uwthjtitm1hXV9uXT_AZhq48Nszz0bWeAhSyHhchgAT_bnMJwxn41G)}Nsmv1>=
zd?NSvt?zK-gvcSw8*7LA>YX9bR0YS?GiFGrMs@9m21(Vy(+J0^OTpV0Ea(0_V7Wi#
zj6~+uOR#TGL2vMfoy*0RNL>Npf1zn~I3%A+Dm1)oDxJoz1A0?)5*3sWsFWglId~x!
zd`wH3Lb2r)9F1!9ks67*JriBe;Q=LE!_y3RAH^lRd#1MU#KMB<D({!TPpiDA;_HWf
z)z6(TFC+Gx_bY1~hrD^Oz%!G)+Q8SWm+i$ZOTNsX45|fuhaa;G*POLetS>Ppu{QlL
zNUC+SB(72<q%&{e81G|mZJT~BD##T{l!0u*R6M6^F_U%FmM_(<KkDtVglhLUvRIr3
zOll5EA#J5d!o)GF$wK*iVc<H$jC+jRdbd~FGcGk3X-xW*jlQEa`&f?E!9)AaxjOZq
zpP`m+r6xXp``($TudT07;1~0-mRt)iIkq3F6jlb29qtKM<?osOsD420jOd*D5;=~{
z?H!~GZrm1;ZQXn^sEd~POS!hQ$(3j=#=>}YtcPk`DZ%o(M3lgQ?}R~erSe205v+2o
zQ%Wr}2gA%8#Y`Vex}#vZudhh(0=3cMLmD^2O7#aGj-~}-lGWrQQ;*l0cWPE#Ks~!d
zTu9z?j>N+}MAiv&5}^A)3E~@MjB3rM&2(<P#70Dm{a}K&JBzqRsPBk^7Ty7O&T~=5
zj>599tcK0Fm?wB3VMyZzX26WyE=!&qhGP1U?oJyvnj#Os$sM_uesQuIL9N9~hEGxI
zcL<lZ!V`Y^(p2EvaP3?+`P925wQlqmb8iyr2Sm4R=-k+bSIG$jgiR1wi_l!CY*O(7
zp;3IvsCs0LcrtL!SeS|}H-<yExozRxlGzgH%a%0{rN{Y;?`kX#?q(ZS!k_sA*$<}e
zL0)zBZ9{*bnBi1)W|ywl0os>r=ElPGO|5Vn=#Lgj_NoeOTxuXZpaHfE%PUy?jEI@Z
zIpTDBDX?6yGM{8iX5dt&r0&>?^oY0P0lv0=8AmC**r-!5;7~)n`+H>9+qXmgj3!FT
z&ZX-HJ8(7$d=0-zd-@#sBvLBt{$358gX_E3<R$;Y)#Q42*?7CQ{t~(F<<8{U|84^A
z+`u}e9IRsVd}PHP@TKThq5l`+*PDfHq}bm^;kt=7!uf&|!S^}$GW~_*fMMYXq$6KH
zuAfb9N<<?^a{Zpbe8S>Eww=u`{F%Y1nbN4W`8mIH-u0d<hGtxvhNReZwXhq#<_NEG
z$J*}UgGboQvBWcc7=tAd4#QVB@%`aq8u$bLqJntks+Epd@!0quB0N9)E6^QzW-_`d
zfymD{yC$BVL@FJftDr!?MrDS-t{2l{C3^<F-x@zfG7l>@UduqmqV(SIk9kdl*l9+N
zNOwMe<$dt8wr=}+NFqNVje~~vx#tag1_%S1zFq6JUQNHNs(zUE=^hvnzvHF`uG>8C
zNLLv4dh${4bi$E<2e6m|4pN95qoVuY%m<nzas6%e=?HgmxU}1|efcmeb!B!S><ui=
zp5pd}MD6yj!f?(w8`RDl1tsT0%p!w{ZpQ@}-QT`essN-vDWnt@qGXV8*h_|#mQkhz
zSB!EG(=r6l;&bB^6W8#V1six5@!1nC*f%+bIo6Tzg5c~uB5O$<7IU|3JXy^$5#Kmo
z97pW*Ply^I$<M_3Q(B7gn3`7kQi3!CH>{1ri3dSJ(LOki!7nlfsma!m#6%mYB*<XC
z2%Wz6mN#RNJ69Iemup|RN}gnuEbL7uRH^A@IR8hb?VP3iU>fb6_&(j5n8m|z$KV;W
zBJ_tMaVWY+N8;>+{7pmDGMYC%VyCXggfu!+IxfVn>PFKoh7jTCqgc9<h6;OT^R6X(
zru?e_G{wVdt7Gi#&e^6ADK%Blt|`q840nMUZ0hgrs|SY7d_A6M6q)FJc(HwiBM?dA
znlE-KWi(Z+mBI!P<mHi^K;~7}T%_jJmO?-&#lCrW!n#QtkARc~`xsKH*bHpkwsu2;
zxj_*Hr_gVnNJjgk${%T`#bTX3rM=q4w|-O`UT^P+dHb{gCK9VX;3FzQdx3TZ3}4;L
zF1_xNMV1=(e|-!G3mC)M`U+&0@meUl)+xNyamS6TuAl40+LkWR<^xGRp7+3Sb1{cB
ztmW*NgP)@Meft8qb`M%icZR0Ce$9&v6QNsbL|UJ?%A6h~7btR5@4>J)j%N)`<=7$>
znFJ=r?s4pEY7_X3uv_j848JfGA7bUTM{v5Yei`m@quZsC=$trMv`t@kiz!8FSjirE
zH7@XePt3p4{V$^D@%O*HmubD8agRK32236J<A&*d^Yns|$^O3jH_T$97MCp>fB!7+
zj-dhhdjr$0Kh?iYWTS)?bqE0IswJIw1HT_nO9OMmMUjS|<KkDS)diaCN@R3@u5xC-
zXN2MJ<9<VOQ37A{Zu3xTvZky?JzDiw`?oBwY}t6(r{q1q=;<4D^p1~XZoO{$fL#x9
zR-8s{C0;RX7Rc6cX)I!FbJKvU8#t$|j)17dW6$ZS`!DtLrm{oiEPaY`f-K|v5+xS@
z2yu+aRXB;=|2eXFn|5It*_zyQyOCg(M$e6HMNDheJAc#N{qr&(2^_D@6OCqi4)>XP
z+o<gp1a5AwF9SWCU@t?^yyHVPr<Td*GU<*u>g<)khs~U%ncSSO>IvDeHN#VJEgSCx
zUnlE>&!B<#*8q&v*BofG*?=0{r?Inr1or0<W)Ey=Zg#Vg$je*)vA<G&wtL$4I)H6d
zs}UM}>Z-s$hL^x>a-Ps=)cE`NTus3V4}Ds~fp#OiTXR7jG0%LNkOCvREz)+!HGlJm
zD*ZxMDLVaxAk@@u624Gd9mP#bjA3~61JKHlJO+=HE=WcpasI>`ABIf@uY-iud}-6B
zGOV&43&nFz=CS!a#2^EnoLE3qP+#9Sg}S=rp_mmbIEFo*BssKX#=wU`hd$v9M@w$h
zDaxJrd`Gl(K)$s}`A6jY(%Dm#<~<Ix<DlD_^rj{_%$ry)i7dE~dtfeJ0Szs>;V4gO
zmokN$$XYgcf3lRfC^^|TvfRnex;zQ6no}{!iL4}1b=Y~68}=Pt)?RENTr^#ffV@)q
z=A+n;1Ok_LvC9o^o-!f%94Nl2dldwW3JPg{@T*LnXrH5$lrn+iBOwdm(N%?!QF$X-
z-@7ad(GlyL7g@Kr&|qFq|7pIp2nEblM3Uv5)x3;6V4#rh8JFe?&WgVL<uT)aK2Mz5
z@oxSIGwbk!>(0D&QS#~exf>V^N6S)DQC$wue9NP7dBfxMNL>CIy0;T7PkH)IVGk!x
z$jrSi7@kQu|G2+esq={-sBN_;JSVmu%Ki42<<g!!0$b8Dq_7X+o_f>T)pPXG<Nf0@
z*2nrbuM9_Zv4O|vL*I+D7Df7p2W1*tp4>#$QvE$Lbz2=!pOg5Ip?vD$qz#$Rmn}<Y
zYpR?<Fr8a}PTK(uGdC(V+z{db{P%sc7LiZ?xI9E&zwdm_r|seHUE<6=UQMJtG{{z!
z6rLk8mT+5P)!h-!?$Lg@6iJTZO{pCdjH793A1S2BTAjag{F5ORpoQb@(^kRrmJ^q=
zN7L0YQw0qo0{9PzRX79zunwS!O*X-B8v?NXiloJ*Q!<(F$jPbRZ3pOve9?K!@Tm?8
z9oT1ApFO!XFfumxEgxySEzT6%6Ix!`36}`S@i<Y}!hW}~BjDd+x;Ml`x{;x2-1e!0
zp%K|0d%GS62zGh-cO=94(#8VMtUtM0;-aQ<Yi&_x<m-9x=;>{1B34TCXuXeEgU1F2
zclbOunLdJQj5~$}d&oQ|Vwz+`4t9qsJMT9;C@If*Go~S4aTOJaoHNP$>RZUs82{X!
zZ?7H(VhDxf<A3Q5YftfvkMXr)rE8=_cQM_OICIV$aS4EVFvIoklEbobQ`(23(lp8m
zmGIdk;~1BuC2&-rCX^%yXq$9Sp;x*6_{(0y1)*j<KzcO-^6wCfT~;|qRe~YDM50R0
zgxE4q<bFc?DNkS#8<PEi4mEtw-ZzpD?kTX)q_u*?V}3@eLZ)Ez70=sMjdbyLTVZ6Z
zvKQ6)8&htL5p=6j#6Q(2X<#tlNIQk}$bhFf;E`p$f>Bv-(+&cG^dzmT4M@I`EVNZ)
z*Q5$_HEq{7rH*Ga(ZA-`wBe=;Gf<L5j<L!4_$>SHq2}CzWEx{GqpV#F5~z(cf3De5
zb@LJV{+LqEto*7&iC>;>!S@a$AXAGiG(L)P{zzZjM!2vL;n1_MR$}hZXpO97<1k-A
zI2(s-#^%o+hLEHnB%<B;j)kGwr7Kv4e6UM=B8!4lJAP_dp`2?tu>>%|u3W|8V<%{-
z`|O;}pI-52x_S*8N8XM=hiRq9Bl>vLxaKJ6u?u67@NiC1iV-+Z$`R)7m*)fw;$Zo@
z0|Nt)jOKK)adCx2gseQgg5$)Vf$v)cMkmPp*Y4Y1&$HIMZ*lToZ%W&|&j8(zQ!e3-
zNB?*7#4*ndCN(gi*A7Szks@Bvj~1dDR3G!pwvqY9`X~K=wnO+(pSDD;fD!%J;fK1c
zFre+Znx?_~nBe*aCE3-fEph)FmMx*h$E$Y9L~kwwInCtBVd1GSXc<K^DRdI#Ka)!~
zPL?Dk@%Y!uM(0PBO61WRZIRxp-wQr2buoN{G4nO&Zcq1p7#O2V@rsFx2@aNbQd_&p
zGL8|z4%0s0Y#+g;Fn)SvN=<wA6rqv}Mwt2%>|NOh%qp}R2LsJTGI;V1t6WT&1XAZ-
z1>2sF#M+g#&qV{&Ehd2IgnrRnsE%w6R}m|H)Qc_u=qP%gTr71cSJkF9CQGq#?;dv#
zzvY&GFdot)gDWV1rf9yV^=*FjZn@Z*I-He4eMF;6pDIzY<wfR(8VrjHe!1MNUJKYQ
zPJ4X(_zeS@$oCs?C0hF|3+j$WGaYpfR}CDI=VrFgj$#Vkb63v)yE@=(Kbg461Q`0K
zswnPqmTTBqXb{XwB2_V>s{&B^cPr>`{2k8N=}_o(Wq2gzFb$Hyjz{uAY6RSZv%eu-
z90${kw*DqsDNM5Xe<a4o(5v}l<&cU!^>;=vs=vg^K|?jji@orQFX^s$V3@WXD~wqE
z^+_1AsXOC{X|{G}nyS&w_}vP1<GYpdM47-s_a?P;?$P3Ro`onTA)qe)#s$>w4M~4_
z?kp~05wk$sEa1n8BHYYxS{lNm(Ch+^3^iUGC_Hm=+6pKMP%R|8sX>*OAB2*@LrN<!
za~nry9xMv(YnED`$7=n!&Ros<3?zM%-5!>Vmjnb2p7oJpttV1Ankdl8S#cRj@^qIC
zqi0B{$>c{FWmA~;!7-nZ^KC99(UWxsrREwSJAEIfEXMQN8^+rpyGHjME|u2QQ-{eg
zfI$lHq(?|)-D7rfkz(>7Q6a|t9op%7fWo{(A|t>pbJkA4t}&~~UPbd@vh+uh770_0
zJD(R;!c@k&*sfh`c#Tp3azXqu+lW1J#RMo6mKwlg<5w5%4jEpO)fcBB0o<<VI!r;;
zO>Db*RD|;yT3SX5p9KL8bV}rm8YI>m4PMx&NDzn|eD=e<N46G6?RIkK47aXcm8Y4&
zfsEG;LUS9U!zPR08k!mvugp#UZ?7)#|C}(TZj4&UP0sI$cG(Xz#DsR2es@o4_uR4Q
z*ONI8Fc4^J4s*Ac*kI9H6)0H5F%Dayl8q!A;tyfK`tYxmaBG7cJQ5j&FuL}c4hFu)
z5HtEt&zPQH-8h8*`{UojY6Pf!EnT5G=SjzD%K)v<K=gC5R}?H(-I5mNM9o69hFnX{
z0-y0^LFPHVN}>JqIw4CK7(vZmgL}pY;iAChEjsN?54M6G`9=z=DmY9|%bF}fSI3&{
z*HLB6(wp<IKI7A;r0)a1(uE1`ud7K_H~yVIFe;nJm>oN0(wfh0gDLE3c>V5sXDzJP
z1R=Y38_xlUqFejxZ4>aut*1wYQq0Q#8Q<puzO8l4lQTm`Nv%aBizk7irqQzFmvX?8
z)&d5;EmG`0Y#Am^fX5Bgyj^9;9Y+sF_2Y$<2mJXE6%~$1p1i)%VOPj1Tz}s1Ge&O<
zBm1aNq=+y*KOgkVH&G@@bU6>L1@c-YMOU|^k&bDU`oJNrEQvqDT09FA=204j#x$--
z$#ZKaB7cWg{1}c_($n3ph;q+0zsA<HT6PMHm6Rng@U0`>{BX0xQNzIn!q5zg(~cnE
zB1Q)VCjk=@88ru`RD@@2CQJ4VWtQpQlgMmpXH*PtgOX(*08xn5#;!9OQNF2gu*|?t
zd?)K)kiWA)Pw(mupfayh>%DnoNYQnZf0$4DIM)^p^fkD}u7;z2amjO7I=SkTL&!b)
z;guS_OvnhG(vO@}^y(|6!lL}lDf}tq{NV2FOIilmb*K2Ecc0DeL<geTF(wrArgNmb
z1Ir@4;wGC%qhj0=xN}tOtv%XB3i2&wvJ(Z3igG@5uMG%hSta40C)ADsiHAeHQ`tOp
zxJTgij>QsUqn!=SUcgsPTZlqgN^WH<V8cWfM}O00s!7L>y~lgLk5{G;>VG`G^~;uM
ziR5RxKi`7GmT<!5u~GfEsZLHD^^LAC`?}(jMZt|-6C0vFE9=WLznw;pJ)?9JgZmId
z%hOd#fqoAWD;y8V13@~Pd)gR-a7y$4I}`0o6ggHXviRG(M4xRU1V;h?-m=_(A=mqD
z)aoErj$p}k`&XsHOiX*q<HagWoxi&2L-CS=vt~`d6&WLbKiRdf?EGJg?9z?~cz4E_
zN^r>ee)TfVbXBU$E3S4^qF}+$yzuv-oxp(|k4+ETO0_#d!UAQStKs~nqEJio*idP%
zg=;dtqH3ay3|j$**oqyE#3obLRIBJjs~!;%+itOml0T2FN0&5jtZTE>i2x{fkKF#!
zfj4YaSVW{`^=pWJAm#;1KfA~o(^D69?Y|@9B8a5@9WD;|Y)TN0bdqr!g!_HeYg=Mk
zP?^N+3;L2LS&$MHu5Xflw+8GGz@w_Ig+xRwxX9rF&fn7D@2#fU4^`FLYHzPxVI+8q
zfu~28GAGqf2N&7C#`kUeyw5zkETR`J7k;LA1lh?cA30lQD4*Fcp6OKsgJ_<s;Y}2!
zKltD2x96$fkQV<d1FEipd;8Vu_M}%wBirg8ohDD4p4Qj9D{s#$d0U8eA(Z{7_ATk(
z6WIUoYo@<wxUHK<Y147Vy;$$o)1A-Ex3>Ypmz+Z@lzVd>+r`L;x1gWlQ_ABzL==v{
z?ih3`G$Ju711?;V!}8*>0E9x2PbSE=sEoe|Sh`b3@g%R#mc5mQaX)}ZOjLdb_v$4!
zCu6%0oddUHC=uC_p@xQ1(<KTSnwg2iK%Gu{s{9bz1jG}{!m6T%KAg!+OnbJF>Lc_C
zEhht2<jQx7g17lpghviwOIJ`RyRub6J`{sK?-MuHSS-}+UAVvK>BR-={EzoBCA>J{
z3^DL_h}ue^;-{!UmI|OEY9RQlMRmF@l=h1u*>$IkLf>W|-Lyaa?)u~Frc$Mk$Nbag
z@GJVftQI)=dw=M8LE)EYVq(~{KlT<Fz8~@Br)pRNjv`pSoTR@~5@l=Jy%RLpg~~Ip
zQ8DXROK9fE(eLY-Sl}hI>csnT#E#BRK`2j2yeZY<>zHG>uxOh+6W~J>pgNzAxC0c*
z#_pM$4+G=`V1mNN*7Ps(8wc6wqZmuvCvAB@i!;q&*~s>Fvc-dS=~Yd2vQOlS5=Q(f
zC^;%HOLgqZ^@vx6;c1&mbelSoJ<}NURNNGC%U@k<N!U|kn6;_TPPs^N>Zb9_Bfi)I
z)(a7nFLh+)azooZ%h&cJ+11UeErFE0q%q5)KzI<L3lDiLBL0sBtm>z>7d%(5|91@3
zd*}GDu;fdLfVPLHfHAkev*{cD?Veoo4*-T}jjSd$??9U@<rrEAaT1VMM9?BAw&%Lh
zB`_sZ3sLIRCGPi4lv7G!OkU!lBEr6t*GKMYY?7gv`F(L-XUn9Sd_6eK%mX2vF2x?U
z9EMSvV<&jUpgyaZWND>@fsn-j0L!7S$87ne>`6n+lX|6D?Zo$%s)6ErL#jj3StPs?
zQXh@6IYfzf*eJeFLxmGk1`W5-IvyObizI1bcD?LdZ^mDGo>4^{2>AWV$|7s|9yu%z
z=LD)-hn9q;#q)l6DA2K12Iahln-iPrCa2I%VA((c19R!YbMaSTt=i?cpUQ9l)4l^P
zHKPxaSg>S`$STGtv)KOZAXMPfr4eaG?hVEXVWzOp)Km5<s^{O$H|D?ds`|qQqMc9*
zh3x*PP=ie{V~Fhh(+;yygAv>?uL(}~{c&qX@+aCaf7>~d-6A=jH9z9yqh+?O77@`B
z&x}V5KVJw7{!hz3cj*R7cE89{;$RZ~hFOu$I<J%&+6hF^B>oU(1$Mqgh!5Bi?&4|a
zd5+)^{1cjA#$jO3)lC1f;d_ExzF~^Ba0+*foOB~W{)Se|=ctB$!->Y4UsTwT(T3CX
z&sT4yuLJ0p4R<CUn|qz;i)H_*1XanLoz01FNoknH4n1?VoG{7_;Y%sOA%liLKih@=
z%A60kLe*}Gj>Y)!pQcccTcTMa@(I=6T=sG6&)~yHh?G%ZY++&!_JIO@wX0i5pCt*1
z?-1^zv%JWdDiJ!gBbS$Uj37oc7=tIfX9SIhpePt&G!sk}hi$trxaG4%gvc&_z-*|A
z0)avSY(>H7aoN=74G7)!lto74K^tdiz*hgP@gT^m5FgZc8omStB%|*#v$qdD_6+k#
zeGGt4e;DNU%rBqeEgN#xdXE47*mcm3RexU}f^V+n`)CHCyUcJXR-42N7(U>Mx_Uuk
z^1_-*-i~I01ZLy?WYn9v)vOp)DQ)$Ql;5jgt|*wZXV^v07-U6h&)H7t^^-RMjk3zo
z1vSOQ>wpvuTiiqN8at?Bu-;;zRYi-Y^s3mV?X!PATRkiV3%4AQ6$X>ZDP~ra{QhA4
zX)u$GJ)&wTB~??`OiBJ6+jxIWsc*n{?ALRhVGJ>+-qqu`ewNfMr)KOk))8HloHsT_
z&S*S9tSb%x1$LY)GlIT@SBRE=MdbuT)R$yP0D9fBU35Ppq00LYmYbif^Eggaa=b9R
ztb?;CTzV$BDp_;3-|ndYZNtq6q&>3$Ox!D)NpDzb!6jN-K^)vkzbCzDSXt`APlpV5
z3S*RW4gOa=RDLOiyG^xud^yoDJXKDM-tk|+$J2g&yA{nAtFPcxWP*!eCJ>ZqcB#IY
z*dAJxlg#He{R@1DDSr%18aJmBqPzOkID18}6UPP&#cGzSslc0_OQkKcV!-E+ucO2e
zAUrCG`od6>FvzGWQ{?raLLVS#`%XWYkEwg~5{jT!s*p$9$->Sy3=7(^8}>}y>FlNF
zDxv>)Ai2JboIh7p#Wh{;<2uSQp}%a*3A8}ct-*XKmra@GoIIF;;12O9jm>^JJSCzJ
z`J_0e#1`R<3WBh|oBr5{ZoHwkNSBbrO&{B@$|yY~oQT=<Q^$zLNv~QKJrtgAr+OKJ
zX#-E|;M>)LN{aqq+HZA>zbeC{)Xh~IEOLuUN@Rodidf(qx};zbh}7Sc8fn?&1|Mge
zo4ZaIqGUTN<l2vXRwZ_fv@FKh7)@AGBUT#Q=EfGXtOJD(7IsQcq3Ou`xQQTC&Ji^J
zIz6M}qMV6b_>gnX<~Gj=Ax94b`rSlyZ6Fkmg^Y`M<HBiIRZ9$>`h*0cCmc)qY>v}*
z3!zv62YI;8|Hm}eUP3}k)Q_^k^h2H@L2bb>=(c7M`!7Kg5?~<Kz99xPUNK66UsIZ}
z7+($fnYya540E1Ol0HFmuruAlIr0WGwDdZab;-#@WMup?cDs*R4E9LLX{kHUEzkEb
zmF3?lSaCT3c$T-!j=D5N)77q&^{|drC`LD5@ep3<q$iR0Jn(R`j|e%`tF65*ZKj7q
zW;qUo7~$5wlrMR5^PlWNEeXi6tR<|rjMXHX04@<0|3eV?8=3i<XXZP2aLe?x4Ae)6
zP7!tT#d*F_(b`6*Uy~#GRO}PF*1`*tlJ<+T(t+WGu@;$Ox%+*QacgR>_3pB}w@uFt
z%h~Uu4N;HDLt8!i^W^;$t-q7?izcwiH`B+5NT)Gn3fr}e3!Ld@@wj#t9@>Gwrx;JI
zz_q8BUk-=h=a*;0-dG`VsKkt#V&=O4-ekQ$nyu-x=~tKx>Tq{lmrk<n+N|KkhQ#Ju
zybjizhIDk&^s$8kVAha;-7YFB%tr&cb-ms`4ysUyUoXduFNJAU1YzNpx2x8w>|sEm
z)@z{vDvfJURh)IL+6PI`I<b7N3xh4hz{<@u^f#$Mlk6k8_e8p8`i)lBc2rTbx$h>|
z$jwYrwkbDc64{1$sm4-Li>k%CMiT}z)!3OJmh+4%cZOkBER0noDv=x>pT+xPyxkPB
zR+c|)XIL}I@5SW2B7_Iyi$Efih%DI}I7apUSSxVF+za9E9iPVN9Y<1csKcV9KX;V<
zHmABor>RSHV>2*Ta$UjJBiRNb`a1c>d6Y5M1b51m&edzw@RI1UJSZWjyo5@2%oF^O
zXG2h0Fg;oL>n~HpRa20{)YxbtHs@q}Ib64WSM)QR=qFLEkX!|NK0kewq^ySj>vY*@
z6lU?ejn!ap;l<8Qf~)#m8!>mg3k@weTDr^o<K&rFvD(sM|8uXn?leP`@ZufkXCEwp
zABWf4jh;YyMntf^?B}WS!H{1M8qUN)V*zRg(__wNA0ef@!wV~c11S&3vHxd*<b5UP
zq4lfz$vw$-k)>rZmbIKcb_L0Vhd7v9i3Q*snNAj>IoKWwU}A$r7O<-HmLO_ndnhmk
z_j?-@fTG?b+9Hk$F9F=3(p5~P@)cQ)^UPB^u;lQ=nt<%u9_Yy1s5_X?%J4wLn$ns#
zYz4cie)8_zxiLat8!+ZH$ZLpldm^;UK2rlLx>2E_ma;f%5``~KW==70sNv$v?sOO^
zGs>iEg!E2)CsZo7x!++RSBG4R43SO<nm$tFqRJ;Pac3Q3pKpoQnDovFeE;80$g}(}
z2iho1)0I7c9gGOiy8jm7C;}!azs}21S;&WSz1`@t*8iSzwXAICL@<fPcclA?-YSRV
zANC~Bynf;#X1Wb(G03qhA24RP@ii8Bnk5L)mF*H$TE;-?H4#CPlQs_(1>utV(h?xP
zMzb!!F;k+BzZO*^tP!H&pqVd<@cu(0kz~B>RZ0JQvT0?d3~i!RgH+y}!%0|FCl4#`
z@DH!BdE}*nYR;<l^1(AFse33{LId2@C=rYoa^6bPn3bvfr{0(3LnR&<Ux7d<xH@Ry
zM<Z&nMQ1dgT7g%HJThet39Bg<kx_KHQV~Q7(S;MNzJ7|=<Y|Usq)aYo&;0J`3_~51
zYFoW4dOv*nj(r$DWM$)8$*zZ`>tY$IvR#a^g}=uXq=MRG8%h<DUCrA>mu5z#e(X&x
zQ&@0lwiht+G^9!t!r7YgLq@Fj`u_e`x~x6b(7C|%JYHTiKxshKhvxFgu78Ik+@NpT
zmB{H38O$*q((xKsfY<kbMti(U1&+nOn*KYAbhDd@qnJ791;rn<Fk-l6+^|#n)DF;(
z>E`q)xl<j)N091u`9z!{CH{kMlTHHC(ac&A8a8MAnyih#z?Nmqg!D{Y#H@Qks3|(0
znPMf@YJY8B#-U%&zu;~^$<A`C;(Ht}8KY*YUxxl$3QXI2IFKmpy|Stt*v@-7RJjX}
zI47X#SZ1NVFsH0?sUav0^6r@M<dHogb>vKl(9|+_ii|*!QXsci;{TX#>mg!Kp*amd
z&N42cD;#^bzqk7J)xVoOCsm$rnd17wE#rBwc2(6H=eVwm7ywyENd`?ma{O3tB?9&k
z^xWc6@<3hQ&s{fQVgm6)sb1aav4%1!EuER5O<@`5QEI+bM6(TlQ<hjnD*H}!5Y&dz
zpap`ZOk8{$!xQaB{gI(l=T9}$c;gh%o<~)*^^vHfCwP=MQ94?H*iluLH?fhyPVQ%s
z414H&CZp=#BDL^sY>k$fWbE9QnhMuPN<nI~-RyEjvVBO`0NfG)Ayq3;fgwsFSWG`h
zzAGJ>;YkqXZR}4LZ>&BMsV@8RTJCMPuThOEg~|ibAz%xx<WN8xrwf$@N00F`Qdje7
zl50=7)6@MO&;4(wKBZ4EXfiq#(@0l8;fF%J3Ek&QG-kD4VRw+LWdAU<5-60e5yg}o
zQ)e&~yhHSLUnJIXrcN}kL1NjK{fkA<KdS3jgocfsRfpi33C(7DMa#^y4~hgkVk40O
z`Z~xq?7tMde-X(_8vH4O|26W-X=oEwQNyJ-yL_`0Q!T&7szb3d;R*2n{{}-r6X&@6
z8tvqjgN$J~0kPxZ@1q?5phm)-hM?yb7k4+04EJ1y&t}d|&(3UA<Ms1GonAt8_8)G`
zqp~Z;xve1{&a;?$y7?$S2~`H@qrTf_oUX(1M!O3F!)@QMSh1H80VqrTN*zCLAZ5gK
zY3(w~GhD`^Guks^Q|9)MtAcN2-i5MoFizUX+Bz$-?$KkD^@5cMSLTQJ<c{e!#1j;p
z%rG+c9Dz-vae4;(A_Y48CXpEC4x$Az<%!&y#8a_W)k?1vI2nNY{B-gCn^k`R{Afv3
zFc_N*tMOQwpk`6-%qb-d;=55+7-&U)=N?$h^e#PjMBsNQ+I)|oVCwQQHzX;9nNd|L
zjQF+wf%zAx^Rh$hQ;`m5lZ9r#W#iBTc?S!V8zU>jk6(y|C~6on1gxA#MfmlhCJY7n
zOy%y95??e^4PSn34dqFtbL!iHU6)W@l(m0Uaak*Z+AZ%Yny}jNzuisT1@`5q{${=R
zMtDEUK25mW{Ih4(8ty{O+(>s9u>}I*ANwZVPdm{Qd8UZw(49VMrj>N2Ha>r~`S#zi
zMTOZtDt#R|NB)>JPw12RM0N#f;R32m{_PibLZZ7I1l!1n<x`pOjK|)WZvE}VjCy&+
zk2fqUQ&d}TcNen?4CHzOcC{=wp`V16OE*}eWLgitb3+PT2V>K#IF1_0`--~RX5CbD
zc+0S`_JJc5(&Wm+{<wA=#+J6W;RAj6B~N23BNEa7|Ea0@y03nHxAjj{1{6v7gPSSz
zXUqEG(4lDone4WWlf<%l0N0?2zO<ce`;x6VB|}l^oXV$>KG^JXT}cA>*}gcmOFCIg
z$Wa=j8{1EXJ}T0X^WZVwrczqF&nz&nHD~U$Hr_LHRDoAC+nimUh`sY;I<au&RC3X%
zhGvNW1GVY%$v=e0VPf59suEmkdY^myJN}6rfrd3}U+37JK5;`PnFP3221*Fh(P_pO
z!gnmMxE6wM<*VYT{>P8SsaR~`U8f538{1|rlMlbL#5>G;Kzy5^R)0GQjI<&6^NBg|
z#D(hLzolV=?9a#3N+H{7Rw{!pr`*2dCpcyTK;g|}{%OjE3rob}{m9di+eN5=5ydh5
zyWcSH@0*36eIz{I#xuV1Qs)Mk_}<lX-Boo)h%(WcdsF8Kd;gnTnGu~(t0SVpu{<+<
zTYk8CzueLj?E$e%%vY7H%&6;bA9U|gT(zUaNL8CA*J=3?rK4Z&hdaKPjq&8+g{3{*
z=T|<DSydH#Y&=ztuA+=T6tQ51X&dxVaVzvGVs@Uef45rFFIto2P)w1$F!`sBa7xor
z1HBKb-XGGeB({;oMP}6<`(kcGMJLcUJU7!fsz1L=o=H5OzOz!8vipm0P@-gRVVMe^
z;PfGQFC&9HX_6~%X6%|0MyC+-(l}E5F(9Gi-;E%bO-#d%s3QQby8>a89-f(%U<vb!
zoew4Ha}K49Q$)n_ebIWMQ56w{v}VUZ+N^hp6|&r<J5|#5QjYV!*<MF73|a3g=HnYh
zTAe1H_@qhziXBv&4D&jBQ59T+34?9h+jCU}7EF=x*e>1E8{zk4-hyyzlF+*R*#&ex
zp{glM2JSsA(XocK5C`=N<=PT=EhZO`&JhzhVzFr_fWvkAiU<AG?Uh1E29-1@&j?qt
zg9pAIz+U32J!byT^}6=3o+jMxkrlr6gj#_}-0JYz|1V|27@!h+CLRr*JF&WoIgo?&
zbizJ3g`8GG`Nd_gvlQB(w0?#vnk!3VC{HxZD@^L%{aYRo%Jg0N{ga+%X1pFstV!OY
z(2iYTR{2J8S7($O#ZB7pbX<A#9YD0OYzA+ZI$nzG2)<FbMZmp0<@uRg{2jO8*8biH
z+sd9HhrpDo$@&snMS=<*IEte8EKB}#;GIa}6@CIF2pGf-zHrEXM~e(KkOPu0h><5W
z6J)v}Z}Qd#{a&Mqop^=A6aKv<-04NEIeS8XjK7mgl`%{x$rRGc=Q+e+_i#M%2EDsv
ziAWFoYsA(0zA1WecZ4CRv=}yOnXjtpxV}g%8o-&_{GjtAi^Y*Al!)H!tIR_nXHawY
zy+GqRtELRP(GJGsh#cjdPW&*wlA_Xk&lwV`<j{bLpp52OE8*8uzl-`|+E__4BvN>X
z%*;h9q2e-~cd-1hKEu1VVt&7<H#UNTsh%iTRn2$Iq@fzebr1nx8P+#4qN9kJV;6&)
zkE^YD$Sv+UV%5?be@hRIjq%43l=z~g!0{uEsVsl+0P#wNN$GHi+lIAHGEZJCtDJNI
zG}eL7{yuedm@u1HXFq$KyulYag>F8|8^j~JOh<^D&i{M#-h1T+C5xTH(`e(<2tJ}M
zv4Ozhk=?kJ)6uQ}|41Ki<%izy%Os@GvtfUH)eMqVaKrihj16-Hgl0O~xIgGr{xc(s
zl4#$96A$d*#DnBtkBgw1f5)c#+Xt#q{%f?v^#oOo*d?pd!}v*O&vq_$f)o#(W-qR2
zG>+rP*e*&T_bRX`IeU1qD@s(T##36bYgz7UkxqEezhSu3i8YTdMIcCFRTxvj7D3^>
zB=C_7xJ<Vt%4lN1teSPiq$ohckR!kHbN9T<T!0<m3`vJ5z|!y9aC|wycjcukpU-yY
zsys#X&ADa!cFw8&MWs{I1deO_&ig{{u7$JTDV;2F{x?K$!GWVLo|QmFnO{kAHYGl+
z5VX6rcLjdABAdc`e<bnDQL#}Ha42?tjZocOvZdEZ5Iq*xYZOu$zi9!@ZqT)mPz&pn
zmgCl}lyV4F(#DfCnx+TD=Cffd7AaVItrwsmf>+o>5=T~TrVTnJ$`hFtm8{=YOggkB
z@<a?RxqOtBq_3n+$Qrc6(9|YSlq^CT5X@5p*pMm65&?>EV&ntL=yxQ`;y<AZlc<Aq
zh+x92SOsdR3B8)T4#MyLjc~Oh$rw>&Kli7lpHp#7T?CaYSjg5MYq$r2g07|+wBET@
z;eiyrT}d04#v7LM;J;&}Af#wb+QDEV+b&UVe9ki@LP;b?j#Kw^VEqG=h^iZu@`BS1
z8u=^HU0`7Hz)Y-trD*Xu@etapR*h28Rh_Pf;+re2Ij5ED{VRX4?%N}Qa-|r3z9wds
zpw$9*VPCvc;)2R5M{UXEHK6NUX%>MG3urs49>ip%Z%mYgk&s+e&15tDo5L~kAH#Cc
z^Lw)YN5Y;6d{b8I(w7XlLR5dduBBbY(DD;GOON$1p5I5+{S$+Ov%IV?Er&>kUuDQ_
zSx2liY=^w)B*jf`0maq>!LiAA1o!~&VdC&44s-(o-6!x5q90QQ3(N#RV8r5)N6xvW
z`GggPl65#ymh;|ejP@TyB=xm5h_xHs<SG~QTYl{w+oUZm$(kfNQ(JGVNLnjZ#gl3$
z#*W-%zRUGLryLj~1z769w_xba9AsvVF#Y6q`t+2{KP#s0%fJB46osG|jrMMQH7=3=
zNq6#IGC4TM<7~0Fw;J`)d~O_ygpo_sj#$}%UeQ7N;|ew>VvbHHO*0r`)}eU^E0S<O
zibk`vSyd^|k9X~0dI(vG($bv?c{!05TXH-<uRtSeiWX}z<jMzZqzuu~NQtQUlp;Kz
zc_~=H+UT=KTJBZ$um}>opp-%}TJ1Ibqc81nH5m4Jaj$A9nYac!g(U_<z+(Ef-oEzV
zH)*-d>BUN?fNN_*BS*$av+_!VwZC{Re8UHSk}x)kKS+gEQ~)qla)Z5C^W#lWC1&Ti
z?6OjbODg&UIGND22VB2MMW5@yW9>#23ipQVubZL>6TQP36|ZFB+(}2(4raww%xE;n
z`sP6$6m-etW;?#sD&}{()5d?kKxeu>o2ig}7VPZRLA7ODoeH^)r5@G;(<LZr63e#A
z3x^hYcB|cC%jRnDqd15UOrgrK$&KR@N&Qp!|5Sza2d?d+?e#;6k-{|>-hWdr?pN0$
z8@J3$q}Kxs$y!AlA4t@TEUO61^ppOg{}|37;H8MCqWX!<m--}BdZn}tnj9UB!I}x%
ze)!=w9Y)-1A*%<<Pc|_?P6U5NL6x7x#j)r3SZJi0`Dz`ufZuE*SPUrIV6yuN+o0LB
z<o{CN_{%meLhnGjyEkON7)RtiIz8K=?{2vV>2bO>*@l|Le_v5XBt!j*Bk%3X)3N2@
zDr$j6;%USy+7`py$w=Q}oVC$Gv}wVH1A^;U*B~0$-A=}D*=r894^e722NgR8Rau(0
zE**0($j~KYaetK&+kgKTPv;mO2OD+kHcislwrx9&ZM(_DPGdK=Z6|GPqtV2+ZQGeM
z?>XOhe$V`wYaZ-p@3roAY1Ns<6H@byWujp@^)@+IzoN(UaJltEh`#m~@;fH?r1b=@
z=VOzGr%76>pg`2$F)53S;miqi1|x#J4}J+JO)cgzrS7SzTSKt7etlK67*=i%6^KzR
z$4Vv3GON6Dy*@&LRQ)AEl2*kT;=Rtux;BIEx)g3ILOq=g1(t4)CWbsRa6LjBg(mpN
zA`g{=owDhZh&$2|itAawX8;+b=LN&;k_Hzh_@`2`n1wiqi*HZ{ClbK*BMg;~Nays@
zhREVKy@@!A|Byvt!zm}O*~E?@=K+A{QUx4tLj9VtvoU{U<2V<H_XZ}5`0P@Rd;<D6
z2yX?7<`#;(xI!NG=XhbH`3WoCk7+`AkQU@gQ4GN)T7nYm9d1Z+LDS{lwizus8@)7L
z;6Y-%t2g_wDGX(TlOx8Ba?1rLpB+g_F@x;5PpoybWn+h_L}=vh&uk)*Lb6CS$vp^i
zV56pH`YD!!Qf&A^bd@6gHcgC}1lZk!zA-&sZ*^h1%aW~s1*IYNrO#7dCt<?8F;l@s
z$fWy@X%lOQd$VT#W9WaW6TPc7n*d6RE;FW&zO<EYr!do!DSZCO<`Jv?(AGK0V%h}(
zz~3#YB|GdZ!and6OH4@(EZ;B<5<@{-I8l{=V1&Aml-I!e@s_;%AQO+S5-*9u_0x(4
zD~{YiN0Os-oEkk-MYL$f+|xBjn-<S0umxfz+>-~Vp@aLUr%j4snF8pRbJOEPgKEwD
z{e`DsDd@ySe*Jj*^2xKYhnUJ26=6^#<|#pxsOB)+SzJ~{Gu4N0$VF~-@|VY%kT;`}
z3f9gYh}HL*IL^N*R5{Yr6L&@W<lJ6OQHr9j-zx4Va1@i~Yh7$0jT0KwjLIJ<3xz76
zk*b+93j`^qAzkIyGnf=Ud`J(avtGu5uy6cW5Q|9HXc4gVR#!nd<?#bfx=DDpmi~)T
z7RC(CY{TmLS$4t6(G}6-xS4LDja2HEi5%~~gHEE@=k?p;be&8~jC8JibWRen4L{RW
zQiv=Ro^3g^o;ZkDva`LehDv`6LtOA9ABP_vVvS7+C?rfs+fD{!9c(6--VEdriNki&
ziAFV=lZDxGt2c-j(^8X<`zRK0j=G~H{99XP8|`TCqCIo-zAzzGJbe6-2zyeO43|)B
ziCS)}EmgEqs<Cfb5{Uj=9p~S;xx5r0InqjdXjV<>SL;W-J8`E3JDrtOFB6k>V}vf^
zFFoT<@`ZLugl7P57GFCg8wWVEw+y#bYo;;PS^&V4v(&8EjdUA@{xSsk|Dypum8?}+
z+!@`K{aa?AT<07nKaYW;FmHWbouZy5aUmgbm?-(ZpV&b|I>b~8Co)-#CawWDJ4`Fr
zzoC)x8kbl2DNGbKXo}%4up6N%kRs5Lg2iPW`^ZcoDLwY;Hj8$XHRX)U6I&PZ*Ypa~
z6~9Kr1cr<}e&`8wVF5Jc_2CR(XDy!h^Ij$Vj)dQoUV8C+hVB}16y%tEU$J?z1x*eK
zswA^$><f!Ux33j{qcxeA7vB<*(gkabheEk+ppOKXELgOHHpL=mzd*F5(`<VL3X5z*
z3QUd+kW3PM>=pXnd-!04{y2rf9pp$$t2>&SdxVM@i|~~>$3Tm^rBwAOMhdKpdIdU(
zje7fFgq3Igb?gSeOiLcp=^6ndB*}0xMjLM~OzUsJKxeji>RQqjba=!jk`lCtZ4wmg
zE+jpT<VCXN!Ue^I4Xhhg;v9-bU4`U^WNsr17QW(0FPth7+3e&2-5JFql01_jK@8nO
zO7)Qb*Tm5uDqf}pXwpXn8iqgVEK6v|<mAMiq8CwBFi@8k#-Uyc>EtRK<jb7Fi0mon
z$ihBeeBEH-qB`kBjoCuyPY9~KA}WCFPKFZIKo!@<0j7RNL8l1i4nKqkF$}w^j0pvv
z2p6vEWU095PqHyvfnE)$R+y_m?{M{YH%uMr5tKs>R8!k)k)dh&h(rhx%Oaqae8)z=
zhM9C<a^Wz0q^)Q<f=ZbJ1*ejse`J--lE<^1$Ne^;R|CPCu4v2KvUZXgxl4ax(zeg<
zmao)@eLfc-eeg?r(bU{ndO8?z`h8$m=x8}yiZKr`M7G3QK}?JJ2Ug+SrTVD5mpy2O
z0{VZH0^71VKLR0YII*jZ0GP=3iZ&(vi<)`nH33sn(g`MA-)%(YBwz&4d}_}{X!$DW
zqo1B=!pJSX+?v0Ov%$4~Cm!8ri-d_)BQoRA)Rz6aDThJS2&4~l#oXj8q!>&Fck0O%
zZ)7A?QhBOy5s|Njj&2w3OqcS^g@$E<k6!PXovqFwKsBHf|G1kt``i1Y?mq7&e<UMa
zufOCsNOq-veBf3P(h!aPf>F^(0}*Eg+~v#SEkhzXH9c;N2%E3gB|W4V3JN&+VlAKA
z8EPu0t+_@Tc+C>n^QJOPaixk_feb!7eNUe{NGZjC!06b#=Y5M<kiXd*IBz&gvSsRA
z&h_-MdUDlR0r_#NJA&WJ9jU$n%b=w`%Hpw07}iH?&)Xul<fJ#lH!H8vNci*4NXS{<
zT?=f&p6hv9*t?iaiOc!$Nt(&nSnS&3*(6e!)J?H7@Vp)W1Rd0TAjh^a>vuXHF8c@x
zH-M_PUN-hFdO%O0-VA`dl1_MU&-!EKpCrB%4?cW+iU_I<1xxwZaDACw^x_JX1DG8~
zx>;$;xH1MKt)IatFxU`&8DJns;uK<t6Awm#7M$wIKM%}+59_J{UJ<Lnkyh{lD~@WL
z^N&1)H<t?Ak}Ig-7q4dqX(CpRBKpQvVs$l|yH}-wCj=M^??%ijfb$c7Y}pi+nj<t^
zvoN6x@`R8nE7UrdZJ^KOh<8$n7SXg#Y|ypVJ(q2Y6`F4X>x|6Y(=)@cMyg_?bWLyg
z(j%I4yi{$pet+i?Y3?LBsGnT2h*d-51MI;TQm4-jZk;oJxTVCqqa8Vh0kFR^D%j8(
zH-o-Z?e1Amn1pAdBIYM0CnMd>pt`XYC{q^ovW$<prf7)csZh@sEHF7KniweL{|ith
zIj_TMik$licU6$UF`V768&nrN;uG0~dv+#wOpl#DE-+ypq~!BBwytBa<NjF$`n}Wp
zm3~Sf#K<?%>Y#wPh>_HerjzmSA$#&fxO{+Bv89%APT3!ZyS}0?%ut4K@>~E9N1Oq-
zH^J2ysu;<j#)aY;+r$CbNlo=qaH(Lf0|b*IT=>d{<;pYTZyE^xciR**RsLWgvEt-w
ztO%T3I6J;OiZB!Lb0Ic5;J~L9{fmcl5+^b(E6Wg^(!t*7<y#u&o&LgOxq^Y2W8dJH
zFGdeT{db%oiN0K(-5oXrl?JYGes(^d;di?1o_@qAV)@@0-J}%oLU!_KRVrI6TsfQe
z|Mc8^zRV&(L_!gr))yQ)NE7Oh@xJ=y&$3^2wY8J8^eCe%h?kRY^x2SCFpqn+(K97`
zhDWcL39pwa7v>=96vUv0U=dl8OJzru%wztv^<(On(8EpI^EtSB#_qoM%x!mzd6-b(
z+hB4cP{HB_SKSz37<l|u&#Q0uW7F25d(Q{q<%YMoN-Af?<&B4cUpqO>=jb*<KBp5#
zL5`SrOf6FR4}I-UH`Z7gXg}7FoNA@kf6<t5rd29OP(T@ZRO2qP3OpR<ktzfnBI9@-
zGFjDM7@^|NcmhXpA<<n=u7hLwjfE@rMf`+#a&iL~OfhcXv&pTSm1D3%<}rYP6yPzH
zsuYG5{tkl<5|NlWTqC!5ilkOxTR>Zi2fZ}n{;HY`Y=ViIUC1)rA~)CZQ`(6OAH&02
z*=lucJp*gX&iczYunTbG!#6fI6d_Cu2?>ejHt@zYumS1J&nu+^3LWz-K{lya!Ao{F
zoTGg%zd+X34`wo+YZriE+FF{OSOIHiNeU&?AR$BqpbM)*qWnRx(m>uH_Y{0;T|F!(
za|o!sp)n3mrcO=N4qc5+`(3N7t&T>TQus&KTmnnG4vw#4j04OUJKqT}Oh=cX#m;@<
z`DK<=^kHy2esg@8>>U=!V^`O?egH^giWkq#;rDLjY%~$uWI7?<EMN;{K$ky^<WTuT
zfwVBz)xnXy({{Gh=5qiC&g<E`O`*5VT{H1C3o)+%dkszR{u6U|1^F6!UvO^voY`Hj
zw+(H6eoeYbE~JUgNE5#J-$u34sdc}hm53Hm&Vf8t2Lm@p%5Xt8I($GX9O?<lxMiQ%
zg~1nP{<HW+lUoQ>3GoRGb}$Rk63HArmCq^to?OI?{@k|DvVHQRi}UgTv3!!Mqo4Nz
z(#36{9`E)Y7#w3LA;vqah|WK9Jqh(eN{fSxOcTkwamhC!nMOa={lnpf?4yckF(tgU
zf2ikKV)JFlKTAWW_dPx%E1u!OiFM!tLj%r!V`V4hgCN4>%qPqLOec(9P)cpESoz*`
zTz}|13#O%W#OR}qdDi<>A~OeRLUCHpx^cBQhDz6Fw^*hO+x+8BJJ){)@Sf%W&@-{Q
zDYW$r<bKZb-*dPId(D$ZY(J%Dg|Ba(k7MRy0;TC1BdrEtFU-FkP}DK@de};)Z!57%
zd?Tu|=WK)8TgV%p_YAwI(sr`K4GhG_FdJZTP<>7oY*5?*m@^gLCTi<8Ha5NoR};^w
z<DQ+M>S{NZqqy44c}1~OnkCFBqIu~Zg)cQ*8g|RipH3TWfo=89b~!Jm1C|sxiioAC
z#<M~itHmpNy?A4ntuj{drmSM=iA^g+a<~Y;MhyRk%Z{|V;2_^|1kxOwP&1>sdg-@%
zWHdb}cy&=~YH5j?iP07(4>$#WsTU}sqbCyb`S<xL#BJZ%kL-6%=l%A*^I>sqUqU7U
z`z!OaTKjRmy_wwi)j&ErA0hIXA24yXl&nOIP@)xhVc_^N8`Y>gqRM^@O6ibj%<^(&
z9yCNNilDM`O>Kw2=9qgjX;S?fZevL73ue0N;(gPmib^rOOS!e1xwG<!Dv8^7Xh%Wj
z1?Z_KKH@qD#S>T?_IXY8O0xoG$99iS-DDT{t+(&+>l@u66Toqc#2FgWM>OHM5wk$d
z+^f#EiE9=ysmwUxjlCqi;h%{zR3V`hWNMiK3F#CEQ}3roI-aLci<5a4R&7cxYeird
zkS$ll?(=d81qw5P{|R#n0@C-+yUPxrx3#T9d?OIF@cT~OvM^6R2lQ2(&4zG_+~@zt
zlaU;^Y6C>^<zJ>tRq()V3%%upAuTEz#VHF8FS%zPQCuZn=5MmPzt!My!Qfe3t8C@S
zY&WuHu&){8^>Ps-c@Fx?EpPvOznm%99QVnd#Z$)P@x+C%H92QpmVT36<A3$!VeRz~
zzJmMY5igvAMwGrpS(GJ9K}B^zBMc{pW4X04Fq*ECF+!lyz(3Q}Gh{wZ;AUTgDCY6y
z3dG-|ShHnvG()6PF_`z#ly|hbVGvyhwuZW=Ylq}b4bu;*sMYSl8@7F*^!TJC#0~qu
zRSY}$#h5qrFCKbwaZNUKcMrbY9i1>gsp!@<x#oV(%3c#E=L44c`~!O4wLal`Zt49q
z9sEAd-kU&o{=?jUNA72SLhqGp^>IT|?gZa5e5=|u*_Vn{MI!%=0esd1|HAZ^VL~9$
zQMg{V1~yA$VpnB_z|_?R)+d@OvezNP<*|{MV6e8$<1pW>0!pO5K@m}Mw2g&}ha_`t
zn^=2(H43^3Nj|0#Jpr3uWW$Q#YNlo6*QFEY;9z1gQi4Wd5zy%Wo#E6J46}FM$jidF
z8Efl6FBZu5n`-WorN5$en$X(KUOpbXd4tokRxhw*5`RXMj+t|=%|S2bX<An47#d2_
zPWT8}aNy0@Hsizq$4l8C!RP!KoXvz%i3YZi(H|OFTmdJu82tG<xb9M5LQLs#qj*bZ
z@8g)Z4<^EDmoq1qqDf?+^bn33s+9~LJQ`)fZ_Zc&wQ~ad+rwwLy@NZKJpBSX!2?sp
zS(yZxff<kS-b+Ev(kD;*m#<?6uqxN+i%ohWDMtkm6@Xoyl9kh$4xg{6p4nB?DoGO-
zxd+=RTap#Y4?s3Q-y{p6PXZ-<ex+2iZ9${g(K3(c?cXO903O_5I0qCVW~zPN80$lh
zfF~HWX55J2+or1`ht3gIFa|G%k!x|PuDf|;6$mXSCnrsL{P0>?S&2Tx>QuS1Ze8E)
znLFk4==`2y1akGe0&jz>&n_?kwdc__v}i=Bi}DS;0q%GI<MUy;(M;l2a(h5krBIsp
z#aDBnzM%5S)LMz|3Jy9Bm}a8dRc;gR_p?G+<LA`w&Z&qd9Edvwx?$e+siXvso44va
zI=F{ujnP;79+<f`OZlATb@lR$-5}X_ag$hS8@~3bRf68JS3IA-?t;N4JtAd!+Z9mT
z3u-BtQLjPjc<PW?p;BRg+|qb-3Resc)=S35-8PHRdoO3gdu~84Qtu<ecd(g^3aNTb
zf^EIjLU#oHMYO^`7hGZNQ>Fq3sdTM73`G%CJ%Z-9MOuGk#i9??jGh@lm!RFPy}xHu
zbJi_0vZRpv5<t)m=)UK`l4U=`BKJtn=p9Z;_=CZTfW=(YdUZpn=YFhboZ8SY0|ZWw
z$^vhOYtZQB=BHAO%U)&g%I^rd0~!85H$p4O-Y$!>6{Cv#eDzPpL&d5Vk9Viy&HLU-
zN0UTrtzPE-u6K~|Pp=zb34e#sCo`Br-)wWr_%KNjrMWb9&aI>qNhn{>K7@n>R^?yy
zd~SQZEqZ|82iS|bU}n_K{l37*+}10#MJ%uTWA?{j&$9!81E|gDErosPQwwwkn#lOr
z`gnAYZ4o~IvE9|LB8R-K+*i;j9tCMdh!*o?JQ)2Y?T9p;K1K^o3s+tzw0KfCFgNFm
zDa?T3&jcKJvufNdXmYRhv?E$G(Tv6$)dfcU0BMtOGB`x%3)-V^kssmnQ#?z{REbDO
zK{K5d=(3~~e6m4R>FzKVy|}?ojb4J|uYCRI@h2|g2VnJ@l+}n&F1Wr~uu0%TcN_Yz
zy4sn+aOW{`;5zyg7&@Rd*CYY!>Bp?xy#;O}h_I)O%$?5!z<ybeE3kz+XG6*dw6N8W
zBep6ny&FuO*9kTNvB`a4GU;~?EKuieJLkO-#1RSZrhr`)7deD}cX^*!VPRniQh?F=
z_alSmyz9&**gAF|BT_P%%C%!{(k9YGD7W?~S%8a%3L>@o9m0Z@G2t@9kQTJs*#m8$
zT5y$Yv9kCodxpk1yqqCkB%O&x?fBVX(g3k{z`JAL$I=yh?os+^L9wZ4*T^8XrVcL%
z7Kl9MytnOI{kC+>7}0(~EMQuNh$$NIWAyjJ-6lIPABFW(cp?-15z~zRsWPIZ^e-eS
z^6r<1EB2`}{npmNyIh+o!~GH^urJhM{qSweCnZNEA`i23@0UIA*iW*Nt%mWaucejJ
zfQh`fC=j3j4l_t2xAT)BM=|Y>xpS<f)O$|P8|dJ$DoelT5h3q&{|M~##9hsCLjfq_
z>9_l&OjIcdy|qo<Dt%UfGC^yg4^Xe3!`8o#WL>uZ_f!Fa!`06iCN*i)3V+G)%cK($
z&L%+V!X74wQ?fE;72LGui0w9E(vdo8Sf*(o(fks_pqB=4#18X?tCj+H>v;#S33|Hx
zV_!nBin{m_;vQenL95i#_<=LErXUuJy;INT#SU;&w&Pb4IIN0cWHa4ngq*L+93`!C
zO=P~&nHmDo5R=_)>y>FE#r0TMeEp}@8Cj6tR*_YiOM7I8`{lL!=OHSy|0`{7yI%-C
zH_`1R__)g9CMK)!?S9Y`HnFhiv+*2MG>b3@p^_T=&3s~)v*&*+{Q2}UFeC*2uj7qH
ze{Jq$A8ojAnsM-MUw&}ETsXV~a(xf?-dBv?<T|hW#>68eNmNuuS5N6cMWCIYiH0e`
zG0^)K$Oud%<`(Yk*akmm6l0PoSidYniH!Kp$nu$cEr6gExUsdPf=4%YhZZzC0g{#R
z(kHAUO%MP|N-z&Ky#hK#+K~=|QbiY9utG908a)_|r;4T2<TBI4fz^<7W5hxT6%mPz
zj`XZiIf{5sP*#fcm?of)tP9r7q_j_+auf<JY)C}|$}<6sTQxKMrHuy-G!7|igrfU(
zzl&DL)>w>^KB|h*E7nK{kMupv)<GCd=1k<El2L8%qU;1SW<uf%%9pHKP#_-Nm<t-s
z&CGb!Uv4{_*=|9zbiD94rh5pRc;Y(wAtb2Eht8caZ*8gKP7u|i$<S2J>KHpA`E1ZI
zfqlT$_#_*JV#o+qoUhJ+471SQ#-WNJ*UKl9OmK|v9c<pmV3`bFpzS4X4wYTgNn_Se
z#BU5Oan<CH-v#iu3YKJQn+N#9oF9WrA`x^`53++=dXyYo3Ac18FYae9knS$*b{q2-
zPz08Fch|Ojg5c9MTi*wu?6=nU7QNR9&Iw3|T2|Kl!@733T8^OJfIEcZh2mq5dq(~{
zl%IWhAa2i_mo!V4RUUm3$z;l~u!5M+$&MPQyG0wHGk5=_vn9jrD5Gb%sh(>z`Z4^1
zQ4H>7KKEE7P?UcFb+zGv|5Fc0_&NuCfRwX=Ca}-N+7GY6SKmDN>g%89{gzmp*Z>t#
zQQLrPX-S8yBVm{VbH*a=0<Mr`j1QM9DeS4oq1kP93X5P+WxyGekz}T``30F2GLi)t
zg2Z<~%a|bZAW3&w4H+&jW=BV!pJ%^%=ZO6ivFDukB}O}Hgtz+^`MTVKu9VpJpV`Lx
z!x~rdb_jvcy=o}QCB~Ofb!qZd#_y-dsCl1By>ODSuV?t{V|1%UZ$6I)X|$cxGP<2F
zhkBjPcSM`krA4&57#yM8&odkDtvWowU=m8wY;wlle(auG_+y^Kuo}O-oQ*Jn^|q;^
zjj(7bawKHrm&q%?@3qBNq8&>=9sJX@{IB8iUhc799(vxlc1}~yq-h<|l2P9Ku!W{T
zZ#5vP_YP2le;jewqohLiXAynyGkEBYKdIY!?_q&jDz3phn*yUpCLhZjwY{ECK{hLG
zaWmA8grW<Jbj3Ki7%gSZ()}livF6&44Vy9aSB$^~skb26-vT!VOE*j>2CFAs4m~^l
zgW8ZHM>}Y*oOueDJ3>5++JEl2VXVf1uF|$<G;z?aog`L`D5|8c(e`fP9|noGac=4&
zbSN+BFa4}oy^Z%6W`q5D-029&ygcjXJVdN5sekK80!s}|Mba>Z#{M(xEekG7jb6V*
z)su{|OBRuWvG_pkD0PS`Ooz@JsvN~Ia3rVmP9G&xcF%B<l3~$|<@Mh*qr<<{xg@ih
z8ykXtcVL0ZhWAtYr1`$i=XaylJX3~R@6AH}5H={DC}&J7Nc&thbqUA&@I(^NB8+CZ
zp0ebD<7++&FwkCuq*7caX&J9@o;0PD1rgdl<zy63E^U=UChYT;Ern6%^oM_ugum;z
zBQju6q6yszQ<KNEeA~y*sF)jY5C=)YUN+)LBQg?Y$sG&^P;=aVv(MiIOFP*rQhp6X
zLXkh?X1xRnn>kNR64v`Itn*{BBRRx^gMd7Wu$2Gm`g$96O#-;=Nk<|HDJby6YGUyK
ze!bg0s{{Gckpqtfhlm9a97b6N?tFkd0*K~wuug?m?)bLCt2<RKhDR?)M!U{=A8f?l
zH)fienkcG?#jwmE(AwJ_&D!i;C-R%`R?86klD_+?x5E^By9f$|Ncx=sor$=0Edwpw
z1x;xo1Al2#p~qld$?}-fZWG#|Du;MtPGyR?7>+8M$ZwqRfG>EEUZPX(L^jpz7dKhA
zLqZy^XKvi>%&*O`TxU8J?Nili8Wvl3IUikDep^Srx!@AO|1Iym=7sscyRF9O!6x}Y
ze*M{ZbuA9Of7x@lZoq?FIL7+s0a26eA*Hf`obb{8bX!>p3=#g`6w3DTaBsp$U34f~
z<LttU$ii`0jUOm9qihaBt;XWg3j7!tNn9)MzCKq^g0dFIxh56nc7U8p25sHWxT9SE
zh%jRR>@s05W@G`E?6h1#sRIP!kAOGNixvTHH*)9?+6}khCL{-oqe}uZgdflzHh<mR
z6C`3ELjQpcr|9SM3>#5XGgnH}AArVwGr#MIJ#FR@iSgG(EN0)AB>m1?x*Dw5zfT?n
zqJfbR)^^jb&P%T+%<n`~{%5?BMRtu=0@v9N^1);E;`r0LdmlKS-@HDs?%>IrTb4X`
zLWH-|KIij}Ec{-TX02NsRpbp6Jm2evUK~J^pbY=3nom6DuJ=**&Sxo}KD`3ea41I@
zi+D|PeYJ_c6G`dCl~6TmiMh~+?|iP`of8yHhCeYFCLpty@dx8^MBx&^YJ2HIN!wJQ
zH@hN*3YnFw7~e*Edu_7eUVpf1qtkq{Bxx53%EX!1(sMJ{25*2nmL3KSsgr9AOlF4{
zE|A7Sg^xOs_3v0MU;ZNJYG1KSvc`l*>k5VI9IT!vC;aKed=+GN{QP1;qcOlasl5Ep
z>1}?W7RJ8k<l>^+o)Nu(DjYBgwMsMcDYcvqCg?XhwPL~mE10AKro|F@U5h}2nczYo
zfh;{eb=M7#>~3UMKZj`v4`m<<--bH#S$#r=6DQRaOp0g;t10fLmpI0@h2&ql=syw>
zGn|Sz#mVXIb;jkgVl;>e(SPMf#TgT_b@MrerEI@RCPm{fQHziKmeaC#JTKm!KS^S!
zbXzvatL;*&_3{d!*NUwZicG&s8%P+!TS9P{yx$g3aGlnRAws(GjCF>l#2oN#e0<Jw
z_-7h?$0$AvP$@wvQQzS85i+)WbFD^PYSm1QuCgf>puqVG<gdJuK6~C>>K+}d>56W`
zv$2lmv;?q9s%dcQ7#1|ymx{MgH?D~-ayM&lY46RJICnZ-nhAa05bySWC!z{R7q#<z
zpBZ{S>v_Ba#O&cu35~h%b-hLTj%D_%2!Qej@A!$5f0u5cozSY(1+Z`yJS>-kkHH#5
zIyk%pQQALak(PB)iD=Xmf4-oKKd-7TQL4)Jv6!+EsG3BGOOO%h<lIUc)V(d#I1BLh
ze?(=@<zV{lv4iD~DWq&8&$CB8ePOw)`?+(3r-kMhUL&V{WNWjm<KN{Q=m$u(0gj;=
z;g$iB-?HOj>=HCWvZDXGijb)8zL?!2BznoZe!?TP&`V01WEZNkX2Dyu8S30$x=yA*
zp+Bn6;2U*Vd1@zJ+#cBIcX&0;SY^xboZF2qdR3;eBT4k5(%werTx7SsV0ju|D}A19
z9rT63znAnpU)49(%kD!xaO{l?5gmb^jXsE<>3d$Ueq8_`*3yovdri^D<?)#|c(-fY
zPJjh@cP*YIqsPDHN-uVvD8{J5V(AZ`h>?SO3y!?P{(4*bPp4{E%!y4kwIKH><|7s6
z5fUwCCRMeG4Q*Tm`RjcGox^h9R!jQ1ER1@boI!PM7qO?@{zr3Rv7m)A<k+({T{=+$
zzWuqp&-KqeBT$^*$Fwu(VCz#4{Dm^mrqPD~`zwA}TgVmWrgrcxj7|Ilq=qe7jujf8
z#XrrT6Q3lI?7n5CTZ`e<UUr#>Re6*V4UYv}h#-Y&vL7bdrj?Y<F6lBWtvbPoMh@-`
zeR5N#6w<n!0qQgb!@njP-Nc&!g@jmAbesWy>deh`0xZ#ZAk{GqW%V`4?82o2BClOT
z@-ER*agDn}9wWykwdvFMSr$TLHe7)#$xa5UFTjC~iF)<0A*=pN%K?h9Kc|^Iv9*hr
zQLy+$GwoIdToGz&Rh4H3eU5>PfoDrg%i3Ee?(S&^C749s_plI{j#GkVI{MoC=)^rE
zHw$|{dn*pNr#3Z!i#Jkr0q1+i)coB1dWBjA(L-OjS#saK)@}2C^geG>jKFLegW-Hu
z49r3q7uzDwKTW0Ma%2!Dr0MCub9C!@*9-Vn)?3g}23V&I#+Nr~{}_y<$i-f8PSpW?
zZ>?m6oR%RZH(8ie>us0Szs1-=bmrY~Q^x$W(;z#A(~@T`B}4g2cbQ*Bl72Zmi!Oh1
zE%a|JUAL+AFIH=P$Jpbct9>djx&>=DDe1O=<oH04AP>P(^~UyLOxEv>JeNUs8)UCa
z?A8pQ5E=$E<8(!9U{Usr<A1yw0sKhM-3VmVQd)?SC%=92_ZQ(Oi;>&xG()eGc;D@}
zfFh62-pC{O<at`u@#GdlmAasG{h_QxQ8BtSm}H7*(de1^HF6)jcsj{Y)B?V3T^LLh
z3E1{u$L*q8lqv*j@^I}=KaMF^Wf)K``xj{dOqi8Sp+3(M0R2|3Vh*evDSbcRoCfN+
zIR(+l3&UM6AyRDqWY=&ePF<~ejff>AY)*Xd^^AU<unIX<(&-f(GY&~g3M+U3WKyMh
zH~O3`#xWkViJN?wGk=!;1|e&<sV#r}>|Ryg-$LT;eDnj><~orK-y+(YNvsbnerLR0
z=H<QyKX&pgYZu1nRd%CN8d<-CsZS^(uWeM(g4)u<H!y?oGpXl=`vn=?xcKU+LNw7&
zv{U_$f9<%3H)@`@oPf-S9Aa!dF)Bk}hCPk>2Uj~lI)X*v7k5(pAb(twSC8On_V*ML
z4RVV_y?{BQyrtnEsOUk#BgKWysXlZOXmXj7E^D-AX0-BZthCl)*RdqEBnb9#l|$nu
zZ~;r4G$Az=D<`8=Us*Z#0T3q<#bv(|Wu;)~Y1n$j`9tSO`wJ<!V_%gawgj3N3(9y!
z|5<YQsTFoV6)C)Vx-mdXsE`MfOvx?UZ&GBL&$|w;G(0(-KhuKg5cKr)9!DO#H$%Ah
zVE&|25Ii=MYQKTf=|JTyoxFl}jLINZE8Nb%2s})V-PuXqa;s3O{5}}WK@zBT`&LCm
z>0-Dntc)3$EytPKP#PH6lD^E}qH^_}6nXYZ?ZF3071_Gd%cH*<QO<46^<D5=S5ZU1
zCOJv)HknrES3<d#%tT0NVQ#ic{*w5CG7uA9Uk%#?OoN)V%vKF{3gm)Nv4N%)FX+uo
z99>+`Hn*KuYo=1;qQX@>t8J-df994^R!`Z?L-L=izp8ra<&SQPDUvqOE1y+e!m46i
z^onK`co|SDmJvk>c^^gwv1nE6oAD7jU-o%McfHMWNrCaYxKme^AJ)Aqhm%^-&B}fG
zze%uQ*b7EEkiyUcf4R%A%brv+a0~h4Ommi2XT@L#JSx!l<BA!kLXs6Oz~QGj;qFGp
z4|;o78g1_EUJ5k#3{vM>-c7@b+)o$z+&CW{m)v1^$>YfCJI27Lt^N7&)2_c{y4{>F
zcRO1ORs@gw9$HS?=1+qY0J_>Hn`^$WK>hba;)cv%L<ki0Y*GgncYQkcQjM5HhQyUN
zzF7=gI|fHXU|8DrBRMEls)cziK)u41CC8G<K;LF`kB#Il4A=iq>k~pr_`w(j$#_Qz
zGz&_f@;g*}kJ@?$8OH{n$2wvw+u5PIkp;IXWB-&*U-)Cbsr{XRp>D|!vmr8P9lOws
z%G0E)Y_|PIA?I%ey|KzDI-QJO=rK)xG=&HOi?%(gWMshkHk1eIc0vq$NQUxBV0$i^
zOKc10=F)ugz>GRa9}5`*AdtpdH>?zB0|neJ$>in!?UY~jW8@1aG!H$F&B%Z!TNQ0%
z2^ZD{9-3cJOE>T_cN5*gv*PH{f6d!mz@fZcCX|cYJz{TczR8S5`=_5@05!39Acej3
zf)j@G;k^|Y>(h66$#j<0(pKN8g#*91H%6cmoxmprGv`r@7snann+ADWX<1BJh{JoQ
zfN97*i!(XbxKOZ&XTgCM;D}P5e-qV|C_Q-3M~kk;ej8b$etR5bS6VGbL?=nEL_o#B
z(DZdE=4@$sxm;wCr|4!<RU@;Up$>>95y>MXqiYa+2ad#XK{DbNb`TGSBX3M;T}A)%
zTXB_&J8*{MXG{5wQ#2t`22MlI;S~jXhgO^SQO?rU)tQ-bUFpBJ9N<hZe0jMjiRbA^
zS$N=YK*iAWN#45*Q?~HX;v<$Gf5(+)#FmRBczJ&Nzsr+vB}6mIiw_vv9RDo<y}J&A
z4H|$DVC|BRX5$1@J&ybVMMDZc&?KEUcbqWVP55y4qN$Qen=8}x_#`hF>&ja#s>ROs
zzfdk|8=l?_?3Im_oe5kI6piQTA9(jXS*aJ-R_`CA8SZg|CY+TCESQvo+a>8}A_H$|
z`uT=PtXO+oqHC9i<|!9U#WR7e^#dnxMW-1B9^;N21q^b!&K2wHurK5E+%Q)<y<_qC
zS`MmNE0Q3gS5CeoTKKU+c<w(oQX1BuH^ML9pZ$-R@8*+~_PW5ib$3gkS0gZR8r1c4
z(AV?Xkyp%VQzA4^eCfXxd3xQp=|-(%_Pw5@N$7q3b@6(-?v$v}2;0G?<xvR%iO{Is
zhlJE5EUc<w^kzbAh<b9}1S)lF@ivFZ{_o<02u*h~q2XKwewf}wllNQ4;kzKpijmjJ
zPl!`taKqUGT<QEbFD2mnbndwwdy50zfp&Y={X2zmABmJJw<zkqsng_-qVhSMJDc{m
z@zL;4a1pM=hbMTyy4LTkMU<*qbQb+N(^)Ha%9Ku)2+5_1zb{NDi~fTyF^{Apuu`xA
zR`2LRs{BY@!U!?9uo!+_jaNNVUx4{b?V<I%=8wF{1>Vjwk7C4LFpI1s@aJ76ssxuS
zK<bcSF^E;!CLDaewy-u?4Dkae*{a~{b=C0X><AwXI?Jwv{&73xJ>}L<tC#d1BLa(H
z!RUDmiIKVm4d^Z+xu`J$VZUtlGIB%T2Mzo6*2zJ+w3X%6_KJQLFE$_$l`3mDl6Dyn
zb<#&D=j&1<wiyB0X}eY3sBcBI*e;l^VhtEBuir7ki@W~Hb_3=cN{^8wguh7ic?_?U
znB8cPlH6y5oe9}zY75(yL#vtiOAhgCS?A|i_h$0j@c$Ala3xeY^U5en=J4a)z|q*|
z@4`Q{cnkm%;c4|5&D9J*xW3jgbveVes^ti>XgDyA3xFqb@Fe%NwKucMN_4=}F{6In
zD6y2PnRz|^4H@-MxD>H(u#(R%dZ*K;Ku7;f$1B1nYApDv%$B>QqT2E~+b|GB<jPuN
zfGMrZo}dZ?0kw$kwS|M%f<{9pZWV9!JDfs<zoaIWrXiax3k~aQK)mJINClc6e>1b;
zUcwMD72+bUIzGj&F%TagV~#WTb6JPzW%_BSq-ZfSDs5}~K#Td;!>vMdv*b8q9#F}o
z5I?uMJ@BcvtxZ8iU0h{&bQbR7@(XI{yoN|~gIL*IpN82TX|-JA3=FQSDH1h=`t05y
zzLrxI;p^D9?wbq?$JGGk?+cMnu(_o)swue@T1(TqYG;}L&vBnbpa!E4P9ooDr|brS
zhk3R4zA2CZ=sEAxU5as;ywk#z+hG=t)LMeafcqDvYA3HTY7BA(dv#qmYm8yaUvt>(
zriT5Joh+O)mUgS5k>TG9Fj`5f+X=yjq}j#QgmksbP7&&f8ykTm#bi^m<pe&6jK(s>
zBJFrSQ0)HGBrxH7IGl5Icaxm>O`M1|N*U%qdslRz6t?3nF`C>l7FBWNN*+js7oX1B
zSLGn9n$$HWTr%r;EBXxDT@o0^!u!JwTa;Rs4)#Zq5Tw8|b#3pbcd?<gEfPgBc~-0P
zhGpNdZ?u_Kh6NC&f5KO3n`fU_x5WaK(Qrcsay0up=-|X@E&BdpE(TW9(~%vj%c3dh
z#H$sbvp;g`Wp(&m4NcYi>|$8)moUgi?l9TdtgUypjl|&NpGUJDVs`{{!|DDCZ@?^?
z2`{gd*-Q6x_|d{07m4O#U{X)(gt1$WR>2us?nLO6*bWV#kQ~s@ORnSx(=c+=dS2)>
zHXDz(8~~DWL)z(z^wmasL*t1wv^!IuqdgOX)RXZPuv<^Q#CtE;M*9!kldO1gTQA64
zU;kT!;C79Qs{18-p2^KJY@vM?2f{TZO^phjcnB$QS(4h)t%Zm}M(htqstJe4nmHPc
za-?CIl!3Y>f1jQN*WYl!$f5eLQJB<@)TBYYtbW)yp=%cZ>xHG4(K5rLhdWlf<j(7J
z-cKXY)D+4qpmCQ;W&Q95nGoRDHui7_$_}BF(oD6ryC_7dWI)o^)$Q{L6N_J>G-*}H
zES(>OG=|JX)3e&0R^gMupwK3)v!kSs%RNNyetrPLNzf+05cj1yY1Tg>(e<uQTu9<9
z;(vY$zjnM!`CmBH{xV77iGFeaeD9eM24RB|M_jy;mkD<SjW4pcNOo9PqnwLfL%ES4
zX4H3oi;0sVmo94Cu{7D=&u4F7f6bk_nqDokky1IP>Q%#xU`VhY4csEV4O}PuM#xHJ
zRo|z)0El+d?_o1{7A#5$by;OKzR-cQHjxz-lxzhj5N#7$B~G6`MI;sfrND{#10Hsy
zw?mS$v;*-3<-#d4e`kl+&Dov>%9Lc7LrE9_qTh)3y~=HZSYvhZN9(tNG5FhrVA*T2
z%6P_{1+U*b4l(>Eq^HXD`Q4P-lESVhx8nu9QSrtZ`j-SaWlf0souTi>dtzq;-4rpA
z;(G`?45j5Pi`!k~FyA0n=OrTU&>Set48UqBHL+z@1~=KmajX>zfSF@pnqm9Du_=SH
zIl1AxKxY3FUc+agnHv{kfc(h=9|8a4mnfOA3;4XEt*LY@HL#2juJ>I(u2;W@)qhBB
zk<y-xB9`1GCCE+NXfGzC<Qk5YyfSW?GjlOQg2|v$lnaj=0?018@YZ#?$<58@Y3RHM
zG-vmqAgH1^&;myT*VVU=f`8uga9B^6MXqNkxVg7eojV6^lKceT@ynjxhtPLE{@vRB
z%zKq_Sw<m;!o!5l|1l8BM}PbW<B0IZvd_{tYv+;6RfRO$u5@jcR;ESDo{+ykdzrI*
zE`aNNJ~d}1J;XHQWuMc&bAInW?gOGDE_b&nZPNKLz*#+If3C+h_D_4>k3nsG$E%O1
zb?;XSaPVoFw&N5FPGj3IPedEeD0qWdLSYdM$n$8oh@9W`By5;{e5G@+H@324nVFh&
z*QDYi<8CVTj)#y(%!bCk-f01F={!=8LWZ}Z1jqB?xOc0baZRxkF$R4sq6N?YiA>``
zgEb$_u<eo5c|M8%-3&E?cMne&ow3(L-}uJ<NCMPp{u+pY<zX<LXow;9RIRk;_Nd+(
zycUR)9KF6qzn51-EjZr<@TsR4(I*kkzG<PqWG_>Sm;RHBq8FT$KDDMoRl)PILk%Ed
z2^`ie7NRcvQHa`?pDd1&pKSe3Y88sp$=cd32{$7TEH}4a@?08kg$=+~ekrXov=Xz!
zvw-Xvs_@FAbB&><;dr!Ch;La`(cqLYtncWEN>04UvGeK@f;e{V05duYSq6B+SYNA$
z(-DoHvAD4hwLQ0qqOs>!ISDG1$<)aWipDaPseUpJUZ#&s#@h-MFwu9qO4Gt5rnfGT
zwZz^OORA@~TbpaSjYc+FXcNshZn)~|P0i@Ou--O2U9U!I;f&{4aBv8K#=|!rm~!K5
z@f%pS_Rb@qB&T|Y3qe#H)g~7}$RkEM_-|O)M-fHgtJ=kN%E`lI<X>*BB$21vp*d8x
zl@r9_Bw_~KR7=fzQ*5}cWJabvPjM?wKY*K?mx9NuwcXt@aNoKax(D8WCgeTe<n1q?
zvJ(038|V7m%5*=>${ly4`w$G?A)5>I6csD-jXgLBD6L;OfdY4IjC^we)ZZBx7)_Sk
zho#WH3QSF{L)Z9C=dyoD7CXtWMp+zArIOL(*j?>@8_ZA2uSh(w4{LHA8?e;n8-iEh
zr{&cUnmX7@S%2#i4mZo-HvGf;G>XM%!@iVKoD*Iq>_%+B+;(3hq-1dh?=vMPSZ^Ix
zh%<NA#N+o4Sq1(uMLsH!mpMU4v6$YCyP0Oy`94OHKZ+C63PMOFSY$k8C%(SpANcm4
zO@xaw+r=ZldV+tWnr?}nh0XjcQYCoC-6A~~UnBgaA@tVKGdA^pKb6p|>US^YpKQ_n
z`Q1Im#`!ZTk2FKWvXB~y$9n9de{Iv}skSq`OQO4fDXapvM9>!faOivSRhU$s@cXM#
zyVG&2pkS>WA7!`eaT^OS1Fp$>@6g+!UF1fBHK7o-zhmr{uJE|oGUHY^9ve3gL(J?+
zBQl{;haXF>-&=d5Uf5QHUrh_h1K|5M?d*Tl^U*`Z_qovXUeS`>{}WW_9GkQ81gv7<
zc3<|oAeec;Sb8{aoXNVIM57g<)};B`D}P!lvx-NHVSb@#!MMm9y~tyu3{$C$v6y5N
zF^#07F|fmAP^SElF6(W*U}CK`!$wvEwJNGFgS8hN%Z!5|hSnjrW$>eoR#YBUb<U~<
zHt08v;`OvVCKe^r*tDMBFwY>9IShPLvo;{U4^=8qpQp6lyrXg%S8>=z%$T9+2`Ar*
zu1C<GFxrD#D;K=qL}4P5W3Vr~^@797=Xv<SEm1jJFo?Xdc6cIdsSVT}#_PFDq?wwU
zk(Xz|mZPe{`N`B6V{7PmKyM_aj_#vh$Eq*oHP$yvU?^>?Gn<-gY`{tw^VcRQ@oRy4
zV<Kedkgl8o*OZ=s70rz*syOSCn*uq4EnV=KRY>f?yF|5+jZPbAZ#jjRsi&~b%`9?)
z?C>q|(<f^oMJPy{#rKi7(8g#x!CN4ONbM}UO(gCh1<<HVXl<aD3{~I+DG@Nt9(snK
z{sJg=jd>)>riu^{LUa_Z)zOd_9Vtm;Li4KisDJWwNK++RT{7lkKPjjt#0M{_q<AI&
z%N892UD6NB(Hp>!=lP7($<udRJBOYLqfeyC1DafalD(H4g&s83gIH&yXTBl-YX)jf
zY+nCMKBJdI*Hk{%xSU*SZp?yP21W+QsV)l`v+#!4`r(;&bEQJ{G|r-#_!*aIxNlrk
zp%z>ex?T`c<Sw$qvi|sW9}<)PZc&eYV%a{gPDDJ;p_Mc$5QZOz8dOt;X7vvQcP~qx
z$9^#w@e0%~e35(f^a$<nmW(uGV1Ycjpt{-Q;S2q`$EVq)uqSiVqI!GjFrDa*w3<W#
zr)SL=evLss9pWF9E91y-7}s`}nUAMWg`{acYi0>(h0>_dy|&A>q=lCo`JP*i^@~f*
zGHbA)-ImWSfsgwzUguXKzio677RdG9zp#AMC~uhDr*`Kw(<{@sszCLt<cK8xgryIO
z3`}ML>`uZ&u=rh6;f_&TmjVJ_ZsZnzsGa;!blKlP-tUU&&w=6b12L$Z@o9v`@+eK>
z+ToRy1G9w`b!up73Gw7KbsV?!U_UC{HB7XVG~fpKvq>@y6HUev!6fd?0&yFqFk_Oj
z(k7J5+-ULb!}PMM+JKf)Wct~sd_^H{(JSxh(QWB%^DRF)Qv=g1X}--IrwG@j4rg!p
z&V1R@+ail1g4gvHK?DUwf>ee;O>;p{?vKy+YeUw}OvjIqC0v2VA9515_UDeDqw(Uh
zMBNrh<b(x*2Pu?&6e1DkiP~DOYOq-f5_)zc_0djTz0$3T@P<TQSgP@xfpw5|fOv$2
zm8G`(b*!9CF_){ZxRfDiO8_ku<$`^z?}uyTErjm}URkU{1jLw!>hu|qK+mu=$AuM`
zlm^8o3(J_S1AJ{8i{5M<U-_O}yg2bMiis}Sfv%e`-d4B&6&Sq3A{Gw0N6(h&w)r;W
zgD56}5kk{;6gs1fQQ0){Gc>Hc-q!XUWso%QWw%UQTX(<UJJG@T-??lar}t_q*Z*m>
zO;gF|=Sa!JRp<1K#`&C!@a1p~E*b4F`;4BRexb$O#4A?yp#78Hq|najBgt-P&&U=0
zD4OR|9W@i`Lc7fk7F(hGu#=sJ#+CB8;$=;ngUvk}UF3E1d&7^pDndw5bd<kJe5EHi
zbEx++!jGe_U`wO`hh=G}>kZ(EY1*=9D9%v*!=0EsOvldiO^I7~PxbRp5MUm5l@Y`x
zca@+=wZ)mhOO2sFJ0=mQup1qTH^02QNYj85=~>H2S+Jq2V!bl?B||GxvsN&7f&xw2
z2j5to<pxfMI`hhF70gh#wYMb~ct6|vRPcLX{?H(5y%NGo!)o)p0P=S)qQ+jkT@%%-
zAn6Q>u1dP81-na^!Dfq+!WNTRl}MMEXTZIw)%iJiDcXov*8+crd;`P>e7AF2G%m-}
z%~ZyvKu^6eYlOElv%#quZW84WNJqU>)tG^U!=Nn4+zl>m{JJS~JFAuqy@$9cf~H_O
zp}?91doT%;Vm&OEjPUnV@}y^$6EoIW!<o_f+>}#okCt#hm+HkQYpiqa;@9Cl?kt6|
zu^$m~=i{r*`MI2TD#VvB+|x4RBI@Vy*iaf8T_^6rD8a@*@*$vuzRIAXhoFj}faCnk
zRFuy$8OFXS>C9`N@%ryTfD<qon5%0$8>63M2L~kt0v{d_Tj3`}kDxZ;O7#HsDd7uh
z-*k5%dLLYaIq&&UFBI_uZ0L{z`rrakzKURwxQ177F5*>kKvE#gqQqHLTCcyNg!O0Q
zS#Sne?}V6D8=yov+hTCz3cWF?p_}B8)n{Vi=KLJj&ty+;$Y%gr?edha*(Vg+MY55P
zIZNt<=7*HijZcO*RdDmRv4kO@M|KvoaJq7duEJ3>h!JpGPpqzDNkTu<ln%?@>qX8I
zkRU)@3VLI8TG-gwV5yDsL%-0#ojg729-#&TL-Oh9ZwvD)<tpvUx#wg?E+@m&4E++Y
zi9KRd1~IX)uv(9D)dWC&QCFT-PmUN<!s$2fv5D=z4?Z=mbwb6}h%%<0`{M9tRQg@|
z4IZ{BAqSYzO^~t^46d123Mhk!LqIVu+y*W+MYl=S)J!R?Z$NT69jkhuftwH^9i&xH
zKEVm{4OX+JF?&r_>v$wS_*^$?(N`j%RrHcehv2%Y=<;QKGylQigdLYV=gYhZo>M)|
zpL&H@=BBo8&zPMF1<i`Y`cbFB^zw3;Z&Mm%Oa=x3UA*H775=dS+)B&QGKU9N6w%t&
z+TksbH;DU-5U6eHjz8~{&`{X_at$sNj?}Wm?(hgpTVIz<K@|cu{)l(Zb`-eCSBQ~p
z7IBAkA|~0vTg7n*j*z1e2HVO0?VhqK<_b6^&x0S2z24I)K&zlR*uFW(V=#~Ma=Ziu
z(l{I;5IFyunwZdn<?t;yQ2|QoHyp)iMG#l;R)c%J^htcTBwWw~L=Ag%3QYxzB1xBs
z2RJO`Z;j<WCx<4_fc*Y)dg-keZdfyt(R^?dP@Pjlr@)lOnb;8rP+MwGN3p4<JJ#wV
z-&F9(;icD5(^B3BODesaUNR`rzQHsUN7(N@_4<nEnX9(M>7aO$sV*WTfdbj6OIxKG
z#bb@uAu#o106SW53yo6QuaUXq819^T;~}1K<Z&gV$GGe9TB^C}4tIv%yHM|T9xbOh
zANt}JY-FjH#=Bw~d6ycppfOO%)>gJh^#wB|o{LdC&1}Da1E^5MrMyTGM>r~CU^!Zz
zzhQ$e#hTJmY&B6Q-Jq;A9No_c24}H2iWi8<ELo;}H4Ns#j@OE;8taIbq|{mDu^r#B
z#D=B@=F{6FwcO|^)G*Be^T!|n9BmKLvi+;`wfwZKBGC`2<11^BVaVh#cxiFqsTGJD
zoDBdT-BJ3zwfJXQ<oU;rxZE~-1|%mccJ%M=qB#fgf<t&QVz@Ru1k5em@r55^mMWgW
z`809Pg_cR!Jr4xP>rW(IU=3^>v4id3%f!Z!|9Ni8oqh>6J*AkdGq&Ldl;k<?&Rdm}
zNsWF{TbOSW5s&Pw{O7HXk`ECgk5-Iw+zyNtEKr|!l@OX3Zioz1Yh{UDA=2CZ8TAsv
zJL`fZBB7zd^Ie5T8;`<+q7Og#iHDxFG9vJ|k8Nvr^TeTIZ}GJ}d(d2<CWVd;43Do@
zhUQP5ykz~ZRt4^qY+k2KT^kHEdV&(HdDVJz2r4c2VJPXNX-~-*2N%t)LkOgx-^8Gs
zyzL?X*RqJh@Vw8LyuC7^JGJ){qfaxcE#MHd@>jViIdBU#a3MuQw+br&VdO3wpWFW2
z;=v6oIXE8N%jL*L89np4$k;kM^4m${@tqR%xu|e=&^B?`M~gB?C%z=ow%h2abv!=m
zQ#_h>qK`mED!iU{zH5m&#D3`=tV^sU_F|eq_Rnt_`_|Pq!XbRG@0z7u6prMlIZ4c9
ztmo_j@Bw(_SGaqs6L?KRhB2f6MK(ODc?T0i3XX`po(sf=K%X@qO?rlREJs%(L&7e{
z&o@IvTmJ7M%%7VDN0T`n9>>pFTLRj9FE8H=^O~-5)GXG|<vgg&CM<QQV<<oD;Y%ot
zamLq4{_%&Y8k0+xQ~WCZc2qS88&FQyi6V`O!lNXN72Iv*1|wB8ZqKL+^@qdKltVP`
z|Frg&U2Szy+h}Q_1b3I>?ykipxO*uS4O-lx#ogTrPLTq^typk(cXuzv-rUdojC0Np
zIQg*0$j+x^XRS5oyzC~mnNRP4b9uC~mS?BJW>@T(m|qj}kPA)HLZ+exego2oPR|1z
zu{8L=P3pk_YG3VdRrF09vm7j}y`x-JtWZHxa&|o9aQ$tsOMO^>z=0~*=pB9W_IhFE
z@1M(UHEvR6VbwfCV+M`i3lqx<43%2|dB`Uv@BE^N#ew%ZNs`J&Lt`jCQ<WKiVONd4
z@``1^3al3Y9U%ht95XJWDNqwp>ChL&<7~%$Nkx>E!pJFzNx_WS<b^stB|wHSR5_6@
zD<DA&nTVk%kz2wbzZy#hNcdlpd<U&7?*Pt3k(yR9br!y>5Ybu-euA3y84^TdIPnjk
zy?lsVx#jjU#n4FRR<pSXZ*)nFSWP+Fuv8|4&`WwqNIK7V)cR~YBMyQA$o@WOHBmU{
ze14Tx<hN31R1ai@9g37gaK6)nx%-OqiZK=Zna;4>=V#I9%eZHTHHWZ)JkM;AJm;%v
zgp!$~mEnA8)4`tA_X9eywav}a%=VO+=K3mxqogFRF`Q#o13nDbkwdOMEhsfETiYZe
z_kU+t*x0b);oST@4tf8#%wj07udaO>%%-itU2s@8iLb^Eg{xJhh^LXr+M4In08YDS
zHK$<HaZ*_iH^rDjsO=jPm`7$URN9iba+HQVN?95aWY`X~5986~g3sWN@|Cv@m3_dK
znMm{^OllN(#b_<=E@_Ioe1x%zko!#pr$c=klvvgV!Om%_BiZyG(~e#{45yzvQ>CB4
zE_MlMTSo}2CF3SGswu8Zqfn(f(qhSJ28szX|9PJW_)#{LA5aGwp^rJ}0mlb#G(haJ
zzkHDu>3r+yjvwG4+v)zL1IN=$u&ugu;bg1Q43;kk9%PJUOd2Ajv=LGMA!Q(2$mj(4
z`tSPU%1rD^GaJ_7PeR79V!xMOlG|KNiK`P`*MZ6CA%=*+;CSba9Sx+PYRE)9L+i#U
zI$j&T6Fx=B42(y?fs;rO2PITu%C^uId|5caPgLT(MK!)5BSus=`@sfE_DqVTJ#907
zzMYc>9E_h-$TYwrVlwkz3F~~juEm%Vi|HNI-(jcC#OxnPqf%66kN4^5M#I!z5_j(6
zdbEuuD5@|#B*cq6;MplPO}F{2`-xNb=IQGr7s<(qT^HH~5GccJKjqM*ucv2X>i?4l
z7~T`kfy!CG1?Gs(<egOBHo!%I&e}OT!h$p1uG{I9HMbc$ZL@`T5K?Zm(CfgkjY$Jl
zvekzR{|73rwmx<rr35Blw<v1-Pf3QQW>vi~<oAj0B#iw^VH5nMvdn-Y$$a7`PFtiT
zTs-DvKF|7tjFm-8yarRQDk4?h9pI3w0qjUz_W`SMS|gweaptIq<9hcDQn^O(U^9|N
z4d}=d91Mj}#K)M)gJ^I{7pq8TK(U2HTHX4=s3fKEC{1*3fwGNe)nc1FpPoLa>ZT&S
zm&t@eR#<oKI~STH-+{6@4l(qyzkYM`YT$Hf$`v42U%*Qm3TDWrut4hBZkHpGO6^|$
zN32_IynTE01e=-#n|dxJw*E&tiT;mtg14Sz7<C`3!;r2+q`Odav$=#=X4AWlK6B=V
zB*dyk3f7vGr;v}h9Sf;+Q`Sk=@`~XdBgujYGZY9QDb-~<bg9-XfQPxJyWbHp7zB}v
z(sPXe=4Na0_UqZpI;nJwW@MJAfM3&_%u5@+lf?IC73x4&CpE1!efdYcqYmwu$v&sp
z&h`1|ig@!^B~PAjc<HC=BIAcujl702%8aYOYU;YQDuDe1-J!AIBN|~>?t`E5Oyk*~
z#~%Do3!l0tgho)<+kf~Y)|s`fSx#QeTs*+o7pNbD=<-u&r$(nJV{rz#(i2!~bL+k<
zL3s}*zMAk5*ERokrC0P0rfx@?BhN!swMVl!D}&_EQcC29_ju}LXabf9+kH;r34`tB
z-^LA;iaD3ZD!|y3xK5<JJuNEVs%U^cFUljc#3z(BXi)FEc1hc)EHr*XTdW>zIZQ-l
zTF;3D7A=@}r|G784<>vdM=bHkZyk(;l5j|fHXJeV3@<|yHRszmwBp8DZmN3BKmnES
zKgQ}`+|v-j4B(A3PS_LG(<|DzFsmwHLB3`&%TzAkkQ7!?fB{bevGR$*?!T1V*61Q|
zz-6E=S#>4C&@eihjt4?C&k^&)+WPvan?!NOWSh}}i}}nn?ozZ9vKcC~bl|ocR(XE;
zWU#U%X&<NQG(cn|123C9W=H%Bzlb<xU`Z>1^yD>QLkrx)k$E-zt+E;^u?UD+h^Z+E
z_39R2hJiYC`a|XhTw}I&_7NYtjp*}dG+Yfi&)k);nX_%z--JdmEhMH5P0n|@TuN0a
zmvb^w4SqPm;gHnX+JsgEcSW7Q##*=?!3)pW*MaWo9zq#V^}iVJA7KQfbJ+D*s^c1U
zxu;aHc(twn!VKgA9yJP&t$!w*JaV;^;tl0dh7}s(2tU%7Dh0Fel=mzJEY9|YdZ&Ir
zE^*MbF*Ezax`eX-=@N#0kS`K6@+Bjn9Bg!_sA1BQSmp6W8#!xuBTI;QO8QC4bje;n
z2h4J)Y5w5H9$7~=5eLgMNvyud0`%}mlwqb>=%tv^3KTD9r?z`KgxTu415Og%i}cP=
zguYDp8ctof0u9LmIV2fEcvFF?we@YW3}npA9}G(JKR><8@j0iJ;D}vGuj}b!#G^bw
z9+0&9Gaq7`EFX>2+%68K6nhR58&|CH4_X-SQuy)-f+mUy7)nwV>ApT5R`Tm=8F`&o
z+_32r-uS4%1ZJRuIgmlNmWx0Nxd#w~s(pvgk;JSV2gKx2?+)PCk?P7#&CG(<*2w<7
zbbqOuRjx9dkg(+uweB)r`t0km%!=U{80CUEP5#Q)_;}@`(s6dOwn|7Y)CqC<yDeB7
ze=7F5jU(VtuoT14HRSBWQ2cIK!1bid#`2<&>$9xW<pt4(<(-@GiX;MJBFl*4NLa4V
zU4t$L8Ho$#X3*w$5^oUbYVzv%__+I#E^JR_`1<?FebxjSW{H&BccKwT{_Z<Ab?>k{
z?5I*MUYWBYTZw`X->`5>t4eYNNfLjr^Y4ewpJ#RQN1hXZ$;b#X+VMz-So#XR7f<Zl
z@}H}zW`f+E?xrdThZbg6aKY>0#&TQ+Q}(HgH|2nfXaJU3XG+iWW4l^laTv5%w<g0%
zI)DH-!y)UV%KBb`o=C7iRWq7rx50uBBPu>pXBmnjp{=W|XczDWd;efx&%kJRu~u7X
z4Sjjm+zaZ-WywRVzR|eFAJ~D72M_zbOV_|FXVWKlF!dRS-7*(25fhJ7qk>5HJA1i{
z9yQ^4tiGXP>DtlM*%(gQW_80?#fvgOvmNFX@wQB-&y+D#|IWx27{NM`T5#}VEkyR2
zFV?C(Q0w9LqESU9V>6o>t@B_SQ2(f67&29&dnBr4Hs>MB03&eEgr3`IBmH-qU^i}z
zqCBR6cX??OCWke>tFZmHK&GE1sl^xdT_QJatcK;jo}|i3O#e3wSi!cGGIsKp!u<Gs
zqs(Y5E0H#DhPxh3YsICc^Y$+26irHhclFqt0Uj1^RgOb1=(Dl((%~jR>U?*sJzVE&
zSS3!22Xy{jb8&W{z_=|y7uoZ|c-*0XQu3Z{68*XomPH#jWi5|O_k`5F_UC`QmFii2
zkp$crzi_u*?V*3;LIbI>I6~iB8EKcxP1{8+1Xuyo{j1xgIgfC|KJE*AE#8mFIp?G(
z18G>*M|o}H6omja0E$eC`aIE6wWWcQ*Pa6O*9qFz!ENB~sNa|UC$hjlfn|qoY2iJ5
z8VJeS-PRt3lvD<ar8Bf-@DL%<o~iJ?iziAI|I??XR_EM<&RfClcZtS%3Jldu)={B(
zT3!I9u~`!7t-n>OuZm1xc>KegBn659UdQDaIMQ3Mq3xZWP}_sA{p~TS$(Vs0Hs;8I
z4C3(bX=Y1uhX|yfu#x6{w#zcInSRx&M_tC0uEGMQU=GlFj9`lViW6$92x?!s@Aavc
zJB*X?#0SJDl_iNu{3$|26h#yfy|TGxI=naaMd%ku;CC0eOslpd%d)!44C2@f5oy@=
zEJ^Jpe7R2R-}C-p$;Ow(%OeH;C>g_t?*ob?DS^_}6oBn{Ls7wSrnuN2k%>ynx<!a3
zILXy**5p&ep+DLi{WsdzyyEWOb2PWwEkn+KE3D<_X3u91nNJ_OrU=4f$&OqV^iv6K
z*Kc#r(Ew2|T{8fRwv~sFjG9bG@L;9i(sOAg6Wt7|v^ZfgVU@bZxnXDrf!4YiR<K6C
z^*kfHOQ^m)DLoBUw{;XBzF5PocFYvMIrSWtti(*H2YM2o^%<ZL^9YaWU2>7=r+h7W
ziVxA;hpk293eHt65cTh-g(Xb9QD3@`jrxq5WD)^^lT!*Xeoq`lblGv@Q4#11u^He!
zQ0y%6ewju+*g&wiNA($Wy+v2!yF-*{fFsRGjVNq7C=EciFVFr!n!*pDY<7JVDYScF
z{oqbYpDpAS+0yd(GSS^|qTABxo}<t;r#Gt9N~acFAQ^&OkcshgfI5XHk}4Tld@_D7
zM7sQ4)rp^vXcd`oo0PkIqrQ=liqKANOWXIE|IUp6J6T%Re^0|o8BWuu9v&ZRnmffT
zd-!d5ZiK(NS)YUSAUo6X>Tl8mXZnR{j;XUtP*sCla`I}xTZd~JJ_S<kvWb<4pO1fn
zdVU#~M6Ok~ut$<^Bf^R%P^>N+7U=2*9`k)8TyxI?@li#43(zy1M5@r2R4ZQFNw>|>
z16yGz9fK$n$c&zPIf5!2+jntm`O#(64O?a7O@Jyoe~v28Qhp7ywD{6<>N5Cx`|Ra`
zG;oJr#iWW7@V@A>{onD3QCH`^z5O1s3N-g~N&y4Pw9fKsF^Lpm=o5+!1XS|(`PGkX
z-mAoG$ks7iDk2^6ye)5GyyGg>KA)O&E5}5n5A5_&6AV0sOesZ@p3tPQS!xMLP|yyf
z{c|<ol%P;k*tDkxfwjbctu227zx&yQ_EklGN$U|Jt4ph|?qcC5t#1gm&?GX<R6V4F
z3sQ+v$;>YjXS957e|Lb|?;7Y*4MITzV)C(mn>)hp<E-Ns&uRapn_~C1rN>q1BavwW
z?syWBoJH0Va7U!UL*Rp(_Lhk|z26V0*evH5SG9460<KInR2U{snx9`-B17Mxv7)m4
z=ch4wN-Yp=+y2>}qs{OkcHgG6zrW`Rs=46v{kL%D=GABl@rA`q9a`4$#aR%MN93}X
zB&9rjAGRdzl^)E#KQoTA!qU$ytRW|osz&dyKU<_>G=D=%o}rXzXAAR-wyUrx{<(0d
zl9Zj_W}a=16b0sbK~#2oy0g7~8b2Q@e^{)}u0Z{^V1tl}lS3{8>Xq?MEd!njLa>B9
zFHxeZT%S{vyHzX=I#STP3`qPsHlwOgS6|Vz(dU3zGiJq6_DYlFo?y6TAtn7WgB<**
zUcu&C|2@_ZrHFwXCLH`>wI2)7$OnGaboM~^=?uCABQ^{@iZgY*3zaR)l$B3+^m)g5
zTdnFVSa>vx9mFyotZkQw&|BQ^gnlZdUtlaQuF|kaHu;VT8{WVweTo06^aU?~MPOF=
zpT7w&j${~3F>6KH1e#K>5DB1t=>GSG57&J!QNij;8Or)!+klml=YlSw91%!hOsp_#
zTgDqwLtYPh4dg@}Slk9RUguoFG5)F1(q?Y|4{m1r)}UmSCaFw!hK7>494UpL4ZMJ#
z7&NA&JK04WS@g<6V_HeB*2R-v-A1G^zMEjP>p31@Vt8r5p%sTD)lTr%c)A+D#T2pX
zAwqwXArK>^`tQWs{l<3G*H`Sg&-Z}RiR^6;|ECR()f`_E=1*}CG4Cu=EtKeRU?X%H
zV~T`i3t+0n6TE4?WAgcfV%@TL<nU!2%s5OnVXl(Yuz2OHuA|<Ce#9r3`c&$j0?%u1
zpe#%CEWSOdJ_t*Gn&Qs|no4*OU*RRs{)Gas{j%sb2}N?B&wZDb)t)FjJVOKkfk_@y
z1B)G2`Eja~V#~r!8_rf%Oph8=ES>Of_ev_X$6NO)Sxe<x&S9|AG|bPstqnSchoUTo
zQ%oS6^CYFkhVD!k>4%wC!f2kdg;;6qU+GXk&?RB*3<cW>**|kRre$9u;(fNwit(Qq
z6$ZRd)>00%!oY5SF-O0Dh_yV{HM1Np)==0M_{g~7qH%E5C$F)kt{0q7v>8v?3+z@D
zmdfqop@X@E`W%y=i+0Gj&4H_tuk9lFQ&mc(TI(dUY07k>@9P;Y)=qU5%J)0uHUlw9
zni#-=4Oes{H*lqCnjc+|UR3fJK$OLj!^srM0{m@6YKat);!OWu1h>oPP|aa_=08M=
zk#h1ObGai2#{J2fDq^=ZDvYS_zt|WRsEfx+c?VEjT<KiIJlLI`US`1<sR=0vaU%aX
zn13nyzPjXDShZczzq(&ka)ydb4AoU`X(3>(dsS3`nV2hDdjuCQv8=Rvf;!fm%<Pj?
z1MY8b+a|W!OE*q7Qi{%Da4iQ6uBCMY7$_tv#Ca$S;v8T?uX^^dzP2WEe}aF$r2cSj
z7>>pDo(Ad(o0Kb0M%JzuBvS#t)d9ynt;j@kFcFceB;p`z{hkx&=dj$)@8eGQdT~eR
z!N9iuxe)DE=F-DZxGI}4GgHZ~6pBE?kb-K(V*_ewacoHV3R(L*db}OG$U8<bUg7eG
zx;3oy%*w$mVXF%YYe-WCP<9hx6tIK*7Hda)Aj;l(%gSclE}tkZPn)j74SBt9mGrjU
z7R^yfSD{y;o3WI-QNw*4-Kf3ta^>C_4nhkkk**e%{8(8z%G;>!5ur(-*uoA<BtY~i
z`sxolo1A!A3`2Jj?b-Zutq>aaXxdc&<C!&)vR2HF@_v*3b+urp{~S(t!uQMrT{O*+
z{B!BTZTH66x@B7fZ!W1&b$|ECd{e_jh2zZfKf|}DF2BRS_dm`Z67*ZXOz$2yjD*EH
zHjrj1VQffbu8f-=;Cp2ZkbAkl)$)bvZg}q4PduzqVS~$ODKCxV(_%q;gSG?vgYlto
zs@cSoiXRL>@}yz#raK+2^pth{2q!uHTC@Uj$b~PUVbK)s8eCIsPQAl#J;B{qPYG4%
zQS_>b?l7+mbgDdxdP*C}n)Ql4ng)-7{y0?-fRYBsY=~J(&yb2*o@G=pj4u?(XCaa?
zo#luHFR^3Im!t?<uJJ+22^a}RHxYsn74TLHDbs3<Gmb*_D|uIqRu&hC+wgKrm<nS+
zMeNyT9v%dkT2+xezzijNTniO_eCsb0)o#MJQ+xq66}(h6=~+QEps5pYn4q=nm(Be{
zr_0}^^;`gdmeSp($g%*N!>;e0jjSPq|MeERRhj`B{a_HX<UT$=()RC+AF1ydKh%M`
zU9E5o_xqd_md~*J;FdFG2^(2zJ!4s$yiAfg9Xtp(9w=C<)D%?>R1Q}X{b6-siNd$z
zIxh@!apJ}Z17+K4u1rZ=2my_R621NCUI!@HPi}^h7*itH$)rK1neL2#o%*He4*MMi
zJ94hi6~emb4w^v~JRTEthjXUVcVz>#Tp9`<B#M($f|?2CINWqQ=dxsJXw$koR#~Cr
zH2M2yf3O#shb{iwf1)KzD~h01BpBUNxLqtj8n&3cx*~rx+Q~jwMRe=R6MZ*p+J1Bk
za!*9`NlZeC6tGuML11fiJ@_}c)a)#fRWy7P?b|M8O?D=c2jN$Zr<tASAU(&HA2O<m
zYb5@nyL$5nczp2$^512D<z2qH@igTA($=%b?w5Kp;TKWNC1n$%`QD0*$qrD#gdDHl
z_ZXEX(9||`9QGMHw(3N8%#r?yBA8Oed$*NJ1MUnWg)MhlS(UWO$&GMV!g@V^k>C`&
z<@)~Oy46--Kb9qnD<gJsdAVrqe47lAY1cK6=cccwOA9?}+SK$k0FhAQiapeeja-@i
zT6Ot{`*xW3oaEs)4imgqVb<#uz+a+yElfh_MwZp*-Mb<Q%4D&@Sa=S%!@DfyD~g9d
zmP&Rpn7nza<ZJnX1ZAN?DSdw>J}n*LuvcMv_+Y|gh5|@&iS*FM1FemJ0sj!A)D81H
zU>3j(EOvfH>ga3e(k}D(LjbiG!Ba-$O#<KAKaA#<Ry7>9E)!_puWP3E8=~}%Gr7SC
z!Y#Sls3hvMp>bsm{w@vZ=!l*huy!$KOb_!Tjnk^UbGuxyWTHHVv58$FK^fVp3Tlm=
zjRLFJHwou&oRrat`y1p^(>7R8EXfKe&Jh2q*cS^L1#I;F@HRi`e1?-u1CnGLz$KLU
zSYUrN2<dg>pV#MWv41E<FM=EhOaI*5zk21K?pZM|u+MPvs(1&-mvK<2P?;B%RUJGp
z3rSJ_4x-&H+GnKKw=OQq;RAWZPK%dG!oWTS1)IVt?kW8XI48xC8?t%-WaW(An4JOP
zvnrpIC6Nh2RrjUUb4*$!w=dKPxUeVyykv_hR4dQkGXoDvanb=&VgKkjkrz{y9HAlV
zIFZqfC&afOU|(CdjkbPuk${7T!?<t%@!YCuNfVN1@gXP-F;q381dg-Ou6+tL0-OL&
z2t!P13f*Ew$)GcshE5GD#h#BHPduPHitCGc83nN1R#{v7)<+a9AZr^_ZEHkWi`zg{
zN_k{NkU$6cG%a>R@$jJ2;&p7w_6*(Fv)csrg!j>zCoyux{J}L^&$+IF*%VzTVO&nv
zZ+v@!#>c&6bjdHOdfktF-S0bk@;o=3zzU`elE^zZ`E!RtK;qnFT~c~}7}F}{c~E|w
zDKbK+8VVyCnMBhezlY2#m1<s#IIi<EnnK~0P$+zezha1d|5T;Z6B0$x{Py|;n`&JT
zBHjZ$!WohQVv1U*;mAxoLvA>atv`#%220*IEqwltZk1_1a-Q&F6@-Q6)tsS+D>3;T
zkcvC}r=SP$$r>Ax56HR32VRk3WhwZ06RgM_Jg&LkEDcLGu~TYxBeP}-&>Q&&p_@rH
zP3A%}XkJ4V_71s)(%mE>?})ZemvPVm4QW78%_-nSLQ|GbKqe*njsua3r|oHj^7S_{
zTD$NGN6o1_9y~+av(D#!b1q5w@8c%XmYK>VZDkd!s^`!Cfs5@bgz!%|Nv8QQOxU=|
zN7r@4pr<f+z%zn7CO2!4)pjMc5;Yw~l>Q^u&gje;$+AS?TxctcQ#Iuy_s-ZBSps&M
zPSo$Y9z@Dz!uUkUoMd^1RpmUq#2`g@MtWiy9!?|GC1ppE%<LC)+uy_c^?AKU{uz(r
zSVF1}EL<At1~;pFWX7NLU4;wA8_U=ou~>2_j@??E&q>$1soHiZCFS?`4}ik<m5Gbu
z1%a5l$-#jhY;$XY!bBRE{xxF;Tsg6eYd|N1!enp-QzSTlIRFF07wtaLi;-3xck<Ki
z;k}SuWfi9wKg?6kYXZ0i`<n(_hLKHNu#{_3blAG)W`ENV$gW~=mf4W4Y~)>Q+)mMT
zygtMU<k6=?r*7vn*G7<~&i(*z!PgBMGw});ALeSUa9YP|n1RyAgchBNYR%(3C2prU
zUSFq)Df__W8@-fP=YH}2KC$adh<bk$zHT4WzkSM<q>hn62F|K5#_Vu!w2nU(O-#RF
z)Yr9j|D7gy6b?8G7$5%dki=4R{paBsDc9>hD_KUObjA|om2h*G43M!S?H`X<8*XaA
zZXx+y!bukgepjVfWP^Jc1^%O_h${zgDlUja)6r0M0CGr733NHwpvc-!=I%t71Fm76
zV=4^$-GFQh_|!lQ%m7-*2Ov#W)Xk3%`<ab{#v^VubEj+3hG{%=wI(`K8AtcjG_bhu
zn{X4lBrq<OT0BjBqZ({dEj?=yy4D)uice1~Ihau|gL5dCVoEAs!UW@oX@91jLMLgu
zl8}O=De(5<nl1tkLSu3=(v-6v*NiuOBG%gajpxP)gQgQJm>5T_OB5Fvu1;DVb~P<$
znwxZwi;}s~b64rN<kTS(y1wl4$pcdW6(8nw9r80&FLYJ{&V{-@%F3#^qp>FW>2N(6
z6H*J_{tV*m-mH?j>Yn)SVPtr?alv40Wm+ACO$5EDy4YzQEeq&j%)#ToQvV<m2Q0ke
z%4%|}$-GjRV^o<}Q$GHA`D@m1@DiBcrh`G{UZ!T!ey8NIVy>T(1s9C};(vWPc#9?o
zc>C4Ithjn}7)IFjOf^sJ-N6?02<KEPV~+}los`CX()M6w1w((&?diz2oyy9lE`BxS
zqA7;I+`r}KKals?wYCg$vwiK9)}D8jqs}x0KZO22;7xyR>15zAiVa4aVEXiyQI7(R
z-kE$I@v%XCniQAt0EI#XJZ>SUTguJ}vFI3Z8(;AukJcd5wnV6iKAK8vGWm7w^xGoQ
zfr!r*@<CImmS=^cuHfsI>Lm<pGCf&V&V9brgtVcM`dlKd|GgE7+9QS$1ZF{|E{GOG
z#W|expYPTEZ><90^6;|X{wXTe*nfYSb^~)iTrSf4r_+;r!G}>QELXCzPmw_wo9L1o
z3a(8y#+RPh{>(76lJMV{;BL0}d&+@$l32`KtOh+3c?sU^dG#1+oS{fhqgG;ym{5`1
z$zDhj`pjtuRvE!3zaQpMFGHU*!>0|g`OmACnJThju~Az;mGyr4EJ`Y&eGH3bk41l-
zr)MYKZ$&~kEP9@0e;a<?v3}hxIB2X{6>;@7d(3?qXnVVUb2)4Y3pgi0Cu{n+_A)R>
zW^}wWw3`!P9lC;`%x?U7w0wn({PK{Gj}Kk+z>)fHo%nR*`p1F`eFSu+AmAR?+mA$a
zj4`^9iP&HJR3zaR@?MpNOk11kx0ag>rhGC;z*E`Pd60raDbrk~FJ>H911@k`z~+Q7
zEjCAqdlaSG?i@Tv$bSug#`b{4t(P--a%bISAaIa?5;I|QL_5uEFU2mhvs|w!H0`4!
z;D3!;7~|$%ZQQ$$uu{_%sh}xjHGxgbI!yzMumjUjDNzmLE*sn``97>cwmWeIF3MxN
z;b7yg$hvO=nWsKoDs(u*bw!mY`XuugnxH;b)^kv*)7d|_KnfZajrVu!6MW}%hf4f-
z$~zgj=!K$$Lq-&O3J6e1eFNys7Z|tD6m#1?sV}3V50?M!vn^1--Sxw+R2hy1v&^J~
z717$B$bvU*p~-9KA3py8QA#V&!^Z3HQ$X+>PU+H`k(_{B=}cjR^ZQhl+M8kC*e1g=
z2I-dcBkLY_kf7H!E{4dRX!=p|)*b7(-x<kjdneHQAI$){XYkBJwlt%}um1+|n5CS(
z38j@;0qhqu-o2H9CkRl%;Z~T*kGic?{`*L<ISkD{GriIOJ>LOxB#=DSz57qs>d9qc
zu}Aoj+Y)y-Hum7^07La(x<9%9PjT@uD24jf^TQEoXl70UodeQx&rBrqzF}-DB$x|4
zN?&PjR2twIw>A|oj$|OAw1*JHJ#w3dpAf{B!VREL0EUCndQ2IXR#_j{9E~?o2poR%
zR&gd!-DX;d?dDQ!bt8T<Jsb97``S9#8;2I27`%HcRVf;f;hzmYA#Jzv(Wn(=6ln8|
zZV53P^pZRIs~>5*l1+YFR<+;1@kaIO-{4FzmpDMRW=$3WPTyDq6eBlj*MIRzgd@Y;
zsKYrZil9bKn^%EvL^E}76z6iq7c-=2&IU%y<g^@}o<{C^J$d7N+X{I8)LC=j7mPs;
zGZ&5?AhIydNAui?2%*Qr{7@b&&Xdjivv8g_LH%z=4t^&Hs8Fn?KC2fJf|lS9ot3gx
zH8m-9HOe968^EinR-IhmYATgJ!@n8^LJ0J}m?^~p!vCQ(`zpkUdjsM<Nn4zhAcYHA
zX0NlL%EoAi%EEfanJQ(lLt6I|H;uYHX^1ikw@rmx8Qr150g2ek!;#+|IiwaAZ99})
zS8*Q_deXYyHQvDKNHzKajX+Hyvd$c4|Ki@(tFUYX8DFPveaNx6J*G7(D6gVrCglny
zl*6{&$ByE~j)2WMBDv_%C1E`i+Rx-N23Z-(v)3QyPUOrWrr+F1#5<(D?b}d?q5hmi
zX5rMR6W$fKDFOC&_gtK~H>CZ2+s0qYl?`PS6!Hdxk9dtJy;nNaa+K{Rwnm90s=sEe
zKzr~Got9g(AP12rsvw^HO0HYaG_%qRN&t;SzGWR*Vg+BaMd3^FMxS|G+xK<u3`;#k
zg)^R2O`3TFSRuiH44HRq=~ASO025Nu><p!!ol&YMUk8a@O8M@ceDza3aKoM05PLR-
zsg0J=&YxhMsyU~j-5zZS<CS(W60?D6x1?c`bbf;*R(vpLf!e8q{kQAvM+ZcD+>rs*
zZ#(a@plC-6USI}^R!raSN$bwnVZJ<X`>>~gU8~2xL8*?fuLt@~1o7bBW%vH^`}C1C
zHkYl2G)k@PmJ2MqFFPxK?%XjK{F-h8y?bMfnyl$|4jL--)+|=!UhzHyX##@QH~*mN
za&a{r6)?7N9^v=YDS^UDR-e(n2cPvnKB~ECyN@fXnjK`s%$e}EtS`oKSLQI!3rjbP
zyPSYWfLt?LP7?iOAu2OGO!1s!HHeuwiG`3$amgSElLbK66%nzCf<4RCOeg;a%El}(
zX?&|3(C>v@Yt7ed_mpAV;pLLFSkZf|4rWmX|M8?UTUuP~|5Ze!#1YB^IFzRBtgeOb
zRLM=a_Ez#Owpuo~$oL&$7mlHoSp0;9I?zN=%g`gs_q@yQJtVfWT4=fnC&IXVk)9cM
zNNfNz5cfP+#<bJBT<hS(`)IdzxHq9nZ3|GLI=WGLI_h~HgGS5pJH0<>T9+X(1{zx*
z(r5Cx&n_3VOM3n`NP-1BXkak`3Hzmikv6ojbpemOMarZGpl4w29dqgqsY<kM`BsL%
z7%RLwD;#Z9xc&))xv?n};<KUzZDUb_!T>A-r$}|qq}GgS*ZQIEhPA*7F5E(-EZY-L
zS_))+y+dbJ#rX30>feCkvR}kj#I+>0WNuYX-upkU9ZJnRr93Z&N)Tj>$qzBS*HbkA
z<eJHIxKJYnB57+o-Jf<`JhsCc{iiUoQBQwZ?Y6q~<aH_MAMfb!rnowfDG}}x>fQ#P
z`4ISnVpdD`8)}h+QfFuP>a)naMC)vCDW@qeCEQoS9*RDAbSpZeX2JchVq)7V5!+K2
zkMN-B?L5EpysPIB-C9{rHD1Lv{E2`YukrN&_O6>DRp;wbkpGpczHA|xJ0Jg{KBmfc
zHcVEqSH&0+c0BF%x*l*7uys5xw8pR>t!?+UwL6U5JKmTj9l4*mrepX+oY*jgFi94v
z<p+Ks_LQf0*F@H<2l7wYyoW0{+Uxd8uEqC=$$&(*aLea)Wgy@drwC{^mSX=&J{0nj
zA|=M3L+VZ=@`*eerE0L(nV!As)8`s?RE*R5&fDuP_x+%qUh_@4&%SPNSG6nVq{+gg
z{LRn9H&1ahEAtL5Iiy19j_SBwFW@rALYB3vxbE))uZ;suQ~>lmZ9?*FEp4*o-3Ol0
zb(D_}Yf-u8&Eokr#H&1VNb);$<xzp5pJ@EdwjC-W4Ft59khb4p$poKeq%0~~YuZM+
z=VtY0s_!*thS_OR_Ab}!dxhc*6xVhOstUW()LNV`)5**k4Y+-oa%{j{V*^)WK`9|p
z<B61-bLkb9H~-x1)HU`5Wm2bl>xh{RK;mLZ5!>$^(OPYGm7ZCFIhV>2S1;|ORSkM0
zE6DTy!hbV9rfMY8D2Hz!Rxbs!etrhb&AU^N8`-ws0S+QG<x1_5B!AZ8U%)56GbMEs
zq?I?6HjPxE3UrT*Hfw(;c7x74&Xt>Lw8ZM=)$0tR`rZN8QLJ6}kC=eoOjcipJ)8(a
zwoq+EAO0SuiyuBZHXhIRd*{fqa|N&x=~4V13$9+w(4i54@>gHSum>bWxEyZrfrt~*
zsC81?DsPuhiD#?<20=Ma_#r`<oWE{5->k8{-%2hzaDsnS*cd776W}w9ga+}F$E{y^
zY#{wF6HlpB1=+w#<@a;K0HOaDhCGVY<_9oG!5n9DF(ndZuR1Qe8Iw7~(p2o+9-GEc
zMp<dM*2g1+)=agvN#El;X5XVo%g3oto8_gdm&Yd?yLqqs0Rxr^A0itMu|^QGfdw{K
zEgO$CQBf$*qZ6-G0^OcgS-;o=qCOoKd2S^LT(^VAx%{t=aJ2mX>Tp^%*Q$hmv$qWs
zn^MotQ5~e}rhX8Uk0~5whOP*YPK?#M9|^HC5*ZG7)bmS~CSl4%>VEaf8Gy+w-C>;a
z<&W%L*Hs>`dob+HT^=~89P#nX_8Pu-HboHfiPB7N@|c=*lDVhNj<~Q54-Tt*B?YWZ
zQ<F~U_MA|0C?mC7!WX18vmzoHQBnWRNuET2H{UPMj<okf4eFU>avU#CR|HEz5S{as
zj#mZx7NW9WF~H^MY3TP4+E#p1_>zj-N|#Db@&hN312wz{lREPU@g`CTJuasF&q5Hv
zIj#E!tD!$8JOeFKiX;&8n`)J@rZxh!^8q4Vg5-j+F){y7xehF=;{1LKNn1WU@DI+%
z)M-Xnd#Bcq1!ciB(NsMQMYqa{WS*D!5bjeYrcN)5FqfI+Rv%=sv^wskZRWf;GUG1y
zjA<c_sooKpcKcd)uPi&E60r>nFRu_D6+-*!dcQMdzau!r@pwGE+FOxELd^v!w$9J{
z{~+Q9io!MD*Dt5!ya#FWeYR<20WcY?%0jRPCL)}qA=T92o$mt6J*%2VA{48nnh@RN
z=%MxY=?fo!Q;Yx0u<+YYkN?OE=Y2jg{}JC}=(A8ug+~JpvCeDIvcA54?CJeiO>^pn
z4T62O*=a3(mo25^U@K#TovrD*Boey9=whA0nYax_6IAxEt;25*&+$=nB4V#20TE)4
zC}I-<_j#{Ec`wilbs<C=*M5<xv!_GZD3*KOh>p!g$tbZKUH{A{@{Y8q8kkGVW7J^7
zuWU~{ys&;RsUMp=ajY{Z1MRJiy}MAhHFu?W$RcA&IL9Ag_zI(j!m|3PwN<o%p;<Lf
zJP^oYs)Fo3`m^WRW9BZ8hSN&5hAL!pIEAFLX02%EjD!F1>Z$9kSnPC0xaX6E0K(8Q
z6@=PIpJEuhxD>Y-3UdU{%ozC9E0mu65!HNNqcu<kdrB2$TWwnTJ(ro6Qz1n?veUDZ
zuhq{i_oyo8zPmFrHlw!4YorJx8dL)aShGkSS*UxL5+pLQ-GjWhuPnPzGsz@(+aZgD
zI8B{=peI(8hrG_<;vPn%Z@&bBA7)tQOh~4wO4boB{ylbUK_z03P$eD@k8Mik%@<;A
zN!B8EYTE(o=i#D*Z4%Phw4~5vy%l-s)z3d}a!j%@7^Nn*lQ6a$N(4NRuCpsOqM4NG
z8_LuQzQt5$7Np&AD!I!^3nQ3fYS=2$Yh(+y$fLG<pV4*sIQ3|E4Gsj{6To<&yBUtY
z85;pxCw9^9zduvp-LUjMOvdSEoQ&f%$0sqnJWo#eCf8$Rmt;pTwK#`7Pa}|@&{B|+
zE5eKTHDdhSxdV>xRmu`k0EoO|QX(ndWk8G0_*CR&`4<2sI8<8{NCg!~01oJs0N`z_
z=7BsI)78gbM|Zz=UJajU#W2gY?aQ{n?ju32&qR$A?nGDHJn6ZLNPsSI%{b#PbZ(@%
z6}*vUb=Rw!;CK<=<(;s`Wa2#IWK&#1ZR<<Sk?YM@D)_Z6{c1FHFJnwmq;+y9!6#L)
zV|shJA*$WgElhP|(x~<ex$h~^fAcBeN7v268_DMv#Jtk)=rT%Q*?!08mJsBb`#VIm
z_>;u#p?v*I4CdY45xZ57yi?A7<nQW#V0D|)mLLn(a#LZ1S8p1#)NvW3D|=)s_?=9M
zZ2LGnpYG#`D#*zStjjJ#o_q|}92Up?{>?PkyH0UL&lzrcWNGouV`5-)z5e0)QtIKR
zp~g4DU6GUAF|i11o`}uZ!64<LWHdnd-oLcDf8bijDrCE{D`P{4^|A>9SO4$(uky*T
zfh${=N!WCGb9J%zL93rNDOca^uV5iY6ZwhZxWu}9r@mAANxowSv0oDPACGS54S$uU
zkJpBV)js1m7xrU{HdW=71@7o)Wb;0wmD(9AvNJ5d>o|A0V)&N&JQWU{2!nC^9QgY3
z$qMV}>Lc^{0QN%-1(TV>);3j0fbc*|>`|}Z>je$P6$lpEn#r^l>Re+kzWVWVA1!zn
zWoN3mVwd%qcFJuOi^&>lXWt{6MZ|zyJ0c@{rj<74Td{C*8C#$H{5qDNoCYp7*9HpQ
zZXLc}j1vZ7k~5><R;`a@I<7W<T%RIi`F(jPWw*dv$j`<-%wx#5IqM`v5)(sHxQcy%
zx6y9w>}7*Vo@z*GmHKo$9Xxi&JHEEM?@^PQ-6Wd(@9^RKL<ZvD7pu<$2VJi+)fWzw
z08kX1%M|eA5E1^zk<p2}+Rn|EgZ2D2T+WR1ct^*&t7(+J9WsM8bj_;r6KoP0_@CxJ
z8XR|&TxjxCs#~PRfd^2*ID3m#O|5hmDL}|b$*tSC4HeP-@~IZyxU~XGztT9=SGUqT
zX?D$H;{`$mQ)&sh>2u_q#}{FZ8^Qo7x6O{KN7$zN1M2d|_Wka7t|R#V@FZ??VeXgi
z#u(S9sz=1b6YH_JzkXy@kDt0~3VGjl9`00Z+1djRv^E3>Vo3N8-P=53(tUy-KCS!E
zB_Y9NF>=GgIKM;;Zp1PLU>E=Y_u|=P`&ymY3OhLgFfa#3O%JB__1Nj~iTVHBAl<?s
o`)^+P--rMAHsZVgzu&(FIHL}-lzuQXc?Wv|WmKfAB~1eV57#WF?f?J)

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_003.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_003.png
new file mode 100644
index 0000000000000000000000000000000000000000..f0f21c6813d8d2c28755e65606da305473eb2144
GIT binary patch
literal 71477
zcmb5VWk4HI*R~532~vs|D8-6H@#4kZ2^4pCrx4uT-Jv+e-Q6t|3+@sqR@~v_`JV56
z&-r_P#D<wkn7#JeYhCv}5z2~E@6m|S;Nali%Sej@;ouP5VDAF~$gpSfH^DX73$Lq$
zrmL!hxvRU8vl*O%k*lMvgR8BTG5Jq3XBR66dk$u9W)3EDOIKG%7d{piyZ_yR*}>U@
z1uQIf2s;U?qqLR_9NY(^|2*K!gi5X8;Njq8#6{FRvQJ*1b5vTX2VpN~UCg3A+C}Oq
zTE9DfS2X1woh2~VvN9Pqew&3v=O=I$ZhyIVk7pe~ktKXD>D)#>N=?wv5H%G0_YC{H
z>UTZ8g<{nxji^7^XTC^6y6$n|+#D5*j4p!*gECRxVqsCocLI0WPys0(D2Yqf>e>NU
zOWTy4z01?x%HHNR@+9o-|7<-yvcvrQKYxd9%zc*?)c${V{_h^p-h+Jm|IhBcwY+8j
zb1dvMUYNH3KM&uU>m1XbfItYQ%yB6O4Q{xmPKh#HV)biAj^;`eF30{oJ!ii?SzWoj
zx=$ZKpU-orrXM!HKQ%(1RRh|qs=k7A`0pmU?Ct%!ex7f92j_l&63@%XyYlO?^98@1
zEyE68Z+PH+>#*oK$@XJVUbN+czh`-Ke?xlPe|ybyItIUwS4M|u>6ZQb@OF{*ur=`E
zVR*3z{2=#H8MvL@OCbyp_k9-+RJ0c*uajm1n_AjAf+>Q<WJRzfgGHi&@PRv52=vro
z-byz*YSezLVmu&Py?kK<26S!yi3v`WE*T@i$R7h=RyG)m4h6y+%|>W4K8;3EtFGij
zN`e{C>#pGfpjQDRAW#%-l9nq2DT>8y6f&s@3t!xI-a(9VC=`a*+AcnJUZuT#rCGM%
z#5p0;30+3(c_DjNRm|xL#(HhZo@w>wul%N2G3Gk%ksV)zoa4|@M_}?EtS!0A_BQl(
z(ZevknY46bL8?C$fjcd8ai0?8bb#x3yyyO+pqWpT7RX=M6}vprq+Q(EQ~M_q^fj^|
zV`Qah|3$S_ak`U6nIq2K-F@OIN73(~DNWxSba0W@^Sp5BE64;VJs12o2fLL2Fu)JA
zt79CuWiE$~0yW<Oj820)I|vKbAvZT}J3I^;(^i@BC0e{t2^HzC4RVZk#4!@VSR`jH
zL^mUA9qxJS>))$&+a{s<UPy)S2aNLQ(0JX<lX%eB)J>ufAX8bk4(k@2#O#E(%CagE
zCfRP@u3mTLU{E2BbG?aa-OJGC(PHTrq@hl$=m9=Q-xZcA?Qtjae>D%Qg*}_RLKBa?
zZkH0^wv0Q02pReb(DR-kL-9Ds&&@5v#R5EY<hH!NY{W{U;K0lJ^cg1}Avu^QwtVmh
zB96)XQl*@24ZoYG?C)N=Pj_cY)FzoXd{&8nPJe3Z=^gNIzTgYq{beL+iPO~A?<aiU
ziu+_ud*|d|yh-+N|NM2U9Q?4BCzSX4pJ?4d%;jqu9*VMPecGZKCbo7PX+u|FHdx<!
z-bbnzulr+*9sH{&KBosUI6b>{r#l%3UYe<&s8W$_c&`VFB0{JR2x=9efRm*UB=?|u
zrVH-ni%E_phl{5p+PX|SMo|{VQwQt<u<y^4Mae74=(LY7jT!7A43$j}Kly{R*Yk}1
zc~F3K2>7^&@J@IKNg>0g#5Pgz<}-X^HdB^#hopY;WtdoKqGsh1vAY`-WPaC}X*FHq
zSi!H9&IDN>l%bWI`h_wT^aD<!nI)E0GsE#m7&f%HeHN&7N`hZVk7|_!?__!?)@a{e
zH<6XOJ6=ABrH*vn?eTaY)@2D`25Mos<IF4gsgEyjVgxSBeT&__eEs6RpqhIn)pw73
z?h|<RW_40ih}1MmKZrDc4IX2WE=tyfLj5nlA&Qh?iRL#Ar;i2b>+0^w89X9HMagwO
zU&@KA48n3pmww#Ki_gj_mkJ)vNXnWMeUaKmfTk3?B2mmR;jS`6zIsJneP_QRYyKQB
zGKs8gKWh=`4=msxdlpvW8|47uEJS*$0HCIA+mw<ub$&fEg(+%xiO3M=JwtXPTZy+W
zI0WnuU%WR7EpjcMV^+N4Wc9fIIUU&E?~ZquH(vsuBh<3>t7L#!tU;x4^^>FbJdLpP
z*%}c;fT@>PntjcAjnpsJVI1iIwMeEpaU<c9_dtgFNyEu_&o0F#^<trui$1CIIbBQV
zn-zT0wRH+L>{jfF3H;#r&Kn57boqVL=A*)G^Es{*HiSqNAMU}!6+Hd9KcR#u0Bxe~
znG@=LHl3&B_gXHjXXMc1mR7-{zs2JBO3>AZUDM-~a{fGCVOv?lLHPOKMn_Al_XWd(
zEYep*2&y9{1(R%mOXAoK9oB4*y-N|l77}V)e~Kv5tWvNlZe6nVUxXH(=e`+cF{qpb
z1<9~=qvGMu*h7+&Kc}R1ZEfJk1Ny*P8LXDFc>*jT1t5ObIOaBJQJV=fAWKV&;P_ki
zz}`CzX;k&CA~AAQgot?G)k>F9Ve)Lz5Ht%WI!8)F*4Byb?JxjFD_k}W9Hr62eWClk
zIdoBTRUpWPVEcKzPdloPfE9ouLKmL%_SI1niY2<g%3BkZvD!m++ZyqAIj{>?m@}*-
zL^Tt$A266afJ4ZE<PnmuT~%9G*Z21?LmIOlWhCV+rvyNx23lxoM5t1(R&pz5SUF@Z
zmugNWlgUpQ0faXqq}UH<n(#*m=P3fxQAhO0^TZOy1+lYJ3T`ibk9e>Q^*C%(LeP)%
zdQrK{yB0uu+Ya+Rw;B;^vomHV4)))66fY!ZBPk|Ka&mQ*ZkRFju|smaM)|^!(68Oc
zBjW_lPZqZcN<=VWf@l@6j!*l9Lo!cKzcx3Y)it)tcR>vQko}6e1Fcx<I*jIY+_kws
zVJ=voxbwzMq6$aKxMHw1EqNiQ4gEKfL<q0HLGxaac2B}h@k|<bEF3n>p|99)GeQpp
zpGpbhDa{24>*uwF0ReTGc3*)3$@f^*Q%Xf}+s`7%Wc|(kDYU#8U;0_aWmPc$8M|p@
z;|b+ITTd8aDPl`Y)wmCWZ|x)pFmrZ&s4^!$PCMTYxfS7Ita2}J3hm^7H+0)&(qQfp
zX=n;?Gpy{@iozublb{^`5sU>^DiSrU60s3_Kn4aV11PZj5xA5n@F?_DO71G%cy{lK
zSAB)K&ajqP->ChJ#proXNSvP_Zjk9*F}?tf=1#04EC78Jg)E^~wqqH&@NW!h89lVT
zc$}9;g0Sr`wAV?~KC>n^7GHq0bL2zg=)MnoLNAYFdmBj-&*pIo*uO3|pT>@M`V7@J
zScZj-VkAutoVMrTvvbzFWt-TqS{JmL&TDF>N~{h1UgS993Tw4DdO%k{K|&M2gIh~R
z;8(%l<l{0cL_xLYKW+CH9a61=^P^|KWxY>LCE)RUCHcs#Yv|;Zz>_OZ7K}CMBwsi)
zsw!QtYsPOeHp`~{DRcn4a3-8JT2oh-GLoT4MlzKsyD(bX(ykUlS^TzP-XN10eYH8^
zG|d5URgwY=9yL({?;=J(XeU#?({m0M7cbK*(H99xo>l#ZJpI~YBq5F33fVza%QSPD
zvLqaez)F`v268qcS-Q%s?+!y;KRZ63t#^!!Um};PPsc}dM6(I*A25OWe!BkzK~LT9
zR~EYilM!q}grKVQe~q7|DJ%-OluTUwUcDc$Xc62XsEG;h*h$1B3*mL{#LhMZcW$SA
zUUz8E>sMCxo|ot7j#wcD<M`Ex@QR6f9%S7b_fRHZi4$H-BSAPCqm{9uBUY{~=PiU|
zCIqS3^NSf2q20chxH@xjYUk_~O$#5h9Ef$0XBRRv6%c#i%)j~jPG5|H<ME68GLKJ0
zl!nlBe?NWPwV7|Nozv`BN;3PoUncZsxy0zbQzdjXj4$O_HLQpX2uU+mn<9x6DdNEd
zS#(Uq0fSakH=oEZ_8nf(enkssT&4%0h)T$UBg2*Hg6Dbgv~{A{=)R$sz>3$>rk?lZ
zm($imRKJ56a*`Q87kS-x+5hZ=SFTdm($O(7G0B;lQXq~&HX=OeosfDipexY6$?r%<
ziJVu=B(9fFqB2ROqKK`Sx5n>hrJCTBgl_o@v>V!W#|Yx<RH;W4SmM-O7=@M!a?Ksp
zj1+?uFIP+8^^eNPt{zL_gXRuiI=ptKeFzu}lC{h=H8r;m%ceX|4qe;T81U8AQ^%tB
z&(^(OpF=6EQjMZfRw1BX78fBll9+fPuu7e6xVEKb<aD(~%V48Ozb2(z!sxm6_-BXr
z^^BBm$1TFK4Wlnb{<ly{&%c+Ju_$iSl*XvAf$jW111l;Um>-FUO+IFumyWdYy8PL5
z9ovha=)A`(r^e^zn`p7}?5|)05KkH|Vn|l7K~<j-HaHbUjXoloLIg{Z!kTy4$VR}4
zl?)Ci#3lAMf1>9h?JYW|>O-sNK8b~rL!vqT>F$?W0&FR9C;UQH<U(eKpYwn5B@3}-
zRt^U4{04J$99(X9psgM^E|qq=a17-20(g8GmRJb*bVZT0w2P6>^nbs=5l{<k&l(6b
z)2i`d#BzR~wK@md=~Bg0e43bWU9l7NeuDZU!JnMiCeY9?n18le`;rp}MEios)x#uS
zBH{rK%l=PCa!SGYMTYj-*y-!QfFX0FXg{l=iAjvjVs(96n@p3o1pSyJ&%fu7r-h!r
zq^g!x6KV|iysQ)T0+gjQD}StiST+%$4un(5$N=*Y5+##2weHX9Vl6rUUVD-*@!EV>
zdD8OdqZofUO@bR}H1Xx3OzRk<XT7a#P0ec(-bwT8d8<ek<_Ta6Rj+`3Ptn{F&Ip?Y
zn3Kor07EKF>Kf(U?-as>0IpDBndmwIz4EvV&d4s}Kb1$iNWvJ&wGO|y322Su!Ms-V
zH>z?3i$BA8yH!<og0J@`1Rpa6*VEp`A@mXzitvd4I}u)A%e&J3KclN{sYZwujwLQU
z4%lV4U+3x9>Pc91IGH_i*VNM$SsEL(Leg{w=D*wK?6=CM`_oAYc$ep%#vt|J3nxvK
z401n4U`0NDR<N1!;M{^bywLZ&J^2le8>gt+eE+FCzO%z{Y*V3T^R1sw@)LT0P4h^Q
zU_a5{CO{#hbXvg_8@JG-N$7x{zRxLtWmQjjJUqCV2O2St8HVF=?QpeFl@&Y!AAR%D
zhVUVh=LjJ=G1tgYWHouFOp5*Rs>$GmmWipx1b1p<!>!uyA1N?anKFO#n2i*+c_sUO
zo>J-_zQni-em1(J=sTRxn!&PlE5>;uk3X_p+1Zp5Ua@I^_YOw<;=FkrYb<p0JaEdV
z!7rp$&ASz+nODHgTN_)<H7)S+j=VADWB^#i?Oq4-DdeIE>6cc`q^mJ395<xYw>WI(
zg4WL8PdqtvUZi9!rlzTzZ?5Kba!*WtIiEhAIryj25H@=!iWQ<l^dn1fIz?8#OMic@
zt)cVv@t8A9;qMJGm^U^yRyq-}s==H2DxZ0ReU@`d<4q-ZSWae`UIz2zcM5R@V?~)T
zJ8_K6Fsk;kfOB$!SpalvPoSGF`P%dI?*#^})Wq1l_hl}V&Ja=*V#UBlEq%RQNv}zV
z`#YaoyO)Z{C48$nWqQvpQjKa|BMXZ(`<Ceor!3Dd@bCdWvtIkC6qy0-6nmg>!oJEv
zU{-N50tG7#wG0q82dmW4p(HtAwif$!Qx_B5;%7oOZNo-M8Ht+B-5ht{K6DM{y3Wo$
zbi)&>VjXo0!3q+t|4ke`M{?n2;aF&PxRrBsP{xa3tP@-J<z+!*kkMG!^w`xoPf?xz
zM!$CDH^8%L2(TgTQGRqi<=v;=_Rh5>;^837kr9*=+wWV$BV@4ZmzF5H?I$RRqg35$
z{<u;5YcfzgplFfToR!e=?)G~#nJqTrAKQfpMv9biN*$dJwcl*(w*-E-yPMnN==`0J
z!|uuqy;=c|!*4&^3i4#X17s}>D@7Q{8>Ejkcq9lHwOLo#6eSOBCvQ*rNlsgMJDf5(
ziLfQGMWX=-Vwo{AbHQamlXR<i1U*>uGF~z1?Yp5Fh7bA?#}Hon_F6K2aN9B>NWcF$
zrg(6a?EzbR!5K=@2lm!;;)<8Pcn0*7pXV+waV93BDl3^(YZg%l{4YGBA({F$Qi)X2
zBZ9@5X1Uzd#gh*|b5-{qK!TaDeo6%odnh5Uoi{6p$H2@cy3ca=<NF71+%h<N5*0j^
za&;_(SyMH>ywW;jFLcK{Dy}Qv3&+P>4t%**8EaV2!>&y)k!W4;CoW)rpSrRxdgRN`
zpFaVG;jnbZMjF=!Iwo);F*+eM`D&E!4Ik|yE&^8<*M9r-!fs|?wZGBDuiNoRIb|Nj
z6U(Mu7(x|kGH*Hh^7Nq<bj9UF(y(NmDkuqvaKnMhXUuxjY}ubpgGgT<k^=YyYQ)Ov
z;wGbE5tQ7tbHa9_W53ZlGQpxRDwBXK+PBxv1WDJ;8%p_DHR`f%Ha?(2YC9?y7b4CO
zg5=7z<9J8vZtvqCiD11z9HTw@D4BC*!L_`$R<wSG9g1Js1e-GjvhrnAiWW@FoR*N9
z+Kv_}LUF2HwjhzPpB*tLDA~LnGEbIa+;1m$n*3e?9HrHd_~A|62mnV&YSHn|p3X6?
z37H)ZVU|(lFnvZ6Gm>oRpi0QLW4H^dyF+J05=BnV``eBuIwEq!!MXv{OmYk@U9H{w
zG6h0J9^p?hSb@s)2A@;X9F=6dXjepFh-))>pZNv#OPSvf9rg^oDk<js2pSamfJ}<E
zkJx3Bg_6ZMF8kwV|CwbES5PejgD)~Tn3$N-iNRf6-)A|$ifmZ!_37bBG-_9o^!^}>
z!@a$Wg@>&*UoaRrKPhBR(FBRa`MQtt5G9n*Z%NNavyNl+Pj?_9K=G!`llJ!|9=k|`
z^q(^P+Y9*aoHVypgm)UGX?#}h+Mm~+e-!JHSvct$xx1&Ke`L;2Hg#~1)32EZgE67t
zoZXO*%p4@QN3MLFe}Bvh=$&Oxh-7{NAn4M`BqsOeOOb@pQCqga;^d02rt6mP#xsSJ
zNQ?xRPJ(2l=qSC#_ZE|pv0}cWDe)MJj*U31Kv_6VW6XUAoh}seqf~Bb(>Ggs4=@cM
zpR*I8>fPUgn>DQ3D*VuVzj61Z#^`IvcigL0yApM_j|-q)X##wK+bz}8VfyuQM@)e=
zOI|r_PS3By?eVsrmq;UK9fRE5ElfpS)9DfGV!^GosryTxf7&t`=Z7f9zpcJ^(QK|A
zdI;rZ3)b0!O(IR&rK`uYN8Z8lsXX!dM%hKKk73UjAN<nZ?(?24q_HU)Mbo`Cf@}W4
zCtRVpDq6~akLyKc#S9*3LyUpP#(@x*0gwkMhKa@bdU5`3Z7>JT-w#kyr%1<Z%2YJT
zQ^$_Vhoa#l>YL7Nd$)1NPFgo|Xl^?KDFegFoy*jOc9_6+_IY4IxnTQHc>{Jc!H##H
zsbF3C7Xvsx3Psxn#_Xg#d7CEUDf67-Vu{qWv~u+deSc25i9_AYlKrYb!$4VB)@joI
z4*exllx}Gyp(3A8=T?!;$iRA9B-{@I$1-^cD*EFiiU=12DHIYdQe&Tr&=!-hTa-At
z-NH*_kFM`>_ttz?BJV8N2+Z%3`{jb?t-pD8!7oV99DgWRrySm8J3c->ocu!;O^AWq
zdAH%^eXO&&lcJ%&EhY2_V+0cbiX|Z)Q9i$oIalY`5LQ(v`#^D~PK(3E<phgrmAdeK
zcol?5mj3qwADPcJ8!S3NiVn8qDHEy>sw!Vh;2z{-|KGN<#qS2`b<1G$mj5edaB}Cr
z9o}nJ(<hOz{ibx0O~<=4L8%cvMF)r@5(wfsMQ9WG9+&)w);INVEk4FNW6F{vchxVI
zy(__9Ekb((cdvo^?jH)@=lq<Ujmq2oyC0Rku2~{<=veDMn$IOZ=Fg;RYg?%j@g<Pl
z(8wsj$WJ=eyhuy3fM4^tY~K3%%x<Phtoi3HkINa!3RUOZnU3vd&s2yU%WiNlfLCkE
zL~dK7ICwfj4vxiAOFmHDT8O+z90{9}j?)^EAsjV`VW2UFd>|bW-$V4Ts9D1PLW}^l
z@mPJbbyGr=haz6XSA=qe;CIo;6e>pIc+;}x-UL^QHGmYQdu`g%yChv6w$C}d+Zd$I
z?>U}3lGaIbZFR`*?tjU4<M#K=4c;3Zbe4cz9|zsINT+2Zi&Q`ED7~JrnCpLvNz52=
z=wwUy3VLZ`0o&IAEJ)o~f-ZeY7RH}GIDLYuYyY}Z6AwscnBTp!a2bhk1pVdQn=O{h
z=4%>btm*9EpGYH&@#bTe8E)jao8>fPJxnSgdhyL;ny_rr<|IaEVrH&yZy)x^c69YP
zX%Y5(-(X%Pk}vvuffy193FVxUo{MIh!0Mkk(KM0sFXM=6&a|G=*pj7KT1F$=aq}fm
z&7(w9hO0j^PQ>w6wCgWqup)DRGi8%NB?l46_eo{BfcaOSeQUV$mn%6A-<PS2CQ=oz
zo;uxxFg9i?1P520+MNk4!HB~P_2vWEeQA090EXXr4evT7i=Ufm%SGtr3XeX*NsWZr
z+GkTn8pHZaq30R0%OJ96;+oE9_VdmcOb8zWzHyF`n_G(C1<NZ^)+d|nm9<{2FhHkp
zO<g8AV06}EAS>tT3Obi3lz`F-dJ!CB6^=jy4sPXdaC%>H>zg=~Dl{zbxW9F@@GKoW
zou83@vN<;6!R1!L&BLVl*kVk{qqY2{s24THIWgynyY*panoTQD<jvnTySCdYcP3wf
zowaBe`{}6z1Nl3I8=}HM9xLX`ZtP~#WxI_ROrfGn-4I%+n6;ksH8vKLSAev(gM+3T
zF`HN8%O{kR&!_(m@?PWRJ{k<t|EO;E85dIpNDy58bzK0~IdmJ3?)_fDMwTnA^9SOm
zYBM$xK&<ZWh~BTb^6M1^lcTlXUfS5^@`h21AyE|0(9y2urIW-U;vr}-IDQO1g*7!=
zE2R;Cz8`OT!KmzZVW`<2of2vr$sb;ro<yaB2P{{!<jP`VVp6VIjMz<|u}s!j0{P0P
znuA(URX2*Rj~eMNTR0a?Gnl`fn_|uu)CK|>;tu42f<sh7pBHdMlPf4vN{{>Lu75a9
z2b(8I&5&M<kCm;v?%CluO>nHl5hBOE`x&>MDB%3CXa-Cy?WE-*jXS#WtgNi8t*@Vi
z+!Batwy!gj`8~72YG~xJu&`o8$rgU{iN?H+cOoQVh$6j~oa5-GIHq*vp%m-(S?lX|
z>Kf{n&BPdwVsB62<J=*qlsLh5#Riqqh2LrIKY1=j)h+TC4ql%huAxeZe+dXO%l;8k
zPz6MmkA~k&A?btmkp$l|cYEI6x@DkrecBT!=j%HzHQn*JNEfCC)FO!8cw|DL-vUB<
zjwp^vsqCWhBNAW1y*D2=;Ik5o>HATIW+dTxf?P6XzZsgDVN-Cxm|JaodjQrCh7|u9
z%XHjCAEcEXU=Vt#(%jMAk?RUfRv;`TCK6S1bA4Pz{ACcc*8M7I#YY*LHQHG}WrQ^<
zv+>=a4_Rajgsi$o6728`he4uB-OLuaV~tOb6o%5m%eO||^f|GP8iN5KIp`EJ-E~W?
zjjrJNA_$a)Jxws$>_=KSWAEcj$YKE^;)#EC$-b-|+-jX@9nL5)jtDK>3E&#ij=v7}
zKb*3+!ByTV<|dOMQ^152Mi$+I1#72=OKNou4M%W&LxWh$>h`}wtg432ogW+|F{~@c
z(hmiZ_*`6Eti!WMu1lLjTU~b>Vm4xUKb&iqC}V#Ym7DVra*#eDpy!k}l-xBR-f2zJ
zieSOQiSYd~=ieI(gul8XhHs5hSHZZnN?IS~heo6hBd3MT`4z>_UVm4W&C4fZXuS%;
z3RKLvSWSaF&)~bW&FMC`Cs<HAo~`k2c<jFSeZ9(qmCY)$Y$TX6iFhV7`5Lk^ncAK&
zq!-50i5LOIuuja}$|>W@g5_Ty3K=Y@4}W%Sk)+6mz}B}oj#BP&H88ehJt8(_DMK`<
zw60CaPw?ddcM4g<fNv<{N+J0-78cgHzdS95q+P#sZL?OP_6^lo)}s62heyktgW~03
z=zK8?QgiI3=`KnCgpgOUnF%&HU@EUA^rCfo=9NJ4L-`^DQ%&WpbyGa}pW!j{tQEnh
z_o*K>c$<84n^N3<s{~JBMEYB+Su82l5^Gl>4ew%;dEZlX@OM-)F~!GlRI)KOGE{zL
zaq@PRPK+s66P>M;OU0l;(R);<AHpy{-J{ICuD^<$2Snf_!~^i})fq&+c)osy6mOAm
zO|hB-`ZZR90}7V42Yxwa3>nJIVg&c28rbaz|LeL>{-+&3TN50dTqgXh|3A{IMN3lb
zlM472S7vo*X7hro#iE%-bybz1uU&VGd-0cZD?3_qeSQDqBXLv|Iw$e_%BJflpT{7<
zfo)iPCtn@1yOvD2s`j~8Ec!@A%vg%|Y(lrT{C_?XLIvU&qA3eTHHgf$LTC6s5`$?|
ztV;EGlAd|VDE<Xj8{uJxU@O@nEj@CMTvDYN@riN_Wo4yEXRH^G_@w#Oc+o@!&^FPj
zJ%Fz9qIOLT^x9fmM~9+5V)t{WZ?3NHzP>!^c6(>Q*e4bu6AMd2b8~9x+~?i8#k>mj
zib1Dd4*^P|xIZ-Sn7q9B*LwWY$dzF+v_%3c^g8Fpyo+n~f->jCD$%B{%Vh7HULQBC
z{9pXt`6~40_lzVvFT$4`Egc#u^ptc5I!&2&f<l-VWK%v?=k9)VSU0zL3NIWGj6$yh
zM$Zw7rJ6nDUuP#6yLr|BvNsY1AmZ2oC3i6k`~2n73=G}cLKIc)y{{?OU(?ZT<XkcD
z2r(oP8O-b2#S2sssK+ETUoWtZG1gB2MD)>#(Kre%i~ILmT);=PwBjLZ0s;nCFI0x`
zr6tRmV|Q57g`DTL4jl+_&HZD0i9U%IeWl=Q#I?s%+>4q~Baf8v?o`+!a}rV0JAK29
z4X922E8R#!-&_yN;nP&Xq~6xk*#p#0Uqy{4wE#$iVFA+28yd&|$T~-I_3Lg*P_TTF
z%MLu%Z9b_M#q56BmG^v~)NItAwm)VW3ZZTDic`W~URE^Bv*gSAE@{OMTC-)I>bxyx
z>E^k*Z}EFNPrKvhQiN@}o!-m=36Ln>d)z!0ebdSWUtS*^Ol_lXC<<&)3_DH!9cQRg
zPg+R9hvlwK3PYa!xl_+G-9%R6xbe10Z5`+A&UbNR_0hY#CbxHYpnqHNr@LG*6^=D}
zywN@ZA~RRX)J#W-*ezAU5aP>`^Kg~fbMR8QC~NaAE4XjY01EJKK@j|0%P8~{A=yW}
z_F|DJ=E+R`o`gj;N%{Ke^zP~8{Jf#PeS9bi+U9ml10jx$>wpnKm)jGoYJH!8IH6a<
zVyY(X5YDOuGy2^+mLe-oYZ$FX4uTId%Jb}s<svG?{?93zx`tq_+{2ZSUlL4fV<%3s
z>Uj^#SkU<lHv)oBgHGQJ{126~-{lSd?-H~!lFBN5MfM>9ICFaKasJW$;X%oP;}8k4
z-?+;3&-Bu=Ago}U+YeQ^8}vF9mphtyjtENc>663Ehw<+xk38*tCK0UqCU<GVg=HR>
zTuCgaCx{eAn9tuBYExixX50}OS%Jd1p@)3wrkRmhOrPcC#w+#{O4um5_@~`D#<11O
zswThyoC&*U7v}2f9q1^9N$MDs$D<Q0i?6A7u;G?!#BNYR5^OZaO|QNh^fS)yb=og1
z@AW9G(+LzJ=_Bpuk)0O3Dic|-)yAVF7*iPGtLy|b2>ehJM@1`OV0FOdYUyETr1|E6
zOZTtr?_$r_1JZF=t!h9R44Z$owSNklYK03`rVE)-b8gv2PIoVfxrgJaB6`$6!|=<s
z6G_^N8wv^vOtX*cGaQAnPwkw;In?JS{URrh?-@cx$@;ZF!Ue4NbfS;~==cfodbBDQ
zur7~kT<G_9T?*oOV&fu*zoTb)FB*I}WGs-=$69)KKq;C;L+|DF^q(^LsG*eC>lOpn
ztzc^pY-}*YXx7fnqJf2<UmPsh;C?TlO-!ISu66-;hOZdKW|@X~<z{~6uiLwvpXup$
z-m;4Vc3J}!ITGiA6^pz}LWW*mS#z)@Wx=|aDelzD?!QW@X_vt%p$Q^E6dsyHFDuc-
zh~FhwVYP53AC5tQMa<T1m0kDa!XiK{M5<7~aypEg7J%p6qTHB2Mr&XY7w8cANiti|
zC$x(HbK*t3e$Bwy8N62NtTi*NvKVWda?x$Qru2GR1?1x_#Ixt~TP79L^C|EyZEl}+
zWkvIikera!cdeV$B-PcH#jC4Sh=*GrA|XkAnZ-ly!iLpW@S86I5;74Bu}lkq(W^DV
zQ*vxm?WAI&jG69_`ToLyK%8P7!TWz-e4k7GhF>~-ToavsBcZ#+pnaA_);wlq{--z-
zs#(e=GhDkVQSFeQ+za@Di9m0wcz?aEeQc|h0W<%6cVT=vWx6`9nZWi_ceziMlqKjg
zcpoyHH9A4@nZS-SM=^|30TzW6;^43-Ld>vh&#wBizB_~*JONWJ$<p1F-47o=9J#jB
zmGij$#Y>S*v2HqGB;rsok9H^Ia`=~PPy;+rY1a+ev{c9}p~L^B=0Bz*7x1Nea5}Vw
zPDszGf52*Cx#!>Eq25I@ibC9hjG&z3J=H;W+}_8Lo+Wjg&+GPKIqQ1r5cO8)eKc<S
zwfyC1%a)G+1l^*(R}4(NfdNvO;>Hsz*LT#_<pYB=d2*-e7asCe1pW;e$nS1HVDbps
zH`Lb$vN3NGNZ?75{ES;(U*EkyOb-bO(cO%xk3aaIU<l3!Cobb5`NGh;jp5r?x-@55
z85$_eAQ-T;`kh9GeMgqFNkx!MPcr~misIjEvyaDU3?)I3psvQ2q*y5;Aw8|@Xqj+i
zJ=j@BzOfD>M-xHT1#N$WwcJL^zaW~X>DObk$;xPwL8M_%9oLT|Ypp&xLVi~kROfo7
zKJx>FeU3{Mzj+8gngqi2(-kcLc699wY9b22F^n+?*+Vpf_G+o3v7sG-01)0rE@?9g
zshteS{3nIitG8#ri4n4%>7J+cw^*;$mhfqM2A;}f!gn$|3wt@Z{$F1#^7o=IJCgO^
zWM%yN?txbkkcGJw+sB&=MK8Vut?b2-<*^4#z8_xBSxRSt55d;2VeamH#e7eFN{_h)
z=IB)+m#>eHkCQeQ;Os0=XJ@DO#@j}XFU$p$#7<pUp8S_(0|fi9?I>{N;j-%rGAJLr
zqZ&JT%I#9f6JWc7I&|P~jkuUnnPS>?rb3Dm=L+1$x3?)&_EhO*Wf)S@Ws5~?7IzO1
z%c2j0jijlo502>M5{qEPKweIc-f&@XpX2EGI2ufQ%?FkPYrbKLvJ=M0;3{NN6DAo&
z$6)}rl>;VKHA&~q_`b+cEy$+VK=hVXo7F2k`X5!b;p=^gG(9*C?Ems*J$8?H4~X?V
zaNFSa?HBM7(G~xvyY}<T3krJ2tlJ9v7KU~Ek5H<xuo2xJ-=Hh#w3D~@B+OVM3Se+T
zYNI2<c*H<?SJ{LpD?-SwI2>#g9d1hc?XUCss~#7Zq7xLXSLqfU;1FPzlkc;(zx6Nt
zCI&YCg!uPYZ%nrTO*z5xo&in7^msr-0MS+m)_crTZUT71-{1P$t~R!{Ojy-TFdm?2
z0ZJN|lCnOO6umMevr3M#2Na9ZUF+{+G5OUh@PkJQ-{bbP8oRj#UGSK_<xkAcGDL(H
zR@pJ6pk;}KuvY(4iyjOU{g}$cQ0YuvhP{x<#9Za*a4hIfMshV%B}R~D!XP6V|A0Sq
zIg11H&jiaFyCFl!ozk-{Pe=O2eaSI8=X=Em#(r_c%y&n)Q&(U7t;mC}G*zs1xr7O|
zd!_XWV*$dFi1=JEP`=1lb1GE7S5w<1H;piSmY^bE)Z=)#^1DjfUwiH1x;VTppKgCl
z*QkaQ1&!cS-Hmlj0H#onHpg@ZQ5EWCBCnH-{XPtYD=j79a((Aj1iboR8OqShWYX|h
zu&IA37SD<%zrc@`ko#J!CR52Z%u0d@r&6*WC|vhPJAc_yrD7qxpcM}r(!RB);Nj{-
z@~-<*Nj9$lEik`x+54&TX2tE+N}VQ7K(V<KC2mYDYPmFf0-rvo(Dkf;s#7Mp(DZkG
zlegTj5a&x2(DRrd!GSnw%aD{|rI?AFUttXRK{fWZSEDPJ*CjU1z%#_}O6ZxNfZ#6)
znft@Pfe&pzPl-#F@_sEIvyospL4Yv~Ow9zNiFi5Mfsx~5x2%M>Fv)-%?eKP$<nyV^
zML!Z2<}Oo8S-;_u7&1lBc6wnU*UPIvU8HirU`xCo0oQ*)Z2v}0Tc#`F78V`$(;udG
zyb^rTSxbQLzbu1;<IQIhuonC2jRE2~H+V{85O|=A8L3m}iR&wsbG1MXDNqx~QeR#_
zh0}vcP*cMQz-XAvO&e>R86$xi%h$*+B@WLt&tQ9$+@x9>QHZv_!Im@d4l@oDo1tXZ
z*70=~{kq-9b3hD;Lbk@Uv*kD{S`j(Mo;X07adz?dqwnX^uUe^wO}D=+ys4wz(VBI;
z_(|AV;+nxm*>Qk)bHLVk@k9~d{ww6OTG6O@uc4_Gq%$<djUdTda=-}O_iy|XbP+>_
zh47Z`&kA@Rb0<i`&0XkJSX#>+7Kbl42U#j+!#VGJT^S>y!NM+87Wu)37m%HMVl)gd
zPbq}E99qyZAmh)Qu}4$$e92iv+e?&WRQ0wlj0EEQ^LFOQ)0-Ymht{?G-kX?HVXl+A
z0@@p5bm&$;C?4Fhc$`$C-^wL}lS+^VyCkIx)-VYyGC7io<8booB6L87ne!xkH(XLp
z;y8J`HVoM9VB16t<ECY5l+iQSPpH71eG+Hk!(8ks(Q<$CFg`H;`&B)VV2%s*@+un3
zXr?OKl}qAV&g^~WD`V+HDy}nOTtLF~B<XPIMqMhN98?*IJxJ<dnFGqVf4R4TtzfNA
zS?Ev}$BP~6YTfQZ6f)n*e$g2HmJxm-zcB6%k02w5UjFk=v4$04)S=?5H0e-O{tgsQ
zn_~bqmTbH^eYuQPFeVY-nZxG6P+Co0d$WAz*_WaMb(;k9&_2t-m#XiIIlRUoyBM`k
z$!g@e@CKf+>?7y*DxP1hAEf_D7fuvabU`=*aC%uX3sI@y_1)#}8FlyFf4?IeGI{Up
zgO6wDe5uaMPDkq+(3|<ZGo4VGKG~QHU}6+cL5jvUwF+%O)jH7BF)%eT`3xoTyWADZ
z=a7Gx$m<!B(_p^ZDUQnK_sbkwFiQrWoZ4MoppdZ=7Oi>KDtIkf>`q!i(lTvrv(4jA
zu%rUz<@#%7oA3*&zWn9{-K4+>M3^s@=H`+{q@$>_8t#VluP|z}Tz$u3Mw?F)PNgH;
zMq%h5r|atL2Q#MT#1|OWRl?rWlS7*cj|nl_b8V6u>R$eS|6r!0)y(CDQv_c*xRQN>
zUZzwgoU^(nIOfe&0B3{CSYRzG&pTmN1ETCakS#rZYd3Zj9QIQZ9xR;m`mnk;k*lb~
zVh~_M_G`=;*wMsWCI@8y9+8-WgA&_`5MLjOPabIu6LDd3j&{|V#vZ^{kSw+6&=@n>
zH-x!J{CB}(;V5+Ca?3Dh>W-r)#pPfw<DEEZgrn+69hP}B7s^yIvd+eav^<{^-t(Xo
zZ(dS4WMuuVqGe}E1YV$@d@w!wf=&!6mQ_u>j!XkWEPwFSPD)A&>w~~BI6S`I;qHxZ
zQL?<Tad2|P!NjRk$z1avugK*k`zM=E*(*E}HFILeonJ&YMl`abXFUAX0u|3;kkDTZ
zdl+VQixl71(u>^P<?`}E>Ac_h(KuE%{H>?mQ#R|ruob(k;v0kOie7LTd6DQi7QJ!s
zXNsIJksRGmU5ob+_io`X6_Iebs7=%qTzVh3u|jTem|%$1n|!fkJP$SK27PmD-TgH%
zsOu|V7i{I()h+IyoyCD&sp+m);?Ve*nD^=g#Dn7+VOCW{A3RTN9e#b*?R=!vk<u%W
zSB<LhhItdSg&JfkREpJ;uEU7HTGma-{9Y(bUvo@cuH-Woednpvg6*Pw7*WE=Fv&kh
zgt)_Ck&b%{=sf5Sv|o%q|NGmHwfx&?g0)PxlyOy%eT`JAl@jX|qp`P=NM1j^P8|;!
zWeWtSHNyN)L;QOlp9A@2QA<8Mw@boofl{T*)u*SYeIFOfVl~GwnjZe;&?NZCq|*JQ
zdeyu}kO*rB8q9c_4f1}NMc;o})PL%qv|;JKznt8?X^Oe@E#*iDRcV(`3!Eu-`8sc^
zr!Jb4Dv_n_?ka-bv}7}r6Qb8>{xJD5y)-0Ae<F`U`%Z?$st(;%mG_w*m}$3SY;0BR
ze$&#Nkq}LO&OjLD!AhMOO)wMm<en+RAuMErz<f(qP1-PuH%tiQZhtb&c}RXW!5CV|
zLJGmBPg&Xg?t*C`0IVPP^%G7k(uB&|=PC7RvamA3sG2y5lo^8!Ww`YP<)DNG{Pcn+
z!H+W}R>s0w0Fz*WvgF@%<boud*S*ev6Q+u<%_?G*UlH?4DF-pLIl4XIg%<7YPX9yp
z^nD-9HoUK<VIm~Y)<4HQ6rH=)pEnE^;9M9t$m%wc#)FYbGXqo~S80-wZfPlo=nL#k
z>n6#XP3Zl7`dsNWE)oTw47Tk39iv>$s@ZYhI~Yk`(54P6R`OGre0jJ!n$82v-^6|f
zTSd<V11Tl^6Y(pnv5?>mUN6mjZ=MW_XDqSkQ2T#X_8*X7`#H+?3N~ne#Uv3%RVYOc
zObX<H)?DV0J&!!Mygk1mAH?_a`C;v*4-H8`xNHa!z`25AE9X0v$VOP0^{&C+LQ@vL
zK^7Mustn!)yX@0`b|@%VL)6oVsNN9}7z1+(+$8bTc~%y^pR#Jaa>$|ntefVj)s}{~
zj_J4(NORtwGtyXF?yC%*1R%XV&pCcDTMM?FoLuMYuubR9h~K!O067{S<lf4SlZ2Ht
zPWC7W*dQsb%gXCJl+Wa+B*;(8)6$lruAii;6oER)Sbe;69?bmhQ#Yfr2>3h3p?CWo
zB7kmLDGw9F`*FrVmw)4z4&%5)NDWucbv=_Qoxd|Bfix3Ap~42k<W426$)dHJbrbP|
zH9njPO@eThkCP<)N~Ehi_JZ2D%V8v-?^>5P2{Xjm7o^R&N3?Z;USA+4{jh=S;!ZK~
zuXP;K({{KvnY!l}b+M--Dg(1Jq|R97XayruzhkMj=_0(a+*3V=)!;D8P)|rQX*${G
z&k0nhjEsyhs5f_hb9DEAFdw5vwd1_tFmD)|nj-DL-N}Qbm)BRpG!_Ffq)0l<-u-hI
zUyQS}Q<s;=_ucnH+>ukh*8ex}XgxMn6S`7?&7^;b0D`DQ4VP^hx^EUVtiG8P)o$TQ
zN2n1eX$hgq3Z?sWZ$Q^40?4lzK51DF;T7ow;0j`W{n4w@{ovymOZJxahMd)?rKL-S
zLs9n|ST&!(E-8h)tZh7h`sc|0OA`fA9H2<tTI^f=e3~}XSLWLK_WoJtq`opGl0{31
zv1g9pxYg`MI{L@ze|Nr5LX-eT!YpvM>G=m+@O}nmIGPgMw>FPocxb<<7PNCFJ?zI8
zyC13fPQQe{i0OCTGMvBdvmDZj5&R{#LXzYIemCXekqnuU*J@B63O02DZ|T;pa00e3
z(Wls^yTPXonGScmZS1sd)*qg{-`dl(NAsLW5rah>cl4+cCwnUBO47TlOLQo{uVrkh
z^n7o7lflg&ZJA4bQumO4`x;8+BwbG5RJ7BIr<wxX>4<!Y_zn;aj#p#&(a&0-Z$lZp
zH+j@hS!JhO!wm?8+1!pAy^5@46Jd9+2so;OK7E5q17YrXfT%>3K5guPuD|BirF0HU
zLG8oM*|^%jhQo^!MjSFh?R{_Vr;&HM9QX?ONr1yfleZ%4Ws3xvdt3GV0s*w7cLFBz
z<r*I)*i=qHNqkM${9^U<Qb4{OOjv59B=fz7=bHiq`fuk?_w@<FTFC9cO^lrn2U#bl
zr_TE@d#P?)zto}-Op&07CoEK97axPmF;`=R!7wl1iAzB2w^XK&MVLY1zCF`V=RMOJ
z-Opy`s-t=9|BLDb-yY3;{hr{mti|_G6Vc=pc(@wg4W9V~GJaI9V?nvAn9=H=DsuG&
zLUc%4XGz)gksgK!2EC6pd}Bg>8@_Ek!6Fzw-i+P|dBHq~^Y~YiAm1k}x&TEf0Lm<C
zv?ALseBJL|Q~%%UW7cl5aUa++L}-z1472$pGj{!6d!7v%n%g4WkgYzeuY}Q*72vgb
zS1*OFu@TuM^eYXoSJ!x3&)c}JhMAVr?rYN@-F@{`G0JAKXo37WcDG_?dMEg;ylGgA
zyYL6+)Kmtf0~m)IHEn0#Gf9J+>4BFx9<YQcJ|$^HqtbdbKUvr~NtzP)b$+2VP--7q
z>eTC6H#pxAhiaTW*jPRd4i%`(fg9(%anW$nOJk=4^+_fCy!}xNFDl*7N}~)9a|Wr~
z1kgpti*8q*D`k{ukxsdc_M&4zi>Xalj#W^9c_&EXfh8|1b5=HFRy^AO>{;0+!lr|1
z(qS1^L3Fwzc#7&&+xB>G$rpyRKt1;{J^EKYWQgz?ighchT$xkpFvwr1MH|Z)9Y%jN
z`YwZ~de+{Z*Zqvi#^(6M-odDfLIxM}!v{?ro!#e$t4`RJzxMp@*~P3;-PxNlSK8+8
zFws)fkV^g!x_F0?>dI$V&qM*B(q0v~IsWY3q;A4G-*%=aNu3ziPL;Rxzr$bc=Yd!q
zSW8gZPZ2Shu3)7ly>+Ivv9mVB$fVm99>p{}=Dc%NLH^BcTZun*Ic4K|PdWd2hd^3F
zQ)j2C=7r($uf(ho_jF7tpbDFCvTWpa{nKC*16al&gcgAkzrN|tf88@=BF&(8dIq85
zTVY@Pg=4s)svpGy_kG^$+}rz_&Ra7qg#Bbcp$3$m(Xo#~9nzaTYvSQEo|l#|zlSvw
z1=1nYFBJzYiG+_7__0*jmS5O3qmwjE2;+w5iILh8%cCqWTd5q`we*z8j!Wybo3x#O
zmh_xc#gzif1#auotkajqrgQnZXMV1(u4aMkU^ihMTo|%$=a^;Wf)u6&pT+rk=|i_o
zH?^xElQt4)Rkc%7Q<uJplg>-Crm63ijuo2SGx-6vt*mm7S%S>8I{g@71GuX~%K?TA
z#OjUBAO_qdyra2&kgk>lclpokU~fdsO%g8#A34&~2@x@Ps>tx%I9r8-I4i@8iROK$
zX@!E2Bsx=g-N-U&tYBUx2AW6)%PJTKB(x)p9ribLEB=q6N4rY30^)T4(^dN;@eiyJ
zHfBe}@lp-Bi`{H~T5&-o@$I5@Am~}3_6od_eE6VM#6H1mMy#s@zqdnsViJ3Io$Z=v
z;*Rs@JKlqtXZPDN;!lt_#5x)0;o*Jxq76*o(gr2S0cWBgGhs%s#K`SG6H(bhezCqS
z3MIJNE*Qu@$U&r0Q84RX^&gu%UOv8^1r0{sPR9fhzd@BJLTbEUXq`!56tC10Q!B|r
zeG38GAUe3@J5YdHy1a{r<0vc~ABs|A@*oGVYjFY4sXQUAtlC?wA`z`9NT{$yHa>l6
zvS?fQ?fMt{=P{-G=SNDB##kJLWAf29`VjmSz19d^To854WE+Do>!^2Ac1>oAC^CgQ
z!_FgG00$GxE$>qI1dIdl95&xAwR=THCJu^(*9Dg8OCUuueM_HoygGPaz+=37trd*H
zYBg@ji2!YNG>=)Biub$Bd$s6%eqh-T`=*4lEml?R+<c`tPpDI(Qgr<&1y(?AXeKtv
z`CuT2#GFQn#jD*y1mPVnr<VE7G9n$u7Efa`dx2$vA&Z9aO(pl1*87&GSeQ4U0rMLa
zGIh$KL$7=(o|l>9u8XRz>kFUa>Gp6x!ixy{J-mM=h+^0FOe0Fi5&{Q6QRJ!F)IvlP
zx}g!I+zF?v$yz@B$-^B%`IxLLPWYGw3GFkKPPIyO3dBNdD=Aa^t+pe(@lHK$hIjL@
z6Yf~QZsYuwN>9tb9YGgwtqY+Qzqgt$8_R<}*kVX2%dx<eDBn#YpDPY{Z&~oqP**;C
zwh#dI<L7bhS^JNl&<z4ncbVe!%VkJmsDdDrpmp7*`}&1LLQS*TN#l57DEjVvegFO*
zj^5{dZrjnnDKsSozurl*V||zxb~~xX(#CS(>M=H*3?Xt^=ME7xUewT4vQWjs9SUa=
zC0y>Hx?d^A&<BC|fU`IDU_bJ|%5#!iGF%ZlD;M?Qx#Px(GnaF;e6GP_YEE9)$asbV
z_+`9>HHL<=CQ;^UBUEHdF0WYGKQrH$=CDin3duS})L#vNzjc`#j#WUMb^YWO9h1L$
z%6@jov$jzPv8{D|#9QXsy>A^5=^5DWgXibhvCNsZhVTdqP(>T})0M1mSg@X9|ETaj
z{nXQ6oYpgnUgL*v^11)fHJ&<eQnsL$jMR6?&ILtHya5q{Pi27)DbN!PPvEiLcpcbp
zb5p8Ud|#CDWwH<t!bc{mIHDgl&O1u|e*8!74MN6TknoCbxo7Or!7Y(|cz>Ug`u1E6
zKBiwJ4M&49%UsP6te9g(>J=c(Qpf{~I}Xl&h#GrZn-6}bdg6GPMeR*j!%}3%+q|@N
z#(=rpw`o01m|$G%f{tSB<}es*{<ok)OMU0~>@53&^=Aq^2&@MUozZ9S<LBn(rJ+{7
zTC7n+DDK>oz(U5g;C;9DjCjttKT3`JY7Li!H3S-eBT$lGE(B`Ovadg^)2>U7G%QsG
zve06I7#jt;Xr4_>1l%uSY|?uhy~YIwYhPbUfwC|olnP<ezLiq|R0R_mc7Mx%EOIc@
z6f2>tv!fkR+z6FSKg`;T;`O9y18ZcMSZZp!yT{JXNQa9?y{oGSv9(fF9KOjUV%soI
zXjWj;MRw+~@d*tFo>rkqZwN<}#*I$@Pp2Y7s{+oS<+2#UUNjBUq-!RjLtN2X`{M`=
z6EZR~Nhw}5kP3xQ70r{CSFBFQVk|}Ja~*BoOTO<CL&g>%Jcy}w={IvFZt)6Z1(!+T
zpOip<M-LOJH1+K-%xxhu%5_v$Gr|6?t*2qLRQ(nywa+d~yn~uk4<GgS$D3Fhd=Jo&
zv8dW`Ga{K=3#`XddC!cu6#KzAIt8ztBHT&7u@bw63DDCRHlE|rO2c%%OtjI@k!;90
z_@KLj?UvD;zoz*|<q<y=?H1=&T!ng+>XW|8mlVs8rc(1J_-3G;XUp^F%C9T>Ee`ZN
z+%Fu<EQBSMuP|lV&W$BA*#CShr569aa4=~@y9$btOme++{lx0eLp&vryYV_>dE1l}
z=f(v(zphhkC{5$UF4#)T5m3Tq<?bl(_GqT3Gi3hg_*L+Um1M%e5Af$Z$xYT8&9wKd
z;N<g7G<lC5EijxK*$Ig?!&}`(2p2EEdhrwM;Zrq^)$fR1v<k*goLT5FhGcAJmO>7I
zkATEpj3%w8V2CIJPx$1nVzEG+92%<{0hAbD3Rs2d6UdIey{<r`iUEAv;u+&0;v}Hr
z%z5do<pVqp?{wXg?=FpJ;OmkCury9?Mlj;Lb(~^sgVH<m3A);U*fe&WJhESfZ9b!;
zf9-yIIlriK@QEUOyB~=jF)5%E1#CVPuHHa9rK|3qd>SE*1b<=mq?2o7m?8&BEEQ2w
ziAgHKM@KNsBC4aK1Jesr=Sr1a?#}c)ySRpn5P{fIY(cNs3peT8`BKNgb&Yk(u}TU0
z0A(~YZ;r>8bv;*h@%HQTxQqX>J#@Q9rel9x+IHrlN`{vX8T^32NbVH>xv}j6#c%Sy
z?<e!@4($(}=0((a20Gp^NTp`%*k0J+olrIMcj8zD2sPK$$&1~nA=Ju=<dNmfU!4yw
z&9ZSjZ%T!V8{8-RAQY#kTr!)iXI3j6dauuMezJZ%=5{7Abg`_81!|KiY2Ab66D(qQ
zp4n$KDg1Nzjv-sdbvd&1YWbJ4oGCqN<u5o%mxujEdo2a^Q`>(4q2*Jwle+h_mMr#}
zZ>rX}#urAmTGi<#Ecqa*h&CSr^uPj%A5KY#0pj#5sYi^6Lt;rk>wf2?Gtx>L9uVg>
zi%J?3Q-KNK?i+VqFT}T}$@Uk?Qu_a8alwboC4AvVQa10Iez0E59<_~8!e_+C8u>x+
zF-=oT%gE)5vy@AzuC-OXa7MfwuCS_SU$@;@pY9-*GUTtuO}9_H(s}5^H^CoeK$1BU
zBmCHs&hL>Cufa6IbU%)x)LcPKYkS{^E;4y#mEaljijlVqCyh~)9e(GUm%A6;ljd4w
zs#K}p&!%<vLWSNQ{vV#+DyXh5+PX!9gy8P(t{ZoEhmBiscXx;2?he5%NRZ&Z37P~A
zZo%E}`u=n7dC3DsQ3YAG_MBt1-uiqfD>E}X95jR<Ky?dLvY4mzOg4cyr+Ix=PxEq1
zPmfZChWI{|A<>MFP^m^@&94(+Ir;P|qVS;&91lD$n)O<-gX~%L7{jjFQ#PR7`Y7Os
zKJYD@w~<ll=l(ot`_mhoRu}xTIQIE})<{w*-6MYE?blUm=YdM;0{NO!db0@0FlKNJ
zF!q$FjiP_=irxs=SYB{TH!d<5ZKJA@$DgN>pfNu}OjD$jiRs`}<npJNmLkcxVVb~c
zHYSeg<w%2I6y%*zeCujy9Nt+YDR>I?_KC?yd#SHsD$!aU=bTFZ0Y|ls?vL(`^D9C#
zf0sS$7OIN}2j@R5%jc^MWOF8o@aObNe`)o-%^7cZD4*`gcY7-+c8YE5n;E}hV4#x=
z@JGM$_(h*)*?evFlXpZwIn#F2$IzzyryPd4JHnnO5+=io!qB9I+8pPYhx;=#?${-^
z7*hzO+(OJOslki6Q(MG{S~^7pTXR-5c&w}}kg`7;xye(5)#~ZItlt@z9HuMzAKcH3
zU=pq?;z)j7sS#os!8R{M47HlYYVDJ{`uf(UrXMGb!J)uCsG`EEl>W1~chSz9tMTQr
z|Co%r5Wmc_{i?5Nar5#?5D%fJ-7TZp?m(Lcon}r@l@w&}((!kIVp&yt7@u5FT|8px
zRIoSZO#WoiEO9Lzf)^e1x`}d1M0V_<Y*&87axUmgr^k^{@ca3DAYr1w{n<J^F_1Ha
z*H7hMLir#Wp_VV=nd$XiG1vWDgq~$!U<SaJ&s#?P=qM&aK>!kh;CfuXd*=nGj^^d&
z?w;wI#D#=tSU=eK)IJltdN4~ZF<u@^IqXRQ^|GZeP91zLN$-y2w}#;1n#8>-5zhb1
zc=lf=iHseXfVacn?_<7;2Bnd!LXT7WZi@*fQjseW8wB66Dq2IDA2MdCv?F*><8NNc
ztCFjiYDE21b|;sSb84HIa8mWB=8U7m>m~2dyg))LRe=FY&OfOpI8|5}-v%0cW!ia5
zF@K4HaKXy+4bo({?A=3uFOngbq`aO$*bCq<2++oeiAhrM#|EFI2RmglX_N1@+8`yB
zb1nBOpp?e7^@uWcP)S$>=M+!QHX72qRA^sd+$5ih7tClZO+DiFPb&X|IOB*>ETyUs
zdzO?eh<j;qHT^TgyIrrR-{iGyVTM0Luh!w<)I=it>ldEWYep^LSS|GV6M6_U_U?=B
z1WWHStfd8uCMs{-PmpE11iwsnYSdtUfL^l8P^eHqkBBgnY6n|*S|xsNJ?hd`H;nt3
z*Rz6i5Sp4AC-zvgtQYRenzVSNk;JGazTxaeX|+wF04?U89H;8ZP~n!^;X_M86DMb+
zo}do(cZ{U9Md5Mxy67rG3AQ^L`n4*4^Yv=;iaIdun!1dgalryWIPr`-hn-(W8v%}q
zz-g^>0El@`jjs$2!rc0WpPmMmR!cfsr9?*JqJ299qH!qEBGLx2@Of3dK<ZvDjiXTa
zpaz(z7XX+;$7;pqx%<<NJ})01zt9R57ZRib(X2*`kw0<}Q(9bvm3kqPAr!eu!#v8O
zd-dM-_^e6*Qe^pmPlw6#=mGjbHeTO76S0$HT3`SXMSy<GgygRkLJz3Wk1p#ReD8;{
z0ZwJodoNb`z(&S5wHzftL_oB29(woda)|^S@|MJ%cT!j2HLCad+V<#7(&NMa*uS&$
zx$}E@Ph=IBy+zE;hbsbbeT<zUl?Dc<h87swZ;gTgZTto?oz`Jk3+$QS{!jRK<-!Ta
zw~Lip6N*-HIYn7+;Xl||GrPMgrTCD&!H=aejCx9aSB#`g{5d{~kS{mB{*7R^kT63e
zanKgzzkD@C#(RP#=7+l);zZpaUT|_AM-fG&zj^ZNH8^+DRw4@8!dy_%6qE9I`;-;k
zv})UPeu3*FbH)}#PaajU{9``ywF(z?c5!*7!h&W@U_k-`jOWe{`MJLJs|5uPRgd0y
zY=(ut^@mtsz3=Ne-J*HMz7IaCnT`vqc~+{k?EqJ##mJ#BQA7VnYPRO5Qmb)nhr`KU
zt<d$oe$%M#lDf<j^4qwj39W5zY~A?q3lFb6f{4O*r#mb((YrZxGS6pA^y>6dOkH43
zCQ~MkAH~ef3iwTc-3NffCf=6b>WblyB5p0HLIwxk)QzF}DdjoSk%3I{6_Yd0-#YIH
zNF}xHpbH3UE6J)I7vNpeVZW1$;*Rc^A52<Qm0ukZ9Ecl{<46e2algssDSHP$#r)a4
z8!7v7ax$WhqSrQu2yH!ExP>74@ZX$0>+xYH>G$ep#ln$`i_4E3F6%D0J;rsFr1b(S
zV1cUV&`{cL0V3og{SW;E;EhZ>l>&j0vpaxK2+V@AK-syBv1^Y*52Ka`fk1z^NALhA
zidl2}4;$d6K@71n{tKFm=t^>Vh?Ac+zJSxAh&X?><-SRxm=tmC7fqb`pQE)6uMH0#
z{uk!0Q9`ztm5$E7@8Ks&NLOl15~T$Ljr>0-jyBhKAY^Yyc<<pf!}DJc%EzH7UgN~%
zB0s{d>{i?`Lecz)6zc4qMF7y4+093O(U(iIrq*{TN2M+&><|eKh{o|PZLd?2+(_H8
zqSlT8FSi^_TYC92LU@*4yq@)csfLG#m>uK1>nsy%TiZ_-$|9~GLhuM~{zM1t`M}g0
zSgI^wAadB=TGY+pY*tB2R=t6)#HpGdp+2UI#-R@U78?oUef!?Fq*ryRgpfie=mMcE
zmjp$_<YUX_p{dwkeJuq@{|O}qZMW~0dyJLh1yf_$7P%Mp?uj+Ti%5CirrAm|bnym3
z`sED|(`A&7NqIY!MMYosh0k8$tMkV*BNk6QOX<x^madML!37GyqV0;aw$|R4o8`cc
zh5-EBrjSZ)Q=kixm&f-(!lDQmI%~;=Jjjs@hN~2nCL#_M??UbC|CGNePL2|jFqW<^
z52)6?v~EAT{+A>-cWEx5EiC-Zd7bYarTXED6sIEZ@wAFN15cLP_5=PP``?vzyoaY>
z{gXACc>pU@;D3Lf4rHL}M?bO|T7if!_w$_RPf>rbfn;01{g~V0uuWAtZJgLsBV=Y_
znLF`nf}q32&G|_6Ziwxr;>gO%B2on82Kh!BHzuCg*G)-V!3MuAa2L3pjg;~@vt?On
z5l0L`-$p((KfnC{yp~k43Vo;+ObK;7o1yvjRAck)P*JJW+=PnxWn%H`JGy1qy0=KF
z*CkA?NdWA+9(BW#dj?FAy{w!ih#^!);ft4lz{Oki%SO^0OYj4oD5d<^4{FYx1^8kY
zP>DJ+fNiaQUmZ_OR?w5hP^;(tfmHrc%`Eb*diLxL_0%cvty<L?w12z6=F>|+JM-0u
zqXbty0&CdLq)&*yLj#)uNp_C@H+Yk+<)^`|wi)HSi0GW)_(V>SoAIiiRX|U4bKl^!
zK>W<oh$|yq-jlc~jcHuK_jeHsvFuXfXL-ok+WtPv7bYq7i1fJ9jAbg_IP+3*KG+9h
z(nbAvm6_=iy8eSSF{8`%NTuN$XQhyhp*g|#t{R*hi-z#9f5cJ7dY$0<mYz0=cJSWK
z5tmji50D0^ULxc^{sb(p6(SK!k|i)ATv_RJGr=AZq;2|2Ox>x|i<CSLcsg|j@l30P
zF(k%|q;Ro$2OfHlo_u%BrDjbrMIVMpa=Hb7vHKf5zpb7)$3ovz3{{m<qJ<>UUg2ZI
zSs^;o{k1^cJR>O(?=oL!z#}Ixg9GOLuI~dz0BHpxHva?&lrjO_YQy&c@4x3J1K{j4
zfH?Hr$?Z>pDLkRLEf+arc>G#@Jw5U`N+8uIj&7c4<=Z8VAEjqckQhbJV7qu+$IQ0i
zL{>b5@7S||I@Gj#6}tvaU=hH$7&^hNV{guoQid^OjZOi?hXS;OtlwTH#0oC>P3gp?
zaCrM3J^k*{@B=)E9EI>CcZVgDqDVqvxyFCLwer<C*3BClOLj_ABWGPJ3AKDmdHgDq
z#6lMA5?=KBGW$40_6{AZA`S`@bG5S=Vv@E$U0_^BAA_;rT3+VEP;d)=7BN#ZbF^JA
zQrgP1{0J9~5WTk9t=h3kwG;)m&spC7ppLy=H<DR9PwB{|e~jmQod%hxszgRjWf`hl
zz$$3%6aIN42Gg%_y;<u5lle=>_ov8RwXyvs8AM43AuQ~#t-==#3R{Yb;oR8lnt%_6
zd0nsTTqY?MSM-u8nc48>zE1RfWHSuq%vW`Fs{IOF?)7?t&|xFqftb~>SB&bKDs>Bn
z{i;K?{)q8$6DDZY0rm|FBf8AXC!U#^nbp6D>KYmrY`AR}^R#NU=eaGo6?%6tG3hD?
z>|9@9(vc^Tbq{*j>)lZ+q<o1JI1u@_&L5+$q;-?W&hw@<`!o<?Xy%TO78?JuLm<PW
z`Qs^<C{;Ev`-RH;o^k!uFW#cH^dt`)1hO3R0zeK~D=esS3}6|MsAgfm2fqWc7$Av1
z!8wx@{4yrw(=;>s^x<H(KvqG%ag|PftPc<hl%|`q<Eu5Yq@ZvE$TvC4PXQ`KA)Zp5
zT6HY}dq*MsBtRxc+YU}N7L_4X%}EZGih_O(A}eEAd2(gb3SlZ$S}qAomZF;#^YZ~N
zQM$oz!^Nmu7!~7UtgP;L-%SpDtc%*#E2>>uAsIHy4B|1SZS6y^9h+vCk+p~)CFI95
zwqHjo(Iq%YKmF<Q{&9B48vK$*mXfnr%WBc-vqj9-?~_sV{_lX8^se_T-ZOy+11!k6
zB#>7%YnrgPj<!*v@A%V>6(b_dLFlu+bv#jr^G}i+q(S5iaKWO2*PTEQApVLpPLdG}
z=MnL+bL>-nM7roUNW;=E*UZP#JMl*#<l;fX?ea%9p4WQdmA$u3h5ACDpl7gYx^~1V
zw070yZNZ_Fr82WefoRPlun<(d8$OctnJGUnJwvn}UYvczH}7o6SL4O)8@FuCL`asK
zS9U?&HqwwhW64r(AO~V8)~4W}nq4WyhB=!|O}<_^(TN&C-@xpnb^a1X>wk6O1W9x~
zJg(;bs&2=XtAz!-i$_@IzVHKt#hd3dJzl8?e1XK3?Jo(wekuj$Hs~2xq>tr3oRj;N
zg$U>5E+H`JdPf203u97;rr+o-bUvHH+yPe{`PAe}2H+p7oTNHEMeq%{O}gyaq%Rmg
zsWIjYb_b~8rvq<|*4GV!J!!Pam&o+EIeIe?vF}fvoR#|kPp5?G0c&2IuC~YKwmpvc
z(h$4$!9j^|zcPigKjZu*sP?a&h!XuPS)9|FTxm!filoyr)J3Dm^<d-Ktf_jNA`RYU
z)4Cx$$<Gc*zXMnFcHM)ZZ-owKg|yVTzK4b}{Lm8$6+?_~?XwU4HIr9ufVX4Dnd!jl
z6JLaOy~*!OPmF8N6RLp-ZJX<|CNfy6u8?ZQ0E8btx5l<-3VNS4_4dw4)czMd&^HE`
zY!Y$d-(U>duOyBd0ra{nF_7qg`TYs|)-Q3+W*0gx!mrmDGNqqso<JK%nVFeo^9@W)
zdQO#bfd7%fmKP5saAuBSV-drIlfY}aLY__@LnR|$L<KASe3SP4KeM5UOrJ+fdKPLG
ztVrl7w*K*&#&xrc=is=Rp8a^-w_*l36+0NSWU$M{U9?A^Z&4x_n>l4XCGgbZ;Y$fR
z^#m7~1yA=Dto|cApR?g+r`;K>xPDo7eiI13stO)d?~qL0gL5q2oO*ct&gZkYH`@%{
zrJrgv$!W-fE+|&}9|%r&EavAB%ZoScG#q(itP+x=IvO{(m(TkDu5P~2n!H}&RWF%a
z23;lhv9BJiUc1u)?y=-JF_o_x9zoFA0bzMgeM}L9@|#Vo_WB1~T68g5<4?zg@4LXt
zgWPI7a!U8%qPOGG`u+Aj$F;I~Ma<aA2(zzUI!V)`A!C;_oNF}8@?cl-0_z&c(Us%E
zi6ebOW}<D9t<o?mWVN8hqK`bN5n|l~A4=o&>uLpk5NDTP|NBv)h1r6e5V)<lZ{)I4
z`55Qxnx0n0ui;>Hv!le6RvUIosY})CP{oyUI)dx!+V5EU=ef<e2_K{Wna?M=cAJ0u
zS;aK%tb4}n=B0uDWcLfWwM{X_^ZyukfjoVI)xeZldJ(Kj>Sqxrv5LDps^6$qPYiTc
zcH{<I7hY$<Rxm3QBx(wr_GLLVjSS8RPH}#s*^R}zh)JWUoNtd*#b`h&$+Ted?kz<`
z#kFO#r!&c`dcFF6Nrm-ITALRnGej{qCTl4mjVzu^BT~-_3up>JLPC|sweGjaz$d;+
zidE(%XaIaN5DaVu9M=HR8G6*)nr^MO-IAS1pQZ1s977=5-+WP>`lae6dnwn1Wtfz5
zIL1=2y^F`sYl#jB1{^f+`5F+a22N|^E)H9AqOjDW0H|6a`i|o2Pe#fOrYghN|HOZt
z$mZtpiu%k)rq-}l(Z5)(3CDQO=Dk?;ddYH=g%TP!nGq)Tw{6cQI4cj>^_Mh{UT$IB
z+DRG4^Lt-nzTXtR%|5Jez8c+m-OG>qX;d~)=Y+l=_rH;#-hY2NZVX79nc>k|`#bWh
zZ#>f80x<j;+&k^mnQ&D7L9Z)aCA+#?C3D-^C~`f2gT1?72{>3av39J7Kg^k!z#5=)
zLnvfE6V@*wH`-rM6qhrQA!kmu3l%r7($o)YnLoSQo2p}OTre~=v$f4Z6ZJ{jxud9=
zKvWLlN<${DS3dgvVsZZW#mfqG)UjvjVGa0FxoNG`V7?;RsAJFl4kO|<`gB)qW7ri`
z;I^E!GgGiT<+|NUCk^@-{n&IfxwB9|axGGNQ0p1krbQk}tJo?8rj4;+Q!GuKsO<=Z
zmP#!@R0_vB{2BiNPrDy{^747I&7Q*0!7Ye*@Y)UV2UQR+fk&#0D)2n+dU}rT6h!RU
z3$C!u-GpE&6j;|5^y+mg=JbB(vlXA@OEc)-6)UF>SACd)O{Jz{>P*!XJuR+_5DD!j
z6~IfYwz!lCYN#knUv~$%h1poSNG6jao_je~<CJ;QMA@HS)bbMek#M2rzP^!Sk-in(
zyMvKT`GWq?$sQ}R<?_c?vURk|2yNgQ;%eI;yHAn6vl}2<#21bQFjf?Wgwoe>ipG`(
zpV<7l-<0K)Uzj4yhE3UrcXJa~cIQF-4=s>XJ;_pRk}|TDvDwf6DQ6z9>EJvgxkF~?
zF#9F9t|+bhu0M<+L{|@yrup<Foib9W4Tfd5Id9M?%M=<CC_P~P5)H@#Yn#qMG3eBc
z0^o>{Y6Zq+(OcsiUfU{_m-iQbkz<QCGEiu$)*yPgSwG+3{>OsN9g~-n%_?ww%@QME
zRRBb8`ufIZfsU~iIv;IWR=*27tqK2qO>%Z~eh$$hdK=q(Cj$wHlrQk_K5ss=9?TX=
zC^Y`!K2@3&5`r<fah>K=(m^DC3S@J;kU(z3S|lV|B>G76*n$sFNoQ~EyRc6l!ykCY
z*?(v79wYiv7Cg3fiQ(F&{tZ8e*6~S4+7=>9uY>#HgnPLwx?D;m-cz7=J~V28dZsGu
zPpsy|pF87)M%Ap~@B;_syd1D+p;CtqQ)^MZ)`z{<qtnYx!RHP6+uuWX@^krV<<7q5
z&sD2#Ian#FLqkKrEQ6Ej)ax0m!~3aJdH&Bg(_Uli{Y!3;fTvwqk29G<yl5BvK>4*e
z#vNPx!b{_BN}a?cbsnR!&+k}&a>~lWymB2?nC`G}Eg?Q=hqOakjP-D&Y9<k`f;+aL
zemzUPs;@u9z<GuSlU0dn)WA{RqP|#zgWh(6vtW~1g4)J$TqDu_E%&zHNU7VHr4Pzm
z8ZXX#6RP|wq-wcZovw1D8bg;du4}PzgRX!?XtPodcY^a&|DQjGb^zEwPp=$YU)G+4
zh{yn7*x-5v{3u|#vJLvot#EkjcjGVFqGXXUL^xT!LM`_Aka)G?*Fsz?jXhID6wTsi
z^pq<W$KEt;z$bsgBw1eo)n`l2pBIsVZ?Ne6<Ss(|rD<8sou{-Z6dQUJl#a~r{)C;y
zW>f-p1cn>KE+82NX~V>yv=*R$EzO;&;k;xttTx!<=;)EsyL{9DNX|9#?+#x4u*|ca
zQC~m?dF#7|+(fZxh|u=G`(`~95E^5>__94pkvP-0|J@UZ3LGqbd6mBdGX|+s>-%5Q
zKzhBBHj;v(n)t&;%#%SBF=}uUscW0oogIqpT?wg~^cQxE;?;nm;QPO`H4ED<loR`t
z^1ADc#^4+-U{`bwQT@~15vjt{ZWqG*q)I5)PQjBYW-CF1Pn%nI2SxD(eNC1~vh*6$
z9!|bezQr(;!eps}c4wxjf5L{FQkg60^;d9Yt?ule&7Ww6z>jE=ScP(|`HiUQ%M^7Z
z;BXJ$jCjwtMhvYJ#X5(4iE;M#28q43N0DI8@q%n|n3E}@fFs3CF`}Ww6CkARMgHsc
zYK<@ljr3)t(7!hIFxI>C364D*fU6T9Kiq5}J=#qpWliO=Jr;f0#>ptR8uGh<ZJr+8
zv*7AC3HbJ}b(|VmyIlSC`N{-adRL)DYv39iZh-pJ3G_bqn3bI-X)FhFla*LJW;%{E
z>hdE5O<(8Y&ams#!V4>!6N-Lj=|@&Ds*mH@&-z>U3NBOz3S`OjxU2pk(#!4gq5!8j
z?uPN$&@lZUFZzywfd$N5mO0`~XqA&9jiY4YfIfql$fjO?)b=d>g*d8!+MLdIF>TeD
z_?EI=z*iNz=Og-A{=l{%LuWvbe#sRXre4<l!iXQ_SDC6Hd^XmDoK4D8`?7Bnt}gUM
zI0D+woV}}xEWjW)Wr6X5Lp^!yO+(Ze%+k1i1f3;%PU(z(KGsTBRyJ3y)Z5*oF|TWI
zV7}}c=@cYVMh}U1k-?VoFXKtB`cjZoTq$F*9kb)`Qe6<p`5z8BUnGiA?H-bhO!M(W
zcs~+ci_XrPZr_jP+h<F(FG8D=eyLsU#;WoZ1SH<SgWqA78w{xs8PRnx7{$IK*I=MT
zNv;&YZg&oje9ZEBy=D8G@E3Br3c$gcy}boJZH8r<%IXRZ^)xfwo$^v<8p`Gqbw741
zhEW^ugmxT4l<YTN|B{r_i@}E}?D2KFyb+IDf)bn?n1xHd2OAZT#TVRYSfMMLG)h?_
zTqE0OxZuPOr_wOEP+8Q1VAw}Z0aLXGT-eO5a^F>jvuWGbUBSs>X^dUnix^eXg_0{@
zT(YCOa9R84^g^TiGfF0P&6vet+Is<}=G-?4N>j{{yf2kF`y}c|Bd<jas2fA5qIB#A
z<j(y@QfG16o&)~Z462$s=_+E25eM4Y{k&woMbh9!!uU~iIAe{{&ZxR}D*-Pd6}i-O
znX-xP!PEoD-cd6kc#sYGz|Yq3TUC7EmdmPpn$Ok*Y*?-nk0>dRK1mG1O=04IT_p9a
z>yc!NLn<U)OiW8qla?w$I0XBnXH_DN;ztby1v!w*%as4gW^Uc<OrGTPKw#&>08Ku_
zZ+N4&OlQn8N%#3A@A)=YR(I!%K=tj2PZVI`9*}+*<lmS{Dw(eRloQb5iD)L5ny_U$
z-21(sT$u-uXV4f2?(qyZNqi4PR8g;4bb+MOSmbj_xr){6OfeZIst(zoc4;qXu6zzf
zIlO(^3zJmH@rJs*_QZ~$CrH<i10@`K`T;Pf>2$ze<+Qg9kz^&8hV>z>MO5#bX3_Qk
zypXWG5w!Ws3sv&upjGI%HS_YH=clsAQ`wrKnwHUQXy8~E*4~c{vd!3>DJW}_FS3Fk
zoL%(eurkp^U*Og8)f-vTXmQJEZ0$haXDE4j_JB(2lAdNtsSoCQJ1obwd?kh8r%p0K
zu;@}Dk){9=M~+zB03fjJZRu$5xjvhH{Ie>$b02gbO-xFfITILqReJl(gH({|VMc+@
zC^n8yp!4UqkSy`o<1_6Q>O$H#4v{LSdaLv-Zz}4O6YGa<A`=UXS7wC&XH#U0I3$K|
ztf?ks{#J#LLf0h^$h@j}QdC`%Rj(fdoKLAI)#ZGY6n&?!Hm($OIP&(heI1W^<!@Ce
z)sq+{=rtO}(f2!wncK3JF@lk&?rVRe6~Q!HrPnJd6QtqEm;0KHdK`B?BzzvMx&HJ!
zKF=9`f72l3Q_YJfQzh}S2hlgmdd&dHN|+?=XRjaEC50k;Z+{1P3!Jh5pF*ac$Nqg+
zwSX5SV{Y!s5RNn%X=@?855=u}Bt;57l^!*Q31i8Q=HP5ePp^8;PTQ87@~gIqqtUd5
zbde%kJ=G+=3iM0*<U+@Cx{43s@5kAvaYJ?2e=diHis(X}3pqeObZ`)P^WmxE5e;zs
z?vLP^&?<8eiHB)WxEhY=<D<%2>JMH`V!F~uG3LN(%NI*%p;JhQw8B`6Gs<_YS_s{7
z(gSF?zLs~qJWXSJCi#1T*Y^U!t1wyp8)BB>ub(6ol+b^Yq7*&l4*}Q7q_>w?O3zVC
zxeWUzB_k;Xc)6-3v&7r1#+KJJgu}dNIlPdIaniloElPy-JnzGQr*pqSsxji54+Im@
z&Q}oh!4c_=z;oJP#t$$mOsIFLHSNTkuV?p%ZAlO3?@OeKI^ktQs4$Vp*fn?BT+7>K
z$WV|yZm7^?Ol#yN|Hl~YXsRg<TrkVia>{1{XF#e13VdSjzXv>(!4FAqBCd5kT&)m3
zdN1ie?j;)a$qK;hJki3|HWKvf8|LU%GJD0m3Ci$M^^naP-4td-9i^IsQ{fh%3nzRv
zjt=UV%>1l|`E`9-!7R7U7$vT%ealp6YcUe*fFls&akH?%vJ&A@{_MAhK~vg!LFPe7
z^t~A<yPQO(?2n5_W}x4}oxE*BTK^-vuJe^&BOjBnn-6Fz&_G|G;k;<AL*?FQ|8ihU
z>TNjtkM^IZr{n1zyLyccXWqQMSMI=o+BQ}D3<Y@AGoU?SO7{@5i-D;dMp3LpG7ZC5
zw4C!Lb@!ZAcTrm5Ys{OnM*Y_pk@IQo!mFEN2!6N*1=t!<9Aa_3S<ybTV3cMel(U^D
z(u=Bo7SAD)acq$Hm<V`Q>OTRZD8NKbted?>xo7)}IS;D1l;0m(E&i=864wZwaZs%^
z9f2LZiV2meU8J%b(nK%UTrGrS9Aa#NGGR0A^Kx?_gh(kfgdtxj(*gNgp^Xf!t1g2c
zP~o;qmArDD?lAXO3Ys!@|Ay^_8b*Gexs@aiL`$%z@WY$+<v;MP?ERRpB8xxXDmBU{
zqe_eUA}W)1?JzYjGd97{1g>dmYa2d3CbYLFs%vz<NNfzoqL3IZOTXWIjtL4L-`?g7
zjRU#_f*NHTf6>O_WX<-7;WHA82oOg?i}$LK)99zp9X%lCiUHnz0OJBU_f04DiI%t)
zd_j;bO#_gRIr4h?Jq?q~d~GpqJ=iPtr9BsftZ%Stdh+nMX7+y_{7B*jw1||0#mEil
zv^a7YAy7DDTA9B<*ONR>jwIrYE`!PZF>c0EHWWv<N9*0P50AhUzi;escu~`a==M<C
zWHlQS?8I%-e=1FrzozzC?eu?da;6OZFSVKDhTfmwGh|vP=EM=w2-;>|$X)*Cd(%ex
zh?&vpUT|sqLDDpobDO&Q1~=ZIfj39T#+xE8(bUULmFU5VNuaQa@W;9zd`P%KKlu4E
z7>D6EQuq~n=cKH24SqSBJyY>Q@+<pVhRd%Y>GKGMb#{knUw&_NDb{w0<3A%^{`;dC
za5-gi5Ply7F#a)ezOvjnjI6Zll!njL;if=a2d^kU4)yI?Ts5(pSL%3s3$lrz{wZFP
zbz>{#q*ijx$Dc!5z1wip$|1(;ZCf2v-fD<2@4bcZ`vF>XS@;ZH2K^C5yc5eYS2;Pz
z#(WfVs+2C3yD!SRC%8=B_R6R~kqJJf$mR#6gd_Ld<)5b=@wv&cex~&_lxyEbRf@?8
zY+2qrohU5)tlr4eH2OE6!JyDKZU0D_j6)8yaex;;;mS^UC!YFqsU@aSWrHQ^zQi_&
z_6yw=hZ%nQ^FNYXt=m>t%bFxH6}+5$B}FnaGF6<)Y(v#LsKeH0Jh)VuvVoeJ1;Z}4
z3}C%J+U0(rZ>VPuQmWB8zUh`saAjs?HcPN#zfi}uX9lD=<aN#6-3gw#+*1`YQewoq
zv1Q3~KNZ(QW_gmvEatf@G-|Q}O<#RIvBb<M;;2fiC33xl<ViE|OhD;3zv5a6LupVX
za8FoJRk169eaHVq=W-rgaxoTjBg7BquDJbJX^j|cjRj^W+ex0z*9RJ>dpnJx^e7OO
zC%10BW1Wj+iF<}9g&XSlI=hzk)_XaMzCXc`8NI_>&pM=9g{?>=8Sf=?&pf={2d8;e
zjjcMrr?1X=o~`x7L(nWaMDZ-wDFcoxx35QdMTdvRG2H5P_Ff*j*81N=1ihVIZ>~t4
zdv8&H8F&7e*H|F)F}oJ(gB1#0O*vVQdC5-hc!1K$$|RU1l|URFqIi~#1_mIe^$>m<
zU-+VClqlG<wXT_!T&g$=l_HTy^f1l5gWs!td0Aa2^C?Pc!|{j{9=Cu0o3d55&*V*h
z45&EaMuVKlE6X1L6e-s&&luOFj3kCqnwL|+&1i=oQ&9OzSVgweKh!_wZUhIzX6c9|
ziMCzJezjREk$#o<?Um=^;^F?KQTE&&nq5_=%=WZwAaYkMl^WpGIhO=U9Onh778q=E
zi`|tOLdoMKO7$?J;y$;k;78IQ5qTTfyWj^a#dk5i_h(N}O$s)`=?I_~SS83Hm0d2R
z?K5Qv3<Z&0Xct>uyOzl>q|e!myBspjessyR+Q3i_x3Oo@KuWm#qNnwF;SA8Xj=2)l
zvuP}haAYA%L6x{wgfj=kBSg98d?i8oKukvF?%|Psf>&wdNEQcy55<FNShD+n_?FIf
zTRU(kcK2AUT)IigWJeLMYD19})S?I4LAC}*xPZTDi-S_D_AFI>H}>(D(iokosX;vp
z0@Nr#qyb#i0jazgJbt~aj4AMAZo1=vTsbXCBEeliEvJbb76;)cbeOss{G@2=9So)^
zCX`i2oxG2#fslj$-<a0YYqYx!!>IQ7)z!_E!X$^kshtq7|4mwCZDWiiiGH~#B}7V$
znPn>cSqeyJ#>q8@ZXZYTq;4cL>+NQ+^+Qa5@5lz<JO|thikhC>!KQ23{Mp}pTNT|E
z+<ap#@QOwg@u?Cb-yJGOT)<;GqQq2l(UR(030#C$n2)gI8QlWThTOoqD<I%CP$z5I
z_xaTO@vfskZi*+`P&|@>8ry9Q&eUNPh%`&zWf7?3V?=QP!04=Votuxxk38o${>|r&
z`)VDF<J%7!kJ{^(mnm8LHuSi-U~;$u6$A_J9bmSOKe`hkHZDY|rnQf=H<y;-pLt|m
zYQode@2_LXCxN|2`u<>cX{Jz&fGRi=3hN>r7t0{bf5fk+;j6CC&|`rPlYh~#T^Xp@
zmh7B{Eti7lA{Uy*t2#qyX-x90x2SWrdILMjZ|1jI<s8W^UqYc>u7g~~*LCMm*uu@<
z&&re}{g+z0-3U9{I1AONQrnTBIqay~EM-1_qtKkF%Jc39>nAFy_49Ky;sKL;VD3T(
z<$?f}GUta^3?S+cLM2Gz8THx7Rg<QOaSp|Wqu8Mau#z0DN<XB`Z1x0&pnP<U<Z!J5
zWqH|${j-rr{)z9{gC~R9j5v%L3J~u$P*z3{e+4ZGw9qVr6*quhR8h;1nz94$H-b-N
zKmid)ANyw^l`9p>K2Cuz61=>>^z>}Pnydtj5W{&kE_E1jN`KlfZo!?ZWW9z<0@%E5
zt{-^~4=bW8Ccx;YLjR{R(6M)E3(l9#zmCaqPsTH`)R)Np!l576^g)=oWO{+^_y0X>
z4ey7+yNQWBrR228r2$7Lt1F0f7?dB5*sda5t@&Rw_<p~=ii#7F349_R$2WeOx9WRq
z$E$|Rg&J|mSBtmLcVdK~p~-&{40r>S5^sN_dtT2>uC5f>Tu)xnK-YuXoa+Z?>w<=$
z;FK(5zu40P$$!6fGv&pQ<B_S@lx&z+1PoI$2Rej&wTwqQs$AQ?3*+XpsMI^6$0om)
zkzfk%hod=PkFW)XDzv9I)`^31yx-dU{Al~(#mC$6Ok|Md1xVf3&Pv-&K`M`jWoQ98
zUcqv$&1zR#Q9~l4&yRWk{+*%2d9s@L#eQ~&PCGcR_e_vAQmGG2RKU;dArbxad`AZ+
znjcR6jB77ca%;<ivO`ev>&mA96!jFba@!WPI-&L`FJfGwk>ht>hpU<k`PO@UPq)^)
zbzvE?efdn3|8o?}&Fi^Ja+R+AK;;{MUeN1<PbOYRYb)RN!_}U=eUpNxPI%)wf7hq9
zK<9QsuEqjk@6Y?L#H#abwgUX98Y05UZ23;EHP{&06F#k6S!&RCq?i<5tVQ(j3b+iB
zhBun3r7qG}+6pQCzfSsnIlJl0bfuXZU>32BNRb&ZXU8~`66S>`X$dvqoASz`K00mN
zr0)~=8IrMOX*NUlF==zRY(gO10_TVrV-&9c1X1ZL5X*p$ntdvd$*Ou6EYub1c0=OX
z!7h=yX9TrleJekOfA>fKm`ol&koDYu9NkerGY$Co?afrGUsS?gv0X#@WV)z7V3jVQ
z+Z>D&QNm_gqXoq0?fx(=fq|hRIhzlXbVKZ+=Hi>PNRZKhCv!gEo=~<`u7okL2?M_G
zF+b_-sOI!CxYr5a<9lFcmMoBAtXiN@=869}1-w4{MS%@uiDy)MBx+d+_>-P7xI!aQ
zF7<LcK*!}-YvW!EIaUiA0YR3Sgfv5a;6Gl;N=khikf-{1M9f72lZyQCp>T`A;;`{r
z&w|#%ZaWnHR3Il)H}&K5Z%d^M6pge0<K?`A&SmojJX7}ZM3Z8KT6A#}S`pPRWZ*^#
zJ&0Q#_jD)jybt?sDO4<(3!N{QoP$$7vK0-+Vkbx&C{|+&#OWD=w=>N0ie9!QjUUH?
zlM>_mO>KQsp1(imgH5*2qRB4iGw={C1_solje`&Aa$X427KHHX>3r?tv@Jx;^7W=a
zk|W~au^APs+c$SLPgVyH$Ny?@-C6#WY8gU6gnEz0?|H|J8rww3bZs0NBAFo$QzH~i
zr4{GmnsR#!j6g@{E-3^B(QJ6~5@aO-M~DRQ{|s2XxQRXnoMo^TYYp!yvbh~yO##DK
zK*-BY>e|#HO)j%%_t_8Uq>H@|ClOD`O6D{WSb?<d`-&H4an``Fd^NTfALh<xq86Ey
zfF7r^fzkB}%Eskn%^btZ`!cE2fy@M=O+%gtJNP`g9VM|+4n;tBTGG|8_fX{1f{yuN
zt5%##xn8FXoS4k7=Dt#@EK|ufYB9^hB*5{o9fFQSKln5JM<?b%Bn<z-Y|5A*?GFg-
zGR*kz9K;A!`?MlF6E#!!y=QCa<G$PabC=u6k86E_@dkX(!1i^-C7YeF()S**zcZF{
z<Y^nm44;cI5g#S0aM77L2f{w8F~Nk!l)>mqyx%u);w?|G3y-=B#q6AA87Q(rMy0K^
zqB&L^L+by9=D`(Qvv}oEi|}mwRUWf|jP43Cspun@J$}-d<)g%ZNgn(J;JKT5M)FrX
z*t15CW{oXHZU>LZ?z{Dh&*pK{AI&D7%cSlz1P#p10m|1H@Jv<U5c>_Wls!KPWNK|4
zeQ6Vco$Ay$Yr{?JpdC0xUqA3vN^QYx3#`U#Oq%pz1*C;xQ!f8^pc5+Kxpq(x{P@08
z5m(mIJULF-WY!g@p+Er>>5@!~)II{=@Uq`{HFqh(=A`4hL&T<GFQdnGWvOKG5<9;V
z-DJRG%<1wR&a&~;A-X+Mo7MStsVqaleWJaLi`#WQH)a@cNhm-BICjjfYDTuiiRzW*
z<?awh?SF*yvn!QA^JqKjDn%)L2spJj*>W%Pk4zHnoSmPCf!x?E5*dDX%kZPI{5|vg
z=?@xz|Ap|);D<`|H4k5Qu!<`N9DWMZv_U)<fd${+UXz#n%WBzF6dMt}VgSXh9Y6Wi
z*!rz|vRc&Csy!|hSSYi+ZO`xl=2>8Ngs$+Ac>bzTQl;<o?g6&pmFAN5boY1DoR2S4
zVCXGd&Q-E+>ESDZFuBu(SWrT1{Ee3%DHXW%rsVzCZ?WALKdZnm|E^RDDzw&ZGvwsQ
zvzosgfE9d3|B-{OqSR?J;ve^wM_u5(R50R~bt`c2#9WkCXT>F!t4m93b3ug!EXu3b
z{;Jk84-?x9bdV-($Kt2I5T=hr_q{!JQ2AXTH<r0nrAglL^H=Mc*-ZbeIpAaHx!G1U
z?DL3Bg5>Wk*AoGKA6W=!XRY7A&)5i2t3fl>)&geCl!_56z+yHwHii=22h*{7I)8eH
zRGO?ntXZSeTbOr^$!dm=8Ko`{<-J@|KVs&xte6ElSm-XyFN_yHs9|SwJN^DkZcqF(
z@yGIur99hIPQ)=eenwwUn(C<}b@9)$^9evIj+nOMn`v%k!k8qGPxxD{Q@eP=V%Lrc
zs8g%3aR#K*dA|gV#ca6@+oYS;Qz>%+^%l?rK0^Cfdw3V!QGNQ)8)vLzva<9vM_%7O
z1cYI5Lh>$fCg9!LmlGB^^fNVLE69i*?N#mn?>)(o`)`9}<6XCZDsuBl<g-;GN;nE#
z074ZtbTW$HA`uTRq+oyb{UXm0MJDDaWh4#JXYS*V2}V(hob0&KG%-||d6)H5iZ77F
zytXnt8!~6jT5dqwRypU;q-K9{W+1wAQ=tg()9-}(yxZ&E8~ZNj<Bhwpkyw3Ilqx?@
z(Og&JbHWbn9wz4Rl{_GO!cbeIi3Snk;q=J*`ucDPZ2d2c6E+c$s2gCP=;S5!FsU2E
zW;i4hpHRUNyG5z`iNezO_h(m`ow|L~;4DG!hO|QrecFe^oqJoBHjon;mx`vy%b`I3
z!^rcC+8amzYo6%e5j@tG4)?VBXR>f?8)30Xie-ZQ_#Q5?{j`uN-OvUHk+<qh<)V>F
zCP-z4oJysZ>SVeS+K;p&9kw-oj>h9N<(fa|1uF9Uvr3LbvGlXC)w698a@aGKB8dgc
zCQ8oE0s;&e9R3`!HFp-_TXl$s#e8{R9C_)Uy*!$@uEF?_vy;cxD%wVEl{O>6X6Mtp
zIZOXaom0AnrJ|Alr&do<7^#Yz=cbVlpz?R*rYu-$7eG1=(6YvzX222RoYB*PQBej<
zJe+u@z9AyQn6yE+E>+8??o)P&JekipJSbzs<4Kz)V?Ft4KC7<*P7e|WAe|&)n+u3|
zwtY2SrSL!ah7L)|LikVi6kw!V{}j(vm!&s)+@`gWIgftUql*cnjN~?$149iTAsVAU
zSm5n;JJKACJ}AtMNz)b$_vFX#{mc0XESp1(`-f)o_#Xbu22HM>zvS=iNN$66*Sg#d
zdwdErGTJ#K|D2-{phN-v0V}7hJJb{YqaX0oQY#4$5>%@W%qJp!+y`i`4u&%i#`gU}
zq`M6mk2GBs{=yvj%F#}d^yf-JGvGq;|LKz(03xZ`yiovem})6k0`&2KtO=3;n1W)#
z0&O@P`Bo;<j%73##o|>2duLX$S+3CxM4DyMvI#VscqE{QZ2=)+E^G9`j{;RY5h~vk
zm!jwq0oNdq`LbOkvJvk3Q0I<lEuYG<GlyQr^Y!?4E@7$=srvsUJFOtflf&SEaDzm`
zyv=tJJBCsXIaK<L#%j1&9=Jc*8D`P#d&!;jkH3nUQC7yZvJ9wgqu_2p6u7A_js<h4
z{x7|XEN^>ru#lp@A9=x#ydQ5ZMt4YcSq=AYYPbR(mf1kRHx7tQ9&w|e{Q(R6wkZHU
zbK+)Z>Vfsekvrf2igY-M&?}CGi<!2K5cFf-rRfibBRmJ(PHobX5FOuUYb_Dnn_28&
zcQ~!d7|h&S4ik;t%LTdpxES?c?0*d0z2Yzb#J=!jWJ|rj9dx2=&F5@VC1~URe8$3R
z^di*%9v%GS{rdfOw$ta5v$5w2c}QKMYE=@dcLP0?T8jdx$FZ??hiMoB3ECQ%Cy+tq
zI9Kh^O(Qi<<W`~0&7PhY8g;piU<FlJf`~aupW{?1;Bp}Jl!i1ly-=RZTi78!N_Oro
za`YOX7C6yHUbmg+{2aiBzvCg`$_hnVhBXHE2>NGsztE+L6S@JQuhCr#jGmDQ31C|k
z0vt}v%Iu@|y9_k3?!|aAp$Kl1RZjW0gdPxZ!tDqSJVU`YdzZxUaK@30ffVbtBz<`!
z;a#q|o85EZu!Lyag|8cygr8~3XZ}W19i8QeLA<{48qKQ1I{}*lp#4eV@v8{~PCFR3
zf--jo9#YgV*K1rh=K#Q7|9#Ngp8y9@1ZtqC$0%M$&Pq0^C$yL2+u{orB1{E)sAyE=
zuv9bq1%{eQJu*9CBG6h1e5=lkN?t2!x~S8HkJ9&b3VBW->k(U;>~)asb%1aGiHS)O
zD9!;wB?>&uV;yAiUzyb|*Ce)>%C7tJfrAEMP!uR&RiGbaw{&!z3Ge2v$RljwW~ud&
zrD9VdUIMBmC?NbW6>=@<N>wP)uu(%4rm@?IC2oJc!>C^L|5sWFcv3G7qsb$OiUP$t
zTPAXxgD^^eR#~;6WKbDcX=Fyr=5q#`+v4-bW4=kWucUfBNr{VT#g4aN=nx^&NXQ>6
zvYc%Q?$#=Pnr63rU$Yv0UL{`dR(!6WIbP>6?s>xej&|1-d`)T+aO-(IrTFjk?0Ts&
zFnn|nQ}ksk$@j!M_)K!uOsozDk}s@WyF{+2S)L5=$-f#m%Z;5c!p~oMGkijW-jj7|
z2gPijPFW#hoa1sbu$CkacGW+}PG4Yd4+w8FHqCp2LPJ8r-W^<GN{RL50P%!@!97T5
z%|eK<tbK2JMiBUbE-EWS4Ic<8%}DTu^My~ow#D6XOYLT9a}nYz^Vw`$x$z*l5qcHq
zm^JUeHFyWq+T%9d<qMZPtw2=5!(-eh|L2j9a6L82vyHm(2t%X<&K2qW^;F#)m#JfS
z9EMOe!-A{kOLOV5vVyACU5W_WO&nrUyYC#d(=prMaZ$W<TWWG=a#5U=DQ@313S>2!
zs~eqMT5>pfPHJV&2oEhr4A4;V=D^m%uzERSI4z7?i3FoOInI12FF}opqDIzvgW7#*
z<^20sPb}?c%46%Q39f~?uP;8>zrOM2QY*{N13_GZo9G+=A&rl-gtMw#qnKmbnc?IJ
z89$u>WxZWmWuYWpMd=@#mhP6xrzZi|cA3^cK>b`=C6cUcw1cCgD^RZdt)pWOykN5k
zoXgr)9X|}cDGfuw$=Dbx#fEVBJFE4N%_Xrr_!K?(w=N*SP?c6rARTF0Of7-C;{Aal
zV9|q{Fon9WY5RDoRx0J(6e{y`D`das6JrKw7SpF&ytw?_IC_YmY$moWhCSJLsgzck
zXKak*TZh-r92x02AUvyCqa~EBg!QyuB3dObmwQCatPmwh#^SL;QASR;igx7bPhG~5
zs3TRoRIKxH;8V&Lm5!3C18_DG0_xM^7xi5pIPXBsYwu}A^8&pJ7u)fF0|*+qGR5TW
zY?C4f@M&<C`6ag0mg4X#6JXNQZ_vI}J2zm;V2JZnZl2DeFiA1)pRJ>^Nyj;BVRw{Q
zyiMwM-EXZrkcpUDFP3cDx3;;((N|wUW;e{PrwM{i$ld@c<;=OM(Rc<>fHK<<_ytY$
z34T-ncsl@3;Hu#LpUq84WbsygsxOR|HUAZ9)HOD?b}#?X8(IcT=1mjptKLW62yIDs
zHA~{LgewX#dx!}B)Xr~AjBJ5KtbJclDI;cTGxK2&CiXj=Hg`J%$G7I7e>L<DE3)-Y
zdiOsI17{V<go%d<)u2K9#@^9agjEMoA2nn<uWpjM)?wFa7Z9a9)=B|H2|Ncjf#w#*
zuPiwDZ@9Z`!>b^Yjh+f>*fp#My04V5M`)VYbc-vkq(4{kaAv)@K+G~!jj&T`M>?62
zp&dj9whSktnVyXN>gfZ9P7GHedst;z=s9g6Fvuon2KWp|W#$@weVBd9>3M#&IPg?s
z_Tdwp!uE43AB(qpre*th9x|?_L_ZNrSHYT0CBt%^TuGz_xUps^R6^y`Z|zcd8JAne
zpJTy4Oi}rIeZ9?aw<(*OC8rF&ys>5OGugM{-aR~w>3tOzD6nW#36y3+wg<dtpB#U5
zi%5XCJ0;pTplS|3MKL!zsMTm8YDujn=!3!kEmBz*GJ)ea;G4|?+V`9``$mAerEu|0
z@7iPv1=>UAg4oH>DtPS4SMk4N6&k73@@|Qec|~+F7Mp%`hgvCLu!7&}D5fwI2udPK
zHIP6e8+0h}2M`DJVTzPgln+LAb6WNs8rqNVt@j55EtZdgWOw|EPVMWjr@hMFs<U5m
zD4?gIsG(z+#6ooBC>e56ng3wY@YBzU8?;biqrN-pf5UDWN@cKXl{3(Y;gnV=hjWpG
zf<?!lhHAZ=zPUNMJ@sy6>1cgSO4HZAzXQrl8X3fAyp7j%wL6YKTUmYHd)&LfHxe>^
zc6T^nz<<4?-gA>*U!Kl%hd^etA&|LPnTlNNLVw);(3e7`*$c5HZssgSz@X0RliBJX
zH4?MjE=SjWSVyC+Ekn&V@te@GnJanx>t`nQO?Q(wm9ezPmskFx{@T(8iewmmWHPwD
zqx4@vuZjULD>#dHICYNb%cY!Sw`mMDY_=mIpLwYkar}fn`&i-CQ1fPWeytP{GIQAr
zabVwl8NgWmPF0Tb^7}jx`PX8?OKR|o&fsRxJ?^i<XY^}6;nop`rIy{@LyQTVFoh37
zo$f{bS9tG%ZyN_9W3nQ*woXBRVaI>EojfUj^SXb$Ie+Y>ukhJP<nsZv^ks9&Ktjo7
zZ<43)_LhxGT0XOl8aeYC-sJt5?_mx|k(~((S$TPo>62UqxRO{Kltc3nZikaIRkfKs
z`i41fJ(5A2Y{^q<9i!|8Ku(<dAwy*d|D4aSpGBWF|B<YjJ3A+_z3t<5KJVvgO}Pi%
zM2n8NHSZ4e`s)$VpJYBKk^XG^^zB*?%&cJi{Z+tU@VQ!RhsK8h@!LUCUHJIsM-h@A
zODDRzcH?g0qF#R{ExHVieR_PgFCpQV+c;x)cW1|Mdp>njx^#t!NdHn<<V8C1huV94
zd%KHh@%ZgpTNgei@vRaxbdA?AwJ5%P`?Q+;>}}HjP-q~CN1Er=J0ml47kThPQ)uXN
z!VC6ic&h0g&P(+R)Yxp}>zZbI5tmU<_`R~CPeGT0@5Djtc2c>(;JNPN^I)^|nQ1EZ
zF<_%cx(wRcgt>mDkYr%Zn{oIXRo=bSEsjQq8*$>eK_Onb?YHfQ+eyAv3NL6?bdCY>
zqu)<T3kgEAROVIO1A@mbbb8)i1Q3jFT)R>nuwr_|-%qVaoLBB1lf`oC^A<+osn9gq
z_pwjsMh6d0v1d0FHwi^J-T3NM1pKka#i?)<IL3%QS}_fKgg8U%eV;I-xf?W>@Ait4
z@@aNZrrg9Ago2c^^)gYuMNQ|IIN>r+F=+FQWzSg2isW<PTF``Im%vFJAQ~07P&~SX
zEh1^q$Q4LcR3mH&w%et;_Tszk9eV~n$WoG>UR)%f2OJ0jQ=~iuG}0!$X$&dx*i&%w
z!clq)GvCUM2;pm#o)BLpEn?DQ829X-l@-dl+{Z@(apWFs>KfX)u#tb*WH7%PjmL5@
z3u+&XP*4}z@X62>QtjX=ezwtG6{X8CCSem|2u*dzJ&S;F_wqZ=`zG?dH%;}~c3im#
zIX5=U?@fyy11224nd-1*&mQ1At4YUQHt$uoV26*y;}XGC9Yp6BkXjt4il;OSuYJz{
zGVVB1?&#~K5s)POC`hKde|>)t@WyH}2xsHMX(}ZK8Ypw_AJN<liu~7VfGA2{{D;bg
zoVlwRWM_%Ml9_~h={U1Hijzt?G^3IG2L>8@b+uey=7c%?62E}nA<g~e5M#Cm=+}rP
zdg*8+Z6-dakcMnMQjAVFJ4$vr@Z)0(eGmTes@pe0)Yzkxij&Z23=GZl%8>#N3`}6*
z(8xq~bF4*wEH9%XJo^J0FxQ<)uV@&bFTmGpO<q6nH3+G?JL%_({sm}&>r2P%zOG>*
z|9C?>8>E(aH`l$eh1vIC3278tqX_6F(0GUt&b&fbw}%)ho41n9_p~+8@3LDSy6Ah~
zox6_C=VgPJf3q*H#sQboL3_G=P$8aQq^g8&7v0zTUKz*eWT){I(-)ZpVcXWoGaC)`
z^S&&qt~9ESqH&gE18UUeEs~4i1#W72g>{Uf+OL(4$_EQLkr<<GB$Eh!Gy#{jrI4;{
zvajvNu_O`~BzRv%oql!X5x2~s13l<r4OUFQblWg&|M~c6Y)i+?M)MTa<EhmJ>2-+K
z+g?&6(=sSbjnA%19s7MT4%x6LZ^AV}KLiD)crQJl)-O4{2Ac!w?#}bqMuFQCX^QCc
zPGM4Oi|6(rv^W8#Prxszm;Ui{E5Y?26nnciOlroJgonge@`#wh6d`LNQm#KM=G=tp
zyYS)S;_AEDBgyD~ko+-xS;WfK52e@tG=U7>Wph$s7!jY7ix+PNGoIWQLtOr)j~vH1
zhD4Jdb0?6%`u~`E%dR-PHfS>;K(OHM4UM}y1b26LcXtWy?(PKl#@!(_8r(g&1PeBO
zzw3Et=0mU5KVV;5>a02n(h5h5Ld>s)<>RhjhZMw>LM2C#AX<wF)~T`ArY{fek7m-u
z%x=ooFNA{w%&K7I^bZH?eosGywzbO1CR4<Vlc2X$CFXA*YwNqWzwCbFbl0whZ2r`c
z$~s(AZXqE{pWafNQ^u@9TRM#I+`*o7Bgd<M$g^$2ov>uID2G^l*G*w7B^YUun}VbK
z{1i6-ce2PT=yxY$v5wlHVqSC~)*8OiepM8T>-I*Z;9vHix&)(MvCfmicalzxtyC3T
z>!del517UNazQt(oKaAgVOz@{E;Im0!IG=(QCFpptg@6G(tI3@)!$S*7Th4#3p>Ak
z>Gpn&ynBWp^)8nmxw-MLV~{vUqSUhI5iq=7I)ilDnhb)$oLyaf5LLXqq5_E$ro>oS
zJB>PvEoWN$7f!1Abeio<lFry)Fqik)yFI1M0<rgC`f#f|J0%jbF#{OCYBlxn25_|e
z!K?Z4Bov`eGw?OEjr}nAZl65hl}%^>QBg>e8SC+KYd^>Eg{&gN)9de!eefHSj+B`v
zs}>^;eYz|(&6N3wr-66adVSwOh2XPwz}{FO37plbPuK)pUp#I{rug$#p}XR!kXDKP
zH|crbl(282`{OY9$Ps4ROIS|fwYAbz<ytE0(+HI*tk~|(#qxZ*i2gs$Nz$s7jrj^@
ztk=^Q78Y<zdm|H;SMxm%+a2W8o*!~M1bwP-)Wlw5BaOl@&3=}i&klL5yM6`?lY#l*
z(1bY3VE{@9^I}P5+5ECW-`6#8#7T1m^f{wL5SgM=oKP4e>hfq?qebF($f0of!4ga@
znP_y<DZsTUw*YhVX0ygj2rb3?Wx8z(qx6zMvW*A#!bDEKBbTKhzd<h@uBecDyb1G1
zp`J#yhi@@Pv>G!lx`g;3sj8Xufi?||Nxnsc7Xeu|{$B6NjJ$mVd#iC{u5$pVDJXMZ
zjbRBdnoKehcN(+Xgqg6Z&A0z+zHid@i*G<`6D_2dZyB;8OKQ@UDRj6N-R1Q)a|sKa
zKDLOmUGQpUvZ+0?<Y46}O(Enc2#RZNv#%}^cF<)Ak;N@6z%z*)$SY>5RfOS^w?{B^
zQzE+6{>GbdDNvfL$xV<b!bU9D?%|ZDc0?$^yL&K)r*3{N1&83L1#be4zWZfq(QDc1
z?=`VK$5<<tb}{z~O?$79StiUzx4Bknb^6DC9M#?Pi?d*$rg+|HinQ=|5gaKyKO^tQ
z*yX^RW$#_f4^rQ{y`9%sIovgNanCAB;ox7G+W3j3E0bgksTy?2>?QmH<eTecKTp<j
z&G8Hdmo7OEpV5}!n>vrE@>loD-dH{gw*B%vGMoGE&?S!(xb<`Ibj;gT^BVbtj&0Ie
zWxk%{lz4HZ(t$}h2?g2}O_9KexoXseA{176NdR0PMP6Sq@7u&KNtDXeqi~3@kgCiw
zx(Z80QC=n0GT6c}z=yBO!te9N{T2Ii@-ArXtE}>Y-!<e@UtisOu-!J+X5uMO_&n7-
zBRCED#i7P)_P#M>cfu5T?bq@_C|-sKqj8qw^SKb=V*Y>_g?Qt*md}xeSWzpLqJoMF
zLUs=m9Mb8*dDOHeF~oW_dkx|8!Pw;{b2f-<?JvS1MT!uuiCe-bx2KLSZpQr+!z)}|
zVts9-RM=X4;y^y8U{d?1ILihM2LC*Z=1&*}S>vM(82qrUJ?D1EKHhuFuf|bDTnr<r
zbqA9BqU4R~#}1rEcYUL92*k1zv-^15V4x`(%U*Jxzq(WEBj_`)UAu_USZ!5j>bacX
zrl!$!hh)RP{Y_`AqEkv!@v?Q+*x{Q>bJ1qao8C$(u0!DPyy_GGsgqqLG1c#sSR5)r
zg)Z_J3P~_~2U{iSkY-<a)GV=);}>2WEKpk^pSA^C?8|TLV;?kGwO%rl(rlK-Rv$mV
zVMv_5wsPVQuz!>J_U11!#3C~yqAwlgiB5rKtwI{Qhf(kZR9D~CI}p+_&73~w_qz_t
zFbv4rFZr*|Ue@%7AtsDwy#b_S*3%tr7G%FX?|&R{nl?uU9qQ?|KcPmyl55`)ctz~E
z5s>%dNiwU!z}*l>A&mhV>(X2A`%y_`Ifi15FL`Z4u@o8JjLc9HmvV=!yG&m$nYw`C
z#IUs#D~W*&a-|IWz6PVi<7WBwu6=*P@BKY@W?NsGN+3&(o}ZhSwfS1=YeKJcPEXEP
zeoymAE?3CpHYSb@N@XNuHI>7-7kH4pI}oa2Tk1jW*q{!h9ekbM$!<0Tzx0*TSCd9@
zM!IbH>sT!sguZAX>3?P~=y@D_<CIXX%a_YwnLBMsn3XPBPKm)0vWi(xQ4)d2g<N;D
zh{)vRqmd*fpH2a!Ewcae`)?$(uUM<z?NZa;MR}1je8I*f-n49;J|5dRMmL+4z~CZB
zZ{!@)mJa$A-WlfIiQwKO;!QLo^=OH`2MF{uy4)MONJBk6cf(@%aWwVoh%K3YcZlR4
zHKfcM+ESS{(GuTImPd#}7RV&<K{m7DgMam9Nv!zhkJ~6t-;=n*c7Xqg(wQ#*nN`^s
zgeO+;&IX}EeFA~T&P;ccnKn(F=jtymH@m7|>U;l9KKdau-rg~~-c_CM?N2}E$x-6u
zuLE-jN1j+*C;AvVh8q_M(5x~H*B1}N#w-bXosLxW6b%UVy!eT0myNX*nD`&p(G2?x
zTh_#Rce|iB`nKWXcP?aLq9W5#w_tJ;7=$c|%(X|>ZKD3LikRx9#4%{p&eBpqcM{5@
zEx>mNsU=xSXCEAB4_F<nZmo^*PQ{(IF)t}G%IYk*czA$BZn8TsMUHZ?OmN6+v%mSq
zq6!SGrGu=se_1BgY;bxJV5NpyiP_c|RvM-yBSnNDzr{QNQzkSlfi;>bjVu<wO>|%W
z+=nsgqjh5|{%sSu4MF{Rd4)w3j>Gw0$D~#kU!259QKG}B%~-s9HMa|BUd^>po(Cn6
z;Hl^571<^<K}nO&*&LoET&a13P1aUU)p&*B5$guB5FAy4WtD*X(1r<McN=9W|M^Ik
za?wSb@&&x+Ri06hnfB!6B_|(W7g98uHg-p9PnukxzXs1BAknf%L}x#QT;4T%t)aOY
z@(4qI%D$nd)GM_fHA2nk;B3Xx&*{}Plf<fd#J+;rg7+&mN-n1f;TNTuC@FE>3b<pq
z-0T=Tx0BHRy|Pp!KFCbKjAy4@3Ul9N?A!~SityULutyny550OS8JJ6@xVvKmD<Ufo
z6ovxerZkFhDSjv^T2I1Y{9%qeytxPecdI{x-0Ceu+O3Yi5Jw>@I|3<yWn+QIeCKio
zKxruZ&BB^~+3|1~5%KsKr+gc22`>YgWc~^r`dJzD*b7VpWPWZ_RKY|n9?B`JJX7a2
zCm;3^z|+uj8y?!zVH>f_2~axu?nm`<qV=G6$BrAPsm2nVt#q!XkI(Sy8KRu+s$>HN
zg?*J*7k!6NePwQm^XCd^U76b(wPsbu=O>Plv3edm@9?bv1ORlU%=M^oO__E&P$Ahy
znA=~BnT?dnYv%ksmFewY{mp1KQ%TBa>BcY_Sm=_<3@nr@A)`#>j({PR4B2a98t7VU
zFE6(QP}FEuZKIm|kW0soAMd`G57>@#tS%O}Z{2G?z@K&N4UTTYUx-DrH;1DxW`wqJ
zv`H&{?xwe{Vadq8B|dV?mWYH=o-ZdlsI`~w($A8T(7(Hhz8vXqn8yr&g%Y+~$mefj
zb8~ar=O)*@yvguy-~Up%(v_{cbi@wEH4V2gKv8!g*qT$#0+PviV@b{J_D9{L!#13_
zA)zp^*h+0Oz0rbC853&ZaepmhsJrp9(224LO$kPtUXizQ7Y*-_1%qj}h*KlQh8oAz
zbMUj<=<0;F7<2VM0p-4}_ghV`(^JSjruPT?C@w3NZ<-?}#_(v<*lM`TpKH(vJB52k
zA#NOA{NpU74LK#UDs<fMT>o~e>0B-b=T|vfRVSIODrE}1Zi@54!6A$`wkm&VD%}U~
z)t|(b@iAz$S_MC`$ZFbjzWtlfYR&paj(CxbDV!pI=arJiGt<CnyjT{a<el$Q;z^~{
z0B!-nqx<#xo248_R~NUzraEigDN?$X%`H<)--H{dV0-^x$E~CxKm%DffSPbn;tm;Y
z3peCC?;^&L!zu+xiw4TCPI46MH*wQ8{th<LVjf9~8arMT$eP(MxO?uRJ@uJ)BYoI;
zAz^~1o^a*zP^ubX#C~d`D1_(EWx?D>Ba=_mj#ep@#1n?syKWgJO~DK!Z9j=Im6P2<
zlMX|Tg?Mp1GeO4Mtw^4r3AU-LE{+`uljq-E6!jTc3;Prr(hy7tlb}Chl8xN&MD7Sj
zqJ^pfQ2v@OC9P6xu&?<P8FULOW~Ir=w}Q=yr^U-4ThS@%O6PLp?q<dG(UgE<_QA2p
zB9}(e1sV~r;lGR?50nm0h`EL1K**j+SRZglmzz(Ne3X91<-xPENfr9!4e6s@@4mgz
z<f%uh#o6$78V|hRer2%Z%98l=ETQzzbA8ps|8c^LmDn=U0J@)-VSp#55q1HjrXIx-
ze%RE;ZhgHaURu?R8dctnw|`S=8zy0(RZRZJz?MioUs#Nt`RS(@5Ig9{79p-g3A}W&
z2aAGx3!$?I$LEN;pUvPpoz{grSKdxS8EdoU(GoU?i?@N~;i5D$X-skDhOREzWZ;2!
zyh=2?yk<(Eq1~u`%fs4B!2X~oh<tOw`Sc{!zPK45O9i0bge?QDs4gLX^SsJ)Y-uSs
zQ;|Bws^d+XbTyFi{(JkC_E1$de4hcf1mLYaJjf}n#44e+OUe|DD3#TkKF`n<Tt${M
zJpqe2Z}L3}N2->W?MMrlwbY=+$SuTY3esxGW%>@OJELwGZ^o=WkV(pQ6LjB_Msgc7
zIeD-b(_iNVy-e9mxHJ~h(#->n9UU_o@)kG`WsJYo=-i1!s^<t0d)ofqJrYc3y)DV*
z$Ynf&$THizIoq$;3Gwk(XV}Lu;(`Y|I|y9$LL*CFUl};u%}rjjX1UL99!&5RuG2vb
z+7N{^kGMIvV1FIIl6^?886vO9uunhpq**I%HQCXrHI)7{so4FB_dm&`F1qsvHYw)d
z`nV*ZFiEN&&tWJnYB)b1qXyum0d1?Rj{Dt$hin-bMmd4eeX?c%UdRmRpV^b9m8Q;t
z0k@79DnZLR0V|Kz7PwZryg4iyCcbN@TV*w6%y5_N;27{7JS?JuaR~FKuW12RMD>Ib
zqOiVSfwbXlB>%F-?WIQW_M3)V)5W0QpT0gmlP_bu78l#g@F&kX4Q`o5kkKjWj2In~
zQYWXD>nDGU0M_=OW}osIh9_K-ZA9J_?qnScjt_MlU7IPS=QSZYCLVu@)i(*PxcR&Q
zTKTl&OpJZsbujVjPkXfXUr`k0<YDJDTL-YR&-twqF+u2hBc}RA=wWp`aJM6~ww-wU
z-x8PR<IKzq6ThYg&U$CgYtcyy_|fn6dY3o-?1F2%y&V}s)zz<8RX4P={3DyL^z!0c
z{ZcY_GU(dX)rCW>+hw4kuCLcO#h~6g2es9WLON_-y8E*IF(7Ft95yAMXI+_-edr>*
z4%$pJw`&dG(Oe>jcu6qP>>*YUQJJ1-L|q(_9=@DmcV$g~E#4wYIkLg#)>cT9Adyb9
zL!4u|su)Mz?yKLVAImW_9d9e-eX>_0PS-O~7Q!9}qn_<?ri%&b{;Yq}%n3#=TSUOx
zpg;QOzVh19K%r@~{EoOrV=8)J14E+BO|i`1FE|g*hCPFdS&cn(nSihfsHmvuu@R4&
zdA4AH==a3uceevHhW=g>yb!5n)xy#<BQK8#<N?Bqm%O>-8*}204O3{_)RZwqqx#!v
zt;I<NoT8{yK$VfWx3grnuPt-1IjtJh4S{n`wmr-4)H8`*q2;iTgA3mAKfv%o9N4e<
zqqNJZBf7{@%aRHDgVbD|`9KQgosX7VQHM=uUb!?uJ4g?$xDKj*>#hvCU=o8kIn)uZ
zz9hsGJ3nrH>3y4|>-K+S9MuUQ$(%4620t)<{i?aT3Q5kp<le0GK8YE{MO9)KHSfS9
zuXEC4tCm}isWr1X!Tu2_cNibT<9c{N16erxh%kTKCo|SgE6o}-j23tm9{qpCeMWUy
z60^3ZO<%<8c~^HwD0^{RhN`a4n}M;1;Ea|!ae9V9S`~p|5PDiU1R{V6W%(^}jE;)J
zbj5YOSX}`X%F#X}BO?WQv{Wbf{=lc7nS^<qg9-X+yjf_NDeyS4q;mh@JXc&dq>>vl
z%F>iw{H;$6c|$UwBdaNZa&G~(#V_vJ*QzAFgLkYA?VeCs=C+u5b6-O~r`%u{s}_47
zOND;_8-I)*jCr7Pce3RleOv*ZXIi_j+50zVLqMR!J>Oj0)#v!4eW5eu#@CGI(TpGO
zYT|m!ePk_rc9We!UUdO9T{W^wCrN%x17f`6r<!q%E+&<2PV3)mdcN-bi9rxKzP4WC
z`4ZV&d{XDcXMlAE*Pw-4uS%0TZ-o_glO&lYQD=UQDgMJXy+w8BAsyh4$gN{6mU7t3
zfYX;_GSU$pg)AUhcf4p{oa0<mKe!y`Y&SjH_a~;qpr(k=A>&jDaAi7Ryq4~w-K00%
z{xdUtwnk?Y%)8-R=z;^F6*_K)8LO}iUEVIl9VUA`ohCTHyhy@=iE{@=QT~QxBN8ba
zTPsV?1#zTM+$O#Wd@y6*=XKab)Y()j$*6#F<P>KhOvAkcqyH)u>-2gQSa8uSj7J7+
z;|99C&UqX)(^1a+50!NaTO_0ujae-!*iuzu$1vvA>VJN6F0f7p8fXpwg^NNZE<;ZS
z5~EnB1G+7F1&-sEw|9Oo8|+_S;}SqU8}+}Z*+QURAs4m!Eiz<dEp^HwV@ux40oY`u
zWg}hKj^E_h8}xo<$<Jv00{4T=5ab$GG!mt!OL0~6+FIB`5E3o$oArugoaNfRe%+V}
z1_pwUud04F%M|lzKBVPzqMq!GmkL`3oB0N-LD@1$X9qX7d*e+sXN~SkJ7+e<9;nt*
zkCKD2BTB6&T{^^KKC48+H9LG$qHbA)wq?2Fi9*-<GU<c*LyUGbgaO-L{xakB7es73
z8=&#KrkSPrl5~4?=IHEm@#S@;;Mwx&+3>lG-7B9Yp65+BFz$Jd$E0bRDt`@-M`R+P
z*8x(@MueQsG0$WM7krXrDyeE(%VhIEf70C&s&b~PPjHg}*3kR-h3zgH^)&qU3AP72
zy>3tX4!*y_3jE0s#Os@qAo(z}SEH6->L|Y}jfo!FQdNME*G8iJ`SY1^k*K3T5A8<&
zY=}(pXW~CuZ*bO=Xx!6!VHXeXX-!{aC(;(`t-z*52X&d#q@jxZU5>Fb9x!4a#GUf$
z{_TAD=?WJ)Mk}-QkIck6PR^B>)XQNt@*J9ejzgY+FIwK2P>f+w9I8{Dt*5vaB92LM
zsEi7Ft#m7;;sVy<X{|X+fhuHtA}RUHf=wYlAxohmDP;k~eP+G(AkfH|{k36qM5>gL
zVzVWFXb2;|qS?xW0UDwx98advLGo8&6J{|<a<kIWr=h?lF^<Y5^djg1PX-0Fq=tT8
zA-2%(>6E2i>$WjDpxwEWuY`Go?hJQczlPt^lFAYCtW;LT$0p=ahE7aZK#=u97D_(j
znk?q|`A;1m_DEG>0$fdrHU8R?hy7D*F*Wa2{_$?ei{HMj;eYi~`^~DQ^BX=#+74TO
z(Q{^qW$0B^KbDgJ#L`bEitNNn?sP%1?rfsh^4m8boc~Qs+223$&Y+<Fd~c#)`h~>z
z6k!>;&MCkRQi4sWlYsQ&w$vigMhw-AfUJSUjsO3>%_fJOwLFppHEiOLl-@ybRPEmV
zOK~WP2^fCAogiO?*XbIgy|yGz`(3Pu1TN98mbJiucXuKZ$dg$<j)enMdlLYyvkN&3
zpeotyc2VQR*V3Pd_%Ku80VE?Q=}UV#KZm1N6C1c_69-a|Fb*C<&0a+5ksWRaB@OmU
z-Cy^PKNv4h6L%Muy^gX3K>DQY94QrbY}Dx8wNj~ryi>)j=LSTqOd02KlA$|RIng!a
zNg}L-@mXM!Ir?<FD_BYy7q7=};X}ueZh_scA<?DTp{sbJ56FhNZTpuxk0TsJ{*z0-
z%f`9xS-P=bMW54}grQu`F^~*Rj37e3I!TN08O!FjXLEk!6!fv2@N*I=PhX-BZ2NfM
z`~6HDMFm!jx|7pcBKaguecWRGuglBdCYp=GZGFOBBF#TJ-C48nuw*}RuA~9HZVENr
z21~fv$VJtY$q)=vjnaaVPP2qE-&NJ)p50z6Wk&RpcJ}BjZwx$asaPvKt^9Xsh-Q-4
zw$&?}e~fes;9Fn+Z@CdH*Snhk=fzpXQ>Y)I6To4SPEN}(+6G^aFuw`LG&&%Zaj1;J
z(bWoDEAf(+N=hZD1=DNNC5iTJ=Xst0i>C*~=Y9DD)G(W%8+mJj69!={MJ<=1O6D^2
zE(EGN4<0TwA@}|5qw5o8%;{JOzk`wjqo!v0{BPfBW4|rR%O~LgWkL|SZJn0L`K3rS
z5HuISUT*&q@0wL(Ss=T!qN|*4vle3mHYEaUo(X}ht6&AcVR0UnqAGe8sYnPL>?84t
z`v3P<GI*^N8gBbI4HP!f={~{K6KQDe8X6P&Ck-nN6)+Lfg_1>7u<t-b+CnC@Qp8^X
zFe}70D_p>ZcF<3~noOO$M4lQ-882``E&}$(G51KRB=06R30g7}vMD+FxQX#r;hvu-
z#eE$YYv$M9lOvVJ0<hEPWWNb3EUB>^9d6kN+%Uc8ENT{*2U`BxyA(JmA&2G5-n@b1
z$!hR*>VLjL(=i7|72AI)pV-G)X(h7eBEx-~@KHkIZiGD+5RrAM)FKu#x*kYhXg(n(
z8avGU=sGica@b#*K8Ey1_a9217k9a@>qXzW*j_FiN_Q1RzTEf*%p}xYI!Yw<+zBw<
z&K})8haE!#h_1Ug8NGkYe!oRLlIVF|HoZj=*YE_t9bb$sEiHwO%?Q4npT2wX_g-(9
zFJH;O?>y4t_O9<=Ne*srly9HCqwr3$R_U&TimJ{cP28mP;Tkg&Ra6dObgZMvsN_{|
z>0@f;0oPl}C$Yu29wX|VbBu9%LqEgaSS&d{)x<P<(J}tSe3|Ie937SB&qjN!?d^=W
zA`_D(xFUkwb8p_x;exlvz4aZNd#O*_w(&pvS<ZRlO?Sz8U6MFLsxw1h8?-}VT2#<h
zlUQAns-ja;FeaN0Am$$WspO#&@^o|YR=v@|HE$iy8*bl?n?2VV7q^ek(r^b^!%4ev
z+)$L?7h1G%oJ34?BHBJX7BSjhZu|B-`u~}NL$R)EcYlM`tS7quyP)Xm<b*0P*<De0
zhIL$fEJC};TcSA@^Q8!86tBB&e5WWz$`$%Ld52yCF4ua_C$8yXUCk>Yeb{fMB{Hn(
z)Lw@1$VTUkTTsy4(lVwP*CilelB@TfK|w*`6D~4UcZ=_LxzzDlYZ=t!CRRA(@uuJ#
zFI@G((ID3yDEuGkL67a_<s68}3ELLE+(q^|k*s^x1XF$*ri7UNX>>0K;y+mOtL&_7
zF<co^fw;`07`cUtEZ#N#?~tS2@c$g`UH`I#`d4!tovt*65?Bi(^Cg&iokE6}TFS|s
zz?<v!E=~<&&qJmjb>~UQtbX~fw>zH&p+|jRN>kuPs_xZO6s+uw0{p<!wM3~Z+)GrV
z02t|f<(zO6+lV*z{qvL0h>wwTmx8=ZE2{#9`~_rUh4wPpAPa3U+Gx>2=-B|X@0o2b
zA+byDm64%u8p$okvS(d$U$f%xCGK@MG&Zb=h>Gr;H9V5n>#td0uv91gP`={giz)zm
z%3Pc`kA){E8HEkj8VaD4oq7MzW$Gz}r#SjZGMX7V-^yUW;C4EQZJM2rgEH$}MmFCP
zH5e5a<hv^FIU&eu%kf<2+}2xohM142taD@VxRzgn-3c{gS~I}6Tw9<G@F+qFxSHjq
zfH8QHQbS?v)qJ{~<J(qzZFu%lrSa}4d=G#DtvvmaPd4@KicP1F5g#jw32SBHRLJj@
zd>yV*^4pN^72r8qVCoTD#|M|If-O`nrjjRHl*fzM1(o6QlzwxRR}FN6>lKoIh6-~Q
z;zIh%iCw(@Cf563*unuIBisL|t@o$u4-XFzYjml#Pf^g-h4mU}WSKZ^JaeCJd41g%
z#lfBjeSwA%Ht%t&vMRdDW>!bC)h+@lVn<_%Yg8<0#)<oAAVe}ZPyX|Ne^)IFhIkb9
zxq=F*VQ)@S8Yv1elZO7(nmP-{(b|%xCCka6SN7v(ALjn~10OSjH+6)W1{W?aysizU
zZjKpU8no2V*O!1`RAqtXO)0Opy@tnXF12t2V>tc)Kb!~O1xa2_19SJ64NAOUK1+#L
zLHmlva72u=8^7-$^J+<i1S@g<x)cDGW_vUOSFq8bV%C&e;?20lXrmSpir7c2#!K2r
z<`?C;(5U^ctC2?Bl$7M@Vy&fO(7rA^I0w8Y{<{u!bIjl;DQu%q=MnuE<dQUn7Hr;O
zXRRNnB{y-FIv2;{PWyC?$mDDv7!zdS$Uh1nVUMpsOyu%(ox6EqdrsJ2vd>epNx&kt
zv4X{;4JUn#tXOEkSQ$SvYcYPZhP)(m?rXw*4LOlNhT;xQ;;)|EAd87Q<MWj%8kuQ#
zV!FUUUt58aS(F4-k|jlGsl82UY@?J)C3qg#Ye4s7DPK&Iq8J4k>S?D$KP(PT{+b}k
zq&#k0cpaX54gXlL<*h+q*|Ma-;2Ie~D$Op8C>EKH>)txdyFQXoy8!q%-Y95c*IanE
z9+}g(t|s^J$Q8Kkde50r?Iy;WQg>)$7h|Y%T3%K|dvwPGX*DRJm%ZE$#GR|{9MxKf
z7h{i#78fnX2oC`Ops^z!iERtkBT(Aea+0yX+|ET>qC6<I)tSP{RY&jtNnUk9<MGrX
zFQuGvNsSF~2652m(EL`}3;sja#;$@;J{`OIoXh99DST#Dc0GL;J7@Ll;Q_nkCX<Xw
zY5}?kdQ_XA7@)0>IgA)VoP`Lqj&80JwX@e;lq6}Ws-n)oLXUGbiW}<$iH3JA<Hs~I
zOo@jn0QuH^pNtxpRMi=DtYJssMEW`F*fxJB6=anuyI;`aFC^Q@S7Fk6F*DffnqpH-
z-KkElrdDOKZO}4vpO%-?e^bA>`HT)DPCMR8Im0&DOyeO)(V8y?FXq`zC6bJ#oRw#I
z%Kl3Ff9G9!ig{4?iwyZ6B|<}*?|p%I9Zs<^snuMR2VF3~T?YO>WmapVBt}Jc3)?o_
zp?6D8VE)nh<r04`M=_zI=Tyon4vdaq8*(ymHn)tF9Nid2t(OwbF1-89>gC5AN^g9&
zCcieckjd4?Fzc<rNM&{hqpn{B#l3En9bUzb6k)n%5|aUo;b0gTT`&8iYR6{YsFbu6
zehH-3jHeqCj1q7FD|v@eDyJM(ZjayX>V>B97s*jzOZ-pIJ%f6g)Fg~MhiEls5&Mty
zZHDXflpC{Gj}m)=6K?l_5!IM5?9Jcjlpeo*tL`rdXv|{W5!_B!gEZzs7~SAPVwnx}
zzJDTtE~eY+zEEb5K)LDqT_L8m&}L6gw(QYeYmnA8DeHz7hY_U05-vMmLC`R%agNy9
zyV<E7YOhnS7i%8PpB;0u@z3*Gp~bPeKcxpDvrY>sy@@t!8VB+yfmjzI8;*8Hu`w?l
zR;@~lid1EyOo}B>$u6#9(9qcWu_oJy2}_v{qm||H2^ke0^2C_wg9JIUhbKUqI}4o2
zAY(BV4Qk9RAJvUOs!<!#F%MNp+Ne&>b0s?ct!Gt9=iY78RoCR3X2y)4Y=$Q?A4TzZ
z@>aY`p2I~=sH;xu>d`v3iY>D6CSF0H$DxsrB(6gAx>S3qsHdDy(Bv}|)D^#uO*DN3
z=`~<SC&8XB^b-Tts!=dfyb->hzH$cQ+96?^-j2K|fc_6@#r^DJcoV3Fk}Lwm_wJtU
z0@-y>LB_lVCP%KYWX!0^K<?Dw8Y$ca#ANnc4HD1)`w*c!{lo;rCHR7O*>__<Y50Vd
z?{kEZLiZ)Nb&<-Fx@?>ZgYbt~1visYCFL?aR9qW_hqsKoM=ZLM-PKMHCAF{;jcN|7
zEtKX+_VQj&jOHhm;yXe|B)A`3$UlHRO<yZ4Q^%+<ySh$xg9gBi$1yaFK}+!7$j@d3
z5?v4{lw)%NGNgif7Y*K!cIKJo=-eS;uJwDOk>Ac2rj34?nWn1DSkH#h(TNzaBz@?L
z4Z>#zth+lJNV2vvwT5^9<kB(u4;{`g-j52-;%QMeokJW<O&p;qPg~RbeU}&7Z#V+!
z*mtdgIjuMucgUCP;cR}{25LcABjSu3VDOsa<z$*Ij$w|zUPaF)hvOw{#U*QmP0I44
z!r{@#GUbGif8I-;>nWovHWLYN4o7(FSzq&|$?z%)>uHmSXG$Y#VJHwT>TJ!-etq7p
z9z%sWXId&bbi(VG5i{wOOSG_XvKeLwnu^aFHfnGuqb38&RS88zD<`(2AV{s&U0-<_
z4=EzkVeYtYzAcC^loGT=C=zbfhiWovYq#lLBU>_KuuwBkTAEsFEs<<SRMmp*>E>(d
zP7o6j;5&|`>2P6++NOG<Q`qx6Dm90v1ztvG9teSgu2@=VGty+pmN(;#LFC5_iZQB+
zIjatPz82Qv5^rUQpc`Fm#7&q~&XaNjeavqd__m(VI>UxUtIRIE(+yZ8CZKQS;$AO=
z6&1^N@efQ0*%XMqXRj@*(m8Fi=%6;`YphE0mN_dHG9|gA=)9QoT|<j}+2TH(9YzfM
zhSkKKL-wx}nWEFbR7nNES7+{MiQd*TWVk2Fdpye;!_SNJU2HD+q)tY|KF|@4MfP)E
zsZ}1T`f@yog_}P_O%me}+}_Dyk;W4fJ33q!Rf*!kGqyxA$!^B2g6E{-!mF!<A8`+m
z2y<UIARE8?oHBtuhC-sEiO7u2K&qav2H&KXgoB?Z{*u=3Yb5yVKjo3Tps(w}7LFT!
zb%{~&<A7vmfX*@b;^O*NbYy%9vz3A<0o6O6h7RQEG3$*?<^u<BX%#2Q<T9BOp;kdf
z3bf%L05+zv(W=O;e=7MlcJ}R3CQ!v~?gO7-n((eiU#zh(cLooLC(dHx`GHWWf@v+`
z1f^ZN!IyV)a-F()_F@+&dCq@F#uT_e^^1h?R!Ei&1Tt#OOu{ym;N#m6;uW#NqBsSQ
zuHcFs3qzw)b&ApG?El1}@R(=4jZay=a^gl}x5rpEjPwM96-QHUy6o89<HlVzq`wkK
zsmj4vX?B;QLQRdJmtm@Fv*O25M+Ymj72_gJ+){tH?keu%y1w9X5;(<<I!oUw^uVY*
zgE;v+{8M5mcmvxTn&M9={m^1~m4F)LaO?p@rq~%pnhXN9zbQB_EtEHXPsIP0GyOdw
zD=7ZhD5uq^i|*^aSk&#X+H@Y*1IB5(PeZ1We^zM{kWnhDvsdlH8=YVNDL4VjfYYpy
zS=#;$Fg&;dZ+{N@j+h!C-%x7(1I{R93<|6xr#;`oV9YrLKk@1&BZ$w28=rOX<$(&L
zMFCzgI=DQB!ZRTuD?JAH&k`QeW6ijgEqbkiPoFSNv5Wb1Ba)XXoNy2vDKD&rp_8YW
z*pCy0?$#A?gvQ(HJ7_8Qh=)<33*C;Ubs^LS#=;hn39v3W#tQQsMu&2ov>0ws%+(I?
zJz@nJX^X6TWRU`7fWC%cZR4^ltt#Dstg23*YoyVd$;dV0jzd|ga;`ub`xPU*Y4@+2
z3>s2aY)Uiw5BfAD1o?D?VB5yg{qy}}0ipjXI=rl6I5u?VgHBiYGp`4h9W42%;F<Qo
ztHg>&<du$Ic>z9QxQJh>N;fh4`jHpdc;jD1Ror7%C&;VrOcD421h?1Mp`qvt)TJ3x
zW#qn?QIdOMU(xCZdz|31z9SC6>cRbFx8zpST1ZAlq}26gkAWd|mUiP0bUJo7;FWTw
zrLc!7c=Z<=ym}`8Kn=Y5cr3U}gAia7Sj{@Ey#GZ6CZ`<wr*E;*O920TqQHV8`7)7q
z0=z-TfD*wCRi&RH*hu?M#s+#68qQmPv|7tlrrtMRxBrF6H5?^fQ%|gkc^c5&X46Ks
zKk>9h?)Gz)cTxhiGd&?ZafCBf9zHL)SyHaXx>EdL<hMLVhmnd$Q{?+8=ML|eH7SDx
zYMBImIJ~8eo^Qi7D3MnMF}T_%bq($jD`9DHgWi&_TVfw{Wqb4gf*WxhhN~|x^@m5U
zHgz_N&;}u5pf6t%D(_71`+cE%ffTAZzKRE42^pxg$2FIjL^U1DvZ4Yq$1ZQCSy<&q
zf6{m3B94`is-%X6fQyaUvz0JKZ@Wn+TtgEV?J<@3FS(oY8L$w`GQZ$om~^wvnTSa7
zJ9#~#Jv&LU;433Z#<qc0w^ogtdZw4&L*GJ(JnAlY%L-3ZyN@eWt2zf)BZnVeh-s39
z#ZmZIXA<-WVQpPcNF!kkK1nWydZEqgU~sC{_i)6bBTVRook!&dohf+#T25<{l;1B!
zHquFWgheyq4PIgqoOH+NK>8sDukv|bTW@`59bZ`!SpLHZZT^Q7v@Okgwzj&fMIm%7
z&<7JLDls+&u#8g2B7+V_A=#Tu=TDPU5kC^>a6_U%0s;rn<PaG}qXtZ`?*23Uxe5j@
zOXYnJ4ld+q6MQ0jfxe~~>S0-~zhYjYg}cY3anXa@Aw$zffcOC$lCFv|_um98BqLgm
zRdJz-1-U7bq>+S?iPM1NKmc`6P!m2CVQmm9F#3}YMrs1Xmt*NB!gY|5fOs}W)5-He
ztA$U19300Czh}y62$ORDk{Q|cKte!v1%%el^hWnV-wa6!$re1*I0RllV(5%%TuuZ_
zuESIrd~Y4rJXXp#76+cgTGH6gn0(ICfaDm&3I%<~L%6oTLo3o2&En@##)2pz(+L(r
zq@UHvph}vlWN&q6&OBp#p{&uFH=!d4G$B@E4rJOapp`y_SWh=~I&WCF+=XIeDAc=B
zykU2y@q6S@FCH@oyfmX&v4A+kkb$#o%{F+JWtrb`oy&%^-~Dex*~^k{)j&jq^I1JZ
zw|Sd^`_)&#)7!51iqGDNF&t#5@ml+S*=8N`5ifcD#Hj1;D$tT_g{2#rbJqThZ>poS
z&_s^|N5dm1(9zEj1DB(Rwx!8W15o;B@^85O#?daT7O_wMFJovBauZ#HUBop^LwXR*
zE*fj1Ra{2X{>uHF8Jhk2!RQ3??mGRo8dsM6T0uM3@{9y{D|}LurbnjrWRCX}2~=(4
zM$cnV%DyS&tu%GW>T-%5<M0idR~pyRKE(pBtLufOh@45P*)I&5*H%0IKJ280#I1X-
zMGuQcn@ii??zGMwU%4zvDYNOokMl6;&WmQ&d(?{6!qzwE=8~8x;2cA*$2p{!nwyt@
zZ#P5|;uBsgFS5p;tSYXN(Xnl$Tr=X{?K0|Y&k12`#J-9?CB=kWrJ*P1J^p;fyH$$W
z2&alpA%-~-8srY!Cj-eA@^5+Kzzvp&E&A~;>qEB(ounoz)7U0)mo@<BQ=p~btrkSo
z(b8qDXkkSPfnJ28c<J`j-8GpS>5~IcKPeWW{UT{sYj-0<-m1^zJsV<Rrk|4=7h`Ab
zq{SH-XnycRw{TEH%V@g@l#UY%QJ(+L6X>p7Q5qUA$4fsND>wAifl}<a1x)*Pze*Hb
ziHFG}QW=knw?ySmv;R9iPEc&9mI#|FV#Ni!#}0qUi}e$P1SLXqjczJySIeAM{ueI}
ziyA+9gJm{R8Y?POX5T-iiophTqD@_QffI|0N-~y86js;NL-3!n!TTN;51)Vp+<?&s
z?2O;7q}SnvW{0Z}xGmsJzW45)ylEGEg!?ezeIsG^G<qG<=D_8b)N=;m@Np;>_+eyg
z{-hmwR%z@hoA`>ykbjt$F*lXH{UTa$ASv{M7jQl%IEBd6JDz5{jH^glYAv$9$nGw`
zu$Zej^IXgHf!+5M(>GlI_7BJF<>ICB{)QMYRuG~cU77tISrWej^-kmV+Oo9q_~-_W
z8|om^n9wmuP>ztGZeINZE&X%%r)&KJUHV!luD|W03y2ok^*gwrrhqr7ePZ=chpV3>
zYV2H#FQ|8fGAQMXj(d!gQoi=DUaH>^`={YhV1q_J0yWw)7rET2^{eGug{5WVEK5>G
zSqg-V7OVB#N)bzyf~%Qkc>)xcib82TiG;D}R=U0#ST#QVVJ=NfiUvht;#?XkhA3SO
zB?z4wpQI=;Y_GSTNybv2o(oo(O}M-{%;<~%7Hq3ocP%RP*IMwQF~Rw8@^h!^WPV<O
z>*JOH*iuztW+s!V+R9^tPD>YY3Bdb;jnn_!aH}F}6RPn9CL?zvwqD{%gkp-sp$zr<
zOas)BFtFkehk;J8UB|7v<c~@P>LgF%{-zYMtA(|kK*v2q(ACT|m+r_V&xXC==Y>w*
zK~ph_t6=@sT(wu@khLBO)+Q6t^hV`%X@@94mX4cyVFZe*`MIvg_gqR#PD*x3o$0eJ
zQbe;d5`Z6f&@k574Lz;m$-P_`%l~c9j)H-Y;`%r%bVODdv+C1+?~$7#refdhrD>sc
zXvNS=OdE*L_y_ocV;mOedF(bxlS9oCO{XYIL|nUbn*|h-M1;}hsW)A~Zq8v`$S7W2
zjLiGC{Jczf=7rU49=~}=dU%gd(~Fey{rLN*tHISWKQMoR6Rfk7Ck_xoOa_}t-f?xi
z6dPUr5s!7nc#k3iUdNun+pjY#8L*QW=>ZC$o)AR<(cw>x>wh^Y&-2Q0V$N?zk%6~E
z#M`i7>}c<jm1LSS1)e7OfJgj~;j0g4kiipo$W7*eSHG2QjtMsz6Rh%D!8csT5xfJ+
z_yowl@;dVo6&O7<L@xO8Jaeb{ertGi*LO?zD}&-n7=IQb1^8`43a1QX4Rxx&`kFa4
z|Lm&yizzbvP#LXoFZBs+=7Nm&s`T7Zl*;aUb2~<TfQlDd(RQ^5rTQ^GZ>gQ}jGGZH
zt;J;F(~4<X|KhZHIiWDL7@U6I^39<}J)GFYHY9ue6!h3Q9vR?OG!=Y7K8X$+0`S%t
zHmR%4nmOL+@hx!aIN0jxT9ohEd!W!)TeV^>qJ}*I0!W6Jl+q8SDp8vrU1CIviOm;v
zQcz$|9)nbuTjDF-)q>(TsasI3Xrx2Q)IG2&Ez0^Uq+19YVS~q_gW9kGI&R4-85EA)
z9ywi(ebd{okoptpWT%F<nYf%bulHvhubUB^=W$8FUuK>JMeGaDv5@LA!vDew^gk;Q
z<$fE*m9&PBYT2Aw$+6dXd{#@|>5hxGvV{W&dqygRk;0|2g`nbBVCmQcb=-bdX#{tV
z&#5HRNRyemDtRe3e6FdWv6Q<aA?pFOqq=I9iFx`lmzsd+votss%6Dt5vwVIA!J6cB
zu#qhxt9<qi$?duMghk)nOpf2R<I_8BmAFrb$cSA5i>8|=!#a*g*=0JK0?%;N`1t%k
zAIF(XU&NcJQz4Yf7?NND#DdyxO^QoER53*{vcw6HSK^#aXW&MG>Q)(NFQUmfu$G{+
zk;vLfLWxaT@$>{0>0@@kop|@UytDs~)7^MYvh}Lx4!HkTC+M59OFagb33%oT48g%3
z^RxTlpwxvSi;16=p-Dxh&p?Fm1k0q|qQMAtf;|y(NZ?lBXJo-A^wmpZZ>#72yBX&|
zpW%<&nRgnHM>v0XpcE_N?Sjw?M)%tid5yc@TlKc_C1#PH3c<+g_iJW$zdv!lp_|8<
z;i3OB1S4M449%EvUEkK^>kN7((&&~uyzeOay7;JT+XLPxA$lJ*7Sp~P!m{Jmp5pr-
zq3M~|UnKjQ0nZuFZSNO>5i=h>hG+ktuZ28xA11s$kgxcxw<HkjNa*f&V2r<q?w2O!
zeO0&Q0B|A#dG7%Pa;dYzhW<a6ii{g&bg+U4___@(YwH&4-w-@7yiU}963w>RVBB-k
ziTo{njn~BC&A~u3nL^PJ&pk<!?o>bDnt5(extvFPOVjdQ_P(;t#<u)oudB+16*H3D
zS!JCqSi$ix4<gH|xFrwkEznEa$|slR30Ld17IF#C^<JkDg#Fz|J}$b@$xfz*PXdtE
zio#1CQ4F>MdE*pm<;T&@%B2AsKZrb0QMz<7zRxdxQ&(5cNjE*NoQmdyWB`u?pK(^V
zHjQl^3kIyDvX3FGyCzSkel{*gRw-7Mxk{w6Vc#Fl@l6gi)mDb4Ay<Czx_0Ygh{U$n
zOMBg6q~kFxqt8}|ruI}yQBKux#Jlt$nj3prU9fdEg?(#(E5`t?6YMS$trj|s60PvK
zl|5^{Y`XgNJU4cnxv>C60}$J-R?bOA@<)Vy7Igw-xFc|I^o~8r+3o*=btGx2hv(49
zL8!!#ZU*gEd{a~gN*9BX@E(157yP>Hi8g7RGsbkV3%NA}K#~ON%FBjDzK+fV=$J^4
zAZ-NMUas$pmM8Q>hj`j>?8mA{G0_L725ORdGpMM>E+XuV@~qial+tW1T&ZzmEy0uO
zPde8RB?7%0f>Q&E{@D>I<*%pP{#n}U82sG<FR3+7ez#)=JClm9+$e@UG2DoL+}(p-
zROFdbR!9IN9no!%B$*JI{(<LJcpG-WGzd@bq2uN+8$HREMQ(pdGfm*L(7WTasL&e?
ztf;&?Uw*(lZNa~!1g8MdPPJiRNQ$|Upzd|2jn@cYxvTu!XzVj2N?^@E6TkgpDl}Q}
zqWJ;E<ab?yBw0cohoi)W^>G}y|1rG!hlk|ftmL!5(7S5)-}wmXh4Bnj6}Ig2>xhpN
zl0(&kclM7zf&Vn0+TQa%_!5v4-oIjOvbz(pefu$BBo8L~7K2zXdoRAVQQjKPf>TeM
z)7kyB_N{c6l?fuCIdterhUgzH-rp#_!8!D76t`Q&6~p5A=PSgM)Lv#Bj@)<s%At$i
z>)QHoFij!%%2It);7h#{;4utvaLwc7iFp^`X3nk>4v%aIt1CtmCjD~@P*2E?Quslw
z8K!5rGBP+}>TfwQSFd?JtWCOwQe;ab5g4|jaA8`|Sv<S$o?nw{@;key!C!-*U-()I
zs!9|t4IP>!WEUn{{jvs0Kj~1L3fZk1Z>je99);a~%0;`lm27HW9U31lyS%)Tz@k}n
zvZ=U^4&wX~SqeoxzcTy*Mh2iVIJPWN*YXWnpy{d=#e^aRP~b*Vrx=*8g`-}B`$8O_
zLy&O#=^FGI7(`!-=n`$Ol7c*7cR$-YlYzuSFn7pUtIJ3;Xr`J<>j=bx#B7`RO_dG~
zvM8nbLTs?)YcOh+$efPBIj=p2FY1@uU9w%&(q$I>_z+w96!?-+x62ojg=sRG)t)b<
zD+)nC{zL9L1&ebhE$~ZGp<&5XH^e=BhY*Z!2YE`u+S+6vBIN>FuzG8gz=dMpWu;1i
zziD((SSWScQ_G9&bg)Mo&!JH{(9+IwpBiK^lKGG3+&yT)18GqCYk3W-nC+dT8HlXA
z6PWB%Jmq?1s$rNc#0;`9`xo$Em+Kom%@b;7TYgXYlEr(o7W4@n-;w<M>AJO>C%lTS
zE-RrudV3z)Xg4dPM2VhO_GA-9xkHklNPKP_I!q99vIwcHl2(E7jvh_nMv(>&^7^UI
z=Wg$0#5*W$;j;?(rQ9`>zwZdW(%gzL3pn!dR(AZ`kgqe$Qeu)Jp0Mq{ZzZAN%e^u2
zD7bM7`ew1odWxxJU=^zN(_*5;(LZTo3oDabOsB#BjYh%nLmt*)z^mBnY?|wt>3tHf
z;C&azYd2)(c9QP#3EUQ#cjb3SJWqI29e8ZZ@5g<7*1B8dT)I_QImc@96++3rX8NfA
z;1IlzA`w5dPQ8qneRhE{cmr3<9glUpALqQzAAeAiyw9TCxC*@_{FjN%1}?w>X(xDA
zwvP>8ao%|ZUz1So`rek`AG`mZ+Forc|2K&vC~<7|sA)(3eD-8P(7Jf&aYh@`OxNjC
zw;Zjyy#19IC9zKHb_YgyIM{JI;q00fjq2Mf+SW_3qid*xKj-gy92itIX{2Q2`?K}t
zT_UPQj!JJBU|r&{yR44gD`TB-ftu2C67{BK-os#2z9Ur`S>zL}e-Wj61@gI|u93c+
zgUdh8Ha@`Z#ulP;EaJ`=TV!Bk``KqzI#RUZpA|d(F=<tpc!6ec&`jgaL2V3KOR?cY
zX<^?68aeP(F=5Qrlc?aMlD*bdvi>ulu7&T=S89}s7*$b3xKHecd_`+LS278jFcZlx
zFO?Qa{<#7|>1`gk2_;P|g_|o2b%VzlkenGTH<(XFoE%9N^KzjWS0kOqK*yUU`-h$$
zw|{;fyCh<Y6xY^5Y}3eip!3<@a~qvBQ>t7QA@X!<Q|xxI_6*5w@b5L+7Wb%lykrBn
zfN5$ub+M-@L7|k9rSEVpfo8jhdqMFmyL^-*zRm*w1EhZD|146^vC2Da;FA$*9pDbO
zO)oh72Z9ow&BCY_29S4Rwz9O5qRvKI5^rjOqRyWDhAWCiL^_$YIauV|cvG|l4M=2i
zYT--#SC|pAA1uk$kO8ir%jCT!zQMnu<7cmckj@G3djii6_csLHlHl!Z&GA*X^Zad7
z?)k`_Mz9CyOE9Q4RX7oLVo=+{@SS!q;sbTdTPZk?PSd-KHMZRIX-+XPn-aRoZk(&{
znW6V?Z5_HoTMox+PARu<aCz$=ay>^rp4$cC9_Ktd)a#St8<1E2`7|RiCL-a|`D&Nw
zQmAvxj`#53AQVz@jPnVH+!cg?oIreYSnn4_^1MrOtoSw&n0=i8W(3reRT8;Q5qi*m
zAL@I$x`n`a4R>q}13NGjJ}x;#{8-;BxO`x25?~D!q)*1mjZ9Rr@RfhU-AECsm*k9W
z9kDJ?!Rt4+3Z`?Bw`;tc+;+7QFjX1*YZ__SxU<Jay?f+|%g~>E@~a5a*F~{gOC3J^
zOUSt_9~{ij<OF8Pnby2m%2b-Ol!DadQBAo(2tZwjwW_S2npz@@F7;#ZRskvg1GL}E
z-i{3e6&#6b-7`CL7XABI#*Q^VD;qT^FN%k{|4xce=<?7_eR*K;#EVfulYT6Bm{n=$
z3+!*6;TKK9$;~Tp_VAm$DUod7nX?u_H3})QCMb{LabKUskI;a^!lV@o^)zRc&-gc|
zuibt~=!ib<iisy(%(uu5s)a!Zpy%FkTV9us@FuV@xT`CtBgR_eenW6`ODJ(>Q12LK
z$6oNkr?Ltgj7|-Q#z99)hZQA4VnB1lM(>XO-g0t|7{&n{Oe~`zuk>=bq8)PUzxA8^
z>lTikJfIcKw(}(#o5B?NoH?Q=FZWE7>fHX&Y;rFfR(Hl(UfQ0-@aFY@ZdD!X&q+q~
zx{{Q@rLTh~!kjYJc2W9_Qeh$}u73KarxdGZV!!cK^b^g*W1oJ?qL<C-h-W;bl1fWk
z;ZX+L2%2@g@C&`JCNQt9m+?Aydp$*(>pM4m@Ebm#PWhdflI*Jr-F`Ok`vc1aLB7Sc
zy^fQ3{m^`u*m>8Ud{_#;VUDA+TCVMSj1&u5gg&+cgA>)G5V0>XR|n)3j^l3#42^AV
z<5xSu%pHS@ey^~MQZh+5Cr_Cff%kp;aczCmZFDnaZbud9YE{;8BQ4l)uY*h<=!S2j
znwbiU%$ycG&#1cb?)s(>BF-6-HwmTg7mvsC++6-WTf-Ohj}xI=%crb@_Xs?suTShH
z_Sk;sB_#VF&mT{oPdM*oLSs|guS+Q1cLlLj`(1JJF{x8fk?)=jj7R?-w#zfgLQlYT
zsiz3(F;tbF`7OiIx>Um>#f;I=ogjwou7(Z02*Lg4ggAtNo?rI`h8jb4No~FDXKyEa
z_6Kt`WnETH*rxO`BTX>`F;_{)Bec>0X}5}?IKR5iN_aq_x>~!3m&xl>&uw}@&J<QQ
z{M<I<-)LQ&WpM;F1`1+VxJZebD&3e*0x5jO0JpPQ$FV1TJU_eIuAkA1fW&bt78f6%
zLAy>P{@9U=R$EOC6VxJh8M?_%;|@I~23Z};Mo8iin+8+mXWC=#r_h-2aqybeMz`M1
zJhb7#6F<3h%83AF7bZfMQC{I`jaqDLJMw1i^MG=rH8w1Oj8!u7-~)FX)*enxSP52G
z{3JGOfAP#bI)fOn#}bhs9+Y0_C&uXtzc0otqv8=kUQTtK+RTF>Tn_!)<p1I79HZlo
z-nHFEog|Ho#<rb_Z5s_5H@0otjT?L7q_LVwW81d%&i|bAV!hw<C2Q?jzrCO5zOO5g
zBTPQn28vHOt+&^0EVfeb)DCC2&v=~fdOau>Vz?0|y!r2gDU`sp94>5)T|7tH2clrw
z%wkoA532vIE|n}wYk3mp;!pUkI~1k_jgPV^4pgnRwRCgiY)x2hITC8Igx$H!l>?bo
zwIvFTBff-LNNPRV<abO|ZtzL%%lYl$_{S1I@1iLQu@_eB60KLZfm$XbZ*$UD7}D&M
zDX+vg?Oq!)OGVA`nJM~s%w{$Z*mczkBt$^op-%9g9lt!!(s_lfGIS&Fq9n};&jg3N
zhQ5Q(9Xo0=W=3~A^gAScuPUCZTh5&xz`5YGg5+KIn@s1=IJ>fKUxF(1oxD^V8u@=Q
z(7*3|>~8GcEf^i-KN-F1AlDd@BL%tD?_v<Y;eISYZqI!m$N3+Pt6d)+-`9KAGA1W4
zfreN_`2zZH?=OxURd4G#d?!r6bp|pf?>>Y3))nbP(UEAovd-2=-DXD@0`IA+o7df`
z(BbSUqwFh_Po92L7}o4#Rg>rj%m>?A*3NItQV|C4E9dQ(?oF$CWoQfbZ%a%)!zgTq
z@6X({QHA4I>LXw|CI#|HzVU{+2hA~{iH+Uv3Pybp%B({8X%&O~e5h`Gdv)~C#62Co
zi1j6DSfKlevKCx*Lal-s;}4&oAs{`el_s<^1oWUo1W?T<m-<}X#Mb7AoJOkaMq3<}
z`N6~`rsXmB3(63PD7hh1lr#ZGN!^NXgWbO{(Sh}lD}F$)<)^Z{uOcl)g<P((?u_XV
z(ttZSseHSW9bf{TYy`3MEoCF70Re`HLpMDIUr9tl>t<0zVYF;?a^+zHAL_b^9m&+^
z6Wz4vH0!qFa+|R|HX?r60vFMw5}MqE13YefeQIg6l{Rp2z4So>5R_hB|5!Q*O2na#
zko==6o02AthB}WP65H(#6NB2sBrb<$VRD*<0>ENaQHBSL+I?f<35tJ@sfWN}R$zal
zsejfL)%s_5TmE7vu}fkNpBaI$@D*2fQ8Vj%S|d+xeFoSnD@;7VjSS>zu09*~sRM&&
zajXr*ZS4M=wpgSM{{m@UC&!%zZGG$DN&Dp;@^b9X`p?U;EF{kCXj7(AZQzxtcU~P1
z!81TJi9(@n{$HLve6>Wy*UVxHI#i!OeBsJj922aN_aahhxlCh>wIM5*LnZ;2*|FG|
zVR(j;KXh~1udhtVFL&W0PfahJR&D(mOrF=_8tj<o@N%kfp6V>?D7!|Lv1}|awlWK|
zb5%aT&ADXx<wBslnl|JQg9VLE9UV0@oHl(1v!0t2Ua)?k6+GF_Y+S3a*FbI~?ZFKt
zHE@{~21iqC=}lV@;q!^fSJ{VXETzafyAK?)FeR-(DU>t@9Jm>T?n{CmS&g~ZBmeFE
z?GhXzc`{5e$mgCtf9vD1_T%I=^kdEFSJg4K*CO?fAfA2a`_LE%w+woUDg>aOcK|6&
zIPi$uOc=9h3od+VkeVwMc0o`(c|(Xij&3Jn#0YAB8-=5`$54oNw!rD%gEg#5<R$hI
z=ygAvUxae4JY<DXBKWdO&k)Kz+M_k~zNf;s>&$xMB(j`5lQCf{zu7F$|C_aI>X*h;
zm2O}->Yh1Ct8PU0>D*2y8wwr;2P;PK@Omwzt6oXM1S&&G=+_*hqdZSW->i9m>SkY4
zm>qmSMd4wqcI@4W8Fqb@K`W%w2H*v15+?{xKV3=KBznAK<X&rsWyqa&$S95c{_!&l
zAHaZpby4+D>?zBNn~2RzH-!f4{8?aEG(k>;Oaqg?7+#cCJTaL>S+%Ovs9lskB=2V7
z8v%@u4Oz{;Nt)eExw%1Q9gMs^$r{!X>lu`2#9$MSvDmNP?Z_i+QHjZcfQtbj5gKIm
zg2keo6sY<YLO6sG!;FN_$;(I6|0RviF$Fi^t(1T^MK9@aLt^2tTE`C_X_k7yNBcUh
zR~3sx`+N5r-jdLW$NyY&7yMi|a|236<l|NP3P0t-hvd#MAQ8o*2R++La?#)#bWKY<
zYqKzd?5?=-{>n!btJ*V~8x%>587Bm!z+`S#hP<K3d)&^`+UJ~YgW@2NIpXVS1(D}z
zsmIriu8HEs5;pcE@$EYULw}oyX{dv=kT{|GjEO5?m@keF*zD!g5yP2N*gdk9ZfAW{
z-|)O_Le2!i&-1f-M(7e^2mj2gJE;d_r_i@ePaXrw_YllS4Y=JQQ2Q;7&w^CCyPo#j
zg#g2T825$;?v`V@&U*>ve{cq6kF(T5NPXmlucQ9A#5)&0=Lb)f{%1XLB2)hfL}U~(
z$2@CV=RYlj7<FyN8pLJWL{Fbh6z?3r3qqAq@X2Uad4DEIqt2xqmW+6L21(sK6MGhH
zp1`7#hG_eE#^2P4&3X+?)GXJQHX@;t{%9FOjRdNY6&7~W1N$pBSHwSYLkNLLGAOe!
zCM<$3Q97av$7iwWbD@2I-A8hh_$r*y5}QDdLW`&v*gooy)u)NR5op@@=;aAew<V(o
zU%q|u=KniJsjy;F)!(N5N1`lgMC>%0Z|;m=h665cWq7+pxTf<v-_Oz*y4}_fGW4^q
zGAlj(OWYH*J@gYo^A%-o#`!-qa(HF%(ABilDI!@3vJzwqc2*&W2KJg~zV)J9f`$V@
zT_8rtpJs+tG;_?)Jkvnt<=u|bP-OU+<K2qq2$E>1O(UrJMlBz{59vVya-$&>gKB_Y
zVLF8J5ww4B%N3$u;^(9E<<#r0?7w|&GPX}~Fqd=mbsykK{)(>LOXAa{0S_jFz#+*O
zP6|CfzTp2))AfA0_J$qzt0fNm>Md^Dx7wW02M+?JLS0&yv?TuUmbftv%Muk)6aIb$
z;%zJobA0R@ymfr^%+2GlTuA$+H?cgd%w6*9PrIUBu^K}8$H(Ek0)2t)!Fd0WT)l<I
zE#h}fIt?D9n~8Lzt=f-?T4C4VZ`IQB(dtvGCR_&FofiDWkUBxGSwpCuP1aLN;mD<H
zEH{#7Guk^2-oXs7s{#9B+feXK$fWNHaCAlso^9(r_e&F(A=)+3rB3(4B~V!_j@z(D
zX~ipR`I7i;_yf3=|8DWLOAV9&<Uz>H`5!m_d6RG3%#J=@_j$Gb_C8lyTC9fI)dS;p
zQw$xJcu7Bb@=kVDPcBslCSWU39Ip^=TcXFuuA}X!on_G6L`bQaK(5SsiN#0HH`?Bg
z`G%mF1cI}g@iVaB_VIA(-N@da=A1->^GZ#DUo|t6y8xJ3fCxF9pmwc}!xII;d!d;*
z6!pR~{K2%*&g-d+`n61F$ipTm!D;P#imz<QGNAK90(A;sxT5BJ8>{o(;NB0FFZJJ^
zkrO3rC7nU^mxQDM@x!{@n&$bNeWnVt7pSTh@55@v?dxO;%5k>&L^IZ?3V(%Fu9Ma}
z#H9w(pL&5Lb2_h`zN9!%Okqmfb_=qGjLh$cU=A&-*fjJ`>cx9Ok;D9M*W|{fCbiS~
zSy!=@Wn7$t*NXGo0B4Ym`7jytKj+i!Tae;j)Q_1f=P%1#&$kjitlxV0$*|Q#N8&7S
zSGua3|HhV3DkEgbi6Hz&m5%tWU2x~c@4oA-q`udV%7S2BodH5}wP7Waj~L7e%=ki&
z)$Ck3#H;G&EX<%N;b6HnVkVAS?aqVIB2g$p%^GT=j#(@%YJ*x$iA%gPs)|MkbFheq
zxyjhu{M&exQ=SU`B{?N{G$1$dM`ePUhnZzBCydRX4$E5`@7*RLBf(PP{r`N@|7#fh
zaJa{uY^P*d3EN*P%J%D_ep)vWk%CklJU|j&Au@}tV<}EI)gQ#@D%{0To{pF#TNFCb
zD!eq?I`4cvWB`1vM^QPHvgHon-4^=`2}NO}Jk46<?eq)X>IB_34P6fD)m~57_w?qr
z5IOh=m>-O>hj>bQ39+c#V?Ysp%T?AuvdVI}+obEi;v$oQl&<!UXp5#kZOmSg{IrST
z1Y{~R3L09f1Gi2dE;b*eYz>z7xrH7GJa$BAG@pe1^Cd|@QCWPDL84d;Yg6eS^V^pX
zqK|E%kAugNaA4R&oRZ&l@@r)N3-5c;$L_~lG3<+(z`ZsWxlN4ao7tBE@$tk(2h@gm
zm=$$<gcw$abo19PNSao0iFOa6q$6Us+K61D3i&yxLh5w8*^wb2$C=*6w$W-@ujCQg
zjy&Sfx`*z#uL5a>oQ<9o?xgbK{S{k?Jr~elgE0DN_OUcuVH~@{@-xcy^O2UF8Z){$
zDhWlGYYcirS<RH%tyeU*XapcAKDj2D4BOQpt8XXnE!XbndTSVKoVgBL{$!>?+24KV
zwDG8A-5ghGJWDZQ*E1n3AyGmg89H0!jmX8uocDE`qRX}q2_QaJ#|G_b3cY|rCoj2#
z=nQHZHU(v5liVL!I`x-wR1>?e=C@^fvnHe~vg=KTfFMbLzNMv@_+Dp(Wm3UHARuIR
z4TN(tkcd>Q7!q)(UTxK)Uk#HPq<)e?tFO^iEwP+|QdvwUht)5hY=N&31%R`$xy%!x
z;#mZ@wF`asxA`L+SVyUg8)U@_isP-!RdKrP5^={QMw<T0<Nx2imH)1!cD%LaTYNk#
zOkX9CO~G)`;ikwc%LFweEBU+9O?eW*r_QS=9N=1@lEE=6Nn$Kl^tRIW(x%;nF(eZ*
zQYh*4rq=Ih-X#>>@Dh7oiTNBY^z_%hz0!a6V-llGs`Y*1`dBi0B9v(blMp2T5Ch(e
zEi84h?XTN>P<nsKGdFPd@d*=kaP`j)4u-b9^nfwfBL^!%@`Io(@~S_|JNRO09@fk<
zu5s+N)QV1b<dc^rlm4ud1vH@z{vUMDcZZ5$y<+4ZkDJBWzZv|mgdW|Wv_AGeptbA`
z$>rD}E3zA<YoqtMkEkb;tKvFuHX<9#a+O?`9npCke=|E8Md+%3!S7`s5(b<MX&Pmg
zXNa6+Q8O*yUoa5{hU~O-T@RIQzr1LCUW{fU(HEzZ$dPAmvG(3EqD)qlA@~@2ajit1
z^1d1rEoN5t+SimC>(9a!MY-a<Vi!l&Px?)Rn?UC^72534uY-Y=J|z6<h*Lxot188S
zzzcc~iPX4yaT5_vpM^)DMvuIcT>B<ZJL~lPqOjx#Fn<SKp}V7Bjk_lK0P+QKe(3du
zz>}l^fekgk$EoNjJKm1^mk*fe@~~7Xza7(Q;=l`5Q5P3;GHPIWs@9Z<xxk{wb0S+6
z1qfrHiNeBk{nYaIX{mkw%ZrAT*il%Mh@QQNTd?DHW9<6}&cx6vL_)>|z0tfYnigYi
z5F#$6^uc6p)&7I*gu9Plf)PTAz@0`376*Q3GS-u$@{}yVqQyS?%7<$^SEmPae;<B&
z7PCln@KrSsn{gTh*w`dk_(oqEEsOmR&m|lcAWG&cnt1U40d~Qi?_9O7uN#}8LibXv
zG8B>sU+JcP37bW^>9DH80zJl#a*m2Hz)Fgim2g|%L}?qoivaVK`6iD%W%i3snz+(t
z<{kNR>(h<CbNh$BFe`cNqrAa`f!CO|ov-+l-5%MciLZSdtqRMX0h)MVE8t}xkl14%
zyY{^~-(vCuZe>Mm-S31HjCcPqq~GJ1tZt)NW?^P1Y>>DNw<QognDccCtx^vzl!PMT
zI3~H`$Hh{1=l#%%03WZZp4I+(<LiqypD%Z94Ur=xu0$@QiW&8BDfA}yO7DN}A526K
zyoyx1OpN!*V16CK`RJ}*NWMY=w_0LN-=xlaXue8J`eG3D=raaj4KxR}n=BwlG*$o%
zI?8M&pJVszRt-V2k=0$Xrwenk@P{tY4N_mkUe6=Cf1L^`#Q$;iUvWh_8nmqpfY8cf
zivRq~wv|k)|JgW=!-%GTG+(;`M*w1>M^IFE^EyQ5=0cllMcpbIYkaUFcPWH{>ecDA
zy1EGL8ie@pdj8y1YPC7OVRPHBL(~K&F<ZB%F%9LAApHJVb0vpD|L74dBvGTT(Y(#L
z47L^z6$a8s!$@Ox?M>{vNjJvyItc%kP|s`cwxdVYEszN6D-aj&QvKY9z6)^rJSNe&
zN~S>8q9m*ARI=IJN6tTWSCq^S7R2>WyebZj2y2#|YMPLBuQ48^RG%(N6|T$>tfH`_
zAeEN4x{578sm6oCfYr-1ODe%6BUwY=A{-N=+M1sc14-fSsPvL7&}_kXlP<zB3dl}3
zUmHB+CV4PB8qC)ng5r~+IJ%_WpZ*`-gQC}P7X$#a<k5%doyPq7tkvdclid~UyIQ0M
zIc73snagQ6$!clQ=!|Mn3`H4c^rFtqRT+6XXiX;GDW+Lj^gC!M#Jv0SEG(2FxM7|X
zYIj<mC-!o{%YLE29fVrH>b?7xIM#7eep4BUG6C|fwIGBb&AEF<4j<E>1eq1Ixysb%
zVn)sW#1e4heQjY;m?*4Rghy{<kWuJg+G3q^2xqimS>Li1OB5TyiUfPSz2<mz4(o0B
z1qZ4_G>jMZ{={2#h_A)(?zt)ztdLi79Botxv)|sr6Tkn9{SCX&TO($0kkI3o50aj%
zLFTUD@OvI|8CV&MpmrJGjueEjX`Br;x(|a3u^78$LeF}`x>j{U{db1!d+O|m&rH+@
zt+5$(6C*4EKRj6xY#Zs&^miM(f>Q|Rdk(usHI<&CiAik5M>^!bOi+a*oAhZ24e=)<
zPM)0+L3nK0`L*Z%TMIo+?H^q{4i`T>?8(EaLp18Aozu<2V*tGfZhFlE%x|?J6e>7f
z^XpGSw}bz@+@J59gTNrNLR++_qu0|2csMl2qqL<3jUc&skCRE%HlyzbCHJ+5dY)B9
zli~|RXHtOA%>9ztSSb-LKq#RQ<Ij!9O9<k(N;6pk6Fb*c(8L-IU5g|hJ(2>S8DY@_
zb(>Sk-DNybd=q@zBJfJ$d5MT7GxNYtL*p$f2yX5z9?zL&p3j*Uj}VOJAnYWhCQC*g
z0~*#Zuc^WuQaFG<sFHXVjuO_6=f3is$<P>Y-xz`rOA8q?)3Fp<2-H`|nuVh+%L?1w
zEJSQaadg7clH|qLOA6VA);H_hQ7@UI%wr-zl)Ci#D<;oSS=!=XEF8f}?h!R)eD0R8
zk?5ARg>N!k0k-06y8F>dApK8y#d<AJi%E6a05`V5bK&|_8Mw&8e=MWyVQu@y-UL+z
zc5=2I6fn`4pR1uFM%Gf+-7LPsi|&M+evLED=9z<}R4+CuKmKn~>p3-~^mg@qkJ)L5
zI&Wvc+1IecH{R%_+M6(j4fy<*sl-lNebd-JzD($nAyU5Z=!b@hyqtd&WD)J+1$JQ-
zqs;1W+v7&2l!R;gvh_8=!Sw{j@!Heyg@5?#gW>jk8E`0i5;)#ap#s(YrstmF|Ac(A
zQ(LF!2XP#-NPiw5%$bsQ+&R_&4SVB~|31M}m?0DUODyPhW_Rv?m)+@b0^Kz9FC)!*
zDFL9QLPKXtLv0E%$Gc`N2gw3HV^{w8V-LOb^AA2jP~8Q8aR`S7g>EemC(y&og$HIk
z5Ry+f$+*$_81e^~K{)AozWbi<e=c6$UbBxr7<G9BAjv}%KG6$koU%W-GL&eA(a3tu
zI?UL0L9U&e;>m@wZ1c8P%{UqMg7ToNw^r@fK|Ps`MyyJA-JbG@!4)_sP<IOm|1jP4
zdn_}SD~{jB2AZI3%Q{!zi#tFpK=G~}+b*K!cd{y32YddpI2Tci==yWO)4vkV+ZE6O
zZH1pkhrV7XdF$dl#yrXE4$d_{Zw8m{I_&1I+1S)5_v6??_n;ZS#mDl0x~Q#gXK=w&
z36P7=wE(q;lEIvMG#w}{x6F_UFkc-L6dNgRldm3>f@_)yPRDGdg`{fcaS6YnpFm6G
zC49mSc1@$qGiE?a5_L;A0F(Ikpim&Qtbe!tRUb&1h99bgCHnhZ8$eSclMF-p$5XBl
zQ&*S{R>r=F%!KtU@o(0j{uy>#)J+0(C|DLasz2I&bM@QHJUho`f9}>;y?25{$YdO(
zarOvGbI^h$cK+e}piZexY3HzOqv(!eEua*tb{x#m;+gKl=>#eg5tn=a&*%pagNg-6
zvXA9%)B$;LWTLR)Bb9eQy4GA^>vw>@wNZe{Qpu!6g2gNYiBJG?m-EKC{gf=LD}{@A
zrm-bF|6cDF)=l)pY)&kJnI$2pt!-*-(?)w|J1aTw)f>gP&_VJEh7M~vDj_Xx-L3=8
z2jMVK<3D8b4$QpbyK;SVni(oF_M>5ktUEb#h$bGN(|aZ^eBXINpnKZ!sOj0J8@y?b
zuTH%MoB{)mkN4y;il|QsUVe(R>V6k%bfLyJ0#W=`c2+UH@5tA9sp#@U<Pc_VWSqFQ
zscu$jiM6%vObu1cQ|H#<B@EhT4+Nx=>;Inrl<sl$bg(3}c(A_qa`B_(KiSvNL}}tS
zv%m}$I#DA{G@Tz;r}LG9rf2AyYs2rZFgVm+OwH_r_*;l3UAx@@o!XzKNp)PvKDWDB
zy)Mv8duvo5|Eu;J=qn#@b-|WWe;mZF()NuNJ7knLLD7U2hT%v$bgJKfl^9P&g5G6P
zAa%NZPCaYMjHqZwRcQq)!7gyYT&NT_zM?`^i#O3GIue|0n&{1EMI3V~?CLF9QWuCC
zV+!mGj`%6yORMh|F8NVM1@t@d89AwTM|$i)$maIH5-_cX%y_=5e#QF%9j4Pu!z_Ir
zLq)JD0r^#;3E=3}DcJnZhO_y`$JlzessZG|6+>-BwrU=l;&8o<?%y(4c^aGZSP`zH
z2(q*jy$z4f^D5-0;tzF3)h3l8`liAswUovIMlTjNr-eL?%xHD##Un<#Dv^4F8}mx7
z9cl(Yangzu<J$jovD-vmT2b6=6YN#%;C%^>sn)O0{US+6mexz#iKcx~NS)^%Vyxxp
z<{OgJTrZ)IiEkk&Qz%R$UT0}irJHBd6h;w3v*Y{!@^{|MJ(O4dT@+VzuV=5evY~VG
zoe!$h-cA-4JDjji2z4cbrNldz|3Ugwaj3?;`B!z$iuda(3t=c1Hb2y))^OAIQV6{D
z801)hELpfffjkv%iyTH9Yv}2IOJ)|rk`b0(uVAFGE03W3NtrNHqC7^3u_2_0U99a=
zxp=jLE^7~s8#!#1zkU2En7oFQ(^YM|Qpe?WK9Ik-)AN8ddb!Ume4c-QeOo+4spmwZ
z8<PDwyTSSB`oQ_r^XTrM`0<YO?hRSapSf$B>c4l0u$sAB#J>eYM|BEEAHstmxH<v9
zkl(Pb20Ou0VIfC4qrXNTS2vz69^3NYF;0y-z}0%(`IxRsL#9Y;66|*zzw1Ps=0nFy
z)n>B%hb!um>1RuXo?I7hRb4y_yFQV|Nf%#43or?^m}H0ERUaad@-@kqpM&&}>EG9M
zCyu+~xv+hw(4V(m-hgCP4OplHVPyc)q(}Z;6+qjwEh#{B^5KS_Y^_}$>&<KU<`+s{
z1}}jfoq7vi;ZmJAyhw$Kx3YIUAaBz&DyW6C&<c%<j-mvWLUqXP)HE(vMv;J!5qU>&
zUTU>a$-)7=)wy%nOUol2LZTE00mm_nj=0L^{rJsLaGzE@LK_w2qs&4RM&YRlVVKOT
zXt@5}lTdE17hC~hqls(1Ht+U-wlX9cHDR)G0Bk8EB#9@|^!A6NT+X}c%THk)T~W`^
zB?a3oSmu^Eh7IBAUYG>#8RY!3QL2jraXk)^RmIa{gNNQiPWS7C3oy_vNIg_Hhdu%8
zAU;$6ts%`Jw=_2Os6Z+*7p~ID#Uaz2Ubt}QALgMLC`KZH3jUgJV1yiKS_lvrz{Log
zsc9A&1$5WJ)*FxK&HH<BG5PemhUNN6wBa&o@UN2(X!lqXJW-ZcP%c|MyI2})^p9iz
z84XHyI7cY+K$?`rE;<_)z1(so%&ERupUdZWz=8!HHM>-4{L<P=0gW4~I5_Mffl38|
zC?UcajdU|Nq?mFp%jrkk!(~X(Oq3^<HUr~20Iq!ZIB!4GqK4&PoYYMgzkLo16UF4Z
zPuM&zCOWl1Gg_1C^*`%<cSTxQ=?m7qq_j(eN4}Fp%L(H5e*X}m*BK!Qz3%mP@{Pg&
z)Y(5q=&9@-^JZ|{`vOqHyC}3d6q~Kz;T79!9FPWYuH*Fa{-&XEk#iw1K%rrt0p6#$
zCNg@A?HQhYkMvKm$n(pB+{?YG(<0oUa-(P9j_CFGYYpCIA2}xnSH=y0;;guVtG3Lm
znb8+_`2ltt-n`i-Usxsgy&1XXn4f12BIr5>KHsJ<>t#D1?j7XxytbU2Ik~tvZMEg*
zRwKrL37$HBUs@Vm@Ox_M_S?D|9i2M%%c@Cr=vj;0cjxmvrhfkv{Uq?5+N`jv&<w#j
zb`ERf7`$j9RdZTnJ=tCPVOF@c8fe8<?<a1WM$+*0YjGiVXp}TooH4an95KjuUOYKB
z6g}Q$7;E>yp@@=T%ys2!H@gm2c&rJd^JAi#tXH72+GqgkpKS@WcVvkx`LYbfZ=bPU
zw356w*5dK%gY^qi<3L!wfsMw5q&)Zp(Ca0fDzR3!2Qn7Dyv6ZCq6icS5Jrt_bNz@&
z7$(t~z?c;I{@t;ZiX4E>tJzb@MlY*<urX#uOSU*k`(iXHgkfAdI%mpZz?g$fVWx%$
zt~x1$%S5lr*+al8Daj0;PstIivS24Jqct?=C<KocpMBOvZ@~I(VJ<vF;e_9T3r3sC
zf(L58-GeY2Zbp(G-qgf<vF$y6r@`w~`;O6hOU5#By(pOUOAYeo69Ks)53U1?^65S;
zIi(EzAqfEdZy6F$$r{?~>@_vXZr<s-KRdK6CDhy|=Bv{}O)ZgytdM`vI2Ef?$OO5@
zUjYAi;H2WWGjPb4p^MxO)17$LqN7zZ$&xKUIBnN!<mn1tRVHdH!4{R&vw+-(9*X7!
zl1AN~qi=E@bx^O%V5XVGK1G4mpv7GytQW{n`r1MZ3g7)MY@pHP8r-CkP~(yUUSQff
z4`yHYc}cj@iDms9L?Q?EMXdKOn3TH)d1ROt6jhpcze`vvNc=P_7gKZLOap}>XuQP;
zLgIMJDB6p8Y7j3EA)8Anxe#`j5}xs8{A98kb!!_h`ARPIfRcP*Wl{9YvSN@QaF;Fi
zDVV|3hGT2Fm-DB7Pp=tUFaPA|D7=P-ss`fz4W4Kd%r+MBG;1!?2>N5H(HrIaKjJ6T
z)lSEqpyhoCYIXhWu|UZusHk5IqOOVF3#zEmVM>S09pw+OFW&gRIBWcK6(Z`%A}mSC
zZD3ba&ubnjuupbnK3%S;nl0|Ks^9R4p9^Y-nsyIOiBDQ^<q;ENzX#F#0s>E?q)D~q
zNS>!FU&cwe)`x1N2&S;Od)>=X*SyAvD2oyEFOH`W*>7dLtP`Q&tYp)yJFj>!;9tPa
z^|=~aQ!sN)n}X9)x6)xA_eWTDYyh*vslj0ajg!$vChxg%FtOsVAYsYXR1zpU8LN<j
zE2gZ+Fx)3|A8}jt1j_=;<+O&1OLm#0oJ2PjCsl+dMU%uNoMIU`K-285j$#r;J?6QG
zIT}1tqSRsu3x|f*rYkUH>JwiUz8+2;Sc+NTB9mFsDm$yxnsrDVEi{@8E62#7Kw_>k
zqR^Ihet!f)e5bKMT~9`DW!3C-F^xW`RcK`eFM(mi(4;%XmRp(aPE5`nPAu(WGN2n}
z>5hg==~P|u#{bZ{p_qW38IBQ{>P3ld1pW^itpb%Mw;yJb1dSn!(w%5+CwQ_;dS)5x
zyN3W?krcE#HuN7YpggJJ6aGv?m33Cb;8N@Nn=Jq(W=p_RgL!B3pc&gQNCO_!lBmnN
zn8$Jjb;4NIn8p9F-+8x*7LRN(8~aN+$dPJsIrUT?{hkV1G8xS(%_Mhksw@0&sT{D5
z2G71s;O&-c=ggqbGFn~o!1kMyqD`#O;`Y`QuN(mY0<N;|L_Iw>ENXJbzS#@q^7*7%
z?r{%!)ox9`jvz=ltzR|Q`Xb%<j>PqEC7e|=f<U0f_2nPl-W?sfkW|c>PFlYc-V=Xs
zFyd^g1<=Oxta%|*adK&c|Iu#a0CEbuH(K2vn93mpd~&#8r~_Z_@47iCxC`W*5IB~?
zjB#!(nwr6S1t%31T2rZ*s>#KyCoyLZaDX>;Bon$m%2xL~W!H%4EyA*NYka6~dLaI5
z``U5eP!0~CsIw7jP_3BJ5}G*2i$RM6`8tRVY5mW(aV!}EC8{jxi)JmVK?i=y;Z9>j
zVG;@dNJI@M0WpGrA8(1FTO=xYyLvDi*OO`)TT~cs0V^IYH>s{+JtDNK%;{WkYj%r5
z9Xs#>VQ>*WE{)W{QR0EW<R4id0U1F2@;9p9iY%Sk;i*i(y!SGw<YzXPKg9kLNu{9L
zKp`O;stIqE^-F`H_S!|bP|d4Wq%&5<3_Re=2<2)p{8#g$=pzTk;DiU;{d2TC(z<r%
zmGpK<nT}q1mGb=0B0d-qIWzQ79I$7xJB0_^XMX5jsh@UNQ)TuJt{E&Ufr(!&El|Om
zF_BMY)v<@a?lNlDWF0D_Gd+Gsrv`}2hN>2h>&{pHH4k8jrlqpV+Rlj|$@K9l82CTb
z(i-?cI+$Y2-{K~9e=B&)4i1V%6DvY|YzTwXU1NMd*FjR`z%nL!sH-SCksZ28lXemw
z=suEE5`ut#LB%6zbiK^!JQ~{+9wj`wvp#S4wC)=~N-LYN^6lj`V-Dm~WGuZR)<NSC
zMb!{ho|6pb^qf3F(b>F8tron+sQWe?Ji)imPB4wO5!D2?n*S0UI<8M+LA-u%Oq}Q^
zu-#P&I(<rS1H}7#_8(NWo%eRjZbjeVElHmShK0bXHd<N&SrrwJL*pBNo-Wp~qr1gu
zt<qF5VgF;Nq@fv4>NT!IP<{BBt3nIo!3;30)7M5uLE+`+H@C5Q5tKMq|MJn#zml81
z)6fNA@|iqfVzzKL=t}Q_aFO0)($54cDs<PF$DjM0(@nuSEBfbyzITdxsdqPa-dBRm
zLLj%dJ>qy~kRGvshs%FH{w-P$%tr7(?auT)d>rcD8y4=K1!FjU&Q>;d9fxdM>~FTx
z4NWX;qC06ICef;q(JvAVe#HaclMQsN^G?9nE7hC5APC~myV9!84ib_tn#?iy67Yhm
z*IJVu5E~s4?TU8f3?WyU1eT?UO3MIzhboOo{no47yuT5RWcjo(9s2+lOQUDeD>{zX
zV`yjU15~0W2$$D6w9wCZ&?#j!CB9W|z<ULX?TSSr7hrUYz38`C18Z(fyn?L84)3s5
zW$$Dns6as;z(mns>gjw_SUG~Hma!$aviJ`~0Lmb;N@%uYr;d3Z7<kPfj7@Q2PnPwe
zR~6EcViKgO4#pu@nSG5&MRo`@E)<OvB%FJ~1R(3v<>pk-sNS9mIgHkfq!Nh?NsUF$
z9r^Wg4nJj%-njp^^xhQ|fuM@UC{~Q3vW8zMAuF5n_#DJ^1Ws2BNc~I=MYU)dXd;jq
zjj%u<Hh9Aw?sJ%Vvbvhdl8`ssWNFi@ZELg5x&f-EJnE0?IWSo5(Unb!8eXMpXEbjb
znc@?X#nYQ3bNc!cl7t2K?J#O%nBae7tP}a9BBn1;6ToU*obigpj&GZAkzefj*be*H
z_;nn*e@m+lg5Qyg*6h>)If4S)U_E<w{EIsG9fZE&_qGM7%+df+p5IxNt{rBcKeW{m
zyYKFk?1iq41t(8s@_haNp3l?l!$yUTJ_c~$k)@XYq${-uPmDMugb))=^w@C~Ai3`v
zpL^=IyG-JP$FqgUvD#MyHl!!BXi1uL8r&$4)OnbHAagW^!t2-AkMo{F);a!GbwE5O
zeE#ch*_2g-HI5(z&97zn;ubUT#3Tj~NjmmO&9V-OvjEKxUEXpi=nS4hD59l+ZFom^
z$SXYG@$rx}Q?*!krs+a~Y8SFNSkK7Icm7xxES+(_77<C?Gm}tcB_t^pJ+`D<!4H1O
zs}MUbM%~qs%%b>uO=$nE2G@-SA60Z$#kFNU99iRvwX$*Cu^D7j8Cs)3-(fq63Jcdg
zUix)?C-+3Rcr86pn(1QH`}^R|NuqH5_I35Cjy8^i>OHLbhras?maeSJl=eS~KT<ul
z05La#e%yOJJ#Pr_eR}D9eA)9;7MvGYzid0khE`L(&qEMHQ(ywMUAhT%r%Lxk+wI*Z
zo*&d0RNxA@sOvQaETSUTb7lOF!a{a|BQ0e(M8)&1=lrTFIZ)t=yr+NbiBZ2OVI!I*
zp3j39ZE>P2k1=K7BRmq2hCqw4%m*eHJ^Weu$Z5TNhGFr-)hayp;qPsMO|Iv+aX0DD
zt7q4>;d>5z8K_UDiH?MVM?hyqg+bp&qF7Jx5S2AV8+Rs(fDsj?_@hT5O!6nPSJ>ps
zAiP`7V&fvab4#Z83)7l`!Fahd*nelsU-8bw`^V283@ba&an|j}n(lW;&Wdv%?tVQN
z=>GBPc6n7%!<=eN%iY?_$ml5W2H$Zysv{MDR~J|fAFsEWABSMqm2T-8-HZ~58MgWb
zHw%byaF!!k?l>sK)}9=OTF5gTcMe8-Y?<iqlF0pu=D9opH-kHr$}>U=)p!45Nd!Is
z&j~D_v{-d071(yA5@_mmCx!A(p46~hW8Ez&1WBbS&8&hF(j0j$ns^2neynyl=j+Ae
z#86XC+TMHYVQ_125N>svKX0Nbxe*r@@9s~WKwJYPN@QgN;+ZY>a~$QGv(6$qV9%S0
zFP$Lj#cY{*)#2vKJPW4#C4K(Mx(y31Aqp8O59Z%lS-m7hR&--eq;Ko^p1|S}Hgr5?
zPhxJt;(U6SER3Yj7oB%h1i?A<+(sf4w6|W6DDSj*cLZkr8gXZtzI^x2`xSsmzca#_
z>sm^D6Ow1$TXT2cVAre}Z@jfjcrMgczKwDVY-F6I=)+$L%WxNrT!h%dvFv?r6jDY!
z?uQiCJ#Ml&hy!Vl5QKbMoe!{4*1K@9^xuGxsSytBpmD%-6vsHKYGLxC#)!9se#Emc
z1l6XKm0;ni&qB^mX)U9d2<4(s&g0hrY=>~l;>z(&9e|rvf^_H0iPiS}uHekaxh&H`
z1Isy&T!WD4*H#dV{Pdz#Om3k=iuzW3O66zm?lwLkVd5lq*DgbE*5fmmTU<^#JlIqr
z)*agdztcu{&{|zCERCDO=Vg%-aG}SKu^g9Tdq*m#Y0QvUZ8yAjCO!EDABP33h97)K
z{C<5lmZ@<2UB*&gJsM6bv}mX(#6DUHwN!F(QDHU!+0kVwmR+!}#N{@|CYlhD6`oX1
zS$7MT1LC6PpqMB7r52i$KobkeamqOwl8Ek_A>eESUD&Q$VWxhTeH(-$1)%rUM`?fR
zI%nZCajQh#)lE7Ik+(s|^K_mG6~k8|+m2X;f<{__Ztc#z8e8Jb6Y>vhB!FHm`~jq|
z=ZeyjUo);mH?^+f#N8MGuwzLGfCq27{?#jAAhc1)NH0Gqr&F%F9$gCvB^FPM07PmR
zf=~N#Tr2NhciQrVG*k}6NYbOaQ7yOXNLgeN(Q_NPbuRbMbA-ISMfbPZg*<bK+u9V8
zeyHEzp|}$#l!uIK)h6xY@hPC{1Ubf4FgUu30j78bF0f!o>r~OappUnL-a6u2JFDBg
zK$8w##w>)&n9y{lhe+HBs+;;4ILM4L4PP%PZHg44eEubs2ujk*w;7Yq$d$~VaU1QZ
z<XxBdT*_&aChCAa{6$Sn^qL?E;@<79IJ1Dz@+az#)BJ01qgN>Au2(Xx8eqTu$HT|a
zqp|E`Tz*HQ@L7&2FrPhFQOC&0LWL<ox+1N3onzv<SewcCEE*q2dp6DgCi^39uygD5
z$@y!z*!}%O{{Cf69TX~7(~uOoeu+kLi%j4wpCsV3>R_leU1XJcOc0d3srI2IsN6D|
zZ>of;X-9!p4^8}pR#I=*b0dmonKcI!KrDp8SJ<77+9}*jJXYiT;X(W7OXn$rq(C=@
zz#~_Os@oBCwSr7XE1YA>OTCZLlj0~rH0#E4v<E>^I3K6=CDDpH%tf8THz_*MPyWbr
z?eLnr97Lw;J<ATxwGl@hsi7;Q)sTPKuJt==**x7b61l^V!e`L;<iBPa?}?pCxy4Jf
zhJ?4F%TY2=e+U1%n!;C*y+kOEHegtbOR<X2gzjvl@OajpkwZV;tb^)|q{S5(GNl8k
z5GFwCeQqItp;iJg5H?CZBhox*W5jp@e@-Tdk1h7<5!r7jw32_sdoW=M3#ERQ^!4f@
zg20eNP{>jtCC8xJk&r*3!rLaStxH;~t#!R+HAcI8&IBXYQN+X~3M4IgPPz@+;_s#g
zkf>MW4U_KT@#5^MruIc7`M{<rw<e^<Q?zhWkfKZFOvrp()eO<9#ToCk(NkXWK5Y%x
z{?GX|PznE4MiCl=?zB)59&-0Onk&}r&|;=$<M%%3`qS1NkO_ru1D(!fAo9Sbe_1_7
zZ3<+}JQW)FTWZEv)&eX%M5!lraA1i$roCu$^BXud9&6gzS)b<(v&jFRVd-@TJV?%I
zHjMj+hafkj&Y@tta?ma{PmBuVQJ(SFqE(s13$LOF{JM#1+wY@d(iO|7kxSBCng^ab
z)95Du1-RJb{`Q~*c!i{AhN|^iMs6_6VH%eRP1j%47I5V>7&9hH|Nc-bozeLUi&CeZ
zvIv4W*a_XA99-_Om(eurphT^~d27a|zPS%PCwT}nM?s}l!H{-LbWYr}Y8emFFmXPJ
zl6L}Uyn%eWmXo^YfrGe~^=|5IpF-#EcBerg3X1+98e>wXo4pL2$7eV+PfbMZ|9<PN
z;2WiO5C^hy@=%4!-=l8X`U`H`^YtHNFRf8DEK#76s&!q>lkkDL$Bsm=Hzk9K)J&wO
z+KG^h%m}5`b^O%Jo=a$wYs!}mDKKH1kJ@s@jwzgwKhRAmt4^KVnQGN8kdHehF_1Gy
zxt*pA-WZ->1i>!FmBu-!Q&<s8B^hulHL2*5HWAs|H8Ofgt!ax!<m+`)dV1sVd&y}O
zo&Zx!d$IKw;A14R8j(OXg-OkK42Y$X5}{TmXoowvY7tCjoNj4k8%$BOkXZvyJIa|Q
z<ML0Q3)XI|V(oCmz~v#>7RM{-jZX@n<bJ_l90h*_UTOJXS$G`uLJuX=`dI^-i%9m(
z=@9hSl;;qq@3lwLb}Xv~al;*_1Qy_Xe=zSDLqvB`q3ki0i-{bci_;D{L=Nz*(i+;j
zCOUJ-N_H<V12zRvQ0QWuW_t_JCv=U}h%|56Mt|7HvJ+&7Kos+TxH&;JG86!d@^N0z
zEE0)d%2~PZ2~}rgCj$MX+8mrS5XuIysG?}J?((w{?EXMX1VOr_fd37`0#nYOt<el3
zV#TtPhr@^u->?>xy$?Yk_L*u-G^`sc>(3a}C+99hND0zRHu?Bb=O9PSMj`J(+sLS_
z#)e4HYrs~sp}AQMP`EcX^bp93Tmq$+trsI{Lzwd{@X~19oYyn?{`A>kGN0*1!}6Ie
zuNnLU)w+0gRp|HAu~_s~(t%`>oL*Gbp^6rw&pz(YDpI{Rr$UYpVI43n^$P`f#>l7$
z*RJp;XGmvg2f8W6HfElMsUng|$UAc1;+?JM(*6DO+k`@pO~%>!9vBBW+2Q{R>sG-#
zjoXRtai9`OHp#ADk!Upr>3H1i_nLy=4wY_x-RdiRsj{yj51|~FN-3q%$ns+)nN+=Z
zCr*-jQ}4{8z7#1XocA7Tr1h~t%*Fta(m@^E;LR+ND&+~3ZV-d$+O*HOA?<2x4%g5-
zs%7pRA1p5kVp2^1JxH;j@5x!e`}*t0Gq>LkwA~E?Jz;nfI+jIHwM4jf1s?yJEH??H
zD=!~h&$XN<U;4Kc1M+V<5PkoqI$|kbiTU;(Kp}I~V!FmKN$Vw;GQVqumYOPq&mUZE
z9x=~nNWq{{o*OImi@eaqfN=xc)Q{KiEUF!M7-PeJ?P18Wrs%>B76-^M+|hlPe9g41
z!lPz-j1Y(A+M4l*niGi1D4hn6UN{k1(#U7kp(M;PiX=y*U@<<`v{cd@yK}hgsUNUB
z*f#=&RG-o4!y&O|dqI@{ex&5YlI!@f_HruZBQKa;WI%!u)g&bRB%*%q_R4hOC}Qe!
zJ6{=FbXET=k&Y^v#<V<A!jg<+f}P2rW9$YZKL#Eq-#Pt=PI|H(TI*<O`#A6Ti;YA%
z^=XMuPDjs$X;GV?o|Uk^xgLBMJ(&P2BbdqJ*EDV-KsAdsN@6iPE*(GtIJFfy$I#bd
zC5X?Q=X73C_LSQpjD46|D~^obLbSPzw6MRNzO%9Y`W)gV@^*=K^8^IxC~KuDdv^x5
zH_SE{lp_s!rC{HFfqlCp9V8Kxjz>Z=^k_=EonEAjxybZLoHwX~2wG|vAlTllQl0A~
zQGi@!GQ+sCt^}u}f&Z}SlL+MMyx~v$MpAR2q%nIJLY>NH$m*qFj%AL~>e{pQuQ*2m
z7H0v2L6DB%5R13wr}kgDX#ugyly*)db^Lj*FdF`t(qEn;GXvPS%a-C<=2ypMR+lnk
zipv(N6qp2J;)&#<eamwPt-O50!4`&_XCJrBt)<YONl6_!PcC5$s&$Cxgnu7vjoKzN
zl?28YdagN%)`@Rs{9i5-s<3?aqK$U9Qwd(L7T$4lDXMB~S@n$$)>`Zf(Ba8?^<G@~
zk7?0-BDxiCc$1GZsJ-Ls<`b-AS;LkXDiErhY49@X*NI*0Ljow~1YS;&yt;B)?7Jo=
zvbkkuQmj*<JNnd2eC?1{nj95nlPqI39errw>rLm{cf6rWYgp%-^cxnZ(0j=x^sD&p
zXx3HIo$=d#(H+=*QEbjLwqvrkV4lWu2s$)MlRQsM$|d87n_T`CUV^iZ_N9Dp?(-wa
z$c;hNT8GGu{9nl!9|Y2&HOb|gJ&7l{E3c-a70ttEJNnXoHbRT`#3zfG5<y6~gzakm
z^gtsTa8x+7IFj#Zmp;Hl083uoKHIb^B*w5R&qKHS)$y<8fYeNi{Sb=h8A{K{(2{ef
z!xQ1sodr!9`#llR#EXSWWcuO`pg>4X9NP#xkH2q@hxpvk2ltN_fYc{SZj1)mDC_aB
z2zc)yOhR^@#?Id`emNpR)S{c;lKaKDY=f=$_1o(DAp@z=nViur%vh6@hk|-rTl=~-
z;y#mJ!vfDuYvcpdCQcZBF_81(thBgk{kobKBHa!P8J$6Fe5ez-F?BEjj-<fDU&_25
zkHFmC0Of_h$iuTQ-YGYZ1jZMRT*g@eRs!>UotG4u#S0hp-EH8ZdpJ+#5-zNnYt)0U
zk*c>ZzSj)oA(q1?x&<oSh7&i0w3N|<P^4rV95P9uqV(j0so}{EOY+<Akki}4|7ciy
zprkchp21?Az+uR0;L%yBlM=<5?o?xsdyiGy^&XJ!C$79}F_M&G%y84L;r)c$uQpNU
zVX7_UoYuR<p3&a>MQNGYL=m46k4yLrO6RwB=e@b%+xEd>e%GH~kH2##llAi(#x7?%
zAiphpr;EmqJPVc)5P!lwBHhVTzL9r&%QdIl&+b1<!BX-+KZttXQQm<T`CT%8`-GG4
zk4jf1wLY2F^~NlRjm{KPI^a*Au+wD3Mbv*}w>-nrM*a1|e%Yb)-|3ipbeazY6qgTW
zAe3(doa?oXbIB)ET?)(PXX0$*KoqDr+$Eb6E#0F&nyQ(<LEklIt1nYVreW-njP*E6
zwcWBv;@E<XX}KKco4~JYxHBz?{<lFVd}?HmBT{0<mXqV4KgXBV7t9c6xS3U$y2`iT
z#WDG4qQ6|qRAWy&H#<_IYm$c>B9h7QbpU310z*BFZd7v(jgkO4SG(O#ZCea>X~<YR
zi|?IjDNfz|WSmMYJK*1WV@(Fe)PC`cw@<(_&+*V84->P4olh#tXN+x+#W0Wqx!+m4
zlfn0cj_gMe|9QDG&7?5wSolNRjYGO&?eNeL+(NCsxKYnv7q%^99|)l;1OOE(EEg)e
z8Czngp7I8q@f_{lN(l)|ph@|sZ7(B$Swy78WU%?1I-Ub@bJ<yB%d~xBq<w}io=@2e
zVxRCt-Zy13>=<ce7Kncc8jjC5whl)Ka{swj;u6+mRM*@!I{%9VC9GpE#o8XlC<}Lr
zNErT`?~YHC4ed17m*NRs1sGfGpo7}HJvC<igT80Da#s0#f#N(u)ydM?XipoJZ9UTX
z{v}~R62*cUohb`G8}2RbGptTx06<%>#flf`R;d>qt@vrmalfJ>3P-5gqp;?LhQzAF
zSto7qbcR7pwyo*wbDehe0rBMBYk5>V10q9A#K6+0FO-X8JHFs0b%&oMncV+4f*SDI
z-3hA7V9Nsrw?ayp;1z;1Q+IXZbF&wu-MI=Y#Nd)RB&@zszI?8_IaQMFzP>kD_hKuh
zQr_qu4kdok$@AW$_P7Q^1FuZ;i-lIVU!8^exP@9q-I~W~PLh$mTBpS{9KHSf2J_!#
z-nT8@@xOO(RtvrIHM$dBou9eO)Ee&Z<-d9Q?GEBxX-{`fwz2bnU3gJfDpijyL8j)_
zBIblL4vrwcByB?c(S-fG3c|3y@$mm^#%IL$n^=kiK9&&l@$@uVlpJ%A?;q#~H0EON
znXrG|d>qqyPZ5_oUvMN^v&wG)zrC&qYCcnZoO2AHx3-^fa9=c8<nWe$;`<iqn&t1l
zN`Bttp97$7d1yUL;NN?hgIyqgIl6dUJ<1n6x8f#@QxvfHJIa0dx>(Qm)-Id7x-M=p
z+?{Uk{V^G}7T%%Kwp6Yd-0u}_gGjl6?d;m|kCQ5?t+Ks<m9>mR!KJC~7iWXPPveIk
zMLY-0QCAqp!oL_GPSHE`EpwZ9f=Tjkv-MY(tUYKHQmKj-YUpRIX>rsA7-En2JfW05
zgCK*HQeETf!7Tg}`BUHzTJ%D?6wPB&D5e459}0y1(-qdi_Q^D;xG+F~H32jd0|EgB
zkjCdFe{e_QPuStG#j@ZAma75l0w*!F&mrFFP~EVwO->A4Cz$Ima-K5d53@I-!C#Y4
z6XPBhXo>?d>Nv_wzotV(5*xcfk~6cbZSLV5NZKVbv1G+z#g3RoL5K(Lt+#(+L7)~L
zA!lsP>_FAsel!^smb}UG?5N&%v9K;6pQTqnsaDPEuMGmn1Ms)*s|mf+i68o!NhZ#@
zcOroX^1p}Tmntm)cuxTpoGl>Wxlpy~uPoNj-;5ctZo_%GYmK(ApT04j*(Q4`@dC{n
z9x}YxhUO8g>LG=;!D-$L5kQi9;)D0D=h)CbpJ$>}O-ilfkg**sPfF6%p{{I`tZ?8I
zHmx=Oj56c#p(!<q3>$VdY_E5dth9@4mRYe(*gS2on85-=#j#-g!IxfZ!H@Gz@2>xh
zPq@9VJcs<R6^b9U@$ZMjs&L3vy>upiOwOpsiI=MHrL0L^G`RT=^lsr+!wBZ8eCd6j
zW3M{1C0_`ig+PSG_u->}fcTQ_E0MvE$Lp2X^OXlrHk-O~GWam|?|;p6Z&sa|8@wMF
z&sVz#5t(a~_j#XE50>7Jzs)y5&7`10dCz2Z9R=B}94syJ#joDpGU=OMfr+%{r8vNs
zWj!C>uZK!LQR3DifS70Bl=;7q1`I9~)C(HwzYWCC8Hxw19p3qRE$6BWNU}ryY8qJc
z%NT=`EyB`XUBH>8uQt@Dr#X9CetTL^;XThNJ(Vk6eZweUUS2FqmRpN|Yu;5e%y|<t
zO68b>XdJ=KG8om5BxBlBx}-kB&tN>Jq?99uZxIjh1&k-{yoHu`b!ei1Vs{t}R1uK`
z0F-Ja7Z?4Iw(DvSxXNqv%?7d4l8jJ@rh|S*vG@}MBW1fD4z0!l);w;EodY<j6y7Qi
z#KkD(S<@u65lQEhO?w;@A8AB|jmv2#Ld71Wu1$g6+{AwAkrN>xM6X-4Kacvja&=Jr
zYKidVH&sL&A%pCxNDh^3-Mdw2ra3W^MOt*n+0ZFW8AUN?$^4ILpcbG?!6Rm%G~)C-
zaTnQagX{;^Xq~aCX>9BSl2<UWYbzpQFJn_CSL_F`1<H^-Rk-k}i(vN$ZmNQ#A}M(P
z?@1V(4A!4g_+4JUxvwo;c(zsSehCScIjg)ut`ONcbs=nMh;iYyhdPC=eXZlswoZ*v
zmwY14Z=3sJ?-*i1V>gG0$_ny3=Lj4$HWf2E8(u@+E|GYFNt$O`QJ!O}<t$+m85mvj
zdNq3UmBG@z5r}H?9|xSNt-ysIg6NE!sR^XJnp8ZMqCH~nKFLAs2h|9<g?K+2{?`?H
z?QyVY;3PTfXyt(gL^eyyRenT?2zm|^fHlQLP{=c*5OuuV{a*pC4pQ+kwk+8k4)8*8
z=E5ahrg23U6>2_^ra96l5EhKB;SXQi<9FWJ=jJbbhEG4|`10@nZyvh0kz&B^Ud{gA
zggdwI@b(+8LZY~O<3(K45c+Ez%nf0!bwu};JCA3~j9~isG1Y8}!EyD)=eRz(&877|
z-*`0Pr3+^{tSc~<kX!JM!^xbgX$^+>(9@0gMF@eitiC1r;DbO2L6PS;=U5w!m`rC(
zCKJw`J@@`Uyuu2naOu)T(smq_wtTjZ>(*EkyKx~zkwNE*?xSinw1tv)7Kvy>(c4a)
zIO4H&5CKFd7=WlI?M$19A_8k|>`-Zi)(Wo_!rQ38bjebUxh(2B<6uEs%yrf&vVFbS
zQFE>Tn{UTzOBHPh(W1pi5r}tg8DfvMgt|2yUUuqyJNk{qU2vTxN!wWyF_mHYn6~TY
z<7hRGQrnaa0xtOWYim*}RNOYv?xVId2r;<jqbTZhqER7od5R*ZpC;t}f^0Tte>P=$
zaL8nDpYs<lk>&-e=pnOwxlmabxnas!cFt^|wWYLyTxgUD)M`TLWfc8_-k?X4XC!&X
z+Hk<Msz?HmX6&557<q<O#q@B@bUb6;OxS&Rz;NqbW_5)sdRUzh(iE4*PEfu+B3~O|
zB_zF^%*CiWYb?m5^||vXB%>l>YkiF?moIVo;zcqos24N#4<7U6(S4@l159nv3i4Ev
zCJ9=`A}q<$_&8OhSwfm8=tSW{gKg@l&JTesOUbhgMVu~J>#3`XG|9=bd?|iVO2rE!
zYnKTURaGsY>%8ogQcHn_ZK)0)7OVCAyNB=Pdj3>J-Nv<>JogKKo-gnH2mF&Uf&}j^
z{eI101Za&;47HV53x&=oA3kDZa}(#G*clOuj3kX$x?y2Sbb{3$Z>PNdV8Va>>LIU9
z6)*kx&r;+)e(Qhz7GHh)b(~eKtA?FnkFS3Dt6aNzo!|cKXSsCs3cvg}9EXSd>_2|Q
z$bs+@?+wNp?mU^12dHnq#lQTuf6l%AhYZt<t4^_Z`#wMV{Bw*l!AvM#d*gMUedQ%)
ziwQ-VQwr!IsE9$~AC*nWM-72sHk&b-PI&2+m$-BL9>!P>562(+weP%rFA~^^inb1?
zN2sUArpKR`^5l5!boAG8E0WV=Tp>^H^W!7d)30-SBz%0EPLF02ebT)1`17X|U`}sC
z<Yu+^PuWFv-JoOtu*Bhbdh&HneXCaK@_yb+Pds;g+zR0Q(HOd%AANARkHEC-*&8=n
zn;8+k;)I~C8x~cCtOaFRE)9RZZ4(|^Rb*EzS!2;sf=-|-tJ*Os9349mK=`%@vVQ5^
z6xtbSEU;RBTf4|9B?&ebbIznk!ii-;Czjkcfk4RU0&CkbXf!wWM+;<2Ru3GRBX{V|
zi;gSkjH7d{iIydudGL`UH~98zJ9>#03`a$s3s8p0SZqSTL=zTVTh_U5JRN^a^z#$T
zoz)g}oq{)fBp*vE%c!db_!^w1K+;Q7^iYusMG_`V=X1hrfwRyAg^~%ONDxZme2gti
zg6G`k5Yx=@jYFk5A}GerQzt1&o{$z9SuZCkQu5(C)xiNm>R8lWxX5feqdJ_Tn32ki
zkfrRGCC%O#D<#8?O$>s{dum{qWem19Dc0AhgM+jnktvDDIV=~XLgE^St1UyFvc0y!
z`7>uYx3xn*(d<3iWisC9$=)u9lY>|!<_W`*W;n>`<pXlvLrEPw#8My>AUs0&*rBTz
z)J;injznv+Owk`?F(z*2XzH4>nW0q7x&@XOBQYr=9`(qvtH}$2R1%$Nq>2|T&UY3v
zW|=bbVX#`yKc4IP)A`g%LGS8C&ius7jDPo!C>9nSG-;C48(6B!k|!|7vsient&xdj
z-+0cQ86uU8Q9oF0ZIMA>>M8Yo!|#0KE?>B7IP=oa@ySnJ<n?cSncw`+|CYy#5>WKT
zh)izbZ2@(U|MlCSWqW&zU;4~1gBMKp_qg}!*Z8?XK@gD?Tb4D=e9kX_>LT?Td~Gpd
z{OBQb6uVFM`Kf{8+Gfr?6}))!IT}1wSu@$&XLDnN-P`Zdlah(^d_QM13<f!As(Ag4
zZ}7@XFVgQ9JlK82tFL|Sd%bBZ`~YHa?+IC&fsk=QkpkypoR_gC3Ki0Xe!qwJq0I_y
zi;T8|=bS?U38IrGAeOhz$|LgGaIKNBh_l2*%9iWa7B^lKf=5Y(($ObP>y!Dbq8sTF
zTC0%w(n5%3(dc|<=-UOchPLPqNB@PkFl;YMVnNt;#yo8dSXWeaKOWjia#x79=@h=>
z?zQI1;h1WbD1?a0*3jmcF1M;J=}sviT2Gm7JiYwBk6>-evy8*>KGqnt(&TB1Qi@Cp
zWRa4ob<Ajl4bdMj)C(5DQkx2uXVKz9fk-u8D}3I=t2k0kgeILYa56<`iAp4Ck$@yB
zB?ZxFSSA#`5jyXa<OS9{X5}267YzCXbWu>+h9W5l!SQ5opPkWwwXF?`qKC|L%C^vR
zNkYG0uwE3=0;XIboM&sW#-+38*cz>YaqK^OmxITTc=G5W`{Ow}RqX6!oH@76+Qx`H
z&(TsNy+R`KB6=8F*I;d=9(ZpfF`n+iMYUYG0@rdgjdd*Q2A!3pS%OlUnEBbBwe1B!
z)zs9+llKHlYov;ba4$lenQW+SLt{VuinLnKzd!L|*Yl?;>L%02jA#37{^*O`Y%Gtz
z`2P?A9v(=NQ6QI&L?xs>Lp>K{S~G2OzV!87UZ@=VrDb|JXQ!9&saLL}t>x9*_xO)*
zjk#5C@{51#@AB9G+F#`#|D#{w^Z)PvP<apJ48<nf{TmF^ZMN6Wkp;o|tsZ~-@BL+N
zymXygufM_GTepzA_qh1-IXn&@EE>V5Up~WVE#vjweWv#h81^%C-eZ3;LnIn6A?p`N
zDe0#fhx-%M?k*bx&0I<{2^&(dZ~fA#{QIfC&-09HSI!}Y;O^b~JluW2U^rl7eZ=qo
z{ujR2o43LbAYOg-HPS@4(OL;gYmzi0OLJ7BBHu+y);HFd=@3G;C7Te-K>A=Ewu<8@
zxePXJyKkmd%NmO_ruBx4<4e(Yx@2o<60<TnHo86j4h+W%p{RcKtu2WUC!e$3h*z}M
zDJt@-T+0PqUbqX8TD7Sc?V{_VErxd<>s&h?b!{iH>zZ{rw5<yFxPYbo>O{+}>$-My
z#a#GSmd}x__w*BS+`Hr<et0Mk&zw8Yy}S3AG$sb$k6LALqVP>a)0B(`BZiyn=%Qd@
z8Yc5O^MfglfK&>vQznB=%xH(;97YP}RSms@EGwARGxiRjuzv9z)9F4|2)4JjnJ+B!
zg#&LH3^zIGJ%S>EO40ZR?<Aq}^z$B7T_Z$5IY%>JP!D??nu<wPqk4UePT8w!bg$3O
z_9kUH$7{{7SFpV{VvwaY)tJL4PZ&Shqnb{!rsnK+&V>sX*xEV6aD5FSC5vjo!`=Hl
ze)xc`tu3}T)?>_*4S3^mHY&c;G$+#qO<lG5tbGRk4fgjBKx9bm5zT@S9JQ&S5!AK8
zSc}w}BFhm<^7ie!)CSfzHrd?XMr8@6?cF-zm^UTPdvub%-ywZptk(07=X(Buin@-_
zcmLiUd{eVF+T`4ieUfr-pYruv9MqcW1J5&O8wRo=6q?B_@OW182e;;Y>5=10k8ktR
z)`%A`pW({c9`C$!i?83?=X3WBcioWZUj76>_ve0&U;kIX#^?U%f3xr&MaG)kVyn2q
zD7(bw#ts{6eFBc({oOx6^u~PpFZ?8nhmZNvfBP@|%+3%~FKBc^-IP>9kPiDio;SSX
za@6LKL^vue3+MQqCljvUA9Hb5k<UxgT4Ty7&nOF8b8%~vW_m!+I!Xa1$X1Q^{c{>z
zzjmD$p1;W-{NFFo?<ppe$K*wd^Ojq8?g1y&r5_aC&$wf=$nb5Rf4`rP_fM@#ryd^X
zhnya%ic`Pmy~eNN)R_19HuzI&Qo%_z?1}N~snzxLHXI*3pPI~`Ao<6CIC`<*PpN=U
z@4{pI$=FjnDV!X4A0NM3d;EU<yFSb@ubLVcq78!$9_?e~RuB?Rg3vaOJA+305XX=q
z79edyE9QW<!Smf1wQ~UuJ{DNZacziQ2;Y8Bm%<P_frW@k7TsUe^{I}#u>1CF$BX0G
z;{eApWaGvKA+W9Qm<@i(UyGf-*fA7cM~<jsi)OpgiGKNdAz(W*<IrWve*0Q0zD~UF
z+cB$e#SzEL&*klS8&0&3#nCpg<Q|?pet(u@KAz2)+8Qqer7_&w-K993u(h>Czu#wW
zEEd76sS#yKp%gkv7_P0M7Yh!j6ZW3$F)eGh&VkSgGS3*Tt+TzMY0{kgdrx@vjjvOB
zM}K{Te7J$qIaTFoyaaM|q8Y8NBecd!P+5ZTU@OC7Hl?bfl}BErRPzN*(_oF|;X4ob
z)33k5TX%Lj=xJWMe1)~0Gu*#_kB#*KXU=ak$a8#C^JIUIz1;`wKYT({I(D{)oZs2v
z(xuC6ZtbwB8g};%+27yi$=(wJo-;e!B>g@jD+o&AJXE#8`<Q$?=&d28ptfVmx@5dC
zsIp;hJTnIi3qH-58OQ8!20@^dCNKIVI%Qtf>_2+Ue(6b4usUV#1*s#B^KDF57%vty
zm7%o$#PR#H9P`zB{xMz8!w0~<d-uN0C(i2jgL1of@7@PF%&TMmF}kge`48W1b<BVG
zZmVPd!*_eXV=h8?|5D$3&;9%N84icv(j8gh`xhYurqd}GE?oE)VZ-}jb<BS-u{!2I
zd{`axA3m&(`41mH=rR95McoQ3tgylgE3B}>3g6#&x)Ohd6;@bbg%ws<VTJEYtcto7
zR#;($6;@bbg^vMNMcoQ3tgylgE3B}>#{jFMZiN+ASYd?~R#@R<fK^eq!U`*_u)+!}
ztne|ws;FCGg%ws<VTBb|_!wYS)UB|>3M;Iz!U`*V4DkN}@HETKm`J;n00000NkvXX
Hu0mjfgB;aK

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_004.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_004.png
new file mode 100644
index 0000000000000000000000000000000000000000..d7cb1d337245076b3c0b6b4db85df4c56f54ac81
GIT binary patch
literal 80624
zcmY&<WmFtNw=M1#+@0X=Hh5rg4=%woxCD0#0fI}gA!tHy3lJngaF{{D;4%;x++82v
zeRA*nQQftwOL}!zowMugT?x9{s`xn6I7moH_&_yfeIz6lzn8ol7RJkU=rAYo<tFK?
zV&ZGy;pppc>+OJ~W$Wwd>f!6^@{aL?gSU^1hr2MJIG->NqqDECr;ijrzuW(-z~|xZ
z#1D~I+Ii^&+f&Wd2MLML_CE>vn_Q&}5;77JP+7q+py1%qwB7;up(r8rsJl{lS#q1H
za=0Om$B-=U9T{@HE)rJy%6kGw`wT6yrRoL8#}e_b{qGE{Uq&?FWN3z@RhMVaJ5O;;
zV~Md-njn4RKoTw1%`%eI5ma4<FAE=yV9K`L0}!iiX;f*iuuHz0xQHJo3m+YJ|Ge*c
zx~yx7@qwKFDViq<ey4MgAltxx_2-Gg?M&_pS>`|a|9+tQlfC%pe^>ZFh0q9DWWWF0
z=)cOsF%Gc5{dY$%W%t51I`S+3qw}J9Fmf=;-hcP|p8@gaNp!FO$5`xFPQQyHI{v@$
z<xsO_h7+q>mOBG8T3SS7h`?8ZkI&_jh+OSTi#AVWP6h&y+!wIUTG2QZ92P938s2d@
z^6;?R%6@)cxkMijwv0O80-B>e`G?S1M=Z902!+^mwh<!_GPi+3s?UiJfkJZO9pV3J
zeFRR)2V=h5gTY`(H*3%S&(crv&rjg9L1~YZJtx4>T+b=GRlu#xarZs)+oF3;*qpck
zhHC&g69p3sO`S7_NX_ZD2<6H~ffAOkdKyJ2@t3<z@+elI+6*ZYf|dZ&q}g3?kvT>g
zD^f7;y<jPiMr*6sZk=24G)lRC8LA!co1Z58^{|cV0ZCMeGBj2!CL)w1*!piKT_Kh+
zoS?wCRAP__#Ds?&gqQ5@Qk8f7{8V?on*U5{DMc<kc8EGTg8p<Be%s?{2f~a*1qs4z
zY6d8yVZ25&y4kpkZw&V1Zae=y--O?)u9(Or4O~B@JU_}Q>|+RAV4$lMtryq<_!rjX
z{!PrSZKpi#sCJ%1?llK}l0==~mV8H!Wo1c|oi1a{LXRRX=%bRma!I#Is^IX36j^UB
zQ20~Uk~N$8Ttko<!aSq9R!}q51Mha70^qUSGLPdbAGF;j8SEq8a7L=@6GhGL^bipf
zwbg<OPjWm@?qY}-jf;-ytv^TO&HXtOKE9-Uv`AzLAA(whhRl{_W;%?7KdwDQtLA~j
zPkx|(A9YUtTCK4Nn|4+(>UZ7d07WF#6yjP0-QhrG&QU6%7R7T6C5?DU>?uX4G52=m
zOV$tmA>)$a_&A?sThD?Dq@{8pwuNB-dI_fn$h#WZv#=UbdH--zyN<bLa(Q~mqS@nQ
zgeSB?sclV10^;fV<N0&Xeg*vbBIU^=MMU*Ef0ZgoJ)IS$98LS~TnwTIN){tvBB?fT
z-ke0*5B*TI2)}iL2p>>cNWS}FT`#;`TqxdoePsj)IHo>tWaTB5c7jRzz)7Q900}(b
z_OdJ7)AUnWKefWfoqPxgKTNgQ+L_dMH6%0_=Vu?*;A4aitgiA_Rh10m7PK_GX&8v-
zwz2_<&txihE{B_vIyr_DQNB2)ma)2g(zkVYZF_oU0(u3pP5SC)UM-4+#N@?X-49_Z
zS5`}qk7yY_Sq;E%uBmW)XxLFvG039jdH=M!bpG;XF<XZ1Zyc1E%aOugbFe*LzSnFU
zeB6&E;-2dY*NnUDH6b}-7~4B!j*(YS@QDcp^2p%|3^LYV;3lY}yA<hN;9~Rx9$V;D
zrdTwDH1|`ri*?(`!?wrdhSnbPpCv1yMT0TH>TKA+qBJ&6E<F>(<eSnOxN2S3VEC#3
zGkxh4Tq)(S^z-Fja8QCgm)}1tx$f1$OQt7tkk$`6#g$*Rru+2}XhG(6?ZlHb8dl)H
zhE32!e=gC}eL=KUU2*MuB)$Ua{;MnUpB`(rB2vm~u`L(~T6Oj2h80=12pOoUwY4f%
z?=3FHZ9DkUl+QXO?w9y=!(fSDvR_;!pXEl~^F&XiJ2?k@uG<ht^X7n6uNj#oWS1gr
z*A9V*l((cE4nIqIPF)aDNfng*YY`fr0u6f#4C<oqa~6vSIkQi~HtNIle5W@>c0G7@
z@r=2vvjc3hQW3IeGS*=XQLGa{62h};Q3*U4Rno+F3S5FK4?W1x!D<&|kkQVkG{yOf
zYORqriCav^a$~m=ps{FO>JBRSKsmP&qZHSk-G{2fgjQ4Q%to))fyT*Znyp6~+A}X1
z4p79ZZV>5kkfIi(>CvzT<|*3{YgzXm?$ZDAuk+dc;`IA{_31^k54jzW8OVD@my}9t
zM%RcJQiR7r?-*9}>x?uZ-2~B_AsE+K{SBmH=|Vv8I*Fh{q(v!#UM*3S8n~2vKZ?hw
z9OX&O{)o<}f}+o?rpr}6cihm`_1Oo9Q>aQstDWNul;ZCtN=^RJ2K`0iQC=!=!^7EF
zA=lbBN4)d^;zss|5&<9%p<~F%7~H#I_L(-ReF}Y$B?O(ys#BKzW$l|H!B)?$55jU?
zi;S+C(j2_ke3L(}19n&{U<R_P6n#x6?VmzFEgJxl9r!{`QHLT|NB1o$=ZuOSY6fgi
zena@{A*R*%^<$Drr%);qF<(&MKJ9n(pD25u8D_YOtz_wG3fk^bCP^UYWv5}3iN_<E
zrIl#kGdY-gfk{o2oN4-c(urimoTPnbF%3I`v8#o};&JvPNLg1q<(!YqEE7I|44eec
zvBf=%dC19TeqrokoMn!bI0)6`YI1p|aw8Agnn{GJH;68;Pk-U%D*$N79pCy@yHO>(
z!>iD%T!198NW`=%Rl{+zpTfX!Pk9dk*8MGdB35jy*_t7V&k>$Tb|*g7=)6&_m(OoZ
zGWAW4FPUSN)De-NJvlguILg#I0!0^X7RhG<Fxzq`;D@&?6VT{f(o~Mfevev3zPZ3i
zd9}zEsp>3pC%S~Q{6<m~CfYu1Im6dTCcjN)Y6KzJzx@~qFu$EVPG?X)bzBsi()93C
zSL!A7s0!FDksB(et1TLD7W88G<{x)M?%%F$f8o|Km=%p)MY6+lNl!<ha{=YC@Um}1
zAUWMPDpn!z!0!8%<J;h&KK5vlln>8poVxl!INOu~iQ#pt&ntxNb^Bp_*%3$o>_}AU
zrfmpu_p;ZHQ9hDpIt&!6HIQ$gO9mEso%Tsqw~TM&e(D@@?Utv}$m1Y9D3hnrkzq?<
zH@>?5&>r?kUFi>bnQ8qy>yvgk#II+lJf8mfh3!>Zv1zdmHgmj6S;^2DB>(6c8M^!+
zL-Z~2L%^g^VWtH2(elD(a5!V&QEcVSb<ZvJ+D#B7%L&Z(irkG3#`lGiT^?(nN*dW1
z$93b2bh-Yt2>$A4^HXo0K@qREuAkNBfB*!0$W3p`)MaiMQ+vP@ErQRrskI@t>C=ew
zpThjPfc?Q=oww0zi0P&}&)qLj{R`MDE=@B`D`PQXG?%;&#Ye;I*t)MW)v@(cxm1>D
z_4$v|he-=1t1qfC22ZZcJx&I|kAaz^M7f@V8Zjpn9q9T@L|pwHvDTgpQ)1%r{{q#Q
z#7sb@Jmk$aj#6K|5m$WT9;>!uys!9_)FY?da#^MXztV@qc}d!uezbajv3}0Xu1EF8
zyrS~w{Mwq69wb+jAXk$xmtTA8@4iG%DX~iWH&5A*!MUOcV+9q;fX}PrXMkeRtTu`x
z_yQxmU>hrfjevQ*8XZBE$Ajv|qh;<T0kF9~qKoVIYwSE3S5HT~5(-&Wjrek3=wV@~
zV`lMIu3s`fm3f)_bk*1PvxQ0hW7H<t{4v`TADi$m%doj@tr24&^z>vzyi8RzpE5?5
zgHh1y+_)Jd6<f`@Pz5v-X<knE!9u7@*fEtW2-O)En8>CVm3#=$k@F#kQy;C;8G!nI
z4fPZ9BtHdm=4KPaS(8&q6Ks4~HU}v9ym$Lvw>OV(-;cAWS2X`kOjnWYo6ob-6QCs+
znZYYy(K7g?0B#=Eb@?=^uSQ1hT%cqY?HwTq*Y_T-k2Kl;z4tqtI>r}kSHR3JSs{V+
zl_}A0oE}99P?`%l{drP*Ch->}#k6fs)1KlH;`KYb{2vfH(v`!Q%vVTlNR!G)R*``j
zC~&kILDnxQq=S@=%3qhrnb2Lb*OPi)|7g9sE+YM#te~!t_nK~#)Kuiv4xJqjjH-<I
zn=%tCC9fxk+-HgDMcuc&2~|^WiA|rMidAQVQX?XkFvV}Se}*SQyYB*FFtHcN2%<wt
zXhK;sE7NCn!6A=(L=J~jghw8S9tTs>eB=%WN8XbjD$wmqI1UvyXTZFf070T34^AHo
z{^-09vm_CzaRm89Pr)h6(e^<~ML~O<t0DVzd97?326?TA^X57cSa}t;V);$2V#=V2
zc!rSSJI{RZ-Exjgyi$cQDntkjy?MHR{y3Jy?+t%2t-g%Wl({yhiCO=6eyMk*tETvx
zO5G7l$V}R<VSdB<G4a^ze6hg5+~fSx`p&dbMukcMtD3~*mAC*VY`x&J$Z%WyyK*_e
zkPBV$8^JPVsUXCIfSBRx{>i^*K>T?{Zfms3sm=2U4@H8OZqk}&+ImfXKTchTs+Nq0
z9QWvGWdn09FvWlEjsV(yG1GZ53N20-txv9;di)zpvk?EYEY+6jI-@H24d+%rt~O)A
zT5hsvlIqWqpvr~z8#dU_)|*qpu#~CZUux^H!+c7+;d7KjnfI8Ji}CqnHZj&VF@{)Q
z=^eSWi^fT(HSfGU===<X=~LQ+{d4lgJ-q7W2F+ATf_7jr^@%tXCH=3IKw=M}{HUZ9
zLZ$cwI{4Q((8i+gKP4sR?e(;rmp&Q46>$Up*e9sx&ax>mHFYgT{Cc1d_gSSOSK)!*
z>}@i%1ju|_;di~4=(VIUX)d}v`wf$0LFS}WSJ<!-ll})O%*TrZUEHBIE=RxPBw5Cm
z-zh^s9X@4{uH1~+8}{l5a9!QJdbWzQOo_MAF_RB9f%dFL_XaEdxt2CYSbJI{-ARxi
z+%Jw|Fw?HG;Hx@BH@)9jVR`_O63@LK9Vs}C{F4`BD|bqePI?hU<)rPAhBa~pOEO&$
z;KtABP-&)CMXiyi+_G-cI~kv;lni#nby9as#IA4`Ojhd1EHscVFs7k2PjjH!?`LvJ
zXU0myV0#SFh<vfil1a19cBwE0lObAy>iv1~cqOzq${p(=l8N;w1vwoOpkGE`?LO7W
z67%J}<NWCb`D*w{H@%Dl(BhJ4qR9E`cRnl7(4Tgt07s$d-#R1kq1&zow=P4%i3$%R
z(;=bdvy8t4qVSGKtK&TB>$Xwr2&Ip&MF+^3*@*Z^L21T$R3$buszj~fgLZa#nG5Vt
zgS=m7^#*(GVO#uGIvjC4EfNMr;U{YxWX{NJb>E>tKYtI=3E$PIuED2Fc6y%bED1Qe
z0_;z7?U;vZV;RHpHh&-WcHeuUNFjj~ID4!?HIgk&{k^@&{>MjW8;SfcUMz4m{24X<
z&+*s6fY)<U8rc=jb}t*#e=3mTH77Cf=~3|U$;E)v{Nm!*YDbY4%*z)`z7f23Pj9|<
zBNbTm&JEANJjn%UT-%k>6ki&j$pREA({x^rG6uO@<!=;hMiX91QYK4;Z}Utq5^4EZ
z&QF;|`|@Hj_`C%w=DnsDK)(`MO)ApX`rS;qBmhos3p)&bI_$a4>AWBG<0~*7+WFMv
z#8=<Ypin2b?kUno*>5l65bi-(%B}5&&#5us`yt-n>e({yeV+#SHV~s9zu;V4xDUha
zW0W>Oykp{rD86a-SNkY^Dn?J(R5O0l?#CIYqOCFOM-T_7&F5a~xa*ca!I!6kB(Jr6
zjbDCfVw5RMSq&NZ31n7qX+w3nQw)(L344guHDb#c$zA$+?1_L_IySWvrl~~=k~5JJ
z$L`jW5gr(~T6O+n?|_QGVJRP&m)W_db*az=61s^=hB_uLJP25f6K4f@Ff`yDCy-D|
zK=W!>FW8qyAh-2XpJWN-{d=xu-J#GXUxl3Qhj(eL%mLR$g0H62?Xk}XqNbd4rh3Js
zr7KO$%~{5T8WpDp0_oQ+j*thh0}C@7?OZnP=;zree@IvUQ_J;rFi0ft{pC}9sG70>
zEIur?@_D$I!zZzw!P7gfpHi+I+7HG$-yu?CQY%xYoCuMME$h{*@0vxm>tfdXZ6i=T
z98>_<`eruP{9e++a9N77;g_7_!S7baYP--}pQ{UuX&0inu(}~nr+7nqto-`+_E-|$
zWUGj1L(XFk=j2{^(v^RWKBI-|(dfD|8pV>dK43A?6Vl{rnLF)sL!4NT<7v`lfBF_N
zXvR?ZF&vkP$c*35ZIQ9;PYBL<+L|~PK`EjCZI99QbrlQoH3et%pm`MmQH6`T<s2&z
z$1s8g_+Y26ekRjjE75v-ZfNrB(XYdTpZVl`ja`XyqN&r!D5K&NkCv|bv<R<D-kqN*
z*4Z)%NzbG7+COgVTP_{TfYEbim`(H5G#Txn^BMCH4^Lg^8IBXQ&`u-_ffanq?s6KU
z<GrqJ#+_X((Rn1OY)W4V<eXaE*MNfMbc~FW-o#t<coibR;je1&-_}`{5BrS12U<li
zLWM@FUa(H;Yi!eOuYiEzU&r13I?oqkrrA?}$z1aEjM>5;hU7M45p^fy!a92P*ltVm
zA9F^&S%)G8bLg_iFl*tIsd)|BxN1~|2=&7iCT0mee4fUsdJ-@R3GC_zQSmj1(#_g0
zd<EL-;laLpm;*>!qVU04|J?C6I~C}aR0OzXnl3bQlT8ar)hF=pD_y^qy949<r;xkC
zke$gB;~pQeb1*_zLk_Pg8MJS3d+x;bs^ryXEN`Cr^j9XnwGc49QC*Ts`dh}wqpw!&
zWq4Vz^mgsvTFR4ZQQ#iu>&?(I!!lDP5DkME_npWDDGB7Q`j*~LLHQ-idBS|fu=JLW
z*McZ^)Xgq}Z%UILr3!WDw6N~6%QV$UoXrwb`#j86Xi*`|F)Tom`l!}fH8$kvRO*<L
zT8ie5idiz3j)c^xNJzWyE?yv$<UB(|H$a|~LH#F;IhL8v(`!6uF8e?+5S|ts`O)M(
zyY{=?&-;BFcE6R9j~`oTQXFAsNuH5ypKS(6xUu3hvqi`ZwGs`Z7JSDKrNtNKmmS+$
z-*`;2m(CUD2ySDscrW8xJtmW>60?1lq2O_k7y_F&ey1n+vu~t+T*&uD2AAj9jnHI5
zHt~-5rVuYT_G8kmDhTN~xHiRx`HA8?s!Us=?fJ0gD@7h0!aaegR+roqZek;YLaU)Y
z(U^34+FK4jzO;YpQyV5M18q!~uIg+wNoSiJ7Ex~7GM5iNYSdK&^FJ(96qhKkRD-`+
z1l7CMfEo6QY!93J2un11^zQ@8av}mFoQUZE#M+l-<*Jx4*du))-j#pNC$UPt8<O7p
zz8Z)XtvOx)m8QB0P-0u^Q{4d3%Ton|H!CuCDBE)_D2@}bT3snw%1K@Frj6Yy{q)tD
z%FEo$GxeL@5@8*vZu0(MAX6_D!*_`fmSFoIb9PH&OfilE#&&c?f7MtO$;TaiFrve*
zLjzY?fW)<yU86m>bI*Ctdp!^Lq=E2->A^IoSi%JE*rse&ElPrTnd^h5&Rdvtk;DyU
z<=%_{h*zDBX=;75=Xp4E{Jq*E6Xor&kenmC41GMd=8>z|=ElIeMbC#q8gX~e+cO8N
zpxwZ#_eY32@Qg^!5Mc!yQ0Pok&q#D;H|Fq=x7ocd=n%EmqP@>kqU^gySt_ZDf>Bn;
zwe*@(s9T|rllH6Y(R&rcBPT-iy;gC|gRB;)K3Op_rZ!5RG45)>BbW8vw1(PODR{Fm
zuXgZb9BzJO14gs8Od*aq21*R3tRha8cb|D(cuM};welV@`<kA?fu?+!P_M!q>vqLF
z*1t7O+-SkUzjEdM+}-dN7A@Gdga<PO{f{~$)5$Gc3=`4AAEMXb?K#$4qquVTi?+$Q
zNrh0GQ(Q%jl>prwE_8*58p{Mfs;<@7qpdun=oOOof<fT9Hy*;4<MLRaEFwu<-Q8G7
zfObjH8xdmMW>`s$ig|P|vtiWeSqk+XW@Foz_nRqnv@YrOEiD6sgCDzA1e;oWTdW@3
z3wOo7871O2G`J0i#Im0DdC@2~)xad?HVPEjQXM$;35eds^Su2Rs8~cn+9gu`*lR#O
z3aW`QcP~1{k?>{#<Ml0L7cK9pgiqTM_8AW#o>}|q3nk}N?qkm3?u5>0>WYUViQa<&
zGVi45n%Y~m!aa`zSNRWu?3q%dQqn4xo!};x?{%I+j#H;1&&o5r8wWj%XuM+%6NV;T
zWR5++#+1fPx@k;AOe(A)Ix#h;_p;|~X^JIfe;A#s@)pV*1i_^(Z~jr~$Na0R`07Si
z6KxDis!i=s-ji(h$s{RLFVP$c2~v|}5T^ZS75|=BC5@@vR;VcGRv9Y+7xO(EUar#@
ziXV0LDuZ`CR!H9s#YI(zCYdt-0O6$+ToSbE%u8rrs0$kEiKF}=w$*G#qw)Ub7hlCG
z6#ow1;IdR}q>P~f+R|9*#|OYzd+#`SgeJ%S{6s0-)s2<UKuGGVS5@uFTc)$(VW6h7
zes$@rm4^BdhKZ`y&hDHjUme{%Qi?mQ1j*7fH}U~|I%}!<hxDSX>QwVy18ADv)QB0V
z?z%jl>DO-V=Gwit0+H%4o3UH{{yn0s7#D3d#pPIs#fD#cJgwN#AmUpbM7CE&cUYb7
zS$BuJv{#45G>zrXGiq^HoZ{I$1a=xsGD*L@6XPFgOu|J#o0@z+(TIDT-R;+%<My23
zhCH2v5$0=m<mTq)B3fg;y;z=q{gkrHvEbZRly>zMw&CI6U;fASU0qt~tdwr4`cwP(
z0f#Mfx623s6aphB;9vSg@DRPI6xqT3wv}q7uvq!C;U{SLDV2L@8zXmdEuKe@U*bYY
zujjgfdtr+ejZ-rN&leuycLEZeDLm-};XZ#aGr<q1Kilu%pv)%DTl?g~964xC&k5K0
zLhi4h$3hk~C1oXZV^jo~9Bd;^1>So4#$-lIclo;NR|JF0-9Ep|EBOS$IyZzp_xUo<
zeH6eMJQm-R4%bP^NJHjoB?1ccfxZoRk~|v~HaZMR$f>@2pjzgSB>r?NqP{gO?s9@|
zbRedYYhiJ3lelBxk5s%acArUVz~uEuatfky(gbf(m6YUUKR!zU)YVu>DtGd4yog`4
zpHw~f7>zt5N@-P>znghkstLTyqd-MXD;=v*MyZOZD&1-<$5E1tD%CN&Bxk*VA0Kth
zCp%J<DBW(Mj;JN+N$}YR4KL)+L=Z8OH1>h)$3nC`ddv)*K94^e^X&<gB^XS!`jI#b
zeu=k9D=SN`uI>$fdm_;-y!u(bu77}tpa<Fr2<VyBhJ%l1s6~-(KEYM*`tF1d0zTgZ
zZ6f~QpWHChT}lkeKF;cD+TVVTccLGo9nkP__*kN}N2e2ykK04o#TDbn!SqUB4`i+A
z4B&R5PUh54YErWp{&WG(F^Nnkm@~e}w95^iwETzNrI@ZhQA0B)>3>D~7^XeN&G89^
zi5kbCm^d~AS+h9|ziIL2q^O1ImuQloxDz07pnCQEY$ue3cy>VK(Tq6_68~9%y4ezW
zYn4e1wl4Y}$H{dFC(mT#b%e&P#1|F?tgk_ZE)S=7u2fZA|5WYKt(0;YW0W`(VB}1l
z*^vE8*67|m9Wl;3Qk7mfoC$)>z%_$ExX4#c!)rD6<F1S4>rSdfNoLQR5L41bQQbMf
zK_33SLes7kx!yjx3!DG(>b8H_b-t^IM^j5fEUDGnM^XEdllyF9RxWsXj#b&51}DYg
zCe40)y%7fe5xVye`L-q8{;hwd3aZvL&w!`<;k#Dv9zazoMJYSddEkiWypy20?mx~Z
z(*(oxw^o4(9h~hx{@~;1sq2?D+XE^51wHS@WQaLG!F1p?^y|0LEpBo24fuWAM4=Y@
zuZ7_`wGtfoN!SVE`MBsuoV8xyAVGUZ({oB=(S5TfKzW4}na<REf($+UD!(#Ic$Lbm
zIzr#~d&?>AVXy9ybgN_w{Ls_WpS&Nu+}#58f#P>uf0SnmvS}u57#xvf=F1{?AoF;>
z38p=;thAyP0@Df6667WwmE=pFFJ!KCy-X)pp&i1>3EHlv{=o`D<n1m%CDhvf+4B#m
z?lEE|T4}XM3We2)AB)^)g0TA@^MO75R?3NAHa^`{0t2l5S@_x_fJh9FDHZ3eGyc`*
zu<&HoUEv+DVj1UFiSj2wQv&O!1+d-yUn=*ts_#{e_epJOx0z0{nVdjowV4CkiVfwa
zICn3v@1;BC)gE|~;ZHnUCALzglOow2tCc+FWjKC$nn)(Y#Wam3h!!7pplfNU^V?hK
z^5UWm;v&_c@`K5mNhya=qb5^~(lBlO8+C;h=f%P4R(vf3yJl$hc}L%7@sY&Gb6dF4
zy#)0gYsyI(;`n;`6e0H+|DJU_Di_#><R62}Ls!^ph!syvz*g+`p5V<r?ZL57q4xaz
zw^Wx8WpSIQKGM5wkNeqd7@wPRhZLsM&dC6{KkjQk1J9{>nN)^QYJ_SbQMf=s)uD9`
zfnS*4BoiC1buvw4<Bfu8eOQg3_CJe1$Qh4aBvmQL&E47=V(pq9S9lp3sVq^}DJ34n
zlB&7%N+r29v@%g-Us`>lFO#6{veM43{Ku(DaHSUGU}uLfOIh4P_|wg!E?(23ZphUy
z(cN70M`&nUtAcnkCb19-4xvPK3aS4l8wyC-=B@pPk`r^akt%6zO&@1;;|J5R39}f3
zT;t%wloL%bT3M5c+vH#b3=n)Txa2kT`U%pQBm<RB3tnoao216Wpv#!szYO`FjRaeS
z33D|tD_K0}ApN1Y&D*3W@9{C}wo96G#>I2V3$)Q-<)_uMD{iqjeFK#kZLDxJ59%tJ
zkrhw2j_)}ylv@W`-&5!B(+uk}XsK`YA5BYd1{V1oEla-8#wO#}^!Fxm-z5FRiym*f
zc>W0en)Rw1e|IZR+tZZ;>fe;PD+glxt@u8-!UU*v+B~0RK%jlvq&J$c4OT<WwYpDY
z`32}xT+FTY-T|x`V_38@w0|tL#x-&Qaa1gGEPBGSa;)$0y24Vt_1-D|)dh~wcyta>
z^xQ}$oy<iV5_Y%B{<XZ?%i%lw^?W3iFmA$dwEz%ZzojVJXkqB?e`jX!me->zFiRr`
zn$Cf4lF}VZ@$n9QEuK53zX?;Xk{L}+bvASP=7e|pFT2AF{7S3*)R^bv?i77Ep@s2C
z?|)>)uJyv*q;jY%-BAM;1CbiqyL+z&3FPkJ`SxEJwaD&EY5v6)7aArF`AdZF-lhi8
z`#KoM08g2(cqZyQ@D)FFigS^vr0M-4&Z_)-$>i_RqhZRRD_&zQpAq9!+o_m+n^dW?
zQk;ytW|vb`udjxwzHTtXun>aZ{w*Aso*C9Nw4J?qW)N59`t3*Fz&iy~h+qrBUHdnN
zPkbrm+_;SRT?&fX+jw`9`R1Sk$*%tI!Za400da+eZWz(@fcCNY^ct=2Rm31sBctdU
zj;xveSyC0ktT<N$Np1(*u+7%#uyd$L+nt2&wbyZCc(|GE92;x#j5yrYW=Dc}%h$x3
zeawtJk*ahGkTLafcboDHpLlg!ukzz05b+8bUAqV?<_Lj&H}n=++a|d4u1cDPNuF8Z
zCdiRb0#lKvUynU>$csPhwxe`2oS6IC<qL5mY7O5`U#C9L-rh%&`L03qxF1y%YF{>7
z2jKQ^OlSZ4smqOY>`RuJ2RxRc5d&+{`iuB;ruT#07U4U5fv`<Tp>(DLu;*CGhlU1(
z2aEgnC>k@Y`~7+9`=GOyBc9ewv3ar&y|VI3E)fcGtO2o`tVsfNpg}f_&X1<v$0Vr-
zv-}Khv=4H^V$yR_3&5R0RL{ZFBwSl;q`-Q{_$mD|lJ)kJrA2&C)kPi@?N|f3Ga@Qd
zBd%gBS~Emv7f9|6CcYig){1``TU+A@ARJ_qJatRMC{dW9S~E6DyPtb89QD6%(Z0lL
zW4y>EWI$aeJUfIztiU)Jz>vHA52vbS*2|<Wl)EdH+xVWSlku7g1x#9E4kpXrezLfG
z1my5pZeP_rOBBDr%HVX~-sQ&S@saW6$F=Z_`<waW2iEQW<2;cFM-VS&DQba_MTc)f
z^c3>_ef)BpS9Br;|5-VIXoB2Bzns03Q=*IxRZ1nD#zCazn6KDZT~0-;)j$MYQTP#h
zkP0El6)82WEBFQe1#ExOaCNIg)dSygW{L#fAIipfvkEx~#U6(t{E9;Ngj`++-U{d3
zY#w)yEqPcXD8fz?Sw|D!;>RdsQ}|kiF})({kXAlk^I*_27os|z=0?lmxB8uzO~YX2
zP<HIy@hsDHi?T&pDGo(mZ!Z2~zUUnZ6WuMgxwaYi1vrIxqH^+pKuUa;5<>w&n4BbH
zt#O^VgDHW#!q&ZJ*vy$b^<saW@i&fjV=a8)SlE3zd!hjzv&A8lM%!3lS*%!q-qn{t
z=D?%=S$(dg^RKVIm+@Osi;!~#UL3q-Qa(YV*{i5<>y!}JV%y(W;<Jz%auu_aDl{-6
z3l6^DfyLIR_atn@)ex&>v*X=`wzk6N7p?3-KFJdkOy)~8)io4&;)zH*A;ht`RV++&
z`lik9x%?G7#z>#SJdzc{<&De6yx82ALs*ecGD1b}FT>>Ftd1OjYdt&v=^lo<6Bym?
z7N$(h1NL>qMY#0-rDj4@#ca}0{K3&tV|Qp=HRX#pPl<^Da9vRPy(f&)IUb5xJ$}$R
z>)c#RCy_Gu^mWxK=(ht0Kiq#7W@GSABD5QV%Jj{%u@ds*ber;cnbWB5qiJ(Alkik!
ztC-rE4oRtObj=IA(G%x<1;pt>aRN2F0n1c4VyOw7&7Je(b^``O#FW&ALImvvEK9|g
zi+2dA2ttjUF_En5@^B(G4c1hYvGl2sV}Ao9$zqa>rCt-tIJa;w7Gjz8ug779MRa<7
z#jWT}n$Pinb$SW-{wMo$MePf9o+2Z+UU$i(#fd{WJzZR`G4tY#)Y3C~pdJ4(;`y7>
zS+Rk6`V(AYFmd2C5?U6smVT!Cw^1BUGSh9PpZXl+AEuC7+kiv+=K(bxv#bK;_SYsq
zmQDiCKB6Lq-rC|dYb6o^%8Dw&o_T(to9fqdIE6gluWdmem*2k7EyRj9pHZhVYUqn7
zP1fdZKp^2!%I7~?wAb1UC$di=M{DSPsscyn?cj7UMGD)POf+}7Jpk-;ePme_4iBtd
zYa8^HLO=>FSA&mbtwJx2AYeJDt62G*bSy0IbpE*IF)W2u@2`=HDbMhEL08wWkbp~6
z!1cn5>(b%;M6YF3s@1=G1UT+`2FSIXN`7MQH+%e@^zUUGb-gO7ToSUvL<|Ud7JJ?f
ze*(xo^#atf3N+fn+;Rp=zRvCY?fw?Tt(xhMLFd(98ne56iOQ&xpXXIr*!lS6)BHU4
zz8Xj8vkd$EXfD36)YJbaJh|htXzk$ysfLDND3}sk`}Ojwtc|}vxDjT;a~ytxY;G<a
z)L*b`54mUPR5E(|VtageD@1Q#b~yXK-v%mB1=0ci_&wNmbaSgcgU=`XJtMyjWqiIm
zT4|>;S3Jby3KaPC%$c{tmAJZ09qoafz!_CStrw$f1RLM>8EyO%LmdntE#AEp(grWU
z&kQJXp%gr$k`wly>(hD@@@rXuXrO)6m(>jY(j;9V)ip901a4`GYH6W2O-uh5RkFiW
zt^^{1tn8~a@}S#(+fVnd)nhJQ<oVUAKb6fGvDJ2~`q0v#-jG-tv3@hy6H>PiAcWHg
zBuHY<uMh@=5aXuJWrfG0PnNncJoUxf7UE`|k^n8u9h;~yF@p>8wM^GIei`eYN=VlS
z6AmXZ6{k4S>EQIP%F%j<*VC+3mUSRnPLGY?$FlTy4Nyt!by;_TIlB8P{@p6kMV30#
zI5t@Wc9&`;=}4t_W$%pN56+<1VAO;@_71yCA{51GRMNM)gH=$BM2BqE4Bt#Mq}ro8
z&Y)|}G;$y^&#ANwgrsN%EtV?e=F3*xUt*4o$JeO}iEJ~Of7*X%-yST_U;f^h{HpY4
z)Fn<1i%wBHFN4xErCF$uUMGfdx369C2CaIh8P^&^eJ%y^*YCih+Kal`SCJ!AT`hea
zv>H>gMswpjmpYY)K;Pfr7YS4{GNKE9YeeP-GO=D_+!YC&jvuGC2C(WS)UgZ+Eq8Q;
z)j5s_$y%i54d)IVDxE3@o=~nhF|l@tV!@sBJWKDUFM8n=co?CM8Jr$nmr^h3H3Dj>
ztk${Br^m6?-<_bouspvp{zdpJ*&<k`D)CTS?&xzTSa?WP^78sN3DX+R0O?%rl}7bj
zNoS>p{>Qg*L*<5=pn(3TC;Fx@gylyEuBW%q@NAFp`y~_5{OYPQR<Ey=5@^k8hFZ(=
zEmM$7!2X2z?KDL=Mub%z!#|v4r?b_t{Fl&roRtUr0Ua2{z{wY#`P%oFT6@#4u?NCc
zc@fky4P7DliB0d*YUT^nDQ02ubpKwd^pU7hO)=Ix#JJq2%w_QQoT4pslXv?ZJNW+o
zg?)v^EAd%%_i^)wr$t9w*1v3Y3g7%3N04YPdF~I2x-Z{FnEShXT`yX?ALaa0(XL}3
zKFjQYNi^69{%8-n3>~eqZ_6&BNhHWG%j`yr6wGdQ`Mv~eLPcpT4hCtBF+889w6#lr
zl_?Nhr1AN%l<qhwLe7L+jlRzcp(Jmjk~WuSRku-A`<a8m-Py`RZW?BoKmK}rkoTuY
zO73Uw7@~_HPT;m@Gvy332_;@|b&ZDrrK?b#I(2D7tAcMHw^q8u-D76UC31c~$kolG
z9-zq?l-_wf?4#>KzlhR4Nvv&`q%<1$Vss4>QuAP$Jfob61DO@yf8zt~qtZv?w=CB&
zr9DYLRd!@~Q!|Nlc1GH@e&a7ZqDW^WVMf8Y1|OdA8fz243;!JWA+ZY)r*PseNj5So
zbJKMrN_|*Yt$&-Su_<J1s|iQ?;r~3cK%8wkEi5j+`MLBj)`RYKOqnRgLYDX`8yi-6
z>;?>ktT?Tt+{H(0Xqcqp$uN{(jI8(vegXAk;RsDgypbG6ic#t@VKQl*+yBIOKQWrp
zv_qc=ij^ly{F5*_fkM$+ggb}1(!dvaS_)0WtG?v={@!lVa6PRx+0yv|w>q`CuekgV
zbAwR%d(fL1D!(5teu_~hE;>7fX6a}y`u2IH`pBbGVr;;4eNxn*F8oH{2tfdvob^A6
z_tY!85Q^qxCs<6xArIRj*_g7(D=ITj{oD2?0M&SGFyCJPa+;Vex9Nvdz^kXlw1)nV
zHn-AAO4W67Nj@YUNT%T4$}*{G{Oyc9tR(Zi8=~FQ@j0Hc4!n68XlUyX*k3=bx;Z-u
z+H^Pj<?c|Lmnj_&%d0q5R<>V$w~#(ON!o=B%&7lHs%5}nyZt$*@MDma2e%Z=#8U>P
zYw~Y%)%${T8zErM;Z2h0>H<YKWIqIec;Q+3te?)H^#Bu9^Z37j1h&`Q_!y{=B3YV?
z$Q1VFM}gN{s-R_KX>vOLck~pIMhKb3Qx3?K=)<N`H~5dme7I)Y(8e_bZPLwpKVjf*
zIYr>(<(!^+!%y*zXhOjH+16NrwEt(&T&*A_-talCorsn}1_E%u*b>)KJKRWL+W(3n
z>;$<|2mDT^2u{n&h3wSsgW&s*NYTwrEnTMV!wFozaEdWBT#i%U7i7Y-ePQ@nF}B=F
zARiKHoGEBRBL3H5M@7AzbVq7UHo~NA9+<bt&y%s6+tu|-Yo)O%6sHD1V<8H@NOe{+
z1cRBLJw<fz0n3W>v~oYI9?&ntZy6q6tvwX3ZSGC{T5h-0$1`aC$vW6tU31BVUc)UQ
z8#DWYNh$L1@iWWu(3vLErjZ6s0N1ztCxp-N=P%(`I?wD+vvRlLjmDIBiX|=m#%?xc
z@7{|SLYUo?eyDJV-sY%c`fc_Lir4A{oU@C!2M6P*1<4NBXr*4`1*WY>RKZM2iAdG-
z7;{m-#+oAK!oKh|PqxNYxO{&P{?46=sn3a1aCYdayg}jNpk`pL%?1??>Po6|^`zXZ
zD5>JE78@Sl&yiw$^`)iOC>PW49hqSp(a|1d*YFuS8oSIGr3K*mdav*I;}`IKEN!ew
z+zP|26oI~PE}8A4BZCHx2IjOBF_XYF#Yq+uC<wsdBe9D3d+_p?riEVwJGa3jd9^0B
zjP?ukwp?!~?InI16;prNYFii*#1nsGBg?XAYU#>`LWsA4aeT~zkbdRqK^FL|6S*%E
z_5(K?SCtc+moA(BeSn*P^-AEt<jy8C{;QB0Z^N+=oopMq5b%d=;|L?p7ME1urzx_{
zk*Gk^CM)}Y>P$*P`Xvu0uk9tcQ~Q`njsvoY%hc3}trHDKGD|pi$k0I`^ua*P?_9lR
z2`*-Pqqqk~^isqzE{!mz0+xVW0LiTKn@}pL17?mPq1+UZ;ahVm`Y8}jq~RE-FOGl&
z*6u(>#-ymXrrN|x>NRmzjp}*I4n03|5}@==z*o^rMMy$!(XyMtjosb!eFdB}G&C3Y
zkGghl3)aF=pS#Df%8U$k+f$?;E_t7m6M>}ibLXP9!h#AaplGj-;N>{xKUh6a++>Qs
zEZyXyMN@}gY^4_@-I3Fy^HgfC)@n<*E`%WR1Y2m(CR&VN%iv|PX;_hFZh>XKo!kCR
zam+l*vac5C9%4oxo8ysto`KB{`st(BSkE-yKv-6B#mU{5+j_IOq36>Ybtg3*kMz;K
z5uiDDug5z-tB_NbtE;Quk}q1Qd>WdA<73-o7CoFFw29u@d5+B;@ctji=ZNkz#KIbb
zZ8eJ)j^jDDqVv(}!qy92_9F3MxY=RA`762a*%y)e)iBnBo{)ZoTy|(K{5H{|&G$7c
zEv394=b~cV2Oge*If5Qrbt-oZP_8D0#p4NIpm8dzWItR=dZ{vv*{&@tuc|ipXw6Xy
zu>N;J120agY9OkAitoi?eq`NT;N}+q&^Cnz4IhWxTe`1U>R#wK!luN?5PX&;;Oj}P
z<XLJ^EPpwpOeIr{JR3GK{~K}0+hg}V>Kc3}XhHvcc43p#H;X~nZ8d}FsqQ6Z%ll9N
zynX?R*G3{I`(-~z`w-X$@K_y$z&sJKkN#`jkO`W232rrg)zul&(0&_}R0O);E+xYY
zkt>nGs1k=WV8@HG$)n|#tHphFRYA%71q#9l1%@-4m_v67ae|`bE-yucy1Mz1@^0`1
z$~)D<1SlI@TB4h%wD2WVmtbLnA^<ejLjG4wA9<kZHKr@Ar2#8#cB_^Mp5<^0Wo6wV
z`ry<7rba;-#>+>&4`trw*R$N1^zuK5P6@E_ve}L6A&osfL*EOP$j>FC;nZI)3l$90
zt3^JTBYQXG%nlVibA7Yq)gUrW+{DJMw+KktzkL8@Z`|L<YQO#QFf^xv!4fo>7Ae7*
zdEvr)!mYgGJzvm2z!Twqc?SInn^up^_|!2=svc#i=}4Bb8+90_z@!lK!KE)&8zZ5w
z0;l@T2+CJH4)Ec2Oj(VnI#;7+Dk~M&?)^WmtsM%gHngleqLQh&idwH}RRtd80#&3+
z*%QyjYuBq--xNQykK_4RqBmv7VJxv|yBQ?a5*@dEb%^poP4i7*@8l~>z9a=~1^)00
zvjVzrV~xR-9Y(1X3*1b~ITu7qWo}F9h_0icvBszp&#$7CB3fETvj7sk@)upFOi{5u
zKj9}Zd(2Lw!`uCh0kn!)`v_Lk(D9!v!h`<BsS3Z`l1klLpe5vhQw{~+7^L4Xx+A#<
zIAI`sG_fpjDpXGYHeFBM`|iReE*d$$bc7}4x<8*W`OfpFS+jICgTmlHR!WkqA<qnz
zcboL_%&Uk~E9=X%9P9ge*UKNI#97>EfU#vveA>5ltLxF%a{pc^fQ`p%z<SrUoV~=s
zX>Txt_w9y(jNh%x%O(uf{&nEhnQ*hw*wYh}6ZD|C@FckUJ5@D)o8L8c(^=}F0gXt_
zn~@kE27!FKGkA-+%=2{OMV!(BDlzU0ubkNkFx3I-ODbc<4qT)1+(lX6Yglhkgzt_G
zD^H2xyHqIheq~cR+%-*TdAEBi)uXS*lxmtHvScIWmfI26Ym+?VBqr&%M;c!m+I?BG
zhDi2b+Zvt83Ak-{;{23UlH+y0joJA)n4k2T?XAb!_TShhSFr3*SwVdRVBMyf5g1e!
zm`D<Zu)g7`m@@3RS|s`MVTf~9Js{TP#|Q;fWKbxd_Zv<Bu5i9Zw{Jqoq>~x{yfJZ-
z5@^g-4Evx66|S2=(Iw3-aI+|EhE*_o6IPv4`(2t=^kDUiAia@DmXqE)$1G`%SNg#x
zuEbT-dRf#UyTNLH6=Lp*xfZ){kHHY{Xz6IByT<}6rrm22F|lC0IU()BrR|4U;R3*+
zAHx^*S3<p#(WF$sbm1M3VwwR>5fem)gzt>aS>N>!3o57Ghl*EEBWHujuH#(A=2K{~
z&tpd=){VflDIcb=fET0^ZhM52csXxKqtRAxrYvbDb4~CU2(?KhtGilXkC<});|j2A
zs%fZYkZ;GsEyKhwy2jm7wy>bg!QJAr3nbPE{P!?0bzK7p3z6CM3fi7~{=nT8wh1y;
zGJb`pEL(+%r=f87KG!h|y`6z0l+++{MfnCaHMdP}DIsB2ZG=EN{2hbHCqJ`b8ve$m
zWNMu0r&UX<Ey?<#xA>u4t;{vmf*4iwHsUGOonzyY=yTyx!&RBGW*_l(<xb_gKKOEm
z9=<dPs_q~Pz^ltca$bbRgRn0z%XV^yOrFZ}**O)x=e!SF(~rzzO7Fj=G~S3dx|#b9
z{}x@0I32M~Vd<CC&~IdABXkI57N$V9L<@#UuI2Gw<nxZDi)fizud(Xrk2Ye+s3o~>
z!C>Trlmvu?-}f=fiQ|->1B<d=R6OD@XDDRMjf^NF6Vm8Rr1*-keUU%jrL!`#LOHU~
zUJp;UGL@RVR>MvPaRf;Ywv1PhOtT_X>mF0BZ?~-BhDH7nLckCaO<?cf9Y&YcaoW&b
zMy|hE_OkiVp{ix*j`hpI2_ZG40$E!+;-Rp*cWtww@F}h3bN-$bR^X?mxQ_RF2~eA?
z(WY;1Zu3jh>+J|U@cr_Sm*6BSC)(y|ZvKnV<GoM*k0aEI8BQndLik2I?l^nXI^(^&
zOpW_wtD_8!U}3T;jk~R@hUL$%(v=&KBf#cJX7a=(<~A8U!C$sXShg}7hf%RI#$c*r
zp7q1s*u%zL&mS9I8%4T`@@<=z?~TtZfxO<P%0|rcoK}x#&~sRR`1oP|PnU_LN^tw|
z6+{fwCE9(~Vjy$hhkHQ+OuHN1nSD1nkuR=m3W=?RoIz%3uvl(R?l2Er-I`jy_#DQm
zsIM7#h}7PUbAzw*QdoG<B5<wd(xtBRs};Fo8?cW%p8<0ohmDfsK%+sM(BXeD`Wk{X
z0{o2AGVZp0exf>5rp(im>yF+<(oK3D_}?E42dD6R|J)=HgBhEGIN3LPw;BCeGB@*$
zg1$YpmADD)SRxE3qc&EaMMYDWys8Qd34&P^Hpa29a%ww-WE2&4F(<gjb#Hx>Oe{fr
zW;1`uOVRbq)NGk{DoZ>iSBKM((uC60$MBav8C47^OuF30TTq~zqr_n$?aUWB3=3LB
zvLmHA4=HJ{Tmvx9A7d}i0<cWrF)hFtyHrccg_MbRf_74}_EP!!MxtGj3Tyy7p(tZ2
z9OzvnPLbCGqCwCE-S}D~NMiA|>#YQjsL<&x3mDTfyUgi!w^2T1{zL4D`PAZ#$Qau>
zNe_S6+<T07&G(Tr@h!udQtb6u1t-+1h|X1QZ0qs44FR|+cdQqRf%rcl!;14KRu&zR
zV1ELMpR#4@B0_KR?qxj*W}K}z)b8?9A*KRon01U=LZ1vm>o7qY)gRR6l+0r_@5@v^
ziFW_46CMmk8?@o-xMyngX?SBU?xDDzDL+y+Q$8ZHD{mEGD;{o@h^;;H`Q|d3K2Sr=
z0ZDc7UBO8sN~dF61Nj$LOOty&9=&uPpqi0l8hII)gDxH=9)=}(J<+Pm>f-#K{L=Q~
zYhb#sP@aepsmd#6b;0P_7ukslWOqh-xx+89WJhoAh3PSvvzpe0sK3C)vDqJQcdzd=
z6Z68>!oz=v5)h6`g%-l>U3xw1Auqi5`E3DfRyWjVqmaSP%<Ql*d_Vp7q`5Lwu&x3i
zY_|2+ZQw<=7I^*WFX?kd>r&<Hm;8bEu{T1Y4tfP6r`g|MdF<|39dM@sXAj&9*ZvZV
zfwSB@p#G-tE~i^j7g=v}KU&p21Q$qxq?`uEGL#zJHnC$?R%8hXhdLk5#{MgOdg7^U
zWb_}BO@jg|OiR$_wfS0K{XYsTUopF98*nFwntp=lq8(b(x=lE^@UW?jOJVLs(nT8B
z>npPXo5(Ma^QXRzV7!8h__g%qk*Ae+i+<i=Q5m7=IpJBm*#eiZ9pB&iP8a8YQ;nL{
zYA7|?YLGJPn|aX2Dhl6Ce0ZpPKs<~+L!YCSbxaxXs7OO0;9n7HxO~<@)Nkkh&XTeh
z*Ql5I6iE5S2|Gc(PM2L?UjE<+&x=wU9F$Q)B(cP}Of8*H(H^oJ$iT-+{+R6Y*g%!K
zC-$1k<VEw=^+ers(atXj&Uv^xe&7QxZbHKkK@TtwT3t~R{ohi(?*z+h$BOr{Y`NId
z2IB-16ul`Tj!Qk%e!3qu1z!`DqPyhX*V7UIAcNT3gpM)Wu|aM#1Guqf5jo;oJH`p5
z*n!j6Ek6r9K&9pymMz_F@2VCS7B<Eg&|kPI3G>PK6CuX>bE2D5Q3EA!K$zMkhKku`
ztI^h+QoOR%3Th15CZGOBzs5pPsLSN<d#*r;l@!A2ya9O4CH2G#(GHeLG`yz}J)c5r
z&Oh&OQ`ih{BzT0JH26-f5~VT%y43@iqDi7ETyixllL!e-c^s=f+}C$*C*5UsJcs)@
zKRZd@#Xhk*9s-6&M~pV!%5AdR@F*M32SMMCmf~LStqt`S+fa!meUQNmW;VE{l(!th
z*sQ}^JNp#TnF1;FXho0|Y7U2xbps-L+>^iJGl$GTUrwyLl^<h8{icrg0vK`@5ME*#
zVs6w3V_^e{8}rMfKHHkxhN5y)C?&c(Bq61!JJ=wI7qZ^!11I}mfBlzWAw}aAdbF*7
zv(HbQeMYJ0{Sfr=(K&7X3@;07`tH<Mk_djKlOFRTozf#^O~BOq0iWARDsxfqSySay
zvAnO?Hv^SrUfhS4WhM=6>N3Lo13vQp$vj(!Qldikch7j7M5r4Y8m0t&^?K&sYpvWI
zf4q|d@xFK+R^YcioO0IMOS29UBjmqB9Usxw*EN(P?WMp1<8!}nbr0e2kA#2#Fn{Nr
zN~XG$ENH3K1Tj35+0xRIlB67Bwtc*o!BhSHv>Wi@_=6wmTLAmhQ-*B()~IW>0?zO`
zX~T4r^q6-6`3tK}p4f=$w)S|$!0bT1_nS{594<3gXYC{+wSEo`&&sh1N%MH_Wng5i
z@A=B!p})M+HRdFDZ(wDWDqUbH<{YqJ;~%h<m9tCnh$$5vkK$LUeck3~N4ldFuiU_-
z4hZhB;eE?t8xhmieHRN8pYzs^z-{;dSBj>^9SK0%0<}JF%smtLlUh9=d3eJQB+D|e
zxOcB%F7~)wDs#fsZh~g_G+k~StQX?qCNu;iJ|ppSa;yeaE+-bT>Noe#vvM|Wo1}||
z9PO?>w!ctDm^6vV8(@_3sS?%@2mj+#5v`jCC+qEGaVtBgeG%=ganR74J*Dbos~5?>
zS0CuK8TN#M$0nC8R}>mVTMWNC3?1B*G;NGa#~w62t1hlGFVJ{cUI!U`MFXPLB-?5s
z`s==~75I*MR+3^XR^0GZlRM$L*>NE79ZdDIM72t%1}8=GZad$3T&J$AgpxPmkiD13
z$Bc3$GLZsF^UJ+ij$tfYYRfWy1ApyfRaPbSzW|;=VZOFWbkE1NK(Z_v>tEruK}-E+
z_r!fzsqX2HDT^<Yk70iNc=`?5-lq|(C#pimu8WY!cW$r(+Ez1POc{?4na>wqpdHYf
z2p`aC2nG1WlcRv{K|`0Tp7TkpSPVA-_kb<umf`T@!0mXguhkH2JoeRpA3u(hj{mVO
z{9*iK&-#u{b<3>ZE(rkVwYJV;g9>tWJ+szPH#M_z&amgeI8xzAdP77QGZUi)jxIV7
zH$J~+cujKDw8eGRwm#`0hTn83ej7DbF-)jJz0gD{RGOmF1Q|eF46lkrdC(|CaGL1a
zo5WD;^2+V7-|{~3lMJv-082tUHR~fJ&Zacfk32n5hivRJIe{>iQDc{78JnA%-PpwL
z67ibq^?F^Ac^Q)7(ln*2s&4L#F<l5VGJ3jg0LS`<XWU|M*nYo{5W<&{3fjHZ>(Pbd
zYnlcrWMB+6QikkGy<D@L1+9JJgb<RpHE3-RMQCLvC8NVb-hA^-#^Xb_w@>*@+rV}i
zYw7h0wzju|6ZGk^awnN}&ez!tk(B6gN9NbhA4i<}2<nalxQ~h;#P@zW7OrLC8z}hu
zP}1ua42OMs{XVntg2~~WM5P=YjCkdhS9tx6pP-wDR=3=~eT(Owd!DWBO?G$h(zXpR
zyzl~Vzw-{W`2u6SUeg&%mM84)-3?>1%%E-?ul>Aw^(xcp^wR-#!u3JVa4mpj;A)<h
zj>!*yMERvJP`va!NKaKpuR01mtz9B`u_E|8qTN{$nx<)wul$9-!pm1KVahSqYR?iP
z!D<d}T}7%;14=4Vk+65?7H3awGu+rjD1o$&#cW17TX@{0Hu!6Q^B?f?=fB_ukTUDy
z`NvrUAM5v<Ypn4^@DBTHLtkgFtw>bj7gQYcQZh&^+R2cuIPZX)bM8oy%B6Q=9$V;W
zZ+D-?e2(bnBuV2J0kyi($UuYpC0Z$8MA<llLjxMmJ;?CdmR45?k&>tc5js{Nd?-<C
zhi(PddU3aI4Q*pt%o}dqJmACkZ!nve6nVj5*k^OIp_lg<_WHzdcPx_-y|;iv=qi$Q
z8-LhVT-ao->u^hC*&=Iq(zWG7MIJ&f8Ke_{9ZPWJST0VF1vEsmCx#>f2(_-`T)Jae
zJqIBL()vvm=a@}r%w{v{x}s^?C0IRfKHkYH@b#y5GO)WJ19uVtJg&(*&gMxekEtG)
zKV3N!B5Jas-BI{l>79zz?>Wv0p5)a(4s(QbAtysA6B!;Ogpvqg&URr~pIqw0jk0rQ
z{oJ;#X0ceX=nIUrC}&7iN+MFPGZPBuQgDt$B_vW|1LGFiB;PMNf*Ff)z9_M7eI3o#
z=q@uqW-d$RlV?<#tYnBLz73dK)y~Nd(7AZr;%FoVmaMm0c|gEQySwF@mZ$`Yzy`L?
zuYrd2F_fH%l9*M8A+D)DNy|Iw5wxu)RygPA^?Dt^Th-I0Y06+Q=oq8LVnLQ=lx5kC
zLtf<QwxvkC)a2H!TeMn}7X?kzbjO#HcFb~Qu433TNfk;-pCr(Mo2A48z#3ooQwW&P
zE4Fvmy~a;?!lY>$Zrr%R!Dz(J&S{J_IPFkb8qTrL*5+2IQF;{AecG&-wbnZT9?5l1
zQ1GPVQ7l31%Kd$kcD7^ej;(R4Yvkj%um*<2CMkWFk+roUJ3Bk9t*>#I%qZtK$@7Bo
zc*2V>KF?E^FZ1CC@1b?W-rg=Rz4Q|2&Yk1K_ph?Ix5u;3KF5_SS9t5~cc_{wXiWV*
zvzX1Os|CGYK~vB0WxAl3_n3_L7#)l_wSDS=j{c!Q-QjqFlm@2_scyI+Th6GQ*_{P%
z{rmsG7tfs{KXt~ZuZN2`P1E2tsuiYAbO0`vjPMJBpIntB$i!C&KXv)DB-LGJ2X~mw
z4r$5-w{Kh{m(c5NV2~8U0d1|BjK{q3=1+L#bDt;6`!rR>V!mK|W5_EnJ<DJJd;cx3
ze(9?nW8t%Sm(;H{A;lltWh7Y_LrKnuNtE(^TAi;82B1xgc8+;zF?}%3gUxa=hp=_0
ziSh-S10&^yv4NdZ0`(Zeuk-mF>wNJ-=USvnu+HF|_W*p)K+$P<tZQ+?XD@>Wp>eLo
zIpG;Qgio$e288h&T1e}5PSQ5UcZ+ZeqoJ-eRoOBf*X-RH^Wg_~S<GvSqF`gQPgOPy
z`vcZO@5Q*uC(75Wb20Q8Mh%=7pvT3*c^1?;e3swJ#vgS=$AKMq0O%iyTNAAO?_TTH
zo~R9bJfKbl_QUtq^<Qssn9zzUCNlU%KfEVE=_pWFN-&!(m`*1w7W0s?XODz4{2Nym
zQMY7x?&tk}R5*3@`Q^u-To$H#On7QBg4X2%Pv^Z0;sP!)&Mbkpun1xAIgXDD2R{y7
zss{<4aWIYm++t5;hk?M(Ip0;r#`E)Ke!61<oFp*yhN`O2LL;0nW?K|JdcD2}UMlgR
zPD&J#c7?e#VUcjchk}hUSQkKF-B8yx+8Bfi?Xm)ZOR|I{&AJe)RK7T`RH20%B2aGr
zx5Il<t0Z`wFR^k?tLG<H9}_EC&~EwMDm$%`N6#RS*{*q>A8GLwACF;36FT`s5Vx9{
z8f(_|dOa441#PHBnj{J9>+6(dd8ChPbjZ^rrKxMGMcpMw^!k0Q^QBU<Ec3;Hjq2Ds
z8|w6V7EPm^Kx^04^=n)0b#*d~mi5=4(S8qV8cpxwfc?W=dc6V8XjBLn@9y5=x#yk(
zu}sbo(qCI^>+AQleESnI=Ry^UkmVcTSnM8U+%g0BICH=J2zTW7e=IY`FdX(dd-hD%
zk9RN_u(N%NT&;2Y=AN%-yJ%^ehUczaVQ=>iZCg>-HM5zo*E`xDvA4H}l#=JKyuj}6
z0egFQ$+NU8ff~!xY^<-dzk3LPk^<dovLxf?jT=1s^ivOX+z**@slb9Pce!G3ayo%|
z5Lw*)(c4_V{D(aIYu`ZidPpJB#tU>~yB1i2I<MJ^6}i<C)G6g#=Y(RFVsM5=NJ@Q&
z!|^_g!zp>5v3~XfRoPJ0nyc@;&h5S1v`%A<;m)mF4A$4#JGe!qE1rMl3;fz&{b`>4
z{Fl8(#)kSyA}}}3t6>rq3S^4NoOzv$vkv1lR#;@3Ad$YjLD-;u>b|9H9a;-~^>9a9
zhwyILbcG0^Pv<&tCqyhp`skppocok~AM)1A&sPFzqh=xS4#FU$MVMGV!f&P?;ITY+
zG8_#EKTrNC-+9AmM^zfiMN3tAap!bWF&fReb@PzvxJ7G;(FU#Ov~A0<U$C>eMqMsw
z>osJNgYtbNV;I&Ev^&oFaBE<69AKmmOZpUfe>kpV09<$MN6j7k{ooeC$77ftP-$Qv
zeUolSm>^65aXw_(`PPn3KLm6kI;$<S`IO0IOjY^LArUP3*NNpDTnD9A&glWOevkdV
z3jB^Ub-+E(2c+%!$nLpiDA>QppJ*)e5WekWhw{^6>a3sxZq|ud;=1eq>RrA7@B&ju
z=ocHuT?o%QN+HOylr&3dbc1#cIG?yMY}ZJ$JUEG|?>QSN+o(sBfps$0fLPYJ)W&pc
zij<O62vp{EWGYQbvJ{p11PGzxuCa8QyYUlCupg;6=Um6w#n*N3SsnkCe@Fg%=`5^-
z)Nx%jWLc2+lbi4CklvH+8bMnGmOgpQ13DqZ5k@YCwO4C})^*(hL9KPyi9$-*wFqnK
zimIxpnwq9**xcMiN=Y4bc&i!Rt8xUXB(2fiIQrhMl05ezU*E(%>^H%w0Ow1a0JB+%
z);@xwwINM1X0s``Zr!9T=M0AH=#Uhnl7yltSR1Z&>-DE9S2$*(b0*_86Di&0zvK0N
zoRPa9EBEoXWd0u3WN_G2DCy(DyN|v(=lJSZ{~TZY+E;n=%{RGm;|7DlAn4U=9Fksd
zz^z-?sj7;zXV0>|y~V-)F3uUo<3moJIzyUyz31J#ce!xk0v9e^WICN7arAqh<=xs?
zV=<p{`O+DzYf&T^=f@&VJ*czh!Q=}Mm;LIXr$=0o`#dY>7$K>G_Ep!0YrpgRY+gLa
z=`VlDhf?s9yb9{#VlM(q2P?u_DX~H!(}FDVY2Mxm-a@1WgY|-~?bGbv-R0dk-{<1x
zXE?WWmaV}yoBcJeU;mKZJ0mu?Pf^#tiS%G?lb`>Kf1Bavd0Jx;y28n@A%*Jdk&THr
zxtIHi5JL8gs5PEKk7@se4HgLAGn6ef#<N1PAf;38SpA()pb~pzfqfh-gRwD-)?#c+
z+$_C@!ysB8R`s@ILxfWZ5$fi3{<&xGtZT8h^%<;EhHOkgDx7sRwWX>wb=3lbgM%5n
zcgBnk7U(8LsRFB8%0+{73-<R9nNFsZiv_xE$<h>&MfyE3Zjtd25SdO0(Zm4eT=zQy
zg8KowxJiDpyI$P$$0s+H_5y0Cu-4z8qSH#b4w!YAY~{T&upd$iaF%%o>Z(w8%ObkA
zVJoFwg!Ag_IF|PTXGhL?b>lzbb*q2hPje@Q@QJs_Y!%Rj(+!~f!RP1Wj{8x}5;FF}
z-!3>677uoujiK0>B;lg`tdjynx=Tt)8ueVJK^slwG-X)=lC_O39KL6#XV8545zpGX
zt~7^nAqX5CA7d>F0V?t9sPf;F<~b@&5Q+5tW0iuy%mE!(6DRzH0X&k07{^qIWm~k>
zX<7w!CUmCopk&E1#&ysbLtQ<(rcNr^Y3Gzu9a|RxULI=UMPL=fpfNmK6a~ZKuw%HL
zb6we!$jDVy)q%S#%cyJLL1I3iGZ^&g4f?$E&O1Ey)KlEJaRWlaMO9VUkQmc6O^^|#
zBude)jF^`t$x!~pI|0td=axNNrMC3?8M}8!407MSr!gf;__l$2dwb01bJjPuaL$*D
zN`YSA>%NXBAN)y}kP;=9S4QWi3IA^?WVdb~2b$JBI_O{hJotEr&jE=_F;;hxjR@$j
zTzQ6H`?bHwv(G-u{{DU!imyuFH+^lmfe?zSYG~Vr{vhYn=^e)70}wD@%(>k2Y;jX-
zc6WDq?ztB^fBphDZ(gTq7G!CPZX1$B(zF#@TkF0bvkT-;h;)n&_Bpt7{ejN?p+Mbs
zKjoY$IlEXOU4{@Q=$`}?g7&cHt$+7B49=gYID5g@9n%JxD}?fO;4s=RUUA~31f{&r
zLn?&y3!E3%TH3b38SU#YDHv{VlVt-;^B%><7Ph~M5h<#_&V0V0KkRey$`#7#Ar8UD
z<~asi7eP2&YjDoji19NoK_-DA@EKxVttNkEAeBSmuuhU#FXUgEAa>4hI9e+xwWCmA
zZ7kd2iytUyNu;0+o~)7r8$hCShR1;ENK_FvQa^S=5wfH3i7m24kRoVVuHy;9W#C!_
z39-JJHxKT7xYmHMUC7ycm5QeJ8oJr6rYviA?;ddX&H=Nj#^{t(^=VD%ft}MFjwVcJ
z^DgY+oI{DAFOvd79ixt65|LR1pIJy?@#0GYV<*H4W4AI#E^f7-Y{6I!M?I?Ly)AGO
z>n7Bu4Pxn_bJCjo0h^Ag0}sxnXBHM^N!$7j#8^uzCAe50F1|;E@S4}|d>q>rpS$N6
z?z1`E@9`x!>cZSCr_^;X2=g9*tc!H*18A=v<`^HwB1m&WgeoBpjioh~RtU7Qw9ac$
zt&4F+AM(pWW@?rt2s8-Ca5z9De)Cp|@+`i!EUJpp;XbE!&VUQqliGmwoAZ1=@3df9
zqWn4=n5aC@Npt0yJ-nZwlnO}@2~LS_7$czb<d5rGl(~SeW6xk8!LXzONUmy~IxU}w
zwd`b<q~RTnWg%m&b*PD_yT6_B=QNqj9(jILNb*PmBjv)_b){KCowz*DyLMr)9J{EO
zi%Axdk!wSn$iu@!Ha9n!&*$B_M=f4i&V5#FT5$gSc`sh?_vsA=0NlQPoAG3VkiHv5
z(=?Q28QMenc2VF9(Hmo!PRowD_Dq~N;irnW1r{w??(ge5l)8wE_uk&Vzm|gjZ&sG%
zMGxH?io7^7fAP0cs?!^O8m6Qq5uyY3tLBN6u)aqZ=o3(Ek6TX{=U9bOE81wbw$}Np
zfAv4(^5sk1x^=Sy=!->3Sv8C&Q!ZY3hWV_dX&M%blJ$)N=g*(x=8da3YgjC1NF~|Y
z+~S?L-(z$zB27{*oV&oaQ)k#exQ&#8ey_ke?|^J?Zbq>O&RVo-!8+#i3D@3w^MTIi
zp+Mc=`1@a@fA<Wp|4;v4m{o)pj3@-!8k{VczjufCf9pT++<*2L5xt(z>=6kvG)ndq
zrzXZ9rBn#1kSfCo$#lHn{WouL_wG&hcJFd<IHGPUUbvX^OJDsAO;eF%8E4O)rJdCb
za*3%7l$x!rQ=GbZiZm~{_Th-q7%pCX-s`ul2^W+j){2vf^u=~<(3iPY5j{r*CE?#k
z*prTK_ZP3Wu&5#Jfw3B6O<?gt=@PLFKSc&gN?&xB$FL#cq%UJ)JzJwxN}6P}twl+P
z6gf&YIDGb%vnI@5h7b)zQ0IbI?*b!X3|9IM5FQwW@Q{;~rRH!nMJfwu?%Y0LZ}$*u
zBvPdQvz<W*4`Sx?n(1U3g#LAxowZ~-{CN7|u*YD|N@gdk44lI`(Y^nOt-St|n*9U|
z&q;Rf!gHV{hojpZjw(t$^Q=|~1rb_%`F^lRLtN3E!zZ{HO;wfDO-0jG9_+y~q!#{O
z#;9cnY)=Ao$Jw`4JI=!0FWh?MeGwFJ^fu}Md+^-5n<^aZy}{yp?IgzVahXMi+H9he
z0az!y<Q&iBc}=8S2`fnm(T8xB&R6$WX%e(}{{B%8tgtkuVNsTz70NR5Ec4?i<I?CO
z90uogRLN2=2cUu{9|mNZ#hUo=y>r&Hi^e<9GBA3P#dXjzYpd+-@rZ$Q;WLayrHLN{
zr#lUo^e^hzxTa}n>Y8#<GM{_R=X5-!Dl6vgqi13!Qgxxz7~$ykdi4AKE)*L>p)vC}
zzAmzR!{N}E{|HQ8?EP9*6@$T`W95=0=}Mt^fZ8ydP8p3xSZjS7vB7{vSu&f=XqpBg
z=)h#tG#z8;SgqgIjc#{~9%rxPBx&tP60bY$q7v3(fWyNPWm&Q#m-nr&!6(_+*dWhy
zrqk&?jM%4XLVDmIi--HRlW+uz9lOCF2H170{Hhb$MVL-_{mQj{(x-t~_9qVWyt%o?
zFaPo{@r5tE%C&3Pc<;UUm`o-(=Qwxn9Cz;?FzByw_RLvkQ(x9(v8X7Df>Wn<*xue`
zbg)m;R2Z#UU-!)L;rNiUtT=P#45!bWWl>IWt|2RO-v<~=o@Z#S!*Tm_OH@MLwv2CG
zeV{RVC{Xw6mtUgZ-(Ym@19rdpZJZM*B#DsJwnYobOv0`2{4pC(o#xcf{v0+fKqjEl
zz{W`s3NPMHNm8E_F`qWP`}X_%<R?Gkhd=rO?|kqsqr*|CD+Is%l~?%E=g-ma=dATM
zNc$P7R?NmXsb&+jU660|NP0PYx9_sMzfUsQ<m}V0pmn{}yR2LkBDzE?7euEC35&R~
z#@BN@CgxmzMr>u|h0JB`7#99QG1_#1PX@18Iu{rvuh8+1q<TEzSF_Y)#47l)Mx5^&
zks?Kg6dFMjPTkK1Hbanu^J)8$qVf)WSeU{_Y)#yJ1+(dbJGXb4&i8%s&cg+Vhcl#D
z0}`B7IOhST6N;v;8IQ(HCsQ9fSYP+!(-pY2L16EJsc;_UbWD$bRk)2Kv$FzX!YnT*
zl-RK36Pb-)Y#%)|@X`yfVy)u#%>mAiY3rJ%(x||~DCt9fp`7^g1XfH4pGcvN4fT-9
zuJ%?JPK%27j!E*xX<b*|<YYkiQF^-j*=o|lJ&Yfbby$MGRXhIpG7R_;z%Cx=%6AgL
zqj(I$M@zicB7pNcI*r4Mm=xjJF&EIY#B1mhB|tdOzzT;-6ltP-*(K8=obkHd&_-%e
zSIic3@}dYp#Ooy;4wOV?p}m#SA$sAvGI)k8gw}-U%8w9Ak3{FRV@kr9ML@*Lb82PY
zt+mu;)y+i&MOl(Uk}{jkNJEL0EEMvMY*b_yN23wr@%Tsr&*pH0Q(LEKN#L~~JktCm
zN!n$&)^*(hsK`Jmr8*s*QtC*(y47}DtJhpC8xucQDb=k@5v*ld#>U15))*F5xvXc`
zwv0xj4g^LWTg+Ytf2kUyyU$)0-F2*1g6I}AqF&g0)t-}ZSOZq#fs$?6+q=tTGC?Vy
z#p^*yOO_REZSAnOw#NSc{*m$fw382(A^l@d1kWzYukO*&J-~6D2<SxI3GNdy_?XX?
zLgAbxOB0OIs6_I`FMg44eB&G3x^;`2w{CE7aKL0T;l&qUW^HYQ-Mfceym*N`FWB4L
z^+`~=r6>w?+vD81b4<pEfS{~Oio&y|RaG$>jW~1WJg0Vc*t>g^vNmK{g03~G^j$$5
z)^}kM0wEMqI+SxT8-MH=!9#+&JS%8>1<(A2uTs7HE|U-MkO+Yk3XKj0QXOVtdH*-R
zMRxu?gJ-UwMA9)*LMD(%UjjfSxI{9aYX0auKjOFk!*BB2-}-IFlY<U<3jv8!G^Qn0
zeh`yW8Eb<LWG+bygK-tTWPoTDlku3*_>kSheLnj$e~~oHa9EVg0Wa2d#Ee4QRlBxM
zV4df!gz_7CED>Og4e@@_ff*OGGKJ6NNu=`Cy>V#ce~b%t-5el7`!VMb&LW&8Nd?w=
z%2$2F#`(Y|46&QHb(Y5ZdT@!#F~Z`c^&6i7PYEawG6OT<!I8uwF}P6B)?kIk(qhO^
zsBXg$LQ<8QgZ(+vnZ;Q{S!(LalB&=v*DWxnA(1|du&H4>tC^P#Wz$mYCS*j>ZPXr^
zh%iq!K)CQbhVWu!HvT&@B*&=LPEWM**s%w~*xhGHW8>Awp!c}+^k=YE;Lsc#e8^}t
z!c_(nif8z&&%oIE7$C8{jBravGXm?Tsj2J6hi+EFnj+AB5y0W1N^E7o;%M9lSjEe6
zl|?wZKDr~n<8d9yehmxw(dS&}Ss&N;b>|R6vLSSgjfpb?Cqkl3H(!A%clS=|Co$h8
z!e_AC0PH%^zjIx}MAr#K_%%TZ5fU4`9>)pWB`UO9`&yU6lB$G6rdaK|vPZpMk1S0|
zlf;KLwK>9^DUslWcNDepU1~&#dLZN2Tkm&7CnmCU!gp5?N|MA>JGU%dyb9)1nIL-^
zCWL~mHki;R$624j-_$jShliA9$zri!Hk+|nEW89DbkIrDl<n<pHa9mJ4%R4=f+ES#
zB)|_IX?{ek<rl8))plGFXsrUb)ntg{9So|f>OMDS$WEuzPD8hvWDyG!M;)Kmx?{nF
z5Uj1OdF}FKOcVRDt{ekSc1)eGd~ni&<1eai#d6U}fTn3ZOQ&M(O~-gVVKNyZq(irw
zG|SP(QP&N<Uaw;vSM8)8%8%2W^`(A}gQ<G~+)?)b<18jv9C_?Mk%fDX=_5X0O79@o
zO+%V0Uj6*%_?e&iY3|&)&Gj4Cn9t`-r*pQpw%FR*=6m1!J~waP=E~Ebp;|PQRn;+(
z)@qVOvAwfJmZc~qsOyTIturA($8b0rc}8#NH0$f@*h%e)oy2we%~&fkIz>&>pqwDB
z79TrK@Q|P`Hh9jqHo5eTU*N4@|DQ=F^G-zT0yB_UXl{+T{yV?NGs89PsdFS*f|LGc
z6iG@NM7~zT8$bCW|Ms{37610P|1GU<kP?b4))Ml4SEbUdtq+*b7u>vai{~y6S?dq6
ziNo{_)=BmbMtpev8k6}9jbQ84C9*_eU03xCBEu5{ts-hhmYG|!W8hqun(hJ}WRGNU
zEsJw1rC3`ZQY4b58KZSz_=4ziwfC(T!`klg$JMnZkzW5}ks-U+p;E9kIID58#R?rZ
zR7E0F3@y$Xl(3lK`8!7dQ9&FYvRxG!LL|7Z!M87*IUZG>O$$3~n)I>8kY*C+DiY;u
zRf$l)X)<Y<Ei}{lg6U$;mRkcSF-9Xv5FT11oX;wB&V-ED;E^5atrbG%;eU~jT|$@O
z5ylPJt-kBXOx`n|J_u(J!Xm|^>rl4OJcm${#w@t{;g3OFMW%+fsS(l(;vL4{BO-tw
z=NES=6A}bg8s?QINfPR+p(<;f6F&Q`_1a8dF31B?ui?^NVQpA0u&dgsF0=UrX0RK{
zW59OF`YnGGLQ2*VMA3d^Z78taVZQJgN`^+~(qWuI;;>R6R2Y9j*Fuc$x?Ik0WTil0
z9uo=SA+QLoz(NmL7fYN3hHl9fuP~%>P?qE1oJCt5*foPo6I7ac&9H&0X{gGQD$60y
zeV^8#S1ZesG)ujn7`o!JQc0}zEMZ*9;_<|R6~brpM`kF#Fs?_-Ov5l64z#tnI?QnZ
zOij~JEkcKw!$Y6_89+}=SQzyB6q)b8y0)?I!Q61@L&Z+e&TDLKna>|xQx}grQAwBG
zx|;nOlP2z`8;kn5sObedMlVg%F7!H^%@9I#pFf|^sjIrngpKXj(k$y(z4?6J5dksm
z8-Zt<hGaD%j+{r_16Chf>V$nZytB}@2B;BPLYnx`uIrM+!$VALMOO51T6aa12ZO;8
zKoj+bpQbtI(6Qbeg!^IT-;s%Y6c?<Eyo7M{BYc)I1}Xf0xi(zm{Dt$B<$@bGud%<s
zM_t!sX+e^teD}NG<0n7<3FppTL`cDOI;E*=^331Q^ZA@!;kBN5o}rCT8d_gp57(BV
zs%wfo@0^}SFMK31^@%*w=@btTqrAbPo0dXKC_jblR{^1wq-6NQi=6w}8;pMI_Yl@A
zP;7APop9tx4u14H!xx_A?3I_046rg*M)#c=wAS3ZvCA7jd4q3#`<ocsq7*#+%w;Z}
zJI9ZH_?p*fSWDAtq)NE-^s`+1;5~lw=4+hUK1-Hm7+rJw&MkiY=1(}BPD%5EVr_>;
zhuTFBv@;m%kc!Y;cxC-v3Fml|7-KPEQk+|AA(op{SPrEx*#VGb85b{J;O1Mebv^&A
z4apKAZsV-Qgasj9P|m4)>FQ5n(h|FDIU@G)4TzD>_Zt)$5<^oPTGIxveTBV=?U3;A
zU3R&2EP0Hjme7ymGK8>PL>zSNP@u`3&w!rID-Op~X0wtyln0P1bRzK|vIJ+20B#74
z!s5{`mx$%RKlvcu-<xwww(b#au(plL6zk({vZ9AfQiN(z%7AjH#-O#o93w!I%b-Ij
z2o%oIT3YP|-jN{@-BRZ@Um_&w2<bn^IdNn|hb18S*eo&UKJs<<hI3<EH`zT7_tuy@
z{@jvDhz^DbgoVNi>%sqOm3$FO;wn(;(HqA5u~UKL4V$u)M>x*N*sa_ubQf>%+M_JP
z28PZ$-$JShg8WovI2^h*lx0PcSENaT6utz8N)lS#`s6LKlqX2xJ3;tL0T8L{7<TuE
zv8*W2fe7Qe!o*AMpI-ot^{kvVAyK3CK&}i-+<ZFg+F|v2JvKKs*jV52<sGuDb83YU
zUA;VOEoE8KRt=`ql#`NccXuCsevW~ZWF@g9hEP|vYpd6L%!I9Lzvjg3)}ko7>r!jY
z#>NIolF+s-v)QZzuW6dnG<DY!ELKB^A?W#h?g3R8k5~@HT0^hsqjjvwcAO!<&$<;#
zHyLBm0T9|``M$cYL$`u9>{I?ZV+={M3?ctXTGq?D2#T@B15W27Epb0xqO#@ABR|67
zhPYS9b9WwyNY^>5OX<eaJ@3Z2TNjuAkLIwXNrpDwmAiB27EN1{Whu9A-6Bm>PVM-v
z4sXBx9v9AE<hf_BFgYx#>IG?*B1}r2dpu*Tm&f$_J?8Tgu&fP-0o*&ZZjnl{w!Y5V
z#s=f*0btQciot+`(ILiq2U19177i(JN-=rBl0Xj!>H-te8ck9ZT>izca`V0SnY@1+
zgha<-6V7i`3(NKIe4i^{`$c*i+eqPSFUdrqki7riRqo!}<Mnr5qcIJGexFO1PxG~}
z{W-q%t#2_eYqaqyqwCj3ynp>RKX>jT7oK{Gs~=wF4}bh!c6WD~jz(B#>8%aOhdpwY
z@a6~C_~8Hb>-^=v_$z$ol^2nruAFaq<I4ubqNtV-3Jy$DYX#PTv7TA;IhVpOq}UJ&
z^XMrNik+=3{_fxTTNJ$^JEwN|NB{8uK-p-VHaOv5>zu>3hPJHHZHsF)R%;)=P4b7C
zhWp#`tic#Iih^iFs1zTn<l7jYJkC>ft`N)mQ!&Y8Dek@RVH^VuM+Ef<s;Xu*I%G1L
zdA6ylNYWlzmckMIwDe||A&FJcxMb`68X=G0vn${GV0IL=gzq~cobXAk$O=@Z0>hAX
zn~+F+`?glrN5?g;j=~Xp55{{ihMW8b)nbhax+~@DTLrK!@S4ePzE>E62Mu4u*PJA{
zzF!vVvd4#KrdQhlt!}idY(@l_$9ufq8ydcM!*<R+78W9cNauXv@viF!tEGiu_4*Sb
z92Ta^gDJNJbt>@=j<0oAlO_pCmWFdTG)+a_)C_E&G|x%1*aY4Gw5_kJmUWSZ==?0p
zKnPqYZ=$-^cz#SY(hKeUb&SBuC9rVTqT80Hu4x;8Jx#`A4i69gv8E}5{($XML$<ay
zDMDB;i-{d8fXO<G)xOJxZd%H+q$+D>`x7PyV{Yso)pvce`MGz#<Fy<+N@%USUa+gj
z7=K4VC#CEdIvp6Us;W>*(eL+Hk{^5swrLw*=vjykOlqx}hSw>j$nxySeUb$hCw_k<
z3H;oyfVvgGdUX*q+9%mr>DS#@u-`dHS(fbY@1u2#5Q)#8b{<p|dBMiIFOOkXIyHRS
zwtZ@oJS{NB(zK0FiWJZ`EoD_P8qKiQg^d2FFZZ%6Va-ZX+Hvr8^xsmsFl<4C?Lv99
z0EU22cpqXx9&y6wfG0ha6f||k-Me?Fss*iX$+DD0rQE%Hm(gg<xpU|F>}Nj*Bvj=H
zDFth51FG84?-wMAqOM9*;*)W-_K}7x^A4O4f~INF#?b5c+1T7*cfUt#Gz!c5#x{F<
z6CWKwpb{UxO|q1_w(Qymo4<!FKJ3E@ZQGI-1@hDu&;Qa_`N2Q_4a8*84VZL_R2mX%
znSFSN*_}IVJbi^ERFSsYP%Wms_4<$KW!p>-4>@;wi!&Ea^I!bkzr#QK=l`7DyCbym
zi&~y4?(R?d-~M<1SAOjqU*&}to@KB$VCT}aoVxrR)58O*a*l3W7Ui7%QOO^E@5j7*
z?OS|!^*Vq3fA!b+b6@!>vfK+1m&pywf|V9P2#hven8O&~Nx|2y6Ep^Lkp(ezI9I6{
z4%c}0*=O0=KFe@xn_vAae}g~%;diOp8Wpr<#`?Iu7gg#mTg-1@(6o=IFXHMJ8#h)V
ze3=qw9oEE9mRK$*-hE$&kz5CpJgyPP!*xqm=zg2JzgCWJ-0>Kl6-*`zrqel9)lk<B
z{eHdzzG5i%*bN?k9edn+ajTwM+;Er25i2Ggy|Gtr61Nf<=oXyFHa;E`NFP!bUKb-p
zD6`Nd$w)fw-qM*}&Zz(qx?@;;5!1RZY1%5VS8>i(%9=Q@PxIZv#FBxEM<nD&6P(Tg
z%gXQ_rQH2qIltAv!-xy#+tEeG;v`^q;#kFNapfo4@#KsL`Md`;<Gd4R5NP3Ot)q2b
z)?mUpS(ZRb1qL+|b^bAja-cjxPJ%?p1Z`T{*05MC5Kd8KIVu5TTdI1FwnGw^(pm$F
zLOP%6XpKb(g>(w(5-<X%{e6f8C;g_6u%Sr0uN5Ypm%_-_A*{n0hasUxu&5$i)lx1>
zpWrZ=Fq<vt^)ois*ExIPH0vAd42NsvMT*9Fr&vlFVY`}@BCvZBFxpU;6>VKJo6Z>R
zA21ouSxhVDW%Usw|6nGBFYDv5s6?Vt<)uBs(dY(c6h%>x=BaOgC>(j7)9RM0uBcm|
zG$*BGI2?8$5a|A1EEWji`)|gYc1_cGz##?e8yg@*r>TqjyY=;T=JWZoE?)qIsY+Os
zzI+KzdR?R?6uOsP<mkwn7nv+)$rDAdSNO*A5{&a<xmIfq$A`>jGd4H25Y7cwS24(P
zHrF>uRf4q*0w{_ei}JyAS)bI_*47ye*3hlSIzv(9^m@K`?qo9I{r9g@E-LD}_4;-n
zXZGR7`>eMS%o6q?FEKg#0vR0o80Cn2n@<|F%dMna`J%~?ooq0bW^d2eqs#J)^XJc#
z=LOTrg3XN`KKI!#FdS@f<HjuzlHuBr+3b+*?JcBKv~2@;9ec0Wqbh5%ET?TXS(<g)
zeCK?^On)%!qySk_kYzd3$t|B0V`Jpgqy5^nH0gsyA|49Vg}n}AW5Z&}`U}r+;io^t
z<lldH<-+vrKqA1E4F~VN&&4l(1t)!3kHuoa-mUl8+xw7<=hnD<@dCq5&ENXFf1ByF
z<@(h-Ear94gh+Css7=eeH}CR)|MlNMNl=L-O%!=vu(mej)b=(C!PT3)T)nl&WWK;y
z%fJ53f63u^pYiyw^DDpn3-pI6R)b7@Ej-U`_@c*NnB72AW3=}$Ob~t|B>lD1Y;2uo
zwdR@Vnlk(1h1=U_`NpsQrx@Vw{w_P4n`jd}cI7wKD7^L8qrkY9x-1`NDo!@Bf&{``
zIY)@+OJC6*0bx-~lgx{+ZA(l}h|258#vNhmtnF$Se^hqliC`}NEhJK=EM_(1!x`nG
zp(<O<`ZDVfmioC)$K@i>Tg?pZj(`rz9YwV(uZRyt_;-RQ*zx~MePtKyeGE=+1j4=@
zUC4@-5-a`Y9S^jd#&}jTB-h|G=46~3V?&LtrVIDG5cZMbMK;Wxh@Py7y<nH4@KJT2
zTpSyY&i8W8m-Bn{WCg+oz`eo>eq0#$_;;-StcI6PdhO$Ai{zZd2^n;BU<7FA0iDB_
zBZ+Kv#L7~5X@CmzBAti9%7L<w3W-t)!b+Of(6*X7X?<NRELu0%g`-~%5otnWHEp9A
z<bB^RQ3mOO6JT7><%R{n(<|BTn$<Xr1)&|Sw$NB;9ZoxDwPsSa7$KNUC)~YzhsmTw
zDoelL<Km^uY;3NvwY^1u=-ZfirDxFc;Wdv^2+u|uAGWn^>jmeFlKFH>vnZKOXN*UO
z%%&BNg!PS0xIKEL`B6%SP2Pv8qwnT$A&JCC6pRfmDy5*&4U5GBYc2hLpFHt3<htU{
z!MToEug>|*+A^%k5u^dMp&Lb6Rm?+uxYg@2IxkV=o~9{9ub`}};EaH<5-(+0vgaq=
z4?=j5=JAP7+{1;rz-#fQvl&h8+m=Zod`7)>j7R&}Mw6w!4y(}y-P8!6SM(^W@`0ll
z4`b)gpT`(S+q5_={eD4F6kVdp*|TRDjYb@fCNxb8jsSGvqr6KYm(dD1A)<1mzNvrA
z|LI8?>wHVU`1}eN8rVD^#cG<CG;J9U*C_G<`}+sX<|Uth_9f1oKF4S@!5HZGdH{^a
z`)qG-lO##d-?v!PhK%ZzG|hb^DoK3OWXw8G64r)mzMdc+a9eAcPG>$8EnFyV;_vOM
zZD6zaKy&bLOr7&yfN&N>=qH-@dG0Ge&FwcopuIT`dJ~66k_km>Vehp!X#diTv{xWh
zPJb=s?(P9qSu&Z9ID7USFP%NZvrj+Ezy8<13D&Z{)<f$S-TLB{+WM{)34T$d5eU&@
zCnZ;J9nhE$G1?F#4bd8-`Sy2z-)q{0<*)qZUm?$OtkWn3LP?xZND@yA2j!Mjgql(b
z+Cko1XXorCy8fx5K(A7gM5UqUYD?lIFI;&U^L5RC_`PqkdoZG(`yQ->hRj0ZiyXGP
zYhTk;^M{&>k2VpQr&S?$Xk`?7?fRzbexnc}!;-jJA8EVgR)Adq2r@i=0*Q5eak?kN
z5fCnAI;|LuCd}t0Rp~Q0O~~SmbI^Uib1OHj;3!&)xs=QD09}U(!4i}`;K&_ieIh^Q
zj$Txs(5!AjNSdmmZEFzTV-rSVTufMiPBIpY)nVyE%Ql2z@Cd?-q|H)T8(3wbl;87B
z2!95U9K)Nd^K??Vo|EjP?6Z@=y$Gn(OSq1X{c%8dJcN3`?>qio_j}FbG`I5OFQ(<D
zB3)?j7w#SrJZdFe%vTP;Pj)&uXA?~D)LY$B3BLeqZD_P*G?{p<rI6H3MK90EVzFOq
zFl|FCd;wmo{6=Oqh&BNJaJg9RHCVbeG_7G)d4N(^Emc)9pU#;tN+vU}_uAPR^6YcZ
zbMC@f2E!pqnxYa9m^x+wP{ASago@WlnYQ(Su5MY(7tF^KpAkEm`Xaxrrq|0^+t^`v
z`V74Jy+@i8CB5D-Ns}&2AIo_hPfn0hb~WhQ)|Z^f(u_nUEEWr@syY&_Su7Szr_(N@
z*B1`X84ia&dgBaL<DKzdujfm<gphC4>8*yNV@SBji>_QsCxsC)JZ?V1$?^SwjWH}1
zB}t;l5`~P3XAa(e`)&3P_SxFrLCLgRqhEjhb>4X6b*?=3Jny~#9y2}jT050c)sN1Y
z-p`~;{2WO|t9=xvsw&EI&SWwTbpcznZZXDr{h{;Mc)T|{>sU8m1l$EC&Pz%~*u(rf
z<BSO;#F8<0;htS}e!7qDoaa>yrbBB-U2C+qyz$0cOs8``_t{r@>BY|i33a9E6$700
z32rJ`W<EP-Fxs=$UXI{MLn)xHH*bi}Br2iEbENc+MUmyqCUfs>s06I>2qQ4Yjn&Nc
zgU!K1YU=1NP9MVaP2!U?r+MZpKgHYs;@^V_0!AZ96YtUOzw;i`Ygbu2a~74P)O$C%
z`qmpD;GK7V!lg^kv$H)wkTaWAy!z_r_~8%#h?_U>pj$_tBq*t<+lE?4U4cI)Yb}lT
z0TCeuGSaEx#>w-Xwr%;&AN>J4J6jC;Lw@yFevu?e@F?AN<uGDvn<zS$O5sFGq7u%Y
zzl=&0MjyTKgz)w0d}#?yS(R+`3SM~OGt_m>@BZ#@(UdbU%2iqSo<u26QHSD}kEhCa
zEJ%8Efmqf{T5fvQdI9c{Yu$zINQ80{%Xp_}>W)z3o+3UjkoPbDsOIWEBeGgF%%&yf
zqVYNd?Ta-_R2MpY5F91o9w}jDQdk{{RXxH9AmYRjUI3Gi3oA5D1=>+oi*7MjO5vnL
zm{{)MSY!AOn3@oC@ZNjC?S5mh#?M7y8`3Nb1wa+FzKe-9)|b;^8NOU)6+Q~ob=i6G
z->U#iEP1|FQ0KFz1FL#0RC@gH`0v%<Coy&L^H<lP)v#{7&L4{%VX2qlG0$atz$V2q
zlrF-!NW`*mv|yzToB~p;aA<8pW@(E^Q-o4j<FiA{YTftkbQaw-WQEUUwp!EL2H`Zy
zCZH@<g78!#&VXwj&IsDp&>BOjHIuU9a6V-;nJ}JD(WOReM^Q*FT{^{;E6;QB;(4;r
z@t`%ngN$v%+z5Q9AA#2hholOl4Z3Y<n+DUgv`x!=JYh1KP?aV1qQp6$`FHy4dA3iT
zK@NtGJ}+^+!`PdUDGaN%E?w8=D>8Cj@`G+kleByNYRfTW4CC>*1B4qJ8#w2dnX--!
zyv1@TF=H0!0A2(MQM*@`WmoL@`1Ku+clCFC-Z|mRJpsZwj4)^%3Wb#ZKG@yeV>Vl$
z!Z~DFMv`Xq`$G;64!FCw+cA!5l6G|rKP|g{{T4f0TYhe=L2Iw8^W|WC@7nIZP!WZB
z2PKL^JPW%_Hu3Ci1nZtmItq$IiJuT=m2x#Q5c~cv_Y*1Ok3@&Mkj}ZTk&-c%wy~s1
zkDvb1=lIKi`9EQ6>l8nJ?RAWSqUbZ798#8ZHa9mYdR|6>Gae7A6e$v{aVV8A35iA^
z==F-FBO#<;mr?Y3eMX~EkbEWJVy3h}>(;Z+$%DbQhXQq;-AeEqGav+6l43=|na{q&
z%^&;-{oaB1E`k7334xd_xcQTxaQVwWN3Iv_{o%jjC*S;c^qzg5x8A+MQ<q<4F>Sbh
z<2pb0b6@4oom*VHah2V@eNT-i3fpKO4sjt&7_J9|>|*8-xi8Jn<l1n+Y&zo)zWe)J
zIDd)Lr?&Z-pLx|U)}g!zLWJ15Kq#MuhEr$*JEzVdl%kx@Q8Ixvq3h?0SjlE>na#%x
z*Ef(d;l-Ce%WOX5yWjb3s=6Ulxi5+0=vbv#)%>xT2mw16FpQH(k%SAy6155T*kU4t
zLJIHs3VZaqNQqnP(5^*LBa}d?ysK`%Dr$d_@S1KNS=whhS#amh9+xkl_FA4^M!%PP
zl(sS?A_9j{BIEc;cip8}0Sv@VnQ)<myL!prp9t>6>hEJA#z(c;;RvT-IzFUrD+rmz
zS(Z~7N0MYz)dFLD!BH2&mnsR4rgf+UQXz2K-;Z^<0P8T?SMwhX)))@g5W-@NMo3BB
zcoDpkDmXJfgVPBS%Ad%NG494jELpwPF>=mzrICdYB;guZWyxFsk7=el-OTEBv6>XI
z3hq`Pi=J)#z50A)XjYwvk9H&;+{Cf##dftUj|QP58<&XK>(U`bs6#2~gy5bLh~u1)
z`r11f?dy-V#)s&XB(xivG3@t{%4_YMQ(?37>q~1j)+&ti8cN|LMnhXWs>;&ThDFt~
zs2V1VIpf8gS-GIp4aUN-NVsr$ou@8c<nrZ<tgmkXj=C-}RwI)X8RF(XOV$sFa1LQC
zrqwiyvP&wc$`W0-l#2z6xn~-r5EMnh`uaL&&z@y{bCay-Q$9fZw~w=v9E0PJMW7c!
zexCbsD-g=U=q0lUbZx?#rs+zeD5V$-28_q!uCVat)+S}`^@p+QzYv062*pL97o9bw
z6jfERSRB={odoFaXV!s|kfsSjdOfFf4y97^qM)iOX7d?1n3oHzb@Wl3I(3TS+K{*2
zdxw6nkF}P%uF*lp@!+wGhchWvC<o))UmCk?Nf{kh-*PL#xwg9x;_5Aew5n^7m-+T@
zy<XliuU_-+t4Fl0rfm#u-F9o(k*sg$5W+jX#?Uc){yEF))|+QJXU?4A^5v&EckTjT
z`N~)67i-Mt6>q=wF3t*0?VP1;TkhVu$!tF1<(HqQC{o`eJvbA+UXE!E2M1%Ml1wLK
z7K;V_{(!Z$wV<c=`=-&FEKTY62TUh(Ok=@$&Aq2qELOKDDcIRL{Xlc@aG=gQva}Z{
zLjZ}A2}(k-xy^-FU*W@Be}L1m2!;<wB-V2H*4reDL)H)8W_0Z}K6`qd-@Eo936iTH
zyv1|!DfV~o^3=00@$}PIs9VcF`)9w-csybE_AbT-4T2NPMRx^6boq@(Zg4QBWv#zK
zo+gaPV}AJBAG5Q$!_!Y);MC5B-`j=53GD%cRG|;0!WqAq6~%yRUh)3xZ*gjSo6QU7
zXj+Xn{wD%Lty|uE?+uy;`okgX>+5{xGe5<6yhqz;bf!Bj5t*8(>v(LTBN1tw<wflm
zFXcn#4xCC*D#4o6H{y0H3`d~8VsZsxs5z4V=RGTjWtnhsucvvkDUu|^0Znb0O-ttU
zinh^#Vs}A6ygZ;tyRUAl(fnq66i^T}c^=e7%I^Llzxyw&_p+XkV$<meCoI#+h{b$?
zj@f&5*?c?#IUzl8ur7xB{AWn*z4plGxxm6%=h+37sxC9pm{0^&#H57?uDrf(_1uom
zVF+NjC18(3Bv$67qYJwS<7fKcG2FXq$FG%D_Tr@L=YCnU$L*u%@!>2w1F<l<2eQkn
zPXtFICP^qf+pkv6K1wXW>lzh4bSP5a16zv_zH|=h99f!Bx4xKZ>=`YrFNtJ?Lu=nO
z+_QDiHU?`2+6v0XGA|9&N>kRF>7r(~SWwMNn%bfitPK-3wtAdDx5JfZFS4;Q3|gnk
zJL^e8o@KNlTTw|t5<;*>TbwbNs-az!EM_xi(-~#CpsH$g>w8}!=r*X0^)+_3ci29)
zgHX!Xz<Uf#sx`I)`&#?s?w*UuZNPLo?aGw^9WaYMVdLwXrs;|e*LB^6a^rOw`%}ll
z#G}!u>l3TBW-^&@aB$Fta#sOe=ae{0mgk*Av+6*7)Vd!-f>L_MYI$ASBd^)t-}goR
zts_bN+IQ*lC4T<re}QY)Z*Vw1?2ZE<Ns~{l|LbtDj&2*Q)ijN!)eX+YHf$wYx6J2N
zh|YLYB@!~$MY)jd<|@{F^TnZ;8fzh4w{B|f(2YLg=*3XHwU+(;ed@ZVs=Ul+71)7i
z?TVt{{P_!f?z6A*>gRuor=EI-#iAk0dW^?&gix%l^(o7$a{`>RY;LXl$U)cFH=$P)
z)MZ7IWJr-vRTWKBGaPQv@B2Q%{@5M5HKa*KmgO9dW&zCkl4?>qa28`)Ql%K?pMs`N
z)|#Zhq?jWO>O&BcwdbGZ^!2-pfB23sO<^4g;H1FZd<XS|e~V&|OQ+WOe6QlIJ1tj!
z{ulYZZ-0l);Si)_*k9w|V8YW+y~uz1SN<0N{9pWIZr!-m3CkfeH_MIx=q7#apq;11
zx3{<0Re}#be4B&MyvldK^8<e6SN^;gMMvf@=*=v?wUrZ=-Mw4P<}>m%W4O6aZ)1(a
z(LNtuzs4&szf6`bi(UHgYXidTWZJex%7m@WvouXj+b#=N##%>l(}=b19>u)6XEl82
z9~B~mG@ZkR1xG1`RSMt7HE8O*-1`LK{E`QA&ck&ddbJ4T5r$+{&vL&7OUevm9aU8`
zKAbWcPpRt7ZZqqYjh-$$c9JXU`EI4P7<Aw+*3WY*&z;1|eH>8mpSaW%KC12Ay@3^m
z{k_}FW>cKC9orc}lE0r;g!o|$lnak(kIjV$pwAkNS=P-;Q@;QTrD(OJ*0qP=#tV%N
z&TB~tz|M8w%yORHvaaIFn2An^A_8DJTDqh=j{t`F7~<^Xz}hMc7dOIByCVJxN4aEx
z2@G7A6VKH7t|tVqda17qj);h929`wnmWom;6qZ&xnoyHU%7i4g@NzoTvvR0LpoOHW
z{HD~lEw%wYx9C<dZCVa%&3@f*SQ{4F(zITXojbvLPjPm=`2R8YUqO~-dwL)CTWf81
zT>8rL)%N!7zH>e9I9?3Kf&>Nt2jEB~5fUC4fl!3_0ER^AK?;eXU<AOB2|iFSQXr_H
zL<|*z$N@MM+z2x;F&NCuwR25-Q&rt%(jU9sa(Y;6pK~(1s?Dvs6jn!eW#&0~&OY0%
zZ+-vo|9xEF+2->0Cc~{Bi44tdh>$f2iq`1hyA|_YJr->Z##x%WLDvm#(ZmLbpqJ~C
z6M`lvV;C2*{lS3Y#)#o?7*jSJQi<gn@yU-yD8m_J$o%<ITzAfNY;c|jyQsEp0qFPp
z=bqDnfX)I%Yt3*tTxL^u{(NU=XJ{J|D~h5xS5{pI5#4)r=X3C~Q|B0@ZUNT;z;(IM
z`sg%y0Igk_(Skr5gVY+O5<a+hpV?wgy$B5I+3AF{@q|3j`RbqeDkmo=c<&jH$4nwH
z{c#QN71nf8kQ8eTWnD9yEtt<|NGZ_TM9;p}UBmk@`&i=xQ@p1C&GTYewj>^VYaLEl
zqzw17RHaJ-V+95SySKNu$HRvYIXO9@EGx8L?}m-Oa=jk?UY{gM!@M+^F`3L64fXQ4
zgwAG*^OoVT&oj^L$EL>x8=Ir}n&Ft`dBN7!766mU1Z!6wkGPNkqNDB4_VzYSSyRrc
zK*k73B0Y_+frj~f!q!%gdOG<;$KX?ey7AlJXa5UdB~NpVtwWvH8qzci9UEE33t#_f
ze(=Eo?qrGwG9cf|`1(&h%N{2rUSPA7mv%?|^yCcNFL?IZt9<*9zQNbN{x#nDVa4Xv
z8@zMpK41FcSGjWKD(}2=8&`&lQsPO{1bm46vaysY=vbTclY{qy$#lZz)+W#0y2Zi4
z0rwu>VRN+2crs&Sb0hfKKw$o&C|yXgEHRy(ol#d6vw6+uKmU0iJh;PjJb{;9CQEbh
zo|BVPzWv>AP?jxkyzwp@o4cX)-oto&!dE{30)rx7f;ww`=n~R~x;8ZJQ@G<@xG8k8
zT3u~EfN-}2;ZiChr@tx_#R*IJ^Bk3r>>QoW2#5C~nnpgFX?!#gif!gzqFc*s7GhYM
zsztY9k`O8=msTIE>xR9^n{~%~otay`_c>c1zcTVzf{5kzJs#AZ=K-H=Ja}*yCk#hN
z4=JlT9wJlgkU}9vH!-a3d=a6Ohzy;KQJ>xRbrVE<tt`t(<IOA_?*QC8uPPzjvg8`{
zpivGWMO-z<Nprakt9{y?MO}=|`d(TGhU-`1#dF|X8Fx(&xBl40W{&?p!Q6R(+wsxy
z91~q<t03uMv1b|arvx$b&n;OIPhjIjTxH1UBPSEWi#Z?4o>GwKy&y2x1}PO~U5Bu7
zBz0>jBd{|Klv9|Oj<ZF@Y1?vUEDH;*56r`;kX+gvaBF9So0|i!jD~FYazw6ZRcO?h
z&1WPk2Hhu$BvGVE$iOrvc-EEWg86(wTh{o>VCrUR%Os@?B%NhYTwAwAgS)#+aCeuW
z!JXiQpdq+7?hvH$;I4t-?$EfqYY6V{_Rjs@ABv(1s(|X=`>ZwR7~=?f1CBiAJS&Lu
z*fI<r$U*SHk)Xhq2%l}PYvZ&K>RWh&O<CblU0KJI<!_iGG|hwdt1|iDt@ey6!>KYI
zPGr7rrzHYv9ZPS+x91ExLQIdRR+W@RJBn)ahY(Cf(FdVaLjc-OgGdmb#6*q}<r2q0
zJxhLLtpPIjK1F-mJ%3j$tNpjt3uEkms*H|~0eTB1cx3k9t@_)65kaWjd)nsl2;O`^
z1D;|{OD<-~Uf+H?cjmS^5W)bncEdTU-UY#WbQiV!FKC#0$@ySfWm?bdcsfB1_%8;o
z!#t7zjZYF39bvQ$siP@0_Ug*)5~7$qTe|lG3b%eVT0s=QH6->-qf07n2DTZmAA@Sk
zTm0XlB1YsB)62@zCavl#28hG2le5-Y-S5L-$#NnS;-&&pi-{r>7_Z1@52G=}npL|t
zORR|_jcY{c2d!}2J?@q^R{ou3Jwqjp(4_MiI3^cYd|xGYz`es+nf75)yC3j>H9+O}
zh*RIjIg<~pltqqiC|qQ}eqn#-@$f(Pfbfev&uIJY5p<30?Lu!!8q_#vZ!O?|T`GO(
z(AoTlB7Th*gI>HQF0d21+Mg=i4mzfrfOX2NaZl4@&DQ~QTgn$aE$OQYWRFq>oMy3B
zer+3hD`Kv{4bd_73hg%qY>hsS_vBSCefV4Dh|9lYx>?*CF;l9HsWg!kLfShU=iB}F
zfv^aFil3~$O-u|r(amR{z0eFmYLGSQd&ay);}}yn`>`;CeOV9@NFn1_unsh94{qb2
zI-LpEz`~CXV<(@iU!_rhV5osrCqr5J<+zwu5d#-x@>!%c`q3#}$O`b(qKwp95)1Dh
zf$EYZsoO~B8n!xO0n}rr>ngPlZSBwbFDLpbvNi78(E(>`ld)uEQja@LT?GZ*bKpTL
zKiR1o)jx~IwN2#;o@CaH0Er;?V<Lgd>cl;R9&u?H2?gRbo1V_T=md!)B;3yW=MO!}
zYBt~T$9e02I@*J~*1P`}kQOASR*qdKJG%QS{6$Ba&Ya9T)c>gyw-&)}Gs+Pr8^_vc
zKUelq#7LZ2f_HTeMdaHBi=aH&->c0cvoIRvBB#`=LZpA27zBjih9*Vw$YTkV)fR6a
z&yFcZb5r6~goa{`{ZP~sjIqp~P5Y+tB`E+N73u;bDrB_HH@m|tV*k#gd2N!AHrp^j
zo1FEZ_u$#_6a(g65pSM3zOxYkb<iB?ht?#qEtM~#8P^CapnCN)*#S%x*5phAM)wVz
z@Mf$7=?sws6#A}j481Lw1$5M!^V*_9#g%4V3$t|f3!}|dWo4Lb2Y-cFt*0#E0-85W
zrY%<8KQwyBwpIe;M2JZ>kI>mh35;8o@#OPFx}=CR)@)!GMppPG`O1-6hwi^mSVF79
zl-AK(6YIh+8(j|sh1L_^=beMX=k4K`L67<7#%w?o)YRKM13(nBj;80CH-8xhDQqwL
zzi#fm3O9M4&Qo07tvYaJu<4F5m*4K_1!6jS`lB%7%^oI|A&&#N^P5O=PxTO%9Tqf7
zBc=;%t6*)6je(gzl=4A$>XqKFZo7KZv<IKE32g1{w+w&h&Nsy5imJG5u)lshE(5p}
zj;^jFE|Z;~RvUpw!GUefax@o>bOqT0yDO2RvJp!4YC3m<@PV{ed%1+@?d|aB$pO^x
zk6s<dO0?G2t6{<(XKFi0t-Jsp$g7`2!1p^23uZe;y7qd&<YnvY|DUnL#|{^-UDr)5
zj}9-+Y7QgjcvuCvJwn9m&bK9RuxC3z3F-u^d@k|j&(u}^MnfHU>(xTyY*u}<Tdw1D
z-XD~`Lalsmp!D*W4$wS7&wSqeJnWqhdO+dH?acS$L&0eR9foYY#0P6r(dYaR4Vx4f
z%w1PeJila+B94jfnt(GB2OH|X>vHm8O(nARiA7eL*vAeUsR{IX^$W=6V&QV&*nuM}
z(~(bXZvbo{KmdCL9I12}1!#OvBqHx)+@uuidtYCoHmv}LN2==My!b1#$u;e#17`~`
zr=neVV(buSU*hAmMK20j=upGtQ>s@Q(1)>yc0YFejbZovIk=U>j!r(t4sSO<@(^3I
z%o(MEUrsA1k$kwUe#SSVDB)F-i1+RJn_81;dh%q)2O2T-YEjmfZC9^If|y$UzlKAC
z4GRMqHVE!t&WV^aneCDfgdnI!%k?lk*2x;uvfq};h~EgO51q=Rtf84Nd~0kxZ_^fU
z%3`tHs<`Xd?+GN{`q5X5+%Gr>|NiSSaa-#bH24`+ic`(BC^j5t^WFaT?XL_bUC6%c
zXl`z4VvQ1o)^92a*`L4m20de4(T^!1GG8TQFQ|7w&=Fb;NIfX!*YztwBfnYeSEdP5
z=N)<M1T%R^lM}3Qx-+4jVPUoo<(pdschWlS>`(2SENr@SKhP?1Fv!L?!%_#=+r!MT
zCKK!i9lG`a8j3QeOr)q-<F3J6@&D8o4<I~2CV5PzY_m&aQXUXnInpm_3+rn6X;r^i
zO@)dj>l#z@*0A)LisO10GMq&yF+{P_6!!3+!^1Q`mr{s3mev5qL*25iwN5SiN^62`
zdhzJu+1W0cfZ^AGW)!9WHe%!j!f1oLFFhmMoC`Dcd5U#q`YU@Os~%^$B)!7tog_;%
z`)T7=A|gU2Jf!i(j6goSP@ME?gt_Vh@T6TlAos_L0nn3SMRvbrcAq!Jwa!<z&F2k|
z$AzXY2m}H%!?LwEp5b#>m-CHhW#{4GXI&jq;&4(p0s<9Vl@7-NhFn7TL@7<R3lt4P
z7@N6XICY|idrPx6lMG`ZE*YD=yq3vRSTI`oEz2Hl-7=}!N^@+ACdq<VG@7wf-Et5B
zZpq%~Q3fc61!c^UBKt?T%(ybxsge}!JL$y(f{MvZWlXw=MB9QQB0WNb0AGFd>l+j0
zRO{}|PDtXm86Uswm=$fQ0tCm5g7ty&Pw79yt&ZMh)#Bh6t=x&FbOis4&m3cfuXC!2
zWh^`j452#M5hZpCky%(@-!-eMb;$n}NeM)xr3^{Llp16%Ad<d9;;IZUl_)m8)znR%
zu5J$e2-dndX&`C8KC)T!x%BiqOnK*de`tD#+AZ(X1&?Ul-LY)}<@x)K&pOmQ229lR
znEH^*n5GS{jEqHbscA)qNE7SYwx-?}B^MFjoaOU3yFTuml6xHSsHGK@c$1oeU87vK
z_6878)f{YpoUkFyBA+@wy)f`U=RdrcopbBG_kVzo@*K`^=;&SSe3B`9B#NlLK|tED
zTAJvdLy1OWhOr^t$Z{C<uf-DDmAp^aJK7HUDI{2BTJk+8?ptWxj{Zxe6H);;rEwW~
zGAdflCwemqbN)qy57ouTILF_A=(S7H0l?Dl4f$pGz_$I`Q_2W#vO!G<tqAU1+px6N
zf)P8}d=-1Ff<R-EQ(W;&)w2`aoxuCh=aa>f;G-2UeD~ma8ZOuHeuiDzwy8kDbF(yU
zg?3c%C+rAqo83jZ2&6nfzL8GgM2+IkOV&nU|C5^7vGPWkfIelnq(e=^-_ZJXJGCZ8
z6m8I@KgNB%(B6}!&CJzUq^904CNVwfhqnE`Puh*TGLiVz&y&pStb<(PoR6tE(6lrW
zy1WXASR+*v@bF5Qu%`=x2@k{*m1j^riWfmz3N+ZX7VUg6%$mILcC^<k9E`kmzz2hv
zK~@CrUfseZa%^lqJXGx-cWB|3e`#PUep#41Z@2;_YV0U{a&v+3Bvnjfwn;%m4SX}y
zT^|)a;JhM8U5G19Dkm_70QWh`Yl-B<!o!t@eX}${P>bGvoCa(Lm<H*J^jDlnsB5x?
z9uL>YW7@{bzLmg^v+UN1Y{7lNyUU~8>m!Dd%Nx`Opd!j5mu>Nv26<eK7u?1#FP%;V
zEK8S<QoCZ7gE4SWG{Vjf_>#X;G|1_H9da#r_Ur#Q7@B1z=plz?`(>nd@8%|{pciK^
zDgnF&pLhsMA*{TAJw?VGV$xg#xjkM?LHS!ES|~?|#y>Qhd0Jt|X4KjTeA)Bjf_wmI
z@$xONk*ed0Hu3@ZpFQ<pL{IhHhArSn1_y@&E8XABV@LrLxo#;XDzLM}R%8kzYIjOK
z9V&7X`Cs;>6Bx$yEy_v$SW}N7QK{YwwEA@WaH6ne%|^?L!uLWIsIw>87sLow*Vw%*
z4!AIVC!|>OCrtM)-v+2mfxgU&mWhnk=K6oiET>j;%f2WJw%RfLiCp%<$M@!P?e$N%
zJjAALrzvH<e|zL_Zf#I~+nvml#+F_Q;YzjTO`-Q#wu+g=8<D3`o7}(oB~63WY9BwK
zf%^3OGKt+DWM2Ge+zNSth;F2%M5|7M_V&2j-^M98mRA0LeLl=7;={KhofLV8cIdq2
z1BAdb334+^PLCtBPp?76_T7+I1cVy|n9iqNaU?Mdj_=dsuv9cMEBoIs<II2t0MF~<
zdw|yqRDaHkVsn<2?icm7BU6`+%eVDa(j1YUiD!?N`jUsL_l$RBMjYVH!Vk+Vm|ttQ
zKF>uaBykK&dWvo%B>9gJSw}xxHLWUF`!SVvR2-{h5~B6tEk8postv3%nX8Y2CC$k6
zOhBxfT1bRUUNs%zI-ARb2v(n`G<7gWWfK**&d(to8pxRpD3g0Scd6nRI&K0kBMxTF
zLQ*bl5P72f=&+NFo?!Yy!aWZYOnAQ_8f?ltULqbiaWQr17WrFutOU&b_{WWKE%@^`
zrS7w0d_o+G-hWszzb=gyF)xzBO4TLf1wPxXI%fBiG~9~?@c%L?2WMc?`ET{to-l^$
zJ+kCsQp@~=HFspkD5(H6kO{D(U@eBE52v=(Qy11D2T&63d;&dH7BrQoI#j!!dy%{!
z_H_8ADVC+M*mKn;&`f%^F9%GWwmfRI>u2rV-xb)Q)6RWOh_oZIO)>`zdit0oBUB^V
zROp#1cN6Iqx{;^c{wD5PAwMg~lHi2WV3FgcwP11*_%<U(2e}DlMv2it>)PJTVNcWb
zQR68B@-DNNjLRhOH^Uz>biCnpk@6P0^=d+a>JWRH#B9I(r3o6H$fVKU=;MOS4oOZ0
zku=w(?<J;qpd3ouDF-(3KPZXntH46LP7O%OJW>+w-yyu3%vW<FKX->Gf#Wzn#!0fG
z>4=8BEFYw#7G9>OAipHC8#Zv7%On4Eaf<pgEF6b=jD~82R1c$VP{0}!8_PJF@IP3l
zqMe~SF7W!Y2)A8r9Ca?DtnTkHc6cPuB3Lg|x4!jQ(3_z#IyxnX+hEJ~INl8Rq<<lG
zu!b5F$k+3n3Khh7eooP2VXCizd=QY2gJwHDcX4*EjXo3-6Ws;$Ac&vt1_O@vt`op{
z5$ZaCJ|KntD`2yPk)DE$9R$M4s#7Z;g0fDHD!+C?9L<6uCUV(UWBc4^n_a+R*VN*D
zk1n6@IN`!4R=~0^V1?C_>-LjCQ91~dDOtCG&md&`E3Rj~K4aRXVmX?kBrUC^JK1pH
zItu{`pjDa_xhC!Kli)_@j3&wWSZTaWFCTm!a7{+iW*R6pQB`OSgg!6+zDW&iBQ5Ax
z8nq1og+Kk#@@&qpWAd}Mhj|3)3}hQ1t^Dul$;++^4Z77$;-&&8!S9^kj|@-7)asJk
zx}KkH4VOQ5xCfV&T?{11Emf#4v7uRG2dSB7n^s5_Aw~7#Jf(QZUqxR&M31+tPLnv_
zdplh1PBLz^@caY4)17x+3!Qh~*~L)!b^o%NtZCNKXiKkp*-^c+@;_Z%T5bcLY3?81
zUp%^onb~vV3gLiA3kC&8&bXQeSB4G<A?h(`S~fO6$K}GL=j|@8HyCGte$bFuyU_2a
zTg=-D<|DZ4^V4Hac*fUmB+tr}`bve3j#oYe1PsvG@_tMEJN4IxmlWM3+OF5tuJI`L
z_HN)%ue+HzWO%lB6Ub6oCbcV9P_rreU0Huq*xS}!Y$-ZEAUkk(_j7#Qs9nC4c{np{
zBs5cRKgjK8(pIwqMR(c_6Nfh&2NIQ%7>8|7uvQR@a=hfX6}2CdG_Y1@Lz##Xb?~K<
zPvJ?}J|_$VFcS3GeEKsB)u|+9Gir<J8<aHhl#q`f*=K7aj|<Ea{wc>DeB9tv&fX1K
zDN1=-I;q)N>y?$(m}U($zJEUR&4E8~Z-k2fTc9190R3+~@AwGMh^5;(kUYTULCjr8
zA-TLB6G<MwvVdl|ph@>2SGn+0KRXp1I>D8=S>{ab8k^9>aj);>*0-8QL2Bssd2*pa
z0oP}8)Ao(X!^pn-tMh;&+u1>i)_oQ{Z^^$!82ho+)glR(h6z|n-mxI2T(&!5_O#Z@
zm(|k}yw&^=X{@S`=Ft<tFJ!amft|MdD^$LzsOM4F9ejs6G0T&#XPu^%o*25fr8hji
z#_eGu4GlozKmS7`QK(5n=JQTC1(tD1T%>Z?7S+Z1b|R8{addxa#EciWgyKO2Lf=*!
zi@=Gt>4k+d>>paFF}&(o`sD+4xX4|w%*7`{H#RVZh<%<5CClRS191hgkmuhR16n6E
zT{{j<)@Q3>T(Uk&4EmI(91U)BTIhwnj4z?B?zT_tUt25|CM7i5rHoUq9PPS{)|%Jo
z!s4SQx3->*w&E!^kGP|6Zx>SRQyf<YYg^>Efs#-q-zC_$)2y;-N4<a+n1A#_I&hY^
z*C{gNP{Edzl+ftm(};_sr=vIf9b!)qr5Eq*?1Y44P;lj_{^LaIE@yZGz!Mpf0P^}K
zoM*++q8fn3{LDY7Yik?4p6nU|tbCKRjc6F;umSfA1`bZ=8y!>4xdw%D`C`lL8Rkg}
z$wd!5XzKY{);0H37wpZj=89dyv&0uvifx8-xQ|w;s;cV9_J&=OUkVXLc1>lcM=_bu
z808ke8R-t)dci%Y{m7cIA|)j=sgR(H-2dtWg-K|`@JC1&z2VfElQUc8U>AnCU|0@-
z97gq%7NGTR#AQv>CFQVCyaB`P1#fuCe{rN6;(t5DU+Ah{mhB1^E@ZifP<2!>F0DuE
z%Q>S2m-pa%6@}vzkK~ozK$7P(Z1M@F4YZ>);*b5u&5e8PMnV3rph*8a?y(KuUzgKT
z6DqO8VHt?p7tdYKs-XRO9nVLUoc+D1oUJqxoffB8rmv4Hs<X$p|7n-xwIsFK9~PS4
zexCc^$9AhSUS;;X4jfl?%3MB*jvPiG_VRn$I}ofkN~}duvkvlXK6n$_VKs>nEHR*L
z)UUPOi+P-_7pD-s{L6W<;Pa@gZKxaOe?P!r<bvpQJ!zl$I#TS5xD^Nr2<6(5tbEa2
zMEh?IpK@Q7UyDl*9Xt=!?iM^xPy9nRo5cOj%do9;v`8_95WUG-q2GM%ukFFwRRsj+
zQ_4>_k?DRlQBy=BT<Hi@{a@22KQ)Mym({c(o2rO-cX&*Z5;lb>11;;K<K85S)?~TJ
zbPH~?Tv-#TD*IaV_`zQi2!0^(wI)SfZ*My&9Mdk@`+Icx*Kb_Jvpb(T<e1O3(qXAb
zZLtuLNCa3IxkYUEJ>AzFev~zSnrNgN1;P)bbV!WYlF<~3cEG&a@T3`COi==-9$Io8
z_*1=e)H&9iJTcg<X`sy2#MQ*hAiCLe#l&&Eka7#B>}pYM!#28S_7zB2@$YDw9xc$q
z`JH+wLP@}`7;V^0`MG;d+G_^G_O-D6(Jrq6+-ljdg&vr<BC-IAs~7YIPL01Vnm|p6
zgIXqv^qUFHDCV#`(=rLZA#ssyEEhb9;Nd6J5KChl=dHuxB<USF_RDxD>5*M4S57eW
z@d3eWx9;xpOeJG{AY&867K2e}Sk8o^!$q;8E(vO^;!PVw4+CSJBg@;fO%1BEVPsm@
z(SbirM{I?=RuCo>ISJC1?VKCp<rHt2oo2<4vhZ-QYMAS2yk`zJqw3QzSX|cEvkC6*
z62{2CU+=dmBB8il!k%K)5wm%we)D-90DmI3wNIdlq&}}{YYX;BzBKQ7EAj>66^Vy5
zeAOcTJHJpqS`!~wDXj=z_oue%yAgR)oE;8)FIQEM_*5*aB}PeMn_v#dG^|$Bi$^6t
zlhl5|O=y3WmLfLD{#v_xaS_MA0Se~d^a)15AfK9})$m+tM7Nx-d(?-Cv=Sf<v$r=2
zzS-%B27}SASj_mudgUd|SoNsWR8=KaxRn_k6a$;W8tpO36IcWs>IeQ>2$W#w#ewj(
zLtVmZjW0IVJwPB31wfHnZL_s?sA*{%2Z?N-*p3~VN}lp?#q%skrsr}BOivx87bGH#
z(UsgpMW1fr`*KCzC6YC%!X1?ZTa^0iTv`gBYi;}2KE*<IfAm<>O%ivm2xfy2n216l
z5s2`pi9?3Sv3izrJo%GnY-8CW#^*x45nGWwf9Jlhe)dt~0!*C3DJ`N^o{o2M1^rT)
zmm=IJbseb}<KE}j?0pW2whk`0X;z$eqb3}=b_SJ(T@z;$@Zz@(Rn@#>rDX|;1vqCL
zj{!C*Iabc^Q1=_&zYLPbHU*`+u1G!o1Dw$}=Xjz7OS)d0mL8+tHP;r5FOD0+f)VaH
zEP8X^iu|)|QW{(Q9di~t?^)jtFy0uuZcw`VM%lZ9G$1^fR4s1DGM%rd4i{(M=hPNG
z6*m0P^FO3nSl!YiaxL`S&&421E6FEcTl&BT-iDp6y}9DBU#GN?*HtiQsHHhRHf)#~
z@kdnFYp&1Hw$AhN1t1WuOnfGRY2@P@sw20v6A%$8(s%tp{rw3xQW_7m!f+K$TddpT
z*m>Q8f>X1Q{%8yFF`k)2lii1e@T<pb|J6JcXG`XUwz)rG8?*P(k~hx+dnIg{4$nq)
zPleiXl$VDTY^KqqV?JAyx;5m*r-e{%#ID<V8dGlf9d&$-9PYJBRGhN6HL++7i*)m6
z1-0Y;RqsA?dM<qKFq_A`Wbid!JVZIJYBEKelqf?X?v4WEl6}0c50eT@M(8r-Fq<M^
z%rn#ew)n8UWVavmfiB5R(=mR%VDtn$Nf5uRNj|N|tgE_?y!XpWLudeOJ_G*;n;DOK
z96ImVK(?E9Yx$J^#KHW;md+LxM^;KEnXwuoAzp1AD31#ZKdeYZD1$Y6)bmf<<cRxM
zoF$`Pa$=D@Dnaarbx%@0=`S)I1g2Af5tKlKE$|EWHkrZc#X`u*HoYj@#LN-?F&UG*
zR2si>FqY3;V+FtL&;iz>5eysGcX{Fjcgs4x2Ln6FD~0H~bftlMI}W1CBV?*>VlXas
zF_rleVoWsJdeh7Z;qv~3Oe~R&SYd$y#W8lKeV)I)$lk&>>CLmC4%^q=ERT)d0dNL@
z#cynMr<Tg81NOsFIiB&$J_FNRSio7{ez%Y@=ho5Q)ARV?T84$gZShuZ&IRP3g$L5Z
zSkwy@wwwspEp8PeH-Lr?nfskP5qcOP1byltm#K=qheWZ_>`D*t{%8b<E;+u37J#^n
zFqWQhmV@{Xn8_C|o=-QjA|N2NKcB6xwRyyDZ-3mY(^Ep0|3FcnppEAv*->I=Q1(M5
zy%;DQa1-um-JYynsx!I%L;kQoe1+OxWcTCBScRP(^zau37*?~|cX^Yn6NjjwB76)B
zrqb4Pa(3=<)yOx9y|C^(q5AOY7@tvOaPWR=9Cw$b@$*M_{%{7Py2Ra_143Zmm4X<d
zL2jG7`KK;@^?g%(C(7IV#!GPWFn!B01eZ}4rbF3q^)D9S5{nF~-&Bho#1$6)O<lwr
zM#D~M&pgtP%bj!0JZZdJ)RDRQvc<WXq~ocZoT;SRtjU&o9CoY+QK{-~<i(&vA#e92
zNwX-_!xu^Lf-@;)0#BoVBlCX5Kf4LQIO~+MMl<$w7VkSE{)1D<z0_Ky$MSmueY)z+
zB{Js)<vZ-g4<*gZmK5W?nFTgrr`OxobZ)XrG%4<lwnYZKr&5NR>rXsFFpPJfZmoWg
zgf(sYWi&Rt%;h`~Zx5%IA@QR}AQnY6*Zw&=nl%!QC1m%1gP#=j4`tSABC@}!mQ`aT
zZJqyaFk4SML6V?l+97?6$BZ5AW0uA{m)--`mv&&VlGo^WTr%_a4mPfe<<2fHwLWHC
z_|eLD^`(i*bq$U+Un|Z68IK;lBA8bRgmE#U_66^HIjHPMuuA^uuTYet8C5>=rmNTA
zo8nh)*~t&&nDWqKJKrlqra1K|-33_ngyemUqq8FA?1#M~<AenXV<Zv1vB5T%fmF5l
zW~Un8vuGs3_wp$>2yrr)z7+rdM&P4c{whDXeYe#=5OwAeYdiOo_Bs|ig@EGN&rc|K
z!mQ(ycc?pdu%uiUtZU27jd?YKx-rXkcP3`)0M*c?3S}X+wu$S5((#G$)GqCBCF-Ps
zzf;8cV5`=PXy)POVWOrOfWrT;=lGbCaXBf0xr4lIq<TL5wA11bJK;_obD$$MBRBT1
ztvH_6eqQu=P&)y`3}u=lbtXlR=8a1d(*<m@B2y?{U)-%@Cmc~|lzKRoy@HXPx^C=d
z?L;0J(va7#&5p0?=h`ll0UwL9df;jZdg7&Hb>*_RwWTa7utSrX{#=S%0RY~c2|^Qm
zr@V+)!+!JFFfuDABdlA`&e24mrc2nwYs3{HwIZ>h=cO@ogjd)DV+T{Bed_`W{Urw$
z#n$6rX>P}fZk9f7e&Ve`9}yRC#xt6}6Oa3J{sk`QB>eW2B8BojR^r!0oiDrW(-i1I
zXDsrm<CZTI;*UmeTMqAhQT@~#+P&{N(D5S~^i9i20FoIg>XFE89q{Z$ASZy!SUB}R
zh8E`R+M;UkJri{pOJ)yjoANHCeTsu!hn~>aCrL-1m{1WA6Dlgl2VuM#$I>u5ssySo
zuW$T&nm6#dMERP>LIB)Qp@6kpaRq>d9X9XD67~QcgBO4A)eX@wHS&EnY;g`9A6HHS
z#>2i_Yv^FxAQRSKpzs^WSFT?P!^4c{9gpbeUEV2P#%2wGhCM1n$00t>>HJJeMU|eA
zHTRFLXB0N{A*|i(6iYFwD2QS)xp2hgR3oQj99F#Ia6hLU`?B<t5gTs;D$cDM`sB{3
z_u5;B_>PtTKccyRgc-<~XgzSM1P&sVcAQ|k&7<kQsVz$}sTeO4-US2m^qQK6!JEH`
z>B&}MNbeguY*g<LeD8+djg{+#<$IZ9#hGLKg;9beC04(s)uUzC+P#9;vq;7U%xu>`
z#%C*e_xS{6l5wZ0G>c|&6I}Ff{S-B*>4^WP9Lz7Mti4s#truLIe)9m8&q{R+6GV{h
zA5ZaevFfv((G?3P*6bhKwCPK5OlFla$m!3nG=U-Z5a73A=HQ)?^Zqo+$*!oPqMFKz
zzt;91u=)P#FF#-I3|)k|-#+3uciE8hM7rp3!bMn&2$kqGY-;T6<<;|W5?xl_8}c!z
zlI|cFmr6;nSOKvH{z51@&jy4@{R;VQ;^G3kFnwefT@yWd)XgWv(FnI6KQc3l`zj%;
zzVofY22QyFnp;6W+OG7NuQi7KcbS;%y_4RYVvVVk7+SM1!UnlIogqam3`foxnYh>Q
z!tiQjY3!^!yAvVT(Z<_(u&Cc~Wg6WP4o={AQ3WQiS1e%Cw7lFRoxBU8Xk(AqrBXEs
znPaM7^KN3n@@;N!HJuf+pZCKmta$Gv2_g(S^0h_O^B+GGIl=WXa^iM&i-mGHKKGrL
z+_er6LZ}He4*7YRnzpuE$*@%bj*wyVY`P$Ny|TjF&sY10OC2DxB~_4u$H$ZWXd=Ab
zUc#9zW0pFf?YUhVZLVR%gu*J}jgYM+$&OXnGouRE(YNguQAwMBkXS~FX$}M+V>!_4
z?WT*t7XOC$T$3?M_5JLB8ZF5OY3+lZ;aJR0F-;G>sweK)h$wW~n93Uj?5gdeVwQia
zaz!tEa{}efe8aArJ9o!%)CqDy$B&BS;Z@x0U>Thn!xYz?<zWFX_PxDcDIj3XRgqW9
z%cZ_AWV2XRau$hrpKkxB_x{gN$Mvs}+e@d#^@Pj2Dds1`jjoa5X)L={XAo$JpzA@V
zOlElB*fTBXZ8*oAgYa)h$pY2l9Y82EcX3Hv_L&fW%i4srR>q$muI=tcaqiw=T8qdO
z%V}dNUJ%>yOw7?}#91)D8cr+b<siwvMvJh$h`L*_D0RD4*juNk173mYX{eGCoEa^r
z!x=H}w&!SB15R?RPg9K&Nm&-6K+C5Hgev9v_V)TLzzhPtLyb>F8^<RWBN-SN*xK2#
zPg*hdiZZ<@X2RGVbn*sXFhq*U!~<i+v5czDiB*TraaWPv2J%dEv+D7;3B02gOA#^I
zQfQxM_zJ!NdS%#FYX*9piI}r*e~3nSQLm$4!Klm8>YdU)lFCOV&~EF>b6$bBfyt^B
zg^ybXCx)kqJMf~50_<{;(e*jh*z&7CGC7BMBrJ$pqsz7C;gJml6IiImxvu;u5J=a>
zEyVsj&OS@Q`;CdWJjcn>tgUTG>=M;AT5@IN{Yl??h#u2NkD*AIOZR2Wlk;>?zcewP
zCX@;`k@7U2aL$fL@jLpW@WYa!K|a}XirALU+7iV%UgLmRRIuA&vzufT(((LqCjN5H
zu@ZOZ{%J>PfSX+8CE&*w=iL!ECC!46?v><waZpO}yoBarJ+K;uDoN!0OkbizH!A?C
zP>odeQ}4K$VQqAOmOK1?oIX3txLwDIaYo_f-inx1B_~OcjVBjR9U8HrU+uaP8BldR
zf?Bzk+*o1lCY7$h$A{c*qdv^GKwK}D&&0cdM4C<$_{Q<g%42BajPBJqhA-VYPf*ml
z)CMP=G2Z~BX|q>1m8He{BYhxlSFHKwW3uVCsy_9@6!`k%hxI{lRQXJ-5Fej=Je$NA
znq}jN*~yKd5y8HK)Z%cCkIqLbc^XOm1Rh|EYwqu#v)*A8X`EA(x>9w+Cl{!dK%@L+
zfsQ~)t!<L9%bUjJ`L~9f6_u0I&t{1{LE43+6m~*#3>}H4X$ttnSkoPHH<kroNVaWr
zk)QnEcy&fBTP5n7_T1Ab#-U5I{s~(Im8WK60vdILz5Fd#5Lk`4en69-wXW7eL*KF@
z*ExIUz@(X&>`DVRRDlq<r$c*E#WC@SSts_*lt2?ur`MK$5{k7pO)}J(yQA7pBIUwU
z@@#XVUdr?#<Upa4o8j3gmcBqtWIEdGdCi|K^*XPf_}q;-!>qqPld5W&{iI%CZ;+tX
zfA2h+U}KUYS~@&b+YJh@FEF{(OtxrG3A~8mQO<|*nSwwm8kJ!N4V%}l(c6)HP){$c
zSC%q3-W3FBZQ7bu6PnXT-UPIiv7P5WGn*B$XQXS(%ilN%FBV7~Sa<{UZT_>b>2zM<
z@$vBq2nk6l$m7pO3$-PSKi*bd5Bk3tp~tEw(;k*^Zagk{w7BikSts;+SL*SB<+dCf
zspK`<ZAuAnR1Bl+fcq?6L51&3{^w-NKbA!YQUYm80>Z-ECnwRhEW>q~z(4Em&V88>
zd~gfD<Joxwc4H&98J*(TW~pc6@SX7F<8NyAyvFwyhI#6+9ZrQF4TUPN`wjJ?Z{zso
zpR37%m-Wc<B~n$_6O%*d8(la??C<ns>SHJ0b7Y;n${=m6YvNVVm9gX>DAExG`H)hZ
zVylu&@(nXiLGw5XHazcM)cFa4Hjfcm=vh^U=_jpRwqG^88I>d|CPh1YyI3vdoZEt(
z4Z5(c35ZRBEY5S)pLVeS^)K<Db+jW4gtqhB4ROWdPEp)Rg{NYh_d=0;>}S7Y(ZL?)
z$Ka^~j>&OKR<E`8_HYGm4T2By%Y;6a5<e8S1%99@^L*Y&iTT<l_V+qNki<ITf;IS!
z)=NfQdIvEAC&8K1%uImN+NvI-EA|uvlWqTLUX<<BXb<E4zyKF!oO{wQW<4vj(Fj2Q
zl5KcC^I0|Yf(FH3*8TBuH7cXZgmQSBh*hItkj<<d61Nx%$`n=>)Yyp6HidT%E~3k{
zKsmvxh(x~_2lzrxzJ$7{3!Dnyu_c*{k`-*%s1&#b-zVo2==XmnVH~iZ=E1|D$kh0S
z_8M<$<I?@MYj=0oL<sKrQCvw9gHEqB2)`f1pPYa-$IF^H9DTW*badp77;KNmG%9hi
zG!S%))M+5<@AG*%oT6<=KWnkur>8hFmdM+CQBGQR8i!|`3UCB`Viixpj05)UIBb`y
z`(yZ>1PXZ?fpX4`9*xUgK55h6r4uZv)<nXLt6W`;<b7F*SvDA?>ZChiD@NfFx@I61
z=dqu8Wn2qjw6lJ;*EKaaFNz%rl%HSQ3oit&$$YwJINeD^U*wF;HW8!_de6j$P;f-N
zkmj4=!x@oMl%Ijc-a4G1>Mg847)oNl)Yax9%r4CQ7|d@R`!Qbb#uA<H3Jva)AyW^c
z|ICquB=t*;2A)ysa1uo|0WL>U8vCj|v|AwRnhUY3V-N(cu}2Csoa>`?*5;CtHmZ2P
zaEmV%1kA5m!~<+gm_y_VPMID{si%h>i1(5b-xqy)zHPe;+gSU+NdC>#8*nEwtC^UJ
zLZgsR2>V%8w7}Z?LwU85r^Gs)4e~8zRbPAls%9cznietvMA67S$*U$NZ;yj8M9*gr
zG#7<E+V;IEBa}%82M7OGVg5G&CIOuTSoi&<)wVuUzTLaiRT}x8h7A6t<>fqUVyy(n
z+aqnU=jBOf*DLm%1CSx9_`|+`_jAgI#qkWS`~4kYdMG5kDTp5ATv_`WRTT%Vy@)qR
zAGYkH;w(aUj0RNrA)6fX7oS#nfH>O3fZ5(GqcI{Myl+^r*ef-?e9)j-J}yIG6X5)q
zAZ~9WtN<2;WlU*GfM&*}FrcHu=*2zu(YD=?&~^BuQ6gG7S5TW-NfNa>wvGPTI>mJ_
zMd!$$s?Grr7Zf|u4S%!KEB@pfC`N?!yFru(m7!V_aY?XeS9gK(xQ}zs;k57GtfsU<
z;_nbe11UBKbX`*NtyScB#xZ(S)i>AO#SqM95*?ozMJ=Z0noX&kY%0UC85kdwQ_~n*
z-W#0AEspkY2`2i%@sJLKU7~SUrH%7&$sFF_JX<Su{L>*8SHD+szoI^z{DM1QOz3*w
zL$Y!iJb8t|2i3<N7^`UD%EfC~OBC?>o-b90@#wrfG3z`=)EDb3sf&FxGfU~n`Y(K&
z$wAgKQe7$YQ!GL!$)O7>MLbxkDvkioLzQVm*hW1O<}7>sAGo}zly^WH1tFJH)nWJA
z8n^+=Vf`DloJ}a`<wnOC7_(w+Bnwc)<bv6f!wzm>ogfY|`emSZJsf^!iL27XK{Q9W
zN?ByVF@lL0O*T_+KuNrZ0&(Q{d-Y3{?d!6IjW-Z*Xun>nQ-lBERtJRijZN}k8+5gl
zN>Gk~06Andw!E#{XUb*ZG9HW;+ah5MYfs=X7Ry2(M(>l}d$Cl*xk9BaD4CR*jSjdT
z1Lq@*y2%3Co8h210zXsxey{>coK{vTS?i)tuTU~ol%rP%r;AWrL|v_1J)UZdzraH7
z`DTX;gY@4R*!d(`piEwPatixg{Y-^1%6H+C-N{_V&Tjo~u;?r(zzuj1Enl)FIA6p1
zRw^D%*e`9&GqV1i1kSV#w>w$8^7(QJB0v7*+KE_Lr2R_wJpG-OdzY}{O-%D8;ASh^
z0CD(({1$11<Mg<gv@uwgWT)SevDER?A2$K3B<W#jtNidn-CwM$#kA^3GQ{nRt#p-V
zKA2c1@{BNC2DM4?W?-e_`@o-QRmaJ@a_7N!tmE8V&pTvq)JL<)I2D+>%qi;MG!(Fl
zmI&>j^U7C#$+Zd8m$9TQN?Pf!4dW!x=(+c}zd1|77Eh+dMHuMd*%{(Ccxm^9{({s*
z!H@mvA9}pLF>Snj>N?(zUqpuBR4g8IW~mHE1(lTnlnuiUuU`aTeK<!1p#x`j3oP{-
zc?^;@uC7L_F2D}H!uDPJjHKi%mqEgH!a$3`cSgj_$|L2t$==!8Kc}65x|;?n$<kV(
zmd8lv`+7duj5CX^sRpJ2I5g22VXymvZzD<^PPNggs`1}t%E*YR%>+X3ZrY)q#de%u
z%*@QQFsB;PHb|~Sz6qRgXN3Ho^b@bHU#~g5bDa_l_fMn&56ntPdAvhljJ;pcHhk|z
zRa7SBW|EE<?Hx~7lmXsP95uu%!nKc(JJy3vYfvee517~XTyDV}J*H}fz{wj4HCBH8
zH)E2Bu_0&~C%@t>vG8o}c&ZsFRM_AAyRGMAVXhom+S~*6XE{7wKB~BiJU8NKdR|65
zMzx#aE7Gbjh`RC--1s)AlA+|RZNJr-KI0OV#?i>+NF#C;Eo~&XS9cA!VbE*AYp?^6
zppb%7=Cz0AoLvKlmjIFiySe)=MHdY!kEVx}d8&TXAV;F-l)J<GClJMpa}sI|_c~;3
zhNb)=lePlwc8ALJ9NPoP5MXThketG;2pt`$-c-%RF_jBstZ<~UJp6ryQHi1F0ZR)-
zBoJv>f~v4!#VK&;ODb1+lS02xjZmt|&-wQUe5<f2#~@4U@MHghnJbRS-v=nfH#YDM
zR!M-7X+573eO{d#S6Z!Zh_)lc^Mmdg+9lKMa3)|JsN@a)R$_BV=LRBed;HC{-jegk
zk!nD&B;Hs)_}Jx}Fnx|DC#{l&Pa}_50UuFUe&>l><&jvVyN8LORw5Ta!$};|?DL4+
z*4B0d5AzfZC0S}R!`!`UvCfeWsr+EfpC``dai#uzJB_**Dc`K?_ls*uw_3#Vm_?R9
zqT(HFDAp+kRauqL=HE6kFi2KMPo#p_ZeVtV7=bC@G#8Ops`%dXaJg{*4|}W?^AV5u
zn~1R_wPay_l%K1%Me4Z0o{MsUXoc$c&_X_R611j{CI(p1BEr&;>}t^B-}aB}@W4E_
zuBJsf8SfH|BiZQ8;e<}^l_0$vXyk%U(BPwQbJU8SZS2{opEQU;{jUaBBMcsCVHfd|
zu;1?FByHg~T|>bf`-Hi+WNM4O2o<M(foA%IH=tHKyM{0QatD_dcqJOB40Hf19bsWX
zQw~CD_(`}sCa;C8>};+yrR2A7M)E?FE@f<yN{l`vJYTKTsq?We`@M&ssDA)<d8>xS
z$@9<P>rWi6?Pbzjq_;$<-g0o>ZS1uzHN!q)!0OuG`EZ6U>M=7o7^d@i{Tv?ui@Yq!
z#}G{5;nPV-Y<py$8Td?Hdv{j9iOVB<LD1EP;W)YMhoBfq;vI6KhGC}<t>0;#;7V=^
zhFns<oql^A1KtkC#E(q8Y4x7`ReUQl&*Zq@-<mfCMtj5W1;s?^#+{QC<dUezHD(z2
zfmz*NLK4R$y>;ZYi4j;q#4|aMuTfGH{?Ac}MxJnM8B;x9p|zy?&B5)HToGt{oFB-j
zFM{4T5Wd?rdmR%2RSC$z0y?zJy1tj5T=NBf5Jl-{#3-vkT-FkL3<1!?hd^3IkyBF%
z8xi)kC9^par>6^49b#8jIh<G~F{0Vm!?ExvvBqo!erGDR3C<Hm_sVq>7Y8|mWUVO9
z;8U#mEux5UQwNlckc>!f{9iP-r{}Xu152_6Pcpi84PibU-M3y0xZ^vxOc4j*@0Uvp
zl%%>koRnLSr;at2!rnf2MK)u1Ia|A5pDB9%pFO6EEe2x_V=<Po60`^#tBgNSB_K)^
zNCipG^Vy<4>#m!M&`5HSqNe3!&Lq+Hg0blD9cDLuB0M(y*n{F@roCDvhEy#;-&ZO3
zvXlu($+LmiucJE*V~|zSCzC7`=2Svr(l@^HWOE*DbV_zE5$YobyQl@Ug$57>>B{qK
zkjQa74N?KvkK>Z`_TU~RG9O{+Z-+KtX7-q%aGE+BPY;2jjyvMmXCvMkz5K#Tn`k1;
z4%w7HZ-ag}D4!}wD0m;gzPuLHn+C>$gu3&rf0U#0M;0%DDDrKCl!Ln4lE0LHNQkz7
z)d<T-{G5O&lg`A)*R<5-oOXJ8n(C6a9XhyGy}{Pzp0cJh20o=rRH3hD>17>_019VM
zZ|x7SPcG!Q9<dpWU#jY$5hR;YVzKTe5NyT9X8aynRJR4-_6X10lEV+^_fuZnQlLn3
z@KF=Zz#`QX&?I^e`!LODS)oSO3$wb)>)+EC)aR4SoaSUrv~|~QC~m4i;XMgR5DsNR
z3nUn<uWFyv&26Lk4sXL0{ROtnKxr6RSrc%AQ5snim9mH!@lc0%-?m8gutqKsozQ>7
zBetb$y`;0vdd5x*1+?k9r%*4Q0;A?}x(!z0`W$nC{qL(lxg+hm&=Wr}zE`kj%mIxN
zC-2*szSL|r%OB%TA5SDpVcFky21iRnc}b=`#8gFq#tSf712+xAb@)G&HrDBMNhe9t
z2H^{Q9~7xBobu|}Y!2Asue)$4gqkS$d#I2W0ZE#1Rd_=(P=y>x=^Fd`r05^9`2gND
z@$ld$pY7Jj&djU^R#s<cYz5Ol?>GG^0yc70;|YKl@8w^}YYF_3VS)x?gagTuY~6QM
z?E?8L<ajz#vYg9>)pJf#)c*u91Ggk1-pBYUz+wv57dE@tF8Q%l!%`8WPle0*%=puk
zoX{e5VSg0Q4@HU6?!G;lM)Q&^m+v2)jLqC)y~JW0`LQ>NAl?@O<y1?h0uV><;4PEB
zj_3P_yQZ%3tt}`|^@}kV;C-iVJXc(D2-b{L0U0Rs>xNIW_5?Ww_TCqN6Z0S1f&eS_
z1QEcp$zbKOvQV3b_@pP%_C1YEIoFw{gM38K_SuY~%NR(3T)&;anI&B_P;o;0{kzWD
zp1G^*j$S?Qkg3AF35N@2ZqrI6k`%f!#>m6YM3P579Kh7nNiBmIASi^1=%&TpAZRsM
zRyXuMoW#j^cAK1Izk;E@2prW#`o;gX4RvQNX#Jdw_wej+8Sa1luB2f*%f?jZeeY@i
z*~wZr4(+qR`ah-L;o=~H-Ia<zN#jT)rU*8!<#dQ%9b)vizL;@^k#{=*YYS`sAa-|u
zQ6tO0)-fuRg$>1aJjyt1&@Gvcq%rvW1|sW=D@W=>QwYbj)b7JxG*q7icW%QTsw!WK
z<Yq<6vlEga95mjz9X>6{#$&v*Kvt@k@iPL-o@Tf`t6nR7GbUnMu3N89rwN~dS~3Jv
z>sTd;bSXZT+LFAa?TD^)P?TaKMas```{DwoP)CUlfdyZCin@pqI#e5*H6Ko1#CWvU
zx~cZ_Ru<*?uI=erb(41l#Dq`?&(~xk6c^%*NG*WW<0~=vP~qxex5@@HB%Js7)Tu{7
zpjb|Dzu8~&^o#+se|vQ(Bxkr-ZRj(3d)llUWAyf_`8}z6;#&a^B^HE6P4{y;;*4(4
z>5{=dsSq2V&yr(ilWszcrHnx=9~)0B)(%}oBLj^w!$_+|<9BObM6PJ{`h{S%_%v(z
zkirxuWH{f$<&zQ+eY2}|_kS)6p$WREgDBeNjkUr9ys=TZ1^v554`{+NgGg#Ab9q!s
z4+aW4QZuCwg6p33@FKd<eliKg=1fRrJ)EM00QQw`U<vLE2Gj&$JzTs%2qI^&s!&-i
z%_z18m0CpMBJ;%tE4N7<i|dL<gqY0MN5b1rlB)}AE!^nZX%^+0h$c-7wOw74dwX;i
z|4C_dS@1`1kZfQ*)&IiraFVkqYsB$E2R*RWB(3TpgnBPVw!kE}&c>3E3Ga*6d>dzh
zi~~VVfXX!&g}qM-u!2OM$La)ZhHI-$NKoP1f!+s05s*4pOkv|pQ~4B$coU|3EPmDb
zWIf=*U+LoID9r{dG|nkz__2ZSpsXJ`!k*=e@KNpV)+<csCBmjO5~N}D*|5VMh@*Li
zb#%mWmlu%RTSEm)7X28WDwEy{>0RPW2a(vK2t;rbv(n1)=RwMaOR^-8<gg{;?+aig
zs!wbAv<jmyGqYoeP@iyRV#6`e9}C(xO>0tT&oMkN`?LP;>*7-4Y1_rk{C8`cg`I6g
zv{R2~=@E8q$SoZtC0(~vSLuEEfp={!vo3)AcVySMmgQjbxR#ciTm!4VI#^Yf#uunl
zj^#Y^r>*&=)HR*uW1@N|WIT(eFH<c<3wmEbWlXN;iwMxR)FP}nUrcxxdO{AaEE3)*
zRYZ_$Gq<;q7C&bwl0kE=rPdism=QPLRn-Xy7mEl5oQit&VhK#yh1%nTs2KJ&xwN_S
zW!BrI+OD|}qKLjKqAuRwAL8|@@g)OfsK>vrACrhpUrCEJ+V2Sq;^R3lL6+BW^cWb;
zEVvn~!nz0+sAO+Co)68$$r?afL3njFGg|*hgUXOY0Cj1e`j^zd++!cS^*$hyY;p=)
zQ%@K4rS|I$FmA}*WS2`wlE!053)ju#G2=M+W)!6XKxLw2q*e)ww<8OB8B6H#><TU_
z%}}9ncFj~+5P?MSC1}i6s9=qlgSbBan^vC3_D6;4_5BCTKeO{(;`Lo(q_$megwlxb
zsrx=FPM`-d0U09|4k7aurZmj%AI*;$g!8y>Q71nZFDQc=%{T~a+UiCJ_TW9MDAlP@
zT(pG{1vC86ULFpZ7nE6`=0b<_Mkm@qeqYUxlAo`8#DcC{oA917r>zG)Do=Z&KD<zO
z`>loCGq{`iI$ab>8IX)d1YiPd4!THbkYqBtE5xdCR&qd<Fx^;qS&{%u?<jWRYYh8g
z{YS84cA3+6m;w260s-6}=EC5<$8JZep+cIJIHb=^WTgRxB?tG{OP2_Bb%NpHgJb(B
zl}26I`QWP%YDI}=3npN}6Ie1Nb)$`N;`S%Sot#=<(6z{4y4t9gu9?Ui3v3SVPhV<S
z*Dge&{lO^d5{3(H#X=susA}wY8uQBad6hWRO|GT(3sN_m7Dg>g6beZ)$_9^e0ak^6
z_A_Kwg)ObKQo<C_dldAHJd&;h;=+#R&0fJD<3}RyvGvGaFK*YDIUYrNCuiPV#wmKq
zKxAUzzVjVsrEz!iwM9I9FwbrDa+>qJ^O+5ULTm>RYLK&cyl{D(Z|tr#*ch&~^|c$C
zL|dAf1sk(tkQJyxeyb?a`|S1;q)zN-s+1^=!WD5A(xhtr)0Wg;JSdaZkHw8N+(kuz
znNXP&5EK0057dwY=GD#4x5zqxI@iq1%yTVhfpe|bRXg~@uTUFKp1;f8^~<?7sD9OX
z!flD!T7dXjz2958`LmvlWYHh%=~h~07j%qR!q^fyMnFqt;c$-aww`MqsetWc=xFsy
zz9GC)R-CpvHALOhz+9h~u9BKb%4v|9?<*f~0iX&+Sgj3I=_r+@zi)hZ<N2woE(U^7
zqv@?6h;85>0GZE!d%W3z?<R0K_?^UWrIuHY=*B>L)9G%{$h|3yl2o!ObG=2;V3C2U
zW}XOz{oreB$e4zCa$GT-K7$H9Ey?RA3HVn@=vQ)g{p}8m>bBtYqJvo^TW|ORDz^%`
zNfk$3>_h7exTxCV>g3TL$0M?S?#XyN@DLOi$2KQD26=A%4tJH|)_1?YvpWl=KT@ex
z$OPH5P~EA;dFe_+2qe)4QEE(kT5rtJC(_@O2~nR<PEUcvUG`^wqWiN6YZ4;+5pqm?
zEYw|H<#x{I-v;B=`nk-EMA681_wLP3*FxSY?g#hYu#<n(k`!i>STqR@wmV{Qy>QYg
z22fN~_r~am|7f4AT#u$QQ}xp173MRM3|Mg&Ckg_Ny;+_@YuWq-5eWJHC55IJSUgkM
zokH7Pl)lG5a_q#%J&0On@MOK(cb#v@qO^fo@T<9!5T(e${qouh@K6W0Jt{&kuWh1G
zD4o|$4?3jw#xtG9EGI&HE(-A39iGoZiKQTbEO{00J#Fu2kPGFM*w-Fs?<b3Um|k%h
zPGMGBA1-vA^SG@)FgU(hhsP3Y5{<@UtJ2|F2iW}Sn6<;O39ujUZnF`XS}3T9iiPVg
z0YmD;YTf5pzXLI<?HR?SV4n0hWSUS{^9V}qZ0WGEzsBqz6|u|Rc)(@=rc;#|zlc^e
zn}hJK+1V{6KjmYg6ry5&zVSrZwPeKnZ6)H~)xyfqjBrSTqUGGv&*jX)#}CCPGixgd
zWcic_WmqBUI3OyH;(?XqV&5z22uk?_Lhq)T7o=o&d~I#k{REXR_&<A?Sdc)lg%*|M
zT#5?>S;3~`B*f_45A9wOZ-+UHpG-_YA2C51h>4K6Oc-1FDnF_#W5p70NG;x+BR%sl
zYj|~tKx_5y-N73eko;GPgy_NaXfFQOT$XA1E4=xLp7Z{X0Cd-~pWH~7rzp16jA_B=
z({nZwONw!R4Y|bXAAr|sY-(xgl43S7Y#2LAubrQh$G#e{h@6Jk)KOaH+o+Npo>LSM
zCUbH28L<)f5lvDDO4*dvv%rNLsE?!<4M|fcNi#^~n>*I2L2T3&lEEeeecNTdf*QZ@
zdP)Ei%q&^m$`9}~pdeOPsBUNpc7v~AT@Z+Dr!eh7Lu8bfVZibx@ZN;mk)(80G#a<I
z;yyY|br+L*2;0at(N7uMNBchj#X&m0bsJniik@V1FhC}f$;lKyn=ydURAC*(Afq=Z
zfa%8`f=>nNyny|$e2MF~e~!s-{Ucs@=_<qBKJ{!$Th)wqFX4nEO;UVqgU*8lUC*c|
zCnzQ9ZSP>bMahJdM$#?}n}ZOS;jASgLnaAnZxB|N7P9^brBbSsGlYSBIH13^!`aCJ
z);a`^-e`bJH+lbT!Sl~Q&l|>~Pbd6TzacjRgbNqk^Phcz%9KbM(0<#jLUKD+#~a6V
zJYm=ykW9}o$8S(X6XK5o75{%9J_vV5qdxolTX+Xm*`iy`<jgReRKcH)1Vx`vqOTD`
z0}kOuoRU`Al!+81sbpi=V}E~}t5<i~+8$8!6bOsfH9`dDPN^(RWG&RyoVso)7Zu~l
zjLBp{Q&}KK3LAu@7?cor8B2vE7~L?NR!k-(RcT_ngbkqEt4FXg9Rt)Io7I>1fxYzJ
zuhc+~V4v2HsOM2f_~`&U=PJ^~B4mpvXbBw}B|=BCkS@5_w<gX4tF7sPkTg(aJ^H;Z
zl5B`b3X&`hB`&8?mWVJE)~gsr<#3|QVGS?lJW*R2R<xcFsl+)_Q^d~BHP&*e!3_Vp
zHNdq5bP%J^h)GndlJfGnbPW;RYkg=JcR+BpE4rt3TF3RH^R!CmVUZzFLg8X*QpA|M
z#0PEN63}@U4T_h49GE$x&4~;5lD5@YmqbHjkFlY=(`rlIijaxw6jDf(SESzKyudq)
z6oOm|atVoaDCI-hv7g0~TEk@4GM$Fb!7Nue9A#UFpmFK2G1Vn(Us%X>+9UyFY5E}_
zNDs?}wsUS3*X$k6`oNHN{$sLZk^{K&Hf+E1G2_~{IMY(qPY&urIXJv$k|yLu9yO<-
zk)tzl?<Q!al2s&LB1w`QVke^3nyRXn+TOA(m!5SUTh|#BH%-G}FhB~yqAG)|MM}~v
zBZ*}~h-So5yY->_%%=o&7e{w3<XN_C;@IEc<I1IdUVQOI`u#o+9zJ9-pJR>X=FOW_
zRmJJ)Dc}0`x0ms4J3Bjg7xq!MY&KCCW1o0-+sg%IU7=4~&gNs%;gH_W4*Q#1T)nh|
zQwfV&GcRk(vZXF-nyR7E2Cou~x6AC@5b0&`&SQ+ln3Ba}!NI{{ScpV9Ns?z|MS)ZS
zScmd=f=a@@J~$_cWqCF1%fK{C0;ua2C*3h~&XH$%?7VGJN|Gfx9#7k7a+Omgp&!5*
z7t8&YR7sRbu;rX~c1AND<I6e988#D-mN^fn6=kEbCfqBHhcqj=di^>lXQ#}kXUMWe
zB!Z!Yyq_`L>_uYZ<Hx6KlyHt@G~(tj{vz+c`X<TVW&Ze1%QiLXppWeL@omFsZ--Hr
zD`+)sy+L(+Ky`AAkP<l>AyY*!&(N(#h|p0&?8rib$n(G?X@hSKN~H|1>@h!^khKLH
zn>$!*==TRmClE!(d+#1_GSOVy+vJ(geuW?W;Je7ln4i6wP}Ott{wC?L&*AX_{Xu`(
zaVtdE(Hl;HR4MBDf~KnI*_yL=-{oX_w44zCRb}g}(Ri2AD{{8B1{l-w%(L6*w&cOR
z30hB>O&r1x5N?1X!;3kH8J3j*ReFK)p~xfzBnpZmVK^)p3^S4(91YH=RJDjYk}O0b
zA(zI+O6qv?Y_V2{l{|oR2w@23j$#Sayc6J4%F;4Et5_^Fb#3X7Oc28w6UvoKcB^Bt
zR$9BR#MF)D&ue;?<==MuZLMSi7eU=RXneAf2#XReQfP#*1m?y>+lXM7z_PnCq975X
zjD!qX%861WDkIHC2-(Mp3=)C!HO?9E0%1~o3(gQ{0*CjZtk{Y6jdM6a8;g#h4xzew
zGZ1V#LcZmBAh+7L=TjoqO^t&FFzij=y<`Uf1@Ut(e|_$?128=I*UQ)N5PU^CO`hxO
zQ}OptIz;TQm7%#y5^{qT0woC<!Vm=WzB9q~tNj)u_q?|uox*sU)`T*pjS0W5Gv|$Z
zI~SHv5`}Vt#D#s24@^TY1*r@PEi#cLsUponvs&9~7W10LB6zQ5c|wu~J!{?8WUWAX
zL8^iotk&jSS7(-H2q9=qD8*-47W9IVN$>QT!n@FQ8bO_PA(~6dz>*|M8jh)t+9VSu
zLu)LU0P5<e0P3s_tfPvPcbW#!mL~a<iA&RzUawEn1T9rph8+%rZAB`B>}EcnFTokG
z>~ihc?9L9vdrz;|qu(on=UG)PHC=h0(-?h@K|Ze;J-_`w3ZUzzoK^F%plw^qvZUAR
zQP(va8yi$r75v;zPuah;&*jUP$+C>`c+4Mu^BauEW0X>i$79O6VltgB(=wW-c`9^;
z^M+Bspmkx{YU_&G(IL~bGmMnv!wveQEv{^AkY65>X9dPMX7dHN4-Yt<PARLB)`ip@
zPXzKV+!JKf7$ZeooO+tNp|0B1Jtea3u}BalDRMa%B#9zb2@X%5^<s{3_dFH5W))2v
zGHMf*hKMlYLdz?hb{JjLHNj-w<E?>OW7?W#F~Qe!Qd5(q0%1H&RWa3`v#P;LiHSRW
zYcODYV;kE#4jvrQ^O{_SDnXGeMuQY7En0uFDA7-6X<~zIMzOWWXa3ADa#<<r_ui&*
zJ@%fzMKw7fFH%TiaJ#cOBq~YR*t^1Fd_r}4L~bo(eMU1oL*_$-m&h~;4HHpLE2Io&
zl`;XLaZckA^hcYtT9XZWY;5e}+6J41oVMod1D<{M^E5{B`LBMFum8ng;}3u1Kj*c#
zzD3#^uK&_s;^g!W$|>?Ri}74~ZH^IW@9-F!1vF(#stVG<HST`!J@BDHq?=%b6ib8B
za3h3O+*gOHAI<&zIL2a8M@>u+)%FKHcJ?;8r9Da7qirSD&M0RF>vR;(izRy{35^mR
zUl+|ykU}9uh(2oDP+HA0i7`#orL-8+U~DL@3h7BwL7GYW!<5}iBla&>G}VO3*@7(b
zk$DQB%i6B9H835zu`Cu1i>?b+=gSBf=PVsHvR;;4FDpLDc=X>t9i}cYhknUYtr*4C
zDsjckiRC_8b<YV@^qZ4H;YBD2h6&i>y+wEdHVNSkE*QZ&zY=9j8TOHhC8EcbIK#h7
z3W~>_!xy!5>wnLZX?ybHa&fF5_f_}qlaGF{&s;C?_z;Q?C?QZXvb}!YgD<j<>$@jT
zB;JJ%*uV(*z=*owIj4<58;`MIIyqXj6jDx-I*;=~-zHT;rU+$M6<G7EP&jAMM$$Ge
zf<}iFAw?oDdI%{nrln~m(s@)bXk?8ElemgLVQH2DV4}yUlyV8^){{<zb0}vb$Q1<j
zU`W|eQh+sh;lPPtQr?97glRRVsqw~8eZua&j~c;J&7u;sln(@Nx+H4qIxn$e39^)u
z%gH*+GKRe&-g~kvqu=i@n-&Ix0fWJS>2%6uGFj^HMx#+U7KE5GqFkUW(N>YolQK99
z_>NVMjYm(~zq5Yn2mwlkZr#(<F}u5hba#jC?JbUu4*B+XzQw_Vhg{m*M|w%W-)B0V
zl4Th`@e@D6=RWs2e&=_7hd17Olm0M#)&Pc|j<@Qa*05KEs5<98PFtLIc%eZ^9^N}d
z<{3$rllA&+?(VR=x5viDCO`X=UuI)x2dy>lz4sn>-n+|uF-L0?m|=oeQ%QxF;GzRX
z7q}y$K*IY_NRDn<B#FZMP_|D~K`NtxxVIQ*!s08uQUsAIm><WK6l;BGLh=qJ0;}m<
zi?)VL2r?zey@RHvUCeP!MPW4pjgf}7shBKkW@W>vgSmj#2kF2l&AD;q277}cM|W>S
zrP(Y}whBq%G#e87SY%rxO!|qA!KW*~`T%5{l3c#^43~9-a}AqHlO_%A#M0~cB7G{C
zH+<uR7`qq^aaPlg$4uMc&y@g3fyz=y61;Drg~nN0mmsYTH&mJ8jR{wX@QBvY%L`_0
zL#qu(Cx;A23D>^*dER{3(v%CHd;S&Fum2VPAOHS8;Il8h%n$Tc&gva5UEk!nt3$SX
zp7wOgK-LI1qqb+D4bvIS7J?gTicMg8`v=Sp57^ll^3wCq^TLZS^2`g*bK~YsoYhPx
zW6n;`c<+O|eCu1^;lYCgn#Nqny8Q8v!-F&C^NJ*qfQu7D#&9@d?~<ZgNQ}-nrer>I
z7#E$b#d%SBXp#`%CC-7?o~mpak7wjXPL}1QNkWs=vB@RJ`3CPp^iC>6Sue|AFihFM
zyg{`vG>xHYD|~ytyHQF(AkzfaHPlVZd^Qh}VReJnHe}XD6XR|?%4%aBTzr&nZvF2c
zRTf#-e_ed+M;$&IV+gt~>JM*IFbQ^RU{0I^b=|}()|kg&BVp<4tImb;f%9%<dKniV
zU8!pQGj*>SR<qr9#JDRu=JUg!|Dr8JD0QsmTSf-;u{y`cjEh~d>z%Ue+PQUL_oz<U
zA7%9S_NcYR7>mH8yhq7!)I!&*#}WXW72!*GyCmv-)XurMK(bz792h6CmMCO*80|3z
zyif5YgcO&+%u4S_up}xR??MVv6&Uj*m85wB(xWYujbSlwDJzFc6e<y<sSJy&c-~k?
zWHEy-SU}LR`oKymrRev1Q1oIMDNu#ZSgh7qWAV-+TpQNeLLhuRZ{h;Q=q9v}xF9&!
zO#pLk)8b7CEl+qtbi-=#Y^<kkb@UMnO&)okqhxxn!#jxTgI=_|rsj(Uy<U&)?QL9S
z?CPo_NJm<9EWtV-O63T_Xf&d#7U-xYY_vu<O@I|5fcLg_OTZ6dzLar+{Uq7m>z7NG
zg;a|yO_rv-p|q^1>WVwJ?{M#f54d~hF1_A>uYUQf2qn35_b$Kpd%wqgKIio}-r)N`
z_&z`VGe5Izlo1GeML}7AGB)UO<8G^2m<2`Dr79$e5F{}xxTh4lX_(Dsl*VxT{X6uB
z1^tZ?#b}$fKj6yc%RK-5i~Q`@{seiRQ<f!n?%w6zgZmsGpD<mNXzzl3lMUK%>B7R+
zBJd&!jgKltD2eqhdVl(uEgS7k2=Sy+D5MHGOL)RDHV$uHSkMVUiU{POGtktGdI@Rf
zuw{jt&#CJQ)oP?}NeLa~Ra-MI7o1g^sr57-N&zL{B@EMom!AI&S2wmfeD^l>bk5~$
z#K1dx#xPKl+&hG6QK@BPunBzp@%fb5uTg7dJ%vwE^^~ohEpky~<_j_rRzFTkya{3D
zofz6Xhp8*lULV~oNR`CehRxv)QYH9A;+vXsk#TUc%f0!K6Vu1pnqjiw#%Rv1ttok~
zN%I_Ax5z{x5=l85Go78Xv$MnQ)$4rb*(+>b+QtdT)fZo8@7ZUu)szoje}mUQFl2*m
z{?spinf+|YKlrDwqASbR_JDu)YgbT}XFN-&T7~NM7&}m|;@|(Pf0fVv^q=H~mtJIh
zcbCy<V_C-NCOm5zPL5Cb_P4*o-}$@$E&uQz`~w~y9P$*JiaybJ{k6MfnIg*-d7jem
z_mDE7t~B03uU9Y}^_fl@#Js_J6SZB-BOE4|E?NqON6|Djlkox}1lD-cBqz@^5}}bQ
zWZ*fcalQ#1e8Q3!2|`##I>%fZ;*3DI2HiFs-W$_4;oL}*e90blz@=?9i$zIUmNZeg
z944Pmh~Ax#7pxZ^F&_8(i@@ShfbMZW|HvZ-Ij@X%&yVw;<HLYd&<VvOxa&-r!$!uh
zz<zc>tJ5{PPI&*AeI=s&K#0!XWqtpi-=0T}i{PvK3%~Yw*vu6$`>6f;QNZqT_A^f-
z`*m>Z7BgdMj}KACkSR}wLq(@H1GGRbB#lULKw=H43(ZFkhw&bb$9RDgiqcqGukb3R
zF6Tgox0beUskO&NYhxd!Cs`&?4y3U3Q^+NxMw4cqJWEhnicAwu&q_|tG)<c$g&;{R
zQq(xt;(dZPDaI72B&TXUc_!&?Y{ZI75+}Eihl&&69Q|I8vMjNtMOlY47Sl9nV{pEW
zCEJulDx3}Hmq1`@L)+BYRtIgHHaKHQ2%y$0MVj}3<0m^n5@eE~(iH1Gb={JrK1v4^
zQab9kq1TJuuSG_ZB>{|kOI~Dvq}S_%l(bEawiazIbyG7LPXYrg1U|%vV2xuspR<_H
zna&o>7Zt6sI2Sz%U8qraEN7?j4B9Ms(rndmOrl}C$`}ndXqqZmXBfkk%U8*>g4?%m
z)3%z~Y)-2UySp|rTAtZr&Ts$D@9>>J{vKDZTp>v^-g@&b%CbVnjMJuWK58>Q(U>&`
zYYjPs&T%Qpgd&fr9YP39qJRX1LtDe7S}?a7RhE=x!|Ol%KHvJIKcd&~v$eU!Gq;}M
z<(FUKr9btj7;cV0C5)$29vmKUbbQ2OJf)c~XtkkkHI+8B#$l|-34u%$Mz@%@#Tkv!
z4(UA(&%CaYDhUn_NlKEWNa;~HiYz5lk~~qd9O}S0ym3@#_i3v!P8+0gq(YENg)9;p
zV>z8II4(<$8qL%PDCUI-O)u7yd)OQHxPECHJ1IFmJ76cx*%=J*^*l&lgrqT!Bol0H
zY_hez4V--JG5B=l*QNH!p@b(zVycp&ClJ;mw4*#5lMM%`UJq|Pwk$)NH-MDrx*{+7
z^vV*DY!xX&DB8N^^kk3k96rzS&NJ+N=_-@sIeDI9h8eFtIN<F&Kj3Tox7pwJ$TVn#
z5|yx+&k){|r3FRS=jE^bBr<b&(~|W1T>tV{Nxt)kJbU9hek12sfBl!SRwKOQ>p%Y$
z?%p}%<ao;Mhb6m<nsKW*$ppW7a)!-z8OsgYx4z4_@4m^E%h$Mh^EqzadX8JSZc+4m
zY;Etbv%N=}^|^9wz@NQxgJ1dA{|(-G<8}V#-~5~Wm;cZI7bmCVxDS7P2=9;n;QOR$
z!eEfoALeXsZO|+Fq)CQu1$7;CTizLj^rWez^?>QheBCAn;2~v-^cmi^)V1O4bdEJX
z1Q{0@c{*ZbeK56EokO|-f-pNt6N+BOa3pXxiDs*sW}!Jdsc348!A4V55xa*J)-}N=
zYPz5-8|u2n7#Fk&Yb;GTrR2HE{Nm%?bJjscH$Dkdw=SN2(ji2$rU86(Bv;35tuutH
za;B%d1_f)Ni>{fWs|7AMJ-!A@xBsQzlUjvUFSA;GS2_ysvwH8xmtP<KIYsx)YEynB
zEzzTa-NnbcpPvR(H@JS4&89-@5;s1<ITO79aCk4VgfET2M9y_3wD3S|N)ZC%J<dyv
za~LaeJ_%vqCW`v4j7(i%>3lr?Qh22FWYUuhM<#6eK9QtR{2jc%Jldp@{RJs%lr&+>
zyu+B5R%=q9;=D)Yo+J&;Gx*Ty8g#RcM8*P^hhCN=oujc1ghZtod7@afZdqc)`2g&k
zqpm8-#e!y0Vw&Kqr$k5xQc95)JycQPdi^IpKGp%&lOaOIAX<~Gvvc12sJ9E{Dr+sM
zpbxF;3TFjPqw%rZwey^7nugvWXz`}gX<+c;qO#NA)pd=DCbbrSu5!>_i%(l>;Lg2A
zD9JaC#u!VUONw4^DXGZwu;6Z5jdg)BQqclLN!Z%jT9#oK^9A4h=C`?X`+ckm!1hT^
z^^JqdgBApq+gKEq+!~aL`N0ZYnjkVshQ}+3R0+Zhwl;>~MbMv(&p10e;_UdC*IxZ0
z8(W*~?q6oGF(U8x+1}aag=e2*b1<Y3Dx{tuXpN(+Tgs{lovhAcbc=4Pz^)pNwSnl!
zd!uL`t%LPWK;cMdu}z6<8+19NDd)7+4A<14TQD^Vj$9>6Fh8kk=2gR_(M%e{#KR1z
zMS@mISajHi4QbfSGj3hlraV33<oyG-vVxt#fTpY{BuJrXZG)GN-QgyeuU{wYeX{%8
zrv!CAihMOb@$`3hF_j`6f^UvV`UR@4sUI97wl@%IPBWQ?6VBSGzesRKqolwIN!@xD
zx?(=w=JyVtC;JmW#r*yW?|tKK=2e5Yg3(sW&-}tGm>XC4;alIspZs(7b`|)L1=qGM
z5HG4UM<oeoC-=xk+w`&l&wcL8yz%>gk5T4$?ajNS|LFJmnXi8q=Nw=7>Sy@c&wP>J
z`n&&(dx!6EGM@3rvf}N=Q@?AOzxFM3YuMfx@E`s4zs@iJnLp3Q<`zke#`<r5^EddX
z|Lk}8>}Nm6&;Q)7@X9Mc5pFhF!ONfj0)PE){0;u>ul_3k(f{s0<{RJo_7l<YA2;56
z?}Q{#6upcjg<d}+&oj0*w~-=4H!z#kG}eWH?N~~=NWz4hQ}Ecr2_z{3Pg5I?+ewJH
zN)?+M1$rykF<9$q>XxPs=ZCR6#()L(uju9M?rzdl4n3DhamLZX+4-0pPl#vHlB#T&
z&F0Le3+k%DnDf>d=g!Itll}5D{D@%eY2TEWL#ps)g9XClN}5aU-P+f3yUq!0)>*vJ
zae4m5ST><_tX^l|P!u6js#`r@++RYh-s*8o-J`Z~{dFAYKjM!9x-J5{k9+Quj@`?b
zQKrR|Q>IOcD@&YMX>__W&RUOaz}m4i8<?9An}rW7iP0|Tf=m|jUUdi=caEiQwKFvC
zw7PgGGDW65iSVR}q=;RQKw!Pis;q>MuO%cRrji63js;R!k}O50o;=M^X^M{?Z#Wk%
zOu{~{b+{*JW0;i-ymMqKK?q4N?*R_ow&8D;!#Gd92$N-7H@Mh{k*b75N)U>{Xh45s
z6Wi-O@$u<K(8>kV^DN6En8|^zVF%JQ_|<h0VOi7<W?4pAhOW@*bh`Xqw?OPVU^@f#
zx~@4rJq=bMu&mH@U6Sz8-il8;&dqP}9G}OS%HeRh)cke&ztLz!6Fn$TPfuy12cb2F
z*=)vx2M@S%<qAd7<Lh7lI`6#u4sX8oCcV7Rym_+tOYOm>pdlJthqY+s9m+vs8iWfA
z7%O~OgF1<{hQ7*a%QE=D5q$Uty__OVNE3<lE@Y?Tak}BnH(ulK|NXzm@q-8CD7Lq@
z+1tO&?xjnNMqA{)K1tM33IWPlq_-Gv5zgQoEKW|)#$nq4^sH{tb%kk5>iGn(8+=>g
z+nU5166u3)WLDsvpsZWw?VPe{nQ6z&z=BX>t$<c4Y?^~oq0N=xNq2L*$4)P&8qY~w
zOQ95PIY%!R$SgF@rP8yzyUCU7*BNbXQk5Ui(!r<hehnT)*3_uBCYK4$8{Bk(#G@6c
zL4gt$H=dDfCuk|qb%PM#w8Ke{OjC+vNRoLb+K?<P-?@K{A7lliqc?HF@aO;Xmq=2_
zWCH2N9*KiyFyjYzZeZ_ymai4xXM3|xZ5wPX^%~>wDo41M!|(n+JNwu0qf6{uy~V;b
z^v#&dmj`@s=K=rl<ct?Sdy8*><6U0;!5xxzfx7>D+&LKYTC1t8qc<_+NpO|_)xYrT
z{K_x>dHThWs$4LePpPVkZ++`q{PB0c$y;xKkAL#d{vKcY!cX!Sf9)^x6QBPoiAu<O
zeSYpw{%ic-|MS1e|Kfl3zvl1%gMaj6yZF8BW4)yIU~S9fbOB<<K~a)r1!<CDt)!_f
zx(zPy-o@4kaiLivYN`ST4NQP>2Jb8<$7A~aoSWD7+1k`pBZIRw_|SdU=$fW(g1dHB
zAW#UOkYzce(TGd?igFI~X-!=>bWIo?Q>9H10?+5cAhM}-Y^L}y0R1RH*Nye>KdQ9)
zI1qS_Eqj8o%-5A_xqlzEP2qRv8AKU<rGnqw3dqG5jz6t{VgTmn*~)dXcrUK*JEo&!
z5S8Lw<AL`d-Y1XR-bVte)uw*L=X^wi1&?OwE<SegxIPU~myY^uT-v3YT*22R&G8XU
zRV^7g>wRcK@qv|-#D&rFBZm%^CD_HIjis&)x(&W=L52`?qRAQ<7jcJ$<Q7O1i4q<q
z99bgCGfAp~nQGhUX!j$76_E&C{GlxGaXyqo+s5Ji0+ps9G<lMe_k$sC&_D);$C|KU
zNRtF>Oz=3H&zEMe3P_{wPdiVwr~<QV9gD>x>W?HvkyGS(EM+AW{XT=OZHnQLMtw>S
z;SnQ=WxO=aNR#ZGA#De?thJn;o(9jcwhg(hRfREz`FzgN;TipYpRKK}4<#2oc<_L_
zu7eqG?53T~X3Xbv+O`c0-36aSSe<`Ajs@Ym7i0$>%Q9H842MJd@mlV#aji8*8@z}2
z-+zC_)H}!B_wRD=-aX!a`)z*p&;BZZ{?GrLy#D&@+_`fHt<4i}tMvicN!dlqg%lPe
zAQ51^$6FhgLxEmKAW+WYrN!HpVXw#L#s)TO+XjPPY=)620Zp}_og3!UDW+{ml^`R?
zu<p#J+<E(*<@QJ!OpJ@jU?wW8Rz2bRlOhlIJ*h~P#92dKEvV*Gs>Or^OQ9fF!LL#g
z*z3bd$+UJf+E6=9<2)76bYBu056)XKb+F@-iXs)<+}z;u_6{cx9+IKh8uXYQosud^
zDlE=ekg&DY<JOJK?Cou0O+)hu^^TwNDhpw|(i(E_sOMv{9=NKet`?+TkPiFkvLqi3
zX_}U7P|#L$OxuzsX=rhml6)|tmzN}2LRD=uz4`*5|I#&5u}NDsOr|sTt}1T4u+0mf
zy}>)L9x+98<1?3d@4Xw`IDM7j&_~)iw9vP$#wQuxd&+n3AgT)Q9E0s0W{qNS@CJYC
zb3c#V-s5aiae8)y){@N)L;2Qs`2GiXxUDU%@T5}Wg<$F&X_9a{pYVVDPyad}+`Gf+
z*)g-ljK!kl_~ewfX^~1W+$eZ({~_P_=D*}ue)-Swm;arAkNwO0ASGA!FY%xKr~fC4
zJm)w6?r#O%##3hPtj<7nmLDnB>0onW1x;m<GTfBBlL(QZl*IWKq8ocBiVyDWDJd3j
zVhoV&rnO+8T2-3)w81nAZ!@ZerD<yxi;{zf2ehrFC<cs%8;nL9D4EbSEzUvP7_uxF
zNg89(dSwM*tP3RunXs6bjL#<2wM93Mwlxg;1yYKoX>JEdx|>>8ie10KN-3BAMC-!-
zb%Fom*q}!NjvsZ%IPqR=ED)?fRM&F9GUi=tj_{JUbxXSt7hQ_IbEs4Xf3P)b%oD9A
zyhkcY7Y8;N43?8$SEA}Z<3%vsY3&}j54!DJ-=9nNLY()LdmL!H`0wLNtmpRSTDNnT
z>G)JemXM8xY+l(1-Ed+x)21eKp%b>NZVT^GK1@zx)qb~{FrvN=AAIQMWzFUwM_GZh
z!6Oh<=%(+oW<~gt(xZeY!H^Qzfi&?X$|Ep1E77gS8;5WTA(99#I~$hJS!b+{@n6B)
zP9>H>ug|bILdLS3_a5tgh*5MFM9yL{XI3siLT^wo91f5OOx@tD4H_&L*ut>iB`Qtm
z_j+t?Zqe`eSIp7I2-WN1l7!Yi`7v5u;{qKmfR@pSZHxDzD|NA0FrUxkW5JfEX~OT@
z#x5Hrwzs#Jpsp*s)^)usyRJ9>D5aLy4V(vAAEn8CTBEzi1=iOtEmgWmG5~7~WmRLW
zL#hC{g9&pmk%!<Pm?Q}|Z{4ET>+y3x_jA1V+H2gtdzZtLk54T4*zJ^2Ye?+;T@_-`
z-8mq_Jt;UV1iO*8ZpegWYjec4D_6<0l=*B<Synh9NYaG5oHIL|b21sz=oVw5H=GYD
zaPNbEhIe6+A}rYH7}qb5^hvTT*po;RZPYw%+tAjUDxND<xuC2T^m-{mNHoEaST{9o
zWvMjOGNTebG->>5X)qWOKAR8yoG1s%S~6)FrWxCLk83-ZD93ZAr*lSmMj{2$JCwIb
z0lfq+Z)e=Rw#Dv7iqJE(uBfZUC))p?vQZ-3xm4TGt4~3f2&3@1$Ce9<EJYL<#%R(}
z0Tx7(AXJ9cEw(O6k_6rAaKY6xMmr-Oj(2!)sU>@N#*5n)UBkcl<0H2A6K0Ed`Kg~Z
zT)8sf@Z^j{D$3yw55_$%Z%>fQ;Iu<naJoVDdZBY`KEuu@Sly8Hhtzfe2k$U_{g0WJ
zinO=Ipua)8_Z;*4@AJX^w|Q4v+IWL)y`v>8$QxVpU;XX>H;D>iy)u?ZobzZaqIs+5
z?5yVS;gp@d6HZT$dHwYt@IU#h{{dhA%Ga<Kwvqhr|C9fO@npvD{O<35nqB`1M~Wbe
zop{0Gz)KyjGB5EyG(oH;;5g~})x`aXqmDp=2<F6|sKe5prh%q`wh3ThHf_0m`#x{I
z`7XNk6#0PRaFfw!M8B7lWC|$~jP7hq{HmJR)t=(^gz}N0sv4TQ4P~Zin3*JMsiRrf
zhJD<46yx+sSg5CUlQzzw<f>fL>Gf6;v=8gT*0p`zcgv&IfrzgB)|EcjUTgg|gSS?U
zwaGG!Hp~6^Q6u<1t;|#(rw2U0y&pY7`Aj|1r1j!>)P#AFm3>;8I+se+pkQ?A5(<xL
z8|pGNf;qD)e@pL4VmVw2KuWyucwD@$?2=V=PEamds*$0W2oRAOjS_Xg#Doqh2;Kg|
zk$6KYEt&G9Dlm2;8ZbEPLPvPC8wz~is)<HLuP*If0C!GM*A;DB<D5mv;JBZV1Tg0m
z)>#(ig7GB8*o}q*di_3mQIJZ75sGTDAd!kezfYd$2pP)xc~OKiXTMLH1r47R0@W{&
zMJzu(1tzG|w}rB36~O8;^R);IIAhU9(>8Oes-kXmh`<sN_)5uQv0$-S@ZiCN<!dKN
za?Z1|`*(f8@KNmWk4?6*HoC0ZE-9hc>ygLL?VO`&+f|l1atVUH_gLr1(hO|akMF$m
z4xfGLWxn*KFY(v@+F#?n_wMk&{g3}QPqckPKq5oBgFuAF0_%d2H(`MzMRYnaE=Ulp
zLpn!-Crt!Lhlkw1cgW_b&$X+U+1uM8Ni$TIVqMTnjs^qzgxprNA@_lD@Xq5~jcGMb
z8}JTc8bs|dUgK2~*g*k8hMeU_8>+gd(I!e{Bu03uvch45H)V{Ya~_O@nHMaar^R7o
z#3d>N6{7Tn^JFqKITb)(!FHA~$_tuV&FSGOsQ|BAv@_(=V_Vqh!Tx5(t*ayUw+eE*
zz|MoQu{Z4}ItHH()DdE4()tWJeg~&3TGgkgwn?Nyw+(hW$9suNbFy4wTZ0pVG%YaB
zQdNr(l<gdabmT%&+muDEIePzu&F5S8Z(QTbxZuvi6Ph~V8~^+re&%OirkV9PJ(zLS
z98y*ZO<j|uA#w*CPHWPlk28kM7|eWzH7&>lsdCElj9XhNX4qiu8E2;94}N&eTc>00
z&uWYUVO#_*YoFVL(UxFQv{Ej`>1Fw^MPQ7hs%FflE!KR8|I7dNKjVM!Kl~r@3qSuW
zq*8IUDEL48jsJvy```U{c;~(Mo@(1f$e!#9ab2cpD4{tQ^fts5l#hj?^FyAWAVts#
zbpS#{L#~itsj=GAHir4UMk$z#7d*Uo%7YKis4GvJHe_i<o}ZEz3H^RbUgWqs2{sI6
zD2ezkB1!nzJsa})+Ey=1uv!P5Ot2|XADYNN&4m8Aa_XbNgYWcGPvM4boyEFGWvPA`
z$aSE5^*bV)BR*Vyh1GFdzVP|?_RE(3U=-UibRQ4sx)%~3`u*eBiBD37T0eFlSKhw(
zJ{MWOjxp%~{*w$N1v1abwl+~x&{Q=|tvR@T8|zI74UdePS{JY{OW_`e!HKZY6GGyV
zG)+s>cnX;yM2Zj^DP%bJ<FS*%g`jcaQPP5MNu?)CJXsPNC<G3oTp^TD6-p?CcL98h
zU}qA1)fCn`l<YE}LpeAwwqX`jQjwr=N+KjoCl%U+_^JM&AJ{zSP-zyrO&29iRgvX6
zy+NNOP3aE?<V6uWtOa-#@~4HASS4{%;FZAHCkNeKLx3?RUIrdwLq?XR%aXE`B5KL(
z@@6gor9?C^7U!6@uGG3->b$s!xfqGJ{<|ONSl<pQCAw{5xjC36r)e-#uBwX3WJ1&E
z<>e}bT<Q(2Ekk+MXpWDM`Q|sj$(O$PWuAHF8Ghjxeu0->dicbkE>#L+qeOwoRz*1v
zE+&}7>PqOS_9RFWg0q4Xp&MHWsH=*Dg9DC_PiR_=OcnjXh+@!3CJO5v#s*nHng|M&
zkO+Y+Ji<FtLLyS<5#R}+%=*v?ZSfdqP^rhaj@q_p16C*+7o-x#2ip|uEKw$l6VP~v
zM&mjStW<bnQj+r|Kwk<*iDD!rLnRrA1ld^5CP$<~k_v~{A=a&sFc`vfS9<Jk_1GO|
z<Wxx0B27!9TZI2)aP8A+>NH^mdUk|z4T<mw+0f38FkX`g!Qy1hU}p!J7dT^ZO@q}2
zoTqk<))>4sc<mYXM<69qW(>1Uk{b&ad4bI}f9i_~^48b5b6k=z<nrZ$JGbxf?l;~E
zlfNl()-bDQ=vI>#DO+26<XMhxN|bI8ZG+K)<sr#AIcm7IzmFG++tVT6Ieb9x(q-i7
zn98~(h`BiYh41}GJn9J@T9a|Zn{V7jpwQYf8Bh4Nzwm2Fp?Kxxm-%1*hyOGFga7$|
zNc|M@5LfRW44R#f?$`0Za8v7SAUePhz0EEH48O*F5H2p!?HAz?KI(PahO(-tsyd2|
zJ*OvUoSse4TH;-TX;MrJi&+K2kfe?*OHe*Tw+`KgcorqoIQa#~a8v>=FuIF*#cZ~q
zsoTJ4h*gebH$I7##U}-Ek9+RZ7}jVsp+mUyeiOnWx-8refg8Ved@cgPbI&4>MsAhr
z;3AN7tCDD#%+~?Y#{;VF9UfJNeq6bBIT4=U>~rIzN~{-Izl-JZ^~WElo%~UU6~Tv7
zC`l$Z80=qSvu;^bHRbUs?RXYSC9%8{%%BxEZjX)09iGs6=!HV~z@C?-rl%xXn&V7M
zTWM4TY8^wD$WUfg(jvS?35Sx9DoLVNtZsLVMP%8fkfC8CZkJR+Vv&1~HlDf-rBiQh
zh&_@Kc==#7tfP@`ugJ+S?V*Gu&w~A`bC$ZUSxly!o}AGi4%yk+4b4_d#={0chrnAy
zA`&tcz@m={DgMb1&+@acz6atpwXO*Zi`k2RNAMq9lR&KBM=mBpeAN3sjdSgjjCK1S
zDP_!}HOo9xYb{M(qofRu0kc_Lyoul!s9Sb-cf&RVqtR$t#vPBx+_`gy=bwL`EX$T<
z<tG`pu3zQ&?2Ni;XxpIc6H<f)i4XH*xAYReiy1UH5ln=;DuYttGXP5yY~vl?FIdzy
z##+jzrmS0}mu#d7!z5vlB@|MTh)}mmg&<QMSpilAP-eWt2q^JP+lEEeuxJdm@iY#M
zlo%g8DLtX#$19fesgjlygY{7x?O4htdcB-MmXl~r-$4&WW(BUb7~3*R6A}T5b13f_
z<$`Nl1y{B*E^qZ1rh;4;gsH)5ywM09Ou#?M_*9_IPETn2InC)qWG^^8V=Yoi(j*~C
z6Z(K|S{$A<%Yu7!k<&Ic##v_N0@Jn#D-o5a*xY3+JLWUHzsv)UF@DY$e)@Cdsbv4=
zHGV=W&L$J?+_}y7-}(-fYuOSbdd$(L<>>GPV?9?lciFyj1B<7fPRUb+t1C=fp{+nB
zJ<g^r&O^xt|Ma~R1~1&=gE#Lmo0c)-^69O*yoYjOc<r^@)U}~3E39t$i+|zYLdleW
z>(_ss|Kh*=Tl~vE{G+G#o<atI=)6Gq;5F7w<Y8hCrnf<PW#iAGQ6kuwgd3hf#+7!U
z$wj<rdT%2y88Tnz^BGN3Q&kJRw@fBerjt3&f*=Q<<6RO!jCV0mS)+Vc5>Ak@X`m}9
zA>xe;g>wyyMagtB4O*CXL{aoUq+$7_H`t3>y6$%upR@jb7di7ZZsJwFz$QsZF%ez-
zmOx-Xv~97nw=8#erSj!=tZz^EdDe`ET}&l#Yd|jmyYRYN2fdXP>YT3bQTyys$m~%-
z>_fW5k73_FsxjqJKY!@;FZ}MQOteaY3`I^+K(@KbU|+LuET?J4sd1R;46H{H7Du2U
z>VmvqIn;OjQ0N{Pw2i0D9a&E;OWI1haJ>-#U8+2(a!4860;CF{E+oHnzJr3*G4Uu7
zu2C8P#fS6@??cp5k>{aXby-yiUAx|S+NKGBZ8oE+YI;S^a5!Ml>yxEHi`6zQWmR%A
z9wVe=b90maaENh%nQN^H@(2-%&Ip0ZGDMaxwN*ljaEp10<D=Fhk9(|J{5<~eKG}lq
zzmHNYoNtmy(=_N6rDSV!gPolnUViyyc6N7}O{bilo$;kFeTm^<z#DJAP0(kC(lkC~
zKX;(dd(YYFDfjN(V>X*{a&kiZ6w2ez-n_xH&)lRcYtF{w&|y58E%lThsEj3v1&34w
zu<2a5^g{Yaj2IPyR)+<!aex;%p+gs}lKlVLd(&9U*7UCL_YP~^!<o;ix#|wx_x9wW
zJGR@74Uw@-5D-WdqL4&Nl%OC!B+92qkrE{+-;hWY1%C|!0tsLu;sgv>#KvHUcH8c5
z)7|dA^Sw2nQ)k-K8r~@%-nI7Gb#C3;Rb6H1H0P1F>eM-Vuj$=uz5n<B{Gb1mWGMqB
z5jL{s))*FTg9B0tp{M+mjCw*D?`<byyTn!oDj}Fu3o7HW2wI{KWe^e*;=iHmA{9Eh
zi@=iL(E*f?884+ESCFR}sZ@xjLA0@Nt<X`LkS7Y2Sd?(I^_)z4F6|Ut+goFMkdaHn
zAoXb5;7kS9f_Je8vwwOY&SwI3_rCP2T=}tom0ZO>g(7%_PNLw@$3Y;`5#VW!H5MxZ
z#oCCh$?(Q76BBIPGIxgQ%rUy;Ssy;6eDsj5TetX~ufD-MZ$0D(zwj~-??2$L{*}MY
z?RRPtF=TyXi!Xfc2K&GAo2-uxIXRtidOTq~D7kv|3Mx%$tf4+SU{L1RYDQ=stubf`
zwiRey@^B&%m)Cf7bi%>Chn=Osi>1+{NL^4@mUrI19a9Fbq0CBt_GkVOqv41@^&kHy
z{N``{7Ul)DXpyRo59<&aw@9-ELbQzTj!O`N>yod1XM;1A?kvQz&z|V2={%;}e!w|H
zUC&umGn9_oo6RRI77Id130~s8grJryuUK(pS5k<p>9=lz?^cmYYOH~(s+dfsOeRwn
zi#3WOwgOuHg7<*syJhU2^tbb%@kIcJ)|m*Ty+eqep%s97tn<T@k85=zf06<+EU#%~
zG<&x7@icYsN!PYfWUST4x1Kpm5BtY}*m;KR$=CJqWB%}VpDtDVq|d>JS;iMJ(Z<^#
z_GJ`0AsMYPmI--YkY+iyHXPOs&emwD5WY)%3(+<7h-p_J`ui}N;szh^E?`W+CLvbd
z$!I?!yp8`7wRB2?R*ocz?1EM?bs8Iz3%rQ;iI4&*1G2M55uGvhQZ67Ovz_XMEOlt5
zP(&{}y6?bjGU0SGVLqK;TEoV8OrB^&o+E|88pC{g%IV33`MhF#YlqSLItUS~V$%d)
zRn*RckX<TDfIO?B7eXw%P+Y`!AALbcoMRQA``qUk4u>2b9wLA*fBDOdhQqF#3}_{z
zbrBsX)QY;P+1lKqX&QF-_7FlWD|?&f`LjmXuV2Mm&++jwqq5+&S6(In_Z~dp&Ye3r
z@2KjUwvLhj6=ipLhyYC~lul4uE}vfnNJaEVtc{^=8=M3>_DwoRV|*<536Tj*wLl@z
zf|z2I5|k&%K7$2D2z;JnM1lzh3(+(jBHOGJToCx+I+ut}Cn^J?Te%HWC<_%$iyhG#
zrET#xFcPqtrHnI0Az~qfa)!{<Tw2#$+Sz1xy=0gRlF%Yt-Qksn;7n%+6QxrQ5K0pv
z=f}r0fw~X=!9QTMv&mMP)3$SRX-U!ouO)437>Eq#J@ZE=9PK}%n$Ey`be6=Db_B<#
zHDTc})-f@XN0S9xdqXx(evM3A=hatV;h+9L?@<(%uYcv3w|=Wcgl*(l^ZIMAARjzr
zul+T~mosj^{{d+#xN-d|gR<nfZAlt~pB|G-#e6<x;Gu15@-&*)CMx4JKsp}r<*$B&
zQc7mt^I~At;}KH;0b?9@?%YQwiof>P{|c8bUFHWr`-A+<PyaMmuUz514{pDZ{Vl7T
zI^8DTBl^@we?UEt&{5%{n}n8s4l8L`qGm^iUiH&&E3!X}y5Zz>%3@I=q+~K#P**Ji
zjaCBZ10k5$BUcCrQJ;g7^u-W#TaFd466YO#k9O~4C0kuLR8=$~Zd-@eXNIF^m6spZ
zra`|lKT9<}{<pI<<m&q$Pj$+rmgRXv5+}CUAJTQ(jcur&bbRrEdKOgx5+@Wv^tY`?
z1WyP$f@eHZLeK$19D6CE9RI9Z)Z@Kd>W4la2=y!h^latFF>LX@=QM+SWCnH#(4>&d
zK91ih&g<}AG-Kk8!>E7^Ad`f2IHW8Gm@G#km>o|zIedhvYm$ytUM^5N9bE{nTWG|z
zO<VvAoVA!HU{rLp7n0bG01Mq`0=?dguBb&IRrf5Rzk4M}7Z=qX>mIL354^iOA;_+h
zJtMO*%@6T7Y8Op|CzDCEhA=H!Dn`R0iPF)1zpE&%7Zq*YkQW7`wJ|zLu-?U1b%4+c
zyd^6O(k!DjEv>Tz(=J8)OWOG&o+<ibqFzflolZLonwDSsrC;L8l`H(@PyG1OdonKy
z24%ryGGS4*xXyIkSo>HS_wwb-T)ldgrfE1iIpOBbn`r&~n$%zY)nDPp)f?Qnd5g`>
zO{%Km;iCh#wl>+lv`f=892_2Uba+JF)L3g{@r!ljWlkz%ukqkREQ073w1d$QPoSCl
z&OkqQ7LiJ!g+d2M8laF+N=ZRLI*0d`B2D6=LP}iE9LLHH9y~hc#DnP`KS9Jo52ZB5
zT2zR=;WZH~Y8k<Dnk%#t@x3T~9b4mh3n{QEVY}3<rHaJYNNZ3Y28raxOCvVM1>;eU
z7M9>D6p?r`wk7zERhJ55!J|M)g#10w)YXTF{O&LOB47C8O}_e#ud}^7q8N^77jwo+
z@TgueJDm`MBu{gOWf}Dv)^L3PA+58NnI<UDY*F%HTJw6lAjpO*<$JvGoBs!Uul+&%
zr7eEv*QVTlx8>5-26>tAna|$9ynh?>jbGsNFIlFGLjr~?SFf<SwMX03Ocjh;i<nJ0
zdhh^$`4(wwXxoN7b&N_)Z9K<i!W+Ny2HFM;o|*SA?0M4zfYlUK2%fg_-2LDIMVa!K
z|JT3BfBj$nw`^@~@x?EG@r8jpXU?iyg*YpPh$&Q{tCk~lhN<4-d_xEZ7h*bA2Jk_m
zNC?4#YY1K;(ReHf*Ck+KUODdGJz+E);+$uC+9Kot(|V9KLPP);0tgpxPAO8ni)qdH
zu<z5GU0dI-x395%Gwy`>tl{Wr!hF`SHg*gLDN@9=Zk+s5SkZl*u{oV~Fwk|a_^|!O
zlYzk*IE*cR#0$}zxW=K9Xa`|!6a-7*adK7nw#tS{&Z-*~u_I`X?^^8O4LFPRIZDK9
z;6#g)EnZqeXbH}N2?P^6&&uGbs*3sK6d?qePQd3#JW6zBHeH`gL<b_yv8qq{_q-M<
zgn+kkd%dfH=$V}U@1Y0uf>k+l1V~Q?xlby{=wB~ljJ$JEk18eTgd{I0RDwbgyl4IL
z5u=CqIjI)7wncc5sE(P7iynttLahjD1xSS_#FkXv#~u;_gmlOdZA`R=L<_VAt$lYt
zh-eETqTMl_hEIqnuqOBpL_+AaM<T?Dk?#A0jBfi%ipVa8KvUN&W^=0f3~vm%P8gO0
zhGjvb6y7@8&gLXhit%{Npj;!(GiuXzdPz@PdsLzsjz*|7_S)1Wap5Ke#<XYxQtLR8
zUc@uS>OCc8WZo8w1=B^%`|p1c&6U%Xs;c;zpZOVfcXqmGsfsCBpe##-kkJgjwbXUZ
ztFOMw>FI=r4<B}{y`^oRKmB_9?jaBFAM&N&{0du}V?O!XYutS47D<{gUCdERG2UF`
z>h&vFYpJS=s;US!f--46&RE)}!8EZwq>z%}VNo~C7d4BnLzQYBO3Qfv*g#Iar;0R1
zC7N6cWQd(j`XWI>$9P*$D<obfNF||dG2Of<g2(xmOv<QrC9>s-iuydM6clSTLNlK(
zXy;SFlLG@_O~P&_xl|_X3=_s_${-1h^NdlMusIyk)-&4n6ud>LfRvr}l<(vlQsaZB
zC4xGw5_GD7$#)!sXVcXAl!tHK;oivcnH!h+=9_o9bMFIw>W8l)v}BNFY`^+4va6VC
z>IPFS0NB_XQ&%-7lT+G6^W_JQ+cS^w4rx7U0-qRtop*od5yRrM6t`aE@~5{6)^Y!>
zeWt(i_gQPd#OJQo2xYLQVQ+hl%R85GCNQl|@kNd^E$@Hx9VXxYfKUDKPg2zti`fE9
zOIaxPo#oLiaR1>ELKPTWQLX4x&PQJ;K(BN9nCqjYATLs~G>f#UbIUd*Aq48$aqrGP
zss3$#>6iaL|IWYp@A3P7>ZkbYfAi;G5aXa8W1ae9iC8J-kRn~OF0m*qre%ZdRxUlb
z>QaJYnYQhK7n0tpA$VBKYVb3vDyH<Rs>L;q5Cn+?ghhna&x`J7cXou`@hlZcLLwD-
zuW?S{Y@lgOG?6^pSO5SzQAtEWR7<wLtE%d@6Z)6v7Ib}@^6~T@u(T@Z!P9vS+motL
zSK~#bUxN*J=^`8HLkBkLdL*9#&R+KxI|Ib`c^~k^A{f_=YoD?cK_I+}&S<{x=HOx?
zAq0e%OOe;=HFy&H_@Qdzk10%e;yIoJU8}!8&-8VB>dvMAUAHeDXZ@a}gZwx><aZk~
zbSw2VBg;$DqChE4<^wp(+Rhd`SFb?5pn7zOs~bY_5adc>$_nckPXsA+_bm@N>rlD}
zY?4Gksvu1u(}6^RQn3k{lp!v3WT%c00@)Qg#J|OT3h(jWcA|0-t7@en2#N3lNtDH8
zNsd6q!iYK+u}L6H5{f)0Q3@p$&RFWI!kD%*G>%5XLMd8nV<AS6SZgS=l&r|H&QUiF
zLQ0CV1Oe084DUT@AZlhW;+bOg9_mfQt%+&U;c!R@o`Zt}@;v9-wQDRE3)a`hT)TFS
zx8Hq_s;ZbyrwG7yzLdQ|@!sw(Ns=vn`abUY_q4H=Mk!bjaCA7~pZwaF`OUBVHd~i=
z*xuRZ^5rXJS;nRHOXO*Wsau+=MtG>JIo`y|W^3XzKZN-G4+yHdWo<TNHk(KHg-(a9
z5M)x|O>6-+7vPnkl^R7rI2&z;5Lf{wNSYv6_?S(STE$LA-je8qEXxS^_#6nHL=X!u
zghYk-9DaCObKC}kZ<k<h3~Vd7GEm&y%DK5U=IZ7e<02u&gLU8y&Fln8v@!8So37S2
zSPxpMn5pW`=DkNqMVe?-nm^STJR_(J$RN4<-hICK=^b9Xu|aN*NJYzVdl#KV<6ozv
zATcX)k|K{)f32ZyD$4aS0|DE|r+jdtXqt);T69_>@VvfzpG%8}9DL&+aZ-DN6}*-$
z*d9AJb~D=0;DumN4jB!HoX#uu5ASevT$9}xvo@RX=<XdZ@2)X7fn+wLnoV%?2}(IW
z5Rz}dcL#?i6e*LF8SW#XnpTQ`;}8E4v{HQKt6$~)+qY@$N9zs-gOc6dU53LUTIg6}
zJ~`!ddfGXZ2PUTr4v$Xwh5!3+@zX!~(|qRBpC-#P>iPwkK8D99!t<#rA^V4_T7tJA
zMSSSX-kerMmwK_%Y2qVpcGp$KnNG~=Y(R<Fx`!_9`-G$Xh@(H#Zb=d8Z|PK5dDl``
zEo~cb_-JdQm*>`lG^zUSp8=Ee*WgKj?&IUdd7y>hyNTZ|{n?yzND;@S2XsB*)Ti&4
z#~yrSRF{lLH#Q;);G>qNuYT=5*DFZQo%8CspSOqUe+Q4JPtSk1RzcpAuH$OuUbLF<
zANk&o+Wv>H;gbQ~iwZ#mO^HrP(hR8+JRq}_kR)V-Av;&DpiRR;2+STG;pzZ=G0F;{
z6MfE<$e2mt5gj8=x1f@uvzbXCNr)_+@-bE4EpUX22F0<WS>RP~O%tMJi0}RuAozeW
z9&1HxbRS}e3xps|GSVa?Sch#|T-(Mpu#jlkRo8}iw|f`a5^D@)S(29}^J>m;eLb>(
zfJ!vMJA{(V7gYp*v8n^_JVFXGonoEE^<672;+bOgd*0iw1OiE-mlh^fRWTe6F~%?$
z4)DP*!QI;0I)lN0>({Sw|NecZvpGqUu)n_#2-e1WX<XbtQ=ccwAaE&>_3}|(X}sgk
zy$8I1_W|R#-$4b*Aj{bpuCZQ@ST9EmheNy)kmRT&AoD;f6xKPas-|gLM%%kwf8{k|
z-(nAexZt-2p~3qW+txT+1C~@q&0bSi$N(Zp(r{*=?}a2kI$>^F=50kA93&c@#ddrG
zXl4sc)e>yi38twr1Zq(N3WX&D3SgALt#!fcmol!73pR4ens#JP;T_mE0_)Z}f-_5F
z_}~N9%4p1-YrOL$X*9Y{g(uGySteul;3?vnM$rNe$^A*gqmu<Y;}Pl0*Kk&He0a#-
zja#5(v@ih>l4LX@0o-Il^Pr)rDyGu~JKpe0nsPEVJUV#5OP_d+MNqW9VmNS&N{c}t
zd_W0e%1K+K)tpWm_D!8-P@7TPtqVnqyB2pXZb6G{aR^e}T}pr;#ogVV;ts`&TN^yM
z6^G*PC+|7)o%hS^c_x#YWb$Xv?7i-_))mjT;lH0yyp6P=2n|bG5a;Tjb6-{%GQc{$
zINZpmc0gNKpJ#U`0tMiLnggVY(F`@44!Tky_5mT-UDq=c&+X!G;x4;K1nU8JqOY(j
zrP=54y4NGTI#^f;RzlPNhfH8;b(L<KIe8Yf?F<}<7B^-=x{@c@H@(Gt#^8KO*86Y_
zKD`p}2LXJj3png-eKc1}lblErB@brpyJcDY$IO@GRR>(4sf7!7*TVx<RYKIlI2D$*
zn9o18^$vzaa|leUgSyWbZ~wwVSMk*aDWsGDNArqfrX5Fvs|?A5OP)T?laD^1uJ6C4
z%XPaM`<a~=GiDR%Xrr>txA<Y02_Dlw?wI@i8*uHjSi3Z>rz0Bp5NPo5FxJq~%jo@*
z`_hHrs0P=XCik#-<3oP#Hy5=1+2pjr{=$Olv81;*;z09N(KaQ<v4O^&#k{J%t6+6?
zumz)~MGHDZkrTv+Ahj?}Eg3RZ`p0rk0C(;0wkSg%W8!<P?!x+k_CNwIddf_}k0fH$
z7u<F?<b;w`iE+|;F*Y4i7?BcqUB5`>%;BZl^#<|L6=i01w&3U3iz;y(qfZIlBJphs
z{}%Q8R_jK<Vm3-pekgMoJ(jBKXi}g?!xx1k{1Z`0+r4D{1)A!>yx5c^nn~Lg!hJKP
z{%BC^xfq9=U%;7&T;A@_t-w~tj7e*9fRD5~A?zIZxJL^A{U*+Sq1N?kk6pHyOSjH=
zYa!reArrFqpK8IhjhD3zOLzC^N&nlX>#Xkm*e-e3NC~9y)*i3Lz6bda?X!gDq8RR~
z^Qamu&2tA@=3R_`c*YwoK9O=$)Fr0VGlfo}4*RNF@hF?u4;PjWhmU?cOrT1UDx+ta
z{6adTMJrt(BpNxgHEG;~w}8>cFYl-XP190QndhHt5^8FVU0C@RB(gxD_zoViPlGA+
zHjfU(&vXB~G-u(?tu~Ik^(saKg%{J=UisMES9N1}{h-yzN~uI^?b^n?jfCh+E$g^E
z?s(S~$=s4=Y+uuN;kt~cK(8dM({=N!c~*4UV?A^E#+dD3Il5eCOC9_Ot>Bc1+Hoge
z=Xf8kVkT?lc-5ahfkRSB3H2GO%=})XpH+?K7t5$U#cstJP5HyUV5RV{!6eZv!d6&r
zKO0hAw=Wa6d_KGnb_7-~g{b-M;l<RQYD~xr2(;{LVS0l+q-O;iV`J>xf`}SKbq@4s
zBG7e`m|mwS@AaKlyW%vmjs<FyN-6!?!(_%vbCp)R{FA%8MWs*x$}no@WQMYu{%`x7
zXy&A)>fL+%*J<%rOR+av@nKjs-P05DBsabqIx_)?g!McP?<tc=^B4P#hP8I0DXPm~
z0NHoBI>#e4#K$>X<-XwvpOz*<-H0Es;7&nPbHKuQZ`CGKj&I~{ogbr_HaGSCB5rwQ
zmeUZ_B!+tOs%h^2$O)A5=Wwb0^&Rcz@#KW~FqJ5?F&nLi>fM4a^PL={aE%u7IHfsh
z=bZ7^ZqhdQ;i5co2#=oU@Ln#XVJToXAn)WDmU4r?7mF)>VoLZX2I28l(*#fV2JmUW
zs=qt){(Kn1J0_omPw&a*mN2!PMunHuhi#3%q%$E1Qmu*_@{qMTmXA^hg_S7CO=nt!
z+bC8dT%5YXQx~Pl%}~`NIJ;1HoS?xvDC83*<gq1bEq{f}I2?z6Mh&OoXNvdffydFv
zd2bH<L000#eDzaeRzr@khcBX1vUz$=3n^L$)fTsg6CRgH!jj%HDUpCVxP@9380LF?
zFP&6^8VN~|qBVMU+8dBWz~dM^GRiIJKn#bf@0=VH?T<Zhs-fV1oGtG24RXg1PAaR&
znA`_B3uWfiw=b61+DGxnXW|w)CZ@oflP<TK@@K<u8hTKTHu4gNR{n5yS}8x)J9p)(
zV%OJ@z@|gc!d~-xXAIIx`@*Vp^3n60EvK}FeMS+tBEfNBdJI)kaR$PxUtv=KAVz0y
zRHgjEE2A^$j$xt_T@UXPf{hkW$W2J=t;WmJ3bzdy7Osw7?pV5xD66fg998tSH?JIQ
z2T@|@7hXWqen8l%yuvo^Z~?mEQItPhA;-GVpJEWbg#_Q|Pe`-8#OAi7aw~%1`VbQr
zr+H0vk2U}6#uLjo_awR)S=uBz)7zA`Ntm=4ddFsHl7Lcn1B)nNwKsUBZj!So=#iz2
za?6#YD#{jPTeHFXeGeVND#)QIC5j}@wX~s@+)dTaXM<WMD>O|heUBL?qIX#3U2hPa
zI|_1l?%ZU%F<BgtEvL18jfo>K6{{Pm=GErud$6(l%^jnmw|DxVHKTqLQU1eul5mxA
z(=?a;TTx))xr&m+dBuW%pqCgZ_`$5ftt_aeFEAfQWv2q#!ha8b@e>2Ry)YURF(ben
z%`mVfzktBf`uecYhyy?EgvJb=<mjS%{HCjzi_hx$Fq!iVFUGO`@gz4oB}wDe9FF|s
zhprx)I0;?B>GxqMe|1<nx1V%g)$YVBMi2lH=#9ILg@CY8wutyrZBTo>;z%vG+;lH%
z2#N*)={|b2%s<u@kD~R(J8dx}@Z(8Mo?4yeG1eiV>8e>`Q|2AgnP_3>devryTJ`|$
z=iJ<A<TDy2^Y?lSBt&)Gp8IA5C8*Q-M)D2+L^~hzEWhecH|6Z}MuNeQH_ww*BQANN
zfJu<pA<x8>fB%!19l9}_rk9R3yQS{pv@`I7bkVe><eqE_<kR|6nHX4lzPVZZ^Y0Hb
z`$A{$vB}n8?#N3-B|j|cMj{|;5?v`9)dZDh!&IMsyK=F}zDusI#rZ{MacV@??{zO+
znR9q^c&6|M7_-+t61vowG?NuJ+)QjOiRzJ;m`bmZNDH%WS}~Q{c2obsZ8TLqwhLnl
zO0l6fvlOj0AE9QC9;Nc7%;)6(hO9@(;0^P3d*DzK6tY62su#?-Mi$d5fhHoX)m|HZ
z>9FK7Y`3_Z;}sVFqh3zoomgM!AtR%NALg~tC+)09ad~~v7k&r)4scZH=o!&q2??U2
zncVFREz*cGl-EkIpr-2lo72-<iwZZ(Kj%rn>Lv3}JBMYB;DrgdR5&nQgU(V;3>3rz
zqFfoHrg?8G_3J(mRSaGPtMI*^Uifta4y&#?XBo-?o$H!oDOCw>IXqM_$%F(xeK?5H
z$1{G?u=lqW1F<P{7d0`ga*o$CmaE}$@!A|4yim&l$!ryIUORjU38YiNk%HAP>ZcW_
z^v}<0V)XF@3o{}BzhAJ+CDTIX>`R1YQiW(DrTTp|xU&D$m(fq)Hftdq3yBI2Ws$Or
zYJBb~V-`R{s6k!-BrZnH1QN8pUw(pth%%ZCh;RA1jk}FC2!mp$32lx4elOE^;T5vz
z$>1KvgWf`!l42+yC*Rm?Z60o4#q{~)5c<es<zkV^Ck+5em6Zv&NWtn0-&k@~o@wfv
z8iw}$50GhgP`(SSKl3Apt69x9Ygp(jru`?xpTvMd3j_H!jFh>5luuOuM~UFM>wUfn
zmpsONY;XU6_lL&8!R4t)qUDIsP0;=3WZ-9ugAVeS_6L_k_l4m2xSHdQg_Ud&`OhCw
z1V-j=OE=vyG5ICVhJ>s)kD}Lktd#5R|J<~Ccnr{`OjokXYJVJmu!qdlRVlBr80#T=
zSYtIY3giZmemXCb2*xT@k{mkrMH1`m_1e60_;&y)Y~nr7+Z2focFK<D@~nDy2dLkU
zsSufj)RZlW<4bvvte30x<{_I_DH~H(#jJQOTfMzrRjkbWb^S=7;eXUo^dxYH4~ljS
z?Rt-sP!T0zBWUuoUA4OnV|a=@&x8z*bx(SySq6m(fQqZ4r%LSVO^1`arktf1MysNa
zf{$rU?ijJl0r3rruMnUMp?mB(q-!*h9e&m^9F<wg3jEbBu>venR<{K(+5^xuc?9_Q
zK}fp^FaQV%I!dH01FEr<*r>dA58!SSAUsHDGu^y8#1=D$QsL5vElowZ_8})?WE<A(
zP2CONGQZ*c%*0&Jp}{MYN2AS)9T$pmLGrUVV6a#&o+&D*l@DJg^7vOXS6C5K+GIbk
zLOleACD2dvBG|?e<e;GC@989-NckRn{hfL_iFm<G`WK2?5w~H)@N&f$X@3Dl$2gUb
zXsM07SvnIx9{BsbhH&9lTARy;vlnrslw`K)@O)RfjvH*4Ky5ku;0VGe<Dg_$v1ARZ
z5XQ*Mfac&iNkW?XWQj6fsE9I3Gr|%7l*i=1*mvR}g{y}JXBR)~>_S|8P!hvt7NKq%
zujl7Aem!T`^=V<Xde;?On)E?M6w{@>PYPyeBWx2L@Nx}?b2mWf8U7?p5Zh!<p-E3*
z{nJ$e0=fO#ZE>Vn^Zxa!Ft9bvoV^{C&H2^b+&<aPZTxlZ355vsaFebeE7Wt6#T`8=
zg|e@gL+)~RDt1Nz?D3pTvKu*Gps!o~**^@5efUcUFKq<FBs0n(E2~l_G9daIq>tgs
z@B>HGRDV7%ak{v<4Ymb5P+pERij6JrQH<D*{~83ey9CMVg(QOy+x%}(pVdvyW(fG(
z_|wmKCwgzllnJjGzYfyfZsFlp!>UHoBosaHY%Dk|_tywv@}ivFYPVlD&7aZPlt;v=
z4cFf3)_42n1Qr6WAC}bOF19w>Oe4S6LK2N-a`n%xZFMCPLg1z?(K^Z0J<5tEv`j6o
z&WnogiwVNP)i!@!v8koMLWySEAtRNlrNs(vn!57z@=bKL4ko2~Qnl}p8j4zcb3SS)
z4u!2t4&x)`as*)o_w7!KGBSf4d?)c7w-qj5bDr!8<BQvPE?8T!`n)RM6G{H?#N~Xn
z+snA+h9->aS)EB;3mi>aj?F#VP|mci9M*{I=y&kAJO~D2tkVh??s(0#++481`e<1M
z{272_;JH_wXd`2fLHuY5g=rNOoZmu#VKNj~Tn}7%zA>9$HeP7fz&*%Jj`GEDv6QY1
znM4@&#{--n*fo|A-du%eRV0$xY8r`pkZI(%ki?4v5>7Y-Mo9$^8U=TS6{MW#&@Vkt
z#<n#$eP#yn^Bi$JQ;S*pO$E0|xpX_*d`Q_V1-!G+PeQ_FwixN0XASKw-}x}gD%If{
zOX+x9#gi3+ju^^YYZ4TKps2<m(v_(!q%%NanDWjwa4j12I0c+(jVqGBpDQ<}Rlr>b
zZL=@8%8WOKZoIz2)pcEdfOt2@u@9!i-G1p#<dn!O|7|r-@^t)sN#b($+PyEfp6=z9
zSgx7VqP4EEr6-SOgp=;fT2D=iKs!Ypge_??_R-`hYZ;h>|AUQQjSnJHEik${swKhr
zp%fuf@m$K>NEfsN=o5BnJQg8S>3#p53Q?@~C(UtOQ36~LE%6kbB0F<(h%zG+bA~Pf
zn)x(bHC76IYxwz6ZVY9wXQbMCS5R}_5h$pQgD`|+FT+aG+%CphGHmp=UC=mG%wVp}
z_EB#-Un2@I$sF|~v6=NXNX8hj=C!Af#D-0bBEzv=S<Qx|cl?lhM|oZ<5O=|FKH~qZ
zZQRkSk|a;D>W_7T0Ha{w2g8lI1oq7?3O`zhN~G-<5Qwo>T)}PlxWsp346&voJ2((k
z#|ofmH1>FtPy>crBcy5ZVv|rsR?<5;8%CHy?O3!bFN`lw0&WRpgW--vbg3sYb8?bI
zGR6ZjgBE}PL)|CY^xE)AetYeGY68Ea<?qALJN^%puiRx1^;kOb&9Pj{)|z`)uvi)j
zJFlk^OuQFEfc1GPM;rliAa<{-`EQh5!|uCBeCytKxz;3cjM8ThP=iE(So`g}UADD_
zas%3EXrF)&me9shC;)0(WPa8^{;gQX?+!R#513X7xG}q-6t1f|*6;p_LCLnR7vgas
z{fT(5x(0&;-ED@~yV~{1{N0YZzwx)~z!AMFMi~@qC@EXt+6Z`)nbyo;jesJDD|BD~
zb16+hw0^Aw{<KI80Vp*u*8r<b_s;e7(>6iu_G<4HSo_lyAB_J>I-q^~uTvo_S8coY
z^tkwM%#trSezN~S)LDMI+<4p7m-2~k__}SNGYuOZGkg|aduJz`487Xah7I$Fn)T1|
zqG=nP@3{;GpAR(ZUha%Ga|dyM5Qas@aXbODYf}v^IznR*%_n?2j%SkSlGxIlRjL74
zeV#HO;h4>C`-En>u3^&LyqWBp&JVgm;#g+c;|N}BZiD&&m(gtJ-e|dD>t=L$z`77O
zfspL{AyO#zT5+K-!#874;W7I^zw7p<n$N@^JIzM-1=?QH*<(DZOdM-=@bbZ-MNERC
zqNd!LX1bM;SeTeUznv^!uIm8V&#Y_s6y;=pGAUYJ371`UpeJF&+p}-H-nQjEf!3ZA
z{%m2}BSd0Yz+G<Ib`@U?_M9}dzh}<nBwyPW)XTX-Pc2f9LfY^&i@Hw3r`EA8N7V>&
zN3upoiX;*i`}z^(7@N?8F<8Dd=A^YcdJxV7AVeoYZDLpwBS$-46oIHgJ=MyWl%@=f
z2`>rdO$I+MGl;(|rMK$@>itHoVQfwtvWTm|XmzaEbB}2PAqq!}T;}r9D*TmFWtP!T
z!<YH)KCK|dOkfbT`=FS{3^{x_L;Qg!#jOE<NXuF@O->Gd?sm>C=-!Ec`^BIuP@JCV
znb)nPt1jbUm3t8K&}`=BLb;sQXOrvZ9t#zUi!a_C+0z-bcE%JUuH|}J-Jf3ZZN7RB
zWsr+e4Ao$+(1!)K6w6Z{h((!D#Oh&*hTKN6)qeeR$3Gro`yhV$9|);szpeNbu!+x#
zwwFhgv9WF`FNyq5_n9W(D-dAd&%Enf>|c5lW?cI__ZhRZno(wf@OxJzHMTC$zugAQ
zDiYP7cyzeyNR9mt2?o1GU<v{ykL()zOu1rujO<RiRN!vz9D%6xOU?nqYtspsDRrTZ
zBHF;X$E-mbYus$G+4&#W())NzACV8wf~<|$5_7&x1d^dw?2))JP$Y-5YWIh6{_N~y
z^l`VP-yLuL6LDGMSXyby@mTd1dwEmEDU9VP+u0DK==~2vxOE7Se{#Z#;_HKKC(Km*
zk*E5k`$daC)n;$sNnb;$)xLY_%+^#n=$5=q62eynyg9Xe*{@qVYSC3Qd4s%LsZ2T>
z$TwQMO&-4VF0$s_X5Vq)(5Y;xE#<-}tHI$Ae#C>Pp8ugf7WhZ+V-*1-1q%-%nP{`e
z$czF`8lIjOk8~gu82(icTU|seZ8}hI^PGko-J@ev%n}VQQ+}pkI#sfzP-#{~nass8
zXyya;d)!#M+nw6uD7xwDI%jUf6<G$M5k>;9Q}LbK?(o(2v!%s|Jd@e65WvJ}N|~dP
z2Q4-Wq3iYq6q_X&IZ1gt!bsVpaw<}49G@Cp!!BfIH2V6OM-BxC44Y*0q6-N8>}nf$
z>BP!Vot!DX8p{d4qqw~Mldi8!G?C^Y`c&K^Wvif1n9J#b@XR$}nxU+Ee*|j=TOD~q
zhK)JfVU&Q23sh4b9;u`XxSkf?KS5d*mbc^jN^7J*w3_fg+}j{%!X63FtlNxF$+->t
z7!tj$etk=~%9@VE=FWXYrClb36gaPu#@wghh~LT&U0PX`Y9+K4Ez2tI8kk4Jhhs0C
z%E~p9&$HYKhNH{!UU30re#)HuR3-@uuj@kvd1hRE&TsFtHf|r={Vwr4UQkHP(nokm
zdlMLbNKOjG>noPu7OhIL_#{6%lVVtIle(EhtH2GTRiaYE)ci>pRKnMn$X*ocQ!+=q
zy|`*fOHwTg8!qAfU@fJ?fiuU;xog(BY!N}jv>!WL(J@@7pwd?C*861qm6qMDL=^=t
zt95883U>Xs@Qt?kRK}%#-p)_4-Cb^LXZ)bmSN?VoFi7-T^M>6_^INU`u4Q8fvhil%
zAPOeRl`Z~7_lL?JJ91~md1Lp)4$~|+7CF2tg}y{)*u^h>8H6+x*wVsx%GN^=nk)h3
zyLtJ6@~fyqpLb))Zcqd!CF~yD<6u#&$|<T;(ns5=8nNrKS?M`bnp?jM3x?xd>llJL
zW#YH))%}mFzWSwYfuu(1e;$!d=bt#=Sm*{Wk<8FsGpd%9r58PlSw>kbDS?IFZVkUI
z&~>RzhQxh8CJz45X~36jiE1;5%PPCc9Kxu8xZM)8*6mqqxU_qj>IfAWNeOma3hd!T
zjW6&>jRbee!%JH;Ql~u~*9~|L!WW76>?32WD;P4GdC_=O?B&Hq+v+Tg;Zws4Gru&_
zC`QzJRUHSenFl@`GXB0YxqWUq>T`R$JL88mc~-3)?V|dz+yF!t15RycI@A=m{O?BC
zbmIhCNgAP3Sz*g_8Ld$jbk=`$zoJU#B}Q_Yo86^+oBm50Y)I@<_y+W{T{so|{ALtS
zItE3RMvKI;kW6M^hO$Y>nwwLLd=rWo6~^=JWuc#nlsK9qC=;x1Af}b?ii7o69?B`X
zq+JwP*qh@S#bnNZ50+zu4R#Wqp7@M^Jz6Pn5t}Mg%~O?>IRSwrMLLisUSWPmgl>47
z<t%LYV)Z@<6v;A^I!8K^jssvEXgPf64v0*I;7aDn$OXT=AXnGKF0KNt&CTUo^a2ka
zdyp450)Ck_6xiBzxwGDBul5`@7b)f*lBH_s*?V6f;J{k=Pu=r_>oD*C#)G56k&yu%
zZwZeUQM_@=7IbEFnHFBa{Y~3$({*09qtRh{(h@gVjUV=n<3qs?@>?q9)-Y>b0)`Ty
z%JshQ(}{rhey&&M5@9KULgjR(nY*BgZ}ZscBSU`Ix8y+|Rm#=<bBN=rGB<=bl#otx
zvkW71PHTjboPAGOd7_`}s5V1NkGM2Rmm~Fg(OPUkDV=kxiX{2&YzhKP=mk3iYB6|Q
zN{^nkJZY8|*DW0FV-yQK2{7fdY{G^Tq>n9elu*9YDKJeWJ%ClUhk{TRGG)qbNn0K$
zw-Y-_VV5siZj<2h3+5}?8D0X$<d72tCkNa0ZqY(=XYNtAqX$4oBC-pA+BLEP?IrAF
zi2t!L*xJ$kJ+sUl3d)e)8~}8c&y_FM52vqFe|1PXWc7Ld;URxpM?~@|&=KDv$QN1j
z_3;d_166ze^yCbVzOLzvX>AkAR9<Qa=Xz@=(`8L8)SmOPGYbrd6d!(Iy?L%0!7gwj
zjd9F?A?$^inHzIpY5F1{6;(%bop7aGHv{b0e{pNz?0PeNvoD6Z$!2*~h`+MU77_9C
zg~rU)Z^%|przy8v*F)JzKnP|2FbYB}@c70q-mBr+38+nN29`7Xl7ED!+02n)8FA68
zhZRi3lBNFyduE@Ucn)tjvJRxLy+0-=PWZ0<#jEwr<r~LD{-j=2(;V-<|E<!m2O6JV
zAEH0*=WuoXjUzP&Y4hFTGj`H``sG#baGYi<-B;ZP`FjSdPm49cqUzRWH0vM!yF43y
zY-k>0NEx{SH`SZo1-EoA4>W=M;qgMP@7*D|IpvOaSw5;p+JXyav3*&D8)I@BHR`re
zs(iL-g(b4yGG)BUB6qfGL=6%wg3^I>Ip1COMS~Ei@p5L_Sc2!iv!x$snK?uN*WL}2
z<(yTT8U4%sD6QO7CL?9%r;O)I)>1AOn0#--r4o&sB&nVdg5yhOFQ4iH7BpR!zr`v8
zEat28C8RJPulSY>ML>bi)t0B|4Dx%CCRVIjl|>b#%ax{Fx$2XwUGAXmgkQrzh|&V5
z>&Y@bQZO7#D)H<L?_?dxtAA_;EQYbd-+FaP!Mr&fD>@Fd@n*Z<-8zO@j;*g#z)YTI
z{JGrOIzLXQwRF*XZs#}lW=fUjJi8531c`WLY_r_le3XrMN^Hk(Z~67pwz{_k{I>;Y
z<iU_1SZ5b~3)T*2Mtj5doTgb}hZoGPPG7jo&SbE2ms<Ks26jC9FUFXg#=q!&SEHZ#
zP>pfVwsq7}1wKmE^*+2=sMjzH+P=Rh4zlLYFO*|8Soj8d;mNn+*Pqg14~4)E*)jYJ
zKLN(YYvdQ-M7B+QBJ!0{5W2z#Sq+V5fn-k~qwmrA)ze09Vq+$|-7*(yjkcFTF|}68
zm=xRrF!8-c`13)attfQ3qK4Qi1}f`t=EP%6In$9fTOspkGb~CGI|ZaIH3WLxux~a)
zfA6bzZoKd4Gm%KMY`&Rokpwe*Nw*v6S5n3%hX5jZ@J$Dex6BL@wh<YxXvZ%}+h>&H
z)jrMVOpzL{HGZ>=|944R<As9gtuO6bf(4@U!c@D5z98px`K538)KAW<1>~*ePJmzK
zdVYb<aEmGL7k=Jhy^-op1XYu7hHQ1WVP&2=T_p$nRdczAiah^)yB5D9daHU1`<rU;
zyag*}TM7UX#cw~h*vIh8W!MG}@px5@*Y8J5;7cME`Gu&`>+8D+XU_m9s-rBIs%Svp
zB@~8Up6-5n-BCDq->th-j#M&c3Tm<dkhCCWha4vibdq88oB#CdUaf=lOphS`acAbC
z^hX)OW{gq(NpTen(=QySx*ytCM=lp-lowcPh+`FRIqKYBhA9?m&+5CIhCwxShX}u_
zSf{iTXwnWhuV@~5KDB*dfs|`c2BT0v=6%m2>VwBAI#$Q3rP0G@nn$(lRK@Fl*TLxV
zm|Ej09X2TxBZT}fZ|}Y%<xIn4qqH{l$Pw_!C5&9zqcnBxy72F}9RuL@N-lT8n{Zo!
zZ9STpmJIAsb1ims=z!qbTI1@7r}Eji25rrVj*mOhB9ktW8I^SReW7ipFeMsB$-(kB
zR|XAyoE-xq7h{~DVNN!drU>58$MALJ{lZ=E3h5LFuVK(^xbPHsqNitSZ8nz=>$AK_
z)bNV1HX!<GX79WNR<$G>LeB{YClD_NIl1<4mUG2MF^<Hc2w|I}2l!4KmzEKPG11P=
zX4tI-!_g6?hCbsRlGcG<DgxmQw@}`fSST^w1vxz20{^6_*v90v+Ud)S|2EXq|1YC|
z%;8Q%+d4S9c$@Zgy*t`HZrbM8?JW|tRqvSgoxDa?*<VQRN^_a1xqJ5$rd)$xo;7Ux
z6M~J7^cm(gR7i4xc`$k&Tk2#^mfeS;RF+hxuKWC{J(D2&<lHsSFHulpjnYJ8DkF)f
z#bZw|K-t>bE~L8?p!i;v-l9T+4#@x?e3@#{<#CL(Q2W)~iOu+fEE!peWu$JFKA&m5
zy4&&IgnVj&C#)|8=j-Fh8>ie&kO7QF<>#C|$cB<Er*%h}0Fr}8LA>hdLuPWrA%Y+!
z8M>X!Ak}q)Q`LIGxrNxj?<l$b?gPK^7dqL+c!7vM@sFE&WN|Yg!**hIU0oq=<PC-=
z&&U!Ib1{Z`nhd3~{YTHWdOa;AhXe*@Pi@)~x(T<->Vi@8Dk<~)TFE}1@4}vkxF^W*
z^~9KJA6I~lnR>UfNcOn+S;4`+4%N=)S%QA_OiZ{?=hcFr&7bkeXS`D&iBg)Kh|cRm
zyJz6tk3Ft$8MgHnGww_N+4+yfZ(=9l;ZEn?v8p=1xD&N5i1>r}#|5O#uJ2UdXIrSp
zx?+8Q4C<Q1R*RstwJ8|2S?*^w&R4<Wx5W}BjsR&!R4g^ja|ma)WABgiUt3$KO-n&_
z0WV&(dY%o+&tT*bu_U@i)&F#<o*!HzlFQ_~ooMh?m-%j=`RXvq3ce2%&G7EGmXO`3
zNu{VY)wgK;fV*uC6GR|I+3=rrO;_~6$UgjRI%kd!G3{W>@MD2*FbGR{%?b5@t7P8s
z&kX`9s`UA;kE0^UdRxlopy5f^Q)e}+P%(s8<6mtuD4|+GQ~=}QD=1J#jRqlv@C;)Z
zke7A}r(|IQSRYC;shc<CF5AbvSe2Libf_y)e-JA-nf*}v?3*VHF1sofiN`9JO9zg+
zK-Vnza!)@;bhO6HH~AZ8-Fnkt#|wetTaae?rzM&zJ~;;SFf=}YS0%$|jn_^}I>OIQ
z+hO_qC}E(TQD(YTB$xUxPINgJ1&l4oY|}O39eEe^P(7p+&VjJ)8i$W07mJo?bk{#}
zCUL$A|3OKDMqA?0$<~qpGUSl}4f=#n2h=2_*Aa@b7s_oTpJc;=VO!Mf_KXt=VVPOT
zZ)U6HsIg+y`HLK@nHhL2$ANxt<!f;Uf-ekTEP^!X9b(;XTag@UFqVihi1$d!YHb~3
zr;;B2K6LJZ#fOjF%dH{E!EpIc-zQzKhS^Uyl2<7Ko7W!wTU*qgMoq1)vRP%YVGIg}
z%rw~zQkS!EK7X$vzR#IF?jMs+1xz2rE+W-nO7-|w@n@K1xdH&+CqMz>KURq2h|y%)
zh5DKLikD)r#6am5f+k(<-^Y1ErkLoKQ8kd6leAEpxzQCRV@+(}?bf*oP}4b_o0{4N
z9-)M#UwGHmfziK}7(55YF5sdcn7uZ7CZI4gf2U_^vc5{FiP^{6-wF%E4o1D>2eF=7
zo{ER)#3TLRj~g}n(MmA>)e@|Bxkjq_U24M^Hr!#sa#%tg0pUZ3yx-yviC%ysAE46e
zDfGb?fm`)-(_i@9yPD(=Nzxk5I?|?lH562`q^&lZk)&qXtNFC~=qI=vc^N0%lIR9H
zB{ap?iViX*))?OoD7Uf$Zk)MA#rkwmO-&X_f9sJ60vV@mqDl(#R9&N^Vz&~&ybNC{
zX_19ELIg7ukRX>P!rPSK8|t(AF{0O#x0n_2x2x5Fa}i$dferE9;5#RGpXUUbrp?Dc
zYF`W{2bt<!lVGWHFD(1zKIXID8-`iC;TkpL{qTNjFRQ%tw&5JrvW!-S#k7Z_;pU62
zBV%GUq5<E@!^7)}4NmFn0T{>W=Ek$~=s^>&kG`R@{5>XoGz$N1B6?OSmFhiH!RaRh
zY}`WLqW5;LGx7dDG<5RZbePm^2(yli8kihT@8-*wBdmUS_g-Mi59&C4WV&wX(!{Or
z`<p{e#?<Ns{poL>KB3#z?qivLvc?`3OrAQto+1V!w`S7}Fu&0@-k)ucFDDQOzvN%L
zzrFE%S;?K(^g;0ZcRw2V1<s;P4Tqh#<mjDOUG~vzuk=4wpL2Eu4QbKzrZ8;s38=YL
zUW1G1lu^$k>Sxc^PZSjvBg7q$rb9Li@E$GkyZEHa_rf^i`D#<Qx_)CQXa_fDh1Oex
z<2UGqO$&$mx@r3#MWG#DVON6MY;ZP{phu2*to@BAqp$<WkU_PrdEH3sa4V9tGMmAf
zL?-?;0w3v<RRggEQW6|K*>TLbw&I>KgXLsRLhWdYkq=JN&8WC0??H9-Vx|1-rf?5{
z1T0N_cuw{-NM2!;tTu2kN@B^jm{;2U<wv_aR1PW`)}w$PnI_08nEOX4PauasY9Ms?
z=PjFboF-ztot)=eU~86!npH+csS|)h;<HCE6Fz<^OeJ)PQG|}?oLpqwsE_S|eo?87
zw)`>}FykuC(=9+q9+HuZWR!8n?D}x(1iQ&s)pe1wbIj!2TnNw3q9%`}*!2`Rl3oHB
z=Sj;|*q(X(9*t;r^)EE@r|Y3Jtr&_5wTA8e5D4bo*~WhM#y!K;(p9-Jwa;$_IN@3J
z@g%b^Zi1QT2Oj-va?VX&yDZ*9((%JTrUR+y(Cqbi30BhE<eN?H0Hw-_Xan7$rh=rO
zh_U3Of?#<!#^@;W@1tn7Y%GTM^-Wo077zif_02zCXVdV9pj;isd%(y(N%5NFT!BOy
zNxM8Ml)Hjdx-5aeKxK2rr?19-;bDiBGkH@#C+B=~+7I*gO!E&*GiFS97B{Pcr)rj^
z`b4^2l2#EOVn27)-eN2LgobP|E<K&&+Ei=Ykb`1T@rC=`_);lPSj)Gu0$6ilR@sCT
zrP|8HHf)>jF^Z41<UiwJzO8c>uP_A~ZMe|cDt==3WJQ%URV`Pa5~E0tRk8q5;b+JM
z&t*mCUadHvUMV>v|Gs5!{Wko(`gTclKYZea&0#L`5!<j|{I&F}#oIYt{#zoanaV^r
z*7^|p9=2kTJ0R!x6ej=C``5eP`*d&THws_jVqw7@Yw<VIU8-UhrMS2_p<Hqvjxc4R
z0_>!Fkp6tg%M}JQLi9t->{5!Zv3I63`08O|3@<Y{0*3Arzy^k#<0vC4pfq67H}Xgf
z;B*b&Dd+xUG5m~g8eNb;BdHjFjd^GwZ6lmOmp0o@R9IImYJM(k8gRCb{SK!nKKDEk
z^<vQKR-bxtox<g~)Ts9BhN-=E?#T)D@pcgJAIOf^An%-=XO(-c5~V+{{2No4Q>v&i
z)<2yallb>tXJP;fPI9TRxD#|%Y-YAP6Sd~s(2r9Xy2uIUGG+R=v*?A#Io)nXJ|;x)
z6Y2jn0xzD0f2q8~`O7tiS?Mb==y&6|3%@B--EA1^>8IEopym~t{&YSF!My-K@{}@0
zPD&g%a+#NVo<w`}kKeP22Yg$j+kCp~3gr3#*A>kMo2xW?614C`JouH>8w-SQfrVz{
z+GSK`lPV=rSc<ZpD-se)eOzBwib0N_yWZY&t=YrH4?ZHOsoIN{H~q39-$hYmf~7C%
z9l*>VU#<}vszMgvb4M4S=5ICLhf*z@&1tKw(P=pjU4X}J@iMYt4KBNbzIGZ6saxj-
zvJ@m)Wil&)ZWn|cfj0=lr6zukgUj_60OEQs3LXml7-ag2Agv%MCOdqMbAwZ#*|(^x
zIE5?D1Jfr_imFK}YBQfxSkw$%&OGIOM_*MRpH7%s1s5L999TI7e}-XOq@Wplhxz?D
zhtLd1rPm}F<)~~H6&d;IxwQ8>eeB9?>|Dh<`qbE`$6etvCb?K{R63cE#4@WO?DL0;
ziWM9^m*}W=EaEfyiJ*i*5!+(&1n6fEbCMl;`_k<L8`-4tonTlAHF0&UkfT9~@@RNc
zY_8J3jc&^B14EcW+6>ErtV*%e?Cn0J`y!#{#1WjPftMg7nOLr+t%8&xqe6{m+~JmM
zFXAh2EeHnXzaZZ&iVHHJ3_!m@PD|OL3*qREI9a@6v)u!T!R|<u=uqC~GFfbSJQa1q
zQEapD$g(m!au_h9bSWvb1^PF*+=;Yu_!G_glhjb=yq=P|KbFLrttfTC3&R41U)Y#4
z_+yS0lxe0&rZ{2Q#4eWDrpL#(|E8b;S#x>V<>eB|-W$?+cXs8~>;0R|EM5E||CI=Q
zxbTJ(aC7#GLBYOfqu4r=9Wy%-GlOxIoFfjyb%%+c%lW6t-JM>oJHPP@APG0(^%VjK
zDBq@8b3zaUo**kf-rW(QxuIo;(xAJw)_t{BZ<&J~(@8CVOPdrRrp7v^zsdltLa_gx
zIMw&?@8q_v|Aoa*sVDPm_0L8^6#-JhKzo;-VLYRe;+;QU#L(a~>1B>Gb6?7)`E5K6
zyr9n&66d5uMFTx3(_ik{x=GaytfO$nvDSL}lg{;Xo%|vS=)NSUqZ%048+P`L+}Nl8
z^zlr;iRCJu4lEi?w{@U|@Vf4oYFY!U&%m)_QH2IC-YraK+CSgAwnVyhtoIt8UmoUm
z?2MEUz8QTGlGYgqhI6qGFW0U}97IX$m=Q?S{M!dt19Ps!sdf}EBJN$X9-sH~n@<>x
zR<2AQ0YAi^l&>v;h-X#P*^J`0Unj{Z;yu&KY=0crF1W17PFG(=V0V18zet*(>LAp6
zK@#q^NaI>urJId{(Z23-^uqIhH}C^<sQMNOk@YwmB)>Dc_I=7;ASa(pWTBocxPd!)
z4-6{}HwN7CltDy7+u-UD#=XM!`_**`!o$M_zpA%0%QZ32%DMvwikE^k;lqBd@*>z{
z8->~<p@)=wdEY8aU*}}|O*Sj`#uLd>GLHe@TUDJ7!CD@lX}8llthB-1DjF|q2;ZFT
zDq1*_QFe0X`j+<i_}IbOIr#tqj3&M2m}FsNG~IeK!%h?O%jAnjn(ONq;8ZKat(^6Z
zFD{Y3$}D%La)QRi<pgI_Q`5xsma7MH!b{L*XS|{9<yOS@%li+d3b28lG_^)(J%(kb
zdTl(lN~RW?RXAP{zD@2Yvg+B|ldafNf~gS1rN@N!8keF%E_YSN5eFxSMjJEJ1XbCc
zVm@%KEelzGZL4L<E-!R&Xb9YA)<C5RM@t~aDa0L?3v-f8nD1R7ntf^P>={0VF^z=x
zwt~ZRD9Ar$!%mVm9ADdqN~_X|Yh)JUU}*1Ps`~C05?_XuKpv+h#G>D_bK0Rgh{b&j
zO@}cO@Z5d+a`yH)Z7}CDRYLlEt)lzQQ-KC!m;*tDi`2{og;_r%>Q+Xl_ZDm3m$FN#
zR@3QqkjQ&~C*{zf&D-Vcm7Nnw6=A-;Dnnd*&V|>yVM4dp_ypN|F+acT?cmQn;4|vC
zg-!ymIf}Q_fOAG6VZXt*+vtGcWFOVxjFW4mmEeV?rH<cw%NYZ~ozIVTFZ?14twNwH
zmG!qv@!fFe*bM#8vm8v&VPa?RYxx8$B>etumVt$MU{9m?)_jNLZ^FgldNJ`WS!$3w
zuQac_EJYbdBj#Hd+6i)S66IS1B`h@afKkn-%s7HCbGg{!pLQU=Jrn}a-6Su>)U$S7
z$Ci?ESWERdYuN<ofJ%;f;W2|kO+@<qC;wjmUdOxEG0Ni5(a$Z#l!@rJ_DcY<N!<dw
z@`M@@rm3_|KFz@<g$-?Of0iFfz4Z&4hrQK|1h`V?KE`ADk=Et^Gasj!Mg_9^xR+r~
zUQ8F228@a#!RLJubnS6l^i|WIhp8IT(BzV7=_XFz3lC+oAgE9Ev5yxnqXYA#gKRaH
z9;Xd(z@v5Z`Z@(>p_ST?klqIWdf+cZ>+XZH&li<taq6+{u^mNK0>8}jc^W0tp+;90
z{it00($N%UDpO8eb)H1e>eGW_NdWRAUK|pOa~jftGg-pBUG+~J!u42FN^{uWesVWz
zLrqh*0INA8o&(Hzb0=+EiYl$<u9!XfMl!=T*&)F5f)vS71ZDy#BJg-RGP3H(1RGKu
z8jrQWOZzd72@1``Vt^H6JFNT-p()7<b+ipC^8CJus9}c^Cu3PJ$H&b|EL5)|f%vJ9
zW9_~)zB@bpvFjZ}U~*IeB1)uZ-ct3@KT0<(6&snM-@lqM$T{HUT215nX){f+`;u7_
zjEaGEZjF7y$0FJPoS0jc`aptG^rYu{`?j%>Rs-vJS)}InS>J|XQGFjDDI4@wj2zzh
zug7Or;GVBAFHChyisjhpsndcb)u*p)8JvU&W&sx+wJ@jioC+QvP?1`D&Q<~TS96?;
zE}9}^>}Q)NuAKrX_r1sFt8_`w<({XoUZ~!77J3H{FHq0H+c1|e?J{D`9c+d}0do=l
z`*-?{PSM1awAps-DrMRwLMth_nAS<WW8|V$(u%%vl%%$<L8vunwR1$E=Vp=^2G$*$
zq(`6PjV*SK!q$i?Dj|+@`Jk2e7pTUUUv!l9j&9&!aJrTKOx$Pmtmj6o$=3_?^5_W*
z`20KDu(WrNRWdLr(QpB~{Cc_D*gxn)CEHF~tMXm_>6&~K%U;%$)wb^YEBNJ!4hRDz
zU@9OfM788ClXU5(>_z~TGPRMF?IsIc*<!5s5K_|-=s~=hpRr*8UM^s{Er#6F2p|hM
zE*YfaPS`V>kc}vJMN;nh<gtyHN8_~7)1C6iu*={1Yn$_ie-?E+aenkgJIE+z`i=7|
z(e_pMO9yfO*~x+zDJ4$7SJ>{l@9A1+8(Vl!1Q8`QLfI@c47TBIrPdsM@z5O)d#uMt
zb3Lxd{pa2Pdz637FNgOea>e=Yy6FBm>6{$!zdmurdF$VXpEDYiG5nterW{24(HRl_
z^MBU8SOnS^umAO>3qNe4f>RofzyE(dQE^DD{r4vf2k#^e13Uiv6F(!)g74qn>7UDM
UNExPCy@S0JWK^YVrA&hU51<@L&;S4c

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_005.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_005.png
new file mode 100644
index 0000000000000000000000000000000000000000..65026a40521d703045aa4a7d57ea76b6c0ffe078
GIT binary patch
literal 92704
zcmYhiWmH>T)HR9|Jh*F-Af-3~THKvLAxLp4?!`5@7b_IE0tJd&(c&7M;#x>4R$L2T
zp7(zDz4u3Ua>kL7vG+P_%{AxTvD%tS1h_P~C@3fdD$4Q@6cjXX<nJ&57IOUJsP2LM
z67y1kdg;2_dihv**q~@wc)2;bdO6u!GJUY|@U(Yz5#$x+735)h@8#v@DgNq}^Z#DJ
z>*`_m3N9`88~G3%H)TUl6ci$h|30V{Qswq2s3<5Z@-lk9xkvx(o4kGoZN&b&-mt4^
z{{x>3*u*)Fr+x(p$M}KDIw+GS3~kwS;cWS2RW5O0dICQ1()YL7lapj~WXsc(s81vJ
z8P;D@JI;JP(qNsP${8mpO@hJSa}kILINbPL1s4k#i4OXDf~jB}fkRvfCHl!0DxsGr
z{*W8=)PM0hWINYWoqv!__x~BrUJ%^5F!uhxhsYCB;4d=k|KD)|syggD=a($Yxc?cX
z(6g|-|L?uXi$k#Z2>$%<NzNyHbDDXL|9$!+b+1rG&i}oW^UJMZ=+?s>o8wDkV`IIl
zw-LaQm0_Vl2@C|?y3419g}0@F0cQ_hlOaO<?=Bdn)PFt>6U_-fpY`8#>~9)qw>CEq
z-rwI#d~nl!baQ`x3ewCIziN51{k|`?lJ)H_nemI`({A*^=jYpPEu)K5?ush?DwN0d
z=ZDSh(<jmAtW_V^@CZnqa*62{;&}(V&EtxTG5Fe0vp@otii^M?M3rSae}%%iclHUW
z%tI0{NNq%$>j~ISfdC_jEr2C@q2`Y_Xq;?tv1)G+8%`gvj0}R?pkC65k64S{Y(xPR
zX$#0lp$h1MIEiHo>7WR{A@hXn#~L7-<5g@AcC;MRUl^E9Ry936R2p!Qti+YBr_4EH
zl*W!T^!|BeG1MDsr<_LY;~E_Q)AYVP%er7&yDr3<+S>DR;?<+DzGi_O;Mo-Z-XxfA
zk*0dVTHHoRRT%<<C$B@}>#^)!_RqxhTs+@lr(Fd|+}_;faBfKaHz~a7sSz^D_1c9x
zwK|-i?B^YYXuMU+So(jYY~KBwx>)ZI{r}He4|3LMv8}ne@p9jE40TN!Q#pb%=nw6_
z=JvD=99(!e$m`TD0{VcCKg8<pHj~k-&wA82i9-iia)10Gvr>NwmaM?*N3#HGq4A)U
zk;wob>(5JoXoo@&@o*IBV(ZXxm@o<_dU2gJl~PP^&lfT@i=eIY=4;gX*jJ9^iz|<9
z^s?Po7YXQ+2x(k}Ld$R_rli0^`-CBD8pmPO5`{zj8sYx1LQk(eZF(lMh%rbUozgn{
z&)Tk`J8VXFfGi`KMEoSWCZbx(v_vJI6Q(2k`OEv7nW2;ee8*$!x#t_sdad#E|B03W
zN{H$Xhy`D6-<D?HQJ7+XKdZn|yEZdB$DRoz!h*PrY%5JOXY*z=c&}NCV)_R%D(U_w
zDj!u2Z-Owrtvc_|-|$L$MSU{$%XWEA;VAe%;#uIEP&TM=-;cf#xGRE~GA)II3L|^P
zQQ^!H3Y>159-Wjp5tmY}etTolp?6aevYe5)rL_s}bcM>5?2Hp)@%`8mD>m;BdvPM%
zsH$efi{>OtgX-kUJx9*ZS>ngJF)T8ezUhD#h#|~?lv&Hf>mL=(P&%~eznDslCSkSI
z;?bk%I?Qa)ur$IjRsgH)y;UQe89mGmmz*oNzOXkoH29fea{ny`mI;a`Dayw>lb&lD
zM*EPc$UWiQMgW34Y|$v8IaCG=kd--l3<&ENj|Or{!7B-E<2W}6B4k3-wef|ENH8FO
zw_vBcJFw)2V24Db8dYR)QAH^q+z1>B8I_mXtEQG(uM{Qz5XT$RNF52F%{e_U|1Sk>
zRQVdqtV=x%Z(63)r{Y^B=m#POC61iv&MAZUN_cAKKeE`)GxF(Hy@=?AMz>H@N~;E{
zE$)Y9qihFGV=!Stx7lTksqWSFRX2ml`0RI+U_6XW_aYVjOt~y6NiH`jdB+m?*v>0T
z56VJ_sr;RJxp-wxswS8Eo{$lls;S3WP6qHZ*}@9RVpu(z-Ahd<9Yt)G9%iIlcLN8;
zB25$gQ6(KWLKYAUr1<+*V7N_z8;~xwuqd8;R!}y#!b;nW<CN8tMl>(W`MxD*tvv`f
zsC%f);Wng+X3U8pJLW#3LS>liR>36w6<=B99j@j7<iP^0(sT?}of^FgJuaTfDNhHR
zY@_C}gJ$8e!5G+sS&7fBMilz+(6`;~3&+6A_T-{sV6VYOBGzjhBahBnYP=zPvQX3)
zPTi+a!q3sCD~SC6CUO-yk$K+X*1tpXheNmPrWO@Px+J34+MB0ZyU(yLnr$=-j1*L&
z>nUu8%4BiSa#@LfB_JBNd+a_cAE2|B^}Q~XaOp^TUN_ikz!@gQSTw{)D*d^)l5tws
z?oRI*MaD8bHp>$489%?zoj%Uy#{4mk%Im)Ah&e~p3~{|D3N?B&yd1BkU0ZuQmNx<8
z;AYRMWJ1j6y~kAJ#0l>$Q1zkn?QUKOEVkjs4~y%yZZh`yXsygfTr|jz8?VUZZsfVU
zFD&_YMhiFwNYBm{L_sEwZgT~B5Sgq49d6Cbp)8d{WXOH-4WQbr-Gju+7WK~Vpp|Ro
zS$?q1#C~OB9{eFQ+dt2tOr0(Y1Hy7q$~{TaBKobt#|oW|2Y&bF^l8XB)mobKT%{J5
zY}Xn2UU~LyK7>irGA^}u9;<dIfA}6mXk6oSQ2k#8JoWCZDb!R}0g`=WL1WE6Ce}qP
zoUWCvQa7}wx!)zy2L8@gi%=~pmBR^Z?lU;Z?Z@$1(`?3yRsc)jq@<@RQ<Gq9$Kqzc
zCN!@;lSO$ZWyM9%bp=OH{TmONfG?8sO4h+E7i?yE%DlY*i!3`aAYMx^)sQ8b!6Zm~
zuN#<E#n`0lk^E#MI78Qt5pgs)YzP%cCzH1JbDaB$@sIFW)VgwK*L-q+Pj1NBAfCij
z8a1DE7n6h#_Cf=@$+&}z+YZNiWcG;XSU<k1wtI&=ycZ6quup|lXV&{Yut~+?q(i8f
z>-@5~1qfuBGGsOXtL3+?vH%&nVR?&@qaz-*csyhRgy)I%@ARQLa#vOJ)@t(_ycgmc
zh)hu)3}qS@n6-|8H^ic-Zjl-mMISkJ%YQy8O6)h3h9)IlQzu2~(KeEDVf*#H&l|Y`
z0tJ`=_$%)d$N}8MzJwkZbqb$tPyhJ+PXg%pFK#Jl$gwb3Fo;tG2HFx<MTy)OvFSx!
zPG(P<vt1{ctFs1}^h4P=?j(SkFLH-s-Mux5Bf=0fDj5dHfzli{eA_dIl%f1kL{HOV
zLR65c-<@Cg-^f?|KS2!*(>B9pLNpgVyc4=S8GsThXC45EB0k~%O@v^&8nGf}3a>)W
z3zv5fG_W7M$?yD=A_t>^0=?~I2-d|riVBijhy2OaoTS)L)Lv*0bF|JcZf{;`&I4)~
zMUw_KVF&9!2FCkewb}*m2G~D$b<wW1owGv-byy029t+U0&C^m;BPMW|&mD^7xTfBC
z@%G{nD0`U`a}7Gg=e3xX-|{N~F_4L%_GflMj+M)YBs|2Fpy1rr{2#9JMSV{4_<09m
zv@-U2+ydn!PI8hm3~(5~Bq}|Jr^UBUAGmqb5Zb>D=7)gVP+E(Gek!dFs)YMX|9AT7
zXxqJFx!U>c7C{O`II>}0alKJ~#TE4wvqCmjj)wtptp5Yeww$rN!3Qs8tYd(<&ZO0I
zAWC$}i}mG`y2hXpr=X#D){#3eaA`f2U9VA(`;H$A=xXsk4agzF8$AwVM>+I(06J1#
zt<urc*EKeeP97S7#g9H-D!uj%j&08H<JuEwHX<k^(_uPjN&bEFu-4&Oa6I`k9OY}X
zfID-k_WQc#X0;-!IeX``K7)`?*t|!7mvpC!^g<iawi7~J`S+Latj=rn+2w7m2q380
z3aV!Hbgecum?iWj)-iBJr_J{9?cjJ_wW`w56i^({QW;1!>m>h~T|SZorVb<fZ893K
z4ceQE6r`}C7eTBm3sa=3ztz-zT#f>UGXX&4@kpd;`-cFY|7Lt`jRyhG-<!A>&w#|<
zwfi9f03bX!bL%|QuDcc*VkU-oO)u+9Sku3+FN+HCwU~u=F=7q>@sA$BL4`MT-LaQ!
z6G9JOCI~5`uax3rzmYHOHLM$6xJh#eKIDO5Mtnpmn!fjaURqv#-Y)osgv&chSgx-s
zta7szm5EX05>#+(IG@|iryJM4`xPiLh)1VN9ySxIpETMgj*6;?ZkNaAc8IhvoYkjh
zVX#BQ&FXb=@qRlgd)`R;xSGi*pq5$2s&ZPpS56!>+hl(FWZ(Qf!UEOt<$^;-1(1|0
zc8)sHKhoh2^~23I02X5{RJ`dr=g^$e722#j-bCo*dlo5Wy**F8sLy*JXSu<S>b5BQ
zTFWR0>Ucr0fgLhrOI78rx=-CBJ(>dM;-kWM<K5nvq;wZ3?P?qdwqs;Nj|c^S$WkYb
zU`}kuOd>RXVn%ykLM#j9iX`i6^}e`?YV}9e<R45HAEIE|^T_tE8LDg+&ZMA(nwtEY
zU*WnkN*s(-*lAO))Mn=$jBx+ONl#d-C$%kOfAP=mh@KOp{Vg$oQ+w9nSJSA+$4uF+
zF44%#V@R|dX!|Bq+A{G6({vZ_YR;w%6#+Y%UH#(4$C?)WDD;SQz$Zbkyix8T4ifqD
z*HV<I_u#+kv*}&`?6pOaUC!g`Sxfl$y+Ib3rZkgW`ri;5`R<p@Kqi7k+;8uecQ_;$
zN}!;-mQrs&+pk@*-#p#=vX+eE`Ox^fTdXg*^8VI3Mo25e>S%Jxj-1wiozrfzYha~w
zkvp|(va%&hHo@;JZ74n=K%X=P>AuKE0@0mZTr3lc<_>_$3YB`6-Zrt8_3GuTMeE&B
zDVgAuLW}g$rT~rZ>%Gn0SO3N~W#a%v<bA&w6?_MIV{(ZqJi}!F_=RbPx--$i+1}&F
zq}1ujC`>cr^DQ#XN@j*mK6G5_rg=KBMk=R3Rcb`kn3G1-qj1SFHiVY9JA?3&CLEds
zq1$c{mhc!Rj|#mWE6OqC6xg!>&|ay=SRrF3t02$D-6Blav8wu8wcGu?I5K!VDkIVb
zZjydQ5S2-fp|Hv60o5@SNH1J`c-+nzl0K09`T_04Oj*Uso@iCdGx#Hs&3q9tZgeVo
zaM#oR{3SvgbieNGUqc|vS~*RHm+emVf)7ie<fAMLD)$8I8I8eFH8~TT<xyH}FSt&2
z*2(8LQMA-rWHu7X1eg9ru1#{#Ia<hN?ddI(qC9}J{qMB6Kd_-|+4ckTmrcR7j@GGd
zo#=(V)B>)s@hUGG$Dyz&l!L^evUI;89yfi<LEmuLh=z<%A8xfqTL^;7VDph;9nPRg
zyHb;t9nFB)xogZd`ds$Y)wAwrsTob7wR9$DW+nw_p!*6^`;+<-%i3P?vkmL_#O5T~
zO8J{73VI21N4w<aEAJ+iw#T7yC)c&@iFPG7S@Ve8*Z6djc-Y#izDYS4Zcxu^y=KS*
zik7M#kU6{_kt>bKZ=r3S(gL)mN-IQr#aBs5GXZ~f=efmWm<i$0?|eBH`v<1L^fuHj
zbcTtQD<Nj5Pr@x0i9v{x>Osu-==-*|6+a!3<*Tu!q3?WF00GanKq4MUC9-`(VdgV>
zQC6zQrP1a%M3u@qdROIP`Sd0@nF*jBCYwM2&l5$XF^<fw{+bqIrus^b>iN}V6je#r
z=-t3qK%mX!9;ux??-2xwa3e|gJ~}q}@rMu+83Tz>BT3Mx(>Je?Q{LnF1&#N#-O;g2
zAcB;UfQZfWq{YI3)SAlGu*4eYrhImX|GNGjdROj>4sR=%HLy?~mDAEq`x}kqdD^F?
zRePP9#oMmU;A~hI@6G9PeN!EH)H-PK(9AAZxZqJdyFQ;44Uf}Jwj|fuZkN6FTUYp~
z|78GnX-ULj*&P1E%hYaWdV=N6&9cqYhPI{{2&y|>aT@c5TD-i@FDWl7l0gLlN3c`t
z()#kw-#V3xim4NLOdGJS#gv5>^N9ZB@6LtU{@QZ1n=12#$k&jH_alB{>>(`)+{qkL
zC1ayMgH;tHen2Ycb-wnhJ0}?hebjd?u;@`JUzbq@6o+Cnwhx@Ih-c!HN8m@xjqTBJ
zkk`S*_(-$?Yu3e4lm+UmwYrag`Sz<89EHF7sd)wcBWrd@Y%JH}m(t}(LW_$;Q&lMj
zG2vSk*D1Rde;4^em}NpN^{vG|t;jp7o|?(STZ{uuA`GwLwLf0xa(fC>1|^MHo<R?f
za<rQ|T1$O<Tft)L>dVUBR?|V_1N^6~q`Y-{LT(JhSu48-n!$5%_zzO8*W+rNw`eu6
z4#_*xIN743%5QxJ4}nc88yVUaOubamO@uX9#^H-KH8qTIS!5<AlWIUBu6JFg3Xw{t
z)L$%d1uWg$;U&qYTd`_Mn|Fv;vy$p{iynKEtFNDyB6DxGTpF2WFgDgGsist~a%|y6
zISGjK9rR5^6Bu3eTPKzXo0@I}Y5KU$4hY50O;Z<yH8Cj+6hHzE-ir%0fKI74fQk@X
zUMIpcg_Scg8`n<X!2ZU+Y4xS}`a=RF5uce$e5K1deA<BWbgfPOGhyrw7v%0}%>tOF
zKY7-<#02or)$yb&#C25yav_tawKZhNl9&^75+*?>;-li6V{MHCLknkiH`zK<)6%Un
zIJov#=0pw?@R#abS;I2!-+=nz(#jF@fR{RdoFVMxSvy*x(m)JmK|$|*1OjSkSOodz
z;qZ^l)AM3Nh-UuK8#fY7=i3uux~ydiys*)$&t4+`gJ<rD21!vM-PSMB+<v3PMZ<h1
zPAS9^tCz9H$(k=cmHYmLfQ+_HR^QsJINn8hXw1Y%V{TLyL_V%?@LB8oxO51|hS}Zy
zJNRjkBkj*Q?oP1koc-zCq0rsYtAB{K?mI3<-+kTA`#ePbos&H?Mwu#kWjbO7H&&Qp
zs{QYrCmou)cqMz%4X{SoSVAJi;ZXd)Dz@Ca>i@?q&^K~jL!*3wnTaVKHCyvHkuh=w
zpio{N9$c(<KWb<tF7MwsG^CP-1?h8@4*WU~%Eoq86%QO<*d92;z;WHjBOHG(^vf+I
z-blF6gsX7=HTb!3Y3=*)!JYz5hInM1G93lZTV>|Bx@;L@?SwIWe;ju@-KquZ#3++?
zfAUJS{XgCfI#I$={x!YoC4Dm%f>YIUv(<8$m$7)lJw<pqq=%3EN$AsO{mOB<*lF0c
z`T#PLA~Y$QNr#aAvi|R63aB~vi`;!$TL5&0=9W;oU4i%?sio9_NI744GBCL1+xc?$
z$ml|!iXIAAE|O+^d??c6;h%dy?cz<1uS6CuhAR_$L5ZLuAb8%ceezqJ;wjRL7^$)t
z#@goCmq>1Rj#=9|151<i-K#FgZqn{rN4mVmolfx!t^tfR5Ffnny{dX8cqHt%b%c9?
zSM{5x%+D?PA;;Jxc-N5!o`-mZeX4--Y&&gnd-<gbx3G}3_nE4O@O(@-w@L-)sD00U
zF~_NS8&^ywvk?hw>-@SXzA{8T`MAa@;`)xAmVlZdK=L*%351)SmshS=<IyB43y7pX
z4GWXUWvp<hnK62B5AD<CzVDCt(D6^MhwwjaE=7;=!`N=F$Ex<>L-3tw@IPaa4%UIT
z01Gw9^!DC{_su=n$MqD#L)Px*p7+zNyB~hd96ia9=<nS3hJPr<#;{ulw&<TqUZD3e
zKSJ~%eS^?^-gfEK^W+h8-_G&3(B6J?m6!huIM2ki%gM|s5ssdE30SUJ7maN+>hs17
z${~K<5s;cQAr+rgqkMJy_ZA_<zu`*VD$2i0x;&>aJf^ISX5BQ$xYgzSHZ6aXy|kYu
zUf7S_8FT2Mb%`Z`n+Srt7LoR1gzi1I_?$I+FUkGjq-AN7bYEK+m&Ym@f=-o_Yr$ay
zjikFdJ!Eqqcbor}yJb5tN@}DT2^BeMuBe}Up6^0N`vPIl_dWLyCOtuU+KaK)k<kqP
zF~`RQiYf(s3Fi)VmkYm7Kck?lAIS|RgtZ!x<5+(adDEij<tHr~Tyy68I^Hc+Tl(UC
zn&YRQ$?YFl;}_offtPg6+2Vt;k=E%SFQdu&sd5OH+iuutbACe2LSh;zU+Lr+N{^uR
ziPPQ1d^>x_1Im6ZSKJyHxE{bu6G}3v;gXKwtF4gvY)eP)@{s`h@u}yQ0BvSxuUbo-
z#sa?rMO7xYS+`tAM%kg-pb;-t+nm^Xu{e?4ljcQ`+E3g_0LwKI5jERr!PAb`e>?-Y
zn6DJVDX3f9*xB|Uq2bvm7<gFr$?xt;baUp$ADO5RM@X>j8Pp?`D}Oh2bl%;+F}=H{
zzh4iPhE``>22e|^YQXc1^Dt;Q7Hm0~&vEm?%!eW4r#Lb|5FL$UR8!J5C%Gmu2D5R1
zMqkCm?20{?%s}4iPbz%>bZ)_sfu4ey7ye2bO)3XON$+pO=MfaYiIk3x85=^T<*}Go
zsKoO`Cig|~@PzyLdb1K-S|_isv52HT-+hf3L3FRUAL@oJq<x5m3#jiqD~0m~;;GOg
zez$&q9CCf+-3)%>(#k6+fiO(d((O_3edv2bMrXj)S<_$7xCI7$fHPO;qznxAky1%%
zBa`y61>e!d{9JNf$GWwUUg*I0dD171J|(H(Sf;-?ky>=GKW_dQkRVPHvy=Tbsv^nY
zMyJS0emBbRx*Z%B3$n=FO04xPC!qDIJ-c@_Q(BT7zaXbsh~c5%GKplRpJgPRW+~I-
zoISW!Zs4`Aooe2YT$O0`yU*6DYq|%;EHRGzMqOsHumpLZu)eliWGd~a3X#H6)kxC9
zB(_hTwRg7CNZx<4Nk8=TITSbi%7G{o>6jR+oV&z+moOQlC$x>}Ka+Mv${At#TFSrr
z+vdtj?q@>e3J1_ZRFOD9Fj3r7kU+DBxAA*Z@5BUM285ctVl>#YW_SM2pAOo2Ly11B
zutAncSz<orsKDdHT7e@#Dd!7RPPA&`wf3$_!6_3K{GjW@T+z+uMOS@}(cx}1c8*T*
zf;EXvQ!ii?aaEP0)4Fq70+v^Y^9LWIk%n}mX87Zyz~T5|uGkGQ;9(U|2+T+kD4cWs
zSjH)?A-1%+z46Z&e^rU@NOXDQq~RyLLX>QlOEnUpgWo0d*G>7eY>&jR`z!9i$2kf%
zHhqZ(y%$WidVtcuxZl+s8-&Jd@G$hHnJ|8o{;_TMxS}x&dcqVrxtdlPQYH9flZZ>N
ziUmF7;~i0|jD(i+s|*$Xl;PzqO78`DR}YYiP`deA#7-X5ba(s_8zkE#s%}n5WoY*Q
z7#S&^(}8=aIKe9f_V&-epv-Ld+htG@DAwb#I3MNT<!S*E+G(|z4nGe^C?TqDei(O>
zafDX}b~m~v`adu@612K@!+&;Pz7p$_oSNOi?b6JF!PsS}bPA%|sF<UIa#e`o+O-+e
z%PI0kNWP%8Yp}WMRjNHWXrJHJ-(|FtdAN_gGU>F&{ie6QJ~D2QD!NyFfM8Jd<93}^
zYjb=Pp@cX~^BcujMZ0W+F<kKT7O9?tFk#Mvb@ms?S(@jXajKzN<giM_2>02d$LU5F
z;|3<*e}y_0h^88eHm){aT_8TVJa<^lGtt;tCtah@hMQ6Q*HP}XJtWT}QGr(OvwxgI
zGbC?XeS@=evJjrSOcu#O%}s1caKuPQ9E*$j7KP(NM6q|{gjgk&n6j8uCJN>0Rxwxr
zN!|vvsd#G@f(F?7K=L{1``Zj10Uh}FFW~uKwR!ac#X##`bV?*9$V9@WGF6z&><cwr
zSx&0!4xt;LgX^R=KLplh;F$!g>qepX(je6j^J#W2Fim0>Qz&0t#Ud^p!?C%uVj%`}
zw<^xki3u0}IkvtbWkY>gyr&RiIKu6iy_nAZ_9^~(CuIEU@YCJn=|7R1R#(zmZd`D6
zIu*Lgj-U@B5YJ()8+T{}%13Oc%&aqu-Mb+^)IHc@5$8i`DqRBlq^dV=^K}6?9`9nF
zi=5E-W%v1`=;pCpAYRkr+IR2d@&bRUf~Qd;_lpzqrqSDx;rw5HgMwtZ3gOeU_Vf*W
zg3&3syVAyqz{1jznuN|zCB4Nc92nQ~<YM^fuC%Gh+8=TgwU<A}6oJ7(MHmqHbnzTl
z()AEgJXPq~+xJLvanYDwMV=~Dwz|%&7pED}yYa-USU&$pK~YYo(H>;L@;Tt<J>!S0
zJhNhbv%Wn-@3cEO=}JT{hNqF8GUn{)XgZxzxD2Vhw>1?%)#bs5&w#hRIM`ofs2NgA
zab$jj281Q=W2eIPSlglUF=tMP^+okWzdJ)GIN|{^wVPHxt~b$sIREj_96$x;X4S~Z
zW$HMO0;AXR<w+|O)WQ3d!5MoC-leG{YqF!uxk8pNxtc1LfJSDqT4`kCDBP@x%gZX1
z7kPcF*0Xrk3(7?)EYiwT<?irbkLm8ywxCRm!*WX0&eiV=RB}%~yY?hjPYK|euf$ud
zN$y{`E+`y^Hcn8ZCq)J=QJDpzU=e%?Sp{a6LaNJBzxf0OwTidh9CMJoyN5PYf*OeY
z#aYedeS$)y0s}x*zKkm|TkOQu+0FGL#<4X=$kqy1MKR?Lcl&4pKXC}ZC>5F)Cak<B
zkH2lzVFjDx1N^&;@1%FEk+Y+z6gHy8b&nlp2nfe!>>TOp@VT4q3b=d-mj>rfOlMn7
z-<TQ>?{+6GH18UI+`ZWA`W9HVu+%D<`{j09no|KOH0kjOtiGRqua{l<6Ka}OuE(Rz
z^A<_*VW|3r#b3zWT%RyZg&daM<UTJQx|(b!elikDZUMjy{r_p?Kd+Eq-{d^*nB9ZJ
zN!B%|4EE8Z>#lvz@%peuLfyosNTW#!E5l>MUjzC~{+%SkhFjQ&RO_(f8QaU}(yI(i
z_ZB{FGWGub4YSKCD6p}x2vaPPp&ACzjlS>QqZMrrcwm5t?IY~K;^NleiMEiCG;D^~
zN&-|3tKwUWHC3I@2Md9_!hX`i45a(c>ftC&?o6GtL4i|Z7@DDtseB3lf==-lt*dLy
zEYsNrB3TSxNhR~shx2V+-VL{3KH0ee6ialvH&|Ss_leK~%hr3^%1e|HU<dk{;}4r<
z+}i;qkZ*rS5t4(TgR`L=gQey5AI9K-r(I-50?DeXR6-{>epY*&t@-ihOARky>CSW6
zw44McI@bPUhvk>ec{m-JAvOa~T5pM-{)DL42pztJAGz<Xi>sNxL`|Ki^7>M*4;-M_
z#aJ-P11C}AW1TM9`Q|j5T^s-{LgfnOkBn=6w^}b|#BI$<U7_ax8`%ZhjW*qQl7b@i
zNTooWGPtp|YxH(C@IB9f-0_9t#S7OymW~eRLb-bFVx!6zgE5gvZV>dv$llFu>C1v7
zD{i4FS6!Jdg}#1l30mF>@$o+DWIwy?AlX-P&4Lc;*0~ai_o?D-9v3OF{ZRj!m{EaL
z+fV8~en&wCgZ65fY|xq@Zt7V2R;Tm!s*RHYX3q}NL%Ja`sYc;RU@tlXp7tj_v8JO!
z`D1M8(=_DeHcZBjztR>=r5>{);#0$zL#xK$q7upn2;@WXD5e;ke6i}=P`MpIm)(m&
zkLg3Bd?nd;cR|D_A=p3L_xj2qQkINZi5PI|^Tk^+wUjkGFH3>e(S3*m<;p0-lKIPO
zfGehVZ6HFND-yO(SkkRU@u5XL;XbOo!2Z0s)}+Xk^XykiG9)1$L^m`RsMs(NyNli7
z85{xxyLOOrp^*bINH~YFdWRhhKPCU_v&D~cK3Q+?{}<5u`wB}QGV9=DXq2nqIJv#e
zoS01=XA|p|dDl)&LAfvH#6S|kT%(_6TB}l{kGAWkDWKlb`LuG;d5S*83<Fl8*A0iq
zxEHez#14RZuz#Go^PYmpaM`L!!Fd+ms<m@vf=hB^z_c*I@CxABfbP^xR?ha{3WeIu
zejH4QMHAMeBVvwt*cE`LSJ)Qv{1m|ZyH;=P@$`B8nZP(_^|9btD<m1V&fpH^C>(gS
zl($?Vi({i{w7I^%Iq>ShkM3Us`+s)RJUfw7RcY6=^s|q>u$wyTy=#1Z{kPu)?31t9
zrAdYrjL$*n{1Usvr!C}==%WZ50Jk?rxwq!f)#GZ#M)G_{%R(b==0j=)4s|ihY1%=Q
z96m)ktBhXnXv$~Ofte1f`3$vb=P<m^M8yg&5OLA^bnC0?5<QQ{Us_!S2+1=t_abRq
zfrQ0&>u2b~zS7ArNk>5SXq6Xf2ReU!7$qc(2T1;h>cz+`z8z`9M19?oxxGCxg3j#u
z5916&Y3JqG`0K}BKL&B1%&WO7p?y0Y)%w=f$t!My-#5B0t}b&lmrx^P8}~0`v$I_P
ztfRps)<z^z3USRqXL_BYF{@-#OEA@KA>61`Pnnn@o?^cKG?UCKZMM_kB%vVPf=B6?
zchWXT(C%k;PL74MUq%JARM4rB_1_Ucq8F;vvpH(4vs<(o&NN~?r?07)A$EWrWyQAm
zHTF1@WUlb!m?e&7Y>*{GHF9B(EpvY#Y63NP_WQWMKaQW2&!NE+pfP-RF&=oyAt__R
zb7wwa1g;Dn!uF6!DVj|X(ijm)<XcI*$SVR;+Rq<)1^n~5frU_>Y;+CFdK0r*m3ja^
zTE72S#%3C_$Kd07t#h7Y+c>&>&uQ(VV|~t*XTyWv0LVduYxz-CtuGimtFw>!FxP2X
z)_n178t+Oxgj!W(TJG<w=9<3Sxx5!Y{t&HBYinmK_MV~8D9)udejy4RhonRY23h_y
z7KG0vO(q7AaSJ1f*^pDh^+jgS3rFzLz;A;sjyJbfV&8met<DKHHh(UP6*HBZ@Xo2-
zm)aB6Jb0~2;HSNCO?Scvs1<2%%`bD*=?nh$p=QImS~;?Ql}nW|zU#21B9YbMbNZzH
z^XVZxR93~q*Gm}zl8XT%q({-z@YJdcLbA4*b`inbf)P3+<|#-{UhKWd^6KX9?~1|w
zeXPlS<~)~Z&GL)kiA-dRlP3$rK&q<xpFa6*J^I0OO%Z)}p1DTGNa-&Me6IQgj*BfB
zImMjC%S~u}@4U2eOQ=?r%(BO}IQ3?z=a%u=tfGXlX!_{%Xc`JN5AY=AL^r!)lrTkF
zeF%q48>k}cFpAg!=txpa%z?hkv*7CZ(W4Slxc-Mp9+jCK5~WQ1;s|&^r;~ejR?s$}
zR1m<pr;5sq#v=*RY3H*%f<JX_30iMg`OyU!sjzCQFa&Hy`|UQny4STTbV!s!J*p{j
zCnhGeYxMmdN7blgGcz-@stjT>3eLVZ+s>ahG<HRPaJ$}-bE5wcBcQ>_i=+@m!Cfut
zHIP!h)JrUCxS?tpu0#O##6)#tXXiDA-1KSC(Xc*YnmG{)&g@61@A0H)dys36XsyZq
zO!SmCS7uB+f0vxnS2p+VT4U@ymrp!%epkI3Tw&c1rX-~*vCZ?AMH29HD~AT892}LN
z$_D^_8t38V(XP>J|Jy63S(}GcS|Xb>$W`6di;z;-0nGOB_YI+6_l*W&6oviEjPU)1
zPd(E0Mu`5Ypv5WLLFQb_@3&kvgGB&&X_<Rzc7X5-m0q#R92t=36|YCgMVIi)!k?!z
zAnO60Na@U!bd||a*74g#JW(X|1c82A?;Jny<W}YmI+VHfE9=|Ku(ATevE@0Gyb{G6
z8CYr)?oJ(f{+&o7#R!oui}>ZC*X(Lg<Gk&)JDdGa`ylegnIRn>MP(erKW{)vA0oU{
z*j{|^M;$V()^<9k<D;=>2ndX{_LcYX>b?FaG4ZOJIh|#q1Dm7kEEH$&$6lZzqBB@j
z8Ys_}PEu;C98#CuNJD&YJ!MfRA_s`Ud%K^xSU1G4ibWRBHnw*jA-=tK)(#eXH*7TS
z1`gE1@`j6v=b?^|tGy0Bpu6#@-5vp@3vzk;{yD&NaybEFY`Aqnc?1RV$~E)O(b8hW
z8rjALR`NumC)sQI*>6tPZ5#t~`b#Jtu(9!@U(E94U<Lly^~eIFm=hIYbLE<6TakFL
zeYaT-9UZH`@#yk?3dGdBXHm~&>o%4?c@uy2P6-v@ise#M^3%1k!!ObD^{`vr)Ul&$
zp`J7q&Y!nGX3sfk1U;Qxosq44#ZNnwkb?JCeAFYo4+|`80)M?~rWGb8r`-PyUHKf*
zA-x;l0)Dw^S|zZjlW?M?^O$>wfdIF?*zk){tv7OI5N2)D!7tT);%~CueTz|ioJQSK
zwGdMSIL*16D7j?SdEH3Y)|1-f+~%z?p9b1tpf=gC5Y5Kp641@TJF)5K55wa-KEejC
z7~!b)J4^~n%DaQ9;5e6ZxvGU?R$;u0#>+IxD}VK`x=nB-Du{mgoE-E^Wv4C6V509a
z>rL2dS}p;h=WB9UD?`Tw`ZD<I*<67in&Xz)(@Br5>uE}MHg$gBACgxg$A0%Ek~)QT
zLl=%wFv*e`pyM0EfE^hQR$j2h6`ws`>?mw!nA*zF&jiFLDluNU@Rnj90S>qqs^aDr
z`s0mMSu6Fw4!En6`AHfuq31T1s_bBLESlFa&(W+l;wu^hEP#DFfI=n`B&o>gwts{P
ze{%@}r>(BtZpov`(nRp*D;A=KKQc$rYTPE5>l@rH`j`eAOfCD!<uI=##DojzK&3}x
ziViH5sR;U1+%EPHef&<^Z*&1xLkrdjo<84g;N4}+nS5mZtqLHF{(&EDSCs)$m?qhZ
zH0S;6_^BP%mjKzPCCCrn!724RIz7?lN#|1(6x1y-ssbty>mbn+reaAtEA`BY*Yf&$
zvES16lWnvtKg*=z`x1M`=B3qZ5oT48_3!o{vbSGz7l&o^ff!K4uqX_+#w`%TBEY1g
zbh#1@rraq#Ur?2ooO-MM4X#r;W!_}NcZ3MD1pEDaH1oj8M&S&SSQIm=p_ttNLFBsK
zqxIhTmh$=XaIL#No-`c%jmc(+BY1~X^7cr9!oZ>QyIjb$mF<V?r8{#Lz6$`dFce95
znIeu3=0H4s4yo4inKnoX`ujR>*HcQ~I{Js}>!nSpt*3pgnh!a_A-!%wpIf{_@R=hE
zp<MNRqQhh70+jHA-5+>F9{Wdzi$ArHZpvP<gvH?S_(m&4ac9mxpg%?W?%#w_ftSq?
zWdR!bi_&84>gj`xl1<0dCPFIfpx)Iyg?j7YV=}3`Mky_-j1-HJ88~m^a-<25MT#mh
z*lp-GU@a~-rjM%s=x9yC-*m(Dp}G2ZD;wNK`K>en#wu%W_>D{Qe!^=0D?bG4Hl|dn
zYSmEh`C1a@9mSvQLb5b!BFsI!;p!)>1u$`;WO>_*Ep^W6mQ)*jB3iE!->wK#ju`lg
zlq40$sOYJ!oue$uH98s-=cBBJSW`$U2?nJh`IUaJvhT!fZL2P$-|V!>H<Qgd-)L*5
zB`M~LU7$!k2!@0)@t8Z^zkB&2&>Iaq7Z?c8A{V*#rTeIzYFVjn`#5-8WS9SCUEF%w
z@q7uE?0#4F^MDN<$u%{hy`ZmT;QP&3)c5<P2%GGkq?BLPh*^qw2q>}K82rTJW0nDl
zkU_4I-3QyK!9fLvPT!9ig9@3SG7aAueu07@Rr>5dTE=nKSP<OXbc?p7mFOHec=_Yh
zircL9b&E^_FmIYXLL2~~lubKR7UkF_{#DQ!5me#^Wl)xynMcw0l?Wm!lBZ9(TF)FG
zH+G)lqkV4VU|CtL*i8T?p`ZvE3ji0RAyhDFA^`wXWhhi}G2JZlp0}{ECoVMzqfNse
z7;-J${rF*VrCvC<c0V>PgL3<nt!BB+Gvm;}?91jGX-B2wBL%Lq80iU*(-Y^^(&Wm8
z6o}h<lIjH^TU++p#+KmEE9x-OjX+oc?alA2XtMD*kbqafgFL26I~g(M<Kxm<Iu(Lv
z5u9-S>}6HkMmn0Nhsz=VO`Bk5$m8ER!P6E|&krZ28N&jXX8<RiN-9f4sYr)ny&m%<
zB&tiY<VUF6kJD(`iq2zh_T7=!^n>cv>QVjML_z<aS@E%P%<gt*6Yf!K;Oydw(<0l0
zlrGyQLsO4^N+2@J@5gd|b9VeLN7;VrY}CURq>fdji@e=0m}7QR+Ly5`piKBXqc@G}
zbBvevtrAWy&h?v49!n1H4{-YHYa(PX>pt%gFg0x-d8)*KlnwlRY;62um5@4s5@T@I
zLWT{^l{lK3_z_UaTzza~sH!QP#JxzpNUub%vZ(Yt;Ik(aW*9C9*&>hHXu~|hkwV)1
zttxOKCSG`_Eq%(Bux=%_lwJ>xXPRYZYiog%@aiaCm8}#QBfWi26>Sn@Lk!L~IL-CB
zh8h|9D@`q|;{M6^J%4SsB-#6ghWIW=&4LqW*yBU!(mIpx7Csj|s9>w(tK#tTch6xF
z`d#Vw;io$9N!oGj$OZy0NehZh$&NOACXT8%CmMO!CJLc1(`?^)&C1yYp|MRL0Jo*Z
z!)7>b=Q;Fka`CkXSqv*DW^_o?sJ$%fuoNq;vc#UspJ$$PBlm2o1vij!G-VaB>#)3(
z`6EPiDEzHib{z$0;Nh3K1L7#u43AjNDwRyz?spt)UHJ;4>8{jbW~fKk;(CXTy=~!z
z?u)HfS`7W*!pnuSHL@!u6HrllF_Z_wOKxanvg4H>lqV4o6^VRXX$2=H;ksOJK?`FM
z7BL<iaNTZ^A?r}NubmJcOvB85BI0MhdR6nh9{LEfG)|5oBkSJ)xl;vDA6l6#>Py`|
zm*{Y^wYF<4pHrj;H@ulFXzzmH?~b9D1=UtGZ43hi!l!@LK-Eh4<tm}2tA||@9QOge
z(8R?TZs8zak_r+CcN6tC8f;`?aT3PtbK=y{1eVb8w<3ThHf9sWoM-lz4raMJacoZm
zs(yQ4w|id@B%C0HRUNK{>4VUk$qv>Mh7yZt{f^ifPNb(rwkUj&dI*}1i{!^mTO&n2
zTauBI@X@fR$MLJhH+pp_F`;o~u&l1G4yO|#GFtQlXqgchE4Pj#G%&b8$Pr|}wIu{}
zDMA0#^SoYrHdxy^R@NxIKOSZILy9hS02qP!$Fm$qFD&>ublY5-PGMtdEeuTTrv9De
zx=kbKbYuO-nw(2Rc+xiB!B0|aN!PVAaC%o_^ZS%8qtn^`;VRF>)QjpDxNO<p)<Yb5
z5WiRb?mMu*roq1!I(Oe3-*)DzSCoKCp}AJbG<K!y7-6AIH>}L5GM8Z5GxhW&iNsz<
z$ETjVi!WPzZouI84IMqh=3rKhxf?we;Yg&_P>-7&i3z}s-9yWQcD}bea{u)n!7fSQ
zByH7{gLGJzhRA8B@<_^pE}k;S((3oA{e23g%=g1w+SxW=)j82l^4c?9m7sdzY1!#r
z_fNb8@3skZ4c?ekT+DPen_^jMOOo%fh-oNRKdRwA`O${g8)@(XOvqV9Nv^M#h;4ev
zY)Yn(y*?RU?ofor8t&gLa<?qu@w5Sdwb|wm__ig=a#%7wk^s%j^hdktJ`*^!V=N3l
z<Kb;k7Nt<|Wz2NX#8h|LcbE1#AA0jdFIu^h*n?IHP!lw{pXQjMoJu7Wzr)I6X;L)j
z>*syAaUuB>2j=)FTPQDGD%)}`viy<~9P%W)p+F3+p}xjWz(V)>yhBGeuJ)+)v=TB_
z@<|d!wlMd+XUZ5>$v=Ez?a29DV`XK-?NMeDRv`c*D{PEVU2aVju$u1s$cpC7iWTop
z;)q|e9RnH6)!vA0#mE@$%-z*r<i9!T+KeKA{l}aC{Hn)oqo4T+?6Y4@w4;eMD}RWB
z6pcJS(z(+i(dnc>DU}u|41}+p`Wjs~80*{LA2p-K4jT^r1}b^p(2}6>w|B9CUtg7^
zB6kP;NX<R1Rw|k%(oUc4XT3su|4Kan>Ur^}oR8*g$nn+1Z&EF0bWBpsJoKWJcIA=1
zp|ljw=Dw6(Dy}`;#>U3e%0fbvj~9}CEJa+uRB!O59>B6A2||sXcSBgG0nZ9(hUU4I
zhK@lG<Q~FP#$x%q<_;dG%Fzwsi4vlG%PBn6#9u>Y>4>r7WXbqep_VQDa_xqtb7Hb~
zIO2(wnT`oqI2K!=3@$qF@0TA4e}A}^{-Rul^p%xE@khLrD^1wO@v$e-1@f39TVL|8
z7-)8dn85D@qJ%U?XYrZYiY^s*RPX*g9OsCp6j^&a=S}`|mM=DZcS&bn(`C1}*2fu<
zYHcr@noyyC2qM3|ktnKo>yXQO@A<2GZ+|jJs(K+fV1!hk8<PuE2-;plUAdwr81KC8
zBj~<*bcW}?*=n~dC^EwRxX2Qjqvw)GHtg9b=F<N#^aNfaQ9e-O<PsM7dNa1mzsVoR
z@*E_}PzeH&TWQwTx?fKmkea;b!-`Uo^_ebNJ@!Dr=|x0DmzS6K;9VU}E6(4$7)rL?
z8W5Ui<<a1zN_{S#SFb#;|HE-V#eA-Ls_;C?@KANAAY5+&uwH|qJ|Kc8Ab|GubJ=E@
zkvcRhr>MTJ*KOLyCm{OY`nz!LnbT%5ai_E|oV8uoT-k3Qk6t~#jllDKsy=&aXg`LK
zMP;d5<xMJi-Jfj9T*R0?G;eN|xt{3g;v}{ymHn#IOkvH`d?nVB>fMb?ubBUk<OX(p
z)6DS~6FbB6RPY!Z3;FLzqb%U{Pe0DCha#G>m<Y!Fg_R2hHDlAs%d~lmy-!ne%$~8d
zctilj3h6-Tu(9Hw-nKQ~z<819k&JqX@<3_EL5P;B*;$$@k}BjTclS%`rM9JgL)daU
z!{|j!C|>iX<z)R!A*(ykvYLsDQYVZ>jeuPcsotc4l8~DOB2!S=HG<;%#o*O1`FiNi
zxkGkc&qheK1YIb^5R@88e$J5ai(@gXa|);KLf(RW{@@t<D$gr5NeIuYj_>>-!?sWa
zRaqu$GZ;@%JRwZz3^ty}58?!9wyh`Mp>p(;l2rtbwE0iD9b&lC>C`SjO-!s^T=L7x
z*x<Qh{f}#Oc#Q;O?xE@3>kyuFoPJ+fe|m1z#D&t8s^n(=qf_T;S7&}fT_$bR$r6Tz
zdlFAiyO=og5PHsdH^~Y~!?!O720LF?osT8F&`_$FrrlC4*Zvhwmd6~PB^@PXtJ&y>
z5Y}H@tepSaL`Vc8$G>rXr+~?9LC3FZ^z?o9E(XN$NT)?NWQb3gE|0}8!y~OQG4;yr
z=8dp$?>5!h7R=j7Lj?g+jLoV;K}^-UJ%5N^Q4gI7cl;^xy9s*&PL?DHKlNwISD(;K
z{k6{8c2Tq1b^K7+^=w_H|AG*mkmx2Gxv7NQR+(N8S2%rKDcxSB=Z!2t#m*}thAehb
z$e3Q_bhC8sAeEK#;o(6jhuZVvEsFTD$ih&lx}AJc-+Q@CTS-(sRJ3@n!o<P%F!#me
zT|rz|@&lFmK}h8D!se5i=kCzkr;U(*AvV(;(QqE<h(L5&_}!p)Zev1}Tmw&u9@if6
zgu8~QHVHSKzC<Yy12RL)c64&`sBQ4?K`}o@qSL-96_*=>6CDedP#bi>bud!jKC@rp
zEpPQLJcW=AW5xHnaIlEAdoodOuh#w-znS4!C|`mHQ9nO--`*5v1_yjtSE^~oZ*P$)
zLKZBh(pjb_Z=qhMZ^ijHJ=!imn5J6JQ6u#WJ{I~tOp=zC*6H}DuCb&4DyMZx3cU6A
zcZ`0(%IrTGBuTF(Fs7Z}?sY&Xb<rp_cs|Cpr<wu6?FuQwxCC+*LU~9P_${a`^Q^!}
z(`%diW{-_#!)){$&%gZo_*lmwSTIQuwQS=oJ3H4;0mdC`U7o%7_pkeP`VzTow}P(?
z0L9_uf;N(TjYFP`bF5ck8DtT%zGg<)|9+(OoEtogGd|pUKIT2IK1V*eL^pCXk(E}K
ztveuPLr_$6>=l#@6KRQt?0t2C@#p*V1#1RKaBL6K>Rg^+cvGbC_AaFtw)WdMj#rNG
z3yGO&8h_}%VE1$!uMXKM$x`)B)hTN?33B4&ibR`t{{kP?c=o-xdGR5lwzmR><E_g}
zU@_PDrGG#|LgCy2-yd&#ms6du`dNnWm}zb3=$plUH`_DN)ipLoE^;<I#Fp5m@x|ek
zWa&E=JXNUa@thYCg@k)1cNXAjNy6&klx(>JmKV`Frfi<Oli<HKZ<Asdww$R^Hf>ud
zj?W9pV*B@ZJ5Asjb5vF;1f|(8yq-?0y}Q|RlRDoGEwP|&8&|=@t|&-2VB+={qok*n
z>Y`GjQ)R=vXO9<~7GE01(zNpS?LtoGkbXWUnA_|Iet4)f3j2@WULJ(Q!GTsjL5SD2
z1awH>huiaY)x2tIJW+nhMpn_!gL)|kk1II2e?RKgJ({B(<L(i9GO40)?l?-AyGV$e
zQrdOTcw(qv9!p}a2NyMQ6iT=@xionSN%#4~vs5*Gox5u$leYZFVlKLV7t)fKcxnV6
zT3b0YxG}z>@ZreTpckQ$A>FWCa$B<LxZt3B{LSElRjDJJd!L=KXj)|AmK=0_D*1#1
zBT-G~`!czEYHaP?HE|#~9=MnMt*+U9Wa!_UKg~m>@7T-aF~f*Ti8)e5!rXjgTblix
zc44$u0^jrG(8Mf*KVZq{$vnh7+??u41!M-V&Kz9El#mo@TdSy&MdC})NA*i~Ju$<)
z<e{Lt){udrAqFlU5@m%FefBtVdvW=(2va-xj`edQI~?9CRCZNVUIf<tFEz}u2UJAZ
z*BLDG@Wo7?h4k^aJbd#}AkyFXA)jc00>z72F9h>1Pjkzl=bgLKxCs}sT-<cx1!p{E
zM34_YKK8Avrz`i*qkjW@3yaq_R4mk^@bC?}q{04O=^Zz%VgbPfHVPX_Q{}#ZPXYe8
zxq@l`G*m)D^6~!+?as2l_3?ZCg6i*r70-~phxm8OI>XbKpOX|f(Cj{Ztf1P%ntL+x
zy31$SZ=p=|&s*3l8*LmlQv3Yn5*z*VB>QCli|d@|wQkRx^?&cKuL+k{mW!mQ_-GQp
zaIJ^nA@S{<@pJrhLeGuQGclvzHOc8iFEvBJEsKRoZ<+u~xegpiAzKVq9OaJ|*xSex
zZ4cgK`;&o7RJ|}tq9UOFzBMUkH%1zhOw~pxd~t@jqeF>@N2rdsNE_dt;bi9crUIby
zIiXi9J-yqgJV=RS<ZgbEhGdtDkhF<{r?%Sa-Oy1VozIPk;TRnz^Gvpdkm}Ov=8j&O
z`j<wWD-MqJ7XxjL`)R}>ehG+XwLe!h%?=qG67!XQ$URVABzyQ=&hDyo=Y>wBc!Qbf
zk<t?Jv{Wa-ceL}L_Ot!YRjjFeJ<PxPf5vd!fr8uxDkPS}^RhnPzsIDO$FTU#h;N0e
z&=j}YrAB3!UV`?)u(;F`Jcg`Z@TuBx(uv6<0V&b(nVGNx^c=(QdAvIA@{Q{Q(8&Y5
z<g;EFqq+MrDn664M}yZlxHsHY^4G4^ucc(o{jx=OTEqP*9>zR1PMuq(p&U*GQG0H*
zUBVATC0k6`yv|$Sg`~3`K;8k))KI%s>RD;mA<w<M8Z;mL%?EGKBev(f>Y4eI*(OsG
z*;Mc)F&)kxgm{-#HnFO)uN6^E=>sA!M9aXgm~9<AKl4~LG7UJWTI;-jeOTq0r8h0^
z1o^5bLldW<P+3v8I3xCb(&B<JK*B&QWSd$u&yPyVcjH$$-DrJlH6vm_x&<#m1Zv^A
zt)kq_uLZ3pA+x1w`4%4Yo^c&r$l^!?jjj6+CFQs>8`8O|$L0O}=T7)rvm(ItJv9ai
z^MpK;`~Bs?X~UiGjgBK(3S`>Z4nrnZJA-<K-CgIeTF9N)$>#FzZ*zG8AtYpe$!qdj
zg9;<GR#vU-oFK_6zx=Pv+~|~FF2FFVgh3_blYh;?(F9%1zaYZT?<6n76!h&7ZJ*Yq
za}w^_4F_%m#Z;P2<cmg;+prH~_ODSVVHo85Q3oy2FG?Ws(qM-x64`wdgHw{%XGM+e
zA8&lukxJewza+&K)~sYv2a-mIl_ocSR<{EoH_a8XE*kc=9d(@vm|^1fg@Th%Vkfl1
zdc1>XK^d}C&#=VFoBV^n{H&^210Pu%BVX60kq)+{KFwGPj2`E#1|Gd&N4yX(y%eB3
zmtO$KH&I*B;!xkpYjVb>GjrVtT|L*ZuUS}iv8B*W{TRY?pEZq7d=M2AUr{ipCb*vv
z`8rU~a`>roF%u^aPiSB31&nH>RtAx{Q;(H)!5I_{*a0)eIZOySyuRF0c)rV!$3%-Q
zo;zr0Z5@Wg2}S=807pT%zHZ%OV{40@-Cgc)ZPM;^>2%v<nZf;A&*z_g#K9uJ`RmWq
z>kheorAeoi@!_YRAy6)#wvghS$!9MMbefXkeQB@0@jO@G{2GCflvdVgltPXYN7C;R
zOGR_*fOe<JC`-9^;~v|4Et1qC=u^-ThL&Qr4j6Pgz-Aij1V(_sObSY2z6TgAxR6w9
zMrp>?tCy)&V?Oku579qtanQZPa>HPaFBCMJTg)x5;cyd^c{ol4)g%e=g_l6(f+!&=
z#u2l#Gi+^dza!`QZ3cCUm?6vTa4yt@F7Uw?8igPgkT@tDw7B_4U*qxlMM|gF049nq
z8OQmJsSwET+T6$g^hTUw64XhM0U`)dK}e7nsf@LR!V)M+5XBT{8q^jS*}i>`m!JPO
z4?TW~B+dv3ND{-St7+|ac;n4Cc=Chq<NQMx7$yUZlPDGQz{5|oa`q97bJ#S?TL5xW
zeZY=YaL3#=A@k~kJaO#8k2&%!SpV6_f#@W$%)eYNYj&P0&LFjv3b|4y-(0dR^VuXW
zAFFgM%JA)8N#KGBPsxA(-JK}0wY|;DuUuzs^$g9!9a_x<WA-_<bcRB?gvByu$jCAO
zFbWBB61krZ>bWvHe^9jBNBq$r{UM!$P3CGLXV07^jv~697R`euH?Q5mn3O^>BuP>R
z!-PQH<kq#TeCp5r70x~IB*Kt$>Rev2Qz#|SIzbx^M&(D9Vw?fxz$s?uR+yQ8fd1hY
z>!%*)u(N|yyR=)Ld$YPY$MVu5g<_RTt-;+pw<(uvlVD(cJ<9-J^$~!~uX80?URj}3
zD!pUb+B<>6!$TVNDzoz;y>37(iv)!-g|Z@!VhTk?sjL`w;o8+r27>|BN_leajVoO)
z4}^{_e#Be+9-VWODJW4aNe7yCt4XC?VR!$4Fbb$Q=DG0DB~G2bK!4CB4hqcAuc0w$
zqlv?qVZX=G;XbX-5wBgo4Bvi@jjessOfWh!%r3;FqtpwPDizjFoh1rGtWLq0T;$;m
zkR`|5M{)}UKc98XZ&mbqLlBZuCGb`IKw+%Gq0ri4Efgy=sHi}irVNLDip3&f7~-tJ
z*<&6UjN6$W)>^J!z52sn^zRc<r1;4{_b}yBgD?N#O#&Gp39wQVL;|NJ2n|kg`_>M_
z;gDLTk|)Vt8IAQK6o>Cv2mva8*eNCRw(fZU1tC1IN(g~55^X&H$>xgMG#&Bkx4*{C
zYcDZ3SECS1ic!e==_Tfu7pYVmXbp$^J6yT^B8BQam!9|u%GD)*3>K_2xtqcGJ>8f9
zj#T~{@WG@60;G&kfnt7cj%uaM`sxbrd*X3SmST;e-|w-xahn&v{%hRdyvzMvM}6@L
z1{oZ++qBv(y1gD+oA*>g_fA8K5NVd!+1myrdpkXrmg)%M&?Y6zE1W9uid3g9nKK~a
z@~dBGZ{u|qXB*tRbC2c69!FTi>7_Mh=9Y=#0)@Chr*%ZT*FjOm*=xM~<~FjIu(%*e
zjo{;-{xrf^dYv{>1V|a=g_yVU{Sly0#HA|b*(H_(#nN1Z>(?$*Dc7)J$Wg~JQx;@a
zpbZQMBf9NFtk#HewK<Oqd?lU<*Lr+Ck4Gd%QN-LkXlwGeg1SaI<iXP`s5h?@r4Hv5
z5x$*eq(B%!iolwP=8aA6JpTd@*MAyOu6X5Xnjwk;^0~1>cxDC8V39ff)mPx0kT@y*
zkX#WUmESn6)+E}HqzOryAgslpFm8l#hEh~udHpOZ2#JbS=1xCIqzdSCKymvXgS5-j
z&pgTb#~-7%yh{J}4Xo3spg?_Q3F{;--{i-<xMLNT=WjUc)w@pQ-gHvtN6v&}e_TbA
zNbARjoeZ%iVi0ogCg*TYFw7)b?7XdsO|edsW!|t>2#pjPDMGC98`=1QD21;aS~(^K
zTi#9aYnEmF)<63^%FeTQw9mi#?U$%GDwsjUXMgHL7$Y#&XF1pmhan6@q9`N?#N^oh
zIkz0_-eY_FF12cbek%k*4v*Sc2T?L&*zd5qvfx{sQqgL)84kLB)SK1(&cFDF{Pq9n
z|G@0blIKVnGikd@C46;Ii+pU;_yH>rXu>d{UYKF!^rPH<^AFg%f0aSfBuO(0g_tzS
zD2Be>6iP|4R3g=yR2vF~2y3!QP^UOH;T6KauXK{RxjE+M=HF96-S*BdDo{N8+&P;2
zBiii(gTCYCmoD?r#j{jv5g-28!~DVTyvo-7Lz)L&>b0sLmFL`~%5+lIHA&d~4^9Mj
zAf;k?X_cF;4Kkgwv$IL7)j~K)qgJO>sZg%gSqTe7QIT?G4yPnpni5DsGUzcgvrMaV
zge;U1B4A^8lh?0o^5*p>s$gi8q1zd9beQnWM<2junlu^t;KC_iDH7xpVvhL>$F?}$
zORhH<@^8NW2Ph$U|D^}%wA&b~5e|&STca33r8W;zF&Yl&_j}AQ%;l=;d=JU5L1VPH
zK*|kJwNBaG+Wg@y_S;0YTI830`4jxJ|M4s6R3K%95D{@Kc=D<9TzUNfV|)_2yW8d9
z@Q_kD#^oF)mkVhoVw#*+IZ+}TUp?_QFXL{=3015Nkxqci&^jd<j<|m9W&X);{tqnA
z7pN3gDFk)W#Bg+UNV#4mE|y3WjdhM<5YpVd!yj(^IuAebEGwrUMTrn!KDGJ&?G?Gc
z2_w+Cf2seO0woC4ao`^)g^0cFiU|~1mJ-{D&f#kW?65KeHWi$G;1nPK%r9VabLqoo
zlf%{#`^^LH-oMYm;StTFHm!D>B+YzM|06Eic<{nwG~!jR-MPnSKKnEW+X<yoc)UmC
z@)s!;N=R??4LD=j-FlNjcb~0|AxjIZJo3oXgi#5sVSaIqAPm^JbA$cE9?kxc*`*a8
z!f^T4K0<0rr78<c7y0nBpF{{ePg|%H&+5Qme{igka}dP^N~JPFfG#G0W0VXTBqO>b
z$8hAp`r?vKEyF?A&$%}yrxEz@%3s%cax(sg8;>1yfZ*(z(|qYG@65S=TS47V{^H|2
z@$d{+{>hMgUwx6lm^?7cPlyBUG>(8UacsTt1~Y5xG(PwQQbk@aoR{qhj>m*q?aSym
zkx!~n`8c5vr9v`mIoxY=u+?P$V29@6KCPo8y8SLmma)=M{M=_QQ7jE9#3c}pdaX*@
zvFO7tMz`rVo3xMiu&JXyJ5Q-xW;7UaaCk^!9QDPsgmH;1P4X(6Kk4dZJD8jkBD{5j
zb;sm;F^=e5{*Ui^U`=vx!9)o-7F)Qyz32pt5=&YT#G&u&`-fwOwZ;oZP6V09H@Y`?
z9dqi&ZS8j(&Gsgz7ETieRkrrFXdN!_`=5J>XP<qHTD6Q52BR~K(TGX=Y3x@e5W=@?
zT8BG)=bK+Z2**PYoMnBfLR>5$Wki}8;sR8wRgz&sl4bOU1B?*d-#Vnz(LC|^1#Vuy
z&gXvnAM>k!{cq=LIR|)eKR?RLtLUuuRfO?W4&@xiY2v6zV|kUZSVoT$q==Ey5{8~<
zIU2S&+}mb;d6i<J%-q~8){WJ){=prFAuiL{oJ%;)%*SiBy1L5UyLaB5#rhuD-rnWE
z`LBM9LR`T(OBlqsQIXePy~pE^Z1M89Z=w@lb?S99?%mzt?AaB9;Mjv~?9p^Q*A?&c
zrqqO&I#voxAt;w?L}5sp4jBvv3<m=m^?9N&W_Nd=xKKgL1fA`%wCt$V>qsoga7629
zkDc8eg1E%o!V;@%r`S8%rcx=P#35EoROF3)D>F4NUU(!Q{pa@|mGjvci#_1@-p>P3
zAtXpe6xI2;pZ*jtz4$U4o7+esdFY{sxc=>?-y1SONb0k5I5fl2fFvCe$MFd@=JD6~
zT$#LrdDLpLx3~AhU-a)6`FZL0Qcs-`AxgwJH{jvN=Qwk&L7+meUf$vQm2EaQ_Bpk-
za-7Usi`5y<&xgbb-&(cS8&{jWyD?UtOCi0`LV2%1i@@p;-PS&bt$kWYZC?G(H@SOf
z7is3W^jJhOj{RKlLV+NPeUhfYx`Z%-AP^iLZ19b*{&OC9aDxXQf0lA}7U%pE?buNu
zZ^;?Y=~YUBQ%K>E(vOp~##fb8=qJZ$B``o|XP2$}U&WarI`dWQjg4Cvqlw}IVGvL$
z7Fb_fL7R*uc_00LpZ?I>vh)T6I^7=kHa7X{H@-=lX7AF1|NA2dLO%7$Px6&NxX0Xl
z#KVuS^GA2CQZB_4&f?gTC<ubQ{VH+R(P`~*>-r6n;gE8v!iPTaVKz2*xqE+uhtHj+
zP>ASsIy9R{EG#Wk84l_74l&t)Lg-MT<l;k5@ymbtKcUvB<Lq(Az}Y;g%DZWZ+;#*<
zW;J1<m>Uy&!CR81jFJ(_C}pT6ospxg@InhSVl+(2vJ7K0lr+Ahh?v-q`B~BirCd(Y
z@>HM9IcoLlJ9oa{&M$xH!;d3^5f^^q6EwHB$*$i)2yZiDaxRy)1|2})0$gXrtuK9r
z$CqjpPM^oAXfnuJ$iVZ<@?go^R^?MFLWKwv!iDVb9&zW^UGCkz&COfaxqj<9yZd|e
zdOcE|f)hNjGQ)>Iutu#GQH)DODndqzLM=d#+9absnI3}Dgq4^=t&Fplz0Gaf-7X^w
zmmd2dnFT?b_r{K^Y2$zQ@tB)DsC=c26J&5ERTz`|#f|@4$z!GO_<=VzT=rv=1TYp$
zKoEL|Ywx+{TSHR$zmdGvzsXvSpJ?F-kZ9gIY2w{Nnsu<wQZ28r-+mLvEF1S*eEsXM
z@smIGL7$D8#Fwn4MC&}03CAbMw_UHk@lAHNuOXb^!3Q3OC}iXI9pX5mTCI>R&NDkV
zj|>8q<`-BHg6-{XQms*equJV~R90NO@*=I~HuH;T5Yl3ez*#?ccswFXDv31$Z{F-j
zUyVjXqGFkYg8{ExAF*0-1S%!XhB?Plkc@`xY~1AHr4LYuVk9u^cd1va2&J$&|0}S*
z(rd{Tv3W(yImgP%%DcB%-ve2y+1kj6!UTjvsfbX;^m<KRedP+*ue8yl<Iy+Quiob)
zA9;!>^wrg|-y0lXt7A&V_<J}BA|I*d1g%z=qgIDFE|3f}!mz;H>^xG2%+9S)C{}15
z?qSW4=I$MYSfx~|&^*}W?yXzw9kkf*c9F%HS}A63ZjorCLl{69N@iy2^oKppoxQ-q
z;_}4MTdCu%&~xg1rN{^66V?O*K@i0(uCG(AR7eH`T1QR#ojvC3bG&IiH_#e~6Cw4v
zIX|XwIKUWRjn-QGpHKFYJkU82NZ!43=RIjUJVusg+_|;Q2S0p)?fY-gIVup!kh$d<
zmR4qX=)nuzxPG56{J~Yyk>S>j4c`C$M+wK`88@y}j%@{wKfg6O*U{#r6yfg`k@J_N
zk?3s5=Kbs3yLE-I7%@|yLrKlja)oo}*BA~G=I6rWinxH?txY<eF3r7tj7dQVX6hCC
z{SG&-d=<-x_dWg7gi#&q5@5XF`&L;Ljl~<vU$=}qGH%kMS1M1a6)5evd+T}9;XckN
zoY7z$tyYsPODR<<2y3xYOe&KAT068IA*G~PEHKlk6Gb6ke*PQp#!~;lC`OBX?Tfp#
z+WT~~BmU9f|0<yf`OIf8VJ#Rf2vqC^F)~Ep?Lb7}sYRWxrQPiC3qSu6o_ykc#Ki{h
zfAVRP{vpCTTKjv<RBEiOt&$nX<*Ot1w)d!2>io=K_!U0+>7Stx7rcL_%tc%iXa2EQ
zox)Fn@pKMkNECUY+gKo(Wqy9Q)*9`gJ#d6&f4^&!k)|nWmQJj2q>`vHaa-heCV|Uc
z75wKOyHu<$FTZ1)>bDit)v9#{Lrb`{z=h9zlpDJ{xZa2$Z=E6}+8GSv+L5GpZ;!iQ
z{5ofUVFocXkH{-^Li!4ykO~(FwDHlaPyr|bE@bo0K3A?^=9Sl8;mcqBBX;&T(Z-zE
zV14g<l%<r*Rl1!Xk&3C6XDG&HY-o@KXemjpZ>G(Z8dzu8+1=*o=#Zmsi__<yWM*ao
zV|_c-mre7m*{|yNY!_K_vc2;`z}Pg?Sg%|i&j3A^Su*)s9|Q{T*)N1w(dH%Sk%58*
zZTxF#t8?Y2^6dzjJF7X1kPhJ-3UE>&Z60jeW8(L_1#J`7&jbwGr#R}}V)b;DKt_D|
zORw|6XC9(jt>l}!&9mcUuRmuUNKG;v(%Ro5lnNmX?bZQ{Gc}^qXV~A{#-u4a(G0pH
z0QPovY4rzKn-YbRYPpKUQYb_q62ACH|B7Gy3;#2O1fw(0t&=L}`51rWCaJ#;gugzM
z(Ey_j-+J*CUb)`ph8(a^NLiY7sKBptGRnAr=L$B1D2|ANkanlz#WX_X?Q@Y=X>#SA
z75UT!e_^bzzqiB^)@f3EfO3ln=P$ZQmgtTQFTH#nsmp{^@NKG)t<5HbK|-Mrd2X2#
z2-HMf2*<%408X9|NV!~?5^%8J!I(CcQk^&|GB-Cz5Gbx*ew~+IyUg8txB1A^@8|6L
zDvh~0(k$fu#yvK-cjzUK7ryxtk6n74d)GHvUYwy=P^iEHhCqdEZ}s`mGv^4y0If~l
zIvm?*;Q3t~w_7cqJdXn<6+uv-(;L$14XM{_v<~+e3<k719WtY_R-z5aVu5OX4s8sB
zLC>3~M^R4SILX@*lK@ZTmHaz*@4V|t*}n_j?uh^Q|KkhH&Xq~EL@H23bXr}uwzjx%
z;WTkU5J#SNa`XB&y<U%6wLED>jd@Jo2I8dSft=hca3XmX8mR@tq{ofR-{J0^H%Nv9
zqC%O#K%-pZ>GxgW!uhk*swMipKC;^bA#zotAsO{CMss+0M5S6lO37%b2@4TBTi1BW
zN<R4P&!NK5duTg7;l3&n<sFw}?&EQ%=GdYF0Wu89l3jK;ULte?jl~#&)iBD2B<aX|
zS{fKn69JGiB8)9z&aw0RhGlza_dP6{`FBCkC_g~2(`PU`WRz;w*UJ3GU;a3cT{?@k
zOzf-twcw|`NU2atv2yAGe&w(KZSKDMB|iAR4^S!8Kq$IL+xc@=(Aqzw)!Zj8cwxil
z&IT7BeTtv{<^PnWQ)m1d*}N^_#Dotom735h{CoSvQdnPP3S_{sR@+x<v+)`T#~>Aq
zV$c>cW5|p~o79^ck9{Ze=AG4^VsUbBH)ebvNap5diQ|ajaP$M8@wXM!2_=w9Ae5wW
z;T+2!xWv|%zl{=ZOe^*k2w!x>0YN6%eeD{hvuBup|5NBngSb@Am4~?qLP%tIY-xg&
z9PW0xa``HM_?184i(mf=Njl=#kkT)TlK4l4&?x0SZMJu}kxL<ku#6(0Sge2(NEe{O
z0=;36-Q68poff@ekM3Z^ldBhe_QkP_`MBNf^0w;nZOo)KYfr>#QZRXVjoT}eq3-!J
z%sSs@o!rJk`Yl4roEWTFSej>aw1=^H!G`k#hJBx2<=iYE)L99bF@Mg#o-s_+bMJCY
z2RN43Bf8NnrMnB9xme=kANvsh?f>%+`O24H;irE3Lr5iY1bK-XZ!w^K+gVs2ED!oE
zX6jW+r2<#qxXR39k6@D#jbat!9HUO3w3pE99Wuy9w0i??-`yfjEJpWnRuRS$tutQu
z`WN`vPy7P2vkPRYA7d2+fnN{jIUK%vW>A2LVzP8drZaxxQ=j2u&wY%qeCZ4P^I!jm
z1Y*EU&Cdi)MhUmBzfL;r6Gkz`VwuCEL*Eiteli8lOXvxGhUQE4!pru7)s+<rg#v@Y
z;N4!h9B#sYGR7ih;O8L{G8_$vMM$VjexdojA<I3yv^GA0Q$DDZ(&hJ<IPvB0{V^Ru
z2uZO#L%**v+TiSfVku&2X^9{R==XbExbOf*C!9HbhEg%~yrf=-cB{qO>9d4|GKE5c
zLLs6VS>|UORI4|MVnG;yKvFF)@sSUII)8BZPkPeSI}mRj`R8C#QF6{9gOD(ch@y~W
z(8XFyk|Z3o4zbqT8EI=N7Rr<=4YD+)-|JE;6^WzBpHtlA;o_CA4k>&-tDWKc^=t3W
zl7F9Y&e881hQq{n8H5*`wT@EWcx{83+AVs$5syB0hOg~iW^cDeyVa#qE;2T8M~=rt
zGHGr5+16v0O3pnc?`kM;+Hv{S7dY6vPrtiMr`@74vqBV{;@s&qmgZ_49_=IWM$~aJ
zCX8d62M2UI9gNn@&eUi$>R4;Jd+#=#Zi_-zLMg$u*T2Q-^N(@*^rPN4FrPXwp~{@t
z-vLJH5GQPIV(y9e);YA%Y;C-WwSuU!PSglEYTZN|OPXpjqfs)*tJx;EKL=qQDhP<;
zFt=GUpbgu*`|r|9|6o{KIm>K$fo?k?Geh3@@Ol2`-}v*aoLay-<7dR<lX}Jhea@v1
z!jQO7;)$o9<IxLe$&T)^vw4ejIHFdYLCA>K(Sh%D;K+>O-sW9u^J{$IQ-9f4-qOq)
z)5|=)bk0wd5S9rE-i`fYb*^mpd!3LzD)TK=NR1EZq!3gqC1x5m21mPO+VADwBT?sF
zx;*%|7M*6^+c3-Wx!-vLr$WdT<53h4hT*#q)QQ}YUxJ7VEPdozx*G@duiZf#pVJD)
zQLzt%0tcp_vHik}l$PeH&MqKa1Bt@rEe~I*3iDpNLys)iuiWH2uY8j)e)W$?bPB?c
z+E`gy@{=F5_SwBP#IeBGjD^KzgpzD*Zx99r7G@Un^EWmW6*Lc;Y;0}N>2*o9p;Vit
z)>tBnLeKMX<m#ufzsxwoW0FN4w-3fuJBxGUF(7u*Ha(eXJ8SbEe*WM2&(tJ4^<G%B
z3-fHYj(n9)`-{w)<7(f8*Mpfnh{p11DShj3k~ym5mhZcTZ@=^k&pq>tG%hc5<;{D1
z@B@!={=y<({=(~g=-Ee^o2%!Wq`^5&6a*)^szgzN#@sx6o0o~hHCmkxrP>_td*lf^
z-4^9Sm8e)I&AftYd;1=n+Z#0Z6PgD*q*@cif_hz1C<Zvw<L;fS%rs_@SaeRx9&>1v
zQWSEPn{{}9xj~=NaLDOXXK}*P?RWXy@BeGIc5YKHKoI$nQd{?Lvb%kkmGci$D3)-}
zE9IiN2u^x?8{vmL`-fG|g+ySCV5Tv{%*^b2D5!%>_Ym<4)((+K0vV%3gmyzL87`lh
zTCPNV?71h2qQH;AAs>$>)ub9Fh36uD_tb`oy^Itrt)69eVTq&W9<5HFs2EbK)o3&t
z%s#z{3`@N4u}4XV1BwN252Hszq!7$BW~tVzEX+3O^?NML*XZ^;L}7tg2_gmk{(ztV
zg}=zD^>Y}~$^8s(ZGVm{MfvuTw~F(?vs9s2E&+}NNRx!YXux5s1=i1wHBL~f)F>9J
z=w!gK->1>2AsAP6a^B&|He~+Z{r-TRo!uYuO21tQz+mkVX9FDGH6IN3?{D$>--o+5
zM-*a-F@j!qMDw7_`ug(3qjD?&@Ozl`_kS4r6xtm7y*MGjdGdMtu*uH78>mo_r72Mu
zP%eb@yGOK=KBM7)-Mt;==jZ7cI#jD=1hbSY6)Keqy>5@;U<gY32Yu)T4N<5V^g<3>
zO<sTfo18iKxW7N$acf-W!T#|v8-%oES<Z_NeL!N3Wi%SFeeW{Qe&Sb2(}YI7!LQx=
zKVs~_3j&lMByUbs`XHy`xR5X|VVxw+eb6?y-=ikf{|-ng`S1rn$*bSk$LRs9bomR<
zJjdGFoVPs6gE!x9^**$L3J3xp3xtIdaj}FDf?+yClpHWSJIC(5O{^+YsmxPJ1*Ot1
ztLN9rtfrg6WAFbs#<+>t!rSk70gkm63s~!EUqL?hQ;(mu))!$WMJFH0=g)78(VE4D
z1)}wJo__lMeD(AHnts|ROETa2aT(V6*V3tGG#rv7$z*h;J|V=@X*Oy3hDtFz)1cGo
z{=nz`Z3lG-Um+41$i^(EKK?96JNu-q4r6D2>)fO|<&+>qFx=W<<JH$_T)60iWF<W@
zF|ROrGf1xz+qr+h?)ENU`^HzuOzNj7U^shbokt$J$SdD@g=VwoecY^LREf~qQ7D#K
zUR}c&%Nti-qgt#`tIiOFf^0NmYiEm_cW%<@_XxrQg;JT+N_O_zT&UKFLhmo9oyiA!
z3#`pQ6M6P%$AWiG+>D+(WUU`VW6TNe-I$BVvD)n|3>_g!6R)g^!ia^%B}VNIgXWg!
z(iw}fCoVX5e39u)`@k>1AbnfIlT61I$9EeyZ{EfYuOL*3{jHQ=|L;FfdpAHB!M9($
z#!vp#hkV7-d69?3j}CKy_ieVEd*BjVcVA_5>mI}5fWcry5SLk-Sw%@jV|JM!3~)MS
zZS6egTSuIJ;4xl!;cL8c{T|<WV+*7u%QQk-hP@s-9ih_%YXwG(eerxyC&5<*tR-_6
zYc#!XmsT&MTB~sJ!bL7Ud5JG<-lsdtn2G)R_q%N_zy9q!s4F6sB+F9j^%_zfZye5n
zwPQ|)$ZIxGC=^&-U1f7~<K0=T?*U^IlnP1Gex5~#gtAPk`xv3YIRYto_SuU(`NYG3
z@Y7~OIe7^F@ktU6AuMCBM##BIQm}gZ0*^fT9H0N(ZzGH$6p~`GNTpn*T&-h_LR6|K
z1BXW&NMlKpAqt7I8O1;%qB3z5q0<q22YU?rBb<RGgVoa)`SefyIf5ua+P4N&{@A{2
zd>Q1T0O^Bt2SF4goWNnpw5HqdG8p#gv^od@M*Av$wLVJ_#-#apuu^IHE$4Quf`t=2
zTOruq*{1cLHXR;!N@B8Xgmqa?R|t_JU}JNi{T)N-YM4YJMS#v6o16Pwx^$5s7*}W9
zi7$KHepM=0BOmis#%pnKuuU>Nq7v1Zo1G)eMkon`L66nNS&XTp6zuG7GCQ+C7(lNZ
z5*Pfm6Qx4t7V8|gkJ#M2PpjQwYqy0?k2rUJ9b*jFuf9q4GYL^x1UDXe=Q~wOdGmOa
zWS%3dJf}HxAY|yzja2k|T@IUfaXO+@jtE4~rF9M!gcOTqka?FyiM%DB`xH8dux_$1
zjFN<Qr}OSC^bdeIsB`y9h8zw#o84zlSbDEr!G8S1h*H5%;0Yt72=e49Km`#hiZB-T
zc8}P&_Z@EBx<z0*{Nkq{0V!ErT>%|1?CdZcG^sAsIcy!VvA4s+ANWZsb#D(MrJr8n
zMUKZ6cRvBj@4Ob7PmOVm1xqq_-tfVl0i0KI3&rVE>wNl?pQ2iuWpVWs4_<tT&wcK<
z*>AU*9SzW#pG=cw87A`Hr^Xtz&hp^y<aM7L(`$`R6N+*Cj-2<m71SZ}YMq6|X`(P-
z_K}BJ`OsbZUw9syc@H%$d=R7%IH?JQL2ALlt5<1%WP_!(hcJ;vkxSK;Hw;EQZeG5@
zpf}?F&K>rT_NY`U7?W}4+-ZLLr#{QqpZ_|Coi<74Taymxbu(Uh<0g+j@;LQ+oyC<E
zkdmucuCo35cW50Qkq(C-1?5_mVy!|fV}vxkdgC^?cYlkIedHrN^MUtKD#ge=7(;ra
z70;i`uZq0yk5mLFvV5Hfv?k~Mj*W={A81M6v-h(p^7g0`L@H!)et}>44}KXbL(ZIk
zfU9r5%0K=`|1G0TW3|svoiKUj(D&!{C`G5K5A-sFGsau2<YEf282#=mrHoA&jt(e<
zixk6oZr#2@Ni9*3Wv;w&k57E!{dw7atVlh!h;b-{a8zn5937=}+b!a#OqymSNyd%a
zml^b0TsV7%^AA2kzuV&S8?Ui_&}M|?)S2^S)^YvT0m;ZQKVLvwsFX^K21BweMajUc
zilrZ0<PDH>&!W5)<g6o0h9sG0`|38<jCkykC%Ag`GP@hsDHZ*3n^bf8^>6c`pZGb#
zC_+lbU@*W>&Tu#<CyM5AfbZW=Vy!nMK6~a2-}=@!-{n<1MwSf-l>n1#8dN|WG#HpY
zoYPpVSy`XqBOkp)xm-Z!RabJe?Xg)boD>=55GPuYd37udi+t**{vw@rhg&ayg_&xZ
zVi*$;5UCgu7D>_}HXHb<4OvQ_dXo(LC^G|25a6&nW7zA_>vVCz>ZynLmB0F*QK~lb
zN}@Q?^2>v^aWFBqAMm>r2+%eKp$MV^W6#Mf%joy}^m;w|gCQk>9|t4^^~M4!P^5`h
zc$Z2g|32caev-$QN&Xz&zH{eCV$i&lA&wE&W_T~o0D&SiL!?Uyl*Tis0Rr}RTNoo4
z3uvq#wgZSs_3otFc--oQV-9kX_Bd#6Q79^wmX^tqgez~pNwrdAsTwg#HI=x+(%iCF
z(pxAM%Veo0jtf|q5yvs-&ON~H?k?NA2OPCWq^V=wIC{N|-Gke7n)@uTokt?_ilXl%
zIHz)jwDpOJ^<pZ7#$-dBD^ac2XdUh|?CmmY4~Qbcwd)q0IfQal;xY?MD|z6cCJDJz
zl29r%zN+tp!U)Osev{$o2XD*$u(42mgrK>>2lOohHzXC3{_QP}uHB^m_@l(70zq8y
z&qU);D#poxE0^!`x!?aiUij7v>>ccqr76Gq3(t|H=Q%n$WNCJlpcG<UM%H%RynBaM
zx6AC>!z?VHBMKBI@0h9abD2D-6B4O{aW9V4IJCC;7|uM=LK4LdMBux3#}-uNwp$TV
z5HV9<;4`27c}}0Xz^kvl#J#O83Q+{skm&?OWm1(F(wmq&=Y40#2E{h_sl_>#7U#Kq
z?Zyv$-rrtO2l%Mf55JaCj4Xz%efa$x-nh&DtxXt{qH}K_=@fy==fxiOXkEXJe%cVI
zXj1tT2%NSw_jkE<<2v=q0!Ph5%EgGd6tjNz9DnVv{S|)i_kNGP-F=dgo@BU@l4J({
z!9V(4n)@9-`H7FPyu5&lD?D`R{hWDlM61<e>;4^__1R#W4Ztcoy#cSjdX<~^U*P`6
zCi{CWp8MFdEG*OjeVjg%m)}nid3|d}OjIr7i4WsJ=0e(IA~29@fyw6`P9AImg1{T0
zo<Dz{x!FakjX7p!X1RLhO^#YkQj-yy0B4RBwOSjJB*|yW*+~V+A6JIU#|&lJyC2j6
zjItJmaDh0gp$0`V+oMse^1){wAxSc#I3Nt7{NbvwMoxG-Qb4u7#OVi~=7lf+9*7K!
zMeBqx^zFSYO&RptbUJNPV`<FJvwv_vuRr3#xwAZQ{w_D~Y!gMZ9Bg$c7Rm%l0pW4L
zfnPV3w_X{Kr*r;+Q>oOr@W^TUy*{^Z-=Yw~XFl~=e(N_5Y4!JL#DImZ&FgG$-e7+D
zG*KMWZnyJw@by~%cH@99uYTv%TPY>$>+6$B(R%?ZN_pnP=lJ3mniva81%#^PN#oWL
zhAAI??lG2E=23wJE6B7S>p3~DO6LNhoD?s{{*EWOYvb9^jm8{*{xALt-)MvgJ)jsD
z(K^K#L9tw9)H|ZzIlvk35v82Mq!~Ngo6OG7P^~vmMlu?X7z_qfiV+W9JjdVsPyUxI
zEuZoHSeGl-$Fv3kQmFhpZbG2;Q%IbjHi94wV}c;`p9PoCob1u=bV!nnV&pIuP=fl*
z91a+b1_%kIVu%nYsU0U6hC<}uvs`%(n+_jC5CsuQVtlJK2=Q&M7^izUr@<LdnZ!`7
z1}rSpCl4{LHO>k|E)RqJ97yTgt;+LEr5z{lK3guA3KWYWgI*gmJ4dhAXM696Q>V_c
zFu%;5Ti4OrQVI(sS|hAv)J+k}&~-fuwG#I?H!%)E>F9O(UQktu==S!#C!Moo$-vWs
z!XQ_)rg^|`atkV7$v1fwQ7La0AF2pcNT;<=IvNp4MZ42wFiOZWO&CaKme*NbKh3?n
zcRUZ;yHNXC#@6CocEX%nu)Vwcp4kt+UsQ@S{N-Qzb2K(zr1_PUruFkwNnq=RZ&7*h
z0#UI_5PF5Fj1*aB`P}dP8-DG7_=g;I4k#B3IHxHVB+oqkAg4~B<=&mUY;0^X7z{XU
z?sM2}VS)ZI<8#0DdA5%NKJ|%@v%EZq6lS8T7SfxWi^vDk);V7kN%N^u7*YeEXv{6s
zm|4a-Fq5^gc{e9O8%>%eWTTYY%nVOH@ieE_&hXmn-{FndUuM`pVlWs`uhhL4qxHfo
zsT9&2+ZT=3c)Z^)F3i6p=l*R5b;l(`ldC$NbA+=E)}DKcqk|@@*&WL}{Z45mN(hv&
zpe3za_ehhW7fS?UvWecf`Ua(<z?wc%%u|dDJaG0jg-XQV_>cc4t=5p;og<D8dl>Bl
z!9Yn0VTi%7-|h38zx#*$*1!G&#ZpYERG?HUF*DO(Zg!5vg&AgQHKes{?{>L)?|{qK
zZ*hP7fK(g)%@=;3!=oncc9T#2#K$>(dIc#h#vFUfoeb(6e#8d=S%%5RhQ)3|y!027
zKs!OHvP8L3pNJ_uMBaYRvq9?!f|%9y^L*x~e~~YK>GL#?4vC|s$;hyiA`D{;X`1Hz
z!f~b0&q%e-k!9}<wUi9ooH;wgotuVAyhL|+p9deA=g~*bO=j52akZ9w(7W^YA|kJd
ze)?yBh5k_U>I;8JuhU^^X_*HfUO=Y?Jsgr|ePX33#S!N(Tm<nDhpi(Hn!7xF;Vv8Z
zcNlbgJpAxuoIZUK6-m^1dc=g+lDAKUm-(1+Ma_cM8SQqDFbY{)U8UDKqF5+Vou6mt
z?k<&OZ|ie(u+R3!9qMx{6pJPHIzFgV>e!dgKLn0H>sV3`u(rBN6h$OS^21)H?};e3
zoWHoj7r!8|CL@R<0#!hXGJ=$|XBN2h<VBS71FVsT%-}0;(%~?9T&Fm3*?$ilS7-vI
zSXw>DCw}tJW4k*joS!M0j_9}c*}V1!aU4@D)(GPwsTh%^nqI%d?OQiE|G<N2ZRvJf
zBs$~q$DicfBOm0!M;^yoUm>@-h0QVf-CGOD@dN`ozR$_0P593g2GOJy>#VoC>GgZG
z+nv0vBhVJ2D5Bb^1A_jbPZS44Venlaemsh9Jo-<NBnfx#-uu&A(02%hLcmY{?8E%g
zA9mQ??m_Nc5Xvfu11x?#oOFW4<sv`*bI<VjrStjNL*It<6C%{4Vm;ntCzEiTz)1nt
z=C(~xhzk_r7>B0W+~dxj4Fs0e<y9Vh=t-8B*SUB11~~64*6Vi}cG{%bh<3M0p<c%j
z&}_A69qo{&DWy_{)s;C~?LCkI)oPJZI?Ah$;PR<5<AR2BUb&0)oNm8&cmZ3vT=OF$
zvjK&|EK7^$Y4;E4-hK`1G-0fG>O-F(iepBj0i{v}B_f}MTk8V`ouVg21If<rd!1kT
z$MH+Q_>27fpZ^fIzusa0g>Qm(#6lslw65PKX*LPZKa2=sR1nZ@5BTEezr;WI@Bco<
zYQ)nIKF;f}ULi>`R21;l=ii`EsFCW7cBjWE^&IMMH{;svO<uoxm#w`czV-6!eCx#*
z`Q=~wC7%88Qv^W-7*rq;($9{!R(g*_e2AhHSXl(2m|t9>TyEsf701s8=X~+R+nS_A
zVM6n0m#|O(6t&tc&ph)SOG|6K_~Mu8^%Ig&9uV5xo-9{1k7u3(lLW_FOSx2fN6!7*
z3+i&Rs1Mq_H=5t0BnuBc%=)GKbieW)gv~2bSPVY%ickbXl5FfV*xRL8okdCOiSaTZ
z2o<-ky}|CzeP-%wEG{gPvd9OY{QzsLtK7f)ZMN=jp;Jw%807MC5C&3hNR1^>notP_
ziRQ4?M~FiN8~=ay{wv6`E6wl4ers)a{K=D<Cw*3y_b8ZxN2Aeb0PSbGhaAi48B!F3
zaY^2g6nfHw9;8=E5sE8aUIs}?<mFtN@wU5X8tvg-0VwaX();86c5CTj?UPmL9+HEZ
zVWT>v50O!YESQtqto5z`_cugwj2|U09s)m%@I&(nhrER#4te*(cj@>01VPA4FFnVx
zWAh36b$8*FmbGt`DYRLTq|(N3NX1B<s8*F}3REmeXG@GtP9vm~{I`a3eT|SLDkx*=
zAc2X|NzR{tiYSg*ynTaIx(UkoHyIm~1nCffxGR*{Cwv2OtcZg9!{@5isgTP%XdRRB
zMj7}!oIO5?=h{feCQ<=H#6}&gO>-k{D?x7z=PZ{S;a7j-pK$8vK~!&-ez(W?sEeEP
z==OUM1vtV&L@}+s9j0gIaL2}JH)|9!87yHD#u<L$mw%mfK9`VYhoC(7oQ#JehT*wH
zB*K@FLOusVVoMku9mDlJM#e|Ec@;VXu(YPrZnL|+$?QTM&+!u5lnBppi1l6NGhiB5
zDR3--(B^Ame3a72$bGYrx7sb1m)4P_P*EF^@=OGdOe0m7V>2V<GmeSX()K+ffV&8o
z@qJ5v_Irq)!`~VdoO_x&2W>fwOe|pKQ}pXQbh~X(G0nXSe%K?I%dok<hL$!vyF2XG
z_Q+&BbgZe>swl0fxAyR|IgX!xieqP<H0X~G?v7sX;WO^r9PGzZC8$wy2$ND_DT!-)
zB-(S7RtyFM`hy<LR*Ru^LaYRCIz=`&f{J7M{XWHF)+7xm_77l6#b_dg)JScNcdOMZ
zyL<Ps>2Oj}?32oB{DCs#SCPh&HkRcfLy1rtAws66J$~UApXJQiBM4~|MIn}LCu8}K
zj-E|3cUM4TejScwO(R{cRH@aPWHKIu!5;N`9ozDlnw-IQJ<5~w%+4Q!SmUL#)SG)W
z>J^&xJ+58-knO#7Dvc%^J1qtgL`snoA?s^v6iXS>*({BEheo486h%l2v^Ih7hA(c`
zV9T-)N)sg<awP<g<&w>1Y4%(6dsTu#i_MK2hI$j5pLcrhFfV-Zt3<KF_XBb{3(Jy8
z!(ce#NDD1uw6uw|LF#^%YJ9xR*MIHT==Ex(C#P`6Mo{Z}q@={PB7$CkiWJguk(Nb$
zZ-*=I-{gY}Z}6GtA7f&AhTr+^-=o*<lXfhkSn<v8zr}Z6c_*PLfFG&-(P^ZH@2I5H
z@AA#>e3L=YV-Wbf@cgqljuG;xghyq(=#n@<+eK@OIEL}^1h(z6x_F1In<6(pN|0n%
z2SJb=&Sj!>4Y_&q3Vxt*oHV7;5p3ILYI2T42T$SqP5fZ69|JlRNeo4lW|Efh?~<l&
zV!XVc1^zdwkJZ#^3<=bd!!2((nHI=QhWW=HV)OPo;YJO#MoEa3Mk(XgZXxN{o9tb?
zPWi}jY&!)`8e!Q;D@70sy1gEaW(Chnad6=f4?g%1IyPpU^Ye4;?rzg=w-a(<#LyZ<
zDMb)Qpg(GEIwVGhN)RX3AR3e2VD_E6jZhdxT)BP;*R`>1K`Qk*77olClz`ae5=w*6
z>10z#oGSKF7?U7l%w{76j+df5HH&RK#%|=EjctfZSe9*Ww2le<0D)j?Vv5JkKSdnH
zEU(-q<GQ3h%QTJ*eHhp`NWDwK*ZU7qe7}re2LrV1k?~Rtv_(N!G~Rv-|NLpPGcyeR
z>4qkc!{&SVl*we1&1MTc{n9TJH<s}Gb=r*`;!x4pUMG_+V!LT<>2TxPRg{h?Pt9U!
z$apCpf9MpCzwlK~oH|c30YyrkoYYLZfytp%`-&i)7>q_LB*xgENMT{y4yjCrbUMvQ
zu|SALt*sbOYa+kT?&b#Fb`#I@Kx?|44w-a%AAuOMgp}f*4NV|~MJAJBc6R2zY3iaV
z;_SJDEZyAV`juM3eX+5HMVbh`a)W613FN^{!ZS*EDRG=6cO;y-<UJko*!I7;hu4;j
zWc85K=oAQJO*gI*25knNHm!P-gGY}dZI`)`ae^Si@mzBG44qbse&44MM|1~0!bo%A
z@LA42_8g&>M1gPqe@T-dv@mL1z&sGfu4&kM-&<y>(Ac)K|2?G?gMOc0uS>h#F~`{0
zSlpztIlNTXATPmyVxepj6AbenghE={Pz<C7Er_&cWp$Oo;Ql~q|4S8yA=OF?9otDJ
zb`pv%TqJ31ET*Tk{K7AOnzQGQ;kmY<J_hFBm*Otp^4>4BF#{jQbPeA{b3Y^ko1N_j
zQa~=7!}GFCl&46iG7Nl8zL2Hwhk&9~7$-M2#cUL!BA*i{&a=I>%zGcc!`}B^V|%xX
zv?Q*ja8nMxAEH#3My*Y^*EhVZP!Wb^O|~qf+>R1&NtKv}%j9{r9S<cXeZNDywFL;m
zz-JJ&h@t@7u{ia>)6C2tU}JNGUcb-iJqWdFTulU-K`|BGUYAb${)rC!tJGJ%_+>^4
z8CDjTadK($hi2JZuizl?q{bIE>HG-Nb{Px?y!qPq>9*j32aj>)p)>s6@BJYXWg^&+
z_)!F5Y-kO_pk97x042@6R25AzpJmV=@XmX0Gdnd;A(!K+CmuBDsnBRb<Iy3F*|3r(
zf4JRhp+y}P#q>KJyzvyxW}S<dFY(ADk6>AXbUIBO$L#EEbMw{>I^6-gdkymW(nt2*
zR=Ytd?=m@70)2O~W-94bf|xKcjlAh01fHA1{#$73K6X0j&`kAS<&qSjg@rdg#r*kG
z#I7<tq2yXCWkT;QA<$vO`g<P`_u5G$m?VYIg&+JmIt<W4apT5iT*oDw8BM&#BtZ}}
zKR?UVRGA<cMsmc-%Vx}i!&nnUnlOe)3BuU?K8TEo<51Zqlb2{%LYN!ox-OAYT)Ofh
z?_YeM?|%13RI61i+qye0yNB$3?<n+M-zj0)2<ai6G?P>F_e2B5+#U7Z-8ACZaO9Ly
z^b@-lgk)l3mPa0bg2{<FD%BdDZr2zt>bpvvA0Or2E6u)NOl$#Of9*Zu*oH9Vh-J}U
z-s9qL|0!Xkfh{e<&?x9^Cy9!YR^p3hzN975mW}0RaLXqNQ)BGxZL+zxh*!v=Y)LQl
z38E0`IP7k3v$M5Muh+!Oruq3_{s%n!nJ*`<<rYF(`^rld8)VM5ZR0=YIHWum&$F>@
zqj<9<s5sbfqK!pDfqX8H<E7Z`XaWU9itVixdfgr-slcJ#ZX1+$=w@zNcLU3l4PiKT
zavX=lhYo+D_v+pX!;qV|Zgc3+VWjp@L5!3ho{pK65Ny;~e)B!zUN6y-nKj~GQG{h#
z`#<YYJRybo8b$-{6Xk7_^K4lbj^p4I%UIb_94}32tjy$r1KhZBgT2ivLCePpEF9%A
zX!=~ba2YKuCTHgoo}OfUa)B7fNYa%?Ba<fbZVaLIV`7fsx}}p-^<CvTShl;*0gL09
zUaw1kFre4%@3X@+kS&z3U5_yGiQ<rau8=hFLq1)cpbH7FEV*7cZ{7Zx;ZL19bAVH4
zrVxo>Mp~A!p%)f8JL36g9^mB31suytu6M$fg?$wMUR5;Q9e3TeleydUSeA{O$`dP#
zPG>;7<D+8D;X{W=rPA!}?DEE&Z*$?p_o(cyv%Y#0zu!W|0l}b06i1}9W1Kzr1TVh$
zJSUGW;5sQbwg+^2p%Ju1AzsQR)-ggFG&KsNySyqiC{5t|^ap){dvaMV+aZn>Z@>OF
zZ@+t;SfvS7jKU(0Ocdel!80gjr~wEIX*mYzBRT9zV|Z@aHn!uixw(NK-2YAfu(H`K
zU;5k^saLiU7#dLVzzo^_bOqC#JB~+6N4)cgzr}|a-eYHbn__VUAyYj4^rsmeA0w4c
z5koSHHWR`jRbqJNQuo+|380Y6a`50hQfhA9zRJamSE*L(3{#$l9&+Y2H_28ANgPKs
z8+BT(Cfy*yj}-5{{|>!wlU}E1CMg2;_A0Ect<h{YDUXelb}i1HKE=_)N0^zKrkKx@
zvTZV68Xc+qk?T<Hs<j!z4*fib&w3okWV2b)>Ga?D9(=5(E{;QEKrHX!$6(0QuI1o3
z87590BR5|r!Z^w!3{r<Mh>NSCb!Ua<?k<+FaEOT(-^ai4F0X&*JB0p#OII!$K1*a#
z*{xD<Hu?0YpXaG3pTSF|$Q5!Rb@K3A`zlzalFprkCzA}#MZ%5w@n-678Nzceo58Ui
zf-vIU_upZ4bB*tQ?}r3_n2_c*I#$M@HohCiZEO$B3o0$w<lIV&(&#u|I!zFS#6e&b
zs@z?)hK=Puaae2mgAQTno5MU7Q&Y2?KmP>fiD^2WE@2QR4@%M;-h=$zOTGl|tER3q
z*ut`G3jPi!tv(L0_Q4Hq{_dZkd;NV6Kif{4oqI_1!m=RoG4fI#UMfRobPn7!bH|U-
zYW3)K0}8n^^YbUjrAAm@S|^jw<E2t$GI@?Y_%u!?3z!YvwjC_X#j+g31rvtDXWOQU
zG|h%(zAOt%2x7mVKx>jzDn+4CB$LTdC}imdg2uog)0LeKx}6$+5MnzH?M?$pge~uG
z>QaJj8Kl;>tYrLXzJK({q5sTt`qMA1HQ)TxclhAFHM45QM%h}_ifPM6D~FZ0FSC2`
z5;_PFT3{tB?vU$wZ_{d;*t`7Hdr^uQ0n)*CTwK>R&w&8PmROEMYNWvU)C{H3F*3O<
z>6uxaiD^a-9b#-|igS;g<>;A{2rosa*F{?%rSgo)7mkL>9d|jh)<+tOqzN6qJ6a`H
z1a}WfMxp5JuYFMz8^6mSV9+0s&u6h*GiJ|^j3F%Bpgcm5&1TJEIQhTDNqR{XMTF7a
z#_`te#h>Ec{p(j8EB^3z-l9>}pi<~SW6Y}*9-3LY<Xby9QE2$oL;N$@aV$%Mh|xMU
zMD+yah*iAbG!HMd5J}2_Eje)HG_wmQ2(?AE-T^6DSXdzCdDQAvYV|7bz56a(8#}b>
zyF`&sr&H(F%}ZRna*37oRkUq$;NTGs94Iq6JxZ+^((VQf`XOPcC=^GSotZ~RAvy|?
ziMU0n*rWm^!njy#!Z;#~6L7(C+1lFV-~P}43*Ud^7B}wfvbD8Mr`^W)LmW5F<ji5>
zD8~2uluD)j_rZ+&?-312fn{54Y;1fIxA(7IpMCZ-99}p;yH&@r9URNWD;BU_+k_No
zP0A6B=??jKe~<qAKR^T_mF+cJl_tGbn?iAtmtOoTxm=ckA0lA?U|@tx2HDbzj~swJ
z*BDI~3wa(mb($dT(P?*BU)?|{fr=F>GJCg*)&9Z4vMk2SW0cEdOpKS9nJBS%>k4bD
zi)?PKv9`WMuiqgELKg2Va^b=UeDK~yu3x>vcqv1-waHs=`~^Sw!5{PPyWi!?)wk(%
zYIi@U7Q``xL5x2LjiIzQK9$2mV;sjx_VwSKSNE~TuW=k<+b)isHe-u~CnJS%(zZO0
z)M%O6)5lrbtU(~qA|}+}V$@xZ0>9m%y|jk&z#|kJYjm%_#YCsZjUY|GBXMlW?VFc5
zdh95b%^k|F&CKj!j-NP#!lL@W{J+WN^VBL$;xHyshV*DIV&YJ525I~C<D2A1jDfXL
zF2#Y*)WjIOdlj0k79U)^z({VC_ul)ECm(+>X~+ah04hc&i?o(^jfc^;d0@06o68f0
z5lgq1P@O)LvomDJO37T^s957f4KqARbh<URwpS_SCviL@pPrjL#3PSB$@Q!6;rk)N
zHRD2k&**pfaNcW1d_OgHe$>XX1D-xRiFJL4SZjJpv-;*0oLrtmKmU0`%f?IP4R0|D
z?s6V28)-Xq+XJ>%w%FO(qE@dmmG3c9$}@lTD4l8pH$OtjO4BUuk;*t^M@G@RJ4o9z
z_brSPVtg{Twk<@$k%^OM2#gBz9?@np7BZt%f%*46FGZ;|N+y+NbhO0g<{q`KCS?g)
z%^J-{g~G@Lwq?`lwux1YV>?JK?`gb^ErJ1BC2{hM&dtr@c^-rN#KxfAt}&99NZUq-
zA$C~hh!v7a;sFP(9=Be8jojoUg%ij3o2KExWoQ{PJYEkEBKu!MuS#hlZOa@Sg)x^M
z_<f?-Lh6u6#dwa(^uYyMdrh>ju~me>U87TNk<E{gPN%8v?9%LYDNY?R$WG+5Z=5;&
z41)c*E}bZ0hf#H)_EEawyQ<LG(ue_)-Y~H<=@=inAYg2~#Gp|JLh{8?KoA5zuIm_c
z^W@zOgOD%?h$BNEK^iB3cDu#e`u*ck_+Kmyo2y-8TOwnmGS|<F0}kLbBOG==xIz1Y
zhw#pw1ZmLldt>m-Yh}a+Vn`_&j$HSjiKGim8lQY&j2FN1>uhZ7GN^BnOQ#td9iv>H
zpgc88-;ZhR?x3T9d_G4Uh3sx^u(G;FDl@`gy!?HN<xvVbkI~UG_G%r+5?Io~ay0GU
zfPCoyQ{`ELK@S}nq)kYTN(^klp!y`4tzZswQaXh3fPeB&{~3Su2jg6P^;?{n(`Y5|
zeMMn(jM4E~e1AZz(Zsf0<UW_spjgKL?`}-m((>|8^RE7l%8`;UKll(quZN04q@6-5
zjg!lgAI)R!Rmo)?PCt2&L)~jc?Ix$P5_#x2KYaTlTN|5b0AN9%zCyf_GI1Qz_l=3U
zV@u=Z7sn<SH`aHBPDzcAjq%JgPcrB#N+X(lYSgH&<Cyk4-ysqW<ot0&Zj=zCbT?5G
zFhqpnFhEBEE2}Gf@ZsC+RCXCDPV+zg%l`*TNlu?WO|#kM!iQJUT2rex`1vn=5n@F)
zogp4k1Yuw}c#(+><DSONOwxvc&@e0bu2@e>`L`mL_;}{SL5Q8n<JgWtzqI}dh#V=&
zA3DI;@ohR+S0I^rYo)Nk5K7~K>g_9d%LfpZON`Lw$V84in=P*F4mf-4C^v6kW^Q(h
zFcf5pZQ7|SK@jrU&widpv(3ee@3HLOX64SRAw4Gd29Ye=LhO%2?>&JJKl6`IIEEWd
zCX?pCfmznqH&|I+qIPPR4?eiW@uLfrN_o>r36=2jG$i93P2yiCl50W|*#(T3M(N2m
zLI?)E9(%PaW20jz7437ngfJ(GZm-MA$};7$#`QdAW~LEBFg-QT?$#<%9Peky4Po2i
zJb$=>-i3SbpD~WsG3B|Kul$pL#Jm6Ee?ot71z$loj9LHwdpJ3d*%x0x6(zQlLP!gx
z*hkaWmUsE#55CW<Z~Tajtqr=p9*>+n!2I-YFh4a<rjQ}(Rq3_cST4A3hDNnYv)!gp
zo<<8v6i104*<I4U;qxWd6^VESX{ZB34(4$55{83hY-YqQg^?23Y@Tv?94T+o3!&qK
z-RrZvv&rP_F@zKZVZ^}qNx5l;*3%M$21#p}{vi>Oc}_=)Me@1aebdy%%KY56a0sI|
z)A2fooIdqP5NQa}^f#+4z5IR7ltytU%gE$<o5O~U5Xq*kVzf~EsFfNvbC!i=xhP+=
zv9-m<+A3>nYi#apL(t*N&mW_Z&*9lAEK5)-7Lk6BPNzlax3JO@FQ1~<?Xk1Hg|FcB
z>4!16B8Ja)&qz<|yHpCTNklA(3KbF?laKn=8KlWh77r5**6VkvHyVUt$mrM@ttdhV
zhG&CT1mS>UK940W;xHoY`@~_03ZlgNNFp$90?R8ae=CmiKdn+aK;t2?%v#pn;fQWC
z>KMb}eyz{#SKgp-U=nX)5=$6#H2Mk7(7oar@t1u&ne>=V@WiPH`Q=~vb>95W?=n3-
zK|Ylu>*Yx0^8|sSTo_@{X)y@es4$?{X;T_2Gdp((q(u;h++Ms&VI)sFC2<^u=UL<m
z9{oYci!Xf%fL^x?2rQY{Em;B+cPA0avl2@a2^U5YD&a*PJ$jtO$By&X557$>2oX{f
zg&HSSB3l@x+ix-G_n4R{CWs}3q1s0zu`F~P({8t^R4V_??%&^b%}Wa|U!wWI3Dby4
zNxN0YE##OxcZBhl$Aga_r8HhZgg#LiGFQm*{+%V#DcITBV5E@84>rjc3rvic2zz~`
zCCKKpIJU#a<}O-+>)HvY-9$Q$j*jBFBNVfeY$jkdKZRDXyRkzzuF-#M8*!;b`uuZP
zhaWLqeiFfAk{BkmrqQVL;l=k@UR@-eEz<1;T)A=s6>F|vzd;ZLEF73;Wo6BX9ioVC
zugk#i@7o}04ZAx#%uUUZ&1LtM2eH!o2g%{yeXp&FWm!y?%iR1MqV*rEsS9^^kRwGb
z+cI%s_cTh75dSr(OooF`KEjQyJ#@XFATHnv@CA5MG1^_nU3rI;2vD}gv4t{zyGLk`
zaQe{)>Gs;Z^26`(>1SVHb9n<xT12wJx4-=Zp8eFPX*QZ%zkcb*LUQj$@C;9ZL)d>f
z_8eaHzpUZ$1D|x-L!+6WpP^c*a_jar#!C}iym*CAed=+vNSfJwd1Uh65psVKidD#^
zOCJzk2yq=3$MMMJ@;op(PPJCy-FM&S`RAV}o%YCPb4W`v7!0V_YbIbkXHhH`KnR-6
zCUG1w=m+?|iHxz6ytv^A=k5cd_hr{%7$vqd_fJJui}ASyPW;MOi5hh}i(3q|CJbX%
zzV#|HmnHMk7x2;<q@DN|*$(ZN&v#z_F2D7=|BiO2k(_1)i)-sNTRm*K$;|jXg?tX%
za}bWC+N!d&e23lI9!F+RVGTEBVdB`779V-|wHCyBcsq@N!q~{%9kItzKpgl8Cxbwf
z%M~b$lo%fyBkg&l(k}h5Pihdcy}iMS$VUqE0(Uw+Mn^`l5)d%PxW>zboImU*RUA{u
z=a`<JV(&g#&<9~1t-+CkT(rxa)5jJTwn%bvQ3Og-xw6FS8*g*u#TSrX3KhqOC?srz
zWkz#m)COo6Cdo+YAcVutPL*3%ZgJzrHLhO2%<|d_jaGxaYxC#>zsA&Lp0u06acq=O
zsMsfVeBxA0&dw3VG3y&^?ACV3j~$>`95>fl+{>9`2#V}~&+t^$-L(Pvk$Vp$G>!z3
zPsWA*U_j^xlt;(->Q}zZ!;e17yKlb22k*X4CR;$Mh%hvWk%|?A{s4b4*ta4AOf+F+
z5WJhW7JsHf+qG6mBvu^aMs*JBJq|iP%}C%UlZ5KEB{tr?z`>WEN2W7a2qMficWgUp
zkSvC2L`K<fIM^atYwtCQwk!+VaXA0<XPGLc@H<s(+a``9q?N{YOmI3nLZS)#edE?|
z*%Z<)4?S?2cDKh!-lI3@kWODAl@hoPguX!NF^(O1fPDq@kMaGK(pZ)-$g?0h(1$2x
z(d+woUY1-TgYWkVBaPA$r8Hrr$d63ocpm*~7o`-LOlF^-4yI{UB3WzVnDzBFI-Q@5
zCSL+ixH+m<E|Fb&klc(B2X<Nwrm{AVJvC1;EFdF;o`t<01R-h1;)K=Y!@ken_8P%p
z!2J9I{@N}NojJ$TPkxFc$4;<UY4FAyKg4#kq%&Du*CGl-dc8g?%d4C^agMRkF>*Q0
z^6lG<WM<jhtT3prGgI^+1ZU8|z4i+J?i$tupCKyEB2`GNKt~b&pvOBOyv^;U8;lmm
zc=4s5!*+8VJbZv_S1z-7`#Li-lSC1$t*$5gt!A%Q#Sa34AfVIfG8p)DnjO|gH#mCi
zD8jak`LUD+^tp#OI<zwx{(T{z{~O<nkJZ%Oe&uBzo-E_q9<J@06;hahT%jdGcpy#p
zT%4O@=COy_`rhjXE(L52(u0%d5A);`M@iWcHWs~sPsYnIleaj~tK!$2JaFz9%h!Lz
z#VZ#$b@C)j*RC^r^cZ2(VRdzdg#!mUdGaJ%+f~#c+JCU5P*{#@jE3Vx?wl;@0wX{F
z378JhN}#l+(P~r77nm+j;tGeo+BSYP;QEba9y$LIsdNfWjM5;qi8eBc4|m5&w(W5I
z#3`hZ1cL$fMvEvC^acSpZ(cXHE1e$sTybBy+Us?A|HF6L-r2^svm8Ejl6=0HgbSO6
zVbE_eU4DX;W2{l`B623qEQ(?RUlD{R*5+;$)%{Udu3csj_~fRhIrMjbp4PwpZT5C*
z1X|(sVg1iuCNnk3@lU^q_FUtACM3Jto4o$!D|EXpMn{WGmd9CNTc_RYv9htni4(_J
z-`Zd#(<9}k@B^RK)m0W(?|`)F^+Us<6T%pMCdyN#!P;NAlkje#P|`A9W=h-@oCrZ2
zhXjKzDK~@GMqD#CRwk3pkj-ULD9Yt|+M6qEZ*J1>b;xA02q9@Sn;e*%HHHmB?8EX=
z6zwbhlrm!%*L68^=nxk#UH)WmSW>YHAcm2Y%WP&qAu=?Bv{3XV2xT_$j4xPy^*u)B
z=gFNriPVB9iPyoh&8vbi7aj}ScCj2Afz9%rbuNE!k?;NJWj?s_0qtIQpEGLbGGx*@
zTFox5Y%o@w#&slKHj^CUTy(F^&e{r<dWBlE$pa_m%{|}OJtz=5iT8rxcf~zCH`CM^
zpSdXM<%Z)wAuNP#@6#-lQuu=bwk3JuiN~3rJH*KN1QJ+YT|zo8QD}m^bD1nr;%^m3
zA^yPMr%Whq3{Jx!_!%)BzFR_DILe}Aby<iv8M8I6u#iGysl?LNk6C}^ZN?VnDLru3
zte2Q(_AZ~&+8>YWd(_awl!c)>TuOmu+gO&1uw538KZrG`;CI`oC_woGf_|6U#xic&
zL&X{?V~AleXd^<OR4R>?&CxH8P^ni41`)0$kSHuG#n-?7PZ=4VAc_NoaFfq!Kmn!V
zeLq3=$D|-ENfbL=zkP@8?FtW^KFA<RFp7mjX_J3DGB!=D741$3$FXr;i!f41VZ=p_
zZQKoXg4nOFuHIJ^n*UOjrNe^dFlcnC-d-j@J4e6MBG;{QhCW&0BU3IxrGZEeOR*mk
z`ieQ#=aIC-hoxz5Ena7PW0g#C9E9Y=$>VrJa^mn2vZ)mF3x_EdM_5^1A(P3lvT}!a
z-uw}{3`jcMxUxpjQQYp{p}l&Q$Hrpv(If<E+)N4?1gPbU)YcZszxW%dkugNl)a`9=
za`W017N!sI{HLEMlgn}A))IGaT;#&ruX5t}DSqw?U*em8{ujo7Hj3%?dW?-t(r9#9
zUjC4)*RJ3PiqF6J66JE4{$N0K&t6T!$q%Chc!{F@PywYBxt}(tZsohLFg<;c{38z_
zTyw&(6vSHH<v$2v2*#en)T3tz_qJ($a04wuMh+Bs{8NvxaB3PY0*DonQaF}H<hqQM
zN}TRBc<b^FPXFC6F?sMZrO^@AclMZ`D6x9uIwy}D#qk7{T9bCCM-+v_b;4{KNyn?A
z=q~S!<om;r=pW;t3&3_P4jx({mnkqcIZnS*<HYgfEG{kL`(5HeOm(ly^xT|L(5VO|
zm1zb^lr+Q7VM%hi9FAq-jEpcbQO0#L)aq5Xw>K!{^9=eu(y0tWKomw)Di!j%B4<xO
zgk`xLK6HXZhfgsW46tpR{-94~cN1Im_tDAWDb_^l#6(eSRK9Vdw7JjLgKDkH(#jI^
zQ=_EkraApfU*?^E`+M}dJ{f7_w!7T?qi-;K?mYR)G?pdlG&<~TZQ%!fN~Jt6zVvA>
zU%o<b5Q5hH-k-d}182`MH8aKX(lX1-E9`D=62>8^Tn5*&`Ll0*i^a7zpL^kXN~1Xg
zyNe+LXOLW^wNLPmhvZgeEEe`Rb}8|LE@?f6M&URv<E1iVr7>Ro>~m-&$4{Q%+kg58
zZ13*UsvCo1%Z6sNK^!ZDhS)coy-sXrB0V%Ob@n%n!-p3>!8`TQisAvb40z$o=NYMa
zto+6I5PpOs1tQW|2m%#>Oc7W6+<y57oSi6OmuD04gSC%x2^-`2AuObAgS6OLZScW`
z5BQxw{#|a}xz1f>H&_CsrI3~)mn*QfyF<#&Ggh2H3MUb|1XL;&R@PQ%^}0BzESY>6
z+i{Ru6EozQiX`suZfsDj61RDbm`Wjtl_H827AC?C1hE!K>Fuv2apFIxQMj&$Z8>Cf
zB@Q1u&4cHkqSNn@4{|7_NvBi$A`SkM(Muf1bUIyDSJys~`}@}|N5Vtrr#W0mvHPtJ
z!ghqT1P(Ar7DFFFQ0ubzgV#AXGlMfaY7i-<?kb*1Y-RLMHg!X4ibzzU!ge9aXV%$i
z{2-t=*rV3iqfy<ZwztP<p-8Ae*dD1|1{KCMYE|~CyUZR~z!Dn2*P~glQ`@T%#{wrc
z$}j!Bf5bzNJ_gzdmSP!y1XU5jY?tOBpovtB@vXGTk5BNWbLV;Y-4EE^-Ny4Aa``-w
zsu6{nSX&fIQ-o1Sr_&;n&EU8;TAPTdp%5>UW&|4x$>Qz5d1Ta2qXcl?$>K>tKNd7L
zw@^`x-Q4DZ!6ud;(rUNJ%}gM4%%IZ3$rq5)qC8c?+79?iDx&P!^t+SXs#RI3+$1Lj
zsoD;EAHGlMrLoJCtlYXoCY!-_9r7aubmTKPbBJ!I%j(`H_#tX<hZ~hm7KCENb<utv
zZAomc2zwnm%?7RdHByr^q@MUHp%CaeA{ca;8Xx7%@l)(>ZPM<wSzKBs=(jj^@-U^5
z0yl14XKQO4%d%;-TD<bgYn(iB2CZYpN5`3-oT1fjQ>pB-wvk6jNweKyaq%|m>zn%m
zEImw0QR?pRHT3#_b8p#?)zn$_4wwGqk2smqq$bDds~D7%oTQQ$(6$g70^cVyQsl_z
zKf}%X9w(;KJp9xNGI<vj_l>wBHg5AeK`T;)0w-fwzP`x^-+qM?#}0A()*bpG<g;nY
z#SvC+-s0ei<3wSfVxh!XsX(>TAd2EdGCxdxFsB|!#%Rg!#ZZN{Kb|%Gs3)FyjI(D?
zbL;v#rBV?cx5;I596xf5Y|5e63fX-BM;s_VPgI;FHvXdtM^Bqrv>~@pKrBpve5e#U
z4jJe!S1*6apx?&#yW}^D6bc1&Lb$$t`!?nAar&JuW8-No8Pce4^3J>O(r&k@Rx3CX
zKKJ}{L!+~gs9f(=&<_9XemmtwQN%kJ-r@K!{t`+nMo%B(*cU&=r9b>0u?DGa`b+Dq
zz4aESzVbD+im6uCXg7B_b?hXkQwKPI{!w0g^)1lGwD<ihclbyD+y66156&_^QDS_2
zjI@_0U-0k;eQqu9@x9kS;P=1z3U9x2fv^ACuW<IvF*7YsHXh5iv9O1Uyv74D4EH$@
zcEY2HwJ<1XGJdlyo8m}`sfj5T4$N`r;86;tGIO)@{MNtwSFCSsFj}5K3X4{!Z5niG
z;SUD;s>(Ql-=&mbnPw-BV`gWjaUAEqh<b~4$k?Rh+0Q+UUmd5ryF=rh4{$I_S6fN~
z6%q(=wMA=bgVpc9#@tuFh)iX$rES)TMBQpAj2757L7=#C;|AaO_MdU<&P}vdc%F-6
zTLuXs#`7#Rit^+X3eDomBE8WLj+-I~dTeg2bLskJy1q{~lcOKS{P5Lxh_%bnBMV4k
zw6H)Jf4Cu!&WL4%2~d}2<T{M?F%eKn>n8--N>d!21aVJOH|V03#&umo;?EX1dgKI;
zoqvj}*FL1)YLT)nQi-LJPX3Zo5Jh+GQlcngw^CuRTKlAL^<TXN96mP6FMa*f#9@VN
z&21LH`zjU+OGtz^!iZ2SbnH^QzRc!3?=%10vk1>dM==74lx`wSQYM`NvG2)n?<iIY
z0hWX1rf?E3L4}9X(@@_)D@DE5U}S8JY;lBMcR;t-W^v^X>+35(%+|&l*<223xilMf
zy4^12(E=kA^ZdPk_|KR-c+~6;N+G3_WOd$c!iVAom8|*NsMOHHMo5>Py&7|K(?mgs
z@lu&gCP$>I#G%Hr(&S5{_=6tpc8kf$GL~g0@nUyl0*&2~A_xMuwzod9+xu~9T-uy=
zjVXEE2f0xtRomitd!1}5gN!1ajDsE+A3hz1M8N>iSgwPY$sn2?KEIG*+>Q8`mwUMB
zEbVTK%k?_xdX0kZBDObaZLBgge}Go2L)yz?N$B?j!d8WTyTx9o#UNH3OS_y(!+hC8
z5+k)D^auEHgx~FwwJlQZCImgSlYv;{NQY*1hwuFPA0pihT1b{w*4f_OqTOk;uyBx*
zCluXQ6AOjsdZZjlt-5LWw^78@#3ZAoaRz;#ci+80v(uqcZ?Lzwmv~viPoPSF%k{CE
zx`2pqZG~&!_$CWq_$={QnUEOi+_fjrmc$19Zkuc_LvenFPyO7}9JKogM<InG?1kjA
zMFu)V`=Nm(QP`e~uR@+VT;k6b@6d3M(x`8dN#)tu*&>(EQOxAH^1=Je96H43UV4!*
z>~rPPMXI|s^Fr7bT4>Ul6lg)KV-!s0Lh58Q2e9^iVjS1OwqbE`3CGUix(;JwBcS2b
zsWbTfK2g}CbK?rFu!eK`F}#K2D0hT`jEx<JFx)6Zs+NgLQlaArts*LWJFKkS#*g}h
zq2T7NYZQtF>Ww;sAmGN$+g!eUjUUO^_~PeZ;Kb2`>~7y-bNv=bK@@ed90#Riw00AO
zNhZyKA&V+y(8zn|^!Ho6_Qq@c{a^dK$vM;!^G`iSdvlN4o0svGMMf&_{Molzc=BnA
z3x^TXp;F!8(83|+j^yytX+{bK`rV<ui(+-VMz!8FBN`0)5(x{fG`&Gclr*=0_{~3H
zarrj?@aw<H=RWsoJU4|#B@r?f2pfffgzuzOY=i)b$C<Dl%HvZ=%VY-v<a2p)xdM%5
z#n@)Vn#Z1e4%<!h*4wYLTiGL*Nz?BQXtmpv3Ppf@UR4-|`|pklB2>b$DHihNa#`y2
z#wT%`eykKH&rLBgKFMy}=g8-t<Lc%bgN+I{MoAVC5o`0$ia@aS?q#wE7Z`icMURe~
z!~|(lLZq~@YzGbOZEkb<%7<LJehIB39LM6(^AEDSwav!H7GWG?X^rp4xTzGAGc)*M
z$lI6Sqf)6*uhodcfWl~rTq%c*&GlRB{NU|3+1jr1((^C!;DaYH33PWyuJ=Y#DIpCf
zHVi@gW_=KbXJ;H-Jf<e+@!WKBP&AEL;13dSK|2w@NDBEPXHK7EZ*P~)^<@ssPZ*c@
z<efB@72+rOG#i^+_<rz-+}*!UNh$cu=N@Ecc81-ZfCJAxO=am0dTGsg(kY!Faxuyh
zIDJ_C(L3Z1&5%EO6s;s03vD^ZyjR7jXlN^gX{3cTJS0nE2^;A+#&*cEAW_ZQC~U_f
zJ2JxP#1ywLU!zuOGds7y*vt&YOaafesn@GW=}@cH(ZuZ5HnCDEKJ(&NIrYe=$d@OL
zU6mRN-Hd?EN`z-o6eZDSB0+ymX{E7jhitxpuxtjsHoiZg-Dy*AwMb_K5gKnKM<$o0
z*X!c@edApk$7YYval+w@_Nf8&dX2_?@$UcE>#UPv)P`Cp2(=>L?{Q%H3br1gY#Te9
zCW^JW$J*?d%J)%mgiPmfY?s(`*<9PB6f1sl-r`Sg*4VTh#ON52CGq13<@-3TIybJh
z$%Y|0$6_mvx!vv)#4(Q0oXI(y%efpb#!P24BN>admm=kQ$XL@K^cgf-j82Z?_UrhK
zJE-D8bfoZm4Yt?TxP18%$Sh;!d5R-bjN~H*{Vum|++g|kGWE`Yeh@PVyL{`rKR`hu
zU{q!gP8=16&(crUf7^ZeV>NXOjjJ@BTZ=5^GE9B;vsfjUR3bN**4=0)U&YwrfU#ba
z@~DHCnD43xECt<0o%Cn{KM0XPH1P3+#3@hEUtZ%&hYI{jZG!_d3%q;bBE@W-!Jy5N
zg=5T2&vN5~%T%`Zc<QldsP64DH91XXuR;`u6iX#aV`E4G>l^DV-PtgT8PCPG1#zU%
z@sJF!2}7TajXfTJ^iu?bm~7S~u(Av~FgZER(&93ain#6fdAPoY|5k<G;#IOwe-=49
zk5o4CJ!Zzq94CQR&D{LX@@-n(2Ft5U96I;_-}}M$`49i;KXUNkVNRbuO|Ku&Y_$<U
zyWK+|aV&@Nv2h};jFY|A#9G~rA&X<v(1kHt8O1H!Rlcgfp)>V=i(0(1NUc^SR~V&H
z-Ns92IsS#`d4GMAaJ@mslJwR#S^W05dHjF<kBL^VQQcg@C(p$6gOo-KeEQSR@$Hwt
z$;R3aj%5=^ihdAdTh_k7!H-S6%l=Fp^ZSceFY^EVm;WQ}Zilb^+?Oa6vk0NkGNJV-
zVK#V-BGtl1DT_!$xjaKEoi(avArV56&SWVTM`$#v^afqrZi`vx5Klk(Y0Bdhy#3DW
z>{M$w0_u&L5ldK!lFpJ4-j%$Ap?k*4EKuX4rTe0((^``qv2Z*W+jYoJPjTS0&v57W
zzKs(N_QgW6Rv;vn2H6hSc=Zj^)03$AgE)?dmbN(nSvFGIblM%();76%<1#8i=uVwF
z!q`}mYuB#ghvr!i1I@R-{U#@mpJsYylJev<b~;D1)uvjh&}lUh2%4Q1%NtEzd+!#j
z8+-ir@Bb%y-7bT{3q0|}LrCi`N^d2(!oso<!Zv(LsghA&GJeBGAsH*rkjWOzU?FMd
zqbS7pjp#vOe5jOCNNrIbpXT(b2iV?RGboeNMje?n4L`o7E2X%3d+EMBg#G7A3d!7o
zX~TK51n$@<hn|0erJWsQJHkptNxqJW1Xu|Al?H3ydzE9OV~EiSw6x9jw`^lX4+y1=
zHHx$mj!{GlVbdE7Xl*uVH`;W24T8W&3r#L<GdnUsOO2Pwv2gS_jY<`7w9LQ<U%P}p
z+`hiZ;S)zWI9g<VW1TS8Jp9y496I(0%5qWR04<YeSenSbyXN{5$95Q+WNB@j4z$(;
zv5)jT9LF3Gl~N{bzFDW;?V;0xK*3016vxZb?RHQ~W83yUP5tn>()*3x>e>qZ!Tk~Q
z_+P3KDS0B3Ax6-QG$WZ5PoEhj)vA~*WXGt*tzrr*=h0oMAR~n>Y*eHWN`Z<<=d(=Y
zawsV%cAJ!4!Ih@w2P;iFevD&V;JC!843?WBj1_BfK)2gvL}+qZhZpAaJdzEOQGn;#
z6pJI2#wJMT^2kUTQCb`m#yv9j6e95H+<XstZX8Q_<O_LfeZj505Gk5$v;$J!7D|VN
zQJ=veB#r_iO@h=yqtpGQZ^2L0$IE`j_@mhYRNudjGd9KS(~p4bnA1!cVkJnG6$gy=
zYD~u+JQ>s13QGuV%R)s8zuCdHZQPVg81_+7430!tl5DBK>ROe@OF816!`}R1276o7
zmR4!h8jP=&IdSYHI~#kf-&y04GY^s-EwQ_^$Ky{tMXp%lM?ZR%58i*9X0wg$*m#!B
z?A#26LXI#9kU}yT_}spIhio>-3(tR%lvkwFsZuDU=!%F=qlJ#a?}qsOE;0)$h)K~w
zZ(qiGJHmSG3k)U}u(1(|1|!CdbA~8DyWL`QbB$_smqC9(6e*VOtgyDePOa9Y+wB_j
zK`GJqoZ$C8ZiXhuYUphjf8?ZpZ+gVNwkG%6_!XelYVrQ}Ugo(k{T%hm4pK@A6O$}_
z?paoU|7C(s5099Q?|h&5xyLx(T4bcT#g$%#wcDelGcKS1!dDm_pWyd@|39&}Q=u0|
zgt4(5GDxWyI#7IM|I}KM%VbztTjf9e&VN84_}b5ZkwPJbj+Eg@CJU@i5(f~5D4k1Z
z$rVS;X*w4B2)gUK<nu+8(li<kJU2tVR%Lv0lH<qDlFt=+{f!^7w0Moptt}27STLVY
zn4}8FbqM_6BNjD73zfN<X>KhpeFAssCu`sjl9UG*VcQf>pJDRJ3a#(INfaq0+90Bq
zL`iUkptrio=EaN5oPH25l{KDRj^R-eYnE=@p;4={xwApBkY#+L#23E!0xy62JM?-1
zN^9(dPxYNw-e+<7|HQ<^IBv={!YeyT|8QMGKjiX_JKS8}rWY8G!+NvMpZvw2Af?3f
zY#x8&;e_-anaDrOGAJF1_#wIWK+^0<3i%O6N@FNJjPo&vrYH*O^}2~_RuZd_SVs_B
zAS4S1k8<SLNzzV46voEzH1WI}j`A$a*sO$M$m-h9mdPxM6_b>S64VkDjvZtC{KIr#
zc@Hb0y9~K<2#t&-)hoBjAHPibi6@Y{Y}PAGlc+71NNA&|wH<^b2>Y;fXOX4FJ6yYQ
znU&RLD)lOZ{s1jBc`waB`OVKVIbI->@^M_7d^(2+Anl~kUJfsn;(>?Ga^Tbn(wPjs
z+c&Yj9McERAl<a_CsRfYV_DMhuvM~{Yr~NyneG_N8)1A#;|L3jR3>NWK3Wq50nJu}
zN_Ee?*O9^filqrGY0+ucNqG*Q>n0whdLOCZr}6;HKO_Eh&m1^FsXw6Fg9v!&SdqDz
z91tTSqvQ=lfTu02$R>^=EG<n;XJQv5q8Jq^ykZWNz{(bpQ*dBL^Wb!WdS4-=!`gO@
z(r3TK&FUUHj>+G+f*-XgO^tIT4`~&VPN&J`#)zVra=DCf9BgSUcVv>}gVuy`Od83c
zx<khA5J`(S-hPkQFI>b|0E&L63-@8u_Ft=yhtwsB4Z;$5zGD6L_bAMaW1lzy+SqdF
zSYca|Owgv-9bhF7yCXDKI%O7q*CYmXDmAiG<2bI1o%ZOo+T@)SuH~`1+T^5b@l?uU
zNh-3ZA7|~xMYgINZ1<aV+6_u$)3|nuOYdJ|YIc?{{M=V~?CEDw+TqB-v%GoyEH`dm
zCRUn!zDT>>VrzSyy}eze5ajb2JU2xcM69f=FflQ~csWO`Vm3E-NIQ8}msjbux-@Eg
zWU^q143U!9md5Gs;=lV1QYSwJ2OlJ|U4&9eb7vL=rDB$r?@(`4sn#kSKYo&vC(rQd
zPe0F%n>UE#khk7?i*x7BvA(gz*2c!Z(|ohlOb`r@UayZI_%xbHrs)1XNdl$sY3}~A
z_?`QulvaH6-~T(#OiYlTog`KPj^#3P@F3!I&#>`l-zO7>xcvs+<(DZJUA}f8OR-nv
z_U;nuLI_$gJ~qSO`Gv3Z+yCj`P;Z*kug2aN%?R<)2pJg0fobMur`XtB<J;f;Hknk4
zFMZ)Nc&U_8o?*rV*sd{^RxT<CNoS3b^!l9@(w0kUcFK_Kg(R0NU|BZpc8Bq?fX%Ht
zq_SxoCqubB&1XLIdD5vgJ3Dvic6&IMvya#eoA7<U&OKxE19LM!?c4L8ty{Nm8bLu~
zx{Bp7{p=G|R@d3PwSuKA0>d^;Mpp_e((GKgO!uV<>9L8si;aR-qsrdS7I7GpOL^q-
zBmBlc_;q%7>g;S+2}8p(OxY5Rq8-HC*{HIzRVC5Tq5(gQ8AQew=f^=XEuhhC@ZBGN
zo6%B{a(RLyM;6c+mGclC(R%1dr_5mt3$PK6OKEJJ!C-(06w-51+HB~9{(yeJZw$ZD
z)EgDnR#)-TdB(;@@w^l>({m8F3_mlD_AQ9+J(PuENTb<ebNl|t`u#YyPh}Aa$}Lk&
zKJz%W#dW$X>*g*E-3Elf(i*EDv-QS1<mP6G%Tstd4M~=^wycB-VN!sCf#ULo>%8&q
zt9<82-(`Dyok-oad=Wt2^$=3iZa2xMvZS08UOt76x-^?R3|eN~nH?Fyb3HcJ)~I#6
zOe~zk@zR7rkYqla#&XCH*UC1db`c{iW1cON2F&D!D?`zg63-hd7Bzn0(`q-_+cQX2
zq>L3%X>1AvblWX5>9irw2!R#?l_XXSg+YOex&1RBk{B5&@vFb~_gVQj|C*kPIeuW0
z=RWl)z4j)qC-6p6s4&DED}grd5~*y4pwUV^|7@h=g5zOD0iB&qQl)XEki=1dB_t(B
zGm&;lkB%|FP(~b@K#q@6*;r?E#Aab)l8m$%8OaleJ!}j5Jq?zi9|hzy8KQ7N90qtY
zkLx&yP!kRUY+FF5$)H^&kdhz0`Ud?$z|aQmzx9~(zg2y#rp`y>0x2w%7W8UO7T<iA
z2c{;lMka_N9}z`ZUd(j0iK}`9B7@&*p$C2Je4gyYDAJNdDM{QPFxah<nJD9BvNX2#
z7#YnWrRM2J=ZXCRIo0J?=N!JdvBK1$leCu=txk=8w?}nv2Pap+ax!eKZSwE`)xYNT
z?|zrl4?n`};loHBFgHC*adezV&OeG}Su8Crv9+~DE|+CvbB&3KDXv|5A1QixPJm-U
zx832=hnIQcvFF&>uF=}tMD}+%lAb1x0%QnQE^Evtdrg85zKxCoIPy5CJ_^T-KI4#H
zyUm@|J8bW5F+MWRvrj$GSb2`Iu`%9x?=9Asm+5yKJn`7WbUR(PHa3VNO>YpexwC~I
z_(XAx?+0``Jr-}@VXQQc<=Q9-N@{d891CdfRn^_Eb-mu;{eS&$IP&+thI?=xEp0OS
zJmHB$90&u#_uk;S&pkwW)I%ybIzLV0+A8jWvs}2jLQKf`^aNwm2RZw|Iab!U60Var
z{H0hK3kp7Z9W>DE50IANz=2s-R&VkChaWIjD)P*y9!E=~ek2K-*TM#=QAFInb(>5!
zi=_pjj!{v_;?f-^r>4kcGvsnPTI~jdL7zseMc<DZ9i3sMG(tXK<cTLg&3o?(n(YoF
zx%~cKH;hLcrh&lSOyQa7$>itx<j3g!K_7&|w(WiOaCW@Rk<Wjc_Rb&BZ*+~Y1VaOm
z!on6Y-MuFDr6tCXpGh`NLA_R`)7Yfb*~PKPIDYg9<+(f$K5&lT`rx<ew0mSywuw$l
zJifw68&5{-F3mwHjUUCvI9LqR3hw$*W-=KD{XSbeTfBeiea6Sen3@_VmrEH-8#G#I
zlujDLBnJ{@Mq|Z75fH52UP1;j(}x!bRE!XkPPauE1~`rl8U_O&o#aD?VL%W>_<lsn
zGJk(J7Az);jEHJ5hzP@&-Q5PAZtoMjw}0KziKUeg8WqJTp~y{5bKtX2QTgpZChD1H
zDMI6EvsQT$`kOm!UcSua=?Ad1KuH@dO`3;gI|yMYGK<%5@!Fd|<oCbvdvyCf5X$fu
z6Sb&7<E11c#XNDOsn#mYlo#;QE|zo%bq^gw!7iY*Vq<-st(^^={3wS{J!Z_ve=I4}
zAcsS|%sd;z62pwMd(bsoSh#j78PzI!-43l*lWMh&7J?{NIJQM;Y|@No`#okyrxKOB
z`Tv=b^Ig=g*X^*qbN@H_Klk}BbM&F}ET{e>UexFFPtOt5SLoHN*o73mK_9=>Bt00A
z880I&$-s{o_%RtP!pme4j*YY>(#w)Y(W_U;j7^X)l<0R_bh`~QnJiwzVnoDT`{sXO
zvwVVhZH2>`h(j~e=*0W2Jy7iIwdr*Q^G7BKBOlAL={4H8mTi)Xl%m`15`-Z^;FFdX
zy(p$OsIj!Z_A}bXeY8GiOqc*0VyOv*M#O@h>nqG(zk&Bu5uq|@Y{t-(MMNr$h#4T0
zQ)zE!7ZnUB%uL{<J>r&&-|8TM^vDRV=MlsanS37Gv55Q>m7N_9S$+QTxf#B3YmaVz
z3@K9VHg*ZaHc{Zy>jsoFX%3E;*}QR$r5jgpyfpdAX(s1paUHmM?IIx9-ri>6;9-iz
z66NtRMn{UwOpGLISVd)fn-4#{%*@0Q{6<8lvBmDvB_0@o=5m+t;2bI2!ImJS2++j6
z4z*YQn7ork&m2Z*A7YJ;BU;Tm)!iK&Ve{E%Kf_FUj6e;j?rySl`x@&jD}3ckU*_1c
zW4!w6TYw>5I=0LF{2^}NzD1?FN4wL-lJ@?%2K(N)ouRSnT}9pfsD2$*J&xI`R^dPX
z4wJw13zScuAd|~e+1_CG$p<-d*kLpiK@brGZp!1#%m|&;8!WjgR+etyNHgx9nV#Xm
z+%)TJ8!%jqwMHuv^iJ=OHULISB}%0-*|f*x*f^^ji(I@s#nB^&C{Il$NS(1*(9%Lk
ziR*f}o`*je&}nyxwZP5ixOwv`TF1=H%rZV+X0Ni#^2#!;c8g4QhZ83(I^7nw?IiXL
z8FqH7WIWHPE8`!}78M^s?8Zh%$Y#>C+MQ4CW+k<}yo?|CSk_%dAW~2`dWhMlAEN#3
zH;DWwu?!LDSmH_$p<?ga4Gur|60v3DV3W<|SiHW??#>qTb7RcR&hYRP$MOAuz#njM
zVUG3nb$mZSg)x@ppmjtLDZ|&3a=$4y9+-OHZ|)=N?>M4Z$kT4MxqkB+Cy$@y{SPkl
zsiz(>e~!i&o*2XgX`z%rC#eTYN&4LZqobo((!vjYKD>O1^XDHypb%2f?+v(l`v#3>
zmtN0|b-LXSgMm-4+vHP^pW8PGH-{~m@bMzTAVw*;v%GR&H0~d(G>OTtxz<CKwy-Qp
z=Z-UT<v1(vUQH+vMn)nK*p^0yn%xht(f?eX;`od~tt^u&B87$D57^$^=GM)deE*g2
z(D(aTmI;f0_@Q$w-M&Mk*+lD@blM`5wwaunV|{IzdaJ=qc^2FDNT&*@SkezdHn-Mj
zciYrEO&)vh@8WqW6dEaQhF+ft!%9YfG(Jvgz2E3*t%<dHUt<*`go9<ffZlHyTAdb+
zX3JQV!~)Cp$d43FBk%Xg=TgQYAwlQN2-~1TQVO=Vw`sLnpVXcGD^(_);cH*{D%!KK
z%A-8<phGbi(cNp3AIU-#688qkSTm^AaMKwqFU9s=hm>QJR$#fV5!JXZs2HzMKx**o
z4N|31LNA3-3QtM~{Wg6+#Fc5r(DX_L#%GFnmPR@joqCnswRHq(7LH8Q8q_I|P7(P&
zVc?T4q!CimYqt&3-fq*|-oqOkWk8BXw@bU%-*4uBmg{43>Liu~A(J7Cwg}n*E3bV>
z`M?3>>>QR*Osjy=QWh0^xLTs^G+ruC)bHc_0o~SsOd*f-QaEA6pkBkWY;whdQ6LF{
zlgVK@G3kClr(Pk~t?~5-rg&|&&dZCCo<7WSWrs$)N*F7eexFKZlc`RQQl>y;N7Nb}
zR_?6wVY)!6G{Wd;iF`g!vsNQhNR!K^>GgY9GDWM^rqk}w>kXKgI)od=G}o@OvU!_=
zWfSZvbWsuq17kF-V?-1YM<M-IlWt{~@XbGA^2`5-$a2w2BUFUn>(Hp|@ywIYarD3;
zbl7KOeS@nP-=Vg<#?IC%I(iaYNMs^1>2^DGx?Kb=-EN<oH*RqI&Jqtl^a#&A^C=ME
z4+i^>p7=37*4?lB^WpT-tcNk>T8H&N{Byj&pJ#005RT)LZER95I-mrxA0QF?LC24l
zeZrd0?{2rabm0w-oP2<EKF3R+d4bp7c!fs2fp9Exxg6Wu73z&PB3Uq%GMoP4!v`sk
zPm%UqEZw7!8%IS0F1-8=KKIh|h>1C)K!qelz{a+1N@FFYhN+1%LRjqX?$GTt+1=S<
zvOGmTUnB}6Ha0ft_PZ1dec~|Yz``+tAT*<wUI$D2_~pQ$8~gW+ee>urhzP=%IEqPm
zS;j_3?~A6c-D%V5w8?lbLMEz4t&q0E?30gBTUuk|%BrCP2!)R%FDx7cwQINOS9VDi
zj)Id$DnX|+pxf(EscezWmdK`ypd2!pEYs7IY;3L)_>sw5O+-kMGW@1}VtJA@zz~i7
zvBzC2#YjGfMAK?DSzBA?`n540JbRo>CT+;hB5?>3Moms)M7_Pe$%PB=pkjj(o;&v-
zgF%OWw}T2bVHnZx^{7@W+*w&9lP)qjG0&kxC&*^9#>l+7McQNky)@iCy+48-9>P|C
zR`Oo&DUD9lnuda)RFv=y6_%S~=GpV?t!&WVsO;Yssf`z;B{aQCi|*DA@reg;<z4p#
zV#U=<m+-q0%d2;2w5k;IS&AbWUVQ0!u3x`Ft<f-~b*;#IE((iWvB;E~W$ES}s(X8s
zMn<uvpxJ72^Y#t4_jZU8<VMPjO&-K`?EOQZ)~3A|iQ(0-c~e>&&q@T+N<0SR*d&l7
z&%f(B2+KCfDsfDw-J;!Y(d+h1Q>O%}T#ih(h(9o*A~Sxy$5u<5=FXhWHFuWo-1klX
zvrj+E!t5OVR*m_`9>Hqv(DDa(g(CSv);KjpAyKDCduIc`)k36V_NpDu94X`265%-J
zx_cfv2vI?Zlg**%5cm2x=`87dkyJWss47L5cWzZEjmPw-5Anm@bzUqgO6eT!MhC}p
znLT)by>^vMDo3mofp5n3w&MWM>vYhuqTlP`=X{*OfS|HOFenhl;%B^_Gxf2WIva&2
z1sVy7hgYaH-Ss+me()wozw!cOai4OdjweC_*TvRGjVN%Co`vUXBrZ`uLb@)2ZQ-~c
zy;>c2tcbQu3WMi)C<(>U5qgb1qFCS!20XV=<ivQM|M1=&f^3TM(h2q&JJdQ27W*M<
z{W_!V29usk&do5M&Y3e-%A&FDv$wU0ABMzA5K9Y-7-2cswugrx7c1P>4!!C&>wDYu
zEt?t3A+4L_(k2`{mXfgF$FVKKAV8}aKkSp;Tf(ZW5KkNd6iP>=(<zQ0JHlA0$nC|O
zbO&7;%_b0W@ZbRs9z1{$F*mMX!yg17ES6T*c<Zfqn3-9?b4_r(>uj*KwaLob3i*72
zPOnF;TE+K$EP0Ph{?HNt2=3#^p`~cRS+{_-TjR<fev9+};8)0IQl!^*@cRKe4oQ_p
z2x?8-RGL@{ve_InYpeX?e1Sh-54rUAYm7`zkuOYe>c|PE7Zy1G_*1;|?t8ra@}Hr!
zpjav(C4_#+-d>HJ?LAJNcoa(nSkhv;JV(DX;KtG#5B~XYQ8;)K@60nWK8@4~e@uha
zf;b9^q5u_!+`4{|R;xz6UL%vq;JPlh<8b4~tt63Tnz@-WM-I($`}Pv;c86}ejVl#V
z7-Ly_-xkG;4b6KJMv6GnC?$}>raU&r(#q;5e7B6irO|4TpPIq7Jq*qeBhWa-5=TDs
z6pi&NqCG%Mi&%lQCAJXsw)fb&c#*NgN3aESw{CIc?N<?w$E7PD^62?zs8yPbPG<Ph
zm%hq{3-7SJx<agWi9<~kE0Z?1uY9=&kxkGJVE=j!OKU|slcF$^XJcc7_4PZPJbH??
zwGECPKa3(tF~iy?sSi=_*|`H8J9-?UF>y~y@!IPb7#kfY2z&;E9($EN8udD7Pd>=h
z<P?VvAE(i1aq)w<SzBLbVP=Xa&Oe+qJEpnw{fH=xh$Dm2_6Gx2*MG)D5=rS~y^)D>
zRS}!qWTh$0&U5hbb2Ps3Dnen*zOm3kU?U)iX{@e6q|lam&4R$Ey1mBM#!W_Y^921q
zvojNv%0+(l*M61PUVELTr4{;v5T!Jh5DfY;|LOO>!|cog)6?TjFU)iO+I8Oe@Etn!
z21>_Ru8r$uNVyJ`PQ=pk7H3Z%!L~34!Wc#C-HcyzxQ-J~IulnWf#g{nCIkqOmJJx6
zy(o&Pw;FVNJ;ETwacyE9lP!#5yD0{Kk95kzbzS4Xod|D~4im+-B#vXQ-?;fn-O|5O
zrBi8s^H;wPDkkc7ICuOA>3Ry+-$D2dq+@(3bz*Ci%H<i<8*H?8u{3y=jg!lQw1{Jg
z5*ET12rY?Jh@Ht2`XP?(P%MmL35cX%b@2}Qd>I>u7ry#cF1`N_SKj$P59p9~2$M%o
zvDer^*p5-#Cr!JwY^3d=lqTr+u@Ur}U7``6tPZFy-DX7<>HZ9;jsGh3@tQiR@T5gY
zVVvTHKnsP8VdwH~#%9YLn#iKUHh~R95m9oYmP4giZ-8!h(J7m3xj<msNIS+-AyK~%
zo`X;kgI<W2Ef6V9VQi9Kt$`~gO2^FQ;Flhn#=5r3ch~CVvtvvaOVqn9>g^VLVZcr}
zz!L*9ttQHr<hDI#%2UiAJxZ<K$B{Ok=ipclL7?e3_t32!`rR%o-9D?nk8e3VR7{a;
z_n4Z<lSw(KIK;Mdgo6Ppoid}9D8f;SyzLV1Y$7I(8h(wo$md2V6^mTH_%70R!Lr$_
zRJe5Y3axev$8|V$@(jh|C_0X?q$cIr5C!zwH9SXh;>a-;<_@5gCJH0!^*Xg?i+A36
zhugRB484fWY4@IA@8{Z30f812Y}mTG#+`3{pHrXv6l0!;h!wi;)Ay@*BRTqeHQaQX
z$gwzca*nxbm62J^x11?1tgW-oDpsaIeYZ+6o#U~G&oeeMiV_whr4pAeeL!_@mz%e4
z(ChY5aYSi6PcVq7*J?=Ru-ERhQx7Rd?-Et+;645of`tc=I4A^K%VeQa^m-jOwpUqR
zU8GQ$=3oB5{smE_IDP6ATRRorc<TZ`di8bwcmMr=$IRq78*A68*6Z~8Jsiv8#L)w!
z(<z4bGjSXdML(GXsFY%MW(F?2|4H92!YHIxt8!pw4#zb@9i@z-T}q3>;X_P4a+KXy
zE@CT*7SiPFN>EYE=KCLV=1V_^=MK>Cf1l>{YxE{(s8kvRy$;napHwc(^vq$t_O)MT
zuUcby`8Kz1+@w-%?0*h}=zP?^^Db}lqffoo#!4iW@;EeqfOe<DAQ-T+y2X*BN0Wa@
z8Dv&~7Q^%qjqOOHFaT}rf^J;DO0!WR41A6pIYzJ7Wqo~}<)u5^xORgteerXIvy)u9
z_!jMM2Z2v<B!@zv6hx6G3}RzHq{Y4^QN7-vai5Rc{&OXTN#U>(E317juhwYWVe*l4
zj9tD(aAyaN5tJxx5*VBqsy7x1yB(Z#3LzaF*P&Lcva`90oXRmiHp-dPhxp3ZzKWD7
zu3uZEQR^7HCtH(}l1RsV_x1O=vG{-F*wKR=IkLdQ!aNHn&e88Q@OvG?C?E=bIvt-s
z{fqB$>1LhZ__e>sBafU#3Y{F>6692i#8yZrexnMICh991dXXh4nh;3oBv(Td_&%*x
zlTN3L5HMaYqgp!@N@av)(d)J-kC*mU&j=&Lu<az1S}D3+pS897J6Yu1*>jvYa*VLo
zCfjb2?KR1!6n>?OmrsK<V227q)Sw3f!e~Id-RI!J16V1SSPL3ei)t%ND|Yc@kFiXb
zQc5Law-`%-W81X*T`G+VmZv!J*-x`o4LET42xlI9ntyZYJ>I>u!sr)%la;1sWFe$%
z$4L6<ZWA?VAbU-u<q?Jfy?z&epwXcMEfLbiE*|FH@4rD1hK3{{h6luZ7n($z^^=Z7
zN!>rfijO&dH9ujjGzv+q4Y6EGaC!=L^CqRjF&sw|41BCepd@(NG{Um5Z3_`A`rRIu
z6d)x|Dn-B3!L}qij2N0qq96`KgwlBF6rtx4#4(+kPi?PBjNql|5prmLw6e`^EODF^
zhbO0yu0x~Oq1I~A>J1Q4gcB>8wLRLGI)p*QID(>-1g^zuug7M4KtGQ0V@+%cik9R7
z*QV6#^1zhE15+_kuZ}%3j}QXaa!_GR7)AKq9!@NAZGo=spu-lBLMuh9UgO<2-(cz1
z6>KMilP+R88FXkf==Ry(+#>5{nVyLl^tzxFQV4?3XM1mhzz+$6kRXic4}7}a4vl7u
zO0z+$)h60ER{qQC{&>j#vTCWArV1(HV4+!l<$b2p5_itSas-Z>Gpe*6SeXoAtBuH}
zv22gx#3V<;RsOqU8D7rJ@yCl57I$`O>~0fnuJGQ-DE9aSb0<zRF<Rn*Q)ifd@e4GY
zb-w@IZz4qp9oyWxew%!zj4M-gb~n(IC89v!wcA)9{sqp?3Ox7%WQrJE!i)l|)jjI<
zJsOQVb8{!C*E(FicA1ThP5S*lj^ohj4Y<8{hw~3VOnGb^KMZkPmtL<&v(v`)ynTxc
zGvYIwy+9^rq-K+uotoOWsQDx-3`1)5ntAqJ4<XFPWZMp|B~dAh`9~k7abt;iyN@k3
zftFYz!cov!S|+aSkZaUXL50T`M)>2UP39)&xN`Yj9)9RiYFiaNC&R|ZE-$_C^L*<s
z{)n45t`iMMQCb`CKmjO|>uk7M`X_`z0>~E%6bhr{GB%@y5|wHdR77=q6M`<v%ic3;
z1MH!-1j%?jLPr6$YK@yWFVh+HsMMMSKl~29@3XeC#Rr$J<GBuB{kborV?`>JLSvNH
znXFOLM^Th~-Ss0JD%3Z&w($MIPwz(lZB`HwM}ESEbI{rcz)I|3tdUV>9zDgK%^kF_
zL27g?u`GoI8cQp5Hr6RjP9dEXQLn|y;vFo<Wo><lY_`Pt!{<41{1o5##-Ag!CYMd&
zxgl{BnM~I(B2bErojO~U2Cu#KAyS(B<WgybL-Vt^w#8PZ!Rp2~>)SQ1-PtCN6e|8T
z&Yyn(gu2I4gv4)EfHF)vtu$e5P&%oN7z;oPiFDmW<e(S~2J|`|`rRI5<73Ru&amFD
zP%M?vS`h?3#X`ZL!3k$?Xm2Bu(R{61qf)7UVz>0aR<>>P>%a7CpdtqK9S&HUbSVdw
zEi4_=-`T?&EfKW3iG7Je`2mpv*LE4nXK1w?ZnloImO4RH$kAx^nH<j%Z>?~!dYQA+
zyEv{#tTi!;T4R?%zt7Z>lgvH+S;jhjvc&>UCP#669M!Dw#)UgvU5RlrMV|W9X=Y}N
zym#{|dG`PlC5Oik#^gTu5zR(KwI&(KJLJX+bRwVK+aI#Ew8Tug%+y4gx#?;0qb0Jr
z9HXP7^!q(3d%M)CRVtM#yY(77yL<S4n8c6{_ut{KpX|Q-v6{M}{X#-b-v2TYE|Nnt
zB`hl->U2=O0F8|%CJsZ=`5eM^ke-WWft~ZvmI+u7!Vu}WSO|1%PDo)GkP>2lB$6xS
ziBwD&_;mXL+0qCpFXB^k8D^Y_*VcR7>Gf$=x<r;Flg)DMz#PK03Hv@y9HWAmX5S|Y
z6ay7eur&QZu+a<XB%WAl*CMMzo+^1fJepxD8<BGc#Gy~VG>+$bIF5s3*-4y?px5c(
z*)BpTM01^}wM?9ugIJLEBu5V&;?~s<sMU9IQxVfMqm)O-(c1hxH*Q?##`Rk?`XPZ5
ztZnV`2Y>c0BCV2>;n3{xCpRxs`hIzy{>l{sn=0U_bV*5QhKk_MDvV9juLro96mBky
z-yPtNWr%uRRKHIw1-9c-7@fdxwfW`a1!gmf|M>>2b^r-M_PztI*PG;Ob;?_tG}qR-
z{>}wP#>-@i1#%+=vY9-cc8|*LD(DQV-D7$614e@u8%~A!lLtt-7P8+*+_^w&dlmQD
zUqt6e&@t#JWNT-Oy~+*;=N5SO>1TNG+@rkt_Ul}|a*1xI$xAPOmT&#}x4FHzWJWQO
z$*FZ52Nf%}cXlY|&Eh^3wy4-_9z*}Kp@^b1GD6Dp{#G&p|E6{4)@`1D=2;vsh2x}&
zq7c`SSkl6>Y|`UprXM-W=J(!1hl)5h5o@s!(5TRR|GQXYF{8y4kCZHychk&1`6%D{
z-VeEc^*V>61+<FU>-yv}B@QkeW^#I(t(`gmQt16tv{pKKFCbRRjArjS+l-Bk@#Qan
zo}G;b#SyTz&B<dY$feT+Ym4+>`zGnLPe5s!KmyW)#}5%ZV)nCnZ{E7b>gF<0Y;*0}
zbuM1K$O|vLK($&$X-Ih<#bS{`e_#}@u}OEarDcRLN+nT+YTwFa*tD;2ZSj}B()W95
zjlZ*nK6V_(PGj31T6G{X9kwljmC7)F^ccHySMV2WXlcC6EQ~lnw0i6=E;4uiNm8oM
z=7sNY?Uk2F%^u*+ogJQj>IG7sL!;g!3^m7(A7$XT*;wBu4mFNtA%scQ2`rFG>`y7W
z{ebOSmrFNSjVk+Ige8vo&MPmIPI;s=9uGWl3`iR7p?!&z#99(3dybBS#4FUOpUt0J
zNIQ*AgcH4956cq#onQDW*<6WUzr#QOzb}(7j-uj-ScT+sS>i|$`T=1O5G3g}LON&-
zD{E`_#k>BWtK&zG@#upO;&&V5+ZEDbA1kO5bgBr)#tI}1e8QfI#4@5ko4#_e(;mAG
zhwts2B0P4IIE*0bQSS|)Z*%(D$Jn?yO0n_{4ovpYm`1qKs8cG9QCv7dZsGu`@es=~
z;*{KI5x0^foh|e5Q&T+g>_a9Go|tTDgyrR3t}go&bc&hXIy+U1=0K30NV6$jc0z$Y
zI?6x&@BazYM@}*`KSz0Tl2U1mOeVAMN}yt6GZqE`%~qXj*RJu_n{V>!Yp-+h(lu(e
z8c`JA7u)a0YU(n!0AV3iv@a>Qkw|5e9f660B&s#gk?CNiH0VOq?-K&jOCeG=PTIyv
zIVgdNe2Eo)=#xr$NF>s6==mMuI7rB~no!3$uFcee8OCNOO{S{U^m;vP=`xmTQ76sI
zi@V%zhUi9zy?P5}IcVF)_B@<ymdscg$4VhY%&r>H8HBi^PQKS84k8?(`SS5`J~!>4
zI(-7aheDFe78om(NPB54+X5AX7Ni{)Kj`9Rvp9huvIq2Uz7J<c5pj-uCe6a3<7~HW
zcB^eXuSQodkxFF<qb{LP_(~H7#;F<<v4YOP|9E%uet4QbW^H;BD`k`N1L~0=Rv~_~
zi3Rw9Z!&@N8Ki5YZ5!3@VmUTZG{8%zP(eVu(V^V!@xMCg@{Oe~H*^}`vRRJ;!mTyD
z-3^Wh2l1PE8g~Sq=MgA@=rwV=9ad`<szE@>^BA<c4Ek+Sjz!vYP;o?a=@zcvq4b4+
zNNlA|)XiW(#`5^=GtW`V*mRl={BDy*b%!s!^dj@~2YB_B*I3`!rru~1#)@bd3l_y(
zy?UK|CeQTjOrp3nF?sh|(hQrsLOw@nq<CK}^m}Wo&>P^|8B(bX?RM=R!GgfH9ZIK<
z6YkV#e0ZA}O-cwNp_!V<ayZo{CDZs?Fi|Y<aG}mdwaS?XPVnxV*C<Vv2!b9HbF*ag
z9_yPGTCI-ZjB0Jreb^U94232z^h--nHUv{sWp;Nf*mj!HQks4{<iNp0)b~0BgEsv;
z@8fSR<DGjF`}C7YkxsZ%8kE{6uGeZ6?kwM?R;zI6(1Ub37Z@8W^TQwhh?$vb^7$P3
zTo%vsSl?VX_b`g-_4?Fmb&4Y;l#ph`tMo^jx*!Nxz7P14f1T1wvH0>gId%95&fHuQ
zm3H^gDXqjO({-7B>_Jv`UPTW=GuF~(7%P;bv$TviXi(T#Wb?xJ$Pap~uC38%wCT2*
zq|-Tecei-_iKjVu@Cd7GD{SxV5l1nWZKLB5r6G!k`dU*_qzRQW6Etaz^mKxDTb9LO
zFyMQye3zM-DW;|-n4Bo1bckgMkhU=&*A_a8uw+CmKnRoj97GynNlN8u(zyaEp-uF<
zU9?t=j*c@mb$|#s^Y9Z`ZkizQkwW0P9{qlgLEog2L}7>!66u(A;M(=;_pKe35Q3lo
z+!t|e8-L|CqxCyTJ4M*q#BnTwZiGmui4jC`h>lDIVk|tmfhM0D<+6Gdb^ZuaTZ|Vx
z4joT1R$gFdCq(#g6OY@KDW=A5<GPY&E6H`YU9!i{LFlu)e3?Rd3bZC;N!os!FlTY$
z?W?#Mhm&V!aUGAZ{GHDb1~LEf-+!BSyUOzJF6%mGIW2f|(8IpANg(rts*ZPck;$10
z%*`HPW_F&^NSWE$StchZ$>sCp@&)qwB3>#-p*X_){2^X?@r(5OeU_J(`0jUJ=8yjH
z54mvR1KR!mCwfr(SWTU4g(xkP=EIa0prm4QtN=luAc_*6jWp^wKq^TP2Z(Njh%8(?
zOB6^f*Tr!hv@j%eNSUY|OQMw~3Ic2^Weggl2o=S+b_zF>$F(Iyij<upo63+$*&Nky
z`rw27=;kha&6sMvK_?hM*r(p_(&%@wCP&%YS;7_$Wh;f(ZF17KOqzx3@cgL>jz~>s
zqY7R`F_$Ns&63GvDCCO>$1y)!9ODRqux$g~OVCn~k~TXVo1~94(RPX`ih1?j_gJsh
z31UUJH=wq*YMQW*?ot0gSDUuW_-9|D^^I@fM4CYmlb)L+7__iZ1ib-vB#?bcI-jFa
zZ(&;wk&3YF2*<T3mh$8$#tC)AOew>K)hgfH?6C=m@lm9-Xmz_L?=SGNgyMSNr#1+Q
zH8@gnJe6X`_L(eZ>G#_d(pd(b4%(7t^I5rqT)sqn<UAV1!2=5fL6=^u#f?Ul?cF_M
zzs=}K3JXojaTpsLWn*KLTelZEckUsYjSjbN-ehZMm%|4S?;A{OVbE@*i5-Kx&5G6<
z%eI-Em|$aT`;)y{z=!X=!R|Y6F#hDzq%#@XEn@_%RD^B2IJSdw91cDEG}pH)L^~A_
ziUWtnc;OeGB%5>0NCrsf^PHX<W9`OuW<K>S^9y^dEH856_$jJuTc}u38p(6!^lAE?
zHnnO4*Kv@RAeA*I=-o;K9Sfw?Cfijf>P4;TcKS?=Eg(cpCIeA_K*~weZEllOAvzSe
zY8&s`t7y+f9D5Y4^9X&{x+01~uHCr8#^x%=kDlV`XP#$jW|qqrKjh+vZ}Z@T53#nk
z!N^DnghjjCquuGyY_<$|E~FvWCn*$0F&giqu<cHVO0D*Z-qXK!Y0cLAA8_T*exI|y
z@*AYw6jtKnrZI|c*Kt8Z;qXDG&+V}L&J}}3lO%-*4ZF8--~Setu5n;2!<j<N@@9jR
zhYoT3=0&oZhpFvWDHO&zbM_&=_{Cp9^G9rN@6fEZh?GG`6cVY7nt8t?7djcGe^k|I
z+s6F9)~xg9J8y9Czye?U>X&d_JF&QdVaiAx$HXc&i5Qju;SnPko17<K7)#FfF>xHx
z>$MR=;@CExmu6;ShR?qEW$r9qqu1||avh`)41$m#4DgeKyO0tUYl1N3#*JG)-TU}4
zYPMA3#YdkcsO~VbegQY?pqw;%(7}%)EZaqTE`vrZnYQ^z1dT?6lt@$Q9pL?x$H~Zy
z=WkrCvAz{@_MsdnPt9@l{Z-0SW7My?bO#|wsMTvoEVPp*yKs``_8Q^xD)MOutzrhP
z8iRg7c`V22vmV<U6{?jAxseikTW_(w(ZCJ1d2{;%HmVhF`Vs9|;ni&7l@2J4B_&Tk
z`waj4|IPn_qbJYcri|qb35WXiH@?pw|H&V4=-^S#pMQeWr_LD<ODV`?Gn_hemXjw>
z@*7|OI<LR>D*vB<`|tSP_kT#UZIEmD=~8DtR#O*Q5<6D=FA5kFNL$BvGQ{=-&PV}i
z38E;%aa{-_;&u-qHF!2L3YVC$*FyuYZIRCBsCRa8U6*dRi!CxpA<Wpcn8CurmNq(y
z2-|HOOAv)%MGAz*krw$(kxnomrMi6SREd?X8XTVo+a)7hmhY@kn>x(Xzy1&S_N(8=
zM{y*Zr}nKsqP^EZ6h=5VU&3we5-FIN8pllua=8MXb_d&biQ^bYxLDF6iUWjeEEUqJ
z6k*S&*YA-}<q$y+)vY3mV|2THE?>Sz6dV4}P%#FWj`(L0V*6XFYO{fS^eGlvEgIi>
z6)#)Fvy1HPt}s4bBsG>tx;9qYgBY?CV+6H2!m<haeY|v*u+>3|kW4zo!RZ-BvuPeV
z9P{p4o15D$D%~Dik;Wbyr{Q{Z;s_N%CNv1om*;alUbIQe9u8exyM(kQR;-CFP=gSm
zHP)Sr#M1|f?E=O82%D=by!-l($Q8$lG~Bv#hwh*U+IUSBvKf}6n3uo%Jx~(Yb1CFU
zxPIdnOUp}~J#(79y~@7&&arJAX`$6mdJWD_PjTVHi=X7J0-#%~@%F#}w>&daAU!#S
zFrC!CL969B#Bq!@J;u>5eVUcu{|2)MGJN`_^JEGp8jM(>qZr5Y$md)hEe-hg8}HDM
zd&K<~Ya6SaI(nSFjV-im@ywG?Q>*MUQmoQxcPNaEFj_9-`#oBn4v3hUnMu5qi$v1v
z_1WCq;eiLvlTPQ!W@F;0gHo`wxl6rLVYb-E5jm8SNMiJr*I>{me&Dlc9ify&MLzY)
z9@lSOV<cbXGtYbmC*{y>)mdJ=&X>RNMPn!&1dNtSq|-UF*&=Vh^ES5^m-yPx{XC<k
zQ5uap1K%g~0|xyboo<Itr%SKbV`FQZcIT6~tNH7c1_ps*`@64`ADtxs#6t+%_`GSY
z(20S!alW#deflxF>znwSd&Za4fw|+;Jn^Zcj8I3$np8H;`BIAQKvFn&l5c<KM=al3
zMiG%tXSlPn$?VJ#o_g}LeCu1k&t4dqR2PPQoEYk&QF`d%X+G<{oKzKsjE;@cQi{c;
zMXukx#f=-cdEo3Zlh7j+Xhp2_{%BuH<6WtQq+FgPmmk@WelrJvm|nkY>`A0V2#a(o
z!}*6FXK#0#N~O;6qX&}DXpDwqW#Z$EXQH51tFykj`P03TAEVA%HnP_uTE0vsYMCiR
zFu-ywq9DM@=ZFVB{Z<D*2#lIEim-%7AwSN$&4Yx%(t3@D&Q4LR#;k4(=rvY|+7>&T
z0h?>P%&I0)FrXO?=(T#JQV!{vdA!jw>la=ppDPffuza8G%{r-m2mA7$GFg}>H9Ch*
z7iokY(q4@1-=u!)1FqGp)U{xM!V$(ieK>AEcIp)W@E`v(#zrQoSE|O6DKu$VAh>Yh
z0&l<dI@fP}$oGEmC(O(o;F+gB!>2#}B2$xdNZUi&l991-zVO8_@$}PA^UABQ@jv{J
z|D4xedxIdnkDEHDJWa5<1tQviVP&kaL`ZqIOmXHgsl0>g`>ZW3GCws+&>NVMRjkOB
za)h=SugXZFV}%lu%t(pQ9}tEEqF{hhIj}8M5F)fd+8$9lM)(mj3TSVykuDU_sT`3~
zs31c55yU=HnsDv0TpFnrPS!z=6mUjP(kah!^z;d;LCg<c`9ASvmh6Ek78b@Bi8b0E
zP?(=4lXg&1AIGwY;)qT^WOOo1yVqqbpF``I!0+RwQ%E7nX7e=LJNTV0BcqyvCD~nH
zBAY!#Ww*-4_HNS1{cOjN{B2hlhFrXTgJ1aiZ?e0!LmbozRRsoZg1SU`HiJ%^R6b93
zdKxE}qu=V7MvDk(J6QP~s@J1e-9=_|q`eHeu*PQ=vYekvVP}i1*Sch%{~B*DE~0C7
z!nfa`?WQ<2Q(#{B2t6R<rO9Tql(Iz=!xk%C;UJ|Z@(0+pO$OJ0Kz!^OTxoN7;Q(*^
z=(|+wA2KpFM=oEc-feK{gAW;W`ox0)5{*CT(P>t2Qy!&4o)@3{3}F<ryR%ERTBF(N
za`nm;s?{2)v_~eBA)85)NvFxAGo;fg`u+POZJ-jzRPJnW<=_2B4*lv^aa@6pRl>Cu
z*shC2;P(fNojXe*XmYL;lFoo0_*khde&i$L7!^lka|LF*0nfyJgj?b-ZZ6TTHtDrm
z96oXs27-6K_d2HzpTK)4OFmcRQ_sD?hZnE%7k~aIoH_jvsdSEXCPlN^Kr4mobg5Pv
zRQI--ogT;WvV8c#Rc6YESzq6xzjuox<q_gRA2*XmYS2-i=EWZ((k0}<Qzm3Qj@VdR
zChq%u>2qHqpSHR1;k#VF{0_Z#6<bJ_*H(yQMZ4YR)~!2qx@|0J<JdO!Mvdk56*e}v
zdGpP;Xtvw<flnAl#8H%Fhu)w0%zxFAIK;gkm;dw+Iqqrfg<0Yt!m=$aX_*nMC5Zwb
zYof%_FFwce@BT3plPMm0{3LTnX0TiZQG!Y<O`%xeNM)D3cAuli4|DVCDuwYp(Y5R3
zC#R@YYfMeeFkUWmXUX{c3BZyP+Zx)CnD1&mOxpRe4=x~=&5+Bc+1cG=b#0kTm#%XB
z=m9L-G6@ALGOBlJf$f5_h@*&XK1V8@W@lrQTq=v3&lxMCFrd?Jn`Tlc0w|>zElqOv
z%)@;5JAY>KUK0f^81-<NaJ8Z+Vtsv`R`dQ;Lg$6#0WE2--r`WSh3i^K&x0^V#}TfV
zLWL1ZKq{Z1v)!Q6ZWH)1UdqPJd*tR0Q%-d`adMvdaf^qady-D<5Qk8$RCxWhiv$hH
zSymB*Y;A242QgmGW9p$#5rqN%;&rB<|2!&I1cM&U^&PT0Vx)cpYxN?7K+q?{c42|l
z0NUHjT&Xu`#4>57K}mtHHBs_>EN*P@fBhf+FQ6lWU;ya-l!`EnXf&Gm{s7CCWV2~D
zHa56^{R-ds<{$IyvoG+4FMN&JnFTB<5tdCZpXJ5pKf}Z4&hc-4>$mvl|KeY=_fr(V
ze5|JK*z+%P{r7(dCk#-+WY<X{D2|qR;LJm090ydFt@UkMl?L%#8DU$b$|ZV@7D8&$
z84qmNoN}VrTxRK#8!ge@ts1#^Z1VP0U_$JL<Kkqq^ac$w**qBwYI}RQxjvcF7+yYu
zbZolSt|8F67Fzps{Q={d6rSHVW{4ZN`N6;XzcH9R$apDFXMF>?vqz?6Gn&sK3k62Y
zWg3+|Y|Aq|fj|@bDI?R41LOD~$An?9&wa?GGgPC{oa$^zN?LR)TV#$0)aoV;!qkti
zF8}}L?$tM6<5zy+@9^M1_)S*-?04AdD%?yO?bxU=!m&JpdJ`4LWXltz3k4R}mzbF>
z5{3a@E{CrovYtz~UB^ym$mT`}dOdX1!iy|U&5Y35y28n+dDcCb!qF+_MMz4;<nwuK
z8DU#C-L7xa9}p-Mw$hB0M%mt2CGvg5?OQ~}DeTdc%+DWUZ{Toew@a?OMtM9ioGc0N
zU%W`*YXYUYv$o0ZAO9)Bwva*^2X#vtP+lqgD5Br*6NW=iK{Jy#Igs~PhHt3B*OKb>
zt8{-aWbO-}#V?ddrBZ-E+BTLY@dsT}I?bsQWeS05wulvjMw3(_M;P?bi5l3;rb+Ma
z@<<`Y%7KIUodN4hx7gcWWqS4?x!f3w*KboAALD_qoFna}dE)%zI6~6*eI9?}N#ZzW
zadDCHv2pe)JN)qbe@@DkXf@#W)ir_+^ks*wJ6BkUYJ^)|)WR&5NFl=zKMtv{ui~Y@
z$LP!<LLt#nK(Eu})Uo3P{Q(#MzwEttlx5jjpZ)B8(mgj<-O5#+^Gr{aCM%9|7Rn-;
zXpAw~;LkSZGuX!VXM+I)CK(Y#&IqBzNTV=0PxnmcSY0{XeDBHU?EU?5svBVq5+jY!
zFz{PzR#$benpJo26W;xX=Xdl7^A|55B^)@opKh;p#h=r$OuDTN9LMI+zJ2W6wu2~&
zc>IZ@oH~1!#l<DEzZSc_aPIpQMb_$Y?qeTi`mJxqo!w5+H1g|K=0RN+2I#EJ)HMg0
zi8t6YDsd|&C?8XmNzw?JDTLM(s#UI?i1_eRM;N>2S~f>Fxj27;{d@N!H`bZny_dz+
zRR)9nnJTyZQiaH~UFgdLG{)9Msrtgz)_EqcUax^xtgkQ8+{~v=$0x>7;xd#vuf${m
zw3I|~jBRT|KPL79f>e`;d}weu95VC<Se8kaWwhF>6w4zNi$!+r-p`JmdngoK;&=-@
zt^R8I^)kxn{KZRuHR=64(~IpQ<qU==o?t#&BbsjDSQ1C2c>N(>yH65Cn2trcF-p5T
zWY|md=bKev!Z0Yi1;QpMl`?nrK~~q+ICJVG?beW$)pep&G3~dRtTgF#`fO~psg=vj
z?mkF;|223km$1AZ*627{mN95G>8@@vK9(^z)?lQT(vMxvb|O~#9Tu8xmU<pTArNFa
zUl(%!`Oj0U+2+A^8=0HRU>G9r&y-v4KHAs{#1+lv5Mx<Vt+}kPZ!&-BET8?{CwRl_
zzK%D%;q5d=8(^AfB5IW~|Kgv0H!phejr{0;{4tIkeF~-iy2H^|L(eLxo4oa9bmlLz
zeE+=|2^cwmfqi>tu+oq?3D9&|I(L@w@dn+sCL<FK?4pAz6lkrkqO_*c7{jhs^YYE*
zzB1DyN;QLKm)Tm*^2x}{Px%8vU=<v~B*LwYFlIY+Hdh&D>y$^wFbggtqvQD99ztm1
z$R~<JJb!>lGECcIS|{``Jw<C{gJxlb^!!<_Qz6qM(~OOb5yT-`8sRt&NgS}T(dOd$
z4*L)7r`PXbnHC80UMPLp)(H>-&kL}7eX5-vg_1+jwm@tAB=>H8;XKBlTb(_Bmj1wF
zWal3Aov){V;UrU2E<|mrwK4!n7!Yi3;IFUaj5hGJg`a{E#~?&bArC{!qhrLq0k$Eq
zD^-U60N?9kDuvlz2k$7ukzJJfJ+|+e$k7~95C<Mvs_;C6AdL~3jiEE*$fr`N(lSii
z>uXHx*iHA`lZYGU=yeBdw!4H%6Ap$nw_G2@=OtqE-Yn?*&t}MB3nQ(`8IRASOirv7
zz9t_adFnW+Z7})j*I*YMl*%wI8^g$_&BuBxjL3jw%eWh%NV@}KAut^qEpzUdEp4J<
z!lAmyW<~R;t};0xSX!Uw;@K0JZi!S{40~Pv=mWn-xmIFidJ?x>r!hXk#j_^~!vLi;
zVQ5h-2=?vW%lhgDtIKVoP%~O^Sv~a#o<G1cGD>C@l|-l{M%gAw7!t)k)xjn*?4#W(
zT5HC~#yEQ9QPx(LiIRk)$Br|<I8VVTaPy0ArQK{HQ8>26*$ZcAw!3tDeY*VtgTat6
zj{ja~;=Yt}mllc8l4PmHrMn+w@7LAPuCv9f14(`z+i}25*k^QC4MQbY0{tWzA_^8N
zN=cFk%QPstCbu?XK6&muzSkm41<iJsoio#1I(wS&?b|qf_0=>tH)u6Gm_{B%ULJAH
zE~A3<7fwYRQnF*`4o2&vxQ=9^K1!?8CWt~-SDQ4(CQu@`j>|c$Ti6@{e&BKL+!?YY
z?;m^di(bT$qffG9=T5XvucQRSC?Zh?y<QJ#NKDgaZf+-TA+JuTOp&J9m3Nzz`2_EU
zi<iD8`}n2g3c>Y`i<bz>Q3oAoTsphVzNspi36-Kt*>;HhfY1-vxO9n>3?WTchj4h-
z#2|v!MRx4Ck@+(VsO$)zy=$Is3(6IT8(wsniOma4%1@)|vb?xPrR1`A&mKy9uOhQ-
zTF38av{6U7CE_?{bN&LuPC#7I6x|vNEy?}!U9#~Si51~_0bZ(5mzm{%LvLF?K@S-~
z23v-Mz~#DH_Icldh#?Go+AW`2txvBzWPN>&#~y!}_r2$vdGSrRQmEAM2Saoe@zUGw
z;MYI!EBwd*{9pLU$3C`&A^Ym-Sp{`^Y>fS{e-q8~=ZO|pK^oYGpixRmlMSS8F<4#1
z)G6hniw-0D%WG)cBp!O`G)8F6Fiddc469ViN61YBVVJn(66*_VL}9dbu^ah~Df1JP
zU33Y00Z9^JyJg0w=IA!p2sc{TjZq4PB4%9@_<f|(7@0x}jn*li7ZYb8V%Vb)+LUK1
zgtp+&-UCdH)zFwUDit!3kwy`%^$j-Gx{TJR7#WB4=793(OnyTqX%43(Z-&+?L!k*h
zkD%GXOiGleMo3dhiu}ib?u)3??QmiKA~#%h73I-MCRI#TcL-NJwtU(#Y@1?zjG*5l
zSYO9TecByOWvYhC6w<Qt_QWW`DpfE-pE!yc9iL`xb%88PNaL94f{U*+l3jaIqeZHQ
zMoYy?yUD_ZHO8ms@Ph%aTSSRGm~?9+RBIz_F0U}YZ98sfgTdxSvgEQo)#bAKUj<_?
zR4HJPDYQ0`nWFdDNknCo(yQ(u5hfroO%o>>P#k)srq1vAQV}cz=@|H}Cgt%7oMMr|
z$|l8v!wuKYG3a)AX*FW^h|9+=TEz7^mOD*0J8k?}5!oj7Y8ln9v9`2^XIW6pEre{_
z!mvy*3^J9H#3@2a3Q0)0*JOEZnUx^mP@%}SilSg?3<I(#Cbca>->0AjW!pss9$00v
zIN{jQBi#M*50PXh#quPUQ>ADZX>M-vsZV~2jb1>QW!(4hV`N%?fuqGg1SJX#Bo>m+
zsd-i&d4fH!emRo7A4zC%bjqZ^&ZLpU)f2H5=<AfI-KRV@LMjBL3KK!GR%2qY#&!09
z{>Uf~K7I^u=%M0}*||B+9DSOD2M@Bebcv#~%y2LODX7${Xf0S-*+6Yg^W|Ma`OuyQ
z_Uzfst+(Dpb8SGinqf<qTD68P1^uOSaBvsOD&!?pk%Otl7ME^pY@C^?8BmHi(X6bl
z(rT`;bNf!BD8>(b*4LML=+OtLRVO)cu!Z9|NNLjRZgBloJBj1y%8*ogxeTACc%ILt
z#pkE9$hEdh5sRVJjFxTeG$e_jHrdFB2DiA87$7XcxMMKzhJ0!@?`54T+n7d5y?H-v
zV}!BUDNa1pr6g-i?=P`!vP^d2B;%VOWqhQ?>e4#(dXYW5c4F8LYH}N1w~e#00Obmq
zSs)1l)=wTsCn;sOMrPJ|V0i(1*A$CuYn(rI318>FH_!IA)L-zm{=fb`|9{3YG&ci=
zgM{HAL1jl64n5ADJICwa_!h<*<HTM-mPG8?wVR*(@t<I%KFV+Z-tS&P*?cwh>^OA>
z7~AI9^SU>3<oA9XIn1cIHj^_|3ZpK8H>B5WGC4Vpv~o&2X+k>iK`9JrBeR&q8z5~v
zKV3OCna)q_!ZKLvCzJ*O6C))QGC$GjybPsn3$su}Zdt2n$D%ehOT5v}{bTC_vrwQQ
zt2l!pK^S0Y$kK=;iU|CKQmI1h);LhCa&CGXPfLLjz@!zCNP)I3(ljRM_o)?K_Fpx_
z=Aea?Hil&qM-eK^uB7M%pwb*5YZwNpiU|S(%O4U3eN4kf7$%yS7gE{fxl)#8-1pG^
z+<5pJlw*Ngf-K1gQvx5eSj<_7fN45J{s4;ve+bfSprzPqRyN5jiNrvf5-ZFnuf`|m
z&`C_OYC@dRxp0Zfp@ZC;xZK!VVn@A(KvFGPOiq?r@3*!*Av5AM0j(&wMf!sQYZvBe
zOid89Hz?KjQYyJL`@?@QyQ+bqR-{TGkrXmmy#F!Ev*S!&eGNh@Oj|OM3^8?|P-&u0
z2VojiXD6{s4ry=5V0o3w*chhkVAwW}X@Fs~vAoD0%j2J3KguW1q<l_{vJh6-?6+w}
z9*?)X)cSoImPx@Z5Ib!|eT31OIh4*w`w1m0&%AY-8?3h0Sn^{cowCm@u+I&+YRbZt
z3PTz=hKW|1VXu#6*w`w>@6VGJijbykpPgnH3OfBE<JKqz8Ihofw5Hn|5^o*sk{56`
z?u$ziYe`9NA)qtGg@>P{v2Q1heFq^55K=Lj4XF(_iM31S2S`ieG)BNS5kfO)wkS77
zF&vwq+r!q9egHME&vj*k&YlDG);2hC^ikH=mf5}g07vhAlwH%?x$4kiJkMu#+YZj2
zJIAH@vou?|A9$YWSHxjL9Ov-oFbwFlecYnU$cR8>MMh@pc-@5f^b>@o1n1h9psM32
z6v8kOQs(`6LJ>tCWGR6kbL{9N#A!?rdieg3wbfPH?G{%bxQ6=3I9FeN9YGl7Kys0@
zD&wu}tx}LAnlOxsqm)*=PrK9oT5sqV$Pt1YtrE=`DiwueCRE%SR0UC-VmY}zqBISp
zZ8Kb6rrqz7CK*Yp*i)4_PQK4`l}(b>53{myke$_OB1@uIFJWCe!S3=Z<%x*p^;O2l
zYV4TXg>6gLJ&*Cx8M2j282t`|+9<YR69pcf^CzejG!vr@=DP{@j%}P>xWvlY1*8z9
zT7T7_%^HZpjMbG6VHlBRia3tBFn^x6zwKRIbM^I@mO+*zOifPmpa0{3!ZZv%_?sUj
zPSdZ-@AvG2I`SsNncHq<<>D#)Pu|D29TQY1ieza@*y>;y28C*s42@;lWJ+MzCI!<%
zV4~8P)N2!TnivAy`UpyyXhWb4n_g<uk2I5lEz2HID$6r^QeqYgs3ZX)P)UYBVox-n
z-NA3Ra7RZmOGSz!qjZNoG7Y9_Q7Ay6T8Hu|j~-oM%q~%{kFbu#V}6(088BL}V1@#L
zMWQsj_s)<?g%<==D<h;yiV*oa7>2Pm>6G_zn9^X-?crD!j$z{j9X6L9$1x``4CAX_
zNB{e*C!czfIE}$Fax@GUR<%yDIV5bgz;;N&5Dgfn$w<*4>c$8RB0plYHNw(xl-RT>
z%NBE$J~bPpPAS<%LL^a^Fz9YFTh2LDd*Apb&OdfP_VP)Tfco?dgUCl1k|>HWjojQl
zO=8?a5z{c}by~QNjk~&tJHDUobF<8^Zt@Q)SZ}F>NIMi|-g=M@Le6~pe(H16$i_59
zY0|JkrV@#ySQ{k?LV}J@aiW1yETVf|{B{ScS|ZIdOjBYN3zVxB;$cW7OL^B-<4l$|
z`R$WkjQR*tjM5K1f*{~b6d{5>=1?%^^-28!QJOLuB}mKQY#1;<j8H&XYF^!Nxusz-
zQ;x7rLA^3Uq2OX{;belKPor8xrkZf+4Al1^lwe_PlQV05!Z=~EcN+MsQ~EzD?_)zr
z3=0$rrA>lnz`{L`P@kPf6zUYEU`|Aovy9XWP=Sx{c_c%R%FG02(ZXME6AFn{Ez=ER
z+;WBK@gk!|&5Vlam<ea5<`}M=V{>($PNz*{Y?|hv#nj9UZ+rJU*?-jytgp5B(1-qj
z$;lnWaSr?Syc{pn?RJ@;U*JVI+|JbG4lFA`;$cz75H^E;6R+fvoO_gHbA`g4Ux!iO
zkq=6Utqgpo(P=^&g&aBhI19@c5VF9NM<2tn9nPLV&*@VqSzTG>{qK7Z3-iZ0a`Y%e
z&*Q3n`;e10l)gOGo(FhIl7p(();9^GXUQi0JJ2@6CQ}BHP#m1fNV5oB7rWqK6dY`~
zkY~ve#DgK0Wm0qtG<%y&YjBZhQf#|IntAMMG-(8%rQg;Fp(wfncf=qeA<kk(sx_v@
zXV6)WB+F_uC~30r(4+L9KF+RdZpDux{7wsR=>ij@4%%=ytuhu?H(5NtL{SQQnWp!5
zbl6gWPBopj&+%jD>GuK#eV>i>P2Tp_xATgZy&~tU#u;Oc2LIuQf0)h9RX+03kCUqB
zKJa{Y<yWoABsjHEw!h+Kte&37X^e1u&7hoyxM7N2DI*F65<j3mGJ;tsY(YU4B1?!9
zg*hzXx7G;SEf5l;QYEnjrWBN2n<S2a^5;2Na3vrTR?dYJS<X8m&Cn<cjZuU<#2*YO
zS`MyTW^`-<+c0o#C+AtYHp}fk>l<xGr^l#O%8cyT$MIuNGFaW<ZMH+L2E$<=+pSO@
z8DV8}bF0Z+kYy=`%&{bweGegTrnC%$ILQcyK4~?ktTI{`7TGB29HhGiY5m7H!WZh|
zrHgE=uQ67ukOUn_15yRj-Gcq5S?<0OMR}m$7E#?H?f!t?aF$28nW!;Fe-L0Ba6S&%
z-aE$O#v+!Av2B~dV2B^|nVy<PUv&_XM7;Y4ew2^>_<tZ++GOTcZ^z%bL@n+j(+HjT
z2&IUlh+?%)u~=l-?$Way#!C*_x%)Y=dnfljcJv=q*)1YNDFLQL3yq--TIbeTy89va
zz3EQsNx+n0AniJ4s!(Zy={le_NjJc-ZBoldhY?wrw_ZfPkL^|{R%#@EjA<COn_YGY
z&HHy(_{(!mPW3Zf!=^S;!*UBKDf7KSkYJD*7E}uO61+rHurka{@%o)r-Z&vh0}t1<
z86TgZTCV1YGNrc;jtWcz(=drTZ3<b4=W9Ok>3i_wl)rkb|D!HT0Hy$2C>#kvmNzaX
znWlMSk&Oo*XXcKZnHhGNvm~l8hNVE;WgIso4g$hqj8ky2oB~0sg@K?{DT9<$8>1wp
zA{)!gym{K-SeOzI8k}nhy6q+%uSXmUqHc%Z`ngrM?cK$ms}Hk%W{UlL_i_D=H__|&
zIdbGlg23a{sbloJJqG<YQ&SC!<pL+4US#+7gRCuY(r+#^RWXrDVh)zEk9?Z=<~Jdo
zxx7_R>8*ZQ&GO13r_UZ|d36;{i4Xkh2e{>yTUlCO;r#p(LP$z(k+t=8ibb0w4X9L0
zsGR6Snr6gtN|L0cNnVCMe{ueKcjz~PYt1sDg6S&MiZMx+B25$FSg1@h9C)BIbfyS~
zK0!YOVNk7CY4jtuPuDPniCuQVmIQ+#B`qiyG(wnUl6;uLSM<D;Ogk*DuP_|MBx%B-
zx4nqXr3KbbKFyBlS*+=u^ap*s`SXNptBg0woL8C)tpR7wT*Sb_Ct*`(xj+5i$$>i$
z%+eIRLCl2<3wWN#a5&(1ey_*+>I!dv+dFY=7o`;A6BGQ{fBrGfU%1HA$4@@zzs<7>
z>co~wy_6QE#uPhVaT_z0D!SWYWqE_`jd4b%>xBIlLS)E-4MuK|tuhT#g47g>C2V7k
zeruhe+r&koof2yst$a%&1F0m^Fu^ndi9+W5AR&<wl*ka$f;f+oGp#Y~0;Q5cn#9;8
zmujs+G#mnw%*rjvRu?aDVB97%Z0uNY{Y!4<MX$Y+2mb62ICA_EuBJq97_<GRSJ3iS
z@RXwDma*(2hBOc;w8}@Wa^E*eqBTM!h|Iw2Mr0EjOFGoY4zhS|nP}_9Ao)B9rub`r
zkY5;L>)rAB_>;?e;g)y?!y!lRzn9m&>20LL6|$&{>A1P~pOhGufo++D{*X*%7}B6#
zE72Q9ocG39n*ym??7e1^UME4h4f@61JaO^?u5lj6DP&lNNt#4hX~v$b4k9PFA*ANk
z_kTMdI`R-3rx$tng30vEVeWj%E^M!jJ6J;|O^SAkT`Z8SYc^N9j89Eb?1h+?wbf4d
z4+m4zQF*r75|HP~A<Y!aj~u5wUSoEqOd91b5NPsKyp$vZk1W&_r$>lPnLiie+{05P
zDcZ^%0?M@tVG`49_HZf{_Kz2-nN^OhcpU92&UKnZZAoSrNW-FI7~t4QVX_iuh$vt}
zCG0W<FWNoI^#zR%6B>;%Y|Elrsc%gb=buMPfhA26KO{|4ASQ`?E?sJJ=KRI4<x)TY
zr9m1RLkQw64@`wZL>cq<KF-WYnTjzaGg5?UTxmg-7&x*30VFyB$Ht5-{7x66UMAHU
zrXev31<E6J2Adn)I#y$<ROOFPraay*u+Z5ch<rBILaJVm)lQoWPd<$r8DZP5Jp|qm
z%eI-B9AkX)Fo&-`#O%~q4!KtWr=Pw+cf)7X=+as`$1b<egqR=;!Kqb{y*1=RA3@*o
zZg5LzCD2)lH|X=kkw;lwTjoVK-p0n}klnlY@zhgK<2W{s<1pT6U`WZ(^R}Mkl7dqp
zx$FU)WO=i^7FQfbE-fy7t#|VaG-o)}gh|a1OpnD>N<}t%Lyn$4LD6<Nuw#s|sRpP7
zzuP17Q?$y^Ns2HeQ?n(;CL0)Nbdq4Z1*B=yY!B#mJeu7h%~r^85RqmvS_v!z$_0y&
zx<zBO!qlFfTs(CaZ==adbqvRH3B3WmvqwpNkHKh_r((si6BqJcJ7E&|F~Qg1ajel~
z8lq6Ky3!<yVxlPE&;23YUYBot@3-K(C4_{%d-w9AKk_4d$9H`<onG%b{XL#_rcNNi
z0BxGg?cc{38kA3|QDJ&|gksG_hdrbykZBEBnnV6XJ}HGxQCS9tL1nB#e{+Ml+k>KJ
zZMBUhEz%@KX@y}~c@w;mFPV{=WFjLLpmk3B$28G~B2x)EQ&>)cG>#!vxOM^UBt&tH
zGEI8jCR&H=*fmSA{{VhF#1WePSKr9?={Y{|ecwv|sq-AT^_864|59|EQ7b4MtuWIe
zMmU7PLj_&%JETG)ghpp6eyA7@!7&XC+awd3u(QVM{6&OT*b-dR#I!BSwJM{H2Bl(=
zFz|_^5a0J11Oc61pESMf^O}F|7rM08eEg4opBv}4F_MBNCXKVLlC{Keix4LW(;*Fq
zBvF74V=S|ZE)*!&%gj_q3Bn<}u5#JAZ!evdkW<|odFW`9JN8_}u?wIDQQ(udo0PX7
zpx<0!V$Xhd-S$$Fl?_fcwz21i{ix9~79*3-J+@5I9_NkM7nm?l(iv``QaGR$w%<fH
z6&4^&i6JE!xi4v!sT`%CUwFG0>-^Pi2~DEp7XAffr06{KB-K|PMw=07kmWXE)iTnt
zu*xo4YB2M4%914aOGc0;8IU2BLJDw81rA)dpG-iODXNt^<CP|N+9e)Z8nV`jh@${k
zDOQFdd+vM{%fk+pQiaJ&=kYFHWOinf89S!v8dS<vq_h}#5vc&fkYsTLNK7fQO)H;h
zv<-CV6Gs83&Yz(_e17l&pRdy7G=2nm77DpFSRgQl38T#wDibbo5MZPdWeEzE5*Q{~
znh|yf*kffJ*TowQks?Dy5z;gWgOEa{MzLJMYxj_uVsF*tJFXi;J#i7_I9?jl?6r8j
z-C=CdXQtp{hdztl0ZV7k=g9-trBtg>EEVvEF+mWKgk5wd8MPfQ9K8q2A5gA9Hs>I%
z5o`%WAzt$siCx3J^$mbPCkf}zoo4>h1#Y_TCT_m@4%V6-hQlrw&!1*yc7{Lr!#`zm
za*FllCf#0_EK`JGgy;E`D-|-ONw?tQS(ZQJ{2*X$ef?{_oqtUN*l)O$r9s8gjMN0H
z&4h>NyIen2;f@<-DHmP*eg~<bTCPyqF+wH;Srn0}44o>XUWazOi`Q&(eluaEm0+eJ
zRSWC_)Cw?HvnaY%T*tz33pkER6i0YbigFyfr%p0CJIT)1y%x_fG1G+B`7=ZqtjaQH
zjxEs&QjBT^Z_vY2m#0;q`ROYFw@}DAbZI_~{N<!fq?FH?UCo_H1f6y-N7kr}Pks6>
z5<lTvzWG}zlwAr|f!DtJwY>M;@8Sc${_CXKbC<Mv)<K;iCFl&0l-vsCB%-owKj!Q#
zwu}fni|8bz=oS!$u{G$Rx9}h0%F9ou2}-4uY9kC=8>D_eXVXOqm1+@;yq7^31}I6A
z`lO*j5}5ePAxZ@X5mq6iXo4e+e6YeWQ7Xm=Q;edCX_^FaNE`*UTANg=C5p8QBUfL|
z@kdY4X*Ur<P#l|K$MrXJdf_aK?Jg&unCHyXn>eK!ZpG$xZ@HP+3^=#ANTalm#<a=Y
zw8{3?35ZS;1O{Etpk6a66dhs|-r{jgFXYA7UC%AI-OBB^-^Sj3`<R%TqEfA5+ZJ&Y
z5``gA6w>W=dE&_@`0Qst$Ab?&%)-J7Q5@$?qvvePpPxG4>2mBhKge}&ehINql*$fT
zN-QbKk`!r~SguPL2lTuFBGnYlGS#%t#-&Y+dWf)#eDe5ts-s<Qzhf^>&0?i?6X#bK
z*)?Th8WzL;klv*WR1UwGl_QT*f73Rue%;sc;E(?x<K<b7JbHq1`C1y2HNNpX-ooa3
zmxmsDoL4#(dVz_D<jf!>GhGf&*ZBAU`Ulvt?*LO%lNgponkKZGO-`OT#>tZ>IdkqD
z^NY)P-eo_)%h}hzpF7&q8KKr!m`|5&&Qc*5DcV?sWZeNe&C@bEOR$S2D3~~v!j_Ga
zMHt0c(!wxA?%f$jNNJ*#BK0HedJ#;Cg~BMfP#qyieD2&`rqwqw+zE=+5!QzRd*1n0
zo?6`C#HnKx$E)lqPBA7WwZ;TG2{D9XFi5fO5xl^oSgjKJ9<Jje1jJEH8Ycw(0mY(=
zCma^n*2$FOA7SCoEDZ=Fp=1?uOEgqoozViRO-9EY#><cndZ;i$v4yY7GVF2*>DU;S
zz$g`Rn0pdqO9La$$kG&@X6Q6SBB<1AWJ=Rq>);I|-ZWcA&2REV3udb$WQIZC8*;?+
za0VW>y~gb57>(U~@Vf)NexJ195T%ODA7X_=!a<)ioi<tu_6xyu*}~8%Dv1zTN)m^3
zn`;>20V>ztPGT2PNzB5<vs}0TFgIOyJ>BLq58Zn|k38^cUi0eL<tUXjV`^fOVzEjX
z#w@KYQF2S9X_gzRkXQ0AXXw&2rQPW;@Vqav;(xcADHPdjIJg322|7c#??S|N4TqOq
zGlOj!I7OFYt%8sOooNRBK8fdH6$)4d7sIyjOp~WC^*FnpQ8X1hrX{;(r*IvUIO$U<
z)p4v6naYTwfH;b2b~?mEA9rFFX&H=77TGzun`~|ej$I%LJvPs;(k;}u*!5W7=wKA9
zWRXvrCL|)SDn8SFD}-QtY>fZ@U;jU{H08q|{xA<e`UuT-`-%(6my!UaAux@c#dz6O
zB)2uzITKIk<=Ae=2fd8caQA(m#IOy%@%`U~Q8aK04&U{i-^Iu8x{HN{#pmoP^X!8<
z!+=a7GcXg48H5Z5>x{Z4i3lO~v4y~t2EsCPPE(rG)3wUbS?(?&vy3#15SE2iERgyh
zW!I$X4X9Uaq%gn|Bw0e{4d@LgxVSOSLg<oIrdaR#lpQdG4x>hwt0xCc))HJpQK%Hj
z1_4?rq%=SZ`rRf`<WsFqP;{&8Jam}a=r%^CYh)@$*cGn3;||Vz?DrXy3HU8MHNdHl
zplq<q1&mm5VSbIrHbW!^)3as1@sL8rn$FN*5E;~@jc?j)rWR^+iub?t4r+U^p;)O<
ztQ0BN>bRvc1-FQ4n%GW}N-clU&ysQV;cNN&_kR-`8=KsF@4fuaZ~r#;-gh4Z&)*`Y
zzcAHR8S87C9Qnw7=mRdrl1;5-Q#J&C-y`%xEXSf)DAB=UIOx#zG&8e{TsL()t47LA
z*EYD_8Rx=AN*E}TSh1d^oNd;aEVpq?he5ATf8h+KRlvJ=k*KxK#NI<><x%!qE7;Y;
zoISCC?V31+B5NyMHdb0Z8u^?h;(Vs~#IZ%3dJJ}tTCK@?Z;ky2u42!ggY4bApWE+v
zDermT*Q1nTIOuWW<T3u@FFwM@KK_@SJ$HdPPQKR5{beiF5Q`MkC}Ly@rpRBsS_>L=
zo21)CX4zJ+nnootfv=F63AV-%P$;@|wMNR^PS7w6k}#rBDPY<TL@8l&4Wn2fv)w$n
zN;9%J!ZZymk>To?5;3x5aOTf`omOoY83d?>i_C2wp;9hm*d{?LXtlbmuSM)RWP`}9
z*peisR4RjpAP6wDAZ&M$j!Tq4v)jwLRsR^!J)?vW)TYLfLyt6(Tcy>!Z&*vk>_ide
z^-0wh&l8PMnluP<7ZTUTuDckPfl6`^eWDb?wlNJ0KZ!tR2(5EZRw1xsn^am%IhuFv
zuQPqF$-~WvwLmaZDl#!PLX>61!y#edbL!*?%rs-0l!T7SIX|Kk#(7VhP>h=<`z@1e
zYB1ZdkRmT7Cs9O@#Hc7_L<sQvXsdu$im9myy#9d4pL~o~r_IL35{C}%0iAK|*h!Su
ztZ!_vw6spM+2WChALVtgc`XIEK&#cthy4(wNqS`xc79=TtMvc;kN{qK!wr<q%rgiC
zwi$4K0d}|sGi3v9SV(dRzhPQr86<%Rp--t)#w@ycVZgBK5hW2u!DMnOV|rqQ>5&qn
zr4bN{idiKLLJZrc+g!yT_A#VIrBueMRw!@ZjhNWYZ9n`|bT6J|F-oWl7&JSaT=S5-
zW;uWA9DSV+wdo`V!@(paO%+?@xvkHXTga6P1-|8*zJ>RG!#7}AChvOx`}oL*|D5mt
zm;aJ>r}L%PMnKUmP%0I%tUP?m;ymk`BnfenkSeZl8nuQ&&u4jg6=@6Zy8GiaY7O4<
zrne%bVBfxdy!&16;-`P+XRbJne5HC;L0$f#WPXy;85laZYBEd|5~oli>uGe7<_;3R
zhg++oZ3`pS#D1K2iV9GMfwV0m-=knV#3H2Z2-cORV&#2wSsD=!T^82&aHPAPU}lcl
zo2O99pn3KplRIZAl?$YOpL-sEguTNihsVxSt+@y%cX%)i149TrZ-^oDW@yW_F&&r5
z?G2=rkYzEJhVlIea7G&xbj<EOyIARM@@?Ps9%>^c)>byKN+sU#=9jZ@X`M99c>3rW
zJ>Mk3003!$Nkl<Z2AP5OJ;KDGQxvS}h{rB2qepkq(-z_9?nd7CS&GF9(^IodPR=qh
zG0p7kEF+^M7^cDW^eiLw28LmR%w1`wXXkkLd*90&-}rj&yXPK$`hWc__df98R(~G6
z&|Wq%nuRRJUhUAGFvxr#V<53B1su!5F1mRA0qtgoC`gE77={7%$SS+LM;X@dV5OPz
z(%0^%ysgjLT8qVv3$zAnG*TDOi%_W|juO(=Iy#Oi*bd&gW7J>t3dXO01(*K#SD0{*
z;g{y1v5T%zCm6=))>HiXnd4kcQu->R&`q!w1{hK>J2%U}{Qe*2nj2r7)AQ4W!JyC6
z$Dg2BtgwIYAzt#5S8~VgFXvzX;J@YWPv6aN{Q9qR|NRf*h2htF*`Ku%G>M&I$TTmz
zl2iSaP#8KvxaB;vCIyBeNqZhv*#S347O5n~G%aeA<79&YX_{d;1yVo4kS3JNMCkyx
zP^Ht_04Ja_I*J^v6E{<&fGF}%I>AC<8&K?Y$eIJXTJYkjQO2gHa10A=WHeWn>8!Qb
zvwIIgJYb|gP7;MV_|-5_SxOQ{C?SdC7{ebBwlC7_Kff|3&o@cK;NTnIL3GdE1ecb!
zAnX!}#wMj)l(=JcEXN=U0u0kc`ypABqKpjE5D`n157riEF!SRp17$kRGj>7<RGK2B
zWPEyxROS?cieIB(+T0vzJ~7|rsr3-AJ48D+u3N+?7O~6%Hj+^J48w?47UaFxhUAr_
z7I!p?H0+47ol<rrMc2i#Y=nkkzmJh<ik3kdw9&;Hq33br(T6y5@+pedNm6ZaadDm~
z4zX@`SY2Mmu?-5AL9e~R)c82Be&s8%EsN7<&vEwLIWAnhNTJ|RC=_rVhl1m9=ED5f
zdOQD`rY0wO<(+qO>iq8zs3GwnrDhmR)di_ek(R(VP5PUg7-&?cC=?4+#>UB{K^#Tc
zwnM#c;`tt9dv>txwO>zTZU@5)$LKxvAk(Ho2rz92mF3y;T4j_{sg5C3-h){vQs2Id
z>h424a^wiT%>}AsV_cYD;F4wGMhUHcL?Ucb&&M<kZolnSEUsPP^x0Eiz-@OdgG2id
z^6lUD?Gy?&0M%-R*_o-l2l#KHVzI<Xtwz0G$1S=@AyBCziXuGE!}C4-(7zI}gGOnC
zPCLM|H_%fVfAD9&#mu%@ZoctWT(`)3-}7F6=XZaX)wSoQ1plmpy32HQrBX7Tq75iD
z>Ubih&?q9p4yNOf6&xI`iMl-o!vJftfmRT9`v};wLeZE~U|J4Q5MtR5rb_T5MG^~w
zEI@^Vh4ltc#yc5*<&CVZ_PGBu$LR+lTET^LZ5%h_`rG$1@$x%ZJG+GFXb$PqR7M<h
zk`E|otx4j9%cU=Am|$3tLATSwC|9waI>qW3GuwCIt(~J{!Fs31Z~w+edG#A^pja{q
z!i?>ECb{9}tLS!y9DVvCEl)G(54m7OoK_jVfk}9=&)})YNa7Tmj2GX23*Y<y`~Z9R
zUrni8#xx9sfG`U9A3y%1blP1G9=w{D-0?~d95{q+=NFuiCgth~uX)|;xZ%bCT|uJ0
z8~Lr@`W=4yXMc|Mjm;P2Fy6T&Vy9i^@iQIX`^p*2G{PtrDORf}t*~n)%AQA|+oNj>
zE?kPqR$JU$e~kS-n`7~{+;`X0Jo4ZI>njQMF_#;zy_WEEcM(QDgT9ApSk&DzX}^tC
zsS%w&Np$tiY`@`V9{rVH;83wg+TCEV@+5sd!X@9PySc)NC?Qf?^l?*?<x?`u?A}Wl
zBwRdqjzY13?b=i-HO`$o&4)hpN9^6Vk5|3&4cvO`OKD7u@t*g+pEtkhO?>t<ck?qp
z`@gyOfd`4>l)q0wy1Vxv+2%Sr^f08r6gfVoIyOda=Pn|D$e^`OeY}b^MBc+vFM;Lc
zOh>J;YZZF^0a=t{6^)$Tg+|&IhHasQqA|6NerpYHW0S(@7$c)&_`@C|&9H1xS<GM<
zL7Fj9v<RFk3zlXo%9yYt&^DnLFf}uVQ2>z`D5Xi0e2QHf25FKYHAH?u8bw%X%Ifka
z^UI4{pEfV!2n0gtEz35I(!3zb>0hoig6g#|Vyr&G!@v4Vm_r|_a?7szXobVq-9)Ws
z5l1~1E-tg@svSfI2s5ToEkUV}gKHCo$~5KCQHJRzSsEi$K7f?P30A?yGOQdFJsd*W
z#Z*wVY{o|#ltQ0xxoLzmt36hG5kZ);zS(CX7MK$?7S}g%l!4jra!@OzYjErK8rK#T
zBm*4Fq*5KFF*Zr5R6%CBJzo?Iacl!iY5et5WaUXL5=QHzoV~b$TMXE~;}Es-C<{yH
z_{`m(VY45QD8*w>9zi_)^wuC<?m*-RLGJ$>#awYFxU!bduWRRxcfN+nsac-3@)nDV
znJ`rki9(m!m;|lxhJADp62>vL#u)Czc%B!NI`5m*kc1I-n&M7Ra?@MiPHB7^qE|5g
z*#mSw@<B$D2&Ey8BW%N_(iqFVqtlqoG%*SlN^{$Zh9O5Dx|e#nfGtc;U6{ve)akCS
z6Dp0RQ+$ykW66_GKZYOluM9%{l?3HNfp@&)9qie+_sZWUam3x9`Xs~Q@Jp_XZCQ+u
zjxjko$>``PrE&?=FhPOu`}F%gHk+H!X(FX048tuZp+ZQ5R%?K5TPUUZjo<tL+h?}3
zZDt4ickSVouXs7X`v-scoIFOpte`Ga87fOK(ikI@lpP1J)1o3HOhe#X62q}kfrp7k
zt+lYn8ze%ak_3S!Q;H}|v1|hhHYN%yQ(zSsrWq>miMlg<ZgU2Iu0(HRo^EH8SH1lP
z3a+5njoEC4OzxUxdZJ1=^jVivTzcvjoQ*#FoE0jiBB@TcxIk(PtpZAz2x(wM9_uF`
zrP`Q)YKE{~>eJh3UwE8iHl$KDxHLb{AO22<8*kdn-u*i{eR`QQryk<WiDhK6$^D;x
zinX-?PlpL>NrtTwOhQbVr+aJDGyIG1`8V8f?JXDu8>B#JjnWFGG-uD8r8j7DY4HS~
zyZ2)pI(Qv#dGotCeE3Ez$3!48tODa>V|@2N{SFRYb%-DQH~*HiXU~7-yZ=1XVwx~8
zGvXnfURvRX-P=G50zaToDpIJA;M8gaX^I&IG^!~Z9S;=^*x}qqw!BDFznooT+h`PH
zMjI81>#K~KXNc2;C=A)TeTI^4<83TsxFxdri^SbcN|Tf5$~4{nDtoqXV{_AFVZDWE
zn=E-Dv4G1{$>9~77tN>cxraxecofH$6pJ<m*Tr>Rd_SPy?{en!NgjUqUJf3*nzy~>
zecbVqS5v9hdE=Yk%I&w^&L8~YAMz7F^)p;rSa{wc4W3CFFt5IzohXj~#fPwx3`<BX
zx4`c0`!Hg`=He2hF-ggxRIlP~Hd#A;mhwmgr&iBBL4|=?D6+BCq&QOBa%RcfJIs<x
z>J88alqaT0yM4lT3#UGcTdv|1%Vb%CwiJ$QBS$89>QbALeIvA;I-gxQ#p|a_R2)s5
z!o=7Z8~yf`a;qjkFbQE0CLu~CBtZ_v*QU!-JEqy{P%sVZl?pre?BbehuccZpQgmGu
znu`kyoH%}xi}Q;N2E!M0Ys5220G91A_L4i<e(p3EKJk|rslX6$X#Xq)m5>fXmM)w`
zM=^uV4kObOB!!6B_mNWJRB9Nmg;p9_EMiwm1kDbHR=M|TnqinWstj5SvLql_Z&Rp`
zfi#dR59}?Sv1im_d)2}j9V3V`ytxC6-TnqX^sx`K&9(48`4<RnFjsRZWdV+z6C9*8
zDHICWwu5P!xdV!kV{W8rVMv2uV-0J2gy)BR`jI2dw<BbCo%vQD!_cHjM4E)eTfNU`
zFD-tF)%#*?{9jv*)au;wq8r%ktsz~T30pHgmgB~fFk_@rCu(;olq-}Q<CwJ)R<%T?
z6cUM$HmQo}4?-4}H(6;n$*Nt#Py899bGx|ax|^sUcoB<F-a~EuERMENX-dJhk&elw
zrFB-LA)~tvv-6tUKo;3pJkQxvN4e!?cjkWDOUoo-LfG*T2ttuNk%UU&wb!2Ua|xlD
z7@gp)Z+{!oG=RJhZe?YK`|i7!B>CbkF#;$TOH|8c#v9{IOiWO(*D+0#BuVi70MoSc
z-bt0x?RT$~WR=or1z93!w}%*p<n)>2{Mm>8fbaO$e@3xV;q|Y79e?s?AKD7~zcM|$
zo;sC+1f>&17U8Ib%=5s6qM;#32$L9*DN-b+<zU*9DD}yL7`yD?wLP*p#<VPiC5fXD
z%d(J;i)$2doPfwMsA-$yy>XgF7q>bjNhFg~b2M9RrguafxNZ;E6lMsc0%wmcF*iLy
zt?cr|AVBtaFxp+FR5lS>k!I;uy;=~*DZUoiS~AR1`cEE3-1KTxmLD@mCMH=363V?r
zZn^PBBCT<p5+_ca<jna6bY?Q_4KYQ8+da)wm(FlLQuN6uz$D~rWJ=I*$17ifY1<rq
z@(I?~mswt0WMzGsey>NX)#Bj?9|kfC1&4CYWn*)lCyzYF%{Se_d*1!c?Ao~>DFsUA
zC&*XaejC5`e}0kg{r(@|si%)`flyu;VB@~ZXsLv%uxNGy7F%oVn%#+8En`@D_7bIW
zD>afhVn-07dYXl$bq-y%hrRVnjBoxnHLpr989^<?H0))jXJa<o9;1yalasTgafpBE
z0;S10)Np{mvIw(#nBKpK^#>N%x2M8W8=Dk%Y-97tDO8f7)iW4_18I`d>J0McSqX*|
zT+ZYxAz0rSaOqN$ix*co^5m1;aN~>lhWCF5H{bLUs-q)(%Qt-kx7_q%e)PZmH$L%~
zpCo&p8hU@e7CRmGeEWCPTv=k}k;k#6q*k>NNsGW&_|0`j8e>GwHbJ{Z?D?2lGYleh
zrm@N;q-9_^1s2*7lRI)$ky04~fnnL`LNO1((hRd)p=eony&i>;2KBKCLT^CHErDTi
zZg~wAE9^w^s&~DaGbgt5^n-V?UBrys@*4V%#0mpkCr6>`EWuC_S-~RS+@#l21ir+r
zmFf9CHkW(6_LZ;VEpK}(x7~3Eb33+EES7++?26E+G-WXGICbhIfBGkX%pd*HpRlyL
zwiVp{{l6FgcGAddk?pU2J<SXAbdH~)V4F;h0(KX#*CXsS8J(I&4Lv%S^4_r|@DW;5
z92>_S8@qzN!E)$yhD??!S0)R!Au*&w>WA3HI@&WBv^FR=8o0FzK^h{JCW(CFG^Q4p
zC=?20t7p0Nh)v0{7_2NaJ62?_<Y3zlrQ$eQ>?43qH>Ol`a`yq1k)|=WtmFtJKSpJW
zEc7svkf)BF;K`>?k|D?xB*6=BI@)y$JoLx|?3|jWSS^!mY!XX_RDwcLAhkd{E~Uu^
zR<VMR5)zGRnPf@E#f=t^%?~)|3C_<i(@O@_3oTxC_#)+zSuUMD&eY6KtnnQTE*+!1
z)#GZo77v|T<^D?@rVmbY_kEA^-EaL?q-)Y`H}PZ5#OzMOAZ9Qe5_AXXM3ZQZ+B$51
zDSw8m4j*R6u3cBY*IIGk{h#C1xzjnR<BLd2Nx4*_R;yF1)u`2KjE;`tI1Xu=(&=;v
z!+<ahh@ybN4~Wf#EYlzqhLB_`!}mgZHpr1OcYW$(y!4KjbNHGYc=3yGVrqJdOY;j~
z@g6?wK?%sYJ2K92gv^P^!y!)DCQ5x$Z-^}n%u)drh9sefQ!Nw33AX8CId0yF8v0m<
ziD?>yp`X_>EE|iMk|hWRCA`r++;rPjlpCW2&6u@~b-XZOd~6R#jvnXW;ltc;^KSMX
znBn-7=ZI9w%`e-_>A%pdbZqLiA(fJyBPOz~-(4w$Wr8%(Sw__9AOoLFr<j?6Gd@du
zkWsmKgj#JtQk$k^P4Mz-X2`U}7Y^&qCdcpp1Q!=haVk&@wML^55)9iUPICWA(0uZq
z&vEqF5qiBANt%1wDxIg~!!RUCbCYMS`^Xij){2ZadbC>WJoVILyz?F3z#HH2c8aAU
z%65pOA&0L%%rE@!pXHzav+w2jiRT_(__b0BtYk40mP0g@3`3u%&R$??BVo^Em0i20
zaSA5>&?6XT6dadbvyv#1s3fLTuVcC~Ww%M1t|L)7q*kR=Dm5(AAdXXpeh)o=5o>IY
zl~WfOn>|WxVh3}F_p<)rr^ymUb#|QNYnu$(U0jj#*uGe}rP(rteuk7yoG5}|$mYh7
zmE}!NoOqfyzv&%({rkU-iLr6^?%B&P{oK#+zklhM`T1Y`WroA&cR!am>|=~iaPZyl
zr?<9=JikP_C~>L~_Inf^3u#M)&QM`Op-@613H>hK`YKhX$lL<bFp05<Q-!nTuc@@o
zJtQ3)vb<5ABtf_ZO41-n5=zA~LP4q(Q4%p6taH^3x6(C6n4X^Ey4zpQZ=XHM@iWIr
z_w3;Ki6MpZHg0>#LDp6`S($H9a18Dk*Vw%iG`l92<IpP<xu9+K-n5TBjamG-&yz<V
zr`2vTHr^o9imAyN>h)1ft3Z9!;g;KP=S4T(!290!UjEyE|8YL@(T|fnkDA`Un$nOD
zGK`I~|2^+y^JjmaQ8QzDcLOam`kR}KjE^C01F03U=cBd4lr|(GL30ynm^rZCHn9wg
zVG7nDpjI&Q<Ejuy!@zVMRGfpr4O0@dx=<I?>kR_0Pu6Rp$%y;`Mi5YlJqC*_^u~8n
zJ%5Vb+bWEYHps|GqJ$_(*;w~*OLe@^qtTckjzYv`mmsA{vlMAdvd|}t0v>qe5&R(J
zg**|xvbM&r{o$YB*fwP$7)jxVR>q-%WJk5gk(C%F+cY)<_RNklJ2grW#hh;rd3;&1
zq{o=tbqI4zvv}sh)|6<8Qf-RO^%V+53#~GOIKzw*tZE6Z6<#mk&rYv$)4ShJ*@F3F
z=lIKC{{TxD&d~D{v{hhcZU@6*fYypZua{dnYH*}x_$3GtN^6cDKgKWq!Y}Ze*S?lL
zdv_5<As_k3N9c6>U$mV`N{MY**tU)1*to7ssZ^p=D&e{=VVEPy9LK@3EC4Lq#)wQr
zlAsesq7sD6E0n$75MgBKH047d{u6fX-pknN2-jY7`0qZbyDX(Ob~3~cHV|oym1(l5
zf;0p=Nil3Qhl@%<97fn<6$~j+LV{r;OdB0%#Bqp}63em}bi0IP*dn4J6khCdHjd~#
z^e8iho$MV;*ioM3LwBF!?8z>&;Bnt)ALP30H`%_UOdQ0lEr(1s#%M`J*RM157H}+b
zz-bDmgd_?u91|rZBFT_Zgc|gTlK@k=IHMEz9u(ps(fNDmwjz3|Bvl5H;jr1*#d4I9
zbe1_61`Lou&fwEZF%18D8u!wrC6*Q!bMCSPDa4iML6&JUosUFcF6{-rN4w=SGFrs<
z1AgO!zsQ9P7x*XN@?DHiPGM_7JoLEs>cjlxkN*VU{+-{&(#rGG2>Q#>s!q6BYWf3<
zp`UTz*(Rfw<mK1R;aC!{-N6)+o%?r@p$NSpag-p0LDcVKmP(|R3Z2%FVc(}WjOg_O
z!YDy$Nt$dR5R7*(Aog8P*9)mFFX0XPIAfC(U4yPaBz3ARU0lME5<knnO!V_#Jt>sd
zL=mj44CwdH)9d~=^B3p&hHv<GZoKiuRHezk`GFr~_paUizd!OH*=#nSt0jHTwXn2E
znk5wGW;yir@1yhczsUGZl^8+LAL5SGP|~DWtzp^@LV(Ir9KTG^++eV=hBMM2&cGoe
ziBjBxi7?F^m7*kCNR%)TriqeyiAETpvjj5}IJQII>(lJEDV7}O4j)D(B@`*9TVl_3
zH!wecl9l<3cv+p=Xc?zoqcDtU`D>)zh*>M9*6g#QBr8VBr_QXg5n4o#o#zd2ek*Tz
z`+M27dk@u0g_V^>e(b;gdmP8&s;jQ!&R4#cz5A}lvK&mu<(g}*=4XHECm9_b<%7TV
zJI||W#;=z2<yy9BQQvcrxjSEqUSH$bnn^A5C@MjDbd)d*sny3QH{{moCnE|1y!JZ&
z#wv!GhN6opZQ7xMo#gy7DRU&2Gz<)<0DwYcx@F43!s~Xht99H`1<UE8LJy%cf*>GD
zQfh94X+xmrW++bAu_Z`L&{|z#_0lq?U8g#lKZB)|1jCS$TLc8d!4MUtcpEK>Q?qna
z!P!ffUcd<Wzpf~biGUVxZY76}SIjns66`7#c*EWTBke9{F0OF*a)6$>f!DtM-JJjQ
zNBP6M{*;w&2Z=+&uHj7B40;_Ndf>CX@0-7s;b6eh#S2_-n51z)5=T6CZUZqf!>ix;
zR)(!6kNn=B^TfwK$>UEv!MFVIk5Z|P;1)}?T3s+8RvAXQi0b7B;J?kjT)cFNpZFjD
zgWvnz-{z{T4>CPH#l833OB{XipO#W!+ZMKMVc90qlt@z|q#(;Olu}fyReHT1uIpau
ziI+m=|E@EH$gSao*2Ed~yAhVKc<{joc-K4U*|B>sFS_9dKKjv5TxoOra`mi&IzyO9
z@<vHJXd?#8=rklqz?I{ehQzXSR#)8Wq2m;3I7rLJGBkP%yeJG4+b)nqiX@Imq7d70
z2nGQa$HdZ-L`gDiNK%?Tj}cceHD+<+HPc*J>Jp|IiEnf1%sRW~W~dY_&UNNl_6PW)
zO&D7wVM3Bb#8HS&6v~#UD8wm`<iNo=#1(?X>tmz|sZpTd7D-dVz>66zI}{5BBr#ez
zw9_)@+M6tPdUOXKEd^3)be0!EKVMhA)0#BPZOyj+;>vh6SN^>G69Sb%7HHO2J-lH;
z5(^%F<bHx6<h}3zCJyX7ge45JH0HKjZ{<Jy@DK9?|M~~;gYX4$cI+1{s)u&bJ9&xu
z^$Zo-Ow?d72q>0HjLpp=O%t6dymk|z1(kXoDFtf_i#&C7f%BUgB{O5TZeiLmQdAUb
z4yI|~IAuyj3t^>@4CxFN_PI;U4Et!egjFzEi+onS1V75aD&l2>t^NJVS6FKp4iiqF
zTEzG7rqycm#y7r|x4q?^l*%Q({=M&`UK{26|JA=@ZR7a`B)zPmH_U?;)1-0uI(EF_
zHQ@Dm>_W(-QDu6nO4!}RbPLEWR;AVw)3hm%PvW&Vh+9oIL-LGUs&eZg(;`b_B2Uon
zyKHz4N@bMngsHMm!BJR_llNK~k|c}i_nI`uCK;KWWA@OkIF$lu&E$attlN^QscG7+
z0<U`YO@yJ(-h(^1;ihZ()L%Zt%BhDqzTV+;kw?=L#4AIz(%ku)*YZ8z^@CIzBedEZ
zoI87pd+zxRk394cg@VPg<4^LbPkofzZ+`{vdDl0yefv%{f^xCQkNoft(dl&gvk!k5
z{Q_^a6Xf<;Qky7SGJWkeRKvZ<rFpW=0gXM^p|y{ZL>L7JDYph<REi}fBgQO)_6Ebv
zb)1ngLJVXUB6SggiD8*+nL#2f1BsLK>!Ji<n3Nk0hP@t+IZ9=u0bOtkMJ%UC(Q$d=
z)F$;2O><&9_qAKRq7+e4WdbjxSag`$F-ddSW~4So9EN0Bn#1n1lyK-%ux%pGgH+KU
z`ZT-!=W#dOS67LKc9JI_HU~aW9KA@@1eHOtIL`a7y^%lt#2@kL2S1BaCM7w^sJnxT
zRijo?oH%-vZM%1}ecN{C&z@v>`WThzGHII7Z1?%hxi&lQyoQm+IHBhgCz{D}kuHKu
z7ccONm%kLtv53<I%aqt|fwWj8>kiTK^WUR?vofvN=(K6|x*R)s5&^_p!<1iaPxHa<
z-0C#VGBTBsqzPdhVp=&1_pgRX)z-OX3od;*+br@Vj@EfwLEvY2Rzgd(dFrX7OwVoO
z(4nibZJY2bq#``K@@obv6A(p+UK2y47=Y;rL^>ey5)=wU2*OSur|43ioX8uaNg;_z
z;~16Z{cVYcl)OL14?|30Q!H8_C0bgjz+!B7mGw()E_O!AYE52pO^E}u0{g&K-20hx
zRNYa6q>p7ul)1!(OC=0bkYye^QzS{uus=YloI&I|1*-LNWR?<peN2@hlNgnx#6n>h
zCYhFW1`)H326d_NRgo3X;pz2ds7^4Z6>B{Yt@6#zdDMUXqWHhR)6e+b|9v6-)gvQQ
zFdQVzUtG(}-O|MSfXDmZ`%T<%{f!_LB7=9o^Id%U?z{Mt5B)jm3*aTv3w-2FH!_k*
zmd~DK%oNz9oS$E%RxcrY0<=OH5*v+ODIqMIZm)+Zm5^?qsj-wDvsJc@Pa;%;VSr^6
z$g&J?(8CXg1e-&~HdmQCw3E4!DruHd92tjdgB3rZv*8m;18qox=XhN;nk0rx7dII6
zA7FE{#ijXq{>e9gE0Yt`y!nl90+sPS-}k+=I?t_z%yXe;t4X`lX0%)eTQYOQ_0$GE
zjKvEy>=6pCLPmYGAcv2WmoPFkDoL=56@p$16O9xJg@T2&OyW3VW245Co4e3E4zQjo
zMn+r+0-jzz&2`oDY;S~U86c%3O>%6JX_{E2Dq1OeofhS>?Tk*$kr81Ka`ECiAN;_F
zx$ee2AS|9fa-L{&iQc(SbFAyLk!CnsV07Cpux-z7e(9Hgj+3X4vAD88yWOJG9TEgl
z?(=LL)JHa0URmPlV^8v(-}ZgncKgdPUCoGT@gM&E4|DYB)11F};cL88UuB8Bmsc8c
ztJ%HExMkt)+JiGUfyt2G!Y1W%G0&u^+(AJ~1EmwR5R^tn@Y+q{R*RKELK?-8X@oQ}
z3I((wQCj9NLn+Xvfy5w76SPh#mTM$gN~u&qXFBg`HBA<lmod@=<wxB3@|!t3e}Sh?
zJj{#6HGMB+a{Er2!!D+6=S}QshLi?|Wf8|QWEp;cNH!Xi^j6tiIz#Ut9Y(GwRhsrz
zPk(o?!B7AC|Ge^ED+rfrVUmgZUgpNPVHQGO{)&UV@=Y&c>FjBaeC(rKQv}FJqJT%w
zZ6LM6=y&<Wpa0*K9FyyJ)|j#-Yq~(qwb-?54^m2oy$)#_lVpmGc83J;vu8IyoGG;h
z7Qd`7{?nOKd2eNw5%>Ya;gCwDLa9{BGjW;9aY%-NWmyPAVwfg|Vc-@DBxynvM`W4K
zdk{5f1)dk<7(PdiJkFi3c`dWEbBv6Pu)4bXEI!Q_>e-cF2OTIDiHCi%UK3dtD4`Kb
zp|<GQreR<@HiaxVVh@rC(=aiFB+asX5>+cSiXe#?4*IAx!IBnP7%?(FO{_B<BVzkV
znc`d>L#{C~Uges@JMf3=?7QYFM&_ot>82YnY=caNoH%}(4}a_k1T&17238UyjT|i6
z?|EcOQ7;r39iPUk)=1-!XwbuT9aI>il8DrZ(L#W5==KJLVMuXgiq*Eo-6vPDwojs^
z#meF;ajNtFpRZzLyHc7k$hdf6g*aA3VZyKf#;>5i`CGaE+8fYXQ7V=A{_p>O?*8oO
zXtuvJJn{LiEXz2vvcl`%{yNrwWfR+9r{X3=K5Vqt8Lx~Yq8Js&)MuuUu1lJx)J8`r
zRVuvLc9<XA%bwT1li=J@%oC4N%mP#zlZ636FrZQzVXRii8Mdjt?Hi~~OmMc@#I2SQ
z(+5~Tag2f2-O4N^Xt8Db{(s&3Qc7;V@kJzQ!s*ke@q*xQ{czD}(oEB8_IdpAr|9*%
ztgbHcPru`zb8z2Py!kC}VQFQFfBzqTlqmknC!(H9<@EO$QI$xgaNG*DOi|sri=yy}
z+w<5`=F@~)lVv%}DpM-Y)<+?Bp+GV95P(fSaI(76;0bjT-i|qzmN)5i2iR*i*WI$6
zv4gMS%<=88`ElkZ4M1J7_{p@uaLOdT4ZP+mde5~u#S&qnsH#4%dim8jr6Ox<n}i8i
zqRqzQ13cPn(Nj4uD;0tu&0y&9i@*AR5IO^)(V56GC@Mc_0fb@3aM-8Y3CL2(2Y>Td
z7z~HJ=GAWiDcQeoAOGZ=zLEd*Uw-TbanSl3we`^qDT$K|Qz`5yBkMOY3j(qbM=NZ@
zLRbc5xrLtEdJpI{&98-G8Nb)ett7LAN~M6bOfYk<VHj$>p~X<Ss4SuEMwD#}Q|GpG
zDvB_bpx_kn0*@f@=?|M+bNfqJN@~pR*v?D8?!ElQkN*cAe(-K)U-=FewS^mZsOcCp
z9-@OjA{=50i$TyOOA`_gK!Vla$k|nTgW=~@S^iO<ug_nTP!Cz~pJX*U$=U7*WDr!Y
zVGyh^JbRiQ-9>gy7D=-N-}jiSIc(cD#p62Skp~~7QYf&k@1shR7vFvZE3S>#_oz?A
ztS&9Ey4J)CqAd!6AmOW<J5@@PrU`-X;rT;6Z+NAT4h^O>iJ}PK_woIlQ3nvmG0|nq
zG!2N{>sM=?U+YQ}#tBa!f0{wBOSM>J$F^;(t*(7V*VwZP>ei1xiBqo<2SZ9xfTVbZ
zoNSu;O>1eL_dpa25JZSDMY<-!mY5}%nvthhLvM%`n3zJ6WEwvT$znz8;u41rAH+4+
zsYe?W#6kKS8wA5OZn$+9hpxK~EfsN+aPiVvPMo^P;-yuZ%^o8YJJ16Mdqq<+7SI|t
zH(LZjgl$_)j!#mm)Df0N90hpoc7BSC!aOLALy|N>NQ=NzC}Cn0syw~2h#D(Xo2c{n
z-H)IWMc&}^)$Op*<d}sh)GRKpZ4s+uKJcr*#CLz^_j1+2!zh(;{q@)L*0;QcU;p55
zz91eq_dN6<LdVSBat+~q9{nVxWTo_5LrO~07$2uJvJJCVMrnmpaP#aJ5g`)Uyo6PC
znSS>_Az6MkYk&Hy6gMs+Er(LE&PZbntu<*F;D$Y_hu^~dZ~Ow2Q{xzW53zLY6oX8Y
zrWsmj8l#h78?3A@bNTThfRSpI|MFk|Kd!ssIzD~(r}?k{?Z5EUv8VsWCroZHBey(l
z*|L_=1b)PsGZzWMkbb|<_k73qas7=q@Ezaw?HqmTX@2W>fA4uU-5f;`i%Sb^o0`f?
zR!9s5$*>P4nRC9P1b=M<t6D+ZHsR1iB`H~!VVV|MlwnyWT3QT(2obq_W^I<}K!w3z
zo|nFEKb9@Qt|EkksWQ%r3gOx=>a|mpi$)Ik%rr@=kcARauZ?cD$g%`sTJ!^h!s&a6
zI-e#rTq>A!oiP^IRyn!3N?)tIyr(ts=h57Q;Pd-pzi{0(`K>=3B%D9Dj3I53<aZGW
zUj2$!Bc;Jx-ts1X`saR@cBlJQ-;b|_(kx|hd67Lkc5ES|5;BqE)GMTnVOI^(RtN&3
z-VnM2%&`%)Fi3(Bo#w1kq$G_3EW3c4<S3|$V<V)*?<<xz$2jfnA|0P1waWOhN5u<q
z7SD5dWQD2v0LO8*GHsGLi&$G<q*|>qd+1smcLujs!Y)?WcGYz}cK_Y1E)00+L`;2r
zg4ew6dSZV-u-RpDT47eYc=z8$s|`|Exb-rt{WiyrtTH}2!nWyY8WZDGYc;f1G&eWc
zXl`<8X=Td_@E<0(@R^ikAuTV@I{o=4SNX(eKSR|vxpAVxiJ3ZAjoOTv2HR$;EVcVI
zH&*f?Nq@*@r^nk5?58*}!k*m+X!rZ{yImHSFA?;EoB_DC#D5nnog^u~@8Nj^I^7QC
zas|h6LFYqsVHA-h@mA(8+d9nVC0LFA!ag|xnFc?MSXx=-(!v6}ckZQJC}J4~smk)C
z(U+lT71W)(`!2?&XR#|KT+2jbV(0|Ju(o_JGZb>m=T%}B3i*IUni2E{Bw2zG0@E<4
zmaD`;M3VK;nSnQma2(0zP_iu*OjI)Ff;Bd~1FqV4kjEZ6!-IF7ruY^I!;(Dp<T-x-
zckag0(-fTwwo~Hpb<>PaR9HU$$27(jUKrABb}$f3j@20-A1Bocg}{$PM7M?Oy2QZ%
zkqQz*vM9l^6rqyXW`QiRX=Reht9J0%y^rSBW`=>0{#~Fa@_0Sd3L?SUY74CuN~ipv
zU;Rb?pMUZ1n4F!$u^hhn8@`c0{*ymtFdY89f1E6@t@68n{t+zGq=ICk1J^Y&_Quf8
zES#{z_DC>`#&rvLQN+<@k0&=3z0BnNg{K%l{m*&RxBqjlz3xV;Z~106|Kt~G4BMoo
z63S4VxU^1MuQIW^&RD7uLUQEc`?>1HFC&x&79g?&AB7con8L*{L_VTUZj&@KIm!P0
z`xtGE@~(HklXGWJa{T!5e7~MSuIo^(Rxk~NBu<H<h$xOpvy?c>xO8z1on`#=&;K~z
z|2;pzi*LM{fB)}(hzB2fh_mOPKczgaHJh#GR&c6OQX|VQww%InQ;3@RRNG)k+-hT%
z%U5i^v{uAvhBQqIj)Pm)1WMEM8u;61D2)|xvpF&nkRb;SZl~y0ID2fFod+g3e{lz^
z8%NkNW0FKEA_J{6WU)lHG>;5?(j-D_NMb?34A@hSKn1Kv1y(i!F0OB|5nnd_{_+oI
zN<-kMoH;$8Kl@b5c)h_jhp%Vnj-Bk^yN`z+ee}6n_UDzf)(i$cRF-Z5c0-Ib!qOVc
zvB{z#1`1KIu^o%J*CSfn#H!awJRhYLNtPkC#*!vzjcpo4nIuXQR4h5RGRxA~i;2f-
zG}pVdyK8`Ca=OCaTVBfp$4_&^<|FJH@1upvhaCc+B#tRICMi$MF@ES4%tC$uAKSH)
z(Q1i)(Pi8I32wOcI?9bQ8!IUfpE*wyB)s)76xIf;_AJJyiY!Qn6=m}Fcf5@qhi_!h
z{(VeO&rz$@kW!GuF@E52;>2-2`jL<Fryu$-7cb899Qw1rU=$?@n@Pf7UJUrliyPR6
zVBEEtF4|N|4%#i?6dY6<po<Q6t%R7{MyXQ8DV4c+_ALEY8=(a*<aTR9{8e=KRg_YS
zD30(0kHMfzyIsUIEuyf5Aq7d25XB)$5}~pbm8Dnu@RZh94)B)`=UPLO<<-94(C7F6
z;E$M_*-oP|%5Asa%BeGFSl`_EvVPZR71Sjc7C3#^U$FP}uOqSYK?gySbA}8MjzR1N
zlnoc96@G6(Z@opg)glfegvt<xfm3oRRI2!yqTTgzRPG{SCMoSWquuM%C`>R~IL*rO
zDmz~NRytR0<Mi?K?A&*bD3W~gFV0~{doc}%EHl}?cZ}UT<~aMIyJ)CKsMkv@Tv{W`
z62=-8re|icO_O%7gQEn50d^W8GR1H>z#-q5ER89;0&Rh9I<$h2#?D#JUtD0l)uC>f
z43xtADuzy%UzV4@hJ1Q5%M_a%9W2Ynz~I+@>jQk>_x?-FiovzlUdvTi9pd3fUN|q2
zBu!}p?G#R}fFXQ}3jrmu&Mgx)Uc0wJ<oVpcX>j0W@1R?4^4Q78x!CrZ2w~5$GoUnA
zUw<R|@XKg@<_{RDk1!bY_=|HrcHO?42S4%=#$Wn+w(r`@Ll54=9b@BE>Z1(iSCHhk
z0lq&V^!tD9chhJpr81Rj70CA|N#eXA_8C5lX&6k5O)@n-O}Sh`XF6vUwp+A1Ey5@y
z%M{B?O@wihpZ$fO;@|$OALgp7ujTvx#lPSO{`LR!yh3HHtS;vdtGtkm3I-I&P{RQL
ziOO)wC8Q}yq6E9>5C;R2B*K(=K&H(cVPtE;hMGoKEk;rUooVj+$UM@Gx$7e>UjB+h
z#DQXcaX?20ctcGX`h<P}TH_Q-n5IqAZc&nuB?)ny;QJ|oshAs|;zX;!ll~B2!qAUM
zzrqQ#TpFby3=`((m#`g+-}(LD;@|!2?Tn5#*u7`>3pS_&$kG_4QV<%UGMp?SOCk&+
zqzZ`w50PcaQV{|l$AmPBv0NL!(?JT2#30U6Y}=t=IN)T2Y07GAj(fsM9AkhEdfafw
zPO7y@Ivudeqf`nemU|UP9)LpYE@mbTGL;YnJ|K{$g*7@!GVCxI#8k((QK^q1jRYwL
z^OrW+b)ZMBX0c=UEQbys;I5B9NUL=Wf3wR|W<oRc$evgvPzEEHI=sSh$WRRWL&nA$
zbO&9w&Fx@na)y__^iFQS{bhXjKl>hj<(Ge%U;e<a(eCv4$Bf<ItT@vw4<nX`5imdq
za4Zws5I9-Nf%!HsiUqs2{NvllkK#r{JgsoiSQ5hL<Uan=(i+k<L&|)p!t;mpdL7av
z#<6Xb%KPf#Bqj<&0^cJFeNvTec_QZn5?A<(TIWUxO4I8H-1p!^sPsYNIAL!490&IA
zr_=2ae+ek^GwE3cby8(4KYkRaQe|%=<>7~pGCfzPF+N7N-p20@s9KU<8q;mH@rFL8
zv~e8=x9XskB2yWG7tm{U(56AXTp)=;CMspt)-wiaLNW2tBB4|W*%lt<(x)W5Z+bZy
z6P$f2=G@69iC1EzB5}($H{801ik8fO<g=8U_wb?vRlKN85cH{5i_FZ-QY_U-qL5(_
zFrp>W^RbhN&PJ1<G(s&+NK{6aMz{sg1p~`;xX|qqRvKJ7b&;Yp$jBHdMfz2|SMyA#
zxDwC-2yzFCB+;y|HgR2xM~^(jr$6;6-uQ+$;}%_B^{Q9#$fJ)vuS#{#r7{hJOfw+k
z&sW-fbhV8MluM%+omD>dnUB)*L#oCEsj-8HK6QrkXHQ|+HaoX%=b|qdwepFrK*23<
zekbQ1cz_>&-@6$(d=<5A+wr|2FS_|=E<Ak<Q<7(bo~Z<<q1oxOaA}@{rK><_+TAu<
zJoDPyrbWG4XZ!4SCZ{Gbq`_b?WMN?eW8nhL_9njXlO!1{OHHIW$<O}MPw^jr=)dvK
zx4wf9{?>2u*yB%rl~?$0ulY+C$+8%vzyyR!iM<{LM-c}ej$vUG9sFS*t6U}%0%;Y{
zNkSC)2mx^rAX5)VOwbDr)|Q$C#2mVIjytZdvo2!RRx-jcL>MWHs|y%902SkVKHku$
zTy&Y(z89UP#N8IPkqT)X5%@#m2>c)=HS0XqYLnDT%s+nO8CinQMwjo)zq&s)VUV)2
zw93=vBRu-Z!@TA-uV-Rn;wxL^=ed$N-XbvxNMlOFRRl32lc=&w<`1z<3+Xy!X^K@W
z=EDr8i7-sU(8CaH;Y5a{(xB>sjteXn4zc&<{Zva0n(J)_UQDr`vTgquJ9p2qe7-{(
zxtw!Gx!4+EtkKQ;w1d1f7ik02t<hP%giaIG!JDy6iy#as8VP%Lk8|P7B1;QPSWb!g
z^Q+KZ=BY~$aH=z4DM_*05@XS$SSs;t-}F!UrtkO;EVqEx8U22b|M|au45boox#gw2
z`Hk;ja(a#(JNNJ-Kk_5oa?8#9%OCu=%wM|nwce(`H_{ryIOX!}=TAIzhU$r*VE2yg
z?B2eOvGyW{r5P)iNpwmz$p{los=xZ7vjP&84~UqSnYWH?nFgdunqzdLh<?9M9LJ<-
zp20TFeBwLH)L%>1Y<*$?48vliK8g@BpFK!Y&YU^(WrOIS-SjIKk*_#+&!fz|d>6A*
zW1K&?%!5xXGhTMMddDb!8etil@kWE~yLVzq3zaFtz|S)o<X(_T7}M)@=?(iBmdVB-
z<&h0ZA~mMUOJ86yS!*F$_j3NzO-hXe%-wL9eLHvJ1tErjQ%@{1cyf_J^9ZB<qa4_q
zAO(2C0p+s6%;Y5XN`oX3gkhg3igT~W&}Z}f0xL&Puy<}Zagq=vF@YCh35Dw#1WAHl
zt#J1A1)?wo!^)8qD8S&0;%F~VNed9q;-ZrhY}>#xb5FU;gC&=%y*Z0jv%c2CEfx5a
zzxV@gxZy^2?%2c2Uh-1@r~OmUiPgy82c<d(7Kg(&fB4Zq06-v!3CS!$mQCY`0(<ve
zMf1{mdVY*9n?y;1?gyNG<O!y(z82>bpF*EIMS0tH;>hEc+ivC4pZZJm@;c2{4^w4C
z&#-}K=P&TKx4n(myyjKB`qg)G?%cU&3>Yr=(pi>CsZ?UDF-EmorQh$9Whqe<ki;=*
znonz|iC}fL%@aJvul>d^@&n)h@A#%~{05$Q^5|F0FnX@k>vzf06w@>@bcXDnM@K`H
zs$&S5qbq_CA#E(zMWh*-2E(vPl8DI9r_)PL5g{e6Fxj?a46P+&J4#H~TwZ(lAW9oF
z+v|Mtu6yy4O{U`zr#xhRZ4<3DbF*_ewNZk=!-_mKW#Tv`81#@?jBA%@hZanXa_q^c
zF)~H)74*Ui0ghv0n^x{`nWkhaZ`khlLz>M^?!M<PUjEWMDLBq^x8To1X>W*B5-l{+
z>*5TSftVDDW!8~8BTF-qRu`m&QE;(^APFOs$Q_5$a7dgcn9`)z8#3xtNaF%${fy=V
zXL<3i9;4+l_usP$Zj5G$SG+Q#U~ATwe5`7TwWTWE-U5Y!CJGV^AuvlNtV)f+6U&T@
zjFM#unb!0Miiz|0F~vB}=q!mUGfZOa^b(iO-N#eS4ntJV6O}ogVQhSig|#LA(~tfT
z-9d+eH=x~Waq`p&3I&VfC!Xe>d++A!-}mj@`N}t7**33w#Vh!^pZ=eG=lA?`*4Dnf
zGTJ|MX`t61@Z{6SdGhJw0NAEPXs|`zzn>3UNmA4og~LA=5;6y3r<o#%LaMbYhHX%)
zR51D`X_jG_2BlJofj6Y!6bPd5ibu5A8kG3IRjE`6!ib5nF~07tZ=+VL(jD~q;BWmF
zCr+Pw*45=_X%i+$ltv5!md-D7=dJrGmrDeXKF#!~%hXgo2S#RT&Zm-shzwy0%t{gM
zfl8B{^<qJ(QNtankR~baxXFjlU!rb8oQ9C3NXJHN&FqLz**wZZ^BnDuKaQ`8XlWsp
zU^?mG>IJ52n@sH~q0xAeN1;%_F&i{S#>mnXFK96s4oOlIl_s<|`#k;RDPFv17q!|L
zz8~NxA@QJ(Od^T}!A6>}M#jd%I*Dl^3MIlpmndJ2zj2+U5Y#H=t-hX^C`tZS<sQQj
zR4Zl5<uaCOA+#V#VxlM}h(h8dzA`;C7({Gtv=ACT`tc9*o&V&Yv2V{F#>U53URrrJ
z+xmi%M&o5|&UYT*V)r4;vlb7X`!sXa3b!~Ro0S|w{kFqntUdi1e&sWFGhQw*>9^UP
z42YwU>Dg`UKd_gj`9)GRT0is5o7S57#RY!tH$KSk{NC@PGWE=}Ga?TLk~Ah&2}m$4
zlXAJt*w`5DcAHka$-wvWQ=Upte#Y8*!bk7=5I5a;3$J?Roy^Qkv$XX508Ou~tPl-`
zn6*)iI7IlH7%JX!k%=)3N!TA!s8&!yY_aMTsn&RL#9-J*r;w!~cD2EryNY&i;rPNl
zsZ5zTc!2qZCI0jeKgY&m#%N=N7w@~CWc_!+i0E{>%uJ3mIx<2a1ftu)jRX8JrkG|#
zgD$QaQz?#cHcMDoS|Rl#hA6_nX^>^>Aoe%i*-{8bM{A5V#&De?aU9d`w&@T0dFw_7
z-OiBHr;gL^Y~qK(S8Ms77g|5^IMGX9iP4xK+E_ux9yk_)j5rvO=?tr25qAgJ<r2y?
zAycG5NS5S{;NI2*ESbVkCdE>bFfCvh6-t$u^Sy+9<C3p?S%W3u>{*YEl@2qr4SHRV
z`Q<KGsemMkNYj8Mh%sG*;@C7u1JQ0`kByNfx%ac*k13TlnfB(1mma2@7C91CSZsH=
z(CahQIZ0G$b!Bq-?D>oQ<j?#s3<FG=o7|@g(kw5dEiZRiU1_qgaFO#DPV+6_{9Ra<
z&F#0{#{c_6Kg19In}0_XMPI|6df_Ne^9StbucqLcEgYX~rD@9A>L#5|58Jjmed;W>
zZ8JVT$;ikEQ4}#44Cr@z3_LGC@NErEz%vJQLV{^ZMn>!0e#h-hPEGQ}<B##;n{Vd*
z?|Uyl{*yn+VDRkv<eybgmtGk+F<4pe62~b8JLAx7kxUI(yc9DwJw;<;k~GUO4I}qs
z6FH>Pv>kMokN0|mK5JV}2ufvyVZxy?lYTEFj1oM*$4sM!=R=}2<!ZvAN}n+5(D6dT
z1f+s;LsKf-NL?jW38_w~RH~@K0IO7CtuthGwZ&#9!5=DcZRUc2$lGLQyu|Y665Z8J
zrlOEA@=-ybD2lL5%_27Qi)#$S6sjY{Dr2BhGVwQGNJ0q48e{yQU;1SLKKq%^@|U0d
zOU|4>_eImGLI{d(iP4b;W8>qv1s7A8_`c838?w<{M}&ef3{hGkwMDPvQ*<5f{`4n!
z%NyRx*wi$0bF=@TgSyX`(uz2XIDYyp$AM1?`1!|ec2ym&ZWI}@0_@CWLrP?zcojq9
zf{iyEFf}<rsx+3;d@0keOey@}Z+>WNofF#o{T|KcCZml8#ZnR5vai_9n3jbU@=8fB
z%V5w?SfR!5{{C<C-~RhgaQkhy@rQr($IsmYeFf@u+N>XWj9s_Al4v+UMFGk-Aj>ce
z1Jf`m)XJzh#mGT{lo}(XK}Z<-47JH%;Niz9mX|P9UZvYw;^5AMJaYFu1_}3ndWB1;
zT}CJBT)k_W@QKfIaBPuYuT8C9Vs_hhqA0*i!RmHsoxgzTR1jgvu-(EnQ&P+1q93rZ
zyn&Jhx~YeH2H5CLH@=922M%!T*fEZsJn@YEi}iY)eS7w^ZQFL-LXlRp3G;$5it-Gt
zRt$y%);CsJ-&kXFbMx<LH@-I7JbsdsfAPocdCNPHX^eFRsV*^O?gnC*61!4DCmA})
z5T=1uDN;Za4u_;^#?bRnsUnF4+O6Z-L-sW4Y>r7%6>#{LS*ot&#<>}?6oM$=>8Fl!
z`qBfq=@cc^Mv$_;wt-4D_4+7dd#)x8hQu2ySl8V^lEei40p2hHVKX~9!}+ewg|<gx
zN>+y+o&x<0gELBN;si+G3#&7NG*yI)eY|19pBaCI6c*p`4d06E7J1LR-p$89eitA8
z#K*t-8}biSf0xZhT5G}}BK*AP{pMy1Ddm;&>gWG=nRM_);l@JK>9nZVtF${UPM<!*
zz5@r?v3)!H_wD2HC!Sr~&9e*Y0EA?yB*T76v*R&YD=@ol2T|Z7rAZpa#BqwS4g4?y
zo%c`}rhzOJkRqQL%K92h8pLsgKlJduh=VR%=sUFf8IIXxv@wF~jv$DLq6nqHwglt#
z0vd_XDMD*hCc%oaO%uPLvb@$}vmG!9G(#^%C4!<YnQWNs*fBxT%vc|UEM9C;aV0xv
zcMvBD29iuD!bA{D$y#bMSZw19kV=vG9x)odRZ@K>S*FQrUh`UB^M*HI+cs}_%bR)q
z-LL1}?|Bb{;qY(X2*YH2Y@C^yS*9kZC>D#D(!%$BR#sN7Wa+X@k)|m^NP-}t*B@ZX
zWgdRy0p9%9cd}#qb{>A{h1;C{l_=BD3SwG8%wy|=JaumwlwBJqO=&!Jo|oU$r`nii
z%qcK8)gT@OL|XHejwmEaN~_&sX=#y4wTkPyR7+K|tsuy<>>Q$NT11K7D(Ol3{e&lv
z9O2}NW4!!jFXK=D>_g8B3o0)Nc;I(_lWL`cQ81Azf0>(>MJh88`PWgHuzqoYVY`dU
zG7Q@y$P}KRA~ESliek6R_K^~k%}=1xJDF}wbKz)A$!u`l)rzXnWRHD<1IA;N3K@Zy
zvSW4^S_zun4lX8(XXY8Mwb*`RmBC<ux4w*}Bjyo2bzzyrEZ{{UzWzdQkg?GQ|MRDQ
znp<zZg{Pl7%IjYLdRp!FU;DhaX|jE08;1^D#rEynbG%R#<*?DMa<f($%GU(Lgdhx9
zU44E!f;?AJnV@;^1DGY3ZU29JZysdVb>DY>&bjN`-|HRSSQ<bB1i)PjNl}VyDW=B~
zWhzpXDObE`T&|jwQ>9c*xyJI;{4ramCiyE>jx$ch*<@QI#ZHvSmSswmDDDz>fB=ZS
z8@=~?-Tn5v-+Rvd@fsv0E&{|2BHyR#RRPuLhx6{~zUOy;%lCU|iZ~INGQrK~(Y7Q_
zG?tV*I(8AQ?G{lGfB?&}aqS#&B2cM8ryZh^1kHq74*HDr7cuvLoM(?;WMj2M<(6rl
z`q4=q_}i!PdjiL?xcTTEG@m@rcy5X9R+HvtoAL1)(=)SJ<CFC3Yd9)K2}u-%v>O|Q
zeuh?(c2?o>%`Qo$%;Jf&cyWgMnac7zGt?T|t&r2FF5)_Wg=O0O#y|WmO2rbt^V?tK
z`wxDPFuwMt_aWCmSL91x?`yBez-eM2#w~<kYpcNvD{GWWC0;mjf)9T9L(I<3a^i&-
z2J%{0xt?nZ>M}L3@A6cJ9T}{wuTjlSVOb`Qa>10iW`XwBHqwyf3I&{e4$E~A=Aaxa
zG#IHOmoFeECx}CzLNQ0b*T--587l|^Ur?MW;<Y<ig(|sf0axb<gMc`UkUJVjhG~E_
z2s4Y7^%j?x8nnA1Nuogu97{7^5bPQ+Fg=;aE!m`jBGZa=yNxmx$8NuY;;u0oLQ^*L
z^i@h4Ya+MArq`wC`51;p&NlIVh2H6<)0pszqh2CGDyqdIpa0drhwIu1Auyyt6!a0V
z7X>otxym~Zm13Ex@hPTvO;awHQJG?Udz&Ee34%VoUKb%^G#OczfXwK0`ZyLm_V~m6
z(l7rCGt<+r(U3Xhwa`G6WJJlJO!kLQU*s$2{wW9d&$72@Q?tM<xQs?2TZtxn6PHv2
zo*z)Jui@kzOw(d)bPSzkq*;QoQ`Sk6WXFYJu!m^|jqMiS`R@IE{&T;^NVUr5*4Dc|
zQ>TDVy~&wB`w9m?_i4HjBteH_wT$lv7`~6!YU6eK*rttZSqR%Ai6cbcqv)8ll)=en
z!tRL_p;9J`4V>+7AXawcxzm(2sNObJ7TH~GVYwNphH7P;?QX*5rOWjC5x3ta3A%MU
zoi43ZP>dqNZVNAyEOb&Dn_XhlB}@}C0qB>*I@6Hc`zxQ~jypb#X_=fka|&-~g7=k_
ziX~=dW|*FyW^8PXPN#!m7&}eTgB!ZiptGIKO%PIllgjYdJ0=1hr<}Y0A#Cp^qHu%2
z?^7D7VWB8VjYuPW&m&7gY5|!-WhqIPk}6GvB1uwQ$7QXPV)eVA#+lMj5-yH&^vE2?
zpI&6~(mKbVYT@_FRK`jiJ+=$C-Jp^@OradJzOu#0NQr~{_Yn(0WY}bDOB5}WSV)Ls
z`s*t=Qc)_@IN=#&<7Hla<|Ij+zOD?>wV*V#nm*4ze+nT4yC$airBD6}ciedg(=#*7
zFI;}}7k$WkhQS#<5bmzo4q+Ja#E*W+zP+<7EG@FRy-m;SBZM6IDPHB7eoaALpf!G$
zk!VHU67(aF<@zdz_a7pP)19UiD3wc?#XN>%VHhS-3WStM4Jyk>k{B-tiTs!>9Rx~l
zp@<<2b^|>xrc|j=F6W6nk6g8iG!63AGG!%5RZ6oPvb@@3alKBx(It!oIYUr#1eFm<
zt!yz`&QmNGu^pQ<OR2AJ(Cv5}==+qaHhcG0DOE?%#R|3@IANIDr5T#2@d_K8ZM;;Y
zErU1;Q9>{>GKHV^==Z!Id)(K6Ee&S(?c>9D-SKi6O=Z0B!V4rnWq6@bC{V3bs8lLc
zD^;r1Dp@v|{JUZh>^LrX9_$QHkZHm|(d))sx-?Ix(`IaR{93kj$SY79TAeP>9Djjl
z82mU31Qvtrj)@qw6a0*%O4IiO!q8)*zDky**rqiYj!2V%>xD`O&I|g=nlxUR@;CQC
zKoDr=_8-77OxEigZ^>u=Sy4bcQ+O*IocN1xamTHr%r7s|@-mL?sZzBCr9z(3iLt>z
zK^&7s5lZK<iY{@IV1_9yZsG?4DvPjlC2HlEO5p-we3mFv*rQ<El1yq^-Gt?Jk47^_
zC2gihT@D`^p<0q`oNF*XJxzIbFNSFob_2F_p7}<LNJ)fNjE(KWj(u9~*3PF8RPqIW
z{nvjT+ZlK+e(PI*P4p8o0)`=}RLYbqW$c_iaKzD?Fbav1c+fAX28P{I2>dAFYyas#
zQ!17)4C7~>V!PgvhDIv`zZWt8<VimDONZHRzzfe@;Lcn3;^YQHGJU;eP+M=*ul+9$
z#R=}NMN_0$akoIR;vTfPySuwXuu`m8DNb>>;_j}65cK4E-}8Ju`IJfaJ$v@dJ$vt2
zYyGaPc?MpW3YwKRg(CopfC_-Q^Sdjo0T60|J;bk^i6*x40MIVbEusE~>wCYUjKAM@
zt)5&@t>OA%oKH>TSZuOOl(zZ~>w7Z->OG}28Php;{|x>~HC!jc%c|@~g6mMn3*wLB
z(QDW`i@rT#D6RuIO*-6S<S0v2FYk{p-}=39b&V$fk;~2|mDj=y{Y@(#jf()>yM<*=
z!p34_kxZiwQT_gvQt!@>@Bv%+M`~RFdpRM$<?Rq^l0hcADPd`j)1}Lf=I;?#>%Xxx
zhMB^NkzmXXF)Z^Aqr_>PX&c_MdhQVfneI6`YxF22Fj}4nnx@IVL6~W=NaFa_2m{L>
zlSXQn8?)5(GA3;bP(4;TTZowgABejWptN@*qdX={f})o*b1RZl^NCJNSyg+a;@$PD
z6aY%qBQx!e0Wp<6*^8?GMZzK*`=#YT2Q3+*SoHkgv98|^@x}JPWoseNV-Dx%8<~Y5
zR)*F!KRyD@RAH>e92zu1_s4lvq4Q=QNU+1|Qi2p`CVk;l>e&93f1#gP;^}XA5@z%#
z<sXxH+@iG4?7@p_U?5qOz6x9V3AZMcsMz!VYX2fyAW)^0!Mu8TklVopCel25q@;)N
zhZ}|dYJ^>e4VAR7#wwFq3<iMb5Lh#-1jxkA+m32Vxp>u?YSKTP(8u6VWbcn3H|o+c
z>5vb7Ai@(as8bb7{BM?__@j8@;_2+AkKeD7<0ihGT&(-aFFcg7ED0le=t49atal!L
z;V}Nr1T~(z?A-ANI~TxM$b=7K`?~2)59eT>e}5U7e6I_;*+9+Dj(W)-$a(-%9ge7v
zT$iO*n;UwLZ&#83{qnk|NPM}9))B?cEsu7{w(u#EA&j7{(T>%wj=p?CG#<)-SFUU~
zt6bk@WNl%wM~Nrv-^v}fcHEd@-y+rW)!!qYko!=vvR-Y6h&6HIzakGdLe<Fug_K@I
zJ6CeE)DA;Pi5UmqXKOqQbnQv2jx$Y6D$AGIMqnwvhZ&8Q^xtD%jg!>eX=y!W+Dvm>
zzZl@ilC1+l=3w-56knpqNE5q4S^TEs+}AGUnzMQuBrXG8$Y}*6tI6ve^@2U-qf=A3
zB&NP2B#os^LpVeI>m5Jw!PNE3Wnkv!bNnX!QFKt#@lR@lb|boR{wIfRGoHwllw$ND
zOt*YuXNS&n!<qRFt}l~&hP^?*aKgzq-=XaX%7zhx8D=J^8F0xTWaCGpogbM}up&xX
z?+@{McUveT$7&J3cBvp}U^-`EkZCMd383^*wt^Ygb{eNbIk?H8=d%T_G?nSub~FAX
z90sUFWqpcB+vi`JvK-1N%O<1Dhlr=a8E^g-8Dfer`z~MGoLe`^5)jMi@+H-;A@1)n
z_jR(d1=Nt*((V|4i2CVW0=i?HBm-s!77N-iuMDJD+5GX73{p&Y*BiOC0ypluAEP8~
z_<%#J|9^8WDB6<xnsrX!E@t&+7lnetadlgw9lmC@M02B-{&Ob7!6xwKekbxRV_L8!
z{=YsJzY#VWR+WmyGL6k19!&c=hL~N?PHBO?zZQCfgO5z{k3<!!g^Yz`@H?6OEK=Pj
zvw{Z#v~Wpe$UQ4BXRk}mKSxmIp|Q?Ce$INbLzvnuGI%0`X)c8?*1=H-I62eS$>1!R
z0>K?bgR8!T#`E}w7ygN`nVah-gr~YXe<&PmDF=}<Vk4|667n6+;+UxxI@Jew6VTGh
z+NONH*Fhn*2Xrk~%-vT0Rn+=0@;dd#Sw-u0R@}Y6HS<+3mhZ7#f&`8al>|&QhwHqk
z+Iy}%&rrY8wnLR~OQ}VB2Op~=$<s!ocC#E?m0DNnF(g}B1i-Cdy^IopO2#W(f0D*2
z^L8&1*JNsw4!4J>7|}L!M5gZ|rgL<a=&&U0xx2Nj9}m{gayC@(D|Bna7f*2_=sDhx
zxpQ+6Xs{%O(f(YO`R^rwdgjvAcLuULVdE&UB;ZLB79Brj)c&W*Hkmbe#AHbl{i-C+
zQM`1$QEy`jGMA9l^V)ttXZr)+L12ljkfxBWuSmud#@mHH)apmHm}3FMuklFT`-pu$
za*ZBvDiDx5SD|*jM4&e9AD_~ZmXfR=u3r-~pF)FX<Gw0A^SPkXX!D?WsB7x*;t4e^
zrmQtkDxzb*gJ2|?K>>bu1M<yju6e9Fh&-LSW@V^!UtZ6{6~@PEfpFsN4fR#K@uOi%
zH7<Quc@d{JDN{-Z?h-!5s^oB5%cTbPCk6?)g!Cg3KjMmEp(wq#<L`Z%zsbDWipMnq
zODmP;@*`_&?-X}SxAZL(6f+=*4%PX`x~sq_k>E>TgN|R5*$%xP#-Iki`zf~jaR5ox
zS&0_QH1w0M2Wc6SDTM~dCwT7f(w$|G{g-2O-mZy4{5H|B4<$xzFu2{OdU&)fc0Pvl
z>A+D^mw)f}!5=5v0-{J8F8e-S8$<l4>*7uW4L<WQ{4+Qp6OhwrZEWH0zKi8xR+ff2
zf6O-vWJE)Bfvpu)u&Fd~oR<AzQJIyP+e`S@M{`?s{6u@mY0!_09f>{Ze6bT=s1WZn
zEF20V(~_-QEM4@y3%)i-8~zDcY+kalFXA2WFKo+<&E&)zxGZ2YHMWGHAs`0`@9)gg
z4Ow}bD@B7>Nh12I(&{;4IaJV}L?1h;jF|f`!_e}RxrEO?e9eN1fGK9zmQ1=fc46US
zUv!*ri^<B&2*-$isUna4)HD43Sxm=lu^+R=ytpK3c8P`kj%iB#8VwbDQ@W`4@*#_~
zU%2<sZoME=I|F9@_FPB{p;1lg?RrLH`vTv|R*k)bvgkt|OB*XCe^&y)iua=vo6h#%
z|Kc;N<@b;5*TxoSw8oOeR{+<6f6@?HtI8gEQLU`-6is#c`+yA?LL5U>4^;FXel41i
z<1HdZl8orx6%xAgQ}FjPG7a3M>iDe~5pWuOw!t)Q`WSWol-BMPevSXt@VT^DK<r&j
zY9)&q>aAn!KekMR2iMj;cXOHgE@zi78j`tq0-f`hniQVJwu>%$*&DAC9Inf_UI<<f
zx!%OO4zEw&t`;BeUp8S4Q=CE9Gdu}YLZkEZ0Ll088i=}5rcH}JOW6?=fry9`K9DY-
z*CE~{6=^^Y!ZccUjJ%6?$EH|P>oj7X0GxY7+BT2EZV!dNeZ!suZax;Ng5%k-FeFW4
z;<W5WJ+Q}`mM>5Y)~>v)x?)SWHiWUi-L8>joa6fVpOBvlMht8$>v1ghHhl;)PqrW}
zzD$+j(GOko{<$cbq==Z$g`bX)ql9&zkUg$S#PY=zpd{ncMrWFm#08&I!USA|f6m1p
zb5>Ubu_h+D#)fM6R>@<A_Ty9vF|m;^JsWg+MMT)SK5+%)Xvsr%XB|BnQY5nZWX-|X
zgC%8TAq+cjkKJ#qiDIuLLyjV&UCW|=hfdup?SFkk0bR0oG8vX$1#<9Rsb7Qt@D5*r
zB*JUsbN88IlIH;T!>B}xMGlQ3496J8hrng!vIr;x9e*M4lL}Csfyb1%W4z$-Hy+Gf
zQC}#CH4|*_&}lbjYe`z5?Ul`BjmG~(xzM6A$Y#EG;~mC$EYR*15OuCA*dnoXL5FwN
zw*4O{C<#Bj=Rf^iauI*qbfvGPsLTwg_c|oh*?rs$aLHL(3sU(ph&=!Edj+`xlYP7l
zr9CJZ_!Xj$Yfbv%SZL3j=UZcsX+WCcy-Qv;OBODhgZ}}gR&ce3=-e@p<@Ne=y+Q4k
zuh`I3;a1m^*wo%0fW4aCY5PleKva4WGXOoOgZ6WU{asGl3Zo*7S9Vf{Bm@|9?CC$t
zAgdVckQ-K@=GiDk5rIZYkJURYoe;y@5IYcjE8nifQ8c3_*u7NTx;!{bs%pez#4q@>
z{d1LWzq>u|^zCmq{#7jx<h1)QmCt?aBu_Zz@~JG!dPf)w;U=BCr9UVn<*_M$C%Y0)
z6?v?fLvYvp{*f3z?>L#}pycMlqi0VialcLlDV7Ar;^mwy7HbXNK2iE~a@L;Hg-KVZ
zaicA*tXR1_gx7dpL*0h&B5@H|k#~0q@24hj{>=>ud7l8ycg+DY$g*;>dPe%?GnI`$
zfA`Il4ppYHBlW5+c6GEAVM}Zz?LlO#CA{&=aI1F02e1=Az~Zu)3I+A|w<<%E_-Pb0
zCoa~XxNotuwS_ySu#KiFR)3`a^A#*t>e6S@VM(@Css2j^RA(P}HH5o!sX8SK_GefX
zQ+-+Vm|&G7oMUAvo_OV<5{eX&{mvn{OtslVtbiGWG~khsKX1O&9%Y;)Vcp+g;NW{d
zP&&3Eq&ys`Qc@6Eg{l%qN{0BNxe7>M$P@982@gj$Sd=9TfpcgTHhw`_6+nI5a7xn5
zCG_a-Q65b55h)nX2u9pq0Fqkpd6G2;XNv5VNYg_`_AKJ{n;#1<_Pxd7FlZIRS?)Gq
zVDw!)R={y8^xGTaX7RF*X9%<w<Tn3#WSs~f)GKR8u=;a($==nYZR|onfjn^{rJOcD
zo(kUmyjfz2o_?s|2g&F8CD6n~Va7&hV>k$uJVvXTO-{~SI)auAm@*kFIp`7yV#4Wo
zt#8+0j4g^XzpvSRcY+*!$ty%;#f(iX?9&3s=oOK*TA}4Dv>y?lO2#rXg*;#J+2*_o
zpx-&xiYRmj+~Ff~Ooe0H#sex3hlDlA$=bgp)swb#@xywpT4hood0oE-_4&-cTeu$e
z0O)r=mh_Z1^ADG(D61q<SEn%Yu(i7w6C_DDose^Um=|itn>(u*kL!z%lq0Rlke?H4
zGk7<|S=3}z-4HSgY}x1jD5}r^Vt^*UCG-`DinzTAkbbzYYW;2S({N=%EC54-D9Cij
z`z6!o(1eEbIeaQG*l=!+rR-s$^))8FnR@XBAgRu(leO}3VsD-y-8P&h23N9UkGrIa
zNr~o)w?xp&;booDp3KFcwrsCFH-T5*T|PDJxJrZnWOaV=G;|xi*x4|W=HK#)?G~(#
zEAl?oAQxKA%_@(qpHzl@gvP3}ppVCV!Wx}I!KvhtNG8}OqCrxW7YC{dSxcsFz;qpA
zSpRd$-@q$VtWT>$)$3-s!%?>GXsW1CV?aBC59jM#7O(eH!vKZPJDDtTaYc<;uY>rp
ziK4?N*Fz*Yq82Gdn8Blf2+OVC*bxqN4!KJ2GuxI{EKU~Mb)kff9?2*2gtQ=v*0@6H
zr3eLD0k0c7$9ZQ|?fp<I1)z>OX*Qco`(|kM8e$FVPIJLx#Ugdo91#X%6GAVIFEKNo
zGYPKTc8|l>mv^~Lh3qt8z1A0<na75jaByH`Ae8!T_e+iU@Li%Q8s~)Z-bMQb!vsh>
zV=*f)YmW~a5F!3@EAFoYCo_m%Ndi(ta)5KcGRE5WKLmWM{exPrNb#v2hm|oQwbbt$
z;14i5B+95X!8{uvD~s--IQ-XzbkTWu-8(JOA696-e^WVaz29c7_jD?5%0_B2o|-oS
z`|Ch;CbahL3@!8Z{@rWh)aKBI@TG~FZG=`?T--u^EN6{s&>PWs>OUW<&JFT0b>bV&
zbC1|I#6~SP#z|ulxu-IPRVa-Q&A?Ya<r_NM34Dq<rYpWKPnYZo)lud{+ch`09KT}e
zqdPpCK`Rf_RsM*5Txl`W_E1`^A9=ChuMY~!I|q6=U;7y-g!8Q3evHdoA`Jmqg-clo
z>64D~gTA@=L5oxdNCZXGm<G|(2lx`j`-U!Iz=;C66Tz@*cM!_SJBhq$z0PNI>Je_`
zK_(^}AQ=Qayx8^XUa~h7>Ny8wNypf`;gnZjHVGQUK`imAuMH--qq&~3ZMJd5=5v=a
zV#*wI9-C<08ct&lcpzzEWFf*l;3_E7Wz($wQW>vh<JbyrZfX0rO|e0KwAPgZuRY&L
znkp&eNt7t}&G#gJd~e5--)}WbG_LglbQ4y^ROmWVlGStDV2C<bAN?~7sKTRMIj=|p
zMhBB_9YpPhvBXrSjlqimDGUtGaX;+`&k@%1WNNh~_YA^XD#xj;YDESf3+Tt+TE(rp
ztDHZO3%FkU_*+ay08JYH9y%2mo9|>u<psO~?wa{MUsi}4*n|oKpPjAlDfb?#cC#{M
zn)QpdYJPs|4l-51C1O*lJC`)1*9Vsf4njqKck+K-#hL~7Zd4URpZxtcsSN~J--UxZ
zEQINhbX!J>rBFu>4G)M9vU79)v^exZqjAoBvlYwwA|1s?M@KQ3)e3gDf?tp%dPX#v
zf59?VH=5lCY*yIJwaAvxN~UR4fohY!!+?ad(claQXZd0f3S)Ff6bCv>dW_&cZ!~C0
zD6_*dRhCC1XZu6mfw1w5dEWK<*=lQH6(Z{L>y{|63;Lz}2KfvQps^;-dO6~9ojMdB
zde-Dh;V>*u_I%jdd_$;pNcQCDpcy=Us@ul9TF|V~7OZy)shRjd7VoY+wsp-+)$Q2f
zvxyA?W+VZyWHR(1RZJ5MQ=AWZtIDtkNUusGAp?pt1?c-19(!_DRHj3tI;B~#c{UTw
zC0iNe@r8<Q?pN*DgllRZ*Z8+N@dt;PX0q!+%N+vozT<i4G?|paN)omKyfek*8mrg>
zunF<t(pR!StCfu{T@m~aIqMsr!L*}MO(Bcd7+5<|TyI=~J>Abbf$*0e3UTL?zj{Xg
zW!)=7P(=0m&Q6u~^*$&n)u9d>E4(yW_cqZT7~bA<#p;U&w@C#)9T;nT$Q({dVz`H+
zFImm~x>evcFhyke)1}4X=7As)59#$wC$t&a9YwxIKB<4Sh?qr-2=4(UecbAMk~$kn
zo>2Q_LoTwAC&JLr8HE^PnIxttO4Gm7r^P8@C;bcqBj_nC@Gz6%FqWU%JE=2W4ofn*
zNF?im-GL$MB5om->UFF89#gh(vi+Axl9#<rYuKx}H4ZxOM3ST#u}fIh&QpbU+nzQ_
zp>Ur?wFF$E^E{LJ%99#-@Qt}#@L#X6_X+F}Ei}liq35BWBojl(9y3o154fre>Z+k~
z)yoN<UTRTD36nRcamGV>2cI$$<}bIzXS$v|dLCn+2}&~|fhWR{JG#`eFw3n;QQXC$
zVWz{+eQZW`@o72xgcD0k3<aC}57*Y4&F)~DzcC%f3Qdr6y{vrPP1md%v;}hiQF=or
z+ft-_D$}}93`s~*97P^ux`{6KEbtf=FEQWn@QI7_C#hT<rw44}?Ksf#bxm*0px}9&
z!&-DAEFWKn=WxMSx%qeAC)bFZc@_e2sA2n%M!7^|n~H0DuqvRI$m_C)!YjjygToOw
z;@w>h9~`yX2)#%`3flgp^;2TXtO)B<@n<Jgdw~z-loJ;@BhN0mx$Ga6EeY;Ld!$%C
z+R!UV-h?BlmPw{y`4F<x(}zZ`arZYl+^ddrdoIf1CWJl(<IHv5U4FaBeal$cQMgG^
z%77&CW!1HM+XRr?=@52bi?nyu4-d}rYCDHXR)y>e{_{`l58~+LwYXogZ~bcDEfMi?
z4B%fkv~bT9FN?SetQ1%P^S7eyZZUe32PLntc&>XBt4{7eKM!PveUI$uCVgmy9^V;F
zc9DX%exoNi{M6W#w3De=myjQct(Y>pL)`g@?4D4HjxPa0ZivXh0avPzTFw$F6EYfM
zI&)n8;WyGQwLPpyM#|@O-YebBf?V!?@*X_AH&0wd(aZsvFY`x$OPW7}<@sb9PjBJ2
zx6GCmX`7O<G3CH5bBeBGDf^lqu4AZ<rRrpsi`%(vS3PLv+bEV5L5Ab0FpR#Wu?bE-
z&-LETPN-ND@wemSxqmujKa7k93f9Nssh^Mps6}Jw20Y#?tHf!3CqirW33FD2(E9Ru
zAc%+eZnMb${oEq`Ui}mO-xzL{BX6qQ3I6p00h^WA=$!8Wd{)f=y56=zvoii*2~ZSo
z5gq&L5U=C31;8^q;RBW?N$CzAQe<C6Ip!EJH$_XoVL{plhwuEKKhK<FG>wy_ViO=s
z)K%*`alo4;v~fJ?JuF`;$CV{|Wo(1hL*CzU;uv7^Sc&(Q5x0P4c-GUYnAIo_&$5Y^
znmse_A3w28PG^@dAM>1K^=GQxFi(kvXIuHvD8a6z@lLlSc9gg3I865_T_iqpe(68?
zTbu91=w$Kk!N?LSCnoLS9)AwQGOMZb<egR5$Uq?=M5BwB_ai$;@vX2@2n|{~C4f<t
zXJ~n<m?q52Yel#z8@O^`UwY#aFX-p#ZxmG%d)c;+HMGBRrlzJ+h`A7GQomsmzn-%m
zoB>uxlB=?uM18YGyC191vdIo##h-PaoeDb>$pc_iOchWQU8a3o(uyp4!W6H@a6hcK
z&1#v`;CSPl0k42WT!qHv{@adx-`g#JEGb?-zVK1dj-VvX#zTB$N8L!;;4OiO_;!`G
zDyB$d^Wb|ki1%Ib3Ves{mkLY=Y$JeJL1sDUNX%R;Xet}gBqQuQ+<#K|i#N?n8x;3H
zXKPo)gVAovr69<Em(kgF&4+EH<;&0*lWW5z-i6UIS~L>gbRHgfTLi?{1Ss2*>wYqo
zY+bhdO(Eu&7GI{sVPv#59RJ0BabVpOk(`;RN>C_e3AJYVLdG77aG!Wy!vFidP|2{`
zCA03bm2tCj9%G*0j)vpQ^ttY=G@44AdJngXh&VjO1PuA-z`0|C;x9a=WXHk5fhWTg
zL1C{(5&;1}4p#ZnjvT)K_vk;{H1_bo(Eavgd;slw_H{b_9of-ngXNqzv(zERlzZOh
ze^!vU-M1+cMI81^bi`t#^vueQo;Vr(cf1Nli%PA2u{ya_E=m~d3vhdzkDLO>>PRtG
zt{YaUYv)1o5enkB5`6iGQ25yAzsHfetM-~?E<+#_Vy7tPe9Mw3AacY1FYEvo(0^u?
z_?lQyym;|bOk6F+$}L?IypfE5_Jn0;7Dq02&yx_ie~QFT?p06}V)A|3x+D)@R)|6l
z_kk*3m4I*T8)oWF?RwTef(M*p`jH=7^8JcwKOb>!UhI30A7mW!Px6}FuJ^pHfhGNL
ze`i`Xy_O;&*`83P4@-W-Uv#}QX}|kFo|@#8+r2xBS*rfC=`smxPo#?BA`#ONHuLIr
z#u=Io>3eNHbqX7M$C+a<G}`5*rJ}jx=PX}&&dfjV6kWHSsK4$OaJB?F6&_#TzeXAe
z>)(R3JuUB6BJ68fD}}wLxLjMvXLF626r|iNQuXn?*-e^~wpNlDP*9j^_f6147Z#YR
zF8M%gC1Jh&{~URT7{-KbYH$_JS2;m?MTjflzB%g_*hL$YYhg!bJC3yov|nnz(HjEi
z<x<Y9UHtU&PKLR1g;;X?Ua{y=l)fj_Hwvjev%#rK8<-G#k)B>%V>pA#4`+$`WFfAP
zoz)TvZXUVV`=qD>BvVS(0a~&Zr+9;;I!ss_7YGRpWScnGFLuOigpx_FhCew<L7ecm
z+_6VHT*BOm#ZJk)XGvDWj~aOD8C4Rs@Pbnek^lpHn~-Y>2)u~EJXS&gtr`+1yKF;q
z#!va!#=fQm&;vC-R}1RE*SA;khhnS#0?Y3n1f`|sd7a;q`{1mLBl<*_SGhz!oiqb`
z=&>A+#`mnth<+b#cPeI59RC+zi$~u-U1gqVV{Rm(eravKm&JF)n>{Nx$txCazyS{i
zJt90gxKerB$w7lMx2Gz-9)l6$ko<r*%WV@*H8m;V{WF=KGoCJnU+sseod2A4KW9od
zWdUraw8QHW;40_TXV-e{(3)h<uR-Q)>zqrGBiH@naFVRnl~ZxxOOc&lVol=q$$@_%
z<Jc2H9tCwj_XQo?+TOHUSEq-OG_}t{guzm_)ag4_gmi`lZz4KzEdu8M-cLEhhvR!(
zS2sU9G~?;Aq=Q(b0b@a6GktUoHU-V$Z0vwe2?vWXWeLZQxJe=1mUtw$B(rRysw#DG
zBJ7Sc(6!`Gc0q3gPC#5<?jcv;Rc2n6xKak0R(#eqsDNN9-{fuWT^HOCaaOQCWG!Ae
za|{GK^0Qdeq)AtWF{tRSo$T61aT@O!zr_ZQxHPJccw|jrws6%spJGL_3Ws!Y3%Xu#
zL^^(w6+cR8H1vDtjYMlUg!ud@!hsY5k2ka~?k}kf>2`D|r$F33oqPI%r8ai|y#&cK
z?>QChMky}&JVbJQbdX#eas#Y$&c_E`8m*M<%(H6W?A2Hn7&^Fe)(dWlOzoq-y+5Cp
zqXYL7FRkUt+1rBBujc#4-WH(&qx!Skl~Car)0`$@tiGJFhN+D|JFQgGp1<wgRLYs@
z&+;If3ACYK)V`H`sW3Q_EYDgMmF(sBzPLfzG*rxf5iZ>Cnel%fDSW}ZC9g)yw;mT7
zw=u<GY#d#185MtJ?v|UJygm!vv`&osOlINSFD-G^_SMc0I$*5_*Lbu{tw{&Z9OHRV
zep1FAY|13Y%TSoa{iivw`eCfVl@c-yXNA;g@89h7v5Q1TVtPOQ;?MuYIy^PWeO*ir
z7Qex4qZycIYS~ZWDi1S_u?vo*Lj!=xXg?_MX(+3tB}GVs!|52~K@cJajYOLd+wF&$
zUiPe}J4-`t=l7(O1%U~5&wmSll0F1UNTfOMv9L`@q#dGewiPxsXo&e=T1%k+o^Z%8
z78w|d6>6H62Xs;6`BD{j`bShERD^Jz%TnTm#@8@;5@FpZ{_<+sI$DF128yX?)*X34
z)eX+x=^bfqjjeA?S5_=UhW4vlXKlA@Zf7jvl5mFs3Ri*9og<gSuy5YLU&7Z1@n_-Z
z0GwzGu_Ih$9tAE&%OY#i50Ao(T+_MvI)0v`u3TLrN5!=Oh0PvjcwOKRl*&Kr(AX=3
zL&e!XvPa8HqHsx5D-^IeZIR62X`9*{Aj=1eYvZ4zmvOn{kM~H7H4JBo+paS{sNQ^0
zT4+5X5uE=)mY_MvBL4}Th)U&M0Kui%jhaXuf^Ag}z_eopF`_jq>tF^kNd86SGJdY~
zWzJTl!d_WRw|W8BGQb=y3=W~IiOG%f$5?O<g^*uR&4jzQPUi4F#`Be&xSyJuB>es8
zS9wwG>MCU2iPz;+<6tLkVyIBYrm~emTW@z1bF9Ap7Co8yn{ukr+0FbhRzeu5ig?-L
zHObW1wSc*I<G}rav9aexB?qDsybIXn8|v$ZfA=UnVDu)2Blg5y3<^?Gh~w54lpmoo
z|G^txirg-H`@C}L7ZW;!UH|Q#Vl|T}iF|B)4epbYItCE*t_!HD*9zvX+suBo(_j<K
zaxAc#meXZY4AC5>ErXZ{yX6?h#bepZ5U43qpnqhwDG#MTj4v<Wz%p1ged}U{;BtJZ
z(b&|j&}9QvXL#>K4{e<9V|gde9eX@XO5t=rzOUnX@D_d9!1TV&6S>_v4ZJSPP(xnN
z3?$Khd^hwYPCLA|Y;ume>2qAiikCY=mzyu+HNe-?R+krWJoN^v_;HSZwY7MOdJ&i(
zczyo@EC-K9YaUK3(w-P%N?}(RMRu4Qd{Or-le2X2#v_!|8l_kE-{OBhXL?;m)=yS&
zGHkHsEoU)zn}n<`+TZ&Z7+iUMPrjZ%uuO>05c=&I;-BiRtxY7@xwaSF99H(Df1BKC
zdzbn}Fz}z#D>9DAEp9M~L{MvwrL@gIEq^_3ez58bh7B*3_nv}u*Tj0a7pjDRYDL$}
zqMt>FG(K{FxeVYbOwuG1!`_F7Jv){M)Q3+>%PNs^y<WqKy09|-Ps(6wC_x#&dx(l2
zdCtPhE$r7KU7^(j0{#aLo0QWduTOBNshnfh?g+J*EUnY@x#P}(J#Ts!kO$WIj|A%}
zC7_G+0``=+U0^J3Ffw{f34=G1whOx79;%)y>O%JS_u0|EZ=W7sHFtFQeOwh;yM<2M
z{W`qtUi>z-8Cdr}M;eJcOocCqi`wfbHng<JR%wwvq+r3?KHI6(Et>oIkg@K6^uPQz
z6)5(9Pt^TJ@;_&+UcarzIEla{<+T`2qLqp;*fjVWCI0_^hyS?u+k#{s>wo|3KqWzO
zm-^p>FW*H-tpCrN|8KExkg2-B|E{~CGWmZjdkyTcEc)Ke$Bg?Q{G%YFDqSmS67qil
D<oLj|

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_006.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_006.png
new file mode 100644
index 0000000000000000000000000000000000000000..9f00ecf346ef3ed079e7c9a1346762f15f6d2f60
GIT binary patch
literal 127303
zcmYgXWl$W!w#KtqaEHL+?he5rxGV&hEbi_OS=`-%vq%W;L4!+hcMtCFkNfA{TXm*q
ze)RNI_4M?d^GSrN@)vXzViXt{7<4&VDG&?{oZI_#5CG}DEw)i~em@DgN^85SJD9tA
zfSt`?l)$czwhpegR>l<WX3j2F4)$EEe5_o|6qc^8jxK_1Y<B<Ofz`p;f^AGpa{ql0
zWJg&Y7Z?}}@P7{2QqdAC7+4q>IVo`s&mYI_9zRUAy#Bp?-s>Np74T3Q%bcX&`ymbo
z`GXWaM`+U3PgoT*e{@sXQ`z&l*KFu;?ww@UB?aFv1yoH{CAKOWswr{taNBj97U&eU
z|9TgWSg0lj_%~T1lKFA=tJ`?Xp75U6`khzEoURMLh0}kd%4$&S3H?W(&i{nMfDXvx
zzg;XYBm?V;PMOEH{y)NCEaX=I_ZAG;vBckX{89fO<ZwS|-2V4Lq6qv7hyODKvp?Wl
zN4yb8o`Z`^k~T>ka3SAWQs<eG*HElm!6_^poovy+J|6pEgJUu9`~~B`4V&v<Bh0d_
zj0~`??Xj7Ui^auDUzvIdAsN}{!o444SM$7Mu@7;d{@ZlSr*~3-(uPc7?#BZj_n!Ae
zp9lw3J|QB@)#dt*?cVYkbp?I@&~lxqKxiZs1VpdLw?qI0S&G6GC6{9W<LRkU$Pgi5
z$Q&_{8-$`b($7hT)+06H)gdMyF-HvL4h3?sp!TrGgD??*V9dM*k&`=I7zskcDG6QL
zDNgnxDL_!ZxRnyFw3J%?r9KcrKBVL`IX7_-g?(>67RXo@832jDcuGJD<&FnohXOnH
zu%yAyO<xh@3tb^+4g6l$9JeA543rKD=IYhj8r;F3?Ay0lL;L3`rJi*KQ6^9B@--kJ
zp!oi@wU?*A)Pnz60!WHcf=Rk=c5D3Y=}niYlj~2E*BDcfqTpz!=YUZ2_S~|g{)dXM
zKBP3UcHan*vWw}Bch4!s0R`+5go(SLpdKIcjg6cDG3$kn6Ty$<ekkKi>J`j|xVUNM
z0OUn|(fQ8JwC0WB<%^UrVEa@Px{uSg+}f@7WJ5(!ns61{N$R+LS5%R>@<;bRVA&#W
z%aFND&GNp0m-4qiO`m*@pFCDOR7^VDjnT7s?Z3HOXG%CFv_NSJEN}KM+A%_tcQ|z(
zpB@QG$V*yVNv7GYyhQpgIj^p`aMMtk$Jn-{*I#Z7S$B#kX!5Kx+D^CZqV;&^+n@HF
z5?K{O%wk2|$?pO#%6+$II6E+<GCdUo0xxts#>Wv_Y-3gYp3fFrOg>Cu)<5HS-hJro
z!BG(zdwuz0$pfD}1pqoSXI1(VP*Isg$0<#$E=zi75rl#_d_<t2h!Cikxka&#cDjXA
zi&({)Z-eL-YCpzRI>z<QRb1{1GtTxl0J-jp>p5)wP3e}LzpI8S2_Z;=Ato6~Za%-H
zHEnXzDKs?n;PjL>X$%tqc2qM-BERc-1+DJNg_(s-tzs4+d1D>WOAvC#<Ifg8;|5rq
zP$jpSi)I(_i<Xt9R?n=7Ks)07$lk&FWrOH_iWEmA7$%4{o*u}8XCIo9v0-I#CBBqS
z&{>}$k%269SW^#vVKT$?u(K;&y(N%`VA7LOLi@R4sYAJ!NgSR%pxiYq1S2(A=Ffak
zOaLBECp(%6oZFHl13XmY$<GGj`?`04DN-Sh0zd7@u7ooX8)1$cA|)OUBI_;%^aQe_
zS@9yIY=N0gkimgw^ofnd+ANOu6NYnkOTkvU7yYGpKb80!G%ucFB%UEEg}a+iB49#n
z87y(aykuP(AvOu)H}Uvgy8izA`OarH0gpXpW~NVJveEz+cC-jPJ?+#wP6F=8d#ZtL
z9BHiLgDEa{!tuDgv7zsG>@L=9sj&dcI9tXL0#CwrqEu8o-55%YfbFvk?beVRAy(=P
zM-H^1$?Y)uw5aPFyRKU1dhCmvkKb>bgRRF4(_IHpW9KpzX?w;yo=P&ZvP>>#%Fad_
zp8R7T%G8s7k&$eRz8sKWIrtwL)#)>0!>B!V#Dk0h5q8T(TsoHzte1<z5BfQ#oDl@{
z5L{+90vIgvje>ZsXc1(X@2Xb%>Uo%!bkg&;WJS&boEg-i$b~qcZ5P^8bql3GCsAjN
zfU5WywDjr^m3|t^_F+W$2YRmA+=v9^V@l)&y}pXfRT$vJrI;1kzyZ-A@ujkb$(G~f
zaz(f~XKLkNwOR2ni^Z5iG>ONrv<+FSt1%s}N1)g_<h@~c<atgnd*?m;|FYZN93A5~
zH-VgsxT$6!I}OKG1dwb!KF5Gq0Tu#Ss%|k)oB3NSZ$hjkZ`OT|O1q(nl~qE^g8)h&
z9Lu=mOCb5gtAKCSg^zeq<#&r)#agE;1Qn6?Fl)aIgH}H0-PvWL{j-uaedBeP0i7mu
zD&%<@TsY`<t0W)<95{DbpCwIU`s^~RS!s;2@R||+m9QtdhO?hhJYh=v;hQcKH^(Bf
zn^-BgWe{ejYzQR(3$#tq%PLmhEaT4~|Hltu>=kImNqtKHwh&=)81NWTmPzU*^06$X
znECs6B5ynLc+|4OQ5eLL>t_Je#SB=S{togsMGTlrUDar=MO`o~iibCHC?VegloPc3
zm^jrTTj%-h`kEa<9e)zEv!`mX32{ger2y!~G>)#%x_N54RFFyk+^SJQLIxTwBPYbp
z^R-+vM@JY2APA_2n@72-84Qf1EJB-H63d?uv@XxMfU8nNz-lfdJ>os>;$?AN*t!2x
zwVA1d9r5X~f|FR9kf+)%$-7<J@en5WVIlWQ{p{`TI#zjti(R%*&BFPR9~BifIyROP
zE#5k1rRtM5MWitv->acT(BWSDnHS&NMcgYy9(rkQw3UXQFlW2kk?pJXv1+l4liCmy
z?&8Im5D7cTOsBb0fir_PX=rktFFYPj=-i5X%5+L+cm%Ns4715}7s?YzDK^~=)8%N(
zdm@q+I^P?Q$^COMg8@_#_zAs(eTbjdKwE!)z0X^%kS@!(7|W@7Wn&R4I57O_>IQ#w
ze*fUD0j7cnuzzq$nIVtq30s_9G<&)9p_D(d;2@q;sT4QHJjvLLh=qq`y-=0@8{oTq
zi(-Q&80?hbU8sy1a`OG9{8>DpH-Aa3dTq^Ke_rMbke@Y&oRU4BGj?p>y;qSTgj@ou
zDV-jy`bFa>o}yLI6mHZ=GLsofAEkIbvDXv0k0{}{O29FP?jKIWX&?%Ej`uTyi~7jQ
zL*ob$;V>{nvV1L%DN7O{UbK}ZPfS7MB&*pFkpYu$GKD~aNT07In`O4Bof?vS(S<CS
zPg)~UE68W>%k%|c_f1}@;rHK_W8};j&fP7_#CSIw8KTvdxa44&1H!1N5$2S`>yOd3
z{;T$!i?GR|vJEQE68UO_jfe1dh&3;tk|O~<6dmDHFjzRqK-|2gq)MCuymT4gvXrTP
z`d8+ZDQ@xd=(1n8Onp+LiNO*rT@l+~_Q}a4t%G7`P-;>nD2k3%m6`SOL;Bg3im+fr
zI&M|szI2#x{ALw(nKx*=8ROgJch@5dVMm)_jFt<IXEMP?5J8Cnsrke*h#6CeOVKM5
zH=sP6`peAQ{lPF&sSp`?F58n8WOx><)~BzRgmUED&XJHVxR8@Stp^MmoIQCIwYCxw
z66fodvw!{^bM)KQ%`M-Rj+uO1u3qNgYvPCcH!^Zd_Y>i;rng?b2lBV8fU42~Q8Yq5
zuGlHYoI+g+Os2aED?<K8Ny>ySH+}ElK7)HEDP!(-KN}chw>N$xlhOrQ=}cUD(iHt3
zd(gxof5iPpt@QiK0PQ0>PJ{0!pG)L9H=3HCA}W7bgR#?x^O`k)Xp+1%)Kp&La}E2c
zndLGA>;_q?a>~iXzFS>z<6r%VLx--PDPv@|Q`E@<aPxudDLvfi-SkolN~ID~KJE6o
zb}CVg0AY#Y6cjKns7!;SC;~H47b?-d*md=SC*g@N3S*gcC)F)2doE1Qgbau|(I%7;
za=x=u<H4X$VY=^7snM2V3a6AVG#t44TtJT1=r2>GY!x&=BNkXJUZ-PWK5Ts1%Sji@
za=B1AKAA;k>g!4<X-x@bFBk(ciU+3FN9g0`?nHE!#LF@)mH3n^?Q8Z_W(N$Q=vOD2
ze1L=E!Nq8mJMVX?E2@ymNGs{<adA(m1gMBEu5p)XVAqG(EpRis$y;X}6plOlzPX^G
zRhf<UyA1Az8t*DM4kaDN^M~YRt9$6uYs*JiWnMsM%6qRK^8ZasG*WhR+lIppLz86T
zihx+D5?)A=m_bruz%Q|%wt`2h`V)BxCx?09m-72~)o^)J*)ch#OfW+u=_A?EwxNGx
z&V*#b!XDDdpAlI*?~O5??4g(!=D!N~A$FPOzaX{pIh1($t_x%C2>FzhWP3)|xD}d2
z;usIEnF(%LWw;J|E#K8Ol$C>cOVi+L5BS?+X>&>v#*;AoA`~mrgw}3hIUiU<LkBFm
zk4OcaAuM{0@rtxqpl`jkwm+iegmx?mo$L*8`wqpk*w=iw(ck+dx1K%pm;2Nh(WVy<
zWkzuLj!W9afP>Drow(3i65S~v6bZ85bTk#u^$DhVu_V4_(xx_Kg0RBIj9#bffiNt9
zN&ZTusrc)UcU=73@`CdrwVBs9cT<^JyDjY>FoOo9kDQN>bVq7Jyu4SKD{*Q(+&^^+
zkKa0F1)ahcRaYwob(<R72J@}^GM#Q<)<r2iPoBLHWGWHZF{U0}&9rmMcLvMv#KDjb
z^jVUph9Hd8FWE>-Db6nQad=A)S@BW{j}Nd%yPJmdXYYP}Wwy95$os>M?ur`}ZwVWr
zfimB~t|-<UJmf*Vr#aFSpIBPB_8K1F@^LJD2AP3l6almlNUh34>MVRUBlk0f7S#pz
zDOaewkKm-26Ai750}}2+>MFEpv_I?kGtxmKe-=fY6D=Yczt}3fIDY6WTwy^oCdDgq
zqIP7(ER*lU6}gH1KNynvI1WMKaPs@)$dNqU?wXox1hsv22TgWTCC=R`Ti%Oj><@SA
z<3roy?7z)F*zlg_Iede26i2YTDPj3!i7xRi9cju@b12h#B+RwDl5y>Lc}Lx9l{J1s
z%F?$}yoFnZvK>iX7tPf}oi&C3{_k`RCE)Mq=zEqNp@@R#Hj}%jqT0hAF7<<i8U_nO
zfiT6sl>dv|SC*mWx*+O4#ps#Bgo=u(B`i-&p(vWhW5O<xl&OO$G98Hwma+roceYCD
zHV#6ER*AM@L90i&;%^czf$r}rDcZWad30G~_raB!#uPTj0$346N)`VK;o_b;ZW$6Y
z>?=EVm0WLbIgWYw-FGPHRB}v@QdzW5Bx00fWxjmnT@omCa#H&Bi4iJGtM7kU>EXDw
zBNt2h`Sz#5RT^rl8OOvrIRTOmF&)s-ix|4xM;!<zrX~%8Y&U)t;QK@ltx-e`pWo-F
z)9{Ov){d9^y$~C+hgdJ1dye=s2MJrt+L~DA(PuJ+GeRoZRtp1^$t@YDThp6hfk6=*
zuH^TfG>fhJCXuD>#$rv{q~-6t=%w22iNRK4KI!-GeiSqy66Pkj)2wor6P5O?%^N*J
zgQnPOyy~+A9TatBiSYt$CeBv=KId)zkqVrd9&D^X3gi%SAOJX6#6>t2jYs0Dk2`0J
zjUeu0;Q8tzxXFtrPMA$IQ-y~h9zKaLM;!?{X~gg;KV13w+95cWeO*QnnvVO*z8NiR
zKrewd(u<1^z{#ae_{i_?O0=qsBI@j#<6NQW(D5O^clZVbhkHUDfG!aRAS;mmlFh}=
z!ZlWyB%{|5g`A!IPySQ(6s;7N{6O4$MZj|;+`b1sZP_#QzC;dv{Q1`^*S90>wZjm5
zh&uHIQ|2ej5v}E{4S*~`en~1nYDNUCv9pl>!HWCeV#k&I<0BtcQN)~x?g)P`q#v`_
zI>$PUPiyDZlTw<IGFpOpIaV$_mVg2=k3vIa)EuwMaC^&UO{TJyd^d!dnRIh7$WL~$
z7n8QZIz@AYDorbIYlxH%wTTH5!+)o0gTo>rhMkE5zAshHH-z8R8gTMl)?4dtJ@ckR
zo{u3gCnG45)$I9I4E}CK5rTzdY0ODT0HIGtNk|AEKgfo|&le*m1gS9*u-m06{?5v&
zG^Cxg+o)x#GTjcnSqWHg7ryG<p2@=6hEIfGZ3*HxbkUWjwCpL6|54KY;j-D_>Tx$s
z_>;rylCZ9hE4o2pL<==Y&?^V7$C8adJKoPUc<%b<_A=k--s|<m;ms$qZ>rpJ)<|Tb
zQJOW44oqmuMTCtPrppqjT97b!`(mrZ;iWFW>Gi2a|04r+5WxnYpY<$pjJb{3b0^6P
zq)+I~TPV%Q9nq-s=}7o-(1cXfD=~EbG3V@SK|JUieFO<Zq!LC<Z@n#Kid4iIf<s43
zwCesb_x_n+fdIK4>|DXU(C9}{VI8CLHaZ}?ghXI>OtVdknHm%HCnyAw*U$wX-e!J6
zRJhB}=#BBtZ`-E+5bFw2sUo>x6Y(Alj|T}K?Cz3?z|7}i;2Y{)5G>xap#K^r>U`LZ
zD(M@?LxNt9n~wZpDHcA*hz=mE$#;<E{i3{n9B|M3O7067-bG26qt?e^;Vx!CmWZQA
ziSM>lkYsbyU-F58m{Ja@28d2`)p2L8*|_(PhtK4j*&<W2K-d8(VN<|R_rQBx9zH#=
zy?e%~zLq|<xKQrLm}uOD*gOLueR#P%NA#YN7*JVnCWkGbakUXc4XeUC!`LaJCBrKA
zlZcp8OQ@DZjnyg)m1KL7Iz#Oub13**R-$!_Y`dYwWiNwyi3PjTps6~4WysOLjEsx{
zC4LQgF5$~A0_{$4>`7}?FS}wrcDVLOU9~+yzQ;rJN-aiiC@caBgGpa4PU=^b0@dlu
zOo({YA`YF7Av1>{d-NGkjU5RvGu<3QUVrX~v8*%&kcc<_nu3WGie-dK#y|;GD=>iC
zxT47USC{`ldVG4A7=IDw)AzAkex=bC2pXD>6t16hqSqk=<TxbYyopL!i5Kxu<}fpj
z<LC}Bn(wHw{sm{^OHviKy}gisv$Vt<aFR9S^2{I4-@Bd6v}fnAjF)57A1L}7SIYd?
zaeXzHDq)PH15B`L@Og#_k5(n8*Nyb?=}D{K{pPw^zD>vq5i6KJQgpsJ=E!T_;B15K
z^|BKOQe|+{rPOQ+vYLujc<W+!8HrlZ46NFk@M!&6>gUL%b($bkD2^+gJYqJy%h{kC
z<J=NV?pq>tbMH+|YrEDKd)Id7cf8`o>d|&F`o3lhp99`5L@yhBI@Xk{Ur*U#DiBiV
z>hV$2L71MvV(x^go&8A@Zrx80u8BbjBEc<Y&W$oMEQ}$D!@JmZE_M0>Kg>tJed?)A
z0Bth$>F2fVbVnT=kQak0F7L>nNFc#AHCU8h(Gs&IODJVuz7nfp8Ql|7QUVcd9tZ9x
zKKQ2?w6blFzh0|&9oP7j&i@-~#X?6P$uIgO(-w+)Ye?HG|J>aTC(pn<B5jhWXkKlL
z3dEIYBPIFl%Ih<oY&u0$dw|PTcXj^<LxuGigLSLDLAPS-L8Qg^Baj(;i2J}5C7eB4
zo7%gK9ieC%xiPmcg*@PBx#jYb>HcK>ndS81pP$B248hHq(FtcX7nPeV^GG+lZ4rAA
zt=f*skm=K2u#A$^@jgSNy<rq^I~WcMw0`<kR;r#~s&zRnCFEg8<JD%TFbSju^?!v-
zv8##^ax?tR-P*Z;Cv24COrXxL-zS&l5aAr$*r0%d@J5`~<P(NWs|`Mk#P73H<SHb|
zRap3gVMppObXg8>IH4e99WEjwvZ8w7f$cco?2bufnNZ&T$O4d)G+SPZH65f%H>=m*
z3@bDJS(=do1YC+gp+pG+vXCyv;FLBD>`V?!P8A6Abs3x0WcQ-R7DfbPPS+0JU90J{
zf-`6{`5_dLE>5%}OegK8m!)%Gwk~oh0C1tj@59?aEz!SA5j-&E_{_nx#>wu3zq5Ze
zgQ#l~p$zB{m=SCWSE1*ErP&&98e9~od%Y)g`4F$mcFE&~j-+=;0}1JO%fr9rK`7Vr
z>X)}rldw2P=Ne@d+9EcL;zH!iVD&E;SA%{={f%7L=pQH6`NZlLH})qBi>BqAGjMff
zCERp*e$9lx7t-<{PW0<M4UN1WpvUV0@oWJnM^!pzG8F6(@ops!-p14(I*jdRbm4I-
zlx>Yb*^qa9@kZvm=18G_<%D&^gl=7Zeee5<Q3#<S9{uOh^6`Sd7zg>(cROpb)=Y>(
zu6Rn{>*lNPCRgrmj>F2;c%mRGd&J8>eTsa9shIL@&J1SDzLjh0%3Vz9V59rEfI+T1
z(XK-BCp~%fFJAWv4(y@=VV(88P%YbDXJ+Q<{%TTPe=_hg8U}izB9!54q`vn7&Gk*p
zgb;i8HpUxLCR_u3UNKxfnk>rqhL=Ly)#{YhJ;PWq^?jY_gI}!Z4ZtRbQmbOTHlI%1
zLqvRftk$CP56-lLu&`DEX~>ULymKnVd>wuLJl%@Kqf=?w2(uPbE-ys|V_odcc5Y)j
zm;v5{qqBdtOsw0Z9vvb2v^`ZXhgBCVK|S|>7ULh7P!R$D_VyTXQ{T%{836pUB0H~O
zkd>V6fY?lT8uwVLRkgcubW_JoMcYfZM~(jrlH&BvUce@SfY%o8(?`-?(H_V4MOH&J
z!@kz5XxY_o0l_kTy~+5Y;knjJl#Q6;fzhh+j{99`^M+bfo3vWa7858MgjxIUGtqcA
z^1uRZC#<gF1dZMRK5RH36$C{sXnf_b+O6mvLdQHshH@cDz?NaL(JQ)UQvC6Qk1WlA
z6|bM|JSALlNl8ROxI8qN0X=bux<?Y88k>P}7G?Jm2+`m#hG6REsI+Kq{CVmajB7uE
zmjzQvz+*LtMus!Fn>et*GxWFz1(WX~TEiwF<xehAQNw}3G}DB<GbAw!E~8b2>=`0F
zDywl@Ba}3iEa%D8SttE&w!!r8xE8y~wn0qniQWVPBlQfHyk&hdV7kAaC-m)m35-r^
z-$~PI{|Eou_MO_rSzm(Hj?Y4k9Ci(wZYTsAxC9l1#v#`ONQD=Fx@v#TwooGoMgxKu
zsxY7vc1f0<YQ^=;a%SF(jaE$lgqt6LS$q!LvGeT--@n||2siag-mNN3J8-O+fYV{Y
zn3u;%aq&a9PJf_$2}fg7>!O20L))Z+zJV?tr(HGJ6ljzc6TeMjrpNG5dltrIzT?;I
z<yd7t-|L;3T6<@<kDmn0_V3V$+ti20bbFkM<c5)V#+eQmgX|k!6isRk%4bXI%9)w}
zVj+kW+r1MZ@1T2MH)Y7~ZUFDP9AKf^{<Fv01HSFD&co|Wgxu-VKfW!R>sF~Dgm0)3
zl((I(ZoY&gt{#QlJHIbA%4w{jaGWXIZlbTepcnXP2a&KZ9nXl$+5FMTv?`*txlEy#
z(YJemZY1S>Y)p!YYij2D$fzXF16%?lS7fMl6#MnTksKI_<?#z1)GKasQ)%TjwTZ&l
z^I0$Ma{u#P;%yp{%kYZ3T8`4^?S+W=U&~50ntOUGWW`@?IX^*MJzD?HfVoTLiZ>&O
z;xg!sYj3{0(O0uX<p+uU`-gwt9iGPITgLNue6+5*x@%C%cFdlRROSnvEOjC^Htpzd
z%Bf%iNgpWY^_`o`D$o4+ZLTj}n!dZhgKrK!n%9Wl9s3)>TgGBY9~Ui-C;(;51~t<f
z+I_J-u+V7orHw5;Q7Nl&^EYxLLzMFnRkYzsiW^zjvXKW9`Qrqe^I8Z&ZeuZ^viZ{5
zzWIXoBbk7Ae-W#ciTof+sZ5pf#OwPPg*R`zL1H90si|lSNw!@OmRYDy&t{!o-BYcS
zR#axK)&IpVoNk+2<E95ieb0*J0;BnwV%2_O!c|bq|I(H2t*4Qq4+4EEQ6}Bl=ol!~
z#1HQdcOJ^Ox%$Xksn@eF#@FDT0S$q>F@=}RL<suRB}QIpVv<SQq8}?$A-?h9UObQ>
zRZV`IHc3NeR8<Y#{2x9c5eZea@U3W5*kT!u;;+?6{9xI^lu8_8=`Z0NXjpsAaxgKG
znoM6)(5%u-jMI6$k3Gkb(Wd`!;&4A>2M5w$vj1)FqTc&=a7q?l6{Lv}q(Yfs$kA}(
z+NBC80{0_|s}b&HX!AC1cc7{7>`H2B5s>@7jXMogG#O|98%~E0kENtD$rVsK9>i7`
zh6f7ewy0wln4qRNUe1;k97iibXrsIN>lYgXIh0W4!kgN{LBZzJdoD)5Nd`D4z)^O3
z9*Oca83!4~+N@U#ChL?ac;8o%kg)vghD<R0QuI6Gl!V5p5@aW)S4aM?;D*J0GFeV#
zNe_)4wMv#Mo4$PQ2F#wO7q|wJ_pQFOEkg#PPILk1C2t|!&KI*JKM0V=uPhsh0XSo;
zHCZ{&FfpkEQ@(h3()4&qzf&!1G}22Jf2uRGnP}6>Gsr_ZNBSrm{)sxgY%cINtvI8S
z@Y(BTaOIO<PIf*C(Uzzbbov=MJjoXzf^3Sy<Dv!7FoB#_iNh9vdWtj!^{NyrJPVyB
z_s?J^h$f^|v*qCDh#Y8p&?L@r#4!=o-M$MCcGO$U5^13kc?#U7pvuIYKtu>_YKklT
zfRQ-j!NKJgTtoz7WScG(GDVl81b7|3K861rxvsa?Z)6=Ai%@%<vf;3W&qpuNlr4hu
z=JK*X6v?%&ksrBB*JZ9f>!Q@MJ(Fb09UZmOSlW^N(BYxHd52o2ktk$zCm4$Kbbt8S
zXm8OkEaK$JFo&gb&hq>oN5%ORkswG@mYamnmNr2)ajZExl0}1l{C>75`A4S@U+0B`
ze+*MqN2qNjYWUS}mXVDyzRr;G*Pgv}BVED5Q%bmMj`Ilr!HCt7ET4r-;ZMHjK>L;x
zx06t~%$p;=q!IUo5j-9J(tM3&j5{<#Ex@0CbdrZc(N}eufdQ?#q}Yk7WMfaqaM^5d
z>F1A`;QWIIbW*6!b;ldkJMehAkigIu8UfTVaukzf#Bo|w64FU$dQZq6w)SqT>s5DW
zcHrRsUYIuJmi!GShQx8keSWzj@{ew#ipd7y14C2)Dv|9j<d;MQ6W=Uh`;Qa`eEVVE
z4$+D`*HgkL*ejbazs9&WA*78p!SYt}{_t31C-gdz8gfDn(CBq;>3t_5M@A4)`UJ0B
zSV-qhO_}8;ICZphwlBiYOinUK_m%He!rj%RWJ~Zg27>Yn@v~ovVj8OqR%m)?B@&eP
z*wNA|Z86?yVdxU&sCR^XOK|eHCEBdUQ&UG%H0eOFE_ap^sEoNp<y!@h6m~ey7a=m9
zEz)B0ALmD*Zb$bXofjlO>imRcgiLMgUEV11f>Td?1>k|UMfd=$#)xe3AL99915!re
z{94oXTR#^{?M~K_zk3J@mfo&#N$KN}Aj1xqXzCUar7Sdqe_V|%{&^Pz$Ts7H+2t!V
z_kI|(rt9c#?BtsuBz$9MW0z1W*Vx&7>JSr5Gxzl6YCK6CcXP3}j@9QK=v~l$e(1H{
ztqHO$p^##_(hC7UhfBN{T%OBu?LnQ7qU3#dFIr~4hRks5C5GiD#I!Hp;cb_iLyX-r
zT$l1yORMN_a^A^%taoa}8H!8dWK%kSG;=t5c5t3E?r~w_;Kii*BvsMy?||;^JVNF9
z^Q`94o6ZzE!@<en=Ce!H&YFWimL#H+Uia-uBH6x-Ro`O`Pjj(cyCzlEUv}Y|h|><(
z`ofDGuk_L1!A1rJ9>(dL9pY&7GGDEw3<HlNr^(TJ^5#Eg1l)>0@*PtroSED~wazQT
zI^+l&78?9e^$HkKquj)%k_?R1yG>snKRy2bAiIFApc5FB8Yvt0jf7B_qP~k;@;C=K
z!2xVN+SQV%Ss|rRzCc7F*u$Kuh1vf3bKCa7+TZ(qfUZqt#}>-ZS>gfZvEC29`9|Fx
z<mH?Rb1PS)J}pNj{+Uvo%5pk&=IBGF!HT>QCmO1{4Od!5_r=!4w0!HXTurO5{d<ez
zeL57Gr<L~Jtv+M?UnSs!RGZ=8$VC0q*(c_|pLr!S-r!;F?FPbi+}`-y?;5d8FYi&R
zMn|J@@$&uwfn!alqOYn87Yw#TIw~is+cY-rFOs68;|5yZBrqRgKir+{RlSZFU6!c?
zAc^{7f!%=#y|ZtNro=)l89J=~VZxhQA1&&McbtjP$_d%1L%%cGJ$`VpU2Y6=akqZG
z;EGc>yj7KV`nK-mZo59A{9IyJ2NX}Q*+sZO9<<!ca!VW%@;7p_jB)>=o=?T^{Edx1
zx9bi|H(nm<Xp7?tnFHE)7nRV(#lv!O(H)$SiG?Fn+PuTB)pjHmCR{zfTA2}>;Nb@1
zUFQh-8(bDqDcx^$+`CK6#dzgWbg{<+cUcR{NLIa{OfvzKjb;4pM!&2woz^z2Ncc~L
z*+<xAX>sZ)$>|cr<i)`U2qPh0zF+%o7?`J}CwiUq-H(NHZCNE7G$rNb8Z--AOB|&C
z=+pF_pnxU6p+v+3DU%8N%R3iHK+f;k#U-lfoquRp80<uWfpB0@a^^%_c=8wA2F!$R
zc3}yc_u5uc*De^BZruGDc{{$mm)-b=70qqguGoMmD%|H%-Uj+xFg6vn9>@;qcV~H@
zD;dV5Kkl||n())%?FpHqNCZZnuzkDOT&Rd}G=zheMfl9hN|)#DMq5jIpdB9>Sz7ga
zUG$Hnf)cD*g(XZF^-W3DCc~|nccn`$(4Q|33^aSWJ|FoEeQ1mxmp)gO;q&I4aq;l<
zdF|yyoh;}Fj4B<A-jf=%8zeKK2UDC@=R+t_B%UdxygyBxrjeAL8Ng6%Dcwxuo7gV>
zhF^9so&EVdME=I{MJ1;TW83HATC@;XcDhKjJieBOl05?L16HH#Z~RqBNm5Zy2pQjs
zY#yHYe$CAa>O$L#ev!>~%0FcD*aUGCS`dH_-Cq_LK|=u#g2}z?6(j-Rlvm9amN$Fb
zHhror@vw~6^FO8&XNVQ;$dU~BYG$gCU6ewRq8MZYPB#~lu-irrD)<OXFKH}*j{x+C
zhN9sbRw!50Mw_H-$d6ES30U9IsD4yC|L%mjQ0>kldK<b-^odko=ZRK^uK6^xy!3TC
zO-GwC$e*EC8AlJ^>iwSt8%zHAZzGuyR=26IExhFSLJfK8+5=8qo(ta?kaJ(kMy}+K
zTbzqQNep#{uY>Np97SSGN>Gh|Yd)u&0k5-f8y!z)Z#jqp>j!Z*6>?d@MGP6%(vYfb
zz31G%P&XT|t0SX<O_Q*PN4~QXkWGi;&gUG_uvpUFKOe|<VujWq+=`$f=H<qdt@8wn
zgQFpw#}RH_M$KcxJmO!I$-~kjuaj}kn;`@Rh@_tG<hc(!CHGqdqT>-b17-Dwo;OFs
zlj5jo+*ZJIIOz?@iq|IX?b)<2yTM}zRx8)i4WB)cvA<vu=;t?MW4R=X>ry_2>+78Z
z62(MW0P)v9HB>G$hdi~MtM<PuHM_qhShwCa4P^m&3O-Z(Hd!GFXxX}&wdT<=`~l43
zkfSj`D<p7*%wSD(B}HT0CaLT>Rv^Ttba684FUufeA<-fWNf-Xx<o~#68>?A9^ezue
z2$fS|{W5`~swTyHG?k*FuFo3C?r4B2T8ioEM#@K$gp|J&;dbG3j5(3gC`+k4sA+fk
z6VfF{iN<9gQ`~(&tPnFLg2Xx;Cs`HV3?F7jY+CDQDdCzIuIlLz(;W^Urv^jWc)q`r
zflL<06p~6vOE_1qa9Epx=F$Fg(stqI5kF+A>9U(lIfFu$%<y*<q@82tWfdyl5N^fD
zJdm&dGg0xMc9R(aj!n}yAuH<oWoSerk(;d<c8HS?FK}zl1@weKIi?e}4Ab<47S))R
zH>sw|6Qdg=-yV{jsNX>{;67nFoWiM8H6~4vGr-qz0ps#DSCz>-@!L_j$%kB;th22x
zCZjK$)6$ID1*(=8e_fuMzumPICCL><qAM3V__#{!mH04Z6b~IabmSal=-eK!-W{SX
zM2TTSgx^Wjw%3~`a3=5GiAe6KUZa}LEIl(G1FM!wxuHWC*An=S;Ft!QU#+oW<aggA
z?D_W6_GGcvGk@BV#pb&k#oeyV7dZW<>Fb<pkX~4?N?{X}J+{G<Ni8MSA*3>VMQDxm
zD~nK)5Zg-N*J5K-!k<sLz?2Lf+x!Hz@JnG#BzY1@!k4l(DaNcGH-n;?=1O?ZA@)x$
z9s!Bo6}@v=XI-bnVzN<dlMSS&r#-)rw2S2+-E+ic+`85UrAT00A2nj6R4Zd{cwL1N
z?!&K?q6ko@X@<xYMtc>uiopSjjx+d|L&i%90IRSAUsdY`((lU$yp-}JzTbW$?$$R*
zM)#LmWplc0;5s8WDQ#`tE%n?7>dqQB)T#O*v)090F9Me<b=Nx$KrrS=uz3#^T4$4C
zNsy}avEMDxr?xYZ%8pm0;$mf|_0_v`DBBB;PmcGGK8b`>Zw{~Fn%o>9wXW=A;>V~8
zP1~9q;{JZOBhRO}RXNrynygVw2<&>>8~EqVoC5V(O%<J3;o{EQbM9N865pB_XSTF_
zR1{lVq|ezC%BXmpcY8#<NC!FAW2CCzS=x+PkIhBKxPNG6CF{pepNEs;M8`njPvb{_
z(+G;6Gbru*51Q@_Q9~GL@&gU@=5bh3?g^?oLpqX1d~caN0*rK%vnMOx<t)OXKoA8_
zX#E8-Qk-yoZuYXf<K?|DGT)!j7!%>vJp`A!5oQAG9hb$gVK4N$7x`0y$`pKPCVifs
zeMZOJy)T+H;GJpP6sx(TUj|J<Yp2ZlXD-(h;*1UWIDQ7nsHihXN{T(e-2HA;KSfDw
zZb|DtB>68I!DdBvm_&)i1fViade&^Q;b!9orgdr7Jy4l9Vex2}SgLi!-x`xIi#0ZR
z*+)>5ko<}+OPX-kl?#|0a*0$>UP!dc4gd~4Ka?GSOo<Z`VwJhyB`%NX=}MavUrvYL
z!!B#VdBWZx*0iif6pFPsBiEi-I9KL-;`zOun2>~QGtK{XPvC83HNAM1HVd<!@vkzO
z`Q?twOeA9F4N^ltr-YAeKA?d*TTZQtm*1<o8%M=I=`|l+-)r`0q1KEp{Zt-6Dkzhb
z$*$qbpLL4ac6J%K@G4avFJl(rdWRX|s<@CX6`a!{GFQ1QTVYtZW?9{A5<!=<FalEA
z`Vd)(P~<u|ts*Sn=o5IdXBcYMYRX9z*=Kf3E<9LCCZs9L9LP<bY9|5nD5SykzSq;@
zwB0CGS(w`gh4ewStaipoRYWeNR)z;$2Y2dL{FliTl~h1Pm1<Q@;RPFSi(O;f5WKQ*
z^*Z(Eo3~e+*Uh@zd~HT`8-I%t*dQb<BR%ym04tNlk7_MaD*fgsOR%*i#f0Ps#SW$9
z3i2x&p_{J6HE(jtvy7f2G>X7dEtLBTeBI1<SqAO{L=9C6f0^5g+vYwej9Zs*iW+Z5
z`~Gs^zU0Fhuw@tSmtu}4bny8PeDsc@uBASMYAEbUbv?dQx%f0UN8r+sRiw)m!VRr1
zT%#Oq)_~1pk+avaM&jN_;h2xY5-DQDRXL-fW;*(A2Q(}0+vw@VEzZ>;v#~;zt}Q#k
zQKcG8?<wqu3*k4Koz%W7+GuX_w;ejK`PQ|*g$3QBWPR7kUEzzJ!fOogXtb)$^G!(&
zG^N5%8xQNtaFSNqjI39Boi|z6&$~v>_xo6pAvA4y(ztEx-4o&i=kiSv9@y1X^fFjk
zydRx95RJS}asSxJrksfZ{yen0JHJB>@gwH2vL#dc&VOhSz`gAl!isfxZN2oi`mp6Z
z|4FSMmR5~+W0%(h<AdK3o?818cVV_7-HRbpIC&`Ljdzr$Bv$E}L4N@cGj8fTN_-v0
zBmuI8(kuaY&OS?`Q?MX*fDizr>Pga|`OJuh!sxi`I=7b;+va1&SE$uD0p8r_718XL
zDW6b&Dv%Z<BlEOF5;7_SI^!<mQ%?5I0iCjHRc5+GesYl^YI+t+wfo6A%H)12?)t|`
z&GTNXSL5-g&%muzAf&WZuB=&himZ>D#T*|++dVQU!D*^^G-5l<lH0my%0RZNA`-%4
z-3JHWQFp*)m}lJaek|qFV`gTpYt|6U#0TT5l_fLanjajE80*lg;i^`oW@Kj@k2Jyn
zw{i-=gn@?;5p>l}c~%2L1SxS_A!A+qlhR9hdvLqP?l$=woQkk-{OXanYzgrUabF{1
zv~GeGV)hVz*gmPxEXKK4%ub3S1MWWLDo_iml%AZ(#M%zQcs?HI<e!Mc%+ru9-<@Rq
zQSddlq?-S`wI+u&BpH|rlx?;OrYKVTxuqwv+9EtFkB&9Ne}4SFgIQqA8@P~ov^|g!
zD)UyR&N?yura)6o&;{1Xd-R(%K=nRkPG6p?T6T&r{4zGbq9<!ARe*A$#Y9h7&Xj2k
z+pagJ?Cw&2iB$&O8t49@z%Kfl<nZ|%J@J6lhWEIu)*$f!L48`?l*?)^w(1YJ66ruc
zhOeN%om!EmgoM*_!7Qe~g9@9d(;msgQRfRFM&{~C#NPr1mQ>g=a40FscA**zUEgz4
zTDalPIa4W4w_<E(sHGXB(4L^D4PZ$D%9I%?<luYy{1%O5i&wtTJ*vu<Hbl(M%{!Qf
zm4N&cEf$SR9oe;XQm{+F?MKLm&hC104yPlM_&&49E0<DDsSz}%6NFF(CmgG(GGXwD
zBqC7$!-6~H3^h3wO&+z$5dy6e8K~V;W8t%8;bX8Pp^czChdVq$xiw{!X3U88Sql{r
zZ;v?8E&A~LU1Mn}z3+360xIK<%__nRp-!|)El1DnC@9~XJs?ezr)hclm(}u|7-eEm
zm>oPdlyc(B0;SF!cbUc0QfHmlW@J{UPknDYJ(<W^_?@<>$sd;Awd$hJRU8!v%S|Vy
zPq*1CH)9?yuMcS#Pn~yJz6Vi8-@yJ@P`_t_E-p%_*WseYKJ{tk66KJC`u9#DG=xUn
z`X%TSKFWOfKx;`=mccfB^!$i+TF&_1C;v;H*>8q`KTrb0E)x6p5L3>FHb45@5vNf+
zt#{2HhVZi|sgAc?_?PlE%Zs$P8@86MOQ!u(GRhQ1mz$J%wepk)D8`E*cGCLOYQR9g
z<Oc@a;e8n?!gh9Jf_yC|>V3kB&u<yoEp3-P99Aj~dO1k4NZ!snG!RDW*`(b_1=fv2
zn<_I$SM=0C$kAm^tN)MgZZr?k>m7wQ);Hm?Uw7r!SkiDI5-9NSNul#Qf5LL|b+88M
z@$NcsX&JG9@QaK{zYC3O(FB1B3t-Be;}V~;K9mLPLPJ;IAiSDL8y()B9;&}Tn712<
zHOkHY?WF<Wa&2k9IGj<Kd^dTAVE=Y+F~d2YP=hix4r$=l>8|}!@%Pe&-J;D~2B(aJ
z78e;&cLVj;#2CL9*-&9?+G4KQZat}1@e26f!)O^)y*nd9GT%$E-+!C241PdjC;*ah
zq7>m0MVRzh3ibaY4~zL7&oSR;5ul#LBf<8K9<@ay9BhSVRgg?OWiH&395Mw=Aeboq
z8vg=M7)pU%ZfFZCZS=+?NbQyDp1kv`aWqKF^}Kq2;kQkj3%4j<)-wvEp8pJa?MU;4
za_P)sYIj)G%2MTM?1<5EfK>8*D5K^qw3|@MisFHZF2TcD6J|}+m8a2hut27by(d?h
ztBO7%sSq^(D{fMK<5KxoAv3L9O}W*^b=B-MtfYNwGc2JD>LhY0X{=PMq#;uxvfOum
zY{GI*;O*I^Y?kEjDC-vT_hLuDp;AE$9LDO?SAD+G^Fd;4Wt8=+1&?!>ckNJ6$E&;Y
z!l5<*xP0`rx}Y)?EQ=eYnaD!I=d_H@cb>uaJdN=jTou?7Let@XH?Q+<x&T?)()~Fy
zdZSg&`O$l)H)G9{ia$~+)+rn@vwn4e4KFPt`$LL&9upFz@}`<vGbs{^5_XO*K0b#|
zZ@2LUR(XF?@OSPuU9nY$00OT9SJcd~TkQd<WczpiMv^4Q`~=H_t^r(ZwsU%o)at+7
zi?)2bx(T{&uvH61pb2Jfo{-G(bL}oEd91;3h+Tp*&GuHs`z~3A_2^a)m*D8+^C(7?
zhYJ$J>_18)zu%`X;>)#oG)1N(DlkG69bP+SUAs*cdZ(X8tbRUt(7H*;Ll_Ro_xJiq
zDlNRtkwgIxb5!eF8#N|uwc}O&ca*Qo0WW9zo_|90U3yqdIRy+oMo4t(+&^b1<`}*4
zb$$mT8r^sAMO>4r@lg807W^B2UBw~_B+zvP!yGK^5@Q6JiH)%XK@f&8g|XF0E0W{p
z?F#+RoHEzzZ<*r(52svj_f>EZOnv_g>9&i;h4wtIxss$ik*2t;s@4ShDdEfgZTc<T
zm^9Y`n?F<-eo>-N+oD@WZ~YF5vSuwiiA*ML=e6bWQyX`08!Go0Z?EgM7USy7%=nhv
z8-0r94gt9!*{K7BoRafrTBHp?(CXNholJ1OU9CtO*xxE4CqkUH^&Vv3c3>Xb|3%>W
zSOf_P3Ghs7Z<d+6J~8t8nRYrxYs~%_O)i^KRRpQGNLjlSfS(q*KpA7xJaO8@02%7`
z#TrL-)PLemoZM{T(7Y^Erg9Blh|Ej!!OcxY)2E9t(}N_afW3kqLcEA?@NHzKi?FQ}
zm*K2x{d<(LBXCm>5*7EXJbB>O2NHwL+vUMdIo1_}lQ>`Q=eAWgPlP0_TI6w&iDNh$
z%ZK*J6mgiBX_MS^SjX};0<8e5YQGp8{hI4(Y4Pk#%z!3bW>6k5ArKw$1V;%ov}lUR
ze^*Q7?`F?a%TrYD_RO)<4-R;On8ZYSxbdi^NM$_~`*GgsaEDK$hm?#Vz40fSb3e0v
z*Jw4?+0f8GUPhe-kf#|rBZDe`nut$q^={H>H#NrP=ZgnZ6f90PGkm9aP1pY+d5~uZ
zs;IV;tI#abwr1~~=+Z*+gJ*ycsaE;;9wk({oKFDP{O8#wmPI4ksY4mWkSS+b)5G<I
zC``j<Dm0h+|Mh3jnV8$MNJ_PE$7RWag<2L=m!0=dPr*IKEcKb_);GiihSgb5XFloP
z?aF5pm<fAo2D&>72}%5Atu42Y%R?R>NsgdoUFn>)CuRa8$4`r{AE)a9J=J7zhAhrO
zvCf3c53Vz9BWxYv+et~nHG2nV%Y+@;5c!LQ&No7jwoP}Wr>F4eKpvYprz5+jNz2qS
zp5d7z^a5)~8BI(!d^lN@L1JDB!hQLSXq)PrA8x<2cz)3rWMG<x%};8;5}9@Xg^LTg
zflZB~39S?)I1cZCh@U-jp!G;2Vl{YY52A)IrOYOv0D@^nh#^4}Sad3^ljnFYec>U*
z^}cdbDbc&k(V&PO$5?%lu-)c;*~y=C0Nqdi?Xl#In5o}w7Wl2;g;FfZg}Kcwu{`l-
zjj<$k)0^|qWRTEq$7c;MY`g{gWkS8iRoY|ngZz`Nm-}TGD>aM^rhy`+#)Bz_GbSb`
z3yn%p(7#ZdD%uR$YKwJ|Zi4Q{4)lbnzpS|T+3T-ly!LC|`@P=%19Rtp)qR@J6BYJ{
z@6O9RZ}s@i6XX1z9nRZnG@{aIo>Lsjo+<Q|@?n?}Dko&omC!C15hM=~{0oCJ6lz_b
zG#l-fyT8t_gxf7q>qlX9>=`I3tJGPXVKk~hF@u>{ThjSDPI-@~r@RFx^HS=`d-4Ll
z1(G;418Mz=`*voy2V%25C$nAZ=eRRr-?ppX2#)3|wzJQ?`7#t^*G@&=(YEeK^sI%J
zBZ=4jxYtC{=k|bct#?8H-%x4cGiYo>1nt)IYTCpy&)+?fHx7NQyQ(-1eZ~yzi4scu
z5(~@+G@E(5jZi%Y3tMBfRoi!PI3qi&pyapM;l?)MF6H;eqJ0J8O{O~0Trh2WxQ{Ig
zJcA;HJL2g=+EPN$>77lYMl{)e5fjqU?~|8$Zd2xqxRpDtM`xvVugxZaa-c1K%tvDW
z%T}XLpCsE1z?!i2)J|1$ID@H39p3_lR8|_cVWwJ`I~k;jPxm<|5{7=K{%H5_H*RRk
zxAu%)W%EIg8%JyN$e{zuuvmgY520qmK%!n0@VhDQ!IvNvs0M~%j8r*1F%C~(K1h-0
zAF1CuCHPrLYSm346$%6sZVJjZI*5iTW$|))T~gmED}$;~74s)KDbToMypu{-1q!y`
zrpqd&Wd)JE3Jm0kbFqNBR@ToiL$jC?ju(h~SLNi7G%{a8FW@nt-JV?1PS5*XrPWd*
z>oI41FmjAZ{qg6=IWNqtW7G5V1Lx<2L1$nA{4a6Y-<-;1dejxie-tp~!sM0EeR#Lx
z`P^5%J=`@Pl>|ia9L?caR9d<c_745~on2DAb&=42;DQ<t$AHdEbvjxOP-D!~|4~>n
zE-i~%Awfzf9uOc(!sk$^U$OD<X)m`kDvaDedPsd*-+PhfSNpB5GU?)YSq9N*z&-5P
zq8Dq6Ek0N~S)S)sxGoKKd^yHyfiN^Ak~?^oBa-UUDW`}#oKJ46JA!Y`or$Zi?(Z0z
zuTlSHp9J6!m(Sg0ADdr7>)M68?kGOP$i~m|N(*!A(C#U?k-`yDQ9+gR8*G&mHW=DS
zR#+k|Y+f6L(5)Y<%L_Y8ziu)~e@XAiJjzAeOYpO%f@`l=K20sgoBGEe|8ru-ey8l!
z4S!D9GOJg8sSl1<9k`WpvK)h>K@Jz=<B$#v>P;g)`5hg@e+|X`P9(=`h2;s9+0XYQ
zH(Nje2d2`RFZC-z<@97)uVP|iC7#Vcv(_eJ=P!k5H!5vBG#8IHD!~i424ogWU$nCg
zkH7#C_HycCa-T1R4h~MOec1c)ic82f-aWIfKWELJo=!C5W6ii`=2Z#}OF-q0L(&8k
zMLT8f<1cmprq|KMiyr21RIXazHa#AaM4w#V{RG^ZRsIhOE3QXJ2Zx8-lH`wCXtx*_
zxvwL+&d0rIkS&_}^hi+~4=oIW(~jVP$6<>E#$Nny^+#~RG_vW{tl35S74c*TX776N
z73V8ETLBB95mW_BvIIwFMXC)g`xc0ifOm3BKZXq6RZ~x`V&(5^&}s8cYU6h)-xI}Y
zU&F|_WNdJ6+~y{@vh!3BKP`3Pw-?v=?a8j5%4ektr+*bq6tos0&N1uHCjsLVgM@sm
zkKS|k3%vwy7cTF5NB3M0pb5^=gU8$H-9p2a@atCRt3X(+a{ufOoAEE2JShnN`MrtC
zx2S>r8uVfDz%n2MOL8n>BK9<InE+A`j){K2NsMB2%yxq{Guz)nWH~lg6VjEQ?rvd>
zlWgh~<ysr|Kbwmv3cHU{zdyZ^j#b~;v<gAY3a|*fBP1&z^kli(kocnM;CEa$P=R$h
zXpYf*Ul+)ZSxYs_RCHWQoBf4Ok=AmA`NUc+1iX;08T|0TsS4br9V&8>oh+!TA|G%c
zEV^Q5Nn=P!t5F)HF))|N?Z^j{E<-^qAXa8IZB`1j4}3}Hz9&dOQ;26}V2yf$3P+gn
zh<9p{3n;pnG?2R)m&e<&$QXmWBtyWY@-hWWs2R3*1~DhfTWe5xhU7UZ-GR?~FEFsb
zj9Nd6UT0(qM1V;Ux{KB{OcQ%eRhcM7qe--y{LVG>ybjYzMjA)IWA@^?F;))7EXgNd
z_*vqZ>)<<hzZ^Y0+;xQM^G1<-T9W0vLd5Fo*@vzpEEazIr$ZNYR%}yq*8EXWuikyc
z00#cM6@&YZxNYtF;AVd5Or(sqDE7~Z_Nn=QVDa%S#KJ6r3B=~gCwnI34=TRrlAV%_
z8Pov3wyt;8d-eYHb_|Q<&hQphh$UY}Q;i}#M8*<X7pu{ldl<E-HEjF5G1BoE;kzW2
z?P_C^e1!yI$h%K8<Hry4lLZ9Pi`(2H!*ZAVK_Gq_K5}<_9~2vmyTSMQT@e=#ay%7i
z#ni3bNv-rI>i_Hujn>p?8yn|5;CAra{!fAbBy`ADfU$Xp4)@IdALT7u%i-_{5jj!*
z*Ybq1J<sW^wZ9iQVcNPmg}Xc%8I0RLHE<PP7Xt3<=-7jD;!WJh6P}#L0|Hq@M@Pol
zMl_djhb(>Xf7-;C@BarrLBYP5=`?n;%BYN(Nq2Lkq!}3=r%_zw>4V3JnkEVZ!%&#E
zg(W3YD@;RRSt?fYGf8sU6v>nzX(zEAlSIlUVOt<HQPT%6pjE427zWK+6TA>BX=0l;
zh5%{V7><Jw5H;$AEuSa~38RpT=MhE`ZZd=AIOs5<RVgtyH^=@3pX1F}zH^L}lAEu;
zn%i%^k#s7#QUnpN*YUxm6ONOo<p!@5j1d<i%+D`StyP(tn&KPx-OJTiT*;*u?j{NY
zp5A}pMTKI$Q1N#k{Wu`_y+8N^cJ4l(|M^#cL8VgSb+3OTsbq><ZhSq(Qi(tQi@%^)
zEWfBEfYss$1=t)ta+tI!QGvi0s_hgi$O}Oaf=01~O1Z4vu!);*dn4!U-h~VU8s#bt
zze$*0#69#BqYFpbk+WE6yp#fxR|b}4ap&!~aP{Sv#Xa8<E9+Dzx)#p^Vs{Ipy}lX-
zBf}#kU6+N$1$ON`hg>elHy?b6V!6tTx$h&TnVB!}p<nu!{BFI$8}58F7hHTPfBS#`
z2lMlDTy@p8<g+=hx$*|)78dxE|NIwJ%C$2dYk@$CSST>H!1n8yhCoRZ*EC6mAq^GM
zYl4$<iPCAha~ZC<>?$t3^a{LIgHL_p<D?uHuTtgYvyTxjFVggV_SHhdT!!v^mJ2gU
z9z8J4a`ojk4nHGOmr6pl1lo)fkS!yPA#6h3jKf??lT2!Ey7@e=ym~i|>k>o(Q5a&F
zvYq$UnF!`;MIwUu@Y^n5Y)_m{g@KR;NEh495k-Q)575dZ2sC~Dc?Jgh(OUD+1K*(B
z@Okp-J#1OOj?HV=Fg89+zPB60Fj!ca<?w+6)ayRSr)McOY7C4F)8Cb0U|^WKH-}>j
zf>2^v2@+Y0P-$#8!}e`EX*L=$sZ$Cm4WzJEix!NwuO6%6Jt^A>mO6?%WkiZ%xk{l>
zW@614Jw08F3=U%_5>%RH2BZKPFOrec)a#XYSTbHpq!gHDJ2b82N~<z0K}LX*C>5K!
z3cwJr9I5N>>f(Lx{%LNz`6gV)A&kPdVyh9-qEejWiATOlR|Xaq=P{(gh~v<xRdKA?
zmS%i>h<ddO2$q+N1X?3ai*zc3WfZY2f#Wz#o}5EzA1MriFd&t6$dbgXFY?F(_i*(!
zw-N*aVHn2!VB2FMrAD+j8q+lCgf^}A2~!%ehn3U@QpZ+1jcNhYv~hF&aYZc#sbre*
z@d+A@I-b{J>y|n<-Ech@UT`s?7ce@$o{9Ba@S9EQ^%}KWm2$bna-l$BX@SY<B{G>w
zCPqd;uZG3rLjVY2tW-6}-i^kJ>brJgoY&oWBcJ}qFRb=P{sDt-hg6RqJI<f}`G4a7
z``!P5BLsopB9+eY{`dVN<;pTa@BmW6>`V!-5i8wVud4Lum)@c~i6CPbcq$^2lJ(ox
zv36{n!O{{V11?XUI)Sew`Hfpi<T5x`f?#=yBhNmC=lN{E@)oi^y+r$G*tsc9(y}m2
z1II~YIX2Z=1u#juE(TzlI_?$|34{s|s)lZvpj~t_g^{!{k}k@!FhUQ<G-<X1P98f3
zgjkNiF&!K?ft}8f&g3aCEmJNop{2le9GrZbY%YgwJ8=ariX&51?mg^txE9gnMDPkW
zlyF^_J8!>*8?Lz$%Q9Chx;v2r+P;A83SJ>}JCI<kbiYwb(QG!@`|Mt-jXIN4rzjO;
zVmjE@&!rb#K%_&SJ#gej5IarN3;5fQe3WLr&L8~YA92$yuj9Y{mp|jN#~vh~%Co#s
z=FqXj1g%!vH}1?{LBz2Dv{XnfsLW3jEYIL5O;eg=q6kyOsWJo7z)afIY=bo$*K_%$
zS90#I-Go7l!onQO^&-hejpTt7m<uO3baWmS!9-V*$#V1ilA`cJU|AMF_J-TI?yAc%
zjJN|*Oy0$cudHF1?ftHucGOO#g4oI{m&r0cGsE8f&$4aPR&KubYVNuJK^6+jFX}xh
z4M&bm@(UmQAg3lLdH2u0k4rDVf=~SY-*ETWzJP_v)QLHco;XU_@>YrzUnoR7Lg)gf
zjumEUnWXc1YPANA=b=SJ)|Ctlbdw((W@>Snox68)`yFo}j6D9~(|^ytefxO-&wY@O
z{^eg_&lR{~!w}_qiSp<$^@Uk#g(7{a1UGLP=gZF?r`UY?`U#&AsY|9^%B2cIxCm(=
zrG*p@LN$?EV><y?UAvRZuRNDj+C>NxMNCQzI*k6iD@dKbokAKbxfwzTBob{%wx4?`
z&+UDTrG**F)h3CgX3TZz>gpns$*^PBP9iJ@1_tQKrE!e7$7>jdwCYvnW@iXBtl7Ak
zndJ(05R!Bf9N709<0FG)(+QGENv#o}wIZ2JGq<orSJxPWgCi?>YTqfe*$Kn0<dSsA
z9sm-g##69&-#(5{&a&^oezt9yAel(wBphnX8b6dc?Z~RqP^nhiNv0h$=C+BhkZ3Jd
zTdV0f<gJv#)bVpWq~MjnrVZ=)(9gY}O&d1E#UP!KFB)MAWTZKI=xLI!Ad^kes8;D6
z=*G27TJ;)<q>I;VGCMm<Dw!ggNRrMa7+bT3TD?wBz6;BMXP-OF!s0x+Oa|LZP%1Uz
z6O)dxESvITg@+%$hYK&b5+TKkJ&s|BSOu1=YC^20#;RQ$Hi#D=-Ok+<N}<t&p--*6
zfML6sZjLAn5W*mn$+2$TMjEvmrj!^4grQI5w<s4D7#rU}SGo(;8^zVJVSpcaG@C7!
zmgYEl;t;b7OZ0YSS2ns1w2opjCrvO`ZD~5=d(rvlaq`qD9)04Kt(tnNh$6)&KlKk>
zaQ^wc^X+e=O;hRX9pzVk`M3C!Km8q^c=8bvi3HPA3pDB>s@)$k{$0?j-F*X%J_8Gh
z4x}^?QAA{!oU?5`L*t{&>^;bByEoAA=9pV*qAi=;#AYnV#WihOOLL5k57X7#Lv-$T
zwCgf`azFXJOE%e$BOMR{%jE*qa)Iuiemt)c8+}I!gorRCXl=!2w*ngV8jWTRtLBo;
z=RiuNZPP3lskb5)7d?8r5+rjbk=8Wp9;M|v3r2x-F3rgLwIs4>Fbxm}ffrCI*O-}_
z=je2ir)NuO4VSG+($|&X3y;md?Bfa0)6>me@Az>pK7ThtNB_^kLawMaoso*Iq2l{3
z&b`(^7=<hqmheI!C*d+NK89rqx_f$%rb{v*xbotQ2wMUBjvjvz<X+SABR=!#e_(EU
zhTr+UKj5}o-oVE`{=c~At6yUF*aDh3lxQ20@P(LOx6#??O~6#K+Mlu{mg}N0F`F$c
z4NdJcny_#(8AK*S-nO~&vMaga`kM$tpL_29D$4U{l*^=#AL7))6y;i-gR>Qy*)$u*
z@;HlC4xU;(W77F@!?rBm^2XP3-8ENWTh_|HrdAvu<ceC0)@Y2=RN1)0&+%i&SS%Dc
zbmTCPe(Q0PuFK~28@TzptGVa?2bh_EF`qMzTFoyO`R#xAo9un^DL(vf{w+6Le;c3u
z+{f7a)E;Jz&ZD&^Wm_mSP7&~<@BKm&NTNtcND7D0c1d*QsMYG!$`u^1NzRgxu^HXC
zo|NNo<k%#aUwIYp_^G>41bpW2Kf)7_KFY-xUBQt(Pq29KIrLJUe_F0%3=J@H^+pbs
zi?~*nTu|at7tY%~!k;}e!*uQC3BR8asjC(>T8BhY7>6N*i6LE7=wh2D7hJH8Yp%P9
zbUKY`NDvxB87LjDB+s8Fz^z(xoSshytVm6)5<4vgBjX#n^7<d+Qy>375EU?$j!BW@
zkk7~F#}Ou@!z1{$3R;CAqSznJZ=#}zp58wC`}?`&#_On7DwIkkk}{;Px0||GCzo^Z
zL!U4Vuq~J6#gHp5+lAvK+KO2xv?rmhGAI?HiB`K}@cj2K+c$w@XSwBub5Id{{tI7X
zVPToBt}b@%-pQT^AEw@_W1GgxM%QfAX|<Xtr3k}dH5pY%w7GiyIycj5%e7aA%$cUi
zP1jw^&-{zK=<e!zUQR`vHmWg^EEi@ub>ca4X~E2?Mf!T<+{!|sfMFOcE%`_(*|>2d
zK@i}30gmHPUM^55Rah#ODHa!T9gBKB{&}t1!jCkGghWRn)vAY`bO?eHg~BO@hS#^d
zldM)%pYDMuh&LI-G**5tgao-7?i6uRHbKy$TAU->YhqX_qA<WTEV}YtY}mLJ&ucL?
zeVpSbj<SB;Mj{n5H7OY$T^n~$k#RQ<TL?_kCXq;DNQbZA{du-+Udv!_cU(}@Cd4b%
zy5jlsr!%&Ol-zp5^~}yMFgZ2#ij3{cg+`;vAN=7TF*-8B%{Sc`FE9oS4UhBTU-@nR
z`hWf>9(?HQB$GCavt??v21*f!$Yop<5QHIq5a9coFl2>z!*?fnLI{M`xI*J39JcJ(
zN?#_yiN_9b!ElCr(&xlvfpSBW9nO&(U591cNTso=MSA;tX{Nfcdip^}P+g)Y-H+wi
z_+G%$(mb<^GZ>~x7-&+76ha9Kr3%%$MJu#WfsbWEF01Ivx;Tc3@<XbH5{YCQ*Rh$c
zm04J7GFyxg)Ug}`H=DuE+PFpr!?4kYpwepK>)6**2!kM0_+CUdljOYZ87|zh21$U@
z9u*bBD?sYjtzE<W-t#WDY~Iu^zzR@d@VqvN&vP*N4r{rABv272Po3h>(IX^M2@-As
zbe7(pUb?$_aqKh;^RrYcb)+fdFkV|5i;33@-}yZS_domyZ+Xj``M`UBjvFtz0^jqg
zHTP1QUcihL4v=sxYJS*O#lCZZu2a&NXm9kcun7zcJDb4^!Z=i3t&$@m5{ikw977|6
zMABleP~vqr-pK3U@J525#Xo%dV?6iVe%|%&_j30qKTdt>6g#?IY}MrG@F;E+lCVs+
z4(DhzJth|`FJ<Un5^USz$KP~2*Ij)DW_!KttfkOV*9|PoYOAslLKrKZaAH-q&e3DX
zXa+u!is<gj5=0TwavAI&<mT(H=9~9D#PO+_7p2Z?pyo$>{L`Q2fkz(UeeZc6w_I@@
zC{3XEGka_b=>^z=_Q6QwhaFSncGrM51dtPgv?0l9kOp{KVxS2M1?<4bu{0CCS(3v!
zdh$8cLX{H-j&uF3H}ck>x(my)_{Nt$%~KED&Bo2UdBa=YK@`?mIJlpEU%8ux6x{gc
zH*@9fw=y(3!J+#fW(@%oDa|%3L4PK}Z#;O6Q#J3U4IXD2eoYzF%Mz^vq~Rh&gv3Um
z7|a-4e$jTa83!p8Xd^zI8X-{-_b7~=9HMsqU7Sa^LhNL_YF4hE1mY_$;<i%6P3teb
z>^jhz2fq9XlvkoBpT`y!j*%eYrjXJhv?FA;i)N)vxiX6kBfM4<Q%YnQkuW8xt}I>o
z96h}~baiK_H)}Xfl1$Q}TCQQ}5Wm%8Vq!hF-Fhdc<q|3X+csU=A){DpSlN_CJZ(A+
zfu%4Ih9r?p5k&Dx@8XNjr(9a5Q7f>0({}bh`7AZBK`s>{l^~KDKWxyb7cngRc{U}&
z=qy4i4teU>3ocfCI^8WaFK6^MpUd&n@BGP?aBJr#k6V_tEoj=6x|1glap1sydb_i%
z8S5tqeQLEDz8}!l(?cqqiF4Ras|Ib>Yc!jZ;&PdCu}rNJusGiUVX?Se!w)3&08C3U
zJeVNYl|V?%)TtRBe&`|I^41MV!;Hz>X(3m`Fyc-d9gBtb-0HL&qkXtIo!BW9Ub9NI
zFiW;~3@Htiiq*lszJAuN-;D2jEH2J+YHEs!i7|@Bd4eEfWNaL1Sd>atA{CO$<tP@H
zdH9hBkfwvirP&Ng*aneS&_3bnZ(C?O=%}E!ZE@l5bC{l;d65<n-ydiMQ`58j%CG(!
zrfG7+jW=L9Nf46${sBJl^S{jCzyM$Q+ULpTGL)9<G#epTTyYhZa+#&2Me4OWwQ7}S
zvqiJzQz+M|*S}LIf_9pN&`8t3h!ow~1nbw1W4BtAr>41RXp~*UlCV{#R<96g$gf#T
zvab)vaiCtoY1Q!n)jtZU3}X5iIRg?2mwKbl-25!F^NTe7kg>5oa+y3ciw$NAJru+?
z%<LvwI*FUm7+#r~@(C8Ii%fJE=*z~}q}lSYZ3n}32~`;-CDmp~s#F6h$#t0|5*BVM
zjh#s0rjnSp6@TPTl{6Y6wUTRxuc?(Pc+Ho$`c(+Qj_up{@GpIk!NGon5ClO$;On-R
zXq>}m7zU<gwyVg+^A4FxDTEeye!zjl2RJr4iD8;Jjzc<?#`BtJ4I4IY043SK{~5ma
z&HE^qinyi$8f>9yMG-HwC%O=^7h<F}g+hVP{olXh{nww%RsC7q^SAMYYqD^1o+#4z
zE1e-?>K2zytWGdfNV){1gdi<q!}%sArfs2`O^j9xMZl<O;N+7e`nwqF8)15Cno^_4
zdw=$4Isb~w@w_UZ{@CAe_~1cqxbbyN?%&Jad+#Ax4{%1hsV7{FkwKWBXKs3tv4qWA
zw~q0T>I4T1FMksK_d+h0<y}Ad<D9p1`-%$N!Rb~tv<-UW=#*&x-)Os$X%-8MU<y)B
zigY?bPfr)gL>eI-v=AhnBx}Y;IXN>&q~d${Oe$47W8DZf$4{N&um0>0`QR-Vb9K%k
zyK56qr!7t%nZmEMXsYnETt$rmh_<IIN|BNVc>y+H2!W?HLWRg6z)jc;jP@`*)I*P(
zAkiuyj~%5Inq2y}H?iZ6HzE=aCl5T$k!PM_{nlOli}(Ku8@KO5xfWM1%yaDid+6D;
zk=>VEh?~r?YugUK_1Vw!=#feM;eOV1HQ7AU#~)t1k$?5g!%S9R8nHVgQg_a-b=)(%
zgz`g#9S^X9A}KV3!4hJA3UmDghNI#Hboe&aNS}7|uf|q7L}&%yL1+@h;WgWIxcurH
z$Rz}er}op`-334p_*elXGAaC^O6b)HgE+x4^qU9`QVXKcCu-Gkj1*d$7>NWyGDWdi
zAPOQvFG5Ao@-@vs^RD;2m%hP4l#Y-}qojJCI=12!*S0K)KP4-Nzm7q!6mjBkrBcJP
zOwy?&l~RGg58?_kDXG;IVLsmMRlCZrTC0%Bbpvrj_%ICH9SNTQ9y)|lsa1bE<7Fz8
zSUWMnFTVeM?B2C=<@Gw_CfWrpu`C|<^NfHKCyN|DxIm-svtiu;VW8>m9i-l9;kBBC
zVSwwpxNZW=azKUD8a2E|8NX5>s5J>%AqY)hZ;oU-&AvnP)M}cA1&?etO(yLyJ2S)a
zBa=uRq$QBz`3l16Q>jyOzfw`$QHzyc9W>opt|D5E60&TO%@3`pYL4SDJUoi$HHe~s
z<zj)!$y1Dtj!`PjQK>Cpm`O}CLswT9Q&Xq->R11XH52RDy5%bFzxQrz>9K43c7!2g
zmA7*}B?z@@EZrH?YOTVP&+OshM;}MOfH1<Vi+JuGJAQ&+{KXIP`@i?Qy#9_ma1tqm
z5ahGHyz?i2j`i!d@tMzljD^KHwrx3wU;p*rB?=>|)e_A{lge_LX1&3QlPCE2r$50L
z{^^TEI$n^alz3q$4I)l)5ISVYu~14=Sz2aCPns>8vUIr(99pz|7t6L0!e?N^8WQOo
zhE#+~Bbzm*musxMWjm(hP@A13lSt8M)|fmw$^60+ffjW243JFs@Wj4lmc&)0ww%xS
z<_)w29NafW-}*ixt?A8PL*@81d-pxeChs}Mdh1{ZH0qjUGD|X9#+NNl27){$QMr!N
z8s7^ELq*oIq1|do#FaD9uA+~k2pvYqP!qL#UfaBUr#HhexaO)WdEa|}np`f|R=R<9
z?r9|Ad%a_JZdn$xy$*ENcBK@N(v&N83WX|<KCy>%#$jM!7}GEaJdaE=g=JfyA{G}H
zc<Sk=St=AsC6cTg8=|i($AJU;NT`5{7t-)Ge8;Bk&;Op3Fwoo0)fZjF^=k&mO+Ch@
z-Y#-Ei$+&~Rd+eDSf{KsLW>pMMhGxW6BS07I#zBnDD1Xs7yxDz5H)=)L(<#rGT7Zm
zcXt<wq$XXcGxOA52KxJW@2~y}>$dM8@>+cH6aSlI&pyLB=U&c5mtM{6$pefJ4^uue
z$)5Q-SG@CfuD<j<{_4N~Z#GV>W2APBt(4e#-YEa!@ktI9zkdeg`CN_<y#GDy+_4p*
zBBFLR^66{Qs#>?&8gnJbQVRmDVpaCoQ5@GHo6RD$CX>mqVf_|dC&k`p_VB>B9_RSU
z;}~d6Y2Zjrs1?oUnFfEa5(kZATb#RP18=(YB6^peVazZY?aE^&8_;Dk9SCN9jnopP
zAW#A&6oL>FNzO7!qp*;1QKn^)$|UIT?IzdX#c0|kXZZ9qYlO>3X)H9rwb_2-^=x?k
z8?e(EYLya;g+;bsay^&b_y)FY-bM)c3yVDb<uB3_ik`848mh(ospEX|qaS1cvxit`
z+h`Hc)s-dfra3nUe}40BKJeu|OxIp~lI$6gx+4ddLAwa0KpAKxcGO}}Dsmo7KDvjo
zHS5W3T!Xe1Qm=-?q9|JR3ZoriJx$qKDVjKKl61nHoef$_i75@NWEVRxzJ_q)8vI73
z9m4VvVTd3iYF1gCJc^S{#EG3kU<ec2u$WnxW<KSzZrz4>u@V9mhBWI9T8##wANP_>
zqy~7$T|dXRofm)*7`nZ<@q99MdMKjftJnUm)v$z$7Y5gLvC~~h0b$^gPNlJJn;?n^
zg~sy)VI+xAL|USvh*q;jG99g)VopC1cfMXJ$Y9kc>hx><QgG=-7x2ONy^o>6!Ikf=
ze(qQiR61;{U<O($48x?Ww~M*y1>9tsHS0ED+a|S2gG#xC6bj3d$gVEX27wn*URq>s
z{$zZDk3!b3S;GbA4-qLrwN}K}KIu$`C!RV=%k!C>Tx9*ae%7xU1nGl}u*@W}dKnW7
zr4(AkM9nk|gcPg<TRLO5^565r*>P?sG-@R<+?ZJF_@I(VBp4lCL#x@K)oM{HE@K#y
z!NDQwwK}EBEE_lO;MlRlJpTBj?Amo68`f>${`<bpGkYH)n@Q5uoo94-<avjJI2%bL
z!`1gPw=l=O4?e)2r=O$I@?UW~ke3TeYmT2h#e0A5=lP{Whk5tAe};j<A*3O29Gh#d
zzKM0~H}aLQ{3FA|!wA4}9b9(^jes#lvr=PzW}eBzM@gGToNE`wmMDQ}8yRbjgJi8~
zFxKCT(>q4et7BItsRsed)woFq+jfzriD@ZDwr$0>Q=sZ7zlB#WLB5w%-w-<TQS-Aj
z8x^J}XP8|qAhe{rD^Fiv7e`O{%sbbx=EmER{W+fe)(Iw$mx!7T4j))RN=a|O#kuFN
zrM~V0j?Z*sSH4Maj|ZUuB|w-gL=m$MxNi3dSsOe*zzckmfaTaL)=9Jv9V!S3d>`NU
z30fXR3OnhdgD_ScUj|sF$&bJFExh?ncaTgZR@V@%Vml}y(6Zx`sgZ_gC(X<Dq1%je
zeO1WP(lQS{`V?1Qc@x)N{|3JG=sh$Wb@I790|SGM4v(Nz#QsA^`N+pU&e3DXNhOk;
zvwbt$Hm@h)IxH^CQY;l%KhVRmQw8d+Z%-!g)T73*O_QrGI*(hgxsdg#F1jb5U}SVK
z&evR8<icf*jiftoajFqf@u7|;(D5E*>Ilm;$Xf=Eju6-wLLe*?H(}G0&5-KJFqTb|
zP76|=hgqCQpDNL;g<QJ(9JasnU8E=0-~}!AJ^D?Giwm58#SQ%APk)f%wd?V(yqx*N
zM|gPdugPxS%#Ck)16_kdy!WRR5B%-lkZd*R8|}jqnh&2p&Tl+)ocZsA)k!{|<Ci}8
zb1MgT6@?g~AA7Ir)qL)b=`gGw?v;uK8l~Xqv7_uid=S&Jaa|Wf8t5p(ka6C3HtkX>
z7ukRK2r5zx_4P46Jjl${B#l-bTPZ5ds9k9D?W^5t{jt`XzOFpC-*h?Gop&)kvxk`I
z$&gAoBxe`dRnREUWKA}~g6~uCG-W?tHw-CBxfZQRlgXu6H?f9PI!!iZA(4dAz-;*B
z{4$$b6?$z&p<ZHsv5uQdu>STp(|zepSXPRr*CGl6wr=0e;MjU5w(ca1LK+7i=j50F
zku4bu_o|C&`5|BX^yk=r;4n+i9pv@nIWAZ~z+jJyYo#%)I9L9h9*1AQY!kos&HYqg
zWJiuOYZF##fQ&GdA`BzUFkn>p3>gN7wkb_6bM&D{Ij295Y4>A@IO*AJd$z3%jf^X3
z(5oIdqFos)g=D2UyXaKf#EPFZ45VqHq=A#_!PF7m`5H^Jlhhkkgoa9KftktUxQ4_r
zEg~hbQwa>yAn+riAmZ4uqu7qk!0;$Snnb=&v)Kf#nHcG#chKT(Km7r=o^w7O-A1>?
z9C>;ki3X+GCc&}n6B8`N^Q!tn)z00J?UrIXzEQMjt8XemX-&&lc)p5TDoITgf$z2O
zTaA?)!!%YqNwllTpZAbbU}%9>x_u+PG<@K_KSOU%_e%S*m<XwMg>D#!es%2UgOJqg
zRl+dh+H0@i(T5+vFanmAmI(bIuKWyRPdLL6IHrrT4eFID&3YNvFzM|bW$X5H86I6v
zv09~6E)s+h(=#W?rixgW=Ba0n;`;#$b4%p&DaQK75LUYy7)fkj9fvbJ3sSsBv<o6S
zxrgmgK}Q;XK2eV6q8CPlK@byr!;BX@!yuE%GCr}MMx#!v)uLFeV45b~T|HQa;G-Y?
zpOnire&Vfnv3c`0zV_9B<hgy1F*Y_#5Jc=hbP&t3=+1Y=g)dSdJE;kVz-xLuci=hh
zzV}{^9y@hf@6^`{t&0_qQl-l8{{A0u_t(G1hd=x)Tzk#6B$6o*CgWr4c-K$8pXsSd
z9(&|rYV`(g!X=qXgN``3Z!drRyT8w=<EP?kZsHJsgLu9rr9nz-#%;;y&=}^L?QGb7
zE)zXD<or=)_dH6laEw%`&PYEvP8!>_=$RPDHckA{Kr|{W&Mz@|-R(#_LA0<)_2eOD
z=cgzvl+Z%c*VjW=R}Zyvz;xpxW(GIldS#A%^B~1VO%y_eL?%5pY|U`axqX~CGE2=$
zv+nZCn0l~G+S@}a;S*|w9~cxHA!*^#Z2F|MAfL&g&~Xxftp>J@v@DdAu@K%~KP=b9
zG%X0jcI7s!sRS<psZ@%e{h6QU>MO4x;kuY+X9L$nDuihD;BA^FTI-echECq9A>&qy
z#cG{Gslo8bdLDoBF>bx-^`w#s9)9?4h6V?j7#$;05syE)hmU;hQ%uj!lS-#}{jE1~
z$pz=LR9NK5kpq;fB}~(zFF%57_wn?B{c+e-ca%v@r&DLbvAFrF%Xr(JcVaC(PX6HI
ztVt%R6spWm&vEEP6|V^E41>*%L#<aa+q9T&`FJ5%2?I0XaH>*gXrPCYp;0oKBnev(
zX-TNT@k7!<nKja*r|DB~lvr3;q@@kkUv&X%?tCY?zO`rx^OGk~mdhnKzm=PAe+#Y|
z8}%<wo#g1=Jt*7Yl3TAsn38I#$a7Eb;n8Og(^c~*Pq)~;Ho<vgqx|OOYxwXtk5Ug_
zxytu^KF^2$<u7u<dFNsnu>j+H5Zkh|GS-%5VVcH@DhnOsbWNk_Q7X51cK=aIl>+1A
z6S$5GO3~lnPd1yuaS}ZA@FRTbD_<oFLk0&1*mcfUHmsXquWRz$-n~o=_pw-NGP_hh
zt%#x>SvhUI`@8e}%ul?H%XSUYdh~8K+9rgWnFGf;bhN<yqM}C&Mr?~juSqrA%l>6S
zC{02`5L%K-%cD^%V--s@XC^5J7NL^#Wo)ilKhD;XJk?5pBd6wAT&~bR-pA<8Zz8?(
zG9oRhMJ<e0g`rBB@YG2L&OM*d^H_T7K`LMS6z0+leT_1;;tYq67nv+Vw`=m#S8ZWU
zUm9&1n5Ki&1_pwbia1ejB8AOWV}pEsa*kGT=EUyII!XXin?%aTtj#eV%`<9ino&Dn
zE>z4twU3^WLB_7WoG6`H$#L?-xHoF2>9Z6DoqVF0D6Fb+9d)jwLP;S&8VJKgOC$ao
z3nM*9t+qfb5>zS`TD2;L*#(lBG+o0(1g(H#vCLAfOkrt>xw#p_&?lQoQY;j(YzNDB
zD3unex2o8Q1Q%a=HEXw?$A(=O5h;yOptVGy+7r30lC{HhreVgRRaR{(AnpMP#Hx`}
zgt|>i+dTy1c3z#3Wk?h${0RI&W9j&LO6fQx?6+3BB*cDyu_0;9s04Z?kFHIARlD=V
z%TO<2nUY8aEB;d}A6&#UBDOGzD@j%4Q!3BlID)a!F@jyY$fj&ki3IzPKSLBY&?x$Q
z`*E!dm12ctA|UcZ#zrPsvt|=R!)r-6DHfI%$!5CHmd&2M&r&Hj80s5fpudk)(&n+p
z574NGgi(`S7mi?<PRGQLjvgo4V<tt%hM}F4C?!G*v}})KC;3pTc21&LA+KnKA>)SY
zQ51)FqBgP1=et?EZZox7mASbYip3JOTAe4JdX#Fd!6g@7NuyR^ZsstK9WgpIgkf1U
zTTN=U8c#pFm!ZBv#zx0tRhf8^Tq+cJ<dKKD_ko94TJ9`vuN^YCirAjV^FtnZ=n?k3
z^PSvy!*zV%=YF1DJ9m@LW-+A6_?mT$jE*xqGtHxqKE!j+KFQSd6vgEw(tQaAtudMn
zAE`rRt4XWY#FCm}6ULl8QSTV^oySZuKp!r0=-3jYBVCMK{}wiOIm|!%7{!xM(4%YE
z`3(8t5iBY24Dgp0s4G~rVGD-oFu(s97N?I<ZZuISvY8|!Bf|u~U~V~0xo4Cp)8L7H
z58)&<BU`&^HUbP&(KnFi&O5K8P?)DthGWmxXssDTt-XlCsY#N~0$SJ{U9K^fG1-{0
zQMCfq(4k(gGB_}RVJe!X5{7AFB@@`0OnhLH0@t)?hZmSUd=Oj4xveipME*m+@BuEl
z_`-O7ZWsC(W?aA`m4%9;wuOti5}oK&l3SLAX;>%}Q*$#MJu%Iet>-f|w1)d1_y%A9
z=2yA?nwyxI7^7aDBk)^1@ZckS<nKSp;&O>xF2^0Ozm+R5zZ9tjPe1(>=}d~Q-X8p>
zhtxJ}M@CpIEHF7gkBWDql|qJmI>GHXUCSM}zmZ-CtRqGGOjur6WMOK7nb`_~C+JE`
zv?1u}OOxNemV;sk-l0=uQ)wzugLKkBo;XfURY)z)Q#@K>HnfqZWF(v7ij5--=h8G9
zWsXfRQ7Dwi^`)7(;yU`Tx*gNW5(4$|GW|WhObm@My5l@jxo&jSV)==?Iq}H-q%7!N
zJHq~b&*9k)d!N|LgP;Ez>rBb{>$};!Ce3&*kK;I8y>^6u<r)0W0|y9S0jrbou~FXt
zvp>UmyU$&54iEsQL<uvNG};IHj$=S9Y{cYFDMho@;>o?w(cL@B9dCR)-~8s6SuRX5
zIAAe8I>y+@C{Yxz0U!IsCn*%m3=a+R<~weqyStmkrCAE4C4?bx%`E3^+5~z(vxUWW
zp7tuCGX(VHGQ9IGcW~KNmyw!!f^jDxtkhXvC^B_&fo44-pEfWoO{S-dfvw|A4qVT-
z9yx&^fu~!nUEjwg-7Y7;_V;W}G{}#1<LPdM)QonS^rc{CZj$E?%~NiMY}mby4R3rW
z?z*i+BCdj_yiEGYK8lC;P+VMK%4;&R`zkp2Fiv3>GxTsR6SudYt=;`d+r=>iI%=RT
z6W30lQAh=gwH9}uT86O=th;<G@46w$=KTlxZy);vjpmt!V$X=wNi8vSgyEMM70Yar
zWgKBJqvIyVp;qWd#L4>}B{$GZ*REYi%S6!*4+^nL*2HR|#Ob8&^bqWdG8D5Fq-kTW
z<Z!eLDC}YM{7Eb)AZ5F#L5U#^nymmkk!D@*Fla$f*F3#F-4u&UaUa+~QLR+56D<nm
z0+9+C->{wan=YV#WPLjn=CxI+c14%g=(fqSQfeh<ZKaZOC5$XlD=J>QA0q84Oc{q@
z;=&QR60+4w6DUE$*JNTzBF;~XBEq1PN2l7Mx_Gl(mBAI^X(3^G+WUH`h)-fsn~1K4
z%}vu>A?TVoH>}yHGIin*QPd)A1#H>81syh+nL0^dZx@Ng2w~ts1W^!o<uDCcw`mg_
zHtk|$bOTyxmX~I!RF*g~HOJQ<c#w0?J0HVvNT)LN5A=gDsFj=SdHN`=hGJ-7BF>f5
zu}9!aWge^)M#P^*pb3do`#Nb6hgaI9PDu+L#rau6#H2e?pryG|e4<ob%%j3MF*ujY
zvv%DEq?AlePqMtcOgfpRuXh;NGMPGcl&+k^`Za4oY33Fda2yA%HRV!;J+seHt=2ea
z$2kNlWO`<X`|iJ=haUMB^=2pR@>(Qx9du6gno6a~XFvZ%?)}ESY~HYeOE0~YtFOJ5
zUFYs5-_^~)z#zB3;muro-Hpslo#OC;eas$xj>VZN{A!uX;xU#^&SF^^dZK{wtjWX`
zSMuqTeRK~EvE%Y#y1P?ES}}cciv9bRsE+rO=s%Cqu{A{Z{w?XA1nE=;ZCO}KQm9sm
z^4&Q3UX;|#?zs<7MW{ewSdxLkejLZ4T4_<Dhl1r&TQJEFNBoPQ`Ek~7Sc4&5qy{Av
zp$f^RGf0_W&yx?Z@3~pdKX;7!sf`F(C7{7V!(dA$#4+m#ZPM@~9Lr&Pa+Z8|7d`zw
zfWWUeF_1{hKud|IJyZ~4*fxCw{mh)4qSdTpByul#ly;tTPJ5#k?JYEFA131OnU)1+
ze9veVuS-%&Y}>)MY_w3EJb8+H?|+ctv5gFmjFU_zx%%qs`0|(kkw+hUhzrj@pQS>H
z{m(tk-+uJtER`w<As8JUp|D)wlb`w|`D~s?AAOK3F29I#cAUepqero9#oY8UGM2;@
zl4kqx<(dW?#z%R}Z8vf0<(H7L4b*{0$yFOPs{yrIjY_>jDkX6<5!w(8j*Q?8ujNGF
z`8@b!gOTl*v+vnSY&XZI&Fk4VHo?-}f6JDk1lHIno=RaE9&7tDn5xOC=@|~4T*iw+
zCbq9*?4~!9+_(c@f@K+0OLGji8W@N6!SERF1(#EspP}*OJrwWz3QnWU1(|@8p=AFP
z`}oAyz6D;ByEgZ-W2~2a#>TW<OkrTOy)}Pq;{eA?4gT-m$(N<YSqI}|WBmHR`ei1@
z$5#ZEIH@6u{ZHF_aHlJeG-O+;6BvetK+^JC%q}j|-8;si!-wha>f;S>_;K#P@2j}3
zW_)Z6trQPF{0M*h(T}rSt}rq(%sbxp7A`vfe4c)44?<}A2L`B=%NV9b!c4Md?K(>3
zWtu^>5=*h9U~Hg|yWa8!uD<db9KV9QG=ot!Xf|4uD<u#iy*+|J1oZay)3<Irb7SXm
z_qQtCbo=#aBo95Xhe9F1vJ7@zd?ia?|9kSX%23z;Pu_nA*_Gw@o#5x3YoB-NeInt3
z0;*8HtGl|ZJxzC$6e${UNJ^a5NIScu-O<KujAmjZX8)PkP-s>=vpd{f4W$t!nx^Sy
zd&jP>>gw_ig(~<=0BMl^d2{_a`^SBmsBTfiE>$yD+C33K0f|B)^Pcza{hsgdYkK6f
zd2BMQ+}Y;xa)(AoaO{zzJovx*2bj}~AWdWx<KMo<#sBB;5by2~p^^C_rye$Nf^Fhk
zR|vXIY#}jCo06Twa7?r%N!l$sK}5kYkfsYdV#n|Et=*7^e)E4#b#eqZQ^a+1%s=u7
z{W#`#|MlmI)sN5AzDH1J;NUUAHj7q_rOgy<5n~_-1q2%0{T3_Vd!Dl+V`S&1fgx{_
zBuK{@-aHJ%uyx^J9Z0}ZJ=48DSS=(-6Jc5yriD2K7l~3>LSY&b({{*I$`r@O*t)&O
z#TQ@a#MyHMPK1d`5G45BnAT2*%U3RQ{^9#Ma^fgnFd)$q!*EzS`5==Er%1F$g%N3F
z$-N33)Pd&eo_6iv<Z?jv9GVAIG91A&jMEs7V-L(B9OQDPH%N^pMlc8!Syv1lt*KZw
zilRg1R~p-K&y<mldnKke4|D#~kN>0^CB&gs{YSMG2b!CB2$s6t7OnaozSm(e=wnMo
z5_;IS!Q}W60xzWBZh_PoCQOV^a`NPZOw1m^b_*Dmh2L+{Y1i4@*ks_vREt$6$`$Tg
zTEwyql!_Uxj&SVA0=wJ0Bw8>zQNuFLL&L+ngM8Dhtb;7QbclDL2~-Lgeh-Q?h&WMV
zsGUnv2eUBKzF(o!A6gx5p6?H^9h>=uBcL@=6mj!r8q?A14tV8-m)P3g;cxxwFY|%N
z9${`~mR_%i7x-9)nI3X16Vo#B1E2jyo!z}Xdi?=G5OK(g{s)WcG(Y;qCo$ApeO`Y3
zEna=&Eq>?!{kzm^RmR80C>9EgRcqL`LzE<RTWwBe6CO17S(qHduxq?_tI2vdpkP`o
zjugq&#)(!fe)?xWz|yf<+MNz}H@B$P7Kjtgv){i+IdAgmj~pd$2qs1=*ttB~kQhqP
zuJ2Mfatvu&1oa()y={bPqtpO5m!n*%0VY8#3Ef5diN%Err)hR-+_@65xw3+rb0}5>
z3yULkcj`1+Ei&aA#c_jvbB8b)Cp0QV@c?aTMqPuvE9iJJQY%Ex!LnS+l@Xdd+jLqT
z@`XIPd;y_NY}Y}W3Z-KbFC^;s(bAzZI>xJS-o(nif0v|q*Wm_npgX<yo=@W|hAd93
z4iAr}Wub*;<@RkZzV!}+K}5M+!*#QV_n)Vp`Xpcb>KFObFZ?0Ljx3VR6)EJi)Z1O8
z5NvI2v9hwv1LyB&a<an7)AwPTS*9l@7#|&lB;oBhFVbpsspSkJBhU2oI3IfS5kB(B
z<BX4wBWTgR^b9kN9faS<S0SxV2Q4&qRv<DCQ*%ccojc0y=?D4Nx3@U5IKkikyPxCo
zl{<X*+b?iqWs`?bk5isGMSJ^oCggxxxkACo6ZHCAS=;2sY6Mbp_G1ro=9m5>jIlWq
zt#Know7-RO@gkkO*Jw3&=}3dRHW@9K;Pq$8cpYpVr}{({F)?BD*wi>Pl_H~AP?m#f
zq!pZoP9;ls2R^GUpV@qo*+P!h&QH$i{J;C({%@F@n;#y~4NTKO6A{0QFgVn*ziZT<
z)?k{1QONS0RTAy+=%Y_?{P-FE;=A7@iXwjMr+<M)eVuM=n^#_WZ3yDq<Z?Nlc;a!k
zcQ$$c1)uMK|NCTI!KqWHL1~gOq2J%7-)d6Lxa>rIl!Ri|<<ax!_{C3uniHpwW2rXL
z)#oX+_XxC4-}f2#K6ci?uwi6uilwC!ct_9j(lv{IH^C5sk9^`Gj-Q<2;>GK{b?FMx
zx%<eR+hqURGgwX+!*u8eJ}Y;-?6pni?mx<t|KR^XZvF^Tniz46zw#F5_rAko8MY&b
z{iMp=Q;$<Va+I5Y^aq&xx5>CJxk3@!2FsQxKOpQxbec_aqoYX6AdWngh<T^(@u|P}
zzh``D329lNHJwh6nVAKwAkUaP%hq7+a2ohmt@i}#$~lYUbCVcXZeoWTtqp7xQV0?$
z5Q!i}pc0$<@?Gw{_!1{epGM^Jsq-|95SdgcEQaidM1fAyhw4E((GbXKDUn8+hnL1x
zfQn;+I3WlFf-pcRMS?-1eG;9JwOuA=7cfi<J5!)GcO2VriNgVztu+Q=mj@p?&ylld
zC{9e#zrBpoF^1_<s*IzwM8_c{2cYa-wJ0JLYHF3bceNO1(+!2f0uG(K_gFNERB3J4
zyCK%08H8c!u-0fT@FGd%=rq<ug(w{pMKN)bAViE5F;ZA4aZfNRg+z*!IaKl>AX)K~
zz;s%tw)J-dx`RA87@853B2GdEy*B+`gKm3=FbdhPThuBeNM(@(1G>$9Lf^+WP3Go~
zbL{AOjvT!YJ6j@%e3CdIibLAHHj#=rci&kqTzG)2YvX1zNDcj7i+-oeNV&?x51b>H
z%QIe`z%(+%L@8=`cf5>DE5Ht}xzMTb`%r6_3YK7~6G_=q3$4TSV;nS3XyU-9yIV&X
z25uq8m8+K#0?wU1$Mn<!Q553)9ygbldHbEWSzF&CiWC0mPyUQjsmRIW$8j?b`^`F9
zL8sl|;#=>qvb@404?je?RHEDMaeH;0_~5YfA1t8zx2^P06c4|zFbU~z?y$bOlgb%L
z4=RyTWK4sjpZ+j;$7Rwr+1uJ7?gx}jMX{7HH8nxo%3>1myT9{gX68nD_|a30O;)kZ
z44?h^IigVT&Rf@Tk=U*z$`lc1TE`W|KAwmfJ8>Q%1dSV)5n)V}D3sLX3t2L57L^zz
zA|fEqWNm_+B^fz-99vlIZMAsr+1DwIBs})$EMuK6#f3@UzP`!i_zZXTUHYm`q8*Yj
zi?q9pW<0bZD3=T}j*WCPRLUjB#>T0QlxgqP8MHdIHtwd(iD6*6CXQp_*y#gRngur7
zU2g5h6ly=Yh#MgUhBVUP?%{z#8mYGV;1DOBRF8^)Fp9W#;|4dEZ-KPQ<nokCWh^rd
z9d5PSNCb~P_Bj9j|M1_EBm)ZB0_V=%haW|Zj!p964?oG5zx)+U%VA_<f+s)n6p7y_
zYeN$GY;Esw>hwwa&AXHe7C--szsc#dXQ&mcn1+c;dhEUa3?ub*vQ~k1r0@nFu}Tnz
z#3|&NJ$j6CZHz<;>};06(_DIUnHQgXi~AoshWzxST)A?GTkES_c=D&X^G7$yb-S4M
z07F{b-Pz^ZT1b%CJoXEp<ifB1N62a|H3P;GdhZ4|KL0Pdeen%cY~oukXP^2Ig_22f
z?Pb*Z3KE4SO^SAo@>~fs<B$Xc{C=ANlaiYQDNs?!AddOo$`+g1F&54~#@T9(fAp~d
z|M>U*kmc32pU7U%&Q1@fM=1kmm>MC>^l+=e(&>I5nEy<}#IYSL%O+A0SFT^>t+%i9
z)W?37La{)hSmaZm`f0xQwJ%}WCI%5ZTbsP_;!D(9Z2+P$<hxHljn)ZIJoW^&(Q%T<
z!?7|vc;P{!AYjn#^7_l$R5KB|Y=z^ePV?9!5AxXk7bxb7sJO-c>)&C??;?~S^rN&#
zS%5GNMyIBkTsn?xSadhmI63neZ@#g~*T4J%W8*clMaj`)6YOvAuzY8q4}9WtTwJ|I
zrngJM?X%O@>^2ljkKNBt|JVN|`KdWj8X3p5ufM|bzxwAiSMSo(CZ&`2apuF1Q>~Ve
z7hlAxuhScZ_+Cgm=#z7DWHMQ_N@%xxjLuABSQeJ!fFY@S38Uv8W%2A;r0ue|zr#QO
zfBlbKdgDzVxc>r<nC2tre~qhmU*h80Gd~uvdrzS5lb<`weJ3ki`xB3~AH0T{Btu}9
zs+iCa>llrR6=*hJx=dwymh!{*Aq^XqXcWWbeDhvSn2JNvI0-RqsxXHAaTbPQ5h+Du
zx6S@$oxS~S>Ww{G%_iM`mmrFmEF1jn$Ip?;dbmy&gr-z15Ox%Cqe~LE>DTMDo4cq`
zQ>u=T&E@b1KKqRZL82**EnqoWqA(b;Y-w(e@$PZ(&~2wh8ZV&qJtwL-Ky<^K-4Dl}
z9O%3_JOUq#m(eL(2Ldg?4-}y*a4dZocM-=CN+qdb<Xs^1UMoRr3OLZz9gMYq66=SI
zDu=&L$Ix-ChVcc8BnoNn@1b=}6np46A`E)e>j~}pJ|`CDn46g=V`nK=YfMfpP_B%#
z*VrZA-NKMiC>OD87uU^VJ2p~iYULufDTmh{j6#u-N|pWkK7~q&%E&b1Gm8k*NWofK
zr@=w&!ok@8hZsz)4m*1p8gw-G9$HFkw6@Y4oJvhEO2=&P?oeObWNLbvqEjK4&9Jh3
z1IMzMpIc;nY>NJ%&-wHBb93ePFi%dfwz0(@e*RDSPyXZIrcfv`Gi|cBx6hzIAPNI~
ze?aI5wBi=eeg6mS?C$+U_U=EJrSG-ChC-yZTL>m5$2t4Rhp4Un0PRJz_F5P^pi&T&
zN(Q5~BBc}49B=fwaczs+H|`=t#Ako;BXkEL_5D7d{M6&D-MED#6$K~9uC&up5(1?a
znZ+YiCgw;CLAZ7!Z3<Blh9$@s3fPtnmW|^^SeP7Jn#C=9*x4?LX;Q4FuG+wdZ+-I?
z$46_N%vs#NvB~;c3oGv<xkIfJ!-Rn;1+$YQbUSslp@@lbg+_>&FbT<2OJqjM#8C{9
zLTJ(k6UV}^(<YQow~MD>VY<rXk!kMk?84Q%?`sQEMuv9<&6-F#G++%I6@*BQT82oE
zumjKI?JJjA+gQhSGniU42to?^0;XXhq(QA(<@W9C)azS3`N0qJ>~nv?!rT!I%jALk
z@8`h>A7-y!r(UnKzrWA?!aN^)>XXz;ISlQQczv=?7E>#lC(qF9cA1`@X0ke(9==2f
ztrZH5Ev8FjBw<897_hywjnazBXq6+!Pm(PZQAt9uw?mfK_~;WC_|ms7^ZXB9V|unq
z9D7Vm6u9wDLR1)I^us^J*0X<1p%~B`^jO)9Nu<q(|K6|j&@cZgxEV|>A@bRK^BLM-
z{Tc-ov3P8Tok52AhaP6+>~R{ezDEDKZ<EOx<T81(xdN7L0uhmbr03J^`Q*zZgh7bq
z*mx@DyPF+qpZaAUI)4V+ad6!NiAwm-%Vqwn|MP!Kv;F?Vi4O@2K<Q|hRv{1XgNBd@
zbx+P?Sr(RMqlDt-@-pvSzJ?Sw`CKtIkts!~RN+&f`dPmCXTOha8=O3TjFTr%v37Tp
zt(|?29a~~_bd(!6Zcwg{vbc1Fv1*mX?=w~{&}gKuh0~``a{1y*TsU)rU;FjnVx&^T
z5Eh9Jh_|j&>(nV$M)Bev^;UyUr-hO*IX}nr!Xk!gq&1t(CTB_$eC)$#dHM(MuzKeX
zr_Y^0sR83tV_bao7PohNmOk~11ke9#B4M+-vCqW0lYH*~^j}k4Jb_4j;(j0R+AAc_
zf0q-p69n_Kw5yAlM^E9{ij}W?pWNmf)W*t;$_bJrCI})1ewS{yMVtgwMkjG|1%!aW
z_pviBJ!vxc$P-{XD4no+>jwF(%ab2@id#3Y(%x>6^p3MseH=eXxVrP~U%9XE3DiCM
z_*sPMbM&bXQ{UVox_Jj7(wLECNXN%2AvVF(Cc5Ku>)GcxJ6^<{JAyWBv@lZBgEUip
z*$^zHEi=QG3d=y420~lx?KHV_Yn8Rt+pOHW$<13g+1cAo!))Uav|wqn!s8E3Q!LtK
zoGi9sBW(k>XcGHv0>4ibdtk&^c?Y+cLo3DZ-7VVPF1~_uXC5X}AP9$fNcY&Z!#4o$
z25SeViQ&h%2eeY4bFZ{L)!*H7`X1n|gTUofh?=@=Q?mq^7Kgc1ah#x(MjGkgksL6G
z2ibHnX`@Bj=n(V6O%v~D-J{-9Fqir>G)WSYBpkY5<2YpC_sL{(cuo#wMMx=7VULZw
zn`CnZPM>>_xv536xg4!lkKMfnnvhOoADtwG{ut$v2}UY4+KoCpch|@ml5(+t?bu)%
z2uzCk9L1bNt7lRh8z)~I0SSgkV^NjX=|My59|CiS@gdqsdvS%pFfF7s)ANr~BymKd
zhT1tHF@#B+M4(}Gbc}+VW$o?;JG~CYVuc{^SX*0RZ*QBWrDKFqM61)~V;}n%J3Bi#
zPKNz@ovYWE`Qn$q%CG<GuTrg!6GwujBPW=inkApj;CUfWfB!jduW$Us0QrBA>+rxM
zc<6!Cl%qBUq47F>LeHn11-GbhDtWSn4A#xp_|?TJTIWu4Wk;~x_u1XrXYF>2=bn2L
zGo$J4Z(^zu5H77CJalSnX)t;0JVF*wewTRr4nm{{7RPZY<O+xEbisuz9d6yY!$*Ji
z9A+lszK4z>9ffcsUVQEq#zscDvD~Is>GJTrqGpZp>eV%xnGSRKI7Ubi21p%~&$?hj
zki;oFBfuXFNR*;jt>Bh&7*e8>d-4onni#G_*0#vl2DW7pb^6$@1g^ZV%PtLZuYF~h
zPVjEikQi3%Ns&rPTAeO$z4LZ@xXKrC>@1CT3qzV@vsnx?ooeKAHd=eAB;febV;EA<
z>2%5GO3cqKGC#k77X-ZW%8Tso?&Et2zTc-SBZvb;5|PiBsf>*9*pp8&Qkr09Y6@*R
zsSmQdiTBQHOj||bzE8WpM`L#fBuvfDF*kn%({@mCM7yz%A4I6l>pb$nDJ~x$V|;o9
z+p)33JR_qH?T*JgZ{OjeGY^y8c>~AnaQkitUmN_?|M4Gi|L1-YZ8;G55cJu9;Tw3*
zKf|EA&rVmNCYD(E`Nt`gOO*E3*n8_mqBtO~YXCt(zP|c7kDTw3%NDU6i~dfJBu=PI
zPm;;z5K<DUh&$bgW1spPoc-9RQe&VLblV-q#wVE>Tcm1?&}M)l{ww$Oe{mf?-_i`g
zw7)maDzpx&!L-I6DozZ;K%u#Q<0dzj?_gRP3@d}{*r^#Z3~4l)<a2rMf8YWB;17Qn
zKj<K(<mAcY)M_<8{>e{s=gu8oedRTtf8ht5JAXf?Pv1}00VDD7x)a>Kvx=c(5*;%+
zIfv_JF{Mic82%2?t;;OrM!=65v|IGLZG<5?cKjIQQ?m#u3Hn{yjXD~WWcf8de13}7
z!%NJ}jbS+^6ZuKP0KUBb9ER}uz=N}tC#KoBeVhD|<NV4${wI{@PJ#;XJs(zYBHntD
zylWGbD!51Q!zhm-!-&0SzR9guzC~xpW~^$lG-*>CD^kiAC=?3x{T@mwY9mvraa)2R
z40PWI!z5oD!*Q~-TAO5Thoxg@n3$et?e+$no6Gb&J#yJ;@_O#a+Sd03>WYOD15aU%
zk8<?mPq4hRjqdrF!_g^166=J71MW!DTia#znHO02Yz0vnK@6F?w4o_w=Cmn@lQfe~
z7-sr())sf~>~ZbpRo;60Ex!MQ@3FmimpDldCsL_G%D|68a=8NCP7m8~DCR0;oE*wZ
zkeI|$5~_5#TgjGDI$?Wfi)OPyv)f|s$ayN2Q4*DoQw^C4R1-BU7(dXG{0RG|kZHA)
zPH%u}_~*&c=B<Y2hQk@!!GAPE5GaPhmm+o14rQu-0<MBMNk3PtVw8?a?I>b+?S`XK
znWnrOkj7%DA=p&O&?Wr-mNC>?80I|ydSISVi9U4s3Jq}>U<iY9X%y4cl!|#&6k#l{
zvV7|nr;cwhJ~~BrcR;B!j*uo_{o0?Cvu$dnBE?b>uiwGRWSJNrXJc)HwL7<I<+Bt@
zMKak8Qko<xA(L^)Iu5&=YxLVq+~U~5l=YASP4tlc<shPV7>z59G*w(m49h{9CQ6a^
zye0`!7)T?P07y*weV<%T(P}k$_0^YYwR=4F=m#kk^Pr*8+@rCw%-fgW;+bck<;IN_
z#>U1N3_L#Y_+tzP9?yRN2fXm&%T&r`&YwF^5GcyU3bAXmy}QdZ&%ea_=GOmz1G?eT
zT2sy!`QU{IP~Gd8azGgR7>>lw>NHENQ~}e$?sn<ieiqk$iH9oFn3KmCI0d3I9GRWq
z?wu`0E2HdfZPIJEFr<UTKnXC>xVA+SMa1=O!ghl=j!9HPAzw@*8MH>nA^EJyM7GOX
zl(4?O&eG8ZzWtTgc=XBRJo?xywz<#D#0<|rw@$AYvE47Bvj+Kbf$(Fd>}`-Cy-tTH
z@~M>z1Ytnd&46iR8Wu8vc;GXT9f~6si2<SSqr(`(F%hPLiXvpJ!4PPzFc7%8-21-f
zLfnfcxmW3UPfu(J17kRj*{wIZ^v*kUdwq(<GOm*)3KMpAcGETt*G&&}v8J(KCy63z
zl`5TX8!1h)nH;k-3oI-gVQO-gC=U6=Cq79WNs=Vwo8S6V;xHhI!t~yz1wtqM%uj!o
z#f9U;16l}C6V|U#^<y$lhDKwL{jE(b+hXd-0wd#77($|=fM$J<FbpuHg;0|C`g8n^
zCqBbnqe{EmCzmUb&$}oEt1D}a)m%<3E>M5x73#6Y=l+}jlCz)uIWR4fphe_$319mj
z*;`jBjE^!lzsTb33B2++j-hzxOWz^6{08&W2D6I`=p?}pLIz%!c4r?cAx;$IGfTut
zI+k?vIpSCmj?D7#M?Q|5FXH(FzVK(C=j&hmbB-)7GC%zw4;}vu<;`onzV@y6yX(UP
z)G($5069Fs-|KcYhMMp&inxB`1~*q$aGVU0(n#sxI4-8N$Yl*G<q|h<zJnnXKK;p0
z@y<JMF)=<(sZ{0zkAINa*?C&cCenmnug&`EUG6`3p1~j@33|kSKsK8vS1R#SKl=+z
zk1bFrW>G>RlQ#aXON@95VI?&7ci6kTL9UQve0GlN=mbIt!a<iteIK;OFbzbkh_5}v
zXCD7~ZYLJs`r3<}y6+fsGgIU<SrQR5>ffcgyG}n}<Y)iUf5rI1Ni;AB1LWo!gBQNW
z-u52G{BbH59w)S1OmD!gul_#l-sFL$Dyz8xYwIz)&6tsmE(=pxW+$rTayhI_mN*RY
zLJz|>$rq{^#T+t-aixo!$w9}W-t6(@Qy=G>U-=Rnw{Ftw?$D3YEL6+P{8+pCo<N;2
z1kw-)LsCAv$izeE*!=$M7(z2F#6po4uB1PM83}gYzCm_jo{<X=5a-J{*<1>S7@8o2
zL|XSEO^}ktPM2#}uk+pKzRfc)JVy}vz&!^@s{H^1jTSM60hvsWt(`68xP@!wFfhqv
z@}LD$n;4c$&+D<XvrVhhqUZJK4t&l}9ZmD&I5?RePVk1LowzqXg~QRTdKXrcBF#Lz
zC&C~8(CUErL+Jyt{;=Cn4Ab!CLD9Uy==cbJbC*P=H7H6e)IpK9Hiib0VS}9nk^}u*
z`uUOsL~@`ZdoR2Dm#<Vac#ssIkS4S!^l&sh9CRWuC>Cn$-(AMEAe+x4w4zihlgnq=
z-Pz~0*I#39W`RPXN*pG9`RiZe?aOcT(D`$mIdvMxa!LFFN(&0v0#g%{tgWuHy}8Bh
z+qbAzOO(nrY}ch$t<mWXXm+}6+`7z>`${B3XYyfPjZ(ykLQ8=mhG3j~fvKixf?<L*
z0fmw&i!9Smf)eQs&GRBIUVMY)<r^$59_0g%Kaq~b!vNc~Xf_9FUvMP3b7z$>2&q)6
z7!rK1%TIms)3h2*-g^5TzVeN4asB2pk3R4K#bS>2&AUAFgO}K?H~IfT0o}jzD(ACQ
zM`jr>3-pyfhycSjQ|W^+C=^Q=XuKpK(Hb*KpnjLAv5jTe9LtuOKetHdiF4fC>~Xx@
z=dCxsP2`k`BZZyO^gC^CUVM|usTo+iN$B?p!?dEOSSVte76L(%#N@ITQ{@fL7cWqm
zoWyV~=8i6K>78Bdq{H<1G&|dCSi&MVS;i{l8Oui$r_VEZ=^CTCIw6XBx5r4uMJ0lK
zp@`>um{MSy>FZDnK_-`lp-C*0&*61@M12p-u}PAIC`zy$izx6Rj!4Y+bQ}Lm*Fj9h
zaCpE$G?M@fDUhW1vb&qxT)BFKL}@CO8kXgdL<!rQcX;ECmwECdpT>4G2qc~MUV2ST
zNxPF~x+RHXe0++9g+*p(j$qpkZ(e*Q&Edmmb?rKiZQ<B9mTh6$7PUf!`%au?abXE*
z4kbFcSe}Pyw8+~wUcXJVy-%PbMkXg28J|LEK^%Cr8hdy_faADnESA!UL5ITX>lBVX
z#h3o<MLZestH1m+xQ>BSsc<IWr~mrv>?Vqj{lou)2R`$Q7})qx8$IwzUVWP6`kS=<
z1Y`OL*%Rl`rb{snx%<^G(!cyXtG6|4>o&({`b<w(DCP<j@&$Tche5B)*whT!e4a!n
zNCcthG0=*UM<2q>6)-Ky#`P75LLPkhVH*28y!rYI_?@FHjXe0iL0u}>xmT$ugihr>
zq$EWe_Mft|VHk1k<_&J$zC|{d!?kntJTKiN(?q6-pF+U_?UT=CDddhIg+Z^^=YjjD
zSz0>A?Cd=A3k&r6Z3_8O+>F6jzy5jpoeoiu_Ot6ap;{_&{@g>9%GIIHImC<xIGr{n
z*CqD*blOc~oiH{%Lv3^dv?d(%Xw>&ew8C++BtwIy=(pkZ?-I{G!JW&ST)(!)gZCfD
zwKYHY@Eq}r|1&CY^V7fe4>|GZhtU{#VH4I@*!}t+vbnZJ?ZcmB`hmyjClT4^E^A-?
zeaeG6`Pw*%cBz#5Opgz^x!PrSSJSxZu~lE?_*{)zt$>p&U>T{#JM=v)r70vDuepKD
z+{eh&G+x-{D_{HrF23+AL30n&(d>6O(`NM_GlTlQfx4lSP)iWD%lHQ#qr0(B|Hd8S
zBo&;Cp^+eEw=|}Pq#v>M(raYLM<`TB5xR`TKpW=p^NRGOIt<p0eZ|ddE4=yED?Iaq
z?-9fy2#skPOpcGI%@1*$j!IO5;|LNJF*-JZFeDpW8(5~xXk{z~XCe)b6$tA4bv8CP
z==8dTaYD9OrC1)vwym_ZTEih!5vh6MU|xElgL`-L0HlnW)(5?K>Rz4CyI2=JENwsh
z{qWC>!#q9%EucC&!rfLgW$EHL?R!n`v3=<cR|3hQX6!)7Dy2jv_r{LWxHt0scV1eP
z#KTnagAudRKXeTgX`=Yd!ZB`KdxQP0WxCxAxoi%vP$HLgNs^d5Yir!RwZeU;vfR0K
zm8@;>%fIx?OizwebX_W?5gLuVL}3pSlS;9MF=z0~Yp)Qgh&X{}y~Wh*G~<&~3<e(k
zz~k=SHRet}f)*+buND#^)Vr?pyTXUXy>3`x3}?0i#6ei}0mwUWVk5C#K^O%*`sfE3
zsnrO5k1!mdq8P`q@B)vOm8%>-c7aD9c@oDpn3$NNwZBI$YcfAG%V&S)Q#2Y)*6wbx
zy0*&6rFr%nyFB~SD>Rz_-z)F__iLr4O%j+Gt5PTzF?*G?M%t2TA8Nmc0t$s9Xhjl+
zXdNS@L99R}5s~iDuiqwKd7ZJzd5+ATW9#)dm@8$u6$K31`#9MgyPI3Ay?vE#eUH)l
z3Io4O-|JH>mdIxF>BAp^OcE?J&ty4ZXX{m7`<l<O51r+$H?Od@>SNi0BgZQ|@X#Vk
zuX5$a>quo1_G9ergpZcjD2&wE?nVeJ%lOzhezylgkjZ4zUezeXvQ6wv7CDT(v4E&I
zKq^fT#Ux%pVi`n<!W<fu7-o09-*q5U_|YI^Dbr-lL`B?LU1xdu4z`n_l+R;Y4!-ZR
zcKa&NeD9mon|;PdM=@=icDv5z<}HMX=yW@5tluS3nsTYi%*-6~3riFWWtMN=U@+)X
ztCa~nkJBg4fYxMPmr}9B$mkf;<5QH1rIbM&GKW?aqHo+J<Mrruo3tAZTJ0vqQiYN6
zDM%EN@6l~F2;+!ME{BjNaiY_B2o(@-T%}r@C7-j|ytc>h{?5O`RFYqKc#-6#-y_Ni
zKKv_xj|YF@*N}!wKkA~pZTjE+0()<~Lfa`a_2EyGnOR6pdz)*leC>}gyIY)?sZz>z
zxwahf#<E~-XOkndS>`8e6mliXl`$;KCXPb11UHvQ#Rl=9Pi1Tl*C~=DT_Vq8a&nFn
zXCC6MH(ux3o8O?lACSw<zVAIZ4kB#i0V8|xS~zH7N?#3O6mj|5HE!RzLpGNun<)^-
zigvrhc&3D58W>U#MT-5MJzU#C2+hv!4j`yiN12~rV19myO1Z|xi?6aUw}cmXT)Olk
zxm*U<c5ob<T(Q9P=oAm%{~(o04Pm4p-qbd1Y?AYQl0lzFt4?Ro!O3T-jEtj_gv1}v
zX*SRrGPy!3^-2^fiHXAj(e^cFDkD7p!6m-;&5Jzq^>6Ywf9`P#x1XV@eP%!WOC0~y
zFQIK074*>?tK9mtKcL;}aOBs1o8rQ8h$6(=b-K@fotg@9s@1e_I5rTLK_-)>Rx7f4
zx5e_!7VCQktzL_{(H65)dE7#VNF@|Yk{}34gvqT}o@eRFPvRCTeC}6%o$r3(54h3T
zBy4s#dS;3^j#=&b1V3gF@p}e!h?F4`5y*0tnI|8kxxGi&>Y#_l0j2dJJEaYQg}~d~
zX5-S^l#d=mDAEV2NUNKKIOyaw@Vx=s>-+3%?eP4|&k-eI+M+<h!u%|!P8{Q{H{YUO
z@1;3zO5^8k;#lKma!gE3lO&4E*WRX($x|#;FfBpk`)qD+vT|pIPQQm~xwzRJp^|Lx
zwmDiTVq0mxT&&a4qDG|IW;*qpq!A?dy6Ns|wuWMQ1xdnCYo_ms^f_Sh4l1nf4UmP9
zgh2pkY|CbJY#hJcVNl;pwRDLhQA2Py)aapg%0fkP3jBuPF9kCOS|7MS-^=d)<x8ny
zII<dQ9}i+?hK!_ED*gV1CX>ta(BmKF``_Io^jjoRNEmt;rXZKiqO`%)YgbrWJVL2b
z<gv#ep*B8+ZMtL~7ZOF*&EgHZpd@Z4!<M&=FeT&F35YbWzVbRhJv)t=FOe^g;MVu(
z1Z`9jg0#?5q4jW#Zl>c`4p?FC=@6xH5C@~v48FsU8|l~<efak*%V2TwC_+Gg&^xSe
z5+oD~xm0&znP>+5_TTw!q+zmh>l%Od`ityrZcxbPC=?5vIe8R+&}YBV;5*O$fGCXU
z`R}j1`LB2VNS?}HYrKj${MitK#hEEe85`XiV4Ehfuu|YHC6?=8IWAf&g1|>9jgZD+
zmuKjO#9_o}afECfa{X(cXKbR(%<&lp8y%ke&R2NybH9v45$)_Uz+><BC0eZ}j+0|-
zZ2AD}ryP9@+V*g>aC%|~f6b)+rp5iIPgBirvAn#+wYT?JzTBXawHaMH&S<&J)n{Mi
z5qFQd(i+MLS>Ng~KUTtZTudqH_qy1&14ts3fJ$&|3zfvEI6_*IUVRU@T*0jt(UC$Z
z$ma4$Q$j4zGJU_l-*sS$I4~6*-aiERQN*p=H(6ibrcfwT$QRL4;Cp>mu3hB$XTHt)
z>L!b)?&HMCQ~3Q3?fre?*r(I&aQE&eQksm8j&kJ4F(xKw$>s{&y0gOa$_;$Kk89_t
z6h_EqTq>0!<x-hUCWC3&2o6HUL6Bgo1hIArxwArjZ<}7dOTXVHlPxeadn8R-2t0c2
z7Fr4l)f&>WNumTj1khgO5yTCW*T2q3pZIk)RyOeV_xbdvPEuTXhTX_#{;6N(fnWXm
z=uD0{^f5Yhw!ZcS)?WEOg)<Lv`jbC{X=Vt+9^v(ibe{hf;}SB}NkVLjxeB#%kL$Nv
z+}#Xl-xyGD_BgRHMzK;%0aF~3=mdd@n=4?uHW|~x&1LC(0o%I`9(w!<o`3ckUVipz
zn)NLPVMufE{ttqNanBU-!zm8}i9~2k5QSX1c7yeeyX5mFvbh{WN%k9CT)%diiK$r(
z!vrm9HMVIqb}&uJ_RcoVW((7>nVz0ue*OqklXL9uZqe&^DdZh&TXXizeHcQLbzQ2J
z3KNsl%uGy^&18p0wsazAHusU+cL~BC?Pi14{vHT}*}0<_ND@EYZ($e|DrJOW6Njm3
zFN}N!VV@ujQLlc3`#<)#xV1Dw&MlC?`!cQl4O-*J`RU*Idq_8n2z>nA4X%9gj~J+k
zGk^O(!>vtVtAy=WzeoT1?_wyADAM$TE@7yNlaMfqiIadRN+=gfJp8~o>l-cBHV0f?
zb=c{2n49V|S}hQH0Zt~5p#}bx*Lmydzu^2&eU40FgoVjzKD0Q6zun@EOS_zynCH_^
zKEhgii$8tl+kfT0zBk@}h~}k&K*!jYN%hnTCLdj;|MZKfC}l}vk%AxtftE3*NRY8$
z|I&5ZAJ|}g`UHus4$HZPkp>16YgVo<GwAuOZ{K0Bxl2BuCrKja7w7o&r#`_8FTOyd
z(<X@0(WC}?-H5j?uW<U*SxTi6W0R90CD*TAXX~9eX*HXKo(EEpD;CHU^ElE$$b?JF
zx4FIZMLzg}5AeuC_mj;!$RQX*ra>#IevYBDRiLy0X<*8GF4K4jXp<opI0z<&RIj3C
z>b6U}DTd=wEwBxXv5`@J^_PDMDJ|xYEOGtHCI0aD{uzE0qv9lWb&6!j9D<4!u^$qL
zVG69G1f3*l#LCc20jcQ!{VtU#l6c4piZt)dIH)R9D$VX=c+xjuY;u8*eEb)=a`9;z
zd+YeVhc`&D%nVAw?*1;j^<BotC#aUI<O?}$Ge;PKN_upg4NSwr?|0eWxl5?w!u<~t
zdl4%uE6mPLLL3tf23VGhZ8#7P(6Nv0+No<=7zks?(rGaS)5Cnh1Dy~lf0u&vl#!L_
zRI4abJ*Xaz-4DW+l}=KucNo7Ws8uT1hKXqykR;T~c?`qkYOl*^wZLfgC|=NEu+!nl
z!W`RoH`!~nXm<MiH6o<<y$-7Vw0@WHKp+0}6Ius#>Ob@kka29zoj8GILbtgNHdxgX
zs$IwTea2>|kfw=FV&W)7$~36ekS5(;7tb3|D^w`DS#GafN9cr!=~+x+uvAL;m;d6Q
z(F-JHTi`eL@LSthckbX=lG!8k7)V5%9O^2R1b^V;<gygpkcZ~CxU}hW{abe^9y`tK
z<OnBDFA*!B{oNk2?-RWG3ZE=&vpBYnk&BU`jrIn(rb{$v5o2LHE-FztP6iSA2t5q>
zP86{pATl<>khtX{+BAq;JrZeQRi_DiJ%XT*X-UGLK&8Y1dkqYMU3%yl3<fMO-=xuO
zF*-UyK9{mIy<UT>Z@<b5&whu-Zi66}lxpK_Z*7wVAp%XKS?BKEO>E0$WMq`ZBgdJZ
zp2xObk|d<nYBM@M!Pw|H<x&+xfbX?QQ8cOc)v{BGj!Gv%N&|xy?Mu(%-@ZcF_c4S+
zrc`BOViH3ORNxau5vJ?n=5r_n!62a9>(CqY2;%^aMq*&A2HCr>^Vo+@V2MfWYk$E;
zXOCd%ET8?2zlV{{Bcq6ZeVbcf`V-dQe3|3F@SD`moX3_nQG18(Gv7e3yiGM*z|<j;
z5?Bcc+hM$BQY_`DRhwMD-ezSxrajo>_)Le%$q~$qgVqs-)C7Y*7Lw-XZ5A?5QOMND
z79GC)g+JiN#aHQUtzlY%{mur7`U#BO4-7FsL^7nq3n>!3AmH-V>uhcBQ79D2X7eNp
z>ie7g;QQa??)oOhTpkQbr`up>XARR-?CjRr+ug^sY{tgMnVy+vd~%xJpu?@(%S2It
zZDnw+EK_4yMn`IljgFGf7cea+eU=<%yvA5bOnmbdqOBXWw(jC}1F#&XCns@C8x;gZ
zVSwSdWD0o_E%5sTTJ0u%zmL`lmT6PXRmfx<xcv@)>(d`&cjYP@Ke$dKj`{Wf=0C@&
zPGBfStG&gY?|*~H5S;y6f1iA@is$vY{pbIJ)(^fzs6)J7g6Dh0VS>^TN~AAQ%Mnz|
zH7wJ@G9+hDOt3KPaq~`(wXKNu^?<pt9#fNoQcZylsR+Rv|MI`5?}t42<cHaP=OQz%
z!*6}?7$=wOJiEF{eQ$$L{>;z)HGnz*(s5rB$4J{jW-O*3zd&PomAzYcVW=xZrl(w~
z4NPO$7u)F3x_O)U!35K=4^M*WX;@L;+u_#oO-lJun)L>`jE$3Zm|a-pH-6*S`TEzt
z&hE}0fj{h01#CkSCGgLG@2k}JI(+2AA7El)6vN4L;@k!1kNdP*EjHKhpoeuQVKe|`
z(CH1hbm=-P>o2jsahJW_79aZHV~ma#f%u+Hl@uV{t1e4JlT&jDhm8=2I=Uo5NOdnF
z0|;Y5lEzNn8_SBcb|;_DbL7YoYSl3c<r<YrmFw58&}`KSlL#v@Qy1LOQ5(kzK@boR
ztJmToQ<r{S5n85SOZ4y18~3B@;CliN!Hw2>n8W+S4qKs8v&`tk630*bJpHZD(`wXd
zw|!J1K$=(sgEua|!PMjw)nXOp$0S)zu2`bI)26wz&FJ_T!h*^91<J)5ot*}INr%%9
zJ;eIW8}vH^I;{pttU-X+ZL@iIm7^ygz%(R=kZ8fs$e|A(Qie2B{pMjSn7Zd))@k@~
z+DefEccJeYS)@!&F>3gIBPBwc<Z^i&$4;3Jr7(1ehyxr$v$S}G?cFV2dF5r|FlKCI
zoc*0`Op$iG|NH8RFin$>e&{g<g8|#Sdo){ZdVP;5iqq?YPIWkznO>{mYcn|rbpE#h
zzaJv{Qri1nzZuH31uV@@bL#jR3?YaI9fWC;u^rruOD>n8P%NhV7DcEy!OCQ?GZ}(`
zPq*75pLNJvIc~ml6W{MrogBlk9FkBmmN)sp<Rr`QT%u+PTq|L3Z5!1QeDa|sa!v*n
zhG;2?lY|(Juq8&MkSak5O(~!7$kBb4`dzkncUd1)XxDF};*hdVn8^7|PVP}G^f65z
zNrou&klIJ6h&W8O$<lTRwIoskX=foxh=UNv&JYD5*an7WArVADL=r}*LY71rSayY^
zAP`2(p!42pg#K-NesEY5BHf67zt8pS*YP}`ndy1ljEnDw^t$_8eDxV#ec>5;?H*lU
z<CaI++S}v5`@jAZ{?2dy29<J=we>Y@JHy!c1V@jaVtjmxOePC}Wo5blzWZ@)o2~sl
z?r!cdKR1P8nyL9}c!1W@NHa>cL?#1bZQ*UK(7JPzFo+nbjbfQLj$`8(2FmwS9kL<G
z<@1C=M6=PLH)sNqOva#M)o=|rZH7?_@9sJuJ3q~xEAMb);}&7Az;FGh{{>cg44EW!
zTASSbi$ABiahv=9?mwhBI!z)CT9;pC>nnc%tsNo_L=k=v3=MPX6iE@Gm7r26^5D4|
zw{ADtuE$)wnb7F1F*TXNv>a4oU|1Q1X|r?X4HEy?kfj<Q|IFWj@b_8y=ARHXJIo#(
zr7`HSIq3cb&R2)&s)s9m#%a7@z~!sg+1qbZt<-QcSrVmb);D<h`EPUW?MqC|9c5u*
z0k7Ysy|+t}1hm?zcGAscnVgvB`0+E;Mp7Bnjhi>vZ#1Zs$2fiZL5lets0b<2{-#3^
zFO4)hl!Qb>+}dUD@{4%etC(06#zv5qNg-b(@w`;>Bqh0GfjEw7?$_xLT9~HGXr+Q}
zI2ef_iUXSa4Vo)k7}u{d@O(D+_c`;~Ut#LxBM4y<2K%&jw@LCj?)!z`#CCH8od$3J
zPya2Q%dZoLK7)Qj93;d-V+jMpv@wh{vZhwck+TcwGe)Jo;l-@Qqvyssd(z<A?QQPf
z>9g0WQ>paG6(DCT#+?Bd|A&9X3%@_lnF35_ByM(;zxn6{AG@Hqy3yj|7ysFh_g;Q)
zFM=UcG|VPD0B}~d%=CvIpt)bisCN(SohjR;B!&<eLP0EP-C83IywogV4#S4gEMLD&
zHX~3;A1Ox2I4(;IbGUh%zw>v0n^wzXXS+$W(Ibh|m!oM&T+1RMVXxcgkH7j|zW9x&
z$z&a}S(j`!OQlk#R;@8MTA@-bB9&rmr_0LPK38wvVts3$Fi!Z+)8C@etkZ7S`RG$m
zGB-Dgl!_#|r^9+TsM844K>~Xekwgc9#p(Bcps5p55Nkm;KTa-RIy6%pzNutNdq+`(
zRvOcEn3_Gp$3Oi!o_Y2!s5cup_V{6C*t<;_Ns<tT;ZR?Az!WAZ9iy}+ir%N?oqM3}
zFjo&!ox+bqrf3crEDgeBZ0aN@&wP;QzV{V^AVvy7zZYUjgWJn1T)T3aQnticO_9+W
zqKIC*ju!_Q`79Y}p+e2>`ZiXn#_^@&Y+bpEYh?*#NZ;$@djVk-5r+}WS6*jq>Ntjx
z_Ef6G80K!J^*wIJ#I|kF23jYGVb-1;21%=7EDdOc(!-Eq(&!7Vhw(Eql@17vv;<m2
zSe8S+Si&^YlbN9s0>6i4Xxi;2SKhfyqfsXq_-t)#riYOsbNKJ;Diw=-=<z2o4U-`B
z>GZnn)$43+Z?U<(Lu0>5xl-ob`SWCK8yzPMJda+#PqW#i)9ujd_3#EBo!)?0X$&dJ
zN{wL&gpmULffp0S$-Vc45IB~>Q%`<?@rh9knUKwAp&5|`0dWxEW?XDHgVrj|loa6R
zvIqp-ZVSsa$e0ygd3hNjx{Qt#$d`%;TVP0m2yKpyI_ztYR%A1iuOamoCr%$HpR-AP
z9|KJyp+D$iWgHx75(Rx6%LECDp|MSiDJN#6yw3^ULWUX{ORxln5nx&hZJFRX2qkC?
zy10&sArvYB+Y-c4gp<t@#3^HJ+ZlAMQ4$QTi98>qAohH`Fhn>u+449dF^C-(rEJV<
zg>JI_zW3ZPQo*|>JtXe$*SUTB7DB`P{1R@)q1Wxv+}q@p7rx2cufIev@E8Q(DFer_
zc>B^j2o2x)i*IxP`E!(tRc2-uSvq=x@$pGq*FBW0NJF9mpT^b>LTKK)^g6|Y%V@1i
z-|MF&E&_3I5Hk#faIuu8zr9YzvZ)oy<T5TOjcu63eIH>6VgZ)p(C_vbbb2_hjhmUK
z-xk!jwu!tJw(t-}NO$0aVN>2*A&8+F#C+mE`TNwC&ZqaRex3chYZ#>*_y79eCzCIc
zgfVx%@h{l;=AWUxF1<*T^c22|5D92h>P{8{HzTQ4MsOUPV&0)z%Cf#y=lV*(wGBmU
z5O89_rI0tUO$)6wa{U&seD%wG;Fo_B+buFxF7v^;ahhulZY;NWZ00zFT!F8?@%;PR
z^Mi5wy^(wR@xJGC<=R!6jV>d#Q8I3pC{AeZuk*q)-{AVCw+TE=u`-&*nCxzXN@%rP
zY;0`cI$6diCOLZa6cZEEn5K<VF>|vEoIQOO!?J0&`V{gRynX|%4GzYDhaldN7>SS6
z7UA9&VXK8<<{2FsK>}IFMGXRkfEbPG*aV(OuhqeIEF5=|`cBA|>+6Jmgd_SmjzSxf
zvFTaHrYr0ZVpiQeANh^Hi?&^K*db^)$>&Ry&Oe2)B)yG0T>qorAzE7|Nx<tTs959V
zEG9~oRC}vInih%9Fz^O=0|W3WI5t^FA(V}ZVlpb^(X(T;7A@Yoy2XvHF40DWYikNl
z12>~EH=2}APk}Q6(gxdPByaMOks{|$h5X~!F8!7J`rbgDPTLDo(3WPSr9fXYdg3Ir
z=ho>y_a;IO87UMA5-kKa2Bwe%8+#0PcgPf~7*b)IX}g4J8Qi*YnVs!*Dy3=0M#l&l
z<Kf31VtQ(d_0`weT;C)PW3rid1!E=<#tC7fFpU_?5DbEtMyrny4Fns+NkR}0BU2Cr
zQA`l2p`I#zC5K7GYj3@V=Xr!-#3w%SVU8V}9kO4^&<&JouC$P;4O|<k^HgcXFy=$*
zp=(v?^uLK9ldDi0pF~J&__+>ht29EU3{uJ#B}ibbI?jU+K2Dq@tSn#0&9p(KqtG<Q
zL=M3qAnq9@4viZ?k|^TPdr#Z@mv8yI+8Bg5)D#{HtHn^OHZ-E7`ECZLWpm-NPZLC%
zH(vQZy?&dH*TWG4!jP@aP4?@1n1(?q2@$$ZtG<Oo(C-geU0!2+Vuq<xXK*uByzO0N
z6j7=c7-$be8uWVudc6Trr0BHv`Qo4b0i8h#>e9|ijcuml)j}altyZH}t1>+`&d6vD
z+mI+?gib+R>e*Jq@Z0pYa$v5|Lo)@4VbEp_q+zUclQUeg-b5(y`X0CL+@jfPqNA8X
zrh+?GBuPB>+dFLTcJcojin)K1YkFdWY$mOF%4V~S)J8dW<XFmz`aX?zi*~z>>$=#s
zMYr39hGt@Xl#Fv6r8Hp_5_mq}d;SG(t*(<X4Qho9Vd&vxEM}(XahwXhexKz#H|h2U
zDByv!C;8}yKZIl3D55lDEJ-T?g=wI~P}8b4DoHRc8_TxwyofOJsg-BBeDOBrs>{e|
z3CDHG<x3=9fGI7s=`w3;f_9&;-@48C;u1?!V~mI{`&)HPp>P}*B?R4G7dMx~$(GPb
zLKF_L3k8(#A!LYfTwL41MM5TxSAr02I4EsE<P(GerfqX`t%+r&b#zKc2+M#d!PFYd
z&5#I&aYz~^B~**}yZhLVlOB7eAe%4ZIC*?+fs?^ZB6?cV{|R)z3Yh|}BvEW{Z?m?x
zM!ryBdTJKOwrDk5>}+iC!uP+-^-Hgz!jK?>wx{Vu3B8R?T-W7+2hLHcR&eYbi;E{%
zI(mYU(NS#6zQ=eABQ>EUnw_m}#%89taQ}JUc=I(r_R&vZST>1{hnab4lMq6r(Jy|F
zvJt`b9Hv&tB$deYS{)=1hY=V8zvW{|Nyc^Pbh_-;J4o#k_#XX!8%M_22GmAJnO<DP
z&PZ<DZ8E!bjt77C7ZHX*Jg5_Qn&gWk)XserEWu!FiyL3~UHaGGNE>GQ5s9aWhm#p6
z<D~9f!zPo>QYmM!WCl>gVT2TdWAjz!XH2f&>9BHZhl`gw%uEj$pQ>P(5ssF;^GCl!
ztJ~*^pZYXAZ@kQrT!z2%!6SV2tt~D$H@R^B=iWD{J0K~B<FdnX`}OPB>GeFOr)IGo
zmoNxZK=-|`v2yt$11~1AvUpL<KmEV`Q-1z4pJ8%pg0;0f7>3R0=s3rYo??7r8rOBv
z8Z0Zr<oE=!=hJR?c=~%!^Y~+rF+N(w^9E?4Nn;(1;WWiS%S?Lc?zN~?YPhC}YnlkH
zQBh13hX^ATGzEhJGSOsRhjypS*3LfJOqr?a5`%t+Bu+52W^{U*(aAXsouJ|t<>RL)
zj*ekO0e7xlN^=o&<A@COufERupZ_buy-m8kE}d={DGjRQ6{g0=aGe~zUP!kcv#|lY
zYM#U`BJFX6^05aV*<gj5z0b^eOy05yy8|k@9H0Ei5<4xQt1CO)zP(3%APKq&)1w06
zn)qRa>!#?$8a|8cAG0FrJ%PHExJaEYLvDtHKnt18GW+oTY%XsQ-EE{dUCA)WRbWVo
zfy8UJ*}Hm;(WT><rVCaEVVFq6B}@eU!GKnKAIEW7m|x`l`TNm{VD-)&W@l&E-QA|!
z?Gnc^aTs$D4Wg7HjAGF8hXIGyB#9zYf@ByCqDddp`du0!4McIw<!f(a+a`u7aNSQb
zKQ~K~L}(HWBW2c<NP+4Bh&f;{(>DrW7$_-;6<ChT=)^RpX{B)^{BXO0IZSU)k_4p_
z!XQK-m>8Sj;Rhcli4*RuE|axwGLDe~M-H?v_p~qvBiI8sY#hfVKY{V9(8F(#v;$TM
zQr+5t3s{IDSWZ(F4PzM9qBWMC=gE)#9B1$U0PAaas5csf-F<Sl=Gc+BwACZ<3A}z9
zeku({CMJ=lO*WSWL$P=37PXO(BpRS{3OljMSh*C)hY`(I1En;kjM&<^%pZU0CA=s(
z)Qeb_O(B<Kd~B43`9-ECXUNzY%H<MEN9H(r{0QZ8{vf*ssfN{ILZ#zZp)nJQGE&x6
zN{LR=FzD1B3A0l(EX^*mSKsFLof~X!th0W1gWES(aAb+mi5kmm*VwKP7)1Ym13G}|
zsmU~^V0hC%2qQI3lWaauZESSt%1%h4m^*8$eDO<PBAapf;0Hd)_~;};gfyESdi_4O
zX>x3Fo`)X1pX=AJaP!7>I_)l#W24maWulQ9EmL7RHox-OpJ8I81{i3mFzxiA;Mg|8
zHVJi-#t$itKw`OB48y|fHK|rcsPFGFI+meSwlE|_vBnPqFeQ=SA>(A&YwWSU)j=1>
z+3<br{VlqxNj6)+v<!5T5bK1%8xVK{bgqD%&CqMqsYnxHIp|oUMTigzX&MM&qNGM(
zp>+aDhzcTflrV^Wn(dI$icGz|DwPJ9P9}9#89Hr?aBPHWVY)6_Bp7}~*zJ<^eGJn?
zIyqv;B~&&BS<IY4D+$oq?EBwzp@`#%_4Rc&H#ez`j4(61h~s40-CAdD<vPzj{YBQ6
zFC*eKi?kC!Hxk4eO8FcQKl}iTM;4ixp6BeD`#E;>B-LsS%d!sTD^gkjNK<2!3RGt&
zv2q!XkDg+8f0wu3ev6Z*PN%H6Fw-WPVQ!w*8lfUYtgsvxQz=vspyG(Y_Yo5O$RmtG
zk|0JVamqd_!ALoeFd&;9CGd*`VT2(CW3zL}#6|@nok&nSa}LX~P~A2+F1<*lSS6Dk
zL7NGK%NJSy`X4d4yOy%h-9AaAvGX?L<s#L}2uY%e<23H7<C*N$eQYyr2r6W3T*D*|
zB1{tQUn;XOm*?%b8eCniv$DNUv6wKDlZ@p&R)7D$<BdQ4pLt@sNFi%5nw{c5`}h>^
z?00zP+B5Ha%MWtshLH+*p3k*w*M{Tv1#H)5(D&HiUFEs&eu<k`E)j*AUI?X$D%Y-G
zr?KBe=!6e_@B`%YC8j3lIez>!;}cWEyuy1u+;Nbwy}C}mQs%zXr?`0Wbw2jdk7L>n
zQ4$XI^(K(YXQVJNR7BB8P}Lfi)`&PpXho;fMJt8p_lZ?HtszXGcS1<UY89jbj$=_Q
zRzXNi%b_$@BMLPMigq0H+O_NaY^MdgTU>em2eg`fKKVC3!{E|oHox|H{N^r$L5FUy
zjp-PS*2bBdn8K0{y{^yZmSVMgnw|Vn8di=cKYD?}L>}dV84K3mzRT^EtL(3QkNL_r
zlM{KoegwMD;z*6Dkv#SJ0(;xr>^EZUjDcxd1b#qY_R{>?VXB5!KVGKEd)Cw$LyGL(
zkrq-wjDa&b!R&*lxViTnL;`dQK9rJ}GOgT7qL_`>FLL6ePhsbaNGWj%c;lIGQmB>D
zLUH~2I~+Z74(V121BE~mhRLw2E}>rEBMj1!g=1Mmfp0{V0O(<6+t72WA&wRC;Y9de
zzw80SC4|6r-BgDbhP-|GEyhMiacq~*efHDTMk+%9crd0*wOsEWg&uHzLWq>LQfY7P
z_{0p3<Dil>>IB~HnoU=dq#8M;6y7kd#V{pfW7AxC;88r!=l04qG8rQsdkgtP4B7#x
zI~d8n`|EnYD;+hfL%=A9xkc*G3NH?YtqjctdieLG5Ma2BPR=ngbp&Y|7$m49z(@u-
z1|(5LqrQuSqE;$UtW>exEY;CTe6NS_Vsz3)+Jc;0K*c$98o!U{_301#xWy8NFsPPt
z*p{H}g{XsT?0a;29qw*#a`pNZ%H=9^v-6xda+3YMecrlwnFk*{%b7E$uq+2H1W1jM
zsFZOwQ*ftsj8rB{gNQv)$xt`|j$`wo4?fAQ)n%4%T;{D;USeZ?gCH_+i!m!(>#S|v
zB}|gP;%fi<vF2uG?!~@{d-Z9+VFZF@r4c}_L4!DoDHe-#yDjRCI@Rhp<=P0>mv2yS
zbukQ^Vx>Z%T%<Z$Bezq)i~0<L29;8QY}P_1CQ<~fuP$@{`SV!C96}o8%N2yRK@tQo
ztkV-hqLO&{wV5Pwi~-mf1ico8Vu3jH5RQdH;yMoPmWK-ZG*pYtdWUt-#8?;y!z9Vt
z<cuPPY})fG4Smm`<ohIXgzY-Gxg6bQll<5iSVk(gPe7qCEImAGrH*wKL`0sSwl8$O
z)NL&Uxm*q-gvj?P<nzO(7!zSxV5FZ@g&~G*V2@Q18H=Ff0Rkm$4Aa3-4qg!BR7-T)
zF{bnWL7ne;Y~H;~x7%fWY?7JzMY7pEch_%m{hgQi{&&92&c-T6qR<k$J~RfBL~AOQ
z0*^j&f%&<4=H?bTf9^q!962`ZfVEQj#?TO#YFp)SL}@TJH-pwOD1-aXp69JgZxDt7
znN0Rj@1=DrX4Q$tH^HzC49g%I#Au}%^!mg~5d?kO-8NpgjUGPW4QY|j7bzA?2qNMr
zoeX7iE~RoA!?lSAAxea74Lmm9zRXj-4jXU3&g)M<%h>EJkAL(-G`{;S>MuQw-)S%y
zv}tu(2mzJR3R9C)<TE)2y@0@1tZyXj_>=VPDXeT39p|vSZH#!E5qXE{YQ#v{LHQmf
zTk_E-7rA-2${W{qxv|w_87O8v9H+}jwo9=xi^x{N5ZJE6g>s&wGo$>|OUv(T&kfVU
zv`myr==FN6EH4v<5p%PPWO5n0-5z^e>pb_}FLCS2JLpK!_h1le2AeyCQOx4v9JP@u
zajcn|oae;JvrJ4(;y4ZmO(ba|gT{7TtgK7dA8_jQ8G2rq*WP%Q^Y>psa{%HcdI;jv
zUhWvAjcwYgEFxA!QAiZWgrQI0>(Ol0QQlx^c9S?xhC;D~sWpiuP}(3)6q&q(P@0Ts
z5C$=kvKSqo=K94;?7jUO&E-2xJ#doxE8nI5{cjO=+w^)J+U*vOYcoDN#pKu&Stm!M
z)njct;db;GE46WYrj6~kxw7*P?Ke$&Jt!3Nl=CL#(<kwF%dFh|8m6<sM8##`bukBN
zyVXdwO2ITK<r+wtP|jz;atMQfcDse`IM|Lu@M9XH-V>-xk|-7U4#Qm!UmS*s)CS7R
zGIr`1TbHj9-`YcJ6GY0?VWbBxTSMp8D(&4}^5b(@B*ZIk5?p_cw+CRA%Ur&4i{nS{
zN5=;HyLGZf&67_)MZ4ADD_`3sU(C~}w-Fc^ra>5|E?A|~2$h5J%pr&y*6kdCl7Caq
zq=}-CQn5%+X~HPtwKrd7YHEh>KK(2|_qm_Jag0NDMhh{FZxKU42Mk*X4k9#S(r`{E
zlOYTvM3VZg7?#7Up=l$%t|_yowZ`*%L{Wf{83YCs6Vp8S;G^`qZF;>n(Qv$=h9-q~
zGk=FMBg1d-Cjsg>%;ifCO$CXOh9H|lyz5~<l+zvj=0K;Bei9D41TD)2vw{%nu$HD;
zo@M{`CA>Js@Va0KOw+-39h8+`BiG7d+Zh7iBk(;2jV>Eo8(^8ZSraWSt}XA=?0Y{H
zIg`nxDcmYfh+@Uw-Y%_ngY~sr96NH1`_Dbhb3b^Km790?_=nC@H3dl+5p|k`frp)U
zkZu7}N-XIRWy>T^4kZObN2w!OLM~tA$lM7oUwo7L{tiLlvE9~mwww4vL&ore^23hL
zzj6b($BG}Uzt(IKD&=&XcL3<*yTLoDKF*M7{LkIZO>VE=;mnycNTDbdD=aR~QK{6}
z+uy|a9$2Qqv13crYGWuJQ>v8cw!7G-$>RJ1LZ70&zs>IMCU&-nG_nZO#*h~F=mMsB
zje*xrwJ}aM1$A+J_&N*25XZ4;cba5#4q7J+f)Le<@jRbI?GZ~8)5;NMEIOGI#WQCK
zog&pj7Fl!%w{B1@<O#zV&mZ7N5eiTuA@cf|`4X~_LxcfQy@fq7lJ0_zAsRB)L-(lG
z2`UOnq7W<V(&~9+vo_PSRZ8UoY5-olNfZRwxl-!TCcRDv2j(P<kfw=~%@KJa(oV;h
z;ocr$WDpry;!J@^n%F-9P`9#j3)gj-ompgJYKB~)$mZHj-hA=fJonwN(B9p{QVPR_
zW+2#VYa#^`lOsIx$OUGnW;uG~7-!Bt!2H4y3WXwuw9r~%8YW6BynY`;2oP!H)j@+!
z9H#~Y&+|BS>TEi49^O~f!6+n2WfC$`I59+n0YN_`@_phkB2)=oFQnNM^!z-o=##ZI
zDva6L+hxDmX0$SjoA=0N%edJrK@?#67PhqMh6*_{OQkT%OW*h^jW^z4&}}jDz;V{U
z{CjjSy-MKq>2^DGIxQ^QWTZOA)Oh-kZ?$^t*JEz)jj`>XVW(8YDOC9M&pbgW61=@0
zQ72&k(k(iBFEiGBnIkhLj4(!Z+MGW+!tuo#*H$*Ue*G@>p22P_W_l`5Vo9PXCgU>9
zkCWKe`**(@QlwqNoi^96Uq?#G+`<xte4bXb$@bbR&wl66xP9#sBnd$Rtw7@?0-+)n
z7iM|*kq4O=pX9{Jvz$J2j)@^dcbJuW&{iQ*&Fs|CMdCQbwp=b;c!X!4{VsdE+iAw^
zfku6pNFXCcA|ysOi!{I=gs3ng=y}A6#)~3)UVx{IXq{jhF;XeI-7bwzi&AEUY$2vl
zDpRS}NJ!|mI%M+|A}d2wa<J_@H=g|ta(jzd#}w*o+<NADh-18ghvyB*W;0BT%`j3O
z#kDdhZP4r_Yy=m$Y)qq^5vrMpx__PhMwd*!Lfj~_y)M{IJSL~gJoxAdy0&2T`WG0<
zw{h$Y{caaGn?YJ8mhDn>9fV`xST=Spht6fwTij5B5Cwk?Idw@AW14naSD_8`P-7%T
z8rEt!4sLan>C?x#z1e_JphZHY!6r?qwiE>29^KpPSod9^)LN&1?PbP#4X%e7yq?4|
zCCfM7;pnlW>~HQcYMV?=AL00kGbjw||N3{y7Ya1?+ays!tfCauY0}Ilo$^KRessQ@
z8H0QJv-H=XR6-I4OpMjo-P@<#>2UGV8<g`^UVr^99(&~cVf2L(fJ#zu2WWXOco=P_
zVZmCF%NK~FnA^*@Q9Y0G=_zux$}pTbb+IPH_J@PM*j~TE*7h33!Z?;=VcRw{GmBh!
z=rOKcd7U7L5Vny<YrJdpd+-2$cNp<cim5wzn5WlK-eV_!2qgcT^L|CDRg>?+9|v(F
z8UxcdQ@5MQGBJOW{kvC*RhJ-&$hZ!UnE@n@nH~1$S|n;fqqEN&Z@td;-X8NumvAz9
zZmewb+$--8#_C=FRtmap%cNM$Qz#S&qku3B0mbdL75aXM#~%D3cQ-=5_%HvEpBfkB
z4WBsC*s+h1b+PgV;sjz8-B8h&4*8|i6qe2tI{7r@QV24R&7%)L#?t&8fBeO-bLEvc
z_&2AmKkl9!I6{9I(A^6l_z`d@Ko0lj{jOp@f4IlikO@ENW0eA=R>)!GKrWZXvMsW?
zEct8}(CqDRQ!M1^_4>p~LMh{N{``4t)8Z>%`4W|CjuXdDGd?;>tz1qIDATiqvB%up
z5pu;c!n9Im$Ecz0B7?y?QJ9b|R?t$QBhW&kbb>b+;P-m8TXj6oBM2kHIL3$^>`We!
zb=eF|R@Uk`L7&QKfr*bSu(G<%?A#GP`g6a;&2RobJ6o$%>;y*&WH`Xi=CCZ6D2Pz>
z$d!u7LY`h@pIlp#EsdbGA`SwyG%#%YurD)-qVzRuS?snVO64L`3v;MAA~7Up;L~rl
zvF$w4bx347_LuK+*T+eWN)(~iSZ)p@SB6lM5Tg=^!vtwL7>@hC_uMoL7UqvKHaSJU
zP-1(1gQvg#C%p8`m+{*jEEHx2I-cZiJtooyi%T;+^6&*_CTCb$JkF^z=a`wDC7&xI
zgqaG+4I7;VbXz-gJFQgjDiY$<R*sZr+V^c5ICdIIgw}_zW0BtAP+DP<Ad`SrzlH3!
zQL%;A5NX(Ic6jI7E}N|Z?Z6;163Tg-3u8IvD-K~WU}vp?lQXDJN!&=|S|AgNmCu3W
z5-O9Z-(vHnXNh;~lqYL+zxxI1+jr23X3+2B`5w7kj)}1;YLyYPP9D#XX>=2AwNA5K
zI?A@pBkAFZRi1u1Vt38Jv@;x896?Wx(RV(C*vg|ep5@p=4jo5ynjOlcBb+}u%R<><
zzp>Bm7K~K02n@V|hi#_M1Fcf2&ih>&O}E=+Wn~4^w3uHwLZvpw{?0acS8ws`U;GJo
zZd}4piby~^5cH$;eLpup#lsI@U~FuH)2GgH?%V|?#wT%xuSFpQ(lod?J=Y*KmSrM^
zCX74=gFYuuoJoO=JecaK;h0?!=?HT;d5ik}A&3uAvBFOhHugGPTiIo&t?9%H!xb!5
zT+U4v$(sRQH>STnpxYYI>4mtaq*^ZE_uJUH3^}b)K_8_giQlHVzQfAOt5{ZoWu;-r
zqoZS#i&ZLx3YKYO2#Mzh?6pg5%kvyNH^pcv$DQphp7{7L^T<b*xN`YAU;g6v*j#ZC
zs>sG>o#0u;M?Zdwz3x-&w!hBYc$|ic_Xk)Rmmu`X7K#M1B8oJGar$0S=@i1UU5p>|
zAphRNhr<XnTfj1{VRgs5*sue4q?8mE=cpaurgvo(5Qmv&CZL4CN<!+(mvQdQA@<*<
zOqZpxJhwJGT;BCLd+aDTmfvA!dV(ktWJ_JTZk;fU_~=JJMXTNA(xunAV=uFMYmIma
zj8ewz9$R!U4m}9$1rERFA7$$f7_V$L!`$378yk07UAs-=)Gik<zRmHY^HeH@6i5j*
z)a4BY^<wCT)x%RLAQ)D50V9<v1KC9gfj=0q*Qisg)=(-=#}Nn7CyAmz=(D<dhtW}u
z?Kn(LO(KL~a$=U<tu^8#AxYvNa=`x3X!Ia%<wxA9|E}v`#F{dL=~aIBcvlDmwB#UM
zSsXSx9Duxg*Ig+^qSFNYgK8Qn1t>++le~O!8|82Dz`1#fc?ZYJV4Dt(<>NXTy4^Ka
zH`aLNtykG=>@&M`gzDHdH*aq6rEk1Iz0<w7-@|`141+Yw3{9m{!f|t`BttQmr&6k*
zwcyU`H7=Zcl&$qXUs&7X!&R8eC=`;w3z1fUYujl^bYl;HZ-->{O|0z|>~kMQS0{;e
z1R`c(W|qnEQQo+EgYUh5k$aunKgzcKRV)4f<?y{_8InvUgPY0XW-?foNf<`-dObYf
z!}r6V3^NOu!!qx8AB0lIE<M@bbAb!W)$-73o<@xX-T>Q59mZ>G>m*81tyPJYVtaD~
zVOaEijcMnZom=AQ;tVnoG+G^=fB6}5xjILVC@_XJt}qCah+xo1W^HnXJW>ea;hU@H
z^%(el_IJ04BM*tdwjA<>3VqYyrMlwzt9y8DkI_tqf?;Cq_)Ncik+_m&_4aiDIYGw0
z+UM?LXRFCXw})3PVh9sk+DOYpVv-~={azC{m&MKJNaBEYvx#YCv9dW56(gc}2*J}X
z<~WLm8f3|!>oGIxq^yOl5IAIuWjZ@sbecQl$ET6DO*+U;(*mSnpp%$(tBsK@;bcqn
z_Zke8i#s-hM$$C|!p&k?*8AS`#rdPm%*~U@6>08l^Oev4kG%5iS4g}*w$NBv=!F*R
zdm+BCxbNHvE?l^u$+1b67LRlM<SC}6XUS!AslHI9<2@Dobn2V<UNaSXr!7fHm8ACq
z(s8U5dZ-mPNb@Yr^!1md%|QZ?S|F8)Rt^Rh!Uiue*q0XXbVBZg7Mpf~u4$r-ggbt~
z>YV|{+5t1;c}AymxJgX2*~e%HR7x(g1g%bkOl^#Ce?WG6iKMYhuiK%2>k4t!$1qZD
zu4x#IjgC_-k5Vp_(vhHH<3}NP+cVrWAEa(+j!e(f^WGvmmZRU_Mp^|r4S0Qfjbhg1
z!N+D1;~!-2)dnN2D@>MM!oVj8d@RSNSS(=+i7hmxvWr_T5-E-E`$Ta>CX>PcktmXX
zm(^@G+1=g2G%aRk7Z{(Irq$SIb@>wC`_`YZx_k{I(MUtn@HCBnDqEVLpXAX;A7*B1
zhBK$nbN0Rm7$2X&b+Qai>_{Q#b=ri1k5&<}O3jsNA38&mB55wHaUdrWhky>9U}&(F
zW)Nr?_-Hgn#wPYPTTP!gZnsz;STt-4RdnfkJ#OtpTx!*sE}P`a1tzOmCZtW&kI6d*
z?VVi=t!egr*2vJ!S2(eFjJ0~3BnYWZjZmu=(+HP1Mxe+#Sscqj#fk()zZcU8&JmR@
zPTW^veYL^4^GES|9&f+8!xJBSjPvKt^W`spf!$5bcx{T!+xu*8?^B#VLF4T=n2YWr
zY>U`e*mgqXb+I#f0zabHY%qRg7Rz*qB26^t)7jl4H2)fm64CAsvYgg}$RBxl!ch3H
zL1nWnJa&QWTYKoHcesJJfIxsF71jO*_Ufy+B1D-6$L2=~x&tD!%;|^D(;sws?%Ah#
z@<ShE^UhriX%Nd6-~RS9eCUHu(r&f6cJ1v$fRiqL_^=V;;Oj~mkAv8R^r6pR3V4Dr
zAd_*>Xl7@psPEUgxqOw8${3d}UFL%yd;~2H3<3;Adc)5n<U!OHiHdmp?TbWjMA(*{
z#$MzL+&4Z#y|K@0uf4)kPd!B@<B-eck%q+geVWY%ahxFY2BlIFgrMDSlO!>|7ZL;s
zz8_$U)RCL^Kt}J<(rMy29@Y-U!`PYkw~Rw0M|5zB50_TyoBS}Fko0<pAG+51@C_?O
z3hWFaG0Y)wK(o5M%GbZ~O`d!HMRxDr;y?Y5Kf{xcoTuaWh?7?OeT9lnr%kKZVR~_a
zbB{g2`rTdr@Xx-<D{o$-J@8YLinur4PG6g;He=xV+*w^?a&nyUu_?l!3o>H7HqG?Z
zEbV@SY_Y)hPKQ^T9Y!&59f=NOlpl~J35He_GFiHg%b;CHZrp@OV;nm{GCGbH1+=!X
zjf8Wj&rr#KgS}SoAZnT)57Zs>E<&-8=j6%LJn{HrJo3;(%+1YGES7LxC)H6#5#4@|
zz5P0O?yU35tFQ6q#f!Z5`s?p&&kuTWQxWyhp=}yz2B0}~eh<NB5Cn$@yC4k6WpgYo
z93cn-wzjv>K&f2j`Ilbg+SRK(^6-b4S(xLY`yXO#q>P~>cD6Qn?g!7YwR4xVr|(A^
z7J^hHFKwHC(5313DHcY^WHK0rL5M=fY1>Jo*#Mm|P!4NdLA|T!)8}Syz*SEmg@Lvl
z_7PZ?gNez-7v99zF|MPzef1{2tJk@Z(Ofut46=}u4pM|@nSf6FD8o(<t6an>l!?QL
z#@-fVQ*#)GOA-!ST@DVhNt`y8Xw7&=Gf_@Y`D!>eles)*p@`pUA@=vkS4I%7jj+@A
zjnazP3-N*oCnFeiJ-W?(wp$*RFh)!-(2jj9w}xRm?`yvS#-^vq7pgS3H~9R&_`mS-
zvtL0+0n&twD;UJEwigoEE{{L-AZJgXW_)y<qe~|_vUH5G@d>io9D-D)VwyHdG)Mv6
zpp}lKk%xL`DGo-f30lW!Wgujd8ncGteKAZ25Q;QYQfds>L{zH8?c4136tA@cF70^K
z8y4NfpyhkSagQKM*i&FgxEu=9rjH{V%w;5xj+R&~xkP!%+WHQIj>q=(J$6S%I9cjq
z^kO=V7E6^fGgXUhKA%nrk^sxJC}i^3rj3p@VKSiE5NrkKnLK}r2WBj;-`Hh*;xwQ7
zxzm(JEWY~X@9^ArR!PE9+P#qHpSi<Nf9?Ug?GLkatI2q|nd-&|9=2m6Z3~p5S}tSQ
z4%ti=3<HV^EhIvVG{^9LuG_b7GdeQL%*-OAlaqwL$BoM`^Sy8WF&oR*F_Q$-()1$9
zc1w{USUfVrV~;+}?9?o$PTt3vv*#EepTKdmXf<qyl7c~Zk6vqcXx1B=su)&&85V|N
zVi@UIR-+CLkMh8LIV5KkVC6~-%mOlQ;(74qKykgF<(72VQ2~vhOVnx8R)UZiGcw5b
z0>ZTrmzZOgWU-LtSjAy#p~~)lmzFHBQ!X(!HUiBpPL7wEJ3UQSCL~G_MIi<P!*s~l
z8H5rDl^_j^Fp~80Q^=A@Hk0Ml{gS)uTa1)v_}Uj<Vso>}&;QaR{PM5<Iy+nI^!ovO
ziw;kJYZF}@MHrJr$tJEykfsT-Pp8?W+4R`053r4dL<P(oS;Wm0h%6H=p;2%ASo`{(
zK;80l-{FDrQEbz}HtnH^N~esImI&db3G7N!nweqh;Ro1y`UT=79fg`2qyr}(Ji=p-
z9>q0dObiBoK-S4JRWO(v)Crnx?mKskJJ-I?rOR({>f}joU%kfk(PKn$kF~W`=I0hT
zdGaJ%+jW$m(jgL%LZtzQDz&@EN~6W_A;d6@_Adc+dgyv=b-I)aMJ7kbv4zE6W1ArM
zxpw^y7aqJHH{+s7Qo+2IDYvZE&;^?QFinf&Cr%-SB=mintqyT482BMKZd^-aS9${q
z`O=}uVKC_N=3B3^y|ayJ<~Xu=l0u<GGSuEFrSZKElcSH~T4^-OJw`6AGmGPdFi?b1
znuVp*`>zi=0Ck3up0MQ5ZZ1Vy>vnkk4vuJt5MCb&ozt6m>Vis?LMVlq3X)j4d7H0)
z^_zV2yWeJcZ5gQ~|NIZW&df}WhtJ)Ilg*=4j6Y0?SHfa_yUri{$yfRQbFb3s`S|g{
z7<kBr{phG2h?PPI348Sh)#@ZmOJ@+GN8fMK>sL8`{4}?2+`^PuE^lvewoqhAhsY!$
z^1#APw>n9v7Rx04KCPWiCNd`C(jH-{Kr(w2Sv`Tk<>bN=7fzq%FJ64*kRAT3^mK+4
zIJUvUk$FD%na}Xq&-@%GPM#s3&kd8`)9b6^kkI#ugNRDG!ou7Fk3aGlfAbfAiTZwn
zBPWl)?>(0?1#pHTu`Dxngd5yDG?8jbyWK8wXq+&FB$vyP$z-TE>g?C+Xau)!-{HoM
z8z`mNYwTm#Cf=#k*tC3Wh3B683p%|Piwj4YoSs7%7D5_mp@=g@e7A&QWXQN#Y$pY@
zhGo$kw29&vX*w*o;mzAk_6(DTVX`^s)7I(uE>Q{*gB`~lvn-}6IpSQ6fl;96^?CK?
zDv1|zw;l0H{Sp%em&M5u=1Y?M7E1(PA2*jJ(J<%_Fm01UxrV0`I(u7`s}uC%cxdF1
z=r}<oF{b5$7L1iFgs_NZf@N5#f6Ng$#WIQS5f8d_6M-;D?<Z1{FO_I_d)Rh{OyaY3
zeT9B@lJyP{3c9OnFj8Z3X6Y}xoxSJfIz@I?m-*x0`zO5l!q<^HKw1LFwF#q$?FMM8
z#K)g_f}_WdF+MWR($aC}78V&99m93A2qanyOv@tiyL1{`1b&AU4x|znkcV1gfr-XU
zALhas>X8+PO)lu+Ltdqa3nfVmn{0WMdSvs>w#Q~S18$9xr8!EmX1w1g9@NpD264B=
zomPwW5TuzPk*xM2vh?xyJM?-E#>@=K(s5csk}nmI%{?k~x$mJR3Z_OP36lsF#VBpz
z+8LZ7phG|y#q4+Ixhsl1aWYG*-Qx5E<G6N?SH8c-$3Jt1U-^|^;>3wp_~M^F%|!JK
z?Y$ls-?+@f=a16ZSj6``Wb+A$9}x!;DvrPa$0`sgMHD6IwAdI!q=4D6?DxIrg+h_}
z`J;?ZPGSg)H(r01?|k|9*}8QdOKD736DWhb+cAL>96!FuqmMq!^u#nLPn_lC=`)N^
zOyWA3AwwsThQRCW(`{@cbcAgffFwaf9HUh-ybT_(vBoe1bO_=FsWv?s)@mByj!xp|
zTw41k-)kqly%i8f1;q4u#>OROua4|@5q_U;ug&d#NIgk#%!I_WS&btykz!=8i&=Cj
zJ@r%c`!OTofYIh2(fVbajHK=5(WL^e<>Htop^!M1lbU_J5X&&oWB`Y;#R+m_F1D@7
zmulSq=t(~RFMmLz5%R)wm$>i28nt2`Cz~UmtMbMVZqeMfc>KXxx;OGfXlx-dwWi<g
z)9m)y*z*YkIJuaEIAZ<k4d#y@!>!hEav5~73|##y_w_x2y4CMK$K>P!g$wr~?DQ0B
zC`g85H#v01CMqFzEG8a0L$tR|=i+s=h$zn$dE|o^m_Id%79k{xSSc*SAhvDFl?tZ^
zEna@-I;a1}Pcy#o4wY(|jh#Iv$11E{zsAX>qgak$ztN`K8xY6Qf#W@W=nh#L7035<
zXAD0+9Mk?Nb1eYVGFVugC!Z}cF+PIVYjEQDaaL}x5CnY^KcT)?XL2STeFzn!q)Ndc
zsYdJoOGrbK&*!lW1FKwSY-|+U%F<}o+1|cOu~5MG24vh+2pLB)`}_M8@+Homz8}M|
zIkI?y#UrQieIL^_@qCZ{-Azp49kP@6#HbRj6XG}_PKLd;!|(n5E<KF;5TxRGDUHK1
zFWl4ZDW%dr-ysVs#Q}S3q<RT4yiIFRNt{Mutncy6)8FSWo_(6-)n$}UZ{FMWHviM_
ze}Qr_$GJ1-$(5=U(d^Y*{OOm!$)EoDH`uRtNRm`<gVKj_D+dhiyT`EU_mRF%7Z;B(
zJ2%b#UY{dNqX^MwXLpT>@hS58JWZoZryX%?ze&yt8OdhR-4GpzWHK(6Vc;bRj+4b}
zb?ELj7^#|MgNW|tbI8+8?CA$^GZ{Yn@sD!p+BLQsZSFM&|C>tymLZs#9Ot)x`)~8>
zzx7*8PEKPQcFH^oO*)nu8g&efnGs0KNu>vg9`-mwE|+=Vd!AlLaR}DZA0|Q^yyi4g
z3!)@pcV~xYtBzqvY|CPHdKxJuJG*;ydubKg-rg>cKKcksDW3lBU*NbFp`Y;J!}sy(
zE6)&k9y2pDoIZ6QrAif~35E$mk(dsdg$tNFYh>~{tYJs6VcTGssK}tviMX*|r(u`a
zcTAS+4Ki9YYZ^Gx0+}!n7FkO$?i#qQ7WK_-Hj&6eiMi1c<oGzPcAJ$(gSB?Zwe}99
z<viOVeClY8LAOn&QYDwmgHq@u!5tYRZgmL<1EgV~q6pJWSr)Anm?kO;$rUn(S(g|F
zrXx{if)!^m-7LwVN0fM|NMRT$OB(c3;4Gq$PP<N4#EgyGR7bPiY24whmu_<Wv!BM4
zCSLUZS)0{s7x~Ut{~0$g{Q#i@EK}fStTds&s|j?0PkiE&%pW<z=*T$7kDp|6dWKTD
zau3iUQsY{`L8q}z6!udV=3UAIHFPj*Opqy<lfq2dUumR<vmuz1!-F3(eGkYaHKF;g
z%MbPxjj%?gR>ZOmq|{*BY_D(QWo-1=1a9Bw-2QFGx9)PeAJ9n>^7$-RqnOBmG8s;N
z=o7rX)u58GF}GH6c5mTGpYC8ET`7@wv)GnRf<T)#87G6(0v(2!nPPg_EX`1wsF8If
z`S~KF6BVw!wZ`stm;d`e{uaOX+mG_tV^47O_%V7Nk9KpHSDwAgW}}6+YV<vWd^SND
z79vqZy*46F@M52Ovqe=Ziq$Hy5=4GT+-nes_kUntoL^#mdK$;h^43ew@|Dm3KiFBl
zj;R%vt4R!#tt~|-k(@qzf=3>Ci0O%GjvqV4vE!$Pfzp{YjX<Q&2fw>dueps%(v}mA
zx+m8VB8>pm9401+181{@G{rGZ3)Ny+*{Dp&<dE4K?{w=d4;(6E(`k*Nk)W{XH*16{
zAc;dJJ1xeW8?5haam`QYh9PFD$U8|wGl9=e9%nr^=mi6gPEVs7w;8Ku$rut6P0(v$
z<jWY+A&6r#Rw-r8qY!Cj5W0^NdG6lb;rwGCA`%8SZme_i!7&z3)Oh)&t0<v)_Dh#|
zblM;o%X8+5$7r-$nA*WHEP8GMVPp_Gfdo3O4v{jc_Yy{P3a$H?ZkeDLa_j0c^Jh#7
zwJP?Hx##Z*)EUhlZ-4pEIq7QL@ew?gfU;6_eGq|Q8VC(x5RfgGSo+jQxY69>#6*S%
z9zQ|0V51T*HCH6VF3mU%nskdrP9|{Y+9nsj{TwHbEwX&`7G4B}T!zt7nbjLNSvYZ=
zIPxeJE7U4Q>iaF?IQbF1k1;$Mra`>P@Yh47-MhxKY54D>k3P)Vv!}UvZG%dsgigBT
zb9s(09V6#j3_1~;Z+@Tu4|V?;tXY!g_hG-xJnr@Hd%Y{~+ned`>G5WFW_D*6OJD<r
zpaUe5A^{4ZK9K@HAVOj&6hVrT7$HS06d}+6!~$4iu?7RIo!PPNO?6e5x7VB3n`1Lm
zAM(AoY8JCVH+sYXv*SibSKYqmJm=(j^8ftJ@y3@a8jI*D=wc5oPUK_ULIGMB^1L87
z2CH)tJLbbH?~}ws(sWF1vq60tca-Njw{PEOer}F<G-h_LgpxVE?iO#p^%le7kaqh3
zU&7bE{KX@%=W&(m;|kitzx!!$taM>8-XTQ(;~s#A-{<Ii8tc*maC^9X481)3e;j{M
z(rXWS=goKd?hn7q^}E-n2LnMA1$ds|-48bSTmSuU@E8AV&fLr*_a1EUKmXQ0<HK9`
z$#a+Vp^ce}W9*Uo)**yE_TP_|5N;GG6_lzGy>8B_v&)pCF+cdxAMyONFO%q;d_3Ux
zhkdlndFjbzYEcNfK&Kf|6j7~In8XuOo6zgF34?$}bDlL>X6O24_BS3<fA%X}dio;Y
z{Mj$^cYgg_#98qZ8#^Ifio(x)`3w9Pf8%fP;>)k1{Aq+9j+^F2Pfo#Ybb+;wPbU2U
zFR(7FR0xDAev)40V=vExAan@bA@9+R(5F#(020zdv%Gqia(xM1WTfesO1(*@b9%il
zVc?VH1yV^aUc5+A6x_LUham8%RLZ>e+IPvbgyU-`ICJtWmo7Yk3L+<x5DE_iQqXwv
z6-@UoUT+8G`Q%B0Hcl0$k!<WG^t}rEHfPu$@~LKndSjNbT)}wC`AYgBVG!V(l)=FP
zcG#z8H5yH4<35Eo=yI9)lPl;fqqDI^Z_wf3Y8SGEr)PbP8sPa6L0BTH)k%_sQX?Sg
zv<U>x{-nfFwzKkl@-(4VE2E;op-w^|d^r^fNsP|vZKX&diNY!(@L-aX<pn`lX51Sv
z8T6RHaE@BDLdjUJ-q>KWzD4iWE%@wLDD)?f3H!~z|9_>keH&{MJWrsM!03YUM6h`P
zjnyU#YfG%I9cOLj7|YA6w3;)n$+iN7LP^VHu*bN!hb|J3&XCy}dlbp#mO>~G#iKpJ
z^IenvXteJHIYzif&iU0DOb*|_-R7QHU~#36kvXfYYvh^cyWjsd_t)=Xazmxjq+XjN
zSUAgkWrinr9&-2K0G-6Ftsdike@Ga66dOBeJWifC!z6CAFeuUVLrkG4CPOj-c#<?P
zC{^kRrO*>sFXL&96%10#i(fhozC~vNC(bR=?oRmrcOR0(WA5I6$lOdob*4qhuaJxi
zZh!b8PrPuP%EEC>Hw2L)rAA3Zx4ln1*0hHPljS7o5GA2LGf(b$=ptj#KKR5npIuy{
zQmS+N%A5Sd|Ng&XfBiZVO%N$m0J{eTomBDc^Uv_|OE0lFzrfiu7dd<G0t<_al%g`O
z8*?V?$?$+-cL%MLBW99$%zDe&miQcb5KhxV5TK^~;X_fK6%wl~%2<r_xk;4|CrxJO
zXYfPI+}s?2AM(RDzR!m@uArn(ZDy86bq2q5fz{e9-tIO#{R6^e!t%-*+vAvj{Kf}7
zUyC?<{2X8WnP24U-d&da1z1g%7B2HPnUIY~1m()qd_W^T*IQ_9h(eH^KHKYUQXR8)
z{v@|P+{Bn&&Rz7mcU^Js)`XWYUZm4adH;<&bUG7M<fDrmmA47Y1=h1nCKLL@3D*w_
zb~4L}`GP{beWKi`k&H5KU%tuOsT1I<Ke3jNuaDK#WfbJ=54iSEzQu`ed<8u_PmV_V
zk1P+Yl6XKm9#W~6X&hVObHDf*PI@uIH%MW~CppzhgT&_8GzV?47#H<naz1;y#c$ua
zOD{M}ue(h-s<FGfO|@2~QLb|Jy?0qTb&9Y3?9Y(rF;_phOna~6#xkD53QMW%Hk-j{
zf-R=ma)%s^=?ha&(Jnvm@jSS5=N`USB?x?GXPaQ*(xoR!<Cr3!FuHM-LEgcC;w8cp
z=P^N(L~5tv5&|)`D?!Mq-GRaC0&5Bm_IG*k;5KOyljnk)x2{od)amxRBw5Cdo42`g
z<r?3WKj7!S{#7oVJ;~nAUA8uExu~FGjPiYq)mR%$TNgPs9kvvO<J1{#jy&sr8bXOQ
z$q9Ua%5iuU7xs8le7MBXhvacM^hhNu#ngjHaOd_Nu3USMcdxv2#Qh5cpTPG>l9)6%
zeD}2*T)uptAoR$JjI^*vV`qIB{2(NOFUbpy5s#Y8L*dJhqjiGK^-X%+4o^Mt3ce~~
zjZdjkWA|X2`}ZGGOf-YxkZoxo8FONxi<l|l)tZ!4385`A$yuITKpM^N_CpYYG|i|t
zW;juA;tlrLfB!o)U-&tG?(5&=!R8i!@S}Gfx&QI&B?OIHg}?Tn{1yJ%U;pbgTeAr1
zgS8Y|GaPo=*u2Z$-X0s9>l_^H(d%_c)08NzxbeNO@Ks2o(O_<NhH9n$iEAE&K2a1L
zi6h+e;?Vdb&pIJ#_x6}uT;au+&rm5>u+~s6h1|IIE-Py%__M$Kr@3|e7W?fsv#ll<
zFFnCmzWin0dixEUjRxJpLyjLi$^6U$Pe1i6eh^O6Qrt_QsUl9)0>TTglE3qN7?qQ0
zOO_i@J_I58Ood63VsnpEYiFsqmdQhpv9;);zyO6dm?A~tQCU07oXu(36d%jZgZn((
z*`l-GVeg<zYi5S!6DJw>x^#Pee(QFJ?R~+^Yh{p@R&xfeGn7<VDe%f6X`J9&jqi9F
zV4c+y!ovtlkr(8XgmO@FnZ-(?qLA9kB4M>eF^Tb`GC?__yRkz&oG|H5*x%XZ<hhfS
zYc))g^WZ^`x8B+S8PL0SgLG#b+4|%)-`}`Oo@MyH8@U&0K{8fs>>5m1WpQPNlWS)=
zwsxAixmn898vfzfO-ir@z0NM<;Xcy31GbtX3r3iu_k$4939E$r5Qir)=}cE2SKHg8
zxV}SANXzZJce!?Zi`L={8(TLyeflB??G9J2zRS|;Ea#p)<8prEl>K&>{rv$w+hq21
zomzbZ+uJ4VA5dRi<^E=y?LYiBf9mUB<CRxG&yE;D=RUKMWH5^HgioOh;(iw^6_rL4
zYc<+x*MMq+l^*?(Mzn`qdhR&;ogu^ih^JpX#;^a{_lN?)`#*X>bH9sf2`+v4i)?Ie
zqtlR@AL5rJY9z4IM_NHTjn>=PEjZ{(&aH)1BA?CMcks_hDhmtvOH1f#6}b6{Y99Jk
z9^AOh|L`mSB|8s3KuL`sDLhZn8|54f1D<{68D9GID=f?{a_;N}PMki&+}u1-RDmg`
zs-z$u>@n!<Q0SCL)SJgACJLnI=sQB<DIe+iFrDN$UsIf`#3SMk2*ch%k00LdQk`GG
z$O+|Iogj+%_IG}lTBE|3e)fx$OI70Gh<0ztpr10S&9l;6U}66uw6{_F+mue9;ojbW
z-}{|!bNb&oMwyK2Y>TWIP|jnDe1w+n{W%y8S!&LpP?*UWZz^1pR#FLaywNTVzs$FP
zbd}3*Z1eSBJj3eAdA|JR5_7efAAS1{n+Jl+H+EP#l@o*!OBb6=wnog0c*@bt=^XTV
z{b9oEdpReXA%4k{<nB5Il>n4atr>B!x5pqJf26v8tftOj35;cQ>kjv-WtP736;vxA
z5>s?cISI26g4AfdJYjayXMV;foW^LG!U@~Py)LC0=j<(kB1s8_#GhXvzQ4}TpQ`hl
zolTCfoZzi@F4L%VNs=LFPMl+9d6gUQUEyGRpI2Y{EbaY0mKK*e*gv4qIgM6}*6b`&
zz{chV_wH_D4FsV>fplT8+MzpITk<?*b90}UKlM4XL{q7RWU4|kg2lyU?%lahVG3@i
zG0%6`Nnbx;a_2)TpZN-M<`~j==*J^GrZf<14AvCrqTufR+YH7%?mxW8sgqChhu`@&
z|I@GjJx-oH%@a>N!6eS;4~7U}IGiAm_{wK)c8<atH`cKhZOx+?GOb-hmusv!Y^EM5
zU(HX!>~<1Q@O_2nE0i*h)NCE658<5Cv5yeaBP8^wshb)=BT!Z#g-^dbW^Zecx88e~
z$&~QhY}A-()Y;qHqbQs`O{z7+q#zlMQJ%t|j&!vJi=3KMx}~I;#;6@`BvbQ`#}P#=
zckkY3X?cmY)nn}J?4zwiYE5Bi@3skzCQSz%coB_U(HRROe>jcA3h|V5KU`k0w6uy8
zaIn3J9w(T*z$=B!g>}lq9^KpTF?;Go{;fazZ!#EVynguu$HfGsg?g>bfAKf|bN=dI
z`|J2YM3EQl?>^-6<#)Mq^*!$1yUxze29rtb8r>|xXp0bvD6CK}m+@3MJ@pEt<fkHz
zc|1wtaMXT?#2@~dv6kUz%=-2QH*VdeC~}&O7ISlRbPl!=LNmKCM{aVWa+!ts1&*yA
zXK8VTy`5e9gANzYT)<NS&ph=U^?C!C+Im>m6d7DBp;i)o@)9InxcVcE&M-pal_JJM
zv0hpsZf2ZWIZqIlX?M5TynlzKm3fwyS3pQgQH4@D!eZIq?Xb7Cizy1`8a1kO%hZz?
z+v}3$1$&(Xy1gDNbMw?@7isr8e7BP#{ghgDgpM_#Uqe~V%&ZeSN#An~uf@?24vm|o
z#0w)v;}J!#Pqo=V14W#=1D~hxgOFkvla40{U*R<>l!W5J!#gY;KTd1uG<hr;XJfA4
z?a?1;YJtM+ciDXV9gcnD>z}yhNt)5DmqBQfIKv9q9cVha;=(iM_=|u3-{$y<Grafy
zHNNttFCwI(*B?-?*U))P=U|g0?&HfTx9YJ2fD=T3l=es&Z>>bzLxb5Vf9`PFAz?ZO
zK?rk1tB8}FH{W=dAo6+R-5(J60oCe0?cNr1ixuV;=h0fBjUdho{6MmHY=M3^W^Z?o
zwPPnqY(cTJh1uF+_0&af-~Et{yZ71q@O?TP_h~g|Feb%U8JW=-Yv}cQH0KskGC*WG
z${2!BQRD@_=hF})>VCxU{MHXScK$T??iyOHf{T{|?p)Vw-YF0lYK$_$jrSjL^@9fp
z&u6AtraK-JOf%30(B3~F&0u$6srVXSO~7jsmC8K0b&unoM{{NQ$ba*btc}|r@LT`z
zzh(FSdq}1YT_8zQ&0ZIdoj%KFKmR#etvODgI>VW>=V-KMrVX6~6_jL(&hBDN=Jp{T
zdBHvsD=NGx4d;-iC6sIAg*XI-roo-Ya2T1X(Y*22J80q4AMDWWwR!fLmw5B7@3Xkl
zV0G;TX%5Cptd~%&m1s1Ybh{(=clSAV>=dJ-Aj@~zzjc?zb5FCro$x!q^SgB3ew*i-
z1(o>*)gZ*EF{3yp9t<f*Whzk_Ul_*mn4nQZ3ybhQ%F1wRCS~L5HJ*9q0^k0F5BQ_s
zy}{4_@(D)6AqA2PPb@L3+MJs&@x343<@7TZ=I2|CUc1VL+I?)4ac8~9Yd7{7C9vp+
zeD>TdXXib3cJ@h=oJtLvGfgH*Myo#0Y<L`gu=y{o>&J93bpB{w2JLsR<IgU!`k5CX
zaIWD+o};EnmC_kAlMc&zL?|_}HK-{}YYIc!9}##SK@^bZv2$?uo$Gj|Rp;S)o0nTv
z^u%ZX*lCjeZ94ZJ((Ck?dpOU9a~Ij&+-KwNIxjr=ER~rSd%OF*{K~6T8!f*3-S6|>
zyKm6%5Al4DP<gDbtx&I5$+HY8BuSET`}SQbl`3ET^3M^44Mw9j^?Jlu6pVTUtcG-)
zyC|y)m`oEf!tPu_y-}cE`Uc6u2|PS^K)0ZA0q{~vjF1e61Gcu-X}9-C;)J3w+`IdL
z^^FZWoj&977_D8jl{Jr$Hmoxy)^_R(XAZsX99nhwd5@<@JZ@|9(-^-7UWm1VJSz~P
zhm;0H@z`m~9K~lnj^c?aZ08<o<BUV4ixcY{blExBWqoIB`ZL41^JnmtU}t~Njn+8S
zEIV`ONmAe|@T8c+_jZbgy2kC$qpvV-vpF2Si>ZuCPE$B6hRI~Yd+)!)^3oz{tnmr3
zRw7G2#nv`?oMC5b#Bq!=F(Qu1vj$YaDnU^gydp<=9*aw>gh9aG`Z`+=AChJnm7qf4
zOX}Gbdi{Ntme%-FU;hl(@7<-_AEP8R%Mt&_|Li~Kul%+D6d@!J@89L%V4rV)_n-0Z
zyKga`^ceJe3<e_x!wHjdLYiob!ny`tfbt}s=X~Tm<r;54_~iB$hhB$=P26Ej*rSvQ
zNntegMx9@L@r%?OEk>gOy<V4%^@pr)ZgBm^dqk1P#dA+^{@i(LwHoDeg)l5}{P+pB
zwgp9|xpd(vR#(@M%7XL}Q)6qPrpYw|thPu~LY{hocy}G$*+uyhoeQoEGzUeA)zcR!
zH$wc9&$~Bn^3uznVrAtR-A;!zOG&d43(Jd~J$H^g&)MJI<6!%M^@r>9iyjNhXIYya
zlkRQN9SymkB|O~OCW<1~X67jh%OCEy39AWTSt`?B-y)4CoISrvy-_6$9Pi4P66Eyx
zx1e#)FcykDcTKD3Q=}<ok`hl+{2;{U29srsjiEBzVDDf+Yhi)4OHbgL3d#@J+u7&x
z?Kak0LIhG9Iv>8z@h^Po6W6@ns8X%@WSJpLEu(QpX9DLgJ;^Wp;=jSCKl?f2iN^Eq
zv%I`SwN_zwug$pECK+~!<364*ks@$ov?(%ocnGliQ4-IQpb4BX!c2L#g~a#lbll4H
z^WAjCTEX_#0mD(u{fD=?d;cy^Kk*E^2M_Thm#=;AZks4-lIsE`4ef&iW@cKPKD$P_
z8gS?4I;$(E>5azhw%drEO;%S=^W8VzrSsku{`9#PbC*sc10U-t#-o^YG$K$wje3(n
z39{aps2n;vkMf9wqUsH4Z|o4sGJo{z?~#6KjaR<7!ZXjTu{aa)TmSHV9(Dq@du{4F
z1!*doS@MWS3FUa7YB|R!NjmP+?oW7oJ7awepILRZAt9YG2<vn2`hCvI2(3Q3rO9vn
z>i>=XjSqkv&le~KwlMVimd->{S5T=|IDh^k3-gOqYYj@}Dz1nUAuWU6F2nvF(iT64
z&K@>%Zc5|C687lBtwT@7qo^`#+-cVu+%_+09}L)B-yt_+-hA&(p1k-pSFgQ;5i!;J
zNw&8K7!#1?1)j%I&Cg#vMWb2a#<lyb9zRDno-oWi9NfFk>d7<Qx_6W8);&%?xytd?
zMT&Gpp*52@#`8TIjRt||xk$(&M|cuX3bX(fST3x!*|>I<Nx8wwN|nux0hiyHFte<f
zj0|fRj<d1XWvAUoU>Q#muK)fmKCiA31Oxg*&6OJ+&eSZmhL4B@)uM|o=9rzIV>}!Z
zk7LU97K=-ZBw5N%yZ4dm`tguDGX++aAWRJ#KYW||${gN>3t*isht&p8Ny^!f#yCMq
zg|QZ2SX3#ZFa`n-B_*SS4wa=j{2;(9g^UJ6YJNnZLLNTsb1_i78Y%8cL*<E=S-)|a
z?e-=+ai7t!M{9POz>E0c-49q=UF93U_zS%Bna^Xa&zX}?@!I*PxN-9e+E{9}2E*Zi
z?VSzw_xF%OP^*;*gNQsUc<|r>3kwU(%~#Pzv$eHL$*=M7{zFECF}=<{<q9ZKrZ5su
zS^V)H>0AGV=;G(#<g*lBfG}p-+)dF2qq%qgF5O<6cISZe=Pz>c;*)&g3t#5O%^PT)
z^ZM(r^UO2Pu(7$#_U0yeo>LSB{XrjN4Ph8EnZ%@7O0VB148zB2;uO6<?j0u{8)JO3
z<tgc2SVog&*<-vo_xFbxt5S%g6PG;92DTbI9mRSaj%*<wWE|`t5KrM`Aq4eC4J8fb
zazvU;uy*=lhp8jICMz_Buz1r4OkiAO8B=UfdMM#Il$OI0tlJ!v6wJ-dljoXlr^g#_
ze2>q3`X!W{Cz%N9tp&WykoVi{87OBlerRcw4O#=K25s;>4=prVl3}gEFIQ;IE`qh(
zzWO2S_wKW@v`nQ^MkNIc#e|y=uXE<o&+*(77y0gMZ{bPFzxS8_GJp9$`j5GF>n4}q
zeU}TDF7eE>&r_+D`Q6|7XKZb*)9Vfyj#IK!qqR8lcaq}KXhVo>N;<Yk@1N5E{VT1*
zQQ6@L|8NZZ2(j}iGEI^uNaf;iYPBj+RAOmyl{05fa`(<fiacdJ9#a&Ca=AjeRK@oL
zd@sQFBTk<>&67_*N19A9Hg}YQLsGo89A*d$V<<@Rf(Uu>v*f!S7;e(H1sjFO%-k}K
znL20BpXR$i_zthU@)8Tn%j5<M<FmEB#nr3t^W2Lur_NmB)T#5#&8_jFeUH@>Yb0^Z
zjqA6#`r!vu8Z#UZBbFvZZjMImqzSioHaNDhOlf9?Z*T2!YOdhR&!0qCgHk@@-T;77
zDZ=0~E2SqfwwN9$re1@da)0kBf>McmI6)+abTDwuOQDfLK$hf`YGqpU3rJt0i-OK@
z%<uovkLZt5cZd=eDGY;$59r;#^@(d9g_6K4;dz2Av)~1ICCe}V(!asyKJ!J&QJu(F
z3?~V1y?>b(pLr5d#EklFiXwLRjg{`>DDU%dh~_f0!EvmtaGBsm;SOF>f-bVDe8M%v
zsA=NF^n@s^<iqQ?K*ERDuaM^j<H?XX9dqK;d9Hr2MY&p`*6>-LUE!^_U*nB8UgxQ&
zp5n!qKhLSNr->&Sy`47E@(LASa&XXQcBYA6s<Cmf&-RYw%tggyG9t@!;xs`@Nwd`=
z3Iaq?kY^K0^$_6+v{WdoIlY|nN4IWp@(Z8m*7Z%^eSL#d=PIOWN`GKEdts54%80Z#
z=8YfivUFB)>Cy?}+q;|-oA?pLh33wBo7WzU`LLgJq8ZU>c;s1zDRRoyD!B)xN`qTh
zZ;<iHr!adP?~)}6zAy1SNuCspM;;sPg38P+PdxDy7tTFHRtUmU1?f4CtnUX1ThKY!
zVLaSN;bP#_W9Y1#7U9rV`G}aKjpIUFgF~vPUX<=WnTAXXdz8*}@BStRxN_xvM#C|q
zQJ<`su(-U!o!cEeuf+1oEVFY}-oN}dZ@m5*Pd)t<pZfHdIC<(M#t80xc#G!AQ_wDP
zu(gBk?X$MB%Fdm;{OG{}GspIri7a{&Q?1k~2f<_R9ujF>x`mYzPb3gFXjU^mdwiFF
zdi8xSzWf=^o`;)P9&+c#09q3U@qlMvd4h-IeY|{tec0p~eUp{e9khb|hg+PebO}8{
zl4i7m04Y5N?GB5pEAE}xn!zhYs3@RWuR`ggCC7ZMrY^-201*mf1(QynJFmUP(@P7e
z<^n~LB8mbPYL?r50y7~KWzxX_o5Xmv8kL0^q>>boM8^rqUYqj5JYl&)Z+oBSOcg0D
zpZU}=bed2#WB$UL&$l)ouypDo!~2H8s6#xS(B9w0uhvn1neFv0{_fxU|M0^<_yeAJ
z{smS~pGMk@wdGYBGjqJ~;-^qbaqr$ewzs#bRx507t+TMO#I+CJMT!ZbpW!PQ4@Z3P
z{uN$%>5FXcbQtV!A>%#Hl$Oy&hRi`#D;O&XCw;Q_eivJ0aOPz&F$Uj_K6TDyIOOib
zyX@?5GuNEs^RIrH+4(hQXJ>iy?bq43f1h~V<CT}5XFM9Sy}3zISSCrv*6ucGmQrX<
znr4hf6Yku;%WP{76?hm4Mp`F#=Mlp4)8f+sNGb4@aK3Wdl4Y4g3XZb<rXSL{mzi`b
zO~Ii-^E4{P8O~r8WI83!GO$kV=X-*aCsx?q+a-(w5EF;u25wX?jHMh!jN-(pdkYPo
zz>^Y*L+|V%mFaPhy6H%#S__$*oo8urg(5c$=7!8P>dekI@l?R>ZkMW8!9Q5%VfTOn
zO*PXzxe`&b85W7B1fl0K9t>&B&M?U`M#DZ;0YSA+Yi^PG)gIdq)@gUz3<rIh^#)Mk
ziJ2L0?LXwJ&ppeH53cger(WSd`m6tY{>g9u8n<rV<j?%_U*xH0o<Rsjz1G54AqV??
z#*>URGZ=k%x}7SG4@Vk@!jMDMviRr9_ph`b8ACrBtK$Gt<kaG1ILUap{(z;GGb}G3
zn|_{)96C7IBMd{978Y3FSSL#h!Z0K(hZte#4F)uubDY2MB>M+_qLN3e-XhlpNuEwe
zH?C2&he}fGng%($O8s-+VEESW;csnH7)_<#AgoBDN(o~#78X{qM$m1iNLc|9vET0T
zz3=@I2M7C{K7E$C`6XU_@g<HQTO*r_85WkBT)+Au+xPBMFITt__Lv#<xHU<*x4q3w
zy~SE<h2Oi;=IO;H79xqx3Ifk(G#XKA)RBQtQ6x?nAPmZrSglc^hxOF-CFVwyLVARy
z3fiU^ouTsr-5!y3N8m}~Fd^ynNVA;F*ALj=8If8;B!uH88;w7nu>amC^Cbibf)Ywt
zqM$}?b%}AZP5b^fyL)>CLCD7X1{cm=;7@<^YwYcAvi<N5fpWqfH}bb=D-gOs2p=^)
zkU1ItAF~)bJoNJDc~#=kl)|Czhe-s^XK)gyZ0{U!>+WrK_ID{)N(?7sr1E+2V2666
zhGxW@Z~q~aaYnn-W-uDDv$sdB-l9@2@yzp2@txoKBbp0KRBL5Kn$X>Rz{=_|?svQV
zaA&~soefr60WyiHl^c|NkH`-mvAnd)`*jv8IrtS8n_W)tZ}Z+8A9CTs8rQCE^4hoe
zSUVTe>!zHzc$^1&9g?huPWxB{>sPv5>fWSTxlJl8Kf1Qd!<_*QU-E3d%8M84oLP+6
z+1Ybb5}@9y)9;Tt@x&>TtbiZ=%SUnlq7}ypwMvLokR}-x*c}!mBH-2Mp5xE{@?XFY
zE4=g06<&Dmd6e?#_4+ibWs+f=et!?8KzcrsM~%Evhlxj)qkpW$Sc9I%TVkNdQ+F7}
zM@TW%OiUYjB*wtQ%`GPJnEUJZvBEGI51GU<$T0{HE&9Cu-gnT2Wp{Un{%FX?);3c5
z%+4)w@rePQ{SLCoICcCCW;|wRZ=2a;$B5=;c=Mo5@a|oH_SsdMK}fw+#>0|j1!V+U
z7*LknXr%8W3xSm>o+>lDobn6aCa=HuU52eyj-NhGW8nhb-hg}euJhXO{E+7^T%dZd
z%ai6lXXdUWy__sh2}MS&;*(oVkXy3cVm*n?azw5vhapiII_nw%Io9Pve}wfhW5NV@
zprs`j7NG^Z*B@}~`VGQY8wgXz;xUV5UKB)mfhIvtAJ)n49wtkuuPoq~Li8XY9gGk_
zso5k5Lo!`ZuGR2-MUh4v?Cx?(#r%iQtng26?K7^;B1ObrZ;w15Qe-KUamIYP#L2mN
zwr*VG-i@pHVTsz}GK*`g1U}rnb{P=t?Cfyj<Y^j>7V~qn%rqOUEHtMmlHp)yhxgyV
z!pg!K(q6%+x6R(Y4|uu>{rh9`lWRntk0(JE1z^#W5uNY-1~tEeT{(@gsq@*<1%rN<
z_TDbO@c7E-zr@P?ESX7Y?`?7K_BA#hJm438{+paTcaHCW|8)Qgt?|8pW5-T$`}Qpk
z+Rh$ANe^QTd7gv!C=%*%CF<h|6F-gdt2F{^BvL6n<<Oy`C`gjTd4@ff3+gBnQ}Q#?
z@r$GaAu;li$6Zt^Q7KjMq{IsYJkN6a^l3ohdp8MvMXBU+BU4>~ws;tVz@tbs##-`1
zqXHl4In+?+`82}nF}b{i*|`?Sj;$euk1hnSe&!2&>cyAY-Q8qw`ymhSZ!;O^)W&^!
zI}f=T8zN6}sw#M9u|hJ6sa9*q>9tSegvlf(td_}=3B!JmRuJGt6^<W2#i-Y3)a{dK
zLp+=?)0`nHS6M0Uadu?+cfasFFaNoJm;dp9{1rB~Hu%eb>96qAGtVQ0<@)u@{9pdd
zzsc3BS4dMsnp(8^lf3O7Z`u#%-S&uN{z;ZU9qAq}V-22AN2bJ{?~&yx{ceYwH$LRV
ziBrtZEU+-Y!0b$uFv!{3bZXzjY-Ec#dbwAsb)I_SNh;-tx8Ht?o!xDg78h7sJ4U0~
z#18_;^f(f|Fdgd{v=FG3(==cG1t#D7CxlxEl<Q^UqQ}AhKIKY4oKEO=Cip>%AoLiG
zd*nqyNf?x}+`fH{>o=~lzJ8BSfBJJQFRgOW&#2bs_{vwm#w6?W#vgo#vF}mwBWn3J
zah}o}wkfid(&|b6yEh;3AARl=&B$Xi8j&UwgcR7RQM2-WR1{&d9FylnK|m3BZgP=o
z;@+50dZ;KuBFM*vOiGk>_pnJdHc7~UiIi;arcA6yj;1Cg0)ep}p_J_3`0x|e8=zFF
z;Tc2V1x)e-a*<)I$7?@&omXD|EKgi|no&FglMtx_TjXe|rT~mR+ItS`<&Z=*_3v~X
zJIjyzgdTEm)pT47k7A;vlxU{>PEXM3bcxe3x9{I1j#C=V2BXOsAr$-Vdpz;PQ!FgZ
zv%YS)b?X+!YMRYDGcz-M_q*Tb-};OH8|LS_JoWTby#3niEH}#Jg{C_klD7AlU0vk%
z`Zj;~;U+)-+$syDkXjhw3CQyTfkdN0OtsAhBnA-}qNvPM$M)&nxsMJO`I)c3z_rT{
z_~5lIf~Ax~7Cia<1>X3hcL<k9<omPCwbwbda1|NA`eui$)8>Unj*tOiwLoPZlJN{P
zv$G5b1JXF9()3t4wn~3EApa>G+RLSgdb32D=H!`W9A^w-!Lu)Zim!j;o1DA&1e38w
ziyOpw#=_z(MV8V(*dmVyV2#TqwdyE&NC<JX{T$ZUOno}FdE69UyJ>(fC~|j@L>{^p
zxa68RiRpER+~0UWZ_uYwEfc3HQU%<-e~U{`USe@^iB4yq_b<OkQ500GW#;GR_|~_6
zi~sOH_)l0^SmWuZpX5j1{}Bb6?r6eLYufi7aQ5O^_J@70?kB8mC%k$*z?g)>3X&po
z?<@qu2&8uRq*1P^7b>OfMX1F#U-c%ew@2K&wom_Rh?SD&ctqV~MD2TgzS5@I+(Db1
zgS{O*YY2Tn1^7iqP%AOY3^FW(ju{UoG#8p=qamK>VF9IGR{BR-ACpt(h>=2?=_akn
zh8Yij@ILd$k0V#tP{OcmGG<y8j1CE`#Cj#ds74XTq-n-@kWjAIkYR+M7bM*dN_kWp
zb*y$qzJ9ri@-?M6W7IvMI_~gqKfS~c9(MV?J5X9a&HaO2dc!ukHuTe&gM%%WMpatn
zI)zuz>5X`B_aX0>>a<!-W@cK{YBl<u4&{1@YNf;^o}gsJU@&Af95I<BEG(TOC^Y@`
z4|%Y4o4WGI_6@cn(OKd)DytDifi805L7(x#9{Fp($>KNveF_yg<(4T(CnI_X`+WA(
zU*zoZQ`kIab7PYaFTY7=Z=K!khuGrNc*3cFgb<9!BSzyff`IWj=H`tX+`fB{=bw9l
zFMjrOARtYWqgS_hEHdozF|0A4{J}Ri$C@0ZMG2=&l%8|EkK>q96ybT!K-A4ll;cHB
zM}h)@Mj<^9p**y)MCA&zbBm4~FABnvVrF)RT6LB;UVEFYnIeQ`I2<^XMoNruqhKW^
zg*Iqw$co&Z6s5!pfhU}Ss~>qxCMgIAgEEc!0w+$K;q007eEOBou($h=Qc1G8xx@DM
z9xXqhyM2!jdqb?Wlq@{C8iAQ$grM1MQ51%~gAUEP27_LgnfV2RN`+y&L%-W*X2Hjg
zN}N7-f!iNmrI;inW5d?CLsTiTbp8bM{Vkq3bAj*t@Skz@=1qS7=YEM#fBIFdwOsw+
zUH<0Z{Li>@<pYMpF>zdA?7!Ho_~@yej=<FvQOk=@*35f;i0}JHO_hD@C9IS}D#2(p
zrdF%7x3|r`ySHdH=LiBHPX(0AO~Rl=p;MA1B@9Z`YYlwAgzp8s^r=@U`972WfLES<
z9xb0^uXDit2ltSh4|(?47ydbW8Yu+AxXWQt6tk<?r$0||<M&ZMY~EidOD%CSrYKr$
zZ?y?4JvKJ)a_iO|j4=%QLq7cQDy33HYqrUQhqrj=oi|a+=lt2T%+4<GOMmLma%^>t
zr=EO?Km68j@a`LLF{6Cqy=^KbMLry`vAaRIe4M}i#y0=kPc7rg0;vQ^98*#re(2+u
z$_P(^c4-vK^U*jH?sPQ5^e31o!mE`L<p_izOLNCPv(xkh-v_1W_EPqG1)1~j!JqPv
zJRylZMYs3K?Jbs<E)OPGfB*pSs7XXYRFRE3bO(3vqAFkig+I?{zdpzE${J5S`4r7o
zlR<Z%VRr{iina=AB%W~5kcR>tV+@*{LQee*rF&;sDUJX;YaJ=Wq57_&^E?kdz0VH!
zu|p27KyuLTF&vHA+S+tciG{OR+TPltHyrcrZ-37<1XB}vftVgb;Odpj{Lb%wi`A87
zW*25zIev_Bw}bJ0VkH^wwRz&HdAgzCc0cFle#Wu$Rdf+k7)x#~RteDH8TS)eCEbCS
z93?#}Gw?<6kSq5MH|{LbSUJUWuUug7pvSfA?~@GsTsn1<yKfA+FxcYh#j8ZogiPn8
z{R38-DY=#8qYOWCQGMgVkcE|HsMo2_%%Z}Guv}uLS!b`)|HL(~)<QfL;8)6Ic^?&q
z$SCIvU-%-Qdhr!>ZdsU_WqEOt_ujukF&QFFiq2ySeVDhdDgHP*%O6K)e|!&d^%p-j
zSvhRvoo$jr07;UP<_Q~H4@uITN<}alPbl(?hZ~!0Ztoz3Aj@*{+>vmc!-ZjIdxL-U
z>;I5fUimb&R*Pqzf0oz3|08B*=jl&IOp=^~-8QvWlg|DDuWj~t`uGWIUXB%((1v(O
zl(ghRlamun-!(_10$C8r2GLwhR9ojnvOydN6mf{1OYl^J_4^13onglQ#sg|)OQllA
zGXc^lhGWOqo3AxkT0X|!#ulA}0~+(Q*diyJjESO<EX~N${3F%%V>NXihEO`c8H|&h
z=f*PL=yLZvuW|MlzRIkQneTN8MNSq3c-C>81U@oUgn>m8P{f7fk4)8HlTMdlwt-bH
zg&_<>jD*HalSyx%LJPtq;fp64T$rozKfQgItP(NTy1;&Km(Hlioj7Mb?lLp%u^0wa
zgEDiasw3P+ir!Aj{`MAWo}<l_6M;r3AI}R35mdDy80<1>@367IL##Yjluya_sg`Q!
zB1KC{9>@5eBF{3c(WH4yW&a-P-~oE!3}7&}pj3)Df9?#ktp>O6++>`L>Gk_S!O4@y
zIeGFpLTGMWzebv5AQbl=tn>QoZ?dv-f-s0^wPpzXO}4kUc(DF}TCL7#GNIFMlcp(3
z9)`S6Pr=h(E-OCyJkG`wL=<9`#Yzv~DztDJz6e1a$5?AAM`etd^1Y@;qVn+J7SciV
zo==f${19rj8c&>iin|Z4vNS(SDezfYn&VS1K1Zk1BaTPhyMKrNU_?}kC`A?8`)yS6
zC{EQFLzd-)p$h^Jqkv;4kApDWym^Omsma%V_80k6zx1aOz@1yyxpD11_V>4W=k3c3
zJAKxwB|7U5xY_Q~GtlyVJOK->7L&~#T8p!UVSqM*{ezg9g)*bbL#jywF9--qB_{D0
z)oIh3U!>NWVd2;@w(i}am>Al93->$J7iSo@*7#93=lhr6XaD{>-~8&IAswdNy?>Se
z)BpKz@Bjxv_`cqI@6zoINMbX!aQqiHXNTNLPf4O6APfQ!f+YQDaWa3>@&XUv_m5t#
z?z}x!G)`?sM&k+fMw3aJaQn^;jvYVAv9;p}ltW*nB1v6@U8z)}UT+YFA)e<EMkTDy
znG8qlx7(b&c#bD7J;mwMXW8D`a#@f@A1URgV2~T>OM$c+fkf37@WK+EUXS(7ErPH@
zzds}>In&#j*#@=7Meg6<#M%^NVbGs27>qdRwDG;bc}eC4`}^z6&o7W>DgWMI`uBO_
z>1SxP>I9_<KltwVc=oAh>Fn*$&o!cmd9Zzl==4+k#@z|ObfS)ocNz8iR4Y}S<3)%m
zzza$kFGUwQ)^mHCE)0GUU<zl#EqxCa284|&Ce;)sNBRO8mKn!}cdk63$Rvd^F5FuR
zB!)=221Dsj{+ydUW|DS^sta6v>EGbO^Iu|iewB(}N1K9fdxv<?Mp8JXzXhQUOn-(c
z8_Ic;O$A*7@I3EmTSB<Q;9>f}<Gz;wN=zHF$51;d43>;er_J8}Hly(fm!M<V+dH5?
z7*qU&1FWM0xb&f=l@+evxXPEl^c9lH0Amcl_uIeALVb>Lr^}8>+1)*$(QGk1=<x1N
zk1ss2h_pE<OUe3pO1gWNf>2nLay&t4C5W675&lfd3-Uh42QfDucG$gpl}v@yhGS~m
zebj^7d^zm1xcm?)GX{emCeJAO79|71qJU;dhQJd6S)AZ$OC>5%E|mzP5R^nqDAlT;
zxaLZF1WKU1h?B<_fevrqy~n%nzt5Mx{Im3XBg$tgeC6}6u(f%gWYi%HoGRU+@6&zi
zsA01RB<6_I?MAYP?<VKd_{b%}_dO2Fb_hgBDZyGiB^Zrj;xu9ZU>7q*s<Jd=cW;j(
zHzdgfrtE1qrH~{^&Ye5AIdSp?w{CyPSHAKyoIiJtTeU0PzxRN_rr~&4<xbjTx7TIi
z%o&EAHXFkp`@>n*!U*AoNN<`!s{~qtEi@hyW8I`kS%r{bM23(dp)Uv`2x~D$k1<I?
zZXi!Aw>Cx$2LmphSVa_L0xKwNhR_;asi8cdYPEtNddw}&GirC(-`ZtiZJ8n~FnNv)
z!;d_F`FKs8G=x$yGPvLuA+W+AHSAuw&Ftztrxq%hd`RX2ML{uDCrYK<=x~x?hhuEy
zQJJrkc^=Z!D3epfF@!$C6eN?Juu`Wmmip`>lTMF7x{JD2g+Ke;GV0nx{$RaJtuo7E
zqeXW-pgSC}pJ(jm386?R5BeBSQr!tznP1}A*|T)IF~0N&LmyxHWSJ%I?_&qM#N#m!
z#xW1mlvMdV*NCVNCoC=0C`Ud<=XhR~JV`L6$VFEb1->!VynuXn3$bw4S!!5Ct=gp3
zXmI87TS(6tZ0;W%@WF>y84L&bfzPFjPts`2V0D3#mMHX~$e0W}gudj$nRA?2JMJ=2
z^MY=-OQ%2J%{SlV_U*ezuYUKiJo@)fW39jTCi21sDoeB2N*U`%2qjTC9-H(+;?Wqi
zrBwA0-l0vgYv}A0-8&@8`pV<sz56W8&-3a_Um)})VKrqq=<?(f&v5SC1wQ=nL(ZH&
z!@|NU>zf-GV`#NnT*==f%L=kXv(?9bNu{7Q*QDNP(e3m&dEy*j`21IS?uD1BS1a7P
zeS_Y?K3f}G+`4g-MzzLb=#g#SA@1!lDKcv162u8-=1Mfvl-%>F)<Zl`Agy9D%t*)X
zr8JpLsE7(ed4xekl1v!(yHuO=EUX@*-|H~w?~z-N2ipb7wJlz&b$NSNbN~8%q_K4S
z1Kxb|b^c$!@_*+CKl&bnaYCL57j5?uY9oYjM!r!*quzAcn?kYo5i<h+q~%ZfbdN+a
zhopCR!HdFq`IVx8O0~k)_8zxxU1N4;mTINpxRRcvC^Ww3Q>j#mO693J_;l2!0+vsn
zpuRLuP!17Flcr;oCy>)*5!WtEeew>y4y_9V71j`ye4<i`@t{wq-=ng-O`7KnhZ6>)
z1H^X12OoS$zu!MnQ9Ev)z@&yWj<Hjus8Xr0wzf>Wv&rB3mH&ph*;#s>4ku2X;<*=}
z;re^mxOCxJcK6rW+rCd059#%GxK%pMKfbZaFC42vo{@})cu`0gRhXnHVYx)9USr%j
zz!%PjqCDS%afTizjx(6MpmQ$@(V@lK0FdOV$K~7WboLUgm5j)UWPm^+g(mbQnU;*R
zPyV#8Z@xoo?i5db>X*6j!dF;WS|ISsBxy=_e~Wb7b4J)gIP`p~Y&;ZZVXbSt55*Be
zOl_B3(#+xKIkmb=k@(|6uEVc0ePy_MV<>XXcsyZ$f1kW?K4xi_(;p0eVngS?m=pFL
zblTLLHTL(nxqjmsue|(e`kj52Ppz?gYs~R#oql$}UboBQ=~E0kZFWa}CgTy2gs4<Q
z6b5Oa5YA#lS;y7IIEyKX0#)GI3SM(UP}yLvVVJ}viiwAvH^3M#nqVmy#X0TG4HOBp
zwHjfhz*vgh5{4yotf@9;7$<wAaYo56k&Q+u&%+9VBKWCT7tPGCk&O=s{W5vc$EX-T
zD0B77H7;NNfG>UVD`;ELYwx0y0m8Es#z#qok^*fW-2<A80)xOiqHahbjz;s3jQ+)x
zUp+<s#bZt4A!SH{u_;*|Gad~Y42BfOaq{-|4~Uc0#XcVWCWlR>t9h?KptQfs!h+&k
z-}+7d(qH}$sW+OOy>yD-y?2Y_^RvY1gd0i1!9kaLV}`8T=Kk)0^A%0CrYP5IWJONk
zg>Kqp3>JmSQ$Qf12+368Zna~OOM#~qMWOLT1yAIxZ|t$R)8mCFmRXvqGZ{*<EJhj+
zUq~8Z39CG$3doHiY}HvleS(c^x9~ilrQ@rJEGI7t=VAI0*2f&bx`$93i(%?Bg!9Mp
zCkAu#CawB8eBY8KDXI_{31Ou)^`ub<ZHUJc)HLJJk0RpH$eo1pVoHwIU>r|M7)w}+
z$it9KYet=v&VC<_;AfYcRIz;b!47*`Iz`>$5;E}VO-6JE0|t`>Q55*v(C_Rs{9r_$
z70e-MNJ$nb9!@4~4HIImNwq~QK|@KN4m?_uF;6cko?g-vlP+HKn0uKjACqhHq97el
z@U<lH1h%t_%?Ch)HHJaA!&|TYh<mrL;`wF#QUm3evAIVwj@jATrV><GUMWZ>V=zw9
zndK=v`<rBGPL}0lc|n|{jK?E-{Q(F4p1awPjFo@V`sCx+w}0#R*?8?T3oEnCoIJ+N
z@l`5I%lOR(z6!{MAP6I}B*&_liXY-!2`ol9!%2iZ3U77>(T09^o3*tiu&>e^Z7}Q)
z84ZU#_2e_GtQ==?X^m#Hfe?y2x9@RqaDdT<`1FV${_sumESsV?&S*P|LY{r*MP7XA
z^USuE=yeVl4tsR>9`J5^i`zGDbL-kogef?-Jj?#wJEZLc>dBb0fYqfHdI$TQ3>05j
zm_d)nEF7OFj6$?=kwwG7fHWPTJi#~~;|Bp+Ta@w;RxlpNgrbL-o8|PGC+M{&*mg=+
z!~2^(u17;2J{aO#OQ9s+{o(g{<NddJ?ML5Zk`!pe#~PU{3AJjOR&$2AnFVHN=Fr-3
z`NI!BQO#ZM>(svJu?8;)0zBVmG#QcSIbm3udRA@o&U<gNvU-ej=Pr>KDN+iuEJG=e
za=A<tl^+vcIP#406^(KTSo*y#!+wv@69|(dgd4#)zd<M1;c#q&<GNJJCCVkANm3<E
zQyy-vv%Ip*?(R0X?%rlJN?dN{<J<qCoyH;WXR4C!wFlg|dV^9WBA!gpML`sl*xTQw
zS}yUaS6|`7cdl^m%mu8`Y;4?P(C<*ptZ+rvIhQG(EhVf!*kgXFh37?Nae@j1Dzyq?
zICT4)&M~TBH0dGIf^xk9DnMF+kdi{@L`g=kuymVqbQ)s{@o;KuWC@i+>6CN_r+!wD
z7az@O{$I4}^XEA8^v`kj*)K3Zzd~6>j3z@m2is)v5M%XJvFgxQC4Hwle{8I#^`V+|
z%J-ELVeAn<@{nIC4^5>H4-3;nsIdZ#69OsuNIh$eAWbs*gAp!@%p=JQXE^>7xMji>
z8jB%KEDs)RQmYO4`+xuMICJ)N!s^-rx36C2M5WB)#OGd|GZ;(=!!mg~Vf$b}xhbjE
z>ZrgcEAlDQ$aGSrutkbgAaOP>M&=*`$2$!SUM<CDLv&V<YC|5U+}fNF54#+jD-j5v
zQV?Q9fi#MdwJ6`??7|Aw#taX~Bf7l-)mDQ%NeI#ud>=iI(Vx7h!}H5D=g!gV-KP*K
zFMsx{eD0giFf+T%v(LRSr7G={O}YpQj1`VnH0ADE`4~E@4F&qh6~P`xe+e;d*qG*p
zKN3_Nj$9wVhur>P(b_T`Oc)GDBxwdxxT)YIb`9N6*e30?=2}~}xAvJ#QXW3oBn%_g
z)=m%wA#2B%>EFD|v1*x(JY}!b;mqnW3KVzSLpJN0i;WrAT<0mG&;*{3)-H7-&l0>+
z)zOI#9Rl(ks|!q_$&(c6DGF(Mc>6v(?H-p-&QOzMvea|vu2OCqK}rxtAOZ(?B?;w5
zjoRuW%Y|X{_C1Uc96xh{&{K@!_@mYJ<J<&?o3cGrHcmf`5TGQdR$3^PQH(~INj4Rz
zYjmDds#Otzj|>Bp0<Rikm2ljsJV*M0Lw~e`X!ATH63(ei2tl=8qcEC0OBu%*l~$7|
zEco17nH9g_2OAUaPGb58V+tiHS1O!4zJ>@q@;JrU8k1@IaY~UH5>rrDmN*k^PBKQ;
z;t4@1P*hCLtF4gdXUZ&93aWmB&QofwIl{p4jeO-Hj6q7lWHci50)#M#{szV1KDxXH
zT2KllXHT8t)`#!W>FyCk1<NZl%+JhWt^0j$+_=t->$m8|IhhfxZ}0Jse*2%g=%Oj?
ze5fG$$*NZSsIU9!h+dCYnnHWyfMlmj=YzWhRgd!00?kv$nLB-&($XAWtBRDN8^vm9
zwq~7(K|0h<$w%HrgJGX1o;<~`|Jtu2sIV}10>m1f-68M4dzsl;_x~1o%=Xrp{evEF
zzx^JUE?(l#{keaS|M?&N9cE{mcuJ6^DM|?D7nW$XX85ar_5VP#waBp3=e_s$+1}XU
zgDYF?Zyw+qORE;)89{${lb!t?y#4^yYLLv$)7#kQQs}X04Qq`Gfs#}jP0%?4L%-ie
z*c7Ef1R;|oAx#oT0#^#_ORSRgmd`Uj^(6BCEtXDn+1$E;76tbvDVsZ*ys!vQk{ZMJ
zUVj}BjuUAG9}S;-w8Ai;QmHXFH_z(IF=m=G6vap5UHprd7y3xwZ4Juv@Ra)q0w2%w
z8IDHIY9|WuJReU8-hBQ0TsnV=x%mZ>Bt{^K<B8)nh9!a^d<?~rkC`Rsblbc1yB&9*
z6c#H5!ulXR*T5--oQ`cABIKg@!Z2WYWr;Y;2*VOP`@2-?RgxsdmuiYA{mIw#u&Gee
ziF)GEgm{u5r^r$==`(!)G7HNKJpKHqc>0xRxP9#g&%N*xrD}uS2Y0!9?+P==PxD5<
zU?Gy+xYOo~3kz6X5SBs;VW1pQo@p`ewh4s6lTfM#^t&C3bWBvMBZ3k#C?P`+;|Gkr
zS=uuTG;M%*>pF2eMp;Wq3VcxzO6SFxX+ic=2zh?;Q@_l)CqKpF;wn1L*&lQm4m+4U
zcK79!M~pETttqA`uAQos;m7$=M^-q2obKVyC(>98CZD3NQwy6Z>Su9g@Q6pqix3i{
z1x2BW(+S`>cX1s5<c98WN?{LQ@~%0E$1&~wHVgB!yztyBwCXM9=Vux1Xx5?<50i|Y
zPMfvXEO8ujePfTsXpRLxpd6J*CS!cxpZ=~EV+x8a!K>8ql%pI-Yo~Tf8euJ}u-HQ3
z$pDolynB0-ev<Lx#Z_kO5$)X_{J_J@6Ffh}DxV-Kp~3(!@|jy&;NkV#40=PVjS5+o
zQNkdUM>3gw;+j{Z3ZAs&Xii-GD$joImsvS+o@&@cSm?C3NX8um+VP~p)qmQQ9a(oB
z77ORgt_y?k<d1W5g%ivjHuOxzG0rE!Az|_`#2uG9f_3-9;oY5OIr`8l<xi;lq2g9+
zP5YpS6!3e$_uKs3H-3?GXD;yEv(NHJ4<3S;Wob0#cAC-~j|ieNo8tlF%p=JXGOekU
z>!j(3Fbc><6DSPUcrN*7`uplUM<+2wnxS)p@_mMrjQ1};AoNn^X38}D98Xm#^8B%D
zV4Q$ZxC6A%ZVyt{c>_l0PEx9r*}Qw7Yad?Y#K|?nO64Qf^<y=4Wlw-m2va!XG8PYs
zG#-^EEYu}Mr-v<a=OZXxIf^(Y2V@u_A`idh;Ya?I{H{P7(mW-KLI-L3K9h8W))~?&
z6qej-{J>-B_zJVDi!M`Dy6iMh2FykSx|I06JA2&j=h)te{q6vxe5~i;g&}^WLV0!`
zUquL^*)s{FBqtDEYLf{%EAWNoo9E~F+Om%s#boIOgQQxiGh1&_3QG>fFdD3&<OihL
zn6Og8&jf{+5Z`<ko}5AGDz$Qn6Q|CzGxXSN4++B#V|$M%Dw7vua$!h~CC@Sn3*aKK
zMoIc{O7l~1=8l$h2C73{VDPE;3yQ%u{X5%igzr&WsM9*RMC14x(edNxav2kkm|0%K
z3rip}tZ-STjx252+1=;-`3u~-d51sx-nV$+xmT(9bL?-78SM9osuAPn06z%G5>2xi
z((7*%`j%72PxHzvud=_t&F$N_81;KR`P8$VJ$)Wsq`dKiKj7Mx_qciaGCSLC%DzWU
zd2|OIHt+OlL=jn0&`&ZJR!=jkR@lCGho?&(&tfUnLS{?Q@FT*|N8(|%Axn0t1{O~g
z2rr_qM`YT?b>)75H+!7y<SYEa?Gdm4?dv>quF9*&&N08U$$K~Vxi`)^(3VIDgf%4E
zOjAG(`6(a2fYu5EKcvxUF*`HE?9A++4BCE-6$Cy(7#_WtJ>SRk4#z5%et&>c9-bE<
zk>v4^t%rBHed7b3f8j;)JOyD%(*#e2RI5#bAb6zG76ybOO=I@=wwR1MsHsZQ7y&4^
z)laQjl$1y%5yF7hQ-oHagyQ^#v)sA2POaWzeRGTL?LFq^X7RiL`zKsg0Y_@R!+0Vg
zrlTJrz*zSJ(FT(hM1v8R-+za1e)E?obWVS-&BfEFnG|C>>6n}MZgJ*|pW$KJ<=IQi
zc!9=xE<VXxgUAZ}Mh!KYP-Gh42*PHQa*{LI+b2sWgn>sE$7rqCs?5`!UqA<S8q0#&
z)5mBZT*DI{p0EVU;wwR37&IOm`ja>EYp0%IX>N&lI3k`5$nz0KXIKN;S`@(4gVRk@
ztP{0KroMEC<5P+A{dQ>(T5GgYC?$?+;jAqRtWXZg<IwWrkb-`Q@VGIjMBxWz1P&=O
zm?Kp1$LcDkzLe9ia+EuYz!*WL>haDyuXF0eIllCTFOlR!e)wDelyX$3?u}`u2~sI?
zrRe1vOF>@bSW%-~s-ksFpd7Mf3XQ<S55sALYB5ED$#SgL&Yq?y@S~9NWXSzn>-5JX
zo<6fc%}ddR!V7#*0u=;Ep=d@Gke>5!lY;8f9JBoayAL;s&Yz`P_Rw(xK|mNqpSb3G
zJJ-M$oILj>p8NEl<JgG{)GIZz!qRDPGwHW6CdU9;7bqz))-|t3%i1Fz<RgUEWw{>K
z+Kt}fFiFG-+vJa_VjUIS?F&*m&Ye35I-Yi+HRJJ5j>!_@XrO!a#Jk2`5JVnQ6}<iS
z8~psw{TZs|XSsXhDtk9?vsf;(o@Mk#eNHYOXV{KOWk{AK6goxAGS(0Aq8fUfW0K5?
zpFHLAoeN7onUKdZMjL{1iNU1c+dsI;+I)nN9t*7!J_VkT?wRq_G&#j0WZ`5$Qc|k9
zgdi(jR<e?k*%K=?=V#g9-DZ1h8~qW3y+2k{7pNR#<+S;j)>MFzhQ--BWHDJ$IElKH
z7;Bu$$w)GtA;txwPy}9uB9o{fa4I_?&_xbJjx4F%1t_u%PeoWE+yhf+0xu#c*9bfb
zg(31>oJTqGIBVgFlh5+qn|ti{HSKPXQI<d+(~ZaU;xTG*hV9*Zc*191MTFx%7d_?D
zECQb|Us~Xdw2U^}5EeA5H7XT%!RxgKV(P4Kw8j?#F@<66RN)m#kKN5JN@x05FQO<k
z-+$|EHacB0Z5U4yI_nQz6LzEq`#4qjUv_y@rJ4|mfzBxiXh=y^2yAW`Puh&v_t*|E
z<IT@7dt#N+xl?56n5APEk=3%3{^NKf)mn|y7oVlm8F1|QF$SXnzwuB00SmLMJazFY
zW@no8cSf|gJ0LVz?J^;)rnB>qnR=N~`+(3W&Mcj#7wogQzRCS-clp*o{54FLBTb5*
z8ct?o#>15D(S#kHI$lPWFf+Tzsbgnp5BhA~zsZHOYkd8SpC{b8i`?J9+JxEGJX+^e
z8V#&~(m}*9fXpP6RoU?ytwmPm@t%5>_r^_r_qT4azdxc}Epc<VU@%-|t$d8lbcbO+
z<)0`=c(>+ItM#wGEUu|5g(b?RGWB}>$mI2ttVc>Tek>}?Sr8a{{UN>=;&}l@uIYAm
zu}RE3Z@tOd$+Jv`BW7kA#BoBaS*Kj7Ocm0H);osLU`Y311D#LsneuWhC@%n<np#Np
zW8dFF0|H^E$c~h(t}aunl{s<zBsXqer`zr0DUb1F@+X|=53k*nwqWfM5~zegcxsy3
zVjQ<oK)XBO-S;okY|ZoH)6ZfHjY$lrk6$Fv0)PJ=J)Ls7uX$pr#AG<8KGQ;Z9;PTL
zv?hoGO4S-!>I54`3aTrM_)$oaX+#*1rwQFyvNPKR0s`rg*Q?AtaT4CYPL#=`rp{Qy
zsBn$z1D$>1nwNr*@o-2{<Y*lutvj@8qaBjWlmq2?4z>GZyj=%iIRy4`ghp#k@y7*Q
z!tDqGbEJ~AQ>zp4nAhc@GFB+?JVBmoe$o^3D5CJuXkAEw#Cyz)9gZJgqgt&o91mGu
zJH-p+y#0sY!(<urp~pdyFv?P7SZAmQNac|iDVhRRDxniiP^uy=<VlWSDS>hxl16Ly
za!*r=tZ)iW-{bn79kw3s5|u3Hj<+b0U~NcV<nH&A3Qzf1EPfDS44A^8^PH$rWA5~E
zvNY$xz5ARA&ruIUvMi_gDKzpqDaVd|fpgD)gOwBKsFzxpJfnNC&1BfdqA@~Xv_==s
zC|FAA;`{6)CF|qXS%<bqX8PVZ6iQgD5!RVVTRY7Qv5z<NkMd#>j_y(_*NDQ(QOW=K
zl<IJglH$>yOVrU+OA0|L3^;Y_6vila_P2QXrB`_Q)i3cIcQ+AUK-rUY@|*%qq7;3j
z7<Ce=VTv?eD%A$YNTl@0b&glAx<iMyWaA0RXoNAAQmx9UFns5YJFG5vU^3=sX9#qR
zR|>IHKCxDk!dN_M@ukA|1In$Y^Puy+=_C~5IK~5H6taBgG_y+!Jh;COJp7l|^<y=4
zg_3ysP`%^|AcerQno#C=p}=p}k;-|Q`9T1AfgVl}(n9E=F$8Gx$pi}oo}yH%(jDy)
z1Oemm7*CXuLb$PMql`k~Ne^2T<ijDp63*MEaD-}KDr)5hqbwmZV}AZpiwD~sIDZVh
zfU*d<fA;~MrPF-o-}+1Z?)Se<is4M9M(3aYV}|=ZM7_x~$65q~JqiO0OLGK~pjxdn
z8jkS10IfB?2vAZ{=nN4!bfFYQ<dc+1JfRj<5!nPYZX+7AjK?uou3V$gmZ?Yjqb|;V
za!;}Uf3K$JGb%JvK!9RWq*&!q_d)^zAdrQJpBI?P9y?n*6z^SS^W1w}`r6ky_Tp!e
z)f$)r=_~L;!g9oHwZuRE$G^kGNS?ZMhONy5-g@hgz(TduVt#Rs*4zwI3PK-53R%%7
zD{{t3f=)HbBx91qjI%N0@dPs-Fqb6=Q!q&~9^?}aGEHnP5=pgEX8HIr;!Jb<{w)?7
zWj_D%)0{kamb;^bV~Z!zckZxK(FjjbtG4hxL1HbdC(kh&Bn*ZLwK<6itEfv~U}x?r
zlFB^exKDGTLTh1$uYBz#f-vC!{r~*CZ0!%&j6oa6$BFYt2*W3>yv_^qG$l*3qZg;|
zee#&FB6s|)!$`2hbQG*1%`zsF31JxFc>!rMV9@K}p}2qVA;0<CzsWmqzsuQECwTG2
z=b4{drBbOrB2P<28c%5NZJ@}AJT<lBahtm_j)X0x8*4jp#i5a{<JJi)9GYP*)oPU|
zpSZ}@&JLAIg>kz>Z!p61y?^fd9vNj%8-_zaH6<U7|CsJ?+F10oM=%P`pFhXa;tI`~
zC4A50{?#kg$`y?Asn!?S-Q1zMa*nrlAMnLvC6YMCuh$U1<I!kq@LM(V{Tyj5HaBEg
zyk?y;icD*^GmqQlkfDm0^90fph(b|bU8cNHr`Ya0TOlDx3qxiE8pGaC;RWx>DVk;_
z#>nZIYkC2u<5S0n_0ZNKc85tBLMpVuMJgTokQqBgXPigrA(AsSsc~sg*4dsI>ptI&
z<m{2i!kv61RvElN5(K_$yy3|H<ne}XI-Q@QH3lg>e9r?cN<jCZ%k0b?z5YIigwq!;
za_sbJ?tGY$87Kbg4TmhY<``OmwE>e}O5}}@2z-332(cQpu#`#;%{ICqACE~#V~o~#
zVSo%me)R5LCjE@rmPZ%{Gy+2ul(Bh%^aAG)Al)-|=q4aMg)ao2Q25HnpPeCiCSddC
zUGChv!`g{eW>@C1A9s@T6V~ZV-{8zsU*^<>r-=L#!+w|1=zwH0z*_D2^41=4-dt4W
zbUZE}31WaLI_qkG*vKC~w}*M$#+oT-{87ZC+gqjv^+%I@hqy+Oq*RJHOvarK`2ly>
zoE`^`xNS`D0Z(~YYn{hnoUygO#lrF;SFgOq=RW%iFMRqH{@HK*Bif@1dAm(fSo*`5
znNkxG2ueYjt8d@t-1!wsCj+p~b5mGD;3=1IljX$yE>>$wjRu*8Z+-hVr&ecp;`|I~
zgGy8MD8&z|jteRUDh$bW;&|qs!7n@EVUeZIh+jz8JWoL@D=ZpIsnwz!f3%u^tftPN
zUnbky22o7O&LFixiJbY>c^WIHiE2J3PFcTqhhs}KWRnD8G)5b$ttz=Er_Rg<s|`j-
z%FPyenvmxSMV4Sp)j^zD?)WoaNKw*=v_KXa!<}_X^*T1HI`valVA2A#BWP<4v(*yP
z8vKfnY}WCc7a7g3a`uS}v@^|jzV~hPVui}_B~F}}rKv48O{gDRrd;wdMU1Z$g?4JC
znZ*jj$(Y$%6>BwF8WWZxq>xl9HTuI{($SdajHRw5dmHztR8DcQ*Jg8P?@-gvC&y?0
zmtAG$5h7?88mSazd}5trg`lZILibEkAPHrFZ(_1h%HXYAyuES2c=v!aU-=rP<{W`?
zesa=ko;<O{{UGQ4y$O50eU@5FJp01u*w}o?hnH`2`+A3y7tgS>v(3W%0;O6FU*`l-
znaOBOG8r?POlTh*FiOYl?RO|i!HzV<+LCB@LiaqM`I#1rGcAlTY;WztXv_+lU;Nsa
z`KABef6V&s0k6IGZ6?Oj&n%}RDAjAu#XE@+!f_}U))v{^86eKQf~>Fc2X_lT{N4?o
zdD-yNXD@K_)N!h{3Z-g^U;nl5fp#jjQDI0e(<WZ<QM~6q$|{PS(RjpoJYqZ^Gair0
zehR3KweIDPl~^O)i*<^QjYcj)sZyySr6Nlv^g9Pgp^1~2{oP&SN$eUj-=|!y5|yIq
z2Ja3*wnzw6fwU59_0i^Uv}?7jo#v$~jWr4@^mMFduy&eWpr;XLQt{-IPjKzVEiRn9
z$U*0aOp=7ESDliwA2rd=9_8@aJG4O%k0A!`--X3BbkolYq(oq-RhuN4Ve8<4<<qCw
zxO11qrBx>5I_ME+KJ^q2{^=%}@$m8#U(FEt&bKkkbHYj)Rf?P~MV_Pl09|MjKV+oJ
zY?Wr|*qoB}@k>>t3K02(pwVLX*b434K4k$)OA=jRg&;PTw)y0dQEt3_h?==t9YS%3
z2Y4YR(o@JOI^#)+5H4LvVzJ6KVrI%KLxT{H2<1|$sUfj^?7AD6MvEQ(nKn5FgF;YP
zH)gNZ>nzPTS!kRi&N8mu-gU(BX=4aZ`6#CX3yDQaJEaM^EZKN8;l}lAG@Cp8>finU
z^0lvhgX3q<asBEI`X~x*86_FV{W4_{Qi@7!-QH!=OF4JRLwUZlw9tZ}6gf_qb_Dl$
zFhFZVSScf-i0{013zHQr%v!7nSZbA!Cc%#a(!?Qo2<5Jil2cDKMNq3zDm!7J1t>J7
zT9q@;K1p|XkIe`784vo@=H@?f%`ZIrvz$16fpSo0*zc0WJ#;a_T8C1LL-cfd{S07@
z!5pdK!9CNS%Z;^6k<mwu{Goc@jl<=%nKQt2{A!$P)O|U}6E%eaUr??@)SGp}ARsg@
zO0Cd#3b#MvT^eJLtWcET=wKjmh~L3}n_j=m&gK?d8xQ#67r)Hh@*17qgu?R)tY$dw
zvs7Nf7MA13W(k3DzekZar`9GC5xTfaQCN~ek0MWr8ZEpq;=R|eb9&L^>1S4GR_aV9
z19Xv)3xOX76nTLs6h<hbD8gzDMEF63O>(3)2tQ-kX=7D@6BJ3x^$JqBD6ydN(QIr!
zR#SKG%Rj^Q|M?%_=NU$D6x`gHY4P-v&r$Y$Fk`kib{HJ=&};JuPZ7<xnDhn+X(^RM
z@B-&MQ)u^klL6J47UR7(N-2!iP6nU3h#=ty_>~Hiq(`|@qpYB_zfVw&DYs?`Yh|SG
zF>a61xy26@)~1Zpgt>A=n8qmI$86r_JAdoHCRsemT&u=tV-vZ%Pr2nWQ>!BDb!O)0
z=^gCjsSw{+WSJ#TBan(*XVVdsCeO1YIJjIY(=KwfaRkvwD#i!fl+R~$x-JdEtv}w(
z{p;V{<%Ol`yKzZhYlQSDiHKp5A;@S+AKyYD1lDS*o{u-ps~T_jxbdsM!DxMl3qSXB
zESx)s@O`XORHWn!7mu?fZ*#9R=E3Fz^vN|YoI1mcue`*YZ@$g`<~E)l(%Bj@i4#1}
zXB=mw$$&8MnT%tM&Y1{7stc0Bg;Ix}&rG$-OubGN`6QDOyXz0|GD{P~r&|%{=PQ(p
zea81MV|!y7$$*E0G5XjlTcb9?$ni*8@tAroq@o0VIixhd#NCDS{QZA$nQk{Gn5#0k
z+#*X2x+sX-2Xt;Fyz}M<s4yqBnY%VP<)-;5)?8ale=wlk-lyKEQ>v6Gl}kT`udGng
z=kWft_7MTOKj<NZA`D9iE0~P>#G?U1g3b!|cK4A&GCMa*DJoNMG^derQ)0Yur&3cC
zQ@NE}4$1Lc<8avI6&V_f)sFWjkD89DL9YdCb1GGjXP&;q!}WEpeE1>VL66aB^kd_i
ze|}u(HvB_A)Zy69B7RJP>?uj$dvv>fvOK3&Yw+N~F2_FeBF)(@{vcs{{Q>RmeZKG&
zpYriD+&Q?$Q<WA9kU@Y@3SW8%BMItN@?ru~ki{{CRXWi;`(A_nxJQ9O1wLU^LJCQt
z6@ID2;@Q)TKDY}-N;*Y86wot<_@{7Bx$8OoAjgOw=SgW}+}MprV=p0n#1a3|9M#U4
z99`rX^C(78TZ6LB+s%qY@;nZmQr;ASb;?&`3#^eY*IF3jct}24=L;{Nr0lOEM9AIE
z9)IxNx465x$B(PT+;wqdK&RlAQ+{1xEH=*BOLq9J-}p7w*B|iw)6Y@><(XOPqF@rI
zWLZi%3YcwHnY*C4dv}9uGJ`BR&V-f_25QQk%Za-kjL}53I`{#<|GnF2Bl!HwCo#Ff
zM`Qb4gbG194>aj1tf$C@KuHg&JSxow1cCEn^ju0qYBKC3K}G>HYs=K<S{!U_u)V$W
ziECbO&EcUK_1Y9g1~z5t)PgV?YaBJf^BmgjRLOFh4CD@bMo**s?4$G-A>4ubu(1<D
zIQ}G$4p1D9-Vc#JbC}jqAgo7WBvHA{;`|Jsd-)=#kI%4s5c8wAuCaa4aecuw9`dmC
zN(ekpO&dIp_FHM8$Si|ikIV19L#0}w-mDV%iq+*MhP^$;qYN)7gDp@}v3hbIlg*I~
zCzQ)2f^vikLO_tjBhq++U#}CCN(^@PnW+la&Yz@OZ8#4)HAP>sNYBHQ9#Z&NDG0-W
zJWVO`0_FSU!yaKdA|H=2#!#87GZ~Cr=K0jJ$67(?hd;57e|&wcrf%`mU!Z&c9-BY-
z9x8WqW{Kt8*%kaEA<JVdJvMINVt#&>-p&E7g;|1fh>0TF+uImxsm;z2)a%m-scB6-
zpR}+H4|=RNS}3K^Q?8Aa5+MbC6q4jArACXnAf&s$%}DQ3oteXnO0;I?ne_SyVad`7
zX_hda3=z4;3w)Ms&fvji+IxE(L@kOtw|LehEVq`KYt0d73A#uL!;n0S+1u-I?@pKV
zm(J54blqr6Kv5Kq06it<BAlH?&>v9m_K7MX<-jSWCb`RQ{nx*J`^n2#jWLp@4;G)X
zE-0yhDjpM^BQ4aGkB^HjN^?ua7u38QX%)#RXX6il!0yf4T=@B~bLp#JrM`Ha(59R|
zxx`A<<C&ce4z@eox%~s~U9Pdae2RLxOkSy@Jj>OqH}NvdT&Y5>&<JZ74~AIlp=B9i
zC5x>Vl_(+zeSB?*lQBu>fT78dg`p_|PL)GmI?>>{izleoB4T41{Nb;YbvxAKgp;Ke
z_Ct@}!U^7zRfO~y>~s;Eo1AS_s3apM<BZm|w`o>0ws$)G{8zrp(@$TZ*E^s+=#a$l
z+G|%ic50cDH{#Z&Mk=TA!T{!@-3cER&yCZBcDF-qze*T|l*%71^!Z=-y~U3s%62+}
z(MHql^$7fczz@+{)9dV#C1WI@)11*{Kp2M9s)sz(I)NW>gi>4MG9Ra7y~74d2#F^J
zLLDmXEmms^>paoa<2kf&H0Co?&$K6=xWI$^n_R#7EI)kfhosp*k60cyZ$kXoVSkDc
zJz_%~3M`})gn^Im`_yVRu3dkh7hiahKmQm0JcHgo>ck1|-@M0(wR7xkZS(N%T^frm
z*0)BS9D39j4R~tGyM;{TE(%G8<VnHqR-bD-yUcv!5}Amwwu^^FmfUC?tDPu8DXJ%r
z6E4qS9<F1J!1p|KA=%Q&Pi1d8YObXlk!w?6rMoCrh-n_@BaWrf79l*R%zx~K?t2n}
za!S`}ZJbh4f{_?uuu5RS5yqu*xze5!S)dJMnlz2k*^rIBEpA+SlTp7-MG2Hq5ENWE
z)8g3DH~7`x_(N_y+;;VynlpQzIzk6se4#pu`OEVHAz(b5(Cc?-clQZ>j}t4a$YO+_
zXU=a-g_OJye@5_3{UUjmVsaNvsKNj%B-#|jgC2R2Q*O5KqY^i-++`-rx$w*>W@qQg
zk_q{+PlBObssjRF`ADHCMJ0+XM|vehAd$XDmL#ZBgfs?Wn8u|!G_tUkQnNt|!_X*T
z{S(zZ@_hu>rO_yb(Gp{dqgo!`6w32Z!gu3S3%+u`qZX%Vwi3+rooJ1Q$9(9FF?h=7
zXiEV&bTU=S(FvFeDy&21jWy&pqto8U&o}tOOJ@kB<?Qh#E}dK9jW=)dowq)4$f0!}
z9G(<Y+pH-nEQGTifurcm48RV!as3J}J^Lwq>2rMP1UGvfYJN_hXB@-^BPC{>;0fWl
zgr4W3|CEPDlMP1%<tkCN%DC+s;+e%cf?5L`3QPvR5YD653IsmNQ&{1Q-K`LWL4bF-
z2WBbMBD99a+8Q?W2?~wP3yLW^S*n)N$w!;Qe5|I<&dqWD8{g#M&K=T++aMKz63kW#
zihK_l_zbt#@oYi0T*4+PgUuam;E|2S*djw%%P7eyWg5RyK`IC73*}L&R@i;GLz<*h
za<nV3gJ&f{xkTKL$@7$;RAqi?jo!fy$zGdac7`Y_<25DmWPmglrQQ7A+Jf;oBhv|D
z)F(;;sw*{;Kyc~o1(xO-SUhHHHMGzaX-a$7nFh`@muSty?!l1i%!>2l$qV#jFC%N5
z`Cu|06CZT(@(R_Z7DXW`kpHV~=m09`j*;)}6D0{X<qTm{n^Be?)+$Ci2PtR>4<E>l
zVw?%eAvAqW4@FjhUf<^G-~M|H?my%+f8pPx7C*q31*KMl`M{$TmRT5$c)3tK>};^P
zXQ|}|u{h7_r=I8DR+n0R4jt!6&!gAbcQZ0ukfjM@oiRv<Oh!3@DJTg|BM9)NM>z;M
zKR3&nreI~JO4-EJrBBrud>_VTff$XbOtx5VTw+`<Gcp<V6Q?=#r+=Axvx1yV_|D(`
z@7R0qN7SzWh;Kac3N!Oh^R3_fHh=KDZ!u1W%(hyDk)}CY;o@?QEAQWAcQ8UT_3HvY
zN~<OW_|l!!3-gneeuv8fgW-VPy<Lp8RO<ClTyuw(OheYKI8q`OMb2<Ep;Rj42Lbtb
zNcUhDVa$>4Ok0S`5mBi`y;`SMtsU{{+_REnb$-+{9`W3SL@8$?Ermr}>!=ll8_#O%
z<~xT<%R`=y+r%A;_qor#O1sl#b9<9J_is;8(L<t@8?VY~(;|hyL!+jnJ9{_|#JC7R
zz>|`|^T@KCTeohpf3U|t`<>t7um07)#!et;Q_z@c5U3KZ`8l??JIr2KU}NY4=}n-p
zR^SCbxh@>C**DyH{~izTud}z=B~xWCUEIJsaSYFkured`obOqh<qqLB5?QUV^4taX
zcQ=uVMhItoyJJ52xE-dPHp|pHMq;O)iAKA|+4o(_$Rm_aOh=#Y`X9bXmFFXsMEM>?
zu8~5L#3LTwf5_hM9%(+n3l**AGOd{z8s##zN{gB0WjsG(@4*f3-n>k^yN4t}XeX8m
zJ;``HVVq}((4$tXva~eIpZ_yo=l}65f0s1ZSnW7?M-82;{xs1@kmvc-*AwzAXZ^u_
zj<2jyE!C)3>ulf?c?zAzq)~|sjYtfE5)8)^f~bY?6rKvOQj(4b<Vi|triJo+MtcWH
zoATWAr>HjOu|ne}*-iy50}ttYNI#%J5`^ycYYR;fl^`n+S&j;Q#_b+?ZV9VZvN)ky
zE2DgmEHAL7h}!YxPh4}QT&*xJ*WKbMRqoJp_&hlNoKotD$sm-6vmKeDVa85*^iEK5
z^jumqRZc%jmJm$2lgdXbAJ20p{e><dPsy^F@%VrTcdzsC_GPlf9RO1VQcRd@Rr%Au
z_$4%kAH4IS`-)m0`5GEyrr!;8;moqLET6uw-N$e=qTZUNT&l3Vw8-XGok=$(Pe$}A
z4KgdCC>V{$_?42Ql6V2a7i7IYR{2DY2H9YQG?q%eLeOmBmqSN4oQBvMZH^jU83fMl
zBM4loicmyRKo*ZNQ%}z*syZ&~U<`QyUI@aYQV&S81f3-xsjeT-r%r)7w#J#S|2*&g
zy}yGTX=<f_#g#hIOo@0rqJMC};^I8gpJx3PImK`S#-L<?)EW7BhztU^L4*O?Y62M`
ze8tWnr!tILXjL#s=QU-gB-_Bpi^@oATqbGg(^y(1+v~Xe*rvdXA|hEQ9FB;S7-c~h
zDS4U_PaGF0D>b-Ku5)|&7#~Q1ieb@D$)&&sK1Goc4+b>KCC)#w!v3(0l!1Gw(iEdL
zMmS}hz^RptHYlYiOhz0l{K<$U9^k0}p*$=Z|7#ex{_*wFzw_rwI(zi4TxEEF8@WGV
zI381&oU-SkwO~|acpg-RLJ5Plf+F*2`hrGa803N+k04Ii`u%V7N1YA6`LoYctN2Kz
zAquI@%@J1X#Jzpa$$*oY20ZQ!b|}r%XbYdc!8TzdAPPc6MN*19%BoDROYDsL3}R<s
z?VA*xW}I4@V{NX9s?<>?Wi%d>$0M}%n1}*p1aVQ276#ucR^}SKcIP&><k4u&py~&#
zEH*fP{yD4)Ir-G5_|f0{72f#uzsslYzt5{@UO>Khf;Zp3&Emog&pv;SvX|hk-{Q`%
z{tpaLXbiP-K&KxQe*_nJDWO$w^5mr_`P9oVvA4g+Klq3LfGq#XQBYcI(k$bk)5aJ>
z)bD-bn*Z_4S8D~v3MO$umK3y>YAC75vxI}aZG^EXb@-we)N57Br4r40i&DAlUK}{?
zjLkE&&N<|>9TDAyKq;qEJoGa=`Z<ijQ&Z~^0O8L1hd-<1&RLqRDqs4-m*@{h{P+LU
z-=^1Z)2x?RoS&sqDpLw1z2St7ojtO=AW+^BN@u42SK3(2^!o&!ht--i%}CM|tmgXl
z_j&)_w^>?QBdnIFG)g4HjN>QHapUIuOh#k6-7YUpLY4}H3?)gX8KstYzVj|OFRzo2
zEoCW*@z6%I@#cFRTUbB_9##oL&k^wRJU`+y`M$^Ug>#IqUL(7^j>VD-!QM})snb(8
zTPfW91?1EZ&?1!+DFlI|%^<+{9a^Ic3qqka(-$>h4IUw#gLST5e~WkD{Q+;kaTy7$
zt=1Wg#`OCZX$4Q8KgP>1y-aO>(HVRn>=5`c6GfCuGx&alv6>`F8I2r&v^VIpv%bk>
zH0Jb$r}*3}&+^09-o+@N!ngy05RT;UYU}u?c~Q8=LVzzNUJ$Ty;v~hepi!>UY&7WS
zV~TV{MxAkL@vUXpA2{zs-zV?`q%X;{m~<2qwps*1$hdQWkb=481%i45B8AQiylJE1
zEWSdNa&b?dQdlJjJV~0Qc%=~2?hynYMVerYrM@&z8m9!IvjZx!9D#Hqu4e5M*IZ2-
z^CLO~+WnrcEtwwDENZ$Bc%Fx~Af-PY32H}UkODQ${<cS+iDLQ%1lks4I>t(Yl7XW~
zWhtRwL8*|!So%A6xcT9Gv^!g1CV19TjTE_x7!HQSNlY;@WaXT>`H(;JOFzTL_73|8
zU5w{Zn8yx!)`@S7by1mmnkO%(Q?)RRm|tF`)~ZmgR)~U-#1p6him=Ks&Il$0;xwkY
zvV>nM0T!KS<e8@0Y&ip0YpK*5jI#-Txr9(2!nlKHVTGGCT0?Fvf$s!hLJAa49453T
zR6e;CC?k;Nfbn2}X9b1wDbFmT%2kTqCdy7@%Ra*TcugHniof#er`fu9oyqTgpJOK$
zs4tY!MM2W;qLd=4*U?yf&t+z+z$5Z}1RkcyD8?P)?g2_bsoBC9x0zWL&@Tc8sbx_-
z;%*qD9ik^CUKC;Ssk?XXG9ZJ6S?F|`9JDFT%-~hZlv^|OMt!se&+~~Qi0VzK&hX~@
z519)pG@C7U@p)^~<1-AIY1Z%(2bbo?a{BBFg*1%gn0l>6F@^3WZVcg&1NUHfQZelH
z34NbX1%z>z{f)N?y@eyP=ub8d{?~hj^NUoM7pR>+PSGC|Z|^a<d7p=O?~v_wD5nN;
z7#2ColSDW*Ax1FBd}@9{#n+5;i;<F2Vacvu$A4uFTdF(sL`bKUwieNxMO6#(EMq(z
zv-^IR3z^~RnSe}WNSPA^nsB~>^gXO8@MOUHW<jsel$Xk`M#6ViDki6xbdhMPDkO=Z
zGi(#*W6EKfiXV_A2|=YqtrBs1Uhvun_oy~!8E<d#;qU(;@o2*N7e7NdGtU?Qga4G-
z)2DdnZ~wQ{KlrCSz4{9C-*^>a0)p%z>)-lqg55iuX?X-{g>j|MgRL#9B%~jKQ#aR~
z;lKW`{u{pW&7Y?=JH!3Ex43%cGVgzI<xia4W_f<3IE#W$u6~8bMrzYJr8bt~aD)(o
zN~wZSf^?EH8uXC3rp`{Av)S1em2!nzy+IU3Q{EaV0ZEo%OaVe1DLD^QCLGD#o#cg;
zM+Epo<KaUSYjBfnZ9qyl+P1<8@=l*w;wxYNEd5cB-~R1i<Cnhn9H)*g<A)Mol_*ym
z^v5x8zju|l-ub|(rj0%F%2Li>*L??T(S@FlBqdpzVXfv5zx_L0yznGv&Yq>&nxVTF
zv#@lGH{XAg8Cy_0agyyx#wgX4!Ukz|g==^Ac;~%`c(Fr7v;efkSjq172Q0tT!!(^r
z*z^2pa>F#C!IS?#>i#?2(&Q}fg@0A`#`Uha*Ur0l=Q!PyHAxbRfDjUr03!j$1al1b
zcVY7#``B=feZUwSg8+ju!3TrM84)BT6i^yTqsh6ur*rS_-Z`wi)|;y8{PC{dGs1jW
z1CsDp*VWVAGu_j@SFQEdQ_pkXzZ<Is(cC<XuegURE8EDCriZ0x|Mn&cc+tgJjI#ue
zbP5*N_`fOTfOW7oFGg$*yW=i}MdwbTN0cz?_xboo-^-^z@#k!8tTNFIsW)JDdX~{}
zM3!Xe#9%B82R&Ll>kQ%!QQ%PWBc>;(n4eo9tTu3@!*I}}+ijCICsESH>YVG#*Ld>T
zr>WE$yz8BB<HE(uZ13~{GiKR|v9Mo|8M9l^!MI60e&5YZ&#+xv#&Zh178JT~w^C!6
zYQm(=Fdi|vu!!sX1uWSb;@%LqTp{oS;(i|`9h9pm*K0UYSR{muHDfx@$6V7wqn2ZF
z0<X|}Ne8c5CC?MmJfjqOWVxm`KM$c_bZ!u2Sx&AEUKrwgF6rkviu<2${5V2D+F~+h
ziiQP3>|$)Bl)Gn`(J#J=H)G(G6gZBLbPA@WE-EapUA@4g4?oJ%(h|mYFlclhQ1X3_
z?44!*zC%pyU1ZqrvT^Mqt}@h1A?4B(zE{FnO_C(k%QdnrV=(NqvbM?#>zmwp-z)jX
zcfO5Z{*~V$PV!<rtzX=CV+5Jj<a)gCwtyIGTl_lL3#c_~IKE4%Qe%5_l}IQ=7}7~|
z8al&`$~b|Ka3v;981?%Er2w5JpbI)cr_;qc0;R^!T}L6UiyaKHdC{phNmg+Cl=|EW
zgVsn-LCGKIq8r>G#H~#cIuS;B*aVW1DRk=I7eP~Jtt~{NwF&mV?lr8QyNcgzaC%Kq
z&SSzj$E%bPK|nSfQExPGf?&7(EObVi8JvE=aC?o>&URt#tyIZefg=Uw&?8F|p!|{q
z3w8@4;TE@Rp~n%@xyE7$niGhy&#>2{<obkRnTg41Jf#RdAFBm2^jO~Mvc9py#M~sc
zN}0yKBb<KzDSE3LyxsGu)u7++;)NB;jRq^5n+T^E*W3g!T8cu%?!LJVu2Q6#ChiZ(
zt0`rp*}i<8jmVOu$ygY?>)HPQn-x{-*u2o~SnX21WrE2Q$C>Nx(7tq))eDztUAj(d
zqfMG=JW)K%gp}l1dbuD}ijvC6j3B3QHAS@uD_#`J14Wx!faAJ_7+1qRw=PmD72!}h
zk~m37dOdU|h{_Q%^jKP2=jxSBj^BBhsO}+zi|07#JS83`jM6?yk`V;~zN7GjBpnPX
zxo*+rc0D3Lz*Qd8;fRBk5&I`3`LUDSxUo)c{W|@v>y#%KkWR#{U;edBOw99%|MGvK
z&wYYse}i^nxPJCorc_RSMiSO$dG2bL`r-jDUt6Yv!Ai-F(SNP?)$x2@asRzcOg0fx
zuy6kXzV2(^!KF)==nnqsm>6pz*M`AxM3Q`dE&9KH-Q-W*bQTs5p?<%Q>v{xXK$0YM
zJ39<|ZIlqW;%16~WOA|?ZB!~%T-Pl+Y2!&~o{WG(aIK`;Jv8J?!rhBPzetwA3Z%Gs
zxic?8SctK1>&1(ev>?!&Jaw2jdlR)NV<PMo2&F8=<T+{51EKiZx4xdg_jPaK;-xD*
z^x4O_cxj1tXF!@8q}9k=Q)Ji{*oi_@qsUT29LHR_dXZPY>H#)4VzkhdqAFn!P_9(j
z+}fcn1bX@)*{QE)SZ<<MKTc*{q{$2E7#3lRM238ruyyGo(VZvp9Us^A$g*s=*?}m=
zW7I0_z59Op&t7En^jVg5Liz=Ip2#r<BFj>=esP2b<IA<Zc_|g+3~GZAZjq*9!Ae7(
zYvSR6-~XLo;@PMEf<~o6$(8gwLz>MRqd4R0wJmf8q3?3+&?3i=9HiN3QYn|IMm|wg
zW^%I0#Plpixr{n(M#BL@3hK2gCBH)Gm)JAENK~ot#OY_b<+eL`$J@S=-}vo6#v~TU
zS~61<jkz&o`OS3EqM(yX(wu4%Mpe=@r(7u`b4$6_VCBYDny!m*Tt+%2%TuHu;FU^9
z-^Ju9Nq>NplEC-LM=?@Jf~Z1w*rODN2*-i3$skWNa--1#v=HO>lPH=LLKH>OSmc>n
zOBSbi<qEkm(C$H+gX@8ilmH|_+rBD3naM~;L(<{j9%-U=QD`D=B1MYJm(j)8-BN(a
zcR$0}!hAw;`LYz6c*k+aacK_97%rWEfhV8*1W!Nx5LYhUpk6IAKU-sMb&Ga8M`wn?
z<`kYtSzEtI;3>*MiOGp3dlnWcHzskF%b?$(+v@;=z$@c94xR1}pMCrhu3S9Nm%Z^#
zT)A?Uzxe29NV8(pZFRBl+L&D}yP&x6`GLpncic)CMl4^uPNiCbIH%d1W_x9YPzA)f
zVRfs+L{JodQWUOQou#P2!%BstEk;|?EF;ZRY7>(<jw<w;u7l&bq@svBb(A6-C8WI(
z{zR>C(cbKi0#ZX*iWv4scqNFM3&@~~&LpA~5%087<vKRCB=KK8t@+FAWfgT|98xc(
zOVpfY-|HS=zEZ`uTdXW^u(vryW3Eoz-9`wF47?(slHx@u1TB`J6yd2obhp<TwYLZ{
zSU+NIV;h$tS)vLZiE_Xx=J$p|WLv~+fDyU~xJ|+4Max}z0a2vLvlK51sn(h#{XQVk
zZsEXMy?&WvQyyA*c&XrySKrIsU-<?e`s3f@>C=yLl88<}W$!(&V|%c|z!;*ijO&$(
zLkNR40@Rq5g(R~UAu@zk4B82L+Tu!|`s8u0UtA_hb0`|;FZOA~=X^&Y&%WvAx%pOn
zY02}2!C<`HRj*Hz#2xZMj!X+4niqrwQ!_McRVv30Av=B6E?ngLv*&4FzkwfXWUlaB
z7toAy2$duhmdscV99kTYa<U?kz!th0&kxWf^!q(L<q(D?rso~3(Fo;Go|q)hQu^%{
zy`2sS@UeKV<oKyYDpjAT+CUnC&SJ7$;|M{?D=|cn<~hFSFj1@1)@|CIE|p4^daZ%0
zJhZWRzDF2J;^h}OaL3)8eCIpxf(B^K&c+H8%^9Q@F@N_1yz!s?Fz@-{e@T4)H1pF{
znlmn122@KPLovx_f5c?qa+A(gQG$;B>%9`aexK9NJ<qY@#}I&WT;BGzU&kN(;U94N
zg>(Eh^1AW3;S1q}w7bTEn^-DqE&V~Erwf9B(I}?9vqqk!1x7#>5r0A8(QG!Ulq*!K
zRh(UJo&cjsvlzsSk%MXs{(Wf!Q&2FWl(6G**^31q6b{v!#g0W_#|*$s=a@0$>6YV%
zSh-MP*cmY##>8n%mL>&)B28Jnah-h!4|2zelYGT%?_<y(@#)Vz#`{0`QO+-IU~-MC
z6k6K??~nuJT)K1>tqtG!O@EI{)#t|bK7)RbN)XYg*O{AN;K;tc)LlvV+OMTunnEWj
znX*``uxJc&d?=2;jttf=F0t?MKEmb{R@`*wI%x7DOguLPUWLlO1!nHMlj|4Gur>Gs
zb!#Y?hpEcQ^OT$LY59`QC=?TuLY^qdauCwuj_LnJq-VxwKl33T{?rEvTuA*9L8$2W
zHN9TW+G?A@z)-2U-1CYP96xf5ndTJLQiYl3BvBNS<r%$hhtAG<`okf<Qj7)}UKk*1
z5cx$nGmIMCe)29ly*}$VuJP8ddOgoSf0jp|JcG3YWsAZyzd2Tx0!K<5C77I=;J*9s
zWoCLWI?0M8qgWFKB`RTs(2Gc|q1ziWqY6w(@jkZXNrKinZYjX%yl`O(!RF>Bex-z?
ziqRxWAzT;JA7HYq=-A|WF`gc$Ysj(D&YHY{I!lS?c*s&Qa&)D_I38hR1~)2`byksi
zp|LfeU%0?uyNuRoJ$6zH2O%=B#Ra<XLZ~9|S-IqSkqV*QqWIIgaPS&kWP4i)$!New
zKJ*?Q{Mh^0-rA(z2$-LXn4Fwud3l}DC`V_SX06PgJ@a^;M>#4}i+rlp8gp}VOwP<<
zq)UHin{K;JZZxGZq+ALJ{R$KH3Ch(PA9?U&Y^`47o8S31u3o*$3m0#o^1_9y^B4DB
zDT(Ju93`nYs!TN}`KmX4Ejo?aS-;L;*g<)cpybl-w~3rdoUnr9iel<v4N`dExOl#c
zPEvH5ARI|HN(rJ8QKOEL5}js6<e=~3lmc>5Oc~H4bSp!niJFt+Xgq_}IsI-IW5KPI
zaVF*v6ANhN(`oeya)EKmc+F{A+t=9EpI<cG%kI>PA_D~LIP5ukgh?!zAyKo!++2fF
zEyTthqzDQ?<oMQC<h)6+IYw(xippe@?&bz*y8|W5+UgFjbjh>45Nf+F(si*4z@l;o
zEi|bBYYTQXj)PSOZ8EGjxPCyMrjQ##FTna4Nt$9DhfaG7Yh(5uSYUMY7{i?yUs#Tw
zyo<ead-$b)_8oMey~MHmU(dy(UxrOJwZPz8gOm4BaUVtl%%}~+7P%NdoORAHHuU@8
zI|}7FXkm$4Yph<qQa}qOgicY#mupofnoXioiFh<5O=5<_A-&OvR;Npz+ueip?^=iF
z`5|H8k?Rzl=8WQ)G|%vqgX{R{&_Wn7*PLML)CtxvTwv?W1-e&O$ou(jyLn(GTI7^U
zhP{U-G5MevPvGGB9**aerzuhhs)0vqeT)8}N3BuEb;hY8R-+uBs+uH>%9u36XwCGV
zX_V_DJ)cY)WSWp?DXt&jL<*#!WQtf9Yc-*@1m!yIZks$y$+X2$0;6-JbO`(qg{1rF
zgESAF!k=0oO<-;5DdO2Z%<esb2rJCpc0X_a{vYLc|K-16aCMVId&*3dA|lUY5NBvn
z@V8v4@P$Mg@GWG2jkh(=5<dO$PxAJ6yuFAK0uCQJ%Bx@ZWt_Wo@vk*ae4*BESN!Hk
z4of^r7z~F@PtW3nW%Sk-n;X|b7xy+RiqCenTBcU5QmIral_DJF0AoO{PDrvCYYobX
zn<lwJoXty`*f%i<H?>hO8Po2@3z?fG(M`S?Xq%&rMj6TG>Lo_q4Rn^^DoIqSG1Z)8
zVRoJ{3>hRNZmh0x@$_kw&DpzuAK&?H@8mVFx|bjL;s4B&&tE3ink386wotYOfkRY@
zIC}gRZolIWu3WjuCqMBqq;60RCHwauq+BjjsZVm({jVV@&lON;ogf^Ce5~O15o0{d
zcnV=8y_I#+?H2Vt`*vNMV=HbES5~OfajIoz@3@otCqK^G=`&xjJ<p6m=4Mx7Vy#6v
zH#?$Yw@YfR8BbmcGJ(-%Ty$H`oq3E8zyEirlnYY6At%dn5OD2!hd~d*z~Q!&2e|db
zarW&!NEAd&Hk#CHHIyT1ZLX8WG22@$o_X?F4jkOi{sTveoQSd)l4l9RfiNgj3ai|H
z%Wa&#bcV7o`KRCgZT$29^FOe@-3OiSj?_KJ!Sfs(B?&{H`b3@P#2&uv)nCr)@>PcI
z?Lr&vNt7$-Bwf6`jOztBj-gx%i%zg8n1iKegLD{!)ksIu@AXlxqS|Z}VeVOu5)#LA
zNUX-@dBHp#WtdThQ!R5d^+1Rs`b;_sw-S+M29acxDpgcv5-9_W<Dp|q+)r^uhH%`^
zx#av!SHUB^`Qzg2bCiRXcDJw>6oxn`(m`ZyaNMGEUG&8*xz=P!#-IP$AMlBf{T`FG
z3Tf!lYY&*5YOt}<VQp=I)^KR=6nEWr3;P%MQm@pQnV6(etAY{?`aPD;U1X=-A@m&L
zL5}B#cnufrXNbTMlxi&aB_6oze$HM#%dum}_!r;vUHt2R_Y<sd_Ym5A&b|k}i&B!P
z9J07~A199Ag~es3yEf+Vf$NoV9G}(UHdE<{Qs|M##biL|DX#UA(j(Uza<l8G6@f=-
zqJb)vitc8XB2A9=6hXN}dAdp79ThpfCZoT-g%E<EQeiL{(C-fjD>YP9#-E%85s-97
zq@j-+H1UKWFUKtBf_7<v(Nqoi#sBHPzU-n-Nm*F!kwjsIa+Xjza1>`@0Z%5xt?Sq<
zrW6JRK(X+=zbJ;55)Ifq!<d|?)}X(=K|bsjY?maXQY{rWH~~T_Fp?}ElE;c{<S;Zo
zNiI+%xPhkRfG<^HtWXZdq^LMYl^h($VU)(C$%yUkO)AxhQmsPc<Vj9Hc81nY@j{m-
z=h=71y_~ytf$KYM&OUjS^UrPKM>WEV$D6+9UQ%bo#p~B;Mn`DQIqaEp*t>lOl5>nk
ziuOQJuQ>!IpA^I3`e~d&%qwocgZm!1pI5%}0S+HI!u0Gcm1-5wb4imT7(Pj2+MPB}
zKJ_G@_{1mq?8A?6?b-@SnvNM^e>Y28O}Sj5mQRvpDe-8;a5zAgODHJ`9EVIBgdgE~
zA$#whpgO-myt&0_b(8L;CE|@1a+F}SrBn}?nyn(N0TgkshmKQRr4XTq=eh{#Ff}vH
zpxs4_;znunLYS&<a=(0^42#SxCP|UnVsndB29u63xh2mNaygd$3WZPthvH`GMisoE
zT*Lrn`K~4{vl?eCUd{(SHa_%6EPmS}el$VF^||uLzeD-!{sE=EM-VDx->vuZw*ULz
z@}BSeS3G^b%aJ`Xi_<gA)=M<YE+cKRflDqS5C-4E(1>y0^>cSsBFIh7Ll1wN>r2<!
zfA}CE@O_`ReDzy-&maCV>+$v%ea~-xR(G{DvWTrQ#!#+Q@jah39njidLt#;_B{Kpe
zU}kojC@NE})Cj^5A&baADIrOQ<avssDBy06ccpkq{_Rco(Rhs~exFlda`UybwqRS0
zTb~7TL1FSC7tTC_?^<S>vnUj`TAjJ+c^b6|e9uF<F1H-LgN@B)uC6T6t+k1w3VRl&
z_}aI>mgio$N-Iu^KLUYNmWhcndlqMS;DJ|g_1YEQ|K2}iXMLNg2JBn#*}HF%UT>Az
zxs%kJQ{-8OHO0lFRH-4O5_XV*D6&+IEjansYVu)9cXNwb*ro9onVLmWCxt>eK0^4^
z=Jqjl_dRSsfBFly=Y{CKU<X>QaS))zc=Rd}Vw})1meSnZB#XsZO=t5OfAPWJryM!>
z;WS2P>}>5IrGvGSUcZm46emwCa&Ye+0ykjTAJC{ZiOMw!5DMgZ%FfOfM!<a!+=s~o
zGY1Y)^{S+uF2>2nmqAOOr!<?B)HmxWCAsy)QNHFYU&pWg-k&kFg4`4bdnxd-wn)2}
zm}+wP&~aY*z-u{k<}7*MMdVol*)R^Sa#<O4+3!QOWXT4_SE=(9$8(WhKxQ>Uh!@#?
zuEH-@3F<YpE$|-Fage@;^j)&C=33_?`Yl7nQ8<+{!WJV!A%LXFOV>i;I3B^&KBN;C
z$w@&-JZdAtGCDVCsc3Ka{?>c0l!H<Z#^^%BIc~t*TrZ9FV_e*4+2<N%q%=rtxNz<%
zKKkMJV6%kSckp~cmMS{!l-0E%@yIeaTjs7ij&aMe6HL@6XjU6cH75zefPSY%JKmzT
z)#9lqo@I7+hEsRkMJP&?y|9pPDe%1jAruGp9$;g84XZQmy8Src@(pj}XaC!8lL$qo
zi!OU9@UU8w8I5u)gkg#O2M%)W+6}tB4Q4APax|l1k5W0HQunz1_&!Kh`0Vqm+;Vu1
z!?jxBqQ&AnK5;xKP$x=Iot(h0Rxwf(v<<1qhA}41iK2kej}UV;D%R3lTSpm9p2lpf
zk1)~&KcX^spfI0=Q<y9#>yH=?E#>)rbfsdc7ZZCf;nV_w{32=U3Tv}LYH%`(Gm7c;
z)|m($G7&>M#1jH%oPB4EfpI(38eK%mXoJ+6JWdd<i(3lFhXcx?!`7fjy;1}m;|emJ
zp$9!W{b{ajOmQs^$tts~w}+H{a7HaARGX7CJ!Wbdp)v%O61q3Sjw$9yLASj{k_@R<
zrzwS1_MbRKZDNteTn%kfgjeDASG|h!ANXA+Wd_4-2BwE!Z(uy|$^j}BT)w)-<C`%O
z#oR)fZ#iKwsioCd^b$o)dJG+p&D_OI%<@fN_9|+JZ>3bJP^y$D*Xo2(nIJ6TI1Zj)
zqEah9JPT+}p1O^1{-%Gx#>OU}`qZcRt>64jKK1EO(;EywU&PrLbiL<K|A6oO*6*ZT
zX^>?JI?rfjMPV035mHKAF&-`p2hR^MbMqv?<l;OBUU4_wtxeLk6}n3|m@I3gHRPE@
zNrPW5kxEI{?UVM}C<6$I=eoGAhnA4r3@7jyY;NNRA#M=jD1~%9GHcMom^{v~49DG|
z(Kr#H$d^>Y!NVaJmc$CIlDI($w_L{Y1Mnm!&ln~{@^pmfxwP9YMxz1F>Q&bN{5>pu
z{Xe9(=P+LVX`cIyU*zs@|0g7~3j`wK(0#At%m4TPoj?7-AK|%6F*;x4)?-axb7&6_
z<(6_7(6JJYB$S$t%^6=3KWAA4CbwK$y28gl`VqeU(03IbZflvHo?&WYg7x(;eo<HG
zYl?3#E$Up?-OX_w_1pB@+c;P}SCAxypnqn1hA1pkuQdvn@;GM9TFYqEC(qJCzcmJj
z8e`F8R>Doq)=fw$W}F{tbF3YsHiX>m6xnen>qYICBF{9ZpL&qEx6b_Z0@ZSrAn<WL
z7tbrexE8m-A2~jbtH6=0Z7fr%){5sAnw&y>eG`=sRRZ=OSmeI@@8g+gpX9*@A0i$k
zSZ!&P_YsC6sUC6HJ@=v95;`BJ+vGV}ZV;hQxmKmW-bNxo6vAy=yxkF6&{^BSBttSk
zL^&@SMYPt%m|7`vlMz&_4CfXZe4cK&&u0|{wkZD2w&-?bdC{G7U9ZspjoEX%9VT1Y
zev)CA-}sgP%9+#85|mtOHUz7Q$`QRzOfnh~d6Ip58yws}LA4TbePs>L?JzYtU$lA!
zkZa;mLbuZ=&QjE@$I(N_7;bl%x#KWlrH)Ppq`Jp2>7b-XZZw|n;JF2!<JGUchu{6<
zkDz-oo+EITBuf)AU6^TF?QK?9m#NpQtZ!_vXQ4^2(_$1Sv|5`eU$A%YB*z!0==W}L
z{_F<#-x)9+l?#Wk5V*o68TIiUhuCP`dX@4-1M9fhEXU?4I<eqM!b+KNqK57#B=HCv
zYqGUfnraWya|?svU_h4Tcwq^@RK}a$i>OVL7{x%3@KZ2u1-DwEyS>Cf8+r@-fe-!d
z_gx#ZykNB5>=3_HPr2KXHHBQj7!bKGus}sJJ+18({^Sq;JHC_Q_$8zx(8EGQyuML5
zoom&AlgAIRZ+<V$T7xJm)0~_xE?BCNRCU`sWG3UD2kyouf}mRGz@d{2*S0V!!)i#g
zQPDAU6xCV{DFm+T^5(C29e@1ZkI=fjjO!FQBAumVd13cywRTutTc%P8Iez>I8yh7?
z+pFZ-(CH13(q(FDhC#bW&6CXDzR1kXB(CET4@VfMhy%4ogYO~zBFkClra<>ti%B$j
zFTp55oaQ9MA*HDXVL4*3v0aR?Z4u=tm4kFFt5=pNm1-p03DZY!qdVNfb4m=XWN9=c
zFV_jHHT<ZG$i4{fS19x?I?u6IgR)4+EvkJaeh{HM7MtZog2Zq@SgT__7nNJmVOm($
z1Q?}|o=Y+u5O_YRh$;Jmbz`Zxg^o_=328s%+WH}$ZtrC@zlVi;XE8>xb>Rv#`xl7H
z0eN@GgHJrl;r<q<CNEK~g$Tb$cu-2AgkUh}qh!%ub{!AL51HB9M7kMTr??iTjvm9G
zXcE|zgNF{V(%Ix6e%CvxH6qqlHgKbeH^23@Tw7Wv&o$3IbDo}7*x`UUQ?yEgHJkAG
zm1XS20Xo(t{^UoIpZ)};QiZwM1!iUzn4X?vVPS#BL<8k0=H?b?)SHFFOoHRV+`=Bd
z;a%_ItH1iI`1FGh@>4(kGkofy&yMwZ@TD5d{`@cgJhz^@jeG9AhiY}QNCMG0J6r1v
zheM($DhdEz)Zo;r4NRP}vAqtJ3Z=vQXr4I4A~4gsf=e5r0Bz8&L{&qiBXD9x))`^4
zoH!n0H3(OsN&%_0_@$7fH$-O{Ub%#D6iWCA1UktO(xK#4sd~=XlmeEV+=9|K8DvFn
zqfHH2l3;U<Qbk5&6qFD`P^;AGv|F^>ErgI{Pd$XY?-&dBy`Dp_dkatf(ob^b_kNAV
zcYQmf<}^}8ocQvuX8qz79{k1M=G@hkiP{?bruOmd_K=_wVyq?gOV~8QwRxepu%D}x
zv>3FNw7Ua-_cwo=x4!-D3`ZkA_~#$s_kaI)xO`>lOS0!L;!N!B&*L~I3`>MT8Q1mb
z>}-<^dpN@2r~*<QxE}M<GX#E!@B1%;YApzxkqmm+ywH?sYsP#zgHiUy4#CZ0e2jw7
z)&NDMn2=8KZ8s{sFyttuKq`dIx%Av4tX@6CT%*q1>>j2jCuy~}81#o^qZCugL1=;?
zC<;8AQ!UqMbvIdGU7^`%aCP|xLQ2Xd4{HlS`+>#1Oij)4{`Y@`rK{I)lwTO6q{IB;
zK1$_?_Sz=1b4Tz?5kkrFh*;tXfeZqqr^t(W0(uPf<)*~KC<ZHSl3pKCu41*sRfS~5
zm<-495ekGb2+!x@=GNx~ssC-)Bac17>%Q#Eif0gZoCsmY<G~^Y!sMnHZ{HNvkKNo)
zKl2D@&pm_1P^$#Qqajf#Vs&+ka#W*Mfl<Flwd^5u&dkgnmN#2OVU?YBmzmKBw-gk)
z!u<h35aPNay>_4Be8S$j{j{&&pnhO4oy#kfyaI%6tt3eY2x(E$C5>a!Y>3Wse8(HB
zltC-8n<}C6oG2<YH$TH`Uh@FG_9ng6>*Tp5Nj07q(wuD4yRpWexib6iD>G58;JZHU
zPKU5s!4VE|mKTS+=TM%UM2xkg%0g}>NoRn{H6`CAs@8Fv1+F6NbPyKeUWdd$l53>v
z6HUzGmaF)c20GS^T0=%|L{yr>$x`}m&WbklD*It-4(WdZc$8X`<vCeq$+7|-TqJQw
zge;uHW_*SgiF-G--ec3GwOBpk@rOT5I%+XBGli6v_D&0s#r`B3*U_9jvB2DPlhJ6z
znHOH*__13svkD;<BF{0}(CPLWjS@zslv_{Sj-#Q|>oRrd5NW$bI^1CxcM1H6BuTJB
zk5d;c$c!)yad2>4#dr{eBL&)MB$lWYQL9#PT^~opT<vV4q@q!;Lgq3%b(qoF^Mv~y
zTw(D_KFW0&#(idH_8^k9c;C4WP7spDDQRnf&I~~%!Vw-wpEwy76E>@f2Sai!)^kv}
zByo(%Mx@CI@)R?<hv(K4>JtrQyNy$yVWfOE;}oiOs?`Qw5aPHVX*T$4cJsycvWmJF
zDeif3dsYxN>kLFr&@3V17LM<u10Ua7l6Hq)e}p^J#2Sd(-SLG?0|f=zaeb0ejO+P0
zCS#ZwvQ#kABTOu~w%+8abU#zCzl*ii4xjn>X}Y5rYvA(59sE#p$14vr{n}TtcHsu1
zWjSHbQEB+ttS~58Ysu1V_YEnP1Ih&r?bbG`T*dS2l&X`=@7>2>?IINy)>|EZ^H<-`
zm%sT=N|D1T*6ckr!=3luLc7)Hndh#sJ+Sn;eJ-nnb4JtYImB1G^qzg3EY0!IyyBJj
z@jd_YpL6)=NuqKYM-}14aWdk^-u;8LcG?_2ev((e>h&Bub^_1y01(olTy5|bZ+a7V
z-gOth@f*LzPyNh)VSQusOJ<_&cDwwWfBUcbum9!0a$xU%%9TbjF3MBdJ39!GP$~r&
zG-7-p8w>TRX;PiCv$KuOHM#H6Y}N@XWyG)tXq4|3+6`+Fp^Nq_NEMI{N02yprg-qA
zqXbte3{Wap>2<s0af_f<LAk{T&#fquVazB_5P6F7930mJp+MU)LP8Y9q!d`!;!8!G
zq>$zX<kokI!XlAjFwtjieU(<LMNPxf2Y!e8(USzVDQ<bkw{YQCf0o)u{)qBdeG@iH
zF^<P8-uW)N*DiDB!=L8j(iRi7CJQoWxfGBnkPcWUfX)aZ3j2LQCw-|-uE>e~?BkE|
z55N68*lF#seB%b2JKMy6MM}aKW8EaM7h&y%K)&Da)2KHoMG;cC>}+mfvlO8$%5jj=
zPzfXIwHl?Ugy#ojdCqV+EL_nFbe6){(2$$F7;6fN%3sXv6he-5c)Y~rYRAu>0t_sr
z-Tk#Fg>oH`E}gYyuD<XX_0VN*YKDb<`$&g<2CW{0_K=y0Y5dH`B!Xdoh;%KE@8CFs
z$%zTpw>IhZM_jtR!C+XpRHZ5a&}$ppTw7VA)oNo5Bn)wc<epdD%i_X5rstdNJ-k3;
zavFm{TV2E_8H1Ju5Y;+kE|@e}>mp6zh_nb&AZvB$FI^{^ogI&z-QBO<^<1QMaGemV
z6wkcy{1<f3f9hxd3wsvkICS6uLMR-!=ztU*cscIY6rDL^ZjNOI@u<V2k9>qE64XkI
zL{*2Cm1QdBN%Bdbt(`6METt5=EZ<n?)X~F~f{@#8-NPTf=e=ACN-Qq!!Sh^(-9Cd+
zLbX=s@WG>Gxn{f5!YK!c%@)12Ym7$Q_<jXP_-ImclN6-?Eak@9G9UVjPtlLFqUfS<
zM!TLO@LY_^sn@DZPc}*75#=Z#?iJ5jqx)1#C91V5Gu0{Ty&a^E32POcl8081ONi=K
zvT@{PCIvyYia?TfhUCWLDuEv5=za<YhP@$~)dbZF6?cMGX8?|6m?TIyEIP=Q8ew&c
zB+kI8VcnEbr%zULF!g{fDOnzN$^4LTW-o4~fusNS1$}OEEJcz-fdSEgCfB04L?Omw
z_ZQ_6H?@F8?4jYY&wi2*ec(??vH^w&M-~p*a@l7v=ny(EF;!z>woFv=vC3m(yT!o_
z(j-Rv1q(7B#(1ueBR%3QXL)^<d+vIGL90WrTBg_A#7l-a${{lexgK#dDzM+$;q>XV
zEU#{m8Cx`+07oc%-vu11wKB~{y?Bm{^{U2zQa%@-Jx_VU<KFvU!HqM|<9R+4wI<4S
zNb-zwy^3_5qH&-cgj9H8SUA9Qa1A&JtSVg1u2AH0hAg@G<p|exXs3p27jIy7!c;vX
znwTJ*s3C(gmxn2NW1f!kP@YQ{i7MBq)F<$6Ml2$6+|LhJ^*1W&w9y!yq4E?JOUk~_
zptVg!COAqkbS297Frxtu7PGdEH`OE)5|d>JEZP{7ILGxA1Rf3sR~v8x`nkr8h9vEI
zKDjy1a8E>M<0|deCU1Plodltv(@xpkj+r^Iz}$3|xIbiF&a(9EefS$)j`%B7q7u2y
zUOW_Tf<{HLF)A6be)ch{&1tA=gcnkuTV&_*6O?qHN>#CR^(w#rTP^Oo=P-wl?&sXO
zWzN6wFz3%KBeP9D^RZ`HTkG*$oUxW^Jd@!N<H&-sUz?lfpZ<UT6?fitA1d%b-h?9-
zE$s{EFVN|2v2^_mpZwGZIC1=TzUHlek5i}a!u5-;gmMF>CMWsF-~GRF;+7Nqi+}Zf
zT)1%Q<!s&GYF)W{g&+ILALGY=<lR)lGS)VVkGWB_TH7ck@qG_%Gy;n;no_Aub83!U
z=L~mR#DRlna~w6qcO3F0XSlvb+#eJfoSu(Utr3<YTnlaxFl@K6IxnELhC*9tHI+t#
z{>~1AtsTNz4Zl<YDG90(HVnw(5hhEpQh{>D_!(0aDAGYI1v)1;;Fn4TKrPXvIzw1V
zDF~UGoTb<A((Uw7D9SqH-0%Maw|?ipp|p5_+MV~aedbA=Y6JI`UkR>3mMh%*t=~?%
zb%V7RuCTVbNwZpGq7hMc6`eT6mP!OlksFIq#UXFZUt$<q7^E?O{*ezO#}3plrK&O?
zbK8V5C=J8m5Un-ksEX(NkQ;V3Zy+%!XRN&w8iZ!Ey~*+LR^l|L-R@GWPhf0H7WdJ4
zT9_Pe7K}n*(1o7qrhbnCXzGl`;Jc$(ZU%S1*hLgd6}g!R&?7FKc?@9^Dq)k^*;&A{
zv$@Hj-NN%KMAar?sfHItNCkP^Cbcnc5EPlxqb?gen_RiNNtS^YIcA(=JREJj<P|DF
zq9+dD!s}lB<s3YGgy}uAl%f*(Ziebj@U4bigVt#GW^j2?$la6=S|GL>63yo2YfRjJ
z3b$M-o>${%uIu?@w2DKw-{a!ti(k+^UtV41-@p6E`SE}M!!+u(-SMw2vY5uds>o-x
z!W0@j49nLq(`&8KsD;c-O>y>xt8DI!Xzg4iDu<+LOeyd%+A{2?IIcq!Ma(psyz=h5
zXtlazS<d$M4(rR;a2(0P>^z?DQ?HfjciU{N-=Gv#SbE_Mjwfl(P2;#8Nj4(WDS;Pq
zW90_F^C$1;Pd@Mn?O{fa-MwUHT2PJ@QB>yOfg?mwiH*$-loBLq4A!C7idj0h#GAkI
z^{Abgs>leUh-$Ss(7S^^_1P)#J<>b}<sn@UAr-@&9=aRj_#RHVgirxiWF-9&xz)(P
zWi%+9i@B7z$}5sG(iBY2&gLp>>wSdqa5^~)C+}o1Sfk|E*vK_Ek{G8pNl>W}lxjGx
z&v5it<BeX<`jy}MO}^_pzGIgWX-uJ^&eP)k>v<lIqjs}@Z<;5HiGpP`?DL6Feu%Bs
z77O!DOr9WokKrI8i4z))GQCcVG#O!YD0vl*p1hUS?T2{e$)|A~hg*&xL25&{KSU~r
zg9i`LoS0^HW0lUZOLMx(+Lepo7=)+r+!8_-Bn;1UF<Hhlm(KCue)BzC+h`MOv)k=Y
zbAdKF<&w|b>>jGsI?q4<G`E~Q#W3!X#VI;7q&nu-+it-(k||*clqFY?k69>D7$Fg)
zCI?5L0v8b|Boa3YNk=hu%xvT)+C*EjK@8m?BS*5mJz(+3Vd~WqQn`qrU=<1rhxa?2
zSzhM)V8GNVpJ;9`wTUTQ$1OCy%E1ppZfvZ7Ui<p;G<7Bi3C3m!o#2}cJs5xkC1qih
z5oakv8*(I$>*IKmBp;$jDPGxUurolXDURzRTuGY5Mf{8(5~>ouKO#|zn)Nu{nPRIH
z5?1?UnPg^m4_n(i%<W4!cKac24d#iHfD6yxV9(q%wQ|Uly%DmzkBRm&QP~-bQS&h|
zU67_ZLo4vCq_1<jPd$UU=gSLHx)L;|XSg=XD0i-N-(7c+Sc~sRoH=ur^Ovq+wL`zt
z#}Nr(`y9_Mo##?w=u#ZM5{lQzDKHJMdj0EgJdbCddXlyEWtP{jv$DQSx6@&Jdz(i-
z`v{;30-tg%WMgxkr=EVCd+&J_@BD^u<-q=<#RpCnfyJ+T<pccpU;G8W=X?J-&pvm0
zJYx9bgN;A&;3xR?-}-gF`yYNcj^hwj8Yo<>HFVlLC@FC~wX1{5^ORClrZGK>(VBjz
zOILRoxI;$nh_u@y>GrW{id8OBD#BEe4tnG!BWhMDPc`YRZ;<!;WT+yOH_K2!G&xCs
zYm;HSgEbITYak^mP)N@s&r;ADJI)rgo<f)s0*lNujPXhG9Fij9%5fxkMS&n4MXg+8
zdUB3<u*9%8Kv<1=>@&3Qe30fHuV&Aey@4~A&e6Q|9J%Y_z2c21<>AjQa@)JU3G<6T
z%hK8=`+P+h`BbYRo11+C6@ja;8t%CFwe<QsoH_R#qxgU1PFXDaOIORiWQ~gsD=<dU
z>kW`XQ7V;iU7yjgO?!I<r7Vss5YpgCNw3@I%;~ed@ikw?m8EOcT78ZkJA&sb)>~@`
zF^>7#9q$$+REyocFy1^k%vitsIZ#|tyeg)c+oI4Uy)7=BKg)LOI$LX(3p0xE(d)Fy
zHn#ZqM?OIwr`+?vD-qQa;p_~hN}cWX8+2OR)Mn}k;o`cQFf7x#y1{UeBa{SFQ0QO!
z*;2}<R;zRW-S=|u9S?Bm@IelrSVXx+$JH3ndA6%N!-6ulVA)MJ=qrh}nrsX=u@nZA
z9D^eSTT9FA+uFjf)e%+|qySOmqbgOzbX~i?#Kz|47kt;BKXZm3`>CJi2fqLNh)Q8G
zg0p%IV-mYMP+@Oo_F7i1o+EM`T(80A`ZB#vhh9fhtHWs2XU{?tB@Ns32&Kc;+6^Xa
z4QysOc4!~ht}Jo!g=g^tk6NX|p7}Y-Q5CB--CmpN$!XH0kB9=w)qrldhaZHXEb(AS
zl4fjndi>h&et=Iub{dT=qNlXj?SC6<NK%bbIm6+IPN&QI+6vWL$e`Q7kq#SMD=ZwE
z;K0H>mtT04qs=l_<Y=devGn}_U%4m%qYHCH7~qs6L?%fG8IF*}I8kb(6u6ndXi&b7
zUkVE$ycVonUq;%Lz!CUSL{y6yD#h}4&i+Y{PMRU77TCt6m1x4|9AT-98-ysw89!e?
zKW5oqw|@E8euJ5*S>E>Mx1gj0jzw!QjNR9^IK#}%sKT30bQD;)c=1`bcdj#0t1;c2
zqSf65JofCF<Km^uw7WejRgZozp|#Vea(kJndW};@kMOBaK17xqhNF~TYnz?+4pXx;
z)M_<?AY{5R$=Z!297kZ4=K04S=ib-97SAt{+k`ZaG1hSI%5{F?=YN|^*SC?5#2H_b
z0T9LKD^4?(mseR_TqMsk(j<XA#~6XOnrfrO^yCc7=gx2_N)fKYt<({YK)GH~usH?Z
z$nkw-6rhBmv%N`4MM&QRZ3;f2C>jt*g@Yv9?jdR+x7>XaRjNW?kPb4^?l!|#pBtSa
zPoH09`D&ZJZ+Zij>3M3C^Ek2qPJ2O!AC|a!{Tlz`|N0l7*S@}NqXZP9c$sQ^A<2!!
z_WSr{k0c+G5BhjQ;Y0x@j>+NyeznXf&G4KM*AI)xu6T&692`dx$HU?_=z6%MM6O`e
zix^BC;+_X?q1>Ecw3V{9vCbeKF*SLJr=K~^@l&U`^WK9TIX2Jfr!JD1oO@q$m~-#5
zth7ApwLX=|D`G;%r=u_i;W{A45kRdLax_HS97ikssRed=n#z@@snvR9wK*bpn%CYs
zkG7Jb@LAv5;`C=e$dzj^@WRN@w|2J^q`V@ZFH4K||3@DDB+oqmG@Z^iS)S}_jPg7u
zj$^VMvMj^eE^-B`wGtD}4%^%7Jp1h9eErw`eZKn5@1PWwFrH7E^f`6%6hHs7Kf`zb
zlkeg5nX@lv3;)K;7{mYb%fHB>J^T5}*S#KXB$e6()@Dcqy<Ue%1-K{-h9rscJfC`X
z0-dMWOp^^eZ1;Dlh6yDh@xl^LrGe{(*fb~Y4#<-fEg`GqxFw%*bAn!)5%>C7YpGN!
zblNSvJRxc}8EkD5_q*dxT@~TF2nDY6$0!Jcuoh`OGLtYG<*e>ZaAQl+PeZg*rW%Z>
zs2(-ZqF(P&sg`l1$K=Emy>5@q)lCM2guoHBKJ`JO1BX#=ndy69MfaJ{5Ipw;u@_Q0
z^=ga|M2GKW@ylMv(g)th=H@B^C1z?hHr6|gk`(2K1j&e{tC#SC@UJZB{+C#`z!>y;
zeS#n$iXvRcW2@Vy-`zpUae%w9D2gsr7?zlvZgQwG%V!^bmS6tOKj7|DN2&T2DK$z7
zgi*Wsve+V%71$NwODPL6^h;QF;%0Gf$1z^UvUY8WPk;DNNjs~QYgJs=p)ozl)Wj57
zrs=jj96x>pDHT!RW7;itx<lGQfLr#+ZO&*I6IJVghwu1QD^=d``Y-43#!fK~91hU7
zIOH0&I(z09nVVl^dU}TaiwBt9H;GU=Cbussdd#>3ClE&O8ait83+$Ai!t@Qfmb=-s
zxh-OilyopVefrl{sUJRsK#*&Lay%T@D_Chla`Dnd;`j>znf@Qwr#|x;e)-pboqzJ3
z--+irV8<ya;|$;eO$AUCoWo(CTC+;6Tp|h`@;t-!p<E3K!;sr<IZmhFVsa`c>9#St
z$J)vj?z-!>#BrHQDWucy;J7~1lT(z+6-=6w*&a?H(22pCoI$rmsT|-}N*IxoCJ9-R
zvAW*n=YRP<JpA-!GMnRgk}wR((i8*up+{5-@coclwZW-Vw{Y;_L5>|g#Om^OlA%T_
zpG;>QJ#c_QYk+EBrM5r9uQ!o?h_*T9S`F(ca-+$yc+~(^@rXx#lm$2Nu}T)nCQ25`
z1zC=?mZU#IRYGoDUZJzqqdHM%qE^C-0#v1h>pFO`!>UmX%M-;2QArZVr`%`|lqxuG
zfd;Zjlp|Q*{QSB$f8EM+%}@W_&k+V8Z+zWris_Xq;=qIz<FWgT7qFWEYHJJB#|vkk
zq7-Q6r{>5rpY@F)@u)>9>|m|NQI=BV)9=S@Z*4NsXfW!IIJW-)t~Mn7HW$yIM(d2(
z$tIJNlX!kmu;W~p*@-4Hjj7c`W*6sBu0(l~;UFPPElbN={I}ou0MDFR#|lkUDWPOW
zt~JV4)EgCoz^7iH;-0(i;>0aSxbNQk+1}b@d1-(UCAN32<9a^66VO^&Vy<G5L4fBv
zIIe?KPN9QU3ZX4tR6zwLtjV!yOky?B<P5|)d75GY*Z0Yj45L#-Zn51FeYJ^SGq~O~
zvJ|k@>hb6^7kTRZ1{ljk#bM^i0cy=Tge|6+VYy6Diny}0#P@#R_kP~p`Il{!fP(n$
zWqrKS1|m;!wM8eD0zjJOD9<SzO;V7?3EpG{B_&2kP!7WJu&E|Z<FUKSrPpo~qwz#S
zAPj>v<U*RzdiXKsgZ&(y%-C0+;ZHt#kqc+r$Y8*yKk->^zhjfV`^u!Fl(prU+2$nM
z($F5(=?|{qyP)$NlY=86OGYT)!3c@SG%`sry)J1s!Vw|<#5BVJ1Zkh-(u1_O6FRvh
zH;P30Y&H+DoM^JvG8f|!z1^_lBB!<=XP@(0EiK*P#`WvtAn^h?d2^DcwMARGd+P<k
zXgFYJdq`uV#Bey`SAXpnxP19C-~Mgi#nj9!o)x710k@qz#ZSKb$NA2G^j+LoS^k@B
z<I7%g9P^_;{cdKh=J=t*Fwr0iBE<9@M&!it03#h->4Gv02K@qWQEre=By?2<iHq#^
zNL)!+YT!s0I~o<ZlE6bMNlJ0hTOshGh{nPk-OWvUI~`OqBnaH%L*u4I6HSJ#9g@KS
zAw=P5y{SELKx!9_V3;NBtVKMrGD$0#W8&ChN|QBWWAUSa&wl0t!Rk7Dch0iVSYoO!
z2m{6J^fc|A9nyYA97$wni|)gpAbjABln)-Gd+j1_XOqhx_!H(Tb=2HZQYDzU@5^al
ze2%r}E+f1R!e)bd;4&JGNL(K>&DPoyef_t;aetxLP2G&KmOM=u4hJ+RCh&t0DHU7m
z*RlDq7|&UaNdfJG)i|z)ABG$_aDZ~D&UU+nuQX}WMY)XKrDIN;6}x|y9*<3gltnC{
z6uUxwtu@MVa8$v|8V*`K^y!cC-aq<NHdn5*XQs*n_ukD^W08f~eKZ;cq`KGIA<ahg
zM<e1qMFuW=_Z=oGMQm)Wv9+;5c4LX!WF6rY47O^m!nc0gH`5+wv^HA|hcQW<AgqUU
z6;Z9kL}P+#wL)0(Fr<)Kj4n(B2ym1rEUgwy0dT@-Lv3=JXljzQyN0lG*UVGUSu73$
zooiOkUSRRwdoYoUHijS!aa|wJ^ASSu%yZ9v(e_>&!|%Q4kJ!6s5AS&ETR}LOTmvsU
zBN1R^p{bghZqjLo#A!;a(<2Oh4j-PU6qP|p_Ut>rtlrP&<`&toOMC4)r=Nd<g9lHt
zu=g;H<`gs82*(RB)-ddM$de%wMOdnm$(Vk-#h}+?(CbpF)~HsijPjK3aKP$TN;eKk
zDNHqaYB4sYUahcq?>v=yM736BdTNH-ZoP}chZYIL0IO5Bw$|wnG|3>P)ovqEES-LV
zlWvUbmMK-Lc%Fk59#(?udzfK@BV6241?9NdPLJO9fNEn3k(;p*un2Pp3Bn>rIfJzw
zX7|i;_|)x%v8!iDl0J@2!FU*%U<TVf_K9WoyygL#M@~?io+S#x0`6-;c|MNo@z_(3
z@dN+i2frZu-R<@Gk)QZ+tTnvu%U)gNFFO)r&G>S1)4(+DsN1|SdJMa?*Ov)YfUvN(
za)oXyEpR<*L?!Y$cwmvOtsVC5si1Pr(v@@Edi-{Vql82I7ujiTGfHD-rl(n4Sj6!I
zbTUHu7GX8H)%3Pon6@O00+M)$7keb50lof^&pvjAvlli=jm8Rt?u`fokNLT2X67cD
zTbQL<t8nY7J2-Uk5P>h*++1eq$`!UZ`uJX*OV=(TN%`Oh-p|Q+hndB_l<Eys7@(9Y
zPGup<Fc<-T6r$BQ7EB1TF)_g{RZ&Tyd6r>-z(6VyMxwKv(N>q8oi3~GD_ptO;p)nW
z&0!y3I+PuVYGtPP9K=w>16FDcTtDF9#~$T}-~Gc}y7KwUtzK49r-V~5zpTL<?I3!~
z*gR&GK`5til*ILl+`@FbgH3a!@{z8Gs|sLfu8VX4&kM*BQ`pUu7|-_^^+r^D2iHn6
zBhh$}<!p5ZG(y4bq|05m&T;uhn>g2GLyx8N>m1lKPbF}<*t*K{u*XpB5T`C#oRMV-
zX%b^IgYhILiSf&g!o``!ghG%Fx~M!OR{=p-BF_cALCQqgrxYm2QmpXV$;({a+2new
zL%TO%dz?LK^-Z7Ei);61){^TY)@=MQc1KR^et+|K6zzIFvaGKT81yr;RPe~7pJ6nL
zdDl1n1CAXzfh!a`PkG?}`}x2B^}pty{~!OH;VAx+sIFGl*Z9Go_>cU<qeqz7x5(VV
z6Zlh2$|Z%$HGP|r2u)D(aBNP$-$O}7DGaf?L7ENFaZVKYIM&4u`UQ;E_lj4<brE(P
zUz22L0n+to%+6A7G|97=q&Hw|eFZk#Oz&MFtkxKIdSr2e9Jj!w<AU{&V=?#U`Ya#1
zzK7t*F{b7lT)ujPxrXAFTMx6jvdZDR_prWF;q19Ny4%k(>_5!>l%Z7esFW+L4Yn{N
zi<Oe}>=Q6KLq2zq@W@Hh$3Mx&nHQ*j_ya_5uj5FcVWq;JulfeIum2pwwJx<Iv)HnS
zx09groE}9`wu>PB66)nXpJnuT++qbuQt04Hr3lyevD&bD;}X7W(GCPbk;N$-jgnBS
z)JRj!*2X5aN|okBmDS}dB%`dzl2jD8oAJ6iUKL}YIO9u>H5MUnrUA%t2h3q-dxPKp
zt)JukxhLtgV)ESJ`vKKzjYhprrQV=itE2OTO1H-HjTM9zq@$G4aKyE1mnoMc%9Rqw
zZavDN*JI<x4XU*oIy2}@P-&E@%~Uyf@L<tdH-<RVV<>u2q*-mqbyDQm=5`#dRg8#@
zkO<{q#~NO&z)FQTHAUs%e%7z9K{o!P#{gGEfg&~*lIt6Emu?UqIXD)37X_ViTymXp
z{=)e$>i*|>&VT-e|H}N_JP+J^59TE-Kr0GOoiZ5)!*Doa*c%kN;Mx+F0_NuCh$?jk
zqY=}yvq)nRCTAG9B>fIs+v{Ar`U3Nd2M8+-+$<#5DV?n~`mG(5QcO0-*{?e1@WMfo
zVUO<iGVSgTJFP8ZEy<h;cR%puTv!<~N)ygqK1XgeNt)5^4=^a=Skq`YgrUz)YmMig
z-=W*uF7P5*jlgN}<O`2cX;ygrsmG}AZ1Rp*?x)h6#0dkOFaT>wvJ~a2;`=B-gaOud
zG1{V&A-PU*{Rk5$1vuaHu$jf^!Z5)gMf*e2u#X>1p*#<zC7T<ooV|3F=dSHA-04$0
zdYFm5`)JPY#R~(nG($>(A4W)}__GgvfFJqEAEDFje8Kj+*YEQmfAS|W#_)#MzNT<I
z0~U?2<2)Quly}yE6%M0*AK#NqPBv)N8t8P0RElAo(3q^ysMI-i@))a|JB<26(%v>}
ztCzU8be>bUJpf{kxnzL0mQtxwyqEeNbe7<|B}zdAx|jr}SwcGO5_h{;VHqST+Np5z
zj#qPWb-*K!Kg=i@k!wwVFrri%GNIwdjdjZH4((Ql#~*(fp$*ogRHJFC;UZbQO}D?z
z$zumeTU#7in8lx(!fVuUZ*p=e_@N@VI8lIf9gMa}-^DMLNVOqLM_A?Il`7<;J_3XE
zBUD~O8-pGt^hX)by|BUMD_uH?VMk{KQc`myk;o|5CsA&IQVvnMj^p@z`m>+m`+o5I
z+1`117^{~ze$`;KfFwb5wooER0gf*ad5>gJe9)B;#H}uVDWp6zjgS&S4p~Z`rkK3Y
zw`CS`EGEwx#xaiYD3#nIMnk%ok;~-4D(g!-Txm_9Yg@ef)`(*Z0`J%@eCp#DsfH7b
zvM#PGG0qJxFGVOvp!J}L1j<tS-5$nRln{h|K(#)F)H&&(i(@o0OEFnaDoo*|vXWLW
zVWHWiE-gb-VrAg--1;(9r<t^dwax%*Ku9qTzW&^U`7i&>{*^!bA4z35|EUf1dl^@+
ztc`IY4ufAB@Qv^K2i$qbT_6lX!#8~W-{WH+{V?zOlRqW@5`pYq*xBKI&pgX3&t70F
zcpR!Bx*AZKZNl^n{?sgXvO!QOgLDcUgLLuS3Zg!P*Bar*15%S9^Bk-}Bn4#I=t5|&
zM;WT*VxVxiItmd6sKCW@JZ6N$&hjb`K6IY@UU`&KIU*g7&}oYFyaJgc3`rWZvoXoT
z+j}|mhS%`fN1x{D)7J?rCGLOo0ali`Se%<-%iAG{sz~MW+;h|D-YFdW7&FrjQK^hz
zo5UmtQ$(Asee``yz2iH{D^0qkX*#jNuC8$7L+_{Z`mdto2G~QlviB?B%;UfHUi3zf
z=E59ntxG5k6(j?p@x^%j_9fZ5&u`sCnpgx`o>Q&Y3Cd-BFC@=Ltgl=^sT{`<IF7&w
zf$LgA7lhWhu0yBWrau@kH#>th*;t<?$c=spt-e?V3P)gt9GhVZvyC06XSlA5&^hay
z*ZG;B`U(2|Ym`eq{ceoo2=?q-pgz%{TC3v+A$gkPO2zEd9GjDy4Eh6396!a}+#<bB
zn{Im>D=e|HY^|-4Bnc*;Aw^(i38O-anEIk~WhvxE+N9V#1B%hH)`jWA>U?a(DsUxA
zcvvaNI1+=_8CD8ZRAJ`mA+{cUhO}dLKmA3oRv?5Tw}yNW)4FgOZ||O6Jc{cCDCw}i
zy}^~MSH9@)f#Gn-zkm0;dG`<hJ5HWBf!6v(wxStdK&>SRC&1=}r4sx1FOuaMje3(x
ztp=?fw$|2|o|-1~Ttcr&(&^!$SYElx^`-OdIdBp`jEMVv`t25%E?#D8b{^y=D)8}x
zD#{B9EM%oBgJj4~r%xVEGPCz2$}90TZ~bbNDMs~YFP$aVno*qa)YH$5bzlPD^9Vza
zS|wud{2crC9AS29o;2;Tv2}x6jvr+C(va8PdkZ46lo#evLLglSYYeXRajYV-7OxT^
z#%_G+xL9Gy;sH(&AxkAhnqY#$3@GCaqa{j1vfU?3a|W~-WdS#qdz`s=o$DJdgyS*c
zD5j6@r#XKB$8$-BBRoGOh$6JH{Nk_u5<mY-KVReme}Px8-{;4E>Zedj@`l&GcHA8-
z9HQfB+G2?UobLKQ^~omjAZNR^gLFLV^?<|Eix>-yiAfgr9>Oa%=yf|}gBq(>FY@%0
zpJm^HW6UlbBrMmkws694tu2!d2RM#LwK<LOT&hunxoVYBdyCfg4O*QY2JJ2b?X%~|
zEle!>q`4t<eI9<|5nBBg*24PcHk;eqy8!jj4~a@Xjar52iD{;9J;lVt1dlxWQSQI*
z6`0Wgy|qmgm6@2D!to?Zy5wV%RBjE@Xk_S+nH(h?(k!DiJq3=3l#(>*lS>Fa7aePc
z8iaP}3`d;1(&FOfZ8|;Z=g`)gPzY+SqNH+MX{ayGQJI;=sWxzw$7eqK(BG(_gO@vg
z?X{p(BJFq4oh@WtU_{}PGkH#)=Qv8?_#T1Qm|Qc;5*+2Age(Gf$66YTVU#8GdtFSP
z;Yt@BCp4z!NUg?K340r5N_*-kxyJNVm0M5kW7uEk$gQ`~*fYyL_uPr{6xzg`Iem`z
ze&A^s&C_rcZk8fd0Zr|8254ib2PGz^=5VVu@-!ytbqGTr6UUe=Ay;X^juJlYPLDW_
zDK%zU-EsNo*%jQqGg#@ea($IFA8UcWh_Ow^SmKf9%H<W()R4p(zw)d94g0O%!5z2V
ziM56(ium5|{a!x$iBGb%)B2+QU0%8@@X%;PMYy~|j|lUabTeV|@)BA(jQj|HcADmq
zy)=*Rr#v-<ipn^yL}-uUAjQqo!T~(Y$dkCpQ*;ZiS*{J*wfLUIjY>t=Z9LMF+TsR2
zPTc_|h_+Hb{OK#a_4S8Qu8YnM?)V_q+7J%|o>`m5x#a|xmR8u>idj6+;N&ajs8nEK
zeuC+#IogkIvU*`ipgcxFm1}B-aC@3sb*mVC2!+s)<(euQ(QmQx@%ORt6<<rTa0FAH
zU}fh9Yv-<W&uw?no0_HORoM52x3jj~<H`4bl*zkKGC#4-wap#~r=XBim}~x5D(VV+
z1xRV=bUFYfLE63`4dtkc<0{(QZCdM>FeW1iJc5!#mMXj?WnreoXCL|m>)S29@=b5%
zt#5fdQWjfh9EVKn;u#=qkr#V&3=3l(8ZWu-nB7P;B^|Ju_rCA9xqS6G4jz~$%`}5p
z<2Vl0T12f@M>>jB8$#uf>l~pRCYn>EnZ^i1f6&EXs5P6!!yfI;O(rJis8q|uIwk9N
zka<kj+lFWoeY5No1w=ScQ?R-)Ls^S3cI<R53c4a`$6)fYYgY)YHdviubcTbVwzxq3
z$UZinxQH+wT8s~ku{lLsklNBdcb>_6Pa$gMV%+D2faL16CA!`2m*jo0zOliN{->Yf
z$N&8gGdDX&@wn5BGd3_*GBr7eR3+j#VPdYqaMWWoiYb*MDnZDw+oi1>MsZB7TBq8W
z#E~Ao{)p$Ff1FaK%HI9A;<_PLNJd%CV^2NB{{083PS4?q44w5E4SHlICCxNQhl$yJ
z;MEWU_8r*ATi^Z`DwPUTk4*9S(~r^}40Z+cR#>vkprodq_bI6ncbr(HUa#}iGoR*F
zuezUl6tMO5lSFO9;_Y|i`VvRDSXpQaji9i#lw7Qmkmn#A+$hAaR6t4MQID`XNhU19
zXoM5slst^F<iim<%UNC@apv4LmNy2p63M{kM9QHiG@++hJbDVL1W7g`D%T0Z621O_
z|L_w(!k>Nc&&a=sa*bZU&rklBpJicwo)gE8VQfLV0E?7G+sX=1WGF}BIu1L7nBicA
zHI`bn!ou7_(T-P&$;n9+7GFUU`t-NA=#M&_f8i;rjY)#20Z96-9a>wPtgfswGq;Fx
zSDBca!}rUD`%@^~agxMflrVYpPHbtC&|1FgtKUSeT4Qc@j!!@O8LnKrOq%ByGtQhh
zhM|oM4Q8yl_2~UH>Qh*eP^*<Vb?dEMdhQu+x$h3{II$mRagjLb(C-gXNsJJNuv8|q
zL!<zmWjN9)^mTp!P03&|Ko4WmQAWR3A#XGp2-sZj^Z3UeC21Kl6oWitM<<k|q~SW0
zq(Rygs~nmK4kODAl=OJ$vk&ooKlpurqk;}zW>L3(`YHT+m2}i6N=8UZg^1FE;}kqF
z*IJ}1fU%4cM4Tf-2jNMaD5R#QhzCRBK_7SI;0Q^kEyE;6r-tn-H#l+XIH9vnJ=q`-
z$LVftFzT;y=lusbar*;UX-Kn-D@zwR^THLbFRik*)uA!H58LzcRxFXahPAM{xy@*l
z;CU`HQ!_-BI>L2HlM#cR9m?erCW(upG>ysf3?W@c1A`F`DyZ_@%5}_Snc8%nCqDWp
zCNsqw*M4!2;(`eUNo=`(eT^*3NVAk*`sM%1KmJGG!!5^8VT|UEJMQ3ZU-LEm%CG(U
zmu!%i1D_lXm@9|OX9nMONi&UaEp9%dd+r*iFI*vf<SAzNEwFI(FqMTFM7cz&A+r{n
zj<CZFlV=DFd7dFWh4B<#wTz5>+-g)nigSZaGfbWr<_9GxM|Ez$;{eH{SNY7N7kTa7
zhtX+@PBR?OBTq(bwVJG`eN;*nZai~_uX@Kxj-9F?T*aP)WpeBD)MIB@zS3fGaWBt2
z`wS}U(+@qi)e+k5C4y)?esC;UNp2zZB<&X-r#rou;J`uVU;Rq1{ptH@)&s(87Z_E_
zB(9HBs&MPq{XIT;{t9cG+tiy)@^;L2W*8z+!V&^~`#<(=|E`vjMTuc_F+OihR0)F!
zp)5NaD|9=n7;Pv;Az|oKD*04vl0|B4_s{XdnHPA~D__mzL}8K<6i`$0oIK0WLf|+K
zu2k6ZkHt9oMHo|X(2Ob49Hc}Fhili*bN$8{<`zmgE_8c2!=a{H@esBMT3=dPB1<ho
z6d9jttlxEf7vTt`D>0(T>n!_C8ncU3XBIGVOs}&>l4dB^;-oFK6M_>0><YuB)i7pb
zS!>4}y78sX7H&Z+D3WY+UO0%22AyH^q!=;R>r5X%!S?AJh=E3nqPwZE=<)wsNK3M@
zLGStvDhvCO(m^Q~3q1MM<6qQ+{V%Vx=g#r&pZ?GMhadd6gh5brSjVF{;X2eOW+|1M
z40;=+DG1@Ry|aZ<igGC&yN7ei<q}ds5P5h`lYYC+#>NU)E}mg>Vh-0Sw6A*(?B~Hp
zAIAwI4()8=yB3?dB=L})c8mUK#O(YceQhDnz;O^#vj5;d0{=}k8?ziaaD*qGd4fyV
zuhH-KFggcq3EdJ$4j<>fyI#ef`31WD73OB@>^ZQX<tt0P<!j%@{d=b=&&`p<+YDPB
zR5Cz{9M|_rvjpW91-_Clxy}j05_WnTon_>4Os}^^rCP@h6=`Nr+9l34EA2iPuWWE*
zC8jsb>E?=&NvSvvHCIp;ny@j6KRJzZJxY}tj^neuw#N7U!1wXQ(@%WS2?qdM+uQu>
zfA_=u*bn^>M~@udHA+}JZr53Z)dnFv^4!7XDN~aZxUPdXIkie1H}Gh+TCA-s<9j}i
z6!_&DGjsdtw|2O?bcu4c!Ld{KQmfB^aFI%})9G_z=^^gA`(DCE6VFdbhcQXqBTf_I
zEN60Zx@eo{2B9RSa>Q%j@EV%UDHazFaOV88Jb(5qE1T<#h64naC<r-x@F=(4dKZgx
z`<R}tbN1XL95}E*sZ!@FzwTQ&dT<d(D%vW*4qIIKix2V4Z~PH{W|)|qU~0C;>|%pT
z6p-Ww)u<68&?wfDjGieIS0*vFIfP%sO2f*P$LaSnlu@)8($$)ZR5Tn#AT7cciB~~X
zW^!Q>*Y)`L$NqvJ_^}`Q8y9r&GK;#4AN??sa|?Kth|qPgI4GN;l#6m*@-Z!b7a|%2
zg+U_Mj5<9sogsw4QHpB0N;*o&bO);ygI<E~OE&wG#au95(d-$mvDxl%%aP+e{_uG|
z`{8qxzQ#wnl4qZ~$nXBvM{(^OCBK5_N1VESj*00C%a{I;=9FO&$82r2PzYuw>r72e
zkz0d7Fic}ad%KAD8TAmkAR|U68NO?Xjl^>Tbmp<GB{R3|<MB^DRz#brLO=G!U&WwS
zTLXz;ZFRed>ajV$_{+b*zx=1)&&<LszU%X?fB#$f!}t6Vy?*~oRvh%L=Ff+1zCNfk
zA>^3eZ73;;k(SIjbXR&@Tsh0uGgk>30TX*?XzrV4vL?uq44uWO@x#JMi_sQk1)i{Y
zu0z`Gk`0U8SA?@G$I6XCBPfR@_D;ivt0_;Nz0O@H_ZHoRJSEK~TkQt*gAG<Tx0yPq
zsMIZM8&`1ri1+-_gKVvq*uOZ<!onnR-r~+zHhKR1$C)Z0VcT0^y<6tUge}l0vhXzO
z+{I{*JW2WFZ~hLi{O*5B^}v4Iy@%-UTqVACp2@=}xt{M3ltRLZCg1vxzndTV(Rbrk
z%lPGp$<BbG)(ovAl!meu^o{vnqOdE5YZhxXN?59u8l`dp9lmkpH1VjAGq`SHdI&t1
zz^|fB&a3V{z`m1T!&~0|jg-QO*3S4)g|S&m6!38i${iPULLgMZeKN)%w8nAeSR1KG
z)13GI>F=|0V+GHZ2s<Q>L1zl1C2^8-W90_RYg=d;a$|L!!-ozs*=*8JVwC3+mLjC*
zWAYJFg_LWPXsNI$w3G<tVi9z@JCv-!6&|Xxr@)cu43lMr<IULJxW1b#y4`WLE!b#c
zoL5|Q?+R$L!62l<wGNX<ju6Z~#^}mA!YcImHP?)DH3j5}VeRxK=5D_SRrPR`%Wyd4
z>1Uq)lD{85_0U87%rE=`|HnW3XS=B}2!ZDWc$Fp-^*N-?aovDgsY06Mw07Fe&(Cu8
z>QzdW2*<sdF<KP2aXKQ7Q!ZXSO*xF%ckl$&@&q%pi+ue%zMjV){VW$>c%INbh41^c
zT0303c7?%cM5)%motP!hi?O%sx@ak>%}nsXYwu@fafTyDj?wOP+1Oeq9wvCMOLJnH
zO0CXhvq2K~$<2s6@3{lvR`{|vzKvQrKw85NE+Xl&{LrU(@;Cn<f;eV=&pdnfO*1(g
z;W!|a0oTDe3O6b-==O*cjj=9m?V={880wr}H|L4>f0XX(kW4EwZP_v@xiQoohpH<m
zsT@U4ZQmkO`wyUM#R%@{XP@DRe)NZU;rt6<(*3@E;|AaVgFncR{<|OI=;51%-I#u1
z3#5sXg2g?D*;;)8V-1yRnZcmTU@)K}C86ihZM7(sBDB_&N-mW~1J4N<_Pd-p`y|zB
zgKD`^BtiH-waF$AKKdAQd-gLuGli=Ru9b}9A)DJ<#A!m!cS&u54Z;>NhSGI7e#b3L
z&&;!Ua4)a8^FBJg9-VGseyLSzL{Xh`y-cMXaCPYtuYTR@D3?Rru+G%{ei9)`OpeGh
zdMi6T{^uVjU5go^=xuGXesP^=T!U1S`8U6j{RfX=(uD0{o3?bxs|vqSr{>lXZiu4?
z#B1B=OwiH^U7b@=0d+?Z$Q&0#=qO6QLNyxrQ`7wVAN)2y{j)zwoc!&a?JujS%PwEz
z+=t)C;jelVnOhhg1X)3JS0H>vIv7z_Vc}%z^ysW_)85`D9VJCeS}FV}B&bvwYQs)@
zh;ND%2`9_hNi{p2F3n(?iQpV7r>pGy^0(2tWs!5IFR}l~MUq7Fk@sE1OAg~GAFUk@
z9-idjzCB#{lMmB0k5aEkTw7X0>x{`}g}M0!JjY?D)513b;t^h+Ahe<1@8KiJvJ9Q4
zltO`Z!E=1JM={O)3tYN#jrHvob>+}AhQSx{_!Z(s?%YiwxfweIH#b_iu7^VL-+$wm
z_-Ehq|HY{&ZoBO^Zn@<Ik39O=7wzx$SFL58@X4glo2cPC#UL?9<GF%T3Ym1-$zpoL
z1TWESZ(ibZHe&wH1#G6l7>p9=KoT^oWWy1z=P`_9lC6ZDwN+eQq=2X(KsgSc>l6@S
z0U|dfDa=j!tgLlu?Hkc7moZ6>?rZu9^b&y$+w{9Jo7;*otkUaRw$>m^a?YJyWpX;=
z&EIeb4?pr46Z6v?-FJwfVdy`tP)>&HE0ALcpib%bw>f_?C(8`-v5zqMmbZ|-@?Kg`
z4(SL*1;KDM;M#>}IdR*)EFM10w|vL9^Q*u98;DYX9S)gP4y#B8X@<h$3P}GKsOt1s
zCIJuxKBX`MDQInO@$};#E{sP;i~+%5O^mQrtm&~ZH_xkI`%0QKlUM;+mW&MqvJf9D
z2Vp@fg>(v-t}0H+moRPx$Ra61D#?X2&(Q7OV5(kYFzPev#|#D#r_gj<jL|61VP;_u
z7nZJa@r4UK(CTvP#9_LfE$X!@VYym>fst5IKuLv^*lbi-RZL+JxVm(iot-v&kKfJF
z+b7VdqIl4HH&wyBM7wG+yE?!^Wcw1Xoftb`Z6Qn&3WNo3vd;7^hq!Wm9h=!*Lq%#0
zzAYT`NhWArUM61MpuDh$G|Bkn#~<g(r=R|kz89eQdj90k-p9<;6yN&I-@F_DCa|a=
zWY2+v#G`dQ-=|WGn4D^I?%YMTc6QjaxQA}HOB6+nMkD&YE@>PyN@A4b&}w%$f8kk#
ztC*WVz{dItx12b^RHH_}zfFJ8XLWOh=T1M%3(r4CRH{&J?q%PU4-^jRG|NCJa-E?(
z!{HML*t2he?Tr?L?g*2EbfH|S;RKT5phuRLx$FMh*xXuUW^NI`6cOWMZH~wi`q$Ta
z=ubYt;L;8*ime-$IsfD(f`&`A7BPA2E$qAZ4usKUgAv=gq~`>fY8_dhM)*~PuxzeB
z$L7`mHxu+lOj{Qwu%_csb_AX%#`E|NGe=J#D@~N}_|Qi_#E<^ekFmC~_9fqMfNM8y
z@clnf6n4js9NPu>-vokd942S?;dm92v`?N1l=4_zT|ucuJXet=F{&Ej2T>uza3QKk
zq{)zWyUp1%PvAK&&FKY#pvEg7IDily7tfug5-5D<I8sU0)>gQ1;RS}H1g|_nr7U*w
zN5)#LROER^x#4sB-6t6iVmjMB;$ec4E`H=vsn&5rMLZmF;<gibo?>fzi;2l5CK~|?
zt245Gi>E&LetKsv5g3V);L9REqo0TZE!3E%XMNiJF0DAj7?*0TMy)b|>-wYuvQbQD
zt4%l08S0D)C!nGvuG9p|QgIZXE6K@F)e67+!T0ksfAHHR$=@Dt^|FdOX*A1EJcD1U
za=4lE$ivStx2H~XY7)J^!?4q%>Pk9!N_%^UL4SxNJwo3ntolV{j?s(;BRbnHtfQ!x
z1F|G$x>9CsUDL}mO4$%AGNK@6F@A=nzmOcf=e202x$tbt#j{&vgNQ~&5|%ygy#Ejt
zE4lh-pP;<;Aa@_DGDvn9jk;8;CFbWBC`C20B&I(a(XbLZ7~*9Kt&J^4QG;5Zkr|E7
z6T$#&pl}_Z%k4ICrODC@S13t^M$<Ef{EK+_{^d|*06}q>WSM1sb&D``dF<(j`Pg6l
z1z-K<w-S~@-uT8h^5|oae@WW(FI~^)1Ll;+t-?nMg))}Vu|&$EjUaPeIy%8ulCtr*
zzOqCANF6UpL0UxQ;nm87wGudjpj;-<hVJqjA}usbI*CaJog!gFNQCR49EB7vGmVI=
zt1TKe!}SYGOx$stTo<VbnGl@4beV}mA$Q*WDt1=ac=FRrw7NN^isaV&sx%uDy!Gw(
z(;u{X?BQpL%pRhsL0BI#v>}nFs8u4;w2QHpK{n#-#U7oG!z*qLaRzHF|Nb8^_x0aE
zF!@#%@BIp9PMqNSr$5894?W1m=l?rj{rCSj?tAsCSy^4@)1UezE8R9uYB;rjk;gAy
zWti<Mxc<&pJ|2?_Sum@7&&A{!AOG;5apu{Fafu6VlYse!Tj+F_aU7o{)0Arqgyklo
za<Qh6G{86wVhj{^90#E!a*PbHLLi0Ojr{?EP%gr8F&NIAd75FrLlk)UR$+`K3>}a;
z0!^-Cwl_C<?Hj%dgXGAb1=7KYjg{+^${}+Li}+E6(at8lUXRVKHl?UVeP$L<dw9~I
zMNV(n<LtRJTt0t+%h%VbPVJ{Tz4(%X&W^hlH)R(VWA(VGV^@w*U_UHay?Yq$)~ga^
z5HNfE7^{yy$!KF(0K~_dEoY2+$*n|ndt86|S)wDyc>0mY`J+Gj*ca6G_;Oa3W&HB5
z|2lj3F7g$xf5UhTTYSbE)BEB4W0b;>a=A(v`P}!a2WanX6Gb78<1iYHaNQzVViczg
z;~}2o5xON>J1w4n=3yqMW|^o*tSn!`8bgw2T)S|F=T1M%*|X;uw))iSb?$mik*O<m
zh6QqMFxryo94Q1!Ny5Nk&)(U>u_**rj>A9^NI6|d3<QLgDpE;o7Gp`lWF);dPrUC>
z=sa}>U&}%OF9b4|Y;9!hZ1)iNOyQUJG8%N)PIed>gA+9fs#R)U4dI3;Jz#i!9TQtx
zCTB;dc*14ERYc0*pi$NmdLDkY!u-+W;JEzSZ~iJj`%6E^VEFlEeSWdlwHr71KY#Gw
z^0PnrpO~ASeaX25z7sInoToXlNWZs=>xGoP68+wY&5bQio;uF;>({8}H5|twNfXj^
zM4luJlL1JFrR!G_!gAu|-83d<+1yy=zI*TG(7ru%yW4d89X5A1dF0WDxp4Ljm8izd
z{!<*?Q%EkXHDo3qpNkoif)N;%JfhNc@tLw3Eli<_mI^<tjdi@1iK)q=!>)7kJO-`V
zxpal=4}J!zb8_k6VUSWV$|V|!2z*2-pgZW33^GKTQ*Jb9OwUpZD<ts<nI;UjcM4>Z
zG1Q!Zni6<2#d8EzS0RNW$u)`4eERZ5{_HcKDPZ_tfc5g+uec;b!^H<5WB#=VSeTvU
z(#2&y`{XiHWuKG#CK%=kuCh!uo9sQfA6L2<ZHPxhvMe35$svnVI-NG1ei!9BZ1i#-
z-H>F`;usJ*16*cm+lcK?ap_}QM9pLDx$_i9_U~sfictbyc=9^Er>@i6dYXyhV;no2
zAq5QjJ<4Up{LBpXN|P)T#BrA-Ns$(kc*y3ZYpgtThQoUfl4co6mNFV7xWW*Iicyv^
ztX8;i?lMW7gDPS>`rLd#{52!koB2OCpYNaNYb+&rULiuouxl*2IqkKUqOe=rCX52!
z^S<BX&O7g7|Gq=K=G9-uPk8@{B>B>0gx*{lcrfiSttOZeE-4P7)wr&})0#3KBV*ah
zQ#{9|)75OYW2SveP%jfrHCZ};jfprxc@h(+B%=|%PP=$`I{`uH77;C2FqT}W<XK9d
z4N*2@@07<@N3g!tXX)}P(+d;CIw!CL%7Nte2Tr2eISX_+@S0objQTWmn}vH1v;X$v
zm@MO0f8}?$^sFojQ(T5SYeXfDwT5z3A#oI`&RJO<va)8`)6ndht<mfD=&dg?`SA+P
zSHB6ReB4r<dtUVhE}T8hhyU=G$bRqFdFQwN6W;ZW-;7X_&p!MaR+g67SzBWw^w`Kp
z7<=<g^LM*+nxK6T*LCr|fUp$OZMFG}5Bw3yppWn%ED40vxV}%hI?KU5Cwb=i9V+D+
zN?rwt&v4Ww&kFLp9e25QVZx63OC4#(Vd75VB2^M$1=?8p-7ekskZ{JwbzN*AiIa?(
z2@jp+T)DiCe<<YY#q;djcbHShPmm=8Lf@lYt5I)G7dVh4X8Fc4D3AAk=!4vP`&}H|
zzn5AiKnPgdTIcf8HP$zK+}M7G>cU5P%h$XMCDm?c;HCRhME#lF(d*4Lm7=IKV~*d8
z9gf}KCk3TFb4(mP%<|^5C|hKu3sa2rGAq!^qVtTW9y!eun@v`?6|39pSo@`ref-sH
z6p#3iKk*aH&(3k%sZ)gs7DKf@&EnqUL~cx@K8+DM<!XiDa7Y+N2y0neUB__)v@L?(
zwSleH27BiZktP}aUXS+H2G!;yLTD~uyTVh?JjJye*XZ{8wAw?GVMgW}q+cP=HQ0Rj
z`3N9wQBWJJF+x*hmlmlYMoW+ysXV|I*~6mXB3e`IQJy8Bb5@^uj%y!#5Jzirsc<nU
zX&Goiii=kYiKd%$`#pNC4#Uw1DSgTl4XU*Xgp_1hvcZt;<#oDQPG6^#l}FWa@rA(?
zhLVKPcacitOihtgs{HKF{UpErd%v-368gJdOV_XS!|(nve)!+~-%K<bK*4PjLJ~xE
z_8mCN<?|hsa;VlT+&MMHrOQ{@YPHz2cQ2iGn^LKS0D9dn@o2<gIH2U05GXd+*GZDk
z@W89Tl5*s7ZRs3R2;wZ^;+cy)@zfJsxNwQMmk^axdc&-6#an~L?z-uPx!Lg+Xpv#9
zM9Q(H$4azJi+C<T+rs!!d}a#(I@*v72V8vUGb9_ExYofH5IBNVD^gn|zt|{31r_q#
zFo;KlwHl4tIqKCWS(;$1CN~+a)eQ#S0TtJwA|X;ao~sBY<T|IFTQaM8A{+9tjcvY=
zD%qDA6DG)vMf678xOAO2+<%mEIb!tKbIeVI%+A)4Qekyo+`^?GK_eW2Qz>BwVDf_a
ztX+tjHNr-PJj)5E9Nv3riMj)69z&KReGh9b3ymRV=NYbTU1aCapI~T8Sm`2-U@mJB
z+H1_!HkmyXVX+L70YMPpJ58F6Npzkw7;V$*_sMbxlV|K~c6sip7kI^?1Jr7h1#c;f
zNqb#nmQV@=8+pbWnvHAgWR8mnBH~_~qCxuq`|y=QP^%Qqoh(gBvW(pP&rKCd396Mc
z<#HL<aS>LKWhqILGD>37EZsdsdcB0rjctU5_y75O`A6UWPdIYu5R+4rEZ<mpIotXL
zSwn01^SI5sN)411pmTiJWuyeIHPjuKtyI&>Q$*@>=G;2v+ndxIlX#T|zPG{E7naz!
zZ<d|)9i+4<EYfu;)oOU<5=Ka5uEAKeNw9g2&U4bZhm2#UB7sy5p1rclEv`p%!egR5
z;I0GvxP0y!gDaZjiyo)XUE|298Ok11W+uq;oR56ugFO9kn?N7NsW{ZakoD~iCj1y-
ziy}LUrdZwRarRu7lA}3zutwndYz{Z5RU%}2mEiQF<QVdU`|wK>yyflhVq<lQM;`h#
zANs(b^7Zfh2fXte-buY)<x`*f7}u^|L*3Y>BrHQK3P5*3quuS?eW}YlM+zi`kk@xT
zl;d&n>>1XVFA=IjpB@AuLMYZ(UO)=T<#dE0#fv78jv|f|`n?v`=0%vjwS|k&7=$by
zw7W%|6@Wo0RdipKOOaSmID=hRQLY3i&%v<*&lM<<QH?BH8+~ppZ_w-ZiJ}rm4jyJ|
zVgk>f#}5Kr-$yD3Z8XX)F*P&8`O8=6^!l7WeTJ=-8{BsLE%;%;wd>ayjADk`VDb@<
zKlEX4zx7TQ{@?cgJWTTJtnWlW=bYabdqnP;wXfY>-PLNfq?Qm8LVy@#knqSh_Vvsi
zGmB@&cgFU;7h8C+Kmr8VAixsB#_|SSF2=@YlZ7O-H)^%I>3yxPu05-A&y2`ef6H0s
zj}w_y4QOesUL@uBdGdL>v#PS9ekbCb_q^|SIdZ}m2sfl(_;Ks?|MsE@|Nq|J#2Y9B
zyGaZ<hZ7JwiLF$aJ$(yTA3cL;r&!@_ZDJhOW^CJ#C0l2;QzpLDM@0>~+4770PWZEE
zcRKvYkH3fazwdn<K71IQBM4#+9y!hC^25aClD85{N0cj7j4l`rhFGK7+T3Jfa>ie*
z$2K;lN);DJn5@tG<}$0@EgpU3QU2ti2N|U!s+Af^kPxa7CidByL0qONirs_}=PXh%
zw&B=OEIYJyW1lgF5O(Z&X%W(=Gf1U8chvZ(Kw&NY_6ARX@{{QHHo_?q0i{4t7)5Tq
zqM1e=9hE^F^3-$6YqK*<OiUB1h|zF>%QAY)E37TAl37b5h^Z?{BnrX+>Y*Z00S-;(
zG({Bh?hm|=&pz_di#}=jO1S@l2l&zV{v_}H(H~=aYHHl2wOFB;Ie46lmmi^&Bvh+4
z!dP+F>%WDK^;H`68r4dLcDqBhRwGF&4EiH1(C%(A(U>O93(}QqY^^Rcy|AAo3AlXa
z3Saz_`?+}e0)xSb-Y{pB7K}uJkqLR>_Y{8?+TA>5pMJuaUSYg1rt{-sE5Rs>RN5E*
zLiuzQ<1YZlSvD`6W%Ud96XZUqpc*)wQ=~?q$5DP0vy%`A+U+fpQi<u=1?rU+)@YxM
zB`oNi?aNCf7V64T4lJd>A#h}crl$i6=lERS=fQmRH7)4ic@}kgH!ei6wAvxhHA!T+
zWnUR%dn{hgnVO$ta%L8741rSUv9YGJjv$J?uakp8ugl6f3Bg)JnikwL70}%_q*=jW
zxXr?3gTW9A>!{WXZmD%ivo`HPN?L%lP@QyCsu9xF(Y8Rlf?BPP?ez#NHCEbvmX|kK
zZx;;v7D^HOM<cSqIt$YkuB~6Cv%Jn+mXc;e?5InYWrTrak%&u+EA&$h^%l7`^sM$=
zF8+K{yAXn@$tgbcEB_LJ&wl2!eEbt1=j{1&U-tGTgrHoiFwvT1YI>SdQX&um!{Lzr
zpwHUIDk2r6sqcJ97qZ<RQZB`O>eC<NZEt-$Q*-m|-@orADe3^$t>$xCkG~W*=^I5$
zX=-7BaRwnAIt<v-n!b@NE#+*UnB!=90Y)j+O2p!DfHlr{<b(hewW(?1dff-B8|Q<%
zt-)!B9;Fn<APFeNRl=|!2y*0!1)jdV&a3JNXht0#d*U3+Gn!F5;Oc6?$<qr|n<Zv$
zdli*hola+yz|IodHA0~|u+U_%G9>JrVPYb~>VPDQ7z_>Pp6#J?%gI9#^HWXoT%%<{
ztKK3CLfm$n^r3rMF4dTS+y9eVbArG5Km9$7hR=QOZW<F)yy?x~$y?s~mk0!Wbo3Ff
zU0bIv6=fXT*8dwF5Q=f!(2HEw1xn!Qk%lNzY;SFH{@F7OM+3sd5-UL*N0e)`l%&t0
z%tjr;sDz5DL{UglWMtV0C4I7jlNPMl&8?KMtE5DLQWB;7L61=QWLJj}22*5AOtdIf
zN~FUcgTavDsGw9XQ>qHWph~yX<^C_;LpEIJt~<YlLx*nSz~PgGQ3>M&Q5-To7m;Qu
zue@`D*T3=23<oK;$jS09ORLv-=Ik@z9Q9g4QKUp-$kR{!2@3}gBM5d6+ug368~am^
z1rs~iF>k#QI~_Zv0xS*}O6rFXP(N~jjrLiLPoHPoNjj#?5)In2L7)<%AjF7_VYcxi
zto~QVmBmGV`e%QZpZbZPpw(<5g&--<U`36MjdiA`TIeEUed!v`XzKOaSnRMu6vb5Q
zReIenx)7w<fZ3T@(yri{^JjVD*~_f2uaKo9!blPXmQn5%@{L-=#B_^FQpVgcs&dX@
zoI^;rE9BVeo?7QHE=Nk`(@JniWkLGSmcMXl3>IDFEZ+A8IuAWgU?h=*YUt25U}Sv2
zyB3z2{d2@giM37#jb&<Xj@IN1l~RSlpoha?Y|iTW%cT7gjWDDl9c8JBLP<#qB!+Gd
z+8Ned#_!yBAIo0@KGh3_`yY6Kcm3o~@KZnbZf0j@e5#1E)arBW+kX=g+cavESZS!$
zYvg&3QUW`^c#lRI#!94&2vp8yf0g<9LkMdrhMLW-RZ=uhJ^eIy-*XRJ?Kb6diE><G
z7^J9*BC3RxN+tg@Z@f^AJ8wc<UwiO&EYAA(gu<uM2rNkDeJ7m&o8yclYj5)GAAOAN
zr4=F+wZQw@c5=TDS*Ora6U|N0&Qk&vGBdkCwc0@IoY80iShTgI!<2M=o0b(c1Fss+
zHDpGit-uM(r?YJyEJiQL&i?$0IshT*TS>pG*=P@#Z6xeFaENR)M9Ki2<>a|WMBZH*
z=f-4cg-jBp5EOZi={iEG$n%V0f532%adQbSbYnKVhB(+@VzNc4)IyMXgHQ`mD41?0
zI3&Vpe|t6(!kj1w7<M(+RyJAR8qpg$`U8zE1m#FFI~j23&<y3ku-Z#mytqlNBsskA
z5P4CckQiY}3qdX=D>|UJxW!O_w3cExAjjXf#g|x(FbH_dTfUpOy!EX_QN&x{_I-TM
zr@n{3_5(jauiyXD3!wt0r>0q0*vH)L9OZJEKzha7($dmym(CcA)*2xsqfx<jw?`<i
zao_!4;QQYGS2%R=AoqRo#T853h$oAb1v%iAVU>YVG-63b`Y}mODGDp;>4L;cwmJrK
zK`|<b!hmWl(M3*>#5f6)6ElR>3f4J<Rj9ywn;F0vgA8NBz@UqQqA0L3pj<0)pb?X1
z85^rx9Gtb>K6sYpmD?#D(kLNt;gClje+m~L<=D}H_VzNJjSfl&%pa`Lsw8YbeunAn
zX`1CKniM4h&YrtUZ&+|(+HvyGJSq&?YOj)n5v59<B#yDJAkPQ1AG)8)9p6Uv$Z_=2
zkbU#}_`(12Z*$+hpXc_|w=y-m!0W&5jjXR;quU*D^_eTg!Z8y`432?y^tJP811~ZJ
z_)ej4DC9W0Zk@c)m_QIz1VJ3~#y9<C7SBJ$MteZkKS!-G%jEO|3P?vigmB2QS*Z}h
z$L@>|+Uw&?{0-Y!gpmI59Y^m82}(#>%{phE9FS%yf`Uf9L8TOMZFvP1I9e@5qh4pC
zQKvWD=AQdL#p91X$Q`e|ivx#_Q>jldIXzFc+~mOi!*n_w27^90Lt#?Vbi^}fpJ9D{
zokp#Oa9xTVI_+&X)~;d-jg!8E`e%wY>$w%aS#U!-2J(iE%b&6$7ZPCwQMt<O%_lhj
z<XKGWXgkf8sW2)Z$EgysNO4gNN@C22tax$!>Ao8L$%7B_{tteTANqTLpK@G60JYXl
zJoLqnaQNVmRwF@YDI1%cj7CGW))=c<Ut6V8t5K~tm}pHi9CqpU`)qepE?hj%>gqbx
zT8RVu=ZO=A);XPam)@XHnrY_u9U&<v7-x`9j60`0%43UjJ7fLp<MQ2tS}4XLb)E+)
zeS-m0K<?N)^C(aK!G{_3hE!xgBeXcH7!(REEY?YEB&i&nXDb_^3(d^TJgv!D!Z2ho
z=#gb97K<75**brf3WA!llmbiH>j`nDpsgi^bzHU?f0z#Fd=-<Nmm2p!@Bsh#Cw`ot
z`SG7%a(c?2?P18Vldod+%Kb#8ggA-F(gCG%iBYeIE^?$0Y_6?RZ%qS^EFa;lB+Z8G
zTi8$9S1hev<KiQa@YFMBNJj&L5TZzc(^%^%RgGGeB#Qj#+=)N6zWBXk%$K`<>9Sbs
zmFP<P(J|w_M>vPHj%+yO;zOTj@rw@;7)LcI2$iB+L}=#+3570jbwSjqB86gpVLx%D
zMwX>$odXsOWLZHzOwrqIDuF=@ht`%HNns2|T0WO`dFTZy=-_!4b;kPmu7Nc~nd0j5
z63x;aVHgnG1OiDCROoDMB9)|6sSuY-gh`AD{5V!PP}+Knf!P`IbV#LIqTlN?Z1<U}
z2u4FebzhZ1XPdCnq|~gCxH9QzM4o1T1cyTf3Zx=6Ay=2%T)MJGr<+m~ekJ12F<lYN
zPghu&EtAwD^k|$BxY@x5mQ%01iR%0mYr;_v%Ji*9=Z-9?u|DY18w^n@q!a}VhZZ;H
z(m4XSp>ZmaAT7;mm3O@DFOwvZKggsa8}$)istUv(2<14YQLQsKJ;%cQ0`+<wYb=}N
zO>s2p)9ZB+BKJ2zQ-CyVZ}*8q`20Pe<Ga55d)T+I@Eoqviwz5WChIdP6(>}JLy&|a
zBV7<lM<Y_0T+^|Zt1AO~y-h*}$XHM=#V7$zDl(&~C-In;=sfquIjj^oDSUV~5qX|s
zoInJj_f~Tf<qQ*5!L_ATibjd~#v-rEW_Wb%1pDurW^r+ojU~hCvsXyUS9ry1Cpi1)
z7AJ4o&&iu6=sj|dr2TP@@6SLvqCj%?N}Fq|j>)Rv)Zt0uYMFl4qsT^7D-Ghff>Vmz
z6l@OrthfrFe(#U)J^%d=@`XSAFrWX-r+Dr6y@P`X57XY-qF$;Il_FmIt>4P(+A`&G
znFk+#hK7JxI6A5Ra*WLv6j~dsb>q=!fHjux_Lf(zB2mhbl&aiz#~ayLU*|i&<*)LI
zKlnMC6NhQGrqG!mjfX*mMqr&+!206HVa5beDSZ)V%~+hF3xsqi<^OJn&ut#|M)dkA
zQ&Tkris3LrYeN(Tl#-Bxhh|V=g=)1$IS$ANT?Xwf&YXRmD~nesS4zyz?qgx@073|A
z%_h}Kgd-pfEuHO@r_VgaaMVXB%gkJZN-bbC%9xs(@O(hb4YvCJ-@%^}H0^Th04V~`
zB^(>AU;j0UvjPlQrDz=9PuPlBxw^$=Tf#;S>ZK;a7__zoN)W1;?cpZIzPMc3uY`|&
z{Nv0_PV<9*^S?s{k|<7i<z4>`=g&RDxeI5RYDUDRgsn~ooe#-!&Fb2KYJG??8HW!X
zrCP7E-PtBcDooGLFf-j`VgDh*FkpH48tu+H<x0%NM3tHOlf3yYZ$pRxVaFZP8;(Yw
zA#8S+N$kY5VRpNzmca9JodIJI)?tmIy?%wqKK@@AUS6Xl9krmqih%9HcWRu2To(j0
zRYa{qksJ2S?WfwB0AMiap>>V`vO+W1>SES9Od_c$Z>^Ra&(G51SjorqfKO*VMlZ8*
z>zBj*4?Msxe(;z0kstnH;y4B%E=>`Y=U863%E5#CaTvB%*XVXSR4OH+C?bv{)>f|4
znw%sEW16iwvaFy#=phI>d+u2-UA}^lf_)3~G+T9?D;V?#jM4$y-He&}LrhFf!S33_
zP64-LFSH}%fE{*(@6P$-*C!N>Q*3vq2AivwS-k%NL@%WrKpaR$MM$m#j4&AIaKaGH
zPE($nV{(2UK~f^mGoO4jPSNr^wnh%OiHjk^S)cMG6~+pTfX}2o9(e%^I(VK%-N-qH
z#!xs*ITZA>0gJ2GIC<m*S*~|Cy;{9SP%We45Tyd&8kG|1ynj<s<P1hB*)aE}vUmeN
zB^3LC-XN#eXi%@0$p!;TP45<8Zq}(=fwr2hZpz}d9+#I_SzYUrW`dFu)M7znLei>-
zOxDX(>s6vSLL1HMwH3PC1CI8G)S3|o4>qYaCvepUk@QIzflJs*1MGBzmGuq-?Qo$Y
z&r+-qOiau%)IIvW!E-L)4v3`U@S#Jz=5=@NZuG_)9((LDimxDEs8lL68x0zb2F*s3
zX0wSghBQryq6mOEPGB&A9dfX7q$5kOn{)pB1-7?4OifNdpCx@kkmLAR)?+HN)QO3L
zfFzWRbWTMD*eIZrYc8#>apmeFCyq6+aZI^hN1K8qYBC%e3SIahiSf_1)?vrwXKf7G
zC`B7h;J{gfv6^&LU~-MqIrUONcYBk`nV6H4j}vw?wl2P&SDd^BaZAARa)<3K<Lc!e
zcfR5zN;@|1{S%VqPjh5ugJvng8pqH@Jau+RHPYO4FrYQnBrH`(x7Gmb4K@RbQ<BY2
zk8|4{Hb((Ze(_<>f9^B%dpW&KbMJ5bI(NMGwG3xl+_$ztvtH-mk>lKU>}KTBBCXfl
z%#-KVxUjPNf~u>(63mWWfVDU@7UxZbh4q2o2$X<}7anDhZ&Rtq7&LSHZlYAFpiS-z
z!g06C*aD06?_29|PL2;;Z`ixD5nF38<AM_KmM2P)X-AqBNEb3OHOZN?7fDB!Ah1*`
z4UQbTnObXtEGq~UL^E^r8y%eSzFNak%9Bq$!r8M=(5Tmu0;XpdIC=7RiX!9Ur3<WW
ztfG`;dZtCG6j7~(^!qujiAi5LI5Sr0URSpYcuqlgBX(50o`Y@19X&Pv1#hsdfxJ{=
zqddtuv&|r=(F&?43Axc&8&WM-X-*_8cb0fLh-hwvwU*!fkH1B;S?9m~zVAf{h>|il
z-}+j%<`3}96ZbOgZgTk00nVR0!}>;_d+&V~6-f>pnxxT~<H+G7bo(h=?G<vJ5k?Wl
zS%%$hy1i}EG$oD_PTX=Qci#0q)SGjjq9H7d$?n%Lu?~duMKjmCrd}|yBVZYK`i1Z0
zSPR4LO_rZ}oX+FtNCZ>@fsz6JBE(pUk$ysAm7#ucp0L@ZHMfsSt&TRDqVN}ODFh}r
z<ay3;V-q#%69pFI1bJb}okClSlkkPC%l-M_YtE;7G4O{U|2T1y@OS^t-=P#I2q&mD
zk8|efXBZ4FFgw%0>4L4TEqeWcKq*G)fX$6fl5!30EQgOC1MBE?wn)kq;%c1}#}Ct*
zm}Gu_AFFFitgSDTltSwDGV}Y7^VaYC0hB7^%-A~U#up{PG48zD8!lgh-PD?$0?+_h
zhcJ%b<~p4#ml$1Lp(G)a4r?P)?Jt!UpF^Dr%jCfYYSYsMK{UQRh}{e|3#6tX2}4}r
zATMw@!XPBqf`Nw5<l8)@hcEc2_<0p|nRATB^J^7K3N+nOkF8FJNGKu|k~C@vr4WH6
zQ2xj3G-uH3kq$>Vd?sE<fEii(+kNlTBMPJv#AOhHAS@*a6=DYij2dF3zuB($QyxFJ
z#nmfYgn3FOQp`03)tY0XqG(iWl*(0tFhVGW(T3h&z*c(|ky>VFo0Rp4{KyHa2al2<
z=*AUvnxW$o86oFJmdR?9jSE*vk!UNh!x1tLdGnjz!joqn<eBql$n)#H6bA%Ca{KAi
z%*@Ststu0OXv7mwJ~574`SYQqB8noSIHpu8Q7)G$l}ZeUL(1hcNs<sn(XO?^t_6Y>
zI4c<RHEU~YwA))W8qJrqQ@4XHYxqQw^MeNuz$hiOLkeS&V(bqVglwmV2cN#e(UW(2
zd4IJ`Z?jFg+8}T+8VsqnnmZz5f+)l~=x_IMSwS!s6^JnlP3&;9k^ogWBxUjf@;pOE
zj^m97=yf-_dhbz&^#xk>1`{%3b$OHR%a;i@&ogZw<-lP_y;Q=Ql(1gs{)d;)Hsz+n
zlKJ^rg2b!2rsof^a_u7BZqEmRmlLuT%bCTT%55cH|Bk=R6MuX^s<+O8a>(mWEU<a$
zDQ35BX6oisOwH`WSj!!+c^y3S2zQ^oz=gH-m$abs*coYPe@Fsh5TcYoNl?<#*}Tff
ze*b+01pLmWM^L32bNfyZMG;E+%*CCI*Dwf?*4tt@>wMCK@CBVQ_Bn>O;EM&VwHL|+
ziur>FnVy-TR;%Jn%%JPYGR=X5by`z(hDFA0hY#blX1l!!4$7r6=_myY-S#FH$tcs9
zb`Rk+>uW1ygCW(aNuGZ8EGi6`nrL$Hz+rm*4&83X%*-qY4<ExiGalJ^2lySUV24+$
z?-Xu3Car?&pFiuodfJYK70x;go|xTk5BU5SpWxxe9z9|L5h7&-(&Cf_Ib#3e8vor7
zeg{AKPd)|fFU5P|%OlHk{_U^-I{W75xa(DSj(L)b>FGn1UQ^}iC+_3o;uUfuSzg^@
zkmWQhHEuce3Z~}vv$(Xv%E}Uh{yIjdB&7=Fat*9uI2<5^<nW0*x%u=PiA$5?=W4y6
z$lf4Y+3CKFzqd2OmvV>WC~<(Z){pcYNGC}9JvNsX>7KiQ><>t!B@iK_BB0O#PVTI$
zBd-KZo;pEuc7dc+Mr%DTDtG=)L4hW+Ic{^AG6lv63<4u1#=sE6-RU+@6zR*XME~kx
zt>t%r|MxN4@PGa<{~M)p0zz=dYrc#1m5V(1*rSxgoVb)Am0*2!1t}$q*EV_bnN^&~
zIB_)N_S1JDf|$*%HIxz<r63IG_Pg|YJqCk5VGwZa<SV%O^qZ)+=6z>O_$JYfxUnNX
z*1R&-DPg^b!tN#2Pi7rOn$hcZu<dPJyGP_C)(M72j$`30g%hOCLb=5Ji4)$y+&a$}
zwASwp0)~SWCnVKool#nVv4mkjnme{Md^qcH+5Fd;3V*eD-kmywBa*_$ff>s%OIh7o
zWqx{r(1I=eXwdXE-OWCo{)jMAw5D2ACz>edXs@iZc=-y|T8qhr8HDwGI2#CL6ahYG
zbCeB8^8tf)L4VZe=_>=ytXj~PLz9xj$7)P8ssw=nkx{Qr6NMF|3P`gN+wC<kEHH*@
zwMMBKP_EVqx3`)3mTzbO>%NtZE9ZIS^PeFOAgMRVgkn)aMfmK{%t4@Y`c5N8mJfgI
zcgTv7Kk#mZut*Vd#~pVPL_X5TImhbi3eTKBgZ>hcVnPVQC?JYLf+#=)0wm<xcRQlU
zD}WtdiXBH00!XveXJtCy!P@S1SzBEpR*EPH$nyLp-!z^XWjsBzkH7WZZ>4kQDXu>C
z1lvnjiTW88snK!7wT(WjqZDCcmWOqk^#<koG(i}WZ><nG2XW-LY-`cZ5|(1LG34n8
zr*m>sc*UVL-nG6c&_<JH1?jNh+Uh1@eTG9vW|=52Q5!AO9mEW`0<zRGNlv6j)Ms-_
zvGN~iZK$^DoL?HTw7AWYdBxmpgh(W5J|ap=gh|5m!V$JtS19@eL;yXjIM>m<^%bw=
z^>26+r(geF{MN7h8>TleQ`Q|;wWW6XNrn?u^2Q|PN`=k5;G=6PpFXp+n^y2r!WloB
z@_H?AK~ybQ#*vPWPzi*|&^?WeV5qy?a`RuOHL)KRhF%1LH_TQ_B1DL=(z^#(ffUy_
zWNUZ4-TaN(S&MZRW1XLx3CY1jhp5*|SW<+u96nINSV^N*=D@yv=%Qfp%4H7h+mDnX
zgZ?&^N|{7Pj0Sy-$jMELQUR&Ss3cXK7A&n_CEVU%V||S{3^;w;ZB(iidcAFou@uG<
zM9G*Vx8tD)&WY=Eie3JeLF>Z*9;bGTTIa^BNGCAHVuYh8au%<w@aa$8#{&;O&Guk~
z#UO=5iWrBFWfC%CeXUQL4cYF!JVX+I9=5w(e*Tw!iJ$+e_j35qA;5QCE7b;fzUI3a
zbT@hC>4(U4g~KQBpwVnHJv~cOEfb7-L~+b$C~-=mR7hTc$#b-pG-nQQ)9E)5m6~Hq
z9J~8$cllG+f*2dox}5^czjGY|yTj+S4!qO~YaO{R=yuys<P0t^5#$afBZfv$SWl-h
z&XGIc&}c3!Fn##27rWT&ufqwixJ>hc!daSO!nJlEs|}G15F((^f=y%iMAqSodHD(=
zzFJso`Gb%A7o0WxAOGwBN)kr|LCEaFahg-JEM0k)XP&&5R<lICImMZ0&T{_ZI@gxA
zIdXiOAgM9T44a)U>znJ023shlu$u}w9WWS;D2jsS<bGcL`nOW9EsVS5;||^R4<l}z
zoNb-GVg2ncJwlFcYryPUD`~Ci^?L{@=&r3}hXulgq#De4KwAKPt4Xb9X0l0bdJ3x~
z#%YB0k$?yo4pWN4urR-mM!CYqprB5|Fwf~Y!H2UpmtVv+e_lmhZVdsFvJgZP;wVH&
zO=mcuKH8=j^;x{U#KrXyo$inW6J>6@{TR&&A0gCRUSnfr6(uFdj~}HrJqbyODKw_=
zkuP?vj*S9EHV_Pl1J0dW=IKqrW?#~X3`Y)I4(zKi(U_-Jsn8#66GnBc6{LAawHlM>
zDZ{}wm2!h5sZok!Lg`RR1=HK6k|=^G!G%?>o<B#cnsDg&ZHO>}a+zU&fGiB!TBL9|
z;Wv-r=u4B8ErNx8`!Gfm1OZYw&YXFQYu8r3Y)q__JK@6ytu>=`M5$cbo%x4hxLcs<
zLZh{RMmzUYV;%ir$}`WNLy4FZ$By&Nxo5r8`b!Aw9RJ~-ySe>4{u{pcfBfG$xUx+9
z+|yin{BgD~UqoKHLfRej;L}|u4;}}dB4a_J9YMK<>9-kmdk6`-Kk@~DlEhI=SgjCN
zVv6Ai`UA8x2xl=xL1rw2bV!zGUa>6-u3cN_a!2#V+vbSlfXSdwt7;L}pez^(#`*)=
zqG?PuX%FF%htDxpbsRf90aBxa5GMlPH5_jzt@$~&w>D{nn#oF;JR7pSxIz$BIC1Bz
z`G5R>V*cRY{3;J!xxi?%!(d~B^7%)}re}EOT+UDZ!Us8b@zP7WRyV?o`E8OA#vSHH
zYl=Vx2$i7%fshV}C@6?a3miH2dQc(4TBHodb{@i$EQRv+5s^faJx3&Qod+clO5Wh-
z=DS)>YZev`Q>#p)+<<)x3)HJ6y8S*XkW5cZ5hn@5UYBRjJWZ`yB~Id9MQxg<6j~#L
zfNrnHVA$pOv0Is%-j5W9rKJ_hm4x}ZmghZbO|8*lZGDygus{YeJ6^)pdX=OQa@PZG
zXH2cN#u(q#5m=ntQS$D1^9k}iWvjQ%lTV!E?oU0&;?*|80<_36Wd0dQG44!D->nlN
z&z?KS#^%enr~_DDS>@e7{ZIJ$pMEcs6BE}DSQXHiUf^|;bG-UnvR&@e_I8`??kW(F
zBsFYf4k0b2q>98}oE8q>!K+^T?Ih*ab*qw{Zp_#sW_K5~*Gco&<J!ETiIfs+4N@2|
z0y|C*84QLvXDQlkddsVXHlT0;8SMI*jd1i#fd~Y1Hy=S&su&?qZfr(9?$~Cdf+8zu
zPEHW_d!!p3;vm3T$*^!-HHHtTTeRKR<G}ULhqaaufApioQN-W*U;Zb8AV4WarQG7^
z$vdgmTbzIPaaK2%+3uw{=~y^4&7H5gi<?fJCd&p4vy?n97^Nvml7Lf;MjC6tiW<jH
zy`FM?c5EA@$Dcj9JG}w#feDN+L+-j7UCMFc?JtVbjv01^X4D@b3PX^kq)W?4Ehwx&
zkN2&@LDv=ZO+i@-ruHw8G@AZO;51Gs9Fox}Bg-^XlhZV&CW&%Q*6lJ#Q`Q~)m$dhy
z7Ig4Di#qEZiBwdCAP@%a9QCL~7({G#N1R(3@W6#Gv4!uw>nNv>PZ35j>7dW*$}+JE
zIe7RmVG<KWAtDYbj3FHkF<FjtAf+HFmvBO$rNWh~WVI1nSJpT<6>xaALakY%TB}kj
zRY{A2a;=4w0%LN2a}Fe79AmAdQJ<icRQ;PkS_-8oh9g>WfHht<bLrgEtR)ez{Pw?u
z77`gJbhf&Lg&*nVI6@~FeYwia!Kldi$)9>Je|*p9Id$tT96Ee}kAL(J>9l*#u{`+G
zf6zILF^0ilK(E)M*=#Zzj{N49rVIuH@;u)y^6WUqY3E=pjAN8({@t(r{}>H&X6NUq
z)oN^RZ9SjmeL*lvQ-1D)KhLR~Z)4w~qs%PKQn~3ST%NIV@hn#!ew6Mg<;4Ddl$!}9
z6;PR(#$=ngq(aeY6X1L&7w>VVvka>(vK$lF8)ySA&%j!8V@ZuhYl}qSvVu_#Q{jjQ
z*EN^V=NvsWLlRX02}+O)i9kc(h>{AE<vK=L?t9=_Qk!%7aE+;{Su&jwB@KiKRh<P;
z98tHegA*(`g9LZiz@TC9!6jI5m*5th;4-*-a0nhC1a}MWE<rN5yUm^d-uLR=*HvBB
z)pfdiYO0Rxeb!zdyuv9wga#gUF^Ab>Xb4)A(tf@A#u#|+O9h=9Z+5x<m;G?E`CvFv
z%c+Tl&B&*f*Ez`1@^6GGFeK0y22Y-KHL^aIe9VIqszcnxCq{vCMMh=%4%EA3AaBbb
znxrZN6he<atg+sgexdZ&>sIj9`1^RN(6cGAvQb(;gxyK=Zw(Nl4b*RBWp8xBC1hnP
z>pe5`Ogx-Svp5oXEan)>4nM=EQS^ih+tv#h4^zN-$@-Ox(<=$kIB>~?$rN;^#BFSn
zORaid!A|*>Dw-5$rK$>UHe(q~T{@p770S_Dw;L0Y+t&*?b3J-SAHG=AZOY}r)%Kb1
z(lF<V{M#EQbNYHrm*ek~`nP*4^tz(1*9%6EQP;*e2WBz4S?SyLd9uf+x(4kmUqQV$
z(E_5pI<9iHDv6#WSA1rTS)zhxey`%=GFUQ}mStuG;oFE-OJ~z(j>%SZIRX?S7j{;6
zRJ7<)s;T5<uIPXB6LrX+EZ%nUFJiVH(Ld&%JxQ&T!o(ZlY=WD*78?At>dHmubwa72
zTO|cZ2u+$pu6eHIPi@Q7eLM@ozi<R4^ietm-p`czN|~%PGxbe=zY><rC1wWYg_Sw<
z49(o8OGtF8z+;j-@k_jaQ0s@vP1s8th8K;A)hRk(Lxg<9Dn*x1;v#lLV=3cQnA8zg
zUjjXgCa2)rq6B({@|0}C19T#l^A*93fr6s4m#lb^!ky2wA`9*!m>mrhr{U@mcQF=Y
z@y<&ZVLYR<y9dLne!s#YD~1sCX^qp@+i&yNv5eXXF7zyC*^D}Iy53jrm==2AEGvGp
zWTnaX#RiUQU*PJI`u<kt`*WLNqL=NM>^x)ji0@VO3<rXr%-MBgrqYHuovt3(q!@#P
z1UF}Pmn<L&c+}G?^?@*&U{{EXCj}89-#+$%W3|S$Z@(m{MXyY&q9C_u{gj*H;dCtV
zLHNZp-l$}3dzYeV8ELLJ$o1ENU&a(OQzW%)!HApj114SqvgJ0Ih^ZSa<25z(8^4gE
zh*cT&#9;~;GL7rdM3sn2!;3!bgM<k6?XY^jcmdl<A5p|hJRp;B8PO}y7UN$(Aw_SG
z24=*7l1u3xUnpM!;*DzKQ!owys|bWb*3+$k$&38e181C{y)+ix$6sQ!8@xNjUU<7N
zFY(V_mh-n>B%T72TtWOil6lxb?hDx;Pw%E%JTuFUihu$N%eK)ysQ3!}4OG?wy2>fi
z5Me|1ERCXQq9h85TlpWBaLVaE-13-ZF=1hJRMS*_<$GahH!0<B;li{pywhVD8F{pk
zw3ifG#pp@H$+a#925Z}EB_HG%iDE~{Kr-k@3B2^|O#7zc7=)%t->-tX!OSv5vIhww
zx2hLIBomk9UC(Uom)9(uT(z@DjhR_rrrCPrYO1g7(-&>+BfKBjXE|TjonFx#MLhV>
z-cgur_9E_SM9KQY#XjDsC(0Jo=fdbR_S+xgC<8m<UD;eQ{9G7QuXp$AO}d?K^PD?Z
zGm`_{w_(P#Z-;_|FPO1UV&3a*FNkae{gV2%s$<K4sOw<U#0ZFMWjEwwCgK(mQO@ay
z2gWn}CHmDG_WaYbCI%nVN|i9-uKaQpyhVw!s5}e$k9`PmgJq$SjI1>|!o`d%30YiO
zA#1NVJd?4}6&eM%j@5m+5s^@TNnH6}?tcCJZ@!aIs=W)jS~HJbg{27+F_8~qZ7Fd)
zcPc*;>|-1z)?UI$+#X<+=R?N7(~Y0TH+Nz8tAW|{EZ&L3U$L;w6{B3T86v^_Ol)&Z
z|JpN3r;H>IMbT7N=g5&nc*MVlO+I|;L}DE@|Cy_Zw6#+-SUcbq+F_@+_AyUY(`lEI
ziQ(Eg`y^Fh(x1Tf3zSYix4>L9tb`18?sk&4EDK<*5IjZxeg}ZV_mkh+%q)6j*&D7F
z8=qi&Il)1yZqb3RW#2Lzx$=%-Mu<*6qz1k?scT%sHFuqa8`le!l>aJiX#qx^l{}o<
z8ex%MuPE_b@830#!1Y_AeE?;*LOA<IP(42b00Tgeutt^?$4KluOIPvw(L)J##H`1x
z-@EOLlJ@rj`b_v~Rlp_`-d@rxl);*9D29`yt&B_{4WcBWs0;&LYSvHap}lYA@+{$X
zbU)_FlKC3?EdzUsCARdfvtd(b^%s_3ues>+jBhK^%jJ(AXoOMIX)L=nM0Tb>DVs7d
z9VNXaNB`DhRFqp<QnMFGFn>x^L)oI}`x|8H6*HyFI*wZMh%A|uK(1Ui%b1X)-63g;
z5iPp274@>ad8ysG<Md+4N<bi#JZs$2yEhq_WgqXCci<kxj!|`)r^Bj6uK<Zgv><eO
z-Da3*TFmqvIMSum$C3M#MW7ry)gQ(5S&sF)#*d_ayw7Zv^yVk|Oe?Lm(PzcgC9`Ke
zI?H|+^!B4BQ)P>rR_5lp;F1p$sCVma?Zd;fc;0D+4)sTNfqkk420y4nx1T!ieDyN&
zceV<Up0kZ#=F?H+Ft_ray1KbuuL1`p9$~NCE*0QBC^|#M<ol19R6$h@DW`*EmbT?e
zD!6IL2;n`0pIZB9ZVnW#P2_7kBsHpIF>iTj^c5)bvUSslSQTQ2CPYJ|)$EaXQCq$H
zOQsVRK8-ra{PFaP85tqIS%8r2>Bfxmzw`v5QNfL&(qs+k<VUJZzk@U?xx+q^Q{8Be
z@|m%lZhHrUSxU+9${6iLn;K<#_k{Fvx1z9>arKn|KU1Wdzh<(J<%jrE`6c}4q(-+9
zARC(;cfA}|m(@=srY@#MXID5tA5!TIb8R`05e@qoF9!9L<pH6HUn(A-OZekKo#p>>
zYsa&oCdLoC;Q&7vS}jIkF_yk_d5m+~d-WBU^>R*ad9Y>O<3%5WTU6S+Wsi^JXz~$I
zh<*+7=@@k8{=Rx(0?#y6aP#qv9IlY<>BnkllxR4Zu~#`d%uqq<7St5M!_<rzf7aYb
zbL(aqx?TMFkL!~U>`rce*S$|I(J=cGi6OFnFUsbu{a8~BgVy)GL5v`Kx_<Zpz(q*s
zx^8j3xr}LfhJA9LxcCe^kl+&o4#3U@VW=9+T=;Hw255>^vjL)q6d)EvyrM5bjCVUc
zJI*TcckASZyKe@cWpcXsBQH_{<x-F3szuUmjQKye75$<lx3nfz&;Su5u8toVtxF%8
ztogfP)pvexNb>;@fJcx2zJ#V?xEur17J%V>G<s>Xa8ZbuSCE|pKC0mdE<+PFtcR~W
zyY=jQ<h2ra)`@$-uIGY4lx~rYSI=H*)W2;DAYpp(6Dgr@3DL~uYOISHIU5ll$m|=v
zdh5cPkwD*-qZ`f1!Z$_hY@93ym$p8Co&Eb>ec`6b+F+`leAXI5GtR3Tn}ln(w6Y@m
z4zn1A0|m)+e;<p1UVvzeyWA@EaE-knx?SBRkC*JBKV>A-f9{ZgnIS4=yHk<m{w5~f
z=w~oap0&CTahg2X6)6l`zR4^@gp+gc3khCBqKPq|v?k+xoGVL@1+b_mfjR1}mb4W{
z?HA1N(?1H_u}N9w!Mom)X_owl9`IB@GrR{6h2zbthdHto&;6hQzwHg<$I0LQ>L0g<
zB<S|{?iEE2Q4U|0p3!vbh)xl2A1{5?U;OeMpRYN8Ya}Ay#G|6`=yh<D_Djmw`#j$z
zuYEwg=mokYVCKsQxZxJOehrTZj7y_^1iky7$CTq?;eGn9z$gL^OKjwf3=S$}-oJao
zirDDZRETPp{>nKb3PDuK5t41masIoF5=SMWLMT(r<%)Bh9_$teyBC2a&Nk=$J8|+S
z-*e41h1aWtZIs5Q42@_s)tn}ZP9hT_2`U5w9$F;`w&5;mHmx!r$P#a58|^qS*g4hs
ztxN0gl~=JetJpFem6NdB<gpUf+AoN(XPB@^KCqjrbL)7~yMAQ;h5I)_h|{#u%6pG&
zG#e#$c>0HyuGgoUv{K-Mp%Z-R_cpYA>cnZf(=)M50%lv-6qA!vh5t4zg+>A2ZM26q
zN^PyF_Rsbs<3O+IUA5!+ZmROk^HGRSQInOwTk_4gGNtb&%Nr0%ym1q|5}V5AnneMX
z+L^oU2L{eY8wd1WbbAg-0DeU`iS3|C>>EG+q<S)ue-)2?_bvH#<R>5q8!l)nN3=Hz
zK64$L=NGMhF4$Df%*?d*`7!dm7LcT|>}$qKC6Jc#2ZAcq$h>swSl>QZCIB3LP}H$?
za&VhfhLboxQ-cm;^(eB0Y&9xW-JVNl=dv_k1va`XEGZuLLb^{*zg4GPGL}WI`+d&m
zI*=Piugz<+ia<l9R{z$o`)%sGg%GDk@!b(7$d!+Jah2;{Kaj%|@>dE@A|4x4x=Ira
zd^-JY2zh71;$PDwfLg%*tx*?Qg%xtnhqqaoyi*)|^ot<?5xQvFOb5kUFp^mFLMH6Y
zb`!Jg7Ver1H;p7Tx^C!b>9MgVKlq-&-TU*?*V$(uTs#x!y~wB?Idry9p{amM(%5i9
z8Ofq$Gds_?VvRvvWqx?lwsGw2-DA#(F{STWOc?v*!(QM`EuO~Z;^qZT@W<E_zrlg^
zuP@cJuZ6|P;8eA5-OIrsk^nMG_6Z_+nxXvsBbg58h0hDe#na9mMU&WuB6WhgnFeR2
zJMH6<Ad}G2w$|Wjlmayi-AC`}j~dvtjkEH9GcvQ8Sm(Qi^WWuL(eJTVj{pyESm2Qp
ztD5NJd0+dHv0p+{&(=V|^<>W#+aaO6HA}*-_B6R()d&Ea2%CUMho$@zp=!R~zd2h;
zmP{HPI)egI3)21G2yHrfG_f!0Q#qL>!Xi?M-lrffmWMUXbJ<25-knIpKl=v}Uk~$C
z)iMpguOB$Wfv04t#5Z(I8P9QDz1u?pAxMPsC#e?dCDK1Q@_n|+pvkDSLrzw|I%Mx=
zsIGdto?vFT{VqwFK$Z5xrZw2PRB;jByj|EqCd-5lSen_)fz-hRy<}HWCd8k^q!Hn7
zWMNRDeBvJ}!lG$(2P}08yif$&MBm_3Y>wYNOoSy{YY*0>*Ac}7OKiuM8gmG{G80eY
z=i(|9ViOhFj*=uOrBf-}_%aP6E04S-Ba9rM<#t(YTcgG(SbrplI2Yw43c!bD$rKa9
zjNZZ&=^v5q+$N2k1B3IHt*eG9A;pBasee+A1oVZWBJXq`4dt3K7hOl@EV{L{XriTj
zRzF|Q1lv~vD-w^h$Cly<jz<=+UL>xI=E@lBCkoJjBj4rB|5Bqlb|kmKW}PkuBE<X1
z@5bRgNZx<DN~uXB<q>$2gl@1FOUlmHo1X_cuQKX@rDNa+->e}nzeL#e2%4=x7KNC5
zLU!EyonM+|LO#3be6zD!gMv^)(@=ML1~ty}98v2bT8t>dnN=i6?)H1L^fwM+GP>WW
zxA&(-%0H2dK5mr(ZUq%)gI!ycD)r#_sHuz$NpuBaGMc8+X&d-HqF&n1^^@}lU&`I%
z7Rw*8_+IU;lIC_899ZkOANIhpArUIPB7qs)EZ>sR?eR3c?Mb}F7|e8Tg#NvIK%Agr
z7a``*KUzQi1`~UInzw^Mq*~V@MMM9VZ4_(R;}U4Q-zM;`orq#L+@$9ooa=<ZomnNG
z_5vp?DhI}yRB@3JOaa@NjfVc$Ld!iTp35mox}T{bRHlk7{)Jd|nv!RPPa<b*e8h_i
zcn}V2Ke;JAASdIq@Sa{ce(*>6$>yf4lgdSSLI;s%IJZEjSP1!sTGE6$$JEg-Yw#HH
zo2HFDAdVNp!>_NYq%@&fsf;g!H`DW?DmEe!xKG3BpG@Q~GW-+r?sfDXj|g9%m;Y0?
zC($rs(`O6Jlv*T+q-<yRFZrqjbA(=@PGxzByLrRkBlAyx9lzT^!oUl`gweU>+%)Yg
z-qtVGP4H3SyJ+=WmO9%%&JIYX6f9G-z(L>-$skYEs3Lq>EvPkfF&rm1sLwi2)+C!O
z>jwA3iNYX!C-KS+J0IE@*%o@dF6qBwJ=|X6VMWd)*u`0qnyMIB30;=w)=9;39d_%1
z*SvCbWT{8VciXfn=3%>dt4vcR^;?$4Q_*BOdZ(7UhpLBSAm$M0qOIuuYL%kwbPl^B
zf6q&?Tp~k!>W9SS<SRg~liu<&AyPiam?`0%Wi7$eqtyx$tHJ#=no$tOzc-_u?qAy$
zbN0N^yn{+N0UJrauB~x7nGq=LLu<LgcKUbC(QGm;tq|6dX=^u*aUdJdzi5FohAAK5
z8sv%SK-#ABv|8qwrbTj%inP!y?Cu-l(m&&79McRs&t5JQf3Xeb{#TmqSem4D@GhKd
z>F71a#NRxZ%Kt3>qTrohq*aHk@ID<fd_42+=-IwSRRE}uh69ibH6%-cl1*V8s8f@G
zI46)%6tZbkZ@GlY&YI;1PGbgk26px_b<QCG`$Gm_Q&YXdF8dXBpArW(7yinLcy$G;
zC4-tmq`aJsX$2z-kS3<7?>S%M+L5)v4Uk<QaKhW+nQaH8VOdQ4IcUG}it~DvYM-Om
zHC9o9%A26HkS#6Qy=OYXCDAe=W&h$NudSR%QdLvi(AqiC@lG@{B?;Uz*7NcN^nMnq
z?YvTP1bQchpxq{?(4DzdfI?RfgjNl+5k*ezfKZ}nk@sY3Mj1gMWEwX#$I|(n?E<)@
z`Fj_HlRYDdy4(aqDu)@uvZQNV*47bQAMV&3o09aG+tR;7u0_2sDg(sESbFDX0#jC+
zU|0Lmgl_RQ#6-%VnW2Y-P$cZo4Fw;^GkDJ{lt_?K*#sSAOtFytcJt+SRd3V4(#j`G
z;+`vC8sEE#5`h;N9g6u~vK7_1>@270#v|bL<qWzG+o!4z-0T&S>_0Q!=$c%c>bWu`
zVd=oeneY-HSW5o#Oh05u<<w?Ah_XH*%zOZB5|MC{<2iCB49mxrG2MMg1Qp8B`REWQ
zhA}b}nrmP{%Te$KNg;f5MUf*?Pc9{)?2`&t6g0>;Q7ro7q?;8LS}sE3Ps^%Tr*|({
zmE>GbC6}0RC&zmx2^c25%h-|8l^x^)M9SjlrTjL(?=YuF4+(tY*<39kZ`-jSP*!ez
zjAlES%<=~M;^>tE4R|KQWp<i_l`Tq){#@rgjf?tCi9=fB6x48?V`{-~Nu(jPIbDeo
zO`kMJY{fl@7f!RMuT)`BGM!VPp5Oa+NKX6}FW{CWAZ{;pFMZnuMk-ocSQ9X+>`v)R
zY|D|AQ>`LtK&%y7L}9Iz!r^U)ZYeE%T>sNEc5qtw<OJs%eo9Ej*xRfKYa1n1nrs2y
zBQawO{KE2}Ivb6JRhE3eeMB!~U9+P6B08CU3@d|(5Xs(=<mNipJ<)V7!`*+Mv!xY4
zuKSGIlpRWRoa(z$&T->LgH}%ci51@Z$ydW~6}|tW7gc6c*e#4G_v{$on;Tsm>frx*
zLql-`Mj=AR3TM$%-#T$H*0l<Si+d4bbkg8tIt;zEqCvU}er2!C-^TbL4Njf?n<txf
zOCO&!ug<H8hXShC0+MSFn9TWZCeuUMyH~`shpr7}0MouG1C|tco@`w`{iaGasP9I@
zHIb$;;j*9pB~_dir-GiNtHM4cx}9NSW*M`J%Ls03lQyV@Y4FXH7%$Xa@B_EEO3`WG
zzgVOL3gPBm{RDp2lpmjv1-@wfE0z6|NXPPdm2T_;3F1ReUrk0bA=(QSf3Ss5*$M!m
zNVB;2T_|kG%^eS`Eq78~WAzne1u4I~UC#am-K_^^mh=ctR<G+T!ab}<_^iKPj*$kO
z(NWQ0NWp7pem|5FthsN~H?O;rV_r2VHD<MWif=T2Fd1_D0~8V8N5Kc4?xF|oqw$m+
z(D^-mh6%?@+yb?U^LxxHQtatV@Z*A+wU}9^2dPtw1%Q3?l{ze+2(O;9{f?SD63v-Z
zG@X<ZQp|#>Wh-zkBFUgKU>csn?q!oTqPn($)%-YAH*+^L^W3x3u!1u%x(qIJaQD}^
zf}ooloEORX=73H6rKNqf8dixUcl-*(VCo(K%nQ3Ay<q43s=Eb~<e}v*FzD`ezpee-
z<307Si6L(ynQuH+qBWfx^$i-dYuHB4r#|!<^~7#{rZ^_CHwMyUr<3PkGnnsB_A7H8
z7Ipb555f-4^;GOCqA#n#Yr0@odN0(!6tN%GF_3Tio~Bf&8;i6!3DUuHb_9+3d~nfB
zk?3A?AXMTKlj>e3yf1x`;jyGwyKUSTUUYRo1UsiX|Ffz}ATxdc5Z}@?SPNC@L4%C%
z^Z+N>*8ld7<Vw)*Ee*|(xQlETuiFN}Yl6RW;eB-)F>5}LDYR*RFy$%kptKcaXb%21
z5@qp`h62vL0%*KtH)WvBD91~s)1uG>+rfVaJj_I|xE~%8A8Xrua`frgP<;5E9|HX}
z_uhBZ%dIqfrmarCK+}M@pbz3IWcE`dL0XLvL*^qs9I!IIp)2<;LTT;?yv9>+DLx8G
ziRG5O*+W`AxRK^J-h0DyCRL|Pxydgs&mx`ai~0Pb*I|=C?Kr7rQd7qv)#IG-p7*bp
zPgSy9$nTg~DQ|}rdnE*M5m5Q1>ffbLegMsWb#3ykFChSD(>1ARUGdmSTc~YeZ?2bp
zRSzR<kb$hqv`vJ1eB5L4l4SS7IKnLsK2!_&z1GZv(7_7Za)>v=>qhyG&7*6_YnbQY
zt`GMw63_lBn?6}xuW?k}HrP15LIqm2vqq<UsmT3y@~s^ZcT~Mtmw<2?+%OyagfN<f
z=o`$KDy~}R{U!AzaPR_)B%ub2j+z3<5l&)2--+Z`PF&$R_QVClX+m#<f4675s$qpQ
zBR6TsW{wc&o-myH!{P}jfP?<llNT$39jOb~6!GHDT&kis0H?>eX6Gk1BZX=sxesOg
z_KKg#UQY<vo@3CslCqpov#!&_$*(eM3_AY6E%hIYRQ*v)2%GFu;S-zN@vypM*NHZ~
zzYcc#8`+X0W3YJL7tUNc50banR$wjcqJP?+J8?3Febc;N)mi%DgnJF58c!y%&`%<o
zT3(`${4-qF^9&lUJbiRL{e84ruKih_=+QujMN&Taf`+!_cR>T(-S)9+P}oVSP189@
z$8?m(+qVp`4onEel{S@c((a{=0`n*&QOke`(}1pDzBQ<F(S%Z~(a6Q|OO_gR%+PQd
z8F>l@;)%Y_X+B?eJ?cD%biwW-hpZ0k8p@3BSdhE0M9g)e^x8-3Qsn0g5j#@)_WtWN
zIGZm7+Hox&sG+1J?kF0pDOd^2kUD@4_u7(0-MqPpsRagY$#9qqV90_{Ej7H(h%qyp
zS{@WE#2BYIXU9P{Zo6F!f@3_MU=L3Z<8b_P<Ll$rpHAH+yJqjR$yW;4`58>i=;=`B
zL<7$=+6P)gVj+^>1P}f$^=4<>_)ltdM`G||c?bpX>rUPKa+3jkDi63;V>(=FtkNTI
zzV)Jq%x6`tTX6$s1c-r-Rrcl?=!l01ufZS0WO#wf0D5|bSS$NDpN=qeda9>>Iqyl|
z&8wgPQw>v1Hw1im@3dVozS-mGjvK05S)s-I!-E<f9n_U+hh0K#^}AKp-VD*afZuQ^
z)vTBKlJ6>!5ewvQQr|ihw#P~3?Verc8?9OEgH%#H<`np^+55Ve6r#=lX~MmFtmvKC
zpTr%xhx;M%!PmZs`&Tc(ujb|@ed8iWK-;44_TFBscvQ&ojCa5uNnq4K^(Q;^G1H=@
z_QrlTM_~mq;_I;O1M3Or?O&SLa)w~_UbT;;lF{dsKrXuM1acbO>nH!#4$d-1k%2wM
z7{V|s-?l4_u12o6)dFWQk~8T3-y--{#mR03BLqJQj5!Uc54chIcg^|-OAVY<68_r&
z6yNTlfGaoLJ-xnUaq))%hf<_>R}1bKfx(K*doDWz*05Km%_*5Zr8bcE@{iT;=<S{E
z8`j8VA4e|+eU2T$)Q?4FRST$caJ?MI^gD)YJj_79exbI%ifj&cO36Lk!C;#og$C?T
zeUJ(u@~ofoHkP(FC0^O{+uJWZCi8s@OdxRVs<O3j_E6r}q|KM}2!-v8%mdpvN?`9v
z^ZFlnp!#n=v5%j8`~u-DPwMyG4ta!D(%^1@U98@WWP~C~o|U;n6mwi1e};Hn{6f2`
zPx-vQlIkP@6GQ0b58K+_{<1{$n78a8_$JdSFl-qI%~_B{=dorj?sCF+`~jvd=`4i?
z`i{t)3Pvt~uQjaIz~@s)s?#=~yzdLF<=y0fZ~d%Q^#ZoK0+A+jI=amzUh<6#KRvGW
z0Ed_)o`5*Vo0FT*_JJN>oZNEBnfmXye(mUT9mwF|M7RiE{#(0|2i6d5hXI(JAyHnA
zRUZZ=6YZUoxFgkQ-#}+MEwraveklbeObccO<<21tS(KNfm|#hx-KuBh(#7HT{#GjD
zktsTO(Ggt9e%@DYk5LW0oy`mFijWBLEbE)>fFZRX<-&s6jBb7yJ_f_s4RzwsB<p4&
ztM3Pd*?&1t76>039M_O%ocTRW<aofrEw2?f?^3w-b3IJ3T~lWDocI8wy!N<Y4S*B}
zmJM=rm9qtimq$W1XD(9lQ`Wei9o(oT>3O<Ld{81_QlfPAX~q<pizqK&HJ32u-I@8n
z`R19HCj>B^@)lwtcY(otaRuXB(;!R!pEkKPQX|@?nb{RIsr_;9pFBzvRn#oRewGAF
z2>XUj8na{oZyeSx{`kY)LUFrp(YhucZP1xn3su;e>RXZZGAI3>&Eampm+75%Pdyl|
zpR?KFlEY1Hj=!N&W~4ZLmpwsefqUXVQAK$UXY@W>2ANr!$1G|S$uZ)iCT^weFpbm~
z_d!no5ffAsv!&OlrzO?Yw>K_h8B@;Nfe2Y9MBUO>Rn|f9e7Lt|ESvzoT@qkxOnjn}
zo&UU?zbn;sFKGvN>%xHl|I;}juS1A7NyQ)!=j8KOwzSYxeD6FwyI2)X?FnNOyTXyV
zHTp1tNM^~fPV(<bfRblye}5061JH+s59O+=b<STUcvXgHDLz}2yGmz~!4UxSKI5BM
zu|*PO7x`4b$4Js;mBw{iE(ao7TIc>P%nHta(efU*H&O)1KxXjLR}5d5p;Tsk{tx5>
zoofkJ%Z`F)4S~-?%q2Dg*%V7%`LoR%@IK=7Z}$RT@HDBW9uT@+v@?j3Mu2RZD<yYk
zve~04WbFv(+p6X+2C+sYK!&(CasveEW*!q?<Gxj#;haE1-nYfwK>X`}I~M&l?|8l|
zPU);<N7lICk!6)kq6gT+MS(<lK7N_54BG;mKA}zXoD&Zz7!{xzL(;mYc_|cBhfbbW
zi9tZXi6BgjPgB$H&MRQ3#%C$^-Ciz~m7)A2gg`;mbFfCRH2yx=*P6rc8zltOHV{FL
zutm}yI^<}+Nl=8{OBTHmbTdHom|2Qze#ehi&HcTQ(AH`MX`r(&B1#dvuo&F12XvPt
zbbY-92HYE-?Iy}md8_B|EYhvy;!xeE29Ee{-f-f)7wuLNudE$d!w%cfU1Ilg)ZfAP
zhvIweSUc(DH1@UK7-fLHJ1JGvZDM%hK8B%3P_2{2@R7m&3tNAUQBV5)t7@1-R8Yjw
zKw6OmurXRgqu4wStMyptiC4%1&b$J%10SzoDUNr0I`9yX{jjyeL->V*(Z~gJ$Zh;l
z>>Vvd)h`Cq^+b88a!{eG<Oke-$&XyBzhNUpjw;xnrIej~yWN4ism<>hrn1U7IWF6v
zA_ilOUq|0PQwKhrIPEZPT&ak^Y(eF#QSj}|f-C;;*i92^#sbZlGV<B41Ye=euE%rg
zoO#~i<5wh!Y%#}e9yl>Q!pcEgH_a1jPW?H0%Z1HzrP@(%lXyQDiPK`$;ctUS2-=G>
zPO3TWEW@+~OsE;^L(>Sq*g6?I7t!>LBEVDRyHjuzq#6S~GY`v5yJ~Qy9u(8~{#o(m
zvXeGX48|_;^v!rnOyZeN2s*CMmWmn@72;l#M`I=pV3n?^U%_9H3da4V^8vt7{jnfZ
z#51T@-vAe1nul>wx&)atoS9}s8C-zboutJ|(Q!lkr5xaJC*S-bDjAtb8r}8JJccQS
zX-=Qr=sHyUs||-GY=luqkgxS5_^-`-k!Ya%J^#@n2IcQF@mIL@6D8~*?+I{B<Zes=
zFkf0)voLo{j-c`<L`WI6EG;W*+1picY<VD7$?uGXOX;*KKRKy=hK&1_bOSy7Q+BWx
z^g^#bkZOcNu<-}`4{+aD`?HS68dT@f_UMN+4ScoOb*v*uG9r!9*%%=oYz7OZ04l{C
zlq!n5!L7LzxW#lsBxl~?{A>xd7lDjHA##-zLPFz15;xQ7yUSf_a7C%{V>*?Yt81``
zO8#I5;t|%M&9lBxV0{CPM%0mkjBm$&!vdNS=GqdPvr?yyoxvQ+#+JTF<HXBZ)5f7Z
zMi>vVJqBv-htc$jWW88S#FQo_<mhN;tk^T!fxV21mx#zw>_c<PR4k;^&O66|=dz9Q
zJvL*B*Kb?`oW^>)ViH$;?NGT_E}YZNK!0SX_af6yB4F2&998?}k`MR!F6Plm&?cj^
zwX>=~Txs1S?m<(HGYUL(xkM6lmO9z*9xAmR1AA;FfZr*-veL|dXp8MJFenlck%mWX
z3uUow>C$SdFIK5um0!neUyh)V0At7dAJMJ3Xf2IjNwcl_4++NBz0vx{Il`?uXC@GR
zM$?cQfS)pVzzFM|<{c8doiRwK#(ZmC0a2RC9*#hvzJg_5om!*akpT|zdM)RHIU(<(
z6Tj8w0CKA5DXz;XH1W}4eL@~=;E2Tm?#)*Tf@)j+SPY!Lb+63`X0NX=&6@)@9=T7j
z^FF^(_gGr;L^K*baeAWQa&1%p1h;wm@+h(b?qVC+!E-rs!#GZlrA1W*RH;Jt?|@sO
z@BGLJPmD~%z7Bs@xCzySq6Rvu)_US8fq%vY-$OfnB~(kQ_B|IBz<7gpoJK~-HMN^p
zJ;u9}&oUPms)4~sm{0fLnqkfTE+|mycXnSwlP%cddBkQxHge^fvLBmQt@n-v@HM%4
zfa8PQbV~S1LRJm4OwqI$M`|qyx%GYyVfr0|du)?Gxg}J&#_MFKH&*D+JTiX>?308S
zxP;<F_nL*7IM@#lO*lFqjR-n#KFA3z9?W7f^}uoFCuQ$dFV8yfMXHFc^_Exe-INN9
zh1=H_pC5V=LjET7?2Q_<`enukY!e`Z)hic)U$@@CU7ogoeO7UKs%}=R)j{wS@$jU$
zuSo%s$-I^cJBm#0jIA0PS6$TF?sPegdGy2pydKt6nA8kZc@sdm@=^WN^*cg5P%aG4
zp;jKf&m;zID>RFw3JF_v$cZMN1rtoL{Qx+Pst0-p>BL!Rhf@VR-#z1>JqN-JagMX8
zNJc$H<KB}CqS@(aZC9z%xa_FJ#>?|}HkA66a)WTg5yEZ*EcHd;6Q=j}h#OqzTWLfW
zZ9ch-=$=@A->;dYXJl%@u$z?y;i08oU_qu~z(~@-7Mzw&k*Y(0r<d{{*O>0Zj32#M
zPHtl!*E$J1DHr_Epj<}VLmN-C|5+a;EN32E$j_Tp-45iQuP;)chwp7IuG)kMED}Mt
zJZ1PnETJjPmfQ<TR=@@DV+YMU@q3~56NHjtB??p<!vt%TBKPjYZ444lrx^~vbU`hF
zKW4#8L_(&C-ry~8<5?&8s&^gh2}RYQJwh)3q1O#7&KSdQ;Kt1ug_wlIiNfd%Znxmi
z!O0mQFxuYyZ}X$Eul?)zT@(D$?|&|DQWp$_?ovlR?Ku<`;o(vfbnVOC`DM6UoKfvi
zZ*Q*~`k*oP?ZWdB!GE5N=n#PY>)Xf*y!u;dcvbtOe6^4Awk|^K*z}$6Qr{@~WNrWE
z|91!9S7!ZxwGe!B8{PVUTL0&^)0<-=5BQV+x5?(4CAP!=_d?1}ggTwMlmCACe>d?%
iGO7LF-@HfdhG7p9JH1&iVWfKlKNMtCq-#JX!T$xIn#Cgk

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_007.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_007.png
new file mode 100644
index 0000000000000000000000000000000000000000..d8891fdcf4d8d3c30a3291e6a5416f027462807f
GIT binary patch
literal 80795
zcmaI7WmFqc+b#?g30Ayli@Uo83-0djZpGc9xVsfE#U;3VvEURd4#nLmoIK}y&-?uQ
zeq<-JGBa6gXZC$xwxd;)q|s4`P+(wS&}C&L)L>xXJfNRL0Ep1B++N`;^o`$LQpa80
z*~;C^^s6O|qN%%!gR{GXtr?l8<ySXbXD3b;UKUO!G8=by7dHV`R>%K6fyMc&HS4sf
z_+RKvkX>YS-C$raO#kzO{UK6e3j+%SBP$`M;hlHd?UhBfqPzdE^C41R`;YdTopZUL
zLE=M-lqVHcQa>_P7Y@_+L`&CL9}`nA-PceP-7X|@3sJb0<PjUGUB!oHuYJy{Icuq*
z?uVCMDa}@$9%oz3+e!C=!(Tj_>FHR!9F$637k<-NOM3DcLbo&G|KGrBJ2ue```=@s
z*OOUFkN+OCez=t}{NM8jCt!rsmjCBcq1#Ii(Zq$j|92~)F0`5d-%J0997e^lPfky(
zF{ERHUiF>TF^nm6>?@Z~6ciQ7_sgyx{;&tx7JE(B0dEtY(xd*c_g-H@gbv=B)OZC2
zCr09ldo75U2YXqk3xp;j(TQ1(_`A*D=A<sNjD7xozG1}ZHV@q2e)pVw*DbBpV1T`3
zeBZ@?N;iHqS1J(xh<*Ks;KM(xhk&O$65+4mxy~KYiu$}Ir2~2)lEIkYPGQN)x-K!{
zqQ%VW1;`b-m-NExW#}(oVKz}!$HnXK{X^2Q?U_e((o(#864g-X5oA+RXpomyV*{sQ
z$ihLgq(^=-#zmzPfRIL$YDxR}CEtJTAV8cftFN#Q6I=-DPCbBLRSgb~ymakLrHUGP
z8I_{?juU>3`a@?tqB*-3j+bDb@Ebr`<R1U;IrdqLiS6)=T%^w#@#e3+@*2Rl(Zn%s
zu45ECk-1t5SNZ(WTs*gJ%h@VH^Q?}!jhDddj`QpN^HnGAt9-%yU@X#qrlRP7XYzgR
z>cu*_gB!%d&p)2QJfFuO1n0>06l%G9fXyj5vcfg5Jg0e^`F!b5YTl>EwsNx!SX%8Z
zO9P0SVQQJr5!86x`z9<r1%_WloWP@s(nlQnyh}f+0zw+-)%1WEqQ#|AQ)5KIX<tG}
zx1(SO+<MhfLf8@>%WRJU3z8V-xFcB(oy?%XtG8+_VS6!@hzbx8LWxE)yC5J*OFvSL
zB_*CiQ)(;MN&TaSWLSFwfY8IS_hCY=dvX@Pz{={-2?+5j{Y|7G#aF}a<mVxm7i;U$
zG32U{Q(mf<T8cbEu!Oz!MooQhr2AGD|DUI3M({xUi|^>`_~855<G&`E4Wa_y9W+5t
z$Abpi!tZ~oT`qGzuLm9qkBpAkv}F36ucS^otcbD1$VBw%z%EgQi_zC>eI>WmA}f9A
z`Ng1}0B5y{gHnb&o?6Q>f)Gkkrnkv%kN!#8Q!O`%2vv(LTrxL}>rAam>jw)&y-G7H
zi@@B$3>(6_9T6Y_oiq&qu#}4+Ux3374KcQl!k~XPnbCz?+<ul23r{T+OUD*V2B=I?
zy2fM9Jck3QQt276`+!n?X~P3iVIdX!-tTLo4kW{jEMWysF)wO!2(ZV45ls!LX9&W0
z5N80W1+0J?KjwM@dn}OU+RA+Y|E%r7^iX*UkD#Z~#2Hxz_cd|>P2k&K`U8lj_k{MJ
z@ka|0wQ%<wX2h3gEYSCOYHq|zQ8@$oC5s`2;Q9nESQ%KU-yU-ZFR(oqlx6lJC_2u&
zwaHA<n1;j)#aLR~vhv+Vrh(_UXIkmU<Jc)_+04Yh>+R9#5o#83n3y$nj2t;QdO{pI
zmsO65M|EbTOyc`d!iLGPPBg{HtE*0tL)de;6xS)_5J2SN5`YhF4B3`7l8l#W`gZ0_
zTAcxvH3hUs8dq_J-6@oIJ-BIAmP+}+k>8Q=GK}@B)j0a}@St(-@n}+l>Uuhe_)qw6
z(T-@v(#O7ki=w{x{@s@Ar%o0yN7!gVz-D4zJ)>zl^q<cA>2f@Tod|qoq71-sSY?Yy
z`Q(TuSC@`L4X96v8?)twM84#RQIt_C$UPAjb<TJizO+{LQJ-*nlCdq3)w>gW81cRG
z9YQ`2|K(9N$tEE+ckqtuNv4PZeW_Od4yOmEsVX0M+J(4WKUotNu8wDX)}GxIIb!hQ
zpj5VsA0mWbQP_~IF`=g}IBYL##{YJR-<rc}w}KL10XKuZxHvC^U77;eH0eMQj}bMS
zPct7i-g?%*7|MjD>+Z3F(K|Wy=>G;|Oc$=uKsUGW9V|tESLfsW-!e{afL)2Bw&TNJ
z3V@L{BMQI0S=g<pTwZ0GrUkyAvO9Hc!}O@p*{Xs};#@QZ>>CD*Y$}N2hsmmSX=UVO
zOK-1$@5LzDU7{v&o*<Zq&r~S(%<)$7y)U^UR)P0yf-Q~~gA4Gk*i9X3u;!SiqL>FW
zwhB##;t?7C<i#6K^YPmJ*RTKbgr9JUTTZPf68^;{I3rSE$}X+CbBk3YkgF7ceqO@q
za_T-Mfm^D?kZHfe&<?r4A;*qjk23$PLfwnezbu73&R{MsY9;{hN~-{HH#bc~8(9XG
z!Ly!*d9LhfYBuC-gj0C3C?K+Q(Ea~5B?I<0Dz(fQ+<iaq0i<oLd8xJV)KNyAU768_
zT~rNXO)11kEU9WYOv=6_^e25x9?9Rhj7vY=P#_QMB>5V;i_=&F^xb#-0_H+9MCKcm
zis}MqG|IwPWQC7I6@*{L*!lIW|BRra#j)`SfLIDgF+ggLqM|IBCG!qs)#XVTGq&N)
z?3G5_L#Tx%i+U!TA!~%6*CLX>ERE<evM+%})B7}U6qf!pSECZZG|!d_3}e7)S2lf-
zljUW%A>(F0^UQZ@>W}vdp#CRrp5-*Yv%{Y7Iq)O>8r}BnkL4rxj#~@@HYY13K(Zz(
zvs(L#xV=!d46N)iVR#ICW&iDJ-T!oIm*K4xna^4R@|Z`I;S}74af+%u@CLE2fwd#A
z1`$PXVzen_xfO#HU@?945tBHnEsZj|yWu-p@}=}#lLkUtI5wUp4E7wQrt$G{j+>@K
z{=i!{#&pCDtEha=T?OR?6(-!jCnw_cl3TLJi+DLxG}cP_mglcb1xRH*aHe}Wk{>3g
zSupoOc2!Nl+ZBuk#~0vBZTsgkU^Nz?YQsJj%rDrgSi3mhL8L_-uOVudG^fcl{LJ*_
z@50<si)`vgw+bb7T|LXeM_kudMND|W?8pt5>l}GPC%kHbR!S&tX?6;Jk;-^Y#3G_2
zlavL6A()2hs50-r3%)E_fU4)BL?amY%Z$_sf$>4p1*#tUVDzr_N|QaKoPbzJ!*|X?
z?h=oKR;%U1c?q7*7Av8hwiF#&8Z|xb26dDVrLwu#1G-9uz&I4(bvtml3-a4LL~$NR
zK|!IuLvQqfGLAh>jyjZXNE<ts2JIG~9<QaLzU|uug1g}7oGFH?s;aTb(c4I5Eyf$*
z^UNsWMt%0tug|r`HgW2=a7%5Ghr>%WBW=1ov6MlpR5qz*>5~$;aMS?<WW|eirjD7M
zqol)fg~zXa@?h$;TTHltnc`Gj$O9q47t6B7m<t~rvPF6jz`raB)Yjs3zj`@r6#V~C
z+F*JJw9@04^XqYAC_mF}7aSuD=zb#;wTr{*a`DdSys%A*y{=`pWzzXT<3ZE0G$1)L
zG(r&mDcBQCovL{h{?e@n2wwoCnx$ktt?WX}1^%~A=IddK@)>0^YCkVWi`_`%5t>N)
zJ>FTTQTXIU#OXqj#HAd)nn+r_C;x)VU#?%<;lRG}c6SKjT;T=bxB?qlb|sUC!Z%nw
zJB+<J^l%bCF$LD}wsAYj8h!e>aOj4dSY}XWK7|jjvd|NSCl<06!`Zax*_2?p;)IPp
zHKB5upJhnO6^qVrr}XrY7qgMgB9Cit@y}btdfYSXV5g?&#bbS9q#i)Mgo}-tXI&vK
z`%m5fPx1G5-A$3%aP)FKObzlIPbR=3E46s3jW1;`Y`@WrgyIbJsccSp4J|tb#XZBH
z@H|y>Wdz+)T<|j1r;KIpfzBY4_@QrLc%<a)&{6;;TlvI_x-{SQI^}|W>|YIaLv+G*
zaN%c8{oicD5+%BpJ`D<-n;_XlD%H5eMC=IIg2UO(P35XtifEi2s-j#i_7Uv;n+3?U
z8Yi>WA%6Cswqy-2pSKm-EJ3fBDg6hN2vOAb0s&KYi$8;O)a!A%@bv@%VwH9l$AE$~
zI#y9{`vNz({fI}S1g)+ZGO-huaW}ZsH=E$1WY<>2r6el$c7xfKVuQ3`Q%sqCa;ER<
znArPB@TRxa-m(Af3<%K9An1BZvXK#mjinZDQlv7s{`CU()1fwf`t#!$Qh|w2!}JPg
zY+PbOO)G7K=Z3fU)cNkts9u!|^Dk3$5K6U-suoAkIUlW&H+DXk4Npe2py9{}^jWB#
zbF|x0$}KBH$e9u_-!-S($BGoTZJio>BoM=F%fAl@Ws^4JN<U^my=l^5upDPp{B$4~
zo;-Rx0+by!Z{UEGU<5|i=zE%`gRC4iXk%ph1(XHylRLrVyTdO(IE)1|Q?S0X<Hr7@
z!(;!=t6=<2P!&(TI;Cm1NTe2u?|r)N794#PPk^0vIPcU-kNe!F18ZM{zziGFS4(Th
zcO_f*Z}oquiZ%*M^upakrpjAP6M-suG8#edU+arMX9ga9*ZsY|X?A6??2&_ze2e#<
zvz0B8?5RljZGJan9f+I<%CbYGlTg7#G=@Rpya_jn;44>F7F7b4mI@s%-~&eW$Z8th
zAv+KloP^xeR6$5cDE*B@b#K<G6%YzzU$hoaNWj$SV;5srA_ne|s|n0d%A*JXe9f>k
z{|NC~99ADKpiNkVH$|?blB2iwjO${P&fFjNEjT<tT&C=10^mm<FiHv+mXw$e7<{Bh
z*xy&z7)Ko{muoe@>;{PCvc>%l+7ftvmR}%7FHM6Ek&)8Nj9K_HuH~3oaLO{_3RsmA
zmGUInr~eliK?kG8w$iG%gpE|DJmt!orOV69GU*gg+ZJuYFJDaDDU?}gZ59z|6Q<^E
zc>bnJN{1Sx!Gq+Q*;yF5DV^o?3`Q5th9lsStq)K?iC4^*H*(1v|0+pJRI!7vmeQMg
zTH$oaAu#{`TY2vg@Q<YDIyw|4)du%V`*+c*S!y)#=jZ1m_f;tSg0BQdzP!i^AWPXl
zzfd3|#Z+sIli{lF;m~rYIHj^3yA7#lIn+-(?h?ToHd!TuzBm0~M<pOvkqN&x;L4HR
z7?+Q2V2=UdbS^OtP`f-96ff|`P!H1}#upLOlHq;|SwaSEH+rNuC18Dmhcl%f|FHk?
zU%){BpBX58+oq&bg|SuT_`aqFyzRRo`Q2GKPm4b$$=`DEdyG>#&n&PORJLyNy_b~}
zPs^r8HifFnL@cL#I+KO-i;;S=5n5&|gRF-_r+V3jL!*b4Af<Rfsq5qweu9d<`){Ct
z47f|CX=Oy6{6o7~hizO1m8@V{!`y(5B6B6#hX}0{Pu?=Ex`sx76#9HK4;Pc9x3_P4
z-LcnjBcm@rcz@>cx#s;t+VIa!`c#50W$kJr_1!UsT`9oV6;)R*1dro}`V-YBHB{S_
z6dR1N#6zFoR1$I@+;#2iVkUWp^<vdx%qr8sm691HCp}Ia5tpk5J2}^j5C4Lu{7U|}
zQX+3dE9G%a|HDq4!lO*(56u(c!z5wB>i~U@2Zl01Trw9i@{OBii&MTHEs?$Bco15$
z=4E*I*2}%52rYa;MSUs3hLwk9w0|=!<R;sYX%`7$q=HxV7EUSOKSZlatcdU1X~*)s
zZPbWdw#F!zKBERP_j8oAZHpiii>8hq3#Cw@hP@q_VgH(>q$J?6Cu#}#&Nsci-R<;G
zE?GEy@_tVac&P#|wH7j0CL5}hclPb@<%Az>-xM*mC=m2qLAryA@nt(F+cYNPQ~Z*g
z6|DJ^H3ov&#)6KVEP=u02LsA}D2aEy@hdp2K~)YmQqcJy9m7DZt@%`YUd%qEpDY5j
zJ$u~pwEjVPwidaRg$$M+822epfPWO7m!$6jueM`v(ktFIg9hKPvK=MMps;FCfL1Wf
zp7A-bZ()I}siMkVkoowI0Tp$EN>;(RvM@OrnMcXN;%3Q^iDimBm~M=+OuD9B4to^(
zY+TO)t)~mV8ycGWR8FoN<PY0{-h)9zlENsM{F{XEt!qQ<TqlkRUxX$<*Rj#jvsY3_
ztHWEk*~8JpV~eGfQ@MhQ)k`r6U<)w4Pc(T4w<%dKSCB||*0T|9c}*KJo>A$?Tj`EW
z<0SQ2W7Tfhc_h7X^x}!WI*^pX>;3RhqmB3lPokt_R-JqEGO7vrzYQ~!5lX}r{dkmw
z3|87C)U0I`;^3r@!xLxgo4Y_{F%k@LjGU+fiNg_i9UI|nC@d6&aBbhm4XOkT75>6c
zzc@{`3{L%~FJ?<5eJm9yH_){(bq&#K8q7eKL}{lm+6ue<-SQ8n-2iqob_^fM#)Fv3
zHI5bG%qstS4(@IQWmNg|?oFbw(nGs>wbro>vkmwse6pnp+0acN2Ao^~R(iTmoa|*9
zOo^eUs)Z->r|Q|0k#Ud!r=>Zw?a$<KQ7v8hxinD2&_mF>AHVOd4|GkFZk?{n4#dX*
zVw16Jiq;lw9P71UtPC%Y>7~Q-r=oGJWyn?MuuQB!n=efz`TYkn7@cS3Wo>o#{Of$Y
z>3wx8@IL+()JVA?j-d_5;+kS)q7Z|NLJthTg^30e$;YUF<ipuEFsgT*CxDT%Z<d|S
zIrXu(#ZE}8uH*}e`z_0aP0l3ut2!d5zO7!N%z!#G7L%fA<9O!<mXdPUU4UJFf*TP@
z`fs^Pc0Oec1x*(~WU@r#+Yp%N1qpo|(l9lFvnP*+_R-Drp2T{(?sDK4#|Lq8Ro;Gi
zg9sNJm!yyoF@I`yTR17oh7&Wu=-FvUeDuW=%RdVHQ6h8MRNU<H(n=bXo6|KfJVCGX
zt;wT#|K~}<9K{e?CM}xM=(0tJlBPWXM_n(s2Q#SAD(O(B0mh(8)7tnVf8jp>iV+qH
z=nCI{k(&G3eNYw&3_mOsl%lmYIG;K$ikb+OpA4LuBlpLM6^)(&>&db&MZ<?D;l_q)
zcRl+$by)UtPATRW=8BS+>OJ_RVwWbhjpt7mT&y+&j(&Qo1+&xPZ9VaS`t(yY-XN_)
z<5W84Beb!Fi^){-^?fZHw*!N=#Dav)c;X53^C-|L?C$V>?puzhbATcw$|!PMoGOg$
zRh-kKnMs6=OxxFsds(BDusikzWc92!7wZPfBdLze<`%#y$ywZm-ArDKEBQJWR$gY+
zlKLkc`n&M34r>D9c_VqbLh&fL(U&$Y=w-kJG`+9{_0^hBJU9KEOB7Tm2P0}*Jv4m}
zK<R^OG1SI!LspvV|DZ}AaoI$1B;)?qr>tF8;DGe|Ic)XEF+ZH<`C*6Pil{HOBv6#a
zxL4GJvgF1?)Wc9LA!@fa-8iIk?K9C`*ORe!(G|%SBjX;bq@G6G)Xti%nc0OZY!8T=
z{^_o3Y4>k)iJ}ya5KH2p4YSM+ynP2%w57X4@YNN2H5hogqO9gP47}gNIsL;kSX-n2
zyUjKxB$+%-jfrL2_4d*<XDTc8@KDE*o1n13%%Ohh@ez0G9B4`=qZWN}ICl3`I)W`h
zeSan-+PV0(bef=4LuT6;lljM_b|QCi4!v=Whp?}dS+i-GUWGwzqtAp?WE4_%K^Dc>
z{ts|foP4H`6|;@GEswpfXYIsdlG>J!uKnuR-46~%F}psQI8~h!J3qNwnT=rrLc)U#
zV(0CWpm+8T*N3mpO3-?HSr__Uq&K3>4L0bpla!XB*=A%y;*E`e5psKrISTor>qY&4
zAG=J6<LlvX$9CxC6yX{&N0gG^*c?m2@TLQwy-2r~*6PYt;xGpWw`8Kn!?Bc@pOsV5
zm&b8yb4C%|45Nn_e)Uv2AFNc6n)<kA*mWziIIj>$qJ8`b)tP?>UBVC*UR}9HzQl%)
zTXvP(%2nuE2Rdae3}k%n%&J)cq@?+0?OwV;8f|?;d(0C2jpl5qb?nCuV<%ZzIQiO=
zqoa@R_`yLx!R<?thHI@JjYz@pwns$X&+UhNK2`~bA4MR4HgbO*UWd0QNeyFHq_CY-
z6QP2A^QmO>O{mn{7|Jj~W;Jb?HCBcbYXo7|8*a;b{X%?kXo{DlJtsn0Lzi105*u+L
zhUbdM+t^r<<aIua_CA-}kcPHq_jcZjsEN3c0jtPo&oC7x=E4%s1K{8XN|G)+-Uzec
zTa$?lMd1A;W!l94M0Y!Mx-b4C`rYBhg6Z`t9@Qu!qEfJE;sL<Kle)s~T0f`nI!rtC
z$OFZ31HZX)UqWM1T=1FrSqdsXEXgYxka=pMl=Gf#-D~?J1YK41e&~p^uBMbU9qS9`
z6j}0i$jYH{TBxT%D8l-b(iT6Pvz@FuGBV;lRdrruRMpVYA!k$`mw+~V%5!vq5j~&9
zr3iTaHm)}A*h)f$=B7J532D&{#nWZV8YyNiHM<;MW8tJAMhxgWNY^&ydYy`T%NQO;
z2@4A&={kT>_CFWP$)^daU|T(+SgoPSS0Xk?v&}KM{EmZtVyM&0o#I3OwMP1YMT(rn
znlFes=w(@%z1KfW%P{3Z+J<@}({J9l4c`ALcXNz4XpmR!F&;bUUOPz2&UbN%5pUQg
zrD(?}^oLB9QPBpoxdZwJq2+Qxyn)xeXF<RL-F)g=`%MAyWB%s5c*1AFz;@ows4s6*
zuC`~(p~brCI=N(;;h03S!F_D1YU{p(!q0N#Y6SHtr4m1J^9hzc(@kbqUh{KH^l3dg
z)TiXGh8Kb4_1T8sG`>VRwIfyUHw8UdXPq+)e)s@CK&!$3RTY+Ad`^_|1<`i-r-r@T
zn8y5B^Hr%tc}^ZJJ=#4x-wfn(j4e!@I_C`0G^-*Fqf(ogE*INNz~N6<{Brj_vv*)P
z&lP?$&oUnOv~PGUJn4X}l)Kh_CWt(ZDw8$ICJE%Si@vOCSTkdnw$kcG3K6+Si+75B
z1I2K`F*Cca7B3ozJ@w<SqyhHBaZK`LhC|uum+ZoI<Xh_vjyANZ@k%~)((SjDfVYnh
zdqi0<qH5Qq3FyE09iTF$<Fh_k-Csib%H#FhZkDufBBTorNQIvPF%(6-5~EB-ija{V
zxq-Lt7e4npMt<MYU)MV*QwJ0^?$I)L$q>m~jI<wG+VprYS$R1Y3W+xb&hJWZ%NK7S
zm4)s!lqHr1Cx|lLS9*`P&_DzVc>=Rxm~?19rr9zXSg41WJGj(nfLw%0V*~C+-Sp|V
z!kWq0pwmJ+E@XofcKrSJyJ_v~XWriYT&M1^%3tU(P=J;bjDkYI_)So8t~9y=-gaF|
zCrRsjD#3>B4zVKHch=03G*(Qnj=vJAp4?L6v@Ojz6P-Or`ULq|eK$aV$moM1T(}wv
zM4b*oZT8qx2$tdmAN$VE7T>M*fNGY$uh^+G?P2Mp@bwc|Th7@Ks!`4?nN?T8ED?h}
zB&(L7inna)ZERfL^3VKQJMaE+3B!I+w7)RK%YUu)dW5$txet}Dj-({FmQ96Vf6BQt
zJ=bY%82TkICqKJuJ1;ou9kui_`uDi=zg61D-_oSzv${0+x18&|Q+G%hstmUeQs}jq
z4!POee!Yi6Xz1w0)uoF%<;hGi8KUT0PbpEtUTMmKr-VF8PShxr{vsCsW}y%MhhDCu
zXRVZe7wKP&M*hQ$8YuDVu|V={D6v#CpLyiOgYc(;ukbZ1K5Rr)XmBr*w%GV^e_5OL
z^QH42i!W*dHj^_@X@=G0fR*2_!${h$SkAi+!KbV3yp$a56pC0j?w!*+mFDg_=n6~?
z(W2W}bcN~H9H^*G04BLmHMrFB=3HihnoqVQ=FEF2yZ(2An}dHoEQeteM&BTZ29p5*
zgGpq7%YUd6kee>n)i<y=7d<tcSlZP4UAo-fwNFe0g&1a%jUoe6O)c3XcD$&5^lXk&
z=PrXgK3M17b;4un6QWL`WS1yg9T46q*>T>Ba$fbz>#giT=Er%3PrqKw_hNkt{9pNq
zc^xd@-@5((32w?ZsI;x;t`ELrAfo82qBV7C9P}d<4n2U*dtQjnU#`OpJWL{;10Onl
zrmBo32_iH7)cqcL!&9)LYI~r6+2u-Nq4oVb5WK|HP@Aeg=*>~wvAW*?DYW|G9Q^Cb
zb(uQ(%lHjlx?I<b;l)9XbAUc8s39tonT46g0gRwl5(jl3jdJS6h7=eg8eyxf8AyV*
zOlygV`%p(zKqgGy+16`BBw{$VlDT3Q<2v+IX5HyWfl6`}ROxc$LZp;<rltfCyGTSL
zCP;qK%1=tB1ugt+A-5D6Sr9~;SWIQC2q=kd?_SLBy2B?%z$OfaB^Pohfhs8ddOXUB
z930y>=@)NG_XZxH6KJx*NvN454OaMO{aKS#1o*HtWqnphg!ZKj-1MtD%Ly6dm8Lbr
zG1bpj*l}QYGjU!Xoez3?&5><u<mb)xdL^ZO>i8*CBq1+5QE70N2unH(c)a?wbo=E5
zq#YStESFb_!Vt43y1<71k&v?mZ8Wz-t@ej>q!m7JvX%qC_Z@fFU5)ep!}}J7uhT6n
z2)6_oQC?6CSuY>2<A}m%gnjkvyyFX>$2C_j@A>xX2}Zp$L^$w0@$FkgM8rh*ijaRa
zc^PEkZ#+pL4UN<1^2HzOj?=SqmMxAeQcWaDLokuav)}Zc1Z6N$9PGY<r)nv`K;bWM
z&`M@9<nZxE<lT*E>R|s5_>1pj%FY}%HH@uLbMzI_jl&$O1Axgfe`=yE@@cft+}k%a
z-s?{cO9rNH8MBfh&#@l2JPN{(yQQLi5Vw<Kxs&5xV-#35^0cUQ2FAjdA3N~zBNU&R
ze=OR$6M2lg#qP;VB395q<q$?j%?x7A@rj8^2qdMPJ)jK8xX%%Yxr0PrRVB|Q=n-m=
zK<wO;TH{x1`O;53GeUs|bs4haN-Z@T6(+XnG3X~dNayrk7JoCnO%kz1E?AY~Mh{!K
z8u8?&%g?T2+JD1|b>6s%YpA6r@v0xOIL6-9o1J@lG==vJNMNDtrI9aa3+2ndKwAI0
zn@@Eo3x(=bwz=sCJ}BboJ=?i|#__)ZCk6`z`|p;Ww+pn560e!n#L4{60~WiXuKra+
z@4Jrkp99J8kT)@_@{dJ#F!6TTT3zA&*xvllh|hfY1N>06ySKMcWAJX-5VCNK__ham
zOOqxIXwLj`%@Kkb6Wwo52VdgeNgq|HU#x3P#S?q_hjW<mq9+35ec!FRmZu-^@$+hb
z#+15@z3SYKpGmc-zV%#f+D;Ge`sm&eRqwTZuUw5*=X&RFaSUS)C?-$UDz26qu5!S{
z^BiX=Qkm3K0jSy6Gfy2sfC}%AfG{KsiU5pMgn2jBDwiquN7B@eSp9nY_tu0OvThj2
zL$ZC04tY{PuIDyaV3rDiVlrGpR<bh{b1!BxP9X#j4TFV^12{B3-Z&AWh<2x!PqtiS
zz+9p@UVirSyXD4rq`b8VRt$eaPXixd-;n*Qk=-i5$}#?1hc_1#^PFe`GvSL=PHL>;
z#=<GDLAW^$R8liB%+v@LP2PI;b5z4box)*soUEO9>jw5RtT}b5_>RlzFw`IZJg&CR
zS7z{W1R++JuH;m#3jI1q*WQ**JeG?+LjTvbTXjx>;;;501bYaC8A*QOf4Hnxso}jh
zE6wt7ehUL10`clSTx*a`4tgA8|8zLs>=B`oF6s@u-a`Hov`4r35^zCw3Flz-)NMM5
z*mEM-liK?p)q7YcavAam;e{Hbe`xCt?NjS~G%+2rZL3&GQ!qYt*FLjXliQ0@xTgQ}
zZd9fA+PVis40yz^nvL}cDmJZIOvx|VE+@udUw9{d8a95?oGuUv%6@2y`qIM%?K?fq
zF-bhR+14Y@+wvHtpdFcftQl|;Q{C6s)WQ<CKo(9zN=|~MmAvdIjXx1SO{d~5Ja7kH
z$9vBoDn)$(5NRLbo{IuT7-Iq;^;Kxn`WDPQC#Mdh_f9{nilKq2Ps`--I>Sa}xw^8h
z;g72)NI-~G)!^SpIx&LpH4>Fcj-7SpU#&3Ud#XBDQ03Oaj)s&jCwrVPe{xLB&d90p
z&obLgyuK#0NNhFgftPfc=tH#}W_0})lLykMWE*2MVih~HUOQTvUyFTuJWf&>gBK8t
zqHA{R_SR8_r3-|jF=Ldh)8$^Mmb`&CIuRmIn~5&Re}T^ayW}IIQL@b=7PH+@Ri{AO
zxM8R@=!Un^XMSX&TV4pJb=_T1pVx7hRM<bS><Qs`!-pzV=FS{FEi*?U4x`@G7j1Ji
z-i9vv&~NW6|GD2_!0Q&qtF-6&y3g=*M$ZCy_*Q}7#s2ZX<=LZ!b$<4vxibx0ni~bh
z1k5KNsBDE1a1VcTSVh9u7J^PJ&5=ugvc7@<U@Cx7j+x{$+A>|~bQ8SzS8!o*c%MDd
z-3jFv)cw?kcStD0tt43%(yFPWp_=eIrI<EHDdcAGQb=iu`G*T(7^XiJB^dIqxs))q
zf-f=Wq%HQOQQ|S}m&i;a`cR_NvA5}^_kf#<t!4BZTA9A!5xMrs@voobL2K}i+<}<v
z4X&s!9}UF=ZGvZo{0KLLg3+}<n+(?G-zI+ix`B?g;<G6HOjFKLvWdG7MUV;R)kEb(
z&C?9-*yOlszAQf{V=pznr3rgZg-TkAQ0B3-RYy}bW;9}YkFn~Vdawgn&L=EKQC6d=
z8cgCva|z7VQ>Z-H%}ZsHi^Hd?<r_<pw~fRtn_|vSSR&8%<N9~F{3?u;El`=J3Z%a!
zc~%YBolp&c#6A{PLv~IXS~uErtLj!lmLjR5qj!0vXl1y1nq_i11QsBz0fWo)+6;#0
zhMmP&2;v$CQjfQ^ALkvtDqAyT9P+E~&uCHB8&t7tdrk3Gr$<vG``jFPoDCQD8mu8-
z0#Bs_!(WUZRtvuEP=B+p6CMiL@Zt~@jG9G^raI!O^qfv!Wr^?De+ay30zPJ^PUcr8
z#ctJhV(e6YTWIqhr&6q^@t|ZKC=MUSAu{Ebs<6qg_t`}Zy!7kMwf}h9yWY7jj`411
zp!Rb2hxhF8=SCRG+YOJk@C#kRl2Zce_TQ|Cg>oI>Ypvydku_o7pRvB8sw9R!n^>{p
zQY3uM3<aVOEN|n~nslXVbo(V~%4TTk!weHmmKhm`^x7FZhPp+6s^=B4X^?Q8z4u|i
zKJln#A(<XsFk&sE*_I_JRdG#N{$z~J`aLJY1?o}xK7U+qP&u=h@z%inXj(vZE^z0R
z(fX66HgMY@wZ;V*W5h_MSRf%ob7u&_+(hR_t>=e?HS3SihFU!&IL7bu+(XG^w}xfY
zdh$6m1SOns{zU;Urf(DVR`|P27JC!EJPP!~nHvvQe{O%aHhw+eDtwFIEKU~XzM>nI
zKmB}RJ$P=<Ua8!@Vrs%RO1$YVXiAV&j%<tFnJrVMS^(PV#PR>R(hXe-&pE9(_PuwV
zRI*>&z3yVd(u8rumy?E4q)7`(H1*3<S-$Qn3br3k6F>cqDzv;mWoRGU-ReIBd?1LW
zrILx}`NX&!t{wWOmhAle?A7S}@G<b^;r(00uo3pKJS>iR?$b!!`^UhG08;a^b+Wja
zzoqg``sF3ssua>bMdD;|!gsTWGCOJZ&RJHzz77Ej!|`A$0(ijwbmD-aF<Z`pJNt-F
z+(!(I%Rf*G_4XIC36U1Q$LRK%{EzKJrRKFe^@3#-b>_7x>*JZQ_m@80g@;lgcV8}^
z=B}^lOUu&sP{oDyyEo}<Ibk(QJ&1oc>~`IMC8wbtVN8v|KjIR1;F*cJ8k(cAoz4^f
zb%i6%@4G{XDj&;5gucUt$oQLwe{5=5d43r`3H1h^K@>6O2ZZuW<H$3I%9dCo#uzR7
z8AZoj2Ph&9HmWe+>gy4wLWDo}f-_!daU}%fr-Hd>CGNRdFqzHKR{!3m#^&~Re!n$r
zJ(xhtG|Z@0UlC}4_ZC*7E8@2O$v#4n?w7C!(qHzOFViiK7#2^t9+t&nrG=Kkp0nJZ
zc(0z_lONid(TRZt@f{)Qk<|H({K5LhUV{2gz3AIM?j7ub-{~^033lgH`ZU8NwJh)b
zy*RhY+8*%gEKC~)3f~rAj5F*XcOp#8E(JhAfP9+2PElGFAi3C++O@%gZhhSv_s5rX
z>)#FLammTTOwS;rW^TG|DWP@g-@&uN`{m-xP5)VMFBos%q)Gi6y=DWkC~uz$7}rOn
zMK00yhdIV?l63@JqpUi<;cZ0&j1Q7Nbuf$fSa|PuM*+l1&F2&>J_$0`#Mbf71#_Py
zpx7U}lmR27q#y5{pK0XEgx`*xg=JdT<e84P%c7u~Pe56qo@1DZ@c2D6-PI7`<4_V7
zHPpceP^+4)UQD2h=ZwvJTlnDMvrA^?0-vmT9PW-MOpdpNQ@SeYjJzL$Z4xkDsJC-D
zd+2k0MEf*pE%NJi&PT0g+0Mv9k48XNp-83J$R6`YQqKTcdwuW~H+G_ioS8Wm0P*Gn
zK~=VKmbqowfLuVjA}10d35g^z{#{uql1?-GIhX@9D)O_3kgE2>v0eeqOAn7yyB&w?
z;4|6As&PgfJPKMup3itqvE<AwCZP=Y#jsMAoo85)CC1?_PZT5*TmK0`hdzl$P?hjs
zBj*YU@#o#Bz^$mZ!~0G;*20PoAY{BrxrB|$7F~;NRolc4@-fsPyajPn&fj?UY{(|x
zgk7fS(*I#8Q*Gx(Saq*tFs_uiU$FQ3^W-$tw1Nn^WIsHQ_h+DY60)3{SbE^mQwccD
zo=(s6;?7jcD-{Yu{*j7+!0TJ9UhAhnA6G33)p=P=#~6+W;<-x8W0#cfxw!d9F#I82
zR}ISkks@z5y`@@p#^;9T+ob*K{8l5zA-Gt|LRU;Q5kIm@=6|Z4Oc#0N1_SvQqoh4Q
zt0FHOw8k^*mtO{mRLvNYoh=byBHA0?Y58Dhga3kgs_D(jRKAH`PZ_^^C@3miR`ovV
zy9?w`v=Tr&F>4<*<W>Z))=t<Ma;Bf%_6E~b_yi}X(<sV>1{XH<2G5Gc1pEj0#GFgm
zeor*F17q5=HHPWr`H?*N4M85Kw6sy>@bX6b@$nz&t~$oLGVi(QJwp=Vsho-q)bU4w
zMmpS3-NsxC$5_8#j_t>oAk6vFi4@N|)z`B@>>SgjI>ovdx3?BNwNplo7FLJAQcW$#
zzAv~-`b`HEwKNDy<1VUtk?e@(Q+nGN=7zb8|Ac7=>XV|GsR}9xK9d$THGbo<_=SyV
z4(?>rO^=o)spQ0-L9|!ok*A#{3O8!fZg1utj3s>Wy4a@;I;G8C>590qPE^xpHn&4(
z7=<*X2uy_LrNt>Hsu2((psiAcs|Mrtbh_~p(dX3<r=pN=yy7jDIeQmT$Z|;$$dzGl
z-kfbtpRNN(mGauwJKq~e$HwTKU#{!6cS!@*H?UB|JDWiI+DY-S!k74o?A-ivO*w2_
zBpgkSMCOeU{T#D6PhMS+Sp<#;IlP)qb#u0!fhyC1&sy&0Bp3<f8nyRc#OG}cHPdg|
zZ|u88U-|ahFNBh2JNYcWevF$^SB0u3nUu6643xNpIJOyHiJ+ipI#TRCs)s**oKOC+
z^k>gL%P@d;qX_)k_n1lSU<tAYsah|k>V}iI;)m{YFd7E>)n?FBCcU;6%UaIurSpf)
zhxXl;HGa=iz6wY|&wB0+^)NI8wQbc=4UZ)exA_(4&TWFrm{0QOhD%SM8K-Sg-;VCg
zRCs&T7itk<kgW}_DN+($^>yPWG<BIc8Rs#du>E&?7x=b!__SaxBFh^urUO)F=NCRt
zA)Uh|H;`mNLs+GXOHd?B+go4f_h&V8cbN=;23Y7ZLcVty9IT)6+r<mG{`~cTKFiCd
zg!!mr&X%E2za0)(s%7s&JwmsynL^bR3<ERB`>~xk&d2I|EfKV`*uO)1Dwh5|;%fEH
z`~@6pYa67qctG85;~)3@gQm!T@7$5Kqg^y<ONKiX<23N~dRR4OwsN*j0}r}I#>OIT
zr}Ozijvm=`lT92VqjElYj{T`Au>0(70gN1IvI=sT@Y!+#ID2CN@)}N;CHw4l#FG3F
zcLJ%}=Ts`0ZK$j^?brL#`KT=NmYT+xKGBWKs_#9%OWbp=@AEg6R>!CsD_k-~DNL#>
z4jkId7>zU0o{AS<zkYLU2n~R-G_TB6Ptj1(L6ED%C7w_q7_*gCBi;p>*+HnD7wf+u
z|MrwG{eoU;{kH~5$rE-by3w?`!jn$(deYD4Uu>8PMi&xQG2vM8hC0M-7#wcQocyO4
z`CXRJ_E4!q`^t#QXv#Fk&1-06nPgL{6LkUGdDu(Fqbd0i1i20cCvz9XYu)BD%-XMB
zYg4@2FJF3ZPcC*(+ubs)O`&o{NfPOn{huq%t^;nbtK5s1fayG;CYCvW{;d?*&jXs@
zB;dZ9d_(tI9~ytfzOqBlMMhG{$bd0P?~|S>$Uj_HSLUn_KpF^EnM1F$!lS87Ybvp6
z%@XOH99V}d^m`1%{@fYMmcyLO+i1P~5h|a_j_*g_xN>GFv6Cq>tJd}nEpX`(7YMy_
zl^G=W%#jta$_RY6I<)2wJmv#ieSjo>yGAgh%r>rW>|T)G5)5izu#5AoQ88}}Fq;b9
z8`w6cRJK6IKHuB=a#HH-_z5F7xi@yTK^M$&?g=Zug%p<%KVP|IlWx7il+R<63QY`E
z8`w9Dw4wdxGNcW1;xnjeB{Vut7r^MBG7aH%S{a$ygVG#jbwgd6LR8s#mcv5d3@vTS
zU@^k){=c@S3DwM#LSh}=!e(m07cU{|@Uhk4f@ISgCXha`M3q5w43}IKd3X`&<7L3b
zlTzT0H8jQJJaVuJU-Hk-mI+^1*9asRQJ>?qUTUwEJV2Y&u24_)3y&+x>9;s0{EA|{
z5v#0VgVz1Z_|N*Di?DPZzpJUN7bc94XqRW3kLahHO;N`0XPZ_|J(&`&{98U>jPF|N
zwx=F?@(l|2e~CPsE3q({R8!$ba%3c^Nf;#r=<{;$0D~_nOGBNi`-$Qdqs>FOEtFuD
zV8|_{w$coG>G1&yKiNS$J{+;WxYGjTI%z14@F>3&qdqqH%+tKxD8J(ui1<HRx*zpk
zlSN3=^G^Nl6E5(#h~T7s4_DrN_wiqmcKDg;chh45MUA6+b|E3Pipc(nMM9lfT~}OF
z%D`{3>z}_ekRYo8suixnrOHG1D-u+?_epl=J7+?~k_jIetO*qC_}^aH_2NHTA1XIA
zH_9~VxHsO^<U;Kt<KYh&!lGqbWw~ujr3gfK;k1Q9(|?3sjxH{&9lm0h?u~6zP72-d
zRWULSo$<(s71c@9*2&O`hc5s<PN`_||3DmD*9Xv-jTMDSDL6SIxtyk_Ql6GzK_=Dm
zZI^op>9#podilfAm9ei9;VL@xY)e~>^o{pF@ZiU-9W2u7{9p8w=d5ogc@lsRxIwRe
zy|T2j?3IGHx<yUn<XH<zo@R4=o)WCHwbxX)9a(T_0s%)qc~L>v<MoAAl!YOcvKd;R
zDi^iK4!@d^*Dhs?_+!(Xh(5y;goXgyf2o$L?cE@F0o8J+rha?G5-|aeCQGqYs+i_~
zXan3NpOK{L(hx7HoVp{R)ZFRtF*^12LgaDqPw>A?f_9UvtGAP<JrST((<(L7fouml
z(JzW|I*=yW#XcP_#Wi)|hF`@>hX;<_)5)=mdCkp4mrR_ym;;MSLFg%>qazbY4nw(p
z@}TpgEx+5(^XBCJr1fdf`uyVaZjdy9-20#XpFXHL(4EC)>3UB^j$EJ_t~ST^T`@HA
zJF9l7vp0aSEv8D!+#&R$Gc?$qTvvjoK=%g!E)Aq}5{_LT=s&-pXc<*>l14_Ri0$Rn
zMknHDHSsC%sP~QJeg811ATj88Ug&B;9!*WI#VK2Cgjm%si|hP)prJxGv4-upNI2^a
z(-Ow~aTkzcmnD7yc55|2KrvbD+vJnNrayDKd4;ZkYr!1>^O4ezIR;?*{;(EN^%o&o
z-~18ve&2oNEsp1LdDJ{U&$PE4I>vpJ2m{e9jFmP^#{cMBm&u9#;lmNs9r;Z1cz`sj
z*~_)w2l8=w?(>eDPc$@As;0??NVS$X(;uNp-N^!>2<{Qd3~FY!7`>8Vb>+|7k1J7p
ze&;~>*sArwHO07ud)T9h&OjC_?&^&*LURsRz)3XbiJQ9~WvQ+RJpq{#0#WofXzkPX
zxmxTn1t%vb8b-V-5rLJ9%QwoXLUT*7LN$*6ndu#Zq+^a<#|De-r|t}^bb(ewWS(3`
zpGv*2aq@|NC9o~eDCAsL%@QT;AQlvrZ1vYU)KRv5fiMoT6hvyA7*6%auo7HEX;o~8
zoKVG$BzLWvCYh!536RUdo;r!oF>Kn!yv-89=gsj0#}`@kkPCXKe@h0S=y`T@wsZ;%
zN%TK8&-C*#;!)u+nICgBLCcpy-&EIev;`kx9igm^M77_K*k@(0Ozthvsj^UL1M+@I
z+Rt&hF5*pe$Ua645y6gF`Sid+PyulGI|WqHI1cpD;841l-;K67B2SbelP~WFVO<wN
z?;jVT6t)-jJYc{3_3-_+FnOubKv=o69Teb`+{NKuQ&z4HEW*@@DH@oNUeSrBQ8X|8
zda$kbSf61K9_x4u#89SBOCzXhSPjPfdw0_z8Ot=ft_|1@asT5X0|D=a6W0;`WxXQp
zeegTFINAL``s|{Ge!4yUQF-6{W$n*^reEEDzjEQhigiL$^EKQ?>Ez!&tDb|CbEqeG
zJywF1DE#!4dCGlbW<XKdbmAtnTCFL<^fm)C-)cNK{*_R(w(ggpA$US24U^84EWNsb
zPkh}`upyk5137g6ce3@bLwG6PWV3Jsf(FM4nqsrQ&fX_t9Ye<^w%!4`vJj=uShs0Z
z*8yBPWxlB*FEgV)@0)?42EtT@iDQR1&>R9AkZbbyU$U@(02%Wv{U2_xlN=CIq5hS2
z%N>oIhr&m?^mL9zyKKMno-oB+n&i3fcL>Vuv9kG+CLATEDe(D2*`^I^58!M8a)e5}
zz1Uscn13f`C`ifRLe{i5Ayx~!U9>v?sockfiuaIfxd=7_<3zMho$w+sO~S52k4uf(
z4=AP5kx%JrhC4>6Uqfx@J+3gOhcmVH<KKIvD!9$FBkDiifkRCM%36@6Wprh#sHbG-
z)>_**VB;a0tpCzws7PwcX8C&Np@vTTQDk5n`a7VaEX{K@IBLH8S9AZ<l0wsqbJ=?%
z4{m+0KVy$WXq(sPb%#(i0{pk`mg$ji^aWD^!)U_BG)L-*?`ax4xp_<}>die_cyW74
z-_z}m_?#jx27s_JG&M_O3#VnlaKn8<zFeTWvgx8KIfBwBPS1g%u*<T(Dzt3YhmwZf
zwIBb`V7|r^_{19+R#ABN`7lWSbgeUy^!4psf(1p_NI<!%9kgysW~W&lyf{P=LBX8^
z3!y|2>S6=!7<`Ijmr&i)yU&(EG-w3a0F`SB@k8!r@k07!ct4pd$fbm~tnMeHYG1k)
z^DK{?*@baqMhIU#EjgF<)+LHO&WoH(nf81m4=+v2wQ|2Cv$fCbeL11b6Ap;kf5sA!
zjaOsqPsk@G%s(xu!?`{WGmd@^7_=t!zWnO^+)I=xbzCja>Qc%)1uTZ4D(UoWy6V$C
zeh^7u)0IRClQN4S6WvCZeOSk?8AkY8MxMq6Ul5h_rJ#=QYM_Z}ve2?VD5qhT-E4%<
zef~~CS<&_W7C%c!P@bq_&&}x5AOc7{&(6&8QX{4gzQ#$;5*&?*2Y)C7F3+{fpa*ow
z%JH)6US=@<8Wq@{r6n!faKafyuJc{YKmFanUKQBBR56*Rt~KUxFU`MS*#I!l6RV%n
zs2b_$1(RtLSa|zmlm9#QiuS0&As3Tn&ApqIF>$cTJE`r=*04_?6$GVKS!L+cG_fZW
z5)Ms8S$f5v(6X*?ITN+^YFwlSxxvf1>XSoi8bwvb9GDp!`jl?xdHI!CjQeyZUY|Js
zUZY$+aTMX=3Mq=CN;wM-wdh~;L-1TOSP;Xb(s%?{P~%*eK7slgtP8iqgPsHxoC&B<
zzAl_-Mn;Z|(Z^q@8Tl-WAwG-8h=DBckvnRLh<Ql&UN&!pUi%XN0d7#cpQ?h;J*Q8%
z^PS&_XlUJlSNe%F!h-TP=!URW1$~sGI`3rdWTM_Exn-r^J-Ngl1sQOR4&@{dmqHlH
zm5|RzO<R@&wD}a+(eqaAOt#vFUbZ}y;L>O@n3wz01M8rj(WecOSIW2NLDKTIGsCXl
zg)9QnLu)8r%pxJK=<+W}Pe%{{BifdTw=Nas)b2YH!_ogj#i&afMWv*bYC?oOuzn+}
zesKtEilV@QhDQ7K%NkEI4Dp*iay&9Jrjy(|397FA3(YpkhUWmPX%!nkcjYle7|2Qm
zGrbFTPI`_Rvvabxf?kj7W*xvR6VG>-@A1jo?1*2HRMB8ACwH5;j!`vMN9oYR!@;&x
zSfhz7*N=m=-h;VN)+yotq~DV_$Vy`A<MVWV-q(u_4P@p0iv(Gg$J=9sh#uTH0NuHp
zMyHeW>Vm-p$g4VWu2@<qvVLXbb2m|h8!b4H*ggo<;MBls_{62ptW`A;ZaS1PrO)qu
zX;=LzPuL?So|w13jZeATh*nX3j?b_ak$nV{mfnH_c8kK}2RS=jzHZp>$-ITyVH1lW
z8r$)vs6rY5Fpc(GvENFUt@(j-x$3y)Qe0BDPYsTAwI*x7rAjfbl$_d!KmvB?j>yBO
zX(y_i3O5l(LqAcUS`=$;S6(TyEP~vv84unP^{LAF(J>_0b`|~|8BYqFYDj!F*kpO3
z-ZrP$n4)a2KSZqOZUVY1GOUn>W%&n9eF}g`&&}kWWdIcI)H=9x$`4dF&0zeY6qU!(
z+$^aS_w}->C+BsG4McD4K~7bWPH>{891hkw8OtBOKIKeo>R2Va3^=gT=JQ^JEssKy
zaTu5LNnqw&`*jG1_>466aEG<;-%-PVZwO9&2)uZ>E)jX#?Z$cT#Z*_He|?4g<K<CM
zXmUC5na1_kiWBwu!sr@jO@2Y|uPpi<b1o$ZzS4!kr+vvzoo?5kIAGI@iuHv=&m0r<
z9UZ!F%yIdLmR_kXzB!jz+UNsnsch76yW8H4&)0}Rxqp;0n<W}d*N&Ty@uzD|i4XrS
z-ly7EjJ94fQA#LFUMH2`@I9v;Dmpr#X=ReTRfeKNN^to`bbfZXio3#iJ^s!IhOaf+
zVbrrZbZE<vs}Vgae1@+eKs8pj3Wb}a3g0y?!v3&sLEx=A9fsbSz?o;NnQhuHMSbKj
zp90@w*@Q2`llR%|JdS^+Xa3T(IEl!xY(`;*YHzq<sHPoi2k!D_8J})!brWpS7izBu
zutK9pKF`tf3n!ncLZ$4ZGx#9^&OHfluXI5-_{yn6$oeg4*Oty*QJ^4^vszO-yByxG
zx4YLr7w>QKzP0x$*2fxJ6rI19#8G5a!&7k_x|}eiGtZ+WwuB{3;)bAdy*w@bVb3Le
z=XrbX_uo-zDc@KU1CB4~&@oIRw#SmKe+#?85Tavw8LlxUa1-+?QI?e(7*U)H#3s4?
zb7Cf+h2gnz=-m2LP#)>tO_>MwkI<=eLEb*Fi({v9ot%dV|N8xx&J^>ioVO=(JGnm&
zhN$9n%8HrADWZ1qH5?PLPaY@1@au61ae!^eCY7F@RclJX=FDm@j=h<)^V7cU$Vhqn
zeGaj0t6&~5Fv~!xy`?oNJLY~s^kUV2L&Rye=XXPZm7vJ;2a$1$vPp5$H<<X^?<t~4
zgD%2m{%cka6_WN(Bw<pVTzn(H%5KmkM!I?Z69X<DQKVYV;OYN1>&YBJ%`=^&)s~*%
zYS6K4w32zs2*WTaIbYDrZAx>w-tSNex*ajaoHlV+R;+cy%-uc9pg|OKA4a^!KeYeK
zaoue+Kku!h?ofO4f%xR2=*ycAG_P~v2YF};IN{0o1erVHYT%(h)Mcpj>`5oMRwJwC
z#BXhT@W6yuqQ*F>;VC0IQnx$1p;|g`Tw;A}sxaPKh_0|{J!#hS_&)$uL8`uT`F^H0
zZ6%2<AN{+J61P^l@rJ#W`UVMegN%%CV{Cjg`N9CNxa)3q@7m4({9nJxXFmJc-tqod
zA__yk@7vzP&DY;ZyV;;po~637L=@%-f`FAqjYhkL5`uafQ}O~Lp)l4G_4lEb!Xz5s
z_XvUjSo0KJYqQ!+kS4R%x)aHOOOmh{;!cA{ANUY)YlZ%yG3wPNI-M3$2AEEl+4*@M
zdGJA!PRiaLgA{@e_3~k!temB9XcL>a?x)l@!rgb=#rXIHKmK2SnsYtNlgo-iIL_wr
z8yPF@<AF~<ft5ZOL%lgmwR@U_2X}JQ4F_4AZ!kGsLJ7e)f6E&f9UG*R8j#@mKCigv
z6+HgL6Fm9klb3s!{>8W?Qs-urGEcRO1VBo8NUJb@p0PbUn0xjx>4_zz^+32v9m*N>
z1{P{3PE%c6q&U6_A5D7vak@ty<>?mq0|Pw!+zEDWzZ#n=mKV$9`z&|daW~b~3LpIN
zBE`NUm2wq<LV1cfbt+h6oQ2AowZS@xlM<`Vy4d0W6faqlB&NTwkA|_tNy4L#Kf=`1
z20s3YPxEDOdOcy7Re+2euy^(?A|rHwP)OzSYq5}G85|rY%;kvV1fi`P(pPfbpjc)K
z*|o}=wU&0LL6US4GKWAhF)__`*WE~?UZv5flEeupv2MJ2apiCA_pY^JU(BRV2*Dk<
z-oX!j-w!g<U*ySO{RpiXmQst9ik-s+w$BW+GbvMAs9~GS<oojktvcJT9iqAOX3m|i
z(8>?6_ttA^H7gvQTVnRiF);lsS5}#ypT+Y%cJJQB)z=(gWPHfkn6x_FbnSj#aocsw
z%<SOY**TDTw(mShWu=0)hFmT{2+P9!9Gz}TE*G(~(xNbQn(0mZc*j?K1t(9P<U@b=
zcRe-SCE(3(d^4|q^=oO=R;bsPN#Z7c8dC$>7Nm9goPiXG^{SiA47ycj3~tEsXx2YN
zYpE_|`N!HOAYr9E%cBo`h*GJ*>gpo(`YHp1)9BVXCr%utytu%Ii9vF{As=WWuSH}R
zD9=_{4M*6%>n3Kl&+yYf@e};TU;P!o{ky-*%IfOnTE3Sw6!5Ow@8I=!-9x+6U}a^V
z^5QIk?-BTlRL3;S%a9lv+7L=bUnY{U(%|Qm6M<<>u~a~LzAG#kH<`V5E*9%I^_ki|
z8xfLHk)~Zvz3>Rl>KRIXL-dy-;<yjNAOrnl^yPh;)r3RW?I0h9jE@$;Y62yh7%x#8
z^q4>WD0ahDj7{z2>T3@2YrpUd{O6zgDIR;`iOaaUS0;RwW32BgM*8*;$8A3Qi6@Yu
zhcKF2W0uwKDU=8}eXPYt-v1P5PL}a~_?GW^BO5o2VYNY6N#H@E1^IlQx4-S}96NT5
zX7i<~hP^~mm*RN=zLx{5GK&a-6dnjrVMsJI%JklyoSdydEU-e8SO{<@UEe^r-k^T+
zEdEv3(!Y9^=HZ7KYg9NA=V&)1z9%{M!ZU2&v7P1F1x5ppsp%PZ?%IbzQT~JfMX^+(
zvRp+cn$#qY)LC4M%{mjwiz&^;l`-IpF2wG}UZdkK6Jx_HE-h27)p+8m#~3IM@r5ru
z$xSyL%tQ_XBLJfvxdW`cU_6ZVoLR6n6pAI%B<19>lbA-E@#!fF!-JVQu~S{?Ed60E
zFSgOFFgJgOzS1~;=o1718#Zj^+UsxP@N-|F+f5LGOF^&~wSLzsb{CrwzgS6~=Xre1
z+rEZxefKxx2}`5A$k`)DsNu6`%Lv=1B4$QCN<n~XE)l8{1UcFZ3;5IH<Z=PI$Y*SN
z7=fhqg}>y34;<!&;uI<vM9DlK`NRh~J2#6qmchOOUUk>2_=b0X9UHc6B(<7-yLU4=
zz5zd%=fvrG9)IjnZolnLZoK)GeCm@QCLi@7g>-A2CY(L9fYyS!bH^B)+=1_TeD8OC
z7w6~ZdMERLCid>x%X_~4yGh~>D=Tw!noZJfN}389sl^u_9TU^1LU$eZJQwwaL1sC}
zZXPxBYRp7O?ncdpwy#@s3xhFTjz0eY&BiQaqr2!U6$pKgp`jr{rP(kx$m?Ex1HOd*
zzC0r%qa;a6|6q~fkue&L4y)zo8J@a|d@j#Bzw#?NxPL$Y^%s7DXP$emN9Hd90=OZ{
zal`Ns2tnFOC@(HxttAXYtkEQKLc3b0m1;DWK@}hcc-DYO@%l<6R#7OGP(jWSdb?&3
z<$Q{q$~{v<i#4UYGac3ljy(Mk&pz>Kd|#3aBSJsGbYoI22!;gZ#WVO)v1xjYp`ih~
z?G_Ug6G){{e#pR3AETq=lvid+lX0TLFq<}Q;=laFPx5pB<LCMKr#{ttU;I13^E__7
z=~nK%;dP`<!Dm14Q5LIh{BVFcZc%N|(~1}HL_kgrVLBd9KYE`2LB-d-`!#Icx`9L+
zY!-1UaYZNLdF<J}o7cbgwfxy%{Ke(o&3_T(|NTcUK*R~23h_PPRlhE(hDqs$m64Lt
z)=doW+=7)^czR}*>)|BjzK$s$dzSFj<A~=z$^dn?jTM=eP3ntf_V3unv18BBX*F0a
zH)yWZsjrr)RLb0Q&#QUkeQ#se&a0T5ns8OFj2O9!S@to$=pwMDdRpI>FH)dp`<7L_
z^7$N_Hccah<@A}8R8|&w;)$oJR8}+7U$-w8a$U)yi4+Ef70%^c;Gwn46CW8IqS)tV
zF*>albMxoXI>DIC<aW*aL>ro|CZ|uIVt!$sg@wh;R9G-Mv5}#{QBtkZI{hLA?Ak2d
z#R}bv6E9v&C>4wR;P?Ij-}z17N+2w8r^VvY!;FoW_}aJZ<L$56%WbzDU|@0xo5q-0
z6VvLTQiH^UG{v;LuJYN9(XAHq$By&NVx40rXUXOI7@ktR`72(-D_?aRyLN4*)oJjT
zAN&jc+kgLg&YgIHcKH;(6hvVG-^+9Lp1o|Dp1|`J8#eD^X3Gxd&z+*x?$S3ngq1GB
zt<$cvv~ZHR+afh7`CN`4dhho$H92{?m-R)DzEU6G|2^N&U|%1V@*FGWC6Xk@S03He
z39eG>Y=A<k5LP0E%FMhqQ54~M9!jdr4r$FaGkauJ8E3%lo_qG>b1cmsB_BoP!hljP
zN7{{1*3hUf(x_EXSSF?>k;+F1!T8h^xk8B~O|d%V?D5Ard-AC)M_I6c&t87<=l(12
z_=>L}ilWQBsxKYv4RYM+7bp%6lh~Anxl^>;4GM(<0?-zOHq>SpNwsBA`9wk>g&|NH
z>sx~Uew6Zw3Vp~hhx7t>&Rh4S^Rh>l<0`VOS0PcJqE=nx;~)EL=4R(eQ%kv0=KlNd
z=lLV2$Ok^f0QyTM78Xv^tk=*fM7bhy98>D+qck{(=LNLeEp)fdb5DMfB<Ug%OpJ~3
z<3IGnyy^9C_@V@h9=`P0ylFE(^<zK9Fa6hF;q|Y-ovXJguGYu6S)5{T=UIlD&oId>
zhxTma>fL*o93CbL6$3*aU;D0CuxHN}j5Z`mOcHm{X%{ef(v6q-O7LZGcr7zqw_NUB
z{TJbqDs{=?0&<`q<taCKda+cTZ1CTL$>-T})3qF#Tf(liGnFX>0=fc1X&7prBRKsC
zk%%##V#nrDy7d-`H^AQO57KPax&PCj;Evm0!R)E?D5*&0Dj)mUr?~yLJE*Q!Iehr(
zb%NutW<Ro)-|Cc4Ys2yD1GJWZlJIonF1cKY#j<hZ6y@bIM~^+v$lw@HJ@qWN-F5?3
zWMp<tD<alB=hn*n&Mfokr=K8sEFtg%d_SaEEOFKN2<6H$k3RZ&?!NnOa=DN~p@>wH
zPN&1lN`*Al$fBaZzYm0<TCJjWN~ayu?P@xmE}n3TT&*=pl3X-P=X|-6Bq2%C%y#C*
zgXytxe)zpV$f0Ym$0Qw^)m2nwnJtk=ZsRniV2SGdQRdnyeWQcqh6<Q+)8*<K<E$-&
zq+4AjKQcnnXwh!BI5A&mZ1;XTOKqk$4zqD)1NVRK0hSiablNQfPhoV*V~;<~@u&Ck
zircQCz0%?2^N%w)ID!`iY~DJ9PP^193vAoAkIz5$8BU%)!^GqS&mTQPyIv=sD>CG*
z9coAHrza=)!SDNC{^O7SSWh_d{{blk?|$dkbIYNdsMl6lS)M29b|~hHByopWJByOo
zq>T8H%X5|%OM<ouo>ByUh>&jhU5M;+4C`VcArx7nGDtzIvBLcy{{U96GB`BIi4(^;
zKfB22&<Ldf!hDfxwL)#BMi_)lPER0|%YQB}FA_yLy6q;NcAFQT`W)k<TZxK;7-J|E
zi~NW0_)f07`WpVnul@?>XJ;?-3ct9q&5L+-*hhb$FxpUEU8c6WLOvfMg+!w8gds`l
zG|E*<!Xqan%1Uyc1`<r*qxwtq^$*}@0tu&bPe2%ia751GB6k7doC2(X(<hJ8s;?4v
zW5S?Gvo(*DKDXa`8wYmp;_$OifS;p~&y#dJtge)q-Y|{weT*@5YE@c|2A+b|)jE~M
zB^GCouxaacXp<5K0q_0p?`F%UP5jz#{btYd<V%cPF30O${aW7hZQsq%zyK?iS!(44
z&OGorymm}cSYq2`n}khQZRf^!y#pB)X*AoMJ-@)<U>`eoZA2>PRjHH2aROzK%B2ZO
z3~8EB%16BYEpOt#{nD?!<h<)HiPRmt|L?hGe3ZZo3B15nzl3!aCo2&m1nJr%{Tntg
zb?6%AKJfslcCAp)f(&8zbu-*_<8~r1#Y55RbjgQ#rb>!Utuoz}DpwuY!KuTa;i+dI
zW6$p0oP7Q;)7y8Dqz%rTInCxxTiCsOH*@o4OvhCOT?y5o{J<FvYeQ-*R;WxMk-5+R
z6NJthfw7j=TAlt<ACsfw1j1*jGEX<{aQMh6uD$MRqFjVUV=M?O$ry)I!)6c9^Vzv;
z4^l|tPKTA%8fhwMbz)w4;jpt^X|yO6``1MdtyY7_pL~@0g?T)$z|7X&luG^RZ1&C=
zOQ&69a`Z+b-&v!q&Cj{nK0%sl;;tc13`RSF%8LavLCBB3>zlahnj48bHQGz(8AuXJ
z%{rlOQW`IitW=PGKv4+tc^`?u<O8%xvDU*1gK2ai)}-B-q}?K3t#Q_rn5)z%<_G8-
z?Bm?*5)VK05Qm>XN~`U1gSSi$@}2K`GuK?TnS8Fq>TH?fz%X$)A@Y2dm*-flRA{xj
zOifRd%a8Ku&ptr^aDmymRWORH_C++CT~yZY6ItZgq3f>Zo8R>=e(BeLy*EhzPvGVQ
zSMxP*ejA-mlhw*Rjrt0?Tn>b$+is#$P0Jc`$|FP|twm;Diz$*)u^-_@?xQ9&;KV|%
zEWb8ant7Bt!$D4;I>M<FC-FT&w_~ZStTHt@&FfzCI^x<2Cr_Us@*}$4HgTsxt68H^
z7-KF|xjuJx9^Y4lVV;$h3eP_E8TMayFA|BdZrtI{+iqdk&h7m7U;1V4f8c@MvHvG9
zDt%rP3^3qXI<`yOcBoWl@k3YrmdN$F$mVjH&eAeDDNsV-3qv3iY6*zZAx3xXBwr|H
z=X1bnv``>C$l{KqxX4R1yNO?)s}?>hwS>SEbi10t!7=W*<u<On=34rTeaxRbO}V^6
zwATK&37vL>Vrh)?=g;B$K67(R<a0T~C`YH=;l%S#Fg3Mjebgri0>0{PZ=+Di^Q-TB
zAFI`uPyE#7f&g}J+s^lW@BhP%hi;<Xta9emb97owx=Tx}KJg4~YpG_=7X7|r_f6MP
z92!KZfFKvpH&`Ibg&=SR53Sd|JFo_UL6EwFkBKoxaQ!t0x#QNG`1q&pzuZf3Nw_3Z
zclzV^Gda11(zRD10#~V12DCNnj_wG7Hkvf_nYex*$<jQvCyro6!oa3JZn*7QHt(6l
ziWsyZH3nZP(jZ`9aFD&NRUUrk2z$T$b&PL$hQXl$&Mho4IX1|dBZt|&Z99G_Sgus5
zH(I1=LYDsGPWqWxz@+Jgxihk_XKmWGirN|k;Q5LzTQ^h8_c1X(Lc396*Up_BKY5&P
zw~6j(%1dP?H*9e2Rg+?*abys%sI>@6q>>biMSP|32L>1$8zu1bR94E&&!4BSRHD;v
zk&E&O0cn!5yu3`Q*w6mGSEE$G%+_6Oo!LXD)4}sR+MPDbi?ev5y<VMMJ88KSm?lj%
zsm^k1vwgqZP{6f*p4U%IuzA~d+Gn39S~|?AUm)-05e0!*t)Wd7D-B_PiL|;(yRt$w
z&_`}^2-Au22lAwGm!w=l2uWwPj*S!M7i*kP^DM3`v2*u0+Pdu9v**qLnnLK|Ny+zo
z)7!Z7mg}%_is{Dm4~(OfPo*)-;_?}~X@~7QcJbMVA7l083F6dJsdVY2hM~bhlxJD2
z)Vb!iefWMzoWyPh1A!5iFMrd0oH}!c_kZw%mwU-yR2YzwFW<h6P<gc49m-2f>rVek
znm}sk#_o=}q&%ds_`+bM#VQX^1%&yMQ@vV^O>{Q<?o^EzD(@z%*tyC~l3EfCT3b3D
z!<J3kc-_6PWyj1;;&zL2d68P9O@Cp4FbYsgIWyZZqP)D!*w`=&3kyhP34P0{<Bzgw
z<|;}9<8Fpa1J*D)Jj_r2$Pe-U4}OT>`kmjUR<B>~C4bRiNP65C4KnNttkIwZZQUhG
z67o?2Yh3b>l9JkLh5GCqwBoFK2RT0_(UvxXZP#AI(Bup#1zvzP+BHd}zdlQhbXD86
zTUn8L5-6pRQj!Zj@=Bq6!OmUVx%tp7Y?$1L@0Uol;MyDR<f+FV<m~Ku@<D-tfdRr?
z&b>EFC6*T!866qo<jIrt_xF*+UCy6-o`ty+432DJ-Qr6KzU&RJXLMwkpZwXMqg<(6
zu0!yWM@|TCA0Flh{-6KE$W;fZES+O<{tRiViTr@d(}!7JuG7&@pqP`2iOC_x_U%S`
z9*IscHYEsL4tTedpiP2v$xvB@pnKo=LDq)B`MmnR$5*}etvvP2^DMoDyz4Hhl%rM}
zJpDI+P35_#h%oMhYO-mTHFcV&5EkNY7jK}SZLhwI!sa3NU$=$Vyya##Y%5`O8=(bm
z=*(i%)LEPK^|5<mfbqsGPkiitcJ0_owKB)@;yD`44uOa`{lZbwPMtJq)89A9@L(UI
zuTTP%x;XVgU1*19X$*EP5b`3$C^+-q8*e<s-FM%`K>rwngM&ml-&tO4+d(1n(MiJW
z<DbE=&mpD8W^{Occ1^H;SLjB33`x?Z)o##j)u^p5@!T_yySc38dCr|bgVr(Dx>E45
zW5;OJ8nhctw?8srb!C?O@BbKo^EZFWpZxJ}^XX6i9nyKQe$i{MwO`j1v}@nH+(L;+
z2wt7<<4uKr`bLKl%SY*d=7Wr_9w%Bkg-R-@mD8llCkeU@1_t}cmx>fdMj<gY7Z;I%
zhsx(LafkZ(SywU-+y^jCVp?g+(fMVPIKlTkX6F`}ot?+?0)!E67*-gfoHH+u{32nH
zC-Nh_pg^b5;>@{|6bm^DVZ`>CO{^}@v$#^G(Q&TxQpF@`LjS-N2d};zgtO34%EOqH
zH0j{`lJEM~Z{_w|ZoaH*{$hdx?u_#E8I2)fr94j@cL>8=HUKM0x*gI)(<glbA@BqQ
zUV>H@8~T(=eQuUcDHjhV++g$C&5Tki3uJc}cV|NEGcq;_#zO#Gw{GFS*Wbs+4O<a1
zN0cArwmV+OwwdktN|F!zh=PJMEY_A>zKD>5=?xobHkvdWEt0s&?D^*q^1>LOJBMh3
zK=HOWe;GgjGe6DV-McUAn!i|(7lKzsB}SyeqLD(9T1BFAfQ#Ug(kF`YH0xDXDof}_
z3r|a=guFB;49!$gnjB~6y|2S76%bzF+!nk55k#oaJ%T904<n@KBYh8NY8c>0Is7n(
z@<S%ahS|S&8#mv0Ew6j^>)5(!3#U$=W^R5NFDOvz8|UU*Ud^r@dk6y`Pvr1qfH9UN
z(S%VRfK8h=lcZe=g#c@s96$C1#&j?wXq%FzF-eke%Z)eki$C|XY~8&1@*a$rASAAB
z|IWfNZ`yMe1DiLqICqZoXO4r_<f4GMQRURbkCA9g*J=vNXHa>J9o$1SIEXeHiy_KI
z_<oqp;;!8cx~%B+8O+QtHVQ)ug%X8AiF~fe%$BWu-B*3(1(p3v0GG6-F)8)4r+NOv
zAE7gQ4xOeXiO#YYv%6*`@qli#jsW^MPI23Pw{Ym@-T1ym3PalJ5TMZ3kaWA~Bz0MR
zVSq6Sx6TYwoIQbQC#<f_q0$Bm3v;whN`Jn{vrjyZ!f@}aUdh3$uV#FFh{#t&nKWKV
zOTH9Q>?<ODdC_d0BjQqml4?C^zz=*p4~`!{iSLC3fzR;p004XT?8EbY(nQlZ@+`H#
z|4WkR9>H(c@U*~K7bPQwKq%>E?S#zA_Xev|tVvm3oaOZCW3<{0R#)r1aP)c3&7NUp
zb%jnGbL53%eDYKG^E<!y`#ksTQ{)1{;`|9_&mE;+Um#6u7~95Jy>6!hk~B3WNlKc!
znL49g7VO0Y?{Z*3D!!sH$n{={Y6=4d2w!-JfxwYoNf%Q&gI+z4NW17x$F)d|0VN6Y
zIfRhtSR;Lp_R1<!yY`xHcd*8h^we0CjqtLWMuFk;kK9jnwN5*$Tna3I^8UYNp}fMv
z;yl&}>dj@&yzmsWbLYrKd2)p!&32Qx+rpxG=+PHQbP6IxYfGtbl&}8!_fRShAdo2O
zG9`UKAP6F`FgVc955M=lY@eCAZ0r7F!qs7sgI)pI)-<ZCR4dEmb0Ja)lnn5MN4r)>
zB`G;6kXDdaE;S|bJo-0o%m|j}VxW)+6`)jz5FSGLNaZ_n=XrR(PY{NLVUCHh5%%xg
z%ryu0@cP%jp6Q8co_Xe39)9#O+Hpc@V1hgEehd5d93*MCDbFp?thZ5~V$;kvT5-a1
zrOZ-!iItTKe&A7FK9AJ~iCUlCRZ=>!lP9=p|6YFKXMcvTdfVFwgYa@M|7C-Z;J#>p
ztujZ>6XcY@T1iv-M12Lq$V2%7g+hr|v(Cc89Hm^Iv|Po*;A8MTLn9SvX&KzVhvL{I
zN_Z&WM|u)J2=M$6<wr<Azz7c=E3ENBg{UAT$QKEs0z!Ep1=Hgr+<Wh9c>TTiv3u(t
zM#m=^o8H9x|K|NX`sjl+8&xJZ@8Oyo?qY0Yg6i@jt4kGhVhDnW;fYCVtriALtybsc
z$<tUX>9iKHqzEZ9vhUWxn3NscXZY#={9{~m@WACgC@&%O3&rcgLEfGpU_&Swx_Ung
zl?9d-&k{u;zUPs|F>{YS#q#VDZDSZzJ|*c9^yQg4bR8OjwuUeaU7<&c^&7w-2#KPI
zC=Bt0OCVA50#676catcyBGIevzKiRwK6ts8<)4j9YQGvRfw44>9_M5+&%{0Vpauh?
zwK)c*)+<(Bqw$gs!>uZ#Lq1^^qh(T12AZoY<c9j_#t9NgJ6%E{@khsKpE}Fyw)XKi
zmGf+x+RUSmJw<<Eg-)l=w#_@3nw;jy6VI?bx5UkdZl%1m$i(<0%S+3oI-!4Xkip?$
zq=0kh&vEj^d8~mT4DdXGP7PK&>ATjJB<XVg{1P`@e;aW}Qz(SQsz9d!<KvT@JbsGQ
zq#WzExn|`o-G`TH9e<9(E%zXYHX@COzL?2p)ii4i)}-h(<;1CD)S9cDI&+e(Tdv|0
zpZX+!^e2DJmMt^v-Mg1oJEmH5M7UmWA&~gWXJmMU)EYd`2W!#RT!<ml8m%p9qOpcd
zm~Adp@GjGg&qft+N7zqZ`^=`64P!nBHsv{S@*EQreS}_s>9jH3E`Bb=cDq=h+wLF%
zk!Yl~_(6cxnzY+Rw>t<W(Z)sfSZ!%ltCR|P+U*rOM$wHEW{*A37ryW~t#*tOlF(OZ
z1J52l!G}NoAT#4d(3*TckIh^5UUv{<63(7E$<p)Ba`eOrKKIB|SUSwil!&`s{IHML
z-1lzAC$=J$%Lvt}SidQd!ov$CVGyu&^A>*aKl}&2?}vZ*ik_z#mp-@oc>>R}QeURB
zd=4u$o*!H=hP8&~av4t>r0`I};>(nJEGTT7V0_zNtnVXKfSaXI?haeoY`)G3qTH-3
zctPaM(QyR()vtXW#VBIe);)Mph?Y6t|CfJ3K+3_ZuVG+hlIw4}mtvuhPkiD7G`e+4
zBjcnf8jS{r4?oXpxyIPY0Nr-X{Mj=Y-Ny3@r0IofodD9b3qmnC*w1%=+qZMWb=UIW
ze(9GucmB!}QFy>zVUcZKNS~h~m7<-(iWamYfnO{jkc2@*J8rYIIEw=O(47-JAd;Fy
z8;q66C}8IH+sK!S2rmEu_0=k&AF?pNOb|va&MvdOu)s>SLa|h&ufN31%nVbT#t8!-
zttDxiK#*f<Vv3Q$VWOyn7lmxuEP2f<?ncL5y6q<Ic7vgjjof<MtNHlH{+uHxp2G(F
zhzfm-7mHNuH5M0}%r7jG&*h17A&o|jq!W`b7O^75N@uBsFdz&wTQ>7kKk_5|?Em-$
zKKtN<m%W4XVuc4B^z+;h6d6>4P#APB#E$jTY_5>cg(&G`jU}nCa`M3^u&HI(3kU^B
zn=-I-l3;uYZ8Z6!tE@VIS+Wj`izJm2jMi?(UTcg1X$(eNeBVPULATp=ZXTr)@BZ4a
z<+;O0UJ_!sOH7n-{%Bqd<;M@>4^J?C%e4?Vbx@ilDC_)!l#Ur{RhZNbLaAw6gAxKy
zDNJhURvQGKM-T-hNgI=D@FhY?3WI%|Ia}t2!6Lflv$S!B&e9x}lV@11tTJ+DlwCV^
zvv7Wib0^Mn?Y@H)h6Y(&SmK5oZ>HEk$Y(zDS)O?O^Hi&KJl`W!9@879=<6$z#BOe*
z)9G^T*a-@SBCojnwM1b*jYgTizKEtsX{^?;8oJGdZo7#rfQdB`4eb0gsL!XUL$9GT
zwiyo(k!dhA&MZ|*i4l@|y~gb9S<2-_I_(Z=YB+h~G-uD9qf)8TY&OwayV*En*3%)d
z&dx#GEVaU{dD}UW#M;kYoF1`eYjQafB?Pd~%W;j@hctpzT1E#nH|-xnB`v%(X5sib
zrngSv`98XB(d`)g0EDJfS#^1iCUqA{&&O&@d3K(_82re`=maT4x{W507gA6@Mr&%d
zCd&(DmO2&C&MY+aeOw9%w7V(q|KKP2malySpZUV)C<b{(Muu6clxcO^sGvyS#0;Gm
zPLt~!U~sITR7-BX;})*J;U1<pZY4>&SgY_<=96YMN#aHa!YGGS5@S-XJ$Qg`dgs^j
zi@*LG#4j=B)RzJwf>(wE3@L@yF>$9!v%X4L8bD-oR8k05mljCsO%xsqK|y74enPSF
z&>?a|BPcIGD20#`aI>S<XiRK~+X)6kkn@P5h%gF~nNOop5nDEGCGr%xs7O@kW8;L+
zm*00E)s<zWwCJQs7>4ZKe*@*^MLzrRKQOzvgbEa`W`|R!=Bd>*#s&y~6+b6ki%<GU
z<&=$5N|LlgnkFdaD#yMb@rt|dWZ&-H{P$n}6+ZTfPh1ffa_2=CadVWX=(!eGEw(gL
zk8+|Jn(QYx*oPE~Zl}%i!Yr*?je)^_@?n5zw+LiRC?riIkQm0UIl%ORYw&VKP(Jm_
z66eo6&$-icguxi|i;H~W(T8~Ug=bi;*U0A!j13KPVDCW=U4K2d-ElMhLjy?PquOfW
z8BOSiM1_75ZRjubabWL0R#z4&4Gtou!6Z!vhsN2z|0YhIJ<jP9$MB1TR9BZcbLuQ}
z^NTd=37e)hQIsduT?{coYD}8V)k%@H`YcviMn{JDsUP`a{^xJ~Hh=b4fBB+wCb*ms
zz@+DKPf%j37Z56m5Qao6<mRo2fnfsWt#?Q|ohD}<dW7WsGDFIZZTKp|1d@?M*C28w
zip2s-Ih!zzAvJovbLHl7onjiHGpjO#u>zf4!<AC*A$`E*LBI9QU&inL(I3Akt->zG
zrB&)omh`FwVb^f(fiKWEHG;Qm7g%S@VYR_il6+jJzu7@a7xCf?i^@f$#z5f8osH!R
zg^3YYO3#HfYIRC}M4&>>oT;)qP~03TPD(>z?+u(i@)UFB^USxaH0rAi4o?zz5l=t<
zG!xU)yym_)aOjrXG1h0>mi;`mb3aF3cm{1OrBXljdX2gHb1W?_B88w-$`b|=Nt|-}
z^l8S%#uynbqK#&Dc7dE<;>@WtG-^#&D@)`HphTY3NIYfnn~QWG{V>t)+hEH<QZGOl
zlaV`@OJ<Db<f#*^td=QPmf5*;H@kQ5<BmJ-=Ew_2&^qDahacv^fdia7KgZnp^CU?^
znx<51RZLbwZnfHU<1VY!Dq$F2bWEJ0_hNFlHgEKg4VV8d5d;CRJ#aM}UpPx>1c9;)
zjiro@ri>5f358<o^Z@zLW9j%w2F51{3wf09k#u9?Mw6u5LhAtErlj3AbsclIe1>K_
zVfUs1OdOLR7$j*l5Li|lZFJ<?b&It+1Ir6+-@l9JpMRcKs{_WOv(%4nYG_yLeDLo-
z&ui~GK*t!`L~Po29l7E#1B0VT&tuoMchT;&NW1R63}CDwX}7ULfep~oAcUpeX=lUU
zUY0TIA(h1U6mNRXt2uGv1n>Xbzg>@qyAmLPTf!34p2DC}LelD3V(n%urSyoRJmtkX
zk|ZV)3Qt%(EK(ZkiGaM%#Qp=WwW?Hh7Y*%Jn^wC`WueNMlP5Vhdlq92rDC3~n`hXw
ze>-Cnqu_anAfnc2QXC$@_X4a?1fI{t=oqD9NWOo7^5Pt|)iS;pv3t*ToSUEHuRiz@
zf<j0)jVTrf$me5*M)I7Wt+4mtE`q3lNmHbCbGuRqe0+T0!(_q?w+@oc8#nROKl-Cw
zwRaEi`<>sVTB}{&gY$0)Aq20z`dV(QE>mc^$i|h}(o8)nsk^ZmDD@LYIZUc)w3dmh
zWkx)h%0Z~BY~NU*n;WCkNC~QKv=7(3@)kyNV5;XibM7n;J#&P5eU)5pglc(__kZw@
zIKOnBIMu{)OtW2QsXWh-6EE=j#~$Xn=bq)wU-2d;hkZn6g}_rR&!3}NZxH$+VLp!}
zWngHSFjrvi{3&Xy6^i*jhK47&{`$N4!$19f8u1ARM+z(~)oHa-I$cYwLsn{4g2FI`
zVn0fH7?EPF%OuBIlB9zHl=8{vBi{3E-^$d~IKTFrzeT&<xx5GMMSy|+0q(o*5Ic__
zCf|0#$+m_@EU9~f1J~{)Uo5zJMhlgtd1~`>EI#xUk(KyDQ1lbpCc)pnjnTdP7#JQ$
zD1p!^*cfXy$`jd4edcLtJ=Rq8LKw8w)RyLuo}zzfjC`RFD{<#_T#V>fzx{1|_MwM3
zcKpO;UbX*!ToS44VhMl<g|UKGrONS#9_6ZuG1R~qY1&1kDJs-Vma7D&MJ)1kYc*`A
zjaMpB7#l(=NgBD?sLo=U{MaaAzQF3-5(7g;q_o^}{YG@RL(w#O#|EE|oIlOP*4@-k
z8ETCR?PiPe(gJ?559Q~XJ3GrCzVA1A;Nu@>?={ylJu`!}F&ieQ=^q;5+Uu@IDaFZ?
zCz+d@qgX6(e)cS5V-r08^y5g;BJ^W?1<iVcr=NU=8xP&d`S}X9rCDTqk!`t2bQ&WQ
zP{jhq3c^;E_=%5U(-^kh0H*DxLbL{*(5lxtapnZ`OLL42jBxwScQZV?f#Km{9{Iw<
zoI7=jc5{^*4_!mE(PZxYdD7I<>cq@0%+Zazq*~MM#xxo&jvqV0@Zbn4@Gugm5==Am
z<+UosMNwh@9vB)J;9KAME~e`ZjvYA*LNY#;FgY&Bg+)BycY~_A9JxXcKMI*Yc8b#2
zF#Y2r)E1U8X&1C>#T%<xZFM+TPAK;Ev3uhn0?3Vwy12JQBMW(=!G4a<F3>g!-EKl$
zS?1Q8zlmIFfRB9W{mjoUp<!LV4uVIXImVgUWj0Tb@dH2b7B+6#OPsp7&@@h*v369z
z1|ezErQL305{ToJAXHeJqLb7SW@9qTTW3n`l$RiUh39*G&6mG{<0p>r#Pi4h-HyfO
z#&$2{YCn%J1fdipMiCo9;7Np16bdC8%{rCJGX13zO>OY7gwl{`x1NQuQHHl}A;=dH
zQnIpej#_1r<45K&*5|>8?&qOLAD~=a!2+SDm>8Sj>U{^f=kB|>>DKFrS5H&ShtyZf
z)RxM4Qqey&j1ij9kEoTGdExj`;;yFA>acCc4#p-oQ7jEHw_K*LKO~Ji3=Bn7Ypb-|
z3EQ^raq~RdVleAD(o*_ZYf#Ds$a|hg;Cl$+vQFRe<~Or_%T|8*eedI$!-p^P5&E~m
z_17HYKYaZ+@Z2x|S2}4-BQ|u5qMT|{4C7u%Pz({qBDysO%4gYFn`L~Yz)<2`47Oaq
z5urAM3NXT@MF@Wu^@Ts<#L;v7>F1s!chz;Aohx%-|0HvZ#~2(A@e7l5;xzk=OO_Uv
zsn+V8ICqN05BxdRT7`GL?KZYf4H0<(=TDrV+i6qqBXl}|l@{gaSe!e{$<wDvQ%jPZ
zW6R78VPSwMKg#j*C#kn%JU?J&%Vz4$GP%6u?D=JGz3XPed<m0iq_SA+1UyD3&PNfj
zMw5~fh7n)))nCOx{{a8}mwx$*9nnxqar2>@`R;Fd55Z#(@%%GS(}<y=CDjz#I%VVT
z8OC<+a&7~0_x~mDzoR9)uJc~>x7OOJa@EPP`*iL`H_&M0K!RWZ2olUe5=GLIC{j+6
zB|SMvwyvcs`}#Ua_O*0vD_gRnsbErKCP-uup%FUg?sNL&Sh;py>;AE;0m`!d9<F78
z<eOt)^uPcH22QPgYR@(2_x(QBmlwep!i)2WMh{m)!HVgnnsj7{n?CXphPUrVSgMa&
zixI*gWfoUz&{?y6%|Y1*U3PnQo_p%cq+x(j0>>?|ZRR?*?K(uUP$g4z2r88lfBRFP
z;vf9upZS(%5WdQoFaZ{%(hv)SNCm5B=a@cw4)@LqLgz793}P4-30|C#5+M7l!rtl{
zItnOHj9}+n(w0Nm>SQnMY8A(IiPD68sf1-KlF(ygb(I~;=Z9{X;PV&O>6QkO!egzm
zMjW?EqL5xUVmO~;W@wnDb7#16?i9A0qcl3s=+q>R4d>4s2LvlCD{P<HNu^R_cxaG;
zYK4iB>dTpu8!IauJ9d(ZkzIt1gid3bwJRsMz6#CRF7eD1o@HZ6kVyg<(q4zH$3IKS
zE?_2hB1{OW!K4YTW}VHoRcvALmN(wZ#PA@I4%l2<;>yJ{EX>XEu6I7jo;`bb;)!Pf
zNYWJBa+sdp!NrRg*w|dB-RWjGL9H_wq9wm6(c+ct*Kbakcr8{wQV725J@4Vj!F`<l
zy+5IqXodz85}`SMHN|yO@^**4`v$ZBH*HXb9D^eRtjsN9IW~#T*4vISSjwW83Rbof
z_U@ddRLUc53)^#1wnflrVicrtK&4j3tFNJg9<F0!2R%-{_&n1y`w>bJrfFYG(Fc{1
zX%Uw<>h!{tpciK&cAYt-#Wh%|0MMC}(zg52a{`^lnVus_`dY5O`6fFAQmsjnh|QHN
zoOu58EM9qrn`hRTy((yQzw@VcWg&T+Tc)ZEo-GJc*i68(1wty6w1|_0&9!Aj8e_XQ
zt}C%bf+HZ%Af;yeowrdLox-vt_00|DE?wZ%OV2Ykc>_;8@hDF`_b`4Kp-qgCf*_8$
zboDZGi*vki;zizf;eI~&_Ja(J3=ws6guM=V+a>M=SZ<Ce30RuHO06<Pu{Ovb{O3Pr
zwbN$X*dW`sjj*)Z#P6jH4LcZP$mfexs}*j&<DP5<q!ekIfD!$byUw;4{X`7gb}(eW
zi$a2J+Z?(65dZwA{~^Eiho9!pzw#BrZ{e51tE~fj_Ve>U_b(A=j?-y(@#C1#*lfiq
z+6ad2Jb4GURdP%$oMS+xDANNw!j1%KJ4NPQu=~>!lcMZClvt*O-GHs-dAg3v((D|z
zGe{5yjBhJ(*IT}i>fi{ME?ws6(dSs0U!>b;Vv2cuKOpi0M7&1PgJP+OYvt*60!H`j
zBI^1?e)hg;)HhjPT4r)`1}uj^{G&f*z1Lyy&TZ7HWh9a`mK5?9_2w#$qgdbEB#sRS
z_T7jjEn>2%&NVnc#=`R?MjMhO!%PVzN?3f?``*vg_ym9dAOGWTD_q83MNN)R@~Q9t
z6z_TGgTx!_eB<#)Xoo4S#G;X;C?Oc|Tz1`l7|Zr(ZLZST%*;xqLXm}&XYojJm7$Y@
zu9Ud>hd#;Vky}wJ%L68jpr7z0l|+(hcwbq;LXsvdTsg`4(~r<<tT8&ioyDbjgmlQ|
z3Uqp1wl*6~O-`|E-;LO|%aOy^^VT=r!xz8$(ChgT{txR_YU(6Pf{?lo?NdbUh`A?^
zF}!UXa&ihKG~+sApjJSqE{>5HD~IcqNPM3#jOexk^5qiJ^|0fFpuUAt7R5?Ao5To#
zoiC#7l$;;Ysc%s1Zt=0}$9VE;ov&Vk-1tsrH&$u1H;GeCGxXWmSYoVGq?Rv}SP5H=
z4s(~Uax7P-R;w~FP@_~T(cIc1U(Qi1<mmZ5l=Nt|+H~3-dcA;=u^l)`N^|}+b4wQ~
zD~o7dV=9s~3NnGVNfAjxn#B06Cf$uS;-~+N(FZ?CqMU4UqZ7hjhsMS_ci-_Qc5mB(
ziG3Cq7CC+VIkwj3SzW%0N$$WB@@3(5x7(r9?ILjKc74vDJIBS#S2%LxO}y#uH-LaJ
z2wnlSd{boDH?Lt|i?yFNhR=NA^GuZUtS>BLNe$hIi**}qr%YAh%-jaT2=-hzgX1|_
zRFi8{t`=EeSRs`OL9an2SD@7mX{C84rYCT`62`GnIUAHkn#5THD<!c`NqwJ!Wiw!F
zHmwpH?Ix$5eun(O2vbwLSzg(|?*-W!{F^41uOKk8FI>}y$ttP(0%t+OD{iSYO-Yjk
zP#J<IO)y&HdATfT-YMYaYiw;UaPFlqvoik@YqL+&?FIDOaM$&_dHm!u?M~-^`FXrv
zrGU42B}Obo(T1*ucBE*=DU&%5<=7aVQeR)D*Jx5MxD4eKqHK}hUqcu{oCt&mH|`%|
zaQ<o5XJ`5Jmma0K{{|WjpL}kMDBa}1bvwwHia4%|-wT<aUEu1%5^)@}w6s8d?E;Q<
z5HFWUB$Byvm)W&{Kb>x!qH^eT+jLtU_D^lcE*5z3f%|y+=rb%X%n@tJMk}VOEP^n=
za`H49b?$k~o5>dkP*Ngog)xdWNwOph!7CRjG`7jiAWB`kAlPV~Ff=^G-}wHIF*7mD
zfBesXz-IH?fBiN!ImOTYZ~ugW!5Z_;I%$%ig`$}n0$|8>m>luA>Ch;<rwWMf8jg~P
zVgbA|VS5XIeVxj#8H8n#cET*#qi;frbxdn@jZVFdum<TieGVL)Bwv)=bjLmTLCB+D
zf1Iy9{0JKx>$F-u(nMohf<n&b@c!+5?8Enw%PEwwNTQfRsf^!A7#f{qVSW~kV0C!~
zX$&KSHDV*U_l<XR{OoBiU%7<k3LH02tJB5HMO5-G`C<txOu7B`yEu5mt;m-F*fNWu
z71`e#lbLxe>zcDt;n)_&z%4i4$iMoz|DFHyU;i7<UA*}E9^`9Kp6Bt7w>`iQf9h{D
zSgR8H9j-q47^`zDY{iNo)eL$LV*`06rfSR-CC<i0!mSmiJ)2T_l+DcrmUNr0N-{Q`
zV{_g2e2DFL-i>mcjGzGMR3Z)7C^R~omiC7p*HlU8jz7<z|Ixo^e7wlm=mfSDlnO;E
z)e*`$m#3e8nA4}OaK~+XXl%`M@bH_dR7d&2zxHvCojk?T^78BXasGGfRW)@M8dnMo
zGULl>L#z$mg*ulXd79ntdMkse&v2uTD`FzY!7|sTAU4ufxQ;>MkoXDGafmF1?YQ)|
z>NtZHjLK3NT-QZQs0>u;HP%T|fg1$8X?ulzLnVIy`O8EFkD=N=)*GvAbsAjqW9I!j
z1MLQ*u9J!3bH!|z?kO58A?wRagmFx&1*udhQiQUxEEg9+G1WM&ReGB%EUd5KD~kzb
zlQT_<xe{p-l1fSJ``DHujv|as31gqa`W4j19O=j|GJ$?VF6Xg#&n^aQ6)s*nPdDiH
zNA3wTGuxP%*@h4)=gyuX3?dMUD|7Qa^UQNhOl-$>JZiN89D9-F<t66k=O~rRbb39u
zHa7{w5GDJON%U2o{r)9NlE31|<-c7k8=HLk&%VH0{gmAU5k)t`2!oOqHyoOwTq&b0
zi)Ot=xl+K*<q)n5wxqjWr`-z>BE=?0r8-2!ZcfHKS)I2TE#;V!m)SGfz_IB^Icby>
z1Ytlf?=n^{GTYK@8YpdU;dgrcwU2*-p5Nm1snc|Oue>}x$HI0T@}5Vb;AQCmfD$HK
z{pt*)t(3~3)cwZp+B8)3E%2$-NM%u~j!>x$F?;DWkA2}630rjz-~JBbu*=fXXIX99
z-1pvhvHOM&|KT_PBk^l=Ca-nfS}k*n?U0uioj9cxNj8%dBcNbs$eFn3leSivj#k-I
zDpG3Ik-PJlnVV3yg|J<u03|ojmzP;SdWzMp78hPR#^}gyV%_EK559>VyY_Pa;#I!%
zrO$Ka$}FqPn^?l&<{Wl!8{tFmeiN1@Fli=QpWLwvjLXRIB*#xa&&tvwg@TJ?D;%j9
z8X92Vt{IkA=D9Sxf=)G_Ye<t6qYcu^amyWV%N*xAMqqFpNrb?VlIoZ^4*C-TMUqBH
zsj!slqY|>9@YHBl7LJokkMr)^1z&$rFu(o^J@ebGO2rcY<fs1$*X`ejU*BMJVSyx;
zv=T)xNf>h!2WC8u>>s3}45qz>k&-lDz?MGJc5uBCVXH;j^RY`slCF<eDUl`#X`)G!
zgs9tPBQo4@>rJ$tdy%|rk;_?hx-m~a`4lHlo?>Nbg?1<C=YVIT_nngi{KVh<Fk^!Q
zNFy+5N*cset7FurXYd!7@f??<FFi{dMNDs>!m}(aX)``H!t(MgJ-@~BT8FDkZ3<-p
z(vYMn&3X%I<+=Ixw_)2kq!d{FIKpcQCqkH)r7*IOKe8+<Gf9|?zHn&&etzz!ev*Ik
z^FPn2GiSeTpN0QX2_e|Gdk;VMw|<P<Z@LweCd8pncXNXaPd-6pTkM;tGCelPa5-Yg
zl@v+^WN!&a7Kl}d)IPaLFp^8z@jbUPd$G>>v-4Em@K*MH;Dd-li74>#J8hJ7K}l>U
z2UdzP8gz`&8m0P|g(T)1Pkx<tt4rRq+1xybZFy{;nxv9<Y1db{ZtphsY#$|`S18lu
z%GoEG-hDeG!xMb$yS|%W_`iSYTQWuF)irfe<4Q$GL)wpJF+w9#SUq`>!O3BEj1<su
zo5%u^gd~I5WrAm|iF*O2-Nkqoh2b)hWg)E;rDGD`2iHdEgrFDW7Rn^rP#zqmx7EOr
z5^YkZitynZ$5Cgl^0oOorNSVil^XSKi+a1odK|GD2e=|2-)f>QNpZzxVt9<{-MiVU
z``FULb!}{A6GaBUxsGYA;&;2ub$zaeA)&IlvEor|_ZS-~k?*@NEvrZz1n8WXWv?a)
zw$_v^hj?`fF|s?mp&3P~Sfy5}aPs(3q~(CJSl`&-#OYJC+AVCy=D_}gR4N0QG(ky&
z=UR|N^x9jvwq)O~J#3%amSwxf3H5rNt!9hoo_mgq7cajYQF9HiaP806vQA#Tin4Qj
zVz0mjf}Uhx%BEZ@V!K(ST+(jQN+N`4l7?LxttKkxQQwG2kc><Xvt(}LQMNO*ZJI=+
zoLTl+b*5-69^=-@E2zXEbjAcs(-5Hzw@p+zzOh1Ft<VWNoP7Ry#;2yZ^_IKXSl?i7
zZh`TMQG&py;JNgA0Xt`QFf=sWzg_F>cBeH-ngS`x%9=WZCQVh}J;f``BOFS#aXP(_
zOIIG|*dxD%h*SLTCRZ*z&!zL1SzH%P?!B2^`wlXbYOc=D^Vu(b`CHyS^=fP1j$M4;
zp@W1+pQRTjbYf_w&?9ABxga!_-=-Y4DK{1<nl||`!c`uLGGv+#jqS)j&^#gR^f0|P
zJwKwG^vG{3Ffdf&(DgU-)KgFLna_NV%d>N|+kU?hHw;(u{LshW$>jJjN&m8?3{;9E
z<OWAryLyR|SEN(QV_T9aiLeDky)H%==9iaQUR`Hs*k)*`NW0x3j3p1e>my7|Z9|)s
zz>iT<^>sGclsAqegfOW7+CE8=%pr{+(HbowjYEz-_a|I9^%-Kng^l3)DVyCp4|4R>
z1zO)`Q`|zKz|a2F&vM(5o9VV2=r|(iwP_kjU^Lh79Ov$v_cAu>qT)L7R)f?JkkTSa
zz_L{KUO<v74`$rS&?gRhxaA@;)dbxx8c2hXdRoTIyKJm9*u5*oavh4L3Ntf1dGV#^
zP|8p&*hFE1FbYdb-hKZYx#7@_Y;CR+_aZFef(6xb5j#<s(NR`TE;3lH(dyKZQh+5P
zNwI`s+w>I2PQS#N^H-5FrB=yhCK>6`Z2H{s#`~GxegJK<Nswjr*`aaTpWvETCb&jp
z_t}L)WS0%4B$jP4Gd;z>`00PdzxsFoo=2bh#&=X>mv>#h_x&H@qaXPg<x+(-@`-#O
zzp+KUzR2|Uao%&=5M@i@ST%5CItz28%{C%R!Lso@m-f~&%CoR?IXu_m&@J0Jc*_il
zW#gax63=bMtjavi`X<G~6j5S`!hmwA%=TT=j86`u?2IcZjUm`vq_w(;Z95c8Rc5BQ
zarlNCFj0?%r3G5e4f4e@1A{dxr2;}oYJ&sxI<v&Z0p9uccktA6&+^1mPrcqB@n5P}
zxqi)Bg;X0flGK1kASKv6jXr;#T6qt)ZHR&pl?b#1w~#|9g=Hy3s`0x$l+4y>w&&q@
zI#^0#;-qgc&XOw9I7VoLoAZcWmnco?Y=vyCH%SrPH(sTP;jy_D)>4UWd+Zz?LpnB%
zUWct#i&if{Bnft^X>P64KG7kL5{3{|q$F|_S9?8{+5vu=5~hY!2`Wl*y<<`9b-8{_
zas60I(yL=trx8NnC>tH8#7RQf?O~@9#}b&WRZQFhJdD<~>RTLr`Wsxia0<)LW9KR;
zJCBJig09cX(lQ07!1zQ$(CdQMNFj*gkd^gCqA(_kVxl;~4??=#4vl7ujb?*Zt4)%m
zFQ3a_uGjYK;bh|Td`oec<1sPmGBiA#HAZPlabN(~bLea|Xw;hs5fk`6$Cnc}x`H<!
z9K+3TXKUiE*sXOuKO%B-+<bSDr@pp^s$9>+<_c5Q0Nn>X$4Nw^(_+u)F!u~K`C3mQ
zJb^>Pl}i^maOe=9`}5Cm`)zkJGBSit61Fy%n4e$dhJ%Nx)yi0wL@I|QPKd&YRHs-<
zBCq90+Sz0}F%ZR3=Khv0ZhjCyGW^<a{3=&Zeu=|7O9&+hlZ0oU_$n)_KJLIS-u~W?
z;N@~8VaR(PxSy5vHJ(2D!fSdi|0}CpD)Ybn{eO)6<inhgG~Gnd(~?#aGiZAha*Dl0
zi<v7YsaYwa;E?zZ;l?JBk`%X1f`O>*qg)pq#t5wm8%;XBfZOi8i=*c+<5=K2ij|cO
z=I57j91DrTbrfj=mXz$C9_Q{m?!>VzRu`AEmw4VGX!mGdxs28^>l^DV%r0^8@Bw;$
zK)1KVrP(W-yL5qPo;iVQ!}L^@dZWeA=ot6i_dVQk*IOWoGbUteFjgk?z6Ku8K;sz~
zG6NH5&0d;n5|iL}nmqgHf8xriFCfA$aRA+JN-wl{=Y#k0w)@}7fBd~a;Nq1l-xdwO
zWm$ac`#!}3Z@ZsfyOpiZ!-%c5H7pz6ddoC7@0p-fC}WpgQfbpRDW=&)gqk#svPqSt
zNV`6slY?B2q}{@VG15R3$Cx-lYC{qmI!3UyvQD8?BuRrT7dkZz50A0AwMDn<BM{h9
z_1{mLX3NL*@<=5~Q-hKYPTr#)tkap_AP5s$?K;O^IL7XMyYTV_I_)mqZkM@*tIRE3
z<>_aS)APGbP8S&*aA`K%bh{}BZn%wiy!*QlN|GcIMrZnBDJ4>A3<*gZ5yv5-?~+b+
zgfc4oevt~TGInd4Skwk;{MZkDl1eep=N|ex>9@IKUIIHTo9}<qn|a`4pG4*yf=&a!
zu}&^d7;UwYItI%S)a#hIi{HBdPM*Q>3IqvhKZ?=B2}(H#AoZh+E@7wH%G=-M<Wo=c
zU*<xr(J7`U4{-6q3q1DpBdl#~;Cguu>^;DD0AN9%zP<m$yy>lP!1WyBFvQkfZol~;
z<CBwYo8FFNc~q-aHrM94eDMOp7?LRL=lpt92Zt$?O4N*nAFop$zMjAFsZVj@)G3<Z
zVrG|D)zn>Mpcyhv-L=285TGPGCTb`Zk#sufUPJ~eNJ-<ET&aj~Y^3XC@qI-XqcZAq
z6vs&0LD65C2-(LS^wax<pja-E=#)4L>G~0cS{2Vtc*9hl2|MA*g&voCKFy6TiIU_C
z1@>&4LO2$&A7ZB|I!bB!AxWeObV6Af{7A6ai|81GB?LJ~QP44W)?AJZ<QXd@6zzaC
z4k^`!a2+?(WGM@wHBt(Coer+$AcRIV7f4#Or1>dG1vyu;d&dqgoPLR|`WlXxFg`KB
z@W4>UIg3Khojc39vlnRiF_9L`FR$^ZpZg+7|HY7@WXvm(WPhp5D}CMT_+5}1`1C@`
z^j?c8JCCCcX}3$z2$2ZvLXP6_Ao+m`i)YUvw1wS^xPM=j@sS#rgJISqn|+7I$cG7!
zKfBDxj$!V3@DQ`7)>xRWF<ICIVG$!}wi;Qii|26LT}QZhZk_qn2I1@_7L-dE1>AYt
z&C~`bk!U)d6{H>@=M*?{{q+=b4oHcT7DiaeG{qz)TW6c>GtYQa*&0@Bn_OW4w_M}7
z=TEY4&ooM;^cr;<tprGD)Vp;3l#%U6c<=ZA4F;-Jk|+itDHQU2_&x7nd1ZyUg>Qc-
ztx}3l{k2bW_|Rcqdicx4X+nfUJ<${_i-{qRyKfp}`&a=Lx6sDI4g{WEAkrz_`Ud$R
zz!=a$h)OkS91}$mNfeTr0=Y_+R=q{g?E#WXb&vxG4)WMz4-v;9mK4}lmQJ(PY}0B7
zjE~haaHtXFyb^w|ODo)9V{MCKInNv4{$|SMB8&5L%q?DJz20W_$}F4pHpOZVDLwY@
zyMcE+_#oqxGx*(3wn|kF(z5%esw6W@kQrtK89GNw^>Rb!dM=HvEslTVPf<!TGI50E
z#g_=`4caY8OpeOH5Ve6J{=tv^9scMupW)#ro+OICb=ihO2tM)gPw<J4em`Naixh?+
zj@VqB#jTWi%e!yo#_1gBn4rGM%3_19Mw`i<V-&}0giBi}BqoX>=b^PG>UME*f;fyZ
zNs37lI;}Q2sR)7q9jCZOkJG15f#nb<F}@%0%yTa>IWdOg=1|JY%xOscI3n*V9{<Ks
zzVYmf?3x*4d1Z;2iJf>}4k;yF-J#xUASK-Uj<-@Om+^ZouFRfibz_UwwRN6(;Uz9#
zUZz&D86T;zw6f0d=n#jle<SaH??)+?%a}AKj*|ZR)dDhejMkbc$|ks$G9X2k*{w82
zsh1O&vJ@gkuiIo}`6aHOYO}WA=DBkb$+zBS@BgQ&NXc8>5;qhJ+<3=bgxxw~@hVQI
z!H`p;<Q9l@7ndG|ox^lC0#rbTJ1XSxaxQ`0!x9!kO42aGD^)VK>{gxp&=A@Wh|-kx
zg(Y5?Z_(5R##9wl%HX)<(Ba*jIGM7tw!t@^dxE9qd6GEbo$q}BsRRx&_rLW4#>OV;
z1U_H=>ciZ2+igrv?P31v6{4jD+@g!^D3Uax*{o;gCIg%Gh}!r;4js6TkA3%d^Y4H4
z*Z5WvQM{_AE^i4C3ZauM77K%gL~4sd6-LUEWUGNm;_T)kv&CTI`^12BJ%ndr=d27Q
z+kaV-YQi|g^IRknY1{O|4rv-8jY46FO^WSUjBT4>aB`GbYm7AXdOa-ZFzB_YljEzG
z*0|V=F^vxE^%h#$7|X(PUF<@E{NOOQ@(?0rO$T&>7)R78^?IaHf-MXW?j7R(aU0$7
ziNYQlNwH97uv{bO=CTlAon|i$+aZj)xP=0CBuK0P|NM(^Z~&1ODdlr)-?5jKw#C|J
z8`s^UYp&pVdE%r?EHt4u#8E_I05qwFP7uDjll&HWrUC=Mc(KJ?6aLnlM<@^0h`T+)
zMjcCNx}6SozR2+QX?ka8xqg?+#P}e_7`lE+H)?V7j%g;w2D!53bMb7zyFXAvX@_vB
zh&EuPMWRD0<tj-K5VW>%uRF+*!5wV;*_Q}0_*XBIB$~qL0Dfm3DGejF2}(|vQelwm
z5ADlt!$KmY0x1y6!X#;y6-=D1vZWw3DZ;iH9NLDRE79w;5jy14pZ;CG_PIY{_fSBy
zkzf#vO&{RS`#-_<y*JX>+9VD9OqeSK+6c<UGVgoG+xe|O{9~Hm=Hc&m-+nhA{hp7K
z#$CcNBsDheBr{Fy9CWzzaFsDPMc22`LS~5%6_BpRmM*zs4V}c8B*}7+Q%&6S5k@1T
zm;_0^*~0O1q$VN?0$QyOwSf^54M}3K04)Sztl4O^_-Ft6m-(rm`0MndkfSd>$Mpve
zA+b<7m)uZ}d~S$*slvwUBJ<13q*jr=H{8JF_Un1md-BwVMyOOOREvYeIwT6iY=x#X
z)Q%D|Gt-zXDJD*nEWk`kkP0I}D2GCMh#>B9=B0<Za^?}PzxhK<O&=s*8D#$QpVH_;
zb#j{8;2@T=sFaF){6in)!0z4r-k<y_>zm(tA@D*79(e2h{NTqvK^jM(v$U7a+6pz>
zV&Ik=Sv+xq<&_p*XN}Hg18uoXjt%0v7HJ$K9YtycQ5-=UVTul}lSis7FI*u=f(XI{
zVM~@58f^4>ShWf(mo8-{Hp?Okd^*0y?+s8YmdWKj!XPFHV=x+tU~Q|#KmCROz|Z{D
zj}W4H?CD3i`T8SR%Et0sasxR$cYxa9FoEA;Wp<6F)fIY)Mk8o<L-OS^l>wV5b~$q6
zecboJ{fv*#0EV#J$r26>7?GJDGCrw(86J-@{b)InMM(CM8cC|rA{&mZEuH0sr+=UI
z<rhe!h(ptYI8+>;gXr6+v$KF(+&uT@N)%Ao!-M2<8t2*n%z%?(D+jy01BFkqV2GBN
zNLw2?B1Yvr(j-JD5om*3EfV!&QjwA<P-x<M7m;R35MjGb97bGS_fh+g;9Zz!YI2l4
zH|!)|yq;3^G;YpiZFQ3q$ImdkFvoBFr~k;*^e8jqIZUs?)Yt@-+BjCP&Co!NrTHsN
z-SH-Fy!jsL%|9n!t>U^KQ4~_&*kEn#93vy6G};kq6rh|UAA0Zmc=)l$Ieqrb>;3V+
zT(7FBb5x8rvfq4UWdvxc867M`<P#+csy~LlrZtdS5~UHMn;;T}V--juiE^9_86gB|
z5(C-IP4-=hNfcoz4<iIRNzh4(V|h6F5{@My(Rfy#f|n=nS?o4&-OOPgJHN(yGi9^h
zpc4fU`_%m|4Zn*T9bkF&3YM@LRvvD*$$m>^X%>#nn-7ezOBy<ho8Tr?iX{q#0{MKN
za;bu_v#d#-rr1ItEGvsVl3=93lNPIsOXPMnF_uS?q&#u-c^0<nM5(6R3)q^!%F7%?
zGH=6oQngdV<4vDCF0C^(GDvY^l$;#Jgb|fenKa7QnA>)513$nnm+3V-Sm7!aTk-H0
zE-=3HD)+qYAg7M5bL#o4bk-tvryChJQVNnV#?2Mc(q>_+&6$g{;FY+iB>2*5mn%Vo
z(ME%r)(+a`0Y)dMNgR-2%H8{m3=9pD7!cZ!Mlm{x(Q!;1M<k>Old<T86xeQ&+Tb+G
z_6WlstIG>4EnMQEFMO7<QWvQVz7IFu^$y<ivA>6%D|7Db36#j5W0ifoqm3canu+mo
z9(e1${Fl#v>DxAOVq$!PpZbZPLJ3W;(<C*9mX>G<_a52H-M8)|Ux-M$n}q9iRH8}4
z03A7qLXOlAag>Xc9*N(;1Tk7`{J^JRImBT^&QV;Qy@J*uahlL>wRrlO=eXg}jqKU8
zpLVlD7{v5~2q_F{l5q6cS$_Ke__zFxAN*dH7S1!jFw5?lT|{w082Q*vgCM*_qt_u<
z9_OYz?x$QEB8gLyBp^wWjE|-@I#yVgg+OFDCzDADG|9_Oy0D*4lk+^JvXGX|(C{?M
z@i=$%8O|U3JR*^tdFhK>cjLXBKK2aluAv=BZa;VgH=oBO36@siy4-rx4UCTs@cV!A
zXPh{5{#z=E;?^TK^ACUg?_mjzP7*@DOM7FL;UFXzh8StUO-UnzbUmt*Qy3j%A|IJ}
zc(oE?lH%s`L|q?YgICUDFmyLI5J`d&5bK0oArDE+V8zC9^E|TNpjY?lReB5$4ihJu
zeY<ya=G-Zkmlo)BdW3P7f9+V3)Cl4vWpR0fU-+fp;D<i(y_`Pv467@PY@6Jc1%Mk(
zE}v&<d6{OXPP-d0vi%Ud58Z~}>TuiJK8SE~fF|eWaqT=w8W2ShQ4o<Rg>8fSrs=KL
zNoKwh0wo1WN`HbY5nNlhN>UQSz~}5sU*h<4|Cyw-f{qM9kkIZ5MyezH&3oU$AN<+p
zSX*Czy^sF?YmG{qdvhgrsvL0&Ei^Zb7s#D^oUw^2!f2G#D7u)&I$_Wxsn^NnTtu#z
zP2mKzR#u1tpP{|G5&0P9+E{juu+>KQdf52_?e#4@+oHSG;*9SwyV(M5Q>^AtcAn3E
z_VYaP#FMOTY>}oJBE_>6>uZZ#I`sqxCvHO)a;z>~MJJFemN|Uge!9I5mL(aT*uluq
z2n(x=<cc|zbTITN6mvA2n>1Phjjas^hFprpB7gIfpXBfUqkouf2fhXBRW)^ql31zE
z++2P3mlOiaq_{H1as_s^j8uXoNw6IU;)Jx_Lr4RzMT*8DCGPbwfMY3gr6Tpt8jj=8
z?RK$59w|h&Hm&4QC@g7Vl7zV3##VwP29;<K23smh`3jvVz|&pcd7#GJ@)qo!2FoEY
z9A+=iu{E}nyFU8;eC3IU3DN8-l-T;>XK1fC5alX2OxJK)Ya|**#)fb_L9tk-)9zqd
z4r!WV3kM}LhnsLRms8I3h<hPDzemX{BBCC;yNRd_((U@3Jb8vBHS{sHuXJ%H`wYJm
z>#85|`K1kZUYuw5?rF-E3fd7QaYUpwe$OZFb&1<P!__L%wHeK?vbB7c4R1e>{P`Le
z&V7Yq!6Dcjrn)}M=;RVw8aA7oY&Extla%?EKyBNL<G0zhaRoJ?$WNDfwiz(r?s4kU
zS!($_dS#Z%o+<8{$T1^^GGVrmXdyD2L>gytKRUz`AcaH=LoQ#VG&luP(evA!J$-^F
zpL&$9J@gfHYmQtAqQvGc4}3rG`N)sZ>2)}L`~_@f<K;b~Fa!)*q$Ff{rP>(w@7>J}
zbJub7#bdAcQN8MNU6&vKTR+b9_ypZ{9i&Ak>|wcr`wmZW>wzJ1R+jY9ZinQD%b1pr
z=*5sEm?Vql3H<;!P$Y4JNn&(2fFLEcA{LfAM7G55wej06Jl8{qG1F5s)Ek>DEiT~q
z0{kc;O*NjaNHh?qoIZPnpZ~?*<P#tN0Gmr!SUUP7N+>+n!*Ly|Ba>Wr>pQ3oO=CGO
zNu2#RgaIY8d|_h@wq^BQiUPzn7F`Pc>{@9lENx{H3BVG9QgsxkP+)245@(+OECsg=
zuAyFE;_*j5&GP&jZQoEFoaD$YcVi2iL>RO*2&u7@&B)L&AN$bzc<kxtdFau{@q^bN
zeX@VYPJZ%7ew=(RYjS+QNoRA7YMM|sF}k@*xtynVa1UwFpt&+n68hvEn>h68Y;F(-
z0V?N^uNILxmo&(<by2%Tu$4(Egqulr&<1JQShj`oeMDk2H84u6(<M$qI(|TR_B@Y2
z@i5D)oA_Y{AGTx`B?V;9L!vo%{wlxm+keDIKlFB1<}dTYnPWK0CfCnywY)sl(e2!D
z_z<I0yRdAVLM=rnnI_I?jW7nCCZr}CSY%n;nXV6!tz1o?5t`@}3@_&pf3s9cNCzjE
zXJvJc3om_@_0_Ws4eep2w?Q0-L_XA88n0U8)?032Y;uZ!_v^pG!qT^<T*E_fvr}ZR
zl}D$7#YE#sxM_ch>n2>TE-W!KRYO`HNf?u^&7!102!n`1v}>ct7>{;7hnFi5`Ym*l
zVtLuK)?Qvg`$2}6QUcFdY%Faux7B7=RG6Q?%FM0<6iQ_()ghLa7ihH`pfyR9V3gp-
z>!$hs55JQmhxRZ~8>G3mNv@Emx3xwZ_wYQ2TzQ;!Ym;K3%GA_Oj-Pp++~5Gy+qSW?
zu|}GvG@4DcHiSWkBu+rUZ8zV<yB>IeKl$uuU+2@v>Qyy$_V75-(lUr7a}5fRsYZ#I
z;mKht6Fc!rHrfxFzjBG`u>qo9fY2#A)f8(*Vk=v(%0y#QjTSPCehLHPI3S4vv@U|B
z&{2#q0%^G<xfBs5$Rwh@GEc5t#(2fdT^1$iFo85gN<o^!U@?a@8oOX4t7Yu!emcXG
z?7nUvn^DRmk3UR0TA;9PjP2V8sip=K29&4A$>(fz;$y39pWg9f21X0CdtC-gMT|*_
zLLWEhA%&z+DA8=M5_Y;&2MlE;SzEY5p|FFEwM`aR*7}XyHJbHzW+V5_Wq>a)w76Km
z!aY|7cklAqKVD?K>f#mhgkivi#VyWUY%#SGbJKN`6!Wlc>njB99;1hDqAL^T&RnF{
zILQseSMY4fmBj@%x7H|?tCaU2qT=TWid9A%tK2wMrrT~YoVPi!x<NNJoZ5=1`<f}Q
z&D}#WgM}hi;%DqR>4M2{Fh*EtVT017MkAC(p;)C@9mfdhw(A@{`ZQmA_$xg0(AS{7
z#I~wIn>_b@;8Q&K;ZL%<y2!%9Jk^1rObDH3FPb2UiQ|+c)kINB9H*SSaEbF5zCDvB
z-uK}9c;Kz~6ZASL%ci-ugz$3Qeg6?|-&sV*0h^1bSYBFZU}Tuu$S85YaSGE25&Fnd
z4%f~hg@!o$J&)svd@+xXLTUvG%BC~h!Edx^wYKnFMKSVu_NC|9*lN%XBD4WZDx^>t
z42cFEYc5`1;Fo^&5BX~!dItmL5=juCH53PDxc%PuVc8BT*^65D*JiShp~zT;8HoB?
zB%d)cp!)#R?8UB-SptdFNFj0a74pRjXV0GI%!{9AW$p!BS;8q65Xzyxv5Frgq{88C
zZ+|aCBO~Z2yWLAE(Mq8ciIg_3m*?)=ZeePCj6eU%Lo6)4eoU;gZSkY;`v5zpcY+Y~
zIt_xA1qP!oQ&yf>XzalXGN=>w*63}l;3-XRxQ3DzT?2lzNpEWlp#)x~f?X(r29lWe
z+6E#u2u-%3X?L+4r!TWGNF6a<$Z=jNn)OvKT)ITQTx93e1Y4Wygng-s5(3Az`jc_c
z2IABZB{4^jU!ai7bN@Xz5f_`7Aba<ehj()Lwm0ME%9v}qw|>qxCfhoteQ<52t+g;#
z*6626ri;V$-(}x4y){PTIWEewFhYP-<O_oo$^)!zEO6@iFLUYm=eX?+KS-3+8JpV0
zLx29;G@6=5M{wsY*W=~$?AkHIPyXnS@b7;8H#u|R!t3@7{MRL=<ofCD9N6knN+i+<
zgfiswhM94=>)<f0%{5vLpS9I3M#hJ*l_czj2;0JhK7#BKAk|1?&~c0*MR~c59-$;*
zy-C#Ukn=oH5)+3A<>K0gD3{~(*%j<kl|nwp^7<Cqnqz)(iC!nbR*Jl9v14+Spa0n(
zCub{ij)#d-!d`^u6-Z@(G&ZefgO!bSI(|UjD=<7VOt%wo?4=iJG`F~X<qBcXr(7yA
zI5LXuDzr>7X=arAzVH1APd)Pt%PT9d@4@gYo4P%3ej8{1<kQ%3l;trBffRzuK#l7U
z-bmiIL3deRSfRDiAe|aUSPE~rMz7IANJB2~g5{u7jZV_+P^Ck0phkCXvmdXLB9uf&
z*)Nx{9qdAZUeF+4D3MpNwZ4v1^vTx-ar>zQ-OVm(Y_J`LF(KVBU?}h5hCa%+(Tf*(
z<X3;5V04C|T8YlWB64+|e9dB@R794`3=9v`*jUF>F1D?RB10T{AQf>MA%sAuDRCUV
zY+A_Y@@ytCsn!Tf;z>n!W10Nkh^=~-29ec&-OPPwH+LGC3nQ#UcXGl=p1ZKjxYOY}
zdy^f*1=OKin6^)mtgrLflWn$5)R-R5adYx8TeHv7ie0uPT?Pi*cuvCEbDQ{~&&c#P
zw%zvua)<9Cx4KGUbb>R#^mE932_?aG1mhD`TCFxEuM3V#tv19!F-N{sfO470_7JuM
zWLln1tHaEW?frF|MC<G&uZ?DFYn{g*|2kiN=u15E$P?I6o$WP=RT$=h4}BkReA@@P
zbm<JeZin&lNer2LFHTazIPI@%e0rTOt#*eeo_UsspLjA$2KhFtL)RVRNB;T`lSCm>
z7&`4H^xE8f$IWb<KEsu(>l8Xm7%dnZ8_tqfkl1+_+s+IwMj9;7Lf96ql~sr%ge^%(
zPzaRcQY_^NnjJ>+MS9IH%Zp143=9&*L6*|8v56sjk-N5)Ax4eC^z}-KHe8%r;a~sC
zr@8UqF81yiV{LPbgNOIvc?G0^IMsken}p1apE2c-DcA5*N=h6%3-wk?{{=3UloDkO
ziB8cd^5r2;oI1tN{LDW>`B%8-#wo(6M{9MRPA8)0hX~8&=DXj;gYW-twl>#Eb!NuN
z;I$%yr}hnLN=YUr#`xY3y^lwqe41xpc<D8F#}<U-yQ%~1otQ#v2-_Q&#S@Hnn^e3S
zCaEK(!Sw`Tqe;5HLeW*Id=dITTVZGj$MI;dt&%1oQO8Gld8A{Lh9OB9QOp%GWDt@x
ziII65lWM|l8%roAUFby-I!S13ZXk@s*7|0~k`!d|PPT<@S^bGx_Bkawb1WlLzVgHi
zEG}<w^9_3#8}eA*in-}cw^JP$A&RmK0!E;9#x|4RlrAICATn_IwK66fR%Bn}>Sg1D
zRFWjoXpucXN?KF~#u*$KVWYmuspr1R#bcktHZI4W{VZ>M%ZE9B;xz46LZd4Xc7=V{
z-GF7==p<oyaDX5Ep&#V;|MW9F{`51iyF+@YHoylSe2_zy&BABCfJ9(PsE))83?xhp
z*2vk4)N{CLZ;5zglO*pT@`XOeMqnJ5R7;Ysk0HReETqwp#5lzw#%Pi-B<Xf>Y#X60
zk|@HqERrO_?{!%0I5??Ap;*FpYzArroILp=Nf_fe4so2KbxN`5F+M)Q+R_3el`({d
z!J+LSC0lE=oH=`%B#oHfxeG5$c>JlSxW%(+`-%&*mzkYkpwmrABe?lUnO--<vI?2u
zM`rrT(UD<(==;B)fBLWf)$7i#dsR){=pFY^pS{B3laHffPzWT3J-a8clYl7pF*I1X
zc!A;JK^iL?)J6tz@-Eu**j!#l8$)Su2&Y`m1nSpxWtL5t7}^^RCM&fpiU;$G9zhCh
z&n1XMa+MlGj!S)gnNGS&abO6`%TXH`rq^g9gdqxhgi%1Z*G9xCmSZz+Vp?;@*<4#=
z!>f^8y1-!_FkTyHs5XQj1*A!c>$=2|&)V7+S1#4rdtfikRvp{2K|qouFQ*V(6Tx@8
zK2Ebmx!%Mpxa1uhjG-52*{$E%?c0C9q9o#1|KL|}ltryjVrFEVC!!7Bc4&%^|KGpN
z)`jzI{q{dc^}3wBu*Ceb!`>Yw%7q9WZ_(^@SZkyh46dgb-nExKANkw36Vq7b3WpEf
z#L0)gNIvMWy4YfKX^Ggll&S-iD+7!brod8I!otma7#))ufwF88DKKe*r6kQ}gU!_f
zIme?qGK!R%B#BvGUgY6NzRW{k`wCA!`3x%QF<BD~PR#I*5B>oA58uj#3#TX+@@(6_
zjWA4!`ctX6zluo`LlC8G)f;^7i(le}mtOmq^It_(%4PoU5C0ei&qc>EL9a`^zRp+}
z<GK!RLExGqUTy*%hsZQUCNb^RRUFU5DtHtI21vSow(g95f~^KQ*F#Ey2|`jpIC)ZI
z$ScjRlEvk9No#$9t@R0F4TVA;*HV~NV=0B>C``)B(RhhAS+gWyqt)T7PafluXHFmt
zw1OJ%dGJAohRbLSNs?md^Q#01A<@dFFQCg}pQOUFG)a=_qjdI9QBoj97BiPC4q~*V
zkS}om{rB;i-~D&YFD_H9*k~Qo>2xTTs@(jhkMrhteIK26onFt!%B+(8h&uXH9f`6m
zKwyl)b6slH%4>e!HiEafC3fVB;5xK7<}v4=Vl-HyXiH?!!m5lQ!vNE36E{|nIzguf
zODT-bz9$VDk#mr)P3(t&APu9esY_z=IX9yn6if8lE%KEDB84~%5L)7R1*Y>BRX4{{
zVwhi;L)m%c;sR5XlXP1hqD1$bat(b>=e1ZuZ9rhqz)R=nIC=3Z<-A8OU*X=ju5)1D
zPK?T?f~K#IxrQe(MxbQ!GGavaovW|pYhQaBGD(s|Vp&!unHDmW=M-xct3$M+7MIRH
z!^PvDXJl{}^~NHhzrjNf{Vt2MmuR+9nmxshci+b3_MHecLJN$ts8%X`{KN0VaTJd{
z`Anw$d>v|_R^y-i<WDj^GRBL)_FHr#iET@EZI2lpkSLX-R;r=}6v`#+Tn>>7S-m_*
zd1!?4@BqTHFmX&OVuDT!l!LStsZL4L809#a#IQNPh)QDel@fv9C08mSjHJD}g+wAO
zk9M?5NlLU79DVT_p8Cd9EG?~&YJ)V8cO~=7t2}e`7`42`uJL^cq3PAPkZzH!^$j-a
zs|*hfGFYo3x&dC!<-(=&1W`!0o8ov5ljB2Zt*N(qjPE=`r8bzY-Z5Azd%E8IrZ@7H
zuYa9yJp1hHayHlMRW)^HXo$TJJjlkSON3XKK`I<27%U_t@fy;xX)n)XnS^3KhY3Pj
ziz^t%BI<N8NrW(lP7sreQtUz@qp({Fp)7L60;^Y72!pVXPf?lZRA$;WC!fP_`owXF
zlPfYjHbrA&g<x$HXK(<|%VSj~ey@ep29;{0Fc_22?M6gtfao;w0*B&6iNFyY*u9Uj
zp-PrnTPkI4n=oW^b&a)^1_RYGYQwO)(WW>sflhU%!O_=#kuzYb#t?M7_#1VsxIl5N
zMv_Pp<p19`bl1uNX(~zm0WQw97^@^KS0*`g<|H@YdMEzZcjK+ia>GH*@<NkSXVw`S
zEwFRC#`w@!2KBTZyaPuVxc5EO7v|`kdVyUJeh-^hukhmU{W9+GLCVQGBbFpKDMCo{
zr4spE9+9RPWzlN4Y58q((>GF1B8)bqNd!hSH8DxE)ulXMCP`wFBxY{*5-+~=4C`yl
zoH=`*f~TlR$?k*K^Y-_CoZ-<)7UnK9JUYVA&=5KmU}AKdX3fI2xHb$ctE>FsAO0bi
zE`NKjT!j#P-}ii!L;Lol)0iM=(cV~QLV??<lNSNCeLD!7%XHTR9HA-X%e2?miJEPK
zdIRY>3>Qi$F9%6V*zKZ|6r3y*Gx0;*Vlhjq2?88xbFd5#3md=NV`X)PFiI&7c#Mq>
zv%bE8L1qzd{UDX3zXlV1r=bv_(L_m#0A{bwGCRA(;85k|xFCT*34!31Y)IMHLrEbK
z5`?soLT1`1o%Cb*43=f%<VqYrH_LVVJSw#cJ-<sF#`M|&QrM&^Shmd_Z~k81{jndT
z-RZEfaFu+%h|!R0jiYQN0;I-Bg;X|DSzsiK%d0&4)YGr&ML+`g<%-<x<*{6gM8~L8
z4>Q)Aqgc!%;}}C5bb;R10w(AnRmLohJ3iEE{m&|Me-djvH%HuUBXyc3FC>y~vyI>E
zP^grM;|NTGkQO?M&~c3Cc{q8Ont_t>Nu);@bXeY4!7jQSIdUsO5V5d$l{nHQVbbTP
zX3s*_TxF-6HOVP)YS?P~<YU97i*vl;uG^4GAo?g8b1e>2kWFxnjgcCo)0cIZmSrKO
zBuSFLkio5lK<jLkE)cl+3gv-O=4LN*`s6cQJ^4pOoh^FU6e=SuFP@{dy20jphi(Y9
z@$J0j{`cV69%&qs2$Mw-Dn&k@<Ad*dJB3`1&wusdJ}CXQtya6u(dVBbo|<It+y!D2
zGhL0@Y6}*chJq)ldNHFDWs)ESDY1$LhNWcw!W;%ep;n=@xrL5n(m2F6SsWZ`h##eF
zE^ScnZD0%7v#m<l>5waxKtmjKQGipfvboft-R;rq^zgeO7cZP;abby6XTQIWoehn8
ze#jsE=@<F%yWYh4OQ+aAIYTR`vpIW-&8<~(PLAQxF^rHTamd(UiI+~D<m}~5f&^Sw
z5vM+3s6mx^!@ciADF>sYmovbP(d6?kf9q49<kXonv|dNDfqYf}vI1v%id}DiC!**g
zrJ+=G7@a8N4dh5gKyzb*(a~X~Es!>#W0H0Y(ikNjECg}4P1tOZ`u@v~L(52ntypQr
zY_@%bRA^*h+t3HWIyRP<&%l!=#<(_>u}N~4OI+W=_&%27;1$cbZXRPm8Ay|mIPBoJ
zd<umUm1>23`7(v^X->!@OFrl%Bu*kyk-;SWR*OnL$930D&}g@i(m~l4Q5d4r6fI1K
zF939+(P<Xuo#+VP53qY362FC|oD7u#^gZr>^{4o4PzhPy>viLAVRHOB_TO|Dwh;K8
z9;s4zxjbH>%FK=l?zmwDzZG%pRGoI$r>iWydmbcz%bVDI`eDrKRc7AyKEh6i7k>5s
zL{)e5&}@Nog#-9fU^y<8N)1a262FIXTvpeY@tX|>rgu}i?pBg0Bni5N-4<)h3(U`5
zW_4|WGndZNZZ|n|`Z!0AK1D9)a?d?);k!QcL5eEn);GL~2R`rvxcL(G`a0XUPct+y
zNE{_ZQ8tAT@N)hxz^PLw`ITS&<!{@D4&ctihxyPu-h-5iC=Bp7mKh4#jA)HpDc}_y
z(#|T;<|26;iq$e!&Z9axNo9BpM>!D1M4ikag_UXGiiH9uo8|O%ZHhDukw)U>3hb^q
z3_F4_l`JgJvb?@RF<0gd_uR|O&KXLT0+v!F+7PFpjd<CIq)ir42)w-7T-{h@c5XJi
zW%nmCN@hB`Ym+{0^ecJ5N@6Js+qQ9B7tir<EH_J)5E3_EV{&FU*X`TQ=`$ypy>fxC
zKlBwmks#3cy@W7Knb~^>Z+Y)0iKB!I7tY}OVHO)@XU$tGB&kpt%GgMvz6P`B`+W7`
z$MF5v&MYB-yK)8I=#>bypsOXUUSPPhNTFDOIL0W6Wu?UH^T@D^n{#m;hqSLdjH8Go
zNlB9!9mSY9Mp_m~nKe#YllDUDORE@Rp<I`s-NY-EF<KJ^ezx9Jg7asu5_Em?R)Fi~
zS#R{%SYD*HvB_KCdOtt+fBZa0ZoZvuGt+1xh_9_&k^j}R@JbV8FvMxX>gqDRZh+CR
z{Pap@Q&<IvzEMU=3#HV{&Hgn&d1j)?a3zKefh1HGqbV$pTyc>3g+>15&;LvQ_n-Y4
z))v=Lsz|+lm932>I-M@{dMktYyA|H`fsZmiISJ;Qq#06XdE+RQw7L8C+xXCX9{gYb
zApAEfNfJK)^+#DhcNs5;7|+Ed4aw4)VxtkWxfSy8@g*)@S|(qs;O27}rNFVN4pdmX
zJVzAxL}5T2`PnpAXBe7JFW~A%hlQmUgC&=pljSU{cVH0P%Mo{dv^KPQA)T(zR?nxC
zXx6qiSX*Buj-t$X=~xt82PLv|Pg2A2Qx{m>YV+7P9^>@6leB^^<<T-n-f$=T4j-h|
zX>sb*2~M1RiBsp#@%1NOpxJ6uDrdb$lECQLcK+rM|09NnCo^o05?Hnkq{MMRn#Szg
zvzw28_jkV>EBSiVtAh^%S_&rayo04HXX$<Q38uG?P#(#XCJDi29i<drxlD?|wj5F|
zP>zLX*$6Cj5|MPb@ar2W0l8`otu2fa80FAR99p4aRJ?Lq);i55dQxI}9y*RQ?pK^5
zkT@fQuvMqGu}N-V0IQHEUmKv&X_6YSEE~@QuUv)V0M8w}%8*l_TCK5)&Cy<iyJ#~|
zEnx)$flaIpJ9kfzNKM!GDVJ&_agyEM$j0!POf!NpSW?k$G;wVkS2?(To%Mz1ajlV;
zNA};2E4RPOTJ4-;ur$bf@3@nzS1)txUAMD!^%8V8`(Wvq#`+RcTHJV`#zrG#DJgT{
zjqfK7dh{OtGm3i;QoQF~^g2FY{pUZ0#^R~=2zT6n6LAz!9!%lN3&eg0&vnUHYOJj;
zBaLQs`wrq;-b|FQ(Q9t;{ITb`x;Rg((;;6dvv2P~_U+lv*)zvEbLIpC168(7?_y|d
zij9pw;7#}4&#ia9iA1NkBIeL_dy&c`PSXD5CNn`KNlKa~^n9O(9{MVu{p@G(egF0T
z`2Sy5RZ711$jub<6?7aEtzBVYYl$K0p;I4EBuEnwZLVM^F<zzC&rQ`B3|29R6%>e?
zo9H+}2bowpiDHt#$0`(vdL5iX4ihKnD8%)$1ev_6m?;=e<g3iDFQQ|?&|r-p{YU>B
z_dW1Fe*M>eiN(dMT)H@)HE{i#v+hSTT`SA7uq=xx3|L-S$}(Jq<k}@m_xEmTnv(Wi
z{jcQx_O(?)qb-evA~6!rEmImCqtnfDpmVOpm%j8lHkKCHH=H$?K`&-#Y=-;a|D)uK
zRc2=|(x^8W7#hVO5h7a`hDn^6#i9(p9K{K7s(Ip>XSuq#_?n*gVaMZLxeC1`!xN24
zgEM!EjU=Gr4I|wgQYoUYk0e6^*#(d8+B%+{O|^74Anf+A90#d2B8fm{YtS@_u~Wg-
z#dR((^tkhuQNmUm$MtaBJZT(e5mBy9u3W|~^;llnqG%<!wu9%`xJii9s`J>Fzs&nS
z`h)D*b3F@-a~bX^$>bWMKdDVk=B~U}N@b`Tou+iUU4p<TU$7Afq|6XR5=eFW3h-TI
z>s=`%QVOzT%S=`}i7_UdxEi6d=FiIG7KTvDA&Fz|zyCf?z3@1VMu^cqwt`M$oo3y~
zvMoezg7<&u?{LRGZ({N497dZATx==gIOAz*BnX9N+3efB>ot8QzPVO5Hu%ybk8_`E
zvvVxY_>#uOvpx3jnIKmv5=NS}<qgKBMp4El)*9Qj07HATNxRb^iG0dl4k;``t(e_N
zm>eCZHdw)u0;vRcK9903f@V8wuw4(YDp*)pVWHEZn`l&{LAg@l`s$6WtgI0DnX5YO
zzng<NC2aI~;^`B-`Hnq2cj5&q#S)_<!#Hcp_`NROAfo37Jon6VoVqZFotNy{Tcz7=
zBb7t0Fv5pE@}o>n@9l$zvq6(YfU@yCK^&zBA^7Nbe~3pOf1ESt&b=<@^Xk^G#*j*I
zD+6r1_a2td&EgK$IJKfECIPu1!6_9Ho=4pCsn%*(o|oNZj7Fp}QLM2#9=**K{H;w8
z5>+Y_+X71piaCck3W4G`C0OusLnQ2s3n$XPOD#z;XuQDzM6N@x-6n6l<Z?v@hDLCd
zBImf6R3LK>i(4&L*R~iKAEHt!Qro_VQzu`fy}ZVI9hXW4I-M3yu0*j`V`+UIVfABP
z(j*J$#`IS)K%XRUD@7Ei1f3p9IijdjHZRY!RxpH7*#BY>-+55FS5XGTnU$w_&^^ot
zAAB$CS1!|d?28-{K5-PHqmYr2DMG(TeRG}s)C~E9w{rfApJzY^OuqYlsN3#iWqzKo
z{fmD_xp|(A;p?#q4!MHFDwg>A@d|J8iWFsn<K?+Jdzr`&*tTaMjoJ{RJWh9Gl{4p0
z6Zs+Arnb=zG><<11lzXl=G3W|ICJ(S!y`kKD>WvkcF_xa3=WSnGd)QdcA1)(VE4`)
zXd{V}EE87iEb}vp6QVe!)oSxcfBb2__Vus5?!wvsnNq-8y&?n3B8dZfi)X3KKTAzU
z2x|bVG=NBYgsml%GNeIB;`=BshwiUDb)4e4IRax4Nlcn%@lTD_EnLgWZq#XlWxLq~
zHH>hJIWC=FAc_nFUV>){9NU4QO}w(i7yjhay#2i&VQ^ra6DN;l(=`mj2#m;>fY(Y2
zLC&$WDPtTFgaK(Ha9s<?v`HCvF5?JhNg`SBH~nOhm$knbG%2>1r#8ABWx3dvMRRM7
zc59Q+?@_c9yqq9PG#ZCD-1l*2_TR|*$^vt<m#J0<@Vp#x9Q7x<h9t?ffU(XrnQ@YE
z>g+k5dH%JHl-q4PJdmr<iWNO$kTPXv!loorE-bAvGFD@7z$NMgxVDWknjmb$Rx4XW
zSvKSZX_Bo^{Uj#mIHXaC>`$;-jV=pI>zumM<L+CxlZ4rXJ69b5TM_#PsjRH&bv>qb
zkF&9{%9P(h_qvQ#2U*`*Wv1Qc<TKx3Z2LCGM<(#xGI5+Bu8nBuE8f28S;!o$0!t~v
zI3`X*;v_`M3}%fnFJp(`nuapdJZA4B0Y+u!7NtN~7&63FqEmzsxVbWJsleBsILe*3
z-b`(<%IfM0QX4kcmJ!A!UvOzRd$^v>?nC$T*7y7{qtn}&pSwg9#aXjp$aHnrp1o`E
z1i|Y%2?BVr+v27HkA04z*Rqik2FD7N$`zcPho@7dWzpVhP^wk339S^^w&Kd=D@bCL
zGT2&BZ->NomC315s-*%>-obX=Z0#&X_I#uXv5rwDrdTQ>a)x@lhifaUIz}WhAOF<f
zW@UYqV=uf&x8rBI!8++b3WkTj@e-Ai$HAS$ggD63IF->!1}%@Fv1$BPgTmkpl@rHk
zH5OS}+2W>KZesh)bsV|vjSLKoqD_L50?RV!RQAt7Be1aTtl=)^i~PO6^LO~$Kk*Z<
z3y$bjHFcsNQZJ=VVQ`G?_uk1wsf=mVSz26U+u$&@@hU-U6CqNh=YYy?!#Xt?@KGAP
zd;v#I(b`<aZ)}i5W84BOYn#~gh(m?cGSie;7LeUiw9p_VQi2vKLfR04NernmD958v
zP$Y4LQ^--S3=(!afF!lENV4Vm%j_F=NR@*V32uDDZQOkSTX^cf{AXS|^(+S|(CkEP
zyY*f+yGwMnrjRRQJ9(5;neIjlPy(aDAc>7Zh!~M7x{Z)@#9&L8>d=1XFD?>B37Kk0
zys~^1f8h`Mp;Y~z@0;-_*XmmbLi#^b+8BQMH-8KF{EOT<=y51-A(X}sL*_dvmAJ!L
zwZhoMcG~-IVc|=Ejwc<e@B1hmxSrXgPx0`-`S%R6$;d>J&ipa<z2P<%7dN@#&|bc_
z5OUkihbXO|;^N5@_`M$6cI~E>cL{ghLevg;^od7!{^&P&+dJOR#OOGi-IUdpRlf4j
zR|)(MlM`dCZme?OeS6U;TFn-&;}8U04({L2^wcCNkn|a8sW!xMOq?Xdal*pF0>AN_
zzsd0v$6xP{`+vG_vJ2ek<Pd2>Yw04nQ(vW4me{ezm<FT{y|p=H8eu6(svzw4aJ?Kl
zO(9JoPC$ZGDig#fhIX&b>_USZ5AP)EG;#Aql;x260U|XhJID6@Gc+0vrdQW-!WN^|
z0oJxwiMBR5_Sj<>DcQg800wyK>BkAf9?eE8bFO~Vi`-Iz;h|cVHqgm*b4ikrFStmf
z$S!7CBPdM<h|RVHnX}axlZ|^I1FB1#%Fr~nlgH?UPN&7v!aSERU*y;eFH!OmbRr1?
z!=4*%<&6(~7oB#4E3+4|EsJt7Loo@_zYO$UyIN;Bro;%g8cn|P^{>CC4(TCj@m8-&
zH&Jwx6fFd`qGs>*VN4Qob)kjP7ueVCFfulXbUi{p#`av?QUOJl6d{Eq+-j4=J%n<J
zq7acLm^2|ob9HfzW=C_~&Jl{rpc8{zE~7j*o9-qFmTS}P)^YO=o}HtRFEQ+_P{XH=
zgON51?FPH@MIQh9!`yP$8~ES{KFYwr2$!y$VRdDNz)wj3D|oXdB}26uVHjc3ghIZ6
zW9N{<L5T3>?NF+9rj1Qg5Z9P<*>`fbaa9?+P8l3Kn?M<DaC0Tx;t;ig8b=Q8=gO4}
z<UIJ?pMQqbZ;~5<IEtx<F-j@+9=ezB`s@FUQgx8o*^8{KtWd2Eki=OkldTk>GoEH2
zb}yy;OF!3ZT?vM-Z3P_6xs*J~*kqBBu?ZZ<#>u$|A=p}2LfRI7r;8;F8*6LWmO`3@
zPAft>l3J}y%^PAh+Q#wLAl)^|_F~Ac(M<+xniP{g4~|2xlWAwuz|Z0w(}bK|APIYH
zEzYxeb)GoZL}^M8r~NHj|JV?8{D?pL!c!cb9OipJ@(FIf`$6I)1_Vl3*uw?hc;EZD
z{oZ#H`8|SOkL=hbuICX40g05jj)fF}$|n0sk|3qTbv<m`K_I#L@L}$M^PBnX7r*$L
zpUtbD)CnXgFqXyC-aQOqK=&vNmKYzek+0-1K@%yw%vPReWPT~KV0289ec3BTX=sqv
z`WjKA33<cH@)owVN#cykt!x`<+ZY94P>Dq<QX&Dy$c#s1Ss0~Bbv6dJU5_M+Aki6<
z#*GQ12yIz38ygrCuzklQ{@#7`wgOyX*n99MwvA8m>wo75XdS=AzB}H`#l3IDL@5<d
z;~I^XbWlMD{4UyWK(9{H=USOGp%-X69dIp$ax79|2<j^=&t5^849J+XEG);SSSd3w
zI7lI%C-8ejVL-3fqwV|Dn=O*$TFz_sbAP8*tJmX?pFhEU58Obz)8V;G>%7#1$=hzh
zTD{CY)7P>8ogZdsY&%|MfS@`+lxi;j+3zv;)i1HToX4r`=B2d`w$yB#c#`GP0p>3(
zu&t61o;$&d=Z+!5kn!nJOl^?no;UEsGiUjuFMgTJbMtH;s$f6%1m)8wSzK9XaG=b|
zlP6hNSmYxgdLKJy_R(s0Np;Hl#u`d!ZoK{wl~S2*y9<`gxK!E@$GWc{N;z@zIKT31
zzs}OPC@l2V)VQ*Ed%lQ}(CswYws4GzoI&bqxQ2vi<0_U(u&X8FdIQUGNTZZstAR;U
zl%<eSgfS^nWM+sqEzVqACqFQNP9mhVa7!hOk|Y|GlEg`XTPRZ&FxalM!wNZ-SQHCo
zF0>k4Z`(Zo>~p;39q;4^e&ACaIB<wBf8{eQE-bLNwn?Nj*XZEj08!MVP%NS>zwelJ
z0Q2$}HX`%!+scM0z(f(o(EmM5`)s<5U8fC{DkFHgD#mD{C}95TEGJIC#M4KgVR8Nv
z*G)j!OUYG-dHef+2<13jId_Vl-(_a|F8bf1nTyi!@~T)!k(tA^=F1N~#OCJaYj)N}
z1n<h#h!RONO;J)(^HL7)aw#~1PPfZ-lNK}E2kEVCQ1`o3$40SS58*hZNrI7C<fGs2
z5QPy*65~llnn)CoL@{k&69zG3qeE;PsbGx8EflfxxtEOx#v}+gN3a=U#W6_~<5-H(
zYKgton5VWnj13O5ytcwfvyRv(Q{Sp{-)(oXW9LqO`B(lg9NT7YcAg-}(rc6uD5Y>6
z$>7jHhQu+3e7*qm^-*Nd;_U2`ED7?O1LxWb-t?Dm{ao_w-$*f1W9Lf@j_n{Z(C&1R
zXoODqz2E<R7Oz}n?_lQ64g3^k+w9zbC-3^`KcHBuurNQ%xwEIKl*(C-ErNtNL!=0q
z<=P6_U)g7aj@P>`bq&w91s}bxN?s`NeS#>)U?>d_GrVnz?na%BwK}#7X%e&533y^b
zP?3uLJ13FO2v4-%z!}$N$Mt(?HCwdf0?WM)H-=9!J=V#7M@bRdpp>HBY*8*2D0mhg
z7WH;WYi)xwr;f3E_jMe7{&@_hpCeD-(VR8GK@@XkaTCuiVk?F2JD!s`e))Wr5O|)4
zlHmJ(X5c|&F^58uBq>5@Qk@b6LB;?PR@OuUT4#u)Pki)ydG>|pSzcXxP0!_3H+4!%
z=u_1Txe~=Vq_kr%*5o9P3<>J<m^dJx^AJk)jSlAJ0AnR3DP#nBolvOM=xnZ$^jaC4
zE{rLa^Pn>4icku)Bu;uHfg<)TdfFvS1d0&bOUYZ{UNcrG3#}tmkf8FJj>eAyqR?k^
zbDdJTK)zC<cJLslo;yu_YXcz!`JoB6-*_A6u3lh%tHGHUW;uUi9k);+S8{m!yKf`1
zd@jz<Gg#Qe;JC%qxW%^3(-5A+_Z5w<qFS-=@-7jY?))jNZon;vZ{+qn@8GVx?qv6#
zJ&cTvQ7V^l9GfT%2!nty3}`ePyzt@+Jo3n+eB<e7xO#Od16k^CrMTd~nmV)D;b$K^
z!;c>tVLjE{_Ti7PbKf5R<X`+#2HyTX^vc70;kSR4?YnnU4iZlP!LQN2bP{K1m@~4(
z<&B6a6?8f_i&syh^eG%;v9BWey6-bA6gOQrPH|$K%L7yV(dQoJPrvXL{3u(uZ3Twq
zl`XC;%yQ40-h!}Pj-NQq?KdCc-h1AF<(3HJnA4|D5cr$ieaCGS^98!yE=inZX#$Ro
zvK)+oC<^)VSHH%8`rY57)9Jj<pYH#Bc|!31VwJqG*o-uzn=3>Y*GWc8czF*Thr|z&
z7-By_>y)V7B9|{>Q0OQm31cEDa3$y@LhG1jqr>Xu722WV`rX6mFmt>r*LiteD;$fk
zxdldGTNbranYZkWxNvHn;elbU#vvEC>TE04IDYgvwb~HVQ_~#0?q)1y69gfRMw^{G
zw{iIJVK!D9gh7LzlDNtt=ej6ivZ;`eU`fD2DveY&kPs!Acw3mHkI4Xpz{{1%S4XpT
zuTEH9Tjs>cmpFFvI7gp3%4E@y#$c4ot#5viod*uFI6uqk+A=dU+p%p$5G8$6MmF8K
z_RHPJr18`@p5nyG6R+uc-|JQ(mBa>$mZado%r;H6lmIC?FUN4LhE#^3(LuV6CYO($
z#LE{MnjEJvSi|U)`q~DjkJ~V5LYl@1p-{F>*N3Hz9{YDpP#YcuVX$%@LRncfq{*f-
zwrwM%>QAK7Omo%qDHjUdv|aMd+8mB8C{`<6Xtvna+~VZXV+_;=85$qw$c?x2+;dNo
z#0i%!&63aMxcd!vkaIm2<`;2nNz1m#d3j3tGG5N>>(?YmiPj?1u$gQ7lZ=OGj3F`E
zv<Q8zDK}_L79}{m9n11iq$HtF;I~;_TfpyhC|ME3lE6qw7^DmjZ{ywH^Y<B<*vZ=J
z5_4BCqm)gtRL!2dz5ydjQ_?u@n@b>y5~Ao8M(XQX2KX;)K1ovNlXvZ)R4Nj+eYz_v
znSnql3L}G9d5^_&=h11(+|@3_uEiaDCa4Y!b1HlT(V+u~i_7HOU1G0Dv7)(h*~L!|
zQO-X@#qQ_l3X3?^_(2yeo0^knHcUt?8$U_uw7R_Go$un(<#W9F(otH?-d~8>6Czu?
zmr5lDYPG&gK9gN!2piGQ?9I4@*R+>Li~e64p|Z=n5oEoM)+j4;$Qxray`@s5sbqX~
zgpYstL;S*j_?6dGg7fN|I+fkVWC~Vnu>62_dzFElMJxh{dN@L0NrkYj>_bX)W-r%i
zmYtYtWSWu$A;Pw?^BzgBOEG7$(QQ*LIY^<v7Q|^x+HKS9jBsUbn5#jKxHQIUqesyN
z%daz_8XO#LGg^tsDUDakleT?~*2sQXd#kZQ81^WaN66>O%p5pGWnh}xc!g9)2&cs1
zyYJ@w7yghT8AETAu5RO2YiI|YqKAqEmuFWvx*i}=j87K%*a3}>4E2tp9V#l)p=UX)
zCpLOujE}tWZYsO4&+r<hJjF_tT%m}U%VSv<j+>`c$vXT=QVt%vf$#mu$5~rj=kdoM
z=l6d1cX|AYCunzj*JL8!G4;qo9nX@C50*Lh&{sI{#3O7iZ1BkEKEqn8!SuFCq6mKb
zH~s^<zDhI7bM?d;?d~Ef3el;iX$)JCT@*CD)G%x-*nHro9VAmjd?~E(M}Pb%FP*vg
z7Z3o;n;WEQmyv3XQ>V}I{IOHqaoerD<Lz&u-e@qlc#?8?l%c^Idv=WD*$#fMM-)dy
zeYhn^k~F2=YV%va^SgZR3txKO_)&k=^|oA*N#U}U8kB$$FQmKLWj^k)=fDn>0?Sev
z#8+g<q0-<8kxmgRMR(hzaRAaLQ4$fyq{h(bbZK`D*Ize|=;wf1`FzHQOJju8C?`kM
z*u+tSu-C(u7KgSCvwN|}$*oPM#zwh1zd*7&$F4$=N5AqlZn^so-uK=Q^4UK#96x@N
z>kr+?p@WC%v^$KA4$y8!6umrBS(J)JY+IpIgOP$Pl`4yP(rGrCOVbo|gl)azrd86;
z5=Ky95?bvBr%xW|<mux)`Q+0O`;;9)7%TQ3Il|3%KS0m#ab@-bW22)~$`!&eMq)@z
zrlo^E3PUQe;QaaXeDO<Pc+E3kt2pG8q9i2Bj+vT<Lcyn$uVzRRDH$9aLE1L9V^j7V
zsw2a!Ev$0!g_DeIpQN^J9NTl~wKCqPNfM+GfWpZYnX5M`munPjmCWz#IvG+yf;JWq
zVWdRZ4q9q#+rlI<GVjp~J4gjP2MZh-8UPbP?7rmL^-YGS#<|*Rv(?(-y6Nq_^xX5@
zdiNc??XC9%n&0`I|H#0=AlDr@gl&1GvCX!rNm`8!@~+4D*c64lhv!(J)9hJ5f~Anc
z!V<Etc}z()VA6g}oZ*^8IyD+gxm1Uyv77?NBm{xa+UhdrFP!DVxwD)*ae~oWwq5FV
zV{)ZIzUv2mnr(Xzv9Y$uh4ZH|NXEyfUe;I|lgR<1Bqojm(ta}+Mlo?}n4O=0O`nhd
zUUgG=w(WE4T%GOPCaBZ~Nje>ZP75!WqrSOLVQ`qr$N=*fmnqr~M|Kn$86INGFQRwe
z!p7z*H{UfwXCc7vbU1R$C|9qnpogcJ-#SHcJV2xxVcA(Qfsn*e%;8~=GnboGN=1S=
zV0C_;g}E#2+_9IH<pnye%P)(Tq!1Kx9@@ZQwZ`=Hc%L=cZ}z0<-`gZwWYKQ6Wf3by
zD*IbAlMqKSX__^4mdwnOMr#xqf-Q4Gn~a4c;4N>si-#V4low8%dd<${RW)^x{m^NW
zWDmE>^7)jq&`4abK-x5zILQ(udR=mrD#o!<i6QDm(08b5rI3zI*z4jsE|CZ*x`I`0
zDB0O~Iiru~xVpNFmm1sfC#IOZZ49jy8yBuHIx|V3=#jK~Jofyv?CxxEXy_8<N)F*>
z2@gssln`{goeUzXENsica&wGs8${YMX&PZ07~Z=NcVH0DMC{zPgQey=pZKBgqEaic
zvb2U>DDaMVy@{)Ht0YOviDTz!rwY^S62yvnUa(?9j$T>B4D6t3Y=TF>hJ4}?^7#_u
zW0Q=IPBJnw&gA4IwSgMSQjCvJQmYPTV|NLb1LKoZeE7S+n|HqR0iJm5G5*c}^@}|I
z)HnM2Jorvn%W4B(p4;MjX)&z36pA)SZo7e2nzHxM4H#_x?Z5fAq@5<qYYoCwU~HQ>
z&eBMF+7R%~Q_L)1aotb(<Fgy2GuN@yjyZMyGD-RuUy{Zc)*3yUt&q#hn;bf@pEthY
zjVvrJ5ycTnl2G(u=hO%Slet#~{k2k(L1LA|#_BS^`0sy(=Z-%AEfv20KT|s`n}c?N
zwMY;cL9vvueWXOAu6h2&tHhmv>Fpy7jSP~e8fn{^=EqXV*x>g%gi%ZyN4Sn4O*62s
z9~;{3h(r6gQ7z}OTpQs!NM-k(ss_t(kxF7)7AA=@=UaOV$F+H2_XwRAmY8ic*fu$V
zzp;*pI*7#QnJ1n?*@C@$c5&#yUM41H$ma`G$__ehkh3O9l7wn$gwc^9q_Rm(#!-Wm
z#0Yd8VUh@KVju>S0CpDjWFS`_#&Sz&lM+WE7cZXYrQ^@@%(F*XTiIY+4YY>J*eG}3
z`+iEL8go}JU|SYLLqq-Mu%8_&4XKjEai*(FbxNaA=hL75Lwdd5YkuZ~wxJ*e!We2p
zF}0dRYbVn|rkX;zNZ9c)+F*Gu%FB^NA%mj>IF`lYrCE$gsg912q%rm87C9vmS`(#)
zh0PYqsWCoMpxKZlX$?Oq;&>soaze4_BW)WY1;z+WtZ`MYZ#K#F+{##ZCg$FqB{myP
zuB<oMyJLpV+8RN(PBEY7*>5~Usa9ae%yuS5hbdP^IDEq$bXqO6X_CZs^3FKDUdXPU
zdl;;hu`NlO8Vo8k$mlFdM)x6iX{w2JHnqhhf5Et5DTm776qcLCN5*l;%E}VQj=#W}
zbEkOd;jfVd8|12jpqqe|<DLgT#-Us9q1o8r%H@mrzR%?3Hbj3?l;KuVlGKnUhEzkO
zHBl53$1$_>^Ze<bfBrRpM*e$sHB8t5n(Yp`(P5nNVeF`iNHhcaB4HR(8X93NG;Etn
z@k#~K1X?|txve(acUHOW?iuEeUFG>N%yE~qmwVqcPGi1B_qie_T16)s;pJ$yHrZ-!
zP#YNK)&tvl;_?g3=PIZ;LdOBiiwoRw$6Y-B*w<0YV%xR}ip2se%S+U%6_%D(Ieh&M
zjExRw(@tZutZ%a4LqUdBf^15wltNi)f90A5(8W3>iDPnJ4rKqC5fV^DNlcnVY%E{l
z%7y1xUpmc^nI31(fgim#=kcnVx@#{jtrJp{U=$Pvt8_(zH<(8Rbu8B<^;}$I2pdh>
z9Upsi5ThYzv=Gprm>4W6uxyvm53n5<OULv=O&kgQ)JF$`tE+>&7|k&J=9^eqZt~>A
zr)c>ZSLpJ^E!<qnjd$&4<V|<8a$x~cHykkMDAim{oEa31F~m_Ohy*eNhO!}rMtu`i
zEaSLU^5r2Wwr!`oa*>h^tMw+o`&*yqZSOcjzF^@8Dcg39a^$w_Xw*9#J8^~0uA$xP
za9M?%(<#liMR27>`}k4fD8V7+mb-4}hyULHhuwP*QYaR&ltKsyLZ5%}FMo#mR)hWf
z5Aue)-^{*!2l`VW5YnPpu5sVn-_DVnZsK=-=lA$GzxYe6uC0H^m}paNIB|K6_wO9$
zo_D^Pjctb*-L;SP%?AJW7k+`Qi?h_D1RqeAO`2#LQ7_X;{H6bzAxaIF?Qp{lH>0Yf
zyzTBo9J=Wy{@K6!c{<&{__8zxf;7dIlCjYtmKLt^;)xfSn%c&}1N+&#dk<jxfYD4p
z6UTAi<r;JH%mse=*MEzvSHCrG&3{#Ow_9N=l62DqPr;5c2Ty^kdmNt0GdWnMyRgL7
zl@)T83gv1I%XPp&x8A_-bo;EnBx~rTg>rK&HaiRt3{k0+!Il^$5hk0k8j-EOgtAF>
zO3uj>2QemZW1%pKWLr`2fg|I5`S=2-8f^y0r<q?_WuV<*WNJ5$eB}{FCPp}P@F13@
zsMIV%KVWUm=g8q(SzlWtsdv~rn#UN4Hfi6PFVJL34lf&+q$1XR{v;Mou0)|cgw_e#
zq%18jaP;|SICc6IXU|>4Rhpay(($<Co_8@my`B2z8r@El{RghgfXZqAmT$7larT2}
z4P6+9eD<@S<?`jrugjUIfS*c=rG&9DN#3#%$_4|LQYg=(RLGMgDLPItQXpNAG)=Kw
z$;9?4Ru)$fjz^=>BnblAKx%7J4V8QiFF(k9V<*QNgEXuX(>rE}x(Uw0B0H>Q_H0|B
zoU=j1kZ7c9gYP2|<Xn$-&?O2Yb`ID0Yj^DBcb+=K`Ffj5ZG=m+b4;~+RHk<GwJ&|0
zZ8KBce%mc*4N~<O9dn67kFB*Fqf=Aly&RQ%jr|AqWouTEIc7zYWj|}3W~dUKMNOqi
zjHO;Nw;%;dSQM*c*!gOf^K4Q!Hr6<G;uxpTp5Wof9%E(p3Wv5?M7<O(6i4p7pZnhT
z{d77lE?zi8v)y23W*1Uf#7Xip`brx}`?_kaGks@b1gq;CeCBhXqxUU@NJ@<1Kb>2k
zY9iir1h!94g5{8yFw+}ju!SI>w~?uhQz+A}H?d^QV9BP{YO%c5VD}vdnRuZ|*s&-V
z6<bE*+FA3Q3P~glzTc%<EaM18n)A5-gYV_n9(n?k7~<t6E}c71p_C(^^LWSG-wR~Z
z_kl{8&CPX2hK9NGwmYcRs%X=f326PYF~*<~(jbME(JoX5euis~<4i)PEeoMke`P*^
zn=8}vTU<VSl!c4Wu)27R_WC)(I7Wnm-D5T<uD&*=r~g+?U8+-b8l#d36-bJ%OSisB
zNrqTT(6eQxk@34&81%{}&hQ|ykmxwhSVdYB1_`#Kz;m$B*r^8Fqm!g)zem`Z;L-I7
zdQ%0OYqK=!>%8@SNAPljW+P&KGhlSbB;zAxf=-WBImX=a+i}-g>~WW}FCi1Z3>D43
zz(ETOAr&g@vU=t@%7Y_NP7zLy>i9HUm!Bt}b|{q<bF;I2`uFPGbn9+*@15b?xkb*O
zeVX&97m@KgPk#M4D=TeI1TiadilbvJ0xX#cX)EIs{K((>`y9F9c9iEp|2C}q>z51X
zFVJjnFgJgiM<4$J2lgN4-S7GkhYsC@?PfWo%Jvu@8scyK^-pr(x&!>g-~R_(_%;oL
z|8G@l3`aXD^#gY@G&I8H%NO~>Km0Vy^Yg65*-JDwV2n+rR7Cq7{zsF_B-Kpq+|J+q
z$)Dxo<qN#!t>ffNMSk)h{xm`Om#<4<9P_oup5p#DXF8kxyZ16ZwT(DQNRsTAIeQr=
zfFTSco_h8ufAGhjrQQDa*U$aeWl}-cq<EHOdc-g~m`$xJ<qAqjl&2UR8zI#x_4PVS
z=jSL5R2bShjq)5!r-L8%5J`g28m$HOUQ8?;%EJX3O-b7yq^EN@j$yFSr8E%WSQbVk
z5Nnj@5_P-SQeitbo2@!hS{&RyPR+~l+b>?_Vz)zOcnW9jDst{D#mOCPTs7d>C<o4;
zIZqtL965Xo%k!(Wnl1L&n+$DS0dp9WDj?%o8AQebQw};a*;6Wcc)22;S3t<VJ{V22
zQRk(Xp6C39GraiXair0d?2IY4@5Wo%d+;W-)~v29v3t*Ma;`(1WN8Iyn)d(HFC&p8
zf)}2Dp2r@0^mRM)AXTK6W@a*>(-oXw3|MV9F;cR-yvdQhQ|#R{iB~RU=#n%gO%kM+
zBP~Q=I^-OOh1og$&_^W+*kEjn(TPcHRpq(0!(5!6X5y|Hw$9ITC0apr@>EB6a6#>6
z<=i)T<IY)%IUB4LrEE;s2Ml2xQm=1NC{?Kza;UP)-3MpTJ;{74U}Sib{Kg`-r7+_&
zTsSjFq3lsEyCgw}m8~}2R+n3j+(D<?VxTn6fRiviF@iRkHcpVO(o!QZNrFjY(lo}H
zID-KjiBu{}5yDU?kB}=5lO`z$&DPci$6k7#7mpw1(Z`?Q;@OMr8WdT23O4%=-@;qp
z^*x~A@}+Y$TXkk;w&OZBVf>0TMC<JRsQXDp;{R#yz2oe<?>g`I>9^Op(~D*_>cz4w
z%QY@>oZ{F{gCvlpEIgED39w|BeKu?;4`pFt7Zw5mLV8GZ94B#{xL4VhWc4;G(|f)B
z-qX+T`Qwb8gpG?J#*RJj*XvrAG@3d0{LVe+`}>y9M+&9Anx@H}-}pL<OG~fVb@F`7
zm;wKDsKNoW#5LU}H*N2xkcotYGOows$_k59i{w*rI`cVDWDHgaX2+OlW_kLFr?}}I
z-Q4o7ZJaw+X5L$7`NTY%qVpgGvZ~_SO>&t$jcS=%y+TEf(02VTbPt`UIK6~YUBO#f
z=J=6QT()Z`(Rc<BSY{EGIQ4Rw%dftgT|2j-sGy*PdW*J$>w5TskD|zEvWz0jVP;Nf
ztFl#L>?5!CI%=wp5lxcLbznpdW+zA3|M2IT8+nr6fos^b{RR#^I!&!yV!0-`V$V%j
z>IfJ94K{5T*HagGtNjlyl5b-KE`Fl{D#R24ri<k|NPd8ajH(%^x{U2L@l6LkVbCZx
z@Er$L(~vY7$F|Tk4cUkiRbm*1iLEH4h0f_pf#q0~XsV9u%Cxn2v%FHIqt9m3CF|J{
z>BO=lTsSpNcSkGfM3f^n6S>+)t~^6Lp(2R@&-Yf9<s}@)qbVeGAyfA~sz;9_UH+!f
zm|2m?wze^8x+E%7T(NsMwh$OboHJ*JI6pFp@Kx%SI;vz7EuZ7W#Cb;SfGS}*QHJmw
zsWJ$-=EfUPb)Dl!k1)S5!_52?vkNm+D-~8&RygqV0l+5`F-WANEG{i@^w=TxUVaVl
z{qdisf8EBgfm05fb#J)(Dn9uSALCbl{Wm#r^7L~eo3Dn(*?$e=3oHEj-+Y3%?_AH_
zPaomJg>k9@$RY?8;sG2JbMeJ@sE+ILxi5YeEh6*Zf9DS=7mNJd&;LIhK6H?O`qG`>
zr(yOi^=6X?AK%AYZn_akRS;5mtpt83qU-r#w$61;9(eR|zIE?IubKY2e~v(*pvoi@
z0o?@?Lra7)w4#t{X+brj=n);)wrK4(NF`G&OwKTM><nF-)}yF0wdD$d=MdKod{5By
zYzEr8StzgLV5ysCx|QyMb-1=i)UJ}aaE?pbN602LkX>X|BdCTqQpa_%nst)t98pP-
z=xX6Zw|{`U_C3vmk3NN*YhkHgW%ATfa`6P+*(_?hjg%2bjTuaj%z$Uno^aUR6TzN3
zi?TQat$jS_$1Vesg5}sG;%PG36pErDOG;?_6nF$eU|AN!7tS$0cAlX#XEAM?)<O=&
zT%yp~!{yh!8O?|=H$Ow6HAgO&3$whT*e0-AZuLDpKiu5S&d%`FJMZ{6l`r|{)O?S&
zluv2dV}3<t%n67Pus#K@7ce%vLbX((ufLPF&K4qx7?Q3MC^DMwQ?J#CWnxqp7SSa^
z&Gf+mKNY1@i*tJU5{{+XsLnU(%)4wF*vd+Anax*cdGMYQrmcV-gIBVD%;Lt4bK!yn
zAZuZnyw+?G(RFl1LD)8OMCA=X^#QhMIsW*6{ZFdZI>}s)nWcHGQiZmbHe9Eb`I=7C
zs$<A1o3?CWadwGj-3Dig+XpjfZUaX%!nPn)!C$p?6tWDmgz!Vn!a$Pno;8M*R3fni
zv2<&=wvwP;uW|P5X^x*b!jYp#Idl3vDI*}EgX1Z*_xE%CTi-`(=Q<|GN0^_VX7lFF
zBoYZM(}h(B8qf2eyI!8vq7xD{%_d*^@)wz$oP5Qukr$&T0=lodiLz?&iBm&tNv&{A
zmquKYab2GilT}Q!&W@VIj)4|p5sU2HgLJk17>U6_P9B`+{JC$U+I^@dY-t=M*Ve$7
z0~%HX(+*u&q(p|Lr3#}>iRR%0T-VpmcZSY$+zl|!o?@fxvFzlCL~BH%I=NJyWOtfO
zvW>l$UB<fJ(C$;xR<mt^gzI_8B3SivUbPNX)Uda1RlhRueI!)@RHDf|dMw35k351^
zpQb$dD3aqN37N{u7?JoO#TA$NMbI++Ty@=B=p0<&PyX`%;5aXD9rmIJB_P~e%8rjA
z$)O2yy^fL4vArf<qmC{qsPPDaW#L*4j8p>CanaQ%ni0jdZA`0)rl_c@f@L+s4~wRw
z;Skp(%vzjAZatS@wUtCZhq>%9zqmldGAXpI=h*SnY~H?|OZN`2VN)lkj}Bu89(%7F
z<lNnY*^*8sT_+jW!(H*J&Xg2{i;1ivd>^}1LN=TD!b9~Hj6xU1non}<80mBkH{C&8
zYvuYKo%lkgDH$v*FLV0Idl{QN%ULs^E>?|>WKeW96c2U6tj+_EKgRJ>$EZ|RRz)tu
zsVdL&uq+GL1J`vCqKZ5VsdS87zQW4N0w+!!;yv&F0JpsDoy6iX0^Pu|>ule)oxlC)
zU-3)-;a53*X6T20QM^(-`OFdKCPrvh8Z<nCEdnHda853!8+3Mcv18{>4jkCe?EL)q
z{e87+^SRG`j&iv~zBR|MfAGKYyTAAQJhtyCCZ~R|+(@-n=brl><mBlyyz{nO*|>f%
zw1f)+9LJ$rt8v%2?%~j}<3I92^b66hO2jmu&MuF5OhJ+rY|kT^PT|`wmT3|MK3c>;
z(gm_@)6&&Kak0qE=p=!z;#v-lZQ_|e>2#isYzt%MjXaaxOxsl(@aC2{w{!vBP7wqu
z+4K%-vpUyyoFyI8z_U?xg8+$!Z4r;gh$<>xqfSI>(sWD;o!z|aZ8viC?)w-vU7AP~
zOJ&AO71HViQAI;iB(kc?o`DQ++b}?zA*0w;g7P$?t@pcS-&K?L$;k;?S_(uXQ5I(=
z>F(`C(hP(M^{8fNrx+O-qEKkzC*JoH965fRryqWRWHQMW*T0Qau8m5igr>p3y58p;
z4J0Y-(Genic)MOGR^4ni_|liZ#L_FCeUXWJ4)qlSMZlh%%U~hFx}FS~WQJ%ag(9m6
zS)p96q19@{6A5G_>g5WhVv$HRLN=e}>_ml{0~KH5noR{9N2k)djn)o9^;ik)63tl;
zDdJ!xYutFt01tnAnraXrXSYx+FOW+Hp=?z&hV6SGNyH*?Q~{>dK(d=mjg7FJc6s9)
zZ=`teFi(t5Q43_Mr4qB168-UM3N3AD(F8=2Oq>}(mID%o$Bo?);%<?~_&M~gR}dhF
z0lpkI$qQtVH3B&-k4u`0tg0AB6jh56OC-a}4?bqI!O+kto<8sdM^7B%nP-ln2%nZ1
zh(IFW-NT!2`)N9R23VY*WoCMk-rgRP$z+%nUY)#@{bx-E!!7`MRfffJ9KQXXZ*%JO
zsaN(Ac@Z4f=A)neD^f<3!LDA|c{xwt`v~vf)kbXBHT1V`#y|83Q&aO)i%l-sm|;Vq
zh_rME%jqlGIIxAG^L5nfDK^L>Z0~CzP?%YqXQjAE00ffEY%_-3(Tf#$bPnw&s!Hrh
zYAi3-nR0xT^QRcF2eGnkBs=p&)g+BZosES&*Ia!$C@Q{?k%HA;I617`v6`**gU|+4
zmP6A7c~z%Sku{=;9L=CX!&Y!bg98VjV&up@?A=-beAGykk<pVpbMPq^7HtBfg&)8D
z1LRs-i6xTUc+J(^_sF9!>pEQYpag`vO1fJ|Hy4pS2hA7wb`n{U5Uz)!tD)$yEa6x-
zdP_2_LrXF!Dw3)r93RKAkYyQ7)2NlpSor9YO+*T4I8iP*Hl_W~&>30BV2ev%qK&UV
zG|YveGIFHBzDJ+tl3h#m_9bvkhxr+c_IwK~azMG6q28Fp(7^XR0xwi9aBUOCPzfX%
z$@P(Ko1j+3aZOYyijixj*?@>s#~yi{@`_ExlkoxtTQOM5_cLSrxTP6}Et6Wf#Ye3A
z)Yh%<-)&D!Ofx+-73wa_tLfX-c)<4szL1~m*8&N%*`T=6B%6!TY?^%fGaqAgbd;a{
znO~&P){ZVDoO*+u+qUz+{^y_Z^S|(mOwaz?+6cW|2w-HfLROV<R!`Co3Z43Udim}D
z_F>-g_FHIaZRPj=`$zb*|NZCRmw^ietQ1YYa>v)$yrqY0uD+gIZ+SPj-*y`x{o8+d
zp6?7z)8y=h5x#uqo&4MfK0r2;1z>Dqg0Ft_ZYHLtf5`9p)rTu2;u(o-QX-;7kR=t%
zY?4T)h~`toWF>6$3BtApBatBRTr&AIb5nCvYaVXh!k0q3rfge+ven7c!A9mQ7A-SP
z+A{^R$rPJ6C)oek65}H)bX<NV$4?fyx_2U+tTiI|5*U#P8V0V}q~<oL+BO%SdJHqu
zNsCkCnw(6235JEB7}ZEf9)T|*WI?N<(}l?;hEFzdkbI3uGKo|x;>h8>D9p$N2m&^3
z+RWLZAu6RZ>)N_-9S2oW2v!FfhA#|ZS`F6s50J^V(cam~p(pNV`<}h@t>21eH88C@
zm+!s=O^e`oVaXQ4eXH+(SNARm0`7m{0gfI!`pPai&=&DYYYq{^!H6knnvQ9DNNOC#
zh$0bA9GYekUDeT22`a@Bj_sf+5RK|AEiQu*<AF(+ye7DHN1oPLjG0Ow+F%c*x%2GU
z+lee2G-v9JPggOLHaFe82_;^@u2~dEbQUM#<Wgp6yJToc!Y7l;q9CZ($~0?rDz?Lv
z7^T`SC@2ywS`$w=oU;9}{|dp$S`D>UC7~$9RE2iKVCT9tZ`v@(x`d4CyZCeCu<3F<
zS%y{L;P4hLW7|IQM2b`@iLMz)vVw$yznW<kBEYt7#>PfCd}Kev!xuPu<OH7Uk~0K`
zA`wYvxaN(w)73YCV_8g3jnmoHMLwU$bv;C|+MqAV$a1KiC|3I{JwKetzW3hm@aUtD
zywaD*OCk`0rVT@5^PHQwlZ^$PZ;UGZ+Hc;*&h^_l_>o^=b9sS@@dchfHph-l1qKR5
zI=uU+3_VD~aY@EpvKbB2w3(P*!g8AAa(Oa61F&fq*7;eSM4HITEZ29$s27VARE55@
z!ik2%;rRwbqi4y)rtntU3A#JkoL70%=B>mNF?=DyUS2;4dm()n-}CT14<A1?PLM;d
zXdwi?B#5P3NF-aaJfDOwGd?+mV^tVCe+JpDQ?4o~azK54j!LCYr4EtY0PlF;&$Djb
z0KV&>D+;&X{3ec`Jjvqn%cntkQ9+#|smO#4zPedNs?8ug3)6+DTtHQ1G+o2<T%45(
z!tsz516kA2lu(n&^L->mMb{&^c7W^HxVD9E7??Gaq@kh-88?ve(ZTgtt~AI-CE8mw
zcJJt5bh?b?`M6D;iSrBecXyJEXbhLenQ7K&N<}P3!?j#o*T%6egc}g(GJ$PjB(em4
zK;T$JB?-4tMe$rbB|;<`!}BC+4ToI9AQn-;br6z4(MvE~Tw<zJp<HXQ5`Zj+B8TGp
z+#kNj3qlAyKeRPl{e!Svx*EiNcXl&u!VG-yO~JxkgGSxObtDcPd=k^Nc;8R_6q`0|
zL6a1G&*7>oujC`Y{oDNJfBsFHru7<_hYcxt0_AygbkN`5&wD@c6Bv;QV<W?iO^p4(
zuMC8s(e#;_U1I3$32uG+d${G6xAPAl|0K?fl)A<zXZYr~zs)<|ek<okM)=0J@1;_$
z{_Ee|j{vp|=|VuIE^~CL&Ro4nlzBE37TC4EmA-*4lG!A(q7V=e*b18OQ>!;fw`8!F
zmk_#6qh?`7au^Yvsd_Kzt?g8%muWPr=$@ddcz9^sbo(ZThMGve&$Ql8#hfOe4znDx
zkcb&^G+DyYBM8^St_4*60QcC_q-B$Vz8r%EfonDJ0bO-zS^@J-8x4=Pe3st!6uJ!Y
zcmz$?v6tu2TnkT&gfbGs2N5<^Z{56w@yR)~bb{HXdGh%JK_D0%Im^uK6fG?|V(}zx
z9lcD<Owr!k$EBBD2Lx2BWwvbENG6@e^?A-aREX7vY(MO!l?XU@ZisK*{moa!)O14k
z>1hw$3rtg{R+8u%P)Mhv;Xs?N5=+I2SUP5@%Hmk4-Y-ciiXOU!C=$%f7HLsbuJ4c1
z(UL*Jqwa$iF)+&tJKoWS-mFoLT1?hWjvR2g`g)c2R+%Lyz{w`C#|<P5RM|twGJ$2G
zDk_qql4^<II|8@nQ&KdVNA}a|)+uD8q&Kv&)$y43T}<D{Z8(fpJbXzeDP+22hfAY^
zcHw}lBFA*1l7lOOB&i5RS(RW3btgwgFL3#uD-i+Isx^|yGy!2iE>>*;XJ)23ar_YH
z&z)i4zWw-tPc9WBtu!$bG2ZazcQUx;QiKSYoE)RQy^XHUj?j|Mf7UZPAe3`m)lL*)
zPo3+!Jn_Wi+;h*jUKQ8KOC`XgH+40xTfz5dY1w-nqf^uLZo8CqmtTwb)P3~yHb5%!
z%+WbEZ_JWPD|q%YX4S+B1dEFSLIgx&IxXG(^xk+Y`KxbcrCed>j)0X)jpZ+W3?rFG
z%BGMM84ZVmZm}yG;pmcw>)CA3OI*>_N-nOVD}WFLL8uof0ttK%-?i{Oo51ss6=`)E
zT|$xrbUjWk-;N$l5d=P7P^YxA#Q4Z*zVP`^<1SvHzf(}FRS0~6tVU?-yM)UI-@u-0
zZa^l$w?l7WAxY%(8Q%G}oB6~SzOvd9_QU7m%C9~Iz67?7R9QxmJQP4RBqXnf-EavA
zP$UVfRK<uzNwl>h$ubfixDKA<5O`sKn=3-Qk-+n4S{ACL6N_me%LrK`Fg02RQY=gq
z87t-R)5~1DBhIES34PO6o_J)KR5XX_R?#$>K%HiEB95X;_<kca8g?D()fzz%Dn&<)
z2&qf~+4pc7Rn)*ob{zuO!;u1%)zoaMX493=laU3@AjWLN;N-##q*`ea0rQmxA|M<?
z4;==-|6KjE--#diyMG>vd^bDg2T-rMjE~KS<!)J}@sADO|GuB%(p|ej3XpvG@%Q`~
z4?lE2cYggFc&{NLJO2ej)Wb)Q@Ug%BD`w{B_~u>TWO91udCQ?f2x=9Z;o);QuE*w0
z8_8ue%rCxhxl0Ja(UWH=mMcun&cEz^5U(bbn8o6v$HbDxoa@magG+ki^mODfG=*}p
zLcQJ~7S%|kk`M?Q)f&}G1yxgtW?Q)L;1Ur<<4x<6C?s%`{WKfUJCLC*n?tYGx$nq4
zv&-}R>W8+|TbF58UDVARsE-{YpK7A0s}cbsAdod27miQYA>T$em0_^oVIUGpC|M~V
z&y7$km9ZQfMYTz#ePmT5p4CZ2qiCvvDuH7*Xxc7DqmE?60Xb}pcSGk3+p)=};~YM`
zAKwqS@~RwDQ)8SzKZGm`GU+VIRDon7O)`;Y+s?~~CF9iUB?>J$dV9JMl0?`E8fu$`
z5AJa4AD~*T@|n+m?o}gH-IRQM*#%FcS&``I^w4w-O^;wGD!%I>$Kn{tBw8dwCZ9)=
zWtvTs@iS-9qY8$uBW$0|DT&sMK_nW(aedUFiRU}WvO=MdBiqu4yjw8p?&rjT6^2hN
zkZU!NP&j*Xma9Am4+}|^@dKBpYY_-=ojOa)C5|t8v~AcxTr=3ym7z<Cq00&@Y8}fG
z^p%?kO`xP@sx2xjmJfkUZ$}FqEm?F4VzCHXGKTLv2z(@659L$>K@fQ8hE7*k4^KY%
zIG1eQNuyXs-!O<8iG<UzzR&#pET>N&rQRrW`JT&}nO&mZs4{c@46$T{%dWnO4O{mB
zFgHI-I-Q`Wr{}xM1*-vn*h?3d<^5;d_5_?A8sf`e{t~urzj7DGzW`4-RKz5gt-G02
z$0$7mn@A^;IJG8zAc!Us1cAlo4Jo490w>ND=^seb-`Pn}5cs}}9|!_XCBAJpoo{*%
zN-RpfS)qH&R*Yl|kA3_v8B7^iO_SyM8DvEvlSyJ2D$#0*LITuyihQnxl&)YHI&vz7
zCn@-n5@uxqB#CEp>4H`3Q6VGA3h`8dc)ArN1wU{pmsS|QaF)}jPw>>f$FP?#(ANf@
zFECBPy20(d>0Q6d;I=EMH)_mGO(6L`(yIBwvu(B8Hf><@;CfD<yYRA{|BEZX)=ChI
z;nb`6m1X3NL?DT9!fiEusVWMpp%d|a0?)^EZB#`K?KOOVbzD0vb(pSAy;ddgTr^q3
zw`{V74jkcQC^o&>1hMW6iabwiA;pgEeKhL}Y}m1tY<D}CUw$cyuHXk2XHK8vo8LYL
zW+z!yL315sCA9gdRvP$0Kqe9+SLi@XrSTjKyH+6@H3%$=z_sxdCmd~;49b-nmSqvk
zwli1McxY%At+x#!Ys^l~;dra2;V+|7bPxzE(`RgS7RL$gbw2g!k0XBi-?M9HxciI8
z<NW%s|2hvn`WVZ_(#v)%uO=$B8o&LK-wS&sUqWx2>je~xMQp1<I+Y}o&Aw<*2ch)N
z$oQ+bG5O^}aXl&xou~xYwJWS|Pchivi4lvVN27SI4N|DLRxOsWEf*;Wh$o`V%`7nA
zlqfkaw{J?(o{i&q4k96=_$lIvCVSq{M`LD@%c4F99xHO<z%2b6^4PTs=guzCEekw6
zP$W#tqH5NtE>@}4tB7bfvZ3K=2GvN4Xgolb1WKbpv*yy$n!@*z*p`i^t7wrZS~P;8
z7$};CX*DU=>(qQfo6|&bOae&_g=IaLa=FO);bB&aE5wosF57)Iwi%Yq9LJ)gvmGRv
zY_5O+lF1awWQs<khN8gcjhi44IG%&QI$#qf;lH~p_dJI$e)-D`kGy&&*!MXN)_XEh
z#~>2(N##;xQ#ow2fxB8l7OqPqokLa?JOyM;CSs)M-!jPD#27*-=$cM@MkW!DBC9g8
z>!SN*YV!+p_C}c;pJaX^N+xUYuJ>NY=FPJ#PMMrPRwmyPL#un}YBiL6kpoQ2qEuO;
zR&QWC7IGv>dw(BN)WC}-2#kPSLLsnCw0I1=X<}szY}2M`Hc5+SIGG%gNTnmhbe(9_
zz|g`TD$lH=Sq^?w!VkipM-h04AfS-X(`=ZWxp09icVC8(#Hx*1uv&^g!_v|mg+hU@
zj$Z!Df7#9_{{Ew^jGkf3rI)koiZ`NY8cQpSB;q<-HV-08GOjOH?^~-GTmeDg<NDv7
zkUn?r9H02aKZg34KZ49vhnZ~_c*k4b&HVBtxz_6`&rMMqIgai*XhsaDQK!8lgQlt+
zzfi$<eKri_Fp_bMfvse&x*0Xo!qUmZShW)Ux4ef)vW27H`~v5`bq92HQJNn|avkcH
zOd_q3OvUKQWbq_{EGuZLK{B0yXbdaWg3uzNuAB(lvdo4}I-SNKtQrd6MUfR!nO3w&
zD%|AR4JO7%IC1<4M~)xn;DH0U#W8wv0>=^Pkp$P@@^jpD`%j}s;*3v@vN%6WDw)Fb
z1<G^1v?A0?l{Id@;d;iVrfD{R`0IO7LEXaXqZpYKj#(#eo5;jg1^hyFd3=E;M3{<?
zMZmO?EDt%VBIz<}JW5(AU^SXpjXIhcs&>1+plMt9PQc38G+VZBCaNxwu@{L*o2f1?
zV%F!m^vZs=Tyhmc4scwTv55<uIXlMG#2m}Z6|${;h?;>uD~KzT2mwn=E10H@u4}Xv
z+K4AJNScOYn>30=5{VdrZG}OpW8r!(lB{7i0s=`ziKI9=J4MivAl;hb@Iwa)+yEpc
z)R}r25rnNvwk4RFn#Xlr9M|C=Kk;|`@-O@<TQ_ef2z++!+Qsd+-NvUr^V!!-3HHVC
zydNA)D@y2c;0NDT?3X2ph>-|yoR<CERM{Fon568HPkAJgGLEZJX*fulK@bF}nt~)W
znzb6$>JqNw;&~F16re>UVo@D=smY~jmCkGgP0>TEsLC8#M;DS8py(PMJ2!IMmd&*0
zNAWF*Gl!RO9TTNyq2%hwvV!edEG^H655-i9p3ANzlITN?q-gEzVeG4)LJSlLbOW<q
zrCzH7I*u!o%_WIP4fI4T^o&Htl_W~V5>2y0W?&1tVc>c$B5-Lm>zuoAh6}^zSy)~o
z8c)$KbrOw67#lm!*!T!R5b9o5YYlA2CE^PrMyTD^Xf)Wn=TbCTp;oPhr8EgV5rotI
zvWz5!n~*1;-p_*%zq)#)4pn@NX~8>_F6%qf#AOXpZy+O~YdVcugIc|g=Qu>ODXel8
z%WM*fCx|4|bZy+k+{7dU5syM9fuyTQl13mL^3f{OE8|28mvP6(&r-LW$hyUj%Tv7J
z#@&<~6Ra#n89F^hf8_$%fjY7pqt+-hF*QoHRw13qlI!Y0-E$-6=}CI_?xlWi7<*(K
zIg&z+NASv3eAgx#iJ&Penr0wLnb3ADqNA%ihH9XyDz0l|TTN=zNIIcBmn5qMo)>0q
zg^%sJtn2Ngv$Kbpr8$bF5?i-y!?J9K&Kzfae1v2&L0fA(>1->O=TNJa*|2>pS6=&O
zbi<%rUct~+wr|~xrWiQS1$58e-&b|;{BZKs^E~F~=lIO$K25Pyd}S|+e+5DWjFnE4
zZ~ZxT@7cl1%nYTc?q{%GL?F?qlxdWUVPYZ?<%*qo&WzQVwsrb%x*fHxo$}au-0{0f
zZro1qdwvQnk>&Ka{)x%EzlhVhhZCu8u9Pikl*j|BA`y+piO1py-$zC!&{Ry@#I|iB
z>Gn{X2nbf~|7PcAXlvVzBFjVz34(y|euaeZJCsVxoH}`wg9o4D>4OItI(3G;Zqk_n
z$CT*k-@<$U{eNcrW!JL2G|%*fv)HyvI-N(6rEm~tRrk;n0ypsReV>+mj;r_X;i1R&
z{m_T*E-I)Se&~K$I=ax4aiW@vKt&NQilU)tS~!l4yz28Rqedd(fQ08`R%*Du`<(w;
zDv`o5ZCt;C@D&<08^e%Us>}3v60J#}ZgZZca*eGUHgjnId7i%i9I@LB6iw#D(P6&y
zPY<Dq4q`?U-H5aOk`8jMNoGdAO1==#uq>9BODIURwPYw13ZXp@5>3ZKDz6ZYMsds<
zk|*I};oB~T7GMQ3x*EZEbyj?twyk{}dg7VT+DuVUo+qgI;Sv^VeGz~yF+aBwM*PC#
zA3pIh{>umdU)s9bF*Jjp{=mQEt9O2lTD|_d3GALlk|ZLAL2JH1q0oxs+sx0;Q7o15
zy#QHJh(zLmgzY$3mi0;>@~fK8f<#&mXj&3@64@3BO;gcUZE&KQ43TsK91mH5P!#6J
z#;GhXqN_Tx6wsOw7!eg!m(g4Yt$cxa>vqb^Hs8AAFr$;>Y}nGu6<6<Ie!9%reakE@
z+Vq(t#3DAn&{<fXV}5atbUH(!dp%w6`1e#!U0~y7myzvWNA<D00YQDG$js;%o~xk6
za}+xA1d4@ZgacZ_^H3r>i^~f%%M}W}ofw0A@KgigIW+2Z9(wE{&X1kP^L&z-Jmr$f
zg$qy6+1bVP^d#rcpG8PM=ZDYp?sxwT<w}i6G=`=dxVBB_Ko@O=0*>uq+cv)Mf4Afk
zfFx_-^z6(GpZV<PUzO$jCyN1#O^^T9tx{YTbhk%n&1F#nnMgc=U8`dk%i!9i^DV^V
zQ5?&~w=E=1qrJBet6V|#edK5iHEPf_>qH^}nYl+<kfU6>tDm7W3sjl`=Z{(}%so!I
zSVXcr=yuEW>&J-d&}deem>#28S|OXwlWpms>(&pjq^8LQF)n-atyIO`*yr{m2^F_#
zGcz`VX*md@NU}k;B}O5iCXm9OKFJedrc4A_4U?tPGHOdFaaBgR7DCdH2*SpB*P-5M
zP#PO&VR;D%SU<Q0KXAEl{uCo4=Mnf63N1v8ByH{8j8BgdOJ~@%Z3oFzntH8_2m*HP
z+(kSV$MJY>NCsB*=vIdzLNDQfa=FakeC%&t)zT#|m14Qf|N5gp;%6`0Ov<UTIUB*y
z70hOpXhbL3y$-k0q*z>_ZhBn$o|`dSdYCzJns)bbQk@;-Zn_O)(=Ke=<I#WoBZ@~K
zXEuK|3x0v>w8idiyGV@hN7oEu(MV|JWj2UL6PUh7X>kEpxTw7wQF9$Ao)eZ^eHTSi
zNGH>bkBrfh%AobcfUx)9v`i)@MtS<c6C65pfc*y#u{^&(cPbzsm65a*d*1LS-uAAa
zBiqu((3#_u%O%q3EXibcRj@v+Ou9jU=Yto7?jYW(gUpUCn>l#+7}Z+shkD*GDyVZu
zCpmZj-3-3<Ex1}Z=pf;uXeuZm846CrB%wqJL_o7rqq49<d1VF1w2=ZINl`H3Q6k9{
zO+TPmZej$Xi-hWW6dj*prAj{1N-lDa+0!Zd-gG;qtv#GOJ;J&T!`QaW19uOj+k>cz
zf$ysf47M=P*Ug2m-%ma`NG20!a$+9ecWKEd>FDf2S5=CY5=J0_WukjFk{?j7*D#Q9
z-O%DA7L^bh=&HerX^~&o#mLws3o9iuib^dAXuOP})3d7L&;CDzP5-_hu(Vi0({vOR
zKKZ$S<kx=nH&BxbJ9qA6>((tCIQY!Vc3iJk67d)t2M5`@We07oZB(iiPMkP~ZJwoW
zHc=HB!!Xcfg-WeKxl(zh5BXJ%ulfXlU4^!ufJ8hM4n?>g^_3#=RDc$b6UYIM;~~pX
z=<6k!OEEt^O<RkgyCa7Y*O3(+A=$J?CMnIHCc5=1o_cziYAwamgvr?xk6{H%h<cX1
zX)>T4BcchGie)CJCQ($4WVS%Z+ul#KYa>^5_K@lrpkA(0SzKVe65us$+IxD4#^O}0
z8ctB9CEki*Xqc5UiBuLBpK@uDjzT-hOdhMf?YUCdh4W`<)+$_c*;RD)_9N*jKKA#2
z%f5Y2)6&w)`E%!RZIhed^d<sio_XdVm+!rrLZO8q2xx07uyJq$p5x)THlFWdnf7yr
z;F2$pM1X0U{QW0B#lqs^E4{?e%LU8lH}}u6Jrd!^5ae6(bhPCFh;@ej(Z1`3?@0nR
zA@F@K)cukqkUT22Dj8Krib+IMNop%A^c2e2Q(t4Td^26CUbbJ-f)Y{4Wn&zE{3Oxh
zI2)yd^me-lp)x!=!OF@K$#j}zww12if12dZtBLsz1DEb1-qJx{(^!#ws*M$<MyHTm
z3qwhw#1kYkNl+|I&n1)1;954SB!jL}Tv^7h*NMe-lKq=e6$Q_85yHdsJ<63LQ&SU+
zjE~{@9+7CA4TD?g>FQ-}W{Qc4Q3B!7)!Bt28??6fp=mm@fOY+Y<O>Dr%?d%_v1iwA
zQpp5fAj0g>_mtv-RUh<Fmrl@Z*7?k5Kg;m&@GE>3{O~w5I?gX{%CNI7i|yGgR!t^W
z>P%NGHWhT%w-#t!w}I5gOPCy;V(vSir=u;2cjdLHZ+RzfI>-Ep<2?G;e?g)&inQfw
zMjI*iT(y}<L_>+#7|R!sttw`%Ohk_mPiJXZ4d!O2$fUF6dbYvU@5ED7T-yw--&_X|
zK})`X?Kr5h2%hH>1U{uwk)fdzoI7`l`yYIOlPAw$HygCY1(}#a=lU(Y;pTU-W6!lX
zuEVL5he@YX^!D`;1QL$xB1BkCbX*VDweV~Q$F^xUnz)X~<kTz=@7qW1W$3BBsGv>`
zd}a<G$4I6a%zGTzf1Hl)4EaI}{z8#vrAA7Vsdx_Ml_HIL6IIrU8V1pnv1-TRV>V1G
zD<y=gkV!;B5z}OX`30Yv=Mr<92+1WLvFNdmGx4Czz~$HD)5?Vt4#PvsxQ#g3q)arS
zbLo}qNeY?qZ#_z4`Ef4Wl%io5G0iHeRE*BfE@JUCu5D2_O|n8pZZy$do6_PkW;{#U
zb8!P7-?NEEKt#fN{Ajt1mCQ47c8r*;;Nw#Z0=$<o9TsLfrSB^fN(8Iow}OSaWuj4y
zXO8XX;Rhe&mbcwXG#2Hin{MLZGlyQ&8fyO>O_6DB$+L0OW;Si!L^7U4o{@>g5~!Mj
zVoMkiom@7DEGe8kd5T)~hr2U+tua)0xjdnx8Ue{<lBg$9UtT0?M6l|0{8|+wogtEn
zVHL|bj!h(yMu|r092g{*ub~G5R0TDrBUuWmm`lgfy^J5IaOvO`R09<`s*=e?s4uOs
zbo>-s+-K<St7DojW8;(fjz=bwBeVC-q_*y1Wnr9j?|M`t&djm<a4IEA*S&+*M4aZy
zdns3!iAUmO)0t2Jxl|-(L{TCUmgi@P$QtQffjSbdA|u%i=4R&^A3H~1XE%LaU1)~J
zLTQDq+qSW=Fvp{hKFZ?a5}xa^VSPXEc;|a)Z6D;wvBOMHPm#}M*t}sqs;pvJ7Jd*4
zod+Tya6|u4MG6NdzWJ?tICS{PE4{ogO4%12Zd$CHZ{k>^T(<{0+gp%*LUVHogcKlF
zONNeVVVM@L?_)JgCdNyo%O<TIZM3zvkjiAJmMU!SvdEV2VsWmW>Ovd7mSEm2Gawe}
zPA!s-3oOfHYG#o}wMjG{A(76~deeJI@3@kBX#q8pCDYP{?^v99_+F&tD!F_&xpqM^
zCgXS>m1YIgDUr_=NTd>2^%|<Apd{jyswFB*^R%?Jk?mQJxA_`eNkVuImT7V1#8HNZ
z&rqpUXqZhBi9Cfu2ia_n#id!!o;}6l(mdI023c0{L=b9-1_D*pF^mYN)x`H5cJ0_f
zK9j}s0z46{26G;P2twiMRi9S~bxZ&0i(llar=NO-SO15L<puot{`36rYu9sbe3{2*
znykC&MzULzJahUPlKEZ+uX!VdU2nku^54<c-iF=38NIKca=pgrSHH-SyS_tj#-S~*
zGctUNzRNEo2prO>G++D93EtN|Kzi;Nsay-19${s9k>cV!tsPy&3vF1tuEuZaC2%a}
z7Z#XaoWn9rOv}ND#E2(StQ*)srz|l$JH@%Pr)V^*tREO)*RI_xEiTerYSPu7Vb4`J
za``oHCY{YOH!}&6ps%l&STs)4v{yyy16<F?@%->t$H%rE%zB+tvCRI12YB+pL9CZ)
z0OsPRU(v9e0mF|!L+ADVbhWoIGCaf6M`kD_47T;<XnHo9A}HkZ^bV{;lQjZAz%rY-
zu7kWPujpD1l}edPy^5l#EY>^@F3PyFKn+0hT|lEPy@Isz1S1bG6VGp=`_k=fShtRb
zX`x7P_Q({qqf^wDkCAIW!=^zOS%OBrMk1lm+15rTna6b{EUSubJIDfTtI5*HB(ulQ
zFxWkS<GR?cgW0grq=0Bt!E{}ksU#Q9jbd9KC>laQod8e>9TT62XB9c5aFXezhFP)<
zx~`z9Dy)vVhc|)MH4ub>-TVsCc!WFe{xX+dx|?-<>$&dQH}L2B|Hl3SR#>k^WLYBL
zlBKPqokB}1imXs7m6(_qXK`^J$90h;nRF_~=E2PbzRw+Z-m&UEyvFnKsd1l$7LVII
zlSH&AP9RfACXswWI@f~JXyDfyc(O#MwG|=Bq4S0WND5Nir93xBG@3&7eG<70<%K0W
zS_1U)<4hepg%sIH5KCd%O^n7cdt!@ZGj+;kla=BMf#;J<8Dw|vq3<p4VQFND#_}xP
z+b_p=EruSshYf?9x%X5BB@*Gb9d9N({xq8J6L=n$(?pge;+Y(lV^LaIAeV0eRil<j
zpeq?<+hTldi2LvUDjT+3%KY*o0TRRG<BUyA)7FxwwKdPfk32yxo#xuBuZF<o!3Vz1
z#!Z*v`X190=h?Gs3yxzyH_7OGKEC%|xeCb(ID27)FWqtHkBl^hDFk;dmFe&9!>&|m
zL^O;<f|!v&QFQ`MCcs7aeWH;lhS8*0Dp56Ej?Od@Qy!eEQOGaQ-;rXlD}f%^bmleM
zMG=L>6#@VNI{ZmQK~z^P<D(#}5)G?Kaiz#~sg7w|$O58CjYK>~=PmD}`-WRt8XIEq
z+%b0i_|GA$2In6C7R(HzZ`#A5+$CJnGD`K#<0z^`GLj@7PonE8jdB^u_le~SxPiyQ
z+!VQdmO{ROmCfR%(*$0C?>QVja)86f4x>dP6!L9sY42iaXpC>&^Gysxr?01%0|)n`
zsuJgi&v5ykYZzF+oyDahh1LRo;N#d1)oPX9mu#n_wT(uj5e9Fv6t>Z@+V?B33gQ8-
z=Wyp;ck$pu55Fci<R_QveC5IdZ)l6s*S3zgzW2u%dFCK9^?(h(_+jG79A{1*q1L;J
zOrZ_Mv6y=3n=Cx>Fy>N`z3W@hVhK#gB4*8EFPvwn=CN*FA41X@5iQ)*)=8sUW_ai*
zwrR0$(?%jKEi8BTqGtPW$|YvzXE=89I90QW={mR`bawP`{pRiDvpJ3)+0XdID6v?C
zf&TThw06_9T^{`Am)Ws%JJ;QOD_z|K_`c8L!Yo}~U1YL3kYrrf3#XZb0N3^Kys)h9
z2R?ovuxy9ZXV3AS`|f3aVc}&x-xp~UCgBDGsb(@gGQ}IO+(;r3$9(1_9l0p&?cwf4
z_#U#NAY=&}A4!!^lQBdC0?!Ra6*Y+G(?qjLJl7*yQ2FM_1Q`_^&jQy&Hgtp#bY+_)
z)Z<Jp4^#ZkVVXe<A#2EiM2A}<DkkYjFVVg}jzG|`8$==z3^h+a+k)?TG|Uxh^*XMn
z5_m4fr79<no@LMae$wd{noX0YYvI(Y$gWK+BC+VX%;U2-xqz!`NRc>JtxOnSKmUhM
zmL$^2L>PcM4z}y!y@Z>&qDZ8Y2@;6}nyMiQ3D<SPDBrelTqnFi2tlo8v$VK^B;cO!
ze3M`J+5f<X_3LRVv@kRMgB=rIYX|}HM3Rp74jT13V`F2?&dxGCa)HIAC0x%(*H!W@
zS+;H7&XZ3)#e)yMx*F79jo1RdzT`6ERM|P>vb`fowxgBSLJnD0NE)4>Dfo^{;ClFh
zz!L(iY2r8zf$uZ5RHtrLi6x4(cD9jDBvEaXHk!0#N;vK_I6(k$87c2Ex46jkbQ$6K
zL}L;1oq6&bwlnbl4`NkH%sp~1JAV2<AV(6+9Y07geuDb?w_sPnaRtwmJ9x|PTanK^
zf@Fq+3h8_}p*nv49L+|P{w=LEaq(KZQ3IJ{XHW5^Z`{lHTn)<|VyW1qVpizt>0+R-
zm+NnMBcJ*F7YTficisMWIy*WT9v#K9YutbT9c&ob!lgU6;Cmjv=Y2P*^L<>`UHz;&
zRBKE8!zVsXz211Gm-eD)xITY&c#MNNmCI5I66qxCds^ts<;Z7~=!T3g%b~53EE9`H
za3zI8Uy3f(BXAu|r^)<em9g_G`PKye-6=8&1t|~&zD2QEWO=2Cf`p-JWYRfO=^W8y
zg2G#VoXp-gQXCm#@_}!#;{(5dk!-=L7P;`PFSD+@k9}sAV-FAEUcH6uH@RpF=aGdJ
z2KIK7z;nqL+L3jg(%ck^57}%1Pc^8glE}UfUW2*$c`UomTW)*{9UUDgu>?(5;rxYh
zW@i`Jf8YQ^LqnW8d7LZvUQSPUKl=|n!}`HXkrg@AIt&Dk*<{Oxjr8^O;JKk|kVTzU
zKwDkdrCYp6ad!w(2vDTBySqC@i@UqK7PsP7+$mNpxNCtT!6^g?u9I)(pP5|bCfT`3
z&dGlFTI*Smry6B(n0OOlNQZ}Iw(#7`P`H~||M!8UjXXSvzkR59QEsZR_!cuL%GE53
zPGoy;Ck#DbSlP!cmgny+MFt2&+<S_)%1Ib)y>KKXMCPGMZ{-kvhKo&HeW)Vw_b{%|
zut<Qitb}{&Vu-j}0gBN3qoy7COST2Q0o9)%x+VFrK^aWf`_cD*pW7(SUq+N*<AO%v
z3sB{3YX){cy$z#!z>YM^drA@v+ak^jlqA6)5)1DIS2ixBJt3W_o3o!Xbr|tD?Ef?n
zOC<>MY~V1l|7?@5)$bLWxEkm13WAc7@%tZbStP(t7N7;2QYaB?5gdtN?74t8_u-2$
z5)IssEaYc<2J098BLB2RgAr)3D!W210>Xo(Ysa6famBIjbu%V12-0nbK}|*^Q>gh5
zTnY3YAyg@@YpFlKuF=V3piLn11xp%ZsT_^cL=bj(571>RVlUd-TmqnF{BCZ{z;wf$
zn|H<ho8>W^y2Z^{)>hwgt~PMa+VKA(Oft1OlvIU&av!NUs4}1^ZQ>14VY#>{Yan|J
z8{n5qwHa%_zzn*(dnfofS^{hO4#=Y<9tjFTPXWjc6}z~u39>}YmdmDV=1kGwNaqjl
zBLWl#nt*WFRA>qERdPVvi`l_D*Y$PBin5-t^YgkDHB|;8%EIEF9<&<bk1{YmfJT)z
z4DBS^0Z7BXJ@3kM=LfO9{Ha%SdbSbLHTi_#T)fLJneb^H)D+q2uYtVYc1`mH_uYA1
zNYw&npqSG{AN-+Rt6ry_Rn-|krH{pxqFkT+S@RF)2Y&vq7Q3h^vcVVrWLs999nlcJ
zmUgSLQuR_!zEbys4S_(^nY*m}$HU`a7Ch3{Y$0oVvzf0H-w~oRG*chQAoq=@pK5<z
z?2U5e@^6W*Y=8mbQ3NsULi~z8ow5c70wLctl<8Bh-~}5Y-(IiM3awmSh!Mp&YOwFI
z2@Y9RofbQHB6(ci;4xi~M%8YO+j3zX<h{u?YTGG#F{)834)kk}V&<#r$p|WMZ5*(l
zpfNdUXWD6FaP@wO`PYv(aiEFwR_a7i>eOSRF?acs+#=m1n7Q;zRJavq*2UWQB&?Vt
zKz$G?#Cy<z>=#&Lk0B{_j_U&c;E;g0kPW&Qf;1K`spFTIAs!xG$GN&910(@`rXh-q
zWUi8|_ouB=s4M1hNj$C(C{S_=KyLIPO-9Ij9XYHl6j2NSJ!K1?x;{(^@OpXX^gw#k
zSC9Qn&G7uAs4hm^T8gF3fh?08B79QTUYOFr-`zCH20X#E3oWt!FVi#Xwc4vacl^ML
zkYbQtHP_f7)PB7S#;8$Vu;;GQtR5=ChH+o+n%zLg$_0WW^j>I}-l0~_%3dnS>)cM%
z7w|A-3UuR;i6OT52MWGHB?|Vo5334fYX<cC3T9)9d@WHPyy5$-@zq=*)$|BGECONf
zT#3y8fND%L@6^$c5>w<Ji{Vx%WM>_LM$Tot_)lKHmmomAmS^m<Zq*F7R^-?y|5_Un
zXbLIzcs?gQ<fc~g%14{qaM&Jer)VZHJ3uB;{S*H2^a!Ing;}HKP+AEJ0!*9pKV~*5
zRbx(LvydxIb$7QMJwOElr?HY4RyzXUfjHH)R(l<2I~6~!;CjE_ouCT~{3hG(q)0q9
z&7ps~gB+4>b;NzW+;U8P*bU`>Cs$4}x>!RoQHJDjjAA4duxf>?r#%=c3X+(|&&j|L
z|53>lTTLZHGnyKH%6qRBLeC5vqG8?dYYl5;5p&hQ<(ELJnPdCSk|lZod;4{I6wws`
zYBmZSv7EOo>~6tf^RcUX{-^KcWPO7sS5<bX?;x!+-PbNf1LLy#`t+=mX4U2_5(8@D
zbiI|OC~z0!=)HRIzC3T!WOs#nF4O#X(T{K&R`5{`6_UOMlKJJ;8~)JT&Ibo4i4~8(
zo-oMJO;P(57!m+M<Cdz9l`kq2wYDZab2uCN%C4zbvS!aAAn;_E!P@OC2#_1$GGbaH
z$q^h334;|KpFxW1+cV*xOhg88dLeEX+hR@G=vK`kmh0gGAULgzqvNB!+x5SLkmosw
z8*H+60$rCxq6sd2mh+bL?BXu##J>&~<4BX%z~IswS<y^}M_}zT!-|tsGceHAx%dVp
zmDHq^id8pubba-Yi7dlTg5~ifNB7+IuMjyd@u~@=QUYuPMIXm=FIrHbTfJ%G9P%fz
zq{-nUYZVk$&tnxf>y1$sG!vaWhzAE&qh&Dir8b|tS<IbL5P7DawQfP-R~KK@G0a&l
zd;@K)7oE>s{1YbryJ?DSbyUfh!EPG3TzT}7|Gq9MjYyAppUk>g=pCFe*GsCo!`Scu
zz2Y?+4&KkA^l5rud@zR>!$b!>0z#?2E5MW}37JIb5w1$#K+S!<^|0N1nrg>oMmsy7
z;N)n7LTPQjy`O-|(;vdi6GCqg-Pym6OAt}_2?;-s<3V=P3;Ul@mkZ!k1Y<-7O4&ZY
zc{7E?ARSFJ(o2Bcvd56x5{kGtP&3D3>t1=aA#FG<stOVm{mNWN({@OL$2F2fzTHnK
zB2{F^sWR!k$dUweL}x~jzf3OK0_oW-?Y5~8WMoS@h#C>f_UT5Z3uQF8QXdG&=ze57
z(0*$%)$>Tq=O6Bb(k_lG*vz53Da6TEk4L0nXJ}UgQa^Z}ua2jwh=c*<(68?nR@P&F
zO05PQQz#ekPhZWWcAdB$C!WcQj~L#_-1=j#c=W9^kju)}bh#h0p+hfK0&_-AElT)I
zelFDX|N7e%WeT;t#_JkuYok?tm}C(Cy*j3`s-3p)3Qt|-9UDqhW78J5x6Xxt_d-&D
zE-pJ-{QFoxh5S>PS3Prsxg9c13K(Y28LX(de_rYRy`he65)2fxw&MXM6`H$Sp-#BU
zS{hsmM*Zr2RYH35TkY|~RG9eUc{XkSi*<o>K^6sAGyVAL@e=}p{pT3+^am|JnBVyP
ztG%7+bPPWAiBCwpN*CyzTx+&A`J4@Pb%r^XDZ_7#68dG@@yidb#w6G0xbSJCgPCyE
zz~3`<lJT=`1QW62BEfsMx{2usjK-o|1Tv0x!;LvZxN0Y@{8Q@r86Dwk-(}tDY`!zw
zT%yk@=0!)FWz#2G()tx~{3>RmV%F^)xi*Zs9FrQ?LR-P70-!6YMk~1E?Yg%OM*}<r
zIr|CU|CP@c#aQ({rpF;?0m_GdCn=%u_D<|i1nAjnR(;7~jFr~Z`Nj=DWXqN>YWZan
z5#?-QFIbW0_b(?uhLq?;fE;5}sKatf0UIj7zAug#=~y!MR$L&TJik=^6P$>tr6uBm
z{ns21+g7GjJf<H{y=YA;W$K_=)2+0m%W|H+^t$P3NNz!Z&~_IqEKIy5qDvztV-RBU
zP=<MR9P~(I@;a_#qI0+x6+?=0PHeEf_l)D0hSWtsjjf_KY7q@mA8HA%WVK=U@4{sg
z*IIjt>r;NBTXEf;R!QS~!jsdZ&($xg`6b<a!rp$w&Zl8!p-z{(@QqEk<pABEkIH3<
z1>qQOse+Q;t`^G=EPy?Sus_U4@hlP~W=yC#L-FG#)pM~nEZ#Va&gWa~`j#86$>U@|
zUT(8)eNKFVaGaJ)v=n+Vc9@l-bqq@3{jPPku_3j0-Q|}J50yz4EHyfsrpdi$VZj}@
z{ep=GrOcw{L*AfQvPMZ9iKTC22Za-}Ld-O$U95GdyFNv6Lj&`0CSB0nj{20QKaEAS
z&+|bqQdn@=T4`=tWh^*F^(-R+=2G%en7HKrd1sl%V+SyQ&3J`$43Zj`5OBqsgS@4E
zN!Pf}*{pgx!i_FchUwa)s-;DG%=m@-`#DC(V8sVzAqhVd%#nQ_s$=%*VaxL;s0g(q
zcLQ8Z?_cHW1U^^-V<B~h)8Hp`RbueoX?5(cKg=1g#>&RY&xPP9YG?3%%rh$vq>6m|
znU4=-eg4dp^iL#u?ep<?S#~xP2VYM#iy{&a`uyDTl%?Mt?Dq~Cu;uRjI#agcJF$LY
zm?VKrDK3l6x*33B4n90`R4}h(N(BFuR}m7}?6PHv437_u9Gd(`7Iby^mKzbF0|8uP
zlD=2$ob7(B#q(XkMb6NUeO@>enR&2I++B8%4?G$yK1EG*n7c4UfHUz|aR8$9<tBwQ
z1E!Y+&vm};NyZP-YEcjg53Vi(6G>0xL#ePn_kLEW$SggU_5mw7{YC?Pq!g0l&fiVo
zF$68M<hq10PE5pEH0w#r%%sc=tu;TXr#8Zp7q#OlFuZlZ9Vf*1eH&PfHM8@{;10HL
z#q9m+$WDmAm@Phg9bkx!wE#Hr46BMD=Dv06_v}6-5lO28*wRFPwPB2X%;T#AQeD7C
zx*{|ERVCytH25-@PE7O39lkYEf+L?kJ=TneM3R%(LW?>>E(I0P-Cv^Hnw3+<&feZX
z@D+a|Bbo_EPJ$*sZLFN@s){H^6K-?%(myF8+ou{<#&<(M7O(8&6lp;pC0jbTKl?!f
zA3HWfUR9Q}PqTdILpXu4X$T-?qIKcgcF`^YVpoG3%*y=p^b~x>d?!q3dLJ?qgUjG-
zV&P3qk(4G^Ay=_rF;C2qjou4^u?)NF;7y6JZMekW__2f88YYAFm&`0IzRQ;>D)c(D
zD?H}xuKiHTuuP3XAVuPVoo`U=oZk7LZuz%g>s!pDXK^U`LRp@#{N89cjU32?FSxpZ
zM@MISXj}J8Q&$Kp_M>huET}B-ryZAJeZ!4OFCVTh*ylUp+K7wj2#abaiuCN78q}>_
zhOO*}EG&MG-{#(>LbjyAqYE|~hZkPwV(w0*-}L`I4T1Fdyb2NoICQPTjsv`se_eA+
zCs3R}p3r2wlYm-@u@>@tABoU3SR<?F+w1U!(!hCtN4zM!lLsqipqq)Akih7OcMJHs
z_oBhoqeV%r8WQ*<gpo2wXGWkUzxw`Y>C1?IFC#b+U{6;8scqsz(cBZcS3qvM`omUW
zk+F2a8AhV)2Rv*kPef+p;7A2u1-A8DL(pCXd)Kc=c8*QmORi5fkwB1(1*zVsx4Ni-
z3|eh1--%LM_AnCX1VI(^ZLw2#_@B!`KzS6(SC_q-t=Pg*Qyu1wH-@0et1<I`B+=}F
zz9Nxx(~luait%hIV-}sy<3@S>>|=wVj5VJHZlXUktDOtp<q_4VHr0_;q@$$umUR|T
zowth*1^k|<JSddUZc9inzIZ!aSp`|8Wo4o>iSN&Q?u2j8{BmwiAMd?HE&~B_y91*r
z)#Z)d95XLl52#K)3%O%fUuTXn<<*a<AOA^E_TRx<h?oYNg}g!M`pm5kZszErJ(C%y
z{>OhAyx$DCZYc#OX_G{xT)rVHjB4tzv>J*g49U^qd@pCvCZ7+QjWpc8hm%m_4G7E#
zAmyHG>P*UJNZjsvbu)nihi&n<96rx(PS7Tuga2K?7`pbmo9%DFvonW0a(<=WhL0dZ
zU8{tGL^AjQA;wh69yJ*RI4O-3O+>wn%WmiKr8~qru=CL@cJFmM8z(#aWh^}w%(l}U
z(T@kGm!Zl<Oqf)ySE9>WqJY{}j4j8If`=%Vu9kkW?w27^qCIGvW41_+U{iT-CEBAd
zbI&Tv+C*bEB&tr=VuK(zheCqkUlzkA@q-b##G?F>FuMR)%tzp2R3yZbB8%$k65&oG
z$CV+v<PMPfedlkRfHXW0mx@2ToI!n%&HF+sd;Jkhtk`zw5>{0e{k4SM9%&DkI76nQ
zG5HcM=EpZ0QW75HAkK;zii3?&J+i}JA))Uy<*W`9$y?gmLr}`XVEMnrV~zXYP-Ohr
zpFd9dVH?{7>6qcKk|k=e(YjGLUKS8qcUn|8cS6(+z$U>(#!2V`?e5aRL{E2qXxGGI
z09#jGb8lagU0ruhXQvOv=G~CK=$(Y*yR>0qVV@O{nqbm?`tXIXCf?Un#UQ`%ibGOa
z6(K47gj1(L0SBGl?2~K0c^m|U%2fGUV`^3ZF8*eB4+P%cxS9&J`tHOz5alqRXL3e}
z@uQbX5;4lbt<b;@TR_6!ZgKjQvT|#iOj~qHKP<}yQfvaCQf5j`GnrO~bH)i{$T%r?
zOETrrUTCe`-2;tD=UoV7EKexO<l-dAXzOrHLWVduXJ=d})8z(V2RlXz=6evO+IbWs
znypptLJABv7eJx@uhN;%<#Doas<<1jL$^hyJ(;_!@ywFu5hvI4Y#54gY=JQ2xci5q
z-1$<_&I8KY+`*gRu>77FfVh@vGsBE->?JgXAwbj%5$HHnPL?0hy=(51O+Fo{Ug7gt
z$<MoAhPOY{h?WAom$J|oJo88JbQj`I5EC4X+pqDDMzWmR#21$MT=CK}D627lnT0z#
zSHMm<1zN%gEOM`pxIv&a8I?MD4Y}a(<BoqKujrc&pWsY_k!2Fez{i6O(I9jGZH6E?
zylmVGWttkMa?7GBegW1^S$b?K1mW}Q?SFT-T8}e-+avnxv>xs&di%#T>NoAc!~dnU
z1`COHC%>leil^*WLz7T-Y0$SO$8qEx5DO~2UiAFm8^Oc--}3~;<!I2SO<isoyqW|;
zl3s~U3+dFugXI?vhb|DtyyG1V+Ah){Zyz$^pn1Yx1nX_Yx!&<Yesy9Mat3zHb!Vn%
zl%Nl|syXFF(&+9&p;%_y!aO1%!rp7k0{^Q||8zns@1?9y{1Li<5uT=jq6=-hjI6V2
zN0Mk%&V%zmOztY{7C)^haT!ssr_!ZN{6G==78p|%BZo5ve3;fQrTQ0zruji{<vrpU
z@`9r&9|j3sW|y9|P1EnSh<Dw=z1@>tH*&7W2-PQH7HRSXv#q>M#b4Lx)91hP>TW*)
zaO8@K({*Vgw5w))moVH-d@v@`Xvx;^P$E%NG)WB9v(%|rBLtWYf=~DsBcd#jHUc0k
z4MuKD97lf1a29grFvYL3tIhT=v#xH|A?)hwK5WEG1AI`@ne)KV(n-<$VBsu?tfe*Q
z5%9$Wvlok~D-l7rEdr(B1_MviR*Qh}fE|zSAEj&FN<wpZG2>p-H^CX}!EIjvHcT{Y
zvunO`TK)0pD;6D$1}-9QjTe`cFP3F00C0XGZLm}z6uA2<s<~7JoT7!XeUhX3=u=Bg
z_34+cIJyXm&X+C}k=Ji*dTyi-lrbAGj%Zt2tQj|M15cU~IXx3$C&|1%gFC58q(rrH
zf<oHAwJa4`24`{a%uou4W92weMBf|XyIY!b(WnvRFfl77AybM3pYqP&8=0VrQDqev
zY@a_-SS&ogXYH|Iux2x)N>^;GtH(tArX?ioeg1`md%B&{|L99k<5dJG-Y3?bf2OU#
z?!zLxRXHnc12|a=2_#&%z^G{nHmRp2EyB}m)`rDkTFcP3&}{UzCVj)ob45-166T@`
zez=@qvDAKB?txgVu0TrvR5spU4+`E~s?2A9SY2I7ii*roT%gY&5`O-CS|220O;50T
zcgBAGV=VXANuIDv{Js>j(()=t@{ihzI9sl%@YLOPugxh98I=GP(YsAQx4|ajY-5@#
z=6@pI2s*O&1J>*1XN*l^J&to+EXlCwKxtolQg6XuYv$1KGVc#FXjlV5x0r>bIS&Km
zCWqTaI(^zZ6ESDL8-7-?-Arm95bCgsCfDu)Q&0-~-p$+Aoj+vd6qx#&#lGF2Qht5Y
zP`QXab=-X(MJas!cPJ5b3wdkv789kxKAb;WC!(ORa<R<NWJ;)JGVJx^5pvMOQyOSv
z!IPrKdf$fcxNIRM_5lE#D4DZy-h#(;IW*exUbB>Ejf_J*$?KPqqNkTT;vj-`PWgS$
z#Zb3GT=@y>FrHXVmFxUd%0OuTG<{>?`sum9gu8w8>KQH;$jQqq+LRm>9SQNhKR-!u
z@!;3i!T%R2*}P^78{~@cW74}6_3w^N2xk}Y88XkQh?6v&G-t2z#Hh^pxMRXYz`9(d
z_`8i<DYB`RW<SQ@=x{*wTG29innIzoIEqQ=X-bkuq#KE)$5?T5KY5akDA`bw=|Yml
zF>=cF0{%jAspi&}A|QU1)v>@2Fl)fz<m5BjpP}(4^yr4hW$0&?G#g+nR#8Yx`L`LP
z`M=dj)7AUuwg(!jPWpMjKF8A#+v|#qcJ83nKg3j9V~Zy%hf7b5N#?~|eo`(Jnn`;q
z^e@9a`=Sz@RDE;}U<Ijqn1=Y<#a^bEr5+QJwmG6TbBhI&u%k@AtI;lLZ6%e6!}9)^
zxFE^uZr0ik7Q8{oZL0-!oY!)3vhccTC<fi6b)YpHHkD<nO>76VIP=frF?(1YW@gnd
ze|iw^#SD3;hj7#z-f7CQKnA{-{_U}1JZ$nZHRE<YihX{6LXNJYxMLD-m3PLQN5nsI
zanH;0AlQ3&HFCPu|99f{2h)k$+O}=&h|%2!l-6!2ki8>htm*+-tZj7zmd4$BbNg;9
zb!!-ipJ1C_g1H6a{}m~NH`DaNFs{u*NVY)KTllgv#N>!-HGq7<jZ-os>Ug~w?fB~S
zn`b?Uxpr1CAPcwf4UqpWJ3p7Z;opdly+ue4^DKIp6oct`O(6~CToqvr%FIq&MSwn6
zc2;$BJLm9r*=bcLg|)`%!vHSG+h!|dDUkVIY^Uc?V(mHw5GUGyn*jgsZ)bvS+0egd
zE(R4!Yoj;c;w~{6yC=77UTtPZaarEPUynej7!GgU1bmX<?~<p2KqcfSw)99ajiB>~
z_3s6cc$-*p;oPE<=G^HzR*IlqcH1{alFGS;`=_91(#Pd`aE8HOF&E1S%%J;0!Q0i6
z_;{6<eu#ij-zX<2k)3xI6tiHx+U&>U=V>A@V;~^z>-lc?7o8!<Wgvq|N@A3iFmeV4
zkn+zSlu7J8X<>lLBxvDOG|WSEAGA<0Y&CrGl$P|Wa?odZc)9e4QMS+gkXPAPL(55F
z=?gTd8&Rw@q|3tw8(wQA8J$Iw+KKzp0hwA{Y)uvzn+y=t6C1)=l}p>*qv>H`I{mW?
z;8epILzuIHoA8LCYaVsA>5l_YYZ6a=MuCGF7i73nQ9Oe8-|471=hEv*7@b9gT<%a4
z(DG;gbw(=y(2`=04>RSE2Fj8n<RpJKVdTIy?cxrCQQ1dvRnE;%+gPq^Pj`AD{<*GI
zBhaA4VYf8b1nDsST50;U-}X79qq-SVC3CtCy%=HL7gm>>czvg6itAvF!Ur;|!6KNQ
z!h++WYfK$ctbcWn%z*nwqh=WvU$pby9SfK_yI%J8sq&3e%#K>!ytDeFemZ6uN4D_>
zfIP$$0xsZ3R463sbuM;Y3@7MG7GkWLJIcdKPb6_a?bZ6(be^0602n$+s;5U%-ojAK
z&Ecy#Yn#*ay?C4*?<ETh_l$9M>@8>&O3Z%0jNpnZ6>WGpK%s-9GDE_naa_K0JO%vZ
zbC}eVRQNP(vWHuE&--?9snTEw?CrOPzLkH|S5%HtSM>6KaU5Y_=)A;mqRO4dL3n>B
z^gQYo^yb@-DVV4t5=)s)m+i;UbLyK#PJZxQGzf^iO>;;nW@M_fFCH$Q#5=ujljEvR
zm9tWFJ(FxV9$v0#{=O@Bv?cm!Ar$PxHA7)`tx!lsaDVM|wBKFpGJtyV$bVU_Ri4T8
zoe94@_KPyIurd`bZL-Dee(>LT$xp<?%``)KNI_mJn`C79`uMIvJN{?Q=>)-O=D3wz
zBdt1yAvA;kHsEr394rSP#flqwd$ADeNF7@c?*l6hj_N<c7>zTp5>XE)JiRYMAp;~+
zpQ~QGCDFq$U;a43gyd(yW>3#((|~}}uXLG0@8>7&X-rc`&j^v1G?7x6pBsn$zF*~4
zP~-*@m@0A)7I_NE<)MagX7bBxRkvYh5;JYJ>vCnOymfK^<QQYSALDOQN}Z|1QN*eY
z_d{gvV52r=skWab618!B9NI~8^d}BAJ*v>9Ya1-4^1`gj2B#)DWvxyZg~%HpcsbZ+
zcLvwp>#k@7-5>j>FAWKoty1+c;)XAVL;}`CY4k{%w>ZqG^DT#73=8HhDK@TFivwki
zx2M^kQdd<C{i7k*4|fNKmXa`f^Cm`sJ7iPkB@@hsa8W+M=@tE{1jjL|Pm0D^+k9u*
z=Hr^AH{On*Q`eHv*7~OpuO(|nLKG6}P5l;q@r?3&Y@vdPl#~`{E(Jfvw<qMZ*V%ce
ziz35Py?&)~#ct8o!$MDtz%*w{%VfZGDVMc{%J()B*Vx<i7o42-H@k_{DT|Iuh9$dq
zB3On*`SkgUuFuz4w-Xk>n?^9pZP{RgoX(MA9ay9feE87n^};{p8&O0XMfB^<v2H>t
z+2VsHBe%szL5~b?#ReA%5teN2<<0Al#Q~(9p99}d*v@|)$`ZII8pITopG`f?i+0;x
z<%KGIai=bd4sHro->%Q-E$I0h=DhtIjmN^k^WPq<_ZYMD>W>}c3Nws^nQXx^1$lj?
zbUyo{11%J}rk?(ow1Z6e=F?~&#c)Sy<>{Env*`@qs-IR-$Oqd4)rxTBQFQi4y`Bbv
zh5Xne%JjNkNHRTIsY2haJD76Y_XfJ=q*-R>pEmtloez;cAnw?Iv0(d^vfug1>X92;
z#Y#>(Z4yS#n>~ER9<wX<st2>HfH&mxJCEs@(n&)0cWHD#{zw!}Et{nz4Wa4B1RTh7
zGs~3gIfD(oh8=}3CLjR6tpLHxRdY>;JG#tmhZXSsvD<fft-4NEw*Q9zqtnxpigrS3
zg`)Zz%3IA?(e0<A?=dI?33u6UhY#aMcY-{h8%Mk@O-%98jSe4k9W721&tTrJjSXOo
z<_YpE+QbJUeu3%2*D*?0SmVonkI3sn_o)zMhfU<M{riN-D#5&qA%)+`x7{&7#QuH*
z{{b>qco?zRcE~#^jIw6sJpREINM}-G3}xPnU#qjr-}+Q}!;hk}_XY0<q%*Xv6u3cz
zf*Yb%=JMqN^`?^Ms~K{p_kiYads+|JN@_Y&aim-|Qd}Zb1#-Ab`>M3qWLqVMtVQD-
zYnmi0UM}Rhe=#<Bb$SwbUMG=b4qZ{UiiAbRhI*s2`Pz~~mqa~{;G}3S;PI4T*Q%;J
zzTOHkF{yP&i{ulX4#x_*ZUw{W9*1md6t>))yMwzF9fcAWyL&7YO%lsHTzcuN2k>$D
zA~e{w2Z@b6nh@@p>c?EAMvacHu`vyW3JqK`O0!p@9axt>(P0Pi{B4QhU8pWsBH+`;
z<8AxG+MKeJ31;E_b%n`1d!KfLLzI-?e!6h{3QL_6r<yPb)OY>K-!GzL>4)nH7EjxV
zDm=!z{bwPQ9ZrVBqs~WpThExr%l`oSZvyf~uofu@c+T&9?@#;Rk*f{ho&ULVsdqtV
z*Uac8^yfXQEmzE@jp`*l$?rQ)iI;Z>==r3(Y_E?h4^>xvhb*u)EbYBv$73!Bht{1B
zEbSL6`T5TvlNbDgKS6sYngnQ){YUrcCcV(p4zY_I4@7VFa4a%mkI0irydYn&cAJ;g
z2mj2jwe5{X)Vq%?R1TXXGtU*2=0`D6_qS<L4@p6{Jf)u9A@c(?CQln8I17mDb=?JJ
zm+T({513c2g4dQO_PSvy285+Z2?RtV*G9TZ1XbLcgzML4J3=Hc@r+Bn55q3R!X6)K
zi<>~vP?<Mqkn_R4+$*T?^;5-kLC{zkCp7{tXY+ST@93Uio+)DoydwS~y}iOXm)=B+
zKCf<HTk3GFJP!y7AD(r3ADf8U`<APCNc;~t3rjWic49yj;MXTgi)$arZbVZ7;eFSz
zwKX}C1d`KbjQOqh#FBjtK@;sZ^&04uQ&3d%;mqaR^W+TH7n&_v#(%YeO4&{#7gQ>g
zf@FU}dd52|f3VUc9yP^-!$>Q23&a<&<EQvF4fDB;ikVAxum`UY8f_4^JU^2vT}Ovh
zAFcMr9r`dbl{U-avBmuSgcY;P%i89Urd&3QK}&}0MFyu-@dH#P{iS#fpw0ygV9uX%
zb7a|C*I9Fxn7k3aR8tinOFp9{=#%qRhgox|RlCj1>FIkpZeP!704Fm#uU<sLG7?>{
zVQG>RgPBGRkm{ENMS3NUG&3uAbg4A4c@#m&uJWQ34P3|>FBd6Bbt}gNbDVcUTwQfX
zokEXs#hQKOp;AUjFx7S^X(}M3yvBaDp+FZnaJ0%878Z8Y^%|t~uC{SxVk%}q;`uhh
zEYQ(fq|aH7ClY9`?~#}^<+f6V*wzE){tIoU+^!Pno6Nopu!(@X(T@=jeo^BWfQ4O~
z{GT-_EwwWk(@F3SW3%+p-5Hn0bwi>Uj5~Zt2G+fM94{}xD$WNRa=o_+Y*2;1XEE3g
z?oPxRv+p_XErwixZNL8Yq#nk+=q;RU{ABzSco!Cw_~V(0W%${^tw<BhL&-l@*yF-U
zy321J$*{(=QHmFHb=r4<=3Y-}dQ=oO7ZBaL=PZ;~A9PH~*8(vPd|eTa*RZ?t3VK+P
z^sgkA(@s?Rtl@=HEcadVyL8VqEQkXgY|DDfZE!X7d9RGsU_N8{JU;V=%=*g}lW>If
zVlgcVYA*8np7qyPwxElrzL`u&2ks$_d~v#MLe!nmSZAY}dJF>mOTdQv3kCojbBJVB
zfPqSok#WB$5|E<a`79yOSI_WFwax6cF-C6kiy0WyEBnk6-rp2Zy+qPxo;K0yJT{$A
zCKPyu_-0h{&GlrGuRyq=u!gmdro|l%4!C{*(?<Ode<SM|Kg@lz?CSWT{2>>{Kg|;@
zvO+@WO<2>Eed0fJ$+R#uJNidv;3X?bTT7Ia=qyCihEWiQ7k+Z8T_U~RWud<=+Y(VS
z7=khjo}0bHQ!V|RX88msp_?Tk!wcPmA6n=Cls*CTSj+Mk&@J_aJd<t>F7^~wAQP@Z
zQ|O3=D2A`#DEPk&$^u}$yPOVXQq^FUi#g0!R0mR|n3lHYohD{p?`xEdZ>h3FpIX{J
z<0KWAn0mjrYK*6!n6bmS9qhjwh}1XIaWO*E`%uBJv@iq>aF>u-(gdzhqlg>*w8jnw
zh86c;j3ScU&}Wb;&omJ%)?T8{Ivc*8k0KJ}Vg!DZK^Ci3?mVA0mYrSSLQb8GfoISq
zDcC9LyD2={a>Fmv54(F_JQKWH#myt9kgnFr&SRzlko#lKT<S1f?35uH!!<bk-muS(
zqVx?n2f+kesiPyuqLQAp|JZP5D|HhOB_j6{fjgP6@HdfTeSii7@UM8Coad_^h6lD^
zejU}EaXH+)tX`M*3o0&htwUp=k6$w;o;M->SqhH6q$+~1DDB;c6~Ch|WBOi*zU*kY
zI?#_;xxgdxW&id9R;QCFZ|_j`9G%F9&m<mq62~s#r&a;mH*OsdtjqZVu6X_GyJPzv
zPy3K|piA`u@C!c2#6%)6&QFdQ94mT~I#YVbS*;T-K?1d;jkQ+&N+&dT>#JwrtBD(v
z$z4YQCLy-_lQB^|Ut1hNj5XIMMg#;7k)g0QQek6xAUHT6?0D*Z;^h&%qGgEjfGOH$
zD2h4*1UQMo>bQc;R46np$2eynvLoRWY3EwTxt@W<Lb0qNdR4otO5xz`<y^^lqKMyt
z*W%oH^WPD3>PjkYE8l4Pq42Rw^fEFG#Cojev<;w&qm2L-(sqcMk8fiBFE4DBgX7hX
zFf`0Po5lk<VwY=4^giNXOLpU*^B()gj=oQ4(J%`SXGN6?8waOwSZ`gX!^+>$L>2>-
zi&<KiKL%(3PM4qiBpF~cw2PNa2tC_76<KdFMgtlEN_Y|j9B`&=*``4gs8Ra`W#|ss
z{aq7)R>5rISOQZ7#(TE^(~|8C5bK%qUrY2$_+pI0gXRA)i7sOm=kKh;0H2T4+=^n-
zd&>HeTv)cyg#K@V6p>3tjpbkW2>}$i7#RwGEbA)ST0;;fK+ztrsKH}JD^EBmRK9gQ
z4$rs`^ZjZ;f6eM*u7XVt%((+DfNcEM(lMIIn96j164)9R+Ukm8gZYdIFY{Ntg|!tk
zMSQJ7?=aeCb7Ne(c|-a51~(c9H)csK8R=;&40M6`Yn_Lvo@bqrGyW&A3ZTRtHWp5e
zYO3(qvQe8Ul5$K8|Doo&KXFtjwPe=T1owj?FG3UnyJCURzURJwn|>Ho?x}NgX!d`*
z!`^NP-=b`Q;|wBCh_L?yZDXw4b)kR@lpE<V1IIv4mQKM$@I_gDKz=hJ>vP=nCJ9?2
z%|uauGfFSSr;^ciH5P#8zk(!pQ06$Fkr5~Od=G{l*%t77AIU5;JQs&{EoLZvS22XS
ziUi4GEeUDqu`?0a2k8mXoPm@|%Z)i-4hXGfF-sY>#w%kb6~8n@NLm>&AB&^8(#Mc{
z%Z9|0&8bF*;cdg*SWVs+ql`&O(@ln67w(8k9WAjOJ+|Qqr0t(Y_y&gViE`U5XKAr%
z0&=rNgvI_u@A4~Qg{J3B$xoH`m4(^&!t)Vgo&E#m85gmo4O|*23*R7Q{_Dh3(oO-0
zF;U2<X^|+Qt$Se0db2@mjo-EEhgFT~jf~JYIv`7Z%mg+4gfs$tDFE5JK;>Fv5VRT(
z)!}*B=|xBQ+C=2KiQ<mLuA7K(oc_&Yi%7@gaMTLN28CWZuV~3xZ)Gh}up=>X%0Rcz
zM_}u5xjn0ZNTzOC?<znLK|Hn6hXSNbIRYgwz&I-C(Naj2T#VlKJx9}beSq_4==U)5
zY?GuB8r$^R8rzOMlcZ4#gz^N1JyCim4^tMYPt;`RfT3eWh;5$jvs>9CKm)uPlOL-D
z)(hnx(Y3j<tpuxWP>w`0;mHM+v%$QALEArBA>k^Fm!)M_hJj=Ig53~dNh<HE_W_l1
z^o;3B6kB%)xEkR97~VLiJm^>3VOu*ofe#YbH#g9ow^!@}e&=Kj>*;2aFUCW^@Vt9x
zF4L@C?4xq)`Vw&~W?XG;j!c|Kmc6NA5fOrfR3+;%t0vjlNqLu;vVBYYbwq%>MA<Jp
z@?%Ys8e?lK&y38jf)JF|pra$qBfYn|UD&VXhNmyPSu5_(5ni1&wLQ5?bIdpThYkfv
z<P|Md)fgM3oKjVN;{iMFF*;m?RB=0Np5CHe8H{k0WVnzznPhm3aJe-%A1j-Prx|==
zfAD&zZ^=wb1%>5rN~sK-$@*11oodB+mCwHC=r*XQ%TB+&$Rq`os&;7Ri)JW7Ktt;J
zUCe3*`Y>V0`BJalk`{i+rHj*LnoWD{&zkgbTmKCfHYQx5d~3-5PgIl3D?;;dEUsLF
zsdA6GH|<nFaSH63fbHHYmrpcnwEnFBF^Qx#GEayDUBgjOn3N|wa;#>T4RfyjYE4?i
zU*X7x6a|?k4-X?IQt(CZJ!zV&<;jy{alwnd<C=nO@hPM#0*l?wmYlX%0>L~Thv!aL
zRVqQKP(-aw#E|3$U{<#AK_iHA+X(Xb4kpbve_No3HKt<t-_$uZX$QXSV4?5|^(c<H
z9?e=h+A6tho-1*+nGCzqfNswOzI9PN75IPO@ASc2v0T0z43v`_zz$9Kjj&D>jsY0^
zir)2fP-u>G4^A__fs@{Xbp^*CF@*e|1Wdr0LB|r;*VoXBH?Ycm4CXW5l73o{+=CF-
zdx=AtY*^d4w$tk<GEzM^-~&e$0R9Euexo5978GhTcY7V+69idy>e8UVp&M1G9`(=C
zaxqy-OUtmG9Ci~dEA?ivP2XZRT_&v1mYMJ55|W7hz)JE1s0UXU+p>ln?)YG7VdS92
z4SgXO36`afh$gm~W8|Px#AH#uEPXdo<6`9-k+j7#OsKNnV&uv)Cma6@b7P`iCdtLQ
z^!HjLDu%-DeF}I|Il*<S=g#$U?^F0im1ZnnCP7L-GF)~)nzK-5%sxN%``dezpgY2#
zDM-Mc7oV?ZG``8s{XLM>k6(@$C${vn`#G@9QjMoZ@&+XbfgJ!SJ@j`lNstW}B3sf=
zuU&7SpPLJT_-Qij>padb>w=$jtGRp*Hh>C=GG8aGT8wn&*Onw&)xdw90>U&v%4&^d
z2X+{EdZw|5KJf~(sq(O9>MJqfJC(vi7AHSu%GXgygE>gq_&JTlmqACKyV(s&#(Tjl
zwp}!e1Mvr)*_=%95!^gVpk(P)y9O31*3|^N?|YQCn@9r*K&}qG(1d!I=OZ(8Epxpk
zlu}joVr&Fn@onFa>9+e%*CT*ZXHmr^o3Z`bRzhP=9;`*u_i+jy2ZtHiIavo|6_gg9
zjyQ6uojW5um#~zmc<Spu=8Y(VrnZ5_^8F6=dv=ME=8`a_-1eq#P^;y76BQXhWR_02
zOI<Rl3{|`s3l5n1A2X-;vk)ifVsseMqaG2a*P~g)tah>L-`Ptf5CZ_1T9rro;kx3q
z7XdW-9l0Dzm@~~{X`i~@!|wk-M-gcHhfB<Ib?c-$)hvAcQ`m$&L*HJu5#=MM{&k%r
zNn1<bHz&QQ6cF<#fW;)`%|<s7j0soSDLNs>`ubNJSGQiGRB9}BpFIxn9whOoKUl!!
zD3MmoZIfquvah0M=#_PM^gmytKQZWF)hPzChE^-s6K}}Q^z;8LSMn&6Wz8MNf6uNb
z9gq(BgByd=Rp)$b$XiUAB*R`yZyl-JGauD9Lh?pvOvT%>u{~2r47$cgpW04w+8SI?
zq4?3QY1Zht^Y|X)?V*(#H-(7uVH9$e@&>sZBh*g@FlJ^*1T<uJyj=;3wc5K}ADO`@
z?uQ-re?H3sDPD?tR=)HqEQKW%6j=XeG?gqaUC=xKb?A8Rw=HHha*zeo(yQW02H2n$
zc%9K%peq&iRb*nmj@MsZ3?^L_cy*&mEEr!LK8bI4yuT@*`L}_pp`k&>=NkMj1Uuns
z;q4rrC}&(h<911ki=fQQi0v%jsHNQlD{iC#7&kmDm61Wf?tmtjp^6{juaHrj+ZK}S
zeV+nDxEE}_KWQD8a*(TOr5nGY=~QU@eB=!j@*#p11jG7@KhQKKmD?P#6aJz$t7Q7Z
zFz|Z;YTxD`tCXZ)CPrQEa(w$<a|S$lkLQ)}ow(U;<N0slet!Gb$lIM?-|%1*<sd2?
zmALx&9FB>Kx>FI|W+F9!|BC(Z8`R1CW%>Qcbb-001-egQY@Ndy!ac%(gw*bdLv86i
zlJ#NSqLl&wjARRtp8HArjVT0OS}+zkG2{^@-x$jwqQYG?pKOUv+|(pFhlY}1YHURL
zAUpJ<ubx~lH7hUA(kw7hl0%F;ZMUl87WSpKkC2F9sq}jl@Qz+wJ3-TQOa}h#eFkf|
zjna^jga#Al$Z(;D-Q;o?lOl;>IUi-i`yYwCAc0iKg2pISsei1se9DN0(bQb4(8=?S
z+&w&cSt*g;2=elcw|PXaK>qh(eUisRVu^~=96W*v`wyC77w3O23PpZMgqC9O7(u%8
zO`cRJzk=!%R5c^EUh}ss-+qV1E4`YMH$z~z&TRc9%fP^FKgbh3&xSYJm}Dq(eRpd9
zdb``jXd~r`pE1}Z>d?o+%?FhQaT4a1s`>__p(!dW!mdK4o;gVrSpe;|yH|5>P;x|s
zkMP~YOfM_Tnz6d--M^UZ{2VH0`}ry|SV8uj;iq{)0qHUAswu?l7CwLlGPYLLGNQr_
zJ@N<%nc+2G2iNqne=JK+?RD#LRsjY0S~CUN%8332K-zw`*JW1+E_u5k6m&jPo1_x`
zO~rCJCi5gqyt&te*mTSbR|tZM)p*QqHyD2Ggc-c|XTLhDlZRTB$g~DMb97U~3*ws>
zowIIf89M<vd*(Z4kHU9sI>cEl3tH{bTscQ=vFktQO1ytmyYIX4w&JkQRD23ywtz!m
zGm3lOxBd|nJe7;YcX2Je%N}9t5hJS!!t6f$UjIf%czb#|c%i>5V>!hZp8%~$!@X{?
znhOms2Wsc;m}I$@4M3i!WoF%B)QbTb;VZN0eDWT~Mu!lX`93>eZ+meJ_upm@;SBWK
zSnIQFb<T6;&xgH9Am!_z$d1yf0bjEw=tgid_|P5ML{?N>dN%+PK-mXH?o**jFaE~;
zKID+kCh`?0iK_p`)Us-N&8>o%Tv+C<OUy!Vx4NeOF$MJ`P;>d&Bn;e<CnTU~ryx+4
zPSdKN087k`r=zw?lVjhHHc8Tx9-c7Q(S4P?hwbIjlnZ<uKGt*}J;r4AFD@=Nf2eL<
zN!qrvwLK{VLjUCSMH<`Ly*~QY);^p<TH6ll`&Qy=I{`y+QD<UUp9KUIII<;u@iEbV
zUcR7ES1#GQdwSAL5H1OKwjGER|AZ|{YW))Q(g<59VtkU*SNg*-;mfkF`Tu#+;UOg4
zt#2G5vHSVIpDwE{SSk(v=fy9}-WUzAiTwY)+>NA*cJlw$eldpsiRke^3t{i6f4>~s
q_5VM5<(-|_=KuZN#u2;VH%i^=iA*;_y_I*chn$qMWSzKa=>Gz)@gD5}

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_008.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_008.png
new file mode 100644
index 0000000000000000000000000000000000000000..804db2505d8f9573a57cf9bb53ba27ec7f7f1e3f
GIT binary patch
literal 222987
zcmcG#^-~<*6Zac@S)AY&7I$}dcM@EK2ZzNixcef(ZIR#>+%0&34;tJScM1ORt-8;x
zx_`j^VRmL~d#28r?bFl!diO+YsL5jhNq`?de85mtkkR_^0p9cdcpiZK{<$LEmGr(7
z1k37ywVkcOUgmCAA5_i3E)LFM2fMG7o>p$|cFs=RZ2WB8ER?oju#3A8JG<lmYry90
zX2U)uA${<E5mXliefJL^u+0Cr!Ig`b*?oZf@Ig^V^0Rl&$%}rSmA>yr<jaDApTM=C
z*ObrCs`|*Iqa&OH<QYmC06dB!gk|*CKgw_^^l<zV7WpO2uBF-3J;`+HXUfPbz>h+X
z&=Ol5JjoauKmqBRZGq<6yz%^(v>$0ZlbzQt`;BGI>7N!>o^sq9Nvfpy2MD%n$_@Fi
zce`LGU3ZVAZ})HA>U#F9uw(z++G~&Qsl^8&Q`m9Pjq>Z;BmDpK?f?EXZJjIZ`TwyD
zl3Hl9@cO@J{m<@}-tzxXBb^J^z~4jzQ&X6yo}{2b@nie{4fpLly*Yixb`YwZk#v<h
zBn`WAe;-pD`v1l=SEWuR{%O4a@AC52(R`(NYOL`8COBd5=a<c_TjO7}A()Z>KJ)!!
zqlowUKJL^eQvc81zUA)+|0l1&)R!*m*Zqxue*RA_L4${fkRLt~lC$WGs4{5gNUkjZ
zhD?3JNn6;iUs?ixZ}owCp2vFzPYIn1`#&(^>k)eBzmbC^qM)`afZ%2V5Tce+*xUUc
zd(r}B9e_nIeul&XeL?F*j3}%QP^h5<`tm&mq@{Gf&LSUt#tq9aZk8hr#lr<r0^m!u
z<~`Cjniue7HH{H8d=?j0j>cce-(GPU@uV2SQlccNcArI|*Z{;?1D13*Ku*osoX&mQ
zKd49dh5oO0_KR9ckHrN?om-X?2m|!_9980<{O+BsV)USKzSd9Q#3F|07sB~wVvD+A
zvua7bpVRD3a>pSe*=}`S`5othsePj;0M35Abde023Uh&9IhaUto)1ThujsV_`xr&O
z?uYtm`Jj`lz*ZZxdfO__v&W`V2DhvJ0<>7Z;g_(5`SYXj`}*Oe62`ZOpi9#)4Ht?!
z55l@^rXXAoWO}4hz#kH=HF}#c2weCNcx1)FZ%<VD4-;rcZS8$Cm)|=+iEOh!hnw!D
zEjkrSg=2?EAOS%2I%aW0diPfKuQ@;pQQyJ0t+#*qDW!HEtU-qx&&vL1?zQf2Zets7
zPd$AtJ=f${B0*>Ei#odhHeMrx(#8qd(?|M0;lxAE+&_u?3=f<-8r&i;)|l))-r5gt
zg%~}Ye~`0p^6W7jOAr<=@=buETiI#7oNWJ`O7&RWF$uNl@H_cL*3j$|%$&jYah}Su
zTm!oPp#;fsV%FUx9j&o|Jzc!fUlafz7nFDfV<}C+sPcHk2yTWq#v>|D9y=s2o1;XR
zlA=Wo`4HA?JL$r+E&iPNO5M+~`*xH6OdqBxLxdPu%PN~!>>;VSTQVBwwei^Df3Y5f
zVjWWlMc38E&?!JuG|>aYtpKc}&2xIqSlV93bRULmpMACs26AatZsU~JzQ&a@a9vSZ
zh~#l8+wj<XG;w~1=UjA*ov!n*zVPiO8ema<5P#cvi_KSxc<t(WKEJXFy4C&H^J@A^
z(sL`^6O9ducOhq_s>6+iv%rzU9f1qlpY9$bHSL^A{ofs9HSxc++HQ^DGmGnS+PYh8
zD5!nD*Fs?qG8zA}Uc8VsH>Zu6h{@SrKoM@#<sWTbyzls3w2^-}Kgb3p_cWPOV$-3o
zcqH**<N5s6f-ZG*mcoDQIv=+2=JiI-x5yv04JcFb@R2p6RZ|x>cC6~d+Y6Y;kM3we
z5=LaG(BP8z`Nvq967_wlA^v;@xjb5#1ML`V$2}(>yBhS4^8)O!ppf!#ZpS!e7O4;j
zd^K8CJ)nnQ3z5j;rAzRy;d+Y%PB@$<UYq?KNvy}ORgxyoUHyG<U><dq$KTys8od;|
z{GFFb5kKF^fi6FqQSLf~KS`zOEci62an>ApQYrGCkB97+Y58y3U|-VBX?dRw==4_)
z&1DgbSfu<xonLg*?!eRC|KJTsfyPNz#M7Slj4FNmk3v|V#J3jT`oyK|yJ*mY{(IC*
zF!tGM&Y}V!62$_gO5(6}S}_KSAd*ydyAs3UXrkR$^Xk_(hnUr#D#hzypl&<D!BlV+
z=3<e>y{3!bp#&>16OXc57J&A?c!#qd=%z9xm9V8iP$^ac&)6xA+is}2(DmbXGm8-T
zi~@aFp(&lnkz?TlJZ<PvFuO^{1#rk%{QyasDx@v7!@CyxJuRjIul%ZO+9aP7XMe&(
zz#{f9``f+M4xe~If7{r+S5<jjHUu}iquXreA0ATs@7$y>N+r+!vAk1~CU|#XgMvBg
z82?BE5hR2*q)OgqiDdaK8orza6p=vh$PAOfwDlujkFI%9Jv2FX#>^vk&iQUZ7&rYi
zuYSGl;By8qk;TkwS9<<3y*ff#&vnlu^PBSPW6<?IQ9fC5b6+O0&aMLKEP!7V5l@H~
zqIy&$S+=$V&2XMsUmG|GQ_fhlYcZsh1V%%G(U8M(ggtC>o=N$gE>`SoFQ<gh|2XhH
zw7gXZ#nuL$9xE^|udMDu4f!8tv-4r+uU#9P|8`5<E^17>`g`sO-(;OO+I{Gj+@AEu
z{$%HG?2em0fAvtC9<x21fK+}~+q+GBGkJ;oU~CIdJ0n+I;(O+hF>P7HjF*2@eR-%G
z&YiAAB^|B@m?%N&pyZE*SzNwU=bCP8+oa{|661Vo5}dU9DKTNNbDQ<%P{OUkGH!q^
z^IuUCrAvDu1MuipC}@xU%_>hMK&etw9`I=s^)SZ+-l3q*T%v$RlL!ZyfU&vJ1G!4a
z_LMwehfEboL9pko6Pt+CWouhj=oFI)kh}90##PtKKZQ+DRZV~we#ccZiY9G>r|cG}
zBlc5b+F+ma`IBh`n}f-gVb2SkZjA$O7_ivbg)(js7%gKC#UgW`yqsaItgO{`R>%`s
z-Wt6NFll=sIn_aPL>dO5PX&9(-TT0-W-1Ju@S!=kzM@X+oriG9C|qVne+<r`Pvkab
z{Ty_FuO%*?F0P7<4>#}sLC<StP<YUdVUKJOM*#y5d$&t$&=dPh_D#ES5rY(CwCvMn
zX{SjvHXIw6pUAQD`s@T^<ooNI8(j-o#vG#pk0BW7ze!tyqB@vj!rH?=Cf`Ot-r;7d
z{DXPA<H@)Y4^L752xaS}$z>`BAxO!mU)Ym?R{EQ3*yr^&o?$^Kd>w)PsxH#6%%a&_
zr^NySc?O>4lgz%+CtFcjMFb>7D$1lR1;vow7pX!F*}U0uemfqLFW|NV6ac7~2+PT~
zX_T=u2QuA9f)#QQf-5i@6GEf-yCIIyxovY)Z@`|t=eO{AKyM`0rvQhChxdiu1_3Uu
z8tqhy-y`KYeVRSrxQ7ds_=`s@BLgz(ZW@C!lY;#`R@PL%c}Qkimgp-L&zBd0ZDM8j
zBr}RWJI%@qGTY>GcnJ^;?od;^9*a+t7tLNMw4DgB22O3~Pm|E!bB;%n<y=+fAQ}Rj
zo}=Nn5Z9RFuGYJ_U=soD178`X%Nj9ZMtWn@9Bhw1Q!dN1st6Z2sS=EBHxxEV&A5Ct
zn1~+%x6ju}T2vt)cHX)j7|?ZqdK_oAlMX`r{|-2PT%%GL<(Px`SoC&^jF6P<@<e=k
z=gJcG*E(hJWaqouxUD4FNlzZSl*_>3?!S<5K|pjLd=H)buZRmrP`}5LoPtDyMQ4-!
zqvqHXk<(O51UFYDb1kW5DIAcjN)?G7nqrn4*jvGCud&IGP~XtGzF~zLAcfv8m6>t(
zIzjw_aP0*pXs^}{43_EAg!L~VCE|YX#f^XqIyyGpf%QDo4)1{MvI?3D|DIy|(y|P4
zQE|)9q;XT(>^fdf<fUadx^<nP3!oSgk~jj{$5R@8%`Ux^QMi*Lc*t?ea~#q?^6Eq%
zF<;y32`!2N>Zasah}}C_ENl=4W!uWvziahm(=OvkQH4oD%F#b3<nxjY8oi_+>=BB+
z`Ix*-*wnR49K)N-gpYn+G+)cBV7P<JK(t?VBo1Ckr{X7Kt)E)>vY2M4z*R>@lW=^t
z!c^w_YMkSv(+`Y1VhH6eK#(rywM$R&gQ+SV=9l?Zx>b(-IkKtG6Q#I(IS=?f)=r83
zZ6XFyIVOo9c-&tPZCEcvW9}O-AOyLRnLKNfO`YQ0?S4SG)Bg#dWRh;h#&@-mVBW__
zWWXl=P9IhStmzxy`{Ypz&lrl-^zx9Orgar-`bH9TWG^o6l?kQMh@k}Rw<l=v!SqpC
z^a#EqS90oPM3OmjHGtuzu*-$CgmyI^&ZSVpQs#Z|hX1O0U0;0s9@aR1I@-WiljOHU
zCX_io>mNj@DG&QEZK|?E9%iXj0gw*Yf_H3cv1nD9g%`5BSzVrMe+>PLnVV|KO*aZT
zbWq8gtkZe&3zvX1kHQx8ps~Y{<5o?X`1AUMB@gw!Z9O<+%-CzIRuo&!_}^0_d>Fch
zf!~*1Eo`4W_8j*6N^_#C*;B(bi+k4vdbblbwP7<1-CE;aX48P&mzRK2Va!~hbYHb$
zhWDv9f1AfKd`3<6jIE5NY%q|=@5zscw6dzES|=(uG&GkDe|>2pK0>`_#i!{)GeQbz
zQ>Sd+koMICY`rg3>5q+hnKVD32mZ4<Eq)%IZ@fV5hPm!BtUww+6B{vA*(7`OTYeZ`
zIS*#*7}XzT-58z>Hyg(0)_`k)ftlv2-X_IYY2UOLG%d(u6JP%}AFgV?)!eqloZNdP
z7LNpSO`L>&OoAUGb7U-Tc(_hY<In|SN7s`u_xcvwfv3z|ChB@3u`7CTwq~gUYiSC?
zq;2R&GqKLsw)RgK^){Ano_Kn64(CQ<6|$e;VXJNT^|*K;94svZz9Sp?oSfnaZ#=Yj
zuDXpgyShQAq%y99;t!L~T%>paDR^Q*%dt$ZRhlTE(|Y^wv2WR_bD|9_^ctA@t`!kJ
zH#R3^7zQGhdmDz;1i5N3>b)`}bG(Lzh8SKpRZ(^6wEV_Qu3TX%CC0K%g1dt%&gWvC
z2SI{ZzBcL1W@!~>$lvtRZ-U<$ui{?TP?|^t`@hKzd<>hc2l(3aw|eiOQErdbUw8By
znT3g4yrOwcb;0)b{PY9K9zha%Fog`{GT^p~m1P)5l#<h!I(x1-2VE~y_=Vf&qKyRu
zw@=S3!YQA9XUt~kByo!z!kT2$SQ%o%s6I~p`!IHlW@qF%WT2crxwIZQ$ejNwvbG-M
zJ{*^!-17zAD9Mf$_i$Ei_$!AV2=((OE?dQ1xNISF+0@!`4v7n&-t$gPpt!jZW2Op<
zg60U9-0vI);4?QL1Jss>+Qjdc!1Q&jOh8nSR+Yz)aIwwXH5TPQ-}Qc=!>GqMUad^I
z0D_k0f28h*IGwvSOT_-3M)|#d^Vt!Xv|$m5#Lf3LPG4*cVzY9%oIVpt7uO9ep=}{C
zZ}QWD5L3Yf@O;KDXSgXgf-?Fx5t*e-d`_>8I=w$K9w(3#(r{_uKN=awV!OLDNdl3E
z5ok+Y65{kcW0X0}X4qQAJM!5Pie}f-%jcih*8~TIMHfhd8sQnGAt?-Pao$t;i_eeo
zJVaK1!siZ2c0yPjiLpF=o-ZQF&bp@$FO6sYKQYE0Y3l1^mjgokZB@xo`+FN&S|&qL
z?rptvE9Q^KyQ4486TuB2$365=55L3wd^5F=GwtOe`9l(wtK-as?AXX~KAQf6ZN7m4
zR3m5SuR(!{qk4139x-Kq*JMslH4ZqLII3!HmsSmm=TSv=4X8EJ<r))cp&R1;N7;zx
zyK^6DQAXOV)kY0J?X?T^39*|UMpR!TzHH<+FyKdN|Gv)liX4U?`xIrCSEC8Fuq2#{
ztj04K3dT-p+c_w9LI-$NFOZ-|NZ}EI>M7(1D44wDltbrtAL1@;do7o2oS)!OP=qOY
z2-zI!>f6TNA%R98Z~+<7HQS(AnW#k)IbrrIg4ehd0;-OESQ-SN3snfea8j`Y6BMa-
zYF)muCFK<t&&cj?x@@%u?5Ve7kC@2@<6t6t0hK1>v-u6CKDryyi3TB}=@`Z#F{4KS
zP!IeSf-a&Iuh4^$S+VJ4mL>$vbP8C)hH;f8i+I%Y*I_rdiIbBUksqjfzm3#NKRls?
z4SqXUAlVVu{R-{zD@0|$Xmy3nQ2Wj@e9{wO8<{TR{uA`X9~3EkPX3(?SDsYCH3py9
zwFCg4p;RY<-5bkpyo+Y~i0+EV{UeSMXzWT6|BHc^Y1dG1FTM*_3D6|Aw9n91bc)6u
zcGVPHpTnnJgo0IaG|>iHx!C4~z4t2M6NS7V7&R{vx2bb6ruKu?f=emOdJSNDfKPFD
zL3H5|M6RkSA5F{Ew@tsnEx<>f%s4qB$MVHvTWuHLhVjR6PEF<-bg2iC+_B6#$to`Z
zw$;Hd4xgd96<Qw_1#ev@S4%7Q=k!jVtpQb-tsVsUJWUkWY<rqBNLi;>KS9kmT<~R7
zlSw@Ls4)&-Wu%PkWL0(<KHNh{LY^p9u-Fjb^#C7Nlwxkhs`Nc2tHW!3PdonwzN?wH
z*DP7g>_RP*A(<q{gcs`MD)d({8f0GHg<#H63LYkRD<{OB&X@iEpTl7JA1-AcrT;$s
zDXFS*tn2<Yb@VMGHbbY)rBJm+w=8G{7JBZ)O{VIzgI0TSIK57nfbc8-G|37n<NfaT
z?-NK8eZ{f0kuuu+Nf2WC2@!&sAe-aEhH2TG4w&%IovdOJ8&@LeTGM|<2krOG$ZLig
z{i+nPg0Cv#ATU?GUDshi%(`EH37}Ygp*Xefm8%_LnS8kXyRmP3&l$f!qFdY5d-RCR
z<v;~^Y11Y$=d_kfkDC#a1>qRrhH+|k#eTs_cfg!Z_wuA`WizcL`x+}BVHK`I+@ZjJ
zb?@X_Ud;Jw6Ohza*QfQH>BXcdV9!K%nGa1blmpieA!Qlo{_L3z>-?GE;RMlw3&L_i
zY<Z5`YO~{@C&^GyY9?@cUA^%Tsb}#XolQ$;PFmLp_IQ!6pVDxvvYtTm*UZ6$!JOec
z_4~{4E2b+9UZ;knhMia2j_W>e2x3!@G6`$vK`@G)!oIPOajL{Q-@+UzW&13Pe4d`L
zmIA>O!`2g(`CVg#JvV|vYb4DEvDK5~yFWm=bUgMLiGNVzJjUkNl0qeORWGIhDL$Q3
z{i+#1v-X1U_udF*e$H07;<%#PWH3Y;KZXf1)&IMHIJOF_Z(T*hMX5zQB%7jACr(7c
z^N}NBQcJ*U4LZk+jO?hYzG-!{%T*>cEuFK|tdyr2g}jc!`E*aSjAl-_0V|i5+WTYK
zyX0EB4xgDpGimj1CXm}0LUV#5(6hdQ$ca^nz((p-ft@4MHaL1vfwA5ihk~$G<&v!8
zJGTg$GbGe3urp_XK4~~vCJh4{KE|Q{@RGt-l@hO?2^|i1lI*9~Vr7X8aMEQ7HkvBG
z+WCUp>^X_r)tq4QS&V)^gtj0CB4g+8TDtaCW1SO6tEgF0;gpKKaK_+sXJWqR+KX9s
zYG~AqWj~k-AcHl)=_<N=6^;?+Viz7C|3&O=U-?FdRF64x61TrVEoF)*0XZj!3?~A_
z@7yiXI8!3?tJ!1OH?qNp_S>p)`k|LI?kcm?XB(7y7YBkkbs{+#&B3oERx?Ww`<Vll
zz}e?Gh%+nG4|hL@`UOV!=>C`#1g1Yjkghd(1)fTK!ldQ|dTqiJhaibsi#DI@FIO~s
z%PclLq{XpePA!T;*)FYKiPe>ylK)AK{W%T(okMN)((EV5RQtuOz8|FJ{(7MgF6GHV
zyAKCf)KF!kXozjj4z22asnC<Th2f?#1hE6)t57+`n)RXH)v~0&8z6q?0jYg6<nYO$
z_#wQha>CMHyngDC))iz^nj}~O+I>#i_hoRuH~8osjTtpwiA@V#O>Heq{a_8Q;6kIs
zXL3>V(%i)h|MB->$nlX_z|-_`Oj(5v;cP8?=3Mx(pJntQU#)%!GFo6JB0{UU7<Dgw
z7Em{N7YZj?egMvMc^W!KN8zSe50XpP*YOjrYDOFbu>j*|3oy{4Mpv;7Yv~DG_{z=+
zcoJ{9T<%sgrgSQpja}D`R+Y+}Uhf+eY&oAmu>51TnU^?C#09Ewth^T=t^O<Uli1(H
z^}RbZ-6-kkDR;hwznv=vphXU4<Q%fjp?^GdQNDxKtiJZTNlLDD_n^OtfrAdK^s3go
z(L+byJzGX5|K@Kmj5||VcDf{9KP`x97_&%$!6@~EK^{F8zptC5lyDk^gy3UwtuDiQ
zv-*$Y3RYL!SX;(1cV}OpB)*t=KmKz(3%U?Dn?;a1<d*3d^|E()?7RNb%-G510t%!-
zjC|_we-hrXyzWOb@xSGU$jabqiq_JOxS5-VWxB(AXeRBU+GVA{WYhXSbY|h`p_<!1
zgy%mUiIa{v{h3CU$zHacuoMER!jz^-`oL9;){IZW)G7XB?<^b_F(mR6MO}}0$=|aS
zy6D)nw3|c3&|i6*_I*$RB@3E>{|+#M{UR_GaVfS3-RH+7<YSn=v1GV!!>jX|%Arrm
zC8r^L?z&1Xp5zM9^<5upALCM<+S<ZY&s((-5cn!|#!ZX?kXCG2+mBaY+Z{_&ZW2;0
zVxU8(j^o3VMX^WUnL$N~j6`n~+Zlz-70Zis-V^ZG2BuF|S2~z<A~O{+<Zu#(S7^wo
zzui7H8WA;QQiye6#oCR#rD+yfE!PWo1lc@%fi?&|Ef2TaZa^udRq~SP5K2>}rOY*T
zKuCD~1GiBzE{>ux9|4Lf`yxJ9QJ3BdU32cM*xF)G4~_brrCNW+o~nNS3X|G`G=D~}
z;MEc?!!!c*$tU-}3D|qR5f?2LQF#4s2rB@L^4zGGTJBgtfN<lNfeDQ0mB$`6YpYb~
z7PA#r&d!mS-ZCi(^$XYxL-y=lF+hABfikrxyl7@e9BB6?3^!3JZ9&2}Qr9Q5A1Yim
z(|RQO1xW&0Xu3K+w5X)jw6Ic4VD+a?k5|7`1g4-T%O+psfO<`mvK)24;TKz$C~n3j
zwAH_7=zr8`3e|DGJ~@<5nZe{2_rL(5|Cox(f@8kLEOoKfE1TK0eGjzq9Y&pxB)T=-
zVfb)@z<6Zy`W33G+u$XCR5)b$Z)|O@*Zq-gflRc}5)VAt2+ni{w{~P^=4Gy7opVOn
z=sU>pdj2hv<p6YbZJmT!1obb}B@ye>^=_XRj&;bO84uoY(s8HT!I(^BGl4N62HqM~
z#a+^gaxo;WS(VMS`-}|Vf^5uGF!h?3KkMu<odL6d<`aC6s}3aMkLzdXXSK6<YmwZS
z%kYufY<341f49(=(15{IP_S+CA&8zHVb+!0u8auhvSPYRqM86_xP!(=Nkt}|u*7aS
zalL!`a_hT#*A2Yk@i7l6FV-eW2?qm>oI+gMbqI|Y7r}Q#Z6sy<gZ;d`4B>FTaQXlC
z500r)c(JMo(U$h_kJRK39i}z&_aLm@9~{RT>F?|M8@dC7tt&&ZQwEAf*I!>-Y!^VE
zLk=C3gp`S}MpgqWM#Ezb|2rlJn~$b2c66w>U2l?WRU6^xixuJwg>m%)WwG^BShjX|
zIR{!2%XtlqS|6FliJ-N=Z-cYzY_vKSbndl|zBr1MeP+?6oW7T(N?*kvA^-1F@>+iJ
z#dtvkw-T<l%5vZp0$7X+$e+wNvQl{bvw#{%o|L_Mx#yTLA5w&E89qd^h3#ry!F%*y
zmMR2jggcpDFe-9U%@j<?Aa#Arb3jF}sAalZuQx%J#6*Q|_2pvm6r9(yoL?uFY>`3p
zab*(OiZ<^vcO9{gI$U&NI5uk>YGxcuqheou=Ed4Z*T7wBfK!)C^l?Uw%k3C#$qjxr
zE*={$?}{j1nGFo5ne|hQX;B<Fys{)YkUJjHLM-c~eGnCQn;J90`wY!!1yOwy+^>gQ
zsH~Rmg<L(f%dMu4<@5ZH?eE`v;zad&I{D<;Z@g`-%~Sis$hn@`5(K*scwuxg@F5(%
zM3N?!|Mk;NdJhX^N=;V8Y`F|32HiZZ#%}D*&XXGT+yc99W6kjO_9i@u3zf^sIMc~F
zGv(c5XC&LYVxK<ClA=BTJR<gne(pZs&I^QCz+U?VHGt8w=I)aYo=PcX!N0KqRSb2K
z!Lnb0!NWzTWIsIW8_b=ZGv;0yNeOG)kFgW{O7YNeE}zL*q*n1q{|blE+9)Z+D!UTu
z?Jzx#<-h6UsgqHh{;_Ye<@F1bjMIXTnLurIY%RhCWvWsKKHsq|I@P$G4imU8ZM%5(
zZ@pUFd8%isX)i1oIMi5~;IvONa^DO?SlynT^trvH$M6vPh26X7rP|cH^@arP8>H}+
zU^@;T1vijzS@5qa3HRXt+%7fj%5zxK)Z&MB8GW+2b1|3Kd?sFBrLVxhp_6<MF_r?W
zdrv)YSd4L0kaYI02d)Ce6h>E8OVM@9KjKv5xN+&3QaDPZmzN(s^VTiAZ)lymE`+$M
z9-eM~=QL;bQ+|o2f=I@B#N^4A^syDNjOPqrQS1m%p-Z>a>aY=C$qselQKn7h${WOa
z;9(NCrc24$tC*8Le&gMl;ne3caA83=Ac;${t*#4Z$JHj#2E_cAL=&P&M{JeeR2OER
z%I<#&YjfHA=~V^ub>jpWbuK_*dSjhQddDF>5)5%jUzVWn$crma<=@|+d;Zs@2JIRa
z=1jGZ5uuz~h=YAV9TmnK)E_!LMXD+|ACXjbbPQGlFNNmpLgrnL@9^_Pg?HC;jRRkq
zYcHYfom2b!S!jUpzM^EEi2w*t&>aTVO0mLO-wux!JjXcygBzytPYq+?uo~Axh?J6D
z=)YLWT@gyRaP0mGNz9A(n%_Rh9dDUl>kn*9NjzD`zFim)bOEt-3n9T$lSXuPtSG6W
z;bEk>UnTkpS}{exe`~2G+m}R@qu@x<C%Ap&w{7_g))7K~3`j^+GbUwC<|aD1n8NB&
zw0Yrj{jYp>e(G{j?7EN%y!v=xL}?6{dHdSr5-w<d@G~Hz!EU}($$uy9oBG(#9k=Q9
zpmUX<tNCAOS#<;{erk}DtP$4R<vY;T$$lz!Uv#5c%tViKCEGVLY{Kb&`haL2&)@U!
z+Nj<4V@V~$h?p}q!AsJW-$N&FNdEzMdtVWR1|VGm<XF+k1rm6ih&#SBfGMr-zfnVf
zlg`02!uJF-4y)A_Nn?l6X!mXPWv=CFQv?c_iD<1(=3lj`pI%&@8uF}Q#Hj};)N;gw
zt$M9|8WGGhDg`PruHQaF4skOQQsJuv?EQ^QOe!`7M{1tiU=L^}nG^c9i)9+mylIw4
zEI+PxU!VNX{WhqkM;sh$-0T|1kG>^R)*)(X;APDeOnY;fS-6KfxyMs#?D|ABBkBE)
z{cmrznzn{tVK#?FG5Ry$HiYGq%iAlK_th?Dw&9}pyVPt~Ya~(r0c32-2*N`S#-`o6
zc~}#!^zPG@9DN9!Nb6LdXX%m-pAc35&Hh#`#1H)rJk~^nxg3a?3y4S@Lo|jF8-f?O
zt0tqiG213b8zp`4J}4P54VdQ_7bBSbK{EgB_2$4+;~%K+%0>dEMVTaOV)6O55ZUu<
z7oH_IK0n+dV~=rIKwf{0=s4G{ihF%QScB>_o=+%NVCQ5;;2F2b-?R6s%{Sv!QIado
zwpw3be2*~bb^O2W56mzbiwi&szi@=xG+~}PF{2@xZHDzl#F$brJ{;2pzBg{z9}`f<
zlR*mP)O|JoYMD?rnD0koc{sHZw>+iN*OmoAtn-)5k^2qUg<0h5CIAG~Y4_k;@SUI|
z)<UiVXY-J%W=-<tdZID%yu2pe#<TMM`lXP92ts(ImoFK`rc$P^;{LHPIvet-=f%%K
z+_!h^kwDYvo$7zP6>4r(tL%sX*eek0Bqz(7%O>cxnzU_hrrTIt`e6$ggh{a4RRJGN
zBpe73#v5&kmST2};dK`CC-v^nH1R+K*r~#&4DT+7D7myVJLXvOxnwlyb(AeMRXe7?
z4X-VjCDMqQ&9rS8jlEk2Z}CO+*%GCubEHvd&HzoWU})Q<9A>GmH2UOqUf<aAhDy8t
zaJzQCy7R7nB^{P47USSpf+U@Y=|S3S@`pkw9U<p|umG{QAw!-lRk8cw%Ku1tYfO9k
zH~s~^(PbYSzqXma`mmWSyG4+k&J|F^_k9X&$e``s&bfL(i2CVVN?-aY>MG6*l_-)x
zvd!(VVWN@bknhVlox9!$hKBHPbO)I|sF1~Y2?^|>IKRP-$oqY#O~6{-@C!-7wQ#Fq
zWw^qYja7Tsf6x!B?ld%2Qc4EvWfaD}`6R-S2-%#uh*W;3K;#-Fy`bU0h@0gk8u3#J
zi4SQC8MN-a#CK<MTUuYPbVWB-4$Vqre}9nNyotxB3@C#$#XHFP5{k=KOrRv0R#AW5
zxHDCFjWNc(cQ>#caNp>DBoU3wHdcX+DT`qdTr_&L^5xF+tRQ^&CkfR7l?G7)BTzY=
zU^iO&jr?sRSbe2+vvT>!?KnK6V)1Xnh@*0X_?Lpt%(Z_Eq@(gn8jMJiL^xsf@)odL
z`20cS0IrcKfNoJGJp4T6_=v(^iKTXGu`GBCw-WSBHJ<b53yRdy%u#^Dj2V}hx!)7<
z+%m+o_At~#o#GBIxk5{JJu9D6Y0nLi=FDQRtmoeyZIheYz*&!dt)!QEzbu6I*6s<i
z8pnP1fZ+H``236G-1Qkw4E#s1@v`rnREnVEive+{um7c1O_5rnt)?y?H}bKHud1fX
z%^TdWVlR}-R;?2&;w9=(#ar@#hc8*COTUP!$VWYmRyrCbWm7T=VW?_+arxaEq~N#1
zM96LmyGkQPG)J&U*tig|SO<GtYNX9*{~%z(D!gIkRx*(prFn^Z;iql5PA!^%C5u?Z
zyYs~PdD`(=i3s%>1Oxf){%gscNC3!=m5Du}<&foHP2C5jQjOzib<kUruR{ol>FMcr
zWhJZkv(Zt1ej96Z7mOUr5AE-RZX?!E&6l}zqWuvS@9?D8=S1EbLu^vy7%udP>V=}H
zZjt#HMt>Q+8^HOrz6?^vdzwb4b8hRFr?RqW9*QT`#UAr6t0|#`l+&T%Ay1RL^Y0~T
z2Kib2+lACR-;g7x!5OS8A8WFzM;3@EkN@_<4g*vexf>+%*%s#(ew-C$+M;?YQ~z|)
z@;V%(njCM(`;{afx!O0>vsg_Ovtp6-#+RAM;>V;lO=`XL<4)505y@H_e(4;eD_LWG
zeJ~pkoM+Z#%|#sIVuucLS5yDw?tSxz6mDdXn}p3MFqpK(bZ>SqTVfZ-7wOJn&ieR%
z(s{M0kws*DES;4^X8y{y-oT}8V<RW1k!UpIbF}P^Ga#BqgMs8Wlk#UO4Ujeq@#ofT
zq|#Lzf6wrIcD{`e>Gi?D`E{4`8#!j)DYZ?hRPhzrXCe^94Kq4}kuhLP-s2g@b^YAz
z7pC|KkGBJ?OZwUt4*dQZePn#n^R4yxa)%t5K8xW1TOPUC4_&TaI=2DtV?bC6=e@Xq
znE;u~)1E4uiGhuUMeaZ-9M;*{HSBGbu+41;7qD0~_Caq1%AIFuQ#JH$gHA?(?YdJj
zX($Um*fXf{r_wTY7`>#x-(4ntV-ht+S?sVd^M5%GX-2R{Pg~;AY#WLymmCEU{;hAk
zJ$vWxRGVoc%#q9@N{tk!GPzpAGEFQ_AGI!mnQzT~QjK%g?(Z3^?cmG|&e|D}K-HXU
zae<BjTdCM_47?qRVG&jQU1tsnK{l)KHV!NBS(s-<l+8F`S(w>sCc;hBx{Exc+ZIFf
z>K8eFrOXK;@R&K!VRR2XL$|=9{WYE3#M!Wl4$waq!FoAgSKzNf!sxX643Bm0pK-X0
z8X^>59uNE*)(6aRHrdT>=~m=aW&|@{&kE14zwX-2-xw?HF^}eIx!kv&{9(PTs{)7M
zm2=SihomL+pereXEvE+xs3g#b`;u?@zYIc8)f_FZflZZ?3Zj55Rme)I#l<O&ZK3V`
zwkC?jy`3eo4^sC+CcAyAGO?w0{N4Z}7{S1yogDG=Nu%B*$Ek&1+h6`#_&CA}OGD_t
zwK9zRz6(F5L0qBKHbq_Trhi<-v6Cy0!r2kTY?q=pGJ24R>2sfZ?g~2#@LdzS71-{J
z#1ar&qDIGCP}1>bzVQ`>0Hby?=bQ0Ky1p#%FH|fZ0T?sm=x?6Nc9#*W-}6;ObM!UO
zk25+aXN2hTjLE<{U%68GlT-4UPvkCc7awBy+dOx^SpFFz38(CjAyYPinpS(nn~vjU
zx_^Fo#ra(OH%6p0ry=h=aVVv@nz>0sYmyqDroY(jOz@E>9KXi3%%ih_=SQ;I@b}Ft
zG}Eo)o`1F1r_OJxk0Ve1d&h4iPbi74ey}H%G8Bz3N+^h2z55Y;W(vvngtv-|LRM2g
zEX<Vz)U8^Q1|?vx+A24Nd~AT7W?oec7iD}Uu#x0t1iPlO^5n#cb=ODA+#27&OeJQ>
z-3_QpNxG{a`|c|fg{EzTOF!IQ!ngKxooDs!v@Vs&DUa6pY`&vg>&vlpn|<xhVIW%6
z#W<z1W@aApA&@QihC%>yZU#ZGjWzaNb%H+0sn*(wzHOyrr5tf1^#14fJwt24iLGGs
z@YrleNSFli6^oUKPF1xd+#jbFJ!4}@g=m}@${L)-qnx}vT-lmlY5;tZK)*pBn9G7I
zE6JJ!RQQLuZK`e<JFJGtl8XXVwYR(jl2SSJFl-XaFb@8%zrgDB<d6SV5d-^cg<8dP
z#Sp^_^<2iRYYWhHbn`K4@*$?*2BEp0VfrmqQlfxtn>Y;T);D9cJQm@5b`z<grI3Zq
zJ#oIzwXWR{RhEd{vQI8<-YP`<4AWk?zy^_>f4iC0%GxB8Y{1}V!G6G4#28x=bQW&h
z&c<}r8mRz#MHe0g`D5#$cvJ&#?9SE;%YyUsPS2F@G4GYWO7SuXJ)6yOk34#HBJ;?i
zWNCA=0vdFc#Jq^oWGJepPKVcK>28m)N3IoyMBPQbxjO>Q5CvMGD})afj8_HEE5sc6
zw>kua=WstbEMF|$>s+fDVIv}%D`<c`Uq44TAcPvYeJ~R&$D`ooS(syfhCn6ey?3rn
zhYb48&b06DJlmdLrAv20j&<4HT`LjzKtG^(`cp?^Ot~hBYPI7Id<cl<L{wI2v3*B5
z675vsRN?>G8dURD%Vjg<v^LpcbCTRd7iS?jPz2;pPO@$fBQH=#_X`Q{kQMuoVuM^>
zVBpa%HhQ77J8Kg_@a2_;&Kp~aNj&G$&&-)q3(Mwlo~S>9OJWB+znb*fuuXbgvY8i6
zV4=c2@hRzdkIIOyDn?ijkNmH)5d&h0XFZ8iXex!H*C9rskaUecu+XmG#tT74tx56P
zk<i=Rgy?%;P@)fZOQu_z7Uk#f=v3QTdfvu9c|V!y=G456I2Vc6UgWFQGh<jaKZF8}
zAX;*yfM7Wb70lqjCsdIQMG8tNWoPI4ZZ0l^lZ3yb5H><5w&yA+I4&t(MJA7onUDD<
z3}TyNzod>=d}vE5Gxa>#P3ad!OJ;M-Nm3}zapga2gy5NBwEyN<NwPtur}yi8x%2Wl
z@46Qd%1N5yA>`w0_4kQ2K)f}O!TwL*)Ks4F-epH%*g%=<y~j3aa|F+ByLDKpR=&(+
zXZIgsF6pf0_oQ^mw@f)vss!E4IC%m1b~;t=E3k}X^+;pcJ+@Yz$EEC1yay#B^S-6?
zPEJjO{OWdqZiS#9atwp^G+G2|XHq6Etu_Q>Yc&vs0*+yZ*crXkB`uY{K?J1!ktSqO
z?1k<f-?As~cRauH2W3yj+vCxJYUZQ2Gc9xc!bb10xFcp<4oHL^Bc@Vto3rG<xv$4Q
z<>Rq25f-VmD1;?P{iJR?AEB|jm94N09QzWhqrv1h-2V+7b7q!r1SntA(*G4&WqCT#
z;TI=|Y8B~;7M8rX60dqLAh_Lu!a}GKE*RrD9wh@&2)QBMneJfCHSv0L!h3I=FxfRC
znVvAE;>|#wCM~1{JeQGPIgD-YCXIp>Ql=Cd5gmC#zFP`WtCmh!XLYTKt$9ui8^o%0
zj2x{nH7>`ktr`D%FGuG7EA%-fXqL<6NXS$jF43368$y44`B~)TVTI8oOGp6XI%*bb
z=GeYCh%$oN7LPha%p)naee2xWvFW_+a=sd~H~o6tk49-vRL#L>8r2QM--S}?e63e6
zzD=mVJuFOgYBKpH!w62`gOkLT!e&F%JgXA=Z~9oblDsISaK;FBwWqF4!?-hwW5+3Z
z#L3?E9D8!MX3ir5ojFXNc%0%u#F4TRD47@3(+Vg1lwbF<%c)ztR<!)>V;H9-HU~-k
zjE^KMdf(9%#G8KI`0rP^cw+`!)PdimT4*b}BL)L|${AvA>|%#Pf}%7(UNdj^VT?_Z
zVfr4PxtMb_vY+61LNQKgcKdS0T)`Q)@oen?fzw3M2FKhJQUdIZ2B2r&yz691?pe#P
zlW^-~s0BTGf<2pZ=Wx%NM+FKAS(juHQ(o!C^zw44EJ_0OzPI?KTUv%!Y7U8I<vj@C
z*O+vvcCGIY-;$D%=hQvzzL)jVCO((&lfz8aIvE^bo}UI?l17!<w)~eRYfjpDNIrSg
z#U!&#MI3igg;rcYX3w$S#pCZ#1}&jnp+=e6(|G|Q|FI77$AI<c*;-1O0|rf~8jk0d
zi8+N?2IH)3nUxQg5{i>6EAKR+yOV)_?`{aQB$3Q|VKGoAXSxXW8?RI6KiNEoo&kpi
z*#5Zl><Qm;z!hk*wtLw0ng11``$7E`S)F3dd`TRXJ;`A$qiMUXG1m~&(sQr6z{mG6
z%>IV<pFPIg?iy;{JR(h*jWM*r#b0ErN1Jk-qJDwdWxDV4f)?I&1$moGPG@4rJMN5B
z7iYpxNiFys^<T`-><0C=oOm!S0b(hlvb-ip_X9tSRkusG=mV6CncLPX25_9vbURbW
zo2Ouax=<8)=MUXqzD<RUUmeFveWoH!mx4`LTa=~}qxf~4oIASvgL}8OIPJd848;k1
zQOr1nueNr7A(pA;?J`NUZak!1<QN3V=D|)x){u=N4~rO<0ve>VpbCeSreU-Iye<<g
zkW~L4@|EIZi){v~+&J%s_ZVG0o|BU|`+Ol*8+2CbVAOipVeus9#X@N_fn@XgQl&kT
z!maz-MHR;098lNY_VXhZ?b=M%boOmvF+S0T4(+to6_xY{l)uvgE6aP!#q>+7W(_$a
zipl-cGPloYAo4>KR2Y$s{?Syt`U~oocZkghye1NeV$Lnd)xv8^po+swE2n`$Ota}G
zRfoGh3SG*%4Ju!YxUye^lKVkXh4rAw62l&eUcQmGtYL6x?1dHe_f&nOa*mMZx_bE6
zkR-JfYyqo&H(>~rM<{{W^5(0OG|TS{>F+~YSOOdS5_;e~4)S*jtR^6`RJI}R$rIu$
z)A6@(?#T6c-=oIJ*osfF)$>%SHt3GFNC7vDH(~&%(KWw_!M51|9T_3a#pnw$Eq*{@
zR@PVa_mWE_pi>%;v3)Ip^a&x3KESt&z)EPGNV<@;hKY5OV<;}P>xt~-^y-n+bmval
zRN^Pm;>I*TYMBF|8m8LGyy30(&8uj4PFL#6h-A1d-Y7{k36|iD=MAelI&H2)#5L#1
z3aRTFi_hp$@;U{Erjn~+esZt#%HUXdUgdwK{gv<kJKOY?w!w06^s?K}vq*?sf%wlE
z7)Rl;Ea2!Cp}$qGu~E&-3Tc2X((j!WA25-Kt~e4-;m`tutK#%3BGZ6?exfvcm_KS?
zcs1*LC-L0jP1#4P_lg^^@v9V_V-3Kh8|~mAsfC_QRl=rAiPie_D&{9VV(aMh)@AdI
z){_njB_UEr__@*wk;U#Ws~A^u)gCQ0{>Rk8)yjJ5E93j|d}9tNdCwC5S&Yq1D<5Xo
zcAwp29d}+IIh+NrqL!tiIExz27Z*WY+DM$MhEl1`X1cPwYZfJ&Qp(mkvr^aA7SjCZ
zJinp7IlMXDpjr(ZQKd|KB^OhRF)N;61SJwZQaWP5Jp5$uK`dM?dMAlR+2_t7@_3Qm
z-Pb>C#>@L5ZF<JIErxZvZhCVfmUtp0XW|fb?=HjXQbgD+<5g@U5lP0lzw;%%kiiWR
zPHsxGdx**M-T|rdc(K+0ftO<ajjfWDdDZuDPQ?4oahP7-p?ZBA&G{YY7`KnSV7(>D
zW35Af^XhX^VM+op?E0PG26_hem=_-=Y{btsywH~VMd|J96@aICUUoJ`3@*zc1W7Y;
z4lNL+J1H~N7`}Ys`R=XaI91Sd!5n^Rw$7U<uUB5Vz$1Z^xKLfNuV{egw^rX>D#AO8
z&ee&<m11q<Kci$R@vA~BGs@=^7jx*-WccJ%p%>@6`9{SrskCSyyEOEW0&Q#~!~dK^
zP|3M|5ZreVH)`7p7W$eJCm?#5TE<wd6E6V7UcemzEf8f*Vuto*vKWIBq`ozi7}!6)
zhz)FE+q0gkMOv3hH)_JJps^~YABT%^)bmBQMhD6!Qiu!ec)qfkw10=V_;)6Lj`6Kl
z4dU1h#X57n-bXX)^8LZ1JbJ}AOhcA1-1!nID1IN0Il1sN)XS0BG(=W2!ObuMbB$Z*
zc$ejJ_BnsNo`WrPcSUg5XE~d(3eSX5gRRI5*jjZn8cSy37B=ebeLnJh+H+@G-||Kf
zui@JI6!x-4Ph?mS@&T6f?XwvwOBgVQ1^OE;l20aPrx@X4zGk7A$_lu$@#E9vM*;c4
zb^|9@Ix3^iY237b`U0Lzwz!Bmneb35NK|r|ZSrPJ5Y;!y#H=F_fC+MaG8?`<&@uh*
zDnndV^yhmQ+vyjAVO`E_rPnz5Q?hTEHLS9X^sg;>9rv?#$*~bOGviG)MGL;Z(so#D
zRcj%KP16NI3U4=nYHpi<b(V9)p;X10&@iUmm1eY))AOWk$f!dc(6EWrBZTrRrIGk=
zH@QmpT|DMBw)k7R_h>@X$T-R(c^qQWp?I7|=nc#DQk-KnCrT=1y!Fj(-S<^7&@OV}
z_@N>wO73=fME4^v#fCfZqAMsQC4|uAp~#)Y6^v0v-?S*6HN=}it`uwPxT*iZDc6Cc
z*N-$1t7<fUsYXlekHbi)00AUPDl8e45%L^5b@xJ#^Y>n#Hnu}Qy%Lx{AJty=&0x;G
zCRL-cn>9)5F}O+3Cr*;KTUQRPJR;|=16`}tT3SLlBi?2FNo|9MaNZOtLbP09mDdps
zvSz%7QTr59s4+5Pvp}gy<%<fLu!S-3@8QRDnHZ8n2dzwW>5xLoIc}<DDMi^VJThp4
zxcB+lyE6B*+U*lR`>L?T;Y1)awbpQB=c5hsbSKxpI#ntwl;P~6HnuL_{Sj!*f@nM;
zdsHQsx519L@YNxK5qEYu6U=8VM4WmUH)A_&%OQiR%8H(8No>xrwB&FxtGt-0X6;br
zj6LFt#G%p<vgOh{HpUc*f^u?Bh-fa+xUfIEBag18?gKmZI`kB7`imLI5R;?NMT<^5
zP?(x+P!$ytCBbODBUU!j0oLIY-5Z=izjz^wf3f`|ucZE$@udlAMPBfz{(0Fd4;)=i
zGPZ(x+=#V@2yJxiECD}`5h?50yWKw8ECqRHpm@E0*EM-!It2F@`<<KNtOosq(@&l`
zGop$)|G3Eg?G_Q_260n3Jp}*Gy$>3Er+tEux~YoBgkM&!j9+%LYdSW^MAd1HL8?BJ
z_<Gxh*(`i_Z0F`YBmoY@4`CI0DB8-0?eUT3DklGk-G?z~f3>ph%Gs8n=ZRO*S;Llb
z%E}Pfj8`ckW6%f&5h{9-$liemYdEMOS8pji9U;Fp=Y0FBiJ2ftX3wk0s3u?pZN@fi
zbaak!&NV-EK-BXtj|yJ0av9__Ul(3P5$WInKCI--^$ca!U1)O+h%MfserI<|vQ@RU
zx}o36KmCqpRJ7KX|M?|_MK!wUJA2R`nx+2O<t1lXS=od&@u?@8g^|$S?k=JQOT1jJ
z*Sb$e#+9LiLrroSs$B|qF2Tg`$FsF3&it3v49z0c1>)w1)?zEXnr*?)PYs=ku_(2;
zc$yUqgylr>oZYmD-^>%0Y3bivQSo%R$<P^qx~$fYV-9Y;Tdx5rhhjQ_z~`y+`mVPm
zk?GvRYdLNJ<sfd!zi-^OUP9#@n-`nki~h)%@4_Q8a<og>6HY1~4)5t{5@1h|FAn)J
zPX}xLWa53(uPWD(R}AOOk_Zp`DOq4&qnpBuC=*sBDfsnUO)r_&A)9Vd{BE_lckg0%
z_rN&&n%kH=YK5mGa#ihSk<8$@CL%?=3+_$lwvoTn7dYInuYqSNov8=q#8Lpl6UD7k
zg?HGD>9xEO0E9o9BnV%Kla~(j_-OrDR#tVl(CWFzB&5Ey@wbqmZ+{;-7+A3=%#+2c
zmQV%&9e{=2AB=IF6Q$ZsGc<FIiJ!!&l(~y2X=13}#S~If(i;@|rWMtrcoHSWS+HCd
z-S?skgqBo&rId+GRjqHXgO=X?wj8@_p8_+gEBOyS`Fn-r$I~^N6Q9T1QodwGw{5Y3
zq8iDB`R9jFGQV7_`iHx42Gt@xFlRJlqwtbU5=`o3AA*Nn{**EBRA%#gUv1t$_yxVZ
z4G%;4|H;3-nC^^=-`m)F=?I7}4cL}x8=v3glqG(;x?--a_qgP;Cp8578mEAV`(7y`
z<s%ciK}J1nr@6#?$qXRc@NkjwwVb7f)3DpJzGHKIL<Wn(-x+ciFHPfL|Io?W5i%?s
zsW#|muf~nu5LwItA2{TSAy%#|OSYNUkwGqzhWs@GvqB5_>6^d@k>|L`S7y%`hWs+t
z+}$fXcEnz-*6zSQ|3?W%<ycSanPrC1>Y=YaS3ZXJqt)*U$rV`X4V+)xMUGr1Z)<MX
zqMM`u|ARX>K|t)p{q5B_#$gPN-NE(gj(1za#n(*=Pxj_1pn=8WrC&AyEAgPZs;=SJ
zQCUIQdj?LY{!vXmp<@1<h~HgYS{hrtTs-CH!FuLbKuFX;u|>oHYWxmD`x?Gl<h#@R
zM7GWpb^MYoo{<{+P%O$?mSf~5Hv-XzK$8<DZm0GC$VR1D95}XPE5nt!lde{Q`9XCP
z$U+=rmxj3~d<OPB29+Y6*9337J#QY5@08(ws)d-I7-ghG&pc`e+X6$f+_ve60iPqg
z)YiUSRwMT7yO-E;UAhmG_bdy#vVmBC>af%_=mTZ(5D~S$ENV~wxQG2w;u!LufO)-b
zVu}5V^QRh7V=1MRt+<%?cZajj;UGQ~w$dUYyvIyQ+9?M4b)c*UZZSztLeIxgoZ^^|
z`G<8ogcAFjB?V)vSpLPP73hU7+^NW-i_JG3{Fy{*RGNtvjIlyudmPtBE<-tS68{?H
zOd4C*I4i3pO$nM243*6HtVHI6T@7KFyscImb+?Yz1@)V`jJ-qb(7q&r1lty2g}qt7
z_flS`a%i+(`tseYtFsqzqzXJjInY<1)zIv+>VLJ5&Dp3c&Or{rN_fm{CPvh%+z@nI
zegn(Kk9ZNUqc|H6Q|B(H`>{q#4#m~k92)rTd6FSN!T&-+4QuvvFlr215%rm2PM|cW
z7u!s&t+XeS^6vsQ$ww1jD|gRC+2*%NJc9jr(xpCQxmd4B#WVGx^W)oSTM8|O*a}-y
zy4CyW#^%5(>=GX}ypgj`z}WuU6gJxPAZBMHH|*()l$DAHGIo*H`^!BV_?7-YRkt-F
z^mrLV@<?~Awc+t`Wecx+PK<DZZ>-M4k1@ixFU*GdgJ39C&fr52A&NuSgOJ!j4?E&c
zqXS^G`*bA>=A$kBS5AE6D|XS+tz7RqJP2FEz4ddi>m-=VEa;qgWCZ0uQq_JADt5Or
zahmtkPXDLAsWI#-T)AWpbw@G((p#w|JEer1Bp!3qdhP7=bjHc6eEwTns0-Mo_wbO>
z{{N74mSIt~T^oKxVkoIWLb}@_q@-i$4rxI;q&uY%hVG7`q`SKtX-Vmn?*8_Be1GTX
z9DCpUURRu})EY!18QWT-Wtab+d3|Qq#SIH#(2g`xiiC_5ZX(jIDQ=FMS&s?rMb#t_
z@O=s!cJ4bolniy1JIJJ>;PFk!lwkU)Ttj7<jhP~AZg>tG&hbWmX;l*E>+!LtI|HNF
z>@mDk0jq#ww4=C_j9ydV&yQe_?7&rW4kDDX>7x92>S-D&>0tJ(J5rHVEojl>KLMYf
zh0gQI+K%SosdGxH&`_WhErP_A&YIxpq!wHU9*CSrM~Lp&56~mn=6CZ~#ZG`Ee5kR-
zAHpVT+Wf`9X2G3q?|!E?s!~jhSR;-tVW5c56c$6J2f4uA&|zsZR7?0J;`ZI{b^T4}
zjrG@pRVxm1MlGtxweH$1728Pq-Mg_^eUN?1XYY6Sy0)*i|GGTR{O_+&|AXrn{&M!+
z(??`UWs{D_iGRV7LYe&IaT9iKF9Mb-v$0jKlmuIjtW6jTJ)OO)T}-fTojCsR<<HU|
z69E$33g>3$9sa@4$GT+&o$~jRS;qBdkLT0@6e0vjfg3Ty?ueRB=RNO%X8OkP)_0)j
zE>WxTcEzGiXQj)q$>E3&3+Tr0Z-enAW{=)Tp<vw)c2+BxZ!y_<Z<My#A8CZHrPeI9
zj!`6J%+?U=xi9^`VB(~+#nPWjkHH1{jRs9l<+E`gmGh=62*<zqn5oE0=7?d(D43^k
z(T=OahNX)h1)44eYp#Mqv2Lolh+tSl-~E*hZVGbNiUSkJ3W-!@Hgy6ky`Pov=00R(
zWvE8@WXrGc_SM)+m#9|Yw+u1X8I*)4%c@u$lZaV%qm|)U95KNngS~D~{eXLoM3g=l
zzpPvaVey+ihvW!0Eut)<Q1m-ZLiO!w0ox0I1P+3NB~ZR8Kc|2Pharbt*b%M`-&@Bu
zW@$sg-7PgMvt0Ug7p#S)y;$0VWx_C32Vysh_ej;#`E%^bQLP06eo~A<pHfUl?Cpo(
zp>Di?ykuj|B^<`wAr#q#eMXTYGHDKxk#l72uJWy<Gd3YW`pV&a6RD+C|6=SR34^k{
zVe1|P#RnNqIXE~-BO|5)mZ!t(p~!hZz?Bv`t*L6q)daMFW@dy|6&F4kec}@XZT`Xh
zFFR_$xz&0sj`{&k$-v9!J%qLPK-YOu`idv<$C*>I-huBKsROnwr9FM<bJ(yw3L05#
z)oG>6qE*HD+4Y5NoH<^vlhYoa9;9TBqNUfL2NE;E&WSYT$n-QNpeTo$ic8o|@f&fh
zbkp!nzo+OaQA!X$#Sr~ZY+#mHKL8>E4pC;jppsnX*Rbn=<jbG8cF5e-Zb$hWvs%{K
zBkb_spu-YDL@oDi(h_Gg)NZN_2)`(*BveS%ju1k6#!?nD>Lb~*U&pMy&raS2szuD1
zZXcGy3i|%c@{qr~d(pjh-?*W*?Q{=R<SAUOptjlnT=U2c?)aLp1seZ3P{3(MuvEqd
z9nlCknNm4d4|62KnC2%Am`CDPiggHL<e=N4j6cUNMK9c<tUV87QCJ})EhTyvVpH)o
zXU3KqQdU_Gq$Kp5EcSwA^3lRJkt5B}e=OQSg;F6|iWXeNHs8Oqr_)LpMQRI65+E57
zQWVfkI@WQYxLsuMmQW^c=KJ+boyTm&GqzMY>+sbcq|mSuMwwfv#iWGw_UZ$oWiTiK
z{+@!}RnfaL<1Z&DF#Z=u<fd6Dnq_PUZ31h0v2F=p_0Ao`QPD2A|J!nGb09^56MJDi
zBYkM-KcSA^o7KM?&bO@JX#DjGq#r?<m(GtU;D(b-@=(`>e9kBtovPO77*DuA8iHa}
z6uYEo=^PzMjQlt6Jqpj~MMCk8--^=6wyGEn@_XWZ=TP(o3T=%?<*mcSQUUL){h~NZ
zS}N|(*|cRY#dGN(<KY>w+O{p0L-#MpG?C(D1Xu#3^4kLG$L6}>(90Biq*>5)tw6bw
z+XPXOqHmB(Wx}QXo}i%IXO6vpSuTA9oKpDq>%W-s<D#yJn8b6@37WLz;uLw8yn1Ts
z$sL{RdTsb=<VY9F`_X1LVzm0VwMPO&ZwWCT`*PPW=07!XHxNcc1-u?o7ELlE$I;8|
z45*^;;};XO)9xPYF+!-zTm?>CYegGp%(ydEO^BG=$<9YVw$@^Uru(L-pfjk<?Vq;4
z54WDSjaqPh^D%xKUzCwSc4l)*>E(rW$Ma(Dms)>qokN37RW|UzEgco2a}ujx=$-Jn
zEk(u&{>vU-=9Bh*%k%q4@Lzli8%4f@;wwCnoSqb&@(DxU?fk!c+wOqtOlYKfWkc|b
zwe|_oMl=ocskrU;XgdW=tZ)nbomEvv=yKrW<swgKU+Aq6#kiXJR(>|8?TUbB^Kan=
z9dyyKskEM??<E2XP=3hh;FbobNxqomCr@T(KqvKO;RBARRUl|Z*J#(-H4&fT291uH
zzGINo7}5s^>$S9@O7?BMjD7z8&N4>yXA~$)P%!^5`P6a0G)ZWQn0tvZUv~*fYp}-R
zE|Te?!7UQev}U>5`&iIx1$tPby<!og3W~T9nsbu9=NT*%OL%?TBzo9=eWn|8z&Ma`
zePoBZ%+}WNpy$2Af7hNOK=Ju*pRj<YwbjeU)>gHW2NVD**FVf#EOxqH`ULHLW0^K>
z+eq|)rljAb{wu;FXU!%j@5hEon0g#A-5;!{9N0ir2sewN@IqeRKSqxSYD?{8!7`|C
zzxZ6Dbl$eMo$r1G3?(F<w@NMaJ5>bH#caG|4G*n&RMvt@Q0atNFmVYJ{n4$vhjq2D
zN&NXx%%~M2TZv{AHBG(=cV;XBaYRlPlIk`XsuqTzEB`b$yh7bGkE56<OpTjY3~g*Y
zgtmYg{Z&o0feB^yNFgCEvB&kP*dt$$$ROKMMKo+4X}{T_s^{T7G6H^3wuHKnV{q=b
zf`uGE4*2^DQGKI2N+voQvmi81q2~eQ@}$l5&igO{R-Dh#Xoe1!no`*0kz7h-*)=sS
zVuUW14OWWbWoJmxd5tUWaoqEEZ*?}OtLk~}tQ1bPC`TME$yaBBc!Jr_tY}-<TP&$@
zg=*mf;snjnS(w^M)h#sxJ8}dTvs6J4=T=tY>;D*8W%!K+(jcj@usmdL5IdA}KU+D1
z)UAlbC}|IAn5ADwvT=FmZXeLru65)q8^fo57rTXdQtDWTY`qzze@iB`Vh$M_;;@)k
zn_sZjnxZd}Bs6BPW&HbmBb>YVud3c96<RJ+x(cttP52NW%rLxY`iLimJNUYIK&R|a
z-)TK4!y-_^KoFUN4Kk{tqn$(&U3Z4+AUU`%;%ZT*!stm(gv|DO?VRrZVf`vamYrqy
zh?>CClU1bw2TOvEn+p8J<q;;a(xO7QQ|Y3_|1zvTG<Q`@er--rtx#PqimN}G4v(V*
z1+xpaf0T`Bq5vh;BB+pDE)@xl^Z4EBlAm@x2h}e=l0*)eT3yre%gGdqhkMIrO+vT}
zY_3kkyd?R;WkF5O_1$}&?h%i-H}3YQKc3r-E55ecBmZMGmzi|Q=<}47(mrVVQOv*l
z^5j;oo%c1#OOt=cRy`wy`rrpG$`;fel|@_A-m_^m)XjHjY6@@KL^@aPf{S*NBOjJR
z$BEqgrgjPz8Hw34&Mj7?x1wWiZ5$!`8gyVHqK8L9dV(8FIczgfz)Z03+G?Q>=xV@u
zZBhQK|5Jbn2T?Si5HG;m$>XJ~SkTh_N{d0QsDvDJ3|v_^Rk$n~`GIhUV$|bA$xsaB
zUD#OAR^azFzf%<N=MTwm78pgfc8*;S);+H#2-=$Yv-v_DvV2`Un@eo<K|&hM&xvZM
zPsrZbG!omo!-y{rhjm`1|4vUJ``IytYAiTn#59ypsk$`lv5Pl4<~giLaU#K5Z)DP_
z5(}nU`#LX@Uwj<(A%s`YRS3X9c%_ZhtXy~gG@B*)pp}b07V!LOE(e+356}oZ&-Pvr
zUKjmdvVH#lT>5$!@pb2lcjwvd^llyjjS}H@Cs&aYBblwTc3yV_#qaLNi;N;R;XHR5
zSK6m2ELA$Z(Tx|>!O*vCUk#`qP`{`MUU8~mLH5MDoRF=nExx7N@O^cnDs_k|a_K!L
zB~EmzdJPJ7^Bdck;w+Pa&~*bbT6<{P2k!ZX{W@KV+|A8k1<e-LssW;c4G<+QygOye
zP@ZtWBz0@jrS-d5eF&Q)jNQ3aKL{K6a{)zNBCQZ?+0By5o<2^fR0uXMl7EXIvquDI
zj^#5#D34NF*1*s5^lFEZ$ms+h9S){VftzkvB0GQP*^Tbst%a8r)-rW1qJV}3V|Y^7
z<k+%LLcV9K{hx_A_-NSazbt-#V64+Eyg?{)8D}`UE<B|#)JxOq<KjOu$)ixfJ-lXH
zqA{5f@`y!PB9bx1K?_iJxPHJ~$|P%^-_~gzh|>B?4BxpY^6OiyX$ar1t{6<qwcm$f
zCp^`vQohz5Ioa97OoEG~h{U9OlyMMoa!b?*C@VW$Gsnk2S*b({KLv(fscp3k>hko9
zD(5_<RIY5->AG-pw_a%y*Xeu;VwLJ_elhRS`WtGO`gxzEZ0|l{(y1kDuQ#Fc!&U$#
zGa8q@chlfTHyLwjDX4&#Z^@j4Earc&*?<kd!(6?Qd}FN`TvL3HvDPx10qxh|o9p6X
z!{6q!<ar2y!Wq4|vXVDt;pAR5Y40vep-<dVJ58;n@5G=d#ZDZFsvfUX9#S8|qM)sR
z5D>#ftWYCh4B<86$gHk)5-}iDr@#(2g0O)Q#z(@x`P2@^7b!8=3<W;k^4|G93okCU
z4b6=%+%GNo{Tv_V!8X#0%wX32(R4uio^jMrM%{c(feJFW_CYO*DWhngNhp>LvW=Ka
zQ&JDngrq033ep7oOE>$*&vXkFNxA(B7VN3&w|K(%j!SS2lF>U=k#`CgK*12r0U?43
zoiPy8UWVDjlK(8VbcK?yT_;~nG1z+e$_~yB%MbTEbthl92L(AF3%=(wTd353{bg$L
zeFxd-H&td7QS`f^2=-u^VFlT|(6y$|=8C^^9RCrs-d#+#hp~sWXR1x7?zX#;pD#{!
zNU$0t9eXqsC%nBBP=GJtsU0T+?IK>{IOgJ%OM{^P%j=OGj}DBehRxr0z3%oyfG5os
ze@;2`lUZnF8l#LBL`7VlqC=umDHW8Pvsiq2I9h4%d<RRk4V|h|;`lIqP^T;!`gnWF
zkqL`IKrR|&*kPbkm95Zkg=Xzl+&6xdKP)f9M#kAX2pR@Xaj=AnM7y>*M`lZ`j!+d-
zoi*khx%}c^a#e81WcxXPEsZJv@#RVD>Dg{_hiIc33Zjm##J-M@wxxGY&Yq;}0Na#7
zP}Le;dO@u^rQu8SP)Xd1CJ`|Y8M|`>U6>COcbAoD``R}bm%v^fJVu1P0a~H!)USr8
z79V^lnw~d5%pOsG{!AWY8jO{K$5ARX1IA|(y4=m)T|)m@-FOM)+5)?O44d?fjG!?y
zTWic)#PWbf`zba0*`m1DCt#x$pxxu8|Jd}J&Z>8|@%UUbZPUr0Hr_UnXBs-=i3Z?O
zYJDNHh8>kss}IO!>%2`pTZ<-l+8MzI=7-=Z_W`%g`yc1hj*c9gi?weJybsWfS{o8v
zwTL;|6FBt1pd;kQQBzvy8+7+I#yWIrpP<oh^uMHfIeaUABdfQ5()=G(Dv`#__-16X
zxn!o6rVq;Tf&&hm!&q?>0;nb#E;4{1<iw0QVy?+|m*6H%Y6Y{|%`j+hWC#DBzwlLC
zyag-RpmE)NP_8&Qj;8y43;|LqH{gf?XVBI;cmY?Gg9H<ZT4IE#L3@RuDAK4IF!%EU
zx@lQwGIm8e#N3^BH{Y%W-;)nzW%^y9!TR_r=RX-DyppDJ2H{tb^MLNJRIfs}^1Go8
zPpOV{#c{Z_w34~(#8&nE*E+rRW3iuF+L^sit}D;?5A;uj?%wd&wNIFEtv}!1Jk_sv
zl7!FO@Z8T92`ku{&Y=d{KX)Z0<Xaw|sDx-doWm)6zl>2aJk~7M94=28Bu`siY>Dam
z^*!vp{xf%fEPn(*ahyTL)C#0BQFS>nvsQ16Xqr5qsT#lPY&;gBSdbdp1fcz=PD_5B
zwjRl2cR5m;ytwiY7n%K!NBsU_R<Pc>Ta&5xX@0GjV-osRd_Z0eMQ`g9h*C>`b^Gov
zCh*xOA?W-rf$Kxuo^i;t@idK#8jTKV{_h~PfnGf1qEu@HE3}K1b2j>TcA2=ExC)hR
zJ{54RDUxJTd};>oJ7qBlu_CB9BQX)bZeG5K+R6b_AFQzvp^iqa=Tg@YvSA;@u58yw
zp_+gT8wlxkhZ=xjQGEOvdF7M}pL`s8&QIQbRw8pa*rc83{wStJ4H73IX!z&HyZ04!
z^1yk&r|rB<xn)y%d@9=%Wr@;<&ywGnq5L{zoVOx<-WF!uzGiPOy4NSVI-Zz-9GPcu
zT3zoFKevVteD%{awwuLU-?~L<6vxzqj6X%6y!=q|J5iW#U~_c#mWUg(sL$J*ko?6N
zoN}r7W?>{(06$`HU~5#2Ph#ia>9fO^%9%nxeiME{x{q{YS1IyJ4D#U(sG5ZN8eCSG
z#73lb@QAkiMC(3>zSHLu{Retv<);num90pguiJB9Z1F<6<HoBR<P(<)R5T+TH@`Y-
zOdGHT*GDbW>$ip<3d%J<f!ZyD-V$ISO*1&#4#lAw<qS;+3CP8Q_*e0DGRb<P*HwGs
z%mGtZC_ViH*AXRj@1KyDJc-;UuNg$ZC6%K5r!|gbR;+1ib^wk#k`It|_iT@<PZAi7
z<uIwAGC4JjcIG$TJrL7QyO&ksVFlWTDWBcQtDj8}5uIDAKX`n%X27urzBBMpEca(l
zK?M9}lGros05GEIDE;ZWk?3+pjdYLB2GbD1#J76@HMyTxtfR~NS(&XXQFYq(HY~5j
zDMKA38*`Wdfq?>R;``Cc=b3t?gw33`m@?;?%#KW#$~6F|$;<0DRX!?JQLPKN|5L@P
zcPuZUucteI@ig&J%rgC77!;9m)%KL{tBJ48`@1JH3dTqJmZ-A#gN)fuYz0qu>pk>w
zbuYsyw`)Ek9Oxj^?Zg<d5o|{BphUv-cr$8E%_Hv}>q;QB&C1RFg^}ZK{<N*vf(waR
zIzFIi`B9z@bsX9n?mW3Bu<5aewhn4V|Il!^SG$sP;g%fqi{&4~<t0N<un9l#P3Z*A
z{3%;*LW}{R8z3+_G!2+v$(ps$ELyN-k0C!@YNTFpq00=|`K09d!ltJmJsA{83>VD?
zF0T*_5d1$ly5{`6WxY;fyh`PQAgT~el=W;HCK`Nkb(y+s?vL+NOfOu#hNWLiG>td}
zo}r!hYk{_u850U;ujhsSjCnDH83r(du>h9KUT2#>qm)=C<sE-O6*bPs<3z3Zg|azy
zu3Luh4JzHE;?STC^8s~1M$+K^AxI7cNfaAL@)N<iu0NyR$3heSw{LUNR5&Dl5&PGO
zpqU!m;8#yAZYF*)fX^JMfG|D6&-6luO(e1if{X-&rK+5xlBiD|fh3e`C~wQehK9s7
zCqvfV=gybg`!6p$em{$%QC2#gdT4$!ULk&Ws%bNHyA9enOSqm>@V%b&NgdvHYkpi~
zY^Z9wp|2>>9uD|o(BAXGuA}P#n#>U9k87mQa{MvZuL`O;)Z1pXFEt815Xh3We(H4%
zy1?~_L7>F9{*cU2w)?Vr%PY7@KPd%rykL2gvuL4Q$}=>Ra3i~%$zwg`%k7BGz<>NH
zpbDC0Iu^$ckReFO>B`FO3|S<MM6y0SJgenFwjas~`1tuLm7@@1O&Mjl#ec<!^!L>x
z17GM~I9^%3i3c=o2T`TZonMSmSyY5j7R|c2(GbGhXE;0!(*aKv5B52TU|S*g3f7p2
z<b#XDl!G*)M>s$Q8Cr(VV60mjEk$P5oYh3fk1lc0Hv@{P@+397g|7+Zn!)&hMo*mS
zvN?dZHB(e}Z98qHXTtTNro}N;dEVpUPMi*o6NgQyikJjZm*B{cqT)$wJ@tubAMETp
zUg+%G4VG(rJ&6`rdTztf2A;2+#U#ZTsat}j2n~zC1jPnl<l@`m^Du$TjR6s9y}$fQ
z=v7WWvlg?>P1TYd>}6$h?$41&ZI``Ldac!B*wSQv?p+75yQj|&r&C9+&E6iD&0eL>
zL09qW+j`{cqy!`GcNh;kBQu-(5Ur}uiEVtI0Zw;Z$O`R?u4*WC_IteH!_ye}40#Os
zHa!Ml#sP~ExEg;Zccx-!d*=5G0ZF7cfg7lkS%Q+ku&eS95+Wl@5t%4tGW1$zjRJer
zejrY3SQH`wjT$2tpG+(v_iP|+;uJce6puPMC<TbGS(XFv3>wZdJ0>=R!yon<!|^f!
zv2QzGAOCJF^;<h@@=WcdRIBn3$?bDey?b*zUO0Lqm$$OeAIRHiv}2GpPRIJbw-`c&
zW}U$4!<iXZfVyZSmWBc0r^gUee3XJgi){3p*0Ies;O{H>(+j)OVoVOrzM#lX3orHV
zCnQdwC0FRd=Zn=$_=cSaOgM54D)A)i5BY!WO>~p=F&UdSJKL#Fw0`oAE1LqKWci_%
zZg9pzY3v@-ceLX97$V}O8pfve;;gnc)y)%grRn+FBg=N*ole`0Yi1uTG=UQ~v^UUL
z(^C*g?RE5H+JBvPa2hq&A<Eqw&0B?L$WVI+aW*oz&%Kt55RZ1BynBupZ^=ZKiHj>B
z($uBG&m#ta2R+0JB*8%{B1vG0SmxJxNkzf{WoDBDSo^aenP66Rv%^o%cE#HHoJA{&
z0`ct;Z6}9+{N74*c+g%PquHF9*sJ+@i~407gQlT&FE^s-Pz-wNx=%2AJnR4{CqjWG
z7j1ckNm8bvO&C648xhYq^q6pMPX%J#|J&<Ch$X-769{Lcw)tt@Uj@SXNc^OFt|6ET
z`_&X3u^%X>bl>Ua>j3VTPzeHoOO%VdG=}(#n9K#%b~4>M#M7`CE&yxvS==lh0Zwa#
z{cQ~7@8lU9&itj);b26=&o=1eoqaf$f?6C+u7Mqk;HhM{+k~YDnV#p2j_{$R^az)Y
z$G|LoSd#Db!;PD7^lN|T?Wz6q0I{2&0FB%r&Ek(T!raJR!_$P`Uh0MhHv6psmzTL$
z#;18<PXMWO%uc+UFXR7-NS9#A`+<|^&yo%;n(;>^+NBm)RkN2y1;3UT+n*=$<gDXO
zCq3w$K)opMNy&mR+iJ0QG*79PZfDm??I29*jmoSc{{%^ky0syUvSdzK%+O_^%1@$v
zZF?jbJ<IOB5v$2C#?PG9e^bRgi@rBMG69I9?q3QN;Qtq;7^=-!G!a{Fe2!6_KCVy&
z^T}+|={X<j8+hO2yWZZqzqY$AE-yPsyB{tgHxC;T7wFaQvXH_xg22CR=K*D0LIs92
z5dWt0a@<e8eQ|fM^JX$P58Inx5F-=A(zxq=#j%q1y*WC}fiX2sxGp7}aZjhE-EDzd
zLfbk+0~A^s6Sb0nhVGzlytRo5G_V3u;65x!cHTE1+MXvGc~@Ssjc8^7XouYANa)XW
zLCZwybo4;gB$Kw@sh76~v0n>jbm5wnPTt-_4-aN8suP8Kdy1<o6wL$8qbobL{<rW!
ztXD0^+9v0XJ*Sc%zqwcMcn}$Vvx*fU7G?)QiS=|}zW!YwKo?xfjG6R1K(5!xJRS_q
zAiFsrFIw?<CGtCa6y8ntZB}fY`R2es_4H3y_>GBBf^Wyi%Y+B*qBENx8t0y)2^D|k
zPB(dBYJ~Z56(!3P#S1!wnUsW_WRVaZyal~c%%CkZ+5_04b9fa_{4k+0m!vK5n9~fy
z%xVMY`C`%7oBjXG%L(Z!P&MBgbNY*}o@3F#Z;=^GhsRd&)!fu25uh~?HvgiMdPD)5
zdT|6Hz3?~M6$j35iCyH<q6nj+V~oNWxdu@l{}&(!2E*)4bm3b$iqTjXbguz#XvK0m
zg3X4dkU0l3x?iFyZV;uWHH_P-mHO$Qw&*q4$d^?!O-{KY7;+JYQkH|~6QcD68Y@O`
z%5$Lkto8REhV8Tc1p4T|1q^7@eQ|q8DrL&kHP0JZyn+H|E>4l?r3?G@JP=LJOocxl
z4`EaGT}wLOWsuA2Ev&0eA<~Pu{YXPS{DRDh!s2F%sSoD6fAs(8Y!Ub!xP)IxdEa$|
zQe9%Hq&O%=aGATWt#$P}n}1d_Ld9zrjkuP!&ensvDI~R&r+<KeK0{Tj)5!G)&M%4K
zWf-NbsDFj#s1r&`E^HV<=plcWrejNg1n=-g&0mtV;ZfU5lNu1+e{*ff77zgw(cu?2
zDT0ILiz<%XN@x0s#fuSue1c3`UcRBv_eh968s#MB*7+aI<c6f~Gw^o~vsz%F8?qp&
zX=soI=AhWR$lUh}F{-iVX`byeaTC~Fi_43JjWtlc${8yz?c*{Bl5G34fBzGs;^=?`
zm>d&qVoDOQ|HK++Xkl-;=f5nD-X1g<=l!9OI7tY@nS;0~eRK*RI6czdsoMvAB&G}M
zt4axXd_0NB)I%Hi0*#=cP(~1aqiT##NDG<ne$bf|DX4I#{FtKfZWuj+_L9*pBKFep
zb}lhuR>yis8PDXsB|;1mvZ$tonI%+7ItbS{+oga&c7F}c4Y_Qu^Nx@A1p~Eey3=G;
z`<`*(OUIKW6ppHtJ(dDfrK=T>drDAOg3LqG&0ry@9$7gG-N6w@-q(9yCBuF+c03~L
z%Ujg5CRbirxE<O$`Zs{rnzX#%Dq6QsHn@*Zq*n6AnrEy;b4brw@9gthKWXt-Up?-0
z=#&#-Duwm_pa^?@ksLD<pLUi_GPAKM!$TW3fvg)s3=42-Ygz*4lU@#fICJ;vED+Pf
zj2pPw&J7)t>6K25%W1>Ev3+gB0MuD;F*}!d<>0I6M=dWgu+*%st=_adzvSr&fYs|i
zJN#5u|4(RHeAqZ4p>BC1W-ryOV8M+4^T3}BC~Mx#fngaf8mSVqa=?Mf(CZnNZ>FMK
z;$-`M(JsDQmT+Z|_zOX-T5?Hdws4MYp!VLd(N&GOEextkCAh=M>6<6MExpTR=yQa|
zWAK2s)vfj}tD_g?^jU~M)yU}>2PK#dw@13BW+~5)pGrZkP9?~n3WkutB~6qI#M<V!
zrxTf1?$>=>-Lk7YI%L58iZ9PZOiw-uO=!3T>x-)AEwv2gXWht&-M){<jeb@{;d$Je
zD1?b>LDT^P6l2CcdC)v#vwp@U#ndq<#$eHk=lQ3Kzbjp70SuU_uup?1c+;sfUs@R4
zZ?1lrD`6?~AlB<_-*#-nC(>7_cK13Q&-xs7b7~KkqQ^*x7MwfL$|j!VRlqQ*w^l}?
z;C>p*AXQ3PqG>K8j#R@sI@q_v@^{ys=`2ds08hy*NJY9<LyAczj%L;h$C}xikd;?R
zfD&Us!a*K;KpN+vdF*X~2rd+7t2^czJsm8mm}S1=xxwxA_z2Ki(iOEVd19p%u|i>k
zYDItR>0;Ome*=ft^s(#Wl2xr2#L(We>$c-@+56=3D~(MV|C&>xu8UuitY$-p{80pw
zi8Fhky61lpm{sRtjfpiQyL`GuGPOa_a>*Y+^{zJcNN!o!W5lp3qMisoMfQ0z&8FO7
z=gv9W?e8-ibGrueGrLEU8aV0PJwR*7HUA*xz@l%d|C7~s8HGk1vFQ`%ssYIUYYa{5
zv_;(;p4g(<PFu1vJOAy7x8D!i?jAqw#Dw5BJs{HU1LD%>&r;4UtU`|i=olyfR*pB<
z#(*iaz>I&KS4haTX|7<^Qq2xTtT23?Ub@mWx6z&H17Z8bo(}$ZLdGtqXXJ0)O{sF|
zK_+h5v^z0C#b5XPPgQu~538sZBGFQ@Z6>;Y7mkm-Li{3osIpMPOdLtdW7l-Ena>_u
zIQ0yM0{=J(VFz_-vvd4LwNc@j>=AHc$=KRv-&lsQFnAacU{5jyuCuD_xfll(6;AHX
zecdnY<7V-maGA;v9_GJuB6oUGu=bt3fH2>oL)Vg3G71*OXpJmrlR=DtvK$%+bT#9e
z)XQT&l_^CEc{~u7S&IpfBYyNgG06(hU-@VrQ)cI}K9Bo!tOhcb*mzhUmVNV}WM9>4
zS40J|?Z~dnX@l=(M>ypmB*66Fh<~zJR<~nJv}gQT6@vdJSg|@-ZpTXPjzRO$Nkvf`
zvX*#jE+PxFQ)cvz^~YB?w_;=gvrRv0=f@g!JQMlu@d&R!ySV}r_;5J8HoIs#NxsW3
zcbw)IJr3i?ZlTVXzrNX0SU|0cZuXFf;G^yI(Y2m%qeu&1t>;=tzt&!cEMtkYPGUgj
z){MF8{^b_}6$KuvVM=s$D&wleTsPABNCst12S<15SBq>Ni`d9_4zqn0<^mA{)KS!j
zpLnb}0Myf9(ZADE6?7wk)VRH<zj(!6*Jm6jMEdQ{fm})%PA26u<eG8vW@7>DmT_@=
zoHaHhT&cL+A8c&&4LL@YjmEf`iw7}n^*bdiR}CQo8+YArg0t+6ju3b%Y^RU=S03%<
z)3mL|`GKgO*oQCrY_#<{@?cX~5vNd_Te6bhO=)r0C$_}IYDUvxR(QFRg3jN|YHx1G
zsD$J;s^hm<uPk-*AdNzMeS1n;K)@y80l{dj=MQXbr<U6W)@sRrlMCjOeaBAW4xT`T
zb1#Xhn9Lr&(O<p^DX|Y`k;NSdZ<>ed5?qy6>QS{PRy~`fhene-p7+qUv}@?;hUB4L
z=#{Kxax&B8*(@vPMvsC#eqO!SF_#e~1c-1A#SvQy(MZ!5(@5Y~u;0@#S8>zI3<kya
z5sSY%k7VAuE&AGzwcox2$Us&Kqob8;+cyzkelI~X>4I{vyq>2lg``+x=dP%|HHQzr
zsR(}jqR!EIa^XXQ6)h{=@<T`6A_zM`>h>>v=C$Q)3!-eE$q#P#Stzd5NaN55YJ#J}
zWJJJAesn!jyKS;!(O?;O!F!&e=~>Z#1E29Edu+JjGq!<=CYI1xc^T&r<v6oOxwwKp
zs&4`$RSp|e<DQv=(XM}%J$@tv7*uXdnYvlPiznuo7pCZFWeQ2i$Z$g_ncI=YQ_uFN
zyaYV&j1BIuCaE>*n%EQ9@Td7(4UL3&?%^ICJF2o86LHulEKCP{R$7oWKzVPUDkL2M
zp_yalQ#2w;vLj(gvgi6e#2C)nf3M&eKdJrP@zwk!-Tv_#zsj_d?Qe_g4C4>PHvb_g
zB%p8|vIa10?(c_{E-X#Fq<V1PU3dnoYjplStmhPMI6p2KBd^9XUjW9sm-N>M@@f}{
zK%H4fWEE42(Zk69ti`f~_;w#RNaw#S+k7`LOoBlg)-UEhbzIaZr$nN=W@S*tY&RRZ
z?fsatmu;N68IWdU46Kznxes*ZIgo|zT@nU6&-oT$XX=xa$}|h_c|hY7V;c-PNWV#T
zEiEyvD$d+S&CDW5vZu3?D5c7qF=WOySj~zPVgMM92*wN*e4XFc<t+nc(COlO{m)U_
z-7I6$)5f0dWZFsW|Dp*qFf8iQBqCGdeaWqnEGts~k;f4bEpGqdjk2s4Q&gmNXcr~y
z12V|Xp5G$`vd>H&ao|&7O`7yFoGtC#<*tFp$AdtuHQ}gHGX<;J^-+s|3_6Kld-J=Y
z{*G;lyghrefNwQo4d)V+@<ne#f!W%Q)!#X$S<(=(0w-SsbWpk|5bJ&Jk(h6`urQR3
zB_8Y>!09{sRAqW$<<YL%)bKH)r_TNBjt))L3)j)D@2wErSK>AQG=pzNI)D1j_xbd?
zbnAVwm>td=*Juy>TWo!c{4p(U%NbjqeP^Ar^73hm*-W^tAn`W3=q}VUOR;iV<b}wP
z0N^0h)pN=B@L=n!r}YHPM*Y(-a1Oo^oTrT@_ZxDarMC&jmKw?l0kgHwT5xOWA2io_
zW~puiiJ{s0cOn7v+cF4EUE-4d(Ty}vxx<<$W)0^#+d_NHU-1VqEW_0%^7KAYQMXgR
z6(z7EaBWZzu&jZ&w;L7Z^DTR!2vWzfv;hSz*<x|vkVOOWtQEV~muGX>wHhUjYA_%^
zQE9Bx7U|I{p;ak^2b2UHJauJt>kV|6c6CJjcJ~oMMpFX^wOyy})`lgh1^uAkbbEof
zf43U_Zdny?ifrns8@NS~Sdrn!xC=Q$pP)rq=&VDReAU9BL#}0J#q7`!v-aNp@gqXG
z$JH78b9aob-FS8+hXvbYP}$N>0*40oH=zhbbV^`x(UYVvu+!+MC@xnUL4`pXQ@pgu
z@HdiL?_un+TYAhdkUBuf4dm5?4RW6R{k%*5<R)A=#8Eca_OwLT`NVAJ_&gvP@Vk*T
zTs&V{BYvY+Z>1kx7R)5#%E9esP>}qw%}I46XR4$8J^A<0pqMMA+xzuQJ3Qp2=a!FU
zM>m>7mkC^Qz_hE5mhT*|nU}KY9E6=VH>a^syr_eK^Sh*8CaentOL2>Uv{1+t*6Dtq
zPAwgoYq587d)x2ML3vs0Yl`h?@7`Cts_&X5<cePtNycP<Ygv}NT;X&ujbApNIx*1a
zFt)KlDT|tq3+CU+B!B*qEPSmo5da!h%=qOwZ)L21I%`!~T;Zoult@HBk->L_ig{qu
z=geO?@8U`sE>67qZgmbw(dJ>J;RI<}GK_&O-wx_F2u`om6QTEZH6l7nQXBo*o|}BO
zTw<QcIusDu3#p>3Dx?@U2kUUA77$va5)6Dxa?DTY$(y3hwC{e$T4*rJ<jGTcyH&2V
z{YUT)vb7QTeUkwi79;lgqCg$3fI2^HxCTbA^zM%b|L#&^M(FJ=AHRU9i%SHg%;|ar
zU-2Do91Ri}<&hra0qE2r;EW}T0MnyNEhw3^?H=d@>W>B&U$8K?trRzlB>vJ1717p(
zWMUFwn$Hrma`R}0iA%CN25la=4^5oogoXmj+Q9B18lWX3Y`t`Skdq;qLzVo2AHMQl
z>wCK>cwa2Oh>xB9QRy~jgmg(j2^b}`+N=6a1dx9XIUdVS)DlT~-Rjq8)2Bu$4%RIL
z8+KfQdjqX_c%I0wAJBHuQc}F&H^VGDR~B9ESf`Hn56mDjEU(XRzg4fH8XEm^8UEyi
zU{_|K3Z3ZIv3;v7gtC<#DCaR5JimUA?;9?iWK}~Y0t7on5h*Z{{c$NT6?aAt40n~@
zF(T?DP~=izh<mmthBs;z9*C-=8UDQdrY!;?SbLN=xvnkVU-r;6MBfuYVG;{Va&k9E
z<wFH_rW6xLlapMWiHg<E<t=0K3XW9kICyhOfO@Y#BwKP@Au^q<beS3)eCy1=5E`vp
z*K^$kjZx*P7D5u@<f6Rqy;Nn@KS5SCCn6h&LPM2^BC%u&)66(bosMuFOn$=O)$>wi
zD)KG)KsnMm$dsg~VMDhGItLZ!`x`#M7AAgDf5zg4bm<KoR9$4!r1LR+A;6B`WXPX!
z&d3o#Rg8{ia28E!=BtB(3gfGtsLb&c6{YemcLC&}+b3cye~kZ(GvLWb2tP%L{pUl5
zxr_wHwRJmtMk{@hDfLI=r#G}SU|SjK7h`BFOGs`kpHT(~r34w?Vm4RpkukC{;%R;H
z5GKBj&9x4AxX~5f^m`?LEg~U5Rl{N}{XSeinxvcDm_i$Heih@X{ua|R3p@CZ6S6nT
zy=+~_Un<1FUHqSwQG{mcDBuQrR3ry97WDYzwi}{u`M$gVIJp^&PVjoyv-PFI{DhzG
z%}<UAWi%cfQCUe5=^xtz=*jmFAKQk5(RG0y%SpSRNRkyKZQ*a*Sm%rItLglLS0c>f
z_%2GeWRC<W)h2f#$+mMfa+;!JAKYP|tc|baezmFTyz?yxv@73eCx0d@l8OdW0{g^K
z=rWBnG~=pfwT#*)clJQat$jX!PP;%Q!m=-1ha|RuZ{|FKN<lxId@{0JS#u+kULkGF
zV)g?CjkW<rKGg!FYO}Aa<l6zD(KiLoMF=SNr!?eG&GG3y1pHzQ<CbGKOVz5HB~cJz
zwA4$RMII#Q{lf4?-X=n_m-0vib_4ONDNK~0fb_!aVDNp?Dr!>3KUJQesvo-r2+VN`
zDMoP}joHO;p_IqA$#dgJeUS=<a2)?bKJe-@?mU_!GvTD_!W1QB_Nbbf11A9L6lvY~
zmA;rS=*wP`z&PqAb~)zO@DGM?pl!l3l-{Qkx{|mm^nNcvj<TIK3k7dyHgapr_+|9<
z;`RK0{<}7_@wa@9{GX#gllwFn)m=8C%rzmr3VbWsXY=H~U0$a`aP8Bn6NwZSib-_6
z2=Gn_YBp>YHp$>V!N7{vb%}L#Zw)s0y@k)`<uOf>ShLE?1SmUlkzq?KsMV(cPvS7l
zwsCOeka*E%VAybIW9kq7f3xXj6_t3PVa<0ap<ci`rqrFfG{FGKH;(aP$3PYut*~A?
zS*=1ikWYkUMuDg$<K!9TMRjQEmPkW-DF=EErU0G8!IJ|>$xfG1jH3mCO*}>>COBck
zI2`6q){#*7Mj^YjOs*5a0ExHknVO=Uck3je<)D`9{;l2>t+b<2oF%oE<RJnikL!Pd
zp&DIOPXhBCd!ek|q7XyJ@wHm<j>q=@Sxwxzj7@bV#r<!VX!5lKR10vQsIw?XkqacW
z(#Glh(UN@UGZk9%<%LH)nAN^ra&o(A`7LSe_c-0t7-`W%Dry9n2e$8j&}2GzJB|fD
zF53tTZ7>Vp^4kIX&wS5Tr82Uj7;6%q+i`#rHQ^R>Gt!uDbpq7YR$wII>*BnjuU%S=
z<GoF`kctLJc(`zVSionK2KmNTp9~mE2ALhT3<ypfsUrdwsa?!7D0SKk#Jc+(-50yA
z6Fc++0^6@Bo3VDboL>ge!j!TkoN|R)nQEnt*c^`i%<~GI`$3p&w)#fKN4Uo~XtGRn
zaDYG6Ir(zJwwSL(*e_AfUt32p$pK?afrX3>{Ng0`?VNTVI61+ltUDql2;Q~~h_phK
zWhag%iY$cD5nT_URd$}6YHBuS^qTEa$JNo>f6SuSt;d>{^Qx>bWIzjVD;}EiS7k9}
z&i`QCAWh0%XvpyofL0vOqjlaMpV~Wn_U%5P)k6&ViO#!kP>K7Y%o0UCk*45*(T^`J
zdlp6{Jg8I)Oj5GJsGn*e5*%WLqc$A=^*fIUB{Hk{j$URy>LSma+6`TV3kx47eK-yE
zmAM_>s~ReFZFCbBV2qn((nsIApKZKBhy-DLy}MwxDXU~bbZ<gls`+<Xs3p-#aBAF)
zzrfK|wcm%fvCMeTY&;gQ=E0_gA3z1LR=fay0<?9t?=S<n<=WIiTDrNS4zfvN3HU4x
zASsz~ig*F_8lFd`u&76QdfRUj@F^%#_>0AQb81C*%X}Kus5iww9q~j3zxs((7$8xT
zQ>h2x$7$<e2jLij-vE={wzB}m_#llXG|j(Dg?=O_mw^jDm6FE*>J5XoN5xSB#wFvU
zx#pW#my5uvOGcH%HMi9NibA|UgLfI`1=`DpFIQB|%0!29#7>zqEij1RNGD+Y>i?ff
z7}^`QsP!8L99NVw8~~3XCCaHV9#u3(%`ONrmm-{cR-J!agRzGfD(f~bTBd9iCc;Uo
zbda3|*}skD01l%~-oi04Rx^^)U!Nixl1w|L?gt)2vkgi9(mVZ23BV*pGe#OnvgF|D
zkYuc)6F<rTK$XC<r@osTvI80_BBH2w?*wy&pgA#A6shCxH#X5+?FwKd#O@lLXPIAr
z_$oKXbRd7sU!QF>{!w{uO@B*LEb7l8!;_8#Cjq&xW7kPa@#ft+fGO!NqJ{o5kC?cB
zWB`d+bjZ^!o)?WzP|%Q<_?xp>`JADXasA;8Kpq_86A}{n8ODHi;7m+2oU&MVYOZzS
zbZ~+H&`Y_52Hir1pQmh4yvfzpKKeFJNQs!K7-13@SBM%}=me&VXE)l0Z495w6i6s4
zMa#qy8U4vPt>m^vOk{%)j}(Rtu;lkog4NZXfDz;ib*kxJm!;NZx6^YQ%X`aAt=nj&
zdz8*|tb1L*BRkG?GN)>ksfbSqbGgC--?)Sd4cy={`3xX+#-U5Du*TvS&s~i+$=G;P
z#-Iq^9x3KENjRUd(CE<m<TOJnLOzJXUZofk{#(aNCEc=uS4eR4JRX3t-@y)YChT6G
zKbL*)EB;f!Nm~(j-0P*Z=vy;6BAYBSWCd3Gw7)V@8dtifM|8tM!IZeKr^$rdB&8>}
z70udgc<H0(dV&!+^R_IDqJW3Y1DZ9#&=YfeD!jgWyXNLY68!DeUOKDr%2Odq@i7Af
zYV%!h_hx{$V$J&}JkVTVwPSMagj<`R5<?V;U38yV`s-TDz-VQMjNLOrxmh#=8WCm`
zLwJCf+h2N+*r|mjjE$F1h%Qrs5Gxep!MF@VnS%mGMHf#A631@gvaVfpn)5Hr5s9He
zEqbOD`{c3Fzu<nJsDuM`an5beXxb#-%V(u3Jb7x)`)TSdN`S=gq^3Et5Z+VZj)0$o
zuuwF4>?5Oh_AYTkCZ2H|nm!|%O<=paxwI3pt*w|<B;acwx*CXQQerVM!C<vjn3xbI
zbq|xhPi?grp+ug3^0R!rq#v$f>BM?m*<tNy9ZZY$GM(>m(>N8J;Y=aIKu$-d5<&Y<
zfkt9Vf^$lK+D~)_>p_x6;<q%BBq|sbMiq(8^&fKAznPht)fq31FB+N}6|u2IrKcx$
z&9{f4UfCaGwM8`CjtYV$O@h4oyxczTc~9PYV>#r@Y^G36b`e#^n<raark;E|zxRG`
z+I*9R>;IaNOfJ0DKmYwFtGU)zCc>612{OHGVLU9pH^b|e3|sPSDV^Z{C40>{E0Kis
zhJC;(G_qO%V1WX}h1&YLt({~$Y{|HhZmk88auds>kY>-i!>4%vwy%5+r4zYYi)sM{
zf+ZEhCSi<)7RRk>b!n`J^&xhFZk|1v(L$U1cPUf6Z*$#W7C*GTT*%cCfBIJ5taADz
zaKnB5wEar#R%$R*o-SVU>pxGyf;k(&ajex2i3^8P19YC)@c|N2@P0m2AXAxFvb7e6
zgx+71UDDQJJH=7^r>CoHxF*6cSh4^*>u3-*jb)EJ*WPsEhpc5rBy=J?CQP123BGDx
zH6#j_3F1reqdq|YNm5I}H5btaWEEo0tI5I<vU~2YXpt@<=N1z@9nUy|bM#ZFgXWFZ
z4Si;D0#3lHv3jb>Ax8@?8vcueKL^{e;WNaIimfVXqv{@x^f5ku&!J*rLjjUJmp*0d
zxIgH$%lJf9pMHoU@?$XnvfOyDUR(O5^1|(9;x+vBO88mWuf*pfmE^Ozf<-*NgXR{C
z5hov*_<$+{Ps`l>>?}p#HnDPG9;ZSCJ2(&xn~Vrjur<}q-rJqvS~{+(R>l-@W5BIJ
z{IRV$?TWr6_AGKO_;rISO;4OTNQ)s~m!t#lr-bzQ?k;y={JYw7KnM?WZ4FF-WGbeN
zoBb1B8T=DRqu8D3oThMS8R%N8(4t}Nb&89Hb>igeDxcB!XBzicm|3Yp7XG&61j~oF
z%G)o$vnT0jRA8(ui*^6)5N48crYjOD?T+n5)xN-tcCXXo>LNw(tlt!xc$Eg$E9e~4
zy&HAr=ue;p^}F*HvB=i+*gxKcW@KJ3E`De8l;NYVqD=@^D%T8V&=@7$6I0)4bEu{)
zFqGT6juK;%LHe+bi2iodekj;(T|_Kv$J=}Fo!?cY@Mhg<gv7gv91n=0CMf)XNLHt=
z42>Q`CNDWOM@<X2EC!@YrH5wqE>A?~CoAUG#t+3oJPXqwcRt{biSe`YoUs6mql_;R
zw|R4;<Bdta4@7L%jxi3rWeKO*7U<vRXfy;BAEWO!uueI$wGtw#oqBs8IfEGddw7nW
zPy=g;UT0qKK4XL$XZgJlmuqIyT#NR^C0DO`O*#EY5|yW2Y1?=6JlPT`Q69E!%F%c4
zcruU`WmRCWxk9O`f(`ggJkYN+J4I7{+&4!8vWNNM!xHx19HRYLVQ{<J)@E_|RQ~Gm
z)X7bg{W<?>$K`jSmg{$7ywhA_mXT4a2A=0mdOkNAvdrQSI@4BmPM3>cny$M)%L|h@
zcwP}YxQ4g{+`EriKsK(N<^8PoZD<Uz#(2YPXk<RAfN>-lH3q}ZEu>p(t8*wA3JBkY
z0xz}U=_Y1A?5*gUY7(vFi3^L*hgMU`QtT6wvbOE9GF61Kdcr-3slV%cR&SqjTq=9T
zIRwgnz8Vg5bMw&7vk#r3aLwDek3E-i!cS9fDT92gW<d_L(voF)Rh3Bd6_<ng^NFuy
zzYYAOsH`jn?Jv+&Nm&CkusK;{O3vh!ye8=PKL~H_kPsfFcQR`)qEx=pXJ8@;$Ly&2
zo(0~z?@5tgx_5#^gAvouP&yyQTUNP~EB{?=lzVw^MFV||tv8izu9*=8jf+3MHd3{u
zsaVltLESO0%cXB(bwu@h#XmOxO@tdzW+lEU6?$8*6X<Fl-+l1r-T$gzyd)0J?tO!e
z!e&_n36jI}1AulE(K*Jskd?buyd{b%3ZKuPIZ$~Y!u3mSyQ9jASHRTo_1(-tm?vDc
zvQ?pc?HdOvJ8>p^_IS!T-IbD{G-Oz6tkAen+dm8!0NcpYccWfPTaritOzH__YAV{F
zTM57xK;&L(pmreAxU7Gsns~7)tNGTU+oNyByjDKuvjaESv|=Qga(-fiYI3-S#CSYF
z7fa4raVRHWCD&v=FL;t`qBa0)t+RjdmTXF4&sq1ayh_{vyG1A%IrZ^nV*U<4$+Sbu
zRA@9+!Au_NQ|8Z-uDVTh?>({p57yfoC`o{2EDJ=-5>)>k(=rqgJP#2(D&6H_(J$PF
zx{QW9GSXJIT^tsz(Pu~Uf`i^fGIq;kqp4Cvil9bHPzQ`1FD{0i4{gfIlo*q<>l`o_
zBWUZ%M7rQS*P_Nd?9L!^*A=1u>TyZvR{B`xexY<<%>Db&h$=`#BMYKIKBIBfBZZWn
zzAR9}(><0tzQ;hDIzb0wOjRt*`l`&&!3ped@H~E}<8f87W`QFqsIv-033H>{NaRC|
zJwGhx*CTsqL$WZ2BKQg!rc_mukZKQtYG?4sNdLtrt0BWV^F*_2cnkdI>C<&Dde|pf
z1_$4k1>2Yv)5dih_IV5Koo`WZsvDOrP|F_uwru1R;_oLE@AMmU=D7L(U2Il5@!VOC
z#>|7axgXl?A=S;Vgy4!#$;%7UU~CLTx+g?Xk{&W_!hjzHp*&(WSGoCXbMLILQ!aWB
z^upQWyMt^G|LjZK7&#<#EY>hE${nGrwIQ<%Jdj0c?Iiv#-(%;u4v4+J)h!5~`~udP
zC~{<C7)7D37O2gZu74^^-mBAimT1Rgbdi_HG;1F9_8hQ$O$mE1n@r)|YZofPd(4-k
ziQq!ELVAhy^3s{pxVXK!MpxPC5x?-)-Cwg*CN5|*7E*+-RF9U>7vOhc^5q>l=|@Qi
zW;kU5LazPv10PmY*i{<4vuEoS_bnmGU&+v`Et@2@q$n2V*pbouu|lV5ow0e>_!=&0
zBahQ(w`SiRruH>t6_WTqQ~<eo#A9G4Kn6fHWvtj`tt-BI_?J$$$*Nj!nBm#*!66;^
zD(#C~<C8x4`8zkOSnTsWt#&+~T+?i|2SQSTLU1E!DgKo0LdqWCtSIDDLE(VQ!1&<<
zDY47A4L9YakAJ_3XP=T8^*}sMa;3oF$y(yb;KiHBavHOeOPYU0G2S&BuX4X4Xh&*^
zMOcD5%e4<BWg}<Ld+b#C3mGu%+Rpjd{vF?iY!-AU(3ZQVOWmHTEWO1Z__E3;nic9>
zYb5kN<uBAa(6ctUD_h)D+nuz-{-cs)^SMBI@K!^O%X{mil{Ib2PTwRsv>nVGPOZJx
z5|({j5gKcn5^;m5yA!N6N)>$xZSk$v7ik;^SM`h6r_NXAICb{)_l9`+ssNexhkaIH
zGq(2ayTQ$Qf?)F_nf*)dp$-2CrnwW4v^?*ZQW3{kfBPt2%d@fZh5M;>_P`fZgp^5i
zGIF7B0_SMpOKoWFl$qNe6+$qhPko(B63b+&Jl#rr>*V|{QiWbSG;MZqt#yKPvI7b%
zD4oxRr_{~&84^V*rYg~=d>iO}blcCqP9?pUy_DgWq?fTUo*&Q}brvWb!VG+g(JT#g
zR?EOQ#Qd=OzSd^sc-igO9Ns3c^W}f}Ro?DLq**>Mm<h_ckV)1SPnKqv+&)G1yfa(T
z$_u}NXmcWG!^rxG8w;_p{uk{|Z)N5JEVPLSb8qiQ<BrGY(;1sNzGI~(Kl=eiwe@`T
zY3mS)(O`8)$^YZ%EW@H|yEgoYG)RYZ$B>fJDBay4DBayD-Q77zN(`OS-2)7bbf<LJ
zx8Lsve>pe?*!R8Hx~}t7yj-HH7D<>TPBDYJa+`Yce&;6QB)-WUFF~fVY-}HwRmB7N
z#I6Fl|KXuj>tSP{7vl}Gx~UU>)&4@1etbN0mOQz$h_Wd=uDPtFH5cs}^d|)sgomh{
z#iAHkBhaGy71e|aWn;b2wBM3%w^B5pr33E8+Z*ECSsvRE(mg49@DR0pmE7it`Yz;y
zFn4-%x3s~ia!&26HE8gDwyt}c0-Z=%O2;N+LJnZMt=h9Ye&iN=YzS|RQ?f{=y*TV&
zL$u6Qb8yE}`!rcPzOQb3_!h=RS8^Rxu|1v8vjJbcv2pPZDIU}F>;<#sX8PO8(nmVZ
zlI@z2{kWy;n1;>7U`Cu-;w(m3UN=9vN;JS1YQQR_(&gPeIY~GcYX10vB&(l2rBj!@
zSAO4~?!`J?Cf>30kG-%YTldEtHA?28Kh$UXF|{YB-?pmXZo0mz?g$Bac6)Tau-gAm
ztFS#JV^g;9%lw0wH}cfm$oEf`>HX)^`8uRAVj4K_{r^==be_a|`?E$5TnV$T9ueFb
zjHo=5<_K<LFQkeF;iDE)!;0i9RtmwZeGxG*Da%7IBLDY{6!*wzT3a%VoP54Ij-HO-
zmvEr`=CwCy=I+sK#e1x&YWV4qo)mWt4y~5<LQ6B|_0o7K@8~<B2{!qERCsbSl{Gpx
zHB_nE4%eO!jqF#!4$2>IDb&lZ(U>k&rRx(P-KB5<DBQfwEG%Lu7~A9D>bA=^hnJz6
zyp8L!oi~ZTkjs@B28C!h(U_D+diG>jq!RJC1upI2v1FRaH&E)%)_}!<>z{UEqD?VZ
zku3B~GxfEAH^0|I(Kj-3vb;nKM2KIastEg%D1ac~;$!f7_LOPPR<rTMKGl(y(abmL
z+m>#=U%Tt<F>1MY@R_KD3AT3r7~#BW`Xc<&P#D_3!FU2&(k+gu%+sA;4(*}gdy|E<
zdMZMui&)6Cs|`&Y96n3jtfAO6{$24t6YhLNdMi9*XXQzb+*g%Fk%I_OW<k(C|1SOj
zhK-hHOc7WuQaF0Mi{W{dl{7w?(WT<$`CxlRGu(}OVJuHNqx|Y7%K0567YXB-Px~&^
zFeSq~d>WzaAC}C8^9!2^MJe+g_ke+_DFGeOEq5IWUHwLlpJ>MKRPd68b38YxxnVsi
z9Hn*?$%d<h9CG<5>@tT&CpH=7YVS}J`!8W0!t2lW3>R;Ai7f5|*X5r84H_?GyXRFT
z;1c3+3G~8Xw%jUmVdSu=nR+hzWRP55t;YpU9aW;hS4kdM811A*C!g$u>@k@@2b#nM
zRw!MzqN<)rE@lBkA(6dMA&cD6#5=f=r9%{pQvU6-fj=1!;~ab>N%P!lf=LpvLvBl6
zYnj90OHxM}g!TKv0D#5Hezi3Q7PsDIKKnK>b32RD<pD1TX+pahAcrBG_1sgI7^wE^
z*r{x2O{q}9t64R~fouVH4+kc>(8kVnO10l9q37)q$@AUY`I*b*#(QEttRVu3@Z*Wo
zcj?OQpZ$V+SEL~P+zI9r!#{x>BKShtikTDicqLy@zmg6`^RS0h3!d}Nfh(d7oAFy3
z>hcI~7Z6HystD`+vX1NB@)DT7%>1ew%-m2!y7Ah?pLc?bmgJ9**0(ul?JG}XCl5cb
zSlXYB7Aiz<$G7-kCKI5$Eo_>@4%7K~iI35LM49W5+W1Mwenedtu!&*8ArL-lC}mFW
zjhnkHuWB7s5OQQsW-9veIH;!3%1U&s6-`?{mbZ??ByV#)22&*}9k}yaHZB8+`=eL^
zRf9g14`Mi_tSs0e<pp)aSey)QQM7ijLtA47Sqqc+Xzj-p-&Z=3PQPrA0Ox3mt}C`7
z2AU!*M$kuP)Jy7;HI>=yJQ{U8{sKo@H!@hX;}3`I!WrF&N-}6lCs{&ch#u*Q+`u$o
ztF(IY4#kqtqscXp*JO)L7Uw&D%QLWcrE}^HMe{j1IjL!Do3Itx%{-nw*M9>CjDLRW
z8nP5AN&YI#MOz!cz$7fZPdWLab*zI_en2%}%u1IOy#h2~-2~=lIaXt}(ad*$$qW#Y
zvdT=W!f**k0I8}faJ6<f2o=_-J2QW3`ceT;XIy?{954$H>%55w!a`w?=6c;B5#8GC
zqgnI3LNN5&^)fEjN=SE;?u}3OAYeAFKFV%uUoRI57$K_IAOiUs>FGm00JPVRp4k1T
z=bfQ-V^8-lZR({gtWE`)P|lodVWqZQUg@EDB20!DX&1-1Q;6MV<#$f#Pe8jH@41(0
zvADKWxbT-Y8@jakN5bNYIuzvC^xr1Wt(-b#=ssPs)dvvWjd?sk^jYcj$*#<BK0gqt
za&5oaV?GfT*w`$j!u$tLoE6*H;=WQy-Vf7b4f=}MizS&AJHcBFI<R-mFbnhX5be%{
ztFAk3ZINkbRQa!_+|A7H(|&5&_eK4<8w@;~vgw~1SbMZl>PHjq2fUQEgHEol-Im-X
z0GqgaA%YqRI^VKg+@^Q-H@}`6sqC1PD{-nlBA&cD@HJ@Z5Y8N(Qa2hKDh$)(mTZHz
zFA|2Nk$_-4;Gm0wy{;HZUeU?(1X?=@YGG%9F-QGgna23l*i;3Mg#uKVKEO7M=n)Ir
zdsZ-dJ(+2vB%{PU$@)W~bZL65<~|i9OfrP$E7U3mMR|R^3B9G}q5T%-A;uF<CX!b(
zqYVCtuPna>Xf=?%I$8s9-QqAPrfuh(ph@#b2fyVZ<X8989Qh9yZK8Yg(#@7U`u(c?
zW9%KkS9=WT*TUbTh1H_85}j?VM9jajJfx&T8_-jIgHGpIlFvp`Os6~ov66>Cb}cR`
zih7K3Kdqhd7bzEoV(lOYIiYE$blId5q6}zSj^krWK97~6$};~IOQ~u+Xf%dTsRmIA
z^X{SgbuFOxZEO8y{W7%rIcYyp09WO7lu!pzKyzcGbs2wGI&3`j+xQ6aqDpbsg@1T0
zrD>2xR>7xE7m1vj+S-9V^Z3u}zPaI;|1RU-dhDd?b7*<{iMertdt|@r67OC4)*`vC
zY{?tFfoGN$>6z()$Rk8+NFX>Nh)Dgi`(^Bk)C4_OY2hyJ1O7|}Ms~CISV4)VPkj>K
zlfr_9^GV7zze&KNk93)HQi=u@TN#sTd!N+D4b8tbmgwwqS*!yU)$68rz3zOz?f%MU
z3BIEX`(r;c)9`z3Dj>w?SNM(5)wv#6X{3eL7mQlj{qnyuV@iQndHBy}d4xbCg;X{M
z1oXc8RihdaYLyOdcpa5KWls=&D0ThZ<n<a_!L4rx0G&%2N>|p`877BLzO{_*R(zgT
zjftZ!`fr=1WW3hw9aPydkI>X>(3O{lU5<Vx$4e_eWefI??bJMWrUz7Cxlsr-W?kK7
z0(gr>W(;C%IzN}PPRk|o;^YKtRfucEXY6v2bg&joJuhY9Y_9aGcDEwjS!iDPG#>V5
zxlsPZf1yV<Ai-ASfj<1^q4X=^R-os4w0Msd2cQ&EVt&t&%<3tc=5Uh7XF};#v-*+6
zLFn_|1yvj>Oyc&JQ|RB=RR7{zlnF>M!_Tu&zTU`3xG5HupY2;<2R*ZB(@|`H4(uc9
zB9E>4Es?)Xf4aqKu_TRWAA<{O;-(#~p7<|*<4&XT;=??u0f-2L^YH~cJYL!JLNZ<i
zU-Ga`vhpOdI9Z#@MMR9?cMUlBYO`CuETX1xb?a5zi^D$fmfrSMNZi5%ykrXR>2#u1
z(UX5xv14r&YolHLzUJ+m%1qb!YERM7>kvPwWpF^=2N|Bi@19&*ddQ%|IkknXC#+~~
z86zl<A9X*Prc$~s7Y@JtO4xbtWBkzgK9~7J$k%``9Am>1DM}0_b4w<9_@rd<ky{a!
zj=+JEb5pkMS&D&tDTL&FWQ7ajo;gfeU6r>it})wG7oXFs>s+4ybMvBJoLufkUyP0y
zxO=A+o9qHDj2--5L?dT&FidjzU<>sWqK~4^^Sk9Qx<&xJM%>TB@`ZD~k|14PDQk)d
zp9kXLZ!!(Y=Jmakzpyf>fj+;pvwOepTgbHK`0kjt6|a8YT|-^mOXl%>!kYD&EEQ!A
z0{-&L)z-pSI_bi`vzx%sMuK1GteFz<W`S4&A(T-vmMPasU(<d{ptZ`l$eOiyUEv4{
zb`-xg%Ai1o`!>jK5Ptny7K9zIKZy9$*hezA)!z?9pZR3t4RVjml70q&OG86_13q;u
z;vh*<pKo*1lr(Z`h=n3otqZ)z?mq3;z04~u57dp`xjj~*E7{eTtYbkVmR}ORbc;3p
z_;Pym#4F}D2h!L6J5)JhljZ!6){0oQx|Ey5?kJXkE~tLyo%GO9N>e||0H(ihdZB+H
zncgj%M1Q2YP+oqKsSx?E6!ZE8bI#+d^|!*em<b4Lp3grsNB2W!_z?)x&Cxy|a#4NK
zay68_c8ejz2ET&~s!AL^;bN3igkI7RBV}?awA~x}t&(-*Dru$LmF=Tsln$yfN-biA
z-+-I0fKdb$HzC+7nf<lb=z`QABx?;yz3&271F5f=0!FLGWq?%y2%zMwqN3I|R$%6=
z^COq7%p|tjSg(_Hp(0$y3~)izaIn3Z`c&7T>qA|Jjw*GTHa!ah#@x^3BJr!XL+v-W
zik;`Xgg}|v7;@E4(in$>trx3_933w}CziJ#0uglc=PqZ5e%;hCblwFC<|Iq}YVdLN
zj><N*6qW{<j106+Tstoe$nbyvwpkX-Zly%~2y*NFrHwq}rV6E2;udWCR#J#tk_`CS
z0wgMC3fF9t=s9qtGathG%`&;j0MdzoDIXD)nS!2;N?IDD62MI5bCG+yhG*u?f}}ul
z2`JY1i&$ta`f~U{ORGv#Ok#XJrrhExoJ4!$qcNvwlT*j`4{}=DGnr6$X#CZrQRaag
zJXG+Y;((y$;JuIG_zYtBx}Xdqqi|0<<CMVG(Oktxb+t607~21R(Sn=kM@TnihDBUm
z0wcPVb`s8OpUdfYkx2rX*=5%@Gp6ykH&STFAGi?DplywgeP8brtHb3<|B-w*0&aMG
z|CK&ZwA=*xZ&`bsg_q=^x!AISZjsGh8rgr}Fo?49(JZXy5ga<5lJ^b&jQ;yauNPjj
zY@U063PyPTT1OiICj3zaLriV&-il2v-1_l*fV_Oi)S74A&X_Ojdn}DC?onfk9xK%t
z%b=dSJ5X<YK=n9V{mMLHQ{6teH^}?<Z@MM-G9L@XJ$%|S2FBrJAbx~LHuq&Yl!__J
zR`uoRtN41}m_T?f-D)iFL9~zvaAmH6PCJe}3ei>#74ZzXOd9Q+-}^xkiM$R~U_)T2
z4q{)DmLL@qnH4K_BrZS!mHg$oeZ#J1&Tp0(e$Y0Gs)5X@ljhUQFRsbml*M$}Gwtox
z882^|FT1F+Se&V}XQ~^9RnuB4TffxJCvSxY<P@%0On7r2<8bv@uNlW++Q)3UmEDBS
zNlQW|BH8ypp?W3t!14_jc)+`RcpTpFGRFED-?6A9N=o&!Mp@^)-zK1`OUs%U4Ts|5
z%P|t@(wu$NIVSsvonT6Z2dAPmbtqG11dxpGWBQ~K1`++fpa|3nxwwQIJzY7>*zyvv
z&iIbboh376lnp|cM$K6Sl~~ADj^4;uuYNz4pX$CND@O^;M`FaLre>$>bBe^GS=Xg<
zW7MRUkX7Z|TOuv=eUeKlp0frFt<!+xIsn`;wlF+~UW{8%6o;EY$Q3g&Mf@k1K+n4o
zE|h|q>@m0NdXx}A_Io`RzAGFVE(FR8PI$?L2rrvIt$2S3kvp{nbmt;b7#Ep%G><vD
zhT*2!w|y<$W`^$T%5JjMx)(Xgm9F+6FLrrk@8|hp@5?FUmesi^-D?>SQl?%GI7D9<
zA!5T#X0nTqHP6+{v)1vgsdAe^Ef-5?qTd|~ez7w8<o4Dha!0<x!2BUN(u9}&lTv>1
z_vXT-lb^VV>OnE^x)^PQ4p<1-zQwqeJii&?-s$FXkjFb=*srw6R?K1C?SX82m*ntz
zEDAhfKlCwswcqzE2l(;Ek^R>h4fk^vs?VV7c^v~!XN@e_9~Qv*k?PFMM`k6lb~^fD
zOWSbk%E&Vma>|B5BKInc{)alKsmUu1^P61*)(8Fo9e!r*+qj?By{n^ps0<1EuSol}
zYEsv{?v6?vq`QWucoD7=r0FNNrwQq;``o~C${aJq0}BG;7Zlhy-$@DX6EC>}nI%pi
zb1zx)0Ep54+8+XSw2{3MICOZrye6DP36mu`xp_ZElW5~2lLTVjqhq<hT8^yEF`t|m
z&_5>b^~DDS)lXSB%Hk**dQ)vv<LXpk*P1PYsktpRlL*ti_B?N5nP-!TCDRc8<*$F4
z8@>E%z?MUUosZ1iPSdqF+q!%3kg`Io6NXaeU5Px!U*x1vy^12FFeU_^rWiCJ{a4%3
zDv4)PnK17BNr{>!a!SNcKgAt^-*M*(R6O#Y5Q_VdE;&O`9P060i)^8)#<J&|8Hu^d
zU^`HT1;~m&5llv-ASzVlcZi;viF*NUJECHnf~Gngr1>ljk70#ykfF!a7Pfs*iv{ot
zU9{vNpU8;W@2}lEW$y%_9-E2+9rR8(+sH1_hJx4SHawgL>NZCFU*HzWRL(T3RaiPx
z7~c$#yDOH!x#WoxrJbWJ%n6*-ibpbsPt*=}b~9FPEe#$-Aoy7CU_*~+sY2-_35i-`
z7nkf8AR4cqH35%5ck!WrB8sCqZ{`U7!B>hmcZMPkfaZ*uo9?22GF{54<2Gz7cY5~=
zLDzmbItC`DB`0UX2%%KN0d{S&(8J8nENhW@Tv_9u7jxA$9-F9Myhi6jgcSJ3blNab
zYLdU(a%J8R%_)`Q@rvtquCMd*)H$>;?3;!_+vbjBHoC&YjO}gl(Yo&@h0+ys1zPe%
zu}>?lR~!SjVk>BtiRYHhV26kTWRv_)0Qf8j`=%tH1vLcARiGoZzO$3Q*URF4wZh{9
z3lwX=SYHCUcQ=aW>S`*e(m$NA^42m=gF~*g6I{>SYuLm{_$ze-R?Oz$|IceM83nka
zCDbPqDakD7B2EeiL{9qs^I|htyi~$-pB{8eh1M&?KPX@d$=?0W3&7t*D{RU*R&E#j
zLaEu$TpbWd5pWhyr0mQmbftL~>5!ggK`!Amueuzhhcjc#-QxcuV3h^TUi`xy{N9kB
zVGEtZjzR^5bD%49^?Uylr^?|Hx+UM?&20dpXN0b~&`UEl@5oDPTuB?3{qsx%fbP2%
z9lCzDY{B@OPP<?)694MnFYRIf%PmGt#YTqh+YuEb5eBOjjTBR;TTtO^!o&E~-xYI*
zJ5)16HnYgeeIsp<GnH6iJUo=&eNF(#6f~sLPz23x%zAZyeYhj)vE?k7St?jKJ#Sgd
z*SeYw4Gw;<G^zAeIz~Ql9y`_@ml>fMNjyc(_L?%vH(EJo^J)FT&tB+`$>??Wjp)Xm
z5cl^xU0w?zBF^jy;QV6BRl;=SGb(Ir@+Z`#1~F=1vsZli%_2Y9^$fZj0bYNSW+}pY
zy$+GpiPyCx9*(3T>eTus_S|vk;V4iz4aND~k%MdM&dWLNhh9P>a|;jL1iYpqYUd{{
zx*sl@Wvi+0U9oSf&{K3ilGYjr2leg0HLn<;WeEHJMS)Y0sh>0BhBBhY%Vgo?Fh>@I
zT-XAw_<~it``$>ycv>aKv83(7Az*v>8uAn$_q$oJ{qEHV&ucJxa&_Y--O$W&d4Fr*
z+aEpEG$OGkM~pgOsN?MZ{dvNxEO^s#ocpTvrM~CC)WKQ35AN<BR&#2AV74_!kxxK`
zE{-ye$J#m~C7*@VDG7%mY^`KK2SrUKJ*gm}{qcz8J5}~%ZIglHHUTnz&GHVP!hd2A
z_G+F|Hs4~cn9POV-uMoUl=)J-7?*1#nw(&%ipV%>827Kghc(m34~Q#rTw^Zgl3#@*
zMD<8p<itm%1?nGWkLUQ?Hn>Aw(#wMwO?MrCQcEo3JD8h6kjty^UT;RhsIQKWFdLK?
zQY=CJk3}&m0+LpWbT};C=?5h2${VL`x<1u+gtU;7ody3FyS!GX#1Yfc5hpM=<l^KC
z-~b<~hCxy0(~y0&Gv;3Ep8bvk)k1Zu#*kQpnF)?f{79Kf7X-j;^Qu9OX|{PzytzOM
zIFz;ucOYx*GyBY`yVaaNT9J(0$Y$aD#}|~zj&mF<2z0k%fhfLorPU`~uU?V{XQx#<
zze10Hy?%b?GxV`kXf0saG%Z-^z|H|?U!q`PBCtwc-LY28B1cZXwm-f7K5RRNqRS<o
z1udfCoI7O2b5`QNc+CBx$g>~9**BTc4-3`8@n6({vY*?$u|lw#NP(QP7MY<I_RgHw
zxm~NTHh~d3b@0shY2-4ZxKdJbOyS6mX@EF_d3SboFfE=_54;O^U$F}*7piUp5P$zO
zkAQFu(*tCtKy7w?wVOX0<8s0H*fS%@>N^suVja`8vTB3dDDwm|R*60L8Ist4aSjQ3
zX-M|Hg8m}_4t#9^u4i9164HtdM*yFuR)m8rRJImuj}u5(t|>Qt$lG!|E@|j_fmjm4
z7-z|=mIV=T!g(l2I=B6|QqQKIrm}>6krLn7W{@vqEUbEslCIqKbhWocXtdp1d`!C1
zHQ<=Sm5C>@d`B4Ps!$62-IGoK<8F1mz^U%K6-a0QR-)=~zH-HWKIy-I*Wy+oj#Ip&
zE0ts&;+SHll;}e&pSXV=|F1BmcxKf!Z$IX7NJzwU_~H<_Cccm3iyGeJD%nQyvbsON
zwflD>nqNN+82X-ta$`Siz06K>gfS*@6Z2r)V)BbHec~-$i}8GYuz0vkH!7GuxHze=
zcD@-^94+`cWmw5Hj>~*#r3Kx&x-Mtx%kOP@$kgRb)Ls0~2>;(X+{NEqa^Ocb<^|Vu
z1%N&LFKOz83fmgLw$1H?5*h;VAwwk6w`=eV!JVjQx{`jvEOkjT={t0cv-9<rg*R!b
z!gm3O)o<bX@Fgsn7F_gz3&Lvo)ICGHUI!`9u-Q;(mw)s%EuH%EI1r{>?;(<GayJLu
zUd~>Z;#)lS;RC~Y{GRY4b-&CIgLq&QPE8U~!=(sCUsYh1`fIPfF6}oD+JMm`{I_&)
zXaP$v{L`$>ZmV`d>s+HAZEemF19o`hF{m2hY!-#6z!uL6B3alW%S%#VHsX0A-@h*j
zy}nN2yA*Hg=uj*)ytAGab9Qx&Ne-&^zp0qjvm+-LqRmzWuHFl+{tv_+BAo-4-1JK7
zS-((+Q^SCh_w6JG>FFsZ9CsPQ*2uS&9q`Dy=5^-R#BS?y1$@@8nEvC%tBP!=_63>%
z#?Vw!Doct|E=i|=MqC%~<m5VI622IesQ0UoSU)@6>##W8uWN)-#y2mU65b}--mSgj
z3{nCD@3CkgX$;u(dz(6ofL)%}L`3f|lrtvxeE-ERv^YP&2{>DpS<N1C>$4s~sIF#W
zth8V**wh-xkJPB=2j%(0E_+uG{30ulCAvv7FCD*s##F92@$vO%CAx1&Dr@H*?{6$@
zev&!-`xONpir3N$AE-nE@AjcJGocL|--4$fjJ(mfP<=7L-XYdDV?Y|QfwF=|gY!>Q
z&W-I{7dq{=<^M4*lne968u)T0dOZ}gJtt{x$yueMS>+GZYJH5bxK>)?sgOe!XU4u6
zKx{S~LqdDlqk@9Nn;&{G-+p?C-UIRzuTaBgMTf$^4VLYV9<t?CY^#amNh`9=kbXiO
zoswjZrzgyh@g=ia+0OxQ526fq^R2!&s5erhw@}^*``jhZ1AeQ_&G568y%U!AEHI*@
z+IaEH73yirU6e&hD%oR{dKPi1tO-bNx2Aqo{2W0juPH;?!%t<)@Mcbmx$t-k*51dH
zu!O93Mj-C})DATA{`Vr()P&Wva{^QvI;83k2L_v)cp1k8)?ep8w?0h|*%r@<Bt)b3
z)m>}u4eH2k*T@0nPEN89mS+a}bM`Z**9Vx7=wW`|B;Sk=eM-~JBJm9=#s#LB&dbHE
z8(5+fcrWn}o8)*johj_CD*e;Ob`Z(D9>~hnLS|(Sv8t{)S<QY|;$bWO1nrYj<ZZ;5
zxd0vi*%uM*wO>YVYVWvFO|;+tnxh=5<_GO8x96_w&ovHvx9sLUF634afL|e>m)p-_
z`U*(#84C2Kx+A`Z1&uWJ^B)xl`x_%jCRaAbj2=0C1btiq^qPh%FR#p22&PEOgn7l-
z{B4ML{a3u96v;@Cb5<-OH}Q_w*ThNCXwHad_mGFk&RT9;3SB-h*}>NHN<{$SCm*Kt
zZF6Ce=T7}y+B-N%Fzox=Cq*NPT<!abG)^WtgBpp-#L#C~{&~;nNfFSi$E5%(ULM*?
z-NhWbZbxWhtuaEQPgF2NKdOz>0A>PQ<W=vom~qU`!4A3!j}PJPVt_07>}BJ<{rYQ$
zd)t6Kdwci(`lAqg$2BTRws6S|5hlgIs|zHcIBRCh5Y|A>k)4KYlZhFBzZ&p}>G?F)
zH41GcfseKA4t*&RO6o|`tvng$L^kgEb=5$GdlI9g!b7Mg3&w7HQixmS`h0i)dbifY
z&+Z=`-S?l5&KmK4uD~yBEl=8Uly-DSB?!UA;C|jFnvs9z>S~p?Pdd3s>izrQ<$t~r
z$Gc!(FxbXi#jPG$oRL=LyH*wzl%F9C(c3jVj4t?x2HYyYVmB61vXd2F<oF6nOABx#
zG;?Pw!nli7tG6pne0-vfo*x%7cM+p^Y>XaI3tx%;W@ppN_TJv|>g#9ET6e<AhFwB9
z#_3I<Q0Q=7&e|UR=`DSF)%wo7G{wv*x%1VaV(0B{^+m^7K)hh{8kIBMs#3C6$B1Xv
zv^|56zYuDXopj!Qk!%zQfZ}4yNaiVf{oB*$K2rStvY|_cb|%6)R96ml`H{-X&og!`
zgamO$^bJgQh~CL@n7CdW!o2T^U#C&TDVaf;Tx}U`tY|aj<BUp_qfG7P>tP9pMNGK`
zAzU8uC+5wHWGdw8;i<0rXijaKJlycq${a4H2J>j>(DWPAuug@6k&G?Sa^%xV1o2Wl
z$(6p)<QBVf#jvXMv)l&ZHY_Kx2b|&ytj@!%aH)1<QB#KvHRe8NV1T~VE`Q5+p4V!m
zs;F>A*H$?TSgqH7V*4nv2`?b$d5qT8?a8@)avf5a!AX9X85>r#oLB-<yFb!|_&G%<
zB`G-sPprzXtOpFQKLO50Tix^9>}r?oOS3<#1Wg&Re|F!qS&#mlVo$cFU$=UkVF1y4
zlFx=ktA{o9FU|Y{{8Y@WT<Vw70r!gL+wUfYU5IQrm7rllA}2Jo{i+G?5<SiQu<sRc
z8r+Ml6iR1)@d<A(-((xK`tradDHkk~a}Xf!`&NkLfWgYcU-Fcln5b$w9n;<=Sdv|w
za4mS=%%uzUijF<pZ~tAXLjc_MEMv0bW-kCkG-Lg#W8F-EILY?mrLxFl=d!&&Wp4fI
zbW$#x>thZlan96It4vBskyS3^r6l#dUJ~tBP(Y_od5kWfjb>U=`AP6-ZDQxqyuj$d
zqUY=Nq_?g`tyDg+PQea>(|)9!BMJxgwQPTpk<)f`cJ4jD$GhppXBVn4o<uPEl^&5m
z=2%r{Mo?M~4*ySHWcAk*?WZ4Joq9R?h-%6{t66r*Q9#yc`0nlGPWFh>O)8O4=NQmz
ziJ|J{3a|^k!L|J*JZNg`+|5e?iaU8ex4NUYqmFzfjGW#i!bdlVszs{IzLsQ?hDMy(
zTofNNluD544x9e&v~EAfX2;M2b{xRqm!X@6%Yahd^~#hx!B&>0Fen}aT3Y%Cvz8ja
z3VL~=u{UV@(wtq>c*6oD?qA8=Z-G&US{ikXOY5T89FYy=1{GDTrpc=E=fJa2=-W6y
z#h3eqiqQEg`uVK*JIn{dt~-Y3xvu+TR{<Wq8-W|{-OFNm-%SYn`ErHBlr4{NU02uS
z<}{U@H0e#L>v9rf6E`tRy_J&cM`#@Rc`}7xcrNqjh$Mh7|21p=Bl;uWKn$N1+cFTE
z^{+6n@8bO;8<<LN>iljgz2Y_6MP%^_{b`4b-yQRxNJ&5vRY(b=`S7Cl5vbUavjIOD
zI|zBR{d=t#O(=~b3wxJ$X0`8)i@~1;`J8?#%{>~1$S(r;{qcTI{SU{<yfaKk=cfI%
zTAlu*XJ`?M#fXrt5JDsW=)3hNl(XmiY=uFDrSSszjPWg;_4KN=c@??B6zVd%20c4D
zB!P0O!i;_UHB%9CPBKhj?0tG_rw5Tz879bkOR*R)#ofE&hqy;GYv)lZ5%byHkEVC^
zOI6E5v<h<+m#mu^*$inFah4}AWcY}3YNsm-X#!F}yQ<3jq*h6sq<G_d{9;nJ*KGMI
z4*XigMjWkc{M}=-l>+wB{=Of}2{iYgqN?I{x9YFsB2qAN#P}4yS*>}Vni+cC&_#`(
zuPLu>3$8x^QO+xaF6T_zVV(mWss{5U!f|JDnz9^zF{=Ogj%oUqT#xW4Qqek4)A7}8
z>OLF4h%6+I(xK|!L^VcVFa4Q$tT&v;7x@5YuGH%E)7Ytz#2;t0wNl}f$`?b5Q9<~J
z&fei1%dE}r-T!qFA*pm0$7@zXq}Zld6g_m`qf*M3TB0@R#_M`z^n5(}ljn=yu@Mr7
z(_tcgXgBc)u#S6(4rr84BmrFB`s)uSoo_>3(u(PL_53r_8D212ML*^uM#Td+V%i&m
z?7wZccJ|KhzWt%6iqvFWM0u*!#R}5m@!5=utYsY8G5nsM;KD-j^_OC;LULy7wrs$n
z=Jk9Z0emZMBkxcL|Lwcs5<nJC9K4COFSOf7!q;V*>fA4TPl^JQaJdi*pV>x9mL1jZ
zx+H1%+kwEa_uNWTjA}FV*r$UWf(jEget6V0UEF-;j_k(7!3japb^Mw|Z_rW$9QTNc
zS_<^QQ6pSs*gBjm=)r#quAM%<D5QA(8Su&;5DjjB!nI9P9l`Rn;Gv(~&lojVd${e?
zhA;?vV%VrY7L=IRx5B8=xaCVtU(k7@JJkF%z9|=_*eF7km6Zb?C+c+y!|HOj6Y`{w
zS<JMI<pU6PEJnsNR$_W|S%96V*ZHeY`)e;g46RoBKnIkQ{EON^r^_hh+C8FsS}Y{$
zFCG4Y)fMKLi^a|U>&>t3a819a8!jo-$mBfw%kc?9ufqwWja-2I&L~LF90uGrJ{PW*
zt2%HmE-x3%S+aJ=^FdT+P`WOeyGA!(g-f<Ms>Usw^$r*>l|rwwqSd^YC0LB>E)wd-
ze-Cgtk<M5*N4k!t+AQ2dJ-UVfi_{g%_Qj@2RNYSFWCzse>?Pt9O83|&GALG|yFVCf
zxAwO2bA)gI;5-5XLDa4~t!?(e*G7Pcv2hcI-ee|d$|caaqInyyNg{s-?^ug`#k`=E
zpNvk+l3{4b`qt_HLf_^?cwc|!2hN4icR=acS9}8tSQbz0Il$(mWaO9I;O7Jf|EKJg
zHlKJ~-l5dE3wVK1suU_kny6mm=p2Wa3jn~YZ&cdW@ztVm$d=cPlibDES)oH!x{MOW
z;-?z!VqKh%+@SV@-MZ3MBa)h3y$PPww`aJt>wm9Vn@C6*)RLbh%X;r*SkSf@CGM3x
z4V8CSe)&n35v=&$QX4(O#tb^^r19uHiExd0&u~A&h29;+Hd@Y>^8T(cJTK>J4*Pe3
zLVgD4hMd=_(jcJ!Jka%cmXMir_O{fV=p5BR*C>PiPBux!!Ba*w&IHI@3&g3aiWv%A
z_BNx=$W8xVC@aPSQQ+$}V>K19aHd@D9JI3}%FS4EC*tI_x`(bpKCt)eq-8wB$dFh&
zjYDsFlMgchb4YzA76Vg**Tu9-nwjgeVhyoAvAf5e=XT?yPoK$zIwpar%vjy?K!I~=
z{<Xk`iy!i&4I<mj7cm)Q{Q9N8KRnLhWGge2oT*9U$=6EDOR5%#+n+Dx4gR=h-ua)K
zD3Pz2<q^S#-vq#M-^^Vq8v(L&oe+bPpIHS;hYgr`E|LNBzkmv6VatjqL$8;~v9kA5
zOasd~b)Qt2<cM%3vCH?o44`@;F&tI$22CR;M59)fW0oGozK3J%iL}HZ+ZOrv(9-Xr
zu8gF}V7TF{q%4fTg8#XV@#Bm+)ES5Lwv3(=V<(J_e(+t^iP!)!mN^bmCQ7Kr=H@b6
zlF`s87S?}V6WzOCiyztJQc_mNLI$|d6H;hYf4JSU%{#)9I7o6anuar#r4%f<tgXaI
zLDUe&dj1F7FtJbII6h(F3*#-PsCvXGxH4|pqNT+HPGaTX1K4eU{{k^T=j25E&ibEf
z@~A;6l%a)baw)DrBC1e#=ugJN`aKiye-x0}F-9naN<eya1=%atNYas=)9f(TuJ}72
zgppRqAX2}<P}w_`dTBJ_GNv04)fmPnySTI}g(}^=4j@<K@5A5D-qPaN-Xb0xCB(f_
z=Jb5uWgwr{mKP2M@xYbgcTAwo{>}cVpES9zz(A8z4`_Q>CA64l7uLs^IGZJ|UM6>p
z(Q{lnFJ0{>o=?gam6(4Ba4s4v5iuK=)f5+h;?K-g&ERMyQq4eNvn(RAXaR<YD^LIE
z+248>B6(?a&ciycj=9M5Yz`|x9kUWVuNC_vkM-5yaQYMQRc0~WpPk^+x%E;kG%?4U
z4s(m(u5;}-0;BuoY8PPbB>Ea0F!Ch&jIq2+o45r8ucnT#BN~thVcc|8b68Bh%Rz?C
z(v{qB;yWX0c}Ug9<x+5R1K<6A^!x6Ka^pZuc$+-<J)a{)@B2h`Ts<tC^j6xvg8PDz
z8f>yQH&Ui)`}n@9LdzCth}{N!^*D(MxkK`o<P!8Y3_^Z{DF=je`R$&3E2Y0TE%Mk4
zf(2dl40)ASp~gKE^5$s3Xp&#`Ez72usAX|FZ$m$~)YSHzy=_uBZ)e)Hzdr7NRF>M<
z=;><{;%gZ@eWI8V<^SR7EX6zd-$y0V)3o$HnBK2O+DR6Q&L_Lhi>J4Y>eo@Df;nK7
z;uARS**$Vd^*>*vcSmtOdwE*0_q$J(%S0yQVw@wP=n9e#zrBkCmXCoujaTARs6BoP
zDzm(duMeK+##ghfDY0=~ys+Gf4{P36W>tpkG*Q0zsojC~$faV$vmxTssy8PosXw!H
z$fGt%m?@5>S|4lE#ZjuU#|ke|OjwyVjpJ=(o}ZfB*{^K6E*z*N`vn7T8HSzZs97Tk
zX16l*@Ni`%D{+leamtX(fDVT&{^BA|qo85PG^2UFh=oPG-?|za8+(Td148Zmj&7io
z3b4&!o|eOvwu`xW#wWA!?jwNeC2{{S-0PQ>6NV&9bxK>aPFq?5lYjjZnKDPMbDOS5
z!q%F*7$b>gX1{7@`SJ-|D6^K^sqfRD(&UheqHxOp1U0Yq?9@=i7&Y*cXOED}UXX1j
zLOaOTypKNsco#LX9Qu`(Tb3E15D`p+@o&vSOGXd6k;9Ig{_7+BrfPq|N7WsZt)dTb
zwpb?LcJnkn{yZCyjruFH3g%Qc$lx#|DLD29kiULwTP;QgaoDg{aqoBq#_;?}D=0uN
z_9L#cx?{A!O;P7o8_;)H-GB{zs6jxEP44-HE0T4Sqv6;a?Mb{nl%M3HTlF$$-N8^j
z+EZbu>{wpxtwvKP7pq}eoq|IhtMV=p)yKxd185U~EzS?b$wmG9M-Dl}kiFj#DJH?w
z@!b4YES)}2l7{_gBO83-lG+BT@g?bam4(-J<MB(T{Yu^cNo0}{GsK7@dor1BUid#5
zP6FH?2!U+QhMYW(G#Z0kv?FG1iPp@=<iZB&Sp8MBXGp-E=+<7WwklI`gG&r+dxmw(
zXq{ThoeQP-bnKa}I)ux^9z!c}(#=C!F0#Qmuy^=p6^674RB6(Z=NNgjLVJc5QUCnl
z;2k~t2;zI<+QM&ljMlo3E*i3d%5&OFSPsa=v#y_SBVdTSA|{;VTzP<rC&IiQrLuG+
zZ8;L0<8M$^DEEaIc?M0oE8w=;Z?d%2FyM-3?PVGWM>8z?##Qh*rkI-d7ab&Lzk}}n
z{D8rg(rwDU$FP2n*s(^Sz2=?#dsV_R79v2U`RyAM!heVQ7S_3yBB^g!osH||CfY>y
zRKcW+UdyAsh!B@c+Fp4}lv<e^?-&V~C14x+W6<uJnwyLEEE-3_#s6o_yt=RcGazZ=
znbJ${_kUw#jxd&7k@+OsE9LRt0=@nrVaZkjk3m8;N~`lP;>MG<zBKRT$R59ynu*0R
zoIqsczKfyXp=68;FjvUoK@VrJ<;G>n`ul{*=jl6sF+w!6ltVq~$3TGOg~$|WVdRNj
z)^5T8a;!oxdm3%kPiN8h$2t#@<MS`mZFBtBdz|)EnN{4ao=V+cZMeofhL%(}2n{9k
zKRoX0Hr)}l=d(D(gno>ZLZ;#C3;={yG&m5;f3}=LB%em)>w2+m0@j#{yLIa>e*W*C
zQrp@wa8>OeXx=7W-8=w1Pm)JM?3>#VBV_2NSa`fD4EyD|by)HwnKrQvR6Fiwv{M7<
zhyFzyc_prkywE2*5t*`}cY3p<Dg`-^(dP_OPwf<?82sxuShH7;7G9<-3t^?J!r{j^
z%+O(3R`J>&SBB^-EW*j^M0~9>DmW4_Dkg<{-om^r6=aBDDx)K;b`o=Z$=tsm3)rnz
zT}YGmSThk3pjV=9`pnxwvrh|cQPC)!;uNb05n>@=U@5MZP~rqnd@7-g#3^!VDWIL?
ziYn+P7B8T=T*8aW;fsBa4IjMGO?+x%l>3w!m$~bLx*cjolZyp1`cPENkZfMhZvw%>
zS@V9fd+l|2Hrs01OKliQ=<)I)wy^KPM-9*Lu!qw=CB<PX7YlYFkymNFybNH7#UTwE
zl4i(&CD3zBXeATowaMn0p-cNwu#!YJ;Lvf0;*UFj{c&FQy-Vi#1tokyEse_@sQ2q#
z-I1!LrzV?=jb)-j^aL!&BHEnS$}yEM4LeJ`CnpP;ZnfWf^;aIwBW998H(~T~tatXJ
z`2hmG%cPn%Skl-6Q4#fsa=Z}}A<PpYHm4ssa=&(jC6Dp?Bg3hxgpU5h0ZASZoXh5v
zXg+(>&HcGICi>(xWRgqcN*=3|ucAp*hZjxWiy;ls0+t{)?b*DC;GFaG?_m38BzP(%
zbz@s4-4qEcu|!3zD%yDzh!Q5mq(RQH>AOfkL|vLKXHNzxE3H)yd?D9&AtJSkxq}A5
zS>g`@5X7TVGl10xXlm4X7FYiz4q}2_#ED?Ze9`kb;q;SoIGZgGE71mjoIj9B-Wy2Z
z!M4YoZho)SF$bfUaT5&38H_Rd!Zi5O4*4KFy`T^+v};!(V)iL_3s#^b=J@G*z?;a`
zMa<>}2tsV@4&V5Z+D`0>CE5)^u0>^oI*6B?l2n!YWpR>pdlQgF6HL~nQKR|z&Vlt2
zXUDjFfB}?IsAbIO9YA^#a3lD3gnKtw+F|Iu>owZG@w9$37gII=g(<QgVbJ{JrlZ)q
zY3iS~P2bIV`jX$B`E4s7oSaOx`Qz&<Cjue<3saAMH-w7szbo(CZ1e-%gwl{tjM>D$
zAM`Gab!^>*Vp&t=Hdxyb9OSBrJdoTb|EE#Z;=Rkn-t|n~+!-)@YaTB#uvkC~wpfH2
zH+(IE0X!ZFkcmMKHL5f(m>R**+m?=b6WmiOX;Dw4iZ^p?2eY0Z0pPgO?Js-B=;7jz
zFI;jqi+}t>fn*lg|Lw2u{!=rb5V3L08R@Seh^C&ZjoW?m+ur)ln9&Q5L``b3I`X0F
zi*kW7r1-jmYf;l~aDElv<6lRG=x;_APSaQgT@?cU?U9&G$Bi~GMp;tnepv;>os2qD
zKH}OpN3alctE*EyB-c*ldE@<PyRsF!Y^7tB1dd(~zvuM^uZM_t`YBpE@Tde991U@B
z3~n8`Y0A=1^ZxT%S>0o;&F&)(RSJn(I5w&Ks{W(@yM#CdywZ^i3vFReb}HtXy#g2A
zhH52h#$$PO8AV%8UI`@u{{CEY>b*7@GMdygbF42?@JUz_VMh?oFL~a9`7HF6dRQ16
zvY5BH>Q&%NJxX2J@#eR7-&CVVmoDhtD|bmo9w*U_Rigj^7VAd)m1eeC;cFcKhtDcw
zvXoChzvdT%;nyUI4fK+66;3f`qYo!BG;!29JQ>vD^5H5X#U&|HQXDV=85<x<YHRf=
zG`>@DbL<HrsGz$UX%sE0C*pK9`kDO=+9h2va|_LM`H6Z6R$kLl#>?{!Qr{N&QIqRp
zn7EEo1Uw<KEn?`^m7VQ*2<Kw2${aS+ELuIV-VuQkK8?v;k{T!)b95dJ%Ys`7;Y>c%
zFU6FA`raKol#~nUadUW#9TnaLch{Dmd=u`f9sKO>VSl=9(Rqy$aQL}v_@zzSprc8;
zE%{RPxi4S@{5l<9i`a1EjHCoAs$s;Q#^ooer>P6~7zTHsl6(_oPZi%2E5&R#2jdaE
zk7gsD<6_-U+Mo>RWh>hgfESBStaJbI6I!So9Sl0R^)t1iEXoC*-kt8iJa$Z|Q_xf(
z7fTXbHVJ4N^(WrG`=MLVcpY(V8b>QRN?cGi#bm}y^nigX8RqthNO{DZXyfm}2kAU6
zcOAUhnG~Cas@m2@#WutJ!TL!TM@I?^R71~OUHf&Bjn^04a*YU=pu>jq$LKF@GkwG(
z-yOvu^J22q-y#tTQM)UJpRDh)kB;Y}rZlU@XXcZf$g!^uIX~PhgsGw8nqmR-NvAE%
ze-^kv4blwgR{{ny^r3!LiL8{ONE+C?C1m8ESOf@*Q$9;xzu&`;@2M*ncpxBuD(2sx
zVkK)inl*H@*P=@<-)D*amb1P72KX(`UN=Q|)?R!>$Ns(BnEPDCm5c6m#z65YJC4IE
zQVpc{Wi@p+2k3(Mq%Y<NZEO9BD15^7bUhZg?qDlR8D69^#*|<L2`ekJ+{HIyKqO_R
z+1_tiqk@rRD$=Z(bqi1}b+lU8*DbtYQDq5Mdgt7Y{W6Pfj)B_~tRV?8QtvS^X$lyn
znAgi4Tujgc^Q&k=6h`)l3NmPFX$7QY7{@?Rq9SC}cTiOVx+S5j!+CESc>W+gVX7z|
zCRl%235&~MJx4PjosmBC?cbL6EBeWoK~RjR#(c|Wv?+SO{yYTm{q#GpLKD6VgB5W6
zdYAPK<+JN^M&?F$Lp?=)kcrlU+ox(xb)StP18H<#d0e^B#icb`n25~iRgSmb)t{I|
z^Rr93KdWrO!nyZjUpu3vbR00$<}tJX6k6DjCzUu+LMpsw!Fgn&D4Rmd2mvSwhBFsA
zif&MZj!QSF-OWkrV;@s<)hMFx|G$sC)aR_`R?ONof6OE5@<t1f{$`oR(y^~3dqT=e
zN>U(k7tjB4KtKpA;sG|e9*v{U;dFX=zQ@fwFz;rCYLO1VY56-8mq&-|6OoZm5T`my
zK228p_H!m2@wam}JpkJt4+P;gX!Wbt)vpqx<7Byi4wxcxfs@nyT_oj_&4|+A8l=Bb
z7njk2f_wgBrSoc5B(|c4e1d*QU4NC@{@m`UVrm$Ve4G(8n9+;tI|lp8#(0u50a1L-
z?YcR0dD*-A67if1)Qj;S2a=`s+6$Zi9OM!F(2QK1i!fc4HNJ3xbTq|ddDbfK^GCQN
zTwHX#Jogy<r@OgJFGuo6cLN(q+fKtIb8>&DmHne`s)(*4RcOA0Q-J)A(-N`Obl6}9
z)&)fLZ3b0J<oL$u+31b^8T^{uHFoUK9#Ob+KFxu7mP>7*ki{FJ5yQ-C_V4fF>^CK4
zIT)vfx32^*@n#f`EBri_X2McJr;6BCE;~to>W50mVd&RO>R{!LAs*?@EYUPXh_~X+
zQ=0#G9;LO`tJXB<X%`O{oA8D8N1U@|kW-GD{P9|A^j1DDBdCsOdq%uyUdEz!p2V&g
zp}0}zu-GYz1=&9>jdSu{U}_MpY&Q^vDjs+FxAKyq)h3z5<Q<E6k!ryw;x(Qc4J=Ux
zq;tsPhoj*jbgOHKp@CkiNhNNE0jG9r{!f6IQ$XVbUkEN^10ku_G3a>Ct>7j}y#8oK
zIPp@{L-$x+M$U%_J5YH*PaS5<JH0%nWLa$G(bj<E2cxBnG7;!t3&(tNJ&u!7^eL45
z&DgAx76*G;Mq|-qijBhsKVpz76aGf2uzKF@$isx+4MN6bA#LqllpHGuWMTm?>(A?a
z0s@Z;KD)_PN}uVKQbe<5Yg*~#vMmW!aXX830=p@(`KYyV+H%rjdWpyQyzUdCgQ--r
zy!VywQwV!2xcGi(LnE-zYhi-RRj(%_ato;E9>%MKl=F7uM~$7*WNO}iGOA8rIQnJB
z2gbBV_HjizUpRmr*2Dw=Z*2A*3Lc5@Rui{yBhJO(0r7dB5p4H>PS?WDP?#yl815xf
zrBEa@Q)9wuQnYp+(336t8&en+Ym^@Gs&H~{l`~@ed^&rUxbfb%sLby1XUjr*y6z~R
zrQ09(2!*K+F%3h`eHI``xm72Tb8cF(l8vug!Nm-xbC<%=U@tyS*Lp@q`A}z(MniL#
zvu@C-zSpDCPPFn?2oPl2l4s`xkN-C$c-Ag20<Q5=RX3%ZsdPS~t3lrY*ljZ-2?fIm
za@dCv_4VpXN=lZEdU^6>J55Zv5|~1s3;qu}n!hyNYz#lPTK7F#RRgu;m%@PawiYpt
zTt4{7b-ej4x$p=)WMwEJeW}g9t&OZS?Qbt=p}xmbn{Rl-bbg}5tFk;6jOv?NOL}t4
za@5|l)bAG2=xXgwqhe$5z5f?VD0`mfkSQx1A*|UJ2bJh^tM9fXZ=0fVqKnfRPS8(e
zUG~qL3W1!6o9LWz)4iM`g66K?F{yEU7k>JRo1X&*e#DBSP)sXG(Nu!U=8R#qR1h!S
z2$(mgyP<b%2`4+_)r$D}AdQwBip1gj^TXNm6ST?8&u_Qudw}f-zi5HvdQNI1#~`0k
z+2sT(O;BSP+K{!T52#Qda#IC2_}|Nts#h#H_BIPdQCI&kKIU0A%eECMPAlO<DsI%-
zO#T-dTA;y)2gUypt6|(La;isMy;_>WU{|?zwN)c@Cp6|(>nfwwI2@uQwFY+eeB;(m
zGoy4l`Y>_!QgL=ZA^He(=x!oyJ~+lrpJUJCbt31JPSKQtdRilTw#9FOMbm^*ICM?Y
zG|T(XulbL<H|Q7J3@6F`H<H9Z<yIPrgRdfjKjdhi$y5(UR4CQH{W(en5tx(}!I#xU
zckFb54<gdn@8SNgrWo_0#`b(L%cM!^Po*>Cke(RI$Jt^G`$JjZDotw_$z0L`Q;inh
zdfNEk6dz??cZ&~YMU0)EOgbw$8JC=b!g=>n6HjF!>+~P}cF)gj?dl{ydof<{v~{D`
zZSewQPH$NTaIq4)>m9ppau-`)7$rhYwe7{tCVVHQBdF)YPm7OtmYR}MEZVL;W9NwB
z=2&rz74I&8te*o?Oe&$h5rCqA$!datY3B0MCaic&NW>S&m3EB+YnxkT2FAmKCV07U
zHf*R`%J)i!97-YM^+UVqfA%Jo#oTmP|J&w9B;S_>>hYY-HA1`4u91$R!trJTxvwmL
z%}>2u9miJ%F*7EG5M(}z>k>3zxd!oFe;h6nN0pzzr~UFx4$lzFLxdz*k-kVbEQ*F;
zub;xehC43QRP9sKya3a;xj7dn)_jPFeC>u80`*jKwebEn1Q8kNe4jtN*vJ76Hxrw9
zbJmiyLF8gf4<FYyk{#VW5@;KaLfqKC!jJ_i>59pIFNSI*qd7m+3u*AbUcebZH_zo5
zXUl8y_9|TW5VN|9HcKLsp`uqvtO!`6d-;hQ1xQW>n$R(bRZ{4h8Vp)5(W~1Z`N2J0
zlAcEiO&F7&oo+wy{}aABakrn)f<T78|ICxazJSLG83A#HPtys=CzO@US#Q3xap?0K
z{WfLGeSnGcCznke{;czfjbi2T=V}wJR@j5W_xG$LNblf`PEXW3ywVaUxVH3i`DBOd
zVt^1VJkW9oOoQE3e_P@jyQ%qiqD53%z74bJ_}_xVG9ZepnBf^^9V&|*Iz^k3ke#HX
z(<Iqb5+fsn%}af}zw$Kw#NEUzE66HAyqR;JP<De<;x?vnFjKHRCwu!lSjfUMEisYe
z3E$#Wo!+vdt2$LoCKN^DreXSPCMYp?+8z*{hbb^5BgFTo@ayf2hx_fBL2VsJH%~D~
zN=>{8*=cel#y*y@s#ycY;rU0YTQ%f@VeQLPQO_@U(g(EEO6nm=2aJQV6yJ2VceO5A
zXeR5vdwx*!;ZGp>4BiiTp?Dn<y4xDd>tOQD^h7vhn+@J9Q@VPPIr61#eSC~s>=1Al
zLpC_T_>mi^^7~inARmFZGjEYBp{3N^{{b;U&c0U0`Bb74zU5I~_{u&n0O>&ZRx=Aj
zQkKCOMT_sR9*&ERa_;$-&YD;YB>Z#uM&Z>ueA+nuG?$h&e(QgDpKDi#eEXX};P9YN
z6#G_4^TBP}<uM&$h>%1WwDQ&B1rf99jE}zaXXJwk&%W^{8_!&&zR^WAY9ReQ8<r_C
z%es@kG(>_GefB5zm=x=sUMj2SkqsA8zE#RGq(=nd@OCRHT;&fzS*64r?Tri6BS@H|
zh2!vGpLwQGT7xOcXLFPVDHKVgO<M@!dehrmV>mr}Or#=;a>{Jx7EQ*>!lziO!?4|I
zG9C|cIQnO&Z13!NeYPaRiI{qvP^;Csa`h@&YdYO+()upfpL?E-ogHeeHbQDI#%qjE
ziB*U1qC(s|Z>uR3tvu2O%QT!0ycugfnV+XP10Hl$D5<g1Voiy_p|z%7Z;+%lga|F?
zLdB+%k~m4|bvLNj8(hA6l{Y{C7T@~TxB2LB&dvlnt%yclpahf#Oz6a5g(GdYc;$0n
zp|i1z0LJ4H<MD`jHfQ(3?(&>HcU2Bf<-DDDRQ($AyfNqeGo)$CWHR}w_RRU~)(3aF
z_1=BvQ->6#r|``BBxZ^e5rr+8NiaC-6ePAFQ4ui)K>=D|ouQb|u|<hg5*x&pwW~TH
z`L&#;>Dv4frq4TePbv#M;dHG2{G6RyeSIY}RJ~5h3XzmmX$NV&&CM5H<>;V88cQ}h
zEgJP2N=h2F8qOMy4j)sN1w}StQj|R2zm1L~>g^6fYLt#Kc}_8(F`LXtY7Kh59ztxe
zv%Sm1hx=snIVVSlOvXcQ-h7r$w?}CWO85XhQV|>79&f+(D*b-ngAb+Q@aTlcCj(x1
z{&~{0fwPXPU$1xSLVhRcxm9bXvbPoTRh`?_`t#MW6*6}&P+eIUPdlHTbr@%X98ns|
zd`53$8=)fqvyhrC%Mzs|&1REYT4OXRIP0JBqj%mVC1<N8DP|5|9Z-g|)pu|aLXp<$
zBuNd2Aj@-fk|KqFj=^9+uiF87R6KMM&P#8Dsy2!1j$LUC#)K?XD8QPG2A_JLd?v@v
zS>=ht3Y_rGY?bqh$%U~vYp9!yE(JSLL>k13jdesu(!ckZ{U5zUtJ|cRUIkV6!o&fs
zGr@qUm6U5`oY$doZ#cvX33K`VWdQ`Dn((E8TC{ngijIUwJ}O>=3{^@}K_wO_B*+9^
z-zJ-!u+`Kw8!bNm@O`q`lt_Dmep!?-^)0*3jt@CJI$&>Smzvg;^Nge(;haTDN17&N
zbBB^1;*~-%oz9ugCfvJw7o{Y-dl!)^!a6Xe#TA|angy{+DI~kwTa-o4<I@S3u0PM;
z_*?%zFTMIE7F9u2;UKC|u4RO9eoRF;wViWVAuy|)8c%3ok$`e|^lW9bEe?5Uag@G4
zrEsB2c2N_y;8Qxh(2cQPR>cuCjzS}DV_=?_6opt+=em&Ma2A~Lod@!4f|4~b9)8C;
zwDQ$}))FkIe^2G&i9i5=Dq!$Ju0pV=T=tLYJZb$iz~G;B)#4tHA7*@f>pRRw1C-Q6
zNgXqJ#B=sB7j#Y{AQe81Hg=MsRg9elmlBN%9{lJd#t+WeedYp}Uwn>@=WkNGbP1cN
zrKqJ?qLfui4G5ouX4exA?h-AaYHnWHqctN{eO?W!{5;!G3AsFD15c=0Yr27KU1dH#
zBpV+yo)zFG)axnII`Uc0>B%Yc*^GKEW$*G;n$0da$9&ZHL8VT3@aPe(-X`;T7EnUv
zQG<Zo8dFl3f;dT-jYk|me#FM6KhS5B38T>|c`-vo0wWc1dyBWf@aO1m?oc}G(R2r3
zQNkn0&IM(8x2$ws9WyEAT3^)aXRr*Qob!E)aRRr35(}x&Dn?0(aV0s1(s*u|=h`X+
zYT*D{vwWY$I|fq1%^TMlj)wgB$3NlDSx#qU+3qT~yMC{ffUc!ny?T=uUVfXG-uM!&
zcE>kK@AsLEebe_!qG26b@_ZtRpKG^L)wn<(g+K*B*i$S7^?IGXy**A&PJY^Lvv&R9
zJ9jxem|{(dQWlFw7h}LNAz>^8GYW*w&@QJGf>a2c%Q3>DrROYWlL@xSy^$6IVWkg@
z@L7pCj-N`u@ZWnfI9f$GSMAZl{)DshsSe2EJPvC{DU1k(g@6+Z5jWV{yGFCs!WEe>
zLX^Jaj81CAQgGxRF&>Y=SxkA#(a~dC-5x*w-j9(&bK%-mIvZV#gZ|MWr}ysQBxs#b
zueWecG9FFH^McdUGv;~5Yj3<kv)RE3K^(;-QA`>qT-?3DrQJP*l+3e?t5>Fg;;mo$
zIzmf?af>lpk`}-r7tgyoc0N6-mT6alEefc(qoOmo&@<dEvMNuzgb-fGGh_p$HI!wx
zI4hj9l-7`y1;&BaF=?8TBnfAOf?Kz5GZ_teZd)OY^=P*WNE(o&HM+e`YV`(5lF;dG
zpj1RzniVZnZ>Zu$8g}WNda&uq`(HS9HtZqm3kN^=;1>VcfBv6;^2RuKeMZNw^nJ*M
z4z0qxVyr1MgvqJp12#p@j<P6ps8$X<9ODp~<>)83*xc+AHycD3B5bU@w5b!JT5}0;
zm&W*LK16<t0TCj28c&kB9b(Ps$i>%Uk)B$0Xr!ou=wQW!^owbM^!XFrq9N{p5j;HJ
z$K_*YSx%%RQVA~wgTrM7!_z(wAMSJc`Zc;cJ5V@`E%?cMKV~$Uva`2GYt$#77d(1&
zz@59daaiIw^5xE^;9!3rtu(Ff24bF576o}WCCg`I`J8DnM=Hgoi<fBD>wIzh64zgP
zlh@w-T;P|4F^9!sa)1*7r8*ZiN-2XDt^dA~m567BQk*}Kg1{&12!%&<N?*|{@D6e%
zJA**5B=lGMQQmp-6<XGYR%gz)oyki_Svn%`1!9#8jk+DydQz?_=QyLi<#jfER3bJm
z1Pj5Dv&+9J{jWTCu!3;m_160Jp*@28vtEsK2j>hX%Mo}XPB$NLxp>SAkprhNLJ$Em
zAeVEYP_{(K2t<s@9mUy{lkta)??2+kXw0=vmpDl<&Mm~4t0TilqiV&_EQP4gqp(Y-
zPOj|}7o6_Jygkw4FCaM1VVql3;5n9mb3w7tF)F!&t~VHu$Bf6vNGV7o&Bo>yNz<oE
zCW8SFAKW8~9#TsZI?Wc%R+~wdleSx&oSjmXB}&EAYIVv`NMo%dFLHDg5d|aMKkI|=
zDb2&9hdeyoCo6NdFI?sMm*3*@)t89V4klFSS8GUw9IO-)D?G<Vu$~&K(B3dOYkOfe
zjqi^jfv`();!4C<Dc=FWIZsPby5hmQz|*RT<9?tkr@Jh3&dyF5jR)jKju3+D*RC=h
zPx<JhJDg<&$<)wmNJ<MyyTMn!{#SVJ#kc5eZqw<tiIwE+<e1^$lx#kuR;!W5@%nRx
zXHg+w@aLvS{wil=k#-M3y2PKYQ54Z?wtnWlQ(f===q`Dl;V7|2fheij8463LYLsXk
zbF3{W9GC?{YW=KO$9z0vd~!x(vrp1#pwbu{M}ey1pYgn7Tg`Pf&8z3?YASHCrz#1P
z03Iyu)$&<Z_n8c~sIoz#0wAiuNJhx0PQBr<A;oEtOtOs8=>S*En9Sx#;d~~@IyQDL
zkj-=MeRP`}&%H$dsL#g6CN4?YzIX|tBK99YLP^b_f5v1sVV=*Jn>kUW8JwN*aQ^|%
zz0gK!-#0eW5pfh_7NXk*?F32G=9!ys@!Sh<gVH!xQVm~)#`$rqX2ON>uq(mhDt%-%
ztHc68sHi~Bx+m7?a$Hx^<RL@l1UOZVxi7|ZE~G=HK*cdydkVsIKIg&yLsIFer(kkV
zN)fIqWKe;w@jA7%L7q>kH++l4`Fuu_rh(sQXt#U|#ivmlyo2e9uipMuTtjcSWVr9)
zaKzvJdw-7y`}?12kHBYi?5y*-Xzzr1nSxwOR57Mm4%n~*_TrL;kc^EZvz{k{lvv|E
zT{FtK^Zj>8Ha6I))rdAVCQ^$EphY@Ihy{nOnxr*YJQ%3O|EV?@i}zRGTQzFu9lOOk
z3c7bfs%4bCvJ0z4lr}ol>M7;qgwbe#5svwMj?QzWg_FZ0PL7Vbe)9%<m#=`7l%hl`
z#m?2M><Gy`%envP5rfkcCX+ELQIw^jq@W(xsMYGUS}pqhJ}OR$q8Mup!^wcrxKCkb
z9x+W5I^8XLTi0oCUf|W&zJ$_saK2i1-T1CY->jtS7BILFOSBVKdaEu@vi0AI=j&Df
zs}^PDlzHb(B14t2Fzbe3;c%XVq5Zu3dovbFi|?bkfTiscS+t@0ZqYzFtkpF_ZrGS&
z{r<xre`1}l8>^>tjeeS|nO;U1i847R9}!8*rs;D<4rmI8kr8ExbCng;am;~IMm}m+
z0ukCx#ifM)kjEc<#KkXuk!X{(ITGP>S(Zdmw6=H7H>6)h7%FA_b;r(^<^xYSb>|1p
z;yEOXlw^p;tH8_&iL;uh-XUpr>5on+jHA(rX>_}^(}W_=iK3XjOP9Qg3ldvcN}%3s
zQ`_97xcxCFCr6mV1g$8~O`Fap_&y0xuhnRGdMF)XZOL>#=F!m+v%+!hg|~U*?XQ!h
zEgTvveTrX51y*2<s|q2P2D&o*tAXUo7#bJS)GO<3@!B8*;K$V?%GKCN6#+rhDr6P3
zr~=W-;r7~fzMw?nf+Kr+ddze-T}Ufz^m@Gf@{5edWA^VI_<pIz;gsUl&;25=eC`)%
zw42oGp2v84ddAVwBc{_aS(f=uH)TOh)~FEj%HQx#rd+HSU#TZN$2TY<RP+X5;jF07
zpuhERx~!Ba72`?=W=ccO=`kU41=2d4OqpASm70WzlsU1`q%I<|4q+80bBxah)Q%2m
zq?#Crx3+L`;wkY$kQW6}6fNdZ2(heEt&r@+k?>?e#1dfedj?TyO+G2|?j8LqBCB|7
zONVc<#$*CX99q^%6wqu~gza(m@FP@|cpkE~v|2slW)tOnu-0jHnGR>ZN>zK*x6$b`
zKRIPKn@~E##`Y!-OIhX^Q(%gc!Re`QLXgIR_g7Jsyi-!z9Eatl*T2MTpZisuu2Ym5
zQkaFF>0<8IoG>3jdkLO(+EqmxT7oiF`nFnk7GYGFZ@X4BS5cdUkOT;TisP!3_7Yt#
zV6IXUMUkg*gnP2iGMh~p563ig@DGF!T0`y&3&e3kqtT&OZxZLeqPECgNb%1?(THVl
z?*jFDeX%c=vS8~4Rm&Wf4~+d;<j>ta&pACi`!w5QaeXGoPGAut@{S$gY=JW+bu(bs
zp0KB8Y-$AtW+CM!E0IAgC63rg^1}fSzWXDZoi?2`#kD>6V-c`~gS65SI&Wy8Iyle&
ze`T+&e9wRVBqw6|JH0dMMCF8N7DA&eBLJ;6r2|8RsvC?<5Xus1N30dvN`%(jeDOuP
z8$B$7NJlss;R-{i(I!c1%;y;@ifPtU=JQ!-!aVW|T}q<321t^m#@X2^?M~avIgE$Q
zvI#asLus>vjhkG$^a3xx_BJxDd80~+5)vofVnCJ~C$z3{hV#a!I+LF?N=shEn!&Fy
zF}GC6cEJ)zDN#Zolne)as64Y35=uDes$abB;DN&;n3s;Mv_#r3jsQ6N)U6AZPZa_8
zoP+pD+F|SK;z?(3)i^C8>d#PW{z+F&wTKa@M3T)%G|MS1jhDHx4izO-$_k}WM8?r{
zh8#tSFM+9?3=1f$nGVJj*%W6i)>|dNoC(rFD_q8fxaYZmXt6eeW50aXb17Q?zbE$M
z;=LjSMpCWqu?2_5c|x;NzV%qM(__|ef;32_K`WwK>iLyLNs`3W(i&x95lYfb8?+i-
zgmtvjCfmIpAAWR;M@NV3Y-|#R;sxowAsWpVI!@TVbP*{9w{L&QJTKV3^gM5V`7aSA
zNd@j=fzq0=FMK;Uf}^%D+zX3dT@~gV{ue}}*H0n8&a1;@)xoaPPY7RKy7&{7Lm4Vn
z0ZdtjHdj?37PzC<Ys*O?k-=Hq+1cj$wW}QLANm$f29kD@XJ2}iq+au9nTnXt3(n4t
znazA#!-_~g8jYye>x;hFexLc^Xmu_57`o~k>lRM0bIWz2l;rbof1W@3<3IjXd*t7A
zX(EI$SPG)@j7!msmNv*TXTwgh_%x;$RZ~EW6dH|25)&&;suN$NhGlegiW*Jn%;uz5
zuOPHXdh@bmRul{LE=iInK!WpU#40D*hCNpCP~|xdveo^uSQkQgj&b3|>XnZCny!34
z@WphCg{82ROb#ECP5QK(HPT4ZtTl=2b(Bsi@`8G!PHBDdMI6VJ&X5&`yfkcYZP979
zh@>K$&*%@%7*9rwry~Z#K6#!o8I90!iY*H-XzVhyG!&-v=%9|dc;#7CQX{i2q)ipR
zN3~xY=Rt<WII?QT!&rpunv|i(i>eUvB?nLIum>EW%G9mcu3!VGQn5a@t8~zZU2z1B
zF%)@Dp3S|Mohg{jXB0&NDu4=#(8xP9O|X{JlS3XK++*j$Rcf^+Nm@sWXu;dVdExbq
zjg9cktk|~YBd^-LYTr~2Y_&d`%_c8C|2%*8z3+X7eexMDUsLBJ1?h5}nb0sZE<^=e
zWVA#{WWj`XV8Xe;`|%05(pij<BnAez@3G%&5jE;Gbc|{=KnltrH?gMHEZ3`$ss&L+
zHRn?XPebOaZWiY^RP|gwl_bzY1)`~1=-@1Whp%)KQXq`>sFajMNkqL?Bh?ASF;Xke
zPW#M8V>WslMD?a`5Ugu7Yi%}K5j)*2=F<_g*_6|>6JLxmoiLqEDYBCJEF(@cIvX9F
zGvr0aWH$E>fb^A0zwk@H&i39-BHh3VkM=lIAe?uAEb9iP+8iOCd?HF(xPB|oT%iT4
z_G~e>q7q+SmAzmEnGlNu3&N5;ha6T8x?rg_x3r6r!Wr_?VJd}=K+vmX87gP)ysXLk
z94|#5&pRYfckG^g?I)sCpW)i<G`O(WpdP=-;X7ka9^I$3jtMcjn^SWJM1&GSw!|rv
zb2tpj`2z(hQ6dx$IqL<ojYAY6)eu#F+wy*^!>~dC5UWnA=R#E|!Kzc|ow1PG7Rwau
zdB?7rV<!|+2X2yEa%Vge2*w$&>*&BKK@_LxDDvo)R20^b0#TG8r6ikW)Ef;D5UB*2
z!{MWShQkpm)(D|_`MH<5b>{=_-MddCO-bU2q9|y#>Lh86OV_S)dUndeqkWFfPO+-S
zYhU~|L|n%TgQ(IuQV>by3Cq?{mBLnMj2CEDzAfOwfV;(7@ok!P7<b=AXVLhZ^^|wT
z-C7j*`2Vd1Un1<svdGq=d~kteu4p!zT)KRfv;HZAVV^jP5mGW3oUyUd^Xc87|0tCX
zHqBzK*Q)bvrSw{HP*D_&Mk89S)}s3aROjFF>*}+Va+w}m7_JqkUP@kh`6YI?w>kVd
zNS^q$Z-5y+HDy;9#1g<FotOO4I8vpsNRCR)48J#IDXBR{Ef#EE*rBn#O*Woz^3ffJ
z4<2#_NT0cm5|UU)l-4XdizvB7vMbSBR`x>WbX4Qu_m5ng_f=cAI<Cvlo{$e<%EJH7
zfOe}87&%=94}f&o$$;_EUA9^=t$K|#Nsvn8gs0Y&W$qEkEJqXLtPAK;LA%jltKH-5
z@RWzgC(LG3@?wrqjwp&~)EY!eF&U4MDrPu1WilONOo4R{DSWFgQ;F(I;ps>&NO-&I
z=s)k&`81<P))WHm_kauGd;mCPl|_TF2P#U540~$%JF8UYGE=kCMF@d)fnq|1i+KU+
z@o2>0?35zUn9iom<}+rq8KyK8IOe6JrX@mIMCpavwQ!8iPPp~{k5D@H)Q0X3wNU6)
zmc?RUt)}$PXMNUu5L+R7tMvl@s@LoMAO83MJHGkNKjrWKy}$RV_Q_}D*j0n$5H2SX
zGcLL_E<~Et2;Xr+#y(oHrT4|1FNP=t6j%zOh^<5C1&8n7qPewA+~^>p2pROvoO27#
z$!b&g^$I2F9T;)0)cF%LW~<TG%4QNvB!W;hA(iq6HFjm=R;$P&j3JU1B^(IfA`7L6
znsqjsb&8@uI>GP#)88hR5jS3Xh0TpEk|?1b*GSVk)*4DvFqw`S4hNj|PwAhXa(sNu
z$M^2?%4;w2OTYY!IORxdDS440Rg4f3Quy9|?d~2r>4X-qf-?8?U*SU3|AZ40HlSVX
zxb?;Ml=1OnUez?MJ8V8GS_d*J#9<X47Zn=E`u(h|Ebov}IcwgKS->F33X9Pdic!7Z
zEs%sVG%jEMu66RVb1V?WCsu`6F|cc!>dCSDS+8IH^Ka1V)EQ5w#AFQnCls@>=K%?U
za8+8tqMT#y1jFFOYAGpzSV)u?;_=Rt5l~o4XTfTu@y>^no@Bh{)X9pw6VQeVl?u-A
za?boA<Kv~^Q9I}Ly-+LbV%>R0YIGE%<uVmqy~a3;a3u)^63ciz=E0p?#I7WXB7~Hb
zWlmY<BvGC2<_3=*?9<9y$SA>;j>r24#H}9b<xA{by+}45aqGL^<<j;gN>lQ1e;-{a
zYPEzUsk3!qkE5eQZr%NmlhFjByL|pje~x;4i_kp^G-&0!54aMPbzmLN7?v(d^(>*G
zx2PJKFDqoF(81UTj?5|vd_@d=bv^mlvly#tPbq={w1#pv!4`95+#nKSu}7S<)EiBn
zx%n(oD$Y)QaYCNW`PdjAfQEUo4kn`!@4x$9(prP=#vVdg>a`l18{0wA|7f8@w~8!O
z^Sl)DUfI_yQQ&0(g<o5%!PWNGCV%6v|6BYQ|K)%6ujz*J=`N4M8DckMOJ+0!KiWxe
z51b9$U*QdKBpfqid^$Kd6mC-B93=G`W;$bXbj0N36d@eXbh@amO{8-~K_K6{^>$WL
zXsEjJglvYeE);!vvc6au!(~8y&ISdiqe7ZI+O4T_{+?!%5k(3es4U8sAOuzju!iZ`
zA*r%7>NT3}HabcC80VIvC@G49d7eXQ8H@*XH#gbX-r~}QJw#b@aQ}cu$Hz1`I=FJm
z;e$hlg8@kz)7$Ez+YM&(3>A$yIXhx98~Y+N>G^2VNt7#jaOXp|FTLooF;M|7ZXwjX
z$kGJeTJIohgivb^v313Ae@L-1@2j9rSbder4~D_oRa@i#RYg3Gkn*)9LTGT1=Q(F5
z#~dFYlI3&qJZC<iGtayLdS+lUH@HNQ${-gOq@<K{49`v&4o|4JYSh~ew2rY@@**cm
zy}Vgb6rs@qi~EOl`}}m86tHk|q1SEmOJDzm|Nl64(khftM9vY5IW0Nn5_6g)l)|^Y
zaKbx8VcB~XZ{gw9!Z~sWsZvO5$@??z{rD%;HhT1$DXHvwPJjipFS{y>a*<l9pd;Vn
zDtt{ybp*G;2G2q)eDm<mKn(XKj}=eC8*mlM%B5CFEfLD$gjq&r3rE8GgTWcJib?Br
z5L2WI?Q^iy>J4NTk@RBz@^Ah%9^AXfwHwdU+uo%Rn$hV98P|C9;68C9rC$_?I3jJ-
zDUKiW;YYW)dgB^zeev_udp)8gLQClP&oDNW=Nn6-*&(U5DMjqjQ)dv^6{O*O>UjkX
zP~g2mRuHl(%B;qCWpg=?kNR;8@VU>!tQn<+)QIxwK3S!Rp4R1H#d)&UJMh++g+Oq1
zStrO#$Sp+S9bP8K;#`4sCB`TpS%(z0!|Q-a>ARXZSTtJq*U~^$k}Y<{uvUk|;@~)E
zlz-amt<S%}a5N^npR;lK5?3!@VD|Alw4@^zKB~;Dp|pm=IYt&{rX&`c(pj7oM1eyh
zZ7BZ8OD3nMMCBBlq~V~|Xsd9>f?7qdy}Zz(hjN$)5u8eCOB^N6m%US^ko+NS0<tAN
z2Njlv3GaYwDM40kwW?q5=gL|`A`CX4Ga3&$K72?y?bB-4(UGRAKF;$DEn{lUHjU<x
zpZw@uHn(@dNVKYR`SsTk&QYtUxLSj=8_#g>2R~%Dw@W>(F`bT?PG%IwQSWxSzkiQe
zku#c=y#1xO*uMA-r700k5G#Q#OP|7a0_g-Mke<WDuSOGLtRkfIl0{Y1@oKF6D_4q4
z#4^>i29j1+WrW-^7^(J=uaNZ;4f$+D8YKwnh21NZOgKlY)#1{WD>Pay=JPpc$45vZ
zn9gRQ52fcCxvb!OfAWVse)xbFU;P}{Z@x;qw?(a1r`_%#rC>In6Ge*6%^t00-M3g<
zEPWqv2*H{Y=Z$sHFAPqQum1E8SRqt@?w5a=$#}wl^Y{P$&$Ndk5u-)Ph8(kLW+X&d
z5n)l@up3JxTxcj3^aEY!c`XEU>!^?BJpA4dp~xxbbH*p9%=<&`|KKMy_b##brOzYl
z=^`+4L7hQ{p6Eg-T!mgM<wKw9u&^dLGS#`UG~NrNE`tq{74qlDQcAA{TR9<8IIxz|
z6eLQ8G@!Rb-Z3c2XG4<MbH~sMt-V-MDRNAiGaFAhJvt^qGoEG~9z3F+)=6Z{qsNCl
ze0V_8PI<KdF^2~axqJIA$A^8au_&#%dgBUT`Nc0W)i9k;DUHSJidN@?CUWrb_IuoT
z@e8yzFD;&V0f>ld4+um^>90C<)$2a#oDEaA_Mhc{SEvD%BMqV28E~F6uY~e|LzyGv
zDA17{lj)e#vtx?Fi`*M)NYfOpHS^qYI<}P7K`MxSl9^CJiU9IFBTnilodgOB6uBWy
zG*J}O@As+IYKv!C0#PexM6l1RY2@YSRWRk#gA1S0u`@*4u)EjhLVN{t_k{Y;pq;}Q
zZ_Jf*C|45sJE9D92Tua^1ZaaoTY<nZJUHaRkAFn6*(Hr@h$!_bAr-1*O&OdO?VS-J
zLFnaVyC6|f!QWKXnvSZ+=>lx4oR+Zhywr*G@VE1opg4y!4xxf<N;SaNVM|{f=p5uu
zlGNMGhNoB;5y=Q+y-1@LF@=QAg<Xsioc51t?d{RqyF}}vq}AD^bNLdjjTT1_A7JMb
z4(@%-$=M_R%5VM#z3nY(ofe{AquJ>)J~^Vu3SWIJ97SQcc;ON{PAHsrO`Rza##iz!
z$hRv{__9jKc`v6*X|AAgjxrFWS-^0vT3FRqTG?flabK+xtRgCB{oy0saucpz%RjLa
zu?n%cr|Pl}3L`0uLrNQdpXVf5m*FtrDtVmJSBCn;A6|dnJBQ27vzjUa7FE40jFb~q
zpTgDG{_M_OUKBihaKyp>DH~f2zW8fj<>UX?N0h@k$P|NPVhn|K6xL##BbFMAqi~i)
zXrjtP5FnhwWhE!KKB9hkhj?cbl9*B$giJhYQ2Oo<LMkeqm@4(RfQzI*poMU(b`n?^
z8+>m=u!|m+axpf3BE6=QFR&`DA`+fo<%iFw$b^%OMq>v3J|KwV7;7v@g%`#50=$LE
zNpy{!y~`XvIN;I!eV%{)EjF${N0g>0YjM*#;JAMMCa1?o3?A$=AI~^BIpyO!AM<O!
z`Ik^hN@rt>&Ui?cmwa^V2W;<Nq0`;=XOl}1((7po@at*C(ejIB>8MrX<<n&jSn{S;
z(?9D*Xc?I=UK`fZ(^U5tnF(vL!aEg3J|}f{0iIUKkF`Fn8$}V$Ip)&|gF&CcXu$FD
zDOJm;V2ZKxIrlzzkB4_}vwQhFy!6`J+<f76x*HdXqL?U(>2zvrZFPvX3{ndM<Cb&1
zG`{PB;}V5(A^2LJJplDu%3u7AU;mjMJ1tVScec2q=5T+Av<4|Or4u+~z&Vr@3uGt?
z0lzc8^UutdOl^$XKjEXZZ)2pSoMk8*F+ZE|!5{n)I*r(T?G;p7^PL%N02hP=89`N8
zC&Z%qRR&bZIvdU%xi~{t4?qtM2$WDL6(fYkN{<9pTChTZ6IErufKriLAGk-c#^9*}
z!WNi(MwBSuRiP{>mFE(Iz?B7$?%d_pJMYqK_qh1XO`NejdU%9GGP9Q6r5zsJ{(v;m
zJpa-Q+<5T?2B&9y^udRG{NZi-g8{8(8=WfBdXv#K_koYI3sf2urp$Q!;1;dTiwkFK
z$r}r2iIBd!e?{Q=NvCe<+*J0Y+KY>Awd(M&l#y9Qjy;zcgo=oCx_GZGb5zm<AxV>z
zBu*A{Bc&wLnw_mJ_Ac&l=gvb$a|Ku;DG=IKFabhHoN(wU@<HC>zJ0+%8Cs&wX0xz2
z+~PG=pjc5=7SF#3B%h?0$xwv&srSjJekCAe#+92rzW9|J?9_7ZenW8hC*QyoAi^Pl
zQ$g=&0e1wa4!~<&*$~NSCvchL=+*<8TOZPF^^n_}<dGth5n8A~==B1;9_CdGl)+1O
zR1OPZO*Le|k`q!j=3Xov;hh7oPv_OWwbWoGB9~-&$zVQYJUU}M8Z#P?na!rmW;3!P
zr?iGxC>rsc>z7i>mLSqTw{A>{mMWwZ9qm>VF)O+A?kzUXGCJKY?Aa6w!)P=iKN~O}
z_c?xe#P;Sc;8Bq%im0VE@+@OG95S9ve1)xr+z9qAUGvp$(okC8t>AnQzSV<uH5yVY
zMymRmUr3$|zLwhlDX%Bqy$TxoNmxcAA!>axV)G7aUKol}ptaAIdvwR61A?IaQ~Gt{
zR(N{V#$Ww>ek|5*aTTdty^Eg{V)wiM^1Iyr=#beo<IBJF0?&T!b@bp@dGDY6OT;`U
zk{0P84+cNc5e~>~iI$o)v_*0rlk~2cfrB5thfFk^FWjJUVVh`c1DgJ|r70F<d}m9K
z<XYdR4jEckRY;psh_VRC>(Z$UqvLy}%B8Uq9_sSWp7IfRHptn3e9EvtKu3;xvqr1k
zLPdh%NuT%L`5yZZ?sM_t1zvycd5U7hcs4=n80GW=xh-9Rv4$J3yvXUJBWk@3Vi}Vk
zpHa-GC;^R5o3x&C^Mx0g>}>J)?rkRX0bl*aFVXFFF=ffcE7wsfqW17UkIs(xli&ZJ
z_}mwMmAy;Xyu1Y38(Sf<rCG1e2Rx5-DaqiG2f1+g*3)gP`|cCgk67-v<$hfT5U_Lv
zMFosY>5&|}biAE4C?zP$P#t@G#KZmjJlMa-@#!%SA3pMXC$th1(jkKMLph)G;G++C
z{O~?!Cx?9Xm;Wk_Ru3Tsy>5pjiou3;=BjgZ^*L7_XJsF+(pJ_{??A8l42Z)gUaI8g
zn=kO%)fc(@`!Ih0eWU}e1h(|&g9_EU&fz(mLQy#5vqwlq){vNlcw~^mkO+mez8lBv
z-ZAg|<KO1;!^d2H>n)NC7m!j=WI0O5KHa-O+p4`FsN_tjdUA`3cwz$};ZyGw^5T(j
z&2*e`cyxqPlC6y`nyqH_E<fHS9332U_wGkz*^J9qE^+nx6_Q%)`$}UmE(0QpB1fx;
zSow3K%nQyAj~Mj_G}1cTJDY6Zyh+-<05ikz=n;26dY3d#X?B~mx3|%0%I4mT3l}f&
z`Wvs{9OyKmyS0h2j^SuTmV4VKWhH~<zzCYHZgBdZoVG=(*`1e4T7Pk+vLM9z-m1R&
z#JT)b)IfmsAb@lsgsp6h^<pa)s5cty?CveZxe*Y}V&>VLmtKB>gM-J+=Z0AhdErQ<
zN8UvMJe+`LtK*%#ATnL~9(h(UnNHZ;Y=r>VojXVTeu%UVAV(!s?0<S5EcvV=`Jc(L
z+ud#O^<RCJ7v8vtnVjKfZ@}pZ`|o`S(xI)_<{}_;KB^~Z9tng6>kI}9LLr?bDlO$8
z=iv{3LVL4EH<h@J4i1SzVQ?7d7eyFV&!_N^+d^B$SrG(8VJU@DZ6;4-L{$XXKpGYb
zp%taIOs5(B;{m6Kr<@)4nM{XF=Of?Yp(uP$(kvr41(_*OR#MkFic9obb%am|?TDj1
zq<WWdYp>m-LxhfN^qOrZCnwBCBc`W)3Y#;W^f6gMR%X;1DN&M;CJ9<d27>{k(U{46
zh7mB%4UOIfHg~W1P7dXKF$Vv-Yn2lvSB%K{LW9c4tmXZyVSS2YRB?V*FOL{jpx~26
zZTa?K3|2TL>rSQLKt<_45-EIiyJ9RHm7xr&!j*xj5Wwmg&R_mMs(ZJ~Ig=o)pdvf3
zj?d3>{nqdNHr@1j6q4~^PHm&j^<VxvS&^~-+rNh@9f@@e%~A+8Hw8#RDrBgR^ZIZ`
zdSqHUi5r#N|C8@>cK04zH?Okyxz}hs_iP|gFXt=({NDQE_oWCaCJ{O*2o1_SO}CJM
zP>4meDpf%Jl*Ax8ess(S@4V0Lk3Z(YgL@3eL!vmQ*X~enHki)ljQb<9`HVaF`y8Fl
zdGqz>NYsQRR!HelN#R5wj~j$FY_+#gGs|>#z~uOZ`D_l6X6MQ!HZESE9;K8jVm_I$
zwY5!qt4kColvzeCt#RSPMUvPTWIQ@O<WK(apYVk*|9P%ndk&#}Kx3Tq&e}Pn<SN|W
zRR&pdo-JDK2kWW=Gu8r$#rV4=_ow=Q#a*kMpw+&`{`dMSd556L=Tw#CQuz!>o@WdP
zeGVQz;Nv?VaymHW<J-46K0d|am}HKz7Bmu2vbt{{z!dW-5AJ@%d^V=u>;k@zvk24_
zN9Z7N4n1&{<j^?;VwH!xYLu(<VK$%t8~4b+>H7cC_NTFy9ap|5_KS$U_nFU~$GkZw
zlk+H25+#wERZ>lF=;zlk3a@QH!^Uf%KNv9l-S8Ltg9jLG7>2vy?$_N9ZM=r7O0P;P
zO_W3p6e&uaGs$F*H}lT(na>^~!XF~`KIdMhN@S@zfFzTb=bp3A-mzk>^<V$>U(Lyo
z$DezEOAk&^n<#PfKL~cOT`#axC4?KrCS39pNQk5$@#XbE2!>h{OGAhv@JS%P*`Q<$
z*=C1Z-+P6<_wVrFKm8iD2QLv1hJ@uR$Xp8`2Ng~PCJ2xL0_772_d1s-1)h(IlZ&v1
zcB{j^J9phP)rxYd>Za&qmVU3#-P?D$dGi{(dpjga!UykPV{&?u$(cz)z$RN<JTrlq
z39t%EQN;}quohJ=vGTwrs*@9xYc<-#4ufkqSXvQ?IHrHF>qfE}feJ&))he~-B*Rjl
zY&an7x@jRvk`PquRBBBsQAB1^v^J!vcH`)bV)@)-%q(9(>vUA*0>+O`DWprxPxzO7
z<q5d4g7Khc{Qvt}l)KmYvy;d7S+Ntn+JH3%sX}DnWQFo=TZ&37udIN8gWX*Q!vXD1
zi&p!H$%zTd<r2fV=dQzyl71(n0+nim$*CEf{;KiM*?nH8+s4|Ih51=z5zrk!SEaCC
zT5W{DI|1h7$zsv8^LxGgXnmr<?(wHC@c82o5Kb6cX_u+frxBn26njTU^zLpErN$u&
zMTyhNlu87F=VO>$*<Cux1A!zkmUw5MwRheloUG8Sgh-*#N+Lw)j}AW&S$ap5k0udg
zhm{4HL*}^XKM&%Ba@?{!Qv!w7FzEK!+1X)pW0Tfli{W5E8mAn#4%j=`Wq*H<gTq7G
zoi<6Fkm%ScRoj4CsF|DjBBz%psZ=7;)E%@c3Vb$U5lTQ{DMcYtTJ|?qiHCj1(@~l*
z5JXu(SdPdtL#0tehOU8Q(CyJVJfhuclUc(cO&FwxCm(+n5!A?ZuIVO@H7ne29WxTY
zv5%LX$RnoW^~K*kIyw%U=f7jIAL)v--k8SUFaKEoj-#IyB^(DoYom+jhj0d(Sx7TW
z76<}H4#ZkFa>gh(I>t?{D8iQ_u*=J6BQ&7^^*^E`B!tWO!RaWDI1qpMy0P_hCaY^)
zp8q;h2!b#|RqLF2@p*>(+jL%ihbZT#c^*}UA+*vL9NfW>S=V(I2uXxM+5j<x?&==B
ztpf%JM?6@o5iYH`WZLAMSBK0wlujXo!11=Ea`uYx;@tSCcT`@q7L+fe$h<S$b5`Rd
z<M#ENy!ZCI{PNXb@XpnD==3@T=Q0W+SMrd8G}h=$Qz=(?@7gkNz5O1KKRC<RzxV{D
zQb>u&W&H-iFu;@oi~wr~^pgRp8W4mQlNxsJ+$A0ki7F9$>zlL>TQsKYOwP{IoSZ@k
zNtQU_-bSOz)YJ^!UYmCRi0z$qmd`whljh8D>ephV7(cc~t|b3~J?5N_1#Q+Pu|)&K
zSTdAL^x$`Bo<xm{w>q%Fc)K!$L4=Ab7;TUR66NMcYV8uvPP@(S-ZqD=19o?JxpwUa
zNs@vPBnDaoi#1Ry3j%e**{Zyu#;w6%=%rrp39|e7Ne7#wy2q;o`Ry6mcmCZD50Ci2
z{onudPhc(gUoU+5Q7&J&Kz%wOzC2|3%v0>$TEz@fgmq=uKndI=6zj^_{s~4xdY5d-
z&}6<5$xUi;HZLtGD^Q80dvk-;mw(A4D+}1D?z92}*Wl;6LXW3E796M{LxeB6THj3K
z&XL@AIe^1hLA%`{jZ-q6v9-AkR!}aN==TR~ZEdo?zRH#N-{XyU-efo&GCw=V>9c1j
zm#YlgUG}$c@~v+?&1au|nE7y;TBS}VN@$az!hof7XP7^I3IJs-K_XFl$k6u5$|*rT
zpc+)MQlQh=brgpoVH6QoD(K9(Ip>+C-RY89h>f8$(3Hy)OwBEF_Tt0LE}z8;1-|L1
z2ucd{Ij7o;l7y3>g1db#5H?ol;oGsj^btRVqfQ;$hh8zk6{S2ETf5GR5y<D_hh}pU
zDJ7L^jf35tVl-hK$0SLL5SCCv=y4^D0i}paW#$%^5J0EB@8XQ8f(jIA(x=@y;(<$N
zsn_ZlJyr=C^Fi+WJ3zrzS@YL{7P4sgD=J%`KqLPX1$Iw7`2f>XQ}k>HDN2Nu8jTBQ
zS$XnNZXULv)pZWY<POsFG^}s^M|kq5qi7>VL9CZq%E9eB)Mx5c>NV7<025aI=p9^1
zL^z3)T!H}M<N~eC^P*s6T}58yW-tn6LHVpf1Z1gYdwY+y)irwU4!wSt!`2?FcW-m!
z_6;_+HfVLbW3%cgHWES*+K4RF-au1&gEq(trLcllPV*=T-1w%{W@s}+YK;*F!pv#r
zgl-chGQmccYOP984H3$Xy&3jm4)+h}^?D>(N^dY=@ze!Qow-Ox05(ICVXbn==-2`M
z_pUL=G;SdG#nAKD7k?c&w7i4$xN{mI621WXdoL$PH*$ZV3fS3rl+Ph{-rpaa6<g@p
z`EEo6K0D426t6UHeph+fxI-j|mwgtf#BnJ6A8Tc%PZ(NGpIva%J%l8z)|h$VJe}XX
z3vTAU8?#qD64wRkN``{e8d9AAf~pE=$bi5~Fap`P?7nq_QvK&#`e%QJRAsWv5SCph
zpOng#O?{G--ybCaD=~O<-MUdWqrle-g9?`z38m1+a<t##<_9;pdgTg#`~AP+&iY;K
ziNkxC#thTgJsWSIX+NfOtHYh!H|ZX&@Y!e2G0_ak(umY50#QOzLMcN~5rnEjq$9ea
zrQhnH0!5GrcJJMzpA6`AkC1_)(x?(d5m}<KWcfufjFW;OEtKT+*$Zgjgd=6<N7fw|
zB_9PSw%Ez?7#blyLVxfdhw(G)MukjgL#7CrM?u~A@8<i~_)sH=D###0`@qgMqGW~U
zTs}ZOO;ge=<>2s;PP<!NYY>Hm^%~t?i;{7ak)k{*1hsn8CBDLqj9TB^;?${SYUSE!
z5{-0eMGkJe>!2tL7n*wG$>52=_t&q#&RcK4^9NXtTVMUtPcze)0ih{Xo7CoJ5n)Ky
ziHU@u6iJYfWQN2j23C;Tgh(nvA?aJUC$uqSf%8ia#BoWW^#FvhfX!&HZ;<XCpjMis
zsU|9m5mJP|KVvJW`B(&21_&?CY<xr^gyTvI71G^J*xA}4i4%s&fHc+|?CoPQBuPTM
zbI4ooy~S&9yhd-(!(!Pv+~(H3yEH0I`n?WW(nAF)&ph=Yy<Ue#b%Ih*bpe%<M5Pet
z=3P)_15GLhMAZlxO;D;-sV75**^pkpN1Cnsx!#(n?lcd})hhd2JM8Ri(eC$1M934*
zy~yIJhX~6x+|9@w&wZT6!0{xasIuHApgX?uxeI`gB?MyB=w@?iD=$Hu`+_{FAK!a8
zd_h?^^0b)fA##ptNVQrc%QS<253me}1CEZ4NRyPQW&@k{5z4DI1d4jSNxfc2Ny*mw
zU6Leb`Sc~`m(L?ql`Kn<Qn0wNfD@;tD7ON>dCgHH@`_hJ!^weBuJbnc>-jblAJ@P0
z2?M*+=NAb}5o!=3ti&pTXf{}S>>+wP+Z_Dn9k7PXx{Fkh50ENK2pWqXRmj|dCqSUE
zgOtsy*Qqp{%$2J!U&5#W?K?dK89Ky4j&7FBvm#&n*4p~0+?QdcL@0@sK4LckSq5wO
zHdwuTmpG1Tx7vJg?Fz5`_EmQFx5#wrH_-8onP0Q?B}Q!=+&??BK<DTHB_)CEp>XmC
zrKm(0mJuoiEs+GINsJcO&9juQ9bGF!eWHd4oHT__Gmegq=(IcZ20b#J((U(AQG;in
z`vO7+KA<rm9U|%^Ek@<VqQU^KgeFGCgZnN^5jf+Y8HIcB5r)kj)Ob+H*uO`?-KgQh
zj^5wDr#*Hkx(7Kr_(s}iqWDZm4Wt&zKK9Zcwl>yctOjG<bIVbYV<UX5E3eud^`bz)
zgmk={<62=3*|pXd^JC>dQgC<v^n*O~$P$k}e!<-{W3X1c^0<<u)*uj0+CnN~t;wvL
zKv6;<tR>MIah4&3B+fKxXqb^9%AaB?Q*6Ke0gF#RN%itWq}o!(O*jxL@a0!o><4Kb
zm&!$y3L$(+{8-}vAoAJDPG$Yx>OJn<xXtS79sc^ef6cwEwc>e?Fj0#4kYiS#XXBaG
zw0i@j67+gKln782Rg4l;BJ^3er4&^O!w4xP?SmtXa1OE4Zj<VSpcWzmNxRizINTvk
zQm}4sC22x$(55r&bK&xn%*>o3*F15_S~zL|@p2-N@8wZ~AQ(BXkCc-}2Iz!Sg*?{O
zHA3$UCkKBYL-+6~t5Afc8tx>`oD~{_)?SoYDx#=Nqd7sn(If~0qA(&Tf(g*tFfld7
zK|CNc4#h+GWK{-Ksx{w@Zm3pj?Cf^9aq||BKKwADa+4RN@2u6@6nc@uSTE?zSn&VR
zm4BxvpLm=HE?wgKjT;}=RPMXVr3e`cveaNRCwiY{nh+>O5+Nbgn%D&N4fJ)2L{e5E
z##lEC*g|SF1D&CjB@~L#=7m*vrXvMX7<8O4*xRSFd>WgkRIPPQc7;uNtwLelnc*k2
zSmCNJDp39&c|mw7KnaE&!~0i0;KuC_xpQ}wND3PDCYjc>yDfJ2cUfCsV{3PdRC@^!
z1i}WygP3%%M}ES8cx!{Loi@$t9BCH&b4_5hF8ppzNg@Ja1(m3Q)@}q=7?cUqm_!fh
z54!ZSu6t+e%J5s;+Z-Jouye3Yd#L&J7ygV>XD*VO(9L%jDT<(IBFOi5zL%x<J!9VB
zqqf6I<h$5Gg-)s{_HHgvKkj=*zhBG)KXGPDsZc@Wk!Wv2xRHKA5F&(Rm?oq;BaR*D
zGBZCzYbPVi29yFf-6oEQSS-DMkG=g}cK3GBRx&ZYKv1e+j3Er1|93~Kz-q8^oW|h4
zr^)$>_-=PwbaJ~pHiGGrPXO6{qQI`wC}D8(GnEv?(t-*pEzYt0)I$u{)-YT7uxDEY
zUZcnA6!fzK_I8J!0jvm629mutn^&(<o|~iAs6nOXYR(3f2plpaoR+k-{xdw9XRJNe
zOM!5FJR|W9wF;YsY~0=9){T1%lP;V48@%+hpK{~Ib&@n@TuB~aeE;RahVor>LPDA)
zj@M9~q|-h`3Wc<ST6F@WCFQV+uzf@jP^r{O(}ZE%cL7|e(MD5AV=|Lru=IL8OlIg0
z+jM&!wAHlxEjryUryh8YMsv<NVVmLEie9vH+;m(M1VXq*7VnU}!c+m3pIlBxWF)GY
z6McoWzI@=F9iy^RZdCcbD$ujhT`Lg6JaBV~-PnO66*(?i1a>ahK}tcX6q02LiE#m)
z6rP9SgF;tIv3}%TekkNpHSUU4K9^Rz@}j#{BcozWAl4WcO-g_NM!DY~w!$#vi(mXS
z&pq}e3-hz|bPHn*I*o}BkI>p>al+rrPz98Lo=(9SLZML}Sr9@HNyiQAX9<lkpyF@0
zwC+2mg8`Yf9wo<95|kg)Ws%nJ2#@~@88{6`qn*4;=6=VauM!o6C>1aqq};rIo9*>&
zHn+F<-b>%--q!j@?ol2P7g7YpM#=x-$~L=)ZIVQgg>GLVz@SO9gea__6r@>7f6%8|
zsZ*)ciK{Vt{T=#tfGJs&45>7ntU&ANh=Z1s2THS)G)u_Rj9A0+g~xd4k<Wn)kV?8F
z?HHlka|p*#1(93N@pUq|%dO8P1o~LnB~%U>vW2KN$3J82-}rM)2t`z>`?4okqsg=z
zoi{#>XLfFZ`GqBJ-@Qq>R2tj&?y8jXG#GCu)><+PwQ`kmweGqpvy31Nsn;8<Z)`I^
zH_!6oBGQXf<7IX-j}+$mTA8wY7qtKZ002ouK~#33(73}MFFEDkad~Np|K0!gf5ZR!
zKmE@>w!PeM<)zHQuuBs6Fv=1LOQbBJa)@STL0ds$TrgZ#AreVq)1rb98-uY4S~x-N
zJOIZhf(U^@BT=+>_lQ~tnA!wsr9r7wb81mi6kKv4-F#n-ab5U<Pfmp6U8#I~NkJCF
z2k&0v)!+Przxm<c&}z4l!V(1zu!z&t3H6UHB|ut*@L~2Sfil#pRch5L80ms<?K^7o
zq*h1*<(dNn<;t|yO2V+jFz%yG>gK?w7Ly4Eg9In3BIpfz4B`Q;?h&nSm-@_k<`z!5
z^T5xgHPY#97CTP3=8&A?;?8vnKQNBu<}3fX=A#MIDCx=1_K%kSi>gt96^LV>VFjLk
z;=16+MaJ{VMQN6i`q7k?as?p;aS}5*JIm3=J)%Gm1rYZK#KWPJQBjH@s-l9xQCqMi
zNkSBbZeRIIP!I?(ZoFR}51rJE%&UBP1Buas(a27Y2P3~{eWJiFhXZrMyQ`L2L=aFv
zwZzgBkFeJM4WctZSr`0emVC;QPiT)nT=%vXVGIGT40p7?&HDS-I6E~(X;G4jge-}n
zQp+zR|E-ZL@rZ+z&QTTJkOkpcA_+M<YH{=Cha~BcyX&ib`@7#}dvAAyO8pK8o?p8F
z3<f2fv__l^FeYJUc8dN+kL{fuvZ)yYRYnq_jlmj4mRX_@0#!!I6eS|8%m76YfUqI`
zVUI!I2_$y99i)<U`#p|Y9gqRjGmA(SI-zJ|38Zq@{gF!^H=OZ2iLsv_e-?J^zfZm<
zM?FTrkIzDX7uTqGNIu{?j^d1OsFPWh5TYo>TA)&`u&^-C-qwZ}u!S%jLD)IEk~^#s
zUoH>C`H1nH<O+_0*xaF8%-FelzaaP@veGo=sJqSSi*uB#75eQq28~S<XGdW~FifZ@
zMQk*IkW^GaYBibl6IpQlEvzN=7g<(;Gh_kE7>v}EPc0M7%wSB4i4Ey6p;D@o2ior0
zcr@FWCWH!HASV^pBk!&>;6%7(py;(yKDct7gVr7!J6rttXD@N<-s<o8#)3w@P80?h
zqtPZM^X1TSq-QXU>GXQcHRq50wl*%f(?)r2tw!q%DFVVEqB$`^P!5Tc7!X9IkR(ny
zYPae2`gHnTy8RwXNn}vw;-jDA@uxoPx*NS&ScB8C19B1Bk<$<b3g?^Zj_ry2qIQh-
z%Si7{V8m#KrRSC$+xxkbq4=jj1tnBi!Hqc^l|Gf{83V{bQms~*nw}*LBL;&Z!(r@V
zF(sIpoy24rR;L6d_vnRnO*1p|r<h+nLs+U{{Me_c6hc?9xwXT>{5*m0=o3C?R#IW4
z@v~_`Nb96NJi1df)G&4rt;Gt-^wi|<GL4VFj3r5i^ty*+>5!maraD!_^m?w;CLoX&
zYc!#BGk=x8wL(Zj86d1Av4(+l+*4(J!Ygp!Q&^0R(PfLRRXE(*p?3Q&m9wWAh9&Ba
zrsHf2FLNRaOxBN3lx{>`K3dODJwgbK7F>Jx8ozzx*L?eXf5lO!g}_y^Qhk5aQ%d3M
zCK};KWC2Jt&3eGU{O4ce;+a$Q+6R75YhThg{ybF#q5{gDn_&>5q~h>skJa^?T>an*
z{a%k+wMw%wL8;^tq$G*y4|>Q@(dqQC7Bb(gGrv#<>t+F);(3o`q4MBP8|{IJ(f%Gk
zgMs4Q9w}yw+r(%URe&NX5+AX{`+*jDyfD%UbZ2~i4O(ZUY2rqK;ii4m8#PK%#Lmth
zk39SkomQKp-Vu`(K@<qrlpzICrOv|g3d<`1T0y10=P8wH7@aX33|Lw?Ll8KC!Wx(O
zj5kG%2X*6I?qd9+acA$z;P>OMPo`rha?P(1ncfX1kce7?nMWUHxZh#>jVlN@YAt6Q
zcw~(eQTr%MRycS+K#~j+c5mEfdSQX6Qbw06q{$F#HITQaduNmj2$Q4Zu7g&25mk!|
zMonh1MsoGNtE5@X#?~6&{?1>sy}#p2u47q{_kF_u-9^#?VHmnHl}?F*5|v1!0!ylg
z1VNM2r_ZvnagXht9k7zw>40*nLKK9AK}Zz2Lt4s^QmH~9HN$v_G=?}C(rNeU^#&Xs
z?PIJfbsV;ius~QY(VU#c61eiA7h%p3j@*bx8z67L_J%NKB;&2rqgjxn;OJOcXe=|z
z@9%ivc)wt99J$EnS)N?ZI7j@VaW0<nVw~233=|WS6SNNY$$AOe2$XRm&DskZ2Ns7K
z<k#Z2mMBWG`Spuq-#uTs16F^ZPS8J6DYif8Gd(wnRTixF_rOq_pJ(pzhdH>t28SI6
zNy0Eo2}D3e1#~m*P*Ne02qNz^D$;T6LXXr1!ZM6t`pgPJ7?5>3XbYVzBao76qlpNV
zqcAAp2+u%1F~E_+B~pb1k^9b41qcM4qXAdmy2AR}J+9uk!dvgX!}`X$mz^0)vT%<;
zNr@CN=y!{3Iv*PEJ~KdyfSJiMQxjFB66h=!BoDAik}Sp8*wGqPKKa3&EJ}rx%4LF3
z`2-63{ejbR>vrh&`-D*mp+c6=J<i34KZ^*e1R*ZC%9SXf;Aj@MgWIDHjTx2ni|==z
zz;4`r^4GZUDT*et{9Zd78+BEV^Yx5R*21WQ5DH^bhkoW7g2L58+-JDGn`J3!nzFgM
zMVh6QY9&rBFHtR5*tm0>lCmfvG1>wW5rv#OeU7=sGX$j?;{m@Al5V$4nr1{{rO?Ae
zh{9%b#3Km&<OKmn=l8`N`|kPJwttLy%hnir-6IBr4#sL!t<3T>kFsMlq#0qYN)$x&
zcUxqgKB;jCj1rD=ffNX997=_tf)m_FSV9Amj4Uw7*#;A*mr+rK5(3@nVB*+)N#S4u
z2(Zc%46Gf210v_I=3H>+i?Y=4;gy@b`}SLW`@4ThyL&|LyMEUSl%m<FVNFbyWh6<4
z(E_aNF#YP6p5dQ=>nmU~x`QrLiJL24E|m#Xm~U>Z&9GLWR6w92hQlE{d)vJ9^PjM}
zy+y5BrCct<#yzT0nR=~G6h>G9p^1<(Ac+SII-C6VwIA{P3*VwSIfp@d`7FpOKzWjq
zpW((Bj8p{T_{{D6EY5=*ZR|(x#t6-31QwhK2*=U7kNiF!z02_rJ8HJ{?}jmkIF6mT
zdA~=RB=mYcgb>7ogrjzgrIpj%`k+fU&S*+E-6sem=H?f<aOq*D=a!LyqTg*HBrGj1
zFy6eETdd9P+*r^C;c*DdNP)o^i_zj3aN_RM{R98|?wgPO%Hr<2gAR|V7-^AOU{I*J
zIp&{yjLyy>-Md?eEE^q&HV^WSN1^%61y{CFNJ>^QY<1cF;2Pm{lgjDSpk0|w*b%SJ
zj1S%RQJ9np>F2b$=V&on69pAMc=tnwgCT=Kk01T``|Rv(`!1=`W|7i$ip|f>Q7c#I
zcDp2LQfRFilR^NF)|Gc+x+uRKRajm*O%MibZEmAfL=c2b)ti(`Wx$a7LBkfu2@8UN
z%%t@DJ=*Of_V#z_4f`ZX!XS<b!iXp;5k^%eCg)LM$>n$6k-3af6lK!<kQJgprAPPZ
zBad?=keEM(j|O$t9|&sfXO4I}LdZt1&E+79z<mtuv&TN$T~{7#DbGi7`)j1gu8pQr
zt5R<^+3$8R)+w_a??8%YY{w3j;uOf8CnJ=~j%R8k?Z5Ht&>7bFp7>-MLceQ$;d3wW
zZ~j;RDV1^slNqwiO$;FjsGeD&ys*gZvt2ryI~;y^o5P!T>2&&(gd`DuDuvI8l@Qcq
zNJ9p0J|mV;078SZ4C6jW`+J0yiVGx$1CFv3>?x`fQwS9n*?;~Fq!$#{zJpAvz>8u^
zvQ+cIyEnLb<2v8|?q9O8vq_p|o*OCdyT!s#p|g}U)nvw%S@I{7AGndyJoCaMoLOEZ
zjuSfF4$a!Mn<X0tuJo9>)6XFdCA4<@LMcO}(OBuqg57S9Znwu_%ax7Fl^RloXrtKN
zX>sQ}KV+iW;K?T*=G@t3#~jP01*G4nW~70ZpZz)4^JJ2c?<b(35}aIxV><COVLe^K
zy~{CYW%0pv5FY!Se9}Yy9?t$W{eF+Ftqu0~_gGuopx&(W_~VZe1p(KtUd1FsNBTFG
z428CC`p?AFETu}r1tGX|GUrAO<Jb*C7uCX64I`V+QHC%K-J>^-8+6i^@-;pZ@()>q
zVVflBkr+2a*UA#JpMI9f2QFf*C8$;rBA~mv#_A7$Nq>C<OXiSn0ooceGPKpyR7l;C
zZwaj>5SqS75ew6-Jo6;=sVU50fHa2k+zg^rqS9!(?;;dNxl*eXPJ_$^4W@WT<37k*
z$?DC!+_`a!@BQ#Q>>ca?j_RPKK%2}bK9bp)8LT#RyFJR~5>g6uCW)eeFtlXapi@gZ
z2>8<HpJt+7W$$ngV=X&-djzUXAj?$BHH`5S6*L4Yq!d*M0;j+r1oZlSE<W%8GYgAY
zZRmB5Xtj>mJ31in07(#rluBhvQJFLw(m7b;?wuPv@yvp!B?U!=<=7IqZcicnjPudH
z&G&G=m$W_Jq&QwxDH`iM+T#8@=kksCozAYH$m|)}E}qwJbczhC#U)x?;MwoHneB&%
z2OJ(9u(7`BkZ&mnR6uiblBiN99rQ3pA(Ww1sj{?unuVp)M5QWLXY~7hCYr9fB1b86
z?<ewW6vB<5)286qY2&7w<i5xC3VgI0_HoxI6DZCvDi6*CXbCC-kqS&uqP()i+!GIy
z?6;8Z1c7y&pZqqCXS(Cg)CXrmDwNR#X-aE-9eI0(rKw4>%mqYI5Fmv>3kkWp_c-i7
z_9qW6;v^->U}J5YEXjEL?Kin|_m)?HA066Cxu9fvdWsic_&n(_;qABIVz}QY$ubwj
zd1qiWsG20p46cV!QLfZL80HriNYjjaYj^4Q`V6OMY1AiBGV}qM1~fV~42A<*og=ok
zx7gg?q}^#VOcJ!Ql*$zXKT0SF%LHN7j}>woHmAR5LL;p#x`jlpwPbu?Wz4v8!RVg*
zEc#?1ckEf`sHn{yiS>ihM<v#i0Ur6_W;6D@i<Ig9D@W}H14o9>Uz}+}sZ?Taeu2a7
z9n#FAl=Y>(kzorTWaq(}_FSZs#*rT+`S0a_&%d)9A%xE~kC$ry(Di@$KmK2to|*w;
zNwc9tp^d==0k%1TQUMh^NBR6&W}kk7&duB0`{}Ff+`dNx#Xu{Dp8tjv1i}s7Ho~GX
z1PYAJFbS+*y}>ZkRHvs9rHJyx1QW(kuGcBo8;Bsph|zbJULqpTF15+|ObVk7hkFMc
z9PRVgJHKUZ<F31g!uC!)SOHNGP>w>9I3_dJas9@EKY<_$6la$v`H%kO>&(n7Fo*~2
z?d{U)bO@)aq)AMg!~g{48c|e1ihww_1S)X6Fsa<c0Rh$u4v&uL^tzPG6~Zt8C1~|y
ze)7`avbw&{?941*`^wk&<3IWuk3MqIC!3&!aW>?WPbH5d*JJYL#a_w(pYbT$N!uxS
zaJYo=<POS@z>lDO(tVE68P;Z~AaV&nfpD8bs5?$#Ha6C|vwD+j*RC^6VxD>YY0jKJ
z&AmITB>jQs5h_wG$c!Pwppy_|9Qunj7@Z?+4!KT}A>D3|$%!Vy%T^h4+>tu2U49z5
zH9p$u<oY-r^1pBOdMzMzL6i}MT9HI`g2fuSPq#`kQLVCk&}Msg2OVeRY;R*VvCask
zqNzfp0Er<|?%o&$r69lzhNQAZC<W!2X~g6NmB|^ZjY&6oB-f<NNA?*>mZs!dX71tU
zfw-l26m#$PJ+8d}E_YUM`G!GBy;>(phh)A%Vs>_#7ryX0hTT4IyzwTzexFo3E;cqt
zy(Oh`$ihsGvnvZoC7GO>W_o6Z!JyC1!4B<Chq>uRCK?k&VTmwsjeLO@Jr9DAYPCub
z1k@T$8qHaT{UK?R(U>p{OCf2ZF&V^3Mx1mA{pi0i5(Hrh8I}pWOi<2^8`(gfU%RA1
zYlF6<R)GAqS(cF*UCiShB`o*RMzyjC8H#;39`NLey+3L=_8NSXP)5KgaLnH?OH&5j
z9%&qt3}cRtT4Y)?IWxhTGiQ)OP>RY7dVP}25-3X)MKmX;eR3p-2Seg`$n@D2Lf`x~
zUYX2&wu>J|phA9z<|w$+vm1TyeEp9$Qq0i=+=b-xSsjXHeex-g^*S>TJwUdx&;A>4
zA#{dQs?BMiJ~+ZPO?Z_bM?4h}N@r-H5Dfb<J2&poSUE*=dC7foBmKeW4q!K?#O<PE
zgnZwTmZmA)c9-_SfWctEaM<U=n;*CiF^}j(QG_)ana+qxA)op5v&8X$o$W0;oi<65
zco@{WZcOW2m5rp=O`I^Yv?!&h)*66iVQCR56n9tevU=wx6U}KRC#Q+Rkf@}vMw1Q`
z+T9LYJ6r7R?b7LXuvmgpg>tn@6h^2(Fw}-&FXrW+{gS7jd79<LX?LjjN{+;3G*&z0
z#ErAbTc-Vv&;M!NhR+Z2EZ<;^oIH$>1^{jCm}q3qZS=%@?`)1KWt-f&SvMMo{Gc20
z--`WeA?HIXsW951k(F|ZG)*a&N>nN}W~XP^+gzu&zeB2V9m(2^1&_WR-RlA!iz9uu
zXSKm(IohecqO66y91X6++LyZAcW?o)`bR(;N>PNcPK0(a?2%?p0>oy5EVBd>H^RsY
zY*eA~=p#IO?ksb^d7B%ryoTCqQ#KL3Oq19Qb<9y3gfIxJ32aEZ)n@15UCOgfYRmI*
z;ViZ3X=E6>CM549lq`&p!u2XQ@+d!$(l{lJ6EdALOb27uE1Wc*a8oI)b>-sar9~EI
zXSsRf25Dx!_^(TUDnURLIFZ(RHRPZD$%{Pv@I|VPhUbLMvbDA4#`<YDRiRR;5e5<E
zQVokQtpx!xRY)5W_``cRh}qlUg(xBjN`wkBk#Km_;)83~=?oKg_O>}{wWyS<G#gFM
zoS857g0(O%)iS<(3U~~S(O$C)A?`~zkW#w)YxC(lqeHq#I=oFBcTn!r;z1XAv_q*<
zca1Ba{GDbwPth<K4A|JX$L8h+n;Tm!%+K?wr=MkMd6CIx6O$#Z-M)p+5>z>2VRnv6
zqsi9JCjB@;XPPjw<a(fa!&H)_92_08v^bBHvXB_cXE{qD&^Ym0H_FNS{o_iI_Qd!7
z$69HYI4wTwKm{8Jf>eP=I5ZX!NMxx@d42}jsG!rfFSScDKVGmIM3jUh5JbwNq)QIU
zT99q*vAN$uloXYTIx|l{&iuKvG-v1C2r1!*hK=C}$Mx0R`wHWHLYAiN?l@uX53XHx
zyffi6U(4kZ{XqvIpi(XI!WTZv6Hh$G_rLddB<YZNICS?hKc4_M7hI#}swf3PAPGxV
zl!|Ch%rHB@$kD+bcQ<cS*{?A_HAkgfCk!J>r3z9=lEej*S(Xw6PJ?yOPe@|dQCqGm
z%25bfxlx+Nac+eW1f?3K+5)?~N6bz)oy3fB-NrdL*#n`taK;5t1?c9J8GXX{<kc2|
zD1?kT`Tx9MF=hnZ6#w7ikZ{RZRna1(N9M)odPYH=^E;(dnM$Qfqu$`c`Ey_dgTb#j
zbLK1y3k!6Pj%YR;#N8IL&d?bondAFP8DMnAVBiFuCz?$cpy}giVh#%FG%KWv+;zJr
zn~Re>z(f}JD)(fc1Y7uorC4_XLi&FT$9Ec9C*4t2B|=p~Ei5wk%%jwHH#xfFBoHjF
zt5t{$BOnu?g>@QtxI@AUjZhXvhPD!EBjSxi_HW!^;^Jv$((?!<$V%FY=2|CMh!6#M
zC;g4cwe#+*t|6shcYBLe57^vWXMJ}AjYF#%jS12uCe}l$)sSy}>l-}t%%}L#OF!bE
zwM%y}aAhRpN+;SH*J)}kptyNslXic=(8NS(Oto5}>^pg97A&e1aQnt>uHU^$@BMd3
zykM#|h7iGEI3UiPm}IqHqg-<Xg+m)q3ItlFB$;4+V~?0E+U;NP@I#k){PBm0Lbu69
zt}i9>j#KG3^N2T>mr~qT286`=*BD>gb^{>ttig;9D)MFq=?ShrtQOKu8ac5Xpd_6F
zx^z<tL{1>~+`rrqoimU~B&8@~YGRU!$D2Iy_!A5VF{Mg{iw|6;UaykHeO`I_$HZAo
zs<rF>($>qE_`uOiSmg92;X9zL*61X4Sz9h-Yz#Kj#n5;69knYz%5m5qS1S1fm$e41
zV}@~`!JtiQ66f?YM?p~eA%#WT427aRJ<XYKe39mZXSn&@pRj%P7Safm_0KCed_VCe
zkkX+$QZbAT{mlbp7%+MM6j8HEr8a>g#3<wG1tSz*3W>^xkSpu-D6OD#)WK+tl8Q>T
z>be1aCuyx#rO{~6>-HGNJ<gmv&*cX$@xj%r4C2&xO}bT50hLgqY{vZT1mF1D^L*=@
zpQl!<BcwzH0TYvx?CtDx{pNLa=DGs|rP$fqVR`X1l}eRTDL^3!0tGTc4>hgU5jSsr
zNI#3IP1I4+Agmkj)M<D9@N#z&ZEbJxqo2IQ?9?oc#@A?0ME(KkV(*Wh**G;}R9Mkk
zdsz$@UMs=qpf2nzuLg{M-#@M5TJl3Z&)B8Rd)EzFGN5~u((fLiqKHauf?91R_ly|h
z<SIJdHc1k5;oMn1|M@TQ<kQbEH$RVj?lXMlt6$^Z-8*cquM>t5)%pafPI>(|uaRnl
zMx#wu2p?OtlTaFStOe4YX4p(4jBz@PLXsJ0R|QHUHOK%sj;{P;1$k1qE?;8;cSc3Z
zA&{Bz_s=B{BGc|#l`6CY4VDx`Bo$>LJqH{j<r=mShLTi72FxHM8w}}pdSs0{3(q}A
z7*$+QCw&50;7<<Z$bGb<mIQ9zIfA48Lw0v|X?NT7hkaxq5t&9RS0$3tjkD94cIR9c
zlj)Q}>?%#8&v1Lj;Ru#?XTaT!O{OQBz6%wiuuOAmiltL$sMhM-x_OIhuf0#8B4(y$
zIDKk`iN*wA+1OfRI2;m2kxS{UD+xNDZ3T<tl3D3=+^kWgIs`ez>u+4eTEqOz9AEn4
z7pT^h)4udU8{X%feD1z&sF-Nt#P4(ek_T}9g@vVz5IPFpxKFc0fE3x7-T4ZWa5Ijb
z;J8I75K16}pb)Xo!3d=Sn#~Co7MAJvI|vD%|HAVuEuW%4j9EIh#7+><Z6A^hJ9OGd
z4AP8tFDCUoIjvLDG$srsQ5Y7LvMkGvq1?IVWX>xr01+GlUd)c=I6qu#?VH3t-ugs=
z-TjO+zj$}^-2)0sCF*CFIsN!U#0M=n>LHM>Mcg>92Pb&y&IJ5OGAWS;gs~83?5(b`
zxU-Kvm!PWxQ<OM7oO<GYlnf9Sdfh(1e(l$M?z5kxebgpPGj?{i>GirODVUs|1fbjP
zGd11d8-Mf_zVel?@V&qLK4~%}O|=`FlI2L1DNrvZoQBH1d%OJQU%teD@~{2{)zUPn
z$|y%=be5vCjOi(bdN86nHATDKW^Zqg-Q8VUhi#I6pGvh#ty!nhY*MO}eei4u{RbIq
zNt2A9{P<@?m8*RD%U>l;W41Q8dH%DXWoCK`L5%fMA0wRVIM?NV&+cS5s1Lu%3D2YV
z8o}cuBBw^>Vw)#Y2^l!VRO2$MoG<V@GY$cPa=JfLQ<Kch%y9PX1>8gpn3|nqa&iJC
zV0&YYYad);m>4qSCRG@%b8*2@cI&@y(aN4Wa`{O?+52)BgeVY<{CnrY-ygDU-qF`@
z5yw4;FdCQHqjH2tA|pXy9KS^h$J!|`&2#P>pXYGzfcE-6k#MACUkn8xvx1TgDcP);
zB4JZQ@2JHvOQ<v^C`WaS5``juasOq21d@0_k|e}&hSr8GGhDrLg^6a9YPG`R!YQJ#
z#2_9rIoV)&X_58yb)vH5n}7T*PM==km%scKd;13-<@R=(?@eo|*GoM2>?1t)+!ItP
zC5&}wXb?t(r4sXt^Hge8qH;u~>d5<t`$xQY>k6$_n<O4GGcm>am9r=*+1}me_Psl_
z2R$Zcrct3I$O|nonPfQhr91C<jWt}k`Yx}ZzeKIx<b}_Fih85s_h%mDi~^D(F%sT9
z1y01-m&4s25yuce$HS5Zs&1`&{uUhiEgb1yj3I=h@{mu4Fo1ImHZ|z9N*)9XA>7_K
zZaT&E<P;abe2!;7^=V3_GCEC>QHhmPXE}fV0k*cbX}8<NNzC@{7Ugo4uv9AEFY{EB
zG>eIc18TJj%S($6nM^fl9Ft~F!M{LKJPMl&6oU&m3i*%RQRj#xf4B`f)TxA0AyO(5
z|M_X^c*{{#B2WsUGpsPEKsmV=ZN0d;p`^U#qYz&92aFM<Iv_OwhDcc|LeSUHTi*d=
zP%<J@7JLKK$oIHSA$_;9^ud}Eio?AFZe6=YyVax9Z`10uFvbu_K@<jnp<bV$+ijDk
zDR)*^sZ`5!Ivv+2VHnv?evWu7<KSq>zyJ0R*xBD_Fo-!k>TvPwc|QH@6Ra$sWpZkU
z)8{fYaBpp$+gIM?)wf>b^wKFNCnk97-FI9Uvj7!9R1Sz^L1qM62PB!K*Y6<F#7V;1
z)*<zYWfqpI>}}qq*Xxs=x4il0`+V;Crx9}Kc&kDbbe5Ccu;OnbZ)Cy4cwV-wID@@r
zWKnGzjT#faw2Dv|1kwkTqfhW-7qunH9H0;c5*4}$JJyURU@o|-)@m#+E;AhTeIwF<
z7hd>0gF!-La)PyccQ`uO;jqy_XDPkmfL@{*BpKRhHzCf8CC5oz$bn_q2%XFC#kda|
zC6X5FlS4_4N{Av5Ow*Lz-QACEC!Y+l^M~5#IoX2OWO3Pt@UWSlsIvIT1N1j`S%32i
zHq#WHTw|LwU&1;<vBVpWFi2rhR+1d_=&f&&4F>2bANcveZdAW^{}i~*qO+8>yLY&C
z<2o;X?nTnrb=$QL50L^I%^I^)6WqCbmzC3VeDj;%;9GzEExz-e@3OXbkE5d&M@L<>
z_FsDJbEI(aH$pK?Gk)~b*SWX8&0qY-f6kLnJkHF_G*PKaSQ4NTrptArT7xu6Io0no
z7z{W%JYaic4QnlxO2sz}K$6B-Vge$v)R3kbhlf2{?LMuo8|?1y@#QalmB$`>jPHN{
zCp`bz=Q(|525X(@;_<SMzvla7%;P~F_yERd4pK_QiEg7jXgU_$dBX5GF*?t<w6$ag
zQf(+%=c2^#D05+5!-@(#Z$(OXcnjwo0u>@;$jsa#pZ)w7Ic&A)_F4=Rr~RkBz_0b3
zI5TcQh+``Kxq%hm`B-A&?ujcmWd)j*8|NRojL{5-3GuM+W-4k+Vl;`+u6e}oYbo4x
zh#W!DIz?N_)cg{Y^E2$;+aXke({U_#B$8A+3#+IQA)FJ|N|Lo>I_ulmeoW47wPQ%D
zu+Ta@pi)+BZ?5y+yYF-T`Yp7!1VKQz+h=ZekxxJOS<bASV`*WDTdTLQ77h;%$#lZE
zzWz=A<WK&T|K{KRTQ)YfycX8jeghJ%DTe{Qq2}EyH<_K8MyY^{=WCS1D&=a4APlKB
z8`PS0!mv!IJLK^25E<22K7Wa2DH-;ATzTg$-uvMdH!&iRG^Zw+o}Wjm080iUKpRD(
zQ=}K~RtVqZ;ADzk`sw#6MIoKe0ndN-IhL2_y^Kiqkz~c~7tlEaMaTke0gC-$>}Y>!
zE4*%^Pcke8mBaBj4S^W#yCT^z&Kb#ETs^;oLb^X>?wK1WeKI{Y!}8(^X*OVQe}~0|
zWdb3{h64_hl!@sXCa0!J(#(mf?ryWUyUo(V+_6M0j{6LTeWa8;_Q<2mOiqy{8N=a_
z;c$2i$sFfG<<2K>uo7PUdOUd=42FE~2S56QZKzV6pi!@3jAk(C(d`}<d|IWH<2DJ%
zN|N%ac}_iaiS_OP(;ns$&xE<CFe7;mok6M<+B?+@O)0cUGRO>3P(eo$q$MCI5U!$L
zCOsz=pP->=A06<+?|qk*r3)m(l&Sg@FMjq*bUSU@og>yZS2=1Q5ohk4QZi(J|9~vZ
z==X=Xs`rVY&i}&*OPs)4@2+xheVZ)HNE6MOrJFPwQ%p1#K!yaRI;YM*z~sVd&Ubqp
z?(cB(+Ew2D;41xopXuo-dfm3;erb&goPJuGDc=9^4)1+%m1?ELpck{gb;Q@czCsjL
zn3$L(P2C;6vwD}ymoG9qR|QhH(Z^^y<3S7@2N_1CNB?&-N*u^hI`7AH&Uv<Zf|UCq
zTU43~U+1fL0z;N*v{oQ}gPqMaCP$wY1|hY2gXx($TJ1x2cKh`EJr);EQEN_+r7@jW
zi=)GRH?B`g43fQrL*g{{rEtY?FyN@wVqtEcyt!lC@BB_sAwO=6bpfw*-y=<(#^i7~
z<lp|c|NY0dlTWN;=Pu+pa-+3A#zC^yVw57Bo@C*Xi|lT0(%sxgX-9G=AJ-&CZ!C}=
ztjbVQ6k@s(iHtS9jUAF+mx?e*0+jF~wn~mkpSZUF&8>S(O*J^!-=JCv5V(@95)Sfi
z)~c+mEU~q{&D`81fBI*C%vZniHImq{yS-16WOO=x2K~e#b>kX5Kq-)f%9Xq{mR@4G
zetVt&)BpMZ!PL|wb8|DCT3KRYVS#3|L346~si`TN%_img6xHS|le3G=&M$FvbcoTK
zEKP}rF==YOz-F0#oUpmm;n#0|K!4z-Djc>B_|D(`4SRe0y!g2n`N_*K^QosF<&j4p
zbfU*P1(DBgJ<e0eq2KYwi1BA8z4Iu#MvVsT_-M!GT#gy$$-^GpVU!s`l31cjgL0(~
zQXC_gAA28v8F_%12U0;4Vl+q@pn?*WYLm0)AK=-~e4bZ-{R)Te4#RjzREn_1qBHls
z^F5hg-*~gcczb+qYz5kIOlK%90=tvrlK!E~8W*G)qkZ#<pZSR)5WyH?lb66P2utSe
z%UDZox`|Mp^F%&yUKT-lrkyTZAQXX235+1jB<=U_vHIF;G$!V$EuF<dt?rq1x*c}5
zR=IrXEO+kQ;?AwxB*P)qY6&3{lpM0PwMvpaO{3Q2!Ap;@dhZTcC&oEFGr`kOKTYeX
zN4qtk*Y!@x=2j!Yb8ClbhLjPj_xAbk{`>Fo?eF}Yb7z+Mvu}NgZ~f6%s8p&LZ5Rww
z{=a|u1OCI`{fNV(4oWG?<%nvf#Alv+hHrfBMUGByar^o;CMG8+SE|IZCe@l@Dlj^v
z+Z(cfc*y3?o@-ukXQtx<VL56Y@R$GLf8c}bkFd6OkLN%0X`cDilY~)-lU6#hug4B!
zS8nr@ZSV~!QaJ*?_9&YLodcC)zkj?5K_g`@@pD`lSxAMkZZ@#V3}uc7n^Gz&l^TnS
zD-3!));CvKUtcGVW6qvEPas1lYMiA6VTpdP&F1=D-uTU{?Coxm#1AlxhXjFh1DVc9
zl7!2b9$;~Pfh^4!3<kt;?2~ujSfZrw;4?*HG~x&4yw{N<OVgC^|M18Bm;ayt<`1@^
zYNbZC<~U5{avdQAX_5dPr+;bP`D3J{v^dX&FFsGXQsMfKU&Rjc2@Yd6iC4Zc(j~`U
zMhfE)BV(<bMVh2o??c>sSO=XM9J=alNeUO#CBp$9Tz!xIy<HxA=t(+9J<>Shp$8r#
z*CRYU+UJKq{VuDUt7xEBts{^uEv{gV1Zz=Jy2(rVUL@a3t_80@Foc}}l~TaW)Fh8Q
ze3@={zz=`?GS_b0!b(A=4O`oL9JV@~JA0PTfA&*6{KPZtZmqGuvqit#L0IGVfwc@{
z%d5Y6pO=627M($lKsfb<#>5P#&zvVviit*pTD3+JI|1-F-guiYe(^cN$N>!Z<Es8n
zaDkknpO;EUT+aOW#?S8YWsF9i=%_7%NCAOz=crO4&50RApd7{IKJT0R$1sRcGT`v=
zfQ9)btT9w;HG16+{a)Ac^@0Ft414=~^ap+F^_rV+Zw#GIhcrzca-93x{NQ2am-Ekd
z4SS9gk3XBNJ>KA#q$#)WtbS}e`DCTo`w)b4QmvA*IK04dfYK<UyfV+ylMivD^IO!>
zpwM!%qf;Ni8<PW_q;lj*VGKq%7`?r{L+|L2hS3B%bWpe~O8LdwSYPMXjraNEZ+#VG
z`$Q4+2W^xRWSOK~Dzm(_$jtN<2Z#Io`9Jv+UVQOICMRe4!S`PxNub^86AztCLq3`+
z=TMZCBov`DIGk{7nrW1lqz2lX2dwQJ@z(n{u*MWRPvC|-*J^c|%?VacEwi*Z!_4#)
z6ZJY$1oZoT+O3qWgCp*(-{#ip8f%-I^y3sa#}kh>4EfcoKc~|^<jY_DGOxb+Yx?~@
zPd)V*K{(2Ajv*#v2;K3H*HM6{eThg2KSNUsUq11iMNb(2$iA@0MHa1a0Y(_rn3-Lq
zT&p{h`0t2~iiQ{WZlgd<YpfQ916x>HVHn3emL$YkMyK84pmj*4S|*SR?CARbzH)1B
zEII!HC>)jh?v~8tU7vX8^oJvMFlz28W`Y_=OqL$4%WqL0)Zqg*1Q?SM)G7pFh)r^}
zHJ22DMPUi0)3UP|kAQn8)JnMhlV4FPRe11^{)}>I2^qL+h@yx}DdgV0+r0V4ZwO=s
zZ00lfnBkzs<m4=CYd2V&KhL?-7pPaNOg3wDdmWyB>KQJazrg0kfg=^isSmP!Ym_gu
zgv#k(x(aQ;Fo`+ZJz~E%;QgC-_^a>yh_8L+3)JgXe)Q8{^7@-sNmCyv2xzw*nzX+0
z1Ag<yJN(f%zQXwjpTuM-{oas`jeFc#+vLXTCi{nd`okfEctDb-4j~iLpBL7Rm9mCx
z*yp$JzQxVe+uUAVWq<#WFMa6?)avD8l1cvT^1Z5*LMa#EAV*5^kqVH$ESV+=Iy+wa
z&-YRxv6GjLtSADbQF7!)W2GrrD^Otx6_kqPNC>B`S*cW*o0})gQik!6?VUAW#!Cpw
zAs}2Qq}8NJ%+bLv>uakRW1OtT{yu>UD3?lZWFt^2mB6~jh$Kl0)IOgg5`~eUNMbxv
zGNR-NsW4dzV$_vB7!LWtk6!vmLA_NN6p~6|6j83!2*S`ODfng-2n-NJ=-F8ozVIxA
zEaApaegj$Ngtz^()FvP?#!s@bSm`q(;r59YmaISI_A9?&^5P{*(=#q{!l3dq%ljbL
zI-<ZDdc7{jXeK8cy#K-5oI7`s$>}Olxk9B};_$GATn+f6ul_mz{@ec@N1a0|<&aaS
zmZ(*mWJyM&UP1taM3APLM|r(BvM!)hG9;3cOJ~pU%u`SB#N&@4k^KMu>Th`c&37;o
zl0<_DFh+3w`YJE~>{opHxu^M;|IxRoHyZ5kZu>1RNK(ryzrMx~e*PB2)SsnNpsYYD
zMHmKX-@Nm|hu3)T$`#I?J<DScKTe}j;<@LZ^85UwBHa1@A743Qm(Q7%QhJ(*NAJw>
zq&yd-7kI`@-Yhp}TfmrKb2e70RH!$aNEM)M`Vn7{BniV|pTVF@rc;Cv3<iDTB&L*R
zOiWC3{=!AJH#Z#UP%gVET3Q#CB5g*zIc+py5Ee-jnHy*4(8Qb;;weo&04&gbZOCj^
zAd=Qv!XV(u#~){NYx`r_$tMi#Mu*&p%;@LB3AdqbKEBGhy11HXvi$I6x*I!ey>$&G
za=Bv2SzUQbYY=Gf2$c0AkV2xcWUU_E-CcA%1XUu#74aq3m9BhYc6N4}-K{F?Yj^m~
zuU|oDidJjiAs=bVnbW6<Dk;OX&5JL5hR;3!JhQX21S%jNCbW)P?Cu^Q5M-%#^?6@%
zkqSZ~5Sim@WF{lC25n%NWY{1;Af4kf4oT6OC}xz#$&hv@X7%0{Z@zuSISHX?HtIC%
zb+RO*HyE<l?lO!G8aG<W1sJlBXK|Y-<K6e)p*QIA&9D41zkTCP;$+BYKl>bEsGMUL
zZpeFa8u$&L&kw|xvxOHv%{yL&z|9|GoZ;Y>3w|gzCLUQDml#enOAwWrnwg_otAmuJ
z+MvX7^kcNKobkDGw8NlsI9>Rz%P<I;oSep<*61uH$x?3K_>hCcLmKrujauC`ws;Pw
z_DvEBholIpJtC3|H`@X#A6>>cT9$Xm<_KW=NlfLw%a!>8lyt(p0;hlF$jZVC{Zbfz
zQHZh59<srJez)iLR*nMUXbcKTNogv9!PtycOHvbng-~V=T|OA{;SYXFeRiJnU;ksQ
z3K7;)E|=-{hICt79PDi~GqXsm)#Bhl69kGd3>glF+`jz*ON*zNnVjX>r=BCDO;ncr
z*`NFg(^HcSx<j&bcr1Vq14$TqlDH9`J#9#>CN`QRvt<1Qohjb=;5JvU-$h7#`D8Tn
z*i%kidD+07)h+(t|K-2)NL~r7<lv}7Z;+6f{9GAz<c+U<>*YydmSyZ6?DL(MzRTLi
z8v6$aeB~=IuyT6AAFO%BE$2}==>g%LP~p6UL}-xG9+P#*vg6V_&Tiak=f0l->*O{H
zI~0;IDpLs~WDrrSPf)2hJvuaMZi=D^Aq4aD3$%_7=yZ-;$90-uqB00c7?coUNMmw_
z+4&__R?e}pwLy}mPAMQ65=Ei6Jvcf#D$1LAXL#-?gTRlSlgf>H6!@X$z8o#B^978~
z1_4|+f1VGo-~5AZNLy2gJ!_*Kn&zttQbvp-Ew=)tC^siK{laJH?jACH{VGF~QBf}G
zkw`M5y!?o9d(?smECH4h0Yfdh^U^OdNy?-D^q<i<djSzfNHj)TROkh~TP>2ofM&DK
zpx<F(X@U8L83z3xZ@>8lNt_Y{W#;A<dHU(+*xJ}-W@?r*OJ@)=V|KdEi!XeRM!iB1
zD5}*8!{LA^5GWP;S-p;Oq5vfWSMqJ8+_<q#tMwZ;H+E?@>)gDv>XK7`h&t_2m(UF3
zl%N0NHSXNG%m3^z{wdRQ%jh&=&`){o_22M=U%W+}<&6-&1S$lrc8k^3RZcB0^WK&B
z_{A$fr`zlD;f?oM-?#@hV|Hem3l~p&zT8M8|9;elJdsgCIO>n^-St8ultE~X$xcX`
zkXQ0_aVF&NpI0sXw{|oQAuv`Emg@vz<aUmbXq_Fi&wjs0x7(uEbB##g8y%9EEJ<lh
z&hYhb{Zp1#&e83(>Gu2F+gxL^IZ3TnL+gwri3!7yiHQjx(pWby*jlt6*EAeeV=T7N
z$sI3OXWBJ2k|V@I@cd_=<$FK;cyxwOR*Ln`5O8h2Li)~B<DhpOncf}5ZWD>1%;fwc
zXC8i#gS9n+odIYsed4(S2upxL3rQjs6_F8IOGtzi24z5YHJ#Np($)b{xsLV_u97(l
z<<!1Ql^VUlkbZYSv)W*5`;eQrZn3m@fo{7`b7G2KQX@zPJpAZoN~IDK!P>psOifl<
zU%N||BnSkxN=Og}euKMcTv(8y^tyHetsTK%XJBmNGx|}di(ypaGsb7-IDTGi6q(i>
z?X}R`hvew>M{rSValD%RDYy+1O3AIet9<8&-{nug`KP@7?p3rgy!hg$38Mg`121-%
zjde0pI3_1aAh7NS2<ys9ZiD4vgZ1TEhZhRz8(f6(f~EK_Sr@ou+JLOkoLZtbF@*{O
zv^GAI9R*G>3hsn&z;L6I0)H47jMZoZRy%TP7=+AD&v{`-K*POzt8DG;V&+{|W3BS8
z#CJoHJ0=Q4622W?NI}l$9_QZqpByipr60c|>31!mr0)!O2%FU@CW}XabAis~fgPjy
zyu$%|H}A0i&P`Nii9A|haU-1~X{aboW>b2ZCDs8#SOQ!_MTDZSHOYRT4}S1N<{x{4
z^4SM4(x9wjVfhScJmBf4pW*1R&F01qH*eme*Xt7m0rh%~Mx)No&H*oc{;NFm#HVQw
z*ExM|h6gS^z~s~{zj^&lq*SDdo0Jd)ic%y{%1a_yS8i~^#tzj5L7E!+)=&xrnfj=H
z)!6MR8XHnSp~9d!=%r4yRl6YBiadB5l|J0Mza+_s17~>m{kMseA%kSdpZ)RIID2lH
zEOq-MM}=GoHm|HWGEW|qyZ3`2aEY)o8sp0<xfrRyVueQ{bIm#nLMl>YL556CFVU#i
z3wlBjl(0yb2#Dh{3Q8%e)f#j2^X%?!FpLKr92`)qRxw)B?{-)|eV#`jd74Ij0&Oj?
zzV-^0YL#-e>?kbS&B-)cvVU;EFd0$`%RXrzm8o3Qns-onX>6=yCXDw3))W#Gfe?J{
zD=+ekSAIpe+xrJt&e-$HDK#Dy?GL&P`W?WHJ1a;t)&(OL2!tf8HF@Z(FVecR&fZR&
zL>QEGdWDWJm&;%Xq=H(IBCH`2iU}p@CzjP8zsk|QHJ<v5|0#=4eFhOk2<3uDDHZ)*
zmrxGbSl=KFBhH;YOH`_$lwf&fj(_`K{nymW4f_2Wy>5>zb7;@<(h|)n$?0>8JpAB8
z{QMWcVt#&>jg8HG=c9x`1>VVP;SXGhvy?2;WSs&1%&@V!g^+l`UytSs;rxIxt~=cC
z)tjpu{OA9#f5(@<@M$KR4SxI9`~2$7tHjyKDMs#`8748`{@cH%TrP9eI&|XUfYmAQ
zUVVpZrA(z<p;oW3xHO9pnGfs(_s2gstuqfijDkIf!a4(!{+!g(pgreWq{&!NBt2@U
z@zCn%td`EVT5uEwV-Z1>YOUd^N*2L!?bako==HjEJ1x51?udF|C7I4J#!a_>;_1(F
z=KN)jjt<z{-QoQY-Xk9N2m?ht>=VZ^Q<IZaE0qF`3&X&T03EB8jhBCO-;^hR<J2D^
zgzIV%P9AD$e)ePA$tM;l9{;<{XW<naAkUL2{%vJ|ji9-*#L}Y=(rv$o92lfAUe(^?
zLBhClqp?^cTz%T)qg&jK*uA^PnVmfv^DCa0;zl2#0*r9;B!g1O+`=;Nz5fAoOQ#ql
zmbcz~g9rY_BWRscE?1~k>zp}1%gV}W&Y!<PyWQdatMAk64d`~e%+Jp;91e<y;~ug@
zSF$W~Xky_I9QQ<PHx)ohiPd9<fm8C&<vnt3o6HytzB5iZQOlgrJ}%_>JGouL%^nrN
z>fKxX;HB^JM_>CRu3WiFr5y46^Ut~NU|+yR7#~;)|3~_R&liMpIg+tMy2zA@jffA-
zoH4o9Y~@HQLXsJ0bkox_OiWDp63{rxoBzHo_E26P&UrgI0o&a(kCtVg$Elo_*UZce
z)@qE-&^lvtbB(>deH!%!K7K6(Af?O^m(gdQ6eI?Ym1XlaKAGZvCFp<XGDZ^*`wR!Y
zLgdwP7#%r#Jh|1{AgracvBd{Jc@4eap@MLvd#oQ7XHdd|(DYM7FH=bCL|FqBIOz|d
zWCJpr(Z0LM-pw0KoxX_j!N%0w5<A;Fl&cfWhGpV3V|`<t{eyiXTVi{si$&Avw)v}X
zZ}au9{u7#VnJA2CwYo@EBTY2DUWXu5geIe0R*cB$8mmn)=Tun7e=)|A8B3D6=M)rW
z={x~BE_j;<+L`vHRiCLt{@O9=8v9ZE$?Y-^QjAA6awIx4hMTuP<jVW+GCeiL($Xn{
zKpiWgxbKuVg5~x)cK(T6a>2bux>__j+U^{yK&T*3L3~rU!041vDWXz|Ac~4+A}Q49
z9ARWT<4&MbsZp!d2@en0-#=htqDiCPWPfj)Kt(LCoa54k%OvrTqoWq>_5tN`86hRT
zVV}4kdpV{uNt#e8RZuGM6L!Yr8f=EZbpry$JnoMQu8xwU<+&LyU%J2>Z@v2m*-k+l
z(V8R~((kwF_j*KO$q5JNNofJIQUs;-#YN6P_c)y&y@pO)$YF#R+1Kto4wYdd%1{^*
z;~Z*LW<*jVbwKaKHEw?IM@(OMfM8~kC=3xn1(d=fdH>2)w$|^GrU}n{>QkIPcY!br
z$<hR4G(k|}u}7ce`iHk&=O~7S#aYfa8ocnu7nz!xWp8(%m6g+MZEY8zg-cXj#v-L8
zPF-_UqP_f-P7$eeqXvyjZj_Li)Ht7GEoo-Ra`LMXphDL7+Wfn}`T-zFGL1IJ)fA33
z^#I*|hu)xDpksj7skFRx<z1E+Pa~D$g%>`<g$pZ0VFcELIK~i$0a6OjVIH%i+=d0@
zO+xwTK?8(El4gCUaNwWuNQ2Z!g+V$kJ%6^BYgL+)^VDj!qWW?2`?&_Wgl@Ov1G@oP
zmQkrxh@ucnM(bdYrIkA8&t5{WXj(@HNFlj;^(}@;Op^2+okB>`)Nxf2paLJPisOhK
z_%p}{rUb6iG|mBc{6h`Op~>bGP#He?NHL$GiLr9t_=5p|*crz`&_WPfi<+8X@zICa
zU)|y8_6E|p0m=xE9Qmx!wMh%43_!R{#d=uxXk(Lu+jm&F_y9pEbWT|Xph^g(P(lz`
z&FZZ?Oi#_QzIF#G6}N8Rpw-$(1&)YXD%EH<n@mm3&}d8`Wr-}aY;EoE;)^d(sZ_Xn
z^$Kfi_q@Zmt~8XHQQ0sETo96`2pmU77oyE4jKc>tHb3yR%bc{(g;R1H%-4qa=so@W
zmc==vS6-{}#TUNBZ{K)_Qn|*npL)vcQMmvz2u2Rb7KWJzS9#WYeDk?yasmo3_zTJ-
zZIEe;v6_j=X(p$qiAp7xEin>F9w)Z`E{6+v{##mPmT?jv3#@`5WM*dGcNw~lyzQM$
zvh;{bR6;3JOr;p_{v5ydoa}DpNPp}k7w=fp#V454_d8cXhD}LgH#RQ+tTc^LC2@>r
zhj9r7#;~_~m)_1UHLF}#pSw^Om<vOSB*ICa7%Y*L#q7oW{j30K{LD<xH*>ZZI04*2
zn_qqRuUNTwiHX@c9)0{FI^8zC?!et6DJYdo?CtOH*4w|~>8C$~k^$}ZfZ=e5t<4=S
zUc5w>X^xJL{8!dwnJyfGFdC`cD5@|F$&6+gCyr01J<2LQuH_x`_}<90AxSkFiLrWY
zFOEH5aHVvP@7r-@`&z9^s1%)!Cqs)|$IaQ(>fPHs{NQ8Uy}QPxOXpl-XGil+NA@#9
zvy3r5O&Z_d&Ytr`I2Q&#76j(i3u_1iMVbs~w+=iKSM_BY3g=iLjAP9UaU3(~_q;=p
zq-n~*!J%_#26lJ1nVg!V(QL4C>NLwsr|9+Cq-n-rI3yi7j#0TBaAxHc<x;6|g5$xI
zTgQ2<c|tZm#pR?8oz(dIhb}R8Hfo)b`m>=@suWdN?Z-U&N8!ZG0V51E7tS)Va+B_z
zJq&Kzh!ubo1i)nI(4<c0KCK!+**7bt);em>>N>jHCt6g9u!K+nfq*M-{T7`K2%;Ks
zVp+YrPOm=%qxt#Ie}<4{o__i>?Co{f-8t}8CPS$la{9~?^YbgTTYa>4^a%Hi^S$M0
zK`t<J{NK#=wjltWNo-)zey*^C9Nd03#tG|dV_iaIUDAZi<KLp5>Fv)=Ke*q@%ba*S
zE4rhtWiaUT^H*MGYI>3%{;1ETi<fxxk%y>MA~rTQXtmlr@x<dan{|)axl-+TuqE9n
z5OVU5OV$)t1}J4lI&XzgzCQz{La2yFeUfU;i7^L3;PnK<JO(_rr*pfFBKO)jj){{Y
z+GI@6Ofx$><L$@N>$Tb6-l7ziDOYMtPtEYi!%woleviF_L#m|+qb<FDk51Q3=}>_{
zYo}0Zb9CQ2zH#1+X#Ia9Ky$ov*6nkr)oL<Le>nKqcJhhj^j$g2-Eig6x|7b(|A%)c
zj{jy5=rE+dxX9w=bC`-GBlD1b4&~<#UwHLux5;F2KP;xxW9P#g^bZdZ<N~_xngTSD
zP3Z5g@rxh)EyK<cfmD>MWwcIs>+Ls~oNQu@rBZGX$TC@K*xfl`&`%H|U~g}~h~7_~
zS|SW3S(Z9zHuIf~e!~YUD8|WHEZRCF%L*fMdR4|sv=yY<8CCw8O!JY4m~ZTp&g%Fx
z5LByGDwPWPQ1Db6(k$bHYgf4X!FwPDufF;^AAa~Dd1?6g`;OGNbE7=T0V+z69?>bD
zzkk*)rdJ3VxbJ8LwfY3pGjl{ynVf&;-oxFy{QAeyGxv9&nHg8!ab0%qMr*A};@CX{
zV~UxuN=D31&vW+7dCs1_z|8a<gMQ4>;UT?VuOMpQ?|Y4mdelskuW^TTGMN2CmwQi-
z7Lo^-4v|U{CrWyJPe>va5y+6JQ6sFBV7x@`gKi^TKnD;yR-tbe5bnURkobc=5SHrv
zJTqs{W0iEXhr=>uGhouRc_1>p`@<jei|_x0N*Hqf{5dkM872w4d;4@cU043m+D(tB
z)eur~bacqh&JNXTm1mxLn$xFOe0C0CP|76_2Hd0v;qO1fOBdmh<MHHT6v$ZXq;oRu
zI<Ch%1y4TP-2I=lBc&A0W|R5(IZiFkvoJSJwOk@ll0ZNph3A37;n4xZc)*=Ix5?+|
z27xM?UXDGdW67C3{{KYY807e)(vPCb`L8*_J8vFxQ#>U7?jgfL|0AE3qhQ7uUqb3}
zc(~8T#wzRUcWAX*PH)ajK)689l4Y^8MQf?m8Z0fIA`Bz?{XT=9n{c2_%7yc1SzcNq
zP@x~eq>sr3WtlFz*2fz|R3J~xlRhqu)b9^IHu3xKTWOMxm90j$n^P9>sB5eZDUjiS
z17!?$I3%i9IrG#*l;@iagdY>>wc-NdgqdTHcm>iDNP#tOPe}<9LEPz*^?OJaAY_OV
zmaKQg&8u$_DmXei<kzpi&ToJFHdn8H$Xjn;rQ5YU{q*N4MGaQ(+{0SPbc(hyQ!^8U
zVVV8?7M)I?Bu>b(lv1fg5Cr}V4@lFDEOnZkQV4>v(sNN=GNuUN677)iRF7x_&c;A{
zFk&1b%2y6V<a>I2#&7O_CMPFpHk&?p8pQ(phr7J`>t8ZV60Tmo#*cse5^ucm2JgRr
zjl1_Y7z`6fAMtb`88}))p1^t7fsY)wVl)XzN|4IYf*e0QB#7#qI(>=7<uf!Urrorb
zu#6DO`;~FJ&8XTar6lyTaI-8UOH*cNXE<|ag^7tKSmDR*Svs9I-EOxi)6UGy^X#+F
z10e?oZM25rFvb{77~~Cd!h;MsEl3tYUatSB?_+1jY{>i2st@)*&iX`wor}M6p3-rr
z_z}cT`0g<w2}Fnx0b#Yu)cIxVD>Gz~zzB=gF0(4c2(52B$D_q#8E@IZ{_VRQY;Ab%
zhr5PASR$KJF<ttbHyG@!a_{<8N`aF`EtO0B>eW}M)~dLv2LYhi+dE`yYln@EZB8w(
zaN+z#%H=Zq`}@T4fJU>9>ss@Jm$mf`B%Yte$Z<G>v10}_)+FK&`26Px>BTaS2X;lE
z`g>NABs3ZgD!!Hv&Up`$m^a?~Ej#-=7;E_HPk+kp?k+;eA`s87=cHr0UjXz`l+Fq0
zD$gkqQelNc8$nd6Ff~2rq)l?p1Hb!y$G>xaC}=%$7<o6I1(HHs*ptnztNuzEGCMQR
z`EwUJefk{J)3d~JOsCr+@tJ&ac$wn?Yi`tqV;DQ&PHGOF+~5DumAAHsVMMuFr&6i;
znWkZJt$AP~Q79R?jIvx|=KMwGAHGNuLSnVg#&i4eHfa@QB?+a%O67)l<L;NX00ylr
zwX>(0n4NQ4fl-C3NQk#@F}Qh!%O%aX9$cc<Yjgd|HJbG*Q7NPpmPlj6&h8<?MqIx1
zFkuNQjT#og(a|B5YK4BkPnKnzKE1+24_)Ts#q-S0O}b&=KHbxPtPWVu`_Tm@K%3Fn
zJ~#2r4G_<a^|mW>$l$*BIL{MLBG2RZ-)KYYsO9)g%S&#8g<n!icbFT?Fdnjha6lZ#
zWBb%6HU@W7$AXvq&^{)ikn7P2H%~g}_ly!Q9}o#AZ<3dA!ziLwE4yxO>na9r|Br-E
zlO&<jX|b`s%Ick)Y;A4P>vhR8huTC@3F)UpG@4CH<ucYfq21GG&at$(!f==}><>uN
zgg{9um5LvOrAgC_G)oF$@6nC*t!j3>geH$ADkp`&w{~{f-r4(uZK&UG6AuQM%<+{=
zr7ERTiQ%wElEqkI2y$c=FiDIU^f8?lde}!q0W%lR@xT{8MYJ?UKM(G3q?(L?NJ3RX
zDF{hTNZ%+@d^ui%$TEh*0m65(D>Nn@GU&9iI-^vMNYj+-*RS)FAHU3Zzxx9q;GqW}
zAx#p>rINRGgSDDag+wYM9u5?pWmHrR7lmKxl9H6}l<t)7Mx>Djk?vNyySs*vF6r)W
zkd9#(knSG7`~7Auez3Urp0m&1&*oUZI5dy{AcGSfjV>dZXPL#iWOM4aUh)&D|45-z
zqZ{QqwRyjmfZq%Ez`DVE@bEdE)K=JeD;AmjQA{-RVc1zYs}%SLK7c~eJ2;k$s!i>|
z<aQ+Y*ot+HjZ@3Ze5TsF^&{kf4|e`IQ-4^{T@y9r<2w=x!N9RvR2Q7U^o(#_VlG~8
zHSL-N3*rF*D-NOrjIe-^l21X5;9yoJd_{;vqg0M71-GDJWD;6#7ToOI+G7cidxQDf
z^UXDs{TjsnilfDiesXr2Y)PE_LFL~q5Y`*!HI0*wQd2~TeS4vm=fmXk(;R$GPBm+G
z4pFd^MZqricr%}{&U;kWD3lCI)wU`Xc%6mcShF|P=hTgCdbDztNM28XWAd|=^)eEM
z&wYi*<mVep%BY){e!9Dw#XPDG=sgpz<*a1ge!KI0c%kqQ-W;G&gy0^N+&^3l+-GJA
z%9N!SXwBLVJI%@!XXNFA$I9|dEiE(iUr)+9pZ>Ai@?`O38DV0rIW$J+jOKVQmmta^
zZ-{&uD;LRi9L>?qO^-~`991UNKI_-Z`CKRFqa-)%<>ocIvB63N-U-04Ed^}Oov<f6
zz)ct^SzGIU1j@fqgB|5-v4z*LW|^eg4Hh#_eT)Y`+HrQXLNF0co+bo;4pZP!f3{nB
zT*%1)_GUmkCj1XwJW0{(W|Ye%<}rVx9$$4Ka7N?-j=!2p0f`7atgm;B$yB`g^XBnU
z2DcWT2+J1`_7Z5B^hs!~Cd*&lQFqJ!^W)JK&g_bNd%-f2Ai`x-Xww|gM=8d&#a%dW
z9$PzwTS|Gn!3(4aqAX~u>s_7k&pPX)3SKurkS*Yj)~0GZl1KQ(qb>4f-V~Cy*#2!u
zaplGQg&jd6Qr@DxnD=R=mP4aPz59W3@AdxSMl-pmt7Bqgo^+wHkydn|G5?7t)+ZQX
zGsHA2Pid=b5ZfVG&*Vb-hg?a?$*Dq0wvX$5JX2Y|Zc(8a!B0)!bsv{B1@+oJykEex
z9~mE7ehOYWRdW>FPmLj6O%F0hHoeTbTFFQG*K2ta9v(g&M(5UHbYNxO`qE1Woc06-
z1TG>&!KZ=<;{^T3%TG!iW>tPHE?+)UU}wP*H@fgRem(d=d!jq0Ag{*l3~~KRuO+P!
zy84L7yc9+Ow%p)mY4Ecy(<ea+W~*MNs@93#on4KXZh(vY?-w2&0l_qO$gdvqv5iS;
zHSt^0;J%-?9i9oA+%ca>vJI3~RJm6(7-!MILl0L4_;JOB%jzea&xP*G`Zb+_rxc-O
zKCL|}ZFsmcUqTYJ?cSGK6CRSD&YgR5{*h+FKn2lqCwS&{#2JM~DQZ{p_A}CQN;$yx
z-rsCtD;(0ZmZ)RWd`_^4Go$uT4*jFbKVO=5*7qy>+fRu?TWzJ3;s^^Ot!i!WvxONK
z&s%c4`{mJw5SFjoyFCx=u>2fj!HLX;bam!}O`*@aQX9v~i;_uAYHiwk^&|OKmJp2X
zH`w<45vlY2nNj-TftWlKf6z?vXeI@A0&#jD#OyvOKF^@dSCCv}zdnByv>f+EM-|qx
z>=5RfM60Z(($LT}dN<8BAke2TMT*tt{!DMJGfidLdv{0P2s}H2jeFJ%@hVX*4;CX&
z$4^GSHijg%jEn6Sqs;vr{TU&w3$zLcR>(9c80E7S_N29NF}TknULP@@$m_;mtFWqE
z&a$0q>vd&oqyVeE409?od(RIIY7aw4VlyuOf79*{Wk7r+aI2m^GMSIp;}3pq@*54J
z2cJqIX@ULTVq)rtIQ)LB|M&%?5Q)+i7>`E45!iNkzlFvYuh5c}JO%!3(aBSB;eObo
z_RdbEmPb{$sIds9<$<$h4Hf{odcI#E##U~RM2HG0ZydRJ-R63@oe&XP-5>vl<Jf)4
z7B32;j8=*!{qyv)N63C%3)?eUI9K$k&d>*`f##`FqqG{K<=P*$?Nnfo;?MC%w`Kva
z*((`YnKX@AyOdPqR%{cPaGu!z*OiGoG$ej`{%4%xILR1C98DB^HP7&10KRyte>Qn(
z`5<9W0VMf2M((D`0peEXA(q)4)_v<0tV87buXU@hWB7LWC1(Ef2{#Pe#mAj(Z!%ao
z9yA)$$wIK(VVPUfE_eIlt_U5&NTVfI2SnLLW|X5QDAGX4K$q0GtY)Ml0tB}~2MH%8
zLFpW$nH*75(B}|4YSGRYHOcLBZO=@J1GO}nIbuZ94pWFmXlN+@y!MAO_#{@UF${$)
zD%4J#r#O>%o}Y~SzE@TDFJke+b%9kNI2k<IedZ%mq9kcfN`<iLhs`a)yylPQjH%4Z
z`JywAGh!3XuM|mS!w6i|>Z&haQIB~j0L<9kSi*;vDZgu7n@!V|_7278WuwxA^a^Vx
ziP|ZPt}y>6ZX5G*zN{0#3F~*-sCIR>^=@2|p2Xc45c!X8eSn(=b_f$>jR)kC`dJ5P
z|CQDl-FUH`=b?$BHh$S#dAPOf+3SHh!dU3(t!+Hp4^=USQzeslSaLL-&oALS{MC6K
zkZGQY>BParucL-%6gkk^6+IpKR5DWXx#RmbuOsTE?~$>wp8#mUf+k~8ok-Zp0cuVl
z+ao>+^d*)obwD}ppD%(cQRHHU6qXY%h681nB;vc&?}s58Up}{K$|AlK78WM0vFBd>
z!Bx*8XXZk6I%mwgCg7k>>ihKT62!XOOl)fA66a&yFQ~@SRjo9g*8zY{y>7p<yIv1J
zyFL3m)YdHgP}Xc$r%o|-b|B{7lK1dioAeT7lItI-z_877fdpf{6<WFeAR{~B=Hvtz
z4k&m)2;#F~?nH4g879^pO3+VA6x6E0@&AHwncZwaydTN1=1@L1x++hqEeN(MeN!Z&
zllMa$+e<LF+blslwzg1WmH55t114h&gdz5q$6=@SVhDEdOf-9LLo%WHWyaMU!n|Ys
zFBr$i*;QKxV8Ythq{Ee3PEpPQAW$n_XdHd$_9QPJEFy)CfHQS~prL)h4m>Q^*ZFT5
z=>aeP%gZZ77-kJXW;+y3GOF)x%XI%A!ce58Jq~LOjU@P&w8Mu-*XiW9<fSO{v1Ip@
z|9*iMge1Q*SFuhtHlMzzrx7o&to&VDwXn1ewa9RBv=2tc!pta7pQD#TiA;QuUN3W4
z-b0v-D?S-vW_!H92FlGL%-{H=B5;2-S<D~76j<u!TgAKI4T&38x%ZK8A-fByo*1Ki
z#fdH|x6#KP=FPVh0xhjsGP%D0@@R=hm64YT?n9<9w>~Bq{>-K9Uj&`L^$QyGvOLs%
z>?lAMETECQI`e!3_8g0dTuvey!=q0qsHx}Dz?EPS4E|k`zL=m9@moi))hTXkS&T0_
zSD?@rB-iQ}{!L?+_4|JaI<8I43Lbx@s)$NN%t}2&I$z(((wD1$yK8^B7Jo$xydPd~
z68@x*ET=t?u)Irt-t|mR$ZlM^xoRB<{SOCi{DVsSCR_4{MmJ%y(lXU+@i&tfb|9(@
z&Gp>=X62LDK$t4>S6CWNA*uBKeWzjD$Yt1t#@$#_<t<7c6UO;NZik7w=M^G>5?+|w
z8kuTsmDBT`{XDA_^^u#<Myqo!^0cTlQAj83c%{v47vs7y*@B*q!46zo-&+#{u}Dwy
zjlEDoDK5b3wTYevnDwz7tnj$SyaNG~xySrC3d56wc*(BZ%1Ex|Nf&;<E>W>*T-;S(
zbeP#cwZeZBAg!yf|D|miR3KZSEyc{3J`q~y1aK@&og*F2BLaG)3M8#bsRwY}rY_Hi
zE|W5!fA1j#d5O5bNTt`OQDXQnAGE4TGI0c%I)1M7|7_e(-m^^FRkeJqY~0zK3C**3
zZN5*~#ADH<Z6Io-siMMOo_WGYIUT`ARkC4?<SOyC`|)-Vo|8XX#|2u<vHJIG<nW8(
z*ZLo)+_&R1Kz3k{aHW<emU<egNk80e?Gu-1f%f}3xNY~!2yn@c-z|VDVPF6(H{U!Q
z@&ptVxTpH}sD1n1y-s5jBhUu#%5g08!(y*gK;Vrw5dd#Tb_L~=EPe}EXM*L2(k}3j
z;56&UuD3Y49?i?F<zmXo*#qS+T@LxGp7Kc++RT`*<H!*m5b=ffq}I!s-(H~$9M0uA
zXt|p<k(|UjOjczFC<96C=0bR1ytKI#a6Oh=7*Q2mt+KSOH#*&ec@|jN*|jQ{aQaUx
z{K{LC6P<oJW6N((p}Vu+?$CzhB>elvL5TvFmZo+MsfD3NPn7-L=u}D?qzy<h#jcop
z6*a{g$`do4c^uDHh~fNkege#HulFRQBf6&81t{uVk_LpMNE-5@S)a!fBmzgBh_ot^
zFl}%Dkbe1HV%uM^ZO(D>tpi@U1BxvG!p?-=P)E3vH$)fRuU=AoVoI9r-#T`5IkpxD
zM~c6tQz9Nsu~gB9ch>vF=ASxHTJ+9o5NcFtIHcI7(6EU;3IHD0+*6}~m-+3>YDZW_
zWeP%2xsHUq41mD{X;LX^g=OxTyr{u~4hAvaV1W;gsKq}4v3n0z_qjgA2{q&rCG{5e
zS89|(qCfh#a(8g(=L(#xRGP+~^G?@VP6S1b+LgoGC)Q7&T!5)a4y2&9<#p{o&$@e3
z5#6*-7I2lEkK_M3!TM>seP-bsNbaB0l3@1NgkjaNy^bWp1lj_$F<P9~wz|K5e6Iax
z)0mcU1XflFil(WJUznV4P{9a`$=b=KsB&`84h5gfMV?OS8&VYMiWnuCWJ=(YN|b6n
z2L}hK!Ku7Vv@9%8AUsvLdBnA9)BbH;po5fsV2izX+FwnTkU0M&X**c0K}_#j`lFOp
z!t!Nc3X0*Kx;$<S=2WKz*T2thpm*pcinF@LX0Um0<&w`92@`k^?1{;ypb1AKx%`w6
z-Mv>beP09ZB(|k$lED?@U%@0P%K42-=#$Iv2NT3vbbTF;JtMGT0xXmEpGj#g+ptfj
zJD3p`8FNC1l+6XrJ>jb6F-xS)`pRlEK;i0C;4pkTxmA(i!2G-|a9WTr)?g=(y9<9f
zM2fYP_zO2LI`&0CPf{on?R5)fa~r*ycPf%zFU<+|qW3$Mi$E$%=xl1K@?5i0M}9BG
zr0Yt^_B#=d`DwYu;9)P{h}a6np=2wc3$0H19K~^~l&LygCF$5-MP}2;REU<cstOf+
zi-dh?y)jC|_W9#f-s2B?!C_zE9sYAN6dSlDJPixk+1Z_(oi%hd53Q|z+;ku$x&#As
zqs${g`YbJg6?0E+&H)qnF2uI&#8o%(wh=G>6wZ2_&A|GxL}vUyQ+!Q?ZerO)+M0?(
zTC<GwLlW*rWzkC_EZA5&BfFYV9y2DQA~zVPs8hO>GmnV82xojYL)-5)bmZYGR7lJB
zYoZ;+yvZ9jh#OafK9S0_t8dWa1O9!c!!j{uyKZtrONR`RVihN?^KgOi&NoNo$`1{0
zqgz{S>+1#p@<v!c2;*TlKHzhQzkJ%q18AD}!>(NUQY8g3yXT2mEOHB}9NEUqnWM0#
zuG;p)tGrztQH!wA$5^E1@2RJ!Mr?)xYQ-8famZC$T@sb<gQ=Ai37|DWYVN&w0!On?
z%m3gCTXx#HrbbI+-0#dwsF#xT-IFx6O*x1f+I_+efR~aRkl^jJ0Z6Vspz)Iz(k0nk
zSi#J}GQDHZE4}#ZQT*zfewjWC@aW`PL|%;Q9o%KGEFcc)eIFxh`6i(JcVnYu#nq_k
z)+ZP+nuIB(lNYb%&`%g@S`L>Y$E*)~J^rJ;6joU<Gm{bPe2Hz@2CbY2x&*h)=!>|=
z9V}u<s7pr8((px5mScwkM5<M1(2uPqh{h`iZ4@;>pS=L(soUYn3Fiq0j-**=a&<$$
zyKa`DVt7tjX+%h}?6M-RyN}^3);6fIThY69WU%jJ^eb5&8W-Do#>AlC^%aw`z!%W%
zS%-3w*X85P#R+%3R&A<nix37fDn0k;XWTS0eFG+8=RKcl#j=Hii@K?m9@vP8*P}B5
zp_)dN;k{)X4CM#sv@1`r0G43fGuiHI*aDoJ*Vkzrd~n3FT+r`rcNl6C4@jrd<?)Ix
zW-7xZLq=YrXWQH{F#fBFZ|GO!XJSpJwQ9D&=N54%`2RxqCDr2lgw|a`+0e;x+~#sv
z+KS+$)g=+|x$q)A^xc=4V-nd~d2vk<UI+L{JkRSJ+hn+ie)2lNe&3iD^O2ICxPpgy
zn-u(ulb-W{IGIl6&~9O|01z*I3J$=rAv<v-WHjv9EZsh*2`O03+53f@Z{}M&p-~pR
zY}-d7K73CuCokGSj?l-s4Q4bc9!^oy7X+r6&E0pu4X<;JYP2R_4NQ~klh%m#IK+M{
zpp>no=&8qMEi|u-nt6GS5RP|han%e`;)gu!1-+S3r3D=S9GJoyJ6lBca4g=Zs*p#i
zK3F`#@LV}-XbH&O&GnsR%Q-+dd1N-N7~F!fbqI(1y*&pGtn}QeB$+Bh3q}Sl(q$XR
z`nE8cf7i8Twq9+$a0OlxzGeMivH^yYR7w;%o1}AW;Xlxkg4GxC66`peF_bpeaXbdf
zt4sc;SK?&L|989aPvmw~=7<O82@p4Qbp<!48P^UBA27f6B};>FS8BH{UL)mn1x3$8
zC|{_=12)h6H+OlT$RGZj4#mQ>zu7k+ohM-6)WF5pB$L3dt80}P5`Dk?J_)Nh!W7r_
zev4V(4v5SsRcOmvqJd~3E}0w{^0-hQwEjQrddv`d%Ej-z2=6iz%^9~Df7a+^rzr6@
zG%2N75oNYy#YdDsimEL%wWc$07r$782lvFi=$M9GKLt^`{5}|<(wp}Ly{8yVQ>0RL
zRVRCPH^*rHp1Zq`k;}MKc%*RFd7~8P;6NEeIR_BrO(``9>Y@hq;R=~jSy`Y^ud`^i
zx;l{~oWCh9qO`}SA57kOwbJu~3zS6;IYfmedi;A=+ib;J>xU<AM1aTHpd08t!$q{5
zm*}7V$ELteb#_<2_%DJ|l4Rmk*pn&!cye0#dYwN2zIS6oBssYeBV4q!lpe5U33Xqd
zHM+q0IdHEXJ458JFL|HNC0+LKMKJ9?WzKE1xrQ_3`{hhks{qb;EU}0GS{yeI&bqQi
z?k+yO05N|+${kxAL}>n=8J#|@&Yb>Y{qeK+v=!`8@O)Y-Iy4gkM&u2am|R{C@4VhJ
zSp4li4#GX}yrxG-j~01dEZg8otgo4&RUyK7vN;`v0l(bL!zyiTjLFIxdgvw;H&Cpl
zjg1plD7BeVy0m5H5s>!z%dG5zLko7lyA13qUA%bZxXqgTu)eK+>PFfUgV-Hn#>50S
z<h;)Yt@(Z`5a`X3?i%TFf9Lf`SPRQzxV-$@&>fYAtvrSEFyH(ifL9=In$wG74B6uC
z|6Y}5bDCS`zSkv5vL=jdu_>k)l!$bM!UN09bt;i;udy{B^-CbEwXb`e7G^~*s~h8s
zHxW~_vySglO^0Zq^RO@}l&Ljw4ftl3>Qt&zq$TQW#QtW({*YC!Nr!?r+@4tg8g*K9
zaCUMUyt|_!WH;Qu9Ld`aXMaV0>N2K+@T`qI!mbSEAa^;qaUBYAt$Y5>?nwsy$c>MV
zgk!kfI(NR)A`FR;@eC}(fS*ggo_hNFjCW>b>t-B>rVSXjyV2|cnE}?l?17JVup70W
zsFd-evV#;rPD;Ga4{f9?DH9+*`ZS$j1rIF3n*drfcayAvr&zz5OH^=b{t8zwXIA)o
zV{>=k#yFR0G1{;O{PObo<(;sj>(h;Gh|_y2!QS1EEQ^=>qv*m|;Zn=|%c#YuF*<I(
zMj;^(h^6>m?et5(jKij1_O&%Zx%3<-aY6jnfJQ@#Qd>-|xyKzUZHt)3H%}+9RMGcb
zCm(`Y${YEs;-4<W?>x;3A!~oA>dyKy6(bL>;3lw)EN(AQ7L;;YS&L%~cev5f*N~N{
z!Zgz~#dJff(Omx0|3*WBlcubgR&Eek---FMk(<4<5^kSHK)^A6^|u)7s-}A!=+${x
zV(HmmRTG_o&9NO<+U0aP5ui&GIBc)(xAgdv{MsSeexT`o0oCG+z#IF%S`S`cw5rZO
zNR0*y)}k`>#{R~kc|;sdq#7}B-cE(vbalMpECO&Jxtm7Q`Qjs8zV}`ieG~Fbf48<|
zm?c_P_JXl$t~ScVMgqhSruesjhMd~s&?)v4T_&A<eGl_#?sG$z^n*sBH~%Kpww6ER
z$SXBoabKh>w1K-lY7l@hmIfLApVj8lrD#_+5Kdw2inghhTvf$Ww;Wr|oIU|eJ3d`;
zGcTbNAtk~IB(X@qhq<jyFsLA;{xPDi{@auQaf%{n*z`PsR)r~i*PnFm0ox<))t@wS
zG!S-Ae88Z;$tUClQlg`i`cQnEKuhl9ps3hJ`ZzBQ%QR-ANuT@gfFND|oF0RI=Y`*R
zfB4?G)YtcesfdfiCrg&Y?3ANNRC1pu>Io=AMxLX3RFNd&=EvY^uKpzI^`7Xn^J48x
z#h!_Ni{uVZ@em1F6$LgEk&)YV&xbQlSk_P|{%;UctZZ;V%ZzHPOdIZeMxNDo(@4_|
zD1$h#x<sQ-PO;6G^S63kaz#f^_(3}YRccq<Mfk5f$VOgg?d@D)3(k1%FRXarH@uz^
z!ABQ|99qLR7a)Tl7_%honfXJtuAy~u`i#vkyHdpM$hoHTzOY#BFa;`hjW;9yW~=?*
zG?fpvz~lA!6oZ_&Y$Q2Q7-!*Xmut7O1qCd_mZqlf{!6yq3V2r8@_Bxr=Xay<dx=>B
zukF7@PqQ#Z_1d(=eitvN-uZsCrSF@f(E;%e{GZ0_0X(T<TmSMhlBu?3QQIMzDUg%4
zH4b>BGY)+H0(iSmJ+d8YrCJvD9@WH;mB!5re-mnm$NdL^rhEvS=90=HnqS;Q*xU^l
ze?;tfM(%kg_Upebej9kpgG^vu90lIybKD*-sE#v`&z}FWp&*?-_E0Z$(I(nnGOLoJ
zg#SW~CRvpd0>M>m{}wBsn2Bfk>Ym)Pvk&D%iOIG|K&JSqtaW#07`}iruCHC)Z{7W1
z`={1KYKRlRP_T)KuuL)zDQaNT2%A`v<zKQTm#}o0dd(NVmOrISLbPbqn0HuuCfj?<
zfU!$unu}9#;2Vd9l}A==vptuI(6o2%qstgaXpfBD$omvDzw|23xZj1%?o05)r$=BW
zZm%{${znDG6s0d!e-<lw(+Cb$S2rb4R&cUP;W4p11w{l3G_uWneZx&jfz~|IC@a@v
zfjGRq55+Gp+yOlbfN2H8o@*~4zpKj#JLr4c_-^03%1VHM+cBxEa(w}xo`{6$>5-`u
zT|mq4xV_(XzXigR$q{p3JC6KlLe~(GW01<>_CEZW<QS-5aXH(NVc#Kb=;#Peq0&`w
zHG`u<0GwxDkC61g;sON=YjR`O-a9X==DoeI(<Sj&gt2N#hBWduM<0Pw!>2h0w3C|d
zD_3T2Zso<wjM=5TVh(C;G(8OG5t%iZ`UF`~@@N7hK=9VI$Bq?Z0YdK^tLD#LBy2he
zWfP&GlQ!;o{gt0oMroI$kVe0=e(Gk;pbDJ`&EO*j4B8BNfW{&*dvv*x=^cigmK)2<
zCI&-kY)n?3;aS}j+*w;Xo>XS9^Ye);US_ka^aS(_OBYp*>eR=h332o9E?Kd+D!`#q
z0^2m!($x7^_UV`T$()$uD>*4KKINyRvQI%s;St0X9FWnx_1gS`lXQM@l)qc62mM3W
zB?{wC0u~-tVV{}qJnrQiQ>LnceVGb{+F`qU1O}s8SV)F}U1WpD<h-bfV#UY&pxZXF
zQqOF~T=eFij^VLW?(?)%Dr9A!RJlsD0iEI+!XB8ZSUnD`7n=^@nia~GOJ?4-nUp9@
zi?^T;34glD0?Zf2)-b>!rb#Yde+$=%vBU7Ih!Cy4_4<jFvTT5mNt|j&RgV7W^e)m-
z)DqK>LGRbjqL|$*y#{3p^oyJA4<UXV&HlG42QxV|KlsL&-AmLFn5C_2gbk<|)2U@w
z8;9KEAQB*9Jnf1>V9N14+oH<-;dG=l<68Y@zBfe!ZQ`I+bp3aQYw~<eSUwYivA|a7
z(ipSpoiC_uD6gD^D?`Oal4NOAH8-icsF$Z6k_hogbNTL&21-TT^1U}Y+iI8CSQ=G#
zx!YX;z)f`IjF@+K3P@6HxdsQuO>$+M(0o{OphC+^zy3tFBP@*(uO!d)ey*C$&sL%J
zak8H)N$iRg&u+v@$5-DbE(QLNKq#ym3YkgWI6%NypaN3tVBn%`7U-GWb|i>2KxIeL
z;&pK~APqv(E1Fxrx?BGWc*x{3^0Hi{prLog@S?Q@iT5Or*3fYuZ|iF9D67Uk0MRGo
zS~T|UFLL~W_~pJ<=hZm-gZ&lBbiSaPWy3o~A(@xmpDt}73m~OKkF^qRP7NY_>PiR4
z4qJMF4dhcDhZh8#xSju{AW~)K=bHxxrUMD++zBiZ`L4Ndw`XK$TSkz>!-Ushz_mkl
z`#se!ql3jR?jv<vdh$yn5g1z!ebe}yqS}_)r1F$}F8iyV(IefG7U5|BPh;rM;$Zl2
zX`~<*(7rawq`W3`{un7Ue{>TM_gMc>BM&oO*g%Ryt=b>YpTx8X!&Ao7&L(s@T^LOy
zUqDqJfP)3p(KWu~loeJP`T?bH?}}I-@IVQF`X2ic-6k#U`A5yB<=zXIUY_z%N*yk4
zDoqkC4GLZOvu}p9>*Dvu`nIvVJ0t+0da(lue>kxgn}PKqX*|-mIBZ$Xo+op|UD1WQ
zGg`*|gy5%%?2aDsE<atpjzUN4DUZPg?uv>bNjc)4eff#R8rU358G|KZ#y#OxC_td9
zY$mVN*m3^+z1KtHb%^O|Ve#_8OQku%ac?N8L{srORgQh?IDS84kzoawaLeb$Im!gK
z0=4-NhuV>~HKE#<gB9Rw#RXN%mtaz0CwBrr@k#;>8F{4|j`OGP<=V^$5)qulw*M5%
zOo?1x55hZdk|}_5aWwkZ$KjQ?6u>C4b8~sa8JUkI*4hu)w7hQSMu;bsi<*^kLJ@6U
zZ*E|VEZ?+KIoOyKwWP9IOSSEGcP_y3;Aa3C*446Y@w7;arM7}PWQ%}P7=`jhp0b4E
z6U6TCzAJ+3kh|xwWxrX8%tv%t%}fJ}yZRBqjnxr{li*a^Uqj3`-`~k%!}XVxXF>L8
z${r`t)zoc~GFa)bsTdWQKM6%@$YO_mxeG^L93Zz=L~Q|XQu{ZouH4Cn0f(fXwT_7x
zc(rc_gd!%-tg%<Lxvx?{?$a(a!{i`wgzf)J#qhGI>ZkmJUrOn#Yn=*hAM5{gD|u#+
zvcOPA?wn{otJlyOwZe0LF7PL?4@J*$<J-d{z`8dd##~lh`Ni_7LRHUTeQ$3Fxz!hn
zsU`*!;Nwk@1O$rwrd8p!663t$1>jG;%$%Q<#g--9yI206+R?)u{?BDOd1*$QR}KS;
zsqWq$DBSC$zRYM$a6%q<Y2h{oI>{#y{Z9S3<dWO9>F==FE>T{f!<0P4B*g1*z(Et+
z#j289szdqz(JuS^B~3t;LmwLn9|+}TWMxSqQtiZ0srR8kQiF_J2KHWf;zj+47ybi}
z9L6Afg74>F*LigXR72bi8N|_v@cya^=sb!YuXzRxE)ODTG%JEn&|K9-QIp>{IVbbM
z-kM&a$(9OPTam9D@tsd$I>a~?+$=0gUz>l|eCHGu{m+BIMZ)F%A1gy{7}JY<03;n`
zO2}i7qzji}^b`IY(n3b2&Qg&Z=yysXWa*#0ClpznpSrTty3DT3d$OOoftC7t%Qvp0
z=loi%$Pm?;%#Q6xe(2s2BU)hPU1KrBQR_pEQW}%h^73E5XA%+ELeJ;B5&4SCWAgJC
z*b38(e34rOUJ$JUj?BMcYGiw1$8G1aM$(mQjt??cJoBD<tjJYJ?~YOh$wk{H`t3}>
zSZ{sXrgm=$SOJgk^it|LgeRa_6fTe}y6=#m!DTbLJpr+?sg_9={o~^wue2{FOkQIy
z%co8@So8gSX87Klw1?D@B%KSk)>Ov~X5ucQ1%%8&k4O(5;Sc>9MT)<@g$>v$hNt7g
zZPT$L@_ghyfD+DD#8#5rsg>(gvQD$R%!}<^&xr3HI|gJ2cuMBU#vtR&<cU}v#IA4*
z*Zbxbs}Db65eUF*>VjP^Nv}kC1!}Z`XksEr&jQ}ASY7dA%F!>H;_b%z<b>?<x#tbO
zhzIiUfKrTKLy`pXSX@1I*ZrDpjS>re3^4^Ds5aFKNhzu=1O5HD1I~nUlCD++AeVrf
zVsWYouHj_PvRr~q)sBL9A&j2{#OxP-EMMeaTS2|2*RJzVn+^#iMFL5tv1_*>kdKkI
z2y-w_y`K2rsl<0k7*GBoJSh`&els#QQQD{-BP!7<(T2ppw{ZM4J7VH15yr^Py8Zls
z5mGqA{s77zMkZs^e|0EP!+o4+=|w-U@Qy(-dg6ajkrk;+D92o~(66<#REl<ghPkyR
z4qP7LK-OF2zbIzWqfpsvW!Wj0MlZI-IF!i<Vt74*><oJvf}s5czLQQoW}fE}b2jqL
z#&oJ?4L79KjYFlc0X~6IfA#>Wv&}AwE?)z2^5!`3m9k;x4`@$*cix6agPDC+BlO(Y
zaIJF`TwOLf#piXjY&XgKmFRYam2K)`s<tN-k4?EPU+Xxz(U;es6R&pbuP62XDvQ9x
zCj!1ynkae?OOo|9#JqvnW8kA>w96p=$fWkC<zc#Gn<{EJX2k#dQWjaDT`Uxd=c9^M
zR%6q88BoHj8RgM2W$h^rSxVowaJJ1f>3OpYcuA(XR(m@Vx<-#4f`NCQ?^r*^k5pCF
zxI7P~6Ww-LjQy8NA~GW8`sbXb$~ez1tkinjP14*}(tFp)JHX1l<8|91@d6*c!;kdx
zrVA@G)5!dGimOP_uB%=InKY7Zxj859I;vk4<^G2`Uq;dI=~2#%+2TKc28Sp3|B?kp
zkd`7S(!@}^B5?f&!aX3c_I`dgfzD6ihgcVCrP|I5(BJrrY(IGq1>*pP!g<ifPL7$r
zhlow%@Zu5S&TFi<|H!<E-k>G)R+grzeH=xk(|fJ&2Kpl0?2uP(+q@yvz*k#DP1o4C
z*BRfmt)KQILCEwGjR&kw{2bmh43jA~gWN1Eh=z9N;3!Z=<BtC%WQgXce=8Y4lPlg0
zSuysHJv}XB%%uYD3>maMTr9-TDK0qVa=F?fMgph{<j%`~cHb=eOgZ?Tmu;GQ!;%Sc
zETGm0jsY5)`VNM=XqTc`WgHyKvW<03A;$=<HSsB7)&}vmEov1yC&P9z)G8lJx9V!_
zC}hl|Fbqx2zws@z*SEHY^!gm|lDKYkdS_d?cJBxtgqZ5GYOdp0Dol>e)fnL^uCv3w
zn|eA-eH*dWLNTnkJ}=AfJ2%7YxPBy44Aoz<U-Z%&nGC!>9`ud@3nNG*R#%?p5@_ZT
z+p2=g8w9AQT#M!+1eB`AoQURI8~UEI*`2Ncs!CT(dXGo+DufwaqO|iQze`OjlcZ+P
zNr@*WYg@uH>#gc=B4GTUhSZXypN0W4)Qn3?adFX=5B9r$HtY#VwyGYUfxWOFdKw)1
zNwzFlncrHf-rfK8xk@2n??(CFf458%r;2%WtLiK=wy8$J6}Q(Dv>eG)Xb{0oU9Ukv
zTu(#uSvR$qbym-Q`UrbL?#6NW=(7-qxcJ1=6Cyx_Vh;DJDxdyx=?5Ip`rSzh2-9pl
z^_vVvMErfy(~bRt!(ZQdmp&f8nq2ln=*C@H7{N5Dk`qNGZqsqfgR6y4Uq>=(j)P{c
zBf2=3qd?j<SJUg~1<g`K4}cWS^)VK3HPc2)5ETKZ!!(BQcQe<Fsgg|&JI@k*?n*=$
zvVa;fZrbZLFJ(J5696*@D#dk;^cvHcyD;;hoZQ^O$L((IWM|@Dga59mW50uTZjZKG
z15N^`GFPvzZKT@__kdpB>A;6ZidR*Ulm8B4dOrk}tkZqZr&MGq@s64YFau6TPz);0
zVOca)4Q3|9-`FH{3bM_bWs4VyjZ83Q{CZu>x~9*I6u8YD%75+?lQC#|xjo>(@+j~F
z8!oZwTkX@3W4lpSdZk$V7+1@4ySyX?iHcR)swL;wpM)()O(TNWER7hlG$A~L$syn_
znhD4}(#kK9na=_!mSy;@m_AGg&0kFl4Vhe_jc?(i@AlhF5M*dLRTdsu<~&P3_vCL+
zybraJ%qMgY9Ds_GX{crYR<;!Jz6$yO$P9wu>sJ$0PE=}7ytn$tyXAMQ!VDAXN+y`#
z#teK2?vVoF>W8a<@fHL^)7D&QX0{Rpz+0l*B`VHcP()FrXaZ*V!~SQT)v#{k%8SkO
zfb>2}NyYfw+*7lLl%zo%@MVIJeOP8DB7HKaGwIyXpQt_NLcQ#PKFe;wnGVOmbS!aL
zMBpVmOqk86eH@*=b8xA$1w|<c4E}%njb4w_uiooXKZ?|Lj4+yP2e7XPt`)5&`i+~&
zSo#cs^O@&;&d+Hgsm8g7>Xost%qP-F2X#|tGXbIC@X1cMXc1OCnHof|sDs9zz0oAJ
zz0S1EEyWZo?vcp|Ko!DI*~rv%$AtJW(zhoz2{UZjWZ$p?>NfvVW=FqQ_kWG}7kJKj
zl)PU3SE+l9!>&e^bxf1V$BnN0!Q|{!)pJiTDb_Q8{suISEb-S|-+FHELEjz;JDrZ{
z<xa`-H`)d&Yda$nuw!L^h`;1aReo)&YZ`RIbM^5X1$OoPZx|=1<Q$|JGv7Oy(q*Hr
z#Rt5d%TgLbH(f)k$=-4uQ?w8-ntEQ330AH%27#<-Wu+1jYh4f|!$F{1Ue-DCFqoYB
z1UzCh|K>jyE2Im21b1M<N!4zebCGLStW;O#Cg`?NATr40*dvt+<WCEDngodU54R7F
zg|ODsc*w0q5aq7(ygL$-2BZ3}8vtz=<n1MTl4<KwiXne!CR~qk%C)C2FFAokx?-DR
zvDm=vF@Hio)dUa3B{ZAgvA)%!4M)K&n&fARI#?$8og}Dro-OX^!Pm#+#4bOMVSlw>
z2>MS&rEDhhxd&amLnJRjd5%7pOGK~`CwA-#Vf)@GMw$~c`hXx|gF3v%L5xM@u^vNj
zuG~kgcpdY-OZMMSegjB92|KsnO-QtBl0mI8^#V?uTf(q}HPZGqDqOxm^yxu9VQ9qr
zO+;I4MXHp3`S4JPW!(K8ib%erv2k(JCN#L`1sx?IUb9P?Cr`aPbn!P`3g6=Q_lt(Z
zjROoPKg7W<*cZDD`Oa$&oIQW{_nRih6psBG{O(Bg5C(DfCNXsN__}Wogber(VqPuI
zj~?5-yE&hZ<rMamiM=lOWOTlMmB*C{B+_CE|6y!cvb0vJUlD|D*jlIBu^@9ClDr^=
zGyo7{Z4&E}`Kqhs3>|nzjVKDy?T9N9*U0%y$k|o$JD~W^P?E4ON+hPE4S3=xmt6_Q
z(3IlWO}KaR5ujldm9ljylom#~-g|$*pBM)&(nbW~70)GjKvRHq?&4ugO8=CcC_J0r
z<Nzz+;T}kORG)qV4B4xlUYSd&yeXO|=jTp7opPjQ+5`WAt{Wv^Cj1I|&a%kc#b03e
z7ykKg=lL%c?`7W%j8#ot6Z8HEzMgOcfXc)?1hqA<-$w{DM8b4T^!{2V{US<pYincW
zVYIij6;3EY^r?@`62)T|O`#tIXgB$|qHq*<cJ=@SzI&dd-y2^v;m1FTL-i3=4bV8v
zusMqm7Q{m9E+W^FzDYiB&ZtLld|=<dP`b{AgxzvAx+-pB)mi42>37RaTX1mW3B|-J
zRi67wj@{Eg8pDnipMnoDGXLb?^OP!yg<Gvcl~^plfMmFDAaI)pHLx@=wE#GQ8bPw(
z_BljNylv!`gx~vVUraAglQ}iRmd{4p`j8ZhF*eT28WeWF4==JpnH6m@FQ<eY6U?e<
zSo(oF3hD3Gz+RJoH~pt_bPp4cDAZ&@;ep-)w`mIzX|!}uDDhJl=1iVnoJ3*?r!(%t
z_gnz~RI!qc5v~Ft4@O4hNb~Xb*x(sPMyhIRfFbemDcgi0%8Y(4B%o!gSneJW18JQH
zhhkM63_GSRcuF?E*LS@lNR0RQlLEM!#bft=%DRL3H7lUQ*5_T)plrt4(@ykxi$?A1
z^8FzTzK16II2ygMuSmeTwc#hh6t-$GdhmHpC9P-<RP+5H@xOvAyxqSw5=6ca){2EC
z1#Rt<q5LLv|K3N61;#jPW+}bjdKsQD>}-lUM&}ZhR^gEif<Bb9j<_57&Cn=m`mGFj
z8qM<*&)c|ch*|oZCa%!*-N^}`_`Ps;Ij5&P^hiT5A};GD`lk8x70<T11`HfKBR_Kx
z->n8bWK3oA+iL%AQpF;|@cn=360mtP-CU3=kXl3YV_D}p^Nwj7SR47Bg_3aY76m=a
z-i|d2wxt`I>ZQZ#@GfRxtAQ$c);}Cix!mxJ=R%ay`Mho}$Zvf@$7uI}sjbcTT#D4T
zTnA2nn1ei4V5@xuoqdbiS&@;rpv)}3$XrP4I(NIjaplJE7dL(=kvU=Fa;ek9AF_Zv
zt8!Mz;lt0@E_>2Q8B&ZJAF|<O90}~aSF&$dH>aoe1cZbcxtUavn%ZgLP;z~nnCXld
zWFRqq=`}ewS5KG5zzMJ(0Og}##VX)W-_&vXv~VvpY603N0exwQ$|}!vZ9@?W0@o+x
z_TV$|TAR^SyR1}tV*fPY*|(J}Bqp9C-(f4#56(q6^U(XKdc>y))F_qWLllV*5V^CX
zJ0It{;<5|5g27J(h}~dZwP6$>rQWPgwmw6!f^08OpUC^oowkC{3xYP0ovr*NI=so}
zXlT5D5p1MNA!d2od<-;a36J0UVO%^M5!7}%B?m9F1xsKRyQHP33kcfd$f2fcM!GOr
zhzPZGXSi%^1~qcaIJozzWTrW2NlOsMCd@-o)cbS+zRxS<mL~6cOtWGTGLe3181KbZ
zqRG_gQ7F;TC0Dy(>KuSSb6dXh-LT6iHW^HF<tN>;LUjYh27)dx77@i<7xqzM9Hp({
zP#XLwoD(-S0R#S_WA42+_b^GK38-X=g}$p_)6CX1L9Ro#p{2YWXH=M1KrHJFQ5@1k
zz-OW~Y{{G<QI)RKr!KK%8|mu#ogqTwK4+2NOY~pajby=f-2wGGhlow<pVnRpru2hX
zzh~-?;qLX!?l^Y`lx`b1k9ZSYEygRM{DQXoy={8-62T7R-(I%`4xY1Xq~6g{IfYB9
z23)@5ZgIe#TyiopII&iUO31tQ_z();k>D{{<wAdnP>r|b@_tvIKJ$xI7z8)(82jT`
z>jKRYmIfF#PGbix#awq=cmEvrb$h7VqM)i-xdyi*ET+%X2yO+wI&!ZngSIt8V3;?o
zE!5kXyd`rnOS|&DrsaLUi@MBFr<eMBeP}tYlue%id7_md{s>63`r>GF#>ryWnQ!wq
zT8Y89Yt!?id%=H#_BvDEoy#U|gw{Fd*S^n?N}{x)*!QCg`sy_)Hq%Q$26b;9a@?fq
zkq}p6?1=Z4M=x2oLZT7qMCkFyGys*e@uzr4NKb3nZ=O*@)bVav)03ae<O+L!xgz)7
zd2cXfKi&U6_~<kKBQ@s=>msKJeEDVnWc1wOECS@5G?-H}^Gyb?{X%|vlueCL1WFgr
z6&P2}o!<!P8G!Vwsnt^?)7vs-fZnlkZR^!*5z>|7mD8)$5sLY9&%f)y=87~q&1e5f
zJ7U|dTDDEQ!De6YS1iO08b+&t-gFjyep{)Z=f^v!+Kz%Cl<DO4Mqae<vg#ZDtww>8
zHj{7PHtZNJLCfXT8<w3L{P(@#M|B0oGa-~2#2#f!*We|76RWboe#imbHf(nVWgJ&o
z{|7m7s^SN8)*e+w1>BK`n()~$8?L?&d#uxC8$URRnsoCkFQ@FHfpU;@jel4bU%JP(
zfIt#9mqBHcDOMVTTFb-Jv7>Wqhf1aEcrbwNb~*i+Dx1u|?Umac8#velbQ~rj-;2@B
z;LFLCmDsB8eu{FNp0eNV^i5hTxQaQldb5l}bK+xTggJTmUqTUTpclS9_eS5Fxx&Nc
zw12x>S;MvGlGpY)r*7uT3DlJM&+qwed*=oRTO)MK?X=l|I4brISHus3$T@r=mJR<{
zBf$jdHa@d|16{K!_)lN*#1l$uwQ_ozf}oDqScp0xp+mL5K2IT+<!ZX9@(VcuJu^qP
z*{KkqyIY9kwEM&)#gZvlBER?ksA+A{PUrS(<(CTW1BE29@NhNnt9_=PmnC-M_z?lk
zq!K0#9_kscEuT|&j_kB7)BRQ7tbo!5se5m^S9YP`K-4d<WOyEM9Fmlkmj6gd1Rz{O
z6Hc*YBo=6fD-sI6t?N?r^dBBTT#yYuG;ga=t~I>ODak#rci*|p{N2-BhNxzM$}Ww<
z)FR;@-0n{IA;A}Dp4Y52shoM%auWKhP~y}?qDU~ZHuONgh%;`~`>`M6?QRa<ip}U7
z7cotSdgTl*9nSiQa8>o>T=}x#QA;E^q8VbTjmdDT)!+v>%hC_;p7t|^zAg*)$6cXI
z8Kpv(O_ctIGq3Pv1B*>uPoY3v{2XD|xpd!dh0erAw?F#!bp%70$Pz1Snint@3`oRh
z^E+@I##emF0y@Pj-N(Pp4&1*7)lFJpOOapq7b)ohbxs4Q_-pt?bkmqzRHelnpY@GD
z&G1V8caVc^ncpahngk7PDdsvkKXr2R`N<_%b3A*B<=uW80s<-VkTm*TT(WyZ4YtdS
zv#C(chQU<eOF(CmE?eJSIq#XV;Igs}`1P162()V?Fe1J3WfMAT)-FdTN94ck&pZP5
z*d&|z5uBu!p36@<2M+i)ag4(rCah=wl?rC*{hmWB?c87&_z$HI___O;xZmJwc%U%n
z(3^<Eb#+}|%f^3i7ZKoiam&lxmJ3AdAO^7Hj<<(6*N!(|q?rgYCov`}NNK0Nq+X^5
zUcA>GE(RRLP5<wPNMwD4tn8;`7-lxV&ci{6BW3X7FfB6+oTm3Mo(_R#k~GdhK)!6T
z0!Wiy<&zID`3+9ZfYkuBctI;^mVsvHupAvYF>py7f}hE+xw(l<I8W{(8Zs2S`VSz{
z52l&cYWX?$<%l2pYG}-t?q(kIM8$I%|5*FZO#4JaN%D)zeY+8#-tY9^WG>5ob8!*{
zs_p4n5o3_e5yVWHC{@hAFCo>iC*~>b^FPioFpw@lPu(UB{RR(b3)N6!Ep@h@#QMhR
zD_A(&ANcE-Ew4e>)=zoUv)*WxO-$l54qHU2!P7@~Pkt{)BRNfRV=7@ZenFHY_wat=
z1uN$q+}zR21dxE6Ln^g=2O3a`aV^)%50|>0nelPvVIrSgEk3|NA{uV|{L|Mna7rNR
z$YZ^HQnPTL6{L-M`9j@>wEjS=5CrvZ4L$p&Z^2?Y%4LZ@-^lBhcXH)d4_h}hsjo{e
zVI(b{)99Zr%cmycU3S>+Pmzj&CPD<eFQ!AXiR=OdZcDftb})(66(rA4B+G(u$W@0d
zFV~QoFDEwg*r_ZfdB)09s^69IaGYV;E)$M|sHV(PGzaxm1yLdyK9GLhYhwfRlxbHu
zx%kNoVsEVFfGT&P&&VfYe#1{sGGebc{)&K+b?Lp~3O%2n-#zBG&$UXYHRigU@W$iG
z2fkWEkQn!bjld4Y8#u7Phsf=)Y25}SAm(72BKpl}g~^Ni&dE8)%$5CtS23T3bA|vF
zYLx!kb^Tp|Qr#*_`7cFAes=L(`6UdwhyH>EC{AMIZK?+vFd9{No`Li2OFuQs1Pdd6
zG9G=e&<tw3Xp%5Rqpwr{Wc3PTtwWL^X5<adFzh)~!-i^xFWA=kK9sAG$U2Wgxg7=6
zQaaB^ag*@+YE*(}VJn#hjgjT_PIGm00RoX5dIjF32C%QS{NLVo5>Bhh+H=|B$;C_Y
zkY4(*0(&7!BoBE&Kk<#vhz)O{&g~DKf=lAq_j*axWWs4J?mxZGr1A*dh9Y$Gp}t~)
zCfylvmmLau*hDTXg(3?ji#E9NVYInys?Ev=Hcjd_A9j#16)lzN&Bq~VAWhVnYm@;l
zXLn^D;<@uXR-bgHAQyaFynwfe*i<1QisINp{mKOT4<SzF%ayG)BX#_g(KMYd?{n8u
z!WpBcTkZ4CO;e;OY3Z1Vt@-HL)XaipGh-oWy8|;`H<fBA185x`fY!K{f%h?+u+{U8
z+c;pC4l^$P@9Ns#6MD;EL!f9z%LCKr8|97B1!=UKjTDhr8Z;b<W+v#o%vhiAQ{NWF
z1kfcNe(=LGr_kfzMKATQzZ^*{^ttnH&ED7>Vi^_;Ot|B9{AS=iR$<|MZsOQhk6R&h
z^obnO&7o_NyL+s^Yd(@h0@^Gp-&2Y8!^2s}saCR7_FxgycLdE!?~1hP2{E7A_6eri
zp0MTCM4g4^b3CNA$jkpAZQ0nl{ou(V&97^IMxep7#hp9CIJ))4KmzJf-f_puz+PtF
zNVo63E<?YLHsqXclEuZvJCTO>yo{+m>=e(8Ugd;lnzI>g)EzURj9zDm#2?+YZ*)8p
z1C$J{pwC)tyovo(D6d8T)rB8Fl9$Y&byg&a%%~KGwb-%D<V$Hq--5l#-}cYn(qHo5
z*xy!qmOd(Sbmn7tFQHeIp(1;V7kgY&3*C&y@K#JU)2;db(whDF<IJUq%Poys^jAvQ
z?ET5ZJQL%~{17D(Sqn!OXB-#8`}o?ooi8=U21Dq5sce!YE~!aS;(lHH`<veA;s_g&
zQapr+@68v{@JHSy&XUczbpf*neXWy61u9v>7Q;?17;S%$u3TMRJ8#!W7>m_q8{2(;
z6&Om8j7^?w5R*|xEpR(FGR2zuz{G1`JPnasKF4r@SZ6u9#6?ze0wbNx1>}aiIf6#`
zy1StOjrR-SejoM9zzMTdv`oJ&Z#@s49djowxsmTeqR7je?S<ND?kOm2$`AjRIzNIZ
zI5^NS&dh;kQCB?=FIMcHFw6E?Jz*;U`gzp3^96b+V*d6m9!vZ&VKbOab^X(_4m<b4
z4{Vta{Z2v9r$2p5zd61*G&XmQe8YQ(Jp|lyua%23jBCS><zZJN;=60jb}zZI$$+<L
z?TVQpm(P@lBU5?w`ebsv0xu2NzHNDa6h)jrTFY!*lRg*e3W;ll)+LiNz48A-GpVRs
zAaXHHki@l-TBDeIdhmyv!8Ys(`#u>IX1I=AO(cRhnp*uQ)x6{DS4js=jv)0AA}-0~
z1q{3?!j<x)5DTt;6l4%$&G9bsyH}m`dLa<cinIXRJmZyuavCTl=qDgM^zE3B=tWN)
zD~FkH0OcQiR-=0~V*XFWCT}Be5!2-Zb7ur{$j>3F@mE(|9l1uZYc-*lE&gs5kPhVR
zwL|3L2D9?mZt~M7&2GQUw=0TQE!cA)Y+IuyHFM>C*d{03uVaJSGM*XixL~jX0RaK(
z3m4LqmxvW`SrXAe<PjbkW0jK>@c7Vdi-0?&t-yV%#Pcc<Zvwme%Jb1(gnf4$k_~_{
zZKfB;z9=@X(xSj)V28X@VyOi7)v#hBL`GEMwgW4kQV+j;@SND$q>PngXGDt+S&X@P
zs|X3r!9J4VFE&j2jy<$m8vE2)eZAYFsi$T%?P?xp+(zHJDUN^OM>g<o0uOf;t4XQ&
z;l-tOj}>mbyw=b+jymM_1ZjogT@4ZgG~)Jwh@^AQk8*tdXpt<WA5%XbS04Y-!w#ZV
zaWcxMRY<G}AI9?zDX4}Bs7tfh#x5^we__!=X|K)=(>aKaKJ#=7tx?mAIRT`QL10V+
z`~O?1>7B>*)}^1q9aoKjq9rZ*tUm^P#tzYD`69yqSutE7YLuQ+(hhFBOia*nDYhbo
z&UUXLhDwD8W#sBdh@jG~x^tAtk5+-{%xAwW(6`>iq~2@dIHZy`>SNE*VRKb(iCNv9
z+vDSxRq!X{$39P{FK_{umGPJJq-|J*2Bt}tlT++2b&WG#JO_KJELGKP-%9Mod~ylm
zzzKAmd4wE??>YL_f2Tz5p<Z$@aXCaUh81Nb$fO52cyoO^>ek-SxTr#qx8)d@{SjI?
zvb)R4e47z7yy3MACsA<p$G}i`{hRAGZkI7ubOFy_aj%GT^!{1%0FQUYbj%a2x_l`q
z^v8c};?<7qH_N;q5nq(pCDogaO@{)GlpE7_(dmb^e16Ui6SxbF>=VmEEcC`NkC;5q
zw=GHE3krz=-DP;W9R<d<K>e#yvyQol;v)rYNF3Vyb^Du0ocGfWv18XQ0bsians=ZU
z-m2Axdv6y#&kTO|xh<#zvtn4dn0NV?tGOM1H*>X2=|;@|qv)&xqU_cv{0m4)cS(20
z2m_3CH;A+#4k<ZwcgKKqcZgCF0+Q050@5*b4@lSf&XqUb4d34HUhi7#na!*xbWGc2
zhpPLQ55~{Uu(oTRGfv(?P%mytV)$H<4Rg99rE5cCMs5%k98t=iTn9XfyP&hhH_oyp
z1W{dV9xs!o4Pj2i<VU();PqQPI1~1t#jIdENwk5N4lkVol+RW=ugR#>$<H9tRHZ4O
z-|M_4@Bf9OxmcvVi=D~^8TidE52Ec^_WOcu9)*aSL~5zHRxoBv;Q9olDdL|ZeADRm
zeO^o>rOIh&Xn*{ymE${@acBax2#D(KBi%!yOUvlv{WV2Gf0{K7TW>PvJZ!wzsmm}*
zs?_L{ha+6aKw&=zfZ)!}N!Ug>Fvc`8<$ves&0JB1or*d+zFfb@3p~S+ELgCgwo^<&
ziEQM;jR+;v@dDNzaJc72H%oSe&e?3-Sq3|S!usd%DdL7hQg(|VZC)KM@EZdS2_2)^
z;RUhX*h3}F`7gYHSEkqnfV|)I0SFHJd#siSJAG3Cv&DkxCw3gB(b{7JHg~<iHp~*F
z!KKt>j13`^s9)R@@=<DC`6NgSyMAJV!;MW^eNut7*O%KCQH`Z4MuDNkGI;KPmZGzt
zXBWTyv9HYk@4TLMPt<7A5Gls9qb5xXd2Xpe;q`0m9v>$i=d>N5NSfiDkm%_7+@io!
znWUUBsz29Vbo2K6AC9erei9v9b8l-S`jo;P_}9HC6^VXrtO^ax&!jmJ#qAhwjsK%{
zLlSMkC~JK+X)p-g2G=xmDN$O~05Q<s?fn_GsYm5yDX6_2b%%=g)O79R{o}p;;{kb_
z+a+uMns1mzTwE(?221FW9DIa(+w_do^xj?Zyy<UR-6@y6=fFRaDCqDEQvqcI1tg)T
zPqcv#1JAF7goI3%+tfNGw{AYtl@#b+Rfte|TvW}+LOzT+|Fs5D&3wMh*tx=zHa|C^
zkwK?^`{wOvi8RMMx1lU{VBWs-a?EV%ycAQe{Hi-LI;yL*Niu&q{w;7GmT5aKNU52j
z#uzte!3h=71kceP0wNM>@9H}Z&4e>Px=FuF_e~>WHdhX7-WDdUmnnhcornjmtRnUZ
z6mNEHNWsj$v!5SkH$N4w_piq=oxkhd2ksgg8n;85&p5@;Jl=TJ)Q1nh=^kFBs=Qpg
zj(y_`A5r5WA7_R2-vx_S+r8~B>C*mo##o2(kdkouaO2Bq<89h#Kj3@ifHf{Vn%@~f
z9`)fPM5ViM?WEYqVHpHH|Fdy}e8(wLE#&&hs^J=;L{ajt`^(SzSjMH7bDwb`6Y{@=
zJ#{JK39_ViMz{UH3BIgAsxoxDR}nT&r!}Q=ZNkDr-@ndAS<N?!%<)H`D6A3FN2AdF
zRx!J)6fd{U(WJivKEk;<D-Rr+4?MkY20Ag#%VnqO>N;Cx^UB(8?~J*ZP1n!%g2d?p
zMjy@@v*)YN55CtpKlZ<QnhCsn_p6GSEIr*8WHnTmQftCG21S3QH7u$${~44#Lx#|{
zn3vZ6Wf(IodiR(eBQ6(wj0JqtY?>Z7{qw?rlWJ6jDxOp4v(@g-9#GJW{y{_2NLS^J
zUK(F*`(uh#aRgUY$nxb8gU}GpHiRXe5cUZ(l6ZPfLisx7z_>FKiK4^c92c7Ci2Nif
z?$_~OJV_YL9LJ__Xo%4BWTkNBm>;eH)V)l8ZmO6+)U|sLR7M#X7+5$4qz(-xZpTc}
zLAE|Vqg8E`rkj4UvWk$rO_nr=zzMqSYL<750pO|kXx?-jf;L~$r<W~1K%s?%U%bg6
zE~3MrKzZP?+X%c{p2yZn;m{G}tT|5sQB=%?7eA+^WB9F95vc@Eqg&_HUMHJ<q&?Ni
zGA@A#$-27YHj(5!!bv5i_aeaoQfUJvh7ofBT*1ML2BRqnkun0}bK>_o&q#OVt8dV$
z_fVcz)FdKu3Nz*~B$>n?S2$dpvCs~QjmL;9iUPqgRaopKlegEs)L4nXRrkKmIlo)G
zm8H4>6ord?z=^8pLmNytSvl&mMO(`+(sLQBZ|g3e16X;ErrvSRVmCRp_g0acZU6fR
z%x5Fx^=vqisQtxQ<yBCSs40O~)trHeDLp3VA3IlOm|Bx(2m<?|i;T{PGfjxTgdX#)
zN@2njnYVg!+=G3uqkiV8ciZEY2tXB5i-`HmVqrc7{d*{T;)*y>hr_Rc!ER7zu%0AU
zF^=_hmJnVCSOhsS2t!d$Qr9XLmUc)}6StYqid1$z`3V#D;TdnL$S+-K#h%I@76^YF
z%Jt2!4Q!4T&H_lm7;a01-d30>%JnTuHS>%k?@FLmHjBB!mkOb(>AX{?*OWLB|EU#4
z%^;Z2+wnB>j`{QCK-BKB;~aJ=5P!+iBnQ(Z<`*x39HL5T2uYo@HG#EBzfeOXMZP#(
zn#(&t&-Q&Y=qq1k#oErDO881cLA_LRuHAiRVUut7Z#MkDm7p_yvZ-k*W^8kj2JR6o
z*;gZC?~0le*`a4CzsFn-5O}kGQZ_^=6%J*e3eU0pB2BipY^}rv9hB~5Q6_pWA!n#b
zYG14=8&y)xYbG9+X?lMDw;`r$O$7<Ki888?qiBBEIMlyc7-G15MSj*M2BUfBj;8%R
zTs!M-`eocM^1|#fDRtX&^%f`-`j@N;tUm-!bD!d7ddSjiK5XSWwT-P#Nd#1=HA;jY
zHaF)DEI&Y$B5QRw|NSFTQ%TZ@Dk4<bm2a|hHjxIfzos=C_p~az7Pcyo9|r;)EX7k9
z?3{lrP<D2AZm*WsHs-4IhlHeS#0%!%T3VICceyyXsiwfv91gQZ=qCFDv~nMDR7%+g
zis?#6UxgxM-iKzzQM`RE#Ic>?<exj5%jWq((006kea2%yZWkm^R9<`2@E=KHDF+6I
zuiO+G6He6}R(W~3$uIDP4aD5nE^;QBtjF9!TjfRZ)Sg-}XCAe&I{9h3f5UhOKZQT-
zgzy`lpUk=!%cp??wac_>yJT;iC49VSuk=+6SLfHxsY*p-;i`@<{kCyXv})mub~4}l
z5Af5|I%K>^hX1};Oc&=#%!D)vnW$lzlH2cNLhN|f!G5rzeesG3hRB&xBjb<@r@2jw
z7Xba;gT)QGsAVCPqPuAmZDM<U^CEY_Zc7d_Yl@|m3a_S_W$mI7*fbT*wFcIwC8`VV
zUsFcnFi=ej_i@EiWb^Z%{_%aCJ+OB)-P|_jyFixb1AJxg4UgWtQm%*k=YvkG`voXb
z>hyjB_^pJ39knzCZyGW*h&ftRRJ7Oi?at0OrG220T6y{}HV(ga{zL<B<3v6OSoEc5
zJwH)<z%gr+nO|z|F{)TD^kik_H*9_JGJWq7FRDmj3cPV0cotb%%~Ri2C)b2X+!ifn
zFOsBPPgjLdDRq!(vcE?&<*e8e_>rZx`i$dkj`{=18SFBZhc^F>_0``DTh^bA!9=?+
zE9NW=yRAJX%%5%YXRbKPkI8J)0AOM)cTFWd3qaPa<?jsJO}}|q6<*PZL^89&#oO*r
z#QpA<)^^5o#UZ{w!v)gB{f-YvyXooSRkfdp6#x0W-9%<wzWR<<_}+|7`BR@$wQ8}c
zq>yQDbvj>7JfF3hMrZ!hsvG`2HL+Aj_ht3kR?^{y%AgBDcOgiHay4+cuvOL;nmcVD
znq(mL>9*Wx>1(}<LD{m+q<9uToRMY#J_<B`bDopT<2Rq(@$jY)ndYRTu`EH7xZJ9v
zm;8rY`Y%oDz%21lm*O*3mJpq&tls~#tJ-OS0w++mdOt$o(g?TeO|rHey}wn-r|Wsi
z-Q%Xtu}bILk?Ym+boM$N4G#}?e2fPd@F#EY1sw(K--<k6&wrb>MUqE=c6YyG=BL-J
zD0oLq&`FwusWxwtQfJ?Mo~4`y?Z)e`?TDW~1<lMrpY{T8^P62yWyOEVF_yO1054#n
zB?}Lah%WpuVBvNfkcFpbzI0%i;Ln1ZIZjOG(vNvQ3;dk~lcH}KX<r50^hg4Bw1hKA
zb6Y_2W3I3=gi%%VtM1GC_xU?1wd)321M28oikje?+n|r$ndhn0d;L^-6<=CBg2fl7
zbJtaO6MmDqTicE51Z!wk#ZP>kw1!>~>GPUdt(^irY_e%q&3USU{97lF_}*t{46vNK
zA3uHrTU+nLs-{V!iy^jrej50~>ME0}&^Zd4L1*o-=bCpm5DqDBFb7k9sURu-jjwvs
zp_#T|RI0;k{6~*EaoqJ?&BZ0m%B2$bKy+g&MbJ&}uYb7E-gnqhSszsnk{7~+rK7%k
zVI6kKN<PfM_>UhBWtlT*7FX7Zdx4?tJk3146(&}krT$f%I7>reDA}Tdel~&~j{LIK
zzmD;ONHviZ-rGx=?2&(9G84dGElXA&CM|XUF9jT9zvl{THuK-0dRz_jy{4Nn$=s-l
zjXm0xO#@U_GvCl+Yy8Xx&7lb5D@9GMV9in*7azAS#01^>6#!J-TxOJfUhwrhi@jc0
z_34gZ67jk>Ui(=xSLJ{*0nD(SBTJ`kv4SD3*R*H(f&Z9(=D)Q9M}Z|oe`LD#GrGau
zuzCc?`f+BBUa|}7xW0N$8?hA21+FBBB?SIOhhGvPV(pNc43y6w12%!fLoRSNGY%vt
z(;JRYx8W^8p_Gi%z^qDi{%40nc<1w<@KDVN7kYiHikOPe;m<cO9CsKTJNpX+BB6dJ
zN6o>XTt(jL`~sJ6sn`-I6vZa~3+Nme+84l75Bys`e@ckXaCk^vqO8!g#FwZP`y0p(
zT@=$jV$QqUHX5aIe~vloLD3j-wQRL6E3t3IEX&g{i~7&^r1tiFL+IGAEQkJr`p=Nr
zRmk19t=Tm<OSo)G`Hh4y0j+bR9D;!TxASh|%9#+A-K)XRFH3y&{uru~OXu`6Bhhj#
z*wJ(-Ig@Gge*Ppkq7W?3WX6jBO!}X`9nF5CWyK})2#Mgl(%Pbc5(B|B<53=h|Hxy?
zFyhRgdhhiBxTVafI+0R~cB^A!_?W!q+ATrP>0_&<t;pg=?ZD9yjjh<mMvrfUQF+_L
z=E%cldguD%Y7KZ%lwrcPzxag)eTyGLEEb64K0LU}u>!y@Yj20p@`QW=TFbUe&s9su
zfbdu3!W0RLz{B_6?|f(=Cx@2Qs*P*~Ko-PE#ntWXmIC{IwKRE~D&Sb|#y25LPa~K%
z2OSAf;fnsHcHUk@U-5_yzqPFX&<<FiBsz$7szAq4R1^scAJN*?V2PzBI(Vck%Iatx
zk#>T;yPYx<W65v$))9u+FH@yy3malHztQEB>*l=SKq+-W52ej@q~<_7Bo{U`H7(Vt
zE{JEulnJ5lx;_pZKnQS(n6%Z6bJetUTj%@t=Re<?8Q2k;&ns)iMFD}D71MNPWy*L3
z8?<egm)h(%$;{GReLFkyfW*W-d`8pxZ-lXkucpfOw+JrZGZ*gP)1bMtEINW;C~q4-
z5{8UQ8B(cb5xU<;F(Te!+hrCuj@{s>vV4xG?9b)ZPFcU{^P-c@+m2F}Q&L$N$?SwA
zq~i)u2&y6mxG*e^UlfCXuvgBp=N4*z-T2Sik#+$8tS5*b2asroywn<v-m1Qw3h!K@
z)+hHOD<51DH_<=`)l{7!j`8AOk((#gKHo4@Z7|_^O7Qi!gs4>gK3W#!>-;C?+lsM#
zzy2->BpsV0()mb6F2Xl`>H$Md0GR4DEGeq$0K|_p58|&xG~oAmxVE|2_#Noma(46U
z-Y9hUW7kJXODBhwj98G6+f<d9W^Y^_Q|7((?f?Co5YE&Bdz3eAMxwXGWM*;|pJ%Wo
zG<k2_jo73XyNn%7lj0?y_u8tZTp?9MQfK9<lt*(Wm6D5JV<=L&k-Sq$n9|1hH9>H%
z$zHPz{E5A8uZ<#k-_<1KmciQh7DKShP0Ef9l;v3C${KYMmcaI5PDD>sLZT30tOwQt
z+1HMN`YJYyOwzpNX75~$@}7eO&YJcsDgZ<)?s+mzZA2Ql{;TbFQPuBS9H7+RxrgXP
zC(m&J;iEX{0cSRf5N*Q5T)Ym|JaRlnY{^f;>B=k!C)^w9q?2z~B?^{!`oFKSVTPQ7
z2V2DH6~|8Zx-y33xM;_H;(<wWUs5FOZG&^nIMDb~!GoQ$%?j4}G@TLywXqTiCN_?=
zvcx}{#CL{}Pc)j(VTVqm?~AUt5R_UJV-ZiVeAUP7PE@0`(>3IoIdM14GW)MIb}&&e
z(zbi<BnY>vrBS!s{UOR&$TIKf`0PA;ZOhFry|`Rc3zncP@oAZ^FbQilWF2_&tMZ!s
z-8x)*Ku@51yS+ar3)f6A=xi7osp;tV?0n=MYZ0wpYnCU_qtN_h>RUCwtt#UEZ(ZVf
zL-IkKKMBZi5y}f-U<~LnaG>Sw7Q1Nc9`IHhXQXfPi?k{=34MgD_SndFifk>^^UTfR
zVBFv9+ObfL8M=@r?<V$A`ubv2R@nn#Q?MAAF_zS@NVzC#$SX*R2Er#tHCU?X2<k^V
zhZU53I)9wV%n;bluGtXbO&lUJ8r;o8X=Tu;zb~V~4dwZJq5C7ScEAQiKSE?wMI1K;
zI4$ej+Pa6PqI}w=+f5Y@8B|rdEDT2#>HI?k(ro+>=U&m|Hvm_I=qufci)Fa}D>>dV
z9pgRHqEyJl1h!!6?oTCQGf+?-ZyJaB<!$KhNc9~Y;n%p06fr@Uu6)Ddz_g2o!o-r9
z#N1_I&8(=1v&5tFeZ*#iRNHTkcktVjj#^TIi=CO)2sCRytWFCY>@*gO7_6onkhZ3r
zwy6XCX2x3Z3L8Aic<)u+h{!xaQAxI@R8C9UrhLPoW%MrJc2ZoFL3V~J-r4SjHx!<0
zIzEtpT3gp9V|OorhmJ3u(+lJZ&;q)SmLcr{k#qY3O^zLd7^i_<kr<um%+l*jG(Gif
z6U~>ys%^-AYQKNTf#qNB*?|w5&y<lEJSTp4SjQT%cJno^fDh77ay=KgJnO9|yv}bJ
ze0^I1yUtxGtuhhAYuPkbw16Oi2;b`y!>1B{pdROJ{X%@X;}Ts*W!SRSV{>rvgkt6e
z2e9Iz1$ZPjsT<x5vpT2@^qE^YVo)_mL76IM@sv#i?8`8V5DK3iRN_{3*y-WF@-{VA
zZq6*~_{r$YR}HgbMoMJ@$|#3I&E81CB@E-B-|ipKdw4M-#s3`QcGm*!8#+~g043X~
zJz}>73?H!pZLK|IOK#rWWPEPy4!?Bw@VM%-@jctZ-WO9#!dh{xIRo>dy;dO{5U|Y>
z6cOqvjU~s5ojASC*%AszQtoQxGpu$FVedcFp_C*=zSg$=kaI>=dL+gvn#B&Uq9O{`
zxl`*(+Z9#LA@E`_FI6L9$>1J<q4Cb`GrXocs3<hQ`y=N#r8obB#uU|&K#)RX5*8t|
zWN%pItV+wNk=fAtM_+$ZMyp!aAR||nT2}AI?pL+yH<D}GW0%WCMyK~GQD0-JF?Abd
zhHRcL2k7dZk@9JL?NV<J_$e-J@inwKt^ZEI-sX#6OKc3MQrp`*F4o)J69SI5SNeQF
z4SytgnA#>u+(wm1=MXS2LJkluyhg!?sW-pzpYZBkFWRnS^O4qpS8SMtiPM&X(c0K1
z6F|8PvEm$`ldWy4N2`8;*8AxOMG7T_DgiqGe}tM%d9dyCR}lY)6pKm<_g{12($;DB
zN2*2RL7bFzTMQ}EuP_kJ%`%J){mcR_G0tJqWyCg;GQ9TdooDWH4wT1XTFw%0%((sR
zUBcxf>mlXyhl{Mj6C5w41%}vxTCtnGV*G{002vZCaJ6DuDewyd_1LYgt$+(|)&Kf6
zKq_eVJPppRd%@S?oEvz9|CAU@olU_!P|gOBYqOuULBOI6_`iN@Y|2Ex+G)oWh?|ZA
z_nB`P&;5TY{BuHQ_PDyb4SYV7y!-!YOWmh~#;j>B$Iu0HCPZr~TcfOfTzETN|AuOo
zD$R$B5Zo7fY%p6wprlV=72`t7P((P*^@Tafz}AFg=zRnNTbDht!%mUZHfC4p^B8*Q
z23oRZ9J>FRLsZ+LFFOmz>^GX%v_C*iL0rro&VvJ3kI52+#C6jK1!B?OrL;t8pCOOi
zZ-I?H-aKE2lvq?WT%p7;?C>_MV{6YQ!KHDC9e(rg<b$KBZ!o#znw9S^r6zFhe!Pxa
zIer^(<{7BuTNsXjzdDFhWvM*x9!sBGJOivwiB<wV8yTm8f$^8|xIhjYfcnxLog$AG
z-Yw>)$@dRR{k0!tS8;xgKgkRbPp24Oy?O<*9tLHR#-`Y48M!*U^-fI*IJG^!fhWMD
ziZZyTzR~~(rQcd1NeHnRyX|C)RLQ!JL{!k=O!F#*(=LOdZ`gHrwpX@@ayS&Vw(*mg
zl-S{VtU&S<`Ukn!$H%8h-{Tx>db~ho09gakuMSXmMt6L=HaBUOTKSN-y9of>sl;#;
z02<Slm$Py6@SasU1tbGefv}mGnNQ0`i5iXaZ^U!)9s}sY=+96ex+I09NM$AuuYsGH
z^&jd3;|*<$qwTMcUhv{e7Pk#go}B+GQ&wPF=y6}d;Ih>$vZ$!Z3uBxbdvBMl91$pg
zO1F`itpO!&0^caIRlXiz#lkd(X3z>FyU3cV4~fvFrBTwpf><VWAoPS;II-ip-Tz5q
zj(fT57+>1D2c#9R_=~mE#*R^*vs-%DrslVKCe$8G8f2A{dIPPX`E?%KVy>?BH0l0g
z;;mZ<0cgXU;&FSg%jsdN36DJFW>vcP_N=~5?#~L{&t4r9w!2ZwTYgtaHeJs59gkN{
zhHHNT34v2Oil35Kpl4j-TJZy9qTJ~FR3&;F^6oqEO!vf8wd9X*Y#NEA3+P&1B~y&N
zKRU9{_g1Gu?U~BBUcDDhbxF2@=4ECATp6B+E^IXo4P)@P-gZqosPu;yf*6EM;jFZH
z`*aS-lS|6ac`y8kN~^k*uywN#mb3xZ30pH}I-N+|9G%QwuPq&VH^*$O%FZnQInzNB
zs3F7aYSUl+wufO;HiJX^9RDds`o;~&rlnQ>`s?BYM19C^`b9dFUCV|-G5-JSn46o!
zMb`X6ecFF_9zOX-gh~Kua$vV*28H^Erk1|?$P9=`aA<L2S$vXCOkcRXin-US(%UBV
zJTtl(NzJK$e7+8k)e1Fc9d+FDnDaWswK2AF(sse8m^sYmUe4V|jjCEM(xHc5HlPaS
zL>+^B`U&ZBD5Q#`6;GP?j8#C8^yUO}-<H~!;`)nulv&kBnqcmiR}zsDVnc~WI^WWM
z<5T2ZDJ6%|s8gU>*cxGKNZHBXH^@4nNxT>-6$I#!Aoa7*#Lv)(Q<JY7rLX+le#z?@
zmi%-M&J2&uSeX!o_{r596|XsbDWPOLCNtOCE;X)kcCHIe3;|Z;vvwjkt<QaH<uk9%
z8$P;w=ME&$@>6*TC?s#k7(wBem$ABbq^ADcZ)o584cZ<?F~gNVQFC1b(ZG%DwFfeV
zSAe8pY^;eD5|nG5P?7d~FT;b&%FZ{PeRlK@*?!Ky3><08pEe4ib6Ck!QPQ3=!p5Z#
zGsY{Eh7>Sd@$(UyJQ%$Yl4FuwMz0d1L;E}Ve$QOtw$W4BT3>rwqTXXzi{1Rk6&DF(
zrfH|@cmLs!f0-~*0rL|_xJ37*fw3{v(UErE{p<W1^B61S<^d0uQs)grt$O3@>-+Gp
zToM6{?AC-q+x+X|r)^`UW9l^Xh=vtb^>>`xoJN!p#pW+a{%$2!eB50>OH`<m=BUic
zm+o6|-LbSCPknx;adQ#A@)5lCd|UJMru_KIsa>N($Gn`sHz<;Y<sMUIkwF`sfrcPv
zZuNy?zGS~y7MuO(x+n`5cCx5<9{qcb6I<Ox0daT*{!$b_gFaCOO&vR{#=Ak+FumqR
zD%18DeQacCqCF7+3s6%|%EA$u*~+E$@gS3-<M}PZ5KPX0ga_E1!RCV&e|HbIS?krc
zl;tRR1rOYswLNd0Oii`}ANu@%rk1s|jel;J`*{&`wNApphG|LRW<-CV!f7ede$Qpn
zIl8<lXm@G&>zFM6;XEDEazQw6<`bR1_CWRBF8j}S_2E)%Ma4)rBSo)j#iX4-*%y^2
ze=ROHSwiU7FGV^OXa*p@S8>3wwCRhokBJ9F@Ff2-DMP8o#Byh#ZnO*;AO8@nr53!4
z!#R5hnyMX$r1KAQdp$f$rFnftSvsd+F1R=8aC|<eI^Vow>z7;Pf{|IeWwas6ph5{j
zY}zD9?PUDsy?Y|481PD8r9c0sM7*&>Sb-_fti^s8CQwkxxA=RCu*y6VOzXNRDvv#M
zcfbM|k8pRa$h1^8z2`HHGm7Ujiv6b<<DqkUX+twu^VN=ZKYxuDM<$HxYIPelHVP-P
zD}(oQCh0D_FmpzKD7_;|wv2(y$K9a<)g77nhF$r%6X%ZcwG%(m@Wk|YAL2Ie)K`ek
z#2Esp&6UFA2eC_~u=Y4)&A3uPH!}3DHY;mo0^%c^TuoDrh9%C?#T4_d0%AR+C`}dy
z$M{%muZfyMv>2)pfB7^88%MiHn4wT_6TFZ=>_r=AqoW`=qZ<)|{_NYEWtYK&KK1?;
ziGeENc%4*x2tx|3>NbJ3AnOqecRECaZne<Zw$^KdDBY2_?)i-Oz8JMG<Q4S*v4=;V
zcE#9Px#ZnDy|j6QEVGRg|K8ON;db)`lUH+k^SA5ir>^Gy$pBOKXWrYQ6;VqNxNKHp
zdjr9vkQ=<>8IrgVZYV6wTl=eHyn7Qn?^KAwt+}~n9KYjPHlq0dub?MmpHN!G$N`FC
zSc!#`XF70Pa{%q;%gq7FQ(jV4I}9bVyCoytYVpkQD#2+?G`xQHUn|hX*6S8Mp%nP9
zt;;bX-+U9P_4m$UQWg<iq+6}|LI?5Asv%d*J^5y3E|;V!`V5lEshL8TCnV8lkc6dq
zGYh*ztXz$3$OKk5K@~~fE&x*m?jB%I=eCZl0?G$;d7--fhyi@(u+1Brbgpz^@5kBn
zTcl<{Kcd|DVYl-}p!53i^vBP-;q6%c!S@=S^Yzi#dHNOv`3e4IAyM`JP@t1C2KVq?
zIdv_9jr@A=TXp#r{Wk`WhrAn@Ci?X8y|W=|OZLz>X-a8BI}M5ZoSi<>1AX2XwD41d
zdXZu++n6ut?;_~OzJ87S79?c{N!`pKIe3w5@CESx7882$UN(}3D7>cpYkwYM5C<aT
zaDmvdS8t*rjKECs#1Np~5$L~<-@_#})-*CP#<y`74+~C!`*<lN#rZ^OCwQi*wz<RN
zt)Y>UfoB<rHktP2b$U<Ae-W4*fSHmYq*iZtrj@c$ArxuTz0)t>as9ZqJDp066G7Mm
z3sE@_>E@@8hnirR%=cXSK_u6QQ|%s_o{yiS6%**rfrci2THKa^N2;}_t5(o~>tJ{5
z)#eT<v-5Mcy4IkOdXu<SS%(tT>KnexPy~lEsKmcbzWN;jFAP#-<dVsypNT|8NhY)=
z;;S2yD;vDi8`hCC!PUVkL6MFC`zt*?{Rcy`<Xol?sk-xHR|ri~w){{&HUf51!!uQE
zZs-nTD|jNjYwN>(s?8`aRT;Ss=k2Hd_*J;T0J7yhezHm|@w6q2(ix^55xyWF@c5h0
z`}y&Df+`!p&feWob(|(!c~CKzLLUhSp#1$*!(3y|{LEvO=FcDb*KcmZO8Zz$28uim
zaMUtFEMHbhXLb`j#QxV48kqpLZS0s9yN!v;b<dfj%8?PP7GM{MN2cz9txh+DsM2|X
z$P|BL;^-T)rV(kI0V+Ah3`PE8(EEYkJZ%YKdd$fJtRsV)6r!108lSoq;%kliFPMjk
zbw861>P(x5X}FXi2&2&9Oj2S%N7Ju%-gqC>c~@X9K4j55%SG)b{=sO47pO#h+Q)OM
zuGD#catQFIa3yd@et*Ty{h@4$ntx9o80Wa8wy~!8firy6wzX$<gJHX%x3BRHnVvo+
zE64R}XHRm%$&^9f_NE;`?A|8|uy9t3Q^94_%*g*>(XoM0kxf&uFznyo$Euo<1)^NT
zcXuq^lRnq+Y_h;a9eB?6oHrqHCr@Y~p_wKh6gfp3EU?GDt1lUlki0?$ENA=oO(%w{
zg`~;;pJdYf5NE)lc#|oknqab&ky(L}PY^V%1jSC(j8Ic_qw|v+s@QZ~W+X-lmCJ0?
zVkf;~6;u4El#~JFV$ZpY=l6^1b1oZW+n$EMS5{RUakN7(zz7gCj{h{M+oo^%h`WQW
z#SO{0tz&~IEsGrcw=s`y39|h*Nu}yTHlBG4&D*t(Et=UEq`>()_t_)NjKp(hd>mXB
zRd~Od{=sc`Eq@^H@tF4zYIUpuncyEc|7u?-TF)Qa%W#4ht*N+ti^#9NvB@3-E@?p0
zVt+r`(m#X-$H~M}`9sqahcnquSdm{4La+h8DRdIbc}+gg(ccVzFJPoC6QTj_XLB&-
z&&H93E0Yw7(A)pPqH?DaQkR)14_qwMt~tM~`M^jS4-}cyq`H`PuF$`KACb8OJEim-
zkf^si+^Zgeoo0YEId1U~$S_lov>5M_cK&1N01&Y9<S$vFW^DL96R|z}>S5`5-|T$D
zSCS!{S#9&h%1H(+?HwN1X!v`zH2xvi2n<VFKfm-byY%L#Zg&98)-#34;b4j!9fdsK
z&^?`cU#8A`p8~awOFveV#f=m8e_ksmO9<hnTh#}HB36-@mdqNW^b>5&3QS)pUWTr|
zxL?w&$fEJwnwq%7Nx%1c(~9pC?8J0ma@=1^r2AHSdQJM;H6$Q?2LCC{{+VZ|<1(QD
z63dzM;>f*BMTSDQR9A*weF!%*Gq+q+NU=iC%G@>_$ahf9QlfOKkywv{tg(86M#yW{
z#ilt=BcT%aM^IDW?Ki-`yuiXP5}yvXKg|M}hgy1P#y8^9#wmV<6q{5qetW+O-r3uW
z>brfZVje{>EiBi4t`+W9Y4?1w<qndeGDee(obzw#y8l4xfr$}JFSR*%(e`E8U`@)5
zs(*kd?_o>d1j&54Z7FE>ahYI1DQB61K!RnBDJ9~Tz5n<crmn(5k}^W8!At0k7jEbY
zqTuHUzeyD|rAYkGJ7wL*Uf1Hs&qsQCRiYAO8Kv^GTGu^XP`>2Jlo+K8w0?Ye7zd%O
zmSP+BPfIy0nhhK<$gD|N%)rOT-IHqgG++iPZGeTQZmTw6_&Wm;MJpP7N?dfD%{Wr#
zPvA{t=mY>K13V5Yv*_Ey+~B00QOK*<??)#Pt}9(=l@|>;2b!vC>`z-vReO6Azz$Lt
zRW_s6&*aliT;wH}r(St#V|84?F4z}d=c{y7l2R&5<+=Y8j~|)gGqk;_mU*ia)O4R1
z%7(RS61`-1F*=*=Rjq$U^TJYn1hq2^6vI5seSn{V;MweP*wkF@X}=Nt9d73pZ|E}5
zGLJp5RR98#CRHmui^pVz<Y&`WOMkdXaj5h^4;M~sZbE|Tv^ZbdK%Gs7nBfZnKg0DX
zb-Kc5Nq?=7rSq9>)Qev`zxJLxYOan&h<ydK#xPRHb1=h_^76HfzF3OdCIOzO!E>7Z
z7VFPm7c5khpqU2|ST}<Cb{O+UpG4CDo|}fQk4X?=@;SDI#K=NXTy0SnezI)s7=UNU
zQ(O}J{R8kj%5JKXo7uHlsPWW({SreY?xaZI)$tbXf+m%@H<y=nf<TO|QXi>8yFlOL
z>gV!>5rkvpK!wugg0<faw8NNrQVtM@(9?6L5_a*&*o==6vT=kWGju5U!GCw=?&S5X
z(r%E&@18bS?{8ecFaLVNUb^=sqa+elu%=+!-UF6-<55L4Y!eIK4?#XH-RHxcjV)Bk
z+-2@AO1lZyfL1rXNo1EzJ<7|gg{=zYDB18&U7ekyPBDO6Y~u@0gkF03nMoul8pzGC
zXHA}ZHrM`)0j1>}{BksZG<Pm3S#w0r<hKtXVcEufZT-^^=tzE4QesVX8b?Vfd?S+<
zkmByi>#J;uv61X%(;zOIS9Zm)@X!pZQmXh+hTJ#-SVH?#;$>|}rZE#C-n501kbDS9
zQDXK$G8SSZ$O6JZ1>6dT>y7DV+nfF9d1&#Cw0j8{IW)&Vnl9ql5>92Or2Thun+y#<
z{Ma)$=l)#d+TvNNYwqrzo0;i-bzOBXtjw4MVo`PtBe)6BI^z+nB>XW}8-5{&byIZb
zjvXJk5PRq;oc3m><fU(lwF(iJyulT_{x>zTx639392Bfr?f7)A|C|%J*1V!s*Y43z
z9*V}U-O(<WEq^W+l+c9X4Tb&W;LcM`Vb;Ss0B0Cw7Lfei`vK}n#gir=U;#8ZLd=|`
z$lZt%Mi~}L!$75FI-P8XO^g?_R;Ug*=Iz1cVR1fhlT7gyEyJX2aJcG|JcU)3U+?<q
zwR9nB#{sw26tfHgAmT9ReO$)Oyz8y_L{!R-n->J3+9*`g8M{I_3>V5Ka*}OEu@nz6
z^&wY<ei??5X`Rao3)j?LY;lMiFWH;^8VrABr~8dm0}tp+gaiFE7PhwGf&cKIIDpoO
zo?`zi&-|(5oJ4C%>Gk5q!g)GpXXozjb94)Ap=iMOB4b^mJgV%JZX;;r5Rm5u`IKl0
zIMa<N*6&_OuWzc@BZ~#ySQD!iDZpEq-N{51Bnt?9%J<Z|9aq-^)lQ<{%FgQl%$>^}
z=5WUgcn@E%KGCN0n)Tz4Pz=a|dEti}Er5{&qDVGJLU}ySq55jXzS@jggLTVWrOL?M
z(IdjNX(YVy&77V-@EYeHvH6odI?n6t)vQUPPGE3a3D_s~$n*`3KQwyCrfK;qT;-i>
zOxTBl=5)R9q5=*1k>a#NpG&nDsYrdNg^Pm`>l9Gj)Y4wf^_wB#8$4=06H6;7u9NKx
zVm)awdemXlaG;$q0X?NI#mJC8T)G53bzrMShVe^Y0u%In2rZY@#xj#n=S6>_FG^U0
z<f+(KZ5gea#S1OUbHI}|ZZf+YK8CBOZ;jPW1WSTBdqq`(l-XpLYvmYoy6-3X_uahQ
zZBu8cFCHSxmqsVJfp$}h_y}h2C=;XYlllot`2C<CU<s%>M@T-t3wSD~d*OT>PJV;&
zJofzMHe5Y1%csf|(_Lp!khPt-WQYRq&dsRY&4`C}3Qf^w+M=(c-%E|;GD9sQPJyWo
zB#W8Atel}ngf<r^sm_E4_h1jV<gNaoD2C#|%AQ-VmCHb5jib4wH)~5TMnRa5<8^!`
z<~mm9057?Gz8Gex=HF5twGk{_u}$6<HL^9aRjj>zG?OU}?hMFHJ$-_zxmWYw=>e#+
z8QZ3)ekiY#iMfZ>cnGROVI4*BTh7w|#6J#KLZx8n|5|zAcpgI+Dbym4XN%wYV9svc
zKv8a{8spU!enw_yp{Wzm9m9`oU_?@$E+No#wt8LK$Qn;<vR~WQ7Iz5V>dQReMpxX$
zP7c&Imv|f6Fz0c=Ub>z7lr|3>!6Z4%OkZ@22OM7vQ`k9us?>=ZoeXGGb#jf|USO?0
zWq?fvxx6eym(11TQ4Vt%Oo`QS@zYP-X?ervCcFF}&6nhDN+i2XGp(0|X=#`nBrf?~
z#axsE)Qv{Kw#jx-zQLM3RfW5%>Qmh1+J#AatxoV?Y0BRZ(}KWv)vFMyZ|>~U-JP_K
z@5J+i2K(vggp6$*XI_}s6+}_%>!%dB91c+|CB~ld(@E8SB|FS%dGayg@AVm%Wr$H|
zZ!j#S_IVu(@JK3X)d)7#_V$g%R`S$LMqLQa$P|()Y(xD21wDp3bn}<!Vq>AuP3ZqQ
ze1Y&u2906gKA~~EzPiE#OzIq7418@@td^Z9!yoIS%$>4}dp@wXOK@!y{Yp=cCk8Z_
z@(Pw3l~BTJ9xz&lNukmdg;tDdld1&m$7Q^p$Wx&UtK8ssGBLI%v0!Rz?BeKRfk($|
z?M}_Th_>@hKx%P`|4i{8c9n~U(TD#JK}){kkR_j9MC3KaKgBk;dtukL=eK%kdN#Rv
zn`FUfDB2b_ix?{xs8-fwQ6q$kz{U+7=*}SSM-ak_0Pwd-i(6*+f~ZxN(AChJIgvdR
zm4E>MO!9y83|@gts-OH3G}>QF;~tYJajrGKWJ=rq)?z6UQ)Jx3W3zwU!n19ii0{09
zy<93UllzW3kLdar&U*#<kV8zlMh(Gzgjk3hM1ciMT5llJZu7=b0DTTT&Eo7IhcTTh
zKYz6T{)2A8U~!lQn4Ws!lT0D3gr%U>T*==tPp3+{X^DZ_Q>h=?;08<G8FR^;7pBSk
zV{$T=Wq+;O@J`lD)NLm2#G>n?AoB=r9(9Ngd#*WlO;4)beBoO7BWgr4A=`eK309mi
zGZAaQLVSEwlqYT7&$i#A57_x@EdxCm_$`mq4%H^@YF3{Xk5P-16KY8|!680leG4Xq
zI(QU#3e3v=$SDqlxti>svj{?$ztU>xQ4q@{)`n@UP@8VQmKY}x!Y1UK&z8JO*to0T
zJWm!zY%><ro{94{GeI*g9Py<O>v4ciy94>=Uw3K954EWdyN2N@qe9*B#zXyzbi0)H
zF4og#S_LCiCvnjs<lA%*RdSyI0+wG~>~EFj3bbQXn{_G_Pgy|n_(jA==pN3lf!Fn7
zX7@Dg_1xUt)S|PSn?XtE-*$OYWqu>OI!Mp8zsW?C!qh}X-*I%;!r=4K^V=qmGRBy-
z_g~UgBh@fdeoTdTK-UtinY*QOq+=Ko-oZ~K%>Nw)?MyS0JDOSFZf8ejik5J}N;K)a
z^f~Ic|Ko9WVUQNRnU=Q=XQIK;)BT8zrfC;?MPismZxU_xozZ0A?r|H7BHpfmZs1Q3
zwryISw{39znBfvwMVXhyl*7W&96r7wA>E>o#@5VOE+M3Pf0HrFm#iot>E|8F{}*8v
zcqXbyh*O_o@XMs-f#}jD_o^VgSZ6cz1Jwu@I8TxHHSL;j*UEKZpLk~{ih^KtKPC=m
ze&TKx01^3b0H&Orl3YyuGtYq`Zhg8jcG(L_6>mk%fRjKgu`^zN9bFXGg^xr{0I73#
zK}lpNHZ4ZG)k3R|U#IQNpT*H@0=*RR+D=hw#3qHxaSU@=(PYjA6{&cbPTn`u(q0m9
zf}6ZpJyGQ>!OXa*nZJuc?0x*2$hVP6Tu4R2@rode5caWx#<C@e?eTU^pscld8i>Ns
z&i`n$sH*zw==&U2pbPc3#2vuWQX!USI$3+n(3H3%brk}(J{{d7d>y^R2iVWo07^%S
zOP+>p`T0R$`m&|L`%Abtmkz*P+`8GrtkL?GZ-_=}*iFuB^3l?p(!pM%RR<{EnrJsl
zeUj0>&aNK2d<8mi7i(Yrziw<~&W1+P-=h;kG91HE1~ghad??4Y6)`_eRs;bhmq^)>
zXkrv{bso)3cb3>jB0SobZyPq|<SAJZMbr<Fz7k5qBmx#J<&@$G0hal4nlxpFAIHup
z2}CSO$PDF9O#>$b+AxwZX(dRW!EL+SSqPC~5)nGXs}U}sk8Gn=!j?+t$Su?G%zL+A
z@K6=^>8Sp7z~dJ5v#VX3AA+@&t^A0_F}Z{|`}2Rji<8lxScx|4Mq<8d)Aol|FkjG6
z<@F37;XrY6)nIMyigu@Any>R+-Zb!@*Lw(6&mZndC2w=sLKxtFAziW+IRQ1Ad00tG
z0#pfrT}VzlY$_uB7I&GxHq|PrpcpH(_^ntE>Rk`kR+fT)uyRvEHYa31uXW`WfdWt(
zGr@yn_nv{8uc$-;fz>Q<PxlZ7$XzO`A~8x-@lz-xNsTYoDf$YQ7^o7wF1JOzko~bB
zPIva0I*_tWF3t3DPLK-6OT5__%{(6*bblkZ@=;Khby-{WVd5?&`fnT1Cot8bY=1Lg
z2l%x@n>-X28tU)iaUF%o2@ij1>bF65czB4`5x(S}m+St{Izi~6A9YCL!0g|T`xSb8
zdGp=BmpVAGD3^FB4GzDcXtdr<cmIF#0A$_ODBo#=?@#k%e9e@Wz*9kJvXun|E0Ky3
zQ=^pFi@n~Ce2BHnq5_q=yEiQ#=1{Q!CdOKQ8&58$ZW>ctMd}v?TX7;-NTQrII4_7>
z@pE`ecTrp^QS1}Z5cuQY(-{c~87{kRLUu2A%P;f1f0T6OMTGroeR~r;BOyX8!fC=h
zKHpqvPaUSHE{7V-r%%6}oFY6I?EyY2GshyP3r3hQ*t@V_fgF;W654K!v&duWx2;Zl
zLIcngy0*E}6BUAVeopOqioO-y-aDfh_yP>T%(MfFJpDakT9#<JAC`xrxYhSHKK+8J
zrEo=bEvn_ROfc<xAmi#5{;(=y`p^Bb`M~~M%6%kZ5~oG~@|RRYgJ9fJ`KTM5@FE=`
zpvL~gs&b7B(WsGc@>jVok#C*-Wa}Lre(Md09B{giC9f5Cl)HbQkQvelNG=sDiEdnA
zM7B9!knybrL{FS7m#*xe<u-Np&3@O1Ez4wD)2rj=>6aNfBr-4kyZcb$_Wv*Q_V%{k
zK%}(loA5+c^T37wx7p9;ov52FDQVB|IYVz(ws@?Eg#~ztm_RBNNy{;8pb&#s5yJ(k
zq1m#6$~KpH6i1B|uXA0tBTv`=9Dop5(!wR$u0wS=P0IyDdRQfqb~uJ<iEPA0Q~z<$
zqzG>rMiAena9O%c^ua!a9Wp}b_V2)6%*6ux&S8h%YSU<vXRsI=tKzDd#V-XxE{!Ti
zxcfA8gyq<Ll{&V~HiU`R5s>k-<c!l*gH1!W&9ybjLeWkcgZyF>YynPbCBXZro5T=b
zOvA+7;^u5+?R`4&l=3P&JtMob>-Q5LXEr|#{_T3`#NgeR*W};YC<GQkUmQCJrjS@%
zcB&0aAMiZzu}b6%F!s^-es&W>%P7ktK<Z!hQDWniUm!EM=XJT<A5MZM-32%m_fzDy
z*%gI!IS|p?z3+!fzw`6eeG7Nf=4AVaYDGQviH0)yl1ILKig){S@^q<wfkdQm40G05
za{iO3@kHK7m%c@SoVT0iT&hUYhZFfaVk&B;7z)=g(*2A}mOjr<vHHBH|FEa|#FQ>9
z;(fie1_T0#(O)FcNp2=joP-2_E|npXUvBlxhfHK<W-^T6Xs7T~C5SHgbWcxn#w!rl
zI+Bm5ssgNnmlGFzxOX`XM)GyICT(6{S{6jZp1u!WxmC$&%cTJwCKJcTd}oOK&THeE
z<Qm}^Qu=r)?3f1;++m`i@E>T#mQp~9XTZ60Y+TV3>$_$5<@<(1?IB_+1lEPEhO)($
z4&!S&{Y4ulZSXKx6+vDo0=+pPy0#LALPzLhGa%CipxSGfSTIpj#*vV88?a2XglUyM
z(_}-u5Tzn<w@#EBi(q=MtxTh-CGWRs(dAMTM<PoU>l_`VZ1s3}WxId$(qa9>khL$c
zq>NfT667-r3@$d!M(D+|$4{~|Iw*nUWc#MRs`kJP)X=$o@ZL<Y2=h8N4y(D64ZjI{
z_ld|l2M2h|=Q=kBS`5%svp~f<A`cftJKUK2zZPM_swT_{6A)q&n3xTb#KA-F{?v|@
z`2UBH2!-`BzpvyO6#)3%$i!41C5pwrX^;tLoV`z@E@TTOF4^=i3UH%^|CpSOB~D$T
zT9NSz{^pWd<XNPzMenXg1Wi?P^T00d)Vm9MZTRVx*_gdYnPVug#BcIs+x7Zk`d?>N
z*h@KHx=({PKCbz2^=XIg-u5bo2uHL(`U8d$p*cQ#f;6gs7;oc(bt)`ATsDqX+BYuR
zJ30JlsFf1sE{q>b7=8bg^P=w2<ws6oMf|YY4{Lt1Qh4&#cka}+wt&fInRTp5aI@4b
z%w|{M$Ius<#80TF7ed68!A#86Ar{Y3cT*B$ej!@XuiuIm8DytP8gRakMhAMe7#mq*
ziTiUI7Ll0N)LxfjoEp&s&_j+2=?a`wjoD(VTo$iV>_u0CY5G)!KCFNWdqs_zBWnkO
z=|WxJcNHRgo<_Rg)CFHIfE83_IZ`UoLOkcJ4ZpQMjQC>sc}-ed1!T5lQd)f`T|`Kr
z%f8>%)r9|5C-z7RbEz2KxqG?EZRzwJ1k&Jcnl~u=A!D9WCX7hdhyeQj4jRSahd~Wv
zLvu3@5}nUcL(ZQRi*yP~|J&aHOH>oy3}JdgP^$Jc7=PX|!Q-TqnuR4bIkFRR@53mD
z^GdeFU$V`5>~c<EJrKbW2FCFKB?OWkQ$(*$d=5da8YKJ*>fiN-+Kb1`8Ua2|3Qcr5
zn4BQo{?C+G*Z31{qfu&0rXdhfJL^e&a_+8OE7QR?Ffu4uvlBYP1=<#<ywi}ofwxdV
z&tLdTJP=JPPTElfh}w=#W8)`6KY!-cHMWo0)wKTxnz>+a6Q;aIe<K<2B?5z(SO6C^
zputtE!)o7T=H{}vy*>22UXI$oy}a0dbJ=7ji*D%G0{0K7H%Z(`WTLR6uoTT}_!0MG
ziak~PHP)46fYRe4`n)h%Gr!FgW>?Cv%{R;qqw#7&OeW{HuKtfyQKWK)-4;pjI2Ia+
z7;rc-)-GXTnpum?<V`vXm-i`D>O`eV@%<Kap}N0_05HliaSB&+fRckpCZgvZf07Li
z%+EP$ii(9<C@8BpEFib*n*Gi0*Q#p5-I&#{p~J)t!C$8RvY1oqaTGB*;Uj)&g_{}A
z0Q#g8wq2^v@%}r`0I|f~nb5GU#(RaW4o!cL@D=~VnuAuNHF7t+1X0ob&3lRLk&#h-
z8Jv*Elc;fK#*qf~TbCoBB>2b^l7owy9Qiu3#8;)nr6KR5%A|f$j^K2*bP4gFuc{Q7
zpD+K@cb@wP#o-(|L+vE+qZ|%y1S3BraKoe7{Bu;K9FDPf`bCs!=u%lA{&!IclHU1N
zu6e23Z+JhnbpxU#46^a@IYqBrKBrV|6LU~hY;or)R1*vl+Zsw{3kwLZX}@#_aPZ1R
zFqi+I+?@l+A&$n+;irDPALjTfr|58;Uq{Z&yi+JnQE*5z7BMR_sjltp?0H(nOJ?tL
z_kUWy4x}4^fgPGC;v#{Mdpd$mGynY1taIYJ<N8DOhqi~8s7_G(259Tj&e_f6w=)0y
z@D^4$llDQAu`JN2P%r;MAv&i^DFr!HE8RyDi%WtXo@SY&=6vTId;TeT*5_r-YJr4=
z{Ld>wPK?4SJd1z`xy}E|qEJu|W%BLpR1UE*ir-}qkz+BYuK&UdAmBf`#tfHjF|uB!
zBe?Vhcd&~U#}yC};74Zk(yK5uU(NO#D!6(Co3;*EBnd9%$PZxce^fKFX!3V(&r4Ma
zf4=R(Xj@y{cGEJq+$SQpA{TW$c3pH5h=Rri$x-V-DGqXyA#831C*IsjbbJ~xaoHjW
z8;qr%YsBPNt#q+cB}Lo-&Sunxmz;#M@=TnP0X_?c19{LcV@c#LI#_1YCf?4atu0pk
zN8~%;CZoAppzF)|WMF3SN#g5z9&P<m%b(nwIt;N`0%<IU59c7uKgXM6V;ksChNdNC
z>&;2b`Ri=F?tD@W_|6fL;KnyP%u0tbj@Zz_THT$U*d~JM#9geVrG%*DFtFyJYj!?)
zxq!AHs|Jya<yYOxE*2pBXWwnhRyuzulqctb6Czi<|1T2a{WE?SZFGF8awc<!+nfM!
z-73(qG_-!`a$c0!UKl5<uNMMOGiV61A_Us=-lz)qDd`RF60`xWf=RBu8kijf_D`Lc
zL^6otGgwr3WC|te025P5I?L9}NHh-RM%Eds`7dbQtRljt=Dd_TX<2^=nJMD<mEyOr
zG0)l)|KKWb0tal2SvAxpUIUF)FrzNP3?Js7uia$#G9X&MuiUU~8XxoEz2FyOrF3P_
zpxyhq1fH?VRM{kgJ^{n5%{?5BN{r$ZiRyp$U(8$vb;b}Fif;EzS+VXNfrUc>&x!5H
zOm%929OX<Cm0DCzqyQT>&tVGdx%{4PXl5*}61kv|t9iQ2r2dX(-w4m<#}HQ6#Rrvc
zpBd6@(qv@Vy(e8!QP`ywY7Ir|w&2GQ*&-=+T*vUXA2BrUnH;uoRsn6i_(|v-*T%Ik
z=(Q|QVhT$B=Y?;RjXxtMlQ#y9Y;K+?U97{!ewy#O`!XA9a5%Gj%R64VYI}AYoWC^^
zqDqi5NL~xQ;$Vx!Kr@flX;Cn!aM2YlGpW$xN+;AQdLMGC3SyYAG20#926`bX&z|m2
z>+4$+gwMX?OIid31jmx|5l6hQ<F*D<3DR7}2)r9zePW1>jqHVu7Sw(`xcW6JoE0V=
z4UPc-jui7n9D&I6!042ug$35{BO@lQF7dYuEke*uV8oImjwmjSx2D3S#8nJDe50Eb
zp+#ZxqB;yL#13sod54<J%|ad%Q~R(yXt5=~L7;)sKjC2u@WuUbGGloo*<UCahWtvy
z(~-_A$5dQzrrq`l6-1%D+f!qFB+5f9e3Dby+{gg^&{NGjO;(i_q*$N*zO`7bnQEo9
z#-dCFnPAJ7QyS<&{(n6-1ZE^jLR#39)<|p7+~^hj&9y*F<%}K%{4{H%qo8cL<9C8p
zl$QKtFIggTs$kv27Tu-bo#Oui073u0-6P_9gWNdGx*jJx59y|39M@yM+vcU0U&Hr9
z!g9zk9@9HIp;B#<=Zd&<%HG}xKMa^%TxS#~<iZM997iHFEY4hDetr#<#<Wi8^*gkU
zOKB1l)T;<IiS$8)H0w3eD54=#8UZ*cq&PQ8*SUgRQ>#=NMPt%5!;un4NGhc&X*@<c
z4%FrmGs^|X)8=YlSiivL`X!bYSMWWbPWOmjr_He6C(bf7g6-WMq~}qu*9b#Fx#V!;
z+7&h~T&7yBlI4kwUmQj39~`1o&h=|A^5$!=Q7VNfr5R7g?C<TPmBjZPYULU)-gt?X
z<u#r@eZ-^3j~S0geEQiZXq8i|T8>;C#XQ;C=H7$PX}8<#?(g%%-}@d{FJDCp!C*9`
z-EMQzKINo!N<4{d+J=Enw|ADJ^xIrtkz;3(R#$@SNHhXvOySFI<=<d&28>21ff{9O
z-+4fJuF1xgD+K)>PBbBP1X3Ct&qHLEkCH#fff7QXrye)jkY=%cXdQ_XKp=4fpCAl~
zCu93i8Suk^G|L$eN0esjOyZcJX7e7$qah_B@Jb=BBT3@~tqcoGD{MV_Vjm#aT6ieH
z3tU_#T$uIv_N`@tgAPA`*5>inrv#N63yVuEEUhx^bvQabq8m@FSD6rGK%N`Mc}9nf
z04O^yuIG|un$F=4&cPli!^-k1SFT^@`sEG!qah;BDOKudVUfu!o{*<8xpGihj!Gib
zsEZkQk;xdHM8ru-C(hWLBphgnrO&db*=&~4Ctb?YZW4{fMQZgL<8hBt5MqozlW9j$
zMAi2YLKNOYcI@s0QYp$o87I#fo}S{>FMx0gC7(PqjjU8^T)ljigToyT_O~&r;F1Zf
zZ9<-Ne0svp-X6ozkSxoQt|SZs>a{wHi_4VDRgyR+Pp!3w6!sY_mwYOrPo5=AMm<Vl
zfa?g#VZg@f8jJID#IZ&9e9tHFeDX9$OGBP341#X2$44K3Ot;%BQZ&w=?dQj@j*wFq
ziXwI6tFDy4dKn|gv1)Qxfa`#-bLx(VGSJHtaD+u7itzB7<KQE4J<W?3muQ4G%1;?(
z?NJ0cXp9sXWjSA!+6?7NgJ~3z$x=kBpr4?}8L=bD$&s#09!<!j2xSCbSjG(kyi(Dy
zWf=xZ5=D4HXvg*iha@kE(9&@cLXqW$z%Szp2gi5O(kiQm!-VbKQ%dz3Nt&@<mi)s@
zv;4b9T@E_O)E1YSuheOe2OPu-+j2=3nlmge%rO}Dsg!HXHk+)it@3Pp2W4#jttSkw
zbQz9D7y;Erm3F&Brc$I3e188iS(LC|uhZRoMyq>7rgJ17&Dt!p^*VXEN$OQ$zJ_cp
z;-Bs^A4Q1%2+#3xbVAuJQ5S}~P*hw^tz^;lTxncalBNm6ZjVy6g7=)CVid>pM*}+j
zHU>qg0vhudDAk&lmwcuyH3%fqaj8~oOyVI)l5nuUPy1valUvol>y>EEFR^j?8Vd`H
z7y<20hrNS+2E!qq@3FMBOmEm>|L_3E@sN(svz>kZ<-hqOUcPmUYgew)oSDaSC86iy
zHX5w0uCukh%bmM-a2>(w+A7sjg@u`U@?3G+ZqsVD7>~!K#*iz;X{*ig$tl`cBjxoA
zD}3$E*I8X!W;h!0@bP1wZST_S4QvCK+xWh}?G<{$js^>HZmu~ydx|Z_jy$Cy(>ZxT
zQOF#PG=lu_gv!M_Dm1tlvN%FL7lGzuW3C*fA+-}9Sy<7;lf-iSJb_#CYzC+!aXk+&
zutmM&IAneRlZd3(w`ZeL<Vq1%$~LV+DVxe<Y$I2xRvGsO2-me1M2;P`8)J};L}BQR
zCZyV9#;*}NK1$?VtktLo752Jq42Hn-DET42>yqXfz2SflPES}5Lgvy0EnV8fKAm*J
z$Y?wRPV0z|fAw>gS5|RCkNv|##3ZKT`3R*^S&k>5HrvE;JQnH;oQxxs0-5J{I>XH(
zWS&DbX4pRD$v7f*0(PT}e%7^ne3moUXwqD~Kqy>1Ur?!*Q7S`7o8zmLLg$J+O$ofR
zrBTHRy0`}taEv7I0)*p%YtxkklFOH_BOHk05uNTa-FBO9r%ktYO25}3PBi7(4Bf$y
z&alhU@-j<Hi)?OQA_z)^L5WPoIJihhQVPo~F3jUP4xZ-{Ckbh0*X(#aAxVl6wz5r_
z=X=aGXDGk$BG)coVVY`fl%e12lO`$0tv2`W-{aum5UmYSbjtthU;Pn{Mh&A4lPDoc
zvm#x|SW{)!s;lGFn)Cm**H<p=L@}-th(gwDYy@H<-=7+8N-1y<_{Jk{4LSb!E|-IV
zFdGm^0WM^Yu;V~2(87-L9M^i<NY@7Adah&1g;|bJx(HJi;Q2VN&G9Q&D|C-e`2Vr@
zXTQ2;`F$VuS<^GU^PbOKr>eTDt9oFMq*@Y5IaDmguo1&{A~<oJAV{1G2S|Vb`2&)x
z<ic0E4uT*75+HB@DzXwwltgiCl0A2Ib#)Er)S33Y-}xEWkc;)~UDeGd#iB@yWOD(#
z&aQp-*~9a$;kUlO@1V1YM4%K^*C4D17hU$9b2L>=S_~qmNN9}KG;K?grp$JCxLPcD
z@Zdf|DZC1_E@o0n7q~yn@cTowu-L~}oNnhl`LHH21^eRxuibx089Z0>E9T2(Bp>7H
zf~PT{9BjNrHWsubAuyGS!@YfSEm&Qhb9QmUi<1*X)kd{y5R3)|lgS9BB-Tlqwqvte
z;M)zx*K~EqMVqj&E$h%x%P9MwrSQshMtgoi-ZUuJareQLd$*>nSC?oRvq6QFq`flN
zHZ8tamA2NRd^G7yykL|M859GgPFT;EB+h}>y?8?Qd6<E&u30bFeDL18tmlguf0P!n
zwV@&@d9_$_adyFSy`(HRG;NCzlKuUCM#BlC;h4l2jFM>SF)|wVPlp2z_jef$GK8?S
zjYY_~>1L@R&r-s;hpVqQ#nw@_73-#Eu~>3)cFwk}Xxi&$=grLi@p=9emE4~N!!7_e
zL;*u$0+}}W(BZ9Rs5D#GVF_eXq68=*7-A4XvbSGw=WvKmJkoj+853QkjO{3E9R`7S
zn~z$2KJs6ki&g-(X;DIAor7GH?F?e9k1{yh+gV5eABB)E(Ad}xw{43~vuMN{0%@LO
zvdA&BO@sGW6nRFPWjGlynL>%{$R;>A+GF={fORFG&o?Mn@T-qD9ACcRVE2&WXw0@O
zan>=&b6$P<C7P<{?EI9yJ<sm$K6h^2;`sE8rm7(XOt03gnugQk7u@^wl;e{#zW%%4
z04X><KIf&~Dch6heDM4+XH`WT{caenw|r)ka&Y?r-E@H02Co!hyi0QV4237h(b<j$
z%cg^2kuVvj?2HUr3IY!LR1mEcZCRmJv@~eDj``&UO|@oXG}@$0c8=iIYl!I)7VyFo
zyz3`d04W#_Ml3Ef=I58JmkX4D#&vAB708nLV#Twk$K1JhkKKcP)|-;EvvcOlOO~r8
zSGTX2er?8hFyLr+O0Zpw^qXJuqxat7?$JI)8trdHe`&dMc*wu?OTWaEXHQW|QKUJ;
zVo1|8TrDDpRVzi3Wh}0)xVX4rwOU8|LV&h&eDLTQFOE+sij1misOpZ78yNptjXp4j
zAffa0@{@oRV1yu338l9zT|`Y?h(Lo8GNSNhNh$)U_+0T4L~NrfA<=!S86gl#;=2x8
zw_F>V3nh^{K_>}83R>F|f=6nD(kAkZ6OFIq`6qn*J0YXELjbO`6q7+zcsj>uJf@kI
z2w*TAB824f{2U)V8t`5sgXi@8l=JnPHng}P*&B@5n@t$aW^C6h%FUW~yQcOIgy7a}
zN*kuUe|Ex?vtx!*VMSnT9j;GG7-)$Kj+2w;EUJ>GX-Qf~rxlYdW1OapvJAYXSzKV8
z<RC9->yl=1M!h^`y{*_-No^Z0x{j-u)kf<DTkFY{;vh@7yEDQs=M=)Rzdy#=8l$7|
zK}bQO5?uRHzIWGkbX|)UlBRB{wp(Q9NqgFj@D3p~NX4KS^V-AD@buXue(#6B!4JOo
z9Try?)YTT()^yI06#EP(d#Ip!_nmim@4a^!4u{O9GmdWEVs~ecEYH!s^8D3TUgA67
zf1CH-|A5DjKIGu=E<z}5cfDL)A8XmsH5MN{o6VZz({o;&o^dr_^et}9?re(CI`ZPL
z+pa*}Sn9KRKf6D2fB<M(OM7!p{rLIxuNt4Yu!~S{Tr80b7ZtG~#0}qtsFV%DBhjcJ
zF|DJ1{(@wFk3peGjKr&ebuFR|gf@D~^x9fXnjn?xg&C3OkmecnW*cu%QjsJks%*VS
zYDJQz)NMnSq`1&$V^%driYT7wADal#`%Ebj3D!j8!ri?+PM$qyv0O5ojHzo&Rkk$S
z2HQGhYPgdX<VObC!5=(2=cKDyTg&nJg3}AZARn;5yUW9S_o$nOv&#!EmkZWi1Gt{>
zE%0qXT1O)~j$gcBHk-0suTXi01jd7$fsiE5lVV}Bpu;n78&2kHwhk&6iLcIk${?b%
zy^`btj+Ep`8(3@!wq~3JKL23Irw$Ez^MqC1va^2>3!lElXtcq0vB*$aLT5W{0N=EP
z+R|nnoog8f%XYm$b}NwMXxDIkIC;x<z2eE^N4)daTU?$Uqog272S_s@%L`1J(sht0
zL&Qi!fUcd>)(xBGn$>d2csQosR@knKq-`J3*lQl3l!4B-Hz<I2u8-UDSRZKn2%oB|
z*sM3Kw;Sf04a?<<s;YYQ?t1+61@ng&(!c%z6&1d5&YIxJq{iwfgHI%6Qn3y#3P~;n
z21ODyLBOqBBN967rUC)#+HlndgcC@qv56w2eGHkBF%AjPP9mI-5pvc>!I6~c#83s|
zWhk`xYD-f=qZMcwIi3h~l3?owmvtze(6wFU!$Qx+6y#a-Udx6BUIo0c&{?uHVK_{w
z>n&1^$nyeW;B#2kb;GYdZJ8gxARmsnv%AY^Jf;{FG<D7M7cbb9n@FV2awel8NBetx
z_~Fx@W2Ji`4a_euF&}=w;ocFO<%(rvxp#O#S}%EidCH=0D80l>i4TIx2Hsz67|5p-
zI>Cy&2&vIF3MZ6=Nr0-XF}4NaxO@LD4{lEwsG6?X5`-i*iq3&i(d4(SYm`w`+mdFv
zrpQxBG^5Fw@xfinJ8xhIdl9H}AR+-JUY3Kxxfa)1cBZ@h@-O}}bz}LpfBCz7_~JQI
zdWKnGrs3*xL%9{v*Uk5?1Fhru*$aN_*Z(EPWZ0%gi+~dGcRo40<oCb%EoM6tT!<ID
zo#~Xjw{Nk(yTifW9zF<y_bir??A>)8MV2uf4sqKpRaG&cFXF}ZV@o0cRozkbNgp>K
z6@Mm6=VnML@j_sIf7Sp(1VnKCx={>O#=<r@5xXZr3WhpbX-$VYx=$>y!XuoEdkPXG
zqu|K(rqDtwgb!48i^Zc-gUl1mAVV3A7lF2^>FS0g%g||xkg*F;JB#uGV+^iuw?h;G
z#n?lI?c(<mO0j?IR@_JX2R7Z=iHJ}Lgr?=rUV*w;P?sALsS#+-sx_-s5;?+cjnM+>
zJ<@xGQf%hu=wisH9=^moA3mY#+UPeJ1X@T!=LsUP-j<Xob`I`R<OSu`8QsMx?Y5*X
zx71o92L<aaWp#dzP=?l8F18!4I!o&vsyAFmK~8jg7#q!gYM4lg=~}Y+3c0N*wCC1N
zLU2o@ih18n+fZahZ<HN<OP8B9uI(t-Te`9#D>9~sy9_2XLNUV9_U8^g2o#a{R2S4$
zMO~GMU?Dh^N*EL)b|<$;lLC;eH*?O9PdR?}f)_7daQE(AiopPYJTEvnIO0>UJml-&
z_zu7M^{?{_U$RWbGr*x$O#f-ximI-eFIIf;_z53Ad(LXJ`7th>6tGw>8BZrzYk!PW
z*UyT6k%S-ne~<WyQtwYbK3QQG0z%0C?Yiw6kviZ18JTvG`{PBtZHOQ^nvK~XYLXNP
z7U~vXl?bPhXl&bo2)MS3oSB{sF0@1_MeqTgXE+_9>Fb<JVpcp-E9y;@=xbw8K_INh
zWC`oKVvwfMDc*aMG|HU2)}fR_sf=b@a(Hyae11hbD41Vf@&3C{*{oYSCo!pHI8K=C
zjrbdPraTyD{PsJ?ytk^ED@iRh-FC(E%>vzfe;pm{@Tpf{Vp}!L=X1_3E-1Gp&PH*9
z&NNCYHrq0CL{>|zbF5p>l)z4ZIAauB0VmFKUbn0*bVB2hNCCNM7->nSB)JnD1jU}0
z3{=aVg8{D{<?JLavRz<1OO_273<fyoFv<`@z$6A0qTPuQ(OJDNOI&A3R7#Q$zzB@1
zs7@a-+`a}{;C$Mv`r<*~wp+IARrIo})&%ERv!>}3QpX6l!C(hvGPDq+rXV#HUEQ%-
zta<CZ-{Z-n4{_G>;`D;mwxa9T$cxi+{@|P6CP@sni#)`(wf)U5wmWR^|KnWb^IgZZ
z{Lx*YKU+lMj~hltl<CHhvmZhPWC)~Ev*}uF0EHyMAcG`GhYt;%wTNwrQ3{KXo@*Y0
zP;{b2)B>3p(m{^WDz<ZtMycpghS3NuL1>b30a74SL(?|#&?0(tDW!1UV>=gH(n@@k
zJ1Y&T*74aoiw_RpTE?R>oAsKiY5F#@C)Wv`Z853FxfW%hC<;uHqIANi7B!!`bbNmk
zDAx<t>m`%flwvTTX=*mx4Xy2{+ZG=bZ+`Yo?%uo2>Dd`wRY#G85E!NDI?J;UpO9T#
za%=a9Pv3orY)j5gkC|^)R6byZ#>H?{f{+|LaF-QtxJS%(mk5=xT3qs|te7atNGM!u
zk->8RR?h8_C6t#eT}#{6?Ck7BUKWn3tr)5y!8(#O>6OI_Aq*l-NJj<e6t#DYHUw~v
z;JT<zzZM<@kHm4fKjpvuE5E{Wx#l1JlV9WU$pxkJ2<K_6rS_iD&LKy)@3UI1*j&!B
zZA>|6nwAefe1dn;w72ylUeH2RcfbAqBY2;i`;5_=TL=4m{>?YIed`Dx1W%v8;Dg7H
zSZ}rzdB)Mf0prn#i^YP|(^HzJ{V90%$LI9Ns<}Vo<)wt`Td)r2qfjr>B_jRD1Q_X<
zf4z1cT585xGL{B}U~hLAk;>2`a^9f=&{+Z^dYg5Y&WC8LWi)z}gBE0?9I0cMBV3<+
zJ|F`z>(p67keDQ;P7GaH;oF#ft*+a>-rYPZC8Ib=q=vSs@okK2R8le=4k))J(rA2W
zxj#{Sst~w3V6a@ej#I5@niA;)sZnU}84;M|hI!SnQ3++aVmKM`=F1Ox_q`8dTNDVv
zkxD_Drlf;0_aA&3nPiwuvv;&hv)Dt{6{_6Add;eCcyF_%vMyepeV{{<D#flbY`tJV
zG>nyEq?4!^4iHuwjJHfR<OD{!<kdS<MsnMW@8Tr67S<$*iAKG_<GPM&wMKNF$<YqU
zG-EcJP{}do{ym)VINRQoKT1JSWbE$jF&vJlw`&IZnB{WLeC^4Dr5NX!!k`k#cr;;K
zt~fhAW^r-FYPkZU+Se$}o1gg%pL+NbAAInb51)L<{%pc$Uw@s!AZNK<@zy&(;_0(v
z)|)M5)zDe@QzyI+u-tB0S5?#+{-KuYAH7Kaw9Drc+MUGKic}vJc3gAodLGg>(h)?g
zIK2=EEqL(KZPFxQRlwIRqN<Q}zp-h7?^=WuSgw(PfKCk7JG3z%1v*P<wLoc!%2SZy
zW^&YZ9r+E`k{1Qwq9J1MEv2REZ9O8?tE5EqFkV@gy`P+9G?{>m$q0M15lQM9Om~^?
zjtQbdD9dWQX5TdY%4=hej!S-c)p6Pi<^+sUSl@AVamLx%F(%2G&So6lxy}8D_h@a$
zcC)40Y)P%7t6RMDl$~X{Tp^{z_K#D`NSb%v(^$t<7uZ;kN--di$UtTkt;`8cO;KB3
z9~yq{uwXBN!%<F=SY%k^x{lT>T5@(q(eXPm2}L$Q^vI=_(c&P4xEU?$nou{fg`gBs
zE73Am0=GKBrvuPa2%_gfz<6BnnK$1c*e&1ugWu-l_ylCUxhCm=@py;X>^5mSplmh>
z8PEy{j<T$|y1L@>@`~UEW#?&wB%lD#dRy`RcVf&~_pvD-!u32PLb5-(8vZyG`;Yp&
zKZkm6Y$O9E(L&&nRMyivhw~AYPNc#JP2&Uc?k;-8x#ELISKQv8u|H5?9jfO#MPD)(
z`5ZRH)=<}?6OBhkZd{QfbVREh5&ganrPf&nDtg%=1YKQ6U&yYF8M_E@0onVonIyy6
z=)I(s#tT8ycIZSi7!5%{aGvAm=ZGL!t!nDF<K>rck>xS>Rx8aQlN{~t^6E*+2c@TV
zv90TaXIpNlo0_t&=(@-)Dck4BLBhkAUSd2OvuK+e<YRVlh)fI4FV4BTykbzb*t1je
zLCX1J&e^u2LG~u<jzk8ujEO0izT>yAD)v^-NQC9mM)cuslA^6+HwnDHSMZtJ3A#Ct
zh>n9~Fs8@~lu%?v9-oEv7^xA?lVmBvD5~w2**K@|T8wkZK}K*b);2NCBxr(*EjtxF
z3P*0@CFs!)-sYeG_V2K)>$oR`$XD&QH9!2|5eP|9NUCN|*;XvJ9c2^SY&VnmC(wcL
z6YsR0<;n9?uI5X2W>Y%nm@gL8O-mr6eW&La<Y`LPG;FK-4=)YTf5Dq|_~&<7N`xl_
zMH}cXmqc`HAb^Mnmmn}wGKyn>QAp6ikc}L7Z_ncGrS}_*KdwM@kx%Bj4&Ox^A-qRr
z8PX{7!eA1Emr>{vRht5xn0Wc=X#s&KpwL+wi^Z<PXoK(`D<W;8ZCdhyMVcg{w7m_K
z>`_$LMbtAHWIX=hIqR+E?tacc_{<^B#+m%yS;fCNt~d@c4_C>U*Z!&9A(I3iNyoZ&
zXaT1ur+oRVKhJbF;r!x)(~~pG&6Yu;(K=!O)?LzUfF+_(Nt%%6M=a(G7RsO}yVQ#f
z+w&8eU=b!N^;4yw?&w;}%VWbY-{0lsOyV0sSvNR5X;P3Sl1#@K#Z(DW<p}i_HJYN8
zCQUMQFLvo8U!m!k3`Z=^UNFoGM&p9nY)b1A_|$8(J3BFt9oKW+VwcXl4k@ESeU=x@
zW@D<>vpl`T&F4%P=YV9jSz}#GF&J^@j^@RSXOvZJho>gC&VTT|@3UTRnB*xJmyQ=F
z$2_=mn-|9~_{R6X$MMD0pO`lqdJ4|<xK<>DpSo85_1DiQQ^g8Lrc5759}vDq^aN0a
zjql@8ND7xAq`@o4py)W5X=p3_rUe%XlHNvx!_ZZc3`>lzaa}`KdSsVClF&JjCMqL!
zsxVoCGBUPmug`N2DnMN}gxuq-2L!J3=%7Ftf-=~;!RHAwNs*Sw*|S~zT<=Nq1hl3w
zLtYuZN@qRRdvszDCZVjh1Px84>AEd<)4*(3u?i_~Ee(&CYf7z2MTGxD*DybO!SU$}
zWSTN8hU^R{jEfOspqLyMAS6N>9tK6%TB3}q!e~cbR|F?WI>GJjmb6-93Am<%vZbpn
z>i|<cUp-3rm6xYH7)=OmpsFshbxmtKf`~C%qda4p4KYDuR1#HOZHMq4L~pO*Ayo#~
zHCVePH7ThNB-=nMBFgEWJwhl6yRYJyA)P{phAh#%`tr-*;jMR`a{NvVGB?JOyAm^O
zm?($WHliDuL70N7OYzdvHZFQNSs(c<o||Cg0CayOf%K0-2>!Tr^4}a_N2)_~AnzQN
z^K2bhFJk)(NTgt>Q|7j&_KsY}HW5LvTxkB;Z@tgYy?Mx+_jiy@)6;V!saW(%@@S0@
zt@Su5uvLe44RVm7vxGcNQK`X-xL!<_kq!rh-uTuG3VaAubxqr}$RvXRN@|*}rt2z_
ze1PgCzP0gdC>0N*;GzXjqR>Y0!F$K3bjWL;`4qE*VW0jY5kBB_!p?NY%d;~M<`ruj
z59+F}Vq{!hqeNg}BmxzYq<24Lf40kXHeqvdi4>CIctSGW;cC5Ru(Qv2yo-JIK7+1i
z-AW!W)~x$BeugELfkC3sI-wB;>snmba)rPM!M-sZYE9uC4|fcoIZD~jI+QDEy})@x
zK(aTOGD-%ZB!glYd8u`Sk_tM<WH_Q6Z&_WOQZ83a#u=S$@xCF1HfB8g==<osLO38e
z5YYs9v)%BUzxVq*dVYevJ`Z{gj|;Hata<nGyX?$TR;w*n%a-Okg!1E)`cFW(4g}U^
z#kwm0(0XvevneY!W%b8>ft#N%e!ow~2l^yKi0eIH^xrbX`QQH#Ld2Y8BLzlETIdrg
z0<)Q9m<yb+P&c?>BR$4zeCzRT8*{H^L~Yvb7UKglPY4=>(I|THUWoiYon?6W5%TJ-
z!&)18+z<_4doH?EnyQPJqPDIuMn%h^-ljw%`hVL+LB^oq{%iNCJIl6Np|&lA6z?4Q
zk*D#VfBvFnBQ(D2*vSlcCmH*rl*_s$lnt3yG<D4%eCr!bW;2Q`=d*9V!E%0uU9Cye
zg27}O(WahvX*<ia#~<?SgC_)Qv05P#iSeF<s3Z@)FvUy3mya_3!53d*$6Ea6ik1pz
z95T2_dD0&%x*b>x)@yc5ffkxPF9`P9630cO{H|q~6%2+W%Jq`TW{gb>{2(J8rZm2#
zt```S5lyRnZ$RHM9u0VK|30HZicx~+rwdveXqtwmZAsJKn#jd?)VvsS>(*U1n-xMp
zKFF}X;qvl|M~^<BtUaxV51*fKe(_J}+K$#a{&mcI{=AQ!PgK~Y+LNi!7hyzGL&V2W
zlIuQkd|*KWIF1fxjB*_pO6TII_wi6l3WVq(aS`*_TxZcr(RD5OfJy?v3!L&KNk*a-
z*7sb#5N%PCB%!IQ*srhSZT{NzJkchij8f6nHC>ZpjJc_p^E{_1tKKA35UeH1a}W~k
zJ!M@H1Sn}p(;<l|Xp#nF1|&&_?K;NOJwCnd`1<1)eCK4tS>>@9luRh3V}OK4(l#Ye
zE?2a~+ZkFRwI(kHWJyMz<Y=8CLI<Iv^Omx6RBcCHE^*r$Wi8SNyb^e&I1Ckk=TrOq
z-B))x3>Eci&Sq0%ZG#jbr6w^6S(Y-L&PbCKq$JIvQ#RLrZ9U!^Lg*EzlCo<^(-apR
zMjMPta81MF=`+M&NHV@nP!ocSK88{XYzUlP&3X63=Pas7{*+aV+?1Rx*F1W5LSj+^
z9)v^4j=HJnT1VIQ9%U}#P@bO#UH|1AJ_y!zOKUBbh}uE)OOtpK0s|S?0N%yuJ|}%t
z9S2QSNq+F|DX-k$BT*7z`)4C0UN|}t$i@XtRiS{a7=Vc_A=62MmWU)q_ec?Q&6L(S
z!29U$C#7UC8c|;@Xtotb8i2^#(VBKs#<s07F`lXMC}BWwBft?_ap$Gm>>ll7eZUkc
z!|@aeG^;guPc|Ge91MAIYoA|OpK$deu<)LyYY^Tu(h3<Q(t;C#AQia4^ADbK^x!s&
z;eaB|nH(JQ-t(uF>nn;xBUQoS{g-L#1=Y63rvt_V8QV{t#dy%pQB|<?5ReRj+rsg1
zV7Q&93<%ua%XlyY<G1(@5C)?asgR^;#%Nej<T;zV!kGAsI@#g8!w3nj<@Ui5ZM9{)
zU9&HaNOD6NYJ%M$yv4YvI_^S&#~^zmRsgQ97JU1=KcICVL!hHj`ev71Y_^p3mZsyQ
z;_pv2{*1VAKM4o^IgF39j>N|h!dOp=o^mDnRF@FXm*~RIph!`^!>uYpkIV@qUMN5k
ze9Q@zq36Q39ZJUn+;!2BASoihE=Un6LWI~Z?P6Ycz1~o6HfRwIq_Z@~yXf~Ag1|Nv
zSsJ}c1)>Kx;!A70jzk*}in1(8@`7wQ#^A}ig28ZtbCGv!HY?t2JgYkJXe)6T?hZ7E
zlblRTln8x$STjfyMA>1>bLPQwrZuzu19orSMVf?VwM7a;Ds#$h$K~mB&OUs~VVbfx
z%t*SLOi4N|X?trc;R7Zg@XEazzjV7m&aYUWtz$Q7cR<x`X*-8-7TNVeFD;Q$qoig!
z955~l@-*w?wxV(KHMdL>`Hn|N_jrG^=4y3?F@|Ycuzu@%B=d8M(Uf-oC4#wwc7ot*
z^33qsYp-(q=zte5p7FyUzRha$HcCl`qY*o^9h5R8M$xu)G!GXr7>>#E93ML7>nl!9
zFDR=P9JD_ET+Rp9RsH8SxBo21CoJp&TE?4*3lbYXt{wdG#q4AN70?=9dFd{$t`KcU
z*IBfT`G~T&bO<4~@kH=9A_oYO{Mxk$?{P8^5{XYWwvz-C&3qwl`bnCw?m7}<aL$pY
zDRo^(GG@Qvl~(wsN9nvLgzL<VC_%Svi}IedC<rbPq$J65d|%Lro_FS*!}RGFsnM)f
z3q-TvONV<ry!Bb`y??^nCtFrFVY8?ZWzEtzENMvujDkQ0*a~RySQ%(GS5)4zk`m_(
z(niU>2mvR-D;S23kx)#E9A`VO+7^{a{=?7S;)}QU(W@<s`GT@p(S~@yWJN)eBn$=v
z@;pZ=O;HT+BAQIAo-cB3QKNJk6-Qb|?=tU^iAE+SO7{Z{l8pJq8n;}KwiP4|K4`8>
z!})5>H@^Qpo}HaxMIW05-~(0DvZ}9c=7|tDFL?AM`q=pCF!f*FQFRV)12%|0Jp7~e
zBSjP=DIw58;JiRt)mzTQi;|F@gWVx1k#BX4jksXx6qwx+DoY5qAxTn#cO=CSCp=x#
zK}~#(^AxS}$WaaM26dN0B707vbI~5gm>8cGxv3fDe2d<cB*Q_xvC2TO0%<g9kz=bG
zSGQ=BFr3cl2)N*>>n+M#1diZ6&2~+e7aScN@Wr~unTB6`a>b<siDGCIb_l4%QQD3x
z*I`A%{QQdDogF^?+NZHD@Z{{AcD>-g`%AyXXFmHG{>5+o2EX_HACd%($O@!tVZA2z
zaiwoPY!QGa#WB*3dn3bFZ{_^le$HW*BU(?_ENJTusD$Zw!e}(<Gny^ld(t$+_j%3N
z5nED8tZmU!Ago0h!*sU8$!3XcBFf>{CooY+4?0+EkU$pKj)?ec@Sbv8v088XHSyEe
zfpt+W{%4S4{NE^Y{jW7d+!rE3eVyGRdM1#>2H3j5qO(|FM`$z(AuL7<2mx2t2p_G0
zqzth&OD}8?0(sqTbq?3H1n)3DA$Vw7kIXctNN>6bQpR?7SJiB<u1KY#EGx1sM~3KU
zsFk9uYqS)kqX9BW2%YbF>)@=zD2;dSrX5(X*5uL<Qq5pErWg$Ax{ju<Fgl^@0-r0-
zcsK}jX*A|eRA9c@5C~*K&^gH=hW&x$!A{CZOP<Y3-a2{CqmvVArD#O}tue}=5ZI>X
zr8_(PJAeCck)6Nb>KngH+jP_>;qdMud(#24LNo9dQEn)oJg4m}(_6cYW;r5Q_GCfV
zS+-?`ZEaMyYDG~L6vY5z67sx=JVOx`#Pa6kk1uqv9eG}G|D{)X^5J{DxI6)^8T%Q{
z`;X~Fz`bdaclU5sqlCZ+g-RrPX9pJq-~Rr)oXsQEYG=AbmgIC@hd`i}p$ma+wWTae
z>b7Q6t~kG1QZ<ecZthZlX_m+J_+-fwroYulMQ1}q9ejvfw;r(*5Xq%75;Tpl2wT%u
zUE~4?2*M#`j0Tb-^4@^Rd-Fc#D%uVk0UfNfpamvP3D(kycoQKZ61!zo(&{v!-d4!Y
zG8&C;0;?h3-gJ^sEtlA)C2=7deD*5eXql0*S*@cAcc8#%oQuSJr8OD}It2~RN0h0^
zGX&9ealYCzIU4hCf8lj*z4wH(MG5zhnAa`KvkPdu*oLx}W!bQHp0m!;r3tTo;pgbE
zy!H0mTsg})O?hqqE+Y?{7f%@q&s2?YMsrbZ**eE-hg1H}&%MfRwPpR}1s9hkZQC%|
z9g^gM>2QeAhRI~icDqGNjWG$s!H~173-aM0%J6$+Q;TR*<B96CJj-w{wiU{<WIUWi
z-zF~@r5Vr9PZ>08gl?l%1_I{-r)TGU|Lu3Ey6%SD{8K%61lKlRKheGU`2YN6TT*}Y
zh-d<RsJiSg7I6si(L9t08Ufp2$!u(cN)@<sE9c9vAE3(EG7}N3qYa*9cS1HD(k?e-
zMS-&xC1vD%%8W)p({%XSlPZBT8Y2w0s%h&QsZ@LmfwnAZs+vJo#F(s}r=^W1C{4Sm
zFj)#pgN&YToo`996s$vgXtxzAEwI>r9|*8r({^1fHXRrr7!*Tp9qjQ-`Gz|?8Nc@a
zIi=6IKiOfREzNp?tqd<V6;HdCjW@ji!yh4&B&y$?<yXG^1^%c1!~cZoz59InOJCv6
zKl*=hetgE|51(+dC^^Ve?v5u|A$YOc@L{=SDI5<o$>00zJzm)zGu2RFk!{JgU9sM5
z$<mbBbce}g%6K%pvDSge(LKMoA{*v$wu)E;cWsOFE-Ej*CrdJ<)|ez?y<XF<SER)r
z(VA7@FoYW%O5l1Vf0*SM6_X-v5ViOzHyzKwpL^%?FW^w%TCTa=;$t^M7X%yYxpEF6
zCA&gX1QUJXM9FM&fbUhWQh3l&(4gbp{v+SG;BbUU{jrvSrwN`Uq(~p=tc`s0R3lAP
zrQ$tJSz_x3r4#C^W;7i4v0E~}Zc5%1gVC6}-y4aS183u<M})|CPNby1x<rqMG~N-E
zCP`8flOmBE?nmBu-8Ag%98mkfkB-l|sH26~=WY+#8;-eMD0U0j8x;hj_=3><&9e=E
z@WUs(y{Neeg3>!g=DBnGfMVwkZ+!OiB)j_@78ZHFOO~YE9*yv2MYTO=ezM^*c(PQo
zvom5c7;|{A8z~eL>Z;^me+S#uY|94kMXzqpVn--17!F5lsv4ao@fE!Bu)Qv)z}u)q
zefiZl`0&xYJbC_@y}IP+=nlKW@cg|Wk?y^MHwTE&$DM|N!cliE-+KFPF4tR(k{q93
zV1veJ(VNa&u9geVE-u(qRTSD0eoScR{+v0Ze_rDg6?T<06#Z7@@i&`Fd@-&+D{fK>
zfeemRgLgGamM}~TobZu^>OD={5`-WzMLhJ{jtoUx*9g^%G!lvP9-;aLpafmp;$^fw
zQ7Gt5BJ)APda)u=Cd$m6W4ql_3<mLLrBo~=I~!r|B&r0pl5|3V^Gru$mg^M+$zWn|
zw#B-Z?YhGRfeOUXX99VWVx)}9xk1XRs%S4S7!7j1`pQ0!pB;02e#*gQN^<KyQfpEa
z#=_=u!MjhN@&3AH{|jH>-~J!`CoH;#fA;IYh8QY#_m7xAKj&}0^pLmyKmTv+haVEE
zhKsdB1<5bGexI+savOK@oYRXF+SZ{>#=U#@$cKiqUUBQ_7FAW_J+!Swh>nAUL#nEx
zZCaGp(HcW4d}j%&$4h&aUY2DjrNMbz(^6LzY1R;gW{?-u%jfuNg&cJFki-bL&T;(W
zgwxZDkLW$&dU1%4TJ|^V<r=R0_{aY8Eu=rm$hD+$(Zt6?^fHqI89Wl-GioJT3M%he
zD*@7xD@mSs-h64wP}>;6<3Q`UHoO9zr$`~_nubv^U{lwq#2}?$kQPWKsYQ)%x@h3)
z9Ik`;ixb*y8Lf-(<V6wN7K5BT&1vc;nsx&s1d=SHT`zHc>qH8L7uR{Afh0|7nhqHP
zuG)|eQiOA~iv{cX6+Q&UfEi3_tCD0eU_2b-+M2&{Z@|m<?(n@2PLOzpvmNI3kkiL6
z_*80mO$3mNwsS1Xnw52o?i}$y`Un3#9)9jKbS7u&BnPwGWWy<6efTnu{>85&mU9O6
zih!aRWGLZzwAu1^Uq9q)y94U8E83>Rs)%52wj0uM!v4V_T4{#EF>Tkf+HP(HEzZSU
z&T$qwiBgDYRB8;ijt0ppM1`>s0;x5w?&!8lOtnUsA?Uc9N~vyYv*-=#7!(Dg!GI{q
z;>T%o@X=cN%X#knsfQ2H`Dm@h4c9K_b~hfD&XFk1SjCGAULqtUMzA{>(p448NYrG2
z$r74%i7i`HZ;F4N7Gh-}9p#bB>^(t1=Q~oT5F+5MMI_OxMhekK?M39eY-&uB(liZu
z-Xla(lIA&GSyIhc6uVQTN+PjTi0H=`e4GnP(OQe}F`4D+;*7!m9%LC+Wf>Pk@-#<D
zi984_mRGpyirdp2%7a}_D~nN%R}Xi1b$@^nj=Htc+QnLAA{f1N!2QELzWxVq@vY+}
z=hiZt-Qqv{U;S@5U9b7p_rA+!x#sl63--nZQv=4W$ww)74sRn6bgjiY$EK>#I-wYi
z3Es0_F39tgvaFd*Ml^LzQjDnUn(ek?XJ^K2w!><*MkgkoH*iA*xkl`SjOS7Cp7ChH
z{Rgk`{=086zr2Wr?d|(yt2Ncr?@-+P0$wBpBtfM}FPUF0`N6v%P=`Qc9h(==d31an
z*QSirA?y4f1|5EEP5tFuT%V}0n;hN2ou1%=zfn(${t9`+wd)^*a)HDMG9xisB8)<0
z2`W|SG+Jy3YmritBnBse*3w)qX>3cP1ST`2lQE7!)6_Q`ni7#T?w#wEx)D(uj7E&c
zCH1E4gSX?qOVczOzN(lXc<s<1l87!kZ^`nUYPm*e#m;obd_Kq7mST5GmZ-?TYhpZ2
z6MFwN-*Y>HL@L2xl;G+u+trq~*)kpva3L@-IaVvqHfxH#Lq_8nJ9qZj-AmYAwG2M@
zDxdn?XDFqi)hXZp-QQw9uUV;t?7>~`yzw%t_KHQ-veOFw_N%WkD>P4j^aB#vG1;B5
zn`dMLL+~})cxGuDi&vAfTrSyeN``|GX_9eub&S^PCgYJlNkD{vNh4Yp6DbnXBmwL2
zbOh%qs~A<Kw8jwVwrk9!!RZV<G<C!CXU{3OW#7-ZzEOYt&3$aWe8idgOS*>s#6w6+
zk!aSfCtg!j?+bYy#}^UpOvfm$##uH%Atbv)&B1=b;9!dFTH56nTXzKO2rBY_EYW(w
z*=X|;WZV=L(eurFht`rH;w8sc72D;K&<l+Mfts>Ub&yncLu(zmEMgi--(HONHnC;u
zyz9Si%tjS5X5FS~%GKos$yR}Ptk-L-i_wE!*O35qS5q!)K$50IR&B+>2>#v|U*?DJ
zJ>vO`N9^3W&+OCp$b%<I1iP~xvLvBet+?nM%fXC$uir=J1>23|+u!;<PUaVo7kuiA
zpXXD%6E2=Tqq@4_;``qvTrQah!Apta^_>~@@i~jLbrdZT_k+FLyUg|r@-%0@Sa9no
zVY$3SAQ(?4Hz<Bq<Rn?b^{rNjXbWR|?yu<M5mbP-ZOM|1lPB+@bV{cN6o)SnRMPLG
zv7s7>Vd%jHw2Wl-5Tbod_^~zfXT*#9OEy}6?FUMXv6Ioy&U;B4Vu7?PB&h&YwA=~=
zhQoqH1(HmYkEfX3F~%gw^os4tIimKc{&n;Mk|K-#nId*}gi*N05>%up2q`{lH26rF
zNsAoabG^6A70cC%BuU7!jH;?gm1MBDM|pZmb-5tlo1s*+_ej$eheRli*P76ExTfpr
z5}s<kV2~Z)d|<nnlbVE7D`YW1tC*^^scIhX?X$bHiwd5Btl2K+2$3*JGFod{EawbF
zU^W_3A5Qq{XI^2k`Yu`7Q3%JYue{8mOnH9tgl@i|di03#OtL#1k!NF8s}*PGr(|gw
zIfuq@xH~}@#dfnmC(+|B3n_%C1a;2S)+JhHcwgddz`MX^yX9bd3)k65Loo?L$2@c)
zqtaMN*(0t3lO$1<-?rGUrD+?+NydvGd=p&FV7!M_IZPkW*qm>D>$@yg8?IaS1dj_I
z2XucZA@cfc@_!}P)F&$JUiyW<&5Pgo=d|Z%@dXocLlOvrh@MhXfb<R_HJ%bPDwrJI
zVY<IVo@+#|qVG@D@%?XooAIpR@ZLe>da1y&ZJ6!tP*)W=hl%#}BuRlx3_(Ptq_EMG
zE?6Ib0x1zm#^hiZgr}-w%vho&Wwj=m45Mql5(p2WwTLXlEYcv*g~rz{>!xL#57^z?
z<Kp6kt7i*}@sM;dLZ$;O0^=;f#o+F)ZfTo_*42;+RFLH9h`Nnd6uC)RFRwV*+ou=|
zND_&o=BnIa_HVIws3;SURF={BfZGpV<2&E}eVWyVEJ;{wHY|KV?e6l*C}$^2@rxDP
z(<Lw6xs5OqB^=Z7h|PA1)ER>;a`4vM6=^y^Cn>A4u*B-H00961Nkl<Z;pN$sI<%Bc
zLopcQyhkab!l|o~m?$qMMjRQFl4S)n9bMZ}ZMI~2LEanK?j{+h@BENkd#~WcfZz?=
z<%Z`kPH3#ZDO#>|YjOM}Bj_(9gW*p%wt2>^=`c1nWpx0MF&az6Cu2jv$Y>?tU59lY
z7D+zIIe6(c2IGWgaf#H5ZndRaZ6iV41%kIoA|ie5obS9v27~Kc?Z#+C@2k}{4PD!j
znw0T)jCCzlUGwzu3x-9`ty_mAT2WV9l+vWKfZlp2COb={f+WpucxKkO(R@W}v=FT4
zS9EcwM(YG?J3{XkA7l$&M0<(4#H60>@{B<-<Wu+dQK6yQoHH&)j7I|$kcNg#Npz7g
zP>PGL;q-^!CEa_8@BiS3y!G90P~<tEesGJVSu=WYm;KLv7DwRmuOIUMul_?umkp#T
zyE0+5*?<h}-`OKg6HIE5iDft_=&WbCsp$kHgM{T`P1`vJqhaLkD)F)QY%J!kBM^}x
zDk5Es6{@#eb|+J2%TwO|um2%(?-nor4MlSBDp5z)S7|{IoaFfVF_+6#-)oh8#P*?o
z|N3$ii@u+g(K*{U6=Kwt6ltV@5DPdNAi<$SO>haw6bPg;@*j)+2NaVT!Nlh|9PiO=
zx43o<F5rBm9Qm$8Dv2_Rd@{x<MKxd3$tbMz!S(7;4@x4niI*E8Nk(Jx>6GQkDOFuj
z6a`(^V%Htxg9DQBnELXPuB=eQh$@;ajl4}|;$66`=-S9<58h(S7GLJ1`2gn~)n-kL
zziCAqmC*{%YJJrkEq83JLt;pbCU}n&fh3PYIJ`yJigcRt>irq!VncpshHqCqdh2aA
zC&!#zp7M$m<bwh)1FQ9d;cysH`Kn_7@PH!E>71wQy-dBdBx&Ryriq~}ORB0yDaUj+
zLLgYL*UV<K8{wntI;@SR#wJN(G1s@rqrr5fW0l)Ab-86!3}gC4h(3&iwdd{M`8D<q
z@3Z&n7ck?8ym<P8|LULrI<0m5$=1kUt#jy;6?R|x3gYsd$G`KNxUNKoIQjiGQrA1{
z$GSHjET8$zXL#wQ*P|1euW+sdAK1><oIE`y&+kPpi4>g9&l&7Yk=hV!OV=xK^1Q$s
z4Sn2Lu%6Jx;wprAD?@-V29xKEcczq!6~nX-1n*l<LdeJmbRqs*?@?ODms}~*!5{`7
zHw~j<!Os2xm*>YUF0V++nrzfzhGVob2oqaVN@<i5*lovRxggUCIyXonUr^$4!E&{@
zVt?-dUzbdVIX*R%0YtSSm4f$w_tzOdc#|v}Gn<Vm`{<F2C(p2(5>a)ec|6Pq5=Oft
zmg@zx{R55;55POh?S?@<z}XI|3`v@jWdklRwj{}f!EnrSeTsD*CW%Oc50Sv@ox}B9
zJE7w1sf{7aa;mD0fA1YyD}<H=B^ag&KX~zsy=sHZ;`7{WmRw$4VZHy@GWmGV=qDG?
zU!BGDuNvP<a{f*}pzsUUb;CeGRQKI5?En#tS3|h@NDzXD4<AzG6N0UH`u=kc?#>t<
z>>|?yd>~0Q)p|p!HJ!7lM8`Cjese?!(!}71Tvb89WI20xj%cfj*0&&{B}1BJ%onlU
zI-QO&iAJk{?b<kRpbtP7ND<eh>mB@qzt;K&2E!q(YtT}WWjRVmp^<eCuQaHH)M%Wt
zw60~nx?(aIvRSW5hl1H`#;CJ2+bx@U$===`u8yYJ!gcHnbF!x4TmSSQ<Mtl%?hoH4
z1UNLF)8G83d@#RYdgmT*{NjHLm1V5EmSlg#)WNH_ZsVGg&O4?@6Lt@V@kVc3ihO`n
znx?E-uPp0jU_70%+$>nF=9t9Xh_XJOFMOj44gqDN>ey(56q2SYky3CtnXxQad~k6>
zy}lyZt)dAzqHkwQs?CPq`1<c~xmXg{hvOqe0j?d~e*9xU8zT_?{!DNvRHQHvy+3c-
zsUo1V27HI;O=Cx+9S-lm!C)}M*9}VdVkK!9O=ryOhOX*jicw++T?+z6doxs?QFa}2
z9CKLXnbEIR8w+fq`+}^W56WmJM+YoVj<H>hQG&Xy@y-%dL>6t+;tGd0N!*jQL0+fD
z7>yMoW=E$foAm;JcE)hFLo%2Wl*aVvt`ZWX6U25yT}734kro6kkTUYjt+U_)t!*&M
zfO8~1FgJ$$&DYtvS~9+Mi|_rDU**{c?=mS4uy<~8a`u8BEm}T*|A6i0lx9BX=-`mU
zy*&g_H7%%w(PWH}0q<&T<FLKrbX#umLXxE!ZP(IShZlk_*m#lk9-#zHQxd#IXKCbX
zhd@(TG3rfgx~}E&;tVAPc`+a#jFA1>8fOLRO3=0eF}p+43jWLg>L2mw`7z<<-1*rW
zpQy0QZ{6XQul;S7XQ!P0;4O65B89~EygErA`2=_&*grhtwO3w8v=I7=KG9Hbx12nC
z%C>jW$2v~um&^|KNK8UoRmdVGv=)z#ZB=)DJ63op5Juyrzi9`@j%XlQiODjOEaoew
zX%^#eoJS?9cMXpRk@ud^7o^U6R1%A!Zd*~emTWj==jbk1=O@txzq=so8q)EUBukM>
zk?5G8nF&wo8hq!_%Agd~wxe?$D%Du;Xu29H6!mhA34y&q%E$(;&W`!!5B_hohj%fz
zAJ7HO;n6;LaOY<%KKOvWZi^Es#bm@N%Sh9d>D~^*=@cy_XD7$lE-LogwqraVk{3hH
z&q_Xc?}Sf(_D#GaDtM(t8;uVh8R9c=z1giGwydr9XemidLPfX<HWx<6R+9)AElCwL
z<%VJwRcPC0%X+)+XT;Cmg8JjeKYRQkfA2r|k8p2&m-P4qXS>KNQIg;VHfU@RRMygk
zcoE5^;^<(?Yp=Y5aNwn*+*F*s=(zXVeX^Yq$_J|T2B9Q+kkeKrX=3QQ2J0PWI7Awa
zD+5BC2n_jvNHdDPJ@T%J982r4o#TEUxvI;nInPhdY3i2QWXPR+cgZIjsWn|)V~V0z
z_x2?Ap1Y!?z&S@g9HWXHp$tkzE{&9$<?<5iI?^<vEl7dvOpGP80YwTe6w<Jv#nK@J
ztjiUtFm$etmk`%5Fov|L`2O$z1{<%Lqz3oFkI*lku%{*6(+~N*|NH+1?_nrBckUkW
z@|{D{Hc+mY9K3vw$*`d9)(kRD+h~#`qi!7+=PQ=;n#b>*@R`qln%(Igj?dm_y^Q{k
ziArwfc?j{KwoMbmyG=BFmP#R%q^c?$(VNi{80_s+jHj_xE2G)8?{BEjUp(P=zWwdK
zb^Vj$<7?_?tEdA=W;57qaBYKEGSV2l#~?91u;7*E3t#*qv)O>hkDuaw;OD>kC1%4t
zLhT5_LD$hW9Zl=lthQLG7~h$ojHbC-B7#CE2AwB}M3W{OGamG@fYIwSzV6r8)zEiw
z`WAeDM&*M7*Yq|jQX!Sb+maxoJA<ujWUeqe1&E0yO5boZt&mI(ZV{wnwVY$G<_uA!
zqX|Zv-f}G3y2xZo?K|jP%x}L|w}*fap4PUsZNqR}paLXa#dKRUFFTHkoML%KIDU`e
zdPAKHo;+EyT`sspu#^c0tz#5QcBTz#G^XiVTGvtJ1x9PM0H*_k!GN{xMMT=5b%s(2
zSBo`W5Zt|YoB4W<GV$k^N+GnOs!FoK0HYI-V2pwg=?!h;smm>Sl9T5Hgo;`3N(x3v
zLXJl79ifK%JbV0tfBE~r-(QY@wh-;>@repMSBxnhzRKON`~sJ!C)CFkT0}Cjx<Tti
z<cmnh>z{f^ZX~|02rhu^aAn2j;*1whpE4d6NCfMv6`|?KOoB8T$N_<`8=AUextybu
zgft&u^86zYts4bRk06AA(3-9lAT^g)3lgQ7PAAbbu2+3JqUmjnG3jMy{Q+yz1mCo@
zwgYQO^8tIe?qaSkSS_xo*DD+W+>j(Ikj9`=L&AWObF8lBQE4k8ZfR{pS#8luQ#B>y
zVvMaSj5dVric#?7nIMb}Ke~K^qob5F0Lwr$zpnHJwrv@monw{@?%bYnFw41ha6q2r
zly%K~y})Fe+Qr*`n)MKPnow0WS62%jy?=uDhMk=$Wx1rRD#pbOrF0*M6!~>G)*}6_
zS0bW$bO^L<M_E-dgSBZun`mz#6qBqV*vOFx0vg-Wb}jy|i@HBFwr#^VTg(6CKmDKa
z{OiBX#q(z@&rWGq3ryrr)A&H=9fi>xOh(+lwa4H3`8S#71;I8rTBH=r&llW%aGNwU
zY}Q+f={VZcbe5bzy;-wbFG=!@d^$nr1r#1{UF4?rUR+3n)EX}vyo?@nX=2b>#$<N~
z-&v~7mKRTtdH?YlpWmC142HDZl6tiwn@r+CM(<;&HA#`P-INF$kokasz~K-QiqV8}
zv!X56nCT8k8Bt>d!beVKp6Bs`;yapZgYOzf;|Yn<=%8qXM`%IaRcH}Nq^G(%!4D5%
zy<+vj`;6NqX^}882}e31HG(|TOw(Q{lz8&lkm>F&c_Jw{(RE*_j;0lS`1p#`lQ~8u
ztXFH+%PqSHQ?j(6s<*w8KjAtp!!|9|W(CfovNU!bghUFD^R8Ds2Aps#t>@NjpG6O5
z2&r(kBZS!6w$}6hhaYf!aT$5cH?!xbA>Kc8l32d_pZzZ>pFW{lUePr*>&24wd+(v<
zOA-<7g1nHt^6*ve-#y@SZ+tek>?F9V2I~>Jqg*a|^1)M1UYtO1>>ci~vp0#>CRsvl
zZSO4??YWY~;0Q6F+D1SjqO1|6#$AI1(e_3sY0Tr5nzpN>*hs}<KYE5n&T6~f#2n|r
z2$du^s%sG<*SYChiot-L!&_)$xVk)Nv$&#j7MZ3I(UJmT5~NO;j%PI0mZm9b+L-<z
zMRb2Cn+6}EQagB0U2WN{SERM2unB`iQTm!qwWY{gcC(bVF)XCyogf%Y4@fRQr0H5h
zh#a}Ht;mN1AoAs1FM2daqog582ArL2*i;)1Z|^W1D;CQOf)(uS&LBj`1f8aIO~Yoj
zWR?{K8B;MJUXGF^BTaLBRTHcu)0)&Mf)W@n$nt`8=N3U2u2u`yW%+M(p8lI2pQx}4
zQsF1N9RA#wd2sTAcmBl^wl%`{6A^L4rWy|v_il};R;MIrtPiBan5VipJ7aly$-{>a
z8Dtsj`4ZbTERWC0CPR`eMF%L$n(b!EYQ01v8H~o{<1y)Qh|ID`EakcoiJM%EFf*VO
zCe3;N;ZyQ#K-!m_QYwOy5#{r7Aqp971_T0a3_eZi+M2d05mKOa!esXlqYd+`bK2Dk
z>0E5Nggj>KB?-gf7~zPSfiXZ3*R}Yz#S9bbwkEhhCJa^^f)dC~<6J`~Et9TfdwIsr
z?IW&y!+dc`d;FAp!Qn#<BBS#$Bh(|APG@Kp4K}CKDP7x8S6jB5HH*cP(-)TYrsA`o
zze|!eoSt9c9i&NilUM4TjZr5*QHNcd0tW(3+fvt-^>RgWa>i)VkrxG7t{7)SoD(QC
zmgsHhu0<vMHD>pdV-w);Z~P)hZ+?!|=?SOryv_IjzrRZR>;*muIv*J4n!o!CZ}94a
z1IB}tOiNr{qKl5UT;puZ>GLxV_jZ}>jw#nG<e;HV4YuoWn+j)JeA}^YTL#M&LzBjW
zqd$Ozl;EP#s6b%SEVhB{buhW40~9bC7U(qPo9`^S``VU+oe8$HwA&4Y5R1D$U)G0!
z9*v^-Vp}17O_HX)nX5o(7)>WE7jxP)M`r_+R3s{4kPRZ?+B&q>$j}j-B~ylhNl2uj
zYGVqGb2UV5K8XmJdd<$QR}fNCiok=zJ$5F8n1rJxqv?bsEifv<cO7N9!Hd9jHezvc
zLGS^o6-E#E);Atul9X3oc^U5_hj26=5@;i5QL3I(7-Oiqx}hnnh}I<uX`V+8p@^ol
zRo^*q-XV0{Cu7!bhY=u<2%!K?U9Wljy${&d?Z?WvKaR%mhgjf$rN{19{}xAo^OsS+
z#noF@msdRd*0*^4|NMVomL*0CI`0V1(ro5YIVu%CMC**;9nx)i`u=-7e(x!J2Rj_y
z+GjAzsmqed-VUA8G;K$alDcZp&JoHj%d1P;O^GHVvMNhaNrsRLtz>^e5_kz+i2G_&
zM}uQy5^U&&82wvt4%>Bzsz#?dGD&0M*IQx<h#XrRsWbUtM7CMe)Fl-jQEH@&hS?ND
zl!$(#(x=qx70bmUb_Jv)NI^#+O;UpMw4r5Hui0+5bX7~$C^{LD{3Od5O$QX~Ep#2h
z2NvxG?{3!ockbV&7$vmU(%LS@fZ7)4JDfH*3e&2psjHH+vzqM&zWfWHqwHRAezsv)
z%oy!ObToum6icm9NlIN+k#9R36Ei=-x|shyoz6I4Kcj6MP=Yo%eAl6UU^F^F7khZ8
z8BTYZPR1;@H9xB-&?hVGLZCGX*{wUg^rbIz{`dn{Z@nFFAR!(QLP%8b9PH%yc0n5~
zI?EBpP<17KHD`Ho&S;oZ6d77->TSt1&nYji=-LJ^pet*9+2BJ*CKSOs+@_=n#54-)
zQAyG#OnjtBL&_MRr8JZ2go~3iuCC?`_GYM5$Ch!A_W97`N<B&?0I4-PNpM|7a4pUn
zgwQCNFc|M(lw!R&qb^G-2WH?ggSe=TNf?hOG}V^2i<3$K+Ci|EY%oBEXp7)&plKbA
zYiYYJWmUtr1Z(LAOJ;Wu&<6Hp;6Xj%i*MW|H41_z1cjCgiA72uJq5dt<zmiyxnR59
zu$Zr?%bZ)cj(GXi1Ip%<s#%hloFvJ5K2Jm~Lx`xtbuo8sUn4|pStAA7Nb0h}mJMFl
zbgjp@j6pVJ?F8NloR?AGDSoz4gg@!HeRP}u_&@$1!(fb03(RoF-Q9iW&mQx^vlH4N
zu)*`im!`aSCnXh2s1vLnqVo(}Z&;ijvt3*wd`Gh_sZ)dSj&8fgEiHA^;4Gv`LZS`Z
zYRlEL=U_eAbc!%0(jshc((Pk}mC_1&!%i8%+Gz8D4+Ph6e6eBKz_o?AR0<oQtty1p
z<f9Q%E4=r(0NH4O)0(#0Mn0=h2uWOPg8|8Wjo+4drO_rO%~BYRXqpCy7YeOR<WXrs
zaGts?2_ic4JKvG?_C=Wzm~zSd>YTD_kxfTk3FdxDRv140^6O}sv33qGATt>{iIzaF
zYw5Z!$`PE82H)F~uYC3aM|XDdPUBognrT+sGn%GiXL5wp3TI=|KwDS1u8W-urBFI@
z8HIq$^Ha)gi9peo73pNiOXCqA{OJ4WG-v<iDPC&85CXh7KIS`bzZ(mYo0;>oc{u&F
z@y+jigJ1lmU&eZk&Qm6N4nlD8?Qc{6-Z#jEqH-P2pMQvUIay*b%8+Mcyp`x2>h%?;
zPoFWH4B6kAF&&M#JU^ir4rsFsB_w&0AXJY;caCPgp_^|BF@}V;+92v2Q#gc@2(1ul
z25CYF3h4#Dv8=9^gr;RWostfRH@r-fB=|1IYSM!d*9HQvwYa9kdkdn&Ig1JjI?WkP
zW_aguU5B)m5Ta4F)(OU>1gQ`vMI;6nJe}?0MIv}aVvtf&H#G_s5j7uZ@kp7*?$cn5
zGZVsi4;D4I_HQws?L={ix70()-u{SkQ<3EZe6W#95-cG|f&=ej6z2K)oaaxsy#DDo
zk=n7mxMY~k7*D21r65F#PY99rqm-f((AG8CD5}pxZ)+3!0=8{iw%aYM`2sBz!aB?-
zA=d-S^HcQXHj}*rKL6%xe0Y5RS0=USla5bT*x?a6K#`E%yU+etzQ*=!jy*XhxQ141
zT5m8Gbg0pE496L2Ao0N>%97>PCAKWNfA=;z5qRkk+LPp(c2ncZEltzVwQWQcdXrV9
zB+k}!WsGM@Rg7fm&9ov<E;ulXG^EtJMH|m>G~~&Pnp-b7q@*Ae9*O87^xz#z=o{e&
zIzVcmO^`Om8{s>gut+V?(=k~h5$8)N8<Z~zQeg9pROc~IP!yzA<2Qi>O)29oOnXR-
zM$7npy2fH#5LS|OfkX;yk)Tq``uGV?-g}>U+pu^47CN6&1kWmTgaEmZNg8D;kTxJ(
z&3bV`z1rd`ODF@oQ^n^#dmkY^^W~b>LXoABtStq>y123QCeyxeNed~FKE&8J=c%iP
z;b6#MFhFTRz!HRFTQA9rgtFdZy`w^da#3k3J{~^)S4(m9ryc*r|L*_D>knV0?mB#M
z$W$Rm1%vx96SQI51$HJWKmWy-*xel>g6$Qq9TrD(al+Y?CunbZ?d7{19>k2);dn?}
zmq;HIB%&!WC?u&igtBIHe2VWJ$#~M6QeUI80TFyG<dur8L1*zz2c1ROmh-DQADpy2
zc(BLrbU?XU(^fSx3oRP)+Rma?iqg90u7V!qpcSEMBE`L@O(+#3^vY@z+f#)?7)3HD
zFiF~P<UQ|oxuB`HWU24{_qNDTpj3+UEj~oU<+N!yd-|03&(3*R*33o2rh+$azlJjd
z9xW>991k;)+EOGb)}=TX(ZTIz&9KOEzNW||<08em3L7lS6jY65vAjf}qf$J?1E?;y
z1Q*+^A;d89K<on8uIBvoB=Q4SE6$(4;6UEv_IL-eUU2gG1KRzU38M)P$$UBIH-7i)
zJb7{UGsp`1lZ{{f$N!LD{Ka2Jh-llRl9YTrVtD@nSKs^=Dg-)f_`!Q0a_8uccxkm2
zU84{x2kEI+a}17gkwh+ZQ<A3!-_%@w_!QSU5@XO=f)uK6vu_B&qD+F4fzWpG2ND9j
z!*>p^6v9M`i*f=w$7-=ARSKa@OlxtDYoAadL^Qt7a#Whcm@*&mZHMm~ppDtX1f2I6
zsmY6R6y}8v>ssn{jcPqKEsh#f<RscK7|*DylI7Jo!Ffcu_EnVFV9`<|gFy;SkqTz_
zUZPHh{OH3o_73+*#s}PbSdh03b=Tp7Llpyl*ls!6dJc=6aaLf22j$~@C)Nm?)sk|%
zWwWXgLa~21!ut;I1)Xi8kWosUv+)86i0eG?aGhUtO=tS3dnBc#Zd<y`D+X!7XgI_=
zhe$FC>Dj#a5iK2~hhO4rU;dl?)8G0A+v<<WIQ}a<K2c$Z2uKmliCvm<@Tu33e`C&D
z|MVXtHzlbAgCo(Ndk>Eo?@lm7jY|TyX{a|Xi}{kg$eHcVSl1ENlGiB|N`P~S;L%F<
z2ThDQk)nwRQ`gc}CDJ4j9n4J3cNOvA5C}{#)KcLDTy07opVoZ7>lpS$3qtnsJ&~KI
z!p9~NAhklLX`Eo9CtnJI>yrY~!GtU%F;BB8@tX~afRJirqESX8q`)K=ZOTZbz2+bW
zXxj$cwODIuou*u`5x!-*e~3^+2Gc!qp)kpq7RQTv%|Ab$^JbnfN@JM04S~i<+Mr3b
zj#-5LP0n^5!$Hoy2e+6`b1t`Y%Bn?4LzZSyV(wgzGQ=BW1VY0{T)X~H-@Bn(Yop*K
z1gv*xJZ<GDte{(5k##lb1Vx5X*<Y7(KV`h~+G}*aMg)g)7TeTR>ovpOT|%D0cEcO5
zA8>GIic*TuX^gQ169dUFo;_h$6zskHGHFpDl)`Vepc1@nF{!3)8w7!_sj(KciV;b+
zE@`(VCNE;KC9k=q5gqfsLr9OYI{Hj?CGBR(^OGgtdv8JGG++7r9emwj>jtegs?Rx=
zTHoY0OJsi-#i9CGI^lh9q66M}((#DRc1=BB;5$z`852y8xc4(h1w~p6XxfG(%MpS2
zd$kpcl%O-JuE9H?4HT0idV6_?(h2t;+-16V#5=2ocWTeLY}j?4or%Gr$@7fOrb1~=
zmgVRqs)eV6$O>F<8dl{Btp(mYs=8%R?4XUowk@vf;+k*{-T%&A*Y?&Dk$aq`DO;m(
zuI1|VjND|5XH#w$BWf?PWkolDbzAWV-~2lN^}qa0s?PrP+W8|!M0&vEkOYLYgg(F1
zYf0u>20}6~Jxe^rAV;+w-c<;tsJjllBR2{uJZYL>y+<h#5$v`lI7i*qu#GvhK173j
zr8P?Q$rnPB<OM28@NJFlEK+rtJjZ)L$3kR~XRMc7RHoxO*F|IfAYykzDv3_hi2C(K
zqO%<ci`EHJ8HCh`5VNL*NhpS6yma`wCas{dC1F!imNi*3BpXjDk^;MPKvR`en{^ad
z5TyiKD-x{;wZjRGHkxcO;0I5atV@rqYCd!4F59aMHqRb2H6T?==^V{q%GbM+H<lYd
zl?JpCbjqNmLWjU+z2xHfIpt!DZ4__3d7tU7W>c*(TGMnLP1{l!(|3z{A#qQ^2ykPF
z{_(u`i0Z{oTb5-Y;yI*3ibkMq$89AzfBp`Qk-Yx!W$qm8^X`+UH*@D_3BCJdkzzui
zgpBa7_YC)TNPhk+93Ma8_#59QZ959rGTceI^YR|)SfRAUI*YA3%H;;@J>%Jo&?|(h
zra_woka!#DY#Y%Asrp5%d(~?oV53~T+0t&KB1(z4xZJgMhwu;#LMl2mrPEwify#>h
zMnErf|44M<LqPq+@2O4FtBC?W2qZG939SR^k$FaTu!Efo+SLlt1;j{V26=DPs?d3X
z8P@1+Lfy0^bxlY^pNS`;y#kiObi`;pMotg-{`Vd;=}yRQ-RAS3|02Kl-aCYyU0&d5
z*9+cgI$kLP-Z;9T5o_>8K#JHhZf%FqaCo@O!R;d&Ygw$8c<;&b1k)obzPFN5^nP>K
zg>8>K2qAiIcFYj<&a+;v2|hkMDI{8Cs5XGVz%8C&rVq(<L7EJ1p5xEbV)|E&&Ud61
z>h+4QsX;{kmKYBRX+kDD?!UBylMM}#3!2IVVFK0ZB~lwkd;6TP*O<18$n{{zwknwv
z1*tJ?SF7j=DMWmoyvJFG4-VT@(Z)fhJzYUY<O7cn4%G#;)pP__n~JZ0=ZtqhtWa9<
zwXfb{e_Bv4R%m0;Y4n?la=+-74?n(n-Z;r?9oLPIC7X91VGPB1Of_Fnua-z9Nv0FJ
zzKgGg#Dx&8S(L;9d6q{<|Ei=Z*9dZ~kZh|pt}^8JC)`@i`P{?XeEH3LOvV!yR`J1d
zg&XcrZ7Sw<#b`I7=`3<!$VUURJVy#*{_>eZB@lFAzFf1dw<s0YewGdx4u^Q}v0V$!
zMx$B}*7v@0&PHWp0F$J=^zc=(G~?NaPY?pys-)GL;54)GjK%yECWo9oz2qPNli%do
z+0|bcb$`q_IyfT8fN%j-mQ+ta<oxX)uzdf$s3Z@8%r!`n)5!#90z#%F!6LN9iH^=g
zQsm^*F)GioO^uM6#<n00iH<qT-uHp`$ezcoV|KUekZF#|2Y3Rm>+nq-X*Yuc5ez;!
zPzp4>f7<ZU=LEsU#G{xo42Yh0+KT{%5I7NYHx-fddL8+A-Rcxlf>Jms$vjkfjcB7d
zLnujGS4d?@Clm5wNHH8!SKFKRuSSwc3BiJINKzdgCTvS$G)@GZ@A%g5{!5&6+!|y^
zlOnPKkI&~gCs<?!-|HH};*{5JPe>bsw;ksfmz*6Rvzo8SlZ-EZ;Q?lnaB}gKx~Y1V
zciI~m$EY<aVv>gt5WMTpUIRkjcw;Il@quV4EhJT0vPlwBm0%Ks2pyu6q*id01>Sr7
zP4w#nZr$49M;|`nBZ1M+j#I-YL+twK59d8j1oR*$&F*pQD_>@Qd`$J|6hp_Id$$?Q
za=>AmE^-MrYr=X?R%8^@30_LlqM+QaNwq<mgrE#ksOS@=dY`DE5JKWz2TI_oHMVW=
zK6-!}dEG+Qcp*SWPqx@@RG@Qezom8I##%zkkKRIvHY}lEK-V8bN`y%eJ@-tAK7GOl
zgs?a*kw!Dx-6d6q^?c4|Ij0DYY&-&~k;Y*1g5h|=<;6vW@pXch3K2YMV#rMzKb|Di
zI^)^J8mko9g=HB^hC;EaJe|q7Oh>%;;t53?*xk>ON|CFC)_48QM$$Pbij>*m4kA&k
zRtvUug*JwyS6y{&hwo9*>%vWI9R(F4hFVAe<)}0+vW%{2an8|H6;+auB`JXp6-qK&
za{25#jK))>ZjoBt6m@@H+<i1e@N`W>RhLL5$%_G+9G@!8-Y)Xu5}%A%`GiD;=&COq
zrVcdgCFA{ltTOE1x=mfxv|U48Zy5~>hT{=I2*$g6G@CU9i4qd;S{xqVSlo7t+iv^V
z9!cU;(vT4FSS&IqToP}`n+o225!gE__{Fcj#Dm*AxU!1gk3Clh*>9Ml53dh#&V1b4
z4I)2A$j~F?4*KBwbUeZ+&33+^+HBByhRHKL*DV==?nOye-QpY+c}_MMg12nT8s`PH
z4oOP8UJ$Acdv|X!8t-8{&+%$YKq1PC_570Zpul=X;xuxt@FuQxFCfVbqv;;&RZZhP
zwY8*qj#d&Q4S70<JfO}ZMQo3S5Pl3<a3SEG?K$v*tQd0drB`U`hV$doD6Ys7+$JSY
z8$NjN>sW30&EJ2A@4ok5k6`?G3+k`ON8|48JMo!2OZn_6Pk--s_~0AgWcB=*)^;ca
zA~761xCd#5Os`W?GGq$Prp8uVvYj0Uhj-Yl*CbLho*gjUaW^@ouI*^HTbgo%cNP!?
zFA28B1u#j57vSoKt||#lgA{_)#>hWu9X14>U0IH|;DyG!k5uW`9;`x$i2QaPx;N)F
zX%<mCBBcdEj4=ysv_+C2y^h@0!7%#Md5biPrtV_9x3^u<#zdw~Z%bsYjr+fdYILPM
zNuDB{<>iABkDp(#TAY)PGLV6`F^Du{cxym#jx0~HLUCEw{BXJ9!R`<<sCe|F54l)1
zD5;no+~L;kT{ca_*~JM`D5S{HN?=Ta)-iEMDupKUxC4$r=uL`K5VCizh(q+M`z%SR
z)*GtLmN1#3jmIQDicLbGkQ#Zl;_~rRiafe&&|B1mzaps;f0FSD3%h8_>wP34lNeM;
z*m?C;Zhz&AymR*Jh^=R|yT|D|U?u}n?`d00T~@Shg~$gag8^0RNRpIhamDUzpTQtP
z$k&;n*BhMZ3I1JY5t%_Ins&RwHVy4&h3ZwUf{Y*(K}J!45)!Qx!y-dM<m~aW+P&Y5
zul=`xn;yRq(OmgPSm8V&h<I2^1uhU;hZUexLq3~=maNt**2@J-8<NooB_(N^lMjXr
zo0c@oDe@v3()LPXt28<_m^9&ZzUJa`!C;n>xD>ISbGUniFFlQubSmM!s>at9U!4pX
zNsCu4QhRJ`soRF8_1wC*M=>6;t!w71C7x(_DTL^6i+!6_fE2wlxZgmoCw^>?b8~oU
z9o@^@romZ@w^89ILPHWPPk;ChyQ2qeTFtsz|M;`~>w^k_5L|EE9dE`#2!a!k7-p}&
z!ujK8oX!I;)?1WTq*+0#GV0S~?iN!HCj&xRa&d9SWISYA3`x=isT8jFcpUEUQH;mn
z9JW{2CDn#zwTSi*mlx#0BS#}TElEj`QelL|nLr&J-FiX2t;l7=-~HOBxqGxrTUIn>
z6$^EUm!Tm06pG&Hs251Y*2?ud4j_p{bpNq=?>tf~vhf(H6qgql)SC^titXJXC3rAO
zlja3VS*{fY+Gx^jNbR86Y)FSwGBM`$(S&D%G2gTmk1lJ*uA{9xtPNbQE|@Q`*v%w|
zMZ+ky6h#&fOc7}y&N>#0IeTCFMXLKRQ|N-^^aFVQeN^x$p`t^+5Qxx6Dt>H!tLrT4
zo=yVMW<r^Sqg%H*KR#tOUr?kO<DtN~iliH`Xh(egTi>N>+8+}z{q^|6<Bd<haYHp&
zw=IjtqohHP3Z!c&*B!5Z=}nGaeZYBDGO7YcqbMFof~Kqm-S`#<4<AxroscDh@pKAW
zqeOaRHzM;KGaQht3%2tsY!eFy?`)Kc6YzD3Ew|Cm%SRJwk01n(@q+DU%ezl&_GSvB
zBB#6STfHC4hr@LpT@_o!Qc6UB0T<5S{5)~4_4{_`@pNb<QF%(9D3Z3N+(sLoG>_{i
zgf9Lr+Vs(x0_&rhvewavTn2DGUwtGRe)01!@sED{EyPlC`{<B^{R7U{E6^#kTlcuy
zRJ`}2w^<%PVm29It>yBnCd}^gh5z|~%6reRNH=RPpMQh()emE&n#o9%K_HP*N0qV<
zfbZ4$4(lT2rx*-Ty|^z(g(Id};X1aPHQq=2nXZmhCS@clRmemUyri=d!5+IupFzmy
z@tK84sfv&M&jYo)UaNm`&bLp<u_Lx4LnJo~rO+-xjVJ89{wZefKIiO5?_x$foVPnX
ze(RF$`7xjU^c{xDluBPR9gQePW3G-*QN|#IB26<o5hvU={ObFKLTCssAe13Z5{yYO
zQd9dYy0)g?ZgE1A3_&CrP6RsNArgbaQzQa|!?n$g67+`ec3sK$AHV+Iz>Sfy=#3?D
z9@qQc2?5H+gSsOiGDEUAWhXbR7i+5R7NHVkmZOECD24<NUDFVwl@#83@G1s}Yk_SV
z9>4b(Wi7T6%!)l~8~H<TeDO=@-CKP7>%YbCoSbkH8ji*jUKyn%vls@yxmvQ>ZpiWh
z2ZwhEDAwB*b=&mONAjb?B1EsH>)$=vFNvEr^^NIfJe$!<5rW4{Mca0?RYO}e<YoXN
zArl#?Ybnw_E>54aSg&uM;|*%}uQ5sgstkcZX+xS9G);}SfuI7)YsU9(v-5>FnZ5P~
zMViuXHe^DeyPCC;?92`s9q!USeV07dOlPy3%w}sXQVJ3i#SF5@akh<$F{MbPB29AI
zrlu~J)K}-EuA`Yu2s%*s0+C1>;i=1t=N~@CI!Ts7o~3kkL%k^@cT8&nvJVoBzf%-D
z^o8p6CMN<xL~9ilROIE}7>dVsth0euNyMmNB6nDv&&dV@luocgaJ}(Mp)gt?L-fnE
z0c190nx~BR_SniH-}$ZYv0Ju`jt<cIgiEivN;EpPY!+9%b$LNC-C=%sm)}`0$WE4w
z73>T%_I5@{5m-Ju$4zeY?){(V*MIXl9{Ac<zQR`@7~IqEMkIuQRN|&hC#C#oKkq%t
zoV7S-W1>I=y-D+&JWZ*#B~4Z12Sc<-$aijYR$5L^uRea(_XXbdv%LA>pZ_HN3qE#t
zcB7!HsnLUihhO|6-AfO+TAp$G_%Yl0hBtrVOPH>uvT#x@ST<V*Msf1wl97VXzV<5Z
z_L2lkI+-Aqih*t3{g`*4kP0&#GVq>qvB7sOmT2ZK>l&2A)paZ!lm?@*2xJ>Ly?DWQ
z-#R6@k}tjSkW3mvXOT!+*WGl7uFoLhd(<gOKnsM>2<tz(4E1}rcOI+<=WyENiv*oz
z7$tDa6+Q&gBEw1uoh40kX0zQ0b|}?fv@}^UM1;s2S0SEh62+|ozWUM*zxka<RK=7R
z;v7AmaQOOX_{tZ4o+t0WgZ=fEvrwXZ!=mZvwiVla%-W3jz30cg^UV`px%YGY+%J5E
zVfh?=`5}}GgpVREe4uMPgye?rEs5_B$5Rl3&RN>FM#z}pQmxlKfBFPM9Yq)=(Yh-e
z@M0BMc3{1wvW{+ih)N4=(_Fj9M3c}zbI(V?k4OFe#aW-9jAM7r4f7zqL=sI-wah3c
zyX?LBDj_%Ae)DtG>ymC&<BKuRpI&e{$q?BQ?CjH+f~sywwIt6INc2ZaJ+5DXxPX8j
zEdzo_YmFXF(8{n`&1svGZo5LsfEEE^C7lxZZOgOA&yWN%<7v8paG->W#9O5aD)RV5
zpL-buc;WHFA+C#?-Yd%qPtXG4K(>w;0}#zgrB~R-;=N!rp0HZ1ajqp%2}*zw8Ce$5
zhNkW?g#i)hRKOTPkdl-03(Ca_FHZvFNzU<d#Xy2zHF(+a#lQVEKKrH5VP*v%e)}7I
z>*^9CN4%6v43g!^IZY>c`Hjyo8t!tndB$oxM+i@1l4!$V3{n}Cmocvw2wozHx7AM6
zeX4}QB5|stvyMP4ij^>g)=_U_VV4ajq)E!QwH(}kgD||q>BU=YH|57}M|^bH|CAX8
z+!!JMB(fLVxPR0Q>!*Fqzu1xufkf*Ryrc^j5uyUMYbs1?_`=tIp4n(ZJHKLP0-`Bd
z%`X@YvY6LsI|kDk+w&?SV<@zYYr{E<^^T-R1_<=G)D9^nCd-j&LORN+S4*~w1&OtU
z$rP(Cg)a!HV16>^#q$Ms?(H!fWxW641z&nSw!f5$RzT92`1Jc0PgJHykt|+D)7Fn&
z8afg48m00CRYu<`>mfK?5J*LgGD|aT=MkyL%WF=rK`V<7HmYI?co|4ji6Q}=@cyGy
zs>@4kstKbhS#w2xa0fe>Q6YHv@HPI{fAAmi#^*lA!R`U;iwmCp=xsju-naS9-}tAD
zpS;Jek#zG7cYgj;G{Uet-GF!e=I=b=&42%^3@=|0nq@pC0@2b!=5bF9*K^U+iH6#?
zBu#_T8Ywk|KwVWmwIQl$0}=$(LGbMPQ`Xz+W|rK{7eN$yh48N^>cq#8oj+=={7Kf)
zU+nb*Fd0t>A)0urQO;m8Av}DUWIM+k?vr<xiER+gio{v8&ai6C*=kMK1n%4!5kgD1
zS+KJ=qv>0tqAz-`FZ~b;y&we1XoQrCdb7p45?60=bw$vM5L~49D2Y~*(6&5&@tljx
z4JQ{R|JK)D<<9;NikJhfgp8iA*OVZoqA9;IkrIF;1Qk`G%7Tp9-S}Q{E$BT+W1N_@
z3M&G}7))ZQ>n4ifdQpPbDdWj*|Gfm>_X#PT?>+YvQiBq49R~y>)A8D_V=}UAW+Rl9
zB!h(euRi3?{X2Z=3!mfv`hWP}v-;t8`O)(ehFa1rpW=S~|HrN2OZ;H9<OkoX`2G(v
zUV8a1pSu4VpSibS@bvd_+d1C4xNo@Y4w45PNt&V3oNZn4?8#&5x}+$Eq-n}(F=w?|
zl6cr|J>Cj*VyM02ahH%k{A?6EHYHg$;L-CZoIHPiGo8cDfB&ep@JW%pkAI!Nc<bg9
zizi|+5D+2S5eG@!yrgD&=ZK@*N2p@Jq%GMw9J6pMwpT0eAKpg@&%rAX*eouXOs6!P
zIkVXqPrQlCzIEfAjo**x-0g$Id3)2wk=h_fV}g=|xu<I@+U<tmK_@v~5;#9y@!_Lm
z?mpN@3qjpjg6$DE8GYx3QlJE0MtANI6d^>@MB%|isUIOY6d@{^L?CE~mml#0Y={JP
zg+^&jlG*sNp6jQiA<IM`(B9Iu9a2l22#DaQ%9<A^Clr}rI?TB9*_Zjj$};kv%1WlA
z9K7ZBgNOVt{`3De>yzjFm;a~#f*-v79ZYa!240-5_}t(B_t<^&i_DKt_~640ZPt-E
zFlmZ037JZfKw}&(*x1GnuJ<eRB-WGEmTUkO9vcESxOgd1f~E>sE9p9LHu@I{DR92S
z21`@doSmN0wU+D3SoW_gA*Nh}>$dN;8j2n<jGx4Hs5SH|T|p1q|DoCVr=0Tt=7tm+
zfs6+UF;BAGtZ6n&rh@@H#SmZDWQjsr#j31PO0&1WPY@kVS&`&}*jl+pU8N*=k5-z_
zcAy2<b_zbk*gRaog@|NJ;Yo)>l-8^lb8OvUiXq#yLNtcp8(heE{&dNwU*6|??_Tim
z)-FSxfQ$-0rByWdhA3?4<By<&&>=~pfC){l)>KtZXardz&?+aG7E}kqMB`NN&_PF4
zYNGHm5Q2<?CEpQx<*E0cjsQrc_6V)0H*2=%Pk43LG15yWdj-eKmT>Zd>8m5Y@C#q#
zl`sE1yL-DxmEeS7XLg(Y*FMje|DFGklYjre&;R*<`=4|E!4H{A$>{V2hp(J*`=u#Q
zezc(3Xny1UHUI8c{yx0(`?Skryl)Y{1K%NggeM3U1Lz&lb~tNe@kZ=ET%4RlYoQPj
zA(G`CV3py~qxWvY?XSo6Do}lfi|Y@ij}EB#S+N!TQQ;?KJk&pu8(tss{o!-nb@}+0
z`AbADmm(%M;02~A&@#tcfzmZ5E7-~5R+?f5L!Q6)4!T~D>I7M>Ihsgz_J%ao5+h|S
zkV3f0W&U`_=6Zw>`Q)-tBxZul4AsRc?RJ9;9x3AM?|e)jK_sk}8#Wuk(f)wl@qiF~
z%sr27BBWIDl9VRMByzEX)}RDJID`|(pa~e9?0I1#fOZIN5Te8D=mY0`1RAVS6tgLH
zRnu&@=t+$6%d)uVh~Alj5a?u0Rc)!;j%{6HI~!HuTGKQ&=a;ACxgfD~9^U^7Cl@u{
z)e_+#9ZV_DPI>ghcWBS&$Z5gATMUBa^f~8i!}-NAAsrwX@#s;))5n@`%rgG|&;JS!
z&wrP&cplI55dA5Iia`S^$?yoyPmg*2^uvhk&m5vJ=6wkK|Ec@$Fins1yc7LZ)f>Oq
zIZYk_17MH_U`}8rNHI&IWJ|JSS(atXYuU1QuXE0NZONAHbv$d!$1B;^+FpAVs6>jS
zC{hvw2oem)7ytuI><OL2*I#^7h4V*!-5@DCP?jjVu4~Ad1~4<-{nlGgJ<omr?y@oz
zmBd*|5Cs(08^3zlh}Era`u#q=yyV#%8(dpi^D?8aDelG}7+=Pu^hIqDNN}Fe;lJ+=
zZG7=;$9mrv?@V9)oP2xkmv3$qi~lsNeNj7QrNb()N|7|0%%&+pRiOfjaT=ow4lT`5
ztA%J&qKgsMR5TkkHutuu)$9JDcg`aM#x3fDKX0FH<2+Z*jD^=uKwMA2wdr>}!CQ(P
zVI|r+`hyH9Ljn~OsR&(FRMsC3`#{!xQn-)H@L}x23IZ2lg`-fV@8CK$ej$(Dv`*oK
zcN#ipaMI`5x>9@FcNF=<sM07TD8@P{i4~sDWsJkh0OtfdJ3HjXkhB&OMIr6QIi@ES
z-OJ0M6p>P-QlU_!wKmC-gjYTAE`IIFM>)5$PLvn4S~JYP_DwwAvDA+o<))Xtg-GNC
zRY}$FL$8bO?qEiJ<gkyda%_=f${|AJ2ns+^Izfdcl%PwEb(X9!m>jZ^qb$5O3h6hN
z9gn5-V3(@1!Oq&chu-3~^`Y=_K3Fiq|4>Sxl>c0nQfTcD?|rSGv%V;{K;<lBoZi>o
zzP@2+;@fkR?XP_Sv$ekSj1**p4$^3*n=NEs5L5+$l-MXCsW)j(O%O){H}YJ^FisFE
zA|DPt6j=Dnl{V(!#qH1auN{Hg{n#UEG1f~PCT8Y5<hb9*4u=rLq*8L%-6wed`4x7y
zyBwKs@a*|DUUBDPzc+&brv&>pawio+gh&il7Z_6!1sWx52n;6cQ*~zOk6NVlEn1Bs
zVG!Y5I<6xO-eq5*g2?A67|&?5)?l?4gBxf4yT)0pE|8_~cobTrlp~HDQEYK5=QwuX
zo49)JGR4{^>#G;J{nf7~P!UE(AiN?FA+S=gaQp-(@3@n5>&xtwB`aGiEI#>p&YXTD
zn=1pVO7p~{*9haAdHJi}$BZ5jZLd@HcF4;fS!Wa3-5@qOfrZcpI3Y<zNKC*8LBG3A
zzq9AFM90qo;}Mb3C41Xj<01}}@cTwPfBr<mYqqq6dcEf1u47b9qh4orW(IJqudlPR
zvPw~up8hX{zpt@hv^au55XU~IDbEUwaTvSbZTKJKqU4-Imzq#TgtAUm8LBcP&xa_f
zh^3^=3nnHfiEqDy&6O2))~?e|B(sy#SOvNmvA45>b%t7_jtD%=*Hf)t)J(#`OaUhX
zkd`1xY0WH>g?r@L0P8ft*b#2Z5}|w4l7zCBv$maaZGDGEEvA;lwCg@wS&Tb%K^!5I
z2uC=6@0Dm%VJe3%L#&p7r!crEL@Ghx41tOcWDX94aSl}(RG?_j&5(6RU`kXNjrHd*
zeog|$MWiIphpet%^XMYAuX}SyX{psBl+@U@^Gx3UTGp@KVDtJF7H>O4b7GQHH{Zll
zn`>-#2OMrDgvJuZDQSI*-Cmz~F6PM5X}bLZrh@KP$ZtMa@SSgcBe%4k!LFUh=^Pvc
zGQeT|?}4Kgr_9tRh~k*(`B{=iouc$Px4q46ERs<n$g`4kF2Jm=uw7WLY!5hhb%kqN
z9kxdWBV!Q<4TSw~s2fidEMXwI=dL?=&wJj>-S^y06oxo!$VNl1UcJh>bI)_};zcfA
zy2NIuLs^d9%-`nc4E$jVJL^x_gW^k!t6AY)eo+um4+7NG990-&EzGoQl+q9e4YaA)
z*;u0;OVT(Z90vjKcQL+RG<=XKVcpm+?MJl6RX8VzQx6O6?d;;J5^a3TTM9*`?fAig
zI1FBla&gvo36(_3(BG_`#<>!l#b|-WP`UymLag&bZfP{31c5;cMIZuj0&Bn+i4zhl
zrN4n0gU&0@^-}&24^)6rn!;#|_O^DR9eaDb1c4w;Jv=z5*EoFqIGZ=txb>#f+*n;<
zX={^-nMG8XqGiA>_r9DWtZ{vNm9`R0&&;ttta#|D8(e#OgFs25cEZBK9IZym>{N}z
zhnAR{s*yTHXvDZ+v<Tg2)ZfN*cNp}y$n}7UA|o6LMCGuRBF`;(6;f0ciIZNyo<_JJ
zpfZ}GyGB=@W4OEH8+`afDhUHxjRrGQ)65@Q<j9HR96o%Q#l=OApEyCiUMJ5+^m<)J
zg8|)cm%Y6`I-L%^L7xlDmwE2FbL{Q)d`{!oD7F9p|05YP<k;yA5Cj3*RFp+d7=%b5
z2}5+L#yOFOIF4zw>L@C*V(7c%LLkGC;c$;MO%VdG?-{$B_q#NR7xkbAf4=k%bLkAa
zD#*1(Mltm`r5X(=M_q8Bl9Yv|DXv~x;pnk|=P&pC;lWoG0YZ4ko(#tPNQ2g1pIO-o
z5oEZ^!N^fo6Fj%#SeTzjhTGIk58rU`H}6meSYZ*7M3*H_ScJ6LF*?LJixoaV%vwKk
zwuQ#ngnqw|))hgZ2toyR#6;}aT3Vt}pW=>}zlL^mavXD~2~FiMV}LC(E`0GbTz%y8
z#DS$Y(E=nZ%NLlNI>gNUA}h;7Ocs&G(_C5^vC>Ep!C@YF=X*$0PIvPf>rZ`|?Amz(
zYteB7696tJ#DURluCJoY+`A$xhb|3Oo`GoNcCIs{43kkrqtT=lhu~l|?&yU|(rh$%
z$tz#Z``-VZ-20N35J&O&`zby1%Q?DxyFC5m6MXs0U*@sL9_8%WXV}^8P*@)bucYAk
zu_OHDAOCSC+7o>6gCFFxpZh#{VX*c;H^Tk)TgGUdt_b6Z!W4{#UCMk&J{V9_0fBRL
zcX!CUmg#njrK3lfUtGWy15DAy6?-`6n3|rZyVs$!+u_*pWB#2z9(jMY6y!zCH)DpY
zgD|8qJxgjPa87#(f^*o)kPZ6GHA+e&nJxn!`N}c`Fg;V_?pu$rI5X)lC(eztnkAwt
zDDsL*D|A(7n73Kmj@ju#Ar(6Dk^|)m7LyK(le;wQ6`@x^Af>}fuiJJgl6r%xGDvAr
zVaS0bX8*#yPlPaI>OoqgR<HTXhIL+kWXIBnJmc7{H}R?8{UV*66&AURXljn{`K$jk
zdpo;qTzD4c9HEfZ<`1#6R-%%Wzwqb(Jf}|I%x-6w%a^Zm?uBKpU0CDe4`1W$Z+Rs*
zAHI$7>NB|R2Bppr&eBWU*ttb&GUDdAO%zC?B=F^HtYPsm=XM83X~`?e)>fanlZz}g
zVg^~4dYrJa*JpE(Qz<o8h&Ye%V%+o`hXsfz;B61QjeqzL{}Cr|zQxxDjHevh_(_!3
z<ax&V7tZm&{nX#$)1P?=?|b=rz23Lq-uXjy>|iYX1ccEDYmq1<0kt4T4*SGOjfL4c
z^!z*_8bOIvmf_|yN*J6m)a$huaRFl;f%~eBb$}ij|9Nukpb<h4IM8{47K$)RNa}U+
z{vPNIs|9f!qOC@o5@$;qjnt2j#Q5(&jvvDbqy^`U&%CR^=m3fUVN#q-DXbSr3sX`C
z8XXoyfgo^>SP23^VkwbhYq>&eKLT@=9k<Zc!RQhxaK>RwL4PnL%ZJozDQT?+LJ`CX
zOUF<0uzHB+uU@5bD8||noR9Z%L^Ng=xZ}Q;vHZ{{iBrY38#myE7nlmCS<3~3VMf*w
zlpA{#W}Ubp`N1FgPOhx3Q)tO&KK(E@ceLttrYGu5P1RVMKg!~%JGgC5u>HFqWx@<F
zl|^ewSt*7&Xss|p5kw&>lxPW+sWAOLtgI23ibh&v;m{%OzWW|tcmM0S_rCi$eE1NJ
zR*N`I2!hZ*MF(aMfBUS8V%*K}ohNN7hNB@@u3X`<M<3<$pZ`3MKmG(OtLuKWJ~qAB
zv6k!GZ+rd0b?|H}i75)4keDE%k`b|V)LH>lC3*-UP;WK~14q4GLuWm*&N_LXQ;!q2
zx3-y>XcO1#zC-8#0IzA>cR34z6XT4^eQ5!bP#T;oDDxZ{Mnv^GG8AO}J)F*|%79uu
zKx;vsYj0QpbWste@p$~mS6zptPmv%*1_Fac+X4_IN;BK`xP(D5iM0{Nz}_HYkV)oR
z74wZflq?WPFAfhqfV8sOJMhQv2OJ3ND61j8?hc*3E(lGl)gY~<2o=&?IzmllgmKE@
z<F{dSNi`gB`Rr3{TsTKH>QnT3Y+gLa=EY~2h!lqoFA;<hR!Yj;arw$cGAZdPhfy&%
zI%~Z2#3{1jh^L>s!tJ-*%AKbUG1;6TT{yvTuS<WJqbtiy+cA9VQ|Q&JjJjR=d)weD
z1R<uh^m{o;y~p&?MbhCW@3?)DcfR{wEG!)+wgo|4qj}TaOfD>;0!fm@OiWJ^M6v%e
z7)?>;R7Ht$j-*y2snvPw(jngTwg(t=_xQ}GKgB=&Xa9_|=gy&xAq)fF`L6Had%x#<
zX*Sy&K754b<!fBLd=)1>a>82oZNG=Ku8?B@v@QyYEThVDTrmdwrZM%jLDt=3ZMZ^O
zOKG(xsI?{tRD*0!(A!?6)oxL5)LCCy!5ZT+JL5y&R=%kDn$F$6*&ztfb{w}Sl<)M7
zn-VySic{)=U=fG4aC<Xkb5nBt${xD9${SzXq?RU>xsT2>rlJ}R7>-)(4i2%}X>+|7
zvuPw*C^&ig1Sd`(L#dRVtpT5X<Vo7ABO1eXnlhtZGaQ<(m~Z!qqY6ZX6&e*P%3%&E
zUUD>^K#K7NPzv90*k|ZPVFE!+<K*qs4<8}l+eH^S6Sv<@v~ZN$?jA8cHGwOK5KS<B
z`@OvDM}Ca|`)~cvjD{I=OUFr%onqsOr<j<Y=Jq>IF+JDj=<#_@pFYX`_mBAGCqK(%
zdx~^wj_Yog+L<Fvi-KT~L6}h08eAOpNn2fHcoDODg%e3cLt3P9G{QPzl2W-o+5~KD
z6`UH3Selz*@A?LLYl5j}ooH%__k8yUsMTw%Z>+Pte1)~OHTF8YNMK=Zp8MbYMt<Zk
z`~~I~mk>gY-(wDE94ZRDj3|f*<Ai%(@-qJB|L1RV<HnD3V`Y^nh^g0`lvTyf?k-hT
zeVvx?Pkw#Vg&i<<ih8pHl|fjCMByrn3JeqlJ6qe7dofe>l%g78Y>rZvDj(8FW6Wqs
zI?j4kD!>`vCEI7G!9L$+TpN$Ob^Es}X)U(&QA-F#m?l(Zh^un6EeL`L6$IYI5ICBR
z#A_k~<y+HAVsQvtVylv}G*ndrZkm2KrZb4x&O>$!$!_MEM2d=9U}+_eR?RZk)-<9%
z%_Juf7AeOl2q|f{T9{FesWetwl$3sqE5yNJTU8l5yPJR`iW0&gMyQZ5jG367;|*_n
zfVCSPj^8zp2n5;`2;l?9rBuA^wXfy!r$0v6keoU^&$*``<ytkvPG>|B#so@GSU<|C
zB~3crJ(}&5{`D<REF9wTN6*pUD!6VoQOaUmimDg<zy9Le=&p9D2a-a-s5F#iKwe2t
zL6-qR5Fi3cVGP4O#}-3o8xeo*&wM|}UiK#LzV9VW%uJ&~53jZ3cck1eI`Ctnv9itw
zWheC)O%wu@iwUEI`|g|O%-#3$LqGV#tZi)Yl`nmnkALEmeD1-|vA(fBPQ>uP|F_G=
z`bNtd<0<y+PazF)Qb#+*PFa#U$4n5=h$WE}$fzI)6}sqP^bqnsgWV0nP!c8yVH7bK
z4hhrLO9dE@T*q_1eRP7M#1z%RC=fu10I3YZ7U&|wR^9<y1__leDT;ztJ7RM)C(A}?
z9JM&29H+$uNsJKQE$xI4u~%U-mM(w~81n4NSPnQc6_aa4Q6+5eg<M~&xV!_SAYn4n
zyzZpo<~dE2NL*QY$%Ox#jt`@NP=yfs&#@?T@;t}r3Z)cDQX`5Inz!7;+)G}|*5l{s
z<$Z!6f~w#X|NiIr{SW>Uu_<ZB0Re_UXl|OD;l#`mjo4o@GhMOQj5soPE1&p{#~Ia{
zG;33I`yHB-4LY0K^g9_J`^4wz?A}39<XpLOgR56o+1cvR&ouYH=_KzuHVw;{e7CO3
z$1E>}RD!)hiQX8pFd8B!X2|M?xxTZ-QyXhUO3_FYYAZLGoL%IOyYJ<eTTg=w$VWp~
zZd|9^?UE!3)>_ur*Ew?dFt^-xyWf|gqCGv!yWjhrOwY{lQ$O?5Jbm^Vrluyj_sqS-
zae@$nTCL7&Uh`UJ=N16L*5)S5%gba%i9IlbeEaPtYwcJLV+n;M31dW6QObhSRAeq<
zraes(B<RtAyy&sDvCi~Fi%?Wp>zJ5dV7R+QJqk(V7>r@Gvx8L18zsU3vmY5YUi;}g
zSRbEdt-oMM=kpwd0F)#shiDY_Bw=YnaBVy1@DUKM#md?iyWIix#TlehUVfqk+Gu*C
zIUc<*!<uPRCJDAKh$k(Lr5R+9@|jORLbp30jwNq+*E_g$?jozJ8!Yd2c<kwQPEI8p
zn<<z`OKL$zqgE29DYmF^$|6GFt#`bL=~2Std3+~N5G4&3Uh*m?U;BDSMS-gfL20nL
z;f|NSltvm-6+@JYL8TnK;|w#iGX%XIf<s3ciW-CNfKpdfTBDR?H0qJ%L!5xuzV<$r
zuU>~(lg0_cMLc<BgGV2^LU(HbRpE7Qk>c%dI?b_p=x;A0g{If<G0Y2ctr!*|l~eQw
znyjBOH8sJZcAdSQEnaumExh!*e}sv}MT{}b&df76zkmz^iqa>P)ax~pTFoQ8*f-1g
zO5?!mf52h&VF|Bz#Vh&XuY8csZijYzl8K2avMl4XpZzpH^Y{Kf>+2hQLn-*5(E8>(
zcH<73w@!~oVh*IU1lAHdfi%)5MD#{f)yQ|~s8IC~n~g|hz@Rj!SdN9>78eHyBMFQ{
zScldYgLlq3V|^Z@_B3#eMUCg@&Q$2a5yb&P5|ir+VKlYSkR~C+%+X2=%_P8CM_?nI
zv}ozm7fe3H6$NEA%^+Lg>UNE5y8+u4%66UU!$)w`SYE!$!u%wIB4cyVr+0OShRZnC
zPC0!jWvbC9j`ye`aFm!hK&A>;RcL3CV|K`XrxRxx4f~9S1EMG<PyvAo38NGxBr1%#
z`Hnl8TP&D45>u4{7DM1Hk^mbhPTX}Ta&nU1sLM=R<MyR_7LL4}=U@13oC-K~&oNe3
zH>f2kL1akU0SiZ_$%>3lZ$wFnH4<Y$p$L&kt+;+|i`W=)sTpR_A4U{qKw&Jg^TALy
zuCwK0wu+Fg)^Wt*-MG0WCTEt|>fGSBKY7g;XMv*LsFT)f%*-q>KeIr4VuE_T?saB>
z-$>(ddHK}{7H@$Efs8^_U^#a5B!><i<sI*Q7nd$x;KRT9AwKls4{_!46^crG{iXB2
z=hyDDKlxp!O@*)$DLpPzIYAnyxX2&US+7S==cG!ICMi+~tR9m0J0!89x3R)x9Aj~m
z#`DznM@9SkI`E&l7>j_dz+*_1*LAruKEdLndjjF(#;Uv`aKekabxA1w_oCM;u%=>n
zug~PMMM4KfUf_T*O%QUd=W^qs-5H#zFjYZWf~i_K9T6+TsFW<PX|8U;MiCPnnq_ur
zg3(^iXD%;ejOLbwK53F*bwL<})DlNoRR@v`IldGmaYCR1v^MN@JK!Kp8XS82cOsJ(
zCr_PbXgVk<DJ#p)^<@@oigIX>)}aEZMFEp(gGQ91EEHwQu5(29zK&;}?$c@==H8dw
z#`U#T^3f0}9F2C9{#KtyF75Etm(O8rh0zv=#JLdbQu>{WZMVTxcgWV-232N3OR7>+
z7|2Y(sB78nS3K98;qk35E89C{Sw<9v#GxWh0Q5i$zXH;<$urMB#qr~}aA@f$T5E2;
z<rH_`eHTfR0<d&=hlfA^5QmN&rO{}64WX0-vEtRQyPtpgv;UIMeD*Wc>vdlGl9ytw
zWo>PZ%a<?ny4SvzzxWsb5{rvVWTPP;`^ZQ5-~aJHVPk9O+qSq<K|r7atTlvDOp;o_
zVq8UTa#n_Wl(i<4tp;hML31=gzPCZI(_tc31c4-K&LOnL4ZDOQpd4ffW6(u`8Rpa$
z=1^habDGDR52MWif6f{^-bZ%aaPZ4Hiwr_+UU{=xS#o5y$;BHROia`m^dJq#k$K~>
zrvR)_?DZ4YOF=bjXw1}*vPo+;V)@1jX}!tL_K4xI!n%S-9(|H`e#hH*>WQaWoSPsY
z4mkJB3+%~|x)kKw>nw^*UNV1?dLzaSa$l4t9)P?L0IziAla<DrW>;nGJn}Hf)C|#y
zQ)nxF#&cMwG2!!Zhl3GuQbPtQwaH29X~axC!&ZO53t#v=Tbno7>lG|tyvpqS1jbpa
z(vTMg))}Uz8jOZHjoKV>3URtktCe#7$_By)NNZ6#<iRgp=FhymjjIgLuUz5A+7`C7
zRC$Foib@zp1mxppS-qZ8-`wKP(`R_>?8BV@?W<^QXf&EE%q?;J_(^8w=2=)=Vs2rP
z*?5UYl)N}_7;EP|1NW<p65~AVC`!5G^xgi~LXOQzj*~Z?WUtfVr+(`1P#ODevs=FD
z<ca+%cE8RwZtPxkUgW8Q05^91jC*3^4v!)TLri5zgx4cr9HU+z5lMoyK6a{lF}P9z
z(mG64jHfVe>|XWYT$Dn^F(MyP6*-cSG_GL^O;zTg3)<~En>$0MClVqDVvJ!Bfsci;
z#$odYgTXv&yGxwgXtE{iloN_$cO2%Wue^uR(2!@EuI}>Icf1*6Gpx<nT<dfFxn-6w
zUuNgx9;X&3IXsy#6K#^D4jHx(QlsM1BSglc_CSq=s)Z%ba$jt!0404V(MNTSBbtJU
z)>I283$J%@2DJ9>J_PlJ1tw3PVD#KsgwdF6z|71Ht;spslhb_Qz3-&asv(sm3<IuS
zy-L4pm}pOP>QtSLwH>;9IXAAa5k(PG6A2sJ16J0$+)_`;^F6AO1|3jXM;1aAOD<Ef
zymFPY*T=ON(di_kK2-f16z5(5;X#<rTEZ}<mexp<gdj)=gOp~g$<opxmX?;7o10^9
zVS)C<BvBLrYP`w3Zgc-2cX+Tb))GibBMi9v?z_3|)?4|3ANW3g{nvh#|K}Hfg)7&t
zqmA*Ra0f4PhW_NYTve8Y<1V#BprROL0C)q6(VAi}2Q`)+D2dK;0vUkv-JURxKwCng
z5K7P+3~<&UqsVjdzUtcOh4EwAfpud|9DW~LiEv&X5L!c_4TaW(afBTiLItJvd7P#+
zT)J?Lc0FNwdV*l=0uQY9d)yiCD92#Rj4~_P?M`rGcZS?H2~3F!B{#YymwGW%CysMs
zp~cO2+{Db>G%MHEdGZS}4_>`OC!64|#ZB4`hcy*p;J;q`oJp?}6a<0q+z2Ts%Yr=1
zXiiC@S`!>ZNs}OQ<f9&Oqs{!uQ(SxWVSKWHq%12sgA8ppuvRfsPq8|qh+Fi`Bu_u{
z0<V7ktGV-Kr}@&CA7OE6mQaNZ`Xe+J8At35hA0{M3$C$VCz)y*qN2Bam3(WPPN%?R
zipof=g-Q#Cxj==I$G7%4XBu4D9<qGt3R!Om&JhKYxkIzO;mr>)F)_pL&K@|$o8S6Y
zPTz4SQtiLC9g~xj%+1Yl_39OFyZy8m)>}Xbf*|7b>ASi0*3(#{y==i+*4Nj_vW%C#
z^yQqo?RJ7F#2L$v{OFG|F*U{i`+xo4SlihAw#4RzK}ZzGUW;ioVH{(vrqVf89->Xf
z+OW^gXh<tcINWHF*cq~&5rxsXI3;SgsW(~-uAe8+6@d^)YcNVsWEpv{M>07<5JlL^
zc*nAHn976pwblqBh@yy4Ne`oT`)9u2tHx-eFhUnOK@d@un!9e9XR28P;}FJQv<wcb
zC+Q3#!WqGxufLu7qYJF9jJUDd;ig+|r7AUfuCZ34RLsWu7DZ8Dj3G@EW|tZ)ojT6)
zwN(m}F<rZj(TyFRdtu0Z;T}<9F=dV!XW}~RD5`?~U=M4|_%dxVRY|sTiL1Z<ADO)S
z<xIWqt(d5e5D{74BNmE&e?V)ZO&mE06=f8$5Z0MGe1zFoy@BTR5)Xd*ED!$9Lj;kB
z??3v;<7}*NQ>&%iedZR9o|vOQ7%&_t=I7=qZgpHZf0cf(LJCRh40&cLjYI2_m9<?u
z+Xc!AOr?FgOFd>K7JP1Ni`yn!5GLH1zL}?Q?DFgj&$HL*5+w;qny@~z#)V5SP^;BR
z;)F)M&aJoK!FPPecd>Zrh|h9AaFu(9E@HozH}(?9@j0?z>)`doQU$#2fd}~K|NLLD
zy}j%IUBTD6XvZJK@~v8bsE*x%`_4JMn$Y^uA{G-0T&!@qj&m_4RLH;)#{v=f&Vh9v
z5t0puWZ4LTquyu|w;E7SF+t$DKV?DD9~|H^7&~0#xP^)p1aX2fLyA#BkP4zOrYth7
z(X^TkmalJ9uh)<+Kp01*EY20^UV$s>3`d7Kzd6PEp<*yO$<(P6%pRK{5b*hjo@Q-z
zi!96O^?Ur{FaA0+Gt;yt+8n>>2yg!02e^9SD$jo9F}`$lojb}S+&q~v@A{<1qS6o@
z2(0!Y=qmEz&q^t@(Ug@>!;o^n^D;h!_iqIvK!y=WgAtaZEQw=Zq&NpEOgMh(4lZ4I
zmZIuW_ARqXgleR$@9gop2fxUBzw-f-nxw2s`n^6%)Tq@GhH}7<{>AUZTEWW74O*=R
zmoMJnU;o0#xVEmj^W;3k>lZOqi9=vh$v{bZ3c9vrlu0I<0b!;Y_O=)n6?r~Fxr!i=
z#7Uh-qeYT5nVw$c@R6g;OwTeoIZ3nGBu)|p9ux4)vrn;n^%B$5v)p#u9UM7wj5KK=
z$6Y-C;Ndr$bYqdLL<S)dg5xJn@>hQ1ukzNnJ;1;Hg@4Cye)z-eb$ers!MAQR{nwT?
z_G>h^N_h%9&Z0F4DLg*IT2v5$7GN||hR8S~OcJy;6jhEej$)KSrO7&72o-Uoi3mgA
zRj}S1q|1^r^AJ_5HCk(rD-vVSwg44IgsCGh3Zk)8KuRbJK_E4WbmUt*Y+k&=RBK#3
zI6J<qI6(V@vCQcA70=wbjSKP!^%KWP8!dL%Z&36~hHb;CS03Te@p)7nlcWh&#vDFz
z3-#s<diXM)`O=r^uOz%?=@QKeh1GhjhvlF!*>^4~rKlw}+N}w;cD68iMz(y3`k6Vx
zL{jOL!QLwA#4Im;=XdegXFi4ATA?{R$BCD{j8T#C{FlGTBdgarIZ@|OD}f?lWn&Mm
zB)8mkl3Pw4=kU=5gn*SBYaBgxh@bpl?!-95-rgQkD0a5@*xMO!@xl$BeQufQNZ^Wf
zUO0c9t(}mO7F3mge^~)#DA}u7)<+|{**2H1tWtD(IPLYrB29Vs+rN{){=fcBgbL~R
z_NdkCv|25XmO5DeeK9#R!LR+=FLCynTR3y(r5rzg6KPTd=`W04`{>5w#ERQ)yPc(_
z!%R=l5QL$B28AR}65jv*_cJv$#XtCm|By!?dxEku<3H=$qso&iB#a^es=`lqf-oWs
zV}i(Myt=?)9dyf*T97lJq^P7xu7?Em3F6r~hMOzIbBEDrA2C*|GtecG4Jo=kWSXFA
zDTINa=!{0>!^~^KmBd-cXfPm3B9bUP==|&t^sKi2-*c8EN!jUjSvqwRNr)}Saeqo<
zvkWtc=@bq*JB_8yZWkt&n$%`$bk=*E`|=Cu!Z%6u*zr6V?s4m_huGQb5H@DHeEk}i
zuAJxnA9x3AYa8rrS%T&nx}y&>F_~hG8h4pM8%?*@q1)>r$Fd%4En1gIDNuu5cE0ji
z$}DH`z2AqknsTpCp7)_SHI{sMU15?WgjO@`b*Q=<eAj#4z@eij_>JHA0{`Lv`VE9k
z7!3Odlb5T@8~o`%^ET2npi!TpJyGWfjt~5)ckuZKpW@>Ab#e`=l2{p{rSo8NCo!eP
z$&e}u*t8X^7M|^ve0g~T8May7%ek?94a$m_zUnm`I&_4c-7RM4XL#-X_mkA>Xl>cq
z*<o#Mg<t)ZU*u&ke+9SSaVHZKQ)3vqz?hQlolT@r%*@R1tET`*r3?Cl9*stuAc%0l
z&9~gbyWjn8KK<!W(P}iQHJfa0@36hILw_)&s`U8j6@07q(Kl7teYLS+KO50I%nSyL
z5SFB#kVYYcjwB!Ti6`nzADScExK1(bk)$am>yy-Lv}!f#NrKKYMx7p2QJ_qlpjpQu
zC`Kd7JR_9KbAq)00!b--hg1kc6%fTKMKxeJ%&0Z$s31g_1wrJ2kDWn|(f(#%1P;my
zmq%pT1kbJpTr6XfQ?uOv=2ucwf(uV9bMEXi{d`|%AwVRQ8T2~a2yva~&p*r2n-6j5
z$Wg@X0!r7oymbS*htTCU4o&Vtq!3XZX>+fwvf|)tXtX(q_j>W`>Z4BxLY!0}G{V99
z>J8ehdGgVSW=$i6_F{Myas1}fT#o`qdWb75A{$VjY|!g>`22&<&>vL1;?=it=*Se)
zvvqcMRvG01NfI;6d(`R)x7>CVsU$l)9pW@)dv{25_Y9l;jG>flM~+pSvtgiHLK%g)
zs>bHkJ$l_WhQl0d43QKBP7~E)PTzAkKllTGmNWOgoY}c0;yCqT>R(-GjV}mm>sR^s
zCq6`{)8!MN_&7(89_QXO_i^vNFJpFg5v6=k{r(FDAw8KLhYEfEFTfpl+{xelJ3q~v
z-~1;2kALyAJn_`i=*o<BCI6i}<&R=Hdr;RMbiRe~%r@c2Jjz;%!3di~o?)gnNQFhA
zq7<7C$i}g5X)UH%pG1@;S*OdWH$+ZP5H=b(Ybdi3Ri67C%CRt9A~4S1*sazik!NYy
z{ZZ$5w3=&2kxM4pidxe0NE#92atD#`o*Pp^WkCeEGAADhp1yWFo5x;3S!HxrHqmB>
z+&Y@An396?m)AIZ{tAaqOmX!10v1OYMC|Pih{J@#FMTObexhP-dz(9wRiZfdUpFBr
zs*=HQfHoCLoRY*T#u)!za1Ns@I*)&W`mJ{%YAw_-M3+PI{x)+r-_B3|?7yYAc7>Uf
zH<L_GgEp*RJkRg`$6sLc>4yo1Jp!q5w!~V&a9E;3sE3MPw@0Nj%1UD;2k}M57+Mo;
zW@e{(;_L+`Y7uJhI!`=ziRDhlmM{!Lka38Tijl798A~^p>|N~9?{A{4Lpq0*lGgMT
z?|tw4`2O$z5$b6h83#0*jqy`6HpIO66yY2^^60~yedY<CdG-l@_xC@>-FMx~_k7@o
zIJ9(>13@Pap}c#s-3n<nCp`_|tF%Ny2+}0xO>elL+i$;(|MUO+dH&VU{~PkcM^k;<
ztOFgc|Ky}pgkemS*8OCssz7V+rgRn^C`=Til7QZ*fR%M>p+KmR+T<MFe3^cp5!Gu{
zmB!>Atq~*<c34pM`q=%mM@a%0Pf#2R5c^v2uQFR0ZN}Z){dbQfj6+7ILa7Sj45p~i
zBSEPnP;mL$E*sTj-1CYvtexLscWs?~caO<dgZ4~Fx1$jR<XORBR1#0s3DP=FrEGWB
zdB?loL~k%)qLtF`4%og?aO-@?#Msbholj57i=1w+gVx$dT#g~wzPK|;Sz#-~#`Djz
z_(16=C3&AbACR;siIgvBgbJwDYS``uCL7S(*x=~NlGncUE-qZ%=5t?s7MYj6L#Q*t
zFrh3BWl<4CnvKotRAq~Pzt7C<96$P(zL($r$Y*%wsmm03&d^2-L&5Y+jm~D@%LBrg
z9b2(sG+RcJ_Y*FxtuX3e#%V(=B|r4%f0V!c6Mv0*twCPoIA@udoFHj5kYRvzFd7Zm
z+1{el>9BnH5>a%WX0yp)G~nr{ALofDA14SDZ+XkxxbNPVkR&zwgCUPS_7$Fg{#oAf
zj`wiOEvFGuQL8ujNB`&_vAerXy;eg?$tcU%+T7y#7hd4(GtctiL!amAr=KA&@^8r&
z`KAgx>--C7d}%niu{&$QXcS<If+{bGREP;ey4^0dL@@04NQA|O0ad98>nW-hqqUD|
zX-&21^tzZ%7v&(V)zE>$8H+VOt5gO7WmS<Eel#hZBTZtGFeIv_6j?@TENNP!stSU@
zBUrjS11jssuLX`s2UO)W%l&!Il?jbgCz(HWiVK%UJpITMWb0eRNzBx2ld>uqjdB7N
zW3&L5B4kROIBvb|Bu9=<@$y&RN)m)ze`byAk6mDEH$m(^%t8QZ;6?6LSsrvdUlg(K
zcRYCU;{sF|1JdI{gk}5c1sb=+7*R)J{Xr)s2F2{5<ES`cuN*+7srI&+oNKXueT%Ay
zc=(|wdH&ov-txdJIeGIUmF|%ybBsnMJG)(sQEYCESY6rRiw`}4usQwREpA+y;{5ZQ
ztaOKLxR4dC*{&>A1=4ERDH)btj4XjbO5Y(9j>*Y3-~IkS&D-DlZsK|agm)u50m8|z
zb&8K)&YPR-lx0aK`<#7di|5Wg&hLEkx47@#S8)IRujlBIV<;7l0j%P{-7K6h^1~=X
zDaoX4^T6BQ%BfSQ_!mF>FZs3K_)YS$ChSjo$$d!Zc<l>x{SU7d5(1PltX{c5E0)Bi
zhSX~?Wm3areGt%`Y$2pTD~OW>t-8c<M77g}AY?QcQf0+~bG6b2W!(6jcSJ$pB|=(b
zBhRo>D)gXE<s}hYyW0c;`U4;1bL_+vMWrc)MjDALLckIPmZF;G{N_AYTgM3!MIFHh
z{^}d(4+=i^p-0(R>7$~6<994^$KA&$@)61gOir|EHWRL1yXK!I6>`(vCs=*vI!k-k
znVqvlK|pCMM%jqf^=lMmK@>)`niDwZ*xlP93_J@gUtMP7BfreTJ3oMGOkxKyd4G$z
zHAVC2De8xh`%$=%*ub)I`d+^GZ~PPw{?fnW%BO#sNt01m7H8mhKK?nh%Q${w5uC-i
z3_-wfXn5#zUuI{k$7nEOuQQ?^M?`uJwfzFWe0G~_`5qe@_Ozvxj?m_$;{()Ka7ZMU
zSO}yNXe8P}b8->_$(3tYn4F!VJu`)hLiTe)4{rSqz|!gL@`;asf{W+R6DI-nW{qxd
zlN+m7`Qab@%bY%amzOi_V{l~PhXG?v(Th?#d+<X^$>RJRKlzh?m8Z@=!>2y;*>89;
z+JAn(l}lG8Rp1Ni15K%v1Ytyy)DZ}>QJ*T$XoLZitp<U$1QTtFY=|%wR##pDVYPpE
z1p$LmMij+p;ZT9Z6(yq5-f^o9N=S?tgY3tVan4z!9~^;$O8bu~r96yKNR$x7q2E)L
zuxMupa^q1zRw79VDL9@EIooT|9Spf}^<ivK5YN;(apwZ>eETbS?(8Zb`jw|B`+`zu
zJbms7j-NbEz1HCQOV4rgmIWrJCb+tMjUblHPEB*ZSfet|8%?4J<2)*;tV;4cM+kX9
zC@^Mhx^f0*HN#5NICY31jIm|T?&?LFM{Z$mehFmgwY$a<N=0Q0j4qj5Jb{fSS=$}Z
zn5+}G71?kPB|zife@Bnc(Vk2hWj$K$Nh)KBlGK|U%#ing@0)nn`(Dq_{i|Q+!LtM2
zf9h76qb-`N+w5+8*6GMvdd|`vX@<Kcqe}aWs}S6I&%ONEU;4`|E*-`>K~j`nKWGH`
zuq22bK^jqSG-))N96OFbgPmiqyUWl2!awKg)ytIC2<HsTS6|@R(VLl`o+rykY;LR)
zgdu}L*UP2GQnU8d6s^e#1OjIb)6Q~uX@R@$Jk58#|Go6OeLni}-{J56{eQrXmDNA?
zeez9f?Cic<!hZnm48|FpH3(xNF9|9`635tTL_W$WhLTi*C<{NamV!}LkgAlZ)u7Yu
zpyHV3WQ);G4`+L0r)xwU$JngERMi38F^VIEv6!-=EK0BCib4Vbd0Em*LgF}LXf#Qh
zvTia8XK~etsuJja!l-QW)K-m4x<+kdmH7AsSJyAnJfdkWHhAEzujA&Mmgo<9Y^-l%
zO~i9&uQ0JN&(f`jnVp;@AND~@o_zdycDA-@Hm5PuOLR9nghiKR_Y$ZsGOiKFG1?il
z)quksbnCo`))=2XskK2xDaJzJ49XT{J1Y$8HRR&S@d(U}tE7PX^c*U#k(oYWo-y3o
z;PByNoPX>&M5xKNWzY-x?ce$muYLV}EFNl+&P`wjBeu7OeBq&IIQQID)>b+sL4#Ua
zQJ0&1`s0s~@0F}5xSE%2mX=I_Fi;Z~sr@q$Nr95WbH<z?syF#FKlG!#^MUuFWP&wb
z)9AeL;A>%#<5iXw*RHJ)MT&YO#ySmX`onG3*01u=7k-b|y!!RL`AzTO(4k{UslKjI
z69}os0BJ!IN8EJs1V8;#f1A1adH&76{RO(c{vZ5<zr{;PagY@8^~;WtFGDm7^(L;r
zP1PTv#;ivbglOSMCXGg&B#l8^FZ%BfQGp-1l>r`PY<+wYfkM<`UrU#j&sHw7aRgF8
zS!ysLp$c(0tPF{h22(it3-esLw#I5lGdnv)IMHBeGt3~M>>XldI7=x50-@O)D0<T;
zNsk}n(sLL1v;Wf@*xEZ!o@?I!r|xBItIN#d6my5?2|^h4D_j+F{pva^YwM&@!mvN$
z@}(=BICe9g)vIiEpC?T-CMH_m!6*HJUR9-!VyjAw^_}llbHaD>on>(4dG;RrB2#a9
zHztS>Mw9n<s5hojNL-~E2USwZfOuw}*L?2})4B2t#id7Z{SB1LS-G*!Z~p4<QE#NQ
z+ii{<T_Ol9!(qYG&pyY^#}>Hb_M5qU=>pwLPqA_SS(b-=cC16nh)OsLS7HrRj5Uwr
zJs&BD3Y6C<2Z~%9a-(_f!Ua~gHt7xq>~^-e=bn2xaq1LdnjTyVoOAT|w)u_U_!U0>
z@sIJ?qmNRQJ`_HR6f?6Elx5EU_h0-p-|_bM^ZM7ng{jGz18v?xzOVDJRw<>|V2(xf
z##pS@Of(w2{q1k%)1Q6t4-{}h3c?^nYdt;_zBMZ^GJ<i3-Xngz`#T6dc-i5o$_(c$
zM<=IgVo3u<K8EQBp`#u+LSwL1K{@DPs*+H7*L5iaR82AJ4GEK!APmUIIjbJ{?afD(
zHmJ}SaK;$QqCi)^a~y;rQLRQ*mLzGNHkxIp$g0xk&zjhCQKj@5%C<vGDjH%E-E6S4
zo>7Y@n3{-4gMufX`aCzEypt2ROtH7;xc#nEoWAQOWB_}+ecICt<ax#V+BP#YQ*3Xn
zu(i9xKtowIurBl8t5P4F;l3gBVmHw_fzb|>!`TY0B97emDsYZ&XPqL?SejX&mbP(X
zoQI+{A(m3-IO%9iFY(+)ho{b8qMr?^H6ljRL;97An3$?@{~OLw>H)P{9cu->?g*(O
zHa1t#rA7!vRceOAoSp4HD<>8?(wJrP@HV~vHX47a9~#Sc=5>c*6c7fAqSD-S>#a;o
zPWhNN1iI8X3`LPs^fPX(T%#}rGYfMp%r8-t1=p`#rPJAEeSMYR|J~oin1UpUNE;DJ
zQt<rwX9$Cfo&+HTX_~OIdYv=I97rP^a?HXTgW*keAn~@^P)`!R=l$;`O=EuS$A5xR
zmVM&~{EujTa~(TdjUlOwop2UwE8I9@Cy^G)0X8o%HbF>DvzfqPgcKI1LMrFQnzqsm
zh9lZ(jZlVE)>5yfB!#9J6^LO*6h*|fgi$r-ZH+ITDp1~J5C)W0Ns(tb7ZF7vQd+Ex
zh-wpzMus>FDI8=7QX#1cMH$X=ezneOm0;?Y#+2a7m9va0AAb~zCZGEKFObCdkn3H7
zBw}*n2r{0hpJ}dKSjHYU>}_=^hZ!eNEHFD2^TeZ9c=p*Hn!}KZ$(soFR;agfbR<cl
zgeo5bj0-sLNOE?+1@1%9t3qR<gvj_Y!U%)i-D73ti=6tvE$FgBB{iH77#xjelUB3A
zXmb^%E!~xi9DBn{SUh^1rNwDhS2s9*^b~7r8+_@(FETMz=lHGLT)ldo!QOywXMnR2
zSzdAYaGmL(VB_2cHg~t_SjF|BWhfPy^Kecjpd61N(l|g0K@f(1Jm>_kdHwyo>b0+>
z(vD$&07{@jMUuv7S1}m$7>$Nxna5}pMalW|=eTldnOYiBt0&ZI332RsXqg%@9QOFg
zM}LDaf9XryfB&0!^P3-FZtgJqFC^bV_PtN51Q>^Mj+u!G{`z12D@;vI@K68QKWDeo
z@n8Rg|M}J~e{gCXUmIU!okg1p>okScn282;b&R~Tg;h|WnFT#WIIl6QH&Sd?Q0(>5
zw!&a()*Be(ye3FEh$KP<s4(=%83Z<v#8|Sbpe#$DXyGinDv8G~OKUA@y#cBsIubE6
z-Qr7+pXbQ&S)xP`x{$nD;0r5DT+?la7K{uiLdQZ4g5G7~CBx><DuwQ_v~&`cC=Q)$
zbM&U8eBq%B?5yRq>M<)ft`h_TAscM2T<0Y(zm3%|F0=gjliVE#6!`*soomE3O}&=T
zsyCTvPx<ky(Nwyksx-|;i!@C=%0mU1utsg^7#2woB*?_Fx3fc8X0&Q;td$T%NLgdQ
zYt@{d=9V)r=c$X2GG+ESb^A@MuI@1!ICeJrY_4o^<@`mW2#%gS$xX**c=H{Hc=XE;
zv%PYiZ1XbD4Z3VwKYj~ci4tmG8j2b_l9hlkRETkuT4jBjMrrn`_R#5c`OV+@Egt^-
zXE}M|1n+<Eck!<GypQSGd92Y~UB1FE{JVd}=N@{Hjg58s{R~7vDLhSJ<;E6O<xOP&
z=@);V-~0XF=8ikgaQgPUIC=7Brl)6U)SLcL5`GWuBZi!{V{om-0QcYjI!@hk6X!2n
z{F-P#AGu_{<sCb>?@sl<I{3Jh3Kf8|0fBVX1E}{qBqBh@F*@%;(PJ`^NL!K(J5>E0
z^k|54j#{HZIF`z&G$I`2R68A#nQ5v(d57%SVQ%3-MpBG2jMgZnh$F9M)Y?$wmN1SG
zB4R2DXf{hWx+T+-KBXd3A)1QPR8-Cz0t%&&p<`kC5Dz@?I-+QYPVXAV<y5xfm2W;p
zI<?5&#ts)RUT0=vhFYybnh0#Q%ii85*REw`!x1+&wkVaPH=1QQdVwgZdrS$CCJAwr
zkd2CQ=X3m}6yx|u98&mrPGc>djTLm|NbAk<v*JI8(cT`T&L)x0v9@F|8W2yn5pMhp
zS<7U*#ULxF@{l|3JVg}4+{|H|3)$T0^4xRR84U|MoeuS+&gzW~cK33o8ZDx5m*=0|
zB5e8)096mka~PJIox;#}5MXf@tW#J77cX68XKR~!y+IHss6bKb3hOLSJoza9$It!~
zM){Chy+)(apsFf1H#gbc-K9SmQI!U53_&26Uu?6mIKw{v#9H5c$+96I{pg3NrA=P_
zsyC3P9@FK=Ar4^;R+l~kvGgWQ<9r<Io8I_FZa;l1k3II}AL~B(rV2ZwOP|v!J@C*v
zhZGL!EYe!yK;f*$n36$n1oeo?c9UX#<U`66!5|w^Gnym_J=P-#X}2cmW?ftgQN2!O
zp{z7vsgaQ)s3llap>sbkcUB<+A4e79gP4mfLjtw5h6AEnN<u*#6(q5MFhHUZI$~#3
z<9ZQO)?nc#MJ0QfVv2>i<18(<nQhN={`m{6tZXyLuX5_P+XxfM(vd0Vm*yyog!AVv
z^2CL+oVn{2IOCX`p61M5DcW7(+EW`W?^%Mh$uRE|rjEq=5Ot@`cpPI7ZcNtt0#9p0
zVM~Z&A}1+?pg_Q&z-@Pk#!gc^{%@sogtdgY(PHzeCN&Dv-Q-A9aCo-G<m?2^$&{(7
zCiN-Fu-amA@eofv`z#Z!DdrasacD`he0`aTDCGED#icJiMsH`ARa0@TEO9cTw7#ed
zzzbQ0AXI{^tZ-5iI)8(mUzq2XTTijJv5HiZS}mp1+2Pu?WtOilv$B4joz6B{mU#x5
zG30s9_VzYLYX*ag!NA{Sk|d_lN=fPgQ5X>fCBxx>m9?vU@k^iM-S7HtUi0eLlcwWX
zE&RL5j$f=;A5~RLB7XRXevoFX$<O@EKVWNn$2&&_2;na&)_iL^4!WwS0_nTaVK~m&
z^`d2K3xv}s>Db!a;b_vNJ~2m83{dqNqm3=R@6@AzhT9#C(5OIB3ln4*pr!8!S;tsN
zz7JUy<Bq8u)9Jm=t}F|JAYeEgFdPmLc-%>?R-<045hXCnM$9xrPApCm>Ws2zaAmvA
zPSm0oIE)fBrfVEJRwHf%2vesK)wsUgC294Toas|J%jC=qpZxe)e)~5b;n?8?Uj3?j
zIC<<gb~g67ba{iD4j-WyDr{Hcw)PM+0Ww-(r+c0Dj3$mDiE1p)AHq54bau$H5jaU0
zdZwO5(ib&uee?UM-SJXNT@lEL{$Lkb4B72(VP}_^XwTuo5GNdgE~vIQ*j>F!ckL>p
z(o$c)z==Cv!p!t6rh>CiJxQ8QBBh|#XtLGq@X9-n^4u36<m{7Avz}*!ddLWe6BeAL
za$}^8kXT`Sj%=t<N>W)zrN`+5&KronzCIcZhpbj3`uz@<E}!QkAN>%w-FiF2;gIu}
zUSMlylTK%ses4rk_zSiWAlp8hba$sm6h@RyL$BZC%JKz1`MV#XQExFfx5!<0-p8BY
z{C18WyUA-<tpusYog9O4;H2cHlQ;2ie*Wk9l@I<Z4}bBCtgNol>-H$h((7Hn#TO(Y
z9AA@-JMLl{Yq9ujab*Rmu}E7n%sNDqb!xHzTT%`C<h!d-`sg(oD3W?g6h>5o5!xC8
zYpAyyWIH{mG(i~nXH6;)9&9fVUXO+)4E-^xbv1TuE40=O`~9(_UC@q8y1RQ!O&r3=
zf}tx8P$R-cY?TuX(-5Z(_nf&E%aEDbC6Zc$ZvO_A88S0F%g**LSI=MJ_S<h^>F5;3
zXnLIywOUM2=3HFgVt02yuV)CXVy~R0+Z9AfM!i<UIYSz!G#f2?-5zCL5``5)NR;w3
zWg#T0Dj9CAGJVr+gh>N)?Gf_5UDCKk5Y-S4Hm_Ww8t#lWt~s06p5^4R!wlypczkOY
zJuuvK)9q~SZqnO>Y&fLZn4`O=dG@*IdG@&r99^8`)<bpdLcp-7P$-&p%beO~@5V*0
zU0J77RMg@DiFEWz%jMFtQTlgy7z9{n$;bGoXP<kPhraN6-ZME(vn`N8NRp-$#fZ;;
z;X&3mZcvmZYS&KyilSgR95Ko&k7qJ|G%Ez?8y(VGN*Kl{RUoDIk#E|te07hX{r~*~
zKKZ*J;q4E+hgZJxb=1;2#%hGoo>^+F$LfrmH&GZ8;+UVC+rJn5v8`{au&XLfFs4->
z=*3V7D?F%}02Rj=G**{Xy#W)Al*($1DNt4*iV_J#L4ZnPT%;h4F#|(Yl+=<Mv9V-D
zNl_JqQ9>BS*r>v4O;r|DrN3zeDnLp}6a_eE7>x!90clzz2z;<_J{mC7t`iD}t29M1
z$y%?*<gF)JJbjdzTc)VBYZSczMab3VYdm-DGK<HiSXhj4Dn&@em%jWkDh&A3cYF`E
zR>IuTJ2-Xv7&oqT+1lJ9uL34#=eX<Mg2DEP!E#C&9A$Ou63te^U@)MbdeP~=P?det
z=6FQoZ%Y;tN4PN|Tv$hyXOx3JvMhW=3f7Bu9iSB9)FQik6;sU+Gu+|E7e3A#Z$1O_
zEo7~Rb&~ENA_;FNNgG5#MQ5*%iYn^0IyXnttX{vw^N&75Z)c6|%Cb_y2!XVEY>e^P
zra&obwFF&w-I{h#mR_j5y0y(`AN~T{on7{Jw^?0XVP|&>W3<nx-0uoFht~R_u*>tp
z=UR>(q0U)`qnzDNmpBb*Pc&(@8Ym?x%Z%q=c$(GK8$9~x7x};kzMqpPZ}Aum1YsEC
zB1frxB$Y&w#0ekxuJ6KG_<Mi<e`jNJ>%e9EqR8lve|hn;@dxSn8L;Ee*Hso#SppQp
z(U2+(Xj_5C(p|klYD=mrrz#3;7!fE%WlQglwGP4%6~}~1p(~AwLNDGP7YzciRYg@4
z<9IELbs&T{(&Tx8v6`|h$%iA<I0txVyTfcdr5=LNmSHwa$INo<u2Y;^n1oc22L+ve
zkB}zqxJj=Z(3+lQW~N5Jx69RQ=eg(1EBV4_pT?Aq?p}wB7q4^b)NvLM9pbsO>+Ejk
zeCFZH$nA)hsc_O#s+hrELTjf-7zQL!Kr?MJJ2lH-Fksjl5ycUqif~r2J1l6OzMa{Z
zyoS;SD4P>HO=t{{fA%BXdduC|i8-7MKn1u_kH>!H=Xv4NA7Rw*)9(!#4oB4ODu?wl
ztIb=HafA8YIkYw;agD2M8<ZwvZf2IVm0)ND62Z(woAzMD$XJ|EfTF@6t;I>k_##EW
zsL<blWcu|_p|FCYbT~4Ks^rPDkMaD4XAr^{6h&E)=OsnqBM`w~YWn@m2fqv0+Z*6y
z4Iv>HB}!!shkbT-H@R@}c^>-w@AJx6-p`qPUdqW6w=g?93sQLy`Z#hbP>Q?nxPv?Y
z&QCKO4%yz>;p{Wd^2y)*JwE<BpJ1=s{g!s^r1BV`efF9^KsEa@Po}DnWkzB~SP@`#
ziPRC~Bed?3Z(K!%lCYi<1tGy$*KB=ckcZlr8WGnpaRkz5o;#;OYlMi9DnLaMI4`Sl
z&QOjEU614Ox;?8krpU>3fif9(&>`zh(rDL+WJNGWj|{UJo*Pb~<^r^^%rDOK{Buw6
z)ML=^<=k{?k(+L7(&=vS+PA!#yY4&0WP66|*H-Cm_c?j{A+Ee|18GxwI|Y|s=%TWa
zx`8~A^hSbqRS{K{kJ%EEBu)u~kYb$n6UUk`j8G!+F?H53*jz;*nOHi;g_Rq0pLmSM
z>t~R00~dzaGDEluog0FBfH4_%^#*qxT443c1~(l#!py=XQ*#Z9Tyy2}Wgh(Oqg=jv
zgR;zd>8*<-+vhp^mFIb8Z4(oOOvR8|!*e@5s<NOEnkF^sm84WTo5pf}Xeb?oQc+rW
zz>iY`YErPfx6RH@hcHTrQ;ES+nUdAD6{@OaFc^+6_s*ka3OmLTi80CqoRj1ktgY@4
z#}URDg!GYi);Nz#>g=$&xz2Ovp60D@eiuLT=YE`8Jw-dO|MdO<>$SsTe@Y{0HQT<Z
z6Q0w$Z=(F8T;H@S_G`0Ty^tMkP*p)3N@{UPf3F~*BD98VZ;v=l$c7_amQ!UJCJ2a<
z1Q!OlKp|pp%P+FryC;*##}_$^w2mMMkxC(iLaG2`Kx>21J}O74fGCPEM$_%=QDy}y
zi5TQ1`Jl&ia~f$YoHm%MO{bbcE>7W^nitMpW_EIsIyG)wUPeWNos}(~d-N=$SKiL+
z-uN;`d6#R~p2oR=uROfW;S=-B&9_-NG|&9P0$=*Vkf)!xKxgw3uX^n*yzJFycyRRv
z_Vgrf7&08?*ie$jp+|fe<8{hrT%1{f|CU=2j;NMmg}2l@ZLw8_s7iEyfUzaomP9%@
zaMINlj*?}PEH+F<j*Vx&z*C?27&|7xshIZUEHjITX-&*9vvh=`N7@9Zk3(acwcTAF
z|H>tHFF(no9x}vmqtf&UvBF{8*cm*2?wk<xN8SZ%_*$4qr8Qr9^l_eg>ItN)h@!wF
zJcPgOI_ul`cHbFnj2{Jv5a&Ew%vkG{GUEq7FAT#$$zHurd!kO7MnrK)RposC;ZJet
z@&(@izVG49Z+d_vNfAimGzA2s{s3z&p_Ig7%=divccUxA-~D@kk8ZDb@b~yutssnu
zqVyo1Xx}tpwZ`NbvMh*9Ntq9k6D_hqm$)VogB~`pj0Qbo6(Ey{wB4fY8sEt-Gpup6
z=jzB<A_GlTX@n9!i`f`VReFc4esO%ZUsx)oyx`mUOG8zaWF{voD?%YTva~=T9MUBe
z)dYDF)3y=2*LJzKwLv%TK~y3(r+LY#CA2i0+w1b0SI%Os!CK7=&tIgo)5ln^RT>P2
z9y}Z-1S;g>ja7DbcbP64G?S1}`a~HKOtP}RMWYd7(tspX)YAq@oFMS|%TX9$bcr^O
zKvYyFL&PbujnF114I|36Rr+C@`7^IY1Tk36Gr#j|eBqaVj#L;-<>>cv1|!XS<w&3Y
zG$&u#WH@;nCr=&X{Q0Xy&6M870jF+0j*M#@I(-)_-5y)(*Qv#pBlUz%HzV`Ib{rCm
z-|K}l<l6japN+4!MhHLi!x}-JTguXq<s*V%L>PD%s5PJ~i_v}$`t9Oj<5gt{l%wAt
zP}yRv%N0a%NSY>N$;yC!f0wJ*F7lfn`bCzOj`GfTd>8L~?{}fnlwy>jq(GOxc#=Xe
z*={pAG0BPJCwSd!Ux!iwzxYePG{y>jYu5gSsH#egF0nfING5H3yTxe4Xb;K(GKz4N
z*x?A{a?EItKtR~2k*E-xm6WDLMgbxW#`-yn!(xR%qzP8xvqgo#=n7#CLMf#5nqjT0
z@jluw2>c-}to2cqfh2YzT3KpgN!i=R)Q%vc0A)0hO1PHIaYNN`%5clgH?h9C#bb{=
zPiM<vs(^FPu5<fcOQ_n2*~M+9CtG~<!@tkUwJmCCokt!%%ki6*$a{*V!%J*Dr>X0b
z$*>?4yJ(l8bq(D3GF=rIYY4-Ls?v-`BZ|Bri4&xAL~)3VYfPWI12;~FvN*P$dV)hQ
zc?GUH31Pte&9@LHZEUZks3%N`m{g`f#GR*aXZmO#k$@{Qvi^{{_6d3gl(PwkY6<3r
zhuB*=k5+XGGh(kdVqh(WvGla?G6i7?7|_VU2GFsFV(jiOwE3E@mj(u9$@<nN*H)K_
z(vXH}f`rSLFLM3HHHxC7C_GdgKx^lVI`?8x=a<D;!R~gBbE`~EH%a1Xe5r5@219z?
z5!MQJcZL}KQQGZEzVG{fh#-j2I!9XzLX4H4{@D$JfHP<A;n#ojLujL49EJZ8uWzcb
z)0IEuzuFCDoJBc9tVe{|5GMx&ZUD9*A%H@m6X_$rwz~vUqOGMCM^sXJAh8fQV=&fG
z8ch(#ga|6FG1d_Va*TYC9_Z@)MiSPbwJ(%{ARtW>&>6k%9wHAYox|CZdI*8@qtH<m
z(${saTscR%GN54&u~v0?_1!lU_tv=W&fA!brupcFJs$r2GS57{#i-h$HIcBlH{|?t
zpJaBf$?M;EH^bpBft18?$kO5w&Yyh}JrMN!1BNoEYZPs{z;JMlu$hqOIa2w<)EeUp
zJ3PhO8cmUpuxSGoMHnfN_`r5klq5Jre}FR;r71~7j4&20y#~u^#b8*1wH%Isre>ns
zr5G6OV1s<+JbR_(Vr3~M4AVCG<YCIB&ak`1LL*{hZJQfeL09<O|AG2*d~7%@#Q~G&
z4=xAjN@KOZq2y)djd<Q5Hclq+5BPpzW35A*0*5#lO?~y<K{(Kr9~<`jC9PIWvsI&N
z7>scUuJMb%^zT_-UgiTI_&(<67C}Ii)^SJ%-5!*hTB8QQ_k7?3Y;JDx&wut`lIQt>
zEBRZooO5HOh}YMdu|~;i-wjZv#1<nOE+;8>F=Y=W6P(fvch_JzBuW#ALsS$o(MWOD
zQVvHHqa3FsGEhi7)<RbmMtGN_72~{P_HjY#fU0k;@1Sa}v6aC{M^Ti-VU6ke1(XOt
zYfLdgQAU`4&dOuYp=7|k42Wyfbe9GU<&fRph{JOsXSasz?6hc3hDa$nd-gnKQOA^y
zz0LqBBg(>Y`QlY3XPdmRdX+{nVlph5Zb?FR7<R7Us!h_mpsF-B2^oxX>S;);*<wp>
zdTD~Ge36`I>^}1-qx;`PbgYdGVlV}H<zR0}?am3pIQ6j6YQ(c&_#F8lXQ5W}V{u`~
z%Zlx+AZCR{_v>t`X(ms<ikF@~#`?g}YNZI_$cmgG3@Aojq|HgwCbcM_As~0=z$`RQ
z@EspO|Mh3hOA7Xd1t6Tj82>%0G?>T{hZbwV7@r`r&%5_2z>2aoAS~Kg2BYi%XA*?!
zzyz?Lz3q8`0iB%zx-`Ub!aLvfK7uf!$a9Zk2||RmSZxR+g$yHF%@+6FcOM`8)nCPw
z<sa=v`7eEa#+NZ`ExO3j!yaxlK#aPmaR<p|1B@%c1xOr1Ly&p|)_xo&<7T}G6(Udv
z`>4t?l6Dgl1jxi+M8!A>$QRV!C|GIlaCKVyvD%n&pp-%hK~+=~x<EKVWh_Ay`F&WG
zpfyAR0zp~Td3Lvrn{rIfEif@Z&1XOT7@ZA)sRERcWJAl7kFC;}l5{q|z};sS38dn#
z({~ev2|L@n?5uR@b_#T*h@Ix-R8DHvFoP#)&G{^IFbuN+d;LAntV|N5RFru^k(ZQt
zK^aP{aNPVoKg#SKufUWUJLk^A&K8YBHxud#Qw<=DXdXVniPyiEbHDNn#Cf0f3zs-@
z{66B?G1E#Yi%lN-tzTz&?rCD3p@M|k++mKXkm1@z+L1xjXIUMA(-o0~t`=+>i*a7B
z7~qIRKu?C0wxV>^m^JqY-`v8%D9icK$3DUfFFecQ;vy6635+qEKYxzR?G1{uqAI<f
z-G}f9EMH&L9XxGl13Ozed%Fce2vH;nBSlqd@{#9{4tg5ruJfS}e~=_edCz;ki&|Po
zsgPncLO8}3TuT^+yyso-;<tY5qkQF&M=*a(#-ndeV`t65%V7V9aTaL{f@*-v`xGOK
zwc0x*Wdk7;Wf+kr2}R|ow_4~TLt9H(mIP4<#$YS$YY6Ef#ws2=j(i`>e*uICa{BNq
z>nYT>8fQm3YPC9Pn&7NuSmkJIhy!TE0b+cuElZ7-AvP|dg6-=W3)2bpsL!GK452qk
z!zHkS(Lj)mN>mciY^O*f$}*z6T_SbN!t@ay{PaT{J#v(}`5B5K*nY~nXU;OBLm@QI
zO)wh0K(p>&lxf|=HH|S??YzL+I*id6T@p6i$T-C~j}Z~ZVyluy98q>U2rSxG<A^m&
z6&M<=CejE-xu&QzjY!f=BmoZD45+FKql>Y~R8kn&iQtkRaPHzY0%_Sb89RfLUHr`m
zT@fPCsDoqUYXFPmpMK2<$?Z=!1!n&s8y9u^I4v)a5g^=w*3FK2TKm7rzDRq&kLk>~
z0IV#XsvsL0CZ-I!@|_5&LLPbKOZ2)!zV~~+k7FlKA_7HH^IDL+KfqL)IEoM`{@f4$
zFw0l3@oT^SYiRSI@1p!suM3wia^}vv54vRgOGpVpfU6W>NvJS;*U`ffO3n~Q67n1s
zg`PR*93qZTap*BRxu?nxheN_f4JCv(eFg#Acn+Qn0%REap9?`Kym;RjLl`Q8z(*)q
zXQ-?Or<k2xA|ymtma3Ye)1Rh042YyemN~7^5TS`ii*!~}rf(8#ZI#@9Tyk={&E*SS
z-u(7ESiU-7?Zyf^pFo(3(a=(rJDfXrfsOSJ$BrDQEMP8xg;-FxyX>w%%-qC~rK2s<
zwB^C?LLgN@Ss7Z5HgS?r<T+WEAqD=BEetv{q?H-#SfScnr`Xyg6p9Ls(*7koss!DR
zW~<6+)FOnEC@V4AvavIw(dcosG33HWpG7a+!I?kxm)P8xMp#3;mhgL@dXyV$*QkmS
zy?%#bHcpw@M-t)Pa$ijm_-5C^@#O?q=cg6YVKv|IaN7qsmxZIM%)$N?(s%NlGvlIg
zd_f146b$+q?TI#@`Rwnpw6w%aUiu1}Nt>~D0t}SJ2x|>WD&i>S-n-ARFgM4w8#lhC
zQoDVIsws0+e;2oNgJ`r%LWvq<VPY&Mvsf#LlF+|X#+j(bSWFxsg4pjv>&QD@M!Ow?
zC?ak&v3Q8Ekdm-gM@5nEtXCzb(zvQZIWyM0`p%1#k~oY&JB+P7Gjc4mvBn?}6qWCw
zN-WMuuI{v1iQ61Iv&fOf+sL<expV0#7X|@QJL2}+=Nawn^30_kxy)I))}`G@Ied7U
zFMj?xhNGOt#YxgyN|D1e&tK%GXn^Wmq1*WiM~_s*!3+XRS(Q|#!e}q4nrOFCQZXD1
z*j(T8x*DsQeaY)seBFC+MaIUtr?~#nU*h<YL(JcGC#pT|ku#$XBAw!%cYZHVe)u<;
zu_ZUoevwaXY;g3Z({y(?c<$_D)VkXotviy506__ZX8~B8tCI)QEU)wtG6mx`r(P?`
zc&qr2<!En6*k_r#Z+snhfV6OR`6?T$*NEbPC<;iE80Uol{7XMd_0pelQOAGz46qKY
zc4#dq3qufkkhn44xM#Fsd$Y?*YmE<m=!0x-Z1Mv?@MoExo&hPz218uss7ixU0Y?uX
z;h+BFf6Pz+-JjvX&pqfT7yMz~4*&V}O%-;55GWD&I3oP$P~Zs55mDJEl)3K+l$oDF
zS9_$5Nz!DJ(8O3>5XF+pIdoA_3`zpy!1(x(B#a4z1cbs0hl(Pfv1-OR5aDl7YAkve
z)?vnQQe6}bwIxjwq9i2OIVLZNr6h?%gmGX&BM>S@xq>i>8Ck<%WVyIfkwk55dxCgs
zf>*ryDBX<?t35|I>r;swj6zz_MZtyhS9sN{?&Wo_djrizi_Oi9stnm1Wb|``35EnJ
zrQ8)1BaK2(>5`Pt>jaC!cdneo2!n7HLJQ)?6ha1wN>d7fGzGHIBuPLq>fl@e+Tx_9
z5+TBr)Eg~i0K<`GZ&)#%gd|}M(jpWztBS(LC=@aZ=~XR;wnk@V8>tL~(STm%7)V7i
z&fwkW6uvlYeX)yh@XK$wJVIf=G!xG8;$>*RIQ**r-i;g8-(Z~cBIVDGBVPnn1(mKS
z%078+a7H0zM4B|Xviv;%;s5>x-u<5Ua?ic@A;XwBt$B<=;dK>J6frkD%m4B>{stG%
zpXc!>pYVl}@EyZH&h;OE<rg{qQ@0(&%Iz;PX4E-nkwp#DE74^aS7nG2bndXGB*ak+
zM&6)gEk=7HyzY(s!65`93J_uB_o0$F6xMoO4N_vnxL}roAg*DGAyg${7!b-DLmXK@
zCrO$FDgtfD^$gp4lXOQlvQi>}P$`5kWY%E_d=nxKx!&91ndM0iA8GR7L)W=_eTTT_
zz|3H(fKY<78b`(I>MD6&Q5GYrLSSnetxE>IHJWjUV~3}S0tLb%h4Kzwqp{9Wt0lB1
z+H9<5WTS#w8X=t^X--j_Swd(->2lay;oOIRm8Lb=olR_2;)06GIZobvhV!5J6subU
z>UA%eZ%&5vdpWw)bOwgQGZ9DUBeu4l<%JLbKYZu+|0Rk%<Ltu^@yxSNk(rV#FX;+J
zW+i<DDCfDJPT-I5gYt?$?CaZ&OGIm-D*c`^R`3UpIKTgl6JYl*NBgD&|Hs>Xe&+u3
z?i^WG((4WA>~{FbM?Q?Tjx%TOBaS^&6Dba~p#V`7bLPw${>D%KHU8G${@e6N*&qA<
z@t=p-{|VM<)M$tLXq~pw1WFv-4+WsK*KNC0;B<j4av%I|43#y=KoLlR)P~;nE+A=6
z&w2eNjzCDH0~yD-I7U|$b^x|C2<guuY=t%&=lr`xIf-$`L(_%CVNiHYmXMCB@_J=g
zR_u(9veG)k?XSCy*2EOM=dLmx?J{|6l4Cc$g5}NYyzX^(p_iXyl7h##3N~$*7oImf
zd-fV(EP2yg@8tfs-p1bEHb3{zejULC5krb%o3b2{4=lEF<P<0+NTU?j^b-&%BuWCh
zI^gj|Va$=&yaQ<+`OYTK|H^;hNK?{WJVg7{U1TEmV@y$!m3>;XOU&KzQie}`hTEnh
z2E7YpXO~F?oQk`s>3~{lkx`R4tRYMR7<4+z#p@l`I$QKbGCt!SHD`z&<onO+`1#!L
zeLKG4QE{-m`A3lDj=ZdpvO>xH;MuTte=I7-C)$5|$IRgiLS2b5S91Kk3D3wK4n}mk
zyFBsamzkWH=1+b94-!TRX-knhkWS*9#^{QB?>WQ&^{@X8|MZ{z6aMWl{9E$>!X)eu
zv%aaqE*27jqBM?33a=rk3~DqW$a=`SLR*P4j#?NXY9TVx$fAqW5re@lX*}h>JI3Kg
z1;a{HZ#J2noA=I<I2{L<`@{fYJpckh{l1RD7~>CMJI-(u3ZX)<4u!x4zI)L(6;TwC
zMhT|0=)fVv03{_-L=;Y-OG(}}oVatA{^~Bfy-gnca-T2`m^*xln{O@nj(6P0wFf`L
zRyfDy4TGK#EbndMqyR#md3u>nr^l_g-OPncSGjayjoF2i!j<#}Awnn&IhMlf<g7zr
zJR~_AQI(bdJha0~0caADQfto=IDxd5TtKWds$PalYJ@JwXbH~ZB26VzMCk~k1RH>@
zpr2bx42^n=*a(ChAmSFWsX^Ex7qGroaHB8`@`{s(=h)g-+}P~VFa1a>i4s4Wvzq@}
zB-bB2R=)P5->6*sFaP=*|1<Y>A2yYsvok`M2819C1EhjrH{>^e^Mh<}Z}Fx#zLhwx
zA;OTrg=lRMIFdNwrV}Um>woR9@mGH0f1=y(eM@vgKlHx$6FN|1nC||vLSlme8^;7`
ziWLGUB~cK7)qq5WUeN4vua(oo9IQo_=HRQdFOkqj6NRD2Ybb@ldeq7|PRtqOUDZ+&
zgeg@yBF{$tFjSHxOem@fV=I&sC<@|0(@Zi#Ww3dRp=#45Atz$cta0*Ag-iEXIH4HW
zfTRU)eCa%2xweANB19D;oIyo!?AR<vj?L2T^tgEO3Mv?ps+eAovK9824<?99NmZ68
z_H}>(mGM6hBtSsDUPlN?RaR7?Aa#<Y)h3ARl%~XNukgix^K<N-KgWrs8SM5|%wCT&
zYT`u1nYTQ^rN<uSh2QxE{efkA29C{y=*5sh=JRr+AfVlxK$tDAUiva${MQ4Nl=Rjw
za-`OvoIS+(=PytJd8IM<B3cCsTUK~dnfm%C_z$(L0}SKQ(TmrfKe)bn-x&6_(87J~
znD@msc9t4%Na%LEeC%T%r7A1V+<PCjTEi;_m4f|unQ+w7lppwkAK-~6pWr|J;xF+H
zJRyH<ODV6nj`IOYIUs5zs9;~#VlZWe(-qP<Dh!N9K4&^mI0V%wr>ZKXgE)*3O3`Sx
z36g}$S%kAFDNxSq=JZ%TQ&y$Fp6(a9Vjs682vz8Th2u$ub3WhLPh>2WF-){3i33TV
zP4boeG}|*FI}g3UT_;YmIGM7!9q{>Q*65oR?!0%F&FyUt#~Qc#7_XSWl`HiotAn9;
zq92*(t?#*;iK&=J9{VzTqh01FYiwTEoSM6jsJ6)Z_9G<MwmEiemchUqYidb~bY7?E
zWiA!Qx&w?^ka&%<N<xx}X{u}+Rpw+n+XVF~L>MAXNLG~CDB{kyzmwni<R>uml8Hu2
zECp#GXo!TKDM>I`84v~$7KaHF)_aC?YkRC$3L8qoG$yJ*IKxckSVLo+&vhQ_cmA7p
zkS{I94Fuym%>vIim19S>7#C!Jm{-2T^C*2c*Z+P#k<WSbT2WHziW{rTWciRVn(z>4
z=cuX*Ybt^uKv$Oe`FZ~OU;jxy{n=0R?70`d`M=D6VSQ7DoirM&{g~AW&z5mjfzE~m
zRe=&UpL?q0SOD+GZDm;?g0b*bf(QfT80c4KIaV85(^Eugf?;fou;T;UL7C;4vhonk
zeQh6xV$8C_8DBw%3SCuXy*^2pqJofGJ-{)hgQHL~BoqphE1Yx89ICSw?s3QIo49&;
zgGN0j4mFKbGupVw{Ov7nx$AC@y!=)^^vj=Qd3nfudx1lTXQ|Z#CZ`*Wh6zsY(%)O>
z(XZ@LjGELYLS)s)xse|u@uFE^G|JK1jFUVBQUnAlBu$$%Cnu3Y;EYB{9}89H8A&pU
zG}a&P!n<<}8eugu2oNH`8o_W_kQbJTR*NK#h^%49wTVgDs+yd;dYOxx>+D5M?zs01
zSw7(6#bsWM783pe{1!q-|1fL6$nW<v^!4?zz&pTbiLNZa`^isGm6``0cn5JDj|uAn
zqkVCZq$zKC%bR)UJKw>t{OYgyQOzH7ad-Nb(?nqnZG934&a({dR~^v`kh(-WO{9EQ
zDnUq5Rm5>XSW7U2A><`d5F(<;LwdEr3a?`{rKZ$|Abhcabd{#aGSJ5VdE>fLJ4a>=
zd69YHcxh20AdYJkWkwn`h=L9??GkGfa$RS*so1ec*`|(|3Ay<dhZyQjHg`b=nzuf1
zE3UuI@e^~r?0fFvlfV5%u9TZdr#N%Zt-R`$x3Rc1$J)vkx8HS`vuD4|?(Pdrh`U%%
zEj9>Y=L{QbPjlo@0~JUF7CCl~3%v8F$a5H@c#NnBh>4?Ute-OUuU}*1g>xL4Y%)2~
zB6J4R?Vu;7aA8E$n&4eO@ssSWtg-sY(*&+$qSht|Q%tkOn2Ip0Qx^uBO%|&md)rSV
zg<xuCl5O4ON-slJp4lbGdRgq)0qwu?d~MhJ55M-+4d1*j^hM>~f9{_Pp-@tdUALNk
zzsFNgJ<0Lo$EnxqSZBR3Li!xk{gH3IUgx#1eGUKV7k`=m-p}v9WeFh;qy!m?QXA5$
zKoI$HwX;a)1w2YfWFnCCJQO^My`ViSFhz-}DvGkAP>R;<v_Ek7;l1vwVjr!gsw%S4
z2(7g*46Mf`N#Wi2`{$C|FDk}3Fq|XL3-Yp}*_>czW*&^<xxG_doR;i9(}gPL>FfrN
znF8%JDhw&gj7Hj`)VmBq!%eTgi`8e(b9g9OY9C?l^v#@p(;ZB<n_RqjnTNmh6>hry
zCOWGb!zAGOk>+^g7UJBpzVSJxr!(3Ub%uissk{_uFc?vmCB0q`YkVQ=bjk9kKEmyP
z<~XW3!(D&w$GG(24^nhC5syDeD@{pfmk>sxgNQ<mICk$Vx&58*=FwmN4OHME*@qTu
z9G?n_OF`%~xw9Cn!5OSjT;1BF8^@e}`CT+8=ID02T)uLFi<hnwN~k00W57!Gb*kUo
z^Yyh&0d0NXTuJX@x9*$k)_wi*XZINNFrID|MMaTURF$QwjBl@xqvvj1zs@H<@i97`
z9p3c%w{Xv0_hPMOZf=%Sr%wIJ7Ix*%7WKv?Qu+40(FT`~C<k4lG6xqU<TyTRe0?)U
zBZNF~;~8raV>3ilOS~(@ShBLjgbKBv1M8F@HI<cbft#^g(paPrR2T|fk(Zfow>#}~
zV^!z_pnHAdC}nDP9u>uuwn7+5ErNPba$@N?$Bs1UHrKGy(4KDcr|-WVTS_inIL9+D
zJi+3b+lY@Ypq0g%g15cpRn#Y@i2}#uOq-${(B0c3ie?FF9sd4L{d;QeG_C1Vw3`OI
zqOn;;I-39%rKHdK>yLba;UGj3s*rlEL8CrJ)SkrPb#)@Nl%o+?Lz*Vos`TPrH)c(Z
zwFgQCNEKq7B+o6wQArXPOxI#cQODVkZ8yzlo_wCYa>%X-xc8MWV^A8N{NmFTSxGV8
zK>KdhZ}+vIt!pjhxn^Z$i>h)I3mRP+s<!46pZGY=Ssr-c?W9qRkP_<*%D;E1HyZqf
zANw&r``|-d{}#l&I0C;9lt&^sttqP#T@{ETLuNfpcb9z7#{@aHmLMuKF5CrDD56F}
zki=N6k#1~g%JKTMD8gBba`>WeoHSr{Nj@5(%W5nb4+xa<Zo+YwrYcLyqVPO3T_A=9
zVVn@fh9Ha(wvTW_;?SXjS+X*sm~h<nnnSc&Ay0g2kM1C2?nuGpe1oOCZ$ib2sFw1B
zKRCsubIar+<iycMn$4K0*%q~S$;=^1YxZt__xFE?b~Q!8aW2~-)znEuH%G0VXKAsH
za}|}Y2t!4^k+Rp_VQ1Uxf<!7wf{<9%3HC>~P7zN|QfoArota>Ab^_U+hI(r}$~IUh
zNvCGG_kp)_?JLhR8dX$UP>&;Q7*a`1ssc(EplZjMi7J}SEu54*vjesjvvvLo+Bu{Y
zL}I_A_4NnFx9=rMBJd;@g_HqBQ4)q3#uzSMy2#nHXKA%3Xw+&r$Jjm1n8E1~3VQv)
zw`2?pV2$+h730U5qtS?Zo*`@kiUYVZQjC#O4pWS9#(5dSIJrnB5rMVXQe%sfyvPYf
zf*kAjobhH6X$-p39;Tk<?0X!A@UAc^0}pzxDvBZ}mhpHr=TR5NIE+v<CMRi4OcIEg
zPJf!`%mON^NLz5*9Y<(SH89pv<OOLxWPY}V5{|v$F3Y_QbXIWFeRm<(H&{LQ0#P=D
z&T{f}$cI1tA!=%xY+!IUq*96<1Z~}9!t|MGbP1z6#*|dLq~9}aZ*O|nw64(Bfc158
zDKwoge3G-BU5>xwyO_A`UT%5cpJw^j{ymGk>ui4JqcFWlEG6}0HzQ8oicIU=|KmS_
z6N*QE>DR~y1&x|O&b4XC0O2ID@$WFA7JG%@>Y(7(m%oaQ-64-Z^?62v9$ITcDKJ8i
zSjiC2?fj!!zQEfb>Hha#vK-cTi~MECLBBg<JKAMoeu>d&<Pm4}1{W_};6MJS|G@R@
zSE#Cj&CLyNK7I?eW*s2~i;IhY<nespRAIOK!t+c_&LZ2>UX)xF=xl)L?$WZA4=s1r
z2i`jmQnk)UJXKXiy<P|Fh{6b^)WN98RR-f6`Cy2W3aLE6H3&oBwW|`;IQZP*wL)_2
zK*Dige;<`Y;wbh|NFnI;2N1?Ir6P!8qCgOe4(+6aNRCpM6I|Ii&%*p7FPuNe3oksy
zJKp(n-uTY@=ydulpMReFUw#i#TN=|7sCtD=6T1CA{oa5)9}z_@e*ITJ%<upHLlCr?
zm`IqO*OdJs4H87LOB^X5qE1DgXN*Rfj}kJLC`xJ68?@S!BvF%~R;Mx*(g?s(42MKv
zfC{AVmX;+Za9Hbkff$ViR1jfpK$cquLrtSr5VZmdH9@}9<105d*~)WrWw_@xuOO<W
zeDTx2kIqW}@;lJQ{g1N__JLCrxnW~tcN_}uquH9xlHd8nZ{wWd9dG*%!Z7l2WMd?O
zQi^--zMJ>I|9$+cpZobgeo%2Sf++C%8>cXpMduj|JIKK<w%=j2wL;R}CYX*7ItRiz
zzaM-XfH00=pRI)RhgB`b2?#92>3EdhAE4q86?mLOkjAJWAPfSOwv5V(tSB+1Mhb=c
z|5Nv$!IEC-eJA)k=5^`w+VWK>6l^z4H`*I^vkfUyEs~-}+L#H6;*LgUrS+PL{jgGK
z%?Bw)yBiZbvs#*1jzpOqky<*@G}%qEEgJ36@F<unl)e6@&wSl6`{CrhK$A^&0}Uw1
z69FIs_g2=Mna4cm|NM=U%uS7x3Ry0=`pPv<o;y#k+aprJ7q47M;<QJN;L<zK(CKSN
z`*Z4CBa0sObi`<H4t|F?5kzT1FYEB`cfOr;rB6}LDT|yW)AZLGpbUTT2UoEw!^Q_t
z32b?WYe|Ie-%GkbWp(8MW9Ou4f>MILy<PV9_E^kIv==m5DJT0I_-uu2eG~611R2@J
z3HqCxbo&W%;|$6Be+1Xaa3Z3rOQbEp*0k0)Kx!tlD&){MEqZyt=s00sDJHJV&ZNTl
zjKOrmwX&qj3(TURsUs=@%KrITK}YQ<wd5Y@NCmUo$%LvZiK3XQD*2sH{}$J;U*rC_
z-N&8h?__;_Bk&#ZOJ6U`Vv(by3@!e*__@7xI2U3F%f&2QMr#|oNm_?OV2PA?ZxLSL
zYeSr-2v6W_E48!_^>{78YkXy>>p3FQs7NpEwZNB-vMea_BD4dSNTm}?t0HGH1fRlu
zv7l51-a}m(>dKNeyY$w21X7EnG?qzy3g2tfU#o~B&#WAA>$$5$NkpVI<*gFODL(qi
zhq!R@J#1gwC7+BLj1JN59;aF>tUq`+{mo6j`js#9^kY-n@p*cE_{0yqn=gI-Nrs~#
zQ9PrrUgqQ-J)%r9U(6T}hb$Iz7I}VLG-(9dN<^Zlgdh<W^JgC8`oT8-zxdlU?|K-0
z`aE{`3hHnlGZ`ZzMX~iVa=1_Q(0fp6ix2$t|D4;;zsU8kJV~lOkxt2E&DeF&B1T)s
z%y;<K^;;yT&oP`8JpR>hBI=6Thc1Eww1C*VAb5N$=FpooOP4G=j0rio&N=oEMr?2I
zu)e;*Z~VruasJ$S9{tLr+`4t0yjWmd#dtg>&ljZ4gh<4E;u9a|SAXr-*xB8C)94JZ
ztFU|J8(*b=`V6f!0o|mi7GNrDk)u_J%dtz!yA(m`sFXr!g((e5k|3Q6oH)R1i6E%_
zd|8KLEe^)Q@+g;2OR{c@C`+iyihMo~$U^Oq9<-1+8FDgf9g3yADv@4N8Ox2E+nn%<
zlc!HYT@%Xzt!R(IWK5IuY<BPA8&5vRnF~#jO)8^lbTiIgI?q~f1r<q(*@Aj7M&~(Z
z4%1o5D_5^@<;oQ<Uw)bGtpl3vRnA<v4^>Ep2L*o6qTy;{d6hUhfn7#M4F&^@3Eoyp
zOB#&^?N*0I*21BQl5i-iI@~LU2M2W8tz%P8Z)$`JXr>i4rYf+`B9z9XFve2V7O5km
zMh9hJuaw-}yG@}57azEbjT0+;_OoB1n9R`52LXeUU|yR8{O4gDsgi<QdQ}<@4hF#*
zqOO(^FqYr^&0j~O`NSuGh)Bnn`bZ6`iQ<@_`kOz+ul)M2bK~aixAbX;FFnV~g^R>%
zoA{&=i0xI0pO2{ycbM(nhVcPCWr%bGk4G)nP>B7I<fft#N6U7-4eH-4qp}uRJCb%r
zl^5K8;R@5ih|c;7E9XuUH&U8$IKX`E**e%`cW0YcGoZ;*D(cEIpBjh-D_Sx&hO0M6
zJlyLqvmxssd=Mp#m4yCE2Q3VP!HhJCu!UlFsCnV3m$>xMI*+{bBmChXeu3M!pW#3M
z$-hR{?;(U{Je-p?n#5VeXfWe<e)|jDdEeVfwdC;THiz5SSn0oygH}r4Z;~&z5y~)|
zRSbti@}i)sD{!ErloM+kG`c6@;ys*x_xplsoCu*}8l4tNzfJOiABD|LMBJdT;Hojs
z8=Qlr(I9T5jAoYMc*aU^jb4(HN`Z)*U~7isg2yi3CeJ7MEN1heJ9yz6PZ1%I&Lgx6
z^X<>U`dK#TQA#eO#KO9t&1T<*Ha440R#(>;jRt({TVLniyYA<w{>D$!ZTF5%uZ0kN
z?4uv$7k=Rvc;))_x0=mBb-0a9PEuC|PNp<ZU&QWT!50>rgg7^8YNT@rBwk3Q2sS_>
z)|5q#H8n|`1peFd{O~#opN;cigW0jt8jnCHC91BMo|B=QQM-^r;H;ynONxAfw>3(L
zpr9-bMPZp&0;6KmMgyq?k(5~A)TwpK#*n(q$@3Xf2u52TUu%S{81^Sj$2AALBN|$h
z^xLf6aS9b_jIkstrmS*4{X1Vmx>ai97)>n375wlgKEh+4c?>zY2Fj5(RuDok7;G~c
zk1%D8)`~dRq{Qe*qZ%<flPKxPlp=PLMifyU?h#$AY4-ZCD{ys5F<&6GB8fHmH@--^
zwuU@&KO)IkJ-NwEE$KBN(=nwHI3MFggw-jp>^ipQHFv(_EMNHI*YVYojv|AY%zGP*
z@R#j&^BrwS{;Ujt+$OeeskC*5SFT>C-EIN!+;h*6=Q&kXhJ+3&=(qc%Sr&k;Nb<>#
ze}aGYKmGzg`*T0f@BiWF7>>r@_mjP@!fxw@7wA3yHSX<qiNz`E(h?;pD$=CO3^|8H
z`k<M$&JziRAh?Aeho38nbs=opJBJgXk{GRZKo@jS6j^7n-UUkv;SrG{Yd1;KlzKj=
z^iYa?>C2>1(jh8KEj$K?^A+<#Fq%8oy+??OD6Q$0BlgA<^3j4uCnFUrTz|I0$2V!V
zYc@`{DRXFK8P#lxaXFK{eZKtp&-3i_FYwBh1B#iT*WTdFnRDEI{wy-mj7AIAdM6lM
zKcMeZs_|Di*)Kxq`J!aLm@%6zuyv5#H=3H>YLk`zT2S)l1}`<n)Kp%QS(xr@vA*69
zgnDam^&A^@f`6Z?gN2E)=%r(bGh(R}jX-Log<Hd#Lew-)ZgAo5i#+}KQ*7PX4rE|F
zEMvjs_r*H+Pdi?FegxMPhwzfJbR6stL#!Ac=MB~we&sj*1D#%%_r2@=NaV5=7Y?So
zFWt?Led;gq|NhyZJ@(RjE5>90)4%235B~sneE0{+Hcz8;gm+<`c&U(55eZLQ36zX*
zUM)%T60aSVchq%>w=w95+AKM*&LOQQZDt51*}r~^tygZ)ZnZePew)e89^K7#gpyQ6
ziH{Pt4-XlSrbLk@jujwqUQ!hxs|pu;oQk<|;5oN2^fN{5D`Gt&jdn?NMLTU!y9sM+
zYmD|Q#={bCHG4Y;Jo3>C7-Pt1iYK0UmY@8KIrV(O*6tybxgm<Dte!}?eEAhlpFKws
zt<lJ0vLwdVS4i9;UN@+gLLP}@GMO-&EkcWHsqS2BH0X8vc%|r_Jx8P4BUh3|RUv9i
znx#Z58;I3)&@qLRxWyE?<iJVqX?44F*1POq$r()w=2b<bwMvr42z<ym&SxCva~erZ
zw$`9F6^qH7R47XCnOn=;y}8P^w`RP?R>*Q}BU6=@;c(1oGzt7)=ZT|;;rfu1CpYQ#
zyFC5elN{{s(at&u#q#1IdHBJH_<Mi%@A8lS$v+AF+_$i}`^NwHGu-vSgS1bcC0<)a
z_BU8Gdek<=%SlJb*+zlVBuR+wb2fB)nw>WD*^II-Nuva*b>L|l6XG6a5Jnuiyr31$
zewVn}z*IH)VopAr1~hP)qyxdm$r?vh*BBdYR`Rmq=0QboW;l2D1igL-=?sw^(vmU0
zUSqU>g<_FY&lJk`K%{sXad>M&F_}`Xb?I%cV@%FqKB3=UC#e<1G-om!vc0vBwd)w+
zFbnv`H=p67A8K*wZRfcD*lnb0QdEXGmKYP-%8?eV^joa2_sAM+$V`)IkF+AFJ{|a^
zNqpfFt$QD2UIj(5SaJ<#i=4R-q*7qCW;I`s;;Hfl#o;cU)N%IIIb70XG#?`ZbDqju
zUff&Yn<-j)is=ID9cYF35{pOZ7==eT&`aOkKilKAT)y!6O(q4;J@*3O+1S`bhycpO
zafEf2^XD$0BekTf39=~VBOm$*?|skvc=Rh@;h+A~f5un7_EqZIyy0u&>niLh4ckvX
z!N!S`ti3%VQW<fpi`9y#2ImA$X<`ouNgPE8t?^1?d{87A=g}&dqbecs5>ymC>x2qz
z*t!b7Yjsrx5rq_prEuT7V7MA(O%k23y?=mbY5M9d`8a10d!jU@6)SdS!uCjTNk&AC
z2Fg{eSBCLq%;@$m7k=PAv}^F_=U%3&FR*`g!L{dhIeYpn&%E?JT4_$5I?1!&y2-!#
zg+HL*K1ZtFNh3+nCT4UyV&Dy^oDY2HKHmBEck;VizmMLzg3X^|{mv7hC9}nX>10M#
z*5EyjG@;*W&|PWJh#NQzb+tew3AU^#geER3m<(w(Q#=||);O~Ozk)FW;|qlG*s>x^
z1WFjJlPHxC=?G~pwQDiU9Yvb(@Vg)6`i*U#f9`r{zX2!-N~0>npRZ=@YZuhg*w}dl
z!A@r~o`s)yhr?0Xnt%8I{nxCvSGeoaJ$PZMts{{MNu2PLKmHT^(l7lI2ZJGR9$5eF
zqqw!h3%~b!q@6D3Kioq#Tj(?+*BQ<A)1+GBr+YM%!0Q-cZ9usjF}8G6;~De41N>}3
zZ+)FeX^OH&35iZ)WD?<P%W&_2)%7(_pE-@OhP|zA_FlOeHcl0CW3y(1JZ`YPKcv%2
z$>I=W7wMoL^Uf2+!87r5-@MKH9y-mL^(L{HvEjElm>e+LK4k6Q6*gBJY~9*N#x>Z4
z!QKvgTZgop+x*a9cpqtZh-kF}qAm*RVvdSB?ClJB@>|cdlAh&}2Os9-#wOo<<_YY=
zGdpy&8a@~!y9H9}aA?*R?;KJ{nvIy=T1FbTG4+Do7oR8DI7#>X9h6mpGL~i|J@&!$
zUf_LAB!UGDOV6yNx60<lvm9L6!8(D~5m}NVQ5?()ZcX9E?EzWjuyMlb*%Q2S^=7aI
z0!r_h7)SmV%$c`h9JP>+E-{KASewr*VwtGmyr(QeYx)RjkWO)MaKO0>7XsRaL%5J`
z^1k=Jk2FcB-{Qk~@YHi$e&Gs@PLu53OWgTGpA2VESq3v@YlCW7X_7{YjwDrC;;g5b
z7i13F-45f?n7k@!W(~B2JYO&w4x#pBNs4Y}pcO3r9t{q!1WB4=ih{hZmPYfTAQV!e
zR6;(k7|(0cH26KvU6Y+#mV?=VcRrNSZ+QCI4*R#Rkv#GculOm&tU^196#^HHIB`mG
z@7qo=t8VeDzw#yC@y_@1?)Sfg%q9%Cc1T1*Zfr0Z)(b2PM7=^BJ9f8guHJfuPk!`+
z9K3Rs;-JqW-=f=AODmW_-DtL&tgWx1;uVT~#9(e|Uc8rd?F4CmjpW2BY^%%6K{Opx
zj}Fm3DC+aN!mO|3SJ&xYyqo0gy;RmPyLFY(^=q8jY_Qth;PRp4neCcpENN#E!+Fkd
zHb9<O1(gNPuoO6{7#c(29Eq1?Qc^od@s`e{w_=2I!&<>`Fz52+m#Hd4zuzZoq<B1<
zi1X*qBfY@ZmW7*>B!U0kNE>|Yqd&;GGiUj){_fx9t6%?G5S+e&!{~JtcFF>WV{Ux?
z>xe8Ry?7@&NvL!}<utJu2<Pw^l$7X1Ba#TAB_fv8#)L{-DI%#bGN><w^9bXxdBq|x
z$QL<vRbi^Y{}N)Uw3RZz#!4e}$a@@27C2|<qzR@lOlFQck%-Db<f*;p*1?oV%9==c
zVxN)5Gcxf4J2yPzD=qGS`$cxQ4|)C>&u~Zcg-<_4uP>?8h!4E)6Fl?87Ju;DFLAp6
z5$Zff*#=frh(xfcN`!Pg`0(94{JyvG%GX{-4z6PdUt@F4((0~b>VnxU#I|_vh+|2s
zm9nz3Myu6ENl9r8vpi=bYcR_Th*NBtgQ;n>V!Zb(@&&EB4i4zfVZ1?@8e=MwSfO<g
zRcNV+bl^l)9!ABS2jBTHrnWr!>}5ua85%)DNxYOyy=RW+&ja4@CygT?I=Q3))W$L%
zPtjT})uEEzdWV1YZ~i6!yZ_BUq_@(eDr({^MheLT4?MsF4?Mshe*W`sSwOS!3O5*Y
z^~+ym_3nG<T(}qvx}y%IG&GYD5r!y@u}Tq}3Uq|EhW#76sA<XcaGx@Xuw_O6<Vm6^
z!YUc;f3%`r<dj9p%E=RW8E%mqXHU~#Ukf%6)-oGU7!60PcG|dJmxIB8*=$NX)^xim
zUJC4JLBkm?_U61aa(w#BSNOogC%N}*i*~!<QhkHxuC0*X`EDAk37Z$zc=7s+NF@0}
zv3G08aPoCN`H3Ilo_p>JTqI{%JGDWlUEsar7ym!M!S!b&KJ=0M*nR3I*A5;>t(~M8
zrNj*|LqVgS&}d~$W^?kq2)>N1ly<vCueC~lBO_Ll%0T|YlU$xox%WqY0{ZI|WrbD}
z${bf!5H)eiQ)>Y}DC2M-qlB#00^!M0O}p7eBe*&AeCy_d=eGCR%S%M6n3L03U1RXX
z^TE7UD1>#?e@4;nKTgX2BZfpixIzdSx-8y`Wu$XJqw2~s986hRnUUz2E7z~^;5#0|
z8b_99#4@PON25_F0N?8OV?h=cX(sG{{c*C@b<Us8s2pT22&#k-A%P{*I2kx~^XZhq
za6->IvQ~pQNiao;vlf)3n9bR}ahu6#M66<3tv1b0o21zQB@wYgMLO_VqnN75F}A{1
zRj`SvJ^3VOG^nVBC$39GQz2U++57sSqI!0pkG<y<?WX09>Ki=&oZyE(`60gc=v8jq
z$Z>MOrF;71lPO!*bMCqSP9AzFVLTf0%oERY;^Z1}6yb~^Nn@tPl)PL})Q2?1X0UTo
zHGBIN&6R|;Q)}G5oDgZKsvKv{a!z<69ivo)GZlM>`*c6_DNg?2k1`yMF)9ck#m+W<
z?>5zNpLjaNl_hnNGoQ`5>o5N`+IK$;x=Aesi-R4W{-6FOW;)^geP@7#Nm;NnswnFk
zq+n8)OqJ*E^QTa0Oq|8ki=5h6X3kLqrE@Hn`PP4y@VcX<B!KTGa_~5;6^w>cwzsx|
z0embFLeg(^ICthe&IOU8a}Hw+S)&n<V&K%NQ@rclkMOl`eErz0|Mia7RoKNs(X1__
zmtMpqii?>hK7BTjR$Yv>GH?vM#oHR{8li)kU!&DzJRIW60;3g?j)I$XS&}bus?nUu
zV8moRCZ8`bbsfC6TrhMLN}=M2Bumk8OxEk6(u_21vcI=SjKGyOI#zh)@kMZf&xK`H
z6wK#CR<jMPQAjajwYgw2&A9dTF1@u49(nIYX4Z1`r2{4lm`*DC>rH<1SALJredaml
zlZ-k~P#C0GV10ry5pm{u*L%+Lu}{34XTEuvmp=Cd&BN=M=~JA)w2qXH!FbGKk>hMl
z94k7Vl-1Qfoo+uIJWEfZSy4cgQkON6(A2Xj{VYYgKyuHE;O8fo@P8<Z9a1>7h9nNY
zc#)P!G{y+FCqtsuHoet8&piJuTU&cL=V?X}sZbQgP&mut&jCB=J1K{b>^zQs7Su`(
zEhWyva5zP&2&E;eBf0$2^ZfF!{Cj@tr+ykG)H2*z&}z5&p&$AozVM|ly=AIk?FEUG
zEN<Ln@ch$s*47bL66=gKE^(thn#v)Xtq=|1f~aV)bI9)1E!NT&qrs3Qj;PAur`6lo
zBvKloRfyCXPcd~(k|tPd@XjE#CTlmrd!+YtIxQMm%HG}qGI5+eae~QwN<N*?N;GLB
zVNlc*^Mci^f)#k;Am)pYUtzU<iBsJUt>ik%q2}6C8$7Vq;ndkRF5G>{D=*%r$`!&V
ztgdEgEm`fIWjZ^+R$HWvHaBkE;8*|sm$>nw=3Vc+#QeEyjKBI>Y<>;7elJ~h2enA(
zRRh+uD@1XV(eO4>dDd4~IkCP;yVXJJlt>3Fq@pq?S2B6=2`&r4rN8>qcp*cW`(%jU
zy+L%o7h(vd$43H`Lev&t8}i8niKp2}$&!?5?s)q4jBnlEWqUCVh>aJVyZ26Hq?y$<
zl>jH9=@p6f<bMW>yQBCp=l<lN@uMV%SMy<5MvY1ElB%pI%Yxd~JpcT2{Me^{9OEoT
z8R8_y16N<U`j(@05YWX~N}K{=3^%{_1nWQeHpcmk>sPk9boXgmkt2>ZQi&x}>sc&g
zv&MrlN=dR#h~o>KIKyN(<Y0SyiNKZ&_6KOCXfzwNyIne~J+f{a6={+v!6gmmvpGd&
zu+~v63PuMDjPPjXkzO-*5ha#7vY5<s`PP(+o0j{|pQn}G<l<t&^~VJ7e$RV3m|dnB
z#Y_(;L^5J>==r^0yUNEubtm_|>m*8bIXswhYcOOq+2_XXt32_<lMHqvR<aYgss^7x
z?9j2|`mJ5Y2L~jKX*VN=!yT%s3YTA{!Z}HdEf-7AN|-Jx%Dg5^9RB)cxbhs+?OPli
z4)J(eal~}CU{)DKvxieHjPTgWkf(m}XL#vL-{g+X7AyTe)>%$=74N)wimSULUOF65
zwKLxS@co=PeS+C&6nqB>3hQN85(oUJ3#fa2<0y6OXf1uW0?!)HWV&E7oscwQQk`+>
z(p{W9aViK*msTnA$Y#hIs=DTt8`pUDx#xg4&Uk-)h@A+o^P;TSec>gt-U^9|QQlG3
zj>}(poOY+p>C;WJw27ob2W~#aBBz=!kaf*?mZOv;pD&nAXN(6!W}_L^QckXfL`i`@
zCR8r70rQ-)EV169M`NO_K{Jieo2yL6Bhn<I(@t=aV7_0H%Zj;hWRXFab8vy1qa#BH
z!|LKT#jMY>pZOB^e)R2p@I!ae=*C=rZouvqJofo(Y~7erSF7M+A^>lY!jYs?PMq!Y
zz}xQU?hEI5?hDWG@>iarv)HE|Jj*-o-Qd(_gVAKnaCCq%C2=HaG*f!LKHXlAIBfz2
z&KQa!4+f_qs3jw1FtaI{5{SxB79~|#2fmT9C@HYcqP)l1nn-(+IEYu0I3kWB=9S_3
zt($Dz_aKL(0XJT{hOZr&Rw&`AT}|m775pbP7k#5cDfMkd*HNq;ggK(5#QSCbYDl1?
zEDVFgaZp5SO`2x>!RLRU2j2Do?|J0C)K!I4A!h5tAO0}CUXR1W!?)!56D4UV0Y%N>
zQ%|#g;S#pDMjT7zcpLVw)9SQvQ5rlDrJ~47ZoYhzB+``ig0;0%jK_OaMZsWam#V7i
zb^F9=N>v*UwssIg1VqA?LF{1MYYGmJ$I)Bwp(4%p)-Ln?ec~)8>9tu*$D~%Ho0{ni
zqS_NTJW5-Nyrx*pp}UHb6MEv1S8m>7v~!x)${MXykTz@PP9Rl-NGSgBcfZW#XZMMd
zl-2bPyL)?l>+9E$^-1o(|0J!!kjb}xm9wqe$kv#_;x;L#kTRj4&A8)i%;8`M=W|YM
zta9$;Inua+wSvM}OjS_SW9HT%m86*}MlU^0Jse<LeMD~+F+E`R%`ekGb%7-5;%$vl
z!T(NLM>!ra+1f%10;%3P3LCRO%NdqA8|P26wy{POMfBEJ+1}rwE=sKTRKk%+N#99w
zEcu_Yb#&x6r<D3*Rz|N~KSu?fQWEc$5spiPU8O=qoyT+K%FA56dWC!Lxo3I61-@{r
z)d~j2*1qZSNUt5uND)a-z4VD&>>RLt^){Oq+Env>9{v2cc=x;Sq$k53C$tKlWe&8|
z%*QkKw|6;ldYyi^4N*cdn{%+UOS{?R^tlTl6_e4Jy{&D=gCUFQjC?Ys)$h{mwuzff
ztctj{wU5Nnh+~S$9E(6ko|Krfir7AYTSdvjh9rvI!uIWw^Nk(S)m7Fz75U*8nA|wW
z{SR)iwY9^&cbwqd`I86)ue^MV$NpfOuYdUlqsf5Xodxrm!a7A>&T#GoE7{#h8zW`#
zie24k@&5PTPd+Z0Zf(<LA0ZMZ;|XP1AOu8FL>wnLC#ignmj#38U*Oyi{4n`aBvlU%
z*n8<kUU>crEA2jot$ATG#@97#D}Bs#0t7p=!)G7o^6!6%SPD+8G(#AG^qgr(PBmL>
zcADIrjJf}P@4{)=zP-iZ;E>6DffT_&=Fd8+5L^@j>IVF&m;bP)!kVe7iph9NtJOqi
ziWBQ6&|2f1Wjr2|B`M9U6EdEi;ag8W!LR+=uk-ZN&tSg~mxkA8P%I@z1=+TnmE8E|
zv&2cv>267+G^717H?M8sImgDj#P}R3B!hzkriUXgoH~ti1*NN~tD4DZL_S|2rKH{N
z&^mQ0!24P+zwi)nZI%#nT~^GdQ)bgy;Gr!Jk!eKKNa%H&%%?M~b?CTBVVZ0$JX^Dx
zyIUO^#!{DaRF>jwxcRMhC)9`EAR7&N_BV#CzT>^LBf;A)T|gRzav6yR<syu!Nt@7Y
zG!Rnrws+paUH6>EE-Ie<;<p&Q@Cs+_K8Jh1$3t(6ICo+-_>>g|c`+r9g9@qB?$GV@
z$<j8UaMn|nIb~H+S&Nej;VoE0oiB)S%qJ6W-QHm;B&VGVL|;X?Asb6whoVs{PaFk%
znMlX%3`Y1)i#pT1{KAXO4kyGYT0$X(B^QpF@Z=7@pSYv{RO9H7I=VUi(Hm3nmtvXY
z7r2;0D5|n%I+>C*VpJ5I;D7m-eu=y8xQF)g7Ou4B&O7hqo_p@*D_?o+E$PK-wS@Gk
z*}eQc%}W<J@vip})rRVYZ*WpOqGlgg7nF;fyn?;G3A@{4?mE9pzm?L6nh5DR*x$xE
z&tx=WN5`}pZJe<j>>tqWb%~M)Z!EQ~L&l{s--gKwA!v0PoR%lqyS>AFJVj)hI8CT)
zyYw}KS_!Vs6|OYA`$Cfwo#yhuuSw+)J-Ed)kFGJWPg47Wi>EfJ?K(HE&iUFG_SxU@
zND{g|cyhi^U3p@?&O6?7hHN^Zdiu9H-@U?_Q%zdwSt>VYUew_`CSp83U^w2U-A*{U
zc?uoJ<W+&U71j%^Gfe!5lA45Ys0eSU^Eo!_5U*}vZoG)tzQy3N&(i<kr;sA>cx7$y
z`GT8Y`$OityC~(+QX#w#UYWM0e`1ZhAAUPldKN{_)$6Zt<=GeUg(1~33X8?kmV%ZQ
z6yGoM{T3|4qk`_Jh?DXtW=y<_%sF=~81Ojcl4}<p$y1x43Y|<x{NgYEOaAuX`a7II
ze}Ss1NYa##e&i#(?f(1u))P;>xexzGhL!>^mI#jrtYh!`A(Img?tSng&wk^1zVY=J
zdB-DnvLY2|O<5Vryx?GGpK`XK+iap2meP;O5sVK9EG82!UbskawT}>zR=Y*3+2r=M
z8;k}+hWiIhM`Lsvqq7DoP3W~+6h)4yYtk&H(~qf*U^b8(EZ}x-*)0u`l(Z2TSM#ld
z39DVn-LXcmbUEGJ=J1)%bK(Pkf$a22R<a&X{NC4*MM<N(!8`AJkjvL!<l3_XilT+b
zExkC~!6aFNQZ+(3PMz0$^rLSlttH?1+~erkg4j(FQU#%+_e7B*%QDg=!+1|=3%qlf
zqCm+Y8Ziib)a2@oeeOJWf%6+DvEDK+=IkF1Xn0Be@{4fad%_f1p4CDGu4X=;pml-N
znk3G+*lludy^jQ*`qoq2y17MO)Y!_fq7>7m()iVR|7Y(kI`T~w!XG2xum1l(K90~h
zYc131jL~R95+^)=`FS3A@Ih|gyv|p@_9%CpzrauY*k2`UwDH1|7X`Jczb$g|jf~e>
z*omNCS>Ax=`(v&>{xmBOpF*r>+<o^QJpT13x$(*#uiTz87%Yf9^qP|UFP=ixp6O&1
zL;@bNIN{W(Q#89Bq9{cw6?}@8hN(qSfIKqKl%#1)r_-XYE6VAd(P+f1SWxA2k}Rdy
zX)zj48IGsyO;@-wfvI)0*AlP-wJeyt3qE&QYB;mDU}xuP>cTVlhGbvD9S^*nyY9Y|
zcRX?zdG0_aJp1f(+;`u3#)By@J-^G#Pu*aCW6olCo3@<L&R-_qeUyjpmTYc>T-c(h
z8I8s?vam^ZT1|T09<4?jB{Ze=81Ja;oZ2oh^)jDOgg7UPipglgmFJ%4V0XmET^Es_
zz*StTXQlBJxj~6wHxp^lQH-;et@)hp*>hYw9C7vfO`P+jI%ZbaB$1-@jsi~&K`F|c
zV0ZEstw||^=ay0uB^7QN8ZC|@XT(vvDg3@t66ZWx3Y_&!rv*_uC2OW+30%4UBES3l
zzs*nn#D9jV43Um$W=-Dr-k?l+D~3d%oMh%BrZdBpFFj7D+amKLdS*<c(L^A~XCr2_
zk{4dyXY1C8cfS1`E4>DhvP6ku<HTu-Vu7g&lmk~5>`!OZ#-Vk@=Bbk?Es3HGmBjeE
zrYwqpPD&N>GV$b9Nz#Zod*@kZ<0*TC15lFM)Fg3C631L~5^W_v`u0AnYki_-1A)PN
zK_@w+Gu~qRMvKaO*3O;h-48y(_F#wK{q$o@M}o~0JwE*Ld#Ur18#ivz?QC%E$}Zy8
z2z%`}xvP7XQzupsQH>CeNF+FO$j<&v%Ce+31&vg&-rpcf5)LLaytOoA6+|9hhL(Wy
zcv~?qEXkRBiB>nUh;Z^nF^#(9#+N^jK6{Di)ZJ8Nj+@LFJ^LlD{@$-+ike0eGH|^Y
zIA2lflC$^U$KHI%<;yR!wY`g(ltdK>ol*-8bwQ$`#GnPs*skw+LL7xcgdtx2-)}`r
z_sF~PD7`^SffC`ugb42Z@G76kd54gSqV();j}a((y$;X4_%uKN^FPZ^{q>*b(xtnQ
zQgP<=8UD#X{>S{Szw>uEICyIi@8z2a&re7Zz(O%uFd0rc(@%Kt;Y&RE_zQgftCzX+
zo+<54kJ@?Wvyy|o5tq)c(bO5ma7;cfF|!5Z;gB?nY4timVBkW^M!VbL+{Fv@`+dA~
z=p@1#Lphr>n@$kECX)h{MZrYBE|`{@YqN~!=QX=kfyL9ZnwE;Nz^%OIOE(8}Gs)RF
zCW@P!*WcoqKWwq1I}D_tRbHb!c$r1g<As|Ka`waTr~SjLeD>4NQkDf;YGmZ_NTiA|
zrl7kP^TU7PLkw^1bNTVF(HQP99X!SAW}9XvNwbKe@I;ZO+vyNR4Xi6EZ9!SU=?^|g
zZfpon5uSN&sB25V-{Hp2Hj8{gua|M^^g0cZQCz!*8||X{Cup6&#OB2l+`fFB;k+b@
z3Sw0e#To}R(}YedW`8na`^qhfd5Kh-rc@N(F!o2bA>#PoVOhL!O3a&m3@sU`NXb&=
zDvvvfuPyFgr3QsRWsPGt%MrVWq-n(C&wP!iE`O87V!?bqBUTZOMvK4vWB)UfM#lT!
z`#v@{HW&<s{L(M~@*76&USBKLgQ`P9LFp{UIf|VTJJ%yl^iOl{{5p|;JGb^GTsfGs
zzhAP^ZE@G7GeoMun3CQ~pGLMx7RBf!h9siIE%Q`soOgKP5Rt&w9#@seHYu;(FcoR9
z$!4cXCtolcj+oA-SOm>Zn<1JPZ%w&bS=y1osRgC%P%LsfQE20M85FjWmTasW^!ORX
z=oOOq0+VMg-@Ls=rJ>dDK%~jygyAbgE<baX@y>##mo&>;bmlKH+50Np<dFN`)}h;N
z;k===C6oD(G)=HBSiY?G`t;ge5}ja;U&`X4taEBp6Vpap!sg)wvs(we{Pkzq-#Osz
zr%todTSZC9v@Aer>bgRdo?>ATrVLX57?jq`7jwqVj9h63SFSUe7Bo~u?kuj^3z0hB
zV(}Q^fxz{9-9x<kpQHKxN6)KQf8q7Nyj+^723w98DP=f#Ot?82Ymh=xmL*<EtQSjU
z4?+sML@I@KlJPhvNhiovP)hS#pZyI!@WBsr?$mjF?a@l}{`bCzD2^zLH=i#g5Fx_L
zc?v5KH+H%4xzBUo+t%>HGAJyqRzy*hTz+|<7ccMf;9YA_9B}RB0<9!x&Yq@`bvS#+
z9e7g)gh@+`b5v!8kP)p`6YFYZ9HC_r9Q5NDU6dTD;t&EA#ne@W!;o~EoJb>fcXyEj
z%F?q_Wt2|vv3vWp*ZLqe!pddzoTHsM&NQx24i89@v(!718!v3IdH)J4D;?I>8r<{1
z2^LrO7(I85L`bGl$(l7ZXOFX<J;mvh8$>$7dO=y$OlMPulLN|Xg7cm%5hOaJ(QL7M
zxZuXELst74t<4@<XpHhGD^Lr?aI_#gb&0!w{I9bxkjk2Bu!r_w%95SSFOq%ZD>OcN
znzC3>A0D#*_@mT=0oGgEsV3GM;{<z$d$c!N)UM>(H=pPFE4QhoCQ35ORAVg&W2sQ&
zrlyoIJ&Ke1UMKo@T5fMaSLz7Ge(jJ_5bKciQ`a@i9pE@t@Tds%K5!al(~9fG4zsm6
z8=EUU{o=RyZ~yPx{Me^{f{%Ud6ZHEleEbJL#((n<{~`aU|I7cCJkQ?}%%w7rMlx7p
zu-s>C;B*T-NgDBvciqdiE4R4v+%0+=GvY>zC!V^-#gnV_x(yPUV4WnNIjY$l<vb_P
zoJM8w^27JX2vnohTt<zVklXAXtxiVI_A%B{Rv{bI7=ssrPTJt|<d9ucVx$BQQ*S9M
zk3r&O#&%w_H_17<S`o)B+S!85(WB`6G)0jUncJLSQCK(Q*7ks{uPM&`;D`CpC+=i_
z`+)763-%93s90gWM<$N@-+qDHFYYmU>PeEVM@g$!SY2Ci>SUKlq%<1|i(Jy}wps13
z<AtWI2joRTX)`w8{xHR2ffEi<RlNMoC+YW^%w}Vrzp>4DwxFMQPMzv;_QY9Q*(Pps
zh;6Oo(k2^sU1I;GYq(gF>5MGah%`pW8H0I2(TKRdeT!;QA$5#Ro6O6K!q1s6FX)OT
z=k7KCe%(WW5(4Q34j+`?Z^FD20<_jN8Vw+<m8xEb2EK-G7u-3HioI~LbQZ?LIoIuN
zdaE6z6qH4cv4%z#^Y8zE|Bes5@57usdyzCwxcBaR_^F@z8+_(7pW)ze@P?lEbrp6(
zh-2%3BZ9gzmhI~|x%*-lz1ii=nN6DgJ}SG7n+#cPSvFQ$w9*#BYD!z-yk%i6x(cSm
z4v#q!1T1+SI#R>}jFR};9DjIr8SxUl?F4a}uzu<UeY=jSYvNSX+RXU;OHblOf{+n-
zi}04RES7u`L1hdg_@~f`A?sN>@q(?<^X%@wfE%Bton4^54V42Y6h>%HO34Z4m|}<d
z*7LODUG6{a*<5Q8#R;B3G#*XIG+PadLa>-c^g2B{?G>~NVf1xXVeJf4Im*IdtqSm~
zFw~{T3BfZ@TxM`{mj@rb$XzGSV>)f3IHf4Vji;_EWDTyEf~`^Bk);|HOZM^tcj6R#
z<1x3lchN@RwZcY<C<&@A>m22B;)^$61gV4wg_jH_)It6q=7#lej%i2~DH@H8G)<Rz
zuMo%a(qvOp6vZ*|+j&Q2>f^)dxLqp(#9ZWt>10NdL^K*H2L}iI_V4^A|K;EKTPUH3
zRLq@s-pNY8&-V7-_x<eOY1H0RFO|FAD{j1SlXg~f?;R)j+N0m(;yovs7oN+Pcew9N
zM(QUB+o0X-Q4|xd-?+*h7cY@yE$Z4+85@kclbA*$1A-@)6#^A$y!8kjA%!H8!9-U|
zh0+1Fj;*Db%@~gcD5Y54++eYo^YpEXuk00k^3E3Rc8i!a6f@G}g$<$|CD>f4=(LAy
z5BAtTe3{`B2A{o?AO6&bkUHk-V^3k9{2F@qaU3Oa)BrzcrG3byGpi&-RJO!<!(uT*
zM=G?5s(h)=lq7LVWj$A}4cOk90Ftw(6k0|&=ZS2}V7kDroaK@K;_px<ZG<!U(IJby
zUBtq0^Xe{>MNNErn`$_qPy!iglC_gW*|WsD43|c!sGVbAYvL0pxPE(wTYLLdUgG17
zcy*P7?Hz1Y;+7b<1B?YDzzcC)9K4YszWcAduAyW&Y@8Rzd+1S-r<NRMArw+cOjW-6
zvn40*Rid~<2+hI%6e$HKPAYbWxA`}}^oxA$n_uO>{HedqrMvIuZ~v|Tikr7?@^e4`
z^OV(_uad32p!R~J?y>ivwWHrjiSSF)YRz5uUf{%uHJ*R=3NO6;62mEo(-FOH8zChk
z>d;>8WAXxtC(crg0AVe{d7KN1Mdv)mRaDh7OIZZFAnP4Yc!)zPko0jN!sEEK5%J8<
zf><eH1vBF*Oz?M&1Z>7JO&u=aNKb<0j+F)J;T23K=r<Lu)Przj$&@QMzQW~S%}MS!
z$9;F5<*vIlPd{~oTf4hR?YMC9B(12);a4ALb^M#$d1_9--9)GcURoA;NCxS&5>Bpf
zkVP4bs-i9{Y^~5X!i=VfB*A#g-g8&ke*RhB_4ZBnZXK}5TC_G!Gr76N?COa9so-rt
z`XFoVRm!RiafS04orYqw+a=Nwb78QhXEGe}Qf--bR(avd4Xo10Zks%>v8Et*F07lQ
zV%i^p0^c2E_+#z`0<t*f{F!rHx^x$_#hhoJdzQgq@O?$zNdajZ(QdbCHd`pAFwH>k
zS(GJpZI<0501u&~jdOnK))1a`GOd_Q50;%zkV3M^O48&OPdxcePM<ta9Hj_hdDlB1
z;l1yDFTeHMzw?Hk`Slfc;juZQWND#L0v59gqu~U%)&>kv(v~Hsd!A;~uygYjPM_MO
z-RaQDx}l)Ri@<X+CbVFc#0!CpqR=i-k~B>q%kXu`cC^OeT-aoaqM%&*3588(X@w%d
z6=Tk<HQBpekjezC!<Y($#+Z^$cZJb(N+37}(?F>#-L|9AkeoJx;rK-khEFq^=U~BD
zNo_!gnl#N>?Wf#zDPyI*PO4Ib669uv^Oo6sO1s@6NmHt_Bux|gt1CoFL~RX4Q8AlO
zm{kMHg=9WA%v_Dsj#3un#Tca`iusUBcb;VPM32GrfZ=RGQdx|xu~AJ?6ezDT^AWBJ
zkvEC*O!67~vdsv=?%pB8>6{2j<vq<_kI7_?vlZ6)pg<DA;CDGKzvm%^<ed*a%*Bfr
zxpL(S!{LykD45OXEXpF7m3|+^os@zoiizWtG;PpmH1OV&#0fSH9&fYR4DWpi=`M0y
zK1VK52&F=?X{{lSVuaAlW(AFANgPLbS@DH0`~g4w@lSI8>_wE<tgWqa>h!6%tgv%{
zbQ<pk6Yp_8<>}|PICrYg-4{=D{pKOJb|>7msoCtttgo-rY&C+{qgZ8du)~e(*H}Ms
zmbHydbR3L|ZC!`lSMNf!SR4Zqq(^CubP69u05n0%zRR@|NWir<RAs?rJZ3tY;FeYq
z%`BzUg;Id<HC0{F?MKY>V##Av7<70ysv?$#bDM(A{+O4x{(#*t<Q!bt##a^k`r{;%
zN7?L_B(Wxn9PM^Qx7|btL1{~@4@i9@YoL;tvMk8rCTW%hF;%QN9F}ZtFIeOW`v(j1
z($Z)G&SPB7qVlYsyNm9|Ne0t7)1suD)m*uH$jz6pvVZM5@40WCS6+OA?$vjb-hDq+
zma_V;AHvS(EM{L|GBy;Yp;%0qHWUYC!R+=vDz7L-NFo>=4k+s!x1=x}6?VtnRPSCt
zIQ<&bJN#L3jIw<%>bpPjWrd)z^v|m65VQB%*HS9*t(@~L%ACkY0eM=Ybx{=Iy{D>z
z8sAz2DDu1vsJH?{$<r@9!T;^Q|EK(;|HJ>8JI-C=fBg^sSN#4T{=w3u_01hp2J?C+
zf|Zr=ki?F)PJ@VW&^zaG0@}S6?|S#!Sbt@gD|=IB;{&c;E2)j;#Hn+vo!mq>8x)fn
zMO7oU#CRW~1*Hfctk&X<MEf|H=c`~V6f9nbj1CT#+1-w+s;G(zZ3Sl%K|=|;k)!sK
zvQ`))h?M7K*Rs-zXm?uFHnjHAsN{~#Bv1>e(Fz=n6Nx~}o7_6M!SwMCdrzOC>fVcP
zZ?e*A@_`?CgxyzmIe7HztWW+U51gA3Cn-W!SnH{4OHt-Dnhm6IG@30IRZdZju!TWb
zNfcGQ^xMC}-GBXWQq5})Z{FhcMgt)X)4b-y+aKUPKlB$+wPP`#Q?xR~$<xe<nM@85
zg9AoaUm(o{d0DeF-DBdRh`PwNH71qdg>PP_8W;E|rP*v@76%m0g!9J(A_#E;VI{1B
zKW)}aA?dZ-lyw#MeII~^yD#3w|Kq>=?|9Gq-;b#){@FkKr~E(v`~QKWEME6I3Ku;|
z90#PX*=*D6^~utds;+4^T1+Mrrn5;f6Lt>c+|oKI46XI?-+N?lVr`hmQYp&PP*xQ$
zzI>TaR(VM3iXz%+o4@*3ev;38_On#g>+c@DuENd(#=CDjtSc|ctziFfN_|lf0ZKZ0
z%?z>G3EYjM<fRv{k~S3UC(p8Z@-(eppSrFQ9wMc&GQ75uGJF6&z^Fnf&;p4@#0gSJ
zoU<&7g28aOT<FW^a}Hk{7FEq1{fNVnz*SI}P*j$N5=^EeTAeoCZYvZxXlk!e%Alo3
zDMwR7mIyZ2VsaxXjmLY1H=aldv~H243aP@Cp)wU%L2WC_vZUW%LB}z*u_zVO>h_UQ
z!f;k_IG8XPj2TY~CW{DD2uv+lcuSfYX7dD7&yi?aDQu)2wsvkZ+8Q&8);MpMViIqu
z>k6+FwwzPdhT3|(w+!-}`T98q;|0T=L*~|@Nw8Ye?yoYL%qYF3euFHABV=QBWtIQq
zKl;af^rIi;aB#?zPdv#Nzw{-(^5~a&>FO0m<54hy{bQl%-~IKy_asR|v(ci}Y|-wt
zX|>u2AWJixbwp82nq<^<O;MKN!cdp5s;HN~T5>t3wAOeBi+O?8l17%WySv4sU-=Se
zPMyUXL%ZGP{JC>{^BYgRflubUyf4Bgu9jRZ=N+|&gL%R?p1#RXeE43nEMg@pIJtg;
zZofy`PLErskg&RWmNXkP9nUyKv3hC~v<e=NLZFY@58i|Ks5nAMiFYE10(?MeYGats
z=M>{HI7^l#RAtFzGUIS>h7>jTcViAmhJ(Wcl2p@QTc;=t#(1O!wIo_w=aDEHG2F2^
zWd6!;laKGlRs&YETikJ9mn7<k%v)&@uEtwH33Mb#;}qjOvsp=1S16T`wVHG~9fZ^j
z#uYDJIb=~}cqFrV!DLp@ZA8mUn?V}I<b~_J_?y4Q*^mDKyW<gF7}oB+lc?7M5fr_{
z!8Y{^FL3exhgsAG>DoGLANmRAySK={ahXw75EV6fr_FSKiks$WJdyKA0eMx1y=nP<
zDneVv0`6!({}axSqj~*Dg+I&i2ZRsTD=9^w54=I!oj~sqWLZk9*+MG8d_JeD<T0|4
zL=kD4k)|11N0{-eYg7n99LGR7j9yjLx&TPXB^r(oUT)TL_4*Ya`|6iDbLvbm1b>UH
zPqgwx2rv?y2>ImBg=k3=L=}|-haiap;(y_eQ@D7*)yumK2LsxjK1r4`n@o^VMA~U#
z^OB-mgwIZej9{S@+PQ$NIfoBQh2yBPuGW~UW-*;2l|~DJDGj2k*hphc>B+Pp@ro3M
zh@cT$Hrt*w(<F_IVj7Y<l(Yz61dA^zgQutrJjm4^bUTpSF^k2Jt=(r(hxd{-?&kJy
ze}-s&l{<KfyUv1+n&j1-#bObVFDYoZ+aMLj3T_`Pm`)}L76@fAg<vo>99(^zY4ao(
zKJk7Ut0}tMqAce)XK8l3oP68;Y>h|kPsaG9$y(N7vY4WjBES7Iv(c2uIi}^9ST>1-
zrCjT=uN2Qb`vTXu_t3=T)?<VujuY}?j&vRkn5BEh5gq7HE5U>iG@DKS;eY+#u-fnQ
zb3ga9ymaMdBCYw<r#{68KKLP`EMa&s<+*2{3TZuW;BpklG3{21G)Y4AXS*Flo~;&j
zT~lUF(lkRU#e6;^&vVi^TGlOMsXh*gTFyEa`66`d>grek7!Y#LOE15`Y(B<&kJd3!
z8uRXVznfF1PIB$qjo1C`uP=Fmpi;GB3Fmu3E(E!V*gBjr&uh9d#8HgytdYcmvRsfR
zD;VRL<$D}#@1SJFiSuVsX#(D3oe$Zk#!%Vi{r8CAt0Y3JZ!-w2%do9VPp~jCmORfX
zs|sf{&X%;b<n1dhZfqOsO0uXxiW2J$-F}DO$~vu9i_vf%Y*oAnZ9M0ZLZb-eY!OGq
zaX65LTN>aJBcxpNDlD~~p|m1Pn{+#U%G$7)SIid$^Sq+8f?M1B91P}6rxj%-uttMx
zB8(x44JZ3e&Yn6?FOJ!@7V8w%-15rJZHD;(mkTx?JQYkOD}z`b-bw|Ova(=MSYU2E
zW7TB7y+wUEX92{*;i8!B?LCSjr}iPddYS(AJ(IqVCS#n$y!Qj|=R+U;2wA(unRbga
z7tZrTfAL4Sb>jxV{L8=0KmX_dobgf?|9VE4+&UC>tv2m;n_jm^yWI{vhpd4!0l7_+
zlsqpuJUpb^>EVSV&-3GAvZ`vN3<Pm06~;Ig^O7Xh#F0R$m@j_ubNt8;|3##Zh?^M~
z?zrPSs&2pE;e{sxVgZQ+<HIGTa`3{n1vl;*bEe;-lUx9=P^k`t_9g8A0ynVTTcgno
zm!I*$AaLlUL`Iq@iIF-SK01zrzg@@~2oE6wouM`cTLoTHoW|%#62)OVaCVL?4fl2h
z*LD+j_Dig}f%As-&66M^>e?^+d{E>Ci^&Y>BU)+7#WNFbY(GY`8*%RRMo_t9z|}ZY
zp|gZago16pu*~NLlgR|{N_y=koo+^)W{hW^?cFigZ_decL8=R^Q`AN=94=UCM@U!T
zDn%A09Bf_Z^Z$N}kF9JFpYAZ;9upf-6;xBp_Tdmi&1zX<tw9S(fg$Oy(7*o?2A5x^
zEaq%w9S&<velW#ZL#icNZ7HlpX$4g%#x!DS9utbGBmb>GY1SOAouj!agrJe6tn_;1
ziv{zdpfWXLc@YVOyYFodN+8l2tu&s1#&x>g5UCbv(zHQY=HY^)G+El9ku{0q1SNy%
z`(z$0F^n-wgw7tLccHBmGQ6E%3Nb`b^&in=W|J|-S$^#|ew}NteEVVaX00bBWiZcO
z?!{gR_9uqhdlOFfGb&S2)CM6mon`}-XlheYE=FXDKt~b%wRODrOb274D57elw0j+t
z49N;AijXS2Hz7diV8SewLN3!NthIP4$-3Q;j<U3QNt2j-QL)~HTXR8K$0#A_Bsr_S
zl$CaiR2zETF5WAIt3yjshwm*!YXwA52!pdV3XK+qMl8u9#Y#&so<GT~coK=Cv!-b`
z8h~VBh9q%9yWOFxDpV9<oM&%u%+(u1h7*JHmPTetQ$b-g)5C<iGKd$RqWAuLnV&ex
z{*w)&`2lq~WBAf5l=B7G4|XY&2B*4fu&8OI8AY+j?$%2*;*`PRg4)dJt<<bGyOiaE
z!*apy!GJNr6~J1<bTGt8Nfsp-Q&BmG^*(&Bf65dQ%k`$D<l%=N=0E@If1T4OPx8r6
zev(iB&hK#L$`$_Nr#^+&8ocL=U;F}JdF-oiNPrN~?gpK9hcr!THk-8DZ91LKtDr;H
zB-Zhg4ip4@#u>!Yy&+9gk~l$l0V+h!Nolaog{*I9X*M;HR=oW3i+tfrpW`DR{z1|-
z0}X3yYdrMOgIvFU^L4MC*I(Fq0bYTC+QY&NN~;--OAdz<($$PeNqiEc;~sLMF_kAt
zQo5_>um)6|lF#O#LUB};CEkR{F|Bn-3Q#h9LCX--P|SPEB0v5h+wC^in!v4#G)0~>
zUpTxlU`p1anprF`MsqkeL|uzDC8L7_8jUtmXS6#_u!hNW4l)ofz4IUqo(ktAk&3ad
z0%!0(WOI5GVV%TcX|`H4+6~6joZW*%#*>oKxMVUdm@O=eT4C*xKp{prK`No$p3v<j
zoLFzLvJ%rw1(6<v$x%Wvx6BV0q*=<vyDxC={qLi+h8hn-k%<_g1*NO7rN>IJR<UTb
z8I>i);S^KU;n0zSG>$13bFej1$dJ=_gm(TZ4W!rp#N&xo#A>(CkNu?|qupr-Bz>u}
zOJu?w7w_Ob?|y_NiDBZ8{^0jK1dzrtomQJ}r%S)zr`2lFZnue|=r~owdyg@O!C*jD
zRSX6dX_Dfdr>bh=I6ijnjw1~MoU>Rfn9mnz?NF)b%GH;6>BSd#=pFAuyO{Im&K;wC
zZ|1OgQaoBf<RKOUBP6vqRF&k#S8lV}Jx7|g$QKKYv*Gaa9=mL#X|02a>Y|{w7Uvyl
zqe0Th0u^EDjxUo4zx3ym;R{;wgFFfmHc;n65IiEpbm@q+(WFt;%xy*LN>(z>{y1W-
zBN=Y*WAlRU>KZy};ALoUH(Jd=j~Ew3LelFrL$Rom<D$;H0^vMqx5>5roX4NI#(Ys>
zoI|2QAu9}%$%N)0$2i4!AuzS2oftM&Eds~kOfW9m+}s)i1<lkjuMDF@OEF7%;I4H#
zjf%NWk>fe_&K6tW`WB15+rcWrDYD+m@p~@I8m%*um32fS$rmN_E*u=@<Of6Qx<*Su
zEMZ=iBw0qDi(}*~l+3SD#eIhh(vd~ak<Ex&s{H%?KL7Rq_J7ZV4?M`HfA`Zo_xy9b
z^zzHxxOEeMECBlMLjtWvlU}DAi2lo>EYT5ZoDwH7t!4`lBuPvZ$HZ|=7N<y|@E)>d
zkvEx*D2jqS&yRmr*3~kPIb?F@i<;qZLaWsX#;V4mMeyvZO~t1_{ks@rzRhFu%^FXZ
z3*MhKLCD~Cf?!%`zVfZB+<l>qQsKfEX{f6<y0?P$4pq;{;t&beTU*7}hB%H%8yUtJ
z%Dg1bQiK#(V=+~YlJa;x2_cS=bAebExuKmFGC~6&Aw>9o`$@^7(PWxiq)uojl5R6$
zeYJyd1!<#=SDG|c%%=0>*VT~<6eRCi<jdGHYHFJ!$`~mfN;p<p8U3!ud5d*5xg8*t
zCd)Fshsk74Ra+Ltoc+U`+uH@RT!M%ZSc)Q87FnBNE6MqFIJ;@6zVT%`Kk#uzciqqE
z3%4k0L%P4u<e8^9ef}I<)}!#~EXLL~ZZINht<d>_579W>V|?`rd$)&l`V~fIm^>ya
z4a$2e7qU-9zM#9h3W;VhI)qw~FG&sI)xYz;E0zevYPZM#^6&k9&Ye302+rSmk-znK
z{w`&alO!n!n9rvCi~rZpa5xyfAu`a>+K!@#Mx#Ni)uP#KhKNTQzI$gJxyoraTPzj}
zthdLUR3QXaRfPgCPBFEiHipX7D5>zirrA++x=HZ#Ef@UE&;1XKM<agx$Nvf_iPn;L
zJoFG>_`(+%jV7;a9lfr?&H)w)PU7&C-cdS>a1pMkIT#M;x4LxFCbcW^))DIrB`jW8
zWZXhUf~u-n94<Fa6^gkyOT&hfCV|5sB|-_D6iaeHOSDW92@Q>mD3aJZL~7aE5GR@>
zt5{4eu5^gnu-b{(8ADzuWIv|c)YwHqKADjA6MU?YLbBREK~WSe=5y+*rmz*YDUnVi
zmy0%vm(L<KI*DmETZkxTJe_j&r7d=MCk)3qdG0ZFjIkO-f)k*HArduNT++@I>%EM%
zwI*r5LnBql+EUh@(hIzl_~DSUnBZg2y?37E!ueIwjUETZj5==Oy{08njI<QKLKX_A
z3d$m8+FfTl8BkAW!5q^IM4Hm;G&x*Mh?Hb*!V%AM5qYBJg8Ji&DrnF`3tlNf8for0
zeVz}!@BP8h^fiQRghvX`t((`;Z@`L208$8I9T6){yV<7O?$Yb^Xti3$s-(IO;)#QU
z1Ck`B-EL7=C4OqCiV7t)wK2zjSH@dB79m3`J6{-d1X5a@gh#*nW#0b4L)2x#iH!{+
zt&hFG-mHVkEtOIZi#cxOh2as5=L^QOF{jo}1fGTqpSm*^Z`~5@v80U#N-Xo0;|Q&F
zuz}Ix5JE}Np{P?!#tM?-+(G9&t~Nm6gSgsIR24;85-CNq-NIQ*?hGr<iupj|Masr{
zA7e_!`v-WXk#Rz+w?f)((`dCQ=LOzcv`}PmXb%f&ye+{&(o9%XDNlauCbNkLCCE}u
zBgx3~ilT^_%ng${Na^S{DlVLgIkC1vmL|;R3tqZ$NM6?rhcKNKXrb_t;bgnVBloW$
zPplA~zQF9(4kAC~;P5)lwBYncLL<qDWs}YG7s7{b9I1{%Yc0`KPEpmG{Ul<3ZI9`6
zPF@*|_f#IFQp8C_tR$rksUYFI1G~Kae5aYBwMJ`Ak|gwdeeS#a9v*quyZGqGKgI_>
z@IkazeB=`!BVR0d?%AjL`~Qo-$Ft8rk1^jHLp6#bdfhIab{DA>RasChk+&+(iFHJ?
z*`nEMh9re{n>0y77sQ2k5D<wb`}+yw$p~kC$mU(r8QfB-;9V$=ceY2gS~*dq(UBla
zG*_>^Oj(vxReg+3iY2=BW{qE+jX9$<r&LDm6_p3$BDVJC%w|(oS6i&DwTWy+xwlOs
zVWqRiO0NUX;#{C7n5rg<B6Jd?v|1{I!^P={wh?71ULSG32p6RAVFcQb52CEH3<55(
z6l%q>#8-xMD>;Li=3r8i7oiYsWC>Me5mAB{p19qhA;W%_qzO8X19F9+-He${rj$iN
z7PU#T2qB@aYFrK083ZAsvKeI{Vg{oL*KY5#y}KYUpf<spBZ~}YRt0gasSOl`qO24)
zg4T-T?n^DQMoO`F6Q1}Mn-4xjb?_ec|KJaie!=*OFVfW?X6?*Hip3n&Y$1fA)m-7^
zU2lh82OkLz9{U_u|HH4dSa=SVV>%dNy(jX3b@=5)bv_!?>i6)3^G=J1$`s#azI(Km
zJo4}({Lqj7@NpN<3xSFw8q1D@K=8~n&+^#UzV?P0-rjravP5`6BW=)acWJlUv^(uE
z1#IAT7e#SwCh=NC-WYTI-^F5q6pG5!RAq^fmUc(8(ocvZg>wck9Mi>+&;0)H^6^jn
zFglGX^MVU^oad9D_&8ts(xYte?0(N@(d#PgYVWXKVer(>Q8|NkA_V#38BNCQ9hAtG
zb(+mKDiUBU)9DnI#G${Om(1r2gcL++LegxIX6Z5fErN_I9Bfi91<*m6f_KDmjBuVf
z)|fh>DhsrUF;$6*G*OxnCn?o@&U{+av6}I$LIZJ}v9h{?6o&C+f@*e&vIbrRE<)C7
zk~SKFLsL1V5jZN0D-m+3wAA7KmlbReDz4qwW@~%IbZRkmhENGY)JR#Ob%7Rh8rsrG
zBN|ymuhXXAZ_{n2B(cUv5p}s>I$bcERWx>ow9{3}YDnxW;+Ceh*2gqjq${hzt1JzO
zX`Cf^Q9&grD=6#mK17r;p6oL(ie=uOJoX5z%aTOPWn)gju^M(+0ez=Q&XJTGA|+|H
zTD<d}@8s<HbKk}pIx6nUvf#><E0kr4l!1fwdPZm!r9{i3<%p+{Wf|>u`}psxs)}~I
zjrX3iEYVR!lBBrF<QOH<T4RjC*m_xn2J=;GJ^8#MOCu_3o_*#SiXuluinY}>;v}Xh
z%h&bX-)$(u=U(6`0E5TiabBPZh;(gCSOd<7jnDy0s)7fUh!Bw`Z3MK$$)#bl53MpG
z#qq^RI!`g5V~rteHUs+LJf^B})}W*ab~{zgq9};9TG|Z>+WkH{(&$M}XR&<UC}wkG
z9b+xVT4wVZv*D1m(L_XwMz_UeaDZ?gCBizb$_gnwSt}uJ_qqJc0rPP|oG4D8Z*k9^
zr*KB{)U(&w+7A5OR+e*aGv<!dr|5K=NF9Omq-n~0QBar##%fl(4OZ4-dTAfuUWHR9
z+4#VR807`|&6n7|c9pc%rL6>;{V8$Upxf#}uf@!JT2ULFXewDVpAN|<j)hPxe9XaU
z%&fFbYEOxv))DPamtsC476POqw-!rK!#V-N{c*HGN<kJybXr|bpFYjs{|Enozw{G7
zPN&yd9!vm%QlUF-9(w2@-gf_eJp0`9-}9j)ggjm&9IsUwV)UX=^hZ%dtJ$L0>#?%3
zLbKTn>b|8|B1sa2fbn>QR)H)oidWmRVSW3sW#A}tPhBmDVo91wjFmk3^y6&ZzQspB
z@&kP3v9JD#oWk#C)WF<(X4W&eg4`ZmOf|C-=G!F)W6R0SCjCyt?Sow!X@j)UU|nTY
zW`Qj$>Y^k{6I7%j(rD|JX&at6NziGGP&)7*A+*FrSfkcDs-g_$^WzDoUKX-4n8T-8
zhOINIMTyjcq6X{2_Y_4DwTlAPNCgD2)$N3W2N47+Cd3@dNYh<gLDmAKWpAAG@(YLL
zW<k4|vDVkDbQ@?DQ&onoojqQ-y3g)`!}%zDDx%;-*K^026*kv;L~)3KbI#-InnHq(
z6q$~wW(9kPhuH1kr|Si$KkyLsBIo)yzJj{^68CyXS4HfexPVPlypX7@O{xV|65)lR
z8jdi9rRc75b?&*jeL&?yddbnn0PnD-VliJJr6!3}KJfkz@w=b>kJvxaok1*BuL&Rc
z;0Nh+yUV%1EW6dWvlgCw;+y1o{(a3g?;WNp@y-)RG3|EfAVDa^t158no6QzgUE%Re
zW|QOfQdQNfT^6Cjg$G9zdAgm1NQ4~l%9^0=mMS2EO5zYqXEQFGzraub&A-6|54@e9
z`}tqs=FMB*)3d+M!p>s}x7W%!YVSa(;PDbEqC_*CjPTZSp%I9MrO83cz(rYFnY3C>
z(q;plrN~(0r9jFcxtBs5xAu;{G>O2w&?0q9eo#>ql#4~+i&Z6lsoHYhA#ez-P}<OG
z#hj=l*4Jds6zgGqeFMfr=HoFdX@*-CB-Sm}P(l(5h0<U|FzKvo!D8+?+?{cI|A5`S
z0fS*qQAz|c(rZ#(qqQZDELq}cWQs--(QCvs8yRUM4MnV0Xsxi$vluTpJlthEolq|n
z?VVk+OK+o!EqC0pj-4IQNi?yFk-a{?3^8pwk~kuywM3*WtY%tD8of4!lZ+<|7Da_2
zD2<J^ltqCO0+hxJM;d3y$TMGzLqq9L*nE#TL|O=9t?6~Uy#GTVBuUfnxOoo;L#|x8
zLRHmoXhYXplV%ysX7d=O3#z2xDV3&a$jp_JJkO6?TUAwJj3JI9e3mhuPLEqcuX3B@
z(#8P5GM`qcM$FF64ujz#8)-tj(<O@HKj!`Q{SJwsc8=16^@75AN*hdeg>)FZERF@l
zNlaaZ@N9&jHU^|b3yt&=gC$a$Na=vk@tQWE5J>0QyS=?6QF|h-Id%4I_@1moD~Xem
zys9aynohGtT@*O$kdej<LE35}k*wE{WhoX(Ufa;3Qi4cDSX<$(K}L$K*`TW0%qJ5x
z!GEyRYZ9f3C`*_YlEHjV(vsYH;RN?RutGbFxp{lWcxFgq$C;IiJI-|3*jOcQG>MXQ
z+14o00#2XYpwrDmX0(<Z9_}+ZENDLbZtne&ALVd3=HMHTF?{JVo83O$?i!-eBrDrU
zEWXo0#u46I;#P2bm$k(l6zt8xZLToN4WmU#;RK5)!9@vKmQu_nm?9^N6s`{FLygA^
zkMQ!_bbNv%H-3-x25W2DS&I*U_``Jj-Q&6STFV9@LbmC2I(=Q?-^Lj7q9BT6q>#Z;
zKxarPmoa(?&1RF;)m7Hk*64IPL{UVVra0$VEEX&li{mv?n|i6<vqVv_7z&71Xz?0j
zkTRgZbxkNjWW?d{fX{vIvwZU7ALC#A%+E0%kB`^?n^oNPgrp-QMm3CWKvOd<5fZG6
zuwGJ5V4lypa5hEt4fA3_FY5q?_R1!U$qeHx&1Me~>99Ze<zXl#LWgr`F_|(tIK%_Z
zR*PQ02Y@pcTi4hk$6HI9W#Anu3bs#09o7S0P}L443|Rx>NZ|<_^r|RGTCL!%z2ua8
z=kS$5T7PV%BVvuzp2NK6vtNFR;m(*yC`ufSOmp^hLcbd^ndIEw$(hVF$cWHet7s>d
zlj~hhom?YJQmlj8SZceVv^lPluuu%k3A6Epyj)Oe$NYEyBO6|F@drOlECpA;@hCAb
z@?d9`X#Gr}D+z&75mIE}3I@+V&i3cNMxHczF;2Pm+!jS)sg_+RB|tidu_k!TIfq2h
zX}7rd?tA&&&-}+f<(m%-yGW~$in1K91r(0vn6-xM*IxO)=bF|LagvZEF<F+8Wm$;M
zThesq^EtEGjDEkrJX0*zTV}Ia$mLz4x*_$$Ev=>;N(;0KJW6XVRcQl7Mk_8|x(5}<
z)W%U8&$+Yb=yp0>yyGG-zkHQ{^^5=JdwcfR^G2-@)XSo-bPgi~+6Pj46zgD=>K)_7
zl+k!Z8pVMOtO8<`HX7)L3@sY1mP%NUl1ox5fDp$GD(N{+7$}Q^`Fzgd;UV+c9AP2R
zF^w#x*=hg|Z*5o*u0{c6T2YsP&)Gkmva-4Y&QN>kbb1^f4yfmIRFZ_^PDw<h@e-_b
zSm7wFVrys0wO97py*Xq)H5dmD6AD0SNHb_A(2EzOaYU<;(Qb8VW)0#fOb!soa`~z@
z6h+NsF=D(kW;UPTB2TB$M3r+I-H1_3BiDKuStH^man!(O4X8cJh8CR_K4c)qZ3a#<
zs0Hcj8pCP9WIU&~V1>dVky5jm&QK7d%e>;p|KeW;Rr4Qy_1}es`loKZ;d@D=n2q&y
z9)9P;O94&z>8r(^V|!<to$c)qLjMN#@@R?DMNxcQ$YohZk|d;QN}8tBbsY|yCAm9E
z5>{4LxP5y;6h-7^dE7=a#vqrhT<0vp2Ol9*Ls>X%?3ql)91ady+dM&*W;7ZN4i68%
zO|<j<js=E>h`?$}>nN-RFGKAHOy`byUQ*RHQIb#;1ybtZkEJy_N{O{%xI19~U{2*~
z`n`;e&2_SVn<xrVe0Uh{?&F1G<IEWlj>Tlk;od&oUN=}X)P~V$im)D0!T#<J-rHk8
zLa7y5Gb77VT-T5<rl@G0q6}6Hwk#?0f~?gB>z5|Tn%>GPz2+KP2)r~nVe!J@l)_Y=
ztetS;%nBPDeX0VUymp%xU)ZIY81A?jae8f&c6SvO#Ym+?dE_BUQcCZMfsJO1S&_51
zyM<JqjrCP}cbsLK=WIXqEe0<>&xtr@6$OhNsWeKYDDAN;C-IFouCDPdL2WF~!NJ1t
z%A)4%iBpUY518dOMre#mNCcSq7?niGW=d7nSfvnN(`_}$i*abpm{)z<ju04OC1qJN
zabvFEzQ)xnSGjQM;;X-R)XqJ+6b*;mym{+&`EEJym`tZcQADTHp_w#^v_`660o81_
z=&khV_5xqE(P$joFdR`Dlu}ex#cVcZI-5~e6<L-Yx0HksEQ-Z3muv}Rp;SPOtJ+XI
z%NM`&c|P@Hf0_I4yO*zg<*_$ifb>Yx#5u*tIC4Chk_h3JE(!`Sg3zHfmhIh&ep@jw
z7OZtwG1gEP6?jj7ZFRZVOQcrEogsukA{gx-a<D&O^~45oESVe(sq2cBm6f2vRgxqN
ze6H9SymwgZsH>V-#&{1&tIcGT6KTuldO{p4W{aGDze7ErV~YZnrpLl4XKc8f`7nE=
z3<(=XIG(?Hz}9}wdfT(!atz0g!-e7U%K{}_P`%3rktz|gz!--X!OqLLAaFVQCW@+H
zFy1Gx<|wDpr9jpmB?N2j4mK0m806)rXubD+tlxhR^<>Cw?`38OJ7mssFdWiZUBxR6
zB1J6<_CEi0b_WGJXHN3s^EW9=PsLK<yUbWMQUoQvkbx%Sd{C<oN8e87{$nF_Hp;T%
z(MP|;XgH#`(hooJ-|o-2v$e(J-~47knBU+yN@r0@9(R7?BydxcBsoU1yWQ^cvQtx*
zIYpUM$MvzbNf5*YG*`HQ<eA#DC`!`AQPvez<%wfOq%t0P<Xt!?xPJ2{ySuxbII+p$
zV93iazs$4GJ^wvF|LaEV44&FM4B(c8`9K(zRCS4$h9qflI2zJ!ws3Z-7>adJ)JE#q
z=S2uX5^3bpC^JwLmZ)4=F`G;ok498=jZ`7yv)O8L;=~D5Em@4`?Cxwa8I7<eELb5O
z2uBpFCBGv`-@Cnt>2%EI#s+y=lO+*})|9h34OFmkk;~#vE9RA9XPEQc%e&m%n&Jve
zf}x?H9ZRCv(`ZWCos2YtSb18yN16q{8!0rNz<aVnpq0ih#UEu|k>_&;lRZ=g{Z1R%
zjo8pVs24#Vzj79Nm*n2N?j$L4?5toBr#N!z`2xQh;>euy%)Q{1>5ROF(`y@y_ID}r
zrTvMHsGY}`719)BGDPJRW5dUP@RL0D)z6c~2}qdEzB_cSh2qYOcXIm7nWLlYctIU)
z`d6=9<!~@~LpCRJsXhaa`Ikr1@gr|7t##lnNJ+QbWj31~qjo}wK>U^xDFv=}$MZAT
zpa?{eBrHW1mc=3<nL9gM+;iXC@Lteqw_o#~-kkBPvoSx>>|xOq)*VxYk0NvyrC@(F
zW3@MAy}!2fzzB$GoTS7`acgUzFFw1)Y_{NZJENV&+_<sF>ROMLwH|2}<D5msG3{QP
zay~^`N28HZ79}^XUk`yC#$y(i@nAq4Yx--eq*+Xo2Jy*!K4UT+vnX;v5~mUK#e!xd
zqbv%%t+66x@tUFx2crsTkxT{7nk6Yo5cG=zttC3vly%KB&t0Qf7^Ez?bT(#nC8OPG
z5l3yD5D4#xmm(@}!_D4#Pl=&iEEtc5G&@bS6g09JHJ_7DYqq!cS#Nb%Q87Ae)4u-!
zlJ!ljsc{18-U>U$<4jFHo1q&SNP$gK2HtZrPO!!<J?cX1-V_C@kuY<DNGp<dMm}Ef
z<3IMFaX8%NOOO2_m8-tPHy#dX;{>j>?C$UIZ~yJT=KUZ3AZa5DU+-%K2HtxHg8`Gt
z<WE22{@C9F)&&)<Qi^uFMRTbjcRs|-C22yECCBT?x!?tAt);GOX0sV(Sx}l9>)i40
z9JP{)vRrzIE|Ep!kP4)d0g*7Cyr{Unb&JV(%<4-2O)u&IE=4T{wI}y3&`yLv2!$v3
z+u_6IO)5!ILSA|%`JBpxB$h1AP-%i#qH9Wox#`5w{sXS6xqWMg(-+Q@rZGZ$R@T?J
zee)*0UJvUmqw$1%x(Kceb)ZSux~7=tRQVh)YFh0sRb@DHrbV}%&}y|%k!Ct8Xf<2p
z(<!la$RrMXz_Q&dbO1lRln5M?((>Zf17hJhvl_Fun$Ydkv<4N^GGstEbxkvoWLd<b
z5)4LjDidsurn8d${ULG<TFo|Tk|2?EvL2mI4`Cf5uc5YB=_s@zP88j43ndL&SG1ZS
zTM4C(&{AURI;?jzIEfG`H@2r7uAbo8>w7E;N3~S<Y9UY#ytB*%R4N7`&<LC}l*NL<
zV1N)u#zEhCmWMsk@b#~Ml|TH<XZXv1?XQOR_-jCgx;FgQZ~aGJdg-M%1S-5bf(f;C
z7NzB}uaS_io<((CA0yvsnvy2LUZ}2W>e`^BrZR@AsyMb?c2Jd`gM*wjEsq6NX`FC!
z^E8+4x|_-v4)zcDolk$7C!c(h{r!Co4i1>l=il?wzpld0_<-0M|0;YeMaUZrgwk*u
z$fAU@TrgYAXf;|G=P~tC{dS}x)1ak9I16Qs@sgsdnNB7Q2Lp;Cr_sph_PT8Ld!%uU
zvkp_&jE5t(uWhk&dyh0tND@VoCd-H%Kza-|*rX^$9H%H5p>;y#LSC#jB?!lSQ6Qoi
zYYL3B%qqj)V9u4D88;3rc_GkJ(~boFmS?@IX~dFVuSL5R<0vW1DeYE|MzW69f&7XL
zDk?9)RylbQNU+s1v$LuL|7>kz1tC2)fjCX@RYiU<p?~5|?)}7HrMcNh)j2nxe1cnf
z&UiGy))kxn^sySx=!gUFFk0Xn8Fg7Bg<$R-nGXs~5o;F45j!{>jHq19&-~oK;P%!v
zYHN9w&G?<>c<Av7FHl19;DZk>Z>z_WTHKOBT9y@;UwEF;WEfPpuVW|`h!Sa<93OZ`
zXo5KM&hnnJ3I?f1t)xby!FW6lB<;}uVm_N4-?YLC4+gEF&=AK?63zzUf-yB)w{PLB
zMJvflf8|YA^*vLUoGcf-U3YLItRsO%qR<kOBp9y_#zR)RE6c(zJZ}~y?C$RIyN_Sv
z#;o97cb?$BGaGC+Vv<B5TN!bZgu+TIyt54Vx7fM81-YToY_WD~la+3d**GBIoqmUV
zA9;|t5mb2CU^|Q#bj%9tIGh0M9VV|axe186loXZ2B^h%&MO2Q&b%<n0z%as+7bQjQ
zP?07}A{>I9{Ryw!*yC__N^C7>H&QO%(Wc!^>GW0+LWN?(8M0RE)q`6r^4W~t?d!n=
z)YK$VLaW)qnc#gF?`%=5oI?spySD}gS{Lr5^R@?>w>pd#3#OAPRcY`D>Y`#?4cTZm
za0o;cF_(_qRwPOgMUtYlcxRYli53-e7c6F^l(ZVF)V1Zyk9~o>_-<UxkghY!D}Mj8
zzsFY}eUy)U;^RlOtz%_61avu$HA)G*eN~+CUB{CD7Dt*!t3{*P{5ECY@vkYdY#X^_
zRN<&~R4#wF(P%K8PLD57;qVCyMu;VX?FC3M^{WQT#lqm);XTy#o1Z!XFs&`l3oIVJ
zyjps{ybvyhc0uXw5zbSVj(Jft%NMlTZA3E#EpT;3IW3V0(kw-!Y4GDfFrO``YDXMt
zgb#6-DpEu`;_%=QYYmHePV6<eZr-H7+9ys@WTMcqCfn?h=X3V=4r#O+oIAS--qCEe
zD3`peqBcZ2q8N|JnjO59;mlAf_)_ZN&uX3LsmnVI2RRql;PmNL;#eZYoHW%m8VT0r
z#8T7F`r$Q;nw^6Y2g50KZBck;(;1NzXsJjWK>>}kL^>u)1(k%lu;kt#V@W$|5na3o
zD|(n_55KX-`h~NkXYXde2oadxhs%$U0^)>MO3O?~91iC=C(u%$d?@O0mcm=+GN4wm
z@whrCRstjchyd?T909c(Os4$iZ~O*->973cF|Rb38+#Uu885u_Jo9{p6tCu2yxtvL
z$Z#e&7YfZInu2w~iQuTX>-BnQt%LWVF;sO)9LIE3H^j9~CQCIr9AO0BOX|{LDvQ#N
zBvD``_uPFijb@w5I27>YHy`KmC!ToS=P111!VX1=rxr_<s380UL=IA^sfA>=nA3=2
zb_aWiNE7J@Eh1cLkj^djm5$n4%6Y|jG-EKBF`mtd(}*+YPILPF9mHvjP!=JAeMK>u
zak#z9-u4b<VYzhQ-JCvu3gc^v#SGzs#9wJ0ZeUU1A6Y!M_FzSb0;(LouCTshQp_-s
z#+e1<>6{z81y^=zhE;|WDGgn)niZVeNZDLlA<Z<APH458Os6Bt%Fykvlcf!Wm&jNJ
z5<8M8iW!X#86NJ0+eB3_$@`K<n$c`^NwlIWbFw5O?)LFTg(${MlNCBAPO`8CS(34{
zsJL<SI<tdYXeBV!J=mq!o)emVV`;DTDC!dOIl?(+K<1!fYYrDBbK!`*!a1njf^R<k
zWxl5=FP5=mLK8<R_uh9uLVc%2=oyU$T)Fl#(|mI5O%(o$@0cL_Z9__N%-@P5g^DCX
zfyD-OsBq*}P7)`_uJ{1uvB+U5xOok-=d43Vn)zbEF>T2oNz%iDHP*5y4cBkpz*|c`
znINnsmLZGLdi%OJhCgX+8N>ZLT#TFuxvi0cEU}zEm7=0WSQk|pV)FnkHIvaHPds;<
z%lk8ABH5WO=uQ@Nx*1tV2GrX-ro$0=v7jt+=Ck0@*y*oO<pnohyn+ytX0yee_g$pX
zZ{w9bzJv(x@wI{4gtUq!T6mPqA}FA&mi#e=l8&mH;p&>IE>T5J?FHKhV_v*9;c%gd
zHFP^E2+7{Tf_&j<MUp#CYwo^djb=Ne)#>AqBxxv0^1Ps|YvSlgBvW#@f19$Hk;EBE
zoRKCC@B$sh%;po+)(zV4{vdHINzUI%)LF%M`%HAqUY@hJw-057cb2NGFxGPV+&Qds
zcw^YQex2MpYU{|FE!LYEx6ApG&m_P*G9k&a7-yNx2mHqG{4%xu-UT+!J9ZBa_;>&M
zU-SMS_)s`p;Wc)3K~=f3(I?xvO;y%G!Av0E|H=Ei#~4Rl*BDb%S0%=z$Aulhd_HGB
zp9k?#@+vRZx#0VFbTNve2x~2AnlhWsSS%Jnn4yB=+AU3ry%$*Ham(UU$ROnDw!1`0
z%I@yon<|<(8nrvJqTrbPtE51B8;rjNp#UYA=@Ui?003rGNkl<ZVK-DnS&kVkCbXIz
zY*8|w7i`}eaxj?FNMp{N?6P*EOPpm`sW{vplciZWjH{9=FNnJ>HcoCb8VqsPv9`X-
z^WS)m3wPbg$ulPdnxh3(Ss+YJS(V5nVLr}D(qNG>Tg;hErsyc5)ox*2O;OF6RW(wC
z&O(|rh|&z*$Z$^b+|@%Ke`=dco07ZlTqA3wI1_|m#t2&NHr6PRmMCkYWK0&-Tu2i(
zR=VUxiCKEXW=V`Z;;k=pwncCd0X#lZWU4_+r(|a?;_i8bQ}6y5mwx0Yk*>rQInO-)
zI5(gA8h2j02fd7R^Fjud@9HXMa){P~-P({kPl6;S_+3uC#Y#}NMvKt?c2c0THk%tK
z&{};Po#8tVhw<>r)mJb}8o_HvyWQsDhu_I>{q}D#naxlHl-ar0SKK*gacHW#K8}~G
zs*2fswiL;v$HKm&q$6VtWm%Ay1;gPGfHX-c36?>@$4<hM3gifYiIW0Xd-}aDct@fn
z>;3NkzrFtsw)8sBJJH`-U*75LeNvy(IRlMGA|f*>QkEqvS&}JBw#TY+x*oe+_s$*P
z>zQl2>~VR@uF=e$xwgk+yJjrg3Ra*bN|Y&*A_)e70EpB88l7|Z$!W(gu5kZY`y2oi
zDWL(#rlfvQ08VwEv-jD1f9qTCdfz8-^6nE{yLOGuUf1#eZv(fNJYlW75gLmpAQT>f
zbPn`Vy3v^zB;<KXt}Qkl(JoRvKLSrN=nk<0$^v>l#YT6;^2QdUV#MxUZSFmBl-XU=
z;0c5$5W-;diow<v-OVj}TU+F#ly+-^d+t3%v(tvqBM$1+CYo5SokhfUz9G(HLTH5)
z5JWyA@Tfc}g`uoUs!_&fPtn&tSJo8EeUG$C5Vm5fKBCjm%uR*N?W$9+HHo~4sdgKs
zDza>Z^y-wQK~TBTky2EpB8p<dD55G$Mx#D)6cdI4LihwhL>NSbVL;#sVm}~=6Qt*)
z_Fg{1)N_P{B&gw`Y0vIv_1P!!iawPTc-A_8i1d(tKxHlUW|LmG58B`vi?E=KCK4Xj
zYUhw|4OVQk7~kThZeRaj7{}~CaA2n}+J?)&vn(v!<i?Gg<ax=Puf^ZHsDsx9Guxlt
zcF_gS0YA_49g|{h^|&+OqQh#n8u?how+*Efp69(GPxhNRO#-7WMOh-_fW@U{%Bms^
zU1W?9&P4Lv+?oP}8=+f6AOwMgNEmiaN|JEk#*P9?qmfaF?@7u$<<im$r&fpL2FMK;
zZgnXKC6fr+wP{E|Ye7{RinPRNOQ+RgYG#I{mbiq5v6v-@Li}1CC7eQ5yuq&v&v&M?
zWA&ddoehML0zZx@%L1!P_kE@y&nvR5!kB_e54f>faB0~vQUO6wfra&r5@Dep!qFL@
zgL_)c&rK6Y5kVYd2xzogU<!)7AP7S8yrdj|wyG>C@)XbWo!?rJ0N#$!1VE4a$o?j~
z?>@=gBj3gDd+$T%Ip<EF<;IPxT)%Xhnsl9{qxam$^!xz`VwBe0IP)4`{pwScLZA)A
z^#(zlFyrN{Yd1(X*5&(dUkpwVar<4V(ion3<{8$OR@l9NAKT9JQd*30oVUrTX&MtX
zhQlFMrO}0by_#|A7~Z^=T(@d_yxQ;gv16s|_Ik+koZ)cDXfz`5gE!G9@;rA|PgzPe
zPK?;**Mak=+wM?kt;Sts_Zvv-j%?nzD{=JLaa2*Uv9|vEaCYAw<e09Z9eD|11tAc5
z0wFa*RD_Wr3MGw3M6+c<L=3VK>2Qm|2Asd%<J65+HqwGd9B|*}4EJY>shK*eQY_q9
z;>5kj$VYuvuH2wZ3+kN~?YSB1jRujQvwZaip0u>vZ3Gr=Exru!f&hdhlpebd?js)*
zXzjAU!&Z~ocH5cmV=zU^aM0)C#ak?-9)*T_ENR!>T3j1imezBQ&Pk3Pog$2r9dl?Q
zJfK1=Ll7hs<&aXRgu*9GYS_roj1wAK!#G9|;X0DTY(!DyL}BC((!x@&HE2Z%(ho6l
z4UXSOIklh0?%jA&psJ9|mu~X>3(u2pF41`Kad+JbgTP{gki7>Eu=&ac)Aa<E4;UHE
z(5kTz(MOkt07+pj`B?pG3yp94eCoG8xpS}oJt5w_PXOZ)E^OPOeE;|VAkRMc48Qn`
zzd*mAy}tF_=H=*~$36ZVyWC1CY-9<-fFdt;IH<;$aRg>W7)9hoL6&9cF}Kc9CU*QN
z@WgiBx};PUXU?5vd1--8XOhljo4@rp|7&XX8vpKR{{!i5IyBs-L9um(0nT+jL=a*K
zBuHt|7~(L*mmyiDk-qOvFjd9KLU(IKZ(VV1Gv(T_L>bM|$p()eImoWP9UARA7UlRf
zm7?DtFk0K9x3)o9Rn+Qr_8mLU<kTd>lNf0bo)bG*;ZtcB-mH~7{TSs06?Qy6@j~a$
zFQhAQq|(F_ZO)usU||%nRzyfIq8?VXgNprAlJ=ye(GtX=CJ-KRR7YDF<r#6UL95lo
z^PFmLG|CwEdrVBUN$NG{uM_xg)M_-fT7xL8<NL0#bH@FyBZ>4e<6SCSTjj`OAIAub
zs+_0NAT8;xuM??0TU%Rv;JX|L$y$q%5)E`Fr|E9>DAN+(KrEc4hUW={hQ#*~%DlZm
zWDBKM_)$QbrQ}&o97S$?IzB#It2ul242z3PRHfPZxo`U7_r&%h)ml@QrL!s-tDtvy
z4Zh2?+|IDvwlvA|Y=>u86vd7tT}ZhTi}c1Df|Tx5QdXKEFf6aG@%#%faQV_T9(w2@
z?m2mqS5BYi#g|^%88yFC9QI?%aqBRa9yJLy-!Rki31o#*kXHp+;o^4EQgLH_$cxwd
zT+B3DNb1(&X#u4ZMPyqcbuR)xpk8n86o;L5o1h+}JY2_GxRHTht5IYXTPrIR=@2gt
zX>>Y7Ndm&z;%KGNWl2>Oc%H-$9C27S68dS6QC5%^Ig9HBY2nO+hneE$CZwv#R5)aJ
z$07ocVX2YAvTMp`?_7;mbC!CoL7_@QFC?xv&^AD2Bg)d{#ui24x|u=XqMc;u4k=NB
z?>RA&E4(W_&qt}8>gAW1`{4ItN&|k4sMX}2hubW!cUfInWVm{rEGzinNB#tO3EBwC
z?iT<4-~9)cvWz2l9U-YDH0n(XafMo=&eUMUNQ-fE$d3Pc`>@t>ec=}W<3Ik#eCQ(|
z=Dz#y<M>_22%~^J&v@nKQ*J~fLR1A>y&iV|?Vfm>LkjL7TNVXnQM`$Fx@`$m6a`8v
zJn85S+qMj8mNH65JHO*#Fkm<w?x@ern0L30?ad6&`Lii&un0z3#n1lSFEG(;QqIh>
zvAIe0HVWWxhgb@f6&PdP7*z-o;mme}xMXTJqE-_G0eG?`@IcBK5Oha<mTnGt>hc!X
zhb7tyysV<1FQAR)aDS4;t#$V7nIiBlON&c1I~{^)kKXzwH(tB2lc(IM)j4_Z-Nd29
zYK`>W`yiymNDJ1(M-nwWtajO^+fi(ua2?<rjaP3n2>aYz+`=eVumw?!MB-b+;a!>|
z`(u*OBlcq_e3O=X-J@2k6UQOeYR16__<>UhSCw{?i*aXZtGiB~b)AWHQ6PX?y@4-;
z6WW9^1PL~Xs9H^$({oJU{~$(LJWq1-<|4oOtG~qBr5Bl;oWhH1C@C>oI~J|<Fn)l>
zBZ?EI{FF&$83>OJT~X?aK)9l=u&=Ko0jM%1-%Nl14}W})tN9U+KKdA86psIR`=H|}
zHj|T+{Fxv51YiEr6Ab!S&~NJ2y*2i#=Xv;^M;yg&0yr+nw)dLhaEL^rs@GFm%CaO)
z-8!<?x}r|&@%Z-5dynUPI|qWiC^1&^^f$i2-~Ky)o9XEp8jU804jtj(fqgvk(8GNF
z*=JGat@}aVUSVg)CaXf=3y;tWA_2a18KufN1y$gMNG~GtJt}3$w4s+7t}YHZvslng
zOA;x#@8C2aeDDx^+CE15SX}jA6%}i%%dD@j;VHrN>@1zh4sk8P0F^OFAzfiJzUin+
zJAbayS-+a`4NF*O9T7x<<GUH>Jyjvl(#M;aAevWfon4~Rn(11}e6wO#Go)Fo<Hv?L
zjtKmOT2vzrVv4Gu(w1hUfx$y7gT{qJ$F(|P9HDX9l3oxH1Tke^5{5BJQX>c=r1Y`I
zalV8UC?U`SCmiVyD0PLAt{}4-Jl|)~-QecpWqSRAD_qAHNoAb{O%MibtqjmviPTV&
zJ|klg5{z=T4noS?x<Ix+cW|xaE7z~_*Z$gH<2?_(hZ867;qc+Z?BBnSnVBh~DBzV-
zr&wKCL#x|6u)BMm0ndG&*1WzE4Tr;>miPAPC=TPX8EVOJIAkyw>=;JMaUoX}#aOv(
zQR4#tjS4DjEy^AJ(ME9f>J9$Izx-ERyLy8tiP^t@KQq(Q1VKPizB`_Fk&xuZ;7gwb
zNg@q_X9+`r7X*~00U_!3hn!y+@an>d#ePYuC6V<Bg(mO??bv6}bi(Ac=ekElh4ex^
zKOhVv;v^!dCm0D>>pUXIqXVQQ9rekEeN>hrGJ#bUy46C;fFO)8szR3qwJ0EnBBbvu
zP=u`GH(G2i+~l=eIZH!<!I99l@u&!xtd$&{5wz>!j#rx(8X7H+S}mcL)SXR5Qm0-E
z$aO}Tk1(nrswb3rK`<^vw#oZJn2@F!K@cK^hm^s%gC?;iL1{%a>ccC~P!4+7`%jSj
zEz+e`9(m|tZeD(cFQ0pXLAT4f^Jh5zz<ZGhZe730*(=xBe{dg%4<8~cN|bdzouMxY
z1cbtpT7k5NccZ)?Wf{Nr>%YdSQ!jDw(0)Gl@sIJ*@A?QAE?nSqpZyHy&Yfd@eT$-0
zY>QCejBok_>$-hX2&ZDTc1O(`1g<-`&6V3;BT14Fc)^=g=i6&yFc^$2S6=T>gkgv$
zovYh+QRl>IE>uklh@v{Lo<7UU(ke+UVSa9o<HwJ4`nA{C>~6ga@7&nWbnH4FN{^bf
zE@#!xnBARFt5rn4aMKg(VU06b4?LeiX}OV?T+R$u`qX`iHkOS{@!YLdR<`=IL!XBq
zY?BU0%<bAu*s5cVp*7XPWF=Kux_goDp^c$bCBjI286ri171AXT;EGSI#9FtxBYdL3
z+j-xVu^7+e;L$y_nl(ssHv0n#0h0|yLk6@a0&0otq)7>p@K7Qms)f{QF<Mu2yPK}~
zH_rTDnPQwI5{?hF(A!!i^do9v9T7D^3PL{|Ul$$-i4snrWYe7C+B&=1E!1#GDd8)h
z{}iX6`y$!KGSeUaIBNX)hk;W7Yi+rD;|4}+w842#>Hsi7&%TO=g)Ra$o*)(9bTE3$
zTZC|SQFTA!v4<b!ul|kyoAUvD<2PW(*KZ$B3akOGzL^eS`MruecS!Pxqlk8^<q})O
z>%Mi{`->0)-wRMx$!ItvO<h_`x8L32ttwUR90cFE@3`*Xcxp60u^BXog0mMcA}(BV
z2LSm5$By1b5JVh1dX%#l&cD@n>o)C9j6qw2kO6`4h@>PE+Qpja3T?fe(OBgBghG;K
ziluJBYYPLecPqL|FdKzDJXzzhy_3u}0(9adiJbd%spxKQq6)>{U3+OwbqJ%#`70TN
zHJUJv38EU(_o+-pRTSjIAv((`MkB&7+Hv5Il7ukwotwA7<+x@Uc~MeU2J69|Jss9I
z`mC<@nQr(@&m`0)>d=^>6cQtS8gWD+LC6YKDUzg)=ZCJ%ZUmkm(C)-|(qlMEDXW|?
zjPQeyC}@xlHi_f<*a%dhjdG-I&%;R2Qc_8QvetQ(MKSminOCT)q+XBd_BQDCHW{P^
zqdX;y8c5&6cpi+SZL+k)D>cGEBz^ieCo?Wq%;)tH%G-ej^0MUoh4XA~Z1C)}&(Lf(
zXf$dhNkVI)$?4OtG8&{<E8g%^Zw*T8osn*(9A_cRGHSKjPRGF*Llj1&!_;x;#$$IW
zB}Gv<1JJ74DeB(XCVwM2_w|viI|dHZ0u?&jh)OA{!GKezPEl_(-Zb8QCukvQ3!knv
z)ICW}dW7RmQCm$C1a7o00-nFP!Bba9M8ab}^l8?hDI{9CYh|__aA?NTX+W*kLfHzT
zbAliuZYIR_1U!#mtUv=d{*~T19?A1|EN+Bw%wC-r7|-BIi7E?{$aT~3eT=nasS6M<
zwV}0dmPWBg(eGl6kM9QrzF;oS*f*n?XeW44j8Gb(J^V1H2m-ulilH#XGQbZ!231K>
zC?b^;x7x@cpb>;rV;`upEL}dRVa%^{-50^OwM$4nX*i;g=TU8KQHG1yOc8b><gQtu
z;>UmLud;ak66ap}I=}g=zrat;?`3LkKdqS=lGX&bme$zl4zXB{A3ugDE7T|>vzFXC
zI+y&tM0>XZi!*LsSzW={ilwCmzVxLpu(7eu*|X<ZUGJi_dCLbQi`#fc=?Ed5O$Ng8
zCZlmzeOrCEP4o@|R~QFDK%>#1)9$d<-J&RpoesqI_{ogD$E<p@kjRe5f>H=}&CIfE
zW`;OUFxon<UZv=6Zt?c-IKLapTC{be-<p&}o=eN|0*GUQK;z2@VO+7|dy+spM}_rX
z&V{v%%6Qxr2@cMLSl?%@Td-7=Y^aL&-Mi26*o+};)_2BdTHrX}zUR6xfln~@z_ZpO
z{Qv`$N}<NWksa?<V|OCr4Bl7(q^OMR1J^@n?QS578k2*N%e|6X143=kR!~qw2Q3OI
zT_mArG16mHR;a3?)@;+7oWvMMO^KxAmX}&N%P^(DcuY<0BJksJsz^5Gd%MEV0u`1@
zD}({HxJ0?tr4luyZ#etH)ATks@q?J|FvA;n={)I!keokrip7Nmbbo*x<^<Bmc+LjN
zBJjmH;Nb0k@B$*?^Wc5=^FRM@|0{=&9mNghVf*^Zvz)Jg{Yif1m;WQrJo7ZyuHN`I
z9M5lG)JZ8_Q8&?IvNK5-MW}HWeGmjYO#^G4_pa5BPnl*zMx!BBRgtDCgTcT>`x(VH
zUu1`7WPzRRZpT<w+AvJZ?U#xWj@owp#tqU@&SQ^1#_IYS8=ISN>Fv6`!p?%R(izoC
zh@?}*8Y9Ox5@S*>f<RbQS+TmFadxrKrA@;~MC`$E?@Ypp0}UDzkf$m2XbPmjdX}nk
z{ItoAixJX+bX=gyB7HgLriEA|$+Cjcpi4IDQ;mk0A}5dm#zGK9)LIR^D0Bv{7Sd5l
zZ!n-J%du&oB?u)=UvqNr6tXM`CMMYM1J10D7_6@oCrz5Q1`R)_E_3!w)R_%y)RTxP
zkj`t%k6g)Q(w&S_t=%}uYC{w!1YSrI*QjvCsV{wG5TQ-UFda}K5J?j}pUirM^)^<+
z`r-<sjWs5APouRWuGi_N1zA}*4?)jEdOp(g9ZA-@plpHbFn|l~mN+XJ=fU^(Z?(o+
z(kx?nb%n6IIlj={2}o&0nmP}qH%a~865I*bj5~Ev7*Z5DX_`{2)fkONV}-0wo;x+$
zHd(c*w$)2+lPSjqUY2F!?!4VGl@vm}$;tuTn_+R@hLe-i2z)vdEs}bTE7z{Fwz2*;
z->o-e+H<j0p_FtyO^k2?1#3{+AcaRGj9A+$dG2OTQ--{EHelD38{_0@L0LGpT%)0x
zsQDyu3omev`X+Nk?J#nAk*3m6xgc;n&m~Q0jUV_#VMq`J7#-lpAz_jry#VP4<2CHo
zkQir=R#oMSEj`ZB6mZ|MX=W!vE?rt9Epi&|I?-g4(d-_&tr^l$ie9@x%U8rQB=MTG
zwaZzI{g~WT<O)nxPy`;<8tP*u@OB|$t=kl{EM<CX${FTbXHfwbV<c7tNMWf23^C-5
z3AE=?sN50yqXt13^4I_NKj5GLlfTD>SH8h-|N1ZU6MyAzF~4s=|IL5<-|~yU^h=yQ
zdk%!4x3NW7Dh6fAqBay(5N<2O--YWy;6y^Is_1sRtgV$?xNx4)C?hW_s@rS}TFZE2
z$eiN7EO&gh!YCn5VytbErU~Wv?}x)7Q4|q)!A`jOWP6gqV6bB`v~BL~6n}b$D+FVe
zparlrn=RTCZB|xRiK1x7hwbR$BmCg^|0%BByum;Fm;d`)P!Iq9(AHqBAn|<a5_|#5
zdRP^qD}fg|Z%^U7QJOCVXv1JsaD8dOQm?|79(!9J`#PFEyC*Pen#_ho&4k@E2}qwX
z_K?1h=LINHP~<s<)(C5fY6(FSyIgVWI#E?=$wvdKG{f`UeTwalj8a&m3F11|T1J`6
zJFm(LWsKvfhXU;fG))bEbx7_(7zh+KDp_Y?BS$qRnB6^39EVIywuwC-wQz%2_labH
zuoXsESfDJcaneCfw$){-GlQ2TSmoBgCu<nvxaU%W63%G5vWCnU1{Tt%pXAXW`~*e6
zi>eHMR7d(gr(SuL@7>y9ZqGr$#~90(zw|{sV-R_TDN2`ZZ8TO&^!QNiNf5UwVtD-`
zch)*T_`^TM{NCL>^Yy1m(-D)ClLTSNwQJY-@|T|A$tS<c_3PK!?DfW?k=uBR@!nal
zB}_D1)RG!VZ^stMeUEkZhZW=dr^5FfpS4mYMOiqGt9=94|66w4w@1g@%@*;d5tjfa
zS`8*UQy9I$i!Z%ItJ(U4DeOw?)TR-JhExRB`LY-r;TuO}mD2fy8Beg;ueiQpxz?8q
ztR)r|hnpe$niW$`MG^)0VU4^PApL+yCPYa{(uncH5M|xPhJ`qe-z+O0jTmh8$a+1h
zJjD~1AoQu#-9}eyHt^yQe9zgK^!oI+x>#eVMbUWnXUTEo_~~ZI#B7VNuJ@ST_b4Cx
zsUPR=haO^LdK&3T%2AJvTUWXE(hHn><t6r|C3{+uKpKn^1O%wundXYIw^Np7L{UH(
z`_#rR);bBC(XWDDnREHZ75e#*&fH!+BWcd<BRX<7@x&Z~_L-cSq0y-$JV>GjYrO%b
zh)|U#5E5Ynq97y*eN<IZ7CGUVEUSfcFE?5cc`m)9u=X8|lPoSYDJ=`K>J3&3<D2>G
z*XHdl-|Y><jelIh7DgdqnUUw&cnlZec_D%CVU6S64F&@$<-*6cR;V&Z3X9V2&$hX7
zLU`XsPVb)A_k3oiXIWog=f0CC`IF!M-85QFin8RV|G)nWXU?5}$9;$WUW7J=R^Sdr
z!jAb<+QZlaVM4UhoLwsDD#?2qf_<|daV;T=5^Sepb#(<#7-ptjmnDun*zs6E>xyic
zQWgcOETJk1f{>)yA*k1#a&eq<83X}g9HWgw1`#q0@%;c*X{^#n9Dmh}1yjlxJl`it
zYOjyyyny|CJIps@7Pf|*6E)uRQ-7WJe*Y(^H7C$j#jQ)1xN-Rcy^R$%u3zQMnddl8
z%w)%{?L;YZDwJQ}=?)2;7ok$h31!L>0o3btwxg}Y_O)VNN5FcLv?|ELCNE#SMC<g6
z-2M0m&|c)mVx?ucHR7-S-~TQ@{rCSLTsn82o0l$d^uz-^{OEhR@4ox_^<VoWpZ@fx
z>8`9|@|;EO6nL>TW1}W_!+fVrJ25_zYi;QF2Bf*0ASnIK*|2Yo53~)X-Dhj8voC2h
z3X&v8=?2yslBC99Fx)xVDCI0Sv~fJudc97gUZ>X!-ek<Z{q>tk>y9TT5JX|bpx>w8
z@9+FBp67A&$PwZ==H8PhIeE`LJpbZL@9>>-4h&XG>Yh*HDU6j^8=w_br67(a(h5T1
zJ}aX&d7)YF8CC|8)JUd%k7iA>|KL6*>J6j_(OMDJV}uuAt-<r1x110XKL}8&0wLUZ
zGw{jFlCjtnZ7R~uE%NRb(psXp{<`75wfJF3z16__lC8}xvOK3!3T4bV9mE3}VSSKw
zXvYyN-2u_GN3GT3TB>P1{2{*Yr~VRC^YeJ)RGh3T7%i@G@$<h*@$#3cdt%IMFVMyy
zC1l+mfraK&2VDyZw8dm4#tOuk{$$3AW@R*)0<~)&pZ@aKh@N?l{r8?^b#aA99(x~2
zv(EYR=g6`ypZ?8H@mK!(-$Z&orqX0ZPFWO?XUMW55FSQ)q;ANFH4qAC2luw-9~N>|
z^0_bk4$qx>k%fhu^m;v%as_Oj7mU)BvZ^S@G=#SUGd|Qx;l}j3ES)c?@9i`Nl+t5S
zUPc&%Na62vxh0OerL`h2-`sWo-Q7d`Ex!a;#Ln%SqtR$`{Ma#`f9XZKz1~}X9k;29
zHTXgjN*6EXIBPgx77a=ZYMwjg<OM8lWGs#}JuUF@m<fEEu?zSP#|j)T2+_upXE|CK
z+LImpAaI+I1>X;MHf13NRi&sZO`$c$gZf01V4{h{;D<h5R3of4ke=r<Evt&bV8CcN
zB&oTWDmx~xE2TgcE<%Zb-BFD{`M$&aslWN((K&bz!Vj?H99&`Mm^*NQ6OTMfclj0<
zzVJJodHEYm6$SH*psE<?lB8C1#B-%+wOW8-l&8qh!{`!|SA<DIsY<S2yUMw%3q-p*
zMD1Ccd+z27FQ22jc!PU>=+Bd@f?6vfwjrCVYXo6}7epxIl$=o%0q(>dB?;PShFOY_
zBoHqCO>a;6#k)63di{E^-{N9@XV@OU30#cTa5!*$2Hz(LLxhlcQh?Pv-GuE$tjGRY
zS)Nhk#hX-w+ugeD=lJH|JBj0+(DO$gd4#H}*x1-$ZhoGReB>jXyKv#QGjG2cGt$Sm
zP!o=Sxjj~rHYD^lp69dDEm#;Tr0|IX1)jy&3awoXU9(xE7Q<w_O%hLm=eVO$7-EfE
zC&geyk){|^;g7w!{5Zk-{_8C#Pa=cBsq4mc0_%AQtttC`=gTF%*Uy$h;0Ml}w9*PQ
z8c|CUi~t*i#FLYR%^42db)3fRG<Xrxz>!BD;mAV|yL&~EvT^A&PyOqEMrpQ)4b+9D
zw9Xc%8q0^Y(&LeP=(>g4xDlT((bnMYs0nZx$X>wKaFdNPCz!vB7r*cnnG9IGdYL_k
zk8<?BN2xWZXiZF0RwKUmM}Crj^S}L5UOx3aNA5a_@O<ixCP|!7Z`4U|Es<*%g%o0t
z<1V)lP#gPERd4ee*iLJ4!SR|>RTN6Oaoz9w?%TiKv~`hDP>wT`L3k3+lVsV5G)?jS
zfT~i&aSUXXc}9QGr&0w*D^yuhRykI<2tIA}j+*bAlQJxp{-DqPef#hOpTPGi3YP~y
zH9gJd<|fTnlb`&vf0kaq$Hhxm-sU?uetkj^AgN1(F#=_Mtk78NIBd=?$wzt?W!%BG
z0<yAnq9re4GK85<jm6~wt>e>3F96@CtTa}Yu5;s!JNv?k5RzIQtif2wi^_GQDrcYL
zM?UR|COYttL5vuu#URFJ>z<@iiY(2j)oVmiNKur|3vR5slv1DpUGvztZ<6dv#ioh4
zJkT8f{-5A`f9fyOoM_@)p&j2fB%o{xc7N#mIK6tAeS<Yxo<}2&uu3x=^vOpf_U+w`
zF$JUwS!*DpkbxRGR|nzNo<dTE32JhVoBb~R)tks+$xr{~pW?_PkI~w5kjqzZaC2dq
zMx(~5mtJIP;THQ29L4i}jvP6{nO9$?R27L37-1=mK|~(0Hx^JygKr((=xu%W*7D>t
zUuUA;q*NuNaVPC<i2=Xeb12Oy%}C<mSh^|^0wUpxHD^j3j*Y$D1fwd&I0ebIBjYe2
z&$kt|PR{=A3(ar+yHB<|)Z!ZNfBbPCdgwu>XJ`1g|Nh^<<=1hW;t5SCAQ~45){HMk
zD==D6R2D0Jl!f(;oYihcPluF1Qz&)^0g+6Zo!w1bo5J&Blr{uG?B>i>#b_|3*=mjF
z%)%8S()r3LrCdn55JYiE7&P&<#WT22bKiqFMud)MrM03gO3;S(M4PylIA!ISzk`6f
zFF_l$)U@k9Bj00lYlDfSfbZ0%7J<Y=gf%K5@pc{H#P|IKQ%COO#ozh`#Au0GQPG}q
zX#u8EJN&OKO|eKs=wWSztiS{j*Os?<@%$Q#Bg?V7_L9dfe(Tpi#lqqm4?g-R)>agz
zAjH|Vyz=a`y#IqA;!96_f$K|4D6I*C5T&5iYSEdTblakol;cULl@hD$xI5r1YTk`e
z(jO$Obph3Y%YhBuX?ZkqBcie>TwdRJL2UEswv~NFQIKaj;JOB{n+%(s&cin^G9jSR
ztTQ`1&3z|Na^S#zp8fjM)EafBXJ<Hi<OoTUFdV%F<<{?kTdHKzkH!RWLp5GM(&92I
zML=F=2<s8yQEJ05H#AUW!#;VTkiKD}*&qxWctOn-I%D_CBu<bBhFe?IYYD3K$jY2D
zPcf=OM!t*h^E`y-6SkTtX`R2Mb!N?FS&<hhwPp)H2q?80k5Qzv60**6qE@d{tJlXN
z@&TT<L{-Vw+OiW_2(X?@eKFGMNxdXt^4LAx_X9u1)nES^c3U+TGI?am0tMC!kivI9
zg`VePmqH&u1usbufh4f*^-Al0-}AR_a%<G1ar}N>JarZ$C5={#rJL7TzIlUC2HgGL
zkFsm;9;!0qVgEz4{_X$4(#`A6P*jEt`a_<3;dxhB8tuCMS%t9z&|^*wgc4ewWLQ~>
zx6%8y_Vtc8W#Ozt-tEKD_~CXs)ItyhE`=gZM?3kiWvNhAK{=*8j7B2{!@-VdW&6<N
zmdl$(6ZQ=<-J&cR3<pe4PtmM5IB?(qSj#;p?q+Lii_@o1bNJX1KJvj2zvIHLFcp)Y
zhbJtR@laMeCx!7LNk|XRbD6Q)7?+KzUF24*Jz7Le)iTnoq#gxGUt_gE3g<tH2oTbc
z4Tfl?h-x*$sOGrL28t}lRu!YIb+TR;AuO%QNjg(A=w=I<)Lhqe94Do<c7AbE&}z3`
zK~T8QJ+Q9xCf_(o5MaWX2WmB*-OTwD|MlPH;qU#^j8gzWdKe+Gy29!b-vh7F<iJP%
zG?#w$-*H%uNIW0U;H*&UF=;iTp45@nQWc6TOY6LR_99he*tLI-xZb4JZex-OdPT*v
z&%Ho%vP0MuY+ks)Q)_GdwSV+ac=*xBxw){w%IYdpQ!ZLfN`dDIjvhNsb8?EC%d4*T
zv<B_D4&8zI8D71yjA0yR@D48bqN-T$Z2`Qi1>H6*kY*Vhn;U3d5hgKF6j7F9oJAg^
zwL)t%_DnP=Wyj)*3L%}5bsR?IenF)*`Zh<x&3c20)&$3o-^HOr2hql`XZP;gQX}48
zVP_0+;1dgj1WIdA*fICcCszuM&oD38>{nz(fKef#5KQ=%=@{mwYX}+e(koY(o113O
zo>@F2h=P#7^C<Fyp>QUSuKii!30I75%k+KUcN>*(3S4b8%0dvwSU;ewoK1#S8f6Vp
zEg_EMov145`)*?Zgwo(i6c)eI=#6#GKL0ek?z)GhGv_v83`p12@coF$7mR#K?cfRS
z`{5twnP2)>*f?aDHH4mz7Y0~k$D`2_-;eN<h*ISQVS~ku5nnoem2+#B=HC4@_MhO1
zuRTG&u|cb-SUPu}m%j9Q?)~t0k@j-F`uQ)ibYq!Z<*aRXDJo4-xR?@wpf%B=*=k`-
zNrfd-njr9*=(Jg0Tz4l2>G)j2z`IFi{)2$lW~@Gfx^0r@8A%emv<^?w?<u0totm~c
z#a^#RRVm~+@}<;jCncb&)Xrw~t%^D)uF#x6e~t(5zYmM$`t|GV-n$pAv>Q=1>hHL)
zyJ;&Pkho$&IHPW31FUt2B8{dVg-pr}ON1(6VKqZ|Fe(L}3222C%|^j=2M}_sCf3+7
zMZ@AO7qqnmVMwW+_{M1EIAUYsFL(rDOnt(m$}>z^VTyukloAFZaV^A3fmM#<rpj`s
zlNJU6Nxi;vNU+9W(ZpI~wLy%_?(O$nU`Ttbm{OBOO&a^}p={4G+_*(1Jgyh8+3#_*
z8KS(BD0WKDa_p%VCNb4$h#mGI3bAn=BPGilU7maGCRekHnY$10^wX#4XNJG__x>U8
z{k|Ur;dqrvy-98iMhN0s%*@<w_8mOJ<qNM+sfr-hJo&Y+ven%r%`%2sG23h~-)wSY
zc>^DpWFW0YS_p(Ckb?5<@`ZTA&%Mi=lNkStb(zcC8m4WEgVGhl;Q*}^LEz{c`8bEW
zsw(GE>U+d-OqS=kq>0xBN^ew_zwy<qt*?)5F>cb`+Tw?P=!e+5cMmT<_dLCBm#=)~
zE1Wrd?j65ADI^W)>{Tl3qpS~BB9vfbv!KzG1ir#EAxc{cRWi(TvZ};036pqq>K>!i
zFuPY!S&J_;7-w4{g;R@$L4Z}7d^Ex+jpqd}*j-hqY{+nRm9js?jCo^WTq9|<P)6^(
zkK0^BKXx<+VJyNJkP_by@r?Vl#I}!zu#R&TM0lN$hngDtD`;PWlvwM+2g^|pRTVDp
z7k3Zr%pJhnf0FC3Ji+m)2CCGAL4<BL7?wjyqp3x8lnQv};v#1@10MXygWS?1%27$N
zwMt&xq$n%AdV`}!@8bH4r?JX1==ISS_U+$KoFoiK0S`R%FnjhLz>dv!JDoO@)0140
zlCCOtCKA3i9Ev02IK<@H4mb9l;18mpdn2^5<at3~^q35NtPKzXvOL>aU||>%N_S1T
zCX%Mp=@5qDII+cvY_|G6lzNN5uYfQNIB;MeAO7Hn*fTfJ+S(dPqt0)A=0Cj!^XAua
zTZNtHNkU->C19OmM*~K9C@n}cg)#xdG^0P#2%QkCfTqz*#DXNMsMqVPuJ%}28Q?`8
zb9*KTrIRH~PY{QZ+epSP>e3j@*ek{#=dKo2g(^~1QD7|yKVaAXeTcvrs2Zhr7F7_3
zctL=Wp4-?2j{l%tD6fzKS}TmO#As#_+&KLr*I&7xyB_;6+EjG6Hpq*T>FGJ5q>g6|
zQ6wo;NNvwSPJGXgaqhQ&jb^=cd=&)3Kvk9`QA|;m1oenamEe1vyS2{qx3=iYgdh0e
z`#HU~N@LG{rdn+lPhW!G7XSJm{C(n?IUarNA+FxK$^DOii2cXi*H>@cpg$Z@RV4zS
zdaX{UGeM~e;(DDd?;@j+**$wWx3uok6O1MB+!#~5%ci^UI)wX=kb<NZ6NNF>TJqev
z;K#N44i9U4QMqxghtWR1?^CPQNYjk6Du37b=yyM#B@BHU^*U2ilU%!Yh26V%bN%{t
zjvYTnsVkx=e#bAXh(p?YcVm{8V60Gfoe)W{4->7DTEk~55tzb&i5M0hnKmpJ8b!sy
zIAp4)s4Ph19O$*G5J7-an$Y)YHe0N(trCSH!b+6E8OvJS#$nj00zKg);)L92Dpe8J
zYc%2r-w#}{x-nFFv13vjCyBE^@tk#r^nFaF5JDrhLYKwPwYA;VDTkYU{x?2JEot%L
zkNycvC}|u$&e_FluGrdp1i#4W4ObzKuu|ZK0ren2;MP%DZ_pd{>G&;F=FwLkzxmun
zo>?C8;UE7YL|rnmv<}6PBlnynY)_I_3Z*R-0;vn)z^7CdzLf0QcZh3OT%l3qIiLCM
z&#-IPER|NsJ-Zn72h27aECn8AsR*#+fOb%xaO2~5d`0}hKnmBPu5?8sXpWWZb&@1z
zVxoz}b382N3g;bfEzjHW1}n=7rM3@F+ZGMqy7ORh0w7~F?RJa%@4uIQ`}WZ9bs6*r
ztZuBaxV+>tH{T&9lbA$nG6P0BUbn@=+Jx?=hx9V`?N`(yC{)E@n6t4}vEGYOHexa^
znTRE$o?$rLVtGAce%CyKAL0v%Fyk03ArKNbD$i1+b=@~%4OAs2EwJS{lQvDs(~P*)
z!f1mXcW9&#c%j4(Kv<7TxO`jbdw57hr72bEc+18xDl3M0iEH2+j8F`|{#mZvb&_Oi
znk>s%Us>hGm8(p4Iz0A)_YrzEg7JMX?o4s5pK+`e5GQTd#VoDk@(B<~Dr32}nIiWe
zAm261p7~wuo}Hs!Z(^|Yx|<9)H@WoUD-0KwNGm9OL18VEQ&ZHF8vQuo<KO)~2;n;g
zu#lu#iohUzNj@&$3{DwbX@k}V-#V(m9|mY+C<@2h4#P3;(h~?_sj7lbr_EG*ie9fr
zIi^MR`u%ZIky96|&2_YT>)krf^XRnO96NfLS{(Aq%cuD0$3DiFzxowkJay`}?l-qr
z*!hx3xEQg@I2D?;V1=a8f>Ex>OOGt8D3nCvlNd?h`!s`+r0z2QHa0S>^$9|cG#}yD
zJf!uAd?%1lT9X&f7sggOfpFw@t0ZOVd}E4HiYGl<Q<Kc?KS0oGkST@L?o=&2f#>;%
zz(;r<!b;F!q`L_TDF{6ut2L!j6lI1k6`>7iD#15C_bFaIcb-y}^an%AN;9`>FOR<e
zeeB-94=?bD0)bL7dyl<`E6={j>e^}Mnh9ax6rIjd(eYaXKS5L_#=^qJfK>%EN2i!R
zGSA-5UCcCFjBYIOt5^OFohqZdv`T+*jo*6hRnkCW<2rXe{1}fu@+e>X+E>P@3nhW?
z!lL)=+0CuRTQu8s)M$hiZUkn=Vu^CRnG5;OZ?1okus9BrvWBu$lx0DZB-CmNp63&W
z0j+k6My)}bW~6D#*47r%^XPOsEG{lG%CfOtibiWYwov)@&*@3eb$GRAtGmTlo_K=U
z={YK;xNz|zH*ekC8SlOm_;JjE4?f1tPk#zs*d0S^Ydp3#1)WJjJ<&8;nypcR)isJL
zpimZF83xiY(oj_%-F`vUU14s1!m0NHpUN17R^zCfh&&&=%LhJ6D@Rul0orP+qIBw3
zDF{LjubCif5uxx924onZj2k0VO5q1S^=1<<@OHX!p6@%7HXhm*L|U_X^9uQ3lejgB
zu!bV-^Xyl@z|~9Fx$EdrRF#8(Lyvuk*~3SWen`@oV6e8vE5GtHWMzf$EV|NTF092W
zO)Y6MDh->XKJ~C7kQ0=~X9L6G6L)d)!S`@w_aT1c|M$Q0fBu8N&wGF56CA$pB)j+T
z!x~M0V~uNz3tYK$87&gFHa5VI@V$Vgg(cS4*SP<|dug;=Y_6}fymX81$`(l!ke3w|
zZk#NHq%v+=@tq258ISlRVHCI#Y@T6-U}kofiDrj|o41^&A#WB(1paoyg!9<k?yme^
zWb|zw5dx~RVs3tpr=EI}DzDh=c3In4XMKI0X0!Q@Ut1DcRA$MAhi?k7<5(;y!31os
z)qsqB2P{S_Hn$Y38#OR7^+2(^tq|73lM-VpZr!}WXf&YR?$BvZ(o7~08ds!@pR-aG
za(sUXJc%w#7h`BGp%-Aai@ifitkQT0$FJ}}20n<`ah-)lOZOfdrMNNZae85u)$WMR
zQNi_%Asd;(;Nm$G-=p2UMD5BV`yc)Q)ARf3w5OS$-$!?Cg~jXF*njAlyT5rZo7HNC
zAA|@As&uIrk)OCA0R$LBzcNhDw|VR%AK}QJgV;(t3X#%8GU6-0{-5ZceuXSGjFcN^
zZ>_D9*K5qr&T#+z_pobzj!G++jqXX-R#)(_B#oGCm^%MJ3lpt6J*6r0!ZmRIfV*w)
z2DY#IN*OlRw}@*A?RJMo-EsFDjXJe@jUbAsHyQ+EW8z9HR@c{9US6i#?e0(--YV51
z3Pb9(8c7`UE1&!%d-v|=(v{16^=n@n=bXQF+*V;%o1esAU&bOx2^a~F*l4sh<W}HS
zlDrHkGC@`O7~v5IhS;~XeM7w#GpZ#0VL%*LV0)xl4KHZWX*AFpq=CryQJF=S6_rvH
z)}T=o!x6<WB@mK&qsipnJ+$`jLWe%1v0y=9z_WOP#2@<%VI50ONr4syEd(B5N<+6l
z<l^cQ*VngL&kC}#B8o$*<|Wo%c!}Ld?qT<VyO^7wrOHxXe&$((E}7fC58(;Iz(>~-
zjy(Q8zVffW&PR`i#Ck{^HIQKg4~wow6e$EkV2h0Ia6sP*?%6ZPgAYDTt20S}WZiF~
z!UR3$1zKe(N=4?ug|inJl_hDK@u9~br;&spEmm3x19r{KvbMZP5K6qrk)zi)H$j5c
zmK-;B2#k6cn&E%2aAGq7Uf?<|N|R<8olXZo^w37r8x9!shvY@hD9=cXoXw35wzjs&
zit<f-AoF{O9)ytN%uY$I)}SgYR@c_CNYXTCW2?*i-~R#H7p}0fw)T!*mlp=KjvgS{
zJx}k-0;Q0IR$`Ilxkqnch-;E4vBY6cn)zs4w>gmlEj;RBNwXGl?fN=e7{WRT88Fd;
zMpS2%j!@%hwWL;~mL!zc6=zkJVMAxbQx!Q@E40y6*$AsE0zae*LZW69Uq;T3LA%`8
zC=8K-hjqIs3NX@OwdLaS8c$tbV11M`bz_xtYimsHJB(6__0=T|0U!9T@55FZ+8C#5
zP9|vVX^s_Wf~mO0h#I48jYg8-N6C0=l?Y=nWllswX<<}kwBm%5b1lAjZG|uY&R044
zz#|-d;1PD;`yeYXKF`lQ{XEg+1cy!>=lH!RxP19Cm#<u*muCFXPyR*P)6;~t2DMs)
zbaa*ZJ-dnGm_w73bQ(=oSC$+%b==i5<9`clsJ??2kHvLGta8aZN;_WU!om_;n_X^P
zU*O{(`#5nNGa8Ku!w_Q(Ns?g3<NI2zju0CNVF5Ra{XL&^JiawXP>Un9(tQ2vPh+&<
z{`((rq39D6l+`=;gG~19CcSWzQg~F-SvXl-{A46Eq>b2EOYp*!sYw_NJ*0_gh9$ET
zIg@RVr6obDVF)~l6cxR053L{wV(MWXp#e`&jXRbCfzg$-Em4BeW*;deo!xV^cJ0E8
zed2oEh1sL9z6D<(d>`v27!r)guoQ&CLtBr9Ea$6N7I^a9b(Z>y(t>s5G=%5MN;R}}
zO3k5u#(YunC%^XxnBBFTVXw>P@*=kut}!*ci!f@SD?!p|5(U09=B-@92)e|V5`>2`
zhD4OiH$#>-d(2JkVwjaQ8!Z}*233*s<d?q0tyf+nsRX0aW$yZU!PCF?NoMap!QrDv
zxa-7mk~kjc1B1>})|Z!<o|>XwuQ95n6vLd@^O)W}#|u|)jgLig%uD;uT|nE3=El19
zo)#5bTSJms17nB8ag30X{MId!c*{*TjG<B$SFc^;(xppp$S!~D5JFI|)tR54AA3@!
zT)KRT)<m1FZufQt^0!ym9eB^ZtUZ^JZVfQXAdPmy23&DBN-IVqMV>2s8{=W|e5i$n
zMg&nH=nhJ(@$jVuo3pmIOm8rxdHf!H>5~>2!f3i%o7CcnX4s&+wMlO<L|2N*cAI9S
zNqg5WqRBQU4ydeh#>rx=><By%2?l3o3K}hPj4cV3hpjwrZ1s8J+BLp$eu10o8JP+x
zwTICa(@Dw9z5su3o87zj@?)R)aat3TEM2?8*2W5#FI^(3w`epw?zHPk4&HS)pE4n<
z!<=?I8FQa1XP{h`Na0K@H~T}j2JS>NGc!l6)}q@R5lD|@qD_1EJ~mz%xhOPUVXeE^
z)^9GdwY0?9GpBj@(MOn{pGS&u$dNTP8x7*f6)15MQ;r7o*SGT;$I6^BAM!hQSvhQ*
zK>EHj>;*xZWsK64#ieCNqZCgDOifLZrYRd68;<xr80_%wzBxYZ+rb!T(m5KXbUK|K
z<JOgx6|^?oT3F!02Oi`{fAmNBCqMm9-_h$jv~NGbM2qI(BdlLubeXIODl43dZB#K?
zMOX{9&{7mR+Bo00nr~?au)8fV0yZ{scI_2xZmm<4ie}?(R8>$E8F`kXD<>ujqX<+*
zURI=bfE@JDmCLHC(v)J9)0phg*s~9jMDG5fEm~=$gdh$PQjXQ9PDG?Ij7AkNU0LBP
z7jJQ8EkifKivliPUPUf_j<8uHjuRTK4v&50dpLaPFv|;9DT;zHOz?x({qisxV<deM
zkd_7$I|VJmBD?@k`UIgw%NnW#D=U0qm<TK#VR`YH7g)J*i+dk`oR9tFU*O-KK94^d
zk}q%a+WOad{^_qX3IgU19pkb0e~h~yd7Mk<&Y_V6e!$GkEEAnJS)OB5MVbylIMYyF
zz23Rno?w5c!%f^J!V56QvT$pWt*xHBo^RdawKHcpao1f$<CKaxNl;p|xwT1eIAE*a
z$K_*gk7|GSbG$A}a_g*COUSc~J@a#fVMMK#(CZD@>ULRNT&Aij=H}*ES>Jd^uWkOZ
z$GEYwLA{$&;A4zHT8j~q%=3sT5T?%ZvLH?*XpfqwnX0Ain)KKjfXaQM=@QVyk%y2z
zzUR^av{9gpGXl59b=<I2MS(6Wdc%|?PS}0qFv-j`D)dlVk!g)p3hB$SuZKj65NSg^
zE6_qAtxu{vZm#ut?(}7zy|BPiYAB^oGip+=`Bbvc{DBU85AEmL)pa%&daP{pI99=>
zi<e2$f-9FUGVE@0aNj%&w{CLe*gXj85jGm`;Dr+#N+IzDXkB5XhZd4B5PbOHH2?hN
zE1Y@t3@6|72(@+xjl~l&)APGg#)+APg_e{YIkcPIiQ>}f7dZB#pCAZhH%j&da_r06
zo@mkC+T^~I_wp*I>8^M2vx;7xlNALPK{ifD`A){`k{}0TESsBK<FuBTJkKdBMd16K
zJb52;b8}d0SzKJ?;>C-UW%;(AHwXd{f^+B2Gdn#s?kKy*J#y$Mubp{~vT{ilZw0qi
z*zLOKZoH!4+H<F{y_`sESG7}NP^c0H1A~uGfItdO<Qrlyrx_c3DzpMmB+hrv^C+}t
zIP6js8BwE+z^Av>#|lfc)r7nv@<JMO^Um5NiU^{Rpw+}gAxc=#+oK^D>xK03Y=pEv
z2!#?kQW%U5xz#Or`qc|OdG;z-2N{+0@uW}cmxQf|a%jjCNw=z)wvyLQpXGaoDN9SM
zy!h<bx&P$d_`b*T@)C#Z6R*eh#5Gz6kFc<Io`Vw!p|xO~;tZtQNDbUtAJR>o!E3$N
zpxKz9y{qoRtyRU5d+z1)-*}ZNtC)>_Y7;GV9HOFtFMRTsnLT`viK!{4;?fr3gHaWp
z5Y&>GN|jg~uPRG(gxMysk5BV_=NER{erElCpW$e@6K7PEC1p|Wyq-6i*nW#*?ya`K
zx4XK{X2TK5(~&DGJs&K*cIFH}_G6!T$FJ+2W5=*aYV*6Xen3!FXlcmC2a}P4wDKuS
zx2U(N)q#LUte9z4%(f$z79mMIYH>+Atf)5|cvi4{>n3}4?V{7DGZ?1y2Loz(ohS^U
zD4cbSsi0JpqnuW|&7ONsAUh3|b+#VDg7FO^ik-MfjBPFekW^?c-Wc+$uPib5$Va*R
z!1u86#V_*QwHGL?VKS`KR6gnYDoMS@z3<tN@M^sD(kTL2;wLo@96F9PKB{s?x>cSt
z=nq-GeuKuo1fd0paqe?fp`;x1gEZ1e62DGwQ1FebtMs+A7g%3jqR3MA9Y03nz!6r~
zPl16#8!}^=K6ISF^tb*SYRv}OC}nA3fjlqB^OSx2_HxgO6BOx)jg3_<UA@f4#wKN!
zkA3yVr}%e#1^nlOF|ISaxUfuBy)H6RN^|=3X*|zkW@d)6D7kj?#+$}R&bs0EUPEu>
z<+WNZCv>XSc<4P3VvNT39M^Sed6`C|!Tt9=$hGU&-|_2e967|HkA0ZyU;GmOAVVQB
zT41$CX^HUw45L!g9|=5bY4`<=xI#!lk$X<uV;$wg^9%@uRs}_#VQAyQ*dodmAyrwD
zXF0YqOz+)GW2!^go}iSV$_kWr9*4fOTd`Jvkx1zyG)OXxU$VAgc<FqPXKq~Q+`<~Y
zLf|D+9Gu?Gd~FKDh;tifc;(_2=dNYMG9i>LDwQ+QnZX)(_0)@OZETQ@226Jvcsh28
zJ>z{yzv&(*3BnlP8W7IAuv8U?X4~8oF7k8#>Sy>n|7ai6lT(PX3jO}a9%Jh0Q8v$9
z#G_(QtH%3}?PYziMYG>$SF`Snnzh*x)p(xZzWeTFd1;BMi4Obr?!{v0-dZPL-$Yo8
zK~mtv6yK>6qc?i8z7a}k`u!n;!Eon$ath9#J+qUceE$6Tw_VhMOPLVZv3F-cr_*6D
z9C7K=B|iS~@8*$*9^tv?pXa4lUwUhgcRNomCMKAB@IkU%aq}Ce2vi9oK$#LFEP{mL
z(9($m0;vh4r5-Ahup$cFnb#Wdg~S`nylZhxR;saoQo+XRDpo1xW@a$DVq<Ncndw>L
zMvdHP3ISdiQjJxfS}UwFc+w+~j@Ree08|B{#CR1(6b$=5r!RH+o%2^X-!G`m-p$Fo
zb}>Oj+FNERJIA5>=TSCb?N-LZ%{2yBZ}CHqew62*eGY%DAiHqkJbn^V8Oeb|<M7|2
z1f^)q&a-;qMfBd8vFo?H@wv1P%c!WhwUv=-$A?I}1G*cV%+?!dYcW=H?Bof)5c=G|
zdxFP~?ZJdK*3uCduis*RYLeNbhbW9<Du|KeNC}ArYw)BX&r8<Uw@8dLt#b;w?XKM)
zuFXl0Nw-z`<`K^Jrn2n@{;mG*b`V7olap=c=H}Q~U#DKLlNTk?IBRizeVwaUufDw(
z@dg~;e*mOMbAFDn-J;4is8DFDD2-*KSJA2~$Sh@U@l1p#1hHSzsb}n-7Iga>TO_om
zOX4sl2qNmW2K|jq>Olxfp{$`^Psq!H_00|Hag0PDlp-At34@3|cON5~o<e(WR2_Og
zNQu-EX)CPsF-TYG;kxpJBICu2oBZmjHTK^BUhX^jFrWG2Z}O$9-(V}R@Jx&@dUztC
zGLmAbsE9at<QSNW%jYk$y0JmKGfks0;e-fdRpzw|7pc~lX&gLG6hu^3;o&(_KA<#s
zra+g9ODlaob@~P`4>h~*KgkE)|2T1@&TszauMzly&g>-h*_dotac$^KY^SDX@dW(J
z&;2|%ZY?l9+2X{}BjiO&RIBsB4}F9uKKolNE-!NS{8jYOvFe<6;dgeI;oHHf!GCwh
zZ8#kA?6c20xBRjA=<OJ~im_qu)MSTtdx8h<yO-&yDXw0<O26MH3?s@?@#rIu^Z74+
zf&Q@n`#<RKgU>wqBtQDmj}j&^*S_%r{q-J2Z-8HELJJzn$S6vaQ)U8RXlhAG7+SQ^
zM6o383q}La<tj%S85r^+C0|{l*{U(!m?E%l8dB=YrAb)H<oq1*R0k0SXk(mO7MJn7
z&1rP$OTzgc8W386SF+kG`0TUGJiX9m!w-pPkMZ#C15CH3XhsbN8-2cZ<|$rXFHu$y
zi;|jG(Tr21?<NRkG34sSa}-5Rqh908xig%6&%;g(Wi&UgT%i$;P5i5JtfckF^a%kH
z{6OFbCC6tbc;(zR{^{TQhy385{z2aR;SV#te=l*f$tV8$U*(_u_x}T4S#Wf3o0=W5
zytqOn1LAbVxy^N^Ta!%44l42qrF0`oEd5@WgZuWeu)IXvYBJd969}hdwFuOAWJ%}n
z`|TapzWzNzK)>JTrI%jB82c^vqPGfx<6R0Nn4D}gJ3GUHg9kW&{vrU+zwjJ??kD~N
z|KmUY$NZ@u{{FZ0n738fRl;ND$T310vbC^8cH@>i;oAyjES{)Q<QnB$Jg>s{3{hxF
z5`*VEV?rS<Lb!+(VFiH~K@@{BtS_&yxwgvO^fW5Z$kGu}65}T^MoO2}7y4-FNT5<_
zs5BA?0^b!;%6G~natwx%3b}Br<he69IDM<n3IRXx7yk-B`Ct77UisQHeDyP*=i-fP
z+~`)k`nfX*;Zu`!I?X1%{w9k{E6g7}#@%<H;MH`%`LnOGu)4%|efRgGD&<r^RY{o_
z2rnedN_<?>L}i`bN~N6<a+WJr`Z<O538Z1PwZX#G%gE4YSY+hs06+3LeBwAARtW5X
zYqwUou-K*EsPUeohqzW|oVjtG=|?(Ld5H}Kf%H*TfzjHTrz%5?htYpnx1|4k5W<;D
zE-fx`^X4t4I&JQ`=LD~wJ`1>0@c#Y#+1PkzBP4)JSFdt#-#((rN%lYXFqc061hQ0E
zQDTIo$Rk!(6w~z(VGXf_$S<kK8MBjs@JNRstfI;-=|~aRU38Jq1{ruxO`cWMCnlKc
zOps<d!U{YCsxUO06LjY12q!1pxJX%qCn;@-s0yt1D6K`ykfJp7(~`A*$@SHg^D89_
zdC1(+_t2i0;E5-Go6r8vXBlNBR#>c6DD6`!k5;G6vAYjZW*a>F<mVA0;L7C-6uM$=
z&wk$X(4)?qa%>6o%GaOdP&0I!fOf@{HqNA6TZ1wdTV$NMG2oX^T}K@}%>V1}{S{7}
zyoa;r&hRt;;pgc0y2P>1?ui*5+CRnNR)eRmZgPI3i(OshbHDf>>2<pt+&jm^k3K?D
zj|l<~V-@Z81aTa(xU`BMRfIy)RSLx$x(nZ_t%lzN-)^r`O1<L@WNXJ8nt;b1eUx1@
z(_FiDjmwuWv$?rNXL1rL1D01-=nn_)=()D>E5GrZ_|oG?KmIZ9+q<8Y*Uq!>+G+Z?
zmM99%Q0J5;r0)BOGN%@6lDhN!1fP1tGVGO%`ZcOb(rhW((~4Rw7-R*Di`PjG#<b!I
z0wK_4Nr)h9G^w}Sh$uq(5?#1)obaT}UzXdxG8Us<B8rd!o{$(Rc<pMRr*3rF^S%%B
z@UFes%^`j+=%+)@tzPEz&9j`_ypFPxhHO#yn}lA7^n8*?ar4qS{_W+LSz25}DMJ_s
zJoNbEeE0YN2*X~Nh09ks^E>~EkIy#Ir87IX(iNFb5o^&}P-(D6VXYt$K1)|`^0m)?
zj%E^ZU+B@AndRO`9^%jcKmP_l|G)e*yik&71Nx(ashOBC%b5y8Ub%FJhrRo$O}qHM
z<)tOgo_URfhYvBzQ}R4zm}Rc%VqNp$UG&EPFOKaZuPm!?d3d@VY-h)(sp8&yPXKl&
z_uTV5W@mOW91gjD<HlQl%-bsL;-pU0?jZMSjy>`a7qStf<=!~9$pv$R#285u3hIfZ
z78}Aqy6|Kn2m+8oqm`yAH9-VXkl-6Z*6k8YNh3}uvI1Qx8k235FLs>3h4csmkE&8G
zY+85->7rj01|tfjv=mjvW~zDmwIyD<TyS|*K)b`oKlpJTIq?9`fAy<;`m?{z=fCzj
z)`uf1Ex|-YLeuj4SexVd0Z)DHDIR#_F_I+W(xr<$@#QaLCA{wgA94Aer5kZoMai(&
zLs$^nqA}QUY!J9H;npx^b5LTfAha4k9dY^eX@2YTzr$8Kq^dGTg8_HV9AN$CEiNyv
zaeb@AgfZ`Z^Z|4_BAJ?F*X%4`d-6%%|JY-6;@t>oNK+SJ9tEMZv5<mDND8B0-<-sF
zXZl07IsHMy_k0RfvAn#*@nc75H0rFcZ{qs_-CmEXEN{z0=-c6wzx^4G9X`Z?{rfrg
z-5(=}LasdX9O<oF6vkj|jVy<e$Vhxk-AidDhFW3>qJTka9PPkb(qW9UIdL(>ph#*n
zq-jc{UPDTcY&gXC1DDcZB)%8W>P+I*V*F+U>H7!;N<mqqRNah?JZCw}*zCf^l_9G|
zz(6ES>^{KE-uu{nw8d08V$fUV@~w-UJ%5H#K5|`PD+q*7E7?W87PD(EW_@#o*<G_-
zICq*G*B2P{M}&=-#f4iu_?}15T9fA)y{%1dy?TnDI68wb1eMkpVKG)=(FkJ*HEfj?
z&t6%l-<sqf{)4~A{^Q3ub?OCv{uh6Nw6}>Y3(8WkX#<w)P4<^teBjs&2j^Q{zqN>6
zF#M$-|MM_AL+VLtwK{Q9M@Wx69ntIc=yyj<AjpiNLcFPK35>nQ-tndIM+RDJHaE6-
z_StW6@4Y9Hz~<(rOAJwp!1uWCzWaFor5DJHcdjhA*7CV0p5S{v{89EDJAw29bSM}N
z2Mm@rDU3%X16n~vECihi$#f^cDnV6Agjdk)7<y}xBCR<mgdn9e4V^}vqN-S1U#7OV
zfvF5tQ4sh(wMG*W1_)20Ohr{Hyvjo0VTFs^^Fx1ZCk0~cV<!dV8m_M93}ncU{KTK5
ze`}R%FTBQcm!9VvH&3xzY>{h))q;ow&o<CyLLnq&Zcvp(6$wga#7V$tm{UvY{K$`d
zf+|lr^~|$e`|PLq=yc3XkhsIDad&+o-0xzH^P3bF2BqTqYKCuMcVO^V7rAlnJSvXZ
zcl;Ph5_910BmBv~_-DEG#AoTG1GkS^L%OlXgO5DUS1zCD{LRbU7oEUw*C@(@cBhT-
zBxf#MWMy>=ooZ^r$45J{BD^a;cHe1ur@Y(k15yxpv?iKtY_2miGsCT03)E@}QVI?q
zImBl__uFszF>kA|i^3XG2ACjb^4Kw^ds_^;PeJPA2@em2FDybwNKfN=1wjbDOtAPU
zJs^|@PZ+FjFdSxtjf_TAM-Ng+N7N@8ctM0$YtgQU*aW;N#uqV0S!c96&etoll0Zf-
z(-MXD6{1W@N1BTZIZwQ@#FbIRstuVraE#+e4&qD8Z+`k0`Sce)!|I^R$SBO%xHkks
z>0JGTu!b)cwLtRB6QAKrpZg@YZY@)01#vBA&#rl-fU3;NvjOQ~$l~=Y%te6<TgNy-
zhgAgH(iN7QgPbfAbgbcM%SYV2M1O7{r_WqwP^3&v)@U{B$T;BIFk-z@l*;n)hxRj%
zWl#?Bl8E>0I!t$NFQ?9*=7YOt@nernXJl2z%)}I{c7;$V5+ks}qqGJiw)H-5Xyg6S
zT3>Gq66tJX0-uG&MHZG8snrr@W~Pavh%;x-u+`ln&)=CjF~C-@%fI>gpXcxVjlW52
z{~?Zk&-c<eyq^nCew7=iE}(iP(n?h2<3$nmMoz0P5MGFufKmwKB=lMds=}pJ3}it)
zP6=hmFwLmNA?avDz1hTX)QFM>z84?@AM3feA6+VRsmaLbuC1}YUNP(|ZrPHXOIuvp
zl=yq^<$HeQ6U@y_bMxwDo_pmv&R#jq;>sdfky9B>>3sk3tWV^%QDqHP`GjEuJ@%2y
z^Ax2@!YDxHCH>7!FbdC;RH~%6xkgatOgEBoyitiAtGAuivBX%zW}#T<73{v}09IH&
z`N?16iQoAgX}3qc78A5<NKYcQ&ubegVNh}~yvFslidwV5LnjX)mafv?Jx@=SOi#~(
z^jTe8;-zmqLx0ex9HxX=az`8)J9>Xp-}{f=-t@MiwNtELy?%r1H*WI554@KH`}fo9
z^@*aGUboM|0|&Y1#67%t>cw~EIs1bFODk*4&do6hG~wPErtd#aum2KJU(pnbnuJjL
zOtb<#ucXoy##*YfqS5dWw$HF1q4FC2&46aJAWT38b#!Ghs-!Fx#wudr5kxWCf+q}F
znPZhe1s04;g^?nlH8F*sMEKIfw;moohMaCzvDPzGK-%r|=!f3RxpOb^^x|_Yq+N^=
z&Vt2;)PpH%avGrnz+g><E-KcSbLJ0D(`dGsXt(*nANvXR?m58APd>qwPyaH1>cAvT
z-$z@EwHkxjF$>?0u>+0J73Y_Gj55VT2Vy$Y6KswQ`5@zGfA*JX&9w2yS=>dQF`abi
zq?wDOls-j&6R$tyXlII7*Omy4$KiX9GdVLu5QX$dLvAgukPa0!Jn#g*C-Ak!Sc}?r
z`2AJ~q(8Rn$`#<o7}8OS*Qlc^&DK`eMU-kM=z8`W&%8~YRaBc>w}y8&(gq4NxNO|5
zK%u2*in|tfio08(P}~a?*HD6#LU1Ws+#$swXo5p<mz0xz#`w?qzcDYqi?!yuTx-lZ
z-{;*@XovQ^d~Ry*9x46Cs+91!tj#+faa$2VB`Z%Ns1B-Syfo^QHGaQM7NGU5ff4X`
zkD?|iUdcBOQW6rn70V>9<QZFA!%;!_Wfljie~C0HRfLrW!<yCisNI><eNQoLmsl=(
z`8b+hpm~dbzFVF-q!PxQ<0ARcokSc-to!_}*X?r+iEC1)UOJJNXNpFH<MGz{22C&6
zEQKQEB`k_2ZGZC(RRB|I+3`b85Z7m2GD#m2tc57GibWi%N0I%b;@;k#&LrzE%@DS;
zn+_9jBo|2v48l$X6qa{ZRtgNsg*bwQDO@!PpDr62+U!lKg&~qxdfXeR_`wk|@2rX$
zW2B`)>qC<e)d^M0#1anW5li(s`rp|(vN&`yCHBg(?9V$JIh@Rn#g`2=#ZIugA09Az
z?UgRd-l?fq3WW5f+N{CrNml7X1MOnr5u?IY=Y8)rPE)-CpBbxpmvy7==}H^j3M$Nl
z2|F4h%Xr}t9IU4s#HLi<y1*}jAStUz4->jWEo=S}`58FL#oec`_Ho6t-&!hUbR82;
zYCK||vvelA>J*cvPevG}Ry$A8avFI=x6!EK>3X&bBQY5A7Z~O%1qcm7%c$pz^uM#w
zne&-$aDT2IuklV+#&?(H9^hcXB{z4i2&d)G=px$lhFb^7Wu%*Qx~E{83m&MW!0+?y
z&(NQ3B|KIL_m(d?^An?J?eaNG_FOJw6H1y|FlLy!kMZ@wRmi=tSap(v(EgS`gDDC>
zxXINNjmwIEAdfoE$5;S10v0v?@s*2AF;*|zeRJ%<!Eu-4J`<)tRkOLp0gYV{F0$-K
z%2Qlk_!Y&P>-d9!iHe+AVLE1xYdBlb?xySMszNmggKYYy(~r;B*+qn2Cpz8V(=VS{
zQd|5)o371M$>Zf+6Ma+@pAvRe+uKmh_gk!4@v{_@CY<Uoe8dA!I!;1T4IK%u%x;-8
zlqPQ3PDm$)pE;ors8Qco7d0DKeigxu6i+S?tH~^t=2S5iij8Dhq4e;nAHDGIM;$2@
z*@kI^)x}kg1l%HV<4p;qdTuk#`-_=7@0RBc4z2~{*A2n%W*p?&c3A?I9^QmfAeE$2
zt~p0|aV7fAmJppJgE#-qy_H@cA(zi#UZ=y=r#B|4QG7L?V^)Jc^La#Jv<fAFh$d^f
z31}o8QGS-SXQmvEWC|4ck7P_oOV<FLjSertc%=kXqzNS0@~By~p3=K`3GNkZPYOI1
z4gLK!vC*g0);23I|AUK5zIxmEfDw_UNof#mY*&d6U_nXd*3@Y$X|Z(?LpsYD7Y#6T
z-#fFR3tA6O(^DVQ-}J{=dq#S>)2wD=ssKuy93yGLtL)tJS$aKHPmkE<bB=CvSt30M
z4;6LD>N*9wl6WjN8O2|4brH!p@=CQJ-tbfed6#RJTH9U+yH?udiF!reA20QWwised
z;=%wz{6p_F-Q*Rwrj`GWm{?pevEw9JHd4Gr_OqJplViURT=;Ym9Q;5M@}>3p+rTS1
zNpzpxPtS9_>>nHv_-Pq)z0FU)IZr+-19HhP#2D$YlS*>qb0!Wve|k-e=t77b1;WtF
z*F#n_jy`&s`a{v2j%l_2f0#UL`a8?Rt7^!^p5Rep7u7Y$y!CjbBSe9jcu6fS0}wPC
zo>zC|5GJc1g54>-PYq0s>}uPTmPi_$RR@U_U7{`{;ytPTcp(f7qYdwE6JL#75*P(u
zLzt6HIGLQ5*;rYOXl9yona~2Jn$qu}<nCt$R@M+8K0Y@OA5Xp8wOMf*2Ac^juz;3w
z)M2h9Xk+^g+#wjL5d2)%)*2!iSvLl5Pl|<O+ezND&YW?{tLU{oDdAX<sHiPr`7s#5
zESEvP(KA6q;rEq?cX*$(mi}uA)ggMi@=DvRt+Tr)QFps+fBA8p5y>L!NCXo#&h+Mn
zP;=&l4*pprv=n2=WHs;5I_+X6^4g!>3E01wE`}!-@mgS0(yRkhe-|Qr>?5Wl%yL=s
z7-EC&4kUQOtj)CDl+^!)M5{-eg$(uMQ`}@eBEjc~<2y>wNiS$%b4>hD;+`r^1j9j;
zE#y><ogLupDk)^+0o|OqK16Mobn{k~N*D=}_)Cf%sNwI$z?EAb)=S?pm-jP@!eX&x
zL{QrD0+7jM!7}shQS*?vt#qFeOC?L_h7M-@jB>~PJSs@XGC)RkU)WT0Z5#IZ44D|`
z{iH8tb10=FPb@IIvC+rYZ=4?lg5BNWs;Eqe9ZxFIZ5_aZ(53-@D9k_MI4O%8*M(e*
zvngeXd`v#tVpVP7y?L?$z=KYvcNt}-Rwi~|vo`kGZE`A;X>t*{@M@(ooaJ$;W#{F~
z=#mYOv6xBx1>N;gY3)rjBG&r;wF_6Le0$q*#4TTJ{EvF>#$f4!@S-fI|K1d?hACAo
z*@XM`@V(MBBlr$#jz|}*E=pM9mYK8=6_I9N)Q~7B5}ahi!(1Vy+A>Xd!YNX6n=mf~
zip%_VYJMTmn*x0xY|t5;IDGHBaUUg}TYD?{jI(Bd33brYIAze<!5D2ThRR9BNEn9e
z@u+0;%x}bhEDhqxSQW{?2$&dEGiXY)5rqEqtI$xpq)G^KOl6;J#GTG}I1^`A<6gZ#
zxxcv$xeoDMo9xXRtMT*#T5^5JOIjUr9620LS(#fD3u<G%RG?K?<6Ug5=)JgjSrGDv
zG?&KIu`x>~n$R7Al#7VyE>!5y3U9VA`51@<=xXc8M1}K4K_rRoQA@oZpl#18D$^E$
zE0xc2$<Dio%jeV6PxQz?YkX6Fc-x21J<_9DN~c6QX)o(cz{3YNWXuG>-8^sM<P{~h
zEg+Ty!di!6@0Fd?=r4l8AcfXhyPDRq@kTYXBeEcbeA4ABTgNNc(kps|8%B{1YTxN3
z2|5%6Z_J0FGev%R$1dA93oruTbTtdS)+(c$m{Rry#)vk4SFc|!%~mZOjwxoiIzX)j
zQXJoR4FYSeKiMQ+ZrMzYtNl14=litfBp&Qd<ORj1Pke>zLfXTTRx2AkREs4zraBL7
zO&an<IV$twXA`jadJ_NAf)p}2MvS5IyJ6fTU3SI11v`0<_aBV2H-e&q7&(9BSu7}2
zRZogGBo))8ma(?Hp6UIZ@nWIf+PaWBx5$mJ7?ct3;7q_TC^4KOQ*ZL?V>bHU7t=E?
zTw7a5W$Qa=G=<_qZNq)rdiVE7+O8fPj+}+6%-Y8I9tT2@@E&3>!TO-}O^LDP5DFC=
z^+{T7%I<AX#aD>wG^+HMCrqI{hhOIN1jraAZCt6owDw&($lFYurC03M3{O{^F>~^s
zL#ZL@jEFIop6l=4av4_`tU82)JLgao>YHHREGz4;6)xe`cDopI|0~roGO890*nk)-
zDopSLkGhGN=i*dI7e&}E=&mA(ewQDY{o~AF$4h_Lweb*p{7BMJD!M5*;D&~?-_BH{
znZW3a>Q}!4h1rh-KAR36AxE6d?dDejm6yX6X=C={;>8}c1E%%UltSfosxFrSCnrzH
z&%zv0s*L>hJhR^!j(txzK97%ClthzdNINm*1NEuecD8qkaGfukG$Tu>Te6y@rLB2@
z3F-IboI%Anth~zH#grAJzXDDl1PoQRl`()|44^_N*XL>s#!u~>yu+D7RH35cUCO3}
zq~ynNhLP|_?q$E@j$<Rs>qH9k^=yg)2>IZ{Ecix?kY2QFvm2`~Utf#6isTIk4l{{`
zuB5ucq^vYh>3ai_XqR$4J((<E2FXTB&5LaVl}qpQW^**KS};1U7w7OlL=wzV2>Y)e
zb>#iL=P=Z(X%g97{y332r@a~XEudUMg5y-c8(+%m7!bS;dthlRv%Og~znd5Ly9<c2
z`9l2nT0{dxLUsNNP+p*---)M+0N7PEAZ2vQLXyh+)oAE$05{WB5{9qe+2iOJ@B5FJ
z$<F5Wm{rNBq;LW3oZB>bV~oJDU~BLwXY0m$$+SM5?F6yAO0B|O!C}nIdp}gPd4uA?
zBKb`grQ?vm7b6WOj=khD9*@Y)<%Co+WbrRAqfW>4&9)v&xG7!UdY&*STpe%!SqXf5
zzbqKEUa%1~+icR&G!)^3R5DCd3m=FM6}DwS3mA(2!q=7#WMb2Jnt&^*Ky<jZyyzu<
z@jF}<rdr=wt4@jgk7!U#e}8}T!%xhxR60(;95ek@t!^a6fb8X){9czCjvdc8Oy-8{
z#FE*y3317jj`C9u>+yJ;WcHaY<ZNLdScc`}Sd_n`lJVmYOqMle1${a{2B&muXI4%{
zd>cGc5ARjje%oaC)py*^>lV`{h%v7C<eIQvF^c8-;&%>Q#mSKmbh{A}dXF({@v<}9
zD0}P&Z7uh(YMW@;LEjO-&}J@`ng*9{!-(UQ_B}?oxCR6oi=|9*Wfvv+z@nq)e4jAF
zZOp^!xTdLL%>EPB$`bOT9>#BCw2r0^$!Zl^P_`^d+7WY2zI+}`DIg(1plubYAeY}0
zz7kDT*VL-S#;wdbT-lAQ=}fjvZ96oy^De^!*2kj#@r%I`LX>Kb)joo_#}X_9m=Pa$
zu8`LWNb7y6wfk)(u9fdMbz11VkFM>{Uhd=mG4P@3TKpY!5>8*Jd+xtq&Q5i4FGK?@
z*qNkv{&`yPTF98-3!OxdAWZzeN}~Bim=Np+^amZXz9ne)<mX4HP&<wM4|S7W8LX{3
z{?F8Z7uQpH`s+VY2KGq?Gis%G*@sWYiO(t`Htxx^l8J|7X2%nzF!jcMtih|3uGsKh
zPenf$fC~=B8K)F0u1J<Esnla}Bmi*=AAeFGeOy9G7=W0iSjb2;FylipeaJzGE4>=u
z>ei>!&Hd~7i0+$sr}pC^ux05d(zR`8sJ_mC`{5m#Xui_F4eX6uy8`O~n^3nPbcOic
z%~SisoDKY6(>oXYxY@NP$5(GVB$2DbxFP7im-{Kv0GZ(1*#q&mjC%1g`{w9pu(DS<
zA5mTwO^J0ijkEpSo_NN!X<XZWYMPj-0ZxY4WF2odwi>2)Zq6Q$vAr+uDt($Mq1WX<
zPT@62jdxZWQm~$d4{ml=cfH$g*RF_f(W|{r*B;-tsv4S_A-A@2on2gr<-!^-<-&#_
zvyM~Q&KpX@Bl}D+(BqLi%C^jIVFBJp*~_nX`!HXCaQjC@`mM<D1)`&}3fyflDi1sv
zX_T~kFH>cTdr)O?VorOqNQ-eyaDUI>;J}iq`r23#E0dgnLT~PEIgW7VQ|C=Grw-t}
zUYoPe`>PqI2y#*gP?#c9q!GSzJ07Mil$9r;$em-cU=CR`E?Gsn?^Y??)Y3m<KW@^a
zuY<Nq%7&q0YycXq%v4p`djak95x5&am$0ANPI{yM31AP~g$+q)`1(VK!8Ze+4U}AA
zQ$??0XhzPg)&MAQ4;S3|Q}xZ4!n=GYng{eXEwcBXT}z&XML@re6Gn&=Kx`2%x=Cf5
zb&B%LG|rW&(?+OtuVw=XkHOu_kuiTySmPv)sx0J6{HZp-p&q4K5f0)qe^U6q+T(2$
z5zoqco5s<W@bo`|iU3u)aeL*%?V3B|a_dL@V=1@(-+xwuJIl)fgtS>$2M=dIxD?{z
z(o1Z7$6<9LkCGV~8MkMdU<+@W+6=eynx)`cr-OsdUf#2Yq#9y}ptE;c02LRQc;K2B
zy>P+DFO^^c*=c>rVgY@9!BV4)A!=gt7tg{0EO3hAbVN@hjrFT7RO!051Leyq1Ylg%
zVCNL;{c^qA*Wp;)M57o!uQyMfDtVSa_!Rs5nw~q(zY-GgSoo&pHoYjQ(qXvKzXRR~
zxnU2<+`%Y)qk?-mm%Q2;s->8BHRDTT!T~*nN|gO9msv(gD^>7O;RnlGJbb5Wpz!<7
zhb@?(h%mv(IvEa)e)UN|>{IA*=i}QZu068y)Xu@LbwgS<ap`o1Rmq_mx?Eh$EuAS|
z<L8D;@DGfz@BJI^Y(RZ?e_P^kL&6%hqz=bys3X4#%-X#DhgjAJ7lUutD<sYGloh<N
z_hnn4yjm1|9c%vFfX=x^)5{V#@TDc`vyUTkYJY8<_Kq(?Sak7m=Uv;Ow%l2_=w<Q5
z$o?y714kw45;A$Gv>zr{C=Q5Tb;se_ac+n__jo2aI7c6rA2JQ5aWbTDUnEC5G}Yt^
zkkHx#RCYrFIMn<F*jPr!KeXgr30`qz@6Q<SQt90v2+;rnuMdkD>;ewiI#y$##BWsk
zvpc0N0t&N}p@dX8#wLr@C!5kEgO6tl0cYjx2DHva9ys~6RZH>d5loCHPA2q{ZSQAX
zn4y@W|33Te%`hlQ(dl7=LK61brt=C_AGC>eyb<{8K%+L~ai!9&5@QJ+E(NxO6>Qk4
zn-)H>fxeF#Q=z3#S3jjbTy5!^8GS;dTVBKAxA0U=uHLFWKzkQt8($h4!_+Oiu9fJS
zwv!skKN9T^VJt9*=vwy`t#2YkT&r*sa5;bIY*6yX3uOG=SPuYn(S#f_m;a&6@EM@Q
zJemXK(JwpM4Gm38<`2T4Svk&OLZmkGx?~ud>G^r!?SOE>)<dL9mzc2(=%Hgx;{|rm
z*)$k_+jg#B<$1gR`U+K07HmPZjUtnG7!cGQSxgij6oY1FJnotF`@woF*C2vN&*J$R
z6}<N5K9{{DeIC@^vQw=^z?~3@ml+T||G;uhw>mneGuCK4?~mBVH8*+LV)9c<u?R0s
zE!%$L)qA&%TgLlq+I4R8JEq6e`d`YT+xvI2eKX~O@BPjCBB6ZfDj#fh#RffSogq|A
zs!te=-b<7OU7xYqV)a(%nQt0Bj^{p3ec$|bKO!gD`AbcZ(BH?WZ`Mry*NlPQS|McV
z<FpkwK_<Dd;bGzeXi55uXU4tEge!powD9IlCVGxzD-zgVm;EZFL&b(Qyxsdslsti%
z?bCtoxw${l9n5gXW#i;q@0fa{;xs9gV2@ws1EvGhLnN`iv%?XWSCUTF)Cy^S2*kT5
z?>v};8l#q)kMn=UiTw-*ZT<{6XTyj;I-uZ|`rxB=LbJjA*WOS4POeV%=3vHLs00Cw
zJZj4Mjd{|_&v3CDD{Lp;@b1qTb>;Dk!8NX<N-!pinA0xd(R%*6@Bi+=B>Arpqi^<e
zZVmr;)qmpHV3`y3UrPx8#H_1Z`aSh?|4*DKVEf^<{oe-64~XgK`-PQm+U5Vm1F$>4
s=jgvrN_uug>ihpW+@FTXo;*JN;r==J$z#xyzc8PYoVsk4v_;te0K64vsQ>@~

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_009.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_009.png
new file mode 100644
index 0000000000000000000000000000000000000000..88c5b83691a317cb53f853bdfe657f3bc70bf047
GIT binary patch
literal 110388
zcmb@tRa70%7q$t(3GVJ1^x#e)xVyW%ySpCTJwTA)?gxS$ez?0PxXZzH_+NZ8Yt7uw
zMOXD&UAr#2yQ+J?@3Ui7l%!FSiIAb7pipIHB-EgwzIc9~ha$p%w&nH;Zl4{$hop{&
zy0evsx2c;Yl%lDJi-WU=gRL2bm!+G#t+Nv+D=#Z2Glh+Zhl{%a8=K?*J%H8O&6;iM
z`;Vi~NswG*blst#&`tk$Kv#-Z*g`==LCH#pY53%yy)7qM>Si9qzP0fk8ko+tN=HO1
zXfje&cQxH=XxamQ@bni__QQu#T0w<D0g`@%eZZ~$>A8yB`MPT3Cesj|P)G-uFX^hP
z?$WS|odDt#Qyb*<oWy`0{td1jZrC3jU|t*>DSBjFDtzVayUAziyQ^*OwUm(XnKf#`
zKbQ3XKixaX>HcFI|EdIJ!?qTG|9?7Re}Cwc|JO*87W|(V|4-!qu1<^n|2qt#d{dH+
z)$Y7)<?l~yK2dbX4YHa39gEn!3d}O{L$WVI>+2(c1dFc613u!Hjet+aJ@;l4AOBkE
zNg?!KJU;^umTL8<&;J><1h=2>CW#(oXdC%fJ@kAA_5FW>riuR-w7$L$^{V~x68tCl
zF8DFJ=i_g;iRr)k#&*Gf!H?v@@!mmKFhR%o!719?KB(5DRPg;Ffb3TolUl<Mf^b?>
zS(!6LN-8P>Y90G4N!h<l`4EEV2jZI67(_;-Dcu0oo%}()AIos!%al~a_3}^zyDn{i
z@W}M}@>>baUB1Xh5Ksfae?edZHDuyT1Tl7k7K~_<nql3q1i(f=;ibPdgL?9ap>4J?
zfFtV)L|GXu+q_)#?Kk1~J@gNIGf2qd*wqe0;~6W&`~4z#Y*Kss55XI2EcEDdiW*JX
zcYM8*u3}qiN@IV8*lem)tm>`<@>lW?<-tc`#nWv_!N>Wwd9|K0cG?K4xeS0Z#>lu%
z-VBOb9EB9%Armtr*;QU<$jtBO^9SKFaAN<>&?tD9TTM>Kyk8gm(v33@h>!r~&$DF7
zAj$wE>gM@(zZ$5j0i|(<5w&r4Je+7SZ@6@)JF{f~SZm*H^vssU7J`0#bJ_YE|HYYQ
zO5po<EC6hmm85>XPOVmpr(ztbPJFE#&|N?@KCFb<wfB!-QX^RnGzzAStjt^X$aj!l
zqXBb_qP#8+a0I)jSprl1#DlE-zbf0xbzDj*EyV=ROgoKZVOoZO(f2J{49W$TDL${Z
z_T!<O_>bjnzjDBb^~X-Iv_2cZh`@mFbOP~mr`tY<h;9$0MvL2!m_ShqXEB>5mPY)D
zN;aZ|nPD9361fIK1E@LCG%^k?9sD8kzK5+<&A(V#a#h&%+x7%LMM5sN#EsASs@KHZ
zJ8L|vzM;EaS~cCoW0Ke$32?H(2_Gxou7}Z;IRBJinud+1_+yicIS&^=x8^6gxiEww
z3rIGTUv90@a(CM9o_>q|TCT6(bHAi-(D{Vi{j&TtZ2Ye2b<b+<Acjvhg-c*B3ox6)
z-Rq2bn7NznQ(~u~KusbvXFuv~+p_m~*Rzin)_C^yubagG8gwrDOfEW~y}j|qozL`J
z$~>-WfFNG%p_|}KqvKI-W?eN>qZHzh#{6(DD=X51gkADGk#Hh_nzxL(+()0%EV|+a
zV$%F`Rx&9RinQ49YOea_b&l<g%lQ4bw(tIawZnNLqr&gyZ^G^=7_x{+d#^SGsmG#=
z2bY{Oz0>J8eEXO9674XU&(gQUzSHYITNFBw?0*Vi6Q29B_x913%Er@?nnr8zR=%L<
zOCNqbZSX_zbDz)W6G9&k{C0SoACFn`9g8Gt&bP;%FRfDU$2^PcE<oEjIFiY^2RtRu
zOJKZZ-P0bgL0ew=$Cn-6o@6O(ui3+0K{6T|Fh&rCfyihEEM}JL@gB^v296$f)x}Q+
z@y$m=b$kWqac+ZFqXXk(rJsEu!l&0q<kQ;+8`SRYAIJ59kj~eOSN<o@{I_)-7_e!<
zAX?F3)U*yW-fP0g(Ph9zwkT%hDci}xWOdK`^LW7RvHuWGjzXk<y(8oRzaDau*X@yD
z?(L6YjPvUkY4XjBf5_s8W33ZWW<-h%G7A9w8i;%nXD0Et+SKi#;20Df5yS;frZK7A
z5e^yFO|oc!H>s^)mcr4Dz2;$~3YX<jXb6J22v&0&jA+tgzdw7s&1*o^*}bY8@K@IS
zHUk7;bbpr<Hc^O}!4x-X<;o{}iy06e;$tu{Q$v<+h6%bI6#YBN76^CwmDE7hrDVAb
z!8j(=Yon=#c&%*_4vX6GNQP0EBd4hR?sD~;(7I)cHnj=Cchaq6cm(T0WkyVrl`4wX
z(5Z!p^p&lxa$s!t1=-y(WA(1Nl;!ZnJK3;}1b}Q;3YSi5OlJoCrI7;D?81AOq|O5;
zvXnu%m!YAIwP689A>1i5HC)P^%1_!ObA?|GKZd3OX%{6@v16eAunbBQX*X^`(MC%i
z4?<WyoX-jmgmsYXuwej|aUos)(dqAqaD`jqAJRz;p%GJ)P!qc<T=Vbhg2(KEyX1Uq
zPPB-?z@`gB!btrBWo>fCvgL0k7XzC~kDdbrYb=SK$yD<<H6p$KG8YjQvEUcAw6~6C
z3wV-0yEt#kWm{n#iQq{N8`KZyY~bVnceEFgjU}nek)!l`j5ME@@2l(=Mnoy9#@d|e
z!~8;$S#Bn^Bboj%Zk(m>e`U;KFp<B`yGrUNQg|>Q(mN<hDM;g*p+q+G87HSkJ6H^_
z{4J#?pajaTwTa98!m$Y>D`ey!+w0e_VFUnWi}zLZknNw2oda^%g7&LBX48)^G`D<C
z$9n=!amQxUF|ZVgT?jkmf!VF3*DGFR3zSGS!frD~<nfc?q#?JfxTHmv69soV>L!Yz
zoVZgjJ?*qL-a3FTv>!vcF;1{44A4wy)W|hMdoV1=@A+y(K+!{zr6>2dYa8+c5UN<J
z3J!Lz<=&o8qD0_Z2N5S(f*1<jU6fZ6k-v&}n=%#!xB;Hn?<bw?vmP`}0YA)?@pfq~
zLKuGd#bnTy&}#yKX377y;Ugc$^^hEf40=R){`mptyJ-y;9Uk0ZY6C&fsOXaFa0I@x
z!<V~3^W0j;vnjOwdmc?W6Vy6p7fs0~JQow#yi_9iyg<Yl%>TZ#M105p0=4DOoq7i)
zI!}K%-9#pj@<x?8Qif>Yfp#A}k`S~3YnfNyc@)Sxy*oKV{_x&h)GES&O4m&BBs6SN
z6DD9@OT@IgBPdEQQZ4+sM%T%M52<?}_m>|BMW4}{hbgyTbJ%XG``)?hjJv^RcgdH5
z4}n_&Z}C8B>M#HefFcZlbf_ZlcogkWYyoDGwE!{T6bl@>7||Ex=e#<ZyI_Zf6;O~;
z#<IFm58)sd|GnKyPK);D7Z%<FC)gWwyJ8zV>{gRXr5OS}351sD)KHM6)TxT`bhZ!I
z`hD{-D6J^$Y9cK#m<X7+^c9R~WMNNHrAX>r>>LF#c6FIW{_X4_j2*`yAo={{iY0%$
zQhwY|>zi66W}{AJFHta&i_OORi4ezzl5yl6nk>N%-aiev7a~a8n=qx4^)QFcZ_U#|
z1He~Qq8}DG9l*0|(SM6|({Q~zK^s9p$S*8(JJz)H`f<b%9oucx!AMI(H^pkQ`NQ9S
zl^~pey{Qd$beYefs8z>WcGk=ReJ}irh&e@b?Q?MThr5=pgsK)6^g8o#7yRh%Of1S_
z%N^OS5RbXZ=~;t|qLqQe@A#I6%o)#Ls9rh~@b0CXF|uLD(0&S!YB`da#npmXVa<M*
zhKMy?2RS3J@8JjrTig2?+Rlxj_Hcq9{6r}Y1eb)0^aQ6ixtXQ@506LlNxHRmQoBfk
z{(@)chF13e`Z<y)#qnH<g8?GDS69iqyVQ8`aE{9+hf&u)zz4ER&&e0HLpkT)BRIOX
zTKW0;Dw@>!Y>hlr1ZKcr>8P=Fk(DT3>)a7Ih>g@kS~=$g$7U3AL%Q<1nmE0GsCJO+
zjJk)*s(G(U8LL@c&)4L8-Vg9yh7r%-CrWe+42A7K797)x*ZF?23ZP7pkch6jh?ok#
z98>wZP$B`~)yCjGLuF>(^q#p(qh=1)4C_TKCoR&%b|IJkkggi06bqZ&Kcl1hs}5rG
z7j>X-4rE?D!S$9P_2a`Bi~NcOC$cE6eA%Y=T?6BMIB$Gj-;(Q93s;Ff6#0ja!LH!x
zgUiD?i3ilY)lv%{9r*|gj`|HLu%=|oOg<}Lr+d-yJzg&}L4p*rNI>i;tON}eV^LnV
zbp0D3Ya1Ig4|YcKY~$%gyL?>e-kr^A^0{xnF>u>BI$ys&2}ZjJ?#rF?`QaT5N{G_4
zCzp<&7tEQkq5n6PhmXhTKIu);xtecgE&(qOqB|G#(XXJs4<>`Y&UkLzJG0UEsX8<#
zEO9xV3hVM1F*Rk*@X;3TqanF+>l+eDFGt-xdf2>zs)uhdk~D~z)M(C;t7pmG-XbQg
zLVx#;pfmsE)%OiB8TUrDM(tnw3;DSGH@FL@UDr3D@BEIkF1RfF1wcqcJ9GqZ(X20l
zAPKL>RVIU2;8o&jVj#6-zbw}#&%U%%;N>Y7CIQNt-&H<7!4CL#!^^0`RP%RdF9dz-
z1F&q=nNhpP1*LC~0$w9?Q%gqzgh9wo1%8%gw21~5{`Pc~_ds!44vHgG88Z^Jg6bjg
zq>Rb9D?eoh^P2(rPCLBbh#W!xuC5}8m+jzY;Hx$rV}el4ZnD$*{1yM!Z;}fOdLIlt
zi`IscVd#bPRhazIbKpRZkfea4(Y0IxPA;WSZsEnd!j)d*93oc@G#XfFgJ{f_(UvOj
z;M!z5U<)0Do%Bq>Hpc1jSt9@sQdughlu1L|>{D469k_YQ5Skvg#-)D=%NX$X3<0r&
zb22Xz=1v8Ni@aMq2l8z59;9Ld_uk&W54l4naQfqRs#BSGomCtaF(ajDWFuHzUECBN
z*eg}(#SNEhPH+|HF=BB_zv$~(`k~TLsor9^EUDuz&R7cyc6(5&ETevAR|T7Ip`oJV
zb=F#ShK+WaT$8D7wXu!!Wf%}Y54_8-W3w@{HWH>^+q~*jG_f-k<>&v%w;LVSw)ZON
zD%$*@IT#E+M?-$vd2;aD8L*mYxGV^sHitG8arrn4F+}IOfleEgWrt5CYnfs^QW{MY
zZ!RpoCBES|ks-SOK64#g@)z%|u*=s*g!OSf=1WW4Z+?+NS3i!5-}+rjKh%b*vVQd?
z6|ej4IzeHrQk>P>xs(=Ci%mDW7ones<dq4!vmIuazc*IEc~B|e2NtciV0-j!LFs>y
zkR_1y;o2L3{}U=7`_xbQAi6M<QB_4i$WVoKO=d1ll$)IDH^>g7gFV}7oj_X%qs~~y
z>23zORt$OYjZ)qwLk0P@B;9p3?>0X9&+oq=4*K;U78>{EISMa1)Wp1=BMnt8&j>`H
zU&F4x<w@_$>aJFN@svfGlnX&K9kpxSxu=IC4|Mqt;Zk_#;r-mZv&lVN+t?S$Ix^c(
zV~u%`zhYjB#r3u^VpuP&7D5Hn#qDOOJ|nI`ns-GdJ3zq&AruWbTn~62O_FgPuEcRN
z>I1iqFRyeq4SNVcGcp*C-gxx83hd2$f22S3eHhmD-NM%gUlX5QY-t=!SzV<sg;#b2
zZFauFUxJqFaHbS`xdVI3Q->vbm?*vSNrGd_=EHkset}aE37E0Ihz31G2b^jL6%1tV
z{F||c`utf1B`e69l7Qi1r4ePJ(@%N3ih4X1f4DrY)q-69+5#m$@+t7l`eyPYCsqhi
zEOi^&yT{Ffi(Vd-9*;_sI)dH=x?lWPNTO{b6ht3zj&tFT4Cy&KWR_ZjjI0Qy3pUwn
zRlS*KPsNh;YHk5&jLlV;OluX$Tpa(opxHCUl~CaAD3&-9pOlx%m+Z8i%B1>!QO%u<
z8|D)P3KK8yfu>#gh<Gqc%i<m3Yiw6q)a8847u}{Vkh4<tJhvz)DM!tO`RuRWS4Uo%
zj9<R7ZFfboAj=&x6cop&#87NeEiq9$nY3_QwzNvB+GR^9#O|97&t3y6Dk>JLY47YL
zL3*>Eq)UWVsu{eJpbCQSjYA^kqajr-Ts+TF1$yL~0|<ZfDv7a{3D*JMl7F~oly9@e
zA`h3M;yO=DF(qBXh{x7WYVT4Otz^zFhoS5;V!S0yPAb!~^VKu4vFYU;H}FoUI;4Aq
z1g7)~`@Vklku4vNCoieMy152mz4aipo}-*04NYrLE(zb1*$=V>`oY2H8t{lBrgWI@
zoos-Ppw7_8r1hF8GLua@bpUxRmD)sZSmptH?o03=&mzy4nJCX#wkEF#gY2T#FZLM;
zMFE0`x;@RCUf1q<hho8sgXUcTBfj^%%fMgqA|tpT5-i*LqGoxZ%J16nC6eMpdh)g`
z)jTsKMQ(nV(2EwZQ~!`fhiG5M4m#MQIIm0m`OsS|O6z%FPG0TVtj2DhC~p2cEc}>~
zKD;koc6+~^hDlN0S~8MO37T3-@LjC0Ap5>H@oNm*nqL&Rl5^?b8-cytpSd!A-0s)w
zc2Ntju7B^FiI>3-Bd-I~eTsicrjVxy)Hv**<GQlpz3t)<?J#elM)~7g;kNLJQF07n
zvT?XEDZE1@cYhzmtG}-XL(n_@Ur2i&mspH^kKrAc-2yxi8BMEFn~1P~jGKAJRY#0^
z4B54o*d{&oKYVbL_kN7@+&nM)zQr3Kp4=^mU}ih<nUs+um0n1mR&ndd+gE8oix1?S
z6;dEv(|aaKMYI{ZxdkL$FP-}6yS&ZODkXkSME;S&5%y=ZyWKV~^ZGs==Oy9rEW&LH
zGl9}F+rI0kLyr3PYcmfu1EP3nZddn07$b6hUTEbhe@cV6l?{v%vk^t|Dvmvc6m#h5
za-EBd&p-fKXF4*F)8}j10}XbIvU^`yI{Or3LAR?B!NO<>WDKQyju5{88$0Kd5l~sC
zMdl`(03JMK;SDW?PU@h>TMHcyU-wOuvy5HbfW<`(I}b0lzYQ}F_#QL4q$NFIN!L8%
zyWJeOWYRN8r6WYwg1li?eP|?}LwlU#=vgi=_X~sli={3Vmivv0GluX>@yA%LHbZq>
z+bP%kg05CI(JpCOE&CFN_%BC2@fey2FI`Te1eeGPBIOmVD@#dGw!5ES1RJyCI+{Cu
z>{Mx+<4iJK(t{mAdL)=8QQmUS45#Pe?BZ<JGxW(pNp8SLU^Z}2M-8j1uFNTOp|$Hj
z?6D9!Rr3l2b4*IQ;`lHmq%5o@#mc<Qy!xIA-jBIyx4UOWB|LenX2SW}^L{<NSQ!nY
z%?`2^p*0cvUEU=#j=T@w#(L|*m*%n)m9(LnoU8Xg<bT!<`t6oQ-vV<`Gi&<Z6Q<R|
zePC?vwy(2)hz5U4U6|oD(`B|xOiIdOJ6NdBZ6G}c2_A}FJ!A9s)gv(bDl6L?D^F`7
zS=V2VZqyInf5-?1Cmj+}dTNT04LJlo(1qXCOd{%x_CX%Y-*YF{8c?77NZ_qeB*UC~
zcOIum<K%Yb12x0kV_b2-f926_D&>RU7|SMV>ZZqnoKG(ggm#{pXFRPyd#-^S{!@Jq
zR9=r%Gd`Frc)nq$ojVYd`H+xb4u~LYhGSV6Gt7!DK6AtS&{x9HE|V}SH}#O~mYSHk
z>-X+xA{5qvq;1g1=xYR%e-^+t5Rra<TZIaVU^c`Z@MgMq6{#J!-4!5nd|+Bjj^~Jl
z4mc^MI0^c#{q_Awuy0cI<M1PH+~3^EsK8ZQ@BY5d<@^K}oQ5Q+i)r=+X7@VV4=^HQ
zA;1`G8j+ESk>qn3-8by!n^X^at2O}LTLSN5T4Hok_d<-rn|%2RI@%rlukh7rhQ6y-
z<xRwm@K<B=ipj84F9NLqhxqR|+tv?r+-lez!5<^9j|G>e=g8JVkX2xp)5@lNat^BJ
z;2w%;MPoD-834KgwurrsHd4Y5?Lr$#)iC=BgVUylpoXxi&{Qm)HB&sNvgEphriGQV
zY=#m;;M?+Cd1A8Ac+2F7w>6ID$bAN>nj^e+ceBwhClF&8<?Hf+jv|feO}|a8HZLDa
zUNc&P6NPMq4Fv(>l(?IW3m;B8(P+-f>N(jDUK(pyb9cm0T7GG3x*d{87S}c@EtWYr
z1BDJ8fJYbOS`0xn%Fk_khk{m`>d-@hkA7UVOA<BWd_D&*6RbwWmI4kexLjRBy~zzP
z0P5+0j2hhye}|>g;a6BVH9A-aHwCbI7waA@>D#rYB)6-^BBH%;*o<dNoqo+K^I^6E
zorY$aZn|t*Brp_%%CnhjSVk3(U*bn!*3h4gHb-gxs%$4m%&tGzXX(Rc#r~>OnWtUf
z*OxpDrSZ8-S<72QdT6n_zXMBTgTRdrg1Ht9`K8P|6}|6@y|3!V<~^;G*T+~ryEhs3
z%vs~WP1sQ&XfkvHV6d@`5BOq@=0G-0V6Z?#oQ6M)EnKuTxRkz{S2#Q+H47ipyo6%n
zMLou6tZY@KXmxz_0Ir0Pb6|tuV^8&0dfK7C*>R+?#f~=XO?~RD`?nTgw@#|45bwnF
z&8|}rAF(nq!2KEZRN+mL#~FKJ)|o9F6NGdm5?|)q)W|ouwy$_D<j&vlGWbwf@FU3_
zuBdBYf6asC9?V4r3xin=ww?pcnICl}9^9rsg-7F*#*N<de2QotsWv44A?oA)*GYp8
zEpswS*yFr}M#+1akguq`X-x<nj@ftJ8~=Rky7R1%<sjrW>;BvUx_X*^eSnVz`QCwQ
zc>=8AL0}zWHW^|-b<uVq@&ct6t(Ze7ATgOJ6*gaE5Oo9OKDH?sZLqdUUTJ9WiM5s;
zgGEWJ7I)h}!{_Z4S4haUs$-Ju=29n{;^#UrJNx5e>pk5&_!dn%+C5itGHa`6YMi&e
zfxDooxjEcDaZcZgp6KV>ECRXv;rzM}`nsTcm1@6D%gFiq13?0lQj#-7u<uFP=k%-h
z)7tcFl~or_IOgQD9$Sav*Emx(Fis=d(d~?<mm*RHECBKI{qW%S7I{n_L>TFDd!t>m
zlGM#4XdZZxS^xd@H;JN-7jz9JimL<KVUr~(HUMdKkJ)fpzjoSmaJxaH)}oy}H~T7v
z+d%>CetCa?XeZTaEq4t%5}H!Yo0~p9w^Uaas~Gk4r`5W(Mb6!xNZ)2XZ)`-&`p^35
zHdD&I$w9y9G}wQ!1^VO$Z{(@BGj26G+G#Q+uIZzC3aUH6e!6%G`8X+aH*qRYXrPsB
zyt<WM{_sk=t;Z(2WJIz#@Htz-9%V`(75Ya+J^{5b%j&jzTV1>v%Uw94*cZyE3i4g|
zTGUXCG&wGvgzA<hOmlG{SXvH4suam_-V9jk5jmCsG_mKhboSTYEOPuY5ld@$glH&F
ztVcteTW;E^mud!8nv+2ZE1ZwY<tZO$SG+U{^jmBIfhIjT_KzN8W`9x9idy7U{7U`i
zxIEj!GVUjdFA!HnMbq#5N1tv#VtM<O%237L*ZF~{-6$x0b8|d>F`Evtj>r^P>=<t4
zV1*BAWTnpXh%)8hW{sQWzhqsTIPp)#<NW!`tUar|#%$F5qL|Q_>!^p9NswU05Z~V^
z(Lzc^2o<9W=J-GrR&R$Vi(SADVUV^1SaE6$Tb5HD|C*vu7%C~cxJ_=(O~woRd)u;q
z`>V7rl2sy(gG$RB`i0><NVz=NA>5JpnZ>wt_3`1)y>&~elhWib?n9`OUOs?@p@<a$
z83M{wz|<A!EncVS9)B{$+70^PlBkDtPZ)$Cby7~IW>C{qrV3lKQ`5C>j)NC1-BD%H
z)2aw4G%-YyjATSkT@G5c$c6`JdS<@Uy%4AiU>h&@m4D<zYVNz7U!ES0AW57Ne#u*$
z&iI}G6;+b6nH*0UtsnSWu6VQ)n|mlvLTT4X0>=h!SL^w^4`b`#3h@?98RNbj%GF%|
zi6hWQ7CdGv*cZCq>L!P%8wbHyL&O*`ny2*Q0H?LJocIZ|E3HO;5eg3ve<XN`@8;<#
zJebff0w<y``@)%o)2<_GW&2xC)r|t5;CY=kb{>7CRDm*}bZ}*f3my`wf49Da^7g;2
zw-=n=U%&0~;;}zJ^NsG)<Mp3IuUPv&1V4m76S9OX=O9)3fVn}^|L|UPZLt&3;}DVG
z%}!z%kBSQs{#y;JBKdvChDr{4d5K}+>+~-`s6cmFGDp$Kz~~NPL%wG<-aop^9;oi-
z90~!9vFn&G)aLaOPuW3>*vA8s!$zzqD&XaeBHqjZ@aD_9HqUX*qbxrWY&iIN%8#>I
z?3JwGpvIS~zjh}5CT5m~cWHgU;!Zx^Wlxn!OcCr(e$LK&uSg2T!5r%Be_Ml*(V2fd
z4R)4UPBuRah$Zv|HhO@u^69yV`e;*S4C4VvUptzhzRVKUEPHs`qejjH?d@yz>ZeRW
z^NQd@A0hkbjwwp82MkGAKH*23ua^pM-ed~K%H~$T#j<s~oo%U=xDuWEm~3o2nmR!n
zSYLp=8nDDKMal7PaWPk1e!h-^PCV`x#;4_`xY^*r)C2j2fk7R?qteceB}v$pkGkO*
zY6&vq$!pYM_BqbT`YlB}zMndt#fBRDl~GdbF0SnyLo7%6zl+t72ROw|*Slxjt@mT+
z#_r&Rws<~Kae;8QDrPYl9mU&|rf#33+X;e)<}+9QbIFj4v%fbMtRT*!&g<=W!9s0p
ztf{ZNji&x1jolzPTM1Cuen}bXk(h(VxN1AJ;q(;W7o|<9xXLL4v$_TxCC?!m$?rcL
zerjM)1`;HO$ezB8ap0O^89KN#495%$xp`<b8%Sw+O7nC0go6X(`gde?hGWY|MDu<#
zc4?zt&%W6|<I5M-=Ftzr_q#OW7)}uH^7Gu2aKZ|4;HSe`W~=92!f=HT8nI7JLoYbC
z6&|1b{ej+V#v@b;AcV4e?cybpSY)<WLPxvaWwhclT~2b@N7UBzz;p2RDn#2^vw{ht
zb_#-!oQG~c@%N&Q3Lu<B!4@@JPgxU+o_YoXDC8N1Z}^jd)6_)|gbs#u3_A`1XL`LB
zwaU@Q328wNeoGgSb@k9L;$<+ivhWrjU#at5*{Bv!Po0r4=yA&NeayIklT7i*96tG%
zZWLe-X|tKV6H)?gW;Mb4=Zh45dO|jijN$m^`sK~r<vX-vvqe*@WbTEU-ne#OK)n9v
za#>3lSD>$N(rZmG4MPlgKCT!d_!Pf~^*xb59uR@u>-70mMtXYDp%saJ7TC{U_OvJH
zzB}TN^pCyE5Bw@yToSqieZZ-PO9BE4zgY2N$9Dh`wk2#z)OT2Q6@@kzTINi7v$$+8
zu>r{O$b(LcYWJ|7R9UfXW<h~Guqu_+t79P4TUSnkONX>rK9nNGLgLuT_@8`#>vYhc
ziZ)umJjoK_M+g2xz$NV$nsM=*h^~MBRxpc{^GLR_l?#elD2^P=-4Tvh<2;x{DtE>X
zWKYz}v8~-wDlQ3W+<yd@Sq1nK0J*BC!d@=(M^vY~Bpla#6SMtYa$(u&yYj+IKmeb^
zUsFrjTFMp{!*x@$Ul><9LtaGmlqO%pIq@2ae$2+x>7U9fX(-Xs8K{~BhRS}!#6Y;p
zROb^XVD`fd6cUQoA&%ew`-bp()qTm%%47uc{0hiCey2T8w{h!Xz^VHQ`BwTbby(Z{
zj|vfjWGoKumY&)A@1YnzJ=^14G^Qw7?&6GGIxasRy6`=RaRO?^`EMr7$ZBe*LH71?
z3ZbQG=3`A%Qj3-j-K^bF5JR0LUOdzZ?OnBov$G5Jgm&3~I<2JC_JFCGgB8_TX2zQN
zolA_a{?lZ7Y<oU$K!bquby4f<G}(>v+VjqFeD9%do+s+Mpt;k=#_q{twS%*vStolw
zN$W4Z(o;Rl;#eQ&*n*issOGkNf{mspD0i*ckYk@X?nJ*u%&vTH=COCZ)9+Z6o_yAB
zI<cg;q|0LEW3Wq}Bge;KPY$vmI|5}C$yJ2(YNz7klsPDMvhAr$Oe?63oYB>Q5A&{W
z<(dw}Kf+TW@06E7h`^NSdyoFs0hwT$q((|f0+Kb!R&9a-H71si^D!Q31$q)gyZ2QZ
z(!NgwaaIg5rP83{QO@SCk~1w0KDqYZJG~T?BQnA?&NIE>P%raKE1%8XKe)GK+?+cP
zo)K!~q9AysJi<C_(O}1hm978C*w@y4*xy=n?qFb&&j2c5u__x=4kpJVmefL%2Cxu7
z7NYU<LiBOmc*pp)oqTU+56hgz`0<8iplN5Sg5ID@uQ6L%<oO6XTDqdg!(6;bzm(Zf
zMO*$tqJ|j5ce+xhiKh8Np}+$=4TX_A;I$w9vHu|`Y4teJEGC~s`z-kBG~jx2*?0e$
z_8=acY)jMxp!=_;;*?~$-q=zjgUA8YJ?_@~%U;lPJ_1_}ODYYiImy{QH*=%czbMOj
zlWHjHQ$rElc-n(8^53n%FpmM(c@E3mS|2qKTB26Bq-10`sDxOvOl1+&V^h<HLvSO>
zn;nt>9z!cTJ#^V;iT52I`+ibqv32%@N$v3aklrs+;^-8U7bqsobqtKvv-2pIkt{8W
z@Sb)B)i{bvew<2%v2P_n@usjU_ZTp=lMbU9E|DCEhfWFOkxNaOinSOdXtHjT_H_6y
zCL^6tAPsg`q=-YbSy!(*LoO<tNq3~jy8Wg+V*Kj;7b~Kx(w$~)zZTpUaVqCf*-V+t
ziNG&M0M!b^BeFwEYoNkJ2p2udeu5Y$7x%?fdBibtfc{LR7F=|9BKSn=2sDzuXkz%I
z$oB1Q6wvM8J&}n1RO%d@Cs-6PrtjP{I+<l;`Y@&m7F<75XeZR9?s;%fZm&F)S5p91
z0n$eCM7n=~9)MD~vRKV&{E%UVG$ymGG`be7F*6Z!9?`N|{jL?vuJ80IT}F)tb?i$G
z?S135`t?^Q)%BmAu})nkPGm(DaxPs;Be5l5z_%8QQld4ac93Y=ZPCPFbn3z~006JA
zmR@+g#5%q&-aL~)J0vmh852?C(44|fkv9sHB<e9uE_GNKP=5W??9#XOnWUXMZ|szP
zawVYFcy4p8ukt9;`6SaErTH2r>rQ_{CcafNDZePYoi%djp7(-;ErYGI1~i!un9$Iz
zh7hN2S+!8LWn}LzY3OQDO$>P8X``7#0GVyJ{F8p3o||)?WU|y(RxlbOy1YdQ=v4gK
z!PIAavLW6<4>{8>%shv#jizt*BJrE!+qX}=5Bw~;KU%D2G~hPF1HzQdE+A*UJn#Ub
zZW#_gB7#}>&V(8z>->$r=z0=O=ne^01L(t>9v7e8n142<$XbG>W~W+BK%Jr#671d<
zoIoJgPg^40sIen%)Ko#b?zfj;FlogdHfNndHejt(b!5(LgUgp^!X`?^-2f|Nbx~1{
z%#;XMkes+zUhb=}7X!mDb4CGL1k=<OB_G5&eC{hk!6N||5G+ZU%+If~AcNykncPn;
z{_^b+JGX2k2Xje-f_*db^=3OD@5MW5b!+Qy(fjV@)zv#NNwjPFu6lVX$Ym8pMcDuQ
zLtWW>XLEGD7|=>>>ut|)>*X&wW;cKQ(~q^TziW5ex-5ydcqwg0*E{{fc=3eGb;g7I
zo3}rj|Bc<*r<ci5JeK^d0edjI`M6Dhy4(4Mgm!jp`a#{?i~m7xYLN`jJA-|)wm-S6
zorLASnv2FD4BA6DOWOoR%M-|&KrVERZk(hlmaK#0`M-^rw}SDrGoxLf_mIVZ+4E`Y
zHE_m$y|?Pl?z{W?Gy?f-pWKu}g`kkxvGmtUrpEe8Q{X7TdwR*tMy8gQ5Vw>|3ga-N
z;HFCltAshG4*3?7_tl?Ihb6{U0wCc^BUG!S5RuF*>&9NQwGs)Te7NyTBR*If&a-7Z
zZF(XcG;nv)X#g!&ovu5i+m8O}K_~w2Ax%Pt#|*fqBb>xpHMd3X)#yY?)1^{<Ci6uW
z+BlQIHO4cyHLte0P>c&~7C3;1yXYe^LBRl)As$F(B!j?PBJ4~w<Os=k2Rw0@AM*9Q
zgoOK}cLcwnIrrYvNh2UcBs0vS`UQE2bLy+%(U#gxsNqeF6Ka{X>S3x62uM%JDq6VU
znEiCrS5ivKnY8jS9I<QtdD=2J=GM74a`~2>B+?lUC3?5Wc17NE!?e-ins)3_#asz6
z=naUis=~lrTl&;1kId3SYsH4bjv)i1B}cTH+>}!NcyN?$69?7MH7m?w)MB)D`PaI4
z!IKNIw`U8x=u42g_U7m&LY?*s*Zy4h@E1Pb8Rfa41AXJ2$#x?%Z*RG{%*E9FmASb{
z5^HS%af9w(wALqtew~FAQ=MTLFyg<MNd&v~4c4|TxrjIm5<mTID|(hu9D?2E=-Hi~
zWf$C79cu>Uw;lo%$Mx<3Ax3WItWS-T)^ii9%k=_XkmR6fWXiFvdB;(BD(~4&`IbNF
zX=()SKoo7oaOkRFd$};`Z{XP-k!HNC@G1?FG5&NPJcha3?awhL!qK|5U#?Nb7dm7`
zDnMkqv{X9*@VOIXY-pG}KpZ6&#|*E{$lq`SstRXe6kdNMhu?`Hj7xEP!BHmcb|S~<
zjgZKVdc^CI1!N8W5A067unu*=(<?gdxv2}UIZ{ln{+^ABUM)9hvV47!1LWf*yA68`
z`w4(dRqnL&W)i%keBZpd;Hl&LgwMQjsrB8?!K}t_C-{%x^?{F<kEfS>5}se_K8OC7
zkh}UoN`&`PgtBInSBjL`Ww~?Cs^>h-Z9g;*MR*5ss!y3vG<0FEl|>3={y4qrmI5>7
z&*yd#;EQf9=(PQw%%ojsK9N29U$Y)dGb}8Z;`r7(UOng*iyp#HPd}1BB~tDL(-z{!
zoIi27^VJex^zr(^b{*HUHNo@S0HXwo7pXKo$B=y)L3EdQD!&m7&J-XA_}g5bo6NlS
z@6KNY60(g_MS3pJk}GNyrYgIk<?VM}FR%8T$wM9c&&HrDz|(|%`R59n%{Fm-BQ$n;
ztlv^*SK{8+ve|-p4-Y?q&HbN&+~KMJNo6F*sCfk$YB7IQ5>H?@ddRw%#iTMia!2v3
z<;szxgrMFhz25ZJPdn!yCPcR4gowHQadtM%D@5IKsEG#q#B{|)n1;)hP%Qr(#aI%I
zYwRKhHG7no>)T3MtuJJ!$NZMS==t3+tQTL8u0Us;cKuCOT??1dgG*5!*t*p>PodX@
zyV4x@)+%bxxzyO+Ju)49de}gDJ^OO$ewv}*^9t6l$&K19Fs!QDL%4i}x|A<p@19V$
z8e&z^44u=CP~yI8AR>aoNLvy!s>}Q))uYEy>WI39*V4GZcNDSiec|#j^#kbniN`4h
zy&nb_7QN>b9o+1Xcd`k_mF%od+Uu0&_;k_xs#Uo+c#6nm7;BBDQE1G^ou3;5Ss3Hu
z+-klYQnDgChP!j5Qvvz)gRWGC!5u!SUiT_SkZNUN9{hkq2$YxT8>jK!^PO-FXFf@_
zKIH0(vZ7+-lM!*rMyq=8;#*>mL-SkKBrnQvyKj8REZ9W)q<kF|nNIv<fzk`|aEFXp
zJT0vMdf174!D*OVf|hsd`Jlr7?>72c<58=}`NIK{>Pbjqk(0|FFVw`Nhi%Xj`J7lq
zS%SAEZ^zBqT_c@WMF>B`?^>czsC1_%TA`_Z>-ER`YXV`$ntXSU)GWC3sPQJfwzp`N
z+Mtvh_pi(5ryG%JN(u26{fwup;AOc?slXjx{AfFNEZ966h9Q*NmZ(rKoyP<wCsyq_
zBa?}!R1#zvM}%8Usi8Yh*E-U4_eR;A08g#ZN*YC_x#)~mF0EJ$BpbEdQ(aMLtm43?
zncND2yS~Sayp0zw7r%hG5p^A9d?LJRb5`T6-ID~Q&s>?Gv*%&Adjj9x>r9I^C@(56
zj&O4wWh);SHRn6Lt8SSbXM$2moi>R`$K+I-wE*GY4cz3IjQ=XyMgJ`lFR!exS!w`(
z#{nXl2r_7Y$MXs8x{bMQsc7PUas&%^asT6_Yaj+&n~SOPfTnIvvtZ}=(m@#Z#M#Q9
z=SrXU@1hMo`G8(`W4-u1jEdz5*9t(c?KZ@M%WI6U{wF=$xi1rqtt(Kbw#x{h7g~Ci
zcwSJA2z`u|A~Ek}LK3FhzvLiHn036Tz)0N4hhFF7rGlTAXkuSO!z?bwJ+qJ+?(=fP
zfqd&G7oZd;YgN!#!8UtVc<mvlnSk*Hsj@pWyINFdUjH?*gaOfa)`2cW#~#52UtMkS
z=_^6EG)KLk34KXT_9U|6(>2Z__X!U#I)@Ku3BWV3Ch~N*75F9O?EN;Q3r#?MhT!Ws
zts-z}CRHMSEeTexG!}cfQq6Un!FN`l9Zef2qRxsdN3ET7yi00GMN3bs`oC22CA+iK
z{^Zy4g10p8L-~CC`+DQU^P(5L-2z7fTm^4BoMMa1G0XdVw#Jyyx9U%e{|Wi><GN_B
zl{9S>$C;#g*Zg9mH-U|dlNB5@lo+m@v<hk&H1f<C4uA)+SIMD~kztUvkW44~_&kO%
zE#E&5WNrPkcMc9;-u5BwIc%xZy!geOqeORhW@rX5p3$z71<WomZQ{GRWdKkRB!E%J
z9(m^l!l~ak;WMlMcnE;IQ(3cq$rC6T*K3y$v8<kC_1`oMnxFE|iMIRdoWeU#Jb&K4
z?mMqQa%Q$c%5Het!RxN;SNIN}v%j|<=S$VgtE;5y%iIVEoM`At9dF0NiAkSZ2RfOe
zSF;aq3AIHcm38I4L&0q^*GZIPcP=YeI`zoC((uxP*%`MP%0_3P95Txa?C;Vl6hVJ=
zR^wGuYis0G9aep%BpAoM${(2uSe2)l^Yd`ct6}?Y;_I+Ikg+I&MaKZO&@IS|6#7Xa
zo{%AI8~LFe=!oclo?6tUFLc*yZB5D=#ww3c+4QlHf^fsQ|LY&iSet>uW-h~@B8D9P
zJ`%01idA%c#vH8DJ70nBRmh>VKlg=6LB(~c#(@T#dgzNz8{U1QA|kIQeAZEJu3uJw
zqR-sjMwW+ufq|Dog}$0WtbGT7-PJy)kTFp4k3`AMm1e+dH3rQ%-?8$Cle0qI+q;GN
zhepM_=yTsI*_1**@>C|5{K(~N-X-K-yKgG`a@$;p2olM*^-sAE0_}ZC;^xC{?d%=h
zJEQ<isXM(16I;A?KhA+z3P&XZ#tb4>nyQp_SnztT5AZwOY@=t|dR_*}T?FcV&rd|?
z5gv4zwS2T2dABN%YB}ht|Ge7Brv7K^9mCSv@lu79Kc&R)_b8^7z<YLi-rOzt&tqvP
zVD#=(aN>Zz=V7Zm;4P16+o~qnj8%grbsoh1Qb&Fibgi4Go(YNCMuHXFjVhs6wc)YS
zHg)_$^sKu$k`}|@-_Qyw5Q|d^d%upbLVR$IeG@ar8UA0^IGM^0NETT1(I=J)V{CdE
zW@iq<bi3bKrJXg-!qaER*;37f+xjwhT%6Z>vGQ2>HHo>|w=#$GyECUun|EZ9@$Hfx
zmTc?pjrJak`3egB&%zJugvmo6JvP5l3byTTSCfMAfrTJch6Kt5c^3POFU7H|bK$cV
zVG$ABQg>}Ycr1G=sZb5tMtjD%FjY;|2mS)3Wx|!gs0kJM8(ls82seW-QJ_3tpXyr@
zp@9FQ+eJ;fS_c^%8inuAdIDfL3+FNtYnP3{gww63ysna?Cs#4U?H?hv1&*xS9dT?z
zWe?0BcYUA8ma(N;hER_`f-?&BrH_wCkp=Z`ZK`tsgz1wkB<^`VVYz2D1mhH<v4m?A
zGtht2T!<}>ukt+rQgg8=g8r<;KtPR_oF+;-yk_2)9HpZY_P2hXeR{?eia$){47*iz
z^=px<=Qynu?iHyIHikA7qWHd^ExE7#w?-RKK9POQ1ZU9VYaf=V=bR7cJ{?-Pxw<*v
z|6^Ga50a}a(mOtgK+na5u2u|JjU7WhtYleeMxL}5zQLR?XMCd+(^z(DSA$uuU)baU
zr1N6<>*d2YGBmW*v2&9_>a@`_wDr!TuaB)JM<OIt{98HHs7K4(?X+uu&jLI;3hde{
zb)=R|VE|1#)x;AvynFvwR}=qzj+){!Y-IY9v4oA_{p~*pevdQc3G%Y^^!OJ-0r8DN
zKa7Vz{>#70PVS_WU|8yE!9j!QFkz4Bz)@wCg5pT3Ly~XyDtzV>Z^uOktl+CKMd>d-
z_|Ar!zAt3pOk(Z8-qEYb)-k|-nL)jAwc_IC5N)r%AFwz+>6~d0crx2YShdjChTvd5
zV)bY<Wwl#K>hAAE((FBQTjX@LRG%O`%L)G1V3wjZsG)K2yWCM*NQlp;#5C}mig!Rd
zLqJ}?L}+W`d>Y~Q%Y4#fspw<K@Wien745j$mg~b5v~$n(oJ;b((->sZ0)AR3>IQ7{
zL`|9eFK)LKK|!j1C$Rtg1p#M2{gkb5ER*eN=Z(C{U@FJy+#K?k@W<rqQ3!@U-Z-_=
zLB*L-@0gjU`$s<|IvLp>s)(DP-#~vqX`Rt#F@q4mYr=|tqZ5K;{8ZKD%ieime_~jK
zrba+#b|hd{(L~+Lun>~birc|gc%Ae<EBY5b=mp_*A}?z-mo<kisR?#M@2f519*!d2
z@3B_09YO%OCTkxgT6zr;8jHZh#u5<MgTV|B<KztJCH<3~jFHE(37Z;TWjLK{xon~7
zgM4|_Ryqw-0$4GA<?YazcAiRVo_Y3lpEbuHuC!5D>+JMNT-6VTzr8}U&ltvZ%WPBI
zMS}B@iZ~Ji^8>v*Mq}pm!DYkT30>&xnO#PomKn2L$@y(-jxMS!6CO@)ghq<l&#G^2
zZSB!jC{z3ns(T6E5RszIROOO7%bK#vZ{>o5hy!k3WBr2&*(D|~H=6hN4=%ebT^Ord
ztI)11RRig`rJKS=f#;rW_g*VqEzzU!(|4!5pPt5W>dNYBzudp09m)rCp`TZU&OL9a
zpBUYgk3c_Okg%2Sg@)y)FvDu8JIC!=c!>s^yTP0L2M6-1YA9pQ6?6@r%p=N7#-@sE
zqVOQiV=BW}Bfl?>U%`Z|UVx^7-tspq?f3q}rONCCO2K%O{L;&9`6O&UE*~-7-e@_~
z0yJ6#vn#xpD6`%}Gp4m8L#)PWcGGuj=3U;^Gzl^#Ur{M}9>rI-SO}Gc6uYHkQ0q`+
zus%O_hh<rHZs*UmRjQ%fnZr#7e+OljxrZwK#fth#O4#!z`Gs@T>-D;LFr;Oxx$kSH
zOdf>0=x_T(lb!@F=|KNQ=8E1;h(L5%Z=|9Bd%hkC?Sh<lvU~YvA6;i;hZd}{D`Lpf
zEMO>A&I=#HoNT2vUk4hmT57qELGLf|NUCyts}>5Tj~GjHIH*^!(o6uK{O(q#R2J97
zBGJ7iozY%35N7cIJwXU>VsfouMM2963+~95t)&_FOIe-F+w5?E)91J%mD@!<{Il9|
z%2|w?Gqj4r#0f^VHVrQP<&Yz*FbLu`RBT?zo*g#J8Pq&jxXUlyEu|+~s;GX;wZko?
zF-;i!tINX_bh`Zts8$b8n9rv?mYu!ofAM~NDhkdkIfBX34A4eD;U1iOI8?cGC%~AL
zR4rai{j3noStl=?S@=wzu`f3LWtLmd94_7V$55R%JBIB)Yb{>TFHAyI`TO;zkN;)Y
zIU+fJZin7aN4pO?1tiy3{m|YQ6CJ^GjopVD{J|P#;7DZHdaV`~W-F$k+fwhx*1k-a
zrPQ#QR%jT0r&_jMizD%$_L_@ZM5u+g;2ne|mg_0>p#PlT?=P=LrQOBnUELA9DxkFy
zr~DGOv0C|LBbVSE1u`M(XRW7v)-hnzyi{&RdKU)uGz&#zQTsF;SxQv*O<m~*(k-Gd
zc<OtRj&3_1nVU1=+;=YYFsk8=I|aot@Xi^@QdBe*tImLaE6bRmG=A$zw8@%bDc+nd
z#hsTG)V5#|>N&B|y!_g9LySZ8$DqV}6S)s1exqd!8khj{pI03NP28-pinXixO!Cck
zni(M5s}kDQeANVTVRV)yJ2kmmTC1~$C29_OiDD<cwsL*nWH%ubKarnk&wK+ia+sHz
zUu-N$r9Xk7wtsKry(9a?ygpBNhAhWfw;{ZS62As%Or5*rxpS??)h)&DEBqCzk+Ad<
zPyWpI!yM3$DA21Lb@_&mTBPspr9ahuljqJ+$)JsV_$<Q!m#Xp<wxIt&@u0*IXO)&_
zi!IS)%4NkUQAzk#x9x*uPl6fKSjvoShVPQexf>uN^DboG4EW@R-z;2plcv1`$PfaK
zBFbY)$futk2L(w;r;gUJ;s?92%t@rEmIRP5e|N*XX_W$ay7%CF=|@PbG5qJ5xpiX_
zXMH3Zg{SHPXBXN!=A9nCMNHZcE1R3pCY?5X!Ue8=0_HsW6teK8nvU`#ZmRBn9M%Bi
zIWE#P_lw{3$5{~8NirVi#M#}IPR_m;!#x1X=I&Pe)8ms$${{AxlkU0Dc+b;!=RZZ1
z`#?H82acggwqx;liGD#i833ZGk-OUq*`m#mhkY0+E>78?W~POqQeFJ*UAX^)#Yew=
zy-OdxwF}7gT)0CJ{bXt@gS4vO#=)OR<mF$XPpSy->1kJp<$00zv})a>Ip<1qjc~xv
zx&ziGy@i?srW4IUvtE-R0qah^xi%K6Pz-w|W#yuswL+o=+OCCfemrQ_$Z8P?pZ>`t
z9yM>tav}ii{;`FrIIo&8hML@H^cX0;BGUF-l-U!WaXfQXINV%AaiJ-Z1tI#O)BA&L
zTypZX-T^@B{CE6I&0#HD&!mq~*%dXNzCUvwzu~%?&Y&G&_fsXcf>3Y0?Tox$Tn6VI
z_=_Gqx<~<|_HQEGXwJ?xeb6O!n?RiZu8vq4J*yVe-0~IMmKOIv(JEQ<j6@}>2#iKn
zRH%}#p~p)+AN#>ieSf7zpG7`4Mc+mO0uBSK^b;|~$D&o|E5~J6vRcQd1>pvSlV?{r
z;+v5}GzF|!b{=acb;C~$g#nmm@g8>ktFWv{<tp+<`8N3`(>mo6lt{5e@uvQ`@98dE
zD$mSBRWjsAp(3q?0l|THkzYes8;a|UgAbk(w((mbnQh!ok3vk=gn48esQNyA2v&yN
z{gs8~F-^!&TV}}7?uB1|czi?M0gVvij-IdK`SlZ23xu@Qr{%W{L_7q=ETS0xwAI(D
z&b_t~>d0?J4&P%Bv$pq}7M?G=IGy`sv<R|rM|H@yKotllwu3fZkq@+%0CBC9DrWZY
zrp$DRwyM;3$gwGncPp%OJfx5`bu=;|3!!hm=ZniLM9O^p{9{C}I|j0<IC|!M-4?*v
zp%{_$Qd=Uec-m?|ZYbM?G>_bDOe#g<W3zwHWlRd~ZAhuS^EeDun2O6-s@juu4T3NS
z)02RLv^R&&4s`P6^s<h}z`XA}ua}D2Gpmh$UMkC8p?-^QenPt!pwF@*RsaXtKYt9$
zq3?*BdL9OodYXOhWNA0hQJSl15?*~gLo1Y6n()^Ca;h4iu8MZu#ZJDdz5jp(2lyq`
zZ&7ubG#cS68rf9#wAL>T#=I5J=O=K3K@A^&4rd-)<!Kbgz|WF}tdf_Sd&HldNne^F
z^?_&Z4bXB^BJEEvuiN;I{?qmfqV;#pqdAn{u*{if^=gfU_X6otVbyM`!cVyHCSq)&
zx?Pc<Q+E=Ep(&B82eFmF4+P31*>k(JDr<pbAV&cS`B$VIE)`u?;!#uS&gWDH_8$|;
zI8h<_=3}g9rw(UC06q7uy_MGNpUbM;#6EiyemunHyGwj<BUX%wo&=G-b=D_6%U5|g
zqXHm!{Uf6-S!u)Ni1KnreY^^Jr9|cIf(acKtp(2?$bNyTSgHagHpheXw#tOE^?Hm;
zl^1D^dM(;CRItxKjbniHQTp{pFdAr7<==sO55Kt`fhf^)InbEGdqB*5b#|N|BX9ob
z{6c)DLS&{2BX+AQIJDYz1{HPsWI{rg!5_$qA%pAQpVdGcSJZKKVeT19YN4BN5lrBo
zefsn?$4_wqa(%A+SztvS^!N_#pY=^jG5Ru{vZtkQdTUE0viaAHSe%Z9Oq7bl{{`hh
z8o##&ABER~<UP1aNeErQhCm3BQ8nRfU;74M`qCHq=)({BS08=EVqx$`5Ub*x%f*VA
zD@XJuQ=eU%MV*HJvQ{@&$N-7wM9-r|;|rsUHvuU`QmP3dQe<nx*<QDfd~K}RLL0Me
zhz5sup(S)JzN{FPj>=o=0I}7+ks)hwL$U<UZG2}fD`PLVn5<Rlk$$b_e53(5_M5Q=
zuWIW<*fp6V>H7_tZg-izV{kpF0h7Uf&+d4W3o@HNmj;tGX3$v6XrcnWX|^E7h>Hmm
zBPj-~RUoGl0xEJ`P!K}Uzeg1qL+Dgz(K$z16gdIsCuM=owj>c_e3ryfgUh=SMQ4y!
zs6)U-Lu>*MZr^2h{}M$xW>hwGizNg#aEw6V3`AjBFBE_+7hr_CUea|9PArSr5=8*8
zS~XOo3R@JEMTsjsF(o|0sH&JuCNynJTQ}5oqo282v0Sa_Lcn=t;ZlSyK^qbggD=Ot
z_WBp;B{U?ntSMIji7{m>{_*SP5!Nn`|I;_~nSkRwuv>!$e$p@e+}3BHwrwDEi5Q~F
zz5ya3gZvImllS`G+oI+2tjRI0qCZ;UBVdh|XEUaMz0Zk0iK?pDZ6_Ejc~7_0bzrfm
zd2s)fySGm`ISe#)!aLFUWKrN@jHBz4vUw*9mdllDAdUikJ!JbC@ue4mv;nAJFgWK!
zI6f1!Z@ljO<D0+FzjpI_{;XNL;rq#3>>0AFpPy@Rl+5Nm_nELdFW0ecChLA$Y>Jq!
zr8zxe@$ez#bi}xH_-RF|Jhm){q%cYA<N+%__kV9S=f27N?2st2fg3hmpX>K*j8A6z
z2G_D#&m%GrySQiixu!s^U$_0W^k3IMzmAX1ncp$@NxsI=c3Q0`C1$fZM~8>ZPiHI^
zT4ASaI${iva^i@jtm#?PrIyud^+aR(l(k>g#Y5o{49-;QY9(YHD(!@=EyG{?Ti@ar
zfBEZZOoi&Xsa}oCotiXEX7E+70}iEED3>gekrK|8_+rexdndpt<IyCGWY%20dX1{A
z*xuUX;o%9#Cx?_Cb|)ilhnA~1U!p2UoX(CoIe5sy;Q_`~jJCIU^^GquJDoFI9?%NB
z^X?%J50>oijEP-Gb1<U|GrsoIe*+1Lu4}Zckr<6L=mA77Z=-GptqPzi{X!Jcnyqbz
z0p5D!P1?}%$?XqurK2;R#VWBq(LnnoiP#2K%cY|A%vkjTtI8@d0FJ`z!G|QGAv1T+
zr%ae1@8Beox&{d$vwwmKf+pY!&VliIeG-%E@1%ebG`~r}#)59yae6evbSu0V_O?qZ
zA8C!y8H*}N<dY<c)<H$}*tF$&VU_VR#$;W$Y-i@A4U=Ct;4cH1ltU(Z=1u3Sh^Wh$
zG?YRE=%Ute)fk=i7U5!okTem*X00Ks_ncB9ieSWHlflMp=|-Uq9ia_aQJuu98qu{{
zF<~f4L>6~-S^Fp@hFA<E7?be6(z0O~nx>;H1aDP0?s**qBJ0v>G4?^(mQh(DTTph1
z@o0-=JY5oAx^^8`jd9iy>xRW@#%ejkVQFH57-*V?x~Y+*$s%pDBE**11xSgyU1j&F
z$r~-$Eb9VL7A51+NJ;CMvb0G<*J-`mt_w76gDV^%CfcZO*ES}MD;ZCBdE?D55XE4e
zAHaQ#t!9=<q^~4!{$PEa>-DqRXT?L=4I|>>{@*-&{_GhWP%&$tC6}&Wzs~F!>V+mp
zq!hqJtP5%w=Xgb3YgvObWdnd5>zfkZJGLiNE?>UH*4BigDDifLvnAfbXk>tiDgZb`
z7ZPC>nV;4?xHsd${W+^8AO*dBXN^VdI=<=jbk6bdF{dXpc6X*s#^WqNy^e=^T)T*=
z%T(QGvT~bk{r*Q|;6Bq`djWWRT>S5!Ra<|D?vz7Pm~cjiQM<AKThm}vgkI28x-*}u
z{}qD?iTdP}qdOj36`0CVjti3axGjUT9<ez&f-InKV%Ej$zgH<SHLUG=-v`|%jCmA<
z8x8Bxy}ysIlgvdXQvNKx+Vcq#n=J#nEPUK+)T&1$J0&wSy*d8IIC@ge7(<GQwvDt+
z%WO7dHk)&JaL8gYCx$S19#TrgpfRQ?32oc3s#h#mOXjN?$A?ExG_Fqt>ZTKz9fNo<
z&QQ25oGA&iQuZSWlZo&*|JGmPSAYFys752Y*bzg(d7C>WyNhz>NAF8du}8mV=pco~
zIa9`tC`Y@*Xz1DwNex{mT)no(_3PIV5&q3@{tHf)Er0k&-{C87y~&r~c%56XzQo>@
zOE~AafBz0YdgmP$i<b94y~FWx&h?jH;@bW;*RQ|CqPa^|8oFR9iZSDzf{=uAw9VJP
z`ZlYjvI~81wl&4T2#H$<=Gu(d)v^?jsFgBe;^<_?!NCcyy?%?!m-l)3#!HmrF{i5&
zu4fTi5r|1xEKcc*lq*wb@L4roD{tuH5(_p0$&ym89STD!r)O#9WCt&SP+N@DSW#^$
zOM<SZSFFtZ7UaoqLxds_yurjsYy)>bc)-28AF;pZv8z+|_io~)0TYR)qKn|fVxxf|
z6xNe$(6U@vt=?w2ZOK1_X4~LQ$eM2n1%tiJkx@57pFyaMf-d;X-|7FdMJhIJlx*2Z
zO0j~^bt*+v!%0sY-N<Y{NDP9_(kUW|rqwc4&N@U4&J~PCV?-1{8nTeF5yU9K(EJkx
zIXdnZq{Da(9&T1muHccc!<d0-e3s^Y0jSoSv0#!>@EJ6DV=@_Gtf%fGT@=P;#dv!g
zTa<*w972oFb#I%+lI6UH5U4WK*rg66axz;&N>pV<L<ph7d580!qVj_fa5<_NS0ju-
zU9U(X5*lUeVveQjVy7C4MiWii5C};K$<l_z*MIt}j3zs@5zSae$0mjj_2+E5IG?`@
z?92hr`uI5i9T&x5t%LIco}X^oW*d)7V9NCySNW;0LTEi7e>kISv<9`-W(6WcnYtdt
zS+mKSD05?+WqW(X&Ffcq<<`qwxw_AITtTiVW2~VnNBBZ$gQIO%tX3-)izTP03yuzF
zoE+7(wZRy#_Z>A$T&zPxHS6ZHCC5iIPEQvsm#PWbi?x0hpgZ3-0p58qe;%AYE?MHU
zzU^@UuQ$9mOJhCCKt0-?$J{h9cpdsPqjzaqYcT6&L63JuJ&6kB`t31)nvs6L0!+eV
zshgJN@iFCtigHpi8khLOQ+SVcp46XbJ<u{BME#zmyRv#i)}^_6xnC|=2Z|^1#bp{&
z@(n0kY*oaduUd6Zx@x0m^2cYOYuAk3P*tU84_PyL)@bZ=|A{dsmdl3I(^D3+8S~kk
z#bU{FxzJE4NjY}E0i%q02!X|7&cWdU4-X!2|KK4<r^nQ*=84AjsX*OiYsAU%DaMR)
zg@6KUDok2oMJTI|U;8U>^DDpl(@eKU7~@GOC~-r_Jf1=7V`hwX8?<Jw5$7yl`{`fe
zyMOX~JiLFKli7;#M7T8?vA@61_VzYk|K{673a($j&gI=LDyIS7Aq48hjJtPk69iuW
z;+MI7cuLs?Ce@gCzV}^Tefb7ETVo~@%Y3OWnaN~}yZ0ZmzkiDxH(nm%)Gk_3y&HBj
z#kr4t0}EDY68y>czQ_9?-Qj!he4np=<uxYb3Dvk_ekioTQf1F-5;&dB=(<3kqTe%t
zX3$29Yi1Kg`d%2SJ3NmK-<2iaILhoujGZci#}skKBQX-XM(w;9z<Qi3aMFv4f;GbN
z@sf`}_=NfFnC<P$6kdq2QyGsW8q6ND_T6UxZcy%=Vdpl4CyflCt&i^*j?TOZk$S-A
zvegDAY5KSmw`R)nMq=?fBVMzml&upCYE9-*SH@0yPB5#Uh{g~`%n;)UR%4Hh)%YX8
z`Gof&%W|kpjs^v3`g^}U^_oR!Ep=sSqh^BHe7JO>%NkzKzS*z!|02$5Ws)eO@ia-R
zE_m;WKv8<ydPQ5Wu)fHDH+bJ?vuo{6gJ7M}4DZ}`-dR#n(B?QT>pkVT!j}cE?AbKY
za&Fe<m?@!cTUN^@tJR9Gty$F#U3O0=V~C<PaJwWlQMh{jCSUr>*J%S<rY@%8@L9)6
zYTI;sZZap&qn!tKo8VxxT^w9*&Yxq?wK-?l-I)@D>UGwu6*0CPK4>{PS|O>VuoDsm
zU7F)vO(*qGk;S4_0W@nGoQJJx$<=GS+_<^V)^0^n3Q20}dQMqQG$g~Qa7+vhp%t3C
z<M{ZLgNH{f7Y)WtFm#X_jM36#s%2`C<cYy>e6-}~aKVimfi6TU(O61vJ(~yiCh#=I
z<Z_F;uZi@)YONm7?>9ovN|Ea{Z@9<Lj?}roH(jg!c6C<JwwEO`vNpnd?xzn$xu`vl
ze5fAINa{g~?w1Pa%n&*@T(GQ--yRZ?3B9xNglaTfXK@jvZCM>1b2=(Cw8A+JQL+vh
z8Io}%tCCTD2zd`QhqYkWK-(}7s*mZM-1oU$T~7`vgO?}h`o(JyxbfKLc%3y`G<Mk#
z*KZjBrOk6guQP_ka#?eDc*Mhp4_VIU#4hIXb#zfZoQ3zK6qwCVx&PokcW!^m{Rj6r
zJU(L8)O0ZrVtiDs;W6#0K;7&}5DCN>D6GXAkFz6E7*Tr9x4!izfBCQeJX>2^ITJ>3
z$&n-s@eLaU+#{Rju<7+?1M7t-*6{MHU*uQ++JC}-{g3|75D)N46ckkzdwVK=9PzyJ
z(#y2-QzW|$j6m0}NHKEx%2lpiyT;%C8-JDA=_$v@$4qSC>Xl0@ngzSN6>Vq<p`$ED
z96kvA2fzGHs%o4Wvm#%!&OqH9B0^kO#o+9<_wgoGxx?T7>)#+n_;3Enf6M**hwSg~
z^Ua@moA3PAZ?mWulwPS@0anX}G7~8gLN@?KYb@f=9z18h?>T6d_04-n={+{P{~97O
zu7-q%6a<@u7%i^wst+bPUE3IqbajTsV#(3T90ba8L{SJ`RF=l&tl1cJZl&bU0BB)t
zLDK8dS(9Zh%#bSHTX`ykny0p8xfNN1x!#|eL%Q?d8|+NhGBWfG-^L~&Ldve(M9K$k
z%<l2L>4^+3%q01W%-S**>4lUX&KKAaK(~o?tfb5Lmij};_F<E`WLvJ9nS*IlUNg@n
z6Eb#a0Q~y!a^54Ies_>ATi05W4vEa?b1nB(FR{R6GD2cfjlI$l9VC;81fm3-H&o*h
zbcU{0cb+nICB7<gF4F*3w{T5D(yAU^+Y&lWGHKgZ<#r4hYw2P~*9F=xQa2I7@r`f(
zEX8=LfNpSpreWjWC=dA<Ky~)t&)mn^Kw#thF8a(f&pp2-eeorwL^-P1-k#82X{cBG
zH1!Ja52@>#`Dsl|9*HBIF_}$nhOAecS!$aXQ{gOC<r$3%%2HhkK^nT?X<COdp0?;T
zOF0OL8VcS!yth;%Pf=Q$m9p5@rU6WeXokNegQS9bm6*?3mdlQ&35+KJtO~#zc{C|+
z&BP=&6SK~+pyvSi@ZB59XXgRkS<v=qyk6dA&OW}OOL!)}E`5&J@bh5mj5#x}3~SGs
z8TjOW4o1MSzS^GT+c1tcjOncPKw&Krs22;;eJDmHrtsiBV`Fh*iA4#e&wH(*kJa1!
zF3bjO9QrAPsKMyj1ZcgeckV-T=0=Bkv6dY@-=Ep-+Ou&RUdDW6(2r>UM~`Vz?C83V
z`FzFk(FsRK$INF7QWq(#Cw3jN>4;rN8(S931)tphn0Md%5ue`ul-c5x5F&|0Q7AF=
zq<{{d*8MtlEKV#U9oCI7CSoZuByMawe&y%C#?E$yH7ZhxwOAL3T~>y+dgyBWoyJA1
z-C8+oHmCk>T1%1-FzY-&_sf43MEK2r_J1Qa$6VgqqcjDT8#5kFv9=<V5xcX`>hy$@
z(>vG@b%V0%794xabbp8ay<IL}xy=5hZ5FFJRW)IIQZYN3<0R15HLtz)7XRVj`Fr?c
zL`?0ug{re6!;bZ$G&-eXcC1_DFwQcbOzENmlwbJypXKEEkmdY<uYKujeE$#slzFpc
zcN(zC5UtRLCCkMTt|-sQlo;19wuA;SL&`N1)rE)tes=FG7_G@@Ooe7ZL?i|}P_gA?
zNJel;@84(FNWx-`rnifgAv<k4shEtX7#~@lM2e!Kk7emXP^&}|;hZdsT4ozEI9YxK
z$gaoCZ1sV!Mwg|eegXIHgpKhd68rr?@`fS_F%ENPsjG!&vX)T44+}BdtvS0YFZF05
zJ>5UJbEC455lxCPIx#RVi$wRviy{k55}dVk`8@B917n0>EMofX=H3F;&wIiIsn2X8
z3G1|fd-mLURb@{>=u$@v+4hQh-M+Adq_u|uy4Vt9Af^tF!4#UU){mzzJ;_0|5ig$P
zg`zAdDzAXn*<9pWGwme1vHOhIwrOdbhPKo1>Bqcj8dgnB+X;f><yT(g#?6;$JJEaS
z{YdB<(e<mxP|w!%t#`Sck8xWc6hCnf=YD<^&^<2mRozH`Ml>C8I$I$)VuG?7vAe&^
zYi~NNo6t4|chZc-JgClOVYK>5X1(;Xa!f0I;JHPTrRx%l#Y#&@m7Y+x#G)liLrk^S
zeRTm}2tL4gV!3>E%1f_YX0=S*`Sb({Y4|Q-EeT<bGRSq^GCN(cm@la7nixxpvKZ=;
z_HA?gw93)CkDu1n`MAvF&x*f%oZEKu`6rF1d@ka{kT=$>ySYY9>%CRT?p6)=8RFdp
zhV-3&DQv{xB>^H>r_x+q-LXoiEVibMrc-=X;zuJ)S!s-<#4}(io5#cTSRu#3^`0<0
z>`@y2*TX@TE$QFdpBs94YjD@|^J035oPHv28Uf@OzWfiH6E2JjGHc65f@l^v5mFae
z&09_mPB?h@klA9cAzK)hwI)3Dp**+m-QjnC_uumFhd)*+tel(RG8fUb>8V%!Q-Qi~
zeDh8I`JIQPHegB(%yuLuM!3;E#N7J?|JG}`D&<3#bF9&NQp<)i|9OkyqRrBv3$gSS
zzw+yUo!znF!N=d{(xrU>x~`?@V7xu0?Pi2#PUseRX9(>|L%+ljLrbg|RBnn`j~kEa
zCR2`%4v1YOG?5gcZiQ7R{Ad5m{}oqn+#pHBCL!5J^{Yf=&_vU_WH53>`m}GIJg28~
zio&xsop5@5K-YFEs%b6rc_QpNq9kRzQe-wiWqbD$fL0{f2&6o}DfO&jzu?_-NR(((
zP^QUS3QCo3k%YB1O<fPH7fCc#yo-<*TZoY^1iUS2>W<I_j2WpYvVpE{DO{moDJdH>
zxVJa@a**xV_$__>q3sv8tgZ6QDjLb<>CiG_YI)`CJQE2vsrFJ15fU*Q@qxWSrT(~q
z2_u&hl!4(%{aCZkuwt0g7OW+Oq|7F{tXl7;w-)C$<4Z7u5T(tQfzhh+dPPRxd(rVD
z=tdu(!)D*Jz1BKHP+1{csFmIki4w9@fXeT5S<lXTsHSC}kD4%IOronhymcC5ipCJO
z(ox5jIyQ(+6y8%-1+H{CW(ADfFp`o=<D@RIIGwRt=mFk#4PDm~VivDPWrUI#rdvCF
z?Hh07%MsR9L~-;cf-(syXY1+Mp9cWvJqVw7C_hbmDnQ4hJi_yC@4tVW-R&)PkvmID
z7VlK(`P%hFQ#)c@u$Xm3A~~8K=oy|~l1k%!o4V!nbj5@Fhq>HeMOipXuSpqIRgqGQ
z<dTLSp@=Y=IQB13x%FD2Ya?ybaB?^&Wh25`hjWmU0{*6h)v9JTpR-u3Xq%3ZnLp&w
zF&aES=e!)}U-KkD?$NP3`|nL5`17CtdF`TjzO&DN?niun%fQ$hr`{j>P@bHTXVlBq
zu}(~LCM#2*j)T60o^F9MfTd)?nu?er)(E2<KC!rW2q958$9OV@(FB~wj7qRE8_t0k
zEJh$$PvOx0gT5FlIo?>v@iG}jf(qBmP=3sqwdSd8xN!4^jU?5QH90x6XDZcB$*z7(
zj_X!Xm`M5j$=*sTR<nkk=|tV7jp4!FBOW|>!0G8R)*H%Eq0&h{k;=sF+jsbrKm88x
zyz>r=)nWt0YBEQBhHd<*K;4gja0oKOqy)(!Sjt#)!zOkc_~_q$hg+||#qJkhN6Lhi
z^&uQ%%<*7r3t*Gk+6)doyNRBy_vx*pnC$cRFZ^Y~7hk7co@Q5FiwR2A$JLAnAH7R8
z86%=`N6vUkS8)6Or`(&4c=OFKDDXFi6auTol6tu$v@Ieqo?hpF{-6G@`Rd!>QnpH5
zPMA3hrg{+4Zv-*sQentiSUG1ajz%NOt$nP4&^2sLw^SKFMuHKV#t@<<l89n+Y)j2#
zt525T%!btedq~J&&FXCS^|Q3DYYC<5lv(HSWl5Dapkh?lpQMB_35nfMSpjc|O(K}Y
zXgbDPPmE+`c(PoEFdB{U&SBMIOHY~GoCtB|J~ESw?PFbf7lF}Zdc&7y6cJ_XEwjIw
z!L(u&(Akba$aM_?(<fb^_mE_8hkDbqWw|bFR$te2{9d@EpA#9{3F=-Mp4L~YwX2(m
zwSsl|`_Z^M(J(YTW{9U#9XfIhj)tfgMWJk1jJck&(53LqV;)l?gn;)h&tXNqQs$bP
zF=<kW?yN`(NR}TmICYJNR$XK!kz6FGg+ok28yo6Q8MQ7&oU^#n<4T7sTn7DX5#X*1
zq}UL;K)tM4E|){O!NH}L7_{r!&|!-SFTMOS*KgcJ6ZQ+bPBOQUe~U0A0N^eJa*t!`
zwEbi!Yfth$F`(-Yp=V#tk8OYW`|q%~yTy1sVtZ!`?+cnb&@~Zft;R^&1V*#9>7%N8
zH(1KE(s&G>Ha5&oSAY<LP?lq=(o?v|Xf#qroIsLR4Q?xp#sw)=gf1b`5xYdwsM^5E
z;ffF<en^Z;xl2^Ha4}!7SS+b)O_qqM#GA9n+opWhMTrRK$N#g&XR~e2iSy5IilJ|Q
z&y%*tv31WQ%tH6Y&42g*RdAtutd`1CaPufm?ss6!5Ku5=t_ofaY!MLkP>2=U`JB@S
zhxq9hj7M5lvGlqpT9-?Pt*=q5PPjowH1$keU#lyh3sy25lb(&E2O7iOouecn$M5$y
z()(M0LD^z0VWMG0(vt|r;C;brR#R4EnpI1RhP$^PaQoxCG+~8vo)}wJ%Z_Lgr^`8i
z`0a1=!yo>ThX)U+>pFj%Z2UI-q<afIZJDs+CD@2dNf%h?c*V5bbPgva#~&T??!WoB
zeB;_4zPyGr>Z<Ztvu)@{tK~D0b#b2FXKjsURP}yrWek+@v(90CL9&i&dKoW~OM7!3
z-uZ~da)uE&J-*NFkKSkGER`>a$xu$mIPd7%NbDl-z4tC<RdW5MS1{HS+pGa2!fP*I
z<;snMzyF{A7yQ&Wel~aTT7}wfx&TF8kTGT#tYj^vZ1MDqH>5yJlH<V~WMNk^nSDzd
zC|<Wo>DC0%3Nb?4)U@>yi4kYDKIq`GCgYfmA8QRRGhgmG#MVg)q+ki?=UMNl%n0wi
zN_EsVA*O-V3aJAlxj?g8(<uiZyP_n-MALRm%Tf>TlyFAy)=-vZuIiu+mg1nECfB$E
zE4}WV87e?v%6{vC%*+j%L^k)67(_aK8V4p4yH!r)(KtXs6Huh5<V4v=BYKRf5cB3{
z-zW0)7NN<us%jL8fX%{_##oHYfE)#N&RRmsu^UOX!Fr>>Vi22fzF>3b?S;C@Mv_P(
z#NG{jam~x8ZC%&Wb$*!V#`H|0WEZbsaC&G8#H>q~PzGLX()hU)5fV`%ZOkA~AP|W*
z;R=Vh9$%=-)z}<d?VZM{CN+{&RlTEGZ3+!<GNRhAZ3xtD!1<D^S8wvwuf0uKj_IPK
zt{cRj?aMsFhw<KIgW)V&^4P;Q|7UPR{sc?zvkvIZxwhWu&wtsouk%0p!*|))-DWiM
zZ0}5|MipC=9lF*qJ8h_&R=pL@P#8z(&Ua<+dd;c;cw$J*PHRGUKv|9$RgSI6RV1mg
z0$Mqw3ki{sKhIDUCEGg%*KP=_x}{qQ5|24PTqq-KO<t#hcLgD|%;yVEPiM^ME1J3^
zZ2|sKFm(}d8;tMI8Sl>n;r&;gf88T7;~UMKW8Oc@eS7|W&)RyIF~D#Ynfc261;%6s
zPS&7peSM$v1u3pUU(%40)CKCrij%`b?DiD5v&CdG!H&n6qSO+@lGK(K$%#9?DE-+?
z)lJjtY*mx>6Dgm`>r3zV_WV9)9vSap9&E(On~elW$=C#ey6#8`_Ac#n_mjJP^x>x*
z+&@r>ENf`H7VC9S{qU2I`Su_G5$}KWL9PxVgC5<Pi+;xTbQL84E1sBI{QNGj#(TU{
z3agmo<wD}_x4*~bmv3<E*MEh$wN0N6+-8k4?|o)GbORVzE;fs@uZy$WtQ?JXsxM_-
z&dw`vTQ```?^8#^>FFu;e8$0@`%JdC*njCJUEOhXbi%{A`nK=hy-f%$JKGbE4h|^F
zilVGIK7PQWo>7j+{K7B&I&b{cU*Zej{CQ#$o#Uc)?@ZpgB~J<ic<0r%dd6}!q&^T+
zG@~&{2BlbK2dv38HdAI>5^aRG6AI}kwx~<Etu;Wmw|G!MfQVK!7~<$M_)WR;#B;Fh
zEkKGI8<!bC@4S`^3y~BRj0`r8h`Loei*W^5XxkR&9Mz~KrJAN`nM`#aG_@wM80#2~
zG~K<YzEclWV#0bi9NrpFLuTwODr{^pMiUndeO6l}bX}gj`Z+18$Sp?bq|Fq*!@3f&
z>v%cS*9VL#*EZC+&txNQYIA=(oOLR`j92zfz!`_nNe)r8j}p2`h~z36kW={A3~vbP
zLR3~x8Bn5X4-WB!CIkDvVn_zJrgs-V4>l)*uh=E5FR`}F%H|CQEqOO&JsOAe$|9x^
zvUZ|gBV>KWqR_re$|zbK1)ic%H=HXxUFb-X^nH>@Q6Neds}5=m)HQWelSDyd7uqbC
zoAsQUmf5Ofd-n=o``XWO`SMF73e^=PVRQVc#(ZWb*Qx}OGwWC|Z9R)2+1T&d@%{kl
zDT4E>rDulwPsxl3>nz*8gzI_cZ}VVzBz})RVzXzav(X{uxd8d2qnKlTI(%`gRUW3h
zcwZvPv04SXF44uPhR()&huAs{h}C@)k2TPB4P97r`;+@jClyyN?XucSG}Ayz4M-%!
zK-)DqV=2lJDeCz!-I{Ra$|c%YJfSYBS9fTeHWL8Zi&K<zVaak)Gdo?fSk$y_xDarh
z2UtCzd)_SC`R_YFuA6NW)Lr!Y$F=8x>9h|SqJkMxB>EU-!&z!ha_k*)VgH`_Y}BjZ
zjKvx~I|`?27@>>Ik54d@F@Cy*8Bd@nDV)cYCCOxuj&Ve%!s~@ZtX%AU>S5_7%eMJs
zAC)E>kM*l@6Re$Yn@SdEjR8m3HEd0{d2s)bH*UQ}3c|ZT{xQeLr?icpD<O9M7SH=1
zyw7j_&TsS4r=JjF_NGWSDf9mRJk~KhC8$&Rhlp#BdBq&^C3`|;9JfW&*@Glx8Ts%x
z{w=%Ludx5kZ(<9tlx3zt&y+P{U7NE9b(<YH1S%_9!TJ(k6hkJY7=tZeLhgM?QFTnq
z5$T4-S;wmGD92m8dF3S#!{z;ZT)BLSqoap8doD3Mol%Z!4o(h;De&qSzQ$W$`WDw-
zeoN1N8dMlI52k2VBPgxj$E@{06w^-tn>}}t7<J=txOLWM%FIoU!;3kyuS<qyD{MQ|
zN{N~*5V}5&E(Zu_ExWaanFt#equsc!XKhwBI!GZ9$z@NRvR1w*DMyuFSKFv1Zw#bF
z>>34T$zYOaG%BgelI3Df8yZr|E*J?QniW))1%)qg&JjY$_oIu_Wfo^p6z>@tAco8Y
zT4e^wafK-bLfd5^n1xZ3)8vVm7@H|63Q-IZM-)pJG@h|2E8QZZN21m(#etNTR48CE
zpa}yhL2MMvIgj-UYK*m5tGa2TIVPHsXHcfWSe!4Y>xQn=8oQ#3CJLIO28H*`B}TH6
zy6W@A4GXV#ja(2$^b<9MBPBhgy%~|{h_THX-WHq!+M=k`XlpfNT+2vlHG!_{NH#K*
z{L8U-Ap}Ayfjn28#VZ#IpfWx|V+#cZi%r*1*ELPkV7<w6zN3pRO}nCQmN}@nWV&;W
zOP6nw22X(28BRHlS6CZi{b4Y;5%XivAN>7zZT9r~HD?~SD7en_gfhg#KhH8UkL&l1
zL;XDHQ}SZHrq_{7&pzm}8`DuBIO?S)xGd)joNZKkB3WWI&}RhZ@?kw&ETZM`gn|^y
zoFo%<o{OQbJ5G+5%uZWk=jd9;V!kA$$a1;l<m3cvJd^1*TU$F6uA*&3W0qXOcsyZe
zdz<5Unv*o_VE9^FkW%90WX9}tLE9wiRm-TX@ZM(ifH0gX{T}JEu=Qr2TWimFl{RYt
zUj)uJ9~+Xo9y5MV>iX^9|2X+9ea%lc&gZ{#b0VGr>ceXoz=ql~GVC#ry9PrBs0pkS
zNE*XxU7_+&1VYoXJUZcY>mkK-s?u4$Q16~|5S8UmNzVuK7=tz2U99Qyv+c|^n$16C
z{m=zW>f;#R{_)a}n9Y~mxc(AlQSrOK|9i}4GbIs>&~;iFXIanq;SYbnzy00c;iFGK
zCT1e1&otKhw>?OHKFKdn3F;gbD{gR?H;O<ZUcnXthoLpl2#g*c^X|X+Exvx`8k1LF
z){a4S%WPF?oKUU;u(^N+>(=C~w*k~yupZ+JoG<VLiIOOc>RtJ=V!Xe{&b4dYdH*B6
z^V@&Gm%sitP1WJ?G)<(Q1&;5X^4;%!mv8>uPxI!Nzrd<l62;<N#amzbS#G}mQ`+%u
zE>LdlfTS5}f#yv*U#B%@$TaOi9eG1DM(=%)<mkOdtgV@t-Dd&AvQ11zdYC9E)WflN
zuQ_KKDj`62_gc*$Olg>OgNyhXn-_ykLKmVQ*4B}-`h4$Fbwx+hHKZ=66a~5wSXGoJ
z#g?)fQ;rHw=P}1yg+T(uSfQ#sWp>GKQr_0-eGhS!PJyv(EILcViN?1{)az7f6v6uu
zF+hw45wEOk(wd}A1BeAzVSGjDDr{+0->eNfZW56caYO_|+tjR9OKf2&3s2#+Hm>!(
zTR72$q(D<5#$++yWawgCFJ^LG)n3}fse1x64B2Y>baum<wYxvYYy>1W<2{VIz)JlI
z+puYa8cx>YT|p90k6x^CcwbUgn)qU^BgTTRTW7smYl$|Jf`GLYxpGE`T?Tk6)e};p
zC@KI=erSWbYj8w~St?~k(`p8_%NH0^uK(J0IqpuVstMPw-k=(d2SDtsU)$<^0@tQH
z^BFLa*<kDXvF#bY$M)|AKzN+H?X%o)XP<jc2fS1us$^j3Fycv}&?KKev><7A;ZSC;
z50jC;4|=Xg4R-x3ix5>`u!*b|iKaG$&eFApqoWxge((`@?>@x2immA`d%Jr~rxTpl
zZ0;DqTCaMA#;PHhTd&_4W1_BGRx34-L<}JWdXf8nUpWt&H-CSg8~kxe73*<$bWAs2
zv-$7geLOn-F1p4?zxT11`oyhg2`#zQWgf;sA<D3kNThp{;-~(qFS`2tUb2@B^0}#o
zw_q?yYks%$CG&$Ls;zCR$%M&hgeyI%D3PdgM6v5~0$m*T)Ae_(o8&m=o-tmZgPY9y
z*?aGITYcY}Ea>}X7|%=~<ilLCOo~V_T;99PcmL#j{OE^2VsB@UWm*zrq%2E9?D+oo
zzsGO<)^Bj{;I4WYvf-#&pLR$@`>g%qY1>nQy1i+^mtMVz``(WzTR}ucX{|9tYcSE!
zVo0&%;K!fx!EgT#U)lZJnCX^gp0*ukY#`T-HAW{CSpuUcfF%Pt$%nW#sy=PVR(psv
zZ9~&G8jFM@N==l=q#SYM*6TQ5P>#2`{OXq|-H6aG8Q=McWjo_%fBu`i`Sq`{bK@lz
z@Bf%20p~~T?q4T~MMA5Vf@~u6Jn&@jAPF&9n=Z>P^d1P<WYH`(Y1yzdl7}0MI^~^N
z#IWitT_q}Ok`(-CDWr%o0c!%*6gj23cSTyPQRXn&Uf(V8959_`IcBgFvq)vOyC{mR
z{$G}auF*Q0MpPp=*U~gtLf2vaB-eM89OSOCbj}Ec6{eF)^~?}0+m!v=HgCo~n9~J!
zb`vf=>LQL>GOA}jMJys?jR_2n*jZBaq+qj{>J~v9U1#W8N0Tf~gf^xuX4~tHN34l>
zV{nyVOMx!33JGTd-Xxr9DO{wm5l4qJoq|m(TK>%H!LmuE#iVZKa$Mm<LGl6V-CH^_
zh#@e@L}+ce!Jz0}c&TUT*1Os>wf=xob}eOQ3!E*qhI0^tXf0=BEQK%7z-leo#t>7e
zEKN!|VZ~<mq;agJcQi2Cd#|Fxs}*hAQMdvm(u6fbmn31G#-OC6v6;hjNwCg?^<2Ju
zm7*wVJJr^*))IP`VSkY4|A@(l*iYDPXWV6**8LOp^Ti<GvEUAQ>~qiB1c%^<F|pQZ
z2#NGw6;#SWHY+8=CaY<!A*MdAR1d*k`5~p0YswnxI<Q)Hgb+D+IODx{@AC0ScQKTV
zs$(XT2TUd-CexDfxI)^3*{o$%*E#zihRQ_7k&-1Pp<Xr2=S!Mevue9+WEMq{4G%k9
z%M<m<#`rvLjLyb9K00>i*}C&>Sj!u~pXa!KRvXyk{_8f#u8Y{^{za1AhMZ&i&NpY2
z`;NKVlu4@bNlYM#(6ua&ju}rUOvhuwc!V!Y(nP-_5{aEx#flhksQ|l<!`CCCKS$TU
z`*q&;t;Pdxtow})NE@HA=^+FLO~OqOZgbet>G6UufB7qX{L!bp|LzA=RmI(VciG?D
zXSrO^v^DR(|0Dk3+ke2lgL{Z1t&Ej&(pgG0O*eG@S#Q9`g>KlLPdY1}3e^40|Iyd^
z##i=v=l{Oqqu=~}e2f{GnSlWf5)eE^Bi#O-@3Q~$E%tuy8(3G8qL8pStKx%wXQj~6
z@31|sZ=KW11g@Y<iNpIRJiL9tgNJuHIDEkI(Gj!7jHc_jx###F{PnjPPgabo2^e8_
zXPfpk5e{d>aKhr?fRm&9q*mCybcM-uO1-Rkcz8$?h23kfQ&bbWw#h7;#sj$x^sYj+
zqMKV14X)1%zPAv1<K?V14(810Y-H_8KY5gOshdPQGE{}AH6U6(Ndtt}BD}_IF8&<z
zWqJvj=Tlh#&`f8fFH<ovi-UxVb-2P)RuvE+Mon4}6`8fRAan|Nwzj4?KcZ<mQs{J&
zquOTUQHl2|5FB&7*d_zjmtm2<4~Mu8K~jX2kkF|fTom2t@w9cq+A)?bV)S&iC3Kcn
zDh?K~42ESJS+*VXHZpHIR&At>F~?ge$TS4JF>H?uF7HgR2;K_gv1e!Om`=tx7Fby@
zE-mBA<1EC~A+e(<oVxEbP!*9u=dUPA#3jT!jZ15T*4Y(NK%CjQUfJAPO)!W}y)jVF
z!F10pg`n4l5QC%h0h^O=Bx%-jAIKcKTrI+?E*bPn{gVIP8y|-3Y0$ruX1F<9;Jvz*
z+qR?4?(z^LT^Cr@iwxQl&O5yKG;K@Wwk(%RR&_(njHmMj$#}N5cXK#LHV(xMo2r3a
zl|WfLT_-=e^|N08d2hfBXWdUSBA><9ZR*rLZ*V6BMEVL0eewurN;PT5Y8LDrzOoM*
z@4XuRVT!uAjT(kAX>H!7X<05-9Q$KJ>^M9)=JaGq*El44x;imGU0}>9MG2!(fj1>>
zEz~Oog4XyAAf(|oQ7lc<GM_D3E^A^)Y|2>xkChmCq%Y3*`{x3>{_mfB4Sp7z_av|X
zNt#85>M@Fx$+A$oHu^lYO})mH1gyHh*D&dE^KUKdqM7jtE2?FRu~f#YaeREjY&2qP
zI>C-gq7nRPq~Dv*oPoJqf3zk><ZBdT)uY$HAWuT|Hk|3WGyUf`M1{}yx%Uc>`g;f=
zFqur4&yTrw<vOW@JD=XAt{d96V`qC?%YTtLJbK8tzx{1K{Pe@Y-P)hgB56=N=x^qH
zn1_C~B?RU6tTlxAWb6E?K;19=($_J*<_mx2*EzU-hweuoV2sL)#2lv(QY3ir;*ryu
z_x{zt=IhsY7`^%?;!4CgO?t3?$X>McH8ZpBOW`oyVWi-}y(2z&??XQN@O|EW???RT
zy&rM!!F}fQIc;b`3~$}s=a;_qGCMmZ<7z_bN^I#E?RY|cLQ^m3!U|lV*s2)qOp%nh
z|LGl0W;5ypZ-4FQ=n@#(VHP~%S7&pPn_)0yb2FD**gdT^*EIfaZD9<Cz0CpUDP=I$
zCrf}Y)U}`i^f5CL`W{O^nKLL(0j+2ucr;V>aqfbrZ4$GKRAgAbMl{x|3*E4WtasSx
z30eJCM3uE2IVLSdrjrqcuV~wvE+$g4ARQK=@-|l?$y9pGav=kFl>syNi7Gk@iREI3
zA!5Xl#1Nx0b0HbJuAnM+5Vkp<Meg5QaC}@dOO7Kdmev!*5p7Awh}IZFX)wOf%qU5u
z*g@)WBFx&rgHKk_HCQ@EBVm8LWN+KDw`G}DW7^QtMi>=|(mR|b>!@`&XEzuF!C3*P
z_oi|!tVN?`ZOQlXT`Iz%azSi7r}7-nXR}{6D<<P{E^Fopq0<d6BWmX~S;_kX?<>65
zDl0B$_d2IlL5vZ^dc4sRH+`nKy6tox^#Y=+dPTizhxnbkty!(q)g5C{0qkUH+m5<v
zST0vARx5SKI%SLnM>U=>nQRea!rEf!XLbF^#^&8y_i61-y6|w?pfS$;ogW$EjdWMG
zR_hMi?5g>!c7CjWGWYBAG2Vohl(PqQPwNd$n;*~K3344+y@UQvGW)tdWQSN%iJ0tm
zubU-{#aye7ICY&L9v-t;)EG*P8AB?GS#Pmv8kWl%ZyY2o3$_-f?}e<b3fv}7*LKWi
zb52j^tX4HUJ0-A=O?#TP@mYO-@oU#-#^c<(pVu~D_wRzOOOkM?k+}Y9HQqPNkZp?g
z=2yVE$1ZN`4^h@n=^vMkTQ4e)6c;O2$Hy!WwlL!o>n+Y0Y*}J_8qS{n6Ng#X8xZ;P
zWTU^&^xmcl|0WgL&)YtP>H;7gQ_}e*D#sL)uv{#8^VXMm_ebw>cyLS#YUDIcLs=A@
z%#Qi?xBrljKK)<_tx~D4{);!SZgVZ`Z|BW`8%Cp&QB~oM!&t+EgQL&s^QQ!LJKMV~
zSBc{KRle}oe}y04yNk?gd|!IUSc1e{?GXga{G<DP_%Htrum9bB%>ETjX6lSphR;}s
zctxQq`{0$gm4Z({c)&Y9`aXa3Cx66m{N}&l?t@RWe8YNN9FDqcnNGKvoz5v;#m;n}
zaWy3s5sN2SOPiFK-Jk4{By#uO9gdC;Ihq~w>YLwWfB&iqC0e6QBE9R0fy(6^`<Ad-
z7ELpT(uT__Wr5&9^?zgh^l<KDt8@~2=d&z_p^h+05TbroNP#2)O-C`g@5t<f0xky{
zdzF=NCB?NB_6#jZLQp|iLRc><lSnQ2!mAY~o)A3AXbc2KvlyjqnHB|w11Te06UXss
z!(!#AiV0<T#C9d@mX^v}5&;r8ylw$WG&SBh#28X87nINhi4<BSEj7?S6@*X_V@Vgs
zbYY9h&Mj`=tNG;qDG!c2Ou5S<j!C16Sq1a1Wz`0Pw0!AHU*zRmZ}9#HAM)e(-lwz`
z-j|fsn9!`ackqzw`}@dv#A-Ijk~of$MN{(L;St+qVykF*^QCRxdUcPFKluT-Ztmc1
zjjvV&mXvNp>N*NvP!)x`l#;`g7FU3Axz=$sNECwT#$rS>$$Iy5&t$3uP(DcdpJ!~V
z)rxM~;#`S!8n8^y#))PgVlB?C<D7cXqZzs`OIC<hKe0tF-=(6^dLHODpAvMTWj33$
zS~ZA32!YjV$#S)1Ro6L|($R%L8#<aswd(4o!CDOiFgUs(>`r&6##3U_TC@F{?E~O@
zkesQfjk|t4+pzw<`9pvHoq1*dZw)iMc>SF}yg%zO|9mj<e44RsuC}oTb;fuFa7K*@
zJ(>dRqcyV}8+WP_R7PEG1QOjKysmY67c7@ElBg?nHb13aHKa6!Xi=FfpB42Jl3J3{
zfR8iAP0tz`(PRxAAvDZq3r<f?na$?x?~ib<T;HQvdsd#5v3q=?!lP_m@4a|j!2N~&
z{uoB?-vv`=3`ovKUytCX`saoha1+FdoQY|^xaBBtvvz#;@s}Ek6!Acak+52^I)c-2
zMI04W<C4mIN@ugXvM=Y>|6a$v4Oh{x>A|yOHdgND7@9M`dk+wOE%8TO;L1RYB)l&;
zI5=c?cOP)P|K5k39?$67PWRv(s`AmtAMnE;zr*R`WcVgydn4pxW^<5TF+y1uv~A01
zRPx3fuk+H4o0QIzlJL*|yMO*U{r;4o&N!o`dz@wO3vY1aTW@pwH~s);gkD`sBdH|+
z7kp>9_a{GO^7?CB`Aa`T*xI9-OtU4BlOT-67HdfptmW|DjCa2O1OEGe{r}}({+oYI
z)7E4u|2!B5XSEugn1FLI9#6S*?+*5Q!Kj$x@QlY>AO<TQSB#ji=G?n?m*dl8=Bqig
zWz9EV`oj7U8eE&&_tGQ!bFL4SBgy|p!=3t=m%eUkN+}Dl>g!Tpx*%_Z#^fN)Oiz37
zxO)8x_0fG|V#t(9lJ(2P<ph$fJ8e<P4M{^*VvJy79}Dm_qkC~nMza%5rs`8lDmF{f
zh3yQ+X%?LKCEf>Bkgygj38GLI2Af(OhRMY9;7B-~3s?3ECKE$kz%*w-cBnfAXR)Hd
zkbKW#lExAmNMeZzQV>Sfh;F&0Z4K6JBjys3U4q%;559ZK{o}}zDaEC$%!BZtnXzgb
z?AUPg3omo?#w)a~@IU#V{bfFUH*oFNRcfg@K0c-@3;whJ{6FU5gC+m!H~uAm>u>%V
zFI~UPKmLFHfB5LbPkDIn7&qP`QqimyEJDMl-?_s_pPuqpfAQy7)klQpA%5I&`RXNR
zCns!|6(%IwZiy>BcuTaFU=7{~#%V$f&T0&!>Dl0On=E>eIFswfJ|`FKd@*OWy~B9q
zXxoOumns+=B4t@*cbmieg0dWewMdN0z;@lhj42zJ!J8zajTloP2wmt1L6d;$PTk1P
zIhM<srm1PVj<(a+lw}E41K;QKIg8~&89MJNiUR9B#yaX{&Gj2E5fe<OThw)fF(z-4
zjh186-Wv=n5r^3IfqcfDO3%E>BV8-PAb6@>zfOj-y~x*Rw2NY``oA}`C^tca>inGr
z6K7)jo}pQ%tT)EwK%+i&tJemT!R;#2i$|P4EH&2FnhX*Idd*0K>M1UlD-I8jbF8#s
zK3mc>9fd=UwHViIoGCQMRIhL_ko&z;WB7X4l_7*kUDqs@E0)U@^{UGP+J1<sGlq@B
z_2TxVUXIPzY`XByKK}G9-SZmL&ueaAtRYhrx$;hb_-%~$1|XLFyaKz=FgiA$H8vYM
z`uRq(=Bo%cCPJ-jolM81{axI6LQz&Zem1UIew#K};ht>SWZ%xV^L@E_Fa5tae{Z(l
zaPHqz6b1M1-{-IV+F#?x-~W)K!((Dd)J;oOs;=kJ;Q`<I&UZLEKFYPTbKlxttM}1(
zPHVyYECpmBMwR5*-{0j+U-}}~E?=RoTk58H{9tahrvr5+$D>#<<%sLQ{EN&!dC202
zA7p>8#@9%a%x(!@ATB!Y{Lb$)xqgN1OII<nhs7bz4;}AaF~k~$I`N}--sRhW^t=4a
z-~1Oep#>v&@3{H@W9~nnEKAZfKkVo3M_hg5hK|gP)S0DacXhSy>7MQBnb{EwfXiJ%
zEEfO`FvBMr{{!$7`5(}@B#i(`U=6m06{8m2JH0cltGjenR#s}A89J^$(%t#N&$;(T
zRAyCHSB@FHEGi<eJjah8_t&5Id7o!x#W_FZxl8XDO&AKa$yi!mMJUPL2Y2zk3QL{k
zBDEqMqJUs`XP3M8@6sO*NOMiCHBW19g`n^XG?vMrb=p+Z9ev4zo1OgKfN&-KZUC4`
z+VsF+GG>yhaqn?|tBD?;!U~vQTI61L-=TE5BmEmaMf)7mAOSrYa1+5=DKUCF2&I}K
zmHw7ZD&=fkaQp_5yCR2Da>VK81^B*C=m(fwJO8@^aWghY=Q&1X2oX@L)$n|uet*DR
zqe-n=Mb{L*QVzy0_&H@DV6g~QAn2kjAaL9ar3J$H1T+{%K5<edtQ;dw=efISc(9i;
zwu@AkRv9KSm$x?<nS|P0jeq<f|HoXo@(zFcAN~RT9(;W93O8<!SXf?RoQx12jFTZ>
z`RaFATVKVij5v03jj*mr<&cH-kY`{07XR+w{)}U*>)g6@leiDf+A?uiVRO{yU;gj{
z&!0QN^1>-X)8h8MI~<u`V-)RzuhAx^64nvI2kT?C+nEXygwm+M?QADi;I^E8OKBT9
z;d@9aC$1x3wC(KfPz^(t7d^sy1#2zi$ayC#rCh)%W+q2n6shDz)v=wTIeG5n3k*|)
zL`YGf;tusQg>;WAlo%5y5pfbv#+Wvm;b=q@MT|xxCvj6a3rOi9g@-kkuv%mN_(_ED
z9B<QDeBbf#l#)}omqXn&Mem9sJ-OeR{4U=!o#a0&Z2M$8yl^~>7(Ywq|HNX1C;-Ok
zeDqm}&f_+nih@H`24QZaS}yb;3Trrnu{pVk-Db0}jFVQNtOwR3GZL#Edg_XNa)+#2
z*y~0J(Wg?YkVYAUzCnnJ3(`dlvf%X@7i3C&S3N08!TO@IRut%lkO-m4GN<oojG((e
zWYCWojbiHc8lDP^Ihq1p76%ZYqK9W*<72IHK}0d9KRnPoIA+T6HG5rVQ?Q3M@nFVm
zW+uRQID!DB9*Ml0CveTBD1&wTkilFI_?hb0hoxXYm19t>q@t|SFLO~25vEB0A}mJd
z#KR%!-Y$BfjjGr1!U`ft3&$Mii)yVmy)T&-;uI3_lr$(Nzo+eC>F~+%JnZNX&9Q^Q
zfFKB{Rs$|wx<Z~7>JGWZ`p!t;`i*N`xq1n0Go%uvIxmzC@=t(tC4e-wQnhI999TS0
zvAn!YtJxw;a|VL}X_|gYUr*cT4(bpNArcwLxp|I#`>X74Y?F5TSUV%GS|YVd;3J6c
zZ*ceh512c94q<Sn-9?_b)ZG$75l3S-?(DGnV3VJ}@l!ILI<ZwWCr_;N%v0xh@16J9
z-5oj|H)Dw#0eNnzRO_rBIf~YX3m<<#v)Z85?BMx=ERMOqai429uhAb3@%##vTAkEL
zHnw`4ZMFz}r!SXVjnT#a(iG=JaZO4KC}vY9WtAU=71qpdPm~nNa?+Sd#}+%z)M;xF
z_yJ4HE5yA%<K6pHabJ^9%<aoGn>B_!a{*Hs{FT***2>9%`Jzoog(!B_3J+rqc|I-T
zC{aRHpmm||cGqdK+XspOT%kf|K~$-P^!q)cg?aoiAgtDqo{!E97AHh(tkX4fDLsL;
z8ly8ON^6yy^fL*WM}O!e><OUC#&(6xZq9>!kDalP*P7#_>zmlR&$(A$ArWKVc;}Za
z9AD+Ee2(!*vcBG?5<bT-fB6ntTh5$Y;K+qlc6K&7e)0$xF5F;kZJCoNPl8lPzd~zn
zfiJ)Q3Y~=`RD%Y88~iQHtq#{NUS^Pts5k0(Cgsf=8=Oj)IJve!VBo>dE{n}M2IEaS
z?JCh|OkhLDNtEFP5jLu@W71Q8#Y7BKmZ^Ie+!&)!##NL(iDrRrw~Oxw2&G7KCt1|2
z*U`!?@KWM*0P`ubG#iLa0~QwqI9*0j1cdHew9ZJ=lsJwUM?;3g5u@>dBu$;wQxUwT
z(U>Gj$nqSErCw`NsZ>a^6sdfKP>kb<dTRkC11yT5Qo&QwIkM!LLl|ecRTIuoVG2K4
zNDmzQClQweJN=myO*!wCoVw}4!-4l5=;w!|T4(jsKG~<UpZH{hI%~ybf-hyi^W3FQ
z9XHX5X*(Wl@mgso{>KifR2aM>m~+7wg=7R+^4u`!kFlnY-;Z4(%s9bh7Eei(&<J7O
zd#o&S8f9u9Ku+e{@-dg17f9`)wQ~(hl8iLf6HCHLA&dpAoNV-G+bqg<Sex#X2=2k_
zaQJg(KX3N2!;e3*&7?UWHHklwYhDKLcIs;Y|3Co0I<;Rb&#+v*X4kax@E&n|#pN=B
zf>Pu@#)xTAr7T97qiP5;V@SpkS+_^NyGPJy5mc+lN~J)q%5}52?xoZivvIF{XrIsC
zPlum(L*e0t5~8pnD{uANx9{-ED=%~7#$C3zb`U_8B&Z@&F&GTEcI_J4I_av3$nZ>`
z0F!%Mh=SCh4S7*~CW>3D(`j=0^l8A->vtIq2ZhN0XSdH4)KT8c1`?YS_#X4mJjL4U
zw;28Smzd0PEOJrUDj~2k#}gWv3wGYWMDOc&SvmR?Ixwi>1cfS6Zn@>!#jA{mF?Tj@
zvc11WqtQU?j1#Ah^ZP&eKEL?oFWBq%NV2lL9)^RA_by!H*=L@o-EOnIwgysi>Ela0
z`0yRN`}?F(1X56MHL125gfc`(&HGnxaAWgd^3AV*ov(cPWooq$Sq@fdib~D`G%Y3r
z<4PEvvVJk5@={>6*z}ab#BOk?!mYsIyeW$Gs1*dtXL)gn|J^_Ohe+vj;`A9Vee^#6
z<A49ph_l=wUxj>%Lk_{@hCEKm)6`8snRdE9LQje*z@YQ&b6wttH6shdV<`}xhr$$)
zxCHA2b8~XE%#gm1D!5jVPpcpcC|?OdwNjy4t+KN-WIT>&_#Vm&kji(dtir3;3V~Gu
zD-2TMc4|69YXe3hG}zq3WIl1yL91o<`kGr?W46YEAwI*vM^$}}oPLITyAL@2;z_JB
z-09!t!sSamckT?=uHE6<)w_&F5u2MkWSQk>Ki%QVmCI<I^Kbv^-!YDcbmr!mUzq3Y
znX?S~0-bfZa;4AW@(Q<aZSvArUf}8Tr&&95lsA6*Glsi6+`n~;YNy8KZkHIvxs_w&
zxg^`W%aKl-VKl%~9(k-$(x3uILx2>GE9p0=i6$Mm2}((~8A|hPiU3Ff`r&DW&)eGV
zHfa>IwY^1>WOO?7wCWAU(HLtCzL%qvN3L8stJ_d0a-F#yX^C><HXpKZ>TqbD6DJ-=
zW5(k#qw$E|V8A#Ux@y)anoNe7(IiPi6h&xb@qLd<SX9LoRjn>qbSQbd-J#xW<AqgM
zrHjS)19F`c#|i$d2BL%KuogTaCK5phV7Zw#Db^`gylJrZu-8$1hlBIV7~)Z6{f7mQ
z4}Bd6Q-e>GQ+FuvC_mre9C4;PUO05DT=t7<NrxmB73-AzsZW3tMGAI|5HWeKu}0v_
zph&mc>4_dB8_fJYIm*dnQd&IaVT=<GkE4V%$;i^_X1MJCNpM!^*tkXXU?4WTN%=GZ
z;D`0$!%(`<>hIZq|2FB`-?nAe5=sgL*nOVCA#0)ts+Gn0V^6-=1!qQ-N%bORJ^)})
z40!jq5w=*rN|gr35thcy5e^Mjm@G#R67t<1daj3QH8Axm+9YU`7yRmiN36jX)D$BO
zwsh1f`equWP79;T<2!)jiK)PNiI|n?e1$~kj+;L04e0N8Sz4Iq7e9ZKEKiU~(mcjf
z5-VYAcZ0#GPqk7ZO2=T`XBu7dk==I(et=REW3q|GqjAN@C<&Enz$>r3%*hic+1Pl%
zI2zI&>^rFYG1}(~>f9+iI5f|ZL5QsS9R1p>>|MRZ_Vs&A;M`)cNUQKv;TO9%q<iHC
z`Ij`F3MR;(Kwxvj?)E0vuU?_uSYm&7k9svAtc9$fJjFlxC;usb`Impm*5)=zTzIJh
zfs$kz{$Kz7pR>E&=XbvIbyioGP+^0oUVN1k=N&id{+*k}0wYPYF&IUEIOhHLFLCY8
zuefvf9@|@8zV*$ovb59!@&hyWBz06QqQ(?@O@fIo+Jr}^HA0$+;H3f5TsTJtrjs(l
zl}$Gq4Njjv&BFXL&AA0Sop~;O{1N-zT~eLlYY(fZ6}5Tp7Q4K#Uds!l&K*~Vm8B8-
z7Z2*J5Jjz<(;1UqfLEk~@~pHu(<D)X)y@XaS00Iy=zKQ)Y_VD+q))9@LkP)u9ASNh
zXF*6GiE`2z+PG9=ant7H7?U|yi_AbS5E{^uJg<_Q74B{e?so?a@yMH1Vr}TBLzY(-
z_`yH?2LJl6e#rZmKIV6S_sjfe|H=QBy`3?aE?(n<_ph?Q-z6Hy42MHHa|`V4?@dy6
zckXTw1U|j)19o=%T)c1<KMZKLT6k|uPM<%+@^XilUw(@FyIpLk_}cIOJ>GruZB|d6
z;p)fNiF#vpW1sx?nDwOwi}h6=?B8OpqUbbyXJ9T<hlogtl?E#%Rs>9U6bBddM=c7E
ztHJeYt<h+ZD8+CvWOsMlp=NU_l^`GsLVUl1?*{}yh%wfQyiPm}i-lcUtkn)twSru0
zlC%^i&WYkN<8j1zJYqEJGa8SG<CrLl+?X<kG<PXo3j|?+?*}O5V-OBqD;7>|>_nW_
zT7e%_z-Y28!t)&$#@c*>f=VkViw;F<$jNU^N!=x4Gws{bHVG#|B!DvI^N6|DZD_;<
zO(+A4+3WSN0PT_F{9+M2{QQ%+=2Nty#AT*|tSy#qAxdxS0)ex#n8OPwqX>4(z;255
zI053~EYLa!q)TsFHC4N1_V?`gE+3z|j!d4h25U-dNe`1}#BoL(r=)3Ca6HX~Pgh=7
zB@g!hq29n<?<q<CFhcmS_dKfL;ZfJ~(*n8A3fiRn;?x=~9cBi&AlJIhmlclkA#EWZ
zEF9GbQ8@Uxe98oeEw5I|zC@H*FGZn*TPv+_5>d_}!g>8V%9BC}A5VCg)L^!IB#j-C
zW);~8@X|U#-bCvRscLw_*%TV7!3X16cYhmqjINrs098tUs}OFSmN#+HzfJ%+hazic
zw4+HmcNCXZ$g`aH-+7mn)fMtYv$MGgXeYQYG`Y=@O7Zcvj}d_c53)E%NQE|LLKQI<
zYPA|+;FBcrM7{|Do^r*Ob8}6edg?s?`9J;NbN$*4u3ou}@+G;6Ie1Szx_!=|u0V>4
zlJhcv^XKO{`mI;k-`Pd&4hp=|U0jSKaw}mVmvpb+CQTw&Ea4R;^jNN5x<IWeFnWX(
zi&Vo3XHFic(g^r3{;U6zZZ~3cW1s!KAv$+~pr<4i-$$d_9*p>({`{}`mw)wRs<qIG
zSJrBDI&&<{FR;ASq0?$1jp4!OfNQsRxOn9{cOLAJ=9(Y<_<vz<f0tfwm+yV|+Z;c>
zhLonTq!AO|(u7xsa76?FS%%I^!D4srC8thEL2d=L#tQXD`_QB;CaJV6%dp1cc_ByE
zPxJSF|4;dwzx{XY?(Y!>D-+AG*&>XRhg4eNIHa&}ez2A-%O7zaAHQiVx;Q6Kx*{26
zeX!2m00Tznq=jIv5(=q2JkKLbi=8yWRq<vS_^w1-d4ecOF{z*ycuwlZ2u!BQgdrDT
zq(xifbl!x)8UdNY<Q|v+t7~MY&ei*4HisdJUn33$V_$RXnKS%vzVj*<Z`@~OxA?<9
z{RaCxLn<=l%7q)e`}QSvce+UB5k(Qs5g{N+K*<^)h{h@H_5%GupK7&E96^?(s5oLc
z?2_q}oxKMfSzp1*h}E@|T)cXX=bwF<@Bi^P_{|%)$*%VpBMCf@sJ}z1Bf?W{YI2#)
z?p>r1G(v&2sWY%Q3ZbkkJ`q*NPtyleLe7celZ(aU61+dFwOUPVqmHk9y4^0@+uL+^
z_gI|oPzeGmVU=pNPPJM`2vuk(7E0;T$&`pfpxfk5H!#<lG)u^`j4VqTjD|%(H=;ip
z630>DCXqOOKCRK#A(NFb#1BJPDQ_&5Fm!Yc>46X=iL)r%?{<0f%{Mu9{zZ--KSt<P
zNunV}=Y&B>o+U-l;R4N)yXi!A#l#2lVSPBeIF|_ABv(?zRMRPU_Wh3vRAyc=bKPcT
zkUnue9}!bL3Djn?L>D|(YfG)jQYxnqjYc}M`s91vLCx~Arh0Y7Qa^KiC2}t&Rkfzr
z6nI6N>{0zR8`w<)Mwb$H+-E60w8=@6jPWQcgsjtI!~1CDZ1%pI`frvfPkEjtBKhb(
ze4G?J97OaEKR)~W)7uxn<|^>eiAGs)g)tM{?cv`b4vuD9-UbgFlg0C8YTjk+-6p|F
zloW+`6WF^z(X=HUMvV6MQ9CtkyG~_(&K1lQY7}Lejwv=8R+vdj_i%d@I$k!TXEYbf
zjUv-qEery-Ha7YA;w7Gc{&|MI0mI$^teIGsSz+1R-(zEQgYBJ7k|Y8gZ+N2TXDqE|
zi+ZijV9+B@BDYB@Xp>4=!ivwUufD{e{K=oNyu8Fu{`M!NS;}zKC5|HwZ)hIXK3`C0
zoqK@`+MG68xno^=>S@+ryu;w9?;uQpNMX=OtPlh!JRwQ$ZZqE6q}rTENkiZ{cL~o|
zT)%vQ&5b*B+DBPlS|(+g*S`8?jvhI}om+2l|IU5#G^bYekYxq3hBVit+Tf`iUkS!Z
z&R%zf5PJyjI+P$U1rr5HmXl<qUA(*Csm^%wy*DR{_TT^hcQ|)$y+FV8WS1(HDoDEu
zVV&=xG!|i8m5#JU)v9s9kkNu_y~D!F8bbQTXO$4Y#TA!&#l)hK!1DYGU;5J5$hGF$
z)k{>Wy&`yWoID{*cLhL93A&SAKTaSvPovLuc^}r&Bq75`35hP45a0?=@GUthI?u_o
z%o#W<g{LGzSixvb66KByB@9T7P=+uJX*3$_?hL4h3Q5&fv*zwKR=M(CQz9tVlBXJz
z0lAMc71FGN=XbcfopE)mk8K?#Rhn)(BKHLU!FRvT#aowo>V@aHb@MLgPOWffXM;EY
z-COKF=+f(ti7nKdHFTCa??I)AW3a}@XdETLh9r>_%=A2GTcfOp5|(isqf_ft#c$rC
zJ>O<waha3x89w;wHQssq3PwvBOLK@gCXK+GTj0H`w|QoLj*7_WCu6J`&~6ywtdB%c
z3)|#KqzGJ9wQ#}tEV?uG#+?QLvt!_kv>*)emGiL`O0vDR&0sKK(C<>IgjB0FnvFKC
z)*PPa#%>sf4w+X1&vRqb7=y_)I?FNEkfbST=0t<L-F=4R5m6j7jz(m8hR)qsaj}4h
z=lgiRvj!AMw_)%l)@YJE#S23WlH3@C7m{R}Tet4A(@ojm-{;)9Q#9&Tbe^Ggfl3I9
z$#`aCRqoyo^>NuAg|?J|jg<~Dnk08-K5qibKgJq7OSV58(3OkL?D2fwl-;K_Azct=
zjGMP76}YBc?A@X)3#lScP7%uK?^3SDDH@dXZGn%LXE`}$r5xgdo=u9_$!k6t!IlBv
z^xC<!mODNrJ&ctkNycC>DnwyjQ0L8v-OqlW<4YY>EGUuM!`I-m?yb*4<YxbV)U|uq
zYajk)pL!^d)xug0tH+3P8869W<yVxO7N++^S^GK(c0cVnigY7Th@KVJal+ASoSTDd
zqR7=c<t$A!7{cxzYQ9ZlaS@#)=xV{^E<aWl_a4OtLQKEUKJJZYT{t<*t1FxD`}De9
zu3fvvU@)N7ZnC$x2NvfapBqrFo~768((CnzqOq&>!xkx4=ixrznP+iviC)hIXx3^c
z#U$PDdcDST&p*c>|H&Wm`s-ieZ~o>d+`M(8IDA{UPy8Kl>MS6nDxhmS`4I?Yt;YIm
zFLVFuU9x+7U@b<1HWp)q<60q!_Iqq!{FtRP=kdG>_*I1Rkg7te1*7qpZf}P$tZ?$g
zDPDT%Wvmw5x_OiJ^<!*pZ8+oHJSQ*wIfZcP&@{`z9wbY;iPDf6fi9#$N+{S}_h}%M
z<vABV{(vCxP@bSt`8`e?Uq|Z<i$*DjtQp|~RW}JTK`+4}6)1&~lH7nFR#;j+is$)G
zZe#|gcb=Ng2wQ87v6?hV5eQb7S9$GAUqk1bo42k~3j(TPfr^w7yO@$L%M@(6FywiT
zegf&&Buxp03+DZ@`cdEnq?JWlgYujMfd%JpY7El%@B^Pb^CsGYqB!R;aDhQ)x<rYg
z7FN;Lk!7VdXkkf(3y#4cEhrff`XQs(BE1fIUgOq-h+DfB-CV*p>+BA8IsL-()R)?L
z^#%NJ0bjMT`8NOZpZ^7Suihcvk7#%eYMzf(g3)M96vZw$F&5<sGy>oE@VtQWIHB2S
zQ>j!*(~K}^lc$NBgmi+=9e*)5Ii3`3ZQkXFKl(SEI&%)Ml5zITdFGoP>}bSjr^Ag8
zFM({Z8#g(z(BjVhcbKn4c!8u=$w>1YGqR{k8(&qiHu=N?QzFLGwBYQ-`-Lw#(hrbQ
z5LPS9FU}*BV87zCyS2mi{yutVhcFDO)f%+hZR+(pI@bs(sD#cLBrB)^$~uP%TclI_
zgFd}M-$|Jy38V3tBy~j)NI?(;gkgvmR`9$KC6!B;7Ab#ExytXb>XTvV_6JxMlIKvZ
z&Y@J3-JM;A{XWB9kMrlxu(&vn=PUBu33batK1geg$uq2ws3Hh0SB!~<W{J)fh|^?Q
zE9QP<XN6IVInmB6$YMH6PTQ2*;}DM$(r|*kQ$$D<+Htc1=S;CjIlr05o9sXJnhRCb
z*|>s;l5XPuT^w~W%%&~`#o}olQf4Pb2vanwgmc_{>k6T4o)v<(UTI|D9xw10w?G?8
zzTU%;PxrS{1!`l9`$7xyJR^$7WSPrixc=hMx-k8W&kE{3&qnF+YYzwD<zutI4?q5x
z-}E_N`|#G7pCio;b`-noU39f6(vueJbU}@)d@jHD;5j}5KDNLF1aPktqFDcmf(ECx
zTEGOwz>z|rEM!pvy&<)JpL`rcmLX=*Iw`PH6iAsYq?100E`AnL2Zh^3F;=B}2x~3z
zIO6`j`&_(u5l?z}Hej#YWjGvT3UvyFlV$66d*~uFCWXWbgD)kC(KMTNR@aspjYf2P
z`($~BGn3Nzo@8xph4W9H=LbLd0nb189PhsO4sX2iOO%o%NkSNev|26F<nc0ZpPN&s
z3W#iWz)Jx+ig0a}^)EfcmF=HGCU8QMV2s3*&T>m<Id|WBm#4n>U4ljnDJ2yW-u|0^
zOLJikD-4$|eaP9<FCr`Rq={3YPgBkM`cb6R?Cx%pCNB3K7QTLvk!b)s1GT%!G0zQo
zksT=00#=9vCFqq(#qrS6ln*Yv$MVt=VNl^u|K#^sSnL#&fa7|VlltsLP-1>U2!wL#
z_fkn#R*w;eAx2N3{lf!IYdx_-iVC?F<w=&8kMim(uM<TvH?Dn5wd!*8h#6CDAtpgx
znUXy?L4C2NGec?}A>Sys=t|E=Dt9x0j}5I`e6%4m2AgXto{#7GghA+#yrSSkN=sM?
zsW%&dU>sXIZHX4HkVFV~t&FkgVskHyz-AIeKpb0wTAN5TxYOTZ=r^bzJ;E<OyvVoz
z-tX|O-~BE>`r9{na6iJ6tNi507r1ocBko<l$<g^G;(SDu^k{TCSg8p@53ds7DJO>M
zd6p<jN%EM%C}&||ku=R1#Y1Y<25DrlB0&0*iuA~{lxoH2{JArH>pQP8io5LYZ1M+x
z^aG;kcUWF&lSDD2os8~-KKGt|j<lchvw!<T=I1-iufIa?-Y-x)U5>3fG%!w5D!xJH
z8s85<x%7M)v>WRjC4`(V4zu=gU$iMt5rzR-mV$z|XcJTeJms;sw@0trV=x}Fv$u=6
z?a*1rTeJASkLP)XMx!P0JOU}nGbbE88jTo^#|1T^kT_5hKMe4~fFKBpb9UNuQg{gC
z2-}{Nc*3GFXq}L`aynEHk{E~*O|JXQ&o861l#f38h`qf%o`3FHmX_ve&o%LVpFB&5
z<HVsi$|Fq@e1AF>9$uzOBKdUl|4B5j1PTs3^r;s^6!S<Cv>kZuByV-#Sfrc^@0Oo$
z%k0S)Y(hw+Cnst1GGG$|hdh|F0-Iv~ap_nk1qh4KIo3#wow{6GBZ>{QC^p-&a4--P
z6i+JYD!h&3+6keYmo-cRyK-EY9Jf+h!6DWKf7uh{Iw8+fnyn^b=;L{6!YvNN0Ar@I
zB3e&<H)s1(Ddi|PMJl@ZrXryIq~@7>wyd6hII1_%UuBBMPU#9$TF-3ztUyvcfotAb
zS!R@JhH(m+K@@g+QaH(!G6-{DScKPy$M-`Hb8;p#V40*y@Xv4*r7gT*Cm@BHxC!Y*
z!#E=8kI>^Wq<Ine6&0gWxMP%4Kya4FIX!^SKsyD+{Zq)?6xTS<GP>O^ySqC)c<_LQ
z&H}z4kY$>FZ$PcqKuU+`Xsy}X-6hMMVnMDAN=b6<T$G+Vf1W6gxqbW2B*W<Wf@;-g
zb#<8+UwVNred$X){q)leh6DccFaMIA?H%g%CUG2*Br$oWiK4#)ojONUEHu?-E~Zi*
z(khH!WBHkLJh<>N`Soq2^*}hJ4&`)@0t@}?x9Dwc(O5Z-k0!tN4#}lA`CttG+#DA^
zy3YBtFJp7X&ekrqmgU>u{w}@l9)I>1TQph?_I7&+6v|U>w}BI%w#Cj3Yo|e+l+F{>
z&Q_ruyt-MI(r&dF8cUjGy!p;=ICA6|fBoaX<sba%-zN-}OHCLTz|5MJyFsN21$#Hp
zb1a?C0%6r5JVLuVAC#Qz49gq4MAodeMDdU;OAxY(K(V@dlrMehb%ui;!(ops%|K24
zHD{5(!>PC@5!6ZHs@Sa7<VGMwUN{>RyDp{h2tYtqh{f7G!_z*#l6ZlSpGvgRXloG0
zd3e@pHLQSf0v3f-{zT8nrR^sgsGub_m&Cb1$Oijkk9#8+s6~dV!Ij&02|F#?tE=4F
z+(Ao4Pz(6ifB8dpc6JGsVt#3!-l$91_7Hha?hiP&zCxqZMhF-VhYW`UhQlsa#>_9x
zQ>nBWkH+XEL3tiXiSaeY7nq7-u3e*AtJ7%uoPYK>XP-XK^4hZ`<Ct!*N0!0O8xf%g
z(l^wX>TF*aQVW`V?@#|BZ~x7Y@Y)@M<{5T(KOj&!Yb$kpWx+t2jH%XZ2<sz+n4qyG
zLTidysU$If8Y%npG~e?Jp<b8nuhnW)DizR<exN*$;c&=cFkmzqF^(gWUf1zLXU3j2
z7Ed~acQ#$DvCd-5cic;-{Zz%Tgp)Lx6#`^|1PfIx2yQ}DQjwXg5Cs+_QOvcg_vrR6
z^6Yb8!DvCV*(9h`xPSjXQ9R=8+0&dqdyZDCffOD_fae8(C5}^OxkRqb_M?2rW3H7M
zpV@%#vr^%QuhU_^-t4CDV4Agz|E8R*DWdkp+JT^{IMo9IO!=Y$O)HPh7NF&XcWfoO
z76r;yd^WQ^S4x8vh!|4@a`HuD=AB~pSf<}yJ|U&W*bJTJNM&fWDl9CvnV)Y_t5pet
zz&TDF3dSB)sN#^vsWZf*P__flGeyj2uj8lv%)aK+ugjyJ|Khv1g(Is3Nlw-ull6y%
zjMq%=N=#Abk~3+`qL|M-gfhUglOWu!8Fqq-8A0LSDoUUliF1jPT4SP|bTA_64M~S1
z%v^%4Yo{L}Bt}lkJSGPt$}fTmYW!)(XbWjnAuPEuXe<~P)D1^{?%cgicW;li<y9(`
z3jJOm-}5KOYXOsM2E9I+cD9+mr|<)ptvGe^IF-QT=Joq{5`+g8U(#;3IC5l-XP<qR
zSHJWM=g*%<O35#N@e6vr9$B6dClPs`F&GSpqL|_E@#s{aGpNh)ya3;;f>qdpXe@;X
z0#q1MnP1}Qv*)>Se-BcD6`IUKfUEBC4J3mhgBy47UwD;v_YR|v-(Y#T$ECDNG?e(B
z<m%-QIeYFbJNGwP3OtS+J<a*2p2MKn{r~<yX*8Sc?ex%@CO6pxg13`Yjb%2Z`Lvyo
zOP^|Vny|XOz}EH-y?&o}-+!CA#yoGm^&VgS%1dsKPz*+lc0mYW<y3eW>lNfnL%q=?
z&vI^Dy@44<tQ<W;eW5eSv*)=}$%eA{L^B-i@nGXNt>y}T=o171$Bv!i)h~aQk3V{g
zB*_qga`(zi>bKlw&hm1eL{R6D7BNMMazjdr)~ZlxcLK$p3@}CdC@(5<J>`P#!1u|t
z3+i;9Bcve=T$(O1Ik|QAaG6QTO-!nsM2tbWU^uhLL|`(g)EBw4CAhJZ(({+NKg>x~
zK$6G&=H0jX!5{t}&%Shyi&rj?nmuG?KyEET%|m$!Qc1>x0sqJU(|^OWFP~%3A7YFl
z(J?o!-s1A*E7Ynr<~vJNt1W){i(j+1+eb=QrF!DX5}hOSEG#xSd*%qsOUDpEv*}XP
z`@1<i+hacb;0l*7UST|PZVB~<$MerV&8cV3lMDs<z%qaQDem9?nDg^blk|t|?cJdo
zcr;rUt5PhQ(O45Ge@5F#V5M8g9kMYQ#h*J>^r+?)#`nSsJn12<Kx1gNS|9|T@6%{D
z>CDd)MG?_>Ot;q~iXz72vEwlrLz-t~){;ww(b_pc1OdM10>errBnU&7a;;R5-pqX@
zri*|S4&^TrLWRqUK?;X#4#o*LZ{FnE^-b>F-Q~rXzQof{KSd?15LPNgQOw(Ky~obh
zE>AsmjuXd^qhv@lPEkrxueHhYn1iXl$6GXK7ybi<6ra)_mJ)g#eq8bC?t99Gs|<X;
za4Tg>igTKz*=CQ`Qr0+5xJ}HQcVQdmKBEX)tT>c<E<dN}=gG?J5JFKpkuZ%NN+hft
z-{s*5QSg?uL5Q4MJz!z6$@=;VE2|5%+AXTp>SQB7>kz=qIUP*3PJ`6Lf%oL-AC}S+
zZhf15{iA*k(kFh4_Qh{;02&K($Z^cL*Qb}J=y5U$xE(5I3s;1a4sa++`i$dnL0XZ5
zEd%pnTp+}x&_EPOXc_P+sqhi#+!A*OB)vY_I3kTBDh0pY@%>6PyRfRWhwiCz7s%m_
zMe)3F1zqL58b=Xf7?NZ$qw$dK?M<#;zC@bFR4Nrzp$Jf^R7sPZJU6bo-t!m@hj`K{
zad@6b;QKV2O_rAyxq9se!_g2S1)e8aSeWC~sgpeQ^wT`^%rl%ncb-mXo~u``@atdy
zhT(8jY>=Vf>l4Qb){24`{MqgEi4Uh4Uag7m`40V>MZ=20ij*`@9cSVE2E&hTF_{32
z!NbA2zE0V_dV%ofam3CC%rW51a)ayl`&`(HdH&p4E?@nSV@FrXGC{39U{Kj5O*6jt
zy&ur+4S4_kx40Qx<<|Av<V9dKg<&Z!_UtZWx)VDLBfGb`bEws-96x@PyLaz#>-G)y
zp4sBvcR%3#*%Nd+O&3UsvP9cWa!x6`(61}2#V9PhfW^){V>xgGXFO(mZ<mGn1&ncv
z@<jSX8%E<1w{G2JX~`0VAxDm^A%tLUb)BsTw_R1B&JR?;mI2qqS>R!nbH8O%!twqj
zOwhp+F)C$0JYOL_&*=_o7r1Gou~y<qMHq&p#*n2&`q~(j@(99!%;uyySmAWntc}rD
zVWi`>ky><OFe8nmK_+0QpV80ibY+#U&JZM2>K=FQ-Q{2Y*}q_IeV(8pSUHh#etwB+
zy^64kz^fooT)lFglP?{^tB+v3OW=8|A6um}53Nqd(@#B1cYnZenDR$U-{PZ>F0s6H
zl!b)_s!L6ps~u22N*L0}u=QY*w?DYU&gK?@@;QENo%2thCC?M?-`~Jf9!Wgrt#>~}
za}`0Iq^I#lIks<Dsx)ado`IfVcV`1cPSeYgUdCvYV#7MsYITy<%&kMcOrZfcyC^-0
zKwe3O#Q6~`snAlBX^qzm5uS(iJc2MJ%X6ApMw+E8EG@Z8)<K^*iiwki@n}R6Md&m|
zOF>Yn(5Tn(0!NvsR4RmFA-r7JIJxsMAe`r{wB^F!e3rGHx;a>IT)iYEi4w9b=hp2T
z^oAqu-Mh=HufEFBqeqbzgwR~Oc7wrS#9$C{`qXKfjXKl3lH*2B7p=#?cODklecHYE
zX#wD)f`@WpHRU8{Vd}a2WYVwYPuT~bJcn{2Ezp-T$eK*NVv-7;kk6U&m9S;<th~lE
z!P)_?R@paYkahTYOtDtmqQX#9t@$i3&v9&hnPcm#EG^H|Y&wLwET(Ycu43}phXcBj
zPxop3Hy01uPy9WR@;UV0$3g6Vi>vipzVBf3>vigj9ooG`MuUOl^Cmf788lnt_>%>_
z$gQdFI+y*W9(F9{c%VdJi#bYz@XF$rB7JI$V6FgtgmC4cO1kIb{)lui#3m^^NiaIY
zSyVFFEJ$WNu$e(X<@8vg;DR2yK2Aj4kfsTFp5b|t(WuYP&L(?1n+yhB>a{BMS_Q1d
z4}8kXa9K=5p65BY2WwGMQma-#SWX;YCypa_cejwzqI2lXw>W*~6wg2ZJm=1xWBu51
z=I0&7=fQ(bd@mr+ifaABMp0{n&iT6$AI`Qmk#lX7rzR0ROm8+RC7uOSt8wzHuX5?Z
zHg-Rvte6P|BmzQdm>=CGxb+(<BE@)$b0?Na24gaBj%UC85~J~epZ@KS`S!Pdm-{#G
zp`;?0U4Hn(zu{Zo{5HLAkB>k8U=rYzqsd(;#|N;JUEp-*=_b^NPr6B(P_2eoEbHq>
z*xlLX%GHZ3c9wbn{R@2an_s~ee3G(Ud+JjnC-<5*86SM`F4@}|fgj-eA&o|p7giS8
z-P_^KH-F7{zx!RP)sT9<fmD(>j@jSeBhNLmp=h@qD$?uq&^jlMQj$bFE|72)xmxRq
z4wK{3*~vi2vb>O;`C`r5!thzT9hcVH6+TRYMi1#pJf;TkMo&x6h4N4V_}U`!l=22w
z%ISw`Akmi0n4&KQWx2bvRs*BK^ILfSBKLci&7LOl>+B~Pwi3{+dkjaL%q`TIU#Q}R
zDMDt1VZwfIkGl^dqzI{2+9+A2HyrZ+{_p-j{DVLKHnoaiXKS0??p?aweJ*``iE7PK
zt<DitJT#iYaFZ8aeu^|5@QYu)PcO3sl{#okHSjrh<Op-^1)~0#L9frogAKAY;Tzw0
znP2|u*XTTFe>>y))m!WjGSbw;W&urEXHI*(e|eLyo>`&Q(R8<a#Bol;2(Sig1wwi#
zKXgcJo?<gysI1$fP{*Z4o)nK#c|Km?VY3Wlv`e=u4{f2=a6B|W2;KOymaK3!Nz#Pz
zcubZ$-`3G!!2bR&{k?tCEG4Lfw3{u$Y8Bu2CL+S6tWO#2n6h@>7?g2Aohsx^Y@tPE
zg~DivqLl7_pW!HVV+G6Z-Y&g<m%Z*D-~Q&eId$qZxiK`GEjG8d8T7}*QO5JnKS#aZ
zWE>6g{gSJ72&pTfn}c_S8{L8;TQO}9<=Y*8_aoD*j}GV#rvl3j+!IT`QU-O8ItG|U
z*xYN&V2sJ*C1zJ{b|sET{?HiN8k}>4`(3&Q2r;enb?A;Pg4mfftN<Y`#^zM29*awD
zj;*h8{NyUDtMfElHG*RE;O>)y#S&65S)3of0sdrg#^ekS{_XlnPW-=T-}f-S-2ZA?
z)yHqusV#O0_x4D#j7$#^7U$w*tt)^MOj4-8ClQf{waIN$NP5_!#M$C(<q(sV`QZD?
z9(0z8WI5?zOg0)ro@2|RAX$WT#s$_TfKxX2_raj<a1%H^y1Tb;bMO8g&YnGuRGKsy
zGU)HIxG=|&wPi-b1Zy=(G$QaMCeJ4Ye{oW9;H4{K3<ICJR*NKyIeKi3-~8q+JY^Xr
zhWW)h=g*(!x#yna%$YN+tsbG%SwJa|jg7l>yM4m2Mi^F*QV~Tlahzg}Adb_&8&2KT
zpZ<tfR+b37kiaX1QHAq06ILQb2!#Q5``9s#y!HwYe*6n^?NXth1sTHAUpmcKUq4I5
z%kfZ*<AhpR<499+e7sAt-{XZB&vEnPpYZ;Lw|VC2r@3+QV~(CZN0tw{eft(Cj-TY|
zr=RA*#x5q#%VB^N2IU7%XjmI^W3i<HFwV;7VS~~dfw8WduH9_0wzNVZe75&CNb;DE
zFWuzTFTG5qTEU_VzLzX!2WM_y-u#~DbN;DkkV2BiG5g&<c`g{oDVHyQ>||GlW15Zj
zq{v}B9`eq6Z?dtuf#=mZed=kN%{IE=-Wg+wqdsd(uT$}zG|F^Ba&~5Uu1OO^ni-6C
z1u9>(DPaq8v_KjsRX|bnVe*vB76lN2gTzaO$6Cm8hc-$TKqir%=TfQ?JS(UK4&BS7
zKnDs*h{!cn-=kLb5Y}Qsg_>KS*Q-P24DWt$lR>ad7F5aOm}*rM<(sS>YqEag7}{Dk
zH@4^x1|&(0wyqy^lHkc6M)w&H6E;5DWw_Df>)(8l`Gp#ZKjP+%8_X}v^YNuioICd-
zl}Z~cEcJGc{lNyC8~4c~!^Vv+t>!#ytIM2OKf&qaXV~BEa;txbjT?K^&&)BjBLbnQ
zdXmM}WjfUwCst1I<DdSTtJk;aw3kWx8Dj&UU*pz(pJ$G=$@c5?M?D&`ATW|3tdd%&
zSeF(x)+v-L&z~&xRv1KK+m-1jPzmX~lxVIYPn`shP|lhy@O^~xKqzvf!B{-4Nz()`
zs1Q_Y6GUb>7|?9Bh!z*=4+a=(sD@Siz@?KNqFw|P($NWWlY?>Of1%uFLQzyo7E1JS
z5<3cn3NTg^#VKi~Cq9}|x(&hg8`tRf`n>k#ukiZoucHk#n{7seA#c3#Yj$?_c=gpU
zbL`j(Y1-!#2-PG*cyPX$O+`HnrGrnV?;h127SKH^2$<bSA3%JbNKhwcf}Vru;6q+q
z_^eKC3#YHQMc0(a=Yr+K{b7r>W140?_(kRBn_`1UvB`6v>lSh;1&wBt#ibU<Pp+_j
ze2JCS4s&xg!irzS5mWT~0ICf<%v^n1n(}bOZua;OKF>5uRD9B`Hr^-BefAs0^zY*4
znfanm9O4se!VmD8HL8mp+JhzXJST}`vN%Ipr!VV#eVx42?AcAw)lUgng)Jx()=r(;
zit#cHlnV|~;Vq6W+#vv~Ef^<n5)TFr@y}DHV$f6ISd{Y#96;kftHp85g^xa<+uh^r
zxs$Y74LoV^ec0da(rDE1{eXVIOPa;STIhmaS2&RZ`6PH>U0Fd&SYBR42urUwBFi+@
ziqE;Tr+NOl=Qwx%Jj+X~wA&q|^w`+g;-d?f=<W}2Udf6)H;hKk>$+NL7QFk%8w{T-
zsJr#oKV@z0B+XY}Km-NiVsHd_QB?Q{C%Tu1KC54Tj%<5_{=1j3B4h4&i?4k1RZcv!
zh7~Dj$G`HGA`b%QIvt)JcX{K(OFa7@{UIwSKcq81$KB0s)|NZmzVtCqpE-*k3U>B-
z42I*vpU@qj6w{m7<T(@G8O7uJ6DpyTe|o;+<f#)hYAse*7Kw&?Jazs&*KS-RNk-^c
zv%9^^+OcC;B*Nqv=>&1zVxlHvNh(RB(ZE*<e{PQD<s|~Y#@_xe8yoj%HJilIm};el
z5KggrXJ?0Iqs{ZrzKl`<r%ye_sngF8$4+h|iX(Qm?&FDQf=-qbi`%hl^4y_lMi=@^
zUmUcV?1<fo;jFxhYNd&@h!EsbIxd0dp`{|z25lkDjU(MFAL+St8dBm3iKkpeloFQQ
zT9k0g+gfXkE4(C2V3cRX)VO+cn_eC;5P|^_s$xm<Z5CGruYUbJ5AJpO_~YB;nNKvf
zRKhu|&55IkG)igJD%8RT)uzJpH9K1)Lch(;8#jsbJ#sVV^qEuidl9+T^!h#S-o4LD
zue`v@>O7~<tPw{Z&z_4|T3p5VeQw{p$%S{`=En6qT)%pY@Bi-a(Q3B|1NhnB{+jv4
zI;T#ob1S<?r?p14(cypk7k@<%R(a4JGtLc-Aml-BkL9^KGHlW7-(e*W5CP;ypo}HW
z4a&2wB2;;}bdSQ?{D~t0agtN5)~MDRcwq&d<t~8BG7J*sSE$w`nJ$WEq%4Gk(<z_R
z8iJrkquC-J51AW}8OLL+VS2xo=JnblmD`l$v8y;VHbqK-?^VDC=p1rg=vzrml4S^E
ziIbd-jcvwbC-@n7K1uFmfrON7?QZhZpZ$b;_wMlrzyJ5CR;pBM4U#0~>a`n;hB04$
z?PX4#TqDyNzVDM|&VDZleDXY-I61hP<ggd3YmXbieOhX-?30HD0M=TUdYy^iCL2lh
z#oYy<ERwJwoYkBAk}@4z24N*q=6t4wbW%)A0y;-^m)1jCA598Y7Th}PyalWqyA!Ad
z>7F?-VTsBAg-}I5fwnndP!jwA*4E}Yc6@~sCstTmS)kbrsZ@M?Uy|r-!lQF2pPU4B
z(<-c)Z+ygk^f&=``5ms?iu0rB4^akcCHh-l#AzgGjoa*E3y+54_0ExD_A^R!_>153
zxdavZc<nl^Bg^DkyTYgKeeyiVw-V2~l<T3GZrUC-?8_*(Ja1bZcabG4T<4r52uvhX
zbjg>)BCX@nCc_cKZkJ>nk;VzWc1i^pfwX4gGK_-RP0!=Bn>VlX{`>E<y1IZh8P!mc
z#xU1u@vC3_l3Ta0adhoCz1|*1y7X%h_+%+A<qBA134EW5U*XvLF>c?v!Ksrc*?jPT
zG>r)Y#naE6<K>rM;i>aav9z*Ez22Z!t6_{_YjcO&x9=c?C{nG0I8F*t^%;Np$2z0W
z8Puu$As_skf5p=kOJ!w|$Y?M=Ld=-ddJ18kXl-q-&6ywk4wv`0d1|%FD_?txS~I}t
z$Q4&;?JjZwXI5G*p4M>l<NLh(!=Lihxl>%da-ArHX1&T%dyZR|uW<6I^W<4XyVYT#
z(_(k0OP*_#nst3}a40w--crk{1bwGaz7wr{{q@&){`qIQ^6_0doi<hvXw(~=KXZ<H
z#b?~lxc|;iINts)d3(jlZJ6nf#mQI|+(H3b7_uyP5+6Dvwj)0J@IB&aK$47T-fPop
zwH%_AWn8^_m8Hc+qTz^z#VSf>bocM`n>XKNFc`4AyMr&`_rCj`1GZ7XB>h@e(3a0V
ziB#NdoA4%!R`U7MG@S&PK@i~i9!eYXJj0Qh9e>IqNwp!j5^V&^Q}_xh6#$o#RVXAm
zCU*#AmMO*wn4rQ(m#<P?evaD03ghlRD$p+Fa<0k;m*3^V{Sj#%5_okwD~nVr0nJ9h
z{6dwZYb%^uKZD5?AAWF~o7XNQJ%RG!;*|^R?cd?cuf0UAS!FO9v$!;ewTjiFOKfdE
z;NJZqYio;~IkCq1=Ud#qevdnMZeqQN)TZ3H^C3oh4B~x;ahI>W_9_pyLyjCf#`aE^
zQIv7(<^`UA`Bi@J55CKrZ(YQ%8tPObm-I&&+j|j5+w=5y1-(&9rRIthM3Gj{wdZ6#
zlt-@9gBRx!sM2rK{@MTIzhnLQ5f&E~Xf~UKVdx4QYE>$g3PBJckXTV5uR>94v<Ztb
zXtXBHbCk+l^|K$~1r_|LR`5RygjqYe2_XwE6Ci|-6dFMVZF7>uX#iSN@Nld|cy)x1
z+3WVWdgC^??rsyOhR_cfW$82#SoB4I)aT08%cTD%;hW$1HYe6k5QYJs=ds`2<8S`v
zZ|J}J0?#~i4r45y7a*nE#rwWDQ;>4FGWB54^|1TpFtS$uJ^PwRJ$3*A6k_6uJbjNl
zrYB(;g^`X!MTW2$m>h)y<E+;Z8Ywl>(*?x>kd8}>a6DR9!~z1}6}L!(HaUW{2t-_1
zr;9mQm0V<;EY{>ajKv6xKsoZeR+MCWY=#vn&1Q|0r;c#q<Qj|1a~xUi(C)Md0*7E@
z+<XFM!I<Om`M#^{wugjVC;k5^>GVfk!zr>`2;7$eub7SjOZnVvWL^e7MH=^zn?-p%
z&fC~M<Yci9MfbmD)7D~r57nqr*BxwbhzBD~oT9rU(9Zk(V0<*gL!9PVKhea2zh}@&
zW3j@auj0U~%5yKWU&S`cmx?q=7!8L+<1t|z5#}k%m_o0$P-n1?t6j2UOV^#*D(r`~
z_FS7FP$=b!7~1Uy!{I(xFJ0xm_uiu4+voJjGYp3#;$%#<7LsKNS>}RNDZyAxXKs#h
zyiBXvq*|>oH`nIF3zt|}Xt27v!q>j~RZg8b&FbnZwMK(5tWv8r$Wjf3WSbBaT6BV5
zuTL~i@Vx-7Kc73)=L+gla<V(OxcC=;$%)_p9(rMk42|@s)jU>7JRlhjs5fe~*N^bc
zKmIx=y$InOq%dUTj7GgpY%^?<fi_qSUKo&=jBlLo@aNaA(+$qj-G4x>(qwb<0gYyZ
zcCEpMci-X2sZ;#k_x~PQ7IEQ&_u1XrbHccu!U{{ZRsk#M569SCQ|>m)9RfJbH(QV&
zJrAy3yMgaD2m+slg*mYB%rnoCL=ky5W_alW{cI2a+1Ch9oW}%n#8OWtTv5oCP*{Ul
zYp^=Unw*{OO>W(~N|Hxpnc(u3i?rG;_PhJUY09O`SNZV6i~L0XoIm)%@AA~ylWc8V
z=l<O*AO(3oLis+%YOD<klBg_-7Mxy_F%FF^HS3<l0`GVyuK!%WP0>N8?N##lQc6Cf
z6DfAykl^~;_Z=sTTw_wV2pFBAq@YpvQ9|HJ@PtHJi8dCYJ)nm4=D58%K!gn{^KI^q
zcFDb*`h0`VQk!dcu7M{we)c4b3+v1^msyx!;K-2<I%`m`Y5c%a394jC%$HhE@$ze@
zx$wbHL1@1Ht#9&R<2vn5mF?{<a;>?2=Q;}ut8|u^IDT@Sjg32ap@;M=K}B=+xiyZT
zn&Z-y8+5n(oOx;$L5=yP8oS&3><@MbssT4{-QdjWlO)+LS8v?p$jK8dEH<b$EbaM_
zWIrXz1A@vNdxMCzP8+YX!1mrvn$0?mriUo1kFzvI2EH?%o~eX7Ks0{>*Zg1q`CkzR
z3eOLygaOrR4L=MBLXUdA&fMHw@zbF*H%F~rr(Um9uhpp4YWTj7kRCzBC!j(egaxXo
zirsA~YXE+pljY8%(`F9&&JE<c@Rt-mzE=Z$Mv?R9-Q3w=dwYw!_cz(x=`k22SRl&M
z$pU07NIUgw9mOLqUA;t>=e+vLmw5K+XOK!T9F3@nkaypCm+@%C3okxTtyTprjYgC4
zcvx&?O0=#lPAL2S(*x;;6+%1;`FYrD52X2oay->TpP}S$Vc~tKDP=JxA=eR>tRQw5
z<H@>Is3~}kXpJ=<w%{teV+A4IBJ0MZK^Px}M{d=G#9sEJEjQJ(s3aC)m_oTFJY+RS
zYm80s18B`PIJUmT@e?Z?J-WnPr$u|NL9H6#Z0m|jT2U6i9Bh;8^%-E!X9eGf|6QVj
z2UEA@-!nnW<jciF9((vP%nmUj#FGH(uu>udA5Ti>ili-Re@N2pGZ-b<%#r^m8kh&L
zqE8-xoqg0!e@Z(shjW+cs{=oT6*I!d1u7^?7bozPN1Eqk<A^LClO>7UKp0c#AG-Hh
z9FkprQps}^pD`aEIkL)2FTF^XjIl<uwYAA`FyPMZo7}o}omQ)ZkWNE!`@sX6%?`$D
z^1Kv*mso9CJ8}e++x*RUI^<bO9QB!NHF@cU=XvhA=bS!ky^iPmj%)9E=-d(n&W&d{
z7!gHI2zq~ipEONrwVEVJTF5RvUi)0Lfx!|O%kauIZZvAFe(!szPC%s~*h=Lfbs+?a
z(Rf+R!nnuMyiZuzXqg<8fl+s#>U@hN&5%GICxk-cFD(<@yu%-!YVmLO?s5Fc3Eq7B
zecJVX;&{NB6X!Uxc9cu+e#p*)ZNB!}H`v|YVs&MWo$VcRozd=e=qxNC1>C)NmmAmb
zVU05d_dJ0v{3wN>h=LRD-P`6XU;ZX(tf|*SQdK7&!ph1TH?G|zH#t|6h*$RSki4<O
z_}WL*zy3Ys{5sNj=viJ%F?t=*-;v0#-@HnH)aB;w8=N}%0zdwnf6xE$&;LhGo;=O7
z&pyjIO6m0n2w*T6BaryYXK`VX+!{R32kX30CpKhSyMQj!g)+OF#3t2TPk>*gbv}WB
zXBQL47f}W3;sVJujiIm@3d4$<EQG*`4l9(BC}YuiOf(*&vy8w~%r%>MLI62fgV8xQ
z19V7kYh<>`jr*Il)=rZJie7KXiRaI9^4YVrmTC;Lqnti}f-JR28*psx1hue=Qicb2
zGTwgs9X|Z<Ly|Nmtop32E%U;&r#S!2S$f@jJpIfmtVkHgeRg-dv|Dr3>NU2uH~I5F
z{|oA^CTmBQSwFc>SPL-32mwJ8o_p~WT6-KnvyP4>+Z#jnySr@fJRnZT2$2(3D|kVL
zS6=xtCr_VXueZ(G(H5P#IWE0>mr<0{ZZ8vcHyI?7M*Apx8`s(Ejj7dZ_|m1MqbNrT
z@E?|LD|f3;AgD_+OQv(|I0Hr~uZvQuD8$eOS|IQ|MWqr_sZ^-dTwvE|G-$ROEH5pw
zva-T_XO3p0>f~+wkkAXzSz)xUHO2_6Q5c<Lv~wjeS|X&gnHxujBpEOq#%yeDv$3_w
zoqP8gjmN}kMroSj0w95zS;P(2nUXtL9%~HOZeC?L?31MlFTVH!^?Jx~I3y8*OP4NV
zjN$p`pQF*NGa8KxVRQ44)W?A(>=W%_>7UO+>khx>@VD8*Y{_M_%1xG-20>q-h1Eh5
zDzQZ&RIw2*^yJF)s@6=9mg3u?Og$F*l}ahBP_Am(ly+efVW+MUW!e={%2^cSS4?F~
zT*lGjILs-?j9R10>e><~P8{Rt5hq<zuh*zn!=gI8?5(mdr=q`Q_M+tTVapBvOcUii
zdhj|O=Efc1G|mLypB1n_?D{^E$p0|J_DMGBf%H+nPf+*pv?N=YXS}pXIv5hAiR%yT
zij`zhY$By}BEq=qqb<luYQ6~Aph&s9E1e4kLL#L`Nb52<#adykAP`Qc)Ja0S%F;l1
zkZJ+}VwjNa_hCGOBtql~Rs|Roxo~!KM!2eISvbUq0tYS&Z>FL3>^de{SXkh}y%E>0
zUE|)JTMP$X_O`bf4~A6z8c7@z$1x8!9&qH?34%&MG>)8grq*a{nVai?)C`6@v^xzl
z9U)}Q>XCI;kE|e-pxJ1;AQ?fm5|U*x_wU`u_XJ_6=yqe$WX$&F7UNMwr4mpHLI(Xk
ztT8jk|LN^>Zt<KxnwP@vJ0IgOta9}0uR`EEsf8><6-brRDf8n!*7T52Y9ebK8lw~@
zHzd8G(?F{PWLbpCHTV*tB=t^<+jn;PN~eJy`)seDCf<I)-i_OI_qr_JUgD{9PqTS%
zo4eQV@al6fQJ?RywYkk#UjG`6c88z*<Y&D5&adh926(<ls637yJ3_0~aEOPH#Bsva
ztJkU58~pBf|A0!^W;ooX)v7QOIm2!rt05U>B+&?22a{?l46z#@qJEvDUi*FG<r8>#
zl)8)>r<O0J#0be?(C7aBJM8Xm5l1n3Zn$y%7I*I6WpA&?Xf&Frj9W8B+OS26v#ioG
zrM8_bk|<v{TlF|AYw{%0uX&!6izzxcn}T&kA>L%NoE(pKE+AnLASeZz#pK+gXv}yt
z#Oj<{py)IkD4azYR+FVMq#jcEq&i@KWJyJnbLXF9?d9kBgX0|<D|JMrhE_RVEqAU9
z<AC?xy2_9L@?VoD8YM0Dy3dJ|Ykc`D=h)iXWq*H6ubcAj+gBLw-QcB{&a%9+&Nzt}
zM?Dr77YM5rZr{4gU;M@Y!e9UR-*fEPG5*{C^M6Zcew8HB1T_!g3GyUDc>$iVtQ}d#
zqsr>hgj&7Mt(&)5UR&h7_ul5lpWWo>@-xgeS20#oZ+KKHhDuPy3oQ5^#tYG^!R}y8
zb7>yox7pd-VsRnE^93f?WNCsF;MHqxhiGR+U}w@IU$jX-bQA)k(MDsm(;xJd<2*`f
zi%M%4j$#Ibh%DPK&{?<eD-}?yRjJnl+N}zmxjB}XmRMR|W^sO=R<nigxr$b!4N07l
zMmb5E5+^DB@ra%MF8A;6u)Wh`FpL?F#*C7fG|SzNzsRmQTwGzS<MM&qbQt3}T-aiB
zBP@XzvbDXz&wlw6(mdwbr=F(Ktf5mwzdzvJci$t+TxD#l-N5q{d4718T;dKetL*=<
zRO`cnfU-S`Q#actp95$kN|<Bv*hAvK->NOGG*GI}U?|z&xAeM0oIvnQO0W#xaN3mC
zmQ|}W;<VOIe6gi)yhdRa%s8O9s#gt#AckNjfHU@hZ@CZ#Y=#%))EX65*A_T=>KN-M
zj<B@UrdF#F1l~;0Q*w-*kLXk=`CwpI3eLLB_MxD<JYNnz=RU_Cc<pSPy>_$L_Ha?f
z!`}O8ZT7x?qWsx{|A>IF65{2Mx7y4tFEH-z6ORU@;}k_mL0EI>ck!PTuDDQ6@v$;g
zFSa~|`w=+baB0bfMF|(wOMw#?w;n<$Bp!l-Gl@U|kwI?0S!Ws$2!)6;(!D-<Fu+7Z
zOx{EYg)TU#vy#-Jh}K<zH?xrfpv)?|YnA4PMo|^tZZ}y!z0UTg;=Q-urde%}#0j@<
z++sLNxNzYke&@TtN3C9G*o(2&I9V@W5yj38N9&BaPL(8$@k5W}CyvmWZxcmh(ngbJ
zyGbPs$<ma)y?uJ!9jetj3-c}R-nqqa*r(s$M--HVXdIKosmsd=c$`{^&xh2RVuz>%
zVPd%Zi??VUS;TwlDX@+#ZMDHul3F^TJ&IkaueJEXqN)|=Z5VhcDH-nUQD0re4+6Yu
z$gn@4=~oC;$nD!bo(>dWt0-<rL;cyWaOcwdJlMU*M$}_C=+aqOBk(GG@XiOU9zDwM
z|M4I3+Sk8@u|8)`KF_btKhLGhAEJ$=*=#cy^m(vxm+kE>q!2WlHNv1mmgd~Lb&KWY
zWfqqjXrsA*f0L@;<o3<m4ErOxd)w6NphS(_NIYfnM_VLs{son%zX>N_BKHD>F_RoW
z)`=S5xOtuZ?k>A~JDflNG*3VM9N+%-ce!-=5?W`x@x~jx_~MJ)z4w3z_wG%c8G8L5
z#u~yfWIT>Y(u8iWM;L}P*Thxy&IWh00oWI9T5GhNPENC*TTWK0&=`Xv2+TA(H<)_e
z?VgK;DA$Hm8{%k88bt^U)j(1U1BA{|zJt?EmLaT%QWf%~!uGx)F*Rz7s|1|}Bb)Hv
zTNlv4C>jyg1t(6NVy<?K_2uKFCoCVn_aVxMm9-WsPubburn}Q8iUm7c5#7B_It`7)
za`f0qj;*hgCJ@KYCTL~(IL*caM*F<-^4FQ4JBlZ&$f_kv9p9*g-%*mFBG55Zt0B+5
zdW?mI4*T6*ln2A{kYB&`Hp}xT*xbFr;&PQ2UwRrPT2%dIe*UA&47W95(4ZTSNKKWn
zzDj@h0dbsBsYvoPBZ*_ez{l3Ah4`?Uq`w}UMf(_SoTk`9wnll1Y9SCDkH^K@E+?kd
zPQFG^t<^Esy0M$3Bxy{X#B{rG?+zGy2O%VtK+&1=S?DZq<cQ<#VJ*>kOdQ1|Nz8a0
zv$3<o)_#xfzzK94>rg5oAqah<G@}3vk<z&sxFabb9sz3NyfQ_COkoeTv%AHQfBZxG
z{Vu=%{og0fQ_?iW^E}>r?|srV;cH)i9q@~>J+<Yt2Q=+wq*#pMVS!rt@5fEO9{!Bk
zz|KD8wNK(cKYQ*7zOUHbt09D8I2e<m5hxG_F_9Z8q)LPn2cPAjO@ogzEhZ<NCbx)6
z$7QqS*^~tZ1%qnxRx{zeNi5n%R4R(ar6xz$msmTx$l_v$Mx#OysF^+x1+wZe4Xg7{
zq{mDCu~!6svt3ty#=-RI!Tx@973<?)&)N1_sry;Z=A#0{-?q6=b|8QXd{nhUwbi7x
zu)uh^$Ee>Wig#TBn^9PqyJ4>lR!F2$2%|qS_ml{p>&rr?PL!KZ=a=m6$(iDkB8^&J
zOQW5*zY-EFELoB=>JJ!o`_u;`DxDPNRVNwx^3sW7ezN5cf|3s9o>b+F+#77pT&F{n
zrXUra&b;H6N{>JJ(|^Ed6mjwLHN2q4rH?ORt)*VCBBUUV(}{ok<XQ?rwOXS;*yYHP
z6^<P{O0(HS;PmH$AV5mR{{BAwexEGQAc|O9TjPTdK4fF#0lnV1I7H`TTnSYnT<p;u
z#GiRxKUYwfU<rT<PFi8S*W=o+-sFYVWz^g<d7dEh92IKTc6$V7Oe$(5{XRC1@S08P
z%kxMj$tw~a#l%~?)RvbBYjwI0wwar6Af@H&UtUKiF%2`~A0PAi%X_z2J@qt$n}+^y
zk7zVzcY71R(L(t(9^ARl|L_0)&-le(|25CP@+wD9pGMl0V{1og&oA=om%fZriW@g>
z@ZiA%8jU*l?%!c~d6kPFyn_^DLO;b<Fd7W`;Jpud{k89KZ)1=C_I+fu#hL0FI!}=q
zs74)Q1>v|y`tA?0c?xH~0w%)XyVR%77!L+qzkQvJ?FTH*E%L3eeV2u$V=OEz@SC^Z
z;O@<vM58XRzxE2F;fM$K?vdw~ah!60^8rbkkZVnnqzs2+u3f#(LT3>bco+#rT5LW=
zY)TaGNkR3qqMc>B#}UUxVd>=fa-QeZ{i7%%&zvA^nx^=^Pgo7GQn>TT-GpsS#(0!4
z->GBs#8t{=29u}MYE?vDMQ1)?Fh`ba=GTt#-j!=ydh<t||MD4@jxKQX#%*rh*uaxb
z-cZ+AXt#Ol>?uwkTW9qf&vWJDE8O_t3RWmKw|6m8l3UAuzelcP^038dDCq8L78k0l
zEImWL?s4trb$WwcPMmm_Z+vac{d+zBum6|-8|TkG#i=tVIs4RUn(Z3xc9qt=kMbPf
zL<PcC?RtWxV^wb7>9MoB&8x3|i7Ov<x%}~6E?m6J_rL!-^}2_Wnq%unXm&QpqLkrg
zm%t0?M+p_tU^uqiy}!xg;%TflEG#ULXDOr6n7K~NnL=k8lNqu+J5VU|MVdfSsk)7W
z8+#dXoZ>5q&h^a3!MTbE1X<>!Pn7fs13@mGPNKykeSwF?=$uiUGq|5{@4=9FFWhp-
zqjDQYoI(n@rP!~#xnAMO;>JK`a?-?&cQ0^UugsY~QwggWn-*2L2m3)u2$V;X<tQl$
z!;m;h`SqJ`5JeH+`_}iFUs!ZI>LlU4_dXzwW4`^JZ_#WvkizkP-NvS<g*Am;U|~ux
zh2rrmD(y@VF%2$+D5?RO3aMHlCJUgPd0lz6o`etR+4DyMf#<`{&MrbYTS&)!a$LeP
z#Vuy4NT-}SfhfFb-TX7<ftBf35YEn(GSylHD-%wcaRWh`qC!@HFjT6F#pMPkPA;>)
zzQW>Cn`W~@r5ZRhc4>>0-t>H?2;9NIX!hsfDe^MdczA%OW~dL38(e=8ly3I8f1CGy
z#s|BYTuK-0cs^dGLao)Ly|hTb+hsV4$%Y1u+b}69kj8>3lnf3G_n8EtC=;)<wsXa~
z7`GQt!nsG}1xo8u{pB6u()<*B)-*{N4hM_}L*mhxT9V>dEz)z}D2q4FYF>(>GQUI%
zoeM>wCtKVzfnUWBW6qvE&)vKC`BA?MHfL#RnQFbxkz>cXdgB(=YL#2JE^_VaB~G6_
z#hn}X&^mXrS;nVQsUl>D-e8j;2$-LrXJuuH`GrLQ78aIRSX{vK{E3fuk|gwced@Ia
zLI~nGc4G6^vbZ>h=lKkWL+bT9%5&rHi3fFB3-CZoOC~Hr3pPK##rnsW2*1`wm>L$3
z1uRP<r;_DpVq}qa8*gr5(wNqfW&CQ0?gu3OAp)q*%@Kqlsm`f2n|Qt=Pb%zeZgNUR
z{8uj?;V-UiGioj%MTM>I7FjkRPZP$Yl%-mglZ#8-zjTosmoDIkRhlbntQ<Q^;KSvM
z?*oF3jSWtmJWadZVQF!J`MEYnmgih5#~OAvHhAy74>_`YhNPP_>^@-Y#s|DG2fdpk
zvXjT0S-k|AyV8F>9<uk-e@)Y`V~?Ch*aWoODfWB&>~3x13y<%8>vuS^v_NWNcDL?x
z<LX83-nzvf{o&u|+_`i7>}PKPZXxx(fc5oLT)ldQo!xB)!x2h)6N73zIZ3b1B4@J>
z6Hg-jIw3R{pFeB0lNVLXC3;sHy9=9>Cr;|X8tv-cJcYK->{#a++GHremx4e_{6a&j
zlp3(wpfgL9S>j9)X~i(jNoCHx?R#twH<+s}@YPqpPP4Yc&gMQ+rz}?6_?h9xrOVv<
z_#^)GkN=d`T!+Cpr9Vn}<L&pjaPbNvXfo)Jux89#zq!n{s~Zf4U6O3bz5BO88Rq6%
z<e878#?Iz|-R+C~?57tnXx2}xaQf65XV0zEYBiXfYvU`Qg_Q=iFl28#rqh{2Nsn5!
z#`&jK`S_y;40{n-tXXI;vb!^;(XNr|j8QVeC`*bUF&=ZZC90K0wB99-V=$H^NyxJl
zYcxp`6!)Z_r0SjQ#S^G_HaS@`!^gCR8au84rd-$DcypxmGUeqKAjh|qQlKp)#sxl(
zKjTEE({hqGVDnEh)XIR)0vQ%e%7H}#$kN#rMH43@4y`iNXALCQVR_nE;xtD{MUord
ze*XhJRpINef8AB~Djx}4yM7BRB;WYP*IAfv;VDTJ#{{85N{=MbPLolJnLtKQf}L`=
z4P~m&y+)Q+Si(KdjEqUHwF_j0Gs~7muvET5P7;>W_9TM3rR5gUILGrmhXC8su|#7_
zq@xgzFosX)eG1`>I2BGuu?#kaa6$hBv18_1PiS>zLa$udbrH~7lm`oo4c6CJIDTS{
zBS#jQ>olp=EBJw8RtCjVL<^LN)-)Y!jF=qm<Dy-ULR$~F$K6Oh4wvrJ9y?rl{YBg4
ze553|t|~p30xAL3PMg-+5>b+{eRB(&Y9s>Xc`mJLtW$+&whp)Gdr=&p5TCdmZ5hB(
z1pC5FMzh-(2~=q<SH_1@pmR+!95e3qNcuxe6d@WJDyS5m;bjH2t4Kwb5^HX<GNRlt
z$ufD25U#+mT5Is^b1(AOKl&>kY>r%d*lI=_o6OBGFc^-A<1V|qJN)QJKjOdsum2l@
zz{m5`!a>DStyJ-RC(q>jAwgK7Uat`NAwf`Kb#;ZMrDZamGaL*V`!R3)>Nhl-O%@j&
z^`qbKW39&bJ?Dj8=+TxN3H_KH`OnFzlPC#7nxfMc=^*9S&);L|_;KXXV<=%*Gb!_(
zI!1>CR${#>VWmkPMI=ees2@{nHIZQjKg)^t_fX2C(QaY2t7`FU4V15`Mk&Mn9U7xO
z{?iv$`T6aAesm40Yp1!nvq^WbOQsFIBw}ahKC8n9omz|B%h~G=xpn<E?^RoLI&;j=
zcW5@7^!E0swW>7gRmRa6B`fs%1J~K(nB~<|1i7Yn=Ob?2ze-Daq}vADmgqEgJ7=pA
zd5+FAqJEFj&KB9P{tYXC|NlU)0<5tJlaq{xba%G-##g_?+2g0MS;W1&_xR}j->|oJ
zhs_7KvH4f=3Po}u1f$WA;b?>)U^I%jeCZNbuixO6mtW;O-}oj7NaA?PcA2pZn@z(S
z^91hvqCG4~ECW1e7p5lCB^fC;jnUeWnw;Efgy(il2B%|ajmA?DcoI(*G#VihULk3r
z6uD5OsU^1oMg<Jhh_G3~$eiofZ_x6(Y<hjh!}l3<BU)jDQ}ajI?e=*8m+xXtkNu5{
z^mf3nwK;v}6*e9m<I+bPL?ew6IsJald+%PQQmHZ;_gn>&%xJc&=sdvMkTels>-efg
z=y@bbMATEPEuF(7;NuI|Xf$g4@|VA4{n$FKdXvS)1w0>)A3ed|PLH|P3hj1>y{!<N
zdFV*8y|K@&8y_$j_la~wCG<$MoN<=pwdauD9BH!c1SrD_Sy7GXcDt5*)=j0D8SRq@
z<c&S7?C?-@<6#KGK{N#1sx2|XVg%SSco)+#<&c}{cL`D2VSVDrKI_Oo{pk`JlCr3H
z_ej%>EJ=CqgZB_Z^3AV*gGx0dNgUz&;>C~ATJzoC`4+9VtL#-uVvI|rn0z+9XtB0f
zs3yp^JFbJo`tq@&KWF+>s*>JNq;+QkIvx_xJ*gH{e1^k>Znw|Qb{8Q$JkKkR<Dg={
zwMC&sVfSQ>3)MuC>Qc%F1lHsjJVfq32jiAgRcsJM5yMTTEgXvPQgj)WpvvOH9LG;C
zvVMG-mDL40^KELiD$0|SLE{cZ9Y6%ZA-?5UEGwAHHsuB%<e@sT<b$s(&-u~l-t;vO
zYvuPJ49X6E$HU%RF5;i3xZ?A)a?@e4q{hjY7*8T9A;DaWR=P}@X-0b^;&h13v<vH{
zC}>7zdhN>j=fGi>iP!?sE>LY-Ai5>mFN7_sRBf63rqubHoTW7+;}L`1Jw|)`B)tK(
zRtLXQbE*@?eKqMTQ3Q0dP_ZlSq0%+QeTTv1nr6L5*ALLzaOTVzn)NC_{_7vHwz7gn
z5{4BvHnvHVgg4*(HUIfP{V(Zs+VuMaN-yLrGvq~4p{IO&-<_k9in+NCtyas;9iB&?
z8^Te<=H@2-{s3#?_U+s3?(VX@yvXY6D)aO6WLd`j`}Y|R2aeX1Ki=H_xq><mLnvLe
zi*XA}W-Oz-`&|E<Uvc)2ewPIuvDDos6d7p{;8~{=C-9M>A`C2&fIP~PK|ty${2*k!
zw@<Lp#wzE+5QZT}LVJFWad(?s3&J?&J15#awb<l;c<VZ8y~1MWDYm<t><znIi!$y+
z`^*oztb_rLpvGdg;f6*<(cMVcesG^8%g{!kr9z<*%E$9ULIe$M2>P3hcQ?4Zy+Ncr
zjwqk1?a`<<(RqTFk}QhwJw=ul$VnDa-@bv`xrJUn0~n0Wsa7kTKX-<OPMfROE;EWp
zbbCD@=j6%boIH6PAvBjhzDN?MAQU%l-QkTle#4O?CkTTIoz6Uge~$+b?sMzTEt<_1
z!||BC-CdF-LCL9HK(Q49Ak80N+vZWt-5}Fo!W5rBD+^$(Ccq6?&sPZJV*?EsQthm!
zjCLs!mos!E*}!+M5}p8Ib&=+kV6vie*b*lh+SWk%?2opP9ZP$@#r|#uKM&|`ju;JG
z%Fs$f7WaS=XHT`LgjGT>qT3&#k`$fBr11z$PQ6xRZx|7H0a<GB{QztYfhG(sNs=N&
zK&{$F=NVG?WU0dW8`K!~Vt(_+dwk=YUm{Rd=G%*`Eg$8h4=$l|&Cw%A@O;I~FF#Ma
zzQE0!cNvW$qz$lXK-KHuv&WT7mx<ybsZ9#Mzl1o;(54QyN)l_L@tA5@ah|!-StgZ}
zo#kK=%aaJ>OYY0z$I9_=c%hq&lbOd#3V=aOw(^f&5KAOt=5cr;=kl;~D37ny%me3&
z^ZtkL5`;dlzVr%#ACP4km1>onH*fIspG&^|?QhVTZ;|H-N_pt~ljr1I3q+9yIvC`Z
zsP^F?uhgEi)_F%x5RaMTcE|q7>H0|($7pTX+wHOOV3&BDk);-;{6hcI>C7qV=1CWL
zy2EsVr^HG*v!a&+v_M2qI9)?UnR_c_$0?jUDX+E^1}P!*JvyBl$JUoQd2)@_BlFD9
zcc|5z#hj8(uviK?vzV9P?d<hn8ZbQyVf(z<%ts-7pLGo$b_|EBTAz4z>u0pm)y3d8
za!82`D_8`+glaZNGfruAb}*xqNRKgS*XPABl@s_=O!pfTgl^VLxxB(+GB!#ymx*TJ
z^!%kPs-<m-;8~J5X0W@@aBq*e*GDfz*m}eHQHzNoSrNPmkZx>B;h{ufS0^0$?Ls|)
z3MAv<h-;UxaOu)z){h<G)br00#R=LNy4^0VRtwK_!ROsOw|M8BH#vS{ojZ5#JMcyd
z=cwZ==Gq;M5mYO6jLDG7qgJaE1R+ZKuJFST>2wzO2mjzd<j?->&lry*tc7QveU?^h
zj%KroF~((Ra01C`n((+Y;^zzMq#=}wp}`4A3V{^{sbTZOt1KK{;?!~-lMP5cAkWDQ
z@}Mjv5wdZN9gMIQkNQ%J)bpI6rOC*n2tprWa^i7DSZ|RVOKV|;@m`leN{rPUYrsEz
zc@1^(Hh+C*pJsi5m3D{yQJ?)mpY1GVGm8mDOs(I;c#_6O$dRR0*3X`0Z$HA99%1O?
zE1xvAM7?cnf0Jl5;?^kQc9M`NpO@Pe8iO&b%S~z(AEProuR#{am}<pIGvqnGF*Ln^
zZ1X;1`E0>GQ8XKKblPn`eE&_P7l88E-r3=Uk1o(3^zj3qXP$nJc6%PHbCk4HLJ#tk
z@nDb8mppam94C$)cNK?O&i?*Bd%Zrt`OR;*diDCW@<)itj`~UM=S2Z$m#H;M;n?D4
zHrf<Eq=ry{6wYrk2m*Iw<&w<i<knU1T5O>~WT^y#Y8ap-j+n0$NDpHogmE0EEXz?|
zfI%=yVro_+mD2`N#>c11T&qo*B+PX@R@dek?CsK?%Rvv>-o8N)wpm$Ppwn(*O-h=M
zru`)RLUu-wYY$-!$^$QO6?;-h@;pVU5UVRFp-{?@W-+yDoz>+P7CJTN=31oBW@NG9
z{=E&h9_;h<)8{#|e3Brnpu{F|lu`|w4E7_cDqy&uaPQ82A|pvnMrt#BnH9Qpiok0T
z$Ie^T7CTvMVH}OARQ*XGKZ?lx+cqaoaA*XHQh%&Wl`-Y1U}t6XoHRy}NjEb{$io7_
zLtioTk68rg(DD72=Qw*Dr5b$kRpdHD=Q;1Z^A=dcYp=ady<TH98ljZOm8;hX!hrAo
z&bO%5szmXaN?0w@&&;6MMbI|$o$hOAk(q-hJsWr&JhSq>)8l*+{qV3h?2p*m+-G~U
zhta-EyBf!Aw6hzoq8Ve5su&67In1Wa3e;}0c;oITY+3Q@f;~x5C0izzg~d~rW;0}Y
zWsYOVmswp~Vt&3wy<Wrf)HFDrq_z(RaAj_{l+-BGujTWHQMZFg*<m8U<F!YjZHL?8
zd*QdahUHxFMcc$)t4N*a##ufJC6Qr>*Jw~_$IP!Rk@W_Q$0IZvRrcR(nexMqlLAFW
zpmDB8Zq6tQ{7ey8DM_ZML2hYaG|{&!kYOJ|Iv6q9*(2!>8Fssb?KVNB0=B3=F2>x1
z9bnz2$s!Q8kX^A9da8nor?_=zga7$o{BuU50r&6RVd?k&9w*k1lcq6YC8S!bP^ng^
zR6{nmcKFS2e#JljNB<d(W{rM#OrB+EZRpI;QLWaoR?uv=Ns=+1=T8b9jTK1gW35B8
zg^)y1Lcc$t(^;U?nFk!V&RRpe-9iY<&dwIY;Q(v@PSUS#uxn#6Naw4GTlD?0!Cbyf
zr*#hBx1@1`$^}M3Sg#_KBhm|Ph(=?SOMv57Dn!E}o|4!scZiVDU_cjEKw-5)7KWr+
zGu%tq+wP$ee1C0@29}@P+F(mde80l!l~rWm(;W}l>-Xu8V?>_gYeR2uo52S|vNUHA
zL0d}FKyiCK=Kdfi(wanDv=X$H<b}YaGam86s^W!JO+Mbon_EW+L7;q0rpfZ0WHiRt
zlE4$#y-jS^2OP4}-{0fSU;UCBS1#cBHT-HD<=3#8M?8wyxc`89P-pE(PCOnt0;&+C
zS;EHlJ<=p2O*7IgCyEnBqaod1pPgQpet$qd5mtWGZS}>Pn?MhyUS}tylD{{ZBrTK5
zP~Z145^QeCbA#3yTH~B1j3o$s!Z09EPL4tdr)5XxCP}3{gph>Q8Ura&zDKPd&}uiS
z*XpRgk1!tF-EG3q!}B{VE-#?$2BA04e0!16XrEMTR5)T7Z=+305QdCX<9v5BR|KF7
zL-ATOM97plj>)tp2tvo3QURk;hS3p9W;7a-tCueleft&8KXrn;ckgoc^cid>XxHwL
zr<!kl<6G?PY?J8$*RI}R(Caf7R_N}I+1bq4-x=cj6=arzNJw=;Gf)Ixh|!WTXfRBj
z72E99jN_OjsS<|aWHBkZl#k21__X$@C1ZBEbDQlVxU&bxM2YB@k|L$kz@rwh60Mtn
zzXCTIf8jIv<oA?mu-Wmfq(W<F%AF(;@4WjKwOW<0z5Z3aSfRCZQTXViizq4io!|Km
zbL|D9D8})&>_kSW_$JDNiN~o}d=!sr0%*(}|KZm^DWq<9r^n7#kI^VY5TFF)c1+2Y
zm9j)33ppjN3!=L%5J_-FRV9Qc3qNo*{XsDmy--CU=QvJMN}TkJrdDgPvfAP3v1L|Q
zI?T;AsnzO5TDO!?DG)l!0J7v4mog<qpD_*GW`nw^eb~Xka1s+f;dVHX{PcU_(f@v;
zXr3r~Ur6UJK~bk71kw-it2HXkCiPB}+FXlnzeg6Q2x|z0#1!e*vXAX-$Gb;nXal&Q
z9$f_YrC#8aUy60QlBKA1nY^{wB2{n5ha*ONdqmwX&Cwz{O(>;P5T%w|F?Y=Lm81+X
zU7A^15ZaI?8J91-$L(uZXttY#zO%ZF;|Q<}21DwN22m8b!aq;)t6%+s-~YiMv$ncQ
z9Ni&~UDf>lewVqqd2&YR++vJyn?55*k_4SYSaot#d7iVs-(@@=6UQ;_c6%~DTmhmb
ziX2CMe!jzC(4*JyKh|aaTtQvALn<8nCO<|9P?A$eIw+Nr4~Lj>ip9gC(OE{d*+2w7
zG7L}(yheys0&NXxmLdJX`M_y+p|dQdQsn!E5Hwm%a^tkRMo~(=Ge;%N`R1`2NBo?h
z-yL&(9MRhukt<29UgzBLV~D^bi=1DaNj1GFAx{ml$!RG|lnU;RQ-;>!2|+bb)J?|M
zIw7yj*I2FRH2fHyB{VyWghA-6W|W681}O#O;gHY^5W*mOcgg!V(Y0fs1=Ub;_S7k^
zeDp4R`&$H+oV6qKEX^-stvjDfmp<mw$5-e^8L1K6d9cO*{O5l~ZjF<3FvZ}0<ODu5
z6FuRTPzvQdb<xF#8=OC3QOa(d$g#GN3yDNvrEz8Os+hX9<6xLvqm9L8Il@{zsR%qD
zsT|jW+>I-fIg}tbhSXSRM<fHt1@S0Br_N|7%~RsUQma;2Tv{a8hIih*&WU4-oI1b4
zty>q+Lg3Y^gw1uV*Cc2*m|txYj3Wm9eIy>0W)%WOY9*cznNF!TDxfTJ788$#DB36?
zsnn`&<6)gY>U-~fNWD=dO>?@v0rT?<R4Xk`pE|?!>(^LXUM8$GxOr<#H5f9Ab{WM(
zx`Q58d5q&6tw04}^Ar`-@I9Y2*Z9IGO>=baV~xdE(rt_=#h)#u<A=?i$hCbEL7g$~
zo_^fd9~zhCqUn%2Q}{271Fw4+O7eG&)B!xSct{wk)sQGkkV-I$hP?6XUs0`9c;WdM
z84ZSogpB0Px89*zYw)$NeGND5P32Cef!&m6G);}q_76;M)M6o=(isi~A)kQfe(O{J
z)V8&?$KLJ$ohfuKoMVI$fYX;N5e{n$Sr`G*7F@X^FtF%{UYpreGqy~(7QYLglj{dV
zwHC6p(&5C(qZ~iJ!s5~#%~pd-rCJDJ&+s*kLz<@PY;Zwekwq|2UW>zeg4411WF%oe
zZA^X|AN6nr?O_D*@3P5~XJ$cKfw9HE5lBzr*D8=V$U1GBEAv$P`}DVW5k{kgM{aXh
z@r^(jQB;=#*2&;dro#_x2i=%*;0;bDuh`TH>EQjcx_T0|mlP9gv027&cb~!b4$ajS
z;^Bax*~G6_D8Frb{}qKsIL@9Y3T-?L!f2$;@l4Jy{{0V8I%dAHL^AC1@rCzTUR>br
zy?gBM?>T3mMvZ#CN)$OR+`I3+%_}dz#;sd-vDOj<0egD`<`)}`<Cr*3G1d@`6XH0=
zIh~|<QHs_E-}hKuS>xQ<^IW)afvv5pEG*2kwzh_pl5Tg8JkP1u>n^Ap_UI3Kk9A!?
zS5Q~;1PFyN`J}TvB+_`)=U}-d$@jY0Jafj~(oJ@G6p;ZktRN~Le$~UT_!xoHc+rL=
zOQ=*rr(Es(jFTZ+r%0<%STd{e1CQ0?M_4$z;yhKQWjr3^$$*7QpM9$Q=-L)ndl|Mn
zWP87lQ9jo5@WK$kUZ=LOgs&<Hq1iGq!#E=l`!vU6beiJ}%ill0$nUNBm|;Ykj4?<W
z^%e`Q4%M*g%8QK#E2#PbNjf5|*YQ(9?!`ox-+|}m5xPOMR^`O0^K1+}wsr@E;T|J<
zgG!}FmXFAUAu*OLO$%W`gElaXlh1Q${>~<Qr2;%<0a-WzIPEA~urZu}qd{jmNv_G0
z6y<vaVSrK|$`klTktKpmgV2_GP-Cnkz&Lpu>vWH7s!>MbseniWflrEaII!A7*$OsJ
zsi=@fy~1#qqiTv|5Yy`p(W&NN-TgJSh9vM~g3wZLwV3NPSwHUc+>7%ZJNFd+ywB<5
z3)oSYdspr;=nL+Sa<1OnL|RKej)>xzRD?+1N1BX^5E$X4YCKPpXRZqQ!iDQZQO5Uv
z_d9r=WxKo0Xf$RJb*VH0YAv_x(|H{s7LoowOUGLT9YuA%MJ@w&`&%q7HyH2rsn%M=
z<AiZy5W^v8Njl2N6GNpQAguGWOw$;t8b||JjWIboPtj@ig|GQ1)~6!ZI@vKFF1EDA
z@U!J&HI=lm)?!NukVC=Wr%FIP0;&5v0o}oXaVEuSOX7T?TV^0mQv5)Xra6@$U_2i3
z)?05<uh%)bezGvv&S^I1c=z4+F$TW#oo|87i-1qhd{6OvhC4TFS9M@9FZ%V<&SCPt
znRxVY5dB5l?sk`6e}t3)*4epWkp}`GQHaKt$>kDZ69rOfCt}6Lk1dGlrRBPC?i{mC
zmSV72!XRX6VV)DmoLj`gLWf4fap<P{n6rV{%qF3T8>YdX0AnU7;^Cq8s6vHLj>tm^
z@h{ldhu?n~z56Vr?nyQ=LF&q%8IN5AQu(Mb#BbH9FSM!dv>5jLXp_3lx5jns3<@s}
z2xLDjl|4m7VRr1>NgBUIU#HU7#rTjOTDtjpu)D{&KVUfM6V7)~l>k+ccPIYHvq-Gf
z1t;0y35yaMX)}D2v3>UjN@r9gWYLiAtqop%`6Yh&t6#9Xwu-e*>7mhRaPPr3Nz&tu
zH-5#-FMXMMz0PQyptYviY%(5=P(hhu^@!t`et&>50;45}ggiF{flm-rICt(GX_|8T
z&TYoy5u2Nv)a!MkC}KDq((84p)<Wv_8bJ^MpB)!`uAnYZ8OF*3s>W7;k%pCp7Nn6g
z@-B9##u|4!8%e5D#3)DPiomOrrxFze&Rkgtbe;i~f~z52bzhp|DOa*=@*I<E0<S_)
zYZ7=8azn+dQLogfRXomGc=qH={N(Z$+da+hewSeyLl&_gjp#-r)XF>$HgDhwpCwfx
z9QAnGQ_jsI@cHgD%bbyx;k{i5bJ~q2^?IFJtwyWaM)*#Q)M$+_1i~xS%q3VUs7Q~^
zd-tiH>0!MJd9L}{n{RP<Z=X~fM&p>hJGT#v_a}*{Kj|h4T_0J9%_@Pj^1&9x4~}Ez
z(DNc4i^U3ql1{%)DMhtfMG22MNzuk4g~IbBp6??)4`adeJfs1kT;<(J8!~M`c<5Xp
zv`3aeGEP`uKgQ*~yHuNQA<pt4eo!UL6uqIr^Qy=Qc1J0<Hm-4Jdz*!0ZGQcmUoejP
z{0D#fUr-rz`8XbO>sFuFzV%K1{oC*H&adBMuIUkJOOo{w$|tA_;@&<oTq22+0x1iq
z)*AG>`&_wvhdldRo_+2(K~Q0?vq%uu*t~X=B-LE_;3`+I4CwU+bQW^H`R%W;zu#qh
zcR-d0c&dptPH<BxL6VH{{0c_tLRQHYCK==8Kpa;^V6;Q&oRwRS&ho;i_lq@c%XxXa
zcr%SbCRNoBTNoX`tPm=n1bfp7`!HH}n2RJPa|gvdKV9%1XF;6}-b&<52tlpp_;F?W
zwS3N1cv`2wm%6@7vlQP~bi4cf{O3QT{(E&+mR8Bq4A1xI_4>U3-iItM%yahaNvhRi
z-YNnQV<<8jGYi(CW1Plx2N%9iEc#{^*|gI@`3rEe9@=kGVss7yJl|q8j+17*;$muN
zyCyZ#lT%&cVga872GU@)DN?C<aqJ!fA4n(LL4ht3cR5&zQi4{a%IeAj%gYPQ&9!JW
zs(8Lj*=dvG3M(w7r|Xp2iCc(DIg_HWs|=>yvDt~sNEr-HgQ#gBSW-wtX^%Gf4wv4W
zxiSZzb7%Ugef+5OYdHplnBHe4N>#pU^2rkr>Wc?^<#-i12^}Y$qbK(<xN)JX0YSS)
zbFs~6&?njJVbYmVUEJr^;sW_HhixqfbsNi~h?#yZ-YW#9X1HUe&LGug^|dV0zFCsd
z-`!=@>(Yp#1N0PGe)J@$!&!zIixL)3SOPTGB#_4JZ*S5JB=^RBw(sAj-EOe`;6A5L
zo}$@kkfsT~A5y7S7z~Gmq0eX>bLrAWu3x{#$&)A9?>g0tMx#ocT70SSl*02pjJ6Di
zL!|H#GQgM|&-2~Jf6zxtk0VEpvbws;VAyAWf1e~t7>}Lbw-AtJE=BD;sRy5VT|ZY)
zmn(@^Xc(1~j1&UTYC@Ueg#v%Bg;at(&+&r*vK&1aJ7_iZ&=>+V*?5cv0#8wGHrO9-
z5d;CF(Fjk}kV3e$X}g9(;Ykmh=VXHcz7pgasN5}bzEm`8ZH8$~#f<pFXFA+^um|VY
z!3(H~fScEEvA24fum30ilpp@=-;-cCQ*W~OSO1E^b{El_<Hhw3L4S+f!1C%MK}FDL
zv=|PCcwT_k8earXFj1$7!0kk<l?vH7VH}NVRvL(Oj2Z1B+6#<E5g&edkzBi*-(qLB
zDKnGO|4N%S1}!K!7sY?VPDr*=SP(AF?vOm^4Qg^rZZ%r_guX&458sC@*W@Of;S5c8
zbjnlM+@d@WUxwI`OUo)xF&>V=N~{)a@9cB_{BfFd4aNq9myzlOL_(D3gno-wdx^nl
zfRG7Rj96G{pn`q&yLVY=pFrdlM#B-8-h6}2t(b-NW9Y#)$Cm4S>GfB5<HI+2=Jl8P
z*{^;>s!~Mglh8#5AxiqBX@ZDzr1UTvu3Wj!jhi<)apE}h^KCx9c$E)7{E$YY!S?n(
z(O42x70*7qPN&o2mp{M2h4;4*rhz0O&k_WZ!1su=F+n3F78#|T5>TXCi*&1Xnr~$l
zXPOqGyN_A+{`l?S;&CvBn2z7WvBd<b!vzP2*So)~O265_u|&_pFr?jX7b%jx!Xwm7
zg1R(KnU2D8fwE|$84mlr|NeWt_FBbNoex~$M3(3L@|VA&Rtq?B;y3_75RfDZK@gDT
z*(cDU&)sGqrbqGao>W^uc9i|y5%=!yV6-Ne8J-kmT7vbF!o!%{Mf}R8N(~xq+~^dA
z_MQ|LD>T>~=jCh*-kgtfr?4>LV!2eCLJ+Ee`F5Sv<$31knl$QFYPE_hSP~$F1|d+g
zkbWvY!%kLm$Kfke*`^RLcC;0@PKbg>Z_Sib>1^+&DcgzFR|!3rSdq|0kToliTlRk$
zsC;_i06ZjU7c&9cfnem&;5n3%74k4#{`s3M(vw98qfiK?(AG}EE`$SLyhergVuvh>
z*-tfDcL>&L7AkjQ<rz)xqlXXwBK0aB24>7`?q+`%K^?}RkVvB$?04yRd(1@<S(X&V
zcOJ9fS_X(tZNcCRgNH_^5##<I<K8~ox306;u2J(P!)}-N-+r6xH*Qd=RycF&ByYX-
zHl6te=I7_RcI`UWXhPrd=iYeZm;B>@^iSw?I*iBniDRc=U`$#_G8IK2Sw@n$jbfS_
zQkmoX9*ss5fNpo6TCK{n&pk`MUgP4$i~Q_oKOspHj4?z}gfW_0t@e1=^|4-6G=FK0
z)Mya7+sO!!#-K#T($OW_N6yebeuDb^3jI+`ZN7omtl-r{goH+?L3L>k*=*ods#t9>
z+ESbA(3o$LWifdgJG8-5m^4FJf$~B`wSo*PcvuD-cVN`R3QKMbd75LA9P$LIb99`t
z(5SNH{r{=^&nQciJiQP5MZ{~b&6_^USC?<!J+8-W&(6#)c4siVi^URv76g`rB!Gmp
zBqT^Abq<fD&!Q+%2htHJ2p~WL&>={yu!{w-W_E1%xb|Ib%iF9<fBk)JMCe1@do#;)
zPxVYSsf|9_nU#6XdtVn%JpbqaJcfN$pV@s4PCR>>$DTUM@kdYbiBCUIo>_FQO7+k(
zkALzQpFcKB&1TFWT;a&+L)2R#wOUN6O9n|nP^r>SMhK;_R+DE57K0FyYPF8Z3bMfv
zB`ggk>D|0SSr%+<Y_qzyvF8l__~`VHV`ytAt)VoUQfmtBzL(ZG-!UN&N+LX$+bg7V
z?=DMCn&o3rf95i4t#K>ANvzm3>())a7E7*kw6z!`$#cEC1T=yutRV3j_MPR;v6Dy9
zA|tGXs1U-MCaR}swM}NXDD8kwXNxF|s8y?!MT!sw)rusn3S>T@=xor6AhaVo>sR=V
zU;Rb)FW33XS3bknfA!0J^=r>#<$&gbM^G=YYKWC7jb?zcBWg9DMl+-}TV?<L1s0c@
zgrTG?N7U*eMLD8clf=~#U-;q)zVzj1xO(Lh$)F%lBzY<*a@W-yMG=lT%JBk=HVIni
zFfMNiDaV<snzAgXi(#!Nzk66oB7!?_%X_-X?)du${k&UZ+1mlT`^EV0_h;Yil;Zp7
zW3M6iWP&b=g27<G;^HDRGc#_%IcD68qPQb=Z+f~Ji#8=$k#Xb34bGi=4{bHIdKJ%e
zKx1=jlUHATm0quh@B7n^+M+02?(uYTzWbWrqv^7D@0&ee+yw&ev%jpJNXmb<vAlnl
zlcx_cH($Zn5keJ?fgj)Nl?2a|C}r7e4K1cXL^{v0sUMtsjKG0<F=5fhe;KQ1QD{6{
zQuUx2`!uT|jcP~~cw>o=%W!p?KG#Swd2cJslnJtoHC^|>iOF`}35ursZVzL32Wu?u
z{o0ilnCSxdo=fE;jyv>*d$#LgrUHG?nAq@6C_#bqW|UIxekl(XhInC2Jkw-mevYVG
z!HCKJ2P2Hr&RUDJl98@-&mOC-J?DgFa$gWGLUOV_Vv-{(CZy|hJEdDk5J-hDB<XNS
z(&>@)hh&2Rx+s_kP#xpvGKU3D8c}KurbHAey0^pd)^)Zoyvz2ba~xZoXR%hH7-kH6
z3Fj|d=a2sIPnesXBMw8ZU%i60nrgLzREk2E<VD8oZ@k9p>P;3G=c!gFfXkJo%Un|y
zhQJG)M9GwZ25B&+#9;6PpL(Og;^H#%^9x`lbF=e&^{Zdw7k}~3P^r|&^8#Hui=H41
z9_YA!B&P1fmwtiwf9Jo$FLI37Mf%Oz7Ee6(G!@?mGvxNoHF{ee^nnG0r-&9>j5<Ap
zw8WJVyug)?O6}G?GN3lwVz{yG#IRZ;l*HuD21ED(eznRd?NF)KsVLal+$5+aR9bU{
z^$ODW7;X>Ig&jMMXAHBH`AS5XB`Dv=tX||hfA2phT{^;itIpu&Dsp|3O3P!mUPCq-
z%q}d@+1kWY;TTz4iYx-DD0Dt9dufUypI&s8xWaZ>pp8Lzl1M6sTeqp4%-LzXGzg|A
z_uldU-z**MPRa)Bz}C2HNlxoyASSGv6iDH^1&h2umzuIDuo{;%AsnD{79|P|DBp8s
z9dMb^0)!fC5~QM9sgN2ERYE1KkQ$$j^-bP=_Z*L&IL+Zl_H*|2E8JYW#_<#T`TWm6
z3r29|@@<lv9cHUl@;s+rok!=AB25UQj39==piPkviK|uGTQ_;?sS`Z<+@pN+_kWKw
zpLv|QR?N?T`7>0t;7|VW`)rKb)Mi^O)@#hQT7=<#PMto+($WH9s2B`;bUJOG`}xP{
ztZR0*w^-S?zz=`;Dret%jdt%6-}&BE+M6Y}uLp$wB6*6lvJsYMz2fXs1Yixi$lbz8
zN~B^>xoMXjn@j)`p!cx4U+??->4Ycl_<oP*dvEFJLngbulj&qk%TxE&<9vkAV0Uwh
zwbt}{J;E?#|Ni|5A?S2Ec%Da56nEI0z&PI0I#8#zWRwnh_r3RMwdQ&Ju_q}?*TElz
zAr~%OU}0gN&wcK5)AyT}wWg)uJ6Q=kWr3J7OS@w)t2v(F?g!)^vZiinX$C0;J3Bjc
zIvqMY0}usL>G53YTnVhzAmrFw-Hv@--DfqznaNHJS;Zc<%W-BBjY48k<7wZ=5D7~p
zN@7tGqlu(Nxz!3*xzZ`d#j`mKd7ty*ZX7rLw(A#oZ>OuBmdzh@?A2vHXlLq&9iO|;
z|GVGk;~YY`d#o7CiLA9);hH3v$UCHm^ecE#PEgUr&4T9qEP20A+U_B2iL%mxb}R94
zwl&rm%yhq6<Rk+YgEkhe6y@}DTDNE@3#`!OO5v=Nq&rWHL5&v%R$Bt02(gTI+7z8G
zGRZJ`fvs3qMz+o?Rd}Iu=Li%!AAq!|Y(%<t9d-2r_02VQu3uo@4|)3VK~`^W(@Qc?
z9zXu^n>_c~pX1=lKDs;GM1fDES*JdlrMr5YI1E`|U+3H3{u6%lH~%V?IHcd}Ff-Gn
z)RtuEy6Ih(r$q|4K+?9L&@P=L2&<^TXKsEGgrwPMFiH(k1%COLzD|~EzWL2>Fc=On
z2t4n>n7WSy>Xx2+p7zx%tbYFmR558vx12b>55G*wiv&xDn-?#zurNnwZHw0896=?-
zL=oG!Z)2>bJ~vO$Xdv87<qo^&la-eKR)_t~7D_2JyBD310zV2#i;TG0Vm=6IZ{B90
z*Qw3U<3%y8*#$<OE<#xHY($o)3`c!Lq45Hr6<g4|c9!jp4Yr~d<)sTeWl~mJE6lg%
zN%9n3W`to#ktb|y>~Q5$o0Dfw((SeJJs$+)jLki+@j@^hCIsCcjdqu)8d3>-u$EDg
zO{0JQG!A18<+v18>JmLkF)$OqIb0r~5aU(e*aA{Y7{{>aGC}19z6uD!kSGpl1QAvW
z!l>fdKx05zv>c1Bjq5CIRAUAS_?~d#(OOceH0X5(^oL!JpFBvnH=wi$v-5L2^~{qz
z`q)Ws-(Ka$wHv(k`UT#4<2<8YN$j`y)vy0KPCvGU)_q1Mp<krTG$Ptbha*ou&dcBa
z21(ZA@UcUzZ*Ndq!-0cKJo|}9IjJRI_`=W82qQK(*J(BzG@4bEC)n7yO*PD@&sV54
zJzjq42h?hH77iZ5dvic*Uh@4He!y!#ewli7kbMiErCN#c!y2Ji635VL)=0Lt@wG&!
zB}JMOjdiC^qvdYAN_m7~G%X>PWl52j<XMR>Kb|vf7fJ$mO*nT<mcn(Z8RLZYds(^h
zQ+M?`;{L#H(v|qr4BvOzion>4M2&TBhOMnFgb-6;Hz~!IWyx;Q!ZCf?I78k-6nLIb
znvS?|;T#JK3(U<fP~`bEpY+E+{xP$&vz$3|1}P<RTwyRAI7W~Am6`VJ*F7ax7=u|m
zDI>do?>1H3xnCYsXZRpvZlQr6c$_?Sh<>k2G8&R4xznLx(M3s76i5Z0@W(O>HD+ce
zPW{RqbM9UkOgI|><D>v-H9jQ~5KBX7Qb_xlVHcasLM?_OM0(!1(|fm^CNT~SBlqmv
ziR?g4**!g>pSY6|o9HPgUz5L2kH-U!`+@C0T|-R&zqbTE*%w9#tcF6BSQR3}3Sl)P
zZqx~@G1=H+!!ed_<zkFMDFM>MjuV{Sn!tfKYhAL87L)$xN%wexvN&H`+~Qv12m@sW
zp0&8<Dd(x-SxcFWD0+SLa0Gb)W-<qPAUvF1ip3N;k+j67Am84g+*-ry?x5Fh(z$$r
zZmC&$<Z+%pb%GyVyvasVFf3ADc;Wjz|NLj^v{x8(+MGFkn%CdE#L=UVQ1?T&H+DGt
z)|-6c3(s@x*im}DF2mt~z;~wI$som7;K>Swa)8zfD0G2Wc%Yyj*8xQ`%CJZp^#-F+
zhA9kR{_-!7BndD6;0L5x%5eCT^>pwNL7knS=j1Q{S+*`+BD;PYq#{s)xoSySY#;-l
z{_Puhwxm{xv1vwcbqyPM<bxr$%n{ZyNDJaz<5#OlC6S&&D37>SW&QdZS(=UQQ<T$2
zNO#i+Dltjd)zE{u#=`OeI$LX`8`}hPvqVt^Z$^@gdPrkY+Bl7eEg24Ta-AXuU7|Fg
zwy#ba2+kZo#qxX;i^p8Ojux6S%h+DuU}LSr?94K)1z6weQ=8p~(T1`p(Nh7l9Os)_
zM+tAW@ro+7<rZZrDaVBJpVm=ojTX)TS8HdRF^MCyTEkd?>`B+{<Ee>9U4XE~bbpnF
zrj+AcOFtlvs|epG@O_493Ra_ahAvBz%#vrCvdr*QiPZyYQHAwEN@K3hv(G<EZ5Gm`
zO_FwKE%@xqXBcz`{P7=ukL{fvCr_^M=o80y{`u$FTJP~EfA}K*>R<mZMcLuCH|sq2
z%yH^v#+QEm*GNVsx66`~&pyfJ^XGZ%;@g~i_dVj}S!%N_=2}f2J@p9Hg=K^)kiN!`
zQVt$Hh*XlG8Zx`@2)4+`vz#lJ-X*MPo_OX2+6-y70uCHrA^%o}@;ztbX)>1fHMqIH
zfeHpxDj|Vq==X+%R$+6C$u+U>6Gpz%0@<-HNHEO>bb!vWO=YPcO5X2B8GBiaDPuFX
z`Z41w-3PG_9|s%Z#KpVpmYZbMG(T~Bdz&B#SXx}9EDC!4{&;D6?{+xS(>0=~sKVOX
zZQgq8ET8|}^E79g>}+ikMlt<vhZkRbky@?J(W6Jm^W2pV@AzyVVotjktBrrY8|+Ql
zr-umS_Kvs`B8A6+gZtUp>d|TU$wnEeE#1OwTn@BGfp#e}Mi{I#;ElnR7<akeQI9mv
zLc$uSp_2+}oPlkCB@~(%OK1!XJEWa0(oU0Ftpag`G=US!k7aAZy01wr)V=usUP!rj
zd|3JQ9>)1T&%L{GXZKjhpKcj4sdC)4eIGYVH+hc_YDC%#fTffOPvLt$UR*)X%rIy+
z$+|u4$fb=S#iaKXX<c)|2c=X_Bo5O>%Y6=+xa0es`$!52rX))e`rQs`zYkf7Upm8o
zBSFgoAvDrxVw2PIEMbz+?so9X3=u0*QIcjUYil{p135E|8lJC^X@Mu;?RVbhsb?O=
z_kFfkS6MlHgv0xm>GehgfhF)fhW$R@{^mFNt>5~0SXo?PeSO^-*b6B0oIy{cq{SM8
zuQfs4v29tFQ|1L>Gk~hogO+7UmgQ8dF_kc)IaB8=U-?DWH@0}~^;ap12TO?hNSZpS
zAUJq{W54t@L@h*0OT8Abw68%l8>2-^cWaBKr3IufkUn4v%6<=Yfsz3pf@0Vw>vqsd
zf+-4$BFD25;Vag91>5}uA(gWUG%i*~3fG<LMHM6hV+(BP(_G$9?1dEV9c+@|1p!g5
zK^RuB)@gcmnNef|l3qf!TBkYF<W!|Wb>$#$$r`s3Fl9zj=IC*PK+@~cti(L}=sr6A
z9#RG<-y_d5*AO7=G#SN|2BXJjwkGTQ6%POau9ZndK~yJ6QvA_?GU?%|0HM6Qv;}{P
z#+Zpz%5_IMH*hD;FSJ{|CrLtYFreG-(Hjg%(sXP*3sN})MX4ltSuh+WwA&rB+%-#h
zeh9drXk!h2=#yos1Byx$2L?}!@JI-y<+0Oe_|!AcGB>kCEv|F@`elxvIKaUp%goNr
za{KlUzyEvR<ahr0AMlNDe2a~(Ys{@E{@UOC75<a|!@tj0f9)%HeubCcyuc``QqH#6
zQG)ps$GG0!<lw0jeEDl%rPi7y2xE4(+O)TJIC$_NCr=#a^vNR}JhG2tCk_)tlAs=t
zjk=76U6OQ&^aXzC@!aR0qd7Z+)+LKeOLV#;Mu}mV7c8ypV}7wgR%{`>4$J#vsuj!5
z)+VFgkVf1<YCAU8bX`xA#w4W_!{K0j3=C<K)9()%jnZ+a=g%&no-jqzsrf$IuW1a}
zU8dA`GjEQ(6L)=m#Lf8#rN+tjOiF65VPfpzXI*30`q~=FXvpD%2WU0wh>0QTm>uvY
zS}x%lb&5ihBne46;_B7QT)%k*=?S7rj1WGRYKyhCE#7$JEw;AWXaiEYW{e40WXGoB
zQ=J)NtTl7{+llw>?k!?heCD!vU6vEmtPeZZmDyd5;?FXYVUEBf3M<Uc&T;VI5vtV&
zQh1bw1uOBr2q}HaV%*`X$0;!etl9M~-HWmu#4`xtQe-BfVcxj(8e*sjO{5F#uur<P
zP1<Qw3<l^l#pVT4yW=D5UXzQ7uFhiFyO7zt1MeL7V}w5{+jhT_>qKk!(+qPDIK02k
z?&BCAv@h=NPM1m{m2~<yPZ8Ft)Msa?&9*R7xF$MbU3}cwp!&m4%8547PJ!Ltzd2i#
zyZ7h#c#qFAPYQING3xiohC|3oWZ?j!#-K$GA|YTzQ)Gm_Em*%r(C<<amMHd#XKF;%
z2wNDgUA}@R44--WaUMN&lsNS142Haa{v2Tx5`-adpM8sGo_vaIG-79Gn<N?H`-*et
z-sO!qUgzMU1I}|5gYuklexBzH`Xffe5lNb&OEcE;=Ja|!($qCpd7dKAbNc-rd7k6>
zK1rH!_{b5y{`If(_~TFVpw)|yW<CrUDcJYf=eTv{eMaASk%Na9X)IRIWl6f-cA@T#
z1{#a+1!yBsfk))|2s})gQx11X+FK|A@k|S2JggE}70@jMdYNTO?B1A-anVpxO1vn-
z6osq#7aD;iSe%2MHlwX=;@MfeYK2N`md>Dyw%~a_Q3O$A25Pgs{N{D$gDNvKE!OdQ
zWz^ww^qHNh<D~+DPhl*N9N$MN4Z|d%QEyQeC1QNzS4ufbS-Rz-Cl&opm(ceKRX~`u
z*}VBGp|?0K0sZNt=bh%wla|t22dpQQ{Q!VLf4*2xIWd*hW$6+|N{!W%ELB|0*x0f}
z2#eMRC`q#nD<#59D2>6E2F7LfD2lPUB~Y4Cx<lxB)I&p*8$9cA{n9qKwyx1^3D$0H
zg1y521CJo}5M8d(-fna0(pB1>Dj3aUPaMM!42KRK=8M1ZX}<J}U*Of3U*o&q|6@iw
zZNB{cmpJ<5V+?eGmI|*@WoBuSxLISL@?kC=YfqpooSS};!_XG^C8WcWZm&x+N;q)n
zFvpHRO1ItPN8f*uOP4P3)?4rL(o1i#c6*yg4xi!R@?r91gvhs8Sgf#bxxx0<8pH08
zNXAtC3d#0Wlu4*Wo=e%lj7whbX6MSTmNZRC({vow_HZWnNo(5loHfP?+oihi1T+D1
zdjZuHyjgLFW=YE3h#>d;`)`7{>yICWp}L!Kah3>96DB<mqcuC*PA7Bt;C`-MzrkRb
z;QNZwrNB>1$sow{5+QP0trl64@z&dCX*BEXUpho*dqAlTaiz(*bC-xKRlf3zUv^Rf
zYw;w2lPf6YuAaT7xD)(g$J>o$s+^quee%O3rGiOU=be-A&oXYU?of@Jq>07q03{>B
zu!<BZP)to`t&1FRWwWua&aL~D3u>2UdfbE%yPyt(XAQ<`d}|1?L_j1U!V+SUc~07G
zGuYmy*_tJeE7%}L$gxiy6KZZ^YvPVALQHLzraMenr~4Ck0hIB%SeE4nvQ!^7KKMCz
z%LncSM<4WBA2jYRhu-<e2in{Z9d|Ci_5kr|M)CMQV+yPg6xM+z6^7JiTj(?;=?%zr
zf+?)ajy29`-fH0(mHU1$?l^z%*)BV7XtsBOa{-JKNj&M~IQekkzD6l!TA*qgPY9&c
zgrcM_a%y=>wssx8b(<g`QR<8WO|4NS3PWZZk|a-gWZxoR{>&G6>-sH%AO3*#oerJu
z4wXtozt`jPl~v-X$?0Q{@ZNjx6U7y@$tg;YAO7(B96x@X0|)kV<HilQ*s)H7o~AAr
z)_MxzJ1wp?mO;Nyr?bQ2!ZL9jF&GTUvXsGKfVGz8<z+nI=M$g&B&F7D?`-kTyYK(R
z<NDFcuhya^gw0tF{rvOXK6jOHuEjfRidvZxrzJtXhKM4HQNm2Cg%?F*F|a}Cg1j(z
zgNV`g8p+N!2#KmUD13n@1+_S!$TOg}C&6OR3GO>zIHAWnjZ$MVM02x<c)+ONr{agi
zagEveMFORWgAl6)G7ebX>9M}C!|cjD&3cX2;S;>`_UrU-Z}4kDNV5rpL60D=Q){)j
zwYiD#Tt1<T5^}aAQcTLKZZh+gA}=)QU_{x-sTs}o<r{2NEm@xJmQ((e0XsPs8c$b`
z?Eb8zn#Rf)J7(~#CAUtGG%*%csPRN@$%_Ij3T&r?pQY4FjTZ)Zo<u5VP8tRQS)Sv`
z0$sGJ#!a?|n;hAHjPvil&&*Pt6Gu-I#wjcFhq&<WHl6K|$DcaN>|BjY7ca1SW0M=#
z)>yl}PPgCV3(r5ncKbHhZd~N4r=H{KPd&@rp+(;K@k=Zne}s5$244vpI8(?<R6`p{
zX~5<=##n+-fl%bBroA(u-|rEI0rf_WIBKCpg)$Ws%Htb<^nL!p-~Wen+FgdD6y?|0
zci<QYm*<F7Oi`4?HOcZyi>=MujJhKNTO|^LSW1$#jVF4<k(=TjXmjD}LNNBGvm{9}
z26YZ4J)G0aj-C4_;A#(uzpG63li7rOOIj1(u>UbeQ53|XkJcp{8|xfCbdY0@9O2#f
zF90qwEH4Uzz$ecOeBYxe%IRm8QnI<Z!G#MKXjEs3qKG^%#&!k)moHyt|GWEm?6K2e
zom4?ew+Q@6jN={qh<oSliO1Vdt2g|x@%pRhXg22YR6u(tVc0Lwg~S*SDFOrvqYW6B
zYGSRKvbPv9W+)U!xX32=@g{7dE59mh@J&gGap$>^#KIB^OMpR?nzT2dzqQ3+W{yf+
zCyXj?F)t%LA)N=^I2_bWv|Ll0xIG}qiiv^LeI)|ZeR>yD_8{l<{T4R&d(4N8>HGW?
zhw@=DczY9D?zVNAoUcL%%=kv&)*==ac!Z4x^(-eDBy`LMd3WR<kHvK;!-pZy$@yXL
zV|O&wUt*^>G%LpGz}B(0z7!~<DMka5UYBIhr!vk44^Y@5K^6nZ`sBS1#m*`!?Gs=y
zI>WP?MzzA?+&nv*ZR)ie^D`BcPN+wMQ;!^^IaBBA&6~XZ^2>C)BeJC6!n^PAm0$P;
zR&U%KGv<b}$XLC3gLmG3i%)*)lO&@7?e-3)G%oYL)TE;vJcSH00zV}5e9BzYZMTWT
zh`HGX8nr5$+dFsIR0V;{Qh)T($N817{~EvdFMt0h9@mcq>clvtUP_<p+%kuM{<G|>
zH?W;Hw^lbeG`B!&Wrno3?dsW40Lrzv8|{!h(h^mw1nL02?RApQ7BL1JR$1HF#%Dy4
zDWs7|<>5RYTw>egh9M;wp%K!D9Bk=2GgS~#tt!eQC#c3Wnsa1>0U*)7izK^!<1(if
z0<;PUa>3JI_%xsRGhgNh|KgwV#yc<Z7*)E1oI{`bdA5hQ7#c%0uHgq2$JSt6e9k!9
z$|BtgMidCG7<MxBqQ#dXGxMjpadDL_FTpig@1Ex3j&DqE*}FHxJIfe%JF)){HO!>R
z!?;zNM4-lbeT*r0X*`5y_ZX#0w01Tq0`P=4{zzj1uC@#YBS<rf(z39)i0AtVA<(5p
z7>y}&q)73Er5a@nZr!9hv&c-PLM^Bv3dPO!+dTKl=h<9+gPrw+Z~oEuc=U<=eD<@?
zaCLc&_usw3_D+X-eUZHIxN&ohXP$kG`fP+Q;Y8yYvb4<9s(4Bf*8=)$ZTh_~l_<v9
zr)mc69eUea<e|i>03}0e)f&tD_7TT1QRHI0ZrohuyWjmIHa52T!4H1Km8&;s&CIg6
zbby7GLkxxq>pL5)%r5fiV`qrtI=9w0Xm5{DMZk<dPpc}BCMQd7AjLXyJh!W{0*bP5
zhTYZyx+F=)paZPE`<(gd#v}@G9E0sLKjp;M<io+py;`H)-`_nMec0UOE`yI&8vDt^
zI&VegA#fSGw{G3y%$YNsI(d?_Z@o<rjIB=c0#8ZGQlq3K&kH)84s&yJ2qCy};|2?}
zD?It=(__no94TREr_DR>yvxA@D>R#Rj4dfkCmt`$66d)k@A_zav`y2`Defvujej#`
zFdpuB{nhhSE9-<oOj&4p-2us{q$~`^NM|T7Jd7B}I8h>u#A=I?#xZr{d)S0UbLDvr
zMk92ICklK@Dp;ygP?0_%5?=_UDZm(XQqbG#(5`P%uQv&!&@qEjVx$wn3#%r3%CY-m
zvhW<wQDeAbh2VZIY99pb9u&|`USsdSC+ySS_qhB0epcr0e$7b_W85vH*!%iUzAy%9
zQedS5sqkWhU#l~-xX7qCB<&~DJYY`<CJi%_*AVyImnQZn_8v0Mo%y@lV}uy{qKff6
zWblwEQ)1GDY|tZ1hm<-aumx3JQYKyU?ixvVlVN8Il?-Tj5^XfP7~u<vr#ueq+s6;T
z|1N<^DTh7YfAeMDyn3CpS8mbEHSJ+RVI-BvNBEXX6tZt=iKm`;jPJhq18S8j07)|7
z+O^BnY88$iJ;G=-aI!MxQ<NE5mZMUiYQuFL2Yx`JGe&~}JKI|*>CtR9Nz#msjSY?*
zIl^Et0HD>HWo~YcPk!=K4|F^~GE*mz&U(W0IB@a=^H?w=s&jQ#R$5e=F*fZYMTE8%
zbh)QgC&88&twAa3^K<kzH^@6(s94r+@8C<HvM4dupnM<c`&j30h$=m_(BuNFb?wf0
z9#$E&DO}kvj41OQN<$n(*svhWbByQF?QCIf%HgB?Nlu<(w38AF%gM){<<QCj{{DaX
z*XW(S#Hr`L#Kn_8hs`z3$PikCR}N6=0Fogl>A<K>Iqt5px@43Z1_KB^g$g{hu%zuZ
zZeP9P0+1!do`)Cs)S3-u=jNPUM=~PMQbwZ@{Uo8??NOF{wKD(XGJAKI^!BcHEn_Bc
zEJC-&`OsMlwlq#iZ5`m2Q;_Od+re-|VJvByQ*So$!T?`+SZzql0#913&53+TV*4Ce
zS)^k!2Au?3!p=sQ=FA)lpW$Fgtun(P9q`)A7g)Q!&508yc>ar@<HE&@1ffU0-s0I$
zK0#Qq7@Hvdh+<@jR6x@0GiZ14ihv>+;!DB#^Y0@oF=4erWJ~6jW~nUBpuz~93#2FM
zwmZD``Wu`*dzSO(&$GF?#f1x(34@qcYnI0zJ4GB<L3j+ZA;(T0=aWzUS#(<P-W!)W
z_vSkkrJzxrq0=4`SjoZG48`CA>F6eN^#S$zBEGNol<+m4r!dAa91h1Uokob=j=qm;
z+^zrFI}z?JvEDiH-CYXZwfT6^@i3YL?0C`dd!WmLg@pw=ogO<oJ#O8)#gU`OSy^6Y
zV{-#>$CP57S+TW@l9Wo4xSMY>;?m^{96NjrKT!C-M=~6Muw1%yiF4-<^32mu5_pO{
z*I4VCIHVGHfb_kb+;kGW=l_%6+o?^6ojmm+jqS~pj&w~XMr*RPpv*Pa?$tor>0~`6
zS||Q<*0LL_E~kLb8s`rN7SCz|3aUa-lRgzG2oZQzAgyqH76S5NPIt3SZ)S%muAssY
z9Yy$2I86+&fJ>yBnBTbjltduR7+%UTxVsOlHesUfZkBjZuyptN?5Z%kf4k@JAMzS|
zJDl%b_}u^B_kyIKVhAza4|@%{Crll<bJi1*Y+OpzSg;1)Q&d__ln|t&lp@K=w!27E
zjDfcqgSI=jbyDqNQrXeCYht`71mFuOaW-00e}1uhzmY;>ti=P0(SWSqVKCUCS@f_;
zL^(_`?RAQsRYcK6=n<Z$DW!0W3}H~(;7gCtSIo^yOjM$W9Y&*+?#-K&odj*crh>;F
zdxYaBPxIo7KjO<@`Vz@#$o_pRtSm3!DWBD~bt+MXciwpu&yy@J&NDwhKVEb@eRxrp
zNG~NFrc~kzRw@(%;d|(^WM_L5&+~}m5VLI<42G22BBV#X<_wF&Fn*xJ`jJ7Mk`iYE
zDXGSFYDGr<=t;c&`w3)5+P;A;QYvxejC#j;LAxb8>FguR0%J<5%@%|04a!l^dDCSD
z^+pBM*hySDua}}6QKpI_@fewqtQ06R{76&rAe=N-C=X+DR9d1cE^aExQ}QffdwY|5
zqe`V&r}fxlyz}zAw0E4%L1lg)hoAm5=dNGi#!iR#Uc1Wqw>Al@P2zgMFa64=$-RV&
zH*PRjJ;B_H$AJ}(L)-5{c8(-bbcTwVrbko>$uSIXyn{DP`Q%ei^O?^+$LBuxS&pAL
z!Q%2V^+v;$c(Rl%P06y9PPfBrufN83zVlsv_@f_l{rWAkJa<Crk8+?M;_y5VUonni
zab1GOG8S*|8lO%tTQgC%xSPBf*Y8U0baO>nVsRblxdy51R63y@^+{1=)We88F9`xe
zd$`Tq>>(~+US)Rv6z`vXmrwuv)0}*4g>S$34SwZUzsP_8-~S;-`^+{TW$o$)ovrr>
zt7m!q=|_mG1@&ggU@)XMyF{7~Q6eYseX1%TH3_}#Et2gVnQLST^=h4<Rio0V6O1zQ
zw9n1k*V)+K;`-I=T)c3B_uhSvOP4Ow?+@^OpA#of@~KZf$;?cPN~KO1R*0htVN}Df
z&e9*|y!rN7)^FS<>q*KoAo6EO2NIjdRN@&z8Pe~rBk0knS)y1>EgdYDEXxT3k0eRP
zpe~z6VU?wMD3&IZII=NahBgymfBy;U!zRN$bKM<}|A`>&F6QIm4M9;jZQ|0>0;@&A
z?X@*3)f!Jc@i?!&{yL-KkSx#9#^5Q1Hck>?jiuM?;dvgFN`+3R&Gj4CICks=VHlE*
zhD1?Jr`zWJ_s+9#Wrc(LSH>DccE1Z}x@zsDJUQi^COe3|#n$OAv6JX3`{6X_Pce-4
zkpvhtrks@W$M<b%$L6UPi^dp%=PR6z$pnljcfIccqppd<eKK#Xor`Tr1x19xL%F<8
zBd}H=g)4s>Ye>2SI%`{0Diyp+j8~};Rbnt6PDH<}`4ZBp_9PZ#EpnV2JYn$eF0TSp
zTa&#_KR;=FXvuX?C1Ni?`-zO4*abZw2JAja|Ms(v`!z`1^$52VH*QsA0V@>NSKuik
zF++2dFzWWPS%Dd(&aZW{y$^5+PZoJXI6EX)|C%)3*zqa9YiomX@rk~0c24P_M>5!=
zOm-;ylC0lB^|lC;E}lpc{@5-{Xfzs$!AB7ZMZH$z;E{uLdVMO=G9M`pwIX)5Qi|L%
zG?uJC;NZ%Be(hJjj<ry$Rtchz!-o%Y;lc$DAK1_O`UcjPy!zV97*q1)FZ}|uv$Je&
zZsPkMx^zu#gI=FFiU|Awr6ji0F5c1e>Gyk-#!?g+>+2h=?Au40W^C_lQ?Jzt{P4qf
z<am5^P^T1VgVf*^7B5NZ_t%+?JqnRRJ|YkTPb!2z?f@%|o79cQYK_qbsWoMqA$%Xd
z5>bwZ)MAgVVV{|L;DouppwI<+*rz*K<jTeZ*VCAyzRY@OL@fj_X)~)jJhs$lsaX&!
zLsYMz`-$r~lM1Au*V!V=Ml>3WRN@9l&OA<Y_8_g5CfekPpw3gD|2*ga&40_hEMT<F
z(DVssT9^QWT7=34m#?nz%4Uj0v9iC$UpZqixurc&^fN_M28_Ib&C<urF7ul|_j#Jf
zpP*8&Q>j;|HD`#cHKMqJ=XnHSg?iInr2El4_V|<h<=_0PY;0`u!V53(Pyfk3;e{7p
zq(2;eV2k+2IU?l~DkzM=NarWCTk9CFn#`U&KMO)zq&AcijB>q&ZsSX!Kq=34O9dgL
zQc)7%g`VrGL!dE~evYxFEB9(p`G}*<IkvKrcHZHQS6(M<N=`g}nis$OCUf&G$|R!S
z%c=VN@kPSwmA6=3{T7cLU*i1vb3FFMW1QGL!J#7uS--UkrC_E~rL{bdsCuYK68ItM
zXhfD~<T~ZOcP?=C-E*w1-Qv>qE3B?=vT?h^V2~oLN9fP9IP(a#N|UC)!c4eCpnbAV
zi47I9>e1Wk(e7<>`}P*;sL$!+r%($OZr^CrYv&ZbjD@Jd!MPZf-eTChNHxsyBM(3H
zD0EJ!0(4or=oo9ri;^VC#vPN+4AyG<KsV8!JRFdeV_~|BB(tsy?ml{=yGvP<-@|S>
z@h345caOd1!w;hoKnUD<he0Vxx7(xDYH{@FA<mt@$gNwqSXx@*<cVXPJ9myS^eM}d
z(I~+Jr7rM&NtP9iMgyWKBuz(Lx^#i#$B*NAk~nr<%3&C=x_XOC7q74|H-`}JKIFix
zbK#$S{=GT6cD$h<wrB2zdJkt<;~_jBtZ<I^V=yL_yOy;n05gsQ6VnptE+}xv%gHi!
zkG;T6*q#tXK+QlbEunw_Io90*W66faHD~Eu)7jc0hy#ctOc>xr0nxa_(jmY}Ug=%n
zcb77|?`OTE+VBxT9o%^??=4Zf*Soj#{DXkt9oNlXjQ0~ttb5tLpLy)rwhv_a?Kl#0
z+=;D?MJPcKxCX8ujL=!mOm9FrOc>=QCNG`T>L+i&bg>9yQ0HzrQ}?>*Bggn~0xo?7
zRzaaPdACE}*~AVu(SCv6-XThQRAh-C7*K*z3AAxC5+M~5g;a{DQf2?a{UjI8QBj6K
zDjsRguom~|_cbxFv$jopyUlZ-|2!8io+r&l1aZjelgCMh13cxErzyQ&52N9=S6*RZ
zc7ao;&QNAK+uPeH;bDzt&~q_^fge(-#ehMV1-f)rSFR~Y(eL+Z42CG>)9(!!4lePa
zH}{VO>fDZ(V|xG%$|5}<r92FhFsh=v7F(1qL1Hu{Zq8r>A5~iNQSMwl1sJ7}floFX
z5(Oc-NU4Q_bz`afPDiK9jC>Gtef=13bPkd1JHY-=FJp{i>%tY5j_jveizs^|zW3@&
z93O1)`1~ar%@`57ga@S*N(hF-0ZJ-7<>7k)UKq1<Xb$NYXr1F*SU7o#aCVN!<~(xj
zD7U(s{7?VJU!>WpvUY0&zgp$1U;iT4udP#-nz!COPhTtSXh>Qp+7-c?&3NU?Dt7iL
zUF(y6_nXKUzeA-`XJvUmOH2D%TwG!Q{{6IOTPRPlva+An%p6L2Ai)b@W&Z*G!e9DJ
zeC=z$%!}Xq9{<Jv?eFly4}Lh-=kdS?_-7qa7!gZ>C!OhVlBO6dQND|<QbJJbax8c?
zn4-ky+8ytpBvS$_eLPPRT8XqyuT@%sQ4O|s61KWIt%d5?0>wiG0l7>Ok;an+3$0l$
zU%AZjW5-BtCQxg!e`Oyktnj;k@NH(|8b^*C;`X&3qk+a23ZWXTv>t<?jmUBu@hrkf
z-hJzRbRk(>o+XYVW@ctECSz@5jn$jiuqLN0OTs9?Si`^g7r)0#KYoSIP8T6PYL#Uo
ze;>2em|8qbqqc~&F<D}0-!P2U1c)4|N(w!s*{Wb|!Jsz)1SgI@#$Wr(f0fG@*V#JU
z<&7V|#qI75T1pe?+w`_xqsXsOpZBRXDui*BZht@+2jpo%QEDt=oF`ZytVCL8@+=Yf
zo?;TD{?m@hs(D&G5powyd<ZzZbD|s@5bt@;2iZs5Z=$^4V?OGbzU~+wN+}R1tkz^%
zPQO3k>8GBg)9KOeb-8r;GEYA7INe^C_4Rd>rx=bBq!4cZN<k3%Xj76TBWjg8J3HH4
zym+2ZeBzT_Pp*>}Ildo|XE|rjzRihak1#jia*ab$PL0Tanv(MF<0o9WTSz?&=0=wm
z>j_}em}W4x1mW^zj4d6G^2T7_xClS14MrKX0pzaPG}a=GMp}(0AQVs$f|^i-Lb}j%
zBoZ~vkj`yNQI^z$DoR2&OzCd!;MZ%2N=Q@*k)A*a>q>Ux<LQ8~`)}fuH`zb#YMICY
zm;&k#I?q08`E?5Dxc}njehZpAU+BYM|9;!_Kuq@ojt|n)?R^hr$N9c?{CL269xCvm
zGz9f3^|=;lr$;dyI^CKvE=%@9-<v5*EvE0wIOfZx3^_0-rjKWg=cTZC!Zl0iQ4clh
z5G8G-Us8^C&_#;&1)>CBdnf^7Y)CAX!63-6$T*<ZtPyyEx-XGJVe1u+&s7-=`;<dV
zFU`4j=`xQ$`6S0roZ#Grb7Pb7kb?*Jv$?s=++2$^O-R$6BpLFZ?|qwEy}`=zKKlI;
zS)L)4PgaiTwmSqtfKqcvPoZr|Q8*7%Vcpp~7z{Xj_AP4l2FYl|^=sE3IQ#V@fx2Bg
zg%WK`tb*#?3`0>8%~cR-8!rseQAlVlS*J^Xkl-)PVGX369s*}*3(jlC^Fy*E#ScO}
zQ!vU5MJ`Bmf=LC}*XMXWKf=P7o@MQJm+yc39ePQMHE{Xj4q>c$`g6xw{Nm?XyKocH
zww$r&sJB9F;TjaIwG?^5<O-6?1LcE;PJ0_wYY>DpR2uW_J9L=g+C}O<thc-TlYjKD
z`Gv1OL#67Gq?$v=mU!mVkJ4!mc=N3*Y!5B{-hj(0<DAiS`yT0)9{sbgP~;^6nooZ2
zGyL{{^tU;F@-eEl8lF-J0cn=-fBMh<kF<9>oId>+U-<l&ICbg_LEtjpg!HI2TKwWK
z{SwbS`z-(XAOBPSi@)=KXMJPy<Iqaqcldrt;0gRNB-5Hvwnb4C)T&kISvOJE3D*%<
z<RxWNKq<+sCCA->Jt0v-68M7LfFA^OGq9@4C>Lm}@O+QZ_wjvqdp5d2%7}^|GuI5*
zzI~Yki>G;KbDP`ODll8+xleqCM-HzLO363=&8uX|P12#o6qd5|iTzom&gg9#iY4Wm
z8gj$&6Gteu=El`qy!6r!$<mBwv&P}Whfto+Y-^TIr_YVc1KOJsC6}0OF4L+XAo3R}
za%am?B>q?<qKM-NB_)+=1u*mnAw?>I5^YAzFVygS$@<zl^=iO+meGtOy2`N7@K9!l
zVfQ+|ms79TsMTW-3O@+Z#*n7zlvT*GjB=cuz_?thl*Hp_2X!!Ej`jk(J#5llkAuB7
zCSttmpG=}3JlWm-I(L+<9un!eSHD-5g-eAQ|9&tSpp@kF>685M$3J3wdyCDjO-`IR
zPPf}ZYfT)7BuO>~x{@dg7>z~@h68Gq8WzpF@4dxmp8E`yN{sGntZ^Ep&CM;&pTEdw
zW<NQ#a@n=zVEkF#KLz3gC-~jZGAX`JRb>3cCq3Yp@+v}1Ge3p(z?KN55PC9Ij2S6Y
zBIPbv(#EmuSl6kov_|Tj(5A#>M6#f&EHy8jW^>A^=1sd2zi=rsrqp<NC?Uyq2K1vf
zqR69+1GEx&O5%k+S_|};A(K*$69_<#wOo_<L2I3azySw2ZvJt)Kdl`A5n7`>4^O(R
z(7Vs;z2k1C&ssY!58fY$-LL$+`^Wo~YR4zVuGb>$GNI$`7wV@x(tF<e4yN_q&nATH
zc2|tQ#@l7A!C25r;z>nNiKw+^$QK5b!y)}FBO8o}y$HdFF?M@&2jj?K2aoNZ>p@Hv
z2MDK=ln5sW(3(IA>T$q<YRo?0QWY6_1k(2~R#O^{S1LT~T-z-YBLs!CWL6V|5!HH?
znMReGuZaBsRl=#|MM@i!U*DlspP}OUY;LS`>haTPO0HbFjFghOxjAm%UL%SFCbm<)
zA2R3<_`wf;$S?hwuX5zb5iVW4OkQSq(j!X~cDA-rO3-LFK)NVHF9^pBsE3e}8#ixI
zmWDJ*xV3tV`T4~MI<7wzsMAJcbb%^!R4S>3A;b1Ib(!HQ!N`}WFu){3JS=8yn_yv%
zLP$(eAh2j-$kGx&P!I)p7<_HOj~J91lZ?nZ`}pqWK1K(sbT_WjX>anCKlcn#Ea-M}
zHn&rjj_zk=u|YZ*u`ZXncJ?!b8$C{hx2RVul(yJijR@DlXM~543Y866fA3`)bBoZ>
zh#+QW<sdtkU!|f4)EkOxSFiHl{nIwje(E^KPafgixmC`;|0B-7yNWC}`Tn=gvbNUe
zt+ZgR&;+KyBgK>MrrTWE$A9p@_>Xz!$<Lsoz^yJP7lyH1IDdg|e~W83-sQV5{2R`k
zeu`iD`oG2Fk3WkadI%)Sk64(W=imSL{^y)|^bG%-|M)-Q!i7s8<<R|1!wQM#`=BKK
zffMITPmTSA1j3qe<cq>+OO_c-0R<L~z<3Z!P<Z&#QlRmIh*7U(lqoj0`}9W*W~!lM
z1q?_$gc0t$go6j>ID2jrk<QSFBicLb3^!)^vw!BZTs(gh?FY;)M9eLObd_e<FR9fQ
z*=gTKiW=K%1KumH@ylO(o&$^fxOVP3tLry-@%u0G&O7h&H~;4Ul0%1%U=1uS&N1j0
z{OIi~47xEhwNo5Ac#P2?N0fE)qKehd)Y?cx7+Rv(qbxJBQloOqPP<F3R!3`(QI;Zn
z%U}E}zs?g+?jxjvMY4W*oz0t9XjLq=Af>Z<nc?6TD>ELIT9v3$1EC1R$eBE+>2xB>
z^L)HGnmRlngvI+HZ>W!Qysc&8g|?UFFh5}>vA1c$S{ET_v3JM@c0t|bIrqNzo>%;t
zKg-^slyWvD%9Z>^qakZ+Yn(oFhIik8k6y3OwQDz6Sy|z!r=H@~S6^jrZkA5R38;nG
zo#eBuV3Z8-q#{i{e)z*5@a3QX3TNMZ%e_zLOhw;#<4sPSIL5+ai?S?Go)dXb5bNE$
zbUJ^J|F!qe4nzTC&Df;u`B0i;Tx5y6{%{dnRyYeEAv}c8<5Ij^a7i5SHsguRn$mq3
zQzA`?M@e8cRjaAXl9-af3Owt%Pc>#^924l?SK!Q{8OIQMLLzcYy4_(IE0Rh?9EA`?
zh$wW1xW>ZRqt4=Dt?YPTP17grL>w|%qD=PZSXVbLJx&%aAI747m`?7eYqFSZ(VeWT
zy<<yvefcJv^%3G~KXmME5}9n%q)}p`)pa0JqTEqO8%trGq$&&pWE4<o)KSZe$TY(Y
z6Y^2=li!;Ou!PCNQI458GrsRkcj*||IgO`4YlG(tnlp1ea`Gsr4=l432rLCrEx>va
zMV6to#WSD^>$<B2loE0&C?!~7h$|K5W@oW^M(hWaQZgS`I5->9E_}LSgKDkLU^rxL
zeT~_LId0urCCgLlbyp>N?X@=uf?%9NBk`2a)hkzc<&{_X{O5j-g@q+Hwr-)(cuFxG
zxGY*Bz>fl~P-rV?&PXbim}axd($X^5uHE43)$0^_!P4S`0r`)nsWT-`9#tT8MraE3
za0nh$l!c@qEpmi5lt?^3#0w-@IYK8nK`mssGenOalZ7wIv(zOHgfX$I5QYhvQZ#MA
zJKY7gDlu_mKv77RmJhJCy~E1kj8jh?<B4b=X%=ze?VB7}S)^Hud99xydxx3rtWvGH
zE*zIx>O{tQUNW+Rz)A+Xr1$!ph)?~3^I%ef*5VS^lY(0J2A_HMSu$%0!z%B-`yS^n
zUB_yVL3e;BGUCoT&R#purOeR7$;PBRrWj9xh0lNK=kbDoH(!5^we?k2*KTlYeU)Cf
z%l7s*KmOs50ZkNz)S59Ho9n#(#w&dKQ=jKA{)OM-=#i5ya!G;`f}j7~XZf%G-rwc7
z|JL8;>|5_lC*Fre$KO2)YsrnXg0KiiX-0pL<Ao#QAV3M_x{!sy6qX`4WVykV60Iag
zfzk#uw(Hl@P-=^mA%!UzWF;HDKEqK-qY_f&DTOUj!r%+nBva}HOq-SEh<C2P&HmHR
z(cSE@dFcx8{ooL5-79SD+~#Xv{dMMNXZXWEc$usmk@pO>%93*&mzoRfH(9;$PnnsM
zJo?mOe(o2a<Jo6VQLA}8a`XW6OY;odnj2Spyz$yOe&?V45wW+-;-LkSks<33iK7P6
zM$VT|rWl)oRag|IF7SeoLYEBFgqhhnhQowvy~>~e%U@-qcZ)PX#`adikH7sAm(ISA
z%(@&}p1}-m(cOHHI7nDpnx#H2y;^IKLXxEkSz3%4FsDnkb}DT|NLLy|Jp>^1AzGvd
z{rz6wE9|{s?kBy*$8k1K$~$r#xu&(|=ISa>Jn;mNKKdv>`q4|Y+ifmizRVZC@C7bj
zyoj}y0|yRp^X4l3{s_;@saC6`Y07XoB=SR|DB{gGU+4KR{2X(0^K5Nxqm+-ehV}JL
z-hcmnnlqmu3_T_~v^ziG-p_8Qzq@-MOiIajahwn3+PAjE7&R`T+C5vcXI8b=MTAKZ
z(#MFghnX&2mn)!bfhrSx>+&}1LQ<C=fwU;bFJX6IZc^%>GP;wJoZa=9%SH+9^-Y2(
zBB)dmK}e|-zL3aBVTHyDfw8-BZ+C*@djZ}3N|<}~brYbo_xt_EE@SuM*WCxPc<$$W
z{qrW<GXYumvfV%Xb-t^Gyc1~N`Q<Xa#RPG=Ct|cKxmzm{!o!Or{MlJ_n$zp{==X={
zv_y`<`@{|lISxu0v+;~?S^hZR)mX=-IbTEfp4M8|tb-n#pldWa)k@5P!wWq1<S7;x
z>X>XqksCxU!uKPr@d$i{$~EW$qf4~4XeBVdOGhfQjH(|xDaCf1nhGiNf-)NtT6kn-
zhPT%{47YC+x2nAQ@~b@Y%#$2Cc!2lbeFy0&9yxZHwYA%vKX;L8T%|0az0;>rpW&64
zUM7wsp8wL%F)Dg=yE%a$kfkYuPLFC7QL|O7RAfnt=Xo^eLWHy|EX;A~(j|7<TeN0c
zlsbQ)<NA>sC16aQ=bBJRN@KBu0bwm5D@T;W0fA6>)d-WO6zPz#Q6tF<0x!l7V~Q*z
zNk{m~!}AnrI&xRCAK;Ty^#w`4%5e4=pZe^h)aGVMwsO`s))}S=3-iZ#<IQ(C{rKZN
z^XW%8acUp$ynd0)lzjU6<DC0_%dK|6OmjfJ8aM&29&fub2;Xyo!&;NI+sI^uwk4ic
zgbVxG>1*m&-k{m+Q#4no`ip$=iG64*8Hteftu5a9{vUGX`un_}7zX1Nu8^PtkGyaJ
z>tOlD_rA-UZ@)peyG>ELOj~1}qj{R96s0pnwYG=61&wBf*|{#;+v}V?`wGAD=l^ZK
z_SHW}rCPxRA$d07@y8zL|M~y;JN$cp^S62D-S<AqA^garC=I0)C}S}=gWx>33<n8*
z!$WxP;?!DGW+{1IP?i$#oGzy{2s9dlQGn8DYbb09zE5E@l0q}eb0Ef=5^F3{`XD`w
zk(eT*EVhZG7R~C2o%MH#?JUDi$P54a8+`RQ{|v`Z9itLQXq&VDU=w2nJ6i>LX^<+$
zSXf**LRQ@3`jzwi_W$~?v3zWnx8Hw>#~wdU6e-qM*Xj34UjETLeEXX}CQS@+&_vjX
zUbjuHI!B(AAWA&t^nu0%810c}pml&Wl0xUiaY$D5P`=^BsUswLm(PFkNnZTci|nf(
zVzAYtySmQ4dW4s6v$py++2}g^=R;ZzpD^-VC#!IJx#4g`n&x1&6S7+~UD+z-n28{e
z7~co7pXrzjcC{_Qu7LR?P9#4``!k);_l!HOJ|6BU%MztrMsZO%W8v+sEv{a@%83)l
zxOC|P>+9RxzI~fZmoD&yFMN*ge)qd5<<n|4&{~sa8Byd@sYE2h5n#x(j3`pP@WKmx
z^_RcS?b~a_afRWqPqk9vt+(Fd#L1%^I<%jn%w5FR9<X%BF0d1RZZELf-6J2D>cU!$
zWx~ddtxT*ySn0kSG2J#}En;l*;@k}~1R90GGZ}uF5Sf&!)YKJJq$CiotF<tU-(w70
z_jEN{F|K0l`Ro>pLXr<g?A+cUuGSG@NURiADFUSsG6bQWF|kmt=qiP?#4vY)76<n3
zH6osXe3x9o^mliU4`%9if#XjacRb(T3EuYt$@{(M!(6<(h8PFNwKlHoCmahVgkyip
z(x9cFl$KH$r}-9=KuDAUot1b}p~4D+fN;SQ_eaD@PXF2s0#8Aqv8BcH1GIq5l;8_T
z6bOs&2wQ?NKpY83(j4jggb30i#d-oS5D3pungs1x4xe7)xz9gMy&jS1A<B~o>4CQB
zQAVI7p0Ie4PpKf&8EKX|uU#uBi_(>`m51lW3^L2sD5uC$tjq{x#&XT$=6ajNC_1ZG
z5u;_UHCjA&@+gg3h0UFHg3xm2v15#eLvCEzAeK!QY88gV5ndp9<Lr+x;gDyZewHE&
z>G%3XeuU8lo43};TTN=sI!Hx2956F8!`y6xYlBsm_bqa)e~4SFtMmpRgQdyzWU!W?
z7!V{Ih%(357M<0dt9VhO0?(NaOF^Dy1oL&Lm2h+d!V9puCePDx_`6TP(;-C@h>S=W
zhI!0|Jfr=im)RE`;rM*P;o1_v`^}47c&~$uhP?QlAM(`Gn;be^BTsVHR#TSe=Gm5p
z&S-|g@H(OI>^Mvbo`fPxP@#tr5>aSmmSOrm@*=?#G2!eYqaj53fb7!u=xk?nOG#-I
znF`sQJIZRNDcY-COcVO!ARIf+sU4)n9sBCqwVT|$al`p8OINO(T*X>jw3X9NGY}-B
zAv@b6TC){KqlAC-U;kY$U%t%0`|tb>7M7L?tRNo@dGfKx`M>;U|7ZSZfBkQ8^VaH5
za~L0J<hctsMiv+Y$_p`O$Z)8M!wg|VEMUqKUAoe#Q10`%MV`TyZXzgsOXz`6k|fE&
z4=6F@c|p6=CSM4tc>?7J&R#*|+Si34j0#e{!)!y4_SR`to7gO&-<SN!zxpm;`-{Iy
zzdPiKCm*3&ud>zoAz|j&EK>+<sTd|XK~y2GwwRk+;QU*!l8y!(f8+?saKO#$8~osf
zvwY{<FJpB~=+BZ26B475)>34N1DzrPQyoug5^Dl<4#q0-$!gNG<hsZ6&wrXHo;pdj
z<@4Qdy}_A7kMXVF`z|-%x=d60RBXZKwfE_4oTnb9%+JO|p(Y3<+G<Rx$uh?Z<#`U)
zpmpIGTH*4TJx}8KPVa>u0)wCFm`^NJq!42r5+Ix=?1QG1y$r|w%2&IS)LzE!PC@p^
z#WwG5Blw=&6}JmPo)uiWbcy{3_i^;-QTqKpy<U&^-+!MoXU?#^yv(gzx2V_a1VO;9
zTWh3gO08BS2m`V#BaW*SMb5Qr*XZ?nG@CQzS&k5bxKd$#{We#xUSnx#fk~3auEuKo
z^BrY3_tU+WCU*kf2}ATS8YL_myC++9{D#xAr4RyZoYZ36ql!U;Es(mzvpK%a2ux0B
z5~`x0E-f|Xnk0mAO3#vAa3YLFD);(!LOHX$e~fVNA4rd~C>U;c*{-ai;tC=NQDNjF
zmxMruA$U%!Y+&3zxeMs-V2#C{j1*$G3)YNnZ1#@3n<Dlek6pkee<I?(Uo+5NM)A(F
z?1wFe#BM46A&>ik)=7bULPlC^vD#p@MtDAQ+_5R7k5`mLvvbt?eTr^}qT5GVOXT^E
zDJu=q!v)7#ixw7b$7N{i%+*UPsMqV{WkH@Nc)r9+*MVJXO>?fwsmC7S`01ml5HdY-
zhV0Vmd8HojOKmA_NogIyB7F~07S4sE#BG;Sn4%;MBmB5RKS^1bn_-xZ@MTH2v&R0}
zDwlc#;_4h(S}^Pm*xqV$<nR%U)O31nnlnuvIra#v*EUH<34vFk8pjkequ1;3-urJ8
z#SsS<juKZagPzmu27ym998wkqQMH1Ql3~3;T(48B#Vjt(0w+)sx2S)B@sU8C60Tnf
zYY37ZM1K`qrX&T#asf|C{J^Iy3-awQHZPGXMEU`~a;v6NYlQL$f`}qB6nRFGr37I}
z(od*|9=??nMxqIzDB0=`X~lx&d7o#WSmE-`4r!?=MgiB(uXFUkKI)Or#r9QJM}0<O
zhcx#o(t@JM$g>n%7)&5BSxQ)IVYI>IDX|a~!yc+EC{;uhS13zCf0#2{3#mj3iX1CK
zcFGzTcQ(1v?$YTG*)|}hvy!;`20Z<jwUk;Q^yDJl)hjXmK6!?~XwZpe{q~UIprFVF
zKYr=^BuUC&`pv(}sS{`Lg+iA(pMCB*{`ddsf6Cwf-~MezN&2y@;<GeG=z_8+&{{Lo
zoFN$%j7ABAL56C$)zL&;tdztLKnRKU_4p2J9P4AOvqteHT3J$)V?>G8&IisEO@tqz
zq=z-T+9P2LJOw^El~6HTQ4Gp94dIjL3Nupt$-nwO&wcSJsxuy){x(1NrDwTvZHrf4
zy-2$|WVSiOa9Ggq^jKJ2<ofk3UikiN?xHEHyzu>3`QCS5Ve|F?p<@EoARiedA!TXt
zJ%uoaC{$nzEGcnZqt{E&`1oExQ5FQDBuNX799`zjnL|AG_)(4=*@u;q?ncHp{^fVM
z{?=t`g<xg2!Eo&z`Wx5q$_@*Qip9BzYE9yK8Yu#D<Jh4*FUDoNacM|`l1?=4`-&j+
zh@uD^JeZf!hYc}KZJAE!V*qKb`OtH07uZc-Y5aQkd-See=3#d8-G8_@l5#hnRtOih
zCY4~Py~E9$tL)pipUao8lII=P*SC24?YDXI$tT&_*+B?Fv(Y3OjcB(!7;OlApHjO{
z*m|u&f6(XZwJXfcE^_1gO*|F2SUeQhuixbK>66qORkyI4>;-%G{;_xbf4n?$+ic8K
z<H#NWARmgr%7S&eW5I6ppL6h+?)n{<Ma@_fj4?=CqHIE_Ga@o#lM$0q6^6PesCWv`
zIxCE^o6GLU+$AK(!gw)d=d`usE=K{S#y66pu=KY&`0*M+91%nzMc}(-wUlG4Brw8t
zUQ5ilq<dEhbx-+nFJPO*Gx?r($G*K^>2z=7!=!}zAnfwpz;054yc^Kn4X*BPw78#r
z%10e{F1~gfo$hD?VSKFjytgriLYF9`+`?Ai0JXIEaYU`vz@!U^L64p;DY~QS5=9z!
zthILL^m2SGTzZehdIGC8gQSa40zY!mb4FMiGj$F>a)1-3kFfv964J92+GRFNsit#K
zS(eywx4CPe5%`{m40dzMjWJkl(AHqB#`Ap|%_f_-RtbWDdaaH&hCwkR@;oXYY;JE;
zJ21=kt=pJ9=L=u{98Wy?ELUz^W_^8w-k`@Y9TKTJCB=C4VDWvQB+0pPZIy~QL$lFB
z>ymCKbGg#eQk10wvD#3n)=|Ds=qZ}j3h}}HXkBpo*7^e-*N;|y)nK%MEJJj+P@+Tu
zUMLV{pKMrQFwSMV-6O2T)Rq>Vw-F^2Ic1)Y%`F9m9Xm&tC8IRO69JWq4^m>Kk4b#y
zA8D|DZHFuES#)!YFFaA@)P6y5>QP?!_C*@;EJ@MB_a(-=$>nQRlqb-7h}DLoaE5=z
zSd<XNVMJqQ0jW#!VGqw}WRYWvl3X|nT3boGpRs>#ju~kgnF_atA#bg(LSvD6Ygp?J
z-Azi0v9$PZrOTcFviJSh-4ogF4P6@;^b4+DS##xX=`sBKL;fd!>96w4)6ar12n~PX
zH~uZY_02!#cYgQxC_e@xyY}`r)gZtNe0<-hUacZmEWKWr?X5PWL4sGW62&oDk`N%M
zR(yo=C}crVD$0?;%QXf=X${7ch|nk9NsvP0DMKhVlo`SV2<5uzw835U!jKm!N-3JP
z8l@>1Y;O=Kj{~y{oLyZfDkU#{=Ov!~+~YiP<`gN0K|bQZ(HS%u{oyvkGbkB?A5g2+
zF<FCF^#uRyzxgJwynL1{mCVl`=2m+L>Bp3*#`gnEIbP{Wi$qeXR2d}+dDf>fQ)QU+
z@xmH1C>RXetn6#?mw)qXbhfuKSw?TW!?0_(`08aYy>XGnu*&|K8fE(iJGU-jl5J+{
zurM1l(+p6aW;kpkL;yxH8V$+w6k|#RE*M@a*X`#AiXiX_!@z;M2Q;Ysup!5<Kjlrv
zU{a~^L}FcO3v0(wc}(h0t^o*_eBp#sZly2AkC{x~a>7vEaVZGc3+^5Y88L}Bn{@C>
zr0@LV3at@RkY*|8&Rt@8c^?N49Aamui?wj!{6&r&KFZ1CCpmZS9M!PG!F>nGlbmFj
z(P-2#!Z1vR%*@VURmqLjD=aQ85LH65)ERb56>|IbCb!o&sn%;a3VU}8r}-6=-C<oo
z{kY=Eq+LZ?z?iAlOpLovA4a2u5N3)nT>6i2rO|Oigxdmx7?)>_bu69GIljpVNvPV4
zDj8K#P?3VV^a!Oh1I~dd&tP{U(Bw-BH?dD}o-<N$P{5RiisvS|(I}(4wmEJjiin~J
zp(Mih5pFf<o^S26tb2zxZjUJOex?6=y0G_vtG!_7p8dxM6Lt6No}GUGkT&lR0+0`b
znf<W$b4>kiXYYhbR>IvACg8W!8Y3)LInB2P@FZR(rrK)a?OQ~dg6;A)=^&%30s`q#
zlm$Y$W2CIbO4lY+N|2tw3nXRkOr<kjfU+Dsw2#w|o#y13N0^zbla@oYNvYPWlx2Z6
z2H*Exc5_}hd4=Z@cs`6v>sS|_AblULHEEHMrzw>%#8^wK)uPwlVK5jnKR=I@g3JiI
zJ0luFmCdaIy4S%}EVSmx5<_cdhK0p_WThdht&tQ7Nug=RGkD5HU#_g|XXWq;N`<U%
zZV^URs`Un?F6ehVDBL|aNiwv7cDRX@k}$3ic^<y!;><i5JkW9cXyw;_8!8p@K@Z*C
zLe2<BksE_4OUkmuQwlE(h_ovmB}Im(Tzh&sM$*Q*OExJo2K^qUEbyg|PBU5yE96!a
zs*FRe8kGYxD7nVsLW3tBKg?*b&WR@;rFCGLPkrhcRG`o%<=uDA@h|_)8<6ay<th9k
zM=D5?gkE=uHins~!tBBdexpg5r)2%Ei{(gDr`uBbxC{`|>Gny}luB!v+dDqreD4<i
zp(U*Jxpm_<`M9LWBp2^v7{*xAM04fxE%Mxur3L@+AN{|vzxCI6`pIXo)=;fh`CEVM
zZ}H9Ve3z}A_Cw!>?)DB;BK$CL9&)}Q4ty#L3;0sf>9om<oXGc4o{O`JqX;kX&?+D=
zHJLP&g92S-SkGd^n7oiAMFG~(j1;ld*gV7P3NoDBJW7N#1fGY)$CEG|4XDK}X6u$4
z>o;iC=Q+?=<l5>@4lF#vxmVvn>miFr4v>zz{Dt561-|>j>uyrLd6S*3l+&l5rOcbW
z`lB1%x@=fm+d|1@ia|~hHtB73sMkV#FCbJgGRz4>pI*01Z#cl}8p`uA6lmK=dLuML
z8Z$AAOEsQ)@)SDH8FhyoS~|hIuf4;~OSif5&Sm_5!t&f4LEh%-^|RP~gO$aIL;LF-
zJ=UT*6OrX9f)QDv>G%7Ll0I3UVzkB^H$*7qyq7%BBM1VbFd_^e4pX;BP414vShOzp
zOg>YhWz618I)sFj&e`AWVZ|mdXeY1v0h7{05`f?P3J#1a>0@n4QD~GW@IwzUl%}N9
z?$GNFICSU;SFT)TI2<tO54e2kGKUWy;<CELpg-Wq;UkO|M_j#j4Xm^OK)`U+r!_mv
z#`bOUBE|D1y<YBEE*a1pB;2@piz7!4p%l{?rd`%;JaOAeC+7s`$~%CF-CM7lWGFp^
z#oRO*MhJ|VOzv{rm}5}VBCMh`Il3s_XP_h$1+guNZAO)XnkuO)h=d`e#M2tAKv{t=
z1=70L)ml4%DqK{RW8`-Mn6_prmRC}ZD;*kN3i4i`&FeStd=K9bs7m3=u$~7}x~^Dj
zr!380a3Z8cdTJU{zn6WwUzv4pbk<%#cW-0QoyTCpo=({7`;}hZ^>Ht-+zY7gZCu#3
z4f<KfomMMMx9=VepLL+yQwk|25pc#;-K0c#3Y06yYfqvoA(5>Unt2qOWSG!TH_^s5
z%V0`;=}pgZVGPC-XerU6BsV!mY4Rc?u2eXF>I9EJahjEbE5x;s!enSltTdFSKqz<4
zYHh~-xl9?mQWn!Fw8?_e^StS2&$5g-4AELssa9#t&C=f3V1BkmvpK^k)ok}N($cWM
z)!^dwtL#7X48!h_3m2~Nc;itH96CZ&ui^y}TU%}3_|aK<qYjOFlSBIsaqP@-YOMzC
zc8~V3&+6?p_AM{cY|UY@3<o_hpoAn%Q#Lm?DT<QC#YLKPvls-ktrpTg7(4rsK;8N~
zuM^HR$ddupEJ0G49(2!Bpe??2&hx4gL6RZT5*d4lK;l(nnreY`I3gVm@DmSDND6Hk
zWhpv0Y+t#_na58Pd+W?(8${wXy^Rf$!5Yszca$?veHJSXc~NlX+6CTy{|Yy*-DYd6
zOKb5kwjUDQvQ*V|tcA_ZZIUD-2z-_nmZ;We5WY{IB@B0VsMRW%EOnqXPu(g)`Xobx
z5gsaP@YbyxnE4vb#Tj1x=1Z8ufKaYn@iC6^!kjF%+_<quQ557w&fotB|0VzaU;k|$
zJ^eVwXr6xhX@2!reuaPdU;j4`y#y<@#^yOiSzrVxPhqMRq9|Z~ZkE9J*xA`;r?XAH
zTE+7n*pWh@eC49z0zoMixix4Dd8rs=h9oWUrN?YFqV7pVQDSm~_H)OmDC5e2N=%I`
z%QC|-U~Z;Ll8(53`NteM`HSeJVD;_SC_+K6y~$_3@+BTwIn45GoliXV7&qr{v$L^^
zK(jDA$N%S_{0rJ!88+|Zdo8-#BaDv862<c3Bk0ma(C8AjHny0ZYY<lgW*T)o<<T2<
zu&SUaGiGKZ4j*0SOJDjqR+g7&)+%^_pDT7&JG}AYSvIcRWIn3!<jF$}Hm`B*`rG9F
zD;!!<99o%Wc{!$039w~}L{hC(@Cwc5`W8izA%t-a5mQ}@aqMZ}6NVvS7!idrT0Pim
z?m@<0t=q)*!Huy9-8#Dxr?CN{Jd{*eVedOZPG9Rj8oG~LPv_o4k1a6F)Skgv3qe_G
z*4Ebf)Tchl!Gi}lckUczS#tgQRgN4wz-+6<rAwFbl*fq^Cm4=~^!k0mIKopN!_kn|
zY>RHU!)P=l@WW}NNgTzbS(j^9ukq|NPtd40rY57~nkLh7;FzIuo@?W{tMTL9zwZNl
zAJV>?Onh!qS13v}4rn?j{{n1@G&z202&5*q1r=FPF`Afy$d&}w;tPwk&bePttHt78
z#BJ?#KizM8%oy}qZ4g2dC>P3J_D2jhw~4DU!jmZ9#}^I&sz^GoJ?<@QTI>`!!<{~K
zALb`_H)D72W5(L*pP+@i=Xbc$@!hZcLEW++=XG+wa`R+NN|{RSg4<~+9@i0F8jJP>
z!~wpn5-5YjFgr}Kc}c$2A<s0vl#KhpG1j=KND10^4H~8_P)e|}yvXS@r+M^=Gc;%F
z6gp#+^$8=7xDt|PC4-?$cfkmBUOECvDdHd;C&G*f!;rxDr^wanUCSJ$1bLRa%7l`%
zX6FDOLEygWY;Ci-wZ+o>9Lo!HlvZIjhe+d7EpyVXO@yp)^U6);mgfoU6^w?{XU-t3
zV0(Rw_un~BwYki(laDYrH^)|chwbe)S*E$YzDAmjICbhIt=U<uHRM_1u6<)j2SeJX
z#FOADNl>W}L^0Km!+iMSH~*OVmHh<uDzWcj@mveL@=?A|S!&F*HZSp_Xlf>zbo&&#
zK#UQ8qt+l#G78<rYQ?ah5r&e@f#hH*SgdOfBx`JT`aF8#G_U;VJU{&7b5wpMMER1l
zuV3W9{iol=w<}b_Izd?F@uyapU97Wu>7O&VU>K$;TU%`uf~ENx78Vv<`47P;PZ6DM
z;y5Nx`iN3ckfO7K&^M$;5_l212-wz=rAH6*$_p>MXfvfy<;Ta+?GAVA(bhmFSi8N=
z#OeL-{e!>DfAk;x?^xQuOz4OF*1!E%_~*a#|D`_|JXB!k0XEAhtU)M=^*yvM$wrc>
z665=dMzcyX7?Kqk0!x6PEG^OtFvh3UinR0?W*!-aevz|1fUJ~MJfBuvqv@5X(xCH#
z!b`yO@O<S~^~R7U8OjQpwI;(<ljJ+BEY&d#S-bGZoILR<HU@^9os`kdHZOnU73%x;
z;n@b2M#%Mxm&gWvKK0DgY}~rV>W%B1Idzge(`;^ZiKB=jhhAp~A$-b0Q<NFC`Yd5&
zNYZU8G9u9lR!g!XA*v~U?Kgg%<0n_}0?o`!%)j`Z-{sTKe1>ZmuAxegtLLsV*yykn
zMjV-qnf3atuDwAyy2ODM#i4^07H1-=u}_|93Y{Z8H<=e{#&9$s&r^i3ZhwpmJyHmK
z-zNw{f*>LcV|+h!i;bUg+}-Ffxp~XG02=JFa2BVTHS97Td$nG70K~gq|9_Za(MF>r
z&Q3{Kv<BsQ1YSUy8#?VS##jy>JjA6-7r|PJBB#^ou(GnkrAwFD+1cUAC!gTh;23Y7
zJv*N41j;Ikq5vi7cDpPs9Ka7e7rbC4Qb;z|H&|W0&CHqBo|4=BCi<}loV9!QE+8g)
z#XD%bhcYHs5{~_IrPe)v#Zq8x4z|Fv89|Xy^(++yHK~b(C9>m1y>%s5Io;B`&l#ph
z!PBpC$=6N*gRA~xu#^aw9w3~plW!$un$g?Xa6YF>;zyxdKnjH?6kaGm$$Nm()V4(2
zADrzyFYg9cdu@8||J?g%=04#2^mU}1?w=2P?7a-%!yR{j*6B8|+e~u5NH(ptE5%Al
zX`Dt)2!RNF!mxoZa+)hC`6yv)IHJf$M4pc@dTe*%gqBtsN?{2?!QA`|OG^u!dF%`a
z4<2wNPB{bx0!>yFlt^6Ts-oNL5vh>S3lQKyobt%BjDEjQmZn%^@Ra8~c1vdvj)f=)
zDe{~!jL^mqRVzfbDw}I-%r$3NnqQzSC267PT-)H_Or3MruHcK9rPd7PC}Y$cQJJZ+
zv$MrpZ@)!c3#ilr_U&7w-l(#*u|blT2rJmx*{0j=VoXV+QRT>yBh0i~blZ@nBRtQ;
zXib^r^mp0_V`<DUP?ZjVKFIh;psu)lopXQu`yBt}U!w3`ql2J8`JUV1p&}n9)KrYI
zhEcaqcYT}A_BMHvA&f>SMOclA>J3KPu+te48kZvB6(u{lW~bX@E?Q(ZI>)Vd8XW$G
zU#0!%LC(E%i6bX2l4X)_{QgCP>^PnZ(c0sY<MTXn_y8Av_m7z~FEKMy<@&WXv@V#R
ztFy9iKY{15(`^$P0ck=|W(aK<4Elt)=p|j|RAPbkA@D-Bla#q5`?+-GI_ukQW|T+Y
z7={m>6Wbm4qxp@sXl>ZsXyf|<3dMi*kN-aZ;cx%1@al>upL~)>A3ej5UwZkWZi_D@
zO1l7bDaN7MWkF$>0*%>Li=t9xZS6Mw;gDM7k4;Dog|&<l$tZzgrWh0@8#<+(D-a=(
zjH&q%Rh3~&Ls=9kQ#xIbaOS4g7%G(-!)~9VDDganC{mPyrFo0Y25epW9?M6b#TQMk
zcVOe<Hdiip@nn;B*5>l1i_AA>_|g|Xhp#-IdEyk^-hlVsKTm6}!mwY`?`7nfCGaDx
zg;M+Udbc@rw8hHG9G)lH+3s@U_@nH!JFKqVX7&0FKJ(m3e*DsVv>G8_`NG%OytT>N
z^>sF`-zM#ks7lFGCk~<pYg~T&ha@}iGB;ar@W2eSvxaIFq%asQT-R^ubXwc(E%MxD
zy-pe<)Of}0dES_vb7j{c2=F`~{S*AMKFXM`#`ko#PSitUSFq#=7(3R}>D}}8?g{L}
zx??BLdHBOAw7`YMTdOI{k|6M@*Q?a(6+9KPwz`S6f+R`UzkffobF*x3Zep!vef>6%
zKmG&@i*sylZcvt*<HwJ4?b>zHJf|!*l}eRtloB;#cG}x4&hK*?KJ6?|1d=>2xPJWx
zkDh*X+Wav7<m}Q_yQOM5X-u)MP-XWdI6QpzX3|MJes7eE?UU}@)K+6{fuuxHP%$Yr
zBdGg|SQ<iU@HKc=jwfn~9mg^`ur>z%cWvw4x;iudYzADj1YnRDf$Nqv;32R^Qx1m=
zwzu)3fNH&tA6F25fRP>=P*LD)QudyUll^(0lIGqH*t=P)-E@ch`__FByY@kh+x-CB
z-QIZjd)~twlh3$!`#sNfvbwUnMR_jiCofXUv4Q{e^G{0L(#bANS&$Vup68mjlrYp6
z7BCvpVMdu04D1?%PM4}uc*0=E@pFY4S14v?Id$qNr%s=s*=&&KLyXmqP{3fUcJi&l
zQW}?>Q;n<m!owDZQWxl=r0{*``POJK91h*wUX)Y0%0wSH>~@i1Fs{(KhKsrRMY4XM
z>o;!T`#!VH2B?VsAm!TaHdUc%Z`>l8pQW=tgA4@v7F@Y;h2zJMQ;i~uB%|Jl(M85^
z*hk2KTQ_ge8x5JCpQG37@uMI82!K<^PhhR3-*v2^pum`tWRzfxrL+R!dBm0K10C0o
z1nQ*GtiJjtVZFiexso6M=uK7*%rLhwk6zzl)a}#oCEYTov%SM`Fv61oaTpReLaZ@p
z<7({P?Kaj^%+w-^EM>7?V{Kj2FAFNg2rCMzQOd#eO|Jck<dIK(5zQhO&gNWvZ;N7B
zrB#>2wSZ@yJ4W3~uKug<P}}+*pE%WEnC*}xJsOP)`}XapQf*RXDT5@TWhHVrA}BK2
z8(SpR7R|DtFdAKE#1Ys?;d>#MI~~&c9M|5zLPaVxn!YiVA0yFlKS8X(B~TQFW&QRR
zaqRQ*8$aS(fAS}M?W<oWuEc!hD_`NImtT3PEJ(9n!_&E&ysWbbvBIs!b(tfCN13Gr
zzR%oTi{WrYKFUZ(1tKgdEF>eJVIfIONKb2a62nMC?8St>Par)!X-M)BUa0Y;bNnyO
zINHrhw1zltQ066mC}}pU4AMSIY7X!7*xE?AasDO3*{3+%I!)8gaIM|v!mDqQstlXu
zEbZGzU_FXa&Qp(_<WIi$4Swy<e~BxXZ!*(d<dv7t@~KaKn%7=E%i+U^c<;S)oIBs)
zFaOq8F}5HX4VY^!A_@8ZfAvj<!yd1`vdzl!A?8~%eCrQ>NIiU?eAJ`2wZ=@vr!`aI
z;KB@pwF`8wzs2_Lvn($Q4jibmIPVeoIS7xU)R@u(D@gN<cBe<D(<KUhq#QG4(scxS
zo+J!>2j~I^>U=*$NpDY=^-n+U249oOPS^=R+U-)jLsxfK`RRke-1PT*7Df{T?T-)G
zx$i<chO<(MX|-CkW@m_k3Q502DW5zmh!?9o^2j6fyB&(sF?GghR#sNHc>WUG+gm*G
z#FHF3e1ywauaXt+{CC%bWi%KdC3s3v2q*s!d>;ttv^!*JN*u@R@^E7M^RD91O`xtZ
zVyZrxl5@Knp@+1;Ce)anaKU;CCWWx${zMJJ5XzF8Nofj6RrrKb;8_>XC5!`cMo0`}
z4Hg0`Cb_Wp3Nt6kG*(E2nE2$mvrFS_FPvGe9qZdH{@C{}>-OpS9&xRXU#TL!0Of@a
z+<7i%S1Gl7KQ(5or*mLNO6jb7#yT}US-?tpUvMSwQz~_1xBU8H%?x)hboTyyH&Fku
zYvEyCC%fCb_cQJJI}>f^?)fRl@(W?X6eUIBDsXnP@Rv?@BRwA>B$YWsGj+{~CQBKX
zIVR6AMvu=cL$y)i$jQT;IC+Gn`4*lp7$rTF5`?i&R^(1{;wg&4olCxO8cHR7q>$sN
zv9aE-D8RbT(`vQKXgH+I3zz+_Jl9;IHARuRZQjEahm{KR%S)_Yy-dH?V_|-YnM%az
zBZu&m&sP5$+k<V^FP~#}w$8@ZIvay7zw|3#WnpoSnQD#oo2w`kX*R%<f_LA0hhdV^
zXf_!3yUflsDa(>K-*}zS^Eq+iI7%won_Cb<k*8ov@+70f4$}81n-5kU{Ak^;_+%r)
z#qYh$zAqkS|MEPSF0S&!*H&4mg*<k6mQk7EE6c*%9ETn`f-ilHHl)djqR6L0#v;w>
zb~|(jJ(TaU(Jy&vLsCc=>mhUj_$)QI5!)|t>04V==T33pna4SC<OstgMG1KSwHx$b
zzd?WN4Q59#bLx136fhk0snrzwmX?^Q&ruYDH0_aPInp|+bMw-5ZoT;~#}7P0Ubt08
zGR*LWA&wPEQ7~%Mxp3~X>ts_t*1`aTi#Yy}ojhLgP8Q)0I@L-EL7-e)y_uB043lcF
zwK!n6woP1(_?_SXXFT)Fvm7~mjOV}bbNuJQ|CKCzFh83I8M90Cc)DOT8j@rgS)OBy
z9BnP4bT{BsCzzll4g$h@z(6Hzcl!*6nxrf!Gz^8uw$9ke3;G6779~f7A(S#E6+YGv
zu)^R8fv3E2cCQ30DN0A5TPa*Sy_6{FCIRIsW)?yUskyO!7Bk!+o;}Kw`}T2Z?H2V$
z3sDqwuWj<;AO47y{YR)(=UHAkN^5q3&FvmWYX0o6KaUkLUt0V;gHfN`-3ry6kW)|X
zV{2o;cfRu$H*arnVEF|7K}s>wM6$x$KYW`bD@%y&oXzeUHQ#Vzp}~AJARF#*<Lvk7
zUVoLk*x|rpz~O_l%+7_x;du3_C1vUIjf4=CCdUd(wOYZ749_3O7#XJr3PVMsS|RW~
zq97pfeFEjXc&dj0>h_K)2%7-Av9w~Msl&=$iwmjl5Qy&;iQi90cgOp%n>71z4Py=S
zbF(ZhE#i3!r6i4dlU8e<_ST3ZFG#b3(P+ex!$-Jw<ud(#8)Hg(y)LcR49#YPt*tGB
zAmr%LM_60m99wcqyfB;sp}_auL}9eEQqd*4)C~H4`r`&I7~?1=NeGjC%JKbTPnFGD
z%e@&B@Zfir4<FXb@z=x*Sd8^<lVEl07IU5@5>VHMx-`T>;tBUy3$A>M#0Z6v$}x3z
zT-C+gvAK6~l@rpJ@mm9KG2?8J$H2=Pq;)o^0?4wQVW-FT?KNav!FmD0kMN?1Qtg73
zd%8LI04cGDg@ZerBkmp_4C<VZ-<`kPjUv0>aF3gO-TSw_%^P>W=EE3wUW<E)gvsPJ
z*_PdXH~~Ny9LO$9Oi`l8+l!qXLjlT12!-E(TB*^cCLavR(;@kmb3UlWAuER#Ir;cW
z_8r}i9|+Jn)|O~fx`;{xN?P)=Kxa9g=M#sqOWMgx%B*xcej%n!5qtA{18)raj3rAF
zwC60ibXnjj*W6bcD2&qu3MC1vRjTzmgJj5(0^ft>`6kv{I-`uui#OTXTIJ$dNo8?`
zewuRY$}L*;Ce^6M_U0!2UY`T|4kBd0Y^zDLS!cVmL%rTW3QMD2rB<!-?z`_21OcZ{
zox&3yTbmosals0g_pLRVzC~{^e4yj{5o5vxg|Uc!!p%!J`0{fnsnx0^FTcggY|QfV
z3{onrc01WhL54<n0<T`d4#AYpKv4Tooof=e>Xb!Eyx{RKFI{8CgS<?kD3M`+wU+&@
z5jF2ku5Vpr=MP?GWGYzcBaL9CXcOD(tTZ=SK2~+zB-xNCiU_?q=34XUvSgTS(;o~d
zN)J;O>}>XU>-G2f<guePoAZoD38Nw<@Ar^JMkNw#lm%;OHm<Kzcs?SklJ+}ezy1%t
znMxsO)@u%c<+;-UJb(^DDM6!NqgJcodp^PniXtb=a*`}1FY@U`+3#m;Zfqkg{OdpX
zm;CjA_iu9I*fHi8=2^XY>!Tdj4;?`@A`&4g^a-+*QJRvC64D}fem}WJ3ysN3bQlnY
zAz>U6VM&LUya!#a>1oYIQP3_m2t_1HgwRN3DMW=_HW1|KJR!?6l=M+XAcTi8V8u9S
zTe$QNTRO436oRbK6j&NFibK9+ed`*djaw9hWggrADBXNWt}~LZ<W}zrX-{+g`c?Lw
znB~&>t3;Iw^NS0N2AYkn>l{6Hh}GL&4j(#5k)=rEQ|1Pbn60%wUfD+G3SWCv@hP_l
zTwL8ILUUlY#*vjekvHP{r8l{B{te1>gO!<#mAQb0`B|DXH3HvrzBM>~Pg#~&N~|>u
zMm_R8C5l4QB1IZ!H(_vWUFds6Q9$TNgh5E;hj^Zcn~HySz0IWTbO$h#(h26LH}~-b
zZp743+p??onNFbmbe0=?J3c?sw&S`*CkC=wYl6@x%W}dn<Trlf*Ex9bATPc25=D{Y
z`vJPNWNASh$CQO;I2^LHJjcS~JXtmzUvE7Q9Xf*YT<LpjYm);94zjd#jrFxv$|7ZM
zX@T}mo4`{@Vex#45`sdzrk0{q^m{#q{SN+ei&AT(7n11$sR9>aG>(OH;Lhy>r4*j>
zDT|49@4al#&pgH?tPm3$6EiNuIvta=1;S?dqM$Bv>OxZy5D1O5a?Im7w$Fi7$>iCS
z09G++0NL}F<EOY!H32SrB!dP^VGO?T@RS1_cJkScV?*<-WU$jCsBhy}ssxo7KMt{8
z1?xH2cQNgRl~`+BuJpKBW$$8Q?>Rr&#~)bgwtG<M2k7GN)#$mEzGEcE<!<3#%Q@Zb
z0=8XG#XA{5A@1MI@Y9UFEaC3gU5^<{VT6=p8y$B@g|-L-w$$UuxUpQoxW*|fJhU`e
zKSb4Q#0zs&dtJ(GKsW7Uvy$Zli#-0sX^x&g41uKI?+}L`QUoXsWto$v&Qhw-Il>C!
zAfggQ2rJQLiYYC+bVk-vI$&qClUbNjV*^7JMR-Ee>2%2RoJt%cq@q@h$<mC$Xh>3&
z_(4b<_|)o6dbKJ$x7XO(T*qjE@+&kd5ubSc6t@NgcCK!)v%W#SHBW@c_rLW5ON&d~
zUb{|rr_0vXCjEAYrR9ArEbL>HrUbzpg)T|5gnF$;sSQyabMew;W|}QN{fSR;@%(wN
zU%kqx-y@2vjLMWE%h^gk2A{fe%Ka;Dt#`=tk|@xe*<V4MJ~yuA%&#mnx41;9U2ttV
zhVIr_ydcDCjkPfB_gEVzAsC}c(}FYe9=)z6%?gI2F8k)13`bBHORZ6GrrsmX+O&r$
zY2l0vYjc)rEkN1^r70+FLA~C<^!xbLI&19#w{LH=*)A9j48#Ekl7wuy$-adutD86J
z+}>n4OG&d4Ch3u78NO$^LBQ1;YYfs78ZC08>6;QQ7!%O<4j}~dbMySY|MLF@z<0j=
z9scMWf5iDq7hN0w2Mi$umAJ}mYmWJa1>z{i6CR_{h{15c#@0F_9V2&Zkk+T$9#M%y
zzWJ>`<X67>tIRL2aNxlHhYIRSYmmN2T&oaOVxl4^OHz_Fq1Ww^<+%f{Cd1RtABf1K
zR;%DgA^oIJZ?nr*Cg}kMpm0%XlOi0}QSinZbeWSR37!g&o{v%zYn+ZkN;%Fpbv|R(
zx`u(k6G*EuK+p(@!+_20l#QL6y!*;6_8ohgM-LulWMgh@_RyVcWPP7^)~+%bjqt;O
z*8D8m8rr=9%llTa#&Z0`aiUR$cV4{C)hk!XJ0&|;wz=@aHKY-QLg0ynYSplBUxR~-
zHFUDh^>g23bL~Cy<Tg=|Gj|~1;6j7BW=y?SB?tmHdDs$!ATJAo&<T5c{T^AK;0K=T
z$SORnEu7J-kj`c$s1SuUqQKd%c%FwuVU5L>4^KqBw^S;pjK*ZLaz9Szsrx+de%hm-
zc08zt%mSYG|7-8hf-K9j^Gxj9(;4pQL%4^>kQos<Rb|$M!cYJK5F<e&8XVfd7MpJ^
zGrh^IpW2N5DpR*ikNu!FlT2okx+V3CTGnb~TT>t*f+Sdg3PmDo%Bsw&%8ba&m^=m_
z?sTR->0zIH-6N~Akcmu~*nlq!xA5Wm^>fZXXYYTn^{@X=0_f`lYkPNxuYBca_@!U^
zC90}oXJ?1SVu7`mqoZT0s^;?Lt8|_1ojc&lmCGC+-a|-|t)WqjMx#U@naz3rxfj^q
zzs!7gOw$Imlo%t4c=Ezf6d5a-jLp4EnYvY1^XT5YJik9=G@c@rp{*;ta+q?6^`uP;
z0_3{GFkqeEbaRRi$=Q99$a>p5;Ur2(luT|nF*sysF|j6#D+aD+gk_>Eg^<YTK}dv<
zM4@oeA&^9$z?@bfL?GY^G1A4}O+bib8z>W)iG4XwN{NoC0^rW^I8*jg_xgwloC%>5
zs0f~RSuub71f6S)0ihy!wvW`Q16K&pQlo;R^By5mrz{d=a1bIP`czu94rMZoRLROJ
zMg$TmRI=6TOLd|L->f$^F@ZHiU$%^C(<b@`7qQ`Y_F2$=1Cma0-JN>|>Xhy?h-(?c
zxCWWwnb-Y!<5|FYo)`F|eAx4#PYS`hAP{>wvp~{T6f%{`i4h@E<4QCFIfxX&F}<>n
zb`7zbAr^4!<(pi&ewnJ8;k#7oud8+=3|^KQCd&vhpgMtfsYGs@)S#KEjI79Uwk5=*
znjre)jUi%v5*jXw#5<N!Vyz|kfYFBKVu^DWDJ5kwKxmUXaf4$t8FTdL0f&zsAfh5h
z&2YTU@X`T4|5Kk~Wi9VKUUL7P+g$z37xC8d-EV)F!9erq;W6WJNoE8O?!U+VyLWm1
z#x1sXwppwy27>`o36_hR!Eiz`81m*@KR}@Po4@%t$n%0n4<2xM_yDamZRn_0e*u{~
z2OuQNAX&~GPfu1{o)%nt?gnjLA*IH(T~bNT6jjqgj3}*9S|N)9DFn9b@bid~imq#^
zmMf}N%V$ftJ<mCsdvg7ho!xE9avMQQ+qQ%NnGx)56(rMNmr}r9g6YV#rkXqMA0F}K
zxMopDmMcfFf<Y#^ysNo!W1oTcJX$o|y>rC4l-#^_gU(tM5-$Rc6?9T^=roJF$5etu
z`ma`W#I;cLj}4<Wzx7*xli&L7-zLj4e*5qIZT{9j|6Bae{=q+BvHWoRCapAkm-e`N
z^%@744;TyvXlbabiU$uKY_fFTC#ek~Bz0|>%@-JXpFjEfAM>~W?%(Ie^Uw1qU;Ff$
zC90}I#|)!2S{n*el4TkBu%zsb&8w=SZJN}ji$E#C)^v<4c9>U!x9%RZ0s<vcWJKWz
zLLr3fK|#_J*IE*SNIR_P$#jk}39NV*h<IdbfjXyw21ZFV5!<HgQVa(aYRo9N?6rzJ
z?|qXWzWGfqJ@*nXzwj!>q-4GlynDZ5r<`&!U$C4#z=i-Z^4{Gigcz8=eV658$=k!%
znV-%uIs+$oZnTXEj_s`xSN66UW&z)w@Z{ZZ^XT3i%#Pndgd=t)frI@K2bTwoi-JsN
zC?yEN;haqbU<uZ><hsCnN7GacM<d4LF|*leTDYQyn8Y=Ua>$?@QI^BLYcQ3!d<X>Z
zvEF?u<9>cJUz4G=Qk#2_HDMpQ0d^n6S|10{eOjg02phl8^LkXbZav3;@E`mxLcrt4
zk5DRgt14xf&mN<sVQae0>HH~WneMUc*ROH=ogdKD6?I*KfSv7azW3n!oSvLe6a~*;
zf1W3gAEH(eqh~l6;=NCNM_m!4L-cbV0lK>7!8<?XXiJk_yUt{HkC6~KJkePY5-)p>
zok-cP7(*|f`l(!^9~<6#l!zE1fItWqqawy+l!E~SZ!t}atSXFc$b~|$8#nqLng}s@
zY4*5forF>w&AFa;w+3kfd~%aa5T)nWi5?`a$J6&&VNcv-Vq;tv2_qr~ob@z|C5xk{
z42C)7Xn>iHky(yFBa|Rn?2)7|^6V^C*z?3BYjBnf>q7_|=yaYp%BEDl7Lh(DPqV2Z
z_wU_n35ve_EH-DAzK)ysJNw=<CD{**xp?ONKb`o|rjHN&{5-Fav#}vYoV901D<sk=
zbb`QyAP`C+lt7Az91IxjOvt)RnEEP{c!}q4T}SCi-PB3THh46NT4S{CpPjV+OQ{eD
zmh%O+^^8Z`spB6(@D}g-ibp(S>xK|eLQ*eRWLd^wFkm@fuv{(~4F_~>i`I(aaEKF%
zy6tG%78x96Ibe5pm)Vo2j0OXQFh~jW(<fZtzrrv7+?SBwxx+VZKjHMrWA<*mz~TE3
z_}ovu%CG(EFY&dneT|c&1(U4;WaRMST^5UDUU=yhS{s<n7hJn~m8VBD=Ce6vIpiDP
z{5Ff(jDPqa{)c?;JKv$}T4t*m-Ks@>L~FQ@1?qfA_Fb#s$y&v|hY#2)4=_fPhXS;u
z&_hn29wC*a91h6`WokFqsT?aJC>JOPL*(Usx~5_{Dp@X;R3}R=4Fz>27+o8&I-OyL
zTa;Tvia4ODYr3XIx<o!!S|t-)Z@BmHgm>S6%;~(r+O(U?O=NE<xUx6o>g55&IK$Pc
z?b;ok5_I6TFMo#7l}kJpk%=C#3=Y>t+G5O;)ttquLMcO;X{ss^doEo>)5BaKm3@$b
zt<i|z{X73YMUicGgKfP;d>9JsABaJoGaXGh*gN3rm8(o96M_#M9UW2E6?MI2v6v%7
zN5td31?id1mgEM$`bYnYzwtN!7T2y`{Zum~owL+UpzS?+mDVY1qZH%mgu!UY*iG@y
z;k<_!5DKO{TbRKPPv;$@(IMirK}vy?nh?@1RVkf_)ZP(XN0z7EuV1ImL)>U1$&g6T
zJ{iJpjxWYY+t%bpGb#<<MXamHB24p0nH7v)nDTVqa`(|2eE0B8Cfk?TKlltUjIZPI
zT-wU9PH}X6LRJ*$@faZ$(MOt0F;IdV`@57yN$`%*uw+nblyaONeV6aQ_kB*DzKidk
zP?(B?E!f>2GaU(rL&31{<azRIjQBoFm&P#ofOD4llNoK>G8_(=OvY!KE}vW*B#KNI
z4DumGma@1ZhE%SJscX=M)G_#pqbEJXnLgEehYoBy<2GY(?z8pcxB2;>aX9C}-1*Yg
zPny0y{N7G9ZXaSYCnr?_UU=~afB*0QE~C+ix4!=t_wL@Kt{bjiy^55YyYD~d%GGNq
ztywG<j7C|KiZNrZUBAkmcituhN7uEC#zWe+Wj3E<t>fU}fbFdvd}xWGr7ZGZJ|QX2
z`F^g7F`~6b5owMdv3x({$?6oY-(<3X1(9niTO(AKM1DzpqU6RCH!JO@yq=$3(-4T#
z5=B7Bh`=)08ZaFW*dA;#2uu8IK=nw_o%H^h&^JP?%a;8-NXklW0L(gbwax~G51K0W
zqqi>oMLqy^dZBmvvi$l=^n5=pBq0X6x?%R@n7k+`#uLJDOeqa!qRE6LYDJU~6nJ@#
zqu3vpTm!nkq3=AWjm<rMzJR>W*!87yvDOvrfgG@I$Vgv{vz?Flut(!u1};8(70z9#
zbJyY1iFHZ$gJs>+J0tMDm%9*2Rt8eFDtL=mK#0lZMt~H_^~S}B5;1)pJW@Jx1B?y(
zH!rhSW{k%rb$yDF64yCq^HT<c5tH$la#$qJw(Zb5M@mUuH#AKHQ8LI|RG%>w5`>mG
z>-#DWJyj<G>b7fBHm)qOZO3x4Ab8JcIzdXss&1KAGpzF{1X*T~LNgdlsEdl});2{k
zz<YQ)o6#*#xH28_GcP}fx1O)P{T;LlTz>gw9zQwcmw)3ok%Nps`lEl1?P_u@Sk+59
zZ+YW8-{!T~zQ~nJd)&SIJ~As2CZ`P@K?LqRy32p_|M+kCTfhDrY>$7Mul<{^5u@YB
zKQ%uZ9}CpgF;aPtje&s?EZY@#AKmAbTQAdgZZq+WCu8(zfXYo;xMeEc$Vkd2*p5}*
z&{kcafl(C05lSeo1B+G1csgY=8PKj)lv`U!tr%=gm;`|fj^la5-TMpPd-#Y)kLNV4
zpj3i!E|~5}wkL+&$$-%$8E-r9cy#}e`E11t%Zl+<#`Di_G2YrCj;3TXIbrBnaNKCZ
z-ju^9r>tBg8b#MO1R>bjIiPY2mW$Oh&sqdBsknLL20!-;U)ekyK6t+U?QdiMjOK;m
zaLCql%5*wqYr4hO)>g7IX&SOD3CiYq0jm|PiRnJl)Pcpk<DGYIGn<`qX?O4A4C$X4
zDl-V#XM$`p?_Qr<r}PGNX<bnCI|(HNxsq7hkZHm8bU;3DsDn*lSO}z07^P4$-4oun
zV6wi$*V1(@SzaInWSJ(2^nK@TnppadwirSp$A(A{{UjV5nU)L(L+Z|xcaF&@V{dzh
zW#f7O-W_g#`(2c}%yjF3>CR39)t3|zOWS8~kOCAaStqz)CDyHY`0yUH<Ht1393?u8
zvP_I)G=;74kgaLXbZjVcL8b&!raiiI$-h;pBpo26qU{`A*Rfd4F~+bp-9jkIVzEfv
zkg(=eX^NtxEDKDgQ`tz!o@*ESg(#7WKdpF%XVZW089Wuw7TL2b#<?HI_>mW%4|0)y
z%u@D|LiM_aNm$r(aOE<;`m4XfjT<+3^R4f3e3W#e_V+JyaB!8^U;ie<;gG$p>nxWG
znx@4%hcO!OWTG?NeIFs9X=(=LxL;gq7K<g<u3cw$XP0U<qv;$bGwAib8B^vrhS<wQ
zcw_)ouP}>a+Gfd<<qEf0F}QR<KDmU+WlvE^*}LF9sVKLGNq;f0PDx1!7Dq=}Xo{iY
z`I~!O+RoV>YjU@wdQ$LIS?0dM&t}LFL6I^jE+mfV8n~@d9P~DdDN<>SO-79MM}LnA
zIP;n9)4!3_q>+3^`|{xCeIi;aq*k=vVOK57qf=(1BVwMqVau!louw<NKtzcn5Jl)g
zf!MfDtod{sX%m@B%Yt<WdMdXHarXaDlr0Ipyi4e9^ZH}{@$qbS?!%s)PwOr?|IDQg
z_Zm=&^DiQ(lsyjp*lH>IagIb0@FcFY6spgpdIBC)NdDBpk%x{FOO}J4W{f5~lmtYu
z7%j0DR1j>9w?Hdm-vkjtAfU)hzt(g<d4qdTHJ{<siqT|(DKi`fsUbLz@PW{C9<58A
zu(p#JB`Nca;0oHdO^tcRFkj8^J}?~&>Dmt8wiHFdc)Ue$u&QdDuTfH?5p;_=I`@2b
ze+$`N=Je#4M|a<3u$3_y-eTn(zxDTim)j2>^6<SM;JcP`V91IAUE_HB&9@kpV@xqX
zYl&5X%B?^N6!7l7+x-6T*Zk@){1U(U>%Yb~zw-ufy!qxw8rP2n>e?8o`x{#siN!Im
z7aX6Sk_kno48?SuQm$H}ReF2ZbgUK&nyMxe@gZQ)_&TthEs5TOut+7y2OzY-lm$W=
z!fJ(A6<$J6iYLp4@4S7?z4wnXUBfU_983kHab#zxn2yGjgAv+f2&M4OvskS-K6!*_
z0+%muF>p2At(O@+{{jVqc|OE7EiNx<F}z&|c1K%0y?u`ci3@^I)yUlN>%aP2eD|$C
z<@@ivMc18i<RWRu_{A@Lf&GKan@OpzYu@<o8-0f9$A^-NEX&C9oU$w#3<i{CNmZrV
z`<f3B`gZ&E?jwXmR9`I}j~_qg<m8y?bn8>i)M=CKxz3Z_*OQ*}?K9=M^`uZ_lCEu-
zRSTjE>}-|n9OqOEM;kp7i542IBub=nN|`x*wo~?=7(Gg*3}KKll`Mnrfn-`ZJz!Kx
z*ESF#&kQOAy3Qg)MV@Iwv`8IM!m~9t6lz3>Im@Nx<n%2LZ`UkV6~%CX4+85wF$$E>
zD5;U8>|xt<<Ygd)Ie8|TZX3!>Q52HNsAN!R$~<E{8emjvgb-1ngd(Vbk4@rw2}LB*
zb}ftLl2u)^wUvOp)|$>)j4@bikrIqj<ax$$IAAy&V2nY^wJA2l=y1VdT>_+^IQlH%
zd26Hu8qWyiqUX@W5PvMN`zZI{M%Z-ywRm<h`SUDX>p$zfV=@`><u8Ak%a<?n&2N2!
z)6<hgCr}w>IpB>q-r)5=eVw2A>7P$7A@fB7-BQxF4SA**jfRZIBig3LTE}E>oW`c>
zsOyT+aLnb)m(%=Gr_TE<GaFC<iBG3`E)U*g+Lnp4$QW22KHzEHp^s)vuhb0pub>BG
zOi|!~&by7}RmWdk1vWjq4wFT;E{)i~GU3)sSGcrWGA$+43zkK|wJTy%({>%9X*N7K
zDI$W@p{@G+My#__eVl`1XWvWYv%a8tF2e<za%?1R*dlV{IZXdvp>LQ_UZN5B&eNQ%
zIMI($gArL#;szs}HtD|50^4_LC+?W)vroPBN`E}-@l$;nd|mE6!(GqbpFge}HeJ}>
zF+(IN=ETo?cHKCCb<T~*8tZO;F3y3;PaE8C_;T_rWA_Y?Rh+#peUkds;~Ispfq4I^
z2nn!B?@_)(xfbm#CVDghqap+cc$AhzG(PsyHwfCcMOe}6apq)sj!L^@Y-8z~hRz0L
zmQhzLv<8u>WGSZwh~yf9Pn(UDOtM(5$xY%RpPru5I-Tt1v|&0bAx7G&A~zXr(=Zv0
zQ9|?BKB8$mid-@%OQ=`W^BI{`eD=93_$c_p@BNT>zW*kJreeIi&9&#Q@|(Z@Yy9K?
z=OYf^zt40#1C`t`mPa#=9v!j0v&-&PiI4@hsmVqIoU^p6hHrf9Tbw>V<~RSwZ}F91
z_yvZO3IFGR{NMkXWBIW&b)HBjMN$s;fvRnIbo_`bdsi_5f=y*5zjVxxmYgnYOs3er
zw9RN|3l$?Lhflcs-usNk+w5N5M})+O3tAwv3{VIUb-SYJR;*4e%X-P1@2`04Q2-aX
zu`9Xx;+UQ32(3McmdSXJ%nXrA)3i0SlgFEu;L&JIxuqG5CYafb{lE5WT>ph%;_3VE
z@XfFO3vvy`WJ)U(cNGi;cpqpZXxFif4p9gGhu{0}Xl+gSU`9j)q%i!{PyIAHGwB|W
zkw=dX`Tjd^;XcGtObCI=G+AcQnNByPgwAyvp~N*uz{g0QCDrVvZRisLgb<l6<~)9M
zNUju_)=7Hg;|%VHL=dS2D}=!Ji<DMLm^4B^0s7P#853tD)`2#9n$FSHo>8v2etC<9
zKjmrFqC`P14Jt&UON*BC9v4DNk6TOIHnd$vT`$S=ZOSr7DTS&PZQJ3!?@P52sWnk$
zI3T)YMTVA0E%Cu&OhhV=?!UgWH$+N9(|PtUd8#_HtkMG7xrp^1+gb!V0nW}7WMq^}
z1_Q_oNl_?<L&I=5Aj@>pQ_?v`8-k12&Lg5hiwy52onO+r7DU8IL)Uejp3bP7n*D<V
z@;u+P<F5g-bE)gg<c4xsVseAjsSM(yCyE3VtV^oyUE48RR-b&(d#Zp~gSwD9j_vs_
zR&mzdngCn^bU!X*_i_GaN@mDEcX=tqK%Ql^U6cNA8s2&59kNVw=gu9nEN6f3GT-^m
zcliEWKjfvCUS>R+aPo9cT~~=aXcWuEg7KAIVvLN21L~>*BIEHGrBu>0^MSG~*xA{k
zEC+z4?HY7u=&Yq{J6s4@=TLc$^GQ@Ygn$TvQflyyxSBIxESWuh!oka*VYGjh!PXAO
z<o#wa0tSQzR02}{^$8;KY1kDC0$C>5-YL0pbB`M@US|LLHrvyJTm|exAY(_{RLHJl
z(YCCfK27&Iuy#EYeVXk&Fhity<V=e%0k}1xe*GG;Nxz*hzn;@d?3>ei(gF1E6Z!@r
zEhQl+LhV>SImHYfqm7}=^S-<qaHS#01jvF&dU{$U4w#6N7z6^9nmJ;ErAo@w2ta>t
zuA2zf0PBOB5pi~I5eXt?{MO@?oIBQ;xO2b1;rN}6%BET?(%FJQ#1GfO`}4<nKqnF6
zneL39FDd$S?#D7lVw7wgo|8ULx<4Zn(UW<Lat*Sp!8OFTCUh-MMZ9lOeWgViLlg=X
zeDCF&Hj%-3y!MF>A_PVohJ!(J9%{N?ZU^#Afy_yxw;mTsRB8j%z4NIH-ehE1qDR%M
zin1)3PA248mYS#94iyDOp3`(KLc;dWHjDX!YPCYd$hCt5o;-fS>hzR}N*Z=YRm;6^
zewX)tc!y$Ga^w0HuI^sqxhvOseD?uWWtr78N}XerCd7vOhmTlPORm0flgY5eddFh1
zB(#=kG2-6gBmVI}`6v9$SALGa{wu%okw)@kfx50IF%E<vlPN1Aq~o+&F{x+Rddc1Q
z9&qPL&FQ@2xt#%@|Kf{m?M(2&v3U5Hr-zSFQu5MEFEHNQg~H&i!`nX9+IPohTG6fq
zRkh;nw;%H6k>F@4nP#3_Hv`XI8?rOK!gx4jxtd|J2_Yom*3l@Z>l&)njNxEPQH&|`
z93vyD7~&T*hJ`|B1<{PS_s-jFj|y(Q^m&BIU@)Lsu8=kX4cTin#Sp0K`G;PgfZ*!2
zYj~fOtEGs%_15>efB*0&&bw)soWPNDj=HHS2PH8kQ(|MxW{Gpw;hf*hg(>qHSym0-
z|KZyxk@ND4FY*1ie@OQ!WNo^xqia(K-<o4$jKO>VBfGkFIYl^2^p4hfZ0j*(Y!5UC
z;}X}cXq`iOfx#e9cxP!{pl|`DQ<f-k&OB|`&~_~*&&e{4^BG;&_17gM%L<y-;zP1)
zic)||oXQ~2cA&`6Yf}ca3`iLmm5^sJ9%gJgi4C3*1=dHp&ZCr0WnG&Vl}u~2iAZ5l
zI$|<Ko}1pBJrcc5O+sjl8DMll^pd9Su)(5?L}|rxx#aQV#|SC8{@iuS;qZ)yCQ|lQ
z2#MAPr87*XFnN}oTZF_%@IK;PM^(4fbwg(zUE6ZJtUvLXYb_E;bS>6J@=S8+@;>+k
z@Dku|alzv{OJhUYbc>Y5j~h8Mfl~ThhHw4R$J!%moskgz{p>^9adRdbyVm;(A*8bS
zn!CpbOT3Arm>Gk0mPd~tQnyR$x?;6jv9q(o<Ht{U`0$XcSFiHQE1#unJhRy;Wto%Z
zk>RjpI2tgY&l9<NIG|jX#ON81$5>~PQqy)VQ3wvMULpT}Nx)*gCo6{dAgEf4^?~tZ
zi^XD=#MO-Ei3>aqimO2|@Bv-VF?g0?#^X1Sc=Y}jyO*D5@3~uyws)zOWZ89ea!6)M
zOo-r8*($2Y+<o%rJVr~N8+LYwT)n=-^&9(4b_<3hgA5iCb3!*_eCZOgs)_Sc>eUKo
zS1Bta040&(Om&amSz<jdtjCidh5a8ay2rCl2-fy!=Vh4Jp|!y~i}NAzPK8Q#RZYk2
z@G+*y8I4BhGAGLwvXB^Qa3WDyV)VdTB1AlMERi&Sq$-}+%(<+;pFa5!P`Ca)Hm9&I
zyGqW)yVska4_=M)pMUrU>C*>wXC3Q8eBk}V8q)TMw2`cd1Oy`VfDQpcBD_F(hiNUD
z3>a_8unaX6GD9efUCzOEL?x4lwK2$$*1PqF#X3jRG`MBL2gVqT(v*2lC`y*C#k;go
z>YA>n0co`Cn^>d-t?_FPXNWkT#FCX%Oea(B-FuHB%aT$@X9?C(<^@`7=JPqz$s~CV
z$H4aP4xO`n_gmj)ess(WH*PW=YG(Bc5WIA-%gJYMvii<#p1kue^9T3XzxEv053cep
zRWMuDM3WJ;g{CETmSLvwO~w7&@9^0#eUZW#jvqZ?GMrM?4O(bs%N2j{ul^-Jy#3Bc
z8p)3Z>OzbZQZW<)EmF(*Br7pG<7i&<_Th?udwWhE;CH_A0$+IP5+=)OR!bfoJ|tI$
z=WpJ`6gfIGh}>Ykr>QD@+Yw{ZnJES(Q6v-iI2h56YmVQ4%=4EtH!lwvZ<UP3Bg%3_
zV=aU6c3+O_(nF;sCeH~$GM(&D7Nhid$beM}Th(mmn&2ZLz`Jk1$>Sp9%fI&T;e<ry
z1*gYzj7??Ej);kZ`X?c?2(@kb&;B?6Isf{PzRGK#dzBkEpW~nYzyArRCyNiKQ4lxX
zZL8Ia#bUwM))sYDBat*s!)mpn>$(ld3%#)7y2NFDq;4Jmm;d_zM!o8|eC2@gc+Ao9
z@y8k7AAfWqVC$yO8YM1Uu&(dGeYW*|1Im3s1aQ&OIge0|u_?GV&9RN;>8wFpFyzRn
zXuFoyf^(@mZCzsSx`w8a)O8MW-HDec)8i(UeghGuGlEO9Cq9tuUWCAlBr)J)nv^9;
zq9w$**-0y*5Sd10S;_(kLGWn;S4ty<B)F6k)k-Bi5pBO1M3ju_H3WnPohgVCYb#dE
zIS7R@8M%?PU1UC=6GC8ndz-QxFr92+z1u7VAw-;w$udmnL?y^HSw29>oX!in&eC>{
z)vBSY8kUPCb<@&zmPPZ)7kH7#uP7xLO@<7I14hFUMKMa73UARmfzGyVI5|1x;ln35
z8}a@uYqu^T{a8S^zQ4{TD`YQMBW2oEue(@fpFLR1d;BOM5YK!rgur`CRXH9$JY+tf
zg9u!`dJXRa_Yd#0yK{-3{^_41%S!IwdzU=VQ{xIL)90K+3X2e)vMiI%l{Sewq7#r`
zgH@Adj7B4zZ;^FFkq>Y_(%RJQW5e2|Kwyc?V&t&1l+Lr`J4zK9nSe5&MZ;|Vn8Vqe
z*~5p753aF&<2f!}f1WxHsOnyhK?>Tg#)*i`KDpm-UMQ|!-{+;5ukhTB11?|LX1JAM
zeS>X1K})<*=;4r^g98T6;;f_N0j{z{m+YRTl*pKz1ys+~f+Tk$eOmlwPs*2(RIXL&
zdr7iI5@eEBkv$tP5(zQUGeU2?E=anGsS{hGd+8Aw1x6@jh;;KM^TWpsh9xRDSUg4q
zirpz8NOYE^dBP^Q6eTrDJ2RDLV@%Q=ve(?(Y#-8|@kh#(KT?7fX`T_i*;$kk5z_pU
z=6&D4cCF|4>|MS2Oq|=?h}4<A7EJ!M!uQ?2q0gK>b3VkriG<#EZ7ncP+Bipn4E_1a
zRQA+9P>7Bki*%{{Dnzel=(@g<C<%Z|tq?{d`wVonE{X3)XhpK`gtQi+140|}EXU?K
zwr&U_5IT#{Np?o9m8o^&igqDx)}`P*ZQD}hIg{}iEfw{uqAFHo#!%)3MN#0LW6{sm
zMUlQ2V?-KFUKYId<{{JJfW7@I7y-ds4yHqX?z1oOXtv~y!(+OW$K=<pvpwEoJl^7@
zwzRFMUbKvjW;z-oeM|6;uBy0y=QgA59iBV5jCF#gt;vUDI@^*N&HMN6eWbDcc%Tl!
zKuJoOI`X^^3B|j>x8Ixb)%PoQwC2}e-QvaF93eA6;#}nV^&99xj*sc4(K=_fsF)u=
z!Y&u&x?r$1p&X8JF>-QpN>$f{5V&<^!t!LnVmRRL?g8V=&oMTdL3_*<Q!pJA2%V#F
zbX`N&S%lCOdO$uLB@a=ln6H-jSRw{FYPiMfWCpHdRady+Qi)U>q}H^H6}nF+dm@VX
zll&vn^@)-1eD@9R+_^(hWJ!l+xx(6?TpqF3(KI#7<$~pM!D6vQN`;J2*Hurj4?U+W
z^gGP69c_%z+K7;v%a^y=-rD*JWZaJzN*de;IeIYxFYg>aqz=Dzw^)oIqe2OV4;|Kb
zMD56QV9Vqj3=OezG&Z7K00k}zTu3F44PVbCoubsS)Sxnz%F%g_oD9+0P}dEss-_%F
zi3q%RY44bZE$!^j`PN2p-aK0-@X48&u3wjfSo4H>{}YiGpWu@Rs}zbPX^@geX&30z
z+#Dj#cX)4!(nIX%T#b(zZJU7k*48%D$%GJ78Qpr{XEX%X`lS75(^yMY5K*+9q_vLD
zI#$bus%crwS2V4sZaTWo;#{Em6m;s^ro$M`crs)#95NUT*xuenNki9J5RuVnz+f<-
zX&Q#ZAt0EY&Z(*f=VEH`z$bm>9|PtRh?7XU&Jd(5DuDM1P)n7-wUE7yPwaW5Jupw%
zx@XTfB5^)x)CCK(*_;qkhD{2Mi<aG;OT6~l7nzK=xqJ5kA;4%nqN<iCEh)+}-G9;%
zg5B_4^E{EOQ|U&gj?oy&vYf50ZK~A*)pyu+U59H^Q;)Uj`4u8!T?a+RQbiUbG7yjx
zh%sQDO{GOAQ5LK_X4Y8h#f;;_$L!zw9HZeD!AP25>68Ft2+c>c5Bf<l92@rcx43$3
zpZ)zECgUN(1YFOIi!wQ8l;hMbNVR0=xvMOinq&I};RPaqwH-=pgou49c3pB@w|ks-
zEl4-QW@<?{0lob3*=za+i8Fh-9_aP2St~uPyH4rLf6_bba>4595&5WOlq=LABa|63
z&yY5$Cv5sSWXdc*!<XwzqU%kUkZ0O-LZ0s!-5dm)6r(shevpXh(|gO5@fJuB>g;!)
ziF1O(oD&=tB7vt*Ia9Z;5+%Ow*(O1X)U7Q0;`jQmCM&o^TT}gY5+Nc&Km~{Po-D4&
zqXpX{Tni%fn=(gX3|=Lm>_WiVju3(1(IfyTgkVsVge>3KxjAd8JBtyCYi*l0t&dFt
z+Fof)&$pJOIbUl{*K5S8UY5yP%b*;vcWIBq2M?H^oU*^Smzok|q-h$o);RA`I&5fA
z<#53M!9J_g+Z@kljHcTZ<$y)q5JcqiWWvvW<_fJ}(9NDgTQSHpF7F+1|H)|*(9Tqv
z+xmM$L`fkdkMBQWHLozz5LM3J-X7Utn4Y5;I6gZ1NMrf&ykCW8C<8=+cR9osDK!sH
zTE6wjVT0tAiRAjOA}<HDO-;x><LxQ3mme`oLyT0_ifVq0@qyvilyb6zE($vDI6XZ<
zOUd?RiuD%TwLJgwb-KUy+x*6V^dB?a*&+Il<>C9h_tih({x`qMUb8}4OHhK~1Ij3h
zB1dH!nPt3l=Uv`DdBVY$eu0No;AS>rYiA1~9C<OJvMrg?49gKko^gEmh=PbCux?oo
z9|0%15I9{f;KL>9o{1P!XX$FSWU)*t)WhMJ7(kG+cb#qNnwqX{aKUcc`_H`#gK~%v
ziog9kzr)ol2Rwh{IsQ-o)qnMoj`=5rk`kjd!m#0cDVYRHh0?@e@i8Xl?mq3}qriHg
zZ!H(mqKFKIU~iD)2Z7_cMYw=h;ElpYh4pbGXVTd&$tws!+cuP=0U`Rn^qdH^Np!cP
zS}ZUo!>F@TNbmvclMY@8fgofO8%2>cj7{p$I!}uc8?&aqbdr{j)a})KMUntWYJy87
z>x5qd&L&WajA*TpA>h2jdk0E08jkvsafWkAP&R}FVp6s?&{;>@bvSEr0bCIH*b#kA
z=K^)pv0AlMbw}GeLXh|%h%qMwdYiBhrwDwM*xkLvU^HZFYpZuq$k^GvL{oQtIW=XV
z>bhpNTGF;HWm&N7br{q2S#yxiAl-R>&>F<8^|+->jw+jo9v^zu{TK-%mAFHQD5;Pl
z@WI}rH1r81Cb8s{`Sd<wOwRGqjMLK@FTU^!`+HYtTTcvv$#^T#1zNGm%B8G$pljRB
z;%c<n2!NiKe94SvJQ{5lcO@0osv^dSmYU9XJx@N;I!jZ+;cCSs&zWYLp%N(R6OYL|
zOpugPV?@L^D^~aCoF3k%K3#J4)z`=`U7`yfUlyp`&@4XLcmwS0jJSNT!`|L5)A1-#
zEQBS-B$X4RK<bQqppi=9cemM!fs^Hu`ErHxtHc@ef^~^e#B<rCKE)+qV>PuYVbaq!
zID7sQ=Z+u6J24VOU-qNVZpZU4>I?t9omjVAar)>HGS|pLV@iY144M&2X;1<m(q=#?
zjYQy+SYBW56(qNZbkoLj**US%s{6n}3V|R2{x$zj^uTUSIy)0JK3{(Q;K{BVPSghI
zN+ElZ-;X`SKQ`9q62e(?NC<v&&r6ZO)JE=QomyW5Q?bd(qhiFwfC&!m0=aA`=n#0c
zN&?qLChlTzju1R)nef;y;6ub9QCe@-IFqHO1|dXJN${SwZPHlD)QoAZO<d}x!wN~R
zG-!qEMW&^cNTmqMgHPkKUT1c;wm5$Jl=*zlrQO7#O=<)SLP*-6ZQD{+6<b?dOs7*$
z1|`!=BckqQa+Kuw^psh>;^u3g<)!BiXoP1*i(Q>CFd19Z5oL{?RahHg*RD%}BE>aG
zf#4K(2<|QoQXGm?+}%>#o#GC~i%Zd>!Ci|>(Be*E=i3MS+yB|0YmPEA@0F9hGi$Bq
zdG1e6-k$M?*27ON8<twNOCJejYRW4!>pr1cwf*aC!P!f2X-A|t1CemDM5hElqh8dB
zuZbgZF&zpo<=bQz9#MNzxQ*Ej?kPLjk8T2{T8Hyl&h6zoD~6-UtGV<XohJ%-lv{=v
z`DICwWisGH`%yP8;OLd~+cOXZx-=R#rX<xSBAbFtO<Kwiw*x{Eu=Ac+T3(-CMOxg7
zIY>g;by(Aexez~JtwWPuFUYW$Ubjd6?#LX^4|<up&#_<OzUy5$uwKsK*9G;4-%ggb
z+0sfT*ZRu#%c&3}jRXMVN_i&fKNbO;@lfFO=@QsGoZjhwKXVvn$d<0Dm0FXfT>gwR
zu{s<C<zerBrv4_;dn(l8eOQtXw~+7SKlTg)I6$WXe1U3;g{mMEJ-R%-N_U?F$$n6*
zYyOn{7TCF0sQ4?%aZK~r;a4Y7oK9L&m1FA43aR&K({y!OSps_QhMWhoHtB?urT%7k
zHqClfjOHK7JZhAvq8fiGxZF5vaDB12mC^GvH9k7E=f?*q@M=^0Bg%`RG5mp{b{!MW
z)BU!UOhHPoiRKgpOK7p{)@XlMuJg4?pY}q59lEEjAn{4|c<0$oRF&A<yYL*S)C&>+
z(6{#Q_oO(M{JQ%yaY7P<v{=&QjZDn3#YHmfrSH?{B%W}EOY#LF3NcYr^^`Xv`ISD<
zoTwTl!_M_5l~SI_&Rfz)SSyGLp3gRVBU~aHP4|=%eOKT(f#W|IFM8GR?wcrF-M>Rf
z9W2R&3`FfXgWeQOnzqC>JF~Qh>9-i|*1p~tmobKQysnE}`k``i35iH50^-BXGZ1{N
zLIOTvi_#Ztd%vw4+yx}tOJ~F`2xRRNML_1cF6=@Q6tXm!vX7nCrw50>&EhwQ)7+Ms
z(H~(jnoHtT$$9dIB9w4*3AeS3ZTLr{)O~@argNa6coDek19utNyN!Gu;+a3+Ev-;!
z5FyJ}rd(jy;HqxYS%|Ag+1V-{3(=?FWlRud^AK6F%B1bmKhCLuw~33bIQqLryy%l>
zci{EDQPLm!1(J#z)lKqevz%A_LeByntD(oLl}D8(O#v&L{r72V@xyY+15|QZxqT&=
z)7K^Q6vG9CoUQB2Ri55BofLUG8fX)I@mV`hCWooC{EaqoOHT7Ti!8Gjjwf>y;R3G<
zR`VfNq$Hx=rGJGP9)+hA0)2CsPK?6^Dz#A)>%D<{iiG)S2RIzSBwqy}dCAXRoyH^p
za2lL1?Q>RPe?Ea*vNEg{`hC|Kl;zS6EPa>UG5iT^PivW?eb~h5z?OrDx1mc4gT??U
zQ_$q^78RIeG|`X!;H_OZ`;WGN|NiC5o6H!8o!B|<bPfTebkhK4PHfY$6&2K@Tv%9G
z9I`V_xfU>M%b||f^8}Q5;_uKPrNvs8E0grT5%3J>%9@#;Wr|+CP|oG=>a0KQ9;s*k
z$7&ZUp`Rd2zB`HqoF$frB5+m>ZqpZ!LPi&<`-rY}O6)nN1WuDgH;K9*h`lXysisHV
zSYc@w>KFU^*xsk_A@xfP{3O#LZTt5V)RHR$yp_nkdd;`&-0dCl_Ck^v=6zfbxsrj1
z#WJk%m^WFqSDj*y{$c60E-pSJ<HSZ^Kz}S<jx9U_cg|ez?q1G>!(~A61W9%}Gp#*h
zfjgpS1RA$Sk1XkG6x@M###hEKL$69~-OqRfF9__fW0O}mjvcBRi!riNx3}yU6F+IC
z5`T&NXZPcZjgg+rqMivy$fwy<*(ZEli*PsG>0lV;Zk*<6KK9hIlEc%r0&Sjp1`L1I
zy@AA%iUNPo&|3tQ%uFyaO_l|t0fYUD2xTkhG{3L{73-DhK=zGLvZ-KHB370XJ`6YW
zWh7IApn^k|(WxXRh0>&ahiin@raAY4?P~(uFFY#m%kS`L55|6W2cpp-kIqBBHuw9=
zEsy?oTX1+I{!t6XRFlyrk)F2bCZIlwEO9NR-g5#UqS9`4j2yBXuw}SyrkF}26Q+p{
zB+%00G|nzeG52(gKg~f?WUjhvT$uJ?RaxCO%x?%TJv}hUw_$OgpK`-GdutD~)(ej&
z$%)s=s!7GcW8zNH^cNH)sjFR0oM*^8W*R||N#rJvz36rI^NVR>t!eP@gAAdpvV)<v
zpA>^RNvpLS_gok5>-I-UlH1#6kHJHpN3<$bqDYJQ86q_*m=X!so#yO*nM7HQ>bLKx
zV4@4zUDwJ03wSNGKayH`a%Q|`6}S%rV9}!%2!KEZ!{G;?22DlHM8&>vefZhQEmKZw
zV%PYkSPqITK~9ru?!i@7V_RQk!U4F>suHC+AeS+s3N@|M!_6LQo3c7Vs`eEBnvE0N
zhkQ^~hDA{;tm<W+R)CoD(Toa_+CzGAJHn3Gj40gmJ8|OZPExSq+VVqLPi?!)9-go7
zDm+O2l0}Y5VMnNVcxndUf+`}e9~1xJ0%vVr8SD1&DAS|PM7Ip<3vU0r5IR3XU4Be?
zY3!+|F?`#RzIVOZsXRMv8-b<Iq)!>G?47sWAM>DYB~cKpa=aNfqA^M_dS<TTBy)Nx
z)KGF%$c&dw1eZe`NR}@Pa_uoGs^gsFO<?<mC=h9>I#rbKjZ@B#<2{wf3||%0hn|Qk
z<7DwQZ}D{LAhET}WABysB7fF-=tFM+Iiuhxq(&LkVVr466I6521lwM5S<}xn#Zos~
zgiWHv(iK(aB#goDaTA|+IiRKtk@XA^-rr1I$@_0Kc6W7Uf5v%e?MTXB$9xCn8@g#;
zS*+OYk$g!uG^Yto?Vpu86U7CnFvPT!l8gFC>!!&wWjNT|k93BJUkA_o1+y!SxC}QP
zPc|4d3Hu1|tnZku2exh=)|owZK4Fy6!2e&<I-^BGV&-gr8w5JS?U>^z)?wT&4s{LP
zlefmNT1KzQwbQKWF+;0!ql61|sbP|6q#Gr5sJZPGvzD7Q{X=ivy(!A8^J5`wzZH;c
z&_dLm1o6pZXe+VWRn>XRs-F4ubrtCIbH@)?qL$p3EMyX~mwXS2me)OwW>O-Zsv7BQ
z2rvCNy^JnmSI@2{W;OJvPk(u~J<fKIG3UMN=C1n=69+!R;gFtfmm|dYTW4oZ<$~r(
z*zmEnA7Q?Y*PYiylkq)z_k&PidTWI@Y{M^ouXE^6;GO95IPBpM`zy%U>qhHsTou6O
zXZMhot^Adn#H`@Tuw~tSbi{Dx=QJ<FPYH^#oKpg$rcK+>3#O}qk88!0Kgvr9yjQAU
zuf*;V8=((lR-&Un0d)rcT#@gI5oIi?=$#dS$k(Miy`keF#fkkt9;=k7hUE`f_01iT
z2|DEZGiQ1hIBX(yiR(l!*gKOn-D4>}hV8xC&~j7xM#wA5oPelyLavR#B}E!LWQOcT
zGRz>w7Ox1GZ8qsSw;O=d=p-AQo%6e!m8?Du-U_)b^M9tLv5A!R3bxnwBIA_JZyXd<
zT0~A!H}?*PVs@!5@*B#`^Dxr1`T3_`n8zH^Y1$oF|BBnIuxzU+ue=u3+9}tLYsS=J
zwqB86l&!)SutMb=ijMiAX<UPz?Y%BbKoKm12iBJM1HFHSLYaQS9Ry8*sSIkl>Yvl9
ztI_N=?)Kt|U`Y)MBOD7nD)pBNn<sv0z%3P~NuIKidifO%-Lww>9S~xNL$gAfBbsKo
z%5q?5!nh^WzvviLbt*)I-^t$@A0#~4>azFoKLo}5ne?u)8{5gUy5z9FJ~`{uqHyGf
z<+n_<(R&ajUwryx&IMj1;wexWOPD;89$n}|dyP$ISk4C(U7{+Pu+DfS%Gttvw_016
zGhe}}G~!Qr6A<Z+Unnk>6r2czTR^=IPhXR8!M~WmciNRrBBYVj_=0^l6EkGDHZt`<
zgiTDDczl6xYUMa$4k{!k@YkL$Q}*TQUy@WFN0T4dGgycTkIAN4;Pns<uP6`_-S^|Y
z#)!yCC9yD8Pm$dA^&+Cl!Yk5Ax0)n8n-xi-4O~EL<NaLixvhA~2BU&ABU!cTw{5@f
ztHn=eC_s+ri$YRJ#uW<{9PqE#R?Fu`xOI-J5KJ-c9>bzubc%3buaEG!Vd;iCPi@es
zW`p3Q>*pHwvs~V^Dl+<snXEjP3I5fb$%V+KbVma*eXFz>t9%DT(%XKcin2LKFwSn5
z%yL8&+qlN|AB&UTA_4^vV`2uPAXOuEGP<(k^unl121^E@ScyTsVB^Rpa*mXeW!+md
z599Aq*a+i0e+^6#3;EiTLg~x9Sm5@zvSIXQPfxG&O@y)r4G&Kb=vDSkQBgm4N)+{^
z?%00oMCsipyb{+;HV%EgOtu`WCOd%ur(rK0m$or)Kd(HYvyR@|-5}$a8RGlC#qLq;
zt~<k;HLp14&o$>)@IZx*i%0$l`|w0T_<MNP<WIcPowsF3Ij85q{BgN&wr~}ziqSue
zvW=H7+vwBIeqW&|kb($|%d{Mcd|P2zOP+a1-_Tt~7sr-)8Y%=aXI8d%VcfseH$6e$
zY-JNig8Kb8Q&j3rYcH3X^H=YGe1Tq12x{pc*fhRdFUOY&*S<IM-qXU`2n@el@9eV@
zYmw0=AmA76+iF=i%S>OYfdB;c8p&C|IvBS9zWqzMt7_tOLv+(g<6x(9(6J{J_+0xi
zGZ{cagWv1PMPArwsm0s4fQQ@aT5%AFiyJ&(I<R&}=>(g2YP3&R6iirY_YJx(0MdTz
zO&D;AbalPGWxI-hnGPHYyv%!H$UA7gETrP@wAYlcXXf<M+!#mkZ>gX~vsdI0^*+}H
zp0hbG(L#O@E=V|~2@_**^^TxRCJ5xs{4IG4!~okUi&F8ER`;}|l^T7C!qZ~(Q@*6J
z4sI&%uS}7&2!j(1@<u!~D>dn_MZAxdd$_?(Dj+Bo?sUGX_;#{UYMEy(YNXu0i0C^v
zhhcY#+2*k{rNW3^?+GW6<<ljFOoj%AqB@lS&P`WCi%@H=PLFpC##~WR)R0&c-Kf$p
zltU`h@f$@Y2Bfm7c=xCbs}7^)@*$vzlzShqjlfEiF`dA4Z{0OXN&5N_c-y^a&bM>3
z*xl<_XUaitrmWEU6$f09qA)IEKW<u=U!g5td`v!ZBSPasQ;HqbP}icUAzwaxC2Y~2
zax_|AfD(ehV-9lXLr?56`9zRnarwa^Gi)fH^6Gr6uZ#PS<C_mb3rX-8gax<gLPj)%
zqxQXfuR$*!8O2an-ie2lfB3n~DCuoJH|Ws=*)9p3fRvEU;b=&W{qp*{EYpCW=#cNE
zHIJz<X_=kD8^;erLkxb0(UYH8wHssU-YJw~j?TyI&UZ-TpsMtFaPB5en|BqdlNKTM
z79$Vc>E)$OIsRptkg`lyP95aqy}h+3QW`pFOJ62a6<ZEM6iOeAX+cRTddt+~fn~aZ
zF_RvKN&yu9B)&rqVo#t6UtTvoe&jvcgOpVw19?Ok1eiv=6S-_*k+=Q<(O3VFtfq^~
zyTJI!UPyXXH-AOn*esBUT-ZfwdnM@6d19?jkAK$Oi*XT9M#u0DV^b{DlrECE#w1BN
zAkM;0sH?rGwVw)+Cp^^Dl-I3vw@K^LcXI_DkFB*O6I#uELyh>}-!aezKFOGJdZoy!
zrP<PupR+gM4w1(KWu4JY-(f~zPC~?hU=poJBlc-`Mj*hPL<*O~Q0V}4S@%a?bKv^{
z{q<zqS+6$%^a+GI_3uW_taR$F!I{bY4VfY}DkYz`>kMOivy4D|Vx|Gv%BHn8p}c}9
zITdd{q2zEEMkTK3?fNRD^cWB)3rA(Csi`iltgO5JA+2U=fB@c}R;ZFeOKB!uA)ya8
z1EvdKe$loYM<gXuObH8`9$&H<dp{tDEl0>Hg_EhYx?1FUuKR{xm`A)nkM6!7HCXQ+
zZ0Wjn)Daf^J<PJKF&Gw9RIM?<>J^VksZw(USDz5WQ<^UqI=(nRu`TfS^Yvw<1(z>>
z(2K&T(qiNZg>p(dUtXitR4kFiqe|8V-hf0G@43`;b8w*bZ%HuvGc~Lp_#G)OCc6U1
zl)5j3p0}BE7hms<FNfivOx$oe)*(69b>BM?hrm}~69Rr?Kf(?7{;ibcR612(h^<q{
zojG&{ok%TriiKE5nBUkxlz9Kl#ou-qn_|-Y^vJ@}9&rITy7A-qK6B<o!(X2J=+W7d
z2u`q;LgDsK7!I6ValX!-qx_vcZ4Bm$#YXA3z4I|D>Gv?^cYU$z{se~;oQPgWqTL&Z
zT3eQCho9B*vU{BL=d=>GG9;MpTh^Ut4WhX0EZ|bM&vb+qi3Jvs4ds0{bq3T8NuSby
z(nofwE7v*BR;TwP*6%V3Q0XiZLAY7Xy}Ie3aNjQNMjL8;c~gQ`08RnIf>52iRG2hs
z{50O#hg5y4SaV_WPbqXVf<3!*l3xuXsz@g*Y`#xwljpF&e{7X@Tm=>r_oASr@^h_`
zMslMqwoS0QO|LrM&kwfCbis0iLDczL;%M>H2^rGOAJUMiCpTZ945U}Y#xEFf#X7s+
z)~MgvIRC@rV)ul;V^;>xJ<^nMZTpc9t1T$2C$R~a4p|Ys3X_peePuIS&NeiO-oiKZ
zWRC_c+w+|#cY^MHFAknlg8$G6$o3~2CK?c)O;S&O?U$api=Q1-?3!5WARv;{pr=)(
z&1?Adh6zJa+g}5G>EJ$5q`CP(wg`w*Y_B;Sg+hD9ZB-J9&2@WS+P2p_Q83;~RW*@y
z%AV1v<xyN6Rn{h2Actd96e~B&PsUQ2{t7FtnVQ)?CIuqO%ES3t>YTW@?lrnfCLwR)
z#O3e~;<`$vk_v*R8vBRhZ6M}+<bj13@0M0Rm?_3$VgYGJtPPH*sHd3=6Uz~z`m^TR
zsif6d(CgP&M@%wmYH(J<9!~GhwW1S9z|u!GCHT0&_aSN%_|n}7W{Smwl@z2Uh<NDI
znQAOJPk-f@%qM>tHcaD%D5*0z1c;&IfclW*sgt6~iCIoBAZCH_%C?ssw3IP8E8=Ou
zVl|WZYS)`I0*AwH%DM$Va1`wYTCW3LX~qDvotZ&L?IgP~X8``?+NK6+pS0f~=EZcg
zMR7LzB**Qvs0f091|X(buhMhTpbt&zND^AdV^^Cvs(V~b(j1$)vfofFG0*j1Ag+qI
zx6<_oxByM27*Gsbc$#}y{nm~{vr>FCsaL4<HT0<0xDa0<L>m=MU2xSIqb<KFn9OvX
z2P@Z~UnQKT1KKemRjZH-#7HKW$YLc{DK0)szbQb@@iZTVvpL7V!RI-(X@F`Htngvf
zF4sn_SUl>BKUrbQBuAlpyU^_U`fn!RmBa1g84Pkyw*STK!N>A!AI=|a)1j2vPA%15
z0gmCGl){_gV$VPneus;kvrcIBaZBsvA68BEY|%@M(Xge)Y!>F?umk_?(|tjf7dvPL
zTz%5)*xAqAeNovFaJ_+>n@i0RvI)S_m&F&iZLDil-6l@~mHAwMKh62z7v$n7$>OS<
zR-)hP0kC%h;rxNDV#A)}H!?5wB`;PcpSdhJH@p84zp`$$|07ywYStWD_v&dedL?Qx
z@(B)n^z9zH$>P8F?LN*Jj}r{I{2qydn-pK*o^}Ukj!-wAb@jb&b>D9|-`(Txu)s<`
zlXpG^vHR~6(zlr~5jY0CvKqZS+p-)BT4<}I0jR8Ttjsq8QKYDJ*wV*{U5%ceZ9DG%
z8dkZFS(fG9cQzDK%|dXVb%(&}I8wFvduT7pau^cbJ(luCN#U01_d3nT$k8Q|G|(7n
zv~W+N0=#Znn;iKu#Uw((Z@;H_p+NYPK6>x1Mw&Td=ho_X<-2B6o4307S-9#LC<nqc
z>`}D8@O(4yDVfH9q_ZlFN}$GV;!jdT$%^X>-V{zLgez6feX8csBP-@<eWP`34H~NW
z`+YnuJG(}QMVlp!uu@BD(I%VQ%NP%uypZ;$;4L^Z_j~7rmD@3<n5>62n&N1qP+Gng
zh_8s&;tL=}&?50u8cD&1b6}G7Di#xF-}HBO|BF!r!Pbe-0e^uTFZ;*O=O|*EYj=M+
zgbcazKKG-3<Ffb@_9WFlQLj_QL2pw1jWH=pDZ|gUM3XV2r>da<mKZIH0OX(-v_+i|
zP$@4C9K4!d;h>1ukdldMtnO=bKuj=;aa*V}sIksW8}REj?b=QKY*b?NH7-70mu+z+
z5mL`{a$*EG^!7Kcw;mYr*V=0=o=s1q4ap6~$eKkJir+36tK|ysuXq0Pdzqi)%Azwq
zs0Zv+GkAs5E>kqAgt8$o5tUPiy(JO;jmULdgHawsEKQw&!t^Fpm=!^$I016NYR_Bb
zOKQFUz(8^3qE3CIS88F%7wg|#y|EO<eX#++7g8jZA}27P#r?L%`o}J79}4ebUE|V;
z+k~P9T;*mdlfx7<CqoWiZ(Kh9a}0Hht};%1F}6D7izPTDJM;wag)SWrJjnOYRD+qc
zvEB`D1&vutm*O+&3K57nt&-c!glmBB&Q{B|a-Ag<d_2fFQ(C)JZC^|R&$H~-ACdao
zl=lSH-}h5&$Os&F(OF#I6J(k})Wn&kBP#9|Fm3;4pD`hJ`f2BJ#!^SNmsKc>zvsY1
z$1SKlII=hj%1J5zGR}bYuXN&%Sdk*#>QJ*i>U1J8iA2y4&|N&USfeaO5mO(;%Mt<j
z^eMljSY3p}PUx%F>;kVq@JAP<&32+yOMiFqQhx{mV>O;r3>rBz?qRkChoXwcker-d
z?BGF1kNiN#*Cdo*3rts{#+$K2d>__9==KpACYouxBMG)Dg5NK%4}j5q;mT~xMt@c%
z_#vrPZJs}e;Eo;FCls+BUxm&Cz%Yh|V$|06i5eX6C-ZmFMOO)Xq*h{?wRc0a<;jrV
zaodyE6PUoizCa~iA-6g6QWFjrp4cLepVaD$2r#|kMDo?1-WNbQ+*0`#YY}8lu;d}}
z)N?r*a4u@>zfu3XQhTmp+%>2#=9TmMNcn2)4gMTt%1FQKlD?^Ejq)4~-!}}}8%MkE
zgaWZ4CINo^YiFIkwJ{^+Y|8l+dhPR3-FJJojT&A>xvr1*xXc9MYVi*biQ5aRy&v&A
z9%q={kTNSQ;2J)7K5qEgx)l|jE9E@_Iq&OlQ=X5*Tm$+WuLya8r&?isN#Wt~59?Wu
ze%p82iUp0zU}V^SJx^Qcb%G=PdMi^pZGdzf=esSJ_U0mc7DF;8Rnx-aVFC+!;5!1g
zk}Hc>6nWkHUXz~uvGU{Wfv=ILAni&*Fzg2lt0dI46c%tz7j?7i_@t_ndYH0856zLr
zp5;0wjuY?4EgX=Oi-{9IiAVNfI5sz^LDs%6X0XEV?LMYf!50)Yy12IPwbSO*pKG-5
zX$x|N;9x+wg5@**P8J7QNA}`}o_TJI;zH5n-)1!W^QoFOO4`a;60D$({85LajWRr{
z1{vGcLh15R5h~hC8{#?uaH#M5KYlOta3lOq&Z^k1{)S)9V)sQEd7iKZ59#Ih`xdXQ
z0b=KgtNUr1ii$00GJ~)PNFm6PhjtPJHy9krSl#rFdLj(z$4>i})tbbz>$3R>c~wIz
z7#?E*%!lt54}SBuZ)zky;Ajw6dj>Dt4chUQReY30ol_aG^MG8zLlNM88;{6VQ8V(N
zeD3P$GO~+$a2MY?O0KXs<Sw_p23`YwOpG`Mt^aZjudxvXf%dQ6f~Vh|B|WR2E2Om4
zFt0R!(IE;+2mz;JAi<K@@(wMC<>P>y{YE9mG=%c3xHH!tX+-5z$#|x*UqvyAaZ_{8
zRLU+hOMiy(GA&%Ha?)3X;-P1d^n3Mqd6H?(yUCSR*-^246wXhjm)eAwNc9<e8s=68
zXMf{{t)MJS!bn*LtT?9Cq5umSWPCkrUh(#%mO~62j8n;+J6l0jEQz`SY{%Ht(%E1y
zn9s!lWi{cQZ!OZqwxD{ts#(Ld$Wv#7gP%!J?{`&J(JL{@>`06|v&6yO*=g=bQQEE_
z9K%yeD2XHcKo5Vy!2*cfkUY)XoVd>TCKQ;ueIFxeu(x1^uqJ1g5WN<2fq{YyXkWau
zLNtR?czC{0+GY%~WtBi3#vni}yotmv0~0L*l1cTlm0~^DXWQm_#3Q6J+0OV+Y@y7w
z0lL6qRgwXDyyA2P=Uz}2p7I9)7$>NRQL(~8=5vcqYH*YnvHD2<qpD7#O|~P`<#5S|
zZ+FbM9$c~|R3Rj6s!$XfS&wKrwOp6ybA+NpI;?jZ&)RXnnla1N+Wgyhr%x#po`%ia
zl^oFR5;@y$$xQ{R$i+<EFA3Obs7tT@G5%);>~|3|yT>VH?30=f8cG&>)Q}F5W5DB%
zB$zQv0g~ygq*5(r%6vY*Iit8-9&|MjBx)e^MeIu#dwshi*ad&YzZjooyOgbMwY>Dc
zX2YH5FCShOUn>LqDW4C!3(*i50ns1WFpbT#4Yy~mgM`s*PO$F7U!ynn2NsPsC-gU9
z2RTB+eZLHz{vaTzMPS8l5{tD=nrB`uY;i`TFvNrlwga3D*l}g?X?OePF1W^AEQeg@
zwt%ZN<BEzdXsckSlZB)Gz>AWlczl^e*W*D1Fe;pUjS!DT$weB!w4QlB6!<)^_WZJd
zlC8YfW>cUiq!*HVy~1vkHFJ{j@GAnEsnp+}iUTf1{s@uapdjkcPiVZyVT_)`WqMkx
zRJ+B&cn8<$ChBqPSN*6UE%_YNsBm;vXN=@SeFAFA?f2a)^|rNVCM*n}D!Z3Dx%rXQ
zWsu1|FKAdU-54uDzt+k&<uh6D(U~PRZl~?+qsN)WU-eOG5v>CswIGq}3?Wihdii)&
z22W|YNcs(nqF1AG&0ED_FrJKgK1Lhv?mq|KQMV+r=_nU<lSWyqkEs~Qd1080;8HA3
z*wecp#wkg7@>82TH*BUv3I6)=AJ?w;^*s<gV=F-SWu;T;sY=_u{^>|!$xn3W{5+wk
zsH`6VRSwovU|nx3-Xo^*7o5kFQZ}I)<=e^dI0d;kTaipWUQe&?BfwtNdWvkwu&^m_
z{<Qb_Ighvq%8~V23!Ao!m~xCo79jVQl6e|VsDUMOAN+d1Jz-ynx{n+F@`&9OnRkKu
z@tN6z_O9JzH9S)uzE>#5F>oU3EyBl@NgTDL+usliB&r`wq1{V>*i>|h?b;Hu7DwhO
zydqafJWp>ylepDkz)T+G-f3IdoA&Tw+3APQPP<hOzVx}jtN7$ME)pLes*Pn!pBt_K
ztp+>RYL$hgq+<6p7Zs?zBwFr2A=s-?jID4RlYNbnt8MP8T4p%1I6J5-jbjFkE7R}&
zWB<KZW<j`_xnK|rg9fRtWQqcj+~x(Cj{Pcu%L_s|o?wrqk&O=)z3W@byL#7i%aBp}
z-Cb%p-ja3k?%(58gDj@PYNOcJiI^S|EP{;+!~dQAngvB9(8ghKsz9=TK%4-f7LiSs
z19w({YlBll-2=1Bs4$ggl9x{}D&P03l5&&X*|~lADM&MIx(QQ{7HDK8gqlA>y>=x_
z(<FN^7s=N6_m_bNe}io@o$=4OO>#gm!8=FR3HRZIkw-(b3)?8wG<>bM14h9m#Hz9h
z8RQ}fcClsTi)uO(=>#CTeKe|4cUrhVUB0(ok0VF*{u!1FdpxA&jj8^s%{)|}$q67G
znGX4~#+5bx1389!oH?1bo>~9r5}o#)w<CNDJ#C5LH{QQ=<%c#fO?Y6$e(|x%JHN&D
z@_XjUIhJE-3#3(V2{@<t;~ziOq|6{h{(P<W>B9$&PWSL+?FKJ^*I$<*d5?ZDx3O$4
z3-inFL{zQVt=~WBua0J+t$Em8o3nRMr?2Pz-!-#<*Wf*WvHjFZ|3u@5663AzXR#OJ
z=a=q(>2tvR;fxVOk5Mw^%i|NfvpSIvco)xGlrLzeP9y9uE1!Q*ifkRf?gnnwp2pVM
ztCV5df1XB_;G?)SpLd=l{*l8e>qSc5bobm)=4#~SG}2@tRaY^r7h6pvTt;p^-n90i
zeE<y+R*tmv{T};v1E_QNyPEj^#L8$}?WNn;LO^{Crp7peK}qhp{_B1A>Vtob&g;aq
zsZ0^wI$axD%Q{nNN=X!O^P5<5$+@USpJXipF=E9A5aoCl9&p4^Fw9emw5-#<P+0%=
zGDa`xhGNm}=T|h46si=XqTPJGPL1^rB!Q4Xl74GPC4*WeV7&}nx!|+YQ|Dl#MrRUy
zb~M)p_>lWX1WowX$I3Qva1)+h!&6=KS8somaHy;-dxB`rT}eZ~Vqv>eItn91!W<1X
z?2U9d5dy2dvFVLGZUQJ3T%sI{C5zUW2uQ8hRb&dwDC9t&JtJtEHrE5jNWx^ONr&WN
zNnsjfd22VX-h<B<J905Y`cz5I|6FvZwRz!+*aLqT1?Sw;(+}`KEJ?^++RhICQTu?P
znk8IaQ1dhJf$Q5=#Fi6W_8Cthvc2ZH0mr6o4ETnFqh@FEwdV_4{Llr`)mx8<Kf?^p
zMiK@&Cb{cbZ1g59N9R(+H;~yhJX>8OVS+*;B{i(IowtG^7@G6wFc6#Jie)=55NXJT
zm*)5(3D}B>G8Nuq$7j`4(dKz33U8I{Wbcg3S2C99vJjKc54lau=*BmxjDQ~L43l}v
zj!M6@3GeOR19Buj?F-;qJ2&EyO6u`c)1k&|k*@lf<4=javK&Oj6=T$g{Gb^bqFoCq
zK0Se#G!^9vnsu62(Lh?Y{hCsHX1J!_!+)`%A%sovhZi{o%K8_ne%sJLnS~xL7tw?_
zhvdurT%n8?zc)krOaBP4l^U(|*BvXJNQ2(;mFE4znZ}DiAo2S;74;Uo6PfB-X4-0H
zxiB6;hc6yYQ4YRypn~p?slh@naSMj)z+6tjcEa?#VH0IsjZI7~09+jyjwIa`7eeRY
zbX)dzJX2)uX#2nWvpjru(L2(M;7J+dxOG&%yGx=x^RcXE2`dthab((@<O5zYjpP`a
zMg*8p`_h-tKIHHpWSX{|5>$m{x>`$L2fUJsrslN1aiD)w%2>ncLCLJ>ZN_wm9b}@D
z;BR6slpING6lM~oS{<-F7>?e^sg0tp@7R6Saaxd2-X8p9v$9~e+U!WJbg9^6tnL<P
zCm;#PX}8rXbg9ZD@=G%RgbSG{M-&ES;A#!Jt&HYV)xmv}bd)O*J+&WYi#UB3Yv}4s
z-4zpU;?kHww7%(ViLqsk4af~U8XsvjOaj5@1L=8Xm$_arU|CI}zOcnDj~T}r5a}md
zQmR?N5p#EcOV^G2dbg(#+GAh}XFMhFFT9#902}x{tDWuqUHe68@tu9}nSFX_;QQ<R
zQ!C)A+lP|a?B(;>&LpWT?oJ-&jTEY|*E!ti@(k=>h`#saG>u^)f-^lkAC*787WKn#
zQ{!eEg!dQ?Vjhe;|L+19pN5`kg4?58BmR5bdkTrt>2Ln`3;4<VS1W^{|8ZG2k~6{)
z+cx$8J6`bmk$?R9|JPB%A<qB%-wW8b*(LrLi@ai2EFmUs^&-2zfqxWaRHbVqO(6dV
D?AB_a

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_010.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_010.png
new file mode 100644
index 0000000000000000000000000000000000000000..6484393f9086b028ca61969c308d1661ca4741a3
GIT binary patch
literal 45718
zcmY&<WmsF?wsi|F_EOx7ySux)28w%dEneIU6f5phAW+<)5Zv9}f;$0<1n1-2d(OSz
z{gG_RUMqX<{XBDxImVb#>Z)>|P>4|8y?ggbL0%g0?j4-R+kP+-!rOPTy|U}uM!;Q0
z*Im=u%H7M%)$*N+nfq4<XLko%b8=5hS2tT{CoWb#RxW088+Z4wZh~xVj{kE3tFx;$
z+l08(;oBt0U*+}O-n~ON``73FZ?O{Fckk-p6{ID!zGa_2drh&(Wo{t8aMqW_y|%S)
zY;N}yZcT5*@1RBy)5S?s0dOosR@7~5Uw%?*V}35zg*tsuQ20y{C99uUIQ{jzm25BR
zDxSDkVet3PuA7&D4PxgqMlM*qAf$9evarOtFrwFKF7)o}Rd$7Di9>GlU+;|r+y#B5
zV7QMcIP~iI;=yEgLt;mt{%iNY&Rx7{eDL_sk8ome^dYVPd+mE138ViE?TUZ*Bfm=g
z--nxW?_|_<-sb!7NvMJ-e>44S`hOkXVFtn4nJr?>935k23xhGX<9)MGo`Os(L9<6+
z52q-K+Q4`1{0Cy$^b_}p==}n(o4@!E$b|i>51xEcju#dd_<gPq(H(YBo>u!S4Vw8~
zcZZDpi{A#U{oesk8viwbC>G(}>kj1k@hLv=T&Klz2Sxt3kJ6ZO+I>b|VB!@%>&2FJ
z-rmyi4t;5jwkan02lFRHa**<Cn7H{*f{M6EU~D5E1v3CeSgZ@CNWVwHf7d3i7}bae
zWL8uVhR895Cx_7gh$=*~ceE0K55hp9@h>{3@DESKqWCZpQj1r~qkl4N9gabOZ7qt9
z7cKr1XS)W8K?48>M`2MX0{B%Vsi;6a7*>d>KRDbPivZ-%#*X{G_UA8_bI0G$hts^F
z!KS{Dsrdc-VX<d?JrPeVip29e!GXBuUV>N-4h>>X;y8lv%x`UHD+CSGx<CTdh^Ob*
zdor~N$ol3c*Xu>#L*8#~#zVuvTZL}Tkt1v$Zd(Hq0slQkj~TQ!*A=v<HT)vAp9_7e
zi>6a~4ZUWNBc?PZTktW|3uBJ%j9(Qtw7Y<AHG?(7@+Yyf6YI{hq*`CmWYDD*(i;qp
zWfOUQ&1x{YA8>A26?;4_dM+h<b%R=2cKC7=@f)3~(%n~WMZP|CJ_?<mWRG4xm&JNd
zTH4S5xLP^9{<h)vm$c+u>_O>utLXWDGQfOBJNL1{ma|vmC99X-s54foM9q3K&%L^E
zIUJ-QL<%~&2<<YG+wZHYTz7pi{`L}`bldZkJ?nM7l%e2zGc~LL&J#MA6oDZo-L54I
zo#qO9`puAjmnvZ+2EhQG2f0`5igCTB?7P*kv9J67Iajauosa0H$?y#mWcSy^EX-^|
zBA*Qml5V5f0@jbfJ~y{T?TyPh4d^7f9;sOwis!?hn|X6O@*e90ZD(G_)LufsUnoi3
zD#o4NZZ{?x8ijsjjD6u9A+~;(L`dcUO}C(HmbN-g8fm}9OP#%96oL8pUfOqZ*mmAO
zy+W^3LP`~cW}pJ+?wRkx61D=L*BfCT6R(Sbcj%?s<rCjNnQ{VkN_XR2K;-S`B9kPr
z`c7oVS$sLDvp5b2W={-;7*!T8N-@kLVDO4uxSYl<LkV&POzxwG#{8>X56bmbVJv|Y
z^*fHu?R>UkuZV#VJ?OTOv&u>d>5ELsdH0}bwv(CDW3Pq}gm2Zcq*_e5!<V1v^7YC0
z6(aUGJ5WtxT|&I6wRI>8JEFz=fN{m9jAwLu=aXuV-u_alZ9#&0D@XXjBS8|e7}v{_
zM3Bc+DuD7`l!R_<l+^WIu@KU+UuVpD3hjHmAOOnb)dtTqeOT8L0Ru}eBDuy)Z(JzP
zI|dQB&KMIE4C6Xgh7Uc|!6w9m3hz?xxn#{^2%Vc{<h=sX;6&|{f(JBdgeMDesDq0a
zI0S!i?30^v_qj+D528MV+`iEGHHNBhGD-2&oO*e-xL>NMuKSYZ9?(hD@}#?T*;Xnv
z=?sRY{CxQ;P4K<XMY)1UgV-!)D<5tR#?>EikzogtY}9Wb&1dD>+z)A0<o$hBFOeg3
z7^SS9hTh6)bl0M1L-)IUKDz_(v*QJbe9RdtNel{(pj+QDVE~0HY_|9Z=WQg;|M<*r
z87i2vdQQ*xwIk1s)fL?w;<$W%t-n_lp}7Bt`}j!wxhh}PB1G_{N^I;2BOQ!%TnBGT
zY!TX;Sf4%d0pTX|5eQ%{IcP}~qH7%E(XK9m(74eJ8AX6&8})tEcH^V+L{eKq;*8V5
z=Np3p)yeQ>Dd(hf^JOh3?TH)p!RJoVC7g>=`wy_I$Ai)HJHoH$dp_$TbP$o>qlQcn
zmfxd(6R@>B$(;M?XD5jq_YgrZRC|kRM95jW_6$%DWeuo<HshGEa`yTD{AC?%pgeGE
z=RrKz0!e7^9HD{vu%*;V<oJ$dL&Dyk3|UrP_0G^U)NVh3QD^<~G5+mOo_Lqn8WY={
z9J3Kj_tsjjxOTitlf@2>YDW=0mbc7tT6eN)?@!Lbc~|EQzlFL!SFi7mE$_7beXw`;
z@fChpA{2cYM$|LbFW_6+I&)j(_t-{Fam~<;v@mVLqd-**Zsc@RmigN&lIy~yos^53
zFjmG8{?~$&OO_~KzLZwTs<J!rgHUa3Wb8*KjGh7&W#V|7=3ARHhbgD?+Pb>FnGwcZ
zNEefeMwKd6k>n-?sQZu~SvI!Y6jN84b1VI6ilaldLM9$wgMx-2I0Y}b?}}QIcGapv
z#s|&}cSL9S-LEc`c;j<gFc=@hriOHh039i8Yo?#?UjD?5xr%r|KeT38IiJHyIJ5v0
zPuN7g4iP@9UpE@Vt}W+W483?)Y8lZ2Ixx9rO37E9Xr#3lAY2dc;i*1kQlgQ(!GExE
zS3HaFxk!7KLIWFyWaZ^GKw4_{uk{kpup;8bxNRrLD*vqJEfZvP21@Uz@p#o+<~@au
z1q=-M6U7#86p09V4IP19PezJFeynAIHPrGpWmvz??1WY!K$D(fp`j7MetI_6-zo9U
zDwtW6SF#MCE`cxM?sx~+QNX0z<@%DGy_lh=cL%IUSuMXS4vaNvRjUi1Od)o~YVGaM
z3q>Ma2NQojaSjT8Mhfy32}|&0bzbktT^_iO@bO$QdX+*K<FGw<Fm!$B8H>-3TR)#b
z(7Q)m4qfyrFMUuYxRph6gy8-($Ow3?G=QS+(<{fzH16gm(O>m8G;oX9rYS+2ZXb>W
zZ0FWJxU)VVdF}>S3=2Q)dO-v`k1_~f&jEW+M(-8(-D*8T6HookhT?Eru9A#OMq(2J
zJfERFJj9-~XWcbN*ZqRTua6t`hY+UZnr}Ul(C+z3VfWM#sK79S{kc#2<QiS&!bFk(
z$%=rrz~_?3=C;!fy4Jh^SV~UZ>|(kC!(fX{n&@rl^~Px2GAEyEhH_WD$7EXL5zqcd
z=*{Lw+n^1HVweOD@A{eg(E)46RoN`;P?JGPV?>15sv4Kdb#PuOF6!^P-V?0H=%dH$
zwvLH8!5Ifp@>K2id|GwAi*J11<6eASn_w&s&G&ZEdd*qj!vNOjP_5xu>&mW77gJhs
zU7>rIpKviXN%`Y2ct42TNLT=lg*C>2*6;A$%X#x@r1Fa*(Cji0_$(M25WVoio0zUi
z7x#7d6bpdzFkbH9d|o?yT3h26Hd{hk$k7;6n;+Xn6pD)wEkQ{eT8!Z-sX|ythm%Ez
zixiH5MS=QrEKv`QUO$*mba_CxLiHB3-X_KEutga7q@s|&uhi{ipLVjyTJOF~KZ8=i
zo7AO^68!To`EDONs<v+O0mYXOVPU)<E?*EI{|yVf&=H}Gzqz#8iFC<(W~|u{Ydu#F
z_HRV0OxE{7_>;BljXkwg5$ZDqu@Xn2HF_cPAv@GRXQOgscSS;_N2pB!BV>r~3}-%5
zs#co?W7z3kx}ja1$DV(_y?lYLb%agBj|5wKct?TmX^QcVk;l#a;c%y%VK$i}P($x%
z`}1tz{gi@gws-o`(Y(dJYFsFt8%7_^qy^rIdN*L?QZpBk_mi+unSj(XN^A+-6lUfA
zCu&T9LYOl<RXBw#d9BOX9uvOz-@?4UwD<HP0#@(d(Az8s^L>Xp5+P8duU5kTxvB*E
zpIGBr-(tlEY<uGtA-z+uXiZMvkJWLWvPEB7iTrl3W|`~#fawQ%3pT8aAYp7qFB+E@
zq!~M4j<KK&M-8F|JbcT|%b-ZIeFt<;MEZG6uAysxbi48db$(|2Y=0o&_CPO(R*4}?
zG5##Vj-Y{pq=@TIJoC9hgGQ-Q+lrll9VN@~+kt?Sp>HS^9qD#oMue`RC|a_FH1@bG
zCx?v$QgX@9vSb8%S>?OZFi#%(1Zs8h#K;6{d0p;Q_S0EACqQXlSCZ(cn_EMs3TxY?
z4<{9pgF^*ooFro`k?g91t1O-(Ljy*-OFB~lL6|t>-l539%LF*Fl|SYAtauzFu!}JC
z9x98VCd|dEQ!^DTcf*c4h4NK%Gb>WFK57IGI8<Ol=SMSOSZwXKsi*4Jy>UTdlAW-9
z+6Q-l4>n0oEvZdx?eN>mkBTT_9C-HG??}nSGN8EU*gWFXXHKsAu9<DP6gylodK$qG
z36VJw?W{d^cJuE@(h_*77ak4U@8F(aXwgOgy5OZvP!>frQY<LOsL-p^$Xi{FrEfGS
zV-dV-nC=rCKo7Afp75ck!R}89UJ64+Adzino%B0S%g8n8i7{_#Vs*QJT9p?NM~-(Y
z>HW%Y?lTai!0jd(dFk*=rcBz`rZNJTgFH$kRGosBH=XR=bE~&?s3K_W4IJgb<n*?D
zSP3O9u316Z3$m7dQ_B{S&+Q)8Ee3q7sw-@wuaBw_c<UKvVTt*p)n&zcsD6tF6k048
zf1+cO)gX2~HpxFqB9?J{_SYw;Vu7~0nh13U2w()Q@(}EMI1!%B_aK)HT3pldvT=_c
z^0Z|S)N+e1J|N~7j@cj0#KDY<Ia2IY_pOSvJ664_R931xQlXIPLMYFkbt3Xc<GItw
zuho?0J#A~q(dzp8c$=cw;-_`D89N@jGpdO7vyttmaIfseHz0VHx2#kzY|q0d#gb)o
zv=snzpla84SH(^r$3d0S(cvYgg@_Qf!XU8F^+uHun!K&wCv9w_rAxj9V8N=`N^S!W
zXRp<xQM>m_60%}=EsyxgoWTAHA#Fo&t{u9;S@A3%k4T<(ic-)JtpNXGg|CVwwu3V3
ztWcI#!Z~(^M7Qk$t6tJQRAO>os@?x(zo@<ayejZl0wK^9&;y$R2bfIxPn}Pgvshzg
z`9VHm)(wAjaax0n+t`}it&pj%<m50-xW<e5@cNsrxplNJXE*RIzJZgT7JS;X<^CO2
zNN@b>w-vrTyCO?foU>^6M-SB!0b&s0N$|KW^7FMvb6Olnxm=!W>gSl5%d@k8T6`(p
zCg3JjW-i{aj9%-YFtG`#cjdb8R#k{#g)*~5Q9Q<vr5JWz*;w=-jF8wH-p?n#V=9xh
z3Rp2W?z%<1Hs##}G1Ewi6@X;dgS@aI!v;M2+32OETK~vtR9}``nAegnxHm=VU>S14
zdHksbHpdyEga3QcPFY3eS2E=Vb<TolyK#KhqH}G4^{2_u0j*fTa*1K?Tj8AK3X%r5
z4DjW>I;ukIyF=ssi9x~@m`LSG=Xi*M;A;dW?f}p1q5LF^;30BLPN28=C=tA16RI5i
zw$0GP&F51`K7l&F$gkYwNN`~Uo8Fo1L|EUcLR?b<a6)nV5b+Dt=~IaRDgU&h8$)1y
zR2C|o*rQwb@A5r=6B;5&-Em&16@){|E`EYxLMm*F@?g|E&(!?q^;98Q84D%mqsXlu
zmwx>7!l`r~?Z8T-!mdM-2gwIoz>39yDg!Ep9K0u=v&+tkx7MacrB<nqe`=~du~1vj
z$Oy`~6(WQU81rk|1|wPNPSJcPGUG@vcEy&rx#^-kX!@&s^<u%Y10|R=BEDj3ZFO_c
zIUu2*YEYofhc-il+bZ$KRz>ogRi#RdLW4sp_1!G3WEMgBE`O9{evujj{p>lc*0XP%
zi46!+vj=9J-SR5>AiW1A%E)o#o^x?^4!sy-L#}GXD}Z4bcZxI_-aw_)xmKDRIOc!z
z7k=ztbtKkD0O0|vy-A{Wrwn@^DQht#g60UQ{(K9FRSPUu0vW2<C;*d$-LUCsTNfk;
z(w7KAr{JcB*Jirg8s!(t!^P!e**R!YmUY2BLI)ix{KuM=iyV!%a17p?ry6mjdS<qk
zx%xK*p%U`yx6@rYLS9q)vrf&IMFskmt~MSSiXd6=Z1H?+<j?PkbQIsW=!@pp_rk0F
zgH7)CI^%HOQ)(lLY~Z^56l59j>H3zPk3rii&LnPtfoCt*XM-Uc&=`l3zj$64D;p+)
zO6uk0IP;H(t8P&dF_Nq-Cl8NY3l7E+(jL<V)~rRe%Bew(G%alt5f!txVi^ABM?Bp9
z=S3)Kb(?DfITFA7w{CBU1v9yO9LkWYcZ7iEB)xLU>FRhgk?)>?$=6PPuExOqyTn<Z
zktI)Vm{DtG<|S`tuFIJ!tD&lG(D-;V)+yRL;W#!OS$aczHVlVX$nsnjGMM2%zB2qR
z_~d;4q-a0ISExCsV=slfKhUMWQRoPL_GWZVtBo-0Dw9lF;K0SFWr#yqPbhPcG;r;x
z+&U)Lo(2|5BL>3v;sYpJzW!aEa8-g84?Y_;(NIch>>X03)WYi;rh7d|59C@T3j{y{
z_GqjHvT6%6a&kQ%^egA^_C`SEs*vHN(yfabKSiPD)vEgosHfdzE&q;B)e$O9shcO`
zvH{cB`s8$aR&)={+1X`x%B*ZJDAr@K$-<s*9h$S&?zyM0*y)KN@PpIshw=cK`XIC&
z&+?S}dTPCMKftHM)AicP-;Va#Z~~FV-ZM)D&o`~vc7G|8U#%bHZo5G{II3<PV3osb
z{VDMVAN*~rU>3#}a58%2#p9LTF{?DL0M>A^H3(F;l)HQ^F;O8d+&MJ5mDccdScG<*
z@PiK6I?nV})?2*IW}GaFaQ3`);-*hZcD%qX2iN@>kO;f+lm)|9Kku22fL;<l*y@qs
z>nYodbyCxaMh2(bj7OYk94DTLEZ7q;KXxvj+=9;MWb`%lq+#a7-K=~4ymZ;!fe39K
zBC}Ppm^?4QR9frOc=6Y%7DDKSJLi9@3A@;XwB&x^pcI!M7O{@Ug846;H>4XdPoLO1
zZLC>2caHx`u>*X86Uqfeb0kGe2Pwz+gOZ~lSs1!hbnxE_$l#S?*x`V(Ypnsn|2TI_
zm$Zk+@l;wxE!U&F?9U;`F;7mPH7+f$fOO6Zi{x+^Qq%Am(%^(Afj&N&0nKVSvSAv1
z^iF0mAj8U;mPXmtYrWN_?ThKcKA%3J17qOxtw&s58yV9F{UL>U<mfGNr(enP+)h|s
z=I)uLb{5$mz5xo_-ffY5n?O$1>47m`_tJ(SRm#z`DX?d-PO>-GWD~>NM;8241m!Lj
zSTLFrd;oY4lY@X5SbK;c1o$rg4?YC@-18Q8L||DK(Fcd?OI;<9`G!@-(#8(pnB0vo
zE-fE_J=w<_jaI_;!Lp}Q%R%%cTkZ@k`5mf=hKK<$m5HfKj9u0F<8$fZJ_lM{ry!2g
z)fXMeU$mH;4eNcsx|((ELz0@Bran@r^*2fm9xpYG_RBxQ5IXuVkug@)^A>Kv*!FfK
zPnfTR6DmWB2pU->)+dpsTi$Q?pG|glqv~}*KY>I%3O^#dX>;8XTQ@^922VxHYO9R^
zG*sfi--{(0Wn-sv!kV*6+~8GIAC;9#UI6WW1Z?9Ki76u9JW@S((v&@4JBzYhw|OBX
zHAvEw4wol!Hz_25<-aqgDE%8$bn<Bp{tLs7#{5#i44<l9?7bQ6`|S>M^!1+2q_~l<
zhoJYqx%zS{6c!Wy8so@Ijl{+Q3s<O|qwe*n_?4A>pqPk`l{;(7V<`GcoBop?6;oTi
ze4hRzscvi|$KH>@rzccN*^0__Q!_I{3dZD7(ol&!a-<qO*}|V0%YuXGBn9(lZ-B?v
z%ZUROScERoKXY8Qb3c5#eKUHF@f|({0AEV2SlV|=ka2uCaAPhqspKA-lxU16oyc{>
zB-5yG_FGCZ1qFcO=~Y$V3|6qwkNJ)OVWJ{XXH4-vOrT58becNyOaROrd1hppWRIB^
zOI-3x=^_hLU~<a#VS%6kPQCxI-@p2eo{#fdoKwbyIL@p-qobSnNMA;7<PX2a5Y6<3
z5m%y<>W`0is8jW5as{1)IzSPQf@USXNc))oo90X+dleqj3<_3MhI4@zAsXVTV>QS8
z#qTep^o{SdsVes;R#f6DnK?A`icfXu?{WS1G|}Lci!{Dzg%CYMBLSs-b93CiTS0HM
zH+jn7(>3=rdymcdwN}P*hTx+wm>g%XPK}^3N>)U1uax=%$BMZ^sp*pZok)#Uf@J<6
zXn<K^m?~VFo~`zh-Nkhm)W3uwLx@c~$3$7$wdMO`V2(1C;4>ljcMXDIbzSKKL=1&s
z)?ZjB1ZGKf{5^`vh$Dfo66-<-PFoCllK~U+NR}s}E7_zp_KykGSvFjP4dab-b>iqo
z?W)|$JfGy^^S(YjWT;I0>)=n57M&~)FKNAVs3`G?U@uHsHtgJIZtl#pGfdSn^!+Dk
zx^CAkdM73#IPH?}@RQ+lUykxhz>*sa^*-M0j9zSv6H>mIVzbG;lVTfD9{gTsAc<wA
zwk#6Iv;zsbL7rFHRw_(yD7l{z5&BZaTyzWn__;LB^||GBR?o>!r176ivgh<!Iy}uY
z{S9kcgGVkCL@jF+7ZY&evB^fj0grV{H~sUPPRDKI+QaPCL&`zN*7bT@hMg~I;O>@!
z?^%=mPEAVfY-y@0zkjVNi_dnpHY4imA(r_{TcFDNpxAT0^1Qvf0o4CbbJp>`Q{}q*
zL$S%jLecZzawmr0&)WJZ&Mbv&*}3EM4`V~cjFvjAi5~uk)+15D1~UvMlFYsx*@?yi
z?dA(*efx1{O}lABM$UTcg=2orBeutN`fX3`Pkb-?klcpOr+$dXL&xRgtKE!f<7u4p
zykZ@T7?R3*cac;RDrwNL4c3*;g69C|@xqs8RPF3q<V>`gOLCo=G$+|Cy$L<v$Qa|M
zG$!vqwJFEve?6&n*xChbn+=X!M@Oqz^#+eR+D9)(t<u~UtJ~a9qq(R8#hPOU3;Ts!
zj+1Bka;E2Q7S;hjKw}S}k@*}FgE7DAlBcSGo4_+2W63BY`RK&=g&o^jCV>%hq~9Vx
zFY=ZqGp)Jp(>u3&=EUBJ+<Jn?l+|KY!g2NO3+RbVl!s-ilQRj>$t)7A66Oq2`UEZ5
z9XJke_`Q!w@%D=&dw(+FvrR$K1UmSY?7gCmor6jg$F!^n`zdl0r9;!90c1d)GaGb5
z`i4YQS&lQ+tmm8)`3#GSiHR{z5u@UD!e4DcdwUFm>v%fV##_JSu>G@s;uU_9jUNpr
z3~su#m>L2M0Z=M_0@0}?ERvG2<v0><+3&vPM#L)0+Xi7G>ZB2@7VBosSyb>50W<Yu
zP7Z0$vs}Jk1hZa}U(HWGU*P{+XA5w5(UCB6gKYY4@uy!T0H&QhzCBiEwY{93Tnm!s
zvJ+4pc~J2#cnNs_ouZ!N&u6&8*dY{7;LvtY#fF>I)HRPGomztw>x+gQ+*dfBZ6ore
zqLURuKN@y=CIoaURf<Lv(?CjgDAu2i%t>ZW&S#ICGIC9d_z3D6+XrtaSZA)H8jIJR
zYQf-%?QNWn^Ism)C9~%K&%7`Z42rN19955YXus7y(&pM6ZPQE2MV&<+JSGV?BChlp
z*Y{RGuQ*?wIU8=)YF^FR`?o$n*M5x2utEQ5Zf?g6D#9rW2tt&oX~Z+bdd@xYPU^${
z%GLS*u?XOmYOBZvxKIg{@SZ|3qKlqIi0Krbypa?f;2QzUp8&_4lDvQq5FP4zrKtNV
zjwq1Af)Y!{`y^~55?+@&F6xGYb3|f6_w>fkEV8|#IjYg)<0J+27G)gcJv}ILI&rQ~
zyLwGci<sg9zbQd-;6-W_scm)(Yv0hU<|<B`v-<k>(FdQ1O6|Vu7)c_TMq+Zt0CNms
z0*vodC#ZkikA(B?DkK!5qF6N2m3wGgd4h4^qR4*$3)m@VH~>g7^1u0=IZv$k1GgwA
z19!64I+kfjWhzv=-C1Ps@n}S1YHdzpl}hgkJ{QViK0AGGX@~*9v1nO6ti@pSC-z|h
zv2#oU7Ooy8aEK~qP#>Ircr9#5?w;~k$d0Bh-dOT6JhqIlxw}^;t|+brzGAnWt2!!K
zSsDHly?*8{Qr!lffu~hIYHoj?E2cD2t!&Ag$>!2z9Z%0GoK%}0)MKQTYVRA!;cy%%
zl^K1>Y0mC&rP{H`D<o)B-d0LQ2}i}n?+kpnnU*8%y%Jt=O#8dCq8vZ^lx_c32K`)d
zKXux0d3ab)j&nQ*aTnbh9`_jg0%x=n9@qbFsg;b9_%0pI>Xk9W#`p<}=O8U!8+UQ%
z@N>td|2!+Pb#7nZv>r8b^4x6PqV}2%1W?KeSXz&RjbLCoGau)@^JRyR2*3@oy<6ic
z7kOgeH}dD;*`YtCMMmex#dSSdMuW)YULC8kTBSyA;N?S+V-Yxz5g1qmOlxvhF5wl>
z^jNf#jX0{EFvzrElIzRZxv!$6A#~+%_S!}~x(CpTeFx=nWei8KyAKYXwip;2x;$7}
zYl1<I*|B43fnp%w+nog5w(o+eo<slOVUg)3l&dC*;GMr9D?kX`jf4`N=-Tl~MIMKW
z{7)}9tOeMBI>~8hYb`nY2<`s!jls51&`Fx{#H|!48T#PEh@GB05ewlJ3IGrCj_J1S
z24dI24pR(r6do%_q5NNTO0%tQq@4gXSL?=3&&wy{ld+G8i%T%UCWxiFb)<46<>fG-
zXV7M8W0?-aW26jKJC5a0xQQ1=7)z<RM~-cQH1y4uWK7T!UQlA&-^+A0!R8ca4}dFn
z5#=!i-N4H-$h!8GQY9|cET6lCBP$v_@+JKRDBbe4EW06mC}h<j*jDP!Uo`XbO;YvB
zV_Iaps!TJnyuP)}^d)J)yMAN|sk5{y4m$`Y`@mg1a(d=$svK7Gfgop$@kgyu^hf*a
zK3%<`KILUyC1U#`PWD{A=<Rg<_HM}C=fEVTw2dwF&Wo%y{Z84V-vu%K@=*)yM79mQ
zJMTx7&5_k;g1?%(B~n|<Q)qA^2VtKOaB)bn{%<LrrsxyQ!2zHzy1<x$r(UpQ=1{=N
zOGnZv3n?5-Eb%VnS?RMGzjH&Jr%ue@C6C4?K=j0lmPU#OxGMjWb#R2%G_Mu_7^tNC
z^2K^!@OXHsglJMtUMpi0tn5F`oE>vqTgHCq46uFCC<~+h#am+iX{7q=N;#I+=drN~
zH~TYQzVlPJ%N15GH=2<`BsN7WqysDKjEsza9V~1)Q*JCFh41p_LPE71mm*@N0UX7^
z*xEjgm>A%;t|sdj?~Cq;7T5;YTANE~^+uaQHQhy6|41Z%-L)_L^&i&{!0OJ>Q~j`P
zVM!W^+Rgl`z3!w*^der<|9B)K5WMsrJq$^DU^Pg|^fX#b99G$})L})&_<py8dArN;
z!f$-?KvZ6qI7LzBU7LdUPfPa;K3?J`N@T+d66<K9WR#01l&mb4d=-!(NFz`72Ss$W
zQgR4Z>;tNhFH2^n2poEGx#hz4&;EI-<PgrY3DL@R5>J>XW|?%0#=+9C?_mSj{Q(??
zH(|02v~Hff*7`^y6XPi)?Gk|~cJ1xp5Ltjp8;8wE7$tDA{CUwx<y4i~)xpcWLdGCV
z-OViQoob%23_--2F*q0nC*k7o6Kbp-=b3$lL<C1@0GZHjH`f9sHMSd&A<<Tkq6^kW
zonDW?LNK|8YuWSf6#yOUjX*E|@@w|P2@Nk%M_Hk?xC3HTN>Hs<c`DiC<>37t9a*6k
zEA|l*$soO2oQH|4FnAXvYL65{Iyn<-T}<m}k3C_MiaauF!xGtpWG^e7VCiD}eQ%GZ
zqS5d9o^AFx;$I5mn%4u4wOa1xVXf2{^VqLsb9?(7rGv@YlW;?Dy=pzg^8SJ%)#@`d
z8(9J0N<I!!f$sJrE~BPsZS99W1a#30q|g8{<l`6##c-T}%!d3^>3Wx+Xlx(?I=RCl
zz<1u$X2DZv6Mw)q{nb@Hv9W-(dg&BtgKnzA8Hc^ZXQk>z(!~jtV>@|<i#84($K4{W
zj1+PnQ|bZw3YGoaIk0QfIceveQeHlQRkK<r8wMJ)c_$3x#F~J0cqx=T%pJcGp*zPK
zaiW=aP0CPCM;<Ps64~CPlN&&#O5;{4ukY?w1M2*_|K7=w>zC$i-S$MgTUmj#l(9oK
zEmoaQL&Zi)y~@}F@Sc<E#^Yg;|4z1j-t>g71V0fA`NixxYntoe8ZE@}=5cZExo7fO
zfH4K-IlJR{&sjvwmTBvFtBDVgGdz)LA3fcXbNdCRIj#)KodsJ=gf4D)w8}!BDiwUw
z>&q@DiZe(3qH^9ae2AGZ6N~rFu(i=>@`#pCB{%GC?J*OZA9nYJ-drpg*p}9Jv(I~e
zLV&icE>GAX%T-xfedTPwZO{Isp)H=!?809&Hk4NT5b*VCANP2kldsCiYXjm~{OIfE
z)_Y{2s}R3@+U8q5-aHG`W+^!!cXrXLEP?40fPr0Gg{847O{gT|omo|4Y+MPHBw2MZ
z`zAk%AT2`Qg6vIIOX-ArKdvlq-8ctr_!HJ<i1Lm0xT+Hb4K!$P>Aif!G_UDzXFcWl
zU^M_NR~bRP4A^|MgzF4RzP}U_8rW<29#F>6?q70rYdO$+l+mx?+qq*$8edTidp^xB
zkAhk^tir@LbCa}XP43Q`CxrZ$Pktxg#}ZmXjpQxN5Pu5zTQd}DE(kmylWQ)Vlv;}l
z*akkAcb+@#_}IOlvuoM0z7@Ukhy6(n853Y{nMvL`F$j!z!h2&?zMNR<kMl|<V*)5Z
z@i;VLgyfYp^fG?iv+)<x@qv;0?J*dn->K;}cKRZz4X%X{KO?7PAn{>$ZeKDJV3kv&
z2ORch`(Dv_HZ9nAAZV4P7VdG@*Vl6seIoU_#^e?q9UUE0TVyCL8nA!UVth$>7ZxK~
zviaRo8{2s$a5P!I@W0*)#E}@Al^OriPjP?5Tp%i&7yN7>G&VtMX}NbHU2>cAdmNtV
z))Ye9P5KV2$kJkSycx2;GVZY<`ih3W=9O&+QY-}&fUS{%3btbvME{(rM$rFA3QKVL
zUJgK^pkDaV@FqjiI|EU3RYZNoKmG?DixRx?2UYdXwY;(?1SX0_7wUB}QKKJn2b0ut
zw>9&;Gkw}~Esl|+IZ=Q+!<qmqIM&A9q2r^VVJ&Q1Q)|%XMJ<=#Q+;Qb^V7;Jf)D=W
z5;Z+zN_vLESUeA@92yDjX;UpN!Y@p`g0Df-nfKu`si~=VQImn+RlzFt%SDVpJpBE`
zSRZAzG^CxM{BcN6oW83ZWRF9I$%-kLr_o{?R>;~hqan7;f|Wml!Dp_-L#ZJQ0o!iq
zK<=jn=T|fX1K+WUiNA-y_=}Z)Ao+Z!nBPzeliDIBItRb<cM(Vt%atPL{M2O=2|$#v
z(s!*y+W${(B!b2TR5JLmL^WapQZOZ{kTD3fiGIY3K9g(4VgU$hP>Q6>2uL!*JW~{8
zf~5n~4i9OAbBL|S6}Kk#k%LKA!{zrb=j24;7ydq-zqpP49)21_4ZdI~MM-h}#Gup{
zEUP-sY13@exR_9D#ZA1a5Ea9Ns|Lu*G5553dk)@K2SnEhMv5TDSi30I^5{B_t*=9u
zJ0<s}c~1huoiV(>up4M9ek)YS4N?0=hVcq~B~6p0NG<|YV%q+cycnBRo3evWpF{s{
zwxj=@U~--&xyh*Nm#3PV!5Z*?X!lPh<L9N3q=~{dO{B*CXP{5F9CJIrHQc`$cnb?o
z8pJErW=bcXr>}<|36Mpdj(KcU3JWRuKFx(D-LgI3zxL==S<SfL`>N%+JQOG^L<KCW
zFtAy|t(1^?4;{^F2$FPg=K<d3J)Guwq#C#y4!J$<E02Qx3F$tLfRbHvvRuwk<A_BC
zZ3j8!$Y7c)RqM8b{`2~Z!R5j!)a!u|bmONL9b`YU>fM=lv^)IMYaY!i){H=D!d0xG
zO64yha!;s6^TPtmY9QQ?r!~9RH{c_|*2mYLqbw7}N?r%Yf%Ph7nW$Mgfc-}?FX%?V
zb5@e4+d<Q8b7m>y%-J7z)sie@&1UCnwUxpIr6}W;Z0+=K*9met9hGOv(DijUM@m^j
zB0jawV~+XU=9#;I8gPAoKA?c(OcB!xFJ5_tQ|?uou$6nD{Q`}p<k|qp_?$tAYP8Z}
zs9JsW?2%R{>P?xVl<bNps`nV>#esExshm|}tMXlN+}FoyuvEBySNM<h9!&>0j^2LJ
z+z{n&7F!3mEn+UXd$gQ^1ng})d-PAC1=%)P5NJ3Uq`N^aVKMOV<f5mqP~PjetJ%4E
z?P#LLQly$A&$Kh=3bRNqMXT#P13bj)3p(WrY-w98L6lfhSPk2cuSG7pxRP#<?R@e&
z?L(*xTxh+JWWL+ox;3~7NJYiF`%IDvc~sZ0+4FUX*fy_{lvFROnH)4TF;z~xOh7G|
zhc7w@Q8eE2pUBf8Lq|hLC^As`C$%s=>K8M;C{K*6=1WyFxfImMqPaBirKd7JZGa;p
z;te@Q=BChEa!8mDY8w{W^X@@<6;hLPd+(YdCz6atB$)(AgDOU_uaMPcu2(aJpO~Ck
zo@vS@lJu6olt|m>w_^BjePCc9K8fvi!hj9I_f!ds`_b3Sz@F_9lmWY<S_0eb<=QBn
zwWT`hWhFE$?W8dy;7`ioHpgc0U58w|HfVi~28Aj{p+dRUNJ_YHYQfZj6MpHa$u_3l
zvtgmnC58u?By0V9XwXdhT%Ia!6xM$`Y`YK<-U2o1s6lLsVMmDRp+vs^iEEoq+C7eM
zvU3<UtDGkg{0Vht1ljU;Bk4(ZVkAqgw~}?!%=R^9eN4{S3dxf<U`xl^dpAQ3GfzNb
z$vM8!iGnf?)8;zA_ja5qxeT`=MV>;vhe}=4-oR_AB1ORCIM+g33Q3um>+pu~_OZ__
zvQfR1<%;a7|1}$t)Eg^IT#qOBqf2NheC)bJjn4(Q;~*aOy9MfD@;%p#e9>qe3^}Uf
zMnMS-3|kdu=E4N>1*}*JgQ)4uEmQf%D!QIWZhBP{H7rik0+}(=lmDT`MWI#7y7lOQ
z!WfD;9%Et`>97jM@VfWoS-yt%57EMH+@bi)zuw$}eIf!vH#r|{D1X@uV(g6RGEkdK
zTGEaxWQ^S@fPUu2C;R|7l1|#~TeY56K#CU6$(b9n3@X;!Wc3sjbK`kK^0g(rf6&hz
z%Ddeh<Awmb{{EsuI{MZ7v)lCDg6q=-<v&cKoliP1f-teC4+`lPXg>_8Q>f5RB^?0N
z`X?M8?UPWy2vy~}U}*Kzs@B=={wYdVK~1S7nC!?v0<!c@v7>xi2uj7Pq#*u;9N}NF
zK+kZYrLRDVytyXE@NopCqV~Mtu!HQh?S(c45EV{G>asD35OYsducs)vhr;Wdu_*D~
z$NPGDQgAf<adFmh$^D2O(~@{SR_Jp}dxGl0LCg$tNb<~M`RfrRz2V$@Q)<GMk&BJZ
zb*}E)B>qp8)1-{^viakGos8We`m;8;#tu63AxYD((w_C$FF8l&bh#pIfuKX%?oo1+
z$M%aKS2IY~oLW{FCe^no<#<HY)k^Cl_GhlX09a)hIIqo<N1sS;03(H1;FC0=49P1?
znC@dn^k(HE;K$7#e&-hRB1zGz>zjB|Qr(heb>->9I!`FWK<9rC$V|{FZ6%y6XJP;a
z9!#{(dTqHX(ipY*Ek8q*5iC2Xd%DZX!stM6zPsOF)|g@Qy|&*UE~&~@O}LW6+tNyV
z;J)5iDe|c;YU;1sqy@zGUK{2LzXZ*~`mpolo7ucr5Fj_4lZqCnUJjvIpuE6$6Rf3<
zae3~Hx>n_F$vOS0shTS<OB`Or*%N<|9i0B`tZJ$Wi`c$SjDJ=&5mP?AC!>DOTs*R1
zSoD)27n2GNJRS_3IQk5=WPG)tAE`N0L>XE;FdJ+qZEb&Z9k>rwngT|jjGy|{{v>8h
zzLX}T<c_F=C?{*Js)D_yFEdUCcB@kKSyG1YK+46A0}Su;m(>=h$y#v%tr;jHjjckW
z7Hnn0-JZ=enc#Bz_wpydR3aDGH`6zE<xL3>YmHfX9oimvuOf8#ABN)#PCoVVpYX&a
zt^CeCh)qJo$)C|<omu`9y4(_~N6+v<K3tttyR24qRBm49(6gDqP6E7cwRBbk?vQEJ
zu)dvQoMtx<?Ce)<?uSmGeg16U=Q5;KIPwBz4N&hvSLDuGz5Y%Tq8M&1`lO(`@yZVq
ze&Alw4SR1-gTU2sIaqJ8V~D2ajohechkyzP9ISU=2;%7TBu=KRO?MGs6IvMQw{b+`
z2}|g29zKYvMW!J}s~w1`S5lUg8SNWHr>emwS?9mw*%`s^m;QGzz(sKPoA+%-F?s%X
z%`j%|V0)d60{w!+-y(d2&@+1{zcZQQH2ie`YmK6Fvw`p<0eWQK2_zBu<RJtPJ(Szc
zMwJ5GBK*yJ(ZA$W+DH~(k-*;Yoj4l*jHKg6CI0AUr~Ex9(D)Up!Zg4$Dg0jG%|($Y
zSd;K;7K)+u{j|w*ns$Fl1l{#vgHBJ-kLVMUtS*|OM{o_E5Z)pYkT$b1RGb@>Kx4%j
zoi<g2!owOt+ZE}u9jUDJi%`kp!}HXPdQhWH4z*n0ba326%#CkU+IrYV<9B{3<k2xt
z+<)`6|3aiD{<gR@2*lQ*ngL`cKIFKG)W5mlWbq1eXmvWC;Ch8lfU`UU2)%;HUy~!W
zNEJ(!$<yloiPFi*exksYN8h@qpB#iGO`>$+Bd95~<x5+7e(x<&HLKigboR?ySP&Jm
z2XilG#k6Lc(G|BW?;EiBWjO^7ZU3c|U=TNm0??>JU?0j$pb=ea`|A#wQNA^mi0{{^
z?}%XxYBaBLuSL}LCX|gr0`h2tuX;%zobZ%l^2vy<(~VU^deZ5C)a<+_yM{u-Qd5%%
zryRr$|I^~pg^1|Iqsgav)dE{$JU&zU@T*^6CP)G$I2G>KM+&$?cuzLS)&pP1u>Y)B
z@bCGgu~~LWtq(CL11RsY;I)B2brZwjm_BIBS0u})+-)iB(HuANTjAKtZbfSFOA>_2
z{yM{O<eASO8=1zYpQQWEoi66d7R-np;m>%&cN9UA`|UwPRpgpnlPTiR2pm((F~Y<p
z^0vkL9Fcz`jh}{KjU}%L(zCBIKUJaI`!<>}Brv1L9mTD%T;mYQqkv#G1~G_k43VIY
zQsk}W;h-28kk)uJd`tGpV^W33V=AN|+dFQD(j{g{QK6$Ij-&T&Y|l6cL?vqd&R(!O
zs+~)vpg*{NPgoi9vnsXEg1Eif7|dl0Y4w=Zi({YX@PEljQq=(^+v!k4Jyq07Jc+#X
zp8J8Ye>v(O4!zbkEOPYL!^<Hf$2eD+{qzRo679`N?ZY<iK2$p3=}}#?>&Ggpma#ef
z3)dBa9U8jTHwjhYe2|ESh11=`@g&0z<t(^7TmYanc`{k?M=+@n<eq0`f8>%;sU|R*
z5^}AtA;6%_E#C<y`qP`*e*aKZ$)$QbTz3Rp??^iCm@xXvg87yQ%fGX`cRr1~9nhW`
ztHLkG&-g#Cv-w-jOxU~Dq1~L%{IxKCRhW0s)-I?hf;b7DkAkkhr7}VNq2pu9E9*@L
zG8S`cH`2U{JX(UD)suEqMz3tI{jWnu2F5BLkprz>^}}l>_rTkAcXvGt5BGaF^Ln5-
z)NT0-%0alV+5q9FGEAs`D%YTgPzmA0@sjL<k+qlK>d}*|KN*suWr*wYO+H-Mh}J(S
zhV~R$gq?^jBlP}B8?ao38(^$qKgFIUx2>g@T$2~TH_r;1DHy=}Jbr&#nCJdLd{Pnf
zMWtg*&T}kPU^Ven1BuFO4V|;qC8q_i*N`X8%F2%*cFp6Ot4S^PdwJ(I=c>6Y@SA&@
z&k-2qIN;QpY<ug`O5pVb-bwV8Gk0O9%Yj+00b`XQIzp_j!vBi6J|dp;YOYrv*R-Mk
ztv?}jY%DQKN}QSt2%w?)lPt+V2J)i>F0I<t);EpbP9T7NSGi%ZqieyKg2;Koezia-
zCgZ%sHaacytmdblCW?)d{2jdZbtC;pgY+p*k!JBmZQv4FU48SQyg_lA>SkJ*^cGS~
z?wrAw<z+UkMGGb@B~0L~vX%?R^zx-x+?j%G4MQRyn$>L0umcHFMbOp|M(Rfzv5eCC
z1*}2LfkT(dHSQs0L}n>-d^Zg7A+cJri2tCgJk&Q%&3C^SouZ8b=J$|vT2+Z{)Eb8o
zH6BT7-(p@EC2~;_;fUVyFn{pP#{0+LT1{PC({RN^(xSA(QM8zFmaadF;nf(qlDc_k
z<Nj49?cL*_7yr24V5L<Vo4ff&J2hv`p)`cdme>2;4fiUmeAH6>0*6ZuFXKHzMcPF0
z$yfUFomPwqn?`P#S*+Q~gkYs=#IImEQokQVLFCzghC0dxKP;P`k=vR-k=o{4#z@3J
zZ=)GAW%$<-SGzg<$(~D8sdGrOK)!hDJLJ7Ao+SV#Fy6;dgFLXETfZ4j_iR%{xiG2x
zce@(g$AIfM?QC5OnEBGR`38+94@5`y#y>fDMe-G^VkR!xaU;6PB|^tpoc#7qvhQZ=
zBbPt+eJfTSviPnhAAa+pR#o^Y?5{X-ji^_Zv)jvzP}|G?iQm~hTYas6gzm{Ocj-Hv
zA7(g!a-KJ?^g~&nmSd`7>Gw+>tax2T$)um)`gy{bc?XB8`p(Y5fRf=_&3H~Ba3n8s
zPFWEmb#Zr8!bQ6N_;?4Q1wgqQB~0Gx7H#bN_!2pP#sZo1Ww%(W3kE$qb$3QdVxbKt
zDo?UXVpCNt1@C2gfI9Lx6E;UtBLW3Wk3NnmNIcV}#`YRqCg3M_2=$B%+g<7O{0Co$
zYi?*R#NLCv?qju^*23$3mPvX;;4|OreNlV!0iK3fM<$+6z|+R-*im_PQQX$<kDyi5
zRbTtrMm_#VtxVT*UGoFLgXLSIJYd$`W{yp*7aUSue&pn8&n;B1<3(bWb4K-8;UjLj
zw)W)u5|@Vq5PIz(BNlD^aqc~~Palb`!>Y0}>&EKwaAD}Pi4O<+V{!JbjIcajcD=HA
ztMv{%I5w(|rS}>29ovo~*;%E4|G2WqwHx2eo)Z+DLgB;*Q8n3BIdn`V9fyI=+QPfh
zaoztm*Pm&#)2GbjMPipghyCeA)I(!HpStL!CMQ)&+*h}-tKeTdUv*}DA8m?Jg2A5Z
z12=RdGdA8rSP|OeOV)>eM`0E1;eO7H(DnEMNJbNGiwD29toJBRPMm>p2S84CfMdkb
z)ypsHnfy?vQ>&kCzUIxP#gx(Fze*z2c>zAUdw$5O7Vr->*5}yjd>R#d<LLH#)dWYs
zX!(fr2}BKEn+rnu+_2)MV4xEXMx8ZLH`aDko@$S{Cc1T<4Dhsy)v}jbq(yb}nD+z=
z-$)GPc}k>e*^be>oS9AYi}b;%og&C0aN7_R^<QlQlE_6^NHiHO2h>bzqCG2%9kE!R
ze!r;^g1h5h(T%B>{H`eHu?O|OJ#lP5pn~ii4fPnOCn%_A)L2x0Cbo~G9Y9*T@be!1
zJO17r5QIw&Wn)l$P^8T@*Y2bLn4%baGw!ilr?y=E>*BZ8vFXPdANo#ZuPIA6w@@pv
zg4YyimhA80FJG)zeuwWX)W2j08|)U&_k%KdGIUjbNJqG^bMf8RTm8+-{qWwrt%Dl5
zw}*dOa@+ey|K7*%j|MjIJ`UdI6{^ayl|*#$)Dk%fL84QWG$*q}VWKrUEm`#j|M9(;
z4Xl6kpqzHsL#Cv!e@d`ETHE;Lb@{M_I!!(h^<nrlwZA^!7xM(-mvQ_TF1AG`EJvcC
z1}I^}skuirIM8}I_4Vip*VzBrz;ExKjq8UL2HvD*1uGj}3fVRRMIvD-Kp|hHZ+iws
zQo*SjG5{N4o?xZv&cJx<*u*%6p1S&;@t!$FurFwp?5Ahy(IWX(a1i*CCpHd0RHjSs
zAPWA0x8urb<kBr4m?s?z<MsFe9-toF+r0Usrt)-$qusT(4d<2lILvHBOm}1KA-I}G
zfTkqgS^gtD8LP0rk5;zqoTZNK-@4Y*#ZFbIpfslxHRVq(H{$!g`cr2IY~Ix+m+c@z
z9i?|Qiknmf6^>sYWBK0RI%-c$^Z`mZ!6dr^2t$&C_<H$IWAFY0EnC|iaROQrBZ`eG
zvDnD7_2k7vd38N_RmE>D#!?qO5~$Py0PanhHSlH&z0$@i=9CUF=dwk9ms;Rm_~ZJF
za%PM_$);JgfD~&1EbkYIWq=dAOpC4@I1EGKv1(Hz*8N%Jt%2B$y@P@^Ky9_GZ{6U9
zs4qM~SkEgp7-xObGmTX9A)sq}hO<{Y>jkP`0GdT`9Oz!xEsSW?1x6O~BR>svo0vBi
z)bKV&G5oKhrz1gv_~jqe7zgD|-?8<;)Ol;CsmB>Uc8%>Vyl84908&NFWti+hNDd$C
z)WeWtDtev_B0~StA03+S-F>M{=Z3`)5?A{j!PQ5Dxf73uIR1dY1Y(yvc?TzuJ3-$w
zyh8e4v@BC+(rp1y6!f)MjM>>)=E|y&WJCVI;G1S+P)ny8KS2=ATrrVW@o};RG6j#b
zVaGt@cwqma=wHm-fWJ~&<utqe1X8rJOi=-GE>`7TOGQ25bhFH}--xG<T?9B%rge~h
z)c{N@C`wg0isdTF)kT&5iGKRq35%9La(kMyWo>&#=zLIndEZMv&_^_I%P>4fk-mh1
z{^l%DhHIAxHmWjf&jmO2=DY1BcaRZe{l@s-{ANVVIe1IX=puK&A1N|Y6I;H$IF_6!
znW!{+k21|Bd4k(dn48lW;@w{|4YX~WGxYtrVQX02VBb;%3&+8E6-ODRq}zCZC|8na
zX|mGlWJ9>Sb3#R&Q}S%~atvONl@jirmO!jBVV21Q$i%hTMWm6BV3*rNFiDgQZLOnw
z>4m!5@gd-13_D(fH3jM1&}&1iV8+e;_i^#=qJ0LI)xx|08kg-0t(#+?fXI(A!)nG7
zwK59_H=pGhJ#i|a5Z6k_-nEf|TIZWmC}6vqJ#lFS$`>_VT3IdDM<I}@V*M)S7MQ{W
zCHOfXa>i$LmZGCt9g+7`D#m3ic2Q(6!_n;HKGWoFkicA@z_ti8a_zFvuFsq{FKa7i
ztxxnyLmo{f8nvzFJc;65%)K_Ul8ziZc|x%6V5`1~_Gq2p_k8r}-SybuJa`PgP?WFU
z)lGlr;9Ly57ke0dT{`mJTT@E%+%PN5VIWyLEe1OKZpJB8vXgYeJl9V4M`9b=*5q4h
zUeUzT<K8n`%b9OTS)_|83%ap7PK?lutR(F^x@Nm2R~2fzon=7-54YS)4GH_qRK*Sh
zBw8J^!H9IZg8RO+N)RLUG(RDIdv{{0P^`Lbx4Yi>darwgRh@HohC&+bz4QhYwb+w>
zsE4g_tL#(QfM&PFz8ZN=?Z2e=@-#LMBNK~2uXuf$d@UB2>pq2-ZIh=fzpZ~vOzqvW
zrrC4vL_aEG+1?Gl?lv#4J6)Bk1%#!f$Y_;LEX$D|+_g((j?>yVR{>)c6jNpV*?)L8
zPkw$g-i%Z_yPn*VG&N`XJpm`{{_ub7n}`zmJ@-|-n+P~COrgvusqF5qPgSIVnm9O&
zVa7ZqkD5`T^*;ANqk8d;0>rs;Q{uMT&&bKy>A7!Sc7|ddBboT>=2*=Qya@wDnDI*0
z{W}@07GTN><tCaR8-f4R*?+gyWrX0dlR*g_P{UUU+MXQ$dbbl_+gTsIaz<R_MXN?7
zTM#CW3}0L1eM7rt^OIWX8?fGtIgKs5P#=?EUFye2Mi6I-8gZc^`W;i?nOMOxaZO9@
z<cl#r$^T>OEWDyx+rNKQM7l&eMY_8|x<R_TyK@*2>6Vs8U<m0LX+~P<ZW!ri=o$vz
zIqUbFwcbBr&#Zm#>;A^)Vhk%P|6mJMC|Nqay4!;w<euhW!LiC<S+G%=Q4No%Xy4<i
zm`Z%psJa?$q9w4YFC$m9)BAv>W!@jndW<^w;C68Z3YuHq;VpbZ8A_^o50W~X)#f0d
zHALC{8_N|pITdE<*$8`+BSQPS;hoOW$n4=vnoS&vm8xWXsZ&HlfLW}Jg2s7U;hkak
z#hj##e`vB+y%;lTrdsc$(c7pE1M>hb?Fw#zK?4=SofCh||IMS4KUzDA+)TP?(3c6^
zn{61h8Ut=5k|ez1$Gz*`j|RwNgv(n+(tO-N|FeH1fT&SpNfb`!Low4b?Y(@b&skvY
zy+zbC6K!Z@^4=!4ya1Tm7mee2G}vO;0hXs2)L~t3#Ue(-wyZ`6_h@r>6I>~91IDPN
zk7--gs`*O-T(ZAd5Dp}~=jG_kPz6PaZ0;<WBWq%KQTT<=kBq|JyZZq*tluTlQcW;e
zKWMQ;=ql}2HlSG(gj<M?8}SR=>tnKj4R^Ej@u#WlR8*_gBT(qO8c;Rp!a!}d1p%9V
z{YKt97q5pih`!mx2{&-3Y*`?6-{Srcp06gC@uI_F9t1Fgtq0e@o1*)9q%%5uAOw_|
zcDz&BGxvXxcwS0YTh1l@an*YtKkhYVC-NEreS$oHpAqEEP!Cj@0Eum4y>15|P5Wp8
zfz#c7;x!iM)XzJRZ8kyiTqD;zVW;<_tq}?h9U(`cSu1wqB^rej9p3Rr4aw^SWU=TH
zX);vR1@|40+b}ICC-G1vREcB3Vb6I-&h90^<oYF_b;&L<K}Sy24!;kE!wy?zDzW|@
z2g2FrV_I6Lp6I;IGGCj5{w$+4dtV*747T)}dsW8kl{HFL$4BSg_BO~2Rp<_{ww>tt
zA+tZXHy2Jx7{}#6#CO0NB}6d8M(kpCkHNKW+e}boT&Lg{D6293sAKV*jXyU{(B|!I
zqkynkZKKqD_U778CdcyeDKpQ!8iADW{uv!l)$AGtsE=;+6uyH|^cqb#eX!1M*c&b8
z(u;k+m2vo<8S?hdk+I78fHdBagoQ}XRsnqxnRm}=p-+H=2M!be%n`l5Kia1VOOe4Y
z-}9uITY$Mn<<j$G9YVJp_5|1J$gc*ger<~XW2BYn)$H-}WehcvFL6+AElMhyGgEFt
zUQk{b<Tu?Ne4RS8FrjEoKnX0X%hj6hN@67!S=@fQWvXdImi-!_pRAK<MDO3C*?@<q
z$|d@jcUENmZ+;mis19$D(0$S>{)%;D>;myF7?`=;475FCMcbj(&HjWLgpe4U#{p?&
zza7%J*e{{(b?QkxS`ZGpLNhid_YtAo6k@>Hisg^tjzVU3@;>}y(<p<Tg1_IuA493{
zUjN3HYw$UF^PBe`^aGk$IwMn7Xu{S|Qn7D@T!{`(lbtL<1xD(2O!$%t4f7a?P+r;U
zRY>LN=CGyMgvPP_;FHgvMgk4y^ovwOp(=C*+Lt}U(-<1!=axRzZ6RAcb2LR-T1tO^
z-$iS3RjL3!_E6dM4SFVU>@WP^b^7nC@D2{b%0X=#Lt|X#Ul?+ZzL6ywtI0U37Y_VM
zT!|@2KSkQ2<l71~1bYpZpz=j&I3Zw%6%wB)UCm|EdPWpR{i$jkKM6}DGvfR1{#0Le
zH5);cOjP(+rfIbPJM=#Ujw&R^@39;3WknpL8)&Jtv!v0srDAKjBk*)s-hH8=5r`6;
z=o+tb?;IwE-0}PQ`)d|RC!R>*U`3@Nlq)C0g(YcGqyy0mr61m9>L+QjWDJfRs2Vm@
zFIg0)@D6g3tbl1gsHj8-J~r4Jn2Gd`pHi)vadHd%^s|a#iBN1|FR2q<to=_rc=HNr
zKsi#<qxNeF&p?y|QHF<ZtJu~;qaLQ@tb9`WKcr@h$XZR38#tJENjW=X(i2ML+znSH
zaCD9?JfGpl%FC>a-O3jm$G;Ob?l^8S3f_EMxZ5`olCIUJ3jBdH&CnE8kYm$BmNrUm
zgoE`(sX~UC5w=sl^wyx5sno1$6ChzpczIF9>FIq!Wd1m<kFKDpmEU7jEnwivpi_~p
zlmAI^hfeqFY+FX$Qrl}}B;S(cdmq=IPh%v6)F_#6c%A2sQ2RJE`Fct4nDq2;F8kk3
z)VDS(x2(`%&|S5<d;`yC--{Z!3+(^@ZLX`TdF}$@zkql3PI-syUwdD@_@PrQ*7AGf
z&FRkaG8IzN<V%VCT$-NgZZUf>-R6h+yfKvRwqET2B$wF^FhOI*YA1lEnRMR~-R8AL
z=Y!RVUM0&!Yk0zz-f%PdJugWXuRGMY%t$NbEXpc<vhP`zy=1HxQu=+L?#tNpE@5il
zqkni>%y>n#dD+zru=URF&^s8uQg?YglbrO5Ck+4dlmRymjjfYd!gcDElXJx!dB5f;
zcw~WV?Zrqmu4Zl_W$s%>!|I^XE$Z3$lbag@XzL#xYf{MeTa*_HNqO7njD<UoD^NSW
zIJ_|BAEq`H&~Va@YSMGu_Z7J!ZK_Jlqu1Db2nRa$BmRIrwU)X<4!~CaLS8p-av8`>
z;!4rz=h1M+(YRM#_XRSF1ogZYL0JB9jA^d&kLAe`$yYJtd|g>~X#RWyDQMQu%*x++
zTIGB?E@gkE-(=5YM1G*3FL6*VQt$Cq3;Nza%eJmdeP}M4QY6nwhAO0gZsppjaxHcW
zY8(W=4PUnSNPmOJ>UQ5^Y!)OwhMu)doW0klIWZT)R#&RWLpSYh;kD!8l6R^-F84Bb
zsv{w)mj9!>;M`fP>8%L;c&5{^L?6*~1RIO?rB2nmR_~R-P)lKPf0dymQJ}M6Z5d}N
zps<=kg4gj4I9o}S6aA0V4h1IOpjPIsfuulJ<+^QTYK_>3rW`?k92p@{p%<~quB}-U
zqwNE5+Kvz4&#jP(=zhGh=l8w$((SwZi-6r%t-sZBE=^i$3my-A?p--pgm1r1?kHpi
z?O6t%uOQv0lGhzRbl~spo<r>7S>m&n$ES*VmqdFm${e=CkH|q8v-)}09dV)rPf<A4
z5cSG&$iJ7UG&!QzRR%9#FUhq`j5r0)OjNG=8nS?vJ0?P(CWL^TU008#;=c{G)NHFY
zecW6mC#SF9Ky%F=YV8hXNj548y7w5H%;%HROx}vjT#9{79sFZG3I)7mO=F!*8uJpz
zLJO;t@#8Ek@j>YJuRw;dDAhUVgxLT-Km7vtM;$rLL*gwVD&8$B?hu*swvqL9j>7xV
z<jeBV-7Fyhtgo8NNXf2kkTKLGkgbh+Xy6pr$M=oDRO1+o!!ML;OSeW($nTg-+Q4nK
zb8!8MsIKdV0RR|lIB;*|bTnO%)p3sOEvlEcOhFTCR<E`tvfHcBP@|2oScElod%mfm
zONhn)GW#G)9)zG0bThHH_6{E#t7)jLxA5Iz@@|UnH!ATxw)sWnaL`EFz7$9NADxOC
zxH_k5+6<1(%a#vhVJ6N*Pq$!uY_a&Et1bo?BA#z)fQNf#GNZhdf2lmpwa}sFu|+@p
zzCR$2cZitO_d#|Y3s>x(sVgLUuls>s;{1G~(GP*6%wn~)eH;4BiX5#Ai{@NBpB{75
z{$cD1L;RLP=#5C#%!PmCKo98F?lB{Yq+`h-AR3t}iC$b=+L_+*H~40{nJ74Z-jFPw
z_-@PMhDLvzMn>iLpP76CeE05qvME;lFzUZ>`on8fTu#pVj+VX=&d_*`(1#fv7J4VJ
zh?Y`RA1xdX=x>mpe>XFuO@mUShgoBd{CG?$@=_$7$)=`Mff?e8l|1Yd=##9Ancl~3
z`V-gx7q5N%y;=LEePjahBAK0&Ab|cQr=lGsLd`C(S~iA2F^oI&fKEZ^YSy|s^x&vp
z<h!&uW+Te2xM!SwVOw+~@z(c!Tm>B!T^xSGp@)uy8)LD@-=wvL&t&d-fJiv3BN+)>
z*gF%qkJ;o9bYK5_*YI}L9}28?Keih5o+kL(Tl7f7O*S93Xi%+a*NdWr-UQ?GcTN0d
zpwhHv+0H$1xXBZCwzR%tL5M3Hx*?J*c}_c=!9$eTKvS(6@n!zQQvH_X1k#5(Mkvyv
zN6Xygx^5awQb(C|21qoR9^5$2mNRBoAwshB%|o*j$FRlf1}oEUryNvEhL%OG+4`h%
z=LYPV;IaI}VIdsMnhY`4YC=2_DLwliq;?RSXN(19S{RGZU3J1R&Id>3o2gx-QEV|?
zxx;vP+7NQtdw;B1*jxa7upaj!<}@v6j;339-}jy47%+2qe`!*qGks#Ncp}`5xM<1S
zDF8moq}`kYR?~gyrVoOT+J2JP=s95*mH1$E&3LoFYHYr1Ln@?eE+Fk=?HNq<Li-tO
zGwfAkK`iS%WR+g3Bl4Oso)PA`o}65EwIb@?{r8G8PJ0*}a)s>7q%-gAimE>d8;yKd
zxWsuTH3IHWGpr%5?&gh<P=RqbWpvvr`QQG!cmZ|peks(#tI>TO_uCNTm=2(otAnbV
zD2$U*4TQe&Tl{nde$b!?7A|)W+L$b*v^^$Ym&*vr+R9S18xj2+gLUF4o75eQ+mYtq
zzzYQ>S=~?}Y7-R7G*$<M$IXu3w|FPHFtjATsil=fBPe^E&tlRaW-ACC2!yVvfcRTD
zI<9v5G|GIAHJ(Q~g@BNAa)^kdd0P;jskvJMQFh{lhM0}{F7jY(s9Rq3s>*%5gx(RS
z-3(8M@;FYgxAFa;HI2qLxmW%6UC{sMnrFX$B~T;p`|YKl;6g13V!YfA!Bt2qgV3)5
z^>?{2X1KME9=RLFA)$mwucPC9V9&r*Tx#n)8efV>yIhsW03+p7C|a06PV08l#df&t
z3VE}6$LDwWyhCLH57x=vk59VQXj0@>)@K>-abDWl*#{Ouw+C7{&VEfi6OJ&hTlu*~
zt>ms$bBuXi_6DAc0&zo>>YEmdXHMP@;lJ_ws8#ysB}0<-8tK+8F3!CFSB-$~&Hg`I
zvsK#FY@A&Cr=B<KfHtoaj1o;;@!4ZHSYyz(qI6vj37r}N)=MJ(w&&7Rq_!0dI<vPk
zFf#gZR&{&=Yi#h0u9OeoDAE9#$f3oQe9O<LB)fK`3i>O<^v>gO+fcory>P=LW-kyc
z>J>7@dDd^iBO1tc$pq>*QlmKfFrv@MFhzYftngo!2m?9FRU$atpXMRE-Ho8m&R@SA
zxI(lTe2=pj#3Jis+su6Ja%Lo-m^n_}z5PJF*3=RzV0?=!8l~Dv5%kVYp$38MoJ|Sr
zg^9ZJWtim95-($0@BvFW8W|WPeuj6oH6Zq_J&ZmHRaIk*M^bd)R6R{-o++c|;cGy<
z6u~`}?5vgQP}!A>)f0S_SL1wA1{bN|y4u;7H?+%i>Fur0*jHo#A5LPx`jEkf1C6&D
zoF!=&Ki(!%rT$UPCoGDjjk9`hT`ywg0O&-_A7Q<rlrN!56rDlB5iHqdHT!o)vJ6cO
zEGAF5A)loF#OywiadtiO<f})otx*P}OWqu4s5|x8HkmY$HopPN-8labMcC70dhSij
zoo_4{um=qe+gNt>p?E7NVrhw+N|4_FO1=x{<u{3-m}JDLLcn3)?FO8P8Ot+N@w2qC
z3?JF|5|a2H*4HF&&!@DTW*2I$Q1QnQzB;Q&F+D#lHoJCS_;$I92}|<ke&T#TTB85r
zg2L8Z^G5uoIc+IuS+QE9l#_S4UG{;&HWi%O8io1zYv}UtaE9}pSuD+2c46U*zfW-L
z{!8Gd;!KxE(hc<^W^s#oFARL%{L28_pcD^{{%uYE*FP<W;d9*c{l_jfo?C;q58)hR
zcG2<39>&eG+D$GTtM4l&G3Vt@o!HRxjm8tRdm$iPnPzdhb$Izi#YZH!-;><>4N0!S
zz=!6N4@qf%xv5t2Nb&>bY+rT$cMV9EICUx;w?Iz-0S+)04t4b;E=gxyJ{l(252Y(N
zYD8Mt@R{-1bVIE?$IJNyp#0oSuex?6GTNBsEhA7iH)2x1-o>k6)j~JR)z#090scaM
z`$C+M3)C4fFscuyJh^aV^3al_^M*+DXeIcrGP$v%Yhd_-MRY~9U&JcNFLUWWu=Mka
zznk~TA!T(~5#is!mz3nyKha(o85cYB=Zmh0A-0FmqH~ODWGkGoI_f(+hrz51v1+3`
zQ=f-!+_{51qn|QbDuLU)zuQU_x4t*Eqp9%3+hfi5?+fi7UXL^b)M}xfgMZ%QUV`d^
zv!4;qce?6%LIbaG#Xg4OO*=7$d6ds6(41#X^gh09;14LkoiDCY8u{v&IG-srZQyOe
z>RHs(B#E~YKm5p-;T8KP<f(5QJ|y=;xMFT~gy|GCw7JwSY8>FrKIfbV5C&fwh<i>O
zp7sD>`OPps6GVhgCCBEjYs<t$e}m`o>2u{w<yh}4d#$wvRo7v|(?QHoE=OtswesYd
zWVM_h+fF=(bIr}8k(Wq%A2bAktect%&J~J3cUI{>@=B(;jjShixW{`prTLW{9`eig
z$)<6Wb%j<X!Kp-5YPNYz#I`%1r_3LBl;C6S9k9I+xFEyH>3OmUFq~^nG*4VY((z40
zZn3+_m=XYdcpAL~aGhTFr;pP)I=I%0U7PT2nTezhMdmr+>%8`PJ4qCh0*)FQ9qo-$
zl5;yJryURDEfeKrQ<%hB;$!pkNlhjc&!T3<)h<O=4k)Kxcy>v9r`OKv(2z2*zki}7
zPp<+=0cdGclz;ME|M=(Xd9W#X0q470cE?0UZ0Az|gYU8hFRJBleDlZ^n&8$U|Bz+D
z%1*Q|L>(e=&Lj$#h~Kn0EPj^nql8Eb79Q@0lpidobUq|D`&fpIO8D`0MY}<jRf6wB
z#)(+Gf2)3D#m;ivQ>mmBN2<>fA^nuO$guDzW?d;+Pr2mgvUK+0XeqmxKIguza-3XS
z!@(m6^I(r>&)I(gtQUAx)67;z7&asL+aavn`P5%O+{gT~eLqC<e%0cylMxitUmu>)
zPd$cdc{_J`cgafe^m_~MPKiI9N!k!oy4dHpNoca}8t4Q6E9Xzrf0-j=u}k!_eqEvs
zztH6S)lU{2`zdw#m{8FD<l@SOa_x3WA}DU-Mg*|nHofZH=;58xWPfUXIWEvrx4wD)
zh8iRwI5NL0pdoo_c{tM)eBwNDxm*P;o;&<|Pes4udH0*Ptn5(MGKMg9!vlFb>Q~Mk
z2H(s`Dz)*cwRDWk>SPzh90;r!V3(NnJ50*Vr^eT-5_g;v9U5~VRu)>c`cUBFa*y{+
z8uTxk^`j}3&)B)15Os7IW4)2tT=Y0O#CBWj9$PnyH>-=D7nvGNtdjoLuOy7Vildq|
z7S@tqpp*Pt5Bv|qPX;-;*#BuK;gKD=mE`&^c*>v1NZmpaE2@-6zeE8<E><I!NC7j|
zY0<J>P*y3jfymrI2c+tPg+isQ+%5{3)gw9!|EEr?c;$5RYy|1LkL9wTViHnwu+}Qc
zoP?FP`EFp(`@`zF9i~qBvh&U9&xSki4|9<hyHSnDQH@Jv=sl@HqFh6)Ul0b2!6b%C
zldF_)t05l=xFxW%VysB(!1n`Lh|(nS-WtS4=KT!}s!I~>kXW5K=-=JzT%Q&39Kec>
zbCGPuqN14XIlwr<>`5>k1}Di2NMA(W7ZbuoF|D<?oEPdEp{ut3?$H2PS2Fw*@KpoZ
z$1rIPcu_-V7JM#86}XM>1~s%tj3X<l?~|;JPKXA{ltBcDr_Y)~#Y;|_!-U>~41e~=
zQ2pQhc?&&3um3;94%1mlBr>%kZ}74xq1Qkvv~D27y?eBeRiz@Eels8<##%?IseR==
z7f@Q$cGk^0`c<cV`0_M`meOXUEN&E5i(e%NyVj3+hgx@DUVR1S5rqP=I4bPQT&mF1
zm<i(3Rg0Z%pieFN@|rkmmqf=(?%g$Wy16Z4o(ysUd3@FVG{AZB3ppj-%`%r_3pWOR
zJ7{}N6}qXEu4U5Umw0{cdw5OeKkwGi+}w+tR^teP${{gzMGFTPyGQQlquPbr<AYQu
zC;ZzM@iQ5^J{0fiK3$UyVkjesHk={Ji84Y6zHcv3CBIAf%0gx~4w1+@m%1jdre8yn
zkB+;<$y}0>6DrV<l$C`#-hKYS-O8bTUqCgqO@_kwq3<hr%omv^GjK+N9o$+bGDh^X
z3r#=1rjq}XLL@rwEm^^k!v5|iqyg_#``rsh*_lAePo#UWgfyYnpRY&S9CxX!%G1Xn
zZD}E=iNvQuBiYKz(BR|P@-<hHboD+MSX906YFZ`8^(3={HyZ<$MP}pYPZQ;umO}MP
z<AZ73A6Xpal4nZg?p&jv{@K&6N|mqKV7oo6Iv@lRwrQU%9R#P8l-ca&ag28OF}~5h
znZo@wy}j6Vt6i6VM(x9V>qQUzCWL5KX{ZJ6U^)U;#2M{q!SV-epp#v*86=%GgKsRf
zVOykDGShJ9q!ys&>od4nb~LVDSD;*%p>PUxnC^b=QOMg-&sRQJY!hB6kg=_;3>v(r
zwD{T})#}F%VtoF48#-Zlx)5@*z<44!-gR=-w(zSk0mskDE;9(+JHGz&@~LL!KzPYr
za1Ga%<p*0`yH4JoNrzkh30vdUen%8riff2q+}w%xeizJ6IBmz;`{=h;ZD8~>+m2KI
zt@p})wZ!x1{I#~c9fc^rt8^7pbqzrVD7bKec9mY${qxj$$CuA=UO9u)?QK-^&UC%F
zMd-TM0}&as?{j4`XsgdDD)aS1t&{x(7^i{M2@ncB5%uNC*rqV^g-!iXt{VXBPY+|4
z^w`=9KUY%cHt%e}^CdXFSN(6-fPm|{bieMdOl4NL_09t}|E-5>CHTgh)oXYALBEdX
z?W-~MatMB3p6yIn0#g!1aQB2`Bm({Hql(H0Hfc<O3FzPbd(3nRu|zIiHmXYlA~{vz
zxGAg@>mT)ywn+47<k{3)mp*y&S+gVl9~NF-6cEApq)Gv@fbREH8B0s6+ofb@3sl1W
zkhW&EI__jb+n+hcvYOFC6jQKBHa*rn0|T3rUQv?{4;+Zj_lC~li<CYT36tPG?2q*g
z4L#>?%fA&A*m_zPv}HTuu#Y7)Sz?QllXpHn7*iODj8%}%xYjoLo~Th3gw0=5O~uls
zi&JZSNwEG^3^%<gfUA*(U<39Edy_sIS0_2P+$^mv_Y+eU-rtoAH4AAtL>;@;c%EE{
zu6BEd7d}-=4rKG%2Rz*qr=^wicN2SD?JIGZclVE<3go-tB3tR)vYIFbqdLus&Ja6R
zRl=+PZ#H&1{Gj~hD{A~dRcNBLf@T5v1J*-dqmzpwEaxId$L6`(Wi9CEt3I&^#=Loj
zd<V?{ioQR>*fP#bzHK~FDF@D$m)DDMUtCxDrq@+sNw>pAX90l6`)K`uja`XQkJl_!
z<*nUKN=;2FrPwu7pY!wi_1b;ys<ivfM$MW=5~~CQyLbkUkFmVH|81rj0RTGXEVIt<
z9>OJ*GaXw*eMC(k5VR_+jObQsNAoMI1|6g67icZmen!Q8mR5W&{O^;dJo}5*id&lG
zuEwBN_ZkRYt9KTOa^JpbBK0i$+++#3(y6c?0XcT7Hyp<ua!vp+Zdo*e#3mLFyd@A;
z=Q+4;_W_8byy#m098b*Qaev591)tmsz9^x7l`;zv3&11HPPH^7y8-aC{wE>-oA&r3
zf-9qh!A78}^@g7(bNvKZ8?uwTdO1^QYlds!a;i;$Z;AJJsh?5(FP%!9mPHCme>e##
zN<;xNp?!6GgT7ndz*5<ZgvoFbl<nE=3%109i})xm{%p<UnL3<7Lg6fyHKV=9hCg$2
zk9~7ZNM%>aj(Hax8%;1O-slpsT=y%~Ry(*QESi~H6UsTzO=^9gMBCX4q*q2eT*hk_
zJb<}*007}(7eb}MEI;MCraunRHK)slzvwQH>BV#AD{=Vf3MX^gLk&9Mv)~da(_;39
zWzURGsC(ZJ%LVmOjp<>=2zzdlFsf6?pp8Xd!nnwyXb2bFRyPNT%{M2%Zy~>>O-<e?
zPn7cu3$30<s7SWaT1X)>{wQC9uT1sz?UrhyCmZq8KRl=E(XM3gX?|<s>6IT5g{e^Q
zo4HSOAS(f5aD97*lSmp6Zg=M?Y7xYb8Dw49aiZ=yF|QqB5wuf3(-N?+*#2bVcR22+
z%)Zv@2b|7qtGu~5;ESKDZPhV{<vueY@_gX(m&~tq|4eH1eOFOX*GB@tgm|~E*4omD
zYHom8p|1MV*T-O!Ly&M8OYc6y9`cYOA_Y>V*A9DSF{xhpZE229-Q1c(oUutJi>G~A
zxus)4qb_kEBhbE0OBf^o4LA)fhqdK0@0?5T+m)S|s2;?fHL5ICSJ$d9KVNG+PnRNo
zmuR#?^O^gBJBLWj_ChG4P<`3{RQ%`)c-2>!(>$^M{Hqi(*~ifZzm&Z7PP>u3i$=z4
z;cuSzL-U*cZ%x{dphSMiod4B`N$2J%eBO;AUcXI}_%d3L!S$f+)<{?}**-IK7B);Y
zrbql+vuSA<?>>$I$Mzs(Jn-9?Pgnf<ST!GPvyN9QQEY2tvMc*|)}NaWY7NFunK86k
z;PhIm%LsD4Fohv|7D1-Pi-mUuR7GYgFt`S1o6cX8vY;Vix3<)5LC&%Tbx@8OF6%q}
zMi1!J+q9c$1&N)jSgf^9aM$UTZ_YtiSI6!!%DjJW+D%H(Zma5T-cCQJW9HM@cMVbC
zg>2A-86@b>f#}a1yOnuf(&nj2)6nvK5s_x2S~f>Xfcsw3{bR7v{d<a}yzVKQlwz#t
z+7CNj9{y7KeCegd&L0`q=?yHuk_kZDM;|xj)H&w3G@~b9ZGIEtZjfQg_!Nzx{W;T@
zDee@n9Z?W{#WyV5AO0deI3kwxhE5sGY-OFDV_pVX`0$m$nqT4YES}m(uFE@(bA7p7
zGuJ*aC=+3pi%`!!VY1;R(;|@hfgH`!W(Acgj7&E6#c%6dB!6X5fl`y}f_o=<99*qe
za<PKM-L6nqDH2JT3vt!5D*Yd-+4w6mfeA@xogz~nF_r0Q%T=BU?pWBIG|MJ3Y-{0Q
zBsm&c#c!IK>AW5sQRMt*r=hE<FW0@WWX*BwjRcaTNi#F!Z9rL43NpE%!8!21VY*Lk
zDxdIp{M}Ic)0q&NDV;E@8dKry`gfdXTu{)$U`Mq4HzFd1jOjY1ilW2;nkgczm&`pd
z^3j3#QT6<uTlNi)h`87C(v^BrOp$xc5dv62i2hIo<wvh<sh0Ti4+3#W4~ot4qXK@f
zfpP|qW~FQe|5_J5V1=FU%S7Gyr=bflIk}5tEG$bH-1!A7I0wP;Z+z|v4}a)xc@i1;
zxwOk-N-+G$#ZGTo)Vc&~nkVI%%t`s_K>tlIHTqTZQb=;NJ|S+M!{(F3`lx0E9@0jX
z&0$I+y?(eeA_#cd(Gr`u{E^fd>jWWg_OvV5>UTg<U_MZ%TJp#ua6xi|>QY^!jp<ZP
z6-Rmlu)xwn>viZKyrJyXlqEre-9{~hiWiQiUBF9@{K#*DHU17D(&IJ4=>LL-!Jq!c
zGlsp%p`V(id?cm)sH9GnBUkygb!Fg6hK-6SG3h-^qR9KR6wp7QUsYN4%0)G_3GYj(
zi|`Ec?^fkB6%EnYlz=l2I^l~B#U8Tmc*sKRcaymL#T1t4$8W@Lw{6SRlgkJAtV&?L
zt<eU$m|t{c31{HJh@5!~Sqic!20LX#)bD@QB}<*z3-EMa%8khe2mI`}_SV&N;`R-V
zNs{NYBW8L{9}FGRr|KmsCA{h|RW|#KhdR}+w%VOkOZ9@iBK^Z{p4cByi@CLXXnYYx
zQ;<W_K-=qTwVS=($V+6-Eq0gYJ>M7OBjd*r=o6)t^W<_ujm>et&fQwQqltdM=YI+y
z0a5|f6@MX3tD2WSfXq#&43-GIQqfKgsEbXctA4@!){2V?E9M_V7P?`okbyo<I0?7<
z(tMOAU$F@XZHgkaT=KBkH>MdQlD?5l^2|WkFXp~)L`hB8@!%F7z+=(9YFCI6>R}r{
zQiHZ~+}H&jWTHL_3;+>?)X4jW;F}{m%T2aXKGFDLi{(?Wl0Fi1gQwXp%*Vwa$p<)S
z3P7yXAlJD2N~EuaF|2LDXCO6zOA;xr3?>U<is_;akFV<u_eL`<*R1h)-0pJm5&P*U
znyw7EI_4kVVbNTrBZx|AkhN5Fd4aY3QE;g0Ipxk8W4E(Lt@9UMYOVV4gbpE|S#7bb
z{0O5L87F7SNOl!y89YJd<x7t#mDBh%kjq{UQx=@lab4@&yaM}aVWujpziG{AkQ6pk
zhf4^<LT>+L>rgXzAH@V6h&TyCeI9JY!KC+tn3W92hidvw3#Xu*WQ02OKs)4z-0%bb
z%Y<GX==5Q$NY}<)(7lwwl7Vh-06l7Mt)2V{Xs`CXwaT_MkE-dupeho*T&pS^j9JMG
z2d|!a@990XC3;T?1NV4t_(YfI;!sI}8DDEr_Y2-=)ePVJPs5&e<IVjuSJpJZ)`D)W
zWaBMRr3vr0%|oWpE%V32Qp6vPV87|7Hc2>4bR=IQTaW8L_jxY#z->#soblxO=K1N_
z<Kc8wsVeT#GZO*75bTy0KKm+s`&^X1Uc$0>_FRMHQ!I~7!9;<D_uE4ZC!n1Kzv=cp
zbsJWRv-M?ugek$j+BkviuFFS;qmiy89zCJ9O<o&ND37~vYS%hXP-~}iEI2!(dxVz{
zSmR}Vw0bpKim-paD)f8!VGSsqa_C3yUwM87dozvswLoLhrQJBQ4b^UU6@yO04rjFC
zN5>k(lb-UZ&|Kr%nsfWdZiOm+bhhC_j*Xv4P0rR{bChxR3D!!j-wj&jy32;okBz2q
zLnL#}NM!x+3^z}r?JMbwSNqz?-Zb~$!Tc-1GD)wjqn|Skh*jfSU(I=Gd%=<yC;``r
z%`c}O2BT*Di`d7GB#`ROyV=()y#~TK5`*SUzZCl}9tk5}^$N)+j=3lDD}E$HrW0g2
z<@@7)RpCbyyhFL`PJZ{2QsE=1HI|lAbpx_1TRv<+%jKC6%O9rCP$7UOhTib~Shq5x
zN@s-X)*#?-e@XDkCeaG}*l9?bn^1(qucsmf?@1Z<iH7b#Y>hj%A3Pwlc(_=GP7_+N
z1n`-l90UT`6DxDiT%&@G5}m-qI@V+x@IT#-iJU0Cm#H}|1l5G;prPG-ewK1$gF00T
z=Yr7tmDTkX#@Q&d(BJDdxqr}jyX)c!YW}lCZ{v|Nf))LFGwmNB9tC0JZFSiPy3W*J
zh1t?9QWQgXQvYZsoL=EiOaGCvMw^sHND!fk$$>QOAsupfC|5z^Vw!RoDU)S0x`Y@S
zL5o5RhdyT+G)W%^P$LGi;^cjE+=NP9f*xlPO{)gba6s}G;!p2hEYz)AxbF~B-{U^X
zgdp*byuBXn^A<4wC;t2s1}mp1g`{zU=9U<2wi-6P^w-x9glp==6PS}mTk-$UziYt3
z5BrFw1TtYbOy0+R;2ZXpT^zdd!=Jgl3VO93bVO<8`XbwAlS<Z8Ksf5Zr+f(`M*fxe
z@l_GdM>Lx@%Z8t4QR&|k;d1f-_X&5Rl1UR<eNT`quw68`OvDt4YsJKTY-fa7MnjKN
zIDYEkDBF~SA|ZBcAMN05kM=jDZyk1}-2X0-WZ<_@X_IIC@GRj^6Gj2lw}am_v4SGf
zU)ET5n@Jgkn_s5zYTxnRPMM00isR<l3eE=av~iYXInO2`PpmsJ-eoHVnX;r>t(v=U
z0iWv#8<Gxi-tT5Q`#gy6lKWQAMU=b%DU&jP02_tt*{5l?42|a5Dl7eTGcVB-{WCt#
z;V<xWFf5Pc8F7Z<)sXdwzH{-`sFW|OZnbxSN1Md(cJ<)kxb_X{4^Ow{E<#Cx{OJyR
z;??xKT=;<qy!iO)gsRJZdcAFHI+}HbxyH{=U~N!-Jk{2Z0fch#8mw#gX0%LY23)4i
zatDrggzk{Hc%~Rek7=tv#>(ih{8oNF<6UXFnZ8*s3`f;gW)=v%7R>K{eW}%sq16##
z+zT?B|EYbrS(8Yv4pmP#3%$24<!z1{(`+)o4^nQDcphtJw{fE<J!NyYe}NZD3>hL9
z)XKW^Usqo{G7q^o9`_2};e57FyJ3{rHDZ3;oIpI5nzFPB@Puytn}q;)kATiYzZwYn
zBq$39=wC<ZMX9&@jj)?N!Oelb(@**T0>(_c^Yj+(mI|M=({6ybZ5fP>qgx$SYBn`W
zUWCq37<$6@(ILq*@Ahug>`Q5p<&ZQ}y>{S|6Q77oiSj{`oMFe?+>WS5cxJh|y<HjO
zi2{{pjpy!e=8OGnIIJ+_8ff!ZgM~=-C!==T$5xP3-VYv?Zqd_c5H3hSL4&Pq;jl$u
z4f+{u_5hSgYl5N8a5K8`=XV$7n#C8I_}Jnl+YMe?vm(RVfBfqb&B_jwTVT_9=lSUz
zYf#^s$dGIEGCjRvjS26TJ>~HBKA*?7loa(3zmvF6xhKqWVfuEo&--#(dG}X)MJBz(
zsLTX4VT4}}+6<{e&SZ2Dfu))*>+3~bGqTC=Bg&aL$EBAP6jeY*?<>;Jf7t3U|6)Q1
zeSTPVkzQkooFe=FL7q|WSQ7)HN7qOMnJ+TRwF(8~jz0-#NrN;&B5S{lCR60lBf;EV
zn#2?GI`XgVf~Bs<ZM%_TVCW*`TmZ8PyS)X5=jNtO>6=}JMnqh0Al@7|N2HN)h9HoK
z%a-sc%9YAQ;E|NX7JGP|Rb{YvQhZO83P|-T+CQnbcKy8`qJ?c0-2VGrYoIw}-TwyI
zPk;YGzP0R<aP-W=0E`1T6DCGjzt?Vo%-#9d`W3t8mXVW@EC(G{K0&`v%LM5v=_9s&
z3kyrF0om)jL=PXY#0xqb3nrL;9-s21NPkc7<>Bz9r|;Fi+ac(4kVVe-kr{ORJ)cj<
z8KxXQ>*IXOH}1cD8BrfmB2iMWQ3s`U0iOrC>YaQZP88hlx@Y$jIQ6)!(a#o;(X%IK
z<KWP$$(%Yz@=_zfnKd=y>p@#AP2Sqg%kaAb-1kZ-_46M>JYu;ikSc*+m6|;yl3GUh
z&Vn7mRJmpp;r`#bB&ojo4&g*h)OySKL@26X%hle7qlM4Z6U@$#^z&<&hCPC)AhA-P
zb#Wu+CLi17FTNjOQApN6#h*kIoj;@|N{8TpIpU&X*4br!)q;;A@kE<&kSs#A*;xwQ
zTf|QjCy$aL63?pOaO%Rfx=jDAu*9Txb4(ExAuw%_E}&?cX4qk-SP>oL7JtE26jIT{
zg|~{-2K9)_(sICws74N(Q!#5}>^}u^M#rkTI?!@ODG$A)xt8%1xP!#<IU@SW5_@=#
zuSt%Nx8(Q_S)2MBKFAZj_;26a-(*VvJ?-+3l;XVy)X&Uj@VU^aV<)nV&pR^O>3b|#
zL8eBcqHLU;?|~`5bv9#KXWm$U;`w;^wPHaERjN3i>9s=@y3OaacjPBAM)tp^U9iJd
z_w{paW|n^qvFfKFc^MJx_o{Ui46Dm3heP9NonG0EdBw8_JR~@uh<wD}lO>W|(K&%p
zEp2m}=9oF^sCLwtKfN-zQIOTLcvUEbL)yHes?WL#|3&*#v_p}!C^*@Juh$_FpNKUz
zR{kw{f&ddJM(wvn(HYz{DqNrd18=yG;G$2l<wn%Cw9Y}b!}fEr-w(BV;kC%&^ccsJ
z_MC8e@j0yfYt_R*D0_=luZ}afxi=)2sU-f)q-*5l=31#k%bTB?+FV$|@hHRzecG*k
z$E49ITF8r?=eL9PLvF6RzCFv_CqCXAW)X9Vedzg7k(TSHO5Fn+7ukQ($J4jSc9iD>
z4gx8!R;~RG6@Rn<>>1}WT&~u(x-%O}mT*m7^TsH6Rm=*nDE(i3W>Q|be8Q~gA`h1e
za$A6v<`mXlo2Z#~<U2*WLH)o5kUMXPkmGZ9p(i_At%<sC&hr}dUQ`{5Tl``5MaYhM
z$ZODc2hQ%iS-VDu+Jd}2*Ki2Lnfu)NnFSoQw5$$Rfm~fm!jNdknMyl0XWti<^m^0d
zE19L%NxvPZ(-lMKDcFsl0&GuJ@@K4;6V!Rf*C3j986YR89p0AuLpDa+cFN_>+qR&E
z?pUADgmFNP0rIjN7U4*M;;Y{z=2B+;(|*nR$d=7bnE~l2-%|!$EKZjK?Z&&@f=J<5
zYX&|K91|3|&3!7h^uK|nMt!6UC$T~$ZS4Bf-c|{K$1e#_USFr;Uw#v<WVl&<eM@Bu
z)wu03xD`ARKYyz72{CVFeb9X=kAVZcGt5MW=o=G-JcDy8K`bIkq^GAP>BVCFk@7{I
z|F8meK7BVT{H~QdSf?J2hjrtRmJ8q_v+PA9mos=b8TV?LE*`Iv<mPX`BJ=sM?O8A$
z##_Vc6cm-q^l%JBzNu>H=e?|o@$s;WNur1g<}8)b8a5v9s}EVQV=fmM4=?YSoozuD
zydHX3R3nG;`eos71Drw;SvKshp3hz&X^Md~YIG1Gk`TMwM5MtYGDeiIhe__J91e_X
zfvEjyVgjwxzx*%EMR+H*dAhIfJVl^9<4pmdWi)_z<WE1c?FkTG2PoQ-yyZKL@yf1{
zw6OCNK<|xE<Z}>tpMYi7q8jQS-m}hu7VD&nfB#Yg<1phh6UB>b)z|TSbh3A`oOzQ<
z;+R&61D^^CWpvuFL@9rS#43ZvS9=M<v~<1@=waH>{~b5W9jg)F)g%7qHk&OT66NjQ
zrECw3FT_?>)sSgfVW{A%m>6&T*_C)ksD;*?q{-7jpR)N`?;Xj%`I-d2KDWXEmQpvq
z4t@6q-KK<I%|?XuV@Jxm!GG1jls+Ud!pfx7Ou2CR5o1X5+7gGqtbPnQf9m4t5|Xks
zM;div8Zr6X>mTKpL?!Hs=+j!rTH7rCaz`#!j;IdTqE=#JjcM?MEL|iYL8aK4*uSeA
zEk7!jqR{b4KOk3gurC|wd@9q^?$=cORK>+FK2fpq`opq@a%v5<eRyf$pPH9nKAn9}
z79V>nc|+R$R2P!CFb<4=Q#o}QMFBX}KE9dXCzYwu8p8d-REbAYjbDT^|5=W}Fdh$I
znl)$)Kj_cq2mvqVt>@7NrMx8;nhilz5^mWgE_&sML2mQ;r|o~Bxw%)kWEqvH8NcY5
z+Gb-?nDe=om;D}Oqe9kBTAO9MQYz>1O0Oyu3e1!s8oL$Um6fv?baAWc0oj$4pG5tJ
z4^}Q_D({wQ5P3O;{+YaH?WQ7o#dW*~Px-D3o0bMDu0oCJ&dmPDjYOzRhvffg?p}9n
zJ%@_oq8Ony<Mp4EU52!!f9JUH3EX}31)PFK2$@|k{G!IY*Zf>xwUfqa?XlB5QI=Qt
zSnfLVo03OonFZ6$>`hC+0gn9fkPG4`>v9<9Q>3J%<1><J_H*WT*tre}$$jAH^t~Sx
zYLc`+NoffBW6W-2zLy=NzWZ?3SNCMO%gaHT$xim~eoeoJU``ptj*DDlCWCw{q2<qE
z8Nbl(TwU3u9=Q~dx;LTCS7mIX%;d5a^47c}=ful(RkciIb+d9P<(|WaL-2E_3eg47
zV$o$YE7-=Z<5uLLo!HLgAU=z9<x+t6z3Px-%F?a;fN8WF=(g*yogh;=YLjz<FR$A=
zdzi0_eIdyv{d=YFAOp)VW%+S~t1C<hn0PW91<D|YKM7T}Jr;Q`@cO<)$Ee=(++OEp
zv*@|191}D!%?F8?_TU&RYRVg9ipVF_Xs{8u<n5!R7dLQe008mzOJ9E2S_5&-ACOSc
z2Gkw6B9!s_H9O1Ar1V>idOd~$%YtZKRM+>_0(T%DPZDXD3KUxxW-G!dl8KVJ)t+VA
zAq^@!518QtKt{RO3l^U2?Q)IweD-{i<p}YNY?=tk8MzIiAa^PoiAv=U@FeCx?k$QY
z<@*lNVEZ6#P@Qyui@nB?E*6O))?ANuD`)@-!M4N>i-_CXcr9^#z`VP=SE*Zt^#<5G
z^)&>dVb5g&nPW)8J?vzFoBr?BV>(mVKvq@h=?$<Ioib3b9b~t*y8@zd-~Pt$Ll`eQ
zD1x=*nfTRx>fdUXr4nE-3iW>bwtU65w#L30B6?m%F2`5)1If$8Qw~d!^(+jih(V<S
zYFiOv64E_3mK{nd{rJ!z9Z2%FIEwqn;YJ`;xQC(;S$#)`n)MHP-}av4a9pobP|Ff@
zMG8&IcLQ%o@`{hx&1`etEY07nZG1qmg?X;5lJ8~6qkRFx6{<<MZ~Xl->8`$bQgsl;
z3Xoi$m;{S(tiZ}{xx|DW-)etNBzl4TV!MQTGaFZOL_y-h@>JLx`6s_4)7C}_(-0!#
zA;Td}jN&s^GYh<a*L{&ucaiAc{d+9eIRR;dgplkq{|BZHO>Ko2sWVTaR`$fRty?IT
zH~ldgpd*k$gQH0QRf;_8vUv~{=wG%-96T&-kC@E~5R$$_?p0A;g=y9W9AAly@_qU6
z>Nl~hi1e*UbGwSkiqZCNz4WYsWz$l**-Q8Oi9Dx^zh2Qh2QhIx>EyaT_MiDtncpc&
zb+q+4%-TCS#X8x>y}gy}zgpF_(=VT5kVV0!LhLKF-9<i!zb2Im%_=zfl#@Mxf;T%-
z@yX-*0hQAoH-$+Lcj5l>mR!|g3C1EqH_(i=4|uWCkeqG2ktO-0Sq2OI=2RjD)&xn2
z%YgLCO)577siZ2Etwpj2B@%XfqNz5Z$iCID`FR244pj=s#l#cog!r>qzF8s6%EkP@
z!}{~TZkOT}hDg#|*N8^N2?b`{@KT-H@5V&r16R4OvJeH^AXM_ZibaAOnMqU>642MM
zi3(^5MLOER*C3^OM>cH9si}hi*~J<U?@PWq{$y?;pNKw<I*SCWy=neT4T>^7&bKY_
z;FA^TX2B^AF1v}PQ~MVt`gc_6T1>S2*X>M|XX(+f*8w1mk)hp`kcfHO?#KbagjP;&
zxr>Z?zQi9R=Um9eLP+ev5*fL(@g0~<@_wdpdpA*S)8KOVi1@%E^5m98J`VB2qAw}n
zewp%kRmo`H{%(%*ec?04oKyH*Mjw&{Yv<^g(HD)&D=HX&L0DB4ot>YJeAEvCFRG~|
zdO4o}&s37v+LAq&=5VsmTNvV2?DpC{;X3flJwB5iiQZFD`kNn#!)w&GI5!6N*GkMM
z-5Iv{qf_^DUKOmD+vjNC_E{&bd9&>y)`)q|>;&1=ovYNL_Bu{0^n&y3K+NtKAv}&S
zHSxb}{w}_D(^PJ?qCiXp@$>|E)q4`TJFM9Jp{+2#xiI*qZ`{k96x$pETYlO3c+$K&
zP5szIEdtz|K(>ub?vgppLxjjSm}3tG@^9gmL8r&R=7d?MFF!H2>G80qIbrzwv_p)Y
z-Il7DN+RzqvvWfxkgV?^i*HT<4wKfyE5w8Q<3p+Gg(+~x{0TD@dHLp*JX+50e6lN^
zX_~iM-+NogZP#07L(5xtxigcu<F!1aOh1>89en`4=#8Hc9T^(~-AaQuGZg_Bu`aP}
z{vC~QXYU}<+^H9K{Jsoz@x#Z)^C2j|L`1~(wOi`bjk4~twG2i4HS8d#9gdB%ZqUXf
zbQ%4(M2Y4K8%RQJBmf=}_Fk`cw`EQVwlmPiTs~oU6(`_P^9ki3C^(yx+ROI$sA_wX
z)P52)OnpLsD;Ak^a~7=}NOkQP5*ase%V0V1i^<DS{Zk5IJuzKy0ySqpjbC}+h_jAV
z%Bfhh%suS5Kl=$HR@ZBtFL{Tb?4G!j+>tj`%2k@deg|^vceA_p`?23Hg5Pnk4zXrT
z?hw#Blkyw;w7+`X39JsYY)brFG@d<>_Hu%gn+)qboS-0fKN(AtiDdtU9aC$Y@I=)H
zI*}woi2R7%?CwH%Ih=O>*W3sooh1hsxJ{KQajz-Bu_k|PeA+_nFq&DfnObB~gM<m<
zsE6rLF;VmJ-TG@@Gk=-JnyY+`Ln*56zQu!0EpTi6*z|k?J!>lrPH-Bt=9;At?HaY#
zko@Qk898z{vu?3JXzno4rQ;w`tJ>o3oG-f^9FetD{k;1pS**|eIrsV9?JArG97L(X
z9PxR>;TQBi6aig?I(tEZxEBq{AvsO<Fh&waJhTJN!knE;$)_!`o4c*Yna~H;$F=9D
zPz!J84#ndCEquWcy6gTT;>stlP$49XO@FNwyZ*$1E!?a9rJFOYy*8YRim0hUrCvzm
z!v}^qJcDNLgedNiq9TgFmS^A>5LLnXZ&HPtmw4uD60*;ViSzTmIq;R>n+Y-bw)YPF
zNV<14TcOzK;WQy~u2gI5O%yEn-aNOf>X;~BqwG?0o>on2T=f~BN}6T1VrW;Clbauc
zqD_Xc^UY9vczw=Yd(M<hJC1yz?s6j3ws5GhJsNN--V=%GkK8bsYZ45{;aT2|<m~Dh
zl#dpS#wSl<;KHYp5o4}5DAUJ1H-Fp<jUS>4MAgOlh!FuMapoHII@qK)%BOT&4a%*P
zxP5&JfoX>9mWPH#Qw32#wrRgIK}I)wUp^m%Ug)<yhr#b5Pc@J+kc$3OOq1?^M=H(6
zDW5L`D*B*2ns~9zd!M>DR5}@vQvF;Q_{(^7=Z^$tG^7K-E1C}#S^Fo~la|Q!3uxv9
zNl{2enoYN&f(11FIYmo#<+FE<oZk;5x7xH=v9Z%DB7aR>FF9EVzVArIL2V_(;(buF
zLSblX)o!+tG5*hz#JI^N4!AwKKFVOo!Sx?0lEJGQn)=k*BuB`D$Y$k)W%CRAZ>C1O
zyEMh??E}2YlH<RPltE<f5Cjf!-D<y6vRhR=qr6TIalqD~Tx|lDD}l-@^grAe9}MpH
z%OxX;Q97RvSzTP#H^v9q=LKMsZs5D;mfKyaRC@pXy&>p3f1B<648z+cT~pveF+E@E
z*RLN52uRq_{DcxW_|QGE>N^&55*sX>JDPZ#`Yd8B|3T;D9buD3L!X>F)T8rNIfS3q
z9*Xcb@vy?^-mfDH;7|Cv=LHH?KdK(Sx%xut4D}H2LpaUuyl?k5CPK}hpOE2%HW>Ns
z3+5fK*oRM%8FRK&fmc%NSA(!^g^)YOrIuT+DB7>6-=xoyS~{~OpVN}hH<F+Jt)|S<
z(-NLHy0!^q^y{RSmT)R42)~92&C0y6b6aV(JXC=(F7UiQXJm0dUU|>x`;v<<j~!lD
zJ>k?%-r4cshj%y(-AQrD4)&7Q5Ac^~fcyC}VP21ZkM8`*Vhx0MLO}OK1!K}98n|qA
zSzp-1&Qu^z-@Ad^gXG{Xs(`;oK;!QV6UbhL#?^`WW2E_HF!eL?cD_OlPm`P*=>%n6
z4$$mB7!YI@2-kF(7mgqDo*XXG=%h#Lvc3~H@fBe7^>E`DY2RQKw^{geZF!ydYf!gW
zzfFeH7#Feo?mo@Lfw7SLZ=iZz4+)w^CaWQoIg7`*EJ%@yT7IMo`f9Xb<lzh&a-*FC
zn|zW1!d;>F6MRq)=hj<Ot;UAR+kvB@q6jVA1u|)hbZLCygVcr%J*XRQ#U1o{5(or%
z1uI`BCfoVMiUWMiyP0_AM)KW_)lzL~`r|U-d1PN*h9f_+>WYs#?3I2a+T*-(rJFvo
z2oNE6-~PD9&=zuULYi$N5Fhr{Ej23ZDfE0pzmJbwHNK(=)aAY_y@LLL@bRK|j$U%m
zx&*a(!F(D$y^?E^p|*@w@$p9@;N4&B?guNPsQbJdewlwZ#*?3k$68DPrcjCCfOdDz
zg0>>c76ARol~>5$Oy)ZKtL#1n9H(w-I0*OlgghBK+wz7tWqZe`0C<zW3m>(Ym}!CC
zU6|&$U(!=R>HsAuSA+`*OrQ@~1eKrGI6&<(-vU0k5i5K%0Q2ym#k|VWYD%IW5Yl_y
zp>qH_#t?Iv>1|Z~^6O{~^fO#dc8usQpX~Jai}kx*>sbYbvxWDWdYPWD3eOu}A=hDR
zuX;QRwRAa3MklFW&eD~m+xF(J`+ZF-)Mj%NX^v{=bHZ_C(agG|kyegy4YT|{o$lfN
zw(nL^_e(}j`k?*dektF$`0(VX=m3W|{D3blP(HOESYn!1fv$<=r-+w(yXqdjSw0!k
zwD_pslH>>(`!!!Oo!%Wvo-d{)H+FTY=UsU%E6;TphSSif(<E7^J6W?ydshtq-WNDJ
z8WDTC3xAx%eezzvU8gQGs(PPKpt-nnmpF7^BKi0#WFt#rR2!L7PW6rYuTjv#S^MwL
zRq!FNHGZp(CKXLox76}gA7O=1k56^Mem!l>7d-pwFh>$a%g>IqY;WFyWHmoqk}UpI
zvv-dFOsEi%9@(GBk{2f}0J;BXW~Ynd|Mm2hQBi&W*NRAkI3Nzv(%mq$bcd9*5(CmH
z9nw9(&@C+>-Qduj(lK<WLw7v$ThG`3eRbcSb?-W7pZ(cE8aa<tnsU2kZBKFQ1Hs8&
ze;?5r7MlnEihO%7W@1V8#mXh>n`HG?-EOmU!Xf#ICfyfe{7_Nz0ST-ZuZw`jVTZNh
zCKQUA4ADxfO=^(lAdA+)8P9%KlUTX<KP**(%`w3ow95voE~}Fr&ph9GBZbNSD#IC>
zu=e1ao2<Xpj(hS?6EBq0nAE`r?;kEvLw&};rc0*o`4Kxe-s|k5z4J8|H+$K)9=-HM
z4hfX|&5Pa}j*lZxHx)PY*>@gnENnP@AOAH$@P-_uO{I2rwm+kFyvLQYc%F!Q&8ZpX
z$THdPY6GRA(_4{|#thNztR%D1$jN^%uMuRq1_H0E0c5EhfvX7$*JRl;1D)DvRXlQz
zlqx!KSK@Mv!d@FAw@JKFqq+BT>Lz2xm-FDuqm06j^#U-Dtd|vm^n4ia(?Mb38^|^_
zwJIq=G%g&5>Cm`hDVgVQsKS=`%mMK!;K4*D`S-hxuc<Ep)76-cK>%7ls9+JlcIwKa
zdFe(qTL%XrlhzpK+l$YP($I*E&dUOay#jfI`k{)qkFkvTttL0Qk?`nn>JFc)l}H4-
z_imqW0CgzD1UsH5-2LpR;jHa(6mfj;bUJ%-zY3mLrB=YpHS1cnD_w*Xa!gI1O&YA*
z<edN6#!O|bj8~*7v+*Hr;YtgZoxL|qLu`5zm5`~NFKK8q*91o1j%pR99aYxlv3P@E
z*D(u@Z%55?R=iY4EL&wqEvRN3K9IG&Gt24md$51BU4fYR6z$&1)c?&^cxZk_dOy_q
zP$PE6&a=Y)xIT7uRLm$sdQstfb@hGO>Q<Xg=|tksypHrOSo`hyEkT}cqqNI8hhEAM
zUX@7f!{$lPa6>8__Xuy1dks4uUrEJm>o)W0#&6vF;6QgKM(oa=s5R^JpDwQ12tDOj
zJ!Erj74|XF=2HQGX79G{O=j$8W-f48<{_>iH#?&<pZ&=v)s@HTnEU$e^m&07AJPih
z#joo=T?Hy12UwCdjYUV9T^ZIDB62_Lj@-`eantjz1imuYj5w)kuof-K$drv>M)-~t
z{lX8NH$J`bMl0ve9x%4m8uMQF(uLCr?1!7~ixn9oY2(8HQONM{@VRZ*6yk+Yvdg0Y
zu?1Un*~$0yc5Lwknji^dTs5NpkLFnhlYOa0>GP1yTNRNFN$1+>hFMLSjydmZyrV`E
zl9<b*_E|rn!jS6Nmg93yb9ceg8gps;vjPqwR}WS=rkA6BQMmVo|F0O4CY}$BdP3t3
zZ95O1*L7*w9b%X6g=)va?EAV#h0Y1tS8WJ%o|U$zt0y;YNDE@dyY24o$!o>;HkBX^
z;fd%tz-}u{zQ#5Vn=|Sn#H9O@i28mHznK#d&WPG!o~~Xe;p2p248#D0Q9~Id=tO+=
ze^H84#iO!gR0c`bSp^#Zz#-X#M;+N$3%7JMjl7=7;a&FBh^FR7+3DwOy-b|sVWz%z
zB!TnA&b%BQo#9{=Piy|EiOgGEaz$r+Q;4sr08T4&jN+kkg6nYggi)Ji;A4hc>g5l9
zsUhIv7E~BgbFUuE)n5<adihm_qmV`s&ZZE+feIL)WZbcmhg5$Gj4rr+Dg-oVrkw30
zd9K-Ybv|e$Q#|}pb%2*uyr6o`cz^ur;nm`|o#gBEv3oLkDzi`jA0z%Ycl24C9i`JO
zm0IH4yL>03NEs|47p??=(sXbs|AI;iD96c1P=x9f1~cM7d5(^xv;e`~F_O)hd|csG
zxam9?#N;dD`RWSh{eRa4r!!fVOvA%Z==jN20%r}k|0wqXmj!xAR@_MnO`(I|i?Rg=
zES^q29v2bUD;<J5IXbR<7T|2f3AKY-GQDnJ_(SD#zgN~M7pP01J?FxcAI6|X4Fl-I
z$pR>U2nzS~;HC>@>vk`Tgmc*-k8}WSiR@S9$(^$Lw^)pC{Ii|@9=`ubgBh3-<F_uB
z<9lk+c(DyvQ&2CRzTTA;8_aZjBCsZ5zqEw?Q@R~RAa_Sru?vg>?f8C#I<^l^39LZ;
zWfEATq%yISLwVSW_#Xc?4e$;28c?SWOL80JfAG4|&}O9*faX|o-1(gc`YXGoeSCvo
ztt!bfm^q;bxAv3vaf{-j>zycM`P4jtDjb<FQd8u9+bytd!CqZoKXTknu2~dR^Ydq_
zGsDZ$(t~T?u#Gt-h^o4EEq=W+;VtsNpE}So1D?|reEg8njnUQXMysxqvPGxLGD_XI
zkE2}czB_ssSA}XE03G{m$&P{*3+Vj7Db^-M+9&1Mo&7jZaTY0dWeizzAZ66BgI#Eg
zUi-f~K54rivL_gr`FCt8MkAM=*?OiJb(GR!IrVdn2n$=IC#5jj5>6u@(cPG@WcMX;
z2!;!gk7!$>QbBeW>lDh@EYqAyuDg&VrZr~9st<grv#w-*t;D15daL-$?i<pJhem$C
zUE{r>NMG#xBgm3O__5#ealdUbyv$XNmDfRm7qU_v7#EI)Z6VsT+t`hZf@Fj^Un@cC
zfD{z;9|7U*ktp3Wf{6;j{Rm2cxYCDry$Lq*l(?j&O7Dbz6`{rVT^_H_lJ6GqaVrmR
zZt3R?{lT6pRFGiMX>Nq51BwBUn`5HA_P(2bCQjxNP{Vclr<<6w;<g)}OuH-E%(;sO
z2aq5?geLuu%nSX~S?3I@(tLZb3Y~e6U*<Cpcx34_5PiLo`sB|BWOIO(QEi$dogL{z
z(KJUc;9N<zJ8|3l<Lb*&^joZ*MJMgYN$guXLUx!FT__}dz*G2UxaO1aUn}>`oT#%C
zugIs+Z6QC=($MUO;_ON9x7%J@NcW`Jb)J}~_=*p4@hokkBO?5L==NyTWbU6EdEZzv
zp|^Xd_;1;^K|wb-FJBnEj{8*KrqGLH{XOtJ%gxtq<uTRoq$Z6WgKKou@N-j*{n+P5
zSC!1v4HIS6y@&Ipbh`sMx4@`s))2bLG927kUV0;FH^UBpLHYxo!uN-LwpfL=#^1*7
zQYB~*eYW@h{@W8_eBg@SyFI-d;TNGjO(&cq7kO0~N|+y5`=a%Agez{g+0A|z-gsoL
zG(EL1x3UI{67AfKFwL%Hr5DhgZ51%E!N&O;;DWN>OsTE)$A;XUsi`*>9FI6iUX5l?
zZ~4>pakqYH%0I(7+tSmUsJJ_!r+7&jVI*9{hGFn)iiQ}JtlTRvl)b;vpgY=&`1+l{
zd1>JG!hw5(Hv3YxZiCK=bYqTW)oRDgB4FM*tyC}#$<7OoowRRT)NqNVpRD}No{jM`
zW83(JrjFp3Ff$!3dNSJ4eOlV7nRSG4lS|)%z-hW-x5E}LtxGY!yRx^Jst-Y<$&YzY
z$LKzH$)fknX^1_Ga@Do<X1Z@D^8zZqd(Ncy%LuA^FA_3T&O%#%%v>`-ljyga08M~>
zOV{P8p#H&uly6upV<!*eG*beJ@^8)Rs;!Su7z;o<qw;lujZFhFt15@O=)k&$@}>6`
z`n+&`WV~voF(PL&D5rxgTPT`<yqilgzqscy(DZPnCSD`^29S5^t5^2U8R=ZG9G?3c
zNH)n0O6uvoq@!h|Ja&ZQ$0;-~;nAWxlE%}$o{so1zp6t!^YDEf#+@O+-*<XT$2yry
zP5XDKk&aG^uhB>aPM!E(nYNTUeXc$*d?4HJRP`qD)f1C;*XDM*{q|f{-lE&ni%&w|
z{vkxI|LuXqT;R2wJ_g=2Qc~qcM0!k>sYT@Y3FpKX?(ZW!W%w7N@HxGe_9IW75Xl0~
z+zFPPe40eY)`J>YYk0Mi-UV_5!7=``rZ!_j-lU`|6u?>M)Ly>IY2Id2`4nwW4-U|X
zk<IWmkDe>zT;%vUMP}FX@C+C1@Q6SZwq9x(`Q?X`*sb2#rj0O7PI7BnOLiN7n7ty_
zMhiJu>!wO)w!W@24uPiV6z0pz72uTEnCR1v+S$J5kx3PCgp?RVp%Zb?O>mU6(P&4?
zSQo=C_(zy&&uI@Y)0=d3@&`IMpPQlI2-u2jf!>-S!t?Ew*zxKRos_|UB#Hah`mJl5
z34tec^0UwPy#|D-xm6T*(cA0ZF6^jT#l=PoRCdfN<(}(ZhL$Tt51)-06Q*|z2*SpC
zDoHA3V{c!zWNW%W#GDy)_X)>D(pVnf5@${cuZ()8tv9u0=4Ujy@s4euK&=~lImH(&
z(8;lIrtJ3KFOk8pt7o!!y+D*jyHL+1GfyX|wrK*02vt0tBZ`AC$iHttRe*@XD=kpy
zilB$nm~^d$sy<eeiO!dFgO?;j&?AXMuD$jnR<oR^d{zY@;Eh32mgG-s)fH>tD#+Yk
zza_t|YSW2FlF^f(rQjP=9hcu((^Ic)0}KN5!aJ^Sc;tR$-&p~Kup8Ola|NiN3)LRO
zJ^L93X<e>f7*?B>X0ob=hLAUdc%Z?ws=q9#!ivbJ4o*9p%35J7GX!m~idm>`|9qgs
zYRL%3yWf%?p{IA^LJ^pc4%U3-mtim};k>Mfzuou_jrTr!@$_Lof}H5_x{tewKmb1r
zdBJ-EJu{s38DE<<FH?Z2n<LS?gqN62Z8d^ocU)1PC`sV$!P+b@|3S=06+079^23XE
z`5K4O{j}xno1KKyXKz-$T6aWzTJ(yoT5Bw-D`)Rt8fvfW8N=r<q9D6#UB7FUmdU*<
z2k9I6aYQy`9@lff{xD*<yEvLGx`=)3cc@^warlER%~WuS?*Vflg^k24IyBnz9|*|D
zLm$b6daGj#4)#uojM=<BTq#-Cmqmo`>fg!9RX7>!cz?YV!1>B;9onJDzSG(1b$K}B
zt<9#HK8BW)Pu>Z8>hQXAB|Mm&A?yDWFAFr^*y|D3EmWLmQ94<_SrpGOK_Mx%Ubu6m
zZDFU^yX9p~s$uLpu+JX%K8sBK)qigz^G068hjclrngu|sPvKR1!D6GR|0*-YNZX>d
zVugMyBWSROWZr&s@x!%X|F}g~o=BWm-qPM_K<p5a^W$Rm_U%TD&u{x|fxOTpzm2@M
z`3AqUSzqNdpUnyD0>bj06Ys`R9^(KVCaJ!i&Qy<Ok2_<e>GJV~YNcs@=aeRP7sp4-
z0OK?UNqp#g7XRs=xdyTFO3=c5zA2f|pE&$->p_E!Jr-lb6V;RoC}9Lq&_MNwRO6WD
z3!t-g(fswjq<eJ)mkkc%+-rOFkIsicJ@Ebq+~7`J!@(^sTI+=IHO>i|j*HnnmzM7E
zU?meiZMg7BKjyN-7Ya#9Q}?Wxx9C!td@`i!tk~Mn6k2ud?WAy4l8@!DUFvrq&J8@$
z%rQFD@|M%TOTRWIjql%S!&KJzV>4AedoY(2trmc6(dxnw!wFl*V5C)G6wHjBPW%=W
z;CFp*anqFjNQ*}XYM;9XV!tKIzH0e5k*i=cn9XNv5>C{jiyT54?<D5s1q3x{6UnwM
zsC!+NI0ZT7p@m5B6M*tufr+IkNI@e3rD4*1Pn`~xQ6qqIdpAYeONbW3)WAM{Z=A~D
zl}YhWep3kF0_qxFy4?r-K&l+klAqILIMXz7oZ%6cJZb$Z=q)o}ENg~(A`gDq^|~8k
zz6s52RoyM;5-~D0zCh7``zaOcMZh1)TiZ^K?s~#AP)7FWToRf;q;yb5veeiJBhb0N
z@<9-OR5av--tV5`dE*_k+gXoLYk!S}inZCE{Y8^iL2v&tgXzYG=HMF^W8{F;uwiI$
zf$Kd!s`z(JfaJW5y{9HK?cd)A*xsu!3NsrkbRoA<!uaO@{MEc+2kBGKF#^kZcIydd
z_zdYn2Zs+ABzL`lA<_UDsgD5;O1~*l3piM@C%HHW!8lZ5d119>mALT?SSQVFStL%9
zFGuSYe>G?*%)Fz{>(XTE!jS0XkWz54>2NQN>iqB%HHVOkw;kfZf?bhWSy?-}AEMg#
z$z=(Z7`p`0vH7znON2nPEMZHf{l4kjyElO_Zm+O)AVHv}T!0QfF`8nyw1w{r)$dfa
zRXczjwPcBs6|8#j$zQj&P`9=#7zonkC3fuNjmTUXdiB&F^JmY<Lw=TvwYRS~H7m>5
z$|=ICE&@E~p4GR$ZsM27rc5gjr3~^%^w~5B&Gn*OEj@i}MFfg<XDzifA<}!y|GCcV
z{zs-1R42C%B{Hgfby{10Up7csZ~GAnuEZJCjxMYFhBNs?z4T*=cI6Ol&@^@*)=TsE
z#hTW@cmgSZ*)pRX_};p$D{(2Nx2jrp@rIqP>8zfn9`9g@!s?xrcvxKi_s&`+4yAyp
z?am(M(Xq11X}mDgmkMgtZ#H;4ASdr8<Fy8HL%Wmn6`;TDsaOZ9;1IPH&fW+$o-!uG
z)L5yA%Ed3!wv&z`N|S}_I__CghU*iIENsfo&pr1hapX&lcF@I+UnC6SNnj{-(VVx0
zF^{0^os^Um03<KyG&VGp5GjT8Ht#hCZf^x!5o>pbf|-00QVi-7JZRKluNK9D5U;-v
z^sRRyD?`KQwe?<?JrWH5Y5~2wm6yI*le9DEwX&`dvD&Jfv!h-x-Td07Xu~_+x{XVE
zV)O#?gWnjcW~621Mb4tR!BL@9Z(URt`XMiTgdjDTmwBbPj@V12pO2X?K%C`@K=0K%
z(LW7s?tO3XE`zZ@jgAQM@Z`fpEB*XdT+zu<h@1p`#V+<#ty@X($o3LvXAkazf7t2q
zC`DGidThLW+<Ao4-(Jo=BXQ^&b+^-hfxHEr5`BaqYAbVR6*KhjTJ@glY%(Rmn~fg>
z0r)Gw5eGhA9@0B|hM0l(Oz_cBv6KnMY!#*vc^Y7EA}!bEvp3Y)GfZUtSR0}9Hwdp)
z@sS;n$X>`f?dnvqk$AbD9cM}v-IbHMXXv(|)27;rD_j>A6}>Mez||+dQ0E-8&Ahem
zv$ZG7`x%rmt+>~3o2EL=jfrdNeUM-bOa7F^^I8Bci&Spl&33P7;@Qa90ha*wTdrkI
z&>xv-Qa78Q(-YoH89Wmks=MmEm$ahhr+~2KP`e;7v5QMH#S*F8(cE0#t%Wu>-kIrV
zV!2@cWFI>V+Ln;a?a2kl!v0nVC-r0{#};BY{qh5YHc&Fw80Ur8u0+N~V*Fd~fz-dx
z+U~8|BJ@@$fPS02k6GJeYdBFF9yjUi-j73R?07ZJ_jNJ%h1)F`N&4TjT*mnkZBHIU
z4u<i9pBaDi&}u_VSa(C%&W5l_m)(9Qw;fMB`}38ra$jp08f~cu4EX>=3qeUO>~;(~
zKo2sKN;{`h*wQ1@NI66mw39OZ8fIRju-mMu$BHwQzhE_y5FkZoQ~P53t>Cbcfw@P4
zGCmel;g631oM<2pzpYHV*8Yf~$^ex-LDJi#ul6m62u~h8ycV~HueCC27gVHq$(UX0
zU_KOV0nzS>eImPDwR(9%G&w3DacZZcpo>uXj3nVIkk0Gsa4ou_tyoo11w6PBBc^gM
zoc`9T?p;|9RY*w4SKPX<b$a=h3izl#PBz7wXtTkJ>GX1M{Mq+vsYf#ehAuCYlKiUw
zu?Pzt-<0<y3XYMC0cxQwe*B+HAsZXpH4K8jTE<tr_=Vw>IKLDzA@P)*M=lt`6v6Kk
z0VI)A$u12oKY7qzo2CS)zX_tPWCjOWQX3{uoy+0I77kePuZB;O7^J}4At+XmfvMOZ
z%W6?mN+J8WOw)o%N0dtOJo0;uY$WMJVIQ~z=5y1orS9-PuE4Un0$ht_06{bbSX6vP
zJ_BE2Jqz>Ls6ZNZN)d`Yvok|`KYrCr_po>i6@2OWNoZ~Z^Zs_a9kNj!ps9K|apjZc
zy)S>n=!;)yybi)sDV8cw&(SMWFV;L3j<Kp!Xfo(^BsELB<Wu;8kCIQw!06C|$;`}b
zV`s;~Gvz7&Y5(7$x<cOgH?f&8*fh*Yuu^k9`1ugY&>=9pHwm_N&J!))f_Zeorf*0A
z=OiX4DHNUP)QlJ@(Au`x`N<zboSF5ZpPWKP52efT89`hCMKfw#WDD`+MT;=^j+2S5
z#QS+}tIDenfpIy4Nr?d%mVCXAhwnlxUk8c8)0tt>FD(^d%ReTnf6k=O!Qo23Z^#aR
zKM3ij3w7mPc?UyZZf)b|aHR6*&`~PW=AbZ#@0OshDT-bNs}l5zs^c*%V~11OAZe3t
zI*_HI?jM9hLy8%~&H<ZPj6Fg4bCBSTIUPuFB`Oe~_H`0?4Fn@=hYbTFpc|CmDRI=Q
z8x&XyOj;(*!5-mI%XJ+hV?r3sX@Bk}B`HbX=2rwsQw@o&h!RL|tiBko{z?egRCY~$
zVN)z(G%INeKeF+<*ZY~BX9iYWOoSR$yh&t5;&61{Y!<Nbnl3vwEJJ5z>!mugJ|DmB
z!@Xrf=9Aa*gWbrLSKE`9&)s>dlCKHJecG6s!{(~-$W^%;c>eWhmO$wZUqk)v<zPhX
z(o==l#S;pQ_Vz9IE9@7M9}7a?y>)_8hLpV`8yr{=-rc2Iy?-IItb*PZJNxtJIC*T~
zb#hy3Fth8?1ei$MnTm9!jjuSnV<!hRy6@twQE`<x{cB|#%u(OaP`lwRhArDuWdu2A
z(p6YD*iH&1_mDUv6LY<v;AC$dDXnv|M+QM~n5Er4{>H_Zng50}$Uys%2+taKOETJ%
z+Utmgu)#W++!qGQ&a=QV3!QgAEcI#ObW=Z@?^~W~80o4wfQOXD7{7O;5ViO<+{Ve_
zQflCJ)5O9{uez5radtNUpwHC6vqCSLI-rk{qqumuP*-Z7Ey7=MQxAu=KB)`We%c3@
z)XNV{B4(p6Y@$k;Ge;e)uY+4*^W$5AkWHFa8@Ez>^zkIu42rGIm~EGV(zQJ$G#Z`p
zHZPWB_+pHUWemDA)of%F%bTUrDUE^gyDs+od*onoY?Ai(<9dAdA>0EqN4&W+%K(0~
z((w7LY%AY%bY$Q3Zuakk`HHCLbxj&#M&AyNG9nkrKD-+s#11UM2cad9k{Hw2c(5t^
zrOzR_#9I2Bw6e?U`^w{(`s_tNSm(<#5eUBhq@%Cb%Y%bQzNl*IO*qJBnGxHIsjjeD
zuv7&B1Da9BoCSPt-;e#mo@qgoR2giAK@GWiWkOv4{G8rxhR>-GFuoKgRIf-5%Y8Ps
zRH#jsYhb6wm{^G$tXm%ET3?)~%u*CpeP94`PFwR#6+w@owm~n^`duFRx`;;I1S1%L
zMN6Bt(rWS`hdRm4oo^Wc)&)v9u6Iweu&@~ND%(_;GUO}CLs7+!i#3@U3gn=u@cKJb
zB>+8*d<Ddl@i$?yrVH$Qe36Q21RxZTbaQi0LyG<RY~jB>^vE9xt>g<BNa%Q}Dy0;1
z36+ZiYz2Nb_^7Zq3A@)T%Xi+b)MlsSL5ziHQ5uF$C!`e$NB+XRs)lG;`eYJp)_G`;
zg(o;97&&mnoAksn#Z6LvT4h$Q)o?a+{^g8>#q+)^!@7{kR47ov@6#md0)|j3Q{$t6
ziRTIc>hExPWq&3dh>(4e1f<hd?Qk1Z#}-Q{fc$QWwFmJUU>iO(o$U;X^9x^1`(PXv
zL7mztPgdOAFMUZ*+a4YfCv{$kwYg~7`5L#kpYQL*s_W*P6tJX&)uBAfj)uu9MUth{
zombvCI02OH-wbmZ_Ek?k2|iEz=|yS;Q_y+evx|l!WR(7<<c@!>Y^`!EVE?Cf`90|1
z6EBFWHTlDYA>lBXJ`NBkN1na3oh1<SdT6}`QV-(LlK|p<iKU6i6uSDNrI@c^l17)L
z)p_Jz%byh9LZ+adYE>#<Fhlt<Vq}C>ctuuWpp{IyIh{<fj_E8<gNM#-yppwK4Nd5f
zOLCg4lgSv0Fa7)@A$5@Se2IF=^hcC@;|bT6{E~p5NgUR!izBvj?+sR{<K)?;tmc79
z(sPbg3D2hC2UfY|Nn*L5#1u}tHB_DcE(#2S(Ypv;TG$-Dzpbr-R4`Do?HZ7B?wrXz
z6Chp4ey}Z+Mut%y+6|jlqP-(%Uf#s()S<$O92hVT(J!-B?fxesb=O+Gb!x+kOrsLV
zoXAII>gHe};+G`XR|zZVSS#!p{TQ!apa8t?S647~$Jl`0huP}=Ak5lXuj8-OkT2tv
zDi8?JVk0QP)Ynhf3F&Zf4g*Dp=fy6Cx&mM~u-|7zq7t@dsEINz$=fNg9X29zSm;T6
zPnc;wV{CDEt@7X&pwQG9gl93DxpThFq|vZADy>>QTRi#Bl1fUVc(FTDl5{dxyI6xw
zz=nlYLniKzVs@Sy&fnc9ZMl52Qn<hMFjuZ-jQ%E4QoUq5GoxKR{f+@MV0qt=FjUH4
zl@N`F>yZdv`es@ImrAJ>@qn=K>?3TsI$R$UJqfpFI-5QoE|xAy8feYTTevfad{rlw
z8_`_OESOGxw@TXBbk6)rB(4C;$baGT`L2)~Gf_r&CqwF{*6hXaU#D_(uh+d1uPj?b
zv)%LU#|nwFO(v%swp#=+A!ZJNiVy%#e^KK>xzOEZO_7Q-?WTr*BD3dUsk(rd;k#AN
z?R1fd^8??*aDopppKnYYuOwqWy4o!J?h+mO7@LD53r3^gW%sLNzf`E;PK_^A8`qlP
zGc+AXzk8BU=`}O{FtJ(iAxpBRwS4Gm_%mpw;moo=*MG`m$xa3?hg|iPR=z=0#$oUI
zdF{54?;*o8W1v&tr@zhnyuuU^GT7^(>a4Tfw3RwBhAdyE<YsFV33m9V86TRCcCUD;
z_PMi7xmiXD=#S6RQPiv^78=@=0h6GqVltpAs~^JaM5>SH7wVc(S;KkML~W(=xBJhX
z=i~H*I^B5|0{ch9jhaS9c2m=#^HIIZo`5r5>8!V>K)-siA!Ok<Fh!eqk}LJ=bnR8C
z_6%O&si+F$oyy+6W##=y)Il?u#f1^^>I&Xgud~;OwEPT`)BT24@v|QDZQ-MX9pSpP
zx+X5}&rKzt^=C(Jr~F}eJS(z(7oYAfXIotxw(o|X?&VwG4AT1Cw{E%fnYTXTdfsGv
zzu)L)E8G3nkf|7cxz%#+b<?sut{C_=^@Wq<r*<T;ERk3TqJ6}ws>J{hu&Zb2N6kUV
z=*T*hhUjt>Vi`&jxh#QeE#xXqL14ikOC=~x3K0%2KSw-Th^<7E1Yu8*alcwkxCYbf
zK2AqMD%()UBAFHw4PrY=CU8(S@q$-M;0Q9Nv)R8U`v!Q&B0cZQPy}97d7HQXVV%2W
zev5)!*KB=TQuv}jo`v1c@h;|(@KP;U_Ae|Mc}TdQ56D*8q-$G;$&~hV&O4{nKkt;G
z8@SM#`X(VS4U~2-S45UeFWh3u12%Ms97!jSTu+26KG}ynm&+eMFAsVKuL8O3KAW~m
zt=SoMI$lK@hpF5#&)SLpYr`~ZYHXFy#0-IU1EnVNHx$j@t$Wc)7I$f~7JlUR*zT5w
zEYbW#=JwCyF{wCo;X{E`w3HZ&cw7)mWhM3O3`#ued?=_+fgQp4C1QDBDuqdbsoihe
zIIHyv&nkY>#?H3N?a4UAs*C4}%;Y>%to#1$BX>&}CzMbU2&%&pf8TUA@NzUE6e#BY
zB0inDcFZqY?7p9!+hdn<E1VGz_9)c&6c>u=o$U$hz3v8DSiK=tQV8!}nd{_>VlG7U
zVvMC@#0cKd>2~~X9AKGt-Kq*L1ks`o)ng#(sl@vCdC$6bw~9o4wh8TKsl&WlEEws&
z{THxcjWWwy%$!K*%Yu<esb0A!4v@=d;oie2+()xR{YY~hz`E#?lBCh`dey!mT;kOm
z>2AXqr{clu4j<4)7tAm%q|=pDX-I<6Uo$oz8eQK<ZLtHYsh19WFDy28ey}{awqgI}
zs8ia`cd2aZUEU$c8`*~jG-lDs)h$_@D$UOdqFlX~Nl%5Zbhj)&i6<O5Z~GVR8tE9X
z-J^^yFD+irAl3`>WaDT+TkkXoHhA=S1M2X^6Y!bjCX-|G3cpKX0_Njck;jLEd;*P}
ziD3EAelO7xBvNx<fQnBN2u!Z;S@D=P)Nu``mbLq#U2eU<$&L!9E-(o&J)lyfr+@Vx
z-E}@qMnxVzu2na=2sr%<=e`L~rMM?P0(nvEXF=LdJVrh(44ctxFL>^{SW>qvr#jc9
z+1qksJEt_oItt^P;5x5Woii<VKiz)vD}X(ww>`jHk32t>UwH$qjm48$K+fBmD@KOX
zl0M=o17bLX{*AsB<VIM<fqiXG{1D^g{@K=YnBTb%;s8FKM(&h7pE%%=5K5B9eo@1R
zb+LV0|KZUW`#r3pK}a`SPj_g5^ZGW}&IWAU%P`YreuG00IAQ^u_esX<XIMJB@bXWl
zYiVbHI<3$s(-!bJ4Z3WFk4RV@T_$ZiAKKu^nLc=SQO4VUjk3^;m_Kt~aI_Qh+RezE
zcGj+7GbbjtBOj+wu}vN4HTxB#ETiJoEtomfSyf=cAZPERdYM~p;W~f_*LbmZ+CKKT
zi=ZfbV6de=pUiFOWY%J<ZHqO{-b$7T;&VP8%Cn-OYR(g<Z{H+;h8&8xT@B_Fro)>_
z`!7tsdPjwr#?jr@scv^BXO${?317Pq?Q7$TVE@U1>AHrQt6(YElR3b3n-sw!zYS6c
z_j_PRWNC;)0$mZq*jvP4efl_=pup|0Ge5sM{~TrAcDYau%d70WyS7R1lQ#)8!~U^9
zoAtJTk{ViYAZL(f*+S0RMFfW_!j98iIRBf#jpWJBdEOI-B+y9N*z_cn1f;2HDpFBv
z$rI+!g-EV`)6vuY^3^YX<cO6;L#<9BNa4?dP{*HZpiHSa13?fE#~>daFFrKr`}D|n
zb}{dyiY!C3GLkqDU42BVn*L$^AgQQ#LO~PvaF$pfE`PnlAb8LVQh7edB3XdtxaqI{
z%S~cNNc67Pl|#7sg{hQ!=5nZ?ai&Na-N$!l-Lq}ItoDt#@Jf7%@n6$O+H)2gV*?mW
z>B075mo^whkSzJ8NqRIhof?N)DQ{L-96y$3VA`)TSiDXzZ}JFAW82$De^a({YwNqw
z;I{#0b^AAcOZ*@1C3C;>0*&J#zc{@Liso+Y9F-Q_qLxU*l@g+?#4_lLsh?s~UIuk1
zr;Zj>Pf=1r(6H*}Cju1YgET=Z#F7g?yz8MU=vlfuiuC-T>(Q&Tq%Ou+eh+p@lQqEH
z(FxD~UeEOTvziQ%8=g5juUDkB(eE{F!<6}DaEr;uwdjV+UH5teM)#S-8zhwmm-9mB
zS477~wWBRQEs&;XnF--gs;F(Dnhm<bX~Wdn5IpWpS5T&c<KG{KUb&UwOQ<m>%TFfc
zn8&*oDIyD`t(^A343_`5zA)taa1lcNYi_NXp|fJq25^Zl)L@A+B6WL|q+oDgP`Ehs
zLl4s$YXyfemjqJ|*RqtB!W<;<IfGb7y$wL|eLS88d;LvlU%f}6sng+weSv!U>Fnnl
zYjPp403J@u)Ylu)Ofe$&z&pIZJCUESQG~AkuHE@@fRV9e(bI3z7%?m<!Acq3!{y%?
znK8z)*J7TQ{J!{_riN&k*jrWm`o`%)NDFlVF2Mub^r{ZK*PPA1=@A;e1ZvXwl%7^s
z6>6#Y{y*SH6Mt(3T<?K3J3HYUn*t@5|E$=b*vnOmM)5VTMmbJGNak<%12yka6?H+b
zsa7IX7=S^(Mmg`#P#UTb+sfC9g>>(Lh5niVc?kd=Pys)btg|PDT|}r)#F>+zZ@IUu
zVZ>p5KJ+GX?2)##6+LB2Djv!ip~Qr*eS@rMX`h>8KzKb}1rpZs&!qa@$}ekI{0Qp&
zu<k(>cX0a9FoKWQ5$~H`mIyXbN}A!jhBObf#&MNXgFKHpUPIflEi3FfxP*r%ZeEsz
zhzlkDhB|Xk9e_#uHoFHb1du4KPA@o8ky-d*IG+7>@gN<3UD$CQHX0vM&$X;z+gNaO
zXrTAwc_IpVORVEat3dXNo1MUMqw6Bc^65<Ll1S;9_kL4?_*l1KQI0@rfqB!0*kcZo
zGL#fsXpk-Ht>MCxBQl7VP44^RS!k_1o3Qh>3=cy}>{_muWY5(_G;+tHv@tN}TA|K!
zZ}C+6jG@RCQD~OR)8GP8TN=7HEh`)?vUKB?_XkZEWBHaZjx(P%?UwYvZJL}uxZEqM
zA=FY8Exwu*KJ`vHSQEjDyRh$@*&@(pA9#;*%KRB)@d(#!Q-l40OHdYXd8Q%y`Qx9_
zR2Be-7gBvoAYeD%Y15mhpEB!_OAt-wbJtnp+05RQF@Tb2`cTAV1@>OFGNec!Hh73I
z&}&;Are1I~imj8sC|jwx(S!{3fAG5$B&N4Hk?0Ly;;qvV`g;lIQpv17=cF43)9LjW
z5WH|dORwg!2;|~J_87c6Hc*+t4p8_Y?x7~gm`qp2MG1FVy^y`T?0&jtPOC5n+tyxe
zwe7*TbJ-APTcCm%n>_=ixD73c>OQF=|Ey6bG8Xb2%b;mz&<_ksjE6kGK9&`QW}1F4
z@+_D{eK8?j<4pX@#ob-!tZYZ!I3(v$Aa&z0_(cYMpH??1Iap)%@2~3Vr7xRFd8APQ
zZG;a-&KHl#b~~m)nF!kj&~(b5Hwi3}Dw^AmMM;fRa`VRkK{(rnRyrJl=u-e>iHeP_
zWX-P|fI3_JHj8fjU-^DtdHB@UWBBTUz(3(bY6#Hym+SWs?%ugSgtWACoO)ZbewQ#(
zP<^&Dy-Kz3HfisKFgL#nm3$mihs1Fb3H?{RL{L?&xP@c9Oj(dtdG(sx>-u1xADX3K
zfM~o$-Ip>X&X(b8&B4~d1RNkADoMDMlK}55Gdfag@8ea-5%k5*E^ezJa*E_zh`w^`
z<AvAb6PwY$?Wc{P+)~}`X28kOmxwqRdCgDpTq<)vAaSQ3ts>TQVN-BbmYIsi(5m|C
z_NLS@=5DlUu?_0<wC@3l<`Zc43J{6cVlx2sX`GQPRsp$_`9R(qBfpP?R(dHO2ii6u
z=!o8AKs-RQmmx1vghx?`n3&wc{b$Jgt<ul1S09(wSrP68kEcvmPa8Y#XYPTZ^Kh1Q
zj#G@FcFQT9Z`5r0;Ndv5dO1wF<H-c}vby=U#vUd9ABUcx9O=Vvb6~nJy2DWr{BL~4
zbc|Wwb%^B%Pnf^zHTbRg+JT&6?^+#S)s@J%z!aL+V+VtrYhXqFZ>y<KPc7~hRg0JW
zi_p0ny5wY<*|BQqBRUcV+1;`@+7vC-e(^RO4C12ID{Z&30~wMt;~kJ9kz=9LfqwLZ
zoi>XDJYP3oi^J8)>*<A))2ierKTccUrsV+tH&{XqXJ^$I`(XFSM7QmIu*cS-5IgGF
z9&^*|r_!^x&CCK%P2yGa=d?D?V!L%KtAl~b@E(yy^<*)=i1Vdlk8z{}TM~*_ZfDWz
z-d^PN@U@b?xrwv=9neqH?w-?0YXbKo5s#g}yuX~y&buzL`F`~B9%O=b7SYG3K@oZS
zjeoBErWpsVxAoB{Yly)G)_9-#x4^Roo#^3oVeO>-dY?|00vLrM$K3L07i-w{<<8yF
zZZYP<^L5h+oCIikx^eYIQ_5(=C#4S?mR#WqxxW^1SKWl)c(&f1J~-bz?XaddqYaOZ
zeKz{R^w^5~w7s)aQIHzR=lAB}jqT&Y9rMC{-PNbs8*2pOX2F2r;$B0Ykv1|i5*a7T
z`bNe76JLTlvU04js3<#rV*f_1_NIY8g3s?L=j+)l`Tt(l^x*et{}BHFcrT;*##Tn(
z|9(}Z808a9g#5o0j34c#&wBF9|D(s$R<*vHsQ;(fBxE>d`oLMt{9iFDq5GR3DN}F$
pyE0+@#!^%IKNkL_(aY<nXL@m#PqNn#@@L3LURp(}T;fZ>{{srSnP30_

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_011.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_011.png
new file mode 100644
index 0000000000000000000000000000000000000000..97d13af68d48ba456a294c556d29e67c57090771
GIT binary patch
literal 251321
zcmZ6yWmH>T)HRA1El|8gTil)E(o)>5!QBJJT}z?Gi#rrAS}aI#3KVE?4=%x>MGJv&
z&-32*`^FvjM@BOCIoao&WbL))nse?i+L}s4_|Ngt(9nofl;7&2p<(!<p`lCT;s7Jj
zulUA*H*s&p_uhK$_TGM0o_1&&R^A_6-Mw9%tQmdnJiVOU-Gq3>c!jtb9lgCjcuDZ_
zf&cFbyzZV3eB*L&_JK+8J}85{(9npj{(GTU%2qg`p`)Rxyp`AY&pUo_N@7;&SdV-7
z9sEVwv%01W8kl{WeUKeA{wJj)lL`Zi6zhuSvgI-{5-kKXBAGpcL`8)v4eJ>OaZJ;d
z_>5X@yMwKLO*b^#JS#WmQ-zY@+;3Xt8^&5m&X=!(1;m1m8+slJ2czbVy(~`0LfZLm
zcDq+Hk)9SPHjACddae)uzfr+m`o-1%j{l#N?=d-UdQbn~@BHr=gL3;5-~Thu|BOA#
zjY|K3{@=NQvD+uQrxC0FpX;5D$tHUeAOD|e|KEFvNDH#``@fe0#?FaL|1o$b>&B+0
zlD#V-RruU<Ch@<wqeN9J5;quGp;h>DceoaG=CkrYvp*J&4ez%SWB9VMu?3%xu!Vgc
z{ogOFI|SUE(oGb~K3002{_ojg_`7c|;ZM+^Q;t}@kK600^?hHA*q3|vTd|LK7NHY8
z4?U=)`#%q)0p-f0E~?M4q$N<viDK^@8EUb*_n26MujgcX?uHAuo-oOx{zR`oAWS4y
zc^_B99%Z+CZrwx2*HCB^n43uDO@nHOh<Tib$hWu=Bt-&z5g@DFvxUM{>o6C|p-?tE
zA8@Ul%Uho(5p`~c#rj`#tLn78lth_?any4Ik=-%Lp*|tRT6ZrTO&B49Sg`t*=wfbd
z)o*E4N4#ase=vjB2QD72yj(kN&DI-h3chtDhD3wkermV45I^agYjSLNd!zMIN^Z1+
z*ZitGlrkZvSD5?7$wBvVgYl8cL+g@e?v<YE+|9hPyOcxIM05LXEvn{>1*>3B{Dc{O
zrX1mrkY*T-Rlr7q%S_KWPh?_d!J!Ch>Ea|vJ0V5}fpTsR-+y`6e0=YK)Bz44(m2_S
zf49938Tp&L{=>La!QX3=M&gEDtx$ohxQ$WOvgO(D%pA?La0}E`*sxX@%IKI*>L<Y)
z;SM*4Rt?vOspt9u)rGE5r`jpih#wQ>+RO6>-}F^<*qb>y$Q^lfj9bj>GfZklk7izr
z!UR!bkMPHk+IpGtMK9Dt&*P0Z)GW4m<NiL+;_+_b9vXIY_HQ$W;a6ik_n5XSQxqfN
zn|^-aU(m&$5V|{z<Lllq>baxnKD*dXl0Ap6KO|~a49GrIJf>No&NGisPMFGm-aHr&
z<$zas_f9&-3bqrG5SyHy))Q7I<iec5b=Qe8TEX7j1aGwfUs>lBt)IX*xBS@R%B!>>
zzV#ctNVJKI!?PM*A9w@3&D`t8Gcefs3^aBXbhsx|Qh6EKRks&lGT}bj2%m?vNgTa-
zzl{(A(KVik=LF8lB$+%ma*N%~cXh)NF!AyL1~{rN(bI19`U}#`#Oc}wl)+6&psA_p
z|8NJag<_fUE{B}*pC0Nd%-or3fnD@r4+~*1vGw4AzCLt$l4gjV7plR>#YN1|+@rAU
z_Wru1mKd@+gR);wW!|(vjfLIH-<^C6&p&Ca&|=s?%7}^S70<<%_*MKa7l;sJUJ!rA
zpci6S4q_tlpM$>T`ZUQ;eZ920s>lV~-4#zA!vl3-;NLyS<ZFIbm}7K&$ivx~BZ3$A
zx;|-N7nAQZ+t9>~sAScpTErtV&UCC5l^Ef?y=oy>pl-t})(w<VP%gdvT1?~(=Daq!
z@~As|nt>?tra0#LVT>+WR#Rk<dV=!<E~C8K1GdQ8?mFbd^etukTJY8%XR?_CZDxq^
z>MrN>0{U^ASF!oj3zIzO%1tIUAQ53}4>vbzEEBW+jUkG8+{?>isTW#WbIQW=IURR(
zZK5tkSF!us%F3SkO~Sk7!B2Xo$rh9}0UeeiCSQlh^u=HO@hq*7Cm_e*kl)Km<YO-_
zK2|=bopv4hJLN(z6WG6olIa>J$s3HpCVHXz9mCh{v3*;^g-D^q2h{R?E!phGODED9
zEyyn(9cKrE)+71@#guB@RO7QUqdL4wqDVSCJgi=!7rf3kaZ$VA7FpJ3*}JwzkCvI4
zsm@B4!YuEi(X-A*_$EBCvjf})mJ;<*)GLQ1h}^8^H?KcAK7{yDe!UJo-I~SVxtS~P
zSyp|K+;HBYF${*rr0bNC2NbLaX564d5j)+rbSDq32gW!4t@*hUb$IwW;Az(jR<y=2
zXov@uvU7DDGG3l!#W}_#(8M*;?*?hlupG)ATa0pzdpI5inKbNima)f13TtG;?BD7%
zqbo>HR7VJepCSKsYf(pgOVBlPZ}0Da9!_FtYHd}jm|3dy(hF-r@35E74g59daouF{
zSZ?l;LUbCkiWEEg%k)2i#-mD+rb?>FH9kK6T1JLROHGOt<2i@4_Va`~YQAJ$J?Dg>
zA7r2m?F^cS&|lkZ$l%-5y(l`3Rz6Z)rYNhorD!@$jP~>b{quGImdT62$zXhYlVP{t
zMHo?(j-p1jCkn)F>-FfL-l-<c<FBG&Y3!Vu_khFU2BBoV9mliwg$BKPX^t)CDwW36
z6jg^@1k)>b&FGpJ-G;2)XPBNanPuO0HI%D~Sius-ANwHTBj3z(*8+E(PV^T(6eI>B
z-q;Jx--ANxpKluM)g8H<5*x#1L_K2#$DT9k6CJN~w@d3N6K=DKg|2JYz28{U9$h9R
zWMWZ><RmMei`qSp(uI)<4seSQzA)rWFnqp#fs64AKjF+}`sLp7F&*2mmbjEu>B8Ce
zr~JU*OLzE8qD#4|Uqq23`N;-E83uFS7)<0YA534id|wd#Q0{Wu*|q=U4BU8EK`tp%
z%)WJbnO0Kr>RhURiJD_K+7t3{BuL_vqnp9<5L&2|ME`e?%7O$7z02Gk({gD~IBo%k
zy=4}cZIA3EmJ>m@f@iIXj>GK?n-gVQ&vF;QGk)NSo$j2Q4_DtLzrd}Ly_)59ri+;;
z!w?7G!d8$N)z5Yxr+GgUNs+5yJ^NZhS1t|91=q8dM5x({i~G?45!ZBv$Gjnxq*i-~
zP5nY%g;{lsw4VuFxZbATB7<dYH)YUTUWdH3H_MFVy`>R3cEO$wIktm?11%k$hrhpa
z&w71C4IBp|Z5DEz_`(R4BQXWGX$?6Rr~E96ECzq`HA03d-?Cd8rO`YW3i6R)8rVdW
z{Uet(qGVJWiPdZ<7g3jmF+t`i*l4X{&g|VmpT8Df7<M5J4GNLSr&p-N#|jvVXXBo+
zePO9*nG>H6e+}NK2DxDh1?BZNy3OUoq)ndtd-1n$)plsQa>kc$IoFS!I5?Nww@O!F
zyW=`s8EZ7Z;KI2OelI+)FI9&z9oV&%qN|U=b!fn?OD$1{E|q)b%mqV=Qs}^Ar{~wO
zqgv73z9bXXfXLh_RwyB(r{Oc<0{Nm^vrki9m)En*)2IFC(A??{kA#l>3Tn`{I=X!$
z8+}M1J<S5{9bj$Dj~I>KgU$gev=%Ra;+I<SGz<C&+UORgixRMr7d+{%&#up)E$rMq
zXxclL^^aokt-fBQwQ<?4Aw3j*@5sr<AthIm)_YM3+D|W=`Pcv1OlJ)v8umr=gUMg$
z$!#Y4JR$Dni7(C5n7RH>345-gkjBD2@qhw7a_oG~7@$J_S5#1_P%y-YtqPoK4!>wQ
zh(Ww|k{rhVdg>0dX>DL*sFDRsHlqs}nXm4-W9mYrWt!>11PV82tc=c>rQp<7?5Dnh
zzDwUkoBX`u-H#^96P+>QM6c_>q%*Vxl&?GMVGw~}-jexiLJe_n5YCOb>8;mr6Sm{^
zO1b#*iIUM7xuSdu^Y1+`!|J)(z>X`Ah#tD!YM@+|yeg0m8v6Kz7<b@L*G5x<xEFkK
z)DxcN3!lS|a+8JL)e-GO(7T8^7_g6{_~UiLCbws_3je~yEQH$g$Z9p^cfzeW&{Wg)
zs>o&2$Q7ffb2iy+A1!}bPL|C4-S-Z;w=8s<c%^}##`){wcNUkla}!Go-h8Neu7h_;
zDEJIK(QvY8EH}xs*Ry`vg!4QXW9Hhq+Q_R&m|{KVEwaVThG<7$dPM9=dEH_<<Y;NT
zrcp?J#(1jTQW@WBF^>mqDhoAk)iW^({v{My_8|7qb^3ayJz7#*@j>e7=qSNveh;L`
z%t_V^8m}QHb6iH=&hM1PtI6xtiLSSRR@gsb*p^X$BJt@iF8&4jU6(yE!Ew$JNuiMD
z$TO8*Us#R>KB^|2{*hSVPCYahr<a(x_UAH{Z~rppU!qev9<vjgYpaE5T4%C7Hxp;Z
zqQDVIR<X<@cTp3Acab0e$BN?{<Pr+i@bw3{Y~JHRrZD&%Z$0>e$jHd3Gx+x0Fw8i*
z_l^~hiTY^_fSu+L^O_YO_W}S&HZ5EWi_!?+xV}9>^_UHVJ?Es#i3mlPj={Sjau;<H
z;2Dq8c5v$a@TFC1i%VoAt9A&(exv6Oh6L4n;lgjQ%jIBqn@xzumY=lEj5qXz24Xfq
zxp3KYk_8@QF<5bMitH#)^y=$0ai?N#k<Oe$_q8%0$G*7DyA3waT{gpUH3@2q-`d?}
zQBqR+uP?X^?8<5Jdf<oKPGXn2t*`bI-HB%J95&d);e$N{e{!~muqzpiW$+hOHn&v@
zHK{%S{Rwvuz16dCVrg2+Gia70uexrTW-kSn8g*jO@d=^-^*vKz*lm=SY{-z~ic%o{
zrjPQokL4xoC2%@{=ckiL-eTeTy3ZFPEL;X~MGS_K!OBG;ByOVRj)O0*J;Bvb_^@-F
zi?eKPi&J~o#~dyyqkv%CN+^*WEo=2Tu`B22E1_SbQEPP{a#JgGUbAW!;dnv1d1M@6
z*UO^$Z6O|+=Uz{Gu)m-2X)qpyin8i3nKE{Y8m)KBQjq8mC6T`>)J0P$$`6l4Fhe(+
z^aYAw27gY{bJ0iZQhqG_SJ+iP8wzdB5veNqH=4@2`GlAmtVn6?UNg!&eqv)v%*-pG
z70S*NqCRVwIat9M%bnO-)%^v#Bd~g47fu<akXI(bmzoLv-1wB$2_v!mpjbn6trhoQ
zmNi=mkSzvch;Ik^Li=1rZ6xyNx#hiHUW62Ge)x5I$~#dYz1=h+11v#VJK^VZ8$23%
zhSe0nIySE#{_vWaa3De8B=Rd{5M;vOnz(tr=alQ1v*3L+c{~GVoQ_N^*L#3465Vvc
z?vL-lE$PtS7y`t~1k7R4^q1SHFElar>L>h?#|1JBxpt;(J(GFsYbNwKOnDQKi<m~V
zt@;*(v}AMS@?wJ<r!ObKGY56Fi-Rf1T)U%;F_!8%=Yg7~@{gvEZ;Ud3di?(TR{(%z
z@x#C~Xo5R@&RR7PX4kNX+9lb#cx4Wb*eyzaU)UHJjb<4$Q>HYS{D%pT4dQ}D<=Hn?
z*oVbkHjA*K-Cc^Y-N>d9@^7vcE*)LrTCxG7aT5jSmj{^iGAmXN^%MT}NoPT6RcNYG
z3@P4&bGC{b#Qq;XC>PcRysum&u2S(cV<5Oww*li9bIkiy<<ohsi@h+_?|Y^(*a~i$
zl{$rBy6N?~=`F%#dg}GuYyI;EL|9)=z3Y3PYEEEc|NQ9Mr@0Q+YWF?4|3{Ewc4}F9
zSN{(+YF|!ya{mGgvo-HL!*ag(UVP+ZN7rEyRs}_?lqX!y4FZp;v(JqkHz0(o>m7Bs
zOf<dIro?)l`ALH;E9k`dyGi82_X!@aCqqgGjtO-`K-X3&@`>~F7E9bMk&2YkuYMZ7
z=gpOO?q&u$bDD-pX90|Au-%=MnN%`*so)_;;kURTE)W4BPbvo;2$g(YR|TD7efKxW
zf33VwH5XJV&U%`Yg_><Nu6@r<`O_m_Hx>vUBa4thTS2RKQ%?_%29Gnh*0GAh(9t+5
zKL7I_JYohJTL*`j)m5|L_4>l$3#n;gJHhP91Mk<bUt981yGj{(6}#7Mz%NH2)7L?%
z8vYhNA+EV@Gd`zXVw8R+hwE7U8jnI$-p$#6m(e^gLgBsf^Bn~MR#e(o<v3z@G^s$m
z4WatHJZ}#3$_-n`{sA9{9?`>&S{lOP1%q7~V`-49rl3_Y+V}6<IXABJ)!b@ZukB?D
z$rPuCtow<6Izqmf$Oetu3T^^84S-_;v`^jJiQ5(&uRT)DnugtuVjz1hSKg~KD;;jR
z12Xnfq2xb;ZVx|i`*}^bS@PrmiHz6Xqot+)(dzf>%2n0aHeBIXq?^(CpDFYS*?|7-
zVVfaI`j1vInu{WlOw5h!`@WL@B3L!3ep4;{RQX8{{r5t+TI47Tq%pVVj>fvIgNtdc
z9+O*yz0JkvE>gtf^JGpv)wj;&<=;xS`)AI2&&Hv3OZiu^J`YbbV8)pG+Rc_UoYXGV
zPd6Jx10YY7YWm7?tY~Ky!PvEif$XYZOR(9QCYN!u*`(zRxKkqxan8Q`m^dqb$G17u
zXW@(k{dV)eI%HvEJ0rn(SHhG;FT3++6P@_!Q`~+T(CO3>@j+Bb>AJj2#GapanVdhp
zGBJ;!9!PyM0iBxycaBBDQM(^46@MdKpl=~WKO(4Miqh@zH5m=gP7M`dQM!Q)g_?<?
z=Q267PvQF~((qa19NBans~rU^<a&Mn!%5k~KJ14PCHIRZlgFAV<;^Va=JuFr4E9Nk
z=CmSCwh#Gw2o4WHtM1@8hNfiyuqim6EZ48Rs=7`uQcG6d;7F+NTOd(pmr|Y<uc&ap
zJs+#rj(Or8h^fk@zdXN+UQh9BLG#0xE6h^k&vX8bf@6oyrg#%VC-rJl?nKEN;c4x4
zf7?a+IM{9O=36GXT)ksa7uiswdUFp?PX*fHmau%?*pDAEuWxR$H7j-k%*PB{;k#2M
zN>X8{y#E5_d4jX(_*MO}x5U403#BU-72S{j5LGLQ7ozsRuoeB#R}njtzS1T+1UU#1
zQ(qUo;V&j<aMlX2W!=biyWCdYG|u}Kql@fZiJlQ!P*X>OZ63_Ng&UWNI^msE=zZ3}
zz?-MQ|1gkT`~yU*zV2Kr)_ZTpxQzV-&Uy@$`yTXQ4`0L1GGnb=ne>29c&XsGfuFfT
zz!#<_!{GP~f}bZMIf7|>;y?nw4c5%Cdx2KH6bi@CIexE}5(SWW-S_8ePl<`~20ma`
zii|-%EyR##X1hg#kgX)eK1|5Hv?L7Q^~b+-kAEMpFfW``-O);<2OVea#QA>dGRPB;
zJmXQ<*G)nEw&xnxtmin7S}hwMU;CVI2D7vb_Ls<)QQfmohf-m(cyvY<|GdEbkdhd#
z$c^E|D5rtmCZ~pr%+F7@gdHz!uS8f|OSDp~o@bI6ZY<oSPmgaMefVe38QzjSX<3^W
zVTz7UfLO+<H}PwYzFtZQm_{?_A)G1s6wWDD^w;WAyG(%~bxqL*Clo}URwY;bC55Fa
zm#eOhm3vR9JXNUNXfNv9#D`fnTxR0y;L-v?@t&7j(eX+#mrsb>+S_a?Sp4M+*DPc5
zyWRvo&<W{eV$1KBP~ZkP^K(&03A+|*R|-Di$sftP%SLM^+jw~hSv+39LOw^~&2K#2
z<fPdiE*nj0G_5b^=3+a1THR|)L|rTOp5X_lGCJs{tX$w$-`<^v5Y5;^%{ACimFd&Z
znH1gJZUvR*J*MWOS?mDHCS!9_wgox+!(us3V(<4Vw&sdQh9@O3>XcFcjR1_O*~;P#
zXV3Auok+Hcg6R|D<o*)!EONOQALVB~k0H?a1m!9o%uypwdnj|Gw+7W;G!Qe9-b}rD
zRbM;KclB}ZjS{+b3T0b^A0KY+D;_Tkvtag@fzm_U0t1`LyPhgQSlCQyKASm6ySVUb
z<hJqiYYhciE}r=6s0z?00gMuM<&4N8yB}QCO>#L%n5_=nQ8e=Tp-h#mvhAj|+|g)Q
zB&0m246kkNiMmXbfvXKh?NeViji5YiQ3c$s*9Zh6y1JT+G7dX+BntS@&u`05O%lxn
zg+kfOUQ0^8ihb$p>q||DV`^$jqNoE-Y(=H}Hu}nc>|(14xz0J@JKiI<a#-Fet3At&
zPwUEj(K~qz2QPpzd|HK%*7MxW_!qD~g3(qoirE2H5qKVQ#VZfUNzTS1og}Y~eF$62
zybptH&~?0deXEH7_md77#F1d!$s}9^Ifgy2@Ojzv$h-;0jAYs^316*dV9lfB`z6!|
z(}<VIDOT3`G~rpNU6#ZKP2((?#l8TIw)TV#Oy#=ZJcO@yB|k~R34VY%1YgZj9iJqS
znun7!J)oW<kvWcEQ&~AcInyWJoL)&-6QBndG7iUw>O&Q|<#*nnCTdXNTnL8>6QRkg
zLr}G<fEAlZZQoPECyd5cKG>koS0c66PH`DdWW|t(juSr<Hw%Y{8sB2F@Y(L|^(ZUl
zW`*w<+NDmvom&)CRqAJ$)NFY9Cj2iD6crUpr5?Y>$FsBwrMHd>rAiirYK~AS-fs74
zuq{Znz~QaV&)~b*8g>p<#T+g}4^@)uCAjQPDpld%%Y%xv_Lwst4z*8gQc)JW2Qx*@
zayHasIW0}!y`%E)H7wc@zfV$k$}wE%kDgAK(?CX;9pB`xcD&2Ab~Pyz>+#wxt5yEJ
z^_HU^e=OO*L-owm`P_BTS%xLynH1^9H~CCf@z(`bI4wp3U`p*;#h-7539yFv+$Hio
zeSL@aEO=>7j}7F>W3SGKHOq_5H(d4ZVd?GFe~zx74J~Lp$IYl~f11Ysb{v<J;BJE^
zGo<ce6mWG2JTZ`!$Gs(X_xCFc&@#vDPL5sE&x2+5+#j#q#U&()7mgaB(6P-;CRyat
z#J|Jp`o-1NS}OmKec?b_2e7S5O0pd&Uuz6pT`vAB)NM3}OOm(RbHR3&8y4zCgA>#R
zM#)hA6iFp>pYhPxxXnZ>$|bn;3E}bg-sMqAS&n!32;$}A3jioz;8p-wIzIjs5H<GL
zWN-0S-KZum7UA$}^tOb+#Ny9OuI~!=((F_U_5f0t1dEoI_Rm2@;ra2B5Wow8q{f!t
z2Q1?-ud$5*Tp#>*SzRK)o31qMj?S1}=%&KFd`!K*PL(sP26ilXPc-1g>|Ka^(6-Kx
zTq8x%Y~gD-W&m@ksik$h6~{-NG>j$x<7x%b=8RdRF;oxkTy_B8J+wb;AZ}g9-!KT7
z%Oo}+PqU;hl8e>8a!pGytJuW$|1(ZErdIOgA@$$E-chz!>^tV%zV{yg`&YbwMoKEU
zcl`9T)Ofarja%KzbMh!$65MHFv-)GsAEi~aR#n!cr8V!9o@>{JL{Cx|#mFrMtXhH9
z-x!1{U{s#Z;&HB6_CVVDrUh%Tsh1;ZrK0}0`B*m1oN*r6@X&4H>tC$gJP;|UmHHa0
z>6VO=v#S{D0Wsa)DfLS2)w=;1r{`WG?aW_ZopYBls1``Q-Z_|I3%(l>>YuEDACclj
zpqnd@DUt@5KZ};Bn#JvUIYHLQ%%u4&<@IJ@il^@#W(8Z=&h5p-zq>e#sk1X(cH99O
z<A07|L&s0LjM%y~x=NA{6*?{-*lO_Vx@vPzmLZ46T*bbVK1VC~=IskhIETyRSKzYQ
zg37rQU7~X7D6CCPdIc{0^snVy#s1NpWEf0eM}<h#sSQmw(ZWXyr)s}~T5+YAzOqjX
zZ=Rp$<sOJ2i5WJZn5A+};PBybW`Knhe%sNlh&pLDw$dgmw)j`$YI2lWJxNzET+lCT
z;DOdCaVg&mjoqw=St!*mM^&n%%+|=yZ1!-g0U25}*Tc(e;PzG=P}P(<0ABJxaQ!+F
zKqxCKE3-8WayjMk?@AKB)H)To<n}JJ8n-g6wJthxoeI=hd^Jgen7V-vf={R288B33
zC`;W<YTzDpiljg{1*G8Qa*lz`AJ=~I+gVZ*cmv76D*2DVVJi$)ppN#LE$uQVR|L2o
z>S&SIV~z35DxbI8Li6+TX1``ygj`Vo%c_yvg`~da?`jjF2AQ^S-)&5@Rj&B-11!HD
za!`f{fZrX~4Z2*kA;u2Hjm*u;>uIvA!X_HyUnh!TaG8451WgcsPr|+u6BBC@hqPNd
z=MpRjKU-zQg@<a)RKYVSS4=KK$ZVn9^v!#&!4?3==;tf?jY<XsD`K5#4094)H~j34
zj$_f=)NQ<o%HI*XQOXl<C}^s8Q<hu@J=2rR#C7t{5p&{vVZkx)1ZkgAUk^o#`bd-%
z=U35~&m}9xwulU2h-}|56V6yg=nQh#c;vrZR#u+>qVFQ)CgSr#n@wFcn$mK?l%l;K
zZnzx${`l7Idttk%1`T94B&}wi)gkxR6_#NRYr)lt<Q=*AoXOX-4Lf4^8~khfGdJjb
zwRMkK<>%Vz(<ewMtyD<=$0x|+#Nl|qceoQdibB&hMevhvBHsP<1AU1j=lSo*oqo<;
zMI9ZzXyzm}CdEl9);G1F_44riIo1<0xOV!cAoQH0FyxvHpiTH{;BDOGRpiP%^PwO6
zRmNU&|A4!;-bCz!Pu8TG#jR?ebn~OkiGarR8JIIdgMa$evh$lI8kvDJC6RFe?z++*
zhfem_GJ<LEXmXl@+0UaEq&hEcS@<YSDvEp+n_!hK0u6x*8!GZS6sBfcZSC#fekC|9
zU}$q*PqvJ7G#+yzz^e0CTx1IRr{xqbdszS8&({S~cS_JSJDagwP@(w(^f3TCO4G39
zf%v@gYDej5Wc2gyr0Lcea@jZ002q~9i17Moq5i+Tjs{?p(4L;n>b-90#yIxRI;pk6
z=m+s*JF+}5W~#=<q$p|Vepk-IElJais8Ya<aX`WUNQQS;E>aV(RGy_UZ^GSsfNtN{
z&u?jcz5XI6Vb})18KQSgV}c#4yv8mP`A>msXjZ(*DKB*zy}A-vSXj7vOOJ(a2l5OE
z7&$oLSP!|O15nC;{8TXNcJE(z59!|}$iR@<$qkcSNG_;p$q-J72G6R8_6{QJ9=atu
zz#E6ZirLx68c!vIPGcG)c#E)|wbS>+Q@>1kz&sYEdfNRx?N)2634~YoTuhETNV~ux
zk4%A=b>}+0yooQRNT*Kx(-d{2vy{FI3X3b)I_8%ss(z=Cd9HzB9bN??Do!i?g0}mW
z89NR%dSsM&e<%AGei=+Jy;kzV+y8u@sR;KzT-ONSRIf}aBP*XfUR>Gf<7snUZx&a$
z@xuFYyQ3<G>U@mfZ(o)Phig86eDwy0Cc0Adaj@JIj9&@&m?v~ucJf-EhRXHbd9kE*
z)P$th1T2QP#`b-`CjslI8dOJcMtt}-X%{Wm6pJu5%fC3B^Hb-Y$eO4>@WwD46Yq%M
zzmcFlrkgy83cfpLDw)cX?@^<cYs`5SOnj29JQNxxN0Md6rmu3cQFV@CM!7%z3b(`i
zBfVYSpD3-@$InK?mt0kvJFNtuR-QinBY$5MNFmc=k_v?um)_`VtjuM2t8#g_v-(wr
zN8`NRRRZ$z3i`Lp@{1pm@}Q04JiJ<PHS^5fvrahTqXuN>fot_D9Gv9ae8n|)NByd$
z*gBm;)voj)@^uILS3{fLe`?O(Tut$Z8gz6IpM5HtJnGUfO)HB3X7Dmu!7C&L<|@h(
z`*H}%>rXjOy=cqQ%JCg>v5te0E?X>;sTL6<hi4$Z$T7)cyjFmAe0Ay>ipqL6EYN+n
z;6&nz6i$5yoj3}0KeB2i270YQ>ppm^`|VYISOQ5TpvsY9p(ioO;<QmnyQFs<F(3r7
z;gpFQ`POkJJ&3KD+9pZ!6};+^t;Hj#%+G80AiB(SaBg^Rk4QFE>l%2%IAeEc9Q`yj
zuX?U(T-qOq)?gVxl?2dCK*0jM%SVE)KsIdLQMk(Q<@;L0D=+^7B|x#!vw9!x)?z#t
z{)R)51j`$`n$ytGP5__?K<-&`9h#e?0Mt*Zkh?TCtpa}fB=;*=%#R81Iy8<M;Cvr$
zSA1Je=2wbgb^|$fFN6Y(=vJ{~f;8a@RxgTLd&kRPc)~{$V|+sAU~*VS(YX>^2t@f>
z!SB0n(Ku`u@bBW<pr*BJbC_bVUV#}|m&rN=(&)eSW=!c+N(Vzchx}k^oJuYtIpD;U
zC(7zX7(+3N5>u*vSySmBGZ)OdpdoA6G!N#Fbndrim195m?k`%{9HW}}l{)O*|I{4Y
zF^b>y<LpqDKE74nS-PDEclCyF+BW~%sR=l|aA~=@IQ)C(7w(`64RgUy=%jTDJ+Q@q
ztV^OJ^leW4pksu1Yfntefx!`+thBw{&aCDQD)6PCPqpVxkxB~7LaJBjBhtbvARv;V
z_takw<gffK;BJ=y!-6!L{u{&O*-bouCQXx*0vjP-fB2f?uMhm9-!)XXbB!``8fVEB
zRY2@m80!}!=PA1{Sv9<xpQR~|2!b}Apg##h))Z$ktEi-K46|_|#_-tK>isYOkrl`w
z<K5gpQ)*|F_msKsXz2G5mdm>4c3J?m*N}bCa)^0X0{WbAwaLs`G`aXk$bnt})kd;s
zm#Q14jE`9v<{{@~{b(e&*)v5kaz(5(Wxk)8OvdTtijkD~q_|P1)YF>ftO5cZvAA_R
ztuLECiPrdKl$)9tG{qXGsWM41<i>*-D~rwDq=TAj*DL<@asT`+>W0rVZ>6rV7&=!=
zw}KP-xMHnG@_z6bDFey?uhh54LW;)buqh{bC<(?VCIT)HL~(-{Oz~e@7mohk?iHq%
zJGsb6BG>X>qfAA|)3uOhT*#4*!kjoGKlS${JLWwBns7j<naQhxvkoHOo?t{Wt7NIO
z-k<vQY@eS$o45er>`p*pS_)s!{j>i}97SteMCyfF6*Cg~BOhMd0z@^dQSSTYNO3BD
zP(>)g4!FC85%)jf4T8T;1eySAnL;m5Pk@F?3P9Zlbiqu?q8`4-kq4Q-P<BovUGVz{
zVfgE-resnm+NS~!II{`J<;C-*HD3Q!%$hnM=Q;)fCC{dP0o$u{uz)h=`uH?5YD2og
z3+{dqf53ZPXDbjJnRl#!IB{$KUL>pkMj^O9jvO!YC%Y8}79Ez%^X;^HZ+Vctzxn4q
zBzthmTcO-{-V%D|JrcWH!PvXF8Y7+!8^@Qixknr-Dg4VcOl1k-El%X*7zun*Y@cGR
z&Nv-j>uqwv2FjImf*KGUP_4>hYl7n_8%gDr(^*&#(GM5Sb_iZ29!5?XH-V{SN%lSw
zb!CKW*LK8!x&a0~mVJFEx4DPMeeC)@!u_>Gs7n3b%quyLk<=LE@#*pO(M>pQEk$?Q
zP`f1TOG`O7KF0c9-zJlS^0|}}M}lFb`9v0FWG{b=uU&dNi}K(HsUop`QpNwscSojQ
zy4#sAgXJssn9fsPn}6_XvGxr34R|ruy6+0=yHVK0^kI<<&u6mAe_li`EA!3|d8#uX
zu2$be3{sIM`K5~oP4*1g7j!f(?J?hn@1Tecw>&Gfz_^zUT~(imo`0j%XDZ_MnihT;
zYcKcZ>_+|x?v#>p!rqe|PDOv2hf_=`3;x1<Z3LeA#+wXxd)z5so)Qj&ac|!@w2C_c
znqvD|%^V%iUl|u&p5!y;gh4&mZJC|R`%rpQTddaa->~e}5&VQ)caJ@6+Qg0(EIqKE
zWqk5E#3k_NZW1b-Xm4xVfaQvrnw!rL^Jc06k=@?j&Xl;lJW)5o51rUWb%Uo-lhb+k
z8<2Yu^zeu7k(bRpmN^oqypyYlCU2@&hbT9Y<cqcsfq|p-VGk~eNl6F3P#@v*5~U=d
z2`z)4xJjiFSvEZEA!TA>VsM!b&(d`-kq@fF*x^<_s~Y<}+f<4=`FoRh_JIXPm^bXP
zZ5<t3TU(KUrfS|33R|!Ug8@^CMmpwjd_X-LOebCd7dh#8qF=?Bj%7eE>$<vwT$?Ob
zJK6|KvFEGj@N0eOcB~#|EKeo?R-|HvLs4he=7yJbQqUJD8*=OO-#KDnfH-V=-JVmE
zKZX+fm<otSZbH&M^?-UZvpnGHOV4N*&enGmNN%T~!;GtdB|S;HU!2IuGxm(9T3R)o
z(HPMg5YoT96JO$&`Lp_2-{e2*0FH~L*pXd8zKMNuh(`0z{-qA%CsBIer&(Y7vvLmu
z>LGt-)B1x7H4I!!eAy_yu`g-AfzhC>{A32F?d8+eoER&H;V=~^o<GHPcY$TjZ0Yzr
zs<1$t?^grx5f&9|)HQL>Zfn>DLtA^d+-OSg*)zi7cqOx^#MihVpN4LvBE;ivQ|6pi
zI%9v$(2^XVoWATxxoA?W+w;OPsH@YgFj;qLH=X0ED*mnAA;+ayzpmtxB%BoRp>5)J
z>C=I%){oDW|00yWRxqo)T|C%OE0oIi5aPJC`Ff7nryJ;c0@TaAt}e{pVFXT^zoRV$
zto^itn5Wf9s~pA?w@cgpH`7i9K5DEIvDCE%SqE*$XUO<^QW~n)z-S%LXYjPO$!S!c
zef_ECWwFZiXUJOu%5I=*uQ%#BczXIu2Dw$QY&|l@fO@sZ=04fsxy3W@ns35((C681
z+isk)M<Wfw-l*)r=1l5>_qO06mJJavSf2#d=(5&YiFCqVYPt?iwW4iVpe-(;n&P2*
z0ax~0Zh_c0{?*&2sjJQ{Rp_fzQ=~fN;N}?Y5a5kp-!B3pI#bERK!BAqwFC=VLbgea
zqt>oHv+1e+b+C4e((OU4;I|sA&3LEWFRo)`?<0C{b9*-L?mD{HE%_~hOpJ_NmWs?!
zN!V_?{<Z={`TCg`!0!N3<7-+$YQn>jwd?P8pI0gs38<U6xBvV}zPPyPcrdIs;E@Gd
zTt+^XX8@bR5gX}uw)P)yGgF}jp!X-sunG;v(9BNwI{f$kX@6P!VxWZw@~2(E`f!HW
zZWF9#KS$!EnmzWpGQ7ihrq%8OupDk|gah>=g<0j_DfH~MxcK*oIZ;N=vKqfefF=Z5
ziz!35T(Jm%kr%l#o!PAruUn?xCsr9necOE%k&o`s&dK2&k<t-sn~Hls8pXjiG+y$S
z>8*;qstR#h(sT~{Xu!ic;X!&$28|-Gj$QOTi?09AuL#dZONi;$bn$NfZujVxugbNP
z#lojz(b{jEUzp2D<1OATGe!|fBbvW{>1|3dHcO5rO9>45VQOK452`RCr8sId^t^Vb
zGBK3X>9wEUI^n>uP4As6h{Ixn_3Bev;$yybCef~SZst`<`OKbkaXu`2e!iRJ4V6)(
zPi0k$;%#s5w)T@kY!18$o?zRL@;8?gJLW$m=2SG?K+;=Pyy;o5kfh!C#MFff@3T~{
zUl3t>khk<LOO7m7265M^q@;0caWtg}fB9@EAa4n~$<bV58q-aA&7?*a#jDe;wc+=x
z=|vqcmG=xbTSa}q&7u1(A~OsYM<pvCNXPi|9V0>f_Z6(Tw0UKC@*3xJ{TC=hg*jiB
z-lrJD*^L+mv)$Ppi4twn@*;g|Kpnl(=!)WJ-6=1u=*F&cHCynh(|&5K|02qz1nd0e
zD>a+*QNpN&k{%@XM8SO;d$y<dB@fA;<3WqR)nWI1pHotVp+K>b&tZEHuTIAQSgHO{
znud$*LEdDnIHc5MEUp4lB(4_8qAXU$S$lt}uhGebyb>a4Qq~#utWO8MH;T|v$m(Uc
zqVy`89MaNOPTg<AHxt`7k=VGI;4;jH%J}3?%yf(B%@C3+lGdAdDSY-3E^}#2*v_i1
zt`1yR^HciBnrbEMuKc-#PlkL}q;J+tJ)$~ud9e+$ZmNwipV7FT9a4q}P@ma^bF(+3
zqLO4$zs558j)Ym~g#!Wl1|SLnS+r(JDA!~X!NwgM$%wn}Edg}MI3EFbC_gnJVA}A8
zz8l7tm(wBl(QF3T!<=$}c`4Xy8JWC8q5I4>6TIX>W|nhMQBnCYpHv&w|J&sY8*wNM
zf%KKO7q+55iWa%bo093fyUegII)?hj5xiePibU)T^J{3d8;u@U=h@MKGQPF^m25G&
z8T9xKbSOHif42q+P$ROpY(B|@fIkCZ7!wUlv?Mi8DD6H-x|_n=Tx7JoPJIiZFxa;Q
zI`_nUQOZ9yqe|$KOk<oXRTv;@M#@MdfB&*1O7Vgp5l@GLjXx7I>8g5`63WvkA(s*a
zgq%+>pED(R^~^^W6Sj34dy3?fR&ezfyxU-EPMS3z%wjKvX}dp6a+0uBeQB^?k9z-T
z-g<fmtiC1Tp!pc5Rv{jnC(}@{8-ME8s|yn?TVG<`i=}S23cL^Xc5b}d2IY^&jyrb}
zXziSY?kv*kKa0`JRqb|`QmV)NIzy1JsHnsMHR$6<x_1L~>ZPvoIjniKUFhgzc9Cb;
z$7g4NU(z9^>1kD<s>Y%KSJ|oKZ@^fk;gjTd^XnoHXybrPCXoMaEHaTQnVIHd2}3I|
zKKjjhC&1g3q$74N<B!<OsJv;sC2+ruoJR>~r%;I(oTX}zb*$3ag3eFp&S<Veo<1$y
zynPjQ$yLY>%65$7H1F_iqng(@8!6#*CNG=BtxfJ<=U2jaA~QD8uHce$KpM<j4%$7>
zQPsrkFqOOvEQp!yplk?~*T#-nJmVbkc-w%#KZT>6UAtMjgkAb;B$YbdL77P2ua^%s
zCA%GV%ZFM}2c%Ki$3m1}=k4Dlso+bYf{?ofX2nsUw#XV>dPMvY6v|L!DS2M{=Wp4&
zqfYB!y#dWMe5cP|&Er!Y!Zxas1jNf}nZD+R*9+nd8eT>rrR1tv-m~FLR+yjcj=nyZ
zA{veND^qpFp-jNK)j6zbY32ZrwY8B|OOI>lY>Y<g#xA)+rYP&gTe#I&Q}JfLdx!&A
zs5R|T)DG4C<@^on3}ob2M)_I#Sm&z9ZChb3P;1Huq}Q-3MIC>d58zn0ZA?PGU8Oc2
zSHfP2Jx@3cVpiz~YL^@xZ9`Ab1PUdy{n!7a^r5!|U~U2y5E6;Rk`GTySzTQvq7%!7
zIm=}+<6Yd+GGPG>Wik7qvN1I7OIIC9abg+-p>X5{BF6^Scrj3{0Cyt^md~jOoEI$D
zccNxw54p`0iSSSL@U(++ym}!q&cd_<-a|ocfAVOsPDe@{O=Oijt>Ub@WrkfQp-3OM
zEba-)XE+`YEgrZ+@Bb<lNCor*E(joa&bpCxIOBEq9ale2oYC8lB$qCQ?zOe&AS6*R
z$0hp>+tdiRSAnO~*zi#$X9dZ>)IYv8_g7<J$4ll^uMpu17LB=*mVJV2+c%mLz@EQ#
zVgcab=AHA9CyfBD4=yHynLb$*#5TfwL0MZ7Uo}Q&ib8BhR8^X6cAXOp8SfbUD~zS1
zI+om0Z+v14UTT9pZH_l@DqN0QV|JZd{4B5dMVD1WjgIN_y8{e^ter+XkTxaW#UMq!
zMaXdLIx0GKMz6TivgJ3a@eiTrfM`J$NEcDWNQf=J@k%i!G)!9ww>@(4>?T30KX_b9
zv#q<k9Dc%2oA^g+dygs_{Rt9zd=}5q6t(0XZKy^swNIibphG@$=BHD&u{tFg3|J06
zB+WSk$foy(<pCA0`-xo94U$ZTUQi(}wZb-nzt(vhA<XwW_#ek=LCoGKWhNbi*cUUX
zMhxQO456R)J0F^0N8MF#XBt%vS;0wilBrvbPp<^h^Kwz<$e8yyT81VaP9Y|Sq7;>+
z?=kJG`ph|F1=%KqFx}G2;Rk<>9`_6XY>aqa(-YkePK1rzPLY=MJ>8z7mnRjXeu^8J
z{1R8)yrWIna_qGxb+}Q;cw<_nHMnfg0G;;vfr`*id<`7|6EN43$@#HVNhPHTC3nf{
znOKRbS?cJJ8@KLDz(Q1h7l&j4-L`UYu3TcB3bjPsL*El4kpS@sxb*-E^j(QgK649R
z2Sgyoj{`|;Wwxv;#J6gO>eb-izawp#tJjb*{oyj<1lQ@k8gy|(Uj2u@Qd3ijyjK0C
z5KB!Cfa@ErpT9=lxPHOL-JN(*UCjCd2K&#EX;8LJ*yJFuMw0dE?DW(vzen2z1~@XE
zuPYXWA{^Y;yIF8C02_cyYGFe2i7)iU*E|JWdYNG547<*Ni)ijK|I2+Hy7VK+r=y--
zUg0%1GVn<c__*WTb=aoMq4^@0NWI~~=Exb|Omy6S1D;=48|yX^on@sw^@ZLa8ue7Q
z`XLdy?trJ<9sq&#Nh6K%eU^Y%U4+1GrnJrH$JGy=aljfsYHbxqR2f*9=yLQCak1<o
zX2!H4u#tl9Ha?!j!GL#M6*GsyBrT<tOoeUyagXMrdeUXrB_dZB2MgatbLTp$quo0B
zr_T7?T>PBiyYwK(b@fI{c3mz1IiD&Q8rKR9x)Z_#4fJnf4<%JY`o;#gx9tY?3Db?b
zV3WU)$YvtDe@wH0RPp;8j=w7SW?A1~8Dss8i(*FkPWDK6K^b1$V7#BZg{S9vG+>;_
zh@${Gv-VzGP$;5sBDUBpLi;~{6v$7ESz<7lADy@pSTu-ApuUhaD`jAmPGU|Jmku0u
z6|L5L3y6OF^a+!nUXH0w3-1Ba_-$}B$q!>55I1G~u%vg8bhEBh=s8K*Q-A<TZ?{z^
z&n~SqT42M<e>$^7w7nWakEGvoaL+b1{W}s@53Gh7<JBkSg`dz~w42J4>%>Jkf&=|9
z-mIE;iLxz&?Z}RuMmbp$>S$%hbS10h`Kt*Y=dqmYb-1+6RDRUGmt;X}IsxhIP-sU9
z@{Je?EZaW$rmV*P{h2;N)U%@B=FJ6rCAK=NP0v(vF|}N_TaDSb{1r2DE9IELhuKxA
zc_z(S;Y+Y0EJtNHkwAwhtW+{FYE`m^f?Xs+F&A_g(tG|U7LGcFX!%Y@`U<6CN1Td4
z=7vWNXomsByHL)!u=l+^#oheuO+ia7U8}S^{05x*V0HsG=9lpn>^OQ);9Dy`F8CVQ
zSGG`Z0Wdh$-li?Px$`-3adA1DfYpZ&=ndEi2RaH}1(*2=T`YK9%3IJT5Br!^{);u4
z!UfV6(Y<^-!WrROJ%{imUN@P-rlux6i^q{vHW4+KGdnUP_{kUo@SQ`U|G90NnuLJj
z#An{Qu{jWvFbG=?URW_jg*+ULy23FAuf3XNbNZy~>Riep9`$gnf@_+b-r016APuUY
zaF_Sdu<J7HV2E%w-_31~J&5V;$$-O0yKo}8T?j0O+w`RRz+^84JtJWaqV1jS8FV_F
zA=~HMV4ti0Ci{{^HxPH1CQS0E{?%{Qj<we2<y3u+uD`u?z1e3h9{QdEp##D6se3Hl
zMusZB)iFM(T4OBEA@r#1Qt$NXKJn>LUkX8A;&DVOgdKv|gOYDR3`X8{nbd*frss`T
zMI1P!)S;9hm6Uh3Aaxz~;@#M|I0DM17a7$Y`I>S$Ke?#{?3@Z#FM#ac)?VNtbF96?
zMK3MUd!ZhiBCK)bD3=>TSNR52^14`m7IeFJ5s=?ZrtVh6(ES+E%D0|k&jE!>VnrzF
z*s>cu^HuQm9sbwNOvqxj2)Ln&*}ttb@Y52C%gB_olS3tm_lcFbjR;iB_M;*XVA$2$
zHqF=K($a`ri~HD#o`1LYr~UZkt(W`xS<B)5FqsuY)1J==tg05@47_*rs}W!dbsh5b
z1C(#+?$8!`5q_zT>Wle0{T)S3^PM!sMDiik+t!{39%NEzpL4+PF^2AsjYYo&&xM)B
ztW%iZDMy?Ik@x@jGTU8P<@Qo5E4)|=uw&Dfkw*|F7YKhTX%Q@kM+Y<{lsRGKBm6q7
z*H@?WY;&rY5FTN8+6#s&(yX~;gB_?=7>h=LBW93F$_EdRzF?VpnCDrp<*Id37uR|H
zQmES!;?G*1ZLYeRUQb`S_(|PS_gN@xU!EXudbSN~vCGt<iDqoKQIoRMlvhkQ`mK)s
z8J6`DU#{Q8;cJ48nwicyk9qeA94T(fI1`Ce<XO+&T(uEkasV<nUA{>M%c@%|iWe6H
zNcDadKSF~52Q|53AK={ukj?HimK?TxFJOoG_m4Up$Wx)WV;1MY-jYtEp|cdsHff`l
zggL}}mr?H-*X1-WsL;>oL@x}-m1#qx%m%L|396Nc|47%r+rxm9KNQ%gC$ae>J(XF-
z*|Wc#v!lA}_$pnu<rcoSPVdmS91}l+2qN2WFRD8U1P7klj1Au|TIFA2J6fHa;B9}X
zO{=3kEBG7Jgb&S`9er7mCf>d0&ivr=L?*H^*;E6e?oP>_N!n$;z|U`>zjpVH&MTZm
zWJ72{tZ-^h4fM!nrp4UY5oPZ+-iiScAeoi;UW*LU`zn&|<;9<^q7OlJ3n=S<f!vql
zM7?t!kg3b?N_qY6YY9g-3}KpQky61iJwtnZ_DnGje?_5{s?pf~U$;{ILf;LPTyey|
zQ!hKW|G6db3okR<&_#VB0c^y91AryTrk@cFON(CirHXZIKl|$Kadh0pg)=mR!(kPW
zBj4070DQ)?D48zvBq=R`4ODRYbcTJ2d==OwvtT54|2JuU=H&WzM627{p>tpjl?W<h
znk3&Blpp1FV5?Uyp^%i6jK3+IKDey1OhT(A)T>AnvA|6uqi&#R!>h8)g2}*HtKqCZ
zW#s)$-RV^rc0<QiEd!1vhaS6;<bEtJh4Q$>1_*LRysEe9EUMX-q_qUG+=#2Ov(Koc
z$0vpdEL0A|l`mf<Q2usj{hdlty-Ulazsqp&M#Q&86y-KOqfFGzrd`IWMK(G;YsXK$
zc^fI5kUb$7RrQ|7PU=Mq$<Tex6%N#_X~!CDO@fHvoHu0zxqn_2c7bo@@YQHdAjwY(
zbyr+E?B#Eos(+k_&ub0e6?AL40ed8NJwbcIE}C^`)i&^(Q+GeJ0iJ7`)edO<nAz7P
zFn7WcF+^ingw5}U-dD@aBy-8t5L1MM1%T--w{!0*-swc0y`bAh$^b;<an1>q<(pyH
z-i7gSLe=NTzy7AV=KzY<RkAQ<fC8yMa4*uWRjq!ZWZ`J)>;|wXg!X9ZkOLK^q{S-e
z$wzbmiXR=hpia+FiGU0FKbCZSTob(90=A~%IVusOf6-(A;@4&)6oy>_Zv}^Uo*)t8
zuxTtGuxqOHI!;zwqidMES)1t2??@i_(^G1|<&eo|abEv;)x#S564)CD$h-Ekwc@m*
zWR51O&}yM$2dE<&;9pbba1nxbX~JKw&0)g++AlqbMgklEAOgpbPgi{_5*A<%hBKQA
zF0YC|e?GmvExJcrC(s_nw4#wbI<tMxy3JreUC>6TC>e6RIl_AZ>YuY!B_E6qXhKg{
ze!I)+Znjc8AW1D+ElOoTme7sKWb<(C-y^v4dhDQ5<l`%o_F~utxk9Ch!sb(nva6g=
z)An2M0L|xJr?D<2o=T^6cb=fd_!K_QM_zBtsBp)YqF;J|cV^+@H{y>j>Aw^yzJR;q
z<cp?PzQaKK^c1Lv-a{euQvPxkGk~RAuy6uhNLw;L@yqK`!D3Fy9^EQ)ruHCv3_u$9
zl6l-}a$cYMuj%>mrUL#AenP5<2g)FcMwbW&?0)C>k&Wy$3GX@syx!I?>>U+`sF2tH
zks9GN{`!+6DCh}**w*~RmB!%<vVZK{<C97LnF)PkoK#&vfs0%=)`yB_LnimD?HRzS
z-{7tZS4+)C!k!%mwiXd&qS>(s>1K=xF;z}3{1oQMm-Xsh_)!Y`DS{THJoVY8w4{4v
z=dOvIhTMCZd4=8f^nC3(ra1;aAq;woOhjr_$_djO6ga+=0f|+Hn-6_xe1F(m1ff=G
z&pA4s|1wpDWP-<4E`F8?mZ;)4_|AL?ZyLJ%y4V=F4$1LJxA3yQ^HL=Xe~%3Y9kXsY
zN0J*kt(R>dyJtjr&J?swT*>@FXvV_Ma&O?iwFs}ao~1jn2<I2<j^-r^gD!&kV8%xF
zSc%s6<ch$SVPJEE>URn3FC@j+35^-pF8z(iO91f+ltnu!dLe4+_09bJeB)tIvVAto
z7;1^#aQ((*ramFm$7khYZ{k0{mos8+|MlU%@e2=^U3<ao*`S|^!!~Q30RZ8*=OBv&
z5h;TkZ3v7{o>I84buKL)j@yNpHkrMcbjGC8SZX7P5<xa{Fo17dc_UX`gg;sy2T&3k
zm(LP8{upG9l980GUNSv$4*+BgfHH<Ju~@ErD(Y&r+v1sV*4z2lJz?EYNoAo?-pzs3
zFq?%Wd-OEPdF8G-3kHU%-YO5(af+&H=V%NN30{9Ki_zu<^Ud48t(}}>UBk5tK8G;+
zdztq-(|7~~Xf&(-=2Q&)g}>^=Q|TA8;5%Jv%@<EIf+3lzJ%?5W?yWm8UY9^KE=J{G
z5`cBTy5{+*IpMo1tRk2j?8Sx4>L3ztzQx8%Y5kh{ES#97RS%d;Z5nc>=&q@szT0o?
zO{JIL*d9Z-BW-{RA*M>C>*^^9V`PYliT_CdJ0JUhI6CX7DDy6kO9}!j-6<`h2n-=1
zE!`c`J#;sSbV+x2>d;-%(vl+}4bt86{&@fMob~K3I?p`!e&ch;uh-5>$f??mzK4B`
zkD?nw3ko>#0M%{G&|V!S@AmBQzwx-*X;GC_1_m2?^3_dg%hX@~oD_&r3i^mAZ&rex
zR@m={EyV1*);FF0{Ubrc2`bGBX6zm*=2{93;*cmwG06f1|Lfirj^^y+$WQ<97k2-k
zhaX9^k10qe1aYyQwEuU;WO6mhdHcD;XiMgcQ6K9UgYP1`-)<#8y>P6^(eIn<-&jI4
zUVP^ogQ@Oo?Z*>|n<$D#CGOobUyqsIp4P4uR{?69(%Bj!c=h2OAHsrPW#CBQojS>A
zS1N7B4{hrDUHc8PYCG}qAt7`S{gLimj2{b7VwU$2e_-b_)6T79*mv2vc<#|5g)h}l
zr<bUHoY+|%iY%iqaijrdd5cp$Llfjhi#0G~SldKB)X<s{(N%4{#Bs^NvUo+%_-222
z1w_b~YS5L^qm1DvZm&S=emTVj?wXO45i@hxF3c5I7`^)4zmd&yak0d+YHKAK-G#A%
z;j>ZBz^xL!G>k+0>drT$*GN!WH$ht&e}8}9!NX&4bQB8=HlZ)6UKMRT;Q^l7Ng(Tq
zDmRPST3ZXkJM}S0J6CsL`7KPW*KV72^aL9q0f>*G(fogH-T$Uj1u7`gqixk1EFi`Y
zN|g256IqqfF=Qa!eA7$yuM4c;o!5PDtdCdkgg{7%d$6u}v$wZ5UD+CuF=fFznE~#i
zDO{h`3RU)=^RT%stxpTjm}}T_EW!sZe^jXav&3R6K?7E$4qzCz9(9=1zdQWM{><*y
zujQSk+4N;3mPYt`6KXQy`NA3^r#sCy;)lkp#=u&Lvn}V+aE{Ku;8HDZBW4rPGLdZT
z7=QhX#$p9xJ)R&XGE9lRVKoe7A<ybGJp+!6(U0+BJ%&*BUyB#U<qWuNOcdB9T&@hh
zRaiC_3YG}&?myApY0itaXXWP2_HHu#8<ARW=gX~<UdsvCrhTPjS?i0c7~<Qukor+U
zAoAu19_b3+`5gt~xUFcrj2tP5-pF)P!E{KHoh{tE6g%!gV7j-q<c-JuF#;o_l=x%>
zZ_Lesz5fUhu2s?nUrH6=7$z<Q8G#t{d2ryTqYB>kW=My-a_~g@KrH2K8LGg`U~)BT
zi<I{E2N?%Y)5B8C7N{sdj5dCqR+YS=J(j@uk_6K$QRi*e*v-ZbCtbRLe6(I!ykCu2
zot)>*y1RhKKW`wBPwquyi>eTYP{62Vhh23t8&p_glULIJxyZ-OWlqTF@H=e=2<SKC
z%;y}44lI|j>Bzj8UN)dclU+y7F7D@2yJ0XS)|b$(9-QK%$8c#DO=7XLQy7wol`u!N
znV^@c9PLd$C!G58ZYAZSM~p>1?a14OKdby7E{dtT4C$M<`=#amrb)J_Jcd0!o#qO%
z@R&+hF?a>~^1e4GPYSeCxREOm1R9PQGJrNS+^fH@o4x&H8CxmZBElP-_gC0+ZGx({
z#7{WJp*eFwj69<s&pZ#U8@Jp?(d+rFnnCW%1iS!1HhvliY)s|dfIv|nK6BfXm6OAn
zF$(4iab@H#@Mr$X?}i@CofhNSCp(k0z)TL$w-g&dr`d69<z~^?YkD1<HRJdcn~a&(
zIlDeF0Z$Uv5541Wjg6ZA*`wybsH~fHB%}lH8(@xs3IF`gw{8|hhCx*5Kc(%~+J9ck
z_$R$Dg^xFe@PHU(4;9-&tb#uL6Y`v=W^71gh+9D2*E6raX&hna*|T%TToeefOULuB
zO!Bu8s_!}k7GN^fNJF!gWcs*uU~5$l_lmBt%w0xwvTH&tk1X7{bHVOX2zm?6h;g2-
zp69s9+gvApze@7u%{w4E7luwznlrwrc{i9!M==&<p4x(yWFM%6f-I3_rS&<>2yr)I
zD`?2Hf5ub+w*>P!{UWDr))Yi3-#Yb$0O>@-%>q|UbYZ=aTO+02p7R`5&4)<1-;0b~
zwYl+jN6+orp`YJ({Z$o7-7L^ibePsNQcDVlB6dgVw;846*cHB}vh^a97QI-rlqMYr
z_Ir+AXoV=VE-k%tC*TWvsWW{9NC%r38j*bI@!bwx35<$gvvUWXlPd_yS-&Rr{Jna9
z{@izpYZEX|cCR==wlJ1bP}D|0H2&)D!K;4Vo6^%fQI!#MsJC*_xt4mwsqVGs`3`z(
zt5`g5)pL{NqC%^X(y-;#QblayV)k*T%gf7?_Gd#;`AXN>u~RrQ+xMd@_l(&>zWx4p
zD~B{@M7!{fvGSO%U&FiTRdc~G7R2<$Hv5(ZR32h!)IB8g&7R*T4^ytf)n-8#CrKV!
zwM7~)<aaTTUl{SV#-NS&o*^N%l|Q7NwQ1cMXt67?2)rm%P~w-`!Ic*wS7dhmkeGtj
zZt$B}!YF$hFJPO5K_0#E?LjdJ)BtbFS74=p38Szs@4Ecm^Y@WdnJBRj9VnSiT@uXB
z6h#C@XWDZDh8I3=tHB?<LUt=V=ln?ww+U0MEUO0_zgx@a)_3>t;CC#cum1fT0cP}g
zI-|G5zXn=5LPq!OksRC+TA@*Lc)u3hui9QBygX`c@b+zFw+mXy$RGp3c96$-R!XbI
z_AO#Pl^P9O^w+aN7$^8n`9G9n8@wAzk#K?Zub3kwAycc}#=WQW!Q54|YQ0$X>F<O}
zxLO36__4yT)1ReNk2u6syWc~gx;soI;t7%g-X!D$K!gPkpf)dD2qd7*lWnO(E>P;s
z^+x`C*nYgv@x0t8dAz22Tqjk71Zp^Q{MwH#Ca(*$ch_r~vo+aIe>cO`*aEZdbQg(e
z&xWer>3x8{a@F2}+{`V~l>?R!vaDtM?&3sLqN^~x{qs_05`=%&waC)@Uqhe%St~<s
zt?wel;9z@8=%>GZ*MDs2y-_D1++3oV9m2ZJEW>2mqyI9TQLymz+gYj^Yukh^Z|rNF
z9w}Vo*%R~77y4<2UyhkvUZ}vdEZN6QG?cp)(x{QviC>D;zic)Xtlv$i!$zweD1=<8
zwp4aeFywZ5vi9b9kRcT=3|?|n=2hF7Zl>n`6|>Ll@8@yz*EEW+ei2<4A?j_eiP1HU
z@=BgLo__K^qPAe}XEmeJmMO3yH(k7FMzkes-~(5PP9ReIUm|_pwquj~hxF>7A;&Er
zBI0M9BI1t!Uc}z-TL{zcT+>;JO1(t)rp-@6*AV#tR_VFDH53XNL6f#xv9B5U43KxT
z>Hj5<6n#aY*}X<s8qU+gC|_b0Mg2_K>g^4D(P{>s4bLIK&M_heks7%gBI;l|`3mZ=
zSF{nyti5q34k-e7dUEDJ2|3p@HmyQTVwpPD+#fF<M-}~Vx!0E%Q5!ueJCZHcbBB8d
znSSLOv1JktHa9ofdbACFwb<y9bPtSsp-h?}EMw%}_I7ebGTUdj3c<JD4d<B|mU`(~
zKeO_4>fE{3#(>!K_e?Y=z6i~s$8?_Bq|V~mB%bP&Qy5?QKikHHz8V)4SWO_3ZEO4d
z(^ot#kO{}Y^6wwdnQSaX#f)=)8GH{)50zK#?0&q|Xsy)v9gJv2Qq=TCV?!9>`Kt78
z)%QDvnr|w9UJ+zyM!&BrTX3ccUf>qDPUK=XuB#-K(OlOgQ!*e7Afi5=kT6;t=Cs4o
zVUFu|Te`pjDt<^Wjkh<$NrSwr5hKfck-G2JZt)G{TQI_Em^)(=Skmp^G}P8QvwT7M
zZH}0ZCzDaotowT%&fs|0%;(u8<N%#>jrIQY-LN#xt>;#4)SNRWCMH#(MC<vjz_^X7
zWzAggSsQ;|t!s!dFK;PALRHR;1n|w)wuK)(4`?xev8F~}HwL~G!Tx0><0LgeFCBDr
zmzk4OtHt`)uig-#JKWqO{5r?8XB;)36fH13fVWp>cJ?f$X?ihPv6KxB@CJrVdw`|}
zx8^0smB8uzcTsb*tLU$@P3pQi!tLJLhw_TuW|m6|q_Dp9TxRHtLl$VS{6_!0(1a`!
zvK%%O{%FE1-LqLZ7j#9E7Z|UQpG_u!8o*I}poZhcE#C*=DM@1Pzgt`zO^VT{^dR*I
z<c(QNAlkZqi`3$^4FahMt^b<-)V}%8<7ADb{Sb*=5pUTvh0Y3mI1)PN9NZYXMDks=
zUj}7m;j}w+87IeCAXW)>acZA`k?sqeuvCABLM@$;IkbsYyAc2ITt6r#?fkdoqRvmu
zpgr9=x3&=eru@<0Q^j=lResjRB~_%*itTh6Wa@DYJer8N7wYLA+$A4ri+Mo9zr*Z%
zDy{^^6qjUD#M^!BU{vg%C8`mklSh{%=Q&ek=h6u{<pOju-sg8TqN#K}jEp6qlv-DO
z{i*Fb@!`RDt;v*HAf%UL$*$xyoXguZxqRa!fiagL@^xUz=nYrKD6kJ-MZ5x^n05uC
zUC;W{r0fkKdtZxuz(%E}-%cUIFEbXPTzwudpg5`XU8f4WN9;w8=+=JJ?e%-HU?p;j
zc5}!l4)TlUG)1P7c!NgS5;#o^k<a526_^tJMjzL51dCYJQEpxPQc0S*#%|@EmEP;x
zd>ev(dd`fJeU6Et$>AoGCf=#T`z}u}e4c@f0V^&UG4bt@jtQ<%dH^!BZ{apAUH`x&
zhSFl{>nS=;8w?tz#2!_?4o#J0KJBl8T>F~wRZNiqu%B_+)!I&h)n>|r*2;7rytWp6
z-WUoj)9<@{osy~`r<ubOPd1wgo-M;ymk4PY*0n{IYGU!Akt0r=>K_%8dC4Y__<9s#
zo-|&$D~^QACToF<j|79^*X;}6d<S1rPZ8ELkct2-Es(URPjFLx)07aJMQ(feu2nr1
zBW#}zIRo!gda>(^ml@6nV#AP2WB7MyAjP3;ddnD$do;YP$|r0*%M7NMg0#ot3WDXk
zyJaW6`c>Adx+T*zXqbjM3MKWPPH%}pA^vaJ9B<5C9la5;6G*Ir^*%g5?5OiGo9K<f
z(<FB!`4%*eYco(lEg=1(_Tn4E!mf+Xo#=m|y`-j>UhTJOz_-&X;lUC8b#u<D)p%+s
zQew6XtVJlP-_@ahmgumZxjm9301|;kWu2$<2J3M@=b@IgtBJJ<9xI=tf7O2Dd)C@O
zo!u&T;+_PdUNFu7ji$jf(T!T&<YY;k$+iP?RfsP<9D#;^)sHicJ@M^Ke(v7wsyl#E
z)%3@A-*<Xr5qW-;W82Ry_Q4y;?eX0IA701JrUfP{5+P@|PkO$KBsLoCcO|SUC(DfY
zr)lnvn;4}KzoGwjKx8e*NZP#?H)E()v9OfsnLz$%nucta*;rEpm}jyu+myvN@8`L#
zMLXnN0PSXn(ea*-+iHd(X`gaM1ETeTOD6@@L(PgW_}NR5wC#R6^^MY3n;RQg_RIGl
zlO=wfw8^-CT}O95H2#Xyhpw&hb|>27`cIr+qgW%^n)l0((**4bV9<`6u?F@~><ysQ
zX^m;_W4ddxY2IHy$QJ1J-z5UyPBr2cqcq5KfSPA(RHzvLj6rz8Wa)p!-x89d4Kv2K
ze-?9wJRlfht1u;ebmz~a5e&yoo;lu$MXgMBwkzf=K12C@lDfAWMHjR4GUSI<zYhof
zIPPhixEF#>ucq(=GXMq*PfqGi(TTDkMXhsf13v?+^S<IS`}1eUL;?(R20wJ`1YgU`
z9M8ra0=o_g>zt3u2t=9AO;z8PX$wa4)uMHOUMN!w2J6)rqnPjU+B7K1s3y;hVW1^G
zW0NPtSdl_SpfBT4m#H0pm$aXj8-^HM-cUB}kqy~05c@+T5Zi6}BUe~_@UNc)g3h`$
z6#}>F+MQ8R)B&e0YF6FN%h`k2_gk~N4d_Pq=4_?4E3NB8C=^h#NPpWKs(+!6f%Y9Y
z{X0#st*37PaNW(%_6nJR#yt390RS3DKR<)*SG5;)<14!^V1Q#4)lLSwEm(67f40uB
z$_s9H92(oD|J2K1E{pzpL+t4`Zl$mWu}I~vr=X%byS>=+s$e4%Up*X3r(*~hyX|&u
z)#uL|L;cJ{8N2rZHRH#;-xb(kemxtOJl2I&!w+B}8)C`z3bR+2znWE6=<)uHU&jrE
z!9o1WfwwaDO+?VHOUrk$FCZD|SLa3s100LdQF$!v12<Dbf%&t}f44SIO^TBf$E?%7
z!;IUEmHUv++ltP$?$6K33`bD@j`Ec+5V>DikQL0f4LrwTiFStnH3ie*TxflaZU)=Z
zmvF%`#5M*k>v!{qeDaUV8EgI7_6yQ*A{&bB#<9D@Gj4c8^nu%eK|;6f-(+F+PG9_Q
zciWE@{ynR`96WQfV~XSDGq!e#9JsYdMP_?Hmy1L#xlnY7Fs1afg@g@xUr5(R?Ttji
zepxzs`3@Z}2N-mUWm^@`G03PUq15iiS5z=ho+5x<L*RR~iFS(D7v)hdD8(vyyQ5vg
zyCu}a^opXF2(Dd=<V=O(^Ac*s%qGc~bmF`5Ji`CJ4&9(zwHWia7p`DncMPGI;^Ti|
zkamWNa*#)PeCr&r<o)1<sDybdD0Fv@Hw??g3(2LCb?z^qD6X~X$)YGc9^}NEY0>H5
z@D#}=(`lVXKcOpy(T%9A(J{zJ$A6zRbwi;K&Yj+U5_y&nDVgNc&EwLN8YK%P2!^xG
z@I4h8@e;O4ie7sHeu(BA-sMY52RAPfWhT@I8W{pQX@$%!OjHJPeVh3e&aIS$1d}P#
z@a>D8D}gbEn48ts&dqLN->@UkM0!+e^cuupHbrx&r<HvE4~-RXq5jL1^LC$fldVDR
zU;tk8<)H*IR%3N{c#S0(;5)tA{SItdi^Q!&ss%(5Lmu>H@Ewihj#WH@mvBgM-@IfR
z$+x~jY_<7<O<gGIg*cw_QTZKd^v!vnMU=bS1Zf1UkVys~R~YTsH>Wg$5{YQAp5@*=
zV@{?~Mv(pW<#b(S<Cv`WRquhDb5gOZ!$0R?;m7UAOrUT%fmefPQY^;^?ENitx{#0j
zKhN89(8i2o9Rcs^a>im{X8<7p$YEUaHeD*oW}{v;8Z7<mC#Z4vaT2x(1oN~1;MJ3;
zhqhL?eZD*O83WCjUlJKK=bxg7eSIjPU67?<Gdmh21kCA12x&B+uaq(xz3phv@w>Ge
zJ=9LN;3DM!za_aZod@~?_r0>NgQf588gF&rFYvhBjY*2>QZ%-<>aL6kWk4A*x=sa3
zrgxR_`~QOWxN3Dx4IlXTxlaN$j;ZKxnEu|t^KYG<ipgf=R8(^p7Xq<B;OMwJd=G2|
zrO!FV<SjXk7kdpi4sedkrmg$~-#!%E(K9`z?Vz_dTW+qIEMgGR#fAcG2JHS8RP9cA
zo~kgj-C(T}4$|0xHjk4a;0ukLK{&q4JK_(VROvav3B~L_afQ<n`7|K#U8Qc>QL<0m
z$DJytA5V{G|M3LOS{8$U*D*tzi0Up%Eyz&Z>=y#1a+QDbh7qw%J)@>b-g_fFe|gCq
zk!pYDE?yD*FSnO5N9fPOnOKiP*UdtrB{xktMqNBBAx^>w8DV597xRv5P=wh$EBgej
z@12`nGksLF`e0}xbPMwe02_6qD3Yc<lQK1o-4K*yaH_W2d@m?Kmk1^xsRV1}$saU{
zcttCnoU}^})1z(2R*VUVr!&C04q`=MT390C%t`yX572;ndo1SDWkX1b#pho7fXy;8
z6Aud9h7*j(Ac1$?AuiKxU2j^iGja5C=dz`q>T|R?l)rFGLZMCO=Zi)iD<ZRSqErN(
z@_3k<_$1bYL8x8LfN^svPgcs5MZFBKP<n16HJWJ8w_c~RAy{xt2S2sE3Z1^%YJw@r
zdV4@yfMxT>QOAQI@Z~?%wmwc=cS(mc(iiDi;M#MtSVBdzD##BGG51p3h+ED5TZ`m(
zxs#9}={-jIP&VYGA)<sHlq#hnX@Xz7E@-{%9YDbi#TLruF>O0bt?9V8d7*xM8ojY0
zP$W!zay^uim^RSr&LdaC5(hyq{~*Iszz{TlMsL7Gd_;M?V3ET?dX)NLRB!gHFg)bK
zcftlQ!;CAVXq3w|={~*B8LnA->0fE!b>?-r4GFy$t_PnGL}q#8c`E<=uS2G=@28Ed
z@w)r=vjaAuD*f>A;1v{`9|l$t&};4iUJD2#W@cqU1JE+Y$LykKqHdSW4;_UY16vv(
z3$q&Xkp73?`@-%)cDh_VX=3iLREL}N+Pp3zKomH=2&CMrMB*}1=gnm6W?f_W_BfxO
zkt1;qVodPZ8e6;SB~R{M0HBMjodVf2BatuI42;?G6;pc`)*6$eZ!B{rhXc522ORZ2
zqhWMZ#?&ddLu#_*H`hEVL1zNw;1PaNPj7AAJuN}PIVgVK-rlaq11tXyR!$>7-3Ni6
z1?zH1{<BjGCf8;uR%&BlOv71YlKG1L`;sz@N`lhq-KVj#qY-lnWCWf%+Gg6Pe`i>L
zpt6Ueu8iO2_s}xNY`I-x-h$g&1^@R?2=s}yv(S5zRC(NslAe42gVffwhrqK(g{)$d
z%;C1Jmw@ODxG+Z8?=s{)?`}aVijhv4NYSOjg$(`6M~X-OPE<Bg(oa>ak!Y>yy}dMp
zQi=RV9G_0l7fgii_L<xLLi+^Pu6Hs8MmR`LP^@$u9R2!?alRI#w8!X6EXSy{R{dU-
zFM{@%KbO)qfsCR<DW=?(whC#3zmd69f2rm)LMvb5i%4~^@!#0E<{q+^C|Sn~p#&|8
zZS+ZN5q!4EB56wa!Ap|pTbEW)d@Iz(Jd2+I>D@$2K5p;TBEBVs(2PBstutmGP0d_c
z)XzdX1ehU6bSRU<w$rt(jj^*u3(54-A8RqI0I-#V_&+d9o8#?;sz_ymJ!}UjL|;HS
zSBYNPtY==Bms$b0xR%+7`0Xdt#9FecUzPJEwS04$ocZeEv+?=ZOb6e!#Ko2S$Lw-=
znHQE8w*NXw*3|Z%46)>M4V=9%oicH98$N;0fm#aAEkbih8*6lD`aBnw-DS?wcCVSR
zp65Giu`+qp3XD^I)qr0au<c=~_a>Dr{(R*QWYOE+pxcfntzC5fH9K}2<=Rgv%@RfB
z`s1V#m%yLZRcHM4g|(2#A(Hh;z;yBpdhJ6Pnex*DuM7D-H7k-fhnvb8dhFYC&(m`f
z#T?%rL#|i*F3jpN6UFQ>#F6Cs!LuN{xu#FVXU7La`SQvn6KbLHO`ZHuGx@>2gR1lw
z4rHHiztdxi0pAxS2tnW^6;${ck_5*0wjD;_iH4<L0BaubivsxD`)V!5!8(ls<m(%s
zvB^ngtS~59c2<`9<f`^T>r7@YJV341Hv>cW>W;x?s(3|@V$Nr*;glocE4OZk-$VNP
zx(Ns|3I@Uj-$RkhxFcO#m%L=(6v?T$j7)5KjyW~c#h3u1p}Sk^`1VEh%Vf`5xf~Q-
z@Ss>iU%k`+R1N;0c1%ZRwN{$JuYS>A{991pDCeH0Bqs;R-`a(!x0u@ODRQ7r*nHx`
zUv-7n05wkNVU$+c##vfec+x7ti+5;9x_uqB>*j{<Xt}+jk=;4rMyG2L^)h4M7H-@%
z98T7}JB&J!{~jKg$zn13C5z3@FrMP2TU)bjR{JBU;?}+c$~!wd`_uOT{u%hs1ZJ^>
zRv!Z89Bdl`-j}vw9eS+P6ciR_Ck?TeEq3fIOFs|}zso=JH?~)LHj_A9aNHgg1n4I-
zZ=I!5`Fo6vUf||E|45=z{^NQGE!W8>Ws1jp;6*qiw?rDCQq+yjFYKlXo}p5lzOr~U
zOqsMCv`%w$^66Vx;FgZ3DUwRd7&<>a4awMRb|j<%Q<__kSmR0g%P>hPt|U@|X@|Lp
zne25px7xGCzmvB-se{6#U24qL3`A}Bt4B-`sebld0H`D_*Cqbdf}o<1Sxtc$bw5f9
zjDu5jui}KVv42$1Hy`0t5i%ty6Pn+Zz1oYXlO29l0LS{6KoD1;Ci%zPrI)=lU}nk|
zt2j}*Y--Q-1nH&AkkbC266OcLPKcCN`l4jmrgq>_237<2pHPI)t8pN;HumNX@nIBt
zD}_SL@^`(l<}?NQ+n0vC<lprQS+`gSN1}KMkwv7jt-Zu@P7dRSHsVh_ED_NQhdAnV
z`aIqh46dGUP`v4UeD8OfP;>HL3?FyZQ^{^iv!$I_iNU%R_Jwrq<9oSmn>YGIMDx%9
zOex8R8m-ZG)+X5G656pq!91dH7lvpdq?JFlunQrOYR%K!hUc>gw=_B6ll(Xn+^Qvu
z_}#>r%dNvaEV@2~&5Yzjh5Ti`U~T<|($WOg^JHOL!0Ucc&dDfXm|JJl((2VG{pZ{D
zCihEn6Qn<(x2U@jd3Ze)i_7f4F-@!&9jEs#;Z6s7_uHIq+R5LwSg%h8sG2)EP)ivV
z8k-d6lZ_nGioqbnI5T?I2BDEu{xLL!4mdRRw4G+p4hYpDe7p`L_fyE$c4RHCHnsR%
z7rN8qV-LSkaWgKyb#iOp6&tu*RLz<{525nA&bv$_`!AH+dl!~RsC0062<R9#oEh@-
zHY-cwuvz;tg`VUK*c>dO@xr^-*Fu$h!E5nOi^3x7+*^>-UE80ROxuRG><v4+b?)SW
z8_4^af7h~o7QO1AwXLJeo<in5cX{Uvpi{u<nzXF@UivoBtUn8^=fO|%M$Y{PV?Fxf
zfjhjDJ-`yl5boz3$+BG?V-u7Ledlb`VU{3WJ8p*%UDfJ)xJ4Y6A2wtfwC(8{5Wc6n
zZAV3fjNnMV(vmaV&%O5K4+D4WA_lJ-BVgp5KK%O!ZX#VM)NIc_&@lt9h-cYiYcy>!
z@o@KltxqMX&`L5pNxd8gcp$s@@)1LZ_CKX?@sMfa5?`0#1yH5VxC5;E!Y(H%meT72
z_#U<tpOmJo!6WbL?W>B@#04h&*ah2eQ>RhmW>a^IsO=k%<)!lzom2<~i#@vA$c=n5
z)^ewl>k-_ohSXP9PE^xq$`p;BsZP|gueGs8dOFNt&VuV*hL}_ZidnT>43EC@kjG3l
zLcc)T_6v>)<h)o7de8K@ZRTvoW5hs4`mlI~0jWJ~n$RE8a`Ba^<0SwCwc3pbRJ`8R
zT$L8V#Dv!j@~Zc|kS+PyNz>WKAL;B7Z9lHwgQ(qmZ*PJ5B`Gh;PmlZC0zRj~>+5*@
zS?rE=qUaLg=0X`b!8FpDv=0H@rHUGrVbi#5&Tf~I!Fhj95(HZWyEp!(s`ORnGXsj@
zHG^0Zy{POMAl+Y_N<|1rA=AsC6+(11CyyUtNSZd$lCPNaWR`8T`Ms~9AKY{mtodx1
zU<ZaPL%XQzqV`CK_txsWcOh9~R)z2Ql6}T`Iz9bd<3-~@m-pnLkz^|2r^a^ftF2n+
z<&K;8FmVT8*XUVL>|FCshv}uUjUXlz4{j7E?(ZL*43*@-YHK4;+!NI8I$jS@s>$V-
zj>@t}f(|&0Wam${_}rg{B_Gpqn{?514Av8z2z%Y{l^NU>EzOm`y_^5>Cs)-bBw+Uw
zJq?AEmcc6KRMY=2jIUkn&AGnDa#tJ$LA4AMcQ-d{2Fr}J{c4DH!wCd?K5^xW!>)mg
zB$9z_I<qjq3*q|m0G_T79v|Q#C6Ab4MICVeZulPgJVRifa*rXdejvn1YeiTZ7wl43
zZ4IqFS`H9mOi1P#5WzC4dtgx-hu{U5WdiUF-y0hT0Ie!}`(!%j-xG@xfCyEtnv=*z
z{%BM4fJf3_m<Ac-2)QGF9k`{*G1QO9cCM^E3k4`3fc%9m^E9Am@1JS84g0whnKVYD
zEElggAV}!$bHfRzG$1}JD<9=-<xd(E5A3Vi)5f=-O7^{1z90iF!weuSARbgkTgirJ
zGw^+e$nN3l^78t8T5(}vq0FqhtWiK}pn0;ny0WpEscGK|Z-V<_Ur*3{ptWKT@GlS(
zM*+p88qR?^`Nr>tbBjkP!C5}d;$zK+M6>qc<S8-hJ=*-;U?D)q+L$I9|C!A6>c<s7
z=2d9oYnstl`RYA=H&N^Es_~^V^m$w$c<v%XE|Hm|@8m4Fer%ZI=YMPvU?!546`U~^
z?K(+pUW2fael)8*1F)FY^G%tYrRfwH0X+fy%ElJXpHh8k6=PzKCrquxlcQfFyyffT
z7x|c6<kCxNFn1V9yyi{$eqGRONyyCL?aq}pKEr}pC!h!6PNcKS?q0kXSAOR8q3i`i
zg6z6GPiuDq%~>B<4jk*tm7i#{OW0Mi@pMJ-qsgO}N{hCkrnC|9bTqP@$s+`!A79Y(
z&V%CtoM0$=NhN69to5z4@g<s{&kfS&)vyJh2#pBPlu3KN#k$2IgAT%YB2ons(&+;(
z&uO+(>?g@5j@))#y{|=vCXV9qQMBV}PDn2AfI@~_LwKDfWKdEIacrpcaX~4J!bNN~
z`Qpn>l3yIvTv2^{tCQgquCtPk%yY<6UOF+%8c_}XZhoRd?~^<gOpNwf{%iJ+mp<!v
z407x|9|zC)bRu84ygB(C#MQjIuco{uWEF3CmR@-UP5QV>Q}lfCdg|ba>iEww%|kb~
zFlN{1XH>(_Cc}QcuJ}V*5C>O}sQ9ICR-=}r`YmFfV3r4=bov|G>m}H3>*npovm}y-
zaTZ9|DpBV(06zPetPE{!Y$<pc5ue`QI2*KiDBIuRBO?IG6g5&A-_g-gS&D!0@|zd4
zG%m<i0<6&LpFg>f=XMD>(0!S+M#$_u<#c8p+~s!zY_#$FXv&n(w(<8rPqa5yHVk9`
z#;q}c@M(lsF}j1x9%}EBcWV*w!2EDIEjyiPP$NY?;0phwug{St_qSA?jLjyn7Ddf4
zm4t_9iH$-tE#km=@k7|I%P{EC1C}kQ{~;ZpZi1aaWEAs1)X%r`0E4fc%&mm-H`8}}
zds_=5O#TWKSZp9FGTJw?pu({K7L#YIp*2W59cF|e;Y>4|h^Hy3=Qg2#&UIua%(E>*
z^h-;lny=&qjos##Upn@KQTpVdX$b^H!5F<PTX=7=+V@TH1EbzTRF+^NhigAz_Js3L
z7Wp`Lc&x`#HdLb%f;$)^!DD?b21=H#T`2-yZR6wqU%%@}-mf+?W>T<Cm{BQG{0UE6
z82+f9ho|#Lw%Y+4{87^=W*<%OilkSSm!=?O<8N4rlb6Ub%>688L8jRuXMT}TF6z1Q
zWni-}K4UM8ZWO(hnfGQNGuP$du|Kmwom(4iU5>7((s|a3#TJp3leT=k*yx~8r9Fj#
ztW#4yOr=roTqh&57qYFPy_S9)jBSZOWox3GWp-LIYRB-?LUTX*Jz5pEnF3>6p(vi4
zXgt>|1yvtKjfkq!hNK|B8pSeh>Fmi}*GWr$qm-<ZRDr!eH$qnF1q3d=!K||AqOmY8
zX0v==L}gsJ7V(pKu2JIX5WN+-`2`Zq;Wr?BQC*?*xU?k5$y5~jn%gS9wy|~GsX2Gh
zWNX@nQqb?m##$n;4xsOvR5PY*cy_P;qZxX&W_TZ9neE`?h$$K`PU2~YXeaOzQBmpj
z7e?+Drk(S<%4C+12O&@Gi33+@=f2OMec8vxc9AK2jRu#O1MJ4d!f=_n%71Kx)jrT!
zW3(;gR}Q;dAY}7xLG~uOn41O`#xER)B|Xzo{^QcHS5dMKoHE~jo>$S$CJ=ls_l|E|
zOh*2Lv@}A#xU5okF1FH(iPR-hnY?2OBOjh3@eSN01bj=4kE;dJ<)Q3>h`R`15c9A_
zGjv~fe*^Z!)ia#`j(c#`2|bFa-w5FKzQ&M&iUypjeBGk=EZZ5BDY8sRoR5*vtuwF}
zl;lR0$2B2`%oOUEwIIo+*&P?DXeigAPFRg@*-`l$pZMkgsqyqI=P`4}x{vDL@pGUA
z*t2b=n<s)}Cty+g`v)z<rvZL2i>|WHTYWau95AW?0jP+rd7wHDoPh5eb4vG3dq#bC
z<tYu;utmsME;Apv>1;o4%h9FY+|(1A4pf~QPhgeN*!kwQwhn<8(X}@#w#a1{BENLz
zn)g|qEl5cBYAhDo^O*oX4iy7(+$}jCGw{@PR$^qN#6$&N0TkH|L^Ob80q8^m;{a{<
zRL}u{Qea&H{O*PEQ@?-4TAj+@#U(fq_Fsd8Hp|>!qleSm2S2DvgLKUJ*ce*xx2eun
z06J<Jg`X$12DI4F{DFNusJvaW;@8VsK<21sg&^cp2+PO+NHSx2Pw1nQ2Fo2tAIgXv
zXxMfmksJBlQl|dfSx?D)^V_b)c2L8XhQEA>KbiNT-(yGf=Ktr%&DTb}JFrG!K-&~7
zU`4TV`mw3~pUa7<WfvZ99z7)@8*+z5>X-b%k7ub|x@z!mE+;<6NYXV&wcvz7_i9`L
z4=17(LprSPPWzU1n*=ES`Ap`#FPJ8C4~7?P@9jt@PFOyuKG){O3w9%&C!EG1U{QHA
ze}p7-!t2V5Xd$R6S{Wcec#T}OWJMWVJx>Ta+a#W(pD)n`n1&Pc)y{AAPX&;W3szHA
zHqUQ;VXX8zmBdF_%p;jxEMXexW?%D8gOocl%&bJU?*rcMC?}IL%U$}7hfX@oRic^C
zVR{L=^_z(IQ8Fn1Fns$A<!m<-7?H<2yPU<fDE7-8d4sb3>sc~G`_o4+Z3b>$DUw9H
zYgHv<J{M!q#9N~SU*!lqstGkV_7=~REHLl3C{t4A(8kdmlv9^J-YyA^NQmGwx8!l@
zRxyK{ML?M1)4Y;2f1HHTO*3m}!}LbAD$NB*)2e>?ym_`{O8mOmjt!^r_L^N)LNyP{
zL|F6_Vy<IP@e57)IhG|2w)$~M!@ukvANv?41f;H?vKbU)!Kh^{xz?$?RijOB5eP&s
z<QNX(u*<zfYgI$chJs6-=B1BMjvugBU6nl&`FT2AB?88{Wl|Eo+WFiK@HZp^v4SSj
zBe;LrHfYgb&8!U{N>0kuEr<l2we~IN-%iFbGs`1x#47o~K``3*Eiq8L$3fjdzvkDQ
zH*a_t3R=?t+cSLFlyh-c$q1@UE1TPMba3eE?iLXoLLw_(24(V#3)igepBD$Uc!p$R
zut^u)RH8r0<ShfOhSNeOWTkhh;X?+!LVKH=o07+$QS{mO%vH6YP9@Mdk0ukUX%qfb
z_%Rzg7DV?pccB;tavI>D@soU$7qwO80oe`kvh4<GH6V%rkuN=;31F1C<6v@mW|3eG
z1CQ0D<qr^*1uZ5A2c_q-`-g|Kn6#<6&oPWE`2e{IY@6WWyhKow0om;4Nuf(}z>ncF
zK7zt)u@YDMmg{{xu?)Ft))8|d+73h_?6UrI+;ADZ-%U=|X-yCk+u^1=>%)`17J_CW
z?*=WUIY{;5;uMAO=;iO(OZS<D_N~w-x^(B`LG1oMCfa+1?72;V15p%Gr<+8X(?r+7
zw;>dtcVPt7tj&#`_*H<bfaej(c(o%GMWD3>C}?t$Kli!3L8y?BjQ{bIU6aX7PTU-q
zQ}^h%-Z;QlnDIgUOl;lE$qp7~JsKHx{+_|dphynTK5mxZGc?eZ-*WcYDsgl;BAsG9
znA-i0C};$rqU@1PEu(?#_aWOX>eiP(Q1L6h_(<xmi)4BwT2Ubtny8hfUbLu}khV@7
z1D{IJYy3Pa80YW41WN6Am92E@V_+xaIL5~nG52<-`hnTQtj*W?N3T@u&3zvcm>`{f
zU3+~RO0z=m9-ioLoA+x`>g0GKG(2Z0MM3Kcs#_zexp=VD@m-W}t?xICaUH;si@gk+
z7eyxz7sVrvnPBqRdHJU=5{Ynx`RoRtnZ9>*u&!AHI1-{=9s;>@d+(ZC)&Q|c001Eq
zQRa0`yUlj3kCmvN$?V>4UT$YOG0=-HE!&xk74Q2e5ooAg)P4J;_Dm81C1@&0DUXG^
z_@uBDNii6^KcD$EgRG3ec!}DJV@$&MJ=Q_Q+DJ*&U1@D71`}3-UNXH-lBIguuFNLW
zY`uy;%P_anw-5eHtOtJOQ0StD9`=-r%s*K=rhgAd{`S9LmfxELzR?p#@j|HBPEaqa
zR*#*JkI#sgeCogrz4sd`0>sOSixiM2BKdDmZf|c13+naQ_mCTAk*8UYbWQ7ad9L)X
zQlb3i(PmcvX1Tulegxes5gF~HNE^KSMPTxNDna}{wtvZs@wa%|3M;cJZC2qBT!F)>
z&ez#r$U@oog;MveuC7+yJ35k|9WK2GA&J559RIO}1#Jqm)>%j2r+$$=5>bc|)ULi8
zH%p=9vJAeo1a1`ky1~)+mkR)yjQfx$mn@4Ie&t*H&74A?t*8=-v-&l~)&0%cdq!V(
z1^AN~sH$uAE)e=~_^#=Y@6bDA*=51_su994?stSL&ULhFP?DBEXwYkBywsOO7G^Z=
zo&&$o{>^a5er*d&bH7keKykhJy^_IvxpM#MAFa1H?d9bFW+*LiWQgd)aCei?YIGhl
zn0pHaaA>4q8`qXWT_zJp?<hkAJf6uAnKtuv!ZAD?nm~?#;D_=s)xQy(={;gR9E)m;
z#8s@bz6e&99b_Zua$SPVXbd0Am)6AW8&>QYQ?!HgDqo`Cqige1_t08SmUCVIko>{a
zH=r{ozKc6fq^B@;jKoRf(i1>^Jame`QbQgY_~An~-G+oXZVA4&DEo(7N>5zc5W(PI
ztXmj$AlxmTo#euAJ{R_CT%^17y}35~i2zj<il@O8daEV<$MvXJEW0mJJ%-m0_DsOp
zw!wl>MWE>bN)u_#;8Z4^%&uDVE;}Q5gA_2y$W-ANblHA4%?*{?=G6G{b6coanP`wC
z=Jw|{i|HTGw_`qyUmgqS?L`uLI42P$%>0zO=9#|?8#)t+Xv<~A=y_}C?jG8vM5p`{
zU<|T5Z}<FV?B^v+exCk9GGDsqE}kaUUp`b-CsooMdA^4>+G^UB^u?Uxoi%;;;ukCv
zP4($PHnji+B{f@00;KJ5YvH+wuhkBWq8fhAi-(Bow8s?(V*+*TP<lwewDY@^ng8CQ
zq8ZDTTng7;($*K($$t{h{65pcRGw5Pe|XBsB<Q_zb@SQR@`mSeYvmh2909z@DEWeP
zUUN`2ybKW&Fc<*|?4x2Xgd*r<#ryaV!YBU+Zt$^v${qlSM)!_Rm&J6p)=JG5-4zXg
zHP&x#B9f=r9yHs3?&^p|Liw8(tF<u0_*qj^6E$Wp)?4U6{edFJY$e~!Aw_w8HEmc?
zB4~HAW+a(ukNRn=R!iGe+hC_|ds0{$A>?phb%4r0kR6^v=1+0-Chju#R(+751-!S5
z=Gom9$+od==V0s$LL(g;5G-n6u?5?gytf+yO>n7pd!MngaD`a{D5st+RB5nuJY4Gf
zT~vI6(nY!7HB_xEKGVaSZ!VsN@*nzAf5VDf7#}@>yHTa++h9C%OyF!aticL^>@6ie
zG~DTSJ-3~7JlHRWLq<_{a`IQwG0OgbCCZ>(&tXbBZkZ%YVK7%%GKo@IC3UowLc;Dt
zT<9#W$iO#2`*C2Qo}8c1NbBVf57PaX2RC9UcLK7qINk_(kNxf6Zr#AL(xDFik%D0L
z-`pH$@f^l!WpMYwjH#+J_m?JpVpl<&rJ`d~vA2!-;|Fn2HC}@2`Z;Xe#Mcvvmiphc
zem13#5l8i(d(pIdx1!OYy%#bwl3VULN~qdG?l9%g!eL~d-o*-i{_?n0Eclt?i(vs1
zH{a;(G~_9(bTMseK`NOq?A<SY@qCn%`)#w6!an57$|v^qmlXZF&U%KR_i1@zFRlH8
z-!Gs9Ok7cYF2jZbg`J$D%%jG9H9uMTJo7%*MsbaV7lb7Yv3M>0w`4~Z+<%7bnWxlZ
z%JqWe{C!mDdvfx9kaCHp3ix!hJa1VGHsN>zu-+;sGraf3r#IyT6G6zx5X!jM`;RM9
zNy8+(sl<pjK{Jg=^l~e<oC_-&l81j)Gw^hi)^yBSK4?ZdK>b}LYA<^~zfjY7G@Aa5
zM)k!$Ml82Nz_}&yVrB4I9@ZaDB}ZRH&CB|R;$7}0mpPh?tUfQ=^HqnBjnhl>^u2#X
zlo58|G1I(rN?xt#MK|v7@;Uv9hvrS7wOwR?1+8Yc{6^(YJHg+M?7>5xmzM{+fY%GE
zJ45{%O0yRK%`O}QRw-ce>W{MDbSS1516>6Ch^k0E3<0&{w*Jn55jMRWDmMrx$}O%-
zGs%Gc%VXf?;aLRrw8pK!7hd~^q!e?&3Pqo}?Y-&v#OVsk5&|B_0R}aiBTCBn++<9_
z^I?W{=8(NJOHj(pFmq=H=~T-qo1Dgx5m~TTQ2au$wZ$@*&#hgky!`B3P`=K)b$SXz
z4g7$mm!yIjDhAT$1(RVQdueB9_xCpx{F*lg;cu1<IN`w*kw5DX7{!23%#a}6?hZ}J
zayI+rH%Ev$i)y`iZsZ*07pB`4itqiVaruxt{R+ws`|Y?Ild^SIHwuShK<?T{-7V6t
zA*hb<v-y>!+=wf-kV8aws)7|E_9baZ?gT>7nARvpvPkd(hK4aBiCnNunNQ*6NgK9e
zD85;YWc)S=+!I8my^0nzpQdLVD9OzE#Ib+maQufR@JN|3NMRlng5%9N<))?)kF~~V
z?d%BY97URuvL@gTIxfD3wkGp|?a+8vM#z=Goi_kTV>!4Jxi0Bt&5w(Qd9CTuda56l
zcsuf!U=l<Mh+0Xf73B`Q!gPyvTwBmqCox(9->Tqj4xAs@oJmwqbn>7kIgjpnhSvKZ
zu5@zhfxV9~wfrm8&p1<OXud19jvDm|e2kTvVaa??x3FT+dnq09ojIFxrw{%}hsUKW
zP=~LFEy&Bb&r8T?v26zwPw#^4%$I&N&rEUgim;3dmLLV9xBqB57*pABmpcT_d5OP~
zif?$nNc>5pY3@J$B20UwD0i9`t@R`IF`YCi$R{(&<V;WFQ40OT|8!08M~l-<M_tHz
zo;oxF!<krM+Mc0tkv}$F8eyGsJsH|HmX%q=wJYY5R52=)5WtVtg-U{E<+e9r=yYNw
zubfCHEfQ}%rI+s#7H)TtezH?VYiZ*-%kZju*p6G~Z=Hz7SAwFl&x>*3W~>ZOzx&uQ
z%#7FPjZ6O+Kd$^<{X5vfpWFI*ufcu~r-5iT_SzY83wMbMU%Bs?S!Uu}XP%?S^M1~>
zQcDMMtYNLq&1ED!8By1#r}1EYe}cNKylYKsmv=WB$mZrLw$hREgFJw9xxXM*_3}N3
zD-H7Sc?I`(i~aXa-FSy|8-zUiQ7Q}vBfGm+Rhj|O_lB2Vvp=j01XtR<0aidyUmr(8
znPRkQ*3rn9t%6Js@&+vk>zi?4_CY9}^xrT&8ae8x-v($Na{1W(vGxEj49`I~^0}x~
z%pLW$oc<2<@apPnYbART2YiV(@2l!%*;zYz^97+sDqB+r;*bmC`mgme%uiQy@2?$M
zm@E-VH8mE)Fo6^Nd3!kQ&YLW96II+633wwdXJzk`E6aXk#W~0)fRoVz9-|Tp&O(h=
z5{rM;_6GjGZ~gRbsp?02wl~(5pZtp8)va4*t5<=x;?>bs(_`G}*9;ca4+FfWk(nxK
z?MB=!*~qg|`rIPBGFC`B%&<bbaIQ$tekX#njK@>jH~s>7crhf=%N(SxyI0<o#WSy)
zf29gq2z~(6FOaJf(8Tu?A@(3^IBkGI)t%<-p?F~qLiVbr(BXP?;;PSWx&n&8(v+Si
zRyX~awR)$2{~=$wGN0Zv84uZrAjY1&;?20CG%S;MC{+$Ok)1iQ7`|8Q@9;t6%aJg@
zPmd|MV#0RP6Sv?LlK6Hf7F!hOmQuR>V?vk$aeO%<!bTa+vE9z}-bHhBnrfkWUz#LN
z<O0@`UDy$s(%hbVbN2W#zYyrwyCCJfie^4doJawUG+^(XSQ`cWQ~%jJ;0uDNiMWN@
z3dAyn=R>Da8M|20xS(gCPK$;DVy!Oxd-tkY)YPLYEm?731DT@BW5rl&N>63eEy?Vt
zV^yQDRK<E*zhu#aC`O}35Z=5{Gpf$IY0m$jH^p|jjYM7*_iR*J%L{_}1hj-t+OjY@
z3n$0Xk(7ZpPtTtV3*=?ot@_eR^2nkdOu1?(o?mIipLvAPxdd^}I$^k~mAh@Un@^X@
zHQEr4!@xX4ZK?QRg0FXH?FoPJ3#KXme9KkeK30A1YV(aUtBpy`kpjP+U`B;|m21|b
zDX8D-vH!oO2v~c&F8rXr%kOz&{C(aK@t}Ucqs*<qYcJMJB;I>pAL^V^-MUBW)?>>F
zt=F%4y3$)6P25na4sILlz(jB$6F+X576T&x4<mlYP5mCE+f)<h!_+bmP5HSynjp|9
z6u<zn{T4)6r`7p!B>8nw*gDu2bxI6C-&27b7x@W82Fiqs&C6H5;9DQ-8QX;a^|8@G
zsD*r&%5V#t?T_kFM~mqtEv4TFZZ>D9AbXl+^ClwS>tyvXauJ!j+O}@=^Qg04nk|yM
zygwHaU4#A^`;B94?Mji=FgR8Q-kiYmi3ix4M3oCSheJMy@r>E`s<!&2PR|vFE@+HR
zE5^@;ZL71u){*UTLY~*xz8KT3{pz~lY(mLL=6c!gis3xH0#^fdKj^D?QDWzmsayh`
z6!8<-BGF0w?r2nGB(oJbPZ7Uxfk?v`OoK)cXF+{D>RR{X3K7P-b?Y_kgwwsQUuOXC
zHoE%@!>X%V;mL8fL4qhf=tnOV3+(Zu&22RJQALMdcZ4#c1Y&;8ro;UCAMS~@L?fK`
z7zuHu1il+rM51l*R<}BA8go_FE9*ZUFQ!#MO`O`ZaCwL`;4NHt_i#Ta{Y09pQOIFt
z_{y}k00(?T6Nj083&>h`BCcGe%CR3m1c>;Hr~gPalPp<Zm8L3sm#+>qnLk?$jBjM~
z>4MfchXpK}^}Z6(8<q4r`F);BfA=PJnN==5QG9yuKW$bG(==_mr~!oDRWyC;Vjk7g
zSG`x<*+jF69`S`PtwGf1(k#xJ_~6#LY9fjs6IhHknx*F}l4NaE1Mye}2ns5Ye=&%|
z{bP3VF=>l<MvLY9MVL`Gi>I8^a7K3WvPRxJ<+Bm@u!aRs<#L%Nb2K-tfoQS&z3^l2
z<KV){OVb;IR^PsoJF%cp6aGA@^i5VxS&d2DBBw-LG@r!R2t2EYq)ofS^az2fW#yFh
zV&AK;Zn-NALiKHx4Yn%aw%1$N6LzY8N{;gQ4fpgu`{{;GaEp{phrRszYF`)tssi~G
z)NtmungQn>1X%?6v>V0@2Cg3XhyDUH1=!_{hf5j_@{5NnN7LGewarJ+PM)K6g`e*F
zD+~q=m8mFd%J?(NdCB38)DwrOfT=oaumEYKUZA8xMtF+D)Yw#Ex#1XJX{gK0Fk7)=
zK!Dv6fU?J9G46Ud!;sSOXIS!Z@!lI^RlT%4OgX*lV%^DCBo{LQMx{g*GV}`H6?Yds
z6=BU}vx|I$(!EB|7t)KAM=h7utK$%tSDJF`VAJ>y9)x8$`izI9yJ3wz@%Ns&c0b6p
z{=Xme@^Mwz!Nmo%JV2=x#7n=?kU`$fn?)Z4{e0&P-H^v?Yd$c!8xtz1sD-=!zV&XT
z_${HVs|$h8ql`i~kFe&^-&CbG4|nT&CzVO%m6obREl8pT#+eU%mP=yFq7|1Wf1}N6
zuJ<Wa?~`KMc#0*rcuGjQMhjgG9wNfdGn!fD@Tn{2#I3)7!J#XgY=5T@#MK0QcONw1
z*0VvATe+6Les{}CaZn_)b%F2F;cyD<Oi~qm2D*-wqg8~Rx)%;_e<m!D{n~E;H<l-d
znbImM$g1tkU0Qrm3;xUcCL3=~Ef}N_ySfb8YddO_du}HJ-7^<_%h`sI41}UxMn!6*
zoZjbm^-6>ifpJs3R4jc?&1+lyVra1j<8;_;?}fApX-!3doOVWb+lzCy#}ZQnv?7kI
zF?7*d^)w}v=?t=oATDMtc8mz8vD$4#ts%^ngv1!pw;dUOq+inDFDazJ)RH)BhYWHU
zm}Qdm<Qb5cz2L;V=-|K_80ZX%jbSWyXs9x}T`HR>oZ+jG4e4R{er0Kw;`GTJJC~l}
zc6yf{{p$$^6-`a@>e-KYEjochXYn&<EGQK>@kjrzDIO139#(E@I&Zi=8^m9P^-nMp
zowuTA=cpv3&rLH|c*wM8q+ene&a{Z;%gkv`r03S3`PeYnk8V|oq#998y|yav{DShf
zglphu|6i@WswO+Q^<4Yxr(sN+Ncbmt##8epHdRYiIy57n1KUMhn`G!~{|w7ngYn_;
z3!+xQ)CS$<n18nE^^3#$k(!1kJxgv?=7oC@NVjix1A5s(8;q5OVy<Sq(VTrx%7S7u
zGQTu15h~TH`UTA7-K*e0{5OT;Jx@x&@v9%25~|@IqNJp>+u_wZiZD8|Sm*-_#%C<=
zXb^ofd|T>!8x@m*8$tt4H=tU7fX1lHoa`^udL0?1sKc%rFclhY5aE^Kvq8i~YVAoH
zRqKTtV8u{?!{pg3eY7WxkCspeGYku@dH^<!ox(!cKkT>8tYfntKmq^z>VIed$kur?
zYX4WzscONOmwYy=MSg1kmI$(KPE^-?Wm6%A*T!7*_gunV8wLIhhH==i`}S(Sbr_=5
zT;)Ul4XMj6#kBL=kOaNXw)q`1K5qCAsWsX<j=2|kgAqZ~XEyPbruAXcYq)s?>6QEg
zV!O}u^&5XJ+AbR~Qq77h_gu_GZN4(+-N6fnyssaruC!@4*SEE(Jvq!?f8C~A>bDNd
z*#0%3mXHXkUC`;;KN96SeDi^LdOcUTRJ)v&9SGIuUPwN|xp<^Sg^BNOQp8dAs~Ih;
z(0n>0jFCURaFe9%p{7IQce!VOaE#@OToCs8F*#|I?;&mPBGx%qqeZiG<@J$f*TtK-
zw(;anO9CU}orV1of>{|(s$QUWdVc*YO#a|(MbbhK<d1AS*YPv^sGqE-JVdg+<hEH5
zJEdU2rU|M`WYr{<A9G5}lSx>XM2mhyFPcOyG;iYXE~lMt+5+l1qWD!!ZEZggo!(mU
zXpCT%MOPL|OX4efUR`3<?k3u~Ki|!;!>f)`3y@^SdyvUe@_>5jk3_GxvD3-0BnT@q
z<1JWe{wOzYs`IpJ?5)3%$B5HOSNb1EX8}}I*G6HGR=T7Sr5n^sN{ApS-QC^Yd{WZg
zEh*h0oze}L?(XjRH~);xI1F%S?z!jez1O?ev!?sEOkZ(P#`hF$O8xACPuEzKyzQ)8
zK3lx@rWC{B9id@(oyWlW=JnrB1TMY;zbW3|3PfcLWB!P^uCniP+hPjspc9uLUx$_z
zV^fIV2n3B%eRW2ixsd36L-{W5+iTJDP)@b6-3v^UoSNlVatdv6S7?3EbqR$LYpJ$<
z$h2bpufs~}`;DKut?+I0Its+OC@Xbr>aOX4-YpVM2_R+n(zD}A&TohgmZw|ID^w=$
zRlH~?rhDcZj*mbOkHwwbV)tB)C)9vBIYzk*T4X;?CDvg90g$FC)Vwgc8Y7D;m*zsx
z0u6yLV1O=PD&-k%NC;$Ma3%-`>gok$t&~NZKblcF*CpsNF;uOrtE<z6t@P<n?kwrU
z4Q=p(q^!DXk+Sim3_%9V@iSk2Zd2c9{CG-ELIPAY_3u+^Fgum}_@8DpSr}6iC{X|w
z%`u$MIi-<WfBPqQU4%xG^IIJ&^+q1~>i_<3c`oo-f8eg|i!m%}eUYecR*`^#^uLx%
z%T!6~OV2}evZJpZ$%dNRGkJq`QC~{6Gu17~=f1e6I9iDu%#F;a*UeH7t&~9o9<N^?
zC{To{RztM$-oe4`iU0RrsP7dn;_qFUEMC7WOc&bMXJ73-?1mhYV<-t#5nBdN{RGT0
zvJm7_atfWwR9V$&DFz)Ya||FMqR}0^Kd%Jd<ZQ$C%i0)4*IM|cwJnMiuNMAIv#98U
zFU|LGFPwpKnUca%vp!H)b^gT~Q`7NTB8hJ5+!GNlW``pP*VjsvEh#b&1w6C0$|B?d
zzL_GIlHnt@a>CRw4pGp1u@C)UEBMLSb{vp%0%<(B)A$5uvUv#p1HB~?vj+f1DUy%#
z?u%Hu5w4iXPT&D4ZzdzI@E_XXRL(FNTv1q8t;k&J-9qM&eo1zkpbwN3v3#mgAeT5o
zuO??eNV-fT_OX(adWwqsZC6P20?vl`N9`>K>v)Go(n_#!A>51u<JkGG<DtNzX+(A%
z4Tu2i)x2W~<wx(J{6vN!vY;_Us~8kisi>imYG>y$7?ZE<lR4)J8Y6o$hMKCe3&t(z
zYK=<1Eq3o)DFv9y&%u!F{%41n&i*woF0a}mUEd1&_iRKSapO(NiC>+OZ?4V56g^i|
z@>jZe+pg1HlrpW>uZWmVU)E+=Z5n=>zRI8YMcaqs5I`$`M~VSPfJq{~mEsSKC!WHp
z$?rldKVkK$NmNJ2-XvB7yt`@6k&IlXN|xx56}^9yqawm2d^*Nlbr|W#{u&66P9B%5
zsIH2fVwe+!CONrkT?S6z*lUd^?6%<_*it8kWy=Ol5LJ!y2HMwP-VTgcmwwi>aYl<A
zNvoe3j0Jxp_%j!+rq5dhHd6mbE$^I=tFMnWp@XBLIm;+zG>w4q^%wWpu~`kMGfeU+
z4vr&2MZBami}^UONpetQIfv7@|8v03UsvUR0xj^xk$ryKgzi;-dkEe-Z|_zkRm~GB
zhBy9~j#Z&yhY|y&VDGfFq}A*TdVoHURvKx%={WEzi=P^(EG~6djdu*^<w+tZC7nX~
zrD<Ot_N(R=!_sqPWCYA?Rsm=B_;A8#d9fS5ksS(8C?x~E5Ht_(&RYP}|0>PV)YR0I
z|95a@PwY<W%&|Ad_tjokocp4AR2Z42{LJg4h3s$b<!qh9UcT=_Q~PV)K7lwlCs{Sr
zIzRX1xopWm{%>ZsNh$9A(=+yUzH~I@t}D%ryyC}mp5Z3?vi`!oer{q-7`CCLP0U2X
zsKr@4Mie^{ksH(a%ICG$+nv44ZLkeI@c3^aqb88i?`f_g1|mJH`J{z^$tSh!9p9Yy
z4weSUNc?_3zKH%s+w;PeWuNgKqL1Sag7vG;cWl+wR$nQRV`|h&Xy1HS<YHeeQjAHD
zoClQB&m(wC`BFdG=v=kfGniCMf_rqP+-;__9CUcrhSe|#_O~_7<~JWA4v*)*%BC+m
zYHBp8&r6mJ1x;JZzb<w(P>pzegiB}nvDd9loM&EOAy`AZa~;5#0I+F<6T1knh0}Rt
z$~EwTe>%Vyi-)I?QY?6ydghQSS&L<+`U2?%3Ln$=!!jiLcih%RH$&|B!2#L$F@~7?
zm%>7W+C*7giT=Vi)sVW?3lhB~)xt`HhSZPE^(#k<TqEiHJvqVs_#~~I?<JH&Mbc-R
z^tjRJO6d3hG<B;NqyoC+$5Kt6HlCrr0TOQ=PEG>3SOnn)JxB_YBVBlVdEoT#@qV#!
z+}uxq)Kpl^*;=1T`$bJPSWiq5Z`E!0`qBVZf4x&fouzi@*2XmzN<XL)*-hLkSqot`
z8NtvD{2*InrcBWhPed~3_My+btW6%sx;GD1<aQ1Zq1z)FAJc}*4g1VCxxS=SD6oq1
zHo{58PhVLx8<d*(DVnDEhSy{&Fb#UZL=FEn%5@?CQ~c<rDbCmkfrG+8+C)Z7%0i@8
zY*Y=gef?Ba=Mi+4M0PY4*Y4`7_<relrzQW1vW*ga+@Gnvd3^O1xEuj^(>C)2v}{W^
zkZF-Sy$378@L^cXqiUlA*PpJ?AIec|hnH?>@38OLPH*t6H$X(lu3=+s?RU^^ty&4F
zZLvHeL`9TpQ6@~G6oY)7f+yrKZ8DFO1E2fMSUqRwgm#@#m-~_J^Bz^<gCfcpUmrx<
zUk%R3WFzTc%qRZ?s}foP?q$VAQ1znP06_M0^73}MFT0E!9Ugx6v4^C5e&irPC9<j1
zU`}xoPXV1X#3{Ugt|5+EO{HsW!P}j+?sI2Fmf3G3EEl$}KF-@S^YQlChx|72nFq<h
z4DTl{@CpC}&}3U2WPg}v+Hb?TK#*Ub{(CK+z{f7lV><q$CE*gHn{9?;$?a8Eo~LKR
zq3#2ZmlAZ@T7HRW=D%Pf^Y(;Gk%CTBb<@~0(RWjsWRokxim#VcN${3%LKhr+5rUT&
zc%l3?e~W(v8d|nH{59Z9;eZj{qe?;U+Dp>8rm<)8z;E5JEVQgzOo{O(?zc+cRL`*X
zo5ZwHCN72wKt7wc)fOtjO`B6mib$w#eZCo}PfWM<a{A96@9l&s@{Bv5ena`iNA2&j
z<WrW@1li{A)?~yF{^Rv+Pt=+&La50DHf#IHTu*SA0ro^()N?A~iuWwav9AC1jP4sd
ze$)7_ifaPGud$Vk^-kxIqN&u?@(~Nj=1w<Y_E62Cf{_B(_FMbVK02n0Dpspnhtv#2
zA^oQ9M(b@!me(RNksml1MF=8dxOW}JWjAV080h+cdU3Cv*kF{9mJJ(omeBU&v1swp
z<XFnX)8cI@-;DgMqa~kvf6kPSk$|6|_681%=<xP;Y~gy3V;~uFsFKXUUZh67m+==4
zgDdD&uHt4vVdSTW5Eie9Li!{YBf!5nS#(}LUF6WVKiwlxXG|)yD-O`w+f~%rw>{H>
zZ%l;y<rBO{#kFi<<){xKq@KREd~G&?ydPH6VVDIqYU)_$;7F0^U^&T2fIsAiuMf$f
zO4)|ov;D~~C|bKWjEOXE8a7uxiZYhUp9sL(yS^uWTwHRQBNl}sE_f-cdpW}PTrJNB
z`}vs`ZLZzV)0J(w-)i{SB<D#raUKh0YRkF@`}!BCARrGA^nEm3CqoZJ(?|Lw7*fUf
z?JefifdRW|B6mf83?@CMLZw>Z!FOR<oAn+vU}#=n^ziUl;B*Got$JCie}a)kg=VRA
zP1n3`7X`94Z{Q3|EThLmf>jlw){yO7Tz<(>oUp-P8nR05l5Im_pDdUA1eJZP?i+7d
zM~&Logo;B4AOg5>Y$(`|0BQS4a_^<1(d(Y+<qJ?q%sMiAHLzT%_8y!X=X4~OrfRer
z=V%-gMnx(@Lyk?${-8C?zM4FAtA&2M&%FA}W|2_$%BkIFo`dGHZH_CJer#@1fGSz?
zhfg{_l>{kKcNPi-Ux6Ns@-r>7XI@PYL*O*x{^yAh)vDe(;_{-HvOtuEpfzaPgp_9w
z1k)1nJ@b5<rt<PXsQ9y=xv!eL=SPAV8O55nQ<>k(5@c}=V>8N)VC9A!Ao+zIdfFx8
zm5df^F=?8Wj6RzB*s%~w)nH{CC$Arhs+GkfCY-NpWlzO~5O^5U&`;<Mu0+|xez@&a
zMvMs`azRil5SsrfQq>lY=0%*xui7L`*@Z)bi8)B2rOQU#8Ft%CNWA=R!@$!D>f}JT
zoN=nQWYUiWmrd0Y#vXxWq8@$P*X3`inguhuToC`fmr|6$*>|&OHr}KcSO3ix>ly?J
zHi!0TFvc^UgRUuKPGeba!FS*7VIn{2@^R8-I>J{n7csYSkp0HKFGW)#B2IRboTn)d
zyIle0<7Rq)jj$Y>Bln>JsnJ45i|z?29M)i`FNS@9mGQ;9pT+z1&C?P8BU9^B_wwmM
z<FU#X!g(%@1aGrsRPO6s#)OVP<yuv>XPoM?YMNE2VdHMiWDaSY!n<e`bx)>$XgW;|
z_fbAS95P=@5Od!4+Sf|-KGq7P`Jl>9wCA}&mJ0p~D5{573$7{(Z-p&<SP@uju9;^@
zIfBGz?WbRDRfctkl+9ihU)3DCE+LSm*`vu|;T8LxMTUW2?Oy-bTd#=YZpEJTsZC-_
zi{x8<Aw76~K!d8t57{q_HRILWv_?T^Jjm!Wxmfy0f==vn<jKh!@}Ud3ZW_EExzBG?
z-LH>MfzcVzAO~Oi=5EiZfN>aL)_g>A6Y@-PI!C#nE^q@8teNkxH(~z7nz7phB>X2x
zDtK<EJuZ?t@0MLTYZhxJEbc)N>C+k5uh&QB?6aFke)iVGULP-SZw;kLWzgojfMj^<
z!zRn^?1)zni}&pdesaOD4NH0#Vm|ws+_1T;jTy1Pc7846rpo28TCv!Y6*`4fLDFCr
z*1c@qXr%5c(KD=r>Y|PCkZT?RImXPk_D6U6KF5)X{;31=7a3zYlHjI+<}cq|Aq(7-
z!rD!BVmTgblwt|wlxX<6*#OVimpJtGxp}s$NM+`5hiVn_&m#lISlm|f-8QpiSw-2M
zBJ~ZT9X(Ws7?%p%f6_2tna6H(op(piI*3HQ9o&9HQd`?;1K0q`Mx5KKOv)R=eMQRh
zp%6hTO*bd!SL?N{Acz@ISq#&ySFy3Nfwyf8-o%g9KM2r!02q=z>4230bdBC$AQA?l
z{FfUytuXA5wvcn9mS9@x=Anog9LFR`rBM4d$?se4j;x&fMSy{dDa3$Pt2}08mY;WH
z`fz@s<vuoFI^yQ6sO<Kr@!w!PSBQW1vb5L)nXz#QYbhV>)OS<qXo2YoojrMFRo(xd
zSW{cvO?&K=A}IsfT?8N`r#HKVVE)gJU0-M+iQ<8F_n9o;Jib_nwdYkanM@LIf`#Oo
zJ9ZUeTaBmP<*<K?BiTMvyQ*5&9@{k^>(Vr0N;Bi(R7z}F_SA6<W%8JdM!xsCMyfQI
zyJNqr*H<mU-A{3f8x6S$)BDo9M1u`RhwyP0Hdh52<<@>WR#CgOqc#pUIG<0cS!Zy+
ziJN=Ck8fP(!GunaDA`c#@0`(H)%8ub6VZ01M8II*wwNyCoxOW{3bSzeAb6B&ZWq+^
z4rc~=6|<FLiE}P!Y#!S0Xi4~%hqei&^Ih&6`0Dnr>D!D613mNd><qt{Qf69Rk3mKf
zSGXcdy|$1Z?BHyMLGjA6fvt=<FbcYWk2xS`4jO$7WdooW(78+OBrTBR!=WpJ(2F92
zA1~RX4ig9Ck+k3O<a(TP01uMQpjg^!xn(Gg$Jqst)n4d^FSC7|ap8^xQ4OA1>*{IZ
zJp(Qs7^y|~X04d-&Hzn7fco;I08T8$|8E_$2(ymO{4n!reoNwQHq~WlP|Q_wWCN`Z
z-t?`y;qbsRFzyoyP<{xkR=iVZk4JW=$yC;R*K__cJ%PPt9^K0FipffuaeeJ_jIY4y
zQvKn;5=Mq^zGt$zB-C9(hekl+$LQsNEPMittLr!Azc!BnH1GLfI6Y+KgST}ojjOHd
z1CtGrA&LmxbPZ0Pf$rDV6?LuZ6J=gPv^~Dq5K2Y3!O!Z-of7aS!Kr51r?n@C+*qG8
zZr$`-{3`Y1lpj?EVjMGm$|o0Al90Bp+}uzFMCO&yPMeYV15<7AZ#_4*PNjG)Zvj;%
zLUj31F;V$WgG!tqEaLVK4q$}FrHbEyy3k^u(PxlQSUG46o;R-jW_V5Nft{h9=FocA
zb7r1KBuX)ieIcFYPZDA{z6=sJ_{z^N$3NkG{`iWz2SO0{J#pp-6Jr^PW#)n>dB7#5
z3y$tGxTp+G;Xvm~Nw@d%5}YQtb*kNX(Cao+)RP#gv<VGBF)fph&eKS3eWk4Hpwm?8
z6?{cjg6-sVm)ONhaG6;0I=~l1(mRPVleuM_bQTMm)tU>93$WFDhbqEmKO|)xh8>TK
z6nxb=QlSloSyU+w?zx)0y>p>NBEX7;$Nx(kjJajRm%U6rDJj%r!>Y%d#JV)A%b65;
zfv8fZx)CmmuBaUtchpwYmZ+>qz;kRSu#X&$J1onk`SElzkg3LKVDI^jFMb(UiDNSv
z^lObr*r8$yS1_R;vmg!hskS{e7V;6QIG3`C2-hf#?9BGza9@wr^>FU<(PHC2a3qFv
zn)5MJU$$Ui)1QmCKG}*oY`kklqh9Z{(a^3C<Kg48YLh$YM*g%n3P9vxgT!l>baZv_
zlV8A&8|6JO33`M3HFHc~PDPqKK+di&asvbaSQhH67R~hb`i#g`P7FFB42X!QcX*>r
zhheH8`aBtad{m9MJ>MD<X=bcAqJ(J6;!34LPS4_0T-a$S?fs|!{<#Jnrxhx!B<P^-
z!n!AO|K081hA~IvUuDZqD}T+HYGH#P!jDb0KY+CEENE64y*LEFz@AiIRn=+o$!ywh
zFz!kXGo{&ux#KR9a&PmBXi3TX&$(L97;oc#<`nZOcg(obcJ2MseQ!gZ(LqOE?GOK&
z(qd}`Ufn7g6H+qrgM&TcRTUm=zcj?a&K22yXr!$FNVa_GzpSv$6`QU=jN`MD^?QxR
zmgUTwa%3|{CBJH4sW_iTC*wYmLs>T_&bqNOWv0-A1m;Jcg~li5$sx#cR%XIQ-#FPV
zeST&8dcB>DlB2`bH`xsd1vD7n>@MV(97?fn8`j_S3{K9}l07`IoV$Kn)9XvkNF*dn
z2Am{Q`A{r^Fb0M*?&0y<ZbE6Y$7Ji4Cp>AgFKbVY{3}~L&NAZq(SDVZqiR{^>P+gJ
zf(YhJK}s|9-(ib3qjxWUHuWYAy~TAEGh_;xite@((PB;<9)4M^TTWa^0+HMg4$za1
zq-mKlKh4T_-d5+>gzvH&Lq`5+{7}t9MWpL4?Ar0`SFNlLbOcM2{9Q9e=a|oC<%%9R
z=sC1szvD4U)2T+!R0uj3f_YkkQh1oPk+S4MNQVRDN1Wn#FY9an3xNp`*@57D`(U(x
zu|Y?EH^3djBEEG_HL)AIX(!zBI^Z%tADhR#=XRE>>NZxfrQ?+r!`W*k|NLUyTK3q|
z4w~@(`(!M?O4MN-HTO`I%;~wd0K^`t&q(z}A=U<8Q{ir<f;pvozq}V%$GP0UyRSpI
z(E9hI2Ei-aDT3gsp7ij3j^+xDP`U0NBq?4ZApm$~cQ{uASO;e_VXByPwl{t^&JV7{
z3*E&-*SmE$q?U>aGhrzAh@UJ%eCT1#^}bV~rdS+--9B)&8?a~5z7Ayr^S34N6bJlR
zz*$BUq8Ub7`(@|lHC(ss^?wro%bpZi8eeV*fm~kc>4HhAJI%c&wGv=fG<k*dR(Na%
zSS%gDIhd?@$sORDi6e9fKcX;Olo<XWL(g$GEq?H6Tch(Ez~gcpWOhlApa(A0+h&`7
z_pq^e?y{He$+P5tJ7V5Ioz6esqxnQy)pmNftH$#xeW22SFN4D=7yFqd9K*%SYpJ~X
zOrzopZX5Oa_e2%l?Qdbsk97LPWf4$PKK+h@5ZA`Su7kXVw*w(ACDrW)5BE|7>OB<`
zH7MhGB`P=LNIgB%q2Gmeg4TcbxM8jAlaFvy`jsnZVFe(%&;>%$AAM2CJv}0%`Q6(N
z8=(YfNJH<;NIO2$HoNv?xD+-w^z%#)W-4KZ7XKZ)oR3qXqJkrwfOgpULEJEEqjR8y
z*{i=oCJ1YhA^yTz|7pJpX50q_%&l+rmNX?<L}O61(Q;WDoDM~P?Zz5WiuEqBCV>ow
z^eBB6BdM^KkDP*{pe%mkH+!4%f5SIIXPd#@l{tk<BwrrG#|#q2Ttchb1G;<ls>2I=
z|5*4d26yxNK^p2gx4Zk^C<%vmj2=hqTn6rZ%HlupwL2(vB1HS1%In@Yc4{9UJ|Yv8
z&A+EF*)e<+nU+RIQ>IklHjF#Xrmb+n{?CRr2Rg9FN|EfLSjqU7z>W45vKk(PI8xoQ
zXKzjCzDt=6q&#zYPKU<W@+7uo-41?B^22uHRgf7PJuiJf9<g=2T$3amMA7*Sgbm=a
z%(K?^eZ%Wk9Cd_CI*ZD&Kr9Ga8%CjAGW{1Ny+MU3iN%;v=?(q1j9740eC5x|i~7lx
z=AJ@KA(olupK`Tq45&Yd;tR(=hhV<nu*T6$|7=b<jAQO}XBMENUsK?i+OIN-0u#?&
zjX+v5(z2W!wqLo_RcC3SnV~H=3pJdlh8lgVo(8VT^Re&br@Q$XcK277!17uJTRYoQ
zbdWKZ|5)$ER&XCkdwRd#sxR=7ECMk5Eg>tM(-|8<cr-x(E{4Utb#hW}%P2Y#d+IXi
z!n<@riB~J)6Ma#_N&l<v6y&Xv?@MZi8WAwNYQ5Kb?LeUnFs=~OglLsPyVh{xO~J<=
zfSS0u&Ah6ohg^<8EI&z-6THfuNGifN(%JpXWUlvv=`g1+fIzj9@@c#0C(lyN#%bQ~
z(?8~Sv-U^FYOLUAoHXZKet0><Mlio4AU82U|3`n++}zw#D@z~|WeYShS>q7IZzNV-
z6_k9LUFRAEO1fbPCvWDR6|qE^!?9+Le4%kO_|kXu@zxcO-uJhyK_2erI&VWCeTf`P
zwouh$$6-~n3LS-;1*0;NQ_8bDVCU|20s~Xe49O=Qdy#zW@TG;L+g58(GQ6(H$i44;
zoR2^a*7fuc5+AlT+rLZ9jfk0z_6h~gGpAoQn9Qy!o>Yr6p>5^UtG0b_oYu!uLcbF<
zU5X&KOUTl7x>SOH1Gh!S8;0F>vc>MgNWSyk?{i8B!+I{NVDlmHT^Bb!nbSDDbZO`9
z%p*L7>^U9-At1P<&7P*Yl8Uz5t|8W%5N#?EKM8T-TWD3&h$J8g4dBclr5A=PEqukw
zXs7Y$&ZfO$(YM;przIF|V9<=^yNX#RH13|uo6Drayq(I+vs%RB-M##HX6@?r;Mbpk
z`kg07Rm6MdALTYBLDRUxRB+)dW8DUN0*tT0X`7&bV^9`_02tsaHR6ix{^`V4v^w{?
z;Qr)OuI-wmFZfkre$ZQx^jA%_g{&x23xQ@ZnbP6X{2vP8#1%E+<o@Jd|F5VJ1p3c;
zs4PX>vHx6!Wx4~5jZMGu?@#qCac{uY#{Z^QVKj0fX>7!~GS4!>#?ixZqb+Ki`A?9B
zET1U|B_+yP=dchZVWJpg^H1T=O2$`Jj5J<ER2CD~wYAxwM7A6vF|z2a^zS&n6d6eQ
z7YUOxqvn0L&@i1`8b-M1*X2xcWF!t=uB}~esg~Y3HMTq+v2+q<TRG;Q8VLAsO{@=u
z%#9cJ!g>V%)^v%wBy|iL49&3>+`Q%G@m&)W((;Du;eHwYDdKXj6IsMRNopLeX16JV
zRtc>9pWS+6ehj<U_h{xHpvb?XTCk>6=-8P@Tmdp>Vi$K?BPK%G0@@|}%4CL}Z6V^2
z9>D_T15I(XeK<Hn_)s^;xj<i^M6*(XBeciuX2Qec^79w)f6lsLXKoU7j>%=)0=zzN
zhGPxp1hjR^inIPo16G1h7CNX~@EbK#9`joU(3?8KTA(sHiUzfdr?gvE1&oiJB^4BR
zX9@e%RFmr%8|TA%>80aVIhH$ljjM6&nZw$#H8N*NxYd=qGMzKj`dXwgRAVG+QXPdj
zEju(H7go4ToH?}(r?ipW;t?S+ce|tg{gzL`jL$7d_21fsQr2$N4p!U{-6M86CnLVp
z&wp#L^kHh=mb``yvt(uF5Nki2zkfLSetiCf<lNwXOr5=o>4Oa$8D4pD;ilTyLmWH?
ziH?vUsBdE~ZGAKR&PCDsrh_)XI8Z_Xruy?b#qnZsfEhv7ouPnd^SDWxUUlAb1}o7a
ztM<m=R(yY<&br_C`IQQjP{D@iX0aZ#sCD{~zK(m)odYc;&y+Dx4Q5N`)&A+s$dc!A
zy@Mhhq0OpY>T>o#Xc{jQhoEXy@3_w0*ol(umC@<@<;uxf@W0&O^k_?lE^<xdKxM*b
zC}cNoNR{1U2<F|sn4&Y3ou0)c5ldd3M2T5Mc^pGdGf~j{iI)C@Et>**w6F!I7>(x6
zh%YZgr*@qfdNq1S^+%S|T<96!XGC-Y)^tSYwB8W8&i5ndx~+mjzwUqIW=rnJ6o2|o
zhUF9uEs0%bLu8C3+7=x}aW{zsw)SlmUN>P2I^9mv+tU>%xR6cjlL82Afj_ajFh)zQ
zb;yCByJ^R}{VYlB$jCj}RZ&avXfw&Ya^~%W3rOJoeNGVdkldO7@ABY<uel@yA&^V1
zZ;8&rHhy{GB_x>mh7}!3T<Mqxs_3#5%EkX0Fh6O~jdGWV&iOzl13yPx$A2Rv;UhtK
zbh|4`o`m=bN$F7d?m_QYQY|cXF5r0f_9b!gvp?zCtu}w>3Y0HJF*x^>NO3eTlF*l-
z_LeB0bz$y07limRtLvN=Ka53qb%6q~k;O)5!fAHqC6%>NLQSet{e%Q_jq4&GwZ_MU
znH_W|>Xw!eLhM?7@TopsqJ)|)jC8CB&h)+w5yp{62XGw7DqmNhpo0Cx?*D-j4h~(w
z^#vw*2CCBpQ_D@RRk~NeqCPV}Z@$Dt_ScrjZww2N=<Q!_&;kppg@DP=h$M2$U=&iK
zUROR1YP^9{V~8O1-#X$#^;ENIGNg_2DC3G#L9yHYxEVS#jnW&mEk!)py-q(8)@uH-
zRk=d=@D?ZKb+w@$@l3by95Eb?X@&vbpTgjYiHV-9_fiu^=-q<FjiRXpk^PPShavZM
zi_UA3l<T-DCVei|LWk8M3ls4en(K6z*7HSAzU)q_%lewIh3_~e5^dO>!wY}cj-mWJ
zkZH(Cucn)$qR=FOzcfFz$a;rqDP@*>@?Q;?c*26)w$BLPGc391g^&8NhCVwXmR`jy
z^=4#2W$L-3kKUft+m1Xpt{o%yj-Ir-qqc#Vzh9gL5l9FZ2EwOWD9B=yHz&jcSE`BV
z_eIBJ+f4?$9l~qcw+388iUb&BObH(PT)vtzglo67>CTdd%8IiQh0iQ6|6A{?BwIp|
z{;00_Fy#F}{4_D*-3}x$?@zk2#V%(zY>JAkJC_a4*VGxF7edj#D>kwl3|0$GvQ^aS
zmGokFp3z-0_%#DAsvvZdG-5o+oF_TL4AUKwkgwoq<X@d&%a#&-!OogIX7JT}W*cne
z)5U0uZ1#9(mcKJ}JG8(fHG7Ff)Ax@D@K(?kpBF1pHCw#3F;MMz@<KF8m5`w>X!y<(
zy!kr^y1Wgx|51s4Q+5c29rr{2#BUPYep8`!PE)}@ZtYkhmzUxk^<+91OQ2=$`})4I
z?<k05sgreePb{(PxBE{i;M0Jpd*7TBu4t(4DDmj=rPz4R@Z|45#Z)Zf!TtJE$q^UK
zOTVD4o<r4fhV6yvJnGURWB>MRI+F#9s@xp@*t{BGF}ICD`*mM~<v>efPVu_@(Ba-Q
zm|)M@QWSebAsG2Sk;>%F-qph|4w=*1Cfjwymthwu(B1%x4NwFvJeDt;oV-i#9nTx<
ze^|!+yjoU{yIQbHfU<cCz>SJzV*weKSY*3$7sUo|`xhXJ`5^j*iDYNSouq4(1=iH|
ziJZn{JosfiKWs}eD~r`pLCgVrZ!<=kPjEl7gkr?r3N$B8O8eMHQ--}(dWAD$1Rx)#
zI6OZNwyUSZN>X8l`V7WdO|gxcoNOWmy0C{o7{zEXRUu3os@B2^2pe)GNvbd>0CDW4
zST&jw?Ve76&B$kNU#w@#aedinYtDY9&^EzcpC<BV@zsVf8B4-|h-JFrtO$u<<|#ou
z;|DvHNR5_5m*0n-B2DPI8T#%cmWS4LMa{#FZDrMEi%?A?uEXQC<N5oElz1@MxL#fq
zPQsGT;7;dv_m{)ZMWX;F&0wmWw^i^W;^w|ih$T{@EO};MLd#_M3QU5=n$Bnyvtm-s
zIh<}cCa!O8Vle7$?K9dlmmR)F6WBzzi%|!bbmsW@RF<Fqk-#ZjRMqsIIvOk$Y@yt#
zMP3YhU#Q7uC6`&<v(qM`IrzHrQ*MUbn^Mfd*>zU20BCUbTNXGRnTX5+J21#rO^zn;
zEi=7epQy_8N*lRZT)zMEG#2s9_8jrN>D|sXFtiW7DSBR(t_O8-<Q^xm`7GKpTL@(u
z{^P-m&0p?1$A~apLaOjei)o9ZH|_cW^9t1KorJH<&JmB3qq@4D1p~wpKhS4!L{S^>
z-D$ITFYgx`JX74Att$8ltJF4gOj#JWe!njxX=dzBY=fr~drj%0s4K|{lVkc*40U)q
zOHe@+A@-8Jx{m-&p6_nwjn&d$|8t7O+H=U{GS?hUk4y2=Hyhb`kskX@^I?bPY%G^0
z=x((~*9cuT#?|kR@l@&URP8~aR&p)5A1$<Wh7ku|+?&>GO+{Txb=a_iXL31FSnpeg
z!rSO3n+;)qduozzW?M{?!mApI0oc!24%NDISF!Qa&XwGj+2*Q&mYu6N%ZJC^_zX`?
z)4|TA?e)^F)-Pjan2~h<VlvDBm=19Qe#~L>s5s8f)jTteyvR-&9%m>oUmR8oCJJul
z=y2n;-}fp%n2j`-3W+30X<A&6X_F`X;JOZ=^2)!6DNq-Fq_&ruqb2?ezleABUJM%?
z<5O%X#xa0e3mP&+9`71k?;9U({_$@uKcsm3*2EKYSe+<)Un(y&I!BN{tRTG*h)$!!
z4*#z4VaDYZC}=mkb%U$3bl+3u$V+w(&?-+?*3a=Y(&?G8Ybz@-B|=*v&P%R9Jv{=1
zmSh5t-&V1l?shV!<ODt<h9v(?LMEcZL_YA{Zs1rg{&np8lxbOa`M2UgjmL6Ys<hEe
z`oK+zhe~zfhph|d=mYRkzFR+9SOuJDDb%mJ-QD4~tOP(d1Hv_Ndbg#b8*0wnV6IOi
zXc8M@uQk-*ky1N2p>kZBcGeuRz(G6ox<dlM@^0_j$im7@@$iTrw3>w7DeBvfmt$fJ
zbIIN36wQfOMEeW{mixE?jyt##$Z>i}2Sj`iI}m;smxPo$bOLmD6C8mQ0#hXUoO08^
z?j5rllQ&XzKAMb!dY9ic=0;}nDtG39;WAggwruK$IE5K?>{d?+Pt)}h2B+(3Y+u{D
zMxR~-rM4nx87;%W*r|kMbfgg>eR40jh6NMb7o$CqlRvE+H0;jPoOjT^FL+;T0^xQh
zN%F?7W#uhR>&oyUvB$}h`@_+(SNrnAzt*)e=jF;ea%_wFxt^SDlG?%o2Vq5A+}r3)
zz4mV569qR%r+^X?P)$xYO;`Nmc1n3!=L<>3;AqhHlkfG-YJ^vy_T!%N!&*Y?+P?O4
zG`ZIuR?L{s`w^hsnW|OND_ywJ`L?XC{eDG4`#l;SV|uqc_OUqy^*dq6EaBS$`%1x#
zQv;uA6mdh7#dT0vO?ubtmB_GUUGA8W_!nKE#gX?*Se5ksw|naRl}jEIAI*}u@2>vx
z$$s$rB>7;VU_Zdp0+8F!_u8Wp?0gR=a7*4c89<*-lDzxn{8u(_me;L0v^YUmnL&IX
zPJdg-i696`8LBrl!(|qkzc&-aWvB}OFWo8TA-R0M2d4PhbM{wqm^;VRoMq!@(x$NK
zJR(w<yE@oC7_WzQa(P8O??<Tj5TNsPzOyL1-^v?9bewNXlcSh#4<z^M?F__da6R7&
z7>awqk7QufYqIUmEL~*LWdBn{dg!(P>ecD?{M~Kg$o$zkSd^NcpY8#L_IKY~;1pRG
zkX6J3(O%*<={>K<yUQ(mU8k<21Mp)nD*~^}OTiNW;{hZDi`VI!m$}Hxw(qUBN4vDw
z8O{qe8Poz1MN*-rl{-`<6!}p{6=Q*mW>q!zbO&wi;ra(e1GXM;;g3&wTF-eNz&E#)
zTkr5*c&{L~PPYyaJ0mHVbUjaijU*;Wy?*2=!*G0f$VL>FTu%G)y(sGa7j~X`n>6vH
zBNtTTvhfRs-tF|$<6ql^<26T$k+6*8)b)b{AWrrrXnXZ7KQ1mu84?0+AO!{)?@CV?
z@yn5~3duGZ3kipiM^*?-3hTQp90zZZ2$i+^=DH7Osh8(rN{XuQ-yvEk_l+)Z$XqJ5
z9r<EdpRe6C@y+_4<B1bjvtRZt%NvuvL5-E=l&BY>p(RH3=-86u$a&&u`O3}~N<)aj
z?-e-0=bZs2E|pauzoo$V#QL2FB5wGj7arPFZ!8w`1k@CM4@Mp;;y!RpaZp#p*{d?r
z{!8tBE?HtuT&T7K+i#L8zfkWQNdB~QeY^a0;hk&8)!B|2n^kC}kZYI}C{WwqNzXE~
zOd6ncmzfvYmg+0moAMgS=>zRgp8>zW==a*rO(MpUcLt+a;*XIN#lB2_9XBkEZR=&P
zd?ouMC}LxwZUlSGiiMb=#MXIc`irp{2@d@VL$%b-YO&=X4gbT>f70pEMuZXRyqk|0
zd^*OJo$eS*>pIUqtv4b3+kJJdTM6FJZ@|)nbTI*MiQ8QLZ;(X{r&}jcckuPTcr|*V
zQ&HPg_#4yY=u0;ReA?H+m3kE_vWm&hR*Tznt5VUFmS5f5t|p8=iWCb)b}g4||Fo7w
zUNUHRgcYG9vZ0@GBr5m_7OM%D-14|h3+~PRe<QcM%a75@H2DAoa<-|gU6m-6wA4Dp
zQ&-AHkXJvhEBTYH-F>tOQ+kv^c#~HcAx@F9Pi<4}TC`g0=j>1Cm14shjp%J*WWKcC
zET-3EYaGo@YZ*qK^_RNE67>P*Jkz50qnjSR;skLni;uMgjn~WG8SCroVMA<uH#Pso
zm1E5lR6tJNTDJf!3Aod40e3E?-rP3&^k!+sYL~v@f*TXMv@?&iaMieZO+^SjZL!p;
zY&PZECUWoKzm4HHU7HBnE`y<OaH>(WEO;;6gBI9{i2JqN_I?$^yVrw{X2c$prroai
zVk9TpOsj!54{#g;naRWXI^xg9$$3Ko8u5m3sFdhS-QB|LqqYD`Z&tGgxPBf_x7#o9
zA+551V50x>2LMRrB{&S!ho^^;=TvYbhJ!~`>o(9ntR|Z}M1$dk4=x$)#k@ieXF;&C
zQ=AF^O=00$n)8w=`uV(q(|nI&Ptj1hGi(rqz`S)GD3mnow+1DfYimDWlJ_rcCeYms
zrl_}jMP-RYMhmUpt*{5SW_-Gw`typme!r$hG*J^LvkK{f$%a)q*n(%>ZGh9^&kuiL
zE1uBm`|=`beOKCj9z<8oAF9?LmY0TmN)t}bx(UCJXsxCRnn%<iXyqn$Wmdn|VOB9(
zR}^J4@Q{OR^jo=9dv?+M=!C8yb1O;hNygXo&p4y?-=h050Ft2m`4@}KB~*m~mDEv*
zwel4-;XZqP(P$ZSe8<OpT4UKo%AdRMDegbb;BVku*q)9ujPRPw>wyPIGRnN37xoYR
z>?xZXk`flAlGASGielc8vUpSjo4?t*5cYyHq7(hRS|iTM_A~j(VkkXdJlzBC7-TY=
z3>HZ-Cf(riHYOf8al{G#sdjj(tqB5CLf=$kqu8y-XZ^nB_%Mtx;+jAqJgxZ)^YBQ$
z_c7V?^7hCKVFc^zQgV{7rdgE>yiJJRsa`Ay+icXj>||^tVTmjWGWZ9FsGR-GDkUU6
zw6W;50fw^C+YI9d8SH&3%o2r3+12S?1jUyGi^lA;Ww4<%VrbL!sq@2jv^C*XGwH<U
za4Iy&2hnO-13Jy-%$AG_xkT0)ibXrk<1=$Xf(+e9!->*-Pmz@cAeOK(P;Q%+(Y&^?
z!F{~cI$-eW8=#Y8%TVzCJjdV8DyK--)Z&g-IQuQy`@6TN^vcDNxj46pwl^fRlD$Q1
zg4ty>c+>)}9FnrZToQaFB#){wS4-Vdi@gZ-y0$I?pe(S<0uDpC>5AIFa|B47VdtVf
z`U{uWF1%Yv{5P;KNN*fJ6-XiTI~q@H-_@&D@0j=R<i#6V^o_DTHSe1b!^$*shV??9
z0AKbTEcD6XbNJbIxhq65ou)l#fIZOzoc7T~<Q3wYwaStrK(hUll<jZ@xJV#!OjM?%
zFWFfvMU4YOj_XoH5#|SJ^jhU&(Vr;90hkoPDna<n{&?xj(;3TiWfVwiXj?<>%%P>9
z>6HmP!Lf-m0yLO|6%ahlwr@o^(Sp`#vNS;o0Oag-IdXr1f$FY4Y|gyOJ3`+ZpV1nx
zwc~Z{6k>;=grdYlNN?~|k6PfDmzyt-QD`y{P|Z$H*?=42B^v3x><Q4vlc8TmU3dl}
zV3t)%7Ma<lbD3WqtZ|WzwAW|M9UiM4vzb_43u*+qFm9yH_vp`Oofe*fmFd8tUs;<^
zm#@lXN{BAPUV>Lg65p@o&wugLE`1+<z}QfH6wE~ZM1dMCm>u_ZKk{~E>CxQdY}UQ$
z{zwztyFZ!Y*wD&D5VU4+vxByB<Ug#gJsN(-E>jk)=6BcUt}a#9&}(1y{PnPGz*GX8
ze$G6jvFk@%cXSr<6yn`+daD29N4mTs`gdf>RHCU34k;Om;iU76dtd6re3->gDAld!
zWPv^0($mD_#Hd2t44M7Pk|V<^Oh&kIfs>i12C9`5#c7V2OtfJo^q_scxVZ;dGXnIj
zmfVnz@x-P-j3|5Dtb(C!?a2A|_d(;2;Dmf0^3g&o(zM%z(_mCuEXGgJyUm1Cqy*o<
zw4+dkJsnk4URV_ojzRyY>TgK!Z`v_6wjB)7Fx{4N#)SV&#mC!bE-MAmoE&!aon524
z-pHGW#lV`zq4|0;ifbZzYnsho9f!FSQLAQG7tQXMJAFc#Gq+k=*gWz@GI%6YRF#cO
z2j_yb>!q_PXlov00#JytV}3XNPB6T1CwvqPP$E)RD9$^_EUtIBD6#fQDr`?H`hbwd
zvF)lD586Uhr$f`;uF%-lO|RnC{x^r&d1j_wHOik7D#&Qy+4ZxIFmTz4i=TgczpV8I
zOq`mtKH+*m2z89Mu47tK0Me&3I61rs*uaWuS&7`0A_DzS#z!Nevln7Bzl8dCCd7qT
z@58m_$n5lX<!Q}-LQO4e4R+^SQZGA4V9J=}d=G<W6LsBH8)_@g`p0<C2Sp{a-tKoe
z!$9=;`FX%@bR<zlV8x`gSJyj!#5dp$Qm?aG_3V|4+IGeHcDJ+hf}wQbwfgy1LdNh<
z)r@~ovu64>xa&kn84u$c-WkEzwZb1cGW)r>Zl^gXt2mDvpK63`OQTAzjWqgrsK+;>
zgIKO}-+u(i-IrK-r@$u$mW>Rr*5v|Wa(=JS>-4ggJ8;z?1czX>#3|c*)6<@|*@pKq
z%u<Gq%BjhOTublL%422kRTCQ=A+pbaV8*;a;{ZNK3uK9BskRBP-2H6lSGMunnaa1Q
zcF3+=6a)>*ib3OaVx)nb4o1zkQ?oQb=ceIMl5?IGV)BCLoy!28_xSXvk;6Qgw-Xw$
z@j98)grclSr{J?Z_FA2oTTyaUCJwZ@*1>YJTyto2!CoxNxuJk4vXn)^(Mv`m`f8MZ
z^L)e`L!mq4Y4^Dws_hkYiEjC+v~VX^48}!Kn0v<`SIyT8Cr!g#I5{WmOCWmy<)LiD
zN;6{&xj9|J8w<r~0vwn>7y<<e2#C`mdw*al(kpx0-4Qz}`YwLEYH0yZ9gdr7M1MR*
zfR-Zr-gUB1mIyHrJ-f~L@|H11h_CKQCdmxYN0yGzT^n>b2Sw$>#P-5Imw0|qL>87&
z1>D^aQb;$2?<H{07woo&nZPIE`E|B@H6)B4g70$rejf8z^j8L=A{V|-9Y2TAzNEvQ
zE-P1x4{d0Wi~M9)$d>HL<S;m1L3F?8xj>?;y})(tE=aI_+QW+=_f~J^-#Atj2L<;+
zI`e(M*vnk5b_gR{9<cYtgU96*GjZC=gaRy-#3-=#`wR6N)MVokOj#^lIJWmf378_s
z`lbs+4;zI={k_7TTP!m2)Q)cgVH{xqb!_YQe4Ch9kf+{s@Tgi*hix7ostj44ud_by
zqsg#&etHx^BS>%}ZUp-nW(~uU3FdhPxx<k%JNjrlJc+qntf=Kz<KHe{dXq7IUbbDU
zMP)5!ZB-f{$g<g3v2B<h2@k}er@ICp@EaLqxBGiYeS;Z+ZC6@D*>)?^IXILd_;;T=
zy<?0Fby(;}!JZ-Dly9to?teZYh7S`<-<Hx@|4o~GB?R;4_xsqZH<{{;_zMm8f!+@h
z&!;`BLVL?r1GN<v&!hwOS6<K>45zTm{jG1j5md-Ee(6|_MSkgE0=o$U%~B}_!y%!>
z5&yu-0vLEVU?Y`$IeVqmY8Hk}%y<=XN|B!?_hW80e%JahV&h$%=G@9LN1VMi!a+Kp
zd!NJKS8^u^OMkdm3WXR#pym{~3VB}<;JF2KVc<js^zbI@)NIf1_%b{$V2_)fL7xbU
zD*n%k{=DrEvD3(k^Ww$(9_E!bf@#)Xtz59siaw3GrqTerQ|eKlz?vlj1=01`M8CpH
z3cMMeL5$WLwzA=T6=RpPLukET&^3NGg={1>g_KhAWD<2~yU9e(EZdx)l(uR0n1R^q
zez|yjYhQ^|aN=Mmn5>wjWKPOwRA$7tZyLQL3msQbM46=7{A4(~AG6)wrsZm;i_$!)
zQxtt_;#0)V+a6D;U8>{#Hha~>E1+XSLLR98FaABIoy$(TPZHRbp9J@@jXiain!cNt
z*Zv-~4|oJ6&=EEA!Q=W49skI^hnwAxWAeWA7k(vNtQP9YmJs;k#+aP-XIoqcX<Fw;
zi=}VU)khAmz&DU>eC&Hx2(b$P$NkHl%X=UbAO?zk=Ik7H>8&vbVw`=#Q|3P#&evBC
zd^7Kw?^j@p^u$^i#YA0BH&AY?DOo&Zz7w}Pshrgm$&<EL1hpTSCe9YpV=0={Tc)9X
zo%6(t>5SiB<BKK$vFek8MvhN^l`Sck^Fhn8LuX>M=Ul|>U*^CfOBy%X3D1Yqe=mn|
zFv#aH>r?5ehShz`bDc-L2Vs3%SMqppWj}AZ02yYxA64obu&&U@j{Ho3dh;@RnVlsZ
z<^qTBMnsZs9TkhhE)l7N;w}h_k-#@c1{mo?a8z;Vvl9E7R6R?lJ3^_XRtB9vNyiP4
z5!f;LG<OoIRiXr5G`q%#I}LLED-=-hFYX^jJ~Ig@6Nu`z=tX{jkweFSQRuFZ7R}c#
zx@*%<T^zz3-_p7hzxvQpigu<?JX!i4I#6FzW$}y1lb6i?@y_vRp<#kGZ9ETjm4Uri
zvJI7*dHNUk=^Wyi)DBDcnffnsU8PlNS=mh9T6IUYq@wFyL4x|lu)zb?`=fVCZC2LS
z8I11V;#Gs=gt&a2JX4EK=Dy;;@^>@CP(d43ToK15D`AfnKPet{!9|!e?dhaAZgvlB
zbhff}ssBFrQ2=2Q2S#iex5@03NWyY|gH+tTI!lb3g)!YhVFv4~)%0Z#=|^+=JJkdU
z3lWDTE9%C~e<wUQq*_ZC>uUwh(zCs=zJm9d4^TK0VaUQ}p!P$emv@(TGXHuXw6#Nz
z5CgTezYn`tQlt>?@e|b)q9KJWUB79l*lrS8e?hsee7+JJQG!2#s|bUnoG*XMTAE=6
z?;E;~!eB}^lQvESf5W(u3TB_iLdH=6BWw@Fs8{QZXcPs_LdGbx{!*xMC8~ApKDO~$
zV1i>HkhD!(Qs7aZJ#PAQ9L1DdXOv;r808NgEvbPROckcWcr7Dsyo5suz9T+(&j+4w
zL6_mC&U2bgzr>rSLWJ~rl94M@Tmkmk+quJ!ZKaqw%2*2R<$t&jJi_IMI0I#`+TrG`
znKxiS$=5HGYFag<fY1AAf~IHV>oM?1vBSTk{q~_{r@J)ddTEC2WKkv&C7l>(<5)>Y
zvIVJqI(}m8c5nRmfeP{y&GHP7DB`sJs$7ZM2ja3RhzHC1R=)VqzoGDRKW*$Ecj{m}
z^7-E-_ptBE4_0Ylgz8CxvtUD)raT#mZAn?i`zoUxnKL<`?(gw}zToGV2TRXd1~ApO
z$Jknzxo>M!5iWu!tgo+c7=h*~^#X#&X2Yvsy*M!XO^Ls9U;pboi&;bHZ%J&VG53zG
z6qpe=K-Y5S^6@$OQ>XVziggQceu1*_&L(7;&<W2zYu!CPVkp>Sz#ukI8*!%DGvbGc
zd?v2qd6j37ouKlw1JZxK$yIS(X)%KxE<9Nm?Z&9IE7823XmQ4?duz}{#h2!@-spUM
zw)}kk`~@V06WA%bR*nj7XA!@`J2(ynmR20Sa+ycRJ*}6x&n*#_{S=zi#S<$d@aiH*
ztE2XH?_29LC!!7~%{5;GRK9Lq8gx4n@*f6=Z$g|^do+)#ocJYz_TPRi8RtLr)pWO5
zs%}bCKESZDzuHS}eF(|WjmFL<ofa}frI_|6by1mW8#XEU`qkbaBZ*;@6Jc&%XJAv;
zn{{=%{a5EAMKfc&30`R#DZLq7id*)S=#Gbf5r<`xPfD-r7m5_Cr3<4_Y5M5r-QeH*
zzpf{h`}01SO>KELY0U&ht7WEXP#+4=r7CN*I)$Wpf1>O@tJ3Yi*pyeQk`GoCPNuCK
zEdQ9qN*MO1pPtj`X71{Ty8@9wid-^fpA11E`$o?tWhO8BcU7f}z5?+9vm|+3R~R*J
zlq}UIUv9L78Oj^A#ifz2Zgn0tIlT#qU8DO5GTX8@3R!Vk1;@F?I~Yg?NxE5i3d11n
zH?5>FEvty0`|7?R?)_&S37a@7;)}0*6Td07nZzW29meJa4;ns%c;B+PK0jixQI}ZG
z$njE%&r@skD6Y$4tf)!-t@yQhJkW78Zm*uNLryLEuLu%`z)MEbeoG}*P`r8Ip=7*_
zJS_HlW^rDa;d|7!NiJ<aiMDb7$&b}y3@r#BZ|hm4a`1RzfTsdItroG|A`Ut}UzeYQ
z0x;6N*OtNO>M`JaYw-=?rgXk83H>{tHF-P)n#j6shm304mnuIrr^erDy})A5m&5kv
zOdzH1BkK`8=oR66DJN!mK4aksodxW4i>6cK6wCZlbDKn9S23`*E<+i^@+N;?QmQ4z
zZ4`%6pZ_&)EIpQMY}x%-w+J=*RjS>(WU&DD>OhVG?FfL`U>iO&JFCk}W({u{&T&G(
z*k)w2otPEU9$moS2(i(9Y`P{!$JDc6kyLm0jX0kSPIlDdNfJx_th?kF1rHHu^b^8D
zdfgX9?|I7d)ZGfSBt8>?w80_E!rRi*J-x%)|AI@{!;gf--l|xTJG^5}?{Ev=uJkcM
z-ZRUD?HMmSd${^f{O<EGGSIhbzBqsLB%YV+VQd~maaB&3rg$nI5czym#ZjSrxYuY2
z8JzE*N1*kxfiq_S%GXU}hNnLSfh>Y)K<&zC#l!MV{wX(1LtZpIWJFK&&yFmcTsj*A
z;#71xgRCKfDv=5&Iqywu_;uH%(Q32x71}NgVQg*k2j&F-a8P*^<>V6hU_Ec&;vFT3
zsv@gj89TP(L+)`C*?obx-6)jqaZj~LZ^%x>s?MAk>Rj)-PA!5I=kK_msO%b*EiP;%
zLOBPYDNo#@zJ0>i6=Xj>F450Tv;X>7=*Ji<2I(-;sEMS^2c&$o@2asTmIN0_oaTS}
zv9s|?3-&OXDGDkfqZ0wu(0ob8W=h=#2VAkAWv}4OQ1<wbH9CK}ZIAV5DhjK*#r}|?
z_vx>+V^n5{cl{B4b<_F0)w-VXG{K+AL7+oQtBG_kjBsqeZl<{A)Xz&6SQ<0CuM)JS
z1@Ygmb#^ryE#6=Exu*Rf8L7{e!htm#HWn%(B;re9N`zW9K}n%qQqUh<NjshN-fSV)
zthNKrP*DyyoSyT2wn-%)2NK#b)KR%O`!ABvQS~23m31?Hd&@V{$DTqmwSvnc<lj0f
zibfC!1`g+aHmvpO_*_n4%$ZoUnpR$_w5x3UL>3TnY|B@iqHf<?MB(b<>mja7321tF
z^T^mXI{a5tr+6yLGhaM<O~+>Q3v^WIw7m={9JmcZkqMA)KwUG?C_&*Ab<F>#e=~K7
zX;ESP=<FoN2yE^jV5Gn5cj-?Ut#<J!O=d(()G`i)p($Sx%a#QHE1j|KhLQC7Hb3^x
zDe+eZao`59E<RtiK3d=U^56LOaOI#;&K-Kkoc@~1KHRP@JG{Gg3u{TT_s9KnZvoM7
z(13)yte@8R<uhmv>(LO-|L7SB`p5XA*%?orPM;RGa-NMr-(9sqT0rZb^_{o1KEphF
z=_1m6scTGgVNp{qz0;uY_0d^23*Y8ch&)+Icn82g5?G$(cAKfs8=A{dS1Q|?SerX4
zebr-vW8|5OE*b9A=A-a-#A)SezqmI!Cbm)Lhk5T9Vun%*DF+tbt9NTFvfrVNXDuiF
z1GtygMO}xZYDjU3buJ_vR*PHLUeDy`JNw$@8iVy-Clmv-T(h@aWZfp8D8#yZ=clun
z8K)8B<rC0j=z*m!fRn_~@U*n>e<Yn{R8;TVg+WSEq@__xKtj3&1Oz0cYv_;|x?2!Y
z`bR1)A>9n!DcuY(ba!|8KK$Rc_`+fU7oIuie)hfhwI^2+3UpWWBUK~W6uEkn{50b{
zU~U6>@?q}H)vIb|R=Ow+DLkJ(Yox!A5A>oVCse8(!k{QMN+%_;)zEB9=${m>%|$07
zvD`(|{!%PL9nn##y}o8HXQ1nMw44jTzgtn$GYDL_P`Oe7#O3l}(@h?dcJ@puMfX7{
zL7>K7lkWRAji(-O?uy9{HPt7RL<3MQJ6ie+tRL_K{iXjSrO4HO#+)_!qm7lhIEz^{
z9v4q8PiQ|nfbp5ZYf0VKR>}Tu9U?TVDf=|mJy)DE6XKu%T4Xn^b1S2#4K4OlE~(E=
zlUr}0#a;9B5U^4b^*hU=IX+qMXPl{j>O8ZxdPEO#O%QyU?4zOa*n$YC{EvyBsAm4+
z77;bRKr(%U96#@Qi`#mq_@)VaU6U82++zIp3u~AD|1Pg@c1yuRmXezKZThgFurP`d
zxS0x6bnb9zqm+{sVX`vaKlc{Ia%M^9Epna>rAj~hqR-K)WV@U#`j%0W`2(;Cu6Zs{
z4w+~OjUZtmi$0mn`EKn0iAV(8d&F<D#7*q&?K`{$&@mjtr{LfT9|HIASs80)7yH^|
zhu3q;QDV8bhwok4Es?q2FljwmR^-?x{(|O3%~kVN>VAi%zU6<8Po5&~M;KOT@9}G(
zdNj7CCuVbq#THEM1<eJl>D0#Bce`{Lx9~dOTS%I?sHH9my#+ptfhQ*k0Y)Ecfuik@
z^`!L&%#y8TvZbHQ3gU)kB7Q@6SKm*LqT7tVA3P&h$yEm#rS4r5L-)EMldPFSM)vn?
zftS3j9={|b!;pgcqUV?mCiD}Pdm(3Tw~G{EeV$`%UBW#MjS30m;op2<w8sp+gUM_O
zz>OUxHy+NmTiWq{8mo|}i7k8DNi1(mJt5k9#wDl&rKB?<vI1PFa!0Cs%xD_%z~*;O
zY!Gw-BekI7E?(T_GGzOyB;_YFOg*F%$4@3nS>30Tg1vtXOE$A6h;yR*-Eb;k^E!+`
z4~1ltvj3VuPurGQfPa(9eECRbxgci8Rwju-zr``x|5*G%d9LXa$we}?6J`Ag(x(Of
zQzkLU1LL~lSA%=wH5%t@qWQ*_G!Eu+#H$;F$suM52aM!69{L6?YQ3r48%8$`QRvQa
zQ6=3(1eWeh<B2+P+n4r2q!s+;Srj23ytY*sp3O_i;i+ZC`kyfNR)&vy@JmO3bk`lX
zo={#C&lyvWY6Y&FTW$q{%{}0P5_Ug00wGLb(5d_n+rfiMkEYT!(=Mh;J?7?6@txXT
zn#Vzt-9tAV20baMXYyM_6!^~$@5H!s<;7n9fW{;7JHCq+0&{D`N22{A%JTix%VuQr
z%&9oB?=NU_)KS{kxVKBS(KsSM$9$1A$%kpQmKP|cy()s(BAAe5HpT#MkaOX3fxif%
zp#iw>U`-0v_e=g)G!GaDi!J_`YYO5pokM6yFZ;JrQMdAr#p>O)TaO|%W|#u$Rjd~l
zcTIw0(}vJ?GTGIK29pMZ{4gZx?M1hb4Sd3k@ZM$E;-UTD+S&tXKz+mqYm|jrWdz3J
zh4X5C+eU3OuJZ%Xt{3lTlm(}ode2O{oR-_pIlZ2;+fM!Xt+aq8BqzzE18qf5W?yH^
zuc0q7%Qnj7d6lz$8@k^6nkkCf*})@r*@56E?YQhO=KY_@h6pq(8lr|wK5GPN6_dXH
zY;wa`B7fe6ukJ2m<4in{I~4DXQL?3pc2-i)=Gjn|;t`|L4Pi-(bNZx{5|{Y;?84Wu
z)A80R24e{UQq-VB;ZAMH>nJDcby2H?FI+#_+BPt|?w;sr))oL{#cx<9i&IP|YKF#F
z6}nkr7jpIFX_eAZqrXngw1#ZG$Pt?>XK`l!>1lEw#)wLZethzv8vFg4urU+5&fNE8
zjb}~d+L~CsalteW^qEaQ8CY(Te4%LkrsT;c@BL_W)4lIQuK&6C>9TBk?8P?$3`T_~
z+(kql5R}0FDo@f@1Y^L;FanfA{LF{VuT=};Z{&HLqhTPID-Mp~FGlga+C6Xc4j-le
zA>mmzv;++%cKxKovtUj)QizMBIGIq&!ZzN=QowSzrT$~m3Rlu*O1Jmd>!-{Vh1uax
z^OnMixY75ochTl<_V>&`H+0|fsy#16tn8YUUV2o9r{DVax7@n_d$cqIeoI)_6YX5j
z&_&ijK8w$@tsnNv){+3`F-^?3Z+Y3+^oT0x{dWJArz66QJ8C=(w*H;yMJzkgY+64T
z{^!IJPa61k!EQ;6W4lV%g)#o#81Sz3I@N*vi!+GoTwi;l+?cJt0YY3MLNg4Lkcm$p
zOt#3#{p$q9uBOTi6ka0D$K7JUIKxH)GN2paH&=ES>g?@(6(xif-@UQbr}SdLdnr<A
zu8;F)^_U9pvD^K)^kK4PB`ZqIT3f1|E^O0u4rAZnJ^tdVi}q7<*rs2ipfuc=E&*jf
zC{UzfABm1H{H;Y9pJ4O@OMk_GjwZ42$?n<~#GbBxqQ_8o-51j(;?k&Enw2^NadAIc
zk=$(2W6N^6FB7j0R;$CkDXuTTMgH&J-oM-7>W#<XOrdkAh;6OzuUWzNslBkC7g2>X
zZ#G4T5{wj74M%d#lvx=WvqB{G99VmLS6`Jtz_*FW<YLMtWcc9jycf>)Jql;2Ro3R?
z4_C-2mNJ4O8gsbU8x)>I4~{$QPaR@X#EYD={GMs9@lE1|F|m^A(^_%PL?_oQ=A<Z+
zofBD+6PRxHYZ&$LIV}Va&|V3dhE&$Rl)b!NqFUG(;4t_!^k*P<b3jX>I~H^H>t#Z4
zhc2g3n>R$5cVb5+SH<;XJ8f&|H_sP1Ui4P-Q-}Z2qw8G;)8(NR7v5&#@VRP0DM+*%
zDD9ytUo;r33LL`-Hl;Lw6r6$lgOR{eTTs-srT<-z&&f4u_O{0$pC^r~>#~i7R`5(_
zeVeSeo$c+%?zaokD>fYS@ZYmhn|V{@$G@a0F&kxfqXp@A&qu5{%<%GA!t<eWDW#$z
z+&A|(XNJUViA_nExZC$d&f14I4x~f+=D4BpO3~k4+lG`djtbRL19zK?lzWDP^#@!P
z46r?{bgOk{jigj#F<GY`45rfV<_{F=a#@+T5DEVO4*&Uf^a0LAg>G?Dxpn%9Y4wP#
zFRE>)c5E!@wFK>_MUEuVibbsPpWx_3H99wrdfT%&E7uf>(AF~Ff7L=RUwWW9O}*~r
zfuKSbXw@O=!LX2R@J>MKn_Z>ENIYM#&zoqxRFvoJtz&t?2AGo8BV<1@Yh$h%Xa~tU
zfT2f?;|CXM|91hcd`_iI>hM(N_k%RU-aqEG5*R?A`sIQ*=ty4`-#8r>UwoS9L#%C6
z+Q!Z~#jt9EPZ*_u^SOFI2Vy!W9ncwIkRoBCGAFk8G&)|4C|TOmjl#Cic{U{5?Z;s}
z4YE5l>Z;ZdEgAy5nBC=lV58oJnj?^Rp=$^|<c1mba@{onEtMk84|ZCIJ4>O=Jg+Ab
z6we9ptf?qOOfmW@R<YgGqrQyMmxQ!Oe^wNGBk-_+xta<Y>;3xRF_AY)iTcVJqYtB7
zY1?Rt3Y%;D48Gk!m$dZd!X|ALZFXUe-?hI*dFJOkp}fjBSfw}yRs^^~$9S_F?}p4O
zf;7uMZqVJhDXwB9(<_@78kuh3Ooy8A_7rabHyB{0uB>B|<Z}EPX8+qwctmFw#)$a|
zBNpF%*qf83jXt!K{&UW+!Ca3`xhh5qoEOhR^gnV42#K|`b{X-sYEnaaX6eatK2+d9
z3ME|(R@pZl+1UzSKC^DAU{-;JlF+DzPS%FdZyz_bT$(;?rQbV?-CUW*xV&HA#bJ~|
zd!~BGWg8h*rdEnFTZXwYi&Wz4u_eU()!mL}aYL+>{M}|#>FrCT&!$hjH=(T7_0Lwx
zwJXmzHHi71AL^%;4DlE7@s8?hrWIhNl`FyN0yFrckvIYd>z3|y;R?1|Dx-`fU&=hP
z4tA0wZe)GSA8c_2*tuL6_|sM)7&UyRSH68F<)7OjA5?G1M7kx#?W6|qz8?Vv*~ruy
zw8A~oOXL(8Xx}b~tSU12`Q7b#nAXd8ff{-ZBgK*kg=8xmZig1DADi-|dGVZdV;f+j
zjr8x@XCaS&-TPDV6I>&LkRY&?ex&^f3!93i18@ZY!X$d4Njg)(Nd;T7BN?7lQWseN
zf$Q}2*I6-VEwI!tNo_<25ITPQJVj{3$cCOP7jxv@a_N0n>3=bNb+sCmd}UYB(-hg|
zIWGW%f4*yEUt;$qPK4T&C`NX&$m(D_*$?P<9TEmiS#+W@_k<L&8<7g<b?)yyDLkP>
zUOe`)N36Q$x7OsRnd&h7JJ=!gbQ5tqfh4kd#Dd-TkOpR7cAH<nV3i{?#02s~jz>Pn
zILL|jR)t$zg40_yi!5vUlyfZf<5;*cb6;pPR|<90(l&fC!>yaXLoDch79s8rD>wNZ
zB&$3_cbSIOYRsnE*mx`!YU7!evB{ZoQoGxl2`0Y>L>$>fFeZ&{T3~3Ao0kJgtl1g5
zlsek4V|raK@~_#`!;hs}SSR%Pr+;)~*xHz7%k|$`X~)y0$cYJqQGmfxR(U%Y^jm+Y
z=Mo@&a(_{QZzq*cTjq9o!6{RzJM<xzsCap@S0N*kTSg_9nu=x_)^3K`Ana{cA5z)2
z(7z<bDL|;5!?_Ki!l^+b7O^8VB?D*3VzLNTEE^qm)^D*~EvH$F>;4Zd_ltGThscbB
z%<^GeZPk%NmJUrkpZz?oyU$Oq+#@_iLh2`c?NN6OUb#piG|EJYTyp~QZ(eh;>l(gt
zjU8raEfpX~U9yA(xDxn}dY6&(jj6^l$mledb_DbL(jTu**@6+Xyi%?H;(LkOQWw-3
zy9t^*Rz0HjUrMwk49z;Z+$OztxyE$sFa`0q?sc*sg@Y@yuhFZFl1{Byo3H;pwrc~K
z3KZ);Hv?@2utBd6;_ff>dKea-*5hTAvLX(C?4BkEHbODJc-Ttsv+zX_l-hEVfg$-+
zCY~kZ-HpnVsnAf36b>L1^;U#G{v2+l*DzyeKw*R<Iee;2OG&W}Xb~9}6cUomH=+T6
z-=zBLXF|u6#8YzJ%6pepc*X{(XTYo5e6@o8aNhC&2cF>4UZ3aO8VJ6dYLR*@57-xl
z+Xblg$JbgjK3#81nwV-9Ck>RfzVN+%64!tiIX1lVX8!z+&w+4cI#Q5qk#u;bHcjhX
zqG4g*t>4)+dX8arWq2_gETXXk-gM@v2~8`?4G3lN&g#TH!+}^)eU5NXm`FI=3;nNe
zhC7XFQWoymr%n~wo`AQ`%W|3Gdzmu2@O54;-N5357p4+x92c4S2p$eBlz-3p_p+C-
zSIyAmyG-wj{*}{(ppr@neXghH6jPV7{%UZE!dbl1KOf>^2)ReSZN$`MFAI6mpZ>$c
zW5SEjI{CBIROlLb`J%pM*xD|yPY~`dYbN(}$<xQ>y$+E&C{|<qY|3o=rE4XW)6-T1
zgrZ1DH^HHR6Y@r|)-P2e;dcttu!6cK8Z7NHPUDL_lHnKmpxgpy0qoOj0!Rq#mm;ah
zGjju{lNGc+@0i<jiyWd^dLmCFuO?R&4G)S5JZbZC+}1B_ih^^9SJ&?xo!r_gvz+Z3
zy<>s#=coA1>c9VP51W+1PH>%z5fdi|?TE~LXM_rd|7rP@%%!GB8`a&{;_0e-RH}%L
z>2uZ>-pEwN(woG(lhl89pLO_-!jyDUN_4jIBZAclzeon3xW3!Fd9w^PUU-Vwu%^Rr
zO|ahxFQqBid9D7uy;D8DC%3?nuhbE;nn{Ej+AR#}k!CBOd}4otTK}xE@?2sl_`Ebw
z2Wpy?t3F6MmiKqnt{{FrqF_p4e)l|sucSCVqKf0r<IuW?P38USDMdm-0lDej&%x(Q
z#P&j&)nSu8Cc^X?N^9XFdPoZ(JNu}`0$2~cAI6nT&mb1g^OpT0V4-5PXU|~ZY1>Bk
zyry1P#y+$a;-idXT_Y;oEQ`1D#K)`2U{#;CV3$V`v{c&#q?VbM_v8n_z%ANEX#MOg
z4!Ce(HgNKeyG&+s-nuf8Ovc!sPGc=$U77i(wtO~tLL7$Qr1tM`FZo;RvR{QkJ6CRc
zeZad0(5%C`z<;eH2v~uyJ~{=jHviqFpMn|9BCYEeLw5RXhB!MkV~#q?V{TP&0{tz7
z4W1ka{}O${T|-s1>IHi?%ViYjuka5`nB2Rr8bQnUdyXdel}*d|tzvG{g@`?ZtOxv=
zY(z+oMNbZ^`IcOl9Q8a|jY!uDa4&@9;nd$T7lazew{*9JERQuZG9m_FTm?c{E;PE3
zM(EEsUKhL!8>RbMe*D@l>2#;RbUGb}3^ydZL#&UikxyX%lO`J7;jUt-1qShMY)m}b
z`?$WsNvmC@#0yBZ;SA-^BfL6x&i{5r3KrmD686oeC{>e(dMFJ?E%+vZW!|rYH3<9_
z8ND0=j^vmBj>tXZ{rj`2?{(DhG}UGKdQ1hg9MN<^HnZjyeU1}x{^;*?76sUV&VVIo
zD+>?Xj-&Cpi|!p_m$3?62g@gC-68?emR6x2BOViX`=~w=fzrRHG1orbVXLN*k&zOf
zH}}-i|CttJu+0ivkZy{2p1r9#oXdezL|gc>Na6_Du-fXNUevJU7sRydoK&=W$|hLo
z3pdJ$a|6Q8wf)0V%gR*KW$pdaD_~K?Wb9`?Wyv0MvU{bwQs1+(n8=^0^m_%BNZKvl
zOYL0R%tlAeIkm9Mv*|c>0F7fHxxl{>gSa9xpxXw5@>0X{?eTs$YDoH10l~F8*JuZ}
zA)S=?GsopF66WF_-6&_4X?ZV06!qRvRZD*dEPGd9?MeZ?rBuw{-y5GCKIzz!L?>*t
zZ)G}HztFY4*g3NjBhcohjV=ESOb#Bn5vMVyFzy7rv@%@be-~?`IWIFZo7!OFvA9|f
zB)@5bBxeHPP*q$qeK4nZoqbPDr2?zC*xKenJokLrRYl0c8;W-+3b106@N~NUQ1LMS
zEcT_z=IVQQpM5^s(Fb?Z)tA!R4ddf#;IB%qS;!n}@ro#b6h-S(7GM}MO3wi&?#%pr
z7g^=+0#R{sCO1|5TeiA?U16k@R8+{e*{rek2KxF(peqV^*5{|^<(_BYV+;}B3cg@P
zP6mmqyJeaOx*Ge%jkX|+#{#%)EBUldiVXIH7AVDo#yx^B{=2y$slvHXz?7fg?L5@s
zr4fst{ao5bz0GjxRnkw)h+oB5BZSExI-Ro%ah(jcI9EL0^ti@&PV*t#1T|b;RzJk}
z_m~ovp8Io_YHzX@jSJAI3-s+SrS1?JcouOjQ$p7BtmSqTDoz5+Sp+^13kyk@zScJ)
zJEEifRx0tUlmB#xTU>B;yR`Wa|BN}m!uBpI_uH%Y#_@R$9N59N<0yqyo&`{Miz~fi
z_lSsLJvN%eQLpf|j`y_#INifI!^p;rJO9wUC5m@(#RQCO=O|#GNGwa?x$gTU<Kvq?
z?k-Q(V>H%B>LoV?@uK8d!l&XE)HVdQWyK6n&bHMy!USpj&YokyOM)#mvF1-f>)_DP
zu&38uiiQDa+Q(XLw(Vya%%y*|1WT2aJ+46<dIn8HV~pl(+x&%9bv3K}$Fd)S!lI`C
zXpzkWC0bn7&40YnWIIKZj#rBl)X-4TWWXCX8nK){<X)IrAG!}YW4blVf{0z8$>N}!
z%5T~PP^O?ROjL6%v{t9eOeXBYFYA=pYF-EAaTnX_k8lRIM|kRgVAzz@OL=Bnf{LU`
z(aLFXwTluYQoM|K`Mb*EcnWF6D^K6(X|l`E=smIvX1^x3$|iXR*;}cRf%95%j`Rmt
zTni5Fr~7{X5Ucb+MEXBI`9!GV9fb0!Mktv+_9<g}uk$QLoR6EA7jSs_&wBuaIFXgw
zz@MsHCB7B6tN3a!H;P!M;XRXudOk#T(!F8;<W5gM`TVm#$GJ{_Ej7FZV$+&B{in-%
zPGBRY%p4%tEQiy$(uc3NvZ89;_I1Y>*!urggaeT}P2688Z4>wkL5a?)6>GI8*tY<N
zKJO3?Ftr+Eb|HgL9NpZe)&e6U?g3NNc84B4nbZB=FCBO-K;6bdlXt}YdTzQOplaHx
zcOq`y9+Y0xy7ksV;R^meOC`Ir{GR@|^G=GkG|Z<~yARXlX{Of|bO@nVW?QfB>K6N&
z(qi_-jkyuG>hHHT*f=eC6EKCu7yY{d=hY`}1==i)&jREpZ%Cw#0vnyx8M{_T-Q<SZ
zU(}<$;<sE>-^(324IPuvK^$)hC<J3sM0a4kx|Mglx%&Pv#802~eB0=|63#5nr*0z{
z)(}xUl1gEoS;b%4N$fJ4FP($~4gFLeavxw~hXhHwv#anp8=q9923!@bAU(`hlAQ@E
z4L?&Ub*y;YI=@3-w`ZkUeS(X374KNFbweif0EkjGUS)%Lypnbx1?F>f^y)y&BLKt2
zB+p0MYWIDALxayn*I`Wd>nXvZp+6~&LdIO_<tf8D>#QG)s<^vv@jt>}hny!!%mM`s
z?4>-235+e9B#fGOA}3_HI1&anNZAs(`<#C8EuYdWY~y~zks;)HF~qQC`&TyofjtDi
z@H<Zel0=?TI6euprLVzkyt?IUJvx`jikO|fv?@j<x`f9U?7tEYT{K&ggxD-+y|W3L
z%51me3N892p8vHp9Uj2hLh`2#`6a>B#Pc7NsOY)}v0eh1A-wO(1qX$~N~QjGVicC(
z(w$9`BiUiN+lmx66n(K}JDUxBxJ@WgeEHD0zeKkhW3GdO1uQYM?_T?DjOT)~TNzL<
z)4fo@+70o}A|WY<VqY<`>D)bB%<ar()W0)igV&qs!-Y)+=V0YM@UC-ENe+~02M>=A
zIth0>S}oQA@kk8F?!dwjFsR!PZnd75^X|w=hwC#S3&hcUm1%qjI?)9{RVKhcr4Lh7
z+lo;0o8lRlJVQsRV6$mX{$Jy3gk$XG;O6cg*VqADKN(&K@Qhm6j`OUKD~qp#DBNJf
z`!~~*=~}(D>`Qb@1vYjr5i@+2N$|<#0vnShze&nz^t*<%d2|n`Cq9iwJyO=4U5J*|
z;d8{h6%D9|nhM<2zQND_G`vQwiv=@0V<#m$BWaK>WveY_n&B0{`Qq@*C)NwXE~S{j
zV5|MVOuEbv>^&?E)@a(}XmzZ(m9^cT@JlyfH=sk6Pg+*|keqWIvHzcKI6-7vB}p^%
zH<S7Nr(}v8A9t&Y+reHcY+h|;O;08Gn(afv<z$8u_}Gqilq|90{xp_AFL(6S_Q~nk
z=x#(%as1H84=E-(@t}6JTM5#e9TPtDD(?Te-Muj|JZO7WRp?4|#QXDD6m)be*uHiF
zr}7@Jri2NPu6In4=D#It;v%~Kz#gNielpx1dO4D`eU#c$c`mYVgB;+JOyoW>2|g#D
zijPRuUKb4xGT&}xbz{8sIX1jA#j~F?(a;&e)ct~uy#L|jn3+JWNcA8{r!?6~QH}7D
z*5jNbunTR*>(AOr$`ArwQ>`CZiA`=Ni)t>tOc!h;{Yh)X(&jbAydBME&GuL*CdxI6
z+wGdqo6FT~%5q6JN=P#J!sW6E4)DE_eHnH(6_H7?o0hI0HW&_qu0ZSV=8qwmF6lPP
z^&4tm6I7a&5B0cw%JQ~hq~Z!TgTU+vg8YC&Wrw=#OL(q_kYhvF*}tA`uEGB9O7+s@
z9MVYFbLk>lI{&~wQ)DVgRj3JWb?v@S)kuVsJNMI+Kudfq?drt^5;>#B-U%u`3ry*n
zF}cM<Okv6F(3h}NyZaBYYpya3gO?)Vwho^*9#<rIJ`cjPB7ID;JqBThBPjTjCUhOy
zdIZM1G<=z8H&{4pacQeJ`*+Rt`BiqD>ZdI_0K|+`JQusFD}%k658VneHBen|Z@R~~
z=6>ehSZ#u+J4D%8*@RN`7xwABb%SBa+YX8_AAWHCD+PwSkHXz-GsoB$6QRDQFIcxG
z3)0nXTvEmU3X|d-`rd1c<kV!-(;H-^n7DGS&Tuqmi<3p4y&W@n1|?9B`F2DiKe^y9
zfv>GAmzst592Oim*jxRYb~bg!03U+Hf%5^W4uS7;swe$t!IdQ7y-gLKglQGu=kug-
zZlf`rQlfasm$yVyu~u8GQb}Y4S${uD7N{2VRR|k`0%+xJqk8SJelKQoE;$0EoT=%t
zQJToVW8r@G0jf!Ud`2S_W#m!i>xz99PcsH|Eh=AYyb#eXtmyptH8W2IR8IFgke%Es
z=8|T#gM12hXoXpPsW#i@m!hh+Hn7=DPi@-GP(4_9Kt2?`iXI&j%el;MGQf}iy3{PQ
zOYY-hue<ArjU=T$haf>zt3l^3ci!g1J$+61vf4T}hHomp@Lr=$6cpSL(=8U;nILy_
zBq&yYvZu4<(#8xXAR$@6z`)rgX5D3G2|1^mc6?{hW;cmJgg$Yck>~vC4<;R78g3*Z
zY||EjRA)2qa-J3EK&F<*oNMcCs{QZwq^&57_*Tpu(TIN`&8LHm0z@qF6F+2gr;{l2
z$je=AtCMRVu}xqX&IJ3b&gQV08g?Y6R?ZS<%5f;-_gnPNV_w88wIS1?Jw;ec$p`Vy
zv|vcGuf7TGbNioT{sc-glKo_+WH{)cycB@JK>9?Vef!+2mAp*uHrSOgBouU9wk%qL
z;~<M6vdj2(6{hmb+wW%I|DHDXTCpq*k&Um3LyAFLau|Bv3sXSyHuv|Z33$Y{F14Y6
z5G^?|lrjve_s=DsZ)SuV1XyeR5f~*<q^8|_$E|pF&angx2pa@`eta7I_w#Vu1+wTu
zmigUE#-?M#DvwJAqqHJr@~6(jA5s-niwHEL8Y+ib>iv@;`26iNj6ENf#dk*HJQET>
zu;cVbE74*n$_o|T%ENu6hxG^ok*l9t<m4Tm8=L+}c`kxqk;Ot#`nU0=xZrwMqTzpd
zZ$ENm<r2oE%Kf$b^sLIGt-GMGwz2_-2JUZPV1mdE*?1w}K}_i1^vPbljxQD&Ijz8A
z;|XafR(L8O@^t=`Z*gw0^lx6QptqRvR+JLNg%NbUI88;enW|p$jw&QLyJVj%b8|lv
z#Y#MFP(f;12r#z)su^l*|9)!szYFy6n^zZ1-f>-_723woF6%lop$4O9^qwCD1psG2
z$MCa;NV@B-p`)K)+kS()!;s0xrSB<);M!H6LgOsVIH3+%ZS$r%F?@*q7dd9d;~U<~
z4EZS#@7m_r(_nGXO3}$#I3Hh=I<X=)zw~$F4;Git6prKO72oNfE~J|D*N0ivH50JX
z&5P<P<p7ASQW^3YyU?->6r0knth|!FGst#}|5wX%g&IxbQUqNOR!+SJYNGONVoTWV
zgm;pq>zDud@DFDl{E}NsnvPw${wq(_umusimJ+w80r3$BcXyBBy%J=1`FutmSs%`b
z+q~4&ENBZ+3%^~6&r}x@w1NL1-L0YC&wdwiRmb(AX=g#8K{h#uu=`_cnJT~}m+(3s
zfkKDm=Do{Xk3$Ms;vDsks!$eRclR31_?N#}z%CDPNco)l*PVhK^@=t1_4OS#hf;B6
z$&nMdrW0`Q?k^WN)9+&a$G%$skSc3bq@9IF3|BjP%-1&Az3D;{L|25IhwV<lp0$(=
zLEzPJGdTAeXBD)>uxp^#yJnK(AiX;AKm*>~TiI~nt#^Hq5T07Xjn|3uy{nE=YG_(f
z&?T+uBllTk8m^~M+meRr=AZD-Cd!0;ZIj|8vcbPq$@eK`SoJ?SJe<9_?~PtQU}vO2
z<>aK!96Qf+9}3Ri^h)*WfqgZ;v@$t;UmlT4_;(7%Y2<9WQ{Qs^G5>s{?wg&f^b^|g
z?u^Fb{bN3~eT#e1h4}2gFUVOUB?T%hxW2{&E=j*#RgNi>D1LUT6ghn>OcB5JzkfYU
zBfgyVOnhZFwc(@NB^mBFGpla1Dq61BvB&J)UxViqyw&+Ec_Dx<t202t%+r<m^uPJj
zlnNH@$sdJ1!luBIJGi`zJQ)^EU;)$pc0|w87MlUX<+=%da)=8?#<qNFk_dTK77+vZ
zK+zN(lB1}9URzJK+~OHLLE3%$W~ynvMUS+)+u&<j1OXyM&z!6DwBCzJi3lcZy{2A?
zu4Vsjy|DIKW>AFQ@bL`?rOGV+Mdub-WASDkei&D+P3wno(!^jlJ=di<uaasB$hhVC
z)!!xvg>>!1+mXng`<maO`z-SL5rdlQj{8r~v@jQ-OKP*K3FfQUbzq;`_O7edBANSJ
zW0olT{tXmmsMN;)))NYHEw>G8Rs-60g$mw;Q95Fuvih9q!(<c8^;`T>0qE~E1>~<m
z9wz<q`1n|34&tjXqIz&UfFjG;&C`Nh!Db561;-z-yH)%i3tkKqs}#Bu5qrt<xhC(#
zjt(hs(g%0ZBF#eFL$mxEvs8RbylsW;SB^iU`D&WjY9bVTlpFrd=Gg4j<AEr{TpB#z
zx1qaym&%?#x#7aB#Y%xxoCD90pe1Wxf+DEn;tT5moE>MhM8U`Czs;Ikc0c#KfofSF
z?~e;I%}orZxyP<w|1F6u2i2?S#B`R+w(&w=?BcA73IxAu+aZc;%b&)}vqN$XXU{p-
zP6_F2K-z13c^a8E71vM9(c=Bl-}0F#S?q=EG#k_ejux;g`LTYBhYoXW71Oo%*AtDJ
z{hip4-Um>^r2X{%VYQ51+&9|&Sum>=6jePTD-CA>y?5EC)WN|8NUI;>nG^{3SUMBN
z8S{o&Hn(u!x+J{)D_Pc`rrJ2(&}EtK)0buF+Z$#%yp7Aqocp0^W3o$OBMb9~l1F@6
z#wWU(!oWTa0j>W?(N3@9v~o+Wy<-GVXToKbsZR?_TfpJa{c6?!c(@ioISf33ABrqW
z)`PMsQ#R@1-J7RrvE;=BUOL^Mt1t>e#?Zs0Y-naHSayyC{Gf^J2+i-wtIsR-LXlEx
zQjd}fbywXm%UCl+ql9~WPJ4H~>!+z&G{Xw=6r1n=yy@zZ?BjjHg7=bbX)r6VD7Q#N
zr@lLH;TGSitApIGq^YX_D}0JxDE~HFk7`ydf{(TuhUt1AsZ(+F3##HiDR+CDz)qVP
zXT6v9b@@v_=42^Bb2<5+CSwJ%ND#l{{PmF>mJv{|jYeMh*4*}eSB+wK7c}fkXSHIr
z=r>0S#J>YC89j4^W{X{H<^E;&4qPDnBS`JE$oYrL>p!_+$*J|d<pXik7mwtD;dF6(
z$-WO6TgM)hdx16y>Q5%Rk2ftUbz%*aTIV0b)F75q>%58g4?G|g0D>2z6)<=m$WNP(
zGoY6mEsORkj3x;hE1RwnvVg2VJYQX14e_5O)8`GTnWymQaP^z#-M?rmsY2GiYuME>
zfs?hAhSi0ozqWpuwqo$Jqaf;W4ZSru_@VeTyNVb2q{w5i^PZ?>6_7lUg&~Nh*F3YY
zaj5qd;tb!aDB!uYm0Xy8<<TaI=vm^Vm;W8t;}Lan&X=oBNZhi46vQ&+El<k}yG%X(
zEK+!i*0U%|H$U^6pECLamvQtbeV0H)<$1o+uZ*b^I)?a4ym-56>6;Ie5&=1{_))!T
zE3Wly@(Xn<?TRrek*r0TPsT-xtUhl^X%w2cexS-ho3=7UL<hI_F4rw{P01_>@h)!K
zgpio3-UK1{J4Gtd$`O@Zguj6V)BT1Yn#RPf+e*ab_}pz<GCwOdNIB{{{P!wOuGVpv
zidxjGy{%YS<o-raJtj~MFw@mZ_KrvjQzlv7TPk~tg7BM$#!;cAvd7&T<};e+9-2FT
z|JK`Q{sG&U`TAYwO$d<oG$}`YX0EZ+gldk{-XK0>TH*a}z18GH_#ZaXpF%j_3nlA0
z4J(_T1e#Y((FSfUveZ9`3ym;BR5t!1JHLVkHla_$4%Bq{he<i7&{rUm$}2Ct4sB%7
zsVGCO<I5IRw(Lj)B{(P1JO1;9ucLqa`D4nP1tUHv`%_PeBEb_i(z#zEc$-8+-&A;$
z3M>|K>hN+znL`Ln1zV`2YOziZ`&;<oOY4J1z7V~olHdh<jA>ZdQwTCKfzOp|V71Mw
zgkjY1rwa7#`tO3wa_}2Cji~|@_U&^GtPSDl`6mrFtNS+%_bE_Q)I<&FTLUOU$kT4F
z#tDk374<7!vy8uTjbWh2V`dBR0Mby}A0!UL=Cw<|!^;mbUqO|@$)nT>+%y1aiwN%k
zR+@2mSXUr05^{d&SdIo*sWcJK_DAsl)e)WlRfzvz;LH6`T7BQ-dCF*&RH4nr*Zs%%
zxY%5T8axClm#<kak~6JBf8`a*Q+#xZKp@JDxg4w(Nl}{lqW4TvvO66~8q&@gr4JIz
zBca+2vjmm)5H@4|^ecS2fCm<%IlD!TkPN=-aYoD)PmXn#$&3H7pU{<8c)Bry&O2L;
zkfbz54IvI0PTyuHcDRC0UDy0JxzJVFri4|Aa>lgRU-P~@!sC(u<Qbwy_oLxeo0sDE
ztK=oD{;p$%3j1agfj*$#e;C8r=LtpC^KyZza3HHUp{!HN$hEt^YG5ZYK;0oMDj8>8
zR*@Jc4P=M=8uZlK|9q)JmzZ0~B!55Msx;MoB>&r`wsA%*9;;eWUQIAeRZO6Lr;N+`
z6}}c4qjDE>bM<cm{t*i&Mi29?C^ww}W$hw%va=puLb8BN_qL0^=oki+A`DT=lch@K
z-ZYVonO^RTQpl?r8fWN{SW^=Uo-E2johzv+f!{G)+?ex4c|%M8*-H3$V$WX0w?;3}
z=)KV>;yrY*{PW*xu77KXvbgc(QFPyDr>tHXVH`KNzqx8h<a%*&T~A_Bg-a`3w0CF3
zYpKU_7+#no=ksS5kha#H(!bL{N#M=UyVSxt`uZDdSBX#`f9IvG8a56pEg_fUleL}l
zvt9m?`r|QKhif7>NwJF*CzxW&OVo2-FdPJTqjQt*C{XO_{zcyy-}<|u0L7v#maHYT
zas4~DOvxPeU+*?iAF3X$Z16d23J0e(D!oUp!|aN$v^nJI&;8x=6@?>>!HrU#>JdAx
zA0wmF@6Z1%Z0)*X$Ua*v`pv0XLOooDzJj8>wQ47`3YFJ0vecQf-80wWV;Q@e^@zpa
z$k`{Oh4*|zKyug=9;>#`wmuO)k*4HEC9C3>_WZ}E2P;959^kY!6){<Y;eK{S4!jc1
zQRjxQlDQo(LfD{p%@>UP02PD(a^-3Sec-KIdy4PbsBCzTsX?Ph+ww9U;LALEXnM^g
zW9a9aeJM<F#P7G2!C(TMVUB=l3-AO$(#pzu1uz(EYhQuMP3-<mP3+-nmC-HrG7adf
zfNU${7Ai8#A2X+C^9VYB*Wl6B-MgGKJez}^QPq=$OJ9FuujiS{-LG-k<r=o>K9}GT
zVauvD$W_NLMyY?EB4hJBU&Pd{n5whZHzNF(2!{a%BqVLRZJVXKN>s*?VpYsKS-WxD
zqZNV9?1op075AlRv(9l}sHPhudmM9fQTuG0>*=_PY=hPy-<Hj8+UIfAiR3)}+@yDR
zz=Y<pji%S6?s)>51!R?HW)NL_R7MqaGfG%LdO{0o4m^Uej?b-$sD*ld$5|m!p@w{)
zb@&jMac-f@O^r2$P=T1ex~#trshWfvjZ7vo=h7>XwXfxSIdihC1@L|?qb<o+8Z~>D
zpA?LfyjXvzUMz-Qg$)-}v|^RXqB8J`a7`$1G85|BLd87&q>+#v8E;vL^^9JpyA39_
zk(=A-<=B37tY)|1AH}x)+wJ1RI8IFX?d3vN3wI}dlDd`ZE~oYw{rYxg?iq&f!|uPE
z77HF~$v3r%F=M;uGYAC6A#)*?NYP1S<d2Oql9Tq8kw#~xfDgN(Ke&)bu7zwv^2%1m
zNfN(kk`f{>2bAkfWwFpQ_hG_&JtmYhy(97!{EH=4Y?J~h+)Hg_!;W_fq}0SUTFOz%
z<Q;jrw|r84+wm*v-c_Pop(djpe`PAmUg%<4cBL@oEZRXEKZ-0Ue3q5K9Ci5py2ddx
z;)coMpkDpOVs0m!e9Ph{0I&eOA1L+qyu0E$z4kHf@S&n0ut(h&+cF`nngadukEY0j
z1G`u!+qb{;RhK)HzvSVf(Kafr&u9!6&Vn|JfxbUenrg5nXH$EbeE*`*@Wto$6<m45
z-Md>`dB*mrw%DsLcX*6jXr~7u4Q>&@XZd(l11$^0GGFz-1>~h9%f~GJu}9-_JdUy0
zca}>KBJuA7aQ)e2dQ4zRyT8B3<)poE8MC;)oNsXFn2DA5uWo8DWr3XfCQE%3iL~!n
zy5D3+Z0sjj(OI0=<B38wdC*<B5H<P4(|!(K2r-06oclkje-9Xy<=ncfb1@IKJ@=QT
z$D<v|bF+%f279Ney@Iz543j=G1&T)XDYymiX<C0h-ITFfw|e)p1RCDy*EfkA)@1u^
zA=apsRdZ9pz~?^BtYX+mz<8QwDZKkd$Rvp&_H!vD0f}I<PePb3(iPP;F9}{68~J3(
zSYLEU&{|pFH}n#+(Byv;#`<SF^D2V^y?sThS)VKX0&5?qn;vi_bFz*Y3MXSGow`^I
z%XQHyq8k79YPC7koYBpA7pQst_>&$-2Fp&j(ALV$J>h8`@($4gfqjh^eD;TL?hVkJ
z=nCHYTo#rJ`toibb-#(peugcZ!1!})#X?Y{JTpmov<@6yj_yXe%N+g<H4t@$cD>By
z---RBqw?yPgUK6Bx{~p(B0Q=LqS<UBjoz0Wa$|s3yO`;Oih&O;tCn14f5Y>g(N|Pq
zWej%Y+7mcuhSoUZVqZr7a7vQHx0$F~?9-#SZtywFH8Qj6EWAAcnBs06h`#C#qVm%A
zEmIotD?W*)*%8A08DybfU5<r7z*Y4;j630`r@oWK5b1!?&pay+_qU2H1z=wq&gFA|
z4hN?wMhl4W*RkX{K8i_;X4<dsp2rt)x|?8-aJ#f4RiAdtTLm>-2vqS`m(+I;hXO||
zy9%RQAQ#G09$H%vma%GHI}{nFU<%fTzTtDh-<vUUc(eaQC+dp(_}lry>3X7}6WB-z
z=n_dwB>jnCve8Kx*gXB?r<^9zCfL*DeMz4C*!YmQ9vmM#!bm_Z-3nd-{!!>gSV-Y7
z=Cr$EoAO%ehG;7*HU=01)uR8z-M59}s%BP82rAQ7L`MBIah6stdRpsQVcK`$7pE)3
z3-7yoVaKyuUOmZTtnBudrnw?n<r_gE{8!83Ug_S4)SMq#HcXYvYg-3$Q~b_xbC^2+
zv`?mJ@6ToA?a!za9-wku6>MPpyw9=@DdZlpCTgxAEB!D!kd}VpbVk*2{Iv)!MYE@R
zuGWaPllJ~~I9&Ts(oU~bQq^5!bFm1zyhEkxRf(R;PBu};n@cEEW58oV#$5D*H{y*F
zlGo|i*}T6WQ$Mw%8f`YUvqm^mYYFM=l>bb1bo3kY;j<JHHs@=}_zQY}YbW`yF$x_B
z^5e`k);{C(>8i$`F5HWH^(x9VQDeSNqs%2#FK`@EtwW+OF0^LC4r8RFP#j@L#wxhR
zYWUkL<C|c)*_fo{XfF3;t&jmM6)W_)42LqD*etL?2)F2=Og09+2P`Bcy;3Sn#FESP
zzvUhDi}cWA7FL*O=(3d7Ip2z{rlMrHU0<RtXbSl2xDsPKqa~JU=k{fqD)odFgx&9w
zcqP+`R5u;cRO_SXLnJNq9NAG>s+E>qjQ*N_WPMj+3<{XMFd~l{*b}Gs>cyU7-b_LD
zD*b=IX+{rC&Y&kbfkvPBg1gWPND(f{ir+kcjFC4oF|f~onuP@q?l>>keC3*8qv%C2
z!vf?rCv_CX6uVpWXM0tsMVU?3cYXp7Z|}4VW68AVl}^xMeWO=Kd$xS5KotXB)gVm9
zm;4{x;0$=fjcL#XybmdI(UDmmp{T%F)Rx0W)xOtN6jm1aHc}P!ePoX*J3BjIRYjCH
zAys8E1K*^!4X8O)Y(SkrjtDO-DERce=NIdWhX?op=0F{LvC2o*wBzpG?8+suZ?iTs
z>K7FUJ*`#vcDIS-zwdd=_g1OZd2f66uUCsz|BK;b58Uc>xJC;!{0>oFJU9(gbX!1Q
zYN9IRPC&Sqoc3B?ULVCagkKUA_$bwpSNb2a_uBSZ3$3rLzbLQ>c4<1^VTO&&7HR^>
zTbmX^d;4-(GrBg*boOhSK+4Ow-+gZ^qtdui!ndPBaNd5nxutvmtw62+a>oJ*83{@5
zy@Yy%RgjYe-HwRsH;l@K*5t(M@GsKAw(6-*sRHG#akqAU=VM4BK?aegth(B{(I$e)
zoayk3Bq-*mjfzNEl?+oH-Mog*eEOW4v)QnV`d{}sSi{cO;!xowd@&rpa>lGx%5q4#
zl*8%i4)<Uds#P{oeVdwLE0;vCO@ZSaAty&5r9r|hVy^L@2f&d~V_=AqNO!d@Ag^{v
z?G-FF4wR#l<dUja3jX?51kV~E9Y|yaVnc=Yf7zH;935tua!GH=agJP94kk_H{v%zJ
zmRM&~Vm+yNS)-!e0WO}l?6!+xVT#ru_qLforcQ>uqYZq#gBQ0b_?UC)QSTHkw`~TW
zxuq6?aBHoNUK4yNADzR%bq7L0{EwuK02}?f++_Tlk8DKT6WfX>dn{3EOCJSn<(iUr
z!j|;@Rie-D3T@RNCe>%`aI(se%*0D_oVhzhTHRzm{Krs0Ar)yyvZA<Ww~$R{`akgi
z@DKpk*9<V%b#f0IPw_m{_zhx=si3o=T^DBMA*K{QAB0!9A|w*>K%<;@w}*&@0A(i3
z?w!O|&0tFMKCVU3v2A~CWWBzI$q0!!6M7QV;{8ufWLYLArbly@qvL1*BeEmGM>nbN
z$#8|edFFEgF|mzVm<?E*cDKEs{^jv1sz;)A1q=WneGUeUM>h82;$mLdgz{_R6>k!{
zIJKldAcWRANJzYA#F<fGKIc?t`>ue~k>KBuU=Ie3&q$d!B5=$NITX6P$wG3@;xKg%
z5xJ}`mqnz_Pf}ltB={rnc(f*2?M!{_wK5j94SGU;l<;-2>|87tQr*?!wlDFIleut3
z9ydm|gqh3hhw#4!WD>WFx4<?Hq4qmthuiOc+GdbOE`Xal?F%(sCKQoy|M5RfXI8eP
zUx+DKy3bdKa%{HCWWXeO+m1W_n9#E#P3+KWC5xc_F4@WWJB0D-1{AC99TUoUoBAwr
zQfJe9g^!|dJ}DbVZyalT+B9^8Kjs#E22oTRs$H$ZMvk|M!pQNB*u{y6%rB|C%z(Y;
zf!mCWGEf(PTT1zzL>8c7rSoxf;173XRAj9{lM$C;bg7fA)iHHnrjZ7Fpmil)%uI^I
zW@YQjx?tU~BXb-#{pYuklc5OY!-d7wJrUEl<jaj+g3SF-vedlf(xV@W`6S`I$@x?U
zo*xUgcf*}vOIHZ&6A=tcxd?QYkDm#Y1gpLPgouctWac(5AV~S)It0ZkmSb7RU)y&3
za9RE;mDCrq?=3cOW~n<pkXop_6pQEPrBGKskQs+7{;)@JPt1o}zcH)PG8cJ(M9ulw
z{|uvW{&QZWc(lRs<-ZN*x}IAY<5g&d9h{N&`(~LXm8fX7y_!d33EGSHSx=TMPMNf=
zEIjPp5y2=YR~!kBo!h?;s5>LVHh&J@;3eYz#Ga=$Z9#X_4GH=cy0(Q#nOpn=tS66>
zEm-u|Ij+idkmOi_G!NaBE=d@g`0dQFWK!ejG-$=yJd(V+{k~@fZd?hNZpHl<I;&&V
zHIFTbD&KPbv<V@zf?HA`MqIV6g6c08_b;_HCd(W_-vNNE@eibU_+o7lF7YQ1KHiIq
z2m|Qh3iIJykLodCDm+&40LE0S{wJajS249wIq7Sa^m0HA22I>Rw9qNm9CSVLof8L<
zh;^r#P7+&T=yxwSzGNW6gKf`KZ*2`?zyiL#JOnhiCUjd{)_p<W^WR$)Sf$u5O+znU
zJ$NTaSm^zMr7v2fRNpt^Ww<~HX=v8~*tct!3H?KPU8e>kq2s+tyhl;eJbIZ)b!lPe
zQc+kc)Np(f!@eKAnYgUM`7-))p{3*8Cd4fNH=C9=Mql<>H{rk_1!c*C>E${mQ@?$i
zriUJe={c;`Mq7B^_Q{p&gO1|UE~H)D4y>@HtY`v{s`HOghDR&A)Z5w0{u7qb*`MQ=
z8qo>glGoN}i4^owKlgjGWJB<`(1?yHt@O*}xGOV;beW7~^KWFgzM_tp&#fxl4DD+<
zZO$ax>`vghCg6J>n%Kjk!?sD}+2#c66xJKR28RTnw%AQ&K~L9q@T{!U=Y8GCH(}KG
zjqmC!s+9bXQW*({1i{L#IKbY4OQ1nxvC=(Kid6#S0;!Vac*M@*syj6lXeERdbvD?v
zxT^v)PqYn9ypT_IZRBoW|2imFWzc0X>!21=82KS#73@r+r<(xKqg?=a{loRk{bM_)
zmY|h=CsVpFvJf_%lL+e6q{P0?e6@6-U34Xkhowt#xvK7ObseR?pE_x4BkGPecId@r
zIlc?K2OEX?j=(p2!<9i2<ycS2pKC|u&wOW#-sGjU>pGxwVnt!^U3}woYdf91j&>Vp
zP`J-EQ>4rIZKt<(KMMQtcX6nARS|Sy-RcUQXvdE+tTmVfG(qctj{AsQcr;{@k<V-a
z@*f5U#^bpORMYT)X77C$1*jV?-=rg=FN~A}lsDppNPX*AmZ^>BD+^L2sezar?p@7R
zTbk<64j3lqg#SW2l%ib(zO6?j7x)fx5`~GP%qbsWbyrRY@$f<|x>`@D9@EE<q#lWv
zz1o|eLq1}bJmkFlR=Hn8)jVMu4efaBiT!`<SAKeGXaJNJ|IJ4c-Qd@8H|&WD;5`Qn
zy>*%sP2l`CEdI}jHNZaTq~yUrcFeWxPK*%?N>YTov%;eXsc>y-hO_P1_Z9Vo#8-=&
zctg4vB2URe46IOUYfI}>v`d7Lx<Ud<7RNXso_*^LwYU^c@ijC(0K!h==MvIzY4`R;
zgvxAIubc}_=^F#Vm{hL&WO(iU=rirK)DYU3^KUQqwgc<soEmpzwqz-ywO}_yN5&`>
zFb{YE`bXHfWF}9wK9{K2##u7*y>S%V?(g`v9MKfHq%*Y=O^MfPh^fgR1Mx9R&^gUr
zT!uAd6`{_9(>~s%Dr)(6;Zu|K53Zk|lS`1L?KWA886MxwJm*NG`Ba2eP|;38qNf4E
zKdYp&KBxO|5w8=f+|69Jla#($+#aN2*QZhC$P!<dpZ-&R8}T-&*KDBcEZkstDW36@
zRq+&aP%c|ESupBTH~CK_z$f$B$LJZ%l5;4}7TT#l6M!3+nJ1s%*4wC`p(;J4Aa{#d
zP~CmGpq4<qUSQ_sHe|wQsgXYItQW0?7VqqHH7aG-?2HjZ9gpGw!xAq`TrkvRfU>3W
zySUYoPc%Ijv8_iqdyX1-VI*gi`1avO6_A$we07&V$)D}-;*%8dDdt&tY9)2yZ%y@a
zHnmNgXWIGh4F<_d6i=v=?RSG>Dc6L06z%e?bb6(!#4gUCv)<*5e(|)&to`}P1EB`h
zx<2Z72CiSFtd;&#AE_LEP~;>a@CfHv0!Y?G4nsa9Z1%#+5|h4qzTa-oiC2M~1x{dA
zDa-*8Wh}?1Cg}UBhO)y@ZhXXzJw@K`lW3p`0~aDNf=_VTWm*iLrT}HQ`Feu_sKaBZ
z_wlvH$G0oZd$Xu$>ic_p4qzl%3yNqm&7m=&9o;=AKO$>+H1K$+$QyQBb%b-7>BuCQ
zQ&_EaemejF=4+p3W9&E3f56}90Z@Q^tq;|HcK#RD>ntjljLvv`It?DwT^KP}zqI;l
zNj_WyM~1OEg*H*Zl*Zy!Qi<Px?p+~?R}Ix4cXe$dD5sK*BkS?%Ahb$m{fO*eK{e)k
z2a9V*H`Gmj>2t*@>|J`qwqsR)c)l&PQHNX}lzF-KWu0QrUB`K-J=ObKu)m{zwmd7A
zt4WTg`TJp?4wdHu-((>88F#G8Kw{(70aw#8b6#*Ps`yOTW#tm_j(U|<Khk9Gpit3w
zm!hjWzSMn`G&Qmm=%+Q*czq;SUPPvCrN0-=tXe$rco?Xzr*gboVK6tyC~&BFQ%G#+
zf6Cy0U)i$Jd)hNVqL%Wke2+y7p7Tq&?E-)BTUN>5<d~kcl|)<HG626&SN<G+5fQTH
z7p#gN{3p@6pyDeV9haPnI|H>jk3hsH%#wonF5mA`nzfXLEJe6{{yBxrzH%`yV%WOq
z(PO`<{Z}dRMwKLCrYmK}MSi_NcOQGbfS%(G+`h46v*onG<0Dc;-`8d}<Fn)<qR~tO
z)m$B-1^#r1hL5i^{3S`Z-Tseq8HwcK=El!8U+N=#iuD#KP>=<Y^dahJ-p-INBXN?q
z$+<f4xlPP!w<(Oxm&k5I7=QF9I)}XDDwm@(*Rm;ot`1%Vd!eb2-c;m6xeea$N{O{l
zM%>3N6`D)Ydh<)X8!+%)bOKye(Dx++mSv6PIZ98yfz=~G6Y%|i9cx0rxf<N9$t1c$
z@y+bI{&W9CdQ5O__C(Qu5L$1=5|mwTFURFegNIb;;(UpX{&30RYzDg@(@35Hn60PE
zeof7GSDpjL({ir({cCXBH5W^_Kg(@oVu$Rq9As6nAq#*v3Q%;xfe{SiFkRluo^lMd
zr`@OT-#ObNNDF_gb8qSs{?z>`Wc&}cF?3c3c%tf&VUKzE#}jMo6mC5(rJq_t@x0~!
z;lj83Snt`7(;68i3PaGoT7HZ~#13JU@om8Knokc@!pohukpJW8Dx;$8wlFQ-4N@YF
z(%mTC-QAr-cOxYoQXkz8-Q6kOAvMw^APjL|?mvF8Sge_I&ffdkPjpPX^<q;IbM_dX
zC3`hy4YbC)ZjX`Lak|yPcTd4Svi@Ibxqgp%r0A=&r(%Pi#1O%)Lh|h%$yYL&Eb3I$
zzjybEpf2#y{RuF!n;Iwd-jrM|uu|Vz26_I6x>$gYDP4#!#XOgNW!Gurs#uC#r<OWj
ztW(*5NZ<344*0B?bcRj)vi*Ul8?0xRbC{Wx$I$x-%g2CNi79y64^4wBI#yH<K^rfQ
z|B+{>$Q+EB*Q~Ti!96bdpNwXr9ir6jSQ#2c!1Iqiq8ZojL3`zMd5SLoNx<a&&0{}k
z#-AT8F2Z3(vomS!bfcZ&^KbLewW=7^L!mB@$=l~tBIy~n_Q5!{%1Vo-A7SpqX)0-X
z;l)Vi^0>L-FIp(J+kjKj*8|yk1Q(6)%a3U_DkV}`CbTv4;gzU&td_dJnP%z@)Rrm_
z<}@{GEz&-}h0o6r?stzsM4v=~&MVjQTy^}vu8~KkaEsa4FC_S8#)a7;*i>{5$s@}?
zQTC<9FI?dz>GrMRFFfMm2<1!mo>fE;7+nKyn?CN}Ur85R%`gtLU6Umx{{d?@hkQ50
z>RNFdlR(K<9Li%APWyQ1cht6YW=LpMl@_pJkE5Qos+s<>ID_oSN<fTNa!+g2*)NRw
z)mAwEzE#Wn;0l#IeLO7~;~kv#jL_%aKD_?Y6_b%;;bzRr%|^|dzP$3Np+2T3hfqP|
zL{>gr^(r%e<oP*bW&cDB$#DRNC2KPBS4-qg=28SO?ma+(0K{+y-c}iNlYz*8ac%ut
zYxZ&;U+E+LHwAI;e!g2DR@Wk?)rrCca)|)@vN7Ue76pT{%vqb6sefJu;X^F3aOUX1
z<#FzLP6v$3kv4$5qZ`P~Lsics^Z)<sD)H{4$MdG<8d<?#=ep&+%geWgVB=Dp)Q9RO
zKuBNO^<U(-S3(ymT^0nG@tcT3g0gs8S(*99hZY!#HYtfB<P|(W8+%s3z`tkU4beub
z=>mK4F9B?y;?ef#v!Q12VjcX9n_|vt{K<kRcJ%8eB52LsZ{7L|y1uuD2)8#7W$@u$
zK4@}qC8x2*DSrEX8M!^Yz3x1Fm><Chq8@Yo?Iu_oKXQu%IjOW-k-(H|&@>)YkU_Et
zRU_2dE`)OBqoe=@N4wAH?K`ireBm<xN8CCp_Q-xj>J0^{kK<)c*jI9ku(7t?C^^ZU
zBg=({{gEZPBS3Ptwl(nBql~>!?EQwBzYr%y@9%Wqk+4F2Pg4}3RK;szXkChJ9srBW
zbgW}ZvsC`Vu1C&sye+jyppo{09iEVgx%=T5M|#HC#4zIaW5Z%x{`2m|KdhiDN<Km1
zl6nHi-n;fW7BPb}S#q)5rQo~xvQ-Pd(hcFA{nJ)mWG1xdUl$*}>#oL==mIV^iM7rD
z+#*@%CjK|d82h7=i6NOF?bDUWikYsynug`iqMHOG=U%&}VivhnEZ7txpbuQ6VH4KT
zYT}g9)I6owFn4}M($tKqg{j8k$(s71Vu~NWX?r;OF$6a%9pytYPv{n?+^8cwg}xWz
zb0=<ffW*Ld|7$brHEHQ5q#sj)t&K@(m0i+gE^0GH(Ip9bt+k06SH{j;b1&S*r&?)@
zLQ;}-`5|69`lbOU=ReRAcFZW(&_iO0`?Br62}kWcRz#Lo9fWCluAdA<*`skuucP_6
ztK(6ZFNNE2k%F-U&}hu;0ZC3*@uzRF3CP>P>I@L8y|GT-7S%vg+`E>C)-G(yYxkPI
z1E3MWiaJGEOQ#s5SMNyn+&sl`e&bjzrd^7BC6PO2<$p89`zF7yfVj2oU%Z@O{JRl-
zbq5G-K_8$awV*+T@^XQZJ@&7vm&c1$NR2V>l0cnV%PBn?_W%B)XSbI+uLhj*70;;B
z${A4A;oTv@xwjg8BC|%h2*~C26W2`#!tH^0STWsn?a(ydGprR0^A4B{|7JxDGUIvf
zR|MjrU^*l<!;52IBO>Cp1qz4H?4zg2=7!I>+w=v7y-Pefowe2u_Qw9o8<8uzl8<&|
z_KYR0=4&u5&j~H8049>$&LNc}U?HpMD_iExUr1U%mYFNq()!W|p8#RLK7tFQ6E{Vv
z&RmJ9I<%gUI{G>_&opgz>{~3SUW2NXCrrP(JpjbXlcLWS?WjUhOHOaW&35%7gBO@%
zzipcxI-PaWf0mU``4@P`C=bgq%ISaqNK!Rd-@<zc7>V@UTn{mKdg*9`CyAa`d!XC@
zs(K#mW9y@Nih0Qfr&BN~zG`wN<ki>fWS9q_jJtWC1zYk%jvA~tS_3gI$o1$gvf!P;
z&8Wjy0)6S0*!|sKDiCVa;y|h0G<+P<?vgGFD260k_J%|L-N+(X(;WH;VK$Z$>*>Ip
z(@pgmHC1+Xb=8`Ud{31)MMVpCfy6_F7mW|f>ae2LjM9VwGYl#uI;7R_anAc~3nQ;z
z09q`}k|f)KR6bChTyRS$N43UA`|2{*8}+~R!D?!92nr7H%Uix*I%dnWH){kp%&S4#
zgLdaDJi;c;HD3n?)=n}e`Ww%v30q1W1rH5%M$W)1HYd)ot3z32W>LA#o7uq~Z-GGP
z9Tz;Xpmj1GB9*mL+)b@rYrundb`}R%0MFWdsar1EzF2N<2&p+i%_hIMhIU<TJ0>v%
z;EE3D$){Tao@GEcGyLZF`~9u)zOT_@emR|aiOYZ5r;itaI<Bp!yE-1Kh#CI)a1UU1
z0B`{)F#tGa_#+ifuR`V*)wXeB+0HhQuo(p3b{inS9n<r8Ue^NRPT*&l-ShKL1HXSB
zt@Aldh<@?GQ(1m-(6<%+VCC5|2EOxhEakRZ*xEO_5qO}pFB0DN>jI$GcfdOj+@gVA
z&zg8eUC?T0=|l|8%chew^ak)rcDQRNzNxkDvGtkY1C<tDN+!$`{ukN1?;WL#J9&-?
z1maCxdyF5|_4G~Mkm3_JqB4CN``)%yOq*Zg%rjc6h8GbZ%Ok6XPcX5XoHC;P=i(v-
zBMLnwT+N=GP&+NWz$0*v-(KY-mHkv;fn9UqR^uN&afW`R%2XOol&Y`Og1i@(89+P^
z0BeT0hoIzqwt5=Ne-?c>nhBEAP5bRi`XgRasI!#I1e3b)n5Q+H0F8Yz)-0kf#4x)Q
zh{WUZ(RBej><&RejWo%Ma3cXRM!d%vzs_YqhVe@d!8dk^U$4xf0x;P3B=jGhwQn4@
z&zwrvC>ipZ0oE<enT2UInv7NTJG=hlW=&tn!q`cQ#*DlZWy!P@(CkZgWxK6rmR^ls
z&ro2#ARtv^VPqUCo_H+3ya$jg6uX<v6K}7Rh)#ZEV=oRT9@eiOQ>9aDpsNcq&nwMu
zK~tv@L4p^u4RL^PPvw79`7`BOgw}lmiFG8B5TT{EOLNvzEid07{#5b5&6jzRx|^8Y
z&AF$0>OT#F(o9Y$incoknQ|SgLy-MyxB-5`8~g<@5r9Mh={!gVPkpQ6dZ(~C%)I88
z16qd_;1B_~$ySK#KzH4tb<M#*`~9v+h6|-@u7xD-;)&oz=Ya9kknsY?G`ARJ`(-_7
zVkYpe0g0hz`5-6lj&Ajw=;)QtQahQp^JZEQIPX9NK}SaiMs#xAdpH$yv1Jh%n;O~q
zziM$@tM}70Gh|78IaApML$=v!_=tc`GQQ|*GA)2Vmj-TI*ETP&@mMuY9XDYqpII9+
zW^5IRXuk&}@*d9cEG-jRQ*+JFf46uKaWx4VrNAvse-+6gwagQPJSxjh6Fe$J6S1w0
z@2j`)Tf3$@0c|kEaKCHcR!j*#u@SCLg_sS+m&?i<`rc{MZG1KMs6|;-bLv_(|1yjM
zr{syWyoUd~O~aWAD?`K`g|*=s`-y6;8OXnXwP@t@T;UJa<G(=4GA_;U>Y5*Amn4O7
zzZlZzWnh!w^5Ug51*9kc^QsY74Vl$v(c0ka9MbN2i7ZpjQzQ5|El4;)@_PPy({sfu
z3eCY*8}!4c#{1F{j&bGoO>-&%snC>7YR@6YE1hdhCMA^xe5%AX%rMR|(uQiz6=u=<
z?00c3q{z?DU&s(@eLDT6=-8mX@4ZmDbwgtmJ0ta~Z^ROcEh^KIY8rtLxHA4&Sd$>G
z7;=MXd8AWIXrM5|7#sRxBG#?Lmivx6TFXLL?g+obFRCB~X@R8SJSefxM;{&+whIjT
zZht7fDR^tJ`UkA_YN%Dwq+ICOyI|x^t&hw-s+&0Wn0y;j%qnA!WM^KTW9{dEhVx`4
zG2^(PzeXpo!NcS_iF_Y%;ronk4%nga<u6z2MBE@HhEv375?}ePlZ97@{3^%<qrL0)
zOpkV?uyLgJvE}Tk=UqB`?k-`@v|Sb~HikZn0Pz?QpjpM(v8!_J;2GEX(-UKo(Vc$f
zeS55lx9tRkd<JvP+zcC4KTod=*LLv_Ef3#FiRq#TAKvi4%~cbhd-eebbo{R&_j!Pn
zz327*HLvG+y=T*8uw87k8AEaFL$8d+`g9ux1QE=hS}-t7420p_Uu;8wNYEcYesE<`
z0&)lyrgF<Y@}B^CJy?FLBGt~}Xjg^zKex0~{E36)Z{N_y>x;K9cwPU|X1JtV@<9O2
z(K;W?U=X}>b#n$d7NV$a&|K@cj;^u1y?%If0=PZ;81Ai!C;&Ek>mWzJykV*?LkJK~
z5MCRzoM12?C=K1zPOF8NX4z}U^f0{eA&ZI*VlydgX7Q-8XDIDkH|a^!PgLfZ+ccEA
zKX-8vG56sq63Rj3uD)rSGN_@s&MRQ7d>3moXY;k8XXKXJ!ATa^ZJv~i1SK_LXcp}g
z$ZK?15~#ZK7g4Rg)7%Ofv?(fbzZ;!+=su6eMx!Eia@ep^Bk`QzK7N^(e@t!Ln{81p
zuXI8ES-ayLbjSO&UnWXj=LX&wP+xRh+xXLj0wV#(Lm#txIVm@fQ1kja_zcOQ9mzUG
zdzsu2_V{9#tE<VNA&#4jP|k0WvU!?!zHaU7yTaM!?gi;-r@T7}%oMhsIo~*U9P7@=
zrV<PORj;C{%hZI-y1D5fnwb+7VQqnu<i6yBi$n1Hx6^ggNS?OOa5F1ufo3#4+VD5-
zSZo?=v(b-tcF31uLM5>yOA%=MXg;9xE%Vn)7)JO|su>1U95HHRwyQsGKE)6+NX)(k
zOE^^mO)5u@@zx9{@n2OYD?T=fpYvzGPCSXx-xU;!lK*ldE1kX(atoxOhNl(zCU1}Y
zIV#rNji0h+c=$gi6gEe}$7<N0h_-}(>2FkNF)^jA`S^`F${4j<OVNISdCUEti-Q|-
zI><KG;)=3(QVQI&shVzM451wMMWwfyW8E_`N(w8ptLXH3K52By^%tw&`oob*Xwhl>
zh6M5-uZL)!otF0@iY&S~wPK3IbdGDKnO0@VoqGNm<d+*f%6QZBJUH`pDa|rc40`%q
z4b@)-L=vfJXP6-&SvQ!7c@WhK4IO>`qWPoZii(X#OXc%VP)PGlXY%G?O~@_KfN*kh
zrcPMeHR+M2-V1>ovL(MZf6`U44uUx8v>7A-jdZS{m)d$8U&5PngZD+(J)K52B!qWa
za<Sevh2xV9bfwaOb+mzB>rB|;`ONu~L93WZK7g|T{{I2Rsf>Of!ew&h0at5#bL(|$
z$K&GE{Tm-3?ozG-|IaGbV0v!Tu{Ignj4$9@%F0@D_S0m>5X(uho{D+mvVCtqQa?-U
z)U(;RAug49-rzD|i@fP(Cy)m;FP7(*>3P9l(*<dw78G?95?J=>_Z2RKYQSZpMZiT!
zggCYM4G&LyR=Pkc(O??vE%lzK;j30(CITUM_s{273_COF(r}=?DUD)!_j#0I3fA=D
zFm;AYGud(krcpeWNwPGDhN<ay`MX$}yT#ZaE{7RosA7_LxTbkPsCrFC#h|d8)IlOr
zf)Vr_#bZH1Z%)hDh~QAjcO>gY0ZSZO6iQljvW!a!D|h+qYF3-1ZrB!*&DK<0YQ#;W
z{GnA%k-!Uey<0+otC!8~?Y*Yupdph5-hlM>h}N9ctPSyLNye!XdHw3T>RtJKt*SAA
z08ZmV)9+0l^WK!u0Y4$SDAV8MJw45@rK4U+ZxIe~k-nZDoQbnn;&WI)c+wYDg(i*R
z6u6oBXq5tK6vG-CkK{V;1Zg@nF*pk|0(RoRwqDs>*`C=`NCUIcYzf2(k$<7^T-jz7
z64?7mdD8gyv5QPYW(3<adk|ADj;nr_g1fI~MIX@-@~pfaAiFx5nZch2;TTy*czup1
zQ7M+6g$P8ZyEfH-6YtY!ssG?VB8BKRGebLV;iiWz6o4$)ilR`YA_~!{IyX^*&_<;^
z>rfZw65@!WVoVe^Tx?x2wHmdGB^%!3ay}_$*=n|}y5?>sW|!gowi&t#_eboqNC=8s
zhpaWCeL6SiEnA^dDT=6e(-kz`w%3}Jp($UtSG8(!Wn-gw)9L1@EMKY7cI{}Z%X&%w
zBJkFYcUh@y;N?Z+z|Du|UnO+oRJi^udsrY&*J$o>)%xC7_z(hlJKQ#0q_&&)cX$7$
zoCD447ueqa(uFb4B?eA(@9{BH_r<(JJIALy729>F?mA!99r}QanPQu#v5}D!av#t^
zL!r<kE8|ZY>Qs#3GU`EXt74}=Jg@@8Lih`KauiA9rBJUOT+B^qygkTAG^!T=a-va{
zg}nC?syF9VtTI{O4SlvXKN{!D_?nG$y2<W-kWw`_B|!aS5%j|nwqa&6;i&Qco_Td3
zJx6<`EAf`4X3-A#D=}?$(0;^KT`5*(^Yi=F#-MfAiE%z6HY>}G#zzaR2rPYycre~H
z!jDYJm|!`~<2)o1STNN!VuTw?P+;Wokf)tA{of1nf?031|IFes7N}3V8Y$ofpcjml
zIvrQ^1f1_n%!M%Q@aF{ie|tD5u0H67s*w;x4Z3EQ)4Na8IV}e1O4(F9&9AZXYbhQo
z>A1IZfauK#??6E6=Z9S#oCrxHzcX#Fb&TTRO!k$ZU*#?XWbsD^!T)JhUC({{tiq@T
zcuW2D-CTg5R3ha3wYKtyF-2B_G2%0=o~%N)Ay$#u-6#b0k3WpfF7Y7|Au*;j`FnDA
z=rpUA_EL0YfK(XlgF#5|44Ee?F+nFnbUb@NH3<+Bedd&-RW&qr#9jmX2I??Zi-=cH
z(mC?kzxfz`nS6;)fvnhOP@<71j-0Q`$K_)!9boR46@DRxY*wfuQIsN7DMzowsu#vl
zDIcOplA(~FS|i_I5ZxfL-x<G9?|J7;s)OCnp=Oo9P@)Lq@NxJ858EQ2m)&{O`rAI0
zY0w9bn^%$R>6)&kS{F9B;^mW%$N$X^I`6LhH3zGhfI#W<1CmHJnBV8VfF+8-#~!#d
zN?C!+q{%G!=Rz8CI793LKA^Rrte0ujx@v6Ch^&dstc#Klq|<pelpu3&l`#(d=D^j9
zQ%Ez=$b7eJ8edx@Eh{SnFooitHsehLmwPIcW2br(7@01kxA1GXHibeJ8S&VzZ7zcC
zXPAAiol}u*xQ@-u&D1gARCeC2**EIdxRanVe(b&|%Q0xuJ1EI8`cH!i*bUCRP+MDW
zWEz5lK85|M2LV!53=Q(HI6!oht;#-NIoQzN4%j{vYcQ3x8L{sD*_xmb0K_+zX39!R
zHcfhs>t~IdOS1h|WiBE9Z3d-k6;$C==#qfq74>|>H52r&VLPAqzWzBU*ZMxiRafPZ
zW8yyN!+h|F&+Sa;vaRPtcTnc;(&l)pRfEs)$=Y<98F=*clm*!8dkExuI9aZ~*go&p
z51s#*Z2r7n>mb^8?+)Fsj?391BK#pb-gi4o+z~S$@jO}{%(dpwYeFoD$f0`iMQwMH
zNn=@i#k+GLutE8LfAmQN(P<2!IrmF@PrT%hR{bn!TW`628myfW+(N%o#$K23-#MCy
zHT1UQIMe)8IA#5z3|LXxieVS}d*aE2#elt5zoTG=uO0KH8Z53ihmMcW;>kCf02#p0
zX>(J`J_6S&tZ|Ah%rfqBPp2^TilfWp2&host%{U-mW$SH@uaC;$#Ei5@c7WDE{6Zc
zQ=t>67E|4taATl^3bo^Nl4ddirSZ3He%P}lV7))>&?M2MQK*+&qE}JKEu%~m(R%}n
zf5ku-YmO8q@+_Y3ySGq`Wyka9P5}Kez^o>w=}KXoNM7dX_d8g%M;mIu3t{0zF%WvF
zFLb{<p}~yU|BH56%Qe;8fW8_}AfCuNc}9?GJCsJDy^&p#%c`qB?!AlEtfI<aC2gw+
zgF6p<u?u`MM)_L~#=jfMPI@s75&{>unKkf4#cO2B+o_68$=yk%pB)m$S4wXugnCpM
zes<O?JpYRhb0R81|8p;-4xjr;jM+m2Rji?G)K#?gjO(_^Jn7|A0{toRnE!>7m0BFD
z6QZ@beOugMQJ&olv#yp?_V*huaf448x<*Dy^K$PgDa%f$8m}41vIXy}gEQ3TF75$t
ze3@R2T8&nP1{0ngFLa$nY}9EN0_JO8QEXIUZ~-F8Qb^gFInF77L^J*d@Bp7+B%=Hs
z6xg7vG2jC)e?>sH<w^Zk#d)NiyiM1Amch0-b}{%U9J_+)FK~%~P!jd3MTU$C+o07e
z&n{!f=IA~Xv7YTRqq=zq9$TIqkB-g8GZPB~u#Up^x-T(_Hl(HJne}t$CETK2+N@1#
zrPK_3Z*!u-E>%^s=LoG|xvO};kDdB`J9LZ>8V3!jtM;F0W?r<(b?^1TH-Tj1u0Y$G
zlja}G-af$2Nx%e=bp00+&Gh(>n|;;hTFByRGcWVEbFajAt2b*POZNkja-nSP>RGMq
z75%Kgq7uPnA5j&{C8dZ<ha$gTeX|^#<Jfika$h`cR42P}vgg;op&sT+e|h;UoCeO6
zxJ`%HNXkS9UVDX^OKy<LEL_8Td~hXYO;VboXb&?iH$dv!P28b<_$+n}SKjqx<N;$U
zzRf{Ms;n<t@{ioo5E@g&lAj>z7~Z<1at0}?Hop_h8R6HDK$k;0FdNK3y7Ml3y`$yQ
zU8J%OTaAi4c<{y_HA1LJbsHf~a|i13yB}%$wMcaLJXerQu*_r4k@QU9yH?zP45q~h
zHUkM?K26V3l_>MY+3@%x&wQBs&`9*5WcY@*&0v-|6;-?&pR+8!<pwChf>6INrZN>8
zRv=KGTU+u<+`oT;$(=&6s-|C@`Pks<Hj7uC=FdE1c8O<F5FX$j{J-~NVS6LwMd71%
zM!(45_=tS?zIQk!NKV1)^DqQiS<{n$Br9In35hkwbKy5tvP^yuGdZPV^@dk4`NKsK
zIqU`=mewl99w0}clXSI_5y3+)#8D3qXBB|0DOOHmMCQ2vZj+^8FkPB7hH6%WE}7+v
z86s;ejeEh~)Tdrb&vn#CV{69&YmscPTPhy&5!*XiuI)>WUg73>0ip&Ifm{Sf({Y^a
zn1*P{ywJyEFBJA>Z&b(rok3Lm(QpNPo)6I~GFg1WvPniv=ra9OWl^GJ49fZ@c`I1B
zIZq8o#%?F<d<!u#;aNnW>Kc<obKYLZxY(dq1Mqq2?-xO6SCtWXo9Nru!^1=5fh6B9
zrK-jMmQQl+_XG2ePXj;NCzhoCH*4eK?>{*=r*@2SOrcjlX$}hW;PqU-W3{jIfZjRg
z`JM9vdAvm$1jkSjfN3kgSatq%7mzPenB>%{<rwh0A2IMUjPgz0oeJBK*QC{DcVf>w
z!IC0C8V^(mJUgl7!l5Q`c7Fbdyv4@W7KJIX8dF+kZf=ReC-W}78s2Knd<>*85Arje
z_GYj2TX?k@eSKd2?XxpSz?9MFs3ZgE`V^po6Wvx2eGo+}CXAD)P`LuU6$JLHetlFZ
zG$2n{=!X}>XtVv=^Cm>d=Dp4cDVncb2`+FgK^^hfG@4=hFxxg<z&~?j&<L}J5nwJ4
z>gr$$^r~s?;a677T~x6c=})7c$JEzV`S<exheWXVxFa0rSc?>mI%!OVM}z$XPRIg6
z9#xWvM5uN}vazn&-SNlqDfP#ei~l5eZFP3toV;4UWHM6`?sX~CVI<<I6h*B1geTw#
zNKmQjQH@dies~mo$CY7?qGeSf1@BRvlm}ZlK0nAYrdXL0KaN@5dv1l|ZmokRPV@B3
z+o-qiMP!}V>G~S&^Q<jT%p&qf^XyHY<(9hnqn7yZowzbbI0!zKX-sFJ6h2HGxND!j
z<71B3v)4OgT?2!|d1aA8(bG9f(9zxF7gvWlk&jGb@Zp+pbTB1XFf||Ld^TEL6X&IJ
z)+TX_yrao6aAIJX2`A7a1Ki{aaTbIwuPNI6^SW{kF_jZnyS+vn`r(9*O5nII-eHfR
zOG&QfCYVx{^rHD+55Yy4;d~yYwhX5C_y34YyEF7ffL1LQJ0H|F(BIGTY?N>ObiFL}
zFi4OWa42g9vHS8Hh7p!gg~!%S01@wy>-SQ)4Dn~OKg(IrHJ_#(Oehn&dbXQF|AzAT
z*w<XkkVbJWkedm2?-3l`5s&F9r2g7gt(63{zTy~~F01I6{;uXSd67E2#<u^5={nO~
zq@FvH@#DU`kKH0u$$wXo6!8*|U9fugs3jcsCnpYut-;vYKQ3mgDZv0=>wALg_{#4g
zIg$pAv}z2fe@KNf++{K#2w7q?>WrMya{aMSoQEHnewO#oy*KJ;=8Gn+XR?*kMk3Ei
zgeBe#lH$v=eER#wAw=@pozH1<lT%Z!z%>9AX23pbEBtt*`Aledm{r%@Q$N3&lmY!s
zdak3RBNI~cY{=vL@~qSD5O8|ieevc>K9$4kb2UUXpF}BvERc%{09i8e^NvCLZv0GX
za@hf}=>K#L4G(&rTV98EcWK^2)57|wF#hCNl9sX4oqGdcW|&lwc$1!D7FpFbYnmK7
zQkVv}{YmN!1Xz|G01O4dc}bzX+OhE!sID9!p64tyD%1!GJ6=eAS#3kqQZ}IJNKa4y
zbL;!GOVqOyL!Y;^YMpNh0`XwzoQ`L*uw^4Xl}*J!rwSch#Y>*=HAr|59_AP7AX9}m
zc{##6dGE@n?Gw@B+}jLLcYr}EX-&|shA~a9x~AssGYMx);0D}L-@C7$+v?>wv{52A
z^-N`#w!(a#*w6Lqv3(~axRqG>-Mi6+9*f;)V_d$>Ep$IzwTi8wZTR|%a8oMmPotv_
zOYVvU$}e1%gv<I4slzMK2e5t@zLhx-)9RqO#6ZE~Zx65h-9x5XJxas)`1OcX23wG}
z<hUZy?j9x^)rdD`=t5{<&kO)&htUVL=YA^s9qq#P#apFrnO{J}3_lWujjK>ZOFO-M
znJlg!Hjs*eFkop(`(ZZdQMvQ7-~H)%`xRbQ$c&GiofsITY5G+WbXsi1`^~dHsmA$9
zkhyq0KN^ZWT?(Ao+s9qQeU7`dh`xA02uX~EZ{+BUaTmwby~4If>eaN<)H~y;u}T5n
z;g)roGg*IX1thGQ0WAMeg9NieF|fveI-m9)kYbT5t{Qpbs)=VXW^Puxnnvo3puuH%
z0cjb=FQ^YxI}J0nSt)1VW*VB>l*q|uc$J3-$2F(os0RlPu<tsKK(?!jXWeBIe!l%b
z$M9}G1x)WFW%k$p)hN9@(1jId97Wg99z5>jJ0<q)SxP7IivG#D=z#!>^k)<yJm;TP
zq^^#hUZDn4(e6Gi<HrNL_{BwTH~##H2mbaNq-Uq7$0PRdbK!-u^6#*j&7GsvkSj>?
z>3)c~tR;LH-QZKzHTBAig`*2ATJuA7l|m?9ekok|>69}K{R9=N8g8hs)vKuoMvA2+
z!YQj1P>1lX5u~DQn)!bHOU;L&>uU~XW@g4rH{NU88<#cGpJYERNo(R#)JprjSeZQm
zue__Pt1nc$2A@X8$5DZEr724I?d#Ig(*I)%0Tc6??bqk+vg#ERHkk{QdDw9MQz2kS
zhC~_o&y=)h`u>?~Fox{g^o#C=<;+dFRuv<jcHOdV%R1i9uh4QT<T<KeC2EZEWPk2b
z(DWLY{&mmP=(oIK1p`1p?gW4@D}2h>d&|lNt`#IdC=ZS3lfr&lK<0z}#f8(I?$Wtw
zR@)8COJjGl`EH=qtMF7Lr=XRS=<3B86v`eoAA5EJj=Pypk}OfP?T?(U9iT#fdlhaT
zmQHRtm56kqQ>RZ8aH1KCAl<EM|IFc~O)RlIuikdb%O6@4BrC#kzvPAEIMV(WhnwY@
z*rrFw>w73E^Pf8xbx2ykX({~YDkys@#r|?{fy3x!u1+mWeHA_=K&So3276Z6mW|J`
z;GLujIOMBZfpAr*`-xorcRoC&uq2ASn6IJOp{jgl0JiidW_1qO({DU}KM>0R^Y!^}
zDjPY;;YWTPqfQ$&>?T3Q@CE#DK6$>6On~*i=uXb|Bi-x2#Mi;CEhW<Zx%#jpGr2B?
z5s(h;%JL2Fu)54zF6`@G5D+3CG-~`Zl&_|n^>rwF)&%I_X!Dt|rE$71zw^s7NItcZ
z=Uu2XWMoldlF-s0&8rvTIyLEe=(9=<P!Fla$TdkMO{g+vT)|O(#_^Vu77Je>#hQpa
z3SXUT)MLfON?a9vLAeKk{!M#z4?G@|c62}+vssP&e}BwteMB(&B*uxbYw-s!7UZ^+
zm6Zf*f1gt}zk&+YFUnI)K%KM74!Hc`H8MJ>o5rS)LiQ(1UGKl^Wm<X}u?TkYAOHMC
z8z&o1V>28a_`})gmm?w&4<UrV)=(KPZxDgLiV^=h_;3}H2e~v!Ocebyf9`?&VTvIx
z<!^}8A|viQB*_(Pgfxm4O<|Z7MLeM0UnFFHQY@#jt}fkpy=TS3&8v-Tct_BZ#}jk{
zOujieISVJA%`GkSLWOKwKT0cppKG7e9hb5;^pM3%agkzVPgxxw*@}y4DS~<+tL>{N
z18f#vUKwwJ-_C(IIb1o403o3r1vpX~0e0pyhMnHCqpe>6cT7T`$D^``KCHlA-kt|V
zz)9sLrmIUBDNGzSxt^)|ly4=d1Ezic=+Z0b-s`XL|1-;z<w`=;m^_n458P~R@X2;9
zoSZ&c{v{+LA^~hPNtTa_#WuDEtnPuCNqxR$RiMpYoH1>*RT601fSuy+_3am)lamv`
z4?9t+s1L|a9K}->x{_&DuXS+m+ic|DO(>~CovH_)ds{=-p1qm_&fPCYUaU?mf}GD7
zDVth*rz6uC3BNJwW9+r1P8?+B<jB9*@BUlz8Qjw|1K0~>u6ZcId+AC&b;8?dy|i-t
z34E90VBA`Ydwyf6HM{j{IzgEVp4U?jmxkdy61Ep(L14N>mgi?Wk#}gOwDy?$k9$S>
zE0d~UrN_lmHu%Q0ls3E;R+sivx?VhX@KCwndvF<otDnqk>vZ|7L|r5UPEm0@VrsxB
z7fcA4JGp-!w3uz$GgGt9ySlwzV!$+ic83e^yeGV7$E7G{p^S?_QDtA{EirG~atS}R
zVGwVf9}s!*x|8UAkf!}+WDodsV#WKKu>t+}oJ}JMI*tEKsdbCo=oqP7Oq<ldFHCf5
zp?gvpz!vKt`*Pj(q4UdL{>{PX9)vk_w%9BGL@HxCwSMH4-r&qE1K2FiiseZZ(BSSk
zlPdnKQwu-;N?D+9C<Ller5S1f6E_e<oCZE(si5sf#}$ms)baZZ{h*|j9S%mxy}l}=
zboF<eqzt%+8g#j(y-@8h1F8w%uBmUIJ7p?P<7eY7V+XmYxsLQ+3)+11*wSRJ+v$&N
z=~=z5fJOnPm7_UdVVF}8>M$_J$R&rCvL-*WPcVG$eC^MrN{5*6KsKUFo*&J6ogwQl
zSky!)i-ZG{|L|bh{dYtDuWtaLhyNf2wj!4L5)-R`L|piFmq4lBYWVio5hSHdfe67;
z)6pqNSVl2qZg{`{ND@lcA^TJqyg3lsfkKiI%sSzLDBe3ahYN5fN*C?GEiJsRmoDq~
z47c5s17;Y{HOp4%u-=(w>lVD<jp{}K(;$EnSERv|MYX5Y1zadn1Ri8KEp2UBo0fIk
zzpub|lztKP3hjJ6uJQ-sbNw&--u<7gZ;C3#IuGjMKIJF^b)MV`0Cyy41ZdqJPI_MK
zjRGGI7N)0egsvZi<gzr3d8W6Yp8QF}R}P~L0XV~Q0}#91`U+d#IsCMSX5^0WXl8mo
z0u2|ib1X)mrqB292YA)n&*xrkzPC((n}+f8u{Lni^MYOgTK)Bh_}S<E97NPr$?)fC
z%?K+j;`4tURBmh8E~0{o1k;2>=#~TI!s9|s>?aPQ7LaztwR0gHiJw1z_PWmqO9ON~
zzz<VVINxyb)aPBS?L`C3OqwZVjBlBL`1#XB?OIzpVj|Y2G*&=~ZM^vX_)nV(#W<H;
zwU5DK%n=HNUt%R@2ZN^}msJE-OWHcV4&MSh6ki+TzkD3tS3tmB(#yA0<gzLfr@P}P
zJt7(KH2+qn_y_&y_g69Ed44{t1v^gCznAP-xS3N*I_Xl#?v%;GgP_LLQCh(v8@|E&
zkI;95{*PntWVq7YV)$ZlkTr@&3H;r2vG;HIuZPG_?3w*9e@mD1S-D!K*L7ZVZ2ueP
z^}dS87)3-B@qYb~jbV<x;nYGLPqWK#WS=u`>p)<2jQB5aY>ydtNt!Os(hGYRg4%iW
zLVvo^F);C2c*Z=@&@3Pzx26We%F3EC<J-RnwPP7<Z;Da!qFL;4D)I=cYx}RBR=uYi
zaJgDxMm@YdZ3@Hk@UA_H_^7P@#gm+|LT0Qnzv6Z?t^9h=x7c~6(&pITNkgB4Uau~E
z4kqfZo3wkBsI~V8utW<NkFSrJXE=F{`_@#l9xO=Yp{sxQ*J^jhoXS@*I1GygX5&y{
z6ME&&1{x_^_}wdW@MBB!Bou~W1qWDA$(XVGW$>8AyESMtI4MA}#D!Vd#GlGN*3HE?
zh;8D!Vy4=MCh4hBuK8tp4e^=XB5+UiiG2QBEZ(xK_xd?tf<H@%j<ScG-@5|+IB829
zag@EhRt)B}@=U2*xml>wb=gOiHVGY*i6|#dtU)_H^{nL_B4SO)a1E(ghk(?_In2&w
z$z*h?tCE{A;Xn;w$LZ-lsH2Nq1qye;CU*6+J^9Gn`F)VDTC{uAZxL-EyK0=;s^r1K
z!g>H+Hynv#ik8y4c-A!=F1h!Fs1b1X_V)gOcLqZV&|dDQX&bR~oDUI=9k{*SvBq`B
z;#2@j<|>vU6tm!!TGrV*I|=Mj#zyD_I+8d4t_<VAEcuL2U@*Av@X+d*%2YhplCUI|
z@f&HRX?x$y^fU)AuXSkUy|48nrOe*aYWMhX?+K^_u;EZFzXgPhfF;epciq=%b(c+b
z0rCDKZv{lAYVqwKF$K2%W|$U>v-5|tgsGMddJ5*#^=?dPPt)rBRtE-JoXKgfs}O6=
zbdZ99?j040HIQtr)MvP)H2a>uL+jKRh;nyubSz=|2b5Dl`~!?<@ri#3(H1ttD%Ri4
zV^%vETk8I)R%is{Z`yA*@txg7u_lhLSgKS#yF76?kz*c5;Ui>TAkYx)Olg!mv8W3O
z{Krd-i)HRl`<vm%{=_u(FPTo*2(95>ebU7$e7m@S#;YcR?=%jcDEf{Xd~h|@vgxj)
z*yd@ct$6Rng01*SFSq=F;5qQOaMMYf6)PXxbdY&SUx9p+&^NVMHSj3IX_;nTPK9bt
z6$kNO6lnlR8SlJ4mIvNEZQucttxO{k>A2&ULqUze1s)?OFEMiQ9_R4H*glll?{~rY
zJ18-qaTAS5inuG}T=ALVj$B<fb|_yD%KnLqeY&jBnpm)^oSdHGP_I(0O=Z+(W?^CZ
z+1fh#^hCaBH&uK!mP5NF-mOrfR-(fK;HvSkF7>JrMM*APa5QEmF{x6>l;U}T#B&ZU
zZ7m%GBl(*aujMGTZL+Y8TxU+-jGqs+8wDL-Hd|iab2jNZZMMs{=tJy{pE*v*P7q)R
z!Kyder^ukK+$lMjd2xd~PVG_K+;A-%x`o&SDf1GxihX=F#`?%d2&a&_Ur$Au1GNg1
z6q;?O0Aip@fkEx}38J{?T<kbSoavxE>;%<r$8V`NHFI3*ZB&VuRB<&?W@3dpg>2-i
zESd~yPzOiD()kk{rmroc|E-?=<RdRteUvp@SrsS=rjJ$65u4H_Xky+m=;KqQQo=Kn
zv&A>>VE4M2K>f8{_Cqv1W!WO1JKnN>$H~VhtEi~R1O6&Nw29tt1=7CV5fZ&=&0O*a
z)M2a8dU{b7(&j*d8Py2{vKw^!XU*CSny{I;x_-B6knP!4mJG-LHWc%d7TEGn0xr}F
z8$cHcyp^bLRphOpq{&49c8xe+ae&p%cUQw*zS8ZB{M8?u$+vPI;t`uVC8^|;ls9i-
z!T-8}3kNg|c1@)2?(Vjc33g4#qpc6A-vJXxn2NO~U<bs3If-<#Mz{viW5bK2uF?9I
zoBRIF(FOoG=nhLY-}k(&n${h=0h$1(px^{BKK=0!@CUf(mgeRk5M$3<6$OQ?AQN0-
z+>UV#^)CJlu`|IjEH_+^2}<Ur25mcVTxiE`Z3dOQpuRt0mK(G#jc3hq4&)<@d<|4{
z0-ddux4_I*Y(I#UM<Kv5{hrsW*0RgGq2R1@!0wLC{zMq#XUC~2;a`zhe?MB2jJfD}
z)%1#RIVOldQgIH(c=7n$oHT(OzBZwb_fseem-j`mzXKe3M}dmIe#fMHO+&A@Ilbo!
zn~^^;8TC<!T6iwS*&beG7sU`?zcdwIq{_e0cQJqU*Vj8=x7H21JE!}fSeT4<b6W&7
zst_@;?lXhzaJExo1SVJZSU-PKW7uzfTKNj>Vb+0iX7n&F?|+iBy}w&Bl9jK&^RIRK
zV#_y*VWC{-*V)|>$PfRTp#PWmZ#t#f)`IDjq!e6ZU+;cNU#*fdh3HHj_GqcWZfO%f
zx&LH1<0RJxko-n48oS{o@DM-}BQ~@V{t0d+lc$p`H#ca&D(Poa+t~3j?tvNR_UEJ&
z{*-=&tRJ<m`2U&768oT=E0%E<%$B>~Ouu(>aCTp_`E{Z%=BTbMUQLjf6#hH7Wjo-T
z)Lv@OzxyEZHXSQx=fwQiCvE>dGGlXl`vHbeLo4Ts(N~3LW=?!)!>IALq0ap;@V|Ig
zz{G+9$S{Rq2qLT$clpd3YuivRc-{t->=do3fIs9N7_&A>p0<WA{>hq{?Y!t)t^+#<
zc-cL_A{B3gJAs-8gLl_QL_DcTD@QTKVnlTkPXOo&$xkolZMbZ`1o2~i{I}&p@ySXB
z6Y7s={(5*J{j}tzzyEuTD!IHx+hy;j61OL-dD<FpXU6#y5X&50mR5Z^?PPLZ#ocYJ
zqqEbG_+ih15CPa5G5hqSwD~Ef5B<yuX(BqghcZU{692S&JFRLayrQDY<RhK8OZvb~
zz{JAR1txTo&(N0ER+&PT9q<Un-W2b4u?oXwTAhD{bHItdLAysxTO0oyR((>_GKaBz
zsL6_tTCq3O=&R?fdGB-1r|X$*DVnNg8~2u1JUgrE`*~ugmHp57f!p{T>*vo6O-(}p
zngJjW0kt{zOP0C)Y{lzma}GRho+pF=y$qNKk;3}2Ksq39bjhuNdqdKAq@Wi_kWz()
z88G<)Uk=3oCTH6MD6q${+!F|v?+woB_v+opbFEh&AWO>ucRqW2TbzDD_bp2nG{|sP
zR#pJ?ffu(qkZ$X~ji`A}nUsw@IXM|;Y0%#oVLA%%C;(_isVwCX7qrfXF==V3OX#q!
zEhv3Kzh?jgSz1366`Xl3gY7`U_oM8ftnsX8*K<iPuM3aaql>UW_@YK-Jw9n{V7+j{
z=>SqGdoLv1oYTOv?IGB)FFCQn5RXi&7N0W0{M`XW>VZEiR5Nee#E=kq$iKGnmC?L4
zPl>!L!5w5hLxfy7`i^4JE@b=VPT(Twl1JIVClXYq)V+f@L_J^{{+FvsFAYtfh*;{q
z+z>7k=EItHPrp|W^rF@60QF!TJxwlNs!P7hsjHsmxJ*$f%bZWaOu+Vgvo?L5T6RGn
z06eV8z7`RHV^#EUn24o5V8iy$P!~&A?}Ay;V)<m1=^8yz+3TLgzqDvW{`?G4<0V1=
zAatZU>ZdV2eCh2TkQLvgNq7iR1^+j1H-7@+x8+*dT7_(#GoDQc_c72);S7%t3gJ<>
zI{aXAVB#)GQ7D<)JG}FJ5q*WeexrPv#ahwp0Zs!BU>G*@@VMMFzDWOx(@L!dBO7V@
zFu{X&`{s0`K3`f-;pfy{QgCj{(S~40^bGSVV*Zk{$u$&EtpIN|r-nV}{`J?SP??B7
z2!E|KDd%Ctn1X)XWxvQS5a`i^U*hho6%(~9HFRW@^RiXR-42fN5ZcI}oicp)QYEOR
zcc2J!%9yH=V;<{-`inMt7_Nsij$r}$ny(6+{}P&{js{DY+7byCR^yMp5o!-x3S%~S
zf)>SUvp1VJ)7{!8hli!Izij!AZD9>6`dJY^Dc+RE?Pbo-&!<NG>~pVaUlO?&(i#cn
znegtq8YX*d^$favfvt?9NIreS-U1zf29_4m0lOMk@|(rFz_g3N`woLPw+OkC<Y62^
zVPVCrI;o-%h!3Qcs^Qsr3UwoSbCR{{@G!$OE4LbTD$ofYo05{UC{<3_tmfS*0FaZB
zkpajvyQpZQ8dJPAfbPA?v;F-1D$0K5CN`hfwr_0?MqoQQJKu=DjJ)nZw_im7$R#EY
z%4q!hxD3?G<1BE?{f{?(PA<})dXDK_&U5Z_Vjb+0(4?EC6)r7j(uQMt&}x%jJg@*0
z`LJFJY<F7%AwK)<1=AEr;|CnyzI{95R>gM(SgUSr<eR`J-K7MuGj68a-~D3Q@f>l@
z_IE$%<uZZXc>(E`2*J?RR|`nYAT}8^xm=>gxpJw%yfFbs_6Ks!NSVV=ahVh;Kas$S
zxE?CqZnokszfs$cs00bTiBHtbYBMTC3^ZzYROz9_;w-&HF$3X96~Y$%pa&U?>S(?9
z{aR?ke?IIYv-+pXAQZFJV$oDIklj!qkdd7jw0nPlel|=<1hu}bmT{4l*}Ql6@D8T8
zG`qKiBd)8irs+%)Cw9>jHL>(e{~UDik%xzefj}Pr+UJ+77EFyTeV@Qdnd0H0HNUVe
zabK`f@_LSXnH06K>2*8PukT7cj^9gz)KFG1%p>RtB*>z$7!d}u`7jj$sosM9ZwBy`
zd%0P77KMbI|8}`}M_A^$$>5f0?!D;a4}+Z)EL3mYT^XFO>;nv?L{hR!y7@Zxd(=$g
zXr{;e2Vyxow&_uS@ka0&X$7GjPkTKtAA+7XDcx*O*w=DD_53@ii_x4^R~G)x2OEw|
zt%}-s7q^$d-pzHZM5M=YA8z@3xQ;Pep0ISG|DlG!m(EegEeqwLk59X>t4_Cthb2m>
zvuGqL8O$V1&DvzbH|O0JY-btiRwI?fx#cI9)7xuf`u9x1+I!ejkyS_i!X+#GkiBw`
z_Gh!QnQ~<c%&>@rZ>pT-K;u_-Rs3v-eP3T6MOa_7QkgthCD^~ud(1Ar7oHVZKCS4&
z0v^(<dLOTz$dMLsRs`Ua3>Sw7D;Dnxu#dz?Rvt-x%!gm@t<R=`VdVMTi3+m$P9~%u
z=<R?L4rCC`o`Gh_Md}^%wCI*f)jtzW!r0QiJGp|up+WubeOm&6-!ka*{tk>JKx+oX
z!H~vBewPGjHFz*$j6=uAcBsFAZIzC$?(YfX7Y+cW1A53*_Qv~`J4JwKG&wWl?&iiZ
z>ooCe)wdnpQ0vU{wHi{jPDqrwBA{=@ww`$IzQ+f|Apra#ej%ZKApKTOvx%@;w|RQ)
zz!J9E(zg_m0CdvS(lR3YdLw%Cd=Zon^liVKS{{Zz2a1!A2;DBI<Mm}Utm4)m2nCx$
z%<ei)Ta|NymZ(RCZ7lnu^``}O0CYc;sz9wmrD}1m+hfF4FAIsTp?5Zy@?59A2MjYK
zg*+N50FXkg*@%m)4GUHHKD;wnnmkh$q%Xa#(@$jCJHGR3z9@a~-_Uz0et_mZWM`7B
zl{!ZTPhF!R#9ufoAc34R3v0di0hV$n>@R~h+UQWyA^@o?Cy<cm;STb9I5-|oUTDIF
zatUBKhhl$`pegRnAQg|RGdDZHixyI22u~QZrQ}90+%os}R#>PkIz0R(P7)!@_Nhdv
z7ZW+l;fsXHe6&wsLZ8dv)huJqiG1TRhMagvZik@F^%tg66_#HZ(BJ}%+GJ<QSXwMV
zNN9Dth2v}>__a2e0DlL%m^-=CG0OgZR|A9(@`<4;a-qf)jUss8m7tja_CtnMwcuql
zhkgu*Hk)GLxbxKW{sJDmA&uw{_nf6R=>-$!k>v9YQ^19jC;)^wS%+a?7Ft?~$3Z=8
zrXrC=xmijkjh-;jBxy?a#}qp7Em-~CgbT*V4U<ni9*->A=z~ZUec98^>y`P-lk$r&
z*+~w1q-5aLx8>-x!w5*e6!z}Bxt~Gf(q;Tgc*9yi#|orlW`jaa%eW%Nm()l#OY47w
zHxcRCL#j5Rij^c0Fh~4k-8?4iA+%$jk&Nmnf3)hw+PI{BtP=desgZH)&G~s>gU5*D
zl$V_n<R{qV`6)Q5g?c6JRb>CYS5zRDM*dx79+6;X9!o{*^EuHt{b#}O)lI4^lYj{S
zf2+q-#Q-#&PGzm!NTDas;^njy&totBQK1ddqyKLN;+8Bf4&FB32lLX^Nfh=wBx(9W
zlT&Wj<T9QgjrvfUkA8r#B)91dM?ztxT@ils9LYpo&e74)6<9ZojmfsVcHDPU>a$PX
zuyuSNaaCwpf6K~qYMGdvR2Ua)B`dVAJM3-&@4Ve7W8<<}o1_^?8z4(G;rV3-&J+*>
zsMmTmgb)6Cfca;wL0kXiBs$yl@XwA8`IdD{VD@|CIV2=tOB7dhu*mNU$h7i=HrH+H
zm^ObRT+!96wj4D8jps^#YB&xJaWHN-N?4-a6~a!oR5Jk3Zdcgk3n&1?T<%f6W{N(~
zivnX-7(nR$9P}WcswVXMbaRy1THusDJ{=IW#2kqG&oU%&eP;CmF(vB#ebUYvyw$3R
z{x(A_wuA~m$;ir*)dkxwuU&!bmfV-#&pYUsEiS@6pRVY;9EH-Z^W?nKLGIhwAa64`
zN5AxdE+O9Nnut1LSc^Y|IULrz{RygMT3S*sBTJWgE#}_xdVDzD2RlTXHl{8ibSGZC
z!?))q<Xy5QC;w&FboF>>@hJ_KY7Waqo$KQ!y!TP5Ejg0RF&wQzh$XV{jP<XRc_9I-
z?yZ7O0UG2u^Wdw3G%BISOY~kGIO-bqF4_zt4Z8*p==;ZK9Qg_Rr`yE5gZS;21mCWa
z^J!r#Yx{*nGK3iFQV4WE*76Wy?fsZlmX(_YU=F}At1cJxi?Y7CdAzR=cGD@*u1Ntu
za-s6$Q?FOJ9v}Ny8pmmzsSK?^LNYlx#k`=A#|W@5PgTFku`A7|q7y1Liy)*ps%5d<
z#V<-xKmq;7rWzH+N(g7`F?Ycpf5`ayv=_7+^l;N-;TjmihWh^(?-RYx-D*`qJbgVj
zZd&kXMm!LcY1{~zJDf&|>Th2oj%MUkDt?C*q#wfkCGT=3W#=Ys;hh_sPb@mGPArUQ
z#2#^L5C>esu_+GR(&U<?Dl~W)S8$Qs48FiISfbLEvAXo}@qtxMNt~Teno>j25mbr)
zbBlS`BBvHct#+H!R`@XaL+bbbel)1V9!^JFZY4#pJ;Y{y*_x*(&SbLwotm5(``EOi
zV6GC`w`#>H%Kqu|z&PEz4k(zO%_`ktl9i0@_Rk-NTVzl1;pfxe^>@*_Dz+zk&_GhB
zqFEaN-N}qR`MKPlqn>^=2Dw?VYf5d>Gcr1SGdghYyh4hX3IkvQD{E^&$Rry-6|ha8
zF#!hfX#kj>0t#A*S`4sy#byD_mztWmfsyOTLtw#<8<>KC<}i2~(Mfl)`MCu{bvH1>
zQe`GbvYGONd5@fGKfwo7y$KPu1K-#8me1N;zCjp(RP6~74Gk^abQI93afVX8kr{H!
zqiTZg*i8EB*0n)-7LZebe0m3YUW3&8uD-N9+mfhNB>=f~4IGoXVds1nxv~bgl>rwC
ziCxmfk+w@X8M%2$uCm|&UpF_2vUynx1Iz!EPd(>#hHV{P*Y`Q{KCuN7mtKDxZ+S<b
zlT&@I(`pNq<@mlofFgWDHW7%Nl#K0&{$yRrVz6fB<o^C6KAhr?{2p@QNV6bS?mrNz
zeN=`?h6M*a11(%4t$z$xp<K2!y@!&LY8OQ}&#|m==;tY`&n5YALJ`s|LaCQxk3!Im
zZ>N;xH(ODZv+vPyx|M_2_Af7kmyMv@WMoy|-LesyqeP4Z6Z>w6%{-rfb~eg6TBS2<
z*QwPkT|Nh6ZM`Oig-HS*vn`K4J}M%x&0|QDPaQ8=m>5BZ!<s=$nv=ALKaG>4qSZoW
zr0s^3Rf>k-{U@H~-+}tQdpS$A1OsUaxsllT<Yh=s88K3VFGxWfLudhAp3t^#`2(ut
zF5iD!e22zCPs-%Wi70Nca<+jF6UOe(kB$iZ{-TNT+v%0g8<H}2T(*vX$@?;m=vyy@
zmW~W2w!Hj<pmxXmABq#PMe>MXD2f=65K59&6?H=1K}NVHK&H(>Q#TuU9bW`~Q_n%^
zV3s23viWS8;vo+jOqPU1={g2nLGJEoz$9D9%|)uHSA#jw7yVt++U5k-YQ8dF{AJb0
zjmJoX@P8a#WmHvJ6sDy^Lb|29q)WO%L`tMP1f;vWK^mmHTT;5aL%Jj$NT+icf9Z0K
zFpqog+51al<AGjWhsyJxBc+cy^3L1o8Uq$5jw0DKi)C{RU>N}WL&3swjSv10{i>EP
z`YQl$J3OQZ<4A)+U8nuK{tNGatBw=)K)dq*=~V7qAGi<%ZWj=l_oA~duc|5%Emxsa
zVZsC0Ho%2zfV305@;_DT=4g~{J_rMQ&5O4dXt^h?d1icT!P*b++h|i+7nt;Uo@mZ?
zQmcPJZ+}@kYacEl)8o``?T*W4du+H)BrvkB+$3*mV?gBD9yta2<wJFNx_4r-+8=IW
z`Y|*=<s}kz=osOFal?&5UDzO^Lo2{)Gx<lYgo|SP_+F&y%eaT$HJkaDHQy<alCAyt
zu0S1m6%P*&!10Fm_Uu;~^V$H9ooTDpFgdzj=dw-^GG0^eH+k?PjCI`K4iLSr{~OwM
z%2Z(PLi{U$ws~q0|8nE)FEws9HTD_^MH$W7;_!Ad-NLZ1T(_HfKF^)*uargn3CXA^
z6Vu~m(S$|3x9hNB6{?2H1vq0p?+FDSP>;|;rCh}eA~P{bT}f~IH=#sgF0ro_-48TX
zs{qYYz{&~aHmJeCO$1ylkj@*PEVJByROF~s^?0s&(S|jVnDKpmZ->FQua+d|hEVZV
zK)^#6K-W4t>7Ede_07IFUyPlc;0p|M%@f(2*tN?fCW8X=V(B7P&s|@eU(N<@+nb6P
zcFP}?_@uG>^7Taa+FNo#vf_``t9_p$klT*~&%OMn*+Xbwfs7K70)Yc^uOWT5En1N~
zcozsst~!jdnjh`VojH6iGq&S`R^%bsjM!SeByY<i2MjJW{@wzV1MC@P!IZWN-S?7t
z(-QTv5{jDX8D;w=?B84~#Wf{8y%-aSNF!(c3`(3h`8Nt(1}w=%8tDf124O&S^F8@f
zYP!x3-BL_R_6m&<+opl9Yzp|_?K?D3E~goqg?2_^n<j(~6$f!|p4>KAR#*!98Ftp2
z6pT1))8A&`{@%7Fi%GD8HGa>^#LDWp*&URKTP^-itnVb?_-d)y2P^|61KD2A&e+ac
z7@tTiN!VU^Os?_dID18aGOZ;RV2kG)=var-(nlqO$lgL-mLi)~!0Q8<jhU5I1jGyC
z5f-VV_%YL#Nq~rJ(Y@)Xk&PCei^6J63PT--K69&g5j)kA)dZ~L7o5cwh}XI_eILL(
zIv;mgL4Or}%JY!2cQ=Ugo#r<dYe5NI<J;<Rfu0ILSD888oacM6TyJ$88pc8S!;mzr
z*}rBs)_iR>GasO1cH^R@Z|7r&0k7p$pGY%t3{Dr=xyvgnYinz}A0Gw5i9%CfJ8}M?
zynNjDlWZ~#!E<e6<+|14Vm8T)h0RNSKMeT7rrEdVd3XZ@B$)AiWnq7PgG$MeQA5Oh
z{g%3hHb;~E-8*?3r=XTO$!}2X2-T5E9fjs)&#zCV$X{>T!dq^+LLpI^SYLEIC-;w&
zW^Q9|d6W5dV^h=5vRi*=v(%zjsNNQp^TheL-}#K4?e&P&8(z?7B9#|wSs~S;wl$w=
zbA5fv&S_Nqn=U*h{c~3W^xPlf7INi{eSEa{TljJX$T1?`Nv3rG7i4(B^1+QZYxyw?
zi2wfH5lgLE3pVD`anXh@)>2V-X~(2~uy+pmBv)je5teM4(nL2u<PoY!Cu(S8-APj^
zUu6O_+k$fLU9|WrlSj7|*`s}pH+K+DIC?8|t-bZ}@r{V2D}q>bcM8B1$s8Tj%&c{I
z8tYO8*P^qFu;=pOt8Lq7THK@0rs!{ntsln)C?vY|<zq?333aA5EplbHU-6aBX@Q?n
z{zzw!LsU>mR+}BdxBTw(x@&IHVcV#846p}zgT%-*JhhjEvD4u3HnFe|+AY8ZqQlhe
zFZtT36v#%)Cp=HyF;nJ)bg0-L|7mSn`PH^!Wukrec;4ex{P7zCxnTx49@JQ^%uH4S
zeBOQq8j-sJn4NcbcT33sw#CEWo>H*1PC2)J22@}1!V$ey@`;hldpz;{Dv{^&n7hjv
zWz}Ig0@5fcFjxSL1~>!^TwLBY<V9?!EG=mXPULBD+Or0b%V3}Z!wuu7%1LvsY2QBZ
zv4eN_*-nCT;U5n2IQwlb<r&SU^6+^)7jNTt1IMdFwrMv;vg1Sp7Ep)YeM$D$zl{jq
z6nI&@&u;!R>T_S@1493e4N;l9#yDKfLf_E*T5`K`63OmAR&4bq3^0F~Diy2)T+c)1
zK)8}e)&rzG(h&p7_UW_6I+{@Y76yIEoFzc>!4)EiwZewlgY<QuUq_+L$ob#MaB~PB
zRP$9fLM>T=AZ4{9j4n+!68(5_$Lc2?wH&i&6V#%hFpChpUc$u0+g;?OKKQ-)WF*sY
zRD7(66iuQSk@PDVAp|b+No`-AsjCL1$hr@r=9dA4@3+D>L;FujGXvfuWb%j)^q;?{
zo>WaRZkW8KgAq4gEvj(|S%11WyX7~YwAl3UH!YKG(4dH>>Yj=PHMUVvQQV2qantYV
z{UE!Oa%_b3;F-U-2ykwduM*H_!*_5{+n=vkAtogS$icz4Wx$JQd`iZOr=||0L_Lw<
zPOnmvdTTz=e-T*`{_)q+@5UcW7MpG1y{%Ip2Znb#PVG={RhRucLz>v#CncOyF^C`|
zwb1=UBUJMUx(pRzTcI8TE^ZH*>sULEfd&$-GrI%-?(KkYBhhqX*FWq*Kv@Ag(Jd1O
zJw;c=>>tx?T3eG)eWYIsAAV>r%@oW5`JigaY_Yo8;J&=3S<?B>p<0Z_Dw>B!K?>U{
zRk;GET1z7eRPsjgiPGlP#&jB)$#`8AfKY&0w)f94e?-IxAm9=v-r%EArH$b|J^PCJ
zbbaNL%g!T!)qnVd%&ar;Epi;oQ)~14+Sd*`BZp27#WjA#AdVAmom$_nQlp4IO2E5;
z7g^yF@oJ+YEa&yj^}{#8RO7EFc6oG$K0YF266{m!sb-54TfBR~va<fT_YSJ09T+2X
z!KjQncvRIn4iMR4Vuk{>`Jr+8^yTt(2q!)&uaMC0e1#79*g@|BP`ZGz4`^Pn42iG1
zOKFO19Pk2vQm7oe?AO@OET^q))cFdQO}DQ1-#?7-x&;UQ2JifnrgbHwM<;%g=c_OO
zoW%R##bU?(E|vu=Y!YhsHZi1U@H;GQ>QCx_{b2r(>}A{dZ|z~xij;a=7lD!SO&!On
zU^P&16{|A<Wa-6F1_V2x1;#=nNux{>IBeD-%7eq!UyS$59$EyH$6+g%8cwcH-o)9(
zz#U5%nrGo<;#Qiy+8-5)|0m08-z-DS3P(zQ?oZs8O4tzfwzTY`*N^;rnr3mLh1emN
zaSyLuF|^6Dnj2C|f;cswu^5UUj(^)3eilzL<YbDUkA@g66Pd745*d@fdx7e2;Eq%F
z<{@d+n$GABwTrEH|MFkTG1CA~VQH2(<oN?uoo;9p>^TZ-L=BkG;<w*gy$GXduG|xz
z+QAGRqot*{J4BlWhX!^W011Hfi-!jfz<#UH?SvVoFH2z4T7M>1UUp06r?vU+pjHqH
zZ$8wYX<g_}$SW2JZM%i4Kd`z=3+qtGq%d~p|8>i4T)e)hEGY@s_gfwt)Eta_-&d}U
z1ZCl^`!TCT#<%(V_I73=0I6>HM$u;<Nw)Zn*_Mct$YFZxo8H@Gg_Y(B*y?L#<Tx!9
zylNT4x1mOu-FYfZAJS9v)GNuKx}XEM?<2E^W;?PP&ra;DA1d5OI+{w-nG{PJ40#D(
z9nbPhmiG>ivz{4?R`85$Z2>W^RoK`XWr12MT{t!^lt;D%b6LORg*TCc`=5#4!x3AR
z8J>I#$3t(Tv!3te$b;~>Grs#zS*a#!$Jr*FO!Bd~%NS}`&0Rh4nYK@!>qT>xFJtrJ
z;h{t!$_nJP{n#sQj0U)B!2<~-kHGs?2;h^E-Dx)5&bu)?OmT&3o9m@{iemt{03;K<
z+ckln%BYSzUGe9dZ7D;e?DA(;h=p~Uj@=EnfQHs&I`h6?sKtUJtYFN~AybnKHOtb-
zx`kzYzZcy+$d@QmE7@=(6%Y`31cXZ#RQ3{aBSz_fTh$G$(^AV3htsK9qu%v~;fEHJ
z6KOIyhKaI>6=9d?Md5}`GhI8&3a5K5jo{{+Gr^eDR042+sU;nEcf)`igE#tvH0_NW
z#x(gNL|4i)?f5I?g1qolb!R1SH(ZflLjS$w!I0#4f`bwin)AklX*O%b?NxF?qBC`J
z&+JFfq?R#U_c6RpRTM+NXy}4T@j~JcX`<6GSR=#w;%X_%@%(SuPaV3xf)aSU>|G7u
z`cP~q{`yBYstA`a2XPBJPbpx2lf=P3J2QRZ*18}6ODP?Qz2De}@%-Z8`mO5~q;sM`
zUio~eD0}-MxaIr_N>Zk7in8*)w+6|8gD?oZacb)81Lwn6Tx{#wy#jcn&peCOhiPt%
zd;<58a%GI>9JWszY!-@L<K5&26lrS)0#jg|;%_P^9cgkfDQlu4XUNQL#EvvwqaYVw
z9f!CXn$qPk;jC?K3pa7qk(r|Sc_XC&RwI17;id6?p*YdXeD~t{0a90AKlq|i%Za6p
zl#<Q-roDi}_14Uvef`K+^_9euaB3w-^@*~2Wp{NOO;MFNB1VCrx`R8;G!@XBGJK@N
zkw-i%B3}}gKB}9@7FIKR=kkd;*)~{P*?L1fG=FW+xiU6ho0cS}<~p3ZqoadNrECmy
z085pZ5_QZ<kPvViG~p7IqDEVq6cQO6Lo>@96|6-_LZ8k|d<4)QP3EM()ElSPSr9D+
z$l;MMv)48nYdEzuHcGXw6)ss9PBaodK4o#D{b-)3JzfHo>4~#PDMj0nWJ!Y(bn<-K
z$;rtq0(~?vY(lr_vW$^Efd5~|yT2wc)jwQp)h2O3$IDNewa-o%1+p8A+}sE%xWQDi
zxi^+>Q919>lCjt1WK(}EyalsxF`svm@XGJ!1yq^`#A5V7i{R#~I1A%;c~LFn_NvDc
zz7<IF#Sp&rO-f4Qk_d^jmk*01G6RpGy<Yn&j{>Z+)<(ipg-yf>)lJs*4c>9`$nNZK
z4iEB?%9V$)ZKfYCCc1cnEek6U=9DaiwAv7&^&2pEl+gQ<ydqLqp~k?^QeE?j<4`0u
z-M@4by)_ug<5M<|(qNbCE0*9+Llzi7RysN+Z#j;GGEh3yGu&apa_z=XNuSk(7Hb)O
zyhn~WD%+4xWJ|A-6FPm3Nh<?f=Te3<c1S3vaO4VS1|?$sX;5O@8=<Eb0b)%7YYT9%
z!;qXK!p~-7e+>x>khRNuJrKq)k4*`eUR-29X81gucWxpLwc2%#jVeSL`2YELwf-EW
zT_Ikg&IK3t8ub;0U-rB4SE_?wxu_#4P0jW-b+t_`EK)z|-7)+-tonxyVwnwjQ3=q4
z0a0UWYPsVe#t{}(k(YU}bgQgwmCLKNyv++6iIiGm0&~oVJt&kjgyi(l+&Q6;AZ>iS
z+UEseAYMWsM6<=08K_mW2o;yak>(<F^1iv&*+JpsA{WXAlzcL{Y}t{itQ&(kVW>)e
zHkwZ$L+Y8$;A5bN^Cj4jU?btX3<V);En&1X0l~>TXxv4MUF8(GQA^?#*buwS!k#*)
z&jK4<9+dS&^+)%1&=n^iTog`R<oZw%yreeSN~B+-OB+qop_J*dhZK<yvU75dOiweE
zl^p@Om0EgcfqEsI!pEPoBuNHvN&Hrcl?_S+{J_lZCG3PsTJ0tjz9CT6d^Lu=Vv#=a
z#4~Myj$T<OUZZJjV)7R4?*kh;4#oo$8xb&D*q0&=Huk{C6yRLBXHEz|M1nY}7ykx8
zL`3d}ly4V&o<0R)zDSc^V852zCZC6<*;90dPGIRv463zU;f+oO!tHj9FK)ZI&-YdT
z(tsbA10NMXN($T0YE5zBV>@_HThV1_Zhrw5CA_c8G!K%)EgUSUON)ys!~EKf?Eu=F
z*Rd-YEl!NjemXV-A8cWc!|8x_R+BSneaHAwo`qLj13`V(_^F_4%9TJxhn5-Qr+&rO
z7?dA=TotQsrHjX2{^T#N{M}s>F!#JDI#Q*NMmZ~(wNf{0t52UM1n@H!{$bqTJPqf`
z#;aPRR(}iBph`ra>P<2K^}0R;2U=Uc{(FjGl0E@V557#7w3U$$jw@*vE#~V-@~)ks
z;Z0mi97~kZib|ENG7du-I&sP)=rXUwGnLnv?BaBW6^Q4E=Tp;EnB=OxVQCzqu^;?1
z@OHS1ZDW2-v6s$CGDgKHEA5|2k{avJrajS2Ma2yI{6*iIadEZn=vT|>{egWxuS2HB
z_>pek6Yq^_3Crk*tW1M{`q^L-*cWuRKp|oDg!ws!WtmSAEry|*J3#&i69-mfejZqP
zXa_%~e9_z%`8W1&IH*Eif<g3N(y}fh(FzxNH$xU-lfFI@E2|%cn&a#>CUj1;;@Jiq
z9AnuAHcZqjj;?$(>>t>;hA@kpOhVRgG=~S}T;f{FH=8wsFU1n<4gD;F!l?x5dHDsq
zq#X^6qS{yoAL2!kJuUXMI_<=5j!z+HLk%%0sp;>-IeM@jSN@$lBIKp;&iDWLLN6GI
zl$QBOn`q$k(R#9y=K~=nr?7Y*^<{NQwpj*wEGZ({CcKPD$e9BofiLPLB~<GBL6|`p
z?Y+q?YITa`08e)5oQ#19*+jzBt2}Y3%~0Da>83dfYZuiu=*x#twq9)H%~an`4QAWW
zX7<=fBSrlFK!ZS(y>Bo(pXaQH17#)yCG2C^Eb}g**p>|I|Eh?Vzc(bw3suBJ-i%{S
zOij%+v$v0$Yhd0>)=1IOrTkKkDV4^i5vVGtDop{W<|?yEq3{RmgG>5HD(=h_!BidP
zP7C^iXS44Y;(E<aUo$;P;hxKd5}(?M8go_X@)@+OdQ2B|uC=b;Y3WENkJWnZaxiId
zFCSf1^Vs8QE<c0S^!W6&wyv&ct<^;f@JMPUlaQOcHHddIVb@AjrAFC{>!Hf~ijmag
zxaP7WQ;9aytdUGQ56Jqm0Bo+Ip)t6B!NkG>8bwEZ0N8_4!e9+dd{It&rIptITC%cd
z+BDWW2KhB3G3bqGSUfNow`+e)!z5ems+weYMJL7sdMXg}<<3T@J-r2|mn#{>tazvK
z7XvK)B}S`2zfWjG@k7*4o3`wGwj;#N*9fOMR4CPf_-1|JOIzF$x}Ys({Nxe&3Pg!Z
ztOEU~X%3mDwS{lB&h{$bdt|oS>Y5P?n9khy{B8oDh~-iPcb=>cIE!QOy4Y`AyoH#`
zAmdh_@R~#LZaOgXNK!ONkrpL45lLwM6jj5zl3k~J4>(ZRLJ~>I)tp9Cb0vq1^Koku
zhA$M@w#dopz85nyspzB)jJ!hVmBq#rry0$DPm_*@v^{4z3SpXv=T$Yg!56UkHtEir
z@AXY#(RwpVK{Ijj)mtez)oeGCEfTT~kb1VhaW>D>Gee`SyWD#r2=YK%K!4?hW8I|y
z$^55JB_B)G<s^T*K5+*|r<7<+)`$nOSIp34N~JrO!P80QLH|fi&&*SyGprSoVG(f-
z0w3GIq@`4NsesU~cMNR~ExpIqNe%^}TqW<~WleaZfckNFxIZ!mJM@!P0{-virO=_U
zLlz#HL56*3dM)lNHw;tFT1mt@%nQD}Q~G4&^2+iQ(N=WLNGj|A=5TQ^OHlo};&1UJ
zt$>%79(X(*TPNt4=U=AYu1Xq1W=F_&4v$=-f(@HWur}sT2$XrI)<#ZL5G)oV+jQG_
z*lWL3372qU0TC2pu>;E`3?^`zMKcC2W+4Jiw)WLc%1^f`&Ur({>(Y{O27?W0n52=?
zMyoFLI<|hlmK4FjN~nkvbKtykdH-2zGnVe0Wu~O-gT+eJMgk7<+r9lanx8+LnK(rS
z1xcEk?#E7(5p^iLQYwkK{UlPkuK$C~wQ^d%%rr797;5jjxK+40ybPlROMp@K>;&25
zPyz8yEiEl!qTM&J*tQ;!PpeTUx3|39Z7yOlCR*0dGeAfO&It%*0D@S)BzvHe1DoZ{
zi8JuZDnc&;BNS1h8pzK1=FLrrrY&nJL4GyZ(D<#fljTdt==$@?I#7!beC4H19w~^t
zJ}y7Ucgeh3y<VIcKT0k?t|1?PW<l9FcG{0xaZ5jXlVC3;lrpi;$ImaPrPEZ=n0XBW
z8yPzXivd?mbhIlF{S^u?z(1WnjT@bj7Irim8<8WkU;8(7jhF5kpbr0BV*!bXt3RD0
zKwK7hqQEx|Jh~+Ya;Gk_bw|@eDg_Vz*spu97*3sAFWuWYDZbV25miQQ0HPHr`mjec
z;pb=2eY6n9;eTkcWhR--9DgnHTlnV|`PtB@0IRo#TmyzeVNzH5KA)lVmHLL@uVWr{
z1YtR3f3>i?hL3U=brRyDuO$|J<BlDP-&;K}Sw$p!8~SBQ6{jdY1saqE;N#~(G||r_
z)7{o`UB9NU{zAtk_>jSvz__#2**fw^JgW0?^F}kJ?C(<4#JMB%!>d=L3#Rzrexkk-
z`g3LE{dDvN%mbmAh2wy2k!qK1o#VNdQB}zMEh;C53D#v2FagJ@wcSpSJf5sSe<_|f
zJ-7*>gRiT*cv(AIT892k$E!|=cOPg%mK<l*(TR$Ztc>)b{Ak-tB1AmGtIL185*LXu
zdi_I8P_=?RHFj`VURyJBbPN#B@uij2z9Bz51m6q7`=4v*WZsv;yK-^lZ!n{!9#38D
z>N-2KSxU=`GUtt&;y7)T_9rfRQuxv>E;Uy&msBEqA`+qRJh9d%2pVw`yu|zbr}TTL
zo+*Iy-eLdR=GsLudfd58l9xEgsE3~rqISbN6bAq!_Ot#IE+=j%!|4n3NX+9acc7k5
zg*&08C4LKOUDI=~ZS<dRpa(SQo_cHZ=vMb@6qE=cuCZ5;#HB?So)35^|I>Z4VqvMg
zvQSy$`JMijlISQPyh2B#BHjdjqVB>$9*@`fSCGjV5XH&M%UdAEy~>;=+grlxwWzQF
z@l~gqH!kzr4Re-QaL_=62Am9_uL<DakHFAJ58xxu=evDi9tXQJFwEsG9I+Flmn<Cp
zPfeAV=Nob1qpHPdj#vYKQ6QTV)%<ssa7uvtn>=6J{rj$6@!j_CypF4*V*)@{aYmc1
z*WA)|zlRNxBGYjGQb^-8-k;8^rir21?hi==Um2`DfG@*A4g%h6IOt5Jkj<|$MKhbg
z`=Zc81(c*9KmSgjz=cfZ9a`j~g9tfFLI?rloBYHQw8977vRn}mq2C`xwA$g7zP9!m
zh#;oPyJk{)FP<~ujajzM9)5Yhk;DSKO<tZve`c0a3e956L9loYM<t-w%H}k6brpJk
zECvkK^QPS#DL1hejxpcF#&ILL&fb}f<OK<YQRS<FsFDNfqJXpWK)YFH3+m=qp9XO!
zS)QycEHDF?jq{6Hm}{yV#40stuHpJC1;h`mb7jp`wXe|08^X-Ym}3Ijd|5Hm49diA
zE)xxk<bO~NyNg=5idWi)Osy%O3547l#5k064R0|;a0*Wdcl43@?BjZ$t$otUV7#~%
zuycG`MmyE*{I&h2iEalr5gF8N+=O;xMMOCE%91X4+euUj{aE?4g(O=vm#Zpb;@l`j
z9P;faiTjf<;V+OaL-M@+<tdbWFegc(&p$QQZGN7o#q|(igmpCdwWfx!9bd2t&&Rje
zMBhbk*R(Gr)^PswmeJO9G|d&U#zM@cGqttF^Yyqn{ktX43iKU=@M5{ob2|=TRgC~p
zd%K4Vui$EtF2mL<=nB(_CLl(inp|GajN&VZ*KJB*<2t|pxP7<HGwjb#QL82JnuqJ0
zt}(}G^yF>Fn+*@XuJ?=v&+)`%d?G(ISU#}eiKS!jne%d8%Nm5rs1K{+a1f)v#+a@7
zY4s>9_wMxkXKmLw8H%J3f-}w39TpH|5|fgEnvhZ8e4{O*1!ri+vOz891O^9Ntx&Xq
zqi1wfFv%fFt;GM*a|UC1w1gc>2xYO*CVy$HFA3vjB;_mw!9@#`M{z?qQYb@UU-#;}
zA_Ioe_ZM5l)s>T6f!a?k6S1@U`^-<ec~5#(?+nb$NN~Wb7=TUmzO2b3U{V6X&m$l=
z{O;7HbL6JuMhJL$Z^67Z0G!?9QAF>Wza6Tj$zOmn6!P%g*&%T&MSxTSUAAo#mN-zb
zfIuO@4PZ)QWk&OL!FcWqtZ!ERyRZ5H83L3}W~K>nnE}5N`@46^dq2H!2+W{kJSsq4
zs>y0klC()j?b~ol@bxCOL57r&p8H_|DB6Q|V@b}Nvjm91OR72Gh!43>oYUW}dtN9+
zb~6>H6}k&ygbJcIkKlQ*j)*EB5vAw7STbzLXgJ2NG@V3T#1g!m=4u+F=Pd@tv0Q8-
zO;&T_>_b9;TF~CXA>O|Y_(gCe&1QdP?Z-{*muQrM??0cZ!Rmu6sf_c(Y^LK(<xr`8
z-SJvLmc5=ESHJD#`(P>woOIW(DGF$m&`u?KzwPaXt9~`A&IrF6MWBYSfoy(a9IdcZ
z(fKTX=O2nlVQ3~#n#apoZPYLMBaU*Z3%);Z{^HCxV_qT?Z|$Dq$sodWG0)Ro^xKn>
zQdRp{wHHQ(_CdA$+=XV=7iz*l)ISex=RVJ{WRH8?jUw$z+xe0ex~3f(y9N6laOR%M
zljg)1{Y0InwvAFXDiY?h#0>74(EMsUPsHL@N--D~+)7%yDMf3xGe=hz4a++YlHRvH
zBHJQQS3U+-7q3S~jz(t{h{2V^XS#1nb4<{6Zp&|mI1u4ee$<r8w|b105PZ8|MzRVi
zin{1gkn)Jymkmq=Id&^2=eJ2CvaW&X%EE3*mK6%I7pluy<y=o<F~5FHfEih#M3?#u
z*5OVt+uI>Zz{NR~G70(a{M$^J-@nVB3L;)wb3gLpO;ES$V-XL<UN<R))(EBU-wky#
zr!v1TqL5`zHnTBF0&8(iTdla0bFkskP87UazbzLbfz#4t7Td?}>9^5HmK?J6zaY+@
zl0RACC_lF7d^`h3r>D0N$R8)Ixqz}JRVBF@X1-dCG)O$mFh_#L?6rsdrc`G(Y4VWw
zcay`ycBK!p3XB-7NJ0uVjUttq>1D{@Z1!}yBUT!%*9t2rQc$_oSdwRWwY5H4;|&Uq
zra3IwH?0&*nX}=es>lJo>5Hcuz*8jy);N;jhIz4pqlbWd#QT0G`!6tie+O++8Sd>_
zrLr~-?jT+P+&a(qN8lj?S_tUr!Kd$TFlvRy;D9O>y$5ONbCy7gb-Rh-GZ^E2N!8fc
z2nI%g#{`3W1r#4kh+mgS3RD@-X=%A#|NBC)uLyPY!pm9n0CkYy-A)3S?h*^?9Vh!?
z9s#@efvA@p@fQ51#pA}d6giVenY3Vh>FWDO0#gjFQ<q9)?h%>Y3pcPOfn)s|WK10D
z);RV=+c0d$yQP^Nls^s)2bR0}t)Q<=fVxDBADbZhtIonW!vt*&mDY%WmR4=ggwxab
zfpeuuYaq^I!Fa!91)!>8S8?5$kxDT+`~ghgF~fW5mNjo=?1<Oa9TB_~TxkD5Zx%Ev
zvDjFpLQh-(r`eytJ@__HQ8%Z!-Dh9#yq>cke$tU4>>Z8QYCQRwA0_?7Duwl*o|z7f
zg+HM_eCDTpxLo#-2qCop8DV|8F~Tw(gI|g-s~{_nw{KO>k~A2@4|ihZ@BC_Q%DUVJ
z7*f0<*dkNXE9|d^jaFMcVt$eD1YoS^s0^O}mH%Sk(TZ|=+4_F@if7IN_ltWm9k_A|
z_&KFP*YwT~&LRAc!&}P+EpTLfCVu~>cz9^Ne6Zm{7KsHpCbFmyg-2?m-uif!=3b-4
zQgrx!dhx?LqUz6Fl_YS{n0(5jo^gC7iQN~YPz@_2-;=-d3iFwIJRnqc+%t@6z!DW}
zSxDBz7R?;0Fn?TJ>6EVfSft5BUB2ME-hPE{Wi_&MK`Jg&<TWpB?=GZ}uHJo4Q_?u?
zti0N-iy_;O4GR^-M%u<Az&`{NRe=jc^?D6*!pZVmqi;7xZU)~V%h2NoVO2N><mDMT
zw2;fne^R>FqM*7M^%TzTCrp^`*|Tln3z{MBbST%p6vUvX?lZ)i7|AoVn6I6sKQcmF
zU9*1;;w5mA{9!t>i$o5vB*S^bRj1_%UFW5GKPD`FxOW>}a~sWC$;MWY`2Oo3OaXXF
z3kakKpH^Re{V04b0lAjOi#2^m@X&q)XCU(K;<pia$iV~Dy0&+D`K{C@$151+o34jw
z)8diM=^~)K+V#Lk0qziDHvv;o5KXG3tu0QUMA(9X06V|1aJRhBA%!0dxWvD}Nb7iW
z`WcMSWS(cxU@3aB!2-c(mv0;h{b|vak^^sa;12@YI&1=$_)-Y4?uk5ZV!TBE0De%f
zDrBDD#BkNy0fS&>xc+^hK=4e*#Bg)@)6N~o=~Pe#iIuN&>lk3;1_lORJQ07&s-`U)
zc)_OgF&xku;Gh<$0irzFv@W%<oe)?Mfi0@Myj%eeW_<pl&0B7_#nA?jT<OEU60DC_
z<0!LpBxoxC`TNAZmf?Ude7f-!@aD%eZbxQRU-Z?c@p#61b&NbCQK;n(Rki^8g!_@{
z3EY}}%{Cke+XZ2KP?H><T6`u)2Kz-PM4u8INbO^V1~bjF{#NhC`F?&lCHZb@ishTl
z_`@Gc8QBmwuTSgEmnSH(blk;5Gi@qG8fHqgJMk>tlHOxe1(1U$))eyxEiUqx?p{Cj
z!yCQYuzo~Xe;MJ)aQ{G6;%UBSY2D7ASo921l?(B<c$&X(>@bIJx3{CL%(ZQT%G|G%
z$3HfI8h2dAuWoAZ>G#ome;cn+H0JolJ&l)e%Jzzvmv`rl$OBS?L(<q-gzPcp(BTw)
zkf&czftw)-c64yPj$-bMFb|O)w$X02`$niz&&$6PgbEqh*@e7YcMkyifKDSDuT;}I
zb}UzQD=(JffSPI5K{&VkZ0h5rEaUstJ}V_0n%zasjisYPq5UZ>9QwBm8Zz{V?~QsK
zuZD5GpO%kSTO31PV49nCpBS~0-sfj5ZCYpXOuQi7ESKMRBtF+Yoj(?c*a)MWTf8DX
zZk%0@YtY8s&wgJ2*mjmT5I?}5u;L5}Qqcd}-MEBs#4f0$UOi6vrk^io#$x8ki69yw
zQBE?1p3M|REPg@rmb8VsxK?!IRmb`3S3fiHuI}j<D!t@{vj;A~p%R8R^5;cA7lOTv
z=rg3em6oti;YiP-MWnj0qi^F{4Zc)(q``jM5#A+zPpXPT30E?Fc07db$tqo=nWIU^
ze$Wt_n?k3QA;e{@)%vhBWt?r_xxNI^-Mo1~6l(%T?w46A*t(kN{k-sL8@6j>hKu%u
zx9z*~xw(%31p&6}k9CTSEX5<XY^SD255jp<bSAW5sQ@loY$bB%8B~AYzCKC7+D(^Z
z0;VNr$=YFmoaFgW)iW}E)Xh<H$K~5CUZnu?f5&Dbd1@uqz?BCe2I!vY-HSUMx?ls~
z7N<mg(M`sw)wZ`s?Yjw{_hLVetv{lHY4+t{JZ_6TQ-PJ8i!dt8d-C4e;ojP3#hXt#
z>g2n{aTbcoYRA)ioRN90_mA%F<6!V~J84)@BX~y`MVBP^673p-X!YuDTmW)r=5uXU
z!r66w%>$T9U@x?Qv0k;)Q~E$_s+IfTCvhWCrrqS6j3;PfL&O|gC%Oj1I<ma<egM?r
zs5zHwG8OvHU29>&CCU}dS}k6wB&ii<FY)6O92Rg?g1>YM%cbM{k(Ffc{Ah!|Bc3B#
zwIHyKb)<Jh+Blx@d4I_Nf(u`(|F6La6?E&<goufOZPx48ru%39P?K~d{wgv)9K(Oz
z9FGLo?Bx=S#1mv})N6+mLdCPTtfPXWxbFvw>LDlC(#Y9$q+AHZ$g}~0QsKQiJSMpC
z+R||Rh4@!mQ35!@?2|_Mlz(%i$vt6U>Z=a=2&>eLO$&QH$BSs>Jx#2_r1oPy7&fk>
z(mih`o>NX%-l_HiwBW}q>0cR4DJt*3b=hZR|9WClvcK%yUZ&6yS9i9Ckh$WFyuE*7
z@n(sUay+@1aDzNy@CozhYL}?8ECBBk;^xKW%mi3(?Y`ce+5j&p!m82_Ssruk&J;;&
z^(<|xd^CJ!uA7<BqKNy9nInmHA83_VMgewtukpZ!8Upe9*UH2?0p@k|^z<Vj?6h}v
z?dtAEjgpFRXz4QN6)j2W=?S17w2$`GV?TCrg{-->xZ0D@uX)Sv`Sn?%SEti-p6Jfc
zk1K6!TGQOT&&KjJ;N3Z!Ty)li#)9)!>m(M7rA>B?d+1PR$)<X<144C1$0p(?l4Hel
zm%~9w8+h_QGz2TwpSqFuY_b`QO8zvUbw)C`F`QZ=ue<x<TNQ8_NC=l-+UHObxW9{i
z?aah<vL`}+E6RxY2>xv<<h)CXvXA=rN6=6<Y2=>xQ17B@9^T^~LI13+<-p=w$GyZL
zay%6zNdD-ibv@3{rYQ>W6QA{m74~|Enco9=P0<?f4VPeBOHu>;ll49@B&mJW0&!Tw
zNsJ|mt1nR~2_kpcz>yy`pEfr)r#|yq^2OSy{C`atSapB~<!!X$>4#PV*pjsKmk_P?
zo8Q~Fw<&T3;st85@79X4+{3}LlK==NrQY|BODPxqtMCk3PB|o<7M=E0sm>6fBZ<IG
zW{y{L0I`%E|DIk5<X(V&01QI6`Kup**T7x+Vczkj7JKjPj16rkYhwTK=HY+ie4DHN
zA8psfpTeUJAaF_a54axl@~H9?5B&z`!9G}E)xZ2*J6*uTw^np!7<4(k1dIXw7dS;+
z_;n}mJT@Fsby1$dO$==GCnz5JH4an53LZj{enlhCEr0vwS4!<a3v@DJ3PLvc(GZmq
z>}S*`zZiE2y{>M(St?2_tFwl4W4MFStMOHna$*SN)4Ek+%DPbI+a6p0tZoErfnC4&
z=5+;E-ePYqV`2nTd6jt4h;!Lk@o#S9uyz|5OZ8!bw_;%lMMF3Xhk;(<T3@tw7+Jl;
z21bP8*xgG2Bl)@)#?JgWC0LE^lFRQ>Q%jFVm|vfQ&feZMc%Ky^F_)NuShXs<w!TfK
zX~opaD)Z?{$iGV(`Jle_5xnvCV|>ou4;OG`K8Jc&4g4Nw87UhXBp0Qk|8AC&oxR_F
zBfK-CTy;L1pgd*kLV!lKU9rd+jT^A(dv`}zpcdy`W)}ZbWCxB-RLPeOI^!)3<3Rlx
zm#G6^0;j5cnteKp1Q0wWGlVt3P>*RO089edBeB6tbLB;^3sPMc_;?r1OiT#sP1%T5
z2Eh-_W5*u`+%5gdSzd8scD|7PW~U<->5zf)a+bm_C1Z2Frnke;g{J5u=iKhZ$7x7V
z0*9Ig24v=m$QF%}_wtNF1}L<FLj$D^qB62MnjqWZ>+bMN5G4Ua17RNBPQBj#!60HA
zm=J$LK(*=}J1j368?{p{?a`)w;64%ei9RoRIPQrv#u+}k5lp9fzw)BW+r2rc)*8!C
zv|2pwnRozjM=(zQD}`HD&nZo_AMBMa<Yy{2>B?fce_x}RlbQV8#(C!GNq85?WkGIn
z;W)b11IwF5A>`(On#ek9zxEJ#ZPDRGGz3_X`k1M-zxuH3@w2M2`A;8Ub2@-Vmh%NV
zIIoAc{hUt8W`Au>SOc=Lf!qWr$z^94GPAO3>+2-|VE+=VV5=qvB$O{*+PB~mG6L{N
znvo-z+khefhgyd=>Lu1qS@<pxoG}Kc*VQ%SP~brZVB!DXr1{?9d$zY$b#OBztc5mR
zS^@*spw74tz7WJafW!E=7hCy-fISCJit`FPF92gdab2arGd_7|lOw)Z3nv6GP@lZX
z)^A!s|K>||d}U=;>KlyGI!0DqHm(zw8GHe@rv;Fq3<6j?ZL+oXO?*Bv;n63Lyf}T$
zvfP3J?#PjhH>>6RB3A>ya+HK1D#ocQN3zz6sGy@ot|aVCU`epmS*bbXlZ)avAoIk)
z0H`K7oJcyI7x&kK`aDs%->S0l<aWJjp!bu2zBe=Tv+e#mg~$iW;$8Y=p)+jGuc}?i
z$Y1Vnr+eNAAK>S_?i4dnNviqANz^Qdwm5*(+=%o?NQHxsz&t9PYtVOkRq#H3>mYx~
z9>Q;!PtHZVfEk4jSAz)$1M@bnn<|nD_B2px2;WnYDP(7aw95NJUhsSe?*NNGevn+5
z{Y`8S)%<&eHQ^dsl{iu8B=My5F`I8%xw&}Bcn-v;r0o3k7pLi++aq5*lEgQ1>bDNf
zSRCCwhl{t_@##)~xAh#^J!gFCc*YIRQRYqjM#QW%rEY53YXG1I1LsQd0&NL*tTpe1
zfV|Ie-VWW@h>yM_$)tHa9}IzXeBsMKk~mTVyxq~_D8FC=7<lnG_XwICwLmr3>D8_=
zMyu_{zTrjAQuVKcNU=r~YxqJ!!=oIf%Jrg1XP&b8M)`QUt<IT{e^WOliDenY3l<qX
zuE$$O$B>hnmd0OE=_?@UC12c7+!+IMwt6wKxeJr%?_4x8rsJZesVA{4+9~^)d&rYB
zMv9iHMujlGV=6k0e`+oF*Ws&`Mf-y{0e5iTD*84jFpiERLonQafzkmKJvpjxzdgWe
zSMaPU#dzBVMth*%Q`i?1puJBU;};MBM2}L5h8zfY&7BI})DI2Ep^TOBq&AxGLllnG
z=88P`!l2OSBAhnQnl$G@>`D3cRR)J%kagnkG~tMR*$zrUehEUerPqALdTkG(Sm(m-
z*;kBj5+yj*jkST(^Qt{CJ$0oTpQ;I=Qmb<hab8E(eKNPMcD%5Bt|uwVUr)L>OeI=%
zaCIG&Ci9Z`UkI<vrd|0t*UE0c)S?JlQ>T`pnu+d%TtWF7$R@J%C}>dnsCw;k+JExk
z{l~;I+p4olox$_O>dST*Z5EUaBw->Q)~_L_&M9U)1<dRqlnj7zB?DGKSu|&P-n#x2
z{ppJrH+<X6xdLun`ULx<dFJ1)+flQm%7}4=@^yMO^GH3lv^+_vCsbwOfDay+j>QH}
ztNvb7551wZnC8(;0Sp_TRxfAL)<1d|4|oE!OxPt2d!RSY^WVdIkuG@FUc8M+$t<4@
zLB0n-Ol?=2{U8sAV&>qF`&k|=S8z5>D}c0ObZV=a;KOijzHnOJpYaVuiC=q0^7jS*
z1-NTx8nsME_!%!BytFGc2*rnBSkg1|L7X(eG(qhE;j{@CVS(dtsoZBM%k`K%*xo}=
zTb`GMLa98}QpxI9cvhyd@P*)T2LzxDy8mA}=gXG}_REaakfl4v-#Hs?=hZ)0h}`>Z
zj`9-0nKtCzxxHp~sKb@J;PyB4gNj*X%sr2t9!SvpOH2E0Y7ZfdpK*!z8==$hEN^V+
zBJP?=8gh0*!>DN7>51(m<efMAd()EcMYO1=Br1uoNu<ql!o2*xr{H@aZ_!bnc?D`n
zFu-Bune=WgqgHud@XfO_<ve0_6Ml>q54=QlLhTTJ74M2eIR!^4>IywP!U^R7?^jIb
zXwNgz&`+Ui66f&_V!u0tvif@+0m<s9OxMRJ+5Z#eHKp+LrO;#TUMeH&YxMUBPgrZF
z5uj0lm($qP`MY6F@9scT=*O6ZZT?K91kcLezKEEb+O|PpVIRv}$z~V%(uZ-MScd!a
zNtoY&wa$^&9$IE2WQbz8xphXEb}3TwuU*0GV5R%mKnX0Zg9H7RS%eeSJ@tDZ=rAC9
z=mW?~T8c)QV*Bqc@c!Oy>h#J4X+Y;=RXzdf+3OqK^YA>MLWr}oHsFN=RAZE?#9C*%
zE`;1`#D{&Yz~8s*XXmIh(PD5X@x;C>dF1u8o{qV0SL)zU64$o4gxh|Xnaa~h1Y$5@
z!~0<R{Lk@cWueyTII0YArvH39xC%zx+?LG-6JCq1$s`%!o2MRFxNy&LTEtPp52^~p
zCDd1$GJ>;f)zqe^A*!K6(d8hc+w*a00FYSRxpe&r!g8rgn;yVmKwxUNm!R9&Um|Xi
zgrKh2u9qO3kQFi?KJ`_G#jKV%&>#|1SbIg`O~Q}yj%+5p*uyZ93g#5fbEuCQz9O{7
zkp`|Ow~x;R47qe$zuP^4RB2YgF}KiR+VN4$nblUOh#+^x(60`j2dI**t*wFjUu|+5
z;fvFl))sFTB9dWt9dL0ibX;@uyn8pZzJxY<(RQg_rUpwxxdj}-0JdMVtpfou0N2rE
zdKpeD_1cGlvn5uowxi=E;4hwolg#Il9RN5le*kTeKq5iV35q>{{SoLlMv__c61TuC
z3;-7}Qg8J|fJ1Nw&`d{g@Ly)yxzAM#9zTB0J_#Rr#%Bgp60(+MwXX9%3-?8*fBv9-
z#6=t#XWx%<N3?#60A#K&_8c%p0J0g$pWMV@<%9=R+dyvO)c)S>XJ0&COcZ4e*=9lp
z1d%DGg$MQ0g(Qr3Ga8iZ*nc0_G1JwE(xblr-;p75Vnww*Y{47f)6;Y5J<rv<TxWtI
z&5~}(DZh9RVxENWcCbNTg%W-+_PAJwoiPY7wJwEeJc-$MrF<W0(<!77C6&j8o@cUh
z0@1Bb|GY5M<@7yhgu}_1r1i#A{p*cP7y2RA6~eK%bHGv%2&#;7s|ut;GFBnSLp~vR
z|5Y$SW7O~+l58lBz9FY#O8zLQEK$X>RKyPow71TEd?3{;=^7n=E$AH=AeKF)8j*X6
zlqfERY#fOP7S+exfv>Zi>Gn_!ORrfycjX0drW0^VN#)R?;g=P@1EzB06<k(~g0+h0
z1a&QX2vt4~6BF~4FrT4;x8HBN+~35ip;Y%*ADr!e?*ZZL#?rptUO>$6BF=+pw3S5I
zI;62NW;S98Vq(&|w)$Q=Vm4wl{64IwsZjBn-mO1`7fG*`Meg8S<UJyTVJ(xu^J{5s
zU0&1N-1Y2DS`js{(vz?2u#|tiGozhXFGUC6-EShV)UXV7emvwdy81`~mt_)Sn4mrW
zPW4p)c}z4l7=7E$wj4o%-YX=@brWxcXz$K}m%w9IpC`tAh@VE_X6Vd4lKh3vWjp5X
ztK-#`-Ltb4`8N!v&svxvN@o?Mb?TJoj5Ua?6o1qx)lz;6E5uDBoL!%_^*$$d@jg+b
ze#ti+ZeUO?jtnU=Q(J=fQ?6Rr0=jFdysTP$u?_4UfAz1?%BlQoi347ipHCHwlX;_u
z;Ne+5yJxg4@%fWyoA#^Jm=^Mk1fe7Y#pT;<Rv3ReWTd=MabZYrDThK~EKCK?rg`ts
zxp0tZZX$F!Opl6E<X=A8d5d%ifet@+s>l2@%SiQOha_Uc_#xnhfubY*3i3ejXtv--
zgabSh3M7#yuKm{o3FQUp=tCTqfF6GdUt&WAyg2;`Yyx!X=YA!*A>apN?R?|_Y=q#h
z17A-xA%WS+kMgD;4GDFPE_j2?OiXNY4$!dl0++KNl+VFwxY!IKE2Llm8HGTPD<U%S
z)g1jT__q%s6|_ZTjclBpJwWdv3(}TA#9R95{XIX}3xQaR@GjeW5`(bm0ou$($z(rD
z0Po+0@hWvn?Gab<2-W~QVDQ>5r{!Nl4@(FgyuGtP1UU+8TNe;<m4#{g06itZqrkNX
za^S&*2`1VW(2@b04e1cUh!wQ9xA|RLGH`rjPh4~iSwvV|yuNmCm+0pvCNQ(!(k0km
z?g6p0_diF!!<&O!T&{iP^Gb_$MSl>=o9T@O@zB}L?(r}0+LJ$T8{j*9`<58JP=EVv
zsUw3Xc|1Mg^ufaMRCRw7Ks>zt#dJiU(S?6XwWv_N!J#j&ct2OnXN2k)EM2M{5XddZ
zqyDZS$>J8DcM%;|e8Z#PXwrRo)1$Y88kzd;jv*q(hEmGw^ji4+m0^DPd4X)K-y3Xr
z%n`)DSU)t^G`Hufd{X5<vBWaU2HU^cOkF-y*_Myd4C*xqWTm!OgODTQqe-KNz&dz$
zZB4&ExkqZz&6?|btF-LFwCmXzgF&aB-0XdV&g!%OHp}Z=lK%<oLExW2MFM{^tH}5a
ztMGd!IB998d<dsNr_A;xB5p_|0+lmlRZ^0@jOb`oqYN(Dftc(aD5PT9pD~|paeBZD
zG7Q@^A3Q%tsxW1QCk?~HeB&w8-!*QcqDur@fa4IUH)^Z^6^(m616MRJRY0#>eG$Z?
znMaC{XA4jU12>bnJz0#{9S1kSS2cSbA-J6W-ug2~>k2u!0U*Ay)ddU)rGXPePu#oQ
z2~6r^7R8pvlCyAjac*gLV~4KX!-)GTsZF2~Atiq_XSr8&=pVRJ=L?}&F$>h$YGxze
zki{k)RPK(Gp->qP<UJRZ1=S@Tc8k`yQ<mHSmDguG_%QfMl|h`Ij14*>&s-u=BI8|z
zw_1w1jDuW1P6Rv%phdZeu(s9HH}Xo*t@Op@Fu>q0Fbo>2h$yf6@<SKYVZDab(ZTc5
zL<B+G^!Er#m8E31Hkmg>W8q&9B=;9OmCuWekSZ)6D+Fut8aZ-EYMKV^kQu@@GGqdK
z%hl1s2U`EWzCI+W?%}eKn!38H5tvsl0Zm2+?wIqI+~7kA-A;)SdFF*budEATz82<T
z8rmI6ri~mwxvEsmX0!wJK&kkJS9U=`0aw%<ARfSu3EHc`#i?EK;%4r+Zm4=Wl7PaU
zRv`wVzI!eES+J3x$`j)OZ~N!UJuUI+bgdPHh=RgH?TWvPMlmnk_A(a_AsWD*B*7sC
zGv}{b$bh~=-C-8Z_PCMCiIi7m<c!(J{Z3_7IRGLLfRy<wcM>*oJD_s`4Ofl;27bA)
zTOl1lL;YGuM+bn-1sl&G<Q)ix^{t5-fhrSZ%+KDph|c-rE^u|}T@?3?SMWwhVJ91*
zBu8Bx6z<e{0csLt;V57roc>~t-VVdjU#h7-sfUiTIWs-_2IS&#!9KOAd_dLnhDBHA
zf&#W=cmXgmVL@JvicMy}^m%O8dxl!h1s(h{1m2x^)ox$Vy^tRHzhZFS!sG^EM~k#a
z@9(E9u=hndZTZKjQE2z(>!CE-YxMYEq!BJ<qe)l>WhGLEjTMGAP>N%@6Kb)j{niXN
zm+73@zMQf7%MO=PUL(#xN#;*HD+izD`iEhn{fu~SiGv+B80*$e)@2f$8fa)gGL;JT
zgByjNOT!#}1agi>e~#3AF-*4B>cOF3q{)qq_>%FPnfzD9(q5qdaanM{<Be8Zbr-g>
z&`<%Rj_2Z23T>8qMi&aN5Z~_EZRk^vm!8+|$S5<*w4jjR*WMp1yI);ieKEHjDN`SL
z+if5tNT?>lw?p>)ypWMaT-R3Hb+Y0?Ohz&Txr(D^kT9?hfu}UHAFRg?RvX@d#^V?o
zK+%qK^-n{}C<tLLme1PrZ9T6eD=@DBVkeM&Lb9F=CMPv!^w91Uy{3unPo!pcH|Uwp
z5xOit)XF9y7`M~OS$TR@+xu`G5Ue1o;XarF`!!dQ0sp5pq55c1nZ#J6Pqb6r#gA$Y
z*8I}U%~b9E<C<Q21Evu&aEZvOd2E}9Y>}Dp2wCLOhLiD{a@2(M)L`~V$fF`Gq8;oF
z{I$rWg)8Y0{}GEY?{B>&&yfOF5XG1&cEJ|4e3&xDr(E6QTHa2H5>1Yh6T_%*?mKb6
zCCjMPWmMq*g%M*tzzR0_vqlq0Uaycd543ajgcJqTWIN~B1GGA1$tQx@73m%H*)D&b
zq6+H#|K2dOn3pXa0phMmtSkBZcN)z)3;nIOEmGaJiB;c*rLhtJVcuv^ZvgP?2h?=^
zOu^DQ1^@$5zjRhLXfvDqWMVXgF8e6Gbc~5tu3aU+asz(S-~t3THt>*=hf@Mm7LY3*
zLm(emlEK4mY-1DkGQbXy`D6gXDVsLS6=wa0Pgr>G;2;#huX;e^4-y3c)C=x#BqK93
zS`!4uBst3%Cr~i-(#-I(b^=@#h;Uw1{FaL~Xoq(US~nNK_v#~pR<D?3Pat~tKulgg
z4=Pn^Iy#~u4<L;J^qT^^06=|SPAMo50;(8Y7U>3U6E^&_0#>h~!R9~`l%C1_b9rBH
z;sGEY$S7HP1D!MD>)2@gKtVLS;`Ht7_E<}#9-z9xkPRdXzwLO5DFbD+ep<U?F>dQ(
z!%%JGfW|xeRy`1dlbx+>^t|-mP&m6tqpZNkS$Mf&t2P&pt<~*&1@UM<fBlp5@6|zn
zeo*_m_E+m+$)bF{CsEqyII(}mzs=t>{}2MHdn5C0%7`Aa$vnC#Z$tb>P>k4$3%SCu
z{_HH!o*|Y48gBWgSA@iYYoLKO2dIRSVZ{DU6|Oe;V-mGH7DS*ij&sE5PZDrg-W&oh
zJ``76+MJS7-^M7a%5#^mv5_a#sh928w<t7({d|(mug?>j)!&>~oGx>RCJFUbu*ncS
zbc-4imNW=EU<)r9AD`mo&sVLgYiJu~yDZ86ZD{w8^w$=Lw9BdevT%Wr`S~i!7sJUz
z6{dTxbTsrEjDva|II-w=zg+_$-uo+~<E)dmvtd1<VnrjrGfBjmwboXR8`Wsk(h|1c
z6OMhTTyGw4g|sV{8@aEcx3rn5FT~M=TYED0RHxk_ur1y=$1f(Nx27T)WW1&B=B+xA
zZS1w4_raeF#OW{oVwDvA&*yo<r{~WfWV`k&ysS&U10E36Q?z7lzd6ltfl?QF?*)X_
z4VFoO0@3q4XJlfnn6@Orb`{Q=#gV_GPDJt5_i=f}ZLU(DNMI56ozIX<4u4Mx<xG$)
zQJk*Tq8JX!RO#<t08&u-L;FhSq4<IkHtdQAj0%H-H?69DM*kb@S>?r*cKM>Ud=<8I
zfgw|2LbL;nNV#Q+T(Xt9r7bs7RpKH(0wD+o*bM;y*STUgNv1;oF;+Y;^o;s|H+R7X
zJNjNi6^mX-${<^>O1hQvw`h5jQPo1x!qKz^H&(Cuh#NyhUEvyo9C7|E0hFMkuNJGO
zz`$x$j~67XurB?kr`GI-(K)aFhF@4%DHY*pQ9s)RZJBv{d--3prS<~Ngsns6MIYDP
z##dg2cg4VCtuE9Ag6>rp=)yrY*3L_U8#u0@!T#6!Xo4!?@j@+yMuBA}El3TOcqm^U
zD!k?g4dQxLz<Dc&ADoeuRao;3F41J;;KRQ!ckPKYJq|McZQaHWh&J5;N*-szj-I`<
zHlRY2qX)r%#dDVQ*mLGw&E8M^kd6lyIeJhr@u%G2V?U_vbq<;Nt(;(g+BOXHM<oA#
zPIzX1?Z)f*0{m3L(LLWww*wy`5OD&>B=CBI?^EB_Hu{%->@sy0$gP@|XDt}Dz=T7j
zn8xI|<^wMVR0LQMdLOu!p;OMh$E6={!dMp)%e3*8z<j}U^mC6WVaQ}9_DU0&o-;E^
zz?K>$PVqEedR&oFqr|hsUi0_h>?{_ve(k$~Hwc!bYTBztwUP`2I)Th+_6j-8cdbgg
zvDP~}tWR}%4fA;`LdM3Ks)ZyH6Epv@bk^Z?|Ns9V!wlOnIi?w=yX%eVobH~O?(Ucz
z#<WdMch@jX9({CkW=EfCj^ESg`}=R#<+=`MuXsM<e!tz>TIxo|g)aimuK}Lh+M*=>
zN#+|g6pnpQ!R0PVl3-%*(P%|AHoOnYDstvbRHWM^{In4TbD1@3brsQavmsn3?`PDN
zw$Zk}<VKCcPKf=9s-)wtAAdJaKmL={5uIUF=6crursaCZA{C1wtVa3US8g*=_qd5q
z9mM%Oav!BCK3K)kHEL|T32t(5r7gwe$I}+fsYz|_xQ?>1O$ja9SouT@&ZS4w%-^k)
zEZ^-1oXR9>R$kWMU#1)CudQ!+tro30O*%I1HIiLQX^#4N=R%f%L$3Sg-0L_zDuJgj
zFg>sY0<JnHlb*32bvpK+^LE4e#h)YcWwTh--6tR-=a^#09AymwP{jYvqUS$c9Bc4i
z>doYx5g-bGsp2RXgxivqTTqI$vW#*v|J?hVP^zgwuYMb`o3(Z;1WptY5PkJcmu2ec
z+1^EzLxs+s`zQl=g%9Ky@b8@qjgHRDFNx-BR2|P-q_+lkzQZMv;L~I!6Rw`$=BuKd
zZABoRd`?T>FloUsMP;^kVj`sFL;lg#e!?r209^d7et40c%2HRwBl70iU>-8#e#|}x
zagzLsR2nSayI(5KVAfVp@Lwtx9+%Bbl>`B>$<d}xjY3%IdXu&VEp~@HpfM9%PpcJ+
z@Xajw4z%m#Ti=6tL9-9R!ATtC9(xUcJtrPWhKHSgOa<l<rovAZC*e(d7i}!;tjEjT
zs<gz`_`<UIY9LLhg-k-|NyAy(Yp1H|ecd{!5)#>SrAb*d@A4W41Dv$NvsClp2NGB_
z4zKHi|8@9T-ybAojb!UZngVefpdbP{9SIHw(9MZJEX>S8*aP;!=HS;$V1Wi%EW<zb
z3LK$)V~z=Y+hOrzRyPn|T>S-Li3UMI>!2V7&Tue-irM;VT|Wdgs*D6!Af5u*7j*do
zICoBg#}tftrU~?_qTnFxs?Vznk>fsH7q<asQMghyb#+o;RJC>lfWu$vlB&CrLJzq}
znF{4g2gQg!tF_`%dB|^pcf0~)@|z)ChuQJhfn4Kakc7VGt}_s(b0TxN4Cq&=RG=2t
zc#=FerVJ|XFQ{F>nGKK{54Ad&FEf?DuhgEdP52(YwPgVeDi*oEf0@7oJtOEv@qtiE
z34Lh#@bg^`w(NNOc-0HxwO$U(prE`5{&|)DPe-kDZKE8I{F;k;$n!o9bW_m3TP*>n
zZ06OkOuG5Lcyv2ck&UC(&aB<i#*6xIUSM?wwopYypgTO*1A@eoCQgIy{cfYV%RuIN
zTP}PGW!jSu#<jik9kynuU%yiyM(i7P%T}$^#`=7+e~T6&5)!E<mr_b)MwSW-QT}G@
zZb&tbC4k_-)u55V52exm)yJW}$E@~>_3?W)s^*5@`#ab;uS&B2sbrL=UG92y^%j)?
zGo!~-ku5D^1{Hz=XX)1WMH*F8l;Lraw1rVF$<Y=s6=<}<;zFpB2&9hvykH`Z1&a<a
z%#HBR9o}a->oZbOO`I6EvG<9r7QG>9@4ow*qbgS1;u6ywbSKQmURmqql-|D7JU`#^
zbN8C2Q!>wtK$A&_G@jt7<Ax&W--@_-lh8;d9J+VW{;y56$;r@0V86bpR+2ie;QjD}
zN~p>(AXza{bo{>4MM2=NdgYveh=^1v(SNTK8kZM?-|Q(v5;3=KBXZx?Xh@_lgqPHK
zu>Ei|S#j0=O481}hLB*)_1R@Y{!_gVUute%SnYaX(wt2b+1oQba;u+Z$O92&JBs^9
zt7ZcGnpt^83l5y*@!tDWR?nXwet%D?iiNsHn@|8sonL(@WA>EJ6V>&;JEY7b1~~pQ
ze_)vCADB)U76tXZZoVW<FZo#ir@GL~nPSaO+$7#G>4nXc%R$Acz?DRPcr@uqx}@;(
zybp85vBc()f6&k2;aAujxYQlnyD390A033>GQN}ln(A3DUz^l-5%<n^{RSivL}omv
z>r;<vA)VTDf7n%mJDwVN=bn*Kg>y)8kOM=w3kfH`ri9!V@AfctC-I8R(c9KsGd;6I
z)OOpyFQ9y|3)ZRwQj!5PKff%0F;x(J4`5lm9Zf6}+RwgJj=U$0^84|c+u5-_oP(Mk
z=o}s*%4h8W-XtR@XUAaenh0bbp@sFBnopyq3;+T~QJDyw!;JJPwCeZnNJB_CtEM@k
zk%AJCcU(i~GQsn0?7e{wK$|AVU{?Duc7A|*fV4D_ylDac1n|@WiQu%{9iZlM;P}9D
zye?NSy53c4gHY2f5ZeSK?jS<YRDiq=bewgmI@QKIe`m{qODg03An0Zdgn)vn^Tvi5
zR`ALuSEc8z;lT3UVTUpl<S$u5Z~KM>5E({^4~wS<8R7p2kb{;AMCAcf;ye4T*I9sW
z2_&=+e+~!9|ImMjKvOn2zqXcVdcRzOT`(DK(CqjvbrYb{!TW6^E}*FLra{cv<On)$
z8<2wTrB`F#-7GO$*Ty$rH<GFkSp$zRaOG-ZEY9_hE!de9FlNeyfNim+4dxA5*4$|E
zlX@S^YNPgnCzSpi)LX#>+j{u|U`gClB_1u4HTR_I?<{Y34b<x3+QH}j)w$eH6$U{n
z3!VZ2>8BwRoOhs3fK*Es4dbgB=}RbTocHljy}6?k388Ru_^C%GoWteWo7-j{@hali
zt79wkuI&&Q?TDI@iKIJaA3W!H`x+IV73<#i2J`78M=t^`!aRX7Vdx3UhVw7A=R;4>
zUHi$yB+2Oahv;b%Xil~A3~?v++%XH^1<n75DPL;Hpp8{ns8L?i-5vJkbUT(?#K$G$
z2$+*@w~Tsa)3<VHhl@}t#O}<Z43_dV%g1UPI!2W~Mk3`tkD$l(eqHN88gwDXoKB|F
zO*U4VgWmBtOxZTicL`c@kk|N};7esvCB|EaRaRo(UUlDJJm}9}e;C{B6+smcEPWY?
zg;JWSB9QCfRDEDxe>AxNVhH2on`-ZM>2a*kD0L{78My`DoX+K;*Zu3IW=B_N=isBA
zXp4T(;0<pP=k2VntXNKw8PvM=dtVcNVxUlOpWU76AJqM7lD{IzkjI^?hALsZ$v6;e
zK74?&Kd|2<l**^_wf!Gmf2)jZt4JTCRG8T@Lz!bJ7A?b8A10>{qhLuy0+B1$R^j{P
zp*}RRIR#E~96=$G^~*oFl>}XsxDm8Gnd&4{cAg*j41x<M1&W$Gl&DR5$A}%`0r{4p
z!x6+g0oUI4_VL&2J*KnimxuJS`v7?eG8(Z{715b@P^F~fdZRsW!l$oMa>kilDlkzK
zrKsiMn&*2LytSvK&tx%X_UKX*rk--CTstUSC}DM;p!5=&>UktBc}<7!Qhsk?*03LW
zT*&gJ(cX2-vuB>qe*cRI+w@oZuV^0ULj3MwBg=$X=xOQcnn_JSAP8o`2cE(!GY1|`
zDtt51&$djrW*4p(Nz7Ue697Oz67mRW`}P7r+YK~=a<*L95A$k}L*;kW`3^`oz^+xJ
zG#ob+5C(yODkCG~foTN_Ww0L|e%=Md^ZsyywauM$lidgYPR3`T`~?9s54C<Ym;}Fi
zIP!^uPKa;&#BU$#+k-_B3TVBRx3*ieI^J$M2)^BNA1+kRQiOfs<~pR_%@)=D=-_bN
zc@&TcKIr`7LUZ;MT<l&jpnFiw4{Jamkdnx|w2+UD`VW662)YB{TEQ#>q0*?vC-r+<
z;N1e0^FaEhUa75lG=cb=?Ljfrw9&!06e#*)v^D5F6WB6A))f#F2csA{-Wh<U@O}&E
zJ32bDBO`0M#)baVjmp!uXW4S#Fe9U6Pp|;Zh4(}FDaTuh>xz_=mtY0@ith(-2RBvn
zt>!k~jZi!~eBuvz?8&AHL$wv=hurw3o__erre5Fw<9#KwmXfh8#-K++wF&KSKW`ca
zHe$6a)M5U;h;Tk1B~igu+>b`ex|r{Ru|rPd@Ve8bOQx7<;vzqk9^Yl31kT2^p)wMP
z4v%Oxb=~5m6PJAum58Fw35-%GQYd}TF!ed+rilJPo6KCka+6vlh0={6fx}aaD)svs
zT%9G2<opSkTQm35eA93jdI(inmj^c2J$_-9G4}5b@BAh<cghW<7h!8+EkL?Z2+i7Z
zLRR0uy}!G^y3bz|Dq0XJUOI9%gX4cW6*=>Kp^i(=r+O(Kbf<}2DL2_6Gx3XnuKK^8
z_=gd2xZEMrykr5NqW3KWuq4`JdhyE?@eQDRG`at)L!gLmVSyXTlljp-+o2BCUv=GQ
zAXohxr{t&jMGadeT!Xmyj5r=<K9{Lc3Q;p<PmG3`UoC5P9F0V_qxJ5b(Uze$GYuMY
zuC2jj{Li=-_T0bb=&j!YwHJ^o$bbDty`zxJL5_zOYS$Plp;r1{;~O8pIAn8*Mtzx9
zhh=i_{HX<ZTJ5Kk2ul*#RP%Vp+$lR=TgoHOi}{q%s*j}X7EDiNP^kSy2x;l#qGPP$
z($+UNnBs}#kN1E1OT-R!ab~K2rZ9KdCTNfq$XBudQs<>VC~c-xz5)n;>DDDNM~b){
zpJabATl2&X!4yZwcA__<eYe6rOHib^$w3yY6!6O3^-v8tqlEQ<cx##Rm5uQgqh7P$
z&H|f%8e&qHyfyG6Az(@*n!f9!r1zYOkr=z8vg#|xBgrHo+<q77Z%GzFI=Ax-VFA;8
zl{<AFqp~Z9b<ShTohh)+cE>-oz}Cul)$F(FAvx>gFi*@h&Bt-G>T9o0_J*IjwG!&%
zGEql09*|Z}(_q>k%GvHS|IEh6NAMZ3IV=!(Tr^4xO-Iia+gD_nWGb~4B@?wG7Cg^<
zJ9#d4K|(1lyctAZx&HzSD=^ZEh=_2v2>e@GqNZ?izT{NGxr>Ac-BsUS+#}b(3=cF6
z;Q0g3e!hMs$Od`f?7}_%S0|iR2jx@;-NAwWgDm(&X?b<Ec<I3PWV8SM-Oh~|i0KS^
zaM?Xz7c~pK%kA7k^!c3KJw;Y@J&5F#G?i5>8pbX5R&Mexgzfvhw+7-<<1@3e{-@3y
zdXuN572I@GIgOUII3lC|3(ysT)im4;M4{v_9jKr+H2z%V`pfUMx(zmkm6eqQ<O{7E
z<CO@BmAl;vRyMg@fL{sCyQlOEx~A82&GiZu=WfqBI;^#lww7l5q8EW$hD9aELcpY6
zC|67Tc|~YtdWy0naT?#oy|^34pffw;^ZZyEBmSch=JzO#u-w|HqeCz5`YThY<7)IL
zmVCw^MS`@#Z&WR%y<67$5A730j}T(j;)x3m0{}Exzfdv$$U8p`<}=nnYVlQx!DF!@
zgW)WAB&_Ge+tBOec;Hu;?jT%$E%4?*k#n581?gkg#EgJ#?O@0!zvn!x)0&*McR9O(
z&$sQuGmpGh*6x(+aZQvhqBPNL%{Y4-cQlIVoAMFK`~7a#F?8=Eot<7Z(cn~7TbiOT
z2D=`*vrR$c2ES-iOXYw0g&C6EYug|Gt?(CClAMG^xP=vd&Qv=35LMyY0$nY=6*EUh
zeu}#!^HjwnrkE7AIbGB*$*k!t*vei!5=~N;kD3_q^ex=wb8W&UC7P$LMJVd!Sc{mR
z($jJM+(vhqvbq@%Lky+TB~Xv3T?qn79L1jhaAICSFnMGLlFSJKLL|F)gKmRvpAs<1
zm-KJc!=yLk;#aZ=7_6+GRVX6Yg3h~7{nqXbZa0qZQSPtTkmwMT4%1qF*MIwh3wC1%
zeqNtLHurx|95!68h+jek1t}4$?c&xGM-!B1xcF)HgvAZNT*B4n_AuYRjj!VY|7OGb
zPG<&0SqC`fKb<6g_FsKXndU6!X(?0dMg-W3V4IdV6sVz6vuGP+`&5*VyH{WEev2P+
z;+5-JxMHG2F5~y3bgl+Ry`Vg~4y&%O?;M#PeCICA>Klej|2j*E{JAcx=64Qh1>(_7
zI%^BJ*w`V`1JAsE<^uQMlyA}O&3p`hI%?NgU?>0P$fywTKZ=fxYIj)fQaxIIE#0oH
zr;CbhrAKA{6srQwK)<jy|EFB@Ny`j(6Dqq$K#j*UyUJQ!3*07D8jAG|p@5bV3&!EN
z*b`D%3U>o5W>~mKZnCxFl*k*k*Pc%M>I8x?;wb;?W7-NvQ2kqVj)QxRKxP9qI!TwB
zsi9~hl83!^+xxnL@%0a!!ke7AIZ+*q)c2U!-*EPt+Hs!H4(e)?apaxWGfLoLV3bPX
zpwfnmAAC9sSWrhbX2YrKKCzOpiSzg}kHHUZS<c05;S-pf-qZQDXhf_g>J36tfFB8%
z!NGpg!^#M8>3<EgLy_Sg;FXk2F>7nfa>`9njdyf#n2BwfYdfs{d%z&lz7cUX@O6=3
zo_!l>0zku(z^Y`;l~H3{4c~m;O~oMJ3uX&;&bykL4>Q@={?HQDJkURRKXe-f;@7mu
zsQZneAHqr^LCB6K)svqdwYq@5z0<wTAB5n)3KPG4jND{LhQc6u(Am23*-t=43jE=q
zEN}+^!Sxr&;QD&*hae*$?d$^`ob*_Ow?S<2U8E|nUk%5r_`g{1#o5@{KrK@lVhj`<
zE63eVp<k`>;_fqoD(|nZu6#aj?rTm0!5anuKZy98hC4PJJa#r|*^HGHp(PvS1pgIr
zgBGBe2ynW)<s6s^p4V|=Bf<_mug^RAx2Fll7X6^!VqaLl^gCFCIo0W1b4|FO`J8^-
z1HTqzOYMO`jrpae@{k2!t-<;Ux-a-yCU|E!sZ`ZLphd8d%*u29E^B>#CM_>1TgL_w
zvOj}bWU1viJ06I_p$VfW$C=gke`BicV(%Hqpw(!AFySBGLQwN|%}*)Q?PNd65Irt5
zTDv^#@4FA>Ochl4a~TFkq#B<T3EC<d6Eacob9wQjhvSV&qLm~ycF`74Gg$F_J~awv
zj6X${r^26D)RWmpHzR1PB2k$7==IeoPQFpNn323?wPE-yZg#*Ft3>&)7dfj~j?$kC
zKX*T)0UOwlzhB$5WT38H`AaHtI;7Tuu#2FrEG6WD38Fga_^~h9<G?Madp_~y<(-qQ
zC`(Ue2ZF7kr_=qR1}}_hqX|ZA2e??TrtkNHHiXwG5$!>MO*51jG^!eSO@Zjowo%P<
zbN3%!@$*{XZ6_BOwnPlD1^qKYK29(ev2b-!Az%Va3sM}}RG*=}JyL4NLB~g$X_((?
zv};;hWr74u!6KGJZh5>m1vip<@b2aNd;!r`DhTfY$hpKBm+r?{l>PijN}$vDjGL)6
zQ&cy-w<@w0^_y9vK3FIX01APgeqCeh2$@qye=O`>!_w<Gwt13xMi2Q?qc!OQqgVVm
z|HU=R5!Te!%3lhA?!4ZR1y5To6<tbKTgc{BqX2PcZVg&f_!Ry&QK#6pL)-elD+|xA
z^N$!$s%xb+!>P5Id!_LIJmR6UqL*}r&HVD1(_XetDY7?wQ{v^WwMhcxHXfX_G>zN)
z`>8vwF<A5isVwQ(D(os55S?W)?=S&J>Z0x2+iZxKSNu#=nDU%`v*OCq(iZaW=Kr*l
zT6c^SndLQYV%7!K=k)H7rs<P#X|wRE5BzJP<Mp@{RE&f(X`DKuu*?uh7LEyWVJGmH
z4sjH_Vt~>0O7)}pPCNd?ZuU=`8dhaTor@wRI=1_4{VfCbS$QiWkqo*H`h1spKDe-;
zUCmsiTgOrMk;-~Vn7AY4(#Orzo;77CJh0oOx%m{}BEB`k`}aJ&Z8J|FLR)y0uTH>K
zHsx4v4um()udeQ%Yz>ZD*ZWqD!`#3yIZC7K5c`H5+28%YB&F^PzwSZRz|)tjen+Ca
z2tn06u^lk9d_abQ&J8%6=YfTo@-{5!C-}ErP-kT&E9htR(q?mgDBk<+PV0~C?lSak
zO!T)Xn9PDcx-SXz1gRQeW1*d!2<7%xjvPs3$Vc1<&GvTaSAzG06__20Ko%AkH{<zN
z*RK#TUQO6rq1f*~78<Q}@SKqQAuUIxaFl3q<ab6kdDMovgs-|R-Y1*r7e{M0PEOTL
zmmg4PO?av->Z-xJkB<Q<#FDN!K0dBnsm+^j$d>LQ^zZy&=?vrlF5XRC9`auafVX67
zbpb38_zbY(nY3%Po4Xxpi|oL%^r<u_kTLOQxNE#PI$G%1^)qfBatPFRdwg8>EbD$0
ze%PbMG|VpC^{S@++;wqCH>#~{#UqX`r&`eOA9;%V{2tDS!79<upOcf5mlb9R7siN_
zN<B)|^dx1~KEQ>jNZ#VgQZ>>Kr8UUNzpwc7DOrFZH5SKKvDY)Jg^&tCZu{+r29ec<
z;lC8d;<%S}JDE~e(L0$xb{teJaWy`~Hsxbq<e&KkFl@PTq;ZYNOF_HYMz}Yr3YU)R
z$29^^Q(kz(;-CC7`^jZ++T48-Og!!7bA6b2HO{|g9dyAUet*n8b(T3VVa{sZxxVlQ
zoXg+`nY$!(f}!2l@a|np{~pSFouJXV_}9qUygO1LUA8Pw&Fj2C0SCWR2YpRZ{*u#)
z3G(hfJLp>V-p}C?S~6L?@N$PoEO>t)hN8qv7!?aO+}lNkR@m-(f(7KGH!;J@C!W#j
zR%;edKONa2XJ01n#b?wjvNnRE=-s(V`GjexXcjAmo%n`)3W+Gr4KBlKRTw5!O#7nr
zMp5W>|Iyu{QvtRlK@;cBCyw|3p0M$%Z;w0cV)nbXmU(5H;ZCUN>u!;(?6}J06Ik_2
zRE>n%AVQ6nHf6hI_;E1lf61sg(4lY$XCM0u6|q=LdmU-~ppeBje~D)=6H}CgN^4>V
z3x^OGKXnasJ3TdXKtsEryzM_OjRu5|0b+|drkM8{@GX!Gs8{WF?qplXW5tyN#(Ie$
z{#*4cB;Axy*5G~qB_QhZfpJ=S1ai#u-!N>kB#}(`KLgHq50A1`lL-Q2bero}2fwvy
zHi^s2QKkZJ&j>xMv3zyJVOk+~<#FZMIxoYo(ES_rAC9(Asvn)Znb3`qgtIn@fG-Oz
znU0QsFK%H5_MnAmEFo39*!icVMLtL;uMx(vX1daTaRa$9{SIsw8A^7tbIq8DNBL_<
zDOP$pQvZku*RJP4?qSP1O#EIDr0_jtXKw+xOGhE=zZH>HUwCCEXZ_r10p%U|!q~{m
z#Crg~0N1|zS|>Xvr`EXI=l<?OP+S}aH@O!YAOY=55QyRZEq-Hjk7HrjJqw^Oj;kF*
z(<Q1P3q8?pDdS;#x?(Lro=hkI%+y~e@yvJ3lFO`CcjjWOq$BI1YzWoYdDc?2nr%u4
zrhpK;7AwwCiCVkqTr1~I5ZgJZbkB@Sbu!yI(kq`W6Yn$dzZCo{oJuF}TI!r%O87c)
z@WBYS?K6IO=y>Nd7Mh~dA6pJNorUda2?<X|yTD*DLqo$E*gp2_hf7we4dn0TRkuj6
zB7I0S)?*<$<9)Ai0`Eq`9}<ZfKLuWqKwH>#EL30O@M|9GtkqvR7zd13qjp-ev?Ua^
zztUO17P#P#cQV17m;8d=$+PStU)a=;N|(bXc(c!BW)mG}@H;TEW(n1nv-dq=FnbiY
z{OlRqf0#*gRBsvjPVpzcu}YDovVE{2{n##R`8FTTVB*u`xX&bJlV4WfQDcS$YiGCc
z$t@=BGO3e%t}p)0rOQY|NL3Im!7NSiIgOnRyV-!%_keA>9y^(g|Bt6o1vqzbBdUIX
z&+&Y^nbf!SpRwP8N9cFcp!to9?z>m_HzkQR2y9R)zI+pqHbS^)7s?@>Cdg-mIBBPJ
zh94RM+P$Y!c(Sv*|1a^YH}~P)H(o*M;@3}tMosRI-*w!qwp&JQ48c^tCJ-<C-HAQ%
zI$wAdcgc=Ce&KTzc2OghADZN$J#px{(C#y$bsJ9Yl00$f{u_}T&svftC@M^AARKR|
zUaSi*$7hvPYy8B;CR3hDmGE-$pip1Lyn&5#;7n|+K3Qapz36GSR^(%=S%)&7<f^g7
zER!3Iwae?J<yJ&+9fzUe=pPS$0b$W^{C_6)8aVh0ruWPmth1Ff#&#|BRGOzH<@de{
z(<wxZ4k!*w2@?$wFw#dazOLi4`Lg@F?q_)D1f!7Dw}zgt*;L$yx3^Kte<K=ZKXQGO
zkoZPNpH{~#QMt6GbwB9Kp(Dy@Vw(H&+SKgzyvHMmkp&d<E_}x>1dCwcgEfL_dLewl
zU4fwaY2Q~Kj}=*P9LlI`8`0v-NtG(@B7A09zqh1nyXf4uMZId^BTVu3d1u>-|MQ^S
zYiqGX>1bg9#%JQ!h+2WcHefoo^RIvB(ly}KW^B(t9eXvyJpS|^otMi#y8uWtSJ_&H
zDv;mfP<DM+<#$vL9VX7@6_(*yK=1zP-NY=P{L0%~qRF`0KI;`kg(qG&;8pXi-K^3I
zcnT_W5LCn6Is`Hg3BZf?)#_>Vpmg@XXkV*biv8&11eiPke1=K=H#Ffp<K=qTixyz;
z0*@Tv3Zy0%&=PM2Ui|M4C=lO_62GN>h&Hj(JN+|BTRhDKPDdCxJAs=1IPcEpo)rAh
z_*q+r$pWan;Z0gyZB2|=Krd(*CoD>@JpBrd7goog>sWl`f3Nu>`R+(`B*SWfmy@lV
zC+O}7P!0j<68eymY1|xxFnR#<*{99uE0#=`PH*R~IQX{u1cP$lUhPOyHJQ$1GVVz$
z(*G9TFz0{1V8eUdsq*f^YW+|s*z3%xdOFW%ZVK{WK3nRsZ6M@_ZyHpmh)?toToR*M
z-!oF*Sx)}9ht;u?V@Fx5Q97i}yS0YtX%~63%LK+ccfN$U-F7_phimLLDaZjqvPdC%
zY|LlhQhya$dGc*sra)pAUq@b`TH@1u|A`);SLFT|yP(`CMwpzQ86M9p8kVYCppKf#
z`>TnLjcXAc;k@6tektIBI4^sJ%E@B7w~931l0ud$8y8vM<EmK;j<~BtE56zBNienn
zI2(?n4;D!Z-+CXF^R~?p`1E`b*zTCJU}Vx`O_x@G@8^H^<gw>Pvpn%z4SPXSlNu4>
z-S``?-?WN?!e;erRx?K~h;vb;rrm!-um*#tSHAI#in#%&q)oFjpbNcz-FEOiMB{et
z-YKZJIsgH^%;`o5?;kF&L+)=)z;dehB$k$KoZTC_&FK_?AX^q}8Jim?(o;{e?~CoD
z-K`hwo5XO*o`9bL{=QdBV>iT`Yis{D6mW@2U1o|#bDmQxViS!BBCZ;kjh&_Z`Mb%d
zi6T)dwX=HiO|%>ZtyS%zxn(TmZ@2Wqc$9rWi3WpNoEo;z<Oe%8_d!=LUv*z)b@x|h
zHFN5787r%EQ9<o{)oL3Bd$qNJAS|vXsY+p7buc)84PKzAoJ+_^XsQ5bv5%96hq4&F
zWE!G!n1p_sAVH<(oHAF6nX_-;Bhj1slI(1zd;iY}qxwJpEeeEgQX>IIGZUus!OzU#
zAr!s~qp``DPh$jB4rimkM;5um^I5R0>zPtYE#Q_vbgR-vVQ$*x^i=cn^X3zjk3vtp
zrP{_h+NKJ%Trq0^bm;`h%8juS#3G~Hf;VD&`rz-E-H=@LlMX%+F6gw%M5P$iI+X-0
zCYW)`7-d7&JemtoRw878=z3e3%N28Jp^2wA!!MIOYv(hZ6-%y-jlL)kq2G_ItS6lQ
z=4F127GXR&I?$9?9>`}ByD%LM;|S`u8)!2x>_9^NJg)q{!hAl;?u$3uW^9DyH}jup
zHGxCXTKiN40E3(nyJPw-L4koP>I6WsK*mI0zYqsz_n^)PwrCNE&-p%SrFJ=nR$dU&
zXN<yE0=QFDk=v@5bIo1-N4Mig&RYY|khhTgY*3#YcX;;z8t_A{54IE!u>@ed@ag-H
zy9w~hE13Lo$Z`zEnZQd9vE(YfTk&7wG1+O+WwA?|Q0lvG-O<c7o?Aca&V!1rz)9=Q
ze2w~B5c>Cw;Xm3#Z+D%L49$)!y(F3?gQ9MYt*uYhf{Gif<Ujx)K`>}pD7~iZ2E<j3
z^L7)91n;YeV!e+qjyL{0TyE|Az*ENi1xbEK{?Sp$xSx-|N6k%h290L*euC`%JMggl
zs?>h_2uLYNaWHi4NfQUpu6U2I{};!5e+7}BP6cDm%t?C+$n50Im+OthGmW?{tB4z|
z-CH5+^EMKgNsxzy-x=@N)<@&P-Nug0>wBv~owvETbxpXQ=DzHqZr5Z*biG)p*vzaM
z;^G^Hi0GOx6E6t4DnhL+@SIUbwcTsV8u@b3@h}VDV`_enV@ShIM_-x1Pk(Ekfft4H
z4Ft9pY5bN@*M(|HJk@bDPNi3PLg%v2@}JOaCGD0&B1~!HIyRedGl_{$lLJ~vJ0{Mi
z!ryMaYMdm`1ck)RF9z5;1M?)9(qhr!J6QUxhvMEj3+A9XCcgbw@Mcs;u)B3~$KEeX
zM9-MYL)NfDU1t2E`$9DLz~OXzEAc}FCwwQ9QH8{Ysa%BK$k_RlshdR#htg$v_jR-X
z^~vBZlJXYi{txB7*Dd?~^nI^QUaQ@LjcV@6?UknS&0@e|w@V%{IT18_-&8uC4h>Qw
zkg7q$@nr3{Cdl@nU;L}SWZ}~Gx<yQ28;^Itly~QJd-+aWA|2hwF(Z!H(nm$_RpEPH
z?1J~}y4^qGHK-LL<w$X5sOQ=!isy74_R&tvNN{Bw2eXPD(}gow%7&iF7;|K?epQB%
z+=Uu9n~-L+F~ueu+_dAod%8$~7D9t5waG}0Lz^q=w?PsO!f!xv{_^?x(edU2r|#25
z)8Wt^a|L|D=$&L0{FjfveMuc26VD-QU>tH#Db(cqPU7Msc~bv^V}$p>$v+I+z&z$d
zImx$YKj2=ib0ni*^7uoDpY5t_^0AbkB~RDCOdFw>3gyh)F&}r`V}1#|S77xQCrkU>
z#E>IL363r+Hz?P>L@OZPz5bAomq;mYEtUt~uNp<9%Z}YMhgTmHbg=Rc6eplc)4jwV
zZ^xi~-QBNns-HPa%|zfs(IE`Fc3if~&VLdP-pbVLB*5LB0(lx0t@K=hwzlA1qFqxD
z+@%V`tb6G~X;O=7aR!q&tmEF_q!wlZZ}b@JUwZtyyT)%|m>s`sah6*@cjqMN2%8DC
z;edMuOf4WF1T!?>IJmNLxjU$P&OYFBBdP*w52Q&4^-whOwR_~n16&A1_W^>PcDdHG
z(>fM6H)Bk&$&03e#=wA6X5%90{^Dxz{*n?Qeh<0l&I`O|0z<&TL`uZ%6=HE~oH)D#
zF)rLlya3n@;4ZqOeP|_Kkyk_Q#cq%PL+(S6aiGTTt|Dz7veXFMwzhd3R2ks)91`if
zepp?~u5og)^4Ro#HPF#_cTDHxnt9{O-`L<APE_A@;3edFXg-@55J<6%kR%QQtjaO?
zoxXv-L~lQRZ!?fVw1}#G2qkk$xV=2Fn1g$)@I5#F%U+$J!>I@jY2RfSt@lv=n3EUV
zby)-~eeLaV^W_cdl%Z{_xN;JO`_b&YJSeAQt**1bzhQ}n=zpFbhE}ku8zXM|MazY%
zly{dGt=D=w)Ac&llOA~iXN_b3u06W%tp;?s8LM6tmz|df{dKgAks%37{zQbDL#UM7
zmA`U?=Y=0C*V<5qh1c7rAW4lY#Y+;Feof(|QBmm3T05Mo`}A7jGl3#%);F54UaeQJ
zEYzgF(L_wVB4qqdg=X>D=WsiC_;naTLI7Tf6b$_vQL0q^*YQ7P)HFD1k`l?7uGmeC
zlXD!4i`$^n0-#jB$p45Li=>_o`EvmQHSXZ+@scP*CG0r1j1GYf<LWUE_RHfG)hp!P
z((M9-;$&fv(&Y&GII{EEL$M&wv19HlDO~g85um?Rn(_&jERogJcSg*miQiG)=ZUo3
zW31h*UbOffiHTqKBnk+M(iv1=EvSmG0~sXrP<$N;ub%#7;17d#_g!?|h(d4g)}I4~
zkEY<5fg@$dKoDYW|I{z&;^-vk_UeAz1T>Qy`~R{X-|U`2kEZwj0qxS}rvIpU>-Q!2
zl$}qdN*r&p$lKaE&K$`fKXt>8i+`h-;nvjGE9Mdd!hnx46<TUhePV`+4GS^pU+bR>
zS5NB~QbEPh_{tT^3HjN8zl%V({Yz51LdLJI;NlY4(+3f=@v~1kWB%`J5)J!An`LWu
z3L2KWbvT+@+B|NB);KX9It?8rx?DtEq}q1$tY9pR9!_O*ya=68r=+J_WRCOS`=e?x
zxbw*1%je1txN+#Wx22*R*S}O4EI`%DIFikQ)ABWeRCQo>e6DExUBNSuQ^-+8lKGV*
zPPfu6v<zFa{53a)i8Ekho}9?PCW-7rc@Xrp8oGM>%YKGUaMYgUZ&^{*5@PyicuoxD
zOCL||{!rgxDzewT;3Cm@+_>A)rdS!Kg??Jyb>21;algy}hbj%-mJN7dOL8|(S@r^_
znSDE6wIV{|TSggT9d0+gz~1|PG2EAousfKCRU|XEn_u$`yOy8Xkx1?m%71j^z-J*q
zKF1ROBoxUD8U&Gm>r@aj>nu>AKh<ghOsl;4nJOit*4zzqppk&Ss(r`_ctE-2-EP0}
zzdacg@jea)<%LaNfGmNG5<`k2OH3nsET5;R=Y&3@1aPn(K!FdDeOthsd%ONb{6Xpn
znt#=x>j&&jYtX&e3kne`dHldF{=l$@;5lGPaC3Hkr1n?<!ia1J(x#Ck<l)gN3}l$y
z?wBCyK`#&d0u&<N0B^L_A4|;pcst?>sZVAhH{H->qMURP@(vdbI0a@L@|!7`Mw0u@
zd%%~h)y+@c+~1E|aL~SB`B=C9Py_w8v9ST~MVg}R&CO8o#VV?B&+;dmCr<m70;_oQ
zC|@Mq4)j4_B48!vQo9LKmeP45p&#pR*8&zGS&-J$x;f;PvNeHBBP2WT$c4PVwpLxo
z`v$%@ecv^$P(EueLgD7_{tE;x-0Agg)FiloElFd-G^J)rrM)pR#Mf_qj>r4r(F~<3
zA|Nwv#iMr1j=Ytv$q@bH7<^Z}z18QAne*`NV$RIOb9<Tp!nUv&D_rr6Naf#48IowH
zpxY$o;a?bpS4L}RcXa%eeV?K{8yl-J`-=<13|#{=H7Q&p-(B7zzsPM{h@M=aZ%ZRR
zb-<@uzKEl>h{Y|G@e!QIImDf2Ql?3kn0`vvoG{5<m<54)noN~cSWfVA^uvYU4-N@=
zsID;?+(9(8a0Twl=|K-WD2(7pld?=D8s?!Lqk9vjd8$;$OHwzz(3@@I%t4;q;gbGh
zFjJNPiU0Y~yJ*IZrM2tl)U40N?+7&0;-Cq`I#WfTM?|b#Kc~Q$43_xk+gnHa|ApSe
z^KLIX5b-83fJk6_K^ZjgCNJ=pIWO#XJ8r>!y#5X3>Y9IbQ+Vyj8NT{D9IJ&S(Y8@Y
zj+CZQDji+#8y=16{hTOrBdD{Fzw3lg9E>b!AW#Q~gy`PrJ$G5XSKk?nXEK3^cpL)Z
zEdFJ#R}KBLTkgDXjf*e)xqjvcg&HBF`&AwLkGxTE*m`^a0z0ydvEmACx3-n9ld@4%
z?sFuVjE&ZvD)Q6WZxZwB^s5Zl#MW>3aZs|ev+*%F7ZyX}u9h1+u=t`#)Hj`TBr))q
z<CwVOS)cVXTT)?KDn&{@L1#`zTR<PlQ=PPVDSNR-y}D}ix@qXIk4e$%>PatNN;!36
zO*M#9tXZo*aFl9m;kGdw4<@kKf4eXhFbBXF@L1m6s8qXKc%^KmWQq9x`^v%!RD8k6
z{JxVDJdlE9dvHfrv<+_4j=S@v?Ce}u&J0ZgpD*H`rke1PQmx7ntYCI~3l(}JzaIiI
z?0a1z!6&gJc8xSK7N}=j4TkCCtpxK`t;W>~W`-AHNbW9$vQ6T6xNhTa>t)0Wi47+x
z0TKE(F!YnLturQ7$akKQ9P#t=%{?Av`?=rjViZ5F&{x8Q@*)wDl@x{OV3!S=T?K_w
zp=W;<K#;6*Kt}Zs0G?<xU4x+KrHP7(F^tqHKB|GjPsW@AQ$V)?-Whhqhrpx2r=dV0
z=C2DOtUf~gM1W_LNBIFz3ci^V!%O6z_+81vH+-07>`WHC^FJMW!+Wz8Fw%7r@AN+j
zl(5%fi}Gqe`hn*SwyiT0J#g#ZKIo`@0MGq$GBHp(%?jwN=7Oh*b3arHiu~goHzE!i
z&Gk8R8G}5@W9tTC-)Z!aJaHTFzYg9P&E2P3Rk^Iw`q0eZZ&Y(<GG+4l{)qzS6=rBA
zsYi0W)&cjQp5Q0%jIsX&&F$;#cRr&O^Z)HGv@@O~3}z($k`{}WjmG_DCU4qn=H|ZC
zbSW&3poy4b_a(*0KLc=%9?;O_h5a8U+xy=iJlo8=?rsw=N<czY?ZVw!^?|(x0pxZ|
zCA%)hD;<2(CxOO_lv|tWoZ}F(e2hP!_dD-vcf5RdG)r=9#aZcdP}K$1l08W$HCY%O
z-)s}VwuP^-XCKu(HlM=^_Y^Arsmt)D1TE2^0mJCKF)`lYp;J8f58f$5{rz~h9vvm-
z5G@R?SBq20U$gx=`!I;?+bm^@K2D7b(R`wUa5XR`G047+kFy9*m3C256Lfxy{rUa4
ziBnEpZZ2({gfz;6oeaSsBNHC81Uj{p98cu;QPPi&$A6|Ft+lPn)#8?%Ip0Q-wWq$z
zao06Q)D@=F;ViwrmXDPb-u!c%lyea9Yc9A}oBTB;qCW4IM3b3s@jv}eJh;1T1@X3z
zWDL4R%=h3Wufx+5o4k5mwix}vhDFh|azgRTjqaU|K1fRdEO-s*9(^wVaOa8!$>tK0
ziv{&un)d=<+o;psxYJ1%NN$L0@|2s7^T?zPr@=C4cppuj&!Uhv!fTu92H!=Iq|z&S
zJ3h`{qg~q_^fRaZqQzwkL*Xfydjb9t*jF8&fTm}&-;F3%P_bT`DTbg-U+t9n%wCsG
zM|UeT+2Wbi#?&&ndv?YTmM}{f7GF&qdExcy%?e;{H&9#YapjD8U;e?9BymN|v7XYv
zw)ZPgFCsg1EfnN)i?Z`_XezAyETrG)PacLxHx`3!&fAGr$<#D8OjMeRv_Et|W$|Br
zFbCUgD8b%KB2vpI*h1BvG0)pcm^dh2XW$>%*@Qc7S@dY{^_R(gA69RB-Nr<!I+&fC
zlVtril*5!SD5LJP#w7f_<$m$8BjvK0z`U6~HxT@Am@J!#2yUFD43bmkOgeh}ITQYO
zb2q$y<~IDDN(KLLwX^SF>4{1?VPxY$jOq(v?Bw**Pb`WV>_kc><IquED9V56Wve0l
zUagLkUi(sS<uSvOjGN(}tx_{fk{j6A`DWYu!7kjn0eZL{3Fs9WQzj@iH+2^uuqFd0
z&SNJQ7D#hcN~{kvM+dcHU$$g>I%?BhL%@X6ql_JbP=tP;L)Rg*g!o=YFyZuNlKxW)
zgVOq>L7H)1N*QulpHW$YHS<Gh0<yUl^XL5*t@H-$8l~)wh$Sz3V}ZR4i~}CxK5R2x
zfn5=}felKi{5p*rpR%xVGO81}0IrAF-9JqVQQu!cT?7<cvv&P20*@by-QT)_k;4{W
z?F-7&ioYiNdQFL9*U{YwFXVG@A9arbDA31v>Y|QRnhT6(@agcm%EhhGW+h?3zXway
z<rbF$YsLW8X<(fGY6HowHEdfQ$?g9~wN!HKFK)1=IbZVI1PobL+P&a22vd&BUc>V*
z1Jiq{KpTS}Ne$Fjg^XAa6VVCa4)@8020!?k>EuDq<D;{q&UMNk!p4_Zo3K2n;`P?r
z+S+~SKzS?P$)Mj^h65@gM9MeEBD2^-e+}}68V#5!pvHcHVSPB$#Ixd<e&s*zE%LDQ
z{XanapNpHuv!TAI7xsHSz*1sYJQwycr~*ln1<B_Bl&ezG`0J<m6N>%0y5zqd9hC{y
zbs*9recm%@%%uMV=whGG@CF6GPP)0k*k_5<h_?II`1@0B3aS(uYk7XSp#j#0=pK)@
z{dg`V7x}sEvm3gcH0(qAaCI^(It>!EOpSDsoR=i=pEG$lNqA0t`xzmBcHNZ3Bopyn
zUr*xW6;FKa$p5XOz*H+EpN|F~%FbVr^$?C`l#W0#+xE32(vxi)uZfT~h~JY}GRDP{
zEM)EXR-mD?`haUD_Z*E&LSM0Yg;pC6aT1rHP(!p?sGMoD-BbvD<^zM1v83C(=hlz2
z&o3|1%P%S==GC`;yvpem+ihe;LN6#0e^w|>*ZiYAo#dK5(w!T7t~cIQOh1((_^sjQ
z>ncrUNE_o`oBvxu|KRyzg<!k)z}{t&cF7>}K)B<qa%~;F(x$g8z-qPNP2%Nr&^f92
zr8tt%>3+cJc)(!I(!wuFj^S5$Ls+<R$)HTr`u@IP%1Nogeo^mY64icE!=n4UrH1y$
z`Uh-gBopvNftG;8x|!R~HLn1ywV?ZeafnK^Ijqmq#ii%u<g<m_6t9%tbst5JXV#gk
zWShR`t|C|IXBrt@c@mPkv+P1bO_DH7np6Tz`iybwj(fy~0A*gS-xg7p@KSahyFi<O
zMdipa{yU5KU333m|2Dt2H($)-7>KBm=CGjKbI5*oja6||u)x%3Z9s{m=LwaFlB52t
zjW0{2IsTMb759beBulhoP)Y~ycQsa)%;0PZo9A0GXslA<mCVu~XRUhm(Z;c4P4(E*
zvyBX$xmqO@71LNrR3Unm#AwXl*GQe=1FsKXcl6Y>OTO7lN{@_>$2)M0bqtaq@v?2s
ztJSaiAj7(-K72mFVDl#G!|R&P&i<URtT+tOan+YnJC+1>$s%&hPJ(n8F?^vOxttk^
z8KZ-RG8mXj8_cxW*b723#F=$G?DvF`_0w+pE?I$I62f}|MwF&y-j+4GaGvN_HjW=U
zXkBQLM$h7uA1&ZS>okhwo?Uv8D#xpF|1crczkxV$eFR4U7)B^0;6X=MM(l#rOLj8{
z|B`aE$3YMse1&Hf*{WODQ11b&R>Qg3tU$BVkG=jo#ZNZ6ynp7UG>;s|zEa~>>7uga
z%9t_z8G9J-GDiz?G#Ey9>a3pzl#Pb9KY2mHk)d=67LoQKQt>e$gFQ$?!I%mB);1Fs
zf#Zj9QoVBd>|YK3NMLIg^6&+zn-3-~&`iPq?LRDBVfVLr2q)xNHz>Z5e1O2nKfkzG
zM*NrY6X4~h>#JPxv^J`|f!Z~aJda{~saZ68b0_vd&1F$UuCCqqv0E%(`Q-dPFO^)w
zlT-ky=s!aC>>A7TsDlnfTh1lM!(3l5JluT)Tbv;L1~}B+0t3fE((;(J&)yUea8<S=
zIzzJ7dmv;%s_c8FWQchuj42*~tih#51JGL!)$MYK1;VG*@#U^by?&+Lf98<$gh7k6
z5vwtVqk{t%urK!wQ%j=<^JTF-fds&WXDK)Dr;~s1!)|Y*Rq_I180Zl}_N^V?w0J$z
z$5CvAg5>6tVfWKhmI~E&{+nOEX>*U;1G|IdeO-*mZ7KR@;Rks)Iv)-1?z4cHh1&w=
zxwxpk|6k+L#f>2Z<1&~nLK-G`<vxBg4Rb$l!s}y`42n*o)(-uAC|cBCOEGX)i$88M
zYQD$Ruck0#r;IP(W){tZVPMDcs^Z&kE|xT|F*ON20yI5YH3I1aBEFcPAKv5QGRhls
zI!j5FK+Sj!G^_<tuyWGDthfd0ardFfH3yTW!Qdmfw6Ry&9pdgigCtWl?=U`15rNTM
za4G7PzmSzk2FK#OX|4b8=52JP(GX=Ig13<4cD8_f`!vo_YX0x!p%cP>t+mJGcIQ5-
z-&KY28TWxNOrH8_M82GjCEpv{#@~uz^rpTY>-{I3D=K+m2F>1gjW<efh*^qX#<2+o
z`j7S_?g8eSUwn6WWobF^4u$_{jsA}puFoSx_H+TvvYbi2Jf2PA^z0K|>iz*vYh!&+
z4U{^@==z;b1iqDCoi5w&`qYC;Wyix6Q+xYzEJxNU_b~(v0toTuwO`%9(DFt+AaO7+
zNQsm&vX27X#S%lVf*$DjF5y%(kEDv!LTHRwuh!ojA-3z^<kRyz?bl~lWhkXtX7r?@
z;D@ndrA>2>KNqL9919D>Gs>v{94@DsM2wysEg?sT5iY}=RHiXV8pc6n8*Y#H_L;s7
znqc_%^-V<H^e2pl^OsoV6J=#zD@T2v(QE%Md*RSOGdr5IglYRxk5vcMLmH{*ruLa8
z3X;u$r#^?`By#zR|H2W!|EZN;z9R9DO~ztIY~N0?9OI}ff7#p}#-b06_`s8>*Hq?5
zD^xA6)$H5;PYK$ya3D}GVv|!`F}<(RHd^FP{!+tOvp3JmXXjayrCY&X5EN66Fbu{U
zIi4^uS2_FFq78NZ!);eaZi~M2N%<>iZ5Z@fx3jnQE>JjuXK8xR{cjb4&Kfl6BUfOc
z*~JR#E$S|@O*fU_LDKzS4@*!3%#l+Fc<=Vq%jcq>U}vbZdc5_VeJtx-Spoh!#8q~W
zsq;{1A0@)Rx>LLUOV_50g+&2h_6iv63a?!~e+2qO`Debjy_Y#h{!!hRvE4sBY;-iI
zyHbKxIm61M|78vU*u&x4!$!wUD91c*W1-&WVd2$BiA;I`s@x+3KrIMl?^SPEJ8ks2
zJzmS;fv5VR(81VMezbF7-qCuzx@tUT?c28t;yA%Lb=2RuzvW^L29Q?O*czT*pwuaR
zvX1l@4F+hz4!|!J3A}Ald3R0pF*g8VD+Xae^?0?YGjo0O2bR}eJa+Ce1}iQ_`J!>U
zJ@m3LwGnEh2|u$!$3)&E?}%d2ffR|C%7}#ssJjZ?F&2T%{9)n*FfT@b?KZK)hwBP~
zkn6zH9(UK=u>6zDxMBO!fic(1z}b^4zrU<z9S$3=aop*A(qk9L#^JI3-&G1sE_`&b
zJ2T_%Jd}(y3j&O??Tyz=to^M_qt1k=)rqME#Myyg9aX=j$)f8A&3lw0CX&2tH{?G0
ziEN%qHq{aq>oa|QmQY+95f+6Q&QP^Tx?*Y49tnkBmXdN*p;~DI^bj0866hL1EWTPH
zDM>tf-C1S_OLQEHk+}z|EUm%Qva(_{D1!p2GU+x8aj69-ONF(YTC!R&)_<vLx_kQn
zH7<P|Pf*dVNRy6HWQ{VDBT%rO=1f`0&Z%Ve^pvUzjg93b>M<4A+~1DgnWXi6et2x+
ze}m%XbZ|062Rm_BaBApjz^&lw%kP%SAgsD5nQPGhLZXjVSVypz?33YggM7?Z?6zA@
zsAtTTr<{@dx?Ygn^=3}_u*)+MahBO{`N_CSKv-yFb36RS>J4W>7}>H9+{5#Q<4YTS
zb^XN{srhUX2rdIignpTf7Zayu+c3_qrK`;FQY9f#P*Ake;rSsKxr6pK<$vTJe>|eV
zV3`4P1SInem^aZSN2-(nCrr68U5lWF%N!Fr0qFtVH%8aho5{qhKQ@#p)jVW<A3Y=e
zvc<Z8F5=n~6i>+6>M9Rriv&c=y>U2`inNQ+Evl&euJyFPRxLHc0+K4V$?P597+r==
zGN)zpwluo;kx|<3d73#l74h4rS&Yed)wNuBvY4)@5*8e$!+!-`_S_xLa@l-hC+QJa
zaXYTh2a<|N5qsoDsSTS%{9YH{2Zy4LhafEIjjka_LP0PGj$D1Vhz@2sXVF`>p#}_t
zDLeY*#di5}!urnovHZ0+Rm|7nM<e@iq96J`ypN+$D(KXx2Lx;%f0iB^*|{An-ju>D
zOs;^!_5ULX53%R5V1Lq89!IuW_hMw$)blm%dbjc1ng7m*^Wn*5R+A^=Rj8^?1pc0$
z13+jAs&?KKPb0g5`+)kKlxn-U??0?{GC|(&;@#%-Ogpczrnl#qjwYYDu(BptiFAJO
zRzqiBVz`Rvp58M=m6XVt`xo&NjD5l5``CH6d2TTiTp-aM;5WUk>^^IL(1d{95s*$z
z9C*Ut?}1vNQuo&7#_jC7<~t*ZGBREXYFg+B;Cc0sU<uwh1O8&wvCo0T>fuvXyw<FM
z4enuD3~Zyb(5bw{?zFDpUtxLH{=g=1=XdTa`moOhpZeCSb8I;5$%{ui5}VbPsycq_
zhp@PA6Dz$r!clSLNuvGRL|~*F7{GY<?%kQl@lSiM46yuht78F1)fvLpk)3296JK-7
zi~P8{wM*_)H<{arKpX=S`D0TruM2D<DkOOoYC^fgs(I~?1Gy=l%Gi&|bPpcAhn6Ud
z-)RmKqx#ZMT$ug8W5gmP1uVbrSHK4AF0i|cSV!~sm|G{7%3Y_U=SSPFVm<ebo;ABh
zbG|GfgJb9PG{HEoiZvR8dl-E>U$a77&z6~Y(w0{no_W-w;?A+!O6Av9ITbd3ohk|_
z#QW2*y?^4?RzT6`n3zu}9FW6h`ub~%ta*Gs0X{Wh?dNda?=<g-?0x0sBvd}{l3-cV
zO<KjEF3QJR<?7s7p=XrEVL+K$U9hS2X=|hDWHGP^6+ZF&_W`r8+TZ0%=aug>i9xBP
zU%E_~p}08VlD~zFG*hcid`U3E)uVZnkw3LYa*YgY8e9A4&iQ1FoqYoz?8lwP>F%z*
zmwU9bH_cb^VbuJxna3&rt=_Fa$tFNC;6FOO&ViD;UyB4@r+RcTjIo*#eBNOqiYG9g
z5D^#hl;Bg8uTNIPoS>hs*A(jzDB%86f3X<gc<7i?31xwCry-ci7+bynrN%}jMyC*w
z@xPdF^@)uiQ%+XFE=8+U3H|Bn64>j@TjJ&|UD7S7q19h(JG>D~oSl;g4#W9X^2vjS
z4?2Wa<!7%7up1`76q}9u^KAk7QQS3k`RsS&naq?G?EtN!Miv`Q<Y_hX(?#Glie9nc
zeD@&yqWif}J|hci{472Gn7{A%<TON2Ri$O}MN9hv>~=cw_m^Eg{G??us;4yJt^<8i
z0s%iNaW+zqn9X>!v00`*Vw62$RfVFxI*ok_sn0erldAEMAVO=ob_lCEJoBj5y$##L
zW@NEauQ6m}L}b_f)~in2wv{E2-gj78@3DU&F#n(;Zu{h<kP2d!lF~D24C)P)NE0C~
z0^&RS`_bk`xM;~7I3Q-S`hf-t&hLUrObe<{QlUEg){S#&oT8PUQStGtiwBMeQR&51
z1tENy-$$9S4@N(i%jjx3<S)4p>o6~m6|JA6H$74(bTzCBJY$M07TeWAIo0W3<PCG<
zW?fniFuAMPHBYl5h2)BYWe->b+dZ0OiTjY7_=F?3^jMp~_{AK7<71DqVR-wTR-7G|
zcOdm?8mU+(Jfu(@gUQNhZ?7ljNonpGf{E$K1{TZPap(w4O0BJWH^Vu3pkZ#+Gw=n_
zr@Hp`k%xBxcud8C(&9n*305*lAF!qwN_lf8br^!)f%YOd<8;p|NE8@S;h+8=NoO5T
z_xrzbIXN}WHVkug8#cr6(bI8s=S)vG(;d^@9mjNyqq~M-+F@#%;dlG~_IT{C@z>t(
z`~AAF*L6LwMKDo-O{`p65iBW$1L}5jX7!Zu;^Jbjq1K`~hyeDye{F58uuSWy+uhYw
z3h2oc+SP?;9v?ohaixy}aR}^yL6fdcNZG>u0rM$e_W1R1=Rdi}(yI6!L7@6R?zu&F
z7>Bnbq>5B`dcgAV$3}k)){j^qemqJpo?APgpY8JQ{8exEe|T#!vVM00z&bF|!(k*D
zWn!1ImGNBGVPI7Uu9wG==>Fd4sl>jP3fCy`wLAU!_pgor+r`cW0H}lG&!GR!5!W>{
zyh>YXa|y{+C=+Ue7;i0AI9t8lnW;Y&;af6Ytddg}sE*4XR}fup!yipNDCe*4{ByMz
zsb{8S6I`+=+1#g$Su^GGjI&3hkE9CgEoFi7fN0%|af<goaL^DaRvi^e2Xa-^e@Qfy
zD8<;({fQ-rPyl3rM47%uT=g3fiewg<UytRNA9+mlnBv8EJRvmUnM!&#;jk~3)ue>1
zGSzbA|Lv+IuxT<;OdOmMEY`BR^S{zz$3hJzA|+(NM6@k>+`YK{jvpEyG-ZjBK@wyG
zLtoa9VO7wsnx|I=_HU|+>B!TQZ=;h)dS}{Is(GiXLa&B1T~jX71t|ImA8rDH!DtPi
z;ziq7ROY*uOj?U+tdAfkl3E$Z-Sdqaf(!5Ra`cA<)HNiIJy4kNij`m%((AH9R&M>Q
zS|uC!as>WNf`g-oqI{4#RYxNNokP)A>SQ{bi#+OQr{~vhcqo7*NCna^yB6a)T6xHq
z4huDzYlk$1oFE`C%^{uwuW>`tElrprP(JBuVn2Kg4X_hn)RT0U7{7k0G1IQo#p>`Y
zZ~Bn?Ma=aiXeayeUyRikPp^YY4BXpvCJK{^(<s$MLygG(oeb)C`C9V1th?xXA!Va|
z(>(E{yP5nNHnQmn$Z^<MiDg=>U6k}gg-Wn058~D#Bh`1mz)j8zR=`H`UyP=_z?Yu~
z_z<%g6=~dnE~w^}Q7Lfunk1J;#nv9SnC|l-JS02x6N)KRq(j;X@eh}G4#9c$1ogEf
zXCaHcN&p>8a}&9fo7c$X3GIKcIo$qU5yrd_q5sH(R@3Sk*!l&p$=kk_|K&Sj;s4gB
zMp69F+b6m{by14uFCP*FDL<ud;LpfR+H{8t>Ba3(z{QoDI>9r~i5d0;HIK7k{z%{(
z)0WE2JR0kRdg#s6jDK}0<eGc#aPPr4=?A$}n2{B!n6BBMe-GXqGa%;$kTQ~B+@Chk
z6noXCYZ^@8LdEgG$~ztRf`Iq#&s*}-5%KzG;?sY7JI#Te8?sP@x%vg=7$Vk*3BK>3
ztJM1Ww~tQlIlw+>{e&~x<Z;XM6cDHAWQa7%Nj6TJ_}T8a1v+KQXV)KFLxB6_as1#@
z`pcTg;bEztf8vH~U%J~<FDw##0qm#XvkgA^M2{<EFOg3AbzfjU1v-Wc`mQA};6qMU
zZ-3&EoR4%6{j~DulY!TWb<yx$NY|IcV-?Kz@Bi&5(3h6S_@3pzJOO4haNs_wBjDHh
zx-O22Vf)uN>wOU|rQY#Phwp*X<APpKcI>I-3hTFW!t-FPYNkD2wm||`Be*4R-u%9v
zaedmx-sPz5Ca|^{e8g7*ItsAHR!1<TS?SxS1D_cV=qR$*N%$}3#q4bDZ>`?P3zWV6
zM5u!WU-?DZo}Mp$o5-PA<TNbA0bsAq7Rb7CFq)K!XPpWML-fh2)8;p=+m5+%5viAD
z<X@IA4fs3gYIyv53ysh*g9#Rqg{^Xh3{BW(p1`NV^5b?@XcPwj^id=%zgnbv74J$#
z!6cDGFnddx@*s{are$f`$QYltR`PWqc~jdz7Cd7FiPFg0xr3)fMhs%)N~D5f88i*K
zBRI185S$Mk>|#j-i3HmNoTBJl8YJYZMBjDIY&sN`WmSj;%NG)rWauc~tbIY)Z5#4R
zV9Ug9C9WJXDoGO(NgV1rm)2tDX{3*4K99zWAs<h6UZlblHWAVzEP)(|{6a)qwl@yC
z>ipgKnEf}O3}1VMG1Apl@aOV}5W?4FWEuJyaik1zN^urW(!W#d0=R{+M7SR|X1-rE
z|1WNWoYc&$x-5C7^&m@5dP=)E5{8H^N|@ubo3q`>(?MLqkOpBrU_EYzeMz`$PsjCm
z7f;{g`>jTihxBXoQ=&OX566M)s~|v5gb}U3p|k2^Qdc)MOMZ)Q^?h-(t_1CjT}FFT
zP>jGh>Nl8PcP2OflP+~Tj{1%`|1U@^AQ9OzVivuJe8_QZ0r40pFbwap8csCMU)s4k
z!7{=|3FcHsH`*2Y&r;8rcR<t##)yqVFl#3lA*29_loU&1&s1V1YgiBc7J8k4zaZ4~
zp~2~^E@Res6|ak0fU0V-aUn6L4J0D++<MTF->g|HwvtE4A>9+O{Eo)nHS+UY-SVdS
z^(1a;ce}ma0JrHKME#Du`7`*m6Z=^7H-&Mnv1Ha1ik6XU*ylS+e_Uz0esaO0F`_G)
zpc3pHYMljKX4xkax#AF+<mQH3Yp~AN=gy4QW+fw*lt~j*JpmJTccpiuz<0~<P~0-!
z&DrNxJT-cFZF%OF>9c0@=q>wxYc3I^)gEow-}LTQFc4s83m$Mlyh^DcO!J+e5q4j@
z0M3yG>~TCd&Mh4cWS7MeoU3%IR|bY_0>(bde>*%o>Sn2k$RxYLySurz6(RDkXQ`s6
z9h?2GT{b=ld}e`)nL29j;<9ba{|XZY5R?IP=J5gnz#5>Tp%;9;!9PE0j;GrSAP|GY
z@Na!@ytdT+Z%%Lh-O3?DcwSoN>H0t5P|Vta>4kC30H@fwjTghwqcxrPX4ZXHe)c&a
z^7!}wO+rJDbpuDQj`fwoblSPim}>Pi5P3OlJ$kgcC#tkvZ$Gwq-`wzLQ+?*(c_Hj8
zH?Dhr>z(!9S6@>Kl`nsPCkW#u{F{)=ZgqC&R|O?L;8I`Col66%hc37B#rTi%sgM^X
z?#Xr{+4Z{Yz<5e8(*$4~Ps`rrAH?vDI{^U|m~aI@X1z7yPoMSk0;2F7J22VXbY9rF
z7|EYr+u9{paU0y*IMr{lM0%eLI}25+-goE>Tk@_0EAHs%sNAx2T6+7wbHscfBmQ>P
z{b;!{ND2EcoRA2Ad?OWAwyut(qE*1Fo_0})D1AVEJ?BaizX(M^MY#2ySbRj_e(kN~
zk5db!&uJiiMXv#+Py1;d|F)RHXhJdcRWKo8p)+NtG-Z6vUw3B53I`gV*eP`t>jPU&
z@&C{*3imak>^zAiz2b4snPCnyfE)jM6WVHiiL$PN+oMuoqft*m7aE(p|2`toa^LpN
z;xe9lAv#u1>{Kp#fOOt@T94K`lQHL_A#Cj1SW!hpe90_B`bTZ{c`Y1olA0yNL}%-0
zlu^&cwBW;y1-|>m=?4!TPurtabDIt1gmbnRBJMqd3xtw!`Q>@y%Ic%mG_y}y6=hB{
z<O8IMsp9h3o~=_4nT1Wr%^6Ehg(F;-f&yt1`Z7)&2UTpiGkNr4v=ud}MFcmC_liOA
z{K*gYA8Vmt;{Z5b1xW<xgxqCJK1KvR0T9SvVjJusx>w;S6~TEen#waMIxyiOmAQQo
z2Gv_4wEP74k{`(b!z2-%ohoP5b%>k_nY{u6j0NkQMWW=Ou`&(zkErEO4UL;tVMDnD
z46?nLHZ>ZWtZBa1Xp(DQC_jeqf`+W&$8|Af5FBg+t8ay&OzZ#Zv!_;mZE941EN@o`
zacto>`E}x9zw_>0<gGseiXJJeInVYf<Uk{4={sJN?-8;<qpQq1X^)1j-tRK_FrAqQ
z6;H>T*K$CT2t$z}LZ?^mQfV6PGv(g8+68tP;7no+?@2)@Pg)MEXLhi=RL&07QAhO9
z$xQQYoCG19^$}lmgVR|8wBxp_xG&}^)Bmh4e)-dHzf1atM6K(m+H6hAbj8F>m7b!U
z?wzkH(tX(shHlm5mHN|hB(Kz`;RZBmNmXEn30$G`IHL8EsqgQ%H4RuD_>wkN-9-5I
z#WFDL6ySZuF<x6`4|*Wy;UDv{@&>)Ml<!o%Zj@QS^Vlu7rF!bq9?CmU;(jh*ozka%
z7kU=po{fD$5)2LmkVJb-p^5+m9oXyc^CBFZH^Cwkc6=`+x^{Q`R`h%%{G+B#rOmQ7
zbw7Ykf)4`dUinCv1uUMd<)0CJ)Oa?GpLr=~)%z7j^L(S<?w*TMHO*wAlw?&#g2BSL
zwy78UrpMnr?H5_8mW_=eSX1ns70snR${0JCGxEjRCYsp{oko^>J#RTLaqa4FgvRx_
z=D_I$YI7gKK4mNg>Dg-!t6m>{tJ4q&P{i5Yja=P+K?$VNl<>f}<-7icm%M0uuHK`@
z&kWvhr^<tN3<+iD(&N0rD-=HR3zD6CAy3j78=jP5v#m{g-E?`IN)XXA@%8OAce^w1
zdD;2?^Me<E_8K5ql=okj8Ap_-`Wv`<osT_3>4$487d}T<Ny6kc%q*?(W#f2v)Znt%
ziWJ85ba)zKk^K%-loEZV&Bdnm!wn=%ORjZxmRThaF_QjBPG`j_mfp0CN=QVjJja7q
z#J=8TsR=8Vo2atSW0hTJ&2OlBo`{J;k6o#EZdW|ZIPO@qfyIq2EyOm#Aln1A`O7QG
zmePQMdf9~~M%-_=qXuh`txu#0Cs<LTQ<9JoZ`6NPKoco0l}<%$He}6DD+_H{#M*ZW
zSE6jznvSPE(1@6K71+IOX=_|w=kUD`?esYdu$XpQ6c#|(9sGqK2siwQtw%s2Y-RL`
z*`FQ~ebQoJ3ZBMcf10rsVkc<!dSF3~>p2U>Prr`XU3Dp;cP3_rCY-<LDxZflnJ1^{
zvL;uk_Pg%cj<yYGk62K@%rdMH!~TBk+?rV?Uh(0Z#GyfnuE$$kOq6k?!lbVgM0%oe
z`)Ofrh<S8bc(1Z34GJfAAqs|WM8E{)h~v=NdZe>gG#3P}wOt=s5tJz=UY=?1>LJxn
zYPtXKSKE=2^5!Py`h<2G6GX<Vi@(vQE2DF1xS7GhJ2DRY6K`#u#6o1m_^QPYOu!2E
zl~^6Fj*ZyFoO_3!C+^m0d>CX88R+wpZGk#m{^kRlK(J?7DadRfBuzo#XK0x)H5+qG
zt%HtDw1~=t=J_>-QJi@a3+sU6EV7Y_;+b@xaid|fJXXG6pO<2zKCQh!_b?ChXS1mh
zEBOotWE%0HNa<-3i{hQAtbA;OBuo!6gsD6A%WTfnNUla%Jt&#9oSsH})Ud;58Cw{q
z;@ADE=H^5ctG&spBB##Xs)wU70I7Ie{bX-niYIBEZ1coz1S}uHwl2uXZIGqvT3du~
zlY-}S`x<h_`}y*GE9f6GQ~bTOG@9(R(?05zo?e|WH>B_S)nR!hb~0NJ@P}VK<O))Q
zU#gBts*T9!a2J47=RN!ObWJCeL8hlt=R43kfm!nrgn5c6&i*_3-%CJV?*^%I04J?%
zC3ZV%Wi&jQAMcOOv)^6;77ltDYtDl^4vTsUGwrDZ*GCT4=$u=~P<>_x^d}h`CT$t{
zd}X=e#+$080r#S@z)$0I?ortou_-3MhHZAnf1ncK?|FR>M>s=K)sG(dsqsL^8SEj;
zLS8s?jKRAoTeXAoT3r%d;hmKe>el!UkT_tS+1XB-)2#|LQ)o$9X01*;cm}tN^FNoD
z+tsKyDQw1%-OXVOnZ0BS&aKOom^avt@3Y?-l8Vptk@>j2><77&8w>pp9^g-Db)9ef
z$2HF<D<BhRTb_d6R@PK@9VyVg=)!d<bTO9Q>U&Wgfm}^#(Cj9z2GLmiq*&mlO<|*F
zt*G_2T0&hkH>#9aioKmA9Fu$aB)^zG^BZUM?h|Z}=M~tZ<V1P|)pO2M85+t2KXJ=(
zp|TCcV=5?NoFYvEL0_eWU2Ba|)ClO}dbx87>5Zt30+Hx~H53ihp&x!($C*+{W)b!g
znw3q$34++wQ`j|gQ1V=HlHRb%rWF#K6?ZIvs6+|3Rq#-T8sx96<-(yWdC@ylej`EQ
z!uP)zty~5gBsZTZYdI?sVGobG4sQFzR37gCT3X#$^>G~d`rpT<H}48tVL$H2NCjNK
zzo7v637;|Jc1OQCYuWP2MZ`fOa*}NF++T@`H>x&JO#Z)+SB2!k;rcnD64FA#7O8Bp
z?IOV4lBwfaQWC`#zu??GX*%M@Ra}A;D^<@tUy*F1_;KtPiq>61*`9hyTWxvC6v~u)
zZz(~(-ac2S-*q1?Hi?<CAzQDknE8ZR^b9(}Sf1<M2XmZ7k5QrF)CwMKg(Q(h)3=#M
zq&z6BRLh~R?|Y#bGv3o2m6>c<!G)}-dGwB@Q?h|pmD3jMumY_Lc5MrE2O=WrZmHL3
z>fG-4u~foeq}OS+tqe3gpqKhtE%hWO<kB+aL70$7+QG_ltL-PdyaGDCVscd_Vv52s
zCTAIa^NQ(!Sx5i)WLIIaeLtpn35Tw2xF74sIBq&MtmVVlw_o0|baSV-o&~hxmjyi7
z6<EO^zeZ;_zkV&fvN{jlU|%FIuu{xHK_h+sx?!<$D91@s8M$G0=sNV^Xjy#S87ix>
zY+B*Um6Q9*YOkSdff_xDv>J=Iik*1w%4WA3v2~S5cCCqJQBAFYL*(18A8^M}^r4E>
z^{NvR{>YG-6BU)!7D%|vM}(}-#PDg$Hf8MnUA>(|^SjsbUYrON?ksV$2XWNO_EY&3
zuA;i7IGF1&R44E`zLxiYxCYW|$YI~<!Rg^5>n@T#Pt%L~dR!KKk*ep*RdpWyo1n*X
z1FRw+!{^|wKRjdy-UC=nx<z?fMP{ee3u?Dx1adt_ZC0tZ$0p$4gS~>P4URYLz*(pr
zP4TU;n}-K9ySlS#l-M;P>{DIYy~{n{3jz=14vw6Y@vv&f^K=>mtUuEgVA!Z5g#r~r
zvDnGV#DULyheiL6`fJLRfBC84a9x_@pL35OqX9tK{B$zON^LpT&p5q#R7()>C=sr-
z1IUH>!P*XTrp4mhazuneoBqVUgaALkb*jGr=7qO=FaaQJ40BZxn$)pwfYcu_vIAm`
z`g;roM1aLB9k!PnJsv)l^EO#;-Fv_i)%(HAqxE|HedlO}gcIrxg~-c|s=91V?UwS|
z^_1rZSU(|c5pA>tY2+VFMX`bj$C$-UIOjC*zK~jZ&*n<u1TywwOB#~&*^Vx3?9d0|
z<*mAwwrIs_$qBU5d^0edW2u1X<svO=mr77RN0WXPie_kwjv}AAr{Zsep<%6H^|XZv
zGl)*wc*rF>POvVtCbwXMK|A$T;VH)OR3gGltbiC(rhFOsmGnHtf^9SbDG6r9Y*5^3
zWdxL#8NWu-e{c>HiyM<EiZ&88yv<ZODen^F)sS=F!ZkIQL4CV-APq+@?>F8LI)S5T
zP*H-AUkH41MpXU;QtxZX(MyFlUiTDa(_BlQ-<vzXWJk?~Zh62XHY_SmNt%Ou0;&q4
zqX!8Si8L08<c8^a8YE<d1NEtEDbSf%CdzTsv+`rc-SxdfWQkg+sD3rMBm^3GCfsk6
z_V*W4KNnIR>$8?ARhSLSt;>dVTg%WAQ-&$w?V_V-3R}CV3+jC$3ZheDKTb}~<)z3~
zF;tbWrRgRpuo=2eEYQk#8Oqag<Th~pM#YGNQh0SUiu#1x)XQ~9pPVb6A-{`=lm&%K
zv#6VrP&$C7*D{KoH3ETal{<Yv*VA=<N`k)F#1=-<Loz5q%Ye=}jwvb4#PE*R(Dljm
ztDYz%HX>O<!$~zdehdN@^95r<;M9B!+_1~Am^W~B?ShL%LYb$Ol*St+%|!GTRxWt_
zmS7><WMe_a=2PZSOAO=#yBN+;%mB<n6vs?19m|Hvks?*nVG{UnB-jR=<;jYpmq&Mk
z1xKuSZLN{j7hRTiCU7~rC6)|P3-|df-6=q)N2W!pZb%f_Pxe;WD+HV5u1q}%Jv^^M
zRrtOf<C(~WOz~j}s%|@KRjtme0ln}qv&Jgwvn|GDzTYQ*d3T1F+U&KZRkvd|2HAI*
zf_+1|unKQ?^8C5m(?kr-E(X!L8CJe8=POygJYRe!V>YhhNyh_mI;F2e{_1OD%y_)X
zW<E3gv0vAwv|8-ht~dg|!75(0en)urz++@w`{*)xQg#3H?HeG<gZuBIFh@9DmWVf0
zAaE7%E5N1L?)$JuJ17XQQGDQEpsjA^fzSLNd$|MFk`!#e<NLxnpcGM3zuMdr3~pft
zCMG6<K*Io<gI=T#EBo?Mxs`|;7USDvOKxt3-v{H%#X^-ZKDl|@@7tZ(fH6m@Av*|&
zL=M#RmD667orAk>QlsDNSi%DqM%;5fG{(nMx6OO%jqWLLPO$=V4x82}mdnTG6Z9^-
zs^<3{KoL|teJP8kHwLlq(=<j<7RKuzziJ9wI*{{m@j2zq9tSw9#W(Nq(~t(q{sa6i
z>Ox{FDys8lk5lXSnqCk3;QBLf!JTUF=B9;eH4eRXI_9b>{qIv<?fjs^zsu@yE+03>
zOK>bi)z<&^g5%02O7sw5e`lg1%$ZDzNcmi)2BUn2loTwfZkii}^z<~r3Gd}QF+5M(
zMk4WC%t}W6QO;8*d1bnuGRyA75~i+QA`{sZJW};^x=Q84a6>jsog9`%CFU?*ywJ)W
zIW+<rlwkJxO)1r%L?(PPMAO#a!RamEPqqo2j#y$cl7U87j{ME=YNrcDRonDxse8*1
zEuKg39E{}dkCmAj!jIP_5mrwmJ)QUl7xDDVE-vXvzv`2_4A&_-IUe~#(qALwx@?=_
z^x?bTg}rQiQOiFQey5L_3&D?h?R9|&=sso?7EY)B)iww(>F?W&JVktqSV$D7QZl3`
zhWowZ(N2n$(UK*2VyxrXk}*jdAT0}JrpFeqNkeP~SXh9}z~(cxa#XtC=NQ#amxu2f
zzc=c2hw+Wf+bD93rAECbj!0}^l97`O>y<{}Mh&B+rzD|=ij~IFC`s$lH83W}Yvf=e
zhm#~ajW~wuSH><9K!@Lsx^ZPYeIp%V+f<KSSXx<!>D<2cxP0fD$Ig}+&xBf7l+>_5
z{gml1pCzZ5AAi+BY8QILxuYaONbg=DjfSU`s2YkCB({fU5|heMZoV{6i%l!zZSnl7
zc&~%m@29ZFXj%U~v@|rM0r)l}zP|7CO1+`;f9r|d)hH4$S}u~7+QpnGMi%bm@}&Aj
zi0qJxW{aREA61CrXJU=t?3m(_wXB*f8tNPN5U*H=tr9rYS-mv9YQA7UQB-n3L!OvP
zd{*i|TB96JvoH{z{tiT$sw36_yYGQqX$xNAeO9uC@ri{dUUf6-23c8|jy|r3u3NJ%
zr16VFgXom(=V`&=NiD`zfBf)ier@OQc6yb8FO=qj<ohZdxTfz)!=~HcO)M~&0TZIL
z5G-pz!0GZ&5C-xKR=3-)<$oepuLTasyV%4>T{RWnH!_;+lykM$)><I^pg7wu5<bdS
zdl~=4|8yfMMf_|(_5oM-(Uo^RJA6LJM3_kD4q%s8Jy)bqtO~Tb$4<r`kf^|61;JM`
zGBOyK1JCn0uaEf2|4wetpgda#=K(;R0Pp={#-y0_<y*Hd@cRoZD^+*rVDbN{c0IMz
zT>0xm=llL5pNhd~4;u>nmbUFNd;`5ds?|RF%S{e}K*&Gs#(1f_$mEuu0mR0%qH}oZ
zpsxGGZn@aItCRWa(g*9YYxAVa&zWD;U%9%tDK$4Y4+7k`{`fR-ZF&UGXQbS(xOeSD
zu%+oi|8aR5?rc4M5z3x2eB*A70(oe2o7k;}eGI-8JUs9R`CFv<ywP+zRG4%3WyyI(
z6NXDi&%d*?2J@6^?ysW`dez$a|I+7c-SNOSk?LorTG&r(E@fHTf`W#mPXR8;uQ~9g
zuHKj7k6A{+5lKV9BiF~ZmLu0XT~tm-#M^BhHOPlD0E6dXm19gnqw4Y;kE_e%+Ky7=
zlj_?mB#sn6zwY+7(^erx7l-sm%7_tQkQZW>Nu3)}6Ea}+S?ks6*(s|Vi88TDmI7Rb
za~`s+>|fSmdD$MC0s_Hu*20F65_E1{>2)+cp=TpoUdj=+-y6Sc%UiIL%7xq^%QDOE
z$B6{R(L}#8hSV0IiPv{dn^<$NyUWILFpc4`n~%Rqe_y<esBL!&tZZMhwL7Z)rXVOR
zxDG6m`lNgVq_Y^ILw|dq1Cd6yMzl1ndS8RJAQ%NE=nmo(OqT*?@Y|#NZ`pTedajLj
zoWICxzufZQMG*$tef5ct4fS&^jtGAy%SJsghc4NYH&Vv9C9&!5&yFphYEnnC<*4Yu
zm8&~WAQ{b#K17af(j<+;J||eL-Q8R}qApR2lE=MaPUrr~vuiN_BjQO9$(XcU8DF5{
z*Y?&gab>0=*#sC4=Up<9{Pc02O)g0RN=Bm8uVEb$%$kxcL88!k5qz8?CaW(u4e_KJ
zOy)Z#EnoQP9QofjmPNaKlN8#9VpdEBiS^dJr(<!Y_&BCDvIBy7Rt$@8zL_Ask}Q1h
zMB~Jt>EP(mV=BZjyo@(Vr!;guFR3h%L63Wb>amRM@TbYClCNk6)8@Gi^jZ;{1{?Jz
zlk`q8f-`oPjzNJziGV3s?NA#BRkC`CHk^1#Cp{6Gu*2aQA7ZrIU|c7A-ePPR6dR-*
zG{QZ_yRm!qZDT`~b?!Ww)L@{@hn4_a)IGV2U;h=eU%w`fSEuDwa5rlCe(#6@e;Tf~
zcJgCfv?9}h;NRE^56@`BiaP05OWFGN`eFmp$<Y>>-FW!5<YYauLm{D^Psm3ZcfU6!
zynSG=-uK*xmS?+;u+9S4+en-|#HCjJv^~v(-9TYcUiI+LpI9>r@KyuV69{T-oP0ZR
zlLxeI3yZJq?eD;9@!P05>>e_HdgEQq9_nNC`6h->EWqGrI->5eC;f8OHyeQIzzg-}
z&6|=ttal}UZw8C(hB*G-XSv;FKm0LAmQ-~E;jjJk7IV0ldJRS}II=)83(Vsx#r1bj
zx^J)CU8f9GAHMsHT^r;0j;GZHmfP~ZF|S=Zh^NO|aov7Qi9)0Jvaud~vJmHsO<)N)
zui*%5L36D3$>J*N0Mc6n`?_Ds>`$MuE6+6?HT^}ib~OJNi8nK!s)rb_65D+ZeD6DC
zh5pyjwNG5@`#yW_vqr=1#^rdGz<`K|NQbkXW<QznFFvGW^_*j}sOq0Aa~){thb9Iq
z1Uh`rs;6F))`-sy5?!{lG}+@uNkYY_g9`TJ7wTUz9^eTpixYk2Oij?j6!%v*EU-al
z9HK5~Q(;z$uS0QFFv^Rs`wLG%8$w9fD(6+onK4n{(q#-iXAx}(gg)CQD_QUuwdE?F
zmyMM*s$ftg3Icaqib|Dd=(Gz$h0^sNn&0)SEV0B0tb*d1XqjZAy3p&mhaoO9)vFc_
zwcTgNJ+>kiVCJf6F|H*lX*}JXmW-|;wc&kkweShzj68VoG~Q?I_O)u!e*CNlJi$Vc
z18@8%#itscy?U4GerBChTdy%^5tC+J<z1q!(!Fp(=1K<D95m^t{xFF~iS4oRQKqnl
zkZ8UV-Fhx^!>C*mh;v+AC20l;<m>C#Z@te#e}Xb-+Hf7Tp|I}OOgtHC!}1qpb4^v$
zRiF1V`p$6wGH{_2M*C-)2WXe2?m(O>Q?<2Zw6HkI3u>H=`}PI+nb_g<aFuNvO>|~f
zs{Ei^{KMD=lVrKq;&zgi)|#(lxl3RqjKdR>VyR>jxj4F{*0j}JaQ|1IunI+KfkN*<
zPMh~lT|<d=exeQ&Qa)4ufvqmu=y0c{f&AcS#mTkc#wm`oGSPonD$E5cG^iB>SKbcF
zn-$48>|^7ColW+C9O6_H71h|?D8Auc;9K@RhsxkU^tQf`qcI`NlA~xBBK(<RaXs`v
zu6$V|ohH#xF8r4Yc5yyBfeJw|vhvT;EO(4ZmS16s0i59(bB*J2%xp-=NJuggVrs~w
zkuXJ~-1hktoV`V@q44rnJZ8P87FDAA!|T)4cDmmOuF*ROx!bIHzih0S_XY}ow%f<S
zO2dixc;GhI?Q5qxyeTI0%<HBh$s)3fT}V9C!{KtV*tJ!!*!0eLOS|2i)}Ji(mRWN=
z`l`W-hXi{jm=vgTN0$q~ewhL<krpdxfjf6OF&ujN-S_3^C>7#Ij3Xdm{r$60-*Qy;
zar%ZX0%Rj%iy$Dg4I3*f#X9?8kYaw*I9TOQaO4l|T*W9+UCPQ828DHC6~M=Njx@Nc
ziO*XVzNT4D0^T{r(wp0T3YDYt?FBacn|J9gz_4h`UoL=?m9+Hp=kM-7j0R}s%`fl1
z()X`@vw>Igsm9wkXQwTK{ppJ`_`BXM%RF^HEA7_x8`%%6!`2B#*^QT{$JW7Z-wOqW
zg-zbRm`98<+8|6aR<+#V*JTi=XB9VsV9~HI&fr!$eAYkdmA&oGyHoL?r}}!Q_G^lq
zu7$;4-|Z{w$jqgG0=9Je%!U;X>*hMBt<2=>WQ^Gt4rvU9#9~|1Mo&#GRN-r{{`#a?
znuwN9NaWgAr_j9GQFvWRPmC?wk7T7sS3no3uWw7(6g>H@?4$*TNhmbVx33a2W;#m*
z2&btYc@+DK+?XHoG<raR9=(!tJiid;0LzpAtb)fJdo%)Slr{x3U8!tzWm#~=6G8l&
z502eK(_{T1hnP~UN~2*t=pgfX-E11dcF6Ee&N8CO_?eM*u{!P7*j}SZP6A?v)&urR
z8oK5;7TUxahF1OY5o!skD4*y#@Uk5|(DD@{4s-es+I7CQr?YP3z^Ab`27}rLCgC&%
z_VD&0wu>hbDJ;-~G652r3R(_X4<k_;l6(;U(MYeE%^R2lTRQU&eL{?QQX(VS5wD>Z
zCh5w;-G<$Z7Br*T{yJh-4VJpR3Rf7!zfM|5$sB@M<b+}I&^b~m51>@)Ld=<H>KJ|y
zTzsYanPe`KPM0K_M8RcmLjHH|j_rxEB3E`?RgoX9;(8hb)}|bC!`$NNT3+v7N`*MI
z_8I)mM~QT@v;R=R@=`fICsD#u!4Uh*(Vv=xJU2EJ`=>g7wC@`NVr{r%$L7Dy#Md!k
zuk095T0x(|`HnV}r~o@o%#R@XEr-lJ1rhoj#ZwbDGZ1@6?_h3V!UiLD+1^qLji~HA
zl_qusQ5AYNtcvV&@iOQjn6ml)%e+{*kU5sazpM)Bt5Vf(v9^ZmiA5Lp6GVbbTHT0K
zvegjS1_$0cixf9pU=!TZMS5P}8nAQvh{aqAkZYcER8OgUYf}!zY#Sph=AS;kMXIP?
zgh>1pUyH^{nscuD7CDCCUX#YTSOj-Pn|p`vaX*0DDb=E<4r%q>Uv$Eb&>k{sx9wq$
zJdia4xAb}nz<GHS{WvDGDS54Gcm8(o+Fc;PC-RCY0r|?EN^KIht8wOG&?o)@_<s@S
z9sqy)h=2q#*6f-&v^Xu@uL>*ZUFb=qI+($y9baI&*<WrBCdeV=@qYrQ6ouqCwtL9P
z7lUJ(!qW27_ot#&?iKe!jZ2@v`fi}E;_%g<g;`)x04`x&{u1Nf3K-w*E;74P;N13;
z!e|GQh#h||W#UMwd|&_wMuFNq5@6WYhv~69xNiW#9xNjtqf@A@MMnmzU-neDHFq5A
z*Qs<PmlhXWdI5<uTw!i;>GxGf-RMkxHrf5{w)^!?7pk1^vRCA3<%!qc<9hx=@F7?9
z`M;-Le=1-cs;_%i2bf(<*%kvN=YEgzxGzE@kHyA*%W@SgAyU$`y7sQngr7fiK`}N8
zOY8~4Q%^Kgu4M?x8z7`qD>z1f2K7@>yd({xwMr%oqV3V}F_wDP$JNYlY8BqiMX*S@
z12Vi<;dK(2^Up#4Hh_t4)KPrP)gic4wz803r#MJVnn-G7+v%qLMNi9fo46ebnXPKo
zO-vT)yl-zuYn0N17?pdYz~llJ<$Z3rQ|goElKSd*LX4JeY0AlzBq3-r5-EPIfln6?
zG6pbLZvRD+|B$sZ7b?UgZe|xWFwYq6m-z3$NGU2?Te}+X2NBF2n2U!G+*HUc)r`D+
z*<$_An+jxgM2fQ!CTj3pNY6I)i|1U~(R#s|T%ZJNU2zH4H8qhCFTpIKO$5c1R99za
zErdXZA}d(6ifkB%`GPT}VMWq>lG=H&bX69c0m-l*+<3Y{bKKYr>HfyAA~BOGPR(Q^
zyAv4#kq$rgxA~Nho0p3g;rwMSWa6zHLf3~#YA4bdJn^T&WI9F3*TXh=g3OG`Y(wx}
zENAAMe0>_*;Bfqy(gcJjzD8lma>O`biY16hODEM%hm0V}`QWG~neJoiC?oaD7Y=9t
z(EGoR4^Eqzy-XlL<D!$s%qLZ*Zf<5Jaj)uTlDbp)%Wct|B`j^KQD7AUQ`$x^?~C7>
z(tb~F#Z*VFnjHO(C6j+#P1~3wrG$)Rjo<JvyN^^osiu_q7p#<l4BD#5+TvyP9k|7{
zE9VQehise8YoT>otRRA3*EZtnF7*)6btwCLEydsV;ZSVms3+Z4%So54qFUovU*Mm(
zm(Sr@@Cw^1z!$yKeP(|Z{F6h(=g>nK<Ka@Y-O^*tlP6(qx;pFNq<@o4>p|b=<N_z{
z=qn=1*R%b_Z&9PHo%2=CxnVMMH~Y0z+HCPN*JmJrY|^ejpXbrpj1>ToZU-$}^WfI(
z2x&)<>3+rn0E_|4&6X|D216TE^6t0S=f^EJYenaLR_2-C4(NjgPy{kY&`1Tij8Co8
z8aeE>Uv8c2-n&2MwOeNQ_p0iIMomz@OS!d@SI0aPpE0}zo$oGkZ1_a=tmVipogxaZ
z^He$lsz+i3GnXU29rMjeX4=K7?|8Oxe9~0>wx9f{hp)&287~%iX=WrjUi2Cyp}TSZ
z*_krr%wZiC^0uk0WI$S3d^8-+L9!N3$HnW{FPS%%<uMG-b8Wco6xhlib|^GfbB)1I
zNR8-|UA@GfxIPh>`q@3)H^GLsuA<ebeiCF=VCa49nlr(rA|*}FURbCwc2P54If|S~
zK)2TL&aeh*5`mjQf|(x+51UX>_8*d+S2V=xL5iHl6NIA#C2E<K!zto9qY29j8sNw(
z3>iF~ABV$J8cbWKjozi#D$Ywd;Hnq0&E>`ZR1rpkCc5B%2+LvBr7Gi$jvZ3h-qr@W
zGOe-=v~bqqS-Zys1?ci%#@&4iuyyW#uOJ(zg8MGfvsFKWu<4B|cApA4V|?O|k6raR
z@JnyUdw+ixJFciH!EIPbFNemclvwCbxASiooy?;D?8ZAJoeKBA4|c!YrrEk87>YM=
z7^f*Bu8AsfJ`S|BTCx@~JvU)NHxiXRC?<x-+C*e3Xq!XKiV#>bA7zA8Vll<!AmR+Q
zvMR=dL1vNLUa55he8VW%r2}kQxty~8v%SQ<{QpsjG3AnAob})trA_lAJ5(uYOk<-F
zhAjK^+-2f@m!f^AP8UHzOhNpFLC?vBLA~(A(@KQ`4cT9XNfP~!E{Ov}aXynH3O%`8
ziTFCFm|U1T0a}1WK$gB?QE|f%sv6To!vuWcK^kk!uCH*W-WJ$Bjf9#F`7`6Ik`u47
zRg)_;v29Qt*=mBK-$g%LZ&Dy{sG+mhf#ik;lDa1<6Eg@^tGsdV-dqnLGa7O1kdW!~
zNOE-a>alG!Y#xI&RI0XF#W^{8NP*H<K^-L(6)gqrU1qH`4=yI0gED7W?hFT?oJf0u
zW@!{1M_IUYackw59$k!1A9b-~O)B~7z37giRXu@*Iq;U$wzuo?Fe}VeE@<baN}d1U
z;2qstVEpkrl_s0+j8!h7^zYGp<SKxx0^7Mqdle2w8sc?zCg6!Y=X%>^x%d5_@!{Q%
zw?B4On?&I52T8ExN9T-)%e|1>@)CI9=l2OE2_Iv^Kh|O&IRc<jk@EOv|3K^nJTG7z
z2AqaP`MJv&5Q1}=44y~k|B9i1-*MHqYO~fb$0vH0)Sxgo==-Tiz<bubKDQl1^qwc(
z%FST*&-9vt+*9ArKqmvyOh77`N&C;>)1RZ_PR4tlw3W{aNE_OkZrX)^j>J^^ccv4-
z6n*jK6!`oC`5DmMvv{({_vy&8hdQ8tP-hkeKl4mcj~ZhgnQ0NDKgw>tnJVFcq5$Ye
z-Z5!NEaOgqy$<l~l6sQ+sjKFNm$TS<zz(1v1Y&&W2FcO)_<S86%*j$G66$&u`vwO0
zoxZTj2XzlhW&hgnr+;xqq7I&V93a`K+EV95Ns&=}BT6t2RWYgT2}hzO?i2blLR9rr
zN%F9}4qk{~$*+Y)w5B4XB~hZ3D^AwaUJd(MV{NsqN<u;|Ur(&6@Vn1g3`dS!QZfH^
z0cwWXteY05vvNxnnf|Kr^I_1pQCQ04ORL0;gzk^7)1;xitY1Avqt3SA*Jmr0WnQ64
zVP($QM65MMIx;UO6ydj0@>SOO!V^npwPU#QzyI1UEr(>Xro+YAr{dq(Des{q<}t}_
zCpZ<=L&mdNprF>$|Jlmn>Y_=rb8BtHU=Z1VV`HZ6vRH}r@`V27yJxcnT7aG5qOW9F
z0}74|c@&DCKPyjkR<D+A;cc$5L}NjYn2OgH+is>pgzE=2GGQSnYl#UfqhU%I8MYgi
zpe;FwYKqzNE)UfBn)FtVl4j=6bJ~PEA8bd~TC!Mawtgz5Z<&!WQqwe$MYNsQ#M6y)
zL4_5yTX;=dpwG&Y^oS3$q_C7Z7%e4D&1C6G2fRiR_2au84j9H|`i(@D%ThnoDl(^Y
zta`fK7tmFZDE-A`BPz&wjL_lM&ZSJjjIqf|=s`v}v<y(PL^FRPS_gu-;fgx4$&9QF
zo@}<zJn}*mXRB`<$s;@pcJgd7HCj_`n~2j7bHjW_1hFZC=!!VSa(cCn3~;w;^GkZF
z@i|S1`TNi1y-xXtdpNY4v)x<okDH)5#{$T%2K+$s*>g6p=P1XJ9F?3RW7Ad0zUiaT
z*I@O~C>rd!TJ`Mut?D^?dD%!S0>2)&-22)vm-4g2T+*!4#ttdA$Zq0gpg_u}G-g!3
zU8{>7;`KtVzW^2n9OfeCt9OTAt{(y6;IB653wRGTsNA1kNEj?QWSj&s-SvP+J6JL~
z5TWv!;nzORS(Ucdu(#hXY~wJ5?z6OQ_V=gJ_!UI$)p>TtJ0XjgU8Cfh-uo27_|9JM
zS3t!v<xh`c(g&XvZa9xY7S?IkYPKp*B_d`}Gu)%=(B)fJ+<ECT@ZUQDI0>|Ju266y
zBgW^-X5rT#UVqyJlG9J;UND>KqGGI!JHB&x&uCO%xmG;@yx#v41W2GKp_(qGqs{Il
z71!084CI>Ha}Qc@P4=u<l557GdI*8I8b6bPeECng+tWRZ9%D+w5pWe*T;|8X@I*sK
zQ^_gpBaedMsA<!k)>+qXvZ_{~B&0A|IDgb$4%L-aE5KSGsAZE<l9kg!7t&Tg!8VH1
zpi89LYgis<ag~x)X3JPCRLhUGg@stZmPb&wr9f3l(c{9Dr;vw4(Sw+2*vc})xTKjQ
zupkpt$eAd`>X<?FipE2E_K9^!ByY+SeZB{?@fJ2Z-#qvD@l;CZ)naCBo-R!y9DCT7
z%S2F+HWFoJ(WZ9x>Gki0M2q40E+*3Er7apXBO=vZq9`6s{i=E0NDjBNO^W<(KdRQ&
zKOmsqB)I;21^Kk^F&(>q9)51EfYPg>v>Ba5h-pgBt>B~~i6EC;WA)P!j}4-w4Iz-D
zp$_F}$mLBw{l&?_TIezxM<i#ddu~^N=D?&Z6QlAWuIO7{j+x<mqF>J$ZTpg=nHLbS
zpA__HWfF^mLvIeutZ<=0rtnm?FH0QLHL}b?2Y5X;vt`$nWk{Bh<w33pl)gHq*hLNX
zw&><ORsw!{OnUzsTW2B-sr+fK1N*=PM0mk+;NHKAT5?soV0sxkS6BY)n{<QxOr{q;
z$;W3WJ0li64YM@VyEta@1dU+{c$pV|t`-u+jY>u!_eYmfH6<9`W9^Wo`Mb)rT~WI-
z<C(BVJXzY5tZrp<*)GrJFV^Xdvt6_>8=0MJO67Mw96yqMJJ>W|{9~zI|2yt$g!E7H
z3h~g}w9w$$)I#vk+GYU&{=U8sM7vhlRo;}PSa@k^terov)HD(nEUl0KQFoS2w+xAT
zX!IA$&F}Mxe$xG;s_heZHs8%rljVIvQ>XoBawgjy*+-ceW{~w!=fb1i!{L1W0+gP&
z1gC;~mWYK=CV<`mw+Fa70u+xLg}o|I<_qE*8w48@XzzL4AO&{k@V{FDf(>A6xlln?
zS!PzhjE}kf-dakz-ruhd#vmQNUf0eB11f3{V~!QH>T3VMzdaGeLx<<)G?v+IILSGw
z@j7D<KcRT7;*IX3C#jsDUtAUnR&$uKl+83%lb`n+5cd_bgI@D$yX!aPxPXD!uyk+(
z!UPWS^Z+Y(yq*3TNWFl3YPI|TgZQ4{)?ZSbNeXsY_8X|;H>O_#Tm5_~^r4McOz)ZK
z4NB7XxX+~J-eDEh=;~b_`DFxuSf{N0Rgzz3H<22SGYzvC%$}0Yh_BAtk|9WSL3@od
zyqz_h01fWq7u$25Khxtkf^aszStc-Y3ab6a+n-FCPPUuZOFUL5LG^SBhJ_kMP?m?L
z`~G7l8M^|rsBCajRkEUr%YYKmCq7WT_bdpTq=^}7W1z81aj<CCj6E2u1gTS(6}D0?
z5X^N%r<7b<OhQ*N`b#vaP$L$q$m6W9Mxp|DF(x2N(2>uDIES#~^pxGGpqE*T6+S<n
zO1m{^9e1o{9OWw^&n0HYRic$>lvFM#kc7I(@A^$vXh61{$J&h|6lnxGX`ekmHdVG4
z$~LKVT&(4kVAl4xh#KR8_rK0~k$pU!9el>$V!Qm)$kQfb={vr(S`aJSob}%kEZx(J
zoS7$|4DItxlgvqmRf)8oE`(DnFqUI$WZZMlCyRfO76avIO=WGOc@4A^X#+@7PA$YE
zr}5Ej3n=<Wm@5nW$J>WbJn!&wVLy$d*ptZ<@?8`)Orw=aux<2cA~#5vn-nUH<*fij
zm0XJb07@A-S@jdSY~NT+n2|2AN~n2)0ltYA2ij9|pvcIy=IV8%5XMN$ui<A`=szZ7
zw+iSnFD8XVOUaE$rl3JCw#)W!Xs{0_k6c6ZzYVf(Z5k^mTko_pp)(^9285$PGx^)a
zV&2%PPoV}9K3B7ZxFFSuF5#vMeiXLJ+zRD?QdwytA0@gdOQ-C#72Q0oH17MpgLlqr
z;iz|b{$JkjtC!Q3FDz8kotKZQP(ANq^T^^HD;E*r2FKpal?a^qvpO67_xNbldo5{w
zgrn<7<LrFhSueBnPP)c83Mw`B<9({Fq;-?cGPloJBdSq8ac7JsCR<EDf@><~4Sd@h
zFDy0w+{|PcbhR&se2fBZT8-bo+t`q<`geVaj8&QIUu}Z}$^sxA9)q1Hczpnj&&cNS
z|LpU;hhTNgDMS;W_|an>GO|jAf0h=^=HvD>-*?=6%i)v{e2UD?+aFdBmcB2nu3CW2
zt-Mt|hW+$nU#9_eKOccDcZ>r65Wa8KJ@oo^!PL@wEeON=g|mm#rqBPV;$n7y1MUmA
zf7514%MB*ZJ+1ez?`qw^C}8pL{H*iTwiaV>bX2*1V0W0`Sd*g3fVbn1%V^W_aBtHC
zkI8RMlRrU>1o6OmOhRMU=SY#lP9oKWE#!~ts_r49r3yKu&yDL9>dmW>rBe%2p%C~Y
zARRzmnk(&8Fvsp+uE21{9~yp6l()AE5%csHh<;kDVJs`x%+FLxU=)fg?yPT|D9Ypy
zhheE2iISsZs#EMzTF9)IKiyz5Hp@KHCz8^YH~)D*LGp||#hyq-nKX@FenK*g06B<I
zoY`4lX`4cvQi4Fk>WyY975b1>L&NFra6|x8SXdw?%6x$Qt56TNfW2<k;Y1qywDLEh
zgE(OW^hRp&LtW|{aAc|vea=EV8n#^g77X;6HIg5R-x?GYp!}WU3yLYA$EK(AgYj1R
zbz5sBK=3(=-Zo?tmij6F%1dx2P!^q0+`YIxyFo?Sp&oTD|1U1Z5G{BPO~x-rA~fJl
zyeT4{ZnRb^ry<7%?Mr>4dy8>ar3?;P*R+bsJVTzfL>i6^T~5SSRFvOp%k;WU`a7vy
z`UZxDMYYD4`(@%#rDXLXG3>+!=Y&|F*Nn=9$l*PS^bQd;m=IfkLOf@UD7dJs=3Ap>
zb9%g2D3F0UoC>xH%PCtiTk)Q{D0+WaKPeS+$l4XP=T)M{6oZ-|9Ii#>KgXnlf<8sr
z!)f@2Jl+hQ`45i{i{egf3JSwhe`d@?#sSB<mI4VCOiP9_JNjYODcdl9bm>4`u@Dq9
zdW|6B=Blix2+<7t*N6(I5i+IpU;{&@b%S#`v=Z~1^As%02g0=uMy94Te(Syb4%}aQ
zY&`_xHLb0lgR`Ajb7?rA6Rxf6XzuINyYkF)%8AvTj82gTecOe+K-E?VCvK`dBW(2R
zM;uT49aYCs>|S%#^P&#i5^YH5v4`Jwb*04*ug{wJXm{6Fym_B>k79n1KSJ!IV(wrk
zNAEy8&y$gtodRff0c{B9Ay5a9A0tv$&UoiZFVh3Af2&9Hh`*qK6Z(25;_biVQj_eo
z^PP+$E4GG9npuPTqRSi%O@I;peLinceOro6%g$GQ`5XQj#VdUFGiBQRV6j8>%`M+f
z+)D)~t8B`1nmMS>XT0Bqu7@DMEMn*T;=cWk`q;-wLE~pSyFjX>ulK*Uv&ze~EWfNZ
z8`36T0l`s=YT#b5TX|rXI`zIcS!(iuB^&n;p%XA;_R4)gu_?XV|5c;e(X!?&5HLj^
zWFPshFSeKV%f)9xh)i4+?I>17Web<!wuIJvpE!1Z8*T73E0%veN^beylFBRblnZ8g
z1gl~!Z1|m@SiYKQi7u)eD*39@95k{*RIJZ-%S!gOB2#$Awe4d7{2X(1IUEzEzfP@>
zNHQy5MUq@vO(|6iyNq9bLY63;wVH*Yl7Ex&KXQxs3boMB5**<f+zQ%ef?@W_9O6%U
z9n#4$6Q?yHNp$Mc<w<IbG2dK2_b$7SeH~~0;WWIHka61M(x{bWn2>y$BquJOYpE?p
zLo6+Z%<3|DZ9EonJ;IelDoay_^0|1aaOsc;oeQi@ARkN#X*jhKxX7NE$YEE?IDODa
z-jp14<T8h8<(^V2Cm)L3rV>5DEVU%FF~(hOncX_-s2!%t5AA?(BGR_32d&kWk_F*y
z4X&}=_AMf%fVCb|w{$?OME*%>ukf^WI&@J;o$Xjm#xRnhY)aaQye=b!T#qQ2R*X6z
zwuB~U*6vkas7rFd+=$}%zz=e7d%rPNf=`jNm>WWGrIicIf@tQ^>$k%5>v^)Czo;F-
z=A~(nA|)Z!!7(2uAxjXHag?W4zd+S6u?<TkrR8&9bG>hAQ11>bo$54{pZ%`g*1&!J
zaU<dfN8*b_-s#r{49}NL@he<b?7^JIce_TW;|8(oOk+OHcAvfJ?)**`=NVE6x%zB<
zkjAQ<P<Vng_cGyQ%9uHiM|)O6IB$cZf(8(GK5tA$29Ivo!JvGi$Mb#lG78((gAzqn
zP#))dmt&RpJMV|T3?2WO0ua^zClmsh6qL-A*L(SGiS3Hc_=I}|F}~W{<EH^RRUzl3
zT=!}2Au#Rpz4qO-r1Vi{t@mk>C5iEmE}b#qBM|#n4K&Ywae?r7pO!Ol`Rknj#r=vv
z>dba~r%<uC(3|KDs4qabqD6z;0ZHwPnJB6J7V+k5ihHMbOS_*wf))(r@vr3`A3v*Z
zc%N3W+@H*Lj^>3h#@@$NhOWJzxsvT-k4<B$37SH{(bR=z7g@RvZ;X&%<@6GL#SZFA
z3lQu4ARIj~Z)2L3A2~hmKK)ji>vi4Dj^r)-OS24ufj|@^0%XjspIrHT{0d@_3L;CR
z8e~lM1KkEGoY-}SPaMG#B}fHJg=1t+vRf^LyyXd1ij1_s))A)CNk)z&LFj;UQ{Ew&
zGl^ydT6t4Q{qvZpJ&AFKrk=}@Gn6pVN-tKeS`(-wD;kKRyKT@Ek-&E8Kh5oft!Pl6
zW7rrQ8+S)7iIeBXi<i%17JMAw30B=FZKm<if>6#mf>GUUC>vUET_Sb`u>`g;30-VK
zOHfyBWns<cb((DkXeQVkQ^zDmo`N%l$))Pd_sgRO`=R&~K9LktgDyUS{$8A3gJxC+
zn-|Kt25o)r<o|tZ(AlpqT##7_NrCM5Wq`g_^-nPIQp`Xk6IQVtOW5Lh6{tJhdm+89
z5mDhR{K40t<S?5A@8|Ty0<w!xIq3~lE^VjRXcM}ujcSI#rfM*Mo2~*knK@)<94Al?
zo0L>H%ppVxiX|`CE`p55;9Wmr?zvNQ(05y;2Zw*N@=vqYp&K`fs#U_!{3_1E%9yQ>
z(^)^<iQdFN*;j<c1>Aky*gLN1Qdhbh{zhc*T>pImtOy*!716#N_$q1H-Y~~=_1UlY
z<uM;d)y$tY@tLTdUix%{uzRPWXDcy&T3UoT4{Sjml|N=OFtI3k<fSJM{CX|*_2=~U
z?ko&1{YzkU<@bH={T~iKD}Y@9I3GxzlK}|UPj^PeVgLdXiu%VK(vPY;j>4zv-H~?=
zoL(X^np_`71N`>4)t`CYp2p;b3aq#Vepzw<3V2_T<zKetTmCH=CV*;ge*TsKCFSqs
zpLuoo!pr<7q75(Xw$iPUx(dZGkD}>uiq*7Hop|1d#_Q&LU0{8^KK_B<44UA2)ox8r
zyH>sgS{9w(-Che$w!!BMH6DCrZcdJDy!haXYq+Ldm>U^~<wp~ZdB1&;DGcp>T%{;0
z*bxolD_(aZmC{DZg2d=Om=zE)4cu_2OeaIEc^p!kxUxMGt0+`L6^+r}r~gONSw=<K
zwQU%X?(S}+8wLdF5)_acKxr7d8zly5q`SMN8ziK=y9Gp~K|mhy+q~aT*K)DYHTT?m
zU)Oma$7&K!wFP~dy)JgjtSu5nFu%=-|5<ev6fGiE!jOTw?0RtMSg3eGl|8aP$G^L8
zFe=UxW!2&aLDd)%U}C`7H_biW=kpy_S+^+mo^P-7=2DU^(j2OZy;{%|^d=v!=0jD@
zp(d9Emlts6dS$tMoYDA@-3YXEagB^X?{jrQ(LaIX>G#a{%a)b7dE(bU=L~-R=qr?$
z|9Q(ckE+zCaZRPAXNjZO71n_UdU#AGS2@1LV1gATiSn*k(YL&3QPIpn#FLGS`bCJY
zQL{)2P09$0ZO3bUL4z99Gg0P@90nzvqz19W+#LY}TOZ9cw|pHwr^J6h6x1CPKE8*8
z+#WmHXmyaXmYq;wyQZ7A^4i(03`f0!3+$qKp-z0;GOobZ@tw{GRfh0Bbok5T1kV<^
zjx7ZSV@f*fm|0A7Ob2E89*}%1bF)Y{4M~_yY7Gu883Zqz?900WH>r=8@V+0s2pzsc
zLUxN8sg^43fAT*VMA&{AaEKATd%c&uH{02oQSnIi!?ke{94~-YRnu=uSld}oDbctM
zvm}FK`d~{`l%Ce+K60&6r}<HMB6N;-6PB@<1CwHDG2%HEGA5cA684j7I+6+d95!*?
z^!mhyb;NF}(`#KPAcySN!;-gW{YZ&VP<P8bP=c0IwSHD)dO}%(rlv{)d~a`$@~!Ax
z#>CyW%$KkKXt(d}RwQq&B0E=v_nsT>pXmUa{#T=hv2V^|-?P0ocYt2dcLM73=Fr3I
z5#nnCI0)9&)oac7vgq%zcg{l4+$J6lKW}VVf58F8j2rR0rMsL#@GaiV@c2-9H4Q%O
zSGV8=lFFr@@jmnRAH?JrMf-cjso($3@^HH=1~tSh@e|2-SlG(_E7t$)>Pr|3Vs2v4
z8DTCweq4FDvR<q2Xeaw`+^V|iA>fbOLuSj!D@(Rw<)EHHba-yB;^2i5f#ol~-)?`k
zarH<@G;tY?B;~d%&YwLCRhN`L3W`h8DuE{jY>uyGf4KGg;y7s_JubXtV120mOGq&=
znW*M+v*C8dugNV|xZ6D1$s-<X|Ft)#*0O7;AjRGsi-SRdYE;~4st>1~F?(VAPp^QN
z6}^;ao*s)~#2xj_wtl)36iPc>eZlw{0%%NJ&llgOO(^N1qv_6zd($HL2IS6SmXl~E
z)MoHHqXX0isKf!U(lM^>M2uF_;vHd)ZjwP{lgTAv@>>njz0XneI{&G9G<+M~Vpm8Z
zf~ga;E%)WQe@9d(X2BLMYGDY})l{&Dz=<u6BE=%{$+HBp(Vj)_M>ZS<2RY7c{<;`X
ze(~o-Wb5&9?bRHS)YYyw0L;zUj9$#}Q>Dula9plaX=D`+tJTkHDUXaAq!6+y$V3!=
z?NIn={u1UoSSX0ghNZ<|15;@!Zj)Tk=OiBRj~&x4;<n)!u#&WxK4#oqwmTtp!$#%U
zPS2Kl|KB=|lvi;WRW6@IcD#n$titGX<6UC;uJBfF7#l|Raw<-9ix<L>g#b$|k?cc-
zZlzqy+MYdN(4cB-CBLt3q7M=IVy51Q#Ll8g?c?>UGmi_Wa)2<1o>E@fSUq8JDE>M@
zq%=`Gav;g37S1jgj?58&T8N)8>l^pMSLC3}kIvau<ab`nD(lDnkCHg2Ip+xbKmWmS
zXYS5Qi8envA|Di#o7y5JSz(lmFZOC7i0(8*^Mmfv0@gO<B`n3Tf0wdHQ+u8+&`fYx
z%>TU!EMvt&CW@?^lh4MK0QaTH_aGFSx4<9)2GIwB=MSb|k!)--FfzC+`uzGI-67eK
zIqvItoH`=)m=rg<7%SM%XT_3u>W)Z(jXxb3i!H>cU4g|5;7x#k<Kw7p)9!xi`<jQT
zPmZL{IUA&pw6UhkLnh9kZDt1CAN(AJGe2Z0-0bYIs&7K;Vc}R9Eb9Y(XdYo&zc)8&
zbN~H!*aisbSNEN`UvMutwMN^I6`W&&oiy<HvNet!>6{%WL8ro_&F^o!{bPp@O<0Az
zvqV_nE^FqBJG5oZ0U#pu!T9{^`hm5`O*9xJbpCsM033CxI_m$5azm%|>0bX?DcvjF
zOKdaqkhkTg40BES>G|dHHNNkNluXSJvgvbTCS)+nU+T-R)z2hoF^4B~$0Tck`@G}*
zYQi(PgAhKc344GZ#!&~dSf|>YpB2sIoz-D`BtHKnV--WfLTGA<KEoSLt>ni4kgAiA
zt7jkub0@V>9u7(h4VkZenG-)tA=)V>++{z@SH){}rX<brwU~+!_6>{6ZK)z>@i`Q}
z{G<~Xr)sEcQ{l9|C$)sB1`X_nGvZ7#W;4Z;3={RWL~0RM<#V}3zY*Ib|NV)}V%#-7
zVIt*PN4^D5wAt@hz=gu5_pz~5>n&yb#(sL{>x=blhXXdI*^8u<`S6Nw#;XYp7w{2A
z>kJAFtSELAiB#T;tFQmv9yvO>LZtSnw(cPJ5>B%#TUOJ@vpN`hpCme+Z&;(k@LF$I
zTaJ<_jFBNNMGfL;V<<$10-Q7(`WK<4ycZ3*0YOqJbGMre#;-=S<l6bFX0uvJ+-?p?
zc5BSyHDE2~pZ+qybX>vr@{0YWy~vQb-i`D1(XXqO_0*$E0+@O^b4^*J^d+M<Dd~<_
zY`{PL&U^arE5$I?cf{yNV?C;>kZMw2q`Aart804k*@O((gvJdai{!I9qc%do78kAq
z)iqI<r1%(`TRmlWd`1#;YxL4LH@N*@0)dc>sp&s4&+Vx{_mCdDZfP;nepyoe8dkq0
zi;ljp{jwX_Y>onoJ54<aLMWFLJ=?6|*k6eh2sRvBa6NF&fCLImbe~#1*$395F@!s5
z)l?c8`zFQ2Ag>FAFMtK)=GI_^6|y%c?#19Qh7jqD{j3$;KBK9-v~scibO00Jh=2zE
zsR8tKka~(1GGhST`gZ`9^O!VlaFa)H05DJyNqxIE*L}>d*J@wCfvPcp4@lo9O7BdF
zZdv&Gb$I$YIyqg<YrkHIU8_9_Vvqz{d$o3zgmbVmx*FyDlGk(q1Qi}|W#?g5mA;$f
zC1C0WoYK`35irN*;Y2HOm2#SN$!?F)BSH}NNSF}tB@eR1`D@aTu9$4K*5XVXf_e#@
zHQ@69xi)7fpGjRCFhjt-%VQp0az({AAKEOk_}=cr<kQnbgxFzWp1I&H|5`x@*VyL8
ze4d7aaGNFpp52f!J$|!*Vp_0<-k!Zoe5rh<@oJaIpng^3w-FFgyoAYCzliPXE*}<1
z3XP3bIaG`0c+pY7tc!_<8&k-IwZ+e{DAlNsF*TpO>0sBFSa^Q#zv%lSw+5S7lUk5U
zedf}Lh{!SJ4+msjh;ZtUH_vc~c@!auTGF+Bj!UdqB%jNxZlfK1cc?>B+$$^XL2D+F
zmgWd|w4|!O3^!5*#X2ZsjGR7Q8duWUCxRWSt|3jhE<xo0W;J0POMM}q!lb(hgR#a6
zIWC~yOZvHeZWf7V6Z~~#uFBkzNp)Sb)t8t?=<t#(#Wnr%?g9~SN@HahyP8<));xEA
zNa|MH;_H;8VSXa1p&n1&NAu4cV|sPE=YK@I)&0DHF(82G#f#nV4<A+Sv73pLLW{Xg
zbs(SmMWioQB>YboJx2HBaA{wd@Mp$R$!Y2rMJ{)Djyx8>*wy{^Jd-J_m}JNv`_rp<
zAcXIs(q~PCCBKsfCZj>7FD(>~j`PuhaZ=4g?Wheh9yk&eC^}16d2!tj@W#8c)b&(1
z{PeR}0zGKH*zWoF@G&o}BJMy!rY5g&KNsqi=X=-R8+{W7qF5X}h>X`WeohU+_M{GO
z{!yd<=k9NT9VSN_HGq=xa}RliZiu$Z?n>5ao@?xh@=dk%jBGvQb;e$nn{X8IkPEk?
z6H-o3LFg6^YnX`$`r4+&P)6*gRsF~CFE;g-h&G=Dj1W*KZbKd3f7Zn?Vc`0Sh>u*@
zcW`hJT5pG*i&Jc3=P-dR(m@QD22TTzp6CP}U~TR=D|c)hfG6dfi;<VQ^W;+_7ks#n
zuSmK%)wOA*KLEXz%S3J0<tV5B<5g}c(5(T01}RMM%e(!{{V{#JdjR|z0pv+wY~TVv
z4_sqPr5t5|84?uyqNV~)-6XHZ1*3<|fitdu)rbJ$4`|&1urDxM)3q?G0A3gH)~<_G
zdu)94;ww97y+~WI)ftM0|Mepb#~cz$`mf((8u`~wJ9fOlgd61$KX~T=73PC1tmz2#
z^<2{4-jl9N+BNsTC1t9`^sXDfbM3stAvPq>2hp5eW?ptZHO3tZ=0o{7OFcA$EJ3`U
zKemjBFq*mqbUPb10;o?IJXxxfyt-D^gz<|(!=h=ZtW89uWE5!&A*MA4lMMWu#izNE
zaD6*7Yl3*$7Q=7ZA{xG__ApG{F_yZD9Q14j%%42ui%=s2#wIjNsE?Pe7BjZ!r=lnl
zOWpa`DLWV*O;DYfwE%;+kyor6`nkteY9Y}^DE=&qZ&%@B1+N5i&v|jOE9katB(2>2
zUpbj8`<cHa<b+3eo=$=v`VwaFofG?AWogo&_)o1?e9UV9`5;eml`4LNNN%@D4PE#4
z7tk)mY)2}1-aeymzq;amWJrHlnP=un_KVeOTJHP^)iD!tLCI<A)L`#~%lVVzU&dXp
z?5sZ<q?jj&Xux9e=v<vVb2m?U1lbG#6g%~{&Ha$|e&+?ma&o02kh%a#s-b0ollSXz
zZK3$%c>u0-XBfL+0z2br<582LyRU<92(DIARXv4@+MFr`eJ48?sABK^dtK}Ll2Z1~
z1|BzAR;zJK8o5ZLpwe3Jvuj*s4Wklq?Wm>b_$_*~?X(NT6gfkY*|GZ!iinqI<1&cz
z&EK*p{AgaYh?0X>GddCO`CU*(X%ilr0G*$!|F}IMWUzDT#N4R#Os1Z~BwuE4K`(dL
zmWE0fqW<l=oU96i3rfi9#WGiKpOV4`ooEKjJhKW<5;Vv+->KfvbP^rQ9=!(V<rgkB
z#X`-$P!o;dK&hl|q&siIvAUXLtm;BpE`^2@qO43>q(ox!?2=Da6b!T2oVd9}Ll>cC
zv9(?KpY0G>gEX-N*=_n|zYVv?@&+dMNFp`G)^VvlxqU?Keu!2kf#V(^iHeJf<t^BP
zOp!W3<4oy&-2zrukb4IlUhpB#982W|o-vt>65%`W7r{TUwy8;nuJoy+;`_6la+r%0
ztEb_8{`KmSd+YzZ)yh7<=96`Y(ERx;rS(SxzxOR+O+4ti-Lm2Yyp<D*<S#*e%r79W
z88~P&lKx0`(5fh;Ve%{3P4oxDGr#pCcQJACa^rzsZ=#;BO&=zK_Q(>F{mwh>ZfZlq
zlqwf_UTQ>c@$=6+`JU6ux!m3oimL}YldDFvpH)+Fm6-MQ04nszW3(W@(f<6j7p<v+
zodMPo&KTtA@lruvQ+EBgCW)4E()mgIrQ}tfns-ad%UNBABo{7Hrm9+d2d4xw6dwaZ
z7(h6UWQ)E>?-Ue?Ws)y)VPP-lTg*?YA@hd4NYqY2wd2y99W+?pTru_cRfJ)F-6G-R
zy3E>l?HwHa<P@gE|4})piJlXmqN+b=Mrd(H6`H0yV%u$zY>j?IldTHVh{a2f)iBt{
z+I<&Rw!x@7n%b4@#cZhe=UwL@ot-&$Nn?`Erxn-kZ*2*>yM{gcZ7jAe`aHUzzm}!(
z(l$XOb--P5Y~N^=d^RQH$Q5NyoP%)gY=?0<8qGeTQ2twMN1mX~9EIo6i9MK*q<Tat
z;nBKXZxf7J4l+v=ZO{o!U|lkoc=>!TAelMBWq+m8IJ#nc<MX&kW07*zo7+7ya>Spx
zMpI+d@rw6|pd1-q+(<EK#g!t+xZ}r~Vla7h1TUwZJ9Up&)%uPSwVxm~rn1Ui^gDQ8
zCxx*{-d25uy$LaA#zSfuilvz?b#{kKzg7B|Z5w&z`Axv*YNjzQc%;n$cM$>uO15V(
zl_hbMxFT;jbZt`%IcYH7T0C(ClqCgctIkixf8~abF1~X+{_+ppPs5hMvjegzmy(Cu
z+j*#>rv6!^HUaQZ0j^gJWYsI{Qeu+*1vI2ntQ|y;fOC*_WlY^;SB3NjQ>NRr7dy}V
zMMq6}8zz?1-Q*h==gF}1oa}64b~H4mUe~rNbv#6PO+%u(X9LGKyo~DU*Ani^;Dcx?
zL>X^sP|b$5MsXcJG3&AXnD-28n*o724_EUYPpa?(CBnYH)DXG0-LCA4j;cz_@f8E0
zFZ?{yQJp!vJ-U4W<hUiuHgMXTXPM6v#5#xrF`QL?%7QxIE?R(*Ddtk!^uJOEPvp~3
zZZw$_Z2qxUhY+Tdw4UN|t3A{E3_rH}F0hBV(gZSaXIU#(=gm<Amyiu*1_{ZxObG4U
zYch>Rf*(8=;(Fl6(piPTa%xBU&#fekZA0lJ!P#F;@?7-go7_iiJrw`6Vy?4~;gx}(
z-Av$et%MxZ8a7js==(BZ_>YH^^zq%-Fx@;d`M*-_Th`{@F}(-))ij!h+N$A61!>jO
z8t3OBUOrc7PR>`8IyIY}!Ky`GER40y79zQV>zm-Kmg4|5)9_UaQ#3;?ahou?^>i1m
ze1<H#MM5BoC3iFCHxgX?BM)EAg=_{<3Z6*7dbs8=m^>_6A&h2uooerdn>KDmY^nIG
zA`O9qy;#&w3{8YHK3>k|<@3f!0>BG_+j#N$1aYVtY0|sPcg&f)b(ly(STfbwNr@m`
zJSr_{5JeW_syLg&Jl~82S$x=*O8#3Eg_K0z0}GKtqhvtksv^Kf&EzUYfex`3hF3eO
zTNRo3yDN7dFWa|FH=ES1Fr9jd8%|P8ln`s;39l<P5uuL2m4@)V%d<=)NxFUq&NADd
zz7IV;4)=)=VWA8)_@i>Ta}?gN?h0#}`k*%J)Utfqm`?ZN``sfGVfEcbYHq10=o)3Z
zy8C~7JSE-G2y=~><5;0R70gwKwS0Qjz)==1^j<4%_O@<;|CWw!<{0M(HY#y}MLaLx
zB!kCWr7L}|Ns27)00ebYe_q4n?TRo&JMQ>|Z6TMWJP)}1pDodp1%Hhq;lHM`hMaV=
z+<BEuzxQy@K0K@$1r#ea)+w1XlhaXYmxzPRUt$Bl!v5e*AxDW_y;sbC30VDt(Ydkz
zMD{>V8bP=fi{2r?$i44l<N5*@4XWmYvl#GOHK3Yrlg-g5eFb0ZPMsI1z@A+p!65YP
z10Wn38X7kF6ks^Yt}H+etz%I>VYVW~ZTrXL70*q>7~KcaHL3C?pDM`ikYe`8jC7bl
zkSV`YFua7HIf>+%x5I}eu#Yq-(<s$$utt<er7@jeK(XRZqmg}!<$c(}{|UGU!DUNE
zRg6?i6|_6-da!ASZ+yETO5E}h$mV^$y)pq+as~k(KjyY<eRz<qQU=<oBbO%VQ>Ymb
zA#sSo5eyg$!`UP6{yUVOG65p8Oc!8=02~*f{n|_gR9~xxColP83ks`%5!q+j7XYw&
z@6#1Yf>C8fWFk7H&gUUS@x7#+RT3ZWMiuDhye(q9#~^+F<cs9(x+eVl;h*f0^q2r^
zq6&=*VNJ|yndSHY-C2OQ5{XC>v$Xh5e}mxYNNan!8eBLhf(Tbz(6g!c%b!x-RnlUa
zAt|A|^wQj&qmC&?IO8Irc1e<tOgT*wDYx*E|H#n{2Z&_(rHa%I;#Z<`3onY%O8YCX
ztCC}jc8)6RrF%@aL`J&9wfO6h-I1TeId&&-ErnHtbKYdT5icIrtI#kcBN<K~dDRW_
zWlk+qVuPzpIxm?r;?+uYcJC+NoRPsvy?v<)!Wv?oa2Lzx5r!Bpn~a7fNyTTf`N;(m
z<l9B&5Gl*1G(Q}*jG3P})1+_L6$kj=nvMpMknk5(Mhd-+D-}0Lar1=8LWk;;2^AuE
z7gQAp6^axzavqY-ls?m4TyJ#9$FmyqOuKx$>7wC7VVh&Afx9-DIEfY6vy9ViD(Rzi
z8Sh#QJ5T2?9Vb|`z}^)UGr#8|2jvLoe~42ru7q6LEu73>a#5Flrpu4aBFpKBl6CcW
z9ekkkSN7uD{<)y`%ibX`n2HBBcl<j|-D+9YoP4MbFG)8eW{2l9UTiuqSIempYa~&<
z*`z=J)$5yT5iMUu;Uv=l|Ln6t6RHF1Ey6FmzQv4@W{t7_df4Wi=StOHnoz%IQN%W-
zKbi<PeSl|ngJ5Kzn#99pGBc!J$WkO57-;w>cYh5L*vUxT9$DP=lY=00lP(OBEs!}8
z7-hfVYvA#B%b9kCq4;~cb<PejvfX%?MVB|MXSl`!0sx34FK=yqJ%3n)g1nhL7`Oph
z%Hw2-p9mkfIv?2}B&4-FTfLT8D2Tyb^qL(9cVxw0@yk+2EzZOl3@MVe6|1+qHx7Tj
znL%ss*RLuOC*)6@fpIr~ZBOOI=qBjjZ_ADd@ueZ6qta{u#DJYO2#y)GyO+lvqC4&a
z_hSM3F0dk<)4z}pf`4c>eGl~FKoKkHe-{UIx+(QsU_e(f*1Wh1Sp?!ExZmw5KRI?Q
zm_~r7{l=PEMfS^*IV5}Wm4%7tGom{ny#_WE;BhMNP(wZtFcRJI|6(_fG;D|T)wx^N
zuI=U)e&)z3gN*PWU?;l#{p0Z@V$t%MCY`iufFfOS<UQ=fye56Kek>TxSo7k6`f86U
zc5{R;3Pmi66`j?RhW?jveRma0#U-hC-uGIg4$s83mNP=mt=KWjL%~*J<^7;(+jIeN
zJe11hr~bXXzDz%x;kj;>UINJQro|@aUSZQ)bG8xM!Gub%7_W=xLGzTQI@z6PMoku~
z_!(Er#E3{!ovlAr)+xhgfi4lXAIHj?=^%<lY>WJZ*t$uxzIQbh$p*JPx3$-zS+s(<
z;gF5~A>Sfn^5q3<aX$OsqFhe(Dn`g((|xfNE<LsSfrQ!MQGs1gw2f27!ZNGK^Sh+7
zWN&Q*d~^ES*TE|5H974}^;IS!xqs5|IPl{OGOyWZvn*2z%wrn(C}QxDKB-W(#ZRB&
zh>Jl?e%~R^^<AFq5kdWyj*Pq18{5K^&DvG^>PeUy2a`|@QY5_}XnfIWzq;WGLM)rJ
z`F=J2VB=RskHw?a21NvwR*whP2R$AlSde?{0$BLB6)m02`^NFba~9ipwad4US#<91
zglH#}^>3c)MDuf{y6LVUg;l_p7v{>YR;O3XZ39f2cpDz<$fy+z2vuYf?hqDctJuf-
zlv2pIg2z=qc4_hlpsGn+R1d$3H&u@arx_$cM>B}8*Plq|J)lbj569gYxy<49_1tSW
zU=Dc_{4xA%=K;g%Ca)mGHqR4i*Bxq*KYJ*u2-wNHfMxr$a=_PVx6dr$0P`kv2C&L$
z%R6y^XzX-$Ekmg0{_DO8{b0;VMqabytS=$o&qx!^62xGH=5btQLgbA)jZfa;yoHMY
zjF0y7@ob4(&8gv<y&S2Os0~@>M?7qE_P?wuS|L%+ok)D6^;NOzh@k$9;1SUZFgc)n
z1ZoWiDd3kF{ry`cmB+Z^CUC+5%l-&x8>1F8o_hQ%_Jcb+3}6;GSt}taS!FPlw}G{!
z%3h56rZTC*U=p97@D)9Hg#g6(lT2XN0o@JPis;Tp*G&5#sZgYB_#X+`Mpc`A5?a2D
zX!q8~zB<PY^h}<!o{~>?2XhzwRN#yj?>?Q4E8&&9v+6D1Zsh8DM%zdJ7F=V{FvI>;
z>oaxs6Z?lQZigFU@OGagZ4^a~j6P3VHc<_UWQgodPjDA8hfun>d&{;^nXAdAlYEDb
z_M_n);h_CG-8@Iy>;xiJ3&Q+YyS=g3GmPZg{MJ%musJwY+Iu)&I~Z_E)>ftot^2%7
ztE-v)t>R8hz`{K{kYM`t^WwW&JN4p1yC3GVQP0n4f8&~A2oE=rB$hUk9j<j2aj8bu
zztmSJWOGNQLa#P;A2s7=Q`al2N4DmuX87qIU!7u2RxGO=>Kd3tP95)#{Yj3M=^yj+
zIA#|t4q;TVRk!fj=^&TKv=p}96Kob~k(|~d5aAq-LbB%4<TMps{3UO!F{r)y@m77)
z`@{paC6{%jW4seCYtsyP9x0?O#4qg9yB|oc&$_0XCw}}UY=&a|)qN!q`J)-?zUBaK
z?V#V-q<j*T?2$5JA)8`(?Y9Aan{6qct&v%M;&8dnm;GJR-+6rGB5BR*Y!-H_+#q-c
z0r5}aGKKv~2)t9jrgm_<s0y3?oppH`>(Fxzwm{jiLavypnkdGk-`vaV_X@8$XB(L&
z`~^%q#x+S&PXhZ6s_Iz`a_;S9!avU}2S7fgF}X{(`dnFUG|`8zhs06yL^drkkq54V
z*l-l@;(nu*0e>ysliQ37kfVY4Y5}n~$iVk}8!@Ce5J<Fq>#${Stk?os%WM^6d&CP{
zagqCFX$~q%O89ILLOL2LYN$U*`@a@+Z@w=baYQwKm%!YFFWqWW$PUgk3%O*nzWo8T
z`ei-x!nf5v<7gp>i6ig=c3Ug%U*-sjh})=W=YxZaIFcbuqDe$@@L@oD2F>*UnU4S5
z9D}OS!Xenx9Lp|!ez-kz14O9;cJRsyacofo*(e;sr?b>y$;vlzu=pwuGB&}!k_5xI
zr7iJH;G8{Yc$$N?xrJ$z%4L**;R=L~Z`|dW%R%f`eY?vs$HtQl_BiQlWq0?t2jmdu
zu&UPrGXik_w6mpqNzr`d-%CUxRw59IjiWNecG|Q${t-U1LSc|Y`1~`&p&ELSyOaQ9
zAf)V$;AnYs^I-7in4Ow+ETRa_p@co05z$An8JfU-5Uue#eX0b!1BtX#lsU<BV_gPo
zXXvsHsbQhbn-{CDxGc=Fn6nF*hq}t#@eoDZnE7t9@=Bh0d}R9pA2)Zyg`rM4tzhl6
z`t$8qob{52GlfI$5;Tx(p8B;{Z*I|;;o}g$x{G$A7`Fd@qH1d<59m>j&eBH**%JJ~
z{q$Ha3ggd?W{=9UKYFvc3xBC~#Ow|ylyWvu$k1=D<6!P19O8WS&exio#IuQkbjdXO
zD*;Ndta9_dNVRUCYkM9M8@pY62`G?j#%mRE=yh;6Qs66K(Ng=rBouGYR@bdpsv>5_
zV4aF^Ra-Min>xBN=W(*HZCEMUh+J7&W!0jz!#2<aUpX(;ARDA@F3Yp?iKV>eZ56qB
z(o16squ%x3f0&FvoY4>tIe_o2Ud@~O>(^af%y4^M8rJ7lZ9H8ftde-YB&|y=1|WL4
zC$rvNuQQ4pBiEn(wl!r*3#YbWBX@^<U!-?QtCv^am+q&WWvm82yISkN1#;7wvm0Nd
zF5l$BLMAZ8sxHR7wb!n{a4`pt>*J*_Oz<)ZMEG*}G=T{?wHxu+4e}(J20?5*`NVSf
z`hJHV3nP^FJpVKxZj+xuUj{2d-&3LI`eq1s@W1|{Ssiff+3^}#_n4)d4e?8@@s1d>
zGg(?A*fw{#%ZUrA&wWOiJf}c(7W&H6s*tG|WTosjtc!Jf_xrc!h5|!iLXT9a8$cys
zSb;NHVtW<!?{e_0ux#fw9lJxojs0of1!BZFBr=@jW|p!m62I&k(45$#KK;#j#R^Ur
zPe|A$d*f+JkRN-tdJ{{C*ed*QO;U8iZOg)-|FC)OgmSw&WFSumTN>|BkoxO(^am}2
zySWQ8Q7MXAB1|flWyjppWV>H49wZYRv}q>*cQdat4d*!@8FhVMY5H5P9G2PGQpa~C
zBy#E8`9j}IdeR8<1NK}SjOwXBr@dAL72@fm#+8lD-^LV;2Qzy_TN|e7TIdm~W{foX
z#qw$@5k=#!_4|<nuDY+QG}j9jjtE&WAS(FKJ$Tux%ELZuIzugmYO&)nSJ_>7g=7yQ
zgYfOy$Khno2X+z$xYumOCo7qs!<md?sjG;Gl)Rte1gZtmvPzWJ(P&;JfwRxBi<oV_
zjxeMRc&)TA-9ty*vh)KvU`?yx-=5(oRn()(C=nq2aQ7{;+Ub3@|I<T9dq&dGG?GyJ
z&q%MqkOjnQnr1tzc-XDIS|n;e3lLfg*Q{WewJ0xeBqH*hw{?kF$Fl0uud-V<nO_SX
znaMb%&qzyH3K|mJl0eC`IP|%(mRA+g0iu-Hl+Y_h$k;4Kv;lO3zYMvMGEles_T@|+
za*=&1ETph5)dn1)%LTRT2MnJ&i@L1%5{hkMS5+NL>m4TND*hxA!E?<Cd=C|oYXs6j
z4jhA=rJ5u6WX9{PGCp~g;CakyWbWdlCP*|6d(p>mXTn45Vz;El7OsZHfQq~aP9fRo
zVX&_TV;sx>5k-EUZyxAd5A^L8Hyu3HkLQ*2J2UwiL2)r#0|a)34O^aafvHb&4$9g*
zBCK(hR$V#mYfudY!_^Q%^2(B^y0)^gEj($b42BGiksev&HrL|E&20$ImoT)NgGXtY
z(2XOaBr-j|6x;VElmni?+Ba4>fzSQl-POF&0_0H-8>hr<<e#%<n#AsqceDH;=}V-P
za|l2q?GJKax4zu6#exI8297sKXZYk?h0y(%w>U^)Zu?BGftXUY2=Pwnv(7X%se<N|
z%Y^v?R|72#4LB94W|SDUxrTzL%kDThMA%4T;+UWMkZr<cVg}hR-(nM+12<kGf1V1#
zvty&X=RXmHwI0Tgqw{~ApEew`;?Z$QE?8}K69?66M!7Kt(XS~s{OI&3<$ZXR(JFSU
z%ETms2vcOFcNcG7)DKV;D8r86lDE*}A{uiJ{*_33ers3HQGWq*Cnr1%cPMuiKUaw$
zL?vovB%E<<CY#h{Z0Vu+nI>iAKqQx8!k11*a1pw;nae6$U_EV|sYWJ%6=|R&dDvB`
zXuQ>AxwCKW3yNkO?c0VItkDI|Q%g!QYY`Z(t=Rxqnj#7<zA0gs*bxQ?LCq-AH@0|;
z{jtu;J4dR)?{CD4h{a9{-viBvOi_gUxP|-ZbHm-Le0H@YER%W0J@Rt%q_^nYZ)9hW
zal_w;nX``t(?5E7Cq{5%<?f*ki<#9y-+V`j&`5jD)~yzHC3-KLb=#pci;tyNInK^X
z<m)hTK5Cf~U*O@B>UY*P_VX@HeMiFT`#_{6>P8LP5iCbLDYo8SpZV`2e!dJxgf^i&
z$(j|nLXIXe#gVN-PU*On8|z;1#c-_=sk<NllGCr*Q*qKtjLL=R0xlJ^GaH<J7lgyf
zF+JLzYOIEGwWIk+34lo*!@69jivn*|1<l#hq{dZC9SvP(NCA;o8j}kuF(qe_^yVhj
zHMps9LXa|DPga}S$&P@r8<7X0=!RhVbFWEH43(%YDo5-dmL?sdGT1aL77_r~9WZ(Z
zuHI1U7tPJhQMx0mYcI!cf5ToCI?o@<@0UUFo~UX$?@%{3anr;dPrxtWiARlV>*@+%
zh-E>DL!ivD$U#To3~WL8)z?R7lxr%H5zhzQyr>@Pq%{;3YwJ02h-AQEdB}l~ReP2n
z0A)7?H_G;P>Yi&z;ZfG-Y_O}Z)cb<;B;Nx6mqXweJ~q#?qyU++lV&7k%r#>DWh<Ci
zR_b;S&NhFmhuocW%UaiTGGm%XpNrkQD0Sx>>lii)u0GEq7>E^AAl7|J`F0~GX959~
zz1N5v?>E8jL1U+OFhM*hoM`vN`1h$KNS6^b%%HpZeySNq6yv?ei)YqC&vHgrt@-6N
z9n0MSe7MTm%SdnZ8I+@_Pq=uDwYXBphlq*-V`2~op$TH9{MigNjmfN3uj?!$CIuor
zk<drEN{vdEg@)RZEE(c+FcOP<xwQEf4}Xe>3Gnkc2J%;+=)I*8Dw4;Xj{nU<!SM#V
zgW0=K`QU%|h$eQNHg)jh9oh=}i8xu#L4RU&_2pw3qT-zt!#AHTO^!N|EBY@<^eh^G
z&?y*LRMz<2@h6aB?ccsrP;`Ii+(UMt)>I3geZ1pT@7R#RI!f<;9)(B2${kdXYtGfo
z1tJA;%qIRET6XHDs+i^@qqC`^4$zRo1;)ZGUiL_F+lCMkQWg`hkEvq1SqI|0jf9uC
zn+Hdwo-3P2v$5t0t)U5&<}4ABJqB=hWm5vNDxwdW(dFHu=1ou#kr}B^0;P5T1HMo^
zp6>g35}x-PJp$%UkV1An<$N2KuID{%W+pME@tg|9?`=k}s(7W57W{F4R^33vKXO|X
zRVG<B(#Jx{<29l5oCI~36dzsCSQw3tdVF<CJ!Y-(FJTzJzpUE)1Os-k)CL<Y4lrG$
z_&nrs7;NR<-|m8B0!E5ru^?YW=B^}jM)D`FMwgQ{5koc{dbV%4>m%w&tT29f1ySST
zvy=&0!Y0CHnM8P%K^XF;JAlw!TMZ9hMw5c*jLy*Nd-EKKr7M~AkmQISg;u&fVvNN~
zCT8YcQKkt3EgA$B?{Z2Gc>I;76GEr3zEZ&-)tg}f5=hGSs1jign9<P0I?R`)7zG*Q
zt50%UGY1EU3t))>@a6$<R6|*u2Po@52mE*sblsu|kNBiC(6G>ewx@f>1!HF^MtG*Z
z0r`%kut$LD!_t=>D^@cEPlik8tXu>wzpi)H3=UG~%UE{XrHSi!e*OX`Y;fK)j)*u6
zD8{Oxp;5P6>c|X((i}z2Ig~hTK-`Ci^^3`?BIIgv)<UNCh~8Dga%O=9au#-u;d&BJ
zertcvNDi#uw=ER7f0T>yW$TT&!i%hyPXyDW>kKqFkQGwm_r;spT?`OId)o*JR0xaf
zNR41YNNmW(uVez7a7|^NhrYecdS4ZobIFb;H(jDq#+q<)!mY^0B-QyZoJzFDCvcuU
zIm_;dM{9oZzIOEwsuz9eGe&mqLE<t(JgKaUeFI`$UtNtMekh&_W3OwCK9_D%alVn|
za0PcVJ|=;<6OxSu9tL&ETd-hobj3Tba2Dk*T7po@6$#1uuyRHUlk>G*lxkzyYrcV~
zIjZ&~D2MjcrA--RXV<5_R!+sDUH=!@e8pwzHNAguQu@O~n2;SS#Dg#fi8K5y+_@Od
zYc@WWl$#lq4*5m)EA~ii?I$Z^4v83*soKy2XKJh4E$efEab5F+v3R`PlniCIZf4`x
zWsjYimjzL01!8vvcKC;5wU!12jjTuB**VTRD>{n|s#wiEkdP6kgjVCaqU3Kc1Kl3m
z<LY~9Km0PjJu15Er&d2;Pm;>qKCN!qY)9bhvw0_~^`|=%@EJTFYV%StUovu~fIko`
z^nx%0i2vk>%D!}HS+fM{pD3^JC3Kd764L^(?PBS-;!;f%##*BWAJe7-JE$F~)`<a~
z>f_&sC(E+cQJ=MF?{0gLS(iLnw_n$W0TKPDDXbNk3-uF`LF_Zv7!RM!gnlC5Q0N)D
zRrrAyG6ES0R+ZrdQk89yx!doEg?T4WYnrwhUG9X@Eh1Edp$o820v9<LjCl3?pwoi`
zHeWnCgRLizy%)k+P*A7ujV;X_VXVq>=?!-Kc>v;e&HK{hCC57_gfs^Ij`DJ>2XaLd
ziT-B;5V@Aq=|}hfoCm>KW6*#Yl_SQQ58iY0@J-KY+Rtj-U&#9Rq=dZwuv@hIcda_t
z1g)Xr0{Io{)TwFvIlC6+Rm>HwJTCo^=W6x$3HBnTMm!Pl?aN16mjVuZjM4gWQg<|N
zd9I4I1f)pS>x_GaEy`mmZVbBAo!m={8|bHr0E)NDp;{Okl(O%_2Yih)N%Rp(LtdIS
zG1v2>_o}8Y&k~G9_tp;eD!WR)tF;!$QmGKA>Vr(pifIaeaNSHlrv9OdcKEsYU#2#0
zA&ah^QF;92Elctwqfvx#_Z(cCkOU*^l&O(9#ckdm#UP?NRJ2ApD47L6UmlAAMMD{x
zKtM}g(LD7f1`0vjHIxF89WCCCy}T0xWlKseVm4C-6C+gKFrqobGCkj=sE?qD#U<(L
zzha!tyFog)KX>@qhu;^^`yNHwqT|20XIY*KxU#sQC)Ish<P~TWsdCrJ5qU?HR#Q@S
zU=Bwygc#Y@sf9Iv>T(#E^SWIoZ`MB$v}%YJawF%Q9=-MU=#wzc_DCzZ&J!0)FFya3
z*U*Heta=pl61|GwgnL#do=j|S3}{I<RVOi`5=oPoIkn|YBMdW&5J0G6v7RtiD>(U+
zaxyx}YJVhj??~x<>H`>_nA0o=E;*c>o-S&vfC<D>ezz&|G@*WiNwZB#O-6Cu8!4RN
zd$m*^0+2is0#O@qNEMVZ4Z<}5XjYsxN<~54;`<)LWl4s~jV}`dwnUA{d1Qg=HvJx8
zC<$gAt4(&o%Glec#^%eCtjA1}cT%t!ydk#09Q-Q+MZxG+46SE*u;9L0yxu^9N18w%
za&>j6TP_kkW)FwG9|Bw@!5L&Zx*s<FXYh*V2bGFek@2cb!oEl0^pW$wi0(DIl^NAI
z+mc4sYinyDPSN4$C6YU37f1E`tAPM6f<2ABFrsd2^CHb0;ww}Wnwl-E`xo|xCc6C>
zwn%@h{?j0${{VS3dpbyeKy*#U=re^6$P--qUDv#`{+G^SC1UA$4!k0+8Sdz8@wx@c
zMRT#s8*4?Pzv3I2+zodbmCCt9IJ-m~lLa$1fH1b#MET>gb-U3l)%6?IgaeMQcTs($
zVn=UO|2Vjka-+!J*IN!RJc~#`V!=-9VWoT(uU&+hF9cP^^R;NY)2pJKJ!?(ODUZs!
z+<(i!ht$ZP&O=HagRN>9E$pF@jX&E#pKaZb{8A?tN4S-bB}z4&7yeR_(vs;FfsPfT
z0n3WP)-7q63#mDZuFBg|r8QX9LXu|?w`$d}Oi&^zMG!tedaj){GJ3h>RjE0<j0#=a
zuRaj~z%Hm*$lzxw=1&vo*JqC|8pc^BOq?{Z<h1DMpWKv!%`4nMzG<AdWwsEsnb#R5
zd`r-*^0rP%z?fh`YzrDER*v9BF{BescvcMX!s24yIQk*nG?un}1`)0a@4gP~+Zp~k
z{f$3c2emAM#nutB2&+&p8Ou{|@|kAKppbd#e?wXRDc^Q5f^j(N^S9Vb)bIDCf{q^x
z3x|D#J7%?z&pQ*+ly`V$cEhaZ*Chk6vOK}~Y2|WBw}->l2x*#sCxeM;F>grM0WSV2
z>WS~r(70sL-RCDq^yf|FGaf$v{6rw6HUUW-eJ)^}fv0FrEQ1AzEnrFo&I_p^|1P4q
zn6cPdbemjMb3Oib1E>X{(~MAezj@FsJm#bRlK7n+&mF^Q@x9-~Kh#?K;?B5Y($Rd=
ztqhhGR&DZ^(1x!tH^YoC{o>jrY14ULSo>G_7Sys~a;2fu&#P8kPriRgJ~W~^o8`g?
z!~S~EW{)(Aa4Fk}^YZJ~;m?Ii03e_*u5vp-Vvp*l*2|7!tz7JkmDZqJMgFMn-28|?
zZLdwbI*zjaB6kYJ9mu3O1COj=k6_jzEoBf(Grui|pwG8>tGKbcVS2va5bobg6@@Y2
zEn6kLh@oot-)@X)>xSjSsECWGOtTOpQf1Fh?#r=cQ~oLa8UgpL+PWqcC%$b92m`E%
ze<AU05z|($jBa9jUD~X-Qq6gdE(1$&MOS(;bpEd=7FR5x-oY|s@!Rg`AM)^NojxD^
zo#dDL6)2mxVG6!6Z>>-o8o#@z5VWwgg*q0lqtRtIW_i$!a2fI%vR5-i;`hFL7gVnr
z?e!=wc1C>aStlE>l;O_L#_Kw|A4wi&%Mm^D#w9wISXJpRjwmF8vB+asO!g;wbon#C
zn1(l`_>EpAMdRA=S!D|C)!=>qSZDE&YKMXtnpf5#g-X_v9}h7Lt4xq2Z`Tj?mkDX?
z21~>I{R9rHXW)l^4xUhg5dD1Ph+39VVsfTow;)c&x8HGywG`3w|1)ifolQo^=_=|!
zSR^%EGyEaOI3}oj6e%<X%5q8(UTS(#=}jrKN&0sF6<NabB@}ipS~*IytygTnoX3wP
z1t<OdM96Mx%T&wx)8&aV9cboeFxHl|8^if{WE1l&>A1~JE+U-zUk|<iWh204dI3JD
zl@ytIewZgs83>kcZi(PB2g#mPQmvKYB+H6*p$7Wp5%7VPq@A;EvVBSV;Y+!aU2sKh
zkBC7_iGDZ6rK|yI&N>GZLOI(AVepz8SQx&vq(P`jHOeh&GW_?SNu2)2*F*?HW2O18
zZ9(~_nM+LL*)qmZNBTY~qkQlTV_-6Z36i&MeF5UIzTI)$0oYptV3Z5>6%eR8Zgm^m
z{mbz?ZChmq4b4%YoDr9lw6L}=UCbB0<qYhy!e~l46S{VmMMn)m1Z&Hu3ASQ1zLkA-
z^{u$SVncxb@|S};`})EObK06HD-~f$INW^w*Be6pj`Ph6ZzHO}jLSf_y5`*2uge!B
zfyOL%sj}XCh~4s5S;Y%yLG67S%_e)>h+)!qi1mE(<hun3*aEIbZQXJoJJi=xnCk}3
zII=&-@cF28s~T{p=u=m}OKn{kMDW>p2(&W5olK>MUswug-%-u#@6oi*igu>T`ex@k
zu>`qcDv{{za!u-2;inmyW}Q3CINLD7wUjkgrlP&xV8UEVd@qv;IfaW!zROotnWh-J
zYGdqf44_9H$iuP~+TAA9yt~9NR5%DANPe`o#}$nh28ectSxW>^K2-)ojQHQs0e_KV
zBQDc!*`&O+1n7&hNogeoe%uJ#eR-H{MzMuKtgaLpXY6jI`t{BbWHxCqj+C<s|61e^
z!23e_aUxsC`VD&f3k7#YH-j|7)TUkxWmAdXuX?@TbLW>kdpDSb63W>onj6QX<|r{j
ze;EraL@rKjm8Hk3jTmU^diQ1&J?^{0r}962xM9Ygzng#Hh3TWs4P*`2M({r_i)7|6
z5#^MVYir)1Hrrvn5o<OjNGLFaR${yvJ}-fueXnAg)tu;5crGJ`(^gze!<NW2y24Pw
zJ6)j0%$U#`HhU2odNHRp@cz{1=U|^KIcfiNztETfZ~=QioV4&-xCil88iP?Z_$VbX
z-z$qiwWDzli&yjZFtJ46gwTqSRidHV8Lu@<Ww@!_ZI;A31oRr8k<a``^v*AVV{fm=
z<m$yv(?N5n`G)(j$a_Py#r!HN0n~neJl2pKUFbyEk@LSVUOMOxTjPCLc-Dx4A=YE4
zAu`zwWPL?jzQ^1N%uhjp&g)sV&ObjZ0X}>c`Psas2?KJr#$lhhZOYuz6wJhPPYl!s
z>%q{%SqvGZRgBqHMAvFU9h`z64)mq*Cfw>CyU%@J47FY5)H}~q6(&`zg$WBj2Iqsu
z!D*O;6u<dllybLDN|TOJSVCEA=&7g8A*W|<z^9cTN!K&{&v`XjOG)JBquTsEu!_~F
z#nD4Kv1&;*iNht8ia3o@2wAwH8z4p%<smh#1Di18zhzN4bc(Jz7q=*3z$VqwRm2LC
z57MR`Lw|`GX|!!0%i$KyQKw(z!j7n`QiadXfUlEQjUh8oyflQd(;<Og&aYuMZJ`@S
z8BuI{|K5BDE{2BwnvhJCCLOmHs*^YCCyH{oPipR;ihJZwYv`!D^@d4H{x9tdju}}^
zt`v(amg1o0q;HJ`n%~D3NM0tpz*Wsa%*&wp(%{vi*9hNa0o>~b^-W=N9tjMh8Rc0S
znq*QW%W7Pr#Fq6c#h!c6M)UIE_I<zqIZyM3nH0s$XU{r;J_QHYY-i{5@ZYYZCQSL+
zVp*36T5aKF;p|mtf1%1n)9CpDFhh(4Pg$o1C9maY)yh#x_pUM-hsrE7keVDToowmD
zUg_j_wa-7M9uif2FC_YmPk2~C&0ZW{h;9gV5~w5PKyI7mI&j-EQ+^)JO`s7t0%JoG
zrKuAO$Zg$_g$3!RurGF_U6k&9nYZEjfcO@j0caRKX8!@(99T#L;E9nl6<}D`7`16V
zlM{R2lUXx_|B@>B<1=T>f;%e?0e5+upQ{;v1X`C{bkH?ZkAl|(_#|YiwvY*q5vp^P
zs1RD(=#Iyy5@R<K$vbXlJp|8AnDCl*jSflQ;XUvcSH)mxjY7VQ2WFVOv7xO*_|SK6
zSgm{GCGwdwD^)2(M4;7LZbejn9_&EOqjw&*{z(DZH7(SxWR3%Tmx2r0p(CH%ud}m6
z=8@37j&7U@S6}CQ+^~KKElaHjL#l9XqI!$1b|Lw1j;t&E%2`RTovw*XpTu;mLk^=%
z2IRIQa$KLN{3RE`)dTIp(=c#Ecv*)dnu|uU2nFC-8;tnAiZpLvUDp;}fk=J7u!seq
zcn{jOH8~M45fx|-2M5*`(kpa!UA4%2?1qedhfhB}ReX)zIeYbg{6<x-HYn!Mq-H8&
zbY#=l6AlZ*s>$<BaLfPbz)pn)C$%V-(+_F*Brx$t@r_s*Y2LZR6nr+hM4euT-7bke
zphizdbv2uIIJ)!INF|$X1y!$YNF^IH9wgx9`2gEn{g0|1_}|7)B^QkEze$wMo@F(O
zVQ(Xr4-_O9Y#hh`2v2@9PUco`uK)w*5YHAh;uemKJb}m=7unJ!>$X=ByQC@|YRp}V
zeIXKIjS<!`9r%3R7AE)139e6)v$LdG3WZh+V>1#A+c<gq?YCr^umM*U7>i|PKgFkN
z&(9BduVLvd4{O{ImCDf<?%i&o?!+bZv;2$Vmwj=m@6bvU;hVLu=QW0-f)(1CIWZ&a
z>0wRb1)~hQM>}x~1;I<w5ClV*OI2AA%}%Rw1Sygo&hjXQHMa`&UeQty^1*5ni$cmw
zIkW?VOEw8z%M39>-^KqQ4v)i0p6IGrys7+waM`XT`|mln)vxPyxcW$~m{5lQ$%391
ze4ySy7iYXb6hzkQ1mdnH^l3DKl!0&_0uZlr6t%3b?tlnUEu)$GvWtrxwz9s#jPllg
zwh*0%4&}cq`@m=97F3NX?PPn81SYfn=bI%RP33!7Lp46?(jzaZ$eUQl8l^IFi5_Vn
zQplSBKvq}qN5-#rSH0p0)u56E{w$#^w5ONT=-^BXt!_H_@uTc?XZ2SW!RXK4b;3*Z
zrbQ3l)i)V9TXV>7@?WjHUk<vA1IWpO?LtQLG;Q^Qt+5G}97Y?3-mK)iiLfN+*nV-0
zEwZ!?kXV<7!}sVOX~^4HpOM90{C&8!`_$mGDV4qVp>^+TWS!f~jV^>k4;3hbN3gxe
z599W4n+>lM)vYAGZ*bl_YNb^F4Tz33fXY<P73&{}<quzyjOid+${1LPS-&8Do6)*)
zU@ytw*ixs#_nvEo+A)sVA)!cN{b;t*-Biol+SEZAzdPk;l~?{BHmewRA)C-qA<OoH
zVL><t|50Sss*<rF1T(8Yl79Rqm%CVAWoy4+Au^(|!goW2>%VC}E$@kJw2<Vwf|2ho
zn#0(%Mt}a@TFQO>C<T?H*PK;?3mK+}BWi_&Gwy0ENU3rDa@Eiz9WN#NdZHN}>z`IH
zoWibJ>mr-Q?Edfqex2CH+wl<KIjOS+tJ7j|hwBnngOCyVL~2$}bQ-r_)o7M-qqC(d
zKV7`?&^znl9e03A{zelM$)sv6&aO7vfKt#fRvgP%hdYW?Nx~kYYhrJp?)`~Nv~d;W
z%-0%MX`<b@bk$0(#9=uMm24Uumd5qJ?}KoRBbxLvf<LEx=u6t$GK3v&_aU?W@`)-b
z@O>FUiIs#e)oPFzp#UFu)0m{yy56*WIFbrY9-rfY<wy~%iw>10Zoi+j5zr3O#KXFq
zoWyN$+5-IE^S@^wzw$Vof{_A3C0P6NvoK_G)jEN-+Bsl5>hh*V4FrRTQ7Z1*$!`=i
zd!p|`!Ul*yXdZc9%F&3oKft;SJGCPs2McFa+kEL0-&ZPEf9z9aGplCW=b5Re&MDLv
zZT`MEHhVnxB~}nV$FbKkNCB^WtUomfQibdl{h4xcE+)Bsp71lAp$?2cyJA)#nd<7=
z2jZ>|t*Cp*cl9UUlB`o#Sg7#`!+kE;1Ptm{LRlRMd7#7xW(Xds(fh6jAI}ndY+6P7
z^u0L40_<{H&aV(^`u`&l3N($yQp61~jg^oN3+k14@uGO{=p57qHB(psO`U7(>9ae$
zN7PI#f#xB$rMwZ0_8iI(su+pBaW>cCo8GIv1bZRL@<MVTBDidB|EA@1&T(LUd8L0m
z^_Sz#Q6uW|a9l^yimO6_uu5r2tT_OUMGb|?YraZ3v`~;~S(8&%+2NAqd@C%jlesH{
zTZ6<xx`RC^&WEWQ6E38xBo8Tw8&yjG&e6=pfkF%ox|R(fP!P*Uh>9P~L{Cz%rCX=s
zkWo`>tOd$y%MUV*zM6+`)CaR(4^F7m-f#Ah?*td>GFXtLo6#)(LA{ps_EcNVBWcjb
z3UwoA$xobyDSfMi-`h``H8UIYXhLLLAmL)DSWsJ|q;1Fx<b-`Q6-{V+wx<r+?;cIY
zN*lZ&Eq=}KD;(C`VW06pG9>TXm>%NLKlwY(GNqpSb!%flgbK#RV%*$H!oW)O$ORMW
z9)US*SwGo6B>_E35T%)hq47rJ$3Ub`JK(KWg*=)J9~|&#5+;&$57}o}Ck^JA<LT#T
z^i&Oq;V(4nG3Q8sh|wotke^U}Z`3-;Hp!Q+@?S^{At76`*J?_)q@!9$q>%1Q18C#I
zH}_#SP7T`61ew~T&kx*1tkP?w`tSoFkG9dypZml+2tXs7=J)NbI2h6b1zkgfV3u3s
zngcNQ;y|lOL4eH+L@Gc_r-759gt5^Kvr6-a75;Ks&t`S%kBqg&EZ)9m)KBdH!BUtr
zO$SY-7d^BQ?LW1Z8)(_3+$@IPlT|Rs+}m%^bWE;SH-t?H{DKVIr9d;e%n{U1JUbSB
zxKq^{7n=qvVQ<mxZKumarw=`>@XJ1%d;bO9Q^3Qgv}sy%c;cj9wKp`#<s3dW87RC!
zod~3xW^9{4)8Q%gD&=Wvn4~3~Nfqk^8b=^)9$O1j+z?XB0;Scf)|#^^h~eUY55DDJ
zM;b+L-1B{jW7SL^8W%y`2rxwK1|0z%-|~zt;IMR$*UH~j&)&@&bCk>~^kw0jzK;K+
z=`5q7?A|U;Nh+a8BOxi>DJb2kAl=>45<_=)HyCtDGjzic52>_tcQgOPyVm=GZ!BfO
zT=#v>zV`lYfeXiu<Au*|IV2~JIn=g&n};G;E2s5;$b21hk9^mK>fU<n^y=)}(L4`n
z+1|-A=J`HrKXq{BD3en9x~1(6NS+V9h*XtR$3%Rw$&@DKUCLC2UR3oZ(M$0SW<HJT
ze|_(wlPCEIs#LyIDmoTZ>69~+i|d3+Q6_a@uottvmsVS?LyI26r-`RjMbgwB>Oy;z
z`;jI#>tVg&Ekw%q#phM|M#{|y&+_o+cj3<wY@*Fh<9aR11ki{g@vsX~Br;^IsEu}H
z!__i2sA-kCa7kERYu~zAJ?<=$Vu|5elJ!ey7bf9gxXtu;UDHcN<F+e!e)Skwzr#@(
z<Fu?-q5GJ`Aw+s7Ec{Kvh?Zb<<NM+9zViJGsao2r<!g1{vxX(Zq+*5B*Mn$P<u5DT
zjSGGhee&%6WW5>5cs8HJ%odjTWP+qp(k8M=XVU}^=7^eeZ3(K~Ig(`_i2(H_W-6vv
z;Ruf02^OqT&W`Sf_SjLm9i;Dq1lX+be?o@FCX-}rw6)(BOAQeu07HA;tL}Q8+~RsT
zex=VbV&Z&)FdLgL)@bnl)Xou`dzq+9r@-~N_SD`*qsz`qkk~Be(FdjFL?%xgG@LiW
zD>}=!1mN}$63#$*kOeLV9nE^}@&$C<uX_Pgzkg<0+ba_(8WIlmj^+mQ0uZ9NI@dHa
z8sDF!&Ksm<Zt`a-jN8bWO2Zdulk|2+{ngxi&yB8c`hBDl=dqwu&x3H_*bnNn5c;6Q
zkcXkO`?so8im25;gM|bIpPX<Au2u^I{UC@P8XC&e0rqQjPk4g#QS;V^%Rl#h4)>_F
z^JRzZli*`LH+L=b{3*52{i8eZV*(9N*z<#(naWW|Mc!D=vyX<^5??I#SQBsvV_g%Q
zo*NB*e1{w3bzOmwaRh2yi{(>2wzAz>gKXeHCEdAp5E8+wI$J_s)uHlSK&4<Sz*RB|
z?#A?|5gQB|>m%sS)Z`Zt$|slHUuu@;&g6OnOJh@ayJRiYQ~F)g>V6r<q~jB{_o1wM
zQSm>+Ofk7hW@)BD7mmUytNe9zJgcD64@zewKM_7DE(&P|qiEWzsRoa(jSGs^w^+W;
zFQQgV#rKS1`AXJ4->Qa0jLblRtMG9LB`M>^a?DzoPL2E<hQf<~#o}KKkTldjr<*00
z$fI@#h+*d96Ow%=fCUJwzF_>cp&@N{g;o|cnexdOH|y08;EF?ophs4vOcGULn$Z%0
zc(Z~=UU9*n1EZm1ngjWeH{60`Bv#C7TH58Tk~#6^oGzMFpIv>6+3-qbDC^li6PgWn
zc2ezl_qhEb6$q&|WL;^y66sLGbq$G%fz2%~7dA|r+|7_5E;YyO_uel+A2m(s`-O+(
z+aca!UtMGKt9pXF-ysud6f&01BQf#Z%{tUHxWAyqtohpem4^<dmc=4Ntryu-tRb^;
zS1U&|gjw{8VaFLU8gHpFP|OounUxE3wF(3pDL#kgs2l1!;@mdk?|J`?;aQRU7*BG_
z@ICBnLQ(W9ymFe4$YYoxLbKaPi)(dJ#4vR!i?<P+C?@MLqiqSi$K5Yp4{P=M^KGK1
z3;lavX6XNm4UC01g<nY{yYc$VPst$s*V(37p}Pu#`1~<XBn;rF7&PqyZKS{hb9^{(
z1xm%9XwLk-PP%*LVu=?UUu8jZlW+{{1GJMQwE9rQZuf9SxRH;!Yn0cN>eu(dyx>zq
zgh=509PAn9ds0B?L+SEtQuHzn-$v2<htfy8ez&YW#T)Bv{z;w0R-APHtqm)lV)d%a
z9aV#?t_!o@cd?ClDBJ%)7$iWObV0c}W9je6KB6TE+^4L*PMhT^z&1B$2%Mp$eD;oD
zxd+}u&`TR=pNFuyNbZwzx5nR&OKwsQvYY^_iP-3NKmwjIVqs^kjVm(P+MQLSqs*<F
zO_ur;ikI$s(hFr0K_>I96#=Vsx?I4xfkTnsN*?nEB)FfBKumSDwYBu-yX6M6AUeBp
zE?em{^!VP>o3n3>$6gAp!Ug@U3SgUlK@?-8CRXQ<jxtU-pb~<K^d`6YY~ze;^+^@Y
zCH;vie5qy1EcHEZ%XaB6-84&q#1zy!T(;SeN3fGBEXlY&tGcBD3zJhvLl3X(_xvwz
zT`Tr7J7G_{_8@JNG8WvitE$gxYXwC4+%nPeB;C&~KxN}=CS5vj2i4*UF%wB8`{!q}
zojCelk@{@k`RZtX^%xFf#%cJSY>cX1^IAR%;VUr~eR@5GfgWCsD~u;%Fy^UZ^=xub
z?z`rq=}@hTh743Mzs{2BA-B{wRE|{hdNQxpmozD*PCEpsX5A~1J7&FdHyy@XC|#at
zWn2V0w=$$uc^xQP6hsD?0)nI*88KFpA#ThQGQ4b8O;G3X{XpdAY;+@(8)jOKkXeH}
zlgCiVF-%p*=kfBG$$Qh8&FQ-H&ikko9zI3k9eYHcdz@a5_$^^#-kM7!lT#o^B`IL_
zLg%@?sQApTnBb;|h)Gtt<4i%p+|@D9XKpMKT6{T4OL8O3^=RAzBr!Egmwhg~va{kV
z%xii2aj|$IhhKPVDoA}_mdzDOB&a&AiY`0@dC}DQZC8m+eo_N}m8QyER^xKFD;l$N
zziGSiS6;-ae^%Y%em_2Dw~YIX>Gyabjt7Ug-nzF9Xy&d50QF*T6iV+skVv3;!XfH_
zJq)Q_vCrQ5VnNgufSI%LL~RD`ctEe1vTx8T*QhV#{5KihZvp}UiT8;#tAGsPDngc(
zS_2Ff)}zzkiUymPo{ydN&ioPNL0s%l7RWSeQd~*e5&$=XBj2kQmzG6Aeu1#N?vn}a
z*(ilGp^Q9xa@^2myY{93Ldt6ujdC+=)f4r+)Y=V7r<+&2U)&L<6>m1#H#^Lx7st_q
zM(`mIVJVwYM}7%(Rd*}_kaHIQCXR6}w0wgNU!k~127C8wFUZl+$)r4+6a3NV@j4cs
z_7pz=`BBH6$>Q5ppt^%NMF>X-lV$~2J=SYe2)v24`OTUXGJD~v$YPYJXovLM9AF2s
zUd)Q0W#IJlA;8$XZg6Iog1yIoWox%AR?PQ54UgcD+4SR!*{<K*FugNDZAhHwOVaDY
z=eLig^3`L(%@o&%*egNZvYH&qNJT|_e?MS#CTHv6A_a@L1n|F<2;@y%CuY_8de`^b
zEkv&-rD$=l%ID^#S<0zn{lA6!QglQ!RgD78;9ja?N?f=4l{Su#S;`TF!zckmpHv^p
zl#Elcsh*cghBJO;Xl+p^9d`=&WLOL!g7oxnv9XnIMx0wHY2sr#uHDr8lF)TnFbfv2
zJ>T>O^fcuBlCQ<gLZQc1(roOe%KEVC&YI-<nn1zTo9Q6+H32>O46*l;esO*G=MlBn
z3sFZw`GrKE@!Nw_PFU)SPW^}F<bqIlT0_PIjl@^w>MF)9UPIRWK4*OW`c*8)qE1^x
z?SsJR<Mi+k9oX?$pt$zK8(!rBJl!_$;IcoRDYUXo0P+&bFW2b&K-X}|c0vkY^SfCU
zU-G<8+a99DeD)nDmjAs@mFpd`?=Ao#;XB0+t};qrEw3+mM_I|gl&f>DD@;?$$}5iL
zw=6~9@m5~Y6}KW+(@^!DS+Vt3naTCqvP`9%O5kUceHmJdO7*Nyn=D1`I>Ehwx(v_s
zXD_}=P-B)H938dz{U3|$qlkp^P5jblJHJc?-&E_C{gR)WCal7*Fm`^(|J`~1vE2K|
zd@H+v_8u4U`?m-0WCKd)Z6MHq`W1;r(Xr?c6waZuSEI4JUktuv6l!5&z1P3Gykup!
z?xfv0ip~0-^DP3X<)^}+60ZgdC|)HM22gB|w@SE|e25CpULs1$(|;>2>~NkNdiaRs
z&Qjk1B1WoCw^sn}Be;@)IN2o<rFD(l=*wk6KW{ZgHy?^F-?G3KV30k`pxv%|f%qfn
zDzt_}*A!t@c#OtY6CnnXKR7oOW>BsHiaa4b;omlfUOyR81O)Q`9L%hof@!UvC<0^;
zV9qA%e`?eQGjLoAy)BLEUvIu*?I@H6zy0?cAhvMd4Y|fqm3z8X1JKT{-%5!%pCuZy
z*DT5{@S@l-+_z$cAQ9PfChe*6e#}fOvGN(q%q3o`^(_OwSb!awnJq^ycU^(D+pR7u
z67XlOZ}LF(A0Y`+r_!(gJ~VN={Z=J8rj-;Hr_7GUh=2K7y4_2nQaXJ@Lc0Jfu!B<9
zpGln;)y}N}O~%Ty0e=iECB;`6B9ab96-&N{)@d)%nwJ&Vl~6aaF6_<HC*4w)`K3@M
zP2C%7LJct2G@|k%>uG;Fr%k$x$Kj*rvr8G)cVG@1S0kY;D&g!=s&ODOL?pB_`<TkD
z4{H5-CzAKBzj-KrC+G~zt@t8Ai7FcjMF}6jK>NJ>*U;E6)LP&8D}-Upb*}SwbiJ6%
zSjoF3dF)BQ&m*A;64ghZdu3=5azhc-NVKn&NGg9iPvc0|i>et%!{cWXK>3-sh$OoB
zWJEX!sYNs{n;k8@UXnF#&de+;F`UfOc<FpFqdvzhO~*kXSvoD=!*bYu&B^xAqkK{i
zxr63KZ9H^~hfEO^-qE8IpN+`pet-BYPZo=XIx9MSk^6uNvoKMLIjj}*RuD;7VNEri
zBEGSfP2Vbu>Xc?D%^Hm>kPyHfTTzH{`#krXOJ-|a!|Srbn!H4!Hznhx)q7_xl0XJ3
zWus)Vborr4&#MY#`qDUC7N5%1p(3LxT22F%5p{D53)A5`RL9j5aYVT<6fMT3>TLHE
zKPp~R?cNJPerk@{krwE(DtUio3VWF}_d(@b-c^z3kumwT^f&IUq}V$EfXd9wtjFS&
z_!9ICMg&$ffWkyt{E6J*CBcj^%m#so+hLDc&G|$Z&1H|I#kHjRhMwq|n6$rnOKHb`
zkeV&J>Y-c91vU0o2dkfd6^p-sZB5i(lnK!ume;HQe8?#8`9mz;*Sux_GWys$%K0ym
zXyoF+GR5(MY)ih#-_CVxIz3nwAEYp?KP1v%JO!}*^O9=J?Q<cIp8B2}uYK<tAf$}6
zPw8b)T?rDIMM=VU`?XKg0aSq;@Liymhpk4toF80gKq(!tR$ABpTC7$Ugp98L#!ktU
zN-<CWEzos=&mD$1_w2I9Q?H!MJbwhL%lc%BR8T(i=7pE-!85j87F8Gj=5AVWLq+f3
zdmkpwbx{PLY#Hyf3Xf_i_|xYEvAr(%AR|bA^e9(UwC$!UFP@V;LcoX$=taXcZHS)f
ztH&=Ez8B&-(a5IC+X|TrzNu|#+e@6653G+VVD?d-s=qI)@-g~@BpGeB`6-;_oXDnX
zGQI~-HDUblqC=p@Wh4E)x^d0(x;$=d(;yxd66tN!_#O;x8OP~hO}(UF^I6)S+2maf
z=va|e_^3e+p(R05av#vuk`r5yY!DAFJjMjs33kZj(3g>CzeQ(R?7yR{F{ZKmoa2fu
z_l_&<+^#pOHf=J%Nd`qtx?N$POzK*#Os(B63e9!NkGc>g>&Gz43^7GBscaJJI*lx}
zKiML8bvpXR(*24^=Auv5C!lROL%7$5q32{`mPIca@Zv3)3fx94AgAAN79nSQbKIdI
zzw9kFjA_5vBtB2ho^1Dc1kYVwz3xo1s9fG0y70NIw5x?Y^w0UsH~b9`E<AW{EMg=;
zTCJ_XK)?kEKQhiE*1cbOsdUDw=(>}chK$3xh|}X_HXP&WG-D)~l6VIG7G#}CWp){3
z$){jMMGwk_GX5Oe6A7V<;Fe`VEN86EDi|3|HTnBhLL`SnRET!i8$(#3^d*djZYbA=
zs_pDAdhVxfKZeO3r+^sh+cwrVec0BW8L?STwgGI~)qW^}8P-<u>H4&_T7-TA*&XMY
z_3xi(_=D;j2!Y8gnONAuDv2?L7^8Fb7sGNh2n0G~F0i%hMIS*8o~mv7u>duGcxZnR
z@EprF@deiMEQ%9gzx28B5ndrs2lD><d4RRT^)D|t25k?^WW0hG(HsGP)Xsad8aQG0
z6cqUB_R91T$VWN<CKBAZ8Uot^@#qBedtOndR<@p~AQ%n;ouWEjVt8fBCqs_bnFF^b
zuPfFE-&_!4o~B&@86<E;xh^)iEyh}*Au{(}qxTJ1zAaz=)v3NM<PD6v89sM>@XcjM
zN5>9z<@deyHnWZ1H=HdkEx@ti%;SNbF?i#xZvf&BnP$Y`%5QEWKF*WHl?JlX3_@gA
z$IFIs7+PV9eXLn`lsxFO-eIm7K5^0@_Oz}G8L;H^8NC}3(A>KeTX8?resLvfsMY=c
zC4yO^r!HFT-qj@pww{<y|4(qEN#oGvGsS1`8;o)pZSvbi>|SeGWt%l6!s54Uje(OG
zJ1#wrg|V{u!MHuu)M|O$WFp^!T-qmEc|tUpE)~KOB&aXVINhr&^d0sVpg!TM>p^){
z!S3Q;?WrBdR5nq|@yj(xY4OLze(O>PUA$Lg5qa4ovpoEI@ZIQ+%bz(;NlfXMWXew)
zUthdRK3-%(X<tWFqdLL<v?W)@^mG3e(H>&q2bHQW{=nBHkzMoFxlxit;xx~szI6x1
zi6Cf&qi?<YBe_H9v_r!X?M$Fquk&hV1ryROuUcIy*u|6k2oUNSHAo2U&oWuC)`~u>
z9sm%(RsBnC*{Dw|Q9W|97AAq6k0SDkC-+<WMmgr_jGpi|_bN$Lv!F2Jpc#ppP)07+
z#pdB;9e2+o<@N|UeAeMcuZwpDQ=6-UjGEC!MZJSNjHT;d-dj3H)zoiy8s-<~C}qJ%
z8Cxjx{WrbQ$H%zQgExx@inpoFE^7oV<h4{N)s8Zly1W(~LPi3)hJ6;ypem%ikGZ~B
zebW{}n;{8LMX)Hcs?V@2k2|CFoP(cJjR;N7CQ0Lm3{_o`%tMpwO~7E`ReG%2<jw%H
zfbj75cQa#wu28?af;3Pk;b@a6KU?AW(DwqUwe$0f8<WNIAjlrw*(YTd$Cw7At0#4A
z`tqOgKBb>QF(duD`cli4$PpC&6w<6YFSg>hI4;5`IxC!i{66@hOgg(P7J(AntRH-#
zV~ijH$8%Ov54Hz@hNg-fjS!cl2K<La{732}LEEF)TzkuC%I5VM+D+z^PU_;Xxm~HC
zqXL?#!!^vMz1O4sk=%+qE`k?b^R|oA_m6ZxFe@v$!W=86Fd=aMK4!n4bnNIEopiXC
zB$63^EbK0bRlLzutF>LQjAFAq5yB^e=Wxxq05w@8Ky?7BhNS9o6U+yM&4ZH?+?RbC
zQ0+TG=LEz76tX_eQ)y+pb|A3uRsz!yco+R@*Z}GV5H;M}&&Y>riohYWsY%Pg1Z<fB
zx#tNK0DW2|&gwPx`7MmF1Dmk`un+6=qpr?zBdQ_vjjFjKw67{(M>+bClac+W1sUBd
z#K1(Bi7^rFx3jf<a%I{F3D{eN0&6Hh4_q`$>Zsyhhy?i#8^w^u_5;$}BsU0coq-_V
z(-&8|!tR5Wh%Ps56%y<G7xLP~J*J9nGM|4h^Ow-kHog*d9E~cH<W7q?jUBJsr4J*s
z{NN_nLVQmKz6}<LR(2&b!`^U`=hjTv6&53ufnuW)f+j4+(y-Vw`v2%Bnm-itz5L7<
zs^VKJhi+v2PX4sUnt)ugPEa4qJo0&X;fQ)1lDb$+Hf1s>l32(lGU232aT^v2&f>`N
zaUo!F9bF0kl^D#Gmn?=J$@5Gy-bgO6wkvR~E1#L+`$sAMt-Hy0#qSxIu*w|M8haR^
zMDiARywVtwWpAmAtA7_&rEn%KlgX5`&w!4?5_a^Tjzc$4rkz^E;&TCy&(f&$KIajx
z1hyYD1nCkvJmaUebe*y&Y0$~3-bTl~s`%?2ffkD6c{TC2C?btszPXLnGv}qdVA!vc
z{h}@maWCgg*u%f&l?P3DP$Rif{fce%hpa5hgg22VY0*W)AAV0XnG5Dp6%c9mF{^hr
z>Ym17s3JN2IT3{$K!PPjEHaYsDXx}Q>u1}g<r^$P4AtFjZi%d_feW0oV#`b2=e-5j
ztfh_ilh)3~W7^N?O{JwRF>UqN<RWX4Oz_PZ2QX!@{OeyH2&ZYrQ%CSHREez!O(ZwE
zjP0+8M-+FIrL1_*7&HH%V7R=xyxzAkSWDK_U%i0|_g|3C&M(a8PaRDjx{MDQ$d3sG
zSyOVVh94Dm5Iie6o?yFpRqgu4BAvxtq2$$)@0MNrEfkXM(02RIq3s615VM5Ygx^N@
z2&~_+46}E_El)fYOAcjeb2s<Knlh{~I_Up-yq4@VuKvNJIkV7QCqNPn;@v=3`!<2K
zh&i*EHF}`E{QLNfuJ-j~%NReZIW6-5V%4+X?+_|#m~9YJ&4|IO#g}D!B!kknEtoJ%
z$W}bPj@X_ILHzomBn<3Cctp9X!j0e=K6y4F&BiueTftw=5Gedzt)*PKhX+SNZ|UzK
znIyw7jE$>2rxqY0#!`rG0zsQ6ZujCF&p6H>qDv!wveR|Z$Mx37+dUqYEZ7#fc&%r_
zd~F&4XEBN+1ibKsO8hevu{ySM;b##2E6T@}Kpc;Q`q|wCCj=iL(qBMwL4pQgf6cTw
z-^AWLUO^2~7m<lx%qk&#8fB&H2#)G;e0Vmye{#YBx*TUvQE|z$1^`#_$qMDYie<3U
zw3V|CE^MySF~Ic_IJp_9pe{sM@!j!#h3NAU^IAdc%~FGhcgy#ie4cyd4WR9nFp5Ia
zZ<#y${)+ukm!zd*PF5?tt+b}N3Y!+c3nR<vaTz;eBVf>0?sr;kZGCuGLH%s-&Oj}h
zXSzuZ?Tz)v@7cIEgI}BSmiIoNvse~0+Oa!iDc9}c?@Nz2o<Y}JM6$4=l5o-Y22K+u
z1*0zczzDb`RK3&S|HT;)T}4Y%jxDHBbc4A~UDXqxH)Tha-18;y6Yu~9Inv@yO85GS
z3*@G2W(?VWMEM=eh(BX#n_ndv<%}m)&rVGR4!UJ^!4)6mrL{_u-qFt_$Y^tie9^Mi
z&tve3tixRS{DlDR2ZLBD!NxD|)=4UA@4ie|BAYQyr_~)A`VIjK7)K^7T}hhX!m?iQ
zb2#;{m?pXi<E&4@7w1#0w~I%|muM7K^=;xpDZ+PM77WABw}jOr#y=Ajpd}bw59Y32
zJsQCIlD@iVW6Zi;@wbX@nx6X{wqEg)#!HxDDG=lrL8vtg)<2r<+#y?74Z$EgTf}vt
z-gg&#xR{4a9w~u1{oY6Qd@9tNi$@gHPBL;?!!p_&2TGaQ!=(FqR$!uY;k@P9$k@xx
z{i7h)mBcT%#}J*z9l%WZulR$U&l|J1J{Y7s%LZTM_5aBv;{Ay_Sc*IFjit!|wKuTH
z+y~}dco!4+REUlZvq483EwIYDKlNb*$jaQpDs{R3W7T8?<*2d>Rt%VdGkDrw<t4ee
z<YQm7+gOEKy#6-&CbB)0=N2<y0@%#^_Q#9Lvwq0EH1OLsG{IN!#`Id<i&`C99%K+&
z>HS;i?*0R`E3d#uD}XZu!NdGA!nEcp6%gl6bxa*LTQn6(`jw&7N9Q%?!ctNDLh*$f
zV6pFiMo`R$Sd;bF*Vbz-$tCp28<1opi$SN4Jx9lo_nJH%tCj8NQs^4J5>RcMbZ7o7
z-?y4$y!_#6u<F4>xUVv4lkqO6NYn;t?ZPe{wx#f09UxPG*rebFrvDqu7HYj*uxSNi
zpKI3W8B7$C&%N{?RAB3RjSxBjeVm0iqY44Rd<|?^!IT0JLmP=q=mVCG4nw%$7MKg~
z4JHFXQaZ^#@<yXV0T;!m?|d3;8OEo{+4cE7GxHDw6du4JUf1;9&;-nc3LSqQM7`@N
z?%^Pl>8JoJq>ids0G$A^RkD8I0Gs`Ah-Gk%#0I_ReB%H3ObF8>TJlijb-DA-PlL=4
z!ri98i#vA!_VL}c@ZIbpyOM^SKmEo_g*}c}-kPmkgyuuX)?tJyBfa0HvVtf5F4i3U
zk4GoZ5_%BF@UZ&-o%>C1huNfF;+9adsbvjw(De&;=cTeMYg#f|HfWEG5wz<ia!AMq
z{otI&y-o|S$5@l)aa{>8{##4*c+}<3jv58l9Y-%pC$#YzUPj4QNya3EbiMwv7dO-p
zt5r_NuUQaO{F+J7jv~Y6*XfU$b{A%e>2=B^Q@xK`^#hzz#!*t^{VDG!i>aBYrApJR
zP-{mxl~|_gXqaBe38J7zo7FDoXS#&fo6WtpJ^wh*kv%t|^dX)~H(X<DA)$a<LUy0H
zt*paTKk)OWfOa3ls`F*t41P(^GmdurcDoG5^wGYedCX$5y*F@aISf{@QuIB;qe(O0
zM?znJ=_@5IGh()fe|nSNk{RXOO-QTMaXH2RX}_(BS|sIb^4a5+c4!Ter;Lw*L67l|
zV$>cV9PXi>Sy%k;87*l^P2*|QNiyqXg9|(Ik~)kyhQJ@btuNUoHBCY3=5cR%?Tk%$
z_)gR#bh_XJ{|hlyA}om&n&IGoCxhrq!t385`~6{k&tl(XJ6-IGC|Q4{8W?(s8P3Qf
zxx=n;^QK3PvdBX0(?y3GN#(QA$=8z25-jv}yx;EcVZPy&_fxQAb%V!e00Fq8MOyBN
zRCY1)Q*zemzjiotqm*Gy`QJKyDC4^s$%)<bUl>FBaVUAOCrRWq<LU%y{1`Ds;HR-C
z^RV_2(MNjtD~hE%KL?+^&yNl8hf?nY8#r2MDB6nc;x8D|i&sanJ(dCC7icp4Y%C#7
zDsqGK^Tc2k27sgn2%n7nLH|M#7}&3{u`z4PNmFQ7z;i<RUwdcdx~5rO9+RzeruAa(
zAtimx_c3a}GS)RN?h<K!>gYZj|MVSGcvT3wcJh8KPmd|0z)iKpYUpc|X}$Jz@f5pQ
z<$rliyKXHAJnS#>$_6Q)WgCW>V5}5*=4AL)VK5_AWsjN{S$)g@H^=-8O~q?xlJ;?I
z`9|n36q3uW=(&-sx!c-6hlTOqw|1M^iYHv<lF!xRLi9bLy@31wh{MB-!*obEZqQ$5
zyh`c+o<{V^LwDjavR4Tmi)3pb$aHA;Ty1^k#x%?cF3!L<ale8lx<U`Gtl$LklrTCB
zz(X+Qci21V%~+{uQnboQFw@GuxN;V~a=u@O+{fND<=zPZxmKx?+M?^ROcbT87B;%a
z>ut~|g}k>&^R<hI^sp9JLJ$Ah(=Pkw@0Liv=nXymiT8D~TUvRV`#1$zvoW|M0U#y{
zL>I_n=VI=wIiEU()v_y9nO_57XFkrtGWo%jWsE>2X+#M-JXlhG;FK0UP>GvJwmEGo
zg1YXLX=#)%Y<YxVG#-;SUd~G~jYEdX$FTaW%W=z_znTrJMeuXLF5{?{PH7)tanv$o
z!J8N{vtFUos;DZcZZc4@o5g%W`pHb%{RS<R9-o#HQ+r_|f+vUk<YVvZEqaBNLHStq
zFh-Y^+^cv(yy6|hc~egLM!03?gk>MY4yU+kSkRB!j~N|uDGan7n`y$ud`@8-WJ9kX
zVhl^S8e6B0zgH6dR=;Hl!^iK9_+qmR|80#Jf<*1l{H&sEM*!mOxaC*jl&GbXP$Hi)
zEYO$K<3vjPTFp8sN=2JWa_tV*hoNEd*LPFv(j9ZNl(r$asi*sn0=KssHsOq_N6b!c
zSPZ&pJy!emH;)gFxC~eaQWzhi!)*RwxQpINV$ZD^+($02oSXzD2SnSXa&RKiQ2&zK
z`@Lf8Iy5y~`GgH8Ju!1R<()|$51?l0l#OyC+9O{7E2v91E{j!u+j`RXUVmw~=1T=g
zt}wA^VOUW)rBk!L*RdcQm|Rns-l3MGw2mJP8o{Ka87V|O*+6U%xihg{^7Oxq-WvZi
zTctX(x*jUb@U=h#NiMEVyROA94)e9u8Uxit{wx2|vG{Tdf?XF6^3Pjiq=7{$t^zYH
zGY>RGJ5pHf3>8W(W+siJ$Hj^{o||0Yy#vYRTlNlwA5%Oc!w(@3c86{7H^Uo44Qg1k
zUAYdY*ESq8m2mp^+^B|l9}I{*bDo;D*yQq+Z5XP7UnBqTye&4xP~7O7oK>NWveiV1
zRYl&iCPN;1Tg28NpXcutyYKu2YY$ddR*=HH66V@(9vlywUp&{F%TaOjzQ1<zvF*AF
zVQ!>TC;p&?APYK2^ef%-T?kR`{d=uy_!L<dZ+R<gC0CDT7PPxxBN!uGDLN7k<;yYV
zfId~Q_ybK5!Vl+D;FXof3adnZeY~-sYT%m=cI;5pIw1e-A0D2ni(aUM7%lWZ6`lYf
zep$|dC$xivOmykMn|bm?IJ-GD1HtUisS%mBU)1*qJG2pcpX$@Wg9>cFXZ;uu>Q(x%
zx7c(=^Lca(nKRAKva`wcJyxIDgWK`OaOxg5DHyeXp8Bad$j(F+C%A5ZqYAVj9+<QL
za%m;euS~c~*|X39E8vC5-8}4czZQ0yT<IMQe(h5*C-~v=^p^31_TwJEh|>qB0i`pv
zm!|6&eS-dy8ZV|^%pqCH3p$C}lBxBf*Ub_jI$2rvFG=7w2ogP)_?+rjLo<PS`cf{E
zV926LD>CAl6j9WqN^)zGD+9gL%*_~9#4la5xQb?mQrYpe$rkx+7@Tt~FZct)M(wCV
zj4N$I7{f#(KNXBiPW`sh8Jh(QRj)CXIK@f~%Cs%JE&Vd^V<ftN4?psxQ4y?;N#1|(
zYaRQ)^qzS{tLkpkrolfrTrp|`1@a>;DaM97b~y8gcE?<zXH&x7?5jBNV~Ds+XKMR-
z!}XFo6t<<`JUmhOwjF+(nN<|hJ)h;ST(bZBd2UPn@P_}}7-g#~Ignf(TMH_VBS>X{
z{nLtOkjHE-R2xvSKlaf!qheR-_bW@0KV=r@4@;7wv<cC`h>kI?i~G?PZxhXvVqe)U
zhpvLBlZ_}R$&WGjQdPSMZ$QQ*i>@Xk)8L+Z{^%Yu*iY>FbLR(2GRt7?>jnegn8o2)
zVwZf!6!;0nLoS0vCm;0U9bd(=DJQY&KgLW_mfyVJzNo+KCy&97_1L|ga#_GAOcW?V
zVKk2Tn`gRTd+qYErBF69YCTZAdDbBzQT0ba8N(o9z=p2)JT;FBI`UXS@_C+C??0zV
z@$a7ejA{7njLqBM^*>&Qc-y7Nw_ReNqFoIr$of-r5!l>UyNwmp7bMd#l>0yjqTjpy
zzM5ZLH1V>?Y-<CMbppW%rrMY7EfI<Q$ZDVGJzpCVTys`9uTs1ga(WyHEHl>1ira4N
z9rF!+bQ~aTKYj*|h`zz1*YF1s_+`T*%ktf!dpcW+$5<oMdU7k(_VHM^4jE)|#h_M~
zfbO!S%KPC0;R`4Ye@0K0mFk8hk4F@w(uLVU$=s^wH^+?>bj9Zt$h)pML)?b&JQsyG
zQ@fU3%v*X&i^xv_fW+p%0}!mmkMh)lOtYpHH56;jB94%0>t3+E1M|rv6HyQbgq(>3
zx0}h-_8tKAxak(X7cWZk-jO$0`+p$o(*@US`R)zi?T|$Ci-?p=?Kp0Yk@NEM#;Hx(
zH;BIcL5d)gX!QbtD=V-BkH#;l*H{(#<l?IV0`_Q4qw|>1f2EKs>z#2=R2{=Mq|)=#
zVYV|N=$TNFRRl-7#esD_^)M$$o(TXckN7z%=oFBKZ&R#htv~tGW-ARqmvn*|arbH*
z7*nyI@Z;mEnQ_qTr)@yAko2t0uw{`B1d?!Ynw+<kd{}~=$)CSfOz&*I6CiqB!~Axb
z6RbKx-2~Vo0Lnw*tr68ege#QQPpF&+oX7#J_IJa>UxG){b2sI0Rha!%Rz(<<m7-BR
zFdc+DkjQIj=r?5lvDLT?e^}mX@ZAFwbJgAttZ0>6bo}tWg6_MXsKs9262jVwEG&kr
z*4&?+$RzZ~D<5QoBC7oG66%=4>wW|o)a-h%<dPN5w4eFcGca}v)@)#=u0DH|#mfKn
z!GLNv23?GSn#!jv5YdY1MgyUE)!5Nca{R9>ERJ7VHOQMX5FbMp-y{M1r{0=ka|kue
zvfemBBl5&Ut|5_9(w`-&;hehlmB#8%d7S|D0kMzsegE`QBC*q(hP+~_-ze%Z7y4|n
zq%}Ts7xhZDX>j*kEhbEZah*Q5fA1VDdgGSKq{(_>V&I)VRyDdYjLcj4yXi4~nCyMV
z$>Y^>_oxOqhumy>VO7|<G`b!T9B>+!*7r2yl@rJi1|9^wLa+WnhhznFmB?TBPr@uu
zik0F}(ZKgfo68h4g}i7Sdf4lZ3syttekVs{e^52zyjC>oT~Ga+yojA&y?{Sfb6o7}
z&ZXT#H`UH{5`$U}V|_@kx7_z><ErYx6K|!4<ng0RkJU5S@kZ|5&h91;u)I1tdUOn)
z;8jF<-i_Y1Uf)NaH;W(sZ4YNz-FMMXwl*iS8Lu{Uwh5eYk+UMhp-^;ej+{>DGD|Xx
zU}EOr&_Iu3lF9D0W~ir(#GeAGSqcR$k+{ZHGfNzm#TP?nW5#X@j3~uSgEhgSpXJmu
zoc$<^A>R}W()eT2l()!4L?nmYRfn)rLc0-#gj$~PP2e#AMuEuQesbsHor*=A+O+!*
z*A<0cB_>xYH-0}MDl)Z@FpTK9>qY{TMzA~Kt1wtUdR${Wy@SHN?l*zcv9kU0{bXy0
zph}nffn^w{=uX|Krdt^oi}K`38JDPH$%`Ln(5^jyMRdX6beS7jrP;}1VHvAmX0iA>
zjS!fPxQBHWO~ewdtW)TbRWSa%Bv@4e%Sdgoy<d=RsnF4xB?)=JHvZ<<A4@S?K3J6W
z&vVQf#0*C>`8-d03HNk=x40jjg7fKD@V0F~4rcrB!Nc9%9n2Fp@MgNEI{`WOse7}n
zZFIU+lQE(I)m?E)EUh%7%v3#!VoY=HvVfz%-2Lzil(!3(eU{MzN^Tao0eNmuzhm_-
zqa;3)&tdZD3V^x@*Q^O5CCP1DL5Ii~Mef7C#m#9dR5!O9Og4xVf=5fA_V6DE=qNa?
z%QXDi0I_Z&6Mi9~()kg?jQPKRO$D>nQl56QzbZ#JJsncCGX408SjD20q096!>kCnM
zL{c_(b~)&?3XNldZ+W+c08bmd{|P&M9ZC7N8bLk%g5`g$hZzgw^ozB7TTHq*Vqt;?
z#~?ze3-!&0-nqkBp;uneeh1(`JzlV>oGyuv-M&~27riUxeEho^uB~w3EeD;g`Yf6_
zc18GE<sJO!uYm0{0^OG;?@(psLStyCW2)&2*-@Dg_t^Vxvno8t-VNUuOxaHyTujI!
zsiJ6`6rW+MZr!{cO5hxIEcvm(gi#59HOcj@U=k{TRv{tdPu;~Y9#5>GRe&p@HW5J`
zp`iZBPx*^@%i{OtBMy?$ZCi&pm*yOfpKs!#t7HzqzVhF*ivLvTlB{y;TD#`CH_Q7&
z`mIJ3F2(~D|Anm4L-w)$dL6T-RxGZZyL7q;uXlF8txFg&SI5R)orOH`fnO?R#GxON
z<IMtrm*&c$g?(Y!S}Ek1LZJOhL^MrVpy|ltWXSIH9GUe(eSlD6CJ!?>;#C#{L+|rQ
z2^62`%U2mrsKq>!6_M_2)-2Z1a?VIDJi-$1q9>7vF{tSWEmB$;wxw85DkghxI144K
zs#{=y&!-^jO+m}tS7L~DdK2Z#1hitK^90WnAS|7}I248iYDjYt{nldL`l2?(xvmU_
z(^cNj5#m+!8uAx0aQa?uK{B9ck&w_{>h<gGv`XIt8yPB($D1unNxR2UhezzQPfpyN
zUw9~YFMwRBTG1NvUs948qBgZ49&32I?6AMQ)(9aHm1^J{9~!&Ywb~`jTf`ccazNN4
z)A>eN4@$Y3zBO*U7#}s)BbmIGL==3ff28%^szJM5dIOEZyCamscU$r>+#>a8$tR<H
zlm!fbcmWEP_w$(A$G^fSBG`Pj=(IWxeJXNy`u)805w9v18$00}xsP0ro6Gl^_jk(=
z?|mO2H!qgF_v~PUS}l)hS6k`Yk81t>7(X(VkM_jwH@&)SItAv)J{~XXu5=#-of@cn
z&F)=Tmw%~Mm_||8O3^bE&%n}D$9?k{bcFL<&u~2;ynxtY!oC5Z0>BOiWCW}U?GoQV
z&&#ulK;Pf^+VpR=JG6f9D?b^qn0oFMr-GIgoF^YR0Eq&<?2U+bm|Dc3X#j430sO+i
z!p5dlBm}&Aw?J{ilkI_j3mE23L{C>bkwM?iU`-S-qsDCsfOVOETp(gm6FQpV!t>o1
z2rT<oi2<qI&{=suqaYYoMmO`YwF-s41E&bp<T}vJTxA!upSr$c?<OYjKyIcI^04<*
z5Ck$7AXPrRQL{921cBeRg1RBU*|!9eovF}!Gv5<JNIYl)1Ds%kyM<X?v8PThu<g1Y
zUA|WY7x}^o^x{w1)kJdjitjBAX99RRWzd@!?e#>m73f?mmmFD8F6(wMQLUrR5lmLS
z0+wn4M-|~=sm;s#dNs6um@mfU@NZK$AXwUsS?X_5pzU^9=0G=H$>z?$R_zS-I4=rx
zom*fwSbos+eVBy+eB!q~6Ehsh=YD!nK0l+OsHma+rVCca*2Z533+pzr3&)FL(2ya0
z)fD5Nk`aI;O|K#I%!y`#1Q%ONFhS`LBtj_G)o~`D9H8k_UJrbh{m;Jeo830e+(4>1
zdCuqI6sFUSmmfd#I%3;I8SInL^M_56WB=rdm+uga_#40Vtm`9_Y*;~>uri`Q4}lfS
zqJC^(H%8)o*4XB_dklMw8c|sdk({oU;UrQfmIUL#Db<y<_>xdISG3N*7LxhI`+N8`
z-*tE81IOKL!`%Gxzj_B~Z>B>l%!RX?D3b97dKwHX>OGCDar+5<<<!paoflwLzsA*i
z8cdj`s4fDpHCWqT)oXvCU*6cZ9lSau6m~Yl;mWC5W+AFWnZsx4v<i(j4ibW_Y5aI6
zwyxXgP=3^GXt%+j-x3~lOwL^Ni89$@KM2%*u8Dt$-=Impv}sS9I(qTzw;sWkw-$^_
z#WLsHt-u@TrDECWz(IsdJtb)_-k*fgQ7SsY{JNBeCGXf}ptvJL?{R#qoAu=g3K`Y3
zSoIc+gx_$GJ=dxr@ILOmJHJEL_dcNomXM*<kD(OsLBegfKL9(|jYCoRQkEl{Xq0Td
z`0XLK7D4|%y8bVL>KxRnM2u<83931&QW?ns=_*Oo1?4S76Kjs#g;fc4?97_PTFnI`
znwCa8b;hTYxDDF-wnuo995Wr8eRLoIxA9ZQz0u`{Gw{*RzCk>pP9|Z_J^RNNzO;Ul
z!gq`NPRKR6zSn`y7c3COSPv%DUz;fxvUNDh`xN!V?@XX@W}kSkmH&71L~tCDdbKch
zU&`kYEuE8dE?7-l?J%M5J^gjP!5z=z!_gyaZEu7y+b&Gc;tbv#Wv*2rkkS0xHCsa+
z{%R<RYNcVYnTJPWOkm;QGI-MPS|CeEMEoiM7cybmFYd`zbQjvo;@1l5?@*`}>6l<=
z&9)>TRIc6Up!${lV8Vb52psTOmCxAmoAFWH`iZ(dghBrw0x&Iru>BGI0SbDIZl^QC
zZ^qo)yMdwxdMMgfa$Xd7(8yl*@wKZWXA>HD$7||)1_QLWGW=Zm449|dZ$=#e$N&1N
zpcqi^0X*yc?m@ejaLh{GhH%WRmYc|T$@FKA3k!EC{@w>=bf^3~koSP96zD-1o{mhS
z*PS@KJ_f9*7d%y9oeepB`i~vpmkwvc(EH&Fx8Sd<Am;&8Jb<19PXr3@V?^MKdHO+g
z8tkh<@}CfTEeT?NKqG-aJ_uIO@iz)(D~;|3A7C|kbvokxhR(b85Oog@%0BqTmoNff
zJSCgAR00iPwEVZ+0y@zaTilym_XvRQP%^tW0iMKHPFxP`xR}6BAbK4ry7J^S9uuh0
zZyUm5YwbAr)38^;nrq74sdw$ws08F3V(>)1KB0g+Jx)V!HSZ6>UlG_L+xZ@U#VYGZ
z*eTiSqI_%L3vge*K|S-W4oe8LZ;@<Sw&0lg_j6WG7NczuI4iCUc`Q6n({60#_)d|&
z4v_^GPagma7t#4$8gutQf#Kc&+x8CpXG6&%1C3K+z12sRT1o%3cC&NjW>|VN2<;WC
znb<H0#qx?J7@}iLvRwaJ$!V8ZZoCzC!H-Y%Om;|}vafgS@Mt#lV(%gHAI1F2!cf$m
z^6VG)oS3>POVQlgs?72CP)4hkVg8|O9fKQLUWa`yJ%yVMgWEseTe0oKd&~ETg*%w^
zuf=+Z*6Bky+2W|FV4O*os-|4JO)Ka7X;Tx3SBKEyz8P%yBjN5kZe~-$=|}-$inlQ=
z%18QzWPokSx<kjsV9_H>O0^)PW0?6dJNa-(V<fYlfD~meS#YmlW?(wNH^1(vhRVl0
zR~k29Vu!KdPw#x#H1)*?QkY)=Y$x#@qDAB5e^ae9JH~pl_Re`92pNZK)h{m_oNnJI
zP4mA6h%V^I9E|pBb+&5vpUN$72Ty4=D-PM;j5b}Zn%uVWX6oqd?;o;fE34q)vr6vr
zuFqpU=;uCGNtY&%?^r8nIXZ$)$}P^wK;*vC=L>e6ubi|i_*vwnkW$1r^i1*PsPD*&
z_Bjn?lyKf~8I+}srkYyC)Q4!`7Y2(rC#YLYWCtzE-|4n|i_X1k%H0YkG}xRuXsKJ!
zD;iI(>yXFzMS9%&Z`@D{t(&#nt!s7St;TTD-IeGHC{TWTdBu9RCZD@;vz4<omTj?J
zclX&X9O|iuzJgZ~p>Tq2Muo2?4@s3*sTRvfs2RXhtEp`nW&({!hoG|+Y?H1b_q*`K
zS$#O`%;$LFnx9Q~dpsHYGq;u^^?U|6wgGCp0f#b-0EEa36uSC!V<56Ump~u_G;&t1
zXA|GHX>h+JfrT<e*{~Oh34bmR;nz(4*HKPm1lrDE(gIB}VrG8Hz(0JJ$l~LtiMpMu
zUdTTL!1;aHdq&%L6|84~vlw_zZh&2kw*4g9;rPG-l6m6cO|{30ejmtMrX>Y>H_f<7
zPCB19uJygyLhic^Df}@YV?ahmRs||(05;TLb6^!qyXE4(mVO7nyIPm+jOVHa<1phb
zg1ik2m@3o-#lBDIxx_^0jz)Hv9aaD#+|b~T7z6?3t9cBVPrl6=k~-$$r~VDl0{RsC
z0dTXe>CvsTTQa8Nv{7^5CRi{dyZr+wz2M>Dyt=?VT;s-scFB`w6yU{MejCJ=IxNUN
zN2H#q*yTnq+XYy~PqCt5gGbJ%0vqr)C|R_J6mc)K!dHw9uR$n&(cT%fA%ObcZXD5n
z;14P2<>mFXtAFyq{cZC~Rr-F0r>FnqV3SjaE|}T|G6@m604C2UQ0ZuWFEk;^vj&hh
z5zW;H$m@Q3RJR#k#=*z}NgY>Z1h$%ftt?j=M9CaHtc_7jg44+rh0b@w+k^s<Ct=gZ
z{w!EQlOkmG-*&C81P8vD$Wi>ZDaW~q3NlSDDLU(g=eEJ4YM5Ipg*ul7X5SMonH9l?
zYJR-?gqXa+f{B2bfFL7BG$!N6;8aggyt8w|OA`Ifs+p&zTwmN3LL~c8TufX(K@wkp
z84~oN3-^O3{t)52w%DlW6erX<Et_dKj!R9RZ&eVqbc0j_={3uq-;Fd~<<uTM9<{-H
z#BeeEt8!AvcXy*+l|)j0!2F0Ggx{z{@F*q*;FbKnMlxh51MJjMdh&Fi@ryi_SnQiQ
zzZ~{dAnCT$@#nWql&~l0S_P=66U}zXe~GNa(C)__tFFRxGJoYp@}lb}hduOxZueE_
z+!BXwlVi3+>jURrs>Y8&5+kyy;o4KBWg>apo(~_$csfHQJ}PkRZyREE&Pb8s5h-Qw
za0>GFPC;Gx;xV^bE2>;JUup`G{^eBNnmjEf%U1pwo+XCg6U`XJ-VyyZBCcCV)kZ$%
zdoxk9=bv&LT(P{Uk5U|c?q{#Z)?XUW&+AIs%S&A@54ok_Q2c51Oh9`3Ois&~n2Uo;
zxt1qDEpy*gFNqa*$M15I$9$_~V$zDO7<+Ka@2LG8n||3RwdJTS{k!Nb`>00G&?b_k
zDlTnGk-kZ3n_|{~GS-Kvc^akh8yS?A;|K4?a&_`*r=;gd@|hdk9g{<2-mD%p<XX7a
z_M2DC4}AR{WG?+h5dE8^KgUZwKlX^$TE!MV(6%bp^aGa>1re$At6KR_**Je1TM30I
zg_9fT?Q9d=KQzqdg(<F4JgxX4E2BCIg|D9^k*@>$Hi(R3hvoMveZ5$atYFpscSG3T
z8~?Vy5#Ao!v{!+{$sw?FJAqBr{Nee>Mcg{Ck!|0#p}LPBWTJ3c%bHaysd`bMr%bu4
zbJZsIq2OShbJLZ&{hV|J&Hpd$UpSlDb+pgX9Mfb4r-xTM3G~|Qw>abHk_V<@H%aGf
zwY24+_sE-ifmRtQ#Ut;292;oP)5GX)W~)+-j?E+uSkJNbjsnj0xPHx>MN0)TM1WxQ
zLh;U@J(+U^8y}zS>FH@s){V_f2XkZN0HyEUy^(|ho~JaCk&zqCDJ51np2?DDdi=uU
z2n`%w<8HU-R(*Ya*K_?Ms2#hz?+DrSfrBbDJw5&9zZ-#zv!>bes;h6m0$SUe1x2^C
z_w=w8MH>*nULSzdfqtRc%v}1RdI*1XWMtgB$%7iE_OAGgzP>p)56qrzdj6jy1GJ&&
z>>iLrqn4JI+DDV3_BQ-E;8d9>QqQ34VlM(r&&PXD02jZ>AS57=+jO+EE8myigCXL~
zmGZr4N0yl8+3QTVu=inb!bORSjXk@8L2}UEc;unD9MPKm7(4OuaR3ZFP#I>^4*r3s
zOCRk06&Zk~Z}fz}`U`fb3Q9_k9Rk`cp>PK2a%hEhJqdvK-|tpFzT;bTc?E=8Xg3#3
zSiC`ZJs@|G=MZL^bG=G}Po`r|_M@8!i3+_ukEp0<NCgXN_pM3YF2>#-lWlqBRMfM;
zLtYK0gbx2whUdEUs8m(?G}ixdc!pGm*t>_)L|H0E<p~9p&AvTy!&|_$lvgSJNc2()
zPm-ZJy(?hK^Vv&AYD_AIB$CVNIzwhTCWRky^(0RC%`+C!f~E-yj75_@suBr3KOPZm
zcw|}fmn^Fq=RzpO*3;z8^y2IcY3Q&RPw9lrv*h&#zkRZ(-ad<ry+>mD<MX@a_#UgW
zy8d^}p<8_Z$9dNBa1Qf+`hmzS-%91m+3zXb@h@-NknsK+N7<${G_?)b4BJ@kKU$0L
zKYFHC&=`Y@XesIbfWDlYRt&A{zU`8wv+i&DeTprOO_P_!SU@aoxij>^w?*3HZO?H`
z;%cvXC=FSR>M2$UE|LjV#NHT3ASEvD7eyjD7f5FEB|-U3;kg%ilv9&;#neZ?>M^|+
z3DxNWB;`(%qpsiH(T`(hUe4=l83VhnibJre4Kfnd#87WaS>=VtL2_bQMcLU;Wk{5U
za{yMVlJowyqM`BWU)&HL)IhS&MgMKlH96X+52q4*(bircX(6oAemRJEGlsdPY(K6W
z$?szyU7jtX(?qLiT~_KB&1@k}Jc(2m(jmr-X^M+opHE^@=(NHgSD{&2)m=nRm6nmV
zt|@w^b=bNdh(e7@YFG{vMyYhey<b69Cl`~e`R^k1Jp5d)MZZoAotM|k)|S<2^SosQ
zMzP|L_GS%ynSQf2|8ayx9!7@LMWy@wlL)p!6grX^)nA^9kJ2&GjaxZ5oP2!!)rI}n
zn($%xn)chB+|ibKQLB4~K6{W}AL6}E_z{Z%8YO@T3_j1Y<r++U$dzqp0iIAsjqG}G
zNU>#QD~?1zF%yIZc!Z4B>UOXabUE-(zYAVoc^|kTdPTy9gPTCZh#2o+v_U$p8I9ok
zsgz3OKQM2n6S2NY<>gQ8T$m_an3R;%OiS5jw5JCJ4V#P-<_s!V!HvSaHaw~3`Mb*_
zD{t>Cslv{;vBF&@vF|mIWZ~N2da=H~{`8~O)k!4KlvPyJI4rk+jq&D-wjSo0(+hx9
zuIrn*{-XSS%LUm7eK9sRHY1~>$#lp#Nbfx^42{2jt=WI2JIz;7S?PQ>%BNmAD?b%3
z$~QAJ<8d*qsimj4p`Lq-KWd&BNyzqff3EtJ=<ZFA93yycY{7E3K#Gf*jm`P=-(Uh*
z=d@%Ax<-IqD6l_i{@ys{!v|6g^X%RFLHf|ImS*kCEBIX|7EaV*>)BX>|M^xFd{I=E
zbSt;}Dqk@nh(>lOp)#qabw(*yWE8Y5du->2dVs%TT29W=1=-QFnX4SEM~}}ANQX^Z
zZ$QB`|F?4n_^kolMnh8*LidsUUTfslbYtRsQO=n>j>Bc#71_cGPpKjnf8PHR1xW(?
zUZPtx4otQ)Da(#qg(QpjCpg|X1PasFW-wsOng-^*nmWKi&%~%dNx93;ev^hcBlms6
z`WZ^K^P<Xg`Cttr47nUtR76$E>^d1z{9q<1w(=T#(f|W>g339MXZ)+silBto0zWkj
zpVxnjRtSv83+BPM{t)s(=VP4wCsB!zSGxK<x;wEoDnaLxxv5Lu(b75ISxoX<bNhNh
zAq4^?hMdFX4a;2y@Jf+k!?*zM*0al&^JZjoi%aLj{numJP0xEq$TPQ&?a&%GHj)2J
zR9%yK6_z$yA3&H3$r&cBNcd08Uc@)(AynkU-?{#qWY>%?6<dee_U6I!vd$s$pcq-|
z@3kB=ekf*cSgqww+WZDl?(R-}a)f<fFbl(NE~r{;Fv^^Yzh)Pr3GFi*6qnrM?;w2s
zCn39*1G7%erbjLdF*Oa)HVJy9!x!(cMMb{V$s~JYGj7@EQdy1{XA&eqERR2<A~_IT
z%jCUrmAhpEJtiu@We&_Svi=sO!BjX4!=*NKH49AD_YO@w(*L*6g*k)rUl)NY87|5n
zg;;#kKYxU;#>^t@rAX+@29wIm4C^iCG(tW#>sRLcXE;;kVeyz&>gAYFA~IEpd=g5+
zN4-4C+^U3MKnE@0la+}<F9IluA0>H$=JcE)Yt0t<i%Ui}HkszNV*{30jz*IU3_-Oq
ziCaEy&7@f^$E<Aa=j3m^_GRBR!i`GkEKbUM;`pqg|0C(D0;24iu=G+AqNH?6cXue#
zDBay9-Q67`AdM*9jdXW+Nq2Yuhws1OLN3;2_dRoFo*Dt1E>&Lw0eBW6)o`e6Ax!uv
zsFaKqfXr8qu(Gp**<bFN70C7oYpCxTP5!F5mhfD$IHoV1Gkl&+`?<-<)uI-pr<y4D
zXlzhSKcK3IGj7WEnRvU}Xj<K^NqcUldXM-bcdlT$RK5S~Rj19CHrnR<ux7TMfo5&m
zof+W~&94J;?$zea$$w|oT+#0V7jWA7M()KJTyem|%`KU}z36p+2*{f!<J)uq2hxN3
z0NlT+yv|XTCSddAztJpF0zt1-5Tki)&Rq?HJzsDE&}c};1pw?K83y6?s3iwD(-bHj
z714PnHWAE?TSa@S#myOerzM0Y&(+Q-$3{$#izOu`i9OYLb(-}2>K3@TxLISSqad7X
z`{dy0=Emh#yP&P{Dd->A1@)p+Ts0uN#{mVRO;<h<&Ao|G@$*aZ0BA3l;~o|=GP0_L
zGaPSpM8xQ>!_8G~OH-kdT4McAF`d0DqNxhkOM+7KwF2XyxjD6`?v)#SkE=?Me}&|A
zvyz`)>OBrfRFL`k-P)keZhfs~xFX`EykrQDy!>J7(SpM;Xnd`clV5K&1tXY?{92mg
zikVR6YHf&GPVrmXCMU6;fiwr1L95|{74hh45U-I4)~Q2Y!=g;IMpjjnocx1KKCXWs
zt0m5B{2Vl<Zb!>9@z=kJ5k-(_6gm!am{7(fSbG(ncz7e&@F&|Ley@yB>X{~Uk|Zjl
z2H7c9k{o=ZM8xPxAQX0Sy2w%-lSsO$Cll4)c*}vS>KT2l{23nznh3fuB<U$bF#SF?
z3IMb2K_ID&Rz7#A0TOwA#fG|pWz<_t16X6?ShH~7w{%t=DUv(=cy^?!OR8(Uj9$LG
zFpftzwqocr802EY$Pv1{T4J*aTwzJ+UkF+p6CLDFI@@@Y=YH4@(ASoyT;cSFbPc^O
zKd5Z3YOfwp;o@q^c!RdsH!CLX{HI6bICTT9<@SS{yn!agMa&|<6Vc(-z$BhVL^qAU
zFJ<UbnuoR_b&fTiaksxgX^!EI-awbc=WTLjW#T=VUyuh$|MIX;K0^(f{EwbdD^83<
z_5L_k@&!T?cJ+NE1Ti5R^b+0OICP&`iqR#=s<N9JI_YB#joqLt1$0+7$QKHFrUKc2
zQ&En7$XwXB-`aRPtvt<bYZ`|@*br?j^<^V5Txw?L;KJql-egvFMR*pkA%t0-)GvU%
z7uQC7$LZ(yK-$@2D7~1nMDxTVBX%W2<#B``T_(a%aM1jkC~MTDv5B^-!yMDhVf{iS
zsF%k6(H?8qb<bHJX}%%D8u}C92I8RTVPXX2mL19cW%#bBsiC=XcS(U4i8nC^Ylpe!
z@eawNdOrXYD?r#6orFrRX{fy&nz88yljw1W>gFVvP}jj>v6{Lt$PI>YF_h(}bzO*5
z=oDTKsvQ<auaBJf-j@hD>64MC;AdawW$w`1FAVd7wClm;j#C+jQ&6{mIc!``@R&<W
z&+xON7f<N_dPt<3kOGNCNd(*R^tl_g4$Eax#>n?`pX`3!R#c8Y|JS7%=M-!@73aib
zHj0vQ@>t@1n97%CN{M4X+#jF8ZEMI7Pp$_hi(`*%UT~+|fxwx><6?7$lcIlvgD3_D
z1^{_}BgfYU{sFW_tWDZz7FM{Xe;XS5fk=tl_TMWi*;HLi%Lp)xywIPV02|RN+Bi6%
zdwFfHNB8CZSIhU$%j?I{`5BE(>MWV5D|7=~?pJiROHLa3l8gK5S43da85>hb{PDv%
znFAdAknE7%U9dh?7BL8P{Yfb)A1be8)A{<q2y=dZ&UY~?3erWwvi||&6hJvojv9|E
zfZYi>!ZqzTLIR+EfIU62bjf1t!Q%i<I6e1#TZi4P$J?~njN|7+JL3a1n!t^2AN9Q~
zVfv|uU=Tts)=Y^OZKI*IoZOF|>V7}Bd|!0Gcq+Myva;3}?PZU^6|=y@DjF*Xhm%36
zYxMrYbCG)UWzT5KmYnz9gtYX_7Y3M?H0R~pFEA4slP=z^vYR-1<?6odCHMXW%aS;e
z<!}<)s_|q+B1x&_Rg=(>M~;y1q=&B|=InmGOe|x>74QqkWJrNX6nagP1#X6>0UtD7
zkwcY&*yuy{F(i{R4I}#4d7Mhx%Zxtp@*`O(sL7JirV>)e<hg~6F(j#){&9vLNr(M}
ztDC#eLQ-ubxlSraegunR<ntOf@Li`QO!6n_?|QK0<razQ2!4DKhz{+aFu9e<2y(EA
zHWqX73e5wAbs^>p5hJQshbtfdtZbj{%2bdVF!h!Nk|W{eX^z#`^u3W=LKTj#zTBTQ
zr{CJPk@%zKm$G#?ug0KhR#|Pt?Y6-bj7Fee!7+4c2{(aOp0EGySNDTqczIqIXS!#a
z7h(C5QV~vDv$6<nieY1=WDYJmC3A`oR|K=`kv@p9M=}YbscOeTIOEHknD?}1n{Bmv
z;&t^6Gw}Nwi?bcI92-L|I>{N5>v(4}%-sB(rvZy3ii21Mg#oFmL5J*R89EvJ(UbcN
z3*$(@?!FpZ%4~Fv{P~(lclY64uuap?W!*H(OqYZ5dqkr_C-<Mo*u(ah{S@z%ew5l&
zB_|2?c7^}=<W6H6-v4&`liz6ErYsI0|64nclJ4o;x6|+5?Ksb0KzGy(smi?NrybP`
z!SrFtBq)5Vir&FjE5zmP_YT@9Q$a39DE&8<#2X4kfXw-QBy>$8j=@qNBSL|Qgbq@C
zrd=bal+@2A|2<0MU-hwefn?y@I2{>2$_}c3_Ib9hiF>%zHe*How1jk2S=@F&*22X;
z>h?~eVobPz$ZLongobP|YHKO}J-{rs=C+@bs4{LKRI+^1mF!#6j3W9~6M{T2XvA$u
zzOk8?OtrQl7cexCeMf=2jW?4P%q@?Ck0SuLqaBu!rXY7U?R-?jch`!YI;b^&r*he`
zAL8L0pk1mWk#Hl^$!fSvvLhUbpqL(e8Wi4?tG?1du%F_gtp0XUNd`)(W~`;TAbg_I
zoiOGOb)ZGA<<|Yx-lerI3>|>jFjEcHx<P4>+G2)4Q&LhUWn}yTIKXOq;P|gq8`$t{
zT_dAFs&p4<pXzG%y-Oajzyat3-UN}*($c<2QVtKTq6oMZYkr>zv5eSDeEgW4l9H`m
zdj4W!Fp$nSDoO;b((Fj>N*zCloyb=*9ws6inZF%D@E1nJDm>qfWCZ5?mlpys$61{_
z<%~aZ>2^wT^6JCw`Nh5GpWF8hQVWM1AEO9X0Ykj#wqKl^ycJ57PE989(h_{9RiO^^
zb7~w~TqFQ^3EjH<-|q2R<u@Ws0s_Mi4}5?EGjeqNF8jw{u>jT~J^6BT+Ptp{$%Rk;
zvYbl@w}zmH85G2Su^$O(w);kspXYykDKtj4X+Jsa^N_D{?$ZA25_@^+NJ&LCaNhgz
z*Uh=du}nNd9eCNzIg^adKB|lj;5}p<$@yS$uPx-n=#LKz1<K<~^oDwI_kX+*(4ZYp
zXPo~rA@L?y_k`lLVDE-FUo&29-PT>IBh4t0wq_SmAVF`}RN{qGj<Afv{eCw}V1aHR
z>CO)g>zeW<$uEy^l<30jbErZX7bLVjOOy#>S|d?rCS0&YO1hTB6pDz1jb2nNya`uL
z{cU(q9+MJdnp!$*RD6}qNAVRrhUza{hz`78Oxa)Fp>Zr4SV;l{Qs;lDP<o0~9GFHh
zYt-bgg<y4a#gbkXqTooDqS5lVwbyh<MCpFb2=#=zw+X9v1M8#l{isb93L{y0$hhi`
zUl|PHGNiXi%v)8~kF8$TwseIe2=?~gK!%jlo`lM6p0VP;&<i*`ds-HQQB8SPy`}wp
zq^W;!1L|t%FlS{TDJ>bUSoe1u<KtS7>uQ&?+D=J!2hWe0nmQ|UziOwp#b&&INQDRn
zRS<EAEyK(zydG|%UhvpNjobFFh;228p_#0;sF)%dK}{-ck&u+{JIwfZo%JGCRrv^6
z_N(wHje&JSW_rO^|KgImZ(u*GAz9@kLaKv`f{$!$_p`XQ6-${pnEl?p-$EP0n4*&;
zh3wNC9#RkBBNOm}3VkY>{&-};wntCsd4TwiK}WBOz5VQr1z$rWK!TZ-zm1<LDa-Rz
zD{(l3a=)q%X5jY}-S4nwU#>j=n2m;GhoAB_VyqnuR#~vcNbnt>dfL?y{v}9g3|Mgz
zZMOvzidb~@@5SxYTO!AGibGY^*2c73Is9PISzDVzF0v^B;wP=d^@8jh(0hZdtj8xG
z<H|PTA?gH1w2P4T5VxG82m!~LZ5%eeOr}^$GO_`9OY4$vM~gIIHjojx?US>==#?2R
zGJu=r@hqr@FKWe!BxI>mrupT_Geis#5^8#Pej<u?g7(zrJ@R51h}cedCk$`Yt01&T
znHd&Zwsm+a<K4}e%u7`F{+j>$uE~6Xv1RvO#>CJNmIM|YR*F{Ow>V(=hs&med1;ib
zTv(0#aC@dl(33me|3l`HQqysuSw%B=rK}(#*=7bVA*-3tQW@@sFi#ox!!v<)cr{41
zyh=+;Pu~KQ3mqNZ%aCDwos*Muu+owN`U(ckI{C0w9q&iSqn2kc0EmKkZ4`KztK*fC
zjScdb_Qk>?{QLK}F9QdV;v}c18(UkK&@H*WIty>P34h@_fVMl77^14Crsw9C0!So3
zIvtKHX=%swE9<qX^RWzkrRzT#GkYtpkITcuvWUa{U-pyb1VM}|z*K%TkF|6I-*+t8
zN2czCC5qSYdA_D2JS72?<=f4PUUyH=i>Sqv<0s$K&AMi@M{+DiZTGEqCs>oWXxH=p
z#R&3$o$J8NB8W?v`KM-D2enZ+6Nq4<G`jB?(OuWN5=`g@$;AQ~?B?a_+@@1wB)T7{
zt_(WOgMeFB)zJa0b+8vG)GLe`I-mPM>@8@{4n+5D&>v1P;aBvJnn<lH=nTMK2QUGi
zO>5Yk_Q`q0f$W^bBE2b>uk<bMx<|e(&yS!2iK6j$3O)os$9P{vKViJ!O|57N`Hu$A
z<TG{Ga0gXm5PGkFBDOlB#Ly8C48xl+8AT~b2Q2NEq8L*ErO0ZVnn&X4Vd|F|ip%CG
z<!GbrUd*HLXsH)td{vGS!>~)u4ED>-(<Tcco75GG!JZv8T1jevSBSw*`AZF{cA_8g
z^P$csiO6N5X~&7MOA*Rt>sHUs{7`}=rVvtn6lKEXLmbm7X=N&1HEE6rFDW=-sud#^
zCn1CRS(rdFf<yJK0nYZFgai|ZWVAr!Zc^VZhO+xia-*5xHNP7^!WJ3}yFef_+-1s}
z<(2c3G&)VU`1;#-&U3t>Yd*sBMvaxjlY;`Sm|}_f_3)Jl7K1k8np`cdW1DuryQy$D
zenv!At1f7oIGHDYF_E($2}NUN)yZihjIlE1Q4?t+|2#EkdYzQ~UKrmQJ3|aZY1gN^
zWA=@OO&NU$T8<22cNJlxQU|&boe1pkDjh7QRO~jj(&*eAWh`cv7Ban&rUoZB-RW)?
zLoPG_N6uO-HDMWewvJa8-#8dDwt9=B&+2u&66o-BDM^jDqR3#RvfPVotYVCiD3H6i
zsZ7~F;^0PNP!DVINH$FqTL%*G)kIGo*d8SQ{3KH6c4YN#+3ma6pGb5le6!(-iVG-=
zPbfh}y`Q)9T7A{^NJ63Gs5|jwO0Xn^2tE?S`is;zcZ-zORkZJ(Z!?RG%CXk)T9s%c
z&zakl|NYyUAAkJK(`+pFk#jG=&Zi6I_{00UvjbrT>Tw63=`iXuj;B`!)^tpdUZ0(N
zNxK|{Eq?9#87m1Z)Df>~8g|;XZ%yN^Jgk?VJs!6VO~`qt%6Y8)NXE(NR!0>f4Twg8
zV6-<~5Ayur?oh1aNPHEN6To|1=il<Y^2GntC-`XMvA-~a@u_~{D+#PHj(#v+bp}V-
zIZsuAHpJTDLtY}sDty^_X3t8sJ7>{~DPM>=$4{eqFR%s8(z3|%{HswsP@pE$Z2Nw~
z5;vL*VC+-w2rq1se6%RBmYzDoPNy&M;~japT{&f@N)J5UlwtHKO~jd*nHfNJ26GUA
zAZluBxgB=U(F7luKzC!Jub(}+k88kXw+8#dz*t*H0oJ4u>%m=xwhq<Z+&t{|*7c<r
z1tMdsYHAvSU$DvGzJG57hO*>j+~HL24nTDDfPMtj%EoL7kAU96LFqmhw~u9lQB$Z*
zNLvNp-sp^95NF~UDLMvz$o;IV0{5&mJ>ee^8XC%SUyP^Q1ST2!+QoJgfpgpd^oYnv
zW`2GxzYj!%C`_E3A1mh!9UQPoU;z}%cwbJ){VjlRq<;7JPcVj^Z;zb6j8|YstvRiy
zjwYJr-<si!b1&Hk)F45de?ji`^SOGiuF3Y>!%$2(D6?~OdU|@kKtwy<T3)v^&2<d9
zE;4xw>w4=eh`@icG?0CN{m&rauTwc*+q&8jkg)&9q+c!AoCC+qxm3%A6ou_oPj|-g
z)58g;)YISVK`jvL1*+h~W6SgP?Vsn1jKsr}tGy#W(MMLqg9v6*uann3O`ldpOpAqZ
zR{Ou!lUBpTg-YHj*I~Ki5+}6h`DL$Aed-P>lJMR8EymNn;fbdpO&`%4#vo~68Z9d#
zO)TH7pYsY{-VpA?2go{}LR@Aft*_aBt3FoUH%oNbBw~k0-Re%f$m#Ct$8qTPB{wW*
z1OmN?-_$HT?_~x=FneT|88qdF>GoN798JkF)`#H7KSVH$=EL(#FcM{H!459DvgX7)
zTGv0zN^DEy$$dJLuJ~q!kT1%?$y-!hY`=FQc_w0a(&|Nt4=2pjS8h-{wyb7kXqmt^
za7*-j``o+O?D;VLd}>2V7K*AU@#oD^+TXMV0bAO~nyn#~k<s1S-HRG@maUs&BHiaU
z+}A|76lHPFI+`LJ#<K)f$+-OOYyAV#(j15Z+3`lO200Kr(T^&GCbV9!W)TI66f>(R
zBSZ`da(v}^5F*5MzgrL`n$UcZQdnXb3dN;5z<w9*q-7F($8UwQetwUO9)E>0G>R_9
z9Jpt{I<>u9YRTGoti1OnIB#U-^Vh<hs&JfL)YlWcs9a9E@B!p9_JsA@?_|@sdrrwJ
zi)?E0IfXV;8U<|=HuqJyZy%j;_e-<x7I2^rMar}3v|zb)t0{<`1r%t4=7l*Cs;M)T
zOGyHW<+SFQLtqqh!S)Lm<%Nox`o4Xerce~oci7mfYR$vdT^0wHHed1xNHUe}jcVvP
z9o*ouv*XY3+3Uo+t&mE-rWu*n#lGzoQobXzs3?j)#I32lXTxKc=#enx*f&%YfDJ?<
zp|HcMtT1=kLVgx(em>IhxSMF{4n-3TK`)WfL~p3E65i1dwbbvk+LQj$UwI4F&GUQv
zsPSm+`3k7672!Shc}GOv&Z=S4TvW<8`z##|7f|*`e;7T-50BoTy`T4V?fa4cV0^qz
z<v}x58*?U|A~>9md-AFzMQ>Z@$IPUXl$_W4wNEG7G4W-~zM8-@*VE;(qr0)1MC}4S
zuink|v6Qy9<*`;O5M0R7tTBf9<>pcpRU1XmeEB1*fF*9`0O^P^=!q%^D)-r09JR#L
zk>^`MlNIpb9JV~x4%~n;fCx^pz){`$Ymm!e^Y;`U5JJ|q;R$gXHcj6Fplf^bPxzHK
z=#a9hvsaFP@NlfC)Bri1(b*o@bE{Wq^rddBd#KWJs)W?*h;R1N?c?*}6pOK=X>sz;
zpB?HN(DBNJfzv4fABqQOdTV<d%wBrN#>Ut{3WBn8=|;fg0I_}bAWC$HZZe#@3d}4;
zrKLY)_j!;1#=i(v!JA9mrsEn_Iz-lh%Xsp<MJpY6({42DnyKPAcz13LB4l4)n+Zh?
z)JITm0s{l%-eQR)a56qU6zGRFRUn{wcuLMS2JoX_+9z*wFIoQp2rpol$jHbxT~!ZT
z2*fc!bMzwDP33cqdqHhdx$N|@a}JH(5sJCx2O?NKC4m(XOU^4pE02xd*fbO?Pgk|h
zOH2psKOe}a>#o(pmOMGXh^NceP?jm{RtuZ|iIrN5kth^zD3{Gb`lS#sjwva;ygP=b
zdA^laC8b<f0-IKhYMGw1?ngR;i%TSm0F4VjtDAQ#NeWFf_JL@5m+KW(HhQR#Mc}ww
zFGRd!77~gzUOkKnC6c0-h`rgjg)Q;?^f!}hfW=|VX(%oDJ;S)j#J*GSou|Y;R3x>S
zI*hvBck!@y7C)ibuF9WbYGVnk%BQuq&#oC+1i1SzW-;z8!oJuv$zYjpZGB88EH1BM
z{&zStjG08AP5f_+>gdm03Q4}W=)7BC&c|a)-309!%qi0B-li*wFY?hagr0ZWC0g3s
zh2Ab<g(}Jz=5b!l-B}$I(sG2l>nVoY--g|ESa?}_uG~KT;T2*t_!>K5IT^B!&T2_v
zY-!Xs(Lo(W()-4xDbSH$tHY;E8{f~A9~OK1yH<4@bzoIcRnt(fS0D%62T0jxGUkBk
z%j0TO^$GjgI&tkH798rqyi7H;zOT`RrCvXeOCF){d5Ce6#e$f>OiUaIIwZ{9`b$d>
zee1cw`NCQ`i7yjIN_Kzq1|=bLNp5cMg3I%coFwQt-8G(RpS0rjknx~0vu<_oF0o{+
zDov{?TL&+*eU8O9Vv-2b3K6|35sUS^mU+E|VcWEe7Mr5Y_FP&^E1NGUZabu8C~3oz
z@a@dsYJ`3r5zr;<lk7WKb9OE%4MiuANEV9=E=kaUy%k82tv)MXBcxar0x1XFF7Cme
z;PzhAc=FI!qFBMy-*X&=6VYy-8g0tzTq2_dJX3UT0(Fed&4Yv*j5=_nh$-wmdk;_)
zBjk@Z<ims-`1+Vo>i36vZ*88JR}5l){{@~iQLEoSf9@?E%Hhv8&1^MZ?>FJm>F_Kq
zFEdm-c=>|`ffN&49oqYsJfN*-BXvnreLH(wZDNBGY7*uVTB=@9+;6}a{`F98Us${P
zVk4P1_p93S*<shi*4(bcr->><eBW^~^d4s5lLZ|<XcQD#9AC_9-@e5GSt%fgZCV~V
zK;j@^xMyco>rd~8&dYlUX?f9JU>4xKx1W&VB*~?Nie}8<zyr&)R!@6~Oh6w{>UCKN
z#+SFYLYrVR1Mq4kK#jrN3y!Df{Z)BGgZdT}*el;WQb1|P_+|U!RLoj*7WR>fqv?~|
zh5Zs3cIcY+nDY{=JQvu!Q6BhS{k@#l@Oa$Up&DRG8KM+MXX57{0`WBiikFLvnsZ0R
z)JXpc?z5?PoEpJ41lcqmxYoH!lL7nTMpOPmHKVrWeY}fpNhLygIGtn1_ga5l$|K}X
zEi(m6AMQ3sRG#h8Qdis~<2M~QGFFZkrVO@9acy8hAXY>~L<~8i5cf<sSMUWWS{-X^
zq;i6fY#BG5R8EfNC+BO=PE0rl&~Z;(@WQQH&P!b@_wL>=tI$DeiDt9)bEBXo|E*ob
zHR=I+qrJP&s#pUmvIU0v8#S3MMR2i0%a9`#^_FvVr1-fJBL)m)7IIolN~EY#g#PM!
zZKlgDwqEU)mQEwouw23d&G%Nj?o3<ZOMrqH`bP#7WhI|&+t{i=QnD_t22{kTdR8Po
zRGglo5Ix=UH_XAG)c(w{<`wB$*vHqsd(%mh>~zdL!lHiZ(~8BkSQbZ})%!+qid-he
zWA&$Z=O6-Q9=kbKQ_?0fZmiboDZ-I3VY<GR*>{X7a$359!}o6ZKLcVnFZz^zQ*xG(
zn%M1P@ix&P-L;aX3N{W4J|R5QR@&Pi2=aBDYHZc--grq!sI%eAtbI}}3^tteLCbJS
zHSyYZot|MlN!IZUJ1xi9dVOSYxu*n;KR$!XW>B>NI5I6kV9e~<*f3Rdqov<;+94)P
zBw~J7$Ucr2Fr|wsNl(HAPexILvkYY(UNt<684R2RhW?Y>ADA~j<h-MR<q*Tcb}9~}
zAb3w^EkkT2*77s17A42mP#O=K#bNTJxoG*A2~)3{z@Pf3apV9=ku127n5|ywNaOf!
z)%fm7g5rA7h(RJ_2dDJpSRNaXV{wjm47NREFsg6M{x%OO%8fxO(v(Y}*ncc=^Y&J;
z7=!0|ta^~$MIGggfPOmWgE21xYq`_7Y*Y9Og-wna=5eqe$+XT88RqL|HyOTrK~*s*
zO3=vE)+*3Y&EZ4!d{?o$h<uJzHq6hSkur;%G_A%nwL;9vf{{jTjKJym;r)yYh{sNy
zr-5g8>&8PEAL?p5k<V*x@BQtc3$ks4FbTd`6j}6L`|?_EKi_%Z@`9Y(hiKRJp<3%#
zzcNU$#V>zfHl39`P%6LG>OkAdRFZj{UTsmWZiG8x5zDteYRcvm+WctXwiL(sURkO}
zi43-kws+=vQ84lOX&IAMo<%ObzOW*6nSZ~L^QNa_I`MGg!ToMerA6HdtG6No#92U{
zjEbV!ff7(&QK7iZb$$QH^0YCS%xY9v(bXjm`VHr0SE?782hsDo_aGQ!KydZO&FMM;
zuT#X>*cc1j+)vb(?&N90`%VskB4^1X9zo^h%r7I+__*RNVE{S^o|&-Z2)5At0a}?e
z7v=M(D<Er=nfE^Q{gL7AE&WJv$Ne4!zOg?{si)HWuknN=**^(qv=OY^pU^h_k`oEw
zjetcckt`PIn$feym6C;1A9&>khlg1}0Rb)3%Kh>+7`--r|Fj0}Pi0L__wev=MvCOw
zNQ<+}u88ATsd=Dy*>unM<3<ZBI6B<B+!%TO;4*n9F;IQo09Dwm{M<4lcICz|AtCYE
zxxR(@%*P7!ER3GJ{~<>sURRZiP779>m+Nh5KhSSjF0}n<B{5&xv8Uk`@Rj8NFr^>=
zM@IM1-aWLFcvhLOr(AdiI<yt4agrRBw1g$RSigLcRvQY3ZK{xO8B#v<_Ae02mepI8
z{Z4#G7>&I*zlYhMjiQ7?l};F^u0mxKc2ryezU|#-jJp+zZzcSPoqN7^<yZ`i&UG5E
z<8_K{GZysLtS^Ob-n6bzybdEYOB_Fb+e_LB@l}L_-gNL$L`Muoh5i*wO@V6&<<BM7
z;|Iboe=X3wStW7eC}qZrZne5To9EQ|carta;GQ&U$uXj@oF=j40vC&4w%I)`kai_A
z!^*&@zta$uop?05$EbZayVV2uNYl9QfnP4%tP&FwxvHLt*WctwM-CBN^=@>c_z#o%
z7T9l|u`5rj;EqzEu)(|?Hz&+q@%HB%X${c5kXNxKyP3rCAA4t^#W=>YDXo4+`q7;F
zEcFvo2$|)-yP?m&XmvE03WPSfCDBQpc*N8dcAx{ha3aev%`}Iz5{cK6j4c9&pxF|I
zfA@aRjwNR^$M}s)Dh96{i|-$$8A@O#KcCDJc7$|mEbJ~>Ce%rcYb1|vg5DV+BU@JE
z5cu<IWc)KUgY$jSc1n~F;lJ_u0{n+%A#Sww5C4kq)~_3W2fI3}%Fk!BL)cR@S7Rz7
z;zY9t1e!bU6Dr<>;_@U^+m%j<-12+JS6HZeH~HCCaGz<$zjnKD=Pp+E|Mf0rQojhJ
zqXz?4uNAcA!+U#pU|T}QW^6DKSsr`rcydaNU!C^HULUVNQUagKeSeQR3`zj^D2j;e
zqBGGmIQ=bK&rd7Qf~s{jCxr3m6Yss+t`6Fk72XP|a2|k}oMj8f+K0D==Bpa#@#{-?
zB(#_-V-F?qg(v$`lQW*$rp*o_j;p=If8D<|wbniQ`YihY!BbyEB(GpSjuZ?JeE<G^
z+;{Kt3z$OCMg;EZz%l+0%Q$KXKRvYr<5*usQN*v4b53Sv1)kuht)uop<Vkb(|1xK0
zUIN_+_*`O6PfsiU>;c#nDBcXiJwC!e^L&($=x@~${6naFUth=~ru6TC-B7U~(?^9p
zT&e6Jx*CnPghfutc+@*UPu0&Wb8_dF!#@4{S?;G6M%RDqQZZnZ0a)jYT8!|EV1|^G
z)XV=vE(w&X3|bw>T{TTBryaw*f<#a3%G>?^t9#c4=ZbJjN>{_tdyY5Strc$4e>Gd3
zZOy9HbD8t*?uF4$000V_lYX5$Deq7~;{9R+9svU!Sg0lI+R0ou$_3tfxm2z;@AH)h
z&Ijk`CGVd0b~%$sckfn)k=PF2`fK$Ix?x?CG5KGel%mGCOW#rAuhI&K!3&IY84rDs
z%o0+AbOg{X2$PycR8hmj#7U%y(0qL*p-%dS1RW=rIS?7*<ow9s=PN{<ON^BnfOEy|
zxKPO{tvVwMrT3U<Wy`JQ*2~lIElDZ!0+ocd8c%gV$jG!fb$!}A8$G2u1l@lMK49YH
zt*dYeTwu(G2s4NJpV_aL#jkrOLXvMoiX){Ko$UQXnLjlS$RBhR_{pmS!@oKU4lGuI
ztR}MLmx0+EuJ?0YdFXngRm0rwdq^O<GvR0S;d=kAc=LG=BP&097f=L<IX7lDP36j_
z1tl3eo+1I^MQ^TfU#j+RobT%wZpr&s7@z)l7MLP-%9H=Dc8W+lK|F}s{VE>CK1?xJ
zKoP>!jl!rq;|W!kNaxrcL5|Y9aF`X_l^M{k8aai@hNCCnt{a%k&VtmhACRj<oe<bC
z6@Y_Ro@oRZn=txGUY+}6o@4QTM7b%PxURxDroj#;Sw;CfMvrHdZ-+cz3}oh)tauji
zH7%~^I1-y%FoYUke;EG5YZ<Tn{<xfJxz+8Mo6qYF3hv(~L2Q)!YDFF!xJBD|X!ZK#
zzne4fe$WIX*mDvLOfd({ic;QpZ)LONe2FpP$BPV!WrmW-PH0<!u>T}k;MI7sLm^D7
zaDqnAD|GDC?n2%sIDx!cy>DX=1UP*Ag%Xm3!_Vf;ce@e)aJX_u|CI5(rUM8Ht=hzl
z*wqfh>k>P6T<*HIJ?dOl4Gr3y&jmabXuW(=t7KBIUM>@k(*9JJnjY3orBP)g%xiP!
zD`-Bnt%qY3Vez+yH)UEQKCaV!Eof*04-$1M>)XPa%)e3oYU$`ujo1UC10es~;HHg;
zib~8CJchk0UGeYZiUJ;300=ZJ9WnfK+#19^YS?G6ZrG({(Ei(7+wy1ys7cqgP%4lG
z$B>o?awPu#9lYG18K^(GH9K^2{~_5*)UkJZY5xTdbVwJaT_(>;BM8r~iAF|{GK3VR
zrX(UI+cnWV--=WndYC&%ig|jLH_EeHp)PNf*5+n$?Xo`Le}P<MT&)2_)cP%4vmiX0
z;rzZfc;!<1GwOqGNpsZRvC@dbb0rZx?6rW?WqCmbVkzpS0>4-D!wUOPnHM`++RE*F
zK-YjzT|mJY!;MJ`F2FTC-bk(9_{P-(y!jgq?|qxog_g&Ks)fT0E{kh{p*Ko*7ZN-I
zC{$m3%8_#H0|5J(`Rz;_TMtXTpsyMV8j><e>WeB8071-2;>wRz!!e7*6vNDg4tTqv
z9!nF+4yAyD5kC6;B&fKxotQi#wej7qlOZm=m<W%30HhUZ2G^jj{?qDwh6YLM>vTOE
zjeXHXdUdm}u`urx{E1QWX7-Yx;l+bKd;`PkYw{sC5o#f_7*bd^2qXuuU)}Omq_KZV
zF9sVm`QXq96Q`gsJ&yQ1hXh2TTXb-WA&@~x^=p>GhmQf678IY2dWpZ;n~FxTz_vSH
z*BvJQl$;=x<A1_f_IT9VGTy?;AF=av4fk^PVNV=R)VHseO1$i|dR;CCk)-R{*|7GC
z;K*w)N4j(?{!3kuEUO4FTBi3t=?9cL@2RJiAuJii{5*+mrRbMP5~2%S{P!d%ux5W)
z113A-m(`hu%Pj-Nk^Qm!#$tcd?wakOe!EkXWsMCGLI30%0cW?f{81maK)xez3@Vuz
zQU<Be9F0(S&{+34o@Y^DCBc2axp(I1M@ofAmmD$6?5AlOoEKqH_Ew+y6H2<$T>PAZ
zq8?{6y>;)tmu^TF#~2D!j&pY1$b%o-Bny&nSw~*%ndE~_Xa#RTwf@YT=Ib?Do4@M&
ztqqNst<=K5Q9sw$B`6nWj!E~gzwZs9{uGjf2iyX5Kk=;u9t&IAn{Qnc#$?^8i1Vj3
zJrr7wQQnHuhht$Hq_SSjJdKxv&n(&d-sYKp=?N{-#Y=s?>9d&E*8y=t2UpuoDKYKm
zb@p$|C$ADRH1_dQR2l_bX3|E=7Bqs+R!S>yTT6yut$k_=riWW(XSL#%kWc;vk*Pqi
zht`0*W5=~j)yiPYu~%SWcqAhs?q&9UkykJCg`5Iy(u8)9txXXUSF1gc-z7kGzl1~q
zJtmMi0YTJ&H8NOk^n#Y@+S!b#3I0j%GGoKb)h@?dRe<YsmkJp&uKe441#Nzeay2--
zY*D6OPz25UtPM2UYKh-_vI+`|f;R8Y%kU;v=bE^pq!KHq%*(5)zV^cu5Kj?}G;QBT
zLz5nh?>*$5f@>7p;gG)Th<@NHZ{{pVdw-dwl5yznykW|G%#veZ{m`)ltXx~P9|5Pg
zt@NS4lu1xfM>}Krh9qFNtM|@g3`ewZYHh~ra0ZC>*H)euyzkSE4v4m0bXplk!_bAP
zNpWh9pjqkO?q-$1>whPyU|^4r?VL91=tIwWtBh^PW59b{-a%R1XSS9psWpvWFOkdt
zi%r!^!BrStY*sVt9E%q2v%eWf-*?iP#7{x6dLlUF5DY&!D|$#MgruxPdHb!rAMB5s
zZ9f&MElD@}uCke8I4x?Hct4I#y@&{CQju{>pZ;&ZDqtaN)Zc8hld~9xG6Tg(5XDe`
z3S)|OWUfF-vUGQa@3UDjujd#T*6Um9r<j#Thx*IN^3`<cnxv&&LM@nm;vFc*3hCua
z)8!!^WoRjt=T}agaoudRIrHVU0cH-n8H8hfONR*Z&>d8L@-G|4vDV8}ALDTSBiI$F
z%j+E|hQDU9%E%Y$1(SM_(05Zir01M1LK(m+`XkbP#z6_|MJi%RQcp9q#494FLHcIU
z_Y0OSc6cmJp!g1=yuIUqY(=jgtg466@QOr%MT86@?MqIK``&j{|9<okzad{QJGUG|
zjPkcNahRwydQ1`_Y?vQU{W~^jVH|{CY!K;iZH65zKkY<CDG`q8(l?)Etm*8EyM;uJ
ztM}syf{VSGO!*Sd!Xux(N3!)-&L^&MIFk&6wzHq{kCCo)8pU^m`E)kNQVTaf3Yg99
z4GIc!nVJ^U8+<~7E9h0I7k>-Qi$^LJM*K_LMllac1V=XAEnK!`i1G34Pu=bGLJ-&N
z_vH@tG@+nhX4v7mqewK1kqX*agR&zhGtcun>+SXS4`VaUK{1)ZD8FL#vGH%?3EX`r
z+^z7uSo85p_}15x;{m#J=>}O{Oa+fV3dU7dE1{-KO<@ZDv?YgB>pCa?GoKe(zTb$s
znzapX`5yd*t~Dd8q9eB(`f6IpQ$1g}8eG6sg?ge<W2+iL58&JDC$T}sYog5d_O>f<
z3IOQH66E0=UrWtoAuu*~?r&QGz0Zp&9L?h&3>zNmQS-g^%hfF0gix+Mvqdz}=vzrg
zYY*FmJU0?<(#F@ijEwW(enC9LYN_;YSWp&_?gR4z9KZT!m4;lsUW47*h>4!cxzF0w
zwM>hz##>9M37+{tpc9)67*v5@6tn?4jV8|zD=N9r$6YCpWfmK9?(~mlLO-7W@@xwp
z96zy*Ja&|*)!7-W%aA{H@cw`T>CLW3X#m*<MtH{OGsc%na-fgW<l61e>s@@;k^$sT
zoL=jkT;5N7Adc0$iWWW>FY;m?=NNgG0#hFsz4w(-;6Nd!wMLw31ji?<&@(LV9n5zy
zhFAs=1$DA&irDc$)R~PF25%oaw#(T}UM;sY)S@xzscz;#5qJz`e6cO+`$Zv)0P;v4
z*BsyBx5}_$dVVOP3CyZF!Vr}1dVSyahGTq-Gz+J^{5M=pNhtaROIa~HdZk~(ze7pT
zC>AS(h9^M$V~iC)nhG1g{fyVn<;NssE^G?N5z-5{fmehy1|wNGt?y6~wX1Qr@ILHs
zcO|0^Y*wf_aZWg2stk!=>d)iU^&FzV87faF6n@D+Z||;`fijoFbHPX#6FOh$Lxm<r
zZwTodzUu>gOKYdw(O_#5h{2uN=RYg!)icB6yBEMJVvwX-8LysDj1d+=?N>GPF&U+*
z6|2Tl-n_V5JyMxlJU#21$~#2yS6wkDh(4{vUu7*tzRqSYr;6bTXCqYwdJ!Y`#eSez
zIU<V?lSgNMRXnEd$CSxpiKQ68UKJZq4@ZWG@93Z#1AivMLQP&jmM0%i&MuLs7bq+x
zN=lIxv9q6~?uliF%28%IX6VUlTwgj@hG#_4H2kjT^&kk7QX8z<-AFvlZvEF_0V^z&
zGoAUx0q2b1pH{hd|7k~J*3t^6;F^j1TMxm#bHSmTwwY(`EjSDE$7aeA_q=xI5(S)z
zL#J1frjeLb%mV{6H%-qsL@s;ze_T)6f+gd+QU76x!&0`pY$1Y$j7$)SVQ;Kx@U2D0
zGoQ&w>RSI)p;+@i#9N>JzK5fLKCZyhkI2iu#2QKIPu@zM=kB=8VmkL31HaDjc~<pQ
zg+=KME$cdXFcFD8T}um83KC5V>Etg>p0=l~k(IWFZ4RbzqKP4SGy9vz?yuCRW|Vl9
zT>ed(vf&UBF={2drk?|u8%zSeUP=Q4BajN?UoSZmt=vu5CM6~P>%!In-|apH)74KM
zc4gow&j7n(<K2&&Y6mItQQ%+lQ^Cdn04g<OCMYKl^NkG5G|n*n6K)tmxj3HlPyYZ&
z=Bzo(r`G!QXOIY&pgW&|4VFw-2t3t4*8dw}EY-N!Y%#CVy;Qv)Zzh^n2Za6{XorC?
zc4D>F7Y4`(m6etIZ_h^dqbmieyi?YDZXZDK+C_>~F4mhjZ#wNXKU46X*H=|_9zPu&
zpMhP=rvM}S{w;9tR>|?~&K1?q2jtbAzwAbs_#<b3xi_UX=OQ?K`lMOj{Qieywv`nD
zesBzgRtvYdS3xa^j)hdo$I;pKvjwz42+SWCq?{PHMOz+rf+50x{kf-1vYnrOA;I|<
zC^YxYQh#kCBbdBTUb!avV})>aLJ&9t(1(RBv@#2-7|ca!vAg6Mf=UQC`@K5IJ*^^<
zGu{%n`}m-T+rLlACLVcP{YwI;F0;F6RFpokan_Hi6;Fox%~!oxQeu2+ORaHhXKt5`
zSC2zR4NHU)7xHSYJ3C(ERZ{2&Oaf_kM~mW7M6L;c7uNgn8NXdT1%z^Kc2W@<YdFnf
zq;O6zOZ2@`n8Z+z!-lh?<GMY<1rQJrhZ+E5MjNq_MPbk60lBmHiRj<kd9`=IoR=M{
z3?k=!runl~<L9BvjBG2XOAzvq+lCe&?8tu2Hd)Qs&)|s?Q1z;(NmO!AhMK9|RKi<;
zrJXnz?n$#T<jkZ|7_J;x@pG%LMojI2B={ES@7Boed=|Olg;L&axW0-dmo!K3f~_I0
zdjD;qJp_*)KieT7UrqrcHNX6?e;#jak}q5Zd&dO#+H1#JeTaujE}c}8w4vTH_x|(O
z&XM&P7yF1@YODp`_f+L&cq_>2A+NUGQD+wyH=>_*qH$1I2wiWOC3?Dp=`u;e49-?1
zjy{(+6Oe;Pp6|9Z-s6h#nF3_N?CF459L3kT*BRq^EP7MK<4q~tXwU_gk<at2wuI!a
z#=WfHjLg)RO;0yDEzsV1a&LntkfQ8#zOjiXXK#N=L;t+PyGdnRiQUpgXfyeJi`i{l
zp*ImBv2rru$PHM!{F+DrAf3Y{rm+Bg!ITH;j~7!e(AbK!tN^k9@+t^N&s8pbk$hDc
zegR>I^;-B9oEd;9{Z<D^-<leUl;6NT!_3Uw_QH?8#FwnC>Gx%J*$wJ4QFCs5U@Az|
z@eH+QWGYESY}#4=q0ety3jG9(JS<h|B<HQ8FwowFH`^!gaUagzYS<*>@bQjUN3P7d
znr@B3somT(0+Z@z*8X6D3lhNU9E0n1j<lq|O1bjJD_tTVKe$3m?eZ>fc~p`HM1F0&
zWDuLBob{4LGjVVz`_=6YB?@Bmm6d`|^>Xsdi}7a@faTt5+5>{t=#NiqPOd>!^Yt>Q
zs<gpILwcq1#pNHI%TdKwIh?pPU7#h=QSL|7A{Fsykq*jAv^K&dh7^)6=cy%pVxcjD
zAO;Z9AB&a=l!upE_2v!%d8?YV{dFu~v#`3=103hc5LK;uCn89P(oh!fy#6@tL^vLf
zB(Rs9HciXIuwzNiF$sl{c>cu3617F}Q*S!bcl3W{rO7B$JjArq=13YMBSyeh_suMp
zEP7?+!XqPh6V-UThs*1LT54utV}KtTceEhdAS@vi*B7UVEd0I5(pIM@XFQt$q5D6=
zuZ2MeC@$-Vv;o=O&7o!~LwG&bJfE5{pi)FXZmVj|lZKHq{yia8Fy!*wQhLXL{|h%)
z9!8>v4B8AaS9o7w+T5@8M?z}QnEyN<<EYzCLlr{<tftwguo&V1^(M)bNz<qZ;#{Fk
zxM@|+==qqdJ41f~1^doGL8;I^G|j|JT%MkYZph|cH=ciI+~htEZ7P%mukMfEDWtCv
zb9_Egl8cO%VQ5hP?!iGNXDKMBq2ji`{IIYurD=ussgWih!hL$uAuSzsdiArkvSyw#
z!~$oXW4R!2+QwW^u7iayIGBYhmo7~<BfFP2LNtUKQweA9lCry%=y8=upB|toU&!r-
z`Uud!Cu$ru&?1SaPKMaR|Cj%@%{$^1tfJ%nh9*r}exFDudF^9l0djRis|veKv*osM
z<a51ACEsX)aiS!RRE}8swvhuwK5?v|NzZfLJ)=bl&fFa)q0Q=q5ap(HRU&ktCF8@*
zdB<~<Nh_c2r2;x3`YmR@$@_=y?PNX+dXTB}A|2Ota!LgHQD7&n1F|VF{Cv>L1y;6f
z>ZH_EI!D`XoEP`Vgee=a3A=xa3jm4W;N0~e)EHL+y-!bnJ+H?N6Hvbg==B0anh;J9
zFu)cJRt=n|Di<mZw+Pk{%Bx6?YunVS_(M2_d}eU(&{y{i?$C4;3Bg^-x&ZGlJU5q#
zR8+W~lwp9z<=OHTZLfe6x>-{bwXzxPtJ>e+2TG&z=H@8oAP@o<lb=ts>~StL4HRp`
zginvR*IHI`H|H!s+b&flz*nD=zundlo%&yzY6SDkq1tp*W0$?L?_VKki)24qSb5?A
z5)uhl|BH3Qqh;(ed`nAR396V&2uXWfc^l7<bR{?rxv|?<VUuR<LGwho@R0YJRZ?Tu
zxQ<3#qzrs@P1Ri+8v*vENa$IB$eyvlDiHdE+b*<BL~Mb`gB0SA!~JW{pSttIrr4u4
zLN=RI-W>DimC+w)$y5R2YS_u({`ma%^)Gc%bchHpdJt4QPE3$+N^CBiq^5zPWvaS;
z0Cj%7kYQGNJ85-pVgQ>(oce1;xQNK%F>9M|57UoDRwZBL_V_MOdrUdV60ucW*S}9_
z)|a(E)w`<8boGxTy+JThDA!+kpr(@Z%+T=H@b#n0Xvl#181>?hBav>qxuIT<&pAMm
znw$RsZw~0*m99XGmGycXPw?Qmgp0Bg>3zEs(C60u#*(#jFv><^W+TKe3CU?~I$rH7
zm@3O1ApS$mHmzW*irkJw4rPZkCT-4!M}}36Cs`N>(Y%HU7YZUSBEi34R-Z7PMD+KO
zwy2<UJf)uJ3RTV2vJCDj7j4HB2Vu@U%J{+xIA?=#+kQ+60lt)pn7N_}uZg95{VDxP
zS95EJD;Jlcv*vo)^~g3>!DDKu9|n>+NL3hg?63{8Qs?~fQ?fE(jh-sGJa%XBq=>Z?
zCUJW0ySm-*Er>@@)qdmPvx^e{&Y!`5Xgl`CwLk$LwvUJ%y}Q;q=vi#tF7KnMyUr9p
zM*&xu_h#wC+u?NH<Jq>1Y$+!|mi<`n_OJOsU66SAV5|WgnzkyA+R7c$wKX*_elE@%
z^WABaZrAB<8$u;p6;!#U0hX_UPfA;TFv;f0Bc;*WE!(&a1+s<8h2Ia8J<R4>tc%&J
z)g#r5mCe_HVtQ_Vel1bMS`XYGK(R0(Cn!@n_tFu*?8|tOP}kRk-8%5V34gFqi$+6X
zZ*Tu%Do69&5mmKteK_rTk!QX%mEB1!^M)oSe>~osrF@&wa<&7>{mqYuHs?#nUbPWQ
zq1hoXgvG_eq0@j@9bukZO&8I`9=nc_L6bq<oaGUV+}HQtryB&?j@0vx$7PkF#Hz9O
zqrY=b-E3dF`Ze!|Fh;A4P!Khp0fi%2Q3@qr5hy2cYdFVy1uXB4Y>rbb=Og=g2%N3%
zmu~&%sX#mBL&Qb(VE5=Rcmw0Jiie%ZDei8dRba6=#JMuKEpW~KVo?DB<<XB#O#nUD
zQty<OeG|Z}*o{t^OE#kFKg{y0pT<q=AX9EMhQZrn=W$}1I0z+bY`)s)43}A0NP9!f
zPMThFXB&odRYW)UJ|xcY6IET;km^1T&-!A#;^&@tI4H>eq}X%=U1hcq{`S}JBAMSU
zGKl?K%N09N1XO|?!fVHnES7vfXD5+=)|3uQ$P<abb~B|-N=>E+d-FA&m$R@O@*N{a
z0S%#3yk7<Z?R1ptn4HxjRq*+>v-b}5ww%q`^P{uQR?WgTh-lHc`T6`uuCji4vs$Hj
zK&AQYBcIEN41w2%a82<qN87{&yzBAfKh*f+AqE0ZV;}Qt%{N?ssFMl9qO+UBDppig
z4K`hSzskW1jwbx|W6Kn(xIC$@S(NqwBRK9bl0J%KOY+AE6AR`W2`CYtadTCH{<*oM
zI+i%cbj;jp)?(P<T!J8s>R-Z1lZf$@zm-2)h$p@dS%*y^DH~mygGW2aMu*$clhY|=
zlW1Vc+;{Q~&lsEk?i5Q5T@{N3hr-fDT&c>f`xp0(1TzeC6g|smSWWTi(G)#{mJX%a
z)a+O0Z+vUMRD_0x-{NYQp9HXnbuxRg+C>s+6Y^Ebydv7CpW}i!*<c}66~ENW>+3}o
zl2+MaKbxr+$C7*sxf#)UgrN)ilYR%e>@CJ{qV7R=a`99*X+rXQh1=<|_ay_Qo!$g{
zJmRl>Lq3w*eJm+1W@csml9*S3N9#>>losABCG@@glt@XwY$j`O!5<agUn`CBS3<7#
z2uxj53KJUA-`Mu&LI3y<0~>AH-%0Y!`ybp-TStP4?t8Aas3MqO43P@tO3H7jS^vIt
zkuOHbGrsK?$M%Uo(FQP7J=}XgAJ+p_PCaPbZi)fH0TM<3%b5mW<=dmCYwee-?+X?j
zIDud)a!LWsug8G_QO=xYB7=fW5X8?1=>P0jJg>77)_P;fjPjig4S&Ik>fuxX+5u<<
zKp4&B{^i;1-EIGiYdftX6@~Lv5+{R8ZJEg`->WNlIKgc~hdX66C4OTio!VNuBMrw}
zgJY=X=y4oe+&LwVYvK)H69xPAirHe=e#!v1KJ$y?0$jiV4Dq0WIx(94;PVh}*Wq~R
zYY!S-YwM#paq3)P5!s9!efcAbswwPM_?^lY2v(zS{|H>tG~Z2?o<-VttoI091N@|#
zQOD54Bkq)ogdjPy%$L^8k9Go!hVn4>nkNEhv)#DKc^XeQPlpwrR<k#dT{iel4~Cgw
zuC_gt52i3dSHM`=X$ATxlGM}oDPLxh>@(=Y-Ag@-LShGO;G))h2&GUC6QMn$gTW9Y
zmSQrc6Jrt6RWRRS!}p>t9t#vP;%@jI3mf%*@^!++9n6|G8kv@EISTX|wGb&y7!(wK
z#I%H@QSYUcM#PY6iI$qN(-98H#YsZmW;nasx2ddi3GK<y-2AsHDe$!WQNGLI;Yp>r
zRo6cLd1>VaH9MBluCTB3fZVRq7Y7%8(t;XF9pRt~;S(Z&q{Zw8n{R)*oYwzvJ(|M?
zzM1247ppYhqdIIoq;~b-`o|ngdd4^TNu7{8e;K^WlJdsssOri2>bhMgcUc2dt!L#3
zm^I1Nn90481@TF-0LON9KZqy-iWVdED4rg@S#5B_sAYo-6bAebU61+Uyd#Z2Gm=#`
z^1m|p!eG>Gf8AMqHu3>>vy}L0Xu~A3B1$BKaa$CARk9J}#IsaR1_pVXslL8lslE<0
z?+{Of*Jnrh;eT(tHD>9k#4`CRNi>O%bx>yU*T4BEEB^9_Hb`w%weX)79V%`f`*jcY
ziy<i@rF(23?nP2IP@A%8I5_hR7m?fS%vJe5B3zyS5WH8@_B=%80Hh_#KPtfzt;X(+
z4MSW$b^2r%rj7W6Yv{T?39Uz5c|S^>sLR$t|LB6Po{=mBqbJuw;r!-@U10>Sv_u5h
zsk5vV!@u3;r4pt68%ksx5BB%kHBZ-t(mzZdJ^D`erV2qU5&#I4Kps;MXck`(ywa8@
z9^hDB2P(Vme{KFAXYm<g7DLC+Psf!&0lY2vd|LtIZxg+FKjBv60E|Dx5RiW$rqKEH
zc<&1QO|BsN#0C7+c|GzF-LxX~-zKM}0Z(JKN2yr~3pck~#g8Y?KL8k;n1hRJdOZ#^
zXkLpv(yc62|D<|-?^#e{JW_Cbo`t9Pg}zC7!R*$aUS@&?znte75b*$;y}nQ7cducq
z2N&&08^V7Eak=!3NY3S1BazedlL-#VhoXokdj+GFrrY0c0yiT~4r?#_+(aOR-VdiN
zEIr%*lrEjTp3n%mT1{R)X7@vOkc014^zX<Msxk6>8^<OxE+Psekc1RnD56a<4@GOE
zmyT@T5toSPpQgkjYFiUn`gRmJuy{7{A(tSLP#@tPGCK+z;ojos;-40jrO1IwNSlF8
z<@+2M)8-}2_3y)u;lwUMvKq#W#V>zf+8o0ueMg^(#5K>*H?9bZq;~Id&>6OA2fme|
zqoWM?1qn}6X=vwu-x?eo0vW2~=D+A6;y>J~#FL_8tT>abmqIg#4tr!z#s{#IQ&v_x
zVQTG8AJ~U8TKWT#X{CoE%NkrGHEbG#b-<b(H7@ViWYK$O-%J%t{qJC!{=~^Cubh>G
zr(R!zI;8AhJ_gLyb-l<VO?@r5I?pRbCDtq~97E+AN92gGmFtykKMjhCOZ43lS^LNh
zHDIssKyo@W(*I~W%cv;7_v<4iB_$0?m$cF#4bt7+HFTGVbazO1NVhadOLwPqmvsFP
z-{14R24>AH=04|&y+6CbJ_+bL@lYn#{~?6?2nb*WP}obyP8Cr*L*r-XiHNZnpfk(E
z%`|kF(y^%zZ})Q7QDo~@7|<Z(iI~t%aQ99pwU;@Y;v|?cs%T2)0eXdxQ@&vtgC?Ob
z35Ddoh=iZKm^uFVWTNXxI;HSixLB9=J{Y;ry)i~Xjcv6<e@@z^R^%+iUe(i#1sYgM
zoqobfxUc!BO<PaH`aK?Dk3E@ewsvu`>&;+J9dIrJ^_4rL{D&PWJr8)K;Ys30xIU=-
zpN{UR9NdL(34CF5U+87MfZ^3O+jcgb38o$ZeF&%oCVC^qI3qCht4pl|`zmZ2m5X!o
z%gSHf>x$3jA*Mx5%3nh!lVD1o)ui0p;M4(b>D70%rez3x(e}{o^_reAn?nzO+iV8l
zs<QwVzLedGVEpSZ0D=T7*a3Fb1Fr4&IzYhw|2#V{SDiL6?gA<S=$aZIuNMCcNCJi(
zP5?lFdhsxlHvzco$0QZu5DvYT(@x}v)NdEI>N+|mPEPS!HjM!^p`h-~nYK(I{Zpn|
zLjO}`(av38&E%jsyzsTlO2^u;a+}o=e!JzZClHQ-GB<-m5IvrN{Tt)W&YAuKjN9{#
zO5ikP@1g2)6kMl;vgsdIr7C7OfDU&9aOmh67|vcq^g1%dLz8+wO~ia&@iNoO>s<jQ
zpb@Na%-Hd|9n|Z^>jI^+j-_&I+&i0iqpxy*qrN?Cyr|+|M77UVZArkv=YoUVFk`Nm
znr>8^C@*E$NkKA-?2up(L54&*VcrP^O`4XRH8kX43P(#JwgxqInujff+3%y-DoaB7
zit)Z2MGqTQ0S|&dP$qznNnwbWNfOjC`szyvHlvlg7K|eLjA4qsxUg0=D<xrQgmX;u
z`QF$bez1XgDHMi({~SIG(h(k=$(xBwYGJ~`nO@XZWfpE0K|IkpGV1z3@qa%9-ihsp
zUV8cLTkO~Q$MkYWxW)H`&05&D4F2x=l;*bPcgW8SxKBY`PF5+*<E&N@f#LHlZQ+sh
zqRpedz3>rxh$-3ms!=rl*{zLTz@`3zr)qRtCp)@AmDDIO?c;Lz$E;u!J~M}+MDC0;
zQfha5j`_VPmz|>Ti`gDyt4vva_ilgH`f~Q54Cv3(-+iN|V_qOL62wD{)%=@ck5<PK
z6QLl6DZ(PSi(^<Th76Z2TH>3KPgzh{xRkCyf-zC|2g4^GT8VR~8fO5XEl@fW&L0K+
zV~6^8#c15bYC0k1FTIF@>>d7WI6Utb(%BuAYP5B=wj%G3w{K)>fDZEfO3Sv!>uZ1}
zy5Tvr0ckWkEJkccoj?_TdsE!9kt_PoI_RMYtItL<8pV^E*~;Hp7A~8@g_?B!qjARO
z0qhrFXnYsuVRW(`0qZq=FCK7f++ONEJL`IHsnpjm@JDbA4GbUwz2=K=&3LktnO-n`
zcH;I;eG`VVpdj1Lv>zi{fma~GDa^=0tBop}RWWJZ(^(0np`Pp|LA7a(hH3IGXAp6(
ztg@bPg338?hrmDG@fCRfHgDBsih$w!SEpKHzcql(&cX50g}s>N!9a6pbkuRPKc?D$
zk4IFTb`R{}tgM(>SiZ+|C_ddk^1TdNrq%SefTXP4C}ai{rr@_R06hSl1VG*50NS>4
z?MC1DSU@2G)UUa@IT|h3qiJR5=4G8NlW%E&3kXc~Vz~;{rF<E^-*n$-viDYk3iIE;
zf8`q+-`x{vOH(u+R_}I}A!Dy3VS2z(tT6~)>)$_CASeSX@6Ru4RX_y*!n__JN}WHL
zJ?Ot`YJE0raZN&cZht<7q%r;RXe~UUKks4k29ssC&6qDPCRlfDKxRg84cf^hpui(u
zy!X)SkiGtt$fQFC{L1Dfdy>BP{*sQ$232H0mk<IYfzqu=7R_m!+c-{Op*|5OgsDO#
z`7!#3Z#4=zbVwCnXg8P2SE|G)=@8!{SVS_X6y7{19p><NiC)XgF>G!0heJe{rg1b9
zG!t0qs5D65X_4X!F@x4*pzzQaLT<~jXiUZiobzdFvofVb(9yJ#@c)YY9g$8jP?7hj
z8o2Ll&AR>jbp45k|M#4mh^rZV-CKv_`;xpX8JlKD;(_|A-+ti{^}Du(tozPnwBoVo
zENr3diRmp}JcXln+npA&>5UzYpVa1-=EQJ)L(cGR+K*7Fs0$EsqfKjMRXhGVq$Wts
zn!~;oSE4=<H;t1>5ft<UfuwPP4OHndBDC;+Sov=ry@`!F%x9_DYketDi`dA;#0aOf
zj703il^0lwdVLe{%ErV<5o2ZyUw`#TJs?U?cBx`TfQI=%tS@sMS%WDuV7MtC9gcUE
zq*9*q9zn#QN|hV;3Vlio-v<HFDUfP?Vf*7pRE<SlgToGbj%9m7(V3lxpUB$nO|X(m
z`UvUxv#mHJ%mfz0_tM4ptAB5Pe=O>dqV_8KMHJeDt?9}ltX6}f-7TLxpOr}ABooBs
zb&KEI+t+X~(p)7#0*w^M!)%*=%!opUlz-E&M7-V|#6QIGbfd+XvP=Ci0`|ljJelNb
z(}}N4=EqY}i}8)Tm6MbA3v^8~*Wh`S7&cVrvC-&12mc#Y`#MzYS+VbX3B@E48~eKi
z>FGMM7<H82W!mK6QddAD_8i)4b@B2gx%h474J_!dUQ}KhF2G+!I`|RbW3MJ;c-h&-
z3MXj=<4%UwhY106c>}a=X<Uwhz~!CTR0u?1KvB+KP*8y2n{`;&1fB;Fk|M>CWBc%4
z6a$z+fI9o!tPK_tp1pW<3|;$PVD`ZIW%E6}va+!$nDSaiBe4pQj+d5}$}c)<0y6ST
zM!?3g4&2A17~fhxp91+AHHt9k2!EufZ`Ci_kjp5&xLG?pI~j4tpPzMi_C_z}kZm3x
zRMRIjs>*%s(Plu9xkT*P!R3Dur=T`)U3OM;U*c7-NMQxQA9CVbg#Y${Ap7^571aIX
z)*ObxJ#C5EC^HM!Z}pN8bc{|hSayKIMeIsrj-HCo4-|lm8NiAQX_3@r+g)meJAcm;
zQqSpUP?w}bplGc0&rTI6OD`WZpGBodfK}DBBKZUBFDSs;a^UpVgk@k^f=kKYQNvKj
zeXh>Ip_Mv^nWA(WT|AW8g`q`S61@)rGA>*%iOXe}#S%j%BLsdN0_R>nUOi$;otK2%
z*^NGfsodZf&Kf85QC)LK-vL2|#<C7)wAC@zF9<E#d1g+1IAn|09&228Kl&Y8%hkEo
z(nU7fM;;!TT#b$W1S$ITw$P3-E4EvZ$c*#6cikq!P4fefM~r^^2V_~Mgk5+hi9Qp@
zN(9r|ll6;YsFx3Wir#I491B}xX3dHARzgL4cbnLN{v!h{vb3}uMmL6-rN<+md-mQn
zPf;gXtz%iQptw#TD*<W5S#*qQ1dWwJ9_+m64-|&qfr#A05>@C(mhQ4sN0<dVVpnoG
z6tEv%7}Y;RE%(WP850P`gCLhe{iIaI-sR25K)%C55DI@KElxCz=!ci>K!X&$f#UIX
zVrl|Ws2*nc%loDv&d3G_o4ByTB)mgoe<681DO3c+kI}FT(Re*R*Mx!(?bB6ezcF;F
zd5n*(1W2?`-7@f-^4tPPm)#oM2o!SE)tiP+o>so;*K&1^3->`uQBW?>m=(|Z2Kz+&
zB=2c>@5N<q<|pYEWTFN}UB|(>X(YZ(^MINWpyz0f#Cge8P4;;!f3o|#IvTsa`d%D4
z0I4XB<ZVCm%u+qVc@@9auy3UH=Iu1FZ3iAp{d%vgdU3TT)*a@0*BRkZ_Bvn%uW~k@
zHXnGU<3ZJ^qrh0vdPNX~MocB61is<}MbQB30#JG{UV;V?*$bMg{Bd(TaMJ+=G-`ue
zFbLJ=-@f)fxt7$?#tR%88j6?Rtb@z~sDvP-wPfm!4nT#aWn}F4r)b*%?MA+07Dz#a
zA;SPn#}x$hAvrXXxFxTZbo*v3>%eS-mayf-#06Z>_bYcBJI$Imr-t8bzzvNZeQ&fG
zmHRsV>3&A@FXxI$Cqh>x>+uc18$aEHfhJ%{fK6><b)6=5-pMrVQ;*w;!nCyN8@ZVU
z9qZr4nhP|cFII?`HwuOzL|hKA4H#=j6)K}ma&mGP{~w59@-US7YvuSWG>@P;>7r$g
z*D(>f(cAu_K$~!kKL{}tuo}Jzm<-yUq%4O9w1J6|-w%#HbfIHNBEUpL2ary}8DKVx
zM;K5b&2Xu~LFcuH;V1G2mM_rQQL5)-eQ*p*R7KLkD(~`lTHs@$uR;>INfhmD?Inps
zj3Y>0W=sV(p~%|6WLIcf=sOL;mGRAx!Im&+K{8<xSg~%D^WX?sFr-JQ(-fddpR8Fa
z{>1X3zeYw##qNCi*KNMsXITmooRjmb5@eJ+Or3CpI~HWKrlspOZ!rkNL~Y;6W8W&M
z+QZ*c%QAoZER%s)uJ^?Ly)DGA=H9&bLO-P_hvc0ACQp(r7WJHVtQn~v{m$<%wkE6J
zS+=SAIB);5Z&VqwE*Mx`&^zX`7`@FEQHT(N-1)UNLu6lLb$zTuA&K0QC&!&MpIN4&
zhH(;QA`85-$;BGrrdG{!;MPi=+6bnIMDXv{oSm%e|8!Wn(JC5^8|vF>aU0)h8Nimp
ziAqEECu^+B&RCEnh$^{5bPEf<x--j1H^t$)>-r<x=bp;|NE)B(_e+<>cRVCmoNo%_
z=Sc8O@t&a4Zo^oJJUL*0>8~plk4sJ-G}3pIE=TlCM5BTrSi8v8+vVpD+efu;spB(0
zkbhvwrc#%Qh+uq2*JmYm-zQ&kI(+Whb2R&N0ljYQ%|CARGH-YZuOK9(t5aR6)6|)I
zoa5oouqde~gc*3#vhNS)@~Ee?`FP^Kwp&qCyjVSeq?(N1&UL%)IeTY%3f#GrQl$O|
zM1cA#Wv66qwDqkG1YYd!np+f{_J;6|N>!fq655`gMSW{o>VGltoSR9~zEDj<-cp;h
z5&($H(*|%my;OjIyAeT+{pTVLSbg&1fd!!Quzc+e-1{{5j{Wv$kQfArwtS$Ydl^ou
zee?mV0buHj-*V1Zw(vfK8m(c~l~b#j9Sp`rYkOSsjF{GaW8A#Vm^nEb@$7#2!-J#x
zdC89t?QijXvaY9^3EbmGzJ=z<Ja~Qq%Fc>*;Ce^<5`L43)rH@qWtKqgv(5BH!QFjZ
z%QvuU_ug^!4qCk{Tm4YM(6HWArYl8!+y8~r<q*$oG~=<#AE#YT|MmA*8h8_1K(U)9
zNX2IP8jVbo++Xk=zW)yrp*<)D%J0tjriQzrqUxOn&$1X_<^&q%-U^67>zF$zEse#j
z;c=<Z99DV(L);Q`IA4`!uCiE$Y0<i9ZlFQ0%N*Hn9{oW#?7doVJ`%1^i8PTk;U>Au
zJ?h%!ge0|;a&I1|a!MzUEiBMUX^dFN3?b2u)^)?<ez|!Tl0K9;EbcMmmMH2AT%$v~
zW%RmD9hWDg(Sba?S97m5^p^j2&dc~)DExcNn5ki8wB^pI<CYZlxSxwKzmfeq(|Gm4
zKoy6_>lS(S?;Up6`0bZHXYArx3Z#96bi?sz{~|I`GniU-1fzyWzh90Z`dQz`ke^zQ
zZ}{?}kdwK1x1v=_KXI=8o3EH7EWY^pZ`4QcKsExEhRPiQA#|i1^$GJSe0&Q4NDVQX
zot)13lVBKa;%^_0?MpvQBHPiBXahNbgtHTL8;H{^x0rMdcU<Pdky7-Uhd2Bl!^>r9
zB8e0_mnZS3Ejdy|fX0s%`4b%wKs0DtIQ%;oj4&g4=H{04?d?NKqz19Xn%hMiW{d>I
zzf;yl8;IW`*B{FfP|BYQd67nU9sKIY{z?y?#|{#B+9GOS&S3Y8BaYj@xS8t}57VRM
zF|aXtlY1Xo*uRf2EFU8NUMCN72MJD`U$0OYE9abjZzT>TrhA+L@dFSy-uAHo{qGjl
zb62hNbo=F6|BkxKyZrZSH+`>MSYW0xiI()8TWC8W%~!n%%5|<!Cnba}Gv&yBGfk{i
zQ<ilOlh@e|d~t&?!g}2Qtbd|{IT~nZzNabzfS6eGa)EAz<I%#4UnYnp@C)(R_!m%+
zZcA{KMx)yq8F0y`@qf$EyUW*uUIR%M#@^l;K%vq0u*KvELfs?Zf0nG6B_FKTVPRe=
zJ4i;)#{v}^ZWQl1!lzH4{>!Zfq$FSm0%MYK%}+odOknu3o|da8xH^o|$Q_^SnYlI6
zf=gUM*hp5R)p7MTt!bolWMrCcSdKU@IoRl~gMyldmJktC&RQg1Efyx2bC!NQIBjnS
z=K=sIHEe+Y2@ruS)IsPO856@R+VLAOP`z`f9&p!=zW_szj&m?fK4Bw408G3B=ZkGw
z@1_&?2BcZ5Rc`^?G1%tmls<wyF?Z0Oy*#<@sE_Wx&DXP^{zyVYep*Dg@}=T32~Z@z
zVHk#aU0EeU7#2gy9X|F^U=GTlTb^k{yQsK7a2v7E29tB9e8y^QWdX!QWnhM39Y^*!
zhvAuiqTSH+H-WQ^jw!(3(-%lY?XM=%X$Tl4_(b1|ik4_c@)1G!Z?_5Q9~ug2Oa!A^
z1ZG$o%DH)b^vKuhu-t?F9|Nem$(WPHwESf0seFXgk@UZM1__x^WQzt9h`r82_E$G+
z!5%+KFQkfeIB%SFJ5|e|h$MgiZC`dF{iLcsr(+87utBRo2i`#5PXC#6axypVvwJAh
zW1;QE<F5*gKNb@iV!J=c<U%YlOFMQ6{Jz7IRux3$j~@L3NPWLwvD^4Nvbx(5W~+IL
z1;K{juP)WNn01)BaD3x88x75xrMUZ<3p&(iCCHKB>NLemQ~<;)-ZWQ`NR$z6tYJ4<
z2pi=*Erv!c#D{k{I2Se>SgwqhQR8xRVLL^aur$`CIW)Gp14z)nAK*iR;IY%+@TB93
zqN{hxqr-CiMa#E~VVgk$Ik@@8Ws_7^(~Vn^i~SR?v;fbqrA2^=4wTb~!NvkI;;$dd
zD7%<DV$s-Kzp;4VG-M0g>p_eJ`bnW+RLnhEu}HZayrg8X{C@~3$Z2}r-`YB!_8Vk)
zolDf(%Px|QhIHXU(p}gi4KqO8a|U=yHHhIWo>;GPa%dV{jx^m1kA!VYke@ui^XSd|
z%SD1Gj1b$aEom|-IaVHpzpeiHTH7M!c;9*tpPau~lm4Xp!Yb-{39W+RLRvm!C!ZY0
z#r0Q_Wn0V0upc2BUa9l)TTm9xXtv!ipPYxtdHHfizO0~utsJN_5>7Y<t1+R5M_<s;
zJ$sXd?%T;`obKnF<&21Ja%HcDpPcV@TEW1P3($i=MGF!gKsH!UHER<P7XmUdn14_<
z|7vY*mEzc1{w)!E_~Ezea-S}7l8J@})o67%n8%Dnt#&+~T{!NiS4<N^qe+$!^nT#k
z872JY#aoFb1n6)wx#MeKLCy4O7R*o-o3b7+lwUP*7(T!i-N${w*1z3=<@i@ANqhI`
zjd<n7HS?>?^J?CV%ge2y_l;DgdqxuEHD}it7%UFIL<GO6O8Rau2*74c0zX>+C*t>h
z<oekebxip2Qj94w0tMd!0Ws2<Ks*fna1tESh0PJx^<0KPXlnyfCWn+V90jbZ)@gVn
z@%zscG!!U8!DLSBcP-l*Vi{Y;epu6kfPU(+%=oCA>BQWD;+ba1FoVZ@d<)CyIgEs`
z5nPl=l1C9tW?4pM?}(LbDo&c5Mq~!5k!k18!l$MwLlh3Nf-M@$iH9?hUoeafR>s1j
znedLXs_eR%!;78Udu=U?wH?k5E%r%}q(H;c^=5@*bv;Io0#D<#|I6`a!imn>!@P=+
z!Jm6BLC%3+ZBIJKjn}dn&oe+Ef!p5In<BK2<6g`Hhfwm95}x*6L@>Q{O!g-hY(LFf
z7jawa7vknHn4oP=*E@pIuUlG`6YB!T8cqM$-&FlxfBeL%l}N&x_vq{09yGy-r;c#&
zFSah&w5GRTJkGR6RI!3IA4VT~$8GpLZAHIBUD<I>SF4mAFD~~2OZj)V_}lnS{eDN$
z3^Odrgyr}CWob(@35PH!1|*2W5=0U?9a*rEmv6MA3=M>_e9Hu*!sx;d+iy)%avACu
ztQqwh`g`}O#CUv&zmQfT1p8BrE^`$z7~{mFF+Ff;-4{!DdOi+4pKvh$I!&HnJwQvI
z?cPq#>GPop4~dNyZ~-Qv+)NkzxmS7*=bx6m2JN@6hvP&L-4ow_K@=l2-es==%G@L)
zj(0}ZUyaPGZ~hR)f0)dU{~)DRr&;EdrYe=I%e0lu%t6ero}(g}aIB7Y2=Dis@r?=_
z^X#Y3p!3nK0I*aBox=B1?v*Ojf8FKRHduJoK>=uDUXJby5EWpSI5@lJ>^uFy-5%oZ
z0|Ka5J04El<w0SfT}Orhc2X~4MK6)=l#bVT2NjN>D$9%Cy70b8UEKuo$N$iqAOgu@
z5398P`aElP=Tt?f6Ety5FE`tZU;Sn0{gT17Z1bYya@xa$0wiJ+>frxD^>)y;L+B+i
zfqD^?w6&UL=Z`8;dKo(?@O~f7>?-``@Vff@#ktgUYj&kdRE)x&A1!fF(0Z?-oA(>1
z_ME_(LjWR8OpJ~F!GsZj)}`K&%4ZbnZfreY?PwTXn`v!$pK@Pue0kdL%;XUYt|mU$
zXI_)iyTg0|GK2S#e>;%L@zel5NbAd(67Udpb`~=GfbT{YFAOD+QDN5C=cyNmZih+H
zHUgcJTqJ}99jJb<T$0jDAR-d!s&uA#>p2uD5>P@p4gm_GVkHu`&BAzMPV9@}{)isY
z<28#S1|k@+n;NniJn~6bubp__!Sj5bIPR$8>-E9+<sykL4=Jm}`E*FFU)r<Atw2*I
zwE?ME{nGpHuVw^I$UMhq!5W<Es`|)Cp;*lbP85m11Re-JL$ZQ`k&<NR<!}y~A)>}#
z{qgM|%_68!=~eSsEm>-EL!}0j<~Ue>&(jUIJe8fCuT%2_MY__z^H)@3^bF-$9;B0w
zL5&VoqhYJso6~A-azwJx*0``aA@QWy=;lX~`SzHfXzoN1`}4k(y+8fF7-jc&Wih&O
zeF=i=wzX)Rqgv0!H)Fgt<r*>#-xnogXZ{Sk@S&;o4g9jNZiSyBm!N(~XEY~L(JX{D
zMFt#A5{lH}hW`E9vhM;78+=Rxgv3|^KB~jij|D?XCkn+VAj5RFkPv??x9l$^Fp@Vn
zbnp6VQR?@kNh?5_PZ~~w@LTg2Nv{Gc>oCGcYtCjnY@MhdpFN@lW2)<_elJ`)kHH(j
z)ybnvd<PqUaV+=%(J&;t8!qkZVish(jmWy<*>e<?KdNxA*6!MyBre2g%v^xP*#8i{
z=a@kH?~goHVY;4aNRs<D^5`iC*gYwRS(bZ($R%YI7%&{<J?(&bDbHjTJLHbXl$|Xk
zX#v*-jMZ1B=axt3L1-Ee4-Zgiz#nQtq#`2BB*Aw%T{egrMT#MDw$z<uFTLcwfD?DR
zjt}^FP^S!o4}q1xYSm0GJX+ra=nOqR2nmuenI3qb1B8<@mdtqZhpDfLGZIP&*m;X!
zUbg_vg$9TY0?{LYU6t|e$^E_ejRDY<XjdzN0Xv@2kJQwSzdd2R&kq2I>$ytDqwco0
zJt|Sbf03GN+wO9I^a~8ez9N*dowQtlDd;mV5Qe;L0q*|oRg?FC&9;>M)XDYUrH2Xa
z5}_bi3;=UbxW!qd9{=n#N|&YFYVqW}EB4vybvUFBcK`dOu2_GPBhdPPuYZOrRC`fL
z+7mhMj*<S4g1NP|b?&U1KA>_j;+YKb&gjiFi73&P^H$c`_4}=lyLh~KK%1|wuD*t>
zcLUvio7%1+cI7r1Z#LsvVgId-EbIF9Azm+$c%Kujpbrh}5oEA*Ml8m_&)gd&8_AcM
zgo)L*^$At;cXqBL5f{5KCfq-zu^CJejA7NV%LF(lg|U@63RIzU&Cp{rlnIn~Vko5s
zW5m&@@J`L2@0QDRClVIrbC~nqH><MPW4%S7E=9|O1t~*b!k$Jb8W{<${hAi>!{{2{
zWnZH}OK4ZH%vfT+_AM)u84LA@jiw-huw<83O^)vEd{$I4@&q2#p{OaubCh)4{vv@u
z9DGS=LIq<0{x!1Eu5P$x4@z|%Z)^BdL#TE$_9gRg<i;0HuTwk_X3Np&V3E4j!@lad
zM<&7SyDyg`9T99_3ymLu0>3y#fT18yF6HQ}{)ugMb6e}${(6~M-@`D?t5E;oj(zpv
ze72~N>q~0v^u;PE4Wq+-SsYRO&p+bAsJYY*0O?Pw92)IXWb~FoLOORmYnmuq!4@-!
zO>II<KSFdbt+Zd8BBE}NS;Ru?V}X!*XssMQT+Z0$o=>g(zkAxpC#YODO*)4J(k=wL
z+CdR4;WNvaY-L$Igj#<Ag4d?GJ~T)nxv+Gwgz2bivw;11i7L%t2r-H)!DX6UJ^q^)
z^lDfD69wj{V@ZEi9%ubXC-Vxxov&XI3aF%Rwd*mksd8w};#kdnSZJy=NB3W^bTX#j
zBPnbhccDC`5y{mF<Dzpr>)C0%kK}PbKP!}6R?3|hnaehn5P$!5Q{id3{od7ydpYw*
zug%G6>J8W?{keMy(*c7e8ZhGEke9RvSlE|N3*aX;hRh1fN=xHWvZ8iR@g1k1O2K_-
z^GoX2;xcyCUqQ7xQm;B`HIv9<->+bA;2FQ;o;<$&u3bfS!nOZ&93_M6;L?MVZe1|)
z+dFTg3zGrR04du3YG~+xkv0Lrg?%J3|46oQ(9ObZ+Y7Z@12{LqPJuR=bF}yNV$iko
z^i@CLECKS<*+yUE{c&@0T3Q}JqyRj^x2Mr({Cq61GuLjgdzFs`qJ`<eHI@%{+=S=*
z`<5CLrQF{>=gx!llHBli#s@D|sb4B<6#4XenX@16&<_9-7to}BhWyLb^eto0sOX))
zF*>?*2B!{aIA81uc!5BT^%Cs<k_13dvSsw_@wxxm9nSgGOmCz3P$yyiawmu>;QscS
zgA|IX?S||HG$Y4*Ceqk@Ys3CzPyBVN!ahASqnEKG02YF{fP~^JARekavd{zPgf|fy
ziSd)j+CfY!H+)K~l-{=5;j5p=Cu38W&?YRJBdFP|UxH?yL9(8yf`xqyU!ul!AV%lm
zE&XE(BHZG8F5Jczk}0MLkaYC+@thU&#xRypgsoFqw7wg0SaF(v%w0AJILgOX$E~tS
ztnc*W1<klLW`xCF{T!gJ=2KuW^|15>`xWm6YIKa4(imK44~u?PenSSr$|@uL1m`O@
z&F%sDJX!;E^ls8#O%RI}j3CiFM`ONdR`Q56Iq7O=YSH1^{(yAP&#7<h9e*^f^1k_*
z51Q0{t^2cfYjyaCrp=#Z(HJwEB=nc2f;E(5jepm&VK!Yl^q%;=(pXPEDTZk@R*xie
z$k{(WkPWx2&4dK!2b^@S$s9L0$9G*gTiT0o%d6AwIeR#H@~E@g6Jd(aPL%1|j+xgM
zO%wPV_#QxIkK;ib(IAB>!b#G7%#Rk_&5`u)FozS$nhb{$pi|DMX<c4l7$DG4Wp=o!
zY_1%NM*7volS(53+3-Lj6ebGVl_v=xjgNIxFF^A{H-Za;90Z%D&$L-AxMxmZvL%*R
zU|Px}SPDv81!)8u2ozxd^j8+v<!1b>ba7D!rHY?f>{9CCdVpI;7?w4u%Dv$~2lv6>
z?lwU^>FX|D5uYF#VlFx@^n_Cf@xyOoyTPhOPPdz}hzPkAb7VIl^mPEhLx2a{IHvT5
zCRD;^RymoPWZeRCtNiHY_LZ0z3?Lr{l6(+L1?oMpzFNlCUp%><2Ygdc&s4A&22TSp
zqtO3Ov}K#wbCUf0PM~I<emOZpEZ5k!c)jz<O?kO0gW)`C@6!0am0M+s3BE)uGwAR@
zRQDf&n?A=Q*k%(v8@xa;x`d8TcG2Zfw$|DDu3R{20PwFb1DB`E3P*73fvHN@-5obb
ztRN9WkOkC{&Q2jj?$22lhx#@1c{E7im<Ux2eE|$@y32w-4h%-Tp6@|?!`7fW<R!x9
zf3d&z4>$IJr=)ka2{#bSUOBjRc!Bhlc)oRO!ctFXH9d=a_KEG@D0YVnd@I2Pea_3Y
z_~UnkL+0!oRl%fLqC3TJY;+jfZZ565W*#0XfNQp}poxzG!!gILw@^kP{nUb={XEf|
zX&uoVWAXY>W5b5o<EiDeH>J37`}*7S^&nx&hlXpCr}E}wf^z#pZ;y_VTeLK~DSaKW
zwC~ZuP9=R4w}atP;#YE#%6Ua#>+sp($I&pOM#|)YDMdv1WkkG<s{j`Kre9jROR6|t
zBr&(Z_@MYcn{@WvyyaWe!AcEo5(!J`QMn1LIw02T2PQUZS-(&8vv?Z|>LwbJEKA1W
zPUAEm5fHvpME>O12Og-V<F}9TzpqeR1v8g8{el=jqNmJ=hyD4(Qop!p_3M6&32zfL
zN}a0jYbqznr!nqRoKqCurMfapVo3UZ<WiTwoT+TgRzSx85&vgAJn-0VOEQRjQm(S#
zEn}-gN^L<fBvA`(x%=Tsf{`g>9fa^18`Vj9LIe`=WAjFknfL|^`7Z|5Qzwfes_mT9
zE51H*>E4Z8b2-^%?5udajX~>QD3PXkb&<<LSeQw@>=DOf6Y*oUFeENK2Pzka`;kwI
zjt=&fNEJ!EEYoz+tOdK~)PYQFLm(ETu<WgieQ!=3%vmJ%7*JascZ8KkK(GtKeCgj+
zR<{iGl!RHG{7X(tCPP2IRyJ85*95u(RZDf>QK>Ew5}hy6Bu*IA(7XR$TfMm(ddDTC
zhs1sKu3WoyfMi&le##^q>y>*VS-xl>MnIi0&PRduIdiTdjt6%_Mmu#96M1hNx_SE%
z!Q%n@>Y+qsggm-9q(Ija9Edg}iH~ARPUXL*uOc)d2P_hQCH2XowOgFN+j<`mJ|wiC
ztHmg01*WE>2%#{4W*fHtwcNWmX*R|$s3rv@^@Hky=pgv10hDQ=!2(M%Qhyr95}*on
z*;@y_ACQYsF4nfTMxf_dsh%tsfPxejzAgI!l!b%GZ!)Upk=ebL0&valhX;6g%IZuy
z%3nLRm1;3C-CYBXiD2`xsi$Xi_!6*{gRC2X$QM<JU-*QERUD}S{C2bhW{o<{E)hWc
zRXfuHCNU<BX8>Rdj0I5v0dViX{!;)W4~t)mlyhF|0k}6qW_*4#DIi}D<o>LJRB=!V
z;BvX0l7PBZb{@4<w_?(=PO@wsoF4$4N|!8Piv{Xp#+^RM0<U@b_O!};dU51PlIkO`
zr_au}zE8G;gM%-W#TV%_a*wXvR#fWixeSTRC>@4HXv#28P_lMZxq|cv2C09bJ?_w9
zKvcsgySc#LdrtaTJS%?LTZFXoaP76%oT9M%yIv2CckW_q+UA#5b$i=tHM%d@2A&Pu
zddIcjsXbfL4i|m&o}BcK;CalrYv(@zIW~;U{x~9}I%4Sv@}04euzH-vX_Jdi%ngUc
zZDoJd^|f*Qro|fYe@utqCN{b``wf3Lfj|VL1wLafc~}9u>}L@=L<A?1!ewC9h~Y#s
zpV^<v`EhqhA5n!<XBwBaO~#h@yY8wQhc+lXUCbc7H;2|Ka;)Cz5dTQi=N6et7W!{P
zhy_J>L{vPk+^<z~I){E%m}$zNJ<<`fl~mlNRy$UgZ&JmonQ7+66Fq*!bnC$92$Gk-
zB@kt`lA}ZzXY_be=8)&Kz{d5S8S)8dwoRhOC3u58$l`#z<iOXg{RX5E-IEqXMv?}H
z>|+qOJ>msMPuH<sG0EBAe-3ir+Tv*R{u(yj^PcQY!_^41A989@nT)9xGmoy3^%=zu
zV|%A`%h2*KX_j`|cp6M~>pv=}2oo(|S{2AA-~t@zr7&NIN3omFSh89=|78}#RD^0w
z$bv@ttgp|r`H!ZSjDl>+w*yQ66A~$cwGZCrk2Vn5APQnIL|h#^q;r2uJI0eBAV$-u
zWJQp25SXy$*rBTC8#$SVu>1?B923Yxzl@ge6)#iYbJ>juAODdodXl%B&kiZEh8G@Q
z5l62oWBd6=BAV4&B>VY!^Kq|+>W}jZ;?vUSde>IZ<K%=o<7_D4y$Xdx?-fPn2+1p&
z5ucolwb;(3Ae^7`N`KB`W2Z`GGSXa-{$%tiwxGx6?H6*Syc8)V5&<Z(fCv$6`63~P
z#?{N^)w!jwd~RtC_O_Hh141~{rcO1rsqPmmZsfdVk^E0-JG_PPZ!;DNe3h!ue%@2}
z&iz?`f62%;C)h|e$CtC%6~xx`bkXb2PkQ;jRI+v^-d%!#13neXzZ`Q$qt>!`^-PAK
zss{b9T^mMANpO>@vQ$_kokuI?u<yDZdbmw){OWDzw>2gHGbFiHH%4`QjjTWl$^O{z
z%)39Co2!p^6&o#&N)kOTt~;0=0k7WT>;o+|g+wQK&<%A++FW0@1TUu35(}jdKL}P^
z+~amKzqL&-N*5|cMn-CWO;RsLfT1z1+reFDjE2(GpEf~Db7X*!G=b>t(d})j!L91)
z6p1@W_SBdHS|A05lBSBAfxW)DX^^Qx>Y`YarT@Ii+mw(RJK{WIZ)Z1_A*?e?kAuv2
zKR<dtA;ko=a0xDvy6#)pP^hS=4URjg#LqW<x93UiLw>K(+)meCIqO`!%&bmUT4&@=
zfSKVXblVn;uE1ecru7R9qZE;?D7I)VbR;sR2CR7B|MGU;@jM;im6FtUa$mkD84cYf
zKUr$*BD8J!?Zgct%30n=Ee5>&tfJFlpvAf89$KLFTAJ!@EQLk0GZ8Vdqn(|dm@Vyf
zjc3IO@r6Xx#ZcAIa4z3LJ66_VK|^!6*RQN>tvdRw<`S(#&8hxXQD2hM7ALxx;n=x<
z4}H%!#JAuS>>5tx7`4L5TO5#koRT{`Xn#78Q--;9bD`!|X52SraS%9yaoqVn_j})N
zU|jsup{<cl)H?G<0B(kXCnPmjW8{gqFpcr2qs%5*x!%g!tyU6?X^&-HH*U)CFhjjd
z2Dt=E*Ar9((vvxXmvNC@d_9@NZ?)0ev9~jxXa#!|K7twnB@i!XkBzkJsd^+Gnv_(O
zY(<v_wgL`<F}*T^_4Afxnf%u9mC8Qd`r+0S=|+xJJ7ul3P;pv;q$zd!&}k0|+s|YY
z9zF$BX-c~~$T1UHka`Z3R_TkiHEYz9RkM~``mQrJS=HBG*HLX9zuF3Dw1;Po+4Uw@
zNsg{}(cra57%%ouz9l_!L^fQ^yc&#)Rmh%W#cnR0Z^UF-el}WhzdukuXy$d~THgbM
zu9XE4ta6dKFiZ1ha^62jn=N9_&t||}nCXaxD15IFW`a|c<1oC2)W;L6Sh0biRD~uf
zx(H9$xA<MX%T!Eu>ATx?U)T0aoa?rG9O^jiBzAc1cBhDSb3bm0Gq=#3oMVkN<P@GG
zHWM4#&`^hTKL+7^dZf2>%<m%#N+nTek*BUyRNX?iQv4RLf1bxK2x1k+&%{Fd^Z}5v
zF{da@U%CvN^yd;(8TC8Y5$le4o9d9ZRFwxl;yY54K$VRLqoJP+zVw!BAx_eHC<<6G
zIhn{FGsPc%LZmRiOv$>mEGLS#J-!X4$4o2HsF2*Qp<D~jIZcjCdHwqO_Hy*e^9k+w
z{@Dm24VFo&f;A+>ZLcTl6TurUu4KM>iJSMrKJU`t6m;rMJ|@>%ut0=&{BU*MzJsW!
zf=P?P5k$|IF41PWDPX3ltRKkHUQ(_(Bk@t-kAxV*!sxFY-tWu@lvk*#k*w-)Zh9EU
zL~Q(shyZ>l!PNISnVhfqCaCs!R&Qms*cRVapCX*b^&CAc{psW+hyaLFKR=kO>uda|
zKA^q@(rdlvyV=`bjy6G^*_ZIS*WXTGTkpH`zP|_pW~>)LaSP~@r3)vo8o)$pqxa2w
zKCk}S_9y3=12&NUunAfWYp`l{?Zta}>#zft=4GRKjd0?w&K5iEofay2P@j@Pg!S7O
z82gI%y|t{YEQkRA`|@XmpI&`?^nON(qv}wYo4v@ArJcU2c^4sO9s0P^>Ung6FqY^7
zF>Ing*$jf)ZyN2oh0eG}mgl3#Q7G5>o61~w?hbw2bgeWZ_(Ls5+X;X4;$rTrp6DG^
zA^g@zqbkfDNbun||7|eacmAR%%pKnixr+SC+x%7;orObaJKECGZHL7i)4D&2?&@Ug
z(B{icE9nXiJ4YYNozWH2;E0vr`SE9)9f+4T{xx=UDoiy6=?r>IRxnAmo0ih6DYgJd
z!EbNBxRnt=!lT8IRpQr6R7rRDM6%P7f7WrkEo5V@1X-#d@UEy9^>)wQ=P+;(+l22%
z!CNIP%5dj&Y~Wd;HnO*Mr(U=BoUb9|2f1MtjS(Pf8A=POc<wjl@#8@UHWiJKybhpd
zi^SSZ4L%g=6&rQjxmhGyy2p83COp*n8mUs(V|B!J8wa+KT->1{fAQ3`R#-3yh65$s
zT|C2%hQmFa`~N&~xNl5X<Q~3b527pQO|skTgl{$W3CSu)wp3GU4lORH4NE&pw+<zv
z9>p~$jx+h%n`?bHkDmL+P1M9%-ENt{i+qX^V)dp7Zwsp`<Tpo|!#+X<%8EzZeJ{_$
z8Sx`xh+d7Wg<sTC%b&}#u$!UiDE6^|%8{LLh?GW>vdON1;w+9e#7o8hgVj!lMzwkz
zRSHc=avAcrTXn1T)kWIIeE6duxfgb+Oba{xep(TK*6gS{HrF)w4LXi<CAT2l@Q=TL
z;Q7n2tEb%M=<u_uR1mpEzP*3`S|p4Cr9<?^OjB6G4vAn|{=sg6E!##>lvD>|Es>H>
z^KE~YD$GNz*#YS;i$*e0O1$RB^anHMT?Lg$PT7|Wsax;nPkTSL-W`RgMszR!ibnhl
zFxI8hsg*M)%WjR|DG};)Pzaau9<R0old4Ne;UahxL4Lb#MY3!mh->Cy*aey$z>=tI
zDlcD6^9qE5lel#{LK{)*Lb@Qa(3IfF&?BP!OhY2&^MhHX^Mm*Rwm$FsPXrMSt|#wZ
zkC)u#60&&7b9mz7A+s<QgWlqrV|ZOw>>Ing!JgO0FYyOJzp6#`uX**+M$3o9LPsMw
zm*y`>c>{B#`X0;|dJN5AEa+ZemX%=7q7@cv(^H(s?dsw}6Pg3(qR0!dPg<(d-cMe4
zb&Fg=J_1QPRWs)=5~PF*FE}4$T*lMGno)=KPB;+l#7a>2_jcTTY9BF<!XgvnRmk~#
z&UKvRdQ*8LxROiUd3kxQzL)s$@BJXPc)XadB!+K@yViv=p1rQTA9}E;afaTc=>5Y(
zIm$F9NxzxASI|{fXMO4UYdFK}GvS6T&oBC<K^GWO623^I7x~4-B>&^&OBNMF9QDu|
z5FQDIL#(1Ry4@cBBIer-ygoqRBCHq2un^8;Md~r+HuLzHUCx-3i=d^ZtRiFmZehpR
zTl6nZca`8Gevik%j5$m$uSC4V$7BOKT-b#XjttY^#*->+GxuY^6d}f2|AyU*#Qj)y
z`=iBD@+7mkt(%Ux-5q!Qbo^Sn532KuAmKAb@WZxNwpBl!pp1U@_Of{LkWVc3(Wi^j
z3LD3c$z%`w#xHf=bpNfdK6#KoRbEgu5J<`5rWWMuBd(Oh^{76%_y-S-ynM@!`D@tX
zw|^>QvvZAZaV0!j>SzfJ^2NP8!L{TQ81C-fPg;6wzpaS3wnaRC_q=JSTr{%v^z+7n
zHj<IiQ8TxeUy0p;$Wl=ccT#w8GJi0gBWBrs(>ZGSjy0&hz=dsO^yfoL)67k0)d%EI
zntyT5@WGD3mWhLK#jAbpJ>$PI2DefQZI4cuzS!6qv;<U#A_aV884#IU2&%VJk<i%8
zv-j`ZC}x}X2z79#*$rmZ$YEJf@)+#eLTYX8QFN!e;JW|vWm8o2uxmqS3a2`Jy_YI(
z9z`v|oLAQ3gIN^=-FE))<6F}H_!g%36RhU0e`%01*N9Yfl+;uvCn9@owz}l{EE&r3
zF=Hu?l-VUk)g+C{25PM)a8hz;Sj6V=qh<nuUaPpqCuh*I&O%y^!$)38C34F!VkT6V
z38Lsqd0{zV`#fT6e=u*hX95$>-W>yv4f_v(4~c+2mEnmsR1=o<v*Za`7z}L0B-oW6
z$WfCN4>t9mh|4Kf31BesBvU5~&#Oyn)|6x(puYNBT{K8GF7u|7_I}FbU4G_IdVPDM
zOpNP7i%^z85UP<wV;Kg$dC`Popn%1Iu!qzIhcdt>0e5tK%-FRaed#f5{Ba%F`7E_e
zy!IyYKMnmL>cbOOz<Oz%hfHhOzcvDPAo%ivJI#{Lto7R3`6twTTpap_aaCe~&IkSG
zgq~%0%nn)-L|L|wbDprC9Jsikpdk7rDR8~_fO1!JyK<SHEt83xP67s+0hwBWTq5=p
zi8FgD3lZvKJMsJXu{-Xbj33)K&Kiz(=yLCQz46v0GYH42L)(-l&d*DoUu_3{-fYBy
zkj;~bORBJ|LvL>4zh$2L9O&cWenO1Dk#empko9kWxVqG2W#@tmAVk}TYexBi*$Tps
zyW8C`n=wbkYkDR^9aRp6r5muXlq;mCW@NZO@G+f{HWVqzJRI6uV4h$LsTWZjFX$M}
z@B8-dx=#2Y07!N32A8F?aLZPXDB@76XecIPiz>xnQBig#kBLcI1X#Gd<_?wD400qB
z%V&h*p*!AVXBM;;zL3qFY#LW9(r?mIrCiu0`Jk4SV8Got%Ir6&1!Ecnhbo-&76oT<
zml<lmsk<?hHNvkjy-diH(Z~)KzEBY<j^=@zI#NH2SA(F6EwrKKvuUHIo!dWAo7Gfb
z_RYGk8B&!%r707cg%a_<u@D~oT!YTuqS4BvkGq#4%;v<t-?nfJVlETst;BS3USjZc
z+rP{Ym5hf{NZ?0(`3J9|AD@<6%?2(6TrQK(CPy>M0lj5dAm!77zDZd<v$__-?kuWO
z7TlUWKzBav-+J_Zwn9grSz5i6x^#w_fwavl{U|5rgH%h2{Z*|dQ!&Fp;+z)-cf=Hq
zEknq-x|UtFww<!(R4@CNFu03(-?%;B&PI6*Ytp`P;z-g+4cxxrPjf#gCXJ=E7?v!E
z{}Q{1v-Cy~{%`EU$7lNOTl*WVnLr!`cMW-EY<0pWPOtto(VyGyGYU}V;}%fyBx#iR
zVxsT-rm)eWM2hZsoGVYR#t)B;4;KEmQbq~|VOHdbg~d)DT&tJO7;7N84SWCw+a0Eh
zZP#y!a$XOVrDe=~Vy{|tdoM=$gC+=`kX5goCo>IgNoo1Uhn*|<{}YF6(nUNzer>r1
z_mE+@1>qvH+iuFd<QV^|t|VW(7Cv2=$}F9Xk3^DAEFSe4qHOCtED1svnUClP%>cHy
z@?>GD50731HH*EHc*qtiRISQEU8KN*T{;UvMMKkak+%|sO+&Q9QB~#^{&)YVIywc%
z>3~`@AD%;_>1a5ffo}w=SWxcXJwD-gP}3?j5_eu9-c!yp9OfyrO4$#9<u9-kLzI>k
zX=#Vdua}eXsG+JCKa`i~#rr0`T<)2p$)h`<r2S-}zoT8Ub>K1;ZMNPuHA|cSB3*b1
z!QkZM+XkvSh8ys=sUgSiUw`CkQIMX*^15Oc`it@nOQzXaZKgl`W+~@WY9PC4)hq+U
zwo=s+cJ@v85#pzH;&Vs-YsZ$GzhP$V_`pXDI`F5EmE69oDr7Xw?4u;xhXLO8l8N~8
zUJRCGCXI5WCFU;JLj8CL3NaQ*qy30@nvl~r(^h&n*ShV^8~-O}qm$0}t!M15Pd?tU
z3JH|<I!XzY1Hx80SsGz)5**FTH1*1qI3B1t4j6Rr3x*SsrPD&R&YG^@E6qypkfOWo
zFe_l;5#`Iv!eqajwp5K(!!Ew!bvEe#Q*2sY3@j7X89BwYmQsm5$|`>>TlM8Z$kFR8
z1NKyoy~G~_uh$IXg(+D?kaFKuU3HJdPz9#Tj0P-+tDw5;D9vOPJFKxeD;PNmB~o=;
zh_z>q9heBhM!~CXsQ8lSe{ir=*{~;i_wKtY;FxeZ{Uve2QizDG;@^F1f+$N{yih$N
zg)@llu3M>2uTk!ZlTi5KO#yT1hn}>Qxep(98r+o~|A=z_>eUdo!fMk!Ujk;V6#~V?
zL49k-+U=)qhuhmIMGWqkDd!(%W(TUa`IV)(pGwPqmcC(VZ|QRHTXzusykt~Prh342
z5r-eP6^j|H><_1n6%*cPZc-^TY8X{Wg(c%Q{k82IXK!hV8A~toA=)hfz1m*5J|C}|
zRH>gXKq}7aD!xWkcx{ys5gj^uc?~(V8aAR{91M`9^}I##T260*t)-Ay+%IDt?yv4m
zW0#PafR$02&M>90nD8*Aw1@t{+By=(YmUuThZL=k3%zX-DyAzL(96|7E2@7$STMGb
z;^+4o7+Sw|T#WL`rTEl1k9l0(X}5cm!){qV_Cvg4b5kYnbm{%*Pr*|#Z8^;zkaU0z
zvkYt-iLcZdlfW0RKC=i*QhePX3bCpP{geu-P%FXJyC0fvXXQwJY<el=QsP&!9L@)W
z#qA6CW&}NO?rk~Dd(1h%G6eU47TuemE*xV(MdF&Z2{4<L6bpR?yB1O}F(RstWuPPS
zc6Qq2k24a9zr0tr^4pD+gF~0i-76pLQ_4N_SBlWJTV9)KrJs*#=%6%9tux%W@xdfp
zvpH_ws`u=S1(#9VJTu#7pw5#iY0zZQ!$k>vh|=jX#K}f&MlE@gYsI-f#;hl1#mbWq
z`!3RNJ0}n6U%DGAPEXf`L9?jI&ri$jK5@97YotA4qn__0&d%M*sH!sU!68c(mw1Z>
z4!)mWy`Dx(dps^r+Q#JMLz&kZpvP_dm(O$toQ`!(Qm<S<`j74Rr#QSQ#KTTInKzkz
zugm0XxHVbxRpJMjbrOe?;J}fdBDSIll%L(XZ^G-bwqiMDLvr{AO@4dv?d-Z`cIAF$
z|Hb5ZHi2F5>KZ(Oh27V!F$2|H*<8Dln=Xo;-p9c~<|A|Rc_HXoNzx;xSbusPmUWSt
zvc6bD+DyRhwmC-vVr}laiB(rpIa2cu&E30j)OF^QB4k{)X5^j8R<IE<?8X6h*cM9M
z5BAGi-J~hm?A&dt56lW8WgR~Y7KJVpElI-2{~C9Uy*?2b!+U20Aq<sdb+e3jz{@>V
zYz28h53oCatzjkf?lCoWb)prsUqpW!jcR;R;U>x9m?IhSm|`?#qf2I*kuE@{isKqo
zkyV(&63uMcH-Lz03+C(Re}KnRUg|3ULc`y&fT^>iTt^mx^4M?Z;MMu$l(}UkYOiRo
zEW$fU4DR{*c?$D<A^!CyD$-7x4vh~7v>&3r<g#u$_q3Gi*ABC~mdfG8_esqhg!*RH
z)#@75k?y?--X~dL8puj2MTGmYeAYG7Eg8+$8{LgK%Ig}|e%Ag}F~8j0A26uqc1%vZ
zj_ZAv#L30A{$i57$h`mBc9YL^dPitmYi%OGAJV$COxlr^bSX^0F8gQV3ICP5Mj{aR
zGaXgwmi1S^ch9I{S;<LyGp1Ri%}5(YU}Q=5)*!(e_ec5MYNNaX#5&+Q0`3d?C&h%s
z%MY)KR;bOf?9)zrn7)2*KSyrwpB|k3SqvTbFGL!E;yHFN6;%h(i0A&eO?6IU^L71%
z3xGn>IPE@>QRJ;w7Z=?hAIAU{9hjZi0ecpJxoa{+AD#vX9()vqUTaDxZrb?}Ru*m(
zeo)}QW)<!^rG~Mk2dwvtzifZ?X;;je{h(w~P7D>$m@*!Z{b%&6QfNeDTdBOYg#1u-
z$>r<D<{d1NB$5)&052MgtWTtN9B=A`hLFIdL!&WwdcDuj!5L@4*t)`W6928kg4L~+
zcVOJ<<kW3grO%i7tDsgVT{r0rL-i*zmWgg2%pYJGFPSbrav+;sny&lfNWL>GbH<P4
z<~3$d)k}%^l7a!0R!=g3$n>HKe9<L=z~J_N@EBLL-F?WsU)uT4$Vey?eo(C^=TXa2
zJ`2I;{^^DzMQ9+IUtynU0`9-<ZPURuq)OZw?c>*J1X8YHj+^*AnSqH|ZRSTgG=I^a
z*3fPSzuhvua6&&}4}oVB<X<HYZoN$o>yzub%Yv<TOCy8I-@Z>!#L`s4P*CCe5(jsH
z%tP^0%pLo#jp}RdK1N~aJW&o!cBVB8L9dk>MD@Vv+~Kz&x3H1#m+>pAB6#(?G3Wt~
z?>*c>WjqAjCt6U&@knsE>8yS&%I7{>2URN=$-#3XZeFLV7l}>9JViR|I=-&h8KJs|
z4HLwQksd>dsA_`ABFeG~i`=zbT`?xKerG=F`7&!t(o|CMlEFMOCv?<PR{mspq6xJd
z%IL+S&<7Z`9!;-=-j|_h_6_H}bxG3Y%9&G`rF3tk3JE$yagmq|uP&w;K!Rk`LuTA%
zHKwO%w!_k@HCa;2l1EfaXPlWy>)|191}^RRdS-F7B?z=Ux`NZKIz~pY%dOvpm=|NF
zAzf706P4pMSoW5tDdC{DtLAlKs8y_F(Mx(c5@Q!$kNzk!@aCPNlPr|Tg^e;&iI4Yq
zOL(1Jjx(nbnF`of<g?k%>I!N<R)U+aiN{KaOjK&Imm&kpJcUx{%QFB@!H)Is&eJ*5
z=}_6um;~$xC#{f)AIM2e)s6;9a^JQmc^j^O-9D~9e|_5fc0GD?H`~6FXOy|Ldu%KR
zgumg7-m_V3p=9iRsY5vp45iUo6t94a^v>@+_e81cpv!n193yPI>6v_lMg!`r%KF61
zCKB0B`?4Alg>Xl<YvQGygVh^uTqbQXckxnDiyk1~!9xuqdMBH)yX4hAE`y&QljQ`Z
zqM}lEt%)0r*cDHKcyUW)g!1T}3C0#kH(@#EC1#$j@4?QZOtr8?gV*-sVuaquCXe)s
zgy?FhKc9I=KL%q|<s~#pNA1=F2JRxeHnf0$kHbhEpMwQ~VOpM0^UgG!1T-gH<)2$8
zL2+CJ7bOUE>6W&!K|*IVSXrv7lITl9#SZ1uFU8KpA5`!h1P`^RNwu$<aS~Pk`qg)Q
zY~x96jJnkZ?xzBSu$-*fgO_zk{o6n80+S}*{utP%5>8H%w26Zzi^(F}I((k21ZMzw
zgCp-$u`W^V{ND}8|9rte`qCMiYG$N@eP$JS;i%`ED6la3%=dgddNKMm&%Xh%p$p_1
zJH)2EA9_P%6*PPPz}L?n7=&9W&Bn(5A4g{y6=l~(VL&=xVCYb~Q@Rll1{}J(hVD*5
zknZj-rMtU3rMtUJMDTm~T+5#>mcyKL?%4a<c<Nd_;&7-%nzFOIdgprXF1(!iX`o!X
z>lyE<m*>Z~aqo;<6*kcuF+w6zP{51Q`DQ6n&)RHD0S#ZVhpy5gT9zbS0464ImuSpY
ze3RPZ6Rc!nu?IrUO)iTXlmW=3fubXkL=JLz&Xh1x<x}^y=88H9Fb_^@w`073eQinp
zJGdgJ97zGZ*A?8O7OK0^Y+H(+QLfRY=u#N2dId?w(;ut>%o!LI#|5_&n;Z%E@qkHH
zkDUvJ07>{nExIbt342cJsCLUHsK87lhl4bRo%ZuFJgyYFSI4xpnxOir8Xp5oakcVB
znG*anR=c!yz@!YqJF}wR$NH-3#=&_+)yO(x6XWux@<pau=AGFhVN#BWf$#xiu$I;W
zbi_-qT6=#_YDAz)6Ql_G=E}_UgE5K@5$B3HhE^RDmq1c@h`y94((v{swxfK=BC&co
zCwCq$GoMk&5lX-4;MCl^8t}<7QohB*C}thlSv}l+Y_HL7fE^bIQ7^5q!(`Boa8(o>
z{t{VWsK>uE!9fR|u+rKw<7dPQqG1idiJ1%LlOJBR-k}6BS&&7WY}YG4q4bMr*+A*M
zcmKG)&Kmq7+Mz~RUwX$MW4S?qD^;j{`)eZO?Be34W%C~}2>C;NyT216n&vD43nrbr
zykDu8O!<kM*O)pM9F0oiU$XgLF<C7kUMChRoOxS*i)R_Hs0J-)W709CgRO<MEyJ9R
z)>`$*sm&+SSnP&^=<$?m0b|kO6PoBtN+^z$+Poy~(tBHb``gJ?_w3(m?dFvNXxjyZ
zF)A3ghC0zPF*;aGx$_QSZ6tXeA8^o~hmc)^ZvL?r@SM$d!^yypqb!}^LtO5~4`FpQ
ztmzn}id4tQ#WazFIF1@g6U?>n3xU$LQL&eQ(arw9Sl*0@l?T(>tNZ5{e8Mn_+b2ka
zu5?NqNQ_RDPIT3FXdcp2xp<na^}RcvHDxTB1DhR?1(rR=Pqg~CTnJxSJ-~keoRe<u
z?m$qb?H}FAf()KyuKLVPz{%O{c}vS{zhOEb=C~e!+Jedd9v08(Y+_>KL13?S&u5>{
zk+f-WE`KHwvXnr2XtD+b!#pkLyexm@p<I8aoqF5(Ew8TbI_Er-0uP|o^IxH3iYdLt
zHl@~-;Xyf6-;-7|A@bc0``v6ILFKy-!P}gNo4kYEkDtC4>sYQP$$DHq{XG<&%K3Ne
z_ZoR%Z*3h>QNgkS(AwxY34mHuNS}e3_;R3uqWPXc3`G^+#x_)K^v-m|r>5}z{M{U~
zZQ$z|Nz5nQAzu(4zSOzX5AMLPG1s|Rlkj?(8i`}BFd`l&rNwsm$Ua99tDR8la1L-m
zmP#vw*R)~!ByN_uE`LmwEuXy7ZmD5KtSPLFm$?#ZW?ia`33@Dn|KK9Ay_#WRX{Hh>
zhdf!?ipsn%zO{6*b?^(i!QtL5LRkXCOL+{~L=_Zw_c2~ZB^jXz>%7@CIMA-<+9^V=
z9d(@~Q7%-qnw%jVZv~zyth6$;UYzK7OHpNyM#*fcqgl=`R%s=C`ETkmS0UqJMEF3h
zS%(v|2A<S0gRFv7z%BL`hc?2(w0B~a;s%aXx>0>GZ@}w;f0j{B)I_RFF3~T<5;@tX
zLAvP7_^-AOx1Vz?Sg^CBz7FoMv)GBa;<`UEn*B0w`RCs-ry4i7ln6_d7GBd@92xKu
zrXQ%XwF{?ag+CJO;oYCF-RmOWqFY;8+I{B%*|y^0X}sRpxuv0pl;X*7Gry`DEDDie
z2Kv260IeT(zUZ46#HXj8_aL(Q-bsmE?+k_Qz}RFyO2*{{Vh2!EHL9vjj$C7Z_-WyH
z&@OUx#JLCt&B=ZAvW)r4_WOHS(WfqgC1fM9c<Q3clg81q%@}0bO}(m;fKhEPORRl7
zS15<pbZ2nqxKR%|I^xpGwr<pHCxa8}u|OhhQhrhh{h&I+=AQ@N7T-fXpq1hWW?+1M
z2Oj+wlatHV-n{}@@45EszPsY&l#Q778rN8;IM!C9h+#kc$Y{mA_Ivz$D`fx9k07$=
zr>jJho)Uq-P^nX+cP48!Zbx!v$=WTe1nnYSH)IKmf6}XCkZ`V8pC*gJOB_}TjxfkJ
z(HeVe1<Kj_CAKi5E#S`XIvmD`!!=p;5coN+F|(FbQ@84yG`VGCSIAT>5(nDtcb^Ne
z0p~GGDz0!Adr&X_sE_xhj@jkUl5AsOV+qI!dV0jYvEUVR%uia;^Za9gI0vr|vj*)N
z{VJVenba|0h~xipfa0bcymANbpt~7INqF9gV7meSW`j@N(Z24v!(U|re~jYyS8{%}
zfN9G81w5zr)=8s2XA!2<UqTnPOT6W4N3hi)7v378D}&9Gf;5j}$_EbGP@P5YH=+Ts
zXW2e3$*~huiX1Wl3fm25$a|Ns<WsqX*O4lGRV%v514`b~K?RMuVKW;lm2wATG`rtu
zqCQe)8Tx{iQs@t7^i5m><dL=c_pevJn_XerkTF8W)HK(Ul%sM*>f1TNkztfO)U>kk
z(-$%f?WhNJvJ9E(wi2*3dy!a>!Kj6A`&6GX{_qhjiCJ6QC~sS>`U?hyVW?R$ETHNT
zC&KP>Eu`3>bIp^Mt|%TMp9+Czspe6_)14afHCadULDpQX%H_7;#6g>T_dv2<E6B$5
zS-h9tL7C#!wn?P3MLIsdCJq_u`^2!HStX39@m$dMI`Pb+LaonADQkiPrUH5noLY`P
zoaP;sBhCt1R^?GYD|(*ci>P#eaELz>pEDV&;U1@R-I#`%2?4ySha>yJ{#}X=E(!?7
zZ(<cj#u6>H>V&(1s9`UNu$FQP6Ds4*c(uD{!f8qqk&Q)i0jPx3v!?exSXCQaaH9Ke
zMqh<U4x7IkQx87((j|(8vviyF=>xPwRGLNN#ALlttHISGA@=T*kyuM+UM7uF_~_FC
zigFK0Rg^ff^je~-kKap3{<T*DW)szx6eWfhX&pb|?I_(pQouC%Epk4$9cSm_j&C$D
z80qoS=rn#f(~C@G1e<%R^+9}B;0WvJaneem6<taRG<{~8bmHvcA)~a{?<kp0>tDkB
zd_)@Fj{M?PAweRcwvec{J6*+92E|=V9OBJ*cWTv+U<<btW$aK%IWLSq#&7VlqVX0c
zkyklgB+fLjabfo^eBK*B1j{6oE&ZsNq3r=lN(q=+tsoTY95WW2Cb^1WEF(A`3d06L
zP6Gst5TQMPzy9QTB7*&I9YLl$<F8Jt7<eK(C#1)4THQppRiRYv8@|tQK5a$-HF#-t
z)tbeibqk(vZl~P)e6r=^{9yk2SZyQ#S#pD|Dh&?|3{<CccXFDob=G?(N{g}E4n<Eh
zH#!XN+fIuHS*B+j*?jlKQ<tvyoK-#ts8YHeUiav2bnAkqq5GeCX;{z7|LPR}7t>1K
z60P6YTcuu>%phbxaxK1@blXKYWlK7Q9wPA-NK#1ze&EN)Bc-l_h65mpm?iXy{WHOD
zxiUVZ6G_JDis&SBc8E%%G9;2|rI&7oZ**;}UH0B}5k~ufTplf`wObpeX_!L24A0=@
zM8gyK^j03h4D-dm_xb#+x+gtD2DYb_1!n3^oqkO!c}KVloQ<Mr*Nxcpe)bD>#4~G#
z<-)47TEW(m?$i&FRi$EUDA7&<Ny%#=Q0XnN(`YqwSSDlBaEubx&(PxvEly+nu>O7&
zg-1~e?z4g;$H21q&(51<cbCwb$wtq_Hp)I#kco%HqqNp6tCC(LxRstkP60>y&Kvu>
zHuncAJDwv>z}28fq!V5l5DsZNs_zF+=mOd<3K$`bXQrCqJh75g+WWPzQ%<8_RALif
zalH@_OdlTT8>wtjzz+VD{=lvzYs-C#bFrS~=)kwVb(NYhl726cPme*4>o0SwXe=a~
zWT6(<8qcRpxF1$e7bdKgPl(|a87#~b>^u1LlR2UnLmbkjRrxzL!JE=tA%SfzypO{?
zHMc0OwM)i9&it14!w%+X!o*6&L?a&WE`^}bFli1hv=U2ZYa6|7xrjPNtq>r7+*3U%
zwOqkcxb6uq<uI{f;7_cvEBP#9A+=*<g;jf1SXA*1c@WB3ooLy;VvG{Osi3x)>XA`K
z|F`94hil5JzrRaD<L9J)$DLNenUih4T)0bkcA(*HOp5e`b65-<8`1fVwuWJTaVyj_
zVEajmn=Fmbfii9xpX+cmd<FC_)1^NvSF^iV*sZu)z@$nbSHUh>aXrD&+Rs&F1x5)w
zdR0%O8m}L_{>pVZY0DUcQdsq@+(s4LO8macTqV!U2cwMrf~?QG9z8OmuUT+eRbJL#
z7KZ+98|*W#qOcS<LV`C$&ixO@VIzIp$LsL#f7J`NR&0^ti>2b4C3fGW;;GGT7z-)Y
zN;!fC66EoFwSfUh)0eEwDt|**W8;?@md@B`hM|k%>sCWb$lS;7Cjb-xjH4U6<!4vd
zXP_E;ouf6-8N804sXSwdhd!5@;F$OC-<w1lu@kaqOlS*d9brg()2^D&18b{=^X#?!
zSIzeLsXNQZ-rl~d!VEuQrxH<_P}wqpc5$g9HK`@Z{3bBk4wYw&^DX7LMD*oY-EO*)
zE{1S2Yp}Tz{~R1$$8qsXUCKC9Ra`Few39^4+kQ_NkA>sAxdpJ>@%1}<B3%5d$1|}0
zbV2-n%uxe4O2Fs$8_0mz|9Ez187Wvn`!9F5#vh38TASD3p3ci6%wEy5x^!S=j*^u;
zMm=dYZcN=SMHb5Itz1Ma&zMa-RTK}xI951iZgc@XrTa8szR)Lp-^f*nhxS{30!gRJ
z%fhW6{Uv6}QZCiYewS*^M?7Ec#&|UE0}4uqhe|XQVa)m-`j%0-X!f_NOJQquDJ0XX
zo@UP17BYi`%?TgSt@LoMV>nQYxGa-7PV8e@Rf@-GZkHox)B|n|cj#5S7*`Ur!?`+4
z3CSl%YlUT}^|yaiclQ5@O;=;AE|Mh)1$pY#GLt?z4fyCgKVNVlAvx2|W!B#OQiK2M
zKZ9yZstDV2#>azS#Qa%{Q*BNT!xAT<m_rMngY=FN)7w<*6TY<RAiN^+z>k$$R#~$$
zezOa$z2NzEZ$-X8%ZPG<10G`^X72uAlN2P@r)=dTr<kEPxzp=;Fmx=<5C+MkgV`|9
zy5{K#l+?pZF&}GOvTaN<wDA4{JkJkRRx5ujY<n-&xf>a4)Phhxv6A+b(y_ax+ikLK
zl5ao9Tqtn$gFai&QZxVe{$)wJ<@yei>B{1ao(T)@l<U)R0DRecH0zgJgux5)+^^0z
zC7YKC)FVV6b&FJ)zNX~TMaCDl!{^rH1xD>hGRp#9aqX&+SrBg8oK$J4;;70Z#q{zD
zwiIy;ntw+*Y@bE$PwR-A@HY2L&I|<r8K2k4ZJAlq))vBnJ*g&omF~Ahkt+fB==tFn
zg2v6QG44)>{oEQ{`@-ox(ex~28-Rvxxfq?Km%3rZP5B+9dH)T)bnVH_EP#>t`8#2t
zs%;o`m0~_X+2R9wy1zGna&@QjwP4QFYIa?Ge&C#xVs~`r+~DA6G@*Tk$Mn{-oVx}+
z)|9n2_n5vu3EBiO+V{P`Qy-mrDdEgv>ij2wXAboYy3xj5riv?v9bCBR-Jx&I#d4Nk
zub2=n%MgD2Fp0O74qYbtFb(M%tAXc2wK%VJU8lrnXgz{cN_%9zy=~r-|Elj_-5(qC
z^{!y8-SalGSYG7v@9=FVFg^X4C9mUl0>&(oH^~HD4bzHsr}j6q{+B$F+Rx{yPc!HW
z+hNnX_lnlW@+c;ztC4|PpFx8h+45J<_;q1Hy=--Lecd;PYb!8%2+ob0{9K&cOr4?p
zRuiiMEfj2*rXEfoeRS`=x1ZUntCWlfUZIkO3x9A;5co)|TF3K~OJ~vIlZiz_zv&A(
zEo9|anaWKsui&EmG@9YD3ojEwk=QPTpx&9{M^I_|3ZhX*g~Zj5F-{P*rktL}FV}hX
zEiZ!cC_TJX;R>qY7kHoXJe~M^N&xgwlvi#Q?~-uN-1l=@neH5ekQ!wJL3lnbD+#V$
zqD=8FPk4(Zn%JIHmP819L%Mk-wu*GjX=2-SvS%A*XvjG_CZ*gt#+BnYp+%%cjm5rV
zD=iPE33E+NAqq8El2j@(KJsgNERSEjzHD;wH|+U547Eb^B4<@8DSjPCPElj^@i2|(
z`;}^+<`^cXS)-ViMo9+Uw$3O$v2?;eab~I1s_JnUXSMawe4`HfdBT#8AQ;WQhklVy
z*h(G{D0hR#5p^*yX<h0jv;$MbdsKi~IW@N0&c;5x_5&wz1;~-D&#*_G9xAh&I$L=q
ze`311iMFrP*B`4t??G8H690Y1qF*k@Cv9S);-E$7r9&FF{n)e@7+3-8Sj@JDlMz^g
zc~!KL3)NKQ<0yt61$;qdSRX!8;|g1Eop28w7s|2y_<5%w<<5H!9iHM$S6o7x{}2rc
z#B|bm<S!tvt(6%AA~)m3OrJX_Jf>!5C>}4?q2Ashwe&J|z&?56k$LI;AYT`|tNXs5
zB2qn?OV_47hDA|Bf`5p#<=CEPZ}2F`cfBM0PcE=ox?Mt|n=v%l3vlw>&UzDo0$Sr0
z*srzdKQ*Za!M!~&$D&{juocL{cTqBx)jt=Fd2*@%&>fp=Own&CuUKSdX4W4sTy&fi
z`?sz1Rf;HOyS9j~SLq=t%-YRTS^=|t5ob7?w)0=JT>F3t%2tp8mT<!x0JE+90=5Ya
zp1k<T{;`Qr0GYeBl?!=9U5Y)$$<=U;d>(j^()_@`C@s8t>F$!##RUPTB5z1-Q)lN}
z;i;_x<*<a2@6V5y&Q4BF!Es(V5+Vv0b>lCF>>j3(35ODYE|Q^7*@SeqYyDoPExMef
zm^!;g-mHVJ-?IE)?dEH3EX?Yb(;j8CPr>6$Yd%gK7uSSs><OBOKidoHvuF8bFC7>Q
z`Li01`g6HWJ7C6j5f%iKpTZN4Q98cn3~TG!j*y@V%HxK9nI0eDP{K#$s-H>a5|<Ky
zk=++;EvAL$tD+$m+9u?C#%4*Aykk|>NH=rWRKt4`pfjk_s#+*d5#K%Vo%^@2syIMm
zNlC^<D1gF4oxT;*lsc|eMZ6xVF<yeNA=uc+OJs^N;3cI$f9Fj^qNB5?21z<xDn}aJ
zy5d_uVswE~&=<F5`E1ST^o6v$G~7fM?X<y!har``YT1u(+qBiNQ@d{Rw)i<61+C<^
zAH~BUAvK_rLB=F-CVNRN`SV?EB6F0Ci9~$=UHp}>ZGNseFzGWc7yCIaT7Q?kmOM@?
zr!M^w|0$COVP-Z@FWjJt;fAIlivlT)?k0G9G8Oz)azO!HPf?Ti16tiNp`?#P$B=}S
zIUtQ#^~u0sA5UwoL><K%FTq<u&5OsasT%w_TLa`dVxz`B9gY*Th(WJT!QGk;kyym$
zN_Yy$R~PG@d_eemH`OtI>3h+{Fou_{$giFFM3wzl8e`s$*w}V>Hj)p#w7$MylwvR_
zS#sQdn~sa(H7N)PE<6EA{^`Ad3{19R@XT)5mo-)%T2Aw&?-!No_>GnFUjz_c?mt!x
znsJ$O<Zuw81Gn#t$GBI=K~EpjHw=3&aGqM)03jrs;9F9aLWD$S5M@)pK)$??V;}wb
z<s}k0;1LMpz?3!**a(T6<p1cN#RcVkfvZQ4gdwjHja~m|%4xZZh$8kMB%Qf&ohxVD
zQo~>Yda`HD+p((02mN+j8_-+n#v~!-zd4WQw#>jld{mrImjEu~P?qJ$lCN2)?dVwN
z^r+>tv$Io9hN3wkjEkUVx~VwbIQIs~e7pyyB(JYu(aOKCCq7f*6*|3eR>AqSZb7mi
zrD6H`>G-i=yMUegi0Q0$B(hk%H6XUOIkX32Snj>niw!B-Pl06upz~^+Ra4%(`6q@J
zzW=JGHUNOKSLAWv?Ao%eoT7|>Qf;P;>K1_j;=tddm{<C)NAIr1;ves!^VcsM-&h@h
zqzAs}w=v>=ih)Sq_peoihd@gAZnjNBWuZ<Rb5px_ZCcW!+@uDN0?uwL-Z5uWK9w#r
zzHKB4FUJL-LOE}6y-&^$5vq#9U6$g7uyV*B^l6Vd@c9@6hmXq*<2}Kwmbig=Mwd4p
zjDn0ruoTUNa(PuvWiMfMgGpoRM}1bAor+it6^QmMQgtFVxSA=C19t&6(yo0-K4?q{
zgfP?S9Hk>ut?S4GBBUp)|2ipXv)mlK$UR0o453Zf;?Yy5H%!GO!Wu+J^o^&+EFSeQ
zD;uPnGWLboGPT$BfBv{Z_>tJMbXLVIdw_Gtl&64EZZPE`bz6&orsM1T3aZb2N*0R=
zh5aw$)Bm`SHK9WE5*AOt#_Jq4i+(`*o%WmmIgGw|(+M?uj~mwx;*(!eC59<7N+}g2
zQE_78%@?N8MPu}hf8;`_u`H#ki)6O7VXH+ikZ=g!fiH5=Ng^6T8MO${M-}m2Q2K>-
zaEdUN!0pzV?}X4=b&O>aJcBqe7;E+Qi~Td)vPE~R?k+En7FiS_2JeqKSQ>0vvfO8h
z<OqqxXqm)fR;F3L9>(PwZ^|siAqL#jG!Q;nkf%Xj>J)zYfwk<ZjO$vgMf!?YeY7Sk
zKIoJ?jDx<nH@z)g^Zs|$KaGDRUoQ^-Q3B35+D*?}?o8KTYCX@8z-)zk609%|pngdx
z$?jZ!U=OZq)6_bm9^UNo%h@g1bu)jY4*6sx5U|510Eq&ZR<|9wx_5}R72x@*KmG#(
zh5&u`Zh}cZpIN6Sd&ZPRQI$YL^^X?@6FgI#l8--BV+DV?{@x{47!fjXe7wkVyC^L+
zws)_p`0p>f^kTFG+{M%Ltlo61ZJz0N*3@C-(C%Y3m(v-mGmLrACS0@s$NbZaSm%@M
z5axw^ozp?Sr!M`fMRinEnE1_OpdLE~Y{C7{Dg>`jJ{u>aZ>&N<b2)Y<o3571#6bFA
z^%K26-3Qg5Tl0-wU2nUk7WYeiG-dea9)_h@KeVM-Rc#HVg>!-Fc{SG<D*x8i8#(Ro
zkdnCte{D@sh7C<-{K(H#fNw!Zh-zH1sZTr^N=i=N_^84axz6R6@(tH;`W%Cahi?6u
zrMyJ-)$P!vMb|iT+x~A<SI^YHzf-pW=jj&EGuJ;p1xB)Vdjk^(VB~-B_{&{{bCAqW
zhNJS98KwDB-T&RWa2$X(!pnskL3e3vGVT=dAw@7pPd;DM$<Bt+P3_1B4Ba*vp0b3d
zTD%|ao}JayYMLJRjMq#!-L{R&l`wvwS!!0Pj^z{CDvs_t0wX1@q@`BMD3cwoydSc~
zbwnNI(uFS1xdY%eqcT+;cx-felJ%ekYT8~&h-c>`#EZM|y0CvnFr<QU0V*Cx03rHY
z@9SAsODT^Fh-H?Rm-BKkrtu_4t94oVL<CGZGL%64jg`pyPFp?*n)!T`(Nu}*nx~H3
zd?h1Z;VcO?+AIMIlLxdBPm%ehN_>Ah$jhEGTbTh21QdONpOhSrsG8@ll{>w(@fW>%
z5%izMXwzo<R94%NYRkZ4wj{%>Or9^qRLnXgpHiMeUMNsWwu%5-DwbGI;XbUMQeN4s
zgF<ViFEw<!ROD`n<uunZVh6c^3;#!1{Ae&-a&5ULp0im5!+mLf6;*uCe{I+`c{n@;
z7|DI)b|QUPjmvAD-R-KR(jh-c*s33UZ-3P899?}k&u$rD46j+Kf0X-~xots+E?qv*
za#c=N;e5eHFwQ7d4&r6gHN6lzl@izDV<d2KPenciHDKVS-Y&yw=(v>63|4y@#oTw8
z>=aJ1{rx-Tczi<>u7-&j+$T~t|EJ@eeq%@x9&6xh)s(cz4fbFLiRS5aoQO#rT|{am
zpkW7$XmoUR6ZxaKCTE}$aRm_--tkE0YCV?FYL6Q=bR`;YZ(zK}$|`PMi#lY9PmpK~
z1Q(3B^p>K0(Y6kVX*_b+8}HR%`=pMgh8G!8N8M{Z5s36y-6$!|Af*(?ecWZ2e$hvK
zc&~pbDhKD$^g%ovT~bUqOMZ&WiQqp9hKx$vFrUu^HbSyY3wrF|S^i`Hp#FY6(IK81
zu;BoB#5w$KLK8sC^;SBHZrYm6LvOyTovF``a-#sy#x<LJbFs3f(-r$+j<%$r-bs{L
z0U@Ec6p}5#6l^Y_F)TE!q2*u1otkRp&K-2)SF3dlX8|j@{)EQ_S{@En3$HfK=LZo5
zg5qUF^;$aLl>aj^UqRL^Krv{&Seeygzhg;@z_Wg6Q=?gbmg^thTB#FN;9pzWST;6|
zg4WR1wi+;G@H@z03vlAI$T1!FV+j-*?gbI(TSS<(o>l6e78tJqFb<;ilLX>J6Wv8S
zsf`A^NL)AVQk8*Lb1q_G%E5|N4+eib6O)1t1&3A!!K$1KJ4|S3Q~p{RP>wbo($O9&
zFb)cfR8#Sd=u~~;K#1b}Sy4~TZX1<6v65a|NzWwn%s@W>@SYR{F<0$KC>QT@5GXJj
zUH?>lj2d>Di&ql15w-HJbtZk$HK@W$*H()af4o>lK3ZXz<lWf$x*n@RyCk!!F9>@8
zcAa1>bABfN`SHTB+OONI;nQ|#WDi%bhb>4TbfdZf+MtW5{+Xd&!=ohB<%?gevF*h6
zHib6<D;IkFfW?p7OzLpWiNk(@o_>yoikgk>hWx=0E5*yyiiNW&tBn@--mGU8GbWi@
z$GPN+{?ttBMs!N4eCr*_xl6jcXY4@nFPRo6UO}_bnge$`wzXVTvN$``Jpu~0f!9^a
zj7yzdYMJ^BvLrTG;X*@N5e?C`%$oQgM;S{}3T*~Wd=NdcaL~hMmXOu!_$!;tymvVh
z3epw+7X1r96nQ}{n83CZ*<bB?Zd2`pO|o)Hmy%Kg8L?hN6c!ET^1t-YdKWhoeCV~7
zus2T;jDS1rrkB7-T=)dyo-(#jQp5B75zvTP#g%ZJnO{2c2itwacbNCz9yjnI5@P(Z
zxwLhDym)8z{5lKLYXYSH=8?fvBCppmxYF2+URCP(I^so%#sX;$j~9By81HTh!)YaT
zPJHQwtfW?_IkC9>E(1dBp!wAVvq-ASl5@v*UmT@0&{+_3$hCF+ZDJ8RtdN@N=g;mC
zhP{3;N#~VzR}v2Zxx&C&`7Q)yK`WA{qWIXSbQg?i6w~TVY0y@lFn;U#l3tyKvkN>S
zWl{<JA$8?Bkr(+?5y~1Gi}QmlDPs#863YIw`Q;hWAViZ6ePC(;yinxN&(G&@n-%{h
zO(AJ3Qq(k!@kBgb0l#8EtjPB+m71Hzc5v?&`U;P(X|1Dv2Ds6HF-H|%@z_&nv7CP;
zyIEjQE|hHl>6Ft|kD?uhuEkxijS!r0=(+J~(-io9DQDQ#rNSIgdGzVpJ32Y_{@*c5
zC3-^zq(A`fp{HJg&S@*Pwh!$YBA2x>vmKxCZ-J~lz@0nEKD=JvU<0$43isW-jtHv7
zNr>L;`wRev4tV+b`7I&IfNS=-bI$uPN8wUp{&7h3_4=cT&!J!Ob0=WaZB)!{`}W&N
z`my84Ee2*%^-n|=_>n02Da9SU6TTL>Y~w~dI6~TI85uZ7xY&R#fOXaCJUcJ^8OTz4
z6*;~C6B%-p5bhX*aMnJTN1Pm7VST1c!Q872N1o?bUS``ZvXM~b98XWdjRH-o8%;r%
zmdNK~*KoDX!#uAEz_mTf3U{-mLvd(sm0ZNM%~MbB6*wt1ZP4YgRmPHvODJX_0Ayva
z<EiB-L6k6f_(D1%y`l476FvgFM_ZwRZ6ROeiwiA#CoeFSK;4$aB$Otf_6y(?WU`VQ
z2m29S3<GQH3A9s;Lh}_#l@{;-NzpA{T)@56$OpN!s^sztngzv0{snapL%sR&MeYzf
zG;-33X5P6k8sswFmM5-YAi-H>><7~BH#6&0qeMEe2@h{Dbq%FFSJ`}I?Ksh$7y2KP
zezOE>l(ui92>%-S#$<>g#9~?%45Oqv^Quro?bFs`#t6hU1d?Rt)0!k(c6K6b4Z}nU
zhCq3f;0|)D7C@D%T0qgZn7!Yk;2Z4X{pDPCR*uT<Wqj&(UBiH;ikU6}SF`G>@$=9G
zTApj)4ZNX2+V`M<7Pe*C=Og{k*IO_JkX>3NCakoOrFWw<V>t3B+)0o9vPM=zE}R1b
zeLfu9|8D$V?5{Ns$`F2}uXvYx=eT6W8>4x0fV77I7uxMvg-%}cP{>ZI`qhXhukR3W
zc08;7at5qETaI<Fm10e<j9fC#3*5)m7dK)!+~%cJD}PJFqopB*vC`R%aDeNbltE+G
z0@1)MOg55h-wxzM5k^Z<-Gm|D4A#CTb+Z=SFZ>wc_aFQ}TbI`XTIf3cs-s6~Nm{d|
z%R2rIiz+k`^r0@5^Awl>X%6)DEf#-7i&bs#s4Z~ph@)15N=~(0zrvnrrwvoi6xWe#
z(HqLIq!w8MIZ^~b(PpNDvV>sP3qN7@5%7(_`NV<e8nA?KZ0LJ$7M*ByD>Qk0N|pNQ
zpR^hGt5%a=231My09Z_OTdtR?8TA8F;g!+_JHIc-6-;M9-iaY#XJfFLX#h6`2~ADG
z2*WNoy>I)Ylv}}huAVUJd#v46Xt|=Ugu2W42Y^^C!gbS!H+$VP0%ugT7m%~uhlpO|
ziuN>h-m`?~TU=`J7WsS~_GKHZ?MNw`KZ$ItMY~HvtvG5m1ed!@VqV3(&WQFt8-7?v
z;EMU!s?A>^5%OK9$*$v2IM$J8(DThR3;Z%xcd+;$-xtU3x&;2vj#~E&G<y2SszbL4
zNJID2k_i`~KL?WO#%desYTrhfUbeunztAu7qIyhh>>7H@D3Wj7a@#kFE!4OUW)G`y
z07YSks*o26!OmJxe(1&ZRyg}0Zjztcu3M*DG3^^w%S~5j61>addLF7OnR(air4?eO
zA#KsJPZ{D&uHk8Qs8lpi!O~cr$t;%)op|kmojZ=Cj%ao`rd+fa89P`&6yW58Uau%w
zOTjmDg4jW0cy$H1#dNrs4_FMS13^I6&`BDJjh%h<q(=FooC8G^v34YPBbGleD(=wH
zQf@ljyHnCm#PUI{kP0;2{cG2tdrUG<Mn)%{t<g;QpjxIAC#h_W@#JXlAsTlqG+OJv
z01<MDGp0XE6WyvdcS5Tq9J@xnhM<Ig2G{eE+QjPvT4+o04_U^NrL|6(_(6I~D~)f$
zc!|M=K)CT2)6|LLaXM)}4U-D0sYU-Y2e?&}dgIi{W2|6PoGUAd&ywjFN_`w)kO~=D
zjq|OizVEYO``w{mn2mJ;G%g_pt#R-jYL(~X7JR}X=~|aWv6Ze~{9P~Rnm|wm;9Z^N
z4fZ_dcS~-#)@`pHYh|se+ubVA(9lF+!e9vhc3bW<X`@_yg$WhY8R(|&Nr^?(fjnJg
zS>CE59awZ~26){$|I5x3`(h<T95i?oI+jvd?QVosgp#G~qhh>m*=?v{eb@{LJ}?z9
zHIo2aZq7s1{W{i(V&J8-1(^?Zlv3Vmpps-9ET9q(jONn!4ln(1YVL7<y`r+*jf~+H
zh)*rk$^iyINJ-FT88+%5zjP~?$!{UO0<+Z67X5FWQNpO2?kDe;Y7N63^amAJP;G_l
zq^6zD&-HuNX~vGf>&M4Ra!f88#svE9bhz%flS8{^pz)I*-5p5*hd7`ZZcs{<F9BLd
zH}Q*r=sNTR%RjRC4wlz6#EjQga&w6h1Z+;X2M_>VGgVFN#0<*@mkXRnK0dx&X1TYP
z5%8z_7OSZ}5+m%5eA>7#OhEX_^jpLL!eo=F8R#-sp$zr%@E8WfB22|86#f;b47`Ns
z+79!l_bU%(wbgd^_D$uH)g?H)d#O(P%IFlgoA=WRQmWSLa8j0!v=Xz`j-($mCf<&(
zeaAh%=u?8|a$^`%cmN<T$wVV*0#qkJW2$4wCcxS<6|aw{AYsh{%xY}ebogvd7AvYN
zPg@~mvaMY@xp9a506pU&?NEC*?9x2*AHU~o5z^U{)#0tg!5u)^09X=D<wps`H?OW`
zxowZyFo`hG7DU+V^{7}2dcZX+?y$(*Ubbp&LRw!7W7OET_4K!0CNkE$B+<A=6R;ft
zLm=z>#C#$=#LGZ`$KPUEbf3JbFBKr~(HWapd6Q~_2L&s^*F*_hK2qnTr$R&YQoujU
zgb+R;{c4~(9cW%rE+_Cv*)oY?6P`uxo{x6b;?UVYQ=lVNDBio}-cYx(vAv#&+1Qy)
zis-itkNzY(^*kMdgH*{pGM4VtL2FRS&(CM-<<sj_1BmBap6)Oz*P1`uFCo1Pte{X+
zQ^w~X4XQ6!N0pLE@<O(4Coa>DknlumJ+?AJkAp4ao`iY$C10d@=)v1__q+!k;Cv*V
zVtGyo@Rp{D;Osde*OoAq#Zlu3wdO$zAM-YOXbeF|veazu$WOUHe)I-GCrzEkjX4G+
z)43q*w8beH-9r{+v#e@byfpW~m)E#Hk5Lx?mMBk9>=)s0uXblZ70iLH1i~%JyP}+B
zOnvc^h<`Z3_*~dQu5Y<;%dqP+@OLjYVIa}AX!wuspFYsORe6|$z|PU@JF}soqwKs=
z)I|55-Ku4DiwYTQByFW%<mfh+dZq$tN_YsaN`|T)8|C+!%!ceyupAhHG@akqOMZ`g
zbop04pMGLxTvz3?&h~uJn&m>{guhhGT$JJNx_C$RHg9nDa3LI*K7TYhA@@5hPAp_n
zrMnua7|MhfgC*!~25bl*>GE*3HWK_snOXPzcl>GD_4B&QdJ70*lo>w+em1lE^9yUy
z-{6>A$M@$p9B#Zk3wivGHP7x0S`avL!HgN6P29I4D%}#5%v$%J#Gdr$hgFL2O|-`4
zXk2K8=siBaxNRrx1mBM5!qCE+4h;?2w$-*)0(Q)Ug7+`<03S`HZ+yfPT-3!L<P-_e
z3>AS};SCl1mKh601~v%V|ItrLG=)k;uXnhw59q=eXt2?tmAC+J`7_IK#)R1$F~!5f
zgEolmmW~GxD2b6*4vRnZd2c@OBrmAmKDAc_a~;%oq`~OLVY-~l!!w5kSGN(ma~H3V
zYW;Lec20oQu*c6GCobCrE9YTV70&NHc>n}RYs2+(cP1O<rcdf(E5NIM`b4{V8gRn;
z>>@JoyJPSf!6oKT2WzMo01Tg0y%s@*={-I;I&a!MxVoz5)=bXvjAvNB@TPzK;{%uv
zM&mb9qE4O&>9my(NfSm2A>w{jgh(Zgv@zk1oOHb_s1ZK@UVX`hJl?$V>x0dGeSIGd
zuLB30hmhQ|7c#lnaST;q+-j5wg&^ne^2FrxAd;n=^j21<ijDzF;m3F1#o77<k@`I}
zlvKH0=TjdAAlcr@x9Lx%d!~(B76K;OY}cdCyEphv*`C=z{~TY8qVJiB8iYy9xoui8
zQ0aV7r{!>^&#X0hSf0#8Y0?Zt5iHuT<%$v~m@?>VO(RAQ9z`;T_53*?l_)A^$CRh-
z_O3?Vhxug1GfSM=Lk!XS4u_JD+}>v~v9lFgt5M?;{1UYP+Rr&bwZK!bfHEoI^>vaj
zP<)Exr00Oo*hDVwuG?~2Kot=SQ6iqYxRUA53#?Z1R!32(9DpV{RRfzU*V;_Ov@x=~
z9;jMabrdwxOwOmu0NK5Ha-p>0y`ONTQM5B)+Ut7@i5MYa30$B6O{|p!dI=8UgP+zZ
zHEQH5iI!0f+%=qcAv(G(uUu9CF3Ns+Mt|fOJ4QbzUGYw|@7Gw^R`gi+Ah4QnVBb1A
zqE`jFfPGNj29A}vBcsF6q(;T-EYH8k?a0^-RawW|;|vN5e!hs~?-<B*dmRURdjRv`
z(#c6EuatQfeivmxP=2{b<$wX^eoj`wNYC^-c56S&-aRV566a;efo~5GCjs_4P7Eg>
zBTDLF`I5Qn>nHMK6rxFQO5E<6hG;&_;Q6|z%oXi=PTelIcP^TLBY9LOMbt}ThlpZi
z``olzrD26?0*`oL7qy#&RiTqX9Vd+;@r|Yc{s@{xE$j(o)MooWXaPyM3#>mlW8hG~
z=f#zF3>eRZ{|kb)09msDP-L9ryA-HcqOzgQ&|SsR>2*CwHI>e2rWFAPIF>BdyugZD
zIF}tC4n%Xla(Yh)AXjW`GLL;`S%QC@wz8Q0x4!KnV7r>;>irAu61Ggw;TPV@nDOel
z2L-Wc4O)DudCLRfp#$9I+Odq|Y3)paQ|9n~(ZGIgtZw;Cc&dxE3hh^L9|hn(I@fjE
zd~f758CbyD_P0>_u~QTX!8D)EJCxRVb^{vF4(~^8Go{Xdg<^W{@~!39o(tvHl*}CR
zFv&t)J7n}*Z)mv+6ETgzryqoLIZrRfneA0KYY_`>JQ)o^>~HQl-tWECUlsrR>(N~a
zW<aQrH!Q8kI?MELg*gilfiO7m<s;IN;3=tQuXqFBzjDTt*6*Xrp!8aLU}yI$s6f}_
zpk=?O*3eIlV@$5JJ@*iAu@O<o*E3~|*KKKK^}~d5qLZcuQjhO&L@;X2K)Mg-?%q-A
zfD>3PIoHtKIs+i%xII1`Vg1QtmlRCo;3G=~<4Mt%)NCm8vPHR8R`k2_`2Pp%9dMhd
z6cwZO<3Dv&S-d~nUPVV3Ltij8*1G)qWwAV)uf|J}%Wzx@!C8Gr=@rcfy`RS&Kr>^c
zfv|<(pklZjL)A}g=SKOVu1eLqRgUC{eHgX+0m?>3230qT%-hUR6@*`xSVvIdL{#)b
z4hkuaL%d0vex)xU8p4Fwexlsa{6dar3|}P<*nX4zMw1YASc?gAk-eR<saU-=@#BLY
zb&A+@WcdTMVxY0{7`;yqalgD7uOMI-Gp4)>*Vg~F1@{F6R9Pk$?a!D6d^83~JPlPw
zmL$()7?rRZHJE6uwoGMMjn*&OmjPJ*&3D`Y>v1fF?VUNd9n0!L24lov^I;ZnUmE)5
zBqT7n()_8LdtdV0_TCWLq=at~=O-ih<tIiEOx?A|S}pGGHsPf_8pHSPSeNpK#)@!K
zbi@keuPwTb-^$9!a3|83WdYTqrf*SRDcR?cJFC6zo<Lvx2td~p5ZFg0az9#f^*!ry
zW`tvIIdbs)cUTl_GrwOp&N$TNUco5~CpF92RzWsR<ELhltn^Iigz4O}kAd~X<Rm75
z_ItyUn$JeXFC=;52ipTSHQ;-c4(Q^6L3p#nHe6uO*0}I-b`wmzU3kNoNTvbvdu{?w
z>UADcOnw1@oxQ!nc^^C}Sckz@YKY34)0F3n5A6Owgx@1@Ouqa@l<Jidnu;@c;b!2k
zpR<Lr+sAfR?z<Aa7a&bzzWBwpX#H<HPXKKQ+{rs>Paq!o?sE1WKnk$ia+qy>xb-&^
zH57mlz2RLd8s#=bjimNH6Q^(o@4&FMY0tOaKR1Ehoj~Mv!QiTW*F;axTbpV=c?r%o
zcTb?1vg3VOX|N%V8De1vQZ{Ty8&-u6eIgS+f#SG*rU~gT(_kb$2Z1!(rD9Vopv&D)
z(tugP_l;RWZ_l~b?3PcKB9KLf2~rX$H0q#S;PU<4r;tb5aBCq{!;?ylNk(KtHtU62
zT3XC1SGuCC8hz|D;H2;DvDM70byLlpw$lBow>;VG`a=1>qk3Rp94o_ewA8TE0bu)r
zh$3eap_4x37XN%&`&mDUMWr{A?M`3-<-qZMVBEIA6os+{UI-jo#g*)K=D-5$9kJ`T
zhB`fg<?L4-Acg`4>w!%hyPJZ%bc^?pOfnGmdVxoJ2#b#)nUnw#4<odJ$`~TbusqI{
z0q(!}{K-ByR7GwfMsbu&J28>##K=WbKd7@6bJ*+MABfdB?_{9TEU?LOaPbImjqb?O
z`_l(28>E`$OpY5;TFYs`8(h#wtF}tT>DY%!OIV?})UEABB8%)JBj=6B=$%#JScaz-
z=X^XX#hi_V_0ZB>7o|jeh=+w0qd8uj+JEcd)ZJ#^H6V?N6;^LY?9D)lyIsRq^n>MY
zE|O~kSml{YM6p=8U<l*lt}=q|QU^6=f>{|><_Ir>MTQhiUC=16er<3Xusp31h$ti$
z>2^4FK*@HfQKUelK`o8=hb(W%_7EhbFSH*^N2#dy3q;SZL#)DS_JnM0GfsvxUkyEq
z%MD-x*a;&BfcexqBC9n^!~k3oNJ=Ra{LZaGO&Nhc!e)%Go5+6}Lv@?f^#^7ZnE=Ko
z%u`Z>KQ}2#qgN-qT@y_Pepwp&Pawa4m;ZbLQ0MC)$=}<5&!Xlvnj%UIm2HZyDhztT
z7ww%gOXx<v@Io$agvfGqgH3=0W_Q;F7$vSCNdBBg{+ni=y$zf<;Nhu9^V0C6$~K2P
zIJkWxx_$%*C%|$WxNhFoZ!mKRIfrxJwOb>UtcxoMkDQY_vMzIawLhR*gMj9%tSI5r
zI{@5A68EML?sVo%QC??cPy%1ds<_W2hVdxgc~A1cIdCvTF2POVWKqmZdIff{rg~dz
z2n|jboODi^9Ua0j07z-`WC{j2fB=BORBI48J#QV``{$*hm8W=Toc}&ZkpP+P855H>
zIFfx&9r&J}F*#h;i@t*?V>CHHz<AhhYDu=&3D;m&S!K%MewO>j_SV*i-vfYFRg>sY
zyvRZXi1cLgi+xn&QCBvlP&>U@b2G`bshl*JdJK2G$@J+WTm%F;*47pKZ+R244EnzJ
zYMn3V_P3)<RacMyUN@x~2Hv=W02AY1yYLiHE1;aIS}Se&vLj%kM^WDV=N+Sz`+eRS
z|5PF=!yPq>Vipy*sDJQ3270Cl*K|)yEe(fihxWFvtt{NsLp)*`kS-~--RBndz26@k
zg3s7tjjHguR6UrvnpOO1@MJ?)HQg!EO7`7Jr;9*!Q22swLYeuPIJt9zTJ<4H#RUgx
zuoB8Iw4v>Qsi8*yo_73QW#iIdyCyc=r%Y-|FpS%zVCKDtP?D?ahKe`jyeB=hXFmd+
zfctS%7}dsz+dQ`a3|%NgYr<_aHZ4-0O(G+fTwOrw6gd>2+Oa_#WtCduR|Kq1N4!wz
zCG=G%`ZhIt_@=6_6~g1;58uIjq(wz5P^Jb4&&ZVu6e1>2Tc@_A4#V#rIoxv7FOM+F
zl9bw1pu&gOVwR={z98u~a9d0qEYLcFcEjWOz@MN>H3z05CjTx3NlDK?W<kPsQeBT7
zMNB_dxB}{LzR>tkSA9i%8Y$g;T^I)?8=1s{tK`~cCrzP^0C#*msbBg$R0<Zq=h(>k
zghY1U|4Z=e?Rxt6N<1a>%=|_JQV_le0hkZqF=0BpIA<rsB)ank%*Pq_J|U}3n=}T8
zd_!0glUYP?M=Sm|dUPK1zs<JstJoc_E>{rB7nmA(npQyR9`z7fTmXsi+?Jn(lbmn8
zJbf}GaOm{4Wc4N6If@qN{R#-5yQ=en1#q%2cD<0@v~O*{iWvN!eP7+wHgxd7vpds_
z;-z4AK`xwgg6>VVnLrWoQ@84`&mu?z-yEy~)TNCB@fHAS7eJ-l*xa7I6@XfIZ44^Q
z6YqgdpB@sBHv#-FfY_q~Xfu2v(6>4EdN%&B{=z(kn#r0NKZx^L0{^Bm`G++NWlj^|
z1RWVcc|-3$w^L8Fv2A^U(bs_Ad}(86(__hZ!T&FJ@x|{aW1@==FO+4IAXu4DKuZ+<
zm+Rc+I>`DFKx+E|ec1Em`T2QrOWY)aGP@J7sd+n%6K1y3DXh5()(QV`KdeGgpb67>
zmt(QdFNIod7n?EB(ZAm?n1D5WqHBBEJZFEQLI)DrHu#1oV`pJ$1)>Zw>^3{s^-mMN
z^kk=C4m&eFs&%Int`CT{Mgg}q5N>h)O#jshST-)(JOJvl%YG^o-NThmzevFj0~{=U
z?{n@Op~Kb9S02E-4+seex_(Om*9p+n0{*(3pTvG$yByqZB86?~JFtnLm>`AR9N{uS
z#t5g(`rJ`<VKoa%KhtF_xx@+Zt>5v5&&F$n2Wm|Qf+t@_(%a_ZsovvS_a!Yk6cAjm
zZKM)3@;39yDE_44Mcp;Q$uW!^<X2Ej8kd((oiL|AU0z->62cVJm`V1?wMtJBybMi9
z<paLE_19DHbmt@*>3@vDSxI+lD-qCuaryz!VD7uS9wuf;zM)^f{Ckm-fC&jTBwM3%
z(gmjW8OgZqn+D<NL#DHBoP9i!G$WH-qT=kr!?0%wpOuoqZ3oTwA7ubYGZyZE>S#mK
zH(4UD4iT-Df=W^~-;=UkUr0KQjV(q`U+d3cG{Z{RLt07*qJ{9#0@KL~VEW{s1<8L8
z;bU%{l*5L=7>IEG9yV0^wIPndfmlW*BS9DTRWxumuYk}%2Qe_xwUb(O!<;8a86WR<
zAajMEG%qPr8YxE*k8pb1V%SDxXd&CD1@qN$(!|H&R83RP?lhSLV-0KCzZJ{+#r?ba
z*G^qSOw4N6iH<h+<9j&v_n09zcF};g5Ev^Yz2;6Wc0H(JK0N-kcJBl|gowUBtezgm
zH5D)aX)Q7DFJQ#hDhR&5luf8da1d+b>tWFOm?p1y@@rjBnY`DI|1$K8q?UTjmt_}N
z$yiQ$#q3fy&5RIvPP?AB8jEp?;U?ZS?qkMOi{C#HjVe9A9W*WX*H&>&f4wFF)nNa;
zN=*Z?Fbj|G6&YV>-=-Hp#{*>cWpI+ABLeRmkOJ15#H2ClQgO5rJN$_%BE~ryAb$a+
z4{+z2uOzyDORqMFH#_5WjAu-KX4{g(7l-v#{xQuW5L=}erINBft?dlN-WJa>0G$@g
zR+4cB7eprmXA8Z6;;V}jOFnP>WyC|>STgc6-Im7p3#avUe3kr+cm#tB$48HKbG1t8
zR|B7mRyeO2WmK31b=6u5#<6&yktEky(tJV(V^1F5K74h7{W2`Hq+Wn*9)Twk>7FO}
zeO>^VpH^C5Y;>$XV30?B25t}@jtgSI?OyfuzATS>3+c@5`XzYW=rP-+tiC8{5txQJ
zg>b#Ra$tTE6yEY4<nv@T-MxJ=S+gw1@vZp#7sz^_=e(etXa+t-;LdsO2IONtXFeK<
zgaQ&4E-HL^VTl7=5n;;NrL0PfD7J6L=Md%SVFbUn6;J%K_&}|ZHfie(f$_9o;o*t!
zL&-C`<o!Y&^ScLjF^<X-dbvJ`)Dy?nR{&8`tI8=n-#8ZI$o2xhwKGQ?=TAwnex}H8
zGpXPuuC@kx*sLvN9ulaUk_LiWClKJ6S6U<Wjmg@E>`$7&C#HHX2-q9;Ow0aSKhKr^
zRg+g=8i%i}6fjj6u(O)GgKpQ7$kvP@0B~&o2-sOR4c>)ts5Odb-DR9|@!`c;Uya5s
z(pZ{;A(C3(Ln;<8d3i|E%NH*T8PHe0Pa=x%;^TV=m86r|h8BQKlux{*aLoH1SK8UR
zUCGWENl}X=f|y&uQnJ-5d7=CZh((mO8~??_M@Z|8{oRnDtdMEqU0x$hAYcM4VlQ8@
zprDBT0JWO@1ucs?JfeCGd_nyl3n4Skw0zeuYVg<{QUzp^82Od$63>8!Mlny|8Ekdh
zIrO8vkCkAlv{>^H2!zQTglPG$vBSdZTaks+6(?BsodhTT-d#x-vXZH;o;3PvIW<#S
zDl><^2oqbGEnBY;AK%?$)nauouP1$1Ii;x;a;QBnP1D>Ep)j-|)#(=a>v)F<zGW2m
z%apLiIgx#&=fwgGkiQ^`A`*u>+G^aOeSMH&AAC27#jSHm1}F~EhCCJN(rMEcZ~v5C
z)bt;@3uHMtICQ@)te(^vEn<dtIo?RGfUC@$nWuFqekvY8?9QZb`zo?4^RJzTAB|YW
zSbtcf46{HhKGXW?e=xMnxoZ+yic-7o#Ji;rYgx{~tLFfuZQcmkYhhb@u;x}~teoRR
zGanIDWgE`2s}pa8z(v)6+sXtJr1ZM8VAy`<o&^Nhh1ZYWxu&IE{FOOGdwK;n)Y{g-
znO$01vS<q!_GH74z<22VxgZboXu)L+aCDE|U@>pnhg)2}>1i-t8DWBlMxCiL5Na@w
z1HKfgp0@m^ww2%F8gyNDf?c&tZzRd(f=u-{+pWxdZ))SZ<~W8q-q?Gxk>l;Qs!fXW
z4Q6%tIc@A-rvhud>nMAlf+Pyn`DMaK@xfhKLC|DbY^lvSDJ4RjvZQQ75bn2c7NrIC
zcIk5=s+FG+3Jk0noW`z4!k_7wF31<ZFhMN6JWg9qSQZBv6!eaqJ$X&)lN5i3?^_Mf
zA15>;mjo;ZJyPuNWtwx5&Ya#KK^fx00uE$(5n~!Hdr2)LF@Jf3h6u@2)M2~#2Ey;1
z#Pup`q{OqT9H3#a###Oie3$ItFI_2dcbwCVg-`XLnC24Yt2DXMWwe40E*7|ZXgK>B
zQ&|j%AL-d_Zniu=`<Z+vvabsTHN_PBJpLv-fI`#+cle5O?uZ_7+@u*+`$E?zMA|w+
zsW0LNA@6Jultg<yqj<1CON&Lhp9asDE6O67NSy`_)^OPHD1E^JqKgZ*kt;Af)eL=M
zcH*KHX+KMG$a<HS8fI0ER7lY<^b@5L+ENxv4aoZ;Gj^9f*^KybQsh9EloT8t?Rnu@
zbm|&9U_UD@3D}@g;`xVhSl9w;F?w(<Z4$P)9uh`2W6flNfL#00ao|3rPSY7RZQ~tc
zHjDr=v)Wa#2@06V$f?hT)3&UShUMa#`QX&K)$eHt$^5cCMfL3Xy0p}^L;Z(-Oey0v
z<8L-v@a#h>h=&q+Lg8N1c%6b=XV#dZT1YD6At^E0Ioh>z&{G-*sd3<cES+^c9^T)^
zu`xN*%}m|8`^MOG4b$Ciy6ekyGo6#u-QC^8bl22$^Blj|v;Su|bI$q1b-k~>zV5N~
zOMm|H(1|5_#Pv_Ky~krrQ-*GahT$%eowfsxT*l9OncbV4=p6DwuB7t2?aj8%wdQ9&
zaQIDp9JjpcFf?&U;JE;5Mn!DVdk;mPm*jL{KVE1n0sAy~4*M3HCI=1+1Fblb6E^oV
z2YY$=7Q~Oy9&}A@W%}j0n*T!+gv;Q$!j=kr&qZq{6wt4v)NZV;eZiuDIf%u5F&$3H
zN3beiFhrDi*WR>lI{<twnu8D&mNe&od-;RLoxsnHAp7U@<mbVYNU<FD_~qrV6`t`w
ze)&$7A$(hQ<cw_uZ5SiJ3pkoDpNtpC5P9S1esFkPy8$N39UpA<vPdG6E6;O@^Yn($
zUh-4crz9QylsxMtM?Qh`48)JP{4q4VO){0?Okvc!)C+^hJ1~BakxTq;=mDNF_%!Zf
zya`ZM%uPyBrtqqzVKVGZQXExK)+x|^6^ZIMY(YmDsKS8q@!fqqzfkKKgAIr(Y!Rjx
z_cP71yj#HHUCvePXvWUl%rbDHQ!|8o8PbbI#8W@xBxZ)*JB&UWcL^I0#^=}ZrfN7;
z<2g8eva_D9WUtDt{8WsZhZ)M&{|hef!xtnS+OMzuBgn<*Hz+;JKQLgzUZ`_~LPA1D
z%;iUXFHrtEiMDCPl_zW!!y8YdbuJPfFxX+n<+93&=K`3zt*@kE3+t}|z3kwH+26*!
z7L^8Uyg$nr?TBN<A|lcJ6$uM|P9P~Lvxk%5xwy+LrR1oc$KT&W(@OWFz)CcS`7}9E
zXXEUW$ja&@ixYboo(BEc_PTj>r?mMbdP!(Kc~mxI`9QxYHX=jhAa>#NR~?1hBneTo
zf=c86N*VU_9KPIzLUlR9{PGg%vL#|%@hJ)D<i4o;O=S!Sn+ViKR*11Dli0c5__hoT
z-hFcK%c?PLWvcKGUzwJ#yj<#^RF>L6qaNb9syIs^O8qp&N6m^<tUPLA^lvhHo38@8
z;PwG4$Mau~v?0LjnOpJL(CK@qizFwSb?i^1;urjr3z~I6sBus`p|z5QmoIYMm^=~N
zR5fTdRo*F)3YNhh_=QjcAC^V^?w-lh!`^f!?e()##E&r4G0ihi(R>`^cth**Q>Q5i
zPVB5CreAz9$-+nVZR;=~vQ3@!g|98O8}qBptt1cOaFVSKp-r`{m5mvk52)9abSKOj
zmq8fn9=l?aB6?9z)LN<ne-PR8Z+htYVYS*x(pdgO`}21DR#f|~tQqVBdemg|zBp2w
z_M}f@<31a<_QV{{19PG@fRbxo()FKQ*&X?T1}Azv`bw=2zi-nH<xdl8JPe+sfV0;;
zB?$vc;?0tiHSy<fv?J2o#&mPI{dTEv@l5yzhzxp2V;1iM;+N*<&0(irW&=iS`T*M#
z6Xp3GPuDXBY<~c$Ibt?t#^Ob+=5|n8{~}(g|F&xK?OPO>&(`l$b>%yC=2mbm&&qZ^
zAl)3ZeeV)+PHtrK#rdd%Ejk=20)q*naKXaCW~F6b5U&eY-BQ!=FQuf$+WSZGZT2zC
z=g-{`?-=lCDJSi!^SUuLn_tst^VoSi)_S>R)Dv|dyE%1^1UU)r)~27K4!Oaa;+uyE
zP2EYOmc#$X6P?5bsi`z40WdtU3sbY@9<2ZoyF^N|IrjF%3k9IyuQmB|<h}kXLGKT%
zUCB_#o_pJG@E!&;9|jG_cMV@PBg6&biK~Cup8QkCooC4q+N#P^;QP<7QOf8D`2ai2
zskgIa%j}3}jFG<J-w@U&0ZpKO(ZC03<&hA_@X1gezK{K0v)gPJp}m+M1MURWp>`{V
z*@z^Wm&S?{N!4OMqfB?H0n%Ap8T)|hU^~>lwhI@2$Sr%|T^<rRiC65|uEDA)kVwfY
zY@Fme!rvPG?&;Mj%xcUDJO>g)r4rY9j$;Z9gvB^~XZV|c{t)XYP_4R~X71ggJ<UE}
zot*CaJeAv>?iWakQ`8`}xtG1wW@%xx?M<2Q_8@M%MR?w?e|qI}(N}*nu#zM~?@s!4
zyJLouPRmCDhd~u3*IG$g`KPMYCnfZtS5pkox8UISV92>S691%-wGxOy3!gA@A4&J9
zZ)|UuXB+uL98u^Pu8f42>6MVatGo7O5D<a+Bg9WxGUeX|3Yjng5re~QhBsH%98s*%
ztCkQVlW8nkRurTa6ZI5Kb5Z8ufpb5DO2pOxB81<C6K_W_7M9kblCS%G(W&ZWz5WFi
z3$HKHcru@~plhNy@QL8z6=F~Od-Iex<wH<5D8)1`fj7mzk$N13<7aC4YjO1~n3}P|
zBFs{p;a<3M!Yh+nWLmrv&HS&l|HjghY@!T^nkcZ1%)yFltb}#1)&7<((|a37$NjXo
z1J%420L5<?pGQ|W+8>3UKjoj_V!=9@n#iSdk{w4LySX&ON(`ZQd9(o-x(qcSP|jv4
z8~m4BA3(B)ixONbu9`E=7|f#!L%vII_4~G_m_ed=CKNCyEofV>O%DZgi~W2hS)Y$o
zCWbyh#LAYXR0y&g<M0n#zj<VO@A9weySOB_Utd3w0*tCqKLFrrbZ@C3cbGTbtgZRF
zta>G`J}$3rk8-bW4aL(R?GCj@dr;!CX)hA!AAl+|jBCwmeCybA@C6}wyxa`+ys0Q%
z!I^Ld`2`>&d!Nggu7g_=6XbT@UXcRTCR?<D>k?;Sd?uFCRvBJ4J_`!PQoOJer;B#o
zjXZN%TI5N*$>TH>PhTWY059$DdJ|6bhp=}!20ZY-%Yx&vZ{vTvUAr_ZzJawlW;IOf
zNdR|3E%g>y#qWUZ!F6E!u^||EkzCh5&8}{cJ$3sGSaErj@VY_ub_HBrsNpRUJMl|J
z^nKCIHBPeXB8#qa&A-n@LL$HLQ8SaoiT18%>nU0!khX*d#~bA8^UF!ZP1uVh2lE_k
z#&=qxNnwd1<uO*&Wi$BOSaTpZOaI`j{+<fyN^0=Wz-j3CGE-|kRV|#lO(Dv+hk<@C
z+*$zRha-XAug8{6NhKHe#cf}%xMMaVAzQ6nL8H2|aXyquUZ@#1mNR!DCYSJ}JM%!E
z%5&$=$@{MTcJ*{h_l9buPt<XV$1~@P#wJCFrsHoA)p|8UWTo|Ib0?~8<fz;yz5J=K
z-Sl~lDUti`!m6{5_46g_E<-h2Kc94p@h=lqt5Oi#!fEx9g~IurM&){ivgDCeHis~G
zd=D^=2VI|>U-6?4)EZiT>lmR#71xAbZxl3maO~<nx;&%L@;u24-C^Azs?<_Na>gJH
zx<avwiLF0mQ+^U>wTop*AC>pl`2mM9WD(d6^DB;liWz07sl#>ezAr0&`m5zNry_C<
z!lk|&*0b-a?uQ9yaW&#p3u99tF{>J(JwQGoabj$RG7=GCC>RwM#Y*^J7sQ(^*JPU|
z=Jrl|5<oMptP8D|CE<q=XSP@6x0caD$2>M53<=`>AeHxElguXu%-@eeJk2??=XVgm
zZnud5tl|APQ;E-5^2oF403LGOM|N2qdly#|9$X6#0|bvM#WMt0nv)P@9;~c6V-q6<
zJh9GurR)1+Y7AHO1IqIyMvFZ*&Ft&HY$%GX^eHVi!k>sXQ@oSA_$1-WJTr)kEyd_%
z#IZ%8m%o(`kB&~~p6_&bX?)IT7K{EcHIu&=D?10W{TF=L3ti`Lars#@5Qu|Q5e*ir
z?_pNTSXMk%$BeoAkUmYf>gcomS4Z@pLm5ta`%A_FB>wAC7;_ttXTd&iWj<h~{f}$R
z7<`!j;HxtAK3pDK)EcLHq!#9t{;O#YfoZnVVHrV-fwV_+vqMtOr^I)gO#vM03ANA1
zE4s<UWzLnFviD%Ye6xVWfAec{M&Kz6FqSb&IbW14ljAgkr$_)W)(Aw;%q{=mjMW42
zuP39<wwZlyD|fcR5D@kc0SD_lS~AbkA2t^*u?EN*&~HCWM6O}5M3W%-5Qb^>-pLw0
z1$tba>MO#qyj>^}waFgI5MHkO(AY7lIu}B@9Yjl8f8+bWf7w0i;>o+%b|ZS&IgZi8
zIU)scc`~+_s<2;N<&{SCUwzv0Qx@mi!=S<F<CWP%MPPbe`)l8De*<w^dNNzpE&`_Y
zEv*__>8GZ{7Ai<kM-j_PoH>q6xM2`C?FwIAKY{X64upQzxDzsU6Nr{eEA_80tEFoe
z$8M@SHCU0JE^q#RkLI@5a{Tmi{8?P*e(J=IH8ezf@FVYE1JN`SE9*nO8vRDCn*U@R
zUkz{jL768NP%D3yNN?bOJ^CW{GbicRqW9@EjeB)|*H&lqP2gq8TEEbxU0-6x#_GfD
z$-g|IJ3i#zY57&kAxsN~>&QqznK4c>t8uznu)B`nZghS`@VUus`y61%m(ITQ2i3|f
zWy5pN=x42o%eO?1gBT-P<>jxf?g36~ExoiP&_kX}wUW|LVN2}TT2Hk9oSugX$p!ZP
zW~V-nn&kKO6ztR;T!w-L+h{&u?zy*m|FY_<5BJ4Ca_;RmR*%^|NPtD&pXk%bR>Gc?
z)pT~!GaawipH;p|pEL{R3I&`)R>HTpTQowF*aTpE<qFEEhO6Ib8wjo~4^z?)yXBsF
zl^1S`=g#il%Nrp!DY$ZvBTWbe*}`f6LwP4)?CI2hO%$nM1sXg4SJAI22SmtW<QX`O
z$WQ(rG!?Fw&B}$6=l%ofU}Gqa&FOeeg#|V-5indKc5v?k9uNNBH~Fe5HkhYWH8#%w
zodt7-1<QRueqLs`r>&EEo8^s{Aj;D)-L0sny(o<<Sv4lw&yz2B*~LhVsUca8|KPqC
z7g@lc#b#sqbb{R0pj%J5lzc$0`gb<W>;M8byf?YM{e>R7g<hL2T1rm#tgNgMbUwMX
z-)ooYwvQ#YKNzQ>P>+Iz93v5(PyzTaaH+pMJ7_bte};$2D5Elcgl~4ZF*P61att$|
z<fI=jF(WsvR=4z;x_pD(S3|Vvd&L{SE4I^S`H52I>ORVD{ZGy$V(BN<Bz>aIY_7iL
z2D_YPt=5OLVb5)PT{i~ij((#<;vvfq#2D}BCnDTO+<b=8TmwXZw*0KhSP1$my3J5r
zYs|)&#K;gZ4AFvNMwZXlr-_)Cg7&%g5mAWcSTi6={>YM*QJ%psXQAu+9GVhaH0cSG
z+gc)9OGn~4`}Nh?mT`0{^3(|nhiDYXXVRKiM>scnSo-0xhQU>+HTl1~BVk~jHaN(7
zjanI!rnq+BQ~h1xrsFr95c`-P^ov*Aw`eDwdsXiwz~h2RtFx!GU<_-A<y{V8cxQ1(
z_>gr;TyRn1*@OHOv>OHQW~2Rn?zdCO9DELgK~3W%%t0E*aIoVa)~}<H)R6`|z_BwN
zT$PlqGaaTVgHc2ows5Z3v~|gk)pnkxSIXgLd?PLNh#U}gGuM76WPiURSOl?kieR>1
zGKf^jC>@1WFDVSAq7H3*9&3K+Yrl7S7#V+p&v+bg@Vu&i&e-NNZPQX(j6PA8;ymir
zv-AkkVWyCr&Od*{ljDC|tMc<PHALY6*?fOTd%ABllhE>6ofz{bt-3_T5*q4F%o(nw
zccfAbuRi<DB}Y9aD+|qJxk%madHkHy$!Q@C%0L&<pySlLG8M4%M)i2`89mVJ%yZP#
z`(gSezkptqX}xt9SX79KNo({zgo1H*s|fL%`!JdqD<NI9MdO1e<rp>%62n)5-+5Ey
z3XbI3n-Lu&+-he_V_0nS2ZYCjVRR(ogYapVv6~8^w;p&_&gG&>na7uoK01F;VO?X}
z1km5*Wr-w7!uo4R!xo?<!{Ak3Rm`4&DdHK3kEsIQFgUH`xnn0M!(R-)?1<0z<$Y4X
zfZh9?+o#&N9z>9X47XJCLhOV_thH7(28{|D>Rz~!;v%5s`2JGMfv}s}E7Hun-vzc~
zr^W4e#Y>G21@J9!gj<3`teZSL%Py<ib3TuAx5=5e1TVQ6rb74g38#mIABQsmmDO1X
z-pZwws+q9g`Bg8i(w(VWS|87x6`OpR<(4+E)RjJyr)eZfSUgO!GJRVWjZr}CE*R@+
zxYkWy>TBpj-eu3cMSU4?f=bf$M6<n{t4|bAb#<&gNZMC{D5I=S_#KHjMo3_io&>S4
z_`inph`Dqn!Xq?ew6Qln4K2S0P{uutX)XOJ`ZWDID~kxIS;C=20;7E@QIoa<gGPhh
zLG?iY3f^X%?13y<5$)fU=AW_6Wt0^-d2LSwhB(X-N)>+Kll*l0RIY+E+qkwoi$tdr
zu#z1U5)uJRHbpy2FN2Mhhs61_Xi|Ogq6!V6r1m<AXYV?e0VRc}aeku0uakM~m0Fkr
zGG6c3vOW*#Ee1O@n(LF(<Ad@_mvidw+&9$tZl6Q*+4H;UgDwe6(JImZTP1?jWJe0{
z*XwMJKW?OqKSn*v|7KaY5c)LF2V-Xlf`%NY%|3pe|5NqVn8cmy?Hh(!X^@Ted0|bt
z=W&zQ=iy4A&B>!rKc?z;0@>}w@Zs_UV*6wBGZy;EDg~WoUt4|&wEz0}G*jr>L(B6R
zH#Jq$tT`Kk+wwG8f0OzA_^mQY{^mOKQPx8B{LW>?h)py4N<@R|9k+h6iiX4KMgYu6
zD}bOBH2fkiySj+z%th+vSU2+`o~o2Zp2dgL)O?<YcNr7jW|-xh;y|{({*euVGdsBU
zKw~*~B~z>fntn$ZQPqzO0<y-YbcInvv%hBp>+DCLXD$nS<y0vgy(R9hu+4r0nNKq@
z3Gvx*`-L=cBAdkN{XIxg`Om)-23yu8K=8ncr{FFq_vm$&<A3Qa2HIcd(Ds%|JnF5J
zU;@h;bd$~i<Iojq!b8lKEc0h|72th9L|O!U;5eX-!eai@S)*N(y%1TZljluH`&Mr&
z6Usqpj96n1QD<oG#nC=WCq99`k8}IsgOY_ZZm3uPvPT8a!3y%gHlbW*$n@q=BQlXi
z_zS;q$JFP_^tS%_YPD`8Ln}v1w>}~se^hF{+m^>l*20(+9f+FFN4ecV#<Dt7qMGGx
zCD}cz%~C_>$-~%*?eo*))jT+|Yn`v!{sl1qyO@?|2l+Cgli$1!G#_@pJyD<EroE&t
z|8^R6X*(wYW0O?A1h-gTAc$mpZ6>9m(egNkr{tKtB~<)rM@mh#vTgAE0hOIjh(2A*
zX!<v2aRNS#SHgIKOY`Z^VyeeZ*5e;B38Oz8cW#P(F3Z^nBQ!ScCljT1vG?Co-0VDt
z_p?J(WCQ3kGPphuWw+8u+;PeUaFnQaW_0n2DyS19%{B(c+im6btNW{9garIn{wy^e
z1uZO;G#s^!R#cAM989@K&+pmyn>5KJnq!9Lj8V7|Bf^#3Z$M_QuR>nla}+U{RZKsp
zJ(Kb;^)p7!DALMq$qC1_&A8<2|A}vHJteJF;#X9n$wJdvR)<SM@Bs^>R{_RpcL$_X
zm7wrZ!jz&8(WMpHefKV|tVd1rkNdRpxHH%AFd}Oc;lg5${`H2Su;}-YFLH;Xhxj3^
zwXMSEBLk_-itCuD^Ih-L_KoxB_UHc`dHTNMBwriZi73x~RhGo;T@!Fpj!6o3W$CgM
zUy=rIN?z#~zMQH+?qlMAwe@7SK4t2~;p;opnK%3G1K+$(9f(H~_aNjBYtms=85Uf-
zbI&vFYfn$)<6aw7b<SKB%AG#X%+ErHN7_5(4^tpB;U>7;^T_h|Y+NFadAocJ9&86^
zVJ&4yqy09WWNh{i69<P8gKRZ<JYNy2z|v)s4M^7s*_*5*;uvJ^PJI7UQsZZ<JV+CT
z8JfQl9AN@hK&?{&XQa(k?c0D4=EVPuD?zVsO2@faRHc}UP;gk2%l^I6U;IGM$dy?t
z<h{q<B^wyGI~ewWah+NjF7Pt5rH<e1aJLVlkUirww_adA3b!-zAQIJ+dD~&-OQ_}B
z)Gp>s(QA`Lu^>$0T|U^fE#3+fF3vf<#4f!^d?kSyYv=Crprdd!1O^9|kF8z%n}Fn;
z^K|wu9vQDZtg)gOtrgb9$|h45mYBTas5XguqI#UDQrl|luJ%8!)X|&S`nFyrd4ZV}
zRu`oZZQP>!Z~BL?At(*^5vw;5b&LO1*nP-IbiirGRA<O*s$8^ve0gnNRJKmGK(_GE
zBmeOFMR^3i@lw(+8K=Dlo#jt|%U@2C;8kT*Wpy=(KYe)lB?r~@W}^3gBCx%z0Xnxv
zvzn>;tTSh{kYp^F{m0k62Lv9Fi7}i1wwl4v)btG)zA-i5@w@3fGx#?jd(b~8{I;rp
zCCVK^r!8pYXh=NK*KmBjB>*Mb?7DekJ-$fBooz0n1C{BE(FMr7a&jn4s^4wCe2P<0
zWY*)Xb#b+=dOZ9==oPc%$g=B6`2xS|nHq}L5^Jt$Q94Ie8xi8T*AcCnkgW*G#N_|<
z*RO&<l0(2SBs+yi7KG#V8uYHOhh|?~1Y#V}(Uhy?Qj1~KIE>9t;L;95ijk)|a_W_3
zWh+{~bWy7jsg>~{hyX_s1K^_3&;2%zbGOyoNID8j#Q$dXUx?pxQmrc<<y}#uf&%6n
zT-MN!v}uaMnQzc=Fi`DZv7AS|%Cb;w{TY?75-HP}KdVJ_<#xSdy>h!*Ug5FN2RxKr
z9>Z3w#%#R$Y{?S`XpOF?$dBthxkjy@#NYLHI_K9Q;9WP~ZqP3^KSpx{Hu|6DiJV_u
zuPll~y$MN6iXfG<F;ui=fyS)exvaCmj*-SOrlv8@|Azf|#3HgZdm^&^Uv>LHsyMnD
z*mz}n1(LrjpIvgh)Z8j2*i`$u!|r{r`+V(rrTYAaNo!#{(Pul+ZM{!f@Ht3#-PGq4
z&+Tefd@rhRsm&v`*J98jzFMu6b{MV{Q>MG>ZQv`h98w5CE5aRYH|@XQI2Eg`ZT_T;
zZOx6nODqRpPf~;a=rhhU|2R*iU->q_#9u#?QqoSvQQvN}IDwTIZ6Fr)#B=zqq#l-l
zqW0Xy{VR@)nNGg1PXx<NkE$f5p0|hB)zgk|H`MA$zb*3fU)+1GxF0aaLU!1No(Rgc
zTl*}_TXAK+g{5_GQp8d|{nziE=Jq#q5}r8VoZpg*uGvMM=k>p#<F@;1?Fz`DT7ogi
z?tPI9)?{HzJ1R-ea<P`Fqi!v2=?pv|=S;x+!P#!Ld>_R^I0XCO(plN3lLsec{3Dft
zA(0v5$R@o{J5|l^kp8VMEp36x+#FEy$PiiphCqHD5Gnv)SY3uoYzj8DoMPOI%CEuW
znyI|~iSI>Go~c|Z8pHJZOlDAonUF!>39z(%d@=!v$M3BDYVL{f`ed~qzx(CrvIV~U
z9Wd!eT_}HCC_lAqzp~V7u>DW>HePpqXEYT==jej_FTTACgjS84acltWUa5+7QnyX*
zsUC|N3WQ0g<&U2Y?D>o9E5PLXUVFSts5oahN!@Z$HZHf;nEk`d@0`hBjLvsYtV|7v
zU}=|OZb+lNO~%mIcts#L{bPAq8^Qm7T18OaZVEkZ0@X`|s;(E%?UhfTT!VcapCQeK
zcb_3_GbB<WGT#fIJbe#N&s>YD>kD#JzH@XUr?T;++E5j;SchO2<kJ>vhy$#ZC`Msr
zPvM87WHD}2^>VtJ$o<QpH37HEz(29Su)mRq$ZKwM<m-<{7UdRM{3s2s_(eT~6_9}E
zx(g`|y|fUm)Y$An(;@0h)jZU08*z2P!xJw#h;9pr=5j`EvGLQ$Rp`qKH~)M`_h&aK
z2R>^(^@ks8DI=<iKCpiPv)CRXCbnVDA8CrWt50vA_Z^<uH51D#b(oW6B6*YL)Jo?*
zQ*TafTY7E<MA<*in+C^Dc#q<X=f1i((0)A`oTz&n=6Pkgx@OB}T%GtjsvaU@+1mv9
z722^h<Af&nGn`rqvuG&G@%2yz9o^e}OKO}yGSyz@L1E)sMjAUUw%c)5QQY{npS(EM
zua3Nu81Q>iF71)%rHMt;T57kRFMaO&|4?R2D1qEq&6bwR;l5d1=Jp54mmC<b1s-{I
z<=hR#lkPCvBX_fMgY00+F#K?jAJv?*ys;d4T1+DAaujH`q*6q8c!nw1lbOO=yQzNE
zyU?-BZ}oApc>>```X6EXZ8<1I;DGmnb8T%6=ETcvGLce-hQ{^8AEUms(6TV=b)M%y
zJ~uq}$u1N&yHC?XP&Zm2lSOyrc6s%dq{Rv7{qFpQE%ok&9<RPl9$XKegR-<nMeVwK
z*@%|u-Md+w^PqERvH4#%<B#{(Ht|b2fA}ySq!+|LFWZNgNi;9Mm5FpA{WTapjyXYV
zuFnb^2&2Lcu6K~U0@Bb-N5Z!8&y+vz-^!t$wE8>=tYwc{I4Y`Xdqe~J;rZ>jPsYpf
znq@IS$LePVRMiat<5&Z7Uvu?V8)S{7X0rQAvdz@h2WnuCs+GeiU9+gQ_^--$txQL@
zBOYIJxEYWmhrwiBt$glbAnHM4@t-#ke%xOj)eU^!QMDXR=Xbd&c)rN2bGSVE=9BvJ
z<qW|KmZ<BU%<O#{cH|~q8RT0HY}4M4cgZz3wmbkASLeG0{-C<@=ypDu#@F(Xvq*<e
zYt%bK&kl#YupIt^oQ^vXvj$4#Z-!d2lpv!!&7t&-p=$O-JZVcNK246M-sQvjID6*!
zZ94cFgDU})*7&ISAY~QkCtoJ#dMx-gEAA)v?4+h}9*N8`Cqk50I3aJ}=PQQRa%~O0
zn@<_=C<#J_w3=As#LH)m6w=Eo{Qk{^U~D@SLnxjv@TeA}FKH+{B~MFCEGs>0DZ6lv
zUk$fiVe|Qj0<2JoBDj9ymCgq0EOa*#s;Yc7j?gTpEha7%ki>q&n2GK!zRg#_ME^It
zXK})^R)~~!VcBG}i*(;rq7b_ltR*X)B5aZ*vGH@h#E26`sS4-3Mm{o3<9=sO$H4ja
z<h_05A7Kr8KqhHSYf)oiRw?-#TUCW!OQV`XH>M^-_V1E=rRRzZm{urTVUu37;A^u9
zct?1)>D8v4ABtyC{EbSH(iHtmL1$`LX?Q-QT;_d9)ppvbyF1OiMCp(q!EdGWQ{e)E
zh|&&u1>^qid37aKP!cSnyxJYcwCzhvMvDIa&2vA=$k^`;f8%eBfGrapZ4^_(7>X&Y
zKpw``t<Ux#+_S3hW%vD2SdT@5pIF}~%dbWR8?{3%H~{89br$JyN0&<#ZE2UI#?{A+
zO?2+@^pdVGDU*5>(ceDHJISk>uC!dw@{W70T)UAHZ@x1@lA^qs>c&dph{UHsqIliN
zq9xea(G`M~mYP{m-SE8_yA9*8{M$SFL>6ucMIq!3jr^nu+qXPDV$s)f121v$S9pRE
zBUI_PT&8jiU@PyI@IL;d(&r4xIe1tad4?QH#GJyJVh#%&(un$KMONZS#0^)!9Ga|1
z5y><Q{}0S1mfmB&+UCz}r)-~b4e}=;@8<&DTVXugbZIjrZ~{Eld^g=1+T>Gqd(4OQ
zSRkaIYpB@?V~j7vDEuH9;fyVDXLdEBrz>(z@T|@-Z56&|Y$-<Fg6=h=@xzIY?%POh
zPl>BWh7Ks(kOaB3_Yu<u(4Y8vj~{of?CfHI<q-3;4D(Wd=1@U08%p+JK3nUACX1oz
z^kG^`?QF&2_(fVIydy598iwDcQgumI`hN8YVasN>@$SQ6{`=4CwyXP&LmkdM7Upb>
z7fEvZJc9T(u(1MUCv}zT_lUz^=FbhEkHAzOj3>L!p%%3og;ljDUcgB03Z+JbE7NI}
zYSj_SUaIY{Y-t$*F0z|rm-b=yai~%TJ^#CJ^Qq>-LFqHqN6x9!=E<{hH1bt7HJ^)T
z4D|JXzP<&e@3v`b#&b2JB?>j{i;ohl0D^zSIya9F#?cI4m8VBkqeiAzS9grt9HWy7
zGAzzbD>(M!RGsY9-<+%AE@`Y|V$U@z77}B~Ba%n<1m>AbwXZ(I`hpB{DOPx55;WKH
zbDO#q>j+D()V$WQ9Py|I&s(+^cMwQhr30Y?kUSo{jnDf|w-m{hn%;lu7WWaS1wJAU
z)Kap?_8`U%a-^o9HkcwRQUohXIKXSfQu#5&<0~1pNFims%3%8)pI?N^+0HV7)|CcP
z@eRxQK#Ad&kK`B4;1ZJ|Uac6Q6`hMDB0;QN6DrV8K4cFqj4l&5sS@d=!4bxMhbsQL
zn^g)?Yk$o{aK<@kHbj(An&k5`O=t9;3-Q$?f>O^Et}{{OXFja|0+XTZDSoGH`R`&R
z$S2$)cX<#w-10->!tIDF1&Ax!n)~H^B)DY=sA&~qH+cMtPFUv5tK$eAb7Xh7)HQ6<
z^RQ_49TPz6GOCAmpFiw@%gIJ_Z1pblhHHEX1~WwbXDv+s`rn|;-EyJlXWPGe16U+v
z;{7*~;(3bV?w1=Hh|zAolCL$Sf1C*u)yUyTA$+$JG5Ux&*mZ`)q}vqgD4=m}?U<5N
zBgUMe2mGUp^(V0wk|x*yH=$GmmSf_@AD9lWMwP|%KlWa#`P@Vg?~Xg1T?}^l&TqX(
z+0OrhOs@B{`Xe#6;D2rzeqW5@f@_}iB=E)K$UV^zks}-M(OEPoyuJBxj2)9%)@}tO
z57v2@L?Bhrf*lBSmHzjVRP;8%o8@i_b{Z+#Y)bO*oT;Jiw@JG@zp^Kp1bY*&8*XI3
z{5qM#N4rNPH?Q3ptnxJAvj~p+PDyW)lg+b3_An)sHIriMa~<vf_$hR04;|qU5Cycx
z<U1#Gk(w1`#(&Ja(GN?sY7))X`GLNQ;Po)IY_SUC`)bBKjZ&4i<BPNu<LwfX2!Xfv
zP2KAi*o|rk1Aaf`3LJWDU77?uwBjxP8@~u-d)zhe8}!o}HkhC9w5iv$iFQ67b$7YE
z_bQs98y+6MX&axK!bCwq`F(K7mT>^atvr$409+y^hQkBrnhA`eZ8yElU=%uhc6NpV
zUk)!f?SC-shui$$VnXRA4+A4(yo`SF3|;xj#>NL}X=y0Xhy)8|wVp6Kr0(*96Knu-
zWZXy(4i2<HIGj>}T>Nz;qsI|{`dfYcgpl{Rxw+;V)D$g?T8UomnWlQzV<}3At@#4*
zjIMQERP;dm*etGb;!I2)>TMgr?z2gFz0#QuAJBFJraTzg|NLq`l0yD#uhjqr^DT}>
zDXxcyhjIT_>+gFwXBwsqf4iveneTcWT|YaZG8;`>xr@qbwAhRbfK>ek$;mu8`*oZ&
zF@pudPnpgsTG^Sn-=1&mWwEKZ2oOVCzR+^2CrliW7~se3SpO}c4#6s<mNSgal?cOs
zdxsk2MR>`Vaj<SSsJCQ_O+%?5L)iaSD@vpEpP@yK1_ux@)a-n@qzjn<6x7M4q=Gm&
zs8Wqy-m9*DL_ArHFU&Q`W-3^0@$VgM6*%I>V+p&oZ1Jg<R=r2kBLt3ZoaT=LUKdib
zhEg&!HBqqP7N<;88@_k&6e^fc+U&&>8l-s|D=6gNjS*pi{xLjQe>KZJtEQ{lUp~v@
zbjKj&wl$?U!p|;Z%B-`|(<r^v;E_5#uAK!u<^^9C&Ew(@tTv7kabf=KeTbSd+ZeH(
zAj8Pp-zt)nOXCrBu*neT_xNrXiH#|bdvrZKm$9{Tt@AUJ_Ipgg2f_jSPi*sh+Nzay
zCxU$8fOOY#Ig6%qb0cm(#%boVAVIDm5|jKPP@1q)t6Gwq)?UJ))O@r)bXKzr#WDn%
zT+<l!$fA&&7o5pnH<n=hF;rG#y-~33Wb;10D*pG;)qsO6cCTZz_S>sw$aYmjgG8+c
z6(Ss|fP0S_2i>d6=H=s<(NiJgpQ%jEth7`FsDX}rB!m%t{1&pWf`S72mAFn3@6Uyv
z9iD*DhvaEr=&@j_$t#Q{T<G6z)qj_kmX_NO5%9127b_C-LN}96UA5jC%GXmJHP#=d
z8Y7iV7^IL_6S>tdGA@=T7_;;dOa7|pzT~=+rd7V%AsN@HRe8_VRH9LuKs0l4$+z;i
zrah9~X<Bd2n`_t|q)Y(ECZGY3gByOkyR#<vlazWNf3IEmCJ#rdtd8fYsNn6OJR3K6
z(7S97OKn|UQ$XRfs2xgXH<kLxZN2-FRS&wmeo&kyWo9yKR^f8UejCH}R3A=&jt`@)
z%$}XGm8yjGN0BA2E$RdTA;#(BiBG$g_2T{N*r$baS1<`>Q#l5a%AOCKWTrMYWnbas
zecqA->`=AmD8ukja4mFv|N4IW@y6yuM&>SGQy*_0+w2`&pIv3yeDaV@{V;fOBR79=
zu>);zYisM7{(*RNAC6EvxEu1A^NfFhLF{l)ARyYEzf5ew@b%&COXjw=AfzSHUAJTG
zpoYik2+5JSs@XjeI*AVpP~LHPrNl@kACOt{B{KFdKP?4HmkEEcrF}rlWmxdb82M)`
z*2~GmCT{tO&=hf#B0vNwAePr4)`j5vt7uvp+yvWV(jlk8>K_~-lluSl9Lyj)oICJ{
zV7~2dn4+S{Q7mMeO9A!`QLidH4oyaxJh+SxaVBhzx{x=$pVf=MzjvtaD#1h7_hiwD
z$@LW*Mm1!!kD*B0a7g)zS>U(q-7zY@*iu<Jc$DtN)r@5K76HC0G)9mR%`Ke6kWS40
z9p%@AU3D$KPFo&2&v?hqw36l4;e}=81U!BaV4GgA`t59Ql43^#<u_qmQgVr_y@GGi
z4Db2YwCB3VD)|<s2Jd$}zc~L*Sp4uY$=HS@C3fCO$Y+*r7K3sjYAIAW$qdVs2x>*1
z-+yJlZ*I6TG5JYV_ihYMe-@jR&icU7Wk$>)<(u=`(ApMLc8Un|@Z>bF{E_O?Jv}Uu
zdGg3bp~|dqfAEkhpNDJQDO#jrLS3(YF>gNY6CYC-qS&UdE2gFD4@mppkj2$;`U?Ig
z798PiJIr-A!5K-S58)!DP&Im!8GG4`ZA1mDwuF#M#&&bJc6hS>OoL~oV%Js!hmx71
z0OkLfU`%68T+Q{(zq6V-nm_(~5eGg$wcl&ET~A4S?&f{-$uczKN(B+f$vlw&xZH`(
zuy?}cVEPg>@q>Kjs`^iP!ey29A9G`>7DHtN;b3cBxNY0obfoGMMR3qe(efCwqYe+{
z@)Up0Sg;BS>G~P+vvVBzaE-Re)@wz*YBCQfdgNP{O1ts_=>VjldLssZ^mz&h*!hb&
z2qVB*+3g9z0>=;B5?@!!!EEJaWeA=nJ0SjjS*Zc-Fwz2pZC!QHj0G^ul>M8MW+pN3
z)j&b@NN3<@Uv5L~js_L2{rfijS7tv-Er;tCoUhN$>}k6ycJ=QFH4q#*2<<@a`LfCW
zd&vj1Px*5P@w?03Djq`WS&bDH9RRQA;NY-r38ctp-fs1GZU`>@or%hM^7R_s75LAU
zp+?O%k4Cc%C06K*9@FkS>g`hZg2M0Ke_y-Z&D?#7P#fD7yl%X{5Ih+_&Q;NWdQd;3
zG=U-$SRK7TM-359>Efxw*BKw2j9cXo528)CfG5}9qJ;m591)8kZ%%AbBq!I)2~qgN
zh>xbo^Pc22CKXPZCPW3EmQQ}L8r5IS4Y@xaQJ?1zk>2-w!8rzmBZZ58*vPL8OpJ=9
zGrze01`y8JMMr+PCJry?P{5}}*5iv;P-bvOhS$TEYLLkJNc4lljO}~y<OkZ8zigR2
z{`$Wf-F)IyZB|U3Jfg%m!zGaHf7)ulp(P&(+OiKm!#4kF)SQA4?&wD*N@+K=a*TVl
zE#4zJWcC&MXf5E`;aV{gC$DW>Hj}+bHg$0Muh=OU;(70AGjNXjeeFR>-r*fJakxlA
z3m@ZLrvvl{rf3*_WQPQ?cx)aoZ?t9S%6q+bvRJb(n7OfE>772>?u@A0E@dBCAzz+4
zfLXjv>!7P+Fg=<efimr>XOvih<PPLJ4Qt!D>hn=%bzSRd`BV3EikM|&YxgHgWO*JV
zH$ih6Q;5=<Rl0G_4q=bo$NIT$MBxMidwt}D21E>DiT^SPvzb$>h1xl-HDf4G@a=Fd
z6r1Psw{H1SB{J}cPV6Q-E{Ol2TUJ&ptD*EXlqoT}w@Nm@4!ypHqW3$o5cX}dzJvzd
z_C7yMKXj5kz+RsSt(`m#nF>y@el4D{SDO!<Ty*EoG8DRZg)((^aDYCkfdNuc`X1$d
zB)iA-s?ezYymQ$*KVUTWwTIE7RVLhK0%g~KmshLvb7=@s{pE{x#lic|=gp(Do&46D
zVC!rL^nd_bMSzO1eW3Bt(iFxTWrQu!xr{N=5Cr+EX=u>FQ*qlYL8i$vJkKOS-2`Tu
zI_0V<{PK&9FfSv(Q`Zq&o?MmU_(81MsxCHrT+exj#Utl1f+?i;HMkTorekTe+}>Fi
zb;T;h7w)Z?Y8vH@&dpw*8P4&*7#T48;&qJQym=%3M}1K&6QK>952|V-GPrzoLyy+T
zan7d}|7Mrj2A9tw&bSU5J%-O?V;he*a}BxJUvArULj$vJ=Bo*&uGZJ+b#IaXm(+Xm
z{B-hc=9}(JlDj*trmpXXAH5yed1v@z&=wW49g9EUs3<cF3E>blPr>-k%ULgC0H>wQ
z!jm!ODi=w?t`)!Dh)hc(=-}Q}6WpZ(U#g&}a7q$ebCba#jeM?a80_aesrRu}QNK_i
zYL_Pox>>QBBZO~%wZu!jN`n{3Gu_*wm7!36lPl%RAY<*Il!Sipp&Yqi0rDC%WCEEA
znv0vn0FPp(jPQYKhE;Bb64}#6Xk|068NIR5VPK&%TKI5{dm&9nEnmrU!#JhqVxF|5
z(y?q%T<Dg#UWIHem~IgreSYzHYVpX9M`L?|oy??cek1;*ZDi=yp?74+^>g`2wfpYd
z8IyDdX|QrngSIMGz^aFiDNgV<#&F=u`KlVWbN4u=;8Wz|`+mrm<|Bpk-<6o+ArD7e
zR_J;U(%XdJhrbWQ6l;ZzP$8mtKWH!0DKw?7rMJDodu(;GE_pK{6+|<hnZJ_euMuIF
z@{S4?y^hr3{7CbmzWrK=1v|wKQ{fxe;52c1K4euCg5<U?|EqRJ8!z_aG9scYnLiPq
zD!NOiVR|2a?8o!tt@hap?eK!IKGp^IZJWBU3l#5Ed<L!xrJxq3(m_uI5yb12&Jufg
z+*3*_)?n0jo+FQ!1E?9W1Zvh$`XJWP%(DO88jLfR>bSacZoH8nkCpW&9!7S}09rt*
ze@RzfsU|0feXzv(*7>gvk4%gfGZe4^!vEKd1p@=aj|39%>+Id`Lm4=*VE*TvPkTBw
zD_t~or6Yea@t?m;-n#&da}XH)(wu-IY<YQE@iiC`d_`slkrZIo51N$IE)?d9rlwHE
zQ1GdMg#-i)81oRHf$j3NMXl?8ULudjl@)E;-lcm<S=p2shs5*`0AnjuDyRs7o2KqL
zaweK{0<8-JGjk&QH_+&0>T8ggcV?tNr?zEIhv<RNNy3Ylpuq3*1O=xAOawfx7{H`6
zTc<qH!qmfCMO77yVjq{Qj;tIkOK)r7@o|68pKF(@%qNxDO^ak~9`2V!&R%rSa_kA|
zT)96{&ho5YtE;K$>FdLizXr`<T26I%v#~8KxC37J&|v!DHGXpa<eMIyE49wdVE>o)
zO`euYKgQ`5X{3pP98~`>T#P8BOVrel2d-UWOC$w5L`*Vb*UWQ#jVC`TXWDI25DS48
zZA!-EG<4AQ$Cux9oL_oZ_Xo4iJCB(*Cxd&Po0G_~{MKzdfSavUB~_~Rzi82SkC4cB
zks^=~6-+8h#o%o1L`xYJ2pbL@6;33>ck-?q2=*TKCj=1+Vqqo@{RAw_-lwbJMHjO3
z*-ZhJH=pV-y9)~h!8Q573f;`kMipXp<aXW<iW)E$Ia$Lf_j;xtDrO}Xr*w-S+khjA
zQi*9>|6RzQ#_Ny(BszY-{*CZ~pYEwE;8R)};TK&}pC9FQZA$X!QCrm4=oy4>e$dLC
zgoxm6SqQUAhn%Rg;43m>P{XH_!&4@&>k&_*IMg`W27Hv-BRh0B?@JwFW-`#i+b`;(
zV3w8WVjUNF)+hCTmX%|6hu-(lCJ+Vq8W+@yyYLvee9Ig?!ti;Xd&YR4kDI&k(S3UB
zU;9IPms#PmxM7O6!yo~-!`q-V&X`tesH@?QMa1G?9z*O$tKoQ;9=PheMTKdUO4dNJ
zoK9duAFTf1O(#ixEnH}T4IGs@lewI#48h&&!H~N(cIP^9afAO_#(9?E8kt0RiG+|U
zcH92lDJzsKSCOuZYZ-Q<x>~bm!EpxwQv#haunaF8IS*=}1$_c$CG%R1yrQpC>Ve7f
zVdVMA!RC5=j4KuW_F5#Cn#yh&iAZRmX+J!9>9oO?9hvOuU(oP45b%Kj%ynDDkM)Z$
z%jAALV(_*5@Dmae0L;7R94eOq9RxiMa9q)vqK{_^$-jGDQd$~+E!w=y<aL<eOMk`p
zr*6BZYLcyW`Z@{TT|FUQ1S`vk3qWID6vd#~pxQK_=~ci!1>^GbS<P>i;AFCes%$Tv
zT)#w3oNo3%{4IZ?dqGBNuHNB-9rSuvfFjhebN}M9(s!w@o)N#|+@sE<<M^AZY`Gmw
zgqS}~hAR-6&Nl4d&YR~qPnu0wae*&IAHaQ*x1yV?mJ)a8XJMIPk`o|O(SDgE-pa(s
z*b&W|k7&Y8(C8eNkQLKo2SnT=gcC@LJCOc0p0y@~ejHI0w2yS(C6Hg~#_2oiqcQ}1
zmUx5yhQcBDKRupo7uyb3p4`7~Biuwu42Jj!ly)+u8acc|@<n06u0M(CYUyldXa=i3
zAjs}EG}8HM=aV!4A={Sun9cw_JIe)nBE$CxNOl$jUn5$7uFXD*FX<edH6&J(e1=A%
zs%AX%v_H>}pP@Y8GTT;EcI$ebN-s6r#fb|i)|N89?fT?yCsEfXm(r<K>X$?BxWmVh
z;&4bzYJ0NcmAksOls62$UJ4qVPC99S&~930;`fMZyk8TPhdQM@>@wrX`$dlSGFoX&
zadhKHPmxC|<;Hx~LZ4IG5{uibVR;zk95UKt5H$^aKZO6CK=`c~<TdXQqH{eLug^Mk
zxmEi!_`V#(a$v;@f}y*Iy$)&3?g1_Jfn%ShAU^^5Y>BLL5uF!`>iRb%PD{`$cIB7}
zv^3@IS6&OH@Pt)x=5n>ml+(P5*vaL5<M>fx4P&|SRKkSRyxo`%lu|_%8MFtx6Mwh9
zDapQyJT)6Jsh7qVdx-P7I7s7ni3FY`h0?DL?R7e?ByzLK(*^R=Hudw%%c2J^f%`%d
z>K)ToT;^P<FaS)?j*lu;XTe8m*_3V(0MjqL>}b(lQ{{AbpHJ{}-<cNMAqIFH!^Lvg
z_PpKK+QGHl(FK<PCeef^3%op*y7o&!mJ4gn3kAao%Z}hNb2(WtwziG{d0}9inll+}
z3Vbp?x6IE}FWNuXi#?2&th<-Y=2o33@S@y3dLvDg5R@yw<|0G`PRAGHdc0HsSbLo<
zIgG~`6M&7NeP!R#A!w*xy;NvpnL@hx5;E*j(*#9VuGTfb9f=(H@Me+K`ZiB7y4A^F
zN0AC_fIT)iu8AYR;B#>ub($OtYa}YAhQMVAP#nuPiN&hr_E(4M;GY8xTg%0i^vL*l
zVut%w%KVf@BB^)(3HtrDkJ2N{VnFrX`dkdB4b%j@j*!rs7d^=!p-Ur&VLiH{bHsro
z9}wI>?Rg#&lNWyd`!$6TnnfU%*65yG;_~sELl`s#k{`~Q;@dU*S@w7M8r^>f$VC~_
zqwHU`B4TNaY(EOitFdhdXEUbmY|R^yeqc7p`yS((5mf*c^CzF=>3~hCB~%`#7-tJ?
z71aC1y$Bg6A7uaG_|ekhl2APZ=e8`P-weuhrWWtdf)E5}R5UnUADAsb@AR`UnpeYg
z>tmMoY`u8P!;T@>73P&)gI9xi9#{W4b4!Qm(^GksnmMa$jd;qNC9UR8ytYR?tK(xf
z{GeR1xJ|wA4xH#WNV43<?|Ob82(}%Z-C2Hn|42RV^8nAJ>jjs$)^|{TX6qH9s^gWc
z^|846%UCde_jlpofw&}w>F>f{7>>WbLvYNWv^Nlq^~cR}@W)1QluOt}6*F}2X0>}u
zn?fni+t@PF8;_GaS+e38)_oeRfQi!;#-Pt}SY7SZ5nqt0e{_J3xS7=dNd6`#+@|Eu
z5=w)!Q}V$$G$Jg^r9wVOefKUXV+Td|?nC=k;rPFv0O{|-FmicGhS;KK;`j)^cf<ZO
zKv@o_Ww6mgQjfI8PkgejzS}RSunv`iHw=aT0XU2KYEwJyAaHbMbNasvtBvU7_i9vE
zRQ!v?Q@-LxPi&v7(y?>au>ory@Y~3icC#<L+b%cfct2IAbrhaR5}J8Mm`Yk#U)s1|
zmkm#yuwEsoaq0eZtG-S1%gBN{w}+*_B2XwCyd6K660NqTlBY*v{8hcy|0J|nB>O+P
z!HVa(EU2i0H<JJv&7zht^QMR9#>402WYUZSlv#`p3hig{<rUS{N_Qc$<ALC<qKlW}
zMcH*DebGv^ot#%jd6gJR>H0i6z4#}K%iBU{MSBvpH`hoqq5RCIk7tLEM&m(`0xi>0
zbyh0?n9W0+5U!Sz4~j8|<0W1gFceUBN7Lw=T0wbvM4DG0ZRW>nM^^uz;r5B_?wttY
z%PHX?e4&eauA}I#y(noxef9wn4%`7q14knUAejS}-OQuCg{Xo5EsW{qv}(U*rE^VJ
z<QFhM2g*s1h|Yp)M_d35T0_fZ1Qkj{y_cT3bO`t1yR1HcvFDZKV~hwMFG*xF@T3&T
zrKZ0WBF94(gr)GPIHgS9ql#Gkpj&7X^W)<6!(N^IOh*-mY^XK$dgH5-9a}WqYB7ak
z1S$M4j?VZs+JG_g(GH2A#zyKQ74lO^OS2Z4;p^`Axr(J9F!neK)Js9KY~z^icmG(X
zqz}pz(}7y%Y{{E{@ndk*bA=7|#BD9T5>Hwh2S3*HgcH~KF&)K+dF~AsG&=9om5uRu
zwhTVswm)0ilrL{wrj}2K1%1b7Lo!1WBUeCFGHMq?(zE{#Lsd*^-nG!Gyz=lQ&A`aj
z=y1>Vyo%R8aFqHDkJKAE+fHKUT0v0{QX-wr@?QGGnHKw@OGq~i6@3<`7||58H`p(R
z!_99Bb6Y}D9K$rwyM#bA>tA6?uHInNo0hXYX1nd%sEO&=leR}(yNnnkR*eD(5>C6=
z<2+&dG+twK^I%#IVz|+{?LcAtTe8OqCqA2iEI>+1`qYG$(RPM?E*Dq+Ji|TzM{5^P
z@Y3Z=rT>Y0uUKYpcs8r`w~!Q7;hijwuTb?G-GI;Xq|_^okHdw{*AA)UfBUsG@carH
zMx#YQ$$%dv4lGFPO>aTxqECGG6WuE~KE8TnloRv=s`vn`ce+DVvv_2X^Mw8jVbxOI
z1LyvC$qxfJw`6eh#!?~yjgGj<SU&ul>2kXz7H17?{mj0C#~I(L_92_Ub?~MX!}*T~
z48Ccpsq~9g&c0;GtU=bhCh97E@$+0e927>nmKyK%OuM>RjKE97z2fT6Ff8TM1RlBb
z-%9d!z=RJ2TKk(gp<66KBO3?$2PuawbLn{V&frV&8eahQ0e-=LprPvP>VhjSF817)
z*4+RUS|))9zxNPz9*;Dxl|i;7=QcYHc+1spKmL2u19g+<2<zl1F^aaU+VWE%iI2^v
z{nm0SR+v!Je{wHKb~_-><MbChicF~Q96H5e3}jC6@$qcC4p361;u%}wmoc?Msmg61
zcU$M(9pB4iICMV>x5t92uBFAWTXOixuQiC_9#IHll{_Q=ZQjyc^JLT{?mZ&4rPg{p
zejxcc^c<C*o}Q|kpD@C8GlCn8`+RgsLat3+Htt=^qA-!IQN`R`&q-2t$0m{cP~SEi
zE#rUppMv}TIR#Y`1Y^YTV<}s~<RbE7&`99v)5T%2e2m!$B^DJkp)|{+WD4hAwg}&p
zBBmYmN5py^%+Ob?TnNA*>uV~-H5N`9M+nJralCNj#o6D320O=5@%^JQIb-(uB0q$L
z?o|}+Vyowf<V<dvanQ=)M$`By{QC1p+V$v|w0vciqL6v&lU;PbR>KuG-gw5(<rKp|
zLX^W|4~yf^v_6+Zy3^>dKBVKoVe#sx)riHcliLPJ&zLJZ6n|}@Yb%(b7S6rbYIgi@
zxzyu<`T4y5BcgNN^*;f02=-5hS_GOzxvX~}<2Hk44@w@{zg4(%+Rf>RuXrv%oEI&H
zjnLDTB^&y_e>hCVfdTc6dfym?Ivl<%;kV4tVA0xFX-q2EH|VpCcZKc4nLbCF4LF%s
zo^f)`ayjaG^GP|1=NO}J>yM#EKi4txU>B6N+C5Xa2qIW0d&Rk-xd#*PNHbk`cvZ$S
zcFuKgpkc)>HxuH5QW@s|s0t-{iJcqww12YC{bVAea11rH&t>Hi`w&3MPJpF8yj7O&
z+x(ogI*5Az*dz7c_6PfXg$ysnmDxROd`gt6Q&^fYY@`yz`5a0!90u#?VeR7L^3?t9
zf^>*Zl@YO0g}gv6-GGO9>c~0j*Ks7hbX0wJK0G#V56N{;^@$<3^YCxlyhs+&*|@Pf
z4f=TG^YioV1G#$Ag2l#ZKUm9a{a3q3tQvZz!rJ(OpnkF69U}p4_Hf-X(BT4(;7k2@
zdwV;@f3o`6?9+eUb}jHyTV1YmBvtpbl_J&zcK=WAiUZa>H>%HnbzwzOmPiX&f$Cx`
z1dB#W&uq27${E<4w*NI+d&v8%-t$dop1NlHU*3bXb5Qf?xUR#uUF0#Z--8A@Os?Ql
zs}9)GL~4|ZrQ`#SYo<xHy59hpPZNhYRkJ&l0l=Tbcx6>p*RIgB8~C7ZPq(|nOJ7Pk
zK?Mow+(ZnMX@=37_2UcFExWMv_Ck}=g-oq^4<C(qa8&-RtT6RcA6{0jn);C73AHA=
z`;@9}uR-7yvjWKR2-Tem>ogt5(hO?6Vp}Q6kuY=Z-z9enZ<5<3sdW?l*O8}lt#$|@
zPL30UM@u(Nkuil7uqZic3uA8M=L}YDs1g^r+MsRoz3z_I#{`O&T7jtPlAI;vcz2R*
zsUNS{qE78!38PZ+|DqP_4h$C|`M=h_`Yoz1Tw6*hkyK(tnjz$+gaHZZl9Z4ELAtwJ
zfuR|?Q@RGEOGzmy0qG7wq(MURtohFQ6Fz?18<@FfuKldF?s)E}WL>f;&XlBdoa_hd
z5pOsl-!bPe?!j5A`wO}+SHJ+Kh2<DaPD5i;a+3I%whM+qC4%3cQF>0efntsOODlTI
zR~+}9UYYWjvQZW?G%HZF`FHgsP>AUExRk-l@L}?+YN)9aLEkm*Un`EGOfgNag^qp8
z^P;(M)QXn1?{1H3jjuV*c9W@*G$r{z4<lj*=kA;N##i0PU6_*O9oiXPn$6saQ@9Zo
z+9se8L;8H>18m6B%k8o739zuZ?BPE;m#Yvt$5ygXJv@pDz-EFmFe*uQb%u@k=QyF{
zS_1kVVUp?}p)1GVAt%T#7wiy-sgW{6k|aLfeJ=f`e(_t@J4#lMelS-=2ErUmbN@jo
zE+NXDrd``eL>|vgMT@-v3c*>_Cw_C0IUKTc>z+s)BBEM)NpX3@C81VAIXy)?Za-RE
z=Zc9^PV0sbMJlX8js4^CV}Ic}e7%`^AeOwX3v0#oF^J3XIo?n}hs({(Ku;w0?qAtg
z>j7}3QjDe6K(P1U;OE)oNV_L<8}|pP;i@lRCYaAY+i5ow#2`QxrLlTGu!Uo-uHSk+
zzMfgHO)CXr=|Y=d2JnXhX#ZV#6YaqRLW|kdHmDacG2r^t@XjqQ)hSaKR!Mv&Qia`e
zHVWPh8UbA5bYMuhVcY9uGf-#AbZ0$J3^)F6PcUDZ^NFEmjNRNQP%k1QBY%M^p&}@H
z9toq35d!~2*v*UAoel#ben&%mqmGw<t<V#gQtWJOf=ra{i&5<HOh#sAh1eBm{-1Le
zmuS$PO@aX@f*6WdDOLfJQ1Uh)@POpGhQwk+2RQbzfsH`udBSso@!g?Po{$HRXcG;P
z4Wc&B;Jc&kSv}Vz4aUrh6k-lm+3_2A%>(@q$1NIHu59Ml#$CfthhX`6<Vb-FOZ{><
zIeu2_Mw$c-LV!)yESARA{)bTPOIT%r2AMQH;zO5ZqVqsJAqoSv{lHH$fk(Vu8ddwv
zi1^6SCc39k;K@i;Y;=Ag-b9aFQsu`8x+g+qYjFuY_;LkKAvsz{Wn4%Td9&Bqg|IY+
zQSvDMPS|SnQEYqLdF<pcZ7YeZ$~#j-&!6*=*R=r`H2l=cQ`7ruS?A3G4oR(#w7stO
z?e4FJ5lRsvRI?GT@O$`1BEh9}buj}XUEgv~f<IEese;wL)Uuu=t9z-mpzoIzbi-8-
zkLI)RwRW+~zWw3idzo^_m82NN&GqGAjhi0Nq~N2ci1E#t-5f=aWI2)pr}s7C2ZULp
zCP`8<%^J74=M3X63#Xu8w*Mbef=Dr^7CAaHm8qAM6x9JuGUR77v-2PMnW;hx6ao{u
zGT>?+*liqP$5CY2QZG|74>>%M^fF`?h@6-mpOWx{1P#Qw8cq(Ka=$ZzggRkEItVv`
z>E|Dc5>II<{0SSp>=A;GVfU;E0~IcNTYJ=Br@+qK8uo8yl;l6I%Zpu(ZT&vI`r@|E
zaxZ^1b<C0ww}8DDJCTvq#nJ!MM)tjUI$W>4dL8k@>bIX=A%qg2OkV<uRU!WH;>aWq
zDr{`VE9&cz_Q|RNA1RwS+s{p~Yr8Ye^5X4bToD_1HtXQe0jdX5K3Yh;z5d6jZZh!-
z531)YL7d2$Z#NI8vmD$s+TYuA0@FdJrl&z%Ab=MMW>nzxVt99ISUd~zs5-F#KZnj=
z27)A@iIa^^C-S1!ZO7|$GH^@;G%6PD!6E6lt1x<i=FQCwuT|cd<smp^9V>yJEwnOp
zQfsGaLMTD(xHXc+E8;f?^K9AWf#0)1v4CDGkig0I(DwP8rp87&u%qqnCU#6W3>GUR
zYB@P>`YKARv({^$A8&_-SOti-hHrTp@MhP;QGa=Br)RQ9w_UF6#e&X!+x3P!7FCOO
zv8Lt3FV!z^#DFMt4@?+JIzsgA#Z{!_X_?Fm&7W%=kj=0K2lK1SkH^yCDXiVgVzyi^
z^tKKsV_}v1@y;hI>P5Cd6V4OO1x@I<=Gaa&i6Y-LCbLV_IPUYfW=u#TN6F8pUz}v)
z$#tuQT{>K1p?Ta)$nGw_QkVD}gCW5A)>OOopB6h)pR&JLW*x(%&os_g)i`zvbY+4^
z9hS)`@s2|0`e3keVpjUq&!bQt)6%}&6z79gqUE`LXA5GUUbFpg_(rJie}k&WhPU4X
zWQI?u&b?hXw=M=9{K60X?tG#G>75M1I5qIqQfOGF*hs^Jq!49%p;dokg?_(YRgAHo
z?0Z6l<e)qYsc8E;I7^_U-+Uo*68YjfVr~3aq`vPOUZwaI#0BM5cmFANl$?ocdK@+h
z!{BV&-G{9DP?5$O>BYE<%D}qAN{+apf-Gi4mxQ_txspp@Lb0%jpA`l@=J<j|ECG92
z@1J!ygPReWT_Q#zAFNGkE|+0yC8e#m`udwoSy#4VOFaX8>5{dax#r|9-*Mu2zLRRq
z7U-6fs4O%h5}{hA=cHDy$tDGK6L`Y7u|}^2{=6w~f>ERlu@Of2Vc}!trPfxMlHou$
z#lyBgJR|-w@~6*Gmd00g%t}xfZhLZgOWkcGpIl@sKF8uCEc4rAjtaS>RzCfsnMIl4
zSuYCpQ2U_5f5oKQPH8Z6R;ZR=+38&h0)dE+n_E~A8HcBF(R^ho<XlG~uj1GWfslbs
zW|Z)*<N3V#{AXY%Gn;pDb4vm<-vDbhFfM54T`RsXV=-TAm7Z+gG8JkMuy}_jH|62s
zsuUWf0JQ+*qBb>!9bBPXehshc4ZF|%0EY(5n-0F~G+#hpH#2dXfNR!$qBB0d$Ag$0
zeNH9X<;<-xO2W7A-u(g%wfbMZliXXQ53bChmuM3x@00bTrq{-wo6Eo+h7RZh^~=lQ
zdg+PO8+=QUXkI%2bQ<8vbY^=WmWK<4cP`(aH7X_9<d5Z!SvszL!+ZSX2^f{i1t>Lk
zi55R->KQ4vA`P`qfUZ%rpXKL0uXzokz+77`J-W;-FRw?s68(Rx*4?h}hRu)sYYOX&
z;3!^k@lgP>9?^=RxrayL?$Ve`X(#p7KJFVUfWJQYce7-891O+mn&4+3l*2v>gF+ID
z@$elt(_Qr+L<KWhd}({wBCqz`IEzV^fRI)K!Vp50<c-r8O&np&DEBxVZ*Iktw2DQ(
zCRk*&pEP`n;|C^lE}wYcS{K8O{PD^vhR1Y^W`@@0oD#MXQh`tMIaM4Pf@ch7%ugBH
z%OgApYv%7u8r&+;vRT1(jbBYqyvXfa`8;SqT$=&4w}0eyHTGd~fj-5BGg~;OlE2CM
zM)!85Hs{yEwi)e0)35<G+5jZe=CR%ytRNq2L5a}}mA!CbyhrdG0ud3JE!b=*)!C%?
z;At-Sqg5)HURP!Zk>kN{Dfk!WslPt--HIzqe4x+36KOLl_R-KyeG*>*LKY=U=<V0F
z8fGfeD^0?p<aw2(<FuY?@n25ic57<ITQy!2ZpqjkY$NC6g;+B-Obp$T9=?YoIfma-
z=u^Wu@#I^vh`e87$8|lm2y^-_xw!gm0H^F!c>EkU0k>*xZSpCETvGm+37BJyDBoDY
zOfB`7yQ|0}Q(navvLShZcOtC`sgzm>QwYTE)tpGJn=kY>PV3dDI3xE;j{6-*qKz#0
zBaF+;?WAbRG9C{49siCC_?X`eOz~rl*=t!1d)ZcquavS@x$N&M>@U1yRC@T}%4gwg
ztzz;MHoo2sD05^LGH)@v?*szquZOZ^qjitq4ezKz=z%sEm68}Qo3C!a0BXcIV*N->
zVBDIUQVpdYup-LYQ7v)m&b{o9Fy?EZks!uKN1N2w*N=}52U53Q$f;|x{5QjWM8>M(
zxINAcxGArpdq7YG7CNw&tCLq(SF6Lh<=KP)NYCCvJ7UDHW=EaD(F%}Yh^h1d5PbZw
z>gySNmb%d`^HYX>r#5@ivB+E5iZNh?G4ny@L)jk&h#!Tkq8n?U4O|lj=JtosMk`X-
zrWbv=z1#u0)Ig<D!uf=wd?vc*R@0ZcC1f$0{p<SQi5z??sPhAG*6^mYM=+!hvw+os
zIRV6vrS;fI<e8b}LD?;eBJ|DM>)urFR@_jfX8S~~H~sChX`DBMEc~@dQz5C&qxNYF
zpmY}pg!JTJ>S^nGz+6{URiOzFG&EXl^%$s$y1DTKsWdL8M6TZ2Q}B(q$p^sx$p;mX
z`Kty&NO#6E;mfZZJ4pyN86dpBN=%lnZU#m}>1UR@DG*LPbGgg~_iKijsf(IkDEoX0
zEGOkPE@0%r?PBUur+cb~q*tORBn?&zYcTb<Co<FwONf%2$9VMYAtg&-4+4j`s&lH7
zUg0j*Ai0_ZNkwFQ43+stttvG^mI^g16_eX@MSPQHQkVa^Hz5w&aD-HcuKB!t@Z}2)
zuO{honu*soW|C~iNGad-?cIg=(Jr43PpP?8_3+(M^IcMk;SKGLt+cT-FZ9a|bS{hZ
z!N*7w5fKq~m7s|EyiJco)2Ga$U*L!c!%jERf)aJ7ou7~Ap{&L5P^Q2P2Rlc{ABBti
zjz>}{^6_B?jY@Z)bv!RmTYtX~er+;!P<cF6?r<14Sls09o<@!5Hp;5Rr@<)i79vge
z!cxn$OG_&Y;U213uZY?t3B`iQF-A!;#pASp6kJ}VTk6blAX5s`t0MlB$r&CdjdW!$
z9^uR<)(i}@g^+6cVou9gT58gyZbBp#$3++?*vgiqw}jwZ{R?zaof|@?Y*Dg9Ch{I0
z+nH}o+jL9P5zY!Vi4oX33J?r`dVX1J+!qC8YQ^|V$pdC-%o^9h-?`iy3NOK#DDp4B
zime_oz=8;j|NQ{tN)W4~TEGJl_lv4{G?RU6M>tEM9{x0oCh+~xdHbDkukT^>u>F3+
z3n9@H<FsV&uLglHTe($f2vz5G-i`&?J%Wo1wZ^G#^;4UCkD39PynM1NeQJMyzox!k
zhFqqiqGAQ6V)fS6j_mU0Hg3wPP=j81!kPFPe*G@Lqrya;X)C8*)0Zzmmt6sFh3ln*
z8(Kgq|7S6Md-&)sR>#1r5S!wlmr$86ero9xFpl2ZPGR;ou-aj#E8ob+2WJR6ID2}^
zw=QlDu!3GmQlQfS)u}Kv`!u!{C_<nxsX?6aIw&UQ*g85d@9w(uEW0ahb_!7FX;=7(
z20nmEpu<hg$dIy8Mrm`TEr8t`5(+566x58Bqr-fu06>pHW=2KNU3L$;xxcxgGtq%h
zAa~Z5ma3)E+Qz5dGh1iGOhue<W;%0HwGvtBL?L#wtva!D%3`zNmz<IUgi=OaOyCs<
zy%0>JE+d&jZ{|qr=l72N@d7`9Pxwo;{q73eJ+=@}9mYb`TVND6g8x3Z;G2`*f;Gi*
zxl}wEAy7Z1XP(hTcf8?(;~Lw^*GY*ar%FXHm#KSTvymWwSZ3w+kMT-1llDDO+el-<
za|vdfqfwx&^_wv}c~v{9inqYx=ewl8T-aVa>M3rTXd`K&h6!&owyGW(n1XDnK6kV>
zLZ)-z-48XDBtuHNdk_(DB<MQ<IyfVgp0SVFS%32nYe{2r$yCPmS-{Dt=b!IpHFPE(
zS3Gx@F}H%ZH`8~bUtQ&;kwu3gd?DfQr9U~Udp2EM{46c=@)Zi@b%Xrzv`+$cb{_eU
zK-H@$Tdr_(753ir+tP@@&HG55wsr}N(oTo0YvnwV4aa~SBZpV~ka!%%FiF`5KE7N3
zwIfqleUH;vO9-U2@Q({{wWiA`54DirOmodb)IP*%mcL-H$Di9M5dH+uVx1(C5Mv;x
z&6=<fvMLiH!a5e|Iu&)c>yM@z;0-pB@8#*_eicl@XmZ#a(ZjzPSo&u0sRou2g#L>#
zbx>knSUexOK?wZKk$xtWRcvNUf1*%kDDx-h&vMOo;j#EAmqgY>zb6S>@>#2oohS-Q
zStb)6hjG!W@ZbNKO8IH%amO={*{M{X5HO1(f;GoC-lsfdIQqGOpg`%Q%}*PxW6qxQ
zr%`Mh+>$d7%MXLR&t@%3FDP!0|7#z)?7gph+l}EqVVi~X=oOpi=iLP5=Ro8+^7NJd
zC@lIUKT)`=tz&HEc)J;v@g4&Q7dP+&Sh;}`^xM7Ftfm_bGDK_gfPakzj4_~x3l{BN
zz($2;AGBG?KdlMea)7Bgjg7N&bL=qyUjZw%+*gCGM*p2g<jfb}eV3f|LBm^THh8L-
z#bHqHh(Fh$`D}8<>3OYJ8L%}EJtd!F>t00{B|@dRdwg8OikWwKYSuNx!T%(K(!_JW
zc?*fF!8v$oA0Z8O{#ZU~E%a#Cl3#I(f7C@=TU+thZe4CqLsgQ7ncj?jFHmd8ETuus
zeKZA*q?MX`OSGX{e0nT3U_2b37upWGRs9H~?I5M&uxYjK>p3#0g{uE*0PJEO01?Jh
zRZUv|3hJt3aKBszhPMi*Bz(iGxoKdb@Etw4y8ZC3GQ2C+QG_9y;dMF#Axv@=_kpSw
zNf+xAHiOol<0K{A{MoN=p|;#yq?kkBcC}wJwaALgZxaz#9>mj=NIvQ9sC@G_t_gEg
zCl&JA;M!4I;x#35o5*K33ag=yXui}algj;>7|WosX8gLPLZ3_1PiI&&Y;^4!!uoj5
zHQmjc_5#VGDizo*C7Vy*{kE10_zsw87`w9!r0LjG7xLd7eD>Ys{OaFPEOtW0l_}m=
zBKZD?(nY@bzv*kkfZKoxrOR)U3A4>8hB#VRj6WjRYbhlf>d$>AJK|2mtv<(a(`vL_
z_$R?pfyGQnUT9%>LfF{F2h`#NSGk?w0is5S_JKcpI~y8gNU)$t3)3>LQ<5PsMg-eN
zh8wm2{U>&fomCbir$Q@d>J_o0Gr+p^Q8zKRkVhdo$g%&8Jg>4ATcR`W1JSZ(QwV`T
zAJ4Xa^w-r`uJnF&+&<Rf7gouuFbYR(S?d~uuFrzf_oVW-?Vd4+bT6E8@NaE8U9)45
z<B`>b*$tiyBvh%9RY!Lwge0>HRXfKIk->ylrOQW-uOZW1V#D(eUsN)Y#CgQ3<RRk;
znq+tbreaE%q%W#Jr>*UnySR%tq=WUwFQ+)ipL)5Rd)jimWD<nE9y#JtvbP15-y1KD
zuIO*xJ8ePtc^6IS(y6Qk{(7E=5LeTsi5j!iWT)6WQfch0<La9GPrvB{588a6Vmr?d
z`-~X|h%U@0dWg3rGrv(kI@@yWF!Ux1wZ^SbgO8DRqgv%!VC`0V)!5Y3zp$V$WMjV7
zz&U6pg9X@QKs7M($Q7&RfEH8`4|6?;?l|v!|Eg+w9i0JoF(H04p~1u>IxXmDS3H`4
z<2G@EIMat0mWw{nrV<7^%=HBico}ZOW)NiuJam}&4Py-^&+px$<1j!Je-B8y3+El{
z3bcNi&k_vl`ZBkZS!5-2H?wp4U8vE&UA0idC?KFcGof~yEZGU+L5s<0*+4gf$^A0l
za(R?puXy&WRfkTgCd^}B|KBA5f}mW1aoZDwz$l&kT9?EpfX7676p3P6WP6CWr5miH
z^<kACmi#5tI|T`pno#rid#4Sx#mTFCr{3sq)*Lkc@pJfBoo0W?`YPm|FUAs0`H)Cq
z?SqIygqr8Cy!v}&#Klyy7BKi5UCY|kY(}c6?fZ`?L-6L}qFj_yg<c_;#zYqOg0|?7
zA%-8Sm~^=uStNrraW&YuE2T_%1>Dc^UU$gH%sf?^3XOkfM;ZUAk4NK`l>FliJFaYl
zjcv0W#J|Ut<s}zHlGesq#KH9WtMpf)FALU#?eAxhe5uLzN_@p4%FU(NS}~MkY+;e|
zW@}^uZd={f+-te;YiIeAJ>a4*$0U&m*X0KJbiZ0Gr>iqT%&jq`iuytYYuRq4L^yl?
zo4OXHuMmEE$iej;3nkUX-^Ubko7*?C#PZ<lLsGvmJ%!(==ND%lZQ-twwlihUvCX{-
zq)XoanA=WK2hMv#cq5gW3{fV%UPyUNh)x_5)3jc;(xqz*S0Uo>J69t0_&8jdKq>N`
zzxj})j?LB7Mg3gs7iX8E?tOR<f-ng79#<Bq@tgb@&;8h}9hv*2cK%r$=11NN`}5}U
z8J1E)d0DnS0t+gghuM1cxKiUoz=GMk5v;4i#rvA_eN>V55miQ1_oC=^F9QKe$iD6L
z@hXh2FpTfyomNXG<vY(DiOygcG!eS;&ipuEe{r;53biPGS9G|O<Erey&8nJRFlI?C
z$7W?+qD|xK^KK1>!5?iNq&O+ZCef%fBlb-pP`QYbNKU`m8-eJNIdea!LIxg--o8N+
zfVS&I7wrk##k!q#9HM(bH3FwTZrT`KLqk@=?s%=*+zujny$m7(qUZTw1jW>yt&dNu
zZ*=GT4lvc%OQfM^9j1cEHJT|T1q^PwjGl7$jBAy;xjDCPL7<6Z@|Zya0)Ycac_BbQ
zf@c{BL4a5^yt|zKhfZ6FXU@1!(C2r(d!;uu5U4cxW7fUgY_Dhw3w)<QFE0EQz7G5A
zEs7rGl;ZC(k+A~o4&MbqGBmnX==*m?p>p#i5KNXrm@*t#UuDuh&2Ei~`z(rP;z0l=
zcqd?l1-9&O8#BpT?9XUv0XAxGRw$nMG{2WD)Rvs}FfbQ8ed5GTn_0h9v;QO$(Ea+;
zg<$7B*&HVQsTXp-_!E^ZQ`d0-Bfiiz+P@V*IRO|~n53F)rdh7lvS>f>T`)r#oDMY<
z!3mk9cL_f;+j9-gpbRploI5{7cs*LKZ*Cp{t`qToi>%2C^B3jMl<aJ6KiAY8x7`lg
zMrVL~r;VSM1~1P*c}}Iv#n2{oXV#{pZ%=6VQUZ4&SgLL@uHx?A4o@}d4qI!B51;x^
zHJ)Sdf)_(y|KA@Q9voD=|L?{B-WRWbf#v@He))gbjQ_u{Pw%jpB4iRAJaY~1fgc4K
L73oUJ_ksTd>Q76y

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_012.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_012.png
new file mode 100644
index 0000000000000000000000000000000000000000..9d13d4f85a3b9cd3864a0e5af8b067fbab05d895
GIT binary patch
literal 126316
zcmY&<bx<5n)Gf~9i%STwxCM6zu8T`>{Q<$<-F<O)4-njfyF>8cu8R{SxIA9ftFPYo
z$4t-E^c|_`d;6SwZbzvo$zY(8pu)hwV93czs=>g(xqTc50gyiSKWr47KN@~lDQ#DE
zhwrW)CeG$CiYBg(whpegR{v4Bn>)K$IoNZu@Un0+Q&_sXI=Tq3vfBMm2Nnls3)U%7
ziKCC3pg79vxWK?*nf%`kwp^sl3I-MiMov;p!!zgX)vNAc%xx>mkK?9g<Mmwr+(yg6
zk}Ggn90s8^h?-g)hd~D?1kZvi&2}uQ6JpU{ST_P<;v$QnDj^=S9CrU5NtJ10Lm$;3
z8QdX?(?18(9whCdrKPph-R0Nze67bZA7J)&;TIdbQI$k1sp;NXbKd0>z^k11J9+Eo
z>lRg2bK1z>m8TF{{Qoptr=a&T|G@t_|9@9^M)nFW{&!ETK2Op=U;jVTE6*1YQG7`I
z|F#uD*mF|3|6k9ziZB+!Jo#UiLT^CGh}g#eei`y$?9OBdj^VntHo4+iQNWeMb0%55
z3|Ce2?XzFpXTk84OBVkC+nWH95!3AV*-!CyZ!!)-zaGiEu9WQF;)2^JCl!_QgdRu9
zF0LLt1TVIFrlWA_HW#Kp`t)q=)la@3#g6d27xR*>y~J-myxRU>8|h-$UXcgZcj5Q#
zo+rBXZXaA~jS&1pU6lOCc~-}N)h!R*@xGzz7RtpX{klO?fk41JB1LJAA4<?51-o{x
zIv$D57jf+pJOmR-qCt9m3i<#;iqb*d7<PhEYB8ds6mhUv9wDArT=U)NJuytE3OzC9
z01Po>N(um^LI7#Dte(|X4$uSyu@5EYBgaTlIL7iZi-S{A0!0DGc`}SC8nO%w(ZS4$
zv+a(xc3ea;^;yihj`xT1oo}%LaZ2|@@rXWl1N27w^8ve1Z4oyD(7KO~f$^&gQ8D`u
z3RVu<<P@%XytkERcFW@WVYWPlhv)a&{QT9+4al{{TVu~-{uNfhmDcilXABUiL1;zI
z9`Is1?R2wfxA`7)@3gxtAdW~uO(42^cFr0k4k*c3(I*OOBGALLw5waBc*#LFkp&LP
zmj1Aje#rO9o(zGZ!50r)%<3GiC%E?uWOu)7WOO{Gf?-J2^UyS@>a!Fvdb23wY1ur4
zxlrfN*6Rrmb43#eq@>W}7r=!jBvCFMaY=kSH`M#YhGfdDlhyF}WTXH5&q$1%ETM?6
zi7hJ|itK+)rz7mN&dP<hs<1V$9m~LR-AOUQ!BNZ*4QE>ALtfoSLVFw<@G>26RcQN$
zCh{Kt!1^{d?XtJ1w>iC}lM(;u60l>c@BfM+{0N`Fzy8=rPi%bU&$4v^1%pe})U%mv
zur*C%AVq(9?)}N6*-a9F<uIW53tTvxoOi}jKrsLxK^8Q6lOr@;H~dJkD3GlKg>poq
z^urMItpOMSRT2!S1~na2{xvra_b#Bi;edQK`qJL^>GBk6bhs4M1hsZ(ra}%m0iraa
zG5RV|gX`~VCZpX{*4G|}VL`}naIVBaFY}GM$g38g|K0BI>6p3c^FPy#Lgqqpa*e^A
z^z^7LH4t`7J0iv57o+!y_ot7~916HGe;YP>_v!S0G!G5s6PS2%PVM;@@RIC%6~lUM
z@qU)S(+15GL&Vmk2jY}Sj%H4V;cUgdXroDCdSMO+)=|(5;@^`55G7KgX%gDLXB!<%
zyg$5)FLui~Al$EpD>*0iT(!KhD0g1I+HdH@OuAGpPEE%^DD&Ae6^P=b(dUDir^jO8
zs?nTti<^Il5dGy0cnx?};dx)lf3)g;IW_uz>4*j7`W9P9;`lZ|xU#ywyE_!a%<_dY
zRC0Mv=}kc<@E&{ToL<UGE#?=A&19mBT7jA9(X08dmY9`I!#vpWYStFQbf0V~K-AEG
z6J4K!2G2PkZazGxS83Ys^xo<Iw=qM$l{I6}#?uns#kYyA^!n(e3m0p<8G#XPk73>4
zx0%<J(6>H2L%+W`rt<^eRc^niu5J%pzS7@Rn?s@6_xg+fIpwcuOQ^Okm(rsTrx+~;
z5MMJxUEZJ2dTJ-PR8_P{jQ#vFIW6K`9QFuSEq`)PjLl<A8R#GUX(RCZX~~_1U=ryx
z4dLhDXEDxJOo5?B+QDM3ZVnLXS&2X~0ruEo)0YB*x7W61pcIH|y1_7_?Ghu%HZW>P
zk|yFrGLfQB#u5}SCHPW1Jd0FD!}JjLx3+4drrk9-X+k`DIDA|LSf_nB#cDKHLj}UJ
z=J2A?kI-->J7Z%+r|}FxEacFjlJ5xU{by6_auG#`5<K&T%Z5#OyJ=AYtE?)CJ*&h_
z=a3)_%MrrvpjUqBo+8rN-ZUCzT6t!>F|r)|GdMX0L#dX<*|$&$MAIytOyP*!X|y{J
z-;EwDpA9b?l6@mIlEQOuLj^G<{3FNjCdrVBwc4|Bs7Z*YU4kJB)&L4*e>z{17X?1S
z8}9Iu2i47ej((#wd6f=CET$RO?GZLu4t__U3c{GTlVZ@=wj~P-YH1tRSTD3`67h|J
zGr&_oKzG=pw!EYQTH44}m9AO2K&xB$ymm+#0qZYTEjmOtb;JL~{+L1Z%HteUyB6|*
z1@DCDn$q7_M$|t>;o9i8I!3SE*Ayy>b(!$_*vHAI$2ncR2zfDP+Du|ggN6Z+e_}JU
zlO#XHa$!o?Wh-aBA{WB~M4(Vr4G^kn32NL!8?{PM58ESQ6N_yAW82W9mz7J)u`E8$
z2!21kc&POWv|*j}we<2Ron`;e^hT!I#W(9i=wkJE20XC@U^vJL4<{5>DzEs}dr@kK
z_c(`gni!*?!y`=PEGSq-KtzOZd2jr{!SU+&9vGh3*fvXKmor)2_ptp{Q9V2vMJMaA
zmxb=ee~Rkwje42V-<KQZ>)~u8nKj4||0LS<mz#c{>8o{hCI#oQmup52)J(Pdo`E&L
zkph-!1LFw6?L@d;M)SbGGXuKgrmJ1+@3%ok#29VfuTn*4!m_fVh1euKSQj@7)9*af
z0WY|rPG^TjGX-cfNKL~>Im{Ep@iMp4RMLCwI(f<F#L2e7>`#ej=hbbrXcISQZg&Qb
z#CnYK-*lN=G}5hQzf@>2%HjvsYoGpRh(mBwF>Ej8EkiV6b4Lo-b~scC3uSaAguPED
zD{%@`;I9ghHboDs$D|eoD)`NI)#BmMq^z-i1#BCx<oZc>%ty{m_aSB-SYyAfyjX0@
z_VW4owICxF#_+GVk+h5cMvh(O6Tv4wi*F&W?0)t|HA5)g35-2Qu3{l~@gWVYn=e>;
zDyV^oHB>ngx?}OB7%#P~UUtNKq(TDDv+r9ykGcMN!&jhR`fW)Gzh=fav^Vsx>!)_T
zA|FxGX^d|ee~V0kRcnt`WANG!3eA=c&|}x!6Yr%e&r&Pz-B`vZ7_=?M7jORrXFN-k
zt-cKDC#rHoQ_2ozCZDx;aDqND`QDUn?Qd2S9=Fu8pV%hEdE1Mmpn|9ZU>-v@h??AZ
zzhjLm5XQpP5g3@^d7V&t1SHQ1Aft=7Bbrr7A;IA@7Ex11QwH-%hM^3YWQT;b04?h<
z?ErzYL^Vb}x&AGRz;M|Edn!giZpc`#NF$o|E*~vAd~iW$H9dl}_nv2l)8icyRQYq6
zZfzlgklz)$$n`8%U2CgsqvnpI1#U7SnYrp)n-E;I6w2c-DQ!y6MG_A|ZX~dV0>eFj
z*F0KJObTpp!ki$vvCv7fGCgG2yX;RpMLy){{G!3>K(WSm2hG;+<`i<V`MtA4hKJ}s
z$Pj8BnN>kk8@HvN!j}4Pvef{jm?@WNW*1m1Rb$jx2Iso(;b%l4z=9MC0DM2a>#rzr
z=?%<tL0;f+poZ}mR4L1yspb|U*Y<Kz5aQHDp*Rw1AvJv%04{j6wz#)&Qf>Tror-Mp
zkhRA(yDXnyrg&BoF%8TAh~e$_{R~-0PtVk(aQJS@z%>v1C-9&LEP@3W@t@-PKDI6U
zznkD<f0i=;nuIx}cSQ=){vYe%7#C@Fwo3MD7A~io#cYT=Jp0C%!n~UgB&@I50Ym3Z
zJu(iLr~|J)3P_(@7{DzP;L;!dHL-<kSuKxkKGqRR+VZL;`zxX17Dqz&u--_TREYDv
z28>at@I_$poYslrXeM+d<g;^qSiQhRfDKBJEXk(t9)9~dd@#Kv5)LuF22oKk5R_Ys
zhL0nOBA4{gysxTq35_q>JTFbZys3{D9_inuR5gB#B7X<GUu`|O<nVihFQgK9I2f34
zkty2oZ=emW^>pve@N0kl=VCTW&Q~Bv5(y>l%0*s(TP5`Tr}c&sQ5z8SvMt=cC0(}1
z0KUES)t=Zl&aSp$rEF?xV;@AqseBwK6dBoEgqNj^un$MY9s$Y#rGUeue4w8T6`D-c
zPLg=^&`Yt%1dw<qmC&J2Tvx|4b2V%7Ad05$;#AemK5|~kiY@RP6{b%FQ89||%n!#N
zRFI=_wO~km&e7Z6b4>X=R>pVh(A#H`hX7p8&`ArOae1%rs7=3EuHjJaMx!AY<paZS
z1~r>LxjwmiO^pUfRuNp}K4wakRXG<cix#_(dWQ)zr+t0ay|8{fOFhpN{f<|#cKoKG
z8RqeSg!l-U#%ic?y=<TB1(%2aOJvE9NTPtB2+2i;&`KaHc)IGL9!yz-u<5B;l8F#7
zNr7aabEX~zJF7Y?Q?KOT_i1P!LC1X3gwa}t-2-<a68<M;BEu)%Xbk_+4q)F*5{Ltu
z>hXqKyEJyy8h$8@c)5PJc4d<Z{>26X49t`ok4w0>d;R9nXgnyy*2I>T?+F;aTl_p+
z%(8JumK5w_q2#zbhpL7uovU|t{<&d`xS`2A_y)QuQ`{C^Ul1#j5yn|?8#DC<!G~I^
zj3Ihmu}DLOM5a}uLNWk+#IIuiGqzs)Tv)!gQB3$4Cu623Sy~%Vx?SMYo4dxV&F_R`
zg{YZLg}F=cE%OPLDg=MA^oItfnHi4{HS!B<z$16GSVRgScK=^C*7C~bHJ$JyY#Q_=
zl8yFr8CSLoz=pKQ@=BS<<xj=AzU{@v%GND&q$qZHAsS}c@wrvFi}&)AWck{9thB6D
zpz-o?!~M2!vVZ>D4ya)!N8iZQ&WVh{*!P_bITtf`Hlghwx01-K;ZL*}OXgK7SE5*M
zM77^QDla^@KP0(OV^Abq)oomad~F&aubF^BRE`irWYfo32MWoxi#8<3rzXOtkPH{a
zHa>P;wO#L;n5+LLnpRJ%9fVFz6toZ+*h^m(tG}&fIZC6GMvn61ZF`UWb$~VTyAMPL
z|62cQH{IcS`$9NiN8||<_c>o;EcD8Fbax$24S;IO02#W};<RnNbs;T<V^-6?gkthz
z6+cAgXRjO_N7gE{R`mZ@?S4b^rE638)4c;5iL2$M0~Lek?u(U@NOQPaaNj1kyZimJ
z9hNq~dlnt5U+Ce>6Shc;Hfj|c{k^DvexE~0+=<FQF-&1}_vG|M%H{9r)B#k;5aYc^
z?(4HzF;5{1dwFI)TA!g$diGMS^VzM=-;~trJus+O7Kp=yP&m@}4M@>QDl0et$8Sp_
zuIDx2*==|L`Bs{HPX9_i-!GL(i|0Ww;B?yMVjY#Siv7g3VtTUYSzAE~M5Y28y3-LL
z8s7w2Lq6N`PFB4S9Ek0ZO?J8Zs$759O$UGr+USvk_Q8{#eO4265C11mUew;SYx{gj
zzhDzI^=eac{+SZD@Ponf&xumf-1CzEeUdjiSZ<sS1c;KO2$#C1!!xj}0W)(L3Gl<p
zxk6OOJ%yxyiu98DPqaQBWTQ_8vSB6a)_K^(@0jGvllx^^SBg`>T?-X!R<?60>#=<&
z$v^=)W_SIy=|@!D5}26F4*)ati2V+(q~e7hleT`lP?9V}C>{$8nG#O+{cZeb0Ciw@
z91aal5g4rWYovhJm`m1bfsn7uJ4^RZFG=015!_36v5)|u(eDQW@ps~{+!!UKS<q=`
zK_AUhp#rng7;H$PpndWQcD?iQiGHget4i>!aqq<&OZ8#Vy<`6DwJUbX&;wJJT95}}
zR0>H{v6Gx?CsjvxWx<rTn7}X5IP*bIA+sB?(JsM=^|s~_c_Y7&(nFFmB~l5wbWGLu
zrmPT)YGy2#^I6h+52T3Ewp*Da-lt5esUoi!p2x)UZ^l((#&XDhH3oiRvB*p8-U>K6
zi&|RRI{Jle8fBW5!5C2*gXO#>rkE?OUYTnS-F+wZjX7-(WV+REp&D@<l%(&Kp}~IZ
z0n}3a3H20!b&b^(aW9{rkFw*MmT>q%Q^)QXlw*ltO7Rdm{BXFSK>|cxd&9l3c)uhS
z>4IC=soty6_2*!0>_QcQj<G6Um|np+6)6@cGuv$780Zrxar{aXEY_Xi`WpXsD$`P?
z5#MnA0yiA&I0;ujVv0-%7|qfngK0XJP0b`sEb@ABPi3rl#@7p%%j<#{hHytG-+eqB
z*WReDe=1imV=QC2e|#4QO<tbsCrT<9oskk9cepskzh(NOS<hC#XJR*MFQ@eu;a{ma
z)>cPBm*nAhE6wTRK~OlD4-W<Z0q-hNv08zf90Ms<ObXU4?QPi4Ebm!fF<jNSSqu>N
zh0Eh|{qQUK8;lsR17qHj9x9n}>^jayt%6esKkD36Pj61zvu|Q&kGO{|iiA!ai?rVH
z$@%n9n7796W{9<W9DK?LRm!IZruAE+p8MBGM%eL_54XZ-0jlsaC>+bvJ~9kAQi&a@
zsIV!W@-14H$zGz<llAOKp`i1Bo_o!;PL~(i%f7NWeNq3o7d00%_3Tv~{&eP#K-&FQ
zhlKBI{gXe{c8@<Dqv?B|!2kY2CQ}Y06Omh}B>jn8Hp_}u)Jxa^7|JJ-Z@|pI3k7>m
z)h1}005s4Aiv1jMt<*Iru!&p`U-T(KtgTOr_`**TA}7JpU=a^@W~`3{6766A8q_M;
z&mQW8<=Z=UHPr>YFlPuRKDRy}Y?X6&Vd9hTYqrN@X~1JgRgIz9!S76u_UIpdEO`zw
zDFvHf!4}mt<e05@2U{jLleGgSGGChX`FzqIaV=1knJEr+jZJN}3%zH4B-uMYoGf|%
zotvl<dSs^CyvHW7p%Z-(1YA^=-fUx-v4a43;?}MNMN)B+qFGnhONRRq0e4Wq-XLOE
zyC{M>Q4~)0qFkzW*r*UPHzgCgS0b(uTs9h-Spmc7sB9yi8~xGczoT>4wtyR9w0{I>
zF)kNz6{5$PwXMur{OJc$lXiM@0;n;!6qBk?5PTZ7gK3V+GFWx5!WL0Oh*zCoeC{1j
z*knSXJTWhcViH&ZZ-ky&6_{YbP&}K^H3ffn$E$70()i0YSb8q2G(7WzR9YIxU!0qV
z%P~N+-@JaDmZF=*zgWk1bguQzkzJaG3KpX=4}<Hx+w*NNZC&2ZXyAxd!fQ02u<RA0
zd<zi@NR#LoNRhWjTT07&mOQn@aubp$$|<A}WSu!Fehk^vu9t3oYV63pIsC-?@HZmQ
zg+bd&RoocjNS_;@@ilc8M8)S?#?`Kd!yw_&tBlRTk7f+MZp$~1nrOrd|LVV&{*JPi
zt<~oFMokx2op?Z;dso<DFo><HFGff_;Dm**_p>?l%2OfB&GU4U)$#t$CWr5a*@8uQ
z;`+pj0ya^3EX_XNnHLwIz9ibBFDuhdgXP>>r5|4<P{syAC1uU5!&9`jQm7^(K_Q_@
zNO0sX(D`q*lBE^qR}1gh90Xh(e#IN-ph!nwIwSrNiXYht10tn<0bqJE&@e`0Gr}>j
zuK<$-(jDw($Q73_3H3YNBB1%*`*S}Q5$&X9t0u&JMx3%aKM1#8Im`AxDH^eDb;mq;
zx_vHsZfT#3T(m4n;D7ayXFuWh6(jTaLq(<9+WPKJ@kH}m0RuJyY;Zrdm<4fhA5cz=
zCN4vP)~rjyzRNd%veKd9Zcr-D$UAt*bcEexSFQLvfvC=DYYLdb43A$XUDz^rbzLbW
zwV20)+Bmj8%QuN*d5cQe2^C+Gd2-?S{%O74d*f;(@$Awz*@9nAZNxLK3`gbx&mZFh
z{T*OzZRM1v$o~6_oCXZ0D5o?>mqnc)Smif-Tm2%Ci>;SUT-)WH8XxciA6h6*lK!1n
z*En&$DW<VahJAK(n#SY^6+PuTs8fc^f{>bRn2_L6xmq%CXztSOA^m-Cy3y<2Ezfrs
zolcpQZbA(c>rU*G87Vg1H`H7XA~#j`KP2D6i`kRU5_r2G`R%*@xhWa>k`w0ni&TzG
zgQO5`!<)@v5Z}TNZEG0cdc8aHKbg)h7D##_Ji~pqVkc^9Hakw3+M!)&Lqp&&^YhBQ
z+W|RrzHrPvKOPSI&2g2kS{q;A+P!2J+tdO3sRsT#gsQbsmPU|@QA8ynscFK7%|;lf
zKb|>sPhA+spSGxVKRl*CjAHqd;-}Kjdg$29`;Ak%GQu3Ziu=arKeN=P#v{MXZM_Sw
z`}~vlJ@Qj_dFz+D9*f+F#jvf(tfU!K#Z`v@U*U_QOZk@ywk|ozY_Fc6HIApHTL-_<
z^7r?jMZhkve|k20m7uP?4(ftehT`~*?z=?DYl2Myr!9$#Ma=7+R|Ls@foyW3UXq$!
zpUt&4?_->^$KV3)5XwaWtB+qKWaa3o;NTFF{`}Ce{r#5vwZSR;g1}Z}`S3PO9YBNK
z51Q`x)IT5P_%cmvYQQKRt>ct)jTckQ@O*9Sq2J)#1H@z>;gFN*d|B06KDV!LYVTct
zAq|@%#zYPjiwtCoq*s{Y%Pk0_fk<2nnI>TIuM3#%ZIjvD&}kb9_xYNlSm70viS2Qj
z?&s6t?YR1pDY~VFp_>;V%LI!VZL1zZHp?e`n<Odl<tnE<x&E_xCfjZxI=9cJTpCPW
zzQq52f{MeBaRy;H#<(#0ZagbE(cf9QQ2<0_L<dL$C@D-KzxZ1m$l5i+o7Y?0Bs^x!
zt?aQ)2Dy;f`$vA5druJn^c0HSOYwr}vY3=}Tw6?z9|i1Hi9YdhREW^W6d7}nU7zq8
z?R)q`Z<wdPX&YTM+HoEprfKUK7O7R^NVA8zCu>(|WS}DhDl5KGxh7KBNf)lIx8Ko7
z7k|$4x_H(xYzs|*UHq=iR3-#V<7k<eSQ=J_h#pAAdlJm%EcDe1J|zj$l}#lBa@DFV
zK>Bxw{J}~WjVYta*yA3Fx5iLxoE?mS{^a+eBP{9@n>|De<xeGn`9epKv{hk}DA@q1
zCa57^2*+BXv=F#-+62@LW4pfG+^Jj6t@Y&}Tc>OyJ&_cy`h5BBZr@bwchF3Mx~5l<
zo?gTeFyr*Vbr3_I*(DZS1)i=|llu_Nk5M@ps!iX6Pn{17we0yC+GLRBX&x&U=DIpI
z<<i~A>s4e6Ih^F!VL)|7IEc1}qJebd)ba8YE2r?XMCV_z!|5Cuq@YT*v^iuN`*|lD
zwk)OC<{C>lSb-%xDQO~sY`p`%@55zOZi$@47_u^^o6ni~1@*0gU$(a|zFcg!hRJe+
z5z!;4=9tTZ@~#uWQxQzGu=uj#lnIEkQ9K0Kq{V2zso1<gTw<g^bg&4y>T*s?&4psr
zxYp)s$-F~jL*3bottrEcd3CSCE2$oZ-42iPF>s0vy=w#B13TeP@VV`erJiv$lYIu{
zsq>|I`sTFfExYRXH~#K9^Z62H=Bm11?c&v%bP!zUOXvbP)u|`G4l!0Yl!Z|?HwhY{
z@AK+Gg!#+!Ekxw^c)dj;;~+^a{`hm77)p9{{nqR>;%W&VE!}n4*V6ou;!HdNQ@#^+
z;#;pBbrBgGr_bgl9yg2*U2i1f5P+m;?V7oeQfB^AO|<M<=44lX1f`??jyNo$S0T1Z
zC=0r#@XJ9geVJMeP*R(?$Q(;?IEHoW9)B^MHQ*I#z%b%)j$<ChOE2(HSxztTs)CXx
zXhw$Sk|m=tP-&#|y3Od2@GT<X_t+3EVO%Y6-&Au+*=hHdfw7uERbK*XA3Ij3q)STa
zs#IxMnDHlZCBwIMa$)P$!Ytyx^>#o+%I!R7ucABK!n-U!uS{XpF7E;N-6bi^t)K0O
z8`RD4b(b(90$v>3xiz5IlCPw`DRdfFDKQJ@wb?`zPimQ2W;Ca&F48QRh3g3$;TlOl
zke(VxE_~RS`0V!8R2#Ep6<vlZD^@@}jXtZR^OMcjFxV<K3g0K3X+);Vtpny%6q!Hp
zj*JK-TDHbcVX#(t&NrJFHHK~AQnJKfw4`@;RvMqUvM_e(DOG}4Gk}Atl8MTs0%XY4
zw1d&XQN&I@mjbU<BaeH^b^3ZzMKj`5;DoV0dD)V(NYPAG7FI2(9hQu^Q|8ag`cFRt
zPG@fA($w%6rt_Zx&DOXk+#Ov8zdU&BC}qHM0+}j!ZSsg~Iutfn#_2MTsrZ3+D=27`
zf1js%!U`}?87E87vBgz0@J!KCVmCAZeRrDi=hYGfx;F}P7jsg-)IK3Mr!7!Ez{^*(
z>?zA^tMaxfyxUf#W3gwA)ya2sP2U|N@zxrZ`tebGR`T3kXB=5dFJH3e`VcSFB!fz}
z>^Mm=+dQDcdY`fpV7aomcd4o64wu*0K^23y0@=O?G5H5CVRm_5TlXELE!Y5M%d8D9
zq0m#THJs{M>5d0FUb3jH(p?Cwfi~@N24#Mo8<Zu&=d)538I}Bqq-rS8G*~qG0Jnje
z2K*Cp4Odd|_n0U&>odh_LF^E~s8mqV;%hL5JfH+se<4ndD3_llJL~n?y<^(GV`6<h
z*r=Sq>I(^Qku33+jinA3j%ndcTtw=xd`+0qQQoq(5F;(R!V`RZM8w(one*l07w=*e
zukU<~DUI!QqRnchhP8qy=oLhF-VkXjo)-7zwG9bf`7T}1M_NxadE*s6iT=O3j!CAT
zXTcnQ_k23x2RIJE1|552%CoJ~c?zY*>bm8*s4r^|NCF^vwC@eHgg-WeD5>9(oz0wM
z=;(cw&ifIndK4^RkWfaflZH);z1LrA*vx`*$Qd2XO>$17Gyd-=T9|AIL=*Sm75K}v
zYO<c^N0tBD@=;w|{q)?++@I-GSDy4z1;~59b%ZT=C{aKP+-?5YlLUihTm@%>t;>1h
zu`IG|vbXtlv;A5C`Pq;56-C)VXt)Bml8Xf6P}c?F!yD+mb0wZ$g<Bjpzwd3rYrwcv
z3pnAZ$`P~%1jJZZi$k}C^WV<{rb^3rJ>RJEDFIYv?Iz=lFPZu2`eT`s1}ExI*;yTs
zVRZz{6X+hc7ti<mJA?LT1EqXEMPY_l-%zzI=In$YxVzS|CM_^y9~sj%L4C*4EBaBc
z(tq@}EUIFmYA9;bmS6v-uS6q<&^T-?o8DZ|#tt1&t*8q=m!c~*VV#sfRiYd)-ny6m
zKKt>Qt#6v1uIlN(oun_X+jE{l9!49Xp4Tr{hlkLLFRah~!GTR(vBynqZ<7=2o!+t3
zfd{4hA3*x#%B{wrL#|i{y4|Pe`$+hnpYyt1#`j)No(nz_8aKIy_zIcb8pm|z*L$7_
z{Th{H#1B2f9nZOP?6qWbY@|b=L^r^%!UrPc!2{;B;Y>f5jmy)alx}`|dg{Q4fV5Fa
zp~~5#HE|;dKQe_H{~?>bM7R87qf&IaiCk3|=W}u2YbDm_;mI>Xj(c!-ECdd$qw%)L
zC^3B+5Z<uQKHvB<XA;eT8r@7V`H191+F`BFkQihpM4Y6gwi<!#D)>18_$>!VjJs;q
z#sg&_;i7fmAiLJt`T!joC$dvg2kG@xs)8Xmn`(8LKPuKB2O6VlRmZ!g<3;Vw>za;T
zL&Tx3g3pi$sx*Q_hUgqoL_Eq7a0rX@hcphLY_JU$oia<Ihc|-#sj-=nD|m^d<Db6M
z?b>3C50Zj`%nDGYEF5zXW1zR#Z$c~NEmYMWP;q7qEha6$E|>^-Yok*Ac){+OD~l*+
zcDM$c1X7Z_)Vbt$wA2<A^?u%8NJw2EDT}ME|Mg3{Y@QAs7sCEA{)(t+HE{Iye>b0S
zSwmM=HfYc(=f*;J>g3Dj#W3OG^0kXyS|=xr9!V$Ylw5pYizM)3`yRlgHv0WH<xNrZ
zv2zcONOsOm40YpLC@3D<!>aw3T>YylKD<W5;|Ujcj6PjoQ_z8MyURs4!4&A3dip>7
zmO<!n%qmB4AcR7-JO#V5OALo4INxQHFzQG=4v2o3`Az}{F)e2{G_B*4o<5)7YT3?u
ztOM;3J5`FH0XuM!?^fMrTCu-W0ddm=oU(}R8sjQ%;gTJz*OGwBm`bw1W>$~mV(S%l
zP-^}(fNetR;%WI2EkC>*FcS7dr&>+WEjr}YFPNTMo(|&e_u+LRiDq_Ii5Hu~=|;YZ
zVYsJNdHk2b0)i4k27Obi+0W5BMng=YLzXeKS-pg+M0>8+{d^`IWTuUfYF-voxFKV!
z)R2KMwF8z>b(VI3{=mb2p$)_QnX;^AtWCHB7j}Ppd1~kHHxXNB)oxjiLEr?A0(yOK
zF@_%%f&SCZ=hcFp(Bl|-h8+7WZoE}bu5Zs66Qz`4uqGuEx2`)PA$lVHHwL8VV5-!?
zgk-4c9k0F1-l};pUx8uue7SU0?E=5g3ExN@dGW>twiJF7Pm``<5q0L%W4~{*<||IN
zo?fVlf&lS6Iq94|`P_L#j({Iz;1$F0<)yaX{Rxgv`78HVV~7kC!mrDGaHayr3WNpC
ztUh`b_+p3Q(H|d?4a~yWs9kjZ!>@`avL!X`=s5pLgs6cNv&gSX#NN|6{`LlPA~LBi
zphpeG3*~T34X4dVa^G$hy0%Ll#xik%+oc8pm}K$We{TCKl;~c33N^l*`Mb)ywLkZ~
zxmS)nKR<1O6W#A5np@nF&xw;_^7HdmSFm7waGb}b(|u|!P)q#q)P?yE&QxB;-dAP)
z{j)H%vokW_w-Y>FbtBypwaO&=A!<b(106H-i;(p$w;1{gHd0JW5BKGj6_XF$yMk`I
z(G4oFcK<6=<enOBL`h^-qpEpXXy8!dSND~dLW;_|;KXZJhv6@zpD6pQV;e9T%re8+
z;taMm1_^|olgg(hb?e@yy2KS^L6eBVqG1Req9{_|T~I5d%=ls!=+Xvq;TPjOC?y6I
z1gj;h%bE&s%D<@B%g{5c3+p}Fpg3!b6><uMz`NhIh=CcMB#D@IzyI$4FXnsYkQo;U
zUHoGJrc#4cYFfv&71b0YlkvSb$iPVZUP@`-A;YHo)TR7e|FNApp5NgglE_z0)wZK*
z^yG)EjZCDl7b&1UvtB~YnW?I4?eJNM&_WWpCPwZ0hIc;KMNhR3ne>PVhDh`h1&gXy
z@(^LjAdJI)x<JR1`6Kpy&fKvnZ+i=_%-*iFznl;_3Otbi86q1P|0|Ki6k(~x{yId2
zs~}0kFM3?CjUgwp<{xhhgSR=bJ(Sc*LEiB_P82ODfy4w=A+Ns6E17reK|n{KNVnG1
z`U@q4*VCMwQl7S{xm}{}KfFCB>6B1QV^dT3YVIz>cu~fm3ydR+S``}Bw@+W&H*5e_
zk?<&p3UiQ)?Z6MI&uDM@<<>|h8lXxaTKk)~Xg|_>GnUymNc_+<NVD}Ptw}vksb~lQ
zrP56UVPF<u$3^++KfrQp&j8!Jy>1D06hcKJS2o&*P*&rjQQkYDt*t)<(sou?(T>Gl
z9CJm$7Bf>_5v)vmG&F5Xo0-N}LsI$=7LeN0C>O*<LWQt&>2OnTl3$`Wv>j}StVP5+
zc%ooPb~v1+wg6XHs%+~0y#Q!Q4d+pw$!zhy`PJUATS1sCS%Yag9Fzo0xVWTkWcb~w
zTdKKXn3^@R9~f(HVS#LbhaMr3qM^a~O*1V$iU}Vn1om~UMU+rd4355`vtVmw|Fq{`
z(D#WoM>3pUI&J8f?Epb83I|RH?`YlldL1|T0NrgoN0wep7H<Rr%s~?WS$yNvH1$89
zfc*NYvg0&u07vaJVHmblqB7GkxT~ojc_eI#0Ycb^w}`JE3ZnG%teG4$#rMul$U|49
zL@+@y-GL=cf8*5r`*iU9@(_PtLEBtmpBz!n7B(L&0jL7Xo-qfGlv<Mhc5`h^XW3qY
zOD?lr_lsFF@=A2&C|ayCJ-HbKU#)Xm@VlH4osMXJL|2Fmu@j9r{Qbp7G5vK#{?bow
zr93|AQr)t(bLVA$_)uC*DkwA=;_i^Y^>}Wy3w^uZlKcrkWzDnCZAAsZ>3=IqdBhg+
zI4*90n^{HS=NFd_{fESebBiqDT8MoNBhlNnriZRpwnCd)QiPDKF`HvlltjHRKnH+_
zLVq^3Ay$tmDA}Vy*ocl>OT|)vk}DExFE8~-g2>oG(TS8uCKVc?wdyFqeLU&w(pvok
z<yY7I^3_`rEB*H7QP4zZZ=K^Yv155|Z)J^d{NAu`)ztB3+jwRE^EcWMmdPQn-!F{c
zi2sgZ2~Vh}Pq|#L%YDE$2(oFvZZOhaZFkh9%iaCGDu;35%5MMAjbt^yw~x=+S!ZNH
z0SWaoX?QV8$2o>SH2=-x<<HiGUb)OR=X^Y@Sr>}{D{#ZE3DVl^ldN<C$A>R_BF-U>
z{p}=B;%_pW<FPT_4}L%PY8Q^&!r3@oZ0S%r{8R|}lTNW_An9a+08F7!La?YQP_wSb
zsrO+8#R-pGY~CR3XT5ipH1i&eDYIG@N3u{UU3z-UF*%GW+FeJ-`dX)ip_!yA`Y$k0
zHZ7i&t!1WYvkG?vTdD>nrJu;s19NtmjaXSl4LHJH{Yaon#7P#vLgrxVD3*fYsyaiC
z;raSh$?W%61`w*QlYuY!Fe3hY?)Pd@HN46`o)?T=u5gJsHdC2s!ViW;5;e8ZZ?;W7
z`xSZqn)t!L5mQ^iMB#Y4%Bt=RpN+JiZ#ri#oU3BpD3VfTu{|ly@5wJ2%by>x^tU|%
zUgOaIlMGErLQ7XtgNXm)X=i30K}XK=xxeGHz32O-`2|)le$K1(pEF5mR?<Xsyk>PG
zk*f7V>QGNuh`i$XtDwWxVru%)z58s~kF}`Go?B+Z>D-;c6SxqQ^w$me@P)+qRs71z
ziH|S=L}=3eA{~P5m6m+KZ#6qaB(GLI*V7|64Ow2_Ja`yO#uGT0<n10?3=zx`e#X-T
zG$qW};1X~e)PYe(oRWKQVQ$d02v~ZPS%yGxBm8XA1;H6?<!s{u0Ouwp)gUG|D;gJO
z=IW=lc+fN_W9rjwnudEXTPt2g-Gib=QGX=8lIQZ(j=aUgODeK(7c^RHDp%j43f_Gc
zp5O#~K;g#tK0zs<*SeI$O`8s;qJGHMNsqyA95rH_!WU760b>W#tJ1GmL~f<sEnBa)
zYxu7*xhZ-V{f~c=G4mtJ4Xaree=g_kJ&5Af&44Oq#4%_$gWdD8jVoqCc`EyeI*G}5
zoK?+(sTn&L8GUB`7dx<(HkC)V{gX@%4ITA;S2u(ShEkn)F7B3J3QFda254f!WIz*8
z-Y|JV-D5#r1Qu=Hek%ebx&9K7@cx1uJgeL)>0;Hg=GLRbX})Z_)RUB8758|yZASO{
z1idQ9mq9k~2lx|pkte+T*K}S#eC^Pb=#SbFMFEJXW9`7Tb*{e%^sf5z9hN|jkl|CC
zURh1C_=VjCCSwTJF4_-5{_h^V5z*1UCs2<XgTrgy9?6!*Fb1w%_x|0+OK3$5l)vkj
zY|oa7hflgmb$AL$hi7N`i7{%)R$p7&c!oTojNdAalR<#UwEWNM?S*i`+UILy(`<E%
z>4a01+Ma8i`^)w1m!bPs3XN8lFN<mV<frO|{S&b5%C@L`=^dw%!EDC(j1X)Ppm0VL
zh`?^5ZJzSWDKh`wg(0JIvaz#4sZII24!C(yXcwxa@pA5uDZBR~R3X$f`QD%Z3Vfdw
z@l6#C10)uh^^OQXp?BH;PPOsL`KoH#lY4b?(BXP_rz5<Of!z}81UKL%(D=9(M*v$0
z^Tz}6q-xf->kVh?Wu4Wf&I>J<@XU6B@9xT6`0tBPo2PT`MIYIk#|68sJ4$e*C!X`!
zx>b^O7zM<#s)mFA%y^g~nakX@E~Y+w_=PmR*3k5_V4^wO!}Ge$Le?X^^Ezz$YIgbs
zePmkXpX^-b0liQUmjdHpDs7iZxGdfNBcL>g?`?dSv3dv*=9*GMk#=z5=Vct++hI&S
z9I><#zBD`-caUAK`EF*q?QWmMr}n%ZPFA1C%ES*gRc`nk+lgvefvfA4thg`3Qdl6#
zu4NvM?cO4&u6z8CDW7I*`BYkq2h;cqsko>K+8ySvwhp(Lw~`3%aAQ2R+Hcwh21VRP
zra5uMchCDe5S*l<1TI~S;JWCPB<p5y+7F_0N$(q9^hIy8ZV8^LBX{rL)2vC^!N}S7
zd`D*=A{4rLE=Qk2`IZGfxY;aU!7&AGFX(%ESA9355z@Fkh<}Ws0uGKi@%t(~E8^mO
zZ4x+DdJ?p?sEb-1)xS*jhsJH<!8Vdrp<d}oa;#Tt>-}xd0MV_*UWno|B7I`(T27d*
z#LX<ZU);tyW~_9w(X-Z;SwPG5t4~K2ORa+Xw~M|^{{jd6>YOw|@|^`37Ha~;aqutl
z_5H<Wt=Y%G9pil4q=p9X=&L#jfj?%+>^?q%=n;OJnp!H16mb&KuD|C_P*wM3FT{v&
zhVY$>`OiL3?dJa9zkwll3AyJePB@M)W~sh7bv_>?jrHx*R9;)?H_lfQ<X8e0jPGB(
zy+2_$!W1yWUCmUVCSez{|2h4V0?u$?nX|2)AwFX#b3bzPkv(>u$@4tnUyREiOehXc
z_>j7)7tPYH?p0dS0tL4x;z($6p6m9eh6zlI6HL%eLUTpGvia_I5w4uxmM@*YEH2eJ
z&U~(G3mBcd#wn$bGXFkNF&G1!Ulea_yWP?<+yxf6grcu>_~bup)&3_fF?=r3AY|e6
z%9m`IukfiMs?wJf>*=Gy81zjPk1_fx^l=E0l6)UQOf-W!w7p>MvaY33qGhA-D~U@2
zfuz_+B$OtjCOso}e6%rw4zAqPP)N&`?Oxz-RlD`@(y$KR<W|k98-^B>wVp`hhiX?;
z@MM+K|KRTd`#B|U_F;Zh%=B#wsA<E|@Vgx2Fr8Lvs}}?Zx^fiiRz71A${a2Lv<#T>
z#C3y33!5`ElZ1bJo}PS_9)MwB0wfxL`I8+!{W9pYvMN-tkhPcpx+{Nn=Gx}sll`yR
z4IF*k<eZ|m7}2uQ-kGa=s*92;?XjwbI8M)FZ*)-2YxraI3EkVLWDn(ZWQybDGsiK0
z$sV@g2d+UfSd6bOHWO8gYR$&fG0+>JzTm(O$97Vqnny&=<@tnl;pA91vMH87S<RkG
zz5l0;WQE$;CLn|`;IgI7b0;%1U!MH{VU=U!9G<UifZs+!1Dr*4&f;l8RK(7eeG@9y
z-@lGp9mb96vj%yJ_()=ugtB3JKKYSvFIV6aF^ds#ki=SY9sXE!e}nfbWO&toj3a?#
zt-VRuvtjL$^8kWE1jHBW!pyz8w`*UA$PW(Bzoez7KTf}2I5?kPknwJIPF|03Zn|zB
znctP(nlfy?g=P3%V(XW@B%W8+I{uUIQDhe?X!OXRH6CoE%PVc^nG}UWp1)pfbSri=
zj<mTz0l|^h+#qsye)kJzjmp75MaB&8n5I$UO9z+dCp}x+b?BQ?wjLJJpP%f?-91BQ
zI8M(~7=~}vMtjRfUe}4G#$mw0f3zn07FOrf)q564=Uat6mwa7dlwI)XjmTfQtedG}
zaYpG7Vbx}V2^GBnQ%U|34H6AwfP@D&W`9J><IWU_BUzHtH8nr5x_1n!#D0ceuEDtO
z^~kfgnQ<YOu35>STh<W#=M^^J1jj+Sc6ai8Niu3Z1OyBom}2?VaX<Py)I>m(5D{$>
zk#;50$%H-{Vpu!E_j%ry+d#~qfPR(upSOzuG8j=(OhJ_hsTh=NAJJ#r>zR<`W)N;;
zE-a28ZCE=oDd_N_&4vT5)4nt*WiM(I{!wg;^LI!uLYFpx=BTN2s-T)8E|iFui|Z~3
z`~!zfm0t>vn9WZ{B;{y-9s9_(R%WC}GfzYoBfOmMmW^wVh@{A-AK%o^3YMyORi@5b
zbE}wq0#OG=%b5+1>&BGjclY|_HNK4ZcxGOhM6-uiYT4hMaS{rFmK75;a3t9l-Wh#L
z)7BYw#CNHv#eDPTjk(ubp9C@>LK8qumCA+4;$DY#L^&E`Cnrph3lHyQLsL^U5I>39
zSct3*Oo?Wd#m;L4y{$KOMTT$R!0dYSqtyvQ2mRD;5BU;2QyAl6-zk?C8?_sb8?J$R
zdNj^SmfNf@hg8h%zB)J2-h=~Wlxo&BDdOA(BQhW&KYQ>AErU1>=AuC8I0&Y@d^;;p
zVbR8g0udsL(h?D(x;pK<`0Oj3AE_Z=ARsCMu_z(ltkGf78sR*z<y$Dacp8*MU&tq#
zF6@A}$2)Us=G&TKk4wHhR0$Vxhs*8lm+@!l1LnVEL$~Kq+CupwB%o5y^M&3LXoA5R
zIYf);+kd4wq;ZSXL1x`6qWi=hpYAh_bj4`d>(xq<QB?5+vZ4Hwy0LqebVpdow?1Ea
ze*GGCjT;2&N(4B>^zLVf-)s*nM$3U!YK#BEHvDNFzqxjYV!4#E^(%z2)Gw+cL0q1|
zG@vD7)!oS3T;%(si_64+Yi7eZgDC$c;SRAY3q7O%lg{073`_{_vy`+cTt}}?_PKqP
zDOFR#QsRrkXFv=7yA#s>_=qIt8HtyqLZhKESik2YbiXOz`HuK8KLqsr-gzUc<gV@U
z&xPg(B=OzZ)eI_Rm838x80P&r+Yz+e-9f`d;A08-lL1%LeTNaMG&YZCRj68<JQyT@
zvl47gW9;Qc2I9Eb>KGdzXAdn;8iXyLjpUPK8|KLT5XN};J+7&9{o_9*abAZFlbnNc
zK+(_AjSUKP3{Apy-y7}*S4j;k5WE<W9<l)N&F<;f>R}owX4h3Vw`f|_1|=>6e+Cxp
z{=7}LMg$@H2rjK`tx<dtP@oNGE2kKY<g57GO(-m0tZKnVLX}dobXwopDP3_xZ{~&V
zyy`!k=Fc)lo;azcc^K9IIaCGVr?g6i<J%b{`THB;!yuOK+r`#1J`YdCvi{8hYgd1S
z$Q4K<={G%o5FGXsCVMMa(Y29GNiONTk`MJ@>6wB_>-FO_Xde6664DeL*2ZsLKY!CC
zCEZBkwEMsqjN)M6m8xDgO=Xv?OJnxulwGS@jcCG*tH{eXXt+&&e*^x`{ntvsE5`sA
zRb-(W=b60`T^twfwCCxm(a;ebP4T5L|D|&|pEP6NRIJu=heybjS^Q2pMck?EN||ou
zO?SLsk9^CMi&H@D!20;fnbi1kQq8}@dZ3q~SK4>=z+Dq*wOHcupuH2+rNr{p57y_4
zuT<lBbCA71fNqN3f#t3v(xybzdRv!qaT(pv^<QRtPSn$sT3X|s9wv~1+4g*M1fZqE
zh=?E;iu(1Z1NX-wwp08jdOpj`dF!SR?<u67=Ioz|x+U-u(^u5i&naflV<U-`rBfQ2
z1+xW}a-JBc$P-?k*(DM3*IzDtDRtEsn>J_SuT*d2I4zJtR%@N63{|&`EQ(ww3hrIm
zGp>_yST@&yu(yHKKw+bRkVtj9DU}KhB4t_%gk(8zv8JIjSRgA~hNcq%;!JA9EAvhe
z+#aJ`)7Ypwb9xH^0L(=VOVCF?6A(hR4J*|u0MdSa)t}bT!{M_UF4{ofy&sOQMQ{7E
zlNL{ZwsWbih{5Uo_>P`DmNC$!m=znS%Pm$xQ%7RO{pkZ5Hvj6^?~=mNjFcecrc(_a
z<BkplYnXD$M(t&MZyYk?I6TpF0DbHIuTo0b$?uciWawj?!1(|?Fq#@=?3&C=fLiHA
zQR-n)GY(l8L$s_vOniH=#o(`Xs0|$Dc=*8&M^4r+!{6y~6b{vD0ZTzt=y<V*9BFXb
zpSZ%gCdBBO<8oqw!WdA;|5OFiQk6(fu-6Z#^0iHT%8%WUMFS-hp$vOs@ok#PrP5)N
z{WJa5u?FD-paxBsV927Dq}&diLFi@8c2gE<O6){uT7}kofDQVwnXHFLx_3eOAiJ}F
zU(^c8#MB|}2!WO**NLQDEPV|y99xyhq)7@9<WxG&$Mo@Ayo<#Ih+R$>zl0_=oqSp1
z)A>XqC*G)Ax@J$6LWzmDQXnHthRc&>?{F>F&e@cMY8bLqi)(K>W9}4KSEEhocq&#T
zhG)ECQnVY*L_10gFij+aqp6r=4V+?dUJ_tsa;{?ar$10fN2(yzvRj-{XuaL00@C=M
ztl$c@8qIt?D{yRiV9sYFB5P>q?0xdi_5Q|ogtFP*F4nUpQE~Wtx|B+lSkKt6@{38%
zp=M^5*pbx^iaU7DSctfx=O(aCTDD7*Mg_AuMcU<Z({Adso`t`C9CcM<=(i$@5}cu$
zZ@GX9TDS1msJ-Q;L!#ISeg8cUHALpg$rs;l9dncJzP?QsaU_kT1i4jEHhLtJ%iGK~
z4^QiMvBvy}oR}=bY#J4hM~e-1>sy+32nMXxEG<<*@DztaB4wx;I8t27#U2Cng~y}a
zBl-8YI9;?=yv;-RLujdHuG?3JZ-4*#zxk1OkN#UjT>In$z@?1JD1rCRwM&byb$L^W
z2@BsgHu}4e?!b0NI;XGiNa8B}%&3K^Ggh4uRiTD+9G2^Jkz>syJ76|rdcVl6nT72V
zi?lPQzFZ`V&#$%B`Ma@fx}l+4yn3!k(@F)!^rswR?qMhgW!2;dZb7N!CXr{_e`U!{
z??!MMtS>d66N_eX);uJss{Y-0wJpp6sL(QP*bz=P$^Tg1Co#R_N}n?sgQB*;FpL}R
z8Y)^`LK7uUg2wa)aY=tu%~EV6cJQfCc}k(^zAC1srdA-!`_A4Y9B*$kT%6d^iYiv)
zB7K}--6g#qjg1|NLTdk+BNKv62FbYIH6i)aHm>TSn0^3zA+k5CY>rssCG9Z-s-iFA
zD`EMhjV?}vxrY2PZB%G54Vp-jwx>m!00X&0)vzPjrF2`tOW-hO<(1V{ae!2pp;JLL
z`(SE9Ng5jwW`a0QXtWbM?Gl5Wijyqq;VAU#bYV@5@!(~daHuWdj+kz#u=gqok!Ys@
zE3uA9o$rllF{lQt6937PbXs3eKG6pJSwP)-N7uK7-bg>Rr!IZ^8gsZ~6~q7P{RRj8
zM^rZjA}bSJ>-(sf=kug<vAJ28q|xelMV;Q`74ae6V`3rwZq8EYlCbeB$^11Rz_73y
zubD{p+1@4fqntzNn1c+H-{%5O9!)WtdGD~f<p_f|$1tzVr*6WLk5+}<CDm(=l1e4X
zSD(!?o=tOp0xL05oN0{j%h5ldru$2ZQhCJ5>B)I(n%gG^#z<`gyE$^#Hn&sS>28mO
zV#OMry4mi}8QAmwP6w2%^KokJTsmVNV_)#z+!Ie>LQemEzae74NKnc(Sv6bMO;P*v
zDVFM(iT~9nq!P&D6=hCi_5eU&d_PDxk{J9}!X}H)Q6KHl>6I)`n77|I1?!aLc(~7A
zv+mnFC6;|*tWde>_$3+OE69C*=RVDiVAKW`lwU3u?@956i<(lW4IxXOjl?ZeX76bL
zCqmad+|o_5oU588-wOsTIY5y(B$HCtLQ-Hk!(PkXnNBf_K#hVh&;sJCSpLN@{8G9q
zIBSw%lCLN~R@G!wt$NCG2Ku%kC{*7XN<cBrxY=+b4=~z!Jm(?I`)d>@-d><j7)}Xb
zD|G%=WwO7|*^qR^(ka2&Y*(%6Vl&F#y1pe(2WvKde`6!{8T-;N;|lBjnYCDQ<FPIJ
z2m@aD-KX~E4Z1m9{V7}H=eIoH6Vk1RRW~xBWsCQM>3(0ELufaa^B6ocYV0?H(*H`5
zlxU5|wR#<%Feh2AQ+CP+>X!C)ag`eyT!Sg)y-s6!kM5*%j0{Q|JEVF<Xwh*FZDLBi
zzy)|Kd}RS_lAja>w`7K<re&;@%9i$wpOMrAzWyHo?m!X0$Hj#O=H@GOIun8*V*OZ)
ze!s)DYgc*S2iILYLuox4=d~7%aJ=Bs=4OV>N+n&964Rr>efF#%(*Z_APPQUX5Xu6R
z>3o-ZjMlm|db4@N&pmpelyVJ5j75mI`}w0lZT?UH#b=Sh94as<RiWDp`KSNl3;ggW
zewbHZyG5hf<Rc$?mS6jgZ}Qgl4R*HLG?tcp&*8xJcg?g9;&2>0%S?FKNg&MLaC(Re
zifl};yGwto&HG;XASc(KARCYP`j=j0zuo7#^<(VaxlOy%BZ_LQwALACV}dYXGRT;0
zbzPtD$p*3#q9a9rlCpca&(^hXB89sQudlDOx>#p-XGEAJoSR>w-`?lJ^*6cn!pEiu
z`NIrrExItIHbe(8Q5<2Ts%zz!ji%q{`k4#Q{|Z@LSeRxljt(i9X0e16D4ecm62}ky
ztT(=Vi$o+?>G$JIvUz=jQzuT)-RlyC5sRyf?A_gE&>1iuOsIO^pTL6_zO#>|tQ5!f
zs20A#z|k}<rNsq>E*NKHdRfZK@(Q)a99kFb?rxDzau$|qjK)2(iR&@ltklT8tcxu)
zI!!UwG8qrBLb@v2%rF{{X%F}DtYd1m8eui080Sc#NLmTw$(WtJU24ahZ<97DvxLGr
zYDFn`Qto38WIA$*2CR`T;M3X_QW}R&D?+re7<>(Z5Yss+vwBIAOtTx-`0SgMUXRqL
zvHklfrKYV$f4}jrPJm9^5C)mwWSkOwuiNL!<$JvP^*apv0fCnT7!Go_w+=Xa_Skd-
zD$|tTI)q2?#1a10+rCe@REVX(=rNu49@XRoK@hUJdq5b~SUrA<dJu8v_BBSM0k!%(
zGWBq3F=5a?@M55Cx`Q6A#bt)$3A=lT?Cl>Q16PsK>vcIe*kXCPNf-ubozw4k+1h%*
zlTYU^Iv2k0t0_~T3QRHO(iM5`s-J`)2*c8Mrw72W)8`j|@w0sW#kY9hxkc)=5ra|N
z^RUd6dnhRRX0Dhdj%Ip*We{vV3gWr|%vvy&3*li$H@5`j?zRyG2qEbYhFrV1#cMb2
zGtjQHfk85yOxQVSb8>BE%AYG`Kc>VBm^z2>9VpBZ-sN;#6T$^AMTVJ-=^wVKtgR6Q
z5xbo>VWq;-`Wn@!%Eql*jK?FY%@z~m58yOo)b675oKdesf6${b-(r+aIOud}cRGk5
zKo^=pzfWgx%tF0FpcGo?4EjTMx3@W0c$*uIE((`o6A~f)dMr7Wxl7LpzgFQ$+S$Oc
zKjv5d_3v=y_HCYStx%ne8THzpZ^_i=<zjKS_VU1Xb7_4LjF*HdkzrR|ibFsjfz9_1
z3{jB!RHKuH>2$ih_UdiE`r-`+y?_Ac=YxL6#{E4mo<A}3bHr56gtxh|mkf~ZCOi#_
z%EMF$!cYMjx){^m-=TKm3}F!QU~7*ssk46SER7)M=Jhui3<ort3n;0b)JHa<x4%oF
zGx~!AI{hy5%d7N9W43np+1=ek1`$P3(CK#A-Q8eytwj`t6eg$JYxCg2U7meD^QjKU
z>#?S6I)2p84MmnyWH~|zLg_h4M-G5~yU+jdkNz1i{_Yh%d~S(mV@R)mNZ=)4W_z5k
zv^6eO9eYlolR$FqmaMf%VczEFrT;$)n{VxFqGW`{5`-~I2JG$~@cA!Y=3jpHo9rG2
zgw{}K#X-Br%^UZ*@a#ozyRu;7NKjXXwjSB`8buDNn_l89porrDYg2UEXXn9PT9=+C
z3M;&I?*U0|j+JMgB?&5+m%hxf+hKlr4ds1LkB1ySxKEyqX&-L0+umd4^eNhdA-C^u
zaQFTLRGg4y8GCy>Jh*e6Qzw^5;+R~g9PID1efKs`yzsH_5K?`o@vR}s+ee<~{Dc4Y
z=lQLF^%XvS;W&%UE{FSDgi^Zw!w0j|A`{{E>1wt5ZLd}GjfHUOekIj(zj>#m`S1Bj
zhEOQiqRU;f1>^CA{ReyO-rhxyg=^zz1;r$%f7oMjWx@0QzJ*(N)YX0JwY=Pvh~#YM
zT*_IaDNM#-)Mp`DAW({KXGo+X=2{C>;(+b_ZP#rjsbVv~*V>%%WPmmW!*s-OJf>P{
zkfk~O-jG3mfE2C>b2J{24>FQQ1r;c?HjE}C4m*dOUeg|_o6$hZJ*o`yjP(~g&zE*+
znV5ZGYe>gqzW$Y0xN~=#)zdZ7BxNuePSN_eO^ydQr1#@;K1zhPY%vHNN&Whtd50YN
zch~#XOwigzwo)h*0i$%xm%jWO<DNk3nu7^NfYE|Gx3~Dvho18rwQCjTeA=1%sFvC^
zN}WbP9@)X6_F))t?AUQO?{-NiBl`U=J3IU2c}BC*VrglOlWQkAefAQ&ds_$~iK<8$
z&>MEhvy39k*x%dc-u+vQyKVOO4%pi2v)j#S%_*c&RFeo%3B7RWm@IP#^7ZRi`NR(!
zuX|?6vjPF0w<dgwO^_D_d6tvw!UZ(~0wD;dsGH&HTi5vFi(lcvy#Ym<v)+VOtw4nq
zB^?(@TUX|c2{E=nr#YiR54p06QUWd1bW@kYOdF63DW=+K-->$vL59gvv@^aZmIh?m
zgqPm9!XOnGRdI4%GQe2D-OU}|_v};CTZI))e7AfODWyvz31?s7Ibtx&(v<EWMj@-K
zYwYv~WD{2fwcl=2<Qa`dlZEAF*49sP{Onl{c6Sg$5+^l8AQ|>L<avS4bJ}}*Y~H`e
zq&J}5X|p#Na!}+nssU1kRKgGu22Ra9P06ibJju9o>n0zx8YSJKn`<YSjZd))A(1GG
zBKOE^?$fYN_NVOMiPhY^b&Icm^A&c619Wa!CZJK|C^S+97{COfv#AUaR-<#xcravX
zaoO=cO0qFP*@F?60>GM?tjI$SYepCi#%h!ZP=t&pBR==LFEi{2gsyu(57An1<HiF%
z{_zhGMS<V$gh!Q1n=5zm@%}Kqz)7Y7?9|Vrn3E??bN^O{$#_Js*Jf*LhcwS<wL(@`
zj&u6NS<XH24BOiekO-2f20=i-e@K>Q6j{pd_6E0aUuV=lV0UMajm<9G-JH33g|v!V
z5>c;&=-e{0BeKjg=#F^v&C7i1j~genT^O>oKnUmOP6^*VcAjgpEF&w9`nmKQu|x4L
zU%tZceE##?x!0qZWSnTgd^JZ!7Nr7=wrCSh4|+?EPIHF6E^_Ud^K%&>z0Hf<wH`d`
z=SA^o{9I=Pa)*^7MTjJ3FzoRQ|Ma)m+mjT<9MXamF<Bw`=BwBF)MtJKh1>re*%ZoN
zy#xwhvE}|;w$zg@o#paO?!CjL!o??^;>OFn3<o_9_IJ5;`!18on5D%PDyPpApL>=I
z@BIk*gBvI-kVylCB=7HICMm^u#Nn+Q*w<g7zq!r*dk?sAcaPh>l;xEGWkXt(3bK|^
z6o!7k&tzg~ANIKN%1eCWr+(&LAa$ia6Gwh^AL+sU2mI16|0}NCIUpa7c_M-3BqNF(
zC8Q`U1rfRo>MZ2xgwDYMjk&q+h}J3LQrki~J=eU*(fZ-#stR(1XAwypMJUU2?q0i(
z7+Yc!VbN~Q893Z+bLNRNu4Ncb?x)<d0lz06DUg`*CTDxPyTe%m6)-<H&rWAdUSy0X
zWBR>5MNv?zRA|*(EH5mwvbavC-$4jS!U%z4I2oa}%LerNUG@(57-u8;{Q;xSn6#&f
zs|kUSgi$~!oD52y<ydVg(wyy`EzBj~>)XyUNUry|m_EAW%oh}1CsbKbu9>x836`y`
zEv{d`N%wGq(FN5xplYzvwNJFhY1>%~%DQyAHJUV=AY{`&gU?iA5ZcXUxynk^|DF2z
zyA0|Kxyc-5Kq!<5F&g&v+eD&Hq7o>}nnl3wPMb8%iIeC^ka6T@#|(8k8qk%RH&X(#
zwFneRTqTS`3Y~FqxXaG&4x{0OGbifQDs_S=1Q~Gt!gEwAHNqgFTAyQ-4#=~NPzuJw
zJ_ma{^apKj-nqfoUVer2$_+Y)eY6&gMqQ%BGB+0!1Q13Mjb_4OdmCeN0wuucX)7$I
z<VighR(AdIT&}$+G*TJga>|ky1+{vE|KvaUPq}{mP5$+-{VVQo-6lS%NU#_w(HhcJ
zqqT9Iq^aBo1j-*A-q2;s6(#LdAI|8RaN@%bV+<zC{XrrTGC%|EP8Si@d=<2}rwZ6R
z?2_l&X~kLXdja}Qy|j5rsGU|zS;|VJQU(QUK`P=(l`sg<Sw`pZfP=k##%ad!TAfO@
zh6*L93a6iXnmDNt#0iyVlgYSGp5+8WF&U2N>>n`fwz+rrHm|;Mne5&jdZQ6WNT+#a
z3z|`YvW75>sKpW8y<KeK5Lcs(*J(SFQYZ^1@H{8$4q9}9G_DgsS-M`S)%Z()=`V5j
z_AS2ruYZ@V2irta5nIQ<GZr#`BP<za_;XKr|Cpu#j0@(3aC%T>T5eXF|8T=<OQtiY
z#U}!!j4(ND@9q<c7KzeeeQ+T{wzdvP(~KmETyJAN6(yY6TnLv&^2HL?qdAU7V7xfF
z^Ycoj=Czzs_V>5h+}va|7}Gr6pk8aBqL4^7xcJlyMAaH1tRR{Tn8^T}PFxp`evkI{
zCjH(4*KS<p3om|^$!l-X-tSXr!Eo3iN-PWWF+m7X6wz!|*x%oD(j+LX{#HMq2`gJ;
zrk}}pL1k$|@8?berO|BhxBm9u=FKaw^Q*t~i`;o|o#cd~A}mHrw1G)lc+`%m&B6P*
zl0N>i(+lwQ5<K<uM-<d)O)(xLr9jFMDPy#TyZ1Jca*<Hg5ClH%iMVlVlXRR?tt7ZE
z5-v1w<me*-kMO9qfa$(3<P_z=8Uz9LMvEkg$@4KA_iuCS=1qFt0dvnR5VaN|h>^8<
z;`e?6ROHiH66IgR0*yT5<lZ(}caN>tUZ(l0zsB&zm)PDqBsY@oaE~N0tS?mw14~k=
zusC05>%mR5*8bxCPR4f+{+3r(`0xJVf6psle~Ewc|M+jXeEAhVw5|}=qP3*ZFdXN;
zlGzQH^0VLNRh>+hvVfHoz9+IVkHF83q44R!vi%hXhW#OlhzX^3dmgU2{IEMf7uq4k
z)_Q*}nZ-w*xJ2qofW~KSj*1maA&A2mr4)rO=nZ>xyItS+Iw6S@gi=I8a$@BqaTF8C
zkV;(jEx8J$5@h*={-DcnGGu#qhub&rvRUjh9!<aqM%fq}Si)+6m8A+oNO#bkRo|Cr
z+K~X%dVj~xR4lA5Mc~b}=e#k7Bu@BapZ;TP?QHS-8{c4fxJM<F5IUL;7IaajcxP#8
zIMOoBpCe`4RN=EP2C&w=tG)eQ5^$UrlN1UmN-<iAwHk|8SVsVwN>y<3)C%D8&Ds=b
zjsLl_n}q!~wPaC7t^Uv|4@O<&?C$Jyc+jO%Nf07uJW2_}h^3`fCX<Z4{T}o4OQZ!H
zZf=nz32Q+@7*`#i2bbDx%^Or$;e1dfsy3LLU*!Jg1MY3Kxpw=2xrLY{%?Wac5H8Hm
zGdDL+s3fMy!Du9QS_~ol)ubf4hC%4MUzoCmilxX4WKmEQnk26B%ro!di6<^{@9teb
zfA1c{p&$?zD;-IGG#sOS=ZG?1pEf-Ai=dn@*G0ggLS^O7qd444n_FEVghI%`H4hik
zNj+!{fw#HQR06@twI#rJ*eD7N5}`trbWS3yOEHNfkaH<KYdwM^g}=2F1^ewbo&JD0
ziIIh7oaKaJ#QfqCX<pFo4`|LWkXcJ-Z<i>JS&k!uq~f|@7)UHM7Az``IaNsrYYDCS
zc{aBnu-WZ#r_-SsDdM~!z!F5zYBp)MS_DF3at+!dEoiJSd#B8`A#j_;RJ78=_F8Kv
zP^>j^l5pXPOPoD>hRux)zP7o;2uXm&YK^uqo@D4kPg57*%I8%OOasfZ3yUm^5ezN}
zDJct&#O?xPa-?)NccPaPtj3Z9?inowwW{Ro*>&)~Q0UzA<^z9r8}H|Dmo9QC8R=eE
z2$XVxp4K^ATe}?Wcc@e=NRcrbO$eiim6dhI<BYw%4z<M<vcfWcut8j@uoPBMNfi;q
zSgTRm(p+362rE1h)rjg%78jSf{opRQ?;r5S%{>+tV=9xJFt<cu%+lf_iwg^cN?`NU
z`MGd%MMij?DrRy;Q5YeE($8IBo##0+FDP<NC8_cJd*9E~Pe0A=TQ~WwTes<tAe0s(
z3-a7D9E^Qsx0`v#nUsD#Ns7|Xaoj4G8PL;Wvq$CUE{l=^0YZid8377Qj@21LIa0SE
zW4<MM{ypbV);e9xB6GTVDsnB9tag{5vipgcTKwoc`iq{#8o}+`_t@CjrQT?u)P!Dt
zL>yN*e(Ee@IL2&mAu3HorpZSGR1~6Cj#26$3M|L~xv)Y~ZE&*I;!iFta`N~IE?>LC
z<y%|4_|^t%%N6RQ39+;!NrmIbkFmUd>>bNRe6N5`2&h)8JpIhmTzujpw{KkM7hb<i
zd*r&s7?qJ1hHke{p64^P(g*F|>9wbEkp#4z?k|sKxQ+}`2rodaaOFXc@Ncj_u+WG|
z&|01YX5ALBpoDUAAJ!EF2q~vQT$!d7NGB8GWkIY@clLXI#>26b|H(8(<_aI0^(Hzm
z82882t948P?Lo(NvucF|L4=Zl0|T(wYJ)(A1S=7tN~qQ9>>nJ`KN!*89uQOmiqN;V
zQ~|Ymm0Gptip^}{UPm&m)(4L)`e|0ag1el1Ykf>3G)5buFy_Ru6RfVSa&WNEn^zAY
zbvDyDRfIgv{Y7wQW#`Y{?nfxb<vwXjzT&&I_q*)<>bpaP0TzK0u}ig5OhpKVkkspf
zAO5iy_{c|}#acsN6#ho0yzZCqJh8IpTKQG-2t2Z5t*;g}C>0Te6?S&IK#xjNrBQDo
z#f0VM72+tSQJ-V~uuG?VNEAS;ny^`9EUzCUjuLwPHr<1L4h|0xDkf>vId%2|{oa87
z=zz?^)tiUxAB<=<E9g8YJ?N7ceV%##Q`o|g=c$)55Ez`8tjoh6;W%NDUhc$L>ndHe
zhRh{6&YwL;R%ATbyo-tyxd<4J4UMX+a52`Pvx4zxG`-mv>r<6p+*k^)XCrV(mT@;$
zE&Q$P$gTdNh7U?rDE!6Sw+q8yt)8aa>Y?NZKlC0x^xTsUohk}Pj#gqCxLa$HJ~$P!
zlsfP>7hcb*1V)TSg)u>#us;~Wa72<+sWs-1>6pdEC89W{)@;%4_vv-pL_j^Ru$vbw
ztS%8n38P-0-eH?gr-O(B;#!4cr%y2)O&GLy$y*I>b~+qRGU~D8*kywe`C!BoAOC*W
zm_5rpv0q@64`=~t%G;Lm!Ind5oXw51c~O8BoH%|8t#fv^H&B73P%yTNy7n82cEMjd
znM}DHWfiz@804`p_g|<2jBqz?J>_=V8CU0_Y;G{xf^u!?q(~4Vz@jnM1t8K&nhnWk
z{^W=G<PSWLwT3M70Uv?iblojjDmO<0HAQug`nj;yaTKkAASMheY;LuI4%JGHW@8>H
z$E>Zb6UPb7#sYi$Z4NsJR07L<wL&LPSvYYL;tECgklx-7`+IwcAg0o4a`xhr^tydI
z!#z?9uiw~bcYn-$!}<9n>yi~+p8v>au-cGk6HH$Exx(nu&-HilbK^v3i&^2znRBPm
zTJzw+JyaBsOT}<(X;qC!;WTMhFdhx3K5qP;=ZN$^4RhqOsH7psqP0VCtyGW3&yAtT
zElDCh%2Gp!2x|+B&3y}3#p;6MfAd%WB!B+T`~aaf*epZX0z@=Rw_4*N<|Ew^zI7Z)
zb&4<|s5H2FZx><rXf#?ZF03MhKBvx{gGvoMw?unmi$QyzMiMf&IEU>;h}N=SvIVqv
zuv=T0xPoZSQ-9A3oZCNS|L{Iz1Yf+i&Aq)LEA=YHc*1DwfXTfB{)0dMbG*A&Yu`Pv
zE8Fx50YMTn2|_xlVJXQ`kwTk-Vv;c#4cre{U-0yuW4h^UxooD}O9L73NZJ9Gg5ZIB
z^N;|A<CD1t0R94=2rfK-p3_gBo@LMc+CRGXtY5E3rK*Gwo?at-z~XqpfeZ*$#Hc$$
zAc&)gT2djEIn8F1APA`@RfdBxgV7KfKvD@Y9Zj<_M;JOv$Y|7OFc=_3Ko}-0Ev}MI
zQhMV7)k1K%IbhPys8%Di(Twv6V?E*0@n^xh7Jb%NwPGb&l>KdINSXzdu+vLa*(ObU
zv~^`^iNYB6I|m4<Fj`_$i%5LIg9Tk^(kz_@h0{Q73Tpe4R0vccy#$QQ;5a_@3<UZ8
z#=8pYYE|fWzy=_cAXIgfh$!sXO^mTrtCk=AvG?=+pL`#&Z+oMQ0u^|I`Ba27m?Crn
z;iZdwBD+L!%5<z$AQPgbh1SC5#!X6|Tb5VmSY2JgTFaNe_`7sR8Q*yIn>=~`9GA|X
z;pC}f%q=aTlwx;xi|bdfG8|^y-q@lu8nAl&7%TG)R#%TPOt*<6L0(A0utKF7QW#5=
zG<fFe=NOMhE~qoc@q|P;bu_E&be}5qWTvIxR=Z9a+R*OyIXLKY`piid7w0*?c8sV}
zp*QZbDg}7{mNu5*u;<b%p12^TAP9VF+Q@0EGV3MOFcM?b^a7|J!2w^9&?~&d1`-ul
zDO7|lCLrA5UkTucKKNcf{(<L+l9(cQ3g5C!xs^PHvO3V))|9`OC0x&xL?D!eI3jE|
z(Lz$>1y<$cgFcJR28)XeSY!FdSH4O=PkH0YWzL;E&AAh2SYOpN=Nc%f*x%je&docF
zi-HHcTl7XfmX;S-tT$O&U1zkvODq+IwS-ARtTVJVG-`D&Joz-~ICY=L7=%<v8@WSQ
zl+ru?@O7L6td))fWiqT3Xk+P(2DE#9P8?fjey+vp@+x5xG3aM33QHi=G<7~245l`B
zh>*p1?!o#FEJlHq4!Ofx%#oHnk0Pk6*PzpO!MaeAFleCigu=Q@2HH}q!Jqn*AK_C!
z`XQ1^g}g}7x<CcOBl*fxsAhf+(zON^WsgPw!Mx+RRIeneEl_Alo@Q8+lI0y%R~J}c
zU&C6%?|tq!Xpd69^3qE@d+8$2Ubw*7vnOe-ETfcSZ|gp9UcO9!kaFYR2JK;w_0uO<
zUubds*hvQICP^sC3qurFsU#7FH7rzGy!X8yU^Ey~<SAMkpYn`cs!s~6GSkhqY%%2g
zTw;aBS{LLUwtMXEA8`KMNtTuuId%L5akWZk(qY|G4_ra0q2KSkjV>Xji;n>3=UA8S
zk)R7Fh^&><=o0qP_<5@a+j~Go5JZX~nn&hUjP1GoOB<G&mjC@<{S*Am&wiRpqk)|a
zAkSP-pp>h0E&cgz<9i=QA_xiQR>-v^8&5EdnB=>hKDEyAvuD9-ZvN8$$WCv_7e4o!
zy#M+4@v-OMPyLxoh{n7_2lwvNd+|$*dJ|r|c8$$`o6}ES<l^cQ=Px`-C%-{8RHRu!
zQftsyh{=p)wYkXqKk*||Wc-~z^S2I1voN%K0~Y5Rgki+_Gv}zbn(ViCITtuoMthr&
zyWMFKf!h;*|MyTz5SkESK=Sa#F+>(d7GVQ~P{cAK5CNGj6)G$?GF*86A{XCto-mBa
zb>a8I@;nzaIt@oqS)aBQ<yG_uS}UwT$dDwcfbaR9+X7RJsWodfn@ud1x30g*WURS!
z=PqmO%dD?2Gru@TrCLKuNvGFlXM2kx)9mkc7>*{?n{zap3G=OaY}!SIu8=QOAtqE9
zV~K-=lgCd|WSMKVXdKYssNP;_!0CfcFVbbuWvp=72hW=wk0uNUV-}YdsMl-E&$m!P
zL^>W5H}QNq_q~&J;xBGKp(5}SNiI`itaRO#jKxS-Q6PPrF!&B`^<Boh3hJu$gl?yY
z5Q#hBWQY_o!i<sD62&>6_`&D+*!MqAt?q9326sRyO2p>7BS^;cRD@L1!{6nWC00qE
zdH(x&<Mo%=-`%9!A5*DXP9_QS^Yb(s4W51Pb7(?VS65kV)royoQBf33hJCiTHYqHe
zxp;}q!yZvyP>m~Gz48X9j<3<ISEyDcgOQ8gtJON&yZg+~pJZ+A_z}_BZ-+z6H0(6J
zR0ey_93U;Jf!D8G;r6{Pu3UYSr=L9S#B?h$gF{PJNMfH^H5PjPK6zevxY|w;0*45R
zS!*@xQr^ZG0=FinDWZoRl_a7!9)O750T+Y_6=JdpV2NzOC*SuTKKlM=sU#6XNsM+C
zdrHo1J`31++YV-o@tcD3XMBl>NUVe>o`0Sjuf4|M&JO*_m?Vmb;w4(m7PVT9C!Tx)
z9VnKT7MZIyh*U@=skocbaLoShE(XJiv!~b{bO;OAYVGFrTO3<jq8=wC(lW|j*MLf;
z%3h~UYkiKD)m4X@maSX9Wo}9lZ7EB1!yoWiupwAMhUMC=8$8(E;nuC2TsU!@q*5V{
zLq`1xxll;lP0R?va5$vM3)}`WvpGWg+oMJp_x%9MDuScP-XpMiy-}fk*g;TnD(*7!
zem+84OC0C?*pI)!5C6yu)aw;rB;z*wfB;o;=iasp&5q4dZ_jZw#Wcv1xHfXnfAAB$
z`tq09-M&Y=KcZSSoT*e;TwJ8tZ1VgIA3%$c_4Q*c&oxK_C;g)H9IN|mK6pT3E$5$l
zhWq<nqCBS_*LeNaS2=s~7;}vZwW_2)bO>9mR%dg2m&K*CoH%|8dqhyY)R8KuReL{I
zQu%n)1+f-OBotOzQVXxV_8Ql3Kj8A~S9tE(bJS{el3GH4$C4HjseI8G(CzlfGk5DZ
zwybas{ViWp2!YVr`?(-63KQBRe*W-5-F&Og*3KSKM+FKMBuG_3PkKlU%}UB&`HO#y
zKmXG|N~Kvr1PTh}b>~W<;UmxSuGXR8gFpV0eDybenfo{1VyDxm*)Tj)uTwj=4vR~i
z|J09g)&{JcJIl(_615;iv|4UO>5Ssmbu=*dffvx5+tiDk#cGSM{qE;^_TqV#=Nhys
z(9Uv<Hnds`+_-a(<x|ga?A*oaq5UvpbFa;tH}A1F*W$?wCup|js5UDc?pelKBDKI6
zfd+;Hht!q+@ca2~ndt}ufi;f?oekrN@o+?lyL<;C@IhUM#bAPhQ|~#?`De}(#-R)1
zj3fL@nC{c1{DyzhS;`nEATOul)4l*fIv(Dc;}^KQahq<x&19SrMj>-yNUc_-k|Z2I
zaT1Jc^H{GZ1fe8~Vzkj@`GiilgE6qYv_`i*CRCbA9I>^v$^2rIxDp~GL6N$>F^*&U
z-2n@=8gs3ACuMRZ<vN?}nf7eCua`6^Cl%?7L2}E32b&!3ciGz7V)@u2aTKA#fP7+!
z3JIYzvam3jOk6*3O8VLCe6;RO-QMW4lU7QU!ezwD)@AQtysMzD-?mums!l@bn#?Mh
zV2Xq&2zcRzbNs*$eTaI!?sT|>MHz*$#oJ)%(&+YE4x;J8wzCs-N>Y@9<EJk2{XhI?
z_~*a)k3b&~80&(fIA(6ni63hMP8>T<Ht0D>UO2L6o{ceDv$(X(%E}5q{;40L-|Nxo
zbg0UL<)sCN>5x_{CMz<EA}0zH4)+Q^{?Ydm#}$v5g|m&+EBiZgyPoofrfcI!fZ`-c
zn*8{Wya%J<mp}XK?Cu^iKR?fV-}fBf{L+gICqtq@Ib{G~JQ^~Yj2&00$ft`y3W*J7
z^EcB+D|t%~f3TOMmpRUt^$m!HRERK!G(*XN7cQOS`#=1CYV{g{%bU6i;VB=`&jo%1
z^N>>MMkPFrKR{=}ykm2$o;||{e*DMyjeqg~gZ6;H22?5)qBy46YO<Kb=s>Z)x<)pb
zV6}EUAE20|7;TwbSYUBskxzW&qYMTE`n^6$VOeO+F&+<SgaLV$Qe-({t3_9LdEv?T
z5XFgW1r`K8{UN8EZ>Q0uJ!1DAY>rcZk7_kOaq$AiT0Zyr&$Hj|&}_DN@~KN)`TAuh
zqY0r>o@*i)kH@5`Lk4wGOc72gT~KGOL3=^pqwgoAbeeaMV6+d7^K+9EDTNTOWtc1`
zBw@(MKl%he{^^g?Y%~c2_gpFnFeaY{7(zJhIcHiY0Wyo}<&-Quz3B_by|L1A>ipCE
z@MnIOU-<d|7K%MWZK>62BuT>j!Xis2Y8V@E;^ZmPUKezZ5*mR;kB1mtu)Ms+vGp~6
z@-v^N+v(75x2ejU)#W7ylL7N{F=_5l*f?Hde^2wnKk@=`QpH$1(|GiLj=&W(pk3Ni
z%nkwvI!dG{^{1Nr$PYfj4}Q|}Yrp#2Y;EtexUk3zANl}a``qUlP6kB2`GOHJ8V(tc
zCun2Hy+n#s&d*KY0zGRSvgh7!)`iE-=fm54Yu{jXj@Fsy^F>J206U}_#r(O?e3<{}
zKm1AROG}QMgoZGHB6St5M~7erMU<ZYJswgB!TJ+V^T+?{f69OTfBu_raGxkQG@DIS
zrG{uM6Euzif+vrkhEWd+4S^?{Wh2OQ8Yj;XT)4=O{V)D9#lb%L_BL}SWA(%_+M^Du
z^A*OEAz3=6QaQ%vPR^(QgC8P@zO{GmA8bNTtu1nLb&)G?-l9JkaeVz4KlbTQ@f*MJ
z&*@AKsf2;6DhDQ`Q3<U(e%(8G%Tnj8tT27pVTZntf_3T%Qb<A(5Ta00uzq2UOV2&w
zIPA&=Vp2H`w^`bd>F=%2C{d~zh*>4<%=~510W8n2@!S(1<i(f1fMG}=Em0H^1|ijI
zjb<|fEm>TcCod*gtEb%7LT6ZOXw(}tn@yg5?kOhYF{AMa6&dP{8sju0tOn$|pfH+x
z5MhRflTVx^3PY!V?d5=`=Md#(-EcHOoqmu1J0)GWo%Q#e1Z#QqmDd?`2h^%HPM$i&
z{ku2N*#s3x&%cH&OUd(Wn!-ku!omXM`Q%o7``%tw*S}M%zq|BnO)>0Btj&=sK?sc`
zLSR^HC_eU~XJ|HJq;G1Clt@*eb3-Y}>UcR`Ntq*BYf?_($hV<+KA$Xc&2m5Z(H{b9
z`QmT<Q%u@nac+)Cgv6>sC8;50Oc7~hbDnXpOSiX$EHvq4Odut)&{P6RZNAC;T#Ln}
zMdlY8495fFxJskyqIqQtvdM(gr_b}pKK)Y!VS+Z<+X5B0hK-N(j}&|e=eWA$UI~Rz
zl1jBk?#q=w{Gkuf?Hn>19PrE&&+z6qE;C3+v}y&?2(+|h#fZ_cjS8c;$&?6HZkCSM
z=1AIwi|9q!-NNz+F7Q+X#lW}%ktmG4dVGkmmbJLXN1lC_dK@8)1utDA$)^y#@MLNy
zWicb1I@>a{x&JEv-U{o33BhyU_kCdD>%aEfnBIW-#ynvV6UB9sYJ!MjSCr5cOa>$R
zd)r7YnM{VRtAmjwGNf7$X*L>+=jUkEtBi(yqNGACP8g2HC}YVdDaTKp=9546X;d7$
z^wLphvUUo;bF|3Ew>2wMXl2jXxKg3;8}WOeeU^T=N80ak;ru0TT)oY3oKi0e1R>TM
ziph{kzlVwf7r^<7>Z!9E1E#>@FB8Vu+*7ce{J}$xMk8R@k*-)GtRO^zBu1h+-VFJn
zk3G*^Gr?gF?sF=YyL6>G>gR^CMU)Hdm=dX7pjsjhZXvjKm#!_+$A0jSgSC9_m;MoU
z+-7BNfk;FIs)9)BNR<%81(mr)igufEX9JZR((%Yiixiq_plB?#Se&1uzp}#OLX-Ze
zN0QWN)+%&6eU#OtqY>xNU*ac!>SqX|3R<V%;^%gDVBy>M_*=XYzUu=nB_;wvwbnpL
zMW_Vd|M3sgIXq<8+vB}YKhJAld4*m&WUiLa{5&5p92}s+7-Ka>QMhVrHM`_a{akCT
zDG<W=VAAp^D0TCVfV~G2Q%s1H2GZsr5<;nYX0guCe&&Z*XvPSgL0EA)Vx2)|uu36>
z1)F<YdM@4c=Jxx)zf_9v`-z`I8u;wr`_D0*9ZoK<A!H0hGbBw0P+xHGJvwmwy=;U+
zL0TZBLd`Fs<`#*LAE&mqi0vOztv6Y$*VtEw1UhHj>+|Gu@8?hahktpt?>t1PbacHu
z_2gNE5KMZDeBvWN$mc)%dmOgAtP~ki3#_sf*}(a(3j77`onOd=aMjTdJ7gje6N|Qm
z<LIIZP*@C=X26A~PZL)n5T!V{VA=`BXAZO%=$*=LTZcn24LoPQG98{LCWYdOvrhw-
z%h$d}W=AyYRiZ$V1R-G*6Nu1j#MQ{M5uI@tDGZY$bD1Dv2_r!mSEyF2)Y3XhwZb@^
z5C#!(5|T|O2;np*R~A?Jz;iF4WQazigqWtmo#vjMUf|2;2hbLT=iZh7=FUU+J!j6H
zW;7Zw%7>gfcAVXf2V|2BQ*jr64Wv3_l#fs%@Z3DVw|luWpQm-oK%Sd@G-!UO%ky0Y
zb?<raIeuqrACnb`$Zc*4LseMT@_oe41A_CXQL*t0fp0;L>=Z_xk`v)u;M?!~@B){H
zih`Jreey?WRRnwY-(+E79>Dd2%c0V!k>!1gbU=|036xVT7Z#)y=psW;hQz9dl>w?!
zA+OeGw-3;{rbsnL!z8nebIV`+%l}K3*VbH>kSX~~GjwdK6(?uGiY-5bM<cZ{NZ}5(
zUT;7c2Gna+dYuFE%;ofrlne%jVoss0!B}@__Xj;1tp&i#4irxMq3mo>K0_H?9F;gw
zk6^cd^6V*gcV8t>HKDgT+EOLr#3VuNj!`EiDslmb3(BXnKnYe)rApMyp1<EHQ;%U3
z@xsSGN<%2xH}BADwE!scjDiYDB_SV<(BlbuG(rhUk){X~LMU{WqsJ*xwTe;_mDDII
z6T0m~v@vK6MkrEiNwNItKmTW0SXywKF0PZrOk{ILbam-<dt@VX+==qfju!!g;gCSN
zb~b}fk3272x>Q<*o*(7N$?nh|4hGccnt-n;EefQ6-*R59owcU3+M{0Zjs+^e*Z9B-
zPw;E^cQIKGaoq)N7-9j($GeCJcL|<2M-a!;=Mg@toZV_YLT9YI;Q7D%c0U(Ke(&|N
zlp;zJe&|Pkg1JPnd+RbwOA9WYlc%TxM6Hg^dob!F@&Ta|6qAwnb4ih<6r%xgy@65z
zbJZ$&y}`lZ9y%|`Ck3V`{T%+<|Mq{!+VSJg&!xpEGfl5^1QC}?O-anQ{7~K(wKYiT
z>@*stgi%1FUZ;EL{9GH0k&52fP#8twQCDp=y>6H0!jeM-wWi43#mML*6?sQ&Zan&l
z;E|-_KJ|moasAEvj0YJ>?6{3e8(K2t(x8L5b{pRJ9;j&7=1{mx`xJ~T?drMJ$dK6{
z@Xm%(0U!Q}KgDvQIQZsQSUq;kzixrcV;w>Mtaj<%Yy?PzmDuqJJM1Bvb07jlr2)-(
z^wtCPBqtl?Xe}6KmTm$6@&EnrGQYaYJ3cMn!4QI}V#Cztf#hIRkT(Jf422al`76I4
z7R5W)ofCpb(m7`3<TAH9U5qY}LFAD@O&D65ODRHhP;v>R(@ZV1D6=5LBXZVK!Za}3
z+h@z-$p8KSTy{Ya2N6$Qe1@n32mLK-^%?+$DF_V&VMJO?C`idnhLi$ra<7vqT;Yj!
z?LUQ7C>c}el)-S|Qkll3E=^%EIeg+1KR~n5a&7knZ#&Cjw#SMi;6wS()@|*MGwuIw
zJf5JGB93E5<H3}+VXPpZ7$PIxK4P5DP9|fjNo{&oOzB)l_VzMP62hX(e&X-K&)?O`
zxbfB@*aTq=7==X==?QDn&`ROnSHH>0Gv{etJcEr4(w0RLuA7AN?L*25I*M#f(@5pt
zG2@FI)o2Q&BCO8y+=qXF;^G;yQO|STGDJ}z(CBfWy?Zx_s})CF7Xl>$A{DZ^yTMK^
z;oP|k?uIJ_rYIN>M@&W|iY&ugsMJ>ZnV<V#@$_>qfDkCF{EhhS!BlCKnU0Mvm&XxJ
zDVJhOl7y%}j})#QQoUAp4S}_$5SBC*6k58LQdVNLW-^{ot(V0T90`hNf1dT?W9oQE
z=lWsBt=%>@h}{CjRj8^_!ErrL)3n&Uyv_Ql6ND$4SXCfxFii&<V`f3d6r2;YTimzU
zefoVKaxMjal?+gg22XwDgXAYpQ4B_I?(-Z`ScEa?;h6T_E!PuKxGxzf!XT!-zr}u%
zaPssiJX;f^4U=(7HqIP}%|f!c%x8Y~&vW5D&r-_m$RL;=>QkE^p$$w$A+1kOlzm!V
zQjs`{iCQ(6GDtJ3)vDKED=5%pLQqKQWiwpmS(>I)roCp%)Ib?bx`~sb)Q6gBlu0Sy
zF+~5p#+$G2V{HXd7_4@Fp6D?rrC=db-2T!xICJq5QR4zu8PpL!H^%8$y^Tpxeg<#P
zx7pm5H)GE)Q3|P)^Yg|cFMRUH$)CDFHt2eB<s6mgSfeS1U3xd)BB@oKpG$!fA#o70
zyS2quy~?>KE}<m8ty{sUKVUo<QDiCBLbb8RU;1nR2cG-j$Go3QqzL_U*=hRNU82lX
zZC?8}XHMzeoy=A{|BI7|s9r+|(7C43Xb?w^WS&dQBo*YwUEZ{nt}1Ubrdkv}O)aLI
zzV#YfLWrpdHb>O&AC;ef>7`w?t;5)$3)kAKo{xBr8mnoA*FXCeF1+vpwN?!wiYfKP
zD&;D?!?IP`5zV}J`&dGt!iWn$@)Ml@@C(?3eUD7%4xs~>jN!q3SZo2(T@r1831ZS)
zw}_h!<ntdyl!Yc)M%wN$==8|O6Rd^W@)`ch-}$e2@`dkvBy;!OMij-QMMkD9!$Okc
zgF2y{c<@pAsdw=lQau`FgTdB-&^kJiG%m-Y8PjZPssS8q-)3Q9nK+2hRuW9p{Q`TW
zb5W_qHY=2HWBS|jJjZkK<jnt-lqgx{+^MHn)63+=*aIONfyN3;VJGzd|HS=gkY!1F
zo(VqgM_d&*w2ss|v$C|7-PIZmjR8Rp0fGWB+ANL6F`HeZt=ZZd$G9_pcDZJ=^Mhu?
z<w8+_%K<?`HySOfySlrotFmNPY8hI`)#@A<`@{X*8(Ccds#poEw`FHmX2gwq&*z`}
zC*JRU8=C}S?1-lkNEr}^0hMZnT0LTVdd8oX&R^5Chf;~Vk1Y(v!+iGH&vS78n4ctd
zbS(R+OZLN=?!9NN=P1p%2i%IHh+H&;08*Wh%jJl|0HqDefJp^-93|FGw5Caljz0>Q
zZ~J$jdydARm+ncr!LpBM*H0!BR&9gS1jaF$hbrZONSF%(0vl6Z+T+#_e#nE<qlEFa
zlR&Z%dd{|N$qYno2Uv_u@6&OZWe30XWSH)<ltF+DBa95ll_m(SW_+~C&iWFqb^~Fd
zUfp7C=??iok_#iW5yT}Y5t1g3rT_NrTg17TnfU{bOgBmBbUI)yhZm-poQ?Qz|G_`z
z?1PWvo}?K>>0+M}h6Q6hNwCKVWjdYK4gwWr6xvwVj+OyfcVXKsD>tdru}ciS#EA!M
z>1Yd*UYDfX#%iZ*5C#J!iL9!<|Cf}4z&Z)0;67mM8tonujY)mL+SMG;BP;}?Nb)o`
zQkLFX<n-`Kf{8qmz_rkV&=Gou;ULQpeoD?CVtwcK<ho>+z03kS|Ez-mBOw|YCeW6A
zE1|N!NvG9BAZhOIv9+>BEJC7a42>n9i;$s2Yj+_pFD(&80h6;ch#*8Kif*q92o}aC
z7!N1;$-ncrIez{D4^*{TLEOEUK1Gb^lfGsS9|QtR*dJa%fL19QCv+}^%hsH@v(r7*
z+6}<;tR?XZD^7zD(j+16_AuHIAidV>fS@!BhB8H1XLDCIl=tD1+TCk|RQLG#g3ufY
zVgf?;7AstP<p(@AJ5Dq=OCSW=3PNx5EJq*sNzng0$9E__{xtJ-JnmHV{XB?b=jY`q
zBBPlYtFygwhgP#j$ym(p2CKJjPzWUX$oY9tE+WE^BvI&8apU?m;(WyX!6OiasBV{b
z+xhvigEP#`$NYo;{{PHFk3S2Zk<0E2&ZP`owc9#Bb$Yfg(F{De+3vf@wDf+y?=Z%H
zjn>+6_*qLQF?18xPSqB)vh+G#l1>|=TuUelLto_-`x0PF2CtA(68JrOKm2@aw@w(P
z$i(S$=hBq3BG0*~L}c@9T;AjT|M+J-JwJ*(w%}UB7DO013~N7n4+a6sf6ZPL0ukj9
zV++{QBrIKrR>d*&cM{CyOD>2m6fj8Sa1O$Vw40)Piuzj@7%ArA)HwtJG~2Z54KS96
zP8{Ok$vnUOKm7|%KlRLCX+Q^{wIWeHj5Rb;>$Iq?<GM_hCBx!sm40H_#QlIxC+!p?
zEY|v}u<bE4oG>yfk;>BEt+2IrllfyMq#SnluytpR$&xV`oZDn=#C<qf;Z>al`~Hr;
z&%3ixk_$>mX%M9zwdO9}Ue|4(PMy8#Hc=!gg#p$G!Z>n815dKQwY`N514c*2kpvi{
zNRkABWo)v{a5><`XTHF}`6CQkX!us@0Qw{FWcz-2A|qb5I3<b9(Y49QpEDMOOy^N}
z1qu(9NG(P=epAPhZJqj%N_<f4u(MfWyFVuOFkbhZ0)f`{V`Jq%nW6+BZHP7rVRM5+
zdXodOo9q!1tTJr8bD4>SSq^^YdGt^j;Scvj`RO=Mq$(W~;2<hyP>su~b6FKix=)4(
zLdVM_1P~Dn&C+ac(ba-_y-v5;V0V3sVtI(M`8m>VN~Ka`r&(inXNQfAHIzyiE*Gin
z?h?g0qBuvjx=pLwATAVm`q|HL^z37tc=#!_u?Pd!OMAFOAR~Eo4?PS**IH#j1`KXI
z8C=$YaV2*dxPZrEJf(s4S-(PSL!u!`Es-5m2Xv}YNw+VY?D+nKziccR;qhmKpiUak
zNPiz3-woV_SdpUED;%tcoCpt*n4WV21>0A48J!+x?(qYdAtxs$rFE0Lc0e#K1J|<2
zz`j=OyWG9fK&C7r!L?UHHw7z%fFMFnjL{ZVdcx9Z)V*3njY6rw$jl^4Y3hvzm1c#?
zUWJ{lO;pljs92<4sX3fR98=%hrPHhv=0YBM>S+!<aE8MVpY;_{H?5paV`q%-*&z@H
zq0_6~S23~v^8o#_Kf)pdKlCB8eQtds;B1~^NG*goVsB#|`?%_*{!+JnRRfm@IZrg!
z`l);Crbcm6boV2ua}*k(8<c3Qaa?V2Bo1inz;WuOmen^eF?Dc`BVYaudUzCR9koVJ
zr9a~5f}kIm*sSs%43fBPe*_XI1|IJ7bCDw$U7%T$bhMyatJ100*<M|zR4y}d=pbn)
zVQ;U3YSh@?++=li8KruR3=grlwM`rsh;ljh_BLsDYQ&`iFMZ}qoOt*d&OY@5%COJR
z?ZD5C^*BF>!oWi;9EQjdP5T7S+dMJuqRwQL!1<){5DcYK$AhS$msny0Mmo5MP8BNY
zV6`EW(GtRQ(OKh$aQp37M{{F^^O4|w_&FeKjIv!yn|FCKsqtujf`;lK0x&(xtv`L2
z`BMiu{I|aZQ?WaHWB_Rov_gb2{Yj!vvxERbu!(Es?rdVk4(atf*s(F<vuB~zMQ>~(
zD_iuIme{#|6_s=tA0I=ktUz%HibeFsBK5`&`H?a|`?r3Xi6@?8^r6T9$^$xZ+><Iz
zNK-}ESlX#2Z(J2h8{Cu(IGcY=RS;Ljxj$3xSO<5|R?<ke2!|@n%?MKE0u-w?waq1l
zhDRBlI)yO>PoX79y*yo3DbA+X=@&D6V3#H645nmV(Mf^`u$eLq0XZ>3x7R?Wf>x(Z
zl6Gh{>*R8IiltGsR<zq4>dg+#W`jz#iZY5qAxFE_Bn%?LAf(Z5(CKxNLCmS+r<tER
z!rb&RcLw>O(>gvavgl6OYycwg_xYfM;4eWJ3;GJaVbB@LU>MfAsafr@a$33jQ3iN5
z310*<Df{{W{eX9$ghXa+A8b1IkdJHfPZZQiD+#P7NU9tVdz_Lraw1^OLV!S7gKcXT
zzxzW*W+oYX_+exedX2d3Vv_w$kPS-g59;=}X8NSHE?Pl`F<}(-t2kB&WIT`ETqMe+
zl;S*Q)>Sq--ITa6#IebFu!4!PO(rMCsZ@5H4s2>@)EmTwF1xi|v{4*7@c_q9J;C(C
zah!li4~yw%RKhyIGPsC7!>nb2=%Ae_y>yfI5w;{C@YACL%pj2IlQ|koY6R`XQVyJ9
zwKg7KqweXyOW%g+m%pt+?Q0Lpeedg2FDo!TgQ&GQ(kt-*ljKCqmUTL(8innaSbb}q
z^7s%VXQmLLa17%9J@xJCsV=e{FB^)v$L3<+*GPc~1E;9qvpi&ooEgKetrIChF)m?+
zljQ0tMN}+saC!=aWNc@L@zGIgl|59dP!^iaCQ%ent?pp7V*cbY4n6cRlLrr?Fo-m9
zmpPtqQ1pAAzBT6TEPd-o^c~VUx6uZT%fgWR<nI`a|6Ax+`cf=it0)m5Y?diH99_S?
zPc}sF+q=e1iCUc<3!bpn+@GK>KoA;(NcK2Rm2<K|UWTkFk5|KB+lJfUe2uZWDJGtJ
z92rFz;c;I3s&w}xZ}^}tkoVX-lL@o}kO4Bv5r(lF{*V$Ya>)1~cJ&r<lroenVCE%3
z5YX-<#HBI^XAYvQV02=Q>8VL}cehA;j?1!EtCB0`*r{xzjpFFpM>uu<dFBtFMhS@0
zuG3?*?vNMORaWtKauiEG==7}neTac?6=~}+tRm=hPXYZZfW>;*KTBc-?bH%V5dIQS
zT04BL4*;$8$|@O*!x*d-uC#|Qj7i6ocR&2x>VQxwN}U~^BjLrsPzrKfPr8JJBxr0M
zF8zn^Fm_~~;V(Unh&?ZzG5&(wAJBcueJBLv3Q#IT48$eE69Fa+QLQdvd6am32CEgh
zi7Dn5<}qu_h;|2^7=oQ`Y&lPRa|NXm#vguy>GMx9G<)C^5)J>?FKZoE&KOG<K|^U`
zK$uLkSG(@9HT|KwzjS4B$bE?91z>GT*VHJCHJLhKh(bZNX57T30n(J+wHp*l!xYK~
zUG<Ye%Q%BoIMQJ^JRy9AI-6|o|Jet14yxkRW&;`cAkTeWL5@<hJA{EIpNqi5DpaC~
z!aOrmV+aI|p&CQwA)3vGml%X@w}T8q>dhv~7^Wu{I5d5f@$xjrfEH;#UdqZ~eT4|D
z*VxShPWL6!pWPYk1~X9fad(${?|o_)=)-f20+m>TJb>+=?HTSaJ7eyF%w#d7d&)i#
z_3!IrPCkC3-7RfEjt+;};8@fqlA$|<aTkFR(6f;4RJrx-@6%mfCnIALIB=+xt~Dov
zkRS;51#<hV_Jfw3kkW;^GDL(CK^PH5?t29g2}2l2!Z=4^Y?Os#$FNrL?rU#R+p4g$
zwo7%rO0TKuHWby(I&XacEt>5vQ!`V<IfoYrf*i-sJkH#)vk*tF?q;(Vu<XCr{k&`0
z`wW7>aXSTrUsAH~b7h+#tF&Z|83vqKSQLVGV$j*9@D-gj9Ypp)aBuLNm2k7l$o_!%
zz8v-@AnYcb?BqEUOcKf11uH;6fDwkoDpW<WbYY3!cDJ9|2;p&czJ;1iI|?ED1nS;r
zeDy>2Te4X|Bt!RXqbMW@oy|id2!$kwL-NCUW)96@kzBrbje4z3rP`ui@6po%Jtb%~
z6W+Ucg?6vY_{2DI>?&<C&T;7cgG?Pe<S1e?Y3nTW^zHJDLINTB?>PvBx3j~51>O&`
zbE}_}WSLxM^uEpA#jS;<V>Q}lQd2&VS7|?!%K{(o*}i)1_vAjCXIb&5UDAf+X>-zU
zaWd`^2ToJj)lZy^P|rfLUE|u<ze{InZQ$oZIuc?Tct6klL3l-!4`uR!n+$h;9=f1C
zj0oe9D2@q3CqvZt^L(EC#29mDPNA*fjqkq3&iXEU%iA<p_UJV<oq9@Ty~_7qd5uP^
z&Gh^nxqJ?dBna}Hdhl5eoIVe^&{5V{cX$VKpH2HdCi^}d1R<e6w}Q~^$E=ee<UX5!
z$OZ#HcY(S_(E12JPtuf(@+SyHKlt6h7lR<|pFi?G`Z+pe#I|_R)Ok7|kPmXs7C2wi
z7+M&5%QZgucYjKHW8Dcy7+;}IGob$<p#9V<tA3Y?h>=O;v9ol0hPB`QL+s)zY*Y|?
zJ?KhoQ&U^s<7>b3N37HvaQLum(TadDS7Q2sCnzr*a-`;eP0JWVsvPHOilCM{)e&o~
zQ;<m$UzMeQ!GElk1xTNANeck)KoGx5bjVNCnV7?2=4^kC6$7NoqPk7C)~=&_HG~*I
zKr&oird}lW-Q|P#ItWVe{^Y|YM^fHDAABGcxS~Z6ghX+X(XlCn6|5}XrPprJZ8S;S
zEktS%sV41oSh{_eq?=GK4--Zq7(paM4o)9od}IzQf__stqi&JDtC9Wg-m^QCWFS4(
zFN57=1qk}vTl#b1!)w1ztF@jZ&kPDK+4(+z&<)gM`b8l*h!;fvxrvYMR-dR+H(H7~
zb!d*@@(tp|Vr@W-i|Vx$2rWnu7#*{FbCtWVz0R4DuOLc8Zt^rq5d{HWH&zIc7_b_H
zBm<Xctt{)?G*Y-APyj|LdR}s)m-HM+o&asTXsap4d1enDAqXSlLYav}r-*|*Drr+#
zU!t8fc>Jk{Id=X5hG*w#-MNXeiXhB0GB)E_omB#!>~jz(Ws$P&1DLJw+;noEFp7Wh
zgDRRt8ZW7`f3N!;Vev9CeiL+3Nt&~c?!Y7%t4LFY(i$NYQYfT|FrNG-%jl)>{BSbc
zREGO;))!>Jqeo6Kdc97R#8?|TvPuH9a74H%3TsMKms+g8bC+YEIg2PpNTD#ww==`g
zUEqLJ)??Ou%`OO>LPC%Q!q#bA29D@bD}&MwCy^$Jt6~NM-Rq%^q7avvJTQlhLgG?^
z(W5g&IVS~_UyJFsYMeiNii0N)Q=S>6v)IO{lpvR*JUNaL(&K_$Ye85~TY)=d`ek?H
z`TBeYBE-EHxR6dO(K)mP5%$$ZvO~ph3SZH3f8P}XWi?U+n6!u03gxS`LMx=!2oYe|
zXLITIl?>ce_C8J0PrJs4A|5(=nDG7E#9a#}B$jSSKudR#_W+~wY+qmF)+?`ae(a~9
zJd88~lPD+(5YdPHTsd5!b=-GWN=z`o*vTMres0}>fl@uX-5$MO7pz8$5N$dbtBK+~
z^G8n*M=_!}%*e5GAoG-xHpSIDw0m`)ec?$?KKcmd0|#keyXuB4;sWDS^H>`=eOK+y
z9_ibzLFVVuAJEo-(C$#rlu;NQurdhx`@>-SK<>e&$V@(<yD?VKP9;ffky@a153N(u
zBt>b35SjpmKw^ALM)*pC5n`X84{*Gnc1@S^{KOX@qx{_p`E~#%q7XP8S|#24>{x?V
zWmex>X6%3e8fX6TFCoT<5w-{29^^~^)jvOjv}mQ#O7|1loPbDo5kBE6ojgOSTtX1R
z^n^b^iv(ju^3^S#`Kd25@wv|+CKoVw>Yx)$TqHj>`Kh1LC;v=Y@&*P$T}e8zB{YIG
zX<<|k%n+$oxcSk)-WB963x~!qxjKPSpe<qE2W&naRSIm+(AwEzr+AaOg@*`ij!+0P
znaB)BJ-7f(Kaq8sui!B>E@+Yg9$<sj8l_zD>n4CL)<{y5qKqLQM~se463Bok%ri7J
zOB}{1-6M#1Y4y4sIC_Z5sRas!F?zjCwAF+%VyG~JFam@Ij4#sA51umTwD|j+OchFa
z=#uku>yy5W06t3qeJ!5-zgd44_!q{cj#5zyhpjibJ)w<33xkpF-1oG2t}^eUMvU<s
zbqt8;KhF9@LEV>r`aBPw8RP2z-eKu`7YMcX2ez+9SWs4B35j}^wbw5(ws4TqCm%!x
zF-E%y%0PzhlJ^0VQJxFT3a?a=Meq?p5Rn?g-d2sBwOzJ%HrU<Urdp}cXf^1iDRZL%
zKk<c!C=@&7b43uAk>N5)-JtdwXjP-NyGyOIg-I+U<C7FiCA#e{J9~Tdv}I)a2vM#`
znm8fvtlS#xuYP$8TjBNJ>|jDB+YJ8p?*Dye=zhJ%)x7&z{UA$oo4?yLBzZwDvi%mF
zQYpr02UOh`WDX{1vI)6t`f?u^+^1Y$eCjlh9+>0K*RHbm)*4{~LRfzdx}Tw~!T?c{
zXY*o(;khl!r_0DV^3)yfLS$dt&1BD_kD&})_7en%AV66~Ww*}OR)yN$4)tn<dc97o
z-6BybbGZVaKX;aVw@WVH10>~AiPS{smPI!@bgC^HyKM{s<*`wUg#x{9N~KyMv5NA-
z0m6dQ{>*48vNmB>oe|b!7p)7vO`oUfL(kU-=KWSJ8PK->1$~=aK|gx~=vhk&MAkU)
zueH8yW03y+NPi&q?P|Qut?eH`8MyY-uAli^kMYF$Nj~_^9=E^sI&shIo#F<cB<_H+
zGA8OumcD<P@xw<LfBq3<kaHKAaMg9fupb0FPH0!-6@Hk;Ri9i1J~5i=MwN}_Ew(n-
z*xA`+Z*PxUqeeGPI5-;e^Iv(4VyQ!%?*W#P;bFRUgWahkRRy)ZLt|$hlUhb6rYH>!
zku=+E?QGLAma(~GMEMeF()A4P*@3xlAjXqBTPqL-jMe_&|Ioqe_sf3D5q`w&#9%r$
z+ZzHpJwuWc<RZ^KXDzA6BT8v})it{mv)9ckxA=rD)6f0%R{tme{nLE$*<)P!zXyEq
z+utUqTp$}F(Ky>DriTg2<XbT}e*b-@9(a?<AA1Di0_aarY4R&;eYKlicGhd`tZ%Zl
zvCj6^HhYy_n!OIE$6|i=Z+woZ)Pf-A0_yQeXr=CPNQr0z8+A%bP+B;MD3x8nzPU=<
zYQh622n)qe{fs{O=W}Pfb5Bb%wRA%$1|D)_4JJ*!{Xfa`yFX|9!N-m=d31#J<&fRg
z8w7#FZf4M3A+mylAkZn*)jO0*<HY5Ij%&?WSJ@M;S0@E-%G_jyC(*BtAtc%VjJC9!
zO`4S|^+t_Gt4X`vqSJHIe5LV_Q;!`Y7pJI5A+Qt*1*9?v-9_mRNw-PTt|P3aSQsLg
z%c0VQMzcwp!f0`lD2P3W<bZ~%-xAKwclWP2#YOG0zy1=>B09L&&;9odtYP~Vw(M)O
zuhS`51vVoQ&)`=6^#<=}aCY>6lDo<g=95)sKT%Nk^t0y?VV4sxe}>()b<!Jm5yH`o
zXkR^0Mx#OqZHTRRx&572IX^Q@{?IXO5c|VEa2b3i5rjZma3$H0APNv+fUptUn-%Wf
zUSw(U4!3UK;KuD6Y;JGSY&J<$0#<N*Zj5K2SYUWKrjRQV2Qe}Z$Pb68ZjD~IMXEXw
zD59Yp`QZ}Q7`9f|sWlpO4LtO~lcWX&Nx!LmPZcKH0R3Q1+_PCMgFwp|<0oWH|G$HM
ztm}Ooq%S!eKM4cbumpsW2VH9I@fQ}y-Rl6U{h&iQfW1?+u!Kkymi<BHr&*6bc!s=C
z9De2))y)cOF(Jc3XbgdnXsggd5!yVqVOjp+ZBCU>kUKbviJZh6QaYQ8eKz;mkB}iU
z5Xewqgk*E4%F6Nz>#M6QFE6sRbeG+|9op@-3vdw}8J*<u`5A`7fPB6{6z7ov<U)aJ
zG)X!gR9AuO5=NF>F~nF)b$gdavrZ4qnTH;5#K38a@1JF6&5WMK-~%W2zUP8JhbwV>
zY^D#8HGP}A@5x?cu#ZH)Ee)<UlK=s(GGzLbmVT@~E5#$cT#V=Z(;8uopTxAD3-Qz2
z{4<|@3=y<B^JAZ9du4_6%3Y*zG$R^IAOe(8DB=CQ+T;4yzs;jF!xT=ObQCyVI!a`6
zS5mrA&Xs17G9nBEL=+;3*;w1**7aN5zI~JH*ROK<#%0zwHfS{JBx!=Rf-?&feCDZx
zjEv?e<cq{XjvyaW6aqRGdYvXJX+w|_m2(uv%2;FATw0;hsL?U-#G}tUKQ|xtb33q)
zkWNLV|0unm8<)@tDX`kvN?YwG2ffWd6zmHIxE$@_Ct&Dl=%o-wgJCI^Di1y%(D?{q
z`|nBH1gu61L+Gg!`kbGiay|Fb5pv^Y&i(zLVCDKPk{>Q2MMxwgu!cZJq&gwZ8Df*e
zZY5m&%|GDT*#h~;&O<(*<+C4aS!-!G5?1eQaP7(!-hS%>KYZ&o?kwJ6ud+wC*TERU
zQ%4T)<rhyeJXS!IOCa)27<brB)zfZ4s|%(LV|kbvb5Y6a5=mu;s<j+`^y&K&`2N={
zrIh0~#L&@}juOO?!%?QGB2H3IkMl|2->3KU11An*V$00JS(=S1y0h7D2O8t1s*Sa1
zE0Ee^yKT1Ct}=hTNDxiANX0sAyqjcq`&>$wj9}e))o;yeL95naXM3CNoo%+Zx7gU;
zqF!&%?e+R?$%&Z~bEhT<3n95Ec9Y;zB4SBmIwWZiV-ge^f`}l>fweU1PXAS<uyF9G
z*O_+cco$gjyU+Xg^}T0icD83~82JCz{g&#7@6|yt*6+S~zE+FH`e6xz3~I5T1}%FZ
zgO;@`Cb_f4oi+QECO)RkKT%LOTppp_F+?*{ocO}?+}zy2HoJtrb&8NEtMSE7ERyEZ
z7K^X^fFnOOh8UZ48M+K{9S&ESv!Os~M-?D~(1Ad0#OmE`uHLx9dl%p1yWjgZ8{4ZW
ztp|554%gL96H28rje3(f$T3_RqmV0MB8?<ONl9W{#bc~Eim{rF&2=i3Jt~bVhmJkW
z*w_@>xOQqFL%<u-S<4ytL)AA!CIRAFh8aLo`?2WEfcrx5{`Xo26{~Eb($9s`)sQ!!
zw4a(Y3S$+n!Xf;#a~o1fSFH&ED+SW{VAD%jecDwj717-`;Y5zZPaj}$_cpez32-<C
zq#!W~+7*^TKx-pm^{o{ahGWEd9z-_m;7au(kQian9&GFQ$t;S1mDOGD-nqr~8&`Pi
z!VlTq+58a@u^^zA_9zrfwAvk_dYz$>kbI%wx$zt*HnClVNEjI|qK#%}Yl~W~O07|4
z{>&+cCng;JOE|R>tVLryBvAA#Ha<0T6*lPycZ5J-`mOlBy)yu8mTLKw*KZSq@RQUY
ziz58%Cf1NcpuJOTr7&6(W*AODTCX5tEW(;ThR+HIJpMQuk{@Rc4-eC6YvQ>n&ivSm
zTwPnoG<vS(VgMO9Zbe+9#Ug0j-RAb6euv{@<A|vlz*XP|!!IE=k|<+c)*gfq39JoS
zy1l^%m)_^XI~VxIx4y>e#xhE)5BYgWC+RUXG)%oxCl}-x8JZx<mmt<yA_5^vv}3=Y
zD2=0yVQp=d-Mt-lYZZ>2d6My|S+vmv5`r*t?fvXb@snos4|-RvGX!8ABt&3c<<45P
z)y~&FwF20W=9&<Gs#Z$EAaH&z1RV_pyWh`Kv{6LXDX&Ng0d6?U@g`ab-GwQQ!zNj6
z?u(y?u{&4dqcc45cfZKnw|)~{Ns0W>7gEwPJyKlT$O^&Ul`XFS_BS{`JB2(n?^_ao
z`FpX}vbWXd#+BQA^V@&Mzx(aq;P#!XBr3fpFbs%jc3Utyf!*GNpol2VfQ;!Mrm=;n
z2StN8Jd4!|d-o=0eU<H<O-9c@MR8*4zWBu_T^i%00>hBN;t1xAWynIJEy`+A)kEv#
zzk)RU(|?|PAxBU32%-_D51nCav5rl<NIB>vWP`a@xI3#?-(zp{Iuo-+1ZAJJI0cam
z|7or8lpnbJ9f{PE`gVo2&2^SmmbrTU8kI)<9;!2N_ga!_^0_=ow?{i{F;pm#i(;ST
z252ov6r@6sD-|%tQms{Kwpuh>ZRREqF;p6Mr2hy%LB6kw-VdVo1xeWfZT%VVD{1?A
zNwb9A^<#h34jF!0c!)$+=+i04NH~c&kEwI9hV<vRdl?tV;p8Cq1;*J=`M7)d6YV4s
z2uINnNJdW_W%h}OureTUEnr8?>8?7259`8IuzB%1Tkn5>Zq!h{#J9EFL$m}SLS&R9
zh;zs&f*@vZr@_@L*ZAhQzs~Rc{_n87u}Zo>pcBG_)h$*i9CcBN_04t9CM^lbQ7nv5
zDvVGnjZz*SCyMe^D-G7x*V(S@(r9;?nLpt<sj>`tVCbyYvJdXwH?%Qs(`(N*V)}n?
z2EXh8UtnyOk@>2NaN3w|*g_J90aLTnB-+1jKWf+gEXEBJxb};KYi0O3;XviV0Oxf-
z&M_C{N<;v|2ZxzFH-Qy`z>ly?T#(|fcDF*6W9L?dy&GGYMh~4he2{yJ9{el=WE2p@
z0WuO0O7^M^Zr;4j>#u*0Z+!D>Y;AA+NI(||K_~!|BC+I(CAN3AXt!Hf86u+sh4Ki6
z@(9J@5lSPYgt<J8W{a(@P4+4~w7Oj;4<E*nSxbZn$hzm=?1P7AvvRn1aNz<OLpFUr
z7;rGYVrR^tP;dHG6mNGi&~J3D$AIyPF%n@g!Uc7%%Bx*lrr#zQ@Y3ObjuegqQp$`@
z;Qq8<Ge0MgjGsNu{4)<j5D@wnrSo&Q-+j1Blm@K5cb$#5-bGg{s9x988~|4K+p#i^
z334$qjv$EG-l}l%y$|@Kul)i4_CNj>cUPA@#>l@HA)E?GVETSuES6bWTcy!#U`+tz
z5v4Kmr7=oF<BW_<5=8|nl`6|C%WUjyQEzpaKXjTvIug~avM_LUUp4M;fPqhIBL<h7
z|J(+%FQBvD&)tE)zqOuy26x!I;g&EAn4X`dXFzFzHcpMgDM+}_o&m9?A1=xKTu8@Z
z=di}^*0?V=&-Qy1F#Y&Rj(zoU2m)d$3B126EXKM{k+6{K=D7Xc%PfEEP4woDYcc#K
z%UVlir^VIx?(*xu@vr!w|HVJ&>dnjad_b4ImoSh-fyI)b1F)qr^qrg7b`^gRx?j-{
zj84Jy0yaN{+1#Oh^E#Vr%k0z}3?Df0|CfLcd_E@yVHgodA%~A2p=|}q2$XR+VPzCr
z@0-f}SnK|@+=Y<N&xIgAG{wl|Az=StfXLW&`Ugv3H1(Y|nza>-X*)HF?Cn?_9KcGP
z#AAe%AyNc%8g16s*7)GcW!`-I9rkLq9|`CjZdG7}YXj$^9F=N~ZnuZDl1N4r;{wH8
zfnu&mDPJNEV!G`Pl}eRbqe-uuGC4f$hb?}zZJ8a%_uTRQ1P;$9$318oYcd@DptZ{W
zKQm}GTc5;d1588ZVSER;U**fF%rYCWFR-)OT`Jhe=PCMT|2XRtRq8U_iIpJYJTuQd
zLt|x!*7dt6ZE<i=f2jG3D72tkDeJFapg1#0d3*|CN09{BAoQ;*2!qI#sjWqI4L7dd
z;+^;2;Fa%vo1RKQSi&%1Zf3?QKPZI|t|FDoIq8t8=~+Y|Sy^8p3iC{jP4^#i5I9<l
z-JM-l)>f!D8zf3o93E$QbcQ&Nd?cQ=+8l1men>n$lLNaKGiR_KT;J!H%M=0no5|dB
z*k}J1WE3wlzzP_jnq;+FaoM|4PVZF@s(m8c2x$NF1}YX(N{kkL>_CR)`7|r5w19zJ
z%;cl<w6;>(cX!eI;w@)$_x?f~qB{{=?{APFEl?U8MU40$Hpq%s0_iU&tfiM&Zrxtu
z+LiZt>#f&ODg`WIAeow&q}%PGRq85R#t;iIDrI775+Pw@W0fdZU~FQ#|D*yB!mn1V
zY;3I4Y&A%=qA)T{X=2iq^=$_0_1s`tsoS4qS@xN=*s{-Q@GbR>SlBvJ;epKu8Q$O_
z(YMpU_AoXv#(J%Vwn1OtcfWkh{;6i+_s)RcMq+$++|Lm2hs~ug2ge2GUVfhX@*a)L
zw;XLzA9}Fg=0Hoc`rRK=nw?@~Y6f8^yq{;8JdUa)aAeB8R?5{6Zt%v1@ALI<e~n(+
z17V56fQ9*aI;|FInmRw!%KLf3%=`i(lq{|+5=I5e<FkGIVi1OiM6kWR&E2I%s*O5{
zQVfkwGCVO)oQpk<4*J}Zo|zASxC<%ze3C}H0NTMMaBZNBvPi2wHYz)D_fr+@d!nAt
z%L|AL!PLwQOMAO0?eLQhJUuuw`WPKeAJ&mo6f-|Jo?aoi4}Ly)ry7RJ9R2yvvwOY5
z-s@KgJzt-|B`=xHV;dl<f}6khDkF!c7@3(u<T9PsA6?d3(!{WGdxx+8+3)jj{==`+
zNjeA&xm?J}W5;P!YxI(yt7a-iDJLN9B4>_b0*ShHmGER9kuUe3IYEd;pq5wIx_X(N
z>Mor$r8G4|t~_$zp3Ns+LFjP8VHjW~96Wl6%d6`o){vwr$`6H@4<!MAT`Q~pIBoTq
z^7sk5^%}j-hI>!@!296ue*l$G-MK@QE29f#pwv&^vR4jeFt{s~VRLPr+sk*ka^o7x
zUmPNkjE)SGCW+ho!gW1FC{RW*JUopRf}OoR;vizQJluc%?A~lPo9tC8wAx*?Hss?H
zrTnlPwDYk@hHti>M-uiG?W8~$F=&b6am$`Rub<3k=do|T{74(t2fi6hCL53|4V6hM
zEztCd@H5UjvCjq>jxQ?^klDs$iQC6#Hu%Ir9RiF5k%Al@=fF$PP}$ffsn)S}z~_>2
zt=K>iA?U1au=4&zMo*k{3K+qF)5&EFPN%cmX|u7s!{+)X-~ZvOq&jgb3TTcTKFHbA
zCwcFk_t@QS`h(EY9g0!Pk}s5)onJs}!=<YiDHnzqE{_pLg0$OZZDWmFcW+T|H3_3U
z`C^I0NH(@=oG1?yNA9pzR(s7wH-X|N+nfPxzY0S?cyh*YI0O%M8W=shi18CPc3(Tz
zSL>3V1AWI)%+&M@-CCXY?wZ5WX@fSZf7xf02i6!=s$AgbgFlDC$p{_ApLe!?qLsDx
z3>slW<VcRWC+Dc`)Jf~A-<r%y$KdpFLlM&3YOryAh2bL;U?vcObVCz9`*1dwbb4Jj
zHuu=s-r@as-$Luu=_eY)!rUCkj~wC3m1|U5ZCAmv&<k^D3vnUO)XX$mYp!3rM5(mG
z(C`SM0M$#`+}dDi=?=|SlOPw9E96O}V6#%?*ieZm2z)gr^;n9m0%ysrzCSqg=t+MV
z`pcbt-;)J8LiX*HWrEoQ6+)7xE@+Lykcr7jy0tc)t-V1|r(LCPe{g4gvS9pWpnbd5
zO%DhCz4&RD%-GeXMJPcqG0EXCJj?Ft2FYH%Uy;(*4@6+G0mMk!E1N97^$z3b&JFxr
zhTv&fq^IQQwAw7)Sz~o|omXD{2B}Ki<!3d=k00T|b7#5m#s#+ObvG$yES*lyDN+?n
z%pN?1(wg@#zRhrPn30ijqDYW-x-73QbM59es?9o4u0WwQM9&1QZtPJW9wiR(ey$u>
zSm6EK7~dwPpP3JA?l70e56&o+Rp|EzxA#g5^=;t~`=r+cYl-8C*||A7l?tt`6;h=z
zN}+X#HmYCsmDar{m87UNbzF=6>FdBHlJfredA6)Hpd-SC8P5LVXIZ<sOlM<{j5S|b
zcljbLCWIUzjq7V%{>Jxt@`(qCNB{gFD(NL`thHHMT;o6e?r)Q%9iq@>c@I7K053lC
zG++OdKcl+4=c-s18s!41vaa$obrhjBtvA1eEsYUR&LVRGwAwUoTx0RQ3+z_+$Q4Tr
zjgHcj5%p?|@qDolV*l%0Qb;=OHdsp`pXbQ2<22T`solIrnx>9$TBRtJ5=POV^D6rT
zzI~tJla=;9dima;u@-49QVuabcZyb{ic;<V&v)mKi{!1ZfTf)(d%MdNN;3@g{X-6d
z8lU{AG@-Fuqh70X<Mxey+tFZ{m>6eneum|xC3?-SJL?RXGz0@tm}7Kg3~LOFtIO=<
z^Arn3L?F;wQK?i|-`aBet1=>vb66#)@3qL!6$wIDfYSQ{HsPP0HT3=5w`SZk^?pn?
z1lh0LxqDyu&TAIz?YEC>MGFrdv4mmB@bCy}+M(U6^jRs4aai2IdQ$d&dskzP_QP_f
z9~_Flyu`;d>nBqM_GMOSpl20v6f%DHG;>cc()!kGnA96mxa%-LV5K4y8mR<3?_Z<#
z+zK-br_r$?z!TF3jxZQyxpn0x?Pizdjk|1Dwip^3LhF>nM-TCpFMp9Ay!He3>NR?)
z?YB0Yjg<E;-QwKY^Nfs)Fg-U1QgZF;HP%0Phiaul(&>N{l!nU`hKI<>971Z|zj=o{
zoBx>?o_mg`o_K^}G3Nw({Rfr4EbDmd@CUm984$|Bl&SKYK>IzizhWYUr|huCH!FQ4
zZUu1=F+DlOfBQH822w^GK6adIm*3~Nf9wBEH&qy=UAs+aUpWM$462);lEiJ4)KP6%
zq5H!WV9@F>Z-@M4OZM+2Nn%<=VZ_Msai$-dWbyT7jB){v7791pEI=v(VGufI??#oz
zLpzKeJOrUaSVtZ$L&q1SHQc^&msUGrYipHSeV1aPh*pXN3)8&x;`6-y_B+%XH4-zJ
z5^F2Nm8C_FpE|{Gxy<DB6hgx7JBw^zdYgK!M%qn5K%rD1UyO<4Ji;ihuH5C$*T2S-
zk3GR7=g(8j$4DDs0w=vBvod)<5OZx;Atiwqm9#jypZ@P6lWX#=Zy7j(VC(wxSuk%A
zgh9;2=r}*|V_!vv5%Y%*a_8z*{_sEk4n3`~8f@sOeuV!S##mIUQK`%9t)WjoEPYj3
z+$Wo7{WW2rXA*Kz#N>l#Iq>2guKvmQFsb+mKi5P;(bEB&Z(pPG@?GYSoOgb1aj*d2
z?zRf9ynl&iz02atEjD*ID3^!OD&_czqx|$w|0Lh}&MWLxEA-UB&l~lW3-4Xy{KJn@
z8XaZ!z(EKkmo8pr_1zz`ySqcuYJ-#vjSf>BE0dRbgw(us?FQGk{sW(X`LjIt?2`<Y
za!BomW28lToLt6FiE!`-(f4!rXq0csXrJ{vKbQM#F3G0XME`M0;xJ}*dWK*5dw&-h
zL>xPLnh)N;z_0zw{}<iF`MEV3i$(|sVlv8*c6!dwwZ<e0s}))Zg3txbLReqbzE8^t
zlxh*?@{GUmFeiTY8ns{lI;!i=_7sg4mI#5BDRGbxq&b#exy*^5z0UNZ!ykF~SZld@
z=_V?PxO(GVmey}GJ~~E{_IT*Qll;oB{3?I=2fxeO_9iW5`dZg}^@P{He}SJo`z$gn
zAZCwIhyylWdyVB+zsCC7I=yNg8AwK_Cn(R3Q4ESmp?KxJ_js$C<8xnlnZqX!6UE_Q
zYb!Pv<~VxrF#orI_RkPu%<<D_c=PF3_@Dpj|B+@d!K50i6`&CEqlR6wfWr>@n_2L|
zM}7S#TmnDsD(D(Fd=n4PGdg>m&BZI&M7w=xGU+@(3WA(KbqvkjRhkpW7%NV>vr~9n
ze8%f(3>#}}sBXeeb&q<hMJ|^^8_mSz1m_<-&&``RXtp|-)OE=apt_p1r7afb4^S)=
z7#<$QBH3JDr&e2})oP)&+w)N_PZZ_JMIi>m+R8Sq^$VOlb%J9jju7W#`a0Rtw|4t&
zRp`s`A)VY~mc0Agi}5rKa{t*bd=Uqi`1dDct&{N_87}k8GfyF9z~sa<tLw{r|Mgc;
zslUs;B2d;5w7%+Zj316eyTCKExNsdz{V}-=KUq-cKCT1$DTR!4$U?-zvyZWNbCK=a
ztMns#&k`+dKp1$v*u5sz8+TBTYr-JzSN;S7s|>r_o7}#6gOQ;rD!Y4>3Ng82j)O;z
z^2@*cOZ@R4|1n#e+w{6#V-bi0Nvh%3e(Mj}-LCW5&pgNM>=Z$6h|>=}#^F<4s?{oM
z%Xj-^iAmZ96HsrqdH?-u+**E}<&{;ox2nAO!ZS=w4Fl?)Y$$7WWvwa4^zRRUuL#k{
z(P@p40|^xaBub!>o~J7d79b2mhK7bXcI+4v<I|K!Cm0(W=i1fFRI0lqIwjH}Ru3j>
zmD2Qjy*}-Q@|8OGzEZ50xN_B?Pq(rvqw&0HxhRJSC9~(JsV&vmTdp%G(BreAfKVC&
ztI-WXeWihUQn`|N9}XhW%23(eWBKk9`Jpjt)f$C3CKpG{9XY_y{`AlA%6DF2XLpyL
z(*1)z5&;VMoo|1eO0&f?k3GiZ_&7nS#IXlXv(P(2y<TT?Weu#id8Z9BpxNzl_0Am@
z-+7yjjSaSTt330><4li_I%zMf`k&oem(5AxaE!(XoQwp0L3ihj2CFr~IC5x&qoGKE
zoPcaxJ$puB7*Hw|Ie738qZ5;qMn@PK9pUz^ThuERQq?02^ZT%w7Ht$slJ;$`eI-vw
z>A3(0N)-2NxyfrqA(uxMBM!av1Upynuz6#ZzS72iKMw=v=an{<E4N5rNQr`Yfbqly
zD%%?yT)%pi(eey?+dB*ubL5LTj-EKlfA`=0Lw@&nf0wnjO}ang=Ys$BU;Pf7+f}~!
zh0ij-FiVsl;ejWf=J@#zdwY8<-@fVmJP7C|P0%6LMw7SRxx}@_SGl{m$j16EU-<k>
z%+8Df3efJx@{kGXtDiCmT$b<Of8Wn_HXxzy^>ZQUDM9fgpvqAgF<dTl>eOkbre+x)
zpJHNil1mpavb(oMq7ot<dFX~Gpw^mRw?~qu-nLF=Oa{_1hy>0r(<HfXHb*JaG$AUM
zARlw~r=MZ{&Fd__eHVo7d%Ca^X=CER5bA_vtH<`+H<<q1=ONrb>F>_FcB93ut2ddN
zI?l%G8l$BmrE-Du4?W8N<^TFW@*BVYueg0{kybnLHn$Xlki<y-`9J?(Sh~B;OJDs-
zW)2>rFg(n}3twXD=@;nkZn1vxT|%P?gMfCs1}0>?TI1{Ad5sIVewSao`uF&SU-;WR
z_}CeSMv8xJfnA)786O|#{3DMtF?#@jiSaSsdFxHq*Ouud9dat_->3Ifcl!s1cIQ$*
zdGP#h>HV{LcFsr<f{YlMK0>RuNn>Zb-+r~e?<g#uOx|ER9g_AA#-4U^fcv0d)@T}y
zI$K*C48<i{tp;%}B8WmpCZ>7r`R925{r70p8!oub;1hwsYIyTI@6u^>ICb_U<&iS6
z%rifCgsI7tcDu!1Wd}F$FX*WRfh6rDtSxVFY4a9)dlj11Ca2GyWN38A&p^29*WkAD
z?bkjvnXg9rL8-nngpcqte4Dibli~-x_T7F0qzuUAa!gN8Gdwg*AveTOsl@um8jV(i
zRB7UU`t44o!r5PWa`J5ON=S$0wI0f2KIYl}34=P@@A+y!?PaYc8Xslh#Yd^^>=NuY
zd>D+oy-fzRHJ~Kb+e;+94!L5<$w~Q#eDm5RiUonuEu@&Fkjr!Y$RYAWG5_En{tDG<
zhs})&mAxifxeFr9#JtgLH(LD1Kl~>D`A@z@p_rps%u_5D85<jAVtj(>sWHZehmppx
zzS-c`(hgT{+-7-wheT=q>|1}zUS*eBZI{oz{1S%_%^{^h>jA&m{-Dm{=r;gJQ^yXT
z;pi-VOr4MdWd+5d8A?MV`^vU%7i6u-G<8~^VVGn7;4!}Nm7nI7?|g&Z${x9Rrhm}y
zAHvXDlO&197iN`0jj;-2o%ZVeTE>II!I?K4j+l9Pg4$jkS@o?(0U$vKP6PytFp~Oe
z4V5Isg*=%u1J<&5>oz4RkSaj}`6$PsxdrmE<mdm^-=N;^vQw#1Z?-)RmLL+6e3--N
zl-+ihKmNh1{Ml>YCl^N)axsN`p5c*U#zx1O7#U%tR7P0AcB{kE-YPej@3LOmBQ=`W
zUi%)kT8&nt&I`{x!-4r3@6?_$M(&rjbJ!E(YVY7%X4z+P&nV-t6KG+`kBw6(57AEz
zy-Q^HIHi=!?xL8P0|$8i3!mqW*S=4sRv`+f9DrW?l6*gDaKx-iWgL+fZEPP)dH;es
z+uv`QpU0D99D4aNcDHs2wtM?pkJ9-$f<S1566{^SL$BK*Un)C4_X8qVF1<%7FEFZ!
z6f+d_1x_75N}(L{PyUC0!roq+wY6P#ck6vWj{-?Cj8HaZqu%2;f9H?*?f?8I6pMLE
zr97pf5)%^>OifNPH$B0`=m^3XR@drWySv5vSFf|Ux<yZG{_tzR$L{Vn)yghk{mK_P
zdThawryDn&Dm^S9<AaQZL`&CVPLmX!^c+oWR-xD0Rp5n0At;W_kSi8|!Dy8mcn}CF
z9X*SID9Um0$Vq<UXa6RD`t?6#Yj1~xag0_PrB#1WMjJ<Wk|c@87rFpiTa7_sJU;7w
zEu$1#X{R@AAf6rP%+Ec+`qDb#O3Pubq_4<m2$8N6qb1uH?_j$<#L&H*aI1Hh2m=sF
zgHS-ZFvJrNJ;=~#j{oUj{0la>YTUiM!TRPNi3SklgMf0JC)HgRt1bTJzy5W8?Kggl
zp`ik$p%TL*!%R-kGI!`Orw`9_baIlgXSi`|mkT$R`R-d6xO``sPOA7<zxFR#TU+5*
ze)(5<@`cA4n;H9S3GA{vBklEYp9;z&WBk2;^iTNhfB%~-Z>&%(9CsD?5BseJOaf_w
z2V=Uw7lDIxB;8N_{<0#V%`m*UT`R*PBeTb8HJZq7^WJaJo+mMMZNpA;2W1p_-!sUD
zf;KkS2=f9h940o3LZ+t2k#WQeFTF(4PH5FybX#3NydaPP#Gyo6L$lH2oi{G?-dk6^
zx=cizizyY03=Iu2G*Y5eEMSDCRqayQYOubxO>M7%QkIKXF4JtcXm?wjd*CeN(-WQq
znL(1-S8d4Jt)hQ-n?8q!|LDF_=5VEgVq7MlD-Oa#>yummnp$ZvmM{pI7@g*!2Oi_n
z)l1ZyEylvYL1{b%R-eYiSg#4L=?`8QgZ9-#^)aR7K3Py_oqB-_+B^%kzsRRfpW)y`
z%QRkn2VrEtG^vqTA&3bGg`~H#O?zvTLV27(8sgBYON3Ft?dzA=+*oF8WP$0aX%eP+
z@|h=Cn4f2P@h#St*HDR~SO~og7YGfB(j?jt28t*Uw0nxZY6~Iu5UjY2SNVeoK`&MG
z(oD418931?7v8%-r_&)xQoi`b&v5eOL0|pKC<WYsZ#`MAumRQ@q_GHp;Y#bLR*j?V
zFj`P3jWIDZhmetfEz!R$g@izc{({s<V0wIp#~*(dr8T#1UL#+qfx!du%|1CTKnw;Y
z?jb$Y29<O^k$FGb<hG9w%M^-*3_wc8k4-VR)}r;+GQv16rp)FDOGHc{EXlT}v(q3d
zmx!!I5+Q<sC=6J-af`jpP3A`?7$2WNw#Rws!E?+^O|yRS5?h<wD5WVzk+XsK%+x3n
zqX}hz5|VBwrQYcPHN?gy+87e6`{c%{R<5<VFH2J<<?@yH>GirKX~Jipe~zOE<}f03
zcEWi5QCD7cI8d)woZ(QE_SIM2pMW$%kSmuNnVfJ`9>#z$eIn9KJ;4ERU0^ppKE?Tm
zA4O}!(w*DXTXo-d_1`A~-wJmDqA?l9Ns)nfwJ~UafqY7%<0gVFm;u|p6#+93KERRd
zOVq#h7Sd#s&!BLFKCx3drn|CBdwq@K=p<pFiA8`6LgFao`sH`oSX*LZ{4le#vm|Dg
z=U;l3LkA9W=hka1-(5x}DWziMr|K;zNVN0wxTi=1Nh@ivTWuo54zPq$ik`MuAqj+}
zn<%=ezSqxH$`3C5fOe-%FG={xpZF4I&KxEP0<>}M?!Zs@L*Pi#30&Z20AHaq7L2kO
z?RdP5g+h6Z;fb+6pxKheetw8$pUowL*{NxsefA}k(p<at0mWhks|SAW<jU|<?$%d_
zE!L*~^#w|4RIkl_TUtZX1=T}Xjg^uEFFeM%_m|lH?N<p*fOcdZB!pOD$&({O)45(D
z+1w>xDrDt&0OBZM@zQ&&+`Yy4>?w{MIg08Y;j2IM<IEp8%IzE9;@Xw#B%Oq@Vy;i+
zg0gg#qN6prAfcBA)Vn?8P7NW}fNP{GrEM$-$DP?u5?YC4?#Cw*-L%Ue{^{>iZ`Aoe
zbpH`gf96SM4$S<O2X;aTf-vXWiq;Ka2_ZOf@-#pH3%|sB@4U%yc@wK4r_H^EIshzA
z(!uuIaR`A;Q&hA0Q9tsNEt}~vySv(85J-fTlt<<mpWCI|yzVexzAI(4hZo^$hH0xt
zuU8`ui^#xX(xi|Cp=5Jwm0F`hd7{j4d6-tv=J3&@jE#+QckeE>N)@9uQ7-Tu9G@sD
zS4@$CMh1d3)pS}tgn(W4+})ZEh<mj%sMPE`+1*)qXKB$FJv0wIbdIU{DOW|bw$C>y
z`k1v~@Jxhtuq!Fh)=#@<!z@;i4~vwCMhQgVDr$&6qzVCn2*{qOKrlKy%!wn%F~+d5
zxlSwYIGY-;LL)LZ41AF7POJ>?XTb-7A1`_PL_wX!;YJ*G$o78;1hQD<;FFKAc5{Vv
zb<bTq(qT`GaoCdpNvBd{`^r_Oj-Mh7^AHsfL5K|UBwEmFx2e{4$mQ}JIee6dAASU@
z1&ep@a`4~*wzf8CG#V(SP>CX=1~JBvqzdeQN{?(h)Q}qCc%`()(Q0G|?jR6KmEzLX
zi^OqA5DN17FLC(LK{s-a15^v?Ds<Kx!en$NzTYh+5Cj-0QO5QBr)C!j!w9W?d(yu@
z>3OSyAavP*)g-+HfnauemM0&77OgaQ7jIIGV+y&z1&(A{n|qBQ2>LD9Y)Vck<&+XW
z^>RU77R*}@Mi2-@A!6>_EL%%^q?_&`THy`{V;$e0)S6yBVej@f<40!QbYec|h=`+z
zR6x7ip<S;L#xV;E2RL)?92Uvi$_jHcGeki^tK|k`98esP0oG_zrF`J_A)6aRDlA5)
zeZE2a5uZ*Q&8=J4iNlaM2*`crB@WClxWmhJ7u{=0uh<~=2{wC2@4ufoKw*e-G2=7S
z1X1V?bblabRk}brJ=?4b=cLaBlao_C`0yhrtyx=MBp=4)gP>mt+CQWrrR?(^xeC79
z+xk9KXD!xK{J05FJRn$Dkfjnwo_may>&qm|6?fpvjFV9kN(hCZy<2DF;$>#eJU|!~
zASya_i7-#1+|*jNvP~{m;KcEhJn`gH7%jMc>lR0j9A<rOm3qDIB=S;)v;#lyB?=qL
zzC2cDbEOPDoB4V27y9|T7vCWlM}(muSNKVeA3KUr3E;RYvuMg@gHRqUZOmW-H4Flb
z3`mW2lh;$zeL`_?ttxz#&S^u3u6=G@TU8*Kn_J+y=Uzf7#jV>{DaJVpxzM%4_F;Hr
z;AoGOa=vXXj<8rM)W@I_`PfUko7)=QZ%u{6WzK!|8E(9FlkT+|!h(?+Z6w-+#8MIp
zL35+Z=7p;qI(Fcm*rd}-Py{qvEq1q8DHKaQ{K%s`@W_)?D>bfsaFO%p&T#wIEp~Tz
z==D;PUP@ryV2w#N-ChEr*J0jg^HeE1LZVEHQXStx$+qmk<~^10o$q~%Vm?P41w8%P
zXBeCQtF72M&FaWu9<4{8O6UJ)&zwW0ipw9oOF3U4A5U`63FWdsN3@*!ik-FS)M<ey
zo%>O#v;8hamMmIVazYR@Hhq}t))u;1A-nO7wYXf_DW0UA7LD3E!{u40i7aBL=}L&v
zf?hA7)ol@lA(InR96EFuX$8AGJB&|^QL9V3oo=5VC}SJY#tnm6epE}@b*V(3KI+d~
zS{qh3)(E2j83^JyVq$XA@l1*VbOdC-wcO{V%LXI-9V@KknG1u6k>N4NWoz!qUHsQN
zTCt2RL8;USb|WsZ)0(Z_E#fF7h@=bP<e()id_~{*fr%`!clyvDuTuAk`40kLBHK^v
z;p9OAa&xmBeEcjr&PGNuBe=FkAOtp5ti1Iuy4~=SV3IsN-hSn4*fhZk!?kN45Jxek
z;y6jqpp_&^GzSkZFgrU%l62{I64KO_DYNnAR9g~dNi`@VNVWTOqFgXwjOlOU{UX>x
z2nvM)aU7$J;o_zDc<23h_~tji%kJ*3KagD@wSN+Fe=9Vr1-NO%5FvBOsKCtZL2|iV
z-!R!T+rLq@b{IKh3>{C`f{;v4FYwr-Pct)pfSuhvnyr@G{MPmd2KG}L?8nz-@9Tc7
ztOePhyv<73fgqkLGIw?g9d-MMsCzH2g%uEBQ_JQD>zH=Z-}JfEaQW3Yv8e{DxqJI2
zQ52IekC1AM5do<-%+Ag-H9m&w_2?yur_OPo*<w-FkXnnvq6MVZpsXde0}P^oAF(+=
zE|((=L$uZ0xN(Im*FNCY*IuVuts;ZalWGodL4EsY&(?Z*I3W;W08xa@<(Qb6C5qxc
z&eCS-*1n~+AxRR9v2?m!H|8${<I}S|@X*6dOwUrO)@Zleo>s$Ub`IO&h>Nk-PsA$s
zS?vH|{;8JD27SaJYxI7O49Oo@;K)<wP?7HMhs@@IbX78&Di&XV8`Z42%smJw+8(d}
z$$xRJBVoAo!8^opj-jCmdR>E3l3q`7<nSRD4$P1w9Xj0}N#a=H!|+}|PZXqD^#9!Z
zM1HP$>)khb<Lw{vCx846wzjtsVKDgmeN6Lqzn|MIF~bw3=Ln)A6EkxJnPQ56hPMAW
zt+%<N*X?19rqybZrmiwFH+zU@o_>+J*~4sa@6c>E20q+x+spnrcdwtjK=ZyV;Yz+{
zkg2*#brdml@&FHf`4Q4w&&jL#Z_(PR!9*g!bQO2L_CsuYU$xj;zW2NTnWWJ`qIv7h
z?-S>8jE>DCf|x>~M5mi_?%Y{U96wB_)1uXC(@PwrB#I(eZPV^oPg@cz`v2WgmQ-nv
zDb9l5VBbd)l*?uE`8=sneDjq*<r}a3DR002F0FcV9~9=VW~EBe?X*3m)j*#)2m;PM
zaGt|QPqDnbN~Kcyi1*?u#n$35)v245H6CAmKUP-l%7Pf1ooPtIXoT^(Lm2OO+1s<J
z=?8VzYARcI(YD)9{$-@Ny?7C86-HRrH`fS)kYZfKD&r_bv|(&~jPgjCRJ+M;qfNh1
zVR2Q#tQu}LSndAo{%)TtgAhLw_(f4f7={>QSzcXbab=lH7q8H4HXXb|dWp_`ciMi!
z;Vgm12blpOfruC#9w!X4&f4IN82ohapQ#&qNRp)AP8}N_<M@%|j17;`Y&PljdJGsb
z2KJC%r!7k&jTy)pew_7*f;z2JM@%Y%{r5LP5E2IwQB-95>`8_WPNC2qc<3s1t}N>a
z!0Wdcscmf$h=2%<y7dmdYZv&zH@-pI>vHkZ`_8Z`U}tNWO0C9AFTKpO&pb~qpJ!-z
z$l*GC#Uq=JGse{ov(0IIs4e^^`EW2bSe7(R7%CNs!iXeQTzKbAme!W}=C{5>(o6e9
zpH>?u5u^v>$*vd^ctP^m>D<bIvGGZAg#t;EqLai;sFGEAO#cGODr?qSI^8B|+QZR7
z1<cMa@c83TGc`R&v)OXE9?xymAJfj-J$4_T4&0ZZ&T8ktA_%;4!NBHGm?O;RnK(L5
zX(mSFNi>D4)MZtQh=|5=oyKkrAta%)=*#OQAFT5B_b;G&J#O5%Mj!>6kV>USy;0|h
zXP)KJ#~&xo=P8zoxOTN8qczqYYT4vzW?zfXz+?tu<<GG>s5GTiC=v#ttI~dOk&VqY
zzW0OINYb<~?__O;CG{n9U%tmbmmrLgQG^Ku!=s}_`5dWI=tN<a=~p<7$8CYD5M~@_
zy-pjI^!$F6Oij;n{-H-0pPHoA>A2VG2lMW||ESIJHvjaOl-3FnxGG7%;2~U=8V3<^
zzR2uDrx`vx>9n4$zzT!5(zhQ3u@zLW-=(^~fs`S!Riu~Rp!eQuyz)nXOw#S}_IqzR
zDIyiJwXsE|TIGvh`Y}H9+0T*B7b%aFL0G2)9ff{k3^!pee$>z9$Mo~zp&{bP`T1+F
zf1kU{clg>Lf1OS@@xqtZUmydu2uCvjNQn$1WROFKd5jDhotPwweGsqAfcD50wfho*
zT4}HEYH4>GeLn}u{K6rgefA}0<`$?on(oZ-9F@k{e%pTkVShjUJX<0Nu_Ey8HUV)S
zkuMRH$2jurIVR3dlGvW(&k_c0BpE+q!G`R<eT&ZePQPWy-6e9bexE=2x4%ud+2Gah
zzJjp|ZDP866(ojV_}hPn&wcUB6iY))OpdvU%pfF;qdupZ!YMLjG(sBs(t{b`e7{fZ
zU)^q(@zGIoagI)}#~=LBf8o}h8@%_!_tEL63yyw-uk1trrqa|^;*Gwib0UNwjv^j?
z>@f}<J5Hrm+xI&AD|Vhh!`l8GVd1_6b$uw0a0#5h$OvRW6of=U%+S~@xskB}j#LPY
zr+9GF?9gs((`nZoA$?%znnk+$E;nynLhF>})m0Y+YG^f^wAyXXoI1<N6Q_v6m^h!?
zS2dLV;Kg+xGrU`t(~1m_z@PKj0e6Pwa<S`TY0aI*MfNHcE`M+tm8yP)?OqBZn>8i+
z9l$^YND*R<q*NRt3S-w1(TRu7(7zvqe_xqQsI`_}+CzDAX^CWP#07RE<x#rb#NFlE
zdJ3g1K-_n>89S&re@r>RPZ-oGqzs9|0@%O>Z3v`@2&5#+<;aguv2gAbChrJGtuUm1
zvj|~{EYN7uxU)=zVWhfDboGZ!H}|mJ4xMJ5D3sj1{sBS@cGfoObnDD79OTKTU*yMs
z{O6gNoZ--s0}KrpK{~xzn{5J{l~V0J4)@_`e!nBS3tOX;9<$REL{UhsUgzES-)3!X
zhqvB(&r=Rq$D?Dw_9+O!$^gO$8%7Z3F){`jQ5q_vQpKH{x43!n63x9`tWo_5H>LDo
ziqda{dSj3EjU_63Ra%`U2+4s1M|tdtXBZxxq}OvG`Ru~TTEAKLmrdejulZ@0)z0z5
zIOkf7{;(HeL?8pgI3gY&V&?cXCRV=XW&CB2w|NLHLvy7;V4%F&BYby@iCTp1D!R=!
zQ4p|l`!?8sy`3uEc9ZGZ86JG>abEuNR~Vg~WPW~@VlfBO@f3mQi1Wdh?;s2uQ``$t
z`S9Uuj6o+UQ)A<VVML?V<m#16tZr{}@q<g=p}^uU602=L+qP0*LxBw?HWy;T00K#&
z=;&CM?=ErY>K$5pb+=D^@TK*jf|>;j?Rtfs&2?(^I^Av?q+s^I0_Pujg!0%JiBc#(
zZS6jqqsI}x^&ZGz--Y(+ZEkex@oxnn9Qe762xUN&%Tt(~;ow83(RtUtu)>mBi{nT%
z#8yD9LG9)u5r*RSBEj3QGF#cfw%fEDHKH)!+U0i<O0v1KPODwzz`?^j|KjKQxu5?f
zre@|jcH*#`Sd_A_G~u`bKji0R?UKLT&sEZ8erB3D3aQkpym8@mmRB~p@WZ#TSx}~N
zRe{veh*+%i^C-qfF~kK-5JM1>FBSk;ytTxg_pZ~~ZSJe0;`t`^z~)9%t8KHoddF4Z
z+s@AqA34r5&wYlmiD|k$*NM2t&of^P!+n1K@iX|3wQ_l{-77!_h+NTCBuga<hYxZ3
z#fQ;D7E6GYnx0h{E0EHVHvwX&LG9|Iiw~M@%&V_)biImhv}x?_P|W4H_~z>f6|r>V
zE{)nYvj-0F!pmRar+?w^uyFVo4?cX3v58?sC`ojR!J-jp0&nw-M#3#2_Yt#xD9Ezb
zO%0~)7AFoJBp>J4+S%r7U;iUEH}<%F_4cQ#)>&m;#Q6~~kqwr)<UrA?@1fFkUl1TD
z779G~;!Dgf9D<MXU89{HqKGhzd?CPo1i|$uUs)g=%FyA;aOY(p0<YE-5l3alCg-q0
z<Rt=y=V5iRRS;N&?b2=TfYe0AJxp+ma=C@I3B9CC5K7j!RuRU~tk>uyUB*VoIC1nO
zk3akb!=(|%CdP?#Zr^2hqBS@%dmPSfQ2MrgIE(*_A3fWxV+<G`86pS*+MN!I%S%*u
zYuvqc*GVm6J!!H(*ZZ#%;8^klgbl$22m<oC0$OXfHaFN<U8LQv_ay@d3Q}1B=KQkN
zZLwS1qupxLO?rS}a(s$oM~*X896~8SEHU^fWE1!@xK~l)<L=W>CKHyVgh7rdj9m50
z4h~Z34;!SUeDn|#r;cEy>c1Il454pAM>=8m<|T4>-$7jdQ$}fUe0qol*`l$v%lVVX
zxq0&gy6q;_-6pL{gGP0iy}eyN_qi|gQ$P6&oIdp+bMv!)^xtI9BZDuwz(Zss$or#R
zyU%$0KAW$26^jK99a=yL%i_`<_9|Pv`|d^d_Nx7ZI_u>Rn4g^tk|FR}u0#-`{WSXI
z*f>MO!w4bhw%e?4tfF;_F)m~1QzB?XtKDL8@h%&i8*FZFc~W7)-0VTd$EHwT?0x?P
zY=))VKcV{<Wssj_*+f7v(DB@Fb0M8(tq@R}FEe^%7%Lp9ZuZOrZ<A0#?baIkYZZ9)
z8pAEcq2egB-H2wj&FRC(Si5@%)$LJhbZIxbv}+A&wHi-9{{o-;@>e-};uO<Ull`f{
zf$jF&+;@@e;1Jjk6}Z>tz9pB>Ge0wpu!i;3Wtz1LA6&jlwO03g*Y=$>Qw7PG*o|~S
zr4e9-L|Fi3L7Ki$21bWRDGn7uK)2mxdu!84KxuDt>68VGF?2c|R##Wq-q~Sych?Wo
zNTz1z7#SPyQyy4r`ji@j_Cq$IOaHxo+9i65Q09oD9GObWkN9~Y898x;sdL9L(%am)
z>UD%`qeLoU=hD06FTX+X#_v(qbxuwVao9Ggt#9+#nNwW5@($g0gUWW3=5C!vWrv-e
z9lr8oKglorjbG-02Oj0%fdz*j+vn$*?fQPsC-d`Csld^rhY$j8-?_o=?j~=(@ix0V
z6|kNgZ{K0<igfN8aF|-@{XFRJi>Zl8hDS#bLXdP4R#!JrS~-BTaJV{wKx<92+2Yo%
z+pMjwv9YnyCl5ccaD?&k8Lu{w`l&dV0IGZZ+@Bd{0LA!JD@%U83X9EJux^qIq6ia6
zCLTP-!qX=)vg^Ae&h9bBNjoJ8l{eplrFHn`pCW5J92qNfwz|jm;s&34@(JF3<5k-A
zDx2#K8k<#m^%_Yp;VWPLasK|V{$rkd{$);`K7mS=!_0c44+8dqpSxDE`;x4C@gp|(
zExnQ9G7mm*28rc^_ugiGZISCA+@RC^_yc7hK1)O>`p9%wO}keWB#I)0a0QWiqt4#$
zE?HeD`<l_wQD)~4kX4J^Zw5>XO1Vj7Df|1!?qeeORLd#?EBi^|pbL<}dRzrBSDc_U
zHsdte-3J9M(jV?J)HK@bh<KT%xJX#+GB%x~Jf4trnjAPd&F0=FsY>WII;dWkPOn9)
z+2Z`!^E~<JlN^{mz{u#xkHUsLyk+ob`c?43=lJN4Cxjr5W5y@O0NCE!q1kS-ba#<f
zqc!O6*n3;4Zb`8ImvIU+8O!`osYpKW61gPlQK?lNX}HzCqV6k+#*idEc6N5C)oRr0
zwf<nm_~-;f#qz$Y-oEq3W>R5ZnL^;tvX5uhPc$fzZfzn*N4)&WKB7MVyOfg9f+-d`
z^2}pgTi?c3I{hp<7SI!rlZNru3i0Bb<VAuB15O^EqSt7XhNGN&;$d3t2CshSTfFq*
z=UKbEN+1K2tn&4*zru?zyhN>9<LcFm{Q$?U?2;3Mi{0;p?10ZU??(kZy`)E>;7Y*<
z56-i@v&)T}SC|}|=Kc3C@xlvFxx?Jr{&;j>pGXV<(%Ph4y!bBZ+bMAr6Gb_OhRQrR
zGs*7W4i_%G!OJhdOremYR2o7Cl5V$4rLu=o8aWg&GBON8P^;C@TG8z!^m>|Zw?`;s
zA2^wM4yz1bqm%>Erm6baWQBjxGJ9RZks@Iz`&nY&=EfarfrzjLIPlOsw|5q?b-xJ_
zL<A&4le2=6%?|O!RdOlVKyqYailnPZ%N0(ZIZeCS<iZc%;ISv4VrzZJ4Fkj>-~GYs
zJn`g{G#U+V-?;8+BnGVtci2mWbxcHAWZw@!`DnklK$0ZnVkZ_oH#1FjugcP$+YFD4
zbM@MFo_zFSY%n07&6HALP;M6jluEgN<qDUzA_^mdFs4{4aeC?$)oO){@4v^>&pbme
z7gHz}9d}-@N3B*x8BJbFhKGg_QqX8LFj~{?r6gL@OA<Fx<qrU*bl>LIIvkl&4m5v1
zY@TeaQyw2BjAUQxC$qW15X+d*3QW1giI*Pd(%Lp=x7$A?V*y<uDA<&-y}N|hzE5C#
z5Cxn$Hb+u#ljO&F_}M3DwrYIi>wm%*zVPEL-&%5h9`5pczyCG9@cA!NsZ{yk;ya{D
z@4LwM`FR$&Wr5wt@$+tvVzGd=h9gG~u)Vp><ty(oH8IQEZ@tUsKKC3}gzmNae(t|n
z2H@c&+N8Yo)*Ga0N*u?;QJ&%QFz2TqU~7Am*I#>`FMat-<nuA*@(_Ux=yW^m?(UK*
z#ZVv_8yiIgf@-CL(uz*IN7B=DIvpY@d_eEowvN^)<EBhWlGG^^d`tlKU$ktk2gBAW
ze*^=YlO?YCkU)f(kphqY%=7%<*6(4~lfKRKLedT_C8Nk+zXsp^HgsDMg^VAa=X3Qo
z$;=$jfBExNcenWWzxnI@;xGLw*DhWs2m+Ga5a0an>-^l$`~<yT%J*OW7D<}y3*=q(
zBJiMfcj@^A&iutTw-(y17UgmYV+{|SJI(Ua5^wzQyDZEf;qH~&oPF|vKNqI);kCN5
z>abZ-5(E*Yp&_QHCrFZ%m8DgVoj8IJf}wJmAaDaBTkC5isUpa`!5^|@$vT`^yPc5s
z5`r)Wj8lP65>N`O6lhJ_OX>C!dcEEt*!>hs>pCJRBExJLVSuXxKZGX*#^#uuKgve8
z2EBGaC=Em56kcLWZn#1k-604SAeopQ=0uths3LO<hv+0-uHCrG>EoxVSN5Fm>K0cn
zUg7kq({$P$RyJ11R3-MWejxAf^UO-#?f<-(j#Q;YvBM;eO^niPHrd$Tpj<4mytK^8
zvnO!ynhaL7|MmU<3yHRxrIjW0q9zDjYdW9HbD%s%qt)d0o!gu}dzLVY$mMcKAxTq3
ztJOkjLm?Ivif$sj)9E<>_0XI&N&Pdn{!aJ$;XZ-a`wKT8^KAb_LEX()|BOdxrijBF
zaTxoj>PGa1l?dTlW5!rU4jf?q$wyiL)(=qHF|~&lWDd?eevD_HK0!WIgaov^J&L&^
z^W}g;?Ol468V^2nlDk*G%lnt!=Iog>+_`d<g%c-9Rg<NqMGhZ2!kIH?Sl`&iblnDb
zgx7{3iX8!|HYj7UBJkO~Z&iH6L2r$Fu+@5lk@7HeQ!~UOVta3cp6YV-+Fc%d{1Nhn
zJQnSRt)<7{nEtrB5Kh4P)akQGAxXMjD%Cnl3EJI+>({S3+LdOT^3X_sa<$!V^3Hn~
z*x1}43`-n4dWLd&1nseR##p+YI&)J`laHbSfYMs5XZ==6lk^NpY8=Sae0m+w<sW{)
znaQUKf|w}G;Sj6FF`-MJ)#ECc<+(9t9_q67+Lo(C5)s-68ETFn9OvO92goZ+7({fE
zF8O?kX<1;tx<{{(aO%iG?kv5-jjNY9aq=W9x0jhYGLPxCSzlXae&GPej~!#XQo(jo
zZ*%ADLFgulqygnyWlrwpUd+TjHb-Ktq0wqGR2pJxVwxz(Q`y@>r9H0QS>*ir2gu2&
zU!Hf89bhqJO4cr`4LN$^I08wp+oREJqJ^NFrrf=An@9$9+FeSeAp$AUTG4JbxqRt;
zws&?2^92qbJx;M)?&B?uwe*@zriO;eh4DVHzV&Txi86-NSW@F!l9`_Jr&(A3^!Iss
zZkjkOII?L^zG+N~6rsZfO4lYCJA8zL&ppoa*S=4x9KlIs1v1i{d+aFBJ$HsUNWq5K
zPLoo;#KCgF(d{iddsQBN>?}7geUrD}d5s4je2|;(f55@BXGv9qJ9lnz?AQs;pFhv)
z+BT+_+1#aRK@_>5Lu*oFF*4}4YyKiX*BC0*8lxj49GICSCn7d?R_Uc}E?&CHGtWIq
zJ|AFFXzR0dE6Ap4_xt&|2Oc62(Cc>Dt<*@Apx4t}yK;p%l(d^IhRe>+Q<c(e)_L>7
z4_H}SBZ`JNb@Dvr5m${30_W%LR+WY6mng(>KY7bkM6`Aju)VG!NsR9Xe0uHI&ENe7
z!~f`SLMVxIMTAj+fwTi*4uU)gmq3g?aEg;Z{Tw%c?c4NHg%*;W2_V;H;n4*SJ#z-p
zYl93CohD*roWqge{K{?GTUB2A%u~Gk|M*?L`Her|h0namyRW{^>1Uomm=<>zS2%X;
zIL|--JU8yFV7iWE*Y!U%VVrlgD@v0XFhPvOK)+(?r2zIVV<ig9=JpO#6O)`ge3U|v
zXKDEc-Cm1pSMPH2k<&!+pC{FI@0yvNVLTakbSmv`KM1Ps>`?Zt_^gwWBnj2(E?@iF
z?{nk!E#mwLk3Rk!W0O<P<}yHAL#wvOf%zHog*+JL^(LLN2*$|@wtJdh&!CjR5Pxc$
zZ|z*=aB-MW1cZJF$2~7Mz!NgSs1{2Tj7=Y<x^~6U7+DCUMHB-jk4-UuavB-AJ)9&7
z`8cLLCK;M&lBR8D7G~I6+GJ^cnT3S~cDC0U8=IwPdhG1&Ff}p7{M;OkY7<EOGsL%9
zT`@sQJf2TF8vhS<WcDv%T^!PGcPSPMjE)Qw2LZKKozxmORyH|u?1&%4ad=s22bfiV
zvw(2;@%h<#gn%?n>9jitBT=blV`Ci|Itjo+zR+hS=q6omFWsS5t&t02W~b&U<_jKI
z?WXCwX`9mM38E;(W<ze)<K#0vT4k`xBCJ5#k9VJcq8niGo3AoAcZBj|4<ce$sS6Bf
zYux0EpN!Q;qjC|mPdq@ny+QrmYgmyodT5xZUU-bdXXmgY0c}vm5Cs7$ju{;r<6OJS
z4?noZx&P*?%pCcEvGGw>Hn*9Z9%Jd+Rn8ngL6j5h?A2&A+bEUxrL3ekvN6UH!)ACC
zzcKrn{)ZAW0fbS&k)wwhDh@L{GfAhp$LUk2xOL|iy<Q95)$DHXGI!vBYr7bQk;Vms
zfMvj_Ed$BW&=64&5RHy9Jv~Jn71^uovazwsaJfvk)25Iwx++hq*xA{kJT$`jbB_=N
zF~^Rc=IF7rbh}-`Fr?G%u(P#BC_4RYeSiDSTHSBa811Kd?n8xs<@GftXV)p6D<g!f
z3`tyokxef<=4*`(6_cmOQQIvV*E(3!qdaVQ`20y0rpwr*0|t~<4x1H6437+RqON%N
zW`&cVJIUO&n~aQ&vaz$p%=kF#w^ulJ=n#Pns8lPoIvq#0?B8F8iMWRx*<b30j1Byw
zR~SkTEG$qc4lzAGPPbO)=+R>=FWn{WbxBi0b+5|Q>@=3Z1+&KbR;vXo2751%l!`?H
z88TWPW@>VRAj(s(RoL0wWN2uJUbjmwmj@j2aAmJTX=sF#XU-8sISw2?!hxg5TxBi@
z==FM3wzdeB8tA}6ze=V0S)A#Eith)!>eiopj{}F#F#Pm`h{$2s0_*q}9f&>fL7gVe
zhs-~Ho@9NM>RUIkw!`RLk*8mNj>G2<AVkmgqqXBV%*8oICnk8PQ{mOOFZ0OX`B~<U
zz03IIIE(8W9GIEl&Icdx;F;4zImy<}F7-x>G`-i)@oh<?Qv99nKmT9s=TQ`L;^Z-g
zOC!wBP1COL^1!(VxPIdr-A)tL)ogEUvvBw@)(V74F%s8m7cQ{tQxXLM!y{$lAS9HM
zxw$!_D97&Z4(n^njEoG^?Q|#=%Lt#nZ*6TcTps7gf9y%ZAkWED4{-9-gLFF1&)c0g
z+Z)S-beyKMr>t?bCIP6#(Y$1Vp6wUYKGj<J&F^sN>?7n}K8*;qmnZZ|9Ik>FNa-rV
zIm@vxJV@{6B0Jx_fzfq}a}lS1;xm-bOe2!I<5x4OqvR;$5p%Pg*r@QsS1$7S|LGSw
z`oMRYJ2=PH+jlu}c#aRg_c|{;^*BL3WNmqi-AbKaujle6950&nctn*tn`^*$?NaUW
zmLFyFTrT3#M;~NtbefY#4^i7%<+*2`<HFl-(rH%dwNk3PdrZwv?+0pq<jPvAk=JS+
z85sc}?IrAPZ!<GJ)qe{@xS(!nX^Dx689w{+mx<yMCr+Pbe0s7UjA(6W*6I}F0IL#&
z(JnTX(#g)Hy00dqj1yxt_Ah>)A7|~XTxO&+Mt=A()(;3}cc_FuIj=DWWdh2R2T`g<
zx3Yo_ExCyzhfW+}Xlj(yI$)tT#;HOCA;nUOxrH`s*VdRj^a#t-%M2BVsO;4l885T9
zyUy(VG=UH_noWA$<esXtGllE<xVB|BrJ8l9?MKCPfIE*TrzR-Gd4@;JbdnZxv$Je$
zZ;`45o2E1y4Ms=CeQST;1gsVPfD->4a``+`3W}v7!z05)GNRdPQmt3W7YZb4LL9|_
zy9=9*CZ&9l!w(!Hlp!;dv&>D+k*113N|H3C+SmiB`nbFdV&H!8U`VT9kNB%-zXp{i
z7k~f1a3*ib&rH%W8cfv3ECfN|_{S_Hy&lEU5srWPvs|xib9%PGqtBkESdKBe<0MJ6
z2LdbQs7;24Iiun3)ivJz`m3Bid6b(sZqrGjTq-a%GRorh8yq=(3YB&k86IO|Y?$4h
zDoVL5VW9dD^hf>mz7cT20F#krSb+?A`spV*fBqaduC6dPHiFd+hDt-6I)0K;KB8Su
zS$pTZ92$8UH8O*C@~VC8j??@09AW~rFr=v>H3q9wy0*pT58k8OY0&GnD6ft%JUooG
zF8aTD^CnZ1lXRLbCMF96GNoEs<IM{fXfzt^?(Pst_|nUt=@VV=pMKjvL7V;C{TP%8
z>XPf<c#mV@336irU9BLL4rD4Fcb1TDvL{Ir3gsLJpFPZ-%2kdP#yNLvjzT0c-8M3e
zu%4^WYS&K54dpq;468efTzTUL#||E3ae0+4nsTwoSb3bayQ?f5J%C9QhK5HND-BVp
zH(V7{WH$Gs@SgU=VH2_gLi|VzR?2`!9(<5fCr@%~X`S-$Ft**ISSWJ%@KH)}j(SJ2
zap4N{&z4XV1+>sW(6_l1_u{%N7L_I_g9F#wF1M~-qt|YebUT!W$_xz+0g$GO#id0i
z#wO^sdyI`035}t#v(3ee7iqRyR4WxCA$j5HXK~sQ`zD*orlvCm+5tY~Q`>y6$=m<o
z|I7I(p)fz?{9Jo%lBXw;VSvO{sD>s+IrURt;PUn+=ccB3;>8Cj4(C9%5HbQk*h8ui
zQcxHj<(z_B*OqwW55CC*XHIbK>J8e;GBQ+TW^A0>m#%R9fpe&|!|3QZljFl|Z&y%C
z6H4i{WCeZ_-@Sgu+c&WIkNWwGFFnVjk37uf5AHHCF@{xj%H?6so<7G=A*NkZEPnsb
zI8goyOw3`Nn8aZ6IHa!w>6FC`Nzx@T7ON6EHsJktF3@S!>2{h7505cAGK#g9BuTh>
z^(wQ|Gj!@LrlyO8BBipk!uNmh1M2k}TiaX25`OY4U;0Qt&)V$$fgSr7CoUg%MeA)Y
z|380&vxQ$GJ~ZkuoCFAr<4cnQ0>?F)wA$onMmh7VU*+P;zh-W@%>47G2}%Od*n=ns
zDg^{2?)$~m1T(~p-MYp1e(U#n^09|_@2&S}7#JTdb8u>wi?3bap=X~Y9|nw1OmSdl
zf|Zp`l2j4(sY={Hn6WOB*IJQ!9Z=z`D}5X_5Ck#*&CmW6FTea5e(;0K%+AhW+B=Mm
zjq~&q&rpa3ora?NgEyG^^5@}0#K|9LS!>x`U8B>eu-L9+gk)-N7I2fdZ@%#+M-CpM
zv0LNdp<zm~MD@CS?>nzj*{ia)zD6Mo_~pO*%LwhG*sQx?+yrfsCVi@zOtSIQE@|4V
zEWN|r(SRr#B1v5^r(uBS@;;<gN)#8FJaCRJ+hp=!%=D2tf?P;y96ZFQRk*2IK^PX8
znS=UPjkUYin3<nuXLFl0vE&Om%0mTqw>OxannhVfu~4Kql&9TnyFmu&s-1vmp8&8w
zLhtb-XS<LbJARaf0}Je~?NBO~Na&Kw<(QqGCeDYbW{>Uabw;PoBJ_x3=&&wPcf}I=
z$s7K>SH5~`F(jtT%K9?Bv`3;63b{Omd>(K-v71|)442EOq({CSL7=I(tK7MBhfcRk
zqtPS?;rxSV+%w+~8?XbqwEkzdA0IpWiTZO1ipKIHSN`aa>8`Dyl_E_CnWPW`8%RQ+
z*J>buk%RNR@RQH+<g;gpB8wCT)$S7G%Fb!8hfbXse3*+dCgu5KV+^g`#&lAum30Eu
zWOH+!j!}#hhq(0aI|LZM^u;gm@FR~fGc!&;3dlzxaTFk=rC4^<Co;<LA-+d0-5KS;
zp>ESgaYPuxty_19!ki;1oEQb*?AZqhqX?C1n%6E-|Fi!@dihPFR-I4_A2?)sk<wvs
zgr^PlZAxntJ6mfkF5aZwX;Q5=xPIdbD{D(ssuj9P!nNx+`SvTX@*BVPZ@7HveG0K)
zYvVR+D>rB~Hc_gMu^k`Pdr1={O1a=sd72ZW_Z|HA;et;|Xl+!u_3g{_Hkw|>C<Aih
zheWJD?7M)VJUzq{k00luV+V+=Mi_%i457@St#Sh8%2oY>D8!}-kB!VzT<&61L4A7z
z+3vBuy-injDHn#harrudfft{Dp0nr9F*!O)EJJdktC9$7$>(E=C6`g{|Lhrou#`c6
zA4Fk9AO%ay%LGwO6o!lrmjTQ16DJA75T!Kj>svIxb`^DTov5jiCg_)cGp@e=B$o9M
zORYg$D!V(ZEich+x2QLp+`WC1t&KHm^(u)<xwE*$>lfbScYp7{aO2uF@<On;w#wG>
zBJI5$OuK<eI&OM40}u&Up;JotNtIotVlXiC>20p4FW%<;|MZ{fE-pD*2IVHb-K3#(
zer^T5Mja509XY_uKl>ujym+2mY!PU5tBcDU4Yt=srwWZF%H_Z$ynJ+=@}2A0cFNw)
zGC{Aw+Ug2ztr#5|;k`HCAhepF`iY<5$)}!Tes+>V6p)XdpUc2f8ZJ^E9d<;Xe~F*R
zxtJh;YuBz5g?ZvQW@>5-tmUDH9wv%o(p1rW|6Q8@>EF_O`}>5A8le_gUx|>?zp&d>
zeLpuk#p;yJjb(1#x=O2EXRlJ@@|BA$F5h8yZ<lT_;nL-6eC_Ms<X`@8{~PbU`!>Z`
zu)cDg<;Ba?s%xmUiZ*ROF`fCjhgKw>&&$|8epV@UUrKceT`HHaa`}Jx4b<gZ=)?)m
zV*%+gN0B5Fl6IwnKvFn;h=>2y^Bj2LETQeWmTkQSfv@nj+YT3I6clpEZj-0y$H-l{
zh-_=t?p!9Q?{Melb#{9Vilt#*`P!cn8_hrbm0#rxU-}Zqjw~<~MU>(k`8Xs9EF)uO
z#;3*z<G^D`{lza`z#RsW=a_UuP*EtjaN%v@xJbT`<G{iU+G?J8@p($6A$q-p?(6Sh
z|NZa5<!g|nAKAM9sAa9C*KV_P>pHdC9vfRbT)FZA%|@Nw-CbIp4nO$eoBZ0p`FH#e
z|MZ{ooo{`ELa9JwXM@FS?^EAhL3el2N!>wOL;%(U3{s8hIefG+P9-5pQ%{TY>20np
zbbEWOt-ME?R$QxJ`J~7PH3mr_Ayr+3fP8V3!w;Nd?$|s8j@ro@MHD*1-TpwAwT@KZ
zDh{145$CoLMp0{5v9jmRz*JMr=U7=?LQ2E=^AB+N&|%8MLxiDpIBtKZhB-%l;^02}
zY~u$hWaf)FZEYlMZEX<-Awd{WEETa>7Umb6dW_a|x+^p~7wG9b$e=@}iG=^`E=1iw
zO>Z6S21BD+V|#atZqlXQ>af1K!ESYrW~+@#HJj^OT)lFg?|ttFtgNmO#Dd+%4%^i&
z8r?cQ+aa}X5XuJ)I57h|=*sw{P49!<e4O=(Ds=`+Y%I+ix41J@WcG8PBN&Ux3pr>H
z_y^K6nlSA$(XKHy9+C5<bE6<I(5hA_j1SXGQY4_dJ#s=4O-<9eyUbUQ4)gnas~nm?
z%!Rk#XQWi2+ih_C@JZ(97P$8A2kfkG^X!w)v%9^;?93cH+dC+oGBP&C*u(@<z{=_h
zcW$p@jT2Z7LnkX@we?d;mNf0Ly1LC%PrN|V)s#v(lAuJl2{SWu+_`lZWfV7i9UiSL
z)BE8L?OT^AJ@+}}_(7xz(ff&QEv|}Z4Av;LQry0KlX|Pl-K9GmJ@O#mdga^v`~UFU
z9654~bLY;{?j+ReE)Z@s+6W|35HUG1iL!<;jKEs7H3Jzkt<l<|QtigSapl8)27e#2
zIfEuP0qw;GD}}pEK6{8X?ve9C?NVkzjU>?tVb3tu(2Ruza<Rpj7>jG)b!tuW<=An+
zA?fu}Vi^)mj?!J<<+H;l_~zCX=I0M_@!B<p>n)N_lS7A2Ff%>J?Q6HG?ACbX(I=_w
z>@qbqMWs?jr77jnG9zQ72m~7&>nyKtVi81PNa)+c#();W>*6MCZtwEoqfgNDhl@%J
zbS+FyO|i7Jg4T+q-WumOhDoY*x;MutJbMsk%1C0DNs?8@-2p&G6sa{UOLuA3E3B?9
zvvB-0uf6^nzyF7Sz=4AYIdS3yoiw4@Y9WADr-P6IVHh(uHjXx$Ac_zcr;uPBj?)-t
zf9+fB`+2Oawf7~1H&|jVXk59)jq-Py|M4Fqn8;BO-sXWD3KCM#HHxU$W=dC>nurKh
z-tjRS5L#n)caT$~h+g6{{!WKTN=9bpXy3fc&mA1$KW;5?^x!dm@WZzmAKInUsdMVY
zISw2+!Uu1@%hu{9pMCywY;UZyu&}_^#ulkc8JnD7Vrmj01WQYc+_<p>HuH1oIiNsj
z=jTb%V|jUl=b!rmy^f+>&eIEq=rmw%{s1?x-6GW~*L0hwx9`&X<`$hx?^F8HR}s@k
zh)jeL(upo)&@Jnx5v^93RB_|RRcehrZr-`c$&-)qCtv#-|N7VdKb$ymiiaM0h*qaZ
zwN^&}^?D0|AP!=tCuf|WM-Ho;k!WZ1M_Sqb<v#Fp2dKUuHdh!5DxmttbuNy4ofE(M
zv&g~_GK>e%bm^+$t+dN{vw_&!AfK~Pia-h90~fHi4m0D>@g&5}HUts&=n=w8H~G@s
z9ACe7g9pz&!XJF&+sv+P(Wq_nz=Kb6@Yqpa{j*m&d~21z`E$R>`r;BNj-O=t?h?JU
z$IQY!^9P+?@0BZ;dHdZfXe%fba>S7&O*J~TC<Ljpbh~Y?UcbXn{md`YZl;Wnmgv|q
zstv=!p+nrbc#A~!xZB<0fvauA>LQ#u2VZ^>#zsF@)o%6nEq3=dc=zHvJo3cn`8WUe
z-|)}=$A88nk3P<4UV4dIv(47-E)v+;t09o&qC5u=93sWIB*z2gu^Oulqza=nNovpr
zSMkPTbkA4nKK-DCu@GYDwKmz^j2Rv~M=C<10GF`rQ*237gA5XKLv8Y9iHt%|E#qYT
zdc77wTtF+Q_@i~|0<&P4MzY7b$4b2Sc9p4#36}1xP%O4dbdQ;dX~suKSzleDS+8^I
z_z4=#IwPYav>FYR))a~*N~IwXP}!@nx4j48)QKd1QX7BgT2ig3SL&QPdJ2;&;#`c@
zIq1R2@Ce(R+h}9i*h-l_x<^+viIY`w@i_v<2z(#EUjcOZt?UP2tY&Lxi&nS8=JqxV
zlM7tBe3>7-_Bzv3Q_RoLlO&3+SN-U9dq^ROf`D>)2yLwEYLY2#**=6QW7lxQ<Pyz5
zu;8yGsKfIFg$eAwbCqafmWAgYgE(@nfixws7K9|w3FGZO=5&*s)O4&N5W-38C_}H-
zbka=um^AG;3`ZmpfuuAx%+m5MPmK+s+Y#FbkI~&;XYbAu)x9c{OH-Uad4|o^ZB}kC
z^VkCqQyL#*Yjc~Yo_?00kukpe-S6@4J8w{{H3*}SToAHwV4mUOA<`s43Q4!y<L1rV
zluAQ<{^cJdpBtgs++}z;PfIA8)jC!~ua(m4w2&n*i6&1IyYT_R8;an`uh5-7Oh}0E
z{eF!X2bYfipwXzawzkaf?iSrn7o`k$ZZERDvcle8jaIAW8POe;fKsYodevH^Gm_)1
zLgOR(?CbWAfoHTP_fJtGz*<SD11i^>gp*Zf9*huS)aTj@{J=<%2u7-cskR|D0X^G6
zdQC=aEJ-`TL=eR>s;gXJ?J(#ROF7o6H6AKWpd-QF!Nc^n8`M^-d{BFrvGECx9X-L;
z<{oRyTbw)lAjOdqySsZl^ynj$MuvIw%?n(9|9u+GmgDjY17>HY87h@X6DNk=>m@8M
zFH<O%c>0CU5EsjI>NU!xn2s^D>K&{Rv^y!P*F%Z`qaki3aD4-*1@gtiBnu$O(bu6e
zehODgj4(7C4K~-;s8)9Ac6%seSzcXZeckcVIi(2gh7OGJBJTS+3ynYI@iaCWg^lMq
z^esWiS|!3^2=9x{-L!9{B-<Bm5X~Iq&}W~7oFm?Jevb2Vl`>J=Wh!hCqS0bNRviSI
zbgzzxVnjZNOxkFriQ)heN=lO>EZ*JWg~>81joCbQn$Fq^yEkt0`pzEH(=$A9?jhEf
zwzzwJk*6PdlJevPYwH_4|H4a*j8F27Z+w$Ce)xSVl^RhLA%l>^M-DJHI_xl6NIKmv
zSFhe+XlR(Pe&y%L=SOMOcNiJT(?rpzR6#?hnb7Mr5joI_B5&)k^bWzd6ukV?^bQ;&
z5+OdB?^kQ3r>CjZt1K_yWqW6nPN#!P4L5Gw=I-KMc6KT>8V&E~nZ#Y%$4P3VQO5Dv
z*?m>Gfu9fdK;Pyv@X(U`0Hm}Ites0kR=#zE+_CpL{nJlFeiF2CTE`lYAz`OOxVi;<
zn~<h}cJ0|Pa^;`h8su_NC_uf9RT_c_0zoi2j9OjgGh-4H53}^h(=>11V(q=_Y^-gv
zaPSb1KKU$nE-!KU?W_FQ=YEW_g&D5jxXGhWJjvM96t8^eRlf7hud}(eOD>luh$7CN
zJH_<WINf#^Ap)As7H?d5o6*s6{{BDsClrUKsqU^aHC3W&61MlY5H_Sy>(Ob{2(5(F
z!rm5qe*;VcU;Pq{mjC>~&RWuLhm|`wSzf+NqtT$38s2{UeJ)?V#M;`Xn`&06Q^3)h
zB<W#2(VJ45)T%zov`$lOnu69SuSB4f?N78zPivCFE<W`#&UAq^G`Cj>7-o3-C{S>d
z_F6mHO9WDC@}(BV(H=51u2q}So>+@XI!F^Z@@k#9ZibM6B+M0P>^50Ap2wJgPHl=*
zRcSYNxzlPhyjNy+dX{>nN@aVSqlXWY%ay3t8yr7&oML{6YuB!EXYn?jPKO`}ky7&i
zqwc?BZST_iKJ2sB`t{w;-lw12``g>pcM4~Q9QGoS6s<yvvf{*cEIU@57y*nR4)R9=
z1c8wN1_IcE93xgFIx<a4EE37#3^{$~?fv$4`aWm({p)M3{IT}EcSZ%~n|Vhwvn~$K
zEjat`{d<1T`aSD;zGbP`C5l7px<(2?UDX^O^ohcdXC8ls*o&DLGr~B4(vYQdH_pxr
zYF)a|W(>m5DcQ!C22VakU9aFjoW4Ag@Ol7&$+C>&;eg3(LRFPmZ8`2Ab39n+(JkDe
z9>4&!eq?9%Knm@FQ_q8(esCavD51r7@9mnu`1tkxb?S`kNC+he3&Zx;UZ%aZf_LQ#
zSSQP2wZT)8q@2>3=P2oPoP1$XaYStl1RhFCW=8`W8%~Ygi$i9UDJ?%DP$Bp3jd?6k
zycj8NOGD#{k8=O&8|)qLu%C^YO-FQ>HwnCmH(z~|jT0yMnP2<`KJtlAW311GvrqAj
z%TICj+FNL2X|*~`rxW(}x9Rr}kwVaFC4@mlRo2|SdzZDfHC9%eXrtNPJs|d5+`Ds+
z*<{XW&?jkt5(%}Dc*^3>4=G;yeWJ%c1!te7_5y@4i{j2MX5*UMcWyBp9dkT5;_~Ik
zc<iw!`LQ4S99OShMeB-he)F3=^UO1B@9eR+vqM!mQ0#az##lobhNNjmQ5KBGW5O`}
zj+iV+0pC&FJ<=}w-odmFy-OtoRm9$_`!stI-uVC}9hl!5=OPfLmR3=sl*C$rXB8?`
z)TRdI#&+{jO45z+JcSGtX_}D)P5iLI?sT6Es>Aa#VB3xeubgE6<{gd>?{hSplTIhJ
zmsjz;kZbSVWMzGoAN%ni<CzaXk5wKg&tK;3h0ENzaSdak*={kLP3a#TFgQL&0F5vv
z2>aA^&F;=F%ga5M+f9r$92^}J#%;E@cbUynro$0&w2X>8v~=uk?`TH#r5l7FI*Yx$
zLhbXYL-F85WenT*?=cw;84iY=yKtU!7ccUWkA95Xw{Kyz=JnTK=dmkS*zfl_JUDnr
za+yu17-I;7fIQDA@{;LfN*Fwf`2+FNN7h3Q!y>JEB*OYVENyWj;Ytw}mb-uUZI(`~
z;6L>Q_>wHku(ihXB+Ya}x6JWc3amv51F@3}!FUKyQXTa1Ppl)t056G{jV84Gh)BlV
zx<BUWQ1RJBasxx_`A>8Et(Vz8xXbQr$aFMdW&IR^AM?t$Ug5;)GyJ80>#y+1&wLIm
zLatnTo~Isvfw$jz6>Th?r7q*~knQbz93Jc=g&+u;MBz47S#k5`O*S?*Sl?))jpqLS
zU6P>9%^SCwjAjh_2Rw7a;?Y1`i7za{XqV!P{{ztrKaP0p!_<C=FlK>;d34IYaq}9-
zgFgNK0Z%^h6i+?<EI)bp)4cu8Tj;vxD_{8vA9~>h?%v;F``*2UBuUL^G{k646i1}<
zlq^pf4u`~H^e7J>k|NB*@r;yWapM1AVs5Py(@rvY=O2HC)w3(Cf1-(%0V1oQHu#<(
zIzB?qbC(c{0}aiF8@Xx)o`UT|IK2u{2uTQo5u(w<Hkaui9B|4@_=H#Uh7l}({#SYP
z%YVv^t8Y;D_ZamL*f@KIX4K{j|NKi_eEbUk*5CSdt~~Q3%JX^l+2{G#M?cA1Z@f-z
zEX`)ZVgG=8_wKTN|2|3xR+0`$a*ZM{c>VP^ICt(W=T9$F>zb?AZqjV7apUcq9QQ{Y
z+`mWCHHb#b6`xkaz1#3le-A$UIXJlv|ANIJxCYMN?OPn|Z?k>>9v}PYi@flm=lPAl
z`I~&>8(*icN`Cvdf198D$<Ood)$3e+?>+LOM3*JQ{vle|G+GTNlL^^8<>=smcHDR{
zQN%p-b$(zA=Ro$J=echA_qZP1Q^HDoYKHqaiCS$uwSlmLs&bySLJIsa$B!#_9jrwu
zi<CaKwn&^vZduIn!X{FA2<21cIZ5ObdOkac8OtXn>XKl1Xo=)0j)%Jp(joOEBP&vx
zaT`x6cJAG0sng}@XP@W7xywkaII*(Dwe=J19PD6?C5aoPdCG7!WHOx~1O$<T(db%p
zbaX_!-6n3w7;701$HZR9QU8cM&6uZCdVzDCzzC;Qq$Xtb8^r!4ggJ#Z4m$EE+HQY8
z?Dv_aGp4gCn;V;)*f_ysv&Y!m-^Uout(&(vb?PLe;fUdAv}nR;(k#VV<6NsmNnO=s
zX-eQfD8ro04fX+U&=r4=CSrfFJbzzBU4bP4A{537(!rRU-*}0qH`Y)~Yt(gtsB2WH
z**qQ-n3Pf^6q5-y&+u9;8f!hIlGKqzXF2)dG0EC0VbWl<*Jr8ML`usiK70yY<TTBk
zzjo5+4|eXdaqcmucMOx+fNY*}+&{o?wo!h<-u+$v!GHB%^R+L0fhV4SffMJ?BW=ma
z%@cHbE4=Wb52KXg_U+s3?d{QQHrUy{&)V7s@4opeQlx}_iLYQjo$==DZ}G8@e1@I<
z0h9hNGCSl#yos($WCg0(z*s?;jwxUJBWzv5g^z;CF!*ltsVmaylw0?1vEScgWod;^
zzxX+pS5LCMyv(;>{wCXZ?vTw#eC#97GoQ`b+u5P6Eook|d$31Q7Svi(6eY7+%FP?M
zSnjT%0?!HGO6#ns;9<4@e&+B0Q>(NUz7RwtXj_wyC)|7E9#@uQqU8Wx7N8Bvl&no6
zf;^*)71cDy76#SuNS0!R5NK6V&T@*8AnGQBNyKzGr<DXqoAbiv2BxS;GR4n(m-y1*
zU6xOuV19STd@>@>4;UUF<2NE?7-9F1_$UA9cX;({FLUw9ORS&TLe-j$^-bEnE>AxF
z6jDmI?{9P1Kco@I9PID0yt>NGcizRT3L<Opr68SVT)lRkr=I@^J4Z)kgAu_r=XBCV
z*EJG_jD4(>1k;-8D>pDz1y5}t%ZggMF`Uqbbe?kW{%wwq4(KlR`0z(RNoQr1rS1}M
zz41DG_jk#sQ=WbDDbkl-;_%>*+8FY(<lu0RvMkWrVW`q9<KF#kmU=x05SQYSyTu~r
z<~vS%A4Jr1>+-yThsH8L8gcc@-{gf8>oit2smdJc8ZXqG91jVDIhF>+XpBiyyrnk5
z#;UW_iWJo>L+thOH`fT8O~&^RXs@)8%JS(CpN2f6Rn7U^>mI*-aFeaePcgn;Gaer?
zpU>$Z>=Cp(sIbY-o%{T~|Kk77pZ)$H^32CR!kNpLQ52jxd79qp8Xx=kCs3Z^`t|E<
zZ|~4)H@SE3HXEB;yz$C6QOt=0m*>;*gjZgEl}~^2r?|Is#N=QbHQVD#cmiFO$O2@$
ziA54lhm>FWr|7E0KJ`hY&M?Av69!!}8;`kq>l(WU+pKoi_=(T_467SwSXo`=YhV8|
zcW>S#oeuc)CqBmPpZ_V__wG?^Lz)-d-@8v<I{)81FPKi}Tz&64tIKPsz+dFK!Pte1
z%ELVWU@(*bb6XJ!j}8&F)l7B<y!5+Y;>XW#5T9*f=QAfhFLT<5Q<&^Ss}21j%raPR
zFNz-po`%^FwgsHoaAVASedr}%C!Bk74K<swIiB;IQN%yJ_9hoU{Be%IJ>_Ws9@Fs~
z?A*RZwA@3rdfa&XI{(Fg_J8E5XP@Q8pZ+Z8p1w>cjClO=6Rd8n^Y{}N!CH3r`t0r<
z;48z;n>RUi`ZQnr!k-YCF>OC5^k8^2;7`8rCI0d+|1I9Wai8IxTlj}}c&v9G^4wwC
zIt^F6^bg^8|3~=gpM*;n|D}hjt}3qGc$*uyuW@qY6hHCPzeHzwg}?J3{f9j9^wYfk
z>Kjas_W6}x_&NUm@E>yRy{lB(b$YJdx<;DMsdUYJo-r7Xc<H5AIC=66eiWdwXtiK>
zFIfEbLWKOGH}Chj>e@9|y@jrBk<aPx-e&7u8x^$C3t1bZEWM=^>y?-~psaJSB~nFr
zVG|()wUN|1LsT<_-@^Aoigb#S5tX!@e5{8y6-!aUb1y31du2qYwMJGQk`^gdRWO}T
z@Kl5%U}T1T?W<qp`n9(?aq<k^?iyCrEO)y!8ZFMAIfs;ze!tIXG$M>b2EzfJc8BfV
zyJ%HGpphQrMb7s9Z7!d=$Yd~Oo=qW0Sy>M;re65c2d<d*QZoG(0W|^VF`~p+wZP6B
ziXvn0@Q}fHM7!DK$;*$^PL^mko7}u}i{s-yd6jbR!Wpi<Q!pBhu-a0UHG|=Zs$NW5
z>YAb`=pXiJHCwLRZw=NypsLtKlfXPIzP~T*SFHtjprxe}7NG?P@7?9pdshix>>x~n
z#bX)Es;G&o8cmK|j1ba;Lrj^|-de+tL-Zt|n9LABytG6RhLpM{X|?ctMO{Q39UO2@
zW&FF(Z1MZo`^;O*ND*;3I;5(m)MY_BFIi1u&aSMod-YvzUws=tjA^ZJvVQUefe+W-
zeFG5e@9%Tw?0Gt!E~_ic^p-kott~BBa)zV*eO`b4Ew<J!P>gD3qdg99zsb`}FupUV
zI(w4H^YJ9edO_n(XAHjlyR`fUcI!OC7N9Lw*G$Gkjt>v;g~yM7`m=1UE>oJE<HKEU
z-*}hpyLb78pZj?(UAn|qzVc0%Y+B=c0jEx#<Hn8a93A(W&gLlTVT^Gns0A<;9<g*D
z6m<{UxF5tQ&KM9%U}QyYYJ~PUx-(|u_5i;fAY>COeU@y4p7scIjVUzJN++gzoMOwG
zX0L(gEBqj&m{&kX)Qa%~pRz898ZlBybfg&_4LOzj{I#`b_~PU)d25Tl7c&?gP|s)R
zaz<8ZmcuqDx+gfexy}6>H}Rv0#!8!&ja33Exc%N$EP|t>K3gYG(dxA6Ep_QM+pM>j
z79=yn;P9CDu3o3Neu~MYrkKo_-`nMiT4I{_sW)2$UWiACr~zfs^OW(Qyg?Ex?CF*p
zO)c<M^YMhy(Ev{dy!f$Cvff>$(wgz%5!*L!vbVd>kAL=aoIQJvmtK3_<(4scfzRg2
z)7-s#hvD%t^DIRvj|T?P$VUvR2i+OFkP*;EKf+^tkE_-oJb_k*N`o+x-M8;>=FN8q
zKi5UrCRX^YYRkH<iHjO-Go&_{swUmu$K*NDsSRWj;&}npV2rR9uh%1tLbR@kI&GA%
zsPdTOojop*@}E3;l7D=2pIK)EB^&G??~=|Y)Md_WexHpv=EC|W_uhP+>u<e^A0>3Q
zPOx?MG+_Ymyzw$1*xB9X{Dn&_^_E#*TV=V|<<v&+Va)qGJG}JmS2?wLg=|nW9qe#$
z?Ny#{TShmglxNQp`2oHHSuOH>KIQli{}HXQi9L52sdF#_t82!?WBU7h_;~#6=YN`0
z>uZ#{;AnrF>+ik6-J7@hE5GuqJpTCO{OO<mxx3D_#`8nYp1s7?_uk|1aG&vXhNnDN
z)K?Y3cX(S1DcrvD9eMshf&P1sxj^`WO4Ss#M%Es;zJ7xzKKeT6wpPGjMR+kQxkZ#3
zQ5X<~s|LynZtucu4(Crg25H=c<0&8z>no7NP^5@n7tfQ>O5oOAE==b9yO%HX#|OKN
z)~}#~9(Q)GkslvYWfMm8Yn*L$dG_>WZe`!%t*?KDsNG@Z!dcEgafQS;?A!`KNbcX?
z=Ir@1v^s4rU%JGVOILW}+-Wy7vzmL?ZtzEc{6#Ka`Us<gg7MyMc3%DpKOPw7lOa6$
z3`7Zt1{4dexZ^(dzx{v1zyI$cPHufyVV6%P?A^LUNXTFM)xSz-c?D|>d-v}1>es)_
zJFmaV-~EsNV?OlJkMa+H>$d?MlQ)cc;^`N7`CH%Q-n}~<4Ttz%;PeNJ0^u(RAcBQ+
zlgsU@s+`P?eq=BCy)2^*0_W+iEeLIx^!Mqmts(t{K>G+IXf`d)mO=UkYbvC0&z7o~
zU`&l4w~)$nU9u{7(m}xkPi>7KL=KjseP*)>oz;que}0MgUP&p-7-baWbVg-zjH#)M
znpVqaWx2(0w9DbqE}rs<;|}dk58=bk-aP=u<1wo%Yb0@lR;x+Uh-oi1V8MEyPDgCt
z+h(QNqn_vF<(x@=!1@_UnpV_x58qcvB^=XRKy3=9^H=DGA=0klHhlm?mZ!{S69Of9
z>hZ@}>Ne4J&15!WzrV|HaLn`1J<Iy~1~;zVbU$Sbp7dB<UghB6fXQq^krfV>^biLv
zziV%Q0OQjSr!n6*Or2xG2<dz%v5In9a`&sRvwHe8^2AA$Fl?HVUblhKA%T@xFD8sy
z)LBMRl*}hNNxOv%BmAl+9}ZB;quFU=wG;C7lP1d7#97H~ctmqP;5VM$;H&qB{P9hQ
zH_vnD=z!7mm`WSQMaI$5E*rBZ-K0(J)eJ^6?%uk`>v5ZIcZpuFORLplJQ$F)W15Ya
zG)qx3VltUBo6bnnoVAT}1hr;-{~hk`-k_~K%D%yNB)ZI<(6H5rx<=O(*<{T8=#c6g
z|Ah6Q|DRJUXYMUbO_9zR9rgL-$3Mfx)90{N#?JN*@4WGC28Z`K*t>_VKaRHmU<)Cb
z&u0q&eZYL4aqa3=Zrr-f^UuA&XFmBU5Kc<sL5PKUWEu9IW7rSE+#UPCl4#NWvx;&q
z*nQ)O<;@kurbb!AYSmyV7MRfBix3-W{DvkdYRal0Ej4i?M1&SkTZ)k@SdwPq@TUU&
zB*fE}uq;T26Oys!FRwho%j0`|r7#2=5j(>j=Cc8{(M*adNq>*kc#TfdMpuT>Y{2%;
zJ}SCSr`=|0sY9dDU^1MLBq2#0k!Kmg517v8r1LpxR?yklz^^^(-6Q&Yw`r;dRpnu&
zLKmfzchCk|8+2vJr#b21i24h!vHVv*iS`yM_u5jV8Pmat7e4wTXHK2N<|PLQd)#>Y
z9Y*{69PjR8pL+o%)j~*DkfkYUnt|t&<vF)+-Qw=fHcveHBp><kN1ZxEQ9eq1D;Kzz
z2YAZ`Yw82()hUz$AvG2W#-o_z-2SuI**t#^dFBirniEw)XSs>3A_NNMSE!_oKhG$O
zf^wV_blZq1!7nxC(J@N-w7N?RF=v6_XrY3bC@q*BAJQ5f@xOTT1Q+iQ_`Tbf<n$G8
z9PBU{_bIhuSfm{8-{VBwX1Up=4m3xD_qcWaHes^F@^X)rm1WwUHlw2>8l9MCvq759
zP%>sTo-mzEn9noTPhKLb4ddHyaO?iNbfiz&x0r^cE;A=IY&4>-Fm*{f95OrHr~JZi
zvH3UtF0~i9@s+O0XA_18`+V**KgHwcFJa3$ckkZk^>2TbqrE%qZr{SzpT=8|%7hTi
zrf$!XB4RpAdHbEWc<<UxKKhZ5@spqX33p%QPTW@D9=eb6Tz?17?gPMAS#L4cs~4kv
zpLCG%=I_43=F^W8J&_<xiAXh~6GP)9s2Dm0Eax!IAe+IgfaNYU3nyo?djLTMt35a4
zl@hvLXlh7v#QpmuyZij&g$tb8-RBRc9<3)n#%p)qWB=$bg|0c64Y+^fZ7z0Bu+`h3
zOmnsm-{SSx-XU&(nXSzYPMp|cd3l+`ojp3eCf#nE$z+7*x9IneI35fbkEWcy@*Is)
zFnRMO-hAsTtOQN+wBs;=vne2<Dxr{&*Vw~jxP24bX(9gCe;52m0>#;E#{TU){QR%|
z4Yn>^0%I8O?C{c`{s}wR-s0waZ(yrm#gpE{y?rzsG9FD(UW4Ii%GbX3RlfboYy9lb
z{2YJtul_m$n9XMo?f#@*;4SOwVIl8u)!#c&f(No8-a_9}LtW+!wr|sFFC+XOLRgkM
zHDMzJ>*IS1Ck!vb7^j)2i;5tO(ZWY6LsiXDQV~WmuHyur3bE2B@N4p7Mi|a{;nN+u
z>vOKYUg012I2ufuW*OR8(yU@UAJJ`vw898&YUY!S<Nh(*G9XD38jS`~95J8I2;$H+
zg~|ej6M8T5oU$lcYAmBfO<MOkoF0?J3Q<-FBdAT~iXh=Mn~bR`%Z#*`p!KV4v_6Y9
zYO$*tswyYTraX4>3LERIXj^hT9C3f|Hskq#!Q>e11(XsG>;IxCD2f7Q6=hMfzrWAn
z!4YTAoZ+#@E(6Z};-Qm=un(Iu4=pOq_o=A!FoX;LLyVjIR>m^l9&+nT-{9ge`~=H7
zV|6qn6cuF<;8~~TB=C`;A`C2&fI6#@K|txb?oB!v5-fMH%DFIvVTh5?=`E3t`qWww
z<~g4^)8WcWi~rBdw<sGCE8Q#fM+XdMBW`9D_p>3r>4^0(pcy2r#7ze}j1;5&g8tqv
zMOC4VKud)}Bb1Nlg@g#2+7Ro-GXMbW%}GQ-R1i!KNRRi~?(dT+k1gdB+cC|!g{}*<
zlvG)U?<uOX#2QUeWi<M?QAc;tYZm}FE{Wra%a<;&-0g7V<~8Q|oY8m;)SNwgnzLt5
zBZTJad+$=@B?!gsyZ8C#H^0r+))~SeqTB5e_&e<F?Q-}2U0SU+vovLJeC%$P2a*Ge
zmhFS*dieoZ$tWQR7c}w*M%vnH(gV%?S9Uo4snc`|pWZmaub~bM3Lj&gU!;(RAZ{Rh
zL0uO}Us1~vKM2UiDM8D_N=sE3f*?R+XtkSUgE3l2C@mjrpXPjXm49;g6?E96+g@fo
z=`)@UxK|yqSB>b*VpjYmlBh*5j<8lC{hDbtW_Wn)qyVhJC<g_?lPE91vmS9B67~h<
zaKzzok6c;SWs4{^G#Ua^mKfnt6&0S6)Rl{QnP(*16V%QaeR>IOjj0V$7;^UX8M>_w
zckkRLOJ@%EQCFPYI>E`!ErikBx_OhLtUwCx@9pr~>u<2Kv56l<bh=%F!vp#U2kh={
z(`YnF^Ni7GL|K*#-KIynMh{|M)z$YLb00$_CDy9NEoG>TVR~=GwLkhQkNs+cl`><s
ze@sA$2}7h2uCqaaiWPp?01;x-3K0e9T;Ycy`Oz`{+7d=dOqLOZF(_DC>yZu*&{`tq
zIX`)7iKmv^{Jod1Q#2FSmY?Eau*=cpn5%ioo&1R9$$+gepcOP&ZFSt6k30rN&cXJ5
z^0K5h0<DzuomL*c7ZD<8)s}Fu&%D3G-Gg1`p2x{3BFT<v$1O~myF5=*0$))U1=iH$
zRZ4T`I&$|W;?!e{E=om`#614k6;^v)-n;q^vuws-I7Co!@%(u%UO0ykhPU5-lRPg#
zD6ZeW!<WDOHBO#BPZY*1_f`pm`|NDr=jNSTblM%J^EpTTK6z2_07muD<Mu(!tMYq}
zc_1teB`{XOf}~R#&G5Z3Z~db$^8EiYK}|En-W&?gksyTcFmzD_UIUg^A@-omAr8Tx
zLY%n5ZoTDXV5S8$nt*}LRk(K>AtThh;DwD9F08EZTQ9vz-tKZ{>nwNoud~y?&$sgt
zZ{=gQ`Zu^3b?CNNIn`YPGr`0WJ891CtJlbig3>^x1R)mddjx($EIhioqH%D<<o+Gr
zxp$R`?{Qfz(Hf6wt+Zf11?9nX2EKxH4poVrjS+9X3442R>eRz-UEE0c*hfFY>CG($
z*RPY!#vC0U0Q^7$zamV*bI&}(bI(4F5}KF4_EqNd45Z?fx8CA+{>R_t^5f5tG+JzJ
zog{wuD%Y=H<&C%BU}a^6!D!6R_BLsnF8o&?jN<IWoG*&vdyctO0zYu>1`o8@w6PSU
zAxAscSvmC-jb=rsHOKQcb?xCxhs8rEtOyCc5Gy<?tr5aQn+i{Pl*Jt3#aINjuJL^z
zV<1WrvT_PWAgDP1c*6Q-#7kd0q_#e-MuTZSXPRd84>iL<N~4j{X?w(hPt$WuVT2-0
zYo=4DQEN1AbS?~1C_L$*q$TugJTIlFrVOT2>QWN51b$c%#4*|~YJs{!Dd+TKt)Zw&
z{IHKvM+mcqSr8Tcz-MiJjYc!!;BcR;C`jiS2E)qAGRrH=SYg@S+oi5*1d@LLn433l
z(e3o`J)dUMWZ(@Mj)wG)j)~)lA}^WEXAdVb57&1Ta(_{l`}-^Eq#=}wnE|~J=d!{e
zH5|NkgXI&eoLg&Pswt%h)HSseu|q25TDED9ozAh5M`N{3>3K*`qfAAeWf1xZQ<J9^
zVWUlLEbZlW(!q#8N{rQ<Y{IWSw~2c99$&aWq}5nvz0+klpD>(G=vO5NRZb{!lF1n3
zNt*j1TdNzKx_FVnFvFJ~Vd&#4pR%-M<34tBKsKLqcb;*tC@7TAbDfChG-YG0MH2ZK
zUEz66syxTUk@H@yYkXs9c>&eIE@JH>)(V7Gw3<tFJ00G7<0YgQfb!@c9r5NnZ!?)r
z@B^R6AA6Ecr-#)wN?M}OgSsT04hVh8l?#_RbMmy)rmAX&!y$w5gl~WQ+uXQu>mig=
zEVOzaeE%SpCoH~{q{eBCfs;9rT5)uJ$kIxaQ^^K)o>K{l&=O6FV@x^rPgQ7aX0fU!
zX$DkYfix0TL0wb`&w_>`%?TSZwKcR>dZd#aUwRl_vO&ntu3tgzrhN7I9*s_u<@O4b
zbj)NvW>BUa7jyhV5zj)5&@_UCwbe~FPj4}vj8U?I9|QzaQPhTFG>6fcGA%hQ#_ZO|
zR7&%Bc!nhNS#7k5d=FFAs6<j16($U*E63dRjili>s1E0dGhM8e2qkE?I&@lXuD<!E
zv$FC8gTa8S*RL|2O&6{PmuYpD93Bxt=y^~Z%IS<yd7M9Uo)eoV9Db@cj3*Puvk7m!
z@dkJA-+!dl^dRQ{g6Qh^6mw|_rD1B6BLXb)T-C7q+I3b>u5xiXL1YuO52Dr%EH7{%
zd2KMW3^Sf#6AyoViOTa4p2pK9I?W*RL2HWHl%U-PZCP5~V%{GRD2XvOrxU?%KEH+c
z&TW2ww@;_F##VQQ<Jpko@rZ-6U~irira8^Y2;)my_d`x?oaF3dS2!BX2~<cJh6Kt<
zrp<?k*yCN&*^FEBIk(G#!t?k@Bc?r_b7HwoBl6L8iSM;2vWzfsit(n___e0(MO3@n
zh*OV&6$m^!on@B0U0!|ZYbZYi&*R|WkXPP#jp2Aq5ClB)%(HZtRvaEtLLB<os$f3u
z6Zwj#9(#-nXV0OHMVA%F#{-T=Bfk38uX6R;^+gv}EO<+ga{YtAS9!t^E6cDJ)Gp*`
zG+rgS^`&=Ny|~Glr4Ae&Ix)toa_#l+h&l5TMjHGYRvQpZpq7wVFr7mvp{ijpfRz>Z
z^XD&O_Xptlh$^Rp<u|UJAbImO{+rvkS>8Ou`SsK64fokQ+UIVTa(jMAoF1{%57C~X
z7q_{5;W3{0=nL%a%?SM_jYf+|K~`AC2Zz}0o6Lt3-kOehs~C{_il1qoVtG_@Zn?`+
z1mqcXSDXY(vjKSll}1zs5w~FHz!i0il`M5xUti;!|LmXQg)syHckkch3xED44v+gp
zQN)jZ`m?NUY(Z7w3us0GHqV(H?$Yo*KKbE~@$3^%QyWWFl<e;Av(rD|cYpVH`S#1N
zKHPI3upkBEFN&A{pIpipc!8uyANj3XKv=`%aEInnht-vk+U8J$;1SbWJn%rTD=dXm
zNC;!<S|Wr(${KAd3<|mEaQR-0HPDRPR7HyNoXugg2Ty;jORbK%@z$6qXwYsZ%=3aY
z&6!PWrjrWgRfIu~VL}vobUPh-%j?W%bG(JJhUa_Kg{G{g7&W7;3yzLTj)xjuD^8vE
ziNcarC&mk$L4mXOsxZc16ix=^SsIZ?U5^OF8e$<Oltc+hBj)bTZIti;$#gp7{`NL$
znxj0$iLDbf;|5?5(z#V3G)0x-`-;uYHC9%73sRjXou|y^8MklU;_#r4dAPMNp5ub|
z^*n!{<5&06dSO+fEf}1#x}O@%wQF?Sm+*Z{nHQ*9U?hZ%7@-uN=SEoBJVi+XQsPGu
z*=&ZVB(|y-%n!ZrG_DZF62=i#7*cA@Y)~-hkI@Kza&w6$mOsC{&!Lw1e#H6p4P@Xm
zN@olv6DDbnsB3&}7!UeP-<(mEH7f`@Qc?zrduht<G$+%VLR+*Fbd==jz@wYad3r<f
z^oFKRhj>e;5JC_rA5&?nx~7<?_*xQp0y{XsRudq?8pC8b;H7VTo!i&n#`6>WxP$T&
zY~_*9Gxm4)Xao&5w`%fq4#psbpsWh^`#Y3{Gwv>{nk+Av&u5Ir6OP6sCX?xc+WxN#
z`3EqHTfDT3xsz+L!a9-9JjUEk>74WNtZ|0aGIH8>q3>`CN+PTz&vNI_D!>ceXZNIs
z({vS1n+edhMp#7<22?>sr7O~+U^L4xLh#A<<0M3Ud$7&1IKm49PA;E7`hsaXWi%Tz
zD++|M_<6~6Fl0WTQtO&79a=IbQ<mMV&p|OJ*9GOm@~tI8&Z$$h(|{`~=Cbk7qZw+e
z<?Mkx-+^9rO_`N=GRBt@dz52mDTMCdr{gJazWO@%uiwCn6TD^v6~$N?QWgV_`p3j^
zleOAXWChN8O;DEwM~8cqbxl>*RJx`pO42lCGMg}*j+rgwZ~jfJT3a;kA^#}P1^TUb
z>3!%jfp4kP992oI0@X~MEtKyeYJ-{0Q46tNKaR-96MSDnQMuE;sSyTjS%J1hNlY0#
zT<!EY=V*U~7J^?~@6j^w$M<*Hw*o&*xU_W=8F&omQ;x<X#`6sFnxN7Q`}<7h6Uwq;
z4M|r>N?&q2OSw18N$ZNjShN%@DarGGNN+adg=NJHtD1UtjJI?aAq1i4V@ns0Vmc?N
z6`m5<gIz>11`%PkVRUrF*Z=I#xc<&-_+bM->EZ<qY~_>BQ}*s}(~KINI#rR)XISI(
zyvj0TcmEzmQBf2nWmS`AIn(Ks!DvK(IAlDYEDYSge|f%8B%{*CX}no5i#(5}0n@L(
z11~=7n4$9&WD}^N)o@6AF90K9G=@lllF*2pM2Pa-DAw48yNH0I1mnBnDV{@_W9Auo
ze+p9a%bVwEagYD+-Fw`ThM?Kz`3sMuLXU&d5j#f*98FRv3{f@V@a}B}!#;UYa#}W#
zG9vRWZ;cOlZ#HFG<;-hMtqg0*<0t$zHiiYC3k;vsBkWEAPd(?x!(rs!Z&@J5Gf)jT
zp55LDqhaA}QH&=1!EgOzUj53S5+qB6y>)_i4`X6xqZv2%uCtVMxcpejbTq~qjRNv?
z%FP>Zljb>DR*;ny)A^j?aL9iDkoyM*^pB1xtLjnAABv`Za4~<dw6@MGTssD|lmewI
zOtnX&v58hrVkcOTw3HV-v;`HyIi(2cfF#QEsmj9Tz14_^g1um@Mq=^&0If?WsGZm3
z^Au?WPd=3pM{xVyDbuMyNuSk~HbN@qS;2gsljkL1@RT9Tr|j-JQduhz#EGD;BuB%7
z@vx$*4RvJ~I);LcO-T|Jte#aYoiUVVPDAw&!U+}&YpAuMt_ngWkV*h!w9T>jfRjbJ
zbNdGS{T+N25~>(a1Sst$IFm_A90btSR8;{uHHu2t45uR}-d}5U?XdAhUXZ34(^*QM
z<sS$R{(Tj74>Zu6?(!qWWg`SA$+@jAN|n^J873{Ucvv*Ls)$=nMBpPAs_tGh#43Td
zP9<CV0Sadyg^*NLNwlzD5<<{yx2TP#DhuXWNu#?&6xMv|WWttT^VRK?TWQAlXilvp
zNu$A~(<c#uN0k-$T4PGhI4h`2LvCu?%953WowQ_TEuIj>fudn5UhIZE-%HqN)HMAZ
zT@|#tD}+Jly#JKvWDTSsoy`cn03i%wyiGm1gHBF@7Q~_C;<<BNf9Ev@!$X3oW^=2@
zYHtN=-RHS_^*ye>cb!pIQ5wPhy+eNc_x_ODKEO;lK>WWRENLI@_a9uY11wHm3xQi6
zP5O*5AS|akU1U|D(+XQc?P<^&T^XWAi0~}Zlc-3b0zf(@S5;R?U!jDLwHhG>by+*N
zg2mCK(QHubn!2pWs*1QB)2KZ@9Br~zhrE*QbHA7|%?4<x36q$!8z&JSRHeq(PGfFb
zq||jyVhWmg<Ww9~LsDCyge3BO67BJPaGEFLEmpN6DNA%w(d=~yf)L;Ho$2+$?=?#^
z0zW`#jo2HZw-siy3wT6QaOUJ`?!0@I$<Z-hty$SxqSb3*t)VVUZr{4i?K}6F)CDz`
zz2hVP`Jex((<-yh%N6^<F<Rfx@}yk2Zx}3Y^lpW~NC8rCX{(3lmDH0dW?o?Mz-m-c
z5-hbK3J`IK^g%QuP!f%ys!Eg};#uMRV1=YE3naodeFU^S9V)G<@|;;#&|K~j2Q@!&
zBH@%*@x`4P*YY{Ty$Q7^Nt$gQKXV2V`IKo+SZQ>n8D-A9H`gT=OPUMr6$KM4z7!;$
zM^jgPwjJ@&Zp?|erl~S?Rnu8nBMc+g)m6TeDUyQuWJ2f%NTIPucc_nVpxb9a3zAUq
z*rkiS``WiSI-U|XYfheCWo>00V>M--^Y)u>@y=WCa-6z6zq7N?KlmShhnhv6Yr7cp
z{F~2n&r>MrVJTg#VG(l+O0c=L4EX>CW5_fF&WDz~gkHyW;bIR6AaRQRm>P-<q9(||
zv96?qEL)7q7Q*GLE7+=pG^Nf8<i-k7net2Rn8)*&-yWyDGRrs|U!_oncDKjJ9(xiQ
zd*t(+sI=6Tp+B8bl$x1NS+xZdVR$bcF`&j5&<cFImFBM`JwDr6;&MG<c{ae#E%Njk
zND?Q(5`>SuSC0n}HxRmpt8c^MGqB-CuznQq+_TT}%|H2P>>k`EnrkjUbDq<u&SG^<
zF`x0ZuY8rSeeGN9=Owv?w{P9y|NLM6mz2i&_iDpJy25;y_0_*J<{$8q(slkGqIMoA
z)GT`c%NtF!tg)s**$S%_7E5h3UJyE*vGA}Gtn@I5hfXKLccSmcT2~|J8Yv!47Q-lZ
zI;Xm#Dl|cm5JZ|YmwlQ`A@{Er3=S)@%t0uEAY^^Lixmo8SNNX0W?5RHYfa@_BJ1Fg
zqhU=^xbd>*OOjA?{*uqh(*g04K|~eWXp*SqoQ0HfV{YLL3Cgm-_dJZS)M7xP`*^m8
zM4&>!`uZxn2ltrHCsdxM*I1?<wHCT_6?^;p9PA&E&oxF1hJy*Oy!3`Eo)@r*56aT<
zU$x#>QI~iEghDK6zygbhL>iCA60Ef)^<d;$vIeAdRcD=LRDcX4MC9Se9)9Fw1bE7|
z?u)7*ibAJ(<NKt=46RF~RVXZ#)%bzO#_27VPpmsnRcT4n6i)^$M-zs`{PE30Zj3AJ
zXhwfH!6+Z=d3a%n-)N95ui~o+AvA|3XO>q4Vn{1Z(PfP<EI)sFg`e8=F|&-aNHIv7
zjW)~eE^!#6JheF12x32=DCdNY27W21y`1dYtMFtGp_{alm^0@tvp@AXJf0GU1LpQN
zQIt^Cb1GpdjHN0|Y75{L#Ao^Y#7umjp8T=r1Ij5}3yZ*@@DL^-ZW?;ZqaIhz6VHOQ
z4%4M`O^rZ=9zs>Ff#WL-##uKRL0Q&>fkX<0@D=sEKvxw?hA1njwZ>DDmGv%5rcG^X
ztc}R>2;~QK{U(zLzH)q(yVVG0Fqj@=lmuTOLxmp)gxw~w@<B+BtCD$L;!i@7Okiq*
zvWm}kFYsyqG$yxH`3xf!jb?+T&N5LHyQE(*(jq^mOy`7&z!NcbS(CrBgYCq~q)Q`=
zIC1VAeFcL$B^Xs?$`FJRI-gUSiqblbGL7p_8!Ty=e}97bK3C#Pgz&&*3l}tlvJz>0
zTAiAWrl3AN##YXXPKrgIrzsvF;usNosKi4g0Yc#PplU;2=EQM~Gy)q0RM`|$7f9<N
zt)<pALFBV_;S?)pwy3PZ3Ye!eJQcDU1ssb8|Md1ASEm(rIOAY2#&|x~_fcU;*ld!l
zZV-4eLTL8tlxbcO4322e=jhrHVENTctNhYxfSIHe`5a><?N*1?rDYn4V{5YzfKL37
zyqFO+n=a3bob1hSAwIH#v>n=wg!7l4U?=t2A54knN6ffR95*PdsmpULWnQunCP5pR
z=I=A+-={ophdzh^p)))okkt6rLz$RHuVBfq5PJhC>&26Q?%vDPK`m$`5Lle%l?TEs
z#E1pV3uyWveTTIhP8_MiZ@Jf~N<^UG%4u2`3_=%RBkZ`vg{aAYa5UgnyT@<cy3Q+k
zf$eA9J9vjW46%`q--wBpTC`7{A`F^Hp}Cb$80H1xenD?MrLHOhB>&;*O@2L&@FppY
zk3lKa$}()7L3FwfOIMe!kV;xGI)v^rh}dE6{`?E@D}N3A1l4SE_QNl7Lr2`13`wf{
zjJ$8rY%NiwM-<YK3PYafl#7a~wuZrc_WpwNeTsPyx_w;;i}VCSIeA^FEx}4ivLvu3
zb-2;*C@zgIk+nt00F5&Y5X!??*Z2`wOkH99z|{#tk(C9tsR^V+O2@pFO400eNt%Xw
zaiH;w5>I-xnjS^!^WK{ijt?}p&X|r<tWh95r1bE-kicuAd<DRGsws;a<)wt8#A<_5
zmdBrKartS5uq8!ZBCH^cV%n_+K@cG1!r&Q882I$79O*^G9@M%bE$@-gAhr=v6tK3u
zK`~NHCpoD<qsk<{@4G!+LtQ(WF>MwHatlTA0T5gMg6n-1b%CldRxXMUx2$j)O6$vQ
zC^O2sMlHl|jdcQ@(l`q#F|QG|BJdj2r9=e*zVACqbzK3G6V+7Cn7uAbJQcYUW9@8H
z0xu#+S_Gbi+7NjOjVK|BJT6*z;_S2h`L#p(W6kk!#H`Gr${1#IM%f&--ed3JHlFZV
zRT1HQ%wwK%ZWe*h=N@0<g0#$bjv=h+G+Q(p4U#0G-RdBG-+eZ%@r6Kmp2IRpuu>37
zkAt0E;tOM}7g5)mue|g!+k+vcHq6tU!TozYfXBcG0*U#)Enk!v8$LQkE(CL|tXXNd
z5qV9m7Z@KQ-H8l|6vDNib&ZggAQ9BUq5_9|u)=BegrR$FPcEJnl}3cx-B;Q;vHCE?
zZv^-fFoGa%5Jw5IH{cA43y*K{?W4O4>Wa~1LR!tS1&p(tX`13K!O>_NPet^CWx{#Q
zs#<jt5x(Z*D`z-umPto*tWIe)nj~>d6vs51Erhqgs~Lk5PEgnRB5JI#L@MF<V4rA~
zVWSqcF}(cNs~k-FRK}2&1*1K;k$;~WEg$F#M2fae^mo2V0Y+*zmz$8MlvV9=TS|<r
zk=EJTl)6AnYecE=y(Z)m8@eLgN`Wpb>_V|#dg@{4((_#BM%OiZt_l5^u-SZ+=Rra<
zYS4&$9<%W5g^%#Z*Y-G=S`Lqg%<>sFt2my`7|f@r&5G^gBYY9C?j=N%Ay0V<p%f&N
zpM7kDE7mZ*+lQ!Tsokd4Y?3q@bUI52KX9L|*7(ZD^L!U`2}Vjn;gQ|FN8`yMCQPU+
z&7Xe#8{9iOrqG62nsIdJ){jh{$6|)I0hL+2VQaAhv?^KO=%A)kC>IZQsX*5_gIxjH
z)BzSf16|+wKqj#xPzdWrtwJnDm%@!<^U~?r3E>LGszx+BupGN_uC|EwDqg2YJG{kn
zNY4KBPw}l=+Z^T|+Xn}XvJrM-*&9vSA5QR2=3L*~A@rM^iB@PF_xY5cAXE<>SNz(A
zEnd_i#r7Umbxg9nf?Vn%IvrSBhtPAK#9F)Wf#;MC(C%{t4eUEN;Ah5&c8slS{@|be
zKJV^rlWWazR<iTXYp%KW9Wl56^6d17VtI*Jh#P<B490cj+r0!tMO`~r2_YofI>>;O
zl1gi&&M`uQ7lNsg(nBbPMWD6D_gzOy2>B3JWsq*tW1NSo=ldux!t-1)7<mCv7~q9d
z{B>C0I?e5?6Xt`O`OMW9*pm6IAk7P`-{g2a!;_Lu!^ih>wl>@-7~ygL$u{e2ma3SM
zTSJ^g4r3bxBvFhNYJo*wh<UpYyKw9gguY@p9%6zFBLcKGT)TdU;b@MoEk#)~pAUYJ
zVf>fY`zq>cCGi#=8L@CMmqOrKO(-k8P~b1MkxEe4HGU94RimdVLRtttG=>08m8MuA
z@Dy>Y$#8Z^5CqKUb3Bo_q97~qI|&N4z+cr=(<#0Z)D@`OjR1YAXeAwHWlm(~{M_SR
z?(PlX@+t5F5)p9c)?Ef0=lR61|J(f0SH4Jr;X<Ru;17R?X@7)hFY(N&F2UrG+Q8bz
z3PB`jHrvdmGd#~(B>BQQDb%_|1g?;YqlhXkNVAky)I^jiW`2z5EHj^Hy!F<*)Y`&g
zl=X<U_`{j(-?_BVXj?A^?u&7WaAxkIF(_g18m^01>(X^Mw8dl=VL_;chMM-N^BP2e
zCnZr5G9JwEeV;7NP>~z+R%MNfLzI-RrEe_tyu=e8ni^@`sLxkEjaG-Wm=mTYpY5Dx
zXMF4gP(4NLx7gm@VSFm!`Ct7EU;W0PQJ8`=%@d6N_y+Sy0WF`$I_vm@0&Por-44EJ
z(rC6^2iOnLM&k?LY3SANAtVKaVMtwQ@;pZ+5hB-^(G0QFCCv-2-n{8R#X|7#xW*6f
z2LGL_mN;uV`v|Q2AkSk;{7~R`+Xzp(JP!kGS))gDq_Pls4rDJhHccH<*H=WHHiO9?
zOJPVpof3-%Qb?+_!0R@Uo`<hIbX8Lh$B**7G+1Hrl}D$s#I($at0}*7Wtm&MeYkK6
z{E!9#H*VbI=*$&9^Y8y1e(z8JkQ~F6Mu(%{`iD#ohS2Tu;gdar!9KOIY@AppjG@(9
zVmh7Tdm(kD2_i@Cs&z>ay3Ib0BdU2$p3d>(CL*7sCr5~tHKwzaS6_RR>PIxsOJ(tG
zjbX933y?zMTSL1}U6HcVaPzaObT(j>fpG!W17A36ttxVVCxjzV^xANHAL0mxBj_Zq
zC`@zcxch>@**I}D25G=npgs594-xApVK!oEc*y6smdJMYsh>TA7e5A3k2`O?$=>r%
z^0EKZf65ns=f5YDnx{G^8T?=VpN#g$6zj`;Vr!M9+jppKMtph`Pjq2*9fm_}(n3@<
z#DT+v=DB<SQbMx{vm7R4=&n#@Q@s8@qP;;moAc*?_7zHPv5UgxLCh_ZhwOtN(%5S&
zkPCO32U@2>2nUY}EmjJwiqX;)X-at*48j!dP%ackTRhKT>H-ufPvZF@`Fw)%6h%?u
zOE>D(y24Wd0*R;G^)YpZa7}zX=@_`4k~E{3tf&xv!KG&#43B0=RuCwX(BtUvh<v`r
zrAwdY^|#-ps0^EHG4t8Wlxd0(37cmc2ycQhmZa4p@gRv}^1Q%b=zmJjLrQ_s22TM}
z5codURFq|j7lj~7Dmy2n<;X1CyBKX3dU8K}v3*}foxi$Cxw{9VcGb54sSQe0te#k<
zvvr=R<zupf`?qg$YNJP)<_P1w?V8;tmFLE*vNl+4FhY_nb*YM+s>-R$9Alc`DNI=*
ztU!7pb*vFZjjT(i`}c|4ZEVz}HU?AHn4$(<EN&iHZpKJ!@EbmIsg1w%7_-$ATzujR
z$ED^=U;ZL`y+Pyj24~JJv!pGy$Z4P2B#C`Yo#87*t!rjkMQ^>qG@Y~DYGSRXEHc74
zLJCQv(PBJ3pqR~B>RH-Ka=3k)M&leuhsW&fA1;zz@UK9;{!lM}rA^&0T+ET~0dL)S
zrn}stb$Wxqv)IzIcW;Nyr7l%op*7CVB?+ixh4d6yhq*8UqE3UVaQ>=wQ4lnIhqtRi
zpxijdSY$1r(#-dc37Zil5^ZXXwvJ>`R!E`HSeBv|(g-{+LN*oN`AyPwkF)1aF|0qt
zx4-#ybPUN#$X2IChbE@bG*{aso{uhad>K;fnt5H*i38GX&Qj9D8bg^E1VM-p9!aCg
z^k__x7PNbUhKv~8?-QTgU^JX?czpaQ%HI|8AO6^fCr(q|-~aABHx@<7>WPyy&t4{I
z`P6An|N1pfZLLtwGuOS*hG@Bk_9fB}94pc}k`piYXlEI9nNgJ)p=mpz)!gN|2tsT_
zBeN1ymrS>B6E7`6>@2-ZS!41FbnX~y8LYM%2xABmAGN%Mzx*_lt<zk5=1D$Y*8I^I
z{t&&{q;-Ca^JiD+)fSs)bk3cmnfT~3CGb3IT{E4RtZX%zPG_ukTA&SOmJ-Fz5vA4a
zFdXlaPiAyh)@e({!R>3bmM+rY?{jZw??*PzgA3;<uI>RPLWZs;v*6pD_`(_Z@M92Z
z?7X6S=_UNF9%7b4t(=t2aso1Pm^6+1JL$vbx?_{h#xR>euL+)T;=V#b+;E<&g>~%H
zx88)+W$3NCF=%O^&@i96f<sg^+8vCp5n+H>^?3Y~PqB679Ica^eCYE(!<YWq@1uJO
ztqUt$KGCLSW9r!q|MDsP(1Ym^;U(CrB+E5nr-d2~9X7D8U^0P53uJ^?UWHrtVKjv^
z7HK8bE3Xosf11(m0atI`{#VERhq*1gSW0!!<p>l`Vnhg}Gzj0)>b7YmYv3t_tvDVW
z(&{zQ+7T5j20x6bw7?1vr3~O?&cd*P))jSKxr4qJx+2$ThpCeu+8EaXm6~}zzz<@K
z4c$T5Xjkx2J3^(YX*2_*FnC@7BBr;wNtyRpUEkoswC0_wSJ5U!MIDw`nzRy&r6P!y
z@IpmxO2@X<hN`G(G}~l(L9-b-g$7;X`#u6e6h-9egt{saNsBlVOs0LJaE1A7#%MTs
zcnyCT3%mDaAb06=pNA;K7sj<=7c}cmuglX<K1bp&BaPx<d!NbC2whYNPZ6zlA@UK@
z5+@-+9N{GqDhwR{DGX@#x^8r<lruV45~Cd$5~c7P4JvE#8!eJXn_<7tbl68qh+7Fs
zr{xrwD~l2eYis7k49`}?MaGgZNOo@UrT^;x#_;YttaMw<ws$BG`Xt4SUMyH$ZL_(x
ziPaWQh4_Am(LPlXfmBqwbR!s}ossV%ag#W4;*iE5JV_KN=0|&2TQV3rHwd>Lw7B_>
zKmTyAGcP`l4Z-J8yoC_7IxVg|`4rL8I#Puk^+(Jm8KyFzAZ+_s&ww>VL4Y56c%g^#
zC01JS1xY8N($4JK7z5H3eb!in68N!CZEO5SNYrdH8BCZBh6rJa;((|TIS{TDDCIi2
z*?fXm8=^wd79r8SjBozE-(_-hn{J~;x<8>Blq4y1gwJxT!P@F7#)9WZ4gl;0R6zrj
zbR%Tp*tvD}NFXwf-Duqy4bl)QkF-C;Rwbk15wmRms9W>HALj2`kNxy7VB>&5)ZdZk
z&|T^A^z$Df3|A3GaeQZo$-yyYQDS^Ww6P4akChgcM9_#3NsI_Xgpv@3G*^1mRfQ+L
zMV>26t+Co5JddE+qz1xPo4C2eVE=&e(LU0Hq|>0$?O?4yR~AnyjH#Fx6MR#WWGOu*
zXx_TYpZ;I}bB=Gk&RVa-^v+$1-2<B0l$BVrvDW3(sgoFO@l}K$Mi}i=lrcy}sS6Nt
zAw*g{bm>SM4X5gFG{To8fns*BgE0mD;{ztM2Uw{ey*%IejbFmFd_q~H1Slzy2%0B0
z=>F_4!Kr7#Yf<00PrW~I<5OS4M#o|4Oy!&;8VR&whn@4J>waynV9OHg2X3^f+(_M-
z412KD0$am!4^}tf?l#=L<#2hO1eTZFsMmUK)K_MR!5*^6U{)ZdB6;O4cK-MOIr&?!
zb9!}!{H^QgyM3bJm`G}xYaRSc7ZAoGk`A`fLWdo)=1F(>$YzVtYYkcI3JZb5*%c+W
zu0T0xP<7)vwk{cL-{W{R{1+%>57v)B%-0^r3Lh0XX1kD1E+uI+IDPUAP1VGgg4twD
znWb1;0}2@=80!Riecwkdyk3RyFxDaMgRnuZjhnbEie6(M7}ZOp3eg%o6%hq7>3m9>
zPeFhRLi`|FF!C&F0p-fd93eG9P!cBsFF4@M_x_M^wnw8CQ5FMCJ;e`m!loc;#dNwI
zu&&GRD<5SQUgZOlO4sgrZVg)34~5&qAjD|x+#GPSFL^OSTg_xPCC`h+wr+pu3Oab7
zMcw+xe~k4HeFW+IC<_XK#B%B47Ji*m&L$LtW43Qxr<KHv_K#575=0?U6mh(_$Nse&
z%nuIH<w7uXalo;jPf=Trk48wLP|7iF%z`l`JcS>H<W)h`>9VqM5}jqtjt;1b3@-}k
z_Ezz{z|pLWl%h<Tr&G%D1aCfJ(`LkX-(ddcw>aA0;oe)X(lI%k-4?5>t0YN;t_#92
zbjH1h1KxV$7TsQ(@nnea`5>UKoj{>eh6jXTKF^qpCrpQ9{2CJBgSDiU^KSj&3;kbg
zEj@pdm8VW3Jb|>1)F6SgC)e-_O*PA?hAI1ZcW6WjlfevSJ^a8Y3PVQy0sXs&WCKS;
z0UpKN7nD{roX(I^EokryBV&Op>_VTsEC@RddK>HLq9#3>Q0FCHAZhn{DBnj|LzQQg
zc}_N;QO!!cnPx2y2(J#w-q>MqaKP@nw`diT)y5LN-ZD|(qsx*Y2(Va=M`Nzt->1_|
zn2pEoeqs@GV;8zX?gM0b#%wlcK1uOyK<qUT7P2z^U=-hfKx^}}pJVIeFCu;KyYl?8
zODBlR3_BT9_xo(W_a5z}!EE~gPg_JB5yUZ*``ZlOevfMB0Gk&JkhP<?+AyHdmVSTW
z{B)HADI3SELi!#-5>b=|QFocu6KBwA%5;B^DogQ_klxC=JF#Oa^Er8*GMi7_$aXs7
zWIZRj`VzBmeTBoFd)#{YTXd_GlifD!>+2-Vh`P#&B6mvP-|O?rE7w_B?l2k+@O>Ba
zs;XQ;_Z^3s**s-D8Zkc};@8j=KElGh%&F@7M>fx!pMQq6pZ_c>@bRp~L!pG_%ySp;
zW>e_*vG;c<zV&VV_7dXWAsFQfxmF9V--2(w26uNM&D@JV5ZQ`C=YDc<jP-n1$XV;U
zc}lqAu$ee)-sUDe@ia{4aO)0Cr;zw?=A08UHZ_a~i184P4-x$#qCX%UPw9N=&!~Ry
zkGXsOJ>LAIKcP3B(OF(WojHeUgoybR8V!fz+&*CEl^cYsONhN~h+_9%i()~D5)Uzh
z^<;wSAHl&9!YE?l1aRl+bRj$Pqq*Aci*%OGBBXTF=EX-|KedYVDr%ik)iZ{}1EM&f
zNM|T%QOd^;eCG2J)A0dSImMcKQQ$m)n<#W;NoNjz@lZiwf!P#7qCB6nstNoComLlP
zO7dcY)+IuDG?SKtT?tF63yP{FD{^$5BUMIgsl@vC$%<Qyr$dg8?-3;xou!C&tA!W%
z)Vg*UZeTW<ad^K^)QFj<b4SvIhaSofUgXBmMNyLH8AYDqD~Lmdz>=5s2V&9o{ao*B
z{AyQLxcoCe&(Y1B6!-Q(Dgq@~Zq(G(Au{lp?%l_;HO(Z(<^_|TeQe-S&gR&<L|Dr#
zuZT;H-)JC}M0yILJmO}9gM0fFd9jdBalVJ{1R)8Mm~5O;RRuxZWOd^tqoaND!()Qw
z9#NFwElIL;;>JbVEG#!_=JS$L=ZM*uC=Y0Ewa5d(;}@^6vC_ffvD|8*g{CeFjt>qw
z+#k_f+Mv4%2S-zyy)BG(KAhhr!f2d5Q$C-Q9S!lS2F;BwbuFnEz|bGnWx9%UKlTK}
zyT??!a}eN5#S#X6e2DNQ^Sxud8j>i2qGZ;efijd?jjf#)*}R?+ErTEX2q_jCb{<h2
za?~GFlod)kCY3c#v=LlUr?Z^86nKr0Wx2-WXh1Q_@jHSritwaIKARzpL>hsx7ON}L
z%1{+4Vp<aV0ZBtqA~}El3@gnpcs@%FXLmC#r;Lt=98czSR@Ui=lB4;A-ul+!B&VIJ
z^P+QQjVnZote_YdsD>cvw9rDK#lLt?|IxYw?CL5{{_-z#c=ZOw?H!Olp^~hH87AL{
z&}X`P2j5gQn+fC@`JElo$fFp~uyu(vhIv+!Y}WYA=0eNNLr9;b-DLOn4p~v)N$I+E
zX3@bB;5QO<F-KPgL9@;3$uo=(c1id8M5`;rNdvDZN#|pvwy4_aUl~&|ofVXIiku8d
z%8+KOL+%Tndj4rPSC;^f71_oJC!2D(x6l6GfR*JFEUyW6567&po_grF>%^CFojYwX
zShDGy^xzm(v}m=KscS(k1onech5Z36(eyd-Kl=p+FTF?h<~>LHQj#?>MBIMGkqF*?
zho~q~ttQOJaAU`{@W*o}Q(_Dh_uR;|+jjBv6a*on+d<#hMW(saH1j++u9r?xfo9^^
zsF{LZ4<3IKZeEA&V>l_H)pq0DqXQO#neOMy3^EJTBf^6|&A7+Xi6v5@S^4BgU~L(~
zCaiR!P9fce+jp7ojYuv&PeUBR(LP*x#ub@WwIG4uBu%siTO&qO*c~EzZRFNEXyIH@
z{-s?1(OE)v*w}oA+4zu}W2_Y@Uy>|CY9}BRSvo*SO&GeVNtun%C`?^otVRe!nQI8z
zC>4Tm&uSr@Q`CGscFqb?IJqioA0CPyXxi1f!dDTUR*!i)MprZNngqU&rv!PKBZWmu
z3&Q#MR(VZPmmpFC-y;Yk>KfKIH|Vqz1d>)WqBe#stC&wGq?43p(t##q<D6c5$yspe
zN0yq7D&TO2by=dTIZB2!!UjSMtPwx-@qS;OI;kKyb&?A|{d0(Bh?JIAGhltIP1K9g
zB4>Pb#QORw(icb{FctN50=h!U01rVmpHhrR=q$rj6;)N@S&8r!`;&^}X@-!>F=dQd
zXoER+suv|l1jbg_(5JI;g4heGh68Mt;ROLvvrQN#SPRO6t_!MSMmEW4G+K0)I$TNG
zG&WE1rfjmCfvF3sx<m^}U6*8&37sV7i6^!gO(#ehpnQ+AD4fTpunz^*O>HpRpp={4
zXIYM)&Zx7A1BNQ^Ax!>9Z3|G!;%_9J`LV~K;l}RG&}XF;5=3CiIkVxEmEIDP1x}*Y
z)bkQ*?X0+jqMqlJlMG#GOr@#I3Qq>0pkL;UvN=L7Sj5Hv!fB;>0u_f2yHHhF-_lxN
zCiD~PafU54%J&GH2|?^m+X{s)3hFc?&ufxKlUA$6`Q{no%_XiS0}jbBWl3FC4hUS;
z<kOs1*yQ5+GSk@v83ZWLqbf>=;j`x9Y1Db~xiM8;l@xh_pIYi!it+=;viuQU(z66-
zPI2Wg{Su-TqNJhI@Yy)g#$QR$<ctpv+1Oe~1Onv&rlJ~+pmuTfgrJ&D$d3n@G{cq^
zwk(`yZQ!vpGaOA)SI~)tBY|~?BoX*{aRX@qQzGJk&WY0`UPyU-gqf%KQAphC5Je3z
z4*ylxIaM}gKF(-1J1lj(JehQ8o<7T~VUzor#Z);}RZt7%@_aa=8z(&b)JX>8G13e1
ze4irEG4+Ey7mGZ1V`Ak=DpQbU8UAESH6J@1r~J{)a}f9!*0}f|{B`L0c*@dk1vFMe
zXbYGhA@1*>&YS>I0VNbU91Y#~sOj1~$9*_Ba$c}S0ofe92Er5U%nj4g)D?8zf<cP}
z6*w7`&XW6cK7&ZY<BvhohV29QdT|6RtFW}fLVUv&mxDu?mUt`cG}qU8rm;o)@hjX~
zNyt&y`P5l{DIc3>a5$jRUSj3S8lu1F^y$KIVGE`g1P8YVQ8~8-;VEoBb)Cb(6poLH
zq9)Q0esp6l!3%n<Z9ENrg!BYa!zW&fv3@{dYO-uftJA^?rzL4HXq{oucv7L1pw>Bc
zm10eawc5!NE$l=m#}%_lv0%V1>?U!YZ;QfP;5TclG1jAzbcq5Ft!F@mlpcW}6Z#>-
zI^E~m)Kof0R~1PVlf*HnHe;f&$L@B-bOgrM&hpF11zaj4NqkmLEs>@<p70#uv97QS
zQrH7z<O6+BA>D|!tSXc$s7!_u3ds-m-v0iII>1Q5){7r!?c*ONvK42~bZD<7=(;99
z9-@>YYPZn{low#7j|yUv<`QwEgP(Lj2(sZ3<!As!fhjBZz*Yg{I$%=R1^LLmy$6QX
zQcAoia{5O?VJeF?l3;Dw&E=1dF=>Y9ha}w|C=YExD4!?_>20jCbm|;8c8<s=$1E*%
zIEa^dDTU#9inbP+3j{uuv79)+MJ)~UEF+3z>Z+!y+*nO11=blNV;#O{IvO(_kFk{|
z%!cIKFB5terPM=q(T~~^R$!E7_4yMlKeI+);pA$ARxGf!rWiZWdeDr}!lD8n>nT(e
z5w%-HtrlLZ0YE;SQjSunET%TtN`tSM37DIDVTJP`=B~vprN9e4H=<F>VL5~(=mvyQ
zLO#l|xxw=T;-$ty=++`Vk09{r^xCvft#WsGNHLt!ZnZdCGQ1@Y7|&-IYf!j*s<xW-
z^(C|^N%NF2jL>zxaP3&!kJc`%nZT1ivuVnFmST;>ALkTr-ypDJF;@K1ToQp+f>WRU
zBpaV{d4BP1o6dRzQ&$xIW2909OG{3h%8$^}#|s-YI$e@hm!Q$bB1n%9$&dPw7wDqG
zl%>ma9WXB3wUmnvoqb?TFC~7GKo|fDQ&_NqaC4O~NJtO*m}!b1gfx~{!1Jg9>G?!)
z#L9^cdgm_j-u5B+pii&YWv{);H%h_LXo|5Gm4mQ8r7@hoe3D8UX4#x1PN=GiYB5%G
zFp$MK&KNut<NlDz(Fjv$!uc`T&2JDXchCK(=DAaVkhFjHX-@v~i!^M-<%@0nC4nsq
z*y}rIh}E9E7D*Gr7}{M}--7i`SU=$kyL)$GXAjcS5hAi2;)Ib(nCZowZ{f<o1DJwT
z&Rrr3z<1{HhyYHmK)VM!eVF9#|J~Za1`Y~C7`YMf*)8<*&+*#b9qOHJHrCfUI=#hB
zSyLSku+{)g2%1zz<6Sv{Fbh^|tK}qF9{8zxi}CQng3a?H?7;vIhmIP)zlZ+v9}r0J
z{NP76<^m@)ytHzW_R4wu(9%2EK_(&AYIHq!N(WwyHlU=3MB&STI7o<tm_Q{6ps1!)
z^&CMB)=f189>q*jWDmMv?m9i#+g<MafVUvwxjYwMlh6;S${E-aPk97E1B7%1wvc$f
zPuytHh*#(z%qXfENs=%hH@SBk=JOnFoF=wHp|z#i?NVt=UX=KOvjo!%Xp~%F|AcUz
zI8O<RJf+Mtq&48p7*%&r{-e6&hjqQb@vF6H31O$l>CgTcdspueE_Zo*U(u{{;=Cqk
zH4#xnm1Zn;yRHaZ=)eeFQC0?T7LgwBlMRkRNK~s$<qJG1XvP6mSpd!NaIoMZKwbE5
z;}qKYSx{>%hG@Bmh-akJDTyBv$4z=GYXnLW2O(AqWE`+Fm~e17ptrd~r`4o;<`Qqe
z^(xc7L;g|_(&@l#HX(>xG`n4P`+bC`kV2q!?SP-yh4TTpu8qN0in7w=vy{4B(lnam
zTlYC^Sc<ZE)OP+6U`2h`vY>@QwL&&OdWOS01N`L{*F;TH)kJlPp9F|dq4Sz{yM-6J
z4cA(Ws5F)FP@yD0nox{fM@}{ZDp8|^BnbklvT`S-Ma&(?U~#QHffNd?L|6ff!*tk*
z5WXdw=fq))C~4AJ?%^v%6gUkc>BGTn!qI3%r`x2}h-shQ;`%#pGe6$rCxRDfVM%8*
z{APos)8O#<a4{|rXl+mr{8bmRcc0DRDUV9m<XM5PN)qEU+8Z%2HRTV9knsOs%eXv`
zmX<m9(?8DkySE9~mUyG@(=2B+(~_Xw!p1T6d`_p=!%O0ag%7&Ilp3!YGe6iRJL)5(
zLUlUWK;TJ9GY+YW9B4hvb0?a$7<c-flR~KtLR%;`Rv6;7736$MI+@T2Ba%j&m5ogT
z&m#^(&;l8U+#8J9+wZe-VvVJ`Mep+Cy#C6!7~kLJuLfu7EeWR65n*GA*2)U^5B3qh
zLM*zrcoJdlVzUQ?19}IZM^S0g*&N+12z5gL>TULWn!L<^bn_hCbx(ZG{f(dE&Nr^n
zIk(J6tf8_ot)h1FB+Z6nhi+`R-%H`fmPHPA6Ph91yan64u5erFf-JC7k|Yr}O@Wpx
z;*?r|;T?05(08wA7GD+awYOGaD4l0)vjx31#JO|e2TmYZx=!EpXu$T)9w$$)veIi2
zKm9B(Z-0TkJMZ!n$%hf_qx*-Hs~h-R8*t~Q>lg|JW##aFuy|h<q(Eo`UVyC%>}Y~m
zn?orWz5E_KPiW@Z?BV_KqZV`LQY7(YOmFoHqr-gyzs=|_1hIxlqr3=&k1-`lk|34;
zknw5I2yIBDPf?Dk%9-oJ$r!Dj-B=KL5AnrfflGz&yi{s&n70dDo9mKEzkwAVRhbh7
zK2Z?SY_#!|A_zR!#g`uAVa{|sBW^Z{;*fT-&b^(x%<>7(zu-|MkmVVHiix5IgW+T`
zB6iPh=c@6jm;>jZt&~L9wHptc1TVB?vjM}Yec0IfVO{U5<P-~&dMSMx%Nv~e?2DXy
z{-apcX7^xD<*m{<^8|&J2%(Wt;BXbvk)XA7z|$ycu(Wm(FNnxSN9gH@az1D0_C0)3
zs-kxIDCN0R6B2`@7GQ<L+F4H^BHu}0NP)H%6+|=|Ep$~9c#3vs8G-wG?Mo_a*t>s=
zE2{xo1&C7c+^0UuZ~iC$8<s!uIbQ1zI2unG&q_{z<g*+bm>WY7Ht~Z5<%dpOQFx#N
ztZ{lom9+>_A+%yXD$upXmmy1Qk8yu*Oi?=DMrXqPz?4}$G--aoE)?PcEuO`9`5j+B
znB}CGoPTHB>u}<uC)j-UJTzMzrWtjwOLA$GQaCEQ4CLbNI?S0C4hIp&G41sgR2YyC
zr|4-xl@{#p_VKEcS{tPBob-w3{y!rHHQH^iN`lbfS%h-0sjWr^K5-*Kmz66Px=RSp
z2dU7qrZP2$_wIAvZ)1Fk5D`y2dzoMPcm5ikkDue+@jZs~3A3tX^J9-QA|taUev;tD
zG1B)yNsN*R-?65wlvGwjU4Zh*OM|xH1u^Y&=h&T2s21J5N2hr4a5H%j*?;BvenBYU
z2YC7yRv?T*s(|MD2`>E9r#SnGPeIsYXMaW+ZV+C6mdZ*51}_Og$w!}6IBaCtXtK0*
z8ZV4Vj}NKGL-cgY{TsIlm{S%t(g>92A$%VpTo={~uu_3kSYIOI02#SFH`d^V5sg+G
zT^0nsqSIXkJcO0hzM`~-?b}y*awSCjA)=H#_nA-dxBj#L1KrR6EZ^=Qb2uC`oE4n@
z%ujJ>U7m+giy&yA!q6$?OAjF*<hi1<21E%)F(2jB+M;Ag_rz1&-WyRC&a3lBG0z_?
z0feVd^5k!Ro|C`w3q-4DFvALc<{9|Rk1yt51tbnS;Z9$UxsC*uy72fDuK2ot4-Wg-
z;fV6ht2DA>bY)xzE({>>UH8vhI8Z3*rZCbt|5)FETu59r8nC<q>D-BbpFE2Q5;wY4
z8gxni(l_{6EU6kzjB2s<(P#PmfBL`0|0}=9z1=&QqhrdfB>IhCr$(c7iS4aHXW5A*
zdp`IP#4X%xGIWg3VubY?=+t7ZfTTm~!=L5d?LJvv@Ice=QM}zDMG3L^ayZ=Yin)-F
zzJ5SU;0{p&N#wU!Iq?`Pz4Is&Gauvxb&F=Og{~KNozi>oQ_i~5IGa%44{0V{l=3O6
z8QP@Ob<TKr41Vn(3u19F$8~6}6*#_uo6b1a>>~dWXk(GWCk!IAsgX+3h?}Uzb*_!1
z(w5`VA<J7H)_8abw$5DOlb`xoTFu9~cX!Tgnp0*q>&uTZok3Ap0u|w_04ZH<01AnR
zuz(e4ZK=^HB&MvvSOVp-9B(i=P9J)2|BE~H176T~h5Ebx{n4-Q=X&3rI)MZQ*7G=d
z`4THwFe#1Y7Mq)0l1_}x$4C)5kg={GZfZ`DvBqd;iriXRX3{^T9E~Bd?C%ZmrB7Ye
zZY<{eNZ-dUHV;*MXrU>ED>M`qJP)f3+EiHWv~%jRgxU}X5jLzS$`a#wj7LXUn{(#e
z39`#qNC!Eguv~uf1x|0C<nR5j{x*|0ZgS=0pX0{mPh(3>Co+WA;MFr!K7(wI$wrV4
zsf9B%wz?+G4YL`9o<ap4T3GVoK6`g=Bdj5i5PKe8;M451=`AnQND}far7UvNG-aA)
z4962^GqM0{{-ALh4ks><(qetV#<@+pwh6_YxY=f{(<KTecA6oBa4{->@F0^|C9t}7
zEq4XYr6$wknrd1gtmSAt#Z$q;7tbt+GZHImg!ND=Mw^OSRNzUZbp^fE4=@wjktwPI
zVHHuBU~5BJmKYDFqY<`9+1l77Kes|&N&+i6_v907Zk*u1|KI&TnBBd>xfh?~_H!|H
zaf&5%4$p?DLZIdqq%~$b$Bt8KB?+y;)`r{|=E4vvhb_^zrZ}3@-`;mLdL*F?@q9(n
zOjznJ5l1mamQ$4_S(cIJDdYK!+AIk5KLn?JYb_E9%3^(=(@$Js%>dbyMz6(*t!30w
zf=EY5B8;`jg)s92)}S?&>#}$rt+iFg2YZx*Av6tp+kFBRFY?^FJP&-V=N=iN_R(l+
zX+baY-1o4aMw=R2<@jOjd}B*P62#cBA}<S!@EHydu{PuE#WT#GdxGq^KnXbg`15T3
z-BbL({_S67{MuEneD-H}`$M0{<TahZxUo?=N2U{)PhmcQ$uXr=gu-Cznt5TE&McAV
z;RPO6TJobEcCKHCMV=?h!w&<ROC45LR)~{?GEXUsg86*TB%L!Djj4W!^8Dcv0(_5?
zAAbR11DMW<R+eCE712`I{;uoLnHnnLWI*Ji7%Oup3F1jOxe2#2I5-9)neH4D1r5@2
z(Pj?<*O?QQW6}CcpbM}isKj;Ze6i4~%RuL@3z*G8OK5dncP&jJ@Tqq063qszTshC~
zhn}J?98Pon{6%_y=RfAlfBsGKmtN!K-}nt)`TUO|UwDBPPt&o`G$2O<#AJ#%+=HV&
zB5Z=GFl9wuOU5c75FQ<0f+?wQUgO5guOW;fM$qyCyeOd4TViE>okkp!rzyHBnN6n*
z(<ukX1By~FI%M#Jj=6iD<0ff%beA`1H7tmVpo(dC8_2+?&N7tn(Z-_W185YB!p!yT
zKoUkxbT&cPDOOl!;~D-!qRd*5QXz!GSWv<eNJVW7wFY68du^*0Ubq^l@l>$zjWvXR
z0M;yUaEe(vqp}t4<u0P!z!o*0XF0WThSlXBfAXdOnmj+?<i-;W_b+1Wnodg+DT6Od
zgv_zFpeCheabE@2jh;)3%>hp!q(mFY>xru&uqXw-FHy=Ph$5QJ7GV%jRt0tKq>!?_
zB%SAu@w<>f`8ToNcTwkLCo~{w#4VatLF?RQyb~t~WI;Z>kF9c&IC5PrV;<s=gmB92
z8nAVRF*S`&m)Y?lbvkj{bVWt0owyUBzyZ#Uq^eWuTv25nsR=1+fug{VG>He{17d~p
zFs4N1H7fD&Jddo*DT|Eb<36o+gQU}<`{a|n{q1)c4vr8)kgRNR=DCk@_1<;v4@SKE
z${nt~*(Ypth+6?a{fi%^^fGSTzt3{x63d$&CpSG#AHM^|RkBPmnk$w%9#IlfVwm55
z8*iTT;b))YV=sQ3PkiD<E?&CC+QtU0cAFsZDT|yU&nb$W(RjowufD>UzVv5&{Ttup
z-o0Ilvc#HyQ@6vvKIYQ8<3|#=Na}><*)HDN23}cG?Dw%n>AZWDS_rAz2g)T*XGmLY
z<wU5HPKWe(L_Nz~+gsNpjl|&-1qkH?E$cj|7M99{%%o3l4T07Kq9jp1f%Gu4#43TQ
z1gduUEKhk9HlxfdM#o1ql7zSsvvm0+*KXWrJQz49ko6uXp1s7aJKtb`a>N@~hrDs~
zfOx4*Gx7P!PXOgW8oxY;t_?Tu9@1_%SiaEZOe^N(-4bql<T79?6~Yq)fub@c*^M1k
zUi0*0Px0*Y&+)?Z&vWMN8J1U;X*3)7zE4#YltoEdmP{sN-hTTnzWL2>@#^cZv3qz(
zRhIl9N%O`Ul#&SRHtSZiORFr1pLi1g^eL3cDf)M@WlG#gke&yXel&U%4pV5Uu~mt#
zYnn?d%ntXcXJdqt=(?ocUUE!G;q3OTmef^BRd`fIKx!IfwZJ3C4@(+K5+NO~GVm~Y
zfygVohKJ|-q(w%N&*}H~>2z8&m)i86f1WqK{s!a0Az+C*E1dZ7C%Jv|9g@M2*S~p#
zw_iOVXmx3{1OD=_eS*AE@b2DidacJ;-Sj!V>2qoK4a^&FlV*xR70?SLQ8S{#FuVCW
z-X!B=FMNoPf8s?x`N>al>9H%Utgg{)weWl&qif2tq{wq7(=o5S@-koelRxIGU;P?)
z?rxJ8#Sb*kziSDJ==Kmf@bPEi$`wcl@XjkR9Yd=Lz7I9uA^#y<$E~U$FI>TO;xyd8
z1^q*mlGJ5RxZHfmawhOVhA<gog$1)jrPis9L{@VGF+_wBLitXxIFOJ@C<O${jbHT$
zJDF16zCo+o#p`uxfB1R!-nvFT9RlDr8Z=-0I0vu4fq%HiYhS&_H^06|*j=Z!r1;za
z&Wlv_X}<c^>vT??W&LTN)9VrEZ@!29T0rXeI8Z%$p{CRDLFbfz_9yt`8K3^xC;0d$
zKf$Ly{b?><xlDIunMR|DT<EMCV<@tW*?h*UufEI|zwjsg`JaE8Yd3F`Wtk&f{eZ&k
zh+Yd>E(@U-5Cxhj-az?X@GIuk07FF(EIR2AT{7J6E~NxxoMocX#9@Q1n4s(2S&N~G
zqlgEsD<PczVXaGaB_Oxdb3t8Nq;DV+gs~(HCAIV31Pe+SSjb)#r4tLD&ZdNMND#y{
z8Yem2-eI;7sg8p#>nrEEw|Br`kTSZt%i#7Aan!<#0xrIAiP~redwr5*iLm8iHbPk4
zN3oA6CB;Y}+6w7;l(k`2>>yOd`Lm}uf8ipRE?s18eVul@Llnh$o}#X6s=B1AYUb&j
zJ9qB#-h1zH^TtgE$0KxIf8WpT@4u*1?gS(?c$LM=a;DP*da*|(awt;*A@HO^_|A%h
z+E8m(eOv7w>{4s$yg>LqeiBiqbDFWo(R|8MD{!q^Ur^}^J)bh3t#SKsm3w(i)!N`-
zl+p~r%ZBvSh$q*ltamD6Wr$h{dYU<ZN$JKmlhF}Hk<xCjk;HAzKK>M)-YL499keMC
zL5pWU`AM$*!9QX}R*)VuH&eo;E+&AW8KFwStvma?)X$M9HcvG9jmHh9v<zp8X`$%I
zfYb}<*FL7V!C(LMC+S>#+S!z}5}KVQ;zpAwPVhXBAWUd=+zo#M&67_(!{7Mpf0M(*
zK41Rwm-z?(-G9fIzw#BP^Az&~$$pEH3ac&BC{!)*GR1sw#F7_L7bzky354%}v`S*6
z1tZYLEc{!I)ABTyy3j~ZIF;`-CGk9lc}hE2LMRW=R8@g4EW@bB+lAm-JtkZ7Im+fV
zgd&)goFL#M%@&(LWG!JFVDiN{LAqUPHaw=xGkVL*#LX5bE}y5lxk0<R=<J0FkG=2|
zw}0;&bZZ|uonh^qR<}c-AZiBGIo#geWiT%R$%$TzUvHkp+M0<}%&ns3`{aRP;ANPV
zh+ll+=V)F&N!*NylaNL`A!;Oqag6d5eiYGIj4Id7CYP>U;TQk%FLOK?@Y1)x%|H2_
z-{qy3UuKr4Kd=qoQyx|u=g?7sKQGApcZnkpBXUGOC&J}9zRNKx=MrOe4O%;KVXdjt
z92xld&6sLFqZRuc9#2?qt|EoAm@4a%dX_UPPI7&)#f>Z^Us`8xFsBt-!gS249Pz^H
zn9W9?SR0adlX{Xm-$o@t3C6=i@_bIGvq6$9apsxl>8zcly|v^Hqr&5<&wiFWzy05^
zPKlTwF}Gvl<z=i7al3(EC~n-`p?{Dgh2-R^HvisbgD5S7JY=FHy1rr_gdA8Oy|%?~
z{@mwiU4DwVkq|}^ouxKOBO!_-l=6U(q|tOS-`wKjrN{WSU;FC}2FHB$t6$;2{ri8P
zFMaV#jOX(oP@X?p|3BjXJI=D}y6;54d!Kmoy_HjUbp*N_86W~Pn1e`Bge*$3sA$W!
zJPywqjs0dk?`iz(8IP^!agxUx1xgf^C{dJTiWCz_5Jb-CM(14Bm8)((`Ru*tk8`_8
zPz(aq;Ox(bU{wKCb?U6M_E~$aUl2%0B=l0ae6A?If-UHE5n?nhb*w^KAQxYg0?G2C
zQ<p0U;-V8*^$?vlu3JI(hxn!GV#FzOh=!P9m#lFQOa1+v3o-`dWiD*SjQJUjLCk`a
za?9j^syKsCDKx4uNJm{@tAJrFUqJPH*n9UP#wTgrb{mx&592mUqrV6CzV&{NedHsI
zD~;%E5a>;svn_1Lp)p=3jbbjGTj$bJgcKJ0_SX4@JsMX_)~yyhx<aF5$jTnEZxij`
z%g?;$UC6zQxW0$yc~omnO64lHTaYJ&TjY4TQlT|5!NTrE-uL|<WYF*P)RRy0(U1HF
zpa1OVXb%dK9@nn+d-VVm3{PV;5{+dvl9V_dQ1%M3eM5?Efn^n=Vq*jtJ!1clSi)L^
z&}f|_goU&m5|!Y1k}OH_D>hOT&74f-7?r`6OZSwbyDqS45)E^tt5IfvKbA3llS63^
zEYB48Q)h)$r7VicVKgMoQ);CeCC_JiVu89pR_rl4#}N*@cHh9#`jb>@Ia>^gOirm(
zcnjOE4=QD6y~Fw-LjaZWGDqGfNiifnLDDz)6`M4*84N5i4eq_~Al1?WuIJ;p9-d#p
zb$wjNLrRHbd6fKu7&z9NVAsNKUjN$H)9?2=bLI?>J@zPP&YUHV(^oCbuc)YlqS7v^
zO$C-g+Rn%mPU86$RL@|toUq$RgaM^S6XQBqxgibHk!28Q%R;&iNf_X}9;t|_dV&pY
zs5v7XT%M2)ODt{d=Y{TWqPYd;@0mesi|w-)ncg!`rRtOSLmqziDGm&`xq0F|^+pNd
z6%G%UWnl?H5Dc-Tg>Bh5wu|kRnBF~(bTX7maSTk}a3kK>IKD~Qw|_6Iy)Ay`?|g_x
ztHS!~CQhZod*1hEmM(9Q=Za%5o?)OYOc)Sn7M-$S-6TA7aRoEBm!5Hmzw#jRk%uUk
zYs}8fGd(@e)YL5V^YgUET3EKl?Cd<P=6JFBOR!y-onPRGe)xxZ?|a|PBM(2!zyA;a
ziLXBP_z0f|->bwoh5|gED~wSEk-@gWwowAyl22;-7-fiuF(NYfwGze=NMoTgRj31C
zFqS|%k}QgG9gob)sEC|x3>C{qNDGxjq@u=2QR0ikKI|>Ad(Up_!XjF_!out{W#1#{
z#{BW~=edm<_ZdyCR7PY2lqq5^g(M0>ENK_ADvpQcyNoxQg%Nd@VHv^rzG>WM883z1
zyLWS08-Dct@1;>Iv$3^9*)Q?7H{Q$2<~Dh*Ieq>T{c?g8h9pwb6K%HAHpey+%=`pH
zSCKq%0`atkU#T!PHO<8I4C9kiOwUZyXf%+PU~+1jMzdA0WMv`nJ2gGcd%o|z-2bLG
z^UTvv^P3<2D9=3qY++*kJp^K3rSq#i!{{6#HJ0OIEC($FR;fa^qtTfnX?GF95U<q&
z-$i5wHB3hYCNP#&;3&fZCD$hv5j9`3nrrH|jSv>Oib=awF0^muvF-xH-Lvd|!z{kG
z**JNg*#o<%)?Bh~z+XJ^EJvaV?yQ}p*(f2slw2!ps~|HC2Lo(j;YbJ9cCkxkX7*2^
ztpsBfQrJx0coXi#6eX3g|L{RB_tyAF|M(|pj+I$k*}|>Xc;G|t;KKPe@=WpUGbiaA
z8yO4<mBo&ySRv-IrOTM919YTE{Kd~8ANoAyN}c)HUChiZFg-KR!omWLRttn-Vse^B
zqjha+AuNlD$tk}710Uqw?|v_jKl)XE>EHemUwY^(#OYO0+wXCOX9^M!R*LYPg5fF+
zG{#2c1r1wW=yqXxvS6zkr|8aI6=iko!Y-`aE{F`n5LZBK3}vrW;NXHTrnE>`+{+i2
zYV7PUbKu=oT3Co@*IBr27qzNQxE1rM2QPAuJIOxzBx0(Ch%b$bItL>yvYj@TZ5K>f
zzk;wF-0|wy{vaW!AG#4aJ541D*?;69rv_F2hyT|HsntrPiK5;p@gqO>MqWC8g*-Pr
z{p<w>%24Ze7+4k7+%9L+E!65N^6>3!mP-Vm`vlT)s5i#gy||B=xkV<XW|^6tqf)Dp
zq!E6pf?p~FUn2@MnoZvR&UbSETi?nH&pgd9{o7yS)1Und!Z_u7TGW{W>!<Zd9N!w@
z-h{+xf$ex?8bFh!5rzcUEn~2dvXBZh=#dOap-q?N365i<@(k?@=o?C<0&lByjuk`y
zCBe>df$Q&MYGx8?BwK9Lnrz{CHfc2E@>fsL+HIJ6jiyvBk)cM!28oc6Br%p0*WT%t
z<<h8E(3VDPRY)J!X0YulmZO-SnIYS5bKhI;#;f@ByL~**;n3}e>F@MWO0%)NPBe7L
ztOS(^f?Tq-oN@F@2aieOP?BUPK%ByNea4zCT8$RXdXrYGMY&Qz2tl*iqFSzwrkcgn
zqSYMZ?z`{dw%cyy^qEt9?z3Ou%$akApZIlNuc)ZIN=4Q>N0}UBL1ny2AaeZiG9vC^
zdmhU7@Qfkp_81H!oau3lhPc~D0J&0t!Is5?lSDC&>tXARFwtbGAW{)J7A$Ry^L)C8
z$+z6e`dW`izkHm26k{}8IM>E2DQ<t=0jA#kde+ZgL39j<&1q^a50e!H1!D|ZT7Z~=
zf`DN;prG5?!K&7Ay(Z=Q1arF=3D(b1b6}&><9B}R)4csXcTld_M6qJ`{%P*G=Lp@-
zkQa|#WG65T`a>>Q38%H9H?WB>_8A;~hAho-QM~4L_wo<^ho9%b4L4D#R<SJ$As|j7
z{>{JrMLO*+hYsDu8(#kwZoKg@uImCIq)oNn;vMgL7kAuwCm;LR@AB{e(|=}TbL)Es
zy}HUkHM+nW7^x{WeIirF^Ib$g#BywuD{-|$I!qYG0Z!Aw2tg902nyz~F$P;mEY~HC
zGi=wz$~3Xg$%G_|2N+z=raoT~JM4MuJ~lf8o_O&ru}U$(<?SxMP~36TB75I{BWF%*
zV{K%-rrV?9i^30f#JtV(>>3bR+7`ktWJU&^E>_i}T+{fi22=AhgqJsII<xfRA%FbY
zFLVFhH&bzKf-GloeuCR?zJcB#;Pmn`ZIzLRTP$TeT-2)!tb{Z$3}3oTo@*R!xcRo5
z_=%tXTkJk?fO4fW8gmHJIObP>`9IR@_t|&w2JX81UiR+akL$SC;0Bd?l{dWcjoft8
zP5jB9{TaXcv5&ITZh!C2?<2uot%^L?mcpBCqV+VURf4dM<NHwZaJ3<8cL{p|?!pvC
zLEPyfFl2dg04eg^Ck`VV*T;!df<%+1f*=pkiR5B;mPhklOupw1;*~BBe)c7LVS?6h
z`c#`zS#i((2bg@vYq@-K6}fG=O}|KeY?S8(MV>Sry%>c++74I_+8BD<TU6aTUfHA6
z7-xRpJ_eUhQg;liJ6(R||NJB$c;H^DRhux(*?+@4_r2kEwzqpc_1sxDLqXUda8?8y
z*NRSP6J6}lJNg89n&BA5efPhCU-+lLz@B{vDVHnQb^!{{vV{NmAO9uooeqZ%-NJqM
zy^($UZz!BAMlX8L_j&VM-_Bik-^0f~`dj>O|NFmTb#49s8HlA(1tDa%3Tg(b61LW%
z>=y(H?H7ez{|fAC0XA%Rix^}XFd|M=YA^_(T*Z!Ky1Ah$e5}DHh6+2;t$Zo-nflRb
zx)+B0(I2eP8D`{K@bq)5G-@&LdjAyF??1u|$J%&XcQC*52%>61IuZ!hhCGSCM$&QJ
z5rbBf4LZ1uaS-6t8!X&#5Oe%E6V3-&-s$om{^J+;@ejY2J-a7J<BVHw-^*>U+0Vv$
zm#2?jVLKNL+M7HZ4|rVduu-xI)@p2>`V48D;;W3eeBZnHnZN(fm|fV7=XuwpjI>sK
z=2O2*yVGIcfkPa*=?<o67LX%(D<K4q<M5jMUdu;+_jmc@Kl&s7zrXm4Tv}TG-tFz9
z9Yq^cSY%lOw_GJn4YuQ<WPq@2gm!QogUTb)I0dVU08|<w1SnPLcS=iOO9zz|G9<ZF
zWQhSKh|&zHU~p2i@siK%-a83<AuFer$<jijGT0cRUBlFYc^3BFM}PAYTQ96KcKZ;g
z;*79+6p9RAM^=!+LX5QIf^dK(eSEu2$*a+7jWb9#C|4v=nDW9`UtrH|d$D{A6)PH3
zEf)6gCJrLDuIv!^4PBkE(N(OjYr@DP%_6esVo_9Dn#H~Qc+WfD&Fs`XzVDCB5d>+L
z^2twsk}wEam|ft|!9&c=&SBd{nI(md@0YpxrdwEC+{0s!KE~%h_eJ`>{x^o^ueH9b
zqE2arRvA{FV#SiG=Mi*vsL2G|5`>N{d|jge8-rfo!JQl@6B3<e2n<SVk~qh4E%04z
zG>+2X_zZJ}jzW^|9ADX*BV4G^+q^`#v&Gwf;12wfpw~^=+KHLoJJ0M?op>0sA!oRJ
z^j^HpJ_o&3YLzm%Df~M12>W3SXkix;3rWDn$)~B0PeENF+!D>%U9>MeOIZ!6)h#Yx
zy2Ky;ZihSXKEQz+_Hg?23TI9|!I={)$ZU&8zjT!K^#RA?jP*?6>I|D0TNaF=#_Syb
z@bCXq?zr_{EWg-r#K;OuYs1+yXXy>Lxx9RWuYC1096od#?|<Ll;O3j}#PJIK9?S8W
zoS5M6{oTLK;UkCn@BisP<Lue<*Y!d8PBAwoN9Q>*b4D^GWuk6}N@}o$AeIX0Ip`?I
zHUiz%I8B$F0$XyGB$uijM_S+tEN$ax3p|%lC738DEKl&?(oOVUhwa`KdOL02eeX?F
zJc~|{uoK2C?44%3S|c1ph(kHY#E{ZO&8=}jrBp)Y#U}H$m>aZJbVg*HvVH6<)$s;s
zjVyW8XD8{tv_dI0R2|96l@<Q9*XOo_i|kok;M|pUE<Eu9=a$#8dqa+{Y_UDK%xQIn
z9W%tv9IV8`7KP+Vb#jsq|E-_n(9O4F`8HCRYZ5Km7%pA9#IV<4_0l<>e&oySzu_=%
zdFQ*i;qXn^zE{{~xi(|17Ju`Hf0zRY_Vcg*?Z4sD($e=N=2tgNrE_$iV5J$746jrs
z+u6c45yBEE&jQy)heK>*(92ucvlCbXRH}-NUn`P0$8{|5JuEb?*5H>JD20v!l3*X7
z-kM{$SfaCbiT3s;@BfiIDOD_b?TodpA+rbOn4f774MNtX;?#5Z;cxXh>|LNyEtBgE
zt&4p{n8NWRvmB5%CK|Hw!lU?O(|Dy8()MV~FVa8xIOVKQvte=J{5d}It2?~*^@lig
z<Nz-nU*h=3ALgYOm$1_f9{%hLtS<L?E=gEb3Rh*=J28$ZYMIu;0{`@%{tIrq<L;N;
z%d3j@Cr_MYFxckO<zsy1D}Tn}BX{!N_xuny9=aLZcCNkXtF=0R@9+NuZocJK{@&01
z94Akm{+`7A>#n#Mqb3EhiHZa3>kteJqM)t7_o1u62v{D#!dP*5O0%MytaMRy*;Y{m
z<^ohsQfrWBF<vhuZQaWsu2$GNr0A`0v9Z(UhktT6&1Q*iSFyU5Gq<n8-rXMYP_R~N
z^2p;i@U9DeT0D+u)FBCqQJ_+VzNIl3+Xl-<tBmluCvon7H$qs&VQKGPqQ@Sl8T6QL
zNzR-)&%gf<J>LJ`Te<5s2Uy?S<jY?<!IMv4p%!oQ*t5@ZadU?U(`_!81TS)MIuV`_
zXbckv4)9O@=YPq*8x9rN;Hw8<W61NIC!TnMJnM7r)RR2&<v-zu!*}wAH@}bhU3-g;
z(<tVpa+x3diJ#!mkt6)<&;D&*c=71>B+^&$e1(IlDikZo@H~ek9bo%8QVMb-u!Tde
z6QpGjW{9w=7%7U0ohV>0I?oHdBsM}CEZZQALlkQ;3EsBP%CinuX_l@2fI-k@&t1Eb
zo<%xLNr#Ho#5Co)N0LNnv5RQcrstfYc56nd?2#!o+S|>wo~AJpiA2~b+gleYmFgHY
z7B1EDI9tO@)ch3b3%0kn2p<cW-Z#U<%p~o#F6W**PiM7FEP9+fyGA>RxU#C~_raD5
zOJ>Nz&5&BP$$Q`W9`-HlEe?&Mc+X>sYuZ@fU=R;k*;(a<(=V`lZZ~(|em9HrdvI(E
zjKH!SYPBkFdF%bm&CT)qzyF79Zfst+xV|z>oz4q+5|bfRf~PZ75P%J3%RrP7XDLEy
zawN9nVY`wf4^dHyTlEOq0V++g9S7k^(j>-l9HdvGWR>x}h{UpJ7?<O{Nw&)+O7$UG
zCYhdDV0)*{>|(-=x9#T^e~viuIs4KT7G|etR7*TJh>-n7#=0w1sy4!Ca&^_OO^~KJ
zp%J)7GE_PJ=U+tJ{q~~pumr8CX_lglYHyi)@4S=57(B1Si4!L|bAAb<Y=*rdwn!*-
zPjmG08O|q~K83`VgyJ6A1;F$5Z@C}ab$Rjm=UCrZVP$=p)r}SUy&gL|J3RULlYqka
zJ*tfon_C+^|H3oebNB1{&=36td-vRclmacm5`z0*_gX&ktN)dM@Q;6<qsNYORiWFw
zO2<2C%t(GrA+j7Nw^7jm$SGk%u86a^fL4eC|6oZ-BZW>4jwcCwF*?n#Y#S*HR}R~8
z5Wa)wJGgP5ROZy2D$i#Ty?Vr~my+jICTC{o^m<ITBnykPIHeLHHm5JFG1(fY;`(fv
zRi4s@1!E}L7D8L7ToqPsqQG-TBEu0bLEd9{vW<9M(S1ZhW1`7go>E!wa_6DLBqqc0
zOPo7*j!SE+Xe<VU5LapHn*pb<oZ_<DArMh9suD%#*^D$5@4D|!Y{%l%iIZ$^ZLqPv
z!sf<W!Gh~{dExmNK`T7hrCRZ6@9c2m_%Uw1{Vv|~fe*5<xVIRK2#rK=&mFJfU;qF9
zCI8}o{a2hmcXp(U`s&5p=nO=`Iu==smnWFv0LunP8VGW-EJb7r<4G*H$n&fdpu+^W
zR${ohjmnBROUojTLR{BF`ejN^8NU<}+ZHY3^Zamz^{Pv)IV8;lGxKw7Z+BR%XWaUl
z8@R)pV-VS#J-Wu?!Zc%zl*fBZIK#b6^e#}XIS3Sao}u!>fhJ9Jh5$xl@{HjV&tSjt
z`?xlao1SGQ%&2T$;tluSM~cDosvLdsD92A-#3-A7XMigbs_m0JckU%lWs05{<+&K?
z721W1#9QC}E=qoh^QTU+v~-C}OBYyOyF$O;qrJ1uBM(0cCZ|;PX^fRv-&o_h=bq%A
zyWhwIKltP9-o0=1q6Z-*Z+z36`GY_BJ$~kA{ua-@@ZvRZ%2zAqFI!OzNmkf)krsto
zDhkv4+>E+#RgkK{-Gw<+UFdZSR|elJMy$hO@o~L{sE^?emI~&&Kf#$`n#;|MnW>y!
z-!MK?BMCEh9k96hwkeVf;zV-kY@fY*T9`!fnXNAMU>|QcFXB#>5qSn_T3GlAAT{J!
ziY-9o3iE}>;k9prl2u5xOia>=5}KDT@Xoir4PiS}stvyS@MAo8><n4zu({EvX(vpt
zo#0E)f0?KAfE}U94G5t@6pV0<;75M?!^|%%u(h$y(#4A`U%tS`<_euohwZIx9{l_l
zkU~+dR+yZwvAlel=bm|-yYGE7Z++(nnV6ip)?pMvaMx>I!ykR(6a4f~{Unb){zSnl
z7xLP{`noSs9C|exWRC4Bv`8?r;K2wKX_k+)<G`})!u2Ol5hh321?fxad<1zB<XMJg
zTiC)yIxd!-l4X)=$zuDoCed|d2}(-SYK$`+1Xy*Ov59e9$44d(?JHZ<n^hd&W_u_R
zXI$LfF(qFjg)TaSBiA957n42L76h>(i<Xe$rV;5#P^r{e-BILzhs6U6<T<#u%j(K1
z+Z!D$Er_ECW98)T7VGO<Y;I_x1g`RWi>XC1@i}nAA*5xqw0wz9zsF9eO}pD6iXw)?
zAs5eIDBLz22fO6ZAN09+`4R{B9pnvnzn+=Nnd0Y-IAu2;I>68Tt)Jv~KK6TDSzfs=
zVSU9!iQ;dR$W-A8Nv;iMIK->EBzZ_43~_~pUGdRzOcn=t^(s-C;o2n}uSAw4L~)2?
z+1R#49EZim;kY=YR2)Gxs1S_p=kC`Yp*lWBw4JiPxj_&|Oit|Qg%^)==;oWb<DPvS
zym5}>&z~dFIrqHj0H;4?Snaqp8$)Upw|EHD=-RbLICgP@SBj+5K}I3U<k-r>o1CXT
zP}DBIK%+4rYs^w{rg-x$b0{MTg~!JBHpd_REEktfaVpXbO%Z+}!E$ZVEGJ8I!0?5K
zzru?zy+E(GLzWkww%Qa1@o^lJ<s;*8(?_mCy-{XtyvNSY21k!R!w)|2W4!l0KR~%s
zM!OzqGUVo)Zsz~|<^ROr{kea@@e?Pn%Y*P8V=g386%rlFXiS*mIR*Yfg&B@?u{?=R
zBzctLRXn0B#q|q$hd9Yd;t1O=>{*g108;@+;$(`lun9zk?J<itzu^e2$p)QH#!h>i
zI7ui^EOPwx84lifkXvp($iDe0UOIglEfja(yvP&Vt1RyfX#^qF5kbd{6vHqDqZUho
zhO`$WVvSN6wsCN$n{<_?a&n2uL6dB%Olizxe(xc20fE$Px3{?P<nyd9Z*X315fT?f
z917QgOy%S%1>n)A9_Q4Fml$-r<Y`j)*;2$jNm6o(le{rQWJgdd`!s8Px;xvPK5>k9
ze%}YU|84ImK=!U7i(+oL;Sm4g7ylRj=l|swIeYHhtNL;;ezrjh3z;fxA&UaJ+eg-2
zR5nBn`dHG!s+Q1EL>2@%tvX4P;*~47<uXAhCX7N{+s1J%qId`tdG4y1sx1hjI{nH0
zyz%}UX-thXSdUp*-)0bnOimx*sb^l`z)gp_`L07O9vI{Nu~ibKc+)#><mev>mb*S<
zwE>Ol*YaFIp_%B|1^FgdB-=ZM{KZIU)R|diyRT@Rewy)4pJZZz%J>X#yJHs$n^1VH
zZ*Fk(!9V5P`Qsc*6$7I&6oRFe<B;ar$k5&JsV{z>r=Nd{Zg-nBjSBX!fh^C6q9V`J
zG{cx4*0P}4Dl<9NVQXuZ7mq&00}uQJ?|kP6@cq&VBx1Pb*4y~#M}M0SfB3_^c=Xt-
z7VmFdA~QwjO%*Z67tmRQcCR2g)I11W==KUh;v|8(U$A0>A@sWiGu3vA!q6Cu?^6p0
z_!5*{q(3*s_q}_9@tG2x4OrV6&<#S|<`|EE^#vy8X4!vmj$7|)(%FiL@|*`gG|xl-
zrOh2-o!NG$(6Y-_0p6z|Y)HFcQyZnoS2qyh$X-kclEpo&hKA;&UuL$`X1HsS`mROZ
ze(Ma%atJJsLA$~ezxPpIeEu08&4#p1f<_^Qz;PYoEG-VP27c}LKE@Zm^f@-xmPz9f
z%eK%!nx+hf0Z9%?lA!fA_Bp}SWSxbDbyim|a_q$?_`VN(nAhBUKaMkk+X%t_{RjB%
zkNplm_BVfwC!TupdywZN$tpCCb=Ceb!E$YqJVEOi$CeoBB6N;Y36AZPrV430BRQ%<
zHq5pPeaa+F!3BYZaB^H9Oczu*#_rp1qEc<31`)mffGmrd^vAifw95SaE*AF9(wc3t
zad{Q3VfxTKJ4gEr)-K~W6C8i!2hUYqxPqZC3coAzEI^19WfYc_xOR&y(DePw&<L?q
z3)ii&w>g1Pf;_S4^(-#0pP=8{WPME&rz2dk03j{%TwR-9ymb6DmzI_Yq9NHxuFZ_b
z!FiTrRPnJfF-UMrE@3sIH|%j~`691-&Fi@H);n=M4<j{sny_nOfuH!vALpaL`P;0l
ztp2s*`ihAXP|V=nY=|3eBJvbR8B|h3S^|^hSgt)19Tuc<f;&+wWG#dQ%SPBPCRL<q
zVKL=64ufu&7=<eme4z=_5@*wd&J#~F=kMXbM8;xunm>K;9A{5<k$%7<4?WIpw{Nj~
zu}T`Htgpn(j8Cv5HQlhuFj&HKK;=1lB%hQe5te78g+ycunIz~zpEQfGMG0?gsyJ+>
zLz43k)7?qv<&s=mB$mh4_+C~LMb=s2TpTeN$;g_KPwg<yUUsi8U%tYX<>hN~PI6>1
z1)!8c896c&Hh>@s1KK+wt+6s;81Y*l{jXfOaDkut$-l$o^bD>Mq{Dz)Z@QU(_iulR
zzxA_!hbybsr^5fXG1nS9%5iqOh*V(XqADZ&er~XAYow@a$kG(2;$jJjmK5szj*Zb7
zSyB|Xj%^e6hr}wvNi#fSv8`>E@@=jhy@KyAuwS|~nhhR2bD4ANJzOFlS$dWu=eL+^
zR!E|h^+8Ir+9Iu_tZu@-D8!{SvVR%T=i>y+F1nql;-zJV8F`vuyDsi{jTj3rRHV;d
zWRT_zg&?;qGFPxO=dzJ^$#=ReCF_JDewoWam?tm0SF3Alth4qtS=p}>5?02bzP`(*
zA?`<Xf{13NL=-1{^5eh9^70Z7{LqguJ~fFe1!<UY<G~yGpZ?kZ$p8DF|4(deZoJA7
zFWAd&JizO&A@c+)Gl;le2tlVgmg^KrB?NgK<2CBomP8BjvOG`HxY%G_hkm<FWO6(m
zQ$jPys+>q8+7Ca;oWGBorZop^)BMh7PILTN8`}^0+~=O;?z?x`v!_ZHr>rce%ui0T
zB^B*>g2CW4zEeeI89FoAmLQ8lY~L+R*|QuO#pqraLyEwsJUK%UKq(#&AN>N|?U=rn
zWZEWieAef0U@6VXHZOB33h0Xh^lXqQtr;Zg%kI{N3zu2CbP4hGd0ueB$n{^&^B^3u
zwH+`qUST+l_~=J}nM;>0@?(GVZ!<nISrk=TbL6I*`ITS!6+ZN#A7*KJ`Bjej=<>Q9
zxUf`o?C@ZGtZ-XM^O0ciNRlKBVZ2c=yJRtnr&KA#iGzM2OXc~nz5(fwY7k&)L#SFj
z+YQ<I(lN%p-MnqFM7dGrmwx{;k32r4QP0@wuX6tz;NTtsT5{o%VfVblij%MtPBGKD
zh-j8ES%x+)jsONdwC`dG0l7j(2?P<0w*c^G=joa<Uc1B0CqK^GP!mdz#3?f@mpFgq
z5a;rk?BZ#j4%XQg3iK%EM$?X8?h<qCr4yWd={Qmrx5N^nAd?8ll_48_A0w^L-mt@!
zwH|Xb4SKyHy<V5~wKd-Po(Cvbs@KM`d-m+*SAX@_c;JBtID78ws~78$5W3a|=_DAp
zPio~j7HI8bgaD(E(n60!hx0tcvi*@gW>F3bVS`D@@*E+>wMI>nDukWmI0hX+uPfM!
zE;8X%aK>UPUXx45wh2}PQYYuk@zZn{JNV;1d6ts3Q)<;RxhbJG49VUEX}bk0-vF|>
zp29v-AkYYrg9%CUl#1)(+I36<wv!O#XG!CXUL-(CRPGRT$LX{cx#+XGm64<cJnAYI
zKTTdvWN2@<X}3r3JyA5Ri{i3?7a3hZKSsBgBuz+$DQ>lJO?vQ=2ie@*<SlQ!pL)Fs
zMv!L-yBFs9(I5R`e&aWOlTN4m*9z+^D(Wm@7owy_<3?@7U<H%ML>ZLiB(^1ST!%c*
zNOyXeG)Gz<(s6MtgVBauDTHO?x;|N=$<l-@iE%xTXb@5JY#bxWv_#=Tmb2XpXq5yr
z6ApLYGRuW4UE*Akg)W!RY_NA>j+*aqu5*c%a6l;9#Hm9TXJlDIn#7n)qg{zkV!Udr
z&=5*vN<xqYeXKkqw|x9knLHN^f|RkUN7=U^OEJQuomV;6-eS4aqdN%L(G;S*%8XRm
zzP7G?W(>J1Y|Tbr;Hu3E*Z%+NcL=ls6&W_x0)k;imI|JH>QSO7=7)du$GP#~VH{zh
z@|4%!cOU<sfBw(;`Cs^X!YF=qfb?SIeP9f3H%6>Sn8c7^;f>d^g)G=knIan$of7G|
zNY}<TMQ1ElG1w9Zk6ahRyIGv#xE?0fR4fl$I;2LUYzv*ib{aC%v}jc<ZkeCsO0R<w
ziYRlrcx8u~@oBueLx1}Uu?j$S>F0t>C~_+yOH#BJ2&vIYj$5x5fZ!y@$0Z-e$UG;v
zHJ<O2BM7pTrdz_7X<_W`dUVSUm%3}L4OSRLAzcv`XF^jX^*37vf%>{=T-zkB#hI6}
zSrtWrzts;3lbqZLjz0e^Ns{t?ANT-!_wC0LHag3=<MuoF?|$wd@UQ>(|CKm-bz*J_
zdxTL^_=7Fn?h>RCaR#oK#@H5)>p-57Zgo&;j<7s%Je<*{n<<5`9XtV9tO_m0FvRzK
z!a+#gcSd<G$fS!&Q`S2HlVyv!Nr!vxSm6BS4so871}^7Lu5)nDYbcjpPV8JH3wnfd
zn=o}rlZ-S=NRk+n8MM^sB*v+afzsr0L|F*bpf_Uk8p`zwSq{A*VWQ?!DOr%F80oRC
zYn*JaajDa#(;u=;F>W<{`#jHHX`Y9Mm3YWtP~^GqeT5`W_@N*AFnjhM04Wi`-S^zf
zzy9C;HGlW-{47Bfy}E=Mtzl&ePCW-f2)MAgw<zj-uVA@uBn9rSRE2W6_#SN_&x*pX
z+9;$=!k{2ZRBF&qDSIwh9=R4o!Y7Z1Tv*H4-Edi$hqt|Ml9QJLVwDs0D;#^F%k4+@
z;klB>pXw1u0m@lpkUFS1Lxuy?aDdjL&`BGHSgncT??^Lz>A?0j?4AQa)~ZC<w8Mmn
zN{KN$Mn?lE3)aWRd3OC0C$^T^?DSX>IhLT%*ZhWCbS>r@Qmz*7SDy@X?Vm;zu^*iP
zQ3mo*bM8`yPA?%&1l8)7!5H5Bw)fH+o4i)o-E{L!{2%}B-|^ue`<o1g!&gPOQeNG_
zgaJDs5gl?ek_<ejj<f`pRoHy#JOuP;Z?M5GaAc|0g&(Z#lBooh=cqixv26?~t}n4I
zNt}aa+@fn9##l8QtL!trSYxo0pq0T+p}*dt+1icm_-u#UL}|>B6hBbtT$3qDt_tjC
z;dYea+2xUlcZTN*aup+`LMwygxEKXVoKvdU*sj2!u&~L89$OnjwtEqAln?}?J$)3f
zZ@Ki>_{o1cF6QMQngZsNbQ5%}5ILN=aF#5~c-_6PXKrS$Xn2{N0|yT9f$x7mzxVrp
zK$>QM=_mhH*DD*pD$q(mk|289SR%&)Y)>Ha0ZEXdDJ((aP9LvaqB=c=kP<--SxTO!
z=)9PRX9jW#fMpoR*ute;c8b%rbkLE*#J)Nkm)l(IjG-Fayy2D#H_i*(8;|hSFQ21c
z8Y9a3IF3ZySGaJwf@KSo3W}FYmNM)Q(Ar=LLCN!}Hz$!QCk^`8S|PI(o#muL7Z0(K
zbOs6Y<Kr}?A=G771CL`HD^Q<e!f4ieL18W|N0@|fNd0g8%zV?&zHz(w+J>Ms3<nvP
zF0L2jZfO(zdccqT@Q-uH?RSFE2n9d%gMWiBJ@|S4_)q_g{8bRRDr5!?Ce?^ej8sJu
z+rl7*0cl=P$SrA+3=>?>p*&G59B`C@oWeOk7m1RgQE0Tvi3c&3<Kz2<(6qKK;>>3E
ze3jMh9+xsueZyPl7FjD-sUIlu$ngvK<tBML#Fi3!xXy|y<5&V)=cAKPMi2y}ZkWLL
ze5zw($Q-g^h^-|e)#y}_xf#;7P*TuKQ>K~|)C|PF$A&97y?GWf=F-wNwxd4!n(4Ft
z%i;TmIR53RezSl5wNEq<BpJ)=TYx}Fn;;B%-v@qxBS&sVm;xL0-gm!;Cmwr*&wb(Z
zFDo9e*D?xpE+9@3?M*C`A%T)7Fj<dykRyb~5rSl^hf^)%&(9Ry8l7X3j66=S5oA&`
z(y3KBVZVzbT`Co~5F3{sDsY%SSm(<5ZBFluleac_%bgW&-z~8Y9p=klJV&iOL6G#Z
z9fPnhbLM;v+ZJd&z-UdDB@Fs~v?^XaWzVNRJ_A`!Htb<5g_FjhGLRa}bx~OC^b>YZ
zOfV)40acdc63<?`jF^~V%4k-*!$OAZRm}5#KjrM%RX`x6!*Cezp$~nSn{K|XC<%q&
zfgk)qKKq$J=l4JU@mD3!i#H@iH*RSKvZVN2X0W}CNffB05R2aGLcIcecNLx5G%3C>
zN(!lwt9EBb7j6*4KBCkFzg)C7+!8_0VejoNj=iwKQ<uk3V+rrOtHRrll!@ji`TUnR
z@x3OSJ0YH9@Wn-*e`y-um56)@T4SP^Y-gt^yle}qb;Q2?Mdx@J78?0k%m{>v<=AA>
zVLObOTbL^B#~RaIE=}>pGe=?1EW35U`F59F7rwumzbrZNwWykxeeExme=i%Is%!sU
zu3@JW^ZbkF0Ex6*f-vGOZ~s2#=678y?B4aBck^RE^27YvM?NyzKd<xBy1+=93?l|0
zG*SYVAQKT}st_WV0+mMCwu|dG3g6UGSIj7NO^{eCP}-oiB26<K%ffb@(eJX5sllJ9
z5bQ*-<6>mQ%#jMEw!v-9vvy*gic=z1DMAR^vV*tnfdxt`3>LX6U`vcJrmpKyDb*1|
zk>(N78eNxGBO5;pqa|^WQ*TwNlntaV;f~G9au-CMYTYpCC0`@DeBD3)I_n#^`8>|)
zZS??wB_v6l@cR2+&z{9agto9OxaXd`Iezju&ph)S=DV!qU)lI|(1CK9bl6ArwvkPN
z76zdU3!OaAu`LVRbMciz=ZYvx3LP0C$(0&001IqNlqC!YeRQ7TNCy=sv?gasjl#1M
zcDJgO7n)ddovFz>x7@r)INadiEk|fA%y9SJcVM{|O2?cyewzRGnHM0Mqh(t-S&FnE
ziX!^G0Hrlezs%U=EKa>ap2j4D9;K3pj$?F|kXvcdauy!l-heocDYs@=Ydbu6ausLy
zG)6kCF0Ya11(OkV{dH!pF0Bo5q_}utl{D2PamH`{)~{iH;-|U&);lmpQ>j$=$N%^r
z^WZ~YVY}VA?$=-qHTaMZW9ST#!pE4x>r~|VC~?5D1fK5{*PKR$ZkV*lbv`;pX|&0S
z(}-{|z~l<s_E2d~V`_p-XH*5ulv>PA*+^@P@kW)yi@V4MLvFfZH_e#_x88IF+Yz8L
z&YZi*!%HWL-6{Oc#z`%ttBI40VHlx}p;WHYoSMY0d8i~M9Y*-BgHCc}nv-WqaXNEd
z`awdPB$OwoSrQqKt)Isp6ByU9xwS*4uUh=Pdc6{jA;}fX>zkxXlVmxc_`{F$gAXW<
z9Jv{zp;RjIQ$P8WJpSYpbo#yPe-XVNRBEKXE~>qWtP2nYi_>Izk*}78^gaAM$7G5;
zO9~>4kZ4uB46gn!%oF<kE;>tbr9&RajL*!I8b!&9*f(CMw%EXutISL{xb3#x#KSFa
zxa}~_g(>d3?-m@-Md_5|FP`L+fAS0}+{L70<D?0e1z`}<?exjDrsY?em|4JSH6e+J
zyIm?}9}&fvG{M+8wh$P*M7!T3iejqcbDZw`eEReX&YoGc<+6PF3W?%-lIKyBbLPwy
zlEe^(8DaRV{KQZGZEm^cc0l0y9{=PQeu2+@;R|eT?_Bq5FiBCcn(h#mmk?qkj+z7M
zsB>+}LYFSB7L`fVFR*i#U3BTRF7A`gAnIc_*9*LzS1BylZkU2RrtC-DI#Xx%hB{I@
zjE_rhxoM0z95Qpu3~tNk;LY<`vKWD#J#~o_Cm$k=Z)L%AaibKgl4Aw~!uEhnYg&y-
z#KA*|>6zl5wmZnO2VVz=aXcSWs<YV*$c6))12=KSn&7vOpU2-fi}qd4oVi4-M$hi6
zMNj~8tr-TIXP-Yq6e|Wn!eB7qeINKC7Iy7H2!U<e{KJ3p^Ze<j{+!i~>kqdu#2DN{
zCo&)6iQ-uoreMROb<r)hB(_tK6Ot^$l2$?9Q~DaCUmHW5=EQM?HVVhG$V^79+$2*5
zt8Az>Y^<>sJz-I*mzdl&L1YF@9ymy~KE~qS-3Ys|0$sbZ#?p%`G_x9(mEy}BghiHS
zBvFjkhH}ZHUTqduk}4z5BOKR8n+$F8Lg*NQ&^F01Cd)FKUW4|U;LM9{DAh5>rr+tJ
z@>k!w%@|Obrnl9F+@LbWLk~a1o8I_F_Ac&0YmMi5yz`y!<iv?n1jF#V`aQ46O?cz@
z^JK|@G#Vn42npA`3oYBiwrw0Ck(Py3_K}W_jB})>crYp@8q<@MTMaDN!I3t$kO*Z6
zqnI#>Id}02N^45?22H(5UChwmlw7{Jj$M`9{QB3@T9_ftGR~Ym#vlLjpYWgm!$<h|
z?|+8Y)FREXD(=vyVl81z#@5z0QIz1i4%3s<RBBCx<B%p1LAy=0T1F=^I?qYdm@Ll_
z(jkg8TG&{Aonx!p=!q(gsU|00JdV!vHT|<!a~X=fN@BzE@;WC^p5pAe)BO5x{0c9<
zbgVcrX~pfg-_G~H|NY;tVf<aM&e;`2nvq2@GRl$W8Yc+Lwy`Y>M;2Wd=}Dv|5lSK{
zwD>GfQX8w{SN+0G!E&&r4Mq?p2~iqzd3_zF74>AmRHzZo1UsQ*d2NeQNpSG48)?qe
z$+C>)i|6=@&wZX>|JcX*)6YIgYkrPKtxOhIn2ZOQETi4-5T`kgZ80%EMX6Rt*s>t#
zbOXx1k4bWLuE^4aJkJo8L##%k*wqTByDOMlM!iwy+}Vo-R^of)y{!=>%CNS%&H3}^
zS-N<EKm3E=<IL&P2ntjC!-o&^wzt0ZRr&p^FFZvS_ejD%G71XWFH`g(q-_^*AIU`6
z)e_RR5m8={fGvqtE76>vp*-Hi_FNoc;|L3(4M7wW#1W^@FOext*{m?Bwy29K+G~;v
z=hkqlHn+U-wX}B4kg1#t=T7m7KllXy;otlUzyI5R&cy6~#wTjn{W3Lu9<35KHZ}-@
z1kZDsotdLrA4570Iu41px2V@Ds5mU_3DXFjWk|~-j6q{#m76@bvWcFoFg88L(Pv-8
z<mP*x=a(+6aO~JIPM<!`M}GakaO&i7pcn?-a_en;@Pi+GRq{M|;&jm&7=)N!f7H3V
zDkkhA92a5xg-l7+EnFs20a(Row(7#-3``Xw)HtOAYo`qi15_|%`@|WHQrN>2Eap8d
zUE|Uv!G&`JN>zv5x9`OpFQK$z`O+mm^Jib+*MIqs`13z`ip9fsFgsU+?I!i!Gms2P
zE-w>CDX!<>?LUB+nJY*U#*l7rV>N2<bvQU-SqQhnAhXF_A3ZV4SFS9fXKbbyCV2SE
zkD`>s$XDYI|K*iy2qVo)C(iN6S0CYp=bz@2AO9$;SC&RVeZj#SZ{$aQ^haKmnD>WQ
z&>|yO2|}a=jC0hmmC_nHB#cN1(kW!X&{$VNWYWPa*KphtLRvVMEbIkEp$nTP8QUA}
z!a=H)Au9tK!bkKqy{#U?b68lsfl7UhTq#z!)_C^mXL;x^zRdHFKF4@{8oyGalxeK#
z1chQa7?5W<mTl2!)bZW&NWeTJO#(d6Me7u86j`2;=L(@ED%J=iab$_rD*?(b<5eoG
zU0E5CD8Dx`$AHc>-FA<))iu^O*Z9&`zQppC<&nySW^wm!?zrQQ@A_A~q8;npgP$k7
za1Ph6P;zVxHqr=e%f)h>!h_ScFhU@N#P)sMY8i|n>J1ondxe~ZWl^uzaosYRk{D$X
z3^ELct)XO?3%r_QL9b)F1CAU#L~psu<DWmx%9S-X*0%V{7oOo?{o=>@)bA~G=9v=1
zwK?`o9OAD1x6+EvGCrvZ;*jm_4i<vxi6)bilf?-PL72vf?hd6=i8LCZazUELs3gO4
zG)7BY+ec+CJ4!Nr<08*Kb-b`Pvn(w0>Yd!bwnhp<Mw2C)^|c+&oju3-OXv92-}tYr
zEG?6(6kFQ-#E*TLTD5xJAI+u5P7^Qnu{{sZ_Ar)%6gHOaAZ>{e8Z8S%xR3(NvvGX~
zV+?Va5)MMLG{dqa)oLB5R3;Y^q)U`*jM20)ghJz*kb{Xw(oWd3dp~PonXf#3p0)K&
zHaB;8<mqGl(r^DMfBNhyFSb24n4&eYkG->pnQO<)R&2sJq|@tT35$t(gYmI3w9e21
zqBuZw5<K4{3qsWB<fk%?i@``6$MaE+PgllFEHrrO%mq}U5SGBw-=oB*=w@fyu+{Ez
z>GBd+u3YBhfAG7kudbu>49jx)fggB)O1XU9fBWfAf12R*8QfBhlH+2q5k??v7vZ^J
z35-(^B#;uTRKltixR#_dAZT~V;uy!WY1Eo{UWH5xOl~pkrx*<Dk;6bsyjsfMd>z#e
zxb4Vcwl3BA>Zeb0Wod=AwGF=bnWy+4fBs|q;YU_D_IRE4m0c`O-oQOK+{$!thN<zK
zVHmKvxr2pZZfcB~>1h-SB_v@QA-1<DmCNKoACn8RBt|C*zO67u;<_a=<+Fvw>`i-l
z{IM6wqxk=kJWmtN%E}fePoC!Ng;V_6ul;A%R<2+uSk^!FQ$J0;Uc2sje(58hM4#P(
zdIRB=!1X~o#mLlmK<Z+yQ?%uwT7r79x3e>Vtu};Fp@lIqgKA8mrHgTD4C2B~V1*iy
zu)u7vKN%o5a_+o+ky9t8c;pK^bhdi*+5;Yb=nVh;KluxO^&^6>JbnYGFC3v&-NS1S
zAK<3mc_t=f@?MwT>LxZ8<FnJSxCfKR7`u3-d0RWUwOTO}9xY)MRgs`AAG_KhElqO8
z7VNugKcD)-<77dMv?cbd*rEH{Dq@}+MVuJUoxj49Pd>x(lP~bcfB0K;+ifsK%zyGH
zf10so^SZ};bNvKKyoGJKIF5xxVvj67uaYB_5hHY{K#tsb42C=_{M1#RBc(;9RKm4A
zPy*XXRH87ZunbOPf%XlR$$%{CF*h@f*OsgvyF|C$q2KLs@udqq`iV!`erz540yMi-
zW~<}uo*GA$Hz`m1#7SC6pa?;&UZ!5Fql`vtL6j6DWXCPAeL{;7_AbZrKp25*JILH7
z7znD(Nv>SDg4S6950kGdQuSra7%;hJ*c-CBzRCL5CSQE$OLTU2P`Y64zTx$+!}q=K
z>gT_rqAt6z#Ocp}3KR6moPy|}RaV%@SRg!$G*7SqX*6-K&+4i3oO<$EPCxM+7oU8A
zi%%cr%E_~Au5L1n742?F7#2WbHOy(JigvG$%00&X(`3gVWpT2~{v*3MeSFCIQ|CDL
z%sIaBr{{2!12}37rAq8OFwefl-E2MkG~@ayn#~g1J6)8@7#puLJ2#JO+w{YLR0~ub
z;pPcKX@<iAh4Oob%2UcEf#-m0d+bCp<9p`0cyWo1oeovY25ksl#Ww9Vb@Z?MSQ~><
znyt+a%a>QVaN!&u`Pi=$42P&Z=hj<q<;ao4*Zo@ZD+!k#ei0F+<X!>6&{tLCQ3zS!
znJgoaMv@K_HkWp|aQreCUb@Wk$yKhLT4VjvCObP_LY32vBcfc93mcu=^zxAIpojL#
z?8$n_qZ>>(We)6{<J@+@r3=d(J9(MUK7E0tvWRk;XxC@Y!Zh<Ub8J3yi5q(z>J^tx
zzmLi@n&lFc)6+PX%`ghcgh3@aPFh@N7={Q@u%~pEQ4$7Mfa{j%CIgzYb(YuG*zR{I
zTP2JZM8x0ec3szHMkAnhx5vub21}PO@rh6TK4B1`mFCcmhd6lPz;*w<>DkL1|I<Ij
z4BKR$Pnu|So+6~dSb%SlXT{4>YeU%Wv3B${Cm(u(6AwMk*{?j!xksPn^3jv5Us+?2
zD7M=HL6DKA64lSxPIGp;U3Bg<DPO?6@EH51s~oy*H%DLWbN<8yUU=d>fBx}vSjl0m
ze3DF+x#93UH|*cb_7l%CoqUy6tHjoJhg@Y$PS#o2wF}R-=><blEzn^|IgOFIX4o&x
zzVj?YX9*SG;7N^V`)mgRllvDqckUvq+uPJ^7o%bL-2z+w3hUeD`NqZ$7tSwp_UtKs
z>o<OdC=9?DZoTa`ZoTD}>wYcjxg{?Dm*2*8dr&E1w2g^F3X7Hi>_8L0BJL&}jDX)@
zwz`;SPNDz&tLRUE74z9AF<*EZ^W-ti#bxqL(eH)iu_lpaRO+&x#B?_|5S1w!$ycd9
zdWl1Gg1cTf#zRj;TspnMGfyn@`ybmPitk0`3nWyz_4XQf+&oYC;6)B>eTMp&Pj7vT
zJPBxxm$45WLO3qE+rdZ)d5)K6pk<+*cl91?O~cA5ONA^qX%D;XI<lMRj~?g3+6pbV
zg3*GmNxtQM`s=Q5jQQ1ROO`7xFRgO)=qXMdKgOSZ;`hn2f?<8*;X~Yc*In2BT2#En
z`sPuLNYKj0<fd?nqS)Ovf?VZ=kcBkldCai8$=2E>w%0Dv+gzf%y-dHe&amGnOEhti
z6vE?5U>QXaD2Brb$2fQgQhL!z{F-2VZjyK<VS9ap&5Nt7J+sVIHbzyKaJ5HkdW`Xj
z78~owvG#WGYd+yHL}`Uz^=UR+*tR5!60`!NQ*1j&Vo2hc!o^6TbdK!{Y)|0YHbIb5
zX^gSGwZotv;<-h8@*BwPulg#w^M+s$(B0{<vAMxRk33A0#AvPAy?YmP^K;+T&wfRi
zI;j;a&%TIPt8-vH=gB8tWOkv+_~Zm?qfOWwP<JG~Jf*wSCK!g;(xv2il<FQvYm`<*
zK}2t-gRw1|Ri7-0nW|M;-%t$ljB*xYL`KDr*%iOY<u6M1-Th`1Q=C1Ta_;0dSx})>
zla#71cigw1nvq=k^M|NzKg?Y>)(Mg}QPihiFEckcPr1?{OJasmM9WBI5aMPDoy~2c
zN{dFGk!gj>lcIy~TR67Ig>IL)HqPZ!7b!~%g<_x!G0InKUGs+g`VD}>fg;Nc8*AH?
zN)AuI@C0A_;um@Ed)`N>T;gqSdmB$Z{mga6d<i93KfZxeF0m&!ym;yylXahFtBLAq
z;$cEfdJM7w{cfKqj<FmM-}UjTJ{XNsia3rK^b1w`x>q7kQpP=xEhQMHT}m;SJf)&i
zZqsF+`_dUE?_Ge|agMIUoL}n@=`s~tQZ6~%erPv@f)k%U$==m2w@*$G^|wie1F9vT
zshMd?wHjHP5yl}62{OuYQs}k&#C4Zyo|Eewm8SR@j3u#MkEQ+!Nh@Gwd6Sa#Q6?oY
z;j2PQxOShvPK5N@*NiDhg&MYZx|AG;mtJ~~#~(e!{cn96zVGqc*S?k)j~>0QS0<HV
z>B$%H8Z8djG9LNzGb}8QF)=+wwzfmq?$L52CQayU?=b8Su`Q3X=TmMp3(Td?34?&n
z)(*zDXw^z&al~xB#>$qapXQX~A%={q6R|&dnsc9$9D2i>(1)iv@m#{mW81{TD&uuY
zrS5X?8~0H+78gJ9Fs-#O^V%D#45Mv=ut&2|Vb`wRRH`k~IHn(l)Qmt6hj?ki&iV$!
z);O&+BQ*sEu;PL71g_(8X3!=YpW?#tb5x~`(iwfN$-ndM`mWcv&GXc-c4d=l#pQ`-
zAK}5zeU|rq-~meIGVgrXyLj}m$FFOi%iM6`Gf&`-jWM5QtbG1SCT2>wyBE-_TbQjL
zwrydei?Fha?hP@e8nRMDlq;a}!Xm8SgVjx3M`G0~n8ARt@d{h%6x(5+YJUjn0GW2!
zlb1R3JBHVM-#%<AyzoNG6Hi36yW>nX6=UN$Z+~}-3F)x>k<;us^Cb=+ERnCBCfZt|
z*{I?lK7yE<gD^n#y4XS#QY>MNzI+8HDkIS;5JuAw3Bp#$a*gGU3k>&;arBuJG^{dN
z9<ie$zI)!`cV0q>m$^w8V-O>Ep(Hh&Kf6q`>LV>Uc*7B1cmG>(U6*&g`&~Tvr7vCI
zy&X+^^D>s@)7YiCa(;<YvrM&GL+JpOMtCJj66Qo<K$;{t(#CZiJjX7&%UYAADRCI0
z5%^`7T&4JRk9bED$2pFdpo~J)H1!*Yte$+FvF7d6X%Jru7_9d35)0`IEZ<^k|1`qH
zET4P~Gr7vbkp{yoBuXQ^l1H=EqU4syRYn}An2H3EU|AVqI3P_csFFhEnq22d4;&=6
z<*>XJ66G~ITU&Te;S3U|=DQ{({1tC>p_lryuNCZJqbSS_T$?kO&a>;pac;lu4jjkj
z$dMylxNz|m!~9+u`xS>I)SP?xY3AO%m-(3q&YxT1@#j{Uta{wEI7XN!IF?~@e4O3;
z_TWedtu%2Il4a>tIasndrPu4y8}_jro6SMaQ=5`Z8f*<hWq`wUV+XPGRnC8Do67i&
zEZlK32lwnDh+-@Or=DA8@cc4^?H3pepXSB`8B#zn98j%V%uP?ztc{apf;jGzBq`EB
z5{GP^Ut;ye6C7CBN1A0MSxOWnI6_k@SwvYzSg&#R^aYYQ2g|`2C<x{jd=qBW*9UT>
zX28fL>aVn<1lP52Y`dUNkBB8#eSeLiDD2jEC{=v^_){O}jyvvT&*FaG^oBR`|9Afn
zlIwWvFvcd0b1px#%=E!&rpLxuT3_e+^-adyG6$QJL}`L;3}ekP{AwX<p|v526Y?w@
zi8$2BGsR%gXE5wxSq|HI!i#xAu5)Z`WTwFJn9T$3?m9g8<x8X!7V|eRuxn<XI7>jo
ziL>imdTxi>$`&_wh8(Wi2%QrS1IiXmO-xa57PfTBaGNAekx0@wp}n@n#<?XHcON3l
zGm<PJNpfrro)pA+LR`tYytG1^>Eh&X6an!_x9R#9b`lDvm{qtF6jc497&ap*ja(bH
zI$eC<<8z<=ZydS#Cg$fCx%;kr_zmYHBN5!|v$U~^Le0r9KEbXx?PmA<G^bB5@zC=t
zOjkT^-!;ZCOK>dB<irGf4j#a<9E?&#K}eP+g?^JM^1Rb-(;4=#9h<cv=aHUIrZXHZ
z5P5>ZW4648=zf7ypV^{5bt4P+9O3Z6Jq&{cOK6TgxkB%$Rf3J@neINuO^b!J%&_01
zUbWaYJ4dTNL6QmLxJ#TQ2n0zOvT^z%OHUu=hW$5^WEpXu5DsG;p(&Sa!Ym_bH8^?V
z3{jke<zqxbUnCT*$yX=OJ=el<oRP0G7+tvETzv-8%&@YuMY&Sq5C7<++;z{r%*`)w
z|NZyldM<H%{W?n8xCH%_W54%h#_yk{u{h7#$#a}PvBO-o#EtXg5Dp6x#KLam9ZiH=
zE{GBR0YqUjj%NxJrm(SrwY7$ET{aWLpT#bb68P8<MTk;^QgVzNhTEL^*nLdynrH6*
zS#CQrO+U!6jN$Pw_vn2k!CQKk8&@7?&*TP<Qe@k0T6Rul|9-^$evm$FpTi_62#E;-
zhA*9?|L{{x-*%|@1C2rCIVF{$9Y~T0HC5%wr=KT?6vpu}MzNz(3>JL5a()zZA)wGM
z{HsO1P+xHHyh3Kfm{C!bXV<XlNoqKIVTD?w%AbAmcQ|tEt+d7`c+;ESj9>Bz!|1OW
z>8~xLAdgcvE}x+?IZvZiV`sC&%Bc?3y3O=rl|0RnmZ4UwVvkSYSk}l(Jw@rlf44&6
z2OowZK^Wsmi?F9zKA}*S!LbBJ<XG5v3yOx_rG5NKR4~EB+&(5ITV$yMH0{e9^sk&_
zD3@{fbeK5ggJnpDG18N?nsw^s3PuTnEF#YpQWPTJ{Xv^<XO)TRDe_#A<vA+N3Trc0
zlBS9{_1RkABv13AlPe)ku43w5g+sdl%a&Nux_Z|~V#Qx$(i>$)g`LA=&pyTC;vOc)
zCpmP(4Sd%69C`lAj+U>`CQOiNgBV0yIls(X@4JC&wL<jtF=od~%*-^A(!!{`P*j(K
z1ck5#cCCyFK<62`QYZ&1;|)r!8hMsenzZ?E=P%Q=A<bjRa-`>CjA6bNQng=XY5N@Q
zzj&5Vmod^oYQb#Qp=6erZEP{Kzk<OKBmutf<Jsekw<b_|P7v)d7!JvD8=YsgxB47=
z{uHm-zn4a1f-sB-vzT<yM`j6SU$B{HtfSam+90zXgkK>ZbV)|R`BqB`L8DeJ3a~UK
z$ujc40#3uS1oc{#YPE`EI|w7lvXmrAiISK!OADt2V;Bq)wl;SV20r~4|Bavhsh{KE
z{{2i$POx(2udhmdE#^j$YmEpcYirxQZtnrgC7<~887BQQll3Y>S{S8}R)MigbYZY<
zl?vfbofqy6S}4jjAHU*~=Q*COdGO*<>cS#TV~lEGxdnz}4n;jzTv|z3S~<g}&?s4m
zd+te6rqh^3Ql_e9v{6Li5YO{)tqGc~36#=A-5r83LRmICQ*?uf)8{X9^R8Xgnqx$9
zLY#yJmL}7bgkUG{v86g}Z}-V<gZ3aFX5^@EThIvvwMr55Jj+P)4E0yQV@V;XmdjMi
z<&g)dB+GKrG$lzR(k#R1!pk;@Q+7teln;IBi~P`sew;nK_b}FKvA%izyaSC9WEx~Y
z=F-{Ay!XCC)M{12$De1R*<f~V3^9@~u`CN?OHvfV5jf2X^bIO2K+4L7>ST-Zc!Mm<
zC<%u@T)s@pHl$gE$#QJZ18vw{>r)S&<ig5Hwm$g`LsPky=X;VJs`&*L%B##TR?ujM
zQ9#Kr;oBCI<J07M&M@4f-yf1Gn><VDZ1#EXnU}cl;6duGNrE6C$Rd(n7n#IVN`iHj
zv0`%8F0GR}4x&^h?j;mQqi>(*&3e56=BFufn*NpZyxFKxtJQEE7okPb?MM>BI3h`t
zQJzD;AG5x;1qMF#XP@9_|K301h8u2RYHE_Dr7PDj=Bgm)$ac!|OBXo!+9Q-3b+QK^
zWxCcx&NLDBBxEVrc3}~g4+|nm!v&KS-mh7PkQq$3FcURI5>s}4E<JIP8JR=SgLDc?
zMJKu>u5feyFw<uy*gP}O7B*6P<l6B1;Wm@;1t!fYT2tlXpue+)U9Dj4+J{)&kI8aa
zc?s3-pq&z=ig>-t*{?pwkvs0dn%!N@iv^h804t1XSc*<Q<h(FkJi9{d*@#-1aBD~`
zzK$UB&0;46lVjsxAdX|gI40HK`K9YvlF9KFV`F1@u7@yI@%>@QARG|I5lU$^hMji6
z#Y<Pfz?VMv86No2ALqb<1I*3Ma_-!P>lbsaQMo~wnC|ut2X2|c^F7jJncA33rBOx-
zi7^^6azaWnU38W$w@_a+np`Wi7C2s+hEpO}3eT}P5w26J3sRL)MHj~RKxH1hT1;nY
ziLKQYvOs{b3KCv9MUBTa77exW651G|B*k$Y>Q04vwT4!jVKO9%6ND->-r_;b+T}HN
z@0+Ajs*<KDX<9I15gC>Xy}qK;)eL(*l(G=UA<JTln(5o@?E*@^UvwsOMV{yURTO4~
zELiKl?-e#`jGQO(oFpywM`MaS&*F?>zi`}n{>A5c>l@#~^z;<fT9xiA6eE5`MO}XN
zrDiNvH@c*0j_)cC&zDhp!1CpkiP;&(r>4o3!nQ1O%t&q7V7uN(*9L+?pY@Rwg4T*S
z&Nw__)9)$bBqIoW%#Ak)LddkCTF*FK>k}s(Izdca$W2ilH&m)F($vXyMs6}{wK{q*
zz^T+&?+jU6+hMDd5e_w!To$5;B-mnZvck&N6}oF%%p@^!5~8C%Ns{2$hGkqXEw3|-
zbEvmSwPK)ilqg!)-z=^|2qwlS_|;$ezW{jX%MbCnFMN(O=g(cU8u?~R2tm12VXQUI
z#N;F;zl1Gp!Z2hQ4B6b?K*WMLj?qRVjYF>!QZ9Ks_@&SC{`Y)86Em|cEX-fmqD~n#
z64A8S?Da{N!qbNRr8+u{SzBGDHPNCuF^*CS%d$|W*yM~c*jGCd7{X!5_Qqzx0MeQ?
z&)M(Q8Rmi{%LoTOrfX$FEOKcm=P`#R?9+yR6p$E$z@mbpT(S{P1C=MJJfl*rVgS2p
zvlaH)>U7x&Gh!tXCBsY{lJ0bvDwWw-TV~Mlm`KJ*;ut*)Nz)A5Xx1Fb%JvpvmO<Si
zS64@`dL$<M%~z{6#=rWP{|XE|`Q(#)^^vb~Y3UMLfBVAD_e(UIEyl)Lc%F}x4sjF_
zh9T|t7Ess@X>Aa;WDrD@JeS8Feu%fe<(-U8jWaVdb6tzNeDqQ?*j!ofkfu2$M{(<X
z1*Q62zHotwg*nD&=P-GWv~08-8El)c%X19FUXRt~r6T4^6U7O)G$q}kM;NCJhh63;
zS_DC%VOXmr+)~*hPBz&I3&CTd4fUp>T6K}s$y7>iGU|;6-5$8L1}p78S5~)K?<vB-
zK-pn7iimq_%uiHUUc1ER$_BgQm?#O+L60Pkv2DXe+vVJ)RR&28%`p<C=;@67JLLJ)
z<P^X4o4<uInuosh5TE{w&v5F@De~`-=aq7m@v$kUr>7{D$~e*{2to#fK5Lt62nkUX
zqm9Bc4xLUwrQ-A1&-^J5{J;;<7@uJG?%mh5sLL41^<cBSy2*h|Ayvr!riEx<!5nSF
z{ym5T`-|euaf(6?iXx~~DY|w>!`2R*I#v9uTv3Ws-dMBQjw^-p$mR-6kAv_L%3voO
z%-CB@<)=vceKKQ%vGBzZw_!tR4B{aq0ZdLIgoH_#aHmImd4tW4A;>&*O>j@Ajp}SO
z-}D%sc$V&^BTQ=r<e2t0Q4r#aoYkts3ui9SiBin;BypawtpieF_*T6)Aq4aD3;e<F
z{{dQQ9(ni?{^Za8gy&v-o-F^?opK=r^?H-px!o)-E>f%3a4eTzuSciT;o{PH5P~2G
zMti8o#zvn;y~7v&{FA)nJ@2PltFd?Q-s@V_T|Ivo4gGe2$~Dr_j4ya79n)^_P^~to
zR_leMkd&ya-(w7xEJm;xLy|=F`kexEX$*OmGrj1L_7!=m2$P6bvy4s(A5~dXj31Gx
z1BzgnqEdqtV3h@q?_vwPuwK-PQpv|eU|BwcZcM*3WU!r}V-1$ecrzu+`_#r<wmNO1
zXh5@(jhY~F(dn`bLEEOi-6M?@P$JjG==D3D!>ZLPKl4*R1;ELZCpmWfB{nuTzoT}b
z>$p@a73#G*j^khnQ5^n~gx;VJWaN2<F&b$s(l8-$ZH}He#+|p{LA};wY;5dRF6sb;
zWT+*>e$I9$V7lQmziW{s3X#$#Pg2q}N4OSYoPaS{wvA<5NZ&__;#8{o21i<?X+k&*
z2!n(}CAiS{*y$@gdz-QG7Nt@PK{E2t1J@BuHhm1TFcmhYFxz!fT-zq>=d7&nu+@$j
zM22CIqccI-l}wM@EG|w_wiO$LnB|K*)Jl>)^NXZehJ{24O`Hi*DOt~L2Fq<i0a9zS
zFd)TXMx&N*wjA5$9q)K2?|9F9a9x-8y#IZ?`@whfLm&DOgW>R72g0(MoS0;8Zl0Ox
z8Or4{wsZ)?kk!@IYh5~}G<lvQgd~bGdi?>8T;a*59^-xA|AQ><-p!LwJpD?-`W<3k
zh>3<KB}@`JVZek}W@c`VB#w}lgUUu0F_s}tGcX`6S)5KhF#@OOXrr;ELzbpQVN9Im
z?5E6<aOo<A+v(D5)bZ<e(kv&*5)2Yo2*!LLBLyNi2m@LQ$RZqD66LVb8PE=6hFTM4
z8l@~sqQrzRn44_kYuHW=Yu%8Fv{;y*A<MJEsX;3e45`#?TM5ISA`%Hg8}c|ICHuDB
zdE2&m{TtrE8{Yf|9M|Rkx4oG+KlVoc;XnFE1Yz)PV?H)E&eZfY6B84ZN+qPUiK2+j
zjZLz=h<O2?R|p}A)11L@h^^Oo@%d+Y%e&sm+}s>596fs7Vy<A6=Yrutv(X+h*Yw%F
zXD?9@AT0+SC#W<-xHfSdBecPC3b_+mE)_9P5^@a=MV=1_1A<|~&5q!VsIxsx@i#Y_
zm>j2E86yr8;!&P^j$yjtV}wNH84`n15}X9ja|nAmOBc4;Xh-y77)F{rx2V_-vtzJt
z?<Brd?DS(UU)-kd3U1xApEOUe<$06~ge6(kHvP*x1fswVr@?>({T=iCz3+W5Z+*wx
zux*=ny!&0e`+e`>2Y%oO==J;G^4>bO&GghXyLatoc5aSprHUgRhC#rU<z<pAMQQc2
zJP*T^ZnuXcF7x;!U*UZZe2Be!_wvXik6pKzXN8hF9j)0JYU0j-a<zta%Wcr>VZ0K=
z0VEkX6mAjPfMXZdVO6UT7LF3YfbAFmO{a^AGTvY;p0y6pNj2rACB)n`%<h3OfvAtb
z!V*xf`vto;O+otw<R#h$&n3$o2FrbR+L|4e(~lH+HbzatOf6>ShG~2YhP{-PD*;uj
z#=>p47b9b7p>)nrSV$Yr+cw)5yL4rarBdQvhmgF$La1+^?4IlLzyp7SdtQ4lmb7^7
z8(+uU-}4UM^{#iZ)Be`lfu8HKxO)$WZamEX{rhP&nmCqAx7Xv$>C;3>h*CL9Dbg%M
zN{hiTVry#~&%kp}KF*up^&a-^-3L5)-D0kFF?>j3L$Z?*4iYL2hq2ap0hJPk&Zx>Y
zM#y4hiMb|sf^d*R6pj-bY$?fAPMRd-Y0l(4^m9SnSKtgNS1R~!8I|QESq{b^9jG*1
ziVaqeFbsm69AR6SP}AEPGVI2rfgufYbf$4UL1n^WY^H(j8B9N8ur<UhOBxdsWVyl;
zBXffcR4y6zBsvO6vLej6O1L`N{8q>s%aYuD)6Lv`%Wc@U&8@fH!L6r`^UJ^dU%oB;
z!LlUvT9sC_MWfNcb-imvU8mD6U=}90MurnYpfW`i#kBiv&RsaiowvV+@$oS(oPTA7
z{3|NzlrHSM0<CZpiz{oZG)pr$j*X{%uqD1-p}W0<v?QfU1+QGf@x3BIQXmXixu#UE
zAg8BD<B&?V#BeYq><pQx2%=C>ov#vfdpMOQrDg@+l!>E=G)|DY0<QXvTExoX%36m@
zOPh52F<DkT7@lL8tO#Z&E6h!o@oO%5q)|qb?{v|&=Exm4Qk|V(QyA)YnW4_fQ$ym{
z*b4d#f)LAcD7iLas7H7Ug~8_QaTgK^Qq!zfdEos&fbYA-2vS-k(Gc;i-VPLxbJ_D~
zRO`%4&M-GSN4;K0E6vW%4p9^mMMDOIK0>4z6iQ_v6}{dN&w)oC{R(e?=ewDoo4W?U
zy&kIo>I(uL;R=WK%{A(ibJ&)H>v<prmR+Ua9U!H^FZsB>gYDP|i{izIz|!ECO2~1@
z;)Ig#5rq+9m@+19Vq>V(sziebyI#R7S-5_!z~!YmLQAAENDGjlY>TyC%9XVNgES=*
z0u*=>T9#m{QDUm(;kgAZU+Wyz%g{RE;J*EoTRv@(Qz_L6@+ER3Nh^Y#V23b@khVq1
zaf!kdQ<M}15b9-}x1x7fFV}d-Ti=1_x}(vAMG{6Md$4c4eAlO1sWCo2$<*W&)oK-`
zHQi2^B#uerm>>us%P>k~lmeYI2ox^x%u`SBx;MR%>8Yvf66>og(~?w61}eu>4wqM!
z886LaJ2rl~1eS$mSLtkQA#ICFwTfFQVU>Jvtl~ZygUkzYY-@g&Bn+t3%M5!1hV4Gn
zu_8zu>ZKCHoepEQG0J0AN~S^-h9psp%oGBHRkp#h$h6C)wGQVmZ_^nhq^huJ@@>VG
zuUQzcuxqAFsqT?Q3XCG(+(AplZTH+tb$*hLFf{B6t2!f13`wQI`mo1f7+~2hW!IsQ
zfG+Y|e~UaHtJitp`#*?n+t<MFNgN`+&FHSkbDyz#i@BM3cFpgi*=V9v&eqm8VHgmE
z1A4t4kw`H{k*gd*PNy@#bK%Qh`aJJ_{|~Th*RJcD=ZTUeTGGj5%A&;Z^hrtw77KZk
zS`}eSXimfOYN1uvY(T9HWgncv=NpO~q|t(X8qw=Qy^PcDVz-CvwsQui2Gdp8*nydO
zXnN3?D{y&X1bLBT#?fF~WL}l_YRc;IHr+5M!NKGf6`3;c<;+f%X*4aYl2eH3YK>Xz
zpjE)W*WZcUH-*6<ERWpeB*q~cpWyOfi*~P%?fBG79;Tx)!p4{sg9CgGmn^XmQZX|&
z!4H1_2S)`>k^L%5klzMoA%I$`%JldY`}ZE;#)CI9IXOw5<y^UPh2fyjpx<SCdk0}f
zm`ssn34)aM%{C<mzWU(jdDA=J&En$Xb&a{o3_3L=iN>}BJM9imrBP^)It~VbFI=K<
zfDjVL^{^cmY1x28Qxrp%(Zxtv*2(h}$8$;Im^?|SjtTMrD)S}MB*eAKNXMh(xMW$;
zsX~lQlI6%>IkOpT57}91lY}`s1B}M@1eI}%`gobjn1^sJ<VYOc=ny3glMB<>ewAoo
zsQDg)z5)e0_v!W{lDH^VUDv{-Kr$kz8F5X2^=jOVfKtilHFw;B<2Z#hpOmE8=-~2g
zmh0G*eV<0XPOH(RR4SphW;h%cqv0$ej$(4;NW592ghEF#(t%TFPjSmlx6o=eUvXSt
zQBfBeL#Py)(Ucv*FbP=RSmB1f2T9U=gdBxhy@p*aV|fmiWh13PNQpF{l_JYhf+!{l
z)8Z6>g<q*+35x|_5Tw)^4eIqWNf1zKHj%bPxml;K1-Z^?_hXh<23%g-U}LjSoCr#m
zpyml0W0F?gVZ2_ZTCd`IF1gCtSXrmv3)nXdsWn}8?`~3Sj$x_|Tv<3_*v6+F+vv##
z>swud++ZAwG>OqdFg7+rm=72Zf|teB0Isyyv$)9J_uX@i(WbQG=+UEO-+{bPsZ?k-
z8Z;UWnvEvSW^=?Ejd5KU0MGL&4C;!90A+|H&7hxh{`>`cy)F~uldmMC-*F`-Bi1R(
z6!@ty9I&>2g*|%?pz?zF4i=Sa1>2WMd(@$mBW52GqYZIZ2n{D$R&<{%iCc1!(q;yS
zVU|*<SE>3wX_!!|mk|<5)e@F)P%5RH#B2;wRyViU4g%s_;A@+TWm9vYQMPH6DpblP
zT*oF?IosQ9hGE3+csFHNGFPusY1A<_2U}=tM`C#udKT!KV7uQVG#N$+vZT<rtB;M5
zrU60lmleJpX)!-H&#kxL`ZZY!tvP+_6xnwmFO<t=YSkLGYL$AuPOVl$DP6G4Mw^K1
zdWbM6#vEF~2-gxMIZI1R3<iB#t+DI6*+<3@<QbXKR6M~T9&&N@GDi*^DLSzuY)rk`
zz^PZTJO^Rh2q{Oz5->^?-MJ_t4pUU7(b)L)BF`5sK`%_HkF{ttD#U{UmGLozV^JAv
zQdfdpXYBMNmR15Tu5GZo*(XdbDgx@Rp;cFm*K8*0b?VJJp6`;WoVBG@y4@kW`$K9?
zpS_DM>SN=W#u%=!a7rGwDY28esB)8)%{BvNFqTakM`$4!pO_`e2K4)bm!04Z;7N=9
zd-rkIy|1|zb7KrI9X(2tetYslty-hmEb@D-HAZW!h01fHIL7ll1Ongp=?@3RI8+x0
z*f7%cx+!PQoT1-sGc|dA3;BSYUY3#QjJhwNyNN#8LEiUfM3Qoi88A5kb)Q1>sjzg!
zl7PhI8lo6893Y|;6bQ$IRvm1Q1>w>QQka>7dI>t+;$vIDWDDdBCJ9N03BApXtusrk
zZLbog4)xqH=4hIw0d*g$%^LRD6r$upnxM{KBI#VhZuh8^3<qb*ShKS*F;?j90a#U5
z?Hc*y1m{*R(oqWI`NYEzEd*1udkB*@?RNKNG1ov@+T3*b2shn+^UK)EXHT6XioP{@
zK_D1wwV0ZkW@>VZndup(r>DvCj35{m-FYFXRx0dt+XcpIl;^`?&h}2k*;A)TqKNT{
z$?KZuS#C&DjWPz;l7vaZR=dOe^b9Ij7zD@yB7)^2q*Y*PN6cMY07jE4Rba%j9Hk8y
ziQ{-!QsPcQ9Ojh#GG56g&l6m?j6vdfKBZED!Hb3&!B)i1R*zvfCJ!~Xt?_D-Qp3it
zS(K|Dp6?-T8>JM3exD>vn4C(meVf`u8Q-q}Wo%0#EE`WcBz;LPN(_1-No+t1l*%wh
zP_NX8^R%!Y`-Wv%g2~Bo_U_+*O`F(g!}9X-w}n^$c%Db8T&7eiQ7)G$mCJ>6O_q%g
zg5Wp~aZ=cm6=Q7BsUeSYHa9m3qmWv)_KM^Bii*0#7$RekQeZifjGTTnpxy1_3JceA
z@Ef&4x!ac5R`KKWI3*Yih{Na_4cfs*MTG#LQ5lE~X$ibC2wUKke1zqog8*fPC|Rie
zZw+IPoojPtsg0AyR9uUhhM-zAj8!Ze)f%O8729zUmW9d{gCL;Y*+9ga>FFlrJR;qD
zkm~Mz_z3!5g*;Bky%Gry=OWE`waNB{E5t~0Ezn_v^c>#u=67)X)U%vEe~L8yT3`)e
z3(2jw+``n%bn#-t5JeF$9e)Y+ZLn3Av~XP)*Ym#i5R^)UVMw`L#`k?(*DdUqMxgmZ
zeOh3&AQ%=8L8sHE(P&=R#E(rzY*M5s4(ORm7zR1Leiz5`iY{%%1<A+&x?uU_sUiwu
zk~qd-3QUIOp^YI3V#sq$o+2^0LQxZ7dk#1jCd^QgMxoGB(N5c(+3B#>4RBIJSvpME
zlCqQ2v@B}XI)1r~<vIu{QK|sZcel5Y#$lpX#g8G~GfrjK7@nvyD1k~9xu20alFO=1
ztKOuu9+C({rZpy32uJdo`(Dq5)5p1d`8-+vb)6UimasW=;~^#{$BP=@7~(kQ?AbHl
z7DFke#Bm&4*TwgJe7}V6mx$sBzf{8WJRHXvjbBG|l2Kd@f-qCG+im*Y9<_Soy2SeG
zii{#M1|bEuEy&2|hAG>f9Xw&<+Ae;xfp9FaErjg=hAc`5dVQiG1PTvG+c;!_X4vT$
zc}|M71-=hL8k|xIVY}$TfZPgD7AVJJV-WG&`8Jo9dw6k3)wP(bYHHIN6J?vRT7z<>
zhV6O?%OcNmdcz^xJ8Q^DGdI(smPVLEhnZX4i)A?k<qCO}kd-RL!sS$|nQD%)d2xwI
zSmYXXP~`bL-u_;W9)FUPr;n3-lRUruw%eGP{MrF3NfM47J&MY|L!Ns@yes7jl}d$5
zxk3;GR4Nrpe(74CtC0oM*XQ|gkg>kLPP@Is#Q4N@&GVto80ZXXf$Iv=G(;r{a(xAy
zGI&0W*NOtnw!keqIYcRDrvv=~!i>7fPK8V<*y&&;5i%J73Ri5=lonL>fbT=62fY}C
z!brm~h&X+z$J&)4l^~*N2^PJOR;f?T0cT<gF*XaO3c|4=%Q2lc`obxM@+eI=5m}19
z^Cs-p--GdaXw)%51a+6db$L1sS(t9JbfU{pN>XLey&=+fdGGr^$cryL!t=p11q=G>
z65vXkyY9S`MzenH%StJZz4*enWhYvejqjHzSIX4tbsCKZ^+uguuSdPnpjxfsmwfWP
z7{6+*kph$!7-P}xrCeEBX3*<0Jw5%hg7UAsvb>PU5|%*P7MaonQNl0?aco(zHk}e9
zW%+9iPNQ;?G$u>a(cW%wtm5#V#xVv}z;*=Sf+GdeauCACm>9H$*2M&++t1jzG@!RW
z#IiG7Us4&fuy$GaRfkHsgy(u#(ncWBMiWN~y<s0m8Y+!4wwvMQGfbBzDLH~D@-R9l
zS1vlY*j(3?nq{IvhddQ1WiX^j%jTvVj&N!1GHV+f->_c`EGgK%u!}~caqWGcrYToe
zu6$byIVnWpx?|g;_qvPcx+H0W>$wHK&a#Vlxi&E39iWTJk4_YW{(xZ+P%c+qah|{8
zPMyNwN`YexQfa~@W~05q?BpDd2Avh7K{d?j?+ob<BOKSFHPNCv*2FS~&iWS1mzSv4
zT8z(4A#{O<)3!jmE?^-PL`guL1_YgqVKn6AQoyMV4Y_7<T(W0>jj=`*+g2bF>a|H+
zr-HO>;v}Ni*(`_!N>Q!WC^c=$)jCeE$JA@@XW_orv%Pel7asXCo(+DzK_V=cEvSfs
zcAOZn^OT{<5s~IopZz3B7NNh9@gy`-INW~wo!G83k|i)~Y^-zo{Hbf;^KV>22pqT2
z{;^#fVHc;PG*7QR6lP@6q>L#>d_bJUTz%JTZRid9Y;LUMSr)Eslcec&A!sKiMU&%5
z7aI$y1yLHYy}iNI^ei?mS{Zbnp>o48OzFoVwk2rPYn1C{EF<Z)2du4bP_EZ#jn@#l
z!jU#wDx@vIDv~yd0+J{q%wxiE$hqNw3rP%GFz<OR)IA#IDwfJ1i>Z!H;g%~%OOnJ9
zy`3%6q+kzLDiwU&$FG*L2Zr%`_c3$VJ~~TloOt4KTqIs~kyL8d<Q64LC~Zhhj-v#T
zi4ck8D_{N`d769^(t|+=i^E55!glOyU)$c=;^L+AMKSixVh%W#jbqz5j)i3jgf!%N
zie*{2&gl6W6~m6>f`K$mMqe*R4*lVPovkfg>EPN9Np^jtY(itm(Uhfy14yL_qKLKa
zHRh*x;V6O5a#Ws@r8%9gKD}Xx<JwG2jnSNFVHs$zth0RnBGu+N6T4@Td5Y^;7)K(#
zQJ&|5C>|0eLk3yOFdTApdB{sU5|bPDHw_1NmHzMc-aFW`>^krJ?S1mO@#WMxr|Ah8
z3@{*sM8G5g0;E8Yq7^Jzq9v6|vRtN1`ok`Z7G=w(?4rdWQesIa2vHO%f)R;GM3?|f
z?&<F79ACfuZoc>Ay|@23Zw7(rK{tDbg24Ax-9O*$(|4V9&f05z-?x}-*08JsAu}4a
z89c9wbZnA1qTO08eX_KoR;yEKI)se|PP@a{1Mgz$;WyD;UgYu<pTn~R!B~^jvbgRT
zs<KqHB^qop20BN?ir@Kz|45oerESAE1WzM{%L5O-p`?FoVPj*RbC=GbxP6|xu8rq8
z*sg`J1x6auJi)RpJl88ba{`ps_+IHIpQQ0O<awvpV|Do!zGdUReA0gnVm>fAxhSYx
zF1|5Xg=RGBQN3~w`|wGGtV6CbNs7r7X*;6m_puxcZ+;3nI{`)@t}m0HK7*cHz}-8C
zNMdkym1<ev!Epk>hHeYfZ9y_3?`(2$Gw1n%4+>6H3--?zR3@6pJV(S5b{~P}blDy3
zcj4N37!06L$cahtLd4``X^Q-oN2$E!&FD0vbNMV*QP8;mjSQvWjMJw<PM+tCw86;@
zT^%7Ji(mhZe@l`KQ2LuX>_TD53iq9S5Fx+j*FWg@ICb_}irb7|r4abOLl}5?o`dB`
zunch$;n)sA5R{fxl)GTVAS^p{qrsM~k)cqAjrJyMt1HyPDuL@UisRczIqnWwR(2sB
zOJG?BDKtr1(CPOWZ#1zjKx+!4F-jA~2}ztGEQ_EL;`$+$gm~1a+inr~6~bB_L%GIF
z8v&x+R3U&QOUd$#Bvm9y%I0Fi#<dJ#3}Xupjk%Dp;-f81K|w93;oBZkSfqKzFzS=%
zIofCf-^X<=T-V33B5JiGj87b*KUm}X&2v~bOgI6mkZkq@GnG>9o96~;Y0|<V@|<U0
ze2!e@f8tqP9-Sq-cI`siUn`Tk{Vq!@E3Yb&kXtKLE!!&Z7a_=1fs_*4wzmXHTcR@C
z20>{_we@f}8lots-|u1D0^5@0uX0%YX53X#muZEKBoqQyg6BF2S<sC})Z-p`JY@04
zGS@a@y8Vcq;~^(b?x#6kN9lsW$~v2?Ygkfp;J`j=Qxo7jsJuYsMcI|uQpdWsMLH5h
z(TK~JS2))aw1$#~tJu3ovvZ-scw>%QrNVI3!*T0qEl9GIYSkml5~5L$O4z{nYXqK$
zBMp{cK@ECTd<)z4F;10RS1vPF_1S&k1j2D442gy#WUeT*Mha8<g=tL`-<|}m5zH?v
zpj3fv+el$JfBq$wmR4WY+)zr%%bks)D2S7oAS@HUD2gq9Ly00UwxrO!^5R$8FpLr|
zT)K=UJPz(Zz=g|~UK8)K!sJ+3_`<~z64!ANDrXS&sSYaSQNrq)VKE&th+}4h8i#f*
zP_Kq4l{0FM=(M^>A=$lifl9LijzJZ=TnMf8R?EKhYD)#gQOu3?7FTpaC)d=BWT&l|
z3oRPc(^P_xXwb(I6^tmAS7EhEo~9&&0YR;SU#a4G9<~81@X*78z>-+5gAQC)Z!A&w
z0u~M(C_7HRB#Ba_lBg}>szIZH^mgMF14bC8XQoj~VOuuBz|||4SzBMf{bN^vlwTXw
z7D|z(34!l}HWY>0O8%|*76nS{R|>k9S4u@`%+(v$k&e&qU3<88<JzmL@a`1(R-RX+
z#1jhNbrIyWM*|vBmn<4^{c6gkjhIe<#GbK`2Oc`i*klu>3;MTKSzlVgl9Gc*4^p3*
zhR{W4IXWvaT2U0`h{<&<l0FQF11?-#;`G2}J@y!r5eF7B_U>phIktm(wZdS~#q}gw
z3F0iF?l(x&n5f^OS{)<s>x6+{=6P79=yvg(lKN3tE;r7dXS^P;|LB8A$Hi1aMuQ<%
zR!|s&)LVtL(L_=2w!e=Sg87|0%C5@ROa9`;^W0oqx~-B<2#{M{>~e!ZP8>&6Diw?=
zy~<q2eI>@_cAMfG^1M)n{wU(?x${UFaOBWoUb=9eG`sg=K2;eGHg(G(kU5ngfZQSN
zZenjNVmfU`vE=fgVALD3s~#}>#$(u1O~_)nwhGs7V5Eif&`H=igYd0VDVr5gC`wkk
zC@oP$0o@LrOUpc)S6ED{%$Nc9`76wiSFjG=k7zbweHr9Lc@i4+A)JBX0M=GuW(Q(o
z4k{He1vJN?)qjP{n&mmypMDn4^Jv`vMwD$~>LI;O4?8pDS|N==i}HEeA8dZ>AIFs1
ziRt;-S2`v_2yWcC#)V5)UQvYp$)JO6S%5+5f-FxN3<o7&x=>_UhUa;ht>#;rq$OoT
zh*zwua;<3(hg`XI5n%_s;h{J1)U!_$e~WS9-9l-z<%=q?9f9rG7@-+ODSi-_LV{L@
z?s`rVWi%#SW_C{FhXI9FWJ#aVppTS-shKG}zlz4j=&gy8(OY&*VB3-`hcrv+u69|!
zZWwJA_)Vzow5Uz^G%F1P-y=#Re5>qEr&&SZI}}AuoW+D*fbaOYu7i*UOZpfSV|%jf
z=4p?1dyOD;nVH;yzy{%xWhs^{O`c0pMClio7q_QN7z54b7+Ps8OM);guPoE)^j?)x
zCrbmvS9}PwEH9fgy4)7?@-(ZAfkG?&%1&=64{vFjv%0#5h0XNLG^?xYuR8hpdfZh}
zr;Wjv78N0|g`zNqx*OowF0Fpd<<$|-T<zm&c;6%YxbMIuj_Z+(hODiw;8_m4_Uys&
zJ#5!Ocn-NzBvFJ)GmI%;c77OOgrJZXCajXyV%j&?*)?giXQo1}8BncN34$s~o)gx_
zkW!#jR*tA_iQ{=_Eos!p3H;LTLP$++S>#d7m}i$ACJomwpJUy3dGH<Yr4SP7`E=WT
zoV>gNG7OFp#IIJF8Hn?gpZ?$f8DD<<F^(NS%I-Zo`J+#Mf^O#<c{$(whc6q=jz%K}
zg8|KElQ@czNRlnGe3oUe6nSOgYhHQ4<jN4I1^?fNKSDgpn3<cUR;$r!x9=vr-xA6w
z0uc}jABzN4WK_KX+wtfoBW^@1oF2CDaCyhB!yFo)z;<oYD5AB#f$R9p&ClU@HkM-}
z9LQBc7MFZV3<xRlJO?8T$^uid$Wl#rqsP4Mv7=I@;yZ+ukRYs(6&XRJUMjD$bnA@?
zw(DYqq}FKSSE_9JXpmc)B1x&+7FrvWhFe#z5jY;l-|;Ag)JV^!Ka8-A2O=gH1-7wC
z&8xpKlOo|4|J5(@%(G8(@W6iN=V$rS7rsbu_$rI!woc(nDdIRH8jYyeYg@byNt&j_
zQEAp~D2u#rj(K4WiBf#@_kNc+Ets61AgomAecf85KYh$K73oqH5?d7HCZ`dExVFbe
zZ^Zf4Ay3{I;u*t_ykS2N9h$-Me3D_Gm8Dzw(q+$qeRzS7^t|%mrV7&G2$hugv6K?O
z5@4jHupCUSPST9oSXpQ9n8ktVkj7YpTBA-F)`;_*N@D^kB}yfw$*fiKa<maNn-hdV
zeQR}qCbw;}QAEwQ%a>!3bNSq9!ocHA-~S`z(!%lsI-5P5qLjI&C>#XQtN9BWh_ZyA
z`l)}yV~>5A<0p=>Yxgex=o25O)w+G28&e(%i=tpS9MbRi86O`fiXx<u#Bt1EFd*IH
z*yMSR(%+Ql+7P7$zwk@{nrM_Uw=hq=R%fGin_}cHkQ+@ydW;Ddhln(b8Lu@EL4~{*
zai-hm6KgS{aQM)}N7=D^0uj2<ZecE70Kbknc>+-h%F(K4L7qT&2uZ1Q#Im4P*&53^
zFy3G=$XL46<~~2?SkohDOd-a{5#v*kMliRd<n6^h5EfWqDwUEaH@_1xJ_~42DcAzh
zA7Et}AV|}MQ%^ibE(FJa{I8+_yBe~#-YJEBa!5jsht1%dyK~<Z^Fcb~fBUI_!uy~5
z0LPCXWo}`P&wTQeEU&b0|7<tqOFPeVx}6TKR*Tu$S^E7x78b+dkWQyV97UzYQ<jy_
zcm49{U65(bAj$a`Kl6Vx7^Lhyu#buHNp3CQyPb_T22TonS0EiIl*S8O9LpviW-PB}
zEM1OpY{9WNOfj{yhHcqoc|y0}#j`EOCnm~0;L^s3lI~RG8O9V_G)z(2S{Z>>78sAL
z2pC0O>botPGagRh;(I>6>yzh-z^#BNx%1`WT3}hFZ)+u};Jbc#{TP%I7?mL_R%xB3
z4eOgLxUS8CJtt9GAcR8_#n@ZC&_aW439|gvyPbKV`29clLoQsp#KOV?<CEi@K7EQb
zdi8=@n{vXEq$x>~5QZUHmX%8v@{BZ1DYg!5N-4C_-!NV^8j4Kw<u5%>k>=E!O?=-k
zMRRw6yDRDtgq9?b<ysA;G-!<|G%w!h@z~9XaogsdN1N=Q@DS1kBt>D^v12!O=%Gwm
zsoJ*3Xb{nFuaOT2c(za2tP@mf6voi$bcx~^tu=e+>I^#r29=P-iCJnhJE=J~VcKTS
z@o9uU!uGHz$g+ei%MrrHw?n*26(c}Pi~eYcG65p=uqsVPogNe!qd20_ML8n19i(lO
z4o28Ul!=MK!n~TV{42pcefA93u3f|TU8E%#4oBqqs~h3uc|nrI42J`T!vTZAu#}A(
zh~wxLFPoysU*TB2d=tFlM4lUjw3(S1W31Wy*81!}1(X5bcJYOS(HTl;kRXiYe6P*3
z(Had|<4w(ZcDXf#ZG)5)8fNErV0ofkQf!qc!8l17Y<5ti6xR#zs{vl%pp2s59}=et
zS}S%`YDB|?QIN7SSx}j4P_+X(y%tkmgPQ9jTpOtZ@-!tY3a}(@SjDUO7;=my8H~Cp
zBN5FevKA2aV`QO;;|Nvc<%7evkhVn@6_f+VvYR52-*z>aF@}qmE^zDCBA)A%+f1T}
z;<m^4@{sp(cWyKqF&vd53?Zy?3qqQd+$?R-TLzW->pBE4zvy$V!Lpc~Y|^OL!Q6a}
zVlJpS7NG>A1WIXy(g-xq-01ShD=`z6%l99sb9mfGSX+6X8+Ps9k5lzgS|M#4%l3%-
zBYGPv<ozDL6Hpm%;@2uDV`y(~mIl#AabTfNzty8(uW)s4p6c8#nyy1FZnMJ*XjXir
z<Cl4!MkGm&kT$+kC8*Ug0*tii4*L|c0**($Ql}UtSlTcg4ajvtZWK5!(s77;Lu@QC
z3Wdgk+YC^@25k(_pE||m%a;iP4=F`yY@WaR6XNCWT)*Fc<@0i$M^QvxWLt)?<v8f&
zmpgoYo`aAMb90kSj86cUUW<@Z7^u50HR)iKq)-ZDNsJbJWx3C9wqxd<8Xuakva{kK
z`~WNmbb`45AqefV6X&`RR-re9tCwN34b?H29*0H<DueY67!DCd&W?$IH*WOkH$tA?
zF-c|rLB{-$u-js?+Q11zM5PLO69)Y)OBD~(oJEX}gGe#9i)pVRObzYNBf}7k!L}?m
z+M8I_TA9U`#PU2w?HE^-l!n}tg?#aCqOP>&u_qtrrL$+LR0Aw2Xm`7%K=3vwIaoQv
zL7!H;MXR;RW~)_-2Q>`({gP6l6?vATbn!|dFN7#_UPFY!ay<6!-@)A6EYH61>}wKp
zt-$pqjw>+AY*}l76o!qP5!YTysRWXJ4}^@(*vPHtlTn7ru}QFOv^E$ik<unhGe&VA
zrD8nW!EpmD*T*P`q7hk^qqLzh<&!5KQR>oZH>mg%s6dhoM>LxbL1-g|gC}(9Ntu_n
zT7m82IzGw(Mly^ejCOH_jV(jcIKrSv(v(7J3RU`=N@=086iaMHrF8H4w*GX+z^$bv
zHa0hK92=w{O;Yr$JlkQ*gEh}d(u5=_dF4__1TP=f^wxy!YlpUPDC&$s7a)a2y%AEW
zgnT=-?bqY(cE1Xniqv2Ps_?+1NNKa&NjbGvkZZ}Ix@5<Mg%^aRNsRUswXp_r%YMYM
zY%qo>8WHu|I7(A#HVEqD*uGDu3cB46wv>$38{{e{&r)_B+(Gu}5Ad$P{tu{(k7HEE
zaP=lP9{W5?FFeLnGD7A#+7e~E)Uiq@XV*r$E?2Hy=W=I*+56wjN~GE2R%wnm5u(8J
z>qL2qYuQwSD!%8^UR}ed<QA2u1#?F((uLM^hXc61CEYi`nDT)!8Vwl?du(oQP^r|m
zOwh~jU6!Y0Nlca|6gq$91^;#5LKsvK!r}+t_de$5X4$oSC;#?WfAx;8`PCqWMJUVL
zL>nKh5yG-r%VN&weR2eQts1kIjT`u6Ns6O2)mjB%m7_1)@&Hp>m~=-t+9Isg@ESF2
z&n8zH{oV*mS~RM4a+Oo08FRB^<ZoZ#(ZBWqf@THsoM@%Z(v#;|edZ?Pv5U;v8jY9h
zSUlH7S|A;p8%wvi*}lQ#0|(hKt1JW-^+td+3b$G%&SD(fC8$(z9fw}j#l!oWc`|Bi
z0Q>4tMlpy64F04z_O{Sk6OTrWe2;FsMHq%?V~_@tIHo8v@-!n)6V%IVgudxJd_O=+
zi+8>AolH+nF+V@cZ+-N)@BAWe154P{qy{5VCIm8s<*?XJc($d<jLq@TuzSM7uT)6l
z2-j28C&rL&ndg?}V2mLd4H<PeaSKgtYz%K~5<3jYi=0log>73*j5SG%oIFj~cXT&{
zcmEKN{*Av!rC!GvP1IUv?aQCz##cViY&68mGPD$EWw2eFAn=i%jpcb<xO9c{omJ)@
zc{?{#%@6t^&5BQ$*KxxpQIX==4%M(m;JIuL`UI#l&k-2Zw>`W86k5?4^yv)l>^Ch>
z<(lDOK)>H%ePgvWN!K8ZW;7g<WhrSAkt8vN`qp_~39CqH@uMI55q2yruy@}ce*Rzm
z;yvfNEj-4p(yXg64X^`n1vf{U&&C#saCy`b%sCOF(b(d%<uE-}TAJ8SS<o3%s#=#;
z%bDfOG%W0dY5-|cj#F(1rluiFkfRZ^#~ktx{QxI_{4WtyD_{(1x5wzK&*44(BtgH2
zNU~R~%y4Z)SYDqA0{W+3WaH*}>L2_d=dwPJNbm(%j)PmDV4yOLQrMnH82GHT7pWSH
z%oOEUn!Gg0|2D`}!D?p{(Y-U`dO6-J=yY1NHdncM^EzW=6KG|SM$zqdNs<xKa6mNd
zljR9&Yh|4HI^(oPZ47DI{KJ3v-*f2TL5`g`%Kz|_Klv^0kGq33;Dx3AkcmMWiNIpm
z$yvFoP&rH=l!Ox!*Y+s#9Ag>0AOIl|NE}CE*y78jBWz3Kc_E%t!Ll6kLY3WUDXD}3
zxl$-qP@f#9UYz9qqwg-qwx%FXTWoBequaSit(GFT#^M-*v?R9cfG7*djrC2I)&@*8
z4iNQVe!?THRKPO$c0iI9IJSlF`PjBiJnG@d(xU4XCF>phd|DggI3<qnd?laJnlw#_
z6PLkofbaPw{Yn_pEJdk;qA2ZLl-A$!UUFRrf#AUh?x#7{WMX21FMjc{JNozUs;CQv
z#+8yxL9T6*!XnBH*OoJSxyy_zSZF#Fgd~L_4183Yl`h`Chi$o}DyJw4EE^hQHLRkB
zbS&@#vMeV}QyTR;uIG{GIeC_$9hWJ$LO*Z_#%hSbM;M33o<kg{&$6=f2v>jam)L6?
ze8<ICKBK`98Msiy+~^GX`18w5pFGIf=g+XV+~v(b{?~CD0nb1FI19V?^TL@IupNhH
zvq7!dWa;X)lGrL7bWz;h4|SI?+K^@$(P%_}&}Fl=PEq8zwoRVr42MIaC@Oz9DFKbv
zyrRGn0vy-o;KBWbq0iN8S2=Lt0IutjRefbp98sHf5`uezySqCC39i8d!DVoFmjHvi
zyIT^16Wnd^fk9@l;1Jy1cHVF6+uH3P{iFN#?V9TDThDps$jTN4%q3Pi(_YpI<0lTC
zoGa_(Kl_2QRf}Ov4av_|Xri2-cZlIhfIe7|`Y}%p*$^4^S|kMkts`oItns#m9-J33
z<Z2Yt**oSKx5bUCkKstzvr8iu?P6r-Zw=!&wg#NCwFGS?E}H%FQT=ZJ-W89UJVenw
zqlCibTgUL-Uy)1>aj=!dW(amSd5gL0&3<Oq>Yvoisr=#Q;}k=asyZ)w&)83I+qqzC
zoxpp3;9rt@de=dh>B?EB;+S_JFQ$u7p_mrq6EtReE7k>**EcQ>v0@$G8~zk61cZxS
z_JTEAE4-OJcsu?gS`jM>=9b)J7@1HskumGP%&>VJyb`J382Eyp>50!>qDbF=YmQ}6
zQq8g?&Dp|aVgj9-eNNuBP*f@vjyUwi^Nq;2vVj5wer#xq{<<cCFn3g3x8=`C+(B)~
zMMrl`o{6!9B6@F7(#&9|07Q;5TwMbM@rNgNtG_Nl!UErdbOXn~cbTsY#?@nR1EJ9x
z1bLsE`$|``NH`brU<<I)=|Q~wfy%+7xXvUO>9$2?G+mK60qM%maRI73a)|+U9chT`
zgMXhS8hH1Pmn)u*2^3-~L`a719?7)wSX;gn=VB)zjtbGviYZdbETf<Ov^}Qwi^tm|
zo(sS)>ZyVi$@+k=q~*k}wI*&$@30|F7HbG90==;HC3SQvMd}z+J0uorI#fDT8WdvE
z@$=yv;Pn!#6wuV}!)xFvwS;fnRL_&}aFZNx-p$0i3Nr*TOuS3AP(rk2;>v(xcbHHd
zJXuZ<DSkzmje(%UVQ97YjD2yJDXyl0H%E#&Y+ccIw-PJVoQOlfGTw6W^mS#Z2i3b&
zhAlQU%eZo2cbFwMRu^vc-FvB!0)zv$9w_5@L&HXhl<g%Z_ibQgBxO*PUo||dR}7R(
z;R{qE;<_UsQ1(UubIu<t5ywodTZPhuz1XR)@bZxmT%(d2_~W(pxss>p;ORB<yCHsc
zaP2SaLk}d|$lP)>yeKJRLJOtp{o6pDtYj7!|H~h*KKvMw%7{)N-380~mulsw@yU1z
zEAz2+5rJj^Jc9rj&QRHt?Y7azA?cpsuG|vkj~jW@gTLX&X57;|2;+k~{T)mpFeBQ$
zf4iC>=l21}rn=4;MA*))ju*rCC3-XJrLJeFv89=Tk6BMACk~z2m>4tQIgp+`c7KtW
z3~U()2;0xjzBoPo%;oAWn~w;waihy_CK@EE*4!K^DK@lX75)=5)ON{A)}e-TH|vuS
zJBlu6izb^eO@c#L`c?#Ns|r@n&t_2LZJKFKFIw?Ob08;EOz-1GJ#X9d**^!XXJ=Q=
z9yeI8QIn)|ANXVQoYzpAool@o<Zq8iQHKv!zo!e|b+^uF-4x6caRFvei98>g>=M58
ztpRq|-#+c@VRdhO5xstLr*nnN_s#75Qdra95Te&k4<2(Qq_r3kU+I6wM+5eGI$nrM
z!VxZkt6?|H2t4HFkkV;Le0_aDI*?(MhL~AvXwDa>aG@zSE=J%P;*tDWO7#~#ZTYmV
zO3AxlesPKhTJ?0g3kE2~S>%Poo}x_<nXn%#*gp}aKJ&zIhgXQt(B=J3Ghk@yKd*K>
zWu|friT~o{wlbI0*)+BTDxlAEnkKzktheTz*H1ktP|R8Lrey62j%kykN|RAS2ephe
zZI1q7UHN@Oye0^?02K7`LhnW-?&%r4)wWc1VD7KDs`+BFIdnkD-aDVv+=6(3aO)E7
zKDG<6(NgLDL8c34*Yl}aHo1H!hpcJ^Rqw|;kY~MPF1%#7tW<sTVPKLJ8odRqHxG!t
zbaR$9nq<M01-#Gp+8-iGABsy!;@;t=^!gFWQE8&TS;mU;tOqaCtPuyA(DfpwE%#Ut
zZfAVu0-Xw$`pH-p$1e4wj?*eNltbazs4DB~zSiw}pL6pIc;@sP)^>|?ea=i0naFi7
z9h`pHbI*E2FKuDqYh0VeGynpxY)f^GPDFp4aQ+SHdeRHpV0<yKN~9=m-~!jTbjC9l
zys)^EX;ppA&d%L1=O5%J2+@?OR?XfU>kQ)7EkDgW6tVCP{bPynpTJ$?bwA@#$uXzG
zJxC}z;S@JnT-K--*)qfK-<b9aM*#5`$}hhVmfAjJK3>JzR-k7-|0UT<*nB9*ywO8k
zI0?kjriRwlza51@m#B_a^#+fZ5grM~*gyVa>e?6W$qO_?ppYXTtc&!0vMoskncBsO
z#{X2EGH;`pnnFD4xnG-pyf>-94O4Gun*JF{RDOS@4E6D1*hV<@qebRWgFK#{UY@>=
ztyw)3Z_QyH?{rfJa27uH?@jA#RYS9vxf7wkKSX_xY4}(BdrJ}h=XFcD{*!I;&JJ2L
zm5XAE#!nj$vWB2SHZW>Gq~9iufvNfx*r2I&<}iAxCL!CgY^}z_f`7hZufd3$Q9Saw
zJc4b3DOn4HFJyX-fi3Sdq<voY7$+_f3I{^~K3ft5*)}1Kw6OpcNg@H4XA%8$QXp_A
zqyIYu9}%4iExyS;<p@bIV`|+W5LXhFJ*p?gNrhoWR4cC1a{Yc1H5`vb8LjMC6E9-i
z`qS@P<RiPO83%3kb40lDGq#rf#JZ5@uGNkE&O!V|?{=H^A>3uph)v$KBZ4gvMnc!)
zbb6Ut4Psj^QJ396W~LW`Ge|NIzQGwb<KDf3CQ-15u7!n#5lEcAahg;WwEf4WsB|8>
zChZ7@2MyDQ_yuv^8q;PLV#_4UP^NtzO#t8O8{N+<hw#{QhVhX|+-^c1PgBm5xo`I$
zQjp*;y+a4x*PN8r#S?ka;A;k5fsQgF$YJb+#FY+KzE9Sq0c&2szqc%ZP<OFbLFM2$
z-t6C+w=AvmIQX^-A8J}2**foWyKe(2d*jS&L<2K6sy#DmI5$ofht6BNQy&v3WVO~G
z@U1E<fAF)?<szi~rs6M}JM;z$el}BjSIJXe9U>6ZZ<9h}%|%i_a{{4^YgWiwOS_Vy
zk^T7to_`{dc<e(a9srCnvGhr-6o#_aQuD^Vmjv6S5!h_-?A)o00#T%E*f)bRTa8P1
zk%eG%VL%$i>9z_5yaGzDZg;dwr(}HLej73nONW<K$utu}7u_ygSiJJDIwd)ky{GqT
zAac=r%~j-342x1s0p`HtX@L#x@yV%=&$5H{HsnJRsODEloOs;bV&r>n9kT{qwjhzw
z0w1Sp@&~dng|2o}$(Dj~jJ8{g!E<+X>C^P&mKL%~5mKb?gnPFgkso&C0iUy~vj{C@
zhnXm5PCSX^EGJ%NBpC;FbfwP*QxgFl9RG=3K$j}nzOnk@{SFKrY@>oN89Hptagtly
zdrG?qeML!SOJ;bUM88mnyOi3*>>8Cb;-(dq#G}rh(Ng8HHz!rqEr|PhgbHHo@@sd5
zQ?88%d39^t%o@L#V~k$gAKew4?wmb4m$bA`Ma%z%$8X1kKmP4)TYDw1D7DNn#vjMl
zpvBafNLMX1I(?$}`&b%`3Qw>(yF<rtt~tL-mtB$e_c;?=1BczNM)C+7t6Q2`WjuUK
z_o<dQO!KXoAAgvIQq*p@;WCvfn<X)i#7n1GU&%=?=(GDhZya9+jI$nX6+XRBICOow
z3)BuJdEUrR{^1;{RBoE2M1xR2!)BkUg2<#{R_NTg7D3HX@0MI8$lW*_^*}^r+J)G~
z!7x=sNqg|c=O5h!bZm194+pmHS0}r}HD(Swb-f=VY;A9lqQ)OB--BJ3Qkp}<5<jQS
z!Wv288H3oA8;0(o2V-bAFt#~qb#A{bmmv4}{NQuPr|`nNPupG8;GhR)xmcraw|gxZ
zh`?tc@tNt=cX_#<E!y(a(OO$<^ghx+x%dDZjk$8*tH`h@VQ8M_cOVxY1MMtFDO>Tp
zy*t1%`dW2Q05N}X$qxQ^RL>)RXHkFd+J(+RI``7*iI`C?zQ}f4$;h~3I}wp+DR*`e
zx0MEIss;CUJ@pt(Ki?px5gX}+Pb`8RTHKfmd8U#h$&t%=&;Zr%+ViuqN4+ErUZ)kC
z`fd?A-B4cH*wK3Fqr6pRzJCpM?u5m<mL+Ye^YdtC)$;sjb!D+t7+HA>aw>u5l%Pbh
zC@x|X{SH}*>$dq5sytUO4l+wnf-f3eTnG0?0-=gturZAD|6-~H)b0(e{<|pkwACw(
zmt>VTPJ0*AM@17o>ueiO1B7j^_bp}`_cuH2TJ4>YXC7hY@Ogion@^;uc-*8fTfnnu
z>JT<&`hg<-uRJa?hfPBWLLZIvyM}WHrdv6CXl@WXeWMUx)MlMCJV)1{M>}u8S`>gB
zi-z(=vR{sIqY-ShqpX{`NM?ms6Ao~z_>yVV3@(<I*X!~PCxzC!Zu@+M-ky6?T3yd!
zk{Ew3kTu1qmQ=FGY4mnWsE-!jZe_?@Qx2#aI<d-E*~tsq9A9}o3loCKp-6XW_BA-l
zX|%oVTX~y&V{G=l&vbBe(z3>=Lovtdn*nw&onrGGPpum=nj(=75k@4QY~jzJ#4j@v
z51^-QiI-9T_L{K}q%e2dvJ9fsnQ^?DI@EAwD*P|(Bj|N(i5{&}!%IB#cKI(<{U6?~
zVUEhE28tF57)$H0Lul9qRblUl7nC^5JQP?*eaz2k)T)h(Qf0SkK>VT6LE*k&Yw!z#
z!(T)5`vvx!`9q<auomx__;T4_LGhvqeMIegDvZ%EZqKygIld1lUMeXb3t6?Q$$N-|
z1;PU}0C^V&<hTkFojNwfv;>7%y!S2Fg?{YZG3Git(6&EKQW=H5LpL;EwB|B4uf1?I
zKLE=WJ-Nwv1jI+zUQS;GUZLQpOaJzY;b(bU33pUwVz*St0pVRj5zG6pjOJsUBbvw{
zv?zJU3>_zkD6D99-*gfF3>A$=@A`y)HO`KlBPro4EK9XwNzJ{9XE6UmetB-qalUBZ
z@{P0tuO_}q%c@$SgS{|JK4G@oGrrtD^QXYD>pRNZb<_-s9;I2#@vxEw2dobZ$hgQA
zghYk<pH)d%{IPM_|13)UrJHmwcz)XR*)pK$YjnWLk*kIW6JTtm(Sf;8E+oQ4Pj4d)
zdzwdfl%Wa`4FA|*BG5Vcj_)5$OD>`$LL)1y-TVE*7wac@lw&mf`9=4*P*xSTea}bd
z`B;yqD5Y1-?3m*=7)p{<_C>TKW=T{z=!I`%j_tQ2RauDA$KOOZM-i)+TzcVbEFk1=
z%t=IbVXHZF82BE-TX}!g#>+N@noWBi3lwWv(Bl%ge8ehx5K|EaE)n%NC?t|sc<%7<
zUX%H;Psa!;=^ZBe_?Tfg>-OcG;eyn9t1E@L`PEWTjqhNLK97$*wR!IqM(<!m&4<&(
za&Zu9ITVPlJHHR6>35xu@$5!IpOgDp8I926w3pI>jk+u24t9GJDqz`+1LVX*0V<en
z02>8AZ-UQ5D1%<bjC9rWY3<l)fO4qP5Kb4dkUip+oN!&}KW#H&$rMnNKY2eDRcr~Q
zyx)@vErfL8e1Jsm&Z12YErbB8o2`rRV|wZpFGv33I^<rUG)Ph>&8Sg{*W9V!%*m@j
z)j5B164PrU{98u+-o(sl^N;r&dKw~c=}PlvsH|^2QvJ$9aMp#j@Me;tqra=L5)g7W
zxI>*B49tPlD4lN3UiuEv9D_R4uMd4h3aN3b9n$uVLw^h%HCURDV!HFiqH$VlbQ4^@
zs#NRm{YQxUdfpOrxcssVUt^Ag!NEiKLH+9oG``8EnwkBIsQFpZN(rY@1qR<>cL|ZS
zY$Ca&cp8H~)iDGFvzMRL3*ym4<7QACU5qGy!t={kY4-Nr*~J5kCqulu5?jbLMQ;ZQ
zziH_vmyFr5maiVTZ#`hv82g}-#fm*9mNZ~!*-jJhWXai*M{%^)bHhE6kw{7D2HpzC
zh{<Eu*9Pa4Pofb*AIWkHi!&L@rQ1xt>s&q1rl_3rAlzSfA&&X_;nB;i$#FbaQvs>Y
z0@F144m=;iAUtMXRyPH;_Rw3h&4YEC(jnYmQOs$vZtp`>ZFUhkVhmy5`up7x9QLQ3
zy_q-j<jND0Q=%7{)4mq1LL?1H>gTIbsZ@#Zq>UY%FUlwF**dRL<hP9Sf_L%4PJDr+
zb;%W15xC0A<7~#=?;&+{!4fQ-T>P0q$szSz@Pd^s-cLr3`v5}`E^D%6JIV*_u@5-F
zwjXR#pFegGRp82@suUMZV{B$=<C`jp&mLj|RG9tMPJ=uc;1FIkO;fw-E0V^bix9lA
zc2Q7!c!u4b$eX`b9K6q<M`6~Gw4$jko5{>L>`kSNFM#8)<Le(7AHZkcU@Wf~n{Cvg
z(T!Q0IAd9$3`RZ<C7Izw>vl6smFS!{fAtTJY7t}deLycXqyH3{^pUNFXQ^7e#c5@I
zY*B0d+mToRr3JH2VB8H?r2gVOfxfFHpjTb-p~{}_SzI{g`IdZVdWqhP|5x7fKUX&p
z*(%>~nPma`*Xe!C=tgs8G{zT|ivYeD;`o#9Toxb|Mko)M6ca8^?{dlvzYq8Ct^ow^
zfuf)iQhRNOWqR28{3BA5jWARBcGa&}ik67-`mJV|fDV^uxwb}6a}oz+R}Vp<cDM&~
z$#&0IgtK9;X95;-*L}ajL=!!;8-&x{z_nYb&y)A-ch+MtO_yTL6;DKI+z&z1>#NwZ
zjE#`nPC{F~Z?jrWT9a|jpNUj58W#>`_zC<h!vcifp{oXf2l%E;*4JS|m4X6`{_8la
z7K3*SuUdqNht(Wme4ORaoiAXoYW}X5B;!}Cd3O_JX8L)3II|6|bTKcTFNPkwZVpDn
ze^qwdM>a<8QBwqZ%q#o~+bw}UQr6zLsc2>NaA32XA~W-*T$2T#((L{9gHT%rlOq5q
zi53R<B-AgL%pHR_o{3}_Pz*Y{Ld61C7=k$(D&$cd<nrSz@Mf1*EGz-2@fKEY6pm<g
z)UIXS8R=59s|pGL=^Tua%QQpV;YuspDF+^b-TLmFy1HpoqRdn(M?NH3lFIGu2i}ph
zT57P{#gV;s)+hQQ6lxdpN<KZf_)Ca7ddsbGh<v>!roFxkS?H>!<<jubdpg(XJ(fP9
zHVgB-i0Lv+RaIXh=H0txT73g+hXAs2>a1UYci13yUwwwZG!e@rMi{N_kiD%+zdX`!
zWc~vNs830+SD@EuEnRRaT5Lr+_g8e`>lNl33K)no@IdPb)|gBAP*ei7pZ)u-vgWf*
z^n$Q&vOB;%%9*MwDBR>_vCz7{OK$QSR?4Oi&1`!;qdq7wWnzPip22+|K}5rW@1Tqy
zaA*{8$|bk$Ek0i8OH(PGR!J5D7!zE#9%y`xI~fWrugO0i1$szWcT{qbhYd@i$9}HV
zeq)B}ke;kZ65e~sGFXj!ZG{luSEWa!>wI3>i}zvKi&;BVQIjB2OuR8D)%~6US&gj!
zR!aPF5F`|QDB6~+eU^<v0uNYECjYc|f05ZearDG?I+J5oE8j9-Y{v#{n(=MZ_3?!b
z@Nz81bM%~VypAXx5`1N}=)2gjSz90ZD?JBPwji!CY?~-R=!ljJ4B(`b{J5SRoq~L6
zeTUJ?ib5SFM;lQ}jf{0<ZEKk>#T*g#DGkt##(+h!3&RfJG+Dj-a^)8{C`p?63tDcB
zxB(NARnlz5Lj*R1U3|l%N14$aQLKlHHn$v(*#uk@JPe!E|NRD~&x=!vt!~YYY`zHg
z#OlM=X$v3t3%WcaO6rIjd3Npa3!V>nkHlLO*Iv8U0JkrJPcRb=k5B9ZI=Z$5NM>wJ
znsEFyKS|w<OGn6g_WIQLntDj8%qLp-JZ!bKM1xH!p*ow40i4`#tQ3A&f5wm#lM<wQ
zCq>5{0NV2no~~Rzb$vx(Z1cMo(bJjUU+JK<bO{i|eyyh5ZEOk3Hm?s0kEAL4>BPf<
zqyXw4f0UWR2(Ndk+Hwg*CGbk?fvTY3qd%b)QGZ<jDmj!jncF0Y9Pl7jSov{sIruEn
z)!H!VW$md)^rl}|(JhwvlbJIh56)$ApQ|8wI;jl=27(gKdj0<z{4ztPB?E+!OYwZ?
z+g<WYjA>>rkbno#nFPg*#}a0v*{wJ)I}}KON9B7W$qPVV&EV)p>!|1Tt=}`jb)Bk-
zhb>AXF0nq!iJtgVyX3UfpGDZp`mu8`DQAJWa?VbzVnO5vp8Kan*JsFVmd$r{Q=YnE
zKAty8mdO-~>hB~xMj%19S=Da^KNEM5VWTH8m02d+O4k5H4!QRVj~e`>>+1m63CCx1
zi<zx$P0>ufw?#VcdY!JR=RcX@2@#X9yZ5`r5ORRf5day9w9Rs&)~0$$D1#adz3<->
zpOk|)G+`<#3UUJ+pS-6w5pgP0c6KZ)tO=u1-h(}-6>SsmNbVYYP81+Op@i(h!peEO
zV)QR$tHkLCCjkd#R4a{eFb>)e+9wWe<+Jf5FRar~qC@OPpyhM@%^ZWbudvi+<<%b1
zG`8xP6H3FLB$rf)0NKe?GwB=!2)b7!cEux87_L7J?cJSGK9P{RzW=5zL$0P;6#E8}
zB=J(J#w{$|=f8F<U4;1T6+4tUb=GXBUO+T0qoE`~axRU{ktDFLvQpX6J&Y1NKPlk+
zOaF-cmbfjn6d5bISg|B6sA>{ku2n%ZQOe6|ugG;5Uxn4HGg!5zOj<{x0rEe!-@&VI
z-~E+r{}(4(jF*7k!E=T%YdN3jSaxzU-Sxw?CiT{S8iE`dX6+2>^0G=m1q)_bktqAm
z&?Y9Op+zQIcDh9N{EeSJVe2ClSl|NFoh5wn`?f@gYI}Jf@GG^~u+8}Q(Upx+PSL4v
zjiaCbUOL4kVSEbI;rP2A|AIHnR3<>DCznyy>3@PljKG0~d?na1$YWkw*|xUSr&_B@
z$B%qNL8nWsS3~51%twkTupn~RZ)uYP4crdmL=nRVHMII&lOSUsF<yxY=sVUJw*`RL
zs_C_FIJqAGaW54H=5T2WK|t56z&~k1qvwu&)7x>Ip)_hq2}4%J*@gC|i*W9!KHj!s
z0$(W2KCU<V%X<G;@g8li$LS7Vi(8JVyucWfhe;E2c8WIw_!^*OUJI7qGJNsZ`b2Iq
z{?SK?yUc01=~dYJ!_PfmHTFT2c)pk!CTeob`}0TdB)*M%v&@?@G-<XB9fPOv-p%!I
zz0>%4qHcM3vZ6}+P7XZ$+<h16vcIGtf-z95nWe)pefC~&$EiHwafI8#GJQ-vef^!G
z*~*jW@$zC{H%9+8@035U)8!&utzMX5JM~F_nd}mp5A9O8)=1uS$jV&m#OcxJnrA=o
zwrLe+-N1WRG8w`F=X34>6;*mZx5z%=^DWWmLF>EQ*G<|LB8QTC)E&ODB}EXQ`XZ59
zz|kz*1o*s5;)xoVmPKc^KU(RdT&<IXK`1=L#Bg|fk9>P9y@*@r=ve1<Pa-zHGgeJ;
z>TP%Q15p9wvfho{EW!|R17SgTcQ+Jw50CIz)F!^|&fT~k@F`XL_eu0-JGJxD(HZbt
zV+)dgD0)4odwSo2`Js|W(a}m!!vhyITA%byipJ<mGOI<~WFpNho(Ske=*+ab<HOD_
zI$+FbNS_)PPegJ%rm@msMJN~I;nnC5CSt9|r|{qKMJmN8MhtQJu}yLFndTrTbE;e6
zkIqJ4DmL|ZA&Q2@_MvWCCuY;+9MN$Nmc2pRk1`|zgda<Os1=|^tm69LDi2qXUuf3>
zf-*X%e{@nsN5VPTNUK;itO}X~<Q&#_;CE2>q~@WqF&1?6%hm#Ai4(X3R)BnPu&%68
zaGj(#%<Ac-lxze|fFhW{%(R-;^#IA_m!=reZW(Qe^)d!eVdYef8=Xc|X{e<0p!UDB
z_2@vDY@afe@P3umsHS|Ya}V@_B=Hn-d#M(DKneHb=}{=*U)TD$rJ{4A<<6MxFtJJ8
zw<0r=2w(n9U+V)6KLv*@q}z|o%Mv7jOEhe>Olf__&utGd6ECaa7D?oG_}I(XwjgOB
zpp@Pxz36JjRm|^LeBpov@HIC%&F8~#S&>m%f@Ljb`HdQ!F9PfECt!Xr=*-dB+%EM;
z$MUz{UNqy$1ax<vi>!(GYMBq4Z!ul^@m_$aucE<a#N|)FYk&hk%|QPj2f3PoSke|)
zrFlDx8D4qZCu)$_$uI|ArmoZ3{tF_90dZpI>uTqo(dnk{%MSSbC>x~c8PPyi=71A-
z+>bi^lb>OKIW#<+B>0}QJ5}t_+vV_pCf~quTWR7*8==}QZmd3hKsvQin>@GbIcw&R
z<*%ASG_rOLY_mlhc#{@O3hf=EDOX&b06v@-pFvzx{}Vm2U?Gj9;<PL(Y_I(&D@O{g
z?_vq5`WmyJtIW9iY^4#ra(V0BVej#9sM5VbB(FKNL|7n6)TgvIR899@b4wp9;bSTm
zV!_t5E_Gs0(?npjIqlT`ZN$WTj>J4sV7_%;w_pCNOV<EB|IqF8b<a>_&{5b^?d$C8
z2xH+Bt$8)q-P|Bu>ioc5CvgR6w;pE#tbJemdh#gbvu38lVQlgIME^S%3iI`#_>;bu
zkleZT4`Z?dl$}oSd`%Fa6G^TEBf7@zm()Y5C7>xQJMi@o?D!U|tfK80kuiuoTwe6m
zMk(jkTc-dy1l$*hu0Y-D<(ZsbE^Bu!vbBXN{t7DW+B;QSb`;R5tv|Z~h~_aly)5V^
z$+?J;(pp5vkZi!4eF5PLK66Hil9qYa9GvbA>TKo0T_;JxYkR$<s-_TJQfsx#an>r)
z)2Xw-;8*91;+}_~poh~}@Y4gMIGmfM!BfUUK;+z=Y*!bXo&AgfwAfWog@^7&X0P_V
zX)E#H{jfP9%WzBiohC%IKZV12O+u0g&8hlo|L+5u))!Li&N0nOM4hOj&vPtP^fi<v
zslfGqn#<xZU;*tobAnvU^y22Gq%}oQtKFaeFc7vW`sjWmVXNOcoOt?iVp8Q|=JhJ~
z_UTze+#D|+g%n-oR1;%i9AU%8XrqnFFw9;L$Hs!X&Jq3Nr<K+ffL$a52@6E1RKMjL
zj5nMfpLN`EN|b!2+t&5s?)Gdw?Dn#Q@{x(ts6)<}h|N(0mt*!si2qX=t`6=`syY_Y
zs~h>OoI-1NK)6k72o5_505ffYskDy0Z;IA4uh+`aQYJ6K<sIqrMHCY1hbqHBs}6%J
zyiC5BdQ0qD8b#h{tq?BY%#SO9XVo#+H_P?Jxpq{*qU8g~q4&91VbKSdb4WgM{F;Ww
zEsv|W?>Ul)GMtN_9r*8UiB9gnCw`Hilcutb?hK_Yyeg)%lq3QXxF%f!n?F5=v`C@C
z0vdb&YHYp0z?F8-_{LaA@y0xNt$87jCpzcHf|hgi*In!T30@Pb1rqech&)po*q7Di
zzB?EmARY~AD6{+Os|;+1{nQ8O;}}*cejX_*@|9K%z_Uy}N%Mi)#U5Wkj<_W)k>7X#
zHwGB`jaj^F(R$4-Gu$jxBF054SCph~mZi5@FN=tg)B^=wmX1pBgV71g-K_jE|8RHL
zqmbL(8<W5bZkA%%o}LiORwg5t<6#;X?`|c|t(~65Z~Btfi1Wilvmk3lKkm_apm8!!
z69&4>d%cAEcS!xEAF+MSnnuy@87fN*$bL+@nz|+27mY1#VshLp__7&nR-|(^zT#ku
z-*NTalTz+hRrbhOx%Em$F&H}{I|U}lx##H<g=&i4A6@btMMYAIOF|NS&M(z2O+5=}
zMn@b*Qs4gM5(<(2r2NfO>vv3SHJWC6(vLV$L?vv-nI~7^Y>Q~QVWc$ezT}<Gm_uO7
zZwVT80n`+-5V^Ht`r|@OuOnDrh}fggOLiq{x#_gih6vU#rf>Rsh$rm3%Fvo3NS2j*
z4!Ls>!i-h!#$r{|$IQPK;O^)&7G?&Ie>a$iVYnvU!kgpllYr`~JThk&Hz=EbCF@FS
zP%@#D<3(hZFjO{`7F&Fz10E54w^;a0EPgXs8Pz1UIg_Rp7l~|h=bxRGWo7;w2`}KG
z(L^f3@$dPwGrgYi&UnxBawnicGI_8pK{_KLiq0pV#a?3!GvH`nq=qR<R<evgc-q<3
z<>$bOnG6@nMg_s*53sQY6<O@6meycCX1^itha_V3_?LvS%P?eYx3%-qfOGW1@(=n~
z?T2^8YQ<<ACT&<N1{2Qt;$bl~6}z`CXMxU^b)AZ*^;N&i1aeO>7;Sh7Jh^HtwD{`%
z1G_uhMnRjQe%7h-mAk_A&VpNC3SZpSE+EyOFGiX!!?rFL)y})fLEB<a6B2Na`F!Ft
zi*){5YRvjntK*<w^Y9P;#N5maLvpzG`QbFwGynGW&OeznQ2eFwHAbG1ff(x%1swiY
z?x&%Hjx9ZR<}d0K5Awvc;*oz<dBGj75rI`&M--U#4`Mi2vOVCLre4?ek$%;Zas$3U
z#mHr?OnA&KYdGiD@sSjM@ZQb?L(pSOP=5Wix%)B0yncA+h>+*6y#5vCiH_ATP0T1=
zXp_;a7W+|6Jt}<aLg9nCysbeAvElfn5n*u#gysaCgoZsSTs=5EEsd&y5I|$s)Pd&U
z4k5Y-YfAWNKAE^rl3MVFdtP@yx>QJbS<C=%)^x>I9lh!fpD=v);-1w4f3ll4Qhn%q
zRc4_z_$_6*TZ~i+9x)5Dt^Z(HXwb4G3ME`UX_&ebK?qK;Q33G{^Rh7xHM60mL-N$O
z`^|srQM1ayYWh?qcPuq-^Z9%W;k}k1r-*S~4s+j7NzvX0$Fzke0fS!h!IhP_1cZbp
z;;l}>FX&Hj6U@JgSuKDUqpQV)`@$Zu*Tm~nU^C!l@AUH6oFnf3ro4M02ab*G{^+-L
zTadoW_7V-4a+wANQMA!lc569+o_bW^J22&cuEAdBe!YzR6n{^S@3{;c7HF=IcXR-d
zF5gWJucqJqpCbu_UK%a`efB>C&yBtxmpGCB&ycN-H_NiC|1$w>*!yjBZ2SM4?7!zd
tedj><zwfbz2;(^T|M!Cr=|(?$MX1e%_Sni@!rs6)MOjsu8Y$Dz{{j|&=Z^pY

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_013.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_013.png
new file mode 100644
index 0000000000000000000000000000000000000000..44de32b9a595af5344a3fbf24e2c681727855cc1
GIT binary patch
literal 109936
zcmY&<by!nx__uTj5+WtiNJ{6JARt|WNOvOw!l==b>L|$(8%T<@h%i#R#wH+LQX3&T
zq~qQ9{p0t0-#>QFu3hJx-OoAqdG7lY&xaQVn$(o6ly~mjq1M)VZhYqsVF3R7J91L|
zw#-w{55H6JS2y>6>Er4j<lyIWN8iE!?HeEeHy)0hfi8X^4<BzSQ8`g55l(l1|F<B3
zn3&i993blB=O%`FrnZT{2*q113(%cAG!FlL2)-znd)y(ob4UBRib-((-|MY(7wwSp
zgr|EEKU$Y(cuw1YRF!6L*Vv9F5s2Rz<ceoKBI3#!)00de((6z8U7+G8a0t<J<@xCP
zt^Si%j(n_ZycBa3v%ehcrbbY}u9f$voKGwxi%kuPR;4qA)1|@0s;^bZ?RtXoks@xX
zRgHcG_OPH}+AfmGp}G<H5Z&K!_~w6hw+1xlK9_{>B)7HMVA=oK@pGUPTmJu_ItkWt
zj86ae%>S+<XAhLx{-2Zn@9{UJiEbK;|L<R*`<J5*kzYdp_v-iq&mM}ySpxq5SR#qy
zPr<g%i4H+Q*&4;~Xjex%%G}ljsXd<*kG)UlH{{+}o83o4Z(9}M61Vem|GQBV!=YV%
z@9ypSVURiHfxhVe+MtDw@Vtp~L**$n(CKFXB)I>|XZe3OXCjz_kuRRotaq;DZYLv0
zBXKmh5^~-)L=>-l(gTH8mCsUau`x8t>+{POjBaB2YT9v)gpfO2w0&ez8GeUc#Pl(}
z_5m7j&#Fdphfl_T&Ivl0iZT)kufXpz=slnGn`fj2ccd6o3X)i4rMt0A&6Q5G+=+c5
zu*DNjvn8C7LC*DpF|pCTddw0XX-fW1;(Qz()5lHQtii^}#iyh5j5F4Mn&*LzP)Wu*
z*dUkJ9o8B-?=H^hB;!F3+L=QKUozc31t{3d-e)_ens}46d>eC{ao<ghGxpNP6g#sD
zNE~BMPu~#q%}*`KZEFtnqMCEBl0dFPZ!c~yQZQHSZW+qB$Q!X6;hXu}`P=MMr7)V<
z!qw{fdLwK?CyxG-jI8ZwSUA5k_?{r}j{|07z!#Z>`lSg~%t~?ePVBNw7H3jT#VUZ|
z4x>hxsb~huQg&C->iKi`YRL55s^!S#B_~KAua=w(0!s_^ftMeYbicIqead2yn}!`h
zQqM7`IS-@wBl|9WZVgK&F7@29df<H7{O4Uh*F-4m=4gSXm?x1Z<hVq6ZLRzIrpSu&
z(W~1dyBoEe*2rT+Ft)v^o?PgL)hds=<1`OFv40BQ+J1$CD?`fLoC0o2kh=F$Tn)N_
z%2@~jXeD}x?dKfYQIyu0J2k}~o?c2}xwG1$gURKE`{8fq*(OYx8@?b6jX5+pZ0EQk
za1&Khj-*L(g!58p4=uQVj~GVYLLvum*Ep_I+Hg0DFT|G@-W*&nhrP3{9hs~Y1MPKN
z{9YEq&<m0PWkG86hRp%HZeeFAj!(!Os{wq1siCZc_m-7#w?`aD%h$`dg8OZ)CAezR
z<;xY^F<H26$ky^xm@XG>N*CLk!++&pa6z~4gWG-~{uvy?{)<OUDIsI{TC`dMAJgk^
z3$u?9Qz@fYAeMj)Y6V}%7#!BH9&`$rbN?zck^`a^wVhr4+s9~8<SvtoPeKX%C(u=_
zF6T)4&!Ilfs{VCB{OC^=iczOLVPg3Y&F|<40@0y^*8;IWXV?Uvz{xm`#`AheQpb4?
zc&@~s1;=g1J!6(LIsyi9TV052I3x<z%yX}goxU_k9gjLNd5;ChzRQDJVmCzwpyT$^
z>tGEy7FibUu<sNW@czXgVKte$V~s~jH=ACfN~eTyZ0iSC>N0WpY8jG){O@U!N%zOQ
z-w!rE9%(ptw#A&v7hmNxHa;+bFJaama^O?A`Ug5r*AsZ!?ICJwG4NWw-ISk7Es>6S
z$NAfv!R$y=ohQ&dw&Hg5sI_3Nx&6ov2vSq8pz{t(K};z4G*Ke%MLq#L@q=iA=NDhk
zg807KcZgh(;KDvAkZ;{Q;^M2bPkuoD`5+?&dl}p3n1NL&vp?X(9g4e2wtS#Z2p*;v
z(B-T9dwSX&fK$VV&uvt?jkIv8>z9z5YA*+(D@p`?)6$SRSe*JF4{;yahi?jI){6}Z
z$AK;ZGOOd6`tZm@5sx4D7+D)xRhS)i*QDe$m>@??4x-|w#Th?w&Ry0HiRwFTzsO@h
z8=6-XhS3<ksVn(xpT^O%;7gb~Y?BvC<5z<q%em|@{Bx-#2=j%x&G;Eu#7wb(v!uA4
zWIN&tL9ypu_9^pCN-L|GOclSkNkC0e4?Q91Wbg%_luy&ce0ju><}~3qpo@#l`uGMl
zJ|rrO4pNO5wUgvM<Tybjhvj^yG-fF+LdKlDCrbQdswQ;4Z$VxIyV0Jvb43BQ@*LE~
zaJv9H;?@rcFqeN9uAq=?J9Ed55}X|m&b1hIa`KoX+<G(Y$?Et<wJO%rK1V823u_$p
z)D{b6yQBqu*iSI*wbqKW8AnKD2^>+vE(=~_b&)BX)<`>?JThr=6UdL2;gKv`vm(EY
zLM4PQApTKgN9Y4?tu<^EmO?-Y!svoErp}P`v62x_yY}wW$hVo*qxz2<)Rsa|Ntw`g
zpXt@tRqd<pc+zzjGMEcn_~$<1kG~JAXLTcFT)sa{AQolG=%c06=nQ@+VnB-+9ZY%>
zW&V9=2p79jIP&K}smCFEcUNDw$?(rPbz7abi5FLi>L0GsuFVCO;(q%&)G06Hef_8y
z$0se~&*L)eiA^nu$MsdWYDUc=$Hd)28MSqDQG~+4$LWs%Fd91dI`$bh8uHkcwUi}{
zV$8Cu(<Q?cYis}{NiZ<Fkd6^(M{D_!WYDono&1$GHXYm5cDpC)xg%^+DbqiZ2x9ti
zzCB&w6ds<*%TW(nQvzohS5Aqm7T&EbPxU_R|EM}s^oNq6zoQjH0upTJ!>~EjWJvWh
zBB`fb*JU-WcAysg(Y!g5>?iLBh{pZ<w=MGmLm#ZX`ed8=x?TT|COLx$%SPFs4Jd@<
z)X(e#ANu8Y2`#QVQ|?mryYJi{ke~WQsfW;e@C}<Wd(rCH$2kIa$98PU(%(I>r)KB$
zLXAG;klYOTBEuS`&iqa$S)qSaqk_0YOai2DF2Y#7xC|%7(JZxUR-5u9eqEo)NUCPc
z7A<geDgE$6Dae~6)s3am@{yBL4EQt4M>5r+g`rackeo;<*>`E$WXWmw%CV^@euzN!
zIeEA^%fF!s!KwA{mekuxk4@vr9~f2E(#E}tHF@zjRTtx3Z~b=(l4E)0HU{Wnwm~X4
z{b1&*Bf+J`9PyfeI5Mf<qBqyU$o(v_->}AZmn^iL(_Kd-gfx(JOUU&Dmb&frhpgpC
z?IFt)bIsse6aSl~r>vu-MkeZL;~$1qM+<k|ENr|!JuFaEEJz3Y-tQE^tp!A6Af{NR
z%6boD55E{P8j1r-7rl*re$Yy)20!@vPuAguLKe6JrvN$>z}$sJvSW&zm9{4K7X|!p
zIB_&A$DL~JlSRJ~663I55h<DuYYazQWXkJC9cI%Z{oxA@$XzDGU*2*OQ`kM2nt#?x
zBq``T3Z)yUnN!LC8@%nWlqb$Y^J|OB6O>y``Y(h?ST&ZiLRLv1>l*QnW+%U!{E84@
zMlUQq;gQni3z!yc@_yQZ>SJ=gG1pkUJxGbv$PdtV<pP*p9bx4Vk>w-GXFdDh0*c}o
zxDK6%bsFMm3B5!R5?UL9s#BCzf>crjj@stNQFNvJhAWo@9s4|`4TU5S8x=uSH_j{?
z*`zv3LraZf(bN4!XW2*Mn~yk$5WzRn<ps;nb-KPs!eU}X6Rs(CtjZDohn3Q~R?K(j
z$lrjACA50Ox+xrBZier>J<||*0uRXJXz95YoPSKA_}`@x8X~N`68m1C+QwhEIv?Ia
zpY{3p#pjBVz*G|Q<dht6$1Eie%)Kxg$s*Rhap&(in5*gp&}v_)w6wh_SBK(m(YIT9
z$TTI7zr~<t-(6Z*#Ir_Qhk)wgH@WG0u4lV4iS*>2WBp)h4p;ANDkrOhvbSBrK(ts}
z9gSqds;>lgcduS5uo}g3d~b_wi7zKf%hf4E`i=K@i}@-#V!@#^*~Pu68~L)arS<g>
zSog*sb;Ch=iG7q!UnqwaFpOs|7y!E7yktKnv(Hm!?KcXMt=X?io+yjjY|0%wVBe}r
z_(qv@Y2nay|JOeITkMIx0lYOTZTyM%NFECpBQow1{(>^6B_V%fOZ+xH@?5l`*$^J$
zP91ux(RPjTnrbq2|3OIFF4!;>`!DlVB0nAGCA7hQZ%GqTd4E7MsMaEkZD4^FV({{!
zJ^W(v>9}(=$_eP<(e(N^E8S;hK3MYBsU}rFMPonf&W@W$W7DUI_cHh<W#zZojS_)o
zj%PWjbUpukH8<HI`E7i2U^&^^7(Qmmw<3E{Mr0`2-z7xf7LMtYv5G@YcK;`aQcGgE
zK(um(>L8UT36KwVd)OK5)&)#hR3wV1@UClFDr%&cWJA8^pU{*XphUV!D%;}QKc+Qi
zj75K8(6idm6lr*Bxm1R9Rn6~;#;>tmiEkv!ra2^-2Y%=7yz#a!oLo{3?xMqt0B-#;
zOHNWvvTneZTX)Q&+8#cV3NmQ6ZJi4&v@04R0-Z1Ky)wL=yi5k~`FC{6tu8k0*MS`~
zHG~VNVPOU^5EbsIDJ&l4az5vmEFvYh?xETl`mD1_Gs{369L@iv+BNg_6YitOB@=Yl
zH_#Q{&z5#B;kzQECyYYm^%}r<6zHc?38=oeEey_4i@W4aakkyOT2L>!xB_7BZeG<V
z0-Q*Okc2E*f(^5=i;7B$a;*z(5kJv4M>bP=^Z8H1>G___F=vQw>A3Cce9T)9RG~bT
ziGkR?ZI$U1-(>WjJ3^y~vg@V8#itevoq?;xG!a9KiX2YIi2`-`Gjq-$uXIcZ%(9wt
zivI<{#7b<(mfz`|;zIDP%?|_HKUbM%?u`cWXqZS;9(_wrXiApNfUyn`4QlcmdK7;A
z1?(&B5ZC{bU^7wWmNAX0g11`^`rKiVPU{UEH`JICeBG-2Gn+u*X{Xzr|GL9yI)aWD
z<T*}8>j)6Fa49?4e!UZ5G)M2-s+OdKIP+VnmkZ5-KSy^CZd&EcJ-tDDNWm9{Rb11_
zkY|oCL%5#=gM19TAnlj)B7xF`P6qy<3<Rzl;T9u+DViUp4t?Zp6Ioz@y$8tqqpjaM
zYY*~Dn*7Jw7;^tb?*vQ7-*m&v1%);=%#-6=i*N4za<K_Zg?Aj!QJ^bMP#5=OwI~rs
zxy5#~IJ2{0%*H<}g5irzX(#Cgl5~W5D(xU~7Q+Wl!~22W%YQu2PY;WQ4#aUsXbs|E
zRqhPZ^_MP6Y`l7pDJ&l*>Td;JZa()rS!LIuP`Wvg%@gKVh&+*of79~()#8jRL<v$k
z?F4=q?wi*UG#c)~F(R{_{PSyMy0>|dNkdzY_I%y2P8u=c{z~ep#<kDCPnlX+Rlo3&
z$21HvIMA<-5iDLS{$09)p3KSaSYh3(^Eng0Y8k#ueLwDi<nAwXt|{?*D1|1f6nIKC
z0$MZ2jBA!rtgVPLCmnHz@oD=fKak>$7D+~gV;Sv6{k~?-5h0&gGJ)=jcMou@2#U~O
zGvuVi3#MyZvbmKVhlKkJA>Mk>R1kB7zl9{7Xrmq=Qzj6(J7GBrno#90xMC!j7Z>4H
zfbSC>XO?-;`~s5OlR}j{l~uK`Dr79}`40W#zvHL3)TJ)bRyn1z_Q2@T`&bWr(Bzd`
zhU_lh1)?~h{#_VWK&&^Nu;OXedl)JmI^X1T*a$4Ae>!-j0#n4O3c`P&$gZJ7yc*az
zlwur8i~(@FDQi`Uzp)r;_^AgVXdz{L0X7^S0n5JKiw2+Cr?e_^NU`oZ9SL@qnH0az
zbb(p*y#c60p|b~3drOb@7B6C_wi!M1u<o^Q?wh{GH;zZ<7!e7kg0HyHf@&PFczZfU
z+^!d-{0OF;WyD|7ixbkKyOV}Yr0_VL67!K@Ma^^;9Nx4R9Q4cC^_fxfJ<%agr0cua
z<EZ}0Jt_b@%p5BNBGMN8kf8*}ih&2$yQ3u3{}f){qh|`1hf|$1-5zs*B2S5GtN}e+
z>GLHPI-S!bPP7isXia~Chn;1g$7Q{K;wYQ<sU%z}>=um(SbJhn5SfEBvvqaaVWWLF
zh^c>ol5kfc&nQn7vS#*lt)KKCs3CpT?7R59Q9U|V-AQ{wFy@9m^8yUqYG`a3E?ma;
zG(~Ixqkmpsww7q?lp6JH_8B;U1xkOZ+EYrIIU^dbGKB((4dr8>Rs7;pB_G_)0Qwv*
z%?IV^zr1&<8Fn;wZ#nqdrAYj_#ej9*_|tvp%O#byKu5xtbtkFAeafOzVoWh<U@+Sr
zJ9jd;Gjg!tdPmvO=ak2wz}g8EPR+$@0dNAc8hcGunCvhYf0&>cPSD}wO?D#f2?xL)
z1^RRjOjTgz!*57rf1rw-Nn-P|K^O*PoGPhGeO{Kg!5Y<xBxW>@ltDK?VA`s`w{$4>
za5J@kSRJ_WvSOO$m@F~DhdWk-SstH;+K$_ekFz9eG)2IrOh`V_UKK5BV<mM=p-B`2
zW!-w9PHyAb!D48y?{L>mswdO_!Nrc)OKR>C5+10OoF?>Ptgo#eSK?h)TgRFVY+!!|
zjkb*3m!wJWmgarOq|eh&{|ww>?7l_b-v1$IRaO1tAcHvX(_^V`A_1qP>+6qI&ZBkK
z(|aXiZbTCjnwo=7d1QZx`K=;dWSh{R+j138R9`Oa6(g_hI|{`(p5$>qvqacf)%Ryg
zEjo&q3}qZP>w50ecVO?E2vJ=S#mCrj$C7%=%g^v@JF@;G0ePl0S7$+WgF*_C*}V%y
zD^NEJ$m#MTU$WG_(**0*<{mlQPMuQw8g~0+f$QM{R2bd;YXzv4T4&JDg4@|!<V|~u
zLj9y`c=axD_Awt^Uwb0sHJw{NMglJO<vD>i+XM%mvWWbN`JkH*ykvW5g?(WgP=iy~
z`n4v(RH|A18$1MrNwNgF*8ngMjl+f0Zt}sPuB*2N@&?J93*5sANHruiAy=meMi*ds
z7sk($a7k2S)7-ODYgfu@nvWvzD)@m~^bH<bU{<SbMH@yO*UogFQq1S+o0{dGb*QGM
zmrGb=Aui)55KSex2%qej7^{Wl>-S03kT(SgBqkqmZ5*&<+Znr0a%L2#(&-G3kzr&1
zo6bv-6jBKda;(9Phn@+*AOrUJ_S8nyFJpg&f+JRQn957a+j<AURNffo=ateG{4*q!
z+M#!WRDv)F>P+1qU8m5EF|Q~8etg9}rTVqo6H5gUZ13Z(gO9M)hp<~SqaR;wk9Phs
zyje$^dz#7DORRS2do)eYE|`}aSFe(BA!IoVjVovk`^hUSUJ?1Tb9w31gj(Nj4(OGK
zT8IC;eDg*rB^9nX_iY@UP$%#Xhhecub5W=>eW+i9uRm|^&$Li@EiCZM;miDcXGu&F
z#|>BYq`qESL+33;LD&JiG3rG({`H|7ps4;K=-BzDK9zw*eYlU^a>qgao|Ye))8yk2
zHAb*`Gf7-vS##*l3uoEKQY`pxL6p(>q4&~4xiah9qHiChHxSJ~8GOn2xJb;Cz9Xb!
zOP&k4Rkhk%Y<w#(aT1M{Th7Tir0Z`f+yv@E6Uo|eC+KrrBo1prWbPH8sOBzf>fV@Z
z;we*dq$SA&;ffN`6$kJgkW3D0rqUYHG#_fGY#q}U+qpUIC`1>o2J0A!{E$!S$w$D%
zFdLdV@3t?!Yji3L56&#tR%Kidq0;cfaWg!6@+2Z*`@8&SE@7W%gIJI_{SYdjOgH@K
z0RhD64p+XJS+f(3Zn8S()1(*drkmxx3q@^d2u;`fe(AoKSx#_lyj9Z)3l-V#uXMxF
z=8mFS(7Tmgm>J^{omH@USkiG;TfHB<r;J<8Mm{*JFHw^S5AU<DHZ_$Wn<!N0a#ssQ
zhd?yi4N6y6%U9)nY0jj%;0A-=nE9wPwN&!OSr(lnw<Pld!~w2OI7*{ZCFP^s1Oo;K
zf*<xM2|UXXOac=AxY*91i{vjkOuUN@&kF&;e!j$yN5tP&lH&n#(TJ`n@u`iYXx#h^
z*WKFbXnYFioLU#DcsB>^MBTVQ`$vZyjX_#WonxK7v!LY^=|%-X%jbN`7dQ?&aBkot
zPmLC9662@_>~*6(Tpo`E-mkHM;e{EctRYA(X!2?EIsUePhoGEff7H;bH}b(EBQSFj
z@hCT^@!7-?v50~Y&qsDT17Q}}Vkan3De6_%f~eGOz*bl}WGu@%8^B>)Y03s-pHk?X
z^Y+@C8XA}aBYt3=J%)pSW7~t4*EN+d=sqJoCQ)gxo^p3;7u)iDeNHaO1Tds`jYF0G
ze*T$QdTr(D<2CjP3#RmeDj(Z2V=v$j)F^fwVKtAoIcNMAKXJa&h39ZsjLG}g=nku#
zQBGU=1HaClSOUzqtdqty$<@8-v%}rh$h64!mLK}v+GbBqeQY)Vhzk-iR#bW6%ui_@
z{~07DN?k7>6y!FGy!C!V*Xu5Q==`iUXz+Tm)%x;`C}dG$h!=x^Smgui4cbkMXWtZM
za3Am%tE3D$ccVt<7g}2(Se0T6p2V=CFQa>V>_&XfGs_$0$`?GMHXZ26+KA=S!-o>e
zjONYB**48c{wP`2iYR3B9V5isT2>-JQ}0kB5P}H8F)qd^l{XorJ(h~*eD<CNUas8$
zmDVRO5}FVp6C8xRF^bLI537=OcakU?oXO{KU|-ygtT+1mT?Zo@pjki1*!rMrX(_e+
z^3jj!Y27d_Bg!TDlsAX78Ayx2FDPkF^JU`3EL{r$ouxn5S<-zAy6QAf|3u|YD}M&b
zb0&wS@$!?aew`F*v`pgkA7Gz2d_NS7T)GawcXBcYC^n`(Ac`}7b&tlW#EtoQ?`TR)
zBme30MqGAh9LIBW9t;JrH0U0}9z~CfOHg4#62*$q^~+=CniKR{C+3xT;jgF6OvfxR
zlKFhWa4z(M=x-Vx*BmQeoaDSI!v+Y*2J(ay=Q?7Ancyq@ER0ybCpn&Z4BDEmsIc3c
zL5bysxBefSSt0LK*F;5KEwB(zt6m+^z>AG`YjH9df1+l%MlnH*ZT7t4^d{Shy9v=Z
zJ)Y7MdUX?_2hrlsv{#}l`0oSA9s$S_c()S8j}J2n>ErgyO)V}K72A|y;y|fC=atL6
z<dx<|l5RB?T)H%EPPVQ^naEvw5ReJ=ruy#N<a9wWz)6zX;XC?uW!#u+mo<w&Yr2bq
z3WEad<yCStI7k0=$vkcLkOd#;l23A47kPRT|95fcTl*n1@?08o;(#$NY{(dD9(XHX
zxV&2j+F^wmq=Yb}ma-ax@nL?Q+>So??+yl*f!+?T+;Dj?-j7p)i&YH2M>)qI8<Vyx
zww+EvC)_T9k^iWp>%NuX(CB=$BfKv$B>YJ4+uZBofQVQ;nZpT$BU0%N<uPBKZ#>7r
zS()*<6X(}t{(E}dP4Jh&bK<1$0>#d_0ciUEHTL4eRF$TYUD%ERHZw{@N=xuP`)Ayq
zLjp=Afj5UPmL1Oj<;8ZGMrUZ=X}OaDjOA!y(uX#J0CrbcNe%1mRdBE0e79lz%>yoD
znoZ!XG)%EQc>F1vX`nr&95*1fovIFezEb8EKPA{vl|`PDOg&w@i0Q+M6_(t6=70B7
zWrOQ?(d}xBVf@5s&Gyn_smMPm0}`F_qweMb>0kp3tbKl=<&Z$=;EU<k-<3*vRx&8~
zoNH)kUPNT<z9zGiyEH4Y(fe9nS+sl8w)E@lnfc{`)BL;$NtR#xr=OfsUY0j)1?}T-
z1vj3u1_f)sKM13W-4jPPNveFw;3K8_7(S7={ROftbL%tQ#k%D0(v%!%C13{ivJTie
zbp5#7+ek>v3V}`tIe~-wm_;)7^82*xX29hcJb}ay#uNKftj$U!6B||Bc&EUhoD18k
zAoz69DRIbp@Lh4QJG23WQ!=w_Td%1<=x-^Zefir#NHXOk*dx&wiF3iY3S#ga1)s*V
zT)oY|i<qUwJadY7M;{2gE<!TXtsYj@)}}vJnAoC%==Pbx6H=pry5wO)7Z*Y?U3Zpg
z263^6!AN`YUK%x_OgsX5bg{@){qG5?p!4Gczpl0!b8<$=&uD}${Dt2q6_-gzx374?
zfTZ`3s@lohF7p0o?v~2R<AL_`ZYE*|+o&iF&$zL->&f6gZj=TCY3t<d)#j>)RM0;-
zB}Vnr!<hQnE2uVk*IVefhK7O3Y*D&kC41|ZK?!8_TZc?h$TVNF`{xt~+u*=h+-v?r
zXixk6-~GkF56`rK9u_v$b?pe$_BQL3&ML1;GC#A^0$m&}Z-bzQn@J_z)N4Xz&S}Ak
z520t54qDZwpBtLgYUj)ATTpS3TFzKMzT~PW(3j%bxTJVU07zw0HChl_JPcle3UAyW
zkN~70LUCTPA2+MNZy8?n_5YG+YTl`ov7}1iiG2&|aNg6la88>ggI2Dt4dv(x8u5X)
zR{fGjF)5B*#KjK%C{%b&hyS5e4ql<di8pEH@H+uceHwypWldQ#UYreo`(YV?oSync
zx@r;4!1L9z7)TS#sq1P_ourX;IMY0lql(4Lyqr`OraUc|_(UU<Hj`_9d+74lNj&le
zRsbA}XzIN}B{ikmKCf;6W^cQ>V6IO~{5yAdEk~H?bU+Od8G_^}#nBfy!Tm0$*u^^i
zKm5cMl1KXvxfGOwk{DM(+hGaU?TWepxvCn|*(GzQfPlU}-NzJuM};W<O<*xS33K{=
zVgfK+5~k0b!cVLV*vAK9I+c%_EQX3rog@T!E8Ad0`WN$jc-r4*o={$E*Zx9qlK<ri
z*xR0@XyZ;!U;T}PNJfejdxti3`){sjbMP&sguGL<V<Jywb;3jQx)Glnm_+86ZEI*h
zBYWP$gRv*<_e2(-eI_Yfl!|0GS-G)!V!e<muJBkF(HWX_x_s?9?yS>jJMi0**!XB1
z=Ozyd$s663c|F;HbQM%tFMih=v|rTdD$lm6t=?$4vHszSP$p~7SZdcFo>=Kvcr4p0
zuZ#HJ!CE8^eU@^2pse=s&L`oIspN@ocSPQEvx|#Go4;fdb|u;5M)f{kcc)c~XiTQG
zmxL81X_VZjVh{4!m(ULIMe=_jRB_}T#RwTOr*2Vg2p^byV)4ND3nDo~L)HD{kWBg&
zlHhmhrRQsZo(ARvjof}Gd^-Bn+}y%xEWisO<Dv$8H%El09Qx}uj6tQE`t9{?Gd>6|
z%oRybS%@?7;Y#()2#U1whsFg@_ieEk{QKg0iSJ*(viv8Tq|<vB4RZ+fOiCJ+DH4B%
zC9MdfingtVOywR1Q%xCvP%UyNQYO#Q>p#8>PaLhcOzRia&KHL+tdH7VDexQG@KQZA
z&loA}&RqKB_-RszcYV(O+d|x3^AB}p2Y(bWs>Ejw)P^C*RbrUirhj4|c4;wGFI~O=
z6~i;32mdtk1U?Q1>`?{?Z6qp+fa&sbtYUDXjL6KCqbI{?IP|<CU7yru7phC+Kw{L_
z)4&eib96mg@Wb^ht!>4c;s`Jo0-)HRA`;UBlG(aQt`GbR%k3!CztbGO_~)5IyBNyq
zkT|9S<&pjiN>5L(=$}EFW2%l2o_kJuT?;Sp%!~;)P}v>;`?dRTFo5)fi);?vfK)t8
z)aQ1)JRcbFL8o-1jVzuAsqXunhj@K%vegf{NY_4Z<+9Ek>yHdyq2s4Opnmywg@qJ&
z<PxARX-Ny8rdSrg5}y{;>a;hgSO2bwpASx7L_}c|43SsIA0EfReUWkVRB;bz&_e#+
zD$GYn{9u-_j^2=@i#2h>HNn%MS`h9cR4+ScTBz+{!9rHf)ZEIBi|Qlbi>d=HRYHsc
zsFJ$z^>aST!|13jXVuzewY)@NOfah=CjFkz^)`RLd^kSt52MNughF?|Dz@x5xNG4)
zyRWslVgF38Q&32OMH+#q^JA)K!s^ww;&KDxO(@?37fI!SuS11+RxTXK0CuGtPu<AM
zlk5C~B2nE$Pn|FB;AP3Eq2OmG6~DY|#%s*Lhlqx%+C~c-E3Sj5OjwNyoJ4<fqm8**
zyKS*?^`CR4iLF@MTFsi5T2u|%&)5bGus+urc&gE;`ul=!Swt!r(J?<Tg+PIhf)o2d
z07D&^8uNQ6e^H43g%NCr|Nm+#rxC$9%GX#q$XWRvF23vnWi_RLh@haK+Zuz;RkO1S
zh4T;n{8}xWsk~N7LFqrK3C1VEf{+V@IoAg|m(eehIr3uLExZdI=Uf~er&1a{;r=x&
z$7-hVNc@C`28DQL9y<SAiAc~6_?X6SQ8qI-$4N;9L>EuKmUv`1G@d9uqIE>jZ(j5|
zb8Ocp-gS8II@|!JpjLB#De_c*ZxJzsxGUxu_w261KET0O)ANoF0w7g1ct>i4MNcw_
z6rLd<_AD3`<CORe%++D$kb6Ss><%jgX21FMajvSy2N7ydZINYM@q1BpNEBetOl%Am
zjTyAC6j0_QFCOR${TuvGxcqE3WxutVd_@w9%3uE35P;{Oq~{5GRJs=x9|0B#{)H^o
zyVR%WNih)$0QE;i&-gHVE~2{T=Ge^k&>g3m$Roz~J)0vD{`}t_ufFCh)%U%Yas)gY
z+X<|g-WJ7lvEr+-xc>(cIF`=Pt%kc*#`!SOy4AdENZK`3U*Rv6#j-X3$bCHTX4ZDn
zm}d>%lc7~AYDP0i@b+iGC8l?av=+O!SI-FQ#vd8ot#~2yOaS{y+;sK)K?akAzNS`=
zlU}?@)-%f2qYq4{WFga*dT^tP-^nKZ({ZQseMjcB3aWwwz~H048w8VVFs}CABapTA
z=7)(&lQu?J4@FX$VW;BS-l8bXprAhnnbgm4IzpiuaSy{NY)uv=o#S_U!vA+6I2+aF
z^-ZMM*y~r7$&95{DV?@!14T?kfkyF4A}xUt19B}`D3j77-yX6UoL0Kt|H#UqOv2*z
zCQVsYRdt%H1V_Yyur2ssE8YTOR!0JI8g^XPm56qspfUP;&=?+tQ2E34Dk_zf2`6+I
zBQ#tquA4B3bV=7G`ONTKxU#Co%h)szo$TO!9E0?#pNTnTfOftONVmJ$z{@J-s4Dtl
zMp%qQlIhPvxwSo*C~R-SUX&8US<R_m{&z2|dvs_2mqMEw+~xU$fv`}gB-d98n0DK-
zhTm%T6(-%>t@N)OS%<3LS2tt<^&x4NU+tAj=N~q-bZF%Ag`M44wH?h-!$WU`&G#K(
z^W!lMRk#EwUSk55gFLK1SqmS=P!3m$>*Alp13(b6=Oovu##&M<D=#5Oxdc}}ay1Nk
zc(v_Dv)!iO=Q**f_eGBIVo|WwE)A9L+}%DuJ%czGO}H#Y<w*;_lAWIW-3`GDvkny?
zunE?<#p#UKz&fm{k=Fv#{i?(nm^75K30x_rSz6sN>m0$AqQDubKJ1i294aaE%z*@5
zDXy<6fT#j#ySVHUz_2DlD=(-2gT2$r2sO!nFSu1d(u(|M&E3CRG@I>QE+2D0Lf;I?
zDJ*dKhPoA0CykyT^8QH77IOO4`zb~JuuF2J0ZMFaub%uZ?J}sw_qy2bbAv?sA(uyf
z9U>rWw!~0!3{OJD#5{eRPM2quC$K>muLPEK3inLsBwxro36HK*b(LOJDQLfJ^>2Ly
zXr~B0b?P2m`<)|X@wa}t9gV(SU4FZAg`08Y-*{X@U~LMyTlj>(8d}hoy+7)5$J3hc
z-<_n-wmzRtdUZfauAb>PXHl~y=(pm}bq)wqu(n~DyD#0uodwWtvUx_8G8Vo34A@&s
zLdtigf4>-!&MR#(A8YbTz2MH5va`UQUzpX3ZyMHX92DPfINhVsSZELbS#f_};MI3k
z!Q`3jftZDN=Y(~^Nqsu0tS5U*_E^lN9{;1VHBpN!x;O$4V1G%LbdJPk0KV+;Rx_&5
z(yJ4IWsBQebQsjnI#R01OI|P&w0A{a?{O_g6nIB=uQ=%OQJT-otH5>cod$wWvz6D5
z`k0JtrIn}rtHZ1lVSS<s8M;B&tNRQC|8x+6L9A+x<IgHmU(Xa7d#&|_yWXVQHbFA}
zTq>E_27lU3FUL%dxittj=HA_VLSiDx^rbuRj_BTui;MD>R(gg=y%tO1^QdQa9V~~L
zW};GT2HH-eb~k4LSPWOtKHJ-i-yire3=R)AHk7LV9X(|m!xyaAGxP03i!d4%Y8t3P
z%esXDALY=vhi^x^R9+%fPosB&!hxAioeN;-UM(-?#-OgX!WdD>8?zF%REV~hE|SY1
z-|c;F^vpAW2u5*269~SA+4-Fx=>|uTSg_TaTZuO@3J9-fv+R>L<RiI>yPNkOzvD<P
zt}uNq@PIz#RrePHc!E8#KW3*Z6AO3S>D7)svbl5Sui9v-Ttwmdq{vw`JLkW59bG1&
z{TDhCeLknrYi8Syd@IhUORao8CId(?z?K^-QDX>q9Oq_RL(}oXkUbC}0(E=t&;@LQ
z0$n6oC3C1gxZUqAbY`EhwBY*4-SnY{WYtP=GbP^t;^{>XU3*uC0e_h>g}KH2^YZBH
znAW!2DY53TEg?9^1C~4zkQH;#=DDA6sej&YzC|(US~nmo=TfHV%LgsOST>dNR{NVP
z+#47$F4eXx?8MUz?j~r_QUw~iaE+U`MIhbf`QRQAC+RzE5uz}T&L5GvsM8?4#}=y=
zAESx|8+<PTIe=D&poh-pn{BEl6yku-v!VgEIXi58EQ#^Qq(r2}bTKQ@V9$U@u&$dG
z8zlk%y+|V%lkI>D%uGlogOc3jxw(hqQ<5pC&!B=4n$57lMAYcf+zm+%UP+pME%PFR
zhKm<Y`N8=oCJ>Ts+kzCXwY(|6MWb2#7Y(*zPQ4F~CP&o&D05l|l68YjmDJlz>3mG5
zzBs{s9q$IU-|?tpRXYFUB`#=@%U$*M)jm;*B<}Io;Hv6II!ULXQ^uVgD{>>Qk3iQ3
zK^0p2C%oD^p`|E9IzLO|02OI?eDQp*NAfY(7jx3J(BKZ7=+f1C__)6Yt*rIOAI9d=
zE*Jm&I?d5I>xDvF^ILMJ&A-BU1i&>V=ujG+5sg~WZ-G5Emg4+C-ND2KrHiK%>%0iS
z+eV5HnFaZZ522HdNeD6!SD|FMb@QByq~<y#jnhzaJBczvUPUX8Urk75a84tMsTQs&
z*i~>f(!@e@-yiO`o34oWg}P+NnIm5c@&(&E9E;fbj1^VueG-Sr3TOqjeJW*9oL1(A
zXfnfiqI4MONk0M&Y3T{=#$QkU3eo5cde(r5=Hw@5@JMSuPk<~|O27U-J@rK<4?o9|
z_E0qN1|)lDO63O9b-PS2EuP2n!i_ym1ax=JcLIycHhaiZ9FylA94In0`4YACbZFJJ
z_?W3ra}pCuLY}9stLy3NnN;{~oL7D~Gle3{IEmA@5(DRre+U$Zr6O|_>9XH-7{-9B
z8R?%xm@5bGnp;#KT<Mk@=^U*oUq;PJ^Pr8rt(P}4&1zm$YlwXFhfrSq5GX=?_Z7~W
z=k~X^egEAVUb4{sPh!Y9N+uc1d_fH7)h{DATowLeR_Q#Oc-Tzy;{zu@6@x|Y>ZOX2
zA8E9B$<k}OSUNH?vSTu*vW5&jPeQ-P6G~GUXAk7wt|#x|Z}21#?WJAy+}!M2ne(2k
zo@mFZJjPhCTbvlSs#$Zh;=jQE;eJ}!^R$v=SjO{AN~YR2El-y?3m}m_0OHXks3Qzj
z05LtQwN{`byWWzh)r?(X77PQj05F~pY=gI`bhK}-440i`1)+9|@rVpu@YewsI4?_(
zcZ43`kz+tIagjLZF`tE`T5$Bw`_kL5j@}Hr&%}fbS+hHMSMgD*<?hPj#UqCyU?AL5
z4HY~$K-u0K7R@0K8!0*vVJH>J^1rD-mSUlKE*P7%**7TbClZ}u8|rhyUS$o;w0dq0
znv&ThIKPIQ?@%*OE|o7?{(BAIELY-~3_k<=Qf~het$JvjApB8I;!V!l-HTQQBL1SB
zRP0mWYD=!sGQIxrSQ&Cj)!Sg$bgq7zxN~G=!Hqq#<QLdoCVlAV7SI<t5v=oKyG?)}
zXAY1nMcItba{99%8it~z^|ik_`A3|CErq>YeW`xFVRYiK7llWoI6Pgg!7Id}L7@0+
zE_U{g<H4iYp(^y8WnR**Q07N_0KN6?y3*PZw5*D0D%^L>xJZBOO}qF~#&>n2XV}~)
zMB^iF&_0)Yfi!_0vOqHpmFhJeXZO9my|;m(>4EL)>t^KFlpRMOqVJ@8FtBG9_WS44
z#&eCjLge9xT?riAnFE%U>Wps8)gCfDrgH%yi}%RY8O}>)Q$=!UHAYv<vMpeHthz+s
z^NEcap1h)5n<RhN4NGc$Hke-|$IxFUz^%8PY)e-|l&~+zI!Y~~APaK)tDXDqg?Vgc
zu&A{d(PwM%ua*C{99uKUh9mt}YHKEEXY$8)h=aR(inP+8yg7J>nMdvO->;Qoyzyjv
zIB5@^${*v->?GQt*uPLkv{(Is)!mDCi`Eh}tFzDutP*_KIuW^@J8!g=*c0>Az`JVf
z%k)8M3jfoKuQB^zdlAYwexfvscw_bs4sn=1%Oc#ds+5%!x-i-4+4TB*?t{l`1pSmk
znNjjAl6m(k<faH#tXIkwGC%T6S7!KKKVx?m{n$&&cju*eg7tvF+V+-<#mCfS@SjlW
z+?~B3k7+~hktE$2+M1Gk8wU066FhI|MiSI1_AA6wz7kYtrlQ^n+c_UX`y?#jFRG_1
z<OsJLqxKw-_<_au&^a_)mleJb7rdxLF{uS@WMC~vPHBHqj=YgAn?|HKYcQOBnQY`{
zlq<z6#l07-ldt<Ged(jS>UQWUsK|M{g~sJU>;ZyMRGHo?)cPTzp(DtwBBIo$+=6ZK
zA<?AH&o^^!lI*|Sz3=ob5kj79`#qWdAQEcR63dySE`l$*h(^)hs;XIrW31!;s=1}j
z5m4GQrO_$YuL52UZ%cXE9c@MaeiNwXt%lUE3rf^Z^2~X1!PM}ak|x`LvxQ!%W%(>6
zUZV_wL19*xRGK45h_S9N;8q%=PCP}Aqq{~uEzx@4?Sf%hL@DvhPJJ=0)cX7)1NneC
zRkUhJBAb$rp@9$cxwZ}=3Y~VB71AqLOY{X4jFRf{^MR)SVI2PRQIOy?H{0x!V?_!+
zH!51<ac^-WCfOfykvCf|UjBYEd=5aPR$!_a{0uAg4OnViY2-uvJIh~*IhkB7GJ??G
z1nao+ULhW=uy9m?1u(^-+B0Q>=+ngfZ7WQyCAr0!PpmPwz5u3D)O|A@y+_43i_f^%
z6*#4&vdSsWQm0|2alCUYCN)bWujs!(+V}E_?7sc?j$WnTV$m;(NXLtupGm&DI_BTU
zZR?7GcKX>fzjSX1y9HxM^A<X%OXpc8yk)ShY}JjztmfoOXq<Y|w^EVu4X)Up+?@;y
zpuKF)+m%(9DYLwFkKx7?C1?3$g{J!M!UnXnWY0^k-C;^?_^P0`><G(g8FM9ztBrHv
zFC9r;zxZOV_MSYgEy)+AbHd*(>3WQ!j|hMK$W81gs<G?rAE`y0$#U6*%N9cd?KKL$
zK2i4>WjACL-NBZHxN<e6x+&zGh&P`I%HI7nerG65HFJr@^NEh&b`q(}n8CE>G@~L#
zHkWljU)Dfe$G-BtC8vRcn?ub_kvl8U6u0Q0<&M5drh7ApfT*eH>greV?a=FXrS?5&
z(!a@W-1g|&%|f_^32#AAGIT&vge5>7k_9x((Yu6(^$w+%TX^}Tl~c&o*zmwz@-02-
zQZy4a@5dgvo8~$L#J>LUbnVHhWPz4z**QGzlQyt%)Z1sbERQNQtu(HVN3JrUarnY0
z$@&iga!Z&@0*Us_WJ&Pe#bB5*r2|u__S9&$!SAksq#UG3zd%W?crf|rcan(D<Z5-I
zuIk{AI*s4tZ=OijmZoXFsTULq^i_Q-gHgU6ndhzF3Hnlk=4A}KT6q*`yFk=x2psWf
z+&R8i{&$~dGf?x*n<~aU)|HE2bLxWJO}KC<7`)5|a$CpFNi@yAAFcK>we-Y4!_gf%
zt?~SqJHng14zjFV)g`YyK5bEbc+Q{eoKZL$Yh6xm_f80!qDd8UI&?&Q_ql`CpG#-(
z$0J~Fb?Vp`bbDyf%#6}L;Z)p+_YYm<NcZZ<er;lxAhkCok@B=cOxOD^d9rHqFoeJm
zFXMoMF{7=@+A*8@SPyceTbR3p7^8eWr2lOuPl{t8nr69kbdu@hUHGbbL(5d*a?ZC-
zxx#st`g&VQGu#B_!0{MB%=XXJ7!Q&R8a6m}LSZ_$v+l799Z)ErnqNOmoOMsEOT*CP
zfMdPI%+)dogHIj)fPyDdhV{?*KH{zT(z-@)p{aAF*47~N<qK(y!UZ!7Qp8*>-)$|$
zmdv8>Ctj10RbTV)xyO*G6yDPe)1=w4W<Yj|z;6Gc_tTIIq2`-cqKaaB4%POuqwbCQ
zD4G+p;}#w~WtTh#t{{;rph+1=_>&0RJihpzVu2U8$`)E8Vt6O>E$zBWD29R)rK3$7
zt><-N1L{v_7*=#&Eb;iW)1INuOMnHgQqIzQp#3Eqyk~|Dy+9eXe!&dbr_ngpWGVWE
z5Dh?mLrvDIV^dsy{|hBu*1hZ0t*O)ZR(!QRDDZ63ZrNG#=peR>dB{26_trMyE|UR@
zI>}{4Eae+J#60(orY%1)yu5B5x(cmkh4tgY0#6msRFbM+x>bY{$oYiz`UA>L?NlD`
zoo)eL7j_sRns}#(mZw+DkcjJT;YgPQ+17a+Cz(IsbY{kV^^)DvEm4DLZ?U!Ku9yos
zmkL9;oj#bV?B_bpyahP&D@Ryy_5237SD5%D|5)=XJ@QEN>YrWMpqW^WQLR3UVW>I<
zR6P!)VnIB{mmb?y#BT3FY_((E<)Zp5CzueJX(|~SDgpPKE-6WPHnvL@jIGZ5yRa$o
zh(Zs;N}r;MhwiP`2tU^d<o4?Q<d3b#y;(_xx@tFrNLoU4x+sw_Oq}JkV*0B+2^mbu
z!Q3evM?CtayM6Id86HHM@Qgs-h<Lsq;BjL%xJYz$Dhh!ONpBQR6i@6TIvD3@4a<8Q
zxRNx?#pS3H49>(l@)^DNi}ouW4wppKIXmzSYf;TiSeI1RjYC1?3HpX9;yimA@y({#
zB9ETG7$MP-uj1-k|G-#_a9P=}$~Oa%BUZRET|kOc4AM9g5}`7gUOV|yIH2gWxPG$-
zoz+U?Q<$Rc?`gJ=$olriup%ZdzVGUGgpbJDy_*|OOaM35^2mAY<T}BX6hiGO0)sns
z{P>~WV60w`FS>m++)Zmwu9Pi+irovfCQ!Zm18viq55?zFT_}3G5_Kkfa18xF?SHui
zoyy_|@-z+6c;1$w58v2aXcRFiBIh*^*$leZa<vnAeShhy&kN;4EbW;k{fH+{%D*}e
zyj!vQ+5AW0{qC(EQK{^yJ`Wvm?D>R%Pkm@mbh&l##|))zoqIHwZI>tdfj%@j$s_<@
z_MEZJG33Zh@p(U9vpQukX2Ao>S!b;VQy!^bTa3u*CU?`Xe|?+Ci+vA&c|W2nnb^ud
z50>q(HN_%Zb&(Uip9}OWrmh^Tzmwn;3%c;e4H_vs+2fW3o13$gNTk?pNW$HRbk(_h
z+Svhn&}@FAAHNVzqRHZBQ$<*0Coyc9oe?01*O_oAE4kVEM^X=U$7i;}r1g~4v3&c*
z0L}MBfeseZkrKet)M3eoc;E44qrPRP5WIQ7e>-{LU)aqvc&~HLJXK=3Q=3k&bJp9p
z&A$+nqKLQhdS6T!;N`#OwywNCgABnlVnv|j?JK-1N-P*Rs+ezg*L{l0BQPJ;r;1&j
z@ZTE_4>y7-+wPm)FK+y4PifXXAf0o#R?yb>Mj*8%p19;d-1051|7pkhsM%q-*}GXr
z{Q~_@t8Lb1BjBIw3*7@P#WXRBrpMP&_+^^)p`Op^4Ox;%|L#6Ya}gZXfc*RCg?dx4
zvJi)zRc=nWQ&jjdgS|()Ktz9NBBLh;zoMuE_2$>wL19mOD9;HQOZ5*`R&^Ol+6T)I
z$l;?}FQJL9VONJ_SN%Q+lF9Wfk$^)bZ?3nF8RXx9{JK(2m9#ZSKlnddzEn4zB;$za
zO}SYuUzlaV18M%;Utc$Bc;oybNk6>#6AmS0HThX44rvjHjX-VHC-!$=Hh*3;=<Xxg
zOlX7K;puONI3^n|7eR*f*$g{8Eq^!|v(T9Gshdal37!W^E6d`13qeIIXy!oF!Rape
zwWe0loX+9lF^gLSzCZ`*zo54>F0y%JuWRZ$I{IeQ-4p^agx9uYg1mSPo;lzw%dPv;
zA3mnA;B>m=<vJAW+tkzLogJg1${c4d{PB)Q0kCOCo(nCKewcwb)~0Q+jK72~q(0zt
zG#R^{9Qy6(w+pOAK&THXF&lK>B^2a*mezNJma$XqP-prE3pwdZmWVr?W31nOX=p$G
zw1K(uHn7k`PAiZ(V!?(9U{mp?npVgQy>h9gkW1D>Er*7|EjIOB%e~#eL1Q)MtRVvk
zjXk@VqTMDI31Jf30+JJ&nwe=m4(&b682e&~ts8I#6wjJFc!fhI=F4Z5eB)Q@?7qV&
zQ-|GHQ-;Q=(%p_*$QG7cM?_$0iJHeJp=>9v!75P_8%F$l!HU{r2p3t{x7N@+Y26Px
z7>f#A(MMqOg(>1OTG9_j)Gd+ihqCq@U2JU)^SfqjUCei8ZLpn`{z?&HM5^-;BGHHW
z*&?C^Ll{sJrs_1mg^)p8IsK&EA{I2x4|B@ZiwYFFugrYH^?*lNP(@gzEC?KXS}u7?
zgdc+T<4Y18K|;bJA7(yZ0dFYhM@upv8yT;Lav7f+U@hW@KrOKFz2(jx5g0w5+d6Ie
zd+K=gOM>4I4Fk%}@Z#mqmwQW#Tq6^4s1zquG-BfX5V^n5(UZVohuu@rko++F#e;Ry
zzsq`_TfEb`W-JHt&+iag(0?^uh?~u5S`ze!$Je}i@KIVu7YR{3B8PFTJke3>?_<we
z#IqsxE-rDwVTmY6<Mr`Lg_HlDL=wLN7EnID=(c}j<2swT6s)>mh;|72$aLnu2JmC{
z9FmNAP>OO}5cU_%6p{rORD+C^&>Z$1t)VHNaz3EWO6_vBDIv`OZ)6OT8jlxXy3=Z>
zcek%3k^pM1#tibs24~FFWC>3Rjrfh}ToUZKfG4LRyxJPfB)S^$g_{TWMA{5SNMVX8
zL5#1r`e+4;>OOLutpiFY7TtUzQ?__9S6hHaokwB^Dk0V_S3;RO)XR`BQ}cT1^>?c>
zqGz3-W9?JMng<&53J%PaZ~2Bo3x4e<|FRT^-6QALZGF^|SjJg6mbiJchu2@qiJ5oX
zK6aBNw!re?Os0&+&GaHuY1ojZ&QZ(ke(OrPGxRO)_C81Cl~Uvt`bN`xxm6`tgU&%>
z{OO<c=S-#%bSbY6j!jl-;t>Bt<DP5tGn)bWkwGkmfGUag#{qYf4_<Whqp8Tq!U@MX
zQ#<lBy7gxwe4ghz&%7oPM$&^Mu-6{Q#l6~vMKG~Stno)6dh_Jo?D8_6c6psCG^z7w
z>-;xs<@Y%&;)v(WrWI3Gy<3SQ`l*DjYg~&xc#=+3&S$yg=p@vc!@69VE!`z6SeEMx
zIV*KsLF?b|uVw6|x0`t^AIewyvrm|kS2a&hK@lo8D2oc~wpgbc{A$fB{>ZaV%*DTy
zv!Q|;;k~`^=uR1$XDXh$xr|7Qi$~3@fnV)yLmX2i$vppc-#_X_${>c49hUl-s0mjT
zF!XiNEB30n7@Rp>@uIcB0Pr)G8M9%eIqR*Zh332aU|_%9HsLykXq>myw7Nw^O1WgL
z=9lFxzM(3r(`a1c9$MJ5u^e%hE}Zs72JJ09CRgIUJY&3Gc`x0_>!D{tXHv~OXcO<p
zaGuzI9Szn(JQQ3D>q6a)aooN7>{9K?!8aNWt^sw6ubxKSF0bL0Y^<B|&HcTl%h-mz
z7Xgaq{@(aR{)G^J;Bb)9o1S*Xj3G4Zga03n&cdO|?``8>5di@K>F(~1fgnhOAc%B#
zhoeR+-5@zyO1ebAk<tQVBLt+SHbQdJ^*+D%KRA1K>b~yl`W(@=eJ$If?U=jBqM54#
z*b_eLe+q?=(L11jc=`yXgT^h_P^vfeTPP2ES=OoucF6xhfQN=V*}EhRiZVRg@js_Y
z3HrOd!h)y3e6!=&x17c<Vx#h=Ru<L6MDi|V*v@E%KZG4D);M9v$(w2OWE_O_(=o0-
zy}18F{2j8%8-9g;$m_gdvTgvXnn$Q}jPQe~Z;?VB=1y%l#8f!Ip%So>$bc_~HvI(U
zLWGo$n-Ke1v{_ZZAX%s}$)3}=QqZhvqICWzC!T)OFPjlpO(O>@OP76W_Sc+|>rnAl
z-<b#*QY#@}dEgd|o`><w-QxNk;<EJKbOMbhq}4j`jerg1<`NGBWW28;KtjovLPJ%L
z*Y_Im-jP@Nep;Jzt?T4p9x!`bjgKf;!S>Foc%3pwu7Hlvvmhxc(2aR-Yt2Dv^xvcd
zgQ;&(b8mopyVz&_coHBF{QCIr!`Yd@yz{O1&#^I(2GPn&u(IA@8`XO3yG;B~Dj3vU
zg5nj0Amm+yYvlPDZF|5`6S#duO$zkQ=v68Ez<5Edx1U3_fAAA44eg?jQ|f=^gjZjn
z50)qX2KOvI;%Cs<GD(4hPtD1q-!)i<Mx%>5`w{hKw+jfOjQAg9lzd{RaM42e<j~=V
z4Hhd$(~i@*MANW_?l?@+?kP70@WT}>{m!H3U{~Uf@`fi*h-`f)q)H^oGnCyL9F>Uj
zzVot>ZN`NTJcf@mbP;q2*o`%1FXLTzv)rx5F&{r?<t%atwEQtPMmzk&K2NV!N`xC{
zQ^Y2ZC6ZUsfDMliuS`$LHkmnY?=nkLu~nnhbaQT{^Y7{r_|Ct$;|woiBM5NDn^stQ
z{n3b@r7`RrY9vhecQqEikYf^YI<zbU62bKK5)-QVjo}A=f>yA>71{Qy-Glt3N?I$E
zx?fwlHj=M@VhSL~$qtNVtXTxQ#+K6KC)QoaoB{LCuTw~60FSYYu}N)nsO^MBtl(k=
zO(MC28y%J09XIL^<9{mUqSSrP;trEcn>UO~f$U=Qn)NG|D#jXlG^JHoN~Byo!w5Xy
zXQi$gn$K@`c!2hUUaN|M6<eJI&?%MeY=^ZgY|ItP&n^#O1Pslh-YDBB*P9MyY)ruq
zY<<q;h)cFuomV>!Ss{muZmmmhT_?lsu#j!h&Hyh=i<wleHW&+On}~%t;pZ5~kV+Y&
zM1Ja1K01xCgHov>bk$HJV(p&YQseOGl5Ty6o2@-k%<`<Kxvk(fdh)6(u)|tBj~pHv
zl@xDRi2)MDZcltls||RWGYR6PC>3hR-q9+5dXmw(^%NMihOre(mE_X_V9)`-SLohj
zbc*v%jw3lxdeNr!C}6QATzy64K$mc#Zm&99`=xqM1RBhH;D?0__pWH1jOwrt5T&B+
zaj6Gjd47VgyEGoqXsTb57izf}Tm0|i=3eO)Vz>(jnnCS`!i-32!jZtDSaL><zrL%k
zfYFV0f?toEHHsafPbhY)8L3&n3x(3Jfmsx@ToX0zvOX^kUcB-@TwPyPtOiG;&5U&4
ze24#YZH>8(00rRODWUoJ{`c(yzMBg_EY#Bx@5||op4JWC8H2)?PoUja79k`(u`1-P
z<8p*o`b-iU-GZ0(YoUc+p=bwl8x<qu#H=Nc-DC3#2NQAu{_cRo;Gx{~R+Y9Y#p?f%
zjZN)R|HT-T@+Jon@YqU9N<NKDQ|+Fx33wSk$#MMoUGndVw~{&q?1Zc_WsH#smO@Y5
z5~Whr!WJq)F<PMJmo2B1|30_fBvvt7_qiphT#R*G&HQfc$Qn$QX!oAV$vO>s%ECvf
zSUM+QhA?xi=ZmVAD5t^t?z_(FQuN;Gf|B=*eZQTkn0Sq}-6XAVo>;yP+LF)CqBi~O
z_vU@hji{|)%Vvz2v!czeLWQJ_@waAf<Wq2;xt+WGMQ1}pOvr^)!|7RL?;rnV6PM%a
z{&&yt0#?2nha?IfsYtu@6_nC_ZsHcg4k03nv1Uo&jRIyU<=3{36!euf_Qes?B-8mx
zmP3Ulo37nlTier!wB0J*qvH>r?OL8scnKGN%a2~2K!3r1*8RySd=#muS^M0!o|A6k
z<yJ`O88+$9^vg%6uW`}Fww?s2X_4Y!ml^tx*1=!`4~c4>s$Vtsf!a<nhV^oyvEmc)
zb|cGKPrWf#yz`docnKD*<0*5h<nW;aSP(wE!W&*cQUHUB1z1>Ren>v!UAsmQbX_lk
z)p-gOuXEu9y7!TA4r>9;XlgzgIr>K=(G+sXVzHbtwxSB(j!hZNtZF2GvdPji=ka6C
zGWqU-(2f;G4bLBMzs*U)yic%HIkyabLY;niXHL1|;PGd;L>M&n`1eQav5RmD*?mqC
z#YHU!rNOjt#+xY(muih&e)=hOFYrU%@+8uV2wrK-B(WDwbQ`9~sBgO(Xl&YkLcv>&
zcg!{Ge6x;=p)wqpP2Yw@YGY%o#^Zyd?w>~T8IUXQYeJN<9EEMt#vkvfrWDu_5VJ^u
z=`*^d&+H$`qNcx2*JV`1+c9Em#}E`&F1{!;lx&j@_er9Bnwyn!HMvXgHZ_CIBH&ib
zm1dY$VU+}FkPsd3>t7rLsY;_-9U}>gUECRM8L8kvZvMNS^hklO!(obMF&2iSFmV&G
zifR1*U3EGfVEJd(s1b<EGBvkRfi4W<B+~~X)ppFe9cQ8j5OMcSx}U8pu9^^?u{ZxZ
zUTkRFy-)#U8v-gJTYJ@%nJec^;Mmmqq)9&oLs$s(kIcYw_d_aLvHH{ZGgUiP?$rS)
z>U8KUVkjM9K|S9Lk}9_PpTsm<BfPiC`H}RpCWZ77%#V!RuG&x_{34)t_#*uHT|YPL
z3=IgWSHy4of!*O8d25XGaZ0N%YrCPrcR}lkC~KUlv%t@3LO0T{@UP7aj&5#bZUPEd
zQ^lgtFSi>;UZ?#PiE`oVa!!b%^6s7l{e-)_;E>xsTy0Jssl(MW1%?*CJ@$OQXKigd
zq8f)}I1&6gmtoPGGh9;H%&y*N6-K$QD-l&<4k5V;?RE90<zU%Ytv^FrDeL^RT3bt`
zsHb)br33B>-Q4&J7UebGR0|h8`<;$ep7?HZ7-+%!yc_ah!TX4&=(J!%*N=@UbyLTS
z;P2aNtPuTuss^-0FGRVtIZg~Hc(C5f4lxkWEgpP0dKiA!bGKqtE#>N5JL;E$rkMCS
zWP4#4MEOMJ7HVNjL|&`Fd~z43=|SPa?gCs}Bg9+WJplYqGtu1Xl-iNcB;CpS2ow1J
z-&4V2*+lY5a~GBm-Q5<Rmc0n~?B<qEq;w)M`<YMKYx&%Kae4GUtLrDDV-0GvwcYm4
z-U(WMSIR5jf|v%{YZ14SK|gIrx@g?0{C?@Uq8B-r7dap}aFRr}lrK*AOj1hnxC}o-
zT$SJ>7xr+6W&F(O?=vX>CSN(<d!FbY`E*$$VgbYV8Dfp|a@Q;5X-_ctn%MS^nhRWB
zegYdyYmkppJ$_k^oqApKO6U<`p_bpN4uWH}A*4caRkQOVb8NY<HOS(4vnM9r&CSoJ
zDK!NL#73^h={8tx{yXQ~_~(LUHRUe#XyE={u-$8gHASNR=tyk*t>K(e^UGS8QW+yz
zbb|C0w-9qp2is{-%g{m#6=1zH?<7@2kgdb0b2w0*q3@4gF0%eA<yS6pJ13S+8WzdV
zuQjNBH#cW^QWmEH0x#_K56x|viWR1e=0FJ)eYOOJRcI1rz>j`&piY6tPtN~#CGtmC
zqL|sZqmT=gp0k4-mo{4G+TsG46i91xj7WmGzYkn*s0juezP<+O<9Qw}|M@JnvAI7l
z-O@c&Ay#>kRXXL#ijR_&@&chQ&Y>tCJ~}sbK4*Xz@XxOQ%_nj=Uo?U)SrU!3crP*^
zjbJK<@jwG;k{yA6U9Wc5*vbqH&!AD|7eT)63;CqXS-!7XUS8{d>}F$$*$E)~p<sx{
z8rXA<_6ogu(sMQZ&XiBr1f9I;+k^5Id;$<4)`8J1Chr$4F1{>~R|!-`UwjYU=hU!8
zfxH?UZR;y$-Dj$hKo)}232DwO<m=hWqlDd4g>Z8B1nRxo+S)+9LH6L6XxI!)z7*_3
z2>zJ05f24y7)JGKgx9a4;t~=r<n^QQ#9o<dkvIZEP|sLE?*oPFne*sgJNIA+soYqk
zfj=e5^lVHn-6Qq7bl^yAhg=*K`+V;mw-ztcZ!L&tH8eB?5)NIri?z-nZEbDFZnICb
z_D}NAGAH(fcfp-oln5(G(aK>O^LTX(g?!+V1(1&TV$%4gUUPVc(Cyi*YvIjJY+YS}
zP4LYxIh_OPivS3rSx-Bx8TSLWiw0rW-D>diZ`epMQx{T8sG!WWn$%Ef*g`8$tAE3O
z)|fg))l=%FNfJz(Du8DaGv_+T0HL!xHkbYpGA*}THTO8En}&q4=<|OZkG16icilX;
zR|p#4E9Ayuk<Tyc&F?O_x1W!JK7Z#{iJYbQ7jeUFxu!V*{Go<bg725hd?ClHLa-Ka
z@|3uunHIUqw<QnSWE>{|n=@!Kw75t+;i~)s|LMR7Ii$!HbRaTR)C1^(40Rupq+=JN
z{;u{SuQekLTQpmFM_coQvor^HnntSM?z?*x^Vh`Zh=e`4=yz=!K0^mXVZp{GPjjPX
zFAjY!?pdS&Sr}d|QIo}+6%@T|lPV5oq^V_S8MRl+b0o&IDk(=p---rYOi@RQCYD83
z_Q};;8h?5DCNSjjg^g`_wBg&6Ybha2vXrH*M}2f!f;dumZi59y8_r!9q4IiUBN5Z6
z#k101t%R@$il4YxwMU~yk}97LD^=*1ta!ey1yY^ab?!x<4I8KGlz;Jk!Wg1yGjw-+
zi8L?;=ZiQOpvR}XNr9wdG>p@plFqF$2DJ(N0dy6b75(KiPJRxFWJ_FyxD_FP@Ds2j
zo?y231?-+wEIHY6Y~cmn$zKq6Ecf<~qf|43T6OC6xa4;-wh&}a@CZoTU>Cpt0S~d8
zM925=FI5>;nL&GSzIVlz0e$A9le=BktM1%AC&L^)SM@o@7H$YXd8e=>L{2VKEynhh
zP<I+8r{T0O*z5fbifcqv!AHNjO~6GmU!0dE*;V;ZFaSeBaQPAukG`K6yZ&M)pQ!QW
zX-3~8zIvR4FLp%=Ke|F4F)W2bVpPOCw=+;Cf6(&Mk@v-_1rTxmRpVE4M%ywVuTg?w
zgSY&{10(tT8qgO<<Mut1GQ|7z&TVp*4=4^`aEWz|;ZH%mm!$bzYxd~j5p5K_ECQ1J
zjJ-bDqo%cH269$~=q!9ZZ9Zdeq2_mOKSOg8IPCkYV%B&X7G{iIBHl4yEVb>c+@aSi
zfRe){B38YB+}#NV2VX20d%bDQ0x_j+IhKGm|2I%IU%>tWEM}=ale&p1&`8eo6QeP_
z3^nYc5g={66&sGu|F!QYw>|BbsYjhp<zqGUJe_;s%irglc41Q$weit5P8ENCE%1dr
zwBOs^--Ms8MI8R2*-t7JE?@B<N|j)&)GcCVAa3RikR>880S!1$EMDW}=jY$99^Epi
z4zS#X71zKQjIA8fwVhj)rsX)0Vps~L!D*Q8qb^lyL;TcSEAtkxh%6p$qNV+V_rFC4
zk*k5vqepk=eALe#{-xf_+yGcTL&Ly>+Qw5<>O+Kv9LsSg!OJI(Gf#cwz%H{8Nh#X$
zG-EHH2TeBc0@?YuVSDFh)y_b=-3^b-xJiFi3Bt=UBLt@KsAp_UbG%jL`R6*Vr<LN9
zgQRAZe&J(of_4YlNBQJB(oP5k<o?>#3!>nM$0u8YMbWMmz+|tOb!}<(!%7}mm~r~)
zM5bNDQm$n`I!#@O{<g$dAl>}BW!7GtRwOq`rY;CD@eH3tIf_|RBAYnX5<oM}zzU_R
zo~6a(IX=#Yc3y05@QJ6F!E?m|78><_|M_>t*OiC59nyVB0CX{+7$Xqs`Vp+aB=S0@
zA<V>Q3kbt#px_qEuT6F>X(&g0qn$kEx$K)+g79}{XjFKxEZ|aJeR8Le?W~rr6JWku
zxouvlc(jzROAJZC3lBNYa5wLwtea7ZY;fJ;;FQL&{q|jAr}9p~QXr841kt|bdq_&Q
zPt9$@tHpNgV<yCu*I$oKV;4Gz(+cPmNGsOsIf)ht=TAugy#!*z1G`Yw6^gPp!_L5>
z+>zY3LT_r@xyKXWxtrN+^R$y3LN6{Yxl?q?h2kSX`IU3h0q2Ch*Ry6xs$`4xl16}7
zaZ@JWVz0bI71I^5;I3Jq&44NWvMGvz6r>tEkNQA@;G0iXm^%d-f9>I6jI3Ua)2)R8
zJUaj9^PQ70=~B#-qiv)PT23EbiF|>a?>UqGha)z<;6a|PNv>CK;M`6Rjs*Us$|;|u
z(YH~?a}d?Cg^j^@kG%=d=1J6o@0iU<r6s_L!vZ{ibJD021E2GMiNE$IV*M=5Yk8mq
z1Ft|<!bZwAh4nN0<{g1WR>1W$DaH;MJMzT?{ZT3|z&vnZ!04k@yxds^|Hs9-KHCe~
z^SwW%0ZwJDy88MM8IZyaz11DcON@G$ezp7T1H#C7O?n+#3`*A7i1jUUqDfZGAmh?`
z(>PC~yNo$K_SQfrI~4Lw|K@`#^dM9qv~NX;_v(txx8c=qyUylh0rSA(gg*MQ<7nDs
zyAHa^+LP*kIjd6))zsA{L+>1U#>c@$!fMv2Np&pjn1h&&&b2PhmggN9=luw!^0BF0
zMZDeqtpv=qXo}~0nBa!smM!j^9x*<17)iBU&q-nL<GIta@RNsr_RzUS53F^v8vLxP
zm`oPP=T8$9jsuS(*I(E-Xl&$asURdMboI?#zD|2KxAhE<xkiq<Yf~G#7`#%A6d6O6
z`@mt~oJ&B7WcqXUe6d_D>cAZCc$k}eWE`bM!}Iueg?CmZ-z;sef5t|#S6HE9@3`;s
z`X1N_@<4SYtn=lYtIRasw{PWFnY-n&r;%(9n&z=sX4-O8k8QP6T4`Cvn9auD9#76V
z;}CXyzm%b2jaK<n66H`^a=AE*sjHvUn?A56K@hC_4}SfQGa;ATt-gyx31|A<Jy5!=
zK|Oc5C>8!eBIbW+2V$}d6OT5$3f&zQgEEj(w7p_Vom4#6tKkfytvtEcZ~7%t8!i*B
zukrWY_!`a(YY~AUc6>gAfwW07Yjc^r@f^#o;I%uC{;C+sg#kFan8`84N2M-04UZ$P
zly{7v>iUt8Rp`gC#sTNRpIKT+pM!4VTz0m+cge!mQMKywh0l&7@$9Hf;F^^2*qJe1
zKKV~;O~}88n8MjQVkzI+`)9#lJj$0Ctwk0)+SSMF+s4C~kp$VKKQfKp3AYmqD~xsj
zf?={HFURJ4W2a)Q?zXH-on9E4d0=gm1+vU)K4y9KRru9TfxW0LXn!KNQ<Sh@WX)Wy
zu~95n$3>D)`wLsCa4GE-cF}(aV^ij^Q_{6tlov_NJE?hvEE!^9D|Uerb{hkhgc<|Z
zz{4>PoI9dB&%#8skk4H=F$&%+C@4Ac1^(q#7ju8>XETI1|BJyk?&LT2B$SNNl_PTH
z0sP=`$JeKe1Wvz=-9DEGs-u+>I0|L&^asg-xV>M`^|hCE>w1pDeesk}YHK>o)Rxgc
zu7fj|FjqU(Y4x``mz9;Bs9*Ec=5Ey_<PS8zKyDlm(KuEEH*vA`DD32uVE`JD0HyS_
zziKZiBL=Wp-izK@#wzwADhP6zY&`{@U=;g8rd;3WLUQc`qGWoB3JD}@(~XK`hfqOe
zN8{6w);=T42S!!5)$2qB&eR+~#y8c!KJl^=%y1N5Fxgt3Q6$&<edF=9w3>!|uLDnP
zj~iVLOZDhdzo<q$vDqphm`(Tt<D=TW0R8Kb5G?977G|2R%Zq@NU>1(?FlxJ9Nfi&Q
zARvW!m^CkW%5^~`;*x@A-XGcjh9djmFG+I%?$NZ2fwT9ElfWNt2IMUKjG7K8L&zp-
zlRDAhPf7`v&{n6yrH3g+U*ExniG{@@=5hU7Lf&Lr7fDv#;jIO1-KVF%F-?N28fYF+
z5lIcLVNjK3be(D0{QhUL*G0}rG19GY(ea1#@P}|~yI?vyiD}~`Ia4vmg;4ni(<?@N
z=puEwH)Hntq{G>)+q;;%l-hPXbDoTo-)Zd>i?mdB4&p$LTD2Aqn%;uwss5nvj?ot%
z=HHunk^K1zjMnw`a$+iUiE6iIwEg|_I<Py`;QbRCl@y@QZ((kvxY|*eY8-R(9|ra#
zVdy1(4#O|^Z<ps6_e6XWsfd+Spwdyq|A-Jv`V^JLYpVNmlZR8+oOq1?iN`0yG%gW{
zT*Zb|vt~)nyRY9>8`n6H;+z3P85zGw4fAht!ygnj{=qPUX|uvkJ@6YnxRy|P({O9r
z;F;e)xm3fJK3};X)X5}*+ySJW{8n~-#)=JMUoxKiZ9n!$EH_*5U(X3bn(J}kAyl{J
zc5ZH2vQmd-;Kw~;<}V1J=RJGDO6~i`C8Ev{1<wgcKLneQeh&EwAOI<g=qkt$&S?RO
zirAzc;(HX@KfH*qsh))B!@JddQYK&`*SCXha}Xh;@*DtpNI2VsMjJJLsu078WCYFp
zUbZmHepS}mcXWHWw)r<Yk()zAm43XdYxrs&1^00A>Ri;<f6m2Iq+i$YO=I(^H9)#j
zS1B-lFOc|o9|+KEPk`a$Pprmn@xa_t?*BPR?#<YQZ480}ybdYgRG?I0_ni{EqVk_(
z5m!oRdf-G5vD)dHHC;DbyJO)4qI;?&@LVd;139V(y*}%@HJWt)8J#eW!bxJD_&jI6
z84&(JEFFB0bxI)GP8pzWHfRY;5l_$ZPqdfx?>o>J5D{|q*F?ehErp%J3J|^8D6@RN
zWmxk?8Um<i2W&3pOEFSCm{*2-@^`_-O+9U<!t;21Cj$w!Y*e};pwMyzpCUm>bPWZF
z>HH_tB&4N#)az~~ytoC(%*&k(f^vYL4mc9Cl+d3u!!uz|?%aO;FY{9z^7{!LOVXpV
zCUq=^7J+B=*#7sYuMq{k%KKvw_dnO!S)O+FXQiQd^bl@_qk5iXEDe(;L(O<DiZH)u
zqqR$xt1L|UjLrIl`g2Ii%Q7c6BL9ZxhJ42RLCU=A&$)|N?mGX@vI<&u{f$_T{;?&~
zGMX6(;?R9t7p>`$_%IQE9lljzWQEJ}#&DvgHJhuCb=6g`tS)eNdC56h%(0Z4wW^W%
z$(D%{z{17D60IMN=Fq@~WjhltcEmFvZEMa%O(HnT0#7a=N-_Dto{E+o^SYvkhe=Jt
zYSf7mq$CE~V=>>v?Iu<m<^RC12<ql)fB$#|ka5kqbn#9;JQqrFu<VLZQA+09)c5i_
z9@}5zf(26M>%7GKrqgVaF!45BFST#g!l;&n%Hr?%e^x>Nt{I3fj^5U*URrK-Jyp%n
zt^}M1oXF=&Uz;ICYY)K>Pldv_^Q(TmGL9r?G!!sDCKi)IaJQfT+;}I=_B7%{(_xAa
z#(-4ixhC)X`q3zdY^7{1LRya}rgR(9oHeN8KKijaQO`#Lc80|TSL=XK)}@5@B+7x)
zwIoFbC^-7w%PDrp;K?aZ#m>G+OA&34+xU5n5q3f>U&xu(nwj5SU#NpBpI;G1)w)!)
zkQZ{oV;F{GFBrgl&kaed{<Gq~E+dM?4L0CqQ8XCJAhjVJ^;IS1vL-A#mG;>aRk%Iq
zMUGotafVToS3Hr+{Ho^L4G&=qjvi|A<<~zx2`z3iRl!pfT%k`>N~e2`11J6YuRm>m
zkpyYmKV~N^6hlt1xu-GQcehIlzWC4}D6$rgeuXHI)irKZ>bungwF5}A!=qC?aJWN$
zPVfDT^h27TU%)6yttpd!_ume_fsM|&TcU1gcq||V(XPi*P+=K}77rPh>5bXlY&nGO
zAKf<wZu6}T27iFa$QNFvf6_n3y9Ml=D)@?=(RPAbs5WcCL~5{%6XUxUp~2hgom*y{
zS`xrYI6@P96b9c}hC~FNRquw_Pzyra=i4AxZfhKeMXK`U80hZj@$!~(FyrLa{T&Pu
z+BylL<g$KNbSeY6##gwkSNOBt81h&FUm+HeOJ{nh?cVIe!}T2<9F<jLwgjxUlo?2k
z#b~?Q{;iGOsfOQuphy4p+ca7SbII5orvgeIu1uwc5BEAbZVFdgmJ9BiPx86fDFqv=
z)+Bmg@)Y7C1x*X%Sj7zSWvqk?6phMnH9s2Q@5=B=*VV6^Ot=Qa*Si@w-{uf0{5@*v
z1V^fCC?Vfv5xG)iR{qpaMu^WhTCw3hZqq4JRVOa|0(TT^$_7_T8n2yS4Lq!b4<|N;
zFkB4s&!pv6?sW#al8<mWOmI`(I{S<gY}+GVA0|YPm6EnvVB5GIB!tnr#lL-(PFwd#
zpx*7jqQUYL@>&T^e7V*tIRWmKr9@($=)_=Q$}@7?{f?EVhQ`>m8E9-se~kLV8+>HY
zKYya%S6BJJBn|O&3%oCI&>Z4gi~{R|M5rO&p$MN`iWdQpaW;7LsUmObFir`Vo<^!q
zK<cP(lCbd3p-*6V&P%0aN*B+kd=md{m?+8R`=AjX5!}|`8+LV->_KfOF>LSiUpQQd
zg3W-3d#2O9`PCnM{>ZkUtGraACiY%1hV0y4RZS`ZdlbCBb|#ijZ?hwX+=n|lKuF3n
zfU+`MyamC={li%+uiSnVNO179Ll>{6uBQSidCYMrvMc*<kG=H+7oTw=@i0-(J~azc
z^icc@Ao_lj*j`=V_(sN$v?RreNQG3R8hl~`-zEhUm#fvcYRnpu1mvCr;kXfiw^|q)
z6dQbr#oMNnM-%ar!AYkQI2fN^_#cs~KB6<@difnbmVTA{7PxB#e`2~_4gcw86&@*E
z_aEFZc<ZBmA#RPmAi;12!o=1yn!x)!lbBk&1cGixPR)@xWwW<dvBLXBbbroJ*GlM1
zrxd4N&BT|y_{%SbBtW_JEtiGwe~|4BOk4!Q{I`UtRZ<M{7Yw)+&dyO%Sg~v{&fJIf
z(fc|TANGY}J^H%HJkB4sW&7S~2D<AzcDN#&a&@jjlFl#aCnz2KJ~6||?G|<P@YwR^
znhsmhmmhPw4rE4ol%G3-3Bd&jEtHaGId87pw^l8<Vqyd&FO9=0>1t*gmOrF})+=h_
z5uGa^l$zeX&GCA>WL)x|HzWZ>MZDkDts-$;fWRFGt(66Q4_8Kq?fjE2BnS1jYZ*Uf
znwI&In*p$r%+OL{7sP>QZeAoRvv#?F8*sy5%easr^Yl1`sZLR7-yFRajhI3Hx<Xql
zDfA+k8q2Vm>tiK(Rxjn9&KN5nk84w%yoasd3$585{E~DEJGSucUSIyFpN?wCSu7nX
z#QN2*>A0S3hN$^8q^#EQ)EX@j6k-sXe-c=1Ob|1`asueOtMap|g;^O~KR1#ry?(3%
zxq#!en~Q)^Or~R*X5&mBvC=Jb`I>??X(}Es#p>lb@%Q76`R})jQd7gMgvNX#nG0MM
zoeR0aMW@SCyjtjkxKf`0A9@=?m_U+#`O=z}9)0h3a-aHV)iL5NmqS~k!yRnv;{FTL
za~peiusK6x0!N4vc}yA)NLLPaXhE#cC&D?9f6!7-{QI0R*QfUlm&}b<=-}8KsIu2e
zt#xr|D#`qxlbkKN0Cb>qv5P;CGAFSnlK@5D`Yf8trZ(T+Fb(>6I74LDx4&tL3-9sX
zz8Msr&Mw9|z}Rh>(~XeV$Yu^z?42LL0QH8o@Q(27VdYj(WB6ug0qmAH$LbEG|3pQ?
zr>Rs<pBqY$5Zq|qbkqz_yw91&!Y@gf&z8MsgG*GmvFCF1Ghjd{8b}`b$Nf19wk-Hi
zaude&5djKo%;2)}hq-b56i?sZG=b2cF_lw8p7O;Ne|v944(cUzX7{VIW_4}C4A8~H
z%R`O;vLfhk{g2A;MNU)ouVJa9?n*_2NjHx((Nm^Yj|HY6{nACEI~TjDIG*iwV4QCG
zFp^Ta;`OF+{@gd!QkRSKy%VE}MzGtl`IABkWA|?<qC#fPu#LLIeGbf(tykiQAe*TT
z09xVAKs5$6_D}~`V$>I-5txEHYs7o|#gJ$qf1*ninhR5qwKik=x=VAK4k>7UNU;m+
zSbf=QR0P7Xff_y12{ab>g>zs3OsCmg{SSFj)!=0<VGrzHbii8;yF@|upaHRT#YQP7
zd1`-rpeQ@PZkO@fbr9^m_3L5=%sWxzQF1zHVv!umyJ`@U(YmLrjcLF<#1f^O6oOdJ
z21&)bb0-TXwYX=$KR)Do@b{Ca-rxZ`iO=ek(Vhdc)c<IbVz*SL`#y&!A?Tj<oS%95
zoc3$!pl!CRcbD;Yp>HXTt)^k4Ox2j%M}-5G`{p5!{+}OOuWXX6c~d}XQIzpmspNO-
zXOf1naRc!F+AT+Yy`Q0F@Uz`u>cNnYW<M<0HTmiM)cm!F*6Q=8)P;Uv?qKLa{Evh>
zwill3u{5pB{qCI`NiEWBRli{0!$|9_JseWcTHghAzcR|;7E6j@i_p||npDg$Xvt}O
zzYCjRkn<bOdd}RZ-?mp;#ebM$WO00Lo5MQzezKQ;I!(fqiUXgUvyI$3!7jGntqdSI
z8<~XWnbP7nHT)<M#DdkEe2;O^8}|(Ba(i|){ptAA$_yEi%abwZTlXItQ=1Xgq#AvS
zKFVeF91MuxcePeizm4q<_!?OTK1>&Oe^($tZC|>h@K2b*Yid=`q`M;ZMf6Cljg~{{
zQ=JT4UrV;l|4<@Vj_OeZilSt)bZI+K;peZ=`>NFDm>DXoKmFT{Uwc9sLzrr3WWZ?i
zwl|{fU#h<M{*DL8v56K%inLGPH$(gSH_^+to_g>kW<(MJs#Wm*Yf>4wf*F4xr4$b!
zgLT6-I->toi?rtnGc^g|6p*p8y6jS;;2HN(!3yE|i`Tq#RuBpr70gB3F=EUvR|sr*
zFQx^bB?WKdG7HRgMjD8Ok-Tw{6oENP7wFfwJJ;9QCI9yrgxtA!kuJ?y!~t9a%5#LN
ziACnT_xiyFotLz3W=?u~TcD>uk&X>fr6H7ZgUcVbY*2vYL6FpDkK>rja{PpPL=Gtd
zDT4VMKr3)jXq)>jT>$gVHTVH+tIjSHb7l5x-pJ%(%<U()xp1rye3`Kfym^xXvr?&n
zGUHWA_R$VFZ$TUDTAQ~+rSighUy*VR`g9D2d}z*lOI3tk9-22^lTRgl<z5sfh$^lD
z1Ev;*v1NGf2&4oIvu<FrN1z?<|M?#7K3XD_=2JDutjG|8j@*e{JS^^p+_f6%WySg(
zj)lot6ejAUlVfHeeO(KdV2|gm9A;WO7aSV&V)>kdE!68Agd?vAW8R{R(aX_B#!K?C
z_-ykpHh6Z-|8%`C9$S{L1KXLu7|=%xGGk`^1Or0~73Sb+hR;Pe=`UbAx+fJ&7BUZs
zZvOekKZ$P4Z)TXt*1C<zEx?RKWfpti^cnRbiIg<9>+hpBR5fKDytijxM4`*%nAQRm
z`2MpZznh6j8`I2x3-tNx|4j5=y5ZcsAPSKEcXXyPoEMlmvqpDnq;T&xJOfRxe+Cfd
z%;NP~ax5=FT!KA#ACJx*+vKn;zhMHa{@9+p*yg!_U8cQX$l8<Btw?N-;_yz)oTE+T
zP!oRrY;o}_(rUkef4pfZTyqpRYGeA37V*y44@EJ@S?_+?_tOwb*M2c5H~Ur?744Sb
zY3<)TfUvri$dK)9&?s4&wEo5uVM_(4>g*g!mQI1C;qOAYkpmdH%=eu<y7&RTksn+;
zC*~Gf)OUSuhCLiNp@DfB6a1P6Z&+DrDgh*RBEQtPdBc+sVvipfR)j@ZRtA;TV69=|
z_p~y2mhZcnmFE#qrV_yxf2M6@rmUVwKz*m4az^4%b4qKPSs#C7m%SymugRe+hv~9N
z5Wr0Cl;TMw6|-ej<4J5(*gbwD-kmv9IzS^)!r2#`28-G4esy?1o1f%$q;x167+p~n
zX6dX7IfOcBZ1z(G=dEe6a41oLjDX<a*`wofr7K3><Lb6N_aaa-m9p=-tbO&cF%u7K
z%`8Pt7NullWcRD2H1Q>m%=9hO&8z25LOt;6M4{(%`QPt3M#OTIUJRO3t^vEht&5t)
z{ld^4sL+isE(fUYku6=^C3<ZOrt%4p4_A+^!lz5#i2o7=p9q5=&jd3Z-p&fqtn~zE
zUYVnd5x1C~<t!g>8`&|w-1gscV4UsfziUfe@nY`<`mUAr61m+!la1f4y!b}(1gz2^
zP-M0Kx57x;Z<iw?DD)8bMbXs&C?~Z7qb63&Fd!$J4ZCa=4gevgU&qe4D&0K?75`=Y
zdSm1z3P}T&5064LtdoUU0jX73TLv=HFAX7XoV*&Yp1D+vK2*%%mte-qta$@mVryF!
zwLrKQX$h*G(faAyZgN7v3p%w0T?RYmAwGrUxX^Jsze9PU8dA9`u+cL+6Evp56lj2_
zZdbd_hZM;MW1X#zU{?>Bs(kRtjnilT#;juP`x2eexqO>ge5$m6LDJeWF85kYS>@%F
zlp$UkbrbE|%=hHdLdg`Qd=}5B0Milke;H_Y7^enVy0{{Bq-u#rQd^!YNpmPlZVdC3
z_t_jmUhOmMsjuetD$$JZ;C$~DEtOS|CQt)p&=p&|r<}5y@|jWdHAC=C6}%5qmYQu=
z5M#7;+^#dDr7u~&$5M<Nwky-lE4k8W$guDuo~8Jmn6zl8N6%$Ri0}-9*3qVH$|>YB
z`j^ntu3m2jt#!HBLp(1Ln<C%3M#Jyk>+Gv)?QiTNg-LZg)Jg=qe(`)3)3{h<IK025
zGIwoDJCN+}U?*?S6B28Gg6S`9A?0HL+po_F)^y}AlT?b$e--e7*CW)OXx?SnHC@^a
z4#&U6E5YNFa9!d$R(?sMJBi=X{TN{-J&-$2<<j6Kx&n)D%VVyYard%4X#xd{gZC+L
z40Gd^eZIZfpnL-UFJ4@S3?rl0HQXJ>!`vP>nqQIjgmUeJ2+gqcQJo}h1))A|Ccw&R
z|2)PqfKZQw2KxZY0ZKI)oTJ00(2*<!Wl3@q7I;o9O!ZG5HgG5urt-_Z#Lx&I6%-Gd
zqI_d4{Du3&AAKCamce($7L8T%Gra(SQ&#e5&}WoCRdw1q$<!PlV#Lvlo=q00YYS9-
zIccu=lquqaCh2A@3!T}1fpl=*^&Tigmh77Zwq7J^jM9+uf9TlppeTpGH8z%<JI~?u
z6quKedYAlTX*bh<hABr*=O}QC9nzu!ZtMuOfL|!rtkOwui@(YD)=;!~UPKIoJ4PmA
z)%gB?gXfo;Y}I<@uKA=uZoBEB`XNi9Lf(}RNR9iHovWB51(^9|Z%<E>G{-0Im=lbb
z7Kcrb2jZH#W4J<;4Om9?BX$^<xZ0#4hy2N{=~wB9=-AmS5jZ4HCUw*tQ0pY5QtgVA
zD4WVLxVcMZpojEaFOOf+0axT3WlMo~WgsbH(O&=9F`+e9*H)nP2ZYD%(Q&;(GjfxH
zPkT_g*KwOs>{?Bbgd*)Y(?i-Cz38mu*GZDVOH@`~B~htoA*BoOv#>Q|yZwJWd$n%b
zhe=a&eji=y+&}&ce+?4qLS%3Hc^)*r{-3YwZl(7@w6UwHANS#iMpofkrf6>HRYz5(
z153LMC9y*D-}`?ShB}2McgaR+yg!^16-&!;wTd`0s197}xIynSJK2~jLPYI$4vD3g
z54^nti*+tOgf)&hi?WW2>U^UJKWewC&@X%*{f;EtutgChzP0ik#3a1yOa1R$q`IQD
zL#Ne{t!S9uzmt7&dQmQw$#8VK9QvPNnVvv!ueVZbX%&+_m8Y|wsc+lmx6G-H{^o3R
zUw=JuN-dur!6vMY!8Fx~Nh|uPZCLQ67VKvT>Cln4osD~D!{X3Ikvq`NE=ETu`uX1v
zB`qmj$+%{-qTwt2GjrA_BxDn-Lh|LX)`>kSOO?wK@GsN&5MaZHl%cOYby>1fNfM+d
zsLzRZ4=!Z>{+&6XpWMHo(H*ge43AQzuHD}g{Mu`f&ooFktksZz{N?FldHJ2ZOL!6~
z<Nnf;IPJ4^X~1<xf{r;y@kD4A@@t{t1R=y??bEYZ$}M=s0v5@hhh~h9J(DGc!92?s
z9*T&eUllUk$RT@Zsnu7l0IR>hDB%awy;uKo!*8S-ytmy44~vsvm>DZ(E)mT)*omDN
zb@hFw#IlIr;TSgB{QSD^dwkY9dWVLtt248gK<ZozE)cQ6Z!<|Zbws#Be>ii%FnvS}
zg%58*5b<pr?L9YgXR2A;RU21~^;wc2oLpFGxue_=oHRtidVfoUww}@@Z<^_u#<z;@
zZRkEfUxKT`Pjkxrl&=9|O@KBhY~frm^g^g#FlU$*^;3Uve&bi3JvqEjLQP8LZD6wL
z;ThjxGkp9d<l=MCd(_(9I@nysX#J2$7X{xg0G>+_Lo9@21|qI%<AmRDv&iFO910`t
zX}Ji4ErMc0I|?7Sw_V_Be*uJngpKyz3%Rony9POx+Lib4w?_Fhgj_o=Nk>oIi{uo`
zX4y6LuOnIF{XxVES&(Y5$*-6YQdG2&N@lZhs<meUf(p(9RWL>C=3ljkZQ>gtqW<3R
zLS@6R2(~V(5Zhl7YBadLyuSBmMkG^}yK=$b5cSp|E$VZ%f5bPwAY3g_F*zvbb}pc}
zvbhf9Pf&<gE@`a~9sJEcK#bL9s59-}sy>}`58OVz3R#nS><f{-_Goxv+0dJ?n<<bK
z_dC~H>B4*=IPiE$hg8p<Xn!K?Heca$OK|Gn&3wG|v80`ikQY*U%a4&kzir@W3O8WO
zRaer&?znf441o~-ArK>wf|Nhut}(qCn^kU+#|?e~7zBq9^>pT)_aaPI576$&(BVxH
zm!b#?b5L`CyOn}U3(cFSwb;BbaT2O_KaD(24^+xVzTCb%ORF*&pdGkh2_{r!vwhrX
zd54qxNKrHQ%l#jb1nx&m?}Nx6owWrHQrU_RfB!quAH1<~M48*O-O|oKxf3*s|5-cx
zKV$%9Z(`&7W@&xBZ<QwB(a|JeyH}ctvFsyX-3{bN1MiY^;&^33Zk}BS2M;ot=R3WG
z32SgAERzKqru(Zd-;WoQd5Bk8d{27DT4!!+X)Z=AoKplIp%aji-tj@7do)<xM7hl?
z*q{*tM4b%tN_<*<{RB=xNzG?g*I*EL8~20Yr2jzy{BXst=8tlVK>E;_+S`{aV?7n9
zd6TE|wJOg&bZEKRB%cD{liLJ%+*dIy-3hnJinX=5b%VVE);!N0r1I@E__ML`xt-bd
zW9QbNHy{>>N7zKkO0fyL#B^T0o5IhEJ01xiK|+R(T7{n}(kC5Pl11#sslRR(Bl47d
zMX%<&*O6wqzyl=pBS<+#<!!@mL5wD{Jtp}o>bk<Zl6Gokz{5qeVKEy&m$VR6ADQ*B
zKmB+&YRfbdM4s}CXK8cS)h|f~X>%{y(7xKOBuQqMv}QBuzOe#{Qzw$ngG~ytKgt$8
zHhX-oa2xf@w&*X6;Ln%TLU`)*!6`Kuj96yJe{x`$bJrcpDBp$L97$bhJeWH0rwPu3
zrF(B{XSG{e*pw9XwZ8qHop`YLJLYAK-R&-RxEyN|m$3^SS3V~`TgQrgw<6Pexrt0D
z4Pl}*k?x=#8jcQs!E2lCI{b#p7&-S5Ev5jw9QcHAI&1ojw;3hVa;8Lz+~+>iQe(Px
zB<Og_`+`L+z-%AuU_rVpr%x^;UZe+8UkOF}BHqGd|At?RR1e9)z*KLLcAi~vFco@y
z(|Vo$+efKP3bqg4IsWaDXb5&mYI+2pQ%%?19zH>Zko6%4Tr3Yl&jF7O^%kvjw^R(J
z!LWag3Ce|MJbp-oj)<X1r46%B#lc0k4VH`;7{W}TOO}}AbO=-|*@{@vOj*84M9yCk
z<mM|s`nD90U|L3>%-IpS-uMPH4C{)7?_c8+Y}&!!61DN9eT-E&`-WgUQb+3zHqFm~
z0>n(wD5)99d;JGfIE5O>$f^n)A9m5Ljy}DZy8_5}6sIa(@q?BP@7HyEMh1eON#xmb
zRKJZ<%!Ko5*8mAS@!$7XjSIIzV{xaftnt5|zETU0vMJ?PB;WV05*84C!kvG|_4-i$
z`#>_%^ElD!j;s~c!x#}lm=r%MPWaOiLx{^pKU0w=km28=E<5~#Rg@5`nvHK;`)ReU
z>)ET06`~cJ)ETs8rFjGCxb$d_*#|+#cb~25o_22v1;m#u@G>9@TOlK&CxTiDZ0OFT
zt5>hSvXFq0Y~$Y?L&df2)HUKZ%0C;b#((!av|5wx79oi_Oc7Tk#?sk91(7496c7UZ
znRpQ_xn~qQy%&_S6F7>&pWV10aS$qT#7|kss#AM1y+2+`m43%<CshC87ekp6oHd!V
zpM3BdwI|3u{&E;6wo~A6wR3<=N`AdeSK&`nGtwUPdQa!yYw@g6@AST}?p;&ZOFM7j
zSsF4tnHA#T59+D+qm`IJ{b_f}-Sf_WYOk`TtQ>57R01yWQTd)Sb_fg<sk0_z8>W5P
zqANS!lcLO)RNko({B(~wUxP%;HcQxOBv%&hCOUQg7;9qrKZ|*`Gpy3r$-N@)c@rH<
zy_Nk#lg|RIoEm3;c%AEr=5#=fQW*>h+lggv|Gb?e;3(LB_w67e*Il*M>+{D?`~{#h
zPX0(~(W-;i@8Bb%;^j-IS@GIdQ%|Grfsb78M^2o(C%87W+MI8398>i@AI<B0tQL-{
zCU$W7nEYBkiz{8E-apZ@S1VlHb{8%|Noj=o<j3NOTW<#+nt@-r!ZS`hFwS?kc`AGV
z9vRM`?Dz@A+1WmA)^@Gia$ii5vD$f-E~rxOC|^5UC>AcGmp`_FWh2bq(o>TwMR!>*
z1c)MO1H|&{2F+1-^eC)HC@^#+V&FwRCe;goNSYPDaibS5O3Jvl4K*42Y~>@Ooe_UQ
z4%V`<Cgo)Y*)onp9rxKWV;g(nl&Fz*Jg!_WiGY|AKS7rC7M)QV@V*S3tT119|JFf7
z*Eh^Hxiz4nWi2EHG1~%a3=NDtv%R_ksm%v!MW>u`z{`GqZd=<xbR(~}hPjrnZ@)Mm
zCda3<TzvLh<y*o#2`MB<A_E~g;Jq+JvBi=GmZXrqeD5F_$_0|@O+ZAS4@x6$>2_}2
zJoch^gtb2kKhvZ_d_6_D(cf;cX_usXK6@PRc|bAx?3<F}UCWKVB(mvpE?*f&>dBe!
z?weuWDJ=Y#O^b{$QYZB1itNSZf>8JvGW<^K)*650`a={66bH(k87KT&y|RUVj#wny
zL*Jj0ED<=(HLbY`W^_dIlB(w2{^7h(&3OLGYi!?*N|-RC%va-7P_N~`6lu+3QQ79|
zvceZ2ASTf}TFDfF9kU8$Cw}>!)+mkKY40Oqc3b7kN=z6YitE)25S|0)(3kk*XUs*j
z$lYS95oa-tK%Ran0WvmHyrnO`C;Y4XM%8!YTu?prkdXY6|4JUr?RZ^6@6&!=)4r36
zhEqOY_$y~ot?ueND1uU&Y`B2kT)ntIF;duUXL?9kZ%Sp{E~SCXVZHv3ZMf|jz0~W@
zB}MD?Owr$8lK0Cc1RPlT>2{y(NYVk1+~TaJX;8O(_r)3UkDF~K&jnH1Qz4xgnSyQh
z)mMSF6mhAhx4fVvOx7dpo73iDMn106?~@}<IjAAE#c!ZX`m9haM_56WMh%%x-l25j
z;J)QCt~87_;KywtJ-#(N{woapSnS^!R_e<H8jOAJKi0(n#!seZHX>P^2#V(m%LTB>
zB7W-l*^f3mk=C8F0PHmQ7Snr6#<Y4H=h^s*J{FT_a2K+(Nf<NOrcKYK5-$9bJ8|ee
z?Yw{UfkD6d2OH94T;Cvz>&Qp1yyBR6c(p2hRt&z1PT&{L^1pn%;apwRJmYz+D~6^)
zsa0o~uyZskTAveG+_=*I&erHLVEz|0*bMdBvZG+zp0f55S^7D?fGn(7^3#K9%a4y+
zd**UBvk242o({EWH&}f+-=_hegrG8@=jc@R`g&K?LIylbnAEGq4j%p4Vg?Sp-TW|!
zTO6j|xW<w_8sh!vi?pp^67{&G_cD2R7}Ki-{_Gj)ISo|5aD3OJLc>(WQj80GHGg5B
z&$1H~Id`}Tday+y_ikNKdl2}u$K&G79vS^2C?7dYlWg99<$hs6dm~a*p#k&1@dOCz
z-AufOdk2++&))m4y7-Y5J`|(^2)W96)nOa`)cZ`t>v)fRX`O3zbcHfHyU1L81|rQk
z2^{-`xP)CSVYlyq9z6=#aL_Vwma|{r1kD#Kq#}sQ5SoFUTJ^G{ti0alD4S+J>epz#
zT@9c<Y^7ccS0djhW`Cp6+0f&Y9T7<X{&I<UlwT;_s*2U%>0>enQ;d*f?)h?)NS-F+
za_AtJRM-eI{7m;!?>{9Q6VBT4@u1UJ<SbDs>aYgpY?j#yiYS++fbX#$6;;_e>A3oI
zPw=r14D-7qU*MDhJeUj~{x{qLo6?fJqxD>kc6XRr_gLH((03Eip>vLq9xo4hA%ujZ
z<<>DLU4JcS;i0%tJIUm7R{5!MmEhO4{XJaf>h`n~4PCmGrsqg0))pSh7#n80J~&aI
zaH@V!?~z2U!jJ#NvMNqdbV<arghz6mM1GOuf1}G(>V+wO-bbOhq%@59w)n(QSPty)
z=N&BqBJKrAkBj6Mfzv8{#7q8CV&H@%s{geNzY$e#(u-T>Nm~Pv>?aA(=rPuQ@^2sk
z{(%zxy1hAYg;l9&_BS(zb0Q!mt){G>&LfyUA!NHCX(;+^0~-t(yxn0e&>qmN5IzfG
z{^#5VOsh$FYQ85jY{Pv(w6>Ygh;rwW(PqPftHM#f_|y0lPUyKzsk0>Z@}$81-VQp{
z4%MRL@wm=CC3@a!p!SS<1)@y9gChULAH@Z|82;wD%-U+q(rc_z_4V9`)G@uE6hq|9
zLoxFSiI@{)3cr5Tdvh7u5y{A`QE8D~^KQ88`MqzA6}P?(G$@%Ti|^mHMaP}smnIW#
zga7z=cqyawlxtTUAO3v!gnq7}(eZ{Qqt>})vJ$#)*pu<RO~|oG{7fPCieCxYoP)FT
zY_X1L2H3B&o?NeI3KPZKc~}dgTbO0Wh~PyG)!om~$PeV<|Doa$VltGglknJLg@XNC
z82pB@Uf1hWjUl%|b~@fyM|nO5VM5?F=^^{T!P~pku(G%1+!G-_pH2dkj&9-+z|%yK
z608@I{=Jbb?z}_zSQdO7!f~Zg|JsTBK6y4TZT49z(eNOThxaI<xvRunI?POeVy+en
zET782Ydy#D5bG#E11XB0ZwP-~C?r>gVHXElvE`U<C*Pv6a&gR5+^%m;nVo(m14X(%
z<fEn4B;7yZNw{A2y_^zB%<}{P{JQ~Vv}QWONZrW^uG@o5^}U~k)(TC-uCWTaF2;pl
zPjvYgK$4PxYcX=?)<O7TF&`9_LK07sG>ugmcFEAu%Bkq}`4x2Zm2jdC&s%FYDE+VB
zomVs!1XO-n-&{TjqZ8vjV<mHS<Xk%i_G10+x9_Y-hl65DSsgKO93vqs;8>M%ZddEz
zROJ046U(h%K5b$+Hy3X=deB+?Nf*t%HE!I-T%aRU#c;ec%7(M~JiCxJYm`(y(zFa2
z9r9>o&Sq16=%=wzpFZbW{8p4;YN<@i#LQm<L5{)Zdx>y$$201dndRm_cuU6XKQ|xx
zn;B}f#<g!Acb$K1s7P$hU5{Vy)kOa;6P~8)wZ~NMduwNcTx0LQ%*(ycrKc&_Bjr2l
z(P&>~b@u%xBB;641izPirC_`omb_TlnJHw64NG{b<}k#s>6C{@$XTRz_j&XXAy~Y8
zOX4Nk|Lx!C!7WUzuAU$Kuo~KRVq7_xWebYO!_wf*u;JN$&wT;;+BwwMC(k1lX~V4q
zaG+hAfK>kvI=ei|wtQP!rTQs0Bkl~JgRM+X^qID(4PS=Sy?O_iP_8?+Z`45?OkGl#
zeHoGfV<9x)*QhkttYF=*lx(us?(Hc{bY?5ep0c@Cz9lu_j~`HfNch%QL4t*X_sV!d
z>&ID59<UFmMYX(oRofSa8znNxIKYL8G+>UCbJt%-s}&*ZSO)Wx1cg_Gm};p~J-3$b
z?tzJ=Vqy9-k=Gv`mI`{CG%CzdCj)BQa=NnLwW1Z0q1P;)pjQ&x`Z>5kf33PxJY=|S
zS=TCOUBGV?ti@^AqT;u|2Y%dYTII;#%RrV-N|KF0$N!I_t8i%QYr_W8(%oGmF_4mw
zk_G{Pgn%@NG^0ie(%p<0NQ=@)jP4vmKtQ@T8ew#P_xlTW&pl`7zVGusA!wPmxF`vh
zv<}*DVtKPoJt_Ssu_M#YD(4;lCSDPz*mz!v3!NuXBW}^<-w18N%lewo7*5dStBhbH
z6wtY-uNq_;`=MX8#8b()pAmvl2+HMq*0a|>#?5YzfBY1*IgB3^{rs}A$*0+U&pKkm
zGNphdSFTg7p{}bJXWW`;tyF{`F7OJ{737@hcSkbl{a9`L2Dtyu9dQWTg?of*88g*b
z%3lHbowgnh7mgy-D^R%XKNQB8ike!o9<H`T;z^v(;pp%2MI-xHo}IwMDk6XK>SSbg
z60&h9GeRv8RDk5>_PpoqJ$^Jr@wag-9FxPJ7Ax)d4;qaol_|oAPNyl}{Y&aaJ&w0y
zT=0Cy$8}ovtCf+6J~A6FRLj6U0f}_#P!dQT$Q41q5DqRtCO%|b(cfK3iIb$oJX@DQ
z?azV#XnsYmr7EH8zpr4t?m^JtN-ajV=oz(t|Akqb@SC<HcAJ`LA2wW*3b5($0;e>w
zMxttz;z;p6h#yS^yC!a(jrJw<ao`DOOHnEp`+A~Tmp_fT9N>+)d^HtNVTjC?^v~|R
z#K>&{V}e&Od1kpk<!aC3B3Q*;<ksh>4SA_z`75hJE!~a|Wr$n@6{;6Fp_*MJHUyB#
zaE7Wiwm4`Gans))50opxM>5NS=;t6>YWBG}%TGm=B^IdvqJFbF5R#8d>`KI?Ja^6w
z^BuEp{miPFRuHRw<`l#6qzEg)q5hZA*28J=WwkJ2jtCkFKEmVx*}T2M<Dzk%3%*wQ
zg*urm>}sOVqe9<w;`Y!`4$fO~I+Q+AV+c1VIKpI&YfHqu8~B@`cy*$jadCbebo@Fr
z+Ql)EoZE^0%9US{Oq259b>eogWBykBsA|E&78((*wjH@Iy<ECNTj$fP_;ZQ{re;kO
zLqIJKIcPA1`(F^<t&zz6*>PgE5c&s)+z1<1nHIGe6|}U)y^nz%n9U#D5MBpz6_M(x
zRiwqHdb?>-hixKzUotV?zx@Ae_!xXZsa_Ir0#7}1+&+4d`tVvqn4q^$laG0#yNKhz
zA~@o0u>)#8MTIxZiM-*<vnLC6MiuWbM>{V?G^Z@@PHCBBntNAnrymfaxv<RyK7IR%
zzeB{hih=9;!$*p^gg`~5(liEyF}CD!40tk*WD3@J`hF3oSxr0xzp5_E6*0t``f_;p
z61(AP#<;2WB+XzJ{blNx_5&He1J;|!R?*kb^&F^2L2%D2e)M?$*73$5p^EkG8&{|2
zkH2=O(Kj)JkJoK8GBb_78$=gx{`;ogxneKDu=+eg$vvt@Q**FKn2{sKKd})KaCs4M
z)PB)?KSIezP+~0`Jr3ellf<9;Q+B@T_!$)=ktO9}MN}sAE?Rd74~5S^YFF@nT}AHp
zafAbS_mnnIx@B<X!HZhX7|~Fcb$?X6PLLH_j8ylh?~n!zvs(F0JGv&j9|P!i*M&An
zt{ekJwhx}JNLuWiQuc{RiQ{rT_BUxgl~O^7I8gV3nmu>;sRhkeCwe0ZRf@`>jxh5M
z&+w8GzVuzP<o68w%)j@}Xo;?Udr5T}EdO$4IBr0R@cWlk7NmdNF|Iru`kMxy@evU8
z3))l#5u?__5z0Mn2%Y<F6)?<IrfqQLF7hF9=1Lzi{h*G}*}V-Y#IW8={9bu@CUq20
zjto3c+-aLGPgR#RY8)=nu$T04Nv*7W`F43q>H-egBc&EJwX=nf^jM|yWc>N;g?(2r
zbLIAiCftH<eBw&yYT18f>$>45|NEGh868#6=c7mE7!Lfhqi#%A>UNtckk`N*Ks;4r
zEEgDYFnDpbvvNz>jqw!d^mTp+ga)^3F+RFh?9&ZfYo>9?IXmi5Mz4vD4c;L1-6NJK
z=tlZLsgPP2X*-UIlIx&RsYv229u{B9uN*HXxqvsqh9wJ!Nm{R)X*1);>TjDK4uX4E
z?hjTbzl$H24fFrkCH%aH=AQYX*L=!_=(32peQ-pba%#-|s-Z~&?4VvfyJ2yGr>;)e
z%YrMD#Pv0Ka4hG-_t%Yy_r<w`nBVQ{%}T+Cmyaf**|QEmNqP*=^#8lQ1m)lgkGBsB
z_auRf%<!<1Sq6ST$+|NQVKT@cbw6B`<Rj47)Sw*MdUm*Um+MlkMH2b9wxmeZV}^`z
z<0c5?wfg<#$KWaw+wF0(2tiBz%v{@bBb5Zxg#m=UG34eyzt{eMj*hzcL0JL3AqCGv
zORVOZg#ONR2E6WjHOZ~bpiA=GVP-Dy1Zxlh@SjxSR0bpKxBd0E!}+&iiE{T)zxH7p
zGGy;kMmcxH?X>wrXe}DB-K*Ytwjj#$-cXH_nukcm)8u(t3ba(7L_1pX(*^e8<v!ER
zn9iJ>MDg@K2Jl+5vTaCFlG?Bd(P67qORn!|Qi(NT80BdHp%3EAz*R)5Z17+uizfAK
zHc#VyT0>(XV%pB{%K4K84Zj;?{tY^TuGK^`V>j2P0#yo1j}%L#C;tva$e#5=+M1~i
z3zgj($EnL;Z6apCI(V`-yM4s)uC=o=OGLw-(VB^OU`+O_^Vp&&2n;Ja<=^<&+JBEE
z;^K|?DqCfp0tTA`$>c{J7mTtGx8UqCs?;4d)UxR2(f!m7R><bGp@1$w>NFHxtVXF-
zd{9Uxs5B*t*TZYc;^2L~*zA?~_JS(1LF!;8@oHwlwWiUaUq}Zun^+U%kQ1RTiKM>1
z{N9Md!W4hidZL%bQjv-XUml{ye>O8SV^qPf&m&mZ*4DGUOw?<c@36Wta5#}qI%qy*
z+U^N_^7BQ;(Z)d=bnURf-3#xIq^n<ExsYpKl<bms>K)8kGucjJ+Y#0(KrIN{xRJYL
zYI0dphvIN&=_uTz$~wC**gyK8;g{$=w!}whhP0*kXrhG{#h!#)W@8oTP^f;u!}y^^
zh=}%nnFMy6A?Yis1c0BY$Xb)Yu){imdJ8I70`AVe0%Y^C!1)`E<rjQe-`YC%>5nxX
zb~gOhJQj`$(nCWkF-uQBU;ILD9Tj29XRf*ri6Yd6>CW95Ql0q5`$y4NNT1~`B<ywe
zm58H%*UB9r21;H}|4GeD<zey0ro~87exxQUOnw)NeEJXtsrPw30m=7So|F==bo~bA
zose}O^m36Y0w1R#Ul3?eP*E8hPe?Jx=K<yahIj7Oe*66m=pj3&g+njx7YZrncCHY8
z#@wdXG(F%%jBYguLr!M{>Cbq{@<+!+A#xLAWEtA}ycEPoOI)c`L%wmWRBElfNY^~B
zbnvCzgU<uCUpwB}hfEiHC5D~BQP6X}MlDTEumtMiZdPui`hG($1=(x^;sx9#wmhOo
z6v9hBXL%HSVdacGV@>q_un74FUFxNhr@#<DTwZv#IA7n1$z2^2O9uK;_9pcmCzmZA
zYrzB<wAVqFsticN*#A;eH0(BLLsjuWaN9cgy1t;VaJ3^Y06_KCna+qIkGX@7k1ns|
z9=PY1h^p$PnQ6V-K8Yk)hq9>d64)pA(3G<k>xHdpNQ^n`W0Gn=&OFzf#vgU<RH&Zo
z{_c~yQnqK3H1FL{fgcL(M#^^gvH!<YK(B-a%J@U%4UL=E<TlH7%GTuTA1MECWSZ^a
zu=83lOWXSVV)wMAND^mM)LkppsDEE{{x<mGn)zyHJXld6lNF|xtX(#Q_SiguoXm;B
zJ@Hv7$AIR_FLAIH$LZ)5^paKX0YJ?zZcc9)gZ|we)xqIa?6CFkt6qTL%Ws&O9>>HS
zPqs*Y`|fEP?h;_UF|-+5JP9w-3**oeU*X_8=sEi90OTa1BLRB5?B(k0DR!V9uehb`
zi{G|LX+p&wp+6V8BgK+xi7EaL2+NKFtZ*{y@-5|z<u>)sX-m()S#tDj<--zIOu~zj
zEdK_IQ;oZx+!-%a(zScXPt>Tfw!R(Ea{ug{)xQM_KCXU95emC&r&~e_(bkr|9%h*0
zNJ$yI@xp#MM&DvwW!dt_leEiJOH6@1FMp3eTdUeiq#$3mP_j%{<NtSc*yZanhs9}*
zIf#ak=G4FKV#yVJzAu##n6?b{p&Mu1h=+WAL!nF(p>l!sN)V?@=v@yR?B#l?2!^`v
z_*w8dzi~$RUa4dt>dDH@ME>#c;L#K?pDoen_Dp!k4OP(<l{00sXKGLDU5L99h7o-I
zGs!mU`mtzbAM~T4^9!URODE5E`7_hx(I1-l4-_6Mh!>-r_tTws;#eV8t^?AO$#{+f
zrfPW%qL)8A?<5~i1(}n^?GvH)^){0ZlT}($TWo2COFMdwM1Li}V{EFCzGR^5OmiWC
z2Yb;9lrygNkaLO*&O9sGsjIlFWaQQqV#D}(u9*G|qF&I3$bt(YE+%;LRqk3u_WCmL
zy7%!;+RxFKK*q3%#IST+&sCuG@GWr#M!ag$H7FUW<{6~A8h<wL!f^!Vb!gi2G>Ig{
z`GDTBmU_Sh-(@OABmKuGAUvFj7oQ35H^#5FTxGfP$5Yf>i}eI{G_%ukdK$F!20fQQ
z5zke=g4CJVOXJbpqcU3{R=+m@N%|;iPXDiYDcJJ{D2Q}Mz}mZQ|DH+Bb~7wAs=;Zs
z^Wj4mL8oN_C&M@o#=vY?%u8KI@OO^bLA;p8FQ7fQ+-NY1=)A>ZO^u>Rp;<cKw6MZ^
z8eAPc`#lvYXhBCdIGKvg57<xNE&Oq8O^e;*|1T5QXx(H55_#6}${TPPv$*)e6E##^
z$*w;`6UxIC#hF2Cne${kar<-tcS62Kg<|kh*D5Qti<te^&HMg$+E<CqUV3IwmL_g!
zlHA=Cv-|P-Te5<QavhnJQ6)~UY=ta9b5^Xt5nGU4yixqe%ML_UTyM)a`V+TTJ17Gx
zC)8p*q`q5gMQGt*nbov%FQChS*<O>;+mTg8DL1q{6*Erj(=F<2OCk^Ki3=T`?YRlq
zjK(x---=WVh`yu}G`DF657pf8zJ=0M;49~D?l==_azdGY(TMNbycoru>LdHmrS}%d
z9-nlXR&6DC&qrxx4HF0OfH4t!--jmViO-GEMNm&Z;imU~^KX-KKjY+$_Yr^czDBfR
zHOwZ8wKVd}n7DzS)95IM@~9&jVv@zz|G(ffZY;~L#E0!0(>7o1(K;Z2ykJkM`2Bn&
zZ+YZIB!U(k_#iSzRuaysxu9mm4O?JyC!|KTMT}5{_K$#EJI|cc^!5*(Cg@Mk#^bz3
z<^Zrv>`A{a_e@)}&wzA*8bZ=ayT{Z>EIk%Yd&mUZP7n%^w9lid{DjgIl5`Sw5gjr#
z6h_!ML0NHoz8Gu(fk(<Cy!sz`fEny+gSh$H;#_^(!#%g>Lg4!fRrvgep*NPQxW&M&
zfV4ok?F(ZLb6)<gM^82~5*BD_N9K)-;--kk6w2*mCdLwhuKb7-aZo(5u8iP}z>0%H
z2D9~e@RFZ!KFey$lXo<Bu`*6#p{{=#>VkR)8tD3_XN;;8UwO7_5gfVf_T?RDjz=Bv
zpc0njo0$pDFNx%^-0rkNop4|@(8fzLHh&qdu-c%ILd@d2dLejBn(%3MJTOJ%a_|WW
z4+o*+^xMv7ypYZ=C-*nZQ_g4aYB|#DQu3v&naKNN4u4y3YLPS;^aReR5z$aJI$Ps9
z+ex1U+_L;Z=$@*44ZBKXA-qiXOVr5hU<0>9WqYMH^?b4r?m#IbXXTP-(a~Jfm*W~l
z|7`!rc`7g)?w$gVvzNkAl>(P=5S9sMU8%tYO{E7%sCP`Gl+a^vWO|900HHj|_4T7!
z4bP7PVC3J`>zr!0`+eiFwkeenYakquvL_39V;xasEhSF2K*35fz8x^q2Q)Xp^b4>m
zXT^OI^u1%WL?g#s7me=)=p=Vc@j!EeQ&;Y&P<7krrsZ2PmS0GmY<NI2GIhRZ#Sm@o
zs9*MGmXh=Ab-nqMkmTW-<4SP}-)5EOl?g!(Yt&ewVlEeUs!Ki!*7|<jyr)jjqKvdO
z+_Dtcs{1cy`^==-B?dU%kh?SSceNHb!{lOO+7qAt&=5r`=_*kuAu8mH)37YDc0X|~
z==yx;NDGyy#nwf{DB7>6KfH=#b6L3~9IO?3DbIIipDWoe&k)J8ezh;o8wU~3*PPs`
z7f|Vaq)F~TqokKuUtU#VZQR&MzxJKUdgjn2J7c-wC$x*01lJ(L;BsbRg)_k_&q<DB
zL!#JB-<3hF<wJi(;CybG1^UcTSLrrI`(HgpvAWnUhvlainf7OAj-q)CtIpyqyx*sJ
zqiMn2zov@rR!KE7tE{pe!qR($2z%a<VWq737E|AP0%7GMrmg4r3bdlTUKyD`zFA%0
z1)ac7qbV&(Mz;yFaLBB-_Cjque{K2FQLsX>Wt$Ecvlob(6GrFi?S(Xj?ei*lvf=76
zHi>uqy^qqODX^2Dh^LpdHOCz^t3rNF<Oa&h(Eqiu(__SbTq<O2{BdJUd|6qo|8SX#
zDrvNthgK1^cll{eD9@5<P}u#Shu)zGO6n?w|6IX>I)>l8QypA3aH$rElzuBtFqYNi
z@5v-<b#?Ne5?6c;1EHl5Y&&GHcjD2kL&~s?WcteL$<|Mrp28^?-SV_(w)~IPWdtMa
z!%krq3_r`D+|Ta0p<s%o03xuf)Y{C_0<<(_1BgHG1yUMnHJAQ#K|r~2^vH5zjCXYr
z4-Iw#l+8N~{o#DtAhXB7XRj*$tv#Yhs0+kzU^oNWXz+Q{eSa@D1U2P)bzv)QCD8HZ
zYE-SWBMH(mG~#IJu+OkA<B)3ukR_O<t*$2ji=zeic3nKZbr^D!DBeIq^HC$|FK{(K
zjB&-r67;$MqX0kk%|qtf9Xg_Fmjk3n8QONj7JOVSv#=Yj*IES7NEg{r=?Qq=f@r;D
zwy~zrtR#M}XToI9q%cZnZxPD8Y=_Y5weqrcMh3V(;3<qej0yYTmk7?lebv#K345kO
zXQi54vb}4MW?fRoY}B^In?0`73o&Bp;UHa_-+NwwIqVG2Nx0PPid>RPcPjNxfHchQ
zrWt9L2}X;8+Z%f;hV*C$q2q^Rrf6)vzp9*WQ^qk{g0m2YacHDbME}T%RCVXM7*F?o
zw^$!5+)4e2KJH3Io9Grh*^dZQ4Gp@R51g<gV|(qgR{%TITAZd7TkcTbS8Z?5I;>5*
z?iLPz6He;S9H_qkJT=(@n>RrA{%WP5=Mg8z#o-cwU)N<ldty$ef-O%{Ut5bDQBP-q
zf?!x-mT_+u(z_7t#-sxzYVtID!9;9kB0}}j(2~xCOM=NA(S!{WwPE`v*MGxfI+t_f
zLGrCjGrb$+{z+MQCHEf($LywWw_a)zST@smXK$rHH}x?PCmlbUYip*1{WuF`<}EpV
z62nG<t)a2kc7|#){wM{o)GX6HgLlVS(D{M)a|zS4W3sd2ofP#U=U@%05`uzohg4$|
z>#gf&N*~0ZA+QR_P!S}3sk~<<`Q$`5*Y7OopleVS@t{cY?X?^7#ysi1Muxf_GLRVV
z({NUP;F76rZvK)+8p2G**^?iL<}m;xlQ*Lv6;UaFIKe2D3sc;X4T-;v@yytZ9saSz
za#u+!Z4azxeouQvQNaV!Y^8SR&s2DI#}>N)R*I<;HxSQ(U;aKjt9ttQ`RYD-pB@=U
zyg=%38c&Q6`LTKM*~=fXj)W2H0~^P;b23cp;#wQDD2&^^M)T>rGxi%MRs6n}Is4aq
zqq9g$J54)%rn9#f9GC%^KUQxCQ=BuDraNhAT+a6Ifl!v8(~<N*ol{*F*^^kI-&;6k
zI{Bq_M%cx9{tP>rV@?&zaS#ee@>wAR@HrSY94J*)Z@fFz_>yZ1FMp+V42L<m3x_#y
zWX)1b<w=4_exS%J>pUgR6wfxhLyOkyt)S2=eOoFK1IuqJ_gZE8g*<&AXM+qK_k;O8
zeu1i02>C!>rY1!ego$fn2rU#0k$PZ5QOz~HGN^(L4nfVoV+PPZiO?Lbw?bAoGlOw{
zjHMle=jTO|;6HKj-kmA>am&0m*ABk_kBl3gH~xx|2aljHT>fw2_pg;$w!q-$-q9@s
zOfK`v9o8y2LZH!Hj<~%zd4iJvrVrtUR_Cs#J67vgR)CoN<JylO2B5%bD)E{I3`_jm
z4&DeuYK=|;1dN5o(JYa&^7ra2+PeW_m7kD6NnZtg<6CsbXW{YUFgNrvI^f?iPQc|v
z*&m`kT<i0D36$0~8;aXAZ(2&HtTXM(=~tN@lckB+i&20gcug0ZPATmggNF3EpCab(
z6;Q{9*7pr;UhWK5(CD)HTCG}#4>Ti6^RyXNA2<J<n-vaf5!!NISv?+hdacsEdMtML
z_Uqz_gOgREO8)A~P13k@jRvPn@I6$6dijkqSLNFrP5I+q=6fh$g%W(3gAG?m&~8oa
z>ieY`$|G5z_Rg+uO50*=x8S`ry{}8|<^Hu^nB|+=cNQF!K<@DxH*-y+cTE1>h8)KH
zzBM?0FI#P*SOG8a!-N=DeK~ol>;Y%xZ_*eKcxa9`ITkD1Ov5MWz-<WBncqJ7RS8Av
zPL$9o>07#xVZevkwe!yTZLe&RZoUw1H^uPRGHs>%1Tj*T9Hgt?u*9SQdO#CvFpB`;
znXy}f`E@8`bzKut{%hr91s2!v+mAhN{+X6V@;iSd_a#^YJW|)}k1!Stq-CJ}!=qaF
zUA7GC2^a4nDr(=2F`sSE52rXXb#VS5p+{XpT7?ojRH+tx!jAX3vl82jzo^i{?aZWx
z^GmODhF7^Ka>iYP2TmsW!c8;sZ#Qu^F@D_!WNAr$G;j(`l_TNPr}Swo?rPuO@o$t)
zUFtKlm55)KCM$ubrBis&&uWFIc5vmOEC_}O%i6j+<k_Api0L`>+Pa6OhioSIU|;km
z(ncVq9B23MrKrf<B7K&0EcN(A&_C!-Z}##pCVQ7JG3j<6UEeKBM6w5l9+(ii-X%nC
z(5y6;PZJrNzm7bR@cfW{HSJhk*V2OgB)wT<z>bTf`iB%uQA+`h9_%^pu|H!#aZl{^
zChZjqhgnFUHwRvbX-re%Uol(-B<S(_r$7q54&1rAgi?_@^11uF;Bc~X&+Hu={N$cr
zPBjR7=pGlA%!NCMcI}3#XUG~V_VY#Azv}Ba6~#~s&7|Ca)554h#>xcdALe<;L=kE*
zDndvIBuOO*D>_>ho{Eed-(7sfN!fcK6A>Q<*(aC2id(B(IksRRmm<Pxc&c3q8}!(h
z#29{rz#1BedMCd0&a(yy1;hXyXeSwl{5LtOJPeL@-$S~ijAQz#g)_~2q>JvR*~;w=
zI-?*bRx=20nuuu2?f9hz!+^I%kAqeNSuq=J=3E$nIkB}qCpdkz-HhDe?M?;W-9y44
z@b2o$ryVP*^ZLC_r78vBg{vd!hQ`Lm_M&upKWA}6ACqbCnsQ3%5RIGn-S(B(N%;t4
z#kC1H9R=kHji-*=^QL>4kC=WFZ&pFXHiKarnxl|fWDz?|r|bi|)S<Sn1^Hj_h_7fD
zu}r8?4`@B}2c{Zc9$UkYv!MP}imWqvGfZoiv*DfX-8^R_krWATS`;3o1TNrSNOrpx
z1?WM}&zgA7y=hkh^GkFi{pGWVhq@XGnU*{I!(*vGG5#r+qH^3rH+x=Da&R3ypdVnS
zvN~nc+(g!Fv`hr8d)acySgH1zzA^X5?<yn6$Z=#}Xi(CPQQdWa(QysZ<ZbFwVP0B&
zebkYc+Yz2HIbtM-`vsH`7pa@EtouSuER!3by^+a<RfRlbc6P{Mq~)A%#;n4<?j|pM
zTV;bNvVjr*%GPOiz@0%GCN*ku6cE{(9~^cRi92Xxa1GM9L$Z0-M}9sR>D=M84RetE
zn?ZmO>ucd61ZX>7JxkQjCXLS>&HfmLyi>w=dV?GNlJ)S6_F&(-Q;bvGNyzU~IQC<t
zKYKS1Zj@qvJ4^*=8bh>;!gr)9D)QF%;juIkT210CsRzH0p8qR*>@M+hUek0jv1Br_
z`jER!LHcZIoyJD9<o&l<=HENs00j`tx5cN?baYbLRBg?{S*-V+9fhBgfS(2xp5@K;
z6jNd7|3gF+6{$~^@uOHH96t9cU_VX9`W21$_vv#Uip*L@oLMgkug|H5Rm6XnuX9=A
zYQ{L$#rN`b3fpx=40VEsS5o25l15;Z2rNhH`4No`6?M7kUtk!l{M@Vds3Pz88_O3-
z?XpGfffcjTpgZi(d)NdYb9`B7O$2ne!8J|&;RMv%qEa+fqr1CwIrd0vDAm%(e9gw;
zZ{ilaSA7LctelTa`X(;>HztF!uWEx!((GKm@lfLvMk>x9h+Qx(+PYgt@2<1Cc3$C^
zawb5zN$~*|rR5GCgKeQzd6s=$Uv1%(6KOnG8Xx7NmF(QFpT)lvI%(`kZ@Ry0RCeeT
z)WmH_H1lSj>D$vJT<^QAuoJSei3sbYtHj<MmsUcaA87o)8zj=%lh!ZSYp)rWuZ*RW
z$%icN3C}MLw#ZH(m{-27do{m)AIG1`1Fh@4fD%v(OgeZ9nPsL=zY-IFh0%6>!!<Q2
zm5+cj?*L4ybqFGHDS;fJ@eKPbD@;456%SMLIlO7bk8<iZ5UMeMhCT5~0a6=k!p{M2
z*&+qMp`RRw9_jLPNa?^Us`J-&8)RD-e7t_`0%)&@RB$<BLkW$F^~y7=;F3Bjh&Uzy
zD2ko3Tq4eed5bodi>$kYAJ8G=-S^}I#e%<`J%t!6F(UTJv8=NC=h*G6`mX<{YfXCv
z-Ren~0IU-4vi^1)pK<|yT{F|wN^$e<y0}zCzZ#0c-UV+miveWd`QbnbP0)PBVCk90
z75D+&tDET;7~IiVrswC(oB)LR+**3xH%wNu;nr$WFVuOR?iq<9gb4XZpY*Z>pNpnz
z<nXX(2h(tJQb|$T4L&}IY;w;J2)(>L;uYS!)l2ZQM$}B8reo2N%YRh*u1eK1R*Q_F
z{e~US`P)tIU}W&I>%)a<hbL^hA${}o1zL!7Ff8wOlJ%^J9t?Vw641o4o<3X*K+ue=
z6br9SJX@a2-kpj_!SZAubG$^VfX%$FweZNJ%a<_1Bt{(m*i@qgoA}SdNHk><d*@}{
z8Q4mgLS8k=Msr%bBF@jrKcCT^`LPRAQdLwr)x4`B>pzyZd2J$X2fx@@${wWt;_9iT
z$lu#t$8aZOFNdK4_cjjY&wUFBdRkF*^E0g{dzuSNKVspcrqR_x=9(JcJs+Gdx`(28
zb>8d?m<#m2^n+1-MMC9UBgOYcnEcz0QB9Xqa#!M}9skcu;dGx(Ogwj2U{L0_FSxWu
zoVohn-;pQ)kfyP5l@&B}sU2`{RXkjVyeXQPY$HFcPZq8}stzlEsX<5{El;CXs&6K@
z`9UDV`e{tGc&Wi_j+iIK>L>!{g*XT~uv*~oMFw>OA(Gecn*w6=Y85E@mE0xlgW3jA
zzL86#Qr#kWHH0R6w-AIDd9U|S{>L)Upr@95H?Lo42~BX)M#yz9K)y6!XwT0hH+z}T
zBfK+XkHB<pF4pk|;6&}8-QPs4ejf_cCPJ?tX9~8D5C>n{0~;5wS01I;LVjt(ac+aN
zVQ5PPET`Py1vI03`JSlZ`tik~O!34g*?v=|F-F<$MK*)ba%K=6Bu>TJnI-!co_ZYh
z;oytV$z#c*Lw$9fr&*fP*<;vW<?|;#DN!gJWoY+P#SmmeC;+M<AZCraTBNzKEm`k~
zgWFpD6}u8C>qvsrYSJD_cP+!C?>2oX3YpMUdKEdc12!X9Q+uL&t)7X(`3aGlWwM6V
z9mvGsQ?zMEK>Ue<+W6Z$GSsdE%Dm=q3v8T`!3Qn1a1GA!yZhqr!#w$2t`b8nlBvFm
zL%Qr6DxSAZXQJpT$hAS*a1GQ3V%EA8!pG8p)oy!G{+jA+xJhHeUrKwa#2!e?0r8lr
zR-k6@Kx3yVdJ993i#A5BHzR#q*PMb?W%oU*_B}=jYi_@Q>GytR-*{#}k60r8s+|}u
z-b4IE;A22I^Zlw05T;GnuebI>_@PD0+|;y7v;j%HceB`eu`TJ?8qhtTw;ezS0%8H4
ztIRtJz;Vc31LEL*Z_&$rMRTA3kK`w+DCq_iV|tk1qRn6EQH$-;loaOSNRd;hyjJZM
zs-YOtj2Js=>u>Mn`tjP*0Je_%?^$z%jb&S8{oP|;QkIvgvGzYM6%&8+xgWr0Y5f(0
zuErmRh=czPGDy!2ifXbI4eA76gMsM?^W@PPz_O^n!nTj_S^mLqm+xqB>-+0zp0ZS4
z$88rkOd`?Msba#V0rKx3QS26{%W~V;MdyRSXM<p&Se)?$()!0PZ}=5uZ}#<j5)~O1
zx*N?r;L@|1)DAu$0xs8%E{_v~HKh(x#!y0<Uw1s!n-*AOtm1OjYK<PTZeQH=x%YXJ
zt=<Svl_zp~svK54^9ur{UCU2qQ0rQ~#*9@BuQAP8eH7eP`>XC+D*E4)sBl`>5ZWjh
zL)l}OS+hLbm)@fdn&dU9U7b~9T<jy2ofMO3)mx9Z%N!c6Ki{<t)p2p#Lp`tRJm>0p
z_klq>zqj}QR@$Wtd&id(GFH5aflW4+!q5Bg+?m6-)!xY@LTDnO8ESl6R>s(OAM=;m
zUuOJ#0c;!3ai_*@lxiNg`>AR2-}ye!>Tv3yw4blJ>!k2-vB>5Rf?o?-hfAG<Ca&%A
zds4-E4upy#u8rB3@^ARHs76?)$_s&idje-3T9`>EOS`ODP37>|6Nxxf5+oG1q+x82
z>*#(Z`<GQ8VQp-nw{pUO!bK>AOz#;+X9tFUP|4RLM}>T;c=287<>c9T?KIX@=#<8@
zkItUB6%-noNX*>9IK=IOD6_gF)RGG<r~fp&Qr2GE+~W69efyp7^5eHBk{x&$9(-Y~
zukWQIZFAzV6^V!%jm<u6t@U!h=d&doRp8XGKo2tmnCkn-XMXYXYa!>OvE5##_|y0Y
z-l>BoiIA5Vx&-S~53{g>$h2AdS8c;-xzy_t&jyyE-8GSE_^#4+q6{Luw(ib1yeJha
z@p{!@3P$Cy4qD6-IU;`oUXVo}F6~%tZd*3Z{@n|m^Me)>_cXmm?T0sg4(}vct}&lm
z`vmJR5q|TIuK0gq7l(_ek%m27k@7tc9MNnUWG-B(a|Jz|s&f4sgJa=h!u7TLOZ!)#
z88?7jy2YxOqzOHc!G3<}?F~FX_dfOGdB!WQ98{}NwzRpnq0)9gzn2Z+3s(hZH#5gv
z`X@k&wwcwg4o!Rn4<~J+bcAA}18=P39oqVk5%6-MszCsCdS3dTA+Cu`Xm6^N;S)Vx
zc}|zqs^ElvtlrS6=_j*v&H*p_C;RIIsFt?6?^;c%3`wJrV}HKb`FCYaQPUD)JGsie
z<^4DmaIeXvF~E1Gj9@3!f2WGBK})UJdO5_a5a+)uPZN51^J1rM;=6t|GaJcwqcw?g
zII^jcDRrzZV#A2r^W#tEb*)kx@Z&~UKi%uon<Y3h@~d1NvU8^Na82+bL_G7SL`JN&
z<aGjm3A^^l8(eBav!nl{y<!?%+lSZW?z4Gj%HL^(to5WfjUU5Y%~@=fjiA2z`Efax
zFI_$69XEV#OoI>DbPn%BbDHE(yXh~Smf)k;;jv*8Q81V@GIx0s01O#s=7-7?aOu9V
za+Dr?Dq1tjY2+(lw`7yFl}?TPT5O;`K~KnwBB(IPbPv<2*Y8a%U2MMyYg!Hh4qLdw
zf)TFmwt{>K_@o}zU^{2p5<sa)gEhswwETsK%y;U`&Bf8@h4f?<S;JhPDnG#C-z7k`
z@?&pZUG8iD@PH20+&{}|<xp+x3WH89J_v4t)5}wddY%}(05uYIn+p2(9Xn^6;+@}W
z-uHuEI!#%LZN1?-e^I{;#ZsrI=kEl_qt&t;d78RkI9bfx;J$(Jd({;wubyaP)u!3#
zJ0R2b|I9yyYQh7q802nRFXonm`e-)*jJKq%W#rSJ(uEiG9WBV~a{YbAKe~B`Sd+nR
zll>*?FpC6|`Or7K+Fa{oD7EL|%wfg)4nNWftG0gcH2o`OGSNAQoK64SB!HwPMSpE}
ze?KLBbHClQxRw52(!8dJ=y?*2FL^!;#vRs85e#C@oEAj##Ly&QFSb5NE(I~q4%fr4
z!W0T}HOGXaHll``$2~Yz#L}}xW*5ECUKDDfEL?<ER(PO$6vAiu;Z^@<%BjX0eu88g
zT0|Coqn!ZOR=ob@te`$~P#m2<4@D#IP$nq1wQ*#cAlBwz4G9+h$qpf!dh*vi`8vyL
z4?wlSjLWY6vmmytBoCLp`kWr*`i9jUh_{R1wMsuY2&A|t1;6(gN}^Gx*57kc3sGIX
zGVe2eoSbIkoarQ`Jny~zT*^09Sf}v<?sq`eZ>4J@f%t5jg4^wmDs|B1Pk&eI9dl7C
zOGA}+RNvIp`-ff_ra8Rjfg!r+t{G1~P>CC`)+>L~ewd`-&T1(TRTwp45q6XrH94V=
zfIP;aDX9X@P@RR%@B8aCVeSx8uqdWv&+t33oI(OO7zMY=##C2$9O$LTK`nUYzfr)$
zAoVhDu^&?lx}0DhP9Njoyr!AqEp$_NcgO;YtAckHUbgXDCAQw)M0vI@bS$qhMk#-F
z11MpN6dHAniUKe$TglUF=eA*JCHIYKt9O<{?YF(3#a7GDZWv&hqXgM%6l)OuG@;|$
zFH7x=x5K{3#X3>UTFrP-B@SUEyI#5U*xN{jTIt}%LLMvWU12l6l)LefTX71iOd%Km
zjgw)Q#CzQulEV@cJU{Z-OZq#9|3p*$LjG5X0RmnKXy*^`+-W@4dbm0-=OP(qh)Bdf
zpws!<55EM+j*p!1#m2@uxgL$-G0npD$lP-!LyPPX)O)7tZVZ`DpGz@f8|du-vFdwc
z*xvOq@Z<+2h0x(YIK8=h0UiqI^W3X*?bu{-Rv4cUo!<wv9vh)dleaY8#!a`z?Em$M
z&JX*U6;&=N!Fy6Y6UkEg3b}1zKvt1hhH$cD{R#3#*cjbcoudLJ)g8|eAjpBrzN)HF
z=~3-@QLR?$kULNMs42#;EqIS<_TJ|qW^2qOjaq$wI;TP)eGcxE`2sCy(Hd${Htx1b
zs?1tn`c{#l<eLpeh}3#j?3Df5bn1%6qYURd$T=ZgGbDXWnX8B>Q_yUEVun2+SW`OQ
zL42Vi4+t~-2~wcbbH#ofE=f#EUn(z<*(;{CuNMt)3lm@kQ_lK75A=u701JMS_n?O{
zoV3y)HlS+gCt$883aA;|5p!Te-~()waX91nim0Y<Q)v*fNNC)zmMxHFk`r*@>UZ$4
z8<>%ylN`K38CN``s`DY4#q(5~gC3IQ@x3D`{QFdC@CH!?r|~im_~1+<x1%i-MwaC1
zLyPM)SnyuJ!*UW`Ke$j4aA8v6Y9NohJ0>=*P73cF-#q~(l+#?3-?C(Zb1rdqu%nBU
zL=)th0xyh>0=@qQWka}90ivb8%@l-69a3n4m=nnn{qVfUO`&?KXvryMbSvc=znuC}
za`9Z4<k9z6FJtxKVa`yf0!hDc>DuJZukW+Os&E1FarUj!$G^-?eui5bnZkY3({i>D
zSp>OKfq6UcJBPSZ@xFc}U-*@0S;dabh9726DR-XCogFxuM|NI)*`4IOnys?TW|Qn<
zuogJAe9IfH-pc>6Q)zMUcdI+~LPmu%zuWyT7d>^R8rn&z?j5bV<pNQORUUXbY24@h
zdz(9KI-XCL;rFM;cXeTDDd8tiN~59_f@RYdRMolTl|JE{;TPo9y-KRzxDi1-3lB($
zpV}*4BHXgdlQ{jJMwQ9~*QuT=oBH%_2<^309HTq(ThRXvE$Vi>i9Z&YFdLBd#}tzE
zx@~(d{Mo~4*_fnH)h;1bAi<KYw2S2-GOnsh(ncHoFfcNa8|}+sNl!qQ^2;ddm08+3
zWR_*^K{D#;MqLrNv&`?zK5#!=zj8)no}l#9-t$>5AN+2G;Ea-0xbVBJrPt?n{8Hd^
zWGfah$QKr*CddUSh0Dt&*Qh<<1+f07)IL|pr}6zVsbBq>#wWQd)z8G$_Ryc~C=lc-
zuon3YZkixsKcm)aJW|5ozN#x37~x%}MWMCh<^PNR`N<W-$nT|go-<Qasjcouqu&Pa
zIVQf*V&)%L`S&rf;06F3KzDv)^92^zNNq=B&-Jx2S1PX_FBf^@M3;<}fW2-OdM7ut
z6f^HCSNy?@^Whr!`}0Msml*||k-1lE9HAJUOllizyWlVE|28*kqLiibIi!N`{xcAx
zine-CMb{Ra3RshoDt8C4p9jExvcpEJzhM;erpjhjC8Lq-iR$GFO|zV`1sDbzDj@k=
z6V2guA~bcSPzZJow8PB-G<QWoUz`^-+jI-TGV&!^+if4pQ?$2?OnK^Cg6PI`zA+L@
zGbrB61E7PLmz__a{lopkV0s!`L)(zj0E;sto7U}Qm@Qk{y;YW8GA8(5W${Q1-j;$I
zJNn3#>XUS=ByQMOyX>o++$hFN(M;Bx+KLl!J!S53wQkHL>x<wB3lP(4FtWP2e<~(u
z&W6uTL14?b_2eMqTnsd}q0juQj9%y>F6(jr*0w#X$yCkdc~fWq&+3mG-AghfTWowx
zDWj-n&wIAHw0Ln`$;(aZT!DM&_h!wC!JnX?6TKGqb+54c)N;IeIQfXOehBBUv~+ud
z`Yr}HFJdY1pQTOmc1031>zSF_favs%HLdAb+YcWz2V7d}{a)%3TOAL&Ww&Ew8GOx@
z=<^$Szcl*2HHwar>9#E2V?LpFHscg7+1Mok`an_2R8JQ1c&;BrFXu~by=Yr+<92qY
zHImK{@zP7&6Qrl5EL`f4g4~kmByn#xv421@X8fcC)D=KHv(2P{1Gh4|CPIjC{gb+F
zL5~7&VjP=-dD(lrbEbOU=juN=ly9t1jR$Ub*RKRjo8R1h(<sli1SRlLssni-lL^Wa
zu*ut7MI;4(Uk_i8=G$=2hn(jWSt5pjOeD%d6wsQyj>DE2DDCbB+OOQhUzjr4g@uJp
zFa$sAN&4iIt6Yt3;^vzxG@%N+U%Nm8?<I1<75StXgPWGdFbDlUXLT{Xu-v(g|M?};
z!)?A%lNF-D)=KePf{1uL2k1xs(*F9By_G=vg*`A|b@Sy2pvhF`S7{w*PyGr68NE#P
zlpRNL6zaOxJy-xj&XA?{r==Y#TqRkFQjD5Y90>|k)Ia&t@cS%PS8<?kt#;U;<B|rt
z>mzx838fR^)~`+5JVU_t;yaBp>LnESN_vRQA^ZB~1T~@KPb+t6B9i2_%@H;G#h5c5
zH4*!~OSiY~-^@ROxS-m9ms`f&ek;>pM>Yi3&GFAq_SoCoXKG4m4UE;jFRwE~T9ec?
z-6S3VEt1=|q^~gzQf7XKDp!8x+5(+69V4z0Tzk&^$3*(NEZ==3re-D+FHE8wJoU$+
zh8FWV{On?R;N;qgf+F+d(l@=A7A+RvH~&$MN@Cv24$e8K-qm#9x2wE=q3rE_aa?K*
z+M#gZ?ys4awfbmWXevO$&7Ourr~x=)ruh$W&I>~UkHbY{sWH+Lz3g&sUbMz$ow4a4
zTdpRs!Y~mCc~M=&uL^1rotbl<xYM){iGf7v|MUBSlxtC0tExYw7$+`8F_^7`WpcO`
z6=#_=#3S+e6VPAvf}TyU|5S;bq~4X@@eamA-1TxLK3oA?XdPEwh}i|U4MOfeiGHu`
zu{JLitHH@HJ^=|31-X}EM52+zOtO-tm|;`EPaxLzL!Yi{Lrb5+l`=qu`(9|Bp=Vx<
ziA77a!quubbook-%VMmmG-Wdf8|<Yz3%ouWEB-ydHIvmiL1yLuhEF>B3t9T_J^l(u
z5)UCf563<8*bZ!re(={C4O_=V_1?V%>`CtAnMc2*hdY|fqfZ=n{xAG~!q&6oe0V#y
z0%-Vx6imy>$gKYk=$>|o);tQ9=Exa+w`jC+^KdRBR*Bb>L8CzC2XF?G%kg!CaBIb6
zf0^Q#M+6+1W_1RgLi4~OA1{(L`E=TYR0Z;jpu_GrQ#wYk{WIilrsP-&xfk#Exeg+N
z@SqvdWHFvx&(wuzAm0JMVr~fO-t%Tj=40#EVj0g8nVc<6BpC*0?j`Yo)YWG|t&cDB
z_QFL3_x0b|j&Oez*(9&<tCWIW1&$d?DbMPss$`TZJhT&^XMjc|idhR+vHK3khlhf|
z)8DLAzN`I)81A)4)G2fy-g%3WL!E?c(KQS*?8`h-BvH23TzLC~rX<&<Jd0NUNAVtC
z)&@1~wBw4)utLF4hl%`Q(c&y}<?EI$o27M(HT#sN%#bY9chkd|uguW+?wCFoB(oS7
zVyU_*I^g}1jRHi-I$1_O>h8w)bvDBk`^*`8i?ei!iBHTy>~ryy1O5)hjfn{b)r+<p
z<ZGqm@pv8Cg4vy?7TJDLo>H*4EI_g+)&~3F89QpYRFn7Mjq97HsoqAtVts)svo0|w
z?MhdY>&F|%Y4=}W`x4dwo+s@=Od^J?vW+3s5e3wbS+zlligah#uUivo6K+&(Js!Hv
zC}KWVkNzR{J@w7mvFRy0)VjYMl-pbn*ZL68cz(e|62`i0qsf)dYT)h>BLgx>U@y@*
zthEbjvqaI0WIkRROpFTVH1a!Ih)Sni2h`XFl@LOT5INS&@fu&h@bn`y>!s|Z;W?vH
zg=efiVrL|msv?=70dIynhu9aK!!@-Ug63R=Z?Qs(=iJ=PG6T4@Fc!?>`RHk2Ja>zi
zA9Nzzdr!hk8Nn;`(h=Dv$3RMy>Oi;d8@)#go&weKe=TQTmvETOqNvIl6oyHHUH7K~
zJu{;vd!zzs!CvG=?>NO?_TAoa2l|Mh)a>320Ne`L($dT8UIE-3|BEA`WA3;J#nPaz
z)C0!SR9>H>WprZDrHp_%FrIH1TwYcTSh3%aA1<Bxq%(>Z&dk1QOE}-67#jo?H=RXY
zW@3-7JcwCA&S+uQbTsOun%tWw##G}Kb=O}Fjl{W&!BnN?NkLz>pnkm~)>Wby&5`x_
zP^-Ux3(QSfP9wt#3XOe$%JRxH=6acF$0}sjJP;tlxNW2+(w=%AWG&l30R<fm;oGnx
zK|2>?s#4y6ZM+FEwp8skDTvdQWnkdf-^9lQ`YnlvLc$S7IfgX`NG=3K80^*8rdwU#
z+SG6HSD6!W_uF2AM(8ZCG4B(q$#rFoTMj^}1@elf=B<M-9#3@*Vg_ucvIOmeG|_^Q
zkZ)NJBj2u<TWQ9p;K#p^hQ^R1zCBc!=JG&Y-VCHz<>xTXLhW#gb0r2Puw<IWL$h|1
zQfdnrXwO`&J$4ZoZ5TEDSN5?GL#j7o?<q@@fEW3(XFb-_g$#={A(b^%ZLMWZ7#8W<
z&I=p#ff}(_#K%Hk<V3BLjr=jIHZR4zEkYh1ZorX8%bj1vk7?rV08mFND@9Fib5QP<
z$xr}RHZ^bW3Osgo2_OTjzWFMuu1b;cK8{J+>P4pI7PnkC|LbDA9`Q3Cc2=y8xY{0G
z%4!|2?g6b20k&ABVL%z@SMY^7jOLB#nR|GMnVHpzO=hrHV^`<VKG^SokU1zI)v9sS
zQ)=@zE4HGt3VbwW@{0)QUQAYiR-vY@O-jd^V$^md9Y5kr`<+Z9(@_zJMAcVYymi$=
zE3+`#!5E>_xfR!w!`Ca^&;grzfsia)snD&h{~DQOSW`%e@9<o`eU4XL*JQ49_x<I%
zjsA8eQ5_45Si=z9YUE0Nx_ukm#I$k#(sOg``zWvbWAsDc10Z0~adOGqm~U|OP5p9j
z5%nlSkvJs8d7j^ZDmtyN&gq>W_EG?$F~i<-Gnyif?PNifjUpQcIp)pTW-AH<65-~H
zC<&O=@y$xFsi#nwNbo;<`dp!e+Fm?L_F_G)<e2|bMo37)FV{%8^R==yDcng`p<HhZ
zhW<F7xb$x}O(c0cgqY;Z_3b78D`T?P&M$e#?@`mx`)vw6IX6lu#n@KDd-stUYcp0^
zZ%0(GpMim4RjZ7JP>&a9RLsjqw?d~~o`iIebi=(XE1vT`;asjVh0@S(3wQTuz+m&1
zXVXo}PR~DX+LdJAb*<yvxo?u|3Ag(<G*4)$QVQ2^5y-`Pqr*iu;0;o4F#d06%p5SO
zhzuStKYN~Gm_eH$J%1oj%Da7rjeCk1wJL~W4u@uh+h*$s=TnaF*4$Pk`7^;Gyom0T
zr`H!uq$Q?Q9T+o&tvFyGt-|~J?2WdB<peHowT*j`ab;`>K0tGbOVuDxjdLMjXXt&i
zd6L#9Hj7D7)%=9UEJA9ugtAGn^|dSWSyM+RWxuaIAILR+JytVoyRSuaieuEMT=fp%
znDp?xVPpcGms;Qg62#Y*RmJylQx1=K){T<11ivv1aU%Y+(JOynelUAnMEcw{%;z6T
z(@I~((W+@6y)}fK(+<&Ji|<^ZGY{hW7kGZ*^T~@2=vPMw_5#l5jxKM4FL_5tU!;xN
zUVjC|@6@BB+ur`e6NDaFgpJnhgf&_@M&;B(ncUF5?R||&I5VjEAxYvu@5+S^yaPqr
zBf?o?*mHkdPv9CGAy?n3jcC)7L|}E}a%Zweu%k*tl1KGV=i78C`bLV(c>Z18acinD
z$qQL-{NaKJB%qdAi|(W9%fx8UTYT|=`*)ZM->6ZG;+dQ8&?iv0D6r>a(g`I@I~4tw
zL66kQB4SKSLp>5btH@x?htvOxk|M+hUI&DUe6nIZTx~lM8<@-#A%2?Q9tLUuudL23
zCMUmrJoxse(;7gM!bhHb?T#|wwp`xh1cO*I$JU$ub|ajP&9kmFrmsZ51wU98V*b=i
ziFYWWr7JOE5DLp2F1E~~{VUUPwRI)5kK+PX-V7jfasa{ki!8sMc5R&Q4w#~QyGceH
zhp6y}u?im<zqZ-bLP`1L*~w0ueFupRaY&8J7GdauXnxZr{yS%0J*S7B$>jcYwc>{<
zX9-$Vef)+wk;-4XW_<XUe;=|{LK#!ScGoR50aq(q(DJMrvI7|<cmen5hXs4Y)x6Y$
zU%mOn{U?J*2ResVh6YB#M=#w@r5kt%g;Q12X>s^{84Wf_t4Kw3$@I5F_CYC^cccln
zYsCybZlQ5jHC}z3CIU6}3#cWJgunF;F1gn8(#7=+iq%ur?qwrev4-D|#OM6mMop^Q
zw7m;7i<UQ+T!S-BYY~D$$1h}ok>Bafb+?;V?3unJ1-@ht%WwABQdweVS+1P1<3YK5
z4Lz6SiSJqkNml%9Z`&X9VFxp;>kQBsC&_EZkpA!{!1$1&n(jk)CTaOAikajElh|sh
zlY1;A7~I@yG+LJ|q_W7AQz{1+_74vxc1~#%2m9PBiawt3Oh%|H9Mwy@4}Qqos$KC3
zpq8tD1#vLIQFo<?S5vyEkEH46D?hU9o_JQ^PGQM$#%#c#d(y%`jK|B)hDDejhI%gl
z`=TZ+B;Df|wQj{r7tMv9vEry!838m$(Ju|KdzP<gaZ{5@=d87ZLsojptN9zP!gi@L
zL-8TOG`vFS3&X19VQTo<oU_aFEp8f1zTg)jBU4kBR5j8FKcU8>oBjR$<YAj;@3ZH8
zTa-_Nm1JFo_pda$gjN5^j7h7j4B_rO5TDESiYk(Xx?S;O8?ap#)e-gnmW}u-|Np@`
zOj}=vAllO3Ka*Rh@srqximFQ1lD;|wJL-5MBEW!~W{}&gAI*JM#>L%cic(YWCD>jH
z3gdaNx+Nl^Eh(us;HO<g0@w>p?A7s<6)Pz@;f2+mvf1H=hLqXgwWv)r{IR~~Ul>q>
zM`<U5Y5Ro^CVA3yJ9J!w^W{5ryd0uE97tV-<Y_}o!uhT+wuDpbfQ_1shcc6_U*Ca=
z__i0$d%hOq!uKN^&0-U|kgnzB$5neC*MA`D{mMWXl7Ri>k#<y}(Oy7dR^$!|C10F7
zy*Ccc+hQ%lOeEXTc-HtRB&B7f``gbfpPb7O;t>niqc4@*i9lT9`=7BNxEtf+*q169
zn3eyWq0}--!Uii|0;1*Xb`g|de+$n9{GUQg6B!ClD6s``7>t9v$3f<!qWGUG(&`9{
zlc9w9XvaQDWlO@C&>|_810}jYzHyOX`{Mvg^Nb&({QU17h_iW~8oU_Y6v!a$S-lO{
zusKmaT=DO!G|69;FrykFU?EiBVk=Jr1|7JBp<VVq7hm4<jm?dXrfFy>C94S7gSNGB
zt*%T_GtE!$NqPx&ao4x%aXkcWmX9tHpf|GWwEKsL^$iWm?aLMp4tShhjS0}f!K^X<
z7wgr7x3;^v$NvLtL6g2r=L@8A*)P<2Mf#*gL~I$lIdaB)=wTk$d91TzmThAvIzO9M
z436o%zkk56-)C`YnNo29ukWD@gKBw^bLYyKmd&~I7s%v_JbJXn?b|o#_lHy}dc##L
z7I0ka81o~Yudkn_SX@NLQ?faeLDwUSX3Qh}iAONg`KS%kA%;2A`MKxoT+Envsnz!B
zbXx>LfDk78dxu)*)jErdiwI#H>3p`f&p8^$bpE2X{rA`T*NfE2q*}!kxV~iX*B?<>
zTg1A28I*3_R<Xn~4Kl%y;&`eDw=0D$6lOX_Bqcbyg?iNLkX>2Cb{wp<%V;npZ>Mlf
zmz|wHR~(b?rcAaBN%q>i?B2V{VSAt2q|a#Bqg+|Vu~K~Y$!Dyruk#oG_TTWW_kMs>
zHs?2A<+qn!<=*{U#8Q#Z7a0x*9M<+|HX9g1kk4muofKgZvAw;`($W%()f};mIXI}3
zw)5;f*<myo)9W<JWWf{}B57cm3VYnZ|L|`}UHLw2zD{I02q}}zoqigmjM;kfm~O94
zyVK&*r7K*y@(Mrv;g7j@{~mE1^8Wkp^V(~#vA2K7;r>2h7!pMh{Xrk8B(Ce?c@zA=
zr`PY}y6!P1NB(KEU$Y0{X~X5;=aOnR-%((UeCi+HBDb=Db5+l6^nG8~P?(VOhZH;?
z+c1)0#uCGIl39w{pnB5llU^uc+7`mGn2bC!rh{cT^tv8Pw%~2opqWcC$yaIAcB!`y
z+4K7JyB!Ke-AK{yw^#~d-ud>syz|btNo5Lbp1#b|;wq0G-$N-up;#h{0-DV_%~lhE
z!uNeztsY8BdfhJRRElCLjS`ARvrWb+P&;fe?2j1rI%Lxd3kxX)mQYx|L!#Ti!1hLn
zGp`amS%f+^3QTERnce*zn(aEn;ed1JF0r|Jjlp0*WnmG|8*~5OUCy1|q}FIrJ2)VU
zB7!hvFdS;1I?K|<^}bKP-zS^S=*3bcC*kAuHOCreUK(?uUE}FbXxz9%c4-0kS_T7?
zAef?JJv%Ws8ENLph7A=X9PP9&q@E}<>Gw&MiWs(qur2(NN7{0*Er-d-XW2D)BW2Pl
zX6Qv_+U+Kdc8$rnN59l1lPe*WKpGN(#no4?asB%1<O&r|t)C;8F0fQx1q_PiG6F@r
z)1s9y_dJimU_>0nOeYh<a7@17f>5+t9kOnTR<q5pKg8>I$YsD(7Agi)L|C1D;(PDI
zbd0g_3UR7{Q1RR>R5K6RJJ_SysxuysIeYdZ8>cTb7!D{Di+JNPTaO=e=JY9Q^;Y6r
z7!!ngCTbi<^JzCz->28>lg(u37EE8n{AFBooeO~&OlW-iIoag}oU0iGCPAQeZWxdc
zkC^MtMZ(-PK<W+sq~9l1E^3`W#e6iJvg8QfbWPfYG=uB{tyYajyT)|Xr`PL{%Pk<4
zAQHM!<<-}&^ZM&=lFcu0di?^$e2LYS4M0#VEue^Lw_CKD4L~uOOz8E7#F4}skMIja
z3Nz--E*ZB-v(aMMA2RND$Ym6!vXC)YBEsq%5a0U^Vmd}_yh5DHYv$@{=DRdobw;BR
zXU|>a%;_r(1_O%4GM+c)(W6Ifp58dgJO~+%G;`C?ot#tOr~fsWzigd{eUHYCJ7kv1
zIGN|`Jn#89FVZ=RX@5YfR7iAg<B!HiIv*eDyj#l94NJ7!Et;KrqVp=5Y)Lo2SYQa7
z%h#^*>MO62D^^%vJ4Ys!qq?}P4{qfJgn-ToosSs|^tiRapAv;0`GSK$(Q0+cx+R)3
zoe!SYd5G27BfhU-;%S|ypQH0IWdCrFMyrlD9<h1uA{!f*=nwkj3kAIKm?v9XoZdKv
z<1{t%FwuF!d^VPC>IZ)ioqw%OosYr+QkY03@H%}S{q{qyuPkAfmWU!B5k;78%xb%j
zBRv9<!5<7z(+O5SPj+bm!!(Fe2JvLdw9zKBRK?9?=^ZvHE#xo^#e45;5c^Yda?Jnn
zl+EAnZ?kgt3d1Lo!KlMz?9p!4v2z7XJHz4b0spW6+kelmfA%x3z4;dF=gwiMfK#jM
z6c-kG>+N?iO_QyyEe;P4$>p-_AMCQUw8EXwKEV(kt{q^TFdmNh?8Ys={jDFdU+XYv
z9$-uwoKLS3M*&6%W-hA-{Ca(YPk)Mv0yzIJ$O#f#hf3oRZ#d-f&SPrLLl#Sm{NTGk
zrcymcrBdO8kKSkR$rC2y9^d}fn~X<e4)^zoB84{%IH(`u2R>1(QwgJy$D@ajsgxHn
z9SdoIG!!a2IicXCo!)^!DO@42g+U|*oo1VzPd+2NT*a)e5`_WC7}u68O^2kUM<86H
z@fhj*Sedlm#91b$YcidTn09)k7s@!PG~;$3k<nrF%jXv{;*h-b`Qx(Ucl|LeuQOH=
zgJGA6r=y)twty6RLpW@=*}8j=GZ!yXU0EZSwm5rclR{~MtJhuu1W%r9(d~9grCeIg
z2GzwSwjSNXG<+<>$1+lMnk}{;?{e|{tJG>8x{ZCDX_FPZjEX{JWME{|pcHn$N%;A%
zh@%j7?kzABN*TIIAr2W025j$b)9y4`EHCoTo9|LxI!&omVzF9bXKM@7h<W3U*ZJ-H
zAL&h9tVQ1GbqGQq6&eJ7$av(jx4X~cLKQpZBE`&IpLta8Wit<z!bM;i2Bs8r>n-+f
z-X^<JCAZ`v!VsjyH6_(xNLo#ZghT92h=VC+Iz=k+voswO-y7ri`lL!F95;p68IU#|
zY}@3_sWPUD$V#8@7bTy0Hr8U5i45qE`_#Q%Os9d9%@aw%us2{j7|^XZSlKvDxmu-|
zOLPA0SqjB67cO2#D#iBp6NZB!Zpx+GX|u3UVQ1?hrV(HWA5+*g>n#rUYMk9DQ9EqW
zsqd5Wn^Z&r6&XlrVAv^;F=qD=H5uzP&bc>|@jyb4AB;mr!vVYdJ9N8kstb#}as4e8
zmo~^|v*hzR_O_p3TZ&g-d6oA+_?Y@(jVPA*en@{b)Teo+2!epgctY*4Ms;xs$90fm
zR&(<QW-f$|ut`C;-e&K{ZF0+1j3pP5=sac5n0ti6CH6d|?_*|;n8&V3&>P`*d5XDC
zk)GRFz)%r6>GQ*)<i<F}tS;lpkioc5W3o%z(K?Tm!KgoA((BRK+h=`qlX7*LLN>+u
z&9mf7WiDO1f&!jAdBR{YAmuu=J1r`eMYbP5z%&9(F~u}&T8$QaI|rOQQ>1p-rc>X?
z^_wh;B1C4wJPj(w>edJ+W1=uboqZE5J#}HG^YL)N?%ocaZj0)|B5%F+Hj7KAb?tYl
z!0wYLIHuy&>#y?u?@luB4F&{Js2k;i0B`JZa8P4$aS11N#GL=kI^VzfIk}Z8=F%D>
z)D15wQ&J5^WR&+Dofj}I3)7wHT#+gjaU2&f=+AY2dcBO5=sYM$Zh1CV<(SSViOw^L
z&bxj5;edX##mcD-N{flkH_wqTmX37(<O##k5Z86-cH5N86?UFH!W039m|_T%X1&GX
zUY*U267^#`U*IX7$1!H-5H;}#!wAm40rpcmA0;~PwW(IByz%OrEG})3$z;gqvh40`
z<5=*@E3Z7G^T7$7hfKT)wZl5q#VU^LzBJ}vb4-{33t~eN3WbOT^}E|_+`WhU-6BF}
zP*_w@R7FH8jEI?HB<Uru-arOZ3TsQ)X_t85;15O!AYCfqxGq5)k;&(=Y?H`O(W=)u
zYfkv@UR&dD9yA%}D;Oe0qt_q|heUyoHx8&~(rhkPIk<O+t$Vkz-8A{-RhCb!<JfTj
z&P_m2tJOHOd5&VSOm(rsLaE5wQYlHFOIo!WH*VZwZRtFIFJja?q_OoG*GtfUGA7(S
zMar_V3^1YyP{iJd&M*HL@^%)rb`GI@h!rZ181%ce8+B}9@sl6?5o^^7ft=EA9I*B9
z4tv|%{F{IEueo^fBES08`v637jBPn=Y@Frc!w0n5O@^Z}rlIG5hG7VnaU%T&N1M7?
z!Td`prd5CeN+==(DpWM@@33<IA?bxO!ptI+K^aNah;X9_IS~-XNI%e44N2#em9nu+
zi(oiLC~)&RY}+PAk;-MLI0k_?C2Y63l!{p`X83I*Aj}romnNNYpW$dkIPvhOF^icD
zDR0Q*ZkxxqZ{p^1IE50$#YIpNJC7X<(`4A|vQk~5R9;|xZG}=P$MRwk-Dwojtku}s
z+9sE-5Dr7S{Tjo=M_fpUgf)-w)H0UiVv3Yzt|ZZTNdJSMk~Un7)0gHqOT`hxgn8Od
z@twEdWn+06KlT}Pn;h&tp<dhNy>EYq^XJd=$tRy7AW6SlR4U61haQbajs9?iYrBBZ
zZS>Mk#?Xjom_MC{|7OgE?(7vx*TMyo=7SxUHXo8Hmykvtp$tkwvS`FOaZEf7kzs_K
zhJ@Y}6-HzhOPH>WX_|y1523)xX0dG>38XS<s+K`8nG$z8Tu#R<<>6)zVy}RT1-;3b
z;b=(UjR^xuC6ggF8RFLud3@^zZoYt-$&xOXka5J}UJBE)iTo)m%S)6>B~GocP$;BX
zEaehkuz+Tx!T#PMu9YVk`gHmSc#Ulq+yL2{61i3E8S^*>6k&UxVZV=SI#_2e>y1{z
zeAMsJs@KW5Y2JSAO*U4S@xuUb*rU@tq}{6X*6VL@;lc%O+_;5q7K{;wNwH9-)9uk}
z)aVU{q}?<EgE)$bZ8@Ji`}B<Gm)a<ygf8@&F%Kop`#UV1)jF3(PBZ6-xts>5MCVU4
zw@iyL(Yce&>QoepR3^=0!o1(<a5){ZT*&ZAM-hz+?8XM&$%w&dMCf@0zGN|*CN&un
zH0zpqp@5ailP#B7l_C4PDJ;t(n2cFjs!}SKSzlkJSjbW>=XLFL6w<6W*xTJF<rE1=
zKHc6Sqxuuhr+nhpl+djn(|HU^5w;H)_I`$IxL9W{>6h+Y=RKN@8W|_UJFmaR#>xuQ
z$Y(t0(yZ^(Y98|TTW@mV`~_}&dXjmmRH4@!P;b`g4~C@N6ha8iJU(L1ADH>GI^Sh!
z^AVYH8EKdZB`6BXq7fy`LsS?c{gBY}QDH>7Qo?j>Ow)Ry&X;C7?{>J9j#<jV&AvcR
z3T(%M-gL}xI3gI22?IqXVLq)L@;IL(RVZSlv!n|}R^pJuy|jL&LZ8*8C5q*;)_Fcd
zrIf=E5Jf(%y51~1R*qmeJ)v{eo)V=_>Rh1+TYC(9J+1T8m-T4Kr*ximGrV>E4K`L*
z@S}jqs86R|qtmMM#_O+X<};l~F$jZvp-j8eqt&d@>ksEvNAvNKbLM;_=3ld>&cHN4
z7)hhIG6{wO+y8Qd>gm%M>!&b<WK{+%l(Wd##Zd;zO5>*TM3V`=A21$F$rSP!ZVEe$
zn07murbVt;&^uxwuroPKJ0?8|7<F6Z#vT5f>nr@rPM3dp1nJdtJZaVG4cmmVr0-8?
zwGLPr<tS$gL{>znH)8wo4mZ*T%H<LZ3uW^8JpE3GOd(A!o5q`Xm_~}hV901V!t<sq
zt(?V)V*0zEvwiT8f@u*nC8}r;2UBe{tYSnI5l11DL7#D}LHOH$&+@<izY>`aN-Bhm
z@VybeR+I02=SN&PeHIl?*x%dd^P3;gY3x!z+(AX(!4gJNn<*HNM~uc}1P<fzg!}jI
z@$m5$Z@%#sKl<MHL4ZG<&Ks|Vl+VC_Syl%^nV1G(=oWIM$)q>o@WVS)*4D7koWUk$
zS%#EyE{3$QlmVuT<79~ZDRCePCJ|01h3Te{L4-f(V+ey(HZwO}vQilg(;*dvjJhor
zrN>{MDYKH2{CY3uf$g$C=rbA*@S~WXHzE~Jsd`y5sVrfq&%mm4nAvCN;V$_?fm|+!
zn{qka+b5GxlS;cxyfKDw7>_+B-b8mPrAoN5WO(q9gT@Y7DN(*6E*r#wk1`E}iXo1W
zkzzXRGwMDe&g3XBE+gHnu9=WA{$z|l8u8k-*V$a(APzighczDGzfHe=$gtnWwQVfJ
zL?oTF!{La*V1#A6Os4_c+uLky?{f9ZRo;F39h6ddV{h)qC8Vx-nN1Lh6Z6XGB@|Gj
z^i-B3<_44Qg!-rVSy)>ozqyH{LY8DeDQ#ma6AOV7Hp)s7`#v&Igj0!~axvW$aS#y<
z1{et3bZTz)X{FK#)5Z-#qF#$tBcQyRr($RLum*R!4*O#pZ!*FUBf6sjPB>w~%a9oi
z3Csq8<D&8fT6=Z2?mZ%vPGeaX>0E|%E`yVJhZ(|QGMVB}ef-H38<(sl86G~S*4if<
zM~JB;9tq;0pqWd}JdPCpaKNa)L!8M`T3kk^@+X)_{)DUNuX1{IooG6yUT?Cyb)R9c
zK`<SX&t|brbAFo#gCV`%0NY6s1~CVRhiq@}a{l}U$_r(}I3gHD#4;HhBz1v&6h?$m
zsB3SOBnZQo!rXkCd2d4PlY3Ow*2tYXgRLT#Wk5OYU?j{D4$@8$PmeKAyK|k-n5SOA
zeB5oZ;!OF;dXYsZ&F>Bscl!<pBOA{f;ztqP(GVw`Qt>im1{vaJ8{c-Q=knCI4|!O8
zNIH|owr$ecESX#eCw0W!^ColVj*-E!CBuWq95(mK$_RW(JT#tRE)(X%{&vEA=~?E}
z3D+)O<ILJAB7aQ1)?oY5T?XAcQ?E}hm3fAFzdyvbQv^Z8{{BAO+q+!2aFNRE6{0vI
z7$5VblnOc1`HZ=g1kuZ3{;bZuBb}ew#8DAT@l59i76wY#STmi6ipZbo+$HwIBb}#G
zi5tX>xpky-C!n-)T<10iV~fee!w;u)M<GryrR<H#3<d<X2BG7k3Psun4YnU_;iglV
zrbT9^bH~O@m`@X(2a_qLvdLJI;o%nb)&ZF~e4)<eT<5!Apz|r0&tGO^bxkv`H`#r1
zpJA^_=#MC5bBR&EuErY-hxGab94Ac}#vC3Vvc0p*xpNmNl}m(Sh#!T-u{NWd={$_W
zd9AgS5<d*S64tM^sk4x{hCpE?-9L&@D#m+V9{>EeT=+MCOeLOB?R9ZQNZ>eFO8dkK
zYz)`Lbrc2;(Imof90JS4c3iwp7pGE0nR*I?>$*q-iVG#YUXv&mxYH>=I#cBGVxIr;
zqsIi<6pQ7{G<$VAqaKeYA-j_<3&S4Eu0zhru$a#22R3EWtNAn!5Aef~SPEjp#3V+T
zHkRe$BFM!OXHds$*Vt>;n3xu8rcGM)$))qekxy(Ggp&!jWfBGfO2+u%gluyQv$ajU
zbRLjM6_HM-xODM6m2#1XkM1*`j_LLLK*Z+eX*M@cBSg%-yLa%X0SJ?=?Oon~{{z<6
z&fq#J%H;(dd!NI@1Gaa!$>$4<JdaMhjqm%Iv)1cb8)mYX3!>=j;C()GA#n`jC@hR3
zjunI57W?mi%Gq32cfk#Nq>xD4)}!K-_LDMP3u1|gOyX&buxzj`%rL?q4loNjFb%>m
z#I|g(406Q+(_xP|RAiBS?^K?Ryu*Kd&_Km5ne-}y>5!2(WOo|U@cWdkAq$R8+D>Cn
z_4Ecg9xzE|^kO3=3Ik<X+BjMXgfcMWh*T6{Ogu)tCI{^%fe<X&4r%F;aX`jCGBk;%
zQ_Vct%!mG%O#Lxtdl$8G3Eg=dlTM{Lefku|e2%@nU8cc=R;PpK^{FhBwEtEdvi)Q$
z>DDsXKRo2tt=p`utYaD$`Fs)AsWBMzX||i%`d~7l)oS5+o<6`I3)9akdSAw3ZwZBC
z97_eD-OUF(O%6Z&3^$jhYQ~g@eNsXpO$$qD6K-Ldm{_PdM)(k;nG4%Ogb~4LjFHcP
zX%L5@_O`SwvV}a;(Eu67EM(w2YZ+EECb{hhaB*yhaWG}#kEsPA-C#h$8dI<>Qg#aE
zd35@HhS@xsOb$1j)lp5uL|7I|MnsW9#R?+~utOiF6Ndc`^==CZR4t2?jL6tRGmk7H
zf2zAGqlh>R2&Y3b^~V_PebnkDK+T!2t*nsGWvJB-@xv+2c8gxGNjmLt>eLEBFkx?R
zm%#Ty7}T3h?%#jFQgsDGnB+1!T*twmP8kdbNEF^=LbKUqG#bynxQ;i|pcLUt(fRY4
z4|iG|d~lP^T!yMKV=j@V*0};q!^CtgU2_)*VsWH%BaHAzBg}mE1<d=zK};#5`0jd+
zwXDt0w%UB4TpTCG*!S@KF^54wCm2w)#uRM}*G|F2qdyqp_41_CS)5FUSSb+NFBKUR
z#oCnHi~{U1KukS`y$-c*i+IjFB;!2GJW7~H(-ETn7^AbF==_NJ`symhT$aPbgE{lT
zpiM5DCbM~pFq*QvyEA8AYc#lf_Z~}2tD1Q>hvPU*CLX;(k61~@lL@V6Gdb9!|2%n8
z=iy7($iGnMyEC0<sG5@UxK9d|Fi+k!=8XAMI(IBYG}C!HVIJv`W42AUkjEblkYPk6
z(fLZE^AAX2*)HQ?$~2hiyidUzQLt^DweURJy&nB+j!Y&`I+vxXK%CIIL`DiD^sz!8
zYdTGI-bM<=qG6FXo}+WiB#vh~AAEt%mnmd3)N6<Mp--dTqT6Yba&6YumI#6g`}=zY
zet-~yMzhJo`wyuuX`N>?IUL6&@TUw%Llh)BZ!sE;NQRmoW3F`{eMRP9Z&PPTT*G7}
zAx=WWN=S?t>bD+JS+8<-DT@q;1QrlQM9EIwFio90^rooc7?rZfRtp4{g<-{*G9;Qz
zz_k%FV(Nvs*#ePN6e`PjogR*1AXUt%9Q@aBtYY5T;b*&D^4SW@#WLOTfbMWWGYqJQ
zQ(Q46Gw35NgIvvJt-8X-g$s1L6Kuo6b!}|ZCI}RheiJpQGZ~NB9#7cueSFjAjbe)2
z&|_sOPbOs};}FZr5l*MbwC<{uQG_iedCMWJA0U=4pp-x;lYFj3xme`Z%?~jw2TY4*
ztHo!Z-)1lzU^_Nfue?IBxPXcyOhb`!Er<fVVF%YXxP1O1XHK0)N<|b#bh}+T{Q)0*
z@Bt4WKAw3I>+ZQ@e}5U5DI|^%cuEkfnOloO_%Y4TAG5Gj;_PZ3BODT%3K53HD)|hi
z-uz2HB%b(?Hc1t;gqDS{Ok@}l`%{GLKok;0A$BT@Y1w4*CH%n{ecp*VSG4%^^F@@X
z^Xu9e+ziE3nz28@^QW`|pXPXqBc`~60oW#mOoo-kB`T|Hj3#5OSm4+$mI;0+nT$t-
zqb{E3(VciSe4j`doXe#tg#oLJY0|ENlo5s@i2Q)qb%~+?8Ae!2k_BYDfml9IlBXt{
z$&q#)?%lb8Fl<!P*8bqpBl^R>9>b&za@hhhiZBf@4M8~d@P<7c)8Ops(^M;qD5V&W
z#*D@u?QV}-w{Ee!yPu?C0W@nP`RDnL{u?pZjZKrMm@7Y`eP@eGwaB^E0#-C6G894>
znz^C9tpp%LN$h)&wr~n*{gStI>L&Jmglj<*A|rvFN@Lj$nOp&XIL1&4S18UGZRDv8
zvGn+$8DP2&S;uA?`uJf;Ckkm#eC&yjBZdgaBA>}%RhCG*f~hybjCJRsWe8*>nRp)3
z>l02U45kyh(<zaF)9DloDrCLnkaEz>1;jx}97cpufDA({DM`y1x(&qgIc=1s1nE=;
z+cJ4@?=yt3P=-aX-)C!Qo8F*@X_~lBnrtSoH#XY%iphA$IOSpqSYKTwpUa^Tbh|w|
zy&mmuk9+s-ad=ptGZ#sUCh1XrDK%E6kk~>z!yG@Nb?0%yJdZikxj53fk(dIjFedZ^
zRLa6FrV*9}mI-s6+s|QMOqhd%<b2UWp3V`g3BPOjn65+4b(jP`(=ebN2DBzV_GF4X
z9D;3<A7)uztddDB;d>LzNOxUYhJZ*i8IK9PKH<cpKbg?+r$j<<CX=Qdhpa8Qq+i55
ziZG=ltt50Bh^3Rv9oypG-Omt)1%^qt*XQx}6V2Q*u`TxmbAe*&jqrxrWO`$5ol?Gt
z0{XoHz5al9x5u43cR4sbWIn$<;Xaf3%P6aR5py+-Xx}}q^AV9DPw3n*3`CNqkNpXx
z&FAPmc&^UVxjg>RJE`;a46&T>VbjNQZL)R|KML@}kZu&xnNG1L(_=c%VppnU9PlTa
zxoz24CPbmcpLi&5KscQ+m`v&V6JlT^(fL~OIXZX0T<0k_jcpn{xOWR-SSZ7!*B`L8
zvqOK_(?b;8%$&KAly!OIA)|rT`P#|~xolSJywj!A>Cx);xPSjXhqc-f+Zppj>wl2h
zud`%tEEN(BSfU-5ZBHWa-=|!-h;1u^sgD^6qycU=jWD$#pok@tv6qDU4eV5k$!LUS
z8mKUud%++f4nu@gxakz3>k`B<qmECf*(XNur>iA$D1NbBqY)d}c8YV$D;SPV&l}Mh
z3>bJ*L=<7il76Sj@UszN5V44$Xcz>J$&Tl7Fq|@pWBgbVn}VWgaNV&edt<Jzm|S0p
ziM%dWX#*hyj%g#qm@ta)#~yZU;8+6HsiVRHkU~kxpxfcY-~NWJ2e+~840gJRX=hNO
z#dJKOc5q16$+Ef@G4;lJuAvYFp--*3PvD0HK}ZlrOr}2L@rYi3K&#(lFc=cejg_Bw
zy@d2@@<he4W-XPr><~gAMgd}bheGi*wiOXh!#Q(1okG}_HU^F*;be*_^k&X<Y^KAJ
zb_<VU6agxVPzo$NMkq*SGQ@sJBqihClx};75C-2{%#e~19~|~*go3o?kS`R$GVsCx
z&z~}x==vUeJfaKqM+3qjq%5IinZ%CC-o&Rqn&3wvL9B>`B5%M2$7G$5tBY`9QDVds
zOhfx5*`^6mOc=!k-b6Du1xBwy><>`?f9n1`$kHp#&%>T`&bQw^_pB<^hAQl6Y>n;)
zx~I44Svcg(Q069ylq}g2McI@YqAbZ_g+k%5B%2Z$a!|HNa}hZb=^*JLHC$&j^l+B$
zMmM_AjeW0u%k}Q}e)~Dcf1I0HSp}ebFamn{A`(EMP>IZZ?(^OAo%6oG*K2?_%%>CH
zdE<31y!9$VMfgz*p<;?uGaXNvO{XXnE6eN5XVXQVp_>aj%VusWMOC~1Pg#;>In()^
z;bg*aG@{50?8nO7eV8N6UCEWePIKhVo3y&1R83t}NWEYlg%ByE^st_BuLlIuQz+jf
zHyWh`)HMi~U*?#HuEsKsF=d6emiZ)QG?)?y$)}ei0#otw?u^}1prl75Nia%K8beW4
zE;gpF@#Zr|z+^t5DoVNry1s|@6+3ys-YloA3o31?tf8skY!GsfuQ}heoa`AqTcV^y
zNr@-jYqYL4WtQU^NvH&3v<t-;>$k94lc#fTT)D(MZ@z*M0V-@ER75&0nG7e)W>e76
z>#Z=GI_6r}3<mp*r(^26rapLWRTX7%g?W}Snob!E2Bc}WAbXqp!9UrL$n*DOE)Xy+
zkT-77?w-I?HFb6;bKxtL^x%NGSTGNK^4U?(apAOpT`-U0*gf=F!+er68qDyKJhL1V
znu?cprtDM#A$^*S21d0gbWKr}<avRqHC{Snv~QU#Jnt?%uLF<mtYCMZQB)<hwp3V}
z$~xw%=Dx1w^pZu{0wp`|Wu9jE--G$pi|_L8TdyKSfETq9G9;T;Oh!{?vk5{=mX=nS
zFPIyx84mkQX5$6#dNJd(Brh&8PY;+6$+Gt{zkPjt%q0RlDNr|W(&`G7tlsN+sF0q*
zdWt%)k;1#vb14^|<3@Ix+swl_!Ib)l=QBbf_|$SpU`k%uowHL5JmHZfF;+Gxt){GN
zio(5Dc<GE0mg#&-Srm78zMGfq%`?iXq|%0Z;rW>;<Yb^Z*RY)I*(08-_wiia#k`5t
zhB8aJdG!kKy!8r*055DIWJo$K7!StG=I`TqUDNOHGn$O37v&SRc4ZTnE?s4wWsGN2
z`u&f_^AEMCJIFvg7@l(QVT1rBIdiOsQYH0lhDl2-9u|$RD&kfX5%|b3Kq>H=Ayx_W
zq9#Q80cufrjBr7Nk#HN4LI|4e7WE?Rah{bldR?Nh=5r?!j`=lTzmak&%^2^`sg)#Y
zG&q0q1S0ULvI1XgOsN@X1$Al2O-)-_vQls}Ety%1Cj@bzXqbv;dLd6NC2TZmntqP1
z3R=At!XU)=eSGC1j6q64I-3!C0YVtW_y+al3OYFfS`delb7#)*?i(*L81528HJitl
zSY2AdTKE0D{q|eD{nopTvWn6Ou5az~5B|xoQd_e~lXgIFAGO_m*`wq?(o0I|W&q-j
zFBh#Du1>;J9P32{qI6wwd5N_OaLL=kCJ5h01U||Gm4u+&_I*{?ZlN!v>)y3u@n`Jf
z|0_ru3EH|Qo4nEljgZv@J{vSy^=e+)$+(&s2GbPndBjmfyVu38xLv@&T7(s3l|~!s
zCNr3dG^^N6bH=(xSP)9GrZk`KMLg7wSP3gSQH4iE61E7#2;cWy4V<-$r|pzLNfZV#
zxkWv?g6^CGqlg2^=Gq#sd^ck_?i0j4*4IzdX|7V&6=jie<?>~&T)xh5mQxzb)^4BQ
z{Mt7#SZZw!>!r+M6p}d_oAPmhK5{83T-*JE`Qc4Q2z*bm-c$s(z@{0xs4!LnE=@kL
z78&?R-$SVg)o@W4R=7Et2#GHhwsLHYzRg@I>E7hAi@G6;ANy=3K5=MSRyD8f72L=*
z{XEBb9!U_;XeO9e6RR~L?rSeK7(9&7Xj7Bt75iz%sH$DPE3g_$p6-R**9ce&Dtb{(
zq%?655{0h#{(w1HiZsPn5@iiyvQ0I*g6^IHtq48AQn$x=JY_ga36c)0t0zgKF0DpJ
zp3k{{?HV_4Y|)=&RK~K`AMyNmUcv}VT`yc;IIoS%tGXl8>@Myf<tsofJpUl(8!aE-
z7ML`{<R!)`w=l13_b%YS4|AZZYJ`+{QXP7(k1&@?5+#ng)`p@o_({M<?DN^su%c>S
z-OIV28Tx64^*xdxq}gg>I&Exyz+6yhFjk?_7+aI&HG65ssH(uYu=!OVKD88c{}JYa
ztce$%e}Cpz(cRNv6rnFzSz6{hlNrPDoG|XOx^|LA(xa$yifqocYgf5({U-fsPGt-`
z{Q=K?`@4tV@4>VHy$I*mi(2RpV*XKHO1iwX+sqGk1qHq*S!*f+n`5)@?YUH-lJ|P<
zp`cpa+Yg!Bh3C<s=an{iQN*#tCkbIm)x5D^a3eSLb3?W8JV_Evqk+|iP=FN5F&92Y
zYP7A%irRTzmUnpmsb0wWB%l{mEX6gEs!77u!gIH%J7k`v`0~A;3#|xzL9g3oG@3FT
z%?aWTt7|7oq7KbQMxM>Ne*HQ(Z*0?_q*TVRHyH7)7hhg5cLM_sc`sbo`lIvwLoMnO
zPq=yoQ@gQPSUe=scr?1O)|S+R5w@<d2BdULfI7=u5N#MCA`d_I@FO21z*7pep{NR?
zC`2NWzE4`r(7Hrgg~C!<jURYyoIJ+z@pU&}Q@RW)PX;VU6Nbe6^NYJ&7+2WQjJ@Fm
zqkOFA;e{c7qd~H~imxJs(CnI=SzZx{A+0n;mo>hy{Qmn^_~}g_Gs`H86oWe|TW<G=
z!x*)w*T4#5KcFb*gpCG%DX6`i?41|k;U$D_(n?}Zow<+gsmE@AN*E59+si~zLRHVH
zgrP8&sw}B3fI%CWErz=PuCI^jJw9HH%E4%)aLYQiI9hJDeU@92dOCr!a%Td<73$ZS
zam<4d5h}dc!;3t36fY%KYpU9~-$x*oM^WWyU7_$6f*CyDV`X!d_Sy<YfDwj9ni43F
z^<>I^?eqNBgbSmBYBDFA%&@+XP(GFKBf}7tH1VZJiD6e4q-Bk)3~lYiY6Xff-qYc8
z%aWifF?ES&B}t=6uhk_9UEw{(AW#I#qpov8FT&GcvN^@IH?j9EBSf2K60@>?f_~<)
zH7p6!5t+S77{-*<jKUb^MM+f^6xuD+%DQGSnSMMFKjamvgRMG?5bj1NtpFs=cEEB|
zp=VQAyji*yI!m2tqySk5E{ihq@D^dDQo6{O+89*e8bqY>DDpd*3t{n;qPx-~>2=*e
zabbwFh)8*?Co}eI#moCC7lt|IOjD6T;3K7?QXW;H+yi8w$gH6<6<JjwWJO15jD>C-
z@a#s+(_KMgDx}r;N)g2g-B#Cm?%!taOOHwy1YU$PV6z$Jjd!u}3PN;fBoW=ERR)=4
zdt4Ca6LPyv5JVK!jJ!6?vz)rhD6|_!pO+>1;CEO2{Bd8Q!f{-9ekXIErPkfd-HctG
zX(|CS2oQmXitl7@jahiET$zCKD6;oq?kSemmS`^Zuu5QX%tPg|)|k?-JYL+Jb75Fe
zPE)E`>O7Yoh36rH01*UuVM1XId$uAgD}*{=uIWTRpF5WD>8>C)6++i|QW7T(y3K{>
zcQdb#Fi&TcH{Qm^%Lvh-8AtS%*B$dwNsx`n>=r>7QI<1uV@R`{s<Mj(ClqDHUOD<<
zqCUJrwK$2f@AbUZ4p{nKd#><8?*lzo%A?AQ_b^uno_9N-6h>&`Y)+&U>xpCjazEwb
zsGyi<)MUs7bFJJ?tO$Ihm1HbDFWo~#M@g*Jbd!K*)?=RT2oigz=bh$J&)ov>ka_rC
z&k=1JaYU!L${-VLPZpkY+w)8t(s@o@r-#h*vZUw_KE}uJJFgG5s0)-^t=C7sIxE0P
z!}@X?%8atE-S+!pxVmecl$5$e%xgrg2)qV$DN#Xy@B3IG&~*hwi+OJnYb<qH;;9HL
z+%Ar(HGvlqBrO6@LT!k=ghrH*L>}iXJaqaoe&?NC_Qsn2aKx<4p~@I$b4J-5wZ6pG
z&SgB|v#KJ(`IrYh<)&E#K3{rpjeDhKcC!y*O{dwS(P)q)3GG(LJ&{U@(HdU}#3Cft
zO0ZH8NspbIw}|f@W4(yF)_m=S?{H%<q|}Canlre5?VvHn{S1Flw0{qmue5vmwQ#s{
z6ap&^OYH#PRMdr&+znl}ZuiSr%DO=4dhvvlpi7Ag13ceHNJ*8Kgh7Zk1{wHntGcWZ
zfdYAu4OHU$AyFe-gqvC-FCq>jnvu_0D>!$2l^1V}*%=i~W*J#kVyc`$He;3~G&dvm
zM}2DH(Gv<a9kJ?CMoF3xPaJP?E)=Ng9PL*$lO}N-lOzevMhkDj+-i*wxLsN)&=?E?
z&j^a0E&O{%5G9ml#Vc>U!;SqR^GYM?lzw{cb|eL65oq~w?zjEkFJBpqUDVGk_T$}|
ztfkxX@kNcU7W+9$04{Lbjbti8You2AevGLkN(6YGhmexGC>G2u(pOmR#>L4%xmk;g
z$tk`c;>96WO0bsD4~c?^Mi_9$z?qFDUfY_nJ1&@I1;wIyWiX$Ui5Bg(CHBT65Q-)N
z;e5i1sX)2V@dsC9&c}kJE<hQQxIq|(j(O6+_k7e{%%v2IIlSO$OS!X!Ix$8@4XU!@
ztxH$9HJC7~G@_a_$a42Uv?#JR#ue><1hW7>^2>QH4?Vw=d4Nxau8gbWlLEw1=Ea9&
zZX9!QlzD0JlaRO>F1isdp&t=1n4h(9W^<WWx2Eil3Z`k{ItJ^4!E8p_Y|&onus0rK
zq@s;SkdF?TH(N1}tR<X_B~h7!(j<*0(E;;Db1@J4eVDUgZY<@_Hr|ObL=DQS=FN+j
zxY-{&&#O6uG;=2#!n)mcW3fiK`otfx>(F!aUe7x%pFp^-Zz&d@3)h)>=(#z{+#S2$
zW?mjL7rv{fzuR*TJZ~jf=^i3OFC_98%#G#rMvvFGXY7tkW?4b0HM%Yt%u6!aroFMu
z{%C{{iWU*ke2-OAV;7!3uo`hbmNayE=y?zXG#ZWX>$$nZ^AR#?QkE5OUtV}#YDAeb
z$Zp(bZf<-2@$a?1|Mj63b+wXs+8qOlMf;r;0?%qfS>c5OzuQJCH!IE$0#_tEO%c*U
z=%FzLXsXl=<qbSV+-fqM?Ggk5^Z6W4BuF9Lv1unkq41=Kt!t|36kiGI3RG=CSbV8y
zB^_pEPGsi%+=D%?ZVlkRd%z1wM8K6xR~c-a<=H>}Z}971`%MZA_cmG#e)U(F?u`)b
zE}yukM=;r?Hn6s_LJ$d>%{H^?49^SDTH}iVB^9+U5rG@)6GstMT99Tbt*D78Q_Q@N
z=qxjzXT0{>o79WIH;cW!inAY0Vf)8<8Q()!i<bArB2)<16&v{$FOv9if>3T@u6*BR
z)Xg)H77;3pbj=QRQMhAUDG3`1WtJnQ8}Fh-gupc;cwvlG3Z>jJYCW5~f>TeTw7aSJ
zQqhQ-WL3ecGJNT5o9*F@nl@JX1bE!o+GT$1ED!$dFYx@UFEUPNoQe|i=f6s}Kg4S^
zIK7<Evl&ur;%<{DQN&S$Jj?O@(5=0NYXH!u1O!ql!XTtBa`G&r6*ZB0ik=Nn;Tn_4
zjJMu?m%OaqvBp8C;?b4-kMb4z=;HmECjt1Oho6K9fg24Ncy4feT7b03NMaNq+>IO|
zU86uFVU`sLA*iYf87)3XQQ?INLMb<arqa}D=H5(v52-;|Lm)jGQ9@A{99NRw=@z?V
zkIJ;MQsD`Yt-S&Bjnka}$)DvrZ@$iOGU8;EP`>m$^>B<A#T@U1ta=rmHH2|Ym?)w!
zrYI_cAOdkP(^#Ul+evec4gqyukmotAc)@(yM}%vPCo|rD=OSrdV(g;!xYoD{n?L%V
zJLdLx$GooG?nbRK^8za^Dv}G&516}=kYOWcmKAQ4RQbJ_*9+!=uMir9)%fawdC9Rz
z&^z5`ca&1uHdcC$`EH;2#u@JWso%#7ufNK0I^kp-Q-1fGRD%&-oN%HWv7$;mqY0Za
zVbUUwV)DFn%sHqrJMetLJPfGwe8Ieh$mi74K3=%SWIW}qcP^3@)gg1V!7ba5GXD`}
z|Ghj9!H<0WIQRh14Knoa@?0ExKF@D^9z^amrY_&#bAd$!9@2oo5UAUppNJK`sLkFe
zqqHr@9L3iDkaXiT=f3<ieD}@Q8B9i;j1#JtzIEvNu};XEUpwY;M3^)_z;l6KL_r<(
zyvTEcJ3L=!@?+rnhg#J6tDBU!wm{VG2~>d81|=$1kFU}>c9y8+W3qzlmoIY9#u8<k
zBNp``&0dqrbH}T)Hdt*iLXvcQR7Flz<<w=4F-`CkrmPTFAia<})`+4;)+N*J>%{Fg
zHfmBEgDGoFQG<3j53OOj86&O1Z}`Y=8^8Miv(@9Ad+2`prRKN3`kUzW291*&oI15k
zS6gh6)4peuB=#|NhOZR0u9;;OOY04$>73<O6KgePkrBobQb-z&7USs-#cWP@$<kJm
z-5Zx_G|sTU+vn!??&3{d@S_6$cYDQ~C&;f}1EG;Dm>Y|)EUTNVbk<K2Ha*ID%4BDY
z_0=v_Uf?QYEpaEo1Rm1&Txh4USRskq9ducujeFWQ=Aiw(LRe6KfC&xQ5=>1#*hNJl
zHf*4U8;|BbpF0`|q|b8VgN3jeqna&RD^2p%6P&yMUUKE}!iz5ubQ?sQOWd>M(~};a
zt!Xuv2m^_r!jl561$k*{2O(Km5Hym*;oU(PA*CdV6UNgSMViuCvP8zv-@QT7K0|+h
zz}C**2jm3)=(A6b-9vWy${};HU>^9ey17bw?F3;%QOxF~`@5{JcBqTewc=X{TQN59
zQNDNBmT#>jYPDRpm2taj*41dBYggMSy`#)E<#_1gZ^Fdw*Z~rWF&b$q8Zuxl@h}#`
zW)r{BWNEcYwtS4!=T1|5J}<xV8jU2Nwbr506!Ze0NLZRlLf{L8(5^GqsKeUcyetVC
z4OjD86$FuE9!Ck|$(%CF5K9mm&0y~qapx?9{Q+CsKfcW4&3nkNTzxNd4^}rE^Qhrb
z%x6sYc3E9<%nxFeqIQf4e5B_s;;z65LELJiixO>WbX{U?d?)j}JTJz5WZ<KM1S8$f
zytKk~DH=l}16C8yz0R5qyk?W7l_uHBF;1U5L*@Cr^xDfLjfm!Ymu6ei^L#>KXe1p%
zUm`_~rv%z4iV`|^Ge6?FV?LwEGTH~8@7*NnoTa}%V0(M_$CmjAdoEbpSf#ysoG?)o
z^BL*>9xJOI*U^eA8F8L_DBoXr?)G?{=cV&pFFaQQU44M((;;{sDvYtxW%?cUyeR`{
z958R<HySLjw#ZgCId$)8YTxJ8H(#e22gub9jVAQ`fJj0!N$@>^lp0S7j6ULdRT78>
zcCp(WMBxW}zQ0A>x%bE7-u|Hm2cG}ZS0D=Ug>h%<!WF1?mU=w;@Z%)@GSVn^ZfrBz
zAEAp1;VGik9z;GuTH+)mh$FltLWKcFNeBa)OFegVs}x3Sgp!LPxCWHMZ#1Z^#c#Ao
z8f}JqdrXIWNC|N(A?dWR7IbA%LSb#qyqMwHinz$=`hw)<1%B(V{CS2~-(aQJVs_&u
z<<1^SF=Htftgg1%Jhq9|7Eguveu&XNRS|(yRNCdv7^A7G^02Tmi4*F&LK}ndBvGK4
z?{8sk$zbTFLAdK+w}vCX{?WcpKJyc(RtxVab6{z?$GLm&BMFz0)?=`}&$K_LE=rW|
z5iNCGgn=;lfx-&{yeLElK1NAI6cTk>)Os;b(HMl1i{m$oP#!9d(Nf|?F+muUO{S#N
z5dupXheXW;fy8Kmv=VD-$}+{%B~ew-4J66_4c_>te~0PK3nZb(e0v)`8nNVC*4jQR
zOD$Gbmly5u0ZIkrxuvc{*RZ2YgmgEUx-PNS&DIUW$aOK*HAq1y1?6ZTtqTUD!GigZ
zKHvAdj(_S|RHuU{?1H)bSX%CK`s5jsXbFYSXm7}LICabek8r7tj6A@gJmCuG!vN`f
zSS68RNZ4*sX^k=ZHgl~VbI(J?5k^T=7~=T>Wtx)DCRnQpLjUkV3N2iho~@~?6yKBt
zbwM}uXis;z`qf`0+qp_0EwkMn>}<wb!(+1(vD|C0w6a7HhWK89ls@UaM(YqGB$cjQ
zEuYcUi^Zp79#U5oRat_t1QN<gAFT_9;~~@8j2~C#CqDgIRJ(nIxj1Bg=JZ*TXc>ve
zaCgXbI6;?g=O$WeBO@P+K`G%T3xpvuaLkW-ZWc%DcQ9ACnR{V`?}y~GIq7Wdm`8z2
zPY}W}w~l#LrFf<!tO~lJqB-8;@^Ahl((NmRN-*Eq!A>Wv#*&Rrz;ds_^6D}{7~mZ;
z*JvGLzbEsB=T!v)0twZ4aL9b}WA3@v?Ys~34yR6@AqkgI!e_KMVm6$j4?OR+4n3FZ
zF3;tG=S}B%z3^Oq56``Y=kqDn*7(5@&yDk(L(dC>x}+O<w5PjV|N5_y?_9&jFx%aM
z`Hc0Z&t^Abxtp-GvP2jyJQp7G@9DW~$_(Dm^YH*(7mR)kJpWLOy7kX|p7p1mM*2R=
z;^GRhoIiIAzs@OV6N*8f8yDWCmBfs;_fgssL?KZW(cju)`<)BScXr)&EgWuMS<k1a
zE&crwQn;fxyXcNVN`$BIqmaBRh&nx1Hcp_kjM@GURgvLE0llSFJTE{9OIf58Wy(CA
zQjRBh^9h?aBfk0y^H;yc{`O6-z4js<le5`tvAViSl0@jbAPhsSwe0Q=c<q%-^p@I;
zCqoyGyXfvYkX{Ht2<G#Q$#}waIL5Cb5k6Q;TDjS+Ke{i0?|rR4{VZ!wJn3%K3+A2@
zoIJ5iA`MkGBOmtJyLp{ZDboHJTh=JwF&yvj)4zJ1Y`>2#3Xaz02|uLNFd5Hr>E?1V
z)YxF9yIJ^QKxH&xt3#`|f-MWOQJ=cV@xp*syGP)M2w|zKoTALg^0_;v&t~-UA?EFG
zGk@bdq@#WMS1-~sIcx0(OG`^MTM0_xW>(gk{rwTwuI`dFLb800Ts%3}wR=+%2XQV!
zkmWhE>6H0+f~PGJ3XCDE3aYyL@n!zxr&)XKaa0r`t&2GDrQrBxkERbgo0AXw^tWyj
z2thWQVu}i}*iD-b`iyVhA|DLVMR~hxTMA?lQd^i!vqdwN`!mKG_xrw&7X)Z5L8C>|
z>SF7PVlrAV_i3~`_(6ogQdb3Km6PXl^n8Y&&uPy3<ga~$`CBhh&Bsi)uG6+9tL=nd
zuS=`lBnaH=v#u%z{W06yW1={qDAL0WuDW)~{)mHUA3<J}%w{vv$rP_PL{?(8AuaQd
zB<%Z#b*(=28P*<u;s|r_q~OG{B^n;+h3EZSHwlC!9ZoPsg;WZGVA}69x_OIyf8Zvt
z9F!a^Jg>}xx&2_~6^5YMqS@|2RZ&ca)Mbtr1T<S+{6o)gGp`rS+v$Mfjpv!aF-*X4
z006AfNkl<Z`69)1#Q4@#+NNZ+)nKXDqt$NW`*Oj&roTUC>(-Ds3CN3dk+1kc%nN4I
zDd}Y5nB!`~vNEUoC?-4nkgnAyKh5eV9><GsGrz<0d`>p(GuXO?F9rF5=SqNu>0rR*
z*7o=CTrit`2+ymEVlqaTM?CkP=X$}s$g@Mw8?!#yYu{l0_RCc1gz44|TDD}R<vee9
zZhKx;6@&hS-6Nh0@&3$(5ai|Eo(rteKNg;UsN+|=vci2o`}^6yc#-1T7Dz>)1j~(@
zy4pnsKGUu1c($gQ#Mr!Ga&sFSc$Bj_wk{FYGRrICQsXxoNF|Y;LMV^8*<k0|Hbq`6
z;!_mHn#1~*Ac@JwncFIjo2+h}V6?wYzS}2QULuMTysjimCrD#a+8~8>yLR)EQs;=-
zm?#fu9&3>Yf(Ost&&Emzi^p=Sg%+;%tG~0$?)HeK?gqV8*x8@bTsr1Dd#W0}*nSm4
zVvWXN$>(!dt5r2<ZuF>YNqra(@Z))T>+78Q{FfMBxk`TF3J3^2$x_Q=%@C~wlf9cL
zU6I5gCND_$`lKU8InU5#fv^Uf6~xP{hs?!-c@RhJrX^Wkx+75Onq-g?DTJF5T<2(A
z6DCcRlFY{g%HbHlV~N5TPe`(ShA<Xs3|NiPC1qJq);ZpMN>WFJenMeuj-R@R<+Tn<
z`Xtgd`jlD5sNbi*H>K5Fq16>k<|)0-%3;=!+j19+-C7qzP^1~zU_?VVNt&zV+68mp
z&XN1^qwjBQaN^mYV0i61`MZ}u2tr@6(v}D@rM58H-$K?Uaook^IeA@?`-&>fsLR5Q
zuFDFdrG@97T7(lTydY$n7BupbFq9w`%!Ncsfft1ovjS@kQNuA$XJhKg6xGDV5C}<;
zyP#sBHCR_bT^5dckj+U<OyEUib;0uHI-TVXffwLOi;+;IDU;!dL4QUgS*Fp3d6p5i
zmmS|)yEmZ)bF1ALwld4fhGP=lBx(wmZRyI)Kc>KukN@&F*16}ipJ#aW8u`1IK?)+N
zSZ)K1Z!_0-F`rD1c&@QoK~x=Kt{iheS}@N_q7PuMX9d<6!bS_JB<XZaHJakJAqpbA
z)`Gb)2(1x!Gf!uS%yV6`dVGWKY8PMm#1f3O3+6)x`!gDiWtwf6W*Obi2Qk+!Q|u1r
zilSc3!sf@B`EAd$qn>wedp_OYLg<n>>Y(rN+?7f>&vWN_v*8x`hn@!?(sMyJn;h{x
zJoH>FJh!$!WL_P3ZW96(lGQoOn`?BIy98c<5*8~hMLK6X95L+ANE*vDS}@NuI^WCl
z`h7gFe>^<@P&;)}L2%Cr?){mcLo`FAw6vN5>&MzeOEFsHjQ96hUtdM~0_g*$a*-Fh
zLdgIRK{cOJj7R9q4STOD7a<{h#rCA4Kg|$QVQ|NBSeHE|0^iLCMPO`&4ShNr$BDg=
zYB<1V8D0<&HQR(?g0-M5=(?aPW@M9$Mx#Zi+u{DCO=I&OUX@L5Wnk)psxHw&Qr9Kf
zWI`v2dFY{IjHVN$3{bvDSrizpF~TmwIl<H}S4=5IZCqHapU$YW37!fN%3Gkn`}p-o
z{}P@gIdzH?Kk<3|r8Y`hI<0`!^@O19QJai(Jf_=gqXLOgV5^dHGQpGuN_cn(>iLX(
zG<11(RfVofly$ptdvncTnmg%{ZWq+znEO6n7%eIh4Kj*oF0B%IKJ{dR%`*JJCyJYd
zVGIlAMpsm2N|9H@Ns~qtb59tPtZi~F>9JjV7`>RWh;lQ0r&HQV#Oc$k<Yfjzp_E5i
zmKd$Ef<xxEc1H<VsH)t}7@W`0=>#P`gi=3dq;Fp$@MxSl&GBcRC0OpFr0alOUW@S>
zf?ChXCu7>37G9uS5=K>0j3?AZfkdIKpqfp|M?>m#jwwrYT_LSIq8rQ&!&%`TiiBeh
z)}_@ij_|<cGpabE(O$w=K6*C8<~hC}5X23Fz%iHB)mPS4PE~56xIw5qHvE9b>N>ma
z74~e1HjCM~!j1mQrc;_x$nkq5eRWvV-~0DR1SyejC8P!E5EvlcAl=;^qedzr(%mpx
zq+2?rYZB5O8(pIr<G1hgTo-@ueQwUV&bi|i>8wt-RN7#9@7}k@3$*6sjg)bMsj_Nm
zq<%}e{^C-K_@EKQa6;#~7*&}%5U~Uy;_&^_KjVN|RmF6ts_3HGFq%1~0kx6j0T+z-
z=hdwC0lX(FaiUhTgw3Zgxo_WOQER`Q-RB@=r4`Ft9|fow^TX)ju@zNO$;^c?AP1I1
zRZZhRV++U~E|e_1e@*};qtExI4F<M}UiM~XzYfaGDW0&~_lnhAH@Ur`tv%|6Tgi*L
zVRF>U`9&4X051<ReLA9GwA@^5%)SAqrzh8e@ocUEoG<ew9Vw>9@UK=y4?!d;O#xHz
z``0<-cilbaG%qc;K&mEtscCRPV)V5%L=Rt#9F06g)2ZPgR*u9IaYs~&Oa#MlA+z;x
zW28ObvwMt4uR>qFo%o++J8?y8cD4M&zoyE6y;av~i&Eb^woGHClAD(%;8-TxX0%TC
zsf*B|FS+M?@_#MrO(p$ZQTM^z7*hp@oCMTr)6q((%yv<P1=zqgG^k2KcN&+GRB7KB
zE};a|CSRq0263iK-pgo=%nW^43z%}>f{NdlsJE<Gu$V$4NWSl|EgM>M65MGIc_NN^
z91m<SWIc{wIq10_>9Flx2Ccp&0F-CO52*}WIhSHL(%;9O5%_Z^!04v7hfj#~v4RA_
zNJe&bF*y<}K_Mf4y=;4{5qaVg5>$soP9c?+E(}Y;F|nzGvHDH$%qw~}7KZ3)f*jcX
zOq^chn4KZJZG`Al_hcX;kb~(%?CKpQeLWQcO>m3!eq`z_n~1j%qIJzXVgaiJhC(c!
zl+uijy=r-dQY-fIH}UD`teyDSkwV!#KY|mRM1k(aqT3hQG=d#U(53L7gmqkHM_o+u
zzL%htcwXT+71z5&x2|tR;(*^z9-Zi9gb`H|*Qj*^lt|d-1q~UyUN!&Ton`yMfSUes
z#bPJFiu1`RS?Y6ZBGA;BVooh4wEbU$KaHjl14w4k*Sq!JKta!p31Qx6OqZ(u*OBn<
zTL65-GmwkX)%S}emK1%OI;txy>e7FT6Oi&+XV0tM#3$@c?IhmH4y8>sp#a5B@|B*D
zUg9@8cAP+oUXDy7r?shi6NGk^?E8_~R4+kbaXwV~^2C>5-uoWeYKn%d(fh!XZ@Cwx
zP7o0t7souRnX;^ZIZ8s0K1~SGAt4@+&t!2K4ix$M77vw$y@MazTGX7HpPsHrmdS2L
z&Nbp&o_@^q{o!AE9H(WSB@DHQ)PtqxUBp(*kGzEH*7-YQT!>|IEw8)r<#T0bU`}jE
zWjhiYz;{4h3&4ctYch{oMVr%0W2H{Iff&pFR>k5q;qA806HwAS|M;s+ooItC5$PT-
zAOpb%cW}UqYrvp2MmIS@i-;eR5Nh$c_yn+#Z|$e!5eYNGg_KJ=9odSy)zC>v>IN%7
z#u>oW3ww{nfkCZ1dQQ_s{Zm9NJ$Inr>)leVt2@^&vW-(w@a%$rGBSiZ+0!T2tM65?
zS{{qo$Hq1%-_IZgmkQ2wC#nV0xXw!oC^5BdUhpTV)NR(WN#1ditL@9ys8!cf6zws|
z02uPr6cLG^TpOAi<O)!HAj{^@Lob9diG6%9Ib3S~ItU!(i9n8OyGFd6GR$5}iM-34
zWJdRhv8^4h?h*?xOYCPVWIwP}G=32IDjE$>E2`qZ=O}f};E($mav;)yr|j{?ssYO%
zQQyu%!G@iow}iMVTk7x(^J^a5D^xe0`~g&5w5KnkNP)>rk_A(&BTZFRO&qjI5)&Rs
zb_k_#v>W%{TU0S!i7mL-5j=h74&XlWOv3mfSa9-Q@H2srXyOzbJhvBLF_}4Uofk|X
zBe@SfdnF$BgdK)VtXui{&v8a9DBdZ9>wNNP;lAjsey*uxDYtH#-nHdGeMq1FOAMWA
zq48E3$zk1}al@E0MjL~qNlmYg_f_oBm4MfWu|*$R-<e%njRkt=p-N24=jD&t?5eJl
zFCW4vHC(ZbTu`9K=za`0%omY)YjUE{Z|Xdh8fq*pS&ubrL-3NE51|42qSkNDY+O<4
z7NfmDeVgd^H95FNRGF0?D{M3I9*GQ}b`Rfhn704D4{poIqt*i*917N-)vq>{7|1Oi
z6E+*YhG`8OgD9*#CZFQ(ix!7$(i9~P=6t!i9$v~B$QZPZP~XlxUs%&rSJhu!u`G4?
zq*ynP#+9pUb7l$|MNSmk5_4g>h16nBz{Cfw6@=ey^u2Yukl=v(Z-F{643u;}K@~ky
zLHkO#X!)qsY?=Gz+#fVLoAmVi*0@|OF`#~6fhS+JzF(|hV-#`I^%RzA7x3~sNaez6
zYE?<mYYzU8*$Do%52ZyLyaM0+3nYtFbJe)|26lphg2;ivEnvHYci<M0u3NwLjeKsJ
zcYf<7zED#iSJR!nN>N5!r~#N*Fd+PdZ0DY2Jn|Osi#@t8QG-fzUSfGdVy;MygBe@l
zU{-&9bB;WJOTRMi9bTI^Fe7EW@=CB{H$_~^-g;Lq^6Q|z9!o@%jAGd5JoA|idg>Em
zt2&FYlQnU1Hl4ymx%X+9U`&PAL<|$#0GimpSO?PBzkl01X-LWx7yhbDE+Cve<<Oe2
z{VMb5%D6Ott-RVpRiN|v3l7cmYMI|*S<eLUd3Iy3bfIVP4?hJ&NlpAo_5p4*5m!hk
zbP<NmM2Z$B&>z>qYo4S%kJn0O$U{rIFA&2e-Oi5h{?xA;-SCQR%|B9NLKOLZ<ttf?
zK>l7`*zWnmmTWo>;|#8S4eibVpL!$F-CQHvW7vz~_VEFB)kEjd#7)^6T%04Hl`8ER
zgBsPq^2`Lx-4w#Y=FB?VS_!$s*Crry<L%CKW05w4-3J3U0$uaz7}j^UjK47aG-5v&
zddJa*HyMhJabPgo5-Rd7b_~|5eI+0C^8;MY5d#p%e4R+H;56TLzi)gxO11!9nt<<Z
z;3ODR;lcp!_8%UR{bd&&u2Z?1Xicjox}0<hmkVnJirRW_P$a;&`dKZ|qXdh*<!Jci
z&z<PX4r-o3zsAjhMKT7rccp>d&8pY!vX6O91p=%a*BA^OzmNm1O|~Nn$62hWcN_Ng
zaZ{MVU#w^82fd`sKZ>aA;}0olnSXdzgI?b3_pMB<(Rnz>8H7iRK$+wg&@~h)*GebI
zhHsTL=LL3<f=(y@Nho54Ltf_lWj=xflVc#>tye?7#@juw55yl>vshWHD)DauGs<nM
zUzxls&vd`VkC9wRoO}yp3Mz4I>}pXrB{>a$AtXJ^bR6?%E_kypb9J%U4NKy~1k()M
z!SyX{pe)~VR4er$_TGM^UJY`xOv-0%Qq+wJH2VNeK`mOv{3El<Voyxl-5Ph<&_2~i
zNm(9_%*SBVLW?IFq@igwl-Df=ZkVW<9~XeeAE9v?@{rQ`*u1ehL(TOi4d_Yu_i>=u
z+M2|WK;av?OiLDZJ?drZ!JJV27|)G=-pRT07(n97XIJk>h*o8Quz%+UlwZC}cm+JY
zS=CC-f+H3qFO8EY<$V3bDO3`Fe%{z?d!mD`?tyXxLm-%<Rlm^?CEvJ$X#B1$ro`^i
zGkT`C{`X-rc`+`wMQ@Z&>RK2+0P~II@Hzyj{-CgA8Q=z4TS?%S_+%l{+OjeT<CS29
zNL;DqeGWxC0_C=z0oe7q+2Z!da&p9y)nX@y6$lHfF-$t#46%-1@K;N+8M2R~dj4$k
zRtg*|MlS;owDbgqs8Y`uf@mxM<EXg#E2V`~r(0CF3DURu`?KFMj8<;W2D_R5&3k8k
zjZ~%~I7Pxuy&!BZYsdO_O<iOJJ7TXpGJkg$ij7=;-rBIf9FskLp2%o*-%vMKR%PK0
zpWpn)r!5}l96u_T_@+EU1}o;DfP-u1B;6Ishx5Pv$@nM%7f%pjE>1mc3qyQTEvIzP
zhDuQiT#8D%vwx~I#%}pZnb2AXw4FUhf()#ZbVEsg(3!(m4btP}j@OH3#B?*L2JjL<
z0`juoa*!4iTcNKQ=jS=y#%>iw-a(6B{Id?l6L<`eE*TtL(EkE%F+cV3<#E-k`fXk<
zWq33*St=Ym_(gK}^2AY^MCOarSVixD#iImPylP-^HIIkaOREGA$#C`HY~rIF#GmI&
z6FKtr-8rlLtz|KFzWfr#qwl{l70r`$zE5dt3&^MQIWnzL6dR_He_wiA^q<hGAu7)i
ze-v{4I?Y(9UN%kVAZYI`dgZT-!ets*lgLV@+o;S6cRU8!HhN841l*-OThL1Yf<NO}
zQ`CE$nx=cj9iC|}<(9fC<g!AaNF-hxb9@zOp_L|!;r@>lUFL*r$y!Aq&k}diJS1FS
zW<me6Kg|rhe&*k*{zvY4P3k(i_ti$|Z;{<Rev&c<I#@105EMoosR33N{O>GOhQwGB
zG{7RV*@i{EFoYJwJ<uxb`F>1xh};ae7^co^7?OJ7F6VU&a4|y-=VCdh4Wi3{og$?-
zOmb}&#~_Q$($Kp^;nZ3BC!-buGLg9OQng6k4%Lo@!j%)E?}R#8T4WdoV4gh4()yYe
zKb6)lli{<gymv@ohE?}D;Pq1LEr+sP@l8v)^N^>|dnFP$67k5JA^CaLH6<=Ga^f+7
zA-vvI>9sr{0~gXfpda(QhD3@=58wMsq)l8`3@E(XB)$N?$|C_fq!=vMvMv{W(R!AG
ziXte?tSj)_^q>$*9Ybq%F)(%%6@}*P4U<MuUHj)UHt^IJZ>1CLSi1!77Wqu0=!37w
z$s8;v2Ns?#-5A>r*^ZUcl_QjRn@zoW_j#b5Tf(%xl7X&0N*x^;1Yjv!i26Z1+L=!j
zNC%g-cOxJYFtVOq0pH$6(r&i3p0pXPu8@O^0iO;*cfnwC3a31n$R0Q)C$&s1**2Z^
zO89xM@vYg$X1-bXCLm-JTh1u~r4N8T51GLx2_Ba}K+CiBV;-KROyk1ZH33KZg1Va{
z24-9Y^$)%5eMOo&-ifSnYLK-5%xj>7@G`@NMLRlPM_OGLxr3kos|(qe%}d6lV~T5)
zZTw)L*M@&~=YeLBh+k_}nadb&T;J*;8;M!nucX9peuQJ?NR>y^h~)GZ@mTVbm95PC
zs_E)0&BmQvp;iu*O#_uMm;AzrMw6IfqzT82BOUvbNtCwmYv2zYfa-2?@IE|=8J@gS
zER#LkiUd8axN*NdW(cP<5FRQYmKa=FSl@dvAIRD0>KCm&ZDVND?S&u^WvpGXxC^z7
ztDhj3KZm1K6H=ap3jA_7v|^#ruEPvUQcADkN3}bXK=J#<oxq1I$oX+2#QhWjnimwb
zZ(c$Frf#rH0#ei1jU8g>e%$uGuikiY>AskRwfUTo4O((v;2)8hS9^Eh&5MZ%6){h?
z(-}(0)y*+!Gwpd%ABM8NP8u8ZfBWQLno8m5d~!%$Bp2fhD42q|vP?M^{gw8+yt%0S
zt?jLt7GRw<Y>c@}>R0q0>jPCcqwZH7eRZkQ!nt1~zzfDSCWw~Y&I?6`qOLjFwdemb
zxYijVTB+JsYPl9q-_*}<R%(+<gx$J>Q<O7o2ddw!nTFYk-~OkYVZ7F+$cTO-<L<2E
z!Q2zgx$$0=X=hF#z|Z4z6!S)#ihMu0_@gjQ@U8|8=!)=E7?hmWtRs^5Mv~x5dOQdW
zi)FD|_lJEn4ZL771>Fm9nzr}b^A}Yd(sBan$ssezjIJ{reEp?Np}I1=I@)ltlJXz@
zn_UwLa|{M2@&CeFCB-Q<VDQ2C2zs2<7<SE>k8{mGt2k`PxFKS;A_c!BH3%OB@XTh<
zKQK}qVarq(^8Wg=1&Z$EwFV9O@iN9si;RnPKGJSJT#Gf;oCoIr(}Ed@JrkgS{=ACi
zAo#YpUQH?E31^pU|7?Ccx!35wt{gwZSuBp6kAuREvL=Sma`Rly%f;=l{f^Onc)G;-
zwSgBgxgBTXQ~46Z(0VJ5+M1bDFRZX$Z%gv{dux%cnLXvs>-m+=d+3UBt6$dInn`rg
zW)Poh2-XC(M~ZD4F=s8oZ(^z1qU^Kxs287p)pB)Sz#KanN)IIW*ak>gb28a0>o`QO
zvP>eRp)^e2%x1A(N5O1;4AnnZ7xKH-)Rm2}E2mFVCcJ~+c;a98zBKM>S2oFDg;?#)
zf#y#GGEug$5hcI;;A<u*s3vhs$5lH|{=)=lIPjLaTYw*JyuN08{Q*h{CE9!Ri=3!!
zd<zP^hnwwoW1hlLYLGjN7d!t8>OcfiSyaI4dD&?Wcn7@L##r~;Q8f+OQROu4oLFBM
zm(@V-`_TwE=u%47%KycQ>dZ#v{CC|GCZLHz*k=N!-9{eo3o$?~o{bAz!&lI0M?e4`
z4QZj?G!}o9Uf~-tAzzT|=;f6dc%!vGL2wG*JbPR3xzP__KA$h`hNC!T3ohqZVl85N
z?aUV+bJp<B0fyYP>((E!@5ab`2D)Gn;91vg1IQWv8@5FVmF=?@fnKJuqsR-QDDr|R
z>R%s~muGSbc-9wtK93P8{Y+i*s<y$;V9KlgkLIyP)Z4Mt&sczRf^Hbjx|4J@-n@O4
z&n#2P{2b`pc@ToJNDansmGE?*>dNkJx7uAWL+USX12EdTJpHh6nOV`?H^o{?8JdA-
zpvgdEn%}Tq{v(avk!jxnY8$O2B=5z$T*j)od?JwCI(dLeTiED@-!=JWIrT@upjG}8
z#~Grr=TCGM-3gzNR=+Dg@#nv>UrIM;(p*dz$CFqE6NK)G@u^fCD5pqfrKXc!l#|!#
zJ!Y2TA=C2jIP!oA$xVig)#0{f%B%K7gV<mgVK{bm{7Sl5JakK5yszh>7U6`_X1(}V
z8=Thj)Pf{4ds<$|oSLOw#xqmGkYi{X<SV!&%6oGm{i6Nz9ENzzxe`;I(zoCCCqA?K
z=m|vv?Hr$#I#~<Gc#A&swg{XI$m8IXXt)1V!3!JZ6_QuJa?oNLzhV)NBj|KglE(kE
zM%B0u{+lQe{MMoYE!Ja_te-azk1b~6BysSg;8{oh&4Yvga60$zeq5v6kZtT5#8B;D
zevG2*<OSjC8>)d9ylTJL6L4FGj@{HXG(8|u%suY-D2AgN>8xflR{o1*#gmlg^t*|d
z_{pn4T4g6xlf2!lL;1>oyvfL=pH+&@yCbRJ5e$MhBBmt(u!p#u&hvg;#KSYfDG*M*
zu{%FS^ADemcoffS0n6$luD6n(gt8D{$}qg<yh^ryG3lCwB1Ui?<Vr@FGE~5xu>^|{
z*8vkhgl7F~vZQ4atMU~7I%wP=anc}WWO?ZQtC$8wL|`DgyU+ba)j;`1-%ZV7kja$5
za%|=VjVluZWf<!x9UJzDX%2v-ROMMa1Px;3$f-_=-+a;wKB6T)czY=9pfTgX^n-C|
zK-Ejf6C`05Vi&EMOq_#9m6I$Yq@p?4lpNk~CI6@DhEkp$Mk_qXkYi`CwrFkR5@G6w
z`pZ&lk$yXff~RVN1Bm&v8zv~@hEOJOc_=IR2v|2#p!>So4S2nbciC`Pn7U45b4vM5
zZ$5Dn8%@uh?ht3gQ0SHR_BO$rtOb25S3a%}RU|E0n7GiW$mA(_dkveWWXOz_Q~Q8S
zbNM!Eu!AO%N2n(qKP)|OM4VmJk1wray1sSqEMCOQ*S<#Mv1<9W{%BtVwsf#s;=w1M
z2!YcQ9-&nTI~e?R9sI%AKZNJC2Xqj!39$=JjwT+aVjNE3<|@rOazSrgwxfywT-_Of
zQ9@^xC?CXXj@grsjdMnYvkt@`Gii743@Y!V^6vmC7pYDQweT5bu8V&g_b%Pnh?Z8L
z9Wq%R<ln5vc@XDVE-L}6!`h$3W(8EW0nC7%O3vZN`^Enx4LG|FoUsa6w}NimbW-AW
z@-k83En?-uCR+z^Lq)}4KVW$;A&{ghu(g8=M&mhRqBy~0Pw1J)BG1Z2$uJ#dn=6xC
zfC3qiAD^jQr0(9yNddLqT^JMKB)Bx9QbFOzO+}j3Pm{~_R#jEiGe^~RO-(9gv-Htc
z0LZY`nAjYZs)lv!cdHtcjv@Bi?lGbKfOxl%#}(r?pS-G8vEI$%PVu>-H|~uWhs$}g
z>@in?dR{*mW2yG*2XzGw>_Zh`19>&nl-fpPO~xcrp&B@pcO(b7tXOQaYJ4Qe6w2ei
z-d^pTtadyLKBs%7(~1pNET53&c(EGCJpozQv%?Pj42OziXMr%I+?*5m$e8SM9tB}L
zHog*DD!~MdFIC%b?9VM)C%5g>F7Hh3-}Zr)I)9WmkMg+8@7?1ryH<hqoe((7Gi#n{
z*ojfcUEHglGyK4BPI|t-oF*%|dj4p$JBzT?lKhdWbHO!j4J7XH`d)uM8uBQVX!iIq
zxnvMJ<T1aH(D?iJ=*7<P!N=<_U`rtUQ`c!j$AR)W+9Du0RN`_NNg;6ylo($KhWXyP
zA!$RBYQ=BkWP@lBbP2FkXTVXm38P<k0wEDQIxp6$xtj~C9a|AA7HVfGLP;}=dO1D6
zhNl()4&Tcw;;SoILd*#n!R*@AN6~>~?m=f1<dw53qSP{#F8nlFjq<I_OtFA5Uf3Y}
znJ1uU@Z}oy)iTc3LQR-KJ0z_XxEeGvd}};6X<Hk2j%TRHYQA?YOvPTwZ#@Y;q!O}Y
zlepJ{)O9a0K!DuYkf1F^6i?gQU@9$Uff7oVv^VjgFZjO}mC?$U2kptP*kABXheB@O
z>FNr7@<GrD&L=dsw@)oUh)mjsb&l+NrRy6QFhrq@_I^?iG<7Ji`aVA)=JbM-Vcj<Z
z$f=FT0L0V+*y>X@$L@0<udn<z72C(|Gb`RYC3-RZEzP3xhPN7Neu~?f)Y2xpu6j3{
z90p2W2A`msM4;2hA7Xlmr$w?-HTd(bt`Kgu`2s?%t$GRC87KZIX+;P2?;Q*K_^uw^
zK`Eu_vVIR(z>%3Y_Ax#2Z{)3Sj?)DD7vkFps=&g^K$|p~`6{0demUU;vHSmq)i^pl
zqeXT{)x`!1Bv^9Djk^N#p5i5;7_)|TM+UA3(Jj7myHUe)5BnEP+g(BNcxs-Op3%+C
zYzKhS5v`(6mY-;56tYp)jelL66^U-^yFFs%X`j%|9{OP*WDcshKs1`|I6;EKPV%~B
zG4JrRQD_4)2)tg3BeeJBu(44ED4rEn{Csb(!q39}uk^1C(i9rAm>~^)5~#`Ii_JM2
z@sWxfHWmErYsi=+a}>SsySA<;B<4{>Xumna3e*CV&uS4F7<aKYcjc1^;6f2jj>g4m
z)IGt-t72!czNxdI4B%;k;p*1T8lGF3;rdto<M!7!x1||UH&=W%q69W^|Eo7!^JF>0
zYVL>uw)us49@BS??F~w~11h#w*>Pl*mDKz>AGa5SP(t?bbB2WTJ+7c6SkJ}V=X2U8
z;_io4p~bbM=fq4=$e;$KRLZZr5=|&wWPDoZ;Bd8{;PC}Gr{Ob=jVt{P*+k_Ec%v8O
znmhvHl)u`3<-t1QGU<LM64v)|sw*G1-3vVLdI<?&Wd9iM#@*ge9NOG?J7dUgcH+~@
zI|kalfWMWf4;6KP?YDh^nFn*|YgYCRcw>YK9-ad7wW|Oq+yiG|0|*j0NDka#b;F87
z8H>s2@dvc-I&1G`&%f;aR&=U3L2QV;=N#01J?VA{2|fui_3MtAr~UZ<2(Le~xPQv6
zP*oPxui9`r7VV)Y;a~I<yo(5**Nh!fv7<yj@Xmi0&!+oo*s)6IrctMFEgVd+ofX~0
ztzAktOUX=bykY<yyyUMm^G`mh<}rJoT*$H(p6@Niec7)MziWi)Kla-#1>e51v<Lm|
zR;?trWM3#SuP*(-{sxD%^z!q4fd&6@akfino$`$7*5atvL<`s7OAEVCP`9VakTJdB
z4Pr*T1dfH=PTxEh@oR$Ot=J5+E|n11EHztWO$;~cNx#D+Gcb419!5n2^Jn;|oy255
z691V4hDx918l%ccq&}NkBif!1C9FPCT=c%ghqpn3qfq)tiRRN5j3x4oYddAZP&>cv
zQI>n!=i#2o5R{&~5y}`X*j(KkQ=Ld|fx;-PL8h9SiL|2di~sj4F>8v=1%ZEFYUz8w
zOfeHl8t=D9r_8H8L`9X_K4Ee7>M@mmZ)mXgRx|N4UNYEr2_Sp^_BhHyX{2|P^^;Oo
zqrOYDbPX4@U!T&HA)$zx6mZ4|o$i?|zr&t?OID?u#{QL5s`b6eo>jGt2+gL@3VMCv
zl9A2cqQ}gU+w5vxtSy__Q}x;9f!$fzTGW4i<&8P3&*xpvYpbyS2x{j9r7?#wStU-2
zKh_njTyI9AU+9!#lK>tg@;Avtw(x;Bh?h-<r@u^;2Au_=Zpaxk+fP&;Hxt4%L08zP
zqR;>3sbR8_N>Tt3;1onZEg}qXvuK3~GrK26M)q7kZ}kr<M@pB4xK9WSIaW2H$AvF@
zCDwmln&ZI-XjhlpvB{biEq|;^Om%>QSDS|ncRa_!$I-OM7`~*BVO8K<Rupgsa-Lnc
z_8J_QH%>Gp2WIm@aArVLjMe+qjJYA_6jfY&oo!QT9$z=nPJX)&%g308xBE=t(i4u!
zv(na03`vKgrVM6=b`Dl4hzHT8d31QuZga+lEGS(~s9!5O#V*0)!89Rf;kMO%3&-dw
z)e`^<Woxs3JsVZh-l}ep!0M#BuPA;y$aQ|7f5%$}$u0ZmGt7fBO3zWfs}8w4LY25`
zV8CP$jCOZD;LOSfL<+f$Lg_Abodw6glR)`ZijPS=+4YQZJ%^cX#R+YQ1mZn!W^w^C
z((AImtZ?!YvhvB_)9@N|=TEUUp#5_Yg~GkI?q2}oxo^*`P08o(L)*k~?I1k)<RwqS
zjT|h>xN@_(r=I;!__QqgXFRRleS1;(zC6Rr3=<eqpgR3n6?tidtzBPWT<N5h^}@!_
zwl|F&u)(uVjLWxqYO$Kh|BLu0uy>OBe@bw1)qSCDAKR?)W}}MQ7g=f}r-160ZCt_N
ze8eTh{oy#MY58>VADpCdDctN{J|yzOZ=zhw&TkVb-+H_LsBdaI1P|OU8O;iS)VH;b
z7AFLUh9K0DY$3N9&*E-fPgD5c+Nf6|o6r>l2jFd$VF3v+RRT_XL@yB7ngbj9hQdGt
zw(m>3XvoRI>W#^sro7{2B@R=7+Wl(5Pq)21$3whfIY-7>=)D*sXGix%_s4&MC^(M;
zNA>i|Dr3*hn@Isl-Sp1;L#9Al5tEh_|H}y;HL%8NXZLu7Z*5((MA=&iXd%@YMG!Ju
zE&Iz*F-CBgoPRe%ez$7ZGg8h6JG}}N-)lim*Yahld;&q(rt$;6Hr}oeAC7$h(<bs8
z<@P+kh^PJEHUjz?dTyK$PtOOmmqR&j8Z;`44_GxDvqfsuanqy~oEmK<l;0;JgA5yb
zwL<4ITx*;JhyPgqenhVv*Os!Z!1)%C!gWa{@r3}xk0Q{eHzD&MRZ1yIu@&NGoo9v^
zZ>U3$0`K;2RN@X@gM%VL>E=QB$S%IIZJs&5^mqiXru)*Ne+?V<$Eb-4yV;v>ZIyyx
z!Nf5<G7(2-VhIFaY06kAs3w$D@rBd8%IKCZvTIt7uGFhJzeV08vgC^ow>_%CZ&rjr
zuxhu<DQcmle>7&#<h6BOVtfBOEjwn=fA%vBr=Z1@HiBY9&O&Z!F9uk|<D!l0%j?Lj
z@P_o)gJ32@vq3ecAz>THLy4U}Oqw0ob1yA4Ngh0aFNRLECtp=lbzX+d3yHy`^-p}>
zlz2!0-LA#ZF2!sH@GX3eG*xj()QNC2$+XXdV6lGDG!Ke;&F~ASWIhu~Q9CO|_g5#P
z@7<)m%XzB%%iJXNS5-KfdQ=iznX&12H|}u<aw?gpJXvp0m*6bFs8b2wq?fCGEoWo<
zvJBN9aFelrv(zcRFwAY*@w@$zR79M|B>50U>=l8OMKt9TZte6Tr*r@wm+Oq-iuvt{
z8*#R9^>4^^<$}Vj@GW@59DXNaWJK(0x&eFGdw1EVYTADFN_53)T|wE$DUP2=_M0)k
zkV|e?hQw9O^M`!@!^S(s*uH|J<cgZiNjp@thuYcA3^v))J-*O$@x1fxk#-B1|HO+g
zcJ=Axlov;qf@wA2%m7&(0^iu2IG9BUg<SOP4Q81BtdT%QtOaBq;n05RPAJz3|Cpp{
zj-MU+!S}dYp14z&II5IWd4b;U<B)zTNdX|01v_@2BGjN7dn|d{K8jqvGIrs;l$b?!
zyfDx+%(g|*;B&U*jLWR4D-r8pcR21Jv}sz7Ow3byxH~=eGDWQW@TxbDmEQp#N*-dr
zYHL_^x$@|^b1$=H<%Lk~MoEQI{Kl*m$eYpxe{tm<@aRE&&^-pF%QZMPAKo&XytYgi
z`#HYyPw?++IPCr)AAy2d?KYyvp}RPaRnSdpz6uks?S9}jGFm`c%e%o5hg7$DAwSB%
z2!4FPIW{U6z9^BOvA_H)jbGp1IWbCnd#?99w6VFu6&P7`x9@sKn+@8us35T*n+S@)
zL0|xb!iDGlX+VcCXy5XZQ?!|**j5|wYIqG2uqp{>y#~7M5~h!Tv+<4t!iF?6PCbdu
zaBeV+{r0}z-_Qf$G*BANPtdTK+Sx;4$eDQ!tZDg$P|X^uqooEmwe=2)pceXm?26w8
za>=*!i_zok8J)TbZ-Fv(EYrn(j#VN*{b0Vc3JN|gdu9R$A2y;Z#{RDHVBqqUE2gjQ
zy7}<GAWyEz@1XFW!Ju2QiP`+XbWGa7*wQ@#zyTo(9*y65aC%0@C(tDg^3v?-sPjDU
z>`OZBbN}fZD6|V~RUr+r*`$9XmcQMc{UZYQ@CNeYpm6RRj0tW?oK9LiO~e(an+vn;
z=g4tCyrU`$z}BHxK;a01<&yh~HxZlLmG`$7tLuCBIVe=2N6gLGD^a)LoWB$H%~B{G
zXGNWix1!TFAn&O5)5acAi^n1Hy?*#w^+6$Qb??3netVI|^%RT>i!#)D3v>WDyS)YN
zk;3lTYh_TmD{$jJ(r^1h9hLVaG&HnF)sn>7#zfMBGXJI7oC*hr_8^se23oqG-bEMP
zDYOQ+W{d1tzKg9UXs6>xb;&UCiEQXAM=FU5@dSt<_!nr38fE)cjP~CrFo9kM+xO!A
z8L6`IXOAP?%(~bXnHsFJ!>>BUO>HEiqE`4Slp4A{LZ6UdC$%tcC49IN<c3FLe)yVo
z^hHeyh)ialsqyXZk|-DF@MqKXQ#qP&{r-euV~}DfY0kuF!Uwvr8JIQevwx}d*T@KF
zs5|}E3^uDSu4wVmVsSH0Xni*?W}$A`N)Wsi73y~-h$0n5#za_z#w;am2FroRckYif
zx_^+PcOtfZm)6@xCr^%6m)l0I0Y5j+;0PfwCcg&B%<%xqrR#iKnd`i~Ok$k&iQ!qq
zPHiJH6^%g>yN^6S==(FW7QuwP%D->yygWRLp(DAd-`mqKCLgR5kkY&RDPu<|Wgn<v
zh1;TEzij^WA0;~LYaK(2aViO%Uoms48DDVdnY?w101>a>zo#TZ7j8~n6xsA|?{8iC
zV<foQH#S7oFXW|%tG3!9Gq?ZM+4Jb~>tFmI)GyqeXoZC(YwJ~u*-WCoPVdXgyXZLJ
zFnnRX{9t!;v$BXO5ykVW9FW~b%4vtuwe%)o<C!Asnt<0lz$xb>Pu#|W5|4S`Iab0i
zQm>|K9Q7bH?T_JMP;4EFm0W&$KL4Yz0lSCgTX7#ZO01SjtlA0Y)c3Ruw}77oWs`_x
zUns)K7HUkk_Vi45M^WH$y>7%~+jDnhR6ju_dK}|cSr39~`GoBGR^pl54|(s#)zB@j
z1|3p`)GhNYkp7$?EE*yv{mJ_4Xsr3{@x^<rFS;o2IIVv*@f9O=Iv&<Pf66IEokfOw
ze5JLvXPYjCR0pOlU4B6Nf5l-^rdH-}7tdi~l3b*7^upfe3=kStz_Va5-xX-(6;|U~
zVs`j(%U#Q^8s+77ZZmb<k8%7`T^I_MpE0rkr+L|~ps4kPyhjY$Bkm!+_WTSGlMGkE
zW%=uktu*R0PX;au3T2y*g#@fyN;s*SZ$*Qs*fAhtw*o`6gvAGm7uGT&`HdSJunw-k
z432zabee@)GuZZ^FX{r!@I%zAUmiD_ZKGa;!X@YTVzRXhdv@KIDf~uw$d2b*vn^D$
zJ1c94NX*jn#u&0>KvCKY;5bJA{H8-eUFMvpwlqX<xoZPmZ7WqWbkFAj)~x!=`pk|2
zRrz1J9f<`x?j&>as2Kq4*wini)7(Lxp{5ORu}s|5FS|B<$Ne3PnE4`Ct2|u$Q@(z?
z#|$O&@)e0l*Ze+nDX`P`C!WF9vrMy|-s?_t`;Gkjr~rs%=_WNKC~n#s<v?iuiADta
z7!U|s*`ObXg%_%?w)T~dlfQ!xON_$AZ%Jud&j_jF&<jb{)N=CgNAppdFP9t09wZSY
zXeWon=3d&|lujms<-;){<o-X%Q3qh5SxzD93{xIJzCUB+*ZsYRut-YLtxFy#ym=>v
z!}TL7$0W3M7+H-ymtIH7eQ9L8Bzl_3f=1ty03+WR6h9vzgk)sFnDUn1;OF_@CEnYI
zjQH`gK=@Gvk|lSyt$PgD-8OhG23+<aQ9C*=g@vbgp$DnMJj=7Q^kdXoCi~~{*%^G>
z*OYPGwneN5HN`weZO(*Uckjzv38l(CfSKXp1)szo;CVb`ulAxD*1ct!ge2#E{u@VB
zH^TQppfGW6$kL9l-}bXc*&Lkm+ZVq$tQ45`eQw<vBRe}9*ATu{MG@V&R$%kK`Pl3q
z@_2Q{KjhZ4ui&KtpogqTdrq#TlMLsxQxuk4R}a;%%P@m|lc(8=QqS8%JPrFj4dRmg
zgleH?N-NKco=y~hPbTaI&~DN6n-X!Y2)Z$6{y3`gynWwD;^G9T#?njR%XBo%Fe{qP
z()1Q|xN{2JnzpiMSCk!@LW<XhJT0^ao_t&%d(G!$G8ONo{@X@JE3M|`asKm^8~oCV
zi-+d6+)-$Dp5u1{98gwKSn;7MkIqf9Bc-khb=RQ|zMknO{!=aTlf>;Kl1`DK51rnI
zXD2C=cB|(x1SyM3@NU)N57cf*@Yb^{C-CoyHAY1Rc70}hs0hOkMw>j1GGJi&c`|_(
z*_=W2#t*ipK6p?kUC9T}Xx`xz9JW&2OhK}bq~Cu12$BQ~X7a1MUlwZdXJpV3#|{yH
zFBYaA&H%yIoZd=+k|QPRM>Sjeu9Zdg9fTi8cwCV;xp^KkwJFmofLOsy|0KMh0vxuJ
z*@jwpuM$fsw2H&~&uVxl3%rk14I5uem@?aY9r(OddL5y(h|5>gdC4i{YO=Z3WNrEJ
zRSXGQYRO!7`#hG*-3>dJ2N3qkK!j-=Iwx1${IlZ2WoH4az;zMvBAjVbcLKSdm?Kns
z0j1`G^GQ76243JwOkQ5{e!%*oGg&5a&4z?LUxZAAAk7dPe#i*&Rq2u8p}=}-dr2MA
z!fKvoryog4B{jf9jzMA~E>C!{H@GPIhmuh$Z0G)k4!G@I(AN5{w~m&+avURpKoSe<
z<*P9Fu+ugj!*ON5W=nC%;#%JH;crTaTmA1Bw_>eQ?iJ(f)r1P;^^H!)$Fc9=K1vK4
zrw1lRdgJ8+x*wtP3)u}4y;F9lM@cw6jbj`wO+H*RqaOlm<H(!s3s~6|2?YvufJgI8
zzkhRic$jkke${2}T3TRyOo8PD+FA3xcM5^mYLc)yh?G_-S5y>LI5u#wj4PA(f9Qpi
z#zB9qt%>W^nm|u{O<^%+cMjr@$K(bZi=XdbXuO5D+2zZE#JyeoPIETAQdEP@Cj@>@
zp>V2PQLOlxm@jmaEDmwJ>*^{U!<*-%w#Qc{<S&@nVM6vFxe7mwdZjn?2o-nC?G4SV
z&L{Le;a3yMvpa84d~M48)G{=GssWt>4zGe-)1(yewHCC)UI^xG*shz{q)R(Jnk-in
zN$hV9x=?i&HAENv8<<U}HX?6p4oZq~m|9%EwR7dD?Y<`a(}T=ly%cs3DXw6@)o2R{
zNY{%T4iDbUJUh8*^LgS$RjeXr$Zg1F8zd;|y4gFGmOw5WA`X8Ih7~t&QEvny&mkdz
ziAj@*vzyMlZZP{sS?3D;0mX&qa(u7O2{0=I`b5lo;+q9b%;v7;QKD3tM+0O2;q~2P
z>MjDNZQY#EXBs^GBi2snMN`g$0KR)@kzl>-3e<ijaGzCik8_-?#dsN#b}vXpe6(a|
z1IkkJ^TZ`1piIAgI6XYT8wYI!sq-jiz~y`?E+X#vqNj{Q6S@@J!11dhx}jnOp3Bd^
z_V~3f5`1=iiW&nkt%G3R-KfUJ_7du{8yaFB!~d}p4iS4afVBw%ZsXjLT;jq1yX7{e
zNWvWD5>MVA5}6r!O?&1g=)~aw$VE6XY)lENnv6wy+#@dgaqCKt{sXFeiqB5}c&kc>
z^WGvegXb(A;}0tPw3>F9s@BeRtbw!lczpBScuQX_Z1z~~cD}e1kPG-%GNJU_4eHEM
z`#0me#<GoIkKBGok+D`7OW_Az?R|@I8Eq;9lFWnfhOu)QmDVOXk;q@U8peosa!;Wi
zAIsy4zR(AZ+qD|lS8S<r?ow*$DSkDXD9_vTMQiiFekD|6=G!(t<NbqXPR7w?T9>om
z&g9%8eTwo*%U>DWA^Soo<bccLV1@}9_k5ji?Hv@>i!-*As}4Y5g>0clruQ4jm5^lW
z7FQ&fKVe{xGRHPsr(?Xt28IcpQ;Y^40a=qATZBpMT}361ghN47A7>^?QOjePSV}21
zv80jG!DL&DKrz#XM<~|D{eqXeMM;fITAeZQ>7SUZaZb@*>={5}Z)QcfOwBt>UGv6A
z^F5bE72}_#KIdhY^Y<<G6>E@APN4Os=hQEhf+bHQn|E5Vb@+zqdoD&#@qwd_qTIRN
zc}z;pd+e1-)tX69<_^8K<S9y{m}X7|LnCF~w#E2CqU7<6TF;Y1o$xyO%#PNrm(U0E
z*M_#45WCp#Y<XL$2VO(w+?(_BWIgAYNP^p3Ktw*4O!`HZdQkN9=En2wz4~+GGp<|r
z(>#87o7v<AET91Q*K0Y!I@DaD&6bkApvDn}`a>Dy>EWa%l%o_R)~e*uL59h?OyW%S
z;SOZ=R-}K{Oy-%R1N`ZnM?}fxkI!sUeL@M?M1}Cm)IRc1P92<;-J!>pT4pj$*ao3V
zc!jk&CHA)u#rFPD!qABu;ZqSQuE8ydJvTCm*6g#Ut?h}(qE@1$(~iBOjxExuQHQ>6
z;Q|We@iYu!M7{qzuc;sO2~W~~a!@$~bp&uMFWh~5R}B}~epQdVc}>%1-&a`z#whUZ
zVuy)uZf}=XR(8s|FM+utL1bJ$eGB;kP|C|Ha{v{5%&9i03j5}4*m&+0LymkVdj{Xl
zi;0L!9yzOum5D}>qG1L)Nj!+E-i0Zw`s2J@Xo^@&9pOf>-N=-W(jZB<C~W>;wpLks
zW<2cb7JB#I^1ayi0~hG^_BtoVoXs~BygbOvK?ZCEo0W!2R?3?y3^XMyY!rc=90}>P
z!>_5Yb_8nFG{3Ltp|pmeKVWVD<N*uD?ST<adbPdpz(79^zf|FJTfu4ye68<)D0ePW
zkw}`34pGE}`ZJ(REx#|uiXhoe0#1p%4LRJf;ivkapJz~&3$ZUR>iE)J7}H+l;+vnY
z-56)o1XD_Cq)t>Oe_p)_T=0K=b2|;p5AY|mKwIznRMeBNmC2)Nee7bOmS=3qo6WbV
zuJPWSLrm`A5%i{=L*rXQ+L-LY?!5{4mQ^W2>>SXnPq|r6LCwF2tJlh7>L;GxhOzbz
zoFffunm6wB%R=>@{v%^Md2jvTyLE3rWY2Hm7Lw4_6@<!wCB_f<rr`Pt;<poxyP#2f
zerTGW*;C)i9NSjgS0yK`h8LFp#6~b{(g9CWbw!j7X$||>|H#P-r0<{pGc%~FpEG1g
zl_!wVjwt+qTk)O+OS-{9j{WMqx`dZV^*^0Sq@NlWnRC@0;C`Pu!!>rqNj2@2agbdk
zA4~jnbRg&JT6ey@<meZm1Fbz$oxRIKXB+Ns10=52Kc-)4K?kJ1OtSoHFVZE{$o;Yq
zdeCWIPMIu&Q$fTo^~yU17p3@vKOFlL`0%QvN&n=H<XPh0CT&&eyld6cwpr`RzrpvX
zK4^I{Ui?&Ewo6OdeEB-srJMI|L3teG_g2qiJws10+5vJLlMDbH1s2OkvOHsl+&tge
zDK;XklFYUZFL|=*sOjul>e>qg(Mr=3`-@U#j<Wp$gY#77XeG3p_vVgqg0JYs01^d)
z%RDV)OeDW+6ar=GNYXS&3^huaR>BB*#My_9RzqyFs8<+pHh8IZjs1dRd{PUvCjygs
zKiH{EhIeYFf@F1+DvEmNhT<yA*+X0RTw8S{3j~;RijSg~aZ&cq?djGAgPwniw~U-k
zG>k@Fvl=^s;;2)E!$woPO>^tH;^A9e-;(91V@)Sw3F0qb<jOi+66(hMH)W``3v6im
zS%a>L#J%Kz*@bLC;G0kQ#7uaxHFy26p5FY2l#t}-qmT<!>2wz2JW-FP5|<yB2>)2c
zSubafz_GwnYG13f6!|m%YEKj*1|c<dvNBn#ATNmf`9V3{yEaHmXP|q4JlQb7+K@Zn
zc6Iq^X8{%Qyl_C(<UEps7;ETJ*saSGmlmIi^q}PD$Ol_nHB+VN&XNy>95^a_nZ-__
zzYc<zZNAkyp|lxp&_Y=YXF>tB`i@vsGv<e>Gjq;m!bXq8u9KQK^286z=1i_KYoXiI
ze)J2I<pwg|Yr<5Qi5;zBY4bkyK%*2JfzwpwWEh4SVn+R9^Cdp93{UsP{S0~c$o%}9
z=WDaO+NJfSd;Oxg^?cAbL88(R=$I2u(W9YlzAZ4al79FbfEhPAkyaWCr9BMD+jBm-
z#cqaN+J*sHL{^untpYP&<E(?O$&#)#X6(cTQtxB4vXb>^dpm2ec7Cq(mE$1VMXrnk
zUIde9IVq;=aR@Qu=~#Kmp|mn(2XD-$CJ%Cviy&8gF<<nv8{>uSKR9~zt+0)Y_D?88
z`1O!}MLRRZZWkB2xIqWO?M9?X{6RstRVbLye{L(A-Juy7t^98c2C47rc;o?bo>_Ls
z{bQnNJ2&o_p{-{72|qgnwVB7uE=IuCT_9Q?Gu5?RPtJDq#K&`vat-`krAN)VcB(Vn
zwke~MY2b>g;ItlG2=!?hibn10f0_vcjcQ8IBntcli$cmMjXJDcCwU)=d0x73Z~Ote
zCvjXqg1UPVVq)s#;GCC`p8l{~E&zTmY{f_vao=+eTUuU1iUlCI;>`q3Co5P9c7jrW
zX?!Kekr6A+><IN^Lce#H${gTdSRjL)2xynG7nn|5RdUSftC*xfg~3<~L>o*>%Csr>
zRk<#WK=0nYEu!{D2#m(X{n<m({aHHjT8Qg!e8xxxb~hK#Iv%HqO99~9Jg~}#4c<69
zN9uti`Btu}qf)^{Yl_8Bj-irB>tAYrL2Bgh<gKR7yGvdZ2SwHXYS2ifs8C_oP_%$^
z0!;s<6wd227sc*Jy_=vAZ=vRumj7JL1*=;<K`ZZkl@spGa^PIg^&xRAG$BOclJ<8D
zA^Z4nuHc=8`^TP3{Bnu-j-V4V*59lIvoFR-6q?4%D_Bx(;`+QhNK=GOH;J$P6(D~o
z;<!n@4L-)>Vx%=|FSjxnTM^0;4c7-864NL}Ug=6P2T*p{C^##F{ZN>GF`~w$y}Xw1
zj21B^J40A|*LXg^#Y+BC<XOZjzs-k0DXK5f@)?!ve+QWJ>xf7nP-$eG{*ZG_g%wv@
z1I8DPPiGtzG~amyakS;C5loVK81T|k9R3kHz>lY>mq+?ri`{~sxA*P(rF4v#cS76O
zz1MPBZ5xGqo?*dBEVLm+dmw(IgpdRBkR%i~283MadgdJ*_F&X=5<+w9Gd_IVrV!uq
z58cFZ*!Z)sss>no(XmZ9e&{o(cUA0IdmcsDkkgB~qn3Qj#JMfU8c1E=#F6<%X^7D)
zaGc?GFjx4XP;bS!M}*}~4oB@UMK$QQ|D%sDla#i;iFsThhw=;tP~tqQIJ4y6%j)t5
zVX)6Ft(K0qR51YGg*3%rOs-a%-GRGO`hlTU$L_Cv(M&3+w0vR>-~6oWl<uMXc`4uw
ze_qPKaH#lg^{p|ifONtZh`RBjhQV}SXsUH%%n<kau+og~@r|cvB=5caBAc(@KTcH9
z43+v?oxlylg8K{)no<Zk9wfGPO|H_!8uH0!1^5YCI#4n{g%&E>vrQJ?ib8FxC)1Sq
zlIs+G1^TH}{kNDqR&ZqEgw0LY1_5fWD5^T@^pV}(C{d(d;%B*WdfPMOLA^(=>o@&{
zBjt!C#)H)-Bw}GUdzlN@p~5pf>Oim`0y_{=yySQhBk_#edACM>*U^I*-=&q`k-2^n
z^JQ2m$SzgbBMiicj1LxFzZ4q=iBqSl$Vdr&l|PLU4w{j!ZEEunW5TAcau)pC-u(v%
z{|099^>@8`2UgzGG8(kjt)F({%EUZQ*i4_wgTYgm@F(C;qA2zO>eGs7d(*DXcM%fj
z@J_^x@BRNiZQV2csvO@JA@WWt`O6gZ{)Ko0J9q#4*uFUdZ<8Ii2=Z4lhE|TMx}<(%
z(%6dC*OsS+Io0Vp7>x($T6%Xg@-U39f2vBA*gAcqrYXBJ<V-CFMlSGuhB5^^(cF+L
zdVdx{zlQ^sFW;nca)nGoazd^krlH3aV(MFZOq%~AGz<Zoc)W_kOaG}Nd?MQ?*nV1V
ze;U{o{`ujTP82Dc%!oaJ2?9)Be`WMCf5Q(>h8*>-x@z+difQm^T)U&NVveof&4ON*
z8l{#Nj<6$#HV~DIAx|5-7wUj-&EMsU`||JkLcVX@Rd3jaa3e-~UWw6|r-VC-_o;pV
zA=5$Lg;BjWB1F#wKI?eL?ZWCE$KuK|(fG+@em-%56fT<=(?nz>*_iXiNP2(I8J*A)
z808X@khYf@Kx<5&%0^wuazL}OS+3q@xzwNpI-2xt74VDXaGFiQF@q^sD9q*_c*XN1
z%3!aR3WVBHYmuckvK7|W*?eQejm66KdyfM6YUpIkH4F>}jpDX^&rIWt_`ZcqRO*-(
zpoMw-m{0dI^*k99^mZ6UiCtKz45lQAMUc?2v$G?jBrb;Umw^zE`@`l0gwwC@LBoN9
z5eEd6M}LBBiUkZ86xc2T@lBkn1n1^0n|+1}(Mpo-4?H2!?aUu}g~ldvoS76lGpUp}
zNvj@S*t$@%-5zk<=YCo#B>wmI%*WoP<|1lXU`Wbf5_FrVd>qHc2g&v#yknNeQO&kh
zDFaqTFuMLVFl4a*&8?#-kVE9>>Cwo=(_jqvj-g`fYOVG6xS<iV_NGjGjm*FzGu~Ev
znUlFaFmSH<_ry<(h1!!ZdS7-)LFrargIcJNU=hyF&gb&a`?OnaZow(OT&52rJ>MN$
z6I8cvxre*)fha|5t82^OA5UM9Z;b(?+ik%e5A83ZaXt;^;6^Cs;<Zica#_nR@G>w#
z_lsxMLTQk5-Z0o*7GaZhIVDQI^v=W*EOdY)Y0e9~2R#1=#rPK}SHSrD1CP$PB$W2!
zhd}1W=ypMVs&V7h$c)>c+FgPmhpg9S65LR@LEm+y=v-P?*nLWZL_`_d{KCSW)$=9o
z&d~;fnVi!E`Ttky!U4#~OPapDH8zU*yX>O1wg)QDgZ->n^&2)!{8bEo%=|~{vnzhW
zlwf7I;`M&=LM>+=N|o?aV90Qccdx%AV=2d!ev--yCEi(>1v$G3Oe<LGJZU5H^>fA0
zM?}E>I{UOWVb<Q3rQ$;Dh+%E_-~7P9r{y;fy>P&v{cO~Qj<bQgw7R;PfJl;<074`n
zqU6DD*yOa18ZGdLnD8}08^K}?YlY;&JMwdYkZf-YSvkpvC^~imP+Y8-4BZ74&%0(N
zarIFcA7TRIukLa(nQ%M*71zdQpF9oSjn6(=hs^d1=8g{QMi=ni1SL)2F=Z8hs?OkE
z<=N$sQAS&^cg)ihPgk?nqJx6__{^7uI<lLa2WtaKg1f5Pnpt~q!ftf~>j})4*^}s6
zyRI6BP;KvsMDeB7=Hbd;la(zsR=U|<^X-8!q3($C*I0vnz0icChOazdcGHg*TF5c7
zm_kl!pKXwHJx^tz{crNza`3yI1URn5^Z!wF6%0*wZTMB`25AP0v^0zck#3OgW^^Ml
zQc9HWW=NNGj_w?cl<pW^6G`#g_Ya;E&wb9h?kkW_apn(iK3r$)p*;18y^VCh$_5|~
z=JWO)ql%BA4KI(_bO6%ZwskoBuGf5c_P<3*4b`(s{GdjFK8Nx4LN3nTPI#7l&eBDr
zNkaLMNDaKqc|V9Q=o;-dH-zX2yWt(;c)!Y|t6|`LD)Da@z2zQ89HtO6aQ;Jja&U-Z
zI(b9^9K~3O^Pv!@FfdVPku6Bz#HDC{hntbZDt^6Y@w4AwZR70XQ_JgqTiC)yw^CL|
z?m&W<RO?-Rp+K?2P#n!_jMZa*oVfJ$5`*i=c|{py2lZSgdw%S>SBY60NjP5#*aao4
z$Fz@PRRO3rpS#mi4N5-gy%mq7h&n8M>9VFMN#Cm<QH?QsTYe2lWB=)~UF7Xj&OT9`
z7*WktFKHvE!MZ2nQ<eHRkN5if>z}z2aTM<x#CONjoYvX}Y=D^!9bzBvIrY+w?t~r~
z9lz8w?e={WXU7`&lD1DqT|GsQbQEO534{?J7&%3J+Jg=meAmV2iu*H0&C7?^2Ht)t
zr`L%P-rsE_Kp`GXLT=}u^aXTvwN6OB?|oOu?<aiu;eK8_@Lt7y=eYSK^^HgRvdaN3
z+!{RSrO(Z>nE^$bpwW4h$z|>9fWMBX1zdK7-+%kDK>r8x=OK29_=t~)om)U`;5C{P
zVrmCH`sZky=ly#jHv-C2jiptZ9dyq00L~{G8IpyOrG8$G3#;?T`1^jOSmq)o#5h*>
z2ek>{G(~`+Q+Q`M(l_K1hCqGc=_lQ0q)CP%J@{#c<tk|8iKXWWRAcs7k`UMq+sLUU
z@?Dui@+rt~4j@@&GvY-sApxlCcF-uGbU&glXzM;zI?uXQeZ482#o`{O^fb-D$yZ|n
z9sl(!^U7lQ-0aQ1fT5Y+#-~?5IjIN~_6)%ph-Vl0g>8+%P9n;xW==j9FUZxjQx@{P
zCDG&I6;<g2IcXQQp+lqB2)a%QJc=>j8((nhVS%HB(&zndPonHOBiKgxR+|Gm1wU~!
zbh<Avk{Kwx%$Uqbp%~wdQsT;V+NH5V%Sg{`bNJZHe^XYGAswVBmH*f#2RQZ_sd&9x
zwRHpT7a$2Q`-L{QX?mUo$KX+{=GWsf6nA1M-tgrav(nN#EE@6aC1w>TPFJ%hP`W-o
zQrs>i<ePL|_bkZUJ8gv;FRTehT=c;Mplz!y%UM1;BPVxA<H>;CiDioZyV>S%fp+KR
zf?LOot#U!|e_%^Y<3B6ZtZ1voK|cj3jgk7#59T|PPq)t+Ki_P={Vnr+Bl*m7x8L)R
zFmuZG%~1?;WwA%u&4sp>_O;)Kw~4>n<Z>^d&aCW7TJMVcaml-0gX9cRVhH)I=x-^k
z^m!%o0+~j9`Dv_7ODg#6C7>?TdeQcwXXmMk$&}8k2Qst!1AL?E%(vIzBQc75M9;&^
zeeu3qtL#aFTglI{?zU9sPXg`#{{4C8A8t5RzZ?uknEVUuyd*v{y|>bxLFMzMNSwL~
zb_A#<U=*32o&Oj1L2=q%PpqE<4SQO8<ptX4DJJJHC5D?*j^g#JIcF8Ypcx&q3EM2#
z&)-xZcJ5CuBoK!KLBv^xFb>;>hNax|r&llLqoOIeH!uF4OWfx692E~bpd#f5Umn<r
z!F=p^r9ZuVuST?8o;{Q`{B8ErH$WPqt%aVKWc8X=Ip_Ol?DY2B5Ty7KV`cZ8WQK^#
z0jhbYrqa^lPYGOMeZKE`C<BHuv#3LgHLH+fJvXa?H-xBO{#hR81Hmt|JNmqP@Tt!i
z>s1RFuxwmf)Teo_W=|?hvgL)#=x*veIGe6W9t)d0Eg_p}bpH=rDFEu<SW;1G&$6vg
z;}<fs^Y#3?043ux>Fg7$b)Cto*!Cx=?eUMi2s+`oS{f$dn$(M%OUV8Bn}0{eZQ7NE
z(NMs1f+P&OOt7An<rs~wlXf#hx#-P0OKEr<4c6RR`a~V-0^S>EZZgTu!pPt@p{I&k
zAee_|fk_akG+bJq))KV=H}?k^mC+=5&pnO#0pasug2Coe;ExN7GX+TtTgw+6WLpM*
zfV54@3d^@^hiqOBp)vxd@07*}GxfnB$ca8o#g#83tji`J%nN$PiaH?lcz*KU2nMI9
zcYNIy>wF$3>qf>cDlwm@g7f$jAcNGT5gG7V*D}em&eiHn8n529N9+TU8jzx^_4(Fh
zOWWt?*?d%FE2?Yn=I^qI3Dd~?Ak#Rr$8P@27jhXDGWu}v>s1iu$?fj_XK_m9lo4z&
zq^MM>_@z7TEdgY}%!nt2joetIao76~bD3eDDT<mYumN_-nNt1Ri@XSVR4c5HBPE#B
z>pkKhA?`{Q4hYv1V^X@!^6Eg?9)rb)n`xs@yC69^kPL0O7(SPIr?hzX6BHbr0_|^c
zrZx*lO@G0<^VR<59scuW!t=N0{66HP(L7m3t=F@Y`TjhribQIYiDKb!lx6R;h{uR4
z+Z()CvEP}AnDI%ltW|*6U{w>PN9q$^=5Sh-91&WJo2n(%#UuyU93~9I9&kVZMLKsb
z>>Qfk(_$pNIe0S85cYX^XUKeyG8f#jEvw@sEOA!P&E6fHk?DM_?HT?VoTOehb>n2r
zs>>s^Y6Aw5Ew&sbVE7q*%-%|;l6L5K4~zQ2ildXIjf*z?V)4%*6I<hm9`y6R=xWys
z+6mi2ftjebn$23pm49yK@xavC({ns`00nLgipg+~qHsN!KSwR8#gyd-XOFxIiKQ?L
zJPPg3*qha@3A)GuJfD`am8b!j&|3PDK}|fY1)7EnzMl{xSN76NG{kWuqz(%|wG2pB
zy*!+}cM9LedfC)WmwcJ!fho&t!mO&MV!g69nz<(4+>6g#>DHJUT0ftXU7K9?BTpCu
zyq0DMuEgU3S=RLnR=20OhIJdI%c-Z%YM&}?T2^g5Jct+Tscfj?Us5Ru%YD_!7LIn0
zMWJ6!?=K~$<9nM;zZ!=2Oz+Q5(TQ~g1+uX<(eDqeXRRj$9Sl=O7uxvReF5nsU;n$&
z|Ht(2`QP3}U0dDYoQCAkZi1v4la^#}t+nS7j!^KO1VPYs5vwIHRTNrHAK}Z08Pdv#
z&X1MH*b;P+j75pUv8>3z_Hb>1a-YC_VQ$4)5=|W;Y!KdnD)BJ3%9$wtl)zUSJ43(U
zbFx(vKg~Dyo@f?Fz5MngIlrC=IJ5r{eC=)zRawNRbm<&OMsme{wRE!^4=f$(e-el6
z@{IC1gX(rD%M^)A-hF2PnDPp}AMwRsAr{sfT7x*$uU<4Zt`3lesnRq^Vjh6?>N@5|
z8aSL!1Xd;q&eUVlN@T?U&TET5FIhXgy6RTo6S5{x=(jzGU$`G?zfn-iac_W|6smg9
zD?=EXWs-^qcN1-d=^+uMhQV9V{Uu?kkI58+N}(_TBPCQSlWp|a2d~HX;Wvwhey17V
z2)G-4TvEL3YLo0PuM=7*U4ZVV|JBmE+n1n*Y2EhcrOF@lZ2YD5tRFgk&aJ5N^mD?X
zV<v1ntC{+cD&tJCMq1b3{>~S_ub#zGw1<F>SI(>b>Evg&Q6Xi2FB&?V#^*F7CaqB<
z{exeBUV-S$YsAeZE(4{v!rt8dAcMXqHVkETxdt*(MGY1u0;ApdGElnD`k;J9Zbd$C
zOt^OX3V!>PfP9V{U&w*BY?X?$$>z@g=bq2+ti<KP=|v0er<@88dxx)7BPezwfy$H|
zwYQp^Z`iDmgfDHBWxDmrxR+3ECd-W4fzkfm1}6m3>eoFtaPMUU0ikzvgjXPHXC0dy
zyp4Wxy%DwOzT)=_HjfdlAUomiiC=W~aBSI&eA6{`T|m%?^eR95fchgQl48emEo(JM
zhnHrVF2%$21h~~5<Vcue{a?kp%sAYU8jedh_Wf$H(juRarewf54j$kB?OpRXYj1Ew
zGQkSfZybf*P0`8e*p;$jQ**(Cj~Ue%pbwavrS-R_A08BASWi+It&OUw+1ab+k6ofU
zm{dt@V`b%eU6Xj1mKiGG(_FYX)t^A}Tj5unfYse=cg2<5{S_$td~u{S?~Ksv1IgHO
zA5pi1%8xhmKbs3!@m6yJ&_V}PYX5$7dtT1J$}xZD3^VGjeLk+e-cgYZu5%&0&W+QD
zTY<TroyR&TxM%F!n>z9;hh!kcKd92sJt&izsrnV29f<v&`{wK*#j?BDIF~NCafRD=
z&V+a=TH6QM@)Yx3#JDMo%fV`OpLayYq2#NJ_$P~BQw!kIi~Tno>JY=y-QLT{4Ex{a
zTS;W2x{BFq_JU3voSfQKt*kM*JZcpVbt#lt!+1Wvn*1^IG0D<cpR->3VZ6kpcNRoO
z5GBO1J1tSi^2o|-27ZI>H`P*{F3#)0s&Kold^<F@#5(WB2}L}9qoIz!9=WphZSDLA
z(#@b}a5Sa?kL7IY<l?5Rs9>M-*w%w6guA@e_$^xc`>Q6aLfi&ndN@vOWptS7#PByq
zMw2YtVoTWq{CHAI<1KWqRhfa}X#;Td43mKGsPi3000P}>LgN$A5w8_}HRtu`<9cy(
z8H2C^g@eo+ys>3{2fSd|)&0v7t84MX<)8I+bswuY3Ryi1fBq;975*x0Ro95S4sryg
z0c*z{HSGd%BZkzI`1`DoD;R}9d&_TerNBoZ>aEC6;r?1wS9^g&`}F+xRZtX#*)zQY
zA?>zM@LzURy6lIms|4iwMNq=&3ubr@&>Gn+lvE=#&3R<nAw^^W6g@lcN@s$wWtArb
z<Ez*xDjm(YmkEY-U@<@j?4nA;TBhlE#yaN~jBKTXdStQHd0Zy$ro}C7-_x-+X|-s$
z+LB5BkJkzN_y0|#14INh7Q+v3^vem&b_T7!YY`ig&pQ7yFfqvGgMXXNfyn`8HaCV~
z9GX3^yy!VwVsfVqoXw0CDp8dT(*qZI_TqUSX&sety>I@H)7IAZZjVKB%r`&eIaEEH
zM``@mJaPrKD8m>Kmhx7kw`CBTNFCj7bKV;hN_WmGM>`wXJTi059`ItAye-t&bk*%<
zjf$71t8b$t3_3Kz=J=wBw(pOv0l1s!ZSmIhw#b3F+QTm=AO^Qr;0J}HOpXM5!BWU#
zrBgsI#XKp|NM;b+&dtB(Vh;*Eg5HbpW?1!z`~QRMEX*4kHMDxA+$YH>++Ej(L@sCL
zuOlC;f`{m11SonQN!arxr;cX<8C*5U0sUWDZhqn(O)4g~!s!+A>$$>T8EBhki<MM0
z_jH2F3KYFi)Y~HW*!`tbRb{iaB+^+6C!)&#0OYF@vKIj+4+nHN%L>B3Pg8O~_{6d*
z)uZJG>|><pyYVas#JN6^B;>wNd|Bu&U1V9`(56@>qN0g4`&l)8U-6Y8bz{JhC{1n8
zWQ7dNXiV78QKdpW1^ckxcWgL2dMUneb7qx^%EjzZB7B41M?l|3CqKiv0hKo*k;G)N
zvSmO3Gm`1CtZN|xZQzey!jE$#clU$j|6L$FPX8|(SvEiTcl^Bh=FouNnqx<E_$Fhj
z;rA7p(G4|CPCzHvG9xQ{VyuK5`pJMbyO|nQ4~1KBhZF5YK#&)um9pwy{j7}RB&na5
zV$kg{aX5Q<sdeI0>5w<3gNK-Fel7{N2(&Znkt2vh@`3JVaURX9!Da#G@Ycg(H!mBH
zFoYv>c&5>$l}9kdr%1*SCmF~H$Cs9&9y&Gap1PaG*GrI4HrulqPGn{i#Mx|AY_i(B
zGwErPqAbBWN<PXhcT~W>xRv#dY4xr7+Tpd7-s|Dwyefa_sIRB-=1Z&~CB9+h(lXT+
zCY|}?uQw<tMC0->i{);~>rKhB{q!3=Nvwm_e##gdqQ>WAV~!6UitFA>>onKrX)n&Q
z2WIXu(t!r4cQq<S3xoCRjx}CppU6ezotm`v9ORQ34I<^)q$J65-=wh?R4EIjSt}(#
zEiifSiT5Mv@*XT-UJ$954N0qo&&2vL&-*O0$s-(CC=4L9B!1b?tn`;mxL#gOG_<KC
zj1ID9Sks9S!~q$LioaS6QEEhFF5YM|sWI!z&36LXp-hH7-D;G~Z%sY(UQ-EF-*$>A
zd7MO_(q-Y6%iBQ-{`@YVqYT$_6ct<0(YSUPJ3s#q?t6dW+w*@#$nb80#E=*KgF@pn
z<H^mN|ISdx{@<doow$33X1!GAB@*)TULd${kLCgKoWS}Bd6Z{lf?gTwrlNBn0w$0z
zQc;6|20c`Bos?CkHP8^Huv>}N;*r3*n9R=JVd#gvVPeN;%cn>#cG}44?7lygAx?SD
zpW}V}Aq!CA?y@7o*N&k>{cC_~dCZcwI54&RshJr=d;QS$7|SR!3{RU^f!L4y`!77s
zv6W^QX3PEizr36Owm$raJo>rz>R=@iv@q3)2ngI2GvA$2phMJ;K=z<^kdkHR*??&n
zChtjY_v8*g?mTT&R~IowErI=_N&M7eePrLK9WlW;+gGceiCE((lZeXo_8|B!V;>i=
zgdC4#EK(_SM;+4hQRHZ@qIM@B%D_P|6Dc6=G#v1nI^cwK7ul5JYj{pj3MoT-P6c;q
z(^{4i8?v#(S?B{Y+!B9Y)cGlIo{OX*#<~$X7bx-q+-tRFr4uE5bT9(=Jy%~mQzbya
zsUvR!vhDK0>9$o$ExsYJ3oq5oAJZ7TN(LRD!?}Zx@#5t3_DUK#d|#!fiz+%;>VT(<
z`DEywL)U*=zMBR~RgjK{J~-!Clqwmfq9Fx)LXN0uX|DMq3uXPehAH*ztf>E-m~3_^
z{!1`t{v{CmInk)j1P=@5oxa+^k@u6jNZv5w;?eI9M<c&ic>|-eziNMCVbcsqD#RDI
ztPj7cPZ@D&MWCowZRI3Kpd?Ksx)`XBJUZ`Wm9!=N@}xM0ej%cz!#YiR+EuU;-2jq8
zhF<b~wAzu*WG_Giq{auVN>NH_?%5csCztHWeaN$O%v9y`@%6`g6>>My1OIn=@dOtM
zx@Po3di|_*StI;~BDhsn&^t)l1i0kD1NT^Ktogx@B%E+i>Fcd_E>9!uNL?~5EKP4s
z3f{v8S>pwc(d@Z?&Tz+IcUpOrrW;#E^B}Wf5GN3Tr-EWEUg8Tr>QmiPjK&cz{lN2D
zPApvl2`+(LHQ&0(=-IZ#hp%QA%w-p_D)-xrGU6$nQiGD_-pHaV$MB2~=B8U>DOLUZ
zu>tJD(MuB<xV<1UcRfx5=jh9B(y+*V@UJ`5{uYjOa*966ejdj7*EqV*I`=LrsPm8g
zAU+%zR>(Tm5G$-!3S++?P)cKPT)ntHy9iDxii-ExS@JlbICyhesIK*1rc=n)ZD42c
zb=8s<&#C#qD{6tL-rKrTs9=A(E=Jye+alj|3MU-pi`+TfIe+$7aXr^qI{V-Cf2!<s
z)H6eRL|PT_4#rG1h_$PePFY{p#iExCtH}fRzfk!mWVwd!QhJ(hEeYoNMCmoErBg(V
zeN&=L7WE!4v=EM7YO5bBL|CrUIb<i<%aMQm_HLk_{lFMJb#9v=y+Vn4x+9-qrX$;j
z+XMgMMac}OBzBI5y!V=Du@^iBQhS?>$f;^e{>h?{3)Nbf3KUj4y2a=WxEF{2+Y$qa
z1?KLK#Pjy>z&BoX_36V>HZs?}$5`@3BV3fehO?t-%~p2!L1I8>2u;1XVd9qx%inZ<
zs6@I0;~B63-%b5kt63!|b0nT_{kceE;^>1>I;H`qh+VL9pg0FB(Cg@1>FL~!&yf#V
zi}k2(a0<9e#w<1B>YE2?@{0HPJ3F3egP!g_Pogd14-V<cta9Y)M4+#`suJ9uK7zNb
z7vWIrRe;E6tn$n#O<6P2TliXM6c~()<SAODxjf&;4ZOhIJXlKF%@y{Ejf}j<>?~is
z4IJDk-Mf3uZlJ|;Xg+}dgRbdH;>{|F;;Q_TQ0<xY`VO>0Tqr)$R40gye}sqkRp+gL
zJZf!vV>T>&Kj#VOz6t~|@J0@+DFF8qsq7N6)Z=%lakRIV$n$)1Y#Y_gNLA$-m^Jso
zntOO7yg5$Bunw;D{bbwa)O%X-&1`>HXz`dvPfJJqKs|$kRPPT-3%;zx^>9Rs%gh}%
z{I>S&idOuM9^~I;2SCiT(CR}^??x1~oc%W8CU$7}Pog;YmFbXml|U-Qeb;De?V6+!
zHzrlt(wLlSE}eYEqQgvJeCQ*Ps!<>>;z#p4HC(!!Mydo-2R))820hzVSzgkWX?rs9
zUw{<bR~VYgMqlQqGX>^pxMrJJy`>y*vM0&?`T8n^mZ>A0UH>f6k`5Mp?cM*BoG91v
z6Sio$X=>dDozO~8w&fM2N&zioY+-QeyEzk=GBw$bHd_^{q-N5^t#9*_Nb?AZjf5c0
z|1SO)_-MR&pwaV;id+9oeDgs!S{lnyD(@V|;e|w^1%Kgm-VtVwmG=45LR?Y4VoVta
zBq)+n<-EKb7Wx;l`FwrB)7k_pkf|d4tXW2oEQY??fpY39UUwJnG8xGX+y0?ZB_`7$
zhC9b#9ajMp!kF-=5`1g)I(|fTnBn*DBXt(^k?b(mR7MB&WSyb{Rsc?sy?~cvvqbXe
z3?;o}vX^Z5&#FqCb?%%roRxuZ7H>vz(X|?uW0p?TDC0e6iwE<)43zmB?7v)4KJ315
zq~yy1?Y%Ek9XMKQNS0uN)-Tur5QjgMEP(~@*)~@Ve@mPay!in8_Z;%r!+`bDENCAE
zsC8XjuPh6V#yr(5;?kov8|AtB#zU7y!ru0hkSitgWT`HlT+r`)<A6`E{1Av$;0=}B
z!IXR==y`y0&zc(8fIyK*#AeVIOoeG_gI2Sg$#&%fOU4OSk<u>KL=w<=-2pcxN6?_A
z*V8ms%f6az#Cqqd++S1|N0Sub`jJ!@Uuu`Jc*a5c=nK~oki-!roqR@Qn{5(0k5Q{h
zbhjl!VkuSNomR(f5=-oSI)oR+yQ?G*Nt-7R)N73W%qLr(>yc}vAd)A}pwINa`ubn2
z=I>M+7rS0zWzFDlX{UG9<pfO^cpA!MGYks%=Ks3>B^p%^i(PaM+=kq}3VymyP(q#t
zWi2`h*miiqrlv^lUZgiErlel^$+|5u#J-EI$YcjmCXfV%j3$`*LExwW-uDL+6P?JZ
zd(83NR|nS7TKsd@!p*b)6@BZ+?suJW^d=P7*M>9cz{<=NQg~<l#9d+v-f|0jvLl6y
zXKFGtjAk54=bxIaAnXlVZ{zyh(#X6zDjT?)`WpZdmpsyS&<ve9s$mgBJ*}MEQW&N|
zsRe1Nfu%A_#~3O(Y{{-$!UAA-J7uB+`zQ2oW|O5H9cAFO8hG&tM_yOeCa&`%xkAl5
zSPnuZ_Y@@kA?Eib+|#I4`oF{kOn};)%FVK^XI`GrP@ZVu_j&+s&qDRe!hinlDh|-K
z128XLN~{7VOit6i-1S@0@d_QwYw9x$MNC*Xf=#JKEz{|m&8yao=<Yxe9$EJvMVtka
z93Eq$nR7dKoomDQg^z_pYpv~v1juY-28p-7zy_1>5B|*>a&hr2;n+-#fafC~C7)h!
z|Gg`@y)jX*J@dT|1k%mC`SY#t8_Z@QzM2oUzq%#2A}P8hZvsR2_d%!(x#v>M_02+p
zkBZWq*dt5<;opW@;K$oVHcGLnf}r#i1XF+Wx4EC~^{YMtz3aD0(x3K39G%^Z<sf#0
zma+hlICbxSy@K1%%TtN=Mt-zDys%h~iD_yR1)j|-Ud4!t0~|-DzA+D%96ytW?gi`C
zd>$n{OPxYosmU~ssTF`k`^QN=XALb^lcO0t#u2@QWD0b05+<Mn{b-3zRa4g5;cMJP
zBke(A5e$9_tuAfhmdT9n-%f!?6XwI-rrl4gJwtARyFzzw{#N%4G2&ei^fzSX=h%Tj
zX(dBUs5xo*{B1WgE-o(IRJ7#x=dU&bFD!wOtx<0?zbnkAUl&hY3JHMi$Tt#gR^J_M
zq8SBA1jdi>8T2{?P;$x9IZ|2Ob(3_Q?WIF2qccZi#R=(lF6ob1WN&3S=ruv=BrG<X
z<72&bbRDtMllH>J{By%qlmDTE3y9>kae)x?II9$ST^HTr$aI`)yp(bqo@mNgz8%#R
zMy5(0v>LhA-TH?4{#M{UV`-~I>Avw#7)-OA#DUR>y`le;a?F_t&kA+2`oF*6^u(BB
zE9UcL-)$&Y&-Fd`rC7+{+Bs)|`<vP)W{sem+xvv)=S{oX#f}rZ6H$W3bwj;$Jwp?d
zVzXLn>>SqUrIn5B4_z4xt_l^sShK|z;#uy0he2jRXQ1G-iJr;v>uZ-G0m3k-d@@bn
zdHD}XP<I=TMiPk3tE4LQtR+&joth8`;#&J#v;-)mWMcwpCNi<fp^OQpof!#sKk-u~
zWF575IMEW3gEVPl1gnpCT>n&6S5;dmlRDH;Mv3g!SpdfrV~)$UD5&Ah4z1$kcE`ff
zNm2AEma%w@u8lGUq$HtyOojzxkwZSVl&%SiYRw-xuVVo+3tW#=u%N~LjF5Y;r<<Pp
zjwU0{uLJZW8i3ukd#~ayRB7=QYvkarUh7|bhd1D8R#W*9q(}IL49SSHq+>DZ3jV$b
z@9xhCK4a$y`3CbHM(?s5b!;7;-RTby&p#he)ve{5NdFduN=Z~mL%gr`<l0T$4Ta(&
zlMAN}SgF(BUE=$XuO+|apR`twiN3ej;>oos1`nB#QExUMHKE2qBdnDY#ak3CD^_=D
z#82j>%?rm><6>N;!fREqnX4`fusjEI<v;GjuNH%EeI1&!V@DX&^FEA4`f?}!1s?GG
zTG6c!n`|E3s~yyKUGbSe+-yF$OT5vC)P8p2;DRODu}K@GA?E%_+G{sgH2jf3S+fsu
zLdukssWyKHk~sE{MP#`mN)e!F2C|kCc6eN--p0$>QrV<ErW77KKzb~*h@s&Pa@B6b
zv7G~TYhs?6_rfh(7QHk%5d#w_N+oT?An1Bf5e!$0{|+`$;S5`&^$eX}=Xtj`YaW31
zymJ>=N!_<nRD<Q9M5{di@@a<?!o4o^|1JufY1~D0Q~xB#FF~P0G!^0axU!%+rfyTa
zwnZ7E<R4}uRp`Rq`Y!9rp*$98*+29YxVRF_cyHuC3S)oAr1$ccBEhAu!Y|y$D%+qx
zoWD|w11gQq;TgW~wOA%sjL{MK=|a%v9~8AMT*8V+8H%7hI0@Uo7iEtTcq&!NI;_mD
z5|Ulb9?1`i7`_IG^-niAMYI>g@Uu2;gsAzyTX9U#ySwZdSeHUSW8@t1QDnxOBxS1?
z^=;mr^?bePz@o1!mP?QL)=~YkVrs()qR=s8<q-TG)!R3!CO2~M@1j#8NnxZ<8Kb24
z#KS4UNsOLh9rW6|zX*xyxpwO5J02f@Mb-mMmd3TJ=B==F^38A!g7hNa6LWkA8Oj!6
z&(@=eMkBZN5)_uk7D^F<nOZ1q)!8zaA<6DN7sP3IGy+IY3LR{IbuBzx-FjvGqUO(N
zW99xgf%R>&Lgg3VwEEN~Eu8}T{gU-MK4X{|^Ju5QYSCbm7<U5LT9%(f2^b>KN9aV9
zy}*$AFTK~4R5B15()pzr>wcb!;cUY&t*T0vZ7KEj$#uQ&9~Comj;>3%pWo+Y32G^?
zR5L@IdfX5UFVJd6RI3Jdp5C#mumg?Nn)vWAf@XpAI3p3;HY<jm?}cY7ZN&Bb;<xto
z&37^+|H+Y)lP79*U#MUH<)zR*z~CGBvq8810BC-j`G+}GQ6*WPcedWcgktfw&J@#f
zmFLEUJ@Nz$Tybe&ta<;)@m)?iU2$A`%YaTUd1oFkctFM>>K*sA-1ngApC(CU(L8#;
z<*~bhr727I<i5|4AyE$BrsXqdetz8_g_+7<zAnRZFp%)c{B<o}@SnLqUNij!{qI+Q
zc2eC*!kg;QepCh9TKU&^r9nTwWrBIz-S5-}hAu-5IVP<7a>~|S-)@SD9(KVb^0Rz~
z0#QUJy~fC2)QjFqbE!Ol*I(%?()1Ph<*mn)$7FY*L1qDe$v!+zm|Fw~6HFv8yxWrE
z*Ko5ue^jI|gK_Tg$chrww88^7y)^g3-^ZQY14U`Cfx|v$(PG$!@7PLVNXDTGnkV@$
zRi@XSA1Tm{&JffN4O*{Dvb&_lUAb#A5lq9BCdRWk9JUlru&GmieJovzLNt2(D0M(7
zw6oXd`2G{kEZ|+wuvxBZ)~Z36<NL35M28)G`4>L-3*3Q8=cjn-Al{RyS^N1!$^8{N
zzvoayEB~9s(lNDzi>p9BO~0DG=0&sZ49Q#c1H?gnaiB#Io10zD^2W(bwbz^%xC%iK
zkJwq63PnikneKa6xSM3iSE;4zL5y4d&2XHLG>g*^X6pkp1aen&@aA`%WR}H;39ldi
zM)0iInSnR(oZFG!wkup_|9KvZDDB^m*UL1elVe5pW^l<^D|$)oJ)X*;zGEd{2TNcm
zC}sBra+%Bg>8X}S$Q*orkNbIL^6^jSva?l_r8^<ycI`uL&$HwH{Td)0q`9Gx(2puE
zOv6O%a+W{=1tMEwp_WAvuP30P8J$g{pt0JiJpP1N9S03t72AKF{sr98xvWZKI{;lh
zT&A84)Rj8wi&E;GbE`wMqm8EQ^wC?F7^5;d6<lT0**9ft2KnwTUMPh*uLCm{ht%Bb
zrjF#S^a1B8`J#p1o65ASKM&-)nuFhYN|!xEJd0%-=Z1hQZ4KEGmNfanjzx%7=QD!u
z%HbXy*Bo-VQmq}bChAx$GcLFRui9pJmS&0qpS1IZ%WZUwu7+vcbeOg-P!#Aqb(Ej;
zdg<Kx^wH-}%1NbcH775(z#n8a7xvffKYazxs1A!{aGBZysriJ&LL?PK9E(>H?PUrC
zyj6IVq+0^M7O^W-ejlG>Y3=o{cl;2KuoCg{W+3_^SYF7erak-fEjOMrZp7!yGBNEH
z`()KQN8ig=`R~-OXen0fQT2zk7~HI~v>LyBlyHduZbxVPr2rxMs$lh3nU3>0dkoam
zT?_rMS$^>}#|U;@3w(B)>nu`EwkVFYq5UIOhLmjz@Z=Ovx=^}+6H|BZEn^I>imc(N
zg>YN7_cy66qvVv4gsW38I6WxME|o&$!c15ro7wlTNmbDz^`vczPsbz-hIi(Z)4(x?
z#DgR2TxjgFV>H(fE2FZ2ny7*6*P5UtLOgRv;?r`bmGm?I{=YSOH)(qdN)}!nq@P(R
zeW&$H6N(Q{UEur0s%Ds4tkirtYh+b^AoJi|7cms}%O<j5*>xA?hMI0Arhc;vJ=A*5
zLe(WTSf0;Ises3V?V)Ampf>p$CP+&is#bNbS^MY*37^&NRx7pY1eR?4&XN?@2mn#d
z*fS&=#E0{8{as_OSX`o#myMJ+CMe_}(rW0elg?B>ynVcIpkQz>h2?_}s5mAnGZCmo
zGxIk;&{A}i`q;Q!yx_<?2TcQ`tR#e6!d>(;_WusyQ_eQo@J+RO?!PaZNmLIg6%7vV
zv8v8bgZ|9Cz{68WC)n&?>-VHR?H&13by1e8>nmCfbZn=|K`<%fHYFMSN`*EV%ztf~
z2CnMX=w<1_I7jF&O;jDe&Bbo;*iNe)B}iiwWioQsHx(T8t^Jf~be*dN&(jEHC1A+#
zk!>@x-K4|&x4|nD6W`=<*%;GLdb<%f52;X5i+ECw$v4a}X^g)cv|0JWKDl;~pg&*0
zNeGknJ1&RpjU$HYU?lm$na^VlEN?Y)^B6pwR)u<)<*BE^?+U;lEPXA6n}l!HNKkXe
zgr>K(eMi?~%}_F!Igb466rfpI3FRhF-zS;JYpFB%l>KuWXzXIrdcEcgC_b?O6O;mF
zmE=Q+cmXzeEW|~|;)EtrttZpv%Rnc(o9u*(RTMjdDq+++I-?X#E&rYY_p_cp(WKx@
zkFEQ+t?%Iql1KOB{qIwpReOBx;2Hoi9~z`M92$6JcPl>85b!%N{~p#Cv~RZCP41n=
z%Q*bDq}P)XooUxbIz)JojJ#%lt2GC!N*U5peY`&bWXV=l+(iqQa)=9$QwDCc{ycaU
z{n7!ulN!r-3++R{=d@--gH<Y;i{a^4HaqS=h8-ItW7@|xLt4hu_SEg(F+N2~pPVej
zxcVDDl~|F;{l;AVrqkHARG`PaH=;2o?D_Id-%^!!DOCwY;<h{Q3`|h;)YObYPRlVg
z{&g|pG7=v0y>+S!cCUgMMdo-?^^zx-<@-lRmJB&W#?mX1Yy_UDrFM6GOK);FrC3B_
zg?%xqmM20(Vf58{1GhmYi`CcDO~V#MFPq3IIluKM=jG9&Y;Vqd5TWPuAo&!>k2ub_
z%F17_GJS-Z>r4R?eJ=CF@sUlH%!8FWHipWOlJY|T2CkT}ckGJ8Y=$JETvaTLmi9oV
z;I3mMF3zZTYHSga>0z+mbH?4-mCU?qUcK}kvFRGzbX$CWfhfHqb<T2qT%abWYckMQ
zoX+Yx%%@RFjUw~4U8aa<Zu>5u`>Syd1@MdS^-wcqlUDVRPK82r<uA%!5iVW_zjvKo
z&Fx6yS)+5%xiQ>}QEU#0sq15$#|6!8oBn%T#AYi^8_2M&ChFo5P3j{N+FiFsItWeb
z3j%D=5|neP>st9SBXN-F|Mu+|uTYMMEg((tA8fB4VUa)Id#PyieUoqMClB4-HXu;v
z{HUJAfCkq4M2=p`T6yb)2hL8CXB!hP<(OsjirZIQ;U`yC0&FU*kReSJn}`^d8~LZ8
z^|ZW)YKewlov0i?z+g0@0vUQ;o`W8F<LD(y$W&H9(RG+nS7<G+C<}aA!aw)x$*fmk
zG>6Rni;8+fZl-i&JD#Hy`%nEp8UV#MpnIN<ziRSDFCHWa$KpX(OGOFwp7u9i?rp>m
z`a2L~n|54PO^h{&oyX_U-R_GN#C<9>z#YkB`>&*`sG^}f^WyxJkMg$zkb0@*Z+FN=
zINuw2ZTKw;SP@ECiu^IEGXAa%h8wGgNp@kFOq@V6E-rFuf>OgOd|G@ta!cGIlU2Ic
zFRh}OYU!Kn6~tcLDkgQn20@Se?_lMt&t!pV?Q1jvK-mhDmU5EqLJi7v?KqCPd&(hS
z*p%1T<RM-o%a*m)s=Ji2#n_r>`8F^gI<ze_vV2s{PlsQM9CNdKzOznf@Qgw1jz!>J
znE9Dw{NjX#zkOTYl>v<s;QZKj7CM79e(iHGDgS!4!S7QS{`oQZHyyle=u?=nIQN9(
z8gXLD5huhB%k*vWuqVd~zo4>8?iM*b$sYSE^?pTt(1Kg`fGs|vSLpM>dw9+bb@SU^
z4?qN338^83SnjnB<Ta4`uF~RQkLn;^7Z%Xs+ksHFf;PW`-=sp5l81)Keumlr18Wf&
zBMoJc&HkreWQVzv+l_cRx?~Dsp}9+;DEJ>_^Hx2te@j_2?v-}7u*&keTcdbSSdkRP
z!psjGmEo>L%gn7da(*v<U^^{`HKp^J=kAI(^(2+8J~>hQc+_0W$966YdQwJ1*=z;b
z@Gu^1Fv5Qj>VjW#aQJg?X^eT@=hdp-TRx*nut`~blk+TxC2*+%2RF5F#-8M}vtu0j
zl>K0TSxBTz{sYsW8uOiIL{Fq$_@-kDb_@;qJUVO17~T>{-CkL?@cuaeNu}YPRL`w4
zH&9Dh90~JUs^L$Yuj}d>LX{HM3qCE?Z<TqGC!$L=adqg5qZ5GD7LWDHmrMVRQ+7I>
zD`%IeUgX)kP^t}|I=gKflGnJT>`QiuA~v$ZxzPpRHZf0m{`nff@tLA7RV53N3Z>V;
z-BFS*0LCLeIiaCW&`;5y#Vyxpw!s3T$VMz}E8~SL&KC5grT#)zl4K5BKw0cutco$V
z{;jf(X2V*&a3_FguEx?B#~}@PLTl0*4-zYO==WP=f%7-bqJ<`GI+Fm6gq2`x3Ut<2
zSqU@7J;N&rAzxfnp5#+>6k9mPa*6i9Qm!9U(XpBKqV+fZz5^39k0tOThg$$B*GJwk
zfBaHdYl|S%Z{N!MbAZr6OWeuI5~|QK$x|wuP8xj%AZ%ZR@h-K*{c3;kJ#yqVOwaCf
zA3dKEj8AQgm&q<Ai~#%hn(SbO^JFt}FVWl>W@go96WCFKwf5U=l)fgCEP7oAVy==L
zHZT%0H1PD@1Ugp#SD^Cqh~P^-2iijTkN4WM;+6oW*x9hE4yu#C#|QP(sXu?-!GNzg
zItCla;(kOTe$@jqI8-uKF;eA3Bp#5LPTpa*xI$`J1o^hl6?Rn4HT74=xWX-R)ywb5
zl3i8{8PE*%RHMKdHl=7T9#`YwWRA*G=o*Q<Jz!!RmZeZGaEu8PYW@G1XEi|Q!em~S
zx)7(Ls{8B~)3eSoHm?b3wixM(zX2?uiX%eI({tBdMO?UtBHi?>HQFpY?Spl2u_x<n
zZodws%p;n((2A~2xMh%4j&5kM2y5}CLI4%HJ=RIh4cb<Y^2`V4@~Gp~oEEPg>c<QY
zr4Jn+fAER~WdI#r#9QNC#EKtb$d@Go5X8pR(<lQ+;GW0>jgUdeK03gd1Zd3IZ?)QR
zh$EGrrn@=Tz*sbBKXxDEww?xc5fQ&0wQl9zTv@q0f`|!asdOBJo8W66xsDX_C`hM)
zHTstCO+!axI0F-8H<^)*&<Z904)_qIBM8pcR9$u5Q~KJ(VFDk%0dT1NhzWBhNs45H
z^jtvulPx$jZw~v18)<d2lFkAkq$aBq25)pAa3Rk03~^hd7cI^3Cm;Ag1wvW(2q*Yc
zRi+(w-?lbP5_)kTBRn<l#VD4pmQ#TkwwPCg*P3)F=E^OMExN#<L&#xHm=KDSNwObs
z@voI{^L7pL;rIA0U(XQ~WN6^~`-$0W(?dzt6Asy2EaY0mp;1!9GQozNytNtQiMMBB
zh-TayZpN=UWkz*}qB;a1fe(y%uDB&C%sFDGE-uzeV-?yg_EYe%iyzU)rCsJ^hI#qM
zHc4;e8Mw^W_VOS<D3n)fcJ6p5w--d$|L{%v=kZW)w{bR+7lY-Cwv!NS9>78~r`<=j
zm+*lkjEUMJW`^VL$g>B^vc8p)7#J`x9^y~3-S(-9IEuL8Q>qbz+(PlRy;^SU#lyWl
z8hce0tF9h}gQc+Vxy!G`)(u}(=jJZQ((esSO4&s7VUAH%93B9~$8E<Nvl}X^%=19%
ze&PM64vPASk)xC{;gG9!Ow@t_F;uW#(z}-oH@g(;s3vFnQzs8$hdYVTgvau68=l_*
zBwAEER;6>2+a%ut;)#2Ab^z7)B7><=uBtqRY5}ZJPBcS25T&Pg*GQzPoMhP7BH?-~
z;U*nBAKRqti<EudmRFQY4#0R9?Q{_(ATt*A-3>-J0tL0)d#jR?+zGlCsU6KjJ?{Cc
z+eeOP_$em}es>*HGU@9Ezrs*UgOkPoIS<B8Oa7t9kQ-aEaIA>S)foVxaXLBdh=rg;
zVY;SUX?mm}sdT$}$L8=4@(aWY3*)lvrhio*0aQQcziLum{SiYD_3s~UGoTu!d}EIQ
z+%+fCVGC@fw`lRx!s)vBb5kURO^%DRZ^%w#y*R85oWUHmu$gj9R_)|i6ljd6Ja^RO
zFbRJbRJJmsi6e0~7D;3vojCH94-eWcv9+V#$0IoJJUIh4Fs<YjYW%nbCIv-<w3*y-
zLr;WLuOb+;iSE3gei+gktAy?^@_5Kr9}s^&SUMwwdhFv&W>1VKX;V6Ax6wIhHEYa%
znZb#aI!RV?JE?)imAeiGbad!ee&<A~Pep$HTGW(`1gDLArGSahbj@aeN1IK~);y3>
zn+mWl4D0j=fu4t#g@}a#My5;?xI2u&moOJxdNNCDZ4K!iR!g4uOLQQtYD34zpFgMO
zoPSNrC&dmIrw}z2h^-a~2*vzfcf>$(hg^Yg>(OoM3-DM>HUXFCu46oW^MJ{1bOY0C
z_JeK>O-4)@*%KO4zS<Pb)pzZ5C|s!Lb!BWizww1epKGo?4T?xA);<{i?-528@b?VW
z+AtvhYnFhrMh4+>nFzz<n&WQmxXNJ7mHq`A{ehCEJH1GUj>>$C58s6!4~*Xe2xvuR
zefIf0t}O%Th=<t1HT%wmNBZlQt_@8fC3?uJr`D?o4&{$dqIu+#&~YzaRf)yY2j8_F
zFfmSz8(~|MF;6C+Bp+r~jQ2r1@(!W^rhxCVGG3m{UmQX;c=+M2A%k&MX^060^#abE
zH591iJPwG}v{!9rK62TJg!W;DV)(S8APM88mc9N+ueCOP_>G2tFuPZ5md{tz-Hv<Z
z{Q<pOZqVyGLsLU~cF)4Q4e_2@xjaD(bEEmlUh?Q!N+Z`V)*Zixu_;>L1qXs-%eL9`
zh5NM24%|%=fH=z=D}#-Icz~E*-ed<szZYxjySQCPv3VVmyXGI+i)@^#)l$Dy56w5X
zM{_6ZheRjUKNc;&#iA~jld2X8PkmY9N%YcM1pt?EQtDV>x}?z!p{-!!h<T@;dF9+x
zxgu-Qbw?o9J<-^6os%m4P9onjBvxBzBv%Au1S|R0IEzX#I`-e3<deqz#%(_0iiK|X
zuzuj_=*>Ukfxf+<VPRbgu@X{EQr;}p!_T6O9$~V{rEHp{Nr?NtcNeyN9yf)Ne2j);
z{}6Bfy-sGe^8i;k>MZ};l88&4LF+FO1l92R#Yb5~Pw^SJE*Nu-V=g1N8WU<gclKU3
z`{(hgP~lfWJJz9woxLNtFF7`u7MTo_e;`R@y-AP!=qy+xBmX$rn88?&>Zj`s^N<dc
zXlT_KHRdwdSjR3LtCg9rM&Ik>Sk5>%VTai^uQCagfC?J$R_iOTue2I{&PZ;0cPlhP
zh?y9dxMqZhA#}CmbIF;ropAp8P@cH^k$-zV`#bw8xx(Pw`+LLg=bPG@RhsS#X|t1P
z5MV^tApjMgK!(-8QFW~oUFb;hiByOi<IyCQK9(=|AK^c&ra>Rmu-=fjF`^(m0Tj3>
zfDw1nW-x1ON8||-+8@`GR>S$=<DZl!(+JGLZ+icY#X3T8J_kQ=bEI}5sav6@(y+I^
z{Cz9uYQInA?!L5Q(@BaL!7i|Hy~!~9#sQ9YSB=qPKA<SIr3-83s=VkSRh*E=ijgl;
z+Y%zu3M<b%;i!H7Dp)Sv=<)Y`T=s-^J|C+?=q^zdwvVO+U7EWFY?ou({NHnrWq@zg
zfC9lxmIJcdEeHw8Vo_2BDCj~eGTx~Up`P89*Ei$`2M75jI$wE|?+;qLV^Rs?wQu4B
zh+^9FEnMFp#P8Sf<3SrzJi+YDM9CVthC3)8C-o#oI|tSXLvg5WM;yi7T{HD8lJ&Dx
zda3pS8T&|z_5t|{?&x}^2g#g+A%{j5X0@pLr;)W|C|709)(5Y1Sm$-8l^J>T&51rG
zI-N$)_zzB8=O{nV6tG7EtO&Sm{xlKNH++Dh5Ko6HN834@PyGG?;T`<>kit~uq(NQ}
zz-AiC)qh`uFMp8D{Bc+)Nmw{O9C1%*qv@*;kDUDTcJJ7AeBmr2J2mk4{L2Y;)9h+d
zqcp~Cn`1C7{I;rJVUtOJpH>1TstX{EpAfDPHxEoeC@S$tBtI*`nd0PQdTSYq(WU<u
zBTB4xhjpnRVHZ2lVx#$IAiU=R(;U>=I<h@b)g;E*{rAjm{jpi%Ug-b~*<3(|$Dyts
zn-5dB@8PItzqe$u9Bo)O6SmF5z8Rb8WC*)}n+eU;(@kcL3o>#<G&lE#XcnCj<RJX{
zkkw~WhRSqOv{9l^3D(x5My!76oR0!PKe=g94_NRps*^PR0+_oJ6ELeD3%J6#W=zCY
z=jtm!LFu&lK9dSdbmUB62+BT0dg;ywzjmV+YkJlS5Xt)bts|XYmkf|nD(f{9ZwXfw
zgVw6lir_M1zo4#nLcfJ&u%mK=;MYdCuCX+B{mUQ<2kd^Hc22?*bxqR!CLBj;O7%=t
z>EfnDjAX`Ajw<nzhM%t2D~I9e>(42sdXy1Oh9<d|rE}XsdP%or!-|eAy%b++t_>oR
z4z-ZKj7BGDcsc3s=PSrrC>+~rUSY+Co}U8LCtuOmbK<gb?6A@U{(Mr3bqctf#HhK@
zavZWDNw%dj>OZy$zhY!|6&zECFq#_l`EZQe$nwWkfL~5{US6_IbdGx@3PQAbDdN>C
zfJKbMXe17g-+su^m0{^77dtGV&Vz`vw;VCRsk&HU3pSqe4kz<^b`>-Kd(f1*utt~J
zfQO5G5wU~+E#O<wQ6n4j%1^ap{euPF$jomnLtV)^O()C_kx!;Xsf_A;<hB&{zAGrR
z*4a0BAB4ToLf7ofBt(?aDtS5^iLZxkRZX!VA{!{~_3KSv!r<0r$6YkK*@n7eRMsP7
z^mi5+oorO!B)BO4Y$er*SV2km#!iJCk%Py~2DZbvm{IVcNbjrvwnJ<t+@G?r<>a8O
ztow-+B%rcVvAsZTP@|$p8as>S$XR^WxFy8e_%gJHI=B)b0E+*aIdPQqr;86B9rmW{
z?j2$(ziioYD}bvjIG?dQASV3yt9G(2TM8ReuPU6W>itpBf#b@sypwkh;?sHrE_(P(
zMA{IDyo2^@5q<df<tvhWzSi^Mz<wthY<zmZbJ@JtG5DQqUe-Sr&eyW;{XS+fx3xqq
zmhavPmT1Xps)|lJ7#Z2~WHQ>E3I54aNW^7l9kI%d#VX|OF7e=?p^!5?5X$Nhcrto5
z;XSG{#9EPwyJ|P%_DL00exx-IFc<&4f?uS7%}6^hQC=90)Y6+5go-J*mae6+Ih*l7
z*{X%%2IoOeNL9nfDiW7@7sv%aH(Bg3@dU2QXIoLcudgr$FwdZEW1Ya<RV=VIe*xDP
z!-!F?f&C69m#rv&*@pA>1KsR~9hXw!UMZp_%DiU?>9^^X6M2upo)QsQs&aW)bcf51
zr9i@B-Fank42=Yy(~8$xL(TTVBr*ftM_{wJrQlOJ@SN9GsCY=I7Xm)xth)#`tG#`O
z(1dI40~b*!>+KP^;)f(aAHvPWN1Z@P5Ow?cFgT>qJW|b6k7={=cdw12LnH5~Ur<N!
zbTU8ZSb-rfcX?6?_KG&pZe`Trk$9w`Pv>8o-`a>~wtt!0^s&|B2k3h$0?Z{XzBxzj
z^00SsVb2GmdsMV9#lP?#m9EvH@?7^5(M8BV=95|pcpx=>%u5&DG=p^O?MJ$FCkE3D
z75y4dkGhFjg%3lvKSd4%4?l}xY3|%}?z|glc0QK!PC9pU*|lb^cpr}Uwd1sHgS_My
z-T?Mm7t^|nrK)#$d|{%Ju$7xzhJV-hnYZOW7uW9?G+tkGl*|v&VmwhhO3~7D1fQ^*
zTY|vf859VS3Beb)=SLDf=8E(cs9574%4HAN*fEpAQ>$X=4qSJmxqF9!X}`_)bbW99
zi{kg&br%|&cqkiyOffRf)mJA1Px6?nw``x404xTxg_>2`<s_ReB7x)dC^FIA-No}G
z5J4e@U=%y~J=G1{%G0-xy>-{syh$vooI*+vPSp3+LtE;H<*BtKS3rDIhUs*Z*$L)^
zmto|Z<km_tM%AGRsOl`*cZi^F9*#d@zNQ)ZZ;V$D+DGS1w^C|RaT@)+f1l}Y{9jc|
z6@K4gLvqkh%qc$`zrio7y$JYFZHrRkoO0JUG{DxL0RR3<K9=phaVJ1|MbuT|nU&-o
zU45~!%k*+(VHw}L5{W57e(1g^#6jOfi=&P=CS%3cJR98ut3#L!lu>)ptSjy;Ak`=Z
z#HKY~xg{kcoIaHFnqXNwfBR_j`ORFmxol>#vs=noWa~FB$k2t`juZN5W#vpm)u8o=
zv|*UjiG)uy+M)|FzI1xMgX*f|om32KTv96^Tx-dWV9xqBpUweN-`G0g>PBc`)cg6i
z+{ctui9{cUDRF#*F-*z$l{J3!I+a6(?HUhM`Kzw=L+1-W$Ox*%^XA{(2udyTW5&6z
zyB^hz>^0IMJXza;N;OEpL|U~I_b&H-4kdx}dwn(K7S4X?<&(3jWrfqj`A~gREo!05
zt78|yX4$l^Wq4|`KhY;vOgMxu<e^a<DfW+`I#MozOYv5aDo;$sZ`=E6TM+WTOf7*a
zT>^A&#Jak(@*rpoj_p}t1`KlGMtybKZ4)%Bmdlk_wA3U#_Jd3as}NsYJ#AznY`P@=
zJak3J1^<J!ygImg<cBvlwa;iMD9}DVzwpY>n!?L}+G8;=GAUkgs7DRsi$rVLNKbN}
z6e?+$iY}*Fq<!muRb6#d8%?`^+oFXc#l1KbcPLh%c!6TYtwD>E0>O(Hr?^v!l~Sa*
zI|K;?cb5dW;ClBx_dDOYH~-}%GrK!GyYtNR{A2@D8!oT@p;Q-LrB0G-u=I6cRI^u3
zu@XJ@-9L)S4Jn)b*x_db8mKepdq+bYRV<VE%eu6dTj7)5>V|u@rzSKrs&)XtRLu^%
zy!>)@>9U2cDq6m~#IOV+lgQS*)3^sj_8YH{W>8E+{T`lkbD4mM&Fx_-`7<joOn9wK
zq|fEzxkQtIiPXd2xR2FNRPMkA?r`8=IL1^+Uvh%6OrGQTIK#*bJz+P6(4n6k`Uj#(
zlS-;`T*sOQ>MwhBH5iAOwE>OtPjjO=cAt{f3P6E$Is!LU(H^BhsU9%X{f@MSS}m6e
z;86#qvoIVJ_T~BQ;}#pLI?(zj4L#npZ#Fm{(~y~6ze^;bX!yWlOcipsR!YSw;g?Zw
z>X&#BL!W!I@c2Yse0Az5Aa^RC)$3q*rjjOHy;!Y!!s7mvgi|rdWz@<mfNs(+Y%Giu
z9+Cg;#IoPiIlPL~#WC|ITM?*ueHWZKb*EY8j9AlQY;;i1l~OK4fUgIw)U)5{p10k6
z1)D#!a0$udeOky_HO?@IYv0Vgv1mOKnp~%v=P<6N?aB2;GohFQ&s;fE>xho-wQu__
z*wTj<Lc3Mt>RM(Bb){A*{Ee8@M%dLWd%B}nO!V5ygPn42x-^B6v6bfs0e!xU`G*an
zgY*%C<JDuYH@tZ^#mT%UHU986RVlQ41}2*MzL1EDa&D7v7T0*ZAt^}uGVe%vKx(hB
z_U7Z1Z7tjH2VI0Z7yD61ypIv=Abh%r@{v^WeHY>V1Gi!6?|$YX0Q(0~FL}w1?md46
zk=Ag|z240u**$;L-0VXw>7Dcnt3PVsKDdqfsf9~u-sDqV6)h`5tBmv31Xt?`ck_~+
zc(S0H(Mgj)_Mx3bekD6PSgl}F&v)~gJze?cJ<Zo9%ni3vu^T#K*77ntqA)R`szwUZ
zI6B7dkv!dq+A0<MYtc-Bq5b<~{4@$jO)xd*kSs^&8FdYJLle!(J;$>k#;Bf5zngoZ
z*p>$nf1W`zB*NMz<-zb9L6AqYeD#v`u&psvRw-fx=g%&8fHQ8f7vCi%2`XAJ0=t_p
zSO&i+l-D35feG6yMYnlv-CAJs{wNuMa0^F8p#-ViyIi^euqam0>NB2QCr~*jFk@AO
z_L`aV*-=|5a+)gRvf7#6rkhjcp-DF$h`J7dl0aKnu2Di<sHH28IU|$YltHm43Tt$)
zYQ)2V#;$rId4CsFx5$Eb$TOEnF4`V%VbS#^;@l~Pe77*&&Z*XK0OtiemYcBbX#B`*
zW0s0XTdDgThs^FkAT?o{D@VMZpX_pEkvzD`4B@4a@GfK`vxN^R6yRxbj~k_H@QBt6
z)ax4?{+RnEKpQ9Oy~$SlL&0=Wvi;U{y){Q~#$(u9g9g&J*c6PE*DM@n19upgK2@b`
z@ur!%>^%SAz1w(63qL49qD)fMsady2@^u3xzI;QRy4G2zYy-DoET3z)E~e(UtEGq}
z{`K0_lZWr}2?v9y?uv)BN!w|jf626QPA4tBO+X|f^#wj`BH`(fd1fkE??_6Duob3y
zr`a=>e|dO)&$Y`Lub};EpthH3%r^OZ52@o;l@5>XG}p3N{lJC>_S1kTa92Kg<s_>Q
zq*x~B&g%KtO>Enooa6e&Kl<Hj#v0ClWcaA)@>=0q4z8vrZD#91Zg0(s+ZwveMK@<7
z1ZwsKYGQM(FQVFqVx+H+MBfK;l4vO{dcQUN+;)O^lX@qNDyFq^t?a(qy}W`y9zF*3
zP81)LrHI~-X`HSD=vfwQfCh6ND#kta%$T!9EAcE`RBPE?a5flA_v3-2@k|k$R)OwT
z2D7qk330jJxua@v^;>7PE)gAhg%or$o>o=jz5;n&$stVR$tliwa()n7v}FQi(#SVi
zW$k88VZlJw1`p8J1M$hvy0o#}#IC}9N%B*^`u3-SGAg#iBNolvRtAoe3xQD}J!suW
z_P3dzhg#oLWc$R3<UVSD=1}af%GyfqhSgZ_WO)TC&w`N6k6|Rb=Ip^qq#j(aIE18I
zYPw142f%Ss^2P0PN4q$8rH>1bq3IyS69mFQ%s+#}UId_XdFPJ+YrV%X)I-oIFCt9L
zZSbx;(#^Z(HJ6O0HI~Ki@_iLeHr)fa=U(yee2msY2_`VoGD>I4KWw7ceyrXFoj>WW
zmwAthdCrBc9Xu_j$yQ<?RT#!{%tU2G)Y7C;W?C!FE3ceoO|JW@sF%jdbc<iux@i@n
zo3AFAa%66#KGk8EOzx;uoyD&~ZVU;oh6lvl3~H)4>g($lE#z1s#)`4>rF>Q%8B=uK
zL#z#i)mIbAB2z04-G=Rs$Dq<dQ!7@tgE+qTbobJBbW%Y}JC%?ta{C1%V139RGxzW(
z_3?JFCClI+Ds4=myCQ1wNwN}xQCaj4;R;A%vlaIeVKxfg%JbDPz7HYjXTJmtGqs2#
z4fFX9^e#KsS-@a%l+yU<^dxxT27y(NAe@vm9_l2H6r}+TzRbH!FP>T)O6Tr6|Dcjr
zA|Y+gT{_ockdluI>Aar7Pa{*QZfMZr3)mcpyuN(+M-fQ^wgX{G8I(YF2D$aJS>Tcv
zgq8u}*&IB_(8u!t8Dzg^Rywz;Vyvs5eJ}&uZr_LYN<Za|Id=datacjfiG2DKO8V?)
zh_~*n?Hs7x8~^ip-NBy)(|7R_aNK#F91*<ogM{G&Y*Wx~;~xnPCzP=!!b|sLF8j$$
zY3%o-)yC<tcSvF{*~c!bmXo`gUlp|S_MLZk3`--_0&x=Z<S(}U9^+ix!yK3Qu4!ko
zFo5&J{TL8lHpFupvvR9ggC9Usx@567`u!PvFXl}(s#;%K5&pvSBFji(=JtHZ6U1HG
zcLB(omkCQ!#{8x~lkD;1+LeIE69>Vt+@1xK59;LC!hv=Ev(iuMQ)7Z+IR+s)=nEHg
zrhGmxQqXA@L-a{2C8{`fV+5|Fn=!uHa9AUf`j(0<xkx(J>-^N6`{2a@uce-I<>{i0
z7-|zbYqYSrgUrcW&a%3OX}8>-bAxwCxhi58e&v07VK+GSkT#{^7i4xncx3)FM3tvi
zzM-!BX3BjLqzGBBHd`%$(#mPC-?*ocBJ4bN3w(&iTbKlNKi!0rg#B*zEGb2OVSGy^
zZzpJHz#GBfaSKq9u<n;L#is~Mh)Ir}eNmaSu~2}=sIuftpg_kNF<-G@sjTeuYKAvj
z(5gtL873YSn(>O`m9n5FKLgC@3zd|t{Wx1u|4W?|i_@2IlwDc*s)xAlrz*kO*r==v
z0iF5yq_E*uzN5w2^*)#PEajaD+ETb*7*U+Wuw?~&K%m%8uo;%?pKwl=TP)V1BJ#`P
z8c&67ymCG}`+i&Tx`*uH`4+G{-Dj>FF<BpPjH;m%8%`aO7#Y}r5{|Fl(4xTAlbJ$4
zTlowOYK6vjWd!jWz~pZ!WJ+GB*qv6rQelX+GwWT0z|19B-j4_8!Ry0_U$Q$UX|>!9
z4zayZjHR}!6Bsh*y?cAaW$R6>so(^A0#3I3AdHGNZ)0Io`L>b3OYzdmebJgM>E}0)
zPKr)zdl8qy=Ou2B(ukM3rT;8k5KLqyXoa~NR@F&vuJ>_gZhX^mGQVV@_CFM$7QaW^
zaY2K|X*!Sadfq4tn|TEqKJr)le?9P?EqaAVRDK~h50`>!=U!89S$4%#lQll{Em+#D
z<e|1H{C8%tV`;^L;(@CKSDSnu{%Y%N%nYGQ^FX`%u>KJ-rRmV+BGFST$zt&?0C3Yx
zk1wFuqDjNiRC8xbUf<7j)=J5c3y-U;YB&u9vj>N%5+b$X!%2-@Y{(8&ONi)4ZN!U~
zA~U|YPlFQjRuoB7Bk#+^qbBS$pZ5JuQ!U+v7A09Euupp`v^4+i?qg$0w%2cF#FM<F
zh|Tp%fbjNYiK@5GSshL$I^10;PGY9++&lrSw?DiL=bOPGdIR@Qvhlx0C2d4r&(EQZ
zeNhqzityv>jbc&%vu%?!4as)s(p4_=@{oiYQQsOC^dkK|fTfD`$|^YOUM%8NHI=@_
zP_YBJJcH|Z<Cr$fX9L}5u{rRibuZ%NkdsMNAFje)36k9^hd!y#-S5tcta<L2gvimk
zv4D!|bRCBA;v&d?w{ZeT>~0M1vOl@Olut-KR3ah{%k~a{8^I<19rOJ_<cw*6{qD7$
z|IW5+ERC3!$wI3c8XSa?|D=2az@*-<#g8#D4J@yK3yGcE+@p}HrDunM*T4(FFix`T
zIL7n4p?1Oq9VtXe#)SN{sX~3xfOaUO$vC8IdgRRk5f?^2eI4_!Iofo>FR@&sIk|Ek
z8J11&Y$MKSICphPh}fHhlcasN_*Yt6I!)8Gn@D5X^?BuDXQgaLB>JY$RgIgyLltwk
zS8t>q#uR70?<pO5e)NNQ!<L!yMU+LQ>S3^b{;JZFGHO<XJ=dBE20e6}fMZFi)4w*Z
z1u+C+A6sLs7T&q4d?Y4E+X`<0J)0u4;E!l<^$X_ED^Ii%OZjtLV5ee`aum?Q$^@{Y
zl*zSHKnNt+H6~!yhB$hJ20T7%t(XSckhh$07esauN6W@+U_Z6^<s&8#_>+^9N!Wmr
zpZs9#kI29VmMXc#&WEO_?b`WbB2g2LiTQ+uvQ5G%SZ{3;7_1yc5YSW)h+ivIA>cg&
zaouRm<eMlB3IsBh{J^QI%_BR}CIc`OhS6ErY?xU4#t2XSs(fH7q$`r(6M%QE(|@V|
z(dL$Pe8YlrkK>P-Z9C%zlm&_5t^_GRG;s)0<g+ZN4xi|v?}t;&lXtOM_<^F%&$*I*
zhfUg>FhlfEu}2TBU~0@6cZyhr9@nnL#ix(U?YBoK0aqH2iTG!2^Kn_$kP8zwoafs7
zcrRixA2fLD`zOAQS-#7kE_nYt>z$Fl(oFZcA7%B@XF?uaS$58ySD{gaVQi|6ZB!@I
zeIYmta^*+S>D#!f<h#8jO~tu+W3~c2N0lBM05VDNn2HOw|GKpn1-t32xotD&9oSZ+
zi1V;EFck|L1XFS-Dmgh+>44eSu{UDeW3%4`B4@F1c4fZCCx{OBSN;6-@b!80A?ZB%
zPc{pCaH?aYp6*>#q?e$i+lsT}$ufbg4oOLDx4GymTq5fx@ZH7H9iDTJ&mM5=F7k^x
z;86)(N%fHO5}#dRI7;x7D8=X>5w-+0cHG#?N@}|Lt%<C?br?|T&)|o~Z&HYOFfQv;
zosJ;B)-Dm!lb{q_W9Ag4RGNjl8$5AP)<lSpb4Jn{Gx)FJfih`h_yV-q=~i|-Sf7Q9
z8{LvN`&gpf;~3I;^_xddCRk6XsqC<UlzU<{Cy&2Vj-qB40)b~_)NBiK0<|4o-|~6V
zq^`Lr5%+#+2E-EEPeiW5ACUo9fy;WxF>MmmCanbglfs&hWU3Oh+wMlt#-=C-yxPcY
zTr>L4pszOjn`fOYw?kZGn}EenZBC0{B&mS3HP*RzlxskEEtQ^_7Xtgv8b_d6*4b}Y
z3;;tH4UL}?P8!RAS5Qr=glcIHW#F^TlVJPX+<K(A8$Jk4%W%SY7m{I^f7pqNyngQ)
zTj9G^FM2zYDShq?APEjPQ~hJW8%YCOR6r?sz*vPfAHf8h#_=VK?CR?3^OT^r;G112
zj0QCgV;Jgl{-Kuj5`KNmtH4;(>F!tH(ZPf}6qLHWb5e`)Q};XJ@I^$gC)4?6I=WM^
zQ#zdU-?Z}0l$&mxUe|a274<mf)5!ttI$~%B+UW&%j4{u6$aTr`&Mh&&U?paYmGlTB
z+9;*ooouWWK%6AYD;P{Q_g^vF_S#Z^M?~JM1mj<Fd~odSb5)^U@!Nzg$zsW@-m(?i
z)*D8%(r=uk74{b5-?ic0kK1S_W<7)s>0Vw<79~Z7CGu+n<koL5V?uSgB8A>#NWM26
z#q}+4oXE(0wjw|MOz@S^R1dz%R{KF8MTH6N>8^*&WrRlcXvD1lIl5bIM3<Gber4+v
z$kQfuh-_qS>RAeY_FKxB)6XF#81hqk>dAjD8M_wa@xpOEF4;6;e>le_-*n3kTT~=b
zf4)jm&@K|Iv!btIQRW~2^N-ab|MGQb7+;vEiqiUKTU2r=!R*Alt&gS3IjKFGJ>MA>
zrH^k7uP2K~E!2<hn1OV=H6ZgeoL0i&J42w6zEDMJL>!CpA!!mwFuGzJa#^lg%mnYi
zY=W(tl^EHFg36Q!=cqv3(LX%)?Fb}Ud@A_)DN-sU6vY;dBr)!vr1E63+D;mX`BLY-
z2qlWQ!AO#!h~ZWh#vrsP>Q?}H2n!FWDLZ<EfjGDjzEK|xIU8OwzS7{QR%M807<xYb
z2dyx`36{@RLfNk#XL{)V^u17A)o(!uVWR>&nFOXR39r*<?G->I2fWiHieU~(laovP
z9Vnx8oWgWmXB*Q@NBXDudw4KigG!R37FWj}@n>?V&1%TSSIaAYqS`2ChXl=$3y(O{
z3iuWd^4QtxaJs+hv*6EVo8>qx0X1Rj>iSxFClyMR31z7x;#npiX;fP%X?!t7xmBQM
zfq_iM<6ZiZ*}@U)W!;;|L@WcZ94j&91ef)z2w-BQNM9@Kq)mwHd6n}=jW<o9A{%?U
z6l6=V6y@G&ept9{nR_>VSAu+LA((Q&;B;~~Oa)q|!3oOO_&H3QpFKmKjz*Pfn^avr
z1>z-lx$Z&_`k0;E%?b#_^iSlTNemLt*>;Wev7~dKG}FX#>*Xk>^ky;g$^wkG@+|M1
zR7+_a_=YIGa#irg)cshVbWA(Jl)f%KQ`yHv!C~dZ5ba|1rHPfc3i&12iFVfYI?}7~
zl4s0E4h_#Rb#jj)hSYpvR$b<!%HYeV3yjOl7DWs4c323bw2p396DIS7mrxHj#~5x_
z<7H}8ctsq<=F`cQqO|Jj%$3c`jZKUHHTCbBZGAD+Fl9yi=^Lm<_NvC?riSR3eYSve
zXEud=srw}w!{;F>wk98v&}S}WeV@%usov>5&OHWG-8*KKL;0-6nwtQ5S~0mHqL8lj
zY!p)Tesd#aAg$v(0)3L8;M?auRCr6%qk-HjtZu+D4z{5oHBwtD!WIcwcwPb2R-Zd!
zU+=Dh_VDFJIZB*p`Lgd54qt1S>p0nnZJAv{<>%=}ZR_67H<Z2kNvb2c`A;^f(n1k2
zKz41c5kuVS@pD7DQBd`K&bU<W`)?w`6$!}5l!J*fOyf3}^sOlouQkOLWPb#atuBE&
z3$C}Mp$cP754n5_(%}NMi&a%y3T&xY9y<`f!cB-dDWY0LHenGPFk!!o0!xs6;&rlB
z0%un(wL=G4P|q<K^iX|=O^>EYOsGH)jir&pQZAt-KoPO7Z4mX{(^`WtOSpIw4Amwa
z$P1V&oh+OfmBSf<O!j!4E0xe6!>?bme-CnU{4Ulp;E&egc2tA7YlFF5u@PjW`Yo;Q
z+4Vl+mR>Rwc-xbl<z@yOT?@{~efAYVILEGIkPre2yO(klqG>6AVV6@a^v%xjWvySZ
z^li8FCwqyG5W;Hxe&wEl{OT`8PvjPgq_-#jS`SpAEx1)icPhra#8=$~N_Xdy)^GJJ
zZ;-uJ0v^hcv5+Y_4mRCdF-m+zx~qFwmT2hpyMBO8=Y06d4p2$bnMg*av}89{Dq+g)
z#vqcsXi7<6Eq+!Uu|$_OWK-pfWZ-s;@$xcp74op-<N{fMJvR7lF6;V>VKz?(YE@?p
z48?{@D_VN~CbBy7AD!EB1dmS5nHuWAs8R|Rn>usvGApWt4&|r?!b8bJvm`u1oCM6x
z=l0LNzZAY2ewiiBijW@K#HUaDs1-*2=8Nq-UJ2xFPHtE9YXmURd0e9jhWB_mdw%fn
zWMpiNQH90K=WpT=sT}Wzo0K*0_4VNqP0A?nf_;m?Ec4Lwj5_z#>+5>>DPpf;IUh;-
z*#_`_I?1gRqW#ewUtU?+{f$V`W~c5m?!o8OQpbToupMJDO#6bBC7=OT8eV7Ar?Om%
zl#OT6r2;ZGB!gPoi<qd}J#AL>T2Lvc?>=_e$~~Wq)+Z{enUfMBmGPP$&sVaEOc_<M
z_}Bt7+Z56aU6QCVTb02-?<i%ytMv;%Z}V7t12^N*huncBQ2Vcu+$~+w5BHBLOH3G?
zyFWu&ZN{wrF^F-#-^t|_5gBs!->L{AjlPw1YPo00wMQejh__uAWsGBZ*sU(Lbc6^{
z2rSOAbbYd~fU{DOLYt+&b3pbdDcl6@viJ_qcSc~9?(W$cNzuuO&i7MyE<eYPT}`6E
zNS@=(jX2E;Ob4ff;|ocLsfS4y&Y=8KnAAPi23u#2<!lk<{5}S=3cImlJRmxbtz+;t
z9@nk+E?Six?e!3$f`xJ*Z@j!%ux1~A`Tp_rXrABt1}Am7>(-5U%SE=drJrDjt+LvP
z`53whE3L{h<XQ-zhh4*Lq{4eXH8gi0CPz`9Iz*3R6)Gp<mf*A<Z&Q0QB4cVx7iZ@>
z>uEf^8FNZi@jRa<*|^>vYpVRt&V&g5ocmQx-$AyDC3g|k7>4;nvm(u`**{yLSbfZP
zk|Iuy+qCJp(`@nYNnANEYLlJCHHf-TzfV>LbAgA?a7{EGTP+7W`t_K?sH%$PP#3w<
zI&mqRiwTuo5!c%=cAknC4Ox?R)Lvm7<mJxBo7r>$;Z&_AI_xC_cwJuzn??#)8;+-&
zVQ`+)MMddAUf$Wcyny@X;l@$Fn@bsn+^F-e3W-eISP}c7o(+84M55f{t?@QbJ5^ZU
zUJ0LXktD@iDl72Qj{QQnr9TeoB#G~WZ#?*~(1-jGJNNbT3r=K>op6|n=n9ESrNc$7
z_CFWU$hHQmjQIthoR$V#P9@-IO`HT+H+B(I<xY$S3G})VAR|nery{#_+cI+LmjY_G
zS`fiBGqbW)M%LCD<Bwm6<}MaIX~OH?)wa4wg-pve+XT@)3O?23YxhW(D-R{aNMcas
zcOcG{oMOKGG-CP&dT#1G=-X8-H1N3?qfHIuP7?Y`m4EyhVinteEnQ^yNjd0mTU*}%
z5v#2r2O^cb^|tfztEANfFw_HT@ZMWUc~MSOQK&Ehz;Dj6XB1-c>KkEOORlBY8^?%Y
zgs;BtJs(yDw8?xHf#{JFUC188!kVC|6ByX53wdBW=5^5+7aZ7mffu}|C%QQgpl>*Y
za>d5zs~MLi9!W(oo;`(H)svlmUyv|UpzlvSBSip@e;W6$zf8Ia)sr0izkgkuk*VA(
zRR0#KVpyATyQP)@p{kM=%gb9HMqCGcCiS6JRoxem3KGlj9o3F3jqoU7V`2My8_Jxh
z+2cFBHpi-Ap33ZO*4o+LB)F+Ey`>tkcXT}CkF382)RG^+I=9`?Xq7$5E!yVSlUAO{
z29>vl*q{wn3q4#so(*OB-l6G@WEXqxJa^l}DAnN1)3nw8#CQcG!l{|3Mb_QC7se3W
zqpdT`&kezXDySas%1nt5oRgnJ^^J^Rf({JqCrmIUnf6J{f#i0j74(%$QU5R4>}WF;
z&XujT%K3BVMH7Ht2jt;d`vcQ15h=ZZMwSBL|B$-;wd#lTUl%H7Aw`gwaGjYJ6F^Ym
zdynqCvRzG?Vp*66QzbQ&V#R}lPJ_UE3%(oPJ?B+JHUYO4C%{}DJi1>F==3zZrcdPT
zU7biWY3k%Ne2`c|Pz#mrid<~1+%f0e%qH$H>Vx*~(6g3Fhe~z1i}HHDCltrJV^QLf
z-4c3Nsx3WR(+`RL5ii{L)TRlvFOC<SP3zRI$XdxpL#3EpBZ#XS!`}{pWe#INV>z{n
z#eyHcE-$N3O-UYKt>H-DbyUzP$bCxYN6oUoHysbS=G>$R?(cJa{kgo;11+VvUppcx
z!5^3GLV>f7*$}57@Y<teS2l`aSggGfdMTnceTEA-+oy`2Z!_MZ0v=XQx<$;a$)SMB
zS(PfQsO*B7d6Aghuv>iLT4L@ExrWp|Zg;kcFgzQ2;9T9Xt&z^*Ko7C34K)Kd2r3ig
z(+ferT9M<Z1qFY93iWjLqZ{&|*xz4!feAw+?E_av1r<$ikAG0dx0PDtoWG6wSUT{%
zd&p85X1Q3r;=I~KTZlZ~pld%1G?R|Tc=tN58GZ+~$?GQ%5q8>O0s8<X{{$IM(RN`=
z_Sc$}qm`qlVl|?2={8v_fL-7ow*BWYjfcJp)HWa~xAHi#vVm!1_K6ru?n_xU1bc$w
zuHA9EwvoQY&hbZ3U>Gep<cP;jtXSCXkRr5BxARD4me$tU6hfZua+~9@jK?$ZB}6tN
zbzz>DE|~X`jeF*)TU@ZxtI$u&4lZZ?{9YNnlrOd~J@M1puf}ocuFkt5N`JJiX(h$Z
z@6`IW!@kFOV!}g{aEo$ydYH)r1%pEC;HSn>5<_C`a~t8A*n7W+plxTRS)$;ufD@j_
z_Pgzm@G@?iM^d;WrrR)7>Oqq>V3#o(s4q)+;@MPoueW$+B8&Ahu7tve64Cf*HNm)j
z{LvvSeSJ!S2hEUr$f3F<a)lC}T9kuNewNt~tGYIKxqmImwR=(jLd3~f4oPwFF|`nx
zLw|T40XMOTpL;JX6ix=F<cFdOZ;Vz_^|iP#ulLgQMm6hs`PTJ<6iG#!^g)gX&z+aH
zQRgv1d%p+LS#uEJ_A{{u)FVQAV^8`)IAU5KQnSTb-KlA~Oy4Acg4OJi6R+GK;(VKr
zhu;mdD%E*<oy4^b79%r>4P>2wQ0@I|G^gqjr=;p4-#<=7ifmdcFAJ~cm(z%-<UV#2
zOGmm2WHRKq%)aidGsr$j(7>2fk>~uu`%;qY_EA292nvV=G8RMD!r|H#2GP)F-Fzhz
zx2^dXt2PT4E05XPF)ws-Mt{|&1+4R}FPW@I&0sS|{X$mq>wQQ?F9dxr&M?h7oi`jD
z5=XDij?W*aEY`Cwnb`xhd|mv(9rZKX+$8SmQvPAZQm2}2CrcG42YNX+X?;o(hCk(?
zCWTvNyb~EV)<qF-Px|(p$4wL1I2adaH}@~wpp3CkOkR>EaB3)pcG#og0zRHyvXhe3
z`ARYLoyY4gUFl2Tx?RJl<VEXLX+vSZL(uUN;*R5EUw?2aqit#<ws;6oQ^_Piez9w=
zy28RJa-bm23Z6(tVQ+8LfeC7BZk0(doV>?aQdB2&boy>=!WfA@F#4%6Q=5pnkda+$
zn^nx&sko)dgkV494?rK_`SMn&uCA`3L2}c_9vg}J%4idNvYiM&9i)<8#dL1zFcno;
z9Vb>ZJbbC<nXYc1lkP!9PeR6%W3!d*X*V)WQBi@qP~~xZpM(V>BF&J{7zpWbMho#8
zc}283FWSc-6ptlJ)hEIC%|Oj<k*)8qYL%@lM_jsl-o&p&4F$A9`W%+U_cLtmynh=i
zJMy+jMzRBlWR!u#(r374lGq-8w3n=;E&uJd-On<|1P+mJ3^M$c`Z>cT18%!sy?6T`
z2RX#WI@Npfil7m_+~XdAG|J{u*^?C$Rj>MAu4mtgBpxQgfX?hC<e`>sERIDR!S&|w
zC6{NYx_xZAIu$)J*+^^m`rkJ)3sx;{BSt)J9oSx&h~clwDHA+QkP9_9dW%#9<PRNq
zn-t+iKF<CLVZ2?FvAIDnV{reN_S<5LhZt4Jj92jo@L;j|+rtHmimp9(Tn*-5`IC10
zqRGa|n-&`M#rAK_QKO?~0q=Xa<#}~4tKuJZRDX>xMcYOdC<dlSgj4G<laU4;YXXXp
zvjHUJe3<SuznZ7YrxdV*m_&)aIl$9OF>d=3C`ZOUk?zcAXok#A4S*49zpBJbE0!e?
z^zCZ-B~kB4?Xv>z%(UoA)P_eHN%DNMrm+>;B~B!K;HgC7;Nbk>u5fv{%fx!llK7<T
zg44>Nvdus45)8(7-RO(CAC3%{ek6TZY;fH$<(0O4AlpXLquPC?@zWo-ZoT^Ncvj#V
zK!)RbN7b)Ph>D6zFmpKgd2IMe6kQ)VOCmO4*rUXLwC733|5z_T%l|k;_HV<eaM<H%
zZ!3R$vHVgugdBgvLLxf+gqOCj`>Hqg)@WV&Ki~fIlUYBc;g`?<`x_1cKAHRfX$K(>
zc(T6s^YFhN{o9o1t%UKU<p1sK4*gqrRMY<+>OY;g|8wG^dHL_L|9c{2uO1$;3)lsd
Ti`!)W0Y1v|>hH>B&4d029AKNU

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_014.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_014.png
new file mode 100644
index 0000000000000000000000000000000000000000..7322d3e4b51697ae884c7c5fcac3cb04758c84ba
GIT binary patch
literal 55859
zcmaI7byS<(_AQKi(cta`C@#SvSa1lGV#OVbySuv-EgGza0tE^bC=SJ?xVu|%=klI&
z&+ndlzcIc)vhyU_PsZ3|J!|c`=A1iPT~+=K1{nq%9NZfPhzt}C4#EBPIv5S*bu6`0
zc6ogha+B3@({!|S^E7d>fKxGXbFz1Iv$rv&@vv}lwQ+Rd<q+iHWv8)rb8~VP=H&eN
zKNC0{U931qC8hUX{{-C$qU#C=hhy^33%*>U%mxl_zCb}nO3N$v=;?h8yY8pFXn*~t
z$OMp{VU_0cRC>B@I-`D}o-qxI<QUdsL;*Zs@yO;`I1b+?rV&WaW!RXHJPwb-4tov}
zD~OhCdHRc+wuQF#LC%4#e?1oE`Y&YaAM&}W;Ry4M4k`7E$82wp@<i#n{7<hqSM90-
zs~B5v@(f%C7&HDoP}wBEy!;CK?{5sxr9W^qZ>1>xGYlEQ58$}}=aUjgAToo*)b_uB
zd!6t|nDnTg@voWx4Cn79|Nn26d9er>5b8fTU~SNQL)-JxSPje=Q)s=ZU;Oy#)2F2K
z&e7AmJc;a-(>o)Hft$zb{4L+-S+(cyzRwxLQ&r!-xm@l|j2&;d+`UvSwRk4d%O{AA
zufEQ)R{FIY_{6Y!(H`)xd+%QAcLOV=KU8TWuH`@GKifSIyxie5x$mMtSPq4l$YcW^
zPFp*++1`8I#TsqR4Q#ShoEfg+(IC<J!b5;KhF}_J2*pY%iP_&3`Y?IL-~>~;50b!m
z<I5WaAmfT}?gb0ughm(?1Bh&6%ViUD^E`ek1Q({xLP|>?9|urW48vu?8TGAf(6Q?e
zWDDyMa@{-a+{8ocqQOo~!e?Zp!39SgCSx!XLKWk<8M<j_vd>mVLoPs$3K57WOJ1=!
zt55Ie)`^~e8QYCbD3#rjw_o_OQ;9_5uGh8l-yR9))`6@bDqY50=9xOm9>+66m33ab
zdpIbMvo9kO<fp5lzxNyi9volZJ`27q1`ZEwmUT*v*?(W9I6r;9S~LI!CI|-HYo4t9
z3T?oqr9i-0iL@JyNC)j=8?J07c`=luRk`cD=X4r1n+%)6svV`cf2fHBhqX1%m`(!h
z-i~?~L5ho67OO*75zkgkKg1}#;dpbn+18JdG#>yg=?f&kHl)7y?F!for6LFz6GruF
z_0&1!8cmq}V54Gy=JPcT3<*!dr)6|D*t*>(_C9ECk7XT(vU*!GGj3Z)O5Am?S{<KM
zVId+E^3k<m0|^yv-2j^G0G1w%s?a<ju>bxO1Nuw<%hStA;k9VLcftk5uF$A!u0%HI
zMTzo2j=ZS}G!#uRwDh4l+WnJDuj{2(yr$KzkjED(VJ$vecQ?Sg0bI<`b3i?9adk-=
zn&bxV%7$`CgTO;;0g{Fm!({m12soCK@q4(q$m$92&@`$lcm~A-xT7QglBphz<Z@hk
zH!n3H5*;s7UuJDGSzE4VBvzFC@3ECJh*T>o5THjmuBRClYN}^fQ_<O+93m8_z}+`~
zimvB78#N^WmV!}i&(+PYtS?OC2)Umi=%LaxwRT%z9QbqP_K~GP!H!%E3ixr{Bz6ia
z!U6`7H>XnOmx9YDL2Yf=YGQ`+3bG5z*2j+D>*jYyKC=NZ-r*vgQi9>;vlC|ID0mRS
zXbebZoB(J)OUd^u%)S-hSZUf^5~CNi%wklOph>a{0MxoZ;rorVjuk>p-HrR`nr19x
zivZAhrI^PIpGshTLF@YJ_gv%{z^lL05HmP|LAaH2R6mZAF5WoPtMVoY4Tq4O3U@Tp
zE-eJoPa*afVf(<wO-h{|y(q=l;b$i%XF!(3qE2_FagW~*959mELHHPzA&UAsd;z~6
zkd6{r+5F*}%QtWHe?mvClKWDE?M~l(`TSutPhzx%x>c>EW$4dX-u860c<191#qzU2
zzMo&LCi5tim53X>)(Uru*HsAc;uOl{W5eZnT(V$ZF5sCT3j+hgCtZ%D;vFE0{J*$Q
zdDXoW{{s8G&k$%uXGr5nCv8tAmki6M7c5Ln5}Lzh-L#~M4*)~)YbJ%slk1h}gNGE@
zi5;mXZ!gx+#3OFzl%2!V){qQ+knpP*!4Y<@EA$PLUg0r-6=C(Y@Y2P$q0v!7B%)Rn
z$8kw^irH@gzu88NOgv14NTAfBk%*@gbfS9@^|ST5j?GqiwDb?V-Lstre%B<k-<H;!
z%km4gvo=C+zWrUmk+3v#O;xms=+6fYj4aN+rn2>Ea+g(~k?GscuhfSH{UZ6KAA$if
zp?aT(lXPSG9gd{JORhmyIKT!OTsO&Bw71A;i8pj!%|9jExMj^OgMIvCuC7!BW@ou<
z$a(F#_xS&gZg7K`lUZf(U}?W=Hg9pGNXS1>i<6{_6|}n2_<OOebM>u%Xc7P-je51#
z%uUgyAz^lrv!<IVtL66>OOoJ{Au<@cTrX+1uH0hlAF`p*@@}3anVo8^BWF(W0@mw{
zFP$1l!Ez+`+l>ELN(L54FSkF3Hj=<vupW>rHCk`3KhB(<o;L>GC=GK8Bo%cOESh~^
zkwgo2r`n`|2B&o@VFuwM`9;X0Bq8LUZK@;`WwOL1@p!p}AUays1EmTP0AfVzgnB*;
z#~)^99sZp1+nju)<<1p$4x``QO@fXuwhzqAa4bH2u-r*>x}U0>I=BGg%%Hcrdaa|O
zhlS#{<BEGBfY)++Gx^wyv}m-rz+Lbhy!gdQ-z2}*OpSWCzPXuPFrua8o8TA}2in^+
zAd8GpI;{R2T}m#7aGuS?2h)U0LWH0iMRV)nW33AS2*eLzV?F+9^$wMjN5VA)Y!fy4
z$;t@S*_8}@9-)2=i*H~W#Q|y$9yMg&4)e}HH`tLPXRU(Hko+u7{5rh5+AR7+RR|{e
zdYcPSsj$6ubz@}UvSKg+hB|QsQ?{Dy?0E%R4&#}*y@k!*58^k~_8YlnqRuh%^FT``
zR(r>#-M*0<>qdbjgfuJ@&vi!HV5d1Vf~F%fn3gpqym2lvwJ857BMoWEN6EF%{W=8K
zN*G}nH_&_?CRPD1o)K{%$=Q^5DfypVkaxt$!wmva8imAJ?v2;3Tq5113|(snS!f*|
zE<r4a7X^p;pGvz68gtIue4OaHL!>;&CB}N7g8aSQyJ@LhNo?}<%~%v{^g0|#%&OpY
zI{J;T=u_#+BI2^jfOeK*zxfd@oZ#1P5<CE0pp^#fkjnJ9bO+Uznc)U?+`)yIl#NUe
zoz}X%kd0mg`|^74gmy}I%g3}Tv<BDcO$q<lxzBsHR4&~{>E{OwKTayB=K9CTnQ?E*
zna3W><(%MV{iMXEOawm^bW*b%SV{kh1h;k-y~7eHX%0kHh|$8w6IxPV=R#vxA^Eo2
zG>B<*&zc-3HUBqT3i2*&44K)Kt`K)}O@m~!W+p+C0EZsxg~13fgEp9^NcbrMMj^u8
z8?9N_dOJyA-#F9k8%TJFq^MX)*6hep8BNc7#u_$;zM9&^PPkbsFoN|YmFa!yel}dp
zjn+Xg3nx{mx49H<r7P}2RH2B4-?`Aolp~T;|3k#BMGO}ht-><Pru-ZDNljv4pg+&p
z?24$d+8<KaOeE@)LHbUJ=Y$6)*O!aQ6QG869EvXdSqw|=fof77wSr=uu#c&Lj*Ot+
zSq5}?Rv|;gjvwDoTfZ~B@7ns5N}ci(sgkUE9<tiC8*+(W(dbo%r&Iq)*QZ1LGVkAE
z{8t&4C)mYtD<^;!xuXM3B~yN4Ba<2UoMInq=Z{>asY(1ARWfRVzyOLqX}H1FN#3@G
zx1ncW>!b$S$T`cQ_(1y0C=p3Ci9|;)8Lbh3q^0qmoBrcI^)vk8*9;-wdlvamM3Bj_
zr?wAH*cwsMn2ePQd)NBqA!P=@$<rHBrU!$&Oa7CkpFPP0YAzc`sl%hxs4+B?weJBG
z==tHhgiu0kf!S#8G1@R2N><DqRJgl1by5-%g~{UPcBu*<gtzoNWu?|9r>@(RsGXs3
zGqdj%BG60pkgepywwzmiE2Fg0mVSFGt3Z5N!XGK(+B~Vu?m_u*2@q}4$*qmL{)%a(
z;kPoP*FB^wkMoRByINJ_HXOh3VNC)<3YaboLy==^&PngkiKEwiDR_b`;f=nW)`&K7
zMsT5C#l<T)Yhh>qu(lBIyJYl%Il+{PsZ3Jwk&Bk}T0^)L`o%0V#L!c+LaPW@&FQ)J
zy->h8-Hl^l&#@hKZG$r^Nl@bVH2ELCS>!hHz4PnAco_j(hL-yz(IZ_uxN&GaFcm-u
zh~rOYX4z_UyfcuzrG<})2^9Eu9~THmMD!^Nll#eoxI!*$Ys`hEvVr9^nSPaflJ?l-
z?TIVbwb1CmYo7iqz_khDilEn8{2330(5600(i;C>C*S3;k3;;T=iBOB!MqI*FL%6?
zso#id5IsHeeTN9{%3ALpU-h^~IUPniQ*lk@Fj|%FcFL^r+tJ7j?g**&A=K0M&YNF1
z(xuFb?4v()Z_o({c!x(u)mby5HxT-San)JVN!&!x94peyDuHzVA+^3st@i{zZEy0L
zZdvPGck!T`{vkK`yVAe4(7X0mE!M~JnCUf%4%z$2ia^-RzSSCvC=tPkoA~!n0^JUg
z8NaSu_)2%2q`P-^+`kz;BLN>+3is<5u~4MR2spMtBqeO$W9SH*FjnkZxq^DEBh}GI
zK})+ypC$}VG(-Oay`+`V8R=B>^1$YKb6I~b_`baPapQS5nfNjFk_In}7ONGJldmhW
z)gx9xMSI(f*HIiXnb#~J9)OYtjyapu(`0|=b)S`S*C+UEa5Nylp^6NX&N47CXQb-m
zeH)bxj|>|+)%}xruOcBQVA(Ka5-~Sw!Zt;7_m@nw=vaS-p_$zU#tLaz6kZptEF@k>
z?d;s*(=~}R&v$-pXkj{&O3GR%W@rOP4Gn`esV3_*i`JLcyHJUO=9Xx7>z$oZr=6}%
z(EBbQ=?3j$9rp314BGxqmJfJXHe4t~v{+qs9fk2dizF3{=dO^Ut$!8h|H|y{A$2f2
zeV96lV2^j`IWi5)%e`LgRl|o;veaNLg-?O~^i|#Wl3h-tbZGBbSgz(TzJSm|QBB{X
z7W(9;lZy-$ESfty=3Gq^IvBJJuQ)`Scy=^vUa)hHs#j^1&mOXrL^Z&cmx>pUU2d+$
zX0){@$ZW4L!cAprL7fK&D810&nP*msr@z$B_rt<!q7ST!Hmw9`j@?$V8iwKqnrI_b
zVe3_O5M&s~h4>RF?vO%flN9SA)2(CSSmemz*ws8B+I)^V!7OaAxPegtm0=*MNvAY}
zED3@4d^`+XF-)Qp=}yJ`{R<U8U0%S<#Y2fq(>QBKiI?NV_aFcnTg0ZXkxpe@4XD?_
zmGg*_bPOd}6Q2<QU0N}mS&P)<*s6MzI;^vs%oX<wv93OR1&cPBvg&9}P5tnG9dTyv
z7ic-4*~c|Bz2T$qGDnp*rrq0U=P9yOEmp#W&!EALbNoKZmT3zRgcR#qS1;RYoceWd
z-H0`Me(PXhw6=L;*sT|uM8&C98cP2hPjdM42Gzuik#H0tdAL=LTXhzKu80{EjxWsm
z-D?A+Sx2z6s>8^U21!9WRU#>lDyDNbpdr0vktfTQPZZ^ItQFuzQH&TA&wuxE9<TGU
zE^n2S@!xsw2Rr;%8}ypzHPVzREHUb=au{zeYXxXiwkJ*|4nYnpP5oxOYDMNWG{QIn
zheI2cBlmT;&f@y!%_rPy9lVD2zZDb1Wmv-zl^3k6HmN)vKFIZEyHqGQP;d(xt<Rk!
zU}9n(oi;`n(Txh(Hjp{jO2QacV9I=7rdZ47=av{Ixw|`q92y6npT)ly>xzHxcs>z!
ze7S5HrV<)G-<V$u1`)Fc(a_fg`~V2UNZ3hpkPauuBNY@7_IaG-Fr1SOt0QluPgcCh
zIq9<xFYFtBsPkx96Jk*1khuZ4xy6Q7PQ;UEygKI8N2!5?;1^x=8>xyLsndc;Vzq4S
zT7gb)#58qQCik|#V?*x-D*njdT@Jpj#FQ}uqCG{K&k1rLJaq@-T~1P;A<1}ECwF4d
zJ{NG}k84El-xX6xc*EzA_&ia4HXPiM%V=RctgNo)<<_8AL4^Qq8W$NtX)D~(C~H@b
z#!ps^0~`{j7yvP{tm!BO1O!cAOtzne2{bWV&w3rdyT4$w+0BAsQiJOIsH9m}b#3+F
z7^&o11$7<_Uu()>14<Bmx4C4qfm{?qziNAs6EHsfjpbpTZZ!u9HZtpECZ9#1RqQ|o
zI`kD3X&2^7Gzf$}*J-~&eM^abA1c=R-?@$V&#tYX$u?nUQy-^J_gImDd|OEd(aR+;
zXlosuZJ#LAVn<aMQ8O;f9(Nl<)nncK^G#X@-rVr>5I{TZrU6U?lLdt;5zM@IJlpR-
zO~Ud``})qHM2ej;F<)YEf*y1Dt+FPNx_DJBDi)7OM_0E~PwDG4SV^|oz|zJ!4_X2K
zx?D8h8k3X|^2ykkOWE4P<8~nN*U0MA@(TjcG)x06XIZg81CDMJCo^t56NxIA*jTI(
zDKM%tBr0m~ofg)zA<iK?-TuaS<<eq7&Rh*U6dH#t&0H~C#zJHY%_*Wo7Ut8vyj4Gm
zXLd)~e5;DI$>0{3AjZm#^S!NIeMZoiod~I~M2C81jA7n?y%Pw#%x$UZxoLl2BC&_8
zt|Vn#b}(+BiKNGBR;|5}j>=Q5YH1jSbNaQlUnD$(jI@`X{qs&Oa~?BC+P=H+Z->+#
zA@OFHMBkg)d4tu$#TKOsKq?j#P}ABP2aWM*OK|>VSlxfHfbtZO;Zvnr?2LWNAiXIm
zU5optT=k-)+AlWn3iSnPm!S2BSNd`cbw%lz7g(5{qd%CFL9`qTrzl7+SWUr~(jFn*
zMTqY_jtzn}B)?nH!^$4(iiNKxC9GHV0i`<(B0Dz~g08}XOjqwNiK`G#gs~s$e9qL#
zu;IGC@K5-${|&+e0nX2okv};DkFi!?QWw`tW#f`kcy|*`eemqqR7x!|-9$dMcs9;0
z+tt>qGa=dgdwdZDiTtdQEOqC&+^rQzGHpu*aUa~~rni{ks;(_1`8HNNZKpQAwi2~{
zlS`aj-p>l7dBUB^cR=A~ro|i)w;{2#DOiPRK)@$cS~)wF;)#OU{bVMfLN*VdO6~<<
z$YS9}+oqZ7>tPN%wOCB8wD{Al|IEqi&h0?|8Q7S2!;!@qds+=nF*=DL1CCZ%T#*fA
z69R@rPjZ7%6Czv@N-$-9%5cuhCz&4OmLSqLXyh8(T8dW&d%|qM6<T6i3Pkr}LNa@u
zTkv+05G?=!am>G-gMi1-!$<j-wf!?8@lKKcE4G{Q?X=!)loowy)_1+NW_8LcGrAv6
z@jO(_S6bx0BxfV@RHYS4BtnA$Y%uA24Xc~VCCc~=Dj+6>m~-AmVmqWtJaRIouP3RE
zjoSHm=i`av5wwagGb^WblFjThZ45g(xjt4zVX#@<Jj~!NK`)*&Kx6z341nb$W-Fgj
zKE$Jh=~}te5~+<$<-0*jl$g6VTBU~fO0aA=T3g_L$%m`F3VzrK5m$0o?_~On9kD7@
zsc^>R#0R_7;bq|WAN-4yhgO22LlCt)(kb4uNN(~bj3LJK{%N#O@%#9x!iF>s6FkeC
z6_Yw`JRG+FPe<lp*gjTMrco3HKt5Bi&^5ekh_(NAs`+*L?z-JS|9C{;=EQ+VLPQYT
z3T5nV>sYzy?b)HRK`b>QiX<eM7-gIB@W#>U;sAB7Ub?YPg@=SkFfcB_Gv8!31?%4a
z<HzyiA6Ys1(m<6$T$2cY%bne*P0*4X0FPJ^Hu`;W$;!+>eR-ME<NBw4B@PRKe@K1^
zhB&hGx)ZejrB}@ZCBA3HwGICbNki!?^z4ffVV$5})-Nad9`<d4n@GN4L9&7_^~&ZT
zUg%mx|FY$wKKm^UiPCOV%p*B`yBGvbOW3=4&|-DRz5pUPGV)ad!fg4f2C%xj%Pdqs
z3xNymYXX8Uq;bbTvKhWyyuy0rafim+nN`1ttY$L##F5(C1|d~_q_Jb;57<5$lsn(J
z49otC2_>G9u~A^ktGXU6{V6Y3ED@GAV6SO*&VX8kfMxNBm4-)0H7lm(mgai&x}7qM
zFd2Uc3Mix?!KIcmbc-izDf2UB8tItY+Gg($U|e9zMj&<%woz$Ja0xbaU~&eA>o-=@
zRK5w*qlTgsm{9mg%!O)1Kku!8Eeg7X)Uuga7(S~K*r6UZcv$lbW|faZzbLCaPB~*M
zn>O9=GtRs}p^dI#D$!z{IQ{_|E<<ZgAFr7t_%abU0HUFn^+Wp~kWaCj$TUj5r#@>v
zr9MFghSamFr*XStzM%?y#8Ufk`PdS8AuzH`J1WEx=n~7E4@$*DhNn$Z3>hyZrRkK`
z&<*g}D&!JC3a%vCD8hDJzu(5`WP~Y%oP{WsfN5vKc9J4Y6?nzOf|Gz>zj|h7Wywe3
z_ZF3|p5pa8_t`))Gb>&EY@#j8r0bg1zWOIhnI_@&FsjwnMI?j>x;)&MF(~9WsY_e(
zB8d?Ni~er03YyhJ&Fl0+io)rGk4h^t;NY%74~L=ykwq38=a=VcoK+$a<zcaC_W%sp
zyw+RME4g@&GJ;$B&)AxO>dX>%;be4Wlku%-@g`~FWGYZ$%FhTk0*iBQWT%vU*rZKi
zBUeoMelDMJIMH9p@ODWv*0_M?vs(r6`&%D5G!(hM{+#j4)V0F7xJaIJbrSLrNk!~C
zoM`n|psh(`Rs`z9RBK$c$Rq)~LPiQ4=jP}8TcS!zHtii9*ZlK4J6^jRg%pNOzMi8~
z_olj^ip6>*6PX2uB8uEIO@NeN1sWfEV6+EL8^dNj&addTZ@#7<A-tdR%#Npn+zcV#
zdTSdKFAgzDPpNvgtlTu>Nqp41)Jm0$Urddi91xQ-g>{xjf{p6uwoj=yBG6#Ve<uIs
zormX-zU27ONDFNtYq36&)<0fi2dd9bDgPIjDv`tw@-<10CzQ5p50@iQP_d10dw5V*
zQ{!q;FV(`oTnNdTo!xg6nX=%=Y;S?H1=g7i(bSD-PZH48i@VW{0dkB{&p;@l;w}1L
zr|hy7o1>t-Yi}p@wY5)O>`fA)Z9`38o7wazEg9|v=L*?EU@xK1XWDRZuuog3@%Zj}
zIvb=U%YhGq3LMoVyP``acz5mLDajJk8^!Q38M_4#b2ZHCu1hCwun94dH*p1mZCZ!Y
zmeGU3mZ3SwAOdYepIkL{fU^SHG#$AA1j1%}r7Mb(FU+1`k`kisg)~!210JI<l+j_!
z*GOxVu~DDsCV#MdKLfhC<-3X&Mv*?iO6E`ps?^lgjm`KZ2CQm+^H%+=TcB4l>GGNV
zkpr>(^H9Xw$c7|n*cWbe8Zh&@axJ<CTrf<BotwW!=Ih73YZRT9H{kU!Q}2Be&so9w
zQ$A9de}ql`R4G;p5-E^9B4Dz_;4K#gwoOAeAeNCMS0b@4;<IIHISFlg(L7mA<En{^
zEm^At7bLn^SfjAZX?##rGYik5i$8%%xwxkJqL42@6mi0ror8SoNXnKghya3TMoP1p
zl+qbF`dCq0udk2BN0-9F5Dr3c#Pdpv>4?meU%9W*rYmN_jO6#@P*M0d2dDav^pVIg
zO8uMcF*fj-NKs4DCoPWj8X)2Sq{$)v%;4$pQJwV@$n~0qx-Bs0wcqpb6jtUtFiY&$
zQ_8lKjuRs>!)`*1{Gy6I6t!w>4-OoMtsc9aPmx-d9%M4C_f50a3nM0ssZK7<!xoQ=
z-vvicGDxQuq%$0o3;T-YHr7vmFt4mpoTMhxJ0|j7Y)t0SYEo8Jp0ozoG(@sc@=6$$
zE!((U?7YeI6W;0$$I7dy(`9yJt`Wb~4;hQdEdhAZ)?J3tTp5dIP9^EZ(}V5e)i5FZ
zLA1qAmA?lo+i<4MruE)p+C^r?7N53M7FqY0k-Qs}(YT~pe8Aj3ePQN=R@pfr9O7eG
z3=tc7$0_Ce_54k}xJ32;X4KN3I+kl};v7@4&-ySjGq<Ush$lObYX+}(h<MUbc<{ZO
zXG74MP?J@F<sIwNPw&si%S6HnLjEVBSGLI_`o4Ux%%hefiHLc7yS!9C$zhaY_JKJm
z5O8IUk)f03o`#X_Q#HCZjGl{`Ztwfx%a?<A(10+*U~gYv{*F9X*gZQ@9Ici3G{lPa
z-3<B%pER_mPklbZ8wQ}nGY%y+FFJU=pai~T9!M9QHQpO%RRm-?8{+&1O<X#yME8Yb
zECQ<J^d@~f#M!=RZX+A5Dt_Cod#?W{(Z>rN0>|+786M8ifP__+IcnlRUrCvz(#09;
zbtl!yV_0SNvD5k1<ZYLn$w)qN^5<LG>$?20Zh9m=Hc=F-hojSv8XQzohf&ibltoQl
zkLTQ9gm<p1zFzuJvQ8^(arx!(BMSS9kmQUXA5R?BTPu%E9^=d{Em+w3qE^tyHPq{=
zczDcYey$>rC9~69IA*f5yJ<f2B5Q>vk3_o`t5+ac^)-8@md^2|0S{T|V!?^cXv<Fm
zFB?XWc&Dw-;Z|xYnpKy3m)GA@{w61*m0fC}87eB!itkbLA;jNMC6B02=_N*g=`ixC
z{QV-5m6?{t=4@#W&8?dv=cfuhCsC6K3{f00Zk17#FCnX~0gY7Ye>Ia#oFtau6u&s~
ze_iYD;+j~W4iZ^~qO$VnB>7T*W?D(J%_MB7Z}5XK(|XT<YueDepP$7HX#MggWxu5U
zd3_*&YLn$NHYfHPK*Fq?&!!%hWLD66=+xnx(lX-t1>59OloC4I29(%X`KCt&Kr1wV
zBs8r!Eeg4!pMsoHwj7dpBQYevy+QGg$@cN7{Y78&<=lIvrCGX6zf>=8SWh}GPUfsn
zp4MxZqI;`AZEkO!C9K1W7ZA^F1{U@cr}F!cp1BJO<a2&Dt}|MOvqQ#?1Ym@h3Lagg
z&T>vTqQk0a;DLOj7!Ip?ufBjL2msh+!^d~%%Aut~dD4cjI_#1g*^k=m8OCJ}=w*NV
zw$7()z-hdA+2MO_tx7i2l-(OkOCr!ur?gGxvAX7l2)Gz$QXqfTiMUOQaZ1>6mHWv`
zi$Z&I|Md*Eo!v))9AG?pxLb@1w_IA$xO&YhvqHh*iR<x##jo>`BtxgZRHnje_1~_M
zc-;6>wpcbPDSDv*1ZK%tu1DeANy?0BgYACD$Fv7ekxY|8Q%W+z&=^rg+uRGCYV;~3
zx?I6I6UFL3;L(xdv<?h@Z3Ll%C=LX?X>Y;nBx-FKdxW?gB)3rqYa1NR+2)cAUEb6p
zMg=)04Z3b@(MdeaGfqz7w~mgdX%4FM^Plqt5A%*%N@Bp%vHW6ykx%lOxVwj!Y`($h
z<!R6hy}KhVVkw<=_}<nw+IHo;)a)MKJTEKC1@Qq{mA_|-+909qA0Fm#I@$v&#w%D_
zmgyQ)a?eItpXle-U!tJ@%8%$PEWvliO4p07v*J7}K8?5xnF?XS_7AV*UpH*+UyrV5
zilL_~abBSpBcYBkLUlB7S$`Dz^>|J!<Wd>KAye5<Fh4z~W&Bx6F&Ll8jxmWYT#k@#
za9Y^G712&?aIv=5vp8W-9Cx%N$qeVt<p0sqI;20-yQ5=-;76fZnb`Rz>SzJKnD!~a
z(emYx`(i}`!%hA6ijab9aLwHm#VX<G=<(vbuxOd9S%|>RKWouNT5{~|9mWr_LC!G6
z5-r(&JjlpCFEKEkRpuw8Sy;+(j#dSq4!GyJcx3-^{N%`a^V&V&`q~#uqIvY$f!q$c
z%`mlTFDsL!fM3R0xysPo(~)u)T93bGe+CZwEkM*-w<#9~j@XIixB!>9n8+30!1~lH
zW^OrOhF<t*zxfY({(5q3ZuE%l(vZbVOvSe|0o9XE=x~u`zA1PIr@F?Bda8yIodVBQ
zIJM;8S1ukGR^u2KZhYh)qs1k{tFR%0bwsnt5OfpKe5ZgB9<@#*WbWtJ6sat?vS%WF
z%8TC_)sNdG_^FdC?W-LjuNZ&-<C5sMV)EjRh}#3Da!LsAQd2;5K)aGH%BAo_8s@tq
z5R-czw50FyNVMUAp-e4z9g|a3Y5gy+vhuY0peBbP1SRC@iMCHNo1tDGv6P88iKj~k
z2O4f2ii5%qry5WG&4)O<a&Q>aT|0s_9E@wD()eem{l5AcCs*u&r2V|AePE(D73)+N
z1r6rPf*c+)7}b|I(P0mldR+GV3_m+jxDpemZydUDVGGXrFAVoCSAwQndo6ds^1>!;
zDyYXd7UyW(KJ{(M_qDy+<^s_f)6boH-`mOE^Tk>4sKd8H=^&y~`{^mVzxgd(&2pAM
zGnGK`XrcFQv>E_jUfwa|q?(3hQx})lj&*sOKc<%s^Cv`$rC1AI;k8(&ViTr=s)l;D
z$p}*aEa|t6EvJipTUjtM{UYg{S@dU3Ls><Yhnrgg#xM#iL?sAN&cphd6s{UhGZ~yI
zg0U*R%UvNw7Rvp}v~ZX}O|-Y!)vDQd7h9s$H^bq5L3sZVAl1ibIe^53Ugk$B>NE%~
zYb1d@81nkaihPN=mo@ooq$&zMuouYhY^fOov&a}n;`3=uoO+)uFC$5^hNqB~{C;+r
zvUo0#@pFssToXZ(+OKlgwPiGQ2#BcIM|paNJEd3LMzG1QX={_JCE*p(w0dQ#Ln10l
zWKlxl4=p8yk4%nn<<GpbeerXyn@xNe+5{eaMJPT9chznyjDF6w4D|e6fNN5OD>S#f
zXciclI3)~A0OK(AWHM0N2YjOYn>U4IdIHHW0qj9G>*efvjpsHu6D76|rtK(10&=UV
zMC3{~xD9Cx!V-|S6@hUzf5r9xgnkie9kP_+nE|2!Nz(mah)dc7g#NovqZ_4my0)89
z5uWryQVhEb!VbL<Xkk%kV7=X`Ztr--GWe&IHl>vagebch$=m5t7z$h@s2GQBXo3g+
z+8;Jvsa<(S0fB7ESZJ+s)q-RUd|jO@U<0Ay0BNyB<vOROhm35S?kkgthjTR805-i+
z+*~&@dV&omWSRLwA>tk3@m0E|QF>$jJpRKEPanPmM*D1@C7aZ20X5N!uU?ywKeVHn
z#`Q@}%E1<L4sW(6#9`b9B)g_<IJ_$<0Qg*S@0cEjC8M^`7{ZEiw}%8hArggMN&)Um
zUokm)>)e~DN-j)MDDE$2nn_{%U!FQr9+NM*qCexPrEGD%Dse908OeZJr9=s;{TIs;
zzB&27p7FqtB4u4M0j%HG;+u&<7YW8(*Z^%bYOHKrUBaxM1*FvLtb9i}_xn()V^xCP
z*jFNgM>(Qi<Y8EuKhs({nZ)`Ug?&G>s^MM4K#<SJl)g)>Kbn8MM9jxeoNi{^ZLe4x
z&iQJCFw4p^vpD|PUukH_nEaH3+4(&_8`0D=#9%<GLMOUtc{^I-mdF-p%x$oM$#j=I
zlZsLl>%ruW4)}+e^h*7;(TwL2iVBk_$n?N)nci$@^RP8iJOCk|&@%W{$iESV^&!!7
zG7P4AU*@Y#G5yC05vsQZ3Y>EKhpkg2U3ggIM`MiAB4jj;JKbzU+y!QcwO=2O{XKY}
zU!u;QvH-BdQXFDe6Fqmjyp`d^1^G$}Fn+}_1^i0KFi&op>(G<RhIt7xsEF0M5)6-x
zt+(x7Q3vK*a;`7qtn7B}YvxLLhZOgJ%FN6H1ABLN$Ywc=N*7g{8ou`hRChm~<GZHv
z@9bvmVb214LZ%=^<uVmiCb`TUGhQ2iV=sN%3XnLkuG8trB6HdX@6WiPM;bz{sE*QB
z9}nm3e$3C?xIFe~l$J5;8tTw-q+4I*3@gFwqPMx{IW)><HV}&n1CjNt3hkNui&a{Y
zY~Y!lRoDH`kkh=p%vO&(x7g7TSA3w0o%cv@713y=Fnq*1^Izs$TCclrNU%|4Siv^&
z{2ZI1aiSEw*Gpnl71+N^&EHAIP4bO~kSAefvusCipwVK=Bpchg=EujYgyFWHZ&&|e
z=@-4Z<Td!^oXXbd;Cgz(`mordW0#eqzjYLjsm038oI#w$@_t@N*TS`Wq_2-Py$hkc
zhrmWY`+C0-S`T0d6`&4uiH0`6YQt&shOo{0lp5Xkw{kc&uS<t31f%*hX4y|=<~ca+
zUW<8nT<ho>G49_~WRTf|a9)?5^2(3o?Q3@gyHc5bPfK^(#FOg^)(mok*dS`%`q2sg
zAA(M_@=tNZ1*-5=`d8WPhqqxYD0#tgNku_U{;->#@U&{bCpsgaD^k|i&exhk*D*g1
zLi$B1;BRmI*8mvV1x6SyguzfD&uhm*{Q2zd55e~~KOWKs^NM@vQd3i>Z29y2?&&^7
z-QB&KIo@Hg4u|pfaqa!kHPY|(0y$rcer<WX!Iu=w8F<d!|B1N#y;9mxiYjNXuY9wC
zg}-agz(6rOJA2~ALOYszaZQydfR1{5P%7jSi_-6QfO+X>OCNKF#P(mzeqhERKoMEE
zC9em9q02DiZQZa4t3XH6^vsN;tGf&rI}1ajK>bV_s<C%#QZh?Q6MYtd>xIJ6dq->d
z$qAf<MIUhar1ZGeaI^1$;^mlR-saEK<yKDj(3`j2bkILvTYJ8;>4bK8d;28lwNXKw
zm8D;#!8UP<e)0z8=(P`VXuR7CRbqh(2nfvCc#te`J!sge)I!ZnoYCK|T{p0?t4u2t
zNuvBZyCqoqpj$S*|L0KX%QhM<Lz`h_cr)qs^-1loUp!1N&qWOsC00v?xOY(mh}gLW
zH0u@p)Vo`?vQ+J7B>4uvq^noW=BS~c@p~4yhj}=ll5N`bF>W*|L*1f)j1tZ%sh-dd
z3!W@)5ZbXs?!QHu$K7hNe^Oo^@gH5r?$o<gP{%F`hLtYzJ&IS>w90!uT+{EE+^9gW
zswf?4T|+gT>({FnupjlByG9~45VcSKk95phiSgZ>uHn<p>bB8S><ji0_QG9FR5$_w
zL7{GutgE?A7VHm3E3OG1)g&yj?$dsj?#|B6ZM@MF%CUCr)06+;<SSN^^_(l2U`fhE
zGJ|fE|33M&K2$BLQm8(tIyV98#YGX0Os37_64*NHN=yf7K7tK)1p~H$x#1Fz#S$j2
zt{DRZ3V8w7bf+&{Z+?by)M|ssTWbTVhm6O^Zi)7FYGFxU{xt#gB@}^qeqJ4kFaaPg
zxHyf;okQ!S2_A|AiCaHzbUY6|eTq{p3$4X&?NB7|x^?I&Ar;v{eelk-;)P<3-TD?!
zo}{<TdL6gC#I!%$ytu0WApwTeQee-o;;3j4Mm&6F>zVdh3-<~q#_eT<L67VmYnz(~
z@}GNe%BK&Nm6gX`v#X9ju!dud>?G;*TT>>(5KArjB?6uhBf0~26v{RM(j<Ena){_%
zUIpE9CXY3PR8HNn-u!@FX;{eWT~&MM(h|Ve*VpfCg)#4La!4zN5Z^4BwcM}5piqb1
z!Y|hP5zcvqcDVCC{?h)@`(#y6xxMng)wfsA%%^}VmY*uoH$W&MUKd?79*D>`hSPr?
z{gn!Uh6PAKqP0+Re&Uv<uhr`Q65m?xkD;k!B^Np%q?c!lCN6B%VI7IB_Csvl?GL=?
zbnf;qWoC+!m61wkA!LK-cJvVG=EGKPC+(|f<1IjY#U(OiN(xG>sqRe7WXfa8yh87a
z7f;n$(kjw1>vJg61Q|l%`xEnb^aD2?i;L-I81v`1aAn;Ki(;jbH1Kp71ndl1!oe&G
z6$Z2o!JL05iSx>u+m)*8x(Rmk*8X@(6!9KSl#{>+boLdL6Ae9a2z&E^c0`Jj+BQ3F
zBdw6o8zNe{<r=YkY3&Gy+4VH(*B%7btJ;adi{bF&kXpbschxoSu5>$fZntiHj5(wq
zgj*h$L2%1&X53PlJsR%=ZvNYaLA`(mvhzB;It|A4yAyn-`p(rk`Yax;M!F#WBt^U%
zu_v>x=bLu$hU`i<R~rz+80QMA+9Q|utnOV2Nd0_FdF`>NrT8wQo9cKaNv}ADFuO@-
zYxFa@--Fuo8z#yE^G8xMs$6qY)}J65q#LQ?cQddl@7TtKfNCWeg3-oEew(zJX^hJ&
zIa3`lY&IsQ(4byAKMmT~P{lpu*&bvNP!a$0t7B2J9NVXqhZ9O4*7cw<`!M`d9QN<P
z=l)|grdG(67%x1jC{z2}*6`856xVs?5l>#%i)iv=PoMwg{$@G&%nj*r3HT?mt#N$M
zxqfV;Zeq*GiHk5y-J%lHZ~tpMxR_z?=M7L(GgY$OJO;`#&C2j<)I2ht<j_M`o|N}W
zGBj2Rp>zY5V;I_MRmkXlt7&WuRe5V^?M#FcU6$|jy!)~jIB?X3Fscu`0NtIwcSBB8
zrSfxh!^yfa{VW}$Uh&7)J&I1H0b1J7E}abn5i914P)25{>zq1?dtn``#_E}2puJ&^
zN8hx@_D!HawcfIbiz~gy3G1bn5r14mvEs-D_mSnBk5DIHwi=6NcaBkaE;WTxqNw$E
zk9ntVj9*cvohD*!=>!DFpNFN%m_|HqMXnak-wadv$1N|5(P;6M;9V|$*eH+S(3;Rd
zHWL%(SVsn}YY58`vSEEBUdQ^e%6#CSm76Oki}67ng50a49)tAK_;THDV&_i@1gK)b
z9&XrvcImZ*p2nZh#zI+u&f;2I(}(XSr^ffM%hi-$WjAO95$g08nN$6=BnTcIVuS5x
z4_CpXLZ~%m-vq@q1&$No+0vX(A~7rJ|Lnp4+qU@4Xdc-}rm17)5@IT1EG0}P-Sn{Y
z@T&i1_stXB&e1+a$pHI{&9S=%Pzy-mw}``Y<;xx?l$2z(^0qdVjJk8hq$cewn*So1
zU?IdFH_rd`4Fxxc-4wGjs#MZx3%;K#2MD<zD3?<ZBUoWQIV4I|M-D`2Ww2g6mbKrs
zzalNpAY7K}?}`^=9M3Gw63`QkGRORMcEnAd7t>mZ!Dkr^uKgt25=DxB*o<kI6$n7W
zRtu`k0NIgVqNaVNciFa*62E`p4(9#S7HaU+U1l|!n~%WVG1RdF0O7wgUslLQ23=3`
zF17fF?lde$$!vs?R}WdI4R<zB1z92`83A0Vye~GBL8B?x*G-o@IODSp1Q~}};yqqf
ze_6CD`i^S<VEOuok-P(v;^%zlYaI6T>=?<JmELZ;fm8wC4Ze!KZUV5MtIKQoFV8RG
zY8-8ER?NajSVtu6%;lGICVnLf`^Q%D!*g+5P-CGmd=|?^r6k;Q6JbyQ!vG7I7|4m*
z*x@qhky<gP6_9FPI*mZ0EmR-Rq#Y9Hv}b9`6>6<l3hUj-fBb7ljZ~z$5_6d~PgIat
z7i=^31sDY$daPoh-|+MD2Zz>YCbbcjf=!_`wC#D}!@oQ!HHCCxZoGK9jPU?`ZfyUE
z)l?oYG$;_3-~`nuS1xX|ur2DoUik2{O`uEKE--bvrICeIK}ueSIf{)jO&b>7Tly(S
z&=Hr|bSEQ$2G+-f{uDsnx_f)h&WnlH{($p>N#%QRXRlkAuo9+z*yQ>Hb@zK|NT*Nu
z8`>mF`5fciPRs2FDC-$~nxT1_d&WkqRT3yd0b-%EUrUs-9%Me=tsk3G=odh-u@m%%
z%SQpIE?$qRo{zX|X<CG(YEmL&7e0Y^<3F6;pp(qyY#;Ndy0s@V=ONRa;85E8$AFJe
zJOkWx<dh1N_}5I=mGypL!kI6}q1H)dZx&#mz54`Pj>Y5V>o7Z2G4$Iy@)Yxqj<82c
zusfDYOiCVXv7bx{z@`NQxgYQiaG980iLa1qQE0NJOQM}r+Sj(_ouj@S<)7_Go_2s9
z^NxQ>Owk$zrRiRb8@Gk5kKFUcca)c4AD({KkfsZQ-VU>6;6c%1YGJsMS`jV?B{Up>
z=va#|s1EVqZhr+kc9T+ZhAR~txFu+;JM<sp@c%ac|D#pf7a`HK_5DOz*4@oqUV|L%
zFS55gP#m86cU_(%$hY4!+BbZ9cLTvEu&l75RLn=bG1Xe92Qz0nBTq^(rlb`n;y`KB
z`Vxma=(XH~RjpEk3G4OVqI?z*vCB!bd&iVPq<{+a^;P*HqT=zRQP;?*H#5wqYItQ7
zTf=}R=GDvOI6+w-t;A^PbN4mZ??`A=^a;80qA>brMxmlo7|GFZM8Y6#;V@GnYbrF8
zR!pSO2v?GPpfkxtZv4xcuEVl%6@6_IEblk3fR8{!rg_p|sA!>tI@$v&4D%6mnit$_
z@%an&$_za`GnQgs<^h7|wCD|il!ApQ*;`$&YiAGbndY1B`~u<Q%Afsshg`f9nd2KY
zPMmUt;#8f%8RF*)@+%b=RJpzSp(Gi>Wpks#(+3o?+ZD5ifUo?e#JhV9B1cu<@CfOd
zn8}B=$d$*2Yier3X1J?75(oBeCe0z)SXr5wCD6*`%ZJvxKAh&O{T73lH7gb%n81C#
z%kVDAj?}w=ifbk;4HH!bOdIEkE=(h(SpS<!+eF4yU`O7Zf-D}Iq9a9t4(*)VtAG@4
zn)WH-g^3%}9#IR-)Av1Q9~H7l%ioW62LC&-LjTm+*Mm6!I886{x8JhP??%mS=h{PY
zW8UFl1pmcety{XA$l=?u%JjRpF^4&mu7*g{9&4m_<-ZT<rd@Qt6<MN$_V2NHi5H&^
zd7-yEN0ui+qhp(pqf=s@X&5=)3Gvq<8FGm=cMe~^g11Ez2ZICa#)X~|_*avDB`<}@
z1yb2qTgTG54Lc2qWu665$_-UMzlGaqyz3HuuNaz$($eh|b=MD8HYJ=9;_g#4@A0Wg
z5w!!PwjNE*u==1k=8CBa=O>*cgOxWUAc`Z(9d$Has-NE?37ILhS?onS{Uv(e%^L5$
zRYG8=ddpB$33^V?uy(MoyXi!2nxf(Tg&YZ9D7J)2I3D4YS@x};8uvOeb;3F?Nc~qC
zzee3{!n&gzcqG4xRMw@%;3>3sj~u_l&HLQu{lnAabu&c2pru!+GNj)p&T>#LhV7hO
zB{n->eH3`_PynY!d%<3J1wcsVH%f;lh;2<E9?-OHy`I4khgsQcf1=8+j9C49QczKn
z6aw)XtS$J6ZUavU`6EpB*A4;2U}n=c_Zko1pXHrf({fk0;{UBONX|(5$Oen|>+Whz
z(Wj8XFs3ELOd6$e_OHbSf4vZ(u-WOReu|4s3O`7ngp~9XN|>y74FJA}r@1_WS(C^L
z>C+OTiq0AlqQa!vq7yLFn3#3M-7!v)Puu@ukBo@@c<(u9v}iR~i?tB(6i+1aFd*XQ
zgJ`*H6^+R~qm;(#bl-Ksp67$L!IJyuNosp^Pfwm)TzlroO7ZKN<2A<U?A|ZzWFubo
zn4|3vbJmIdG<%HLA_9Iyc5iI5H|k!}?%8#~B>D+rrQLA%esOIUkZ2KhC*HTI!ib|0
zwlrPphao}MmLY$PUhsRjdn-vb`=oz^uV%<{`8Q+ojcG`Ib5|a?=y`nC*+HR#z!?~<
zVqgXx%E>)rl0y5`=CAFrB7-)@%PFB|=8%$AV#7ql%%Vlun{8?OYn1rcrRcOva#j4>
z6q68qY`e&%rblc$L5XmNH^#WAgg(V~l%&^xrGXDw`~r_TZ6O!@KdYs;W|gb`ugUh~
zBn$qgnMISG=751J0NJP`2|I}gb*z0>!Gt)90QCT>&pZD)Y$4izR~&tT8Z)J$MK?;R
zFXQhL5Blp>9pzjIUjjg5gW!V5>s<ZA7hh!-%=lGFtuKX8Mg=@Z`1xERTFH$LeS@<*
z+%SR`Vtmdz_&z$8c(`0{tayt|EK(-vFjpPG3{K4oG$EFqLM(U^&$*)SvevcdOO%37
z+RVwM(t_wqc$v`E7Qqzpr@fV*2^T0Ov3yLs-4K%0poX~fJ=7c^yadO{4c(yyy@khi
zT(G~>!6yiDUmV3sO83_L3|+BP<R`5ZB0rW;5A;#;$tiO`%uoOdkS-UkwA+{ZekW?p
z2IOs=D~_r^79hAiSi*m_X{BHqjQa_RluTVX@qHdiq89oet4~6I&xxCt9yR?M30NmL
z&{l7fMp7@SDDcC;DET;QXT#y1Xc=#~uGiPs;4Ao%t5|xFF`VH!88Kq<Wz*CzQ{WG8
z9L!V{CmritZ&;%#3pRP4dd<={$xa`S5FG+8xvQ?;u`-X*g!g5`kF>>e;mh?G0O)@W
z2Tynek!0*(=3iC1aJ0Deb%>vRqg0J>21lx#nw)<fuFS*DgKkAd(AA41aYL<(AXa{u
z@fT#fnlPMFX*9fhvpZakq>~Pikd?R2e^J1qDgRW-<?5#qX77V8hK!efanHa<frSNo
z`d*#=ctJ0|iSfGoKZ#ie;)6Z0YP9d->Zl@R6y8Lt7L78PT($ld*8j@Iq9@Bjm@M3X
zVwrk8P5I-;?EL()X`o}GN8<>Wif*p@<4m;yOzT9GJx$7s?21fb$i!n&=MSu<PTJLx
zAj6AHlQ)fx5Ty-%v84s8s*Kw+q+FdzB_eop9dT7U8c{YWj4za!Ui`Pzb_|J4(Ae|e
z$gj2}Rq7tOH=BP1#dA-j){wCx5@0MWGVA82ZT*>mU3;fE9N@MT+6w$S8F-*s^s+88
zRpO;Pv8lx;8uP_D+RP{bzF5n$=1FJ2VH^$1Mylqrmw8UVx&K+8dc!_M3n98&7%@^*
zwPLe?tILY==pzZ>qh_T_G|b%gTj%g3UmY)|vf2m_w~#^=v=~>m@>u_QPb6dSMiDDQ
z>@39ctBOI!UIm$=V7ao|pUhn8zL1?0fi{N>)8^m9QvhT}Yljsk%Il~N@S0at$4$+1
z$C?N@Tj~~%daD3DlK6_%UZN0rK00k+H=pRT+RAhh@sI%43w4ru5XZRQP_^PBRMEq<
zI=RI!&xFc@QlzXN%NxQ({#oO{kD&f@qUPCp<1rZ9owMzc0gMB>xT^kM_fyX2SO6UM
z-8^SIAij>|&4QhuUB?LZufEg^_n3aw^>=Z^tZ+&w9PslSia1H`4G91+TUg@U!rbQd
zh{#f#Yb?27^{Wk$U*LzpzMC)u6FeuV4Ky&s$*x)5pN3&!N=R=eTl<YWl&M=PW9m|h
zw8%Etv@OIlVI-YD`&a-Gl{nukc`BHumHCYH42|~t6ZuC6$y*krQ=vJMU4)y>Al#$A
z5mtFc_&9RDh<R58l4Le1dEJK81`;!8j$w5Gluif@5j8KI)~(~(ngQj7l+lmG`{F9G
z{QXYz^Aj~sTRp_>8fhi+8c`7`A@UQ(#}2g6a$=#Y>GyZm#l34MDq$+PD=5az9gH-*
zMXGS-%ZHk=f<&B9%Uxz6=L1lGLjKleMP06#`=`5kp)b>{wFy#Q4wc7$`xe>i_0XS8
z5__1MdU&i!-D;%@b!aCUYqINk&BqVuVCD|q$#PY48eJuct8^)A=%?uU+h4-`#q4{!
ze<jZ5AtJI{tOZcuhHN+KT8!r(iG5hxg#dZ!#ppCb73yr6OUCtREiw!DK6g{j@Si3_
z351t1v9RNBhr4%782&@u{|C-|XqzdR^_01&%e_nIboMu+TqP~2Q)dI4w)XM}+bIfr
z3N~Ma&I<h%NsSsHmGSsa%;SXwW^E-RA@`X>E0ID!!((e~Z1OwS6u&(~47l7pt^3sw
zGH^>|zF44+H5sN@BG)oatBrh+uAmr>!SNl!C#Q%{9MYM(jh@?B=~!9Y*f?b9n{QT@
zXX)ZXbO^qEr|>3>m4TX%=GIigQ_u6aF})vJb)x9oL%HqytbT!?8{E}l$~H(Gnp{;~
ztUgGHn6jvS5jQ-eYpt|glrld&Xf0^WAvP=>O2KYfF2W)wftOs9wt^K}ObUWOnw~%B
zJ6Q(6YrHl6o`$hf7!Kw>2U1R(xl+6=uq{1zi#@Igh+nL=Vdk}a5}AnVg&V$GA*-R1
z89a~H<;eAqdnG8ctNrc{*@robd4Kg<qrd%R3ixyUqhk6ig9qemi+tdmY$-;hl!!dL
z^rTgFV<I3X{7#mMxh%6%#pz}T_2~Fxp*f^Z$0|rYoNwiDVukvAkdb2!4X^8vUQKC0
zqd8!wHvXH}Wg?9)=+^$!ulyS(Iz|iz3VuDtX-SD3hKe!Yo&7O*u%jhuvL?uhin%T_
zNeh)~OB(Y3pq@zKLL?}9Sc-D4&_EX^vF_7^2+R5aiuD?<?r(AS0##&kYQqx04G{0x
z|03%w*y3uoZJiJZ?u`X^8h3XI?(VLQyGwxJ4#6#WfZ)O1-CY|exI5hb_TJ~Y`=0v~
zR;_xf=A2`Uw*;cf6qIN*S;K(4E9e+_gvr9>l&$!7i~-)E=N)$kFI%^ZxKy9EswF9u
zFaShMNHQX2YQ}b^$UMZ#(X=$|MJiZr!rn2Fk*-gKoSX~G4yUt69#;3C`P*GT1I6U$
zL}=J;GbVeg!$jB-_FBYz=wLo3zXcBQX6&W%AZV+n4Q((va`Bt(j!hBUxch(a-k@3t
zx?Q#vA%<)5Oi7Th!-mGFF7Wm76atAXp&#ssxZmT*I(7(b@ZMEzn!KLM();asRp6Kf
zJI6oJn-L=wMetuMW*2xoqpY5Hgq@wRow(Jx!DW4E>n(yk4snqiUu?Qq^6Gf}TXor>
zX;6fQ6VywTfO`B**fXYK>fA#+%dpR`5NU=HTdZ^QBiO0OX!hg0O3LFS4AoyuHCc6=
zJu>17gGo<dLA=i&1C!mc=iqvoVuyTrk-PW6>t%(=Ph{oE+788XTxJ8SL-phza9Bm{
z4pAE#YUG`q84v5w@Jh(gT+unnQ!E+a2=czRWnB2n1Zk2_4ECNw#_ON_M*pXal!QM(
zbuYyvg*~ARZyI``Gc5EA$UK{0#m(G%hVK)rt!}t8Y6XIX{4%dzkQ&OgLvn?GUdY68
zsg=W}b8NGH8f+Nl@;ndDy$psZX!+8@p7k2^6O&;I(*q=|IbZWhsu|7Q?LX#I+%KMC
z9+A~0xfrvBLO>A0)phN?uk+9KymlC-K!|@+aSU@%87|<tI$ebHTPmmM6x>`GUBWjN
z?lw)l_nw&Eid^{1IGGJpAV$bEae)mDlaY=4_mtOm$6s6!byMI11-d(mZw=OF?p8-8
zq7IgDs)(04xw9e~V8DGb92m10Aago|l;T1?MLT+r2{&M73?dg+C&TW1fNGDiTWQtx
ztcmL-tHA55ssS>Bn#!UQ>u?mVPi76bK!$I;FoeN4xTe~xOK^nXh5<|d81PhqS>M7w
z?*?xyJrOC#r$2X`!B;+U4WHvIy)81xWhF~BL47*YpwpR^O%mxDv^VUJKe{hqne0OH
z&`VJ@$r4V3&^UELJPIYgRDDc!iT5o2J&W!H1E{@a9*uvrbJCDDS|@%D&4~cIEP36V
zQ0Gcp+g(t~P-2cB34U!m|0h5EpORK^j<hF2rI3WWnENV)skX60qIR>#Qe5od^@063
zg+z49Q2*eq2sNGC1sg0}s)WZu^iU2ErTP3ybbSA3S!;ER+M(mCO_CB}$X?%N0l)hj
zJ#pLpM?Svnsr!Df&-1vZ{QWASp+)8uZoa;cVm3vzw<{F97=^(f%a6;~n*4$_8iQja
z0i#$L)Y_Z2&czzkv7$m>cI+;6(799BRsVe4c)Y|5s9z!{2NP)F8<<k!?VYimGX&m|
zY|)2M;Guj}B<LTT7-zHkb9h?ueXSK47}LuDH6(1Of(f8h{b3Xc*dR_$5vn&Mign#0
zcYJPLd!JTJ*Gr?!DJ5qWtpS*vG$0&HUnc*+=61dW-c=d)%3g4)eT{+j(T%$QVNy2M
zGmW>8bMQDLTZK`^tp=Ol<aa=nYtTF}J1IDR<H~X4k@(GoS_IUcw`uO@22(_#sOK5!
z-YHhz9wsfITr)J&g8?zN6dRP8%zO+sNY&!S-v6TK&?q%C<spNJ)uzF(VsW3m-Vtwd
zG8z}i#fqWwO$E&1j~mU+yZrWfiiqMzn@3+L6bU50TRDy#8N<7GRMb_YOj_jBB=G?f
z9}<`3<wvUx6P*G^p~0S#6KVXfzsoyES__#_=5zHWSp2QIrtfeKJG<ZLDiW!Qfr~0e
z^R%Aofp8_BXRtD=)-3%W`8nN#a!DygVXlA!E<3ZwH@u(DDmN`a4oNl*kqy8>MWP)p
z|5r_~*O<j+I}W^JL43YX+3q8-#G6}}gE@uMR?LmK-Zc8nzM&0q8>Wy#xNs8|SULwv
z{;Q5q8yNpXcG1BJv=gvBTGiHrW8nGSxjB%a|Il2DC&bt1kH<e?iL$AyJlcM<57mM%
zD`0S|C`}vE-qDrP$F_@aS`uU9V^^&K??PWk!gop(T=tS=U%JoV8<A9k*|2M!kGHG?
z*-+S|n58iQ<PDx6F=_X&9nfJMN1yD@nUX6Cnv;~!GAKAs{JldPxZ+eF2S$fzX#U)F
zbL;T;SVwDepQbl^<_sK5E@?G?lQG+}<PK{L@{PpK@;^b^vq>mGq)yAuDwsXWQ|rst
z^>~~WdPX$xdc-RJi;R#kG%z9DfPHy`S@O6$8)wCtY5O~rNalbk%kpyAGCeJef-$Ab
ztEH0jRU<4Pe>;17J3T2Y>s|Nd;db$HdG&nu7}==+w;d4})>Td_)**?2I$^RVa+^Lu
z<SG+l$j9F@=o96b&%aJc8+hN7w3Ix4%J8;j_!7sR>U^+|JR#(tu=&~^AkH!)RY?LW
zM<YWZX%NRTbtQlu6GzBbvuck|s1W&ib1=0@_FrWcr2m2B6pblG#l8kdBMzzbU5!rG
z_`IU%a%qx`ijfDylDCVN+jgwkhrNOtnJ0DYBL{Q732()iT7rTz7ptaACu$q(WQzc*
zG&tPe+Z3JmKJR#&Z+DH$YfC2FLl#vZYZ42w-<Hqc(fM>4I_~B*wY1VCuzqEAJ`fB}
zA}?)jTHbEhwYf+ErDuq^CCX=aW{T+6n(kH4)(wpz?gGuu7YPf?<G@EGCsH*sLN$J#
zRIF#j0m8yTFAos<Oib?$id1yh=ciK<T+P6R44a0LGS!l_oQwiamhs(PfF@%=&&C@{
zgnUE=5x(RBl#nSmfk~}00~OzL*s;tdtDbCzKgxh$>(o@d@=hY@@7&s}fmXR}o{5t(
z^PvZ-rmb#nV;0U#avpW_?Gx!6J_OTVNAueFrm4iEWr&}hN4Bo6fA<Ek?%z|*wYzbu
z`|@`Fe(!s%y9qHy{SZ}-v)rUyKnFgAb>8dBO)flkqZ(TJ`62S1IPT4~t~KkJnV5Vl
zDxgMNHt1KzD=ac#aJk&}QXqaMlZ>Nh<KyZE3B3Vzbs2;MzBi%KCZN#f6plr{3hXFc
zLITk$p45ZJ-tc`Z1yHYdB>FquxMC%<;QjH;lN0-yjZI?*htz8vB1>nJ3@wSmk?5F$
z<e*$-8o<=D6jip@1-OMQ<oV^xbSxwLe{BXG?tm^aefL^d#%qlpWC>MF`pzeMpH;rn
zY<I1;6F@Ygetn9g5ydmtc_PwtkY1U@SM3|U&MTI^#I9vVl`$n+0FU1`Zokc#(97UE
z;@$nTZN+&_<7#(C9=DEkq;g1q?flsl)B9oTi(<f)0K_o1=bYf&Jhr<F4RHyLjVrKf
zSCMC%+7*n)w66MkpHkf~{E?2{W9qoKJUKg^vOT+qsR{@+9mGzUF+LSR&FGccLg!BS
zNGD%&RN8IabdP=*3czKe(L5+q!#BO!lR^|<NavMTZM(M8H##~5`tNbQQ3*YOJ(iYN
zi_j+HB%=wa>~ui?<VVP(`4REIczY{mWWcU&ya&RGXp9igAd14xtovyg_{0sTTc5Zx
zyr}G-l8NEi9aCBj_LZ6Tg{qJ!X9`nb!%bubd^6Z2SBE)hJVJ$q+-IpG)ZpCL*>^fj
zOv-&6s`RX1KV3bAagT8efe8J!l|7r!l@3>pZv$M1_up^4KY6zk`S?7aoSaoRHw+q-
zhbATVKYl5R8^2guRwdBSB3+0yd>yD_<KU4mU>U#fcqc#G(C_7mP=RePM#1+9l6)GI
zjU$6lmCU}J`9XW1^XqteXCt>{Q*Bq*Omfq1JWt6RAVlNUdY@BmKF$V=E_OW8$tfzZ
z$Z-x7grFqVrKt2dWnIR8MTf@v;{7>s`k;E}-%1_qKlez$uH&0g7%y?pm~oOAQ;INY
z{8&zB*L2jqRp%X1k2$p#0d@q=H{4K&qVme%or#}pqe_GKT?DagI$?W%K5kj7bI{B*
zy?L1jW8V`ZxK+2+_qFq#S9?2PCE<=x#w4fv)x_h<nh@&Xkx?EZBey>5hb&1r5}PuJ
zC8cPQ;}(B6=KTv0yh2djc-yjrMEj)^ugW{0)F*NTGO~Zz4V1%~Nib2U{Gzu#i=PNX
zIhr8z)7abgeON@0Ns%r-z(Hpn*#NQkPLfm5K-$}O2`CrUYyS6_QCv{7(<?&13MN5Q
z1*?Vl7;?_^#~Nq3n-C?ZlpA6zpMd;uGz*j{LT@S<pY5d<=X9UcSQVdJ?Y1#~bqa*A
ziyY{v%h;X-?Gcyj8;H*d=evlHC3D%{Tr+i=A@_)ZOXzr|ya}h*afw4+ZOh2?&Sagd
zD?A~{+b`H9{Vy=QflLF#l%wo688M?Q0%1cD%=1kn^4o+Ca0@7&A<F)*FD5tee1egI
zIGqVDaRM&Ie(SfUdiB%psD~J@_k_J78@og#k-18xB2d@bZ~}WeM`q8@YU*mfdsyEY
z6yW(oop<pjT~Yv0me$s{vOPA0>H3e@4)AX9NklBL9LY`W!vhJ>N(V#ig@KQF4mYVk
z{%d+7^{?qkkle4@izebuXeez)mz#b|t#<!hS96@ULFwv=9lgWhg7};J3|l|%^yS%u
zsTMe8_<VL1Ea{58)#IAYwPGb2=j{iGgXIaY4Qs90FS9^0uDZEdp>=%#H&}Wst?+La
zxiqp_p(r;=tIn}hc9%s_U(VDiPsMU~Ptm}|<sNE;Y0S(wlgmYJ{}Y;pVo!RR(4;0z
zks25i+rX~{riIYSJ<<_QXjqYBjPwO=fRQ9rFBcV=dIYu)l@0$u+~A?Qd)A<s1m9T~
zwCp#V6?kHSSZYf3%8&0ubs9DVwS?-ExTe*)syYLFnW2OI4||tQ25~+P?UOU<<l>{M
zB+xTKiBLc>MWUzz_GP2H8bsUmrbw1gQqP_sM|a^0{uJl*hykQtaNw$k{be`oIeGHI
zLj)#-h?m>k3<C+~>wsJ<@DkLDlKT?ANAn}WnrRkMB;1r%m<N)$x}nMj7?F@ydTsZp
zQbcrrnKZgMSZnq{Eh2+P00j>+_<06;y)1o*Q6ZW5xImUICMy;8>Cb{DGZnHO9Ez9A
z1Wz<F+lLetseA}}glO_<pVy)N`m+q30)dbilk_2~oo8P3$h%}tL&MJ8SgDKxD24--
zxZ%IXp8u$R+$>eB@<Mu_>0vzcY3LapZypD$oemY*BhmP%6UaQd1Uc3M4*YDZvSZK^
z0C-ntsBeH=Do=2e^B(y&O|eQ286Iy%0!75$;j}q6wuN-!$6$Za3?1nFHaz~aTZ(}s
z9cG32l?yi$sIIS+$1340N|GZOZB4j41jPO?AJ6r@<Ier>B~MOHqQjnhx@l4!4Po<V
z6XHlhglS^pVa(J_{J{{T4C%7~pFv$O<>kWntQMD8e~pcp__1+@>$N&279o$2(`<d-
zb$hFj!{8cx{bz9PeFCemfmkjn`RC{08XnQ`sn%iL@1Wx~p957jj3-QUcL1>0P3Igv
zVSng)GH;~8RxRFCIOw3`4nLgctLO0JwPIo+{Q3r?lgQ6{@>s0<v&!m@4haGr;183Z
zQeb#HBw;k5NCiv+6oWtlg5e<V2V1q;v$pp!Z_jf!CPoq2VO2%`@j9H`z0A!QgxT5I
z9nuX)<`K%ZMUhA|0OG~QPGxs1DdTBfjei{2G3_-60tEsmdKN)>e3g*~f_4HuZ0&zv
zocTD^)_CQn9|3=*$XI#Oe*2$qn@$y4)B*oLQ!(#<iF^&#R$4UkmOndXtWB`cCCvAU
z_+)Pn>Ay1z<De_tg-=Z}-A%Q4h3)TC$<UC|?+^GORke*DZ$6?NId+8+8zPW!M4pZl
z-et&lT%}Z9U*U9$=W7+}RDHYY9-Xq$<Ji%=VEg*P6%Q2x+<Myp_Y0vx;<9RV=424J
z*Z0OvpQzUI`-ii8e<@bEPjTHoZOB;q1yj~aAB1TNBjagNTrLYuoNc~~mWd}8E>XZ^
zF{E>Gk?qDNcbQ#dJO`s#^Y-ze(LgXdUT=@7Y1z7kvqy<kKu&zN?e_QR5@zqsUYV!r
zpK2kE+EqnjvX+&kH9+eX)9YD~>zQ3Xd~oe$ckp1?Kz2mnQ`rvt{(kjF;?Jp)+#<&}
z7=y<YLsD9R*WZY!kA*9&Y+U1J1>JG}H+~xrM~V=mtE&#7Bbhj6G@<68G~PBm`G*mR
z6*wC|;Ji9P08IqY@7f}ul7u0dfs^BxU@9y5xX;xgqZi&|WI>8XtL7S`LeV|-X=B?q
zvQ9cJ;E3L*brtE^Z<|}R->Fn@^v+6r2>JL6zdb2*kx2>-EeWM3{qZqf+x0}It=(_s
zN{#SsxRe-73{hVyvVXo2wID3zp92`xmGs{QSsca$@r9CQyh@n1k2W4BTw%D93I-Me
zpu|Gk|ApXp>4zYkSoR7GQBN||dcJyxb@HC+rA|rLNBMo;)(4SX{}PEp%Ec=Gkm_%z
z^T<!Z*s`ONUYA9hkK>y!tq_gc<)jb&*DdQ1bM6EtQq(;LKWWIB5MpF6g5%PM5XbU<
zDlzQ+;s`SNy7}!_+4I_#mT|sw-#|9)P0CbYX3fKiOS@Vgd0VV7juY!=Jms<l<}&cY
zExz9Sg(+_<pQoj$zu+;eeVzSuP=$?kgNL?m=Re$#b${59-Jyge{`(%BGQ}}w0tno*
z+j2q`o0cg10+nmv1tJ>`@L*b;nwu|eISKajO7ar0F0GsFC6x~a!SMIU{xm|qG>AvY
z(<pMkq5&rE_bJaHGSz5jivnbmf4QmbaJ!sIw_T0VkE_x`SOd+2x(*!!#P64#CYC<I
zt&Qii*0)`uJjvgTv_c$7;7DCBwUr1q4tk%7XJ`?yaI`kpNac}Zyt6GC1qWwlkS2xR
z*;Oq4QqAs^rfKn~KlNxc1A7XMK1-zr${x9|97?%djag6d-$KuqK>{K{VM%Uyn%bHV
z(*(obc9H#|rg94MN<j5)zO&8F{#J+g<b^lRP|=VD1aWMN;hr0E#WzROvB|-u{|10T
zP$GqRWYW{WgbNECb?yXr_t=<3LC7j9d|uH<F4PxSdF4$WDDouRUpb8ro(2K-ona5V
zlWBS5N6oaOwX!}gVbqg3`%kxVxHr5f=dOom4c6!9)vc|)TpgFkFB1;$0u#ADk^M!X
zj<gfltg{VU`h-yarA9nkCcb}6C|es(Li89@x!rB?Ncla^HcUJmlI}O(<JZ?;$JmhW
zota@#LgGF19ckK>LRff>+hXdN*?1%x3?$*Y^A=6*{9r(;S!{4lO7eRZ1*^x-2NQM+
zTno!fOJf;m2-rBWtQ<Ui`_=@%)9_LkiGLkFdWqeSM#hmx6DDS`d-jWxa9@x`vuqKZ
z>Uu0Ud*g#qS|SmFEYM{3zf6iJ`MWRr>K+CIvi)q<mQJd-Y3KHf9}icMC$hW}O`1pg
zTvJBOVpjUtRaIww&wT<0h~HWRhEvLkYyY$mwQY_WSxhEZ--iVf)6Szb;pSG#jh->g
z1{+jOuD8;8GC>Lq3e_YR{_8PSIeB@r9OI_cV6V8Xb}`SWKC(8-=xO*JyTWg$?fSfx
z*c(2Pd#q82qcm8gMFY>lHne3FDbArggNoJ6on0?4dJDF5b$|Th*4ryt6n7lW*R(8A
z)X!eq@}JHBSBfl8G{V*86VEgp>9!d7tal^$ZfnF`!S|BqcZD7J{>gYcNUKXpScJE@
z<v!xPO+uN1yOP_-_lI$-<S%N^m(Y@<yA_CE|NJsVvvc_GDOR!?Zu4X=gs<y)J8GEU
zyt#X|hdPKTTKrjrRE$!(>G!9eb2u~+>bF;zFH~YOMNmYO(*?38w&#wgPbFh=6HZ4n
zgT9ZWhFwu|kCJOMPCmDvaw6P-#QqP0N3B370Hy@YulX;E<CE2Ot>dc>@6lBOugA$U
zFtjPOD$<aMkB?6)u4#h0M1e9aDP-`W1;V|-{4O?X#4H}A)NNWxDL-4o%6-5x5MK<m
zwh^4AxK2&6vL|Y+K^N@{>sb$Lhb0%j^Im&9YRxjtl92lWRT|hQ*4;ZL?M^2t727$~
zH7I4;>~0w)S}A9C_qcnCTAB=MdzyE+C|K+mory57sI%hgu6kX87?}r!ezHt&^@NrT
zd}q>qIO)cu9q4fjrRjxRGTwX){a%fRb9f&H2MxK}Sq0d}m)YWK+8c-aqZBW8h~JUG
zbzy5#*#oEZsGd|ZU-@t6*u*C#D6B;@P>x(ze4A><Z|0Dg+Ad%^-e1pW&ry%u`M8g1
zu%qM@6!IH-+~~(G5Zg+ot@4tKwIYE4K)=by9kHSul}MV%fv<)#K?@l)k>O-9b$6Ss
zgRuWP?On2xAVw^ni$m}~YUQHgGh$A5F69D@31Qf5?FBgGFQq6z`1{qxKms{^`vW~_
z(}&PhOw#651beIJy5wP7){@31`0F%25)fyEejYIB&4+ZK!S8lS-^lbHE=$}wYKp_?
zcsx5wpUWd_!@Jh*6Sgy`;0W%4Sy=FJ4Birv5=ID%+@4CC;*GxcOi0jz7uWF*g7Tzx
z+^E~MnJt~y&dR37M%QgUMj8(s)wI7}yPi~{$GTjK7N)No?mS}yeyk*)$PPdPQVbh)
zZTKUaLbr?b%$KK%rMs6%R<7V`wm?$M8_!$+90b!XJ!MHarSo5w=rzx?ET<f5xSuil
zX_8gq_#2fHiTA0g#3U1Wvyej;!cy3(-e|erx8tP8;`=DN>F&rpfA)Z1(e!9g9X2{A
z|LUICJj7^D(YoevL9L-NcDf|xc6en6P2<w<nFMWA58A9cH}cCH*LH`c{&w2#(QJC?
zbMf_lWThr<?HaiSiMNc@m<~3(+;gTf8Cp0Qf6+$zb^PIgl2cyvlM3au3W@Jm6G}Ae
zN0{#hQgMK95#BKtt^F$tpCc60vvu<Fq9E+%ZyuXoxgr69!sgH>sqzm$a;4#Mlbmly
zO1RI%qaY&MU#n9fj_E{pE7}T++GsuBxRZo#v)<lviJ_6P|3}l7wKlhO_tuCJZXOkc
z$Rb{lt^=!-*l0+?0k+Il?@M13Nd&pH)%M#3>cf9^W&hT+JvyGhL`6}T3fu$8Abh^P
zF7Lr`tw=O2OP0{^P@Ei+b)AyiH>@_PbZ(6UL_`1Boj8VV38)3*EOHeD%t1Lhg)2{o
zTrQ!Oo}GXL-Lc$<a0un<%)z#Xn`?NTF16Y52}xO`x_z~9{m58uq!djmGT1eUZJ$I@
z{_d5&VssL0@PSta0x8Y1X`bkLL3*8FJebVh8K}fZ3e!L7s;z72-A*7UB{)V>XvyOb
zl#DrzP)oFFkSPRA_d&b3)b%N+ueTFI;E<(@Hb_qbGX1QgfzrsUtE+RrAbf%syyQq8
zw<~^XMS%e$W?_(gz?7wr3l_tZse{FC8T3YIKQ8{SZ?+c}La=KtFPT*{$EVZ&_hP;n
zf|D{zqO4>;wgqHpZ<x(rPvasia7u+-h`{7`=YJlS<2LsuN<eVL!HHHt=}HM|_-9Du
zKqY6Zh<fsRxbuw@zcJU$!6TRax~{_R%<1F>hH<k=f%gUyYb#YBwIXE6C^~r&X%V*Z
z>yjjW_W#iVK5#t$n4|*%(u$QVn)sP!s<=hcX;wKu?OG#*BK#^%%gMGleS3{}Pxx|V
z<|*|7iti8g1TSwkN?AkU^RunJeL|r?F;$7SmKG!jtxBJAgzD{y$~!YQHa2y}{&<*l
zU&9IBucTXRc?9;GpqnBUnM9;FQ${nz6kr#Gi*M=E)NABFDZ)5t@V{LDEA8^{Z6fsZ
zU;V(!@dcRpO{iJv25?|`$cM-%8Xo*c3$ZhVnQ<5LPjB5+1;mv5=w=zxT%L>`02fWN
zPRT<jsP?{bQ6-Bm9_v**K3~zEzdwSqgnSZ@7wu;h^EH4ht7%&k^Pnv50F<OV>pAN5
zoQk@-_Ba;oA{37Z6aga=_OrU>MKZ)qq7WCcx~^0M?*!^x-z-q-IUCP%*DB|{Oo~g%
zqzdrsjSjH<{59M(rpolo^%8TeUDm85Q9M`C2RZ_48>#Hx)#RY_t6@oE=EFujl-4<o
zFLTFkZDH*4&Hk^vb!==meEP+fGBoM6osA>94xRF?>lT)lnNmGfey`=rTP?xI#GxQb
zEUtoFKO$%wKJC4-&BwSVel#S`KR{_1*8b<<$=iW#7>JzG{z}F5b^@D~nYB0KoqNC0
zBwCeCq7Hu`&xwv7vcP?STd$+1;key{G5NAmuGj1y#b&9f9GSO%S~tIT@X%bh*5Qmr
zm}2$kZfP0BBES~3fB-`E2u5wsAz?pMTV52Doh2x=6IGq!+VqWz6MPD(!-=hKTof<;
zl5HPW&h>kmurT=;U3Stxv~~QvAQPWvuv?HE<!&=<gno}wxiF$oyfpl0>BU2mqALA7
z4S=>jH4|d14GE-)<aXId(p_Cm{)OfK=UEGfg{8SKim6SwQp42S`}<sXZoqq(>D@8-
zZ)0zY+rO6K@BcbP*vY%6Iy}M@-yc?yG|o+9Naas)#YN%T60MAu?8pNLrYSpKR;w<y
zM{&Z*;s7YK_-F;yf@<o^b6ZY_qR}ouLtYMPw_7U6xV*FRJL^AM`XQfKf!^NE{iYcB
zp~3JKE60ByZ~vf#=4+%$VFY<-e2z0ivdpLzdm+~+aSd=`A=%9*3~{HB1;?EM1~Pq;
zSaL{~@q_0{4fT|Ah%`es9r5f_RDk@`u&($z_6kB4YKoCcNaBN%FgtCihqfVk@7+GF
z$C+Nq;(2xF=KkLX#pX`$)b>|_&(b|9<%^GRs0KTo&#&)%5MFtN3jLBEq(piJ)vw~?
z&h|VCPK3=|DV0&MP|;)#kNnI=F5K@$xMapCC?gb@FztL~h?JF7(!rFlK_WXB=;1|D
zDxJF!zTTh}dh@8dUcgV4Rt>tw&}U3EQYKSVFOwpVPc#0Oj4ssNENOWJsD&UBo5SbT
z_YZJ@((-kS;0Vi{ezM2w&Ic6Yws%2%{8l8DQ4$V{vVK01DncEyCRg@ozJHd)ESvJ?
z(^a8#B#YzIjkF3Ig|rGJ<M{F6W;Tw+Ix-e8f32wWKHn*J^AJ^-!u!>$5H7`NA`G%5
zG~BeIsgNZZA6vtK1dsCdezCGB@2o8ka@V>>4@XExI*mQLSUx;08te(Zx4+}X2{|SB
zZ>Q^W_s=BM&DlU^<C2jWDk0by`DD}iZl0xw6|X^RL36%K7edkhD`y2LbMZ7eIMQqF
zuCz%Hb@2Vffu{;sw)Wo5hVP!zcI7o{Su}H}{Ta`3?D9D|`0S0SsEoIl0y!(XJN{d|
zc%93Q!{&}I_3;b%3}Ns`?Dbu1s+HOuF49Vtp#P7ee)HdL@93)_-L{wQ)TQMWGh_dR
z(@MK&d22ZZRvv*EfMdX(V59x-EyBJ~+)9`Q?l$B?9gu)EubH_yWcOc~Ck7R%0OLy$
zf>FjMnIKSIyJ$8&;yW+*bE00Z#-sIzkP*5~VFf}7-WY)s9vC?bazmsAT>Fg<h4R+P
z)7D0AuWew*een79`vdV?ASSiK<-<<g>@hXF?nmxZchn#X6?C>)Hr(?mUa0e$4Z>)K
z8f=gNRf8vN(E2t?+sI^)938qAurx<td{`mw_S_G(HI@0|Y>SQcr;P}$oSIq5nh}W0
zJdI;7`AbE}NFT-PI(8^GLRPd%P`*c@Zf3Fq!QN`4&dbCctB2<KMUM-PwL3Ks8?@Hs
z5;Na`H6e6O!4_55cDbMkxmry9Ok>iQ@#-V3&Q5C@>w>Qx%^u;X6&c7WD0-ny`5~@(
zghHJn$TJE?US2<P+h0w$fBsx0#8_qQr?s`MaJssIub;~TCVM(u(S)X2(B%4OogHuX
zo^JF?2lM`CZP;Z%3fbIZ?X2Aqrqs=D&te^y4OQ2<FTtC8cc+=YSHslFvMe#)p27<@
z=KQg9#kS|R1)+UufDmDLFXkL6U^2#`6e$I?dldf#ZQVc$D*6A-{_rPBA|0S|K67Nk
zmSQe>owH8gKtYm`uugzq1v7xXPt9t}A!QqkQGdilbe=ZyLxkx<*<y|B#40$PSg7Qm
zxlKtC#);{YAas;fFk+IrEtARdA@H*<=wzd1^zBTL&BKx>;0_9(PIs--EAAl2Gh<|q
zA`@TN7hIHD#!2-!9p^vx3`s>=u-#zN(u`jvQ2xE8wN4}nHDktgetnXMv^1M9Pn&sK
zR#QhON>w;V2x69<p^eNZ#YoTkW-xjtz?tr<fRLwjk@g3Cr2W`}>G0?}>q~9xwmx|2
z)^!uNTOvNbOr)nz(M}McQJByDd0p!&1l(j&iQIfs;U~OR9g^tDXA^n&wcYWgX}k-}
zxa2qm|9XAQ?jXV~hcr8BK(elFxp(V9DlO)w1~8T&N!7us)6FrFSI6U(kd~^r#|0(m
zs932<N0Z`Fh%3XFVHM^-nnJ>PH3hRTwpGEjC1Vh7jFWppS<+xyT87c{61L`VeG7lD
zALJzvSN$OI)cU$9M4!xUfBc-iPh|KwlIpd;%rs>c;x8%&(rWIMDW4~eG|ln=M8QmC
z@klhTf?GEs_VZq<`M>Oy3xSb~YSe$>U_xcY^VPcDL<S;1s`o+kXeO%@-2fS<Yn0Ey
zlTFJA2`K}7g&uP@Za;p+vI+=D)~JbthKE%LdAxVwR8^5_QxIgzy3&VNV4+cDX(BWW
z_Dh`Su4Q6YkqFM*QDL*EPy^1v|Ly;agb3AA`U_xf8WH2eruQelxm1g(gUMi2^R^A%
z`o!cU&X}_g5TLIQ8Cbc9Y=9@6Lq>*owN(w<0>YEgr=7BHD3g#dQQtaSr@}GmAC>55
zI4b3TgE`Z0?pv8aIDVzdz7tHEjp}O|3On~3>bYX=yj<<<kE&{myhblk<R4a?6!>SL
zV4Pi|0fNvBwRAft7Z#{g;A|RLLz|>BaL?A7d)L+o*y_|P>MDNQD@EF@wfm(iltZ>8
z_8tO_@&I>)Moww#@WrW6u_t)caoajN7Jb5eWUD)jD!kEt)65|tEPLG!(lT^~A}U^_
zRQYB@3Q8lpkxYjUNPmjc+*v!YGJ;^8VG^CcY_r`~phPoud<$`o-(|`pg!beM@ZiR<
zdB<?$<VuDMI+_b?cz=f(PSDILX=sUQc7_qBS;UbssVKl%w9<&wvP=+-_YOKmol+6S
zkf|+aQ*~Wzlum-4C040iv?}?8!MD3q^EnlB6#vtxH1(1H6<cgh=g{^ID`L^-`4wm%
zJ`jaPm+GWo|I-LUz|;9#c%wl8GEl{S?Jb<i;5GdEAV*wUq;<Veqr9%kF=Tm}Ckyhp
zkN^Y;L9;+0<exP@F8`;Im!?chC!-|y50Cpc?!0$sM47g?esxFK4-``vYxCG1#LnTZ
zt2po-tX0xh(O7iH2lR~W9LReM!x7fSB^FKIx2&m50b~D6b0>27;#AO?+HWvY8TGe^
zuE?TBN5z#x7BarJ2LTenj#ABfKhbc~tk`i;!ou!$jo2p+IkxqDiOQP(S6PJOCdSXA
zBo1!NYh2wOP~6-v3+0*XBN#wJAIL-a(VtCKbDR*)+N*&mbXWl5pmkA|qGXYXI7P*^
zJyXibEK>Si-T9}#PYnwwP}SxL;3H6S(|_wu2zRJjGY0R%tTYl<PB213h}o6MKWzxH
zQnXteM}*!>-WjR_?pyiaPF(17c|&jDC9zYj<;E}8u9@<Zw-+jPD=I^4=;_Mu(=v6p
zW>GiAB8ft2NL8lnJRUBB<2;Z0t7>aIWI-UT<;`7JYjhBX8Nx4}ZZDDiMGCL?$xEuB
zS;G&DgXp13W%5T2t7~mONw<|87E~6Vw|iA=0zM%}N7ftO2l8i|9rB&6ld?>LvW>Im
zcIGgq{E$}u@4%N{v+$Z9VCMYUBd47w`C@8IrJ4b0$C(}Zg^KCt%d+?D4+6rZ1wT=$
zrNBT%4(``}oCilg%+}6`1(>#lrF;{+77?=1z7ENJ14)2vntSHiKH~)MO;^5*M5N2?
z4QC_nU5gz3L~ZA4H&6Gm!=@Nkq^nuzJJ^4u@kd;p|1oO=W}>%Sq)O2kgb!q05-JaQ
zgGql|=ymWtT(jd*UeN1YACe5|i%5G0(Xo3pdfjI?)+{(*?Mh?g^lJUsD=L5}D%VS@
zliR-O#$S?jO6Pv@-`*s(JuaYX8>IadYz?6scPi-oKEuKYwR%JrYN=(9CUgygBkYaI
z4Wfsfz%8N@$si5|Lcac<6SCFk)wa8V`{m`9$%zRRh_r|ysgRYYmcBLKxKW@|S>4nY
zz3|PM#lqS@K~zteP9h&owekm#-#&oPXLYHMgiwkajTbaR+t<xMGzkX(2O#{fS3H#a
z|9Qou0xPduMv<1k2Z57u6MVQw2QaWF(Zh?g_%d0+3{VX=4;&MXYLAiEW6kSemsF!Q
z(q-w@fc=<EMHP++g8`yhv^kVA{w@K*qH>o#z3fEEzFamiv@}>3($qG0N@x6G8ICvL
zGXA5t-8V*Gt|_>;kP;VxKxuOwHAbvW6}$DfB=tl0(-ZHYV%E>;G&6S#h&26K)&&!f
zbg8DjV*-}gw{K?6`up?uiMKDAje@&n@KzgVi8<V&%92*_Gq_!*K7{Uu&y90G5&Vw;
zJ*eH@Z5;I(J}0*JXz=4)2CrXa>t-qJyC(K&8-E9kyz+TGR64v;SwbEd8Wi%!{9wwE
z5X<9zL$O$8XzF7b<z51bW+u|ht4_gQ!$^`41G`JfZx7f6hb)WEz6~Klru>H76YYox
zcpcMyqeM`>DBl>W@Wj7Jpr;n0ja<77HA#WolIn%T=P?m(X^9Tc&y$eC`ebH7jVxsn
zIG@MHCS68<Jd~trJdW4ZsjcvF3mMs)#M1oA-+|hb8PlkmJUwOjF(EWCe61Ld@?6ig
zu+-Ekn%L7k$Ox4a8VYl^)*)TjJh*U&()ui~!FA6jT5V=w98{9)pEy5`51?Y7kC50`
z5bO_op5j&}ZSl(ul4e24S_l+QDVJ-1;<7H$@B1s?MiYq8n|mAsu_^3Jw}XAEUIXVV
z2~D)kyzL{BEQxl8E+NmmG`^>S7JpyvF?9k0O95rlO&;gikeIQ#h4&AghHq|=C5j|t
z{`V5$JuBl}PwtTB{!h3Ql9HE)P^?N6ep0|4UBF#L7G^shP!luzoZvb?!pU%B>-Wr=
z!B`^p8Em^nh}L{Ry2nGoY^nUX=Z&mQgHZ%=I5Wyv|BCU(t)_!kZjOwkFY*_)bjY;v
zNPRykgYkFvi0f?nSk$k&wy|N<*)C>7IBuMKda91it4{T9c=GCzY`?6%84+-|LEYKx
zyiV}!uAm}YnH4V##j><Cz-eTPZ@LrOYlQmN%LZDgNPp~&a5BQfH^s-VjuChmB#Hl!
zpQU3{)(RmekJmW2KlTy&2V}K0>P}&_l;Y2U+7+w*{%UF5&@Gv85o~q$%sJ<>E)W<$
zGwLb2PJOglD_%Ti<ytJ5w_#aZr?ehq$7iwr@zbo&irs=WC%%|RmQE7B-q>?bn^9j#
zR@W<87ke=wjkPKiK3l{pW*TeRYRJ_m>2HFe<zs6Apb{#}4XzwG10Q_KuyJ+4$L(}T
z^LwPp{r1bG;3G|Irsk4XYxFGY4t(>015KI&wOp|(^v7B20mKqTm%A4jqNaC|jmsa6
z-vY@d9b{PHpMUlb$u`<m`cZ0Jx;;15VFZ0LhST{Nw&_f)o62NbrQgpzoPo4-s+jOA
z4}hJ|7U>&}GZ<nlTtJu*JFrAnawYI{gJ8gls#n>r@RUHEw_7ce9>wYoJB@w+p{QiR
zYc!weTmWTx#VeMce%5Y$hfbP0T3c&r(Az~+h!AWY*KD8EJBZ?B$oX@VxVm-KlAmZc
zJ(HZ!T&@^TI?Kow;?elIY?DZVo_UNzBOr0Gl})hW@hxE*ItqiOzLew_j8tfk7tzRd
z#46zzMA-+_04Ds-$7|1GajBprBIB9D*6N`HqDO!wlcDef%QNc9!GsU7iTx4DXn7dg
z2Dz2x{R=W-%-1A~f}n!J>G0;$IiLao-^h+4|3Hc0y|yFT*J*ias(TE&{8XaJHr88p
zwh7CuN$y-H!JU_ep+^S!zx?$10B(w<6|TzK*rR#D10hKF({Ug$mKohIlB=YaEE0HN
zqmLI9hlFLk6%TzVM;;tFqxv8s)v`*RcD|)nxkB)rR72GAw=D0HD85MALEDaFm?|Q}
z+qLF|DkmAXcgfNh+iE^MN5=A|#M^XyR4tTG86rKRYZY#mWc^0%Mxaw(33cO&F338N
zB*z_uhQMVEU_^6Dc`P)yG$<7F-vM)%kgW$8q8XW0RLt=@N=ipCERiKrrk7d*S2{wp
zc~P_~X8WA_?5N||Hjf?e))3{cfx%$3DFPnnaB@i$(h%^?^`!DHC0u`~tmeZu4zc-W
zM{kIT9Si2eqLyy!L=j!8Rl|zhO!9g*VdPI?X{Um+&fh0%qkd(#m4)}_V&uRIVIz@+
zx9|=CJIW3B-9#AAGd{7YWWTMO4`70tMz+56wtxWZ4>O>sOJ%#C=np-7eLgK>Vl&a+
z1e&pp+ucu^c<k@3pAyf=a-0FmeI<;@JnaJelbvU{-RzdF0U7;gtS;k>gQmH`PI2oR
zsWCszREljalb)D?T&|j;*qZm~Pl7VM$sD+V2{kKyrL8m?>16||bVG30HbTv`9v4>n
zA@lG64U@Qb6!D<i$`<TLQCd+Uixy8nog;387fQSoIKiBUDT`-NAz&zCr*S&`iz^>t
z*UonTs9u{V-T|P*zB8B^p$jjTNN@;$^8pQVYy6}EMWxrSO#b~Vh(gRi=E?svonmZ!
z67xIhKnuS7s~gPyhC>j2w*M&4KU>Ln;4%6d@1RP4!og03njv3Jjy|nDYz#h%z54P-
zd;gxv*|jXFav+%~kE3Pd`Rke62X*UMd_|`ONz)qnBQL4RJloI-Rye<q`CV&?C98tU
z0)bAs7Cw@AjAJC&jEj<ULkR(o^f>*aS(~xCwWI0W*uywUTXjl&bp;yd#A&o+>5OC2
zXKYPw&u!?|S0qF4ko>qR-;tRF<VME=<l3WT+1cn(36Xr?vvYCwK_HzhL*me22Ersr
zWsiq5)~1TH^0KmM4@M1t5nYzrQGq1u=lsdcv8(6PFcW$Ehh)gr#HlFoj8t$5Cv?%g
zbJu$4y6^b4ppYCvV2bj{Gj%F8i`PG~7Z%01QcN?$zjN};v9BObaAdjez`+W~xi{o_
z;8~NbJ_UfCfibrsMwQh^tJU!g&A0fcshUUPi2(}PlgOa`K4$&+!&MU8v<->#xGP9J
zk3)kl`sQ*QHxhwb;AdY+TI@xkQGJ~V>tazoHV~jmeX*IV)w|{teot`1l5<wgaxx35
zKWbQ^fBHD;FAgS~afdBqNJ{y3tb44-<MX50K+c>)LRmRIk_4?UTSD9?y=i*Ok$hhX
z7b7E*`a4gOM$glJ8hcgzh~zQ6KY+BZ?au6mpB9LBj1k~PY=3R5Z^ekTGaUHlIJAD>
zK~yfwoY4=tyPK~WS1LKxwJ#xZ7Bj&aS#ns-x3;M*7>Y`s$PwqKH9SWBc0C`#C!%(P
zxMqdSFe2}8OzPNmKw6t|kKfL#Blj+~|B3GCzxL0+&!zx7XaA_dDlWAaLTtz-hAsu~
zZ&L#q$yWJcY<eVn1<gRwk=pjbsq1FLnGP2NWf0;qgd&HiF!?E+Qa&FzE(F4;s>2<_
z?S#`J0b`77b*o!B`ba#_ZBnUX6%BPPXQh0FwwJCDSTFDjG2JD}zgovmec9Cd-l`j&
zo1v7F0Rv441b*t@W8Zt?re^!z!yS#`n!KCSRgkLCY$@S187WfY*LEfnf(AV@sPSC}
z6loplKoLak*oxm|11%1rncR9jUXdGa^`u)~yg&<`<-I^_dhO0e-yVb5zkM!$X7}QS
z0VRkkX}}!oF49Cr)>hEa$^TfSnhpkspMnFYt)dt9Ee!kv+As!LA!`7G5rskGq=l$X
zme9EVHNf@zFAQkq@JoUjQI>JX;y=7jG8kMK(QZZz>N%KPnO3Ig9k7rP0~Gwoy_1TP
z8IQj*8jXYtqnUk8aL?s*!$(YL2cfo}#9B4=iN_c>TkUE2jIoB*`XsmmFEyE&blpNX
zMNFZKr2&@uA9?HqG8lIaWF|#$DZ+SzQ(8H+9Iu7ls>k0;7O)Iv#87{4Wfibv&1gR4
zNG?g0CKt`2urx+rtP01*C%QYtE(*O+#pbjj>KCakqp23b8DesnXxvq@GL%&4--joD
zmWptT?LyVoUQ<MSf(}yL^668y!&pD9*FJQ$Zq)V6d3tSL_kniBYlHfKju?pv2Wu6N
zy2^|vq_=@gEU4t&!ww>ul*3`@%pctR1fNya`zoo(E)t9bRkk?9vM_Khf#nECunnK%
z%iNrP{i48tk~|Gtux*_BvoVotK^bF^N;B7XE8HV!c->qMJeZ9J-vKHqqzk%-#bD8b
zBvY^+28?8TndzpAC>t~WIGYu1>FBUJp(w&_3r|;LIs6UfXHKo2nPPt>P{peEMhp=N
zKYxB05nEb`=E;_WYyUt>Z{mq1&=M|<5{TZHzuPxmkmYp;@BW+#onYa%Nf70N$gTO#
zU!o{9d2}$zJVYx)*Hrsw`28U!E-!l>@{k*YqkTA9hiz?t6mdtQQ(fhk$HZv9LKh=?
zoZt5>R9zs>1nEFwawVEgYJ5dU^H2tv=#K(Gw1NB(iVHtm^@T6OzZN>LUIW*7pxsC>
z%-9E#Zr%OKQPS>UcS8p)tBUoi_&1zMz*-oy+E!b5vb@>O`&a6EWtThgGC;Gu!#+|2
zl8C~=A-L_X8X<Tqgdq3W3U;}oNt`E`xbxx)MFA@HkSno)SG&mCykE4m1x~e`rV1JI
z2>RPueQqdQ4RLuh<#`0!#ymb?DC+aJ+~y4o^RpUvztqZgyO7d=9Muy0gSk`cqXEGn
zR{}l#ZAJgd=gW|_|9p2$2fIDdU{mT#_r_pPJH{3E|5v<VC<ZFQm4on+35YPP1-|HO
z*~s3UeJG*Jf&m`Kh-aIt1tO7)?}7)3{C7<y){#SM0O}%uP$`O;d+8|nfn=!?C2m^F
zi%)SWV$W%toq9gi(vlg)o~%ZKU#6+|3l2DW_$nNO;dNipH6FPqdlAK6z6P85qZ>-V
z886lIB(;&D`RQ0k0y#PyeedlM+Z^Tx6<9~e`pK+0SH>EB`i!Hm09v0E(4Id^@KD;a
zU&sdCIoIM4g|+UZ6-(1a@c7<vE({W7@|pHHrk1q|11QzJWhm)G-5PR|f?JI6sJt$b
zSua^H5m2wV$mUU?f|CsFQiFv}YgE_>yY`2+$KB9#;^K|${EW8xSp<!u2vR}6<W%>t
zX6Dz0X`~OscD;i7P0AxFY=5c+Vto$o!ioGfS7F#OG0_mS0-8ohq!3w6%x;s9i@ql^
z+!h~{bj_r{Tai*Ft^bMxGw&+XQYT^VoTW-X9B#3z+CsELAp%2`g88zKxcV}o6KXpE
zZTzi|H9d9-?{+v6sQ_6DrYAw2(s(K_O`!}btb}}fl_=2C-N~khd%VdN45OiwQS^u{
z!6GR}Ksl-ETNQLpkq>};Xdph!=s1mUI#L_K`ZnnOuX*PiD)E2+=iz^fR48@h?6uaV
z_H%|Vnb9j>{E=vtV}Wwa2sdnaDLSBhM4C;A2DHH^eHk^ht!F_6xai-3r5a-z7?y-B
z^p|QOm{8>$8u(C5WKufNInZ_yZ6RNi<Hd=n3D&%J1qEMwsC3186#}{n$MQ!GEmc1+
z0~+`TR|a7C!F8o%JxB4Y0_x?4u!83;rV%?cEX{vD<rmM|PFt}{C5E_{2VcBCS+adL
zgzL*e%}v@WBuON~U;K?pOuXF&SBaeadZpRavRcCa!#bnRXU{piUVzb%w5&{%l6D6Z
zp_z+jo>VV;HYOV#nkDGqR6AR%Fkb7gC7II4-(WDmSXqf@P=7L|x;kl=n3rJu`MP3M
zJQNe$6qO`**bRSS35}aVdMvO3Mlta?**Yh2KcfyRr~N^F!+#Ycim>0eN|r6WEAFTt
zxy9*IUCrp<E9Ls~B~~7Zg5K+uIof;ke(Idn-7GBj`$;h{OICEcLP??Zh!uaI`uv7V
zS!q`9>Ei`FVh^u#x%Nh9+=(7+z6-zN<-%fT>?aoiC`xMzbQf4uzf+m#`{?0->bp5E
zp72g635|b#UhhBUAlrmtRMzH9p#oim<%xW+VQGwmfF+!HC=SAlJ$@Oq4(Np>P{wos
zD9Q^bfb`>oSV}QB$pyfvTsoju8Vy>mM3p~bPG(O#Xj58=hW`j$U;HKRGDTQ3uO0?`
zQS>v#H7%JZA^R$b&s$GtHFOnaL86RoYx`05<Tt_2xX=Un%Y>znNYI{VLI0Vgf6V2=
zhv_rt^KPPUD@;3MYx5uy4E9VrsvY<o9HZ!NoR2cJ^+5C>VK`|52{mhxjH3FVR6pfv
zW(mbDbB@?|j!JUH@Fn$&whhKweBfU<i}Rx;;Js#-@NI%Y41mzv>Cl2N(bWTBFCumA
zZ|C_s@Dm6$><H2kJ&G96N;YDgIa44;<sXzWUq7AB=#=rJlzj(&4*v56?DbHIo#Nyv
zh^zPwQ*D7+A&fmZT#xhN!o?4zYN8ua^Y2eJP3PSu<Nbxn>qo;kn2q}d4|FmfE<(2Z
zxATIArT&=D{9Z+h%LYA3)y*C9>3md4=;Iogm9ohyYQOumf=i(EjvOZv_jvp<68qqH
ze7_-Ln3)oBt>r^n#pg4=i{!lDDqh3Wk)H6cT-#U}n0`ej|1RbK`H7vJJe@_-*_}-?
zUVg|SU&B{A6~07aU`0LS9G&oeox7^h?oL5G;BI1-Oy7qCRS(RZaFn%q^J)BDL7YO6
zo;g*<;!#g27$pE@ZUJqzNwb8#NqITLMakWUJzo)=T4SOztUxtI1K3~Wf|8+3jXLni
zv-=$$hV1b>efX04U$dI70enT^q})eni)N9XDTFZ8$yp8*&E0y`--&&K8kP?-`7suW
z>SGLCxFxFly7F0U$r4|2C8bEM*>#Wx@c~J|bskAwY0fHxylL){yM2_!1+IGxriTpk
zH4w?KN)8>wMu^(?`>v`^#w2TUp@HH1;$%`Bof(1i#i7&higOGKbOqYNM7bwSUwz_M
zt*Sey<cUO-rw8MZb4gE;kYvoM&v*~q;OdsHQ2^sjwSh2bk7U^@WQf{Rh3pe5o_C&T
zg&PeFUlI}>#e97a%2&1E18aNzos+tfbX35(?`o?OzG`vmVzaAtSP(`yvx%#D410Mz
zq)bQ3xw3);>F`x`q=U0Bp-BNExT<D63);oAEZGaJ#&z9Bn#`;}iatrED=<}8vroRI
z6t#=2U2=>zWnP|gbak>@{HMi-^ktx=Yh7m;!N=d{fi!(M*uYZ1qmVqd#hskiGwREO
z1dSRex9sV=52|W#is%e?Bv<G&=+vYTLNp~hEmF0NDDG6&@w{%kjB$)esdxE?Hlhre
z>x<e1m`jHuhV!+OVc?aaFr+CZ-c_gJ(SA=!6r5$DBX^-Ep=ajmC*picpZn4rAo@O2
zM}MOu_uHUgZHWv>HP-*Fzn`=GS;J`nT!yxZS~E~WOBNqFk_<;$y`ZGNH%>N4p)l3b
zqZM^Qp`4dyUj^11w(r!4`@Qq|p>6lt;hg(AVl%WY&l%$w@-<neO{)iGmN_Jq?$tBZ
zzDA*UqBvMj+;&5aPw|@EbVD=WSVyO=V}UY-HA=~H?$5Ugu`E(Wj>Q|la8$mrsLcX9
z(9~vGj$FkZoDwt~&_Sg5C563CcD1L7(QI(wX}C-)a8f8s@JPa$P&Z$|$@w&&FZJpx
z%lM~f_#ZgZy;QB~rXWz?$}L9-hLx(4I*f|?gE7j&R{4hm0u@?cp{8aG#X7*`wX(`~
zJtY%zOsIMYT-xzG?WOSgrH7cI3>?~(M0**&!=?qx|FqGNb2IawbJPA2+*tS05*FVP
z=bVQoML?~1(!0pBXC_|L+B;uoX+_Brd~28N1g$IukT|KRta<CvF2>JHtF3sSZin6Q
z5)5@eu|<vl`A5v@MB7v<;sW3gNYsQ;E*b0@`6%~;WqBiymbHy0kE=2K00s_B6P5f^
zF1^@v1PtfEYcg-%mzrbtG?%($&^fSILR@0%qT$1Uje{ed2v>mE!`#O+xWCV+DCCv1
zp_d6RjXQrN5eCJ$xL29NqkTJKH)SVaIYYhVpy+`;kSbp}K3yeWZzRHx^Ua8<;*YGP
zjs6haO9pwxhc^WWoqaXL<(h!oDbXooT3@@1MkF0)*K6YCFD+eTc3w?e2*TaU#l03<
z3)v0yHKKk+`Kc1L+yCL}E#unix^C~5;_gmycPRu2?!lqB7bsBN3KT2u1c#!<i@Up9
z@#3z*EfkkG*L6SVy!Uy&?0n8|uf5h>V~+8k&xrHe^|_;HG-7m=H%rD{k&))<Xk&L~
z(G!XfZ3fn`@{<Y~ua_z}z8+b~-YtJTJ$wXnLvbftpt`y?N6JI2TTQn7?+}c*9$zFl
zipCrXM-8v2=(_5A&i&9vD|Grkxw1O|;)P6oE0ppEV$JtyIg7u-IBYcNj}{etgo(G1
zA6e|q#>e0P>S+Ip9}5hSCby>}Lc-?Lw$6-GR|peFRlrAh5ZA`k!52J|yXb|=O+jVe
z2xqUJgJmI(3noU*Ob(n0?h)4QnN0ywdHVQlNhWaflIhNskc{z3g8^%K8JcZK9K3=O
z|L<R=fm9B)%`V@2ZiMdlF?*iK)sr^e>Ggyp^}~1wnGHq8a#E5Q<dj0Cy~JHjqJY?|
z&fEB`XcXk{p%sL(#BS=<bu}aV58qqPP&;H67vIlju7rIcG&<CeLOFWdP0AMQBSgP)
zwxflS&MH!$V!hk+_uU0j$m05b$i{;*VJsyhupV_@jdW<@Paib3+ET{$FO|Nh(hNFI
zNd0XU6-ar?j;j3)BmDR<BO3mkTjU4AGK>l9+L5F^Db`^1X2vqpxIB(>_X&+65@*K2
zTd7gI-uVZuRy0;=MQD<6+pbsSd)@7Vm}Izoy7`E2A2yJYKL*==9nMb}`OB?-=y`Tx
zGv5@3%qP6VJ<8n9W#Q))QJBop7Uce5P~ATDsn(mKgf9DhPG3n4>x~9Aj&uy-r6ks=
z>W}i<OJG!(?>Oc)*S}wuYudk}C#;@4{iA@;)ANq}bpnejomxXkfBgHNuYFz%F|AAJ
z$_g7NCyj5w^;?77AzCg15TfUwm6HLd?;|{RTIyQ#*@?&|HKL`9Bv!~J)x+P~=1~fM
zwdu7pNzhbR_J)XNB|rBY%g=2LSdtiHV}Do)HVbt9Nlt8}Mu1nVE}K7{H+$;DVl%5N
zGOSrU@)y0F<Cj|m4sL~F?R2;bt}7(#kvLjTp5&(wxR$AU&@`m)^t95zzGg5lNpdBD
zZR~OuMTl&x<V|jS2h!H9!-C&q*atP^meH1t7cK`4(+X1{^HQ@Y&BnMIM_yfZ_uwF5
zqst;{oTzab#^}%@u5!ot`>(CP0J0CFos>UA2DCzx8z+5O2layb=ZAQD&5YI>ilF#W
zQmB;bAJG@8;t%1v1j542-SiSvW$|Squ}$HbIQy|O`cKmi-F`c9?YaqlYW*7GeG=@;
zEzjA13zt;qX)M0#z^lZY?Rz7%rcGw6s8op3i(P_0gA!cE%On!<5!W0&8$BGR&Dd-Y
z*w-^V%Xa<Z{d@ZD<;wkk6&?)=6sCoy8yW9EA8sw+wM5c}3D>9MM$hjG`zC2fpYkIH
zT)|~tt4k=&oZt!w-6<diQyn{MVt&nKs<sR%W64A%KKib{g`Gzz2VXt^XTt>4&IC7!
zlv|zLr>9WOr;6WSup+$3*9VnKSSM$~+7V~UVM6_3pjY61!DLlsp9ubX@KxBZAL=6*
z^5PWjjQIl%UNqVLM63~uoqd*#HqzX!u%vX^Uzu6U;86bU#FoN<Czl0~e49lo?e9pH
zU&xdzQ(<R>_Ko9Y_4oSNV-wHdak)+<4j&F<Ql4cc^=vu5jXq*2QPNx|Vj~*65N6()
zk?(dw1)#7^a%abYpTbCH;pjfW=d?-+L##GoaFAx3kL-n|U0n?aKqR98UU?I$g2L5R
zYEIsS5HAbd{IbCME;QMEcz|X+1flKN0KL+1v;PGl?*<su6Srp<w!k=bT(2f+N*k8n
zn;K{fsoK#~pp6}=L!Z38c>X4}Li|p0v-}<P)cn-ZS8I16WE1s%k-T?R`BQM_N=EsY
z3H!emK>usE`}%Z5E-ShCV`oZuV`BL*Di|t~t*681u99gRsEvmMtD<F_7RnMCq&t%B
zSw=^<m1$<O+-uh;uyN<UXx_uAqNY2GYY~b==_3~emQN(^8e*fl8eTrMR=$MqL$EJ{
z-GB_2QrpcVtG&1Z8vPRCyl^B)hAR{`H7#SSM{rr%iR)X5K<M>O^aB5VPGFo060G!<
zSj}5HcJ{SJ+hQ)jCeA+&)HC#TQ!~Mt*Y_!8VwDnAm_AJ?`f!rj)%&?mJxRv*+1ogx
zst{Y_CmVjB5r_ZPY<aejWX&w3lAy0YYM2J>0_^#xbml#{J???!Xjl#I;X9=TTodjI
zwp+ZM2zlRc>0sJZB$oGMNt#~aT0Rne$THMAmJ!rq&EA=tqr9Ks;IK`o*DuhLNKsx~
zTQA@XEoZ|XHLFW_cC}AQ-Cs0X`3%UO>T(NhJM;TKe}wyLqQ^xpS+`X5dr;}Y_lBKg
zyS_&#OvBiSE9w^*^N3Xvq+p32P!*nzvd&~mLjWDZtmPvJme4hMHC{I{7JRy0Jx+b@
z6MN_}|4$YF@5W~v4>edW!SgO^)Og_A*8x0D2sm#B2yA}LI409%af~0_5@ybnMh=RG
zQ^TdxC6@)o1v1%883}zSlC}u4|LrL*nB{WXYw4p=DoI-*&4LBQiqoaj|Bm-}sT~SW
zbA=NC$ILa68yGH|A5sE>Q=LBhP5aBl(MMsA%~of^^$U0RuQ%$bXa2`zcW7ew9zXn{
zZJwWOQ%|?jwhDHqq|PmddjIGLR!30MF^Neth2Xva^}Q8pq9GWDP0-6!6a<$~(TRS@
zu{M$1KrK@eu>2je@9({)akak^({qZr`iWCSVEx6jS3N_=z<?VD-JN0V>Yf|w3g)7{
zVHddgOE{}IiHjQLikgvX%BCBKoYY`S;U#2Q?L}a;-@k_Nu}pEwy888ing#c1st*+P
zfwUfx4L1rR-rR~PXB{@iYG~6fl!TqeV<kk9|MXphOtz1L{O_@r<z<M)-%V~6$7CkW
zZMY%4oM+tE!{S$F;hb0aTh57^DVu2<L_@#p#NUe6@(OxNzee#j2>`H}#6ir*Mf_}?
z*Z{~*QtmW{diKv+grISVVDmCHI`Aj|X3PCXHhXl({a14*SI+fg{TcW3Q!wNIs^`~B
z@Z{Atf#7#dSr%`R_zSr=KDjr@%|tlgV>8o4yP95)Q_fTO*pzes#IFePl7$%}gQ)3Z
zVZ#!A4)a-K+zIy{<u`_y--OM)mn8<holy9=RvD~GXwI5h1H0RBC{CpS!kx@y)KhdP
z%>j#pEt!3Cw~1?CBi#fT@ncuczQswEA>G|~{Y@nn@ru>N>WZps>cG$;6^f^KBUDvY
zyN9oZB~Hbc<<KIqF|GJ)9DXQKUE{M{aP`LDF6gy+zTILA2wsF<z~y`{9$$R|#?Vnu
zYK+n_Dr*kv$@&8%kICLiJ-fa-hhpcw|BjAa(elts(uhNaS0rpcKmSnBhmPZ3%^`>K
zcC=0K7YZ9PRtY|FpQm@&%{Hz?)4LK=OZV7OZO_gyd|XYf#ZpP)HvyM?qs(wflDbN;
zzsZG5?;D)(Q6`oMep8{AM-E@q44bN<Xx2zPJY12e5~s^4-GU>}DUc}JD|*aU=YHD%
zlG1v^J98H76Nty@gNVZv+ERdxd-{P}9d8?v=S*^@&y^4YFX=@WU$7J`Nv2c+1MmOU
z))jHVXQV3(QL=T@<0gq+Tw5|rFOqha#FmB8!dYVv-IqPrV*&!crEU*CznTyKukOcK
zq7Xej;GI0FbJS<8DWPv<>o3pnwN&ROJz9ba<a2eVwzrK3$6<O35pOdqtF>njuG3EJ
z@@Ca5EHyvYFQYIazy)iV(kH7-x8#iiHOMrq1M<4T=n+U;F_KDqX7&-IA6PCgfZ+1q
z%u&%<S?v0!hzi9r=ap@q`_op3=MJ>NRC<d;TaO#vhhtXFwZfY)qsCRQra^EFrZelN
zu-F|Uh;!#8nf14T3QyOOv+3N<)I^<HCc_whP)Xj(Wq3jy?%4Gxujg#$!iI;Kqt9$#
z`~A4%cK=8sTSfWQmnIaroZ4aLtIWJ%5_BZs&7HZ=RbtLI8rUa(EVW8A$#RYQ=4kg7
z3!NH%P#+t$TB(j3Pt3w+3*oxKHi@6^h(h`f$0lW=%!t|Xy$k@tHJ|rp>t6nlo8rc1
zf^&=hRKJepngp!DRaB|WjaV>ni`zcDIg~C<7@XaQt+5yRQOY@gbJ@B8ohv>};e)>I
zHWdxBIm}{C0xl}(<9@cv-l_~cD-dY<?i%a%D3qFghg*!9wYzDUD*PVUYk$k8p%iAZ
zld$nJm5S{4J0b{a=|{Gim9ne5va8pN3hWX+6zb5&Mf~Neo=SWM;7QbqbGcU~+C$WJ
zft>UDL<I2ON6$Duj+Ubxt))Sx?Pa1tr)|yRz!JTnI`@GHd(QSb?oQ?U21}@Ugg<i=
z82zVJ|9`I-q9rT5C&`HIqg$4OJC}&AHFESWOqy`i7X^}c#zxp>?8~~D9{`$V>Wb!5
zUI7=rFk)x1W_r!gjhj*v0aXZ`iP{fTUJBk!HfJ-}_%Sb~IejzpxEb`Gv4`I!Z#~$-
zdS@T|M{yf)e*^mT42ZTtJ=WFA>IL;HigeByv>vO<R%cWlbC`oE;Xem_9lg*veeSwy
zsaWMb(w4c+n!+%stYv>`IK6BOnXHe-Gil!n+~*3jHM(6hMmpyn41fQFwm2jD)geUe
zwy|kS&>2giOlAgT{k63}hCw}Z)9bxN+SJh>weumTg|*u&$m=RrT}#V=l_-?N?>^Np
zV$$cPF1ncc25i<(l*fT6t56PiisnW>{e`_Q!^Ve5B2G;xtPxS?u$5{)(Gx^dAo)w*
zeVrwQD5SYG59Hcj&_g%N1`~z`>E}s)k#}CH#}H=iGsvYj0`M*GcHC3iP!i$XR?QdO
zygug-3{ElfCirCL>2phtwqh>`Ew3T;a+&ch(~Zj-`kdK>w&@&HVW{(TC!2RZv`jj$
zAic1y%!gTIhuW#5KMdoJD#W6Y55<;WJPaU5nVl{Nn|xEDc_&e2V;vHj%dX%ylBYxM
zv2f3sb5Gt^ry3tSRIrmY;v~+W6=%h)Z#zn32`o?a!WiIH6&vw?;p`fonW1s?URk;8
zp?siTznO2E`9dcka5NBw;x*^`f4lGhM>fFxLP3=OVS+>r^YET&&n!Mz{T;7cGbj;L
zHrx%gdbkZoSx8~osuM=t%@Ts{6X^!R4^E;fCPC#d&)+-G$~X>m!FBc80jQ#5#-9pz
zzX+6@R(yK=jgSI9DV~m=%1=`D-&4#?vva-e6<p3D#+i?t0x>6HD;47<ee@jdyM=4#
z`g}hHoT=g}>i4MjTw(5Q@uJajIk{7>#A>=56T7iGxF#((+30uV^L)K+4;oA5pIW~7
zqhn%fhtH9fpcCmwqw9fS(&>aD;E|LI9TEG>y=x_=kWjB{v^05j<(HD8z0~C|*3t5Q
zRI|MDtau?yjv0&2^KJQ=Sa;4#0OjU8$~z9x$3s=<>r<V7kCo_aNC(vCE+1IwJdp>)
zqQ)l<m?Idprc0s8$rFOx@}^tk=dg*xKnkzz^$et1D(iEiis?xqFq@&`M1v#ZW+org
z;HtbiGH;Lj`NP@zp8VE>6VIe;w1KmNwh(4TRp;k0HWwPXTt;#9Un2vo7;YY(?nH-@
zRQ^idh~XxNi3jP#wo|g2hqy)-c82Rc+M*zYk8q8~0OguCDGs=aoVM=H37Xdz$O~ED
zOH#70B21B@ypK_JKDgk?E^dZ&u2?e%$&*TV=B5g1OAjA|&*V;wf6}t#d9))v&U+*j
z!4Ac$ni}Vr@$nBZsEw?D6L>nbp~3bVNtiglCy)cS`m|mitDaCPUoIRk=bO5+UhmJJ
z*w=6N1#l4E(Vi*yBK`hL^1T7TB;U0demC0>gEpCggFdMxQ_@frn)96yY6eu4tbAg0
z5>=k3J+I6VA1=u^%f#JV3IUKXe7#u8V9!Pxb~Y{MVWrS<gj9ijuXbSXcH4UkU;AN*
z=p}kS+mIf>|7$82|LW^L9$Uu_dwEVv<bm*+Q?booNkc6nJ|W1j(MYmmsuG}@C^7N1
zhuS#5#3H>7XZ87%S<$hT&3*MW<&|RBMTXDNBPErMr$x)w)TQNn&%P}52fWUQKL}ro
zVlS<Gy|kQ`Nkh%SPrc!ni6%uXAKr#8blt~1^}k*vW;UO4kdA+@kmR7^<ayZ=d)PQU
z$83JSY4bcbSRW|S^uRu0u0=bXw!8H__wZxH0^U(|-9>f%euZ+r(352t_**Co2rN$(
zUaR#=5vQG@TLZv8i~}+vZt6d;0ws}q9m2GEi<TE9<ilI-$v2r1kn7zKN~xl0OKZgy
zbLf)p0%*0ld0fc#*S7*TX$*2aqm$<_9laKe>UWT-w-zG=Bb6@uA(OTDC6hHSS8pmA
zJykM4t#V@f88x#7cVqcDbbpFL)C>r8aD2~8qE7g3W}1J2(<HCPCD0jRtOHZ*ns#IS
z+;7);ri39DKXuan;x0A2aK+YTLqzU}BeMP}!;sQ2bICzH-}E$pg|`Wq(ld#v^XZ=R
ziud)dSn#ozSBWuEKV($`fAHx!=RxdX+#SXdvAR2fi1r-LPO6_V*9w>Yh<@<7Zt~9<
zqV$fYeA%Mh^?xz&PuwdUc(~gJQ^A@(|G&R%<eu`2hE;h_I%=WfM-H1%>e&}ecD+qz
ziKDS)B#w;jt?FZ8_$cT{$@zhrKYxj2Dlgw5sDBR|5FmUCkKkWqD<nSI(KL9J$KoS)
zFt(W5*cREieWT7sj0tM|>XHIJvvMy{7FAe^V>SGC6%(96s}j0k^8vTM?C6uR?d_3R
zH!2o2L1|j;<!XLCB=x5~wuF+$U#d1pm*gjg`sGMPBF?J~zZO{PA#1bWUJl^<K*|Yx
zMdk#0W9Zpmwt&S(9v;$)uOk1BVlSeC$yHIwV<(P8U-mcdnBO<w#{6P>!;D=?xLQB&
zydT1gKD1)kdc~IW@H-4<5PH1ZynSLm>wJ`Wnv$YY_v_#8h4X)|dS>!i^+NK$Eq)P&
zxlX<JvI8Bl1rWXI2{i7BIlHA(6x62<#(5j5mjMZ*^L}9GcQN9Xzcn3BH-!y|hBOvT
zg5%}1NQV{q5<m*ST+*{82T{y!h0b-`Rb1RbR)@%I>ok;)4MQWma>ZoT0LmKYEmGSI
z+Pu%bwMLMO4wC}iFRn$4L!Rl|C|w-;$u3nd?+RuCwp?aGd&2KO)z_iWhOqJ?+UleU
z-w0{USpa#+MO}i$pg|(<y|3A<Oqz#FG#^_|Z^BwmC!Tfv<D1$ajkUG)SMl!4?;<rG
ztZC}qk>UFqj-ST-|AIZAMmRrR;g<JB=sRhGdPY`MW^CA<?l!a9ckC%QTOeJdVSYQW
zdxyo8`7C^xQ6(;ojZOCpulo?Mvn9w~@#_I4%yxd6nylnm_#yFs-G0hS>RKJFz7q7}
z2RHUA6>m!9ky{XWY0*T>&|W@abH?Ed4Oq}i@LIGA!&;}==d?Zwdb@X>V080O$=I-=
zwi<4R-MoD=jfrW^BDtk>W2{Jv>r0k@oq9?hV_jA2!R_Gy=(QGRImfikAdr6|F2C<L
zM&-2iH)N}6S@UQFc*_!nEGnmji0SC}uPocS=iQ`epeY0WeHZO%a^tNSEK&I_a<j$F
zC99mB5Z6@aL35tW&4@rxr*uZnj_#IT9=P%qTkRCnjt8b~_VaIy#@H8gLzL#eJERNq
zRZg&<{H?gGEQ-#%5Kg#ICRCfRpZJ^jJ)a^tp`l^xy*V)CTtL2@gA^zr`DgHEmZKjY
z?C5~SkM35)MiNy;g7V+P`e?odVoJz|bF~P)LnU#N55HvJ@fFC5pWxlURZ#%4wX(px
zEZ4Xn$<<U)NnJ5MV*e~d)zJ&&$}45Di;w@2loyT?W5xtJJF=^Z`)rZV(CG-n&W-jy
zk=v$9s`&pUSKV8P=*>*@pxiuKX6REG45}br>Fjs#5Sxt{M2RQ{CMA5QMF!%+h*O0n
zz-6q}Fz;)Rcu=78wSWKs|4~T@>3-WBjDsqnuG9jZ5N<xbqYj?o6IQNA8)+?WzP+#~
z$D+3vj}%ddL(^E0!^wH<XPmYi@x=lTk6!uRRFPY3T=^|$VM3wM_Df|)cziEoE|Sk|
z>>SYU=PHOI*e~|=1{Sa);^_R|fN9eM7-AHAMR>UaKLf;|lm6-6=PhLux9c1OWjO!u
zPQ@+&60ew|QIuk$TNYZikX%G(a*NDXK@RS<m0koEj}`}?GU(~ycyjv@YZ4&nWpA@$
zO=|9X=||n&zZIp&Fo}yHQ-Z`or!vk81jXbUjcPkvdiW-F4FMB)CcdFCD0K$Xx>Iyw
z@y+G$9V_PYeGEsjo<l6b%G-CXStu2<-?<5`ZYtN7za1NY1DG>C_*$fV${gg4baEDJ
zGnYE<#d8xfFP&h)@Q?6Vt;5=(D?wg@>GMA(p9zUL#3rFXEVsX`QB%Vl8kU76APj)$
z!=dWsTyc&mU^RE90V&(PA;Kve@{zQ(w6_K4N~m5K#T6>`^((*U)d(l_+5ZaZzWcEY
zTo7*>6jz((bZg`Rt@7J9xmfUfV)Wc!g1k(Pynfy;;$k8sV?wpMED-0*z)wt=QoJBl
z23bwbl9R~XsRw=HqsH_BWvLU8_z~mEp(IaR+lS{Ba3EHKSm271X&sbpPO^8VIe}vq
zR;ixUrj|B8Wbx}7d}ANa+oB*je(6J9ZrQxWRjv!~7?BHyQbm1JG%4{<j+4<tl?P{W
z3+W_JjoOm_&Uefq7b1I$*k%3<4$}_(bWG~XK5%#Nj2_1Rv_(4m1@;lRQ_!<Tu<9D?
zZdNw=5XoV!O`O&gFvG(|)R}^mH@)uykdW6OZ#<kf)^2AHfA4rVEEM85=y~GH+`WhG
z6WI1AqQn@QUp<6i(O1{=jw*=G>kub&3;aZS$S0G6AhEc6XqCP1_|(6#z43H1>9mmL
z(O&xhc+_Av*cNrTZXr1}W#n4rV2;a}K7959+-6WdoTF-in%X`>k0csBy|%oXvHgo7
z=_rpJDB}Dv8tI)2dH_>+u%-TBCNtUt7&zK_fXqn9f+>)C-vG|q`|bM9f^V-b=Lzn1
zt!p5+PkFT5;#q%U^a)OQhDW|lsVEi2mrLLLgC}k&ja0~cZFy<SIGF$nFjE=2YJae+
zVu5D(D9S*KD(@mQ@t26kNXZbpamqb?vtJJjfDEhXc=3v*opQ?r9TV3o4x`r5IaZyx
zgog@3Cb(|mFlSs_M`TUZTDQ?N6_6tbwKA{Yme>m$WHJXL_p%MHX>OO?gx|9m?eJI|
z7kid_P3yYS@sBQBsPeT{?Gb=Y(*kuVkZ^9be+Qv_1PqdfU5F@^-1f^;NZ@NYfVp8<
zK7@Q*C`23w9;Vfnmgdc!;^s0L$)C~1Q0-V?&v@TaS3-yc+fH3=YI|-wAn=Um2cMvb
zj=qqLyCb|P!%jZx@ORoba^>fCypHs17UFfz;dyX-Uh-J=2x8IPA+kbIPy&AR_zM02
z%6uc)nl^=*??7wZ_x{3b_i?eX1x?L|5G72sNXhcF?x!Nn&Cb>^`tj6cDf+3ZMJEi4
za<jW9NLTTBe|4OOq!1pYThLCebQqq9=n6%7UQ42jxR39;+aVYJ8f|h178{x$6e@Ir
zA4gObh4TG#*b}|1bv+AWo~*o<LjJwU|6`4T*(->v5dJYU76*q(2iCzaaeVT@D0Zzy
z{nQd(Oy{?m?=I8A+n-EYB3VGW$r@INX^$@H1ur43rCVD2lf@;Fi;m2yUJ`3VfrXVD
zUd^U4O!HffsbvqO$6H&_-(9>CLyOeuFa#s^{h}P}mEcO(aD$r?%hwUb9F{6rKe2Y$
zIe7pP*I+kz{QbPH-48r8i4PRHZXwz@>X|<qOo8k_Y(wed155iO;*8dIPtFK+9;6U=
zn1u-o+yFbBvhujb8V5(@%YXQpy#1YA4YD%M9or>MG17k5)bLvhmHYz16pw4cS~dvM
z%PN-Z(JsZEq?y}D#(TfC*Hg-;cYasr&lVLFf7yY^CY;f{e+ld3k}k<B8z{GA`b90i
zqS>=)>1}z5_V^&-c<O4JmL-fL2jhE!wxWyJT%Xjok*NKM)d>-tSw9<EvzT2L%BTmP
z!!*~aN%G?r0<cKKV9h%t8r3aE`Y=a>{oc9bXlNu}2knWfKAl4>)2|C!$_*@(UgRxB
z3!8W|Y3@VXWNAd4AFg*X{DaU-C6(+#&+Pm;<i#xff|j>p;?04Aj=d2rGMG0)ZUz|p
z^t!Crr4!pRzgseL=5DyNOHX4Rc5en7g#<Rvec3Tf#0LWR<`83wz?6XXyU-@=A)Kx&
z6TkRbtL%Ii#61L`t$4XRvDfq0jGPxnF(X^K4%GkNJ2(ezBcKMtejiL@zUPUfH>peJ
zzv0A37AZ>8)@1`BU%iBLMc8m~QWr@7%w?IjjF|iD<+&%mQ>~x4FPM!ik4st5uRxh`
z&qi)r=WLUT_AIDmmC>|5yZOgY!s$pw6dF-izav=tCHP#i+#?y9O3I{a`WxShO-qZ1
zB&NpuD=%yk+!+T=Oi^a+MY)7FedA10@BI7kDl?QsCpkIAtP~~6&7v4yUv^0AQ+1tE
z{%+x0NyrDR8BF_*cRYyXw4^?t+!megI|bBmk?0t<R~tKm6~Tv-s`qtb7kxmd*B~mk
zUf3=DMQmV5>_+(MsOs7OKu6+EmMLyjf_Pg9jTmC~#t4wPKx20&Pd6x-XPiMZo@Qsa
zT|Te{yV)bWohdsu_Idjr?$C=)J@_lVjebA_vH|r@<SCRk1#8l|zA1{7<6^0}D6?~)
zNmXoYW!*hcy;1R`se~o!4-(+3XXG?yAt^Hb<@5K?Z6T|586%OI)23);z(83tc7l1V
zxK6oEi_^DEY#rLe4OtY3+Hu24zA=H`DYt#riT2cbhkV$$nH3#yCis+h&#z%*pSiYg
zhw-1*tG-<lg~m<U-IEKpYETC0DeuUUJuR5r%e?pB=61p@-t2lkdL3)qfQoj$KE5yl
z??<t2|Ev6<zdgG3xjFrPUqR6o`xr-;l!)`WIpzel+zHvXrd9jyCaD+g9gTVQa-5co
z(QlSr<2H{q=Iv?=tsV<t?-8oWLmDh)+YCBH&t(Dpu}vU9$lDcCjIuv#dE`2@2ui$N
zRXB&;g+#93u&5DdKu>s!@Bd<ny<HOhmfc$gDhOm7BgQBuwmp<Zag8dwJH>4M%{(Zz
z4d~w&Y-E`#?MHb_W5Gq92oY3p3~4aR0gHH0yiW0`JQ#M>%9hxAlSki6t=%~?l)LZ#
z*y_DMPAV)MW_{`7d@@)yxl>&AKYs_Xvf(3rI_-LHx?f}z8@O?JP3{WUCb~kAsEavb
zGeyEYgjWF-=&<_Onor3nRN~KQ_WS7))d{sNoEdopqtMk*&e%1?XRBJWe#D>I(5J<Y
z><u<NQ<bm6<oXID1LLTQ_(l}8{Qgm#UFh$gB}eH?u1T5{-3+bN^f^a53%e8Ywg2_>
z#+EA9d^rmd1VZ#0J)0+;vkoVc4z9q|)27NtJ`D3@5C<St{#>!ak-@R?&zftAyj`sU
zA6+5#YM%%<gEr@G08D=qdfH+)#CHAa2*o+?4YuMmum>XZhi-7-6S%JK6~(S}a9$6^
zF8UL}w80iaSyz*<_hJx{a~~u$*A7SeHqQUHwGzaLu}nCQ=zZP!F;o)CZOJb|NP_F|
zEuC2v<{|8ilP^?L2)-7vsEj9s0S;o)Ca!1{s)A5`d=9rzBDiAt#uTD5>xe2pmDAtw
zA1o^`e+ro{+MNk;Z!g+CYif^J4<HyV-45YZy@b)7Uw2Gi>ij3Go>H-94TO|DDiydQ
ztbdIGbZQzJlv4=c117Jwd1ADQwjPsY2HiYTKlAnw3TiJ5AtqO1&oGX$X?y<qHLkhP
zRD-50q-d1=e8jsrIzLvSBS`LEAsa7TX{ZNjfn-B|FU>wZbv-TspRYClF>*P*KHow+
zD4#aVYGt~77T(>NMNv6?>LW%I=s=6%x2`D393Tx+aJ4I}i4RNOjFwU54J9)uc&08Y
z|5Yw49HkjZ7c1Wzjwgrpvw&`P41E`uwn@c%#O3Y)_1h<}2UnAyuy@^k+j6=q;`4g@
ziUPZr-;X_Y=RCf79#=&F%9u2sSt1yWEgBVt*6%E@r)_6v&qaadq$Ptr2m_m@DQQ>`
zQ+GTpcHjDBR)m{JxREA;o2@qBT)caoT)VSw*Fvt3%~3t~3K5J+CI2*#n$ZC~okE1q
znzIsNG?RS)y(-t+wcE{1DbN6(DiFr7IFS0=Gr_uwBluq~uBE&D?WYvKggVMciW#UC
zN*-d(`PN08%4W%%y+;6jb#^aG05&dXByu_mK#rfBmjs<T3Qm=I8K!&1sgXqRMjkfz
zZY`$EGw5SCq9dIMIdJVyBsTH3&+MXgxu)CdEG+7-Y->_J^Smy+zOecnh?#7Ka1JTW
z-#A-reKq!YGXAy%o3z0`3$*lZ153ipGM3KHHS>uQaVdhIs-?(>l9q_`5iEb0M@CLH
z0a#Jmku?!ch)1qU`h4~jC;w<OfpS;N1oHj4$uYy6Q%bALt4^J6e(Oyb0mmkv+j91n
z_gJkpw-vx$7eQjuWlKH-NKbg@UX&v>cuDt3<uk~9+JJgyZmke5%o2=V!OXe?U~y&v
z6})q?Ox+4KvQ#~HGz0A@&wvhLE_zG~SiE|%o=-%MTiY<1HQXV}Y7(dl*$fuwILCj)
zDz9yC9Tr}H{Pwc&x`TALnIkrAUo~l6X$;xEf%@#NCSLC(p4cgns`vIvVg;F-dw0*z
zYgdqk(4x>7dStVWA#W3oy248Yqxwd|8UuxY1V=f4=%mEVrh2ZCot3hKqkEA5qHe6E
zMlGXhphQrDw6!6E&R6h5;LOdTmr-!t>y!kL;Ah3Y*z*-9ceax8C286J)NsrHy7tQY
z?%652v>}|X4<4fZ5$6voEL|MY_#c@z-XYL@Oqz`@imZWIM~Bb_z7n6DsC`E0O>doR
z$H8k@E*Pv8mPfI#5IWULUz@2GCg-MP=O)V9f8e#p^dGvN>4H)f=LinPCt5M1j>w6*
zrThd06iY^50=|9<Fb~6W+F$Ss^LnIjT|-RetT0y^==F)3^L<1Cg4fJ)rcj4@;f$!F
zEqO@rXnvKV<|`)XZu%g%$7Qiuy$t(J7&ZSMRLQBlA~qPi)?(JFlnN(u|Kd<iSm8=+
zzO;O@^Y|8ZpzC!Y2jckLztLmMeAaPmeAW)d?7S<KGnux%1XFDPc@!O9+IX;kD&_e8
z^)3w*^j3#BQ1*Afg*>%_5WCPA)C~eb*~VtnTKgh8K0HK=OaJ4T0l!u|b)j>}@Mqd_
z{nCiQ(zt5Pc_mJ&-|fPoh<Avl2L)vN2VQXQUI!&mf+Sle1Fnz{GU?#@GNYy9xAC|b
z?`X=E=5la>3c9yDkJf90jc>5s7ZO1|q>5zjz%Hg6{~QGucK*o3Ni|NM7pOt>dk9CE
zsLJ{2fI0=@jUHV@PamUW3Emw%x|0zpC2L8Q{F!9oTq38HOpNY-ALgIrf64obF2WDI
z-%SKQp18JOVqkUTE%5&5yVeo?M~!;T-+W48aGxQq^xZ29uHa|f|6t;?TUP%5fTyeZ
z`_dvkMXb$ZLM{i%$q7$ZfhP|YrquS=@H#?sj3qwv=Gh^yNC)0L7MB!#<#_>gbEEi=
z%qk!G@-l<&by}|563nqW-u2;HWwQxBuJX^-OH|QEWW2=!`eddg8a~CS&aA%hRSD01
z);~$Gq+LD@_I<ob<P-WRGeI|YAQ+KZ1g|CG@0w9QouALjV$0P^727R?x`8GrMjY3)
z-*WwgfLp>vJUE|c(qUzKz9{80e$(|+yS$QRXmFX|`2>1^2fn!6cs3HfcsZw5`Lqtr
zzP%`v>w1cGm^CO|d~JPX8l;^#^K8G|dem(bGPSZLYXHj4*l?G!QE>hg7L&{W@zY*G
z8{Z-yj+!3CvMsA*)fqMw+Cw*lMaw^gr=3E;KFRnOFBtk`A4eze5%5Hj0tNe_4lppD
zV*s3o;QY)%{&ph_^k6ZU*qUnc`)m&$Jx^g1L6JP137jE&+5Dtqm@!Qy?9NP@H}()M
z%APLCPu|fcz;oH^;Qme{w>^#xDV~hQ$f<mVp7h-5?k2xr{-;`fLH{G7`Eyj$lkNHQ
zczR|PbeXKGjeE0;&m&g<$NDY8SQ(my-hkAt*zdNk9pSqRIlmyb@w>$#(Lu8J-$bek
z-roIBpyA)w-!*razyCDfYI4b}F}9>CHrnYyZuvS22bnP5K?56^*O0g@Yzb-@ZL$t+
zKLNvHDSA>x=XI~x3QIw|>pLcU^f1u<mPPQD2>(zzr#*<}@PZ9l^R4Ju7DfMr72x*L
zo)lUPxaeNX1iA#<q~yZ`Ul$1m!6s(y&+x8?`I$oM)Cfw9=IP1<oqLw86<(#FI0$WI
zp0p6tRZ@j(YRk)<m7wz`v89&!NBX7uZlV-`wuP3=vV-&fwO6EJ(3LYLgRc174b!p4
zvoA<*d!JuA)2fIxB(a%%s<b`3)Wb_}JCj}iVD&y=I@50Jh+VS+FS>)s1${D`y8hZf
zC2kfoM_y^qJYYzvk-6r0K1vzB{z0g$aCCj@_glZ-^lM~5m^56bzm#O5U0hz;y1I}0
z*eJ=OfX{g6_kHac>ni`dO0i<`(%{GR!jvJh6t+$RYZR4x@qt07_X4l{v0dH)hZ8oH
zqAz?_9L5DWa>PhYVxDpH?~rWRS0^2=u!|=jbdh2{M~(D5`I-mycp`0hNULXXp;yE8
zGMQc8QQxe2Fj`e*GhX;gicvWPe^g1Z3|+ickTqAy7Lq0QGGz(=&VbA1p0PO=+V;bd
z9B_$5pwhKSX33sZ2$RZsA7cVt9nTF=Bt$l=ekUrscD@FQb?<=11_u-&ch^>8eH*O)
zxC9|p(yh0Z*8gdo{im?=_jdJ*;x`K%73v4<9JR2vKW?p9^OISCMF)-ye~*HQm~AU)
zO7K-sQPF$xJS*S@5XxBVxrr%Bp%mN3Xht^MGGG%ikBCa7!in6OVXkh$p%-J&>?2ER
z(`4;qd{x{sJfm*`LAkc?CyyU<6NIcoWB~HnPau4m;B_Fh10i}|sg7bW5heMiy7iNX
zZNDa1Z1NaPSvadlrOZ3SQcMnx(~`hN;ddVp3LPIK`<M+5C*9q7E8{EB=|>atbN0(T
z3aesdspR?niEq9dXv*P(4S&X6HrOe!l!ZAmp?sLR4A+)@LU3>=uwtAdIB4T(xC*M{
zzdI!8^#nkRLNjdp_WbXnKn~;VY50UmE;c+w=@l?y0eYGeDL?&OUWHZNIL}f$NL5ZN
zy;Uy-)|7r%0NG&<DK%~7r8@>+`he{ep&dh5yi?eoICt+GnQ$bXDss4{Z=J2T;9bxC
zDI%0`8hb%Q=^M0oVY*}-wSEqf;#0_EyIv}}9*c*L`9QENP_=BS|CVnC`oqA^b@E*k
zbkCKD7_&(sA*~;btZ!dVnXn&AKWT+9Fq3QPW3+X}tX`%llJxSj^*G2fDm3H-EJ50*
zqPq44GMnTX{JpB;9poleO;cj{*xHK{c8&z2Mv%+G`+ZNsrG~H29^Re-s&spFy?Wm_
zbzRhD3VOuC7<-uHK6}!j|Mb9K8vd;gY(t<1>3<G0yoV!U-lkCvJcLl;8%AQ`{n5gN
zkU4_=)RgJLP3DlnfL|Z{+kbLMyJHsZ;2e98j?9x@wQt8FfTyS8hzMrghCA;uXYpB%
zts7a-tYe0wM4xR`j05VYGClax-_A5lH%hieYL~@OfG^3|=O6Lr=f5r<=@j6hId=!J
z<)=B+IM(zuRFt95H|jFGT+s}ChV2>Mx8yp=&7PK73YEEQ`SqbSbHwV7Ii0KKRKYN|
zZa~iC!Ykr;Rln`TryDu{45Epw0iU3+Hq`J6tEQC#lfBzC&iFehU_Rv^N{n_6j#MrZ
z*$@~HYr<+#Rbe|0?EBPoxB1wW8j`GqNxt|RiFvag!)|W2*tANI47}2~Ty4OoHy{U+
zS9df{Ozg{MIhtQ!H}DtZAJlS46Tp#H@UwcO9%``U4MSS&mD}>eC39qFwjL(pWZ^pT
zm1IRHV^*Ziz{a#RS3RJ#6Xc>qTvl1nnISyUolE4UFY3xe<=5%ul@oP~P=$2d+JrJ^
zrA6J{hxw%ED5Qf*j@tLK(q@6tFo(8$Ap#P!po>$;=iQ!Ps#FB*hq;EnN>?srE)en+
z;gzRFlULg}?NFohUEz(#tkthi@F3skJX;944ZmemSKrM%p<Z@w7J~wAdJKai+F0{2
z><+0RU;gXh8h56AQ{?=GL-ZQPKJoUADbuOy$P0S@&spN{|Bs{gMSt^Yp27W%Bu`Dk
zh<x2zPVH*MqW=Y#ObSh57zb!ngBDIEooJJM6sne8`XMOzCyV0m;VnfOJ;H)S=EM5U
z8{!GD<X73ysg=(TopR*2*!cJ|0S(x9fddY(yb)ioV0rvzDtxw_(2SwMhMBE^d8QzP
zW=sXV0+n9&Ex^Y!qMz3;+O^VfF_%Cw_C!ocPm768R`&{|1{!I-AvRf}C!ne*1QjPW
ze;a!~Ryy&W-03*?r9Ah2eMAZTe(QhXeOC$gCGvlmd?Fc499d8ezS1?+wb<x<E@!KX
zqeIEQR+#<g$ll1DBQu}YrwyhXV+qaY?ra}~>4RRK5bv&>zA((o0XgFqnlD)0dg(JW
zcHV5)q;}}8l}0Vn#EiOSm%ZPqE}ROfBP6lm@;}IC=<^N>K2eD`AS8c)myn;ai{q2b
zp4`-JGwrNkx`t~L_N^@k)6$_!6$ID1*kN3e0=Kq;#aYm_#7jOXteC%!0;62HD|Dyq
zJmojyQ&M6m#tRiQ3iNAMBLSXn+0LI^JWp{0&dsZt7$bb-Oxanpi`<=J`va13#cZ&t
zj@WNgqpvxWbIR-Y^C_PxL?7b8YAN)T_fw7+2Iu$eFOU9@j`iu|8y>2T?HXj`^EZV5
zAeem3v-p&!8)+19Z%$V{T>n-k^+SP~b^SG}#cO%Bk6t@u>%SeJ9;#dpuPoQ^E?9M>
zMgCK63jI@U`d2vnDJBY;+qQvAKEQ`;A1%ALUZQrM<18Zs6?Q)m+nrM-(>d@VClZUH
z;uv@MvXV2|M<gP;ct!B2+VGjfGLAO>y^>OKg@OAo46{AhUDNbu{NaX@*_JSx!XHE@
zCe?tCyMvG;{hX*&0g`B!(ozQAYl<k0o}*7ezGEortm0OLJ8a*qzW-(O5Os>&KZ48X
zoxZ_k(9Tzs!)aJdk0n;sOuCAA2j**N?x^z!8+uRpreQ6z-YrA<Wf!Rn%!G<9`LiHz
zm?6CC_K%w54WpTS@zQ5Y=MdkrN9(2)UGt@-CF(JaTDx9Ppkz31SWg;iff*nZ`{80~
z(l_$lbHk&?f1|VNu<>47N+GHW9ixOvsxM&97roZK6r?#!UQfTE*@_0$h0!21Bl+@2
z(K0K+HkogB6l&cf*B0r*af=e8KbI-Y*w%`;_eYq!*h}WG<fdZ|BWk+D;|_33;C~W9
zK#t}&#fUEjR4?xg=hC;HB5zcUu(9;a|EX#Gu=gMf66Lu36oHHD*ctUF!HQCB+`>MC
zo{{)4sp$FwP9hDgXIBq%9z>T6gOn5G;StpuA|DjvY9<1c(HZbwhQj<}fSk?~X64k;
z8Gu`t<7XlAI6{QJk?_A2?v+(WR(zz?$rgTo*#MRB9fk^RG6)<zYN1bcq)M@ZdXK@z
z6{g><hhLQKLY2R=&RP@ce<_Cz|E8mPAo4M}E-C#)N|DSu{FTlt(~0<nol8GvzcgK>
zqp)17kHY%Jakz)wq*7YRN_Cku?A)!LHR|~hxRLL=J@sdf+^n1MA`g?AS7g3L7OWK6
zBP@UUDkrE@8PihV*fY2v-j3Z5nnFd$5>%)<<=s%RZp?r;`_7*pV|yGOsu`=Sp}+QB
zDy+4>)wTI^nE3qYQg+nhxX8(;u)h^=HBX&7^VnR<v44+O#2s-bUS8YAZc$*%55)I{
zc^zyCtin$&w$gVzyo;nDi$Ppd2~#PSYML#CAwHw2CtU-n-cX^_rm2`!E#mGB^Nx9b
zbU9n~3hjG~290V;m=WXpL@l4xOO#OW!q<n0lzI#84dD$AfZ-u*fu#A2%tgPrWX&u3
z^r8r;l6!C8kvdDK!82987l$6g@1W}hXw#Yk$$3Z%gJTevOux^Cx^aD?qsYEavy~%X
z-Ppc3;u%x4&zp)T8ESSy_h5fkADCdW<VjZ}q|d(6UGI`?o89!`H3KrW_;_`DR!60w
zGuv0hRmn^k7Q`D!X%n{b0Rw)aOAu3MnR|F-_1jL{eK^XvN|0@uA>}Y`g#luG4jDW_
z^X!Ly#6IH-eG91JD#e<JebjwhaXn7mFk(Bi48BZ~<#Ri}3}y`C34vBVygHjr!@F2Y
zG1=l-OK+=Vg7N<`4*&Qk+4SMxwSWp4Hnexn^HB)-oD{P}fsPzZqb90ethJ!+iK{yI
z+CHIJw8QO{KDX^XzrAC|GiCwk`GG1O@-dL>?H0-tK$3}4IP25L$awda=Xp9vNnM*^
zM|+gXJ|ElwNN&nbaGx|%rA{W3@ABzQF1{?;;4E|7GlsN-p7B}2LG;yuLV8#%4mEKI
zhMV!vzBIU;x*wLEOO}sMS}t_UsCMbpQaK(SI~_pilMcZ5%xm;&fY%+#Q_|&J<hHuL
zwtazaZ@3mVYuk&|yzAw@Eg_N$8M!wO5UY?7413Hb(J6%Ig;OC{5B>>B1_b}YD9$2K
zVoZk-Fdv?cgtE?9TLZztRD`&;6h)=b`B;PB?rljEYUz<afJ$4|dYO!G4LdS#W3;1~
zB5dQ8^~5+?Qed`&;H#c{-5n+kr)70Z!HJK<vhOT$lMrmsqsGI%Ren2fPLXS2bF*U{
zOAT4DEoPC1CUWZuN$Ai|5GNH|e<1z+H^n*L!rWZ^XaJsQf~&<Rn_R9fZCi%gI37pC
znI`2#36YIlSUh+j5%Wj?V%|1h&I6Oj+KZ41MOPNpFIy`Iwm<cC!23U%c9o8Qvec2c
z0sCidXR;$??)(DFd&cPI7?ye(ga~>m@kPwA2OWw{dIChyQjaFN6zvxdK!;2P?a-P}
zBqb%~;O{;$CX#^8MV0jY|8Zc!nvMU&f7eTR0dUM0SyH5m`nz@~4%MyASQ~*AM-|e%
z)6riU&>2Ylejc!2TTi&FPfVL^Y>yU>@fnEh-?CHN`T9AC1fau-Q=?l9B(-Pqoa!jm
zqD&fPUp<8)*b4$V1zp+)42vTmB-jY2^E3e);w<3yVd5KsV@YXqYkq@%BXGgeASI^W
zw^?DqiwlFrPM=st(MJ?Xth}#bl1!CM*2r9g8h{J8k$xL#m}}JW87q>dPye|=Ku)*d
zqUH4k=@O1ou0wxiQ5SOE0+~MBYqJc-z6+4UzVi$SyMUKxa*u$(nxbpmyKi>LVNU+<
z?ZeKE#^MIY#IbUA4&`_(GMirbmiLk0DNzC>)%4});k6YKxITF~{Aj2+{)k&(LP_$g
zNY1-cp@~E@IrQag&dAvtTKJA+8O-;h{T8cMkOu~pd3PQSVLLrI#9}H;GHHn**~wxO
zftDS$pp{g5&jdx`lqksH2_~a;MR!Q_ph40czAc|R@$4b{9KsNY`$zN$V!9ho-jUnl
zQ!iu7!!Se_;FkHg@xXzKac?bufR2QrM8QOlf?&=0lNo5pcE9QItRN{S3j$Xe#VVGU
zd2UAxO<9I4jzO-jP>W+XVsG8ctukFd78bFwx|KLD2ZVH^P3dc<xLf^vlY}kcN)m!2
zxn50->!xp?lA^;Do%<)<$r0hMV(>pT=RYgsAHI@eV(SbkkS+on9=Eq!PJWy!Yi1%8
zhKgoQt#mbGHRl2e0#coBJBAJ?8o0H2jaMwZec#d3Gpdm1Hvq8IEA$;$Vj3Ene!60N
zMt)+EpJle1th`)}HMOESgb&S^V?(+XNINc{`NEf`6m};;W{cpZH)R)^WT_%VF(WS|
ztsVtHzburJ!T9YdPpbio(nB*0MLbfBdABbj+Hm~dC`bzZpkWPaBxEL6`7?xW!~f8F
zJ4CbC<S#a84V;uh9fD4ZjWyUF`(#<YdBaTK5UYM#leS0VjJuO5WgBz;$Vs@SO<Hm6
zy=&@NxSLiWSUj026A_m<r0Vo=LNZ#~nta`92d|ME@JA4YCZo4iE+=5a6)Ak@oEQ{#
z>+Hr|w4%~nn{joW2q8X+Rh!9gXzlME<r&n6P05_}v_-noZ!o-nuQ9(KC@EfoVb&K<
zhoO?GyBF&<<TJ~|MnJ>?sx-`~6*e>nX3f~S{Tjv55Xz&n;9*7%3@?~AOUE36a_gz9
z=>?1w|I!Y#ahS4<{l!2I@C$sQgUl=yJ8PlXfM)DWeM1=O=V7TyFiAo)M!U4>bzw`N
z^YikDq^l`?^Q_IAm2->Nmt;q4*t$c2*O-4F<&2+LNFs?0oXjsR&Da=+iU6xR<vyJY
zCE@cWhHEmani}@}aJ3S2-`-W6Q-7WMpS1Z#V%xu<4%&YL=MzG5bOadG7r_$3a>C)4
z9pAIm2bcabn}~|h)L*PQ0tQ7G2`_@;e1y!*ePh(_$%S0t#ZX~iD9*#j!MI}PBBtBM
zEChAjonUdr5XE%-is-em408PoK<(<!xfq+q^nT~W_ELF$kpg^T%|L8`>UJ3QJV@wN
zaQl~@W3(b{@Ud3<G@3|x2?8}|wr}h>(`JB&-y3#1o@m&6xG4K(P~X2LEO8UVb4XQ&
z`ZD?Y_`1!T^T_70?v4}&Cf{umyD~$3z7#O6zR}<t<jPVZ0S~#g8P)Ph(TqSd58Xg|
zDFT#{q)9|tBdj~asN?>3;4XQ7c0FUG+_^6`b#30PhrUZ{uwA0>k6^u~rz95(FN+b+
zMyz1X%;w*rWqe#es3OGc=mZ8k!~T%YR#mQ%7CA=5sj-L9TJ9thQiZRc@Uz#{ab+pK
zn50(_Xp4Oq>_pD%VI48Hk{(5-IT#c@;whP;9PX3jwmPIArtuaKHkDG!6gR^|ewwhC
zsLweA@-e?iO!Lc)2T{?J3pyhydX4PlJaJ{$!7!D=3q@DQJK#4G^yHTd1|jDqYJ+~m
zuQymPI_oznO;;zplg@|R1f9@l%AYH?zV9w?mlf5SW%8$g!kC@J=;4XPQS^~Cq%nvR
z*sJ7E$0p&Nr3sH9?6CL#hnz25&a|fH_F0#nhEaxoQU9|t$n-M)%}Wp5S12E0(}8dm
z!|MKiWtgX_-y2bs67?3eoPc^RP#DEdh9Q%@!L6V`!XzJ5l#H%pD!gcqI|NwPw@G{;
z!TW}z(4@R-otu@vK^ga%WL`hEvhm0DP&65e&c2Q{J7vM1H{Ae7nV&Rc1h#Rwq-F~2
zbfXTI+ocbVVNm&=969Y4=*&%nq8`IB`-1H@!lyW`y<kw94Za3OUgA&A2`M3l_4?9w
z96r1NhBPOE;~{87;dyUL;1frZN*@L#L$nG<pwS~1zb$w^1GA|LC547-@hey}>#U9D
z$|$WIMc@4Gdn2=lJ!9_GB3HHIE2Hi522Sn^+YbU{JxT*A`iaC<@8&)OEc%l#<BZv4
zP7cN0OSsjZ*_%gmiU;-LC=uL<6=<Xt!8fZseswart}{L1c0z-AfyPyexI^;HE!F&a
zOSaSLMdYh`R_)Fq&V>qng7c-BX6ZX_0X?+HG=beig36Vek%#bGU&u1Q!b)-y4tB2D
zw=u>@?cZfHHx;Txg{BF#ZF{xfv45dp2<fI$dn4(SmEsYA?=)%T;d*_be%0PjDp$AE
zeS2(z`*-o_QO<YcHasuwbDd1xQr_H`PZd3=;XCk?P+L6v$tn-Q;C8{Og&@)3&ThT&
zLD`%aQ#GvbRpKx|((Bt>tN)V(=u&xdcJ;Geuj<XGW(NdJQ#Zpgi&WmCb++b~l<zh+
z?sm7+MB_yQ`$9vttAbW-b#AoQ6q1WHXQnwiYxr6zwr|Ld@zqJkO{azu6O7xWELlox
zH>XD4B{_Z*JAaAtPaIq&pGZ=g=X5=+ql7Hu<ydm((3GK;!O!a^RfXqa6L=0&ZLWqy
zWu7|LxNP*57pBlkB%T=(QS+JoVW>cb+nWy0kT&Z+dqBj&j9Ft<a$j`Z?)+j)qBMB%
zm)@?EA@iL%`|M%xrN!{g4PlHdnw5mw<jq*=&)>YGG;%AbkB_(~^4o}#Or4Ont_<HY
zR&pVqIR9I;t_x)&s6eOa_#Y59<{kl+Wcl35amZ@Z$(G_68tvE|#sAmVd4|K)MSDDi
z=rtmU=$+_MM`Xxgj1nW!dkdn3(L1j)dhfl5(FGAbMo$PKL@@g3ZG=(6oxJaT?x%a!
z`LI8%^X2Swp1t;7YyJMp$>rfXaYA%!H`E=xkeAt?QU$<FT>G*D+}V_oHl?N>1^4+9
zUD~2rGj8nR^#Rvx>nGU=0&^@e>uus~%JzY<5}~9%r5m>P(C+q+tF54;MIfsKGZd_0
zJB-2Cb#yXxzuuzJI;A(f-Qld^Nr~ME?>o4I2E+usSdB_kk@(894jNe)6_>@QGJME&
zNU<sMXmx-Jz`reJeB@tNZ5Bi*LjAe8tZdSL%F)Zfz`#cCF(;SI&g;PCC;}Y~YTW6V
z>P>gBMv@&)Tkv|&^@eA5_D6fbHfivV_H{I_aIW{iD8cpLXrW&0Z*^KGF@%k~Jap7=
z-6%ODfov8M`MTGgZ4UK&l52fF8A(&Yhhs1T=~k;kKEJZB)nW1Qdy?kD9!RL8VW>w3
z0#+{a5G&&f`UD^~tPKp33<J;H>LcKs;h1ga_W1kz4}MTCjbE5w87BB83{$~1kRAZw
zAl_CS<PS;D1=cg;Kw(58QSs~GZ#=5opFMC?wC^>n(E4^|4RCyV-=|x<?N{;4zWe?)
zVVVdKYf$UDud--`Q;{1@@sP8XWNGMXj{fnW&+O}fjd&HIV2cWv#Gs~2r@pViHhyR%
zbqJgS$2=T|ts+Un*w}9A|FBnTL5+Y#y_q?~5?rCnTiqcE=1n?sbrb>sD2H~EovZ*k
z{1J(*S&=6AeE?vh-=5)Yh`}|kmVBKNxQs)i_D{l__>KZu6^WBxkLbyk=fAf3BPKlN
zR$}FIOgNtK!?5cFFw}iG)LumB`#J@Yr-F$7s(xfed7cw+e)noF;Dp>;;hpr@)hED-
z`7P&+=vru7XT;mgaktyW^HM}ipT)3`sKsq#Y1U{MBB{Vzw`FUCe#IQKeYmZ4sUine
znpTxlQeq(Biodt7ml<)v4Vb80K(yl|1Y0c1(UvuYT+pTWvlbIgDzMEjvQv`ixk>9h
z1^=^(7oXc`aouD2e*qc)7AWfHVvKXA`IE+>?EJM>d4@Io!%)x=L5-w<5yb#(qTUUy
z5GG@U(;YG?&!#8bB5l-{$<ZE~t1jvIpaJuZuI6hTr<Git=cRaA67-&Ijo4J{Y{43E
z)t)$Iu%sF+x;)(YYfIAaLt=r=%_3?d9d*e~31x2ql7hd%M_M+lwawQ|rY(<=`$Ff^
zjHcJ*iWuqJdt<!X)K6;Pop`#gLlPBL3T!2FDd;olWoF?dFT>|k_9>6buRrN=)W2YG
zel8Q%TUza2hF?x+m{A<hPJ9=D;;dJ%@tn)l`1!cStpSR|1T$8orgc_#IAiIFuBJmC
z(6U0<C~6HjUaRFXL{{4b?no3c9efN-e^cmgbHM|8%@L5O!VrdDJ0n#sRUzBHBA1f>
zTbLGw2UQz&(7R&CADp4`d`}PohadVR$)irpDjv;ldv=%rZt?pa3MV5Z_5!Z;WwEA;
z$Oi#6l?~%iEeQ_olY)+O7=X3O^UPPu4}$AGh%dxu-p0gBycv<6mBplsoJ10_aK%{j
zr1Bf*{$eh)OJg9_rk(Jtsu?(@s4g<64Q2qo+~9u~)-;U`3V&)5$DUvfDCdi%HFe-n
ze6AiEFya+<r-sma8F`#4nM0Cn?mZKT_}}$CGx2ZW*OQH1TmPHOw%i&P0Im>}{}V(i
z*Jkjjk(pR8C&hupmR}8nr~AG15t0DEX$9V-=4K0iisPDv58PKJq4)RHk6LKXItJw3
zHz1lf@KWVZEbRHxzK=(ku)6x-OaZIs(sA|M$=q~piRiF$GUv_jR(~1Hq%8f3=e*@V
zMcEcmGT-rmMVc$tpu2|`2ZsZi65|S&<f({UlMbKw<ze;%zuk_H=mhx&`>^+Yqf;EU
zF8>7hYnVBb>OT_iF<QB^PcZ+(d0Z>R^WMHtdmV#zrPDqoLwu#dc?;E@5IS0Z$J#Pn
z7GljO2-m#Q_8p$1QL)|T;S6H8Z)KxPc>Y`jPb5VxyIDu>2XBVT9pJKBi7>^?o0oQJ
zl&e~~Y#P+S5R}GnY+7yeL%z04f^I6W2`iPi;qTP&3R1qb+-H&Tg0GR8DiN$C|6>xo
z*W7t@nrSH0Fq)dE(C5+FKToFZcGA`1q$(H}!@t1=;-)QubHbeK6i&FGo#pl<*dv*P
zw}f+uGUkG96gLLO0GOZJN7Jwg_G3N_|8A~v9FcBG1ij8OP0W1K#SAH)&*znqPim$z
zNlGgu;Ps0SOceC!Palja%CPgHERN~^yF1lM=bcxJTh+^_+vw~~)Vge+G#p1|*7`qk
z2jc5rnBu&5p4HJhV_A#6x);#{OAdQtRa*Kl+!^B1SaCM(p&D1SQdo^<uuQLR7?O-X
zugKpfdVCeX=~qbulO?zKiH_uR14&uqKY6IYJwV~J;QW1XG$ma(!odA5r7um}q8HI_
z=HFnH(4_|fp$5H((r&z&fp~n*s@CX}^%^9UUyCaxL!55vC&MyqP&!QrdFF#we$4^h
z!dvq5t{c37ZT9feMf)N^qC<}Es|S>!n~xr<1|e2Lj(tYc$7b8#t%!qkN^EcfCKHoJ
z!Pa<6?BJApYa;};szSoT7+uBR=Qzsz?O_mP_Nm{2WLH}kiy$`(9?Z0m!&4)l?TM0?
zE|{8Mz?hc%iC)H;1VFK5cDp2=bB3_^9l5ZJ_`>`k<zV<!KREWQAi3^4+C=z(8PXfz
z6W7N2Ye-3=SCAt8=qH`2@NV8iGoOTX<ssRCb-oEjiJ~vC6G)|D`=Mw*M?V!Ub=Za8
zaDd?E=QE~1`A_%ixVnNuvAuxIdz|<iu!&sMpnlJ_O!|xrFN0<!LoE)aSbWI&&Ih%=
z^G8qls6@j7F|)i}9LTKT2Nxw5_4fno(5?6`&F+yeV6mI9Wg!Q-@)@U9U+30VDXc0^
zWGiTFR(kDvCicIoB-|eV>0fl;;_-QRB3e~;4Ws|wBS8@VgLx8`5fdN77g7!718^uJ
zAI99A(aULOzkS-M)cE1>4$rBm8yrS&`4~veUP~oxT_ng7=n{pJ?HY^r4wiQ9{(3Zr
zRvd|xO{CNO;Z5r0tzz=YCmK^(NNDc1j(wcfgnTM@TKS-EZAZ@o(f;A7C6r{)*mylv
z>_H^1d3>;O5=80osIx(Kbv$s4JeFA?K33A8|D7ZbizY2_-@MB=@bNfqS>n~i7zfo-
zhsnf<4t{fJV~ppGQ{mLh(C^;Zl{=qAS*|RL*IE-DcM;QM_`Fs#8rahaJfRIhS$tc6
z4cs8_iU4(TS$YG2a{ZgX5$}~soABAWv`9zAhek<XXv|`Kw-S7pyd4voS#LF;-G;69
zN%y4MdXn$*)O?Bh^0f^&o1#NnizMk6w{6}cjc%r%%tJlsXPZmD3yc$3MzpPbm9;#L
z)>L)51DF-+-~b{^Cw1~6mU1#^u#6VSaQYNNsHB9i2Um}fa@<ySWGAhnZE@g~ZMCLV
zTtzba?=_+oybrq24WiLgO}JgTb535Ohmau?s{JH5NdGqbV%NlOLfc}{3T_nbG^LV8
zfKM1XFg{MlZ@l8^wORpi(y-&L$kIA@bE_S+(9b!J0Es7%wMN(*S!CHvffuqy(yO)+
ziZe6jE&}WX(cut|6_>5XS*ZG?y_A?{Dzsf=9+Iq8IfBOIJ`>5u7^|X9(AR#6T^Zx;
zI2`omb!5&br(2hMtij>Bg7s{fv($5Hj5z*b*;I(F<xjjId+JE)Hs+mJhz(4ZgXDNz
zQsAJ5Y8bzo#>!Z#m^;}rVtdTiOwrV;#7f-#uphZdAkuR#URlW-t&}R&464jrKCSwG
zddRX|D~+--MrjrJKP!1902LBSPI#_%m;Y{<sD(>=45GJCVdBv2k^(aT>daC$w`mgZ
z=-ny+P9+oxXz+|{xleS{RL6@SXF*J#;{)%9(k^`1zw06uT4YoUtxIG4DgAk24*;Zw
z3e-AZ&LJewS$^l{%uc~9rKJ?05KDt;N6Hy->d5fWcx$i_8B?)#Nk{CA70ve$mEt&i
z?C#yB5V$cex$%IBA!DdFLvkX{j^T`BP!XU~YKV>)GPCfoApRa>5Nkyz;Hg5DAP#fM
z;VtxwiEV~;UdDiVQaNXTf_@U%s*(2(wmNxvmD!WNyYxE#UU(a~-|iytq|i{Jq^myj
z`<qI>tC+MHh9wDP@ZcbL;Jc)eCN1d>p9$UIOl&81{q-`B30B%)*Cm1@T|N11DlRd;
zx!F~|)j4_OgY)Hr$G9iXCS_-3Z<U;+(9*5i`Rr<^&gHw^T{VyYg!phGq4Z+ExRXg8
zq?An_IjS*HfEg|sT7hCoryjlx6^+k0pQ~*}F{o2xA6!R-v{eef+f#ZzJ8XHzIybsL
z&3|(`U&t@H69R^oxy5<eFc2yB`MY+3RaND+6*0_e!1j-yudxhKx^8JN8Ps+>3-xLT
zVMMK~Z!Btb?3^8vnRqTXIH<}rE@8qgSr$I$Y)`npe$w_WR};?VR`ib!w|tqvm5XZr
z^8;LNVkMS3(yY7ph><5d4kp`!#3JV^5}nmRfoHT3*1Y2Vqd>`3`eI@SLd~7mWV0>y
z_mYroEYb}^M2u?gt-zqFu(<WhxI6S2*6b0Vy#RS;x%E^B%PWtiAZw}@*ir~N1<QKW
z#7)S!4?g^Tm3X&FUjx(fwC~<XrGXi)S##E$P&s+Lr8&@sRTV)IPOC;vOA=)uK7wF1
z{mkFj?UU15=aTZ`)MH&nCq9c`3fxL}A6)^rT;^e(+Lw(%AWm`bT$tP8d+*Kk?dlKB
zC8ve-#683eE{BKwRsb9v<i#m8J8f@U+4!I;3YF2<XJxRue=SHsdU`rtknl$}$i1|G
z&n1HH<`274-)ePezTQWbaOZg>Enbl3vH^X1#@L92Z)6V5fS+3Vt5OL|m<8&N9)tKZ
z`3cl`z1SMsXH{Nt)p#N()w~N!w)E%7T(X?N1dv?H*-j?%1E^X-q)O8Iffu`^-Q8E0
zjAzF)(c~U<k5h5tN0)*cM;U1&HS_#0kldPRwg}YFU;rX#;1Xt48q@Hk3{T5Fkn@Yp
zm*Q7K)|(g`UHYf}y<%){iDQaixNqpOv{B7^ZlQku!%px}Sg0YZM`IL{><ocjV;ePD
z756=-@<(+53VrQaAhhy%34zBOZlK|}G#((`x9VRq;0V@8pGKbZ!+p!Crz6}#of;`h
z3d^O94oYKIWv=q5r5TkT-SzyQo8+DzuW<8j>|2(ea~hNWJ2|gjC2!383e?(WKzsv=
zKXyXVBEeW9^ibtC>%8(+U69R$mvH`wwS^UNEcL*#1SAdeOj$v;b4e&YM|-X&HMa64
z5YFCLU+Wrf`1Zoa-Jzs_MGwIa=0E1HQ3s8SiyEM^e$R_IO09i|(G_mw2Q+U>NMx!C
z3?w}6@;G-YHmYj+q__F4I7&J&K@R=ba!YaZd1JT`m*@Jvs^bLTG^<d#H9&luS%K(W
z*PI_(m}fcD_TD7J>Ch!v-01yBP<u7stKGb;DmQojRU6BU<TJ^B`A>5{^Osi#7{|Xn
z^nJs=th6rn@m!EOXiK!`&mTx9rccp9Ejn6x(7sZkJM}|zyh&@S6Hhe9>el3aIFc!f
zrK@nzDm2xVd8JP)vDfumlhd_ceg!qMlPkg~Ha0mn8`A)Kk(#7JcYNla9UAepRWJ8r
zi<}sW<GuczOQeONlk>!6bxjH}=MFx;Et?KO+vJ!<QE*7HLWk_}jaUjvs(X9LBXd?{
zes(U=%`JX2js6W!_BbsG(1Ckm)E|UDhJ3ukyZhXgWANIF)|_+{2u}?&plj&i90_V0
zDOeF&6SU>AtH?I4S!?lZ_BKx7^>g-mX|XfymeA=Iei(c{v=BUjSK=7oWDkuYt&!a1
zliuX3qvy+TB-T(-WwqFM=(~ze{kWLV9B`_$r0AXOr)w3CS7H4P`j$@Jw$q2ap6O*C
zi9O%_hoCXU$#e0MA^yIBVbw^yQ1U9G6o~<*a#5RKr{rQHu?<4_#ukrgBB_z0)apk)
zx#B@~`YuGU{kmgmCG`<3lAv;E{CI|_XYlg|hx!2S*UhQ``U{^&UoLL7hvHVq|Lo%7
zO2xW~bMN)_bvo|wOJpUdun?t|?fLbvsQhiv`4dx2hwF9@Dsb2D=|gqXo|#N&Q$s@t
z+BcWwdit57IzLXVv8$^~-<uD`XNnQw#knWeT`a23ULQ11=HNb}?~<B-Zx`1#yIIh`
zL&<$ud9K{Q!w>QyE<67@9NhWzzrG!Jw<@&$eQcky=>K!PCEOJu1M@zpNxp--z#y#`
JbxIat{{x5$VL1Q*

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_015.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_015.png
new file mode 100644
index 0000000000000000000000000000000000000000..5b4850b0f342da7a46d68bd47e720edf748385ef
GIT binary patch
literal 51565
zcmZ5{WmFvPvNi6mgN6Zu%b>yCA$X9%CAho0I{`w14K_%C0Kql5ySuwvkYFF@Tj$*S
zp1W%Gbg!PL>&Mhob?sgIiB?sX#Y7`TgMop;l$VoIhk=1}f4dGwL3;ZvvQ=_<JMp_o
z>$qt+TDW<dxO|5BXyWE%=jdi<ZA#_w*~Qh`(SeKY0~;3$m6e;DldAwbyZ!$xU~_b_
zWFHfgJa}`0>LjP@3Il^}@~;E7M6}o%24-MMUP@fkE9>O>A%#}<>pE(H=IqE5R^nsA
zz<{9h`Uh18P3?7qRW9?CETS>>u&hO0pl86&(Af)N3k_!rb+>MwDxa(hHow2;pHJHJ
z!Ek6|K<pigWasrw%7AwkML!-xn`NblR6ohVa1mH65^sfIc)GgsJZAi8y}Nl_!=D2y
zh9IEg(aI%JiHE$F0AD4s-!A@rr*mJ{bmsm4|LX0I5Q1D>^Z%@kgo##5|IgA394d~Z
ze>*`yrG%rEG58-Jl0iKsivRNduSdO5HvBU}<2dZ>?bR3va3r42g^Li3aLVguj@^rj
zi<7T9Pwv95U-^=qUScD|u8o9fvtGA0r(Ir3<C-ffSoik!+N(O>--{9*A0Lm5jfqTn
z;*Gs-%XB@kQ{s&UUMBtPF-ZSIG3OClI2_FD>Fe?D&EL;~uOh=`id4YkcrG2K>$=~s
zLJbb<dui!L7B4r9shT>>RVX-clfJNWjEP1PR2*^`TCzRWec)IeB1XiVcN@NpsaH5R
zzuA-2O)Gi<vFdo^cWMeiC{>K<7DE@<cZ&}`2nX%Gf8jaMKl}?G5Md~!ny7*Yd|Ysj
z1_aTGT>4k(Ai&K>byF8g$)lo#B{>MG=M_pQ$dmVnt+Ijp&sJnYs{v)GEfiYmYuihS
zufokODlM!FqTIFlxB<@%uc#W=#kgJ1UuQY$KVX@0_w`6Q%N0*U$!xaP7V0>x3aa}i
z&fM-DUs?kFd)7m#HiiCePn)|>ub1qv=TCtzzh5c-Xzd=<DADyhKK)H^bl&A1|NX?t
z<$8_n_W}v$zF<fayv#uqXwx<fewD(#N`z%ySdE0Tf^OTlS&hS`c8g2`up-iEE-gE4
z4$j1zP$@7|;C_9MB4qKTf>Xv6^B`gfx`3JV%@&&pTlKA4j^GQMFb$^ISbAaox$=T;
zN_`-<FS(!qeaQZOcgHvuKq}dc<^x}ETlG`C#nTbS*Lg(ny!G0Z|5a1QAaF|gV|;ZL
zsC#KoxOURq^=p#`;#d~M_(dtXq>O<G>B8@wSqB)@c78{1^180BfI!88QpVvNHP?LS
z>#+40K)S9-@lgT>N|3Yf&+^0&*0BI!WX(aQQk&|&rQ;4tr3{AjsUV_^XKSDZgB+Zr
zwiiDB!e=)l==-+T$SScGDEutl2~LvOEPZp3jJ9H0=pL6~8h6Of6v3Aewf)UQQB|-t
zwh{{m&F}L7(${XE=oZhP&U;<&g~mP6i+R4S<(pSHZ_jd=&=t-e;za<R$?<JJ637C~
z4m^V%S%9};*Z;A|^y|e#6N{p~#yZWz)j-JLq}txT3Wun$*uqSqdMUeGtOq%ugTzx7
zpt@5W34>aS6vb=KthyMoiiJ@6I5r8pBqosdo<}J08za@sIG_Mg00|4r#0<H)+BcA3
zfbs4@1K!w_AyTyonHv6qfPfN_<f#X>MwL9=l{6n-M=v}ZOS+=;(|HInfA>&mT$oQ;
z_h{z)!_*;{95i5zLIKvysUBnd!^Y3fxlJM2+tu;iZ4DB4fBm-BzK7W2`^&W9nas~Q
zK6^rrtq%bF?QQO5<dw$uk?-Nx=2O<DK#*2Y9g2x6T6r3|X9gk4p~wf}^FCJmw(&3t
z#_+4c>m0s-D5$0^u|-Aa7M+wGjj35DT3D48K7PM}%`E3^D1}WOXG+R;oQm@117hn8
z%%L4FkPr@=5mTn$p_3CSRN5GKjL{uGDQDGzwYfnDm9Z10ycUHvhaPt-epDy+2BC{7
zwX$j{2##4OW0Xe$x{SiDgKZawjL_wDH^yUF5s+3rNK(#&a@;CyhOq<?qhaxCr~16=
z7~HE9i}T~d*f<0^#VP0C_HyC;vVC{veEX#PRqJb%d#2+4yUk%EpM<Us27#$^hZQp`
zfu3s((N}W_obCpOsL2HENKQy~IoF;W$q%9}-v!B%ez<WBIcfo<=B27x2iJ)OItS11
z-}OFA<aCV_3=+u55k@492_R!E0Kj+-<zkw_Ad#2Vw~`;@hr&`0Sm+?}t`+y#*b-36
z+Iz{GT9YiGEs^J=uc|HmsSnjR5ENlSAQ7aw?VG}5*Tf#gOnk{|nej!F#ziv&xdoNj
zPWiDI-fLWn$*jKKKAulT54kNfx|QEwJQI5G4n4K4`^HC?QmYppaO+`U65vYY#=qx9
zDiu@EUs$`LtAsIG?SV^zi>s2wU%rF)TbETJ&L?srI67C6YKUqyE|;b$?@IaV(>5}8
zi*~A&cC;z-<r^1z&4?kntb%x%OXBl0CvqBH0e(%E@HIwwFlrp4J;EHI9HfJR=e)y@
zW|DnJkI-ke#pm)tjR^-0NGk#8@K*mC05IBM^el_7);awN)mo%I(!7jYe|-PpyFVkt
zC&h9Wc1{t&$Xd9v(aA0Pa51?>qpuh6p@c-c-g1Pnz&uQGzn5spf5p%7CB4jR4J|nV
zX85!>^W@y|s;=^x$t)!lV2j#)kIr#K;DkyX`4<qvy%9{9YsvugDHoH3MKHhf?<Ox4
z9IpTkJ`We-6zx42A@9?<_4ui3n^wZ+OR!heba8_o7pGkC#BKU+;UHuwv<A3>AQQyq
zfUN_^h|Sbdw4H^xfdi-lIl*LYSrVb);yX1Dcdj~4MP=LtA`k=%)*&4;E9#SzKBOkA
zyz7zv20^ZlYw+kO{v<Avq-I&3yIe0829<7EHMi}yw1<uiwfV34-U$@Ir;Lm%7Zgg9
zg-$)ACsB1M_@o%ngo$mgamf|Br6@4|{&1KmdHozH$j&M-WfIh(6WTFlKhxxTEXn)f
z%+xK@&G>r;z|zk=D|k|<NuQ`(B^M4{M@Oh(;d)5()zQ$>^e;F1J|tHs%VjP5<RfSr
za)2bR;^}PKBBl_%?|r7VwB397avFti;-7b^2_ujX)|JKsrb*o+O@9KIBH)qrv#_mh
zKFqK6q{A}!yN$!t&b6&0(!lxU;c)ilN127rk^XU&Y(w4ukNy6ZrPw70!{$TSkLwG?
z%On#luRObaNo42x0@;an-7iJ$e6f6@2(L+T$%^8I`cz`6g2R+C+D;vqjHkZJux>ra
z>#;bT<Z6a1O>*)IkEmKiU5?-sG{MFpm-=KR*nEj*I$3M&+nSXwQi;D_FeCo{D|Db$
zN=fg6)2aEi-k@@e+MSh(uu0AU7XW7wbjWc*i1OPLQ53UF_|aLa=~uW&jNDy2UFS3o
z+3?m78gWhlB~XOb6R(wgB$1j4nLrV8riX1lgUWc>ilL4WM9Yb(PSEsslj#pQvn@L+
zbDr4yu$TTaO+r2v56&#?dZ)Y>&6nKe18J<djw^R`>KD1gZSNOUb_Gh&9T~yul1GH-
zF0f*O?snzN&E4W+bMHfq>xi-no(wRpn-dT!hlo&$A%JCFZ(EZ3K2vepq2Vxs&|&?F
zvTRuUyMb3$_;n%|uQQrlXq3<9xuND97zb{bu~30R#cJA><1lg$Q-Lp;m9P3I{v??D
zE67f*;^fO9`-VIudeh`b8d^i8p|c@PRZWfb0_Grr3&1`Rp_M)_RRTTukTsk-MjWM3
z9b)<_^1sanak6Wqn(<@1=T8H}!by9ziUu8U>#$?CaDN!ij?*Aawu<*21_%>l3y(@F
zsO2+0rBZyT<)+iSdPx-L&Z@XbOr92Kl<40%qhES$eNQ*!Dbp!lK<>Mv!fO#^d_uVI
zboC)Fo{*n5osHCOJ2m(Y7i-z?oVV8mg#=!?_j3eBNzNIiK&acEx%wU+<*-8?BM9bP
z7dsig8inH7aBx14H9_O!OEB?E?pU_V828etIC1gH6D75aSMv!yPLGd~-x#EdI3ZV9
zw^YJ_+YQjmL;qEfi~);j>iH(>yD(WfGJBe@=|ei(lq0=RgCFrs1)(%3GzmExG#wN8
zAz&O{nFRpzU&^emYvJO$q#yf>P4ete&%!#;9DD`GT-z33(X@lh?iKdKT*FI3awmp{
znY)MGz{Uo{wpoGsFWo09>2gez0dr2b49u|EonNp5$~5+J<=9Q0`^dH67NOf6vbcC7
zDLHIg$?`P7ao3K*hLFfK`XD$#>L<*Sl_|wGa`5$Dd8GQ(;suf^{*=S~Aun0$&F*>u
zpHRYK-^nTKFRx=H^?J3g0=3HNLpMgOj2z-^A*p>eaG}dz0$btoC|LOuk}x!m>P|mc
z;NSnWC}G!;iLRD~J5ET7UN>iZoz-l1-u{u<d|T>Fy^P0r6e2y}`9;wwFHpw5rEoPe
z=ui^3shd%awZKPOn%v`@aMnYkdEf79rJt|1j~TUJ<EFrW9kfdib;{iQjvKKbAz#W(
zK=9FQ&{QtIXpq939|H>)-yp`9By1Ev^3%`NN0CzpRBIO3eWTuCgP`O@v7)>c*kza$
zI7d=I_vx^O^H#HemG^`@PC>4PQ-tB`+8>kXC3l+&Qg)$NL4wG>OSu|@rop*Uo~O^2
zm0Dz&%_dQIxiS>pxfa1t755u>5_g@$sXk~TPRxQE2d|87g9clev1Lo&n8)|ctt9-Z
zscC)V<fYsSBr}i<jnj8o2KV2I)PJpaVThJ9bv7jWsBmPat+n`tgp4h1Q=lykd@LUh
zcm?(OxH?B>7#{Q!f`egG)e&+E4TTFA?S9q&8oS*i%0UhfCNHg35fC1jn`1jviUEI)
z^=yCL&k}CvHXCmZR;Qdp37b7@kT#0l?Az(OJbNul_h2K$At`h-v=?=A#5h=(I;@W_
zz*=T5OTA+rQC7nDSVj;J!m$@p!Iz=vs!~iEn~E__3TaF~4OeB@Jqw>jvS*RG;cU`0
zvU{?Irv0bdqVE4z&4M38K;g5V$?gew2pa8XBBtw8YowS48lUfH&3-Qxeh-dX-W$1D
zx7y4P?WR)K;S$qPscu>|_Vi8cd!f7Ib(}hovI7h@l?<;OfW+CT_V)*Vuvi;S{?uNy
z)zn?mFLH4t-qtO8W{e4qfDO#5tAj7%@&@Z6Kb;!+L;4$rTA%hRfv)MswCFTsR7><o
zbL;r^4t5^M%`Hyod@V@fq@ZA~mxPb>wqR~-&j~fD&;GG<;weqj#nmQcoNOGlFmHc+
zX`lv8s9ib}c1Lp135@wK$k72Z*HO}ad!jn5x|bMPg{0r&D~EWQNn=Cp7!9#XK_PxN
zDSESZY`@iN*`Qe>H%5N3<3Vw8%L8o!r6FU*^-*dUjPp!UZOhov6|3e(@~HWsrGFp2
z%gDGACqKVDRK5Rh8eqmifQE2L(EeqC&$Q}WagyNh*&1a#@x-Z{6>pg)6QOyxnT5v}
z1@Fka^#?k$n&NNY1<qTQ&GhGpjpNZ{AWDHB`VJ{r9D*n4Br*YIw8~b@s#ElYh1PXo
zo#oH=hyLX^rDsZ2M$5AD*g&eZUk40**ZxSdeS9q17sA-FzI<Ug_X@ypr@YrS$L{Ap
z`68ltWBC8JcOqjH=WJnf_dP1ht4Yv0cSFi|H1o?LFXq?h0RUr6zA|9Tw67~6aVfKA
zHT<V<N}(eYYmTnhTHi-~6$$7vy$U2Tc4<RlvvMd*eeA+cyEzMV!}(HcLwO6dsB2(X
zHB<6}se3J_$WOGY3?|}}{+v<zt}^|(#X*;(A@^{+T6@$R$r0>b31_DX>*=LvaDFJI
z2L=($)N&o<rvK7|wP|}y6BW?v=g8cpLmF<2x3!`@J)-u_hVLNJ<?6^u0E}3C6L8PS
zzS_kf8yyeEQDy0xm?#!_3XW?Yyu_%;h@|&Vbu#(`2dmX+5Ln6~<Fr3V<me9-r^|NF
zvo-4^1ywWGTCpk{dr6f1K+Mg4on+@`Sn;mIK`F(h>yrWs>M5?YzQCF^Nb^GacVZCK
zuMT?o_blj1AmqUG?BZ1{)F~0tg_RX4+m97m-6=74yu&v$2;Y?wz-=;hKFB~!MimK+
zYzMhIwJqfJXMx|3zOOv$2vIsfw{!Kd;Ul=o>9`(YT-$?M8m*vuDwa8q=t(NSCp$_)
zv6@*4vUrqCfDrGzr**{ADiNbRa>&nQ3=E7EWke|fi3c<xZJlXTgBa1T13f)jW?{Vw
z(HC&79r;}<4aooKWn+6ER=grYzvNY@99^ewW5Y`6pf_mo>iC|K^o1FdS<d;9rUsel
zlVdmu)B_GeeoA;P<W(%K7%)5U_I;G9iM2=NoHkIW+c=a({+7YWDZsB#&!<>JnSAMr
zGVY?ohCe6q!}n8E?l*Q$F8%}s<{OjX_&@$n<>FCjhM-_pNnOXw)!z?#FE!!B7wYhh
z_WRd-TqTai*-=3rP<$M$EjMS<jv!C?8JH+Us!*WhGQcuOgBro%^c0V>R5JU)D@xYG
zDz0{aTV8&~(@g|3)#ZB?1n#YGYJ1sr^?PL3UI}G`3ro#v;gemtT3jpV6#AOKI8k14
zsv%O&A^g4f_Btl64By=C)r5e`;8K<mT82uCFeevxp#9*{r1=$Kg3#xtYt%G)N_MVR
zd75YTvF?lM!Pa(`XidjS4=COuEwl}9)DHc<aK0P|2ZxLR;f6P5uW7PYfi!_kn3~jw
zpQdr|vq#m9*7GgEcBdyCTpwi1TUJ^FzMsYuDrW~EFe88m3Wj!wTvMwGQk?`)P>nhM
z@?{PVEU?0<=eNRa{yD}&O-=8*<E-sM3qh%9VHl52iaivMkfmP{u8Lql8#KRSb6=y>
zmj8fKc!fhW!GRKe1s^ig&jcq7`mZDoJXExN=oZz>uSi6XA5k8iw713j@%idI?2^eq
z-z~3i1ur$E;3??0TGyfkg)y3?%qeU)?#$3vGGsFS-;ZWYoA%{@7}8yYc-*6#V!4!*
zF((A9rVUYnwv*yg4W$E+F{)`wMs+zOJlV6R8ekaf$3hJFfaK6!PWOVQrMA(76NXcZ
z*p_tmntHFy>BQR_)kt3&`;Dzwk}TcA++2OC8o1VVK_wW=CD3A@^u5JJDlIFFeRP_N
z3?lb-(XqwUFw0jl2de&w5saz3%QPQQ0an)BrpCrw8vDkQx#>|V6ed$661gCbmi@k>
z^%W#U3x{ogp-<;uD646&@9$Z-S3@DQJ(MMTWZa}TdUkU8z2x~$0<AnsP3u=pbx)h1
z_Jmrf#dc8D^k+ZQ9FH1_hEh~P``bHDYu@8HmAHabS*Bfl&*eDfID7=<or4DvaVAM~
z?p>Sk)oj+GT|8(vU5g-vx^?3BCKk1Rv&z=)N6{y?Vq!MaG>Kb@Q)ND~P<QuOH%Hh5
z-(vd8$<NW&`tSe1!~-Gh`I;Gd!MI2<+nS`0wV8xAv!v=F#9sPdJMwgvdmt%>@#0m7
zlwnLKPQ@WGIv5Tu7;5Ccikd1K81;{Kd2{=z0E%g|ilg0Vj|jrRdK9`F(Xgx3W+Xy;
z+Wqib1*&D?Zb|HYTf2<r2b1c{j(Fm#U+k~|)?yN5VQ0gSa4)y#A7;;%wkKKop;SC=
zQqwA3qqptmMVcLRKtSep{&^5WDuo(83T~aCVMp(=x@G?C@!^*L;H;_*JM+_p-)Q<6
zg67ee_){#=`@Qulr%mgya|pm1$+FFzpYsN8+lB7vrmUUs2I{_Gc=IWFJTZu!KP|f;
z>D_i!izKsMm2(w(5dG#KbV~pkiL|^lPN8+UK|7!Q4hQ8x2{REo!z&JuHb)`(E9`cP
zG^PIyxr&hWLyf2(#HJ~hBSP=PSMR54S7B!cDg<?ToE&Bt)SJh{CYF+%v94RaM!w)5
zvEgLp5xsiz9br-?QB@)OY_d!j9X93h@|Hp}GDx_brna_elar`n14+uhwacs_R2|?0
z4@soRtE`Niq9rTU3ayga`NJ@dsYoibQr`MYzy08!ENiqF%Lt`#>hh`TI^1yH<#koJ
zB|noVNquK-dqbS;UHzZ~7F=#_a!e`W3gU;x;V#zc6Pkw}813mY6Le_`*Y_#o9I7m;
zj@(J=>W<t1N)o{<mqg+*L!$3^UJ}&iaZ&FTgmcX~|AC1@t~Z0Z+ih?T32>;SBDajc
zYFM&WJN0b~23{MG+FMSRSFU}2If2{aA_YFL4x<p)fqzBV9?uWRe)ac3^o3_YLF}WB
z&ribqHX}){$*nlrf-x-4$qI=Z$}QtCRo#%Lbf%XfW+@(rn3V7+l&a@DPwZ74hLq@{
zM~>Q~4dJ2c7`A>a$3}T7#i!Ba+a^kI@%@a(=3$)jSk`{hT+In<?u*u1zHHvEZ74<Y
z6K7K(;l}e<FO|^M);7Xn>hg9v2e?7)X!^aV5DMkNAYY($%<AYjFf<}El3h5zQL)Cr
z#N0fodXVLNT<$006}^?bNjnhLwL^ZR%Qgz*WS=qXz#4+~)K>l#5pzOWoLRP+F^8Lg
z{m`Qx27!>vBDX?($gEu1;tJR-z$L5GFS!;1+_k(&QxQM`CCh9kqq@|R8mN~Nz{5)m
zWb?jJ_A*R~$)?FA_$IYQd`T8opmCEX1c_wj$1t_{l4aU7N~hQqknm+VnOR`+QSrGV
zaAK~mpgIT_F8nPg>&7{%Tdt=}?V_mTQ4PE(H-FSV=H!Dw5%2qCkmB>mWT*&;g6D8i
zHvh6SZJwne28-!LUx__H0=MY1{$rwkSw>!V5vzZanD4M)(|d3uShpHO&J(!=K2fEi
z9e9|1O>gjOBHMzBzSdX5{#a$34doHy??Ona4FhTd+X?+N+lQfpBq56|+sM`t%LOct
zo`5><rHUvwAeCgWa>2(TY*iA7*2>0N_Ogx<NJu@`d$EA5AEWqv!M1f*JdrhOFN~&U
z!L&`I4AHyL-nyC#u^zT}TNrEt^NvKxoh0t1hJ`99aWJ%Q9CBnX>wQ1ql+lqQ&|s`=
zt$KlxP=SK&*WmbO(R;|6WHh;4>=)S8uUpv-j#oPci8alwigekVD-Zl4&8=N;_{XKR
z)7Uvz+e<yKjxHuYCmr6Iyk$wcZ7Y>Nli>~+CPo$QF4v*~UA1%hJHjs$i5%6P1Ql#3
z<q^mtCCdjd6)f6qmlA}22R1uL71OE-7CA3;>3n2mqJa<B`r`4SNUPG>t12Xwt<2iW
zDaw@NW5t@ZH!MCr{0!NrvsJyV%i=5~3Axxwqyk@7vp6xm`2~`*b&Di*<M0#d3dH2_
z1qt$Asip3XZEs56w^2-Ic%U!bu%S&`9TAk^K_agq4d~QHwE9?-E<HE@^QNZY>9K1V
zg89G52m9UrYxk=cA?}}YTqICYXZ&%d(6xEIhi$A<^FY_Vv_qwlNu_2!xPtQv5#SD!
z7mey3&g$W1dVxxGMQnY*G=FYTpcNY1ZRSQyCN;}O(#OwA6lINT4B`>j<Bfi1D_Rq`
zM36x{GRYkE30Lq>(=?Vt1o38?g7!^@sIEYQh_l(GW%CY~_vvF4)->mliDjtsZf@>Q
zGr?SY%CMz|5iDolljyqSVt*wOhJXXf+J3K>i+^!p(nQx8ycifihJReV5R)<3o!Ia-
zIS;;0R#z*VYr32y)fUmo)4jV0>jcQx;8xcdY~36l7DQIR8+wtC%1Nj9fjk%9!fj^!
z1)Ih>Xz;JlXXf-?`_<nn+>0kffA_vJ>=?SD09PP2J>mR14p8Z*AdP%)N-<?ep>99J
zkPtU%=ee-VUK6M);UiK#H)Hcj0czjZhjihOT|sm7hiQ44Oy7rF0tGdHo^0Z#dmBy3
zudpGi0EMDy#%=uW@KlKWP$eTxAn*b?a12?rb(G6eW?&!84UTN%gx!;aP#=7<X;5~`
zLQAlf`9rK=%i6^kAGz->yJfzCZvFnhsCdH)X+C_h3HBe$hi*+i50oDqR&!UY+r}|Q
zm1Cb*B_sO#17m=0q8114Fptr9!YjOXI1GM_*2~{|{~F|sh(-@ecrW6a`^l($<l68q
zUj-Mg8q%a>6tj|<yGMc&qC9Jisd-0{eCaHxxn%05cZ`24cysmroj8;5EFP6r%Zzo+
zoywJ9I49AXzw^pk`=g6_3!}7Dx7#H>0}_Kj`DoQUHiM3E_U<G4cis|l(s-obul<3e
zX`akk8OaN+VO7Mt(|p~azCp|cWxL-UKdIjR?uU@}xlMqLL2`MjUbW5k0T;KQmUx0^
z_-U!aBPT=BcI^%QF6I5!q4IjiXxyaR4ts+>f@6wbWS7uc8p&yieQuwF4i8D{waVNC
zS>Hm@(t<3dHXxQaaBx8Dn&O%z;Gf*~Y>$wH^-Jzh%~mU_D~S(-e_`Fk{+aUh2#Ean
zzP{o3<P%LyD4I3FqFd0w{m&@#mX835nzg;Rbo4t+$52pEvWxBlzzW-kilGHd;q-E-
z0dNlii|E5X5*rlW77=<4=NHfK+PnL}p^ao5JTSfbaaXJgJ^-`Hx$e5<%Qt$(lLj15
z>f#3nMF5u+M6-t<M*E4&&9kf7K68xU*ZgyCx&0JC02|DkF|7S_BKk3p(K4LWc41*<
zXA7Sgw^1_$G8<Ei;WvfyC>`xj2m%Qyw15k2TV_0&?LH_@{K7ET7}-_k;OFTaX}V6G
zzpGMD{wOLA@Tz6{q(oqr8FF%lqjt!)J0|j{f=zHZLK;40h)|vq?pJ7t@<%FA?Q#`S
z_o?8A2X~76)>^X3K2D4<RX>Ctlq46_>V~|Th-E&FbA4h?ZSgobGsg0|mA1N`7SX>!
zHmpUnVY=1;cq!6!DqEQ2&A)}Pd10zB2I_?qE0D-b@5>;WY#?Xeu?91N!aV@<5jhfp
zS_KR}ba$6QNo-q{Y+jAh6yUq0y6tYm@*m&XDI`WY_=A7q8w4jUkk*UsXG)>YbY9)k
zpjftOV#2G!meP;zmqckfE-X)%(tS5&ejQrdjF}JWNe}U;QML|D_t{Q@rb!3fyhOS=
zvZyo8Pl6|GKA}#1?d&ATK@PxC1;}ZLtaN@uQ9%VQdFSM(PdRpy`fK`W$Z0!G>nM`j
z7IIEpx99WGrL$kKzwW-qzl~|H;wxWKkC%rF_Xw<6WE5&69JCASI>Gyle+rs%Tw<Y6
z;@dPlMp}ht>8UhgkSrpXDEof_nsZrEQ_I61$jWM+UsYtZS=hP8Gr#3@<CTNE{*w5*
z)nVcin)n)Xl_k{|+=lH;%{x{o)(AF+>IjAz9Y(txhoBG9rvX?RoDng7Pxj<3Kg(=O
z-a9LDZXLtboz)SZAw|KHdV^n*O|V5K-cmZ|n7?R1kjU&KV4n>Ek{x36s$q2dT0mJv
z%M$ZP;%*Elg_chn<>6xRQ@H8uJNLw|PexTP=e&WB@s3Unnn2_oiG<j{tYq^JaVa%Y
zoHPUmwhKbZf1cg~X`>j-!J*Djc!4toPdjIGq86zSYsIPCx$?vZBV$_|J(;m~_m{AL
zHFSCW2G%FV$TS(nL`~tg>jc-g1vypm6X$ns38`8=13yt?H@6UFdh^%ZzKIEjRltzm
za^bW!T^n}*Q>;28E!&d;IVXc?w=TP45-W(0yX)^9l5J+>iY^-mUl#Rh`~~&CG!`}{
z2OnQpB`sY-hZXIu9EjP6Z1-8ybN$*N{Kjx#4WLNgJHSVmz;meiwV%>A1f6)SroMS-
zy`o~$4o%3Unv*(#rNV-2wTx{IV)u-}5>{v=HOQI;cPABc*7ho|2mC*;P5R`nQ)Um*
zOO5gCZ<*QrOPo*M$<XUIE(_=E_j69#8Qq8-<-vcoC|Z!?#ED~<BJ+c{q>{g&qTZT5
zZ8}Sxm^EPK;&2HGhsj}zfiUXh_tWJ@38*R?OI!+FDJ5w<o7#R4%_+zJ`SmzzkG_aw
z(F|&a>CNjP(Yl4XL5kBMHpJi+>opd1I`|2N5sE!7%sV78kL++DDT+W1!B7#tBu{5E
z+&No)zsBq@SLa#b??&^{&Vb)^nc0Q)f<#9rclxO7tgqPOXnzyd_xCMR)FUNPFdidu
z0)3wjJuRWcmvQufLv6_ZuJ*frup~iChA7-_#(VV1X}ll;JkLz?^GCMS#H6vB6u3C?
zQke1asun#&8}lh$v<GT+Hvk#g8-vHNgUrvMU8*^ZKuJ(zUB8u7U23f^o&kPNvN;S=
zZwgaooC+7ZE*@8_<SA)T2qSV}P2*>%W56<hqc<Bg>0^$CZg1Gl%p?iiIh;@G$&%jm
z1n+&@9`qGjrEk@@;_Iwz4UB$+Bw#F6Aiy|D-g+&yU}`+d(fYikURJ<dZud8jvhdSc
z1<pBYT3JL8w4;r_<4)+qW`#Jq&)O>Si~`;N!jm@)D&ob*b+RpE&^LVXx2*EIi+Mbd
zNtmLsL<^0cS<pI!$T{2STHVyun)fRZ9DWJI*_}k31R<q$^mh@Dd2Yi?Q;6?SkYVOX
zt?bm$z38v1``GNcJVVCQOleN9kwq#zYSJ?3P^&Qi*DjSNqs@5+Llc`yBrNd$P8m2V
zQeN3v@0@NQG@4P9*S1y$L+u7`9f?MdW_A0)(Zu*I8-jq~+hVU-X=qyDx%%8fXj5_<
z*JD^kI#J>$mq3IF`-go%E0C&yEj`=^+8|tx&^}ZkC1$s*W80zgOLw^P>8Pgla*wNU
z%9@%YZyT^Xn8Pb}D}-}C5Q~ozcjoOPe7<|IK1W^nHd>M+o&@8B<IcH7Qx7lPQ{m%p
z>)3ifKCTxnp*_6G9P|ZzBVvn*zhy!!oZJ9V0$<<dl{WWS3odhS*Gz@7&YltOZAnBr
zx*}^2PdUU8hOhSl5hFn=b}Zt}9cGCl@-4G2`0I@pW9s+)U0SncZBl_8$Dd_&(Y0D^
zj+UHsDm0ir@z;U+2L^DsQl=5UW7K)1p)!i0l_T1-j9OsM8%=Jt@(1t#sb`N9vavLK
zFNX<;=0LMu?Jn%uhur_8sRiyciq|ZE5Jvc^tasXO1~<N6F~7dEs10$|+koJa@$u`b
zR3o+hZs}kF1yS`<v7$$Egb`i&f&_QDIQS)u3D5l=g@;Co&h^_n&4`Kl0zI-~@reOw
z8Ys9^IzOFVPw8eYdiL53u=FVUEc-~l%Q*uBlfjmV@3AI{`nZz5<nk<Z2GNh_^ck*F
zJCI^H_<rhmj<<(jp>zq}?%6F8RCUWG)N`vFo;$-T-YX)Zpfq`deV*aDrUvO7?ZLA=
zs`lZ^8cX;@e<hL?b}!LmVj~Sam_F9~h+cc2v><aH{d(Y@Rlo#Jp_C>~$;J?L)h5s3
z6;ij~%%!y6_D%0xS%+PK8j-dC^wh8w9HhLHBR(S<+YT)V7pJaPxLTO5bb4O$B|rJs
zWOC4))%B2DQc{ut4UsbNNS<2+A0I!C;JPsN7#`5ik!_KkeadQzj=s<S>(ftMDUL<W
zH_GBCmnheun201Vvvn<XIgfdbx5q6OfHcF*#br-!_(Fd3JfNgK&|VfD4hF9fYP+!%
zPsUPd@47wrk+RJ##*R*rDJ5U6KCGH{HnyD%z0N0_Att+|F&z<_DuEJRx7GNZ;J!@V
z2M{`LUtVXsZ<XUy`L_k$^Re%+jym8ss4?(4sGo13_8yOmjMuP>K8qn4ykpfjNrKfU
zd^-x2H%ULYUJRtL7&Kw*66uj5v1FmED;3epG54mV=j3J*#wn?AegK6iqyM5tUS;E5
z`MhmRxEw57oMY@MDzzlZY2Dj4?8zs}IX)2-`A%#lyc;7fp=%6DX-|faizCEJruOjj
zcEA*3SR=uj{D%HH(I09b7i2Ujonzny?6Ke*OIkm|r81H-)@;*baE$ed+;O=cJZnge
z8-&O%g`gWIs`O%)!)5D9vUu(i#T6r3W^?t%2L8?KT>?~?`7tJUNe*(;DdT%lj*A9H
zum|o+Up)xre4%0)Vu*|PU4q%q{gcI=FjcQ&!xlv28MZy>2sO5N5psN*I=?!9-B=G|
z+ax)rQ)cSj$Yo9{z5JG)1GJl|qC?8VCCaagpUg;}<ml_~f2cEhLHqMBJ%@x=UtB!a
z^a(E`kw%!a9Jwq)K*m+T$Z@5JG*jB4_oo2I5j?x!)^&Yb;NGk(J6UvX$7)4H_@^ap
zNQc$^wCJni?d=c-Cg$Y~mx{{+vS{I=HJ%q*K{j;Ff#=X|31AkrePG3rEMeNV{WxzF
zl@Op3@Phge9{<86zzJpS=BVs9+=^C{IF;AambBr^6`5D*jEZ3IxYUZTwK1vX&W%|}
zH9=N(_TH?~v-{82%lYJZ-cTf$UW`&rv~!L@)c_c!jObF11$1%U5|S`Py(udwqjEeb
zd-SS>y<MVd*}PI0$c_-cC{*ByG*EPF?U}b0%z=Hod@)?i{$&s$s_2lIeY1FNR-j)x
zve+q4RxC?7IVUjegA!T=>-yuIAWNkkCy4v4l0i0a{}3{PQ()_TjIfzM<})B5Ae~B~
zc6nBIp|~{mbiD9JirNajaryPn85-PWc(yc!3N&FdcvK)IB7(}cdjNbmfjXcHtx)O*
zfP#cLBi(pXI6h(;O#U%oak%`;^ItLHL=}{BC&pcM@79YW?GAAJtM)5%k)DaUQ<B(y
zp=N}VKx8W8mKY6kgw^~HP21%Go$1rs%R;HXj5zv1OiU2Bj=8y8Sl%Yws7I(KlOixP
z-_SnSPHExL>1akbV0-?RG6y1)dC?wyO1<9)Eq0Y4V983W@I_>Xs-nCfG<TIEEyQWP
z*>crn{tS-COEmC`%M^Wyl4+3UAf|~IQ&a>rSn*)11vnWowWz{1a0BSg>y4G->PUfq
z2`5;M#KQhT@B(*)Am}AC*L>5%{Uly&BB|GYhTd1GPfA4%_AiSZ!Ld`>Gv5$NLv$`?
z=i8+U{nnm`Ma5^AcIaHpg?8(O+Qc_7Oa&qic~8zIcPB`|k7q*Q2oTDlGo;ihDLS5d
z89t?aiO1yyfj=;o(_RXp9=&+q;(kf=K5~EyPhBv|U!NQ63_Cs=|8;yTu4#5ndMBtF
zX7e+u&-ep#rLb%<cz@0`Mccx1vAJbXS&u8a25}zN-o|YSvoKvkFI_)0-swmZp4)nh
z?YupK06Xff{N{KaVXQBSbcAtGJLT(y1DRP;CSgrI;m=PPHHU(nPL0p*SnSvfPi+3)
zd`Ngq*+qd|V4&)I30$bR-tK8LS|B_ge;}cxgrka)ctYU}CpXcgKjV!mpv%GQF(ES%
z2L~rex)sYS>{d-le&&=~B#(@I?^YPyLRyfA4_->%g}gvs+2cMco0iLv#wPEHU>}v_
zsIphA8hkz{^`R`|MPk(V(5jg0xD&o=VyRPzW{3mzL0dkMMtG^_Rdnj?zn>%HLF+Yn
z6Cw3D$MLY42fAQ9OZR`<CY3ot$C;-4C&~BwVt}#zvf$i|q>seCpT8o7PKLrRv<+tK
znsD&+cvd+<2TqAte%xSVo-UGP%m!<YXi%wWKuR~=Kp+F@JF-V7i)77wNCkw!hJZJL
zL$J@@U|$RqwXa3QHk;`OyuchvF1}>V*X`H8<ALeb;TjvgRjR>FW6ln8_<RSe5#9+6
zM$CLIW#Pj~hyzgtq`eX-XuT|>SFmc#J5}~L)>aLx^XHowJhqr=;Oxy!f;ov52gy5}
z5S)eXA3oz*uYO)s1rC=UBzB}GaV>draH}Tjq%j6MzIU9i=$4E*Mjc2<){kAf+RiK1
zNGs*_Umsf`G$s`|^a<6nIR>x0^5GDQt-<tmCBJ~+s!zv<KnW>wnu!CC?B}}o>Wdh2
z3WYcfo45>+infm)uGV8&{E?|aIN$h+R_!LO>!P}GY=J468}D9I0y_&-M6e}t%U$|i
zU&D@%L4VOR=Gr_s+qm6fY_p8{vX$|7B9q{@N)$!X=~3XQE26`p%Y;4Pd=4vS4)C<K
z=fK`$2w0gVt3U8vw8y4grOTCJASLZT<7Cv<68<>`Qd=9_QOIGZ{}R>7WqDzt?7wGa
z083mN;}iJKdfw021!EIBfH8K}r5zA|&_|Kcthg-af`nsVWpCFc(b^n250C=DDg=cj
zdn5>E4xfzdeIFYKA|ZI;ICw;7Qn=NHAIboX%}WMEpZ8zmira6>P%3|CKXH3~x}X-5
ztj98&-JS$z<$Z`bzVnIRW-zfk#}=jr0LVH~q<cfs#&}O|lyPA{VZ?*8^YU=R5ZKe<
zR5pkB&f%12`C@9`J>H#&UmDwYF|j6*e#USvzubz_#pQ$&3?XUG4Y9O6v#G>kZy|0o
z06|2ph6WIk6<Q*;xb@%f+fEo6X}1_wP7_k5yu;NQ3-`?msnf_hYce)Fk&u^dx%J$`
z!oi;tE9yQ^DSWx|tHniV!Whz!%9D}J@V)zrq7I8cU4wF%*PZ0if;s16V-Ytg4_SY-
zeC0_Ly3|kP*^k!M_j)b?HzS()gwtjp@6iF3H<ln?)zOoE&8+C99nJZfS~o)6_$K90
zeA4E;PSS~~e4emXER*8!S<?52IRZa<&*&bW*%=Xs(8{Yz*u?^q|0%|7_g&}+FYxP-
z#o<88>P3BG-sRL2y$8A)k;#~&nTO>*!|bbUE`{qRD3VE@C+cFWhX#qE48)J%jn<<4
zLZ10*yYT+qWQl3_9iPs62XRQgs2u&#M*L2)u9bt#w*jZ~Gp04;j{6aWE(t5aZw8^3
z(^=U=AUfBl>DR5_JBP%-<1Rw*X(k)v*4Jv7Ita}c5tyjB3H(-GY$yE<Pb3_OvAEfa
z=n~tRO%n%DOwmoosgKv$(cPj-=YA-8*y@x;*CJL<JU$~y%p{4#$tp+Q1xBe^!dMH}
zqF#~;jvO0Uz$@zAaT8XX?01vJjVN*6?b#mq!A=gxp_$E#j$i9Hw-BJI_UGXAQ|hgU
z9{Vz>qa>l_36*P}Ir?b^e33CrOdy)E;Sq=zfSwBf3Kx=`){x4M>h5Mjj%R{~!h3ks
zM{OqsOMLBCM-`RZL}p}4^Uuf@5U1-_tZW@Tl+D>)(O}Pymj+a6nS$D)SFQ`ow-+>W
z(0{w?gYXdgZiV=5I{>E3dW}gsN|n2@r-zKDwY*`ZT?1Qscfw}n%e*LpI%ZkucG*#%
z<h74cX{uzwv1@SFKC^?OudV;-N>%s*QTRoP;z($1N2)tZ*IJ&ZE7x_+FP`rhTW^sl
z{#;@#yg1|YP1ANA?%Q+#erA6YawaF)s6{TYboQ_HNn)JodQU1ymU=21*Gt9AotY+B
zqLeeaXl=JRWz%q=NtUYR$*{_7LK90ldr~()Sg|vZZ1!QL=PQe@XfIi%(C`U9yUSrO
zJI#y=ou9cc4G}qs2Z68lMCW%?#)J=vh&Z7?I>YX&?GF>8qpGanoh!MAfb<O-24zi4
zeF9d&oNfMooE<`OhO)+&jxDR3<1ue<xWY7sGBuNMUA4WC@-KKXtmtUsc9`{0-RH9w
z0(uz|pCiV?bCZMcq;S5Ri7g!^UzaTi#`N85{aLh(4{XjkjW0wGnAFMb;Ylb;n2bL4
zO+NZdj0ZjSzz#o`0Hzj*vv&Ak02CRC{8Rg_p3EO1e8+;?pxT@6jSKfJqK|WlxDsKa
zH0%65)fdx=Jtf@Y?jt`W{Tn-bsTJBo?(M^-S*d!)$9V7|mPoiFVvf|7B^kF?LQx!c
zF24?cqEF4p*2IStTl@UL3cZ7L3x65us;5!pzX>>s42G|KAHzvb`e_rB=UCeBU{#b8
zsK-RdR4l9rocopmY-_}uaWSwfbI1Ps8XdhD1hI`TKki;LZv2pnEEd;S#QO~F<~RXL
zoBbpL*+iXQ5qYx8c?IN6HaUlRjJ%o0<Vc^o7E}u&F1Vr3l1j)(?*zvWi*rm)ZoFsc
z0flWwy8(4Z>xH8)H`~21V>m7EA`)F)0s~h7wyc76I8n(nPIM6LLU97)S|lFTW6|Ax
zQN*{oW0F>K{wSjd-*&?V6j>eUjK7Ye4S6q6Qgw6sZenrejBpQDJ50SrK3AylN>w5I
z+e!cOtpv(Ed)1u;Qxav@wOUT%dCNIk2QiY0d@Gf-g;mmbF3Tz{r-#UQs;h!K#${5U
zTO=Ct4rk1mT6h9(Lb0uC1>!!&;TFQxrZJ>#oj~`$30$KfwAcJRb*|S6EPs{{s1^Go
zqLJhBdfryJ)8o!xsfpsuLUq<w?wqo08vPcP(6jK3)h=9oY25ldi25CSlEbkCrS~8`
z@hH1R^*=Ibfe$%0zXvW7Vm0wct?YyD^B6Wq8a%Q1iTLIfY7zh3&;wtAX%Gaohy>?F
zjosTXmUT$+;M`dF9j8Dni+Be@mYY`T`X{t3^|tw7md&Fj+UB!0uRv|O;L3d&;fzB>
z^4kt2ZJZFzkeYF!@tp@*wLAN!wO3owYK|wy<YR+P`s{PdumvBIH^m>vK>>qFL{2Zd
zDGtZEGgBbBpr5LcYM}#Y-_;}l&3&6q+^Sj&*EmTc7b<2Y9@TLWG_^&?Eb14j*lQ%Y
zbHyge7>7dw+qM=W5!4ej3!weET5~;Bo}|N=JT$BHWylpRim!;ww%H`zVL=)YDwAiA
zY|pbuHM+9`H*}54i72zjj#I+zyg*&kCc|8Qy%MY9ym#3T_?A`F)EZUIF{G$Lp3$Kb
z9WOcYS5DAbDER%x(xHxRRlfFbxuLZIejkwpVb*BwpLHYtFJg{qHH}|EI<a*j9<<Tw
zbI|ro95b_g2D&*?plM-La#uf!Vql9HS~S5}3->Q0sp-0-hyRA#@5T{@gn5-~-&hGz
zz0SI(Hp$EEcc8mjqvGRYmA>?Z%E#gWy{-iof?{92uWY2zbq5R9K(f<_X*SctA3<ys
zzZAq^Bvml3!~BtcnUypl)~XifW0}#2{Hf}SEl9jr=qm6cM%Q$3$d6<V5dLUlcVd}v
zR3pCSW9o+|@Kn3j6m(2s0kiuu-+8+>vNdJn87$Bx*lWB=NrSoYh(Y5=RdA*qu<xAH
z$lX&BIB#+Kc)o4Dn%UaL&q!oSW_Eee^0mVACl@zY{|34W)u*6IqEIP$Y-O)^bRLKg
zZ@<{6J}^y!@g3BU+Kl1+6aNBJbzqfbX{*ZaMUqdA@`spjq?%82+7Q|l`xNER;yN+#
zh+P>#w)sitG>cNHKTg>A;F#_Lb60+W3M}MZP6K<!1on;MUs5K-ariBywrcdkv|*XK
zUC?YY->eGSzcuaYxUPXVLeGLTuA3*x{o(GE6e#6E%JVoyh>Yvru5;imzAiv~D!IK%
z)S4q-V>^w+%`nCQKEz+-znZ<*v%tUPmgAorpCf^ul?E^0_j%@ic|8!pPhTUb2w|#E
z_FC^x!S#F#=_EL*(??CBVE}v&iI&HqoC9|`q+iK=&u)3NJ1PI=-`J^a52>P;B4fON
z>fVRY)(A%ppoh^9Bos#99;xzo*z)A9KuNSp|9(V@^`5Ch9!<&~qjVlt=c^NA61@*!
zDNdy662`c@%e!bL$hS-iqy|H7igp^`EdtrR&G3Vlr`Rqxc#hv>FusRr{kVpau8wW;
z*0z}kXJ`_Qnr^Gh=ZW!jw;eT<^>dqR>;8%T^hk2!kj@`4%&GW&3ggx|KKKe?&tP>g
z1|y(Qc$LUwOJw<q-e-IF?@(XQCf{?mmPcQ1Qp`IXozBHlprLm_L~wAB5|d&soO!Ms
zwpUL!)2ea`gF|S{<u&6JJbyXIiM@kH#rz?_>#CYK3Ik<@e{*5`=yiE?@g<Ci&+*3{
zWq-;fAoOgO;OMnt#~D;7>iJii-{7}6><wbw#seQm#oXIwNI<OSnc$c0iQhO~gvC-c
z&|XPA*Fc`D@z8$?#jiiC`+R@y=Am2ri-+vHOn!Dk^>8R7RlNLp*Vob2!3~AqpOhIv
zgz6Wiz8ASBT7*c*@kiny35hf6EDA)UT(j=JAuO_V?MB~6tcC{1Up`gJZHC2U9?^Bn
z%~>o(4_7IYN@tk47PD$srACOi9=T^<bW;&O4LHTrL@oUCuk2|T>EMOmTlc9W_J7Vd
z3FcY3JznsK6sn^!KC;dS3~4_i8Hsw|-Y++9u@hRk<?SkDW_8Xs`Qmh+y|GZi63JT2
z8!B<mXJwVujV+<}ZWlhA?qRg`f2~@H<jw38XU|(lyBx8B5i+>Kj>>ZTh-jr%IaG5$
z9g<aZ$|TMFd9`nw-0$eLe;RDItiSRKK4$iS-z|2+sWJ$t$+2nY8rtHN5WqJ9IZMg=
z=H-M;8BybxVE{u|rsWovo|)a<oRNRft}Sy+t*kKY^#=^(v7N%-vx+%RUE@<`h;L!I
z2IGu9*39{j-^9o@y9N4&Xx<q#sG?vv!1)fXO2_6F0PU&f&h)FJIe7a@^`y?%g6oig
ztX*INW2o3Ir%m}}o6@iGw>`E0vB<>h6{`#_{!nuAot4Z==k3JTxGJ(2C3w@vT&c4b
zN3Q4Pf++pqC~{L9EKLMcn5o3h6kjJnI1U8#?r*g~hps+Zy_J@L^gD;=Z=Q}XK{f?T
z%Pwv2w8$a=pN&2MjL5}u@Iegvd(vF+nXG(49bQ76h(dn6ci6hT4N=-=dz9We$<c?8
z7<cSQM8weJ{o4RMiNY{VG=Xf$(@D=Yb*J#|b66bP&!lLl-XlsxXOwBo1BdP<cpZ0a
z^C7hNNeWb{<`*m0bdW$iW<rG^0;=#HDv*xUIb5q;mG3E*uePYfd|8jdd;(O|gas`u
z@<yYCD*VP@6?QjNpuu5nFLV8c?ByBq;ydTb8?CJD15{PWP3?^VK&es&tJ_*W7A$s<
zMpH}kYS5-Lei_D82LSr+rZusS?d%<E0(J7&jzAp#Sd4@}(oF+zSQbPOxA=OnWH)`y
z#O@5|rWDN@ZZajw%s_MG{?j_sG*hMAVNM_OvZ=JUykn{doh4QUE8VRx`)suDr>5`$
z+da*A`q}UIF@16*n7nU5fw<#tUEcq;2g!k^6L(*nkdZ~#tbixDzx!DWaKnnj+&>MF
zDh{ct#B&P1yrV{cVe~p77vhZ>MHR@7s;gr?d%kB|f13Pozl{A7qfl@ajYJ`)9a;x#
zqg3P8DAP!AH)TUeK)jKK9h%f7Nz+8mUNh4y-iS+GPg}r6{OTDsrK>;Nbj)|UqSB&*
zD-pEsFJw5k+bD!Z-SaW^g|BVxLGWyb5(I0C^$XjJ5u2bCzw7i-50zSu>+)!L<^z`1
znl5mTXdY!D`eQb-`6bJNMbE9^c}g|_3KEK_3Y8<!@@~6E6WXQm?Iqq$*gUIvFVcEN
z(|-NzI32%=c_Gw0(-T&PMcI$C+4@yeA@}2(9F$f<p11zPMH{!7<Te{Q8-jv23Ia3~
zk4%qjhSa@FCxrrFVeAw_by3}C%yNU6DZv+PD_XQRuW-}7-k+1*3D(|yEI~@1y9*zv
z;5ci)c|G$l`uiECunP@9N%pzM<v^&3&qEmYPjxtZ?p?>x+Bt=`K)SkVZLX1m*pKn~
zwfFxRob8aqmX?Rnp@@!NFdJF86pZYh&8a*O!A!*Ske&N=^bV>WS>_;BI1wKDL?qzI
zzSj0lmGCY$SNA1;4d`j{To-22VhzhxWtcc}UW@FMC>lL@(A89^s;?d@9o6vS>P9K;
zB)t$TW?BB@PdZGd$wXoP!I&?dj*HV-=qoz{(R`*|q#-#WeG;=@SG9H4*`)Amx0&?N
z<0lv<D?3pvn^6_J-wMWfU38Z@;<gu&;^*Seuw^)+RcetiG~!ZJ$dTWlr-t9kGrpG1
zLQdn1&T=}NaaxD$#M+X{_7M-#c(yD6M<GPD|6@*?1d4R=s;T93(CVo{z{S~Jl-mK5
zeSt0?Pd^Ruxa)2=bE*QS_Hn<Cz(WvUeIXY=U9siFlGF1X588Jj_;8?*5stfn-Kkx5
z?Juc{#g6Og!Ww=Fn#IL3m!BZ9jTTa=>RG^P*KXg<&*+?~F7ij>XH=x0Lya7{b-om=
z82)WYZRJe9)CkMh2VD-tRR{qc3-<`wHt|~4!-uT3;+gj)n>Fqd*yS8OY9~aIYjB%W
zmxOba>3IKVe1fd{%#iekFHH8<y~2+m&xrA7{kLx=V%G;`r~SiRv{lLBFG11N|KsW%
z<0=o|weM`(wp~+Awx^mV+qRo*+iqoBt0vpFHQAG&wfA%1pZ(wW>-BE^E}Z9i9LM*P
zi1jJ({XL&ixk7lsWZ&f9HW_bELg)&1U}7?<fa-XE%s^|)-waAO!d2jD<Ms3GJ}qq=
zI@l#oprUC;URf{OY;5ZG$hOQGr=TN<Olh~Sv}(0yNYY;;HeRm(;jbN)LBQR4KfKA5
zFjP#Zvp$d=d!K~p?2>3%{-f#l?s<yPtQ-<cAKVlEoh#+44<wv3&>hUvk_?7`L{5~s
z`%@%WJB3{zBw^>oigXRD-{;{Oqf0N5p_gV?2-PvSAyUIZ!BuS)J)IyuQqt8><snLd
z;g@>SYNM+Q$lN~gI*sra?sItW)f~s?=ACEowq$U8N!p{>p#dOVE57k4CUH7?#fTTW
z;{%!DZ!72%k~Z4xn?B~k)lpF4H|!FEHbnv_Me<S<$V-m@?sH6Z`467g^^8Z)2rKer
zi#^dW%~Kx;#7bUoBL+4=l=LT$fb>C!xdmYT=UCW;7+p0Mo{i(<gil@+Hq^L<uPZ3Z
zEBysV@28aeZQ_#f*UN|h+s;CyvBHHQCZp?27ZfJpzLE@4m_2(0mxfJx*be4qkmyK$
z%bd)AoQqT6YA<3=^@PMQbUOzh))72*-fsQoMunKB%QwJVuWOyHJqUGf{-}#=)p0H)
z9uLRYF03|eZf=#V*OK`0u(OZ<+V;+K_<i2h-1TfA&xrKmXcekfr&NR>j${B)0a?2V
z92=gGw_Ij$<!7ca!*4v8WRpcfwPPl=bEtn0_aBR%=^iw35E5m{7g%rnynqWBas0uD
z%4V3$b5C1y-YoY6uF?0Bw<oM6NIqU~_w|wReUJ2$K7{#K$uy)OVn>h1V#q6nyQDZY
z!|){!DPB6JKnN#ecD{2ns`+9EdO?fS2s?)iIj0b3ouIszQ|gCLw#ZC$qLLn2O}`p2
zFX#v+wQFK<#!3*I$SWKoFcul`p?Af+4afj_wAE?S;bqL!CYf`N^WT%da_am)&Mz*n
z7A_xNFTn$oBnUnqdp}&{-ne_-kE;}v;tEy0ScG1f%2-^>)fsGS>I`E=<z(?%{?J*+
z;{2-HHn;j9UgqUl+o`%4K4<0R;Q<?zF7$HTUUTvxJfq1WeE+d@Ls=8L3}+Qbz)DCK
z15FTlwnMtv;*%#Wlz%dr$r*zbuZBltWNDcJT2+pk^yq)#416^Ga@?>-vh3UBH*IDx
zpS!>D*1|DvX%?MEXa1+GO?mPxP8T2fOTVbQbGYy7b42LbNsmfZq-zP#?Tq?i9B~r6
z?IC=?XYSz_uG3bRoPM(Aa776i{1r8}=Jcuj^{?&4<LfG(E>+0B($eJR>tn1<vnti|
zUS649=$n)gNgjc35@&e43iFEQHc+~tv&1)Y6?~&{@jQ#M>YPXyIP`jjBj97E|B{M}
zpN;>K6Q+l6F`NepQ4diBCwOflm79)?#|)*wl0%+5=6pQq6(w#`z1<~k<q<wAMaQHa
z9&*W5&{u7I?QXfKr(Wy3@0&Wqwhqhn*LypJ{Z#)rz6HnMGa-{VZ||lcgkfnb<jhM5
zFA@WS8fh?T9LU{opRLXlej}k=YOU4nmuo9!=!!`%HC8g8qnAOOi~7&vYsp^}FmLX0
zk}%*5N!&Hmua?w&T|9TOlKo!i|Ec(S%GxVQlTDGWm^@{UPe5R5>KSs|KpqAIlQ3Kl
z)UmL-49NDlZs_q7!L#{%((zY(sh4Qc-e+|1i+p*ZFX6KE)Acw#9BhMAFo?mIJS{g^
zoY(U)@E|WCYYnpf9i-c{inx>yH}_nbDE%XZ8)rtI{kun4-RJuCO>XZS3}Aqx!f?O5
zW>U7Wb~%d;$%i1NhqnS$qH4-x`NhFF|A%{l{UjfZTxAITSl=saeB~>S59#`iZ}`%(
z)M>8xZk-2v&M?U7)hBltj|C()ul<>AAPxhN)j^bog^!^ezX1K;^Hr)^SGQ{mA4iG<
z!E)J-mA_sgdp=`HlkfZ**8&u1CQcm#EBetOi-QlPnoR9t*UgiMRAV;;V2KvbHFtsM
z*K?;_6-4&;!w&wL3n$aRQ3jL_6l$srVMToYRK(-EUVVy^*K4ke;6xhRKXEyH+&SFj
zI(%`%hWkvKA`UvGdvijUqP!-wBYOnAS`h8Y3TFJU&<nNUZP5E58Ywot*rz_u(=Gc1
zqN10cv>49HRHH%B36h`LUo5O|P^;ld$%^lGAROk^wR@cKcL|&$VO!vl64ILH;OhU*
z%RIAM{v~W^;}^{rGiKhcUxkB4(65PU9b9JYOKs_9GM-8!mV$1JW~0b^1z4gym15r-
zjji7yqFzM5Ht@=K2bnpg-tSm@$6oGTzN3zXj_7URgkodRWA`&ubIX{Z54Te!wWhC`
zR=@S)m^zEcD1eUn9f5Q{xo}*Jlm~Bq<XEWTa9Yo+j$aWi=2o{nXA0Ozc$cn(Eb^6Q
zRcIr|>y%Ec1uf(#8~I=Gvj6E^fE8*OsYpQ*yhFP8;O+4E81^IOcIwDiYcE$dc@7vK
z|2}Mii>9^o)=8Br=k#1Wk;0>a&+ikpubJ2J&ZaE%CWu&8`m+68J8SLNXn##*YzOk4
zH(^2`FZ0CvFws%prjo_Qy&XIm)f^4)yZudNXlb9S`vMxATv+g2cpF(Tk}n(5lK-=+
z84)xkmliHwH^w6&*zGvUqr7nMaC?I})6FWtHZnm>46`MvQz>@*NBXC{Dc}?k-qpJ)
zPnNL8<iaXAwY(2AXO3SzB!er*Emcq=L2gB=vDWJq-Oz-4)6Mk{KjbcfL9rZqN+VkC
zF+iPRs=s1E@NscLlUY+w%gEk74G1Lu`^XRCz=X4ZJUE#k8DRx~&jj&Q$T%dV3VhBE
zaDMG=e1{!Lrj9bH-RYD<^Q2QwQ2IhBJ3g7^?=Si>?qwj=KBTY<)$E6gtIN+XXH-JP
zvs*SFqd2|mC8WJTSa(Xyc$*?dKjUSsf$1&|6*P7$tM7S*?OLs+rAI{wWvV*P%;L9J
z;&`GWM;v<ghOg#CRNNbY=J!8$>A=W0xcUB(0jqYU9@ZcvHCDla&sYA(^@a*`$e(c=
zYbykbtyi0Sn^kp%N_Y|T&!^gorp8&FAsR_)h@7D45~n6%eP`$NzkhgdZ`A!aO?uJM
zj-MAMJSaTywZGx=j(Po0Y;uHrjydt%i$w^k1KGkZ(1e*G#mVSOoUg*n$xybyq@|@i
zUygpAx>H_lb_ZH;EpU3VDSB+5a$4;Pp`+?fN<MCuCs*vK24F`mE>gM7d=Em2A+((F
z<dLEX6@d?oHRcg&kFj6!u`{8v>U{qv;2#<*R?W8A^MM{nnlOQa>5j~w7E)+#Zyz=`
z#@=5qUPg=|Nf0P+1uI{|Y-AhBH7JE<^?c37GYGtCMXS}H8w(p1gX73V5=BEJ9B4WJ
z_3!3yxbJ=T4gge+bGyRCMX7Y2^P{9ce~6JoiFv_gioN2~(45EsRDlN%bKeDC46QC*
zv4-*1KOXGK@k#Ud)-3dmjeXOCRWqfU&Y}#oHzt~$dp)>XF$rQm+k4>t*RU%(kSX7?
z04|WX_s_*aysvHc?cs18d)QHY?T@NUmj@Nisg0di!lH6XDx~MDE%MYvy6+;3wS`?c
zTUYgLyi<p(w5iZ9`-Tmh8@lFa(+E57ue|h^*Z#&3qcdT!NI^s)o2#@AUHcsB-wW2`
z${&%IG>YMuoxbt#j3cskPhVU>)PVFM>LTbG8rNL?+i00(KeA$_^HJ=Il)oj@SO7tY
zqk}iCVbX8PzwKl2R=Du@y1#`m8slr0Z|ML!+rlvrNEN#u$U(EN`zPKW(Zx&0%h#V}
zb`jXHKGY~xKYJbkZJtL*D>8tEL)bgAntP`2ZYlI;G55^OEbrFd>!{>Mb3XaD0G%Q_
zs{Dz4A0Z>G`_J&}hSrv8-rh%#kgd_5H?25T$Qs$jjvd(#0V*;_ls|X9$>oM-Z|Eu%
z!q{SFjZPagygLyWFrPZ#{&ut4FkdPK)BQC@g2PHq2Cdgk`IuPcIZLj{ez+|gW|JD=
zjKD?y^I`vk*Z)!}<G@m%;eJOYQiP0*0$h-gwjDh2b&Di+;MR6l4gKwwzVj}lZGWKO
zZjs*aZCo1>{IuWdQPPlG=QnNJG<9j4jIQe)bD06kZJxbz5uQc23+isJIh&lpbiHl(
zo;&o&wrzaGShzx`#mlX)#``lcyUN)qT(qWTZ|T!x<{0q`3lB5abq|9!)gz2+mSVZK
zH9FBvgWB)oAHS_Fi75f$*JghhcpK1V_KHgel>suS@$C1f&<fh%5R6EP+4yiqx6CRI
zDhB!^BP!L^2Rq*Bw^H~URtxrvH{ZwhbK}vp9qU`<yVOj)&Es(Oy@Pk&r*_OtU4vdi
zweBFb5^8A+p?~Sl%UR&(-|)QgJ{*<^Th(--bxDQm(Tqsn_{Ce-3?Eeqo7<cRpbE7I
zMe5<Ca08)fmRGm0wJZMIH{_w}6Hr1X2zB@d@<nOs>l)v#yAd2=EDtIr7wVNJFWWow
z2R4>!ur>jkhK)MJnzq2mGjqRBHO=%+{@JD5vX>z0^5%USd=lBII=3);oO$_wz72FH
z$SHQF{HY7KxJq|_fB!5g%*a<<`&CoFROIXey^l9;v%29wvS@_unk`i&Z_4HI%K_aT
z`K?M`#z>RTE$YO;J9X!u@)tUY@--!+8^6hlf<tG@0Y?sZkPvUbZE9zd%y{j8gJ0w}
z!ngX(@Oqe_)z7~%O468_jzSd(XVbM{=;xQ~4OGPHC`n%^!~L(riyq6xRU^>Mf{s_{
z4X_;!YlDd+<;f6w{W2}ySgUCwY%_W<n2^+e<|W0#WC-X2d3i-8Q6a^4&lBeN-BcM#
zDeT|5gBQOE<XK4ZR9vTezQ1eX;0hNlSgX);TaqkS{E|zVpzgU#H8^VPeM0T<+=DMI
zrOl;X%oYSb4u!J$U;^uar(OB>lTRikPrxEuo`LswBr#2ml{9+NE2oz`?(w6j`{S6d
zw1aJbvXDW9P*J1MehPHY+K8gCxk95mNJu#H(`IekYuh5X5O#Jk#~hM6<SC8hr6p{4
zcaPPrt#j-=ECt1Bc0n$wYHsZjX>!k+$HeV@Q=FAC$?N<7mEqW2HmQ#y`c(F3fkIM_
z<zWBr^`PH2B2DtEkJ$nV;KSR%`x}h+D}ijm=r_egh?Uyeu;4e|f%$P_LHzpGg@PY7
zbLA^*8;rMAy{`iA!8;$Yx`G~gdwcIEJ*CMia{FcwQIJ>NT>!mt0Oa<|oR2zVYVpbO
zNuRd-W^DYGk5m2VpBpy1$mJ~SmZ%n-Sn{;6$Udx>Jt7vlO>?U=6c77t6DKD^o4$Lh
z`}?wtY*K1#aq%vw6j;maSI%3J^=?ZC+;5r^UgIq(-`Fz?Xi-`W8?|usf2VD1ysZwj
zW?PDo6)=)fBM(d=2Q59zM57(hPv;X9>o}l$1q;5utAPv{J<lFH`|=L&7&?sEbQKY1
zp}jBJ*P#B%ph&yI`5cr@l%4JR07{OUp>{ZZi*FKM+bY?`q?P0aad=AVHII)=>4{U?
zyF8QU3Y}{8fz#)8y%E?z=(;*u9N;~iS2KC7;$VwSdgN~+^nc(z`4suAH-m$8fIIN(
z-Y?2+?{j93Oso!Fk_xI^rEt)6*{M3wOR|9vObnEbo#QWEKsj^g&Q-!yZm(vp;r@X|
z%nUZ%{wneV3tbYVmG_;9|7UIB<@t=#k)BsvaHwo92R6+YQ6r*++`#saB31Z<uncjK
zSSZrQP8ju2dm|#3R0NG&kk6k+d68l6EaqnOApNwN;e9uJ=b+_8-|Xz_JA@UyWR`>f
zy+HTvdb(BH`li12%jSwJhBl$EaBOOpIz!3EWn**q;LQ!)ubr<1-k#UxviW+++yv5|
z50>0^aiLu5)QJRHvv0!AX>t|)sT03EFCSNAWMu;mXU%_vGmhFuty;stgz>p}qcB6g
zy(#{gzqHN@4@wms$Wk%L5QRmxTx9-z?cil@9P>vW-7~PsHPlT%G8HZnf}U+yH7D2X
zHPmADb9rNM{U`Y|(<qj%_Z1RruWMX0?m?lYTc@d-Rw-iCi$>&?Q7Ubo;LLc<we#a^
zyw~A8b(+Jr@$sftQG4I|Iz!npZoBIl#)bg3!iZNaMnZqV`ycG$#jsV)YRB!QKP??E
zRES#yfemSLHA)R^e->Av0&mt>uV-TgHamg+O*7vW`ICo>*l|Kql@yF)Co=v51!qqB
zholq4Fx>fxG^__33YOqITu##DryQE;4x^>lK)i}z)Vl7O=_{1H^n=MWc`O7;SLJ-8
z7{Ed;JNFZY)ed0hnH#4zHkK@o+h_$%3Z^(W;^m&=KL(qiRQmM?cU`P<0w<*m%8M-L
zoTN0gYZufB$;sk^z^2!U^x&|{5H@+<GnPxd^38=#Qv{jwoI10NSyar4-|@2t{ZQCQ
zzGq66|D#^*R`)o-#iHHoh*IghGq~NxMJjJbw!jP^`nlcvvJ(U(37WNU8!dsL8HTM`
z+DKstAEY=QzRd1^7_KUr7dAV8f$C0~DOMJgCu)Nmz^O4$S6ac}$gZ;a*Q4CFMy)rx
zPr1e!luPI~o+@0dTT!4MmYKPL6)hQO=On*abGzv>b5Q1Nu=>~!N#Irm;$o?(A!6nE
zL7?+FDB;g8lZ7CnQbMg;?x)Ng-p8X~UaNHZ5;9W2oWOeCK3aTx9L3TIDL`&u|0jnI
z3mz;kF1TKKb0yv8^>F<KgNaDW&LucHd?loIAYQ%IGX!Wlxls!Wc24y!mL!sXXxj(y
z;iP2o2`$si`t;(Cf63nKg5}j|^7%#ix(AF!{#cLYz-84nwDO9yZ;QOLlzXEtHMh`%
z+aAQT+6)5MTuj_mV~`cfANhsg@0f6V%MKwd|586VJxv1{3X$s`t-v3Rh9=Dp3k!>N
zCrRS}V3mTv=kiKXHX|rjw$OJkxSq=c>!agY<agNj1XNpz(8e5DHw&^l7hGl)asOWx
zPS4#FY9@ZZ+0Eul@$VkmA5bVju2w5FDfZjDCW*HV_3PSw@QyhmWuRNySoFZC{t6Qp
zny3mR8z0*^XC}i=msfaM?<)?XiwE8`a#L6D%<3SY$MyEXT+_)d6;YbICD(>D|KJQF
zJ#}#KHxALb#6<GSCCKUgtdnA$w#W@;xgUdlpCp`3E<}lMoYUMqc_Vgw0-81zB<ibM
zYh<~s^R`R$C`bWFv&-`rVTZ6%UdJ83WB|U$Wiu{&iHN6hG-+w>T~o`iC|m!c%N0C?
zL8_}-ICAaB8L11sovd=qJP$!kJlY7iTSyri4oC*RFidRfeKorxx^i*i7~rharP6_8
z>x{+HuM^-{<#Ru~hJ@6euXnbIp76i_&XOlQwZG-lOi6dh5rYk@1$0RdjmC{A<{1hk
zmkNWSBUX<viB+X#4DRX_o~^Q2cxyWiy@(h8S1WQT&kFo=5HCEwr9+5>jNQ<wB$R&A
zzjl8(FI1yFxE#J?*7Fa?$qC|8`-b?8FJUA|k6r{(9BW8o*w9U4ga?%5#pb`{2T>&Y
z%@aWksp?R=C{0~mQ(M;*_u{cpOBiZOhH#zk=g&6~HdhV+=?mX`x19Z;SDBotLfcgR
zn(v2hy#9Z6zW*Y4-3OW{KZDpBIbU!7r2gh%Hz7_3d){$rfC*=*HuO;T(%!qD<?rMc
zv}%HPsS70tz%)l%$P)4bkhwUc>0anr<H_smdi!|u0g3Tk0OS!y$Yc`Fq`zWeao(G!
zF1g@fV+&}0LR^dTJxr9(;pd*&+L%B>F0opnI#Z82naML3q+W$b28p6JrT&*~hK@<F
zHF_4tQ_|_eA*TW6O;g)SSrwW16gy@~b#*%?-p+C6HDfoMSi^(1cDBwh(%18UB`z*(
zDX_aAUei^!PvSfC1|f5qsY-i##;1D!@#oecyAokau1o%X>;Fl88(`lQ2NvbB(EZu+
z%WE%H{$GzIg81fNw1U}XG`B7gNXhs{6?|Bt+uc0kefhhBK#_BMY!u<emHh$_`;cwk
zdAhKK(Z5uKPX^`hA6%H210{Zv>(6k$p$TcWeSi`9-kprKM4A#tii#33xb&h6SfM;v
zdHIQd`cY&z`1trxyNS{y1NVOWG7If~x$8Th4idcE0}jhK&8GlHof0%S>f8~^k~eOf
zdj8?8j9?(M#xGo}7Ldf_cJ*K5=4FELRB?Wz(My47-#3AsgPr${uRy|;x+HigAA#r9
z^>%q+k37)|TWf4XZBD?1Qu1Jfj^VyOL-$;5-rShv?*A^q1Yxwp!+jsx3}`p?x*X26
z=QvS9Q0K$%5F#@6BH{Hvrf9wPmt0C(;u9cR9VrB&G1r!uDC}<Rz-Q`i<;$RgVPOsZ
z+!@Ry76EGO(h_<I6<jM0`ZPDp2ERWfr*j0}B27{;d=306k+jI6rp$wfh6Fr$;!oZb
zrm!Pxb|2Jc$Y8=7GzoYe&_3e*Mb1YYOthDj{gi83F6^rtTu&(Fd<3dlK?#{*I}jjl
zuCi;MOF#hD_hv8reJ}r=p0wwUa{Wj8<6{>S%Z|C9MeJ-ComCGSi9ZHNZRFI1FtOUy
zGyaxucE@NSiVR&MITNfpep!Jix?<T~J|I&yzFMzb`Kw@VumFjRG2Ao^_B0F~BSJt=
zRct>#c0ph3gjWgmrtC=ZxJ(&@n5<KsB%VK8{O0pla6yf8v*#ThcH^!4uHPo*={X}O
zW8d&AKvNnkJPeNjUlaipMK*ldrdbwAK)4YMj6>GY2?48N{;_E0;*yV>@1I*=Ja(iS
z!XRA+A{wYMvDTnvXAIeATcE6*_4tU`oWXSb>J1{w6DL6Rqd3|)+70H0u`zZfa&a(A
z)Nbk@U*{;p&D+)4dExoG|GIl)V#?W@lO_c$p;Rny8f(?Hp~?6jaq@G>*&^qi1cTy$
zaXL$uvCC<E^O}$lC&o`ru9Gv2ZMpwrwnZHC3Sm@K)!r~Z5ZEyRO=1ueS%c8*eXuQ9
zHoz=bMmKQc#~Av8pqO)hu3iu~L*$HL*WGiqZb+1<mh40GUv9h~i^U07+B>`3Cq37(
zs$LDJWWDKT00$@g1$lcvC}BoMyv}2JE7U9Luz%FJ&+(g+)zP7VMokT!-ZE`(O>f)U
zHWkFnpk{?WBsQ{{I{EfAEbFWf|EikBxaC->Y~mgO`QG~v=8@EZ{7s-YNZ9@9#C{};
z?|8VSis&MJt@84IwV3q=*yjI-^y4Mp$=W_`gwzkp2-=9X0-)i+eJ;pKmO3?uf(${r
z2aP(qT6XXbE}%wNzZ<IZdV}*hO2t_xgcuml9>y1?vK*8^N6fLwzzHBMDJiu>mV*-n
zur(Cw*r9oni1L#~t!HQkv!cr-5fT{M0SBYSy<8_O{I8I5KkUXQY}CiGrB;U9JwGwJ
zvfpesbxzhs@+f!(2e+?;SbH82UiQ&;KCYfS-uL3mHO@VIq~*LOCzUwa5@->`OA=i9
z^>A>n6DN*d0Byf>T`OgZe;6B*DXI-}F>qIh*>e_c*i*O|iYaEnxSvoi#v3mp_DWg2
zW08;)l`Sl3O0Yt2!N~&kO>As(LC7MdE^#8lwCkG2o(9nlUBH@cQH%R)r2diPL8V>i
z*cNMu)$3z?PC~2y<`}}k$Ko?<?O)ITtmiuhVnSoPCQ5dGrzf^;OF+@D!#%yh;u-F+
zvAbXjZ+%;4hT)@RfzoiI=w{zlK}X*k!=ivwIs4%*|JI$vinh*m6bVlNz&H<Qf4l68
zsja#B_Y!S;aq03%xO2Ef%gV;}t?%g=Wd`)8tobJPejIyvKIgVYQn<LHg55!)g$C*F
zqL(aOTrAyo2c+x2nn4IF1Dh4t%Mz^ewO`PTO#10)uG~QxKd`{Le>!aq2G{)#pFA;n
z?nXPx0WE3s(=cr~xK^t;QY-Mq87;BMD?0591zT2V*(Q(|Wp3@U9sVn?C;Mv7pN5H@
z`ac~Cp(<*P4lYsgGDwEHJkhJ=w_4YC>1BDU)ok5WRE?^KaB)&)F$p09!;_MRMKd=K
z>3*Z_nyGSGyCM9}5&>xHR5{bDbSG&r$%|jj;cE5zrC!d{2wXD6dWRM!(nGLLwV>9q
zf8!je^Sf;G+*{&hTRyqyKj?PZbF$xgfwN-@6-)F64nIx9yuJCzkker`mIQJAz-#L4
z9RK%^uh_(^dCu1EGU*nvx~6)roVz0BdF)sulBprlb9|ro`LyQ?w}cglf~bt{1Kyvp
z4Pn^V*`Gea_{>+GLYIo#?zo4M&hmXYz2^a^?Rrz{Ji=qgFP%2uHz{_?u9cTJvM7Qt
z4rE%#vjNQ|o3-sg_Di`IIfhjJzQ~DJ{tmDQ=vLH9%>RiRehuw*8w^Q+LDy(?`-Wb;
zL#@dx8$sDV2~Szv%DaW}>O?HeYf%#X^3w!^CaW?U6PXS+3VdvAzFrzzW1`2e??%6y
zd_m|<Au;<YgQFxoJPzDo;yw?tlYKts<_In>Nu7BXYM*}B&E0uVH8^?X?TtwgivRw`
zvGaBA;B<e*-DxDj!Ycvn8M(VdIJ#qDCn(fAls}JHYT1a1i%U?_>co3mMx{4(-dx)*
zL2oU$0~tzM{vmZ%HABMbg?fD9vGl-O<OS-3@4X#I#&vid5G;8kGrG5Xo>{Tve4CkI
z?H&I+kKnX<xO_=%ZgB#VIza_g&XSGg3c+B55fkj2BAb&$+mAj{f+G`rp?s*9g>!gN
zt%2{R#{`}rWmH}8v$pYJ`c07iTO0uqZCqCbpA>*g4Zsj+g&xJUrH#B{b%2d<I2@iR
zhfzk|xmDQoi$Xw1iXpqW5jeB!NJ7tCCnOwx<vyBC^R-qzv%>o;Q_wH+HL7r#Zo$5l
z?(%y71XksD+T|njXPN)w4rq4TJ61d+I04d6We7P0xW%13Q!Mg*NB~$96BD)$4#_ho
ztDD*uzP?G>D?$=cIP}7Uzh<j>+ue^S^}Y8Lc|67!pV|^GzfYT&&N@ZJh#a}#G0L=5
zhNOSN5QaNGNwV1?^&rNdM`!@2fh|547Y>~KeB*?5-9w|ROq7(A-Oqhpw$;_M_8OjN
zp%#`rS+)-L7$a#Kw5B#5+4if{%USlkC3BVy&7Sdzg+J7Hji~Uf8l37x%37rJwsBj2
z{b$nA={ITyn0X+>tX12p=LzKW%rqr}_(RqA=%V8k_*+7pWwm-OaGkJ3biSV<DFLDd
z4G3-Fd$;64g<N`Fk+I=7DMw54#ZaQL6yfg&dxHMmZ6Ly`7m7Ti`nub)HF7z(%$v}f
zZeDH;@#j+S=P%!Dft}q%+Pvd;0HUes<m3(1Ss{7P`w?sM1fbpd1tnW2`>>I|wbi5D
z`B9pQc}uba?+m{?1!}|0$ttrf&pXPwMZvF4VpH%C-XFp2oJ^2xA)0@AZ^S>e0e$}>
zIwGI#!W!_scB@C;)PwF=`eppwf&KdnU8X`J;{s?*?HjLMHO4t3H~>rPJIndkx0<`!
zt_BUiymhN%pU>FOEK+{RDCE5z{KctR2)hnMKD1)~1If&sSP#d%3TzqQtl22hr>9WB
zi5`JVV@FSpGv*>TWzE@K-x>%ZhAADn<)n9IF3_<d3z*y`6J?MrH)#xo$g7L!vO;n9
z_*S$2{!3Zu9@pd93P&n|VoKjEo~{xzTu*x{p}{sRbdJ5%e#i9X$?aV|tbY&-vxyT*
zXtTo$8|~ua!qIvUX=oqKq?i&36(^H~^ukO5AI06h-T8>JbYy0}26R_#Tre9PM2eQz
zaz)~p%lZiALE(3SRoZjY`Clv3{rh(DfFE)h)T&9WRz;$^zA=PLV&HW!p%5R7mW2WU
zGN<7>G&XdOJA4c~IDfu-^?ZGJ5U*(;_<*~(lxgX9Pl#IOdlyG0kqqM%(Lr4ELHRWb
z?~XAX8YsLK{6{+zd!JzfS+M-}b#;t$z{{%UyC)B~CzLvDAp--oG^*vV0fFQ0Aojr|
z$PW{Yu?$<-gSf&piFA++UQa+zi-$LgHXOXT(dNZL%<l0!PIzOzIA<zxJ}9?74CLV}
z)%_+RJ7&uw?RJE+1CrYvP8wqr6xqJ9Pxs$QUTwa>84tZ)W>4n$<lJV0!o)QA_)|Za
za5nuo!;~uefR}d1<|ln(R(A1^sa=9+gzn4f?R>%^(kL0>AVA}VRP?~5FvUlU;2LK<
zskCIxk)ubvcCiwFLig(hVGdA*%6i#q8AY?lC;J1a_xmyZcO)mGMaNwqEbv+qjvUnC
z3+7%?r2Zv0s)mQi*X&ZF5jJakhI)N8{w+kJqDK1zBZNy6N>fr3RqLY<d-(Vyk+X@;
zD1(E1K>S;b61ryOBS*<R^uBvGXxnO?ClJ$-sad_#mAYt3(Z~Bb{o($J10kF({gS01
zP~zk%(xY`O>~EYzT1|Y{CZCYWKQP|MlIk5PLs+@R0}HF8z;lOW1{(p)a0c}K1#;J#
z!r+0t1)iP}Vud0tx-<d_<oOe`)@>>rprKUR4~JQ6H@tj7MKqhrb6(9YZBr4ikT)2e
zl+*fys2|#~5hkugbG1LMNG{NU_hkz2$NsX_5Yj1;XRQS4C{JFON4(#7{Rs$L);3oU
zhnJq|u4|gQfMKx)!<ZQKahx6mQ#3*U)ko%Tv}tv>ffDG#6WMh@6o}5;>ABzCY$JAN
zN@0IvUnm4S|6>)!yWko&@j;@BjAX@<>I2wZ2)Ka(@kXLNV>!69NZ#N3Vg6th@F3OD
z*y3mWN+1bL3mxXzw~vd#5043qg9^JzB~N8ti1NVKAbc-=iJrV%`nO@9>}sA8tR39x
zhQMWqbm^T7Zm3Jo?-X5g%+Q9ja=YC-HhM#LFpwN$iom5UZ0;i*uL7Pcg}fuS8020i
zS>81{oh#)=QOS$b;nhA+-ZC5N4#)h2M+T`N${0Xxsk&e96TWLF>HhQuUlF*y#opYB
z?dq0|o97i;W<dkr#?#)Rel(IUT;f|V=~WgI#Vtnf8b4&-B3edF@iUDtP7COfJq1KK
znaMv$P1og`CBY^U>_E2BW97FCWRN&g9y$YEbR=OZSejIL7DiIT^P7B!+aB9{n+_is
zl$}$WM+qqraKT_Ro!Vj$q?xKE5Ybd*_9LTYWDi)s6N-KcQQ^@Wr&cZI$^qCgq6UJS
z8Quo96aKn--m51@7ejxZpkARL&&QzXnf5lycl<D_Aa2+eGHus4qQFi-ZHzI$t6f&n
z{;_t}*b9{8jhdAz0EiW$kB{7|mY^NBke6T1GhEg;^or|vDKO^8)FjPd%{t*SQ~&6S
zy`~3Ij|dcbt&=wlq=>7FMrI&jOiVzG9o*@BGf-}Gf1WrimnPL3J3>n{!@|gjgq*9*
zVXI^VUX(8F*NWHk)^kQ4DDW@C$;G9x>(B@U4*h*)2<-=p)rfYkkZ9YGl%4bO^|s#L
z_C-C&&!tZJpUx_J%nK)dV^-h|T`nBEwfD`>yoYBpr<TL7Z4X22`zA_((<9j209jbB
zWc!AImH?zX(j<o?j&l($y~`t{i+kRdbE;vvIL+!&%@{;HlERIHeGR4$iKIh>cbn4E
z;05@bh=z?+vpE~q;k?!Q{D`|a3eR-|(FoT7nicd9tYV}RJd+|m@gL@+QZQW&l3g8B
zaaIS6@j2{J%WRofBR&1wtcnMf#0&(Q({|PUuh@KVmGU`g0!U`uWw_JHqy&4#LyGH0
z(as%6KL!^UA=HvxRyC4a2!EY%>At(z*<^e8|6?h;PtLGU*ENFbT`Xwso%p+zjistJ
zo*N-mEVX&HYoY7h039KK#_zuWZvp=fWLuLR`053I#kC#^Uc2<2yfIvLKaW+sxhjp4
z!9yS!Vl>e?RUgt#Xg#yMS*Ho2*m!kw9dtAax?I|9HGEI2X{e=^fB2^#Z^ocfk^vtC
zg-&Q$KuD}qmb@_C<-GJfMe5cGo&*+%1C9Z49^T>vk_RjFe$Ld!wryYg`PWe*OCdCN
z_YyN>aRk?#q~Yb~AMUC{ry@lvk}1l%6jPQM9iEv9OQ3#lwo%Qj#>}RfpJY=PBY@h~
zgiFPEg;xS7RdPMvTD>cw>d2)`YnJ!VY<`cvedjRHK2xoV;oP#E1Ie)AknTwtHB^J>
z<VT{UW^$F;C@#sI7Xr>9O1yuPXyQza@Nt}~;a>$#`)~qJoVE}(<0vQw_OHU^-A*|<
ze->Y8moxQoRZ$s5Fm$~5_HYfUMrBzg+tyH)#uc=PSO#mi$X3nmc5o__DH0+j=Wglc
zefLP~a||%k$q{rSj2*_~muD{-c2kfsc{hR`Ng@*VO&JRPpaaMzr+{p9Wx~6+Bcv8q
zXX8`1EHCXjxW9XF7&34t5s#togI==ZJH+*+mT5g`I>?cAyL@9KX-Um0^(A9^ez6rA
z2~60D*r)G?&$1?$J-Cxs24jjC1}uqDQU7g$hPlKZwh-Cn)QNj}*--pEPZ~*g7cVj6
zWM%l*Fe5|j>GRfB#LWRW8U!|gHu;m0NLSNz>E5w<$TN$gF(4s?OP2x;_Fsdt4*mKe
z_@+@Rdkr_Io2Y6y*PP+T;g2(_1fx?t@nE}D^o?^z&+C7WP>**MvCC$)f{)CbGC>Kt
zkA8U_o+!eT^$ak#g#)Yhv4*TSJOI%wYF*S;$AGa+t>Ix>uqmk(3vr)8r%3?f$(gGU
z>p980wLHCp(+l?(pP5{7f++zO1ux74_TPO;?Z;xxG#V7j81(`wzEq5UHa0~_P$xGt
zk^oUM=&Tns2e*)LN(Egr#{dbfl|+^@Oy5BjC!S*#!;hYuD}*0}I?Yp2<aPQ=vynO2
zsEeXFLuMD9CtSpd$95wwk!5Ta_1hwRuZc2=&kRCTT+VpIaP3)(Gz>~hD#eOJe3p&M
zW0i(^=O7M0Siqb^wKsZgGaL<Kqh3W;DydW=FFsY)R;@rl=hPIe-4cN3Dj;}UBi-(O
z##H)qn1P$EnuDJm9vCq7h|J&~y4I00`z%6Vf;K2Scsxzul!7(_@05fQf}Y1)v_e|H
zam|j0^n82>+%)^tE;;@0Rp8avZTNX?l2>()p(DQ#bNK%A!er#hUDSzlJ;5hG9e0Wy
zz;xfTdCC9qYfQ%^94*5z?(W3rvK5d-rbyhV!0|xg-s*=$dqmj;Q=56)QWIc@Z3v${
zCI@}iDPc6q78jZ1OQ}`5c1sy1FK?>4fM9HW82Fc?zN^iGw{jMW4V5e8*pc|@C)b)=
z`8JFS?T%M`j1YqsfF!6k+f*le)DU~Gm5gY1O>?h7n&jwWo};zdY5b?0Y}5tJgUTt`
z_%wfP<QcB{DnZcf^Gy(#tJuD@1c3^D-U;Rl_y8MK+`;?fjzz~IGqfA8uQQj|)=+FX
z1_aVrUk6SYsF0ST6P7Y58XHXtQLTxpXD%@@!86n#ZYCdXrclBtF*6~X)w&8zp<=H^
zv%JIiT=I@pH=}Y2!5viBR@9|E(nB;mu1;~Ne$~B&eC;BsAgBMuK(y5xP_LLo)<4dn
zo->C^F|UbUnbKJZZ`_wf$~jM<1}E1lDoHIk>)dWpJ*hwqunskqm8jtkG!f8D3pLAt
z10<;K#!K;)#-bt28Bi1JQZllN(9nHLqJ%|tKHFSy?c7+kAHUedOE$r@%q5)(MLg+n
z{KNo~k!u8{#@hXUMirK?#MUZ(0{uOT2V>W1zxUStW05QW^Hb|@zrCm)J%<HgCf5Ml
zt{@2iK<x$-P@;xBr5QCCrU|Zca&nQ(O3aRq44KieQTZo|zy)j9zz<4@%4M0UPaxZN
zO#t5v8<FA-LR}clw&dA;^R+uB7Z0K4RiS=|WSS0&Dnw8%8D`J`m`uvKbFcYm5ZgCG
zVJkZd8*>9!L&hj~<&p)FvY=F@s4mJ=!ctZv$+XT3c=HA<xzR6UqEv|1dPg%6Cv3fi
zo@P%LFDRyp-^OZ%X3Pnk?CfwGdVDyWj=jLk^1n6VR-OIIb$Uzwn-<jBNH2m1mwH^R
zHk%<+!coCxRLLSM@ZA+;=99@Jl_G$m3UJj8u`y0nU|JD_XG4bMkW7(iuML)<U`7-s
zNHk+tCB(DD$TeD{;e0y*T@=$NSF*9!ZyEFEo5Zt#L0lGxrbXphAiBU7_;~$nY2+E7
zW-7wbtt``kWJHV|HU|Qx?Hu6v&D1H1nu}DJr!U-`d6AAR&KE1m8Zhk%lpLcs%8=6k
zQw~i%K@;^oM-G}1r^H;w@s?|oCv3EGahw_lmBhrwvT=)KpO0A34||0<xr8KAMM5rl
zko_g8gF=lBp;(9Vh2KvF;lGaV|Fda(u6^FyI&Ii>n$Y$sW@+jKY#^x(p*ElTqqF~9
z=pWJ}agpOgGW|X5BR#D3Ao0=!pc%tvn0tGtot^#Y@naX2y;jL3Vl$)a7U1|->^3zd
z!Tuolng=Upm<>XgskGV+Z}NvZ(&^ya=;zZZvJ1wm+i>m5d43`J8e+_w!FdP3$4YAE
z;J~prcqq^t{#d4Ve@oJkB&UGnoTHk_9nka@K%zFu?3?oopBv^4Oo3|WUmb#nEn)B?
z|NKNTUpprr1In~vEv`ug%_rfmoLhm*F_ey+qrnbZMKxF?lsAik$lUu(Ql9h_-1Sok
zR!r=+gcFUkVhI5F-J%pR0*o3iD<TsD7Hv57tfG0O4cCSVZd*06O2;V<>^6lp?r%kV
zi-BF9r~5ntA2IH2BXlU_@TV{&z&`^SBqk>345jrMT)IScGw=(e#4~o$&$&7K^@}mf
z>$@dfmD~)V&+DA&T@zSe{l40WBo7y2b?kypZAtG47+YTF2MqW%pCy&zqYB?rRpqic
zmEOE3OpiPNF6^15{|NgDI#W^p`!l^y4ZjbAK$6z`eYf7BNgZ-|fUX9T@=?g9knRQ%
zFf+WLhN+u3we}Ptu@%yXwS$y*Oc|?hE}?=RY2bb#8L)qOzbcDL`L3!L@_J%IcWcK6
ze7)R)x6s_Bgg%(W&Ea#Y3N5HsRU6sY|D*-MFHj(gJtc-)t@C*8#7tn3Y!7vhm9Rq|
zv1?|a@fo9(UtP)*;tl*M;Yv|a^`g<Teq~(AwB1JORH)`g&Hp~~X9Cg%*{l8x*v@bp
zUZsQZS5@5D{O@ss#b`KjHvd2(d?{spmRkwA%mUU#E4l1gn2?yYqXGvO4}RozG57`-
z-BP8i^u{U(;^>u86oW;fT*Et#hnF!gQAR)J7-0rC4;7;xIIB5dpkdDzC9ZpP+mOs$
z>)^V(D#T_t-haHFdhTkDavRrsgXsIYSBF-}r&}@ia#4nV^?>$B<8ftUgxV^LtuUov
z*cD{mGHuD%&t=!hG=RYq#wF8)XM`A%kfNJD7=HtqpQR%?C^T+ENpRq){|96YRnD>!
zIo{Idt#!@KAK-2L&n=X9+y8I*Iteo=!U@TVR@;(HQWJ*?zMRL4QUV%|yymdUfZ0$X
z;Cp;jt)U7nYFZ3XI)3x2z43Vyq88%qd0E~AEEJ0eQ!1GR^6wJ2kXL*EeK1e$oCy3F
z9&?_7K`CcbeB0NB!Lbdhe_}w2$q@}<{s)5HPT6|3)32%!YY+Ak^9wdhX*M0^h3D`4
z^?_RBZ0zigf!Kyf<V2N67!^cC4{(&ZWQ68}PNW+Q>@pZEx*boiU|zAV@}jxi-{iY{
zz$ql*;2}&)*(CZ8->G6d#M`+Mu_dZ?fs;L%?8d?p<w&XYl{E~JnPt0Xm9v>*@Uk6i
z6p>;{R8+CzDV0H11^ZBi(DSY`^7s^TiB^-2E}^QxIQOa&xx#%cY(Rpf4E-nvUR_#(
zw$V=})26s3!9GjpJj$}Tg*eWh*T=d9#b4zY`9@AoIp(EvKV<xNn2wS{<sg{`$fi`K
zMEh4=Lkq9j^{a~9q==yTsB!5*uI|`R$7svLIiiI{<gur}KN(VkFYMd2Ha54rRvi|}
zx*J%ZaH|aU*;xO(PIVAEcCV70zcfyOh2T|CrK)EWi1c#=CqSvppVQK|*dOIH_5jPp
znGGEmvZTYKj9+mF*ZH%`=j;O2ncf>me}~Y6?)A`mf+*emY_A+_owRZReb_@!f{)L`
z9+t{xpi$0Os?QlooE=;viBmef=08GyCATwR(Q}7pRNQ~4o;yIDEoDpr5&s&<p*9sl
zg3k>=XE0{tD(5c6ZxD-q0X|<q!b7A>H4PGoTYINBw&Zi5Y+Rl*O-|Nx64WnT0=F%D
zSb;_Hx6y77e!?)d!PHo)0ay_9P_Yd-9-HpGW;R3>NUa*Y_!m#3MNmcr?a9oJ%pr>@
z<0$T9Lrh(Gaj_g!d^}4H8hP8iNa|_iNRi`rq9b@P8UV2(ZViA`hiX|@(?ES!aEeN0
zDm}0D%bz0&6FA5kwQkFh3QV~lA`2KPD{24!)_u8juFSB{RkCaAQEsmn62;;%!qd=#
z-prjU;*7Q8(@YB<XG$7{wn{K%7lTU~mr43(Ze-J;PNRs)0YEDqY(ehcLT-*bb9EWH
z(amz`J>B_d0QP?_B7HXhdv*iQukd=?Y_)Ec7Z^6RI_J|>3RM9z?oa1%%dXfbL0858
zRVp?VMd*7D)W0Z@ZIH}p(GW%~1ugN*I9@k2qrqjS*E24Mh`kGkqB}fZops<uP&nYp
zDNx~1MmFxmr0Ee(*c^3842<%=_XeLX{J1lmzgo{63}{c?=x8Yk-1{Gz$7~PS^eHML
z@HP@8z{^D;&j$aQ!r4!*Pdz<h#;At&@rUHWrxmDk;Jfk6X*Vp9V(eHe9-?Z3E+EC7
zFR}#(FT(joHZD9ks;pw^<NEmoPcE{*&02b$5;eL-Z5`2xR<<jBZhGs4gtq#+{Tu5v
zU2{EM9>K+m(R54A9zHTghIlX_qA8s!g#zS+8(cD6+nlknc_0%U-lct_1bC^<0zvj5
zmtDy0V4^bCo9AajyYQa}n908Z^N*aKGeW~x#{?5Ixuu0-<~rOow6yf;dW~HDPDIVi
zwhkViSCiP<CT_36ENNDOqtcm{xv0NDGa<A*T((nGJoh(E`jk*;?E~5u-(KEN>d*QA
zZ1G+u{Yd|GU%G=Z_@~HQJiOvo3qdvne$y8Be}7sZVSGO3`tlFvfVIKcoa|P(I4%d*
z{w9xW38)lJ;j;9QV*UH~=OHp~0&6b|*g0rEl?eOOBESu*shKZGQ7EZ8`;TGCDjGh;
zM#&`2y*UVM!exDCv@SqcoePYpQ5~kSJ*L%BKZl@SsmLY}x6$AcpUzXX;<k1*4acA!
zRFpmfvgkvV1XV+J6{8Vf7@{zZXF^+gXHrb{e4*RbmDyXp{8UzOyr3ktc)l|$Izmo>
zX(h+2D=8V{nQz6juQvRfV%-sfLJ49ZTW(O1c%5=dSx(U@&<){h0D`!v=DE;7@5yN!
z*?1+>#XM<WAg5@+Wv=k99$|<IlgUzCsHTo(LVI^*hiuqg+_LMo#jK@3eS%eRNtGsu
zop`wGE98Vm{2=9Ru9#kev+En99LdbIEMUj4TN;&+B9LruYhByEU|-HD<c8Pj60bCI
zDOX6Op<mn1z#*s7HlyPzJVC#3XHQ00%EYl^kWKWrX7#4~?;}kGbhP!pWavm+bbL?>
z_Z2fv;B|F0Ws@5Y{}4%gt7=Ink0_tE@f8lLX{y(Q-h$l+k_3Z4Tm`67^OEO*GeB%q
z&yg}%KjMvWj1~6~Mt27Z3VWvd-d%8YOl*4XawHhCDMwOT))oiDxqz8qvZEE>uFj%4
z1olPz4dl;Ebso83KhC8w+^Z)IAyZ&=vR@j|2^~A*Xc<F4IddU&@p)U^dh5V-gdsDI
zc!vYbP%c3MF6@1g(H-8&KGQhd=D~b+Ai{;;*(H`l<rHhIIluoJA!baBO!&NFSy)Jx
zktbsop)MH!H-!&tW=E&mPAK1@M1)XJN}ysdO@o^-bV-;cfq}CsNbri%M=Ti?$WUmD
zktlEta%UpoyiZXGFWC@<1uwNb&^YH)!msI@O#q<Fa#7$xsWWgBOu>vyxG^-nd1`2A
zp^AWhAV`Ln%^HRZ29I@)R4Y3^MTV8JGSQ@Q!YXV;qUO>*BvnjhP%mXe3RWM25{)>D
zmk9=;z$uYju7huK>0fz!{<9Kbu@yDh!Ft=iH}a)52&?A9IjC{sqnJI)C|@x%uazYm
zhvOAR3)<cbjUg0)TKkDd-9fyCkPwg%l?w{U%Xs!-Uk@c64l`iiHEmHD4AelZi~(=W
z11MwGU@+>Rl$;C(u3%!;PyjC&w2YW3w8vhRrj5}s=#B$5EDu=Q2#YpB1F&UliQ<(3
z!H9y&B<cvx1BdS`&EvGeRseL@hK~UIq@4uKyR`mN(sq01a>x{{$?uNBYR|pczi5dx
zZB_Cak^%HaUC~B(?T`HU5ChF~^Ps2yQmPc?QeiYYz{Sv9JtBRGBPi5!3X-AhRv*-|
zMN7~rcn6_W7{Wo>NLM{JW;7UvIZNt>vSFu7j8A*QBiM{I@eXd7FxgU3Je=BdHXcq{
zmmSfxWp{X3yR0b~9e`8bP2mP<l2t4cG(@z7)iLT+qFg90R678}Dk+KTB{$3nuYg+&
zVU|FR0LjNM>_dou%rS^%?{6|+=`dRiFs@P9?>~>KZ`zc}AqO6tL;JZNM(}<6pqN7_
z#{+j~n;am|Fzvj2br3`G?m|1dyTa)A^-O|ChdxwNx4ON+*M?9KPXcS3T#sMjuM4}f
zhoY08&Zyn+cWY=Ha5Q+V%0}8~SI}<MyMr-(fgO<52Bby6^X6gN)#hPO2I6{5=3<AK
z`dJpS{y=fJ6ygTaNhGp%rR^;zO0s8zRc`36iTc)5;f4dZ`Dqs}CQKuOogFX#TsNDE
zI_+dxM*@=z|Lb5|-S#lf?fr{VQEp`VP<r>8C`UaE;es{mBiu?fuUxS-G+G3g!~Hg1
zcz!!HjA9Mi2B=ajb$K3HBJd^LjvN;kJ$=m%kAzX@8JBsa*nVPC@f4|l1LS5Ch3b_Y
zJnW%%cqz89x$vfA<~5v#HeCa&7f`X<smuF6riDDR_J(<ctpvCT*a(p8<+&?saLxq;
z;1)y>MAHCx91eQv`%-`P3R3zxkoKu4iC#(`Gb+a|4YZ89iX!*vcKpe=SBE883;(;P
z{I8}3G#32!&=m8!Nx_M_IrDdkk#CBB#x8`bIHOdFE|r1GExS2eHIegA|8~RnY7_-8
zP7iy|1XNs7_8Y!B9lBYTpbZW9o50gP0KyoPPidRaiI~oab0)Ea*ftS%yS9Gr4s4~9
zoKP=+Y}CAQe{R*Ea3*i3u!(=S4AtH!n*QBW#ALEipY$ZEJ?>Zv?<Kd1tEX5)X`mj`
zTFNLYVQVScHjQ&1_`}RWeM_}e=>nM}ZN7*@Xkg!m)Nn8FZB{DxkvV=ar=-y{N=6>^
z8dNQE7!*8$A`P`npib><j`p8sy`XKDwR-6buWCKL;eOgk)siHc0jLEiDg;>uk?3v`
z@ob1aZMY3E`qAd$dOe0z0xb-aY?#O4fBNJlR83To8n)HCd;dxl%~vBd9iO|d4o>`E
zt~oyAo^KzN!nZyXKIzQAHn!z;0?V*k)rJ};@X~_Wkeb>0>gN!+k|b<GWY97LSE5H%
zW2uR=)Mz!=PBDk*$r2)PKha;rw|zHn!aHxtaeX}P1%LDXugyWPZi2gD%QlyDyeqKX
zw_nlaWRiaFm#?CnFtrJp30@J1wY>cqh5H@pj(SmG%x0e_(8Y4g$ioA21qJ=L1c{X5
z74R6_{qM2!DU(;aEi5?$SFDNVepqj=XbD&Lgbh&}F~xvXR@A|@FQV{Abg#rE@%2IC
zHjaMdIGfxz-{*~0SMk<Ra~Q;Af~>pkRwDh~f?ETu3pW9@SC$BN0?v`O)hCB#-JTzA
z6Pf&hP>{`5jgA*|n3gI+ska18P0Zu582lHMq1o}nqx0FMj<Ha=MU6;E!Rv#3!MS$(
z`lRDoQ?V#%%`(YY0PHX`a%Mz3Fw2-lDop$auo`L@0tHMvI*!x=&d{Uw5+|gJBGWEQ
zTJk896s!5`%ObbY{XVZClaX5i^0--|u1o!=8J3&tukhaf-AS{!oiCxZI3OXh1($AU
zF}68>Kb$n0XL$ka?{syUkl9nFAyn*QbiUF+qij<#On2V3c6aPT?ADMZE8}&XC8^wo
zTPKDi4x?R6Y4V<HP9%uV4{$4Q6CA_eo5Bj%|NkW2tM7ekM0aqP4swf2Mj?1j;M&yS
z!Xy-uElXjQ+A@PJ32tHUnBm^z|D)<110oOGzTvFR*sy7GZJ5lh&92QhC)>8|+KkQG
zY}>ZA*{*l42lsQm-{$)<=l?v;AI}>}tJ8U9I$_q}yk6e1t?6KB-M>~0G?^Vf5U>Cr
zKKR$xo}wu?+HcNb(OM9#8vQs3Moe83X{6JejiSz$?ep`(lkgoQhti*(_4Xe>b~HMD
zT!IH4yE2cyl8{Dji4|eS(I(0z%pio~>Eb4!82WxAXEd-F*8Pi|X_hc%UU~J>R#9!`
zv!<{XBU_|gtXmFN{}@a54}S?uOrwcF4A?d?ed+XRJ;<Gm!*t4)s!Xy_P7GV^F-p6=
zU1lLWtIAD`a~PSp^hz2%#d^5VMW%Ih0tG12CW}QXcxKL%?0<+V2vzFB^+7=P0)I*-
z%u}PFBy{(yp(@Y;!)n%}>JTTRceVONO7X~Lm`A}76J|fa9(&s6@>jVU4X4Flo3`ls
z*qnm36;T91cf**0w#&WV8%kB4AKAYD2yk;YdP>0%gBAZ<_-ojNUr{p7;CFHS9CnQ~
z_LVJu_!-!M?%X4Mwr^B6D*&;b%B+Dqp-qpajrMo{SN9=a;_cy?0-1^NU*$`+){$ID
z;k|J*NGJE6e9kYloi7X3*TYT3ChqRPX<}xiKBpN!|JUd6KaWuGrj`iN3TKbRQ(&4g
zFA6KS(F3TAGZb0X5;B)vsoJls-D)v(SMt`6<g0H)hB5!92zfCvqrcen9<oRq5ynNk
z1A-PeRyPfxtmt^|i-Y%i@5qtWhmy^dD6%@IvP0;v7?}@$ZN95y4ccf=o1$Gx=e~ET
z($?v*x~%)#(Erb<EzKF)tiAWu=i!G)<Iw8DY0__Kwr{9P@K!O_dLfoaeti3XWormC
zll*#;O)8dmw|fHD?f+rBuDer*#t`Of*eaGOKUgs1*kfZ+Vsu{Kt~(yN{VOfudR@nF
zbX_%G8O!col4&BIKSu$j;HQNTA`6A|F$_vkg(CnSYJ<Cz;Cfjcney2lU1sQn&03&a
z##v^8Lh4^tS19j)lu;(?*h<u!k_SVkkd9!o*h=`a4J@2Xru1P-T8C6-%P&B(9<!L<
zHCuZOJCWxZ)2wTzTjAv<3ytyx3ml(git^vTU>_*{hNiYHVQ!3Fv<j3NKxiA(x%8sO
z{X7<61z=sRxmkqGS>;*PlrxIaH#=A-Qo9IIz}4TXeC$dFN3d_i_W)VUCBAL_U6Ev%
zUq<%5*tjgT-$xy7AUM!Ww<wS-RUBYQW|qZTqgArpC)sjmv1xmk`*_QFmA10Iyt=RU
z{Vo4K3H}jEVtV}i2CVD+?^i86<RJ3d9nEPHHo856+N2S2l0JM|AVNHzMBnK7zL1T~
z?F;JjK7Q*DM}xdlW%|Z7#Kb+0mQvu)h()4ax0s?07U^2qGnx9C1?BI66oT2me489_
zi3gCyKy*S15{*267J{B<A2D?Q1wTpJ^o~7Zx@h-*j+NiU2X|h#CJ&h2`S)114O{zM
zCJMMa(TW(d>H^QxD)g33T%G+WVJg@DDa%1<#wT!I6Q`g{Wm?aPavnaNG<AHb{RG;S
zimY(Zw4lbEZZC5<RD%Y=9+-z1m#t1w{+sw%;PIo|;ds5*18dIC9qSajXmyVNHJBuc
z^L#W0MJ(du5r8!DLc)c1n{nzo#i(h0Z2hi&NJUevHUbr8&xYb?(Q}cZL0Tf6p_aV;
zWQG3_8k34m!7U?&7}tOLZx#)*FV3mcwDdnZ@sn7VJDieFpl_AS7*D#7@~BDnhRsvg
zs2P=PlhEZZpltBqLeSkd5rhfSLc&c})8xWW-uEQF1I7&LQ4)t@_M)Qq{-p_rqSb9~
z?NonE<P^=6)wFa*tCU{NkocmFEj3BZMVaNa3<FuGJH^xRoW=rViLUz>*zcAFUslQ9
zR#W59e3Kno2G7R%=H_l?WG7ez1mrC%f(;19Cnx7DVvO*pB!JdwzxB4wd@hcjmWTLz
z`_O*$SAQ^lCRF?0#&-~)J_#0I{67N4)QHhxE5@MnXBpt~tRh8_F3v>4nu0g(xyHKp
zi6Se`-Sw1lpWQCOtD-;dn{_pmc@kDL>cxJx3IebBkHNu7!YwN%%!{?Dx9hJOhSO~|
zgS|8FN#D`YygIMu6AAycwwqW*M^x+=oI=nfTHtmO`JtrvX^cwL4?Kl;%ou3~#zhfv
z{VdWfAPXhratiqU@U`m-%{NA@p!elohu->nvg?YmN^hNVOflR24)a}!N(BcKs=Qkj
zW=ngM>^E;jYKw8K5Am%GViTKO5uadzf|-v^43>C|zM7@6#?Jy~!AjMx^q#Lm(^k>f
z+kt2|G4RCOE39TuzqdZDv)wORmM9%v!Pujs#if~4vC?12<ZG&I3Hzb3u$ssdF5{}e
z0(OP!DLZU^+1LV{pbkhViF&W|^(3e{hp+(jF&Cgp010h^I=V3=9X7pfv?C9H_x|gJ
ziDSVOXQxAoAbLKIw#L8hB$f-GA&lb9TR9G<O0w3B#22WBlt8%~GSy0UNiQQlMECeA
z13|bBn{Nf0LoOLfr&#0Xdi`tK`KCRpO;eX|hh|T;v75pzhL}P4uW0g)Q{E-V&f<Kx
zXDX}Nrd)XT17Xd%EHA=|SO5HW(y1kOu`{EZI`_`s&buDR2)SLpfwduXKy;{Hn(fbL
zfp?@0xmYY}n?3MXiDW1?r@!$Q6#$SeyZ=h^b`{!kwCVYLzV(6K^}_UtqqW8>P$O3U
z-<JpZ2&-*tZ0w}Ysto)CQD8z)Gqh<~9OX0|%(t4=z<)blySPgalb*5bd$Ng4iHH6X
z*21K)&4|-tmO04}Y_Sh$ik@4)gLs9B-`A0Pe#V!jMVm1hEQDmebo)?ky+X82RXdte
z?^a!ZQLfOiVM{Y!rT7L>#OEgo0q?Jhui%N(RO@Fq+T6gy`f7GkY*wml#IXeTHdy@u
zAo)D;tkIFgMna{ZbOh(&=GV_1jVm32JZn~NTOjBm*9eSx$#i_PjjPwUtJY9&)js~^
z7?~cTSDkMDZIDbz+j^<=749Oj-ya5CbiP*7)6?tYElzAt)>&V;k7;p%ub`agakHoS
z5`NL@B~RJfkMMw2856MDLBc5Nm!dOayG>?B6cEMVcEsyJZ`&YWA=CP7#2yo%AQ|v8
zkyaGDSiKRlDmuJ(6kDCKfK3>dU%@t}`h0wcN%`1TuV86o2Rf%6zKP=a(a>cTFb98>
z{<{n-#;b83d1<Ph$jq}_!R)tYkxUuZ-_385<S?3nm{dXpc=1Dvx3NkjIRPFaHPUQm
zk25Vd6bDu!l{Vlm81|1xdFPf#YjG}zGYDM93j6D~re-%Al=QQ!Xr>$kN5|oMgI?1F
zgL0TebR{IVjKMQP-=nV2SMY_p#;pCV@TYn7EK(60xNs_8VEy1_l)LlcUSM}WyzQw!
z_LHq?%ZDk&h%62-Wo*A6rt7KcgNIuF_x|O_yKi@qbgJGQnh=;Z`roj(uT$C_ToBbf
zd5Ae(+*-0R;+^<5>I_bflfJ%+w0p7KH=qYGmt=O)ml5*znN2ZUr=f(Lk_<sHT&HhL
z?SV|%#ME4hCN=d^o}Ph^AHkkr{Pr8wUrzm(D_W|fGLd&;-W3>2v(4a!uXa<N>@r5g
z=pw5>&+m0<x|_P0k&_~pTc1?EUcA5*D~aJzCBB+1`&@8W(B!fxnVzhE&3SRv+pE0f
zAMX(sce`mRulW4n<bIrIx&PbO)ajCb?te(e<WPCLEtbo(Vt%vqF~|h|hY%qEbmw8>
zIlHx69Q-LxrZ0U1&2nCUpc3)*ygz;tMn;C(!1)Rxj2^HE+$#wT#FLAIV~oynYCAhW
zG7LHzGI){?8P0eFjWJ@RhWXHBXuc5y1OdY_(#b>(HBpNh!WDDq!)nV>b%)3!jbz4z
z)DyUnYU#h6B+=lT5K@z}AQP+}TKOhyID{*DizG-Cn|)ojc1_cX;__KCxUf4yAWxN^
zs;mR@iWX?@mKyjG(z7gT6)3V}XvUk;Y7|2V5|%t^>RXP`NaJVuz+E?z2%8`w3tC!a
z9FXR%>g}9AZb}!j$0+Trw;Z*&x#Ry%k9r}O!BF_8vps$vsU3q2{qv+_&>V3zDR<}h
zr6*Id_3ijc30SCYWbGJyU&#*l&VTNbDNf>!J#42FYEjVtF!{A4$0y)*^Fhw%!OSN+
z5IzP_^ivWTwB2#(=J$EO<Qv`fF5NYrZZ?%agP{Ai<6*7xf9Mj}TTT*3FBEY>9f@C>
zN>nS3w37bVbr=-1V)+2(TmI0eL6%blb(rkKwKi&Su)H}GBl`?xub9%&$^}JRG~E57
zAKM=ww?N;;WL&y1J~bMVA3WV%NW#I9D|gMWg4F=lb4HL`8gu~;)xT%HP321ss2I{=
z{ke*{1ghTHYk4NXR>D>;0*st@jFYqxQ-#Cl+J&ekv{b^}l<v-3uyc@~%~~Jjz(kcD
zsHTsFX#<VSuc`fjL3z}KBc<Bj+d6Pejj^V&QIhZxFB*J*9TG@lDl+;jkT{&I@=Wo4
zVn42b$ZbAu=9qYR>>MiAB_*_7c4q25C+qFxe9ZZdHdrp{VpTM))5Mcxi45Y$#1%>v
zh?8fP$tJxNZX#&mrQzV>giz##Dph0lBQ|REI!I9{h@iiyhlGO&g7R{d>a6Gu=~77X
zRIdh6ZyVv|6o^R3m!ntKH+#Q8Q&FSd-m#i-+rOT#J8l7Dh;#PNPAJKe9kv3+o3G$A
z?yF?$ne$a^M_7U<A0BjQgG>V0W;Dd+&&i;3WJ=P~{aLx?&C(RwUf7}s;|iIj2Qqel
zxQ@qz!kGvg!7A<7=qY>WXTBro0SQYsBtTD%>@Edm_$^&96W<*>_}&G*ymLrq#NsRP
zkPs4u03?wl6LRk%T2SxB(rd5z&mX>;P4e}~zo&O~5AnZ9jwZ8=x`Y*AG;O)?eje(5
zN`EZ+J~nZCUOX@D2mS|t{@)TX5~4%AmhdVOLo>wUi0-eXRS@h~FJV3;1`uDMQY_Ig
zA*hF0ojBW%7CVjlkPt@xB$&`TvkpoV+*!iAjfuyyZl>kVNo(k=kV(MF%-im*luXuM
z&htNz2=u>sh2NJGu?SzMpq(G3%Uh<j;bh!1enpuTIZzsfF{Uk4rS;8c%|G{}AC4<p
zN9W1THZ}S7HC28u1$us*-y-UI`^OwYAH^)ysg^aP&+YfD%Zr-PF*P=e?fAJ^$B!FP
zo^wX>t*-Q)4>=U^ZUhf=)oRz4<~#h!-@Xwfo}W|TKb}BDJ!gt6LQk#H(9_W86UnXo
z$B4kr%l)aZL$+t@ffyo05xWaZFKQzOC@L9L_a}r6hDVaw%5f@95j_;+K?FruDk)As
z4Pj1f2;ZLYuzsFC(x6&gB7}iVJMU^2Gj%uYqKVjS@<(@kc+kK<ww#Yy%+%~lfKF^^
z*9(H{Mn~Vr0G&O0ealxCR(?S(IuD=3_@*c$!Ia`Z*fNNc*!o;lQW7EwfC2<|LUd}#
zVw#;O*&WaALGt`Dv>8kBJ0;07WU@X3vo}&p1+iHjBU)s<i=xc?P5M5Cr*oURrRwUs
z*1CYC{aD-<d$V+^Im{w9Pm?m3PVtiSIfCzduaI5OT3v(5-_@RiyWTj{v$A^@1KjrQ
zkF1++1wQ3Jqq-hAKd8!2x7-yv9{vkw3Uj9nV7Y@miw&LcWv}r@$4G*ziQ>aCU2Q@6
zO*<0pkJ7xd^*l>n&&Ds=D#PuUz+A|hv0Zn>>T$=W^bevwkv4Q4vGb)xXOUPSnv?kY
z375(4Y?GK3UMF9DMt5O+QOk{H)dyCfLt+C;5B`}}r5i3OM(rm8%a7hLuW?=6!H0v;
z<t#%Al%SoxEYFCKrtJLhFnLLL43REI0Vb8d`kXo34!$P)xc=<oZ7!Otz&y6LAU>&N
zC47fpTw3$m@AC~Yt#^0iewfY^e(UIZd;}YJpK(^t9~i7Zd7Z&~D?o+<^KUOIPuLCt
z4+<hC&4{Fou;I#APzHL_>oJ!o0qSgBEpoJK2HO~VNVs(yMLbzJc>+zp3t@o4&k5$*
zzZ41y$lnr@Haf-2Dbmtwnjd2{swy53=A!zWT72vy1HU-VYeyDcurIy=I*-u@ze=XX
zQDE3(=VJxeQPLsVmoX?8|FV~mM8LH51Dl3JBDLreiv#j-Syi2B+TbXdHMyK->7;)X
zskVB12*)-;(X#W5VI43nPMB#Xnm!+Ac61FZ{qw|q`NAmLBHT1e`h`vF8}TL$ARIQ9
z{{ba{bpPmMU2pekCQ-eYRN9ER`P1j)THr<e?dJ0&^*Zp%`6T_nW*y623|&h`UbJ$$
z?bPc2o;JR1{<C-#K}E;|lhJfP*a{`iaB5ZrddqskzE91)EwxLTG|cE*eDfc-VLR4h
zRo2p2ZbJ7{{>Ic0icYiwdk)4Jb-w1GBY)L5R}D8}C9)VoYK7yTxKz;&N^$8`RJ+R>
z98P!LMCOL|@wV1+n^gN!6Pj}rbYXZQJP2gP#(rs&*wkb6NBKZZ(1H+fvoDRtY^P7f
zyx#T%COf@fZQxu=<bLl~f8KXZc7fYj1NL9>L|worJpLb<pV)wxHaxNc2pMtGSwMVh
z?k_wcbQgH81I&dJ+MuMCawp8hL1wxMOHxb}T7u-_Cxb9rt5K6Lg(A{laMWc8%j%sw
z`W7zmG>lSRHxzcB+PnFEBC35&!aYDyr1<(8m?l*Y2Oh9}*v9Fhlh~4&tOUpkQbq`Z
zBle}7)bScpXDC3G5E(<|1k5kN25H8cvqWO!v9WVW1bWFDkA&G9nt?CaSemsfkJCs&
z*fiK%_<O!btnUd^owtdFwjaDoKmsc5w$^l}a0#r-Ezh?VFgmgB_}D>;O%|B*T$*z@
zruV}4c|X@B{=$<+^#4Fd-V<-m16$vR62IqSKYp6ZnpW~ZAiVQv;1^OOcq61`yIe=a
zGfDMjv}`}d{baMg%r$AJO_m)fUh6+!YKng%VC9V#u2Bxg;Gf+s>06^}Z0>KieTVs?
za#ss0=W7(sk{H#FQI{02TJ3Nyl&;(y<Pj#6aKisSsXLGf*}(m0GU=yzonR<ER6L3e
zg?Q~c(rVqYGbX_;kQFYop7FPCV7;1)WVtLs%Fv(7j>DtJB5XIionH3Bzt2pc14oHD
zZxo@ZM3dC@5+m^ZSCGWs`xu=s)55;HkIfASO9r^3Cc86!6<v{5NJ-NOMZZ8&8O3QK
z{wK@L8X;1w0TjJ#GROeQdiqP%S)^z(2{N%hNIKb?LeZ~cf2qvEokNU7&TLg^nb?^3
zsgx3(FPgnEeE$y2@XF0-rtMhUw!!yt3@Z<})Z;$A3wRI&aBCL(C-+N02axJdKMp75
zpQPYtW(|~aO6<j%FPA7g{4xlWB2y$6`vTdUw?u*&0ytPi{zeqZ^rI`L!Iyx6VNs1E
zOeuPnKvz7}yok*p&i;Os$G#%0H|X;!^ONLm>J#Up^OpFpsxlpp&k5Mb!{mK$dFIku
z{Vs?7KNvFm!Z-hx2af5we&GGE*eSg#VGTVV(lQi#F%SjUR91Zr-X3mkFmiCg=k<97
zE+r+8Z=7244`WvI1Pospd(Qh8lQPF!{i<4^X|0<TiNdnU&OV^*7Avdu5j!MZp0361
zn#}iYpaf$O%2`?gO{|%0@LRFl@3;4HDnj00gMZZg8MC`0t&<GgjJnunQzWEWj{VpD
zUu41^1;cH1j(23_4UiI}&(D|M)_s`nUn9)2HFlq_!{mK=_Q12`3Ps&W(aKi;>U+)f
zt|2FIOYtsXkn1ijxUsl02d0qhf%%xQ5~J?^m2h)xq;u+;<S5VuXF}+JUTh5(jK=<o
zBy+sPpltc$=o%<2C5Z%?J#m0Wg}I*2!TvclO}Z9afwi`MK+gk5k9nf!g~0Gg8}UzR
z(JTt>4xezxqv|>b@4$@>{>oC(yXhX?CFLlSzahVWa|ogvtbLtJoKv=t4Kl$Hhtw?*
zZqO7HHTW3738er9#C`QI`PJKRcgok+8IbX3HqGK@Kb2UR5=n#~bN@aGE-8R!zM{e^
zfdbJU^Y+qrZ|f$~chA=GU;jBxvi;+R>n&E!6U^sM)5o4=cJcq+ssf%Z%Q>NL4=lH~
zhY<s55X~tz_$pnD4jc#wDZA_S6}wF-V|NU1{`9~pew*L;39nu>ihz8RYk7I6)1Gyc
z;z>yF7+kh7kW78W;i~b^mFq1<n6UlaMn(zU46%GkjoWvJWlYT#0Zy2}=yticTeqP!
zt*wyG$oadlD0lw*EviX9WDLxL;N{tLO__f5@KXGDcL=&$F}?5g%=3J>ReI!A^Jx{T
zZu`5*m&7sG(7(_N!TERW`80c7?>2)|Z!*E}k?=P!t$jo8mu)0qn&GP6H{W_p&1d9`
z=F9BSbv3;~8;(>;_@Ib|KWG^*C71(dlx<oFoF3++b7q16m+BY||Nf<4Aqa#wU+!$x
zCX*oPWV(%l#g_3q-Z%*<OLH(_Rpc=wi^n72Ln>LE-`Brk0i}_{<!Ke1a^U-x{249q
zB>!F8E5F~|C<Ypf+HGuy7X+J%m!s^KH+5ly3*B=kk~lIlvM!9^PZAoYo2ghTLWn$~
zC`cG;tPo&C`!`{^SUP3-FYCzNwP?22nHB%XAP{(WD$vIue`v{4DM2=HD3)0yEv?<z
z*kAoQSY6xLE|f8lHoJfH*{fHmTInNxKlE7rzcscV|52iS_OJ@P-&fbn@yd_X{mDcT
zX-^IQ22bEQR!n*G^FmBOM5Cabmgjb6g%r)_9?y<9ozq|9sS%_4n}7G0A^V9W@-s5N
zoC5KfxempnQYKvkN9W)7l|-F419NXopGTi0TTiau9K`8^S{qaTE#&L&PaJgGCKg^2
zp73y<E;iBJX|@ar2%?xt(*;CPzZog#02k5IXU?f*2udK9pDD0(HrwzVtnf!qi#IF^
z4}AzH4n?JhvqdH;RRV*)uXnzM9reMB1hgppPkFdJ#<nU=Nu|<N^o~2XzTNS$UTOWK
zkp{4cAtnKkz06@9eq{<mpx`Z%*)5AX%ck%^ggllYgKcFgFoNV$e0=9cn9LiUDJexg
zd3*;S)ut5Hf~zi|Wp+K}6L3-Lt<B|`4lSM<MmB=`p^wPR69D2eP9Mo5$0F3!-kvQH
zm}>TjGOh2uxtFb@H!y4;JX_~$1kIkf-txu)yXDA5it(aDxDXO30MbUIl>Uv~97L;v
z!%zP1jCwkz<r_tyH00!7Pj4wT$j*uRUtw>Gs+x%=5x>tFQ%sEXuev@Sx{iFG);`6@
zGkSizZjO{HQM=#O4(ZK#)BdmbPIW9;?&0{f!(3fmZU6E;=jbYwKWd1pNQZ=64r79^
zSqU}FAnXs<@kr9$OI*(=V~1{zKDGgEFaYH&qmW+}pwi95In-r=axPO(-~>|nYOa#A
ze^ZYdDez#>+B9W%aK}<Vv?wV&GWU7U6Fsw$77{qj78)a0<GY0|{dq1O6;7Ro$J*jR
z1&eZk$W^-~h%^F}q%k8<B|jqy#|9FFs7w4bqR@Gu%s_!`RLI9Fikg^kv~f+fX8PEq
zK8DX_17{6+O6<|{zqfVu=)DE&4LVB`@_Qr+hoWA;-8R`+I)c|*WqeC{{y?R2M{sB8
zjO~xUGoi%m5yZMmR3Bf7jfA_oG~Kp9&Ku?;*%5o11){%-#&qOYP%r0<t8C_~Itet!
zWvV8Q-;L6!lCVe^FzIZDJFoDod=(sytuFya5z=OcDcM_;*K~g^5)>B>vtQAa<`OJ@
zpH#})(5_SVoZJ5hqa*)W8XKROP^w~+$d6M%pPt1$d0v(jarI2ioz0J@qM##}CYNF~
zU+QKds=D(CYOt@$>5|d_hfpIYlX(-hndtJdUL0=4w<S@Bn_3#BDvs=4mvjxJc0Gym
zKg^JnC^KIBqj~ua5&Q14vzOEV|L4l-Uv8FDqG<Wck!>pL$JkwksmvugNE~^M+>{Pu
zMP*{KaY{eQP%1QGkL5=^U=~KN=uW(*5u<>ph6=C-S%B0>EJE-v5aCCs+(rcVAU@+;
z{C2<A7QCyh!@G9x=v-T20o6h+a(%Y_&4%NA(f;b?4<Yv}^h>bO5mnewVCy2@=xa<1
z$pDXzKyd~A^7x-Y3}N3?o29&}fORvf>sor6bf9VK)8Yk3-#@M;#>;loqwCk?h@H`7
zew@d0zL$xntL<0*sMFSUaArfb@2`^DFt34G;aJwk?7;CU89R4Sw0O3rQQbegwOu?P
z0t5Vvk!$-05*|NF@S_3&I)@@W9swpUi22<o_QTxIAl(X0iFhrALeYp0T_bzwx6NOS
z^J>TEu{Pzw*m}FEbHX2X5KIWnFj>A9!8<27A#L)khXRw=<cNxXoCJ!M1mm*ua~08v
z6ahd&bWvtSVeFSPZC%s%bVouBE|Dw>5lNYCQFS2&I(!((0KIDalU|&BAZ(wIC0?G}
zbyQnw(zYpl%yjjcucNSHfb}=^V#J8dfuWO4Es(6wwQVD<iuT7T$(t=Wl&ynMoo_tw
z(u4DJ{R3BI#?lpD@qZ8s+!NVeK4>_N!F1{z{f9~6aW*aOuGXH^)GB6`v1X(8@kP>s
zaf;|w<lEal2@OF6*epo0zag+FAPC;InpWrox;rQ?@4bEzq)O;QHVr(1*y9Svk1^h`
z+C9S7OPEk!hY<ZF0AWs~-s7@&t8}$zM);G#rb~0Q5e;JnG!-Eq6-Xlz!C;r!n752Y
z9_9DF`i8pEeXpo#?36H%P-YuNIkYy`KMprgXl`BizywY&$!^`LG%aE8eM8XJ>)oZ@
z+HYF}L&F5mi<VG%?|({#AP?!a)&XXN01;3>Q2+i037pM{w_p*xeQlOb^qrY?PDBd&
zcacQjPiaAs5Fo5+Wz@sZB1BM1yv8ql+g}z75{XH(g(H`GopN(QTqpp^_?YO$jOP^j
z1+S2U4N9IIZ%n=9nOPya9;?}+!qjx@GoWL0fY_cWan<Q=+Fn!>)$lAflQ}D4w1gs7
zGL@4H9G~Y4pY0E?I58urEwrn|I5)1JR2U{|HNUB>5^6iz!T2gdlMvV0IrnY*TM>P5
zu0r`CE{G6|67gh=OTJ4z1n>vkyIih9KPPnTC1Hm>0HYw01rD8@)?_6VJF{xSAmG1U
z**da)>-8O)=?hIy%hI>LZ^$H9`=1GOlH5jld01kLfm@|-dW>9u5q+QWe1URw2|uYt
zM2?4o6k=P+1T;Q!AyfcF{Lq<IMVVcHvT~ydLJ+C!pyk~Ep}{jjA2QKa1Bi^jsmDe<
zGo*yzuDrkDhr~;oD*s6UCPvu}u%ogyayd{`^RDB8=ul`hfJ*hhoDgOKZRE0tW^PX~
zn_(qY+K2vqSV`qZPK_aP5BgmFD$J@lE=E0N2rDZWTQ@Cr^pW&qx-J3zmU-tY$7fa7
zLpkq1KHl`+%IO&yXD!%QR#rNS&es)xrEW^RTLj)$65D6(uHsW4jTu`EkKU(+uiy`s
z6ie14&jBT7o?lvL(;5ROS^mP=a_KuxF=&-W0cE(j4OouNf)oOd9@D?sZa>G-$Tno#
z)p3wUwK{dg0U=9sd8$g4tqZFX%Yu}t=PjJ)?ZQ+_YgdSBYiCSLGPh%~ldQf5_zkZa
z*wpo|PSSNm6Pf160dEv~{AeO-WG0X}mQ*?LgK6ir`ss}RD6+(tjO+cI#T5cF^Y>JL
zuKB`utATUhKAyh#jt`7U8*dee>LLiXC>vC;PxkLqP&Nh8%6M&=?ME+fZU_Yh0Ff0m
z0S%&Er%H_6UMFKYdsrmzh*YMgU{kfPEYsV6b71iM{yK2<v0jSF3Ju<J<AMnu+FBZ>
zR#qXZd-~a8{p*pgu)xO0VZR`Wb35l(q?mE>A&(!iN_3Zz=cu`Jh~O09Ah41TW7-PY
z-O^hkjlTZJG)!%`a)uZY0cFiN9VD~29*bI}Mx8-Bs^xo_r546F9X!$8^{8<3FRVyu
z)q_@6wu#mkb<^jpcpw6uoNhHdOjW9X^tPAV<sBy^`}a}Rg$^y_0&k(8=tbDEd^ZTO
z0w3T$;v<!uBh_-z0S9n5=bd*fxbP<btW7o(a`xBliogpQ$@8$hqAf>ycDBdwKpqGH
zKJAEkT;1AznoaZgr4|K}S0S5ZK^m7FMx#H8ST1`A$S@6V);V4%uAh13r;%hJSIm_-
zaVM^QK8zY1>bYQu(FF^~X1=loAS^5&p{TnNCQ%kQTI3izkDV@2e_**dTtGr&&8mJ5
z5iB&1PtC^^TuG6M&7^0PQ{Z523;}#sTbq|kfWD3FmkjWyz={B3^efh-XCSNbJV6yI
zQgn((iCq?L=t-r3ZhFvs!q9x~3YHru#09SRy!NF@cFe$31AdGc6Iw4YJuA)F#s*w5
zfDa~QivM1D=iJ}lXSzCQ6oZ{V01CmlX%ghxeF8AzB<u4=?GrKZ{_1|GRjngv{XY!W
zme;s#P#*I%s-cH{X6Y;+JTCG!0Z6rH$_t&=9QrQKi!Til!Om|PhM`uZTLg9&gT7j5
zzt88PxO=dsJ%NRpxxX<MLJMJWhODoF{gwLWy}7pRcVU1*Au~OEg5MFxjwad*q7%6U
zI~&C~=gyN4R(&C!33Q`s-pTSf6JfVJdzGJz<qN)gsXjhdb)vk58+*?>O2_$T({+q%
zJ9Hn=?WzafeK(x?N&NGeq+@p;9spC%T2IL1gwQv>7D}BoxX7W#S>1VdX>Db5J#Q3y
z?H}utG%^C(UN7HNYaYmDRC*4>l{Z<A@A)^#Gck*tVzNtvD-&+1dVJm(A5y4Xx}7Ag
zUKI$uP3oA|FN{!Z3CPMOL-nRxz%1B7=;AfZCqwWObnc8;IyF+sNXo#`)}ca%+64*~
z<LBODQKIyY8#)(^H5^!b1jbR!<if@jrB@)BQhy(pO|`e=0Of-Hsn(Cbig)LK43$pG
z)iAI**N;mS71mNYSf05-hh)bZs3%!rZK)l1czlT-*;E6>p^1voWh~J79%AM6J6i{#
z<IR#HiG`)U7L$GgQ;9;w8Yaj0ksa$D!TCqZ#pX?GF#huBQFy5E2pHgi$vPNgu`Yq$
z+U*@%uV|9jY4Nn2%-2TL@=LY<7R&fv!_3hvXkTa-vb(}4(ZDdK(1OyG;&dp`_KR=H
zTWeo<YMbJ3p@e0bHEyRDAfvxpS@ld3$PA!Eg?`gVb~UgP2ovt6n5F_Z$<V|HG!8=u
zl-fC$j}1KdO<)9hhVsG@GVA*gqD~4mzzO2*h$c+F5VVCH5|s=BsfSkTuwNVWRS?8z
z3n#!f&dkY83C_rN$;V`4i4o7OchV4pbt{|jz-<+VzQrtpP&A@qZHIK5NpyE%9>Lfa
z@A@;BCVVj_YsM_lgz%O}(4+4@xs1GQM?}Y)%ZB1FT=VE06ADd*Mwo(OBNNMLBluX%
z0opSnr%K(viWFwTFMYsxMbbhMi(eXI3H@$@68l5OARp$IWV#TbIz))BOyx|jE)$kJ
zagiw&h#)D``JYr=S<)n_v5EEYeKbR3BYaSNt}%i|OfEAOXn|qy)THd5uF)vblvc|!
z6ZwTWRE)o&c;e$-)?$L$sllJYvB^oP%}EH0eMR4~BsA6HuL-(1Knh`S(DZ5!P>}a%
zU4%<eQ`F->AiC-D>CGtskv_`hc!)gm9lKeT4ksfTE?O+mv_P_cg<_1Q7ueru4M8{r
zI~CXgrNttPM-vj5HkMjFefgM%m(z|@_&+25#4!HYg%uJ0>HDX-#ywhz^^vsSZ~ZSL
z%md`a`o62?{z$7w=b2ogWPKR$Cj75hF-M+EDLYOIoF$ebLt0u>Try9(<`57SiXDbs
zYG0$?VA-s4&8)O(vOzbKbdhbOZ@3imuN06Q@;)S2aWRONUMIkY+Gr<G*n%<=t3E-~
zlB7=&g%tx2o__tyHnj4X_NCJaSfm>5SEX8xHN>od-MP=5fuFNK>3Q+eXmb!Lz2({R
z9x_O!iC5gSJI`+#OCJY|Uy{Qwdu+^=(sNa$rVrWdanB;S&Ae&RvHR*(Vnh2U@-Nji
z0WE9@rKAKtj4)4!?!}vxdUG^EFAz+kv)xRT7({UrV!jC4H2=Z=N~MvDcQ1ZS4;fF2
z;;#f}fK-tX38YUtipS?qhcUrBixSG#`B&4r-V-5A*VrQ?{Hqc)NCy{f5v|w~0a`)m
zBb8<i$rdkZ3r4`1o0A;{{q(w`S$Ge%!$8hL0^!B;P?FK7Kn=Jxj9etE2vL+`(g^(I
z`FiO&1<&pIA<3xCFHhJ7p$c6bx9&bf1hz#6PM%W+l5V%uPl5ZuKEWrvyV2TeC{LMA
zfiaafSqdDI+@d|`tYft%c8+6V!6xLV2vrHQNn_&anQ7A`TrlG$dwKZZh6v&tAqahQ
z;OlhN;N7l@rkr0l>>*^ld|Vpjxy8Jy)ZJSg5I;TyN-*SnIcN<>OgI`H-vBuVCE0_x
zoJS5nNaUaOUl+e9<z!GSX7;MNW9>$t97{0_Aw!^2DCi5-P~JjZS}_ERA#$t1=z*pF
zv)y)Bae@c3+{WPD8$fZc=(4$AD3AyXp_vYbL8JJnsjFP|hQW1fX!hVsMOAhCJ~udI
z1TVVwaN@|G(0*sPTnm*UlOx377k)J$s6Q^?Ivi2#lvr-WKUJ@`Wa)J7(itQxv(9p>
zH92o$Zbf?&&z#;HE7ZuNH7+-4K>JHFn}1S_Ht->Alcu{4MH7pm%t#iw^2Fw%#=$$?
zqH^GA9b4#t9cvU@Vhs*i1Ws@tV~|*v(*Qfl&1YrrEhRv?mWz?wu6a<sAOd7cO7^co
z(yBawO{SM7cWYxZQocUQqCu<_q*%3B`{G?p?8fxvu`6-oV2tnmO3hBj?i7X0pvLg0
z=+6{dG89t+)>&GzUvE7}KB+W_zy9&AMg<EYMS`hr+_}c-eTGvltxWA^uk1UM!AS%_
zjTP2fBhVVl04WZ=gt-`GxaOXkMkPPS(-;lWOEW&v?#wS<BS2Ww(GYHgw_GfVJyabY
zZ2I39b(b{Ym`Gh=ZT>++y;tNEZ<?Z0{wST;KFLs0vSGGxXOeU*BCl(3bk^CT;p?0d
zyY`wGUIN8;zyur>TyaYz`r4CEK%DfQmzxF7Bnm*)ZaH{QX_{Uu=@c;?wm!xm-KZPJ
zfJOhix>0>*K2+Lozm=~N1;LxD!>_glADg^*XW&w~@g>?+oNGIQg4O|(v3-}{38k0(
zmRnDY!i6VeO#Rth&PaVg8~#hDRqd|Mw}=u&h+-4czk~5qxjm3e_Ih^_m;Pi?Als%|
z8T<OhtA+pM0jQg#NsJ2VOWRp}0VsKNivKuHgdtb0s&;gy&gHyX;Z!|*;6<AwEKe*=
z0^29rGPUp&9l%N^i^RQcO=C%^ltP-KU!FwD;OJ|2sI5nJ&DHJa#@rtXY-dq{@Yt`1
z-=o+~OI7Hi+WAv-#AyZb#FfiY?Fn~1z3C3%s<aCBxzrVnz|u*BoCr((RmRn_IuG}7
z)diF5;d6ws5C?qUeL!NAGHv0sdESiF4t!wh4*{f}^*26^6vGTEMf1rVfuYsa!$(3&
z8j}m3N@cN(Vi<#;rP@N4vKhtdDb83H8HI%)pBT2Y4I)c1pW$F>X}sQ@;~jEs^;VHL
zq~NUo_syO${?N&Qg1IXQijlIrr{i7wrhN{2=`_^Cxmfm%Py`{1h6!kbwN2kZ>V(U#
zNe(823h*I*A$(pf7zu1QIfFd%0`OC?HKlp5S#*89v|O5{N`_H18>Gw@G;qcqK@7$C
zPZ*ZgT<s9Db6^312wYUb8X!Ggcv^GX=LPfH3fA)zgNA!KWdewIr5L#{d2=Iuq$qJ%
zeAsCXHQZpScH*M<n=%4$Loju|$@rR~B0(BS9u%;TBXu*&cCb4Ti9m7d=vZNG)o#I?
z3^t@EEjG{QlED-YOEeVQq}Iq}K9K5z=JvwBLleT1bX}sm5~g?85eeD$FCN0h7}S1t
zL((tx{ACC-iZ{m3yG#^i)(lL7YSI?^B1nhg)FP$)<(M8^-rBq0?G&t8FVdh|q+Etc
zg?OO(g%(!hYaqJ0k|8&$lqA7Dvy=@E3PPy_a$rB2IX4yE91ZMJO19ZrL(5SI2{{ht
zN%I{KbOnWc@Q^t@148H0LODQ!?8El+@iV^Kch5j>!zEz=`t@mV>I0+WA})0*+v2aP
zu|~*|J94OS#srl=32A(gLD-RdzrELOB<ZKaJg0z_WHDlX(&v>{m+ba{%+%zs@&)rP
zp*83KIk{kw*w!@45jWqq2tV<eBE;zuPV6PkAgtG!MDm(xxrB!32}gOscXgk@;9#R>
z8)yLXAE<1&A7py#DnEnU5#$SKjZ#U|0xB)-)Vs&ia@%P%{@^9`69#ZpIV0c(%4i{2
zb?2!v*QOfCq}1#~xff~Un&-CTk&o49)vEF}Ifpw`PqqoX`PjksHZBuBSC=vG4k&6~
zSATAF?OcnK?9Fcs;eE^5J=OC(R>A*isb9ZTz|TAOhf{x^XqN`wb{m5lG5d2?7L#ew
z_hsvoRRNEcnc07j3r6s&?BR7WfCt7zSTTX1@&=nj3ZOWUe(FuE`+(0XyXjgtvSSYq
zu>P4)oDm!>I$kGSj1yv@pMknLZ8p6^hL2(|lj=|}SE@+904@>??OHN%uxp(B!)vTg
zThGETZ0$k9K2}KRoYtUCV1jGFgICyZFF{BR#&|MJvZQbreo6-VbCS(bimdefVY_bi
zmN07Xrd9!Ei*Yg(f-q5j$T0jodjAFY;^`iBbR=$>{o|ETty*`$*}4#b7tNQThA@^$
zL5U14RsblNiQBOYN`g}qMt>><1uRCe$=>klR@&&^J>x=%go{Zeo}7R?3=%J|WgVse
zo0We?k&R1rH36jkm}YNw`$`;f8#DJxy*)1z02qS&MyvRsA`h@wX!p;B$^{VS)|)7L
zq#?kt?(0-BXOxRrVIVcT61SrwYmXd`(=WX6#T&8UTh+<ODzQ^z2pRqP3Wq;E#O$T1
z@<n(OieHNJgc8uWK+VWWlS(v8--n)IK2N?tr5W-oGvC;s^DWVYqp7j!DR9a+WXk)w
z-h7`S9nIC_$1ay?$rfdMlA^N|asi}9M!Z)n<<_@Z0`0N<O>=Ofr{4YR=IWW6fnq$;
zUs7qJX$-}T3Rm*%6)DN2eRZ8Wpv`vDsEC1|1B*o^oI|zf28+xz!PesyJW+*=0*HfV
zVF;*`$|Zer-<qa43#MS}f+%8^eVzQD{9|uEy53y>L`CP-r_CI)ERlS$Sjr2G8y^D_
zVN4LfPEUI1C1;=&7HCbo(<8d*0+rQuV=APYOJ5BX!;V2VV1N1~8DyQNETy77R9xQ@
z97!d55FBByYbg2AmP99N%l4Kn8po8f7)Kj?6pa`2a&~d?t8{)E3FetdH%p}7;3r`)
zIKBiimPocmSr8$R<nw7LJ@dE0=Ptj`9;ffE%K7c1V$WOhlmhsaen|ais9Waa05piV
z4sjOLM(98sH<zi1eVgrVY&<cnW;hsxhl7dfMP2F&SgVrUB|qfAuXyl+$%53~n7;;h
zU^t+GyoCJ4%bkxU?x#aO1Wk&Nm5A^^T(Z5gX41ve@33TCmZd)d=7wv`Le1tsCX#=D
z(YAR6br_`_EFT5hzzzZJezQ=awEv3NxJ?>I4FTd6`rb&Q6ND*=X1jacN9?AZs#V!N
zEi?Dh7-GB8U`uAT8$MPT;7L$EEyIH~T%GQ60<ZL7%9yFM?6IRLw~8LSF{vM)Vw6G#
z<xSnu`A2civdmh*bFrvYI&dLLno$EA%Z<PBRG)xDAtkPuHUu$R6@|PKzyQ6DX`h8I
ziV^KWv}P}n03N48%0!`#!LK@+28*zS$my@q&_i5`IR-{sh*e%MOI+p_+GB$g+U+|C
z8g&4a=1rpO^tn@+4fmw_k{lz;j-hwo2PPwSsjQDd&NMx~@wurV(I5@BOcaX=DoeF0
zLv_rcV5vDRUWF!_N;U<;Jx}65gV>D+4p0U-*+wz}NDzJHEg)d42!&MC?7l<_quw|&
zQxFQa+*fIL^n%5wHP)F|uUR<iKwod={{vJC$Q;HowV{rD-%7v^(>LMNTV6*SUjX31
z3GV)C)a+)QhR%pHKQ`W9q=pDZH-1hi1qsnVe600!$Z7iPL-n&cE<pBj&@9nO^g-qF
zFpC}tlNh<(;H3)OTN0vLrsWIj#=1r;vWnLBy$m0*s<lR4t_?klr<~jwZQk@apGgIS
zi{K3&_9f+VIz&v_OQS96P*R<Qx2$B{bTVXg#kAVmWG3VQ-#m{bEZYT-JY8g<bFsg?
z;@13AQRU!V@!f_9Niwb`Q*yHu{0*u29h0J#1rH7>Srn#Iki0&<lwzI^ZYpPMd!8sD
zfj8J_<(Oh1l1u_CNYRh<F60kI9yhzJvquC4IGNKxP3qZ-&*74=5-`dZgNm5#8{xS1
z1eJWk-P%D6k4jXa08$WeOS*pi#QvW1O7sv<I)_Y4;Zh`$M%Q__PLR3*mnbd=+}JR+
zJ3oCp?YiA!^}a*6m`4`hx^W@qeaALH2e%QWWo2wT6a_{5C(;GTp-GRX5|P|$(nzzS
z6RdGaCZL1%XFSkh3N<!aY&5A1Fbk5zI*vJH6VW0$kkwng5N-FdsN6$wwk}h>|JPmV
zJMo_xb}6K_b#v8pKJpvDe25X(os;XOBokEe=Kx}>9reqCa~mH2Ip;<$@|!32kA=M@
z{ky~N<%gKjMKH<qOaImfg79C2!{JcCm9KvpM_jM9f7^*084ymgl(l$8WqljTF}fG{
z^wpqP&t%fdQ?Z}Xyd4^|o?Ogmzg>o!@Cn{5zy%kJE&297?Ywhg$zC)c>1K1kBU5P2
z(_#J1khSt8`T||V{&6~Ee8a+mEP8xCYaV4FC3*H56o?TL2S#FJSmy}lhSl|2LO7Aq
zGBR3EaGs@Fi*hIJ*#ybL{*o?E-vdx^q*}nIF}<um5)cFHv)nhP$##y}=$9)1&Y^N5
zOpNl<E*amC4_r>HY)=%opkEsV%KrUIg90i(M)s1^(>T1biA9igvJP(X+#0A|+Zw(2
z2j{x!$(g*$RX=pGJ&QY-hQL|Mi*sg?KCXryqJ4^-*7eEHe|4Sn(-Sh4L}1L^IkUmi
zavA=?ajk80h=^J~U!ryciB&0aJ_DzB17MJa2^y|jV(MJlhl?&9HL!q`GAq=?C`J^X
zGAfxf0^a?4S!41TZgVOeUaRW5J76->L<QR|a(7kn-9H{1Y+-Mg#Q)D;`_X5t>z`@b
z4pHEjh2pLc1dyr#$8%Yw@JSNq$7x9cdwhxmQ7`Bff@2P}TkbGag9fsP-^?K<IyfZ-
z-7^{mwq;()Qay>Y`D1JEK65=@p|g0L13I_%ro-Dx7l+cgJvg`QKK-ukf-};fGVsjv
z&v>MD`+M3hcMHHemszyuhNDj$`?jGXY7ZyV61UT~tx4;lSTf~r<!n7~%Y>ABgo61t
zAiiZwK(N?1Dh0d*#6aNKF(vbh%DNC({@=Idg}wGyB0oy8wEx&Ms-P_oyxIPvj-BIm
zL@dRoW&W)jp!8KasQ64IDjCN<%|J49nxw9!Lu~KYZ=0@A5W;A&a3Smjv=SQ?b=(9&
zycC8Mna-C$Bm_~ZVL8tINO_%|yzi%HF@}n3-Lr^FtwTO@2g|jIR*ZUwFAe2c&6VNj
zt<8}`D_(~@C?z78uet2!1|pt#c6Np-i%;|pG78ze5;d!)wHnlunAz`#OD(6BkNT2W
z0JeK6*gm%Fw>TQfX@prVZTT)0kUrO)x`yu2zXg+gn!z~kY#_4a+@0q)bR@|ZZ38ZX
z_+qeZy5#xQ_mkwx^|NQ|z4M`5|Kp|f|JoeX$HziZYwZt4CyvhPt1DSYQgzMPw-4-B
zYEaZB;&>lv`8t9Idt<C-MP~W<LKn=*F6?r1Jg@DuykwVPb(Wm65ue@;a5`J(Oy}gg
z*=fyaS$(?KMk!IV+cO0{e^)N3&{1ni%nfq+aF%ba?i&w~=7@9T5*a%Q4~&G{|J#Gd
z!1{XFOV+w()M%wNBIb3PDeo}k{b-oR2oAiP=%!t{(=~ZKwk&1ym&RR+bAYD36^mq(
z)w_O_>C4q%Of01BJeqlYdE4E35&d}ne4EmHcC|HuNAtZB&LRmyuC9nH1n|^JR)*E#
z0CgqptR*m^p|O@6YJvn^T|?&|@yk^ddi5=A{KuI`q);`#2hA=wMk;?j_S4O{db+jd
z_s#2zfAr0AB^-16m#Js+y)6J+g<}ol_k-W*QkCODFLw<xDj~n**P&ynz^c^TZv@d0
z$qwFT{WKWJt}Y1T!k8?mS$u<eJ?vjIa2RdSxY#*m1&{!eGd4(c1p-$Hdr{vO{#3hq
zRW}Si3VVGP4JCUqAp&Sx7&TJBJ$v}b;>d8&rp0ERm+y3GT|WO=Wo$*k$M4y4^vzp9
zjPLEoDzDbhO+3gJoeloZK4MpaFsuHmM?M9hzDtEjnkq<^cv20voC+Dn3S!6)^|+I&
z!j!M}Z%-;?6Ze%r+Ks;6?fw!;p4zJ5m&60z-gtxYf2%x}xH(jg)~LkX5r2tBC*X9p
zLL%VG%sVFTb!D`AE=)6N!Xz4|cF5u5*38l>zohU)DD5dBi+j&zhqA8-isAd*jT&|6
z;-0J%9H_RM0`#&k(ZrOAA_WtGcrj@V4DDrG|5;+3uMmbdKZ&y~J{c{VLc4`ZlcXr~
zHQ{Bj=A^<DB{PIZoUWQ0LDA9iv^27dGnDXSjq!e{iu7}2>^O8p4kkw3X4xmVZ`0fJ
zTAQLH47YlnH+pV|qwaO{xuAZT#ZkW7c~(L`hcJPU^&@qznqm2m^eCS$VKcK#4>%#o
zBq@p6pwl*TZIqdYv9~q7ceM(q#y=~l@GOB34bB%lK2fdYc!+^VOj$?A=OuWs&{B<&
zOu12{BDt{D?%$s+DgE`?2$=Cuxtw2VtMs%0l8P_yn~xpm|68pz`0GD=rzXpm0Jqr7
zcHoomT`1>|X_g!M4Ob9M@=DZs`43b=AfT3Zsl>sThoZlm@<ym5LChKFv;!l@-71;d
zXyQi+aqb1j)B6zrltX(=#O<v$!0q212z9*S$yU!RtPd<R>i-msv(YJbGzpLR<@rse
zpGV)F8jWY~wEV5>bD+!Ed%ifn*YH<y)1E!J1CowCk|HI`6Wm3%nLZs&raflzl<?4J
zD<f(jg*!yZXBT$Ap_32Jz<(M|q~BKc{#TgmrJr5ev?+F(M@<7uTg*|*km|rriL9jk
zg%4XWS7HtUB*_zQXj&G+W=!0WkRV2@_$9$y;&1lVbL1Fbdlu07pWsEOS@tb%&Tf;>
zr$8Onlb@jIy!}2G9fP3q;dkFRZYrw5=j9<^%q0D3P`$$iH*rPEv=F(%laI)m``iAH
zTM-mDM!e~@&E7tJ$ELouaya-QZ9Xl#n_*@{>^u@?b5}gehUNA~F%2oW^j4F-UXsEx
z72GurF5exprW_v0B1J<Z$4tj5fk$uKHvNx5<^S_y_Z5hMd^|Sz8cVXXn_ds@Wqd}1
zYmps^#ERtoiFKA>{GE|m>)X%*BtE;7q`?zD4c0jt{*S!pw;MV%>yUBc&eYe3iF&7J
z4gU9i-0;pyNkKz>L)eIFpL^-?COWUHK7p7#_U}uL?uV<+A2QE_>Xi^o;dn&}Ief;u
zOWxVC9!?yb$6rE|4Thmp_RYynzXuqV#N?laZ6+tlEtP1{*3Wr9hXZUWRvLqo;IuL=
z3HgB1m72T%7KkQ)3%r*3e1bu>9=p)ZKYhzO)ZCaQXI^`fK_Qh@{biQg>EMuK7jD1}
zObm!6N{5)YZ6q96uvdcoN}OmC>1iVnlwO?G*4XRYw!UY<HVE(gfcvyB?-k&C<NN&T
z^*mCIQN|}#9nMTe7h1Z4CBtTF_gCMgeLUyglH}m3R9O3mum8hA`S1kJp>3vMe4bN8
z_%>0Ktb#39*47zN`%<x-E6pZd4s1%Nk_u3w1r!w;^4{@6M@`6Ts$j`W5h60s9v31F
zL|-%561G(J^$g$_qSG@~*Eh5Xf)SsEirWcW{x@&iuK#r_eO~>?O4YSE-sBvy-Z?qH
zaZ4fPc(77msCnxJN6kXhb=<%b5&<|)rO?wOuz&T*$+MkCNdYZ$T=F;n{0ft$qjBg8
zpHcK@Fc8d!`|M(o9DIU3=E-x?OsEntH9wOjQCa-sxn!c--tKZfAtFmqXIi|14<;DO
zv@FLbqZ6AZtMvcBP5k)p@P&+2o5E1TTajsMCpY9AjGCsH%c*oQkCP|(lGf}sdDT~}
zZud-G^Zs|uUxk!eY{_f6oH)y@YQ3VxmU=aN?L6{%eQH4Y;*R3lr&d~h8CQ<a*!ey0
zPZ|6FKcTxOo-A<Qm~XXAE4k!kiovBfJvDNBUwyn$b}ZTYx`k=n^C$l_#g?C5vo7xG
z8?Sc#HR?L)rhRwI)cN;)cKr8s{eP|aibtziyb>Q=nDDq~y4Xr<?w+k`nWkD9Msigz
z0$p|<x4u^`fAAV#|KE%99}cASs&I7xPe}X5*?M4I=#7s_VaDRgYuZ$eoPwHk9)I4=
z%(wRY4Efd_^ZI3Hs{ET!>&CV0<ewiBA2)bgp8de861C`pO8x@x{}&EA&#|a$0dA3-
z;dL{?;Ko7Sn<-jPUwsvAcpuyH+Xyju!tcM{{PV+SrgL6L7aZ4Adw2i*g3HO}rBnF|
zCW=?dGPE?!ZF~~FT9f~H!LyoUxupl+P7apI+8J|m-<g@K4dc5KStmPxX8-Y_`rh&S
zKmPkqzI^Vt|HWO_4fC~rTjXq6v8#AR(Sc<Dy2(Mg2bvF8-^{AE`T0}-)555^D$*0R
zm|qB=l2{&>beDU_I@N938-f&iJR8opb~#5&7BhEp)U451B*!Y__nTw)p@^$Xjdy+B
zbxeNAGBtI}O%{g__uKVey72n@hPzVRUT@gAlXnl-vtOYZsi{wo`TBhP5&R<}GP2Ti
z2KTf4MeYaxxW5lH_<yll<V^h+x%Ee4GX9=gxG$*d>sR3Te3aemwHM}dWv-RvYX>gF
zQ(Ri}G`?aY*Ry)7P@tAzpk9C0&+@i=qn@qTTA>}XB4D9YE0?I&mt)t~w*COIAVT*h
zXf^Eu%YgXija>`YK~;vl@LiyU&{-wCg5woJ8w);d|LvQ<uixQ0UGWwJ5O})!xvX<a
GXaWE(K=_yd

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_016.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_016.png
new file mode 100644
index 0000000000000000000000000000000000000000..a6ac50d21f1c7dd9a3a3c263553fd1ea44985d28
GIT binary patch
literal 48231
zcmY(q1yEa2*ENh2+@-i%DON1FLvSljkx-<#yF($k7I!F6ym)b!;!xZzxV!#5@AtjW
z*FQ5kNhassJ7@OnbJkvKZHTI}ECwnGDhvz^21rgy9R>!@_3e8g0O@V~VXf%&_5pN}
z)^X9WH+OM2ary$IWa8prWA9>PWlH7t#mU*q-tHqCAKOP(DoYm^2WJ6xcH93>V6%6!
zU>_5c+<UtTii4c4GYkxt$$#FkC8EVvFfc7MASrQ8kIbWIXd>N0S`L~Bl#D2kwtj5t
zcUf7zU0o1a6cvyR)7`_u%Kh|YwXMNTboE)#t!m-9x`n%7vgyw|Ody?w7`izOx>*4%
zx1PIe%6DDe(vMtq4x*-_O|n|N6DxaCeAWGd9bZgznQ6M8g$8;)`}ds^i~(iM)4qJy
z{XfI2QpVHkEv(u9{!aUHv;V>Qfa?!y?a2QOFR(hhGyk2Q_6%P=Qu^ONxgL>?(=<~4
z@4eox#p>iLX8XTaa{W1?4EUmK_}`~3h{9Zo3{mP@>+0&}8L-iY+__R}nCHX;p_cV}
zBcr3#J1<YSQTP5C-#K58YoqS%bH0rDKOQA=y-F#s3=gA5Mn>i=bbYzWs26lUseXG^
zLCr>9mm2<eCM)=5$)Y3bp3&)Gc!vw-b?Nolf5-pP{|U%vZ-k73Xde;(HD>L#IM!+Z
z;!C#I1qhQ&Bs>8Xihv0#K{bUOkaV#s37{i{^raD3V2lAn9ORhKo?zC{)P^NFeA>n2
zA<mfA<nkb)HCg7+q#*v;J^p@RBlMINMuI8nD!x7Jia=gI1e*zY78u1z40g+TAJVDU
z2EAW*03ZxeVTQ%w($iC6OE3)363DX@CRtWMkl>>xhphGCXHBxGNv@iGpLxbQ?o(2`
z7}~Opod>3(2kx^(?@~r6_@l6gynzLd`;I!+)mnIm;eP&|R9t&k^U?Lh)(*2%hqw1e
zuSTz4DQkW<cbWd@#jnEl?LPZ_!>z}mPstZL3TgVx)#_XaWZ>BvQ@V-{oj%rA<Z|)>
z|2ytgB1(yp?b8rvIqSqxTk_{)z8cX7oEuXt@P{#A6inKMx_nXj2TV$!ljy}Be$-u)
zJEG|$fNdr`<!5+iPkU>`f)C;m>ew|WgM@x<%gOL1;3u*wmOq87IYRs&+6%aS^jOS?
zW%p;E)$vq|k&nCYE~7mApFWzuKr(+*cP=b|Fv-&`_D?XPc<>`)`hvg2kmc-l#<QG@
zseRBs^P#lv0#@%|hGEX0y{gM2P}Kk|ss*}<6Wfh`q4<b~orbnx5(cQFj$iswKQV=L
z3PdMij>E?>?bF9i&m~XMo;lx{1e_(zB}q`(Y~EQ_BqFX|R%@A5D8(*OZzGRe?41FT
z)tQ6{Ha#g!;`r11Efvda-1a#1>S8AKrVmlm1+~xr5=sPCv4(cmOfT*pBwvyN9p9do
z7&)kGbKoFgk+{W^&O|wRj1`^_sNh*7HYI1w;Ln1&Een|s8|PeQLiX<&NokAIk#F7L
zWD3=voLOyB8nv<<%|#uEzb}ppqv(2+$N9t&&?|?vK~~Vufn>vHz1Po!)CiV~?GXr*
zx5pG3UwV#HPo&=5R@=rilKw84i9O%Y!$YQwcEsnsj_5HEyG|||<sdkJjp?<&Mq7Xs
zJZO>$kMN`A+vB;>`YJ0`MDbbnGudV!s-eon;6EXl>wLlkx=MQA9@EF<2n}YOwknlr
zYz2nJTsA{b!Cqe@l<gzYPS>^v^R~8~fpyqAjGe%cY9_=n(1}ty0U+&Z^+$Y}*`LA=
zJk%om;Y5&Gls6+?(V{RXlr_f8u`Y6QHV_Z6!J$=f9bjG{1GbJfuk9Zh!Pod$>*4o~
zy*%!{CXPDwS=^Lqv@&m1y$%^%n4Fy|E^*v4m_H1}R~d&}-U$j5+w{bxP6XGEQ40q;
zpzgfStjX;}iB-0M93TGypz~v8_QS)OG<)p`sqnRTZjT6!>4qrK1kUP3)dz_4pZSRI
zCEUV=xnfdldS&LcWBoWFN)z&oYxI*Gv6esopfV*_3eVs9awXez+Zf;|<$&%eMs>ly
zN&c4%9sH1$Dz`(dmYz>sJyDX^I7_^yM>{h9%jJaXSv>gouPX_uwPk^$W%+g7&yZB^
zK~XPQUH*{wI}ZGrFnrVw*_{M}6Zj=WWCx|N${+`VkkZZe8`g(t`z@QdR3<O`OUP)d
zxGd&k6x`tm;OyR8#=m|`3n6ee=YO)qA7kWpXXd$B)IZwK06c-lsdqkm*r=mlc^>9}
zfj(XPsSvR5{yR$RCxDt`Hm^dOB`$$3#9CuOF~Gr&s1o0}Z{kw>&M)BfQ7$dyJ>rqq
ziX`Lv1yvBqj0v{#21vH-6WI2eu+1*Ge~bZlG+q*oH8WE}UcAvScyyFR{~8l-8jVc&
z0()nedp58J&f<fz4jCb2J{lEuBVjNbTgQ4rK$%dD1sj!`PC|$3{3m%2enqFokh=b%
z0|PL{{F%+MBhcdcDhs!|9p8qpq1pA4L<22Sj?tn2M4&+(5RxRBf+t$dzoD8Prpbwo
z(+ub>?l|**s)+1J@~oOJdU(gd=hta;FXGCK&1NVZ|5R^1=V%vKZco-3A<3$@oN7AS
zm6t3KdHiQvI#}EnFZhQ@N9WBRTyn&^P0=4Y|IRV9WI_Fc<)&%lvN>2`u`K6s#y{bA
zteaJ+^P;%(3b-<z6@ul2O<BaAGn>LlIuq>g`F}e6!EN+Gy7N1@Vob@Y#k6&C%2ubk
zB5mq;kL|Ysi}XL6*8hplD@p3<pVC@ojy)h%|76kJVfC-0T9>l^3et?$n}bJ1O>_Qg
zj<5tw@P}b^$xvJbeh}&^orzjPj;`4+F^Sxy908Zxsmh*LMw}*(#(qqMSzUw}@s_@1
z1$#&LdzRU&LByDaedZti#U+YKOoV9bK5zsRwr(X$B<@um1EZ0rABEjhB?nH6^|F*`
z+_?;G1YI^+gX(_Rsjzqu?;>I?3lCfd=j{tqTbrC|!Z7?$>xib4VS$`|aLcW>M8GZP
znX5F~n6$_qNP%UlZ83-GPrdT(ihwa<3Yp=@*VB*UWGcD|zSTDWTrqOC+D#TY<#?cN
zM)5v}@<HTKeKQ0Gxq;r1JdhNp1kb2DdbjJc6J^^Zc-}DzsyEVhPPj#??N}LxL5BRJ
z*w`l``YU9q9`5r>F(1hpfShI6sbtyYAv|A0s4KdQ)!L%J#(I4<Zr%>#b_HcwI`HX~
z)eg6}u;*{&gj=d0rnh9i|Bv<h_Y?2d4cMPmaa)-0=?vqajQB%&;R}BndmsKA6TSb$
z1W$d_)3Xo&1F~5P(e%8nSgK1G=AAWVYxj?4q!10O!j~6F!q$0@TeS+zbhw==on5a6
zSh$nZ7160bcw~1lN-&&eomE?M-ecf~I>)#b53TtgkL&6;_nsF-vW|>m_u*V;k?2GM
z;Db0~Cvmbq;}Hm;_K~%HjuqF0Ui76j`(7FnVrK13oYyKUQzAG~ZF1wM_aZoAhKPBh
zmUerOr7QJ>p-)#5(kCEAY{NNO1&8#khv&0JKM34UQ48&V6y}vF_zb;hIi<r9H@3A6
zwZx8RC#vb^v>BVXmS>K_p5O45Do=cXD~ikM3Vj8o8u`7FOm0E>tMD_X&U3Lf%L7z8
zePrcVH4{jupp`eA7$}6xg78<?M7dRQ1Iqan%v2EpJoC2Etf<OL+(Z7iYtuSyy-$<U
zyXL2y?7Dx%-&Ow|#M=wsGD!qM_X<bqE5UQ$frMH>w=GudNh`K`|A^yO6zD6;=AGK0
zU<ei&vfmT0xh?YJX?FKJN^9%zy)ZG-Go1Vod`<?83`r-A0L24df}p8+em{S-czr1p
zEF3Mz9m1@e$A2Q{|8@7#>7Z_cX{`rO1P~ie(cib+`nhLeo2tstsCd$jgOe+!+VnQH
z0!Bd(e@~1*+JS_R)l3t0{5C;Apj4e*)N!xy%I$Ek{IZj{T!BK*`k@W;wDlRU{c(ko
zLnOh%?N?uVB=Ng6<F8IZK6wF@A(miqF_E0&H7nZ8JRJasK$r{8;*rDF@Ozh*!)5mq
zXGP^KDLE`CO=hA_N{C`I6=ZjTo9tlzg9ie~Y;JTvRuG*!R46F{6dH(VOY7A}7_6MZ
z`1ua+Cs<=X&>Xx)GRI~pHcTP!QK$dm;TiTX4jS)*szf-q-d-F_=_iM=9uy3@Itv_@
z?yr&0=n_{sA@`$)he8XM4(=|Qr?ehxUH(}3ekhl6%rCX4HTeXSWvDH54%X{PxJw;T
zfBiGKa}AzQ3<QHT*=ZPkSYlm)*aGI2Paz$TtL$8qS(#!`)CttI@DCZB<%WX=GE0iS
zfJ={O0_D-lME^VLRiLP(!Q=bhBLBc}(!qjHA({Py3~UeY#tdPGsTy!LK_UF`<pmeh
z9_P}VCn_}!fdnLA8+AnLu0&Yc9V;_?$YBM}0gp64=Wq#rN$0<4PUGWRibhx>gf9FF
zM<VYS=8$_hR4l>NK;QOPgC4kCARYfx!tR=*9GG(~pkBuso6fL^=~~?X6FR*l3>lki
zI!w93j%0b>npJgeZQtUsVWm-9uX4p`?liPF-fGu^k%=HulZ7Bj9U;5$?|k0$eznUX
zW|_P8nGln9L<lg)z%C>?+12Sj&fGUN=VkAZOt_)n7t0s3NGOv4V3BZ3*dx$VH$UB6
z*FRnq8jDmKFi-OZw}$M;mkPuJ@@H`@@>|_D*NjE;tk8X6zIZGZ06RYn`Yyjo1s?<8
zg!sD#iFv|G_HALBzX->4qFi<2-S&T)&<?!_7WjSIeUZWNx52k1$5{Un^5}=+EX_65
z3zLLsx-5<nPA(Be+uPI0Q)n5a^Ml_Idj1F(qhdd)-}WbzG(|MYvqRlPgGtX2MBUgk
zjYlq!a$4<W9V+7y+dLdy-Tq5igmTKhf`gtkQ_$giqXBH&y>>X-Qqv!mAo89x+w#9+
z@Rn2|z0HS!YnWeNgx9ecHyJ8DS%~Gru{Edb1{qQm$eJ8OSAoF}w_h-o6!Xhq%LKv_
z;9GOR?JG78#pLImfocC>4pGj_XP~Q`&zfv046{&f7$;6UYheeG%hGq!xl@OSZIdKL
z64Y||#DiZtV0B-@M*SRysl|E|G?&fykQKc6DS8~iqUwMcZ=0&Mjhj)0lFouSc2$-;
zMI|C2hiK2p2>P`b+S<5DMUep;-nN7fy`zA6p<cd$(C|Y)QacgH-tE2%$;_dv*%H{#
z<eQgm^ql_#IRg<Um#DW;Gkgv2J`7x@5?rBoky5#m(^^dL{YTz@ooJ1OLWo3ep$hT#
zo?8n)pYi^~A^-lgPKJ4HGbDjRLX22IvHaX8C;lAd_SMyAya#sCG6efhp5A0#Ghrv&
z<&W@ju8Mt5l;`aOxr{P}xdv+Ww*=aKx)<WzYyu$*^tCJFEx&cYKyNKE*6R0{P07+n
zq!@*joQ^ExL=<-G6-Fl2`)UusT@CNlu-RX#Y|a)ey>4{LO#L}*oIHL>qdDN!@;R$h
zlanQ+%fIYS;A7vAaEBA&LJtHTP^9j(9_mKiub9kRwbyy7!a(xC1Y}hwLK7_qy8vfG
zFc|yW0MSMH2du%qw>3F?Ii+irNk2nVjoLgD%jWDIpZs5ow|fl!iy1%e=7b-#XZ}tr
z6<YZb#^{?QmwkW;@-w6oBPk~^Xs+vyWUn8BqJs3M1RMy<m8EXg+lbr>v<XcRjB$--
zey9n=w^ri7QDby}wIQo%aLS52?GWeQQ)f!9?qdl|$^<qKZg!(<eB8oEL*Bcl3CfIx
zeIq3zIlr0LLKq1`B?6lltS;_pMzs=UwooPXBvgXFy-?*(uI?sN`UQ(VB)-05h6~XG
zBxcJes3S}1VU<h=^)r>CxxhuiZ@ewIKj7>fjb0Nb%Z><)L33by-sEkOrEHe)B|gD`
zidY9!#3jRq9<CI1LEr(g{MAE$@qyp8Y;w#vv<Cz^gKO=~qiLgT&55|5Dk7lpEge;_
z?Ccrid__s4>~SEUr{KauV8brK3tA7z^C*$4KfW}(7g!<^P4{#(&-}Up(?B#ieh<j~
z<(Cq~$*r$6-(&EMyW6~3#-!WZ>zg^(U)mu)NHGN|<V}!zoY0HZj!`9m=qAi(Qtiv$
zVh4+MTdyJB8g8B_Md?!?+4u)|+Ph|Tg>kGIEE(5M7ET`G?sos1XE<}dK-ckhhp5x%
z<BDDg4y=THOqPL^yRX(MFL4I?ejSE_LdJ|+`p780Lr=C^hGnPjM<EVaE}WZWbp|hc
ziS_c{hz*3Wx}A5?)o|9x%gwH7e~3O8{9W#}lq{g&^15Ihh$S;M@rrUQ?walA&5|IZ
z7v9w26cz@tp`QuGeyx{pT%nIK&F44ICDl0`BtWe@DW9k_)SBf@H8POOfF)Y<&-lBe
zPbTC;!iXR<xKfH2Q;3Ho#vH%hOCQja?w$2P>!7zQ5qAd>U|*s}tNJ<obZ1}26@jxK
zUG?*nvHEB<2QOz<d0=g?qm+n|yqVf^8wxRDpe98(tW0nmlPDFy`LXLU@22<Tz}@7U
zKi4k%5jRj+W8FiKzf6?+60Eg}_6KnG9lI{D*0K^Ys5ul?*6#D}iD1>_3-+$YG*~JN
z={ed~YB*CTJ|mkB0ACK}<W-kGF*$L8zvmo5zW$>amA!HDQ@n|7*ibon$U^Q4y)lrj
zAHCY0Hgl-UB<v|i@z>giLHpQFQn2x<@zobE&Y4B{asoN_-ph+Ckq}Q{xvKq<F_N!9
zu;0Jgm3se{U8Hm0-?<1U4N^-3FeAS7OdAOd^y_6gHFxv9JlHny>MiRo8hbf?<r9WD
zw@8FaHpjNbnpisW51Cj>pk_`mG3x5Kk43*fHQwaSx*=5a@0_`U^~w?mj-fudonJO6
zp0?uT5e`+SFxe|W7H0TS3<#nETCFhxE_-KHt!H$N^z*cq%)N_y9`tfJpqw2#R43jx
zGqWDgjH26Td5Qc*aaJTeRuL2HQ<HS1&U~#J3*Ipx<Z95TdU*8j>ZK9tK$-Vjnm;vW
zmNKpcT+^c&)PsmTGWmUD((a{MIfpOzuf6ai32xj*_@SM0V$=>TcsvQai7J%&9nDy~
z&qWwxO3Ga#X_U?}1!Fj{76CHDPw9#H->WoQ9x&Z<Ms{*#t-Mfi5RwAgPc_gkZR|-I
zN^9HBCr28;9*$&*3>k{{$~+f}_X9)b5n=+~C@t+othDd3ON4{K$TP#AZFC%HaQBsc
zm{c*e_}|ix^E5Hl-Z5%t<v&r6$nIiY+*}hxskuDbdbOGU8<z0Hb~l-NJ(+tFYPxLH
zp-@<jkDtaef&I{qE5KnH21z}vnpv~U2E(%-WreHLgalQxlqV>xEDlu10y6yRo$kU0
zIe2GV-Ct+~cHv%MEV?wN-DHyB7q8KwLl<&DvfycM9k{y#6gkBtCof%j+2)sE5qq`S
z{kG&FmI`Sm8;Bu?ZVydwUm&JeuszGs`T{OuWV{K+RZe8=3|r*zud%Iu@}Nt08;G$H
zVTXL4!Vg^fTTVKb(K5%+nW<66gnw}tsFFHPTl7}Z5H{LnT7Zwa^fSutK4S*bpgU&@
z-vW}q0!uyP60};fSQOXLTp45QBz5kGT9-Rjpc^N(W*3s>s)MJI&BwQ#k7jI{k~)!9
z7(j69>^Bs=OLgu}1A*8(f>eF<A9M6DzcKU0rG~~Grv=CKHn_XkO+-rU6VPJTryQP)
zIgBSxCr*v%NLs^qM5wCO(4LKiNA!n$q&CQ=bZ6i(?}K`xdR<zCH^rqISH<N67k*1a
zX;aIdi1kSYQ*~&S;{J!Y_#b-jp3%I$D$DU{hV{^$XVeQ13(Lr>*GsHe+rq~uWGD^p
zzBP9Vce$xSre*pEjRI3q!r*tgq{PIo7AnJoBRSawN2RunI&0n)L4MX}zu3iH)Z|~X
zHF1)qC=N*b=<eFz`8huJ$(72FjpIFd`s^g4!P(fTdTa;NtlR`}pqFa?ym}O7Gi;C2
zq}ck);wairB3YjRb}8&twj~XrNx6h<GMwj1rj`}L*yw4*krmYUsF_;%W?>VjM3FcL
z5Ws~>F|ofZGIjlRF=b*HLK7C9p!$JE?dV2Wx|s_ixn&yJ#T9Of&C;yTG1MxE21u40
z03gE!I^gaEUWh)2`k(EQjb<gU@jGKF)UOqEG)uf_8?&A7Axl5q1o#nee)_4oMKOMt
z;uskAQ_J%eoQk&8hvLrebVLWC;Q{>+r;v;Ma+yjH`rIJ)As|yx>gP^FXB|T2D(I_x
zQP?=U#VQ6|h=Mt?wq~IC%;-O49-iaM+}K^D|1F6;zPLaVI<X6Z=<~VsSQhZlamL}c
zvu6ys(RN#-eNBU2LBq-$@&ZYR67&;q@+%YJ)y+cK3|WMPqHK<Ns1e{Gyf|%s!JmtJ
zHfRS@iOX^e!Vk6CrfqJ4{6KlIcd&ACSh0l6MFF}?NHLwG%Jk~KX;$95xwyb?CXjuu
zRYUrb&5FcuA=uT$Ea@wcY%PzNC+zezf!;9^pPd2p2p;5{5r=8bI$OBaS>DMXT@Pv;
zUd8{F1jk56WBW6l+!9$}wzt1$bIX&04KscU4*uJQPW<o>{uHTu`5h;cSwcGJ+k~8@
zI!O<m0gj`jb6<Z^57C*IlSBKL#nj?jbCHB*r8HT}p4Rp@@mtB%rR*Qo(?zG~0+lB4
z(kh^boH);_Ib)yh$x{!2g(GkC$K)Pm^96ATJM|92IyM+nO}=R%My!?9Q3<Uwki9c8
zC<buukK#;=08q`Q(mp%;^zYUa^`8^rA*Sy(|7z5?P?MYV16yljub};M{`AFJNS_|4
zn7p>(_vl8Ca|<%B4ac1)Ug-e))Y`L;A-Tl5Cu&%KI>mLC5^6J|(744HFsUdLm#}|8
z0+GYvF2U~=-o7a&r?}RfN#8}mB}=@?{gVkxzS0<#sr~DP`SP0P`sgd4I!^8kjD|+-
zw9NALM~eR5Iz0?6?G&IZ!@Gk>IcWQ$Nr`6Q-<En8ff!E5sUJu8B34?Sjf8m`_UA<Y
z=dP=FLtiKaxRguhi&j=Z^u|Ya6+u*=j@>}#R>aHGSR?{nZxV=-L{m2OnD8fD&Ja_Y
zN{@jY3T4leD|4y90OSqb!oyLY2_ssL$a2K)VyWK3$=H4FDjMnQ?6l}@6<X-Ogv*TV
z+BtKw=vwN+n5}VpS19R4$O%1rr7|C$cB#)b@i8FkO%hv9L!eGg%{G2KQ1W|t8ThDj
zI*Q#bE04t<*K(Nr&%QFS@M7MvDAEsLC1l1&jXlX7crWUg@>zpQ2?s@P!z6lrcJvd4
z{}Wkzm0atjl+WB^A);5+jXW-Fi?((c?PN4HisXY|=kl`9`$kgq2SdFIq{Bn~H<VIj
z@Ws^hTVI+qCWXjQ|1JyMAR=Pnm8IK3uUbe9O#|<o&Tr@7&)~3%QClr-eH(7mpR}k}
z1fd22gO3m~hbmco@#4Q*2Y(fWZapItaLdRY3vuFhn-d+WamCAd9P#l;qw(S1cjCd>
z^BtHsyWFFSJ`Aboe*Rj3%vnoNLdY}h&a`h*(Ze5Wb`=rpO7by$zae9=)G7&=IM0nQ
z5TDm}9fmqlx2dP$NX5E?#d-g|9Too*=SVEvOBjbXv-1w}$k=G@3(m+IMaIT;mh_Kp
zkaCW0U#ziU3gi5wB7qkbH%mmxNx8|<OB#{)2cHhHWqT^&(AQ=n2lwIB41Yv}{@;Xd
zGV8DHyDs#*l-8q$)ORg#<*#Ug?dYwmVt}Ep0%VWd>svo&=l6(%#A>*y4q_eT)7j|^
znn4`@m$3fNz8_oTE8zNlUIwC9r^>Ve#;~zF#tz2?CM#S{Q-uulKVSxnE2g5|kR=)0
z`}a>MtQ=0D;8tXsWY{MpYHxSL`K=cy^}5*xUq`$C&JqbDl*hlMZ|y|hcf6Vhz7-bj
zDFvz|d4iUBvN-MdR{|dz_~xwY-JARlNkqM$KOhlRs#Vo<$V?pb0-R#v<FUu#MFC9O
zPBU(^)x7;05SRi7t3a6jUNmdBP~uR02tuT@rET2si5o_ro(603z0pg`E2(zX=k3#u
z^SNc+Hc(T^;6}B&DLOh55}0)iB_Yt0nIqYNKV}<KP9B1fOOJ=lA0JcGVmg>J`^Eu?
zvl*`vj|^~N6`Fwn=2-;+R5@G&bna%T(l4~BDj$+`@aUI4bF|jB1W9kFTR&1YXzy2c
zNaS&a8D+~3;LvWdS35jO_ZJ_M9T-TqT@R(c0VW;6$R6ZZrb_idrYuXEe(bM=p0iX8
ztRHACb-Q4p9$myQ{k1djkO_)*C?^kZSDxq{`K;D#)797^<<ML0|Ab5ZE&{PtMk-k-
zw%D8LpD<z7`fBN;mj9K@i|dw@!;0g#aIj1m3zU-LH^OG6NR{V}<s>fS{C4Fjd2Quh
zrNF=}pg=ANk)#QcWKR2`3g(?{EV1PqggOIorVlaCo&z8dtB2?C(F%rw0z7o?87-}$
z9j6fw;8S&gI&}gaz8-l!ljyiwm6$5iM=pV}tmH4>Pfi?mc6n*R)Xm~iM2tAyHIr&G
zGBW4My$8hQ;+t+6)VOPG7+I)+r}slwk=*3g6_Nuar9671P9}kmCoMU6j4`z?kG!Hf
zdim2h9^D&eQ8DBz%2<gkN{rTCRV{?n?@+htiQ-`rCdxtnFD54EMx%Z_yJN(yu_iIt
zK9T^v&IW9uL%MMq&hDB@y6G`$k#F}$z1#)7rY?D12StLK!e=T<CN1fW1t>`MpA@~{
zy_YE+2+AauzPc>gLOk(q7+GdC76A;D{`D#J$L}t#)}}iM5o^fll?KLr>a8(b-AI65
zRwd=ze8^KkXl#Qgd|X#+jZQgtu3pe#Fe-I=YDd1?kmYdiN&1GR4vdOhzVE`aREh2K
zjs34YDg=IsZx~;w+IIo0^-!zQC`!~?mVewA2g%cSK|7w9R789Up{zB^SIY>f)L~AP
zgv{copw0cYKkR~xxC4F7b^0^;ngLL#u8}>A(nT}ejyeJJgQBuB4ifdw6hOZo=~^bb
zGx#g@KtxPxW)>~F6Suw~cSGZRr7f+dT&2Ng=LWbg-BxeKmIgVca&CU{M7?60({aSO
zMY>#n`F5SiaS?X_zI+zM@(!XzuAVk#r6IrJAZNB|C>Z9*mUd3q2Uy?Rpp_CZ_1xiQ
zQSI$6n66DiqZ`YuO_iHIbaO0?iE+h@=iLvGM4fW8JQHRQJY|)M^LogcWJ)u_a}M05
zN?@pFli~`b{nE~gs4@Gi6z&eNtC6Vj#7%VZ**@fJ>rnVy5w44ss7T*Y?ZEsUA4Tvk
z4-2Dm%~>C&Y>$4(bKGAv`p<X+9+r8dgKseRxl>YADEjqmdQg@wAD`A2;$wzHgmTom
z`2bh@6PsSPm~P*baGz`#C{K(U32Y$udG2^24}_lI6{yfA-O24H28*i~TY(r8I*kju
zzoi@Zb!Yyor1sbvDI3<nR0!~6h2Y-(48et6ymlB1a>YlDVfVc^5xISme5({{Ae?#y
zLTS5@F9*?HEjR*!o$1{2DKbCYUQfv+SmJ;|1_WRPNBpg$0Dd<`<!BCJ;Wrh6@^#Of
z`!|k#`z3F248_LwhU8n0g3V0Pi||QF%M99FAj*uJyDOg#R(Pi`ofV_Mr)6gseAJ;u
zN2vLJ=a=o%C#x?T_uhFe&H3v0N^fZ`@8l-VOB;=BGE87zLAqPJfNXvXZB75LY0H|n
zmG$*J>HY!Pw45~BL8#RYiJ}6m`Z^i$;hsx6U9B@Ys8D&`)9UNu@QT_O<sj~WWSugH
zAwO}5ivPx|K$@tPu`b|X{}#{{fvuUf#i;J6y*okKwXCEvi4#+Wv{w8nqJ+I;G*ze@
zdP1=&c<Fqk`GZqnVBdn|>GLJ-{Y}8qpR#%7u$naBuJacX%AV!%WE4uji0(~lc`(VE
zZn}g?bnG8A8#M;^67qGz`A;YW)#_9~$SfrdzANtFBV#FX@d}JbTN5>TTnk`{`kc_F
zd1p&_icH~A<(-lV3R+)jp&JPrY{X0A3DJpt`ClW#Tc$ZTuoqXV4o{Zw;n0M|p)&gx
z0&p6_Tw<uAJWu+Zg)W<snZG*$HcUX7xXFa-3@iDRehwfO4Z2m)BRgD3FY}NumPp?t
z)Cg=14wWgBZC}?J*&!w;f8#lwZvvf(eSQ=dzpMHV_{4>$Y^~HWZEEX^c)UJwo>)GU
zj!^Rhd-tt4^~D?+S7qK|#-}^Bi*o)hk=MJ}Pgqb8IWjVGx?3;GiZ7tpgE?+Hw*W5|
z<buwc+2IhFp-^Vf<P!g}7*l<a2yjV#oK8EHz%g2`hdink$v+F~kz2$`1du9Z;@!%W
zHPcDiRrEOzIfs?5zhMZw-GbQ*dg>-x0#TR~Ko>dn9yLmQCjTRT!P_bxp{eZAF~4xX
z*T{~?kBfUcMMdwc{iN?Zkb||a6KnSgvA8(hN<Wi#MRK$bKIa%P1CB=<b^8?l9#nqe
z(3D`vWN?U?9Vc&bA=HM^Pj(<(UPCQU4rj@wI}eUY21okbyT622T=PQ5ga&1bL7%-S
zhg!FpolLmydj%p7B@Nc3ZYygpMn5%VKeD8ZB>^`PV!^8a2u8*KiB|oVgEBa(KLwXr
zC~3sRHKy~@<7$$&^=xV@NuFMaUYtMalTPupkRz_ol5O~cM%ec!HsQTI>tW%6F;Lf2
z;FizHrGKx{OJ4_UZM22qCVnV2EiFVRzYA<)mtC^btBcW1Mb@i^WsSatm#C<ube_!K
zE>*Tr__<0WE`12XDG-eL_p4kXZCwn?6Is2|XswE<81e3Ocv=ocEkX)eMm>@&7n03s
z_o+b9em_<)s?mM-vktt`<HOpo8a1$sOA<1Lz`#<~Q;2eh<T`UX=z+#-1@qYt%ld=f
zKub(MJ}z`czauZv(_ZRNamTv9#t51O70ep&r62$m8!eC^h{KIl8&lQu32XkYKgTz~
zlO0p{RfU$7J(THhx1c@^E6ErZf);8f<GD(2$XNA9fg$}qTKpe`z_001^@n81>Bgo}
zkXU015ub17_*6xk-*ZYJElJIIa8?*_iLG;)0i`a7?F<fraEEj^OSUI|kK=d#tG+~{
z_)lA+ci^!s5lMINN7}S7zs|uyCTd_fERxQ2_dMq2iF9v09_!jIhiqmKEZ0AfW9BBs
z?*oef2E>_DeFwXn3I$uod^o~_q9axwD20l~hC`Jk4>rvRv7+Z!Os`To*mL6xQf#r=
zR(<?%{UZH)f=<$o())qlA%^mK#lI8tASyNr8<3;AGX#y0dzTL%Tpl%L&1=c69l!V{
zBb*gVeU)5I#Fu>^6{GnshuFfUT(NkL0Ua#~jk-dHbcb85W4tR^_Wt1Xi8?!{XMPV}
zSwDfGz7XJ+#UiC{`c<9Nlo?+FbP-BZ12PsqzoX?o;Q4)Qdm`vaNwnq@dy}hBUlOln
zTu3`*_&bX{iP9&xLYGPWmy(V8%t5DKv@;U*x9JaLIpQJN5^66psX0UmJ9({2zhg$l
zHX>l#YG{{V??uhF>O1-{4x*x?&a!Xbu|oI%YSr9icnC@#{rPl(MV{}KA0%u!n6tm}
zm(IjyZ#lkhwW?b8r_=t1W45T;?t*wme;5a4nvh&9>cf(@YL;`;qs;{*>H9Y-Rybq^
zNU8wBjMzzlS62W+OP<4N{oafncvHl2+hc1*d!E~sLe%GmHci1QEmI^UTHub(iB~;Y
zPR1mx6MuS)1KQb0lKB5hF3+!T>g<XCI$+IkrCWvshYm3TRY9EJeDg!^1J~+gaLGiF
zoJ_FK#2*){-yD<Hq{A-g<avgV*aMnzg%t4Y+iB1aF&_5$R_FMd8a1kS<%a`t>_m`F
zLazy#0dgVw2{o(@F@eBIoF8k{n$nq9s`o;Ud4pqPybQd1fn6kfvN4~J?nIw*UfC7L
zMvSw>^}>>G;3sPgZPV)Lfk1_C7-jrWj#(P%DbFRFhI8Lo^%}9-b|tbtos2q>#w9}G
ze2y_wLtI#!`i!J9FbpwxLT_tI_m_99$)2j7x5~JTFH&C}<4KOGwd>>D@G=}ZEQ4W-
ziPMrC#Lqfy73-DI&~SYp>KAL^4q2XO8pXXEpxvD1ZF|6gJGE-vMDvLBby@E5E)V-i
zl}Y7*)BGsFZG%&DQqz_Iy{46>IDVzM-0j3!er>U-SeB?0oIfR#X;+ul6q7^Npi%}(
zjhPsIhfLiOhf9jVdtmpfg>{=Y@|X^&l_rIQN3$}l$$Za1a`jmvQO%u>N9%BJz71&@
zUThEWz*#;{VJjuhIQ5k2%0I%w?cy67VVz>V?gX&zVj)ZDyh@AzegO0Rvu{S9q--gL
zXb2kX6h<Udq7`nB_-&lI&aG5~TO@0kD0r`Y&`?d1nLb<dE>vwy_#E!xjym7&V})xZ
zwjM1<>gUMOqQj6-p)pV)Hq!<UQ%-VpOF5A<8PExx8J4>G%_!%x@7}V<)D0XYD!xB`
z*?CR=2&F)zR#m3=A4kqAHHltva3)QRK~<oagUM8ikalh+6Su-=78`(<@52?x#wHCW
zjOh}~jqWw(R5p#yTRYTX!pNYXcg9(H^lPMsceBR1t-LO-*(sZ>%DRuhZly-(b5UB7
zFfHo6LAW_a+w;phz9Fn4%*nrdG&1?UWGb)6hJ>T6<!DOwDwlR&tZkXi6p%1Kh=YYs
zlr&DncXA;mGSK)v*5m|^T>K^VTMn2l$T4UrMW@5{9Z_R-{<r#9!cF*aHwpmG`;UX}
zSX5>>76{$$CGX&g#Ae8Ds^~@nnpD;$K>pY`^={ReLZaKdEvM&a%U!WwL~1$y92e>9
zSWvq)eGb89#My+!{?5ayU1;15{8t3hSXxW1_98M<?<}tGn&w2KG1X#DLMvha+w`)f
z-R%Drfjw2??MQa`xxic3D~<<}<Ku~Un?l#mPgIJ3%1Rm{57I>4zt<@RL~ePEaqv!=
zTf0J(8$%r?%I&Lvxg=&Lgk;Co1m&j6dPeYbrOrES^}7af7<<{lQVT2MPaa-(0x`!_
zxHz{i8DD!|NA@2nK`A_tUl^CmEyD|#<x0{)iPZ+1#_`<j=+u^FoU`;B)};w{I#36i
z==KHTL$c>Vu5i||pm<C$70DIk!7!7PiTP|scxNBs4w+INH3^UZP&G&VS-Nd?iRhA?
zy2M9f7XZ?b`>c@W9@+_jB*O{9JRZR4x5WJAoFbX6YF9Y#X0<LuZhnmy#Luzc_)(@A
z>^}zXbNzi!_<=|@9HGKBcol_-2)*~4g`HusyU;o#EgJuZS`00ccWSJnSDh@y`pqmV
z+kx26M5eQv6V%z8#7i7FW~}@f4FIP2wn>Tr;48q9lnB0&#pBRIt@!C`Y&1ZQO+N7y
z9P*;Gi#F|Skj@|UmoTr(C8cJTsuz811n+vIO-P7_D&_yT8D=oS%0W$RLgqei1?XB(
zJ6tKI8_j*MZ)D>T5$)N$95<H#BDF}5w{OP#6G~q(uFfR366xf~PQn*8hEbrZ4v0g|
z6Ux*ay1grzLO;Gi@};%p&A3|n(4fu7l%x<fHi~--c5@@KsVB9ux2<Vt2%5!7b_&co
zwtb@N#gaI<RuXt76c-uCQxP8Dn8tEG3LM~^rDo-4teKwyDHNDlv4<76($x<Oa~jJ<
zFbCH0Eaw{@B+)p|yF7&{SnqK6>(Icf9i16o2w5BAO2j<yTTg%0%qYzNlK&V?(6rL(
zeeksCa?xv{s?lP!o2>p0(s^vcCWOU|aEdA&mA_7=AW<E#ou6b7pTN`>>?~p2h<Jkx
z@A0bNWJ$`I<T~l<wf16-ne7|5WfzZD3EFUpls&bpo@RE;6IIZO(JkRieKYPkMAjq|
zqzmM0vYj<Gj^0R8ABqpcyYuZAjZk`S0xSRGulYUW2Jz#l5%HSr@CM>a+)*L%lPxay
z{{Dgh4(M!V$c?9wzZJt2K=ZJ_srHIHBBvV@=r^OlI<FaVZXXY1%tok19NRc^oZ-0n
z(sVblj&LmMa0O31^#90mq53xgb_$gAv7}@ysKil%7Z7UN7buaN)#r-Q=zquZcOOND
z(gSASTW9Jj7yCAo$FZu~sY;`f4^i-Gm3|d<3hm6tzwv!9N+9~K!(rojPpWBL&zM}2
z=G6nIczbU=1n3ZSKIzKPKQO<rAivw$|6QwAI_|V~Wxii$G!T62c-32=I7-<E69gNx
zQNotMmZq6Bsn?RRMIn%X+dYQcLfWS=RgIebl!b4c*nwk3p3RA5L0&`PLMy8aJ7%Id
zMV7Z^LY}2`c#C_Sv2s;eyC=+S(hO}9sqiHnIh7ltr}i+C`z?1dHc1rph0gs8&HB+)
zSTy~~Ay4lyJx5BlS3pdnASIN{lw<wIy6SR$0TIbht}$)vTf$5+e$jPmS07E4;cUnY
z2{I#5DnD00H2=eof^Z2N;Oy58=;H`q!@kbws9Rv9Hg9w7uP6?}(jad2%*x_8fC)@i
zA^DV`rZqTIf2|_JWsC+<u2gl{^ceNqXklQVK-~UX6Y<oKlvq={=#9Vy+-pJ5VGA}Y
z)C-_Ckj~S@9cCENh4&BEKi&A<p_B{QcZ}nSxij?H57d-PP%=hDC+pRBt@J-7cyQO&
zh$oao_@=bE&V*Gel%(!W)=W2rPzt8oIe{+~^8$>^dWy8?2x-H#S9{2BB2wQ^;_Swf
zbpgAUL_LH1`!V(3FU@ElqAz~)Q_wK7zo^w-&0RVE2owfkBM~#Iv25>^%BBiCk-&*U
zTAIDHntH(G0uC)>77v0urQ6mIaS6gu`jeq7^hnTjSnWPTXC96VZ6t`#a})pTCVzj?
zYtlpRWkzAqjvazamE_Zc!QG>Uoh_~tA>eoaSwvFnPoeZ-TCIqNBunb1jJrlL_29kl
zf=#0OJu5F*#vK?9kZDetf!-SBIcCoeKQjlK!sk((vDaeS5<Ud>zpLvn%NV<3yu`x0
zc=)vD-Z%_jUJ5EisxTP#rNz~)nI+lS=b9XkSsLbKH^rK+D!<O#+Em}Hnqfhg**+F1
zUGVtQb<MaCR_hNe?p*cQd&I0~cIjR!q(L`RuQcd|zu9Jzt}h#zjM`qsB0V{1;QT{Q
zWiPhVxYA1J;31o<5mn+!VM6g?37HXOP(yH}0Oct2?QSXhqW8?jKaIrk6+loc{I@oM
zDWJ1NKyT1T9lDspJgV~}dS&@S`77tbi-ZH-ig%VhR3seXNC0b2rl>=3W*+u-s;*GT
zg-?)A+BrNUKogVPb*PGMGFngQ;trk+=uMfN1{C0{GRczh1pX|T)?suMfCCfH_N_RN
znf;Qs{RPaFtKTnFjLkG0yv?`7wMwip@s2+tOqIa(bHjupwCUi6u~`K-;LKWW>J&ta
z1ETRsn0?CSydas*5X@{jnlLR)#DotyU9PY1IdjVWLe+;`Df*7K5tTjo_lqQ0QJg`)
z>`>ezo9<XPcgL%Lys`Gy+;V^m>OjWQ?e~E94ib5b-*8B4$>04J1lo1@175LwH+KZf
zTQ5W!){I-Y<)8QPqR2?4g<oK_&Bv=B=78qy%YDx2c61A?R`j&1tuRRv;ogh>Y#15A
z^T~@i%Jf+Svp?*3TVfun$E3oE-y13QA0O%dp|ceY>Ur%QA;;V@=deTGkAa6`n<6be
z<1}i0UOy<@cKJuGI+`whoDvoI^IR;`d`0Qa`n=~V1XdqVJYoxQBMA!=btr$j@|A%S
zz4JWsD@3jCH;HH3@5&xb_=`i8Fqo<-R~jdumlUfmQd2*}OJY-DQ`=nM?`E6$Fos#7
z94SurQ853mjXjgrXfGZOjf(8Ad|HXsWIdq_-=2;A^!!4nOw7SNLa1rHU*ot|MSmy%
zPmQp%9ymvCm<(<%9jK~z+ZQP~1-b;9*_A^n0?F91BBSM1Hr;RUlfFThD{rF}@?EJM
z;2z)=lJG+OHv0sj69WBuiQG{~j$Efo4b=fl6{Vy<>P686pO2V|RfL9fABNJjp+t_c
zE3xXZU$%6_)EguA61rAwHcv>UDTpgmI#HtA9(Zpm_&9$=*QYz>b#_k{jKnGy7>IRD
zgC|HQ_UqpDat0TXvSu?m-cC|H3nK0be%Jb=8I#p(&2n^w-fY#?Rr2RqcwnUlCEfjE
zn~^>?l4OzJ&gPv!8>SyN15XA`=^FRRYM(u53cVZSQ1G-qqh$jw{3^0(j*M2y@2)29
z1Y%Zo^@tG#5#@9~?9`2Rw#{sTV9N%Qg%4DBBEk+7b^iI^L+jpN`10-0QW0_`+_(nb
zY+l+_mE#(<g<?ZS!a>d*izIc(*_RHZB7A;lTo%Rew=v1ON?{l^e+^D49Q;wY00{(W
z{$11ts8DB)?>=BQ+AW*!-zZhCH4XnI)z)2G&Rwy?+XrN>?;SCz2n(nb;l^R1A+U%c
zMHuu+-o8m?<fbd|Mdu3C&GI-FA4uJ`t?KlyK4i9>a+;zAZS&xH+_K<>VwZ~Yy3Br4
z!rLEjf?DcYbC$M`oO#oqx*}oRqa%_e>iz<y*s2?#WasRy&@H+kS(B^NkH;@795U$f
zrkmx5_DJSC9=WoJMjKkTjzBH~eK|X>gHxE5v{jK~s5Hg2nMRaD9vA(4uk{_aa=@<(
z{BC*yT=UVd5%;_F2DB20-v%fJ{%o9ooN*^vZgM)7OruOA-@&5HfNnwpR$T%uKX$IF
z&0u+!$d0R}?rmT_^cq4wQL}4DglDh!6}$24+1E@>8|bWdCt}LN^aw->TYq<0@fK%X
zcApd35p|hJCi{*jSLj0swue0}dSvt*mH8-PeG5rreCM4pn>FKjf{`g0kokd{0V5@(
zpq!3<{p^3d5dZ1vc+nsAc_WC642qU&@1_&U9=ksp=o|IaKSCHGFw}1*Hsbymy5k}Y
zJO9@KvWH-`-aQ1G=oyr>WgI}ML$+9kRH8ByRuT-<`F55erSkK3*5&D052bHx{b-BT
zJQpV{6aF)Qz7Fg3*9H_#xgvgf5mr;*;9f}ukXfSGnWYKBA-eTp+0oI=ywzWz!#1pp
zEF&i`sF-CR7GhbgPzo_)^r3n%k00`Lgri=z>ng4R6-O#%kh+1Jde~!TQ1Itnz@99+
z(qXywR2JeH+x2ySb&5eO<Wj%gKzBNFDF8w^Wz<5YEG6qHwha}F_l<M&X9r|CB4_V8
zEnI>+?F!a>;Q5%0b^C2!5jV%jHlO`qp>`YX-Z=L9n@WJOEWv@Ub^Oyz|5tFD`&Rg2
zC>IpT_Q43V^z|9gRz`gh)2GaHo(%wC3JM^0b-a%<&fGx5D-+VFnzZ=~?4VniFQxs0
z>%^x857ItoJEVSG5v(gT=tP-%K(8SMesW^!ppE~9zzPR-pFWqMo74;I+Y|vhGdkFj
zT6C#PhleFzb^h#mlC73S6wSa%8Utp>{}+F}+P>LXMp{+jN)aSLGBM=lyOYlRrK&-N
zf>Ls<G{71cQjZ$r=O+T*RbQ;>uer!WB^7xew^gZx3Muo0u=gKx;lApBU^8qTac{r<
z@N!}8Ka^G9((BC^fXzg&I=m4g!#oI$EU&DZI;nHi$_FrSm*HQ2is-``PUiFxmGv#|
z8$Mvk5L~MJ?fU~7L!b4*3Bcux$9L7!+jhl#^uxRYF;OIwHkk=T*1Qr{+cI8fg_%<a
zrvgY@rj&!I<Bn@6jw#}#UAY?4c$ylj`#@-IDX6V_<Kis$=-rXBi{s?|j0^nqYB}pd
z9+T6Sw1U-V`s(K#EwMR+lq`KI-}vl-r-hdJ9oVA$@wBu8vN*>3Rjl?$g*1gc2>Ou$
z`-OMkCx*`lR1p;B>wBN?QL;L&_sV0d)hXab2Ri1h&bJp?j<ak@jw##^1&K+)M-%TT
z+lS8dl-T5MTHW*204%<M_yfX-4a8IG9e(vE&L9@!X*T<m_Usn-usf7AL9$typ-}rr
zenD7GoH0{xxaV&Y9W)CK|3t^)P3B)YlgdKC*-rI)0O|eFcd1;KAVmh%I#2CpeB%&C
z{()EfFg?etsG>@dA=QLKhg#gaB7L0{s~GC)y5@8}<+|zxweWeBT?6ORVq=bLj3}!x
z%1`RiO5w(%N(R<&CTe^H(CdCS?7wwVj8Uc+3l(F=3N-e%&e%!*l737W&x_9j)|B~y
z0!L|Kn*dc7hs=N}Pa;g1Za&xSaD26JR4!`vXdyL!7HgVYG&B6RR@|W>2_Y(Vfa_;q
zY|>u?wR&Z_AnH1^FER!cn)DQ&y4y#DCgEBmNTv~&-h_+4MC{fT)Yw)d`?>Dc7J=g;
zt1nn+W?P_DN>NJ5L$dv-WNCr}EpOcpkhEHY%h_9taEhc>7q2vXA>3aV(;q?-sf0(w
z)_D+Pxf>_n|Cv$Wokc6Pt-_UdOc*1CWMD{6S~hwkVoXE51tUT4047uQ->q66H2~*S
zLv`k%afu*E*U2@l8}K;HGhw$;EXvuQTHc82HGm$F=PRx5A;v)@5b|FO(!Kr1OJ6}(
zKsBxux|3N{C^Xy5!ZxvFwYMl5ezw^}mqZ!G2ca#Q=lh!Q_+ZTz08Ijf{k8E@iX{{~
z!msMjV%CRQ?K#h#zbO!<Bt^AOhrM;0VigK#R^zIPcT3h@ldfzBer`Q&3Om<FzF5xh
zx9nNPne1xJv77fnO=}ikY(ozlE2od@&Y#GdCE^F=XL(7;-fIJ%q>PjWgLxo)_O@Qi
zL#2#aRgz?-m)y%7vq79g;JJMTjYhW66qr``L#hxu5_A~Zb!0_lrD=QaeY+~UW_K_r
z?+UtAGzHuK#AJ;!ul#S_gwYa*lWfJ2VqUZw$2Yt2RZT%jpX1dMxkKB^iHSA(*gN?m
zTxVL@z4`(p<Mt;0@r7wd|3J?wd~7t?D)O_NSm>DrtAFbq#ayvMInjwl+*&U76|m)Z
z_h7{#>**Ke!O2i~ZZ^X_O6x)xl@dDCXvyIjW@Qixv)k$6x0tJm{S@3JxFQ*xO{<Xb
zP96srY|C$n>fdg|gSIj65<jZsyho78?)=jcE`Vh5dCCX$?nLMV#J3NJF8IGZ;j7kv
zdNoqoMW?zut>D(cM^_;8oF*k`i)dH>5a??>zc^}OB8dudA)!s1cUu4SAI0&UUIaqv
zgN2`k*f_^ce&-}BZaIS1jq}S3xtc5{iX&Y;$aljj9fPs{7vQg>NWt^RKX{iaQr%ZO
z#?q!OO0~)qN5_Ja4BB1N{@$+Ux7SL}90ua!zyETPjGKp!Gj(=SpRsK4$}aSHH7)eA
z7mSZe&E?u32bqvp7v)<<ggZo#3uQcYE!w++<FFcbJ3TX`##cryucfPHp1QGQ1SE~m
z^wRlXjK989e;^ZbO8P^zx-<TI+>uXdUCtXfTuc`)S-qJ5h$9ezE6dKy55|Qg*QLR+
z$_zYqKj!Ts;81L8aEyz#A;|t_It6^=O91nNK0<~)6=TK|u_8J;eeC?=Whs5XW$>-t
ztM0ay%?-t7=GDw>AKQ-_`>9+!t#h@t*w@t72A^DWs82irceU+TzvTObunXU!SGC?C
z)tlqJp9(KmQpW^EBbG9r#&rksSPEvc!hK%+P=o+jBz~j}AkWQNCRp@K@1Cc%r0skK
z?#83&WTrz;-&(;byO$M2j9cHe(i(MogK{yyVj45`e@pDHHrx1?6lKi(Um^3U@%#Nt
z=5I5`!c8BNcOiqV<4+n`8ZB-aqonTCqaSD!tBYMwRf;ZmzS<sW+vihqda%q#KcOHf
z>%Tc^n&0Zgzf*LAOFR+x0oB;QJoY$VFN|K8*Y5iH=5^N0CR8j+;Q|wSkzTGlo&}aN
zls<dBSectMsF^?#p%lC85Rq%ItYpDPhgq5;x`-BIf>E~=_qDc>YeCVSu?s_9z5}0=
z1;#g*8jKg;F)C>L__O6Fh73NXDJ+6Q_rtv?ak&=3kU6+`!qh_i%g=gR0Z`93Auomu
z*nKyTzZ35$BRkOKvkxbDZA*i<MQVR5QL~@f-enx1EL*?Ft)4=r9hEp#(|c~=u#tFn
z&pq;}eX7b$9X*CPMM_PEDN5wDeZ?v72n)G|-9<yjbiUY5(be@$j*d1mtHdyPU^lWj
zQA`}}wc1pUFMIX;KD%E_Ch8RN|8ezIQE_cso539lcX!u7;qDeBxI4ieg1cLA_W;3z
zyE_yb+@0X=&~<M2fBSSl)yp0=#;#R!t|^~f?J8PuJ0u<!We7i<M!$mQSg&?(YMd?Y
zP%eeOv?RFVY@QgUa@gH_P}zD!TT-j$^el(79S{N+r65I10=^Zah<YP__RoFd`FOub
z&&oU89?pA$IanYl9z(~<qkiy~JSa7Z1CFcYSP{$8XEd|l(BFLjZ@cDilW{RuE>9`B
z@RRa0#@721{7s&BL{636ygQ>)*lZb&Mq!8u0+Nnucbx>+5jsv3JgYUHBd}WEFHg+C
zND+>nW+a?9j)YI78tYQ{HU2~T?;XjB4gcx<3SN#ao3gBPSA1{n^509+u6N@1@lC%p
zAM)ZaBPw)Q;KjAEk8O7OsEQOLc5!26b#qHs!{A^APfLB9WHHL-`#DefneCjF?agfT
zl+#uZld4WkYbky6Vz~=^rHtS$aA-ihcV}-c{Ii=FVgn|Y+#HttpY@=zb$NyyF7#n`
z5T5|6nf(>K;p3R_&WrG)caz7rklx$#k=jm0ydk_oI=K_{utc$>2!1~4>;<6DH%ucT
zn48BTjOZfE@=-9Et5B61J6_<DE><+4UKf51F5Ul~@guu2g)jE|E@j5z3h}M?X0O^N
zH9S<vSY2JN>5VkyqjYJDzh5#ZWq0-yr?S9gR<Bp}T5E74M#;eWVAn$*F0fdhn%0+e
zA5D@O1|F>>j%*-rXARL{sbuwO$~BWe0!{BRGIdMvAXRj2yYr#uYJP6&YDL$<FQK(w
z#sN(GmMHl31Z9)1kdgJ@G;2BWu=JkIm!`lfLq@w|CCjJH;dPA_eRw|o{fRk5ky2F)
zd6aDPQz$J}@V;Uj;mO{(DV;S!hgn<5A)~#XSKeJCe~iXqMrJmJNz`zzPx4bM@A;Zv
za0)-;k&N1ix9mgdCu6eNun{eRlB<+_{^|L@#35zmT2h4lBBej*;bH460BB!}Fv$`l
zz20&b!MJ`2l?mMLM}gsZ)}J%1Ne4?wCX#75%7~?oAc`vSrTH6o>$GUMQBBGbTxtBg
zTuB*WU*<YSHwr4c<C4#wum}pViZ>&a$_268&yC}I*C@3{imFLzu020_-JNPwOq!jb
z*oF#=Ho7kT$`ArESGNxQ%nxR+Y1FzoBYE&m7MKzj@QPG&&B!lWyenkAr`OE=&I{nm
zE|TfexP{nnHbf{Tc*?2Gusd%_78`eKz%-D<T(4X$!fNKKpJG!CRY|V0dw-~Xl4@HA
z<PsGG5)m0i5W<sTsZGt4W=`Hh&ayO}xRneUbGFkg?jJZjl&<fe9=;d^N|Sj1`15C9
z{{RH1a_JQ7_T2_^P4LsdM~wavXXEAah)9^n@q6tJWq?2?EG?f?QH^S5%+0Dmc|^8Y
zH?1tIPCBa29M~*eR(7vj=-f0Y;rK5ks{0W!`EzHg4z}oFyr{$i6HsUZ5u8D*&GvOI
zEj>d+1XUVHRiF80kxN7?8zt&sn>A-mmx#YbR`r}VVbFe?8qHkSZPBvX=yuH|y99%a
z(_S#((3of#X7G~bC~P#S2z3*<q>+o#5@cfYu{7iwV1y^wc{$?`pYZzO1iZn^I#s%T
zY89Ym1S~pa@5DTEyx=vK6rufkmlQetNI1iCa<M)eg%?W6ygj1(5=uqmkdGAkbK4oF
zzuZAe3PBKsPBYp0cFe)^pK$@ldnTMwC6}buL?|LY&-6c;7RLwt6ABIfMH6{m*Z53v
z74kePAnANIn&bB$APT+fI`-F31aU$2IBaPN?r>8Nk%Hp~ccevbA*w!Pd^S!IXl%_P
z`+(@(*yc?Uy7+LC?-7l6UlBz`*nR=tspvu(`$TH_9f1Er2y`}GAUHD!o^<{uQ4PS1
za&!+HlvXl@L;|5M;@VKF!Y2b-wCZ1nR@3bni4kZ<*|#PyK8RW39Q&M~uDzlCMC#b2
zxN7m3BMXalEi5d4(EXjekQ}HJO&CcDpyZB|;WvR=K{QEv6`sY(`N?u&J$q_r2H3mt
z3L+EH9e;Y*+^+K2+~MDRB41ryrBg1Lb>AuIadb9#P%Y^9h9-v=^Mc2xcoUf8j1DnQ
zf6rOgukR$+HT^-rYs1{x&`C(1k(qh89iD6S{aaww8NrznEjqV*Kljnzn|r*SvhVt3
zixAQrBVl`E02sad0Ft!0xoK%&sZ|baSudKy+mhGPZx2@c!W6^K=M^JP`!iy9lj){n
zvF38b7me++P%Hm5{&}ng)cGInk!{=ly|}RqRFFacJ?n43_X~XF=i^Ok^4D!R)sXxT
zmkHXRKQAHC4w)RfZy4m|9XN%!2N&F9Th<HcevKT{n=Q}&YYp;BX=|vI!|p^O!i4Sp
z9sbK0<rkqYL!-a9PIA#nR>&X(K%`~73^WQ9+$3SMwXJP_NZ)XNP50o|^~vkw#Ye8^
z^x^Bv1&y1HPF-W&q*76@%T(uxnzE31`nbxkWRW(iZ%5_p)%Z3^FEQlESql*G^dA0A
ziy3W}alkk_Y^29rWh@&5|C7(D8t0ce7w-X90NQUL(6++tI;TP?4Jt7img<uXHkFHg
ziHWOi5L+H?l!*csHSw=w7%nCd)jG6}O}5MJv)ztoKqJ`~FjDHD{KUrlS-yRpJkCUc
z&tic0A#h~#((QpSCvPG78ov_th*t>v=6sU+VB9>WexR@Pb&Hee3U00KlE!KN@V+sN
zi_a}vgD5M=l2oWSNm)p;AQ`Nsu!(o$S23AX3Q9!lH)E>@z^Axc(e{D*qe&cLZA>a7
zYiY!E`jza0MQh{%&t+^PbdBROa<Zq$SGn!1heVj6ve(A=^u2ujXoh~#RaNZ@3Q8PD
zZg@E)(!@c79lOacq;#PFc2QnKK2|TTf4Nk@KN^te{YZli#1fK|du5~I;U0a>z+B3P
zgjUkp88{Jz55EY|$Xb}>f7JSQ$N!jSVE#+L|IYh%MbSsu`E1M^o^nY?yL8ckgQt7M
zRM{tP^KD(&#K$BPDjbLID<WCalPfC5wo%$a)5DtAr*@J%mV3c=pllw20+*4+H=@l4
zE3T}LxAUyZO53X;!V4p}+3n0S;ov#gw62;YL;+H;q|tsY$ERa3bb6-l;GpqDkN-pH
zY=s;2w|E0YrC&L;q;E7tkXc1gVIzM!|JU`q9pFc3XurVgQIfh84IZ*dgVIm=7Qyn(
zf^gx}mh~x`N1)mmW%Z)f$wn!d&yLDBF$aA9G~<*WNw@?wHW=Gq($ZZq->_X=vVMSf
zfng#y-Bs&mKPcK~2W9ic-l`D~ZbPFflNZZT{9Xu%L+68CmtKLlRVZEWyAc$6*m3cB
z@_)vZlRgB@ipwRI#V{h-UqT$TvP~gP`Q~EMqH=O`gniD`G<YKM%f}<mF*?n_yWkBz
zG@?o2o%o*#ELgMX5DfFoi7)}H1$q1JFt<xYcpQ~~&xKP}74iwV!VGu3&h2Zc$3a69
z+scYKv$??B_>SbXofR{^bEF>WwOjxWlI30LR`)Z(=XLJ8Gwz%moT_K;#uaYX&Rg^M
zt%r~FB-(RJcYB(vD79g_b(_f29I;JoEQNx0E{`q`Oe$sh_ogo+E?YeUVnBP=_2Q=b
z-c^UZQjH4QM82%-@y^z2FdG~pRB}6n;;LCN2ExY3#5G2>BnS}X*F64rML#C8PJ=cy
zMf@1k^w^but;;vM@$U1&B9E$yaqsB_g=tZel_{IlMVM(Udwc6cCvlrMYk;!p`Io>5
zR9)xG@{I6RT?xZ?;x(jtTP)bypK?s+#$C>DLo7EfbESF{F85uSkD>!RJLpP8-4{H4
zK9-hArfT&unia=6#?BtV?2ojLpOV1f0V>0iK}I<N<$QtO5<<;Zw>J)!DEj2_HIHYa
zT-*d)FGPyhPd9Xbo$gQHUVDY#XkUx`Z!x-R@Q{%;tM$x^`J}7|*e+}=DhK~N90wl_
zIF-d>Tn+I#7e`&IbFkP>^@-^`Ui9D~5`DimW~s5?asNJA2o)$31tD&Yf5L>h<`<hC
zQJcH-*eeTL#Zt%92c`_=&)HpFqjinD<h_6iBIkw0V&i<9&kGMvYxPrLaA=F`@oRN(
zlKq^kI89E$s1vmZsrcn6<W&I@R|M?d4LR!Y&JNM4?y6p=dt&4a{f6&@Z?5mXQ??Ax
zM{SdOaSf$e+e^<<`pw}cjUOk?C6teFu{wjx$1U_~BeQ}3n`Ou2PBi#y22BvksownZ
za&icC2+d?wsx(C431VS~te!g~dV8iX&@Lrr6K!J1<978AK|0|JnZUS6)10Pqf4&NV
zN2(ms*K};SVd0P`5xIF<BNCbM$#=fszf8Uq#igESTU4bw`f+3EaYC2T6S(n2WH{6Z
zmP#%=T}~SI53N2=Z=3Kpr6k_DyWw$H&)$dVEKZe!mn9zie=h!HVEv->x%22!uq%<8
zLS9<7Co_$Y9YO)ig7xAWzEa$M56%${VZ;H?Ix6P2!fQSR3pj~{g%BRG5Hk&G`jg%u
zSP>$}{uN86;JatNF|B)2d@c#vKxPXjvYAN_!^s(gTr5Gh!qLq*uyvTk<gQ&<DoY}9
z#f3~9csGKc>$8K$tGruqbPtEs2kBxy7<kSdoCEui#}BHK7v5Wc*xb7<S6l1wjC?X3
zj&#c^DOn{FaEwqZRjO-iian$C0AM8a6~2<86L90$5i1qL#8OczKPI4TRrc^ncP<y|
zRI~B(Dw6qd+t?tx=E<iyTwPM4{TM%V#H*yt5vP^GD!f>i^~{}$Oe#%EB!3g)zoXGX
zYI}=d)MTOl&3MMLuj<x3SF&n%v%p_d-_#X$h|TN#h_&yGowTr_eRA}WIiVyZ3Q{gs
z346LP7SzpeBqz`+TryEqoqo1Hi_GHhxIW#j#I2@mxAY5aeTJSW0#pk-hg=9c#Z;Hh
z&>_A-=o$z5XtOV@4>b`IwSM$x;6{<>5=z5@8DD?*9Acids>Osz*^cf2WeovKG+q$o
z!U-5hf54f%{(=qf(Tea3{a1LXc-8WrT{PjoBbm^dCzZd9Hm)Uak_W$f`gV+?202u9
zioa;lw)ou9MAvhc{i~hvj|3OW?Qc10899OyPunl(qC3lsNJGo(6yhO0TEVEk79^eV
zo-VNmt7%RY7JDM0!+E~x!KIc?6p*Gf3p<RVsE|@p(!6t3H-Ej;cyE-u$8HNA@Q4YG
zy?m3F%T-cY40q;dOdku3Di-qpoEN%HI969<Jwzc@Ej;`M$@u4nxrL)Q79jN8V27iU
z7OR)BaSDpo=<n`DV(N8!<@DwU9}_n5G-v2*$;?)0n3gDVqiuabR(n;I1rOx9R39lR
z!8l%wN0@K?tCdyjC8{iQ7j9rR1b7I(1zS5i6_&5O-Up(|h(PB@s_^*ipPLt8{$ZzI
zd{r#Az|3lA7XC)L480R5zfra=C`SI=Kqwftyzb&78Tb8GS?eLbFz6gEosw!?mE#$G
zqAH@CwCr-M9?j+ImZ^9`&?Yja5F1*Lka%CDS?RrVJf!>KD~(W4D~iAhv3SM$ODy{o
zf8Mu(k%j!|o}J?G1`KlI(Rrcz7cUIy%<;75h*<I_SY?L7mhr?S=`syQI(n^a-hrG(
z^Wrlfx*HjMzx%(+bP71|--BinRY2K-F|?ikKDP3+@pVmpJCOfbO_{!D8WD~q%90s+
z_;>3*HF&rgqW~V&{i8I)d{4)i`go!pM*l}FB<XRjIdMEuRo4SY9oku%6vR>ssdW3A
z%ll@7gT1SMu5T5#c1&~Uw+Iq(tD4n?FU+oMMLPKAJ`G|=jLDx$?v;GMJj@mxbHVb&
zl%5~61xheoM$(e(4&oXb74r=q9nrORD3^oFn3-KH7a^KBZLgE?$EvTzTrKm&8<TF)
z-h){j<xNy^aIW5Vu<JCF&osCj-qC(FgW%!GxJMcrO&@Q>Zo184^^ASfaU=ie(@6R=
z2$8k>mAUJ_v3qOt<Ei@qZu5n|n37n+`=T>^d&PKo^3o$ulD(>X&f>tv@URZ4xmI}C
zhm4vhx$gyV@Dyyi=CcRze+pN2da}Yd{@x-u5xQVS->O@Fc|QTadWOv>?3u9)rbFkD
z4E3EcYO2&(7=OzZE4U>z-l${Fo|zyLRub?)Jn@lWIO^0bKDlRe3`lzPlOZ>xU{c4+
z$~`=m&h7mRk)>-BqbY@;Akwufl>BtMhlM8O7#BK5QM8ZH#ypwg;b8GERzhUU`FMHs
zbi?`D!Oo^n@cS5tl+6zC%E}`eFJ%A<HU|f?j0JRcdLcY%s+`mJK&-0e+N>NPr;`r{
zrBLal)xp-Yl5QCnqpH&ycs4lip`p??`TWcvSW%86*YR<$OUGpRyo2kx)eD!AOG0_b
z=-ef}#Tep>H5PdrdApyhC(ypmBF7LS<KrxRc_q-|_eRr&ivcaPyG!{gTgaC*_oB#D
z{N?(EtNSf4g0&-T(|QL#&<)lxawGPKtegSzY9TYoD*KPEU8Up8{pYq$?z`{NjGC;!
zJAMx+`G)&pIX#Vi?s|Aj15d~9Kl(!-HmOPBxfkqffBU3p*brA`X)X4*6*meTtN*aF
zi!_Z@GFPgU?d$gD87QrTS3CsS6?Q+tZcMQ+iUB@|TmmKsdF@}{=X@^8)oy*lv2kiv
zX~hb%2T75?)JYjxhe7lWt?_Zl`R>C$87MHDchz-`ee)L~9`$zU3=Q8OJQSYFcmchw
zq!Vrp9KWPj#+WviRPto_=8v749@d)9S;44M<rpncWAOBqRL=3pyCMoIZG6D<Zl7^H
zKxcbz8B^?_nF|T?^&JOGN9ZVKM}rm6+}{2bgTi3Nj}k-XxE{V0%^H__b)$olv@b;v
zyoulE<j$ljVHdUreTUjnXoh{nvLY^YP{zbcyTtt}`^cga1J2RlLkIG*3q4Ly4F&bK
zZ5rE&()mYD9J=HiTg1CJcaC~@CniBIPiGHHI_rDo!@0)ZF2Nbc+D+We(=<Q%_{NSN
zfvP>%>*n?wEz$LW5Qtn_ww0Y2ad@^N^kJy;jrF}N={)Mcgi(VDGT6t>#^1lEW-U4t
zR2{734~z4A;B~Y~K0L!jDR+ku>4H^T>ZM;3=3m?wb}EOgCS!FM!5rB&owm_!@rRrB
z3;%fGmj{1Pfw(l#<p6ozK%yuE*jJ^28#tgTgt?1pWqBn)n%xidc0kTx$J?E*>(vH~
zgv?IwD#PFi$D(9-R2v*FOtD`e^k%9a2~P0r2z_kY2d|n6LD?iOkbTBO8hl3WtV1-$
zGKJ_C&ZnhnrY&vxSRhVC)dXg6_;Z>YoHf(OD<sw=dtFrBOib8bt<B{XX>UF^vRyPV
zU%5tf2#Kt+s0EWW7oiMz=5p@BX@7T!>RrFlDp2vh$o^r&LA-r{julOmlUX=db;jX;
z)<|04);0th)A{dM7Ob(ud-FEwCm9IE8g$movK>Fg_Ho+}c+|3)af)%4EhykXH(e*F
z|EWlaB50{1wnZQMCsnjaamx$3H2`>S+z26z4#^a!q19GDxm!f7uxROW1^+hr^&3xG
z+2UUXCPnA#NBV^e``<phsA!uHp9bqxxEFZ5UL#BCzDhrBqxlc4I=sb5QO=|yqzu3I
z5v9mQ0;3YdW-riM+|OVKO(LhdANlwC7|!mwQ{>{}rW;NavjatGEv8S%qKzoU<GAh5
z1m~x?{?^q@=kY~^hs)S-qp;W|rD~B|H)d8=t{HoK2m95;FQjJi*t=MIW$09imd^F!
zCGE|sVOwX>vC`y+Qw+P8*`;B9Mg6d^S-Gu6d9GI@{;)bQU2yg%+AW<ZiS(?{-3_yz
zOAG6$$0MJXBX6X(ZHN?zr^DSE7DQYDykL}4oy{OhU)b)(rlub~6Ded|tmzf1RH;?R
zd=Yd^&st0q>Umw5C0<zIp~n$VzSPpze^a0E`0Exj%6VkVdE{1xBPtcfyfSY+<9`cJ
z0~A01>%~PtD`RF2D4&m&<Qv}JrM*%Nk3}=?s{a?0;mLB!$3Kw|2{rIbkfV|I`xT-3
z0nMdSE_7PM6`$!<axtK>%>9|JycV>L6h3x1y&$$#P+Z*kwFt-=NDA%wl;B=3xjcnl
zmQYdxTM9X-#yhBJO`q$P&7Zbu?)JiB(+PW~ku!6p4aqwE)C}U}6rR%b`*`p?J)Sbo
znVJ!|d^(P-*Xz0_8R8q;XQ<Lh`qOC6&BK!*_Jsu(KA6a<xYhTCTTM+(K_QiGq4xK0
zr4814SDQ%pdhyOC*#dYutZ9uM`tZ$Vqs!;b1pYfFqVJ#1VBl@k4nc|V2n6;UZM{mq
zS!XLav`X1K^W`s|=dHM{KKmR)10BzQE3)}L;`ZAsDmOaiiH5DJthoqBUt8FUSI^iZ
zVCM-@`q<X`sg6IbwBv^jrjk}Sbqc=#4|0<P)-L80Ggiw68<j|zv1fTrw%MMD493=E
zWUVWtEG>e7@N7>BtrpSd!qSSNKGx;Egn<-kYRM($#%M0<Hf%?Y;Aq@A9BZE#RwCM2
z0q69Un9v~FJYTGJZ-yKdEz4sNHwa&+=_WE`RW}R#RMa#!i~!@z*v!L8NPNt6o)3}V
zJ%hpguxy6kCwKs0&zUGEFTd3-4g8kUaoPD?Uw2%l1gBCMHUA5=$z}g&WCL$Ed3w-4
z@%~6%v#;vC%j*KYajKOi(l?NhHAsbqEuX;#o?%2n23ue*UTmj~W0eji$Q8O^%$>|_
zF8yTtEUQJQYPfD@RvD5US4yfsfc%_n8~(8#G{tkiVR3LtuhYyl=^krPI(~n^=T{zd
zja0dzy}kL?{edm~4)OrrO!rTa@&>u0Bh}xi57T%)CQ7TwIo%v6YG`WGr$nvkxAxD^
zBjZkWhhpiA3HNDETQ~B4fV>5+mTuPcD`wX8MDMwQnR$ap@c6+D*mX((cR~pVf22}M
z;4^TE1V<PlB{SW=VlcLZu+pxYgPo)IF!y`n)hNu+@UYietfqFCTeLwb2s2(#qIBK(
z?p(jY-Sq{X@uUK|5Ml)nHaNn)ef-Xb2ahyI1cpQ-C<V1ttxoFDz4Lvl@W4cPLek39
zvR$;_9YG?nc3I-q`^Y&1G>@kt+mS7C_q@m5bgZPP%8I3}x4w6;71`zLb2B4(ny+lc
zAH(tp7EXEsEeye>JFIn%&y>q534AC0dBk}jb%ci8>YUNM`G(VXa{btk_Jv8k#p{6p
zyhQx<hqlFO|F1NM;c~xELf+ud3B~dV*Py_^7)0ux@BJ-#9SAHU7_|FvW0vRBPOoK?
zIg?+dDli+&Lv1wuM}(jLYHDi<r2k)h{!>jTSdfFy6x!3sY^NNFQmflFYfwH7A@3K3
zEkyV*Fiy5$bdW%tG$RbB{e*C-OxfHXQJB;u6C&7x5MobS3cfF4#0qhrBe!Ncm`L~J
zgh)g1eb~k2b@GoQ-@a4vNS(?iS)A^Xl}IxtyY`5SmC28kaokp~C*M~#Wv}n4$Dv-3
zq}6&pxPZ#4DyPe7^k08eXlRG<3ClTnKL3a!)M;8+@I1_UNYCOM%dfUx)Ld@y%6*Z}
zcd8Kgu?oyS+?7qxfkK_LQ}ceRp?{HFTSE^a3~&M#S#!HDZD<YSlCb)o!yM2IN809{
z#bVnVw2xe$t-fw(`lsy<$KMo@b|MCF!)_vJ=?P3U+GsAXY!v@F&o2&dOa;dECbe%b
zb=L3PGpPFGkUuMrbp-7OP0yRe7REQ?DBB>O5RQ~7yVyqoNeeR}jO$~ezEe_i^(9PO
z@$GZScNKDsIJx`gocxk@w2w>D<jHuUp5W&`C^a;E|8%465El*-ioZr)kCPm=lVYUI
z=5@hXB;8)w>7z9f@~sfRnub)Il9vCm@SW&Ae-g@TczyRlt=k=jR_qI8fzQz4Pb*vI
zU8ML#)2ZtsQU{evrQaQTo96HBhHv_)|H2%<<8y3UD&|PY0%msrTUaWLJNJwiY+hTR
zoc$oEnd-SsY#SmzlLsaNS(Vcni@XKF|8=E$<pNcyFy2?Kxy|x|Wa)C9-VBM6Mb_jm
zUF4h5Ru{G^f9D_PUMxM$t-~D*!(UI)?d_`69PBU&<xQteZEQj_+#_eiV9u<+-ElWN
zFkLBi^&K6-o0=9SkDC)T5`@a-s<JCLFG`LQO<F{|S12~GT7ElV3*_cH*W0=u`^+##
zxzf<`Ye3>n!fFXWWC4$O&W8KQ)(-G-YoqXS{((>hne7H&f$zFH-^}fGFnY-AzQx9J
z4psqX!QJqUT^S-?&`fiDmK7jRv`%lHvz|M51wgy<30CWMO*9hncpdeId%b5^iuUAh
zb}vrbUBn`<5;is7*S8eC>`!?2T?-OAnKndlk|!p?vrds|4xQY=!=Vpa<k^?Rptuaw
z(KM^|)5WKJgE9UAaM7cy+a=`oMz81BT|1xme0#@_V?JZ+RzBFCVxLMwOlELtvCp-y
zd5X-4SErs+$x8)*Wg~QoTqk_$dylJ7q=uEQJIAnowFf6a(u%9A7j`v{zG6<@m<q}s
z7eEQ{aLVsqXOSA6|9|J1rgmtIP|`z=>C1Z(QB_pm(H{%$sZF~dyoMCxN1ss(A|iu6
zO;NXTK$7A!^!1P<Obf;$LsBSj!;!DAp!RxVgIAr%#&{;w>cdh;3p_JYfcqB#|Jt$9
zNDlw|A(TRsxX76#+l9q)@PPVaaP9bvZQ&OuF>G=WB<7Q9p=FZV{35Jvbv0XrX$&@A
z`4_y9>#L3gfas`J&;j52+2>TA#UuK7-u)yG6uBo3DKs`!jEC=!xfF2}*`JYjNCqdf
z%Boh=%ZmaA`4olncEK{iG4UF$?r(@We&OGXAgoY{`FaY(;Q)P(PqUJ+Z}1^I2WR}?
z``Bh;l9&qFIVLU+{n0~PUw`{PXEk{odu5%a@yrf37(y<}RlGQdfc{11<ZbIMCF5h_
z_R+vMygC9Q=+l`9esuqESd{2iEwra#ED9u~GCCZ2=nCqLCw2jqX!jHuaw0}`Z0x0&
zZ;~SA()okep;1a`B!|n3*|~F6N+X1`8_XLchr|9j^!Ehhmh-y^!BA33Ba|ek_{Bdo
z&m&1A(KP`sGFNnp2SATkx;fLYQg^VSCgHtG!3!h@s(97^-x21i(J~jPH!!D5v_sP}
z&GtNdFXH9p?Om#)6E;T(h}gJNn1rh{AbfUzY0-d7yP;Fkh!8+K#!rr&=_Cme>ihP*
zo?fdZsw-{olWAY2L`J=Uk0W<EMUpM4ueZbR;ZBD;oKYS3d(Qgz+_&HO(v`!$Vl$=F
zwnl=o*34%1&S8V5!)0f;mnG78&yQoYQAC|XG}Wz2HFMkUrNgsEJt9v6T9Yc@+%hs4
z)JC4PFn7JEcl=1Rrl{*z8itxvf8zpbYRG=ZZD)%BC0n`rtQyC>FS_DxA;EY-UES>N
zOQA^(OyNsj-@ih?{;+4y&>fG2k{m_YS4bVnE0G9-dPAYKPNwTDEG@%zibic}I)@r<
zc*0gfkB^eQ4TX_j%qKdZWH&MxD0Xl}$JW3e3?&H*J033)8gPj>M+zG70%#}rF|k*w
z2cU1x>*=C(L6*dtL!}utf3%Y@wvn6V*xt98ELmBu#)Q(R6hO6(r+Bf3iBDpn!_3GT
z1BC8P=>wNX>unn|<`aVeVeDNN_z*VWlTZ$sa=DH<wfX<!x4_fmlXtwhR{~yVQw;@5
zl2JeEM|V1h&Wt(vY@*-6(J|Lz^qG*btCw=88k9!=fGi%E+COSs1(Rk>P(Wu4x>-Oe
zE$Q?r^l<V-_68fMZ&@hFL9C#-FMU`2t2VZHkr#(R4#8k#AFvKZqeYTMbNn>HrWQe?
zqR!#!K%xbt=vTQyk-GV_-?M+`IKu*+@TkieM&AO5t(yDaIHO)Hk=xpXd45L#f<&H{
zmH;-B`E!FjKYB1*R`^)#A$hYKxjepgywBx1-u!*>^n(yKlzf8GJDs_>)zZ~2!NbbP
znY(Q#t_4$2R0NFdZdhi>XrfFCwdPLHj$Bvf2^E_N+%hL|bh7kWx~=jgBi?s2SoK@`
zocDoHo9MT2csZ~3UnX{w#k>(MnVytzlQ`mCc(2d6bz~*R^ol3G675!+3g#7Ig<-`S
z*Ku!ojtQg_Z&MRF_=+7=t<a<i!`vu&gx@7fW&<R~q=&)^7qIyBHDU%lAyeQ4@9|tU
z(M^$<6ee%Z(LN~twejAh`v=f3K8Yp$BbZdE5|B?7eki5FAlAIPWFOMsZ_MVT(4{8^
zEyRKR1+halLJe}4Nri~=jcg{EJ(5L<NO~`%CvifY@#~nCO+*nyj)8=eYlvQ%mwQ1^
zc8!;L>v*hv{E!A*6&09ru?vR-2$WtU+#(E)MULJ0M6CU+p{5dNK-IDuYsDW<3;iUb
z(!j198z#V|Vf3?%{>fKCAYXt_6ffklE51e`q)~?csA)CCAOL_fqb0@TV2p@l1pV~I
z`SjAb+Q2hw1q3Uq#L3Syw6_EeqCRwax~66^Y?JH6#nkT2`GfL)v-p8?#x%6laQC_q
zu#FbVeE#-O-w<zlp1gOzMpqYgxngI=Beq4(RtEg3$+8VeN(hSw(}qQnEW#@f7zPT9
zPy*;(gyW8)&A!xQ2A!mE1b$X8ga|blIGaI-|7@s7=h(OS`Uo}@d9)gN)e3ZxRQzo2
zXXhGpMn-k-YqGyxX)cDC`>nb40Hz^?x>R!{$CE_cAJ&7R*udWD-x!2Vwtr3pMmgXM
zd__{77Epj{_B)oe(JzXbMn0usAd)Igpv3wKIuyqQxig91M*q~47DTS95mT@ySlW>y
zQD(E0-^x@W#1%1=(j8X#GM$%g`0<sit`T?))+YAkFcMfa3m4ez>gsKjim7N`Ub8ky
ziN6u(U>jh?p^s7CsUQ-=<F?In7p4L5C%5Y>6kmnzdMGPgmS~f%o};6*X;GD!gkFu&
zJ+Z}KV`)JtB|JdEs)W~S4a7sQoXoXf<E-(7u~CnO3c#V|8CC7q)vZ$?+qkYNI=4dB
zodb}@d-%Yzmi(Wh>!#1q?xn9eySBpbohH`dD>bMLs^X^E`+^iRygNyRjQWXK4$%(<
zce3NL@>9g{<*7-&OoeJuEj-X8uh3x{=k#q|<`@G!UvGPe943@^b)%ExbehNOIkrOe
zk|?}VYaARO)Yu+Eb{T1Dfb1t~)(>>ZZB|q#gNVXqF|!Ca>4Fn!Zvp)(;YfoNDpUZ2
zyo{%thAE-wth?5O!;Itq0vf@*i7fLj)>4))><go}Msp812&czah!TPnRVw9Lxo$WJ
zGK`T5AxBmr8qfyl#IovutrsCB_?WoBwU3(459C><_6z>03*N((;ZmM`U(H5`Pm^Kf
z5YrLRG6_1)GP0DBKioA4Fz2m;R|>dIOiXzeI<zKF!4kS!-&$uA^yKPvCL_cX<>4?w
zJl3roD%$llG&G)lPVa_KqiJVsMI?%l)={_uUQSxRCK9J8uXp6VsK4g9O!0mkS5M0A
zREnoj=SDe-88r@l6W@}Aj^z%d71gDv5#{_g9U%bXYjM5y6S}!ZZ5QtcTlo}E6pS5N
zA_OyuGEe?)^(9t*WGahb!*@<A{khHrnTKYmw-@88z<p0X@2YE;IY|>M6|B@Aeff4g
zWr5rUfS<~`??~AyCi_9MO4uAZh3sSyvJf`qyYFOAd`lloE$f21;Gi^vT5;8EwOY_j
zU1OO$;l^ezmH$}^zA{<HvM!+UM^AE##BM|3ZKh|*05n9(g{h$_OVuDlWG!k9(+oqV
zeM{MARYEz|LfZy<v`ND&0$T|bk|=>N)a#^&n~`yTPjRrdl!(ef3xjC_^=&=$6W@XM
zmdou9#YGCLw^i4lZ*vaZPpU<|XYRO%FLQa5cnst+0`bJ;0E`f<;@REDUONL|9%T><
zvwodn*5W~<mz@FlWAk85MWr-lfHU=2PZmOfaM>FK7(-;mdhA41MQB)f6#bMk6r&d6
z3iyu41QEud=p?d5zvCYLZWc}kFY^OhnJYm|AzZs11q)I6JFu#J$kc~=8k^2e6zEfK
z#nYxRmOVn_bGw6VlLhDaj^_!1g6_DxLI5{j>a}V3vFh{cl2_d0{B!wW3(Y_L>ipBp
zvD0Q9e$Wi+O5e`4fGCw!BcH+$rT}XXUN9?UvbvQT1!N%L2QaE^Hi%2!TAEoR!z9U9
z!Xu7q^q*<W%n*cg8^%M$U;FPUo1<QWH`LnO;X%I^dl>>tKmRFcB`9qXcL^;S*ea!m
zq!X0fuxA%uHWr4rqCwE%sukO^g!$SWC|Xd$lOxPdqDCL68OX~uRkBkqmDWB^WZO56
zYx&qZ31y!nEVYQ_X#l<EW*57C%8p9n84~{7T%KLR&M*A}evNM)J)))37fZ;r()_2o
zOM1D((a`aYeA&EQ7X<xU(CVt&85p?m4GC*;6nxgAwDQN&%)dCOSsq3Sz}O=EMVp(I
z4(^)C&EE3KZl1|~o)1cY5`W{+Spon7ITL)sLZQT9Um{Ffh2;^6u&9fBq{@%Ebi3_F
z$_|T&gy~Fcbk^I1GaLh~-IzfF=^`F+n<nwE`$rtZI#9WNsn~r8$+#3d@c0s*z#-j}
z1Z1Z^u(M4Xfv(jw%&e7y0movL_9at=V&3u})I*ZC==KZY!DV?|&&7UW32mT4h3w5E
z$3w>p@`W&`{5bXmsqXfk{;>5_NP`ahiWQBv<g^tBNHs(iXh|2194-kjTfaPngFQ$k
ziXbg5E7e?G<Tkhg5+c;)uc6=E{M9JGnVarl#3DH;FO9RLCD?!Yh;Y434Wi*)wnbWx
zo1O&F8gtB1=pGrbGP`0?#z+!K&t@63t@hS9cW_9PWRCyt;|;tL(<3qthL0`swPFeB
zug9sGFB9864w|(xGq^TkwAj+_TDoPgATA6)FZ8|VGW2r|j=YJyywX(!b}5{!Tc!(h
zB2{nInKHgkc@SIL(CkH5;l)^RS1j3Z@g|jMl>k5hEPTjl%pZ*wSsvKf(MiyhM1*qA
z*1LSU)kmkN_1gbLkYxEJ_|RUx`(NXt+FVHARU*WEj1wV?0W#!AAdut5I@<K*VS+}T
zuTAvK#wRs(cx80v36VX@VoLiEXs3|Hh%d$gUorXu7EUK_XR1`SOaXExlMTnra@#yv
z>Y#0&Rjy3p=l1rPmv$Khr*_ez)gfNC6mf)5m(1eyt=6cPk)x-ep&@LcJDU&Yw?^Fb
zJZsL@M$y(z`J!|-LL?j<zRJq#t(bl_@-9-7Smm!0WY{V6=we1y1;55LX?eMKV0AR{
zY|wM55V1x#Xh11Mx25gU%Day%a_{_xfBfq^u1DTlx_dgZbGDH@gnI_^`gI1&b#*7s
zP%fho1tk5f&WkB+LXw?A>?P9KL`X7~Ei8ZbzsQlt_R>dk;W$|te9n3VseE27o<!wr
zZ*O>M*Fw;47oD{W4z}1oLV{6p@nAPTV#sCW5IR8`8b$HD8c8s%GxGa6xmJZju-e{Y
z(r3Zs=~d95rlFb68ApuGaVsvcn}m%n$AF_4i31i?^7~4GI!WskjIBS@gmWnS9BAU?
zfH{>ivGBihTz9`(mcK*hrix1Q?Fd@kZS?~g7>6uEo&W^p=npRP?wd(C)lOA0)=b+E
zAUEBA8A6vJSHY_5ZXI11*OY1UU^D0SRbdy*K2)<Be&6^+k0nP>6kf$%Ea8bLj5t~m
zRL0-`daa_6%D%k%lYygy&d9<YVhRb+r-DN-nen68bZi@*6~AtWW2ezlCf7H)vkJvt
z=z&T<tDa-d!q_-ain8x0hz;2ZR9ElP<1(dZYSGm1I6wU01ITY2YCTq3Hfc*lmi@s_
z5Qp~={zyR32X`n_VO`Vg0sD}1*lHkmUk&#GbrV}(QOrl19ut9j@AWvwXv=M)<Zxyj
zQxqe1O=n{7lC+;zGrhJU>YLMg>0{Bjo7)MYK^LP6Pjo%K<`u<&82KxxgTH^{T17o|
zW@UAVmBSV5^7Ir^ob}T-@M|#<vl@xg@MAx^_h^LlXga=uvHQtN#X;iaa)$NIQz(o(
zA-Wju8nU?Dq2~t(Lil>(bM=&ai^sg>%k<hD!S>H}rEH@%WTDX(E+Vjto9t&~)X9HB
z=HmDKKPU9wY3wI0nFEIJzbd~{ZC^8hv=f3v5Dqd4^v{g;0=Dn@a_1Td*!QT5Vwv5m
ztLi*latehd9>Nv4xkvAx)n&s*5vO4b1a>r%xGLzbA<3y!87k}HdDlCDRs#VAzYv33
z2K%m61Wrag?q{s!u!dZ}bJ9^KE)>K_BJvP7%46IQQ4W~N%nRpJbE%NEcuv8j#*>JD
zUY(d^p0;gHz_@SrXz6Skv~>u}fEl@FH(nYbh_2>3s5K7<U|6EyNas?SdHdcPDaga(
zG0%*6+@f4gstVj=Ol|n%e0)CW!ie}eRN?-`xu@>TjiPkfeqf}dxU+6>r}6WFLpA1o
zxaS*rRjlGpr;y)|0@TTzYoOI<-psGJ*I=z?4Tj=+An*FtZvn1{iXaaAx^^G12g(dc
z$RLvg(RATEXL|o=OG(`t0QHzFhiY4Tm=cFpeCHiPu>b+GxAeTVhm=?_%)Y6Wy}74&
zy^wkvx7ovUY6T~&6aN|Mvf5|jtQH)fHM3>nF0@n&1juo%R0XYz(x^kX<~<#$T;ISc
zt)1yVRf4}mIyic0v7v|KoDYVDqpU(@j;)#@Q#JS{>e;3CkRR@~baX_XA&;S&#Nwrm
z7?P1M0qUdaCcqvu{H9-Jzc8j`Lu!@?zMjT&hNx@p-Ck8-%CPI_{FOwZf;?f+JDSrM
z5<3l5e}mFX-G+fssSwBGc0^5vDnj~YBE~&t#l7n4t<S>6$E6u}RcY(Q4^@H0pl=c)
zaq`2Kt8g`<)zvzs*E1v;V(ft~8qx+?1DKue$5yvu5-Uj`+C;6Lx?76r%H5|1A?6j%
z)3=DW*mYwHmn!&&^6e_Vub_1kv88dPAU*a<0#w&2NTREge~2_u(CT4jzvtRbA&ifS
z3RlJ?u7U0MwJ>Mk8By?8I=&^aJYnP*(21!<?i3gwM=+<zt{)2T8#XRI(gEo94;bNa
z1tc``>7f#;qYj}8=sW&SceF@$YFwr{akKVR>>Dh~PGdeF75NKkm_|o_eEj3wYQT^3
zKrKzfYteHc6&df5L{x<=RuqxP4hscn=Vb9rN8UiKlFF$-n>{Ht@qcQ!|H*6GDsWo#
zTD}*QOhVW86q-)g5@{e4sG`qLexxN>Dy6}q`5m8VeboVTR(BCu`ma*BA!H{PDyXRW
zbhYN86%5&kcz**_*gJbvm6WG+INx4M2BIl0eE^8MyxA#W4POaSg|_B+MNR?XVs=iX
zv3?)`JOmQ6NW(oM<{wOn6}+Ai>NH3@^Ng!0o+gcB2a&r3_LTmPqb${&nf+mguR>JB
zsQ6X!rG~x)Y;B0OMU<-^11CPzPhQOtqMi4;up5#qPVd&zcGS+QA!k-Id`Kc7ca!LW
z-iqi|IDT&eRsJ~^3<8s5KXOfgA2J9?5Tc2Y)fN10k@k05;QRue2wVn;sbh=50Eq#(
z=Z?H@^Liwiy7V(O>}e_DTC%M`X_5i1`};)euzPWX%na(UB=x8<*UM32e%@Yz!b0v*
zv9W@EqY&%MBu-%KR{Ct>I2DkwOJFZ;bDVOHK?yzc#VAK^l&N~j724!vL2DA^{?CeP
zW#q692cU#{Sa4;0%N=kRT(Etx$p7zhn()A*p#<z<N=beSa()#T9yr)13~}KU=q|B3
z&IT?*pW#q|tJj!~iHQbyOcOr1N%iEE!7Pd57hN;B7c#0}>jl>^xmXQ50l%067Hr%*
z4*p0Y^yQr+(N4I2*fjJSM;=(9Uxl~|{M-&7aEe}jl7uXY2nJX}bD%B(@#;EVdQQg#
z*4?8sFb_T7TWRqf&M!GdWtx$aMkC1V=Q1<4@6TEK)MUunq>A=j@$0b+iNmn9Y_3uY
zE0Le5Bt7JG7Ha+Na&mqa(M2CWAzkHc4jiCTChtiQIk}z5Gx;6)#<13~3-k0(+0z@$
z@NzGnHt9;JNH9*e2yu%0S+QgfW<=%xK9ZF%^^bm>UTffg?8^TUq(U^!$JMNAQj)DS
z3GC1SJ^Z{Oj>%#nUxW>$&dVAc#3a~dFDLy?*!(#6BpD``2uq%Uz69K=4o2SxLD-w;
zhA}T=GdB%WgEe7J;mB9@wbEl+aGG0%IQys0&v(3{x&U-V?5Y%wTI6Vpd<gReUSIK3
zY3Pkh%}h{S78f%9bITZH{ZG8i$m9TRRar~G%a{yI((kO&q&|41d}|v@WgRe?v@9yX
zNB}H8Og@!NEifjT3`V6$*5QiI0Fa1s&$bI;{-f6EjYf$6YeAw0MG9OloD35N^UtY{
z5_O<583idiT>|7T5kmwxXQlt5y2Q#kCki@#AT(6v9Y1*f%)$hXhfOBe3q|UgZuq+I
zKeF~BG@;fGcZxEMOiMkyc|(5BOtpX!!u*;ac-?=TO4A30y~PBiN^yZ#A&;TJ$HF_i
zG?#I>Vp-q`$0~t^#<)8Ch&(1Y6zpo#oYoYEh!D@MA=%7m!Q;-Nyf(?XTjm}Xx;3Cd
z6N;(fd!2U|>(?a{c8ctgtz=6E#|IGnLv2`bXbD82Qtb4Sd=ns(!8q^sOX8-lp4MWl
z13w`nz@F1=RU+mIiTecZoc4)mB4je{)W52JkHT$pF$+H!OLNWH_dQtz3K8b(iwLC(
zxSOU4cV^duD1?8z^ek_ur-+UYvv#JzxlBr$deYdkqY~f=5737Kx$>%C)0n!3UYGKX
zH24|tIkY>XC;!v;{HOi-KuzDglF1PXf;{vh67-MrB3ep5(923m!6#V9OG8Vf=mWw@
zbwBs>Q2}E=!zLFB3J+U(ggdi^gqhmhd@_7Kk~7Vbd=i>{a6;@UXb49SQw?yz2!)#J
zr;o0~25G&riKS6n$rDMm)E3gogS~BtPPg?yk!sQ-AVV_25roj)*L5@}pMAQIo$gou
znB3N}P-&rwlpEm)MwImt48G`6+A<$V?Vw56g!>sWijaPM#Y|lqOnAXiM1dH`y*FlV
z2R&(^JQ?}DKu1sMpEEXo9_jiSvMJRu3R*?AA#Evtpx8;xeQ4JE+U96PlToGInvSfG
zfyq?ft|U>UBD#wmiD3r$*-(T}1S`dSK`Z7QNzl$TII+ba9v+>Z1Bfz7*9G#aeNh==
zU2%3_v^J%QG6~NLLdXGR*#ktGNaiX{VtDe{Ql8-l+JG4}#=hSY-$~F=?$bSZ`H|$;
zzDTFvVI}PZWHlQY19P`WR0gG+o0ihO1N3Ts^C!IOI$@NnS-i67?&GZjJdYDnEz2vV
zX^8r)ev7+o)7x%sBS|-R56&5yRy&HJb5j=I0!;XF$YLmMwG&dXlent>X@>skU~O)A
za`;*PCMRzplUWbh>3kJ^nfL0Tl!iAG-7p;f${P#JjQgg|pX2SWHOfHG9x#W!@GWy4
zpXP=C$dzk7tGdj}gCKw2)w)5|h&n}SrnPWPK$TF{DzLC89vKC#uZE@a3&}!inO+eA
zN$z1{3a@!$U3a&9`&{vi9jD8~y1>U0(k<FOKx{hQaDM;_&E@HUv62+o^D2_C<Mouq
zxL$q@_3<iRP~q5IPutM4+AFRVJHCjKWca}_V24BdCmvOl!6N0ikYueNcZXL$vOP9G
zJ_%oLNPBL+Y>Y3v9BSzq{9Hc<Kns~Cs_pSgCUWATdhU$UKF*g4P2zUng27-NiYlqV
zlP*&S|Cvv_4%?aRBN3BYSI%GAc-?Yw<A}%;Vb<UI3dfe?ijFPf!Vl)CcOWUL5dm#t
zI*PM71&!UoX>pe+#b4Ez7Sr3sT?|VfTPa2aGP#Q@U?>sI%+fvCp8oFX?vvX^j(+~R
zT_70WGLgZ|!E$~vd(4H;B~Z<GG$}zee&`A6ZRsAr>7-5-7uQs3nP$beb4!S2SzrYI
zR(C9Piqc(t!Ab%w{&wvI?DE@g8+=aFkAIVrff|2(zKKxj)2j(bff6eiVF+x3k<?JN
zj=lED>=Fa)4C`xHLQCQ9NrF6q1(m=4Z|rvOu6e1HV&O$|ZGA;$9mkJv&WIIcNEFY&
zfDfd_?tA$8GGQIf4Ti{q@G9$MOsBm9FMVI7LX<Fd>V^$CelDUWhy}8iDTK-nA=5_d
zZE9~>kf=4LEt3>Ch+X~`s{hurE^nIJhF$kt;tbXzd%IF{?AY<#s#aR1*p)AuYX_B<
zbwk|a9*>hzKZY4yH0x#a!_VM7*uTG{l5Cx}&a`g>nL8jKyFjkDkx<vioJYD)i+6U%
zL9bzHb|#N*D726fN|n_}^JQt-{Cx$jEAM>S+qz+uYXwDToy4hL5{n0H??q-ieXbO4
z6>rq?lKb))|FN=ZW@(u=<#%8Idg6aq^LTnFGz^rj+YI#R2H=?H21L^y&Gl$NjWN-Z
zIJf}3A1O-d)M&&K<U{W|j}0VNcJQNXVU3YzwJO;?y#jB1e@TU;(VPyEl>T6Ua}N;Y
zdQc2ai_uPJa{g|an;}6)kos*l{|6Kw^~m()*lft1y`-V`wL8>j5MaTiL`=3<8qV7_
z8f#H|Pi}s%B#qIvW^yP{Wb~8f!P*#=B-53*(D6A*|Lj&Xav-ghJAXPuk-3lF4mqY|
zy|tFLEv13!Mk)j|>sH78@)^raM|1yihTC=JCsP?raeq&dOv1`)wzlFCns7}m#lw{E
zKP*Zocdi+l81RUQ5^yB{Qv&^`W$4}pj}_tOEVra=vN`jKW^Yu+`Y8~KXe^E#Q)q}|
z0M=q_XW%tN39-3OP(~AFog6t-9))5=P=e}SgII5qTf<wwmt$Lg_Q2YjP`_I6u33dN
z@AW0e0>iyQD`<9<l1mr;d-8S0FzNoOS-kf_&Ft||gyYJ#D19hK5LHHgai1DV7w%Rl
zyZrEBf?p^*byslc4KBRpVT9i3F=uIu;M%_b(}Q0in3P_%%Ktb?L;6&X=ImleScNhr
z^u}Qdwz3d2pnpc`O=Wx>yF*z!r4Iwqe%#U~<h@|UN@b<&Q!+6V8#_6LQcB0Maq<UW
zV=*#HryoxiB*S5JeA0*`+CTglGu$pBeM1m>$e6lReg`kzr~I~~!>4YxS9x78rpvG+
z5c?%~V-AC_6J(;wA_@3hOCsSY^)eIOyd!#st)wprT*|SyFA?TK1mF=QKpJ#bTH(B0
znr&q118eQRVW!^od+R~loz2z=;_jv-Zm$7}4z({)b3AFD4`!*#6Ncthgu8u+=FC#`
zKIB#;nYd%SPQv*U3tq=Fgz4X4OT-u;-sahKzD+m581R@#g~#k@F3HM)qCChC7;X7W
zDu+B-B$D+rK}!B%;y0_+Teo*8LCb0aXFk8arKLR4sI<rw%}7NAP%+CeD&|@c5uLj1
zf~hK~sMAuxF#3H~-DW=1wDwI}0`8D@V0pYndPcQ>&78SB)aSKr(EJziN}UihT?8=-
zA`b{6I^uD|(&?yN<4oAh1V%5@4jgf=d*LChOh1Lw!sL>wn(_Jpc9x?WYSl}~IPtl0
z*Vb}(N3W@nNc|$JCdu<Y_HJT+6IpJrq(Jp8UG}_yen0j;)ZdLUOKdvL=>HVN-1&7J
zB6UdgRr~6XY_zIZz7HGQ(M6FCaZs&v5$Dw3D@-a~O{<sMb<*n^zukwYnRo%2qT|H<
zuwBbk&&zFj<Y&xWzk9xyNK=7KM@4MCrm*cL%5EW(*9pxsImLkO^@|q>zvJ#_b5L$~
zmPJKqDVCUsf!V9!$JIyfdv8Gdd0Q4*$49?ji(?`g{3pO{(VvX%VZYa5ugd~3Hh=SZ
z@ioc+d9%K={>4t$l`<!Ugw8j+b&W$cVdBmTQL$Q3FFn5>fE|hjLPrg2pvC_)OVlUE
zg}43bVp>l=@Ls%+sT5HRQMXQ4OOHN6F>C?`xA_qG**a;%J7<sV=l560_8~aNL|s}H
zV>!nwhW2RcuQ=BQ?;LrqC$<wI5#lbJyx@8e7Fw0O5@I`Dn6E!hw|EUm#^`J|zwg>K
zEaj8jIRQ4L&KAUWvAtuW`Hehfa4=MASnvyad|RB$HEZ9&q|4_oRzK0bN3(1_auM?a
z?@mkau;1U{Z2tBMqx8!$MxK_sCa;zq!t6iAN$4l$vdujG`narhem6A1+qM5G82)pM
z81ihYs#sR`W+IlztFu^#lFRz%H8EW#lilq%)YznMZP)6)KqI-Ri<wj)2KH<HucE#(
ztc_?}yT#p|6nA$GP~5e6ai_RLp-|l2y?9&P-QBIlT@olRMFaWLbMO65eoZpTer9K8
z)~vN&*~-ZdkoS@VJLXsMh!PgXm;%boNR<JfebR63Qg>%CkzR}g4>%<h2+$)!Q#N1n
z%09^J8=NMoYqcEmILt!c?9N1a>^iMYBdzHO`K&mrpya``0>iYw)0jUGj~-32ON~;6
zWNl4wW7UYM4XW7r5I6`rU%EiO3M02XG8)0^FmW^zXYVb&NuW8r!P)hrgnvYJHTNut
z@MAi49U+Bq0f7);GhOHPHRy^J=(~;mt@uspd!a}#B8PXiT@bXvD+-4b&dcN3*SEwM
z9fgT6zk-Iu?tx*HpHb*067{nR!4l8vP(7nA&*Y#7R%n!-N&C~%CSonVUnT-=wp%P)
zZVYS3&_RkK$wC_ZNcnr#{whwz!8h6Tb^v#WLU(g4Qw1xXT~{UtR1Z}zyPEpak%oQw
zZ<y0;Wup3rW9KZL=$4-?qQ|`T9fGpM!$|`yN}YIBoWD(RY;KWGr4+3=-K2`LLe?(@
zo~MxtD(!>?-_;Pq9e>LWX@oly6lxZlrQ?lr=?$sm(Q1C5=DdhYi&dQDSV~gszfrJi
z`&tF+2~Uk9<}a!K8lmc{E-{$exxE53$r_dAcF;2&I-u8&6g8qQt4zr>?U`CoBA)dy
z>hO<FWlZ!qTVr2xsrf)i7CD?I9NPc2;itZ#k(^&lclXPGQw?ZDd!w+Iu6iq@8{q&!
zSjzw%s_&czRzpk|3+_m9vTtA%u>xAYMMKIr!V~~qR>RQEmoQG|SeW)U*<7WHEX-W4
zh`^Zm65<-%EvK@Pt)Y`rc35(HMp@@xyfg4(+a_p8&q-tAcTOa)M|9j`Q9xolHjiq9
zOwRX_V!)tjp2YI4q`@aGMwC1&3%(M@Lb)!meEmoW-V_^$8*il@=JI23Bt=f36l!kQ
z<v?r~Oz$=0s}g|^)WpH&xS3j;G-79??dupEp1`YO?C*b;fKYc)?Hn@}nmPD~##ehe
zySQ+XVXba&gC|WPJZ4eFNxFhIwm!Lxg->YMz2Fzfz$}!eLajhvDLDOG*pBptKPPms
z_JyVQ7CSK8w%|*_;Kx6|zd*MBL?L3oq6|=?Ua9uCRBJ?nj*I-X)3rzTsOJNtu;_}^
zGz}D2(d;SczBkA`i_c^t$f_MauW&KwV5V?c>M}oZ>>RiTv)!Sry=!s4pxLA#p2;)2
z`sG%m?%eP^!OAuT$}{SDFY>XU%JL$ecPc?NkHT@z#ZSmKI(~#SR&tJ+KXXC6upz+)
z;4xvmAFXM*7O<x8Eia7UyC=g-E7m&TUa4I9``|hj#?^!sE`1o}o+T;1U>?qmj_W6o
z1Y|Hvh=!Qj8TejEv80G8Ca-~Qb!(%vxA*0zUkRbd-3A7uI*03H*+)lbC$nm4L$0Oz
z-;@m<by5TojvHH5|1Cy(xqM(gbLFB<V+%NF(Jfel**Gn5X!*sdsKFY((<r%MZ9l0M
z`{Z=OVB-BqZXL~}L*}0AOQ38)#iA+%@}@AQoO5hMHD#5z3*4P)H$9zlmhO1wn0i2p
zeq-zV=NR06Qs57JOf2Sg)=k`~n-taML<SwPWd_Z<byCiqsM!g^$JGmY_?|D(1Ynm&
z@<tSI%iA!I*aJ&Yh!b471e$6kWLrhc|J2Og5<l*ASCTL)%8uP_QOpLtCbk9M(3?!K
z;e%cE!w?94)6&aX_;(&sX9IF73jJetT?>_n6`1R>nwAdy)XIu}vgDCdyGU28n|V8r
zy}k;}0&fTd&uyV1lrLkZ<@1hap-HfQ<m~4CCG^1-ng+d<fYKKP?GC$;LgQtu*y|9}
z>#WMm^N4&0sIS9{332`qaAqf=g*(fQ0Y_)?W7J~Xs+hX7SmLf?Lxj|Sq($Bl%oq6`
zA?FnzL~!bp2^u(tpUzgNHaRa&*d8OYIvx@(H!{vWb9d@t`v&Zb@J{OaUJTxI5wk6{
z2jzFqn(ovD_D!FMb=$muAiG&Hv_KS!#ffni9zZ8l^0l*W%YFhV3@Z5`H~+P+u`m3-
zY2I7+AVs}TRH5eG{yL<$mGoF40rexY5j}YMlE<u}lx%D2AP)G_?HB3Y*}r+u(fHXa
z8C)2Y%{QKGEs4{zvbAn$ZC*eKUfJ0A`QM8JO7-vJfGJcpR8bn<g>hWP4-z<);<YK2
zz+JeeN=arDBiuNprRh^k`k`?4rAIVOC2Q?Hi|_Y;1xQn&(U>bvWa&+Wyk=5Z@{5eG
zKE~iz(+7g_^?VP9-|{FuY*h4&y<nfZLu6ukSXbA^3<`2>O%0Lk6BXT4=a5VIbwe48
zSa`ILRDF_aFdCTZ_HnJ#ccZghy_5NZzvpQfy<mjCgcN1iRWfOhoo_G;ph^GU-cc&x
zmV;Xqd&p~6x&Gr&^LuTL-MPA!mVR-+5)<^s5-_4z_3{Z@STchFa9S}nC7`evxBW0}
zdbrl?06FXZd#|r=$lw~{p5nR<yApz)N1>R|@$DD?jgIas(0R!6q%-kdtFSp+VJ(j{
zxKl_+lT<^Nyb_^!A!|7c*-uh<dqg1`$me77l?hWa+p?_m`pa7ATjfX)ytx63<%IiN
zQTrYBwLBB&#(Aajg1pFPj_QOxm1h8Upw-w|uM*>t2H`tLpZC=4#ETU2DY<c9vX;pz
zDk`)INy)cp;Nrgx_MK@oiy#EXd6O;)F)(|2PGH9s*tqn5TS&3gO+hPc2Jy@aQ;w|r
zn4R~~-wc<<4G&w3Ly96FC-h9rZ2cp(ghF&vKjwY}7#kF`Dc#*|kI3`=FXN&4(WmV3
z9if+!ir(-Jo22(w5|<rMHC;(I;>)xl?0}{kru)b?En1zZTlj!O*QKI(Nv1@qsPAaX
zIz-yCYB&*A`*Ge!yP}HL^bv1Zrnk7v8Hm_spMjniXv@x+W6ecM@Jl9JpkMPkYss_e
zub2W0+&_e<IKTYtD*lN!K{KjDM6TeNjNf8YH)1iFr;qZ}Dr|w!P*%nbz8Z1A)T)JW
zA&NkoUM&o8M5Kr1ZE%528i#+!u5gU?F)zT0vrVtp9m}cQX&MVU>goFz=VjKXD*IgA
z2{UV|zB@Zfh2^K#<)OTUdd49nz4uPA-SwQNM}x%pdP|;SH|X9Jx(vC7f-)5wAx;Lo
z*4JWIAMo@G9ab4i-0_P%_Rx6Rzka|YfAH;~S)rs+ir1lD!v3BntsiNNq2`w+8D^>v
zY3Sp0pWr&!^EI@}R9k4q^S2XMBmBtw_7FmJ)m1k4&J$6a`_a39A{4H+b9m<m;*hQk
z2Se0CMZVsYF5>mcRBW+~lB6S<Rlz{4_gq1utvgaS6+2sKX{nzb$4@*)8Z6J)Tx8OZ
zFV!xUkE@MNu=H4>PmjVsf!OqO_5^7fv8#haswL|rT`ISq=KtcmFgWC~vb}M%7O4tX
zgCJ^`LdlL`CmgP5lt3|jxHydJA`yk0yShv^EDeQEo4$js{<BDLwfi?W{c6P<fI0#n
znYco9s<^;Sw#25n|Jm1*7XuZaMC^_Ag8^Jg*Im5AmjgnPD>eiEfqVL;&f(@}h6#L<
z)SbAppGpiR+)E;9I78+#6|G<Sa(GC04MMAw9Lt!)Wtp+}>0aFWliP(qFC?!awWcQ7
zwhDf09P4Uy#Kg}-q(&r~T;^1KeX89sefYiH*sd4EPsFevL%!Tvp4eF?f6a;a`p5J!
z7x>U8O~p3QHw)be8Y=v<w-4#ue!M?wHQy(9>y;G%XgQ?pQpnk1e3K4dv@}!<A1n^1
z`jf~+#+_Owo0kUb+19cpM}nM?Z-hn+J(y<=h~Mr<B!2n5VR@IP?%uq27axa65FM1d
z%FAfb8<@T|Wv^trVv>mCz&+hXwJshH*3nPhAyqD6G58Yy=uA>S4~!LSdpk`U*Aaa7
zH|f<0juH3gkp8Hv(Og09KH_oPtbou1R&Jr_#Z^J%#h^$VWP;EY&(_@1Klb-;ou#Fv
zf0!b>;{OLq?jF20BIQEec?K}<lf;4vBUv?@<Q(EF2uw(DNJD=|RIyy9MdIN(O?lG)
zY!qTM(BUFYtm_!a6ldVfGSDybjL;wi6xz=Vtx1ow6A_DS_|Kv)j0=FzyN9Oct+#Fp
zkTrt`R`Q4odZ#X4`#&UA^XxN2e#)?nHN?$x_3JBhe&Xt^3)wy{p3JiV(j`bc7Fbro
zRV(G;P)Q*Ta{Co5W*OMr^nvj?w?-s-!cxUsmj#)d2St!a0v0Flsz(gv3rK5BJ5>z~
zW4iX1gZ81}&%?cg@b_X-9vL=`;}Q?Z(68zNe-=;Q-ELodXzX^6fd~@-LS4HQ%b($x
zH#{wEt5KV6iU~w8#^NPENgEN`9h!#?@gn_O3>qN=xiW<EUE<0-q0_On$>jx+E=r#H
zcQmzTkLz~h1oyD`TS5UWXRYwYF4q@X@8#t&U$_q>2Zm!(T~i?dUg8$}&x`72KSIjb
zO8o37WH4V}AZY&nwwEDU1+dj3P?8s@Ur=Tb_=-sul6LF9WrH_z3_wSsLy*(enCFY$
zulTApM*bAk1O^$&a;n8lHYw#UkHY^8qb!45oO+BwunOKu0!n9IE1j)sM?{)3;&~MY
zKl{N9(Bq*{rV)?Q0GI4wbxI8vy&QM^6Y-jkUY+3~JCczu-K|tMftmez*~Ja!-V@dB
z^m!O7xh-~n25Z$S7MW6bc=&lcHZ`N+XIsx58fTYh{@-TRySs=fwt~KR>5%n=@{VYa
zy3S7Z3_^t9*@XeTX>aV^=QLS4ch~oozx#+Do0!-=Sd6Jl)id;~E;TeGZCz?D!5=+|
z{eiF1^7ZM6MSZI-$Hl$#LVboz?xRmnL^?W}y7bUJ(^rqz#@=UTqmC1bDt_Q)A3WrC
zRD7fs3W7efQ<ewqJaRZ(KqbjYF=8;J;oa<N#D?dDSip5k=2+9lHTCt8=yV;VyxQn8
zib`oJD?mO|?u3T1`Le?=%P#e9v&N-c0?DLx`CB8~#%4YBL#oa<*Jq4#Wr>t%pW_WI
z@asr8>VJtvQ~EH=hnVq7Rp#;mHd;Fqr_r=`ZiWW%T1CW0_M^VJ5&sml=-hmtAo+RH
zS5l2&1cfxoniPJQ+cz@1P}o1Glapg-1GQ$`P+n=c-)wB)6hkow79<~Y)&Bo2r~S+L
zacDH#R9I@x8Sj|_3;U2I?OL-ZaS16FczVNfRczPsUW&Y7H)}#0h59nBc=3;h4PrX!
z9t?dV4lA<qp%E@ejCC5EcZ}Wa&r$}T@3F!HGiM-#Cq6Np$@F?ERqsI%`h@@B94@!Q
zfRel60Yz(!brPs~Ri9$lPXWTtMxs<5-?pW>`P(x#B1uC~RKK_z;De227?JitRx!;j
zdwV*9N!Sm#yLhYP1=ODF)zR|m=`^Y1*4kp|jS8}Ne4_?Ri@?3xj72_Z`Sq~&+VtfG
z%C0P*f%=kU3TwZHLXWp!1op=d*uUIRLWdabt03<p&a%jRBKrp0&`JX0F81w1Whn08
z(vFyN)QW#cjJU?xvgF~|m60@y&>2XYM?BPW&oZVcptr0=G_zTIFIWIOXAb^!=Ly}s
zfj5vcoJ(#z$ZeFlX2M)7UkGu7&{WryWg2)Bv5ns3Etycpnh9<1r=p|t<@E!|sSOk7
zVE(3ofJJi8WB-p2VZOT<npAG#2%N<M$T@}L{wcA<>Pa~<FK1t3e!lk=kC#))vH3@d
zDcAbfEqh$gcS-!0s{C%wHF|xxaJyp+%$q3W@Wjz(+RkkeMklmRnb+c8GFaHn+PxDG
zefcs`m30^MA*PNLzge_&#GOk4^{gp2OwS%mX|UgQ$f#+|(KCB})<@{7L_!t1z|A2M
zl%5K%(I7~Xc<g3H7&ZVBTEvYWH{RrlyQS#`g7>FZ<&Xd7Y#6N%WW_wT3$RwoqW}{_
z%?szy#(3e3sp718NbzaD|Fpkokz;ceM=%)+%;5F8M$xO?w({>)SNDsnruX|;UZX}k
zRCw`$cE~fP+zZQ+tyA<%fW-Tv2g<+`(@WFW1t?i<b=yVHGtB;n#x(G@zV+6q%R#XH
zdB1n`05)SC35)7F0`0{m&Pb4C3b!Wrf`Lt><&5KnM_^E4cokJMQZ+%U&6HuMv?E)n
zq|&0{g!GM~y*^38N;J`GoC#<v-rslcf$a4U(ev^;8|Tct4{0EjM(X|&=HyO44^d{D
z!OVtHYs{fpnziG{(jL6aARoQNBwvcmyHTS8>>pbv(!B##8P=I|#DqPR#a$ox_88}d
zt+qVnZy$ZWwF8UV?r%|hQV5Rf&4nlyoqZG*PuayX|I?<}E`UW%;jZp~;Zo6I#TQ~f
z;2qiMk;55uY_FSem-R7g)R8g?Yt%DfTq6_X4qJhNxd0<-FsTiHuWDFdI>QR%{#i(U
z-a%GdcE+M}c&Wv_ij_<!2A^=639SFh=Mcmq{8E5zp?{EWt5nm~CZEoiZLhqi+ZPHm
z&^@CQd<`lrZoVho@+y+n0wh_}isU0zR?6G)b;5?Do6K^<%PvLGkqoj;tP+;i)6-ru
z^OOVx#}f1qyE}1Fxy;tzEqIY>nRCTe%l3Y`@7q<V@coPQ21i9yLp$Bl-y>1z&hPXG
z=ap^HQ$@fnt%U4G)MGN`BQC|4Q&yK_oAd>;7K)%8?@nodwHsd3&pDeVVjP@P_|BBB
z0YV@@Q+JunE(hq>Lq#P9_`AWRQ4#6|L!zo(tCXwr7x97$u1Ju1r)+~0$saU>G8D~v
zibG&-@09WKY`xnZSx=U{-p%)UHrcfybUcwWvD5jJroE`L%*4S<)E?iW!Q;I*66Cm^
zWRTrIxZXL&WU@1>bh~RIl6lIj0bM)4W+-fbVdxDY22@^t=H=sI;vLrEv^sp}*wfZD
zbVVC$CFmZ*SF&gqFfBpmA`44ex&(s!|J{+7cRAaufAj)kez_m=Ce}aoe*hR1fd8ap
z_B;$Bp;sq$s!yfBNIk+tAdpbt6J5`3{YZ)}vyKi8F3PtF<%q$zgX1WN{-<&LiRTI|
z$Cqp<AKr>9Z+;niwQYa0_2&a~xZd{r3;A5rDQ1))blnb#>@zNtUWg4lK9La5=qkJ&
zw^FGr(>FFOUM?v4iQp%m!x7}IalqWE=i>f+Of2Mo%V({lsZDR7wS%q<U~TwGbJpZ{
zc+DSU6ngE3#i0NGZ7r`A>%$*iew6!>^}kA}lHV@Hu-R%k*dG5J)p@A6gjIjpvl;4r
z?uK$=i$CBns^{%SieH98TKj|KHPd_so;n_Xwm(i;@=5r{R8HP(_*jQtMq?UQCvD<4
zZd9PVANmG5CrALlq!>M+SVlJx^s>}Pf~va?`ggzHtIA;nU<)q+J88$uEJ1S{yEnc*
zl-&}$HMX;<&9vSb_eWceksfHt%^DxkJzFlyMOUfSd;9PsK^Hl3S07_I+1am*{IPoR
z@+Rbsn0Q`SFr%;2&|wpYE^@lw{CY-RW0koxv7t;x=(Xjpr}@c3B;q9g7fZlJaBa$S
z&5@9lDDs-1Wx*xl(cwR_^&k7amtk8L<g>z{qG1q8V!P4NF@r7PUZ7j4c(>A6H8sM<
zd7q@U8z;vSo*DuddWm4t8LZs*2h9m%Bobtl>02IO!*>-jdSPN*0+#&Y@tuzc!*#~Z
zilOXKY<jwz6#&`YFKr1<6<<Kp{V87GIj-cBdC02@NGMf^Dvp}VwRE<jDB)LZnf2u<
zTA<L?1Dq!K-i^c1o9qvsoj0ZSa)yU#`?w7(b}}LFjO)YA!61QjFl{lzqi>(N|CS2|
zGs32~1Db9IflheKb^u<vR)|`sY~}BP`+zWfZ5>6;G%Hy(gLHidw;65<;?wg}S^(sg
z@|7d#-USNI6ZMFA4EGa&Iho=w2|Cu)*4zxyf=OF*0uJhJR}13(Jlo~{?3~l36ePd)
z`@wBGQ{g3bNRde&4_bN{7r^+%AMMG|DSj6{#?093<-Bn9361XaGCyy5<WceqyVXst
zv}EBX1QR|yi3E3!|32s7%(ute^Q}`Lk-;6v-B;$lK`?AgIrA{*37)Y1VYlysEet(S
z27U$+4B7?zpMNZ!Hy-e6N!F1Bpz3-JpSRj=i9@V(bbhaYnhCLV^G*)E%32`VlT+ab
z{8z>R{ug48Yq8)(hR@F4TK|ZK8yu#^jC<GEqs!CVFtxFVw?+ain-jv>sTxF=%HtM7
z9`_wuGG|0^!1Ob5bKA1y<2#U*O~EqAKS?RoFp#hO;8rh2!})ZZn%4zqU&If#Yk@00
z`Q{YvO?c*wEK0F2xbFCu(b4(5alKZ7k!bI0iGf`5X+y`HmE}EA=(KT0Y+jc_{c2P(
z)1}y`$Xmgs)~o|E5#F>Vt`;)d^1+Irc_*qcvq0wp(`PVL{f+qFarLXwpzBdsn-~q`
zm7$DIZ8xN9V++bM)u&3Z0PZH2GpAY&%xnqJ5Fhv@co7{3M7*TQ0j{GZo@vGJk_&AD
z{lZJ+6T8Y}Hyx%o@4mMEazu)Vr^Ph%o*G?JDljkrWe?M*WzW=KVois89Z|`Ugu2%8
zP+T>(3X;72!9N8iJT7{yFW%D<IY<O|@M&FETULK?s}w8#cq^QF5cl&Ku57d1vXL;U
z6AQh?Vps>bc7D0I&&?^gzOh~CYT;4%&9sOyW3p$oK9#mhW|t@j*ybO*9#1hoc<&BU
z=>OY|+ro4SjD!$Pcjj}%PR1T7Eyh_O4N{YT%yr52>m=n8<sCJ1Nfe49?GA^o{P(=(
zgM4&<y+s)xQevQ;Hjyy@LGE_2f*QjntAuTwmR2T*5W~mWC!7W@0*1%Yl;UVxN0SN%
z{9u|4@k~i#K=c`Qbs}hx;0fEJX!PJ6y21zvDQ2o8f-|j~i*;29TSNiZ$%UBW`BEG!
zG}brIo=c7hd8pvUS0lu~Z{P#F?G4dj&en;|81(k2i)Q2j-9s9FYAD3j9i)FtOJP6b
zHj$trq^d4c!dC*EyQ182#$5P8RuW(615arwsZrp>uj>o<7{#yfArMXo94sd#mu5-~
ze?-*PORVc^QYo45OBJ`zaeux$)2sY!$!6#fmSklx?8-NiNgnVx&w2Sl;>9CRC@>F}
zEmVJGyE63PT>u<28lA~O&g=0PseO;RP16v{Mv!7girTM$aTRr%VT5qKY|L^AlUno4
z=v(}XC@CS=b6o&Z@tw1)qYaEuUnF@+eF6n^Y12z9PARH-wNMYfIIDTiabV)7(#Ojs
z@qWpXPar0IRZUbGD)Sjje=mxp+Y#aFlV^f@E8jhm$zbavy5hVXW#-jaqG@7cVuOz#
zbJe($Vm?&N;*4^qwvKP4&YiPyi)cK&w*ES+{XbL>O2OM(a>3=j*27vK@l9i6txJ{D
zWaM5<iaeMxx{D#S`%7uc&@_WN?MR|jnBW%xD}c!<nq2i@fyTOE!EqR-0a|zsdn~6?
zMy#e~PwB_$U8Ozek2VSgPBPuV{lEl>yU4M)?|V2j&c^48V#o>W9QL6~ZP(8cRI%|H
z7=OZPhWE7Lo+-XZ)}M7h?niIZ1+C&HBrKa*@OO9;Xy$HO4GAe~PB3Bt7k=$w6?TgP
zv9^icEZ*5KEI2ejtR}u#7c;#_2~LTj$))O<Hy*ro`$CIx+#<rp$=2A>pak>pA)9=C
z6*tI8vBn?S$CXMG!Whw}!W`TUn~R+N^128)GJQf#sH*8H!~fK~1rf__U5(8oeWqxS
z9}^qzb6D71A7xQoa@69p_{lWnw@B>|?z5Y0n?F!ijz#dB_^mtDIf6d}W6ND96TO;6
z+1-tOR&`&dn&2V24tY<G<*HpklZ3YY09Ze5^z=Y>v+MD)+j2ll31^sP<6m(=o+{lz
z)n7Z=2Tpom8<{~5aAyc&)NODt$TD3fy4F8QqR48hfmQUSX%mubc}t<XS+h@K#ch%V
zzW>)S^>Yfel{E$desF8gEL<X{(wKko551|F(M%)5((`E&b?VZECwyW$q%M1}5bbFe
zU-VNcg{?wCIj+ED<@>4>>aeC+URqP9F)aO*O~JYUte`%B<aMMU+>#TjIrB7<)jaLg
z;v;tuPNf^-&2%)QWFsQ(k&gsh@{9^&POcHgmbc4t{hYFJjtG$0D%s2(Ztg&EUqc=A
zz(sD4G(Ku*7NwfU0<*$;t7!eFr#wF>cu5&eZ%DS@25aX5-9qkx*O!$rHV{AXW{OvP
z+v-ZHakX^$Y~jrN*vtR+vsfHA3^u_*coSxhpa>V4%FdktY>bnq*Q+(L6asKbw7N2u
z`Zxx5C_J>EVFteZ5*~7x8qF8z_}v#KF6XDu?5bt)j>*zamtMXhFT82Uei9J#*CsYv
zBM-?)>)NPQxp(nfHeH;*SXG6B|GOGPNjv_@lDtlVkt%2B{fju9@V*ZrLIm24DT<?D
zn~Hweq|dm_X6ynTEn{g1!z6Gi!&YNJR}?2_6zJ1P9X}K#YFEYZ_q!APDnzBt-BakE
z(Wi+);MG-8-@ow{X%^*+<t*DKHztjxn~uq)+Yk7b|0(_d)s!!mFqwt%E+E@Ofmx%c
zREl?E1b~5rAO%}&`1`sXQ}`uaK~7KCjB?6V7I4~CD6aXzH*1QARL!*<PLn#F%JJ(Z
zLLh)TEm0LOPKg_@O6ye3y`bB@uR0g`!CH-1gUtuAy2Y~P$5G4LP&`oqB~{vh-Jp=7
z(2)zR`{aUFV#ExSdVYx)gbz1Fb43P>C=$nf*`8tmG8DV3*`$##uI?)*<+**lL=K7r
zn|{ul)H5w+gF>=9;DOZb7hC=YMka8Ao1eHQc?$erh2vgEX1CL|5hUN^L=3sXA}LjZ
z<pN*Xyd91#t|X3Ge70KE5H*56tE+C33ZJN`=<PGIXJ>co?c&CAt^;311Gf)0k`aE4
z*5kpZfEsfi<9sAY(}YKYqA|%vMyRTp&eu2l@WC~Nhw!G@q8K3*$y2Of5@4R5sEn7U
zq={!yrk`_=*XI8I9`yE%rvjJBQ;%!e+fk5Ey=BeFp-BPqCat&x(}xk2@{da@I72@3
z^Jz8N=Y0B_2v##<;R}8gY^AOy$(oR5ep7uz3m~ah#;~c~R+Qz)iNj($0}4!hx^&dK
z8NgG3Gdo4)sPkXqjE}gMtt*0-%6c%Vvf+UvrDJ15U5rb=?qZP)?|j@HBL)L$;J2*o
z!Xo~4X<_*whs?jtxlL=lQ`H<ar=H?@e-U_4-h^`{5u*BKC*86#tcDT0FmyFJNq$tD
zQdBFj=UsQ&^+S@8tmuBfBbYov%gGb*l7B2CV`<mCHYAoPR~ae+Xn5GdcDqf!Z}hog
zQU7wD8b%?L&k?Rs#cGGfHKHYS={Y`RFJtrAE%D?O!!o#5`_YDtP3r;HPm}fyPPFU@
z@bmYNg`Hcepim+@DHKQ_ElNNaY#$t5TF1~0FvP3_ZsqNW@Br2OH-5?cr=E&U%?Z1{
zLo;_oMOp|2@|qbg<K8hr$$mNqT^DA#`QbWGL{`?9Dzd<Nz-D0`*Ekd0FbIONY~!Cc
z<L2y;NomwM{`U?;y_O?o?Km$wgOMrpE1awi6$f6%B*JQ~{x}1mH=ApvMZw7+D>aMP
zNuZNGqQKmkAfUec{jOmBgk{f|Z%|O=^9{fnBe7)-xOkJ*Bw`gfd^sU98~AE){#h@y
z1gussLAv<Suvm$Zyw*~@r(MqAQ2*Y`<v{wTl<CI=n%j}Epy24M>%Ehp`a+oNJZ^{c
z|J{=&x!VSsdWNBqrPK()VO*EETD%JQBCAO(xph6>xhI{Zp_&u*t4Q}7J`r1v$d>&3
z)9>zDS~RuX?@sO4&^>S-vy?1Z;+u*5+e8E|31n#$7Q<Ne1pM4g>m-sr;Rk=DW?0JS
zOnK|s3FfM#PzTQS8%mlGH5pP{XQwhq=frbp{$jwA4mj4(#OI}0h$Aq;*@Y0b)|i-=
z6T0o4ilIn+l?(Mup<)ef@m@Q4sGE1oiSOC{TaTg(Ti%r<9_**E>w8>Ayt1<Ld9%e`
zOUdl1%fvx7!*(d11Pm~A2-prH)Yc=bu2>%H-yN&4OaU1M9-=9Q!HSQJNj(!7X;gH#
zoA#!RZ`VS=N-(#s$(y`6cb=jwYZa-m`vOC>@f}>Ha8LUnwxd(3;`l|O*!0K%fa6*-
zT~3G0p8eVtQRkR>K!$1|EYW23z|7Ru)h_U~RF=~I`gxPkSs<5<GUr)$jt$-0ka~17
z@19QjwQ2s){g;gd26Ne<jA~k85`zYz@P79?o#@GFyw=m%ji>z9NiKNX7iEKH%q#pF
z*tDe2n3bPzeg^v8x(OV2pJ}mZ0(*RNPpK5F`EP<c`mgXX07sP;Gb0pm*@2G`xO<4}
zBqrJ#(l65;Xr<649xv_B#Y7V+!w$y7&Z<?V4m}_TiV&tD^Z=Qo52Vrs<ypl;!bw^<
z{CoE5!arQ&iDX6hcR*O9U@<ige*1e1!!shc4VqdOvNO_B0+J6QI6~@sk<?2{9r*@(
z-^1a<nzWTP%Sd^PBKff#c1O~QSEJ!5Aan9CX8K@Me?LlZ4pwZ2PypkYiTZ-1E+M4w
zl@n?!{`@%Wc6=V2C*&6=@j57BHb;3AFR%V?#vqQHPI15&SiXFcTt<qgSOBI|lrqSu
z-6B{VSnd{~`$#IUYZv$;7!=FKY+z_mzUkZ;<m$@U<>UCet2r{FJFvJc_dv7|seUuG
z>UT|5ny&ItU#*IjHcOcg9G4`}LKGLcy^5z+`k<{z1JGw4Bfq-+kf2QvPqgnX2Zc=<
zYubwww0}lK?iO43bH#bqtc;3>UKb^daTP=QN%vaDHbe4P&N(qNdyf`hWn6bZg%+x(
z`0mp0OkZ=$xNh*KP=79ukM;22#b?18R0c&k0^e<cn{~k@g;W^1r|kc@Ly+eU$V`j~
z0w#TQ+2HE2E5i~^(_wZqbzx^V&xv2agGC~*2elN^R~d)V`4YxUGU^e7G%a5f1s*lh
zca9%aN5Wvno6FSBE%P(;7uX7o7!K?ex*{r1mLt<hCmxecnF;fEVr0W_Sp#KSajCm-
z!pqq_CK{wr62}~H1U?GA)n$<zN7%QJp_SqHr+cA3N}*Gp*Hw?y7Nb1&BiKinMRi?l
z=0>~!{H&^8o9F?2hIknD2Ie!NB`I-b(`o-canaV-$pOx0sf?&FSdQ$YYO$1tvMJAZ
z3<$?A{r2QrT-9^=dOZC`_7G>m(%wEI+az(h{p6gf*0$%4?8bxGAM$DxCC(==8hCMp
zIw2S3PE<{>hHjtmK?HfdiP16w#z{U&;$Qcj;a;Sy%d?)+pY&jTcYcyMAD_iJj*{D)
z<wl)pG-J~_UM&}PBZbE&nT?-DNV|L0uxz&ANR8mD<#s*?OpEC@SMv$PA{AI!T@Saz
z>it7~*Yi+5<>ItqF?Z4XDR6(Sb=HWbF#XV=gV(rbtL>i~!GAK&AWB$RywKa7Pb13f
zJHjyvPNQgjMpSokU||*BP6p(E2y+XinXM}-5?hp|aF*C(#m=Ku#b20z+f&cyZ@f3d
zW<sDzHf=UHnwpZWmN}uVZQXz6A<w-1j2rmi^7198u~Fy)&uYi`8u&FXzCwIds(`8z
ztrB5LKdm;Z7>CezjB4l4Uvi`4i-rIwtz&scs$}ezBQ`ak`7i2fG`Pe0xQ{aCef>)U
zxLEP|YT85Q+wb#BtF5_62g!|^_5dUfiYci3<4N7!#yQghGa+V%XPG*Z*|ur22`z+o
z5=<+A__k=An8K&=LMt!p?0v?BE6$*O!L2#rUX!h;$Mys(2Q}^l@-V*$49t>jj=Q-o
z5k8kv)cE<g>Gh|#9d!8yo&>lgsxEW#%HxI=lNmoyCPLah<lDvVuiL>UBc-jl3ykMe
zuJ=Ezx4gJ_WF^X~%F!K5>$Q6<0O47E(#UTwx&uRzpE?~2K8O8n28ga$H?^wkkZo*e
zT2yAA2IUm3_F)IejURN*0>;%}4%8dEy2h^XM<4%gL;fjrdiQ^UwjpG{Vd#Y5q-g&?
z$$LZ{f|eH9_bT}sLi^*kz)mZDL^i|_S}{u1P$N+?w=ia<20HpEDIlRn7Fsp`MCiE+
zdD*%GY~AV)8HBQ$b7!6~Q^21;zfa(uAhRK;%GIj_7v+X<2Od4u3&4)H+0&LjZf8D$
z=^qMcui<I%lj9i2gx=)WFBM)st8dv{^>_fG2Of^w6qjSBFCQN^vlwA~poBc3!4I=y
z;o#@n&&F%r6y(b_B>|`dqo2Om692*4uPR|uUZ6AQqo`}`4P7D(25agNrjCd4sa7nj
zD)m{guMFzaOMRvEg_(Lr$kFbt{dLF_^Ut6{8QZp^ToJSo-@YN8GgA}w-bc|%8&Xsv
z4Ywo2$w5<hh(jLC7LaR&H9aK_aETBH)gpN?V8X#N$~scMc5RNU)lp8%2DmUA?KGeh
zkJ>oQRsYVfe4qOY-W`7)pq@4KR>?N+Og*ZgYa0E&fY$F(xM3PzWSa0A#C8$5BQ8V2
z)p3Pqe@!FFblj5TNd97M8@PAiwkJ0fPcA4FbZ!JoD}t1NO#cUTcx+w&7vT^W3|Q#O
zbmDkp`4J6%9_4&Jw2|%n;d@o4%CX5BRU#`T(WjGB%Jg0j#o?`FL0r9#tczRp(gKZ+
zrbRxz-N>T2{a3ujiB_(JxkD>X@qY9|D1MG|<t;M2T7kHFtFScqmoO;s0jv)#2^!{R
zDHTwlM$O@FZwc6I$E=YvCYsqA)(VQ8kDtAsg)w0J7up+TVkh(bm$5I8Qq_}eNns|f
z;lqA40+w(DMJ-my@Bz%d=B&xBG;-uP2~?Nge*%7#`|nv9L<mGsVO$0zcCpkinVWuS
z5G{`nO%Gw!Go{WhJ^1a}jOlV04@H^+=c3&#Fg%&`GnUArNxaw20Ts22@)fF&cvi+_
z5@nKbvhNYoFBnYNs2r*?Cgu!oc1!QCxC`HZh7UZDrZls3>06MBoaEdz^KgmHUvE7c
z&TLbr2W}IPIMqqHH<RK!Oip=_Z^U&NiF_gE<rQ_T3Cd7j6$h)c?;5^!{_8yTmml)r
z+!Pr7?ROuPgD-LaYo(s4IUk@=<KSQY#a{NGZb&P~w&IN_%b$U{WSx)!R3+pW0#E09
z0SjgvwpKbtRBL91o43)~MZ`l0a1Far4GM17iu{Vv14!t$>bPuEDtcZDk(3BbvxCCH
zUf<#m&C;$9VBp{;xyBBYb5-Nh9I{Z*Z0Blc^Y|BRvG<K)qZ5z*euwHd*|V)jVX7gz
z@TrDl9?pO`q3f|h$e8*oAIEsx)5?vC^I4snW$4a<+xcr>Wc&S|EL#rMr-rV!_npgh
z+vKU4IRnAmF2qm*P0`?%Me7;0uU<%JZ0$89jHsQ+@N)YICp~vaZ<3^F!r^xCgYIm7
z1INPAedgl~RZ?zM^Vi9_)v>29CXkH;dAsn2wBb4iaT^p2Sd@tvnK>gdO@RI}R^|~q
zb6sd~abW7a_aZ29*5+yMf7W2S__xSH3G5thcfrUj<f!K_rW=E!5;p`VC!B6)cSw^g
z7FXXvx+|rQ{2D*#r#vE%X{Uyf4IIR$BnP>c1F`Dc)f+tCcWm3*?@FHgU%oBa=s@1w
zHY*}-cs6#d>H5l>8rFv%{4c6C{sOtE|D&g@JN>JY!Fz~Evt1!gM5^Qc-)J0Bat7v6
z&~3;wcmyAKsMJSc#Lqdg%_0fF4zuJSI&<DPANS<*0bs*&c^#QKJ<qi!pQJvYaOX>&
z#Fss?!glZ6ZpCOUf}To!Pxu6XT=1-^)d@&v$=L{8)9H_(DdV{Z_ffre9@FwC4<dMl
zk{G6wSDUWo;%yCqxw7f1CvUESWoo7ESqfeTY}=um*n7&@)PQ^&{6jxFdo@iRA%vXe
zLcpWV{?{uZ^B0g)jcG5e{m&$PIDP}D{(Iu`o{27AF7?VPWQmrDDB~gB^idu%7<7xh
z{mXAN`cX;*l$?)QlYj2|NvMyde+V5{?ebw<R)849-7Qe`TZpwjIZ>GJe%#;`oaeho
zHr~9x0U^7gZd4{c3C$b%!$Jv}k9>qemiQ-tyP)8GeNHp6I9SX}j6;uc8wrjRZiB#p
zs-YhO=yzZDyRq7lB|oP(-goC#++ir8cAPEwv_Cfe-P*+77_^mmSrzm$fuY|2gPl6u
zv@ZU8yzolgW%}Bu@%FT+m3bK9etQXtDCAzQih}#j5<WnCEgevFba7U^n6~PT&Potl
zDbHOL87-aHh;o7wf@1&0l@#9iY9;nL0CN4-Ht2!TZ`*GW-}HfC9Ym(pw6ex2v_e_$
zZ5i*7Y5K<0?RrX9#b7VT_(?JRL<9g=W#i;CP>L7l<1@}^$%4ViSpdF0LX!7aHQ6F{
zv-gTNPnIroOGa`aNx-{axcr?R?P6*y5$}<}H?0Ue@?Jd@Nq{Vg<&n2Y+z7s03GSrw
zq{~wmDCm?^Gr^s~We$iFgM9C8U%_nZ*qTKcGpd(ZEQL)Pxx+RiGC?&!ku!T7gKWAn
zYAkL>tFiGsr(PFycnA|;lecCWi}$+^*5}7ZuN#NqALweTvdAn`V@)TXs`=L7<>3=L
zEO3cRj5_)T56cZ4o20(ZL^;^%&U4U;)T{7_sBfPIF6W9m2tOC?2iz^XoOgY?Y2}1P
zs6fVlt{kHLy(PuYpkD?v&`V@L3utS-KRkjsOqw|zqKkY91ojrs;Z;-LNiCecv$3(|
zCfW<f;=sBJeS_MjvhOdx)~GKAS_4IG2YnukwHz+<`SMlh)1^e<b<n}F1E+aYafK4m
z8fHP@p1oSW*fb?{p3?3105uR~3u~6XMeY>!#6V%KZz*0NbnwTgK^~5;gxmyOZ#M?k
zsfwZq+Y>y)Jkz1!H=a`cGE6325IpdC`w$&DE;8<quSg+)VEiuY^{}qrxF`&%s3liX
z%J)VVXCopyWu#chE>F#-B3>EDK@Xw88>DGuld;YgpNG&@g8=kaB9SPeQZzU=DjwCt
zN8bY3V{Tgkd4S*NF{`S!%eiM#@JYa*tJ%7q?!WvX^)H9aStnGZ@K!!VV6)xiL{~a8
zEh8vkMya&==qP4q4!G4mu7s0x?nE1sw8GTlykQft3j|B|XVd}CP8-^v-cCSTdy%Iw
zD+bYF$MheqY>;g!?9p)X%V=xd)TZp4+ciq;MGtu`?gAd!tRF(4Ov#~TJoWWEvvHpI
zCB%o6&<PrUr_vOC+aGguvT`Pe7%A{~&Zn0y;x8D`&B7k3pj{4%hazk@r+pUabnUAp
zh5wbG5@zxd&P0j+7lat3Ldd>^MPOh;;_JceSHd`~JA<HIVUQuWnh;$EA3IZOSfSq~
ziY+nqtQ>}S%W6k+f2Q;KKFj!zRTNGfBcOkQWrcc51!Bnyyp63*d3-|IkpKG0lVGMU
zD-8+(r*R;u1N*d%N=8N3<(>y@hPhgEfuSJ#@+;5&xlFcqIimEa`Sfy&Y&^v*wgSRC
zEAj6rUT=B%pa<5FY@PGxBFY2MnFH{Jv2gFt$ViYGMH|Nr9$B?OsA9BT7-a5Q-PQaA
zO_5a&C32W2lPFAaz@Gv*V62RzLkmM?6Sp=PtLYpXaXD~K5gW5W0C12#&ClKYu*bXU
zX~CDZ=0{Q9|CAUo*CsX}J}p7A3!W?hmp@c?CL?EaRwBW*ZZq@Yoje_8&owF;p)3uG
zSfU&0)zQtzjdXG>t!dw^y17fg%YsE;!<&J4uUBv-d<}6UWQ7yr^m?rx6bIdh?&>`>
zj!LLHv*XMYHBJf6&S8PFFF5c;`X`J+d!Gse!gw>)JYLEsPffcMHVnaIo0Iw&OZGs7
zi%$P!HNk_jYWR`X=cZh0lV3Yjxl{qztkjMp3e5IM)Pte8pB!SdGvsOLi^52mh31GR
z^*;J_dEm)XSE#YkhVOu;*BxD)7CyCR9eX_t5R2cJn?etKGxG)Q3g|+-W@*`IXWr$w
zl;xOZ?eGWm_|DpgEd}h1h-B7cwHZ&%33c`<ZyY?~xB6aZSC}jZ^qkxbSIZl$Z|uvz
z?iYfG%rYxlmPj|c<A$j6%inW#_$O-$9w4`3G4;%BJW(hOe2@F_^)C%*D1>@_yJi~u
zKWe{6AMt;QPjyyrQSo^t{*!ZUt@&h#vu(y<Bn8!`1l5;gk(U9>E_>k??S&M2jyn=n
zKxb+ir>VNH=ji3pIVQ)nSDSA@GQKjV^Jd4)%?UDO{46H7eEPBTVn1>VDh+8geP)D=
z>xs+Su*W3T<XLivOELl{PN(2_HyN-7ZN+b_Wi0!i#c#P*?>G`)`!BAx!Uc4aH9w2P
zjG53S=bRgI(>8c;KefbjVhHC<k%;#Q?%kffm1hvvOu-|d&=CP;E$sLKvUAG-_x5(u
zEZ^K=OXPk=-bAWU&j;0@+df_sOUO4Ck;HpP@4d0)UtV5%YZ4>{5**X(wGBW9rAC)<
zE27pg1~`x6Tx{*E7)6He#~K-by4<NgnX0C|441xH^x0N*^PbvKCg11^4BhDNdCcs&
zv?e+~yMhne58c?dzx;La>MuaNh7pT#{P>+`7!79cmrAQRzGQ0lI^2zWzfHl_E+~H-
z(}iwI694#h!KztU<3H3fQR;txJ$rViVA9{ee`|@F{$3gTR@K(_;(}mSWk=tqd!Y$y
zu(c)sv*BXHk@pNL>uJtdZ*^I8ZpD=fXp}GkzrzCq_OG&F+}+ZylH*huao+yZ%|$P)
z01See(?PH0rnfaT^(xfzMZEs}`tbprpD)UKbpqS^pO3m|KM;HUXV1U)+webPJO1a3
zu(ub%zhsU7_jeUD_wWA{*Zlus{HeVDJL-*CQpfko{~htyBgGh$QsvJuQBc`;)i<!0
NqO6)sjg(o){{u=v(vScE

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_017.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_017.png
new file mode 100644
index 0000000000000000000000000000000000000000..94c0d9298c0d561c82731f84a2cacc99bc2780d8
GIT binary patch
literal 29575
zcmb@uWmH@5_AQJz1WE%0hZc9&V8vaE7k6za?poa4i#rszV#OUwuoB$eTeN6je&^nE
z@9Fv9kMD=<CwoNp7<u+uYpyxxN`#8C%u66K5CH+<rJSs!8Ug~6>+|yf0NV4h1ghxt
zyzsk7X}hS~Te`TLI9VVlnYcLE+Pm1=n3B6$I62$c+i|h-v2rn!Tf4Y8I18|`ef&Rf
zV6}I$VuOiF>_2}Ax`V8aGXlaZlYbtFWg?|E2nbG>a*|>i9$Cka?~`qHywYFzd7qi}
z%6PeVROqd`ra{+Y?DE<J&{|~!!J*WlMUraMz2Ob?T(5K+F4;eaO<9NHghA;Oz#Z5S
zN+~2`WZS%ss_HMSMy~b-XVWs4zqh_-gyCk~;Lno_s?MU|%>OZP<l^F_GR=K_w5sy|
zqXpR`HT>rBYjYrqfXxuAX?X1f_1|aTJUq`fV+7d#_v>kEFaG~I!+D_hH?OjAbxpBs
zhyZO4b<nb%B|-KGG0`e3E6I0yj@>6793rirUc3{!5b@Z|_|ud0CgnEH{6|%FwbSu@
z-OWcw!>7m5D!pci*X4eI;a0{QyI<e`tVjI&Bj;zPPgF#%<CeIex~|_p$viFo`T4HJ
zZ3`${hKixs#N@LP=NFGbz;?deFyZv@13g&-LgxSgA#r#j$TFP04Hh*`(9U9*TwKqR
z0E8t6l1BrA2#b+WylV(#8606IRT2zxpiuhacr_I;u>lyecX}&`@d*T^kuWDDmk3G3
zp+Ff8aK$pi987<l15(s#tzCFIKu`jmKw%QH5s5e;0L9pqY)B%Ifkqd4&fG_6(Amsm
zOa>=9XWki1A@F#7@rQ(ViBb+luRYa~V{Fgwnn3%lACO{n!by6I_NW$aTvam3Z}T}}
z2TfR~y$;{(B}_Mwd&mDF_s{j)uAp$M(+uz3oYsFWVNw75a3_fRhB<rNBUH=Zobk+F
zHnnZdN6?l|usA)qWKya(X0imb_v<=f9dBQGi(Yr8sZk<CNI`;T`3KWS_EO|Qp?JOI
zt1po9k|_a!{(X=1!QBEe$+6@J@$^LNv<?cW!0=w62`NS3xo|NEpvM>qZ(IK9{Tj|#
z*_7le80>R%b<7a%Oddk1!XX9%$rPIs7jlFQE()VISR3Ru(Gp?YwWk4QdBw%bn45;{
z1os^dvgi1H*B(*Cv{xlzrgXr=ES5Paw<nKPmgTl*ES&fwW(aF7yq*v7)P2+R=DV=-
z#4^{vmInupKMB!o!k#B7nOTDVfp5Ou0fVNn;)ESz<NO{d@Con<X_mfS?eZw+3duC7
ziD3k|B0cYMl-hdM_Gyt$K>W_xN&hs9rscU{zKVQ4l8)~5atArX>{AH8m1A;dp+Ue`
ztSnJLSRHW{*Ifsh*V57Dy&y9SOKokVB8V4#6F)_lgaR)Rfq@0SujW}bunB@ldWtfb
zh<Xk>U~#GHQ_yG!Bw+<m1E|GQp$#w!<NyaC9%EKrUU$T&seRgPm=p?{K7ukXLPD*+
z*PiR!ANSlB3Nj8bN_|hKTDWN?DrT=32x$7=4<AzL49KTF@;P&5|KB_PujLKa?hDSt
zrv(wufW~<W1=DD)Zx5TUaQm~LGJa2e%&g4`l(b|q8Fo$B;>MK{dQlqWG{qQm%A!C}
z^|CnvLpQMb6+b3bFT3REv;hMxP~|OcwfYN{HYGB&4RC-9CYoSnjR76e&HjlU#Z=Z?
z&*Z{Af_OFdQ$q_VH0<Owymg>3i~kX298L)js6(=1QPw8KkC+#c0&I*XeFAAiC)?=x
zR2e})3Tkofyv8qN-mC^<d81Wd;TwvcERUyTk&QCOeIs-!#u5zsmje1-ZW$qkL-ES%
z(JNipMIZpfjYcwiJ!C^g?2`s+d{R4UzjxQK|GWFkTq303Uzz{Vp&Q0|0fr1xkA%-I
zQSjBAIXoWXvkCp7b*`d`ePbXd7gmRFajhSMg8~^r`6dPkDDZ(%Ce$w*7xzXWExXQ~
z`WopT5BMp*MuF>~#M(mEvoYA5_pWIw<;o;ucLq@LEeWj0dCKOX>*ADn9H(*ni)5$z
zo<UIiu93}A2IU;@U^nm9#>*Y^hLhJWQR!t%9o1$GG-9A+y@ON6!*+w2n--4{3Ki`1
zK*JlfM5J<R%Eoy`lPLhTGL+TGGwhY$@2y}?B#Y}C%x*%iTm4!}N-&Uu2P*`QZ-yyf
z9<+Y-pLaZnOC(;Eu811&MHU+|u@@`C=3;S-kMI(gcMfCK?}Bqe=#f5Ojs8GI+_;Lq
z%4^S|cg3|!H>i&?Yh_KPm|86qMa&~kisj6(md9dqT@sNRxJTNjnrrlD$G`XP5EYiq
zKRA^_b52>>-i{2X+zh9$Hg$o~9~7F(R2fpp;naT{8))ttiGkKQUT*hg3Ei+?MA3Z$
zhMR5zStdY`7vgzHAsW97THKtLjVF|#ss)NI%=YAba%C=rq@Hg}=2r3+A#&sr_*Rzt
zuI>6hC$r}$h?O-piB|lxmSj*;?~$yVNHri8ikp*OxOfmqwmAN*FXrt3=C_>N4P!s@
zLS>8?ATknmAYfRIy?T8yg)EDO!1Hz#t-~a)wOl6{?;iRUi8-lA^hK{t@C>BRYMd#M
zoqp?7O{wxh4Y3IGdg~O9YK#d_c~!+@!?`*8zH{t-=eet8=@;gSSK99U>gvwS-zQY7
zW&wqSADIDVS`|fw6um@;O@Kls*|frO*7s|-yj#D}UJ2i+?h?SL%s)Bv!{jQN3`*ox
zbpN<*V!0mE2_X?eC#7+vRc#j95sJ+-OpL{rvZ#sTV?u>4B63TG8$OM5XJ)3g`fRWi
zDXLH!LkQzbh)x(VzDCz!P|%iQglSk*2Rb7vTZ<MTN2%~@=^fv%7yO<4*q?8Oup|rg
zI!tWw{Ewlo#5<;WE^^g1S2{*ZCMCiKw3?g|U0d;krdf#4IygCfg^J4I%Jr9dCJt8j
zc_m(=?l!sBI-a7!*X`{~l{Kpyqs17<r<fzs3^&<^+r#IiF<#kP6(MSo^zoQp?a|Uy
zc7|O8i`kli2%oD29BGQwGE7}vgXDQi)(m#M*X}f7Sq6LgbZ;Zo5MjVzt{tnWjuxNR
zl4%}2zosM{^YTx<lop=73(i}*7}A2%%whx#&^i0shS{C{O`vBgND5n=8YO^&P3>HW
z>dOy80^%akcZ~iZVDV^&!=PzpU!<I7+@gXI-Z!gX{}~S|>>}x^Vbqus$&vAt2sqIr
z=oTO2UT@{^9H!_AIXr>oQ_;iW0U4JTb%Tm`4K0rzU(@^BZsb$OI`HlZD2tg`e2Qy7
zvEm4jG`BdOATa{VG@;xdC@`X2W8dhQ+So+cxr-8N)xxK3c_SD(@)IEj?6Y>0mANKn
z&PnXy!_&!AWiri6Qdr2bqy`1;4X~)Sd(I1)cMj&a`kqL2Un>!ivRKH9Zc@ed^aeZ0
zit7pWj4KBj<T6|sR5+<)8KZ~Hvc_6Gccii~531PeXV=6>5GcrF$x0?sOoo3P3$+Ux
z^pbZYE?NhJg5@bf=vTUvCPLkBu)NMT|1%~`{)q`YoBYw1H=tn<&1f200s`GApq5*k
z$z!jVG56s|{D)$#HjXM}tDeL9gB)j3qK4IXy4*<Y#VjA_v^wb5<^TclXeK;ghjolP
zdY9}wC39lRd5Gt|GG*mo3;9Qz$)ss(e=ar@R8HKMOaUP;w>rNt!+iA16=nR;M9ef6
zrBUH}7xcDcRS;kkfs9T(&0NZN{PV=#Cqd-;qAMJXEzT62p&v=l<Rhwh{^F84pM&iJ
zkcF@%3dtIqBm_~IR0TvDbH6sL6x*g^Ms&CP$Udsn%A%$gfs&EZsLQ3X8Qu#nQxQ|)
z)p2R5UnY?;;aMcAbbv5mi%Ej`HL70)&iyYy@Go3&A@Th|RC-0`tz&A<)8f^6fmCpO
zV#8is{tyM!-Rt}Jq3F=iOupv%g_DMQjY60Cwg}CdI!k*KB=ZG;ioDKs7&ej>jn=WN
z59`qfQeqQSxaD%2UqY09M|WOV#8jvupXJ>{UzVV6Vqp{)ri}1cKEdb@X33cJz=6$8
zp%e%u0BP}rcWWb^l5IFDQ1PY~u3A1zxU5@jKU7(5@3cM7(KzC9KmK03-8UPz^mR9S
z0Jc+?LQ@~-fo8Z`L3D%yLKu%blV0VIVsY6BAf+TZz4s4*1_&2TbI_AOEP#WEkseoV
zDTTpu7k^CHA+=z*T#LklsHuI4V~H}JQ54c`4Z^{bmUP@9_xP(;_Wx5W4LW=>ZAtSa
zQ5j4fRP2f0UM9uWx2_g1*!~*uzXdL3BHOJjn>+fMtYt?qnH|&;G~l0k9k=4&2#B0T
z#3RvhkIKRAjW!MiYfK|43O9H?KiAcIWSRe<At5^FHi1HermVYrH<9E~`yIn!IJ@`+
z;RR*i_SvZ;E5PvDnWOr;&QX|h80<7W<|?@b$6;GLhVte7vd-SXnO~4P-NaV1xbkwf
z1x_*A`|DZ`T#I?Z@@4NBi-Gw3<1b_;wcNxXlfSU9d@i0f%oALuByv5txsyc~13RoC
z$JM)R6Q&~J1yjq3cdTx%MiT9j5~9<i#CBjUTht$VyHEWei2Mwk{2=LXhRdMMnPm<H
z6x0Ec<-_42#k{ZHdZfy{c6}HX@lRM)4sA-sus7J+2!ko(pdwPl%Lf8REkAXy+BVfw
zTH-k8(g!qI)SDg<G(nKl*uG)$mJT}y<Ni2u-K0mba!^(;NBa&_rZDe0Jbbyki;rC(
zUyKt^ZB{_v%RgM7Repd*r=_+3GVIdJ;4Vn7s8u=?q)FDXK`f@7BVBFemnrlve3Mp9
zE?EtNiR8(5dOCXR^o%Dw(e9~4NyD9j<;%cLx+y=l`)3)|fC5W%3kSCGORsf4pa%g1
zgCW&Fe-^&Uo3-O4#7a<mK4fHMuZjFx?uekpladDkBI1ejY2?Y+EvEm9nXw1a&yJ^0
z7QCm{cPnoZ5%V$WT?Xk7I0Xd<9@Jt!XtKWsR2v#vL5QalW1MD2;WGdR4)`YNFuE-b
zS#cPAqkkoK1_DANP32aaU(hX{#Vng+r0017BhTJ|9V*4Lv9QlttD{%+o200OXcPeS
z2*HT6WiXF)>>-XStxRsE7KdMyYN+48E`e<VctYJOdM`DCR<f}GQpGrl1iq2BlKM6&
zk&MUGAfsx>cP0wZHbpn?YEq?51SS+<go+ry{J^2pUU6R=sT68vTD7j^LY*KccpNT-
z6~n}grjSZN4#o=LKn1A6Hq{H3nO-0TIN;!Y0<PWvPjoT<yV(B{N&bhJKyvi(VC0$a
z-pALzKTYiPm-2MWovxQTRy}qEj-Z0aE}@D|ddW%vyg|aqf^_X~c}=YM^NPfHr<&Qz
zOyU3(oR3b4jWV3Xj)#{zyOz59LVKfv1P8CwmA}86IkM#>ra@t}{34)K;*3PYfHUsk
z9_<?HR~HlwmX=FYoOyG(u+kZuY!Nbc`Gfn!E#!$@0u?(diUqCj_4^vbjg6sEkCu2W
zC^KAfFibnZaVOr$qj^k6J^;5~sF<9d)cQr+B$&1v2L}bQIy_+-iOBm@n}eaj4VIr=
zFlDQ}ycp<oYIyrCTLXNu!*Q%R8f@YL`u10AVfs&N(bk^IK1}!KHDz(Z8p%&M37_Xx
z@liAyVnD4n6VXcy_?$qNKvR8Jt#I?MJswBh?}2UiE*`8n5K2IAejROod%IMXQ!+U~
zXXO|q?TQ~qT8b^H-*t_(^()@Xj5s4D*UDwop*e_`zwj4phe_`Bu}~qkg)=m)>I91u
zFf}-NcC+sSPtUc-4n-UUW1<Pe;dah{f>6a}K}r0MZWEecx(^@c+gX8%s;LYwOosb0
zI&!1f53zn18{KId-knP2K!g<v-&#Uevy-Ej9`#XsilEI*N$<Ay?f&F-WDfF&)mnTL
z0R%ouYSOo>cfJ1$SvatZ%pbqcG>z^yO%Trq5L%(HJ;9*Vz5(2>)Z|j1cmj{+;2w7e
zGUyUgHZiH=ca^T)1{mOU?e9(^SaCu1uZvwILQVQdN3pf+FNdc&g2dutVm?Mvmd@I8
z@iq6?6`N&2Tyw^&cu-U#KFAVHzHwzZC(YiowXbr%fNLhMvM%Y7!Qp7O*T6+}pbcJ=
zd~jK)g+vcE#e@mBY>UeQ(wlqqs6?C!GO228_6TQ;B9JUf2psGwX2PuosQe@f2=rxI
zaoX4v^f_u@+qxd75Ma`_6Kk(l&6gNKfRYWOLP+O9N8Gebgs<+0L!^>%2=lK~M1g{G
zzfma07CZ;j{kM`Y1+o9#YX4JNU%j|MCU#Bj-KD2b5e?Qx)>~?qED)2@OjrCZ75-^F
z#-P5Eyjt>xSsuZ?0&-pX!y8Xj<$BZ+c9M&R5-G1w6)oO4a&T;0U)S2>k?eGC9pu#{
z1_~iZhGqFCe`r!B)F|UAA+(Z+6rxKGH0RE1XzDR;8A~}zyM2mcOX_gVIE5jYp=>$E
zm2IYfjzmWS6oy+?^ukUB;P@!=pLli{Bo#;7$so{c5)PmauTdi`ze$V{16SeO<Y%wK
z5&vIC4{7>sm%LrK1O6jbP0;=!FdnIvd~^Ua7onM>GJ;!|D6Ps)%ohDSpqhM9JiX2;
zyG6O@0_4|Y>)%Z$=J!|W9~o-3s6xae5h##FH?w;3#|9`7B${X;Za(=Mug%!C%rKMY
zPF!<Y*-UE=&E7FM;Dkv6!ql8Loml3V=PgTT3U#Z<CB-;lfWGk7R;g)jD`@+~<t1rG
zX!a}*V+v3~EQGx}Un9E6CFE-<3uC1dZ5j6LXGNt6C+HO~-`VjwtO-k<izB6yoPHE{
z;QJ-Fy82A8ys0Z9oZRt#B&nEHQNL6igl4k9;=S3~1kcBa`9WOuGNolu;+1f(?@jjB
zZVZXf<F1l4l^8jxGX8!q$7`B&xzYotcV}8#<ka3TT}Lxz5sIO3#hgDTcE8p#JpN5E
z{Zl;&)89B;9~EV~5UzvhvD2z`xZxQi1=KO2>vVpHK|=TWVU94h{rW#qqA5YnA?f`o
zVn8{YgR#dbC*@><!{ws3{sIk2kk-QdKBDiJ6ykAd5(c&1J%aGUbW5|ylkC?kt5~WC
zC%Pe;*<4-U#7Q;xU%4H)emyvL0Wp?L+d4e_&_R2o2yzTt?g?QMv~?eB)YDW{jdw)@
zdv#01`J<JxNF|swu^LRs9!m!7m@Dh)ygrFfI=4O|9c*<P)H5Xi#8s-Cbd<GJ!Kac?
z075M6pRQD8jB7~OPl_IDky|4@A{DP+!&C*1?x6&Wsr0(zQ>AO)oz2h_{S^$T7sy1e
zSHzfY>qsDZ&=)j$N$a%rEYzd2SyGW(^>^n}aUx>+<tph&g*=S$NsziM?ZTc;DK?Nc
zSKT39>haSJjG_ohhvG{khqTHzD$EibV&-QOIWT~l<?OSkXH?7wyX~o{arl!llO^Js
zXg2gd{d1-m!@%w?aT=?>G52Vzn^nxI$M9wX9)3wCGix>bC_I8nET6_iyLb^e1;^I>
zEOfbENWJA)j`EfCm`TcmZ$wq;?q_Z3e3CrTUUx=)!`HgwYcv-DVdX5L!Q*3YnN+Rt
zFPe99aZ(JTz$|1lvO?8dCwLB1kAfITQGY?}G>+F1rwFmKUpK^gM@KZ`#&7F?f~pE0
z#>+O};4<<e@lg<iX|Pf}40{0e&HZ+v&+o!Gb$bmEWw)AP@MmhZ)L9x4AIIV&^<=on
zF;j%(X7YCaQ0~_FoRP7&bIK}tzOFNAg+_W#PRWAp`A@SrQ-F@XzWLV`a;+UjNEOe4
zb@$yO<6_T<^PbMi0WRKWUHzS2J$t8zVVo-I6kR1wQUZuo`7CkeGgmL{8(myXb?<oc
zW*4~PymV@?)ajIbh%lX?<MCM4+~VdO)6Dv9Jm@AjPpb<1y;-PWtiHl#1RY#p{}mcI
zgcVnj4v#~gAvhjFcN%NzSRA#itFS&gciI}eR`FM0{0ml+91CR45Eh~rOGYBlFlif5
z0FmDQ?wRq~RDU;P%?(IK*-+6k$<FDIP%!f~t!*&s^|pT?o99zHQ$?Xem5I&WDirxO
z^n1<k;4JU}r2XE`l$47{lUIQoh17sw{O$2pC7KBcD^jS3xbJvb2m=Gd=b9_mbn<NG
z#I0Oz{9zfh&{ndYEa`}}3ZtO57nZt9^~Nh2QmxE<>`-rsUag<mgeJq!yS*B-_I@*B
zb4WF6!tV~dGQj9**2v7+IqcxT=C${xs^RUmJ5e!JfvAa6hD=7ZOp#4s6Mj{c??9sX
zW?=h$b7}H@F5TMw%fG1^Q@`4uPy>!bxh$smFM%L5x!~^l9~wTt55lZ@%!+E)!Lkt=
z#Om6?@fO--ToH3=jyJwi@kxv%t#eG$rzh0P$vXpTs6toy?*g2Cj^_R_@E&mr@<>sO
zn`##8mQz_o$70s=C{3SD9o0}YQ7GUdF(fRiyrjN5z@Jh7Xt`&Zkz>OtAKLF2W65GU
zNm7tS`%Yas3(~hIaw7xi1rF|>1?qdLe$@H$T^PNfK#_wdX*WcTJ~&YW>Iy_w1LSz#
zAz!u$Mu(BTZabq`ed-~>Y~VO)lhzGLh}DlHigQynxvZ|Y%ttBytewRLc>({B#RaIF
zd%5)Mt;hafofh>!otE(VyiI5Q*O+l0N=962ywuMQ%y>)TF-s^`%6X<31*9*}?U-hi
zmZL2E(>m%`J7Sj7-k7EX<=<O?`+a_I<(3%W*{*nHW@OkPL{ic*=Pgzy5g@zpN)k{Z
z(M);<I*?~~?~^9QhHH}1A!{_rt5!6$FZF=YvWY4KxZSBkwkDo#yM`a8ao|KnuG%(h
z7D{I~sTs)-t{yH1haUcCMwDSxKHJY<mJp5mkmqg%OUlsrqt2ZWsqSTRFC@x!C&GmN
zc*b@D!8lqB8q-jGf4$0c%#hdQe!oz5Ly-ah&(@)@=bv_aCrWUg5diIBBuyMHVwO`7
z49cA7>a9z%vd$HZh+>{Tp;1j;DINP=m(HB!uG8zib|%v7?^vADbpt4(tW>(#=gJk$
z6_k5>`p$HVUmS{pmV`Oq<{8qWHNYPMMxsHY#H(-f42$s!nJJy1q#;Mqf+kL-y9i0|
z@x#QLI)`%~LjEMbc=JdO2ypuH;Jw;$>&<a1yu)kBscWuQkGXD}A*5Tb5l%cKi`X`7
z{ewhORni1g7gDd77t?Rdp==sMp``X9rej<Fv^C=N>053$|5QhN-QCUUm;dJ6=F@9W
z35&#ZqTr#T1f+QUWAgd>k~h0p?>jGq;+Pe&B{AqJYOw<4;?jf}0wSl21+h8GixR{s
zuMKi{*5q_@Kt*c9<o)te`nfa?p&8{AOHe+dm=}N-kEgU1d^M`7i4XvX7HGDjOjLCn
zO`WjXutOXy!(2@&HH?t7M=gg0bKD#ja#o-a`LpMj@Z$Atx4E7|#j<Lf?v0?8V6EG<
zMlJW20(tUCtT(7}mhedarb%Oc6fIg=vC>t~IGbvmMaRw&KiY*5*)uh<)GUjGo&oS<
z1vW$T(_N$pOi$+k&tDCD!1$}wy|@TcN=7KcVCSVHtae_ub>McQw!|@-OT9{4Grb<h
z8G8F<qF5M_t}(LRmQ{?8VsX#_2CBGMnV#yNdBY!aW*I{e%0Z0Xa+*(y;UAbeNbM4d
z#k$t$$guV+KdM1_(+zb>WHH>_y=`B4yL@fdF*1@m%F=2Wv@4Y$90*obY;GT*Cj)6#
z2wf}>@n3Z(W%(Z4M7shoU~<fENUo-u?P|hEc*)h;r@XIF0$gz`R&{FePiW<%`%UP!
zM3$+)R&xv{k_=<W?2pdr9LW<n<!n~6#7V$XtD!pAk3GO;zhz&ihIzTt!yCV~ze<|y
zKRiK_azU2ZjzMfB*=drnX}-2=9?06t(X21VUg({-W9-4uYl%?A)CfV<@OFoRmHtuh
zly$?jlhujpZvy5Z>|hF~b8k8nPTii5%jD4*hpMRU)T(L;MX0JxJqN9L4uJ~?FLy1i
z=2uo&_XhD&#r#DzlnyQh2rKygt|?&4H)8m)Sv?4Wm_w`9uRyR3S|y^YrhAT4;en1L
zx72yxiY}KWMvYr{wOT%iw&LdXAYqm(Sc@G-ri2O!<FMc?X1(4#Erx0<d#xR}2W3mG
zZHcGeR33Fi{-qiQ-3D+=2uLaRqQm*gV$m%S)F>EIx`D%_;3b1<pR1)JcQ~z*?r6U(
zRI!zUQS=K!@ekp3g)uNK3B%44a=9p4IJqP$g^6~j7D9e4iad@a$Ls+^Y}VTt=fL-!
z_c$Hn24CL`y3DniOGpSzy71lk{8?h-AT9mQ2E!e|9+XRt<ahq_Li_C_f~1Mfv+|H4
z=nY*#?yRXyS}PY#F`|3L(7M+)iW><Ed^DdP>07{Q?90cFc8<zuF;k$K21^BGKBynA
z@F^S&rry()c-!aQ><#`Ew(QSTd8yr(*G_A{EPm*vg*$~gM*^b2NK~l304Ef<qzmiX
z#+`fD(BsJj<zY;{8Vwsr6ETV3S^QIfAeT0EEGOa=dH^mC9gSMbqjM2()Vc+3z8Jql
zq?1EYIgrQBIslPi`BLk0Ow9<0_@ZPPw<c4>xbLIrdP2M9)BPgHFM8&ch$Pe|WU6UQ
zyHr&y<oc2lE|&%CNq>ag8KmRT%Tz$D98IUkq(?_a>C9W_4pU|IVu@M_MhtycRbyZD
z@MqythZKby=%9UcuABjfNRtGCh#9!mg0(m1oyim^5(<QDv|Hy?>PVN)4F~>ba(?%e
zALhCJWW)OK2glCuND9e}mR!oFR$x+j3jp96I=opmzZ>Q#$?-xdr>u0;rj|!@h@<)7
zYIvH1n*@Uc%fa#?^muC8V=&9OlUXdir6U)&uluK`oP;ljBLjCB1}1#&RKGL@%j+X=
zST}#(_KjL4GIohmEtRl$$ITR6<Z}LT=sN8<dJV(MaBL({#u?;^6hhXceXF~ta-%yI
zj+08S5MR}<*GhTK-L2Hw;$87!eknhK|LnTTUf*UVs2g3DOTiGk-^6A?aBs4$TQ?~A
zxc$q(HkW*zXlrQww*Klx`wi`1?8U!?(HzfH6Ulg|>!dfO4Q<kn2Lj^B79*B8_p|Uk
zWPtvsqf-6HJ3dd3I;oxCHd&Dl;wf^8ob+2$N#iM4d5xQzWuO4@miVd|gKDFWo^g`7
z7N;wmv}!|BGicI=gKV|4Lm=1KjL=OFRiiHUY>RXerFl_VBa+J^m4Xad0I?XIgd;_Z
z%!(z`9lrk(tqA|M?>++%<5@M|sJ`mB@g1>$Rh`#ZNjKq05Z#2&b->ne(~#)S6m<Rj
zn8%vgY69JutraIQp0jS6u#s2$DE^C2%yu=2Hb9k^v7fw{O0<9u|IfYS#b5dAfAN|;
zOVPLrEs%+YcmC;PBu6bK^6ZXHl+=qvnBHGY%srfwNQ~~W==|M?IfTQ~wXelf>5wS!
zSV7k6*R|M;2sGUx%dMRQAqw^vhetK%G4{SGS5<=6X4kgnD0wPHQpz2x=96!5%G0%c
zKY!343PqFql$yAl3R^!|1jC8oeDLQ~X5v1%^ZQ`v8N3h8Pd|P!M@UveGbl$aCb(|a
zKEjtf;K2i^=AdLT8ClW2L>-$x8;QK>gwV=U)#9RJXtYLQiG9d370W#DXq?q9W#A&~
zH$Us<kBZ0IAO$+8k)KHNS7m76b@2T+UGprACW|r#s3ev*lxp9t7gqXh(Ih)4qFEbb
zYcFs61Ask)M`tR6o`J1&xuSd};m|9m3wpB?Km_JlswfqWAPF+-H{S#{{|g156Tvm=
zG4*sGU{Z`a4j7A$TVAY5*XpC$%T_`)^dowtFSBTn02f5YN_NFiTieQ+6L%t73AC-`
zAuZT5Rp#P(s<ThS&F9#6+OB7^mI6I9VM2vm<4~Tggf)1-K&@n2so%Lr52k;Ck~Wf&
zNUrjn2op~2OqmJaKL@_mC~dg1QlgRqO)U<)_pj^o$7n>8>(RPb;(w_eecvC9rL*bH
zx|IXN8JYd|b<modi8`FgK9VL6z#vDo@LDPdV`)_Ans>tDCh2&FXK<!?fdk|G3*s^m
zzfFIs`dBf>dym`Mvx}zZEmqtkCl$-nHq>Vpl!C^9icv|1kjyZ42~t7wPTnWHCMqU=
zUirZcDDtQPl<X6%=J)n9Z?J}&^5LKJd75OwArNR?b+8VjWX5sJp3zEF=2@%wneHwA
z-e~N^{iAiXXIlFY9lg^r;YgI7{kU^qJmv2152WsG?ZuikuKaIc%fe)UJjI{fNfYf$
zSGw{MJb@kcPUWX<;r^w6DP7O%uW=!ks%~-t0zXgRRGjeG=^egjb6h4hvF_e=l};%N
zMWDHB_EJp;?8kyL07G03ba>irR16YNpf!Wre{87x+S8o-hc;wnta^-4IR`R`yf^Xd
zaI9wf0CTL94W7VD+32S6#Whtc);M}+9HVBYlGmE26kS{_p`^09#%bpc@2-$SOgmq#
zB1v!L&N8lH9A8!Yx^N0+taKpI<mRM=ivVl$CGod!$UlaBWFb_4d!1s=<~C8(JGipd
znXP@}nTi-oADwQbO?`4j<gPU~-(E*paTE5ZjdMS)0pDO>O_g?3E!C1Kt6fj)>Hm?-
z$xKaG^oFx&fykDxArt3Ze1yI`?ehi4w*^wIHo$<S&!U7y__kec39Dgt4IbM<wDt6~
zF;5x`#-Njy^J|O2zGS_lxrN1*qA6=*#hwWDq1)qkzM*G6H-g8P-sVyVqb}VJDnAc1
z-+TNF2(xKkVrFIwO4k48kkG_uSkm6EkV8$5dZm^0p+bz=ESUS7+b!`N@00u6^ZZx8
zE?>MU)rv~M^%$;aZ;x(%H;?}8w<)5Gcdt;p+(u)serWIxI|JrBs2OJ$A6p1JmI{a!
ziR16?-rkD*a+W1>+PoxU?YtuXvwikQX<FQ@DL}5()pHu5f<<6_%F~ch+QeZnnPwqr
zTcMWFfD};k7lKK<?THo`&1N4oZj_ehs3&&XZlzmj^$Dt~a5u&YlMWOm&!$vKx^)ti
zXiKZk7cmBCH2SN709jvMzVfzs-Qt`K3bkjG9nOZg;5#30Gx%)~Jb{fKHVGD%=P4EF
z8s<L9$^nu6@6Y})W>>w<{rx*kTlcffz)P6O9W~vl!hu_R<!W1O13m+dCkI-^{;Q#H
z*Nals%XhDv?bb}74P!DQ=gb!uETU)aHu;S|&SC`;c6QUE*3QT~-njFjQJg%13^BHK
zjpOG#9y69T#6XkQ+z{V8?glOl?$YPJHfLiqq<Q|tZCFXn8Y!z)FQBlAVvO_BTq7z+
z`+BDhn8iG&2L5lSM82_~XrY~{j&4+jMgU|FU{wM+pQ4*zIn5IYa=M_^V%I7N7DdO<
zo$l~T5>go59u-wo!zW8)>y)cvthEdVJ8NtKt(0GvPKW<?_TLaGT2h+VUbAqw3ZE+8
zM3GlRm%t3I>IwJ1t1#O2|I<{XQZ?}71KzV7JTs&I9rnGRU1Z#)BBU|3DF*Sqo!!NC
zamBLk675{*+V9aWORHvynz;@2+y}y;I5w){f%JA^>r+jnu<po+D?@O)X-$ub=ZQCm
z2L<XP`iB^)!<cKo^PL?0a5<~P0(u8HHVOyRkn+XB#1lp5_0?KYL%Rei%V`{8E1%5_
zp+wGJl233)Qb><LjB9Mc<cq$e{}jvfE&SLEVJR!I3d0s`cqpIDpB^4by)W^D>^q}t
zaWU|t#5RM;#jyc$Zr)M9cF|^cG39j&@{Xm+f+H)=zKsmtv6L*yXr~*ITLqNnJd(?3
zX|=d+(P-IM6*wzZ5tCb#oAruTzih2Tk;i51fy^Ac$N=hs)_E5RIfWVOq~q~ESJJ=V
zAH^W&LuKo{qm)aG-$Tv4{XWF+emdW;%W6=(19%mJi^G7|%aGL|djx`;R*ALAr98Lf
z@2Se*r5OhL(!&nthNfNIy0R*q>HWtqakSHQwvpd>V12}V&uUbpb`pRpSRT#5WJ$~6
z=Ezs)Br=7|K3H(EE&8_}{Ouo>f?xF6m}q*gehUJaR}I#czP7qL{}?GKxK64SSce5g
zm6r!6nl}%@mk82DoZpD41k^^VAgXB&$2_a`GbbN$hwz?r=zv+Aa4c6K$y|-|@w_T8
z-y9JpnswDwt7T2{ysm<wbviX3l6#eb8AD~aqI?9IMf0B1#HpQrK99YjM>aiBB_L1j
zR!*5zB$NXcD@d)F#!jkyK9DoR>8uf%oF)&fCExhWhO>9M1d{{C&dvLu?uwF8<nuLD
zbWdRMR{Tgk)4e$FeAoG-dG)&-GjzeKF`DJpieHQ+Z(qxo`=*^3UPZ%+FW$ENP12Bl
z-|4!4x61OtRBcTUjTse^AD)&5xej>2o`7m~b4>uHQ3E7{TEw02i{G<vA?^UM<wBz0
z_?*l+O2y))MiH!$1E~(1h|a&8Hd0T#9Ev3~v9ZY!dMB5ZRlp7yJ-2iK4<<lmOhom>
zOh(U~KThGkF9jj-JGP!ifC4`>$%Cb=p`MT9?>%pu(r(Ywy2MIn7T%m<9P?2#*Bn(J
zxtv9l2aA>=P?(u(TR0vNI+DJ7f*qeNbgWdxFqj>}bra=0`NZpM9fIj{JwgXU3T@rp
z2xFx!s3hNOtsYnT`hUFX-@1Wc`TgY%?YhZ$?q|75=t=P92mA<$R2a$Cp1U@)jR0rv
z*zrP!7yFd*W3N?tFQZ5=YgY4`TO)8HK|!78oC^a4w=nxBTB+3YyUdytN4u)|)#KWw
zqhED#zKMG|kL-@c;)Sya2czoJdDj(|l`IoPvD2<Lbe;&}qmO&1D-Z3P^pRQ89nyrs
z+abFHX-4b2hb)~BDK$w}QVZh?{8@Dnvm1lbVk87{3e9j0RepkZj3rjP;T4+a-@3Ow
zb*4BYdY0s#uY_Y=<p@hCl8B`2(*@C|rpEpZcqP>L(97Ak5l+d*p2R@~Yjw{CU$5R6
z()iZ=H*^p!RHm{JCDiCC$Kga`5Km4Hc2u;m%%OU<8dgOwx*l9i?~YDEnvaYofi`D{
zZEE9gk8S9Y6S9HDsw5lg+bimHo6QH3uI*?OOT5<McZx2}xuq9*obiudXqAf16_PL1
zMAm_oI;-#!yWo(9dn><YuWLC|MR1S9Pdn8L8sSRCw?yyzWY{&qRP*?0cJ`y8*MVjw
zE0bwfHhm^<%gCNUIXS>l3?PQeL={hQC5>xfU2V|WyEC_D-ZImpV=%a<q;q9TyHYS_
zt|{SP+qZb9<mTQMok#%~nJO&xI1ieb{`{G04A{uKP8(JlQQ<MsyH+vDACB`EN&Wku
zdjQu{EYGx;oVm(kZ?xs1|B(m<QC9Bahoj}0U__Fae_wVJiZ&>4a@&QqOU~!fqF?XN
zX9lz^9cu^4SVCJjkKfBgtVVu8E<#01M4XFc?E7x>3kN^T-#vmw?n6XixA(72JT1x_
zBk4%^r<yo5osfLlu)5sA40Fe;#VELVQUI|kt)1V$r~n%+ju*s&?U~=sM~{w=T@=Pv
zQ|FxAu}A?Fj^ZY&kbDPO6vV+#JnVeJ5{0wk05O5r4PX6Um;ozi0X-{fHfBUIGRrNF
z*;g^Hyu7>&nIK?lS}X^ba2zk1*=467oUq}IcUN=4-YUs@$6t{U_RM$yX(Gp8ZflU>
zl{tMI7AC<+k`#OKRcmXz;hmG?M~vpo<V-9{Bu>P-#_!Q~xGNq3k;8QXo68ZN^*!m7
z>6<;T-VRy|t!Enj0lvB4e^=}Eh$YY1czj;W<{J~|w{3RWCd{hu;p;DvUM5x^Z$Ts1
zT77<5_o8!|+;1Pv!c)45c*}uQ@T<60dtdRGJr`d`H?*c>MCR=|x3XfgKU=eg;;PJW
z#C1R&YE9P@aNrfNfSnT4j5vliju|$!u&_Ho&^19Pax3>`Hr=ZuKqHY`X<=nW3>suh
zzj%TV5UZK%BUJjlZJt4)#D%1)^^?$yKGM{PsM6KCiXk$vU^0pG>i^X-v-)4~d6+up
z(1%9^mNlk^FtKRRVN;?p)WWCUdL^fn<`|zat%+sUQ@O5-4m$jKp?q=7WmS&y8DRi2
zUN*p^^*M5D3v0nHPUf}n&=tnodJ~_2H}cKeL{l@H!}c*hP<VTHoFoh4mw12HmD_ZM
z#mn*zgcILe&|7A!XwAGRkVVg8CB}aq{H6c_2CQ^;xkp|c#~-_hlB6_;T4gJi>bH1g
z(~10qq<laP>k$AcE7Sg2toO1!2L#67!$dxT6NQu(POhhP*_EAFJ@>?QN`y&gY#VQw
z3<kF-*t+g0k4TPC#Ik2fPfHhU&1mbqpky}FR-R8pj3zqr9R&7lGM@pByxSQxY|QXv
z^r|yWq5tDG|F_$C<@Y(Cj^SImixC;nv$vB8V^AlU-XH`JP=`hq#_ao(BC^W$u13E<
zBCvdmCFf~4Da*r{h|Y-}|07Gj+CIJM2d~dBh75B2w<updO-(d-oWCrwDrxDc8#wRf
z=3_Hlf8w%2H|x1pj_YxVhNC)}swz10wN8HU<x?KtzKzgxpFmo}MMTNp+%CuE{=kxo
zzb6|inALUmwpzbkJT->BJilSq<aU2*nheULwW#Ka5N68O&zACx_CbtN5T1eFULjR&
zGppvlyE`)7aatSB65VnNlzs7rjt-@UsojkiZ)D2DXG&qthnL7jHu4}8YMb8cUB&>}
z>vvq<bSX~sbz)F;EmWfSIfkoYNDTi)6F)1WAKZtJx*phCx;D$0VF}vEyprMVkD8Ks
zAgpudLf+Zk@9%sA-QAiYT^c(PdoXpVw007?6njQ8hVoH8xNUaR4|$HQ@_aO{bL7^!
zw8iIuQ4Vq3E9LeY%(vRwKXP%C9~JBd{k(91pMAorOOo8ytl+YMj5a#xYkA(0-eP~&
z<~(K@Fd=jSekJlip2oUt>D3m)g-m>fc9YF7(}BZzN>JJP@G`bpB*Ko(bkx!Ip2K&0
zWm}1%8=#<S4TL(Dkh@Z?dTnV=n~SE#Sg+d5d}wmOiY!<^-x>QpF`j@a^BtzCTA$b6
zK8nHCIc&_HH?u&S{4*){;kUWN%F5MQEA)oZ?%MChk9x-IKSlF@sF|?LLZaoio~1vB
zQAHKa9J=b@lFP+Y8V*J|1oJ{fB<nYN8&pQWLGOg!MNIaeqhg2w8@R<Rm1+Wv>BH&A
zhU5GEzT}RXt(Y%*zJBPm1TQaKwz9+k`g50wt+4iC+v@GOS)2%La}zT~MHL;nK!u{H
z5YDVnY|+A3)Xd!;!b}2WEC-1r1+oLtky@N~h)qDaQWOmh>IgzVyCqOe!dWr<qU@*t
z#Q019f@1_1z74?Ewn7L4BjuYDWfIgsa9cS_Z&g}8%d;fU(@)BwF73#xt(UydWI+AW
z3eC7PYo~KYjo(S;!%EkmR~|>>b7j(j<5v}xtDXJEwE=M7N`=~LW)?Q46l&eH(Gog~
zDR^?n$5YMl8_NIY?#_SckS4fdvKegX77k9Az76>tTNWO{Q>CF(MTwHZgCaG0+%oi_
zC~G@r>8lk~&whBSmTi-;gNie0Bh5^`@qjdPrRVdc5oNU6+V?93_uKEFIqNxhxfw(&
z<t#IcTL6KPXV6I#OOE$p-*W?~qpNS}4JG*qT%4IvoPbfHE_XU-Lf@aQGfO&3f&kH9
zAzL7%Xh*QrV78#1&|^<rd}*kcFztKSX$odmR_BvZ*Tz|W7*5W7ht%~0?-rB2>YvHd
zc&?#%-tOg@2at@-m0nZ>aUsw98lw`CL#M~*fs|~Y;2NKkiifQ~SGf-7+cvL#eu@ps
zDc2aSLV|h-1E@79oEx02t<VEOS$z<1ciRXK!mY`W9vx8fP)vtBksuM@xnci*5xvF@
zMyl)N!}%BIsJ#m`u|E59;o_A|`vs!>FT@KJVc$K%EV!ssL;H1-K`8PjO*BU@BLY8_
z;i!VqwMYOo;&O*GEOx7hKQCR7xv&-l+&;TNW3r)q2d6F>?fetB&NmH?R#8ju?SE}c
z#Hr?HSUQ_UmF9ZJN_&5{_u10>c7A?yaD|3qB3g;`Y=+LC{h+eVuO>Uqy>-4*@#E;F
z-)2&7Ax<+9mSCeUp0qgE@wFpby{cGhbi8YAYkn`N`3ddtp7`pAl^lUG2t-cHpNSTh
zo=l}jL4<i%S-n~?vp}t+*7#Mw-7841*}kD+#*CZTB$~nO;A`Xej|-l4C>?X)sne@o
zh9t(Dr_)ut8E=cpGov}jzcLi^MUdwqGXzB(ee=81y&&6^ZUz;^|Fyo(;88S<N}}%D
zJYjiNHIr;AFqURjUR4OasN9Hyci7!DsR<Zc!z}a>alBT%$kcB2*wMepqq?AlmgVY2
z?k;>_PWx!tMEHP%)FmUwfELL3?dkl8nT)^Js_Q4$v!ESkYA~FU^y;GH{`KyrpuKTr
z;ess};iZmV*Ek$bs|NTyPVA_WKg`2tw`^=;63}{!Hb|`Xp06xa>DhBXQY-RoEtjgo
zlK(NO^{f>@7OA~|8L77Zp>jCqZ~;+NX43~DzvmVE=Jv?gzJisV{qeN@%<r`855?Q-
zx_3^$vZDyqe?DAHT<jOdIo-pEq3)+DU{qbG*W+ur&L+fCNO|}?+O9^1bm)eif$7t!
zW@v}`*qhal_s0K?pUDeZ`>+x5z;N_ILE&njGpW})x}r)fd9I<8#CRmBg2I%7v;;i$
zes9ufQO0SnLWqhI_apnM84fJN7JG#ptCZb8MwMrvsR_Kj>@iDcE;f;Qcp)Rv>9L2V
zP<m?RurmGTG(LLpIUoTfEE?KoCve6;`}Pk?ofkW77wfq1oVG7*ipKcce1fmPlfLs?
z527z_aXTS~LZRHmuBVl&rY0s4%|`x241@h9%4Npfgy(k^_CLR?zuAM`VLeu9n%9qH
z9CAwU4I^n3#<~W2EK~XKC_u9NGD>@BwR@*EKL7`V5Z~JyTKGa^TFd$0-YwH*WQ&X*
zA2YLkyA*1!F`^@K_1rC|BYY`SxzZX0k9GXq&{ij%YGrNqZ0g-`_?|NMJ|jCvLJCnU
zFa@iWZk#xKsrha#F6(bK_0>Pdp+NwkTULgGv=_k3y1YIg64gJ^g#9&?${2s>+oT3C
zhC)hNKWZyR$u?(NSBJcEFx^iQ{)Y0?;SaDXX%dN2)dDpDPoSc<qppV`<tM>wgAF5t
zMbjTyQjx&S6{SeiPcW{*!mL}}lqBbiN<*JaD8U+abOkp-nvRfHFnzH5s^3e{^W2T|
z=C;k^iOue_cOv6MLp?iBc}|^eP0ht&g?(+kLuh3Mk=xw-w`+koMX|HH@Xc*iMlo4(
zu@5gH-#vz5e1B-hYh(5JV;qPG!?-^C*{8vF(eeSmf5?dRj6N{8DjnEXJDr>_Eq6Kw
z<=#HLb3R?>@4TwoL!2B0p<xfjW^bPFDA>AxCB&4WCo--orUwGHI|bb)0@u_7J7RyF
zP5hVP3i$$g_lI0E3Iz)mqH@A3Cp({e&E2>+GOzg~aZ)WZxws65KBo`M>8sFGb5F^M
zr|v}GF*Ir661@jtD7Cc*sUgO=Rgl&)p*#;80AJ2hTeY++nKkz1dYMPquPyKJRg1db
za?dsM?PCdm2dBqyzBa$tA?8i5Zki0REEU`yoXz~E`;kLfc;NeY?s3bcRTx&2R*H1T
zD1@lcJjBnx+kxF&NLfc)K&t3vhn~Ho*%c=XUc9}IzMkoMqpjjTrZ_a3RrXF;dMOjU
zc5i|2e<vbd0k9(L{ZRia{`4CEjo<ZFmarf2J@s;{Z&<UTcR=>Io^9ny|A7^Hu%vRL
znoJ=@-xNTs-#YmxY#>U%(<_GVwSQuq|8FXCas~BkTX&PQSi;u1kI<8Q$AXLhoRRrQ
z{Zu6|jwtS?=t*^+0Ui;qtnM7nbq+1$-lAJ7_4>G{X@j>_CD%tgIRimNOq;_L5t9+p
z(%7DPfWq`&FxDwslG4%YaYs@!nntS<LigrLX;3M{8<DBq2mK`fhiT1ouaMS*3tLyd
zc_O#gMDVh_TLzJfPBay<YE0{g!ec>oRhK!guW+rANF*~<X+|(wzIY%xt@I>&YFSrL
zO9Skr$`uE%wm|$#wM^k+|2`%ajZ`8CEs}D=5gw0+epLl6J0VFA;rcWnpwy(h(dlfJ
zo^9uRJ1Q7GjlE*{YBEd6Gb@QMjw<*cnsj8EG<b6{Mo;L^14d<K7I`eA;=2h@h`Y9)
z-p*hZ33nz-=oW`yHkM9meo;DoKXrq1@7uY%|2ZsS`9n^D8BYSEq|$W|J&L>b*Pw-E
z+;`^>`LQf$+lYBQZMI%P8ed#^V%3+8gDX&e?CEWbeZs^X#own6fJka939yDAu{63i
zS`rYQxkME83-nE)JuGjHl{MR+@94F2A2l$4EnOhHS|GbzV07Y*Ox7;?aVKm`Zc{l_
z`JK)2_*loy4Og59G(?dP9C4PJUh2K#lYJh-ipH_CB+3AKely7DAZ%!yEYd=+D<KCE
zgKZ+Kv_^G4>f-7uaD~(JYGGqoIVzK96)dEiwyo^O%o$3eVV8ZXI{9emf28z0tz!Z{
z?|ytWWjC=NEiETQgC)8b{X(!mfBPZBISb-N3MFfsJU%_GpE<hd#b9%|J)iAZ4N00}
zaPR7T^WTM``zQ1kJ}AJ9v(-%*kJDQ>(0PG1Q{^0-PxX!~8DAUSu_IC?l3WE~CMRu{
z3?`>BDHi7>#i5U2;)aAZpfRTxh-%B|SE=-Kv~pm&bGm)?B1Fm6$teifBH@ECa3``?
zI(Rvv6NR`8Nx0ql-fe;NdGhNn-M$ijds{Tj;|y3-OO%Ps+~TMRO717*++vrX)Jl{o
z_;m0qTqYG9$4qqTk!E~IC=Q^+GsTd|=UK2;w5C=icMv3LDR`yf0H0o4*%&zUO@pSb
zQUjpDv%HLz8<C$}l50ELq=mmUd@GO}qN>GbK=Dkd>vRnhVj8{({5DB#Gic3Sv~<+4
zax{JRYI5<|`Ic|z(c$kolV_*waM{vP#wNYlPw!VuDD)%cDK^QYQ+*U}Sz7=HLPb@T
zkGvq&@Mx;!z&z}&<Zg(9A$XWwoWh~xI7%JNk4NE<d@)H}MY~8GDP)$s{iC$Btp9Ty
zX;mH0hfZJb@3W5L_TIr-iMk!I)-4%qRnY|V;6Y<&3AT~RR+^^IsZP|b>oADbK4}0*
zgGy=zUx<(FqF*tlxUB@cB%QL5Us#58AlO8<3@NxtmVxk9Sie{>2!E<Q=G18DM{M9T
z;u%>`FKxMmGaVn>;b?M?F>!sL_`ad<Dg6BB0qdI~3NA^3%RsxTkLMHqS96mv)9CEv
zoc2fBU=i@wmiv=?hX023oQEF*5b{A<X=dCGD6&N_%ii^fD-^i5K@8TJWpxB_ToY9g
zIIGDS5lwi=D}%KW1JsO}brMv|`(>#)F^Q}sqDK;x^Bur2kTjygvmsWvUK?hd6#L6V
zk(|ns$=9v?oa8-k>oSwW3WZ*~Zza%COcXQ}d?nDd;3=*t2IwnMA}T>mm@npkZRVsz
zPb%WOMk>-uY8r;_$va4C^F#0Z*OKx<O^v*6Vu!#Fgb6!Tn_k)M1pDMPo_iJ^)@UZg
z+)z#BthTr-v2IH%QPMMZr@1nX4`44z3g!$~uxG3r0=}6^y5$S1+qY{!z5hE%p3##U
zDtu(blrER0TYhZQ7UMivANr$6C&v{ZU6yG?DVESCSW8ljEL%YXj2%j$42Wj^1cd7r
zB|?rSJ(cqqAr!T`W=#}8M#&~xRdO*sMzI*LV(BUz2Y62q-J30vGu!%EE*H0U%nNX2
z@QS%4A5lKOmG)ALG`aY$W{Ag+ttZixm72o%cW-OEi4F)?4@4_%_&z_+EIo@hRhA^!
zp*u}P67(pPfKeh(oGh@0dG}zp<`*lkmb^fCsC~z8oMv~okr)eZ<%L$)hG5>h!L`z_
z5+eBj$nZb^U2k4JS7{<wT{$GPn-@>w{J(h}LbdTx>d3P`P^5@Ll?bI&tiKya$75|G
z1kv=WL=Qp8f#6xopp^|;RWJu}3_$WFgXDqb$2k>xB%<}k5<*NsiYP%|BkBgAxIhQW
z^<ACOt$*)L`tIN&*0MWhVEE_o7QZVrL50%W$=nAnp94K3nm(tJRr9^MC)A)S)&ib^
z!l9CgyG&~g3<YONJh8$lQwSjr3icqOICg%KcqS<}dT_i36JDg_AcSb9k{fhOD0oP|
z*bps^ZvX8U@Bi1<c}K(9b$?$s1W^Wq5xo!5qD8bsFJbf&y+s#9C%Wjp_fCl3i9Y)1
zo#;IXLUixszVGK<Px-Aivu4(;>z_H-K4)KN@9+0>76rtT5Nf39cg-Ok$^R>?c_+}J
z@T~<yf-VpZc~UEZIYpHFt>-pxqIswyx81we?&?CaGsmzh3~NZ_K8B=ni%DUFfJwCk
zG(D|o02)Rsxyv%kN3A^HCMMvAW9&Rb%^f){7;vGhrpvBZ=+{7!Zf#T4Wh%qMiD6Em
zUQio&fiE~io6(ZJen$Woam!M&yISP$2-7ZC75&!xnW$EQEMyp})1OqT{<>TVkSIYh
z%E^+VUCslFPO;^4K=ewm9P9moaUSim_EqA1V6GRNmMnuiN7mDa#fWmggN3&fioU|L
zj@va~wf~Y<Bx7E0I%VJSsu@cd#iAgrCN!2~3g-i_qg9_+#A;MKdS*U+B#kEy6;nyg
z4wegziLzp6EYkp35vzfX5(5SDxOm>RcqDd{2PhL-59Ky8OKMfB3P@|&L%%f6uPho@
zwFV_9+s>BMr*HpU>CO%8793@!AZ+rtO5!BOwsPe-ukScrt@Yew-ziGRv0tTGozF|c
zO)bUV+0}(qLN+7ll(~#RCW<n#>SC<e@pOTmI<Hl>Ze^qDt#B)zMAV0hK|cfYk6vU9
zFD$UPN+)l)HKN>ohXzyqACe)G?0Zl+l^Zr$Ff%5$U$5iw5A0l}!4U77OuWYD26@_F
zCf}_}kx8q4E>&Z7o}Jb5-Q~`YR8+`Hwin|1_zpkry|hJZI1sxK7&T~#!$ROkTi#nB
zDOLDtba_%mA&ztTd(Hr%zg+LH5n*wBnn1cSnCW%Ogqzs+L!yWqNNeCh<Q%B|!MkDZ
z#Nj+L`TbJD_CDkJ`8lGo8UYX81nq*;(_gloE~*z<=={2A4HO5U)XfTL;w=)J(UD2z
z2l`Zw88%9n<T6S)L^-O!J#O3mZrA>M9m@InB|Uxf+t>r|lIGW#6+P6_&;-MVuM>TY
z+}LF0E3^)2QWatHJU6bhjw)gz;C1IHhbSW*<xyOt_9!(%GUh<+621eQWj!YOTG{6M
zZlUlPv+|c^>areKlCcEp6eG_m3tQWjS{7*1GIOD&1PZcoj#Td{hqMYD3KK<#1{K3V
zmn8)Rk%^4&>D|_TxF_u_3ZR&$D@JhcLGo*>2X{P@Q%f0&c8yFt#Z1i6tsyPrDT4!e
zH1u$CNR%@#?ZS5k<^$fOq6wY6*p6c2wN-}y&N64yvgi}li0x157#%qmR46Snv@ek)
zF_l%S;1@z&s+S=hj6#OeAolde5IZ3PfDg<^)~*bfCA3jd1`7JchbmWpf4+x97^tRY
zy8qP$g=NSpDF^zRw(4tVAI(uAomxPOu`6aWJ(|j!!U@_aYXQt5zX>u)KKgy8=|jau
zRa^L+&UAA0X^Tx&+NT+^AUKyqo{Q4t$<$YM_n%$FEde4>sS-}=G466Mt#yYy;S?>~
zYFJ)^1Og8ri`D<yd6G$)Kmwge@~Yx^arO4U?;dA9UKO+ofwF-J--@$-UEBK1R}n-m
zWjzV!kYi!3Ai?m%rp>@nQAyM>3A1Wf)Upo1r*B=_TA>`^w$Wb;Ln)^95X#FI3}jBH
zL(#Gx8cUALf%;LS<yHmtXv@}d$}^O4e?K<t#}Wsrre`4&8-fAE(9+%!$S9A?4`A|%
zP5k9c1;zS>68PB$=%XiUgwuc;HGrjjb6}sit}Y@jP_90VgI7`oZlEgX=LBHs6htWn
zIeX%GywWoYjej?}^qc?_uW-7cW$)+b{}6BlTz7R>!j0&AkI15YH%!XkwJ8a|98EtA
z3sx#EN-J3qQ)2~g2A1yc6Q<#tjVuDUf*3#=AlF&hk3!V+Wea%#bUz?(8G&RLa-^q_
zer<Lyb9YU21gF`fbzPTH=?I*<>@{L{U(Z>jW6R0YC{_TRVTO_Ul}jxz?AKN|_l%r@
zkIV!1iFVx(^Ny_C_ADeX<s(IKkBCaWtcge3Oa=R!oEQni2&uuE2<@;}{N<fG@+Clb
zBbg8rXSQM^+)TtuR{pX^m~HNP?}-mlI_lp7+qJ)y2Y)P61FB4nuWbu)O2;@HH;Bd@
ze|;`?EbBXfh0>-J`)r7FGZA!Cvk@?35km0j7-Yl?H7u;wnZ?y;_U%kC-knW^zMQio
z<SrWEos=uUrUoF0J$dK?DzKe*)cA?`1S~Nqn_$*5<w|}oNhKh&&XG3S_Msv(ONQCe
ztU7#58bnO<p=*=MJ<Z^1_))Ny>W043=BR3gVw_?J?0Z1u42dXb2%`v8!6k+P3LR*X
zh}R(k^g&Jl73v7%=~2!_f}xbspxfs9S-RmD)G?v0k8iGHZll}v#XXsjxVmb8U#<L)
zQ_%AOjS$#Gy|(AQo6_S)shVwD(<ZD~@vg7%RY8ei25SEA=d(0j@5{MzfTPXYH@fLt
zE6=9E2w!a=_Zw33LkKHQtWl^c5?&!{1m_k?Mm#eZ4OvM$g53c|2qSjR3TRNv$L8Qb
zU@9okezhzKB!AeLCWyO*^L&^69JKi@Bx~#HOyyzL_tIj&<73eIm`~jBZ*O2S#Uio%
z4~XBhyh$2*I-AL$Qw|-OKnH<*vf`w5{0*_(JR~~s{;`ntTqR8gL!2@$ia&F~gdG>p
zOZj#UIz`x)_K=&fc?|uJ%Nxsfe$uaP$E%9}l`BQ7REi7GERbZ|7YkHOdEc%`J8S`B
zM`T>>*TWKY>Y+Mavq{wlpBcUs^dc#dYJDkfEGGNNH?irnFwDqi*<h6e1mMg0B?3Jn
zl!zq>IUBfANtn62!*WT)+LA*{QOZYEN~LB{{Hdu2RnH)~Qww<uWvbTvsXlu&-w`a4
zkFE^_63Xei>%E(AolbWc?2eq>Z!zcGcQ4pCJkqUrHj2Of<aLj6xSnM<Kio}uhbPbi
zc4#{%F(s8HlyogfvBFfd;u1`nsNlzvVxjW-;fT=CYK#2J-4*aW20xLP=hbD;W2PU2
z*6=b6LtEN8MBT12y^gF6c3W3Hv(ouB2hYG3D<_Lb|7i>|^l`|y<F3*q2VhIb6yRV`
zjHsbv$4ZDW68ej=z5yk8zps;|2(QJk7_yt51UaYCbjMU_h(}BFmLV_YBdd-ASb%~4
z-N2NIPv0H<tkchz6}|vR8TeXE$hpu4Ie5y3Ux%ZTETOBLmGWAri2EaSSn{rR9j%H<
zk?-qCO4M~Tyw;2sdHPfI?M!T<LeqpV*PoulNgp+LOp!P7=w4E1rNN}|U$B<wIb7Tb
zJ+~{A0S-%(-<4uQ_*{SB5eFztBoS;<8_S2xr!x3F<Wm)SUlLf2GS-_%N_ip{S@A+0
zIsi&ulT_`y7RzWux4m`gYtg&AY;z~4aC@Kg^t*iDtEbmHjo<nVd0`J6%lH4Cu(%I$
zery`o!30r<XaevHQ6O<m(EeZGuj3J|-o0*b>R$zgOC)^U+5|_SBWtjj0^=g3)$&sG
z3V4mXf?arfWS8|Uwt86k`Qbn1ehUl^sYD642iAuj$<JvCy9avY`yWSN!{k?Ye>kLv
z?0y}$(j;(PAkbk6BJ`iI42=sDLk~n?G&Hq^MZ$5OAOI?qBde}iZstj0!-C#Bz%T9W
z56cEun{~y9<&}1Xvxi>C79#Lb3JObGxBD4a5GaTuTINZgonqB7w|?74On#@ZtS7P^
z4`y49-=r#LP2Kka;o;I+T3R~;v2+S!3i&8~XJ-UO#pWukNLb=%!_rY_PQDv&e_oII
zCfR#$Qrsmzu2-Ioo_M(uj)nMUfA#$L0>1CbAN-L;MW)vbsMrzvpjI^^u!vjtbwMkd
zV?=3hFFe-86DJsJeFFlQSUjRbvc*{g6i@?9YUmjvj8Sws;Sfm<83lhubbfwwcR|s?
zJEm+??~4A$jvK~%`abO@5BHEJE4RDX7TO4fpLT~cMfIUu=eq(y%FQtD+zCPHSkfO1
z>~aWhK$n<nV}Mm%=mv{bg`h@Mwqc!5syTq7_0_}r>b<7djniA@FPQYD>sP6gS#CGt
zEv>Khz3<60V@<}XKB$}N)dhd2rf#3L__-JkY-a5K+4b~Q&+~_5-_+XQuimszP%C58
zP>&KmP(7~r_8coE-=6q}MtfbwXNW!`sQ{6wO~vwGj9T~{6uWH<GF@Xnu2G$8J{&xT
zM2=SY+Dr2PS9o>M<-@fCUZBhEmZ(jJnn<-$rg76L`#UlZ3pxyjB89v=!5<xi0zR{2
zqv4sbYnNfsp(lnS$y-7KiGvpn<GHk5gDUMSI7M`d8{IRHk4Z#NL$Y(A#J^eubu$zC
zRHe1_Eo1TeZ3Y<Z+AeWU&9j|OSw}r^kFYb+BbW$Kb~_O!X<}B2>eI_7h+0^V8NNzl
z>9)xaJ@3zHWRo^5c6uFf(R=Au;p=H__;=U1aMF{X@Su}ZxUOqVAIa*&kac<GQ@2;_
zTF-G}*#R<G1)b}b&ntiW-n@Ne%X0fgrpAE6f)6*!?9=Hlbmf?!uyNX))%ChhH`wyK
zrP_QVdfxALwa$dKiRAYth<Imr=PLSf!?zdpsYs{qgTcK4!bv}LTa5l6L*}uAIb2%X
z0<rjr-fi03&>10xFQzCcgFQ(r4GNbc4CE50AN&Q1-+B|7G64s~?;;wf00#FY!eAQi
zuTb?D8v-<{VGZGKwftirE9B~56fo0=%S$gD|ETh!-Mllga4V>hNK<e61+$Ifcs_2Y
zcHKoSe)*Pjs{M640d<BXMI;vq{#LX3546~LoXDcbvoE2=B9Sjh5)3_zV$)^C^IgL&
zvb^F(1UH)Cy7uj(FE#Bp6gsQ;>G8>CjwSV5Pw_Ol<*BOdpyiM*3_>s>Y_h3CY4yFY
zG0W|j*Wowp8vMuOCl`CpLc)(LyU4E#38ySI7nT>-wl(#)_KQVEs2-aQ)}{w<W1xl%
z%rsZMAz7RDkC7el=MRM)Bi5B2pL-B|s${B=|4<bn+S^UaC_{Mc-2A-~c@;glA-$2I
z3egk!G_lHP7ubA?ERzI|U&P}`--su2eubKn$qF&5A&ck8OgS#}K92wlmWKG7DqHY6
z1s7*|#=w5y41R>)>|o^c8u=XKRJPqP{uaICd@Jmo=CrHncD_Aqs&W3SAO7J~6a1x>
zK=p`b)FY6K#DdYfk%Rz)8a0E<w6%4>G{P~hlX>Hs?gtLesCDJI-Baep;prZ9k%s|n
zm@JIe+{81TzvKQ;uhre}$!D`YFkW~}PTr@sabPnBxZ^Yu(FdWZ96xknb{0e(YxxlB
zkX8gF56kOp7WqJ|b41Ql`${MiSKsqU@?nqaI^OHS$u~u(Jvh?%%x0kH>8p{)-<9|3
z4;#K`;ceHmOC2Ux{~lzI4@N#~5?DxwU==QUIX(g*CSJVw7qN0{*pXrWvhfn)S>|t?
z6cZA(XvEP&sk^s>h2O?P5fK((mkfMO&&}G+J@iFyi05O+Zxpt~_>4?8feEcG1FviD
z<nPC{Y;@l)>QBz*H+8|H6Vz`%%$^iLM&T2&Nylzs>wAoj04eM8U@?hUh7y5J?9a?N
z+{0A?BJQ#Hp{%DPD4PUoXEY0m_TxvUzv|xTN){P6n7?&o;sLh0Hf1p4#E31__G=Gw
zvVX7FM6@e+E-Q3pWUuj!MVTf+A&!LBl-u$CrozU>JmwxnFt?1Ku8-dhb`NI0+@(Hl
z8W{RGejdO&TMw0{@<@4{^Zos}C3<)6>$mj9^?N$We+1OtADNmetr%^WcAhCFS5v?g
zi!8Fb;M*K}BVeA3`1k$w50PqI+!27p9_<i5j9iqw5{X#tFyZ_GXr2Hg89ndjr*T<q
zHm*0ymJObCMPK|2n`4-$w091UI7<0|o1utmnZv;z_Nd&z_R8Th*;7zB<)m<d%#$!$
zx-qDpQFUyuIg6tSxiy6EZ7H@m*FnCsp9V4O47tt|jQosLDTRu>oB>B#b7}sjw%Wdm
zm1iqmp;728l<)*4MCbXDX?MH`9Kh|K=Ul0jvwEbZ`Z2o!Lt$fHi-L=s1lYWGed3!=
zFMCZS`dBRJdyFiR%PF+mY>~bFcw_K5cenkx@tj9#l1hFw;eYo?807>pW})Bt<Sc>s
zy(qburuiu0l33y`OH5zY-$OX43*O}94otX5&6o+dIpjEMRge|4_=T3tNW>5_y|y9b
zc5Uf{MN)n}f*UR_d2PiC?M4Ig&Yzx*R_Yh2XVS3T1m^Z)G&XzWM`5GEA{<~)pkU7f
z0vI9(D02*&ofe+7JX6W;BsYyUd7{VxkwXVDEl`?*Q!~xvgkTM@1LmEJ$d}_fhH&j!
zPr3|r8g!Cjdv@7}3eje&n<@Lgoy$jllHh<p>=gOmO*o+Y3WCS3?5io^)2n!DNO$Mh
z{e&n?^V=Q6!$QYZ@*dUeO^Scz|GUP?TYi>~TPDO0k1Iw9tNEcJsnv@!BdVz-DJ86O
z2Y_|8PJTTjqR|jl0t%Q3E*Jo)eIg{ZRnTtUfZ)kPO>r>2%Q*mMp|U(hyUBx@G#X_6
z#T3&~O1WD?PtMt&xnH4eu4C*zbC222mlwX6CMuCJvKi8~*WXI+xWcAqQ@{+X8i>V?
zqt9KXFe*1IgPNH8q*l#H?XoQ7NY|5_OaXOGEfY5bkqD-aaAl?v7Jm_1zoaqn@l5=A
zMXMBC{#}P|4(|{QI)Ax%MI~A8Qt)|waYp@YIj#`hclYE$aryb3_A}pKSQ)j5lI+8z
z!HvAvwY7nTr^WAkgTQ~q!Na#d8_UP0zZwf}=VlwFB<|b=Sqw5(W;KNY)XIk7Tql`A
z<$7#-GTNB|t52cy;_Y&}qw!HN1j5h=!J?Hl0&hb4wvyoWOTs^Y-m3-R7S$-SZus1!
zr_=YiWVu}ep9)`sm2}SX))ph>+g#Sk8|})g+gD6H*3s<ni7JE+4jAEGvl<}?+-6=(
z-?ikyD9amAoCQ{XCM9_!IiI{d16&pR9kr<pca0W74lBF9L3>c9^|*e@M&WZ`DqSsn
z?2LO#+nZ=syeg8KX~)A^AxV`OQbFHC-|K8da=v?hws!y2q+<QFpK2E&pFn?ruioCC
zw5EUmm)gbG_fH|#S8rrn{NV~Ao$kFlFw2w6v;wUGVwCVp0K~xp3{`&JSbq43(guyy
zbf;k4_fJNR(phF`aubMWwmkBnSb=;HE;cx-BG<G8-Eqoc`s-#3>N8<i(RN+jZ?_1*
zLvuR}wn$>|7)g3<|CnyDq4{vi{@uNL$iX%!-s$(bA5S!Qrk?S~@7!|zHufiZ_z<`v
zm#C=0WxO~iT=h-k>v|oAw-_MyfpcIyac^#)wk%MMQ}Ru}49QsbTH#5Dc9et)_O_1~
z$^AP=0WeHEY^3Jf?D|>ZuwJuIYR75Eoy+lZTM(b1AI?>f=ykCrA9)7@(OP@>Ur`{6
zpdbIm8}Rl0Q&rH^Yai@;evfS-daDXFN(HenD;ptUzbPM(#==$a6B=&f&uN^<UD}Uf
zBI_m(g!W#rCYpy=SO0MC?``-YL<vmGTt=YTcyzT5i$288@atw={Ti#eUE|{T<{lKk
z%MouDw|bR0{ANKLh1f))bHCOpL&Pyb{&7QoWbp(+rr6zybctX|Nk{DNKLj!Z=J$^X
zb&=mo6z~LA<Ak6Zqe-e5IxAJMI?%ikS74G&edQj=HfL8ev0x^Lj#F;-%S84ayJTpq
zw^bl>wazXltfszM9wfyK^u49JV}GbMxDzGfC5{%kVSXxdg*4{-81&nP{jWUw==Mkc
z)P;Dj8vG$u6}(t>-&fx-53j9F+wJ(xIxJnBl9}CO^Mmsk*D@$B3%>vHBM7%2jXRIL
zUrm<_t`qTsK0i?3f|QiGh@3bKVwSsL=-tCPj$UsZL4<xGH&p#g-R`bv>z683e^qDU
z+y{=1c15l-o|ObO3WB%7p20=&dXF=e6Y~!kR06BU2}d#X-Z~ZJY;4jcGfP12Ax_zp
zkA#BcM8ssR=}Yt%?-yHgsxb0-%7`6^B|*J<M;~R=YnPiGBGc4BIH=YWq9T-Hfa=G2
z->4cB7Zm}%*@jY$*ZOYTo?hGhzVTze#}*~CWp}-A@6c9#*ZY3gYZv|<3(x-PzMTI?
zIH1Z5um)jC!>ew7iP6~`1CwA)gs-_(w5i97ZYP(BC&ERl1bkjwi)ehsDF<tr$i{O?
z#?lc0a%7x5+7bLMDq;*9Jl@sS1mGlr7Jd<u1j!QRo85H1>mk02g>$g?h7mkdc#v6E
z1%=~ojU*u1o@?GGDE0J7hNvzzdPFcPId0$32CeF-c9=v=k#`M5S_r|t($%!9xjE3y
zc;%7S#yk*~VVXdIEni<I6Xx&uWw0<@a}bKl+A9b$Mn=EW)obu?*LVT=ogaOeNSG%>
zfPt)aNu+YIm7kHX7I=X-;}PV0$=1<J^j3IqkjUa{URq?xm+HP_u;z|h>#^ehS4vmu
z5E1e~cYJz+agG{=MFgCs!N6*)T;5O}`!1cY!sNd0l4MZ^VOOLn>?Vs;!)8J!!yo|o
zo8$7Me0fsGm4bgzYg;y@=$u;$$GkgBq)<SQIRs)yM$lzAfm5GN0ra<bF3mpr95KKe
zg5zl>v6I5zl^rt<AF(IFryT{P7DD9;CBP%xFJFaBq^V3Aae|?HNMD|OY`8(F!q7$B
zc7R8h;7!NwCYeD@Z_`+nF91H5E}`JjbPAw&XvwrJ=DV5?{63yB661D!a)$0o8bHh2
z6Qyo7xp6B#OJ4Fo@lXf<yn(@*40`#&aIiHyTo5B5qGA-UyQ|}--e70!G4(Of$pX&F
zFWR5E^aZo);NRU(C#ZiW%daeoaTX2!maXO@Lb7K9(4KkZM!T&M_UBx!RiS*rVl2FP
zXaa|TJdgzryvuUj!h2G1LFaWEi7+SW_zBr$xgqayHTRM^Z*aE|d4)e_n)jql@`{R_
z&ZoNZ6qjiVg30>v2C-Iu^V%Rv9KfOCtZPVlh0FXRJLoC98eJ44)`Det;<b=#gWc4-
zC1VzO*;+2X{dhAnqqQc<-V5DB997(k_aCf5T9E{Wx4%M4{qX&p7X%+*kH17wf7Vy}
zMEc%SJf^vR)q~yX4iyIBPnWrv7g?mHXBX>~5{7;{bsKTuHFfetjm4hi?Gp-jrfFZ{
zs<#?R?l?R2T6Yp1UY_KZj(+BDVDI&>hl_vkW&va_ND|z2aYhEU?Jte1->nZ+6PXTz
zSWuml;&;gjSYYol;JN-t#LT9J{y=})F)nXuW@7g&u4hr8Xzt_6?inA2V*AhlTXXZ`
zQjKB)Q&a8Wdi)^&>r!mI&kp44;8_`fQy#V9^4zlaNmYG3rwq6!7DnQ)L$95IE8uAC
zr(-w|Dp+or&DBoK5dPrb#;hz8QB@E`ogC*v$=OujqxGpS$yk`Sf1QzIo{G^3|7Ysh
z9ewuo@@{Be=$QAq#nq2w(c!WCQ&G5~z}=B}$*1dPcMXlGs=B5(6&eUn$>wRw%*>4U
zm+zqPS#lbBS?KuTQ@}Q5*5iZgRciKqK0=wFfml2r7^eS6yIjK40RGKY6U?l-tgmZU
z*762~(c6IHS3o$uZ|52#2d$X<Y^~lc7Xgs_UEz>ZSE%!aj+As@it9ouaouph&2>QU
zT}%XeBpt5Q$}=PHumiQsurINUSng>w^~hjWWoqJ5tmgKH;fwTi=UW!Z^-ipfaV`T=
z$k?2=cDc%Xsl8s~3`y6P`OdR7qTTn9v1%4@MGkD-%K3PJRk1<J8zY}fE=Vo^)9L&M
zW3{Vv=T-v}bAN@*uv&f>)pa*j5%e>@L{0&%d~oQhk~MSBMZG}Lie)u#NAJ)Xc7?_h
zYe=S22t8Rm{cDT>5s57U92k%qxBYvJK%l+%?V;BVljv}4Vxp`1QrrJP0%KH?1^VYf
z*cPQh{ziNovoOlzmo8C|InL5!%Z4<xNOD#@6;?4hzZk;(PI5$FG9oU)Cd0(JxjNAU
zl>sR2*Tr1CU^itoGayA*rm8w1rjl+oQG;Up;^lpUt}Di}){>z~?b|f0GP&^x^dquC
zXExIY_52Ad{62z3OTOdssgp<ngvlWzOu)x0Y-bwh>jYz*+3t@{^?6y;A71Gf6&!R&
z4!N{r&~p5d5eU~TYH+|rYxhsQ(1MMv&qTr(Kc;RIIbh_7ZS0FMU+Ef?c72Pom7*0p
zo|!CTD=TahocC#jk_1SOdVUiC#=#_zbea~%eFmH6*U3Yd)bvbI0QBGIkZ<-Mty!|5
z^adD;t;WnU{wp5&<qAiv@*CEINjusE3A8w|MElCAj0r<AA*G=m&sjo3L&D)I{@zrS
zH-2jVO&GNJfiu_=LGu%Ud>9BzgQX#u1tE%FzZ|9P7A}0olC<bAHN~Z>Sx2E*R*xYL
zqDkns>@nmuws1v_)gd7*F4+&fn-(PRf2%4!9(}mS`?_D9TY`frX#oaANSEVjCo~Yf
zEpEK)t$xv-bW;6|jl?Hg*DEP4s3<h<zhV(}N*y%K26BbCK>qJus3yL*SFLej%weG$
zDG1MAg$4rK$Nen2bjC`Jl^DTtC0AwC*uY0ACl+7>;l4;uPe<sb#_is=R~8HY`=I>c
zCVC#=M9GAG@CvPts7;8>#7)}O1hCKNIWwf?7Oj#^QZfd{8hhBK!A5{Xf*<D^5tdTM
z$Q|BVwkH5x*qAMKj0zfapdUR&><*n&D|(OVz!JZ--h5RKJCivB(-?2+dwu{4DsBEu
z1{3z73~`J(R{P2*H^Br5;TM`I{iKsGIe%FFbD2rt=10f^;<--q+lBfD5^Duts`4^a
znba_X!3lpaBX=sa_l&LMVQerA&HAF1Us|gG;C^??gq4Rct^KX7x*FccMmpXx3P7BA
zVFbjNq!9;FqaYK0$Rt8TR#DTy5$qI@21^2{N1++)@_7|5zc9S{_-8*S(qg032tURS
zfBg?i*4-cf>Gmu>bt#rOJx!EE;Iwl;osMiYE!l@d80%aEVaq*Nzqyt0-5yzyxen+H
zqA6F?{>?8guGZr1R7Q4?Tqp}~(YTg%b-Yy?v$82&mzyv`Q2OHgF+k|G>`U$pq4$Cx
z3*ZW#spjo)hiSS99W6GQa|FelaZDAK|Lj61U9M2}MHE)yYxRQ^@n{c8f;Yr^^J|xh
z!wUSY83MPQv=KQPCc<EvCjkv4EMO&pjQWW92m4l0lCGgxIt1ShaqI(_l%uUL^X1{=
zZ1o4IT&GcmxV2&&qnfpPhLdFlR*wT!2?wuEvgelJ2j~Z!eiqW_@1#DbAX4*((=&#%
zJOP0Ie?$q_ALY!fAS+ab)L#r#=Tw)+YJ@+r&!neMb-FPlQXQ1-bHf{*kx3U}JUI~)
zdGW%S_`}_nY`2^Oeo&}D0cWtYFw9gC?w8;%R}n^p!cP2kmfD)}$MW*lm0ssypU`M2
zYovLARlT?la5`-Yp(<g-q;c@0k1#1utkdODW5Weg7ZfUo8E0@(Vrp$xk^BJ>kIMBD
zWnT&F$e8NowoWlY<Qgrb=$!qxR;4Dzp{94M4=0c8t9O@;Za;G|1su;nk2iDQWL#Z4
z?N4602v~5zILRXs=e#H=D#k_-(IY%+WRX%^?%yhb+mRcW?`nb^=~H)3?%I9DxQDd$
z^z#;MCKnb2|7TbbSs7p@2TjDYkauOArLEhpr+D?v#c4zV|K}iPkpK?`ohh13-N(nr
z%i|;ds6&yptKTEuu9s5Kwml^|AOrO5h15n<@KTeEpPH>+omgxma8K7y7~MfX`~lV7
z@)HR<c7fz7N}XglFlt4hl&-ZZPDUgD3*{-RN*hP=%i3BeK_?|D5GgK-yfMp*8fTHt
zHb>HcYdk-y&#6uV&U?&CGsV-`OAU_TZ2g3B+e&CJ!8amh+ZQE)E(NnTwk|R^-U(2J
zANKh6l5dO>3v&^*q>$29yqWS*AO0L|T>Pl$Ncu6=vo^#*Bb}uiAusBX%odzIxeC8h
zmx+9w&ugs5<u0Ck8zXmoCqLs@4Gw`EtXm>`?5=<)yc;iyT)#%o-u)Od+u7LxpgbKW
z>2-|GYr2&PlBAKH5^)3*^b@E6O|<!vf_+5Ahfperp(HPT?=3Gx`8(mGNZPpsNj-ae
zNT?MguFZ!Z`*`_B6`iJ*|9qUkl|ElNx@QGT$or8bOk!plY2yv+k|k>}1l6QiX@SXW
zPTACJ>Y~r&4^S1$4cfl`Sj4kbjQb_iV!I3_Xh>5d4-gO+)0Y#<WCsr$=S&nZKM51w
z9QgctEBleK(&xfDFd$%$n@L~WbbWbg`C$#*^Zw}<X5|L4MC_${EEK9Grz+v-bf@z>
zFYC-|3;ivnJkW2eIt*O^&1nFJZ0!01o>8IHnrEG+2G+RDG&~^8a@t!@vlN&q;Qv;?
zdnefvw-EZvgG;4t6b2JA(voJw|DkZ}9vfTieFl*%L%Cq{dod#yiqF^KdI`b2jCS1G
zGkUmOTK!h(^o?Tb?)%&TVh~RL8IZfj{Ohm5EoI7XNs{^!xnSt~_^q_&?Tn*tlW8Kb
zO|wi{0aUdDEkRqUBY6#7B<d(O5gjULgPDfvVx!&v!4|-}<u0CN#!G=KyjskR9VRT@
z{H#Y`rf~yGRzk;$EYZr_Uz}nS;kf&g#r*NQGR1+{?~ot?xt$qS=0<%BvFY>ry}?cQ
zHhXQTZ@8Sp)$NqcpwCvL>FT%@EbNi-l*9MaW=oxp;z0#Tqs4C3yC;F+aUWbst&-~|
zP2H6y4<{mpY;3;LnECAr042r2{q(XmPAiZn%FO5oq`HX!<gdF98@-NYGoKckwHcY1
zPKBeLqdn2ncAkd3w@XXkDQZT1i0<vQQXJS+)#Y{4YVxTV<PJsn7*9c7!MgoDES-*S
zD9y77)%*#(=3bM==_mO?@a2&G>qv9M{AUN4>n8}pAky!HNHsoQ%A(p-^IV%3!sb}+
ztx211j&%d`G3O7QL85$;Jm4HINB>;S$IwFl2frSmR!uJ<iM|$1V%Uknx6#p@!Jk_!
zQjN2F@W)Vl@1rr$D=t}1ZrJBBuc&j<=P{SJ(^fP+bhk}p8b7N}5xnZ-HJw=|JBBmr
zgc`KmQ{j+2nQEIV+jO6fXo$S<YL6o&qMySOd1FA=YHqP<xD=eddKapn5kvCQeyM8C
z!0WGRKn40|HiRsAX}-=pvAzkSp7Pi0(Lmu=XC~%6z=Bzkw1dy|VhSidwvpH?{h4w)
zRv5+zowHPXvHT3o+&-Y#yBY3Q;K^@#JZsp@sTX@6Q<%4+*d)+8FdB(RmS*Gg8Fj2<
zWIB~}zR4w|INLX=N~HIpwD@YTLj1CjjrR_}Qln4-Sio|A#DQsW2kG#>9NyE<P8a<q
z;bk4Ubl!Lhi*~r#e*Sa%7ccf_(!hvF998?E0t$ln9D}?G+dV@aL@HK(YA-7Vt&!!D
zWh!LM#HSCdzaNiBdNAJ)^CNv?fIy5;f8IB)!W_MovV<i;)1jV&S@2O_Ujc=GmI9}#
zJ8}Ni8RfYY5!u}ik!~oqtWvA$nkIVgxY~V0R~hYFdA(seMOuwnyD8yWL+jQ6lP6_S
zZ;Q|?P6^U_ZmV=$xETnBy0P_aB=E6ihi){6It)^_GZ)PGHo*G}O)Z9RTm?4J5J{r8
z#w|YRT2Lv|ZXF%DYa+f}PbX(zllcs-&YWjxKH^5<$I@8mxKFvb<V7diy>2?#_Qk-k
ze|GzCx-9o~LA_ydfFd1EHD!o?TpTevx!>0DQ|aSGjJ%Tm$qjBqFWyu;%JyhWb_&R{
zy@KGulgwDj0Q+SY%-q}(<_sP&3?5-~zn%4qu6s=W74_*m9mY>bU0+O7K5m>iw5Jn}
za;OXdGu>tQ^A9e`u%Dnj4eN%%XdJxtjB4wvqZ9}+4qqfxkqJ74PP#I)iRiEePP*xT
zD%2bI7OGk6pW|~smr5Vh96Os-8w~pGGhihc&R|t;V_wJK3K@CM>AA_;#JAaB%yJH<
z<Xbo(-Yb%#6O$#GHQA8Buc(ZN5uh_{PN{+q;Z7yQXX`r2si^rj7A!zQN*QH_rIlyz
z_0LY@KR>nqx|&<hC<~3FU%9Xd!dLVER0DaBAQI%K*fkE%@6Bh84fFcCA?~dg|35eQ
z#7y(`N@p*<YUKkXDc>VO{e<{hUIIN1j}>mGqr<Ylf{qLpaX*vxI}<~P7^t{eli$~?
zrmPV3y>GoPdIkB)+{Twm$tU`YJ36L3=c&o1XzJh215O;?UkVl0y<ZUQxnM;Qi;a!X
zK&zgcxOkCN+x4Q51%yLS-ymPbUztRX7+7OyM8`NDGrGD}*>Rl=sn;y^6xl2s-MQl)
z%gD%Ee`0LpbOduz%1EU-b}e;7{``s6ALvIST%|ck#p_2MWzwROoXW$47(*3c+|dy{
zt7GY}6eb<;_X2j^{=+U)NM4yzbYS2~@qXJm5v%v-d`e;d9G&=rQhJBnki1n*nN*`@
zk|PN&K4^apdx8ctJ$Jm8j<PAi!iTPODvl3!)rO`!#97|Q)N`V3ovZi5tB501kkXZT
ze^$^Wl#6L>W3GAzc;5zJt<+Zf5jroWL*BKIu+?MhSVDlZak0NW+Af#HYZ&#8UH8i?
zCVR}y@<(^A_=p0KU%lmgB`6t1tH|AwZi|75p_SU9I8(Ei6xLq3djE6U3OUQd=^hr6
zk($bg!?vugy?IbS?D=3fbvI%9cbI-7ExOKsU->6YUGiOo9Miijic&`y4o@C{xkfI-
z-L{q}UwZ4`wONVJ5?w{TPQc9QsOc`PE69||J9^fmp1cc3R3YNNoXg)$jp8{iP#fHB
zimpHXc6WQ%m)!wh61{&4E&TMU1!QXAmeJA}?LErn#Xc{pw$0{y!w@wFHaWh*<+Dc|
zlLvA9`3n?ASjd3qpR5!;Zh50;@Scd>UgfEGMvh!76`dZYnIDx#6(oH8e8JAOT2lj5
zBKQmPWYr&8emn`guV2?)SjfNAVq*`KgpaV**0y{I(P@W?ed+Q%$y_@B{Z}G-IJEft
zPy2`-TbJBk`F-0iL|4n2u)*QJysM>8w~ae=vN6M(pzK?p$Vf}fXTw|3s(r@PCOWry
zr!#nc;cgyGu0Tb$zGs=J+qlzIGrd!>?cY}epWZKa-{#j}*Y!PD{C+d|be}CK^6eD4
zamj~-VBgR#KO>AZnzT6A@9tD8M?WX+y^|gegmxc~46g4SGHDOvUvg+?#bEnsKGi<i
zUS>u8(IxbiSvL09a2!8>3t0Eqfl~>~151BLnVulT7AvZ!D_8u?GQ12Tiw0bj$Ypt!
zyR<gn%cYc=v336_EUi_)|8WkB^;u!u2s;FM`QC5jt>M2Gq{o|-oucc<(&leZ<>lq6
zp4RPL0iPkh{f=MY^^L%rx*L3{aL#nXIhx(JsXrPt`}^5NA{6rcxoYJcfyDA*f24Ha
zY#MWzfWsP6lG)KgZP9gc{QHM3uF-7aqU-owpY+u^WeZTS5#lAh_>q5u;;*luXnLM$
zvcG@D82RUdQxNCt%s&@`U?!b7F8^GJ{C{0|Wl!m}L~dSaH;gzG6G>K5>CIOOL;wE=
DAr}_A

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_018.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_018.png
new file mode 100644
index 0000000000000000000000000000000000000000..94c0d9298c0d561c82731f84a2cacc99bc2780d8
GIT binary patch
literal 29575
zcmb@uWmH@5_AQJz1WE%0hZc9&V8vaE7k6za?poa4i#rszV#OUwuoB$eTeN6je&^nE
z@9Fv9kMD=<CwoNp7<u+uYpyxxN`#8C%u66K5CH+<rJSs!8Ug~6>+|yf0NV4h1ghxt
zyzsk7X}hS~Te`TLI9VVlnYcLE+Pm1=n3B6$I62$c+i|h-v2rn!Tf4Y8I18|`ef&Rf
zV6}I$VuOiF>_2}Ax`V8aGXlaZlYbtFWg?|E2nbG>a*|>i9$Cka?~`qHywYFzd7qi}
z%6PeVROqd`ra{+Y?DE<J&{|~!!J*WlMUraMz2Ob?T(5K+F4;eaO<9NHghA;Oz#Z5S
zN+~2`WZS%ss_HMSMy~b-XVWs4zqh_-gyCk~;Lno_s?MU|%>OZP<l^F_GR=K_w5sy|
zqXpR`HT>rBYjYrqfXxuAX?X1f_1|aTJUq`fV+7d#_v>kEFaG~I!+D_hH?OjAbxpBs
zhyZO4b<nb%B|-KGG0`e3E6I0yj@>6793rirUc3{!5b@Z|_|ud0CgnEH{6|%FwbSu@
z-OWcw!>7m5D!pci*X4eI;a0{QyI<e`tVjI&Bj;zPPgF#%<CeIex~|_p$viFo`T4HJ
zZ3`${hKixs#N@LP=NFGbz;?deFyZv@13g&-LgxSgA#r#j$TFP04Hh*`(9U9*TwKqR
z0E8t6l1BrA2#b+WylV(#8606IRT2zxpiuhacr_I;u>lyecX}&`@d*T^kuWDDmk3G3
zp+Ff8aK$pi987<l15(s#tzCFIKu`jmKw%QH5s5e;0L9pqY)B%Ifkqd4&fG_6(Amsm
zOa>=9XWki1A@F#7@rQ(ViBb+luRYa~V{Fgwnn3%lACO{n!by6I_NW$aTvam3Z}T}}
z2TfR~y$;{(B}_Mwd&mDF_s{j)uAp$M(+uz3oYsFWVNw75a3_fRhB<rNBUH=Zobk+F
zHnnZdN6?l|usA)qWKya(X0imb_v<=f9dBQGi(Yr8sZk<CNI`;T`3KWS_EO|Qp?JOI
zt1po9k|_a!{(X=1!QBEe$+6@J@$^LNv<?cW!0=w62`NS3xo|NEpvM>qZ(IK9{Tj|#
z*_7le80>R%b<7a%Oddk1!XX9%$rPIs7jlFQE()VISR3Ru(Gp?YwWk4QdBw%bn45;{
z1os^dvgi1H*B(*Cv{xlzrgXr=ES5Paw<nKPmgTl*ES&fwW(aF7yq*v7)P2+R=DV=-
z#4^{vmInupKMB!o!k#B7nOTDVfp5Ou0fVNn;)ESz<NO{d@Con<X_mfS?eZw+3duC7
ziD3k|B0cYMl-hdM_Gyt$K>W_xN&hs9rscU{zKVQ4l8)~5atArX>{AH8m1A;dp+Ue`
ztSnJLSRHW{*Ifsh*V57Dy&y9SOKokVB8V4#6F)_lgaR)Rfq@0SujW}bunB@ldWtfb
zh<Xk>U~#GHQ_yG!Bw+<m1E|GQp$#w!<NyaC9%EKrUU$T&seRgPm=p?{K7ukXLPD*+
z*PiR!ANSlB3Nj8bN_|hKTDWN?DrT=32x$7=4<AzL49KTF@;P&5|KB_PujLKa?hDSt
zrv(wufW~<W1=DD)Zx5TUaQm~LGJa2e%&g4`l(b|q8Fo$B;>MK{dQlqWG{qQm%A!C}
z^|CnvLpQMb6+b3bFT3REv;hMxP~|OcwfYN{HYGB&4RC-9CYoSnjR76e&HjlU#Z=Z?
z&*Z{Af_OFdQ$q_VH0<Owymg>3i~kX298L)js6(=1QPw8KkC+#c0&I*XeFAAiC)?=x
zR2e})3Tkofyv8qN-mC^<d81Wd;TwvcERUyTk&QCOeIs-!#u5zsmje1-ZW$qkL-ES%
z(JNipMIZpfjYcwiJ!C^g?2`s+d{R4UzjxQK|GWFkTq303Uzz{Vp&Q0|0fr1xkA%-I
zQSjBAIXoWXvkCp7b*`d`ePbXd7gmRFajhSMg8~^r`6dPkDDZ(%Ce$w*7xzXWExXQ~
z`WopT5BMp*MuF>~#M(mEvoYA5_pWIw<;o;ucLq@LEeWj0dCKOX>*ADn9H(*ni)5$z
zo<UIiu93}A2IU;@U^nm9#>*Y^hLhJWQR!t%9o1$GG-9A+y@ON6!*+w2n--4{3Ki`1
zK*JlfM5J<R%Eoy`lPLhTGL+TGGwhY$@2y}?B#Y}C%x*%iTm4!}N-&Uu2P*`QZ-yyf
z9<+Y-pLaZnOC(;Eu811&MHU+|u@@`C=3;S-kMI(gcMfCK?}Bqe=#f5Ojs8GI+_;Lq
z%4^S|cg3|!H>i&?Yh_KPm|86qMa&~kisj6(md9dqT@sNRxJTNjnrrlD$G`XP5EYiq
zKRA^_b52>>-i{2X+zh9$Hg$o~9~7F(R2fpp;naT{8))ttiGkKQUT*hg3Ei+?MA3Z$
zhMR5zStdY`7vgzHAsW97THKtLjVF|#ss)NI%=YAba%C=rq@Hg}=2r3+A#&sr_*Rzt
zuI>6hC$r}$h?O-piB|lxmSj*;?~$yVNHri8ikp*OxOfmqwmAN*FXrt3=C_>N4P!s@
zLS>8?ATknmAYfRIy?T8yg)EDO!1Hz#t-~a)wOl6{?;iRUi8-lA^hK{t@C>BRYMd#M
zoqp?7O{wxh4Y3IGdg~O9YK#d_c~!+@!?`*8zH{t-=eet8=@;gSSK99U>gvwS-zQY7
zW&wqSADIDVS`|fw6um@;O@Kls*|frO*7s|-yj#D}UJ2i+?h?SL%s)Bv!{jQN3`*ox
zbpN<*V!0mE2_X?eC#7+vRc#j95sJ+-OpL{rvZ#sTV?u>4B63TG8$OM5XJ)3g`fRWi
zDXLH!LkQzbh)x(VzDCz!P|%iQglSk*2Rb7vTZ<MTN2%~@=^fv%7yO<4*q?8Oup|rg
zI!tWw{Ewlo#5<;WE^^g1S2{*ZCMCiKw3?g|U0d;krdf#4IygCfg^J4I%Jr9dCJt8j
zc_m(=?l!sBI-a7!*X`{~l{Kpyqs17<r<fzs3^&<^+r#IiF<#kP6(MSo^zoQp?a|Uy
zc7|O8i`kli2%oD29BGQwGE7}vgXDQi)(m#M*X}f7Sq6LgbZ;Zo5MjVzt{tnWjuxNR
zl4%}2zosM{^YTx<lop=73(i}*7}A2%%whx#&^i0shS{C{O`vBgND5n=8YO^&P3>HW
z>dOy80^%akcZ~iZVDV^&!=PzpU!<I7+@gXI-Z!gX{}~S|>>}x^Vbqus$&vAt2sqIr
z=oTO2UT@{^9H!_AIXr>oQ_;iW0U4JTb%Tm`4K0rzU(@^BZsb$OI`HlZD2tg`e2Qy7
zvEm4jG`BdOATa{VG@;xdC@`X2W8dhQ+So+cxr-8N)xxK3c_SD(@)IEj?6Y>0mANKn
z&PnXy!_&!AWiri6Qdr2bqy`1;4X~)Sd(I1)cMj&a`kqL2Un>!ivRKH9Zc@ed^aeZ0
zit7pWj4KBj<T6|sR5+<)8KZ~Hvc_6Gccii~531PeXV=6>5GcrF$x0?sOoo3P3$+Ux
z^pbZYE?NhJg5@bf=vTUvCPLkBu)NMT|1%~`{)q`YoBYw1H=tn<&1f200s`GApq5*k
z$z!jVG56s|{D)$#HjXM}tDeL9gB)j3qK4IXy4*<Y#VjA_v^wb5<^TclXeK;ghjolP
zdY9}wC39lRd5Gt|GG*mo3;9Qz$)ss(e=ar@R8HKMOaUP;w>rNt!+iA16=nR;M9ef6
zrBUH}7xcDcRS;kkfs9T(&0NZN{PV=#Cqd-;qAMJXEzT62p&v=l<Rhwh{^F84pM&iJ
zkcF@%3dtIqBm_~IR0TvDbH6sL6x*g^Ms&CP$Udsn%A%$gfs&EZsLQ3X8Qu#nQxQ|)
z)p2R5UnY?;;aMcAbbv5mi%Ej`HL70)&iyYy@Go3&A@Th|RC-0`tz&A<)8f^6fmCpO
zV#8is{tyM!-Rt}Jq3F=iOupv%g_DMQjY60Cwg}CdI!k*KB=ZG;ioDKs7&ej>jn=WN
z59`qfQeqQSxaD%2UqY09M|WOV#8jvupXJ>{UzVV6Vqp{)ri}1cKEdb@X33cJz=6$8
zp%e%u0BP}rcWWb^l5IFDQ1PY~u3A1zxU5@jKU7(5@3cM7(KzC9KmK03-8UPz^mR9S
z0Jc+?LQ@~-fo8Z`L3D%yLKu%blV0VIVsY6BAf+TZz4s4*1_&2TbI_AOEP#WEkseoV
zDTTpu7k^CHA+=z*T#LklsHuI4V~H}JQ54c`4Z^{bmUP@9_xP(;_Wx5W4LW=>ZAtSa
zQ5j4fRP2f0UM9uWx2_g1*!~*uzXdL3BHOJjn>+fMtYt?qnH|&;G~l0k9k=4&2#B0T
z#3RvhkIKRAjW!MiYfK|43O9H?KiAcIWSRe<At5^FHi1HermVYrH<9E~`yIn!IJ@`+
z;RR*i_SvZ;E5PvDnWOr;&QX|h80<7W<|?@b$6;GLhVte7vd-SXnO~4P-NaV1xbkwf
z1x_*A`|DZ`T#I?Z@@4NBi-Gw3<1b_;wcNxXlfSU9d@i0f%oALuByv5txsyc~13RoC
z$JM)R6Q&~J1yjq3cdTx%MiT9j5~9<i#CBjUTht$VyHEWei2Mwk{2=LXhRdMMnPm<H
z6x0Ec<-_42#k{ZHdZfy{c6}HX@lRM)4sA-sus7J+2!ko(pdwPl%Lf8REkAXy+BVfw
zTH-k8(g!qI)SDg<G(nKl*uG)$mJT}y<Ni2u-K0mba!^(;NBa&_rZDe0Jbbyki;rC(
zUyKt^ZB{_v%RgM7Repd*r=_+3GVIdJ;4Vn7s8u=?q)FDXK`f@7BVBFemnrlve3Mp9
zE?EtNiR8(5dOCXR^o%Dw(e9~4NyD9j<;%cLx+y=l`)3)|fC5W%3kSCGORsf4pa%g1
zgCW&Fe-^&Uo3-O4#7a<mK4fHMuZjFx?uekpladDkBI1ejY2?Y+EvEm9nXw1a&yJ^0
z7QCm{cPnoZ5%V$WT?Xk7I0Xd<9@Jt!XtKWsR2v#vL5QalW1MD2;WGdR4)`YNFuE-b
zS#cPAqkkoK1_DANP32aaU(hX{#Vng+r0017BhTJ|9V*4Lv9QlttD{%+o200OXcPeS
z2*HT6WiXF)>>-XStxRsE7KdMyYN+48E`e<VctYJOdM`DCR<f}GQpGrl1iq2BlKM6&
zk&MUGAfsx>cP0wZHbpn?YEq?51SS+<go+ry{J^2pUU6R=sT68vTD7j^LY*KccpNT-
z6~n}grjSZN4#o=LKn1A6Hq{H3nO-0TIN;!Y0<PWvPjoT<yV(B{N&bhJKyvi(VC0$a
z-pALzKTYiPm-2MWovxQTRy}qEj-Z0aE}@D|ddW%vyg|aqf^_X~c}=YM^NPfHr<&Qz
zOyU3(oR3b4jWV3Xj)#{zyOz59LVKfv1P8CwmA}86IkM#>ra@t}{34)K;*3PYfHUsk
z9_<?HR~HlwmX=FYoOyG(u+kZuY!Nbc`Gfn!E#!$@0u?(diUqCj_4^vbjg6sEkCu2W
zC^KAfFibnZaVOr$qj^k6J^;5~sF<9d)cQr+B$&1v2L}bQIy_+-iOBm@n}eaj4VIr=
zFlDQ}ycp<oYIyrCTLXNu!*Q%R8f@YL`u10AVfs&N(bk^IK1}!KHDz(Z8p%&M37_Xx
z@liAyVnD4n6VXcy_?$qNKvR8Jt#I?MJswBh?}2UiE*`8n5K2IAejROod%IMXQ!+U~
zXXO|q?TQ~qT8b^H-*t_(^()@Xj5s4D*UDwop*e_`zwj4phe_`Bu}~qkg)=m)>I91u
zFf}-NcC+sSPtUc-4n-UUW1<Pe;dah{f>6a}K}r0MZWEecx(^@c+gX8%s;LYwOosb0
zI&!1f53zn18{KId-knP2K!g<v-&#Uevy-Ej9`#XsilEI*N$<Ay?f&F-WDfF&)mnTL
z0R%ouYSOo>cfJ1$SvatZ%pbqcG>z^yO%Trq5L%(HJ;9*Vz5(2>)Z|j1cmj{+;2w7e
zGUyUgHZiH=ca^T)1{mOU?e9(^SaCu1uZvwILQVQdN3pf+FNdc&g2dutVm?Mvmd@I8
z@iq6?6`N&2Tyw^&cu-U#KFAVHzHwzZC(YiowXbr%fNLhMvM%Y7!Qp7O*T6+}pbcJ=
zd~jK)g+vcE#e@mBY>UeQ(wlqqs6?C!GO228_6TQ;B9JUf2psGwX2PuosQe@f2=rxI
zaoX4v^f_u@+qxd75Ma`_6Kk(l&6gNKfRYWOLP+O9N8Gebgs<+0L!^>%2=lK~M1g{G
zzfma07CZ;j{kM`Y1+o9#YX4JNU%j|MCU#Bj-KD2b5e?Qx)>~?qED)2@OjrCZ75-^F
z#-P5Eyjt>xSsuZ?0&-pX!y8Xj<$BZ+c9M&R5-G1w6)oO4a&T;0U)S2>k?eGC9pu#{
z1_~iZhGqFCe`r!B)F|UAA+(Z+6rxKGH0RE1XzDR;8A~}zyM2mcOX_gVIE5jYp=>$E
zm2IYfjzmWS6oy+?^ukUB;P@!=pLli{Bo#;7$so{c5)PmauTdi`ze$V{16SeO<Y%wK
z5&vIC4{7>sm%LrK1O6jbP0;=!FdnIvd~^Ua7onM>GJ;!|D6Ps)%ohDSpqhM9JiX2;
zyG6O@0_4|Y>)%Z$=J!|W9~o-3s6xae5h##FH?w;3#|9`7B${X;Za(=Mug%!C%rKMY
zPF!<Y*-UE=&E7FM;Dkv6!ql8Loml3V=PgTT3U#Z<CB-;lfWGk7R;g)jD`@+~<t1rG
zX!a}*V+v3~EQGx}Un9E6CFE-<3uC1dZ5j6LXGNt6C+HO~-`VjwtO-k<izB6yoPHE{
z;QJ-Fy82A8ys0Z9oZRt#B&nEHQNL6igl4k9;=S3~1kcBa`9WOuGNolu;+1f(?@jjB
zZVZXf<F1l4l^8jxGX8!q$7`B&xzYotcV}8#<ka3TT}Lxz5sIO3#hgDTcE8p#JpN5E
z{Zl;&)89B;9~EV~5UzvhvD2z`xZxQi1=KO2>vVpHK|=TWVU94h{rW#qqA5YnA?f`o
zVn8{YgR#dbC*@><!{ws3{sIk2kk-QdKBDiJ6ykAd5(c&1J%aGUbW5|ylkC?kt5~WC
zC%Pe;*<4-U#7Q;xU%4H)emyvL0Wp?L+d4e_&_R2o2yzTt?g?QMv~?eB)YDW{jdw)@
zdv#01`J<JxNF|swu^LRs9!m!7m@Dh)ygrFfI=4O|9c*<P)H5Xi#8s-Cbd<GJ!Kac?
z075M6pRQD8jB7~OPl_IDky|4@A{DP+!&C*1?x6&Wsr0(zQ>AO)oz2h_{S^$T7sy1e
zSHzfY>qsDZ&=)j$N$a%rEYzd2SyGW(^>^n}aUx>+<tph&g*=S$NsziM?ZTc;DK?Nc
zSKT39>haSJjG_ohhvG{khqTHzD$EibV&-QOIWT~l<?OSkXH?7wyX~o{arl!llO^Js
zXg2gd{d1-m!@%w?aT=?>G52Vzn^nxI$M9wX9)3wCGix>bC_I8nET6_iyLb^e1;^I>
zEOfbENWJA)j`EfCm`TcmZ$wq;?q_Z3e3CrTUUx=)!`HgwYcv-DVdX5L!Q*3YnN+Rt
zFPe99aZ(JTz$|1lvO?8dCwLB1kAfITQGY?}G>+F1rwFmKUpK^gM@KZ`#&7F?f~pE0
z#>+O};4<<e@lg<iX|Pf}40{0e&HZ+v&+o!Gb$bmEWw)AP@MmhZ)L9x4AIIV&^<=on
zF;j%(X7YCaQ0~_FoRP7&bIK}tzOFNAg+_W#PRWAp`A@SrQ-F@XzWLV`a;+UjNEOe4
zb@$yO<6_T<^PbMi0WRKWUHzS2J$t8zVVo-I6kR1wQUZuo`7CkeGgmL{8(myXb?<oc
zW*4~PymV@?)ajIbh%lX?<MCM4+~VdO)6Dv9Jm@AjPpb<1y;-PWtiHl#1RY#p{}mcI
zgcVnj4v#~gAvhjFcN%NzSRA#itFS&gciI}eR`FM0{0ml+91CR45Eh~rOGYBlFlif5
z0FmDQ?wRq~RDU;P%?(IK*-+6k$<FDIP%!f~t!*&s^|pT?o99zHQ$?Xem5I&WDirxO
z^n1<k;4JU}r2XE`l$47{lUIQoh17sw{O$2pC7KBcD^jS3xbJvb2m=Gd=b9_mbn<NG
z#I0Oz{9zfh&{ndYEa`}}3ZtO57nZt9^~Nh2QmxE<>`-rsUag<mgeJq!yS*B-_I@*B
zb4WF6!tV~dGQj9**2v7+IqcxT=C${xs^RUmJ5e!JfvAa6hD=7ZOp#4s6Mj{c??9sX
zW?=h$b7}H@F5TMw%fG1^Q@`4uPy>!bxh$smFM%L5x!~^l9~wTt55lZ@%!+E)!Lkt=
z#Om6?@fO--ToH3=jyJwi@kxv%t#eG$rzh0P$vXpTs6toy?*g2Cj^_R_@E&mr@<>sO
zn`##8mQz_o$70s=C{3SD9o0}YQ7GUdF(fRiyrjN5z@Jh7Xt`&Zkz>OtAKLF2W65GU
zNm7tS`%Yas3(~hIaw7xi1rF|>1?qdLe$@H$T^PNfK#_wdX*WcTJ~&YW>Iy_w1LSz#
zAz!u$Mu(BTZabq`ed-~>Y~VO)lhzGLh}DlHigQynxvZ|Y%ttBytewRLc>({B#RaIF
zd%5)Mt;hafofh>!otE(VyiI5Q*O+l0N=962ywuMQ%y>)TF-s^`%6X<31*9*}?U-hi
zmZL2E(>m%`J7Sj7-k7EX<=<O?`+a_I<(3%W*{*nHW@OkPL{ic*=Pgzy5g@zpN)k{Z
z(M);<I*?~~?~^9QhHH}1A!{_rt5!6$FZF=YvWY4KxZSBkwkDo#yM`a8ao|KnuG%(h
z7D{I~sTs)-t{yH1haUcCMwDSxKHJY<mJp5mkmqg%OUlsrqt2ZWsqSTRFC@x!C&GmN
zc*b@D!8lqB8q-jGf4$0c%#hdQe!oz5Ly-ah&(@)@=bv_aCrWUg5diIBBuyMHVwO`7
z49cA7>a9z%vd$HZh+>{Tp;1j;DINP=m(HB!uG8zib|%v7?^vADbpt4(tW>(#=gJk$
z6_k5>`p$HVUmS{pmV`Oq<{8qWHNYPMMxsHY#H(-f42$s!nJJy1q#;Mqf+kL-y9i0|
z@x#QLI)`%~LjEMbc=JdO2ypuH;Jw;$>&<a1yu)kBscWuQkGXD}A*5Tb5l%cKi`X`7
z{ewhORni1g7gDd77t?Rdp==sMp``X9rej<Fv^C=N>053$|5QhN-QCUUm;dJ6=F@9W
z35&#ZqTr#T1f+QUWAgd>k~h0p?>jGq;+Pe&B{AqJYOw<4;?jf}0wSl21+h8GixR{s
zuMKi{*5q_@Kt*c9<o)te`nfa?p&8{AOHe+dm=}N-kEgU1d^M`7i4XvX7HGDjOjLCn
zO`WjXutOXy!(2@&HH?t7M=gg0bKD#ja#o-a`LpMj@Z$Atx4E7|#j<Lf?v0?8V6EG<
zMlJW20(tUCtT(7}mhedarb%Oc6fIg=vC>t~IGbvmMaRw&KiY*5*)uh<)GUjGo&oS<
z1vW$T(_N$pOi$+k&tDCD!1$}wy|@TcN=7KcVCSVHtae_ub>McQw!|@-OT9{4Grb<h
z8G8F<qF5M_t}(LRmQ{?8VsX#_2CBGMnV#yNdBY!aW*I{e%0Z0Xa+*(y;UAbeNbM4d
z#k$t$$guV+KdM1_(+zb>WHH>_y=`B4yL@fdF*1@m%F=2Wv@4Y$90*obY;GT*Cj)6#
z2wf}>@n3Z(W%(Z4M7shoU~<fENUo-u?P|hEc*)h;r@XIF0$gz`R&{FePiW<%`%UP!
zM3$+)R&xv{k_=<W?2pdr9LW<n<!n~6#7V$XtD!pAk3GO;zhz&ihIzTt!yCV~ze<|y
zKRiK_azU2ZjzMfB*=drnX}-2=9?06t(X21VUg({-W9-4uYl%?A)CfV<@OFoRmHtuh
zly$?jlhujpZvy5Z>|hF~b8k8nPTii5%jD4*hpMRU)T(L;MX0JxJqN9L4uJ~?FLy1i
z=2uo&_XhD&#r#DzlnyQh2rKygt|?&4H)8m)Sv?4Wm_w`9uRyR3S|y^YrhAT4;en1L
zx72yxiY}KWMvYr{wOT%iw&LdXAYqm(Sc@G-ri2O!<FMc?X1(4#Erx0<d#xR}2W3mG
zZHcGeR33Fi{-qiQ-3D+=2uLaRqQm*gV$m%S)F>EIx`D%_;3b1<pR1)JcQ~z*?r6U(
zRI!zUQS=K!@ekp3g)uNK3B%44a=9p4IJqP$g^6~j7D9e4iad@a$Ls+^Y}VTt=fL-!
z_c$Hn24CL`y3DniOGpSzy71lk{8?h-AT9mQ2E!e|9+XRt<ahq_Li_C_f~1Mfv+|H4
z=nY*#?yRXyS}PY#F`|3L(7M+)iW><Ed^DdP>07{Q?90cFc8<zuF;k$K21^BGKBynA
z@F^S&rry()c-!aQ><#`Ew(QSTd8yr(*G_A{EPm*vg*$~gM*^b2NK~l304Ef<qzmiX
z#+`fD(BsJj<zY;{8Vwsr6ETV3S^QIfAeT0EEGOa=dH^mC9gSMbqjM2()Vc+3z8Jql
zq?1EYIgrQBIslPi`BLk0Ow9<0_@ZPPw<c4>xbLIrdP2M9)BPgHFM8&ch$Pe|WU6UQ
zyHr&y<oc2lE|&%CNq>ag8KmRT%Tz$D98IUkq(?_a>C9W_4pU|IVu@M_MhtycRbyZD
z@MqythZKby=%9UcuABjfNRtGCh#9!mg0(m1oyim^5(<QDv|Hy?>PVN)4F~>ba(?%e
zALhCJWW)OK2glCuND9e}mR!oFR$x+j3jp96I=opmzZ>Q#$?-xdr>u0;rj|!@h@<)7
zYIvH1n*@Uc%fa#?^muC8V=&9OlUXdir6U)&uluK`oP;ljBLjCB1}1#&RKGL@%j+X=
zST}#(_KjL4GIohmEtRl$$ITR6<Z}LT=sN8<dJV(MaBL({#u?;^6hhXceXF~ta-%yI
zj+08S5MR}<*GhTK-L2Hw;$87!eknhK|LnTTUf*UVs2g3DOTiGk-^6A?aBs4$TQ?~A
zxc$q(HkW*zXlrQww*Klx`wi`1?8U!?(HzfH6Ulg|>!dfO4Q<kn2Lj^B79*B8_p|Uk
zWPtvsqf-6HJ3dd3I;oxCHd&Dl;wf^8ob+2$N#iM4d5xQzWuO4@miVd|gKDFWo^g`7
z7N;wmv}!|BGicI=gKV|4Lm=1KjL=OFRiiHUY>RXerFl_VBa+J^m4Xad0I?XIgd;_Z
z%!(z`9lrk(tqA|M?>++%<5@M|sJ`mB@g1>$Rh`#ZNjKq05Z#2&b->ne(~#)S6m<Rj
zn8%vgY69JutraIQp0jS6u#s2$DE^C2%yu=2Hb9k^v7fw{O0<9u|IfYS#b5dAfAN|;
zOVPLrEs%+YcmC;PBu6bK^6ZXHl+=qvnBHGY%srfwNQ~~W==|M?IfTQ~wXelf>5wS!
zSV7k6*R|M;2sGUx%dMRQAqw^vhetK%G4{SGS5<=6X4kgnD0wPHQpz2x=96!5%G0%c
zKY!343PqFql$yAl3R^!|1jC8oeDLQ~X5v1%^ZQ`v8N3h8Pd|P!M@UveGbl$aCb(|a
zKEjtf;K2i^=AdLT8ClW2L>-$x8;QK>gwV=U)#9RJXtYLQiG9d370W#DXq?q9W#A&~
zH$Us<kBZ0IAO$+8k)KHNS7m76b@2T+UGprACW|r#s3ev*lxp9t7gqXh(Ih)4qFEbb
zYcFs61Ask)M`tR6o`J1&xuSd};m|9m3wpB?Km_JlswfqWAPF+-H{S#{{|g156Tvm=
zG4*sGU{Z`a4j7A$TVAY5*XpC$%T_`)^dowtFSBTn02f5YN_NFiTieQ+6L%t73AC-`
zAuZT5Rp#P(s<ThS&F9#6+OB7^mI6I9VM2vm<4~Tggf)1-K&@n2so%Lr52k;Ck~Wf&
zNUrjn2op~2OqmJaKL@_mC~dg1QlgRqO)U<)_pj^o$7n>8>(RPb;(w_eecvC9rL*bH
zx|IXN8JYd|b<modi8`FgK9VL6z#vDo@LDPdV`)_Ans>tDCh2&FXK<!?fdk|G3*s^m
zzfFIs`dBf>dym`Mvx}zZEmqtkCl$-nHq>Vpl!C^9icv|1kjyZ42~t7wPTnWHCMqU=
zUirZcDDtQPl<X6%=J)n9Z?J}&^5LKJd75OwArNR?b+8VjWX5sJp3zEF=2@%wneHwA
z-e~N^{iAiXXIlFY9lg^r;YgI7{kU^qJmv2152WsG?ZuikuKaIc%fe)UJjI{fNfYf$
zSGw{MJb@kcPUWX<;r^w6DP7O%uW=!ks%~-t0zXgRRGjeG=^egjb6h4hvF_e=l};%N
zMWDHB_EJp;?8kyL07G03ba>irR16YNpf!Wre{87x+S8o-hc;wnta^-4IR`R`yf^Xd
zaI9wf0CTL94W7VD+32S6#Whtc);M}+9HVBYlGmE26kS{_p`^09#%bpc@2-$SOgmq#
zB1v!L&N8lH9A8!Yx^N0+taKpI<mRM=ivVl$CGod!$UlaBWFb_4d!1s=<~C8(JGipd
znXP@}nTi-oADwQbO?`4j<gPU~-(E*paTE5ZjdMS)0pDO>O_g?3E!C1Kt6fj)>Hm?-
z$xKaG^oFx&fykDxArt3Ze1yI`?ehi4w*^wIHo$<S&!U7y__kec39Dgt4IbM<wDt6~
zF;5x`#-Njy^J|O2zGS_lxrN1*qA6=*#hwWDq1)qkzM*G6H-g8P-sVyVqb}VJDnAc1
z-+TNF2(xKkVrFIwO4k48kkG_uSkm6EkV8$5dZm^0p+bz=ESUS7+b!`N@00u6^ZZx8
zE?>MU)rv~M^%$;aZ;x(%H;?}8w<)5Gcdt;p+(u)serWIxI|JrBs2OJ$A6p1JmI{a!
ziR16?-rkD*a+W1>+PoxU?YtuXvwikQX<FQ@DL}5()pHu5f<<6_%F~ch+QeZnnPwqr
zTcMWFfD};k7lKK<?THo`&1N4oZj_ehs3&&XZlzmj^$Dt~a5u&YlMWOm&!$vKx^)ti
zXiKZk7cmBCH2SN709jvMzVfzs-Qt`K3bkjG9nOZg;5#30Gx%)~Jb{fKHVGD%=P4EF
z8s<L9$^nu6@6Y})W>>w<{rx*kTlcffz)P6O9W~vl!hu_R<!W1O13m+dCkI-^{;Q#H
z*Nals%XhDv?bb}74P!DQ=gb!uETU)aHu;S|&SC`;c6QUE*3QT~-njFjQJg%13^BHK
zjpOG#9y69T#6XkQ+z{V8?glOl?$YPJHfLiqq<Q|tZCFXn8Y!z)FQBlAVvO_BTq7z+
z`+BDhn8iG&2L5lSM82_~XrY~{j&4+jMgU|FU{wM+pQ4*zIn5IYa=M_^V%I7N7DdO<
zo$l~T5>go59u-wo!zW8)>y)cvthEdVJ8NtKt(0GvPKW<?_TLaGT2h+VUbAqw3ZE+8
zM3GlRm%t3I>IwJ1t1#O2|I<{XQZ?}71KzV7JTs&I9rnGRU1Z#)BBU|3DF*Sqo!!NC
zamBLk675{*+V9aWORHvynz;@2+y}y;I5w){f%JA^>r+jnu<po+D?@O)X-$ub=ZQCm
z2L<XP`iB^)!<cKo^PL?0a5<~P0(u8HHVOyRkn+XB#1lp5_0?KYL%Rei%V`{8E1%5_
zp+wGJl233)Qb><LjB9Mc<cq$e{}jvfE&SLEVJR!I3d0s`cqpIDpB^4by)W^D>^q}t
zaWU|t#5RM;#jyc$Zr)M9cF|^cG39j&@{Xm+f+H)=zKsmtv6L*yXr~*ITLqNnJd(?3
zX|=d+(P-IM6*wzZ5tCb#oAruTzih2Tk;i51fy^Ac$N=hs)_E5RIfWVOq~q~ESJJ=V
zAH^W&LuKo{qm)aG-$Tv4{XWF+emdW;%W6=(19%mJi^G7|%aGL|djx`;R*ALAr98Lf
z@2Se*r5OhL(!&nthNfNIy0R*q>HWtqakSHQwvpd>V12}V&uUbpb`pRpSRT#5WJ$~6
z=Ezs)Br=7|K3H(EE&8_}{Ouo>f?xF6m}q*gehUJaR}I#czP7qL{}?GKxK64SSce5g
zm6r!6nl}%@mk82DoZpD41k^^VAgXB&$2_a`GbbN$hwz?r=zv+Aa4c6K$y|-|@w_T8
z-y9JpnswDwt7T2{ysm<wbviX3l6#eb8AD~aqI?9IMf0B1#HpQrK99YjM>aiBB_L1j
zR!*5zB$NXcD@d)F#!jkyK9DoR>8uf%oF)&fCExhWhO>9M1d{{C&dvLu?uwF8<nuLD
zbWdRMR{Tgk)4e$FeAoG-dG)&-GjzeKF`DJpieHQ+Z(qxo`=*^3UPZ%+FW$ENP12Bl
z-|4!4x61OtRBcTUjTse^AD)&5xej>2o`7m~b4>uHQ3E7{TEw02i{G<vA?^UM<wBz0
z_?*l+O2y))MiH!$1E~(1h|a&8Hd0T#9Ev3~v9ZY!dMB5ZRlp7yJ-2iK4<<lmOhom>
zOh(U~KThGkF9jj-JGP!ifC4`>$%Cb=p`MT9?>%pu(r(Ywy2MIn7T%m<9P?2#*Bn(J
zxtv9l2aA>=P?(u(TR0vNI+DJ7f*qeNbgWdxFqj>}bra=0`NZpM9fIj{JwgXU3T@rp
z2xFx!s3hNOtsYnT`hUFX-@1Wc`TgY%?YhZ$?q|75=t=P92mA<$R2a$Cp1U@)jR0rv
z*zrP!7yFd*W3N?tFQZ5=YgY4`TO)8HK|!78oC^a4w=nxBTB+3YyUdytN4u)|)#KWw
zqhED#zKMG|kL-@c;)Sya2czoJdDj(|l`IoPvD2<Lbe;&}qmO&1D-Z3P^pRQ89nyrs
z+abFHX-4b2hb)~BDK$w}QVZh?{8@Dnvm1lbVk87{3e9j0RepkZj3rjP;T4+a-@3Ow
zb*4BYdY0s#uY_Y=<p@hCl8B`2(*@C|rpEpZcqP>L(97Ak5l+d*p2R@~Yjw{CU$5R6
z()iZ=H*^p!RHm{JCDiCC$Kga`5Km4Hc2u;m%%OU<8dgOwx*l9i?~YDEnvaYofi`D{
zZEE9gk8S9Y6S9HDsw5lg+bimHo6QH3uI*?OOT5<McZx2}xuq9*obiudXqAf16_PL1
zMAm_oI;-#!yWo(9dn><YuWLC|MR1S9Pdn8L8sSRCw?yyzWY{&qRP*?0cJ`y8*MVjw
zE0bwfHhm^<%gCNUIXS>l3?PQeL={hQC5>xfU2V|WyEC_D-ZImpV=%a<q;q9TyHYS_
zt|{SP+qZb9<mTQMok#%~nJO&xI1ieb{`{G04A{uKP8(JlQQ<MsyH+vDACB`EN&Wku
zdjQu{EYGx;oVm(kZ?xs1|B(m<QC9Bahoj}0U__Fae_wVJiZ&>4a@&QqOU~!fqF?XN
zX9lz^9cu^4SVCJjkKfBgtVVu8E<#01M4XFc?E7x>3kN^T-#vmw?n6XixA(72JT1x_
zBk4%^r<yo5osfLlu)5sA40Fe;#VELVQUI|kt)1V$r~n%+ju*s&?U~=sM~{w=T@=Pv
zQ|FxAu}A?Fj^ZY&kbDPO6vV+#JnVeJ5{0wk05O5r4PX6Um;ozi0X-{fHfBUIGRrNF
z*;g^Hyu7>&nIK?lS}X^ba2zk1*=467oUq}IcUN=4-YUs@$6t{U_RM$yX(Gp8ZflU>
zl{tMI7AC<+k`#OKRcmXz;hmG?M~vpo<V-9{Bu>P-#_!Q~xGNq3k;8QXo68ZN^*!m7
z>6<;T-VRy|t!Enj0lvB4e^=}Eh$YY1czj;W<{J~|w{3RWCd{hu;p;DvUM5x^Z$Ts1
zT77<5_o8!|+;1Pv!c)45c*}uQ@T<60dtdRGJr`d`H?*c>MCR=|x3XfgKU=eg;;PJW
z#C1R&YE9P@aNrfNfSnT4j5vliju|$!u&_Ho&^19Pax3>`Hr=ZuKqHY`X<=nW3>suh
zzj%TV5UZK%BUJjlZJt4)#D%1)^^?$yKGM{PsM6KCiXk$vU^0pG>i^X-v-)4~d6+up
z(1%9^mNlk^FtKRRVN;?p)WWCUdL^fn<`|zat%+sUQ@O5-4m$jKp?q=7WmS&y8DRi2
zUN*p^^*M5D3v0nHPUf}n&=tnodJ~_2H}cKeL{l@H!}c*hP<VTHoFoh4mw12HmD_ZM
z#mn*zgcILe&|7A!XwAGRkVVg8CB}aq{H6c_2CQ^;xkp|c#~-_hlB6_;T4gJi>bH1g
z(~10qq<laP>k$AcE7Sg2toO1!2L#67!$dxT6NQu(POhhP*_EAFJ@>?QN`y&gY#VQw
z3<kF-*t+g0k4TPC#Ik2fPfHhU&1mbqpky}FR-R8pj3zqr9R&7lGM@pByxSQxY|QXv
z^r|yWq5tDG|F_$C<@Y(Cj^SImixC;nv$vB8V^AlU-XH`JP=`hq#_ao(BC^W$u13E<
zBCvdmCFf~4Da*r{h|Y-}|07Gj+CIJM2d~dBh75B2w<updO-(d-oWCrwDrxDc8#wRf
z=3_Hlf8w%2H|x1pj_YxVhNC)}swz10wN8HU<x?KtzKzgxpFmo}MMTNp+%CuE{=kxo
zzb6|inALUmwpzbkJT->BJilSq<aU2*nheULwW#Ka5N68O&zACx_CbtN5T1eFULjR&
zGppvlyE`)7aatSB65VnNlzs7rjt-@UsojkiZ)D2DXG&qthnL7jHu4}8YMb8cUB&>}
z>vvq<bSX~sbz)F;EmWfSIfkoYNDTi)6F)1WAKZtJx*phCx;D$0VF}vEyprMVkD8Ks
zAgpudLf+Zk@9%sA-QAiYT^c(PdoXpVw007?6njQ8hVoH8xNUaR4|$HQ@_aO{bL7^!
zw8iIuQ4Vq3E9LeY%(vRwKXP%C9~JBd{k(91pMAorOOo8ytl+YMj5a#xYkA(0-eP~&
z<~(K@Fd=jSekJlip2oUt>D3m)g-m>fc9YF7(}BZzN>JJP@G`bpB*Ko(bkx!Ip2K&0
zWm}1%8=#<S4TL(Dkh@Z?dTnV=n~SE#Sg+d5d}wmOiY!<^-x>QpF`j@a^BtzCTA$b6
zK8nHCIc&_HH?u&S{4*){;kUWN%F5MQEA)oZ?%MChk9x-IKSlF@sF|?LLZaoio~1vB
zQAHKa9J=b@lFP+Y8V*J|1oJ{fB<nYN8&pQWLGOg!MNIaeqhg2w8@R<Rm1+Wv>BH&A
zhU5GEzT}RXt(Y%*zJBPm1TQaKwz9+k`g50wt+4iC+v@GOS)2%La}zT~MHL;nK!u{H
z5YDVnY|+A3)Xd!;!b}2WEC-1r1+oLtky@N~h)qDaQWOmh>IgzVyCqOe!dWr<qU@*t
z#Q019f@1_1z74?Ewn7L4BjuYDWfIgsa9cS_Z&g}8%d;fU(@)BwF73#xt(UydWI+AW
z3eC7PYo~KYjo(S;!%EkmR~|>>b7j(j<5v}xtDXJEwE=M7N`=~LW)?Q46l&eH(Gog~
zDR^?n$5YMl8_NIY?#_SckS4fdvKegX77k9Az76>tTNWO{Q>CF(MTwHZgCaG0+%oi_
zC~G@r>8lk~&whBSmTi-;gNie0Bh5^`@qjdPrRVdc5oNU6+V?93_uKEFIqNxhxfw(&
z<t#IcTL6KPXV6I#OOE$p-*W?~qpNS}4JG*qT%4IvoPbfHE_XU-Lf@aQGfO&3f&kH9
zAzL7%Xh*QrV78#1&|^<rd}*kcFztKSX$odmR_BvZ*Tz|W7*5W7ht%~0?-rB2>YvHd
zc&?#%-tOg@2at@-m0nZ>aUsw98lw`CL#M~*fs|~Y;2NKkiifQ~SGf-7+cvL#eu@ps
zDc2aSLV|h-1E@79oEx02t<VEOS$z<1ciRXK!mY`W9vx8fP)vtBksuM@xnci*5xvF@
zMyl)N!}%BIsJ#m`u|E59;o_A|`vs!>FT@KJVc$K%EV!ssL;H1-K`8PjO*BU@BLY8_
z;i!VqwMYOo;&O*GEOx7hKQCR7xv&-l+&;TNW3r)q2d6F>?fetB&NmH?R#8ju?SE}c
z#Hr?HSUQ_UmF9ZJN_&5{_u10>c7A?yaD|3qB3g;`Y=+LC{h+eVuO>Uqy>-4*@#E;F
z-)2&7Ax<+9mSCeUp0qgE@wFpby{cGhbi8YAYkn`N`3ddtp7`pAl^lUG2t-cHpNSTh
zo=l}jL4<i%S-n~?vp}t+*7#Mw-7841*}kD+#*CZTB$~nO;A`Xej|-l4C>?X)sne@o
zh9t(Dr_)ut8E=cpGov}jzcLi^MUdwqGXzB(ee=81y&&6^ZUz;^|Fyo(;88S<N}}%D
zJYjiNHIr;AFqURjUR4OasN9Hyci7!DsR<Zc!z}a>alBT%$kcB2*wMepqq?AlmgVY2
z?k;>_PWx!tMEHP%)FmUwfELL3?dkl8nT)^Js_Q4$v!ESkYA~FU^y;GH{`KyrpuKTr
z;ess};iZmV*Ek$bs|NTyPVA_WKg`2tw`^=;63}{!Hb|`Xp06xa>DhBXQY-RoEtjgo
zlK(NO^{f>@7OA~|8L77Zp>jCqZ~;+NX43~DzvmVE=Jv?gzJisV{qeN@%<r`855?Q-
zx_3^$vZDyqe?DAHT<jOdIo-pEq3)+DU{qbG*W+ur&L+fCNO|}?+O9^1bm)eif$7t!
zW@v}`*qhal_s0K?pUDeZ`>+x5z;N_ILE&njGpW})x}r)fd9I<8#CRmBg2I%7v;;i$
zes9ufQO0SnLWqhI_apnM84fJN7JG#ptCZb8MwMrvsR_Kj>@iDcE;f;Qcp)Rv>9L2V
zP<m?RurmGTG(LLpIUoTfEE?KoCve6;`}Pk?ofkW77wfq1oVG7*ipKcce1fmPlfLs?
z527z_aXTS~LZRHmuBVl&rY0s4%|`x241@h9%4Npfgy(k^_CLR?zuAM`VLeu9n%9qH
z9CAwU4I^n3#<~W2EK~XKC_u9NGD>@BwR@*EKL7`V5Z~JyTKGa^TFd$0-YwH*WQ&X*
zA2YLkyA*1!F`^@K_1rC|BYY`SxzZX0k9GXq&{ij%YGrNqZ0g-`_?|NMJ|jCvLJCnU
zFa@iWZk#xKsrha#F6(bK_0>Pdp+NwkTULgGv=_k3y1YIg64gJ^g#9&?${2s>+oT3C
zhC)hNKWZyR$u?(NSBJcEFx^iQ{)Y0?;SaDXX%dN2)dDpDPoSc<qppV`<tM>wgAF5t
zMbjTyQjx&S6{SeiPcW{*!mL}}lqBbiN<*JaD8U+abOkp-nvRfHFnzH5s^3e{^W2T|
z=C;k^iOue_cOv6MLp?iBc}|^eP0ht&g?(+kLuh3Mk=xw-w`+koMX|HH@Xc*iMlo4(
zu@5gH-#vz5e1B-hYh(5JV;qPG!?-^C*{8vF(eeSmf5?dRj6N{8DjnEXJDr>_Eq6Kw
z<=#HLb3R?>@4TwoL!2B0p<xfjW^bPFDA>AxCB&4WCo--orUwGHI|bb)0@u_7J7RyF
zP5hVP3i$$g_lI0E3Iz)mqH@A3Cp({e&E2>+GOzg~aZ)WZxws65KBo`M>8sFGb5F^M
zr|v}GF*Ir661@jtD7Cc*sUgO=Rgl&)p*#;80AJ2hTeY++nKkz1dYMPquPyKJRg1db
za?dsM?PCdm2dBqyzBa$tA?8i5Zki0REEU`yoXz~E`;kLfc;NeY?s3bcRTx&2R*H1T
zD1@lcJjBnx+kxF&NLfc)K&t3vhn~Ho*%c=XUc9}IzMkoMqpjjTrZ_a3RrXF;dMOjU
zc5i|2e<vbd0k9(L{ZRia{`4CEjo<ZFmarf2J@s;{Z&<UTcR=>Io^9ny|A7^Hu%vRL
znoJ=@-xNTs-#YmxY#>U%(<_GVwSQuq|8FXCas~BkTX&PQSi;u1kI<8Q$AXLhoRRrQ
z{Zu6|jwtS?=t*^+0Ui;qtnM7nbq+1$-lAJ7_4>G{X@j>_CD%tgIRimNOq;_L5t9+p
z(%7DPfWq`&FxDwslG4%YaYs@!nntS<LigrLX;3M{8<DBq2mK`fhiT1ouaMS*3tLyd
zc_O#gMDVh_TLzJfPBay<YE0{g!ec>oRhK!guW+rANF*~<X+|(wzIY%xt@I>&YFSrL
zO9Skr$`uE%wm|$#wM^k+|2`%ajZ`8CEs}D=5gw0+epLl6J0VFA;rcWnpwy(h(dlfJ
zo^9uRJ1Q7GjlE*{YBEd6Gb@QMjw<*cnsj8EG<b6{Mo;L^14d<K7I`eA;=2h@h`Y9)
z-p*hZ33nz-=oW`yHkM9meo;DoKXrq1@7uY%|2ZsS`9n^D8BYSEq|$W|J&L>b*Pw-E
z+;`^>`LQf$+lYBQZMI%P8ed#^V%3+8gDX&e?CEWbeZs^X#own6fJka939yDAu{63i
zS`rYQxkME83-nE)JuGjHl{MR+@94F2A2l$4EnOhHS|GbzV07Y*Ox7;?aVKm`Zc{l_
z`JK)2_*loy4Og59G(?dP9C4PJUh2K#lYJh-ipH_CB+3AKely7DAZ%!yEYd=+D<KCE
zgKZ+Kv_^G4>f-7uaD~(JYGGqoIVzK96)dEiwyo^O%o$3eVV8ZXI{9emf28z0tz!Z{
z?|ytWWjC=NEiETQgC)8b{X(!mfBPZBISb-N3MFfsJU%_GpE<hd#b9%|J)iAZ4N00}
zaPR7T^WTM``zQ1kJ}AJ9v(-%*kJDQ>(0PG1Q{^0-PxX!~8DAUSu_IC?l3WE~CMRu{
z3?`>BDHi7>#i5U2;)aAZpfRTxh-%B|SE=-Kv~pm&bGm)?B1Fm6$teifBH@ECa3``?
zI(Rvv6NR`8Nx0ql-fe;NdGhNn-M$ijds{Tj;|y3-OO%Ps+~TMRO717*++vrX)Jl{o
z_;m0qTqYG9$4qqTk!E~IC=Q^+GsTd|=UK2;w5C=icMv3LDR`yf0H0o4*%&zUO@pSb
zQUjpDv%HLz8<C$}l50ELq=mmUd@GO}qN>GbK=Dkd>vRnhVj8{({5DB#Gic3Sv~<+4
zax{JRYI5<|`Ic|z(c$kolV_*waM{vP#wNYlPw!VuDD)%cDK^QYQ+*U}Sz7=HLPb@T
zkGvq&@Mx;!z&z}&<Zg(9A$XWwoWh~xI7%JNk4NE<d@)H}MY~8GDP)$s{iC$Btp9Ty
zX;mH0hfZJb@3W5L_TIr-iMk!I)-4%qRnY|V;6Y<&3AT~RR+^^IsZP|b>oADbK4}0*
zgGy=zUx<(FqF*tlxUB@cB%QL5Us#58AlO8<3@NxtmVxk9Sie{>2!E<Q=G18DM{M9T
z;u%>`FKxMmGaVn>;b?M?F>!sL_`ad<Dg6BB0qdI~3NA^3%RsxTkLMHqS96mv)9CEv
zoc2fBU=i@wmiv=?hX023oQEF*5b{A<X=dCGD6&N_%ii^fD-^i5K@8TJWpxB_ToY9g
zIIGDS5lwi=D}%KW1JsO}brMv|`(>#)F^Q}sqDK;x^Bur2kTjygvmsWvUK?hd6#L6V
zk(|ns$=9v?oa8-k>oSwW3WZ*~Zza%COcXQ}d?nDd;3=*t2IwnMA}T>mm@npkZRVsz
zPb%WOMk>-uY8r;_$va4C^F#0Z*OKx<O^v*6Vu!#Fgb6!Tn_k)M1pDMPo_iJ^)@UZg
z+)z#BthTr-v2IH%QPMMZr@1nX4`44z3g!$~uxG3r0=}6^y5$S1+qY{!z5hE%p3##U
zDtu(blrER0TYhZQ7UMivANr$6C&v{ZU6yG?DVESCSW8ljEL%YXj2%j$42Wj^1cd7r
zB|?rSJ(cqqAr!T`W=#}8M#&~xRdO*sMzI*LV(BUz2Y62q-J30vGu!%EE*H0U%nNX2
z@QS%4A5lKOmG)ALG`aY$W{Ag+ttZixm72o%cW-OEi4F)?4@4_%_&z_+EIo@hRhA^!
zp*u}P67(pPfKeh(oGh@0dG}zp<`*lkmb^fCsC~z8oMv~okr)eZ<%L$)hG5>h!L`z_
z5+eBj$nZb^U2k4JS7{<wT{$GPn-@>w{J(h}LbdTx>d3P`P^5@Ll?bI&tiKya$75|G
z1kv=WL=Qp8f#6xopp^|;RWJu}3_$WFgXDqb$2k>xB%<}k5<*NsiYP%|BkBgAxIhQW
z^<ACOt$*)L`tIN&*0MWhVEE_o7QZVrL50%W$=nAnp94K3nm(tJRr9^MC)A)S)&ib^
z!l9CgyG&~g3<YONJh8$lQwSjr3icqOICg%KcqS<}dT_i36JDg_AcSb9k{fhOD0oP|
z*bps^ZvX8U@Bi1<c}K(9b$?$s1W^Wq5xo!5qD8bsFJbf&y+s#9C%Wjp_fCl3i9Y)1
zo#;IXLUixszVGK<Px-Aivu4(;>z_H-K4)KN@9+0>76rtT5Nf39cg-Ok$^R>?c_+}J
z@T~<yf-VpZc~UEZIYpHFt>-pxqIswyx81we?&?CaGsmzh3~NZ_K8B=ni%DUFfJwCk
zG(D|o02)Rsxyv%kN3A^HCMMvAW9&Rb%^f){7;vGhrpvBZ=+{7!Zf#T4Wh%qMiD6Em
zUQio&fiE~io6(ZJen$Woam!M&yISP$2-7ZC75&!xnW$EQEMyp})1OqT{<>TVkSIYh
z%E^+VUCslFPO;^4K=ewm9P9moaUSim_EqA1V6GRNmMnuiN7mDa#fWmggN3&fioU|L
zj@va~wf~Y<Bx7E0I%VJSsu@cd#iAgrCN!2~3g-i_qg9_+#A;MKdS*U+B#kEy6;nyg
z4wegziLzp6EYkp35vzfX5(5SDxOm>RcqDd{2PhL-59Ky8OKMfB3P@|&L%%f6uPho@
zwFV_9+s>BMr*HpU>CO%8793@!AZ+rtO5!BOwsPe-ukScrt@Yew-ziGRv0tTGozF|c
zO)bUV+0}(qLN+7ll(~#RCW<n#>SC<e@pOTmI<Hl>Ze^qDt#B)zMAV0hK|cfYk6vU9
zFD$UPN+)l)HKN>ohXzyqACe)G?0Zl+l^Zr$Ff%5$U$5iw5A0l}!4U77OuWYD26@_F
zCf}_}kx8q4E>&Z7o}Jb5-Q~`YR8+`Hwin|1_zpkry|hJZI1sxK7&T~#!$ROkTi#nB
zDOLDtba_%mA&ztTd(Hr%zg+LH5n*wBnn1cSnCW%Ogqzs+L!yWqNNeCh<Q%B|!MkDZ
z#Nj+L`TbJD_CDkJ`8lGo8UYX81nq*;(_gloE~*z<=={2A4HO5U)XfTL;w=)J(UD2z
z2l`Zw88%9n<T6S)L^-O!J#O3mZrA>M9m@InB|Uxf+t>r|lIGW#6+P6_&;-MVuM>TY
z+}LF0E3^)2QWatHJU6bhjw)gz;C1IHhbSW*<xyOt_9!(%GUh<+621eQWj!YOTG{6M
zZlUlPv+|c^>areKlCcEp6eG_m3tQWjS{7*1GIOD&1PZcoj#Td{hqMYD3KK<#1{K3V
zmn8)Rk%^4&>D|_TxF_u_3ZR&$D@JhcLGo*>2X{P@Q%f0&c8yFt#Z1i6tsyPrDT4!e
zH1u$CNR%@#?ZS5k<^$fOq6wY6*p6c2wN-}y&N64yvgi}li0x157#%qmR46Snv@ek)
zF_l%S;1@z&s+S=hj6#OeAolde5IZ3PfDg<^)~*bfCA3jd1`7JchbmWpf4+x97^tRY
zy8qP$g=NSpDF^zRw(4tVAI(uAomxPOu`6aWJ(|j!!U@_aYXQt5zX>u)KKgy8=|jau
zRa^L+&UAA0X^Tx&+NT+^AUKyqo{Q4t$<$YM_n%$FEde4>sS-}=G466Mt#yYy;S?>~
zYFJ)^1Og8ri`D<yd6G$)Kmwge@~Yx^arO4U?;dA9UKO+ofwF-J--@$-UEBK1R}n-m
zWjzV!kYi!3Ai?m%rp>@nQAyM>3A1Wf)Upo1r*B=_TA>`^w$Wb;Ln)^95X#FI3}jBH
zL(#Gx8cUALf%;LS<yHmtXv@}d$}^O4e?K<t#}Wsrre`4&8-fAE(9+%!$S9A?4`A|%
zP5k9c1;zS>68PB$=%XiUgwuc;HGrjjb6}sit}Y@jP_90VgI7`oZlEgX=LBHs6htWn
zIeX%GywWoYjej?}^qc?_uW-7cW$)+b{}6BlTz7R>!j0&AkI15YH%!XkwJ8a|98EtA
z3sx#EN-J3qQ)2~g2A1yc6Q<#tjVuDUf*3#=AlF&hk3!V+Wea%#bUz?(8G&RLa-^q_
zer<Lyb9YU21gF`fbzPTH=?I*<>@{L{U(Z>jW6R0YC{_TRVTO_Ul}jxz?AKN|_l%r@
zkIV!1iFVx(^Ny_C_ADeX<s(IKkBCaWtcge3Oa=R!oEQni2&uuE2<@;}{N<fG@+Clb
zBbg8rXSQM^+)TtuR{pX^m~HNP?}-mlI_lp7+qJ)y2Y)P61FB4nuWbu)O2;@HH;Bd@
ze|;`?EbBXfh0>-J`)r7FGZA!Cvk@?35km0j7-Yl?H7u;wnZ?y;_U%kC-knW^zMQio
z<SrWEos=uUrUoF0J$dK?DzKe*)cA?`1S~Nqn_$*5<w|}oNhKh&&XG3S_Msv(ONQCe
ztU7#58bnO<p=*=MJ<Z^1_))Ny>W043=BR3gVw_?J?0Z1u42dXb2%`v8!6k+P3LR*X
zh}R(k^g&Jl73v7%=~2!_f}xbspxfs9S-RmD)G?v0k8iGHZll}v#XXsjxVmb8U#<L)
zQ_%AOjS$#Gy|(AQo6_S)shVwD(<ZD~@vg7%RY8ei25SEA=d(0j@5{MzfTPXYH@fLt
zE6=9E2w!a=_Zw33LkKHQtWl^c5?&!{1m_k?Mm#eZ4OvM$g53c|2qSjR3TRNv$L8Qb
zU@9okezhzKB!AeLCWyO*^L&^69JKi@Bx~#HOyyzL_tIj&<73eIm`~jBZ*O2S#Uio%
z4~XBhyh$2*I-AL$Qw|-OKnH<*vf`w5{0*_(JR~~s{;`ntTqR8gL!2@$ia&F~gdG>p
zOZj#UIz`x)_K=&fc?|uJ%Nxsfe$uaP$E%9}l`BQ7REi7GERbZ|7YkHOdEc%`J8S`B
zM`T>>*TWKY>Y+Mavq{wlpBcUs^dc#dYJDkfEGGNNH?irnFwDqi*<h6e1mMg0B?3Jn
zl!zq>IUBfANtn62!*WT)+LA*{QOZYEN~LB{{Hdu2RnH)~Qww<uWvbTvsXlu&-w`a4
zkFE^_63Xei>%E(AolbWc?2eq>Z!zcGcQ4pCJkqUrHj2Of<aLj6xSnM<Kio}uhbPbi
zc4#{%F(s8HlyogfvBFfd;u1`nsNlzvVxjW-;fT=CYK#2J-4*aW20xLP=hbD;W2PU2
z*6=b6LtEN8MBT12y^gF6c3W3Hv(ouB2hYG3D<_Lb|7i>|^l`|y<F3*q2VhIb6yRV`
zjHsbv$4ZDW68ej=z5yk8zps;|2(QJk7_yt51UaYCbjMU_h(}BFmLV_YBdd-ASb%~4
z-N2NIPv0H<tkchz6}|vR8TeXE$hpu4Ie5y3Ux%ZTETOBLmGWAri2EaSSn{rR9j%H<
zk?-qCO4M~Tyw;2sdHPfI?M!T<LeqpV*PoulNgp+LOp!P7=w4E1rNN}|U$B<wIb7Tb
zJ+~{A0S-%(-<4uQ_*{SB5eFztBoS;<8_S2xr!x3F<Wm)SUlLf2GS-_%N_ip{S@A+0
zIsi&ulT_`y7RzWux4m`gYtg&AY;z~4aC@Kg^t*iDtEbmHjo<nVd0`J6%lH4Cu(%I$
zery`o!30r<XaevHQ6O<m(EeZGuj3J|-o0*b>R$zgOC)^U+5|_SBWtjj0^=g3)$&sG
z3V4mXf?arfWS8|Uwt86k`Qbn1ehUl^sYD642iAuj$<JvCy9avY`yWSN!{k?Ye>kLv
z?0y}$(j;(PAkbk6BJ`iI42=sDLk~n?G&Hq^MZ$5OAOI?qBde}iZstj0!-C#Bz%T9W
z56cEun{~y9<&}1Xvxi>C79#Lb3JObGxBD4a5GaTuTINZgonqB7w|?74On#@ZtS7P^
z4`y49-=r#LP2Kka;o;I+T3R~;v2+S!3i&8~XJ-UO#pWukNLb=%!_rY_PQDv&e_oII
zCfR#$Qrsmzu2-Ioo_M(uj)nMUfA#$L0>1CbAN-L;MW)vbsMrzvpjI^^u!vjtbwMkd
zV?=3hFFe-86DJsJeFFlQSUjRbvc*{g6i@?9YUmjvj8Sws;Sfm<83lhubbfwwcR|s?
zJEm+??~4A$jvK~%`abO@5BHEJE4RDX7TO4fpLT~cMfIUu=eq(y%FQtD+zCPHSkfO1
z>~aWhK$n<nV}Mm%=mv{bg`h@Mwqc!5syTq7_0_}r>b<7djniA@FPQYD>sP6gS#CGt
zEv>Khz3<60V@<}XKB$}N)dhd2rf#3L__-JkY-a5K+4b~Q&+~_5-_+XQuimszP%C58
zP>&KmP(7~r_8coE-=6q}MtfbwXNW!`sQ{6wO~vwGj9T~{6uWH<GF@Xnu2G$8J{&xT
zM2=SY+Dr2PS9o>M<-@fCUZBhEmZ(jJnn<-$rg76L`#UlZ3pxyjB89v=!5<xi0zR{2
zqv4sbYnNfsp(lnS$y-7KiGvpn<GHk5gDUMSI7M`d8{IRHk4Z#NL$Y(A#J^eubu$zC
zRHe1_Eo1TeZ3Y<Z+AeWU&9j|OSw}r^kFYb+BbW$Kb~_O!X<}B2>eI_7h+0^V8NNzl
z>9)xaJ@3zHWRo^5c6uFf(R=Au;p=H__;=U1aMF{X@Su}ZxUOqVAIa*&kac<GQ@2;_
zTF-G}*#R<G1)b}b&ntiW-n@Ne%X0fgrpAE6f)6*!?9=Hlbmf?!uyNX))%ChhH`wyK
zrP_QVdfxALwa$dKiRAYth<Imr=PLSf!?zdpsYs{qgTcK4!bv}LTa5l6L*}uAIb2%X
z0<rjr-fi03&>10xFQzCcgFQ(r4GNbc4CE50AN&Q1-+B|7G64s~?;;wf00#FY!eAQi
zuTb?D8v-<{VGZGKwftirE9B~56fo0=%S$gD|ETh!-Mllga4V>hNK<e61+$Ifcs_2Y
zcHKoSe)*Pjs{M640d<BXMI;vq{#LX3546~LoXDcbvoE2=B9Sjh5)3_zV$)^C^IgL&
zvb^F(1UH)Cy7uj(FE#Bp6gsQ;>G8>CjwSV5Pw_Ol<*BOdpyiM*3_>s>Y_h3CY4yFY
zG0W|j*Wowp8vMuOCl`CpLc)(LyU4E#38ySI7nT>-wl(#)_KQVEs2-aQ)}{w<W1xl%
z%rsZMAz7RDkC7el=MRM)Bi5B2pL-B|s${B=|4<bn+S^UaC_{Mc-2A-~c@;glA-$2I
z3egk!G_lHP7ubA?ERzI|U&P}`--su2eubKn$qF&5A&ck8OgS#}K92wlmWKG7DqHY6
z1s7*|#=w5y41R>)>|o^c8u=XKRJPqP{uaICd@Jmo=CrHncD_Aqs&W3SAO7J~6a1x>
zK=p`b)FY6K#DdYfk%Rz)8a0E<w6%4>G{P~hlX>Hs?gtLesCDJI-Baep;prZ9k%s|n
zm@JIe+{81TzvKQ;uhre}$!D`YFkW~}PTr@sabPnBxZ^Yu(FdWZ96xknb{0e(YxxlB
zkX8gF56kOp7WqJ|b41Ql`${MiSKsqU@?nqaI^OHS$u~u(Jvh?%%x0kH>8p{)-<9|3
z4;#K`;ceHmOC2Ux{~lzI4@N#~5?DxwU==QUIX(g*CSJVw7qN0{*pXrWvhfn)S>|t?
z6cZA(XvEP&sk^s>h2O?P5fK((mkfMO&&}G+J@iFyi05O+Zxpt~_>4?8feEcG1FviD
z<nPC{Y;@l)>QBz*H+8|H6Vz`%%$^iLM&T2&Nylzs>wAoj04eM8U@?hUh7y5J?9a?N
z+{0A?BJQ#Hp{%DPD4PUoXEY0m_TxvUzv|xTN){P6n7?&o;sLh0Hf1p4#E31__G=Gw
zvVX7FM6@e+E-Q3pWUuj!MVTf+A&!LBl-u$CrozU>JmwxnFt?1Ku8-dhb`NI0+@(Hl
z8W{RGejdO&TMw0{@<@4{^Zos}C3<)6>$mj9^?N$We+1OtADNmetr%^WcAhCFS5v?g
zi!8Fb;M*K}BVeA3`1k$w50PqI+!27p9_<i5j9iqw5{X#tFyZ_GXr2Hg89ndjr*T<q
zHm*0ymJObCMPK|2n`4-$w091UI7<0|o1utmnZv;z_Nd&z_R8Th*;7zB<)m<d%#$!$
zx-qDpQFUyuIg6tSxiy6EZ7H@m*FnCsp9V4O47tt|jQosLDTRu>oB>B#b7}sjw%Wdm
zm1iqmp;728l<)*4MCbXDX?MH`9Kh|K=Ul0jvwEbZ`Z2o!Lt$fHi-L=s1lYWGed3!=
zFMCZS`dBRJdyFiR%PF+mY>~bFcw_K5cenkx@tj9#l1hFw;eYo?807>pW})Bt<Sc>s
zy(qburuiu0l33y`OH5zY-$OX43*O}94otX5&6o+dIpjEMRge|4_=T3tNW>5_y|y9b
zc5Uf{MN)n}f*UR_d2PiC?M4Ig&Yzx*R_Yh2XVS3T1m^Z)G&XzWM`5GEA{<~)pkU7f
z0vI9(D02*&ofe+7JX6W;BsYyUd7{VxkwXVDEl`?*Q!~xvgkTM@1LmEJ$d}_fhH&j!
zPr3|r8g!Cjdv@7}3eje&n<@Lgoy$jllHh<p>=gOmO*o+Y3WCS3?5io^)2n!DNO$Mh
z{e&n?^V=Q6!$QYZ@*dUeO^Scz|GUP?TYi>~TPDO0k1Iw9tNEcJsnv@!BdVz-DJ86O
z2Y_|8PJTTjqR|jl0t%Q3E*Jo)eIg{ZRnTtUfZ)kPO>r>2%Q*mMp|U(hyUBx@G#X_6
z#T3&~O1WD?PtMt&xnH4eu4C*zbC222mlwX6CMuCJvKi8~*WXI+xWcAqQ@{+X8i>V?
zqt9KXFe*1IgPNH8q*l#H?XoQ7NY|5_OaXOGEfY5bkqD-aaAl?v7Jm_1zoaqn@l5=A
zMXMBC{#}P|4(|{QI)Ax%MI~A8Qt)|waYp@YIj#`hclYE$aryb3_A}pKSQ)j5lI+8z
z!HvAvwY7nTr^WAkgTQ~q!Na#d8_UP0zZwf}=VlwFB<|b=Sqw5(W;KNY)XIk7Tql`A
z<$7#-GTNB|t52cy;_Y&}qw!HN1j5h=!J?Hl0&hb4wvyoWOTs^Y-m3-R7S$-SZus1!
zr_=YiWVu}ep9)`sm2}SX))ph>+g#Sk8|})g+gD6H*3s<ni7JE+4jAEGvl<}?+-6=(
z-?ikyD9amAoCQ{XCM9_!IiI{d16&pR9kr<pca0W74lBF9L3>c9^|*e@M&WZ`DqSsn
z?2LO#+nZ=syeg8KX~)A^AxV`OQbFHC-|K8da=v?hws!y2q+<QFpK2E&pFn?ruioCC
zw5EUmm)gbG_fH|#S8rrn{NV~Ao$kFlFw2w6v;wUGVwCVp0K~xp3{`&JSbq43(guyy
zbf;k4_fJNR(phF`aubMWwmkBnSb=;HE;cx-BG<G8-Eqoc`s-#3>N8<i(RN+jZ?_1*
zLvuR}wn$>|7)g3<|CnyDq4{vi{@uNL$iX%!-s$(bA5S!Qrk?S~@7!|zHufiZ_z<`v
zm#C=0WxO~iT=h-k>v|oAw-_MyfpcIyac^#)wk%MMQ}Ru}49QsbTH#5Dc9et)_O_1~
z$^AP=0WeHEY^3Jf?D|>ZuwJuIYR75Eoy+lZTM(b1AI?>f=ykCrA9)7@(OP@>Ur`{6
zpdbIm8}Rl0Q&rH^Yai@;evfS-daDXFN(HenD;ptUzbPM(#==$a6B=&f&uN^<UD}Uf
zBI_m(g!W#rCYpy=SO0MC?``-YL<vmGTt=YTcyzT5i$288@atw={Ti#eUE|{T<{lKk
z%MouDw|bR0{ANKLh1f))bHCOpL&Pyb{&7QoWbp(+rr6zybctX|Nk{DNKLj!Z=J$^X
zb&=mo6z~LA<Ak6Zqe-e5IxAJMI?%ikS74G&edQj=HfL8ev0x^Lj#F;-%S84ayJTpq
zw^bl>wazXltfszM9wfyK^u49JV}GbMxDzGfC5{%kVSXxdg*4{-81&nP{jWUw==Mkc
z)P;Dj8vG$u6}(t>-&fx-53j9F+wJ(xIxJnBl9}CO^Mmsk*D@$B3%>vHBM7%2jXRIL
zUrm<_t`qTsK0i?3f|QiGh@3bKVwSsL=-tCPj$UsZL4<xGH&p#g-R`bv>z683e^qDU
z+y{=1c15l-o|ObO3WB%7p20=&dXF=e6Y~!kR06BU2}d#X-Z~ZJY;4jcGfP12Ax_zp
zkA#BcM8ssR=}Yt%?-yHgsxb0-%7`6^B|*J<M;~R=YnPiGBGc4BIH=YWq9T-Hfa=G2
z->4cB7Zm}%*@jY$*ZOYTo?hGhzVTze#}*~CWp}-A@6c9#*ZY3gYZv|<3(x-PzMTI?
zIH1Z5um)jC!>ew7iP6~`1CwA)gs-_(w5i97ZYP(BC&ERl1bkjwi)ehsDF<tr$i{O?
z#?lc0a%7x5+7bLMDq;*9Jl@sS1mGlr7Jd<u1j!QRo85H1>mk02g>$g?h7mkdc#v6E
z1%=~ojU*u1o@?GGDE0J7hNvzzdPFcPId0$32CeF-c9=v=k#`M5S_r|t($%!9xjE3y
zc;%7S#yk*~VVXdIEni<I6Xx&uWw0<@a}bKl+A9b$Mn=EW)obu?*LVT=ogaOeNSG%>
zfPt)aNu+YIm7kHX7I=X-;}PV0$=1<J^j3IqkjUa{URq?xm+HP_u;z|h>#^ehS4vmu
z5E1e~cYJz+agG{=MFgCs!N6*)T;5O}`!1cY!sNd0l4MZ^VOOLn>?Vs;!)8J!!yo|o
zo8$7Me0fsGm4bgzYg;y@=$u;$$GkgBq)<SQIRs)yM$lzAfm5GN0ra<bF3mpr95KKe
zg5zl>v6I5zl^rt<AF(IFryT{P7DD9;CBP%xFJFaBq^V3Aae|?HNMD|OY`8(F!q7$B
zc7R8h;7!NwCYeD@Z_`+nF91H5E}`JjbPAw&XvwrJ=DV5?{63yB661D!a)$0o8bHh2
z6Qyo7xp6B#OJ4Fo@lXf<yn(@*40`#&aIiHyTo5B5qGA-UyQ|}--e70!G4(Of$pX&F
zFWR5E^aZo);NRU(C#ZiW%daeoaTX2!maXO@Lb7K9(4KkZM!T&M_UBx!RiS*rVl2FP
zXaa|TJdgzryvuUj!h2G1LFaWEi7+SW_zBr$xgqayHTRM^Z*aE|d4)e_n)jql@`{R_
z&ZoNZ6qjiVg30>v2C-Iu^V%Rv9KfOCtZPVlh0FXRJLoC98eJ44)`Det;<b=#gWc4-
zC1VzO*;+2X{dhAnqqQc<-V5DB997(k_aCf5T9E{Wx4%M4{qX&p7X%+*kH17wf7Vy}
zMEc%SJf^vR)q~yX4iyIBPnWrv7g?mHXBX>~5{7;{bsKTuHFfetjm4hi?Gp-jrfFZ{
zs<#?R?l?R2T6Yp1UY_KZj(+BDVDI&>hl_vkW&va_ND|z2aYhEU?Jte1->nZ+6PXTz
zSWuml;&;gjSYYol;JN-t#LT9J{y=})F)nXuW@7g&u4hr8Xzt_6?inA2V*AhlTXXZ`
zQjKB)Q&a8Wdi)^&>r!mI&kp44;8_`fQy#V9^4zlaNmYG3rwq6!7DnQ)L$95IE8uAC
zr(-w|Dp+or&DBoK5dPrb#;hz8QB@E`ogC*v$=OujqxGpS$yk`Sf1QzIo{G^3|7Ysh
z9ewuo@@{Be=$QAq#nq2w(c!WCQ&G5~z}=B}$*1dPcMXlGs=B5(6&eUn$>wRw%*>4U
zm+zqPS#lbBS?KuTQ@}Q5*5iZgRciKqK0=wFfml2r7^eS6yIjK40RGKY6U?l-tgmZU
z*762~(c6IHS3o$uZ|52#2d$X<Y^~lc7Xgs_UEz>ZSE%!aj+As@it9ouaouph&2>QU
zT}%XeBpt5Q$}=PHumiQsurINUSng>w^~hjWWoqJ5tmgKH;fwTi=UW!Z^-ipfaV`T=
z$k?2=cDc%Xsl8s~3`y6P`OdR7qTTn9v1%4@MGkD-%K3PJRk1<J8zY}fE=Vo^)9L&M
zW3{Vv=T-v}bAN@*uv&f>)pa*j5%e>@L{0&%d~oQhk~MSBMZG}Lie)u#NAJ)Xc7?_h
zYe=S22t8Rm{cDT>5s57U92k%qxBYvJK%l+%?V;BVljv}4Vxp`1QrrJP0%KH?1^VYf
z*cPQh{ziNovoOlzmo8C|InL5!%Z4<xNOD#@6;?4hzZk;(PI5$FG9oU)Cd0(JxjNAU
zl>sR2*Tr1CU^itoGayA*rm8w1rjl+oQG;Up;^lpUt}Di}){>z~?b|f0GP&^x^dquC
zXExIY_52Ad{62z3OTOdssgp<ngvlWzOu)x0Y-bwh>jYz*+3t@{^?6y;A71Gf6&!R&
z4!N{r&~p5d5eU~TYH+|rYxhsQ(1MMv&qTr(Kc;RIIbh_7ZS0FMU+Ef?c72Pom7*0p
zo|!CTD=TahocC#jk_1SOdVUiC#=#_zbea~%eFmH6*U3Yd)bvbI0QBGIkZ<-Mty!|5
z^adD;t;WnU{wp5&<qAiv@*CEINjusE3A8w|MElCAj0r<AA*G=m&sjo3L&D)I{@zrS
zH-2jVO&GNJfiu_=LGu%Ud>9BzgQX#u1tE%FzZ|9P7A}0olC<bAHN~Z>Sx2E*R*xYL
zqDkns>@nmuws1v_)gd7*F4+&fn-(PRf2%4!9(}mS`?_D9TY`frX#oaANSEVjCo~Yf
zEpEK)t$xv-bW;6|jl?Hg*DEP4s3<h<zhV(}N*y%K26BbCK>qJus3yL*SFLej%weG$
zDG1MAg$4rK$Nen2bjC`Jl^DTtC0AwC*uY0ACl+7>;l4;uPe<sb#_is=R~8HY`=I>c
zCVC#=M9GAG@CvPts7;8>#7)}O1hCKNIWwf?7Oj#^QZfd{8hhBK!A5{Xf*<D^5tdTM
z$Q|BVwkH5x*qAMKj0zfapdUR&><*n&D|(OVz!JZ--h5RKJCivB(-?2+dwu{4DsBEu
z1{3z73~`J(R{P2*H^Br5;TM`I{iKsGIe%FFbD2rt=10f^;<--q+lBfD5^Duts`4^a
znba_X!3lpaBX=sa_l&LMVQerA&HAF1Us|gG;C^??gq4Rct^KX7x*FccMmpXx3P7BA
zVFbjNq!9;FqaYK0$Rt8TR#DTy5$qI@21^2{N1++)@_7|5zc9S{_-8*S(qg032tURS
zfBg?i*4-cf>Gmu>bt#rOJx!EE;Iwl;osMiYE!l@d80%aEVaq*Nzqyt0-5yzyxen+H
zqA6F?{>?8guGZr1R7Q4?Tqp}~(YTg%b-Yy?v$82&mzyv`Q2OHgF+k|G>`U$pq4$Cx
z3*ZW#spjo)hiSS99W6GQa|FelaZDAK|Lj61U9M2}MHE)yYxRQ^@n{c8f;Yr^^J|xh
z!wUSY83MPQv=KQPCc<EvCjkv4EMO&pjQWW92m4l0lCGgxIt1ShaqI(_l%uUL^X1{=
zZ1o4IT&GcmxV2&&qnfpPhLdFlR*wT!2?wuEvgelJ2j~Z!eiqW_@1#DbAX4*((=&#%
zJOP0Ie?$q_ALY!fAS+ab)L#r#=Tw)+YJ@+r&!neMb-FPlQXQ1-bHf{*kx3U}JUI~)
zdGW%S_`}_nY`2^Oeo&}D0cWtYFw9gC?w8;%R}n^p!cP2kmfD)}$MW*lm0ssypU`M2
zYovLARlT?la5`-Yp(<g-q;c@0k1#1utkdODW5Weg7ZfUo8E0@(Vrp$xk^BJ>kIMBD
zWnT&F$e8NowoWlY<Qgrb=$!qxR;4Dzp{94M4=0c8t9O@;Za;G|1su;nk2iDQWL#Z4
z?N4602v~5zILRXs=e#H=D#k_-(IY%+WRX%^?%yhb+mRcW?`nb^=~H)3?%I9DxQDd$
z^z#;MCKnb2|7TbbSs7p@2TjDYkauOArLEhpr+D?v#c4zV|K}iPkpK?`ohh13-N(nr
z%i|;ds6&yptKTEuu9s5Kwml^|AOrO5h15n<@KTeEpPH>+omgxma8K7y7~MfX`~lV7
z@)HR<c7fz7N}XglFlt4hl&-ZZPDUgD3*{-RN*hP=%i3BeK_?|D5GgK-yfMp*8fTHt
zHb>HcYdk-y&#6uV&U?&CGsV-`OAU_TZ2g3B+e&CJ!8amh+ZQE)E(NnTwk|R^-U(2J
zANKh6l5dO>3v&^*q>$29yqWS*AO0L|T>Pl$Ncu6=vo^#*Bb}uiAusBX%odzIxeC8h
zmx+9w&ugs5<u0Ck8zXmoCqLs@4Gw`EtXm>`?5=<)yc;iyT)#%o-u)Od+u7LxpgbKW
z>2-|GYr2&PlBAKH5^)3*^b@E6O|<!vf_+5Ahfperp(HPT?=3Gx`8(mGNZPpsNj-ae
zNT?MguFZ!Z`*`_B6`iJ*|9qUkl|ElNx@QGT$or8bOk!plY2yv+k|k>}1l6QiX@SXW
zPTACJ>Y~r&4^S1$4cfl`Sj4kbjQb_iV!I3_Xh>5d4-gO+)0Y#<WCsr$=S&nZKM51w
z9QgctEBleK(&xfDFd$%$n@L~WbbWbg`C$#*^Zw}<X5|L4MC_${EEK9Grz+v-bf@z>
zFYC-|3;ivnJkW2eIt*O^&1nFJZ0!01o>8IHnrEG+2G+RDG&~^8a@t!@vlN&q;Qv;?
zdnefvw-EZvgG;4t6b2JA(voJw|DkZ}9vfTieFl*%L%Cq{dod#yiqF^KdI`b2jCS1G
zGkUmOTK!h(^o?Tb?)%&TVh~RL8IZfj{Ohm5EoI7XNs{^!xnSt~_^q_&?Tn*tlW8Kb
zO|wi{0aUdDEkRqUBY6#7B<d(O5gjULgPDfvVx!&v!4|-}<u0CN#!G=KyjskR9VRT@
z{H#Y`rf~yGRzk;$EYZr_Uz}nS;kf&g#r*NQGR1+{?~ot?xt$qS=0<%BvFY>ry}?cQ
zHhXQTZ@8Sp)$NqcpwCvL>FT%@EbNi-l*9MaW=oxp;z0#Tqs4C3yC;F+aUWbst&-~|
zP2H6y4<{mpY;3;LnECAr042r2{q(XmPAiZn%FO5oq`HX!<gdF98@-NYGoKckwHcY1
zPKBeLqdn2ncAkd3w@XXkDQZT1i0<vQQXJS+)#Y{4YVxTV<PJsn7*9c7!MgoDES-*S
zD9y77)%*#(=3bM==_mO?@a2&G>qv9M{AUN4>n8}pAky!HNHsoQ%A(p-^IV%3!sb}+
ztx211j&%d`G3O7QL85$;Jm4HINB>;S$IwFl2frSmR!uJ<iM|$1V%Uknx6#p@!Jk_!
zQjN2F@W)Vl@1rr$D=t}1ZrJBBuc&j<=P{SJ(^fP+bhk}p8b7N}5xnZ-HJw=|JBBmr
zgc`KmQ{j+2nQEIV+jO6fXo$S<YL6o&qMySOd1FA=YHqP<xD=eddKapn5kvCQeyM8C
z!0WGRKn40|HiRsAX}-=pvAzkSp7Pi0(Lmu=XC~%6z=Bzkw1dy|VhSidwvpH?{h4w)
zRv5+zowHPXvHT3o+&-Y#yBY3Q;K^@#JZsp@sTX@6Q<%4+*d)+8FdB(RmS*Gg8Fj2<
zWIB~}zR4w|INLX=N~HIpwD@YTLj1CjjrR_}Qln4-Sio|A#DQsW2kG#>9NyE<P8a<q
z;bk4Ubl!Lhi*~r#e*Sa%7ccf_(!hvF998?E0t$ln9D}?G+dV@aL@HK(YA-7Vt&!!D
zWh!LM#HSCdzaNiBdNAJ)^CNv?fIy5;f8IB)!W_MovV<i;)1jV&S@2O_Ujc=GmI9}#
zJ8}Ni8RfYY5!u}ik!~oqtWvA$nkIVgxY~V0R~hYFdA(seMOuwnyD8yWL+jQ6lP6_S
zZ;Q|?P6^U_ZmV=$xETnBy0P_aB=E6ihi){6It)^_GZ)PGHo*G}O)Z9RTm?4J5J{r8
z#w|YRT2Lv|ZXF%DYa+f}PbX(zllcs-&YWjxKH^5<$I@8mxKFvb<V7diy>2?#_Qk-k
ze|GzCx-9o~LA_ydfFd1EHD!o?TpTevx!>0DQ|aSGjJ%Tm$qjBqFWyu;%JyhWb_&R{
zy@KGulgwDj0Q+SY%-q}(<_sP&3?5-~zn%4qu6s=W74_*m9mY>bU0+O7K5m>iw5Jn}
za;OXdGu>tQ^A9e`u%Dnj4eN%%XdJxtjB4wvqZ9}+4qqfxkqJ74PP#I)iRiEePP*xT
zD%2bI7OGk6pW|~smr5Vh96Os-8w~pGGhihc&R|t;V_wJK3K@CM>AA_;#JAaB%yJH<
z<Xbo(-Yb%#6O$#GHQA8Buc(ZN5uh_{PN{+q;Z7yQXX`r2si^rj7A!zQN*QH_rIlyz
z_0LY@KR>nqx|&<hC<~3FU%9Xd!dLVER0DaBAQI%K*fkE%@6Bh84fFcCA?~dg|35eQ
z#7y(`N@p*<YUKkXDc>VO{e<{hUIIN1j}>mGqr<Ylf{qLpaX*vxI}<~P7^t{eli$~?
zrmPV3y>GoPdIkB)+{Twm$tU`YJ36L3=c&o1XzJh215O;?UkVl0y<ZUQxnM;Qi;a!X
zK&zgcxOkCN+x4Q51%yLS-ymPbUztRX7+7OyM8`NDGrGD}*>Rl=sn;y^6xl2s-MQl)
z%gD%Ee`0LpbOduz%1EU-b}e;7{``s6ALvIST%|ck#p_2MWzwROoXW$47(*3c+|dy{
zt7GY}6eb<;_X2j^{=+U)NM4yzbYS2~@qXJm5v%v-d`e;d9G&=rQhJBnki1n*nN*`@
zk|PN&K4^apdx8ctJ$Jm8j<PAi!iTPODvl3!)rO`!#97|Q)N`V3ovZi5tB501kkXZT
ze^$^Wl#6L>W3GAzc;5zJt<+Zf5jroWL*BKIu+?MhSVDlZak0NW+Af#HYZ&#8UH8i?
zCVR}y@<(^A_=p0KU%lmgB`6t1tH|AwZi|75p_SU9I8(Ei6xLq3djE6U3OUQd=^hr6
zk($bg!?vugy?IbS?D=3fbvI%9cbI-7ExOKsU->6YUGiOo9Miijic&`y4o@C{xkfI-
z-L{q}UwZ4`wONVJ5?w{TPQc9QsOc`PE69||J9^fmp1cc3R3YNNoXg)$jp8{iP#fHB
zimpHXc6WQ%m)!wh61{&4E&TMU1!QXAmeJA}?LErn#Xc{pw$0{y!w@wFHaWh*<+Dc|
zlLvA9`3n?ASjd3qpR5!;Zh50;@Scd>UgfEGMvh!76`dZYnIDx#6(oH8e8JAOT2lj5
zBKQmPWYr&8emn`guV2?)SjfNAVq*`KgpaV**0y{I(P@W?ed+Q%$y_@B{Z}G-IJEft
zPy2`-TbJBk`F-0iL|4n2u)*QJysM>8w~ae=vN6M(pzK?p$Vf}fXTw|3s(r@PCOWry
zr!#nc;cgyGu0Tb$zGs=J+qlzIGrd!>?cY}epWZKa-{#j}*Y!PD{C+d|be}CK^6eD4
zamj~-VBgR#KO>AZnzT6A@9tD8M?WX+y^|gegmxc~46g4SGHDOvUvg+?#bEnsKGi<i
zUS>u8(IxbiSvL09a2!8>3t0Eqfl~>~151BLnVulT7AvZ!D_8u?GQ12Tiw0bj$Ypt!
zyR<gn%cYc=v336_EUi_)|8WkB^;u!u2s;FM`QC5jt>M2Gq{o|-oucc<(&leZ<>lq6
zp4RPL0iPkh{f=MY^^L%rx*L3{aL#nXIhx(JsXrPt`}^5NA{6rcxoYJcfyDA*f24Ha
zY#MWzfWsP6lG)KgZP9gc{QHM3uF-7aqU-owpY+u^WeZTS5#lAh_>q5u;;*luXnLM$
zvcG@D82RUdQxNCt%s&@`U?!b7F8^GJ{C{0|Wl!m}L~dSaH;gzG6G>K5>CIOOL;wE=
DAr}_A

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_019.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_019.png
new file mode 100644
index 0000000000000000000000000000000000000000..19d8e667a78dbd6f21428726512b780e384a6af3
GIT binary patch
literal 56979
zcmZ5{by!qi+chBFLpRLOoyyP+!_WgFAfiZ1mw=RX_W;s0BOx6E0xI1@Dbh7aNH>xn
zzvq44_?|!J%sJQIXP@gj`|NwKweEGt=<8~b5Hb*AU|^7FLR4WG7+Bs9=OOsG566#=
z+8z&oQeJ8%UWTr=UOtu{HW)gVUT#jVUQQ0L+1}cCcsjVchzm&yi3_mVdwIEeN(&1+
z|4#=YR}VX3l%mSk!yp815K~VK3^L1qewbwnr4ASvOBkA}N=Cl9ySJvb)~0ViMg{tV
zH!po}QHDJUegpBj{G~&c1W!LO+DrVjuP0HbNHzP|$q}pfBj;2=yaOVECejgOP!k6@
zJ((SKDGG6s(C0ClHkV2gC6jbn@H?En&27Mw+eHVlPuNgk0a67pUfSjZ=m8f?X#c(7
zYYWHQotIty2U#;cTu<s^bevvO=FAv6AKx#NiTwBFKP~;ugGiTu+WxECuIqapA@BdW
zbb1~9$jPJpzgr&$xn_KornUHg#$3k8G5_<wkLbC>omq!T{_i8sCBL7({LL!0IkT{U
zu&7}Vzq~&e*V!e=y2#fp?f&zJ4<<i%8@P5-I>jY_KW|4_aKG0OGj@MI`|{^zFj8xA
zV}r}l(NS-AmFxWed0X(kJhxtc0HP-A@@D(<zxwrT246d^oqP*CzaRYPju#ykcq<Cm
z3iqWqcK4YIE2xg=Y(>SiSuXI)wr|%A&uU(vt7?MM4ej(w5MBNu8t=)Y*KpX7vOxqP
zJwSKg9IN<V|0Hr2UL5Jz7_U$mL+92G4h-6U0p2u#7MI{7sbGZ13$Kou4sKQMQJ8Gu
z%7ggy4%P^L7${uP&uz)7s@F{B&J2i+%8L5RBFhrOM!a^2&uaj~!U9_RvXtw?fXE&?
zm5h%jj$IdTX>$Ahv@%FreRehjj_-n_bS~(KvHdj?gXH^N?@$K~9)ARs!wY^_b<&Ql
z=JradxQIn|!zCI%tj$(=n%EtoMCnNX>fN8*zux_(Q8)4H=HUL|{)58Rw_89|!6Ggv
zWP=7$K|mXPqIT}OxcXywtf5vq$><7LFf4p+u8Gixl<Pwa-FaNyxi}?ppIw^sYVtxM
zYlJDrXnR$y2qQvQ=Dz13WI0@mBSo2DkZpL2#IXGYQtr02T8V=Vvv1UeKmpPQs%Aw<
zj<>J!tUfS$IQIK%Y7S^8Hhn#8?MJ65zskcqm@otI5`<rvPR7Eh3}A#+9JdmsGIYr{
zkrHEd(<^zL&5@n2zE{Z<xL=nj^nILPmm+SX?;AVWZN3MJYiU^quO3CPcA^7j#QUVQ
zT?P`mO<66LR?RO1r&$lj?vn1K`j7u&Z&K#!xyg4voak|*?xC4qQXV}srU83yVDFH*
zSmhyMXO(%?p8IPQrKraPhH5495s9GOM+cqMFBLonwk}{$Y&LiNhxgEJ_WY*3mY}#y
z6_Hk_f&quwSKU{gmXKp@lEB_I;wJNyA4Pa}<4*P%#PGxbqyzMLvAZG?UdIy<-Hi`R
zMR->Gx+EpP3{of|S(Aeljq@ov>QF=_I8=3R;BKljj#z3Zibg8E9GCy1gg6Hff_|6v
z<d>i@C98wGTD8Usy{Qz9m|m^6n)_Nf4h+%XT0%2y=&VGhQ*)R$nMw%Ug7c`Zw;VHc
z1w{Pcq)+d8FMKEeMV-*(Kr=QX=2H<eO5M~aVL)Y-haZ)Q0mK@R!<2>Me#5kPO-8+n
zS7s1DZ%qyLC|BmKF8n#L848Iwi*IC5D8Oq>rOQNjsFnbfDF*3wm_8o18jJpfr9(ND
z`UyF4V1)29lbvMKMDMR)?5=E0;E~v+xdsbwcY#_;myDi^;AP4fyOB94(drK}T9qH#
zT4em`btU^S$uvE#X&>9DV)NGLUj}~|Koy~Sb((SQKHu-I9%qc@s_}d8dfK_O7x*W6
zXDpb&tZMsSJA+RX>U=l5yJSC(af1t10zyv+P~6~PY4^KlUw$O)qRD=>`u>$T2tEbo
zEdhl)JW^Z(cBC?BXtznu|K2pn<jQ!ac$9|&ur_gglgLSt-~9)V5T+r?s|PRKr%0ZE
zyp$=@hZZ$5X-Jevav&MLi5@WL*Mj>t={7rw!VgXCM7WTP(uk&SI&Fai$H$<!Vg9YM
zMBR@(PZ{vb$?XwdV8^ZvZ_+4@6#!6?0QH{S$~Q^yNp&(s?su&u#o{JaNFPSJ8C$!u
z8jw&egf`p4H7_j85nsTUWT`O31ngpdAKv`@TUXaHLAwt=D+q6sX<F(#NOlX7e0kWQ
zo{~>q;fv}&2>ZJpMq-PqJbqQJ`z3Io>$hiaAewlp!8_6+DPhMy7wjIY>L|uvXX&2Y
zRn+)hCgj3$Pd40sNVD<+Z1;8DllO7TwCY5+`~^O5bfY>iPWYoS|2TK4M>%4REUTQ#
zingBG`IB#$C#1lI^V_wx_21`)n53%>t0l^Ql>(qZeIA>*EkUNWY5#CMpc=q0bnUX1
z&Q_&<A)HK1*kH0GypX~Xj=jS??GF+(fF}14TCuBg)PjoJ+LW#jiLjXuIS&Il02JMw
zS9rv2FYVr0%3m(v6rhO;WQeff6so-}#8JwZZp;OzNe=(BW+`(O^1#6et;OJbJO#TC
ztf{0Fl?DNa`!7CAlK#E^cJVmFm;sU^vU`@5^@c50;<Gzxxgu1V@E3I=6<dIbH3si@
z0)@98RK2PO;c(y!-dGZH3$;H#9`%N-)Zy2RPfl$!)pvMjDt%Miu_udJUC2tyZEuf;
zhX$QI@%N9~mGmDG&Rf4}eO9I?xGHPHeTZ6(=td_;yHf#GA!AVz)x95N1&WQ>sFWe$
zZi+2W6#`5gs59oIlA2dL8gs}7q68b`VhgC47&SYUcHX!7xYYZ4kZj{DeGw%`kmK`e
zJcskV^<6bEV}D9&Z7u)k=rZco|Na;k#k@92;Zime=$dzj0^Cf;X$u;s@Bvwgd37G^
z!;;O&P#|ia-%1s3>I5NtlP8wUomyQ$AP)4+kY)@=q5+@NdCJ(+lw#WWL~*jN2nfm%
zN;tQu!r~<NYQ}lW<+HvCDmNd!03{KQV%lG{Lf2}Pim&%fQV73z7ou*3i3T=c*~#jf
z<Tk28PFpzEC!lN*Jtp05;GRcFk5C@~6asg{@51XHIDipAq3qanJ4#W;C%TTbGIYfY
z4EgvWDwp`YvJO5uw|Ey>&rNY@NhP)LlhXM}gDf%Q2qPH7;rNWCIOx5X{2uqiPyPwa
z|DNrsm=ldJKcs{qcq{`)r}V_{oLp2Zm*v*?_vC`^rp7)Lm~H$Rd<s7`+0=jA_?LqU
zgDjDZ(GYMq0!$GTsD&BSVJ1#BmMljx`pZ`jhJE*o`kwYy#{AnACfO60mEHHST~kvt
zYj>BF^1lo}-{1tOR=&R4W6|*$2zy@Ou^BgJVkS476(>AyUm2eBRe)wbA$LEvyDym6
zTL39AL?WH5)XVkP<IPnORZXoub=auG;{)Je3KRs2(xN5}Q(Q|}-Kg4WKW}mc4AQ}w
zYv|uMy*1x(194Fh0!dgRL`fZ;0Y~!re!{i%g<@r&MP?-yokv$~qrwtsa#)vNe>UD+
z7_@o49h1vB-2F?_`hCbLkm(qBoM3&}YC!msY2B<J4kbvLUX`O+z(sZe3uYS+TNK$u
zP@r(T85d9V+~X%KPAj{~Jm({*u?C5$SMJp#=Ipo<F#`_XhSEzOJJZh9!9j7qe^R4f
zw)c2h0TAkZii8}iVS^^mRQ=>{n90?$A^d!r6vd)F;62tB#!!eR2*->7jxVq0Mf?!_
z0^YN!@zA?jLw%*$-mJVnKX|!0`Kbn{zzXoes=$h<z!#y<10<gesRi$(46&b~gTMcK
z*ZpU5@4VJ-QiQyfp^ZZ9YQzwaUz|mq{j3WmQ!KS@c_+Pgl71%%v@vjo(#kx_RpSFW
zi%azBrv_i9uC}94KF$;|<(;h}#wOXIZd7DrKpiFr;ep(uMNaUU@ZI~p(NE^G-}(lB
z+SJRI_<7-@fT<tAyjD8uuH)*Q`Ov1GS0Zeq+2+@4%=?RfKcv~^Z;teCnwm(Hz!lfk
z_!^^GR7jbS)G*F|l($+-Cy&mKK8RfC(fa_UvQraT)9KQm3q|HKVHp6ehK9y$o$A0F
zgJ`z*dwV+d5+&v}Yn67Y8v!$Lm<iAgM?lp@5g$Q_K=(fea>AY_i}hXOr}<12qfoRm
z)_Is=9l2GE7l4`<9}zfnGOvMy&ftj`22Tp3!S197f?YDjkfNvY)2qM0vHh}?oYY7T
zLcET_!FXN}V+;*o$v<&UC+M(0xaTfRMj&5qu<IDq`q*n;=P;4S;KwJ?KHJ^1C|?{T
zOnaJDY$RvlCLrv(xu)J@gJM(Y#d^;WZ*CdL+!#(4+E#U<YL?uS%~Ju2P~?yCd}AI|
zi%OBv;86b^0rJSqU-0)}T2xAU>sbKXV}hy+jOt7|GuONC$zcN*m*m=$1_hhmkT^zy
zg;UUvgq^sPx}*V<H~8d2H^s3O1C)pVyO<x~7AuN%8)g@-q{bi!BF~P;oZeh$^^W#g
z1-dTH75#J;RgjBvdwnya``$_w8)ucIi?N+<af4c#Fd%l^K@2(!Mf^a2X**%4?EQNv
z)+fSrdb!ojda`(MhvPYAV@xCMbRECF?MFkq!TXF))jhNhlj^remSs)cVa6NIzHEv~
zJ39ZKy&VO_^73jc?YGXZHJk7sz6GltJ%5wMo`$+z`=~Xp(^IlwRyNQ%Mz4Vvb~{Ls
zxEOeSHpIBhlNOnh!P^)1BSLGi?)`lo7mHgaUOb+n?YGjzQM?HbcxpaIE0t8tYJ6_g
z?Z132F=8f2i^Ou>Te$Tz+)WUP<e;uC{F1AlrEv=do`C(CmkKLo^VSZwi36J9L^(eR
zYr&OdDYb4=lbfpp#uUs6(G%GajiMW%l4gi8ppI~&eG<?`CzgrM<<t?s>JTajJOxMw
zokzj>#ECac!5k&iAFSTFWW3=8CbE0b`ic{seQ-+lSQkX5c4dooAx-9OZ~Oec@HAFn
z)5Zl#f9TM7_@8P~SS-Ryya`^b<43yjf60~crMHv6j+SLoG#aVOgN(&?N$?(}xz8LV
zpX1$3N(RM6%yL$GIvp%no=iuvJxO_;Ub)PnBW@3_1)eE@j-Z&35XDs_t`$aH8Yz(4
z!uN)fXIT5cfcF1c^1k$D(Kg}aglJf<Bu@bbkh8Vpr>*}^@@${>ih=%*<s-;W?NTVw
zqI@U@75wTr@M*kRd^h3s>lDmf;kkr7n;A!Uug0<SjYlqvEkhHQnT?X`ldYV(l_Q3q
zjI*+{m3h@FIrF9g9TrJvZ5OleYGbDL3o26KASlw|iKGutI_2D`PSA0f!um~wbH4v?
zMjqsmE?YqK2leK^q9@(gp_D6ekJ#LF2AE-y_;&C<Z_>&m;weln)_U&?mMDK6U7@N}
z<fApPdNiil^ZEct2C@f9c)ReP3hg2su3TPAv-D+rmNT}R<uNE<Uj9&t#-LrDJ6lN_
z-+7_;jsiA5zgOQpWm_lO?M~!x-<jDLi4xjzL#@<7c$0Akr^TV};^z(g1D`+pDopRx
zUQ{A~N?UKPywA&1VOi)HJ+!Btm-}9s_pz}aQf_YZ(TppR<2S_3_lM25oI20oxRO2Q
zb}IPCPpX>Q=Gn;BCQb9|NyojzG}f=(oe_^L{BL#m`1(vj+_5&D0ZshmtF?^G$1DS$
zS}vXloLUYuO8(eyv~)xc`}P<9A7p={Zk<1qU7MSm78ls^yikdz0?M>7IprHx{gToa
zPX=T?wkmL>{!-e`8?et5P2Q>?TA=5K`ro}f`b4Ac2Qa1>ANd+F{vE?n9ArxGmlX8N
z{dzhuzaxN#Q*EL`r)br%@rb7Rhk@L4s-}H|PaY=o=uv(onPdRc?foDg)9%9NQr@RB
z6GPr-WlupgnH9OxbtA{eVxzWEJbbIUB^EW5nG_(-BQCoy1|=49W2k+hltj%am3JL9
zqqdZ@KXbvY+9z%~KEcO!0=crIG3jwYEO#L<49ukn?CtKVaa6=yv+(5a>E9-Hhfolq
z4mT%zYK!W(<^F5<+4F-n8&yaDn2vK6DDI?ls;`sFxIxe90+vRU4yY>5D|qFlDij!j
z@369F*PYoE#xmbd3<xL^rZP>^5{C;V1?FEc%lP>CZFFfiQu%hR=0A@TCv{YxQv^QU
z>rjj(zuoAgS~-LOz_^w#lz24}k6Q;cr-Ys}UbHrUjY;A}f<d|zd?%$20=b2CwIeHK
zIs#U2^Q<b&K2q@`QcZTsw138YdYpKEuFuD%hlj{xzZzvXs?o7J<LC%;mcMgI`)T+o
z_hlQ?m$h6XW|Q}Iq=MLARxp20i;Z|p%963ZR!*LcU)X?qA10)xLhKpbyA0amE>!lH
z?x=*V-ui6Ve<mXr!&OvMX-_E2q(oH<9KG_-RP6<CiuOutGIKDK&$b9)^e-fFCpTLD
z7xnrV2P>uAO!6v!ZXqz`>ev{sUpx+q#ks`LNAyZ{o_)#*xR!Y_goLJa33*JRuqWgW
z2-e?6LE?<b@k<mjg$P%kKL7?$Zs%|M`JCp<wT6p<N%?d$l4s83T7&7sj+J@lw(fq3
zZ+=sSDXE*s)_xDeBdeXBbZMS|8I&8Dyf7_(_}=#$9h$1y5$qEb#Kcqa#ux8dgYJ_q
z@xEQK0@dP02rc89LCR5N(V+^j2$ij9e6cOwpQ-g^zOH2h6+=m=Rce;|rQ-(PfhEZH
z55|X;`P-6HVFHjm>o;#QU=(TdOw=Z=JiJGjYvjl4s8{X11WB0`MOI$B23dYo7kA2^
zMtFAFc#EZU5;Zv3UBbJx9_iKxIwqlh$^3v)R}z`&o5VHaKEqE+43v-6_aK*t5U??0
zUW!m$!_7n_q@FN(1fy9bYnFPf>D@a*YmD^8Uc2m3i1TY{3y&D^J)<KfDj;fUp63wH
zux!)e!2V)p_tq~4nmWFBHY5FozyhR-t^R5<*Rp1CZ22$C=6yGjeY#?`o7(4vMCr>$
zj#Q<M!U8ifagOQK0e+Fs1JbAT(M&+Ck!M;Q(8+zWZi}&OFzZ7Yi4qX+y|z#+r`yDd
zn`#*008G<YCMs?#V^E3FLI!6CHp5}@$%N%fe>CE!f<kouBf`J!`xU&66@1t4)15;j
zstgq}x9C1)|IE_(;RR&NTX1K`pNXZ(^hNcGh8&lWeP4?~EInXu<;4nx38Ug#UQZ0V
zjH6vA%nfT3w!hrb*EANaTiOx#K<dOTH;zxtWql&kv>Lqp1<!rRww_+`ln;9LPEKMW
zIn*6>N+t@I_tNa7O!L#SX;6O9bi%yroW>UgqkAuOE2}J>hG+`{2d4X8=oSRh2xA)$
z)Ipp24eY6l;rNf;30B<CTN-*bDW8PHd6?O~?4SGS7J4ZohEE%~LYW*5;iTa|)E%;C
z{~~K~mr^$Ae|rcv8eCiTIz=h+B^<tg%*MeN8<RF!(sDsN*D9IOa@+l+_^{O|wy%{k
zTA;Z3SFx-iJ1d>pa~;wgxgjl^8}QYUiK9sH;HKq7moyaUW*j2CHKB|_D$FrSP^s1G
zbHU<MZDBLM)XS)KU#7ZuY-6lfBl2y$aQVQ)YaC(30J09Ri;QS>NChivX<7(QF<T>^
zRGK=D_xtIYe{yR)d;ieOzoqxbkg2)P_q{QN&CaVpq1<or>7r1x$%S||rGzv5q8K>a
zBygVXaAqNqa$Ol9{~f_K!B8%nlIW&e>=im~GdU6oSvUjKM<mB_AaOQmY3ho{(PH#J
zJvyQjnOk|`I|T1uy7V+r7kTPwBn8j>C%k{#x0l0+2uz^hL_of)G`aX)|FW?5Y18bG
z?`6F`A)8&^a|;g4L`L+wOQ*T}{3x>({CdiOq|q1}W2c0zL@4<Up7#m5yu8QH@!9si
z7>wkN(<-Gq%|^&+Jz=VCZdR%JW$Eq!KuTy*SWaS|XvOuoLxOjBK)OTgD36&p^$i_^
ztE-rJ8gg$xM{#hD#xxQB)o<``8EsRzleZf)@8gdUn5+&_b@<*E>fh;$PW(}zaUkeL
zjFYHI)oa4kWcCON9nl^Y{b_=jU9=i^oOEeQq426~J>88HC@|?xDe-zB5`|`YYpaR$
z2XZqhUbZVO_RRV0KekxMNPm%uI)o<MghOhlDW7?u0?-tTx1{b(eWMEZp?4Nhi`P0C
z8yj>sF^j^7tpYvrVOw^hUwUP7F;%xtQei7k6TM>h60R$kzs5b5t6#!=GFm2FokDuG
z9vDVXYJaeodOoi$D2StjfQ=YGv3}>02&;7Gv8ED#nhGgsnE&J6*r9W8UtsB))Va#S
zmsMQ%4kuIQYO$^T9?g0*@|Orw6{AXLXsu!WinXs<WT_3vzN*Yl4ITLYSS$k^Qelkf
z5`R&kTpnJ2MSNCnns}vCzXa%Zp#(mLWtw)J($E^yYQ}X;?wx(8&|}XP?)!z|>#>s<
zXhGio$bx10zp48_IN-m8ninMo8fqH&zzM!04l9~WM)OvW5L%1V8-@GrwnkFJTu<}c
z8`3capCUFwEM>NI?4mF_Hj(Tq8tK=u+5U9|xkT2z$+X_SM3OZ@SVHV$V&Sw#pPr_g
zgf;uR@}xiZ^qj)aYH0OK`umqNoX3#Q0XIzEoe3?6aOgYLPMG@isqxDTvnsAVRA6M5
zk~(>&T-82}G~vX+V$RECn@4MiL?a3$6-%5V%df=_R>uXf;}IIb#-eI(oCGh1@rpOn
zYqrM2L&I*{m{@j`TWPk(0oxRzn3}M)zqBuHDhx`e_CQF))Sj43vq++EqLPsp6>kU)
z1(2i*0)Ka7UoHJ~Ro||2(_Pt{ujn(lcQ8(C%Z~Q%%FAUAN2Kr0%$KIo)?sd*UO+L#
zE-ImnNX<1_x%g?34rg~XFkn2IaOgy>?Q63yd#leLjc=Gv*cjRahuJZd4Z+Fh_{NhY
ze8}Eq3HhWgyigm?Gr5|q`V14|sZwS1`?%ci?L4o7a}&6Ixt~5Mn!*Sx>^yAz$$q44
zOJI6utzPQ_TPz0>V%ecwJW$2{Sc*jtLG_!X%Npbs8^D(6?`WV5OOI7YY`xP!!qg%5
z`^%-?Em3vM+HVXhutti#66^|sW75sCMkQ1rnl%cqP%8NJFG46|$d&vx$tCK;#ZCBG
zt`skK=Klu{{!46U-Ii#%93hk;Bi^%ROyEk*UO=4U>TGO$G&QN#7b3c1?%d0n_ws-7
z(OOU-o7MXOQb|;GQ1A9RH9xtCq&*~MO+rm|-JLk<5$VN~tP019bPDKbSTn@cueKFc
z{#)&%BMH<Xi-V7{JQ~J?_J~t9O#F1(#;chs^7!W~Fq7Na++)P#$YX}}ZYDdygmMb;
zsmF<=WQsgcWVC>5TB%&>q|r=S7j=YhA6`0n$-Ebvj2WwXn(udl5M>aY6faBM@;9>o
zFR>ayW`vO-i07?_nL&ZscdNtJ?lnjQQ+OA9?GrqJQBpV|-p51Cn=UOl?in`hlxWk|
ze!p?x9Vg-!hftSkx7)a#ZMC-Z*Kl2IuCOqf>qdV$1P`7FcDZvn@hf*0w<1Sf{BfaV
z`ds`yG!|xW&5~j^sXgnienlqF&mkVUB%jVr4aSQPFD;7aK7eBRRW+$O*^4>)=;NpW
z>0nEd5zXT0ZD@>3_DRDh8GPPyJ)Jt6D{ukjGh3dYe8h9tY_E+F^awfgCyGx$68&_4
z*7!Gys=dxK5=pKfYgK7vt)gTbx{(T?NKmB)eM0#MtSh~1A0rp)w2?xQVp8^qrhham
z=taBt>Ypr_Br_aznlM6fBgP_V@%54=$?Nss1qY1Q|0BfzTaho&%~hCmI83@{RnUmx
zgCwPjI=nneV-4(#dzSwI?U{#=VzfflN83?8vrE$JhlRslRu$wxBNNBLliUd=87uAa
zI^J+XR+ciE`f*N9(I-|SPNG|9VE=urlLo^N0_f>8$Hq1lUEz^&=^5lN#=@svUVf23
ztnE1m@gnyPST_atZpi8RD^rPKDs596_V+@tNYc!#ZF+_Tiap)fNl`5=L$5D_F3Vqg
zPlO}hwIpN4SJM&i<$b>rS*QJ)>yZIlAzH%5_aN&F8zgD{2?jMzU&<;$@euKob;b4#
z=SwB<V&fWGhdq%bx6{q)Egps=I`up;11h;#|KOsIYhjlnYdLax?|3OAr<+5*VcIeG
z!xj8I4($RntNTAM-udf%8HLZ5zuCeyj2)qiAB`KJ$rW9BO5R{pv4W1D`_et<z=0nk
zDm7cHGB<Cz+NJi!liPhZd2y#8?R<Vj7&WOdV>7c|;uQn6!A&dWL5e<7dA{kJGgZfo
zpaEA6h;<h(e*6|JQy3W!^G-mR0ryzgtd)O&Rchvc{~4WcKOUl(4uM<x`sTM4i$hbh
z;uq+#i!U3#(@nruW0ihm4G?7SaW$BHI#R{20_u^A@x(<1Q^WrN;f)A59EuEuFb3@B
zYVcD}!y?Is(H>E%75x|gpW?&)kK*%7)<adAKgc&69{cuF_hv!$$%OyImM&q2&35`l
zYW_K1u3VdRP_2R4iEs{&%T4h^^3^;_1HaMOml#EA&$4I0o2R-HJ-Ya)xMgTE#_)&N
z$fC@z?Ta=eNAUrP8(GIluOw2DCndiWihnv!eRy6dNmt+IGq7}r3Dcl+bZCCg#N<?0
zE22i_P&vBoW#;7Ud^nPxJ7F$wb&O>%rmBOAcRb)4tMjuroHpPwz-CZh0i9+u-yJF3
zsVrR|J+_dN)To*(omb#9>s$QriH*moEs(8%>!>bh9A9F|Kb+_+sOs!P8KG>-k&-E`
z5!FDgvc^C*Xs>am^>hEyPw(SIWkNjTSCLYqi!_do4G?-Lb{EyjQS5=_m3M$*MB(~n
z2X<rlxrZQO{r3B%PEv7xNBg2JCg-9OzTvnyZHLjkv$oS~IjCWSb=O))_ohl@6dpJ&
z^*nLwLdIxggZ@0qi|m5FQ4&xe-%RgtpRBY=G-PO(Y=$;he*#|XoaP#nEr<Xpnne0V
zHnW#MO5pDP@V$+mRJ>k-2+vDw&B3?wVr8{<7TtEguGez1q-Sa_qhd~jxMfDfEB}yr
z`!0IAlE{%tZ34_$gHXj!@~k-YgADqw(v_zv#B5{zM_#6Jw#Fs82`h?7e0%(K3RB@`
zqMvvU%i$A5kuaelw9V(UFZw~(1*x?p#1=a*JnEO!=V9<iF#TFR3O|(qj>i^%`1zC4
zX~{8#^x8{W@a^04!ISv%cVU13+DvEs0tcRpwfR5CFQj^w44c}2JlH^-X0v<X+z@pC
zBy1XR?T_c=>^<ZBTjC@2BrhM5&LA<gaOIU^hEk3AN?*dKfRjvSC^YwIJxfW^V3QQ_
zDuzumQg(RI`9ts)w$*5fo0gAx$J;#Cn@et|N*_@N)DV?K1unH8{0Dfm&Vo|CMiuN@
z>!AB>UR_A22>zY$@e($yCG?O!x(tP9@%?3D>Yen)E%|}h3KJwkB-?Vy8Xcxr_aesa
zx8fERV2ObA38|=XUX*GW{t3>WiahI>9~Blip|ZE=y59nh#Bpol2MdeUl^{HECXwCG
z+q-)!+PX%9o*r*wf9VQ|wRojEKF4F@IO9?J6W!sGWBEpe%g!gbVDx$iG0N3Z!u)QO
zIWkP8_?RG^zo>)(6mx{KtY<-)Xn*PLctAS$$F4w7#Gu?_P$V|g9h(P)O#`UMau^aB
zHh{gx;dtHi6C-YO@l$il(f*jj;+Le0NT?u}t|Cvea)kG}0fdjOc4<{iMS8bHm9EW7
zHQtlvK#7_Vt^A%pV4rZWtiS0$0{$8jmC_b3LS|Z|8s1Rb9IPtv25uDdx=UEu{;f|U
zWGWXyrEsbrTBj}5-~1y=QgOwc3ZM@l!KXqZl&Li6a@^fohEe;Zw@b<5+>S=WkG+^e
ze&l{T1pW3%hh&!h__$sA;#hz^HB_mlyK{*&Zy(USzqD?55TtnZS0_7)xsv+yT5v7Z
z?wz#KgO2;*$|$D0#5s3(0{XrINC;5#Z^k>=HHkJDVEek&-cJ{2L>imp>q7j~Ebo3@
z;qS6Lx{vl|oAuUn-*pjaY6w3B^VZQ!{&-!*-@@#|boGY%y?E?9Bpd>0gfmf9ZIn!Q
zQi6B$`e!R4IgZv+U1yxPRqTyrCHw%ar!7yX77vnbXr2*z-o_u@e;xy%sXCIewlRhK
zQgmF1L+Xp7q0~lHRyw*q!BMCsX|BamnAEUWt7<s~P-jH4Ajj&98G9@r$4{w9hbReV
z3SPCE`L+EGQoKDH*2Ru7?t-9!Fd5dga~HG@eHwwaQ#AhxgaF2^iZuOF71&PJ{V{Ja
z1%>zZkvuVjPll|d39TZ})0Iqw-<FFqZ|((gu8Pr{oAcnCZ#Fv{HAjONcT$UC1rsJF
zN<g>bcpqFxVyd-#`P@)^?7B$^@=%G8@TsaRw4!ShJ5OYjF5$BuQ_g>9-3wN3wtN1v
zi#g$-=pU>iMF7|%iUUCwN){SSlOgKPx$@pxze*CUxQw+O^*>XqI`LoR)pM|fkHV=M
zVfB#KjotlSPv7xkb0MHAs<41UP4whMi#Uqj|JvEsc0-ed$;HKhFWqFCvltNyBS=jq
zbRGd48Ev#zlRx{TJvQPV>=|~3r(yoh^ZTijGVkbwUo?-co$ZJGd?1X%&clr+hBb}5
z+X5`auPROWW>)QLS}JLfWRfnv4xUOlCU=3ADfGhnc1iB;bzq)_USUN4FjJcURR`v0
z{cKXlcMAz&$?@3U>Q(3Du!hf<CokbN_d90bLn`Gr;gjGsksRGQ^0$6ubY7a{CV9Pn
zQFS3!CvgERAm2Wltiye(CA<O+X;r~Kx;cM5;yE2Jb0Dph-<#w&iRIfi*^h?L7J>XH
zaR}Z8sfAKAKKH-Y5ztn3ip<XE5{W<k;u0yU%dlToQ~rGBu)pxJsafL$C6}5-Z}g*A
zE6Iox`t#|~89fDOooTEat{_tu_4;H)$%u@(yM20^V*BT58tG?=O4&tIxkV>Enpv9#
z!HL6}sD{?M(W?suE<H}=b+&A9PHJi1I3Jl>6`vb;jRUJ_V5)0VjSIU<G~;Mt2wnUO
zJ@<bhJUfcj{oJvLi`hXIi!(JjyEmMaM-gAMitF<C%ix5XueY*GQPBckGUds0w8dro
zQPUT0Q>^8gQH?^%k<Y(UtuR4*x5`FPx<vMdLQl(nF+fDEEiF?xbjyByoK<j|aAKx#
zv^Rn}WyeRPKYh?)+dNfwUKNCOKMIHP{B+Ef_7Bl*(<%*C+*tinrE*SoCGl}+MwVM{
zD>;XBtmEdBFM8mF%1#?1$L`*Xo6(#5IXU(MqMr9E#MQDgLO=s!b1a`-oSre~Lcb0~
zo4xeipoaWOlF!VD$kxRp*PbY;7-caU6*wr_DdIh3XC>gEHgxUdeRv<9ffvs9if0jJ
zc${+FSBYpseBqd()CSj=tGoQAB0^s0;U+3Y>7)}3ottJA<Y(6+_%aYiNW1>(D>e+s
zPB=v4^tn$6HuES2w<o@+^_#aLZz@fzeu8CpYv*N^c{SuRW`c6;#tMpI6_c`azn3;x
zBroQ+z^~ta>5KbG{VtCsY#;L}H~34#;HwMyqf^sLhu8k3zxIz8o-OsQN8Z)y<!^xT
za1%snfQm$Mipll~?BqEKFq6%JT=rqlt8V?Fzevwd#JT8Oyqp1u)h(4R9e|Sth;}oY
z#|-Oab92VfC~~S%Ruv|)NKGx0pgM<lGlY;6^2&yCul&nV^AC{s<N^mXxZgoCzr4Yb
z);VMdr%<hkq!8^RQ<k3b<D(v{xxIRJ+1G3V<V_!NPJE1uv@zfoQtyVyeVrh}M$t51
zt#QfRbzI<=D`rG%kyKKzn<o13`t6&V<k~to01o67t~v{*{@AtOPejRj-Z5p4<y-Fk
z;F~&W;6ctTF0S?c{B-(`-iO6$V%Nu1x^8GqI}eF``fmG)rocb~N9+*16cm5bgOnb7
z;g_QHjrB<gQCJ`{I#DyRVtm!Iv5wvoa`=SD@p-?&r|wRrxGB5o4>>ujJ;!|Rt=j(M
zuQdsiu?{+=^=f``jm+^6)WHh9Q-<Dp$j<c)lWU_nP!Xcyp=^nNTRu`o(T9^V13CGs
zDL4sQAM*LHWWQi<eszx+t|?$oq!i0cVGNrNaV-w#!+=$0l0%AWBgQz3c6L^I)HM@e
zq0p*OA3{5_@BLt+^ge-xlLqkS-_qsaB#Gh6wG%yX;09Hk<ECFwuiN>$teJy+$G2Xx
z8t2J9#`Di_AyI-1TEj*kOYH>;IEJ5$+eAz~MpP876~b%KHYkgLjIfd?VNUOxXgD9k
z-We73ufvJ9{ek1yn;Ocm5#2IjC&|zIuM7$n7uhYq4I^2-^tQCLEjc4;B>RT>`IvwZ
zG7sE6-OL54k?5B|U}9X+mT@Xk5(Qx?c3-$c1K*`A{ulYL|JP+jwsYFJ`cDEs-7Lnq
zA%UDT+z3JsR10-`gt+L@L`i<T1--oP8#Z#}3?Bz~<>=$2h<Rn+tVI!(>j7NuMV2@y
zVxFaLS_H+7$^VtQZp&$uel$cqW>(%aGD3XzE^qVYJAhWt<Rpb>0z(X;_-ec;u|M<Y
z<|m>oDD(D_4a*#`q*1EaGllzjL+*^Miu7_Lj>VwFi$o~0?|ffgT-;GWxPb0gQ?LP_
z7yL7GNAuAJ{yZjv+xu{Ev%u3XFm9|Ss59M8_aUhS86sFz(X-{yOo{Pknad;fbHVCt
z+=@1U0>gX!-DP8Yqo`UvN%y*c$uC%*gK5&)M_%E>!(JLGQ*n>8vWX=(WHdF8dH!u`
z_`xTt%wsMTSEf>kV9=m1DWxwjN-|sO5dPAtY9h&Ly;8{2J8SpyH)^wB%5{m2(WMYg
zsfwh|_~@AXf!EM)l~}z`;DN-J87uN`@QFHPtm>qVHD}Y23Fpdm`fEh(W>N4}xZVE|
z+|U&;tT3MV4L2`UEZ}h%cU>$JJrUE=8n!e~u0w~>vM%sq7}|TNnNgcZCrkh03D@`C
z+ota<2tLHgVQjm(jN6gD-=$xJ(Lb$9IXJKT97)0O(oHuz?@1t<&>~8MP?N$TTz`c`
znJyk4&-2*bT4nZol-xB(z`y*Gl)2*nvQ2XSi`7XT`?}UnlWWFMF>q>u4MEe{zbs_>
zg#$N9-ZU|oX_B-2rb02Qe~~47G4r>1ZeD{-?n}0OBcPos7wnzldULF>^6k3+o^$3=
z<B268Eqr^MEyI2y*IZ6%+8LdPX$fFCc6#5C(fSvMhL-kk!pmFIytg^F4Vr2UG#hr?
zNI8at@DXdeG)fIl(hK63!PlIr2=j&7lC><(%R?uWsjQ{bKsS-XQcrK6-WqQZ`kBRd
z(@Wam%l_Sc7BO-0ZmpX3fhdI=(GJTb6aHKhyP?8mD=^baZuf_P_BuqeZ(w^+iH_8r
z-|-QqLVYDADgA*50zc}*p^Xv0T9tZbdqC{4tHnpNYVw_xlLh`tB6-?~F(l8%U2Ab#
z)Y<|Q>kSY}AM%nQ=2_^^DO%s{oNt3k_rEb)&P|-Z8pnd0C3KL++$VsSQXR@WWf79N
zoF3TuN;OfUiN9xgg@knLZuXwdI-JeAkc~5JlsY21NsoRoeQpTzB&Gfdkx`kp_@;_9
z2Q}zJ*IozO*!$yoIJ(h+I1Tn&EI=t`#ivyKy<U?mnZkL!CjG&8H9-R%czEzXh61uD
z6#jc{GwpG(-)o07Ue_ww4w3GbcO~SZWo$X~fXx0*`?a+)v=hSTl*8#ivU<irRebLR
zV=km%>EvCY7yvk9C$^^Jh{^DbV~gGo2L?*-+z>)IaQZqK)R1{!9I4Z#IQ)k2t0x*$
z2rK*78D;-kz9aP=BfI+<9H*5tqN7I{7p<Dkn^^(Ldi6RkH`{XDW#Wx1FoBbqbQ}gs
zKvr`8%;L5E$T%`OCMy2CRB|hogZfW;xbCO8<47m)`swz`P2Td&^;TK9_iSb^Q+-+D
zYt^$E0(?S};_q6?%0M{6ygYT@?Xcr-->1gxsM9t)S)XmGufF$&UmOb5e}-fJnoP@*
zzg+_G#eVxSE&FGLAz}Q8C6}myP=jw@**9PDX=E9P+oPCg^q>lC{cU{K=M_&ovuMW*
zs(2pkZZ%4TaIpxAHf3m1|Emr`Up9}*MfN6Q?sjRKX(r_MqDi^j2C^pT(ZLG(;MUN5
z`lE5qN6_P@GHBBRhG}OyG51Es7SnD+Yj6QiQ{y|iMhpjLq8MwN3RG4E?);wo$ys=q
z5)^I}t>t}uN7_|7BVlE^P5ABj)4dQ9MpCat=Wb;3!BYyP`w4BA5lXCQ>q!TQW*-b_
zichBy<otGiKD;33bYc*k)O$iJ<?EU+yT}QJg%|Y2#LKeQE;|>cQ6#Q-S_!7$n{&DW
zBMxMzBa4<{;BY2DI_v%Se@l1&Xn6mE^)LQd_-->TFV6Mb-~woG2$-l>O;HyB8OE)X
zQiGUB0QIl{zmvb94mu3+H*55Q3o}8TLCj74{nD$B%U9G4Hl0hi!XNVTFxke$i<ySb
z_)M2N5ihitRQN*%aq-C}i7E$cyIEo?){yGy!|UmM8X9;r3$I~u4I(45VGAb;w*~jY
zrY07BesNwb+>4nmQ&!Gd#l2ZwayT~F2Z0lWtfVb(+q9@0-uxuo+rae8K%3hK?!RTW
zHN)8hv4SFXn{_%wiOAo(YA3V0y$p(+z$M008kkjx`TL$f)426IjZ0pVoEM8p&S0(l
zQV=`?lztqQ$ZP>d2ye)WWv9`8>1+2f*<|(D%gkfSzt;+WPo4^%{DBr*zB+H_;SOC)
zXp>;@Za=ZyuzJYw#1771DiP&;F6Zg2UDlA&w@9{)w5%Bmt=3OXMHqiv-a9E=>}QA?
z##vG`DLIsX_%`nDBhx8|O#~|R&sh?>+{w~YK~;1F)pH;Q?`j@Nzql0As8`~n<@|%n
zB|}$A;%<=ke(rpynIdz&L*fwA%7$Huc>#g&$`J2Fr3}2g1p2@2bj=)CU!C^*zC6_n
zb^KSr{{QThi|udmzlOV!=612i>?PZ5+aqf4rpQcXJoEc{#V^4a!BknMj&WCYJS@4w
zi{bqoT0oX7$)JE?){X~TP&&QEgN!0VJC=MlujVYPLKpV^H6HW%v}=md(B;~pfimT%
z2cZsc*GMZ@#s}!e9sZrZyYQFY%z;%s@j;RKi?!wZ+xy4QECNU#Dj=xu<X&3b01h>B
z{3QtcX$6NdmT->k%Z)(Qqii@IWRb>IfPT@|Cc0Hk!;m_PxbAl`4H>JiAjl`6XUF8x
ziTA<UE9*t|eJYN(;Lkp#B-QbULLy**>MbxmmGenxSdzTe_PXX!DMhXq5R1KW?fFi0
z$!o4+e%P7p`exMJ@}uakA?9moVjN+|gtC#7>3LW28H-orJ65^fEyO8yzL`_5tnSwv
zf()Dyo>*1bq5yAX%ex;NCwW}oBwhex@v)82TV$Tv!S9FLd~0~0+5OF1zJBTcJN=gu
zghF_nA^^}4fj?xK+btbsyeenQMa*;I3l|D->5ls4g~;OL<axEl6gTxam8XwRuY&1R
z0;N_IpcL<uZ`?fgqw!j6Md+MiRq<EFjB6K*Cp<s-0e+EPyZv`eZohn{M}lH?);4J%
zn=H=lri*(b#jVyC3*%v(GFMoiP6`?#iksT<1#n`cC~artIC5#3_>`f|A_iWQ5CbOK
z*YMBO5PA@HITcI;Tn?%5w0F(Q+pquDq+7GB<ZCbKGi&}$MN^Z4*)or<I>W#$#f~QD
z`GZQXmfIsxm9Gj(7!|_}!}J~7rX(5>*0@<xxJvRpk9ZbmBZyz6pKKlHRZrHHGfl($
zoON=|GHCh8_6h1$9@(~I`FA|!YT91s#fYq!?vWH*(u42Nq^4w~O1&kkt00uHQQA14
z<;a93NRwCJV&w&$>TS!D-QR^IT~fJC(GsN<;(SjN+&HhuvSoem#rYa)D0>0fB1FQI
zu?wb9dD%=DMHN+HSYZi1RywRgM1;{4@#oRDbZVgyT>+(<s5;-91T~6Gj|rd4wS6v`
zQPQr^+1a(*`RCfa_R)MaYDjn?dqPtiot2+YV1b9`$>bc<gFU}OPuwlPZ-MYELM`Y*
z;WDZ9c#YrE?XwS#(&OD{oSb#3D{KvsBWAfi5KOaPUuCU;x-3l{bz7BBUs%-qe7g=e
zYBQD`wcEm8*o#a?d7vf^F1F9Uv~+F$hFB6QH#LU?XK}c~F$8gR<)j`ARR6|V|K*UK
z;DTaII#v#kFFNbaGVAu`hY0@J>8rc#+#Ja>vE6(AvU~bZbg>qf_{VV$LO>oBp?-h!
z>qOUslKGphX}$=Ia&jmd$v1fisQ9ei+%`N;IJftt+a)Tj>IL5!gZL-@_3qq~5}NBp
z8ocXkCs)_k+I>bHT@#&{229fJ*`sprx=GzzM{-;P;~K_-B8Y@L07Lks?U`V8W{uve
zFDpTBkScr`R!55DU<DAaj2kIBPA9$x9ec$3wX7<KoWN|rLvOwNm(->Tp%i0eW7`FG
zV4*U6(%%hBp}LLsm3sv242_mg)(8x5rmyeNd_0ddKA|gNBa~P8@-`t4DP{cfTWI;T
zzC~dK!Znnz4fbU#-VpPW@6vbuWL&D0&?Qk-Qwf2s?5K)xH!C8?r>A^ijn`Y4I&Gti
z-ST*&m^;@mGCo3cU*;4oKVEP=+xEV$WPLk3BiCi3(qyr59+a61hqI5WmkzTr7Y<RQ
zzMeB%ygMg!Z9nHLe`F~~8o@hIze?`rD{P=>s-X&7tyR-hjuh357_0Kc-*yuU7|_JV
zPolyyyj(DiVuU(Pp|xKjd}sS*;s!k?l~@n3g4KPKus}~6VZhkx?(y<cxS|XuR1WZ;
zbNoP}Cbi(rWtD@P?gN~&AV91R8DYo~>2mYJ^_`xBz^M4iq}Big%`+LZ_JKo&g7;=P
zXJr+>w{;)r>jPmfDZw|r>7d1;uUYfi2XIRLYzco(8rA>m17)UTZq7>_;SV{pmO(BV
zB_+2al0rUSbIfbU$zju&)AYJ8d=pPHEv$}u{)%KKQ;pd-;^7}0Mevk(x|pL`oEOlI
z4Lu32w?OAl>ekPB=}1Uv)E1Y_;(YQK2(Ge9t1%AhKC_K)lzj0A66xD7Gx@-AI!X?&
z(Go{#08VOrN5G;=#Z@%@qAI@*_eC@o6S08IL9n8;OLnUrs<@Hw@^SC0*T`fjZYXaB
zxQDQSUkdk&z>o+GnoGs@*Z{5{D`7?m$KVGQTa)ZBeq^w#G@=-fO(pk=Ohj0qKXL1&
z`1Nj}NH1(F7M$O1GitfW=H$Vlxec>>=-l5!zfanaM;>+{@9Yq&0oiVH(l<#O@905x
zY|R;kcB*XKLxmq|tjTejmeLp)v7ssOCS?NAVk<U+`)Pl->D3pBqK~zwxf3Z5P3!Cg
z9I_)lFodiL5e#<9<VB-$!`)Kv$x`v5N2<7i|FK+CbRTU~6Fkyw7;P=n<9H0Aej;i)
z<NOD>?CP;I7Cdm+X$+jC0Mnr6YT2Od>GJtRJUy4TguTj8sURRVBw-u?)lfCKG{aN{
zxv_ZV>gQ)|^t5{V3SS=fA56^2y$(KkzhRu#=t%#Tcr@>c!u6{E1AwQ=povbS+u?+A
z<_2#(HAcx1xl`?L-?ez!omR3(DU8q>JQOhzexniQ_1g+0=-ZoM7jrmbXu6A95>-<0
zoB=U2K1l(Noq&+6eU-Ar8OePa=;&=r47r$Y3*V4pz0m`arz}(rq$Raauoqd-of6NT
zZf+i3wMDIRcmSK6E7kbb5FijUz~M!El)(=!0H)k(R$S9*ZyAW$5n5Nwz)K-GfyLKl
z7%TBR*7j{{=clneuIX%ryFu^$-P6{$7w;4N@tB%C&E&K^zm1*w1V=J__Iry|Dxt7U
zfutWEHwe2Ob@`cBippP_YiS}lW2RN1Md`{h$_3EyvI;7PR0yS`PNhVQ@fPTWJo~om
z!k|QF^yuz<30?IFboE1I178E03_qmH5!Y^*Fa{c@nnXk!^GdN~uxJ3qL9xeCu#)bx
zqyJ<&@ba8$b#v*a(5o9<_l+Y906JOu6;jbeDf0X5hp-&7mTPpup*C8Q{HNKL(V&N_
zsiLTzE`BojkVX>5$h|Prk?d=#>I-sWzyw=C5;#X;lLBhsC^gGl@*TnSmY(A|dU#^=
zu=j@U3y?9V_H=1`ldi(f;U7j}_x-<{AIvO}KAT)9M!3+#2h%RU7)g2)nd%Z&8AB|~
z#@Gu|VM0Mh-Z*Yy+2QVHml!Y0k%4X@`ZNuKJY(Oai%=z@K}y!S!WWSzk+PC>j({$<
z2&GtjZuLb2;JY-CDuZq!Hg)$*Mt>86!_?gL`|{UBiT4TPLm<R=esUTCELSZr$<Mxr
ztDELmb34%<I1xRzT;w%2@yQ0!%I+(~N|69XbVL_l#1BGh-XU0tSNQ2Gag}A0t8REu
z2F_B_xV4MoK8ZMzFB4U5nfyTRP<b8kG@{*312_B9i%yCi$DlC<|Lin7W2wtEA(RXS
zbA!aU_iAfZn2eKu%)&Ea@l$KlewEF>fTcElC9ro-*!yJAMR<Muc`9AZala$4hJR1e
zd2+wMPCg2dDHvL=U+<nFA!tq9$eoJI$&p33t>_2h?57}9#}Ko|`SJf(AIPTvt8A?1
zwqw(8HH*IbE)8=>!I#$Ad3Asd9D|1@VbOuVdlGz1FU~#vA?Owd5$isUekkHxEy^hq
zH#LzHCbV&+21!vD8L93wMYRc@OOnmZFK?{4C2dRq@+_hx#!Rz}Z_aDlA+2JCa3e)u
z21D=q&)rOKYd@+Tth}F|5E43(u-z;-`b2#FAYPMcR+ZNIM)H&=+wrz226!z+HC}Cf
zj>kt)6yePNMWd)i=P<us{w_+SIiRo7ch9>(&r9YP>%3UOAvV-7rr<DWne{&q<?5G;
z699Ejd}Im?CXCSostax&7k%@W_214M^E{uVlDw-OThd@H@u0MwPFUi{vh!?@9arh>
z<o-2lVP1TG;{cyrpSDWjp;uyx=$^vYrm25)$>vqjq^_22($69N`V})C8R@KF?eN_a
z%V-JBVz*GY1#8Z8hN(lwW$E6VAMe$XNwoP*+hThq@i(;zV*(oUZI#WWTV{YB0Y+-G
zJWNT_FV-ke4~O&?^P&R0&=&8nTTs5}1yDQ=#S$99-SE|3k982tbiF`n3x3Bc|6M&5
z8+{qrGn)+jcp?;_mQ9WuFi!}59<X{MFh(d=uc&DdorepP?2SScuFToTH1{t$Bsu;^
z68fPyFvnFcX<S7p;QBd}lS|5SOB{KT;4DqkOSFJ`*g3;8i^sL_(&}F%FSHc3`ZS)F
zFPQ^Ldbai29f`+dRK*@KPPoc`{x-42|C(jzV&84&oWy0}YbR<+XwfBe#Cip1^O>#h
zP7|fJPA@!-ur4;Gu2=#o+D)@vOKVHVr(&eVqx19#Rd@X=&bYLDyfbdBLw8@1{44h%
z=L)`o8&@BSggDlXmuyj&y&t?;8}W5y>pop}LoxnFoWh>CVG~;f^ei-yC!D+cl3`O8
z7EzPJR2mXNu2e=I2M<A8sz;$H!H2=yJHl@3Uvh&^_`S|AM0R%Yi(-{}*^phcOx&H%
z6Alovz85c0V^0nE<G|FbHrUe2>YBa<u);v!5apN;2r8(^WdB0PxBi6r=9$)SgPm3L
zyNh4dU3cEMFLH_NAj)ffJ+enLg9~~iUw@t}q2cVI9%aJ1!ejDkBd(WEPJ>gOy$Ki3
z7~9{iozinholM3FPkLf$S{#DD?s}RsxmAP7)$2Z7DAhn=hr#W_k^b}JD2oB0thN)g
zpR$&e?vy{apX_&Q6({)!BWCqH_dYsx3XGD}WDg2{CK%KsZFSIK2td@k^=Mu5^53Ps
z|HogRQz5xU#xyaBMH}?nabaP$&nIdN^fNY@&b2OnKRb2*O6cP21v}~nMf}>9#b9Cg
z!Rq+~w%YW4rnW9(vJm?+8qepV<sF(x{%W25>wTDlo&N!5-<+shY)PnPzCqRWc4J!)
zpJj!#id6X;C{EQ1!Qly{mo?1l^ZS2nodrXj&6aKhK^lkP!Ciy9y9aj&?(XjH?(Pyi
zxCaRCPH=C$ahKcQoI7*R%-p}wT~+U{y`Qye{C)~cQZgLhH7keb;R`D(f+qG{os1Rn
zCoZT;l+t;kjv8L_qHS})&TwL`YwZl)wK#@4-_;Q|N^##(?TUeRsM;+c3`+(%25D=Y
zEBFQ`N<>JL_p(P3=b_MgOg7ME@X+O67XH}(hfYrAEbm0*;n!-PJJS|qcZUJC7>=s|
zRW5T#PxXNGq@4?aY{8d^15g)|iVA?JjYC-^MCb$%Og_#J2#q3*fZnvM-CEsm#LIX5
zm_OBhOLn&q@JKH_H&UXmk(P<7#E4m(P(yFYrn##aJwCSln##k`03c#c;a?41eE<i@
zCcFgJxlD?A3pP%#UI)kNCns55(lkY+E%Cuf7(FUnpw|d+Gx!E_4-;!fuONyg@+NMb
z#|b!6J%E9xrKtOnbSey03or6Tc2adoLKS1_b_Qd%&Rx^qVKZ=f@I3$!e}x<>dK^??
zLop^p?UIlure*e@ms{}v{V$9FU4~d~8&^)BYy8@p@z>MagCglCa@7TB5)L?o3g^Bl
zs^!WY?#WjTy}pGqY1+klj}UEM5z^v4EQ^wThw_0%#@Z`$&@0^5%PeonU(~34kp>5u
zp#Hs3VH)c6Dw}f~l5|oTCCtz=-P%BGsX8U4$cuvUsVR<0x8#cg*>b`C(nSQRN&wCD
zED}H0B(|ldBjU0n?K1zq4vok!AD3jJ-Yz_7c!;iu-52P&vhbiBnpy3JdbVjWrgJJ{
za?~~OG(!`r0sWWF29gSb_h)N3$ow0C*J{`g1N3gBA`^n;O;eGpUF3KGFA4G&o4x%k
z!lxScsY}u{=<qz7oJxVo^??4#A;L-yvq3QdT8LH^4A$3>=e_0f3v@nnFMsSx`n<mp
z&D((iKQA#Zw!GB^3}qeFg<zS?U-uIWv_8N7Qj(VGSrv$jRSHB475IwL95*iFIoQ`t
z>Is$#;xL_+=kPm*X;yA8qYd}N8qOx_o9?VR%y`E<?1wsZP#q~t^DTL^oQ}YYd+p)t
z;Y&Uj^OG7N%8@{8ThYp>{P2WWdq)M8UCCnzp!kI{=_|`91~?h$;!KjI63?M&foCbd
zU#PAj1pA&W=>Drx7W~2G-z1w0$X=91b&M=+?UgY%&fmQ2#KCL(93;et*lHMmZ{W?X
zBj}kvHph4O7dj+gqUcg4p5Uii2z|)o#Ej$K$v1(=BXb`RtiSx%2waOAO)YJBnb{#?
zv9><S$lo0~p@XI&Nj5gNsK;LrSfl$gMG8_Rh`Ck-4UC<G-C9M>=u<1T%F5Cs_5dJw
zk}=3B5vX5#)V(}Ev*VQuhg*HZAB3+wt0j%k)8|Sz+k^BSyAp{}T?QW1JRlkoZul$+
zLv^~D4Fz)e<B<EGXd%E32;)U-8j2}GSwz0X7I)m$({89jGBdrNeuHvpFA$+q2!_v5
z)R&!0RXriWy`#PONFd&RV>AFXE5AdEMEbh`WAKl5Sa1{nc5xM_=jBeh?g!^qDYp%Y
zYj$QT4}vT(VQ4p1NaBmV^MYLJI(OcWY)P@82M&}9a?>fsnq7FBP_8Ic(ZSpS!qd+0
zUfqOCT(q&4hvtWsL*S3c5NuuyDsk>G*u0@H?w$?LoX);Oj1B1oOHn5Ts746)wF}lu
z2iT*Yu0PhBpY@S`j~Akh>e`q38hyKJn)5wsU?BWW*~eB<5k(H0BIv{?FZ@^R61=g<
zF}2GO|9=|^{v%!(N#_v*10DGLxAauKPik>34<lK&KEiA+oALIIdT^vr!IQQe0`DQm
z2-s-IAxS$9)OIDx(4^<!NbJkI0Iv@3>t#78VU50m!I<0GE7qJ*f4>+-V0XI!&qY7S
zToYkWQ|5qnyo#!R>FV#ILPff@+@kqRA~6y2-}TM?r45}X?;kFzMXHra+@tq)(Fk`m
zyVOUsIQEVlCGxp+n<_g2iD|PViA9<ipn(W{>jCR7X^Cd<C}g}3#pjl`4RcF)k6mgr
zL&r$Ns_!2LtP7hp13avR@28@(L?C=hPQzl1Y>Fl`x-LOGI)7cfEAeV5L*a$0iQZ?(
zx3?cQRp@iKgEs{jfJ^u$R1?@WeBI8`*G%>lXiR9h&y_*C0Sv1Tu-yU=%R~<K&cB~g
zHVf<AE2q<D4l@3-Vry}~VZF3&n%4Jz-#j;7#XYcqvnIH+%Eg}kQ}Xor{jrg^dUMmL
zP@N*OKyak3!+E6nMd!A|S5=v^R(-U3`+f=HM#a@i+bCdx*9jPjxgRnxkCg{c;7Bq^
zNP0!=Y0&*!QwnLH!-3ML;Xxd2ts|VYKcMiknE@{dD%C=URkrPdLhb{3%4{%@NMckI
zTLLrGQ7%n5E3pjsFVjFqqJ>q+LowDH6>8sqZDeqVO7%Y&?ucDC5b<3Y*W84FcV;Rl
zX!hkI-qCZ$hnbW<OJuuL7$KT1=q<i(*2_D%6g|D6K~xFwR!PhhakUOk1-xA(ysWwZ
z<y|dRuQIfqI&Lg~jMe{8^Sj+6a$K=XJS0gy4z!~J7Ejwb9{dGbvVEnmtqw@h?evPR
zQ4lQfbP?mg3BXV-KS?RKB4-hn8Xx8aIhX_UukKp64r7<~%9N!=Otia{r285RE^|@$
zeg&ls|DY-R6!49-H#{5!UwFw<vKAa;o+y78q$y56#dzucak(Wv7BLT;az4L_e+vFM
z`50s3<T5U_XvGr$xrG~!YnDNek0{XUfwy`oxi9OWnJkzDj{1&6OC2i_Dw{{7mO0s_
zULa_!JpY|0K{0S6#kctqsDV|9*bm@YuATRL;$C0HoQt9;#GwG+XHyb*?fjjR2jsI4
z@3Fd<ewdl33*Be(_6|93y%IUTZJ-<xZBp^Z;O_rm$yWwVfw6uh?K)jEQ?oSSv<We^
zU~k9Zrup(Qr%uMsr*Lx1g-G)gv0qSgzg$mYJ|^ArJN5mOdtVM^Cl+LnjGPf@6rltU
zQODp#63|IYli+b-Q4n(hA|=ROC<k-|I6UCMf;LuBdhsN>094c50#>=_le-@_g<txj
zGxkb{W1#y-_y;;A2%Hy37C<!78*v^l;5FdFF^cIcg?N<*sA9-aZ;#yEHuwtkasO9I
zt^ghupDRmh!dZV?!ZFDq9!I0ve1;F_1U0z7Kw!99M7en!V`^jG93XeH*{}C?S<TWC
zO2R`1a>fOB6WDEa2^Y8gp6w%9bdpn+*5q|g`)7a<AoyG?Xyk4l@1i-OM+V?H#CnhZ
zI6>iFvA&Vr^2$CqfL2xByN=f1xh*A`G)ta*alc}4!7%kC$|>47zquM-Zgc&hTHo^c
zy=j+h@AFngPcH{ymQxn$pdv`<N;<F`KL-j8x^26#`zF$Hg{lW@B4LoY=k|zJ(X_N+
z#XgnGJG_2IdUwoiW@gszcIO<I1#7}V_3NvBJg-9xoXmb1&Z1FNwy#GmKI1^+{?i`O
zJ(+)2(OMP0v}#GF@6(s_J6IQv7Sw*C@}_PfkJuy_8v2%v;?DPgzSFQI29iN$gg#&x
zY&P$Dw<m&H4ePofd99if(&X(iozLk5IF$UB+*{z^_Pr(DaP5ojv#R{Wj##ANo4Nl`
z{Th&0|ATg17J;7ar0Y#4bzt*rXE*@|hZ~a!Nz+gUMzzt*<X=zF4`gsvG#P8#;+FmT
zX5Za+oSNylfa$>*s9$VC9hp)}?9TFAL&!`-Tc8!$n4p~*U>eyh0UwDOBoE(?@S3Xx
zfgnUHU71$z<&^ms4BV${WaQ-@zQ=Bv%~%@dh<kD$I`1zf0mIztrisVvmy%4w)^So&
z_K_7K(u3)}A;Hzcf15D&KauI3L}i;M(h_p>W#u0J;ndr)`1s^HV<@MJO43>tA_NC3
zG1t{CpN0(Ct{ffo>!{uHoR!uzu#vwcqdX^t%l{CgRC_}3S^ss~f&=!<hRD%nWiU&L
zoSa4+4QpJleAxdb+<!>$We=oWORh>wL7NKC$~r$WIn~g{G%&G%F~{BF#<Q-y1&qdP
zZSQSXFr)uC?EJ_YDE_Qn^SS5d+}`PA0bZG>nt!Rl(*xZD%C6)j7r5d>?truTQUSwQ
z^4mkWv_-6W2`8CRveSTVooCv)>&7CtzQZj}|0%Y2_~Ct5ft!nmJTiTVOoEWo7G}mA
zn#d3XoPOfuB6`G8=ohiv<yt(;kyKF2@V(=Ho)rDVgzyTf)<uQ4+EfZ?yyw9~0c&;2
zY(C>cFWPkN?I{S9E8XKHqrpXtVVW@rimEdEy!gMyAGIQ#zflIiiQ4qWJmi^JhAG>^
zQ7QA(aQMHQ3m(2~y`VW(iYeDB($bZERZ%@Ll~Hl7IKiKiY`RD0*4;9GSbbpwfjJma
zyZ3Sbp>2zJ^~OQa``EI_`zx(fU}KMB6yw6K>CCU|0Ngt^Y00z$K#&*LW^$~~B84cU
zX2le9{-2!!BH^Ifwj3d^`z56%wDhT*o%dG1WJbSR){YuNo}YU*INy?WWX<S@V17Tq
z(paXEmHw8N&cmbEdb}LXTY!E2_pl+}KL-aec-7ItX+{~2+kCYDz+BV&N+8!=K&WjO
zCK`cuQv|hmMkLAZY<6p-P&Z#(TCzI$YjYe%Yn}hl)V;6SS1L_iO>*PJ<=*G7RO%Y-
z{vdu1ki8L$XlZqH13}<NBCf0%fi^BF#Zh_KfEfx2GMbG+H3hqPG<y&mXGuF9X;lK-
zw9Ql}QTI4l8Zz$Wcc;-iI2dg`Xv<w9GXOvd#Wu8PzJv!6%smzOUDzl}JnRTHZSGYU
z=;j*QSwfQ(?|Mf%S~n#WlrUJ`W5Ll+F<hNgfy0OgOOw_qW8WMG@r6NAahXL!3~`aE
z#zj;HH3KHWl1?B-7w&=(1`nvgg1w?(i^DTG7;MJ+zt1>IVmT50KmeQJwqu8{l)HI&
z<q)5qw3SF<v&c}A^>VFKQ|otIuej^8FDgd)&ibBl=u}vqADiVaApK#twG<875t*12
z6@%Z%VJ6`;KTRAHNvD;c&oM#_bnv&YNhn!8c>g@`B5Kog3s1}-k9NZ-e@QhKeB*8(
z1t0Z-ud#JkCkM%5BC(NkvquHk&>DZDi9-v7#D30H!TaQ_d(8zU@ayF6_PY>#ld!!Y
zNo5p!^vyuWAx9zwdHVUmd@OR-4>fo`KqzsiXow7_A~mvL0&sooa-em<fcY3)A8=s!
z{(r+GKnM~{nW){5ep%v#RCw#U-igCWq-)slZGL$eR4!~lBmC%g#o#8kh8CAUT#YPh
zDw+(LTI2-0*PFoPk^mUgn>cRlx}6*Mplfq~V6e8P`n$1FHh*jo(`OQ<MAzUPM~_l$
zB|#cIyg?qf4renfb4JsY_ImUq^y6^o5?p0Q7{|^9DIbI31F+`AnGdV9|NDAmeFJo$
z9Cwr(u}fkhhP+HnFf~m>y$@tf@l0L}j$l5$)E}8+2kVD4`bktk8g{zp+hS||kkiBF
zcUng|ZY9HFOdn*iM?#rO$1%4Be&d)vfxureHv$vAc$KE{zQ_WR8QxbclBYTuxyAiJ
zNW(twNQ|ACkxtwDx~<*Y((#mC7BM&xSu4lLK($*~{;*t1+Q<w6$21Lnzr>%^V1hgQ
zc?{ZRFTiG|-xP*(0#z?a%HjCanze8##|4`x)NqGSL^Jc08dDLH3Z~2*%-hDdiTZHV
zAUT@VH;#*7qC%rlL8T;_*YA3K?n!b0zPJivJ{`2x7s*d8gid((amqnBv1(C0p;q`e
z>t~1!s>aOovve-Cp^n^3;5@=@7!7>i4IM&8>*lt?xx&id6-s%GQYqtVaU~b@sLQ)V
z<&wd+%vr)#h7={jWy=5L#QtS&8~GCUvrSoW9fls-qfkmcP0RaE%jWENCv*Gnb`csK
zD8IoWZz7|YT%(YP60tGxiKg+$H-IqQY@Po_l6i5AUhAY@BFQwZxfz|FkukwkG%Og^
z6;8yPH3UG*nztV07_=U?Wj<sWr3$Q|JKou869J4JusGry=j<g<3+x8R3vN%ppS|GY
z<Hv-*Kn@ohZ_8HBE5#6p2vNr-z_HLaHg!oP#b~dbdrT6lx3+LBEaW;PM%~+mm^fZR
z>3bf0Ps399<?kTXx{M$avB^5e0l%%W2gJ1vn|0=gi*`BE!%HQ~c@SQcrY?Gdan@ki
zHb&UQ8oBnsD?*_m-`uzHf95V*W<aLx!{=V!d?Mr-zwEE~BzxpzP?#k=YzjwC_Ozk{
zR>(0Lgm-34A-m1M?j%aX<)bidaBw)c;pbqst%DgX0+$9$lx#2#6c%MRs{{oN7+?I#
z8eKI)O^WBT@^l8f-un*ce`)VGHgtj3fC`qZHDC}Jgjh^^F%4FbkigSGVg}Npi<|0Y
zWS1AIK0d^R@DjBsK98~nyQUS(=LaHvfzwdOep<z=W8oJRb#EQ6PsZXaSG>kGR6Gs$
z(K|A0zj`blZvA&6L;#EcmQJnm0kCB5n1HCIYz<KnMexzndL=C_0VIes;vw1mq)n5<
zgsZLKLDz;joOsCK`g6eTYUk^b!bb`p!S$#)Q8-Vl(iw=*VZrKZ(%vt-wda{VNjY*=
z6{!S8DNKE@78fXzgVu%=no)pC0<HJ@moYCB2!@BWo0TK@zHYv41FP*ffzv%N*d~^Q
ze+avWb;4Wia(IyNKB2-_l9ogb=<-!&Wad)Qq>ui>&o(VAF0JV64BF>jg(`Mr4mM2Z
zQ*cCviaIzz?|MGTrl+r+Ilmui#Mg1)>3YzQ!jRZ{*!lk4mWCTWN5W4)3Rd(rsiv4@
z>~Oj6V2qB5D2}?3qVh`Gt_)TbYz}=!c6XN!DG4)u9F#oszhbTR1I06YP6cOc&{zky
zCdE$jU!0#^P3Y+FEY|giVNm=@<xZdxp&H@p(~4Hg@}d3|`7pfa+@)Ax<zeo4j7l>e
z0GJkpi5Gi{8JfUn(IU3KUwMC&JiFlA?tM=tnkO(5e`Vv@9v}hv6iP5cqVddSNezoB
z#b8Du%&RPrP0m0Y-2rZ=!cmaEiORknCERb$Uo;=%1rG)%@PXBl%liJQm4h&rbqe+w
z%o7tHU5Ftlv&Q*?E=>G^jeh^5W#>`JDpeWjg4+czFF`DhvyD#ySNa!PUbj7j4%7K0
zgbbxB1T}Bub1p+J??a>mTh>El?R(1L1&JS*!Hg}s<>=&TK54aqzjJtB{p3j`yPp3I
z-s_V6+vmf>bS~FJ^5#nT5vKepbdHXPLLXwnD87O=N5LkDSJ6(c5}j`xvRZmMWKnE2
zZrLb;e)*G=zMI}4vK%-9=1t7(Net0U2u92!Kpoi(MNW*cgsQcxoL!sq4Hl>1Q_RH<
z#Ac>qerw_BAePl^@src}6p7B-DLw(mF;WLHxD}UoKv(CRT6yd_$2N?Ehg=)3JXnnP
zULgup;^_O}9d`~_&r((;Ir@n`8xj#wLO#nZT|70>j_|0=tS8}R10h91unO_Uu+h1e
z(vY*VRq%Was|pj&G&0Kj%_pVa8OTGL-gM7Xq7{!$wdC<M$@zHKd!v(LKE={}&vNBL
z&_vma+dYzW2noXUh`$UaE_onk6G=^wcHKAq<sy|=3PHbAB$mEXPM@bw!mx~carVpk
z4*kOJ{Kj1H1Je3!1-PUlrpCY&7xITLcJXUbi0&U^%o}OB3%;%1#?gEqm+;)BPvOI#
z+QS!~8}xmI<7{2;pFDhn@m_peym9padR+clKz;8>6R#+FL5fNG7{QtTJ~SwmMa@g6
zTLGf~d5BA;^CTWvq@o^I5C=RnKbEFg)CkL*m@6>5A0z5zT?KgqXSh&<M$GWnUt{(!
zU?ZkK*ymlidNub#&SwAeZ$Cu?+dXeo!rmArgBiMuk#i3r!oy*GfV{m^pw|um`0;c-
z15;P-^{D7j`g`p@5T@~PJk<NpbkN5fC3Oq62!J+~rCdH&2r%7!$BR#Re!UZ(@8Ge5
z$oV1y{px|N8v)a;$c3Rt9Tb&3r0oFC5}Q?i|4tHOumWmqe*0wYoWly}bF6A|y?y4t
z*ysfJ`#E@dJ3|452Bw~|e=|c9f?CaT;+mJ@tU)O5w|s~J!_3@FDJ~*3GK9}P#h>-V
zmi#Daa<vQHo)Ps9g%ai6kVBzq6a|by7wl#|j{!&B?p(rCrGG8UcF86-3DV?q%|M%K
zx~paW3CtD`mW1cOdK!n9d!CRspHsKY+|Dt7-dI2K4Y$OP!1yvjzA|TvN1$z)MPPe^
z8?2X`iKW4gF6K3Tyl}`xl|1trdEqN_s~z%70^7Onbyo(xm=UAmDK<ZFNcQ3O5YHom
z$tbV_Zy{9x8G3p7hL*K~2|?;+j~2j$LT8hukUEVoVxLY7xb_>f-Jfo!koBuU$*bGr
zjn?HA-xU&pVc-FrKJPC+s*j}U&!hL)=cm1K|5N_*mEwAXVBh64`>q~eQ@tW8G=G~+
zD69S({Bq0vrDSmZkq|vGW`fkPjeYRr<}^(tVmT(Z0%L;P^S)OWU0t_W-PTTS?<2$3
z>kwljxKfm%jj_3=^1tTqxw^BTvubJcx=X$aQnU%KCyAZ}iNfRkS-CO{d<;cMF{3$f
z3L0)~X&LK#hv2mH>wBFbCGvm8)&F?L^H8#RM{8BZ*peJEdI)so&iA<{d3xb?#6TB^
zYu3)iPCMl0S_f{p`wzADzS)i)gjh9WY43%L#>F8AdwG1HwU7%$ZPNZ$&5j6*YiWgx
ztkMJ^;9JtZ<|DF)0e4t&uK7^TK?=k{AX~fo`WfQ5#VZF;KM*+rKL15#Vs~6)g>Plp
zeXbvm@Efs4kQA5t3$N;hpRk%{j&fH+N0hzW6N1vtpXlktvd<UYFUCtf*GO?oYb2(O
z+Fc*r4<7H6%Nc?^`wnmDHmf8cLg>m**bAMZ>LJXm;xc+EeGgZ)@d@)DoO}$-05lm6
zy*)8`Rhqr-V#j+em>PcMi>(q>{q5X`x+~D@wZ^50k+^|YOsndZZxkHQ{?auz$nQ7r
zl2R1en9_u5_gLf;!ulrfcD(<+z(=zFJ4|C_nR?Ba)lH;ad01=uQku!*1Iv(*++dFO
zW-yP#adQhJ+JvfUJbLXTm%~ApkNs79sDqgZ==gsF73TkZ9RkZy{xL#zK27!Rq?SS(
z<)WUpN%_Axh~9l6Nh<i*<BLtE((--7Z0UI+i$Q`7z9&ZaI9zo0xtl~tX%a3zLQ$cW
z4qfcnRs9IKtr$dL)Z4#U1l@q=IyYym(+<Gss+=~5#B~=f{odhY6?2cnO!Jjhf!|1?
zP$UpE__R^MOogsE*lpM%P|ySqGv7W85k9B!Pr8Lm_(r9MQW&8%xVL?`3IlV$>2+Pt
zETv9x@kqoEzcOA=qtqWS=XiGK4$Kopl90mj+1GH$j!;-@^?SdrHg)=*G!M1Smu`Ws
z`y@fXx*KDc)TVNM!@Rl!zggW)d4>gZ&>`3>M~TqgRcsPUWC3CtOe)F738r8~PWM(}
z1^vS9Yw49{F^x(gVEc``!3F*oiY~xG(DN96<23*9&h!?7mqtk60<Bxu`<e86ZaM=?
zDL(ut({u*oqGq*5=2W<eq^-QJz&Oyo^>6zdjkEQ1y}NIEW+MXP_wYMdcdiDvux=F%
zu9erAJQp$P;0hL@3dBsgz-B2>{m`Cx5b9SND-Yx74YZE%5NA2=K#L_ub{w~R=@SH1
zpnQQzrk*#7@r%#xAk3EE3$)b%6p(<s0Pts5F$>N(DI`D{<LVh<TKOh!huFAk(bQs_
zn)UT@6{)Kg-zA)$^TeXGzMYJ~#K!NMdt<;hrS|$b5`lb?XNsppl?yl;Mu26OS6ae0
z_1E~<?Feuw5igv1FlS(vFMqiH-;2h7yb+}roB7=XV`I1{`koMI;o03z2Hm0&3AAE>
z>c$Lmqo-!z$U}>zzXuYdQ<=i|tD)uI{$q2EOVZDHFjSH<hPxTZ;GD;8_UrAuweHhK
z@4#H9LS|;JnX!3%>LBh?-H(IJ8|;WLeeZ_R@~!dAPabxK7{mJ_<`a&&S-aEGO$`2U
zD3CLWyAyxs7id-J87Enf2%VnhDC{=c#k20~MaTO#A?T<S!d$9ZxsLk}A^4({BK2vd
zYt+e<qb$6#1_z;cEBbe4A=cJVGJY=OB0s0dVupN5#=C!bs9gTa=YL?D62<L)&tg8!
zKkmRKW<6=ZW#I3Q{taf&!VIG_r0@@xaa6@r*j%JP<aftb|7QlU0Lh}u1Fi5A%2@Cf
zGL6U!J7mrs&bLg@clvt=@DvTtLm5lG?D1Sf?{Kh@j`(wYpO+*(R5SU`!YyK*iRA0i
z5COw`Zk@CL6<N^<eQq)T&-q6D^%b+Lbx22g={161!r1vp6^l?_UrE&<Mc_BgabvzV
zSN|@>ZDIe77oJfSqw?1yUi}M&j$FP+YDJ5ftsxAn1+(`vsbuzp!{317Kr?e-z+9yo
z-<M;0b)IVCipX)rJT##aiW^qfpUTQuK`NU{Dp}qylIvTWMXEAfUlB8Emqn}uT5s=l
zj^^-zy<TvYZ01urz*6(tnKM)J#G*x_`esie8(3o}IVH*jk${x_kfSNAS3HTj;ii9E
zEzv-aF$XZ7$m6<jknDdYcquokqMc>FP2Uz#{{VRzst!QW%2Lj>b9i6&k35z1*41OU
z8PiFxTrkcdAP!}qJP6{XY!U%4p_tH44bG=ah0UGo!N&D2$82HbJFCplA-0Fwt?h{e
z*$f{cW2><COd90-Y#ifYh8ax?9*f_fjBWF`FVU#>xJ;``4ktPFflt6XhrJURR#v30
zbH%EqHkWtOMP;$x+{UNA2kym3j<SXjq`j3P_BYg((ClsugHW<*ByiR*oAs7o;NzAo
zEEw5xwfMcDZ(#vrQ#22_-sPDb`s4|-LFjyAnlfQYXnTF*zEwJZ?BGzZK)$ow^HyWz
zuyJ)0Ki6Cwmp!}14+q(Nbo+Se?LT&~qL^7VdxyN!A4Y_9ils~nAaGgq%51Ld9^04_
zsAK%|`aC|(suJrcAjunw<68TdsmrTlcKbUl%7u;7n>VPz<pNR1or75-A;tJ)Q487b
zjLBbVX8L)doQ^GxVb<FL6ZA4-fmtcLqt!eF6t?R-#Ql5EvyZH%4^2XyO7}-PlCQbq
z1Ww-n#uQ86D>`^>Ev5Vf(I&q{*U;?%aZx$)KJj2x!M}g-S!dr_HgIY0b0!aBD%wsx
zBTrxTtxIen*nEzIoV)!(<SUbfJTgbpdSPGBB8Ug$0py<aZl&SG8u3=MYE<Fk-0z#B
zB<g*Ed$Y@*+96T))bl$PuLnc-0}e;OYYB*46axM~G=wS6^~=NKuKqh(x&I#TN;|X@
zJxkIsxdg?m(CVpTjhaPu&GlXH3y1mbCp7L7hyAx4?dHB2;3FGKaIE7IIi)MNJ}>70
z(PqbY)*hi7i)fk3si`TfX0KR@(jjvqzuzZ+8(&xJ-r+q3wi^`$4h<fdT%s*PvGbft
zq4Z`h?xUN0sI#I)+P|tYe;Edc*|i61blf9cV!7hLh$;-qw7?7gu|dG-i%Xra1g{`@
z$2HIBv22@1R$&uykb0aVS@WI4qbKMR{4|20zkRs(gI!V<geOCr-yQ<E#f#4#k7I_r
zT)VYDuh6^wavlD9Sogt_&X{B9{=$)s$0-CKb6i*i0DjvXE_^}MI$C*j#OtDynBZH7
zf8#P6ar6eqM&400--ohSBw`UmRf{6FDfHmU6OdScmS#L1;O4z8;dP~ImfTD61+$f=
zsE+!qRO5u-fklb)O}JaSxuwm1*#nP=S|1!TjI&^g+KF6jyG&}0g55k<>z{)<UxI9)
zjV#XDrUWIXSWy7J+RuR0l|zPF>k(>EOCeJR3{_I4vM-!Wl_Wpa6YjK-L3L!R1en$`
zu{7jCi#O?`a}W%z_&+)aB{s{R`w`@FdE>va<H+DOb&8Sb!2xy>84_x08a#~K8k@&+
z?_8a-4V~KjZjc+D=h<|=+2l3)pPM7{h&D0tJ?Qx7H<qVqZMBU~9l`*l)5^u5#JzyU
zLE~S1?5Y_{zTmvPNhcoum(u88tt#+g6HeupcM+?zUa6+}KIN{+%1oiOUKWkfjT*F|
z<sx8GyJcj^K}z%IIS1}FTJI*}`b;&$7B<6BhLn7^pES7h;fdG0{1H1Gg{#>s%YEx3
zL*g!Cr+(mR9$V`NcV9D#psBxge5>BaS6J^(dd8ai-q!dA?$}IRv%|v#NU+@70eJtp
zE0`;CR@)$UV2zmCvW99}g^+J$FLt>{o})>%3NG^}%US>29K5^T#vs?gq|0wQ$p>lw
z4uy{;peh)~b8<sEX1gc366F*3{JoRg_jTaBlhg7cYyC&>z+YCz%MH*Q_s5<84Qvj#
za~z#U?F^Ag(<%Yf%<(On6e%XNASXD(Q&WtZ>E*s&YHGXkseO<aAbBC0B|X@dILxe%
zZxkH6X49i#_*|nQU6u7qQ3_Z4TvC8@O6eCj_8iIdWLm|G&_xq+Q>b05s4cC7KK=|5
za;5ZOW+;V#Uhe=XCuc7ozm}ENKP)z0kg~Gcneo)Y=$y~GN_PGI7VB+;(webL?7<*?
zY=S22+&71ne9?0E8Bzq$@q&51X<k81`7<?w7OmgICvPqzG>62lP14*ucrm?4cvIuZ
zDTX+_(A>{6(Yuwio?*|6XZyU^c2&@>!T${_dh3<e$uRb3@_y*JWNy$VI$J!v)9Ip4
z>+!=2evLho_cP+M|1~n`<KcrMw%J_E=N@-%32hC}w&c{Qa6^K4Mp-F0A33O;N0>|Y
zf`66wYy`*<AYh=2p;J2&!v_I*_isZEGKWjmo!42j-Y+3Et}w7fx)fPDhkw@e1G)OS
z!$+M}9UIj+j9dIiqRx|~#((ps0K$V=<C+#Jma^I~l7XLd8c?M}l2pyhG<*Gj<FT^a
z2pZX+;}3LyLk3mP^VzWU_WVAGp02+!h>u-+cp7tNRLadY1kX)5H+`bDjl)3THxzsx
z`=%|Ld5)x#sV5_otK_~fjx-|DkPjG+O=h=Ns$0KaxB4e5&lRlk8(94mQ^4t%<k%Eh
z9hZjXH^J03qnRDF_m6tc6z+b<L&R1no{fOKOk=uHkcmW3k}sb;JoH6IcEd#u!`W+d
zej$9|<ly47nLWJ=B1}y2aQ9&yhRr4A{UYu{_AN)~C;nS!_+pu)?Tmt)<)(F)Gd1W~
zw>!>?9@YPVPb_L(bB<mT^SksMk9_H8JvUsnog>e9Ea8R?8p+NL3Z*Ra$WfbZ#0X82
zR1-h1ECoWFo81Wv5K2{3I)f3<X~~*<+q`u{EC29nCIbE$p(Cgd3*1N`uKmuuZhzg(
zsb9#)KgkQ~8t<Ooq&g=eZ7$Z%H$7Zn{c}LeF#Rb~+WXA^)i-Gf(R}OORaKQLwj-s?
z$jRxKAG$e#=d|Ix0QCs#02{%pTlV8{*khG`dAnB>OA=EPwTm$gNrJe9aqnLfYwz^U
zcQmt=VLS!rJ%&jg{TV@<M3fg|vRxU><<s2>hY2)Cjued?*NFaUiBPG96%p*wc1Jif
ze6<@B<d=UdcK%VSX_wiNvsFS#pR7F2ddt;J6*KC?5UqN2yIN3Gnh<>LCi$5WjSUx1
zo6a;PT&<)$w;@>4!KZ*ReB>n3x=~1&8ll$z>=kWpG^j%{NN`-C=NbI6EMRuJ4Y3xC
zwIQFJ*^X)SV>cN^!0_o0oZlYkoujVz4JjucJZkmyb2|2{{qccu%6I1#4E~Ro?qR&W
zJEce}mg&eHBl%d@JKf_42gMJutu)P_{2Ly&`dZ9@zJ!m}%{^^50^IBP9WQPd5;i6Y
zpWp%t#3GBNw7!VPlh6I4o-MvwVK?p6t!?0AZ87Pfrqj+$;Y$g_pvcPn^i_z=MTnE<
zVe)Dn-F;JE&+vVv9FG_&s6*tliN<;B$tL>A0{4p#$ZJV!$Jf7it<^p9jgqr=!+=K~
z=l-E`Q8@DJstn)3Wio%1xnJl}-a}tx_>5ja{k;Pdmjf%>pQ`0rCY<}DJCwH5TJJVs
z-pk`}SOKsr6Ubb)J4%(m-v>*yzqG`$rlX^(7P^!*>k;Cb#)K$6A5%n+6DJX3_~^>b
zorBY3$JC`~^Ax+-0-~=*--e<Nts{7=>k+ao4fZ!eN-;~KXq+Oaut=fEwv=8_<T{NA
z8NB%{19ZsiPfil9o9-X`5(q>eM(a$5v(}T6{gbmrbmJySdi&W+olL$b|EdNkJzU4a
zklxNkFn3f-sy1;!hiy??dz}Y*V%WTScUReNbU6{|c@1UN?S6%C`QpNBJ7zi7ch8aB
z*H_IkRJFIf)F8ps)+||bT2UvV3CuVHp+5`gEW1$;Q4G4|w_O0la{FbB4l1ZbhxFg^
zH@u02^Sa~L5fnA)3s7$CVh)<=mNSs4O1eL*E9CD#{%}vc6a4Mxj;^D&{(NoU_<Z@2
z$Z0?4Z%d?tBU6(puae{Xm{^nonQL8^jInitY+VdJm07pA<ntr2CLvs0aLZ~6m^00t
z?<@fHnR?Pmy5ZGcz~q7?76CJd=kaZ?2j(6Iw$O`O!l#gh?6hpv4-!xur5Nl{S2*eB
z9+Q-Jsn8`!M^Z+=a(wC}ZA~-l)y><w70e3*Q{o=F-l)q)ELTE^xOZ==be6SvJyDXa
z)2^Bn(n@ac-LLSzrRW>K9NNAQ3O;T<?v@J-cE|IL9agGHX*ikqCg1meiSIc-rv~Gm
zceHD3Yhb>&#cs>Q)HGH@-v^mSS)})mMuJ>f*W;^f*C_#!iFNt3#|H<?+MY{S+$_Z}
z$i)idTw*!N6#cjFISf0uynSYxuYZ*)E2XH=YuDTQr`|bab161IR>HM(1^Rb)NfrDL
zr0srmN~9rrF7pzSWK!DL+ZHyHP0HGV>{L=%6@C{{gMUp1{f>SkcQ{NBO*nO)903e#
zmy1}nk8CAph=PVkNJtE1VQ$=pmsi=4@o!xc(M7bsE%>$^PYmQ(r$$tRufR5SmR<Ge
z&_U>u6x9aFMC45T+ar3$`JapJE#1SjgOoCquMN)S)Ymy_-=6{}BlNer<!lcie&)tF
z&5@0>{<^}V)e~z=Xux{C?|t;t^VpTO0Y5>}yeBzoD`9OKYt}j5xpl|DH@(+CD0{ZC
zY1lAd*6N%+UWIVhbxj`xd7zULWW{6nky`i5?;{-dq70*Wt->lBeU{QwZ8vqsilbhB
zZcMps+Rw6##sySNt)sj%SqtGJ?$5P97P}_*;BMi@({;)p-_YfiVpcvrKJA~{7wa1%
z&HX$wkP?iRsjf*kBJ&+&Ag6Yu^-a8%COM|tx|2=kdr5hHQ05rFK5ta3-4dq9Blfn=
z)Kkt4H_-V%)}Ft$lb>WXk1>=-<3+v<R|wi!k6UhWT`rYRb!<I5K4COhS}5gDwlI88
zkEV=^3&Zi8r6G7fe4`<VL64|UyYk`0h%~WfOE$Yt@IS=x&n}#lO<x~_@m|LUeUlDJ
zmBr4G9x8sqb<j&CCmm={M3$P`<@={r0riJajI*jvs|o*-U+(l*F@6fl%_WG-=MwMS
zF!i)d43*E0O`S4)h&eDY4+m?TB4bnMF+?X_+9o#_u{{eu7hA(EMv{r;#Zw7eA#us^
zlzu$<%g)#BambVCLJaN1nDKp`Sz8;#CfvGvM`>v2lCpI6@@{SSI##p$(K|&Yc;%bJ
z@0~`K=keu}En*M0!|jnY!;g*;Vf*}1HeqQL>u+jmfmB;uac^FAZFL$Yeyp{oXNFU`
zO%_l%qmL&@Np!;iLY`;`60E@>=k;&&8_Ft4J35l3?A~j<M=a*9wy9i*)ZGbJ4l$or
zHy9?F$QLoA&#?mQMs_o9Iy>=`G5wF%VR}6dIIx4hS7@6Y3oLDxWZ0<KmM?I`&Fik4
zd?@!#9UyopNi{Xe`MSpbcJ6xI)o=hH-+STnx&_IXi_ZKsG|n$*hLm!Mv+>X0NY^V^
z>d1B?uAI}^*O{(gz`C_lkY0|Tz5je7QxGFjKpjnkE%>r#1LJxdsY0z21YwT&_8u^*
zM-;lB3-ZM)U&0#wK2_2>y#1^;#A)v|{UR{&Jo>%X|4J4>)1Vew?vu8J^|!rex{0$S
zrz}F0fUbsJ27dFh$>jbWuk&%a6KN}C)bIbTs$O>P-hnKh$Ry(Z>RG&UYATB5-b=pm
z;W!e`_<|7AjWl|fQ%&?QT#+|M{S}c@EKx<|xWEPN?<P1q8d*Lwk~HxWo$RbDn2^0o
zLQi2_beYJ=7UaI~6Q8Ypc7_7BIuNg(X=(}1SaNPFVoX&{@bTBpY*!0(tPA<S`*R&U
zr{eJuwH~?g($l+Ga%9Im`&!jxNbplmDyPKY#8b&psm!sNo1T*ga>Us6^xv(eThPQi
zZSAgpQ8{Ea6fsBDHg*x`AWXzRQ}TV<YIyebho}k4+-+aPI8NE=69D->z;ZjEk=Kra
z)y0E!jC|kyL6n_qoDR<$9wMYD(aOmtV{Fo>P~_4>p*O&Gmjl{(zE^4@AJB`CP$JM?
z3KQ;yJvy;Yu*H{Eb=c9WOoN=A__t?<<5XW`-LJNuAYHxR*l91hJGUZ8HNK?e^5P4-
zlr?O|IqA1$+z6ADv@w-~S%#nh#iwd_9!=wV5oLR}NZRk5bbkr)>jlnlG%kJsms^n^
zZx51o=Cp=deaC`tGS(GMIIkpM7NW?+W+7WDq=sk)@OaNcTzA!j+vZ6tINO-Me)&UU
zh_8GIE#tYr7sHZSbdzJTgFj|5wizwgnm^bIX-j6qPt53~Ekax^VI^Kl=QThAE_qV)
zACWG8JhATU1vC@Oi6>HGy!l@EuNCt-ymIZMGwAhBmZmc#6WSzSbi*G)q$EfQ3V9Hk
zrX`tji+Etn2ld@);OKehEjq_Hyw`RL3#p~r+suT%tTMlyqIdJ(vb-)C|J$O%<nApN
z&kd(YfeeS@rCX}5CTnNk9^2DDoS^E>7Ef;A!$*yZVgX#P%V|h#lZlGsaFIjs%#RQk
z++z5lCf&nAMcd|NRdR$IMx@T@9VH+?%u1jhaue}w9$u7Eeq7UOP@kg7Zgl=7x|qvH
zZqHF$pMas43#$%;h%FdWUe7q55{ufA?|Z+ujr!TI8JGwkDojl@jR1$n7DIspO1gJz
zYwi^-<KswCIFn%8rIf(Np{e|ZxTT_M8rc|DG*$u97bnmiuFip>Ujk>{c555Uwa!}F
zU7jGN^O*2sKTSrW4|6&0;)vi(_1@Ep>jf$5cKan$<@;pC#_oj(i?wzNZQD0=`(~eV
z*3H|rWR%go>BSdOoAL4*P=49pL@lFl?n%D&8&K@um35Gx#80ak_$t}MX>?kdzVdpx
z#%0A09HRMdI5H5h0^fuQ!0_9jN5oftOlg;ZCaFxH1&APnH(52kK??Ec7i~v5lAS+B
z>6;$$A_za(fXUei>%B9F`P~kQRp&fcC+^8{wTb3HT*4H&{Ts1{BK5UDC>>&ku!ZO*
zfJHy)7H^?a_9gggMS!{RvlJRx=x<Ks)N@57d#3g$mQ`x24R*)|1u{*c?4+V3axEGD
zOQ~yoYqrWMt<5dcm2UZVKbTX*A=QPVfK(O9ie}>;rz=wFMCL3LK%u35PC>UPehs9Z
z?uH1S><MGN`zcm>ey5zZGcF^1M3!s?lSzkvc5tqM5^I?h?XSL1DhGg3!@NOITV!R_
z_5b?tnMd<j!j&3X^-Xx?I`)$yo%Ogx4o;fN*Df?HhRl6#lHlc>_f@iTTgW(UlMagJ
zI~o5bGcD=`V@l_eCVWIzO{qZF?}cDXP&24D_nsgmQ_YHR`5T45E6(c<CclHbKuh1G
zdBZdz23KwfEK2B9lc#pF+Cy7PW^)ta0Lxsx4iVpk!x;JdXNENdOpiw-UfQl0w2~b7
zN86MzFvfs8Zeqqszgp%>u%0wkc07@O@(mBoR&R<}`&e6gK`dU_{u#r~K?jsuaPYXS
z+3I=i;Fe$qT(1anZ4YYbnepN7!3qE`+T-24A<Wric#38||45R54(Zvym^bJ4+g~~J
z=$S)=`TShWtRVQt1~x!?{^-1%;W@Cj--<z<+m@C#G=C3g8gHLrYV!%bIDwSJ(rKfm
zxD&PzJvRHvJEnqTJeq{a+Om=*>0T+m>q)&u)kNW$MM_FkZOD~7W7R?vv#05v+TE8>
zmSR!Gtr0Bf1w!+?8pC9`_(5r7s|c!2j&G%r=rk!Mn{-p$boKE^!{&g4S*~=CpI_yR
zmhzT5APN4o<0D5=k;E@$Urc2@6#N-x6nXxcwmc2%a-KQY4qFhw+(N4ET4pIxj+LB3
zp>0iS&9xg&&g(sL$GQ#mC0@Sxd$Ahu`bjooAV=XL{OpHkC<_s3>Tv%RY@{|peU>&a
zm5pcHm~x!6kNz)@UAFT(GA@LcAd`f|xG!#c9#NY&`Nb6&Sdl<zDcC$$`~5?m*#9h?
z@}1w8_Fs|f^v$<cHN(XTY3&e7pah#&5U`J0rgmTK+-_^2)kQSp(PVZ!;SK4vl8W{x
zAP3_~e;s2eNb8oG|NOfY0Ii}1I0fuQ(1L3y!&Sy3P&^YrpkuQzaL;+!eTDVQRy<d8
zM;M|{*m26c#{}0Tn)V*vct!8@{veH7rIyG{S`v%mx|E8Yf)k4|b<6;@u#Pk#fUm={
z|Fd-z>=Ggm!M1B0X$Y~RKnshiwhB2-Pa;&_?>l$LMVttSB^JV4&%<&3-IKxetJkl7
z>k?1wx(A6=xV){cm9VZ$oNS(10nJh*)Kx*bZ{Nm>is<RaWRee{rCCh5zB)5~Wg%3U
zu?H~w6?-^GRhV<PYA#H;D+Z`;AE4@X4rQ)xEI^L4jYrBvsw>(rq2`fZV$Ny8B-0dL
zqA4rK8++}+u`BFES*d@rLDZB~O9Px*NTSUX6FY`=J5JT?M9gs$>gbkGIx(E-5;v2`
za99%ge%gV?kdC)$63>(~NGi5c1mp)BMdijts7h_Wpu)(_3(UWfWL{NVLlIDCO~O#^
z@i_8qgD*2yN=le(P_d0tG!~wOEi>fWtXXAZ>dH4zEz1L`pL42qXb0Yf<4>pgr?;1)
zr543t2tulNn*QwL9zRm{VDF8gXB<PNQc0cHQyP0E82$XBl6Le<1e^Q7MT+B$`3RH<
z0x8Q6%MHS(8E{~I3pu>&-pIZ(|MyzxKR#O<r&%-?Gkgbk{%(g91EaAFGO0($7};rD
zeE2XbYJVnSNG}r&1C+PfMRLC2nA23jAHz~ZO-;Bs{`^VKv#^gQGGoC$yQG*3Ev>E$
za*Gg|e2H6Fxn$RGOLd}{#33LD_dFI)bNJn$&^%PN0__o7*kXP`{?K-+vZRqzhi{on
zIxbvN#-~1M*=?{JAwytaI>`#lH~tx9<{rrexbh{^_XJocDksR7hv)^MVlj7w66y~u
z#08-JDcqkJP1f}}sJeyqhDtQ^H4Z!orO3&C697*44m~Y1uuRa((6zO94>P_K^@h1&
z8J`)RK5^0|xa#A<+Hb!4WjKeT!ZU@>ZLl6#^E&Gsz8^qC)|KGI*INol#XE=nqSriX
zBDqvj653|JJcW=FwnUo$`3FUWv|tV^gt$t^=c>c0)1de-DQi{}?GYvo!X4HIDxs!s
zAq_jWmuOgf0si-TO+Nj(4K6rCbR9Elj42WBJOkMoj=nnBBNo$DBE>YA1FbHJWG3$I
z6{GMyIwWnS+~2Y^wu4T#p=PLam%b-?@lB0bA9^gM$4}ITipBiD-qpsQDT}_ak!cEc
z2!E_%kCdUC7*x1oMV^jt)3@L=x4k0<S1~5qU?WCBKiB;*aeCvCRund*NrIuwP>w{*
zAyI^=g{QtoG{&a~1#<G1F*;HIT4psP`B4AYHUD3CudS$@t=Xd5LpW3`btzgi<ru;*
zmi#9KnwHWo{t@$-Q$Z<>!D6|lssV^(V*5M8gns+R(5$6S#xy^JsNoZ(N<hh2Dls&P
z@`VnnQ`@kF*25f{>$<s~_0p1AWgaXB%wJiF4_g^Pj+UxBVJi2q#?{x3dEwY$k9R!H
zgQh;1Bq4^=hjKSjLWmY+G-V5q{a3#3->&#i1>CZfuxPNPi#LrL7AcU8Q_6pTbqHEP
zp&kpV*BfSpUL7B0;DdemMl2U>_U-U-v~wlX<Lr(y#4!KQ4~i%&Ri45xX}b=<p9H_<
zGJUhK-P1-nJ(K+TMHC>v*hD6<hFb5C_dhP^q)lcK3v&HdZ_J^yL=(!xRff=06_pTh
z+#rjBtPW801nl9ILbD)VeZmdF-$$4EvCWG@Bx%NeHv<IQ+{-nVrD_JHhJ~}hI%r-i
z1Ix>a=JDGh=vtcu1-so>RLKZ{Q3=s5d-@eGQb^FmL5gYQglv#gge-w_%DVT1BjJW_
z$q1DR&*0ZVqm@Hd+r-iei6d|OeV<trX%~^TIu$yKcKtd3JIIT74!uRQc0#AyBj&vI
zIr6nmk6efY<-nh=^>Q<{BIy)nSw^n_9l1<$^s>l3>vUR`kzN>gafpl4A+c%SZ^!x%
z7ujh4x@P~=;(4tZeIodF0;y)0%jeS&E~pcbnzlHIm4aKk#5!n8(=bH}VM$R+gBB@a
zA-=shTPg+((F58G`F#W*QNJ{`&`~3rPLUVoAY#xwWF@*S4%JK<NF}U8S|`E(Qz@En
zimUD6kEqW@Kj90Rs6CU3_MxIF;;#_Hv;uYPmCf!bDxWkpv5c6uVHMW9)8D`WpgPw#
zg83T4kSw=i>2AGp^{QE`7PKD0myMWAPe`v#xdWY`<RB@pGrJ8+0sekZG;O>*k3+ik
z6BO!L3TfwOPWy=@H^6l^Is2V0oWi7=MWfxx05)wO&A3&gbZWWW&nO`PAB{N{b$~fV
zBm)|HEIlStuk<m3k0AJ!n;47~PjSvDrlH*`Em?o16o5h?YRanOSJ|~_i={O-U+C4j
zyVGhhc7{9AEWoDQt?27vvn}#aoXJZ0FYh+#eEKWu;;(w>Wx*PWO%=1h72GVmzAndI
z-N2!2dPm9QFyrVBJkdnYqIDQ%+iC7uRK;HVrljE1!1+j@c|O87LD4}I9N+XW5h74Y
z;s}<_q^0Y5#C0EYh#CT5Ihp5mxagRAe%~N^of#%ms+(A_LihL;eRP<=dB?_bmcE?h
z4ga%;fB)Z&9Q4mdmZys}r*G6A%`B|NfSkq^(b@@x?41{(4P5#v2q>7#kd8_~7C{N7
z_^nt<gACvkU$<e(xIlw~kWSi;{Hh%Bm)W{AOk_tIrJQ}yB&ZIm1G4lRAe<3%Qajr+
z6a8<9Le8m^Zzqa70W@O)C1mZ7c6tAE4A9fO;E=t*6)rq|`RVuP-g{ZO`|t?$aM*Y^
zxFK%~1UhS3sKkO<3+7z8tFBMRh&=#B<-+fg04NcuUFsqkggN?BTy|8`sPA48_oUKE
zS97v;MX{gvtDS*&)t!G?UkC6d;8PkKnY}ZdNyx`=F(@3VEnL>56qHplo0<NoWjUn~
zr;3UbQl!Cz@qKq>m#)BwuA#&(_g)A}2tE#x3ScCcPD6J+@t7*6#MH(s+oN$+NK%Wd
z|HTEte%zpB-SA@{>L3OD5_O`v+l$ArLP2*UJ$!zZ4z@Hlr{@;R2X*iLAfqzZXd;){
z2f{~OQ}6eE@th4GK(?*%bwqJ)@}EGHZ*~pI@|Ca<q0DS1QUw_8-ztdY9nP6Z$7Q@M
zWiU7+s~}kkc6g#=%D8nHd!#ZSO+2%!V+whL>H}us7oJh^Dq;;So(LLGh=MZy-`Q;c
zeF|iu1Z1pJ+1!LFL=)-R`CK73czwun9;+huohTb~y~``J*cu_lGOixRnm*Ppcr2S>
zSjN`8Y(zx))TJ^i=jYh0X<ReXB}Pj%syuNwDgmBYVPc2LFX!a8#eQ+5rsfr8V+5o!
zA(=qt`jv>psoiB?lk(85_VD9tO7n~EY&}bl*PHyl>ZM8|oxR5^ah{=_sGOx)SF48j
zMcf``i{eZ(OoS>|O3QwuM&>FC8kT-85qRnDO%t}ir5ThiSoq&4{&u>idcBdudh@rq
z-6Nxac-4#%`Rso9lG)T1g};PzYY+>o!W_fL)e-KjB%-sOLZ~TSzsAb)bu>(nkD75W
zhaC)Zw14sFi8nWkI#LIubs&^@N~oKbn98r=UXrUR6{30wOX;zCct)~dwzk;Udt}_K
ze|4#>ghy)m;~c~m+RR0Xf$4~|_ss>hCVb@okaZ61k!anv?sNwgbgYhT+qP|VY}-yc
zRwe1!>e#kzTOHfJwa>-b-?@Jv&r_*YbImoz9Pdn4bD#R=R-+8fcl)HyAnk%hjcrX=
zRcFqsB(JH#DLjWTLzZ8Onz|S@7VH#l(l|u<Sf9pWhI#lSQ|mh^$(TcASt4)f93^>*
zn&u;qUQ35)UFXQfN8I%!UQcsx8F(`08uqw3HLQebB1Tq#oo@85Z0vuu={lPu7}@Vv
z;sh|)Att8$<Z(hE1>mOJ2SNE3BVMh>%PQ=;2T@rCAdf`rMtFlXi?k#Z$kAISTEfsx
z1#KW;l*`S1q6%XLCnQ;(n!1^X>f0fAfaBXCS+QwY-8E%g9ScNcx0+v_NPV_FP!QV5
zAfqgDuOy;l;_A$4bO|5&?AiKh|Bja2DqGalNa6Nail&Ti*mL6;c^~bP;owfy|0rC3
zT^a!>$YUZE0+h3ceT}a(UfjjfQU~#ivPfW1M|#*MlkT9H?~gR!F7kfy>jdp!5A<Dn
z<3AB*RRh3;CS?lS&gt;m4*Adn+3+G@wVPIsABS&{%SbkoAau@%gA0FzhzLp`hPVgB
zsv;wWmebNEsG!BCrG`q&E2e5Pma6dGn+DBuV%?KTN3*5C!#B@t4XFnfss$~`!5)H}
z901}p$~cFWz13^tig9ki?<-E<_PRbFms?y@{|b%&^vW*plFr9*B)Q|*)Y}0G_Z)}6
zOeD@i0H~Aw=7))C$b=B9ShKbLIDdTU7Egdh1&-)D(ptHO%cll9f_BTrm<aJ);yE=c
z%pHNqKf!Ud-c{amalZb4H>cXxJDA$m1AM_{J`M#q^3V!1_7hieMhxW(udQyYL>)Tw
zI;X%7al{Z2MG-r%^i0?k3D-x4gIxFB9m6FLz7W?t_)v*l1*sj5`oTEl9m+Qfk}8-y
z*ObBr;@>nzp4HN-#_oy&ZU#T5U=e#kZIEbcLIKL?3?}`Lx(lY=Q~J2}CsZhI?REEK
z&Q&(Ejx}Qsk5nz%-lISUIETrarlrF|jLK<Ub4DzLD_(qX^9l!B(|T?zR1}aF=+0uw
zgt=o?0*#+RTw$tb=f^SYbLg<E8(ZT#B@CDn=8BU5Plk%EL{Uw)PD_M?<nh}hO%CMJ
z=6>x*4`e7nS!%9t7>LB6o8H<IPdUJz>H(WX_{}kPV1R%inu9vYS>M1xsn{vb9P>jB
zV#77%RV`LwpHth*%h*E-CoDrtgXqq?z`j6ruway97@4e((ST_;LsC&)-{fl6iR88E
zSl=_WNS0$mW4kg0O-oTi&6FkR%N`8|ByU#p3sZUpI{VN!%g$K)I5jlTaz`>g!k;P+
zE2;3!8>c8%;>s@)P^om)2Y|ziYK{j4$?^YdXl2|sksZ1@zn>&0H=i194fPz1p|Up$
zQhg=WD7SKM2!rz`TVub@C{m1c@aS>3g2HeI&Vd(8gXS@BBbr%2t*Vk_K{4VevF-dy
zM~!eiM%eix%<0)qr6TgYVXdN=DjK*$&_M$+9D?icqbGPItcc?GD&mn@##t_4jl_;=
z*;_#Y3#TC$(daiUY}p$D`-}07a@qz5cN(^IOoM4rK|8nDj!Z(>OxD)1YA=S|nrcSo
zpOq|sr)z});dPLa#PZ0sL122S#WjnfR^ehFxp-f?^nWjy;5|>f)%cx4E=<BpAQXaU
zPAjPRv@>ikKF9_q(TIE>c5=0~H4u)yCK)(_nbK;kO-Nc`wVWW6k_0MA9M@>r-K?Ik
zMU^4z4{Q1au>x4fJn|Lu&&Z8U<nkBy16)QH-gQAje?0~+tV47&^|JgRg_+_cfxWqQ
z4KJ;vYyf5G(;#^SC_K66V}`#Q5CV~L+algj8cepuzdidido3T^nnkW|_1fd(XMalw
zYZU8#QV~}<XM&Pp`2fKWQ_7AGKE;`z7Cwpf2`Us(u%#-2%$(ktl#kkkfB$7gCf;&$
zl_tUj=8JQ@cVr8&2o*T9V1t!m)8>U>5<7}PL6uk15&Hz(fobQG8!@0sLAnF}7z;6C
z8P>vStE#AmACmN{pG8r7CR&@?M`hqHaRs2(*`m|`nRtk+><!^NZ;&;lFTlEU&<w=@
z7l}qU(CFh-fRCapiJ)BJ$k($2?XYcWTg>b91!9;4t^Lr@RX9c_CecbU-eWx18t-_t
zWypw%XA*eEGf~9x#N^8OZ6j0Os9^*m{n<z+A3PnSW=3DD_$zGg2)HMD&31j;3E8%*
zbULYdB;Nf(Nqg@(e0&dIF@Oc}gWVG<LtA4AmLv;mIl`G`@Hj%jT|_9Z5IU&GpGkJc
zM6_}&H8$t6CY_@(5tR~6Yuu48qZ<Vo2~xxX7h$@xKLm_?iCy#NE7?0A*~G8DxDeBH
zHHB|bxk0@u`zE6l(we`pY1mdXBGyExhn^FxZVh**RmIf)q6#UdMMy51QR-|9lgHTG
zFbR(MPe1-J;cR6xhlkJvEB-zS!?TmPnh->jpKQ7h+N8;+IsFq^#2ntg2j3k2lI4GV
zSK9N8R_B}UXB*}1uje9_yZzT}6-iiPRyUhp3Q-lyLIiW#zmWT=ASzoq|2nmK)AAw{
z;N-}`dB~-xHVG-C%7DQ<_IKDpG;CD<TJ75SEkS8l6v>?r5K@6NDHWI`nnys!dt@*N
zH-%8@f+6662_^I_qBz{U%9NQ;-cs)(z;`ioT)z^H;XxMCV;<-Lp9%}km*irdwJIPj
zonRv35LL;d%39{GDiz=apzP)Z<;>v3WyFYeOaOIY&lOdOp74e925vLj^gK|7!OWNL
z*susZ!RkpkE9SMWO8K*B`Fi-IXceZgwBViihQ<Pz3mO(K8n*SNOIjS@P<?oNn_#G%
zs)~6oA-7)#3#UqcQ`&)AoF!0~qlCO*xgAD;T2!w=ZC*)NgSfcNf0UmHY{jpg))bJ4
zg$Yx_J7?JLpJf_al6SItWSdirZJsgV*4m#d-Tvz?6iz%Wu&cP55w;i`zq>re%w-^<
zrdANfi2&?x(L4`#fR}eaEB0|MMzOXqH}Oe0R}TmFMH&V&<gMrt0QUH+-`>Oe|I=$V
zTdJBgAPGK)i19i4W}V$7*1UXhfGa9|*xy2C&W5Ncl4R=Kx}iB6{Ch()J)*YOeePF%
zlknnYKwODw<S2%_()6!ye`>#A=@+|CV2eieW*-j&nCH60uyK;j9l(&}CCO1NeI;<%
zH8ivc6(TqQ4q^m*{uLj;H#+!WPo?YAzqbxe-P}k|F|KUx#X=IEv&j@21vX3OuN^#m
z@bxM)cSnnjXkyo)R^j)-lxH5(CnXVP7agfd_F;Ld+t#YPp_(vyy$ikdL@uveXo!<S
zoQIgOZ9TKWSK0~|kJ+7bpvwxzsj{4W|7BQ9v*+#>PZ49L#qVlr{u?dL9l6f0TDodW
zTtaf%lK7h-32DVLId!oBio6sLrNgAb)^JNKGIFlj(4`92Da}S9gbJ5|9gE38tY%C)
z&Sj-SV#EMJH&%c?Y%>qu4EC9V_G*r;<DwkS-t4o_$N40>-zfh`IQ*FzF_&WE!oyej
zOMHTTn!^?lFq0^=FA;f+CU%wU>v4Q~gd$@R11);F$>$4QP^c)z4|u_jl|fGNXI9E5
zJG|Y~2hEE$3Z;8l!+)7iAV(8qK5?)#B+V9gHV<AS@D7UAg<!)scQg#u44AZ@;@&tQ
z5W+FR2uOAtM1P|c<jT{^6u>oDUn9lFLo{+Q$jt1(<D*o~jZ*{3&u1C>hVdE(f)WMb
z)Nn#C3W0O*dpK0TN|w-Csi037D;|gjoM9xQ`0E%}Iz2+#QocpO0Hk{QF1^U@Ub8ND
zBR~HhKA()qJK(#B%_SIJJ|TAOvxPUz??L)12up@^@0NcdwMWNE808EuK@^LJgB_3P
zOCiiG^lJ97h|HV!7nlq^-|2h9wfhn#j1v=IWtZa1`|-0ILyn*#ObKk_0=ewc8%P}e
z+ZpI!RW>9HOO3N(&5#Kn-;3D4BV&#H82ZdCHCxJVL6MCO)rJH2Lc<jqebUafJah~s
zioGuHR_E)@K3|7Jrlk>*0)p>x2g4wKjVh(6AiW}1!;WYlw_vftv7w*MmiZ#N16%LZ
zUVDq2F8w_NV_QA&(s#<|Z6>xoQU7znZ>(6~J^b+(7WWDe3jal~GU^diLzumM4q){Z
z#>uWmklC%lK;eiZ1Oyy1O2J0N5GxUGJUHkz`z9K532J{{rBN1a>WOoNb#GHMw}}hn
z0H>U_;tR195SvUr+yol?n$95Z@PGfmQHXyO{k^vs`4eztRF0{=VO=xL-T>KBdNxh{
zOC_tJX76BBq(M#wEQe*u{kUzGMNEZ=zq3Sp@4TZs0Lz%nbH2{InJ{jagI9^IA<8dd
zvv^{dQecyksPH@p#WaIz97flD0NvyqS_j0P-iSP!nyevncYi0&A$i+`4HtF`esa{9
zwhJ$y(W&@TnBP6W>07VKh3{ini}TW}Lt!qirjXh79cq$1p>*{M;w+e!M6f+azNFKb
zGfZx=LO>W&jCXb4Lqi$B44R}LB=tS9tkR;RW0Wpg?rTbE?iW~$l8j+S4Vu5V(yyOA
zegwf;;UYt1b$jBalP_|_e6uol!{({qES$MorYs6(Dym1HmLR+i?v@IvLrYo&6mrzI
zci(sL4)!6{nzq#Y`QHdR@=`0nd*jms7+_H0^{J(3Qoo#f3*k&Y)@0?@qcx6~hB11i
z5>t;EVj7J`IHW3j@OdTrIBGju-@=`YQ<A-+#rucJZB2~uKcc+3eJ;pfk74K^>@@lO
zs;IlnqoGOn6~zUVp=z4`a*miBV)&6T>Mv^0qe7BIID!_RAQBLU9%H?3=Kanz(HvWB
zDWdod?L}k^wqCnTq=8G7k+dn9h<<n$TJe)dGdW*+pc!fr#&r6#=a=@xe|D$sCCW;H
z%>D&<>=r5z6Nzp+Hnzyaf!_8g>Id_?fZTZlg92`A40L8}^zLIAE7p706bn0qOi)o{
z6BVsf{Z=U8uK-hCFiXj$u97NlDGWoafcg}l`&aPEld$Nfx>4i+OZE1OC$@Z?GBQff
zF59$Bu41FWYs;to?T#Am1qwh|LTh9Hw&UjbF(_*L5Z3;3#!A=TEU#gQ!v3xE@pVh`
z*hHt*BXQtG6Mcy{rrZnXwBzf5N39H_3~FYhxC$&~4GFvcptR%W$8=x&X^q=$)+5hg
zYMq&;rgr{{B@#HFNxbqtr)WGg&UcT87Bi2x>#3VUdj64U>?X=lWU~$YWuP6+lN8~4
zF1{gVoh{X3Ou3@AEfUp){d}J)C+t1u)~zWK5U24M@l{fon31!f1=Zr3uI(ed#4II1
zS~bZzjke#QBJhqUBgt4*C{40fg{dA4B_E(0j$wrk%%L{JHDA1)P8BdkQ|u$Z;f0qP
zsvmCVTrd1VKU3bo;d)`q$KTR_zT@8nHlY^gh3_6mZ>zg)Y1e%|LR*=cos);v5IXhB
z?~2f(4|=9&7yhM0kM%6DLZ0b@Ng}59NzmC@;NiNDE3J;Q)fF;<Mp*w8#Pc2U@_23A
z>?=(i($vz))2u>y!AoK^78O)Xj!K}UZ-1lFf7(5CzX0XrJZ&P|aq^0l$yV^ko%#P6
z!uvm>URuW1bKb>twK~=nRanY_3|SjZKVH&st_;FE7+UY6X)DWWUi5%w5HmBYr3D-p
zJz{69e&n=;5b*<t1etRhm=eVmL{cfkNbrnvNSW8z`moWTzgTU3Krw(G+fTvoEp{%w
zmTg38$t9Y7GE5l_5~Tt#gMHs3hzb|9a?zVO94>rUVa1Otn;08Kz-93~FTGl>%0Bgz
zjUg;z`{e-Ky4GI=iG`8^sh3>cP5H(1G`6g-%qBIhk1=J?5`|`x!Adczt9;ISvHVWn
zFAIJCf`Q&24W`jzZNfTshy$!VQ^Wu(mbE&g6<N5)jZcElne3Okv-9)Y=PAD|ziTJ{
zp~GzVK*<Ed&o8Flhd2S3J5)hTh(E-2RL%;OCq)8BF^yHHfeA+nPNP(jbIG7Rf{Yv$
zGm8Wbq*IP6hzeB@#uid*2$*ez)X}2eJ3HQjdD0OUql?Dj>QK0JDTJy{@gE$ycvAgx
zLg|!t`Vt%q6q+i@A+AA=89$>3_{X0=+_USks?oCxP*~mX0=$msDry=xyA6<f_cHDL
z5aiz<=&P+EURCj>r8?$BJijVdp{PNm*x_AycL@<_XhuHg)}fgkNWh3@QLo;|)nGWb
z++3q4KkY;D^{j?7&!4jHPK&dqKRL*tPFVjU0-w<O?w@ZP@+zOT8^(V<w1N%qu{%=y
zG~oB{#Pj%6wbw{aBBzx41yw<|%jmz02j`(3?>|Bz6f-VHKI`Pf2}HqRJ}^uaS<Gd8
zc!x@s3x{|%xOt-of=`Njv_8WI8bYHmE<iq>3?#%*I%4JiT;fXmINRJf3UtJBGJ0gu
z$HgOF<LLd%y;QB{V1G?jM$T5sPISj{0HWALFwcFv&~KF~!zJ!bC8}3~AqAIY5|s=o
z=JBTQhBZ=)r6qcYF_Q9z8gtu_ZLXHsmy*lm8#$lh`%J<AOssEnakahNSvvK8OT^zF
zyw|N~0}XCMkcsWg2DdB0Sq9>&UyU!&uJ=lcO_G0}701c*$W<&F``9EdEXb;<Ev~Gu
z7qm@_dPLmc%eH2U0&fHYF}h#=BKaPL`Tdsn-ZkR6Zkf8C`y89PdZzb1zT(60R9<35
zzbYu8U|qBTW`L5$q^02VKH1|MGNs38!YWLKYv)3UyB4vvzqjEBu%eZo47w>KkAlIB
z6swP<>7<XIka8$w2kVQUPI?%#B$KDDsBt(OB5ES`Ph>&$Y1SMZbuMGecLyz|_&eWn
z%Rs#Aw;$-Ji-;Z|%VF>tmDKR-F_KrET!+^rjiG3GBc#3u4yQ+&r&si}WlR=N52)JU
zzu)2;)nH%WMB@o|42tt#vpv{-r=adDqt`RDMw8?#oR1($)*(8y+h`8A%_71ee5up-
zpW+Y4XVuivHZuXiz>-8`Y-wueEs>>1N)Jqu(`n}oFp?4?{#lW6w1>ZkvUP93_qm_=
z_TKgG=KkW8ZCYGhJqy*zoXGocd4%ZDt-hpg5wCLrv)BRVBoY@nHyqvSm$yGg=x+)V
z=NLl{L#Bx<bqojen;lFrbE7X|<Pcf-O2(YNh`TEh_0=kc1k69!FxHoK_gikj$MAiD
zndeCQPvc`zbvR}_qlJ71*6vTdW*X1iUMnr`u#EN@PKOq*W}$Ng_LGW9XjK?gQU?Jy
zJ>B-x!mL{024PxO;n7AzlCE?!cLr_@yN4ggRafz`{6ZBM@du$s#p)jWnU733G9DiT
zWp~I!#ykfRR2+8S^TWk`UB>rA%u1wW{ddkUJuvv6{XVTST2A>+q9L=C0AL>R*PqY$
zpW2Y`gpcb^_?KRnVJ)v0PT+!H{NwPV)ly=8FCTA)^J8!7_+ROtj<cV&&o0jWhSF@K
zT-*~<Z6fA4^lK5DIJ&p#7ZHzxDPR~KMAo#(`Uuzdf?oz<7JGj(h;R-^4etC(B7hMR
z)086y9ATg(O%B5j1xQYZZx;r^%eS&FZaT&EeEc@+^v)uownw+syW9y2^up(zoEf?(
zb~DKydcE?ktX+(*4)bt@m&*K7SC+fN{Akb7=*7F340=>Z)XkXM?2)``FPT3j&rlrb
zf36?8KZ;)zUD2u<a(H+rxplqTCVT)>^jimf|E$P`-l<5*9|9A<WU2u7#+ZGGi!wZ>
zhMuhaK89LO`=Qeuf)6JuGiq!1yWU<ufBX9NzMIM4#JlnIf+PVHh>sl<jQRXump-4b
z9&5x1w^M!}`nPVM#o4B(i||{N8!!K%aKG`JE1H$}q`?$QtI$~l6OQDFE2Fzh#l1NG
z{gy)N&T%Fe`_m&{O2S%|&>qu@(lHvJ;@4Lmn$DVSahZl5Q_2DRps5SHIJ{Ek)wV(V
z>M6I+Zeo#|pDZ*hmj~rp>&8BXqI|L|Nme?3JG#MMMw6jU7K>Kq>JzJfD|oF3AgOtR
zRZ55q9v6x&Xe$NA2m*gkz9_z^cM46&STv$8Rq5{Ozej?_cJ9HMU2ye0-)if)pO5Ld
zo4{(G--gI{bx%$PW86k$jv_9@#Nx=q;KV$<K>8i<ztg+#_qgTeeJK#&2y;QV!o{H1
zF=pZ1ewk2zIy>}Ba(d+HfAqVNxBq;P-3`P08h%5ZoRa(`@4HKHf3fNGwgrMBCPZS;
zcX{M*r;8MR4}dCMFjwAHMeB5+c_B>v_?0#dF2w{#!W30n3ti)MRKa^<MO`4Ifcv+-
zHrb%k0urXz|BJLN_ZyRO6-8<R^niQNR8;Ho{jV`EZIX8|{LfA=B)+IYw6>1mdz7+r
zx`y9!%q+<jZ#eoF(eCood*=Carqnd>dsG63nfVFGxg>c~EjKrH^>|iu;aU_^W(FqF
zk?L8+6~+B7Lk|<$Z>wEc4mY_jY;w3e1M05sZ|eHnivzrW+J{oQo_O{)+1m1<xm9Cr
zS9_T8s4~#fmXg_@!}&k6Z<fD35)a7C+(Zj$n*Gpg>%RCozB#evz8e|qbW!Q`)N}bv
zVaM;3HvO1dN-2?yCF^@L<O_5<yg2y0&VEOw=Y6DscPoi|J|FuZgW+bZE^0yGr&Pg|
z0Gkj=<g1(_9Ah*ORiY?S;_Ej=`Z!ds0b^1@Vv=L>-?ja>KEd>qkvLnA@dh&Gmca&M
zxu`Z6S2yOVL0U5FX-cXI3k}3LcQ57m^v$oib3BMPaTrJ#pr1<oJ=$vYZ!_{Aj@}s|
zJLvbJVfN@YEQ1;X3dw2n&|k?@u_m5c4v{XO0E-$5ffu7x?6nbtM`V|tcZx8FfIhCO
z_8Te;x~=XXvB%~S<S@HC9-tRsz*vY0oQ7M;Uz#3IWSsWLKKx$TuUUQ<ny#)*%d0IT
z=Nq(x`A8jlvE*HMBtL)U#ffp7+WDm%cD}$}dJlO$C-UFrXYe>+D8ZD2lY--WP8q%J
z`CYj^k7U1CwO{Ryg+<Tpe&5Ju=Ps{Aq!L^EY5;>(*}MNkswWU!U6ZU;$Xpl=rk5ZO
zU1BgOn2Oe%#mMymW%Lip4KI&A)3_5>KrF@}jVEi~&y*S!nzcByIlBC1^77*OC}Ss|
zOzJkH)pSa;OWui~2G+swA=s)jM`xo{)4nT1N7%&5TFy<ghU5I;LKZ9|*6m^HK7c3z
zs!_#?*N1AJn=j&Tw6{t+`ddTIgzt6^7XeV1+(CCnjdXfOr>pRj6cH9N1q2d$Mu)3&
za9)%_tpd6BjooZKIvRbd5aN7KiekRq$|;rAO?49!+S=CV7YU@$`4-u?^u9=^JV5-N
zCBv*AI9IdM^w&py2)AZkpNSdVj{DL4$XTe~cN=!a@A4l=pU?X5)I6WPiNKnI<Ru1C
z$>jEDb=Lo3=CDh3%jAkChbl<ayEGpItC19rP%~wqgL})wmwAduQF_&a%F6-mnEX+t
z5rH)F*GXZAls7z?;ZM(Ce55Y=ak3{Qu*DujMvt=^It1JlD=TZHBh1Mua<>bI`hT-t
zFAwA0vUbllu-o6(Nny%74xuLPKT>d;pujR8fTq0Nn32q3RU2BtmjQS7g9>nv!!!dN
z(PpSSH)QYOxgPgr^=l}99=KfmLJ;_>0~``mg3s`9u5K`VGt1cZ_8P{STx`|is0AVL
zObSa!S43{LvtMbZy6#lP^6`?gJhnC2nUN-MGo*x7RXNzjHbd3g277mZ>Di@V=(&gD
zgAP)fy!RVa$%54~K><zCy5H9OTjuAz*;9DeU1Iy&XxV4a$mb*U>6x`XEP}lD$!xJb
zR(eWsJkbKFXo6B=L<zb<?Jt&ekVbxlMmnLDDTY(0j(S&{(W*4>Bgvl~?hM24-#rWC
z11Xs)`3Q*ZyydtLp*!~NdDx6=ANMdnAffN@j&I859tgwZ7+I5dd+Md;K5L7B!sni_
z>au8Iy4Jv#=U4@ljE`@M<6tF9ZwX6;Glv>43-3Rd7wQ9A$E-*^Cbw71STM*~xOMZ)
zQ`;`}dlBPW{jFZT);<4H7A<9&d5JSGe$9t9oD!~29*HD0Y6@aTf84jLbvlq7Ef-KH
z9sZKhpN6a(!+`di2?9>2D0o6NVo@>c=?ss)>xIeh3?+L%jPFelbVnj0TYb0VcSGQR
zZ~B1ee@^-MO0VnfU}|I3$-DOSpLbvkHtCmgC-4Oz!J6<9@)n28j0jag84J~wDN%@$
z!e59UB@|LpH3<d0URe9WTpFhW;8W+6F$c*I;;%pnjRdB@hMEjdH*?k^<Q#P>4(&Cg
z^gbuZ82Y$G`N8CPpU9xW-l>|s@1y0$X4Gaj4&hn#aC7i;`;fT^Vpipl@<OV{CDVN{
z;U#dj6xE50CP_QtzwHe|SOQ8ilyp9D`t*7vr0Z2tsf_)uNwh<9v8){^r+yOzq?*Ro
zC0ppd$C$aiztGxlv~$i`0GUrym#m1Q9Wcpj7Oz_c43f$2yeKz9uq@K9<v&-p4n~ux
z-<GpqLBPW$Kt@gm8IL>Y3{8Aq`|VDB4&Wayo>$n}pn4~|>n2KQOP5PqE`VE5S&9+-
z`NB#4hXCLchfMx6?T|&whKQ6=DDWmeE4K7(e0og+E6dP|#nfMHB#<p89==7*m_)v_
zSIj0T$#$T0m8k1}z{XYw$53TZIUD<`6v`84%{pP9)QcO!YJ;w{{_Y0SYfg97H**#<
z0~%2&K^^+uos1@&?H5&knhfO<3n5;^1+pk7DUEWGde`_*t^tj37nuAU^UWcKvJx@%
z71^p<T-v+`EYwOdp;$rB@9n7{R6<PczCkw+;DoqL<#Bo6wE|F<sRYMJZ)ke3B)5ek
zer6Zc2-`ffw2nWw_}$-TbUCDFzoGDV#~0EeGYnoeI=Nkcex&<FcD)$+1;8Ha_xh$M
z9sMhDx3pW){m)1i@G4HiKAGT@m`s}^Gf@?B%!G&Z`Wt=)_2?!*-Le2C-q*hhXOgR)
z5paM6j-+Z(IvV(A7`ivkI$;Km*PFbwydP1ZkQ0;C*{EkVVJaV?-05I3P6zuB%l1o%
zt{mP<&W~puEFB~2%(w_Xk7MNnG*wjvbk54!WdYFI*Xo%8tVmiXL&A~-Mk#f3G^7F6
zdUTnDgb*<aS|x=f4y4(o?{xQ%Eu5H)86<Z!636zYuI4ZI*Ms7dxfA1%2UjR{dkcL9
zezG;DMI1KYe~;I+WyZ-+|53WySJwXr+4>~^u6e=d1+PyLMdxw0arw9Fx$1-ZW7Y3t
zNuus;ZI=9SO!ySf%p6rY&)R<uB4~8h)<r={0YSK)?#Dye(P)ke$L;Cu*YX<UK?2_F
zB7d4Dnb8Cb3yWv4_ws@R_FtLCQ<CLZ0vK?QC|HE`+<XY@b-ni$ND#?XkISvKQ-24!
z?uW^7^$uORa!GQh`8+fe)bKn~Cs2DKmZ^xX>3(M9Q-R2M`*Xn1I*y#Vk_m9~$P78u
zy;r>7;N)0QvO#0{yrCU75i@aFk<hyM$A&OMQHYq(deHaY$%>gtce3AgLP|J;$Kpf*
zi`xZZ;V9)!8u=s9LPwZLWKZ1cnmk!Q8}?SrviGVEL-9XBTE@=akb;zdo%i-k*mKth
z`zPz?hW-b1f^l#C>~Hq}t|*{^J*#WA-M7g~{qGvYITVQb-+<E;-LMcs@)QX$k)qIo
ze3&SUhFE3P07@kD5-Dd*!}~M7cE=nCjS9(I9C`40%P?-iV<YD5LS<Uls>c>CjqZ8`
z_pNU(7n`WyZd+rA{7&I&?E-4i*FNF*AiBFJ?g4a^q`fJbJ14Gu^<r)?0X(r*rb)pJ
za0GJ-%UIQrT9%SV$^(=9hYSoG2F9_3xTV#)oRHr7M&Kfh^kglwls$Sdss>G08|kx~
z-+t$gco(<LMf7KM$Z^5$>O$tqb@QFXhe2=b+g0{Kd>B520TQI|5!5Gm_Q$;44Q$t)
z-OLu`4z@a%Go`&pct+H95@9z|2+T1|(Ht^Eun@x$g2lpVd1ciUxLC9x5KU7rr)&kt
z0s)g#8|uqOMwEjs0-$8<9WqV$JHs{d0L*kiS;D&gcs;?<MWe~;NT1?!h*n$gJsz=M
zI=giA;ufN+W*g8F*GjsZV_dW>z?i)uUR;Hhu%aj;wp2qr`Fcnr#=*ni%^B;UR=uKQ
zD9RCf0r2wZKkp$esX}Ut8rj5T3TFjFdC`|%aVEp#@_a`bxUxd9zj@&f%Zh`1%1rOl
zEZD{%Z0G>D%7&m(FPQp#>%w>&z<<Kq^n3gf9OwSkcR$t_U;k}9dwZ_yihcyj?!)+B
z8erFj=6hxA2ZNI@$g@iXo3{kcF`7F2p8?A40ai6y`3-~0F!XT0qIxWa)zI8Wg`vaQ
z{!VwV>{c7A_Bb;Qp1|n8{BF-_YnMBQ%cbI2v{B%m6&%4;OV;MG|8$r=rZdBDPPvF@
z-Mo$wR|x1`&QRCmo9cR9s1j-vTvC2g4OfXX`LU-MSd%QLVr1Zf+qEE;!Z~0pf)lpP
zGo%KgmBc|P5DX`da>G-+sB6@Ad0VP(w}-DiAep_!-b8tyi`3vzr}(O+aSF=iFHkp8
zN=wJ9za{#;bnW=w@&Da{!tddf?$A3d_B|<nn%~Pl$l-UN(+|SF&)p=*jA7^l+cfk?
z`dI;HR1!s|6u|&<Y#Ch%xA%vn9<LJ!T+P`S$5%2sK1~P$fpkf8uVz#SuOGNjtN<oG
zmTiJQ`F%DuQ`ZY@vFmMejOW)6kC_I0!%X<zueu@u;wIF3?&mPspW_e-#^KBAPpv{y
zLj+Tp1>Z7Gl^>@Z|I`W2TjWJX9UG4d;Rq%h$w-oVXOJW2^|8<nmt_q6#li^Xqu1>S
zU)6v>9jQnuP2ie42H14Ta<b>Ru^wd*9x^5gV4(TE>i=!Ve<s1_a6>=LK?-^6*Z&yf
zdDiZN589vb!~It`DjR)?`i}~;=JU+iD;LR>u0+p7(x6#W_-#u0Cr+o6M4>_hp!RFL
zmGK$j*KqY%i!zfN5>ol1r_?ZVk|c)$(HKhFj<O~Ww@=>wO}b3OSc&yMce_~Y);{7(
zzLAzNygo^3O;u2sD_&i%ZFJ&v4TRP`s>B(R6$)I7X>AQ~s_xn%9)k(&HgoYv@~jKG
z@);!}K7O=dv0Ns~{=UmpJc}muVjMmIgMilHe*GYOrWEub6;|baLTPOxNb$tB#KqCA
zMd3I!k<P86Dyo_E*5`Nk_#_Vkt~*}xJ{nc*?1bU}jrIO~yp{i0;rE-XGNtQ!;4y2H
zt@W15AHKR~towr|R0v2CmjwWof}^pRMhL2|EMZDX4Dtsl?V@{(rTn;Z)y7-r9Ibw&
z;@C)Kt$8Iq;&?&XG{`F>Xv7t_%yMr#whparb&jlE_+jSav@GhhxFo96=^WnN`!sqg
zi<l+09uhvXvU)(DTk2rr#U()pEbTj+MJJhJAH42+j5aTsmFj^$X6KAbEl~p`VH$7%
z{TGC#VkEYg`p+7{_WIt2tq)e6-Rj)fA3r*E1FPY~8VPL+2;dzq(8?*RTes~#&*%|9
zml3XG>WUq@^B=SJg!!+6)}QqKAPvsb|2VPyOW0q3l5?QD+>ZpccI=W>Iz+!t`a8q=
zgt-EWzXC7!fZd)j<o#y=&IkY&D=1sbV!(x?hz^mZRF6d)m?LIFGD(pj8PZq7pCr>j
zV2u^CjJs~%+hKNT;9S>};(N=Z-PE|Uv3Xe6br1LX?)Q58cGJ~95@Kdpw1%Cb(Hhp2
z=^JP38b4qnoluN}1E&B`OW^`xnx|UZ5erOWFdu&H{box4S)W7WB%_G*f^J3{fMM{*
zKP$d`(<h^f=l*6KMwAlx`cQGyg$^Iq$04n%;o9h>)9M~RwnTT0@G-MxaymGXS_Hg%
zngMM?FlWQ+Dns7Zj*XP}4d^+D?Dv&fQs&^{AqN0sFFiC!r?dK@15gtPVoKRhHBK?%
zhR)z1Nj&7vJaFt9k3$)gAj(N-nS!0*61j}1Em*%7+hlul9IPKF%Gh~%pD3raxLd_q
zaYMZgn(1v<eRlb#`rTI9U60l!7`OEuALR2mpTITjLQ+@!QcSJoZV4B0DsYK<MNcVZ
zP#LVwmY;zV#&U?Jk)D=INQ=m$xc2^5BzDAFO-6F<b3!g}|2k4!vjVOK>wXNlwn;kz
zV<CmIo^ks|HiDTsDC6e#!zSD8a?!;1Z}z7Zo!zBD#(;VR{zp9j#LFty#8C{xhtI#0
zqbt{$J8v;^Cx58fIX=_7##3W`=#%)Vqfk{Sw!T3SsQ@L#ga<UtB*jt0+CZ%ao{1Vk
zxgdq8Mr?Q*EvU*NC<}2bXFgELcCp-v8=>c;q37MM`?d^R$G#y!KH?p`jiqopJRpk_
zK3C{>iYR3CC=(94XK>Z^m+px-KS_6u<oQI%6;DQ=a$_pj)Gr(R7$tV!9>qcis|jQw
z9ep*RnL7@SOLZ@rA#Kpu_6D@a=_S{>8zzpa+f}b&Pb=b~SV?w9Ps3ibW@0&*(;H>>
zVS;uMe%V4$p!m3+`Vjs&k^eArmmLT~;2Z7$o&AyGEMH}PX%DYRjIJln3(u`KiJ;&0
z_47J{+a4UyL~$gVg0h1WR>#qU>#J*fFC`#D4zW?FG|p2aU)hSWkre6}EKAw^dzR#H
z8MKZWF!3N_>!+n=@6N$g1i#-q8|gayenGHiwmbFvO@8lOz;ub@os9kcR8_W^%In!G
z-+)T50J;30fp^*{Sq)p9ozBAVX<0)_*TKsifs#TRSiy~F(Syz*1dM65RDiDtxByEg
zL;k5>>@iFr)yCO7sJcNe)$A?^TJ6earCZz~`VWm_Rv3P8-PbMSh&!U$Ald?$6`>oq
zk4JqF+*z7mj623-|HCPx^OF!BZ|iVEg);NMM!7lvDe+yD*O`CcV$fY?u^KyXcns<W
zQbi{c^?kGMiE+_WQW)BNW@mzv>3<w27*xqK$zB@^;Dq*9Aezw*V%3ru&0}x#h!!I$
zqH$7BKytT*hF6xTQp%~!=V#<U-|BqYjFh^0Blf{y=u@=Zoq(w-6|VvNqy5D<ebR~&
zbzb@LiXE;vH(lK#3yu;^jI=7Su(8wUPi+1WnWdRHXNr(f3aT<Ja}$)gS)4|Urlpc8
zg3~EI=?y3C#I?5!-!D}vCBg(57}CD1u|pyXwO?OGPcbT&S5ggKJb*cz-QI!TP{n?)
zP0#&9{E|A6&li^;wppGpsqE@G^FJOysN(9yf{n(v&#afPL&?dUjny(CCLDM*RJ^uF
z38(pZLDo_Q-dw&iO^bsZvKF{OJSJ8=-BhCNX9`Eqnbm#g4vFof0PFBvL&9+mUES@Q
zy}<m~cG&Y|kMn~J?35Q34}rGaD@!m0UdO0z%E`|`<@uMXs{kWf_&?uj9n;0?M_LL|
zXco9YLV;eEW4~m-yOgfF&emG4J}jr2noK4cy<?!jPa(-|2yfPv;y)EkzQ-Osc3|Y>
zK+#yHid)aryYwGIfK^y&K<U-{&%cG{aC%48t=*`y2_3lY4EtsDzDfEXZ;@|TkG{Xf
zP;WY8eG%k+c9<2s^Z%TER{wjLU(UOjv#ahx-8*95@z_be?z$(@x{}u6c@~7?bB#O@
z8}=E(kZ=j8!WoN<#Nl@Ns{aC_xz+*yP{jkl6%I)AXX0j@qi%Q~$kK`*NRpi0(Su9B
zWG}cP%)IiX@cRh(T-(O*IP#mxXAq1y0{xe=I!5+e(q-T~M}=Nk+_FpKz9t+8iwW1%
zv`Q7XNT@=CII)UA2`JI5TEAz=rfk0uW^p-1H8fGx1X763>i!y{GOAg%#K^-cg?%hw
ztP{h)SC~#s{@GJ*Ua;QxSG$)2P@3~83o4l%M)7_p)-h&9VEGtniP(21#V<=7&jW}u
zB!3(5*pc0Z^frYm>OpStUbLelkkl7l#WOV`F7Nm_i#27tu`Nxno2Lxx?z{)!<t9yz
z@xn;fu+GXO^NyOin*-L<D1Jp{=I=k9Lt?*0%8FQ~R;t_t)cA&o<jDjOZ_C*DWBG-s
zXJHbxk#k&k|B%Du?`5IB3fX&u)jn%?huz?pXZH(%V*9?@<I6<~kDYODYoFKIU4R^Q
z!zih`OZcN0x~Wq4pf`m6`f5g&OteP=BPx^<Ss0U}R*JcjmZA9trwIHH040T_k%*ye
zo6I|ZqxKHs52O%B!F_mbT!B=HmhB=$iqGDm_AP<!>%11Dug<qU`=p)~80C;0)J060
z9SN^jcfo%8kI|~Di?GYR@fWO1k2Bpj=YmZYgscC4;eEJVqwl`K+w_7bx%`qGtQcc|
zu5{ZH%~vG%dJ$d3FkD>Zk|?zoBTnj-6buk_|5~NlN6O5NWdk=D_F|acye6Wa=DxDG
zrQ0bt6r^;hU}228oncg$K6{mQIM_o93gn<EY4=IuTR@yH_%Cn#?C6+UJ?4TJS0Vw;
zrytSQF6;K1%O>3amMK<Q?K_G1`1p;H?G@dW%_rH`A(CcVpeo1tgSaeEi)48YjT`g9
z_DsoxkLXXLd}vtcSeKhtA!2=cv}W&ULf)V0hev)kbtw=;#+(J~DJ#}K@DCMbL_+y6
zWf`&+?Ib5zV&HoRUT;qo$YI!{|5(=FJv*MGp{1{JIeo}1xK9s5!#qXQ9IU7~^_~)b
zoW4qbv%IFkhK=N@o)wjX5Zj{;K?-EVI)>eK#pMo=vQo04DEL{rQZ#3tuogiqcfD{(
zk6igg(%7vTc?p~0!ZSUtDKmAebqUFx?HN8hU_RByJk4e#R5x+`;%{{7LPgU-Rd}&f
zQpP%2pE686;k#+ZFHMmtuBW=bj(d9tJ9YK3hfuqyW#GrNlOr<{N?gJuq+e~qBAhEp
zuQ#xmk~7nDRzNaKIP4dOJcCUL?((Cy))5>D{D^4i0Xf1!5)Jq3vzP3{Ih%Cx6ykiF
z1nXE!sgz7=?id%ByhE6->438k+ol=*Q{mPN3;%twUodmA+^m+inZ5n7tn|Mekn92N
z=zoYz$4jwbnY#}cakcXz0Xyra-;K!E+x2SeI|p}1sa7$03a3HaJZfCr1>HI92~(q1
zI922t7vm1m4wHEXiG}r4NREF+53otXB)PC?p&h0f-57QgjEqpnp<_kHcoGY=qOYo%
ztIt?BRwqLHc8pB^9DQ(&eHRqvYCwd-8DWYr&W=9d(td5iw7slT+us+Bm>5JA4Y=gA
zMYccx3qPkLTqbSt9i@!X(AE1TIy3+=uyBC%qi5^&^$X8O>d<ra*2{M~ozCEAvh!IX
zfTcrb$OVz~*jT?}6gkf<i%FWy3iq<;ejv{jD1VqaWge##Da?0&tbh2*{~@lSrF}RY
zKf&%&+Z2|V-e*+YB(9W|dyb8#VXnABDZkk{xU%3fNoS{U5+UFLbZjiFSwx1I89y21
zbR{#&+G_DJ_q37tlaVy>izTFhYGrJ&-6(hB0u}jG<4+0?y8Le-u`LEfxB~@PCZw|=
z{(+WpV*zs7f@fldh{9vK4yJs|;;92e<uFQ6xg_4CRDyN?>MUf{cly~i{t>YR#z~Z6
z<v*4YNrALG%<CP3ssMA>8m+{8cuY4|+NQC~P<;SsuG93kU-KiZ>zQ2u!1q~L{5-pt
zZF-YLpE&axI<;`?v#S-fILNrW$falmjn*%*lN$;A^E*M-O;9LF#$FW?1jm!S?gdWy
ze$>5%U%mszHsDWBhgv>n)7e-4AGhm&#~d0<z9A-*Cdmw?1qt?AavR<rIYiKiN+IY1
zfrVn=5iIoGoWR3eo{%V_0z?oK<OjERDaBZzb6Ol3#ACm`3C=7ww$P=2$N-(gx~PHz
zD#W7DHf%CC5`4Hl)(+01#Fg~dDPml!Mpw_XM`p5Aqcn4C(h`V-1<}~OA@vfE&<)AS
zi89`_n*IL24x@jB2pM}5)B}~IT<eGSO|w<}>yxQuDIA*nh?nh9I}fjCKo=(d`_m9~
zpnt62C;9VoG2bA<`w0Jt@rze>rq`&Ehec|1wCtkVQptj)`z*Pvvh7yq`{18V%*aJ7
zLK00iMlAWO19W-5dr`h;ao@PTp_uFal&y9m%RL-GY_Ei;^4P&nuvZ%6r3SAczcipg
zJU)rUB>axo(<8<{ru!D)jnJhg3@xaIikgo^7%`7}7YRLQIP6($*rZdGVo~{r_hOWb
zW}yvkHqbyZhjyN)hmpP?5N^f_BW^*;L}5W4TT%K~scY(kgqRjrv#3T3(KF&8bZTXS
z)X-#vX(GCV@B9m%7Pp*M(i0%6Rt3G@KRy|_)>O(9fC9O+R;Yl+mB7KXZ6&^ggtzL(
zvTB3(niYF}jlcuS0ejyBUG6mT-Qj)o`rQ#T9Q*Vs`&i&O=oB}21GUT&wcK#|W8z2i
zrI-_f)-<$tLGn)|8ibHQQW4N1_9I{qR^KBT|7|nY@oZT#i>tlUgZcj-Bd%d?i|xk>
z6mqb7Wy^H`zV!&O0nZ}2Ie|m)lpqa1b?McL!Vdx1xq4IMP}qc9x$K|7m}Dv0_5Xt|
z>H!wE5=cWtPe6{?2kT!Vx}Atrie;GqJp$&LzY`u%#zm|@t*_$QpAX{oImoGU+6PX%
z(JVonj=8X@S&2-1$<&OhJK97%)x7Z8x5YY@K7$q?D`RZH(&dUM8S!N@O(>u^K!L)u
zuyhD>yJWt)$DR@+64c^~hV*-T_4B8IK@`d<$Qv&DwPiKW&{kO$LU=9~*>J$hw#aVY
zprv7SeD7(jDU~dnDl-UB+-1O24q%BAQx?L{dg?WM9)9$Wnuw0xnz6a92(3XjLMvo*
zXWK1&Un>tg?e7#H65FN}nmQc4;FF?E#&+R~=5oLx<OxaodO%PDC*8pHF^arLV?-Id
zGGz!k5?4-10W%`i)H7+`U#gWdU>gK+ayJ`ALVs1?GlcAkM-fuThXD`S-dHz8F)?FW
zw1fIIT?x&m{awz*lEo6t#KNIhDOx$B-2G`>xIU!~M?xcBwlZf1tA5K=i%bQ|hlEyz
zVy3e|g<8Wcan@$H8D4{A+9dqXTyb0L_@nPr8{qwiTozE~HbE=rDr`$bqw_dtLyI74
zZGIKIei0*ljc`#T;am75#`Hn3k{n`fH)jwfb1-|(@3UN<r(I?Jzps2ZgfUqhf;wf1
zW|U&5c2{#3JpTvD_P?{@L;n~POotNDxsJ(c^uN)e+1;9&Pxmz=l<>K5h3gjjt>zVS
zW0mlfV1j`X?57W<e-cQ>C})xL%@ElBuqy~Tfd3{Top4Q8D-(++>Cc3pRi5Xf)l$o;
zDB_=NX<^-Yd7x|aGP)5$q!wgGt;ZnQNS-n*6U$#90DP;TIkP=yz$48_D`MdW8;Wr;
zcSX04W(&xmH1nd%%l2A6VB>xjMn1Ri%%gEEN2H-Ck3WH)lf2=UJHkbj7zp1R@)&B{
zdKN2=tO*5Ko&hJ1l&<AVr_QT3tLJQ)u3>V;L5!h1Q_z85{28JMB)6-Vk@tp~$BcCx
zpKs8_@(HCQ$J`QTx7h2QRFVd<f&io)Ob4Yw+TLf&8>lAtp_H5Pa_L%iG5C<+Sz$nA
zO7PBf-k^Fa&S9reP0><XF$yikAOQ;SKnq5bikTe^VpE`u=7n>7){s_{tT{wWMKj+>
zb2InBTMr#~5RwNepnxDI2bUV%D7K+_QmI0a7cgA$*V``ry?)9_Q>yMqZD;+sWm{y~
zQg|?lZNCT`8LhWYnV@TRkSh!Zdrdx)oD@2S91*ejiHZHuvsr);r1S3g<k*azOoXk+
zLFT&2>)z+v-*4C$%K$>ePVwbq<bIo8*Gw(;g2gCPT*Vf#dj#@$f_^#LMe|@v`^G$`
zA0`aDtM{j8vPCpi<wS)-QXxCq*=PT9t+YYmED$GhGM+OG`1L@Vie=AQ$d7{rF+$*3
z<oxeA!gQj$<-@s)UV;qSz6Be~RvkJ%LZW6-KINne0aXEVFe_qKcPw%6O^sxrBBfZM
zYQ($6WMs<Adh5&(z0*m6VFhSj0T<Rqv~7l@lJ)BCOD?JyHkqW5Uv=a#ZCM$vC09O=
zNAh5evIt<%05=okh@q(aP{~x&+L@!-F%l=d@oT#2@bqCH{u>27s0E(X_l>T7TV-ti
z#zW|F6#%87sSUxNh&5gtm7?bheJe|222=qT6hI8^caQ<?4Tv$CWdUu%MG{34L>#wN
zJ1c(?1yrkTmlLE~K0G{N!oAe5WwN_K>1(l5!`zdOHmLy5_DNI3d<=`tJ^3~+;FJnc
z1K7trvJP=%$lzvCy*PHlgm#=qe%8vHkf28c133^SNJ~Npm7b4Ca)HV)lqKE+6)cCx
z*u0fzaEM&QFsQKKB`YAuz{;1noC6`_L4V2uuBi@qI)@8n%doAOrEx@$#Wk@xmsfh^
zaqR_UCK<~HB`5TVpy2D0t?|XGxn!s=Gl};j#rht->vWbhGkM3%Y%XP4ZK6sbIW6hB
z)z!@V?A^}gZNG)UF56Y7qbnhU6B^V-Mi>F{k^8|6a1g->O*9b)`KVyx^!)44i@$57
zfCt1f+gL9Y;c$y}@IHhITUKaJo~IjI{ae=r4J=<vX7(USFxDY3Cf5(IoL|;<wg=et
zsMY0*FfT-TDTmI%@hdo|w6dUcQ5<5PojKjUe8-xgi^Qe|aJTU{WeGrn<sd@V5)sm)
zBnS?$q~DxG2QRtOnvY$Jj)c$2qOX4EnewMFXH9{A<5Ke27acUOW~UZm2^GtiT)!@1
z)q2WDqK@_XMSST?55-P*yFkt-d?Yw;sSFs9g>V5I)PRA?qD$q99;WGfJMEHCG|_l%
z6F^6&^v3gjQf%+eM#JSEy10Npr0)v-;H8lWMtlkoM|^S*2)hE8pl}I4^Z@TEHyUOv
z^MVfg0=6$2Q&*R4ZiSma^1EXLx3>?viDHahRIa8Lo*XAjNdjtC&3GYrDb(fS*j9l(
z@Xx%tu7XP1dr#cjN^o~1M{V`EXOr|cj=HO>hwCfMX*4>ydd!ZqXy2`URyXXo4AmUm
z4juyy(la>eP?OQfSEI^%nnuTRI*J%tHqA12#aVv}P9^QUAM;nd*@E#7HuY2W<I{`P
zsH_%a{s&@-FARH?ZtaOq2>iqM=d`}JrNt$$5if=q<fO`jPeO(vqUxk{KjKBcy(tC_
zeNJ>46ChBiyfi?8yl|YK3ygT}k2nFvo6E-=C-*Hb@1zOl{%VYvvFtcqnZlHjTjjal
z_&bzW;Bp-GrEyAmUm?Oc-6oOePYA$QnqyI6%y6a9jY0pEciuOGf(jvf2457%|Dwxg
z-1hyWi@ebe{&TTNSzX1=2NT30Lh=b&hFS<wV=I~{+k$|73-3`o90x~2(XaSCvSzs<
zHEAOccu?GJ6qKA!e^Dn6M1Oh96f!0aYDMy%6+!ce#sg+qc4iB0^v<C@l_j|uGI$I)
zTQ$vcJLKy0%+jK+g3uRXZfbL80F+}HEI$;h8p89)958pK!y-<8kNom1xC#UuvNyXt
zBX5ZHUo$@vzkRV}owW!ohg;Pu_$})`sWp4{I5<MIyu99z8GYy>vEzQ;puW7i`drlL
zs3g^GP$6g~n{3wKfO&|cW#zr)N1WR_YtHK>R%go^Ruf~9=Jc^k3IuS1h~<a$x&`H(
zF1b=ha1p5%O~o{w>f(Uq36$hoMvvsH-3Rm~_!kgtCdL`d`>7HNbVdX$SeqK%TI{QZ
zDM;}7e3S1BO+;{%)X+>c4ay7Wu?};NJ>bX#-+6Xc&b4)LpQDsx)3WP2O*t2B9LsA4
zO*;nVxbl!B>XGio0w)-VN*zO2;zFv0*-Pp$L+1;Kqp(mDfofl}Q~>n54-2?#T736)
zZWAYNy;!PQM48>POd=aQ$H>}+B#DfY)I&ZUwn^WlLl@E}8qn$++?4|TQ|R^co18-G
zr|b9)yT;B2a^nKrmPJi#R+Z4w2;l4&4WctkTZy{^C}DLpiGumDiB%=Hp+}AT%P{DP
zZH5Os9uxUrNl3O{{%`Ez->0M8NkmBz<EuPoy;FT4<`(QxG&9{f36GPBjQP&%rbh(S
zB+U*Kp$C(Rn_0TTGY+@ISALl5ulPtWDuZ#wltmN{@CB$M4VdKMT2wj}f-~bsTg!XP
z2}9uvToZCK`6~o5*!!xY6)U1300t=_;FiP})3Bld)IP;vF>eT1lmhdlm|JthSmCX9
zRtwcvBENzX*W}YeM18*u{G#+HVnn5JI#G<bXcFOT|7?FgFMa;6%D(cgt+n~ONGL8v
zin|04Qd~-LhvE*!TZ&6@iaV6x9;~IfLvalrT#7@n;_l>^^PcDY0q@Lp?JxJod(Z5d
zJ!`G;m4Jt_G~aZ%QA)ZbuiZA7Sqr865i;0BxRF?5D<*+5RQB9O5Zi!dl6Hj^lQdGd
zDMEUv|B|%_FdC7`>)JgJNj@0V7Fl`t$Jx{=R;_h{FQ2KoVFk(Zz_*HVF7dz27?KGD
z!3fzaB)zI@4cQ*~)hpnMXos43?7sRyTfJ9S<LlE^1{S+eIw^em-k?!!?C)B`ZGDLK
z2uwLXIP8Fo^9fB9s)qFC)erVH19Giut)`ZAj@3khq}og)aR4z^M44}%da;04qvqT;
zX|aFcn)@Z1am?U6giPqSv-91X9<{K?iI=Q^NZJBYoNpPZ`_<O*-#-#fQx6cqZ6i6G
zYpnH`*8}hN^ERf{C1EDE>MrMlY<YKwTd|HfgEx%L$hawqFG7iKzaq0b&*jIk_A^Rz
z%OJW?=OniIWX0%N0B!^)d+Z!K-QWdwDN)!Y=;%xterd}SRnpO>)D}R*)&9u&{eSG3
z2%dk&X-2UZqDC`xKUftxsK&?R`D#cYoZQAVXMJ(W0N7UVu1K~iHUaMoMnlnfO5u1t
zES~X#<erOnNs7bGoyu==@O=mx)MA=fWkjLoAjUD?@`h|<Cmt%5^qEh~#>Ry$KFK8~
zn(X)G4+KYsp{MVNGHXX!M=<Hc5Nb#IPV1=prKBucu`vV<_&O(v*mC*GNgT39;-c@b
z{EBupRbGjCHHY*-oYz#Ge6IbnxBAl%DtEJGGMYi!!TbD>pB*h47CQFlVwNH~pd7Vy
zm6Ez({g(!*2>!4eNk!S>F?3N7t;+%QR#Q0iQs3lg>d5_;x$ro?M_6iI8F|}zz#>OV
zD5^FU%h%Ufyi=Nz<LvY)Yk8%d0naV>(@3kJI{q7aM1D<C-vb<~zH+2IOj3i^>kKM2
zjlHW=JLX8kMLR4eOKHe9u;Wv`9_mRinwBf4?ES*C<u}ZuV_8E0X^VVo$BdYZ{&>rV
zfKxdZMiH2YZR(0J_9Mtn$EC9wMZ)?q?1XX#iXU-$ZluD~&(Mbse|3q(jIz60+L(MM
zhO_u9IU!qWp8tMLZToO<EivuLAS3JZAkzCeqz)(GMmDN0m6sbBj&iw4t|I}-RvQ}d
z=(7+s$k5R*?51KHKR~usKCkZ*mQQ8I=9TwPnl`{Hbt=~?i}>O2u8^8fH2@DV69`Lp
zdwvkh`|ocmTznHR^Fs6@fSB6Dz8fPJx0_gD-^Ews^s%y}HmTJxi6JH$RH^xgOn{%i
z_46N#*4ipD=Xicq1*pHDWvSQiz6wAauz){v!#RkBX+<|{SX<k=yQWSaNb?}HcD4JA
zd4*6qV9hvQUcV<qK;X;NtFu`KfA+Asd?~L$z<1$+zP2QmLmq3pCo`rV^Q*o-smJ^K
z`-?3KL=8OkeQ%>3rjo!f!V_PLz6};b9R$B%<winA@*dudK-YNH7gR?lKqM4HXGFrJ
z#4Kw_8E?}(e(9XLYU6X#Iv?;iTdY1(zlKq*&elA$91nP@yw=lwH?Ma)+wSoUb?3l_
zZp>}I5?oqRk$A+IW9X07Di&UrL616^jXEcS*8jNG5u1lJ)I>$d6<nKY%M>14m=?cx
zM|<*J9hdzCOs2<vdHK@B)3dQu`=cEoxV=7ao7LbIxESO8_K!|&bJcY{{GLA<2;BeQ
z2}Wp1!g*`wjqeTrj#)UqK&4I)36LyE4HP<NL+a*bR44<kVeNKb$=}e~lOdR57vkj&
zwKhWUWdw01&mD%NH2xOVjtQ0?LmL}F|4M`4&N?0)uV*l`WT~R@vW&h`QyJ5Excq*%
z>GUWtYlu@pyfV%;&t<ZC#V)z=h4RBTSr}cS(#9eQ4z$~qgK~Zwrbp5|)jD#y4+ACg
z(71_(`zI8b_MZLi^Uf>TO463EUJ=p!1Q<yK8ql`AII5aFGU!u?d_G?u%p-C3oq)Kp
z_h7O4#t9Z8^>gnDRlZ?#k#oEndP(KVO}qbjv!2aY`KX>53*6pLdx1)Z2aLuI<H8RQ
z`P^b<ADqEFGmv2Mn|e23n{NghL6%Ft8x{z6!fgX<?BBJ*H21y#pWne@58W63VZYu2
zDSK(zY4u2=_@j8oHVx$jz0#dgBeDdNf@D4-uN5lrZGw8vj5((1I9-d9(@FTlKc?4)
zTV$=@SsADyy5PSal=Dn5w{s5ZMO;g@uw17tyjlE?hzWtX4DX%4;Bp%Au_O4EYP=CS
z%<~j8p<1%*lq!*D<Opf4qjum{n#KEIILo2Z<F+p?kfL?NZM*-qu$X_a-kKlrNTa;~
z#Z0E^@v3QGJr2Hr((se&jKri8*oI7#bn3usVR7Z%FeS8h@FKh0{4J5J?}ZI%&^uNQ
z$#w`jQlq8I`5%nnuC43M&pSIi&Xf^<f*zy|2Ufx9$+TD@<)Fx7o4TANg*2sp^YWh>
zJ%v?7s!#@1KEo%ZhA!wElTCVtF{VZU9|I0Yi=gQSAk;ukQZd~gSMx%=fP+Cnt7r>f
zMCZ3ISgAew^U{mhIGTd9J8tKTTT032xxfw0C7VG{GA{mucCPTFchqNwRWO>)hSA#$
zaGT!~j!5f>-*rZ(ss_&O278t3?VcuyQ-|V;$55tfn@`TOVOvv*YQH=gxfZJa^3vnB
zcle>=CHI38oIBqtI96Z?OrE;L<mjaJ2T)=SInNd&qpc%NvDNoPQH{2i`s5qJA5}VI
zG8)~zS5vsqF*cqficOQ(UnxMl><zmLpSpM{``;@Na_tug8Ap|7cg5zlo=PD8sSw2g
zmfHXs43+ArRC1QPT?cpF47hxLnWNmp56EChMn+X!klz*_JHz-%xb??Ad<ygX=E)Hl
zE&T)cT}jx$EXG;Gh_U~f0UM!Ky(pE9&91$6_zxP%p+T*{wAqtKz4#={L1e@<i#Sh)
ztpmPe;^>Vtu%pH|xmwV4iM)2p0#A<uyM5ULXFu*iTfS`bDpG1y=#o;<Z0q(zyZch2
zn`gTPViC%ogHQR10Q9&m{XgEp7QyzJeJ-&an3VI~wVX*}(yVd!Sl>+Qnp~>wd&Z0e
z(O(iMb~ODCM_D~Ip};mtv_L4cC6U-seUl?`+<c!ic(xp*cT!pSIVNI-Co?`~l<Cuv
z06mOnW_5bfkv~6Cc-j)IGRvYkm2Aw)))3U?3c?pUta#p6lvRatAW%zKX=V^%kpbOK
zSXkk-%A2g%F%`at7Vgu5Cb;145a<G#0!(|luQcX{HE*|sJGPId?Q7&NSb0{}`TU02
zHhv3R-Q^GhZ4P+y5OI-d&@J}fLvK~CvLx3@9v*6(KV>JzR(7r8gdAVV;pNvMFCKI&
zL|bG4$t=^!dyzB#{Gf+cJ76y{(YOyzMONn_GUJP|Jj;8u-y{x|m5F$-YHJ9vrvveH
zo(+^>$_NkSDc&vKJsB`7-hIIIs4R<elllqB2=_$g)dX(yY@jojp;@3AFh?fgTJXqB
zpShs+txy|~L+WDv@HbkJEZv|k1L4=Y9;D<;E+9^<3V>1w2T_G-U&<2u^6&k`0>w1l
zjQc^t(Y(Oe9MJ=@%Wv{OX5@tU1HV$haQQMFG`!e_W`pS8Co*z0_)tFgZ9R^C?~KH?
zbZkE-2|<WQnf!|3bBlFT_e;gH9nVh`G7TF#<M}v-Ii!}y2reRz-jqrStcrY(RrK3v
zryKw@p>3f5Cj11lYMq75<7wX8a=&;XO-kB>y8ZwYkFajd7V*f8*0aWc&R!$D{(3}W
z<x|z94xx>fQjF3?N@3&o3sJy4A2k=WuPe-)84b$BKlJ>CJw?jISHNGeftoEE5d8ZV
zWRc2%t=jKZ!+}+S<65B@*6kD6brB=j(y~JTN|?K<g4c5cY%JuJ+8@XHJXiQyh@XG>
zaq*(<{^?{g&Pb)jgR<$}E4Elh8@IZV8A6&ZESMh&yUJ~X9yhQ4(#GGZB~(6n;j<wY
zQb`H*_eIT8DffFoPdDV2X?kuE2RQ*tm*DDY-;TwTh)MyCDB1jL!`F65N`KJvfv&Yy
zUFKd{tJeaCQcc83eLN>^rwkuxv|8*`9L3?{*q`|VuIV1lG4;hK7L;H#*#F%7KaThN
zpQL}Eq2+Gc`4vKx6syYu#X%*8U@RwZy#2-mXxvDY@fT=3o=UHnKK86ukuq;7GMvaJ
z`_3Z}Jz4&ZElL9|+}83v5+sq)p^31=>rMc+W@>4hNJm9zv>SVX3*|_-yW?{xU;cpD
zRWZGrDU=%_=pP+7pK9MM;Z5>&?(ka6^MMB{kH^FBif*Y-phCscQ8(gVr>6U-?Q9B5
z>~W!RAv_E})2BGt-XK)-UFNPGYLaiDh%dKY{u-WviGihyOBnwcC=NcquBCK0LV=GN
zLi=Na20O79y0Cz1_VjoDc6h!60X4qkW5ICZVkj!0WjQQK#|ircT&Dg0mt?T5i(d@Q
zJ}c|9kgk8|QjS2SHK@b{EquLik2C_x;+*Mm|EsH31$(K7`2|E#LH{F*-$O%LTdjLs
zKz2b`9ORXlQ>gv)dQTLYV?~9ti_14#a%t4c-o8T~U9Q!QY9oup;(Veq3nf8nkfeKh
z#u<Z<om_;v>+ZO$+#{VfW6Ska%&3?<H4HRa(1nE@YM<owfZs5&O1e@V!-l)cwYduk
z`#OVmH0DuIosa~W%#Z2nuXWfAh@gi7quXaS6(8$(%P6reK0HVmbcPM1di$nz-l%u>
zNIpb#dgs@q<Vp(t8|un1nxcP#oJHcso5Tm>X{qcFuhe(kDK(74^lwV>yKi%|()GVI
z#&D|q7F`+yerVQ}#Q>j`G7weDZnWA};45tjQHR$`LEWv!5mfrS<0}$3b@8$9J6K$u
zm(l_;OQZ$E1CC0cnxN-XyiHYRfF-M)KC);IX+84#UdojTJ5=D%`E`&K*PfFT82kPY
z(P(ytZDUj4H=D4gCZFt_%=|3oMCE!GDPnhDIz`>n@vJC3uK5#419SAN0^s6}Z5aM-
zuDm7@qQ3BXLPD_7hN6a05)SrQB*$Ho30DlHtqFdjYXF+n6lO=NITORe=s5i9f+7AS
zUQJP3iG(TS@6)rh#FUN=esnQRV@lhz#m7K6>Xi(@!m!olotA#8amh2H?>2dx_v!&E
z$)-{Z@{zw#JVi`Q&02bU;t-lvLtxu+@ILZ3&LobTwrU|hI#bIkbt3s20na6wkVGmb
zGI3$~2_xGgi9T1~l)5d3auzbSaZRas$QyfGtPI9<#rTv@_c-nI%9rMy4o}9>@1{=b
z_q<k(L^0=YYQ$+Oh#Wm4>t9ddS(ev?IML|%Jg)=Axy9G;{t42DD?c-c<K8Xhh58d_
znim|uVOq+(FuCZxV(A>xNKdWXdwW`@+f?QHmi~!TP#AJxdHl7B1GM{vJ4X#Cg*|KX
zp_^|zw8`wiwZh5ux=?-H7isOjyXni(HF6lgeI}V@L(n`|=fGKu!Ski)+Q#50zuCPg
zV8t<J?waHZIh0p}jU`QZL=;5~XMc+GdFABMNY^N9hB7Uq$I8n-E<$}w%L{;F;d*qB
z%mT}^#=kkGN>F$kL@SjKxT)~(dXl3h#YRJHNFUFY966t2^XA(36n!aib6aB&oxrlZ
zCl<4+8er%~F@Lh<9kgMYNl*&{63yV`RjhE@9P}1p%0d6`bul;_hoa>uh>bAw5X-J^
zo%<w6`nC2>b(x<Z%JYn2I<=oaEE~6lF6MUKh@#48IvAVKX5{SLUr1VQ?rHL87Uz6U
zozo>_BP!t?N?}YEW<QdUlPv0N>=J~}3N9X2T0D4!S1+zKcQZ`y5~N0aN^kRgV!ptG
zf%{Ax*obWKpmy}wgeIlI8HWvdYzq$$=G1#PQC)K+-u$%tS7hRFp_P7F?cB+aO?ZCS
zQy2d@iu%5P{$<;xA3XP@N<D`T-3FKMc|lkNX4aFyl-lU*kdV)&_$ta)I+8w~_<ad!
zqZGTIO8h)X1*Mo6GfIUioaB#vo;&Amt5el!VM_K{ojQW`9bxe0ffv3Hf3$7fLI;&b
zid6ChCrK0`N#0JMfJ8)ks>Gj|G=fb(rOlQ(om<VyW5s=?kA7bZ`lFmc&HTA5%Ozoa
zj+u^`LX=a&^o_8@A%RFRpWB5MFbP=PwDjSNU1VHwD{8n@mBtAkG!SX_XT(-IhjC!m
zg)K~@^tRl|!R6|A)=LR53-oM;LjLg_<`#DV`#4WSnDyM*N;V^`T2+dTNK-d)5Osui
z;XSLHG{Zv4SM6wI#cZo|Lwu;P3-@lqs_FVvPV&8U-uCG}M&oPKZM^4oih7;!lz8*j
zE&XRwk;}npHdp|h>F1bc?%ne4HzhY)|Bw_xQ%f#%je@Q@zzQStQN`(kyc*#f4gu?g
z>bj7{<0&U69xbo*s`&+X{?|J;Ln^9G6oekBrvnMEDmg#QXQ3LI6sSOOCaBF(Fr2Fa
zwl@Ge{#m-$A+b51(nvc`Q9Ucv>fIt{DdzF~o~JHuSpNGuo(y!nGpXuSwVn8IyTLbV
zopkg%FnFSRQK0&O9JcUDVjor59_Y*&6R_iQR)M3&0;o`!Ith;j2qfg@^(Q5jDVIIM
zjHpPaEhKKT0`t~xH>)nL0-tqqABOV20RdlT`ap##gfrQiauncZLuWh2^%3~Gp0JB&
zYJb?l;i_TZg8uetiy{#~NtpvRPef3yeWSK|vY>OWMAxOJEl5}F?E=a0i>6IWQZ!;{
z@zP%lM+cC013C`Md=Q=VMAFBqc6_|&vZbeWZ!YL}7M@UZ#8$GgIFFwVY|>*Q-0hcz
zD0zWRow6)ArP*259BG%8#!|F<*BmN>Pyt?!UZ^{t6{g41heRR2n<I+l_%{|}w1<tP
zKW`u7v=Uws33D%->=Yj~Esn>Z?{mXRI-ft~ZI}f-nK^IW!KF6!i3SM%UZZw@6L#<=
zsuBqOYGVgnr+JzqG!_!=O{iM?aInLBy(3V!s}9!Eu!|~GrIz$hf+(f1A5kGTnQb<E
zgia#_jKYt;QD~I4q?bsocr~|9mMoa|mf;ezOO3(EsOmsQR7K_ab=kUn@&-3$=g4$l
z)U#@`QdKJ!tqx~cRV*frTBMp=<B#S~!{wW=Ht}%9e*GJp4^oJh=kB?gCU;HihwpBI
zv5|?KHKI=dfoZ%$?V4oks})2W_Nv(9)vUDRu#b!`(K3&u`bJ{tUN>QKIdZ#r7L8|E
zZmFoFF1OST8kKS8nv58*ejLs7jbxF0VqCk$>+G#d3s5L0V&=lJ3?&H6{N)?E0pSTJ
zC*}vhB;maDd#9x^F}1dqv3|*C1_@zZ26ao`5#}-aS9H!Fi5q)kMZ+S5jeAJ|7)i~*
z+heta#RlAnml&p`8$~4EGrK=nyK9BScva?{2nT}|-d5reWj^}5lZ{g6ZC{oOn?7IA
zXkw`Nw0{;`gJ>l6ZzHONlkJl8aa)1;`3dqzTJyb8m#J&6{c~Om7%dmJ+`v0yh-SEY
zI<T=<Ak8s?qoQ}}uWX5&ypX>`>fI#T{4v+*z&ILKQ|o0ks~iG238jIP!jHT<hYlp~
zPp};*1T%~F<F=?WyWR0Nw96JcP=Jn4w>klxzjuZQgbgW-mLI{A8(NlNeMJQ;gUczT
z8r_*kku1&e#`?(==^$o3-<6>%4@cg1p3x0Q8iMdI-qB=ub|gByz4<&MvXGbhax{`R
zz$wl8jA_-Ag6sz^zRve_FkY(;c-j}0k<HV^2r}V-uUCR*edHe#uLTK8RCBNGIZd82
z?vEn=z3zXWog&9EyZ6-kCMNzY6RA+MUP}*;E7M`)xknWD<TvnBD(f*xKY4=-ytOmO
zH?IvleP{HPY8o5mDs-jM&m7^V=MsqNN#~mF$B&r3pzEu}+MPgg)=$qHh2?h<3AH;k
z62}cYH$;Dx{<GD!L+;JA<o{V4H><RX;=eZZGKhC3<$ni$FM$8w2e+N@YEXW6dSc%T
Ogel6Z$yB~G5B@(T*C<E;

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_020.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_020.png
new file mode 100644
index 0000000000000000000000000000000000000000..a2db0496d2a891c6daf5ea7494b4b8a34adb12d6
GIT binary patch
literal 346468
zcmYg%1yEc;vo!&N2X|lGJ+QdDOK|rEg2Upv1otJu2_D=b1b1H~cyN~lcl~))?|t82
zHB-0h)}6UCRoyeE&*_fVP?P_FMv4Xp2lqiyK}HJ>4&Dn6?p+ZI(%T!oz~9nu4?#~^
zeNSx{TTifsyA7PGg{P~Ni>DLFlG@wG-2>#}%*!Fb!OKQ%@9F94A;ih~`9B|UxVYPK
z!X%^*-X=kHRWR^?gTu1;pZ@NLcqs_(9UPpZjHHfF_TOhvt<6H(di3*+>lfk9$Gref
zM!FCgOB$4rjd7N8pq?UqnvzzN(&sYdNE_Kma__rT8gea2FBMzFYV-1E*)gp@Uzex7
z!Hp}K6*`|PE$bAL5EbF*xM`C;I(>g#w(e((-t~#yc7CIy|HS?K9s0Sh3{Uho7vO1S
z$E)K*O%vpH&4Z=X{lD$iG&Z1%YC}hD_rLALOfBb^wqV+S8dgZQkk|h|iV9B9>*)U<
zA_nJvG@CP8mi>P%0B&ctO!&W3Xy3yPmuZ9l`{UA!ci7KCPa6B=Sy@?)hU^gR(;?dI
zQ7WnSTMQ(G`Gtk#i^RRxyra8yEWx>RWHU9dr=D-x6v1b^SQXp117&|!R(785FHeUO
z#nSR5PgYv{T0Qn(6wExzw}ZQObcUz`{yQxATf#a~Y4x1LyR*cX<E}~ZhniQx^@o#K
zYJ5j!JD<I!*nk&SDvMY1M7hst9{U{BkX}L8wQ%ePd&`FwEi`8OUg4q^A0x{uK)^3~
zlqpRqa(aa_Y?Usxf+o4A^aWN(cPp+H#=K$CtOCFdW|1RJJ;g#3qpgDru<>C)_1UkT
zx2SqAg}>7=BpmCpX{Ch<xoWVxUE9t^IDPj{D*uxdIcSSXN-6pylAWp#lH>WBRwk~I
zp|?QIMUKmp$;C9%Jz_ogUrOIW+g!8f%U2IrZZOl6X-JFh&1E~7*pVv&!0LV>GDIBf
z_N9Prq#pIP^!12p{OY>l;4b)9{KX^avg=vYIrx$u^1G4Jz-o6?=$rT@arjykAE=?d
zIZVb%^ru7DNJny%j65kleXju@lHMM4?vw^1kM@I}8MGqdC<`QoXD3i3$CI7Yyxllw
zqGwix(M>W}$)d(XN$uDN4IoMVLdBmd@D^yHDNM}MAp!|2`e;^XA$5eOBE{Uqc0K-{
zbQ+@o2=zQ^Vkz5aAbJQc4k&vj&T<(fQ%P@N!90fR$R6>*Z<s;QK-1lSTS)wFZ@rY<
zWC$NuyW6KRaTk~B<sx{n_volDA*V+;@e)%V*ihUAWq5L92pA`o7lYlsKDoRK+!k-Y
znngizs|e!#!ub^S4ULQD9Nb@41ZU_r|2_yQE5}xHu>W>wv7bFcVXht2wcENo36)Ta
z*WzZGR-jLoqQ=<>9u4%u-2d!F75H;~h87}&68iDIje+jI$~)X*d^d7Dc~mKsb}Tc-
zPuCw|o1p=~jf$>w@M6~^TCeoy6O72mNT>DH-CA^HX`iio^-IUDmpt4dQawW8KC6m5
z@bsKgL<7h)mQv3|Y)7)>qoVos=`#49^M0DlMFcxgwQH~I{xLZ5b&u+ak~8Qui<-|7
z-OlIao*-!R8>Q8oh5Gr0UhY$t*BV-i-ZAT<tC>4@I~;x(CF6SU?43w~B9A{8<;8A(
zjlW$i1-ix~K?5Mp6X|PP=E&j#AcK$JwE=WWgZ*w487vT76hyRczse@@&^cHJp8f}|
z)}tl*wRQh$#7u>`C>m8`3+P8Am$C1hj&5s~Dz%KrWHpzgv9ET2eI#;wd{DAX4!*I!
z?Bu<URpxc!2XRoI^f<)Uy`viR?V;nWo!2mwwos~P+Yf#XehyywmZ1Lc9J!COi&Je?
z=UzNu7n6u1XM1kpkABv9a-!G-qnKAlS{gz7@Z#!n>8D`$GMY*}{citAE$lNOlR8Q$
z6fGLT&eb-o!Qti-8l#W`0jl-JR*F@BlEEKPV20Q5iyeBGFcZqdU2wtE4vv!-{e>({
z{_k_a$86CSxr&Z3ZLgy$dWUO5pNIw|ZCfAYLBVV=%Bk(392$1Y(Na%34O|lb2E7&&
zG(jO1N~ahKCl+?t;M4B4Q}TWI!@_z8x^tcksJ+?eu#EK}{$ZunH*JVA7&kvmdMD=x
z+~lS)9AB2<R=qjT2vba0J)ixjt<{c)2LsQiL9FJ(Z1FKzuZVY2-OUxAaX?Vc{VqR~
zdN60<suQg>qz+elDtt!givbN>8NL!T6m0I{1W0Nw>P4U-n66wXD{jird>`c{Rj4bi
z96e-}GvJHRYY``Vi1JSaKG7q0#I=9A2M``EDnkbhc8*BMke5~6+xrS?G(@RUP*&EB
z<3@6~;#$kk8F<_n2EA88>Z>lI9^;L>-4tw>8H)lYfcVH7k3m4Pkq#mL`I#{fq`t7t
zHqLk8Xxe!5Fjr<p9XpZ3nbppTdwHrJ<!D!=Q$*)-E0;Q0ehzK@mwGntLCdN>{FBgw
zJG*<KV*DmPdPQ8qJz>5hFIaVrrZOJK+va<p;48)|JcTGo<B=4}C*x${G}ZAVz&~_c
zw_<sKMsQj`%yn`qf=Xzj@jNRv4a1WZ%5c(>`EA2mdL+R#K+oRCC5lgJ`Dbg|&=k6|
zpys%cjh~>Z=xuyvo88$@U}L4q2I5ofX1bZMA@L{WqHBI44&%L}w)O2>tQYQ)TDs|&
zlg)QZj+s>q44k#FZPrrP7%!AJ8%^I4l)W~ELU&|)mY0{+y^c=ghs6C_$0C%E+b{f(
zkC6ketn}{s?dJ6j47$g}+Q@_(kJV{(!n{ZkfG(QL?<<?j{l4Lr(Ty5GSieI?o8iCf
zsH7AT=%NRBQ8mV~R$`v=n>W3;VAfcqDXSGC-kv^)df6m=UKRc^?0)jZ@}VU`k78?{
zwCRETCyL#N`M5J(^X!2wY*-lhT?M_uW<uTa9`SLq)@F`W3`*6fh&fcac~Uw>A9t)Z
zLxwlAqyVpVBv{L^y(9Q5VbrTBX1%>9Y#j8YsDXBf0SvvQwCkhr<_CJa2?4UK@nzLC
z4LwP=S345sCJVN;{hxYjhfZU)1=50_IoGVA|12&bD{=RGr(J#5xBPdBFLq`$Sqc~9
zma$=tY3AfnELv$=)Dhio0ZH!xR^f2W{gv-!lp^MFH1R)04$uO1sV)6}l*o^5cSsj#
zc3S%%GZ<wFq+17$2J|nF(tM*5l&+X7VpDoRK6SOT%5ppW>9eT_%kfWX6ijr3x9HhE
zo1xB@_=FSHO*=5F0F2}IIgRqdut$VvPRO%4`mR84mt5|FSxQ$;?*0=@@efWX>daLE
z(a>itaQ}nYCAwL`bS45KW+M3aflOGE4<f00n7l(gixPH>qUlOY!ur$0XD7E1PCO@<
z@xhkr>9&pm`JjK>=~CwIiJZ!$6HqBhXZCsFMvQLrQq63*&A|7yAw@AyQuulO?{w18
zaE#%saSEEx7%DI#v4q?R|ItD?mll7G?ad?Y{#4h*ujFWI4Jhl|GFJM!|Mz5NV^0ms
z$T0Wk71KAE83PEGYJg1-Bv(tbM>PC#6yQt4oUCP@L(hqpeLI_q(vB$s2(NbXs>Z=)
zR;&z5K(BB$+=@wAdoh9j)7iH(7Hr^_BEE6%P`p2%OybW(LAWVO=s>hMC&ouhlxK?O
z!>k=13E!VmV3zjgOUQ6+DVuZ17WDYy-8ON5N%UjEv9hXa8q{RZ?@*N7JIU52&m20>
ztqCni%gym+mX>3SDew<rx2{!E($@K<2Nao4!|;oATJejdp>jQ0p=mM<ciJLRRhpkP
zDfn$lACm=Wu<o0bN4$d4JgscVK!lkf)G;*F#FwtviKo_}n9&h@A4WS8g!Ugitn?{S
zx{je^>VYwR6Mrge#HFg1sE;U<2fwcM5cWytH4t&uw=_jSNsmhm&_b)@a<4fMnSCHy
z!%UL2Wdf7|NY`essh{OiRmZIpRub>`7Mfe#yuNka>4_<QT^Me_5Mch%qOEk;?MvRw
zlWysKTXY5przOWtB%&;5i-!K1c~tLlP>h&k)J;Q=yhYlzg0L<w`6cE|fQK7sRm99c
z7x$%&u{LI^j`G;Kp7FE@fQi4Rpzr5x|6SUT&SHhCQCIcH!gDIq!H+0bR$m#|th9qa
znR5*|U@iLB8tP;`yK>Maf4y^80Kbaj9&2F6xUy+bHl7&=?yHL3w{mJbPSAe%c{YGT
z53h?tlB?R^4=;lch{df>8QFQMdzhsR=K6d7cr**>CVf}m&s)Sa_}4{#rPFP=g0oAy
zYkgl@FV`}tGrqSVhtXOy{AL$_ffZYKuE&)HYE*(FjZYze4+#6pTexA&{TLLMw0X8x
z*Wwi^VCo<36d*y0PWo|T`A7sk)+=)&iY^*kN=uz=b~}CKAgU$kjK-zwCJ<)Y-YbnI
zu3E*NIEBWn6Sa&=kB|5Xik&Oz7y-pGbUgdC*jFbXME)GLElnq(Jg^u?j47bOG8j#+
z_(7Nwfil7zY!}<thhXRti$W*0xOAIcQq+NUo`j1F9iY`LmMz;pII<%mp^5v^--21^
z9r{<oqn6wHINoTd51>cc+I?i1MJ(+kYD1#6<`eq$_ZS6DvjSoV8NDVCaX=LxC<WWm
z<dCpYi-{z+s}>JJkgu~LEra=WwJodtdhi%y2w@8`h0hjLJh5dQudDrr*?c5<6BG)g
zUV!_LHSMDn;y&>nM4T6AeTkYCV2wKXVvP_f^UprE#|b?&(X=#GKL&8t<Qpy->~`rx
zX^5RZJ!2M!WX1`y$;U-zDWq);zybEo?g1#_MLl<CnBk_0wvJ*lA-!-&?1+1i1W`yz
zA;zEv%dI@ar&qoV$BF45ljmjXflydLY!k0zW9Dbl2?kIHA&+&Ma|x#UFdJ%73yNLc
zr&crxodo|okDA|t6Bm-Z-`mRj|0-uEp#VrQF^3sxPgeZ0oP7Gz<R=>%uegL_m@ZM}
z+gg@Zmv-mp^pa+Q9w)eCb{t#+U2;%5z~kNkTV`IT_z#}Jeh9uwcYN9J{f<(u_lCy?
zAl-_eEL6T{DKij^?myyB3@tV5#U_#DvGX`mXb#NjiU}SGH^z|@ZQ$r48J!FV5#oJm
zBzZW)h`3NZh23a~_2<vexlY&NIHBwW!x%1D!<Qd#n)g}*1IH`Y*UhJnFIRR;4fO2`
zA|xNt{-E9xYub>8qbot=x9TqFbdQ74b^Ws^S&cYxOm*npho<*Ftpz+Es|*Uy;!0OA
z=o{wPS!hl>Nv0rTcq^ush!qffwmyWf@2)O%PAm_Uwz?kD7RB#|%<%-AzTw)<iVf}q
zp=>Kni>6$`EAi;Lw?KeZk~y9iOcKvOg=-9NsyGg8|F+n70>g}qe9X=>DjU5Kr>Llk
z(v*5%c_UcviiK36{*U|Yv`aL|EJp}HV7Q<$d&Yz&P{3sL^r9C$44WrDc6VND@9Rs<
zm4#RGVT0dibw)V3%ko7ZjIh^0D~gk9Rbz(6*bbVaP(a@URu9YZYp~#UI_&KBIbQo-
zAE~IHL&g%bOFx~OSM6??i(3OS*|+*N@vR1BzxnZ6a?PTE1drDC4i5NRYF?ABm8rJM
zumYh>28beS6yMNuZa2Yip(WlA%xLhC9Dh-WDi<Wg12BVaX=r+h@qmmUvAIHPwsCfV
z$tm)z81&lmBK*)1NSCxv;{Vtt@^p|If?BNVg#nd<Z0|w3mtB*SkRUR{$Z+^tG+q?U
zT22`nUumHr*ULY8B&P|k&DQ%Tn2934VrXP4i;47!XvFnG9imb&*7UQ%_Usx$fEm&%
zJ~!xh)=0^+bVKqukcE=Fwk0ctBhKczlaupVQ}D(5w#wjWvB#`~hrY`pB}G1B<DQn#
zhml1U;3#8JnWcy>iUkSz=gav`3TG4l$R_1)z9)T?Ne4Dbl#U+QJ)u0Xd|KH`n#S@-
z+#G1bCAtQ&bO~LZt^%M|jL8s9@{pp{>+dZlv+`tXWN>KT2|v~2lt=sBT(C;=Dm-~J
zITui9qqlgh2PLhiERE4qrPvC<){6%5)Umh<thiI|I`chQTiw;BdL_1-|CX$UP%k9y
zjufV?0d*(c{GwA4iHE5$U?uroXV?=~&)cgfQUOAm2t!G(JwwynTWNn`&|CM-pb_5M
zy5}|awU6)?rcsl2wWy@8noD<V;=TuTO!L<~pOH=<qk5npi(DOOi##k|GKoeGL3Aqh
zfG*f~Ns@%Przc)FCGEtKWut5PPT_f9(eq7R+<bf_vYVOePXXBlEiLiqUeF8ssvoGs
zO0f-Le5l1g+K0;)vNT8l`{XcUiT4S()W!5M@V1Tn%l(=vraQM)f69D{r+!6`8u{?O
zm&d;s>3aRMu(Z?&L>6#a+nw1ZprhOs)n&;h0&<?wm^V<nij^rjhBlNSs3`}I58+bF
zTv_RVW*G-7Shiojp_8q=C-5r~-=neb4#vL+a#%Wv?(q$|n^Q(Qw{)9)6??QCZblmU
z&06tmA9(TS3oh$7u~9s#2R_59$j9SL7DxPAe(M*4I;HPugYDmRz8xd#MkWe!o)Hku
z;Qk$Vt(De{1NtAPlHWL=pO=bF2m~LS(IoKDY3$(L+}!#c;82jS1taJxGxm&Z5VFR8
z!=((-20J8*6W`ng=4jhR$jXwMxXGaLR5q-1h<<0nD7k8XKKXoM8Rm(5DDjZgTl#<p
zltwyrZ|4jxm%(8!(8rS%0>Ys|ae1dSl=3~pv{0oSk~Hv1{M(EIwvW3eZo6)U{M%?p
zMeKMer$8-&;1jl-2uy0Ah)BPFI3doGJx1sET!*iF!i}Wdg*^-cH9?UrX0J$nYYa+J
zBuiq%?JOam04&%{?Wse9J%TYjj9GC^qo~E~2rYx??L_9H5hEn=elje3>QDEtAHU^2
zuu(nVasD>dmpm$G+aO;p<S_Mtl#dst*T0j{$*5+EiS~@s#Zi()p_rICyrrd#)Zo?v
zsj!R;5TCO5Rf#q80k;?1>tuDU6-qVc3p?Z$v3a1b*#g8_;Uj1K$QGJP*h86o(Ld&B
zQuMKhlKR8=W@MCbSY=5U6t<9*hEU%3jf6~8eKx(Hk9F#}{#3ul-^?_8$z3IvdfjC6
zRp<lU2_;4?tS(*YFCD@+ywaHYUJ(#4fXvzBv4iS-G_h0t)jnJwbxR-TaKp#PAWct)
zdRYl)v!hp#DwD9XSbPLX{DGJdZ_uWv-!G1cnDcC&m(4u^DUgQXlIr=$Z0oR8yoWPz
zE4ER{4%;CL8}IC^oA^Px(vW5OHVbP>8pAxQ;Q&)Hf&^2<D~Ycg<L>wa4b|h0&7c8M
zRcSy*y23F@(lHA<PwXgbH3?!6#^Q!68d!bsON_oKd$B(iEZl3CO?WTcj_2gEO8kzf
z{l0MB$_B7G9l5~FVAb5xeBkw{e_Vd4Q{d?uviwh|%225-H>LkeU&pP_Y0l|c$G4#H
z3E>*rH0fiMs@Y16fE6FW>4*qYZ)4f;9Ofwo>}$Hal!2}OesqeE6iA6yN;-*?$z<b;
zff03+^iV>pTKpWZovYNeQxgd$GH<DVo=>TiPC^o<aS>BO)K-*tTYLhgQ@gWubR3_d
z3=kiWyW6L4m5hF%Uvn=-V13nA%a*BPYOpY;;s6(u(<x$(zBj%{Jp{K5IC!L=|B-1|
zWAv3+w^YGI`UlY5OVB#lC7h`1oGjqgUtsCa4fGyjxnk8RI@IcM%v&i26zV6IR#6cn
z2c%Cah_dGJ|NBKlX1agkZ?=0x);03eAAz35h?_lf1W%tmCCSdyZ=R2P;K)skPJqN_
zpe;+Nu~*81Audv0hG*}x>qWr#u59vhW#|>bFqYj~XIeSC(0{g1+04fNsdDEps?A>s
zafK<C6`L~4DOBB+uFn7Eak_T(IHy^pVu%5JFp_owF5&MnWDDLTSMqe|%-B&Cl#yp*
zTuwRR%=ela?`niOqBkn|9{6mIYWV|lD+=uO)PO=On`icUCq3`vk=vlfg?}h`N!9gD
z4S_hSh~G1JNV34UKDSjB)=N>g6>=T9-=Tw~>Y5UT|5{(xDpooIvnGSi7t)U_H*#c@
zVzZKKWd{TW_x_C+xVa>wjMmfam2W#m-ok4eGNj_{TKAtm#|A6io{4%-n)L{i#U*e=
zLmA29e+Xp84fU~PSy!k?+dv!P_PIG^Zhl`ZhBHlsVtkVsJW#M=r0=xFQPLAvcK10>
z3KB2uOrn7+4v2cFlokr-=J(|G6=of2J+!Ha`_;E8y1B)KOk4yd)`=>3GfD53B~hV*
z1v2ix|2EMir^Sngc?@G?8&JlDA}?}-bn9D#y9e7AMjE&k8{}pbk_@+p2wm8PZdN*O
z`2Aa2?w<5Idq=lOvZ2JE>o}mFij(GAt8!asEDQ|MjJB{b7{4mW{$#R#R|L0@xZQ5e
zhbi=O8oYmeh3Vq<+$#;8r<I9GGAUAFZoommq)70SrxqQ|EazqOOqP?PsVwxwJgzWa
zRA+X1VRL`k!o7@TKJD<24B*K(t*|nI6FwG|@+-RdG+)kQp%Yw$o!=AQ9f5(l;WlV%
zm@DX=>vT|rP&5_Nou2Z|^g*cWQTR0`cAqG!KOLx~Z)Q|#T9Z1jHMku&Z65_rm=Mk%
z-|YaX@%?LMAOnM=wup@f|2fvwyKM{QJ>tCrSjULYEth>+Y`+-m7Sl0HgQkfiJ?ITt
z`s40g+&G4OBWAd%582pwTU-;WvOZ7-?FgRyZ6E6GC0X;^&a%(h?69){MR@GPaOk}X
z_-u5%ni$+-E}hGP5d3_|_H*201U8?Ms$ms>;q*bLmoOexewsuyA~PQ>&F52ZBYynW
z;G9@^kX(nhG)f{A6QDBHR2K!Za~ps@qgLK*Qlh`6>VGaHKeALgbqx<c;phlQ`1R-y
zjVS!{F;1SIH8v;an3zW#O<9bjp`&GJPN0T&M2^o0=h_xBLn^EV^;M}daX;>GY4AP9
z%zN1iR;lnHbRn2&J5ym)au)wkQ6-?T$*dd>B@Hg{Sn!WvrbrEd<iz=&=RIRxwYhlw
zeBoQ$&DSB5H1=UA1b`YQtl#`n+oA~p=YUQKjih^q!b7AO%1b)*e0cX3RARpcl}14X
z@D#;N1hf&oKXH$hiZCb`;CoJPlB%jqz7p<9(`^`UKo`-dE344>hgOjHMdpdWm!}~!
zXu?TtRuC1r_NLk)oU8<7#s68bhc$s_k7u<9BG5nQzMlOy2DS~m0=rug466wLjh-#m
zQ)`7`i9g~$o91o0Lgr(h!$lv$DOXQIblXDN(NCP)C%|S8q-GDJH2|H1QM^xv3rl68
z>9`#1gOGV!01IIZ`l@MuCsoi;<Gjoc9)s_f1_WobM@*a6v4vm(D-lxaB$;IDXt+}Y
zeM4Ix|2VIZDSb2DX-QY1Y=PUxTlF^8Q==i2BDG{8Io1eX`f5JFfJxw>Pw@EK4Ne7m
z&QX#unzj~cWD9>>8On0DAWg53l9laBJ}#M9f`k$koCa&<6-K0qvvC{jW}Z@!m0p#m
zv3lCLt$&G?%GS**dc;*`LDGQ#yMYxxYRpVQ(atnu%@l==ikwyB*DI`B;#(w@FY<H~
z{MAg0e-?Ki*Fl$KT~8HEIHBk(fYbcOuRmBi9IARq>)Xw>@KJUphFLBeLQADxy@?z@
z(jfM+^<_}D)I4KrpIl4^3A~*|At+_|S|m_Uk>=)JFmfOUJSS}&d-+U>j4}bKQ1-xs
z(h%0L+pXG^!}xFe6L-(r$<baKCFROX6g@#N0&}Fie%e|TL?t-fY@qjEW5TphTyn-m
zdWqe%ugOPb#P1({drn{`rY|qb#)05SeGrEUP}<HpL9S4;!xY&rIX``JRk-EQ>(P1%
zV#wa(uh{a3esbt9)!$1EcN`0fEThx&$=5^i9nPoXSAzIKnsjU%e<IgG)v->gFFbYS
zaiJb6fxj~FHrVP?WP|{!q>X=K6#z>sKL_mthJ9S48n6$Et&}paNrj@Li>VXv+Ogvr
zP~YfVfqZMaBAJQ$`WwoPnz^G{hr__*>mn4o<m!B#51TaER2C))&=2sstlllk&HR}9
z|2)epRf%uSbB6w4BSy&VyD`?hMTI;Ycig&TB)Uch8Y&5k?Ju<ar|rsxzY9J&PARYx
zQO>lVZN_Gsx$u&EBj~mmHMUEGB3RKa-%pD9M+CWay8`9pQ=Y>q9>T>|oz*-}FrXwx
z;NW3!*P~Iu=GZH@R537*%F64b*mkL9Q)9#Wk`X!wnt?4Hs4AX`m*Jk+sb{aL#^N90
z`qT1j>bXx8Tmg#1rn0`i;U>%na!zOsaihP}@YI#E<kr94)(?CwdHN6e6L@g=$+=UY
zfbpZ2Jp_MzVqqjwNJeZlRaL3lbaiWVn<$B9&{xX(h+-Ze5cAIXJsfZS$9w`&M7|p!
zEr3NVnd$9lU#NMBay;$$0{*hO5j9uLTgkzu;eVpIw@=m&8Xr9>gcj2(ew0$YxxGzS
zVIZfV(2Uh~Xd)eBRYn1f#G)D*krhST&Yo^f#a)WMDKfYBXP0@_|NdH`;N7XLz1Q%5
zStMH^jXU=-`+|}9?YDY&8|oKafaAf>FL7TmB5B(Z%siQ60sg{zZnp4R*p7<0{+U(2
zZ}Tgm$uTgh9{(6EbB9@|61SK~43$k)MZ_ez{p<8<zfuVAK>`%3LE_z0B9QohZ?H-E
zkJejX4vJx!8OS9=!cRnJ{!C5VhpPn^TOE`464Sjq_#1~MmJ>^#G`jHte;aUW(lMHI
zEYK}nFWI}YMf30e{TAz_)@fr@N)+ruHqNam8tFq;q;$*(q-R=lIDPhboD+teYnuFf
zby?<+)b|xdUB=+XcORMad^#~O+hd=Tc2uIE8MaBT|E`==SW_>z(luRCErzB(|5ynn
z(o33$2I7>|OiNQtV&gmPP9VQ%DdBQVx@Jv1)BR~$o-ZqQd_fDh7|4Y;)2K*CklX#9
zyST?MoIuPv7lK%Z{qO?V5O^H`-#2xg$(lZ&n7}Icj5|G}i*mPI_2oZ0OKN#J=-|@$
zpZ3{@9LBK}mlne$`197#xjoP-P}cV_BITW&R`y(jk?8?Sd?>Ol!sT}cINRsxx^3~d
zxlVjVn(82dBj_L;d@(J*fOA7rha#?{cZ<NZA_XJb|IkWQ&l~;$d;7sS^fcKyup%_f
zg4<@Et0Bz#27YVy^2-C(UICvJ1?_oULuI?9WLx0!JJu0ZDyC^eNZYB(N8~Y@h@vH@
z)!+78#?-39RN1yw*P?WeA!%u!oL7Fjt^djy<(-`U`J7-$&6i2cI@Gw27ZU{W%0IM)
z8dRH#?u=@)GMUxXE(3A~GRQSMSQ+sxW9CW<WOYrVf4l1(<jtbkoELQ420EKOQ!cFg
zzt6VdQxI(U30r;=nb@ba3=(5dA{eVkv`|t{VS#K>R#wY1gQ^B+M&YHHE@-SA{j=!O
z;`{^IaQfW49cx}3#jcX`_Q#haHEis!&S2Z`f&JXN@;!@>Jo12EQOtaDskmCHQ2W9Z
z2K;ce3X1YY1_6>t;Zq6FIWF!8MU3k`Tu{}Q6t4PBO6151J^CQ^LyBV0MHCEN7i}4>
zE*FZt<!!7l|HCHz@*rAX@c~Dr;G_^U4~bkpNs~L-VIW!QPL7|$vV@HTZ#lOA3T9qI
zB+_g4;suSoYRJL)J*YGF!VJ!8@g28Fhx8@n=kv$w`>-kjk{j^IWwxzHvWFea_54XF
z)hVH>>VtO+_>kf{S&K(tq3Aa%Nyr<g-`fknZ87M|XYD!rfu6J5uLlOhvFRLYMc-FQ
z5&H9n-%f?scl2KW+C*6v)i%r@YLuWB5<^f{q7U<11}Vw-NO&lezTCV5)uIa!8F9br
zQd!<A;4pspLM9&~Ih6ZL#pkMy(N&=_VYog{t}tF}Gmylpc1>?kD-s8>Kn6vS)oLY?
zE%)<Om$1|*F|%bmHk*my<PK9Ixy|giJ>aj-v6E{Pxk7o+!f%#QbdP0~kA<Let4(S2
zp9GX33SGh4aT;-EL|uHQQl>E+m#%9L;T9Lf6Jyh)MnVH)18d*F4EOwrO&EV<TO@`b
zEktK_KXbf^boMt?w@R)}bjV(d2gdWO4=@kS_m%dMi_E&ZRoa6Y(<nJ$Bw+g7|KW7I
z-zxkG1mJ#EH2AV)N33*=%>=<D=cY$U!k1Ot07VeblStx*e@&6AlEWYHN#^Nt&+V(0
zV*GG1A#F|!RCLdXr1XVnrPn0q%}}v#5&Ee#RE!O^F^FHBQb2{JY1m|(=$)((pN-yD
zpNF6R`|W&sk&!4q6rpVAMv4CpfmMJBiiN6*Ig{3dGAaR2_(>q+$K8HDB5rmm2%P!8
zl6`T#b7bP!CErrEEu(e{<a#&*&lzxqN-5MOXF#|?uUM6Eo%{SOTG@2vC+gsa8#QFe
zovAXVD1~3T(%LZg-K@60e*7qdcct@^q3fFSm0Ml>#bQqnmLE}BV3UX)DsE@u(+pZ#
zRF2@a{US;jDz_EUy>&K|I_uaEN-g+XSb%682F0W>SXtyA@w>vj)i*U`OwcBc_xq>-
zUe*+3NB4F5$S<YPOZ?5@i&^3HPnV8dC^{C%xnoz8M^Zz2lw&f$I--w#=(~%7iQ@(A
zSh~8iL6!fPN&_YNQpmuqWm6uEc56Lu&J>0G{R~s(;rR*Ky5;tK1qL!1erk%GM~1ac
zn)1HxNQ02kAMg^@DKC&5sq%#HV7%qD=g28(nkt^y(;qr-{vKb<66`3ILFIPRiwjz@
zwCb8<U4Ib@KAIGXNVKx{#jW@Ni(o70eqZ?&(qO)V+G%f3nQTaHEknLAtr<fQ1Iv&f
z_6PqMP({iDO9DMHJ{J34y`j%}q3T*cXZ|aoc~jMio3Us(D)}}tc6QEx92&g7)SU?=
z4#)_*9G_4YyS-1es@i?zT-}JLa`reLBt-TXb=XfROa_q#@Gm{e$OP9{Ug4SNi_9=P
zcib}g{}DL%mf}CztfWmKB)P!^j>@O=>>s!$D)L*qfg*rPXZ)qZLu@s<fCQ_!IvKa#
z;wCQ!o1VFfQ}^e+4+#&n{QQr^jQaW<Wed2NH;HfNyuP7q_N2`x%xW(h_yULjdm7{S
z@=K!pa{=k>&N&?jA5$Y)tENeVMyd?IO3N_*zWg_jH0xMKT7$^r)8zr3to%DBc(CGR
z)WVuSB!|*6Q43oZzN+wX{(S9Py1k0lI!=dK%h16ifi=$;*N(j<_-9{FTf)g|Am;7R
z?4+*HjpUFLB{+trB4L<$q{&^OQj}X*L8<+Q2@71kx?>JTw&PKm9}<q7kG1Z5*EfP~
zq(V&bS5r-n@0;$Fkgb!43b(BVC_{m{(o4qha!0xXA9k_#CTO(zrjD2oAGji%5;Wa6
z9Q;rdYo9RU>{2)#Gt2L&m13Wl5yHqNfTla}v|oRkR~umoC%h1x(eo4dq9#495#tA)
z`tcF_ZO@4Gcn<NHfpc)v5-jS0Cp=WaHz2LzDK{xoNHVdOoC<>FKKY;!qR?3lojt}b
z!htI#23qGsP4KXf{m9@KqOMDXc@8$5*mQ?GqqKywY(AnrL__f$DU@E-tPf5KTd2X7
z;}T3BXe7VJA0my~&kMgD++s#;rK>gc6<_WfIF-((rmwDw9$ot#cz=9<dFe5G>{SGP
z%VE1<!?>DI9bqD?ZEfn&e3KYXLeBetXHumSNkhN>bza?@k8jB^6O4rl5xa(245;TY
z*(T%=IqQBsU(-W+SINwrgTUG9<)7E$5@!&VZ_4ZWXi}6;Dwf!g%QM*>2?CWcrCSTS
zyP=lRRc}23_cfbGgCv=PTV09f=jTt2Y8ZD{Td-<(eT#nN%iuw5oI$<&T~|KcM=lg+
zdlj_KCgL0BE(-sQQ_C(<SHkC>0Cmft#CmS<`Xdgqk>0ttjM&ju74`HNO_y(`YeP_y
z)$qXjk@VsL`YomcJKoWN;hBu4H@^7iptR%uu*yi=X~#rURtxY*^=AjkT}=*wJZ&^^
zk+=P!%H@23^GVPl=NeD{b43Vm9O5tJiueq{@^=+|Nq>QJbei!$Du4fIN|wsEp<u@|
zjQ73SWq^;Iul@C_n|Cr0kzgUoP3o(vn)yfG%0b-*E>s|~>|Ut(b(!dk`QrBNT-oC=
zdg9D#+sOBC*s)QA0jYCMn0Ax77JQWMkAnbx`CG32!E04rT_TNo!*}$(-KzBcY(hYW
z_)!oqEZRQLL8`IXlfI?8EBwbE<$$J=j^1oa)C~T3lM&sBh{r!c?^}UGZ!A#D-qbHS
z_MB2SrI{im;As8#bnC`_p~|Xix8p%ozg2*uDz@;XPkkcP%kZC&>>48(804j!q?Lb#
z&s3uQuaAYmz^$hJhb~M<MmtvCFFL_$MxQqdktN<$%+B%hGs8gA+Vtv_qTYv76BEI)
zts~N84WD<Z1x0*>#a7u34erQVU(Vq}QQk>?r<6cmm7c6T_pVVE#LN-NNfXSp6JpCd
z<QiePI=$%@x(I^Z{0k)fMrvPNF`gxVDDlFtEv8&-MNc;q?n>}(;XMUyxW%~1$b39m
zs0$)fP63T9ba0WUblN$88}qsqGyGXqPX2K=lcMORXoS3k>Bro$KH?xL5?-HD=^NDI
zd2De`O-$5K#5+ZgaiTiikDBp~QOt9iEpDh)Dxy=QlSU>tue35o%+1nG@k=o;E?^q1
z9UYWB5ivxWVJcvFZ;0+my7>(0m~g}&NYVu3*(pJyB<+Qi9kY#03pn6GT}&S$+=|M2
z_3@BC=m-%i0Z`g;Sn6M&W4yjP>DZK1Zp1IRT<+Gy8KlRSiWADq-ZgtS6SC}lEm@a=
z;LwP9o_$G1`hCs-$l9;Nc#Len85LMs^GmekK_aPJbZ)3!%r{9~{OO++^qSP#+Bzjm
z$Xb6MvYSTWaBU8WYhq(#Z>VkTU+WR;+e|uMIH7JTozvM##d$=gkN1O|y{~ybpwE3i
zusi1&kG@>nRFTci6)J>{!}80NN`DWMo%ZR%*|9UQ-ch}dbZy-RznXC5l(_|Q1CnW_
zE;KCS7{}ainorMu>K!QH9Y@uX=^3aPxcx3Eq)D&~FaT%E{#ZE46LQlMyEy;&%ZT<%
z2Prpd{Jl`c@dsG0(VIQHJ}y&m<(^oKTpzbpR>D5%v%_QO9pIK^G7_KE{W#%PQI3;H
zo92a<?7*59l7k?wxlC#bjRm<fP=344RXWAJgGOP-+F#KLS`LRR+`9hiAom!N!qp7B
zK${%lh0~dTn2L^otnA%LregaR0vk64F|qH{f@Hbckr-tXo7vsaPYp0ABP(4shCaoY
z>X0jV0>+{`nXi<|azHj=qnhM2yf5;?`o~g*h_%1!VZF;jZV__^nn`5x<)7>gL#C?4
z1{_dmqj$o4j{Y~}s6x91%=KN=%*sVrWrXtjgG+~3JY%&njT+yZ7C`SU@C)PN{X+HX
z?mba8^)}13=zAV2B)2UDUronQ9f|ivjL5ENW>xph?r%z{W!mrD<0<Di2j!`k{MJ$R
znKJkeP!#K0T&SJLNSN=T?Gwr>QG&DDCpp|)l=-hCkb&&aPO4bE;iB-ObCbTf$Cuu#
z1FT4U_GrtW9{UUx$B7XP*GD13$1FUc^j^__@EFrA{emn5x+^ry+`x7ek6c?KBL;$@
zA$8u3em68N^D&9)584&)rOL}IVvf4{%-29h1?iZJ?dpe=aZy<~+q4mMC5!xjJN&b*
zm0A0vcm8@D$ASIvJ5oA~+VH7_$avg%#+f`g4i+D3n@7l$D+ek?1pYy?qlr+k26r$X
z<rO}cO@Mz)%+;oH4{Stj-Y|^xVTH$@6&LE}$>(co5~DdOM@P{nAPE|V#C^;{mUjLO
zW@ZfB%7j0SazJk+THx9Z30TSU>R9dci1Z+;9)}fsVOsUB@1jx+8Cr<=7zq~9c!rH;
zBvkM;Fw8FCE2M42BPZ+in?xqywrfFQ+WOk7SR{kP`oakB0exbF%&BH5p#gA+j>kNn
z1E@Gm&h6mgESL77Qt^9ZI0csq4Cn`cP*n~l=!X*5=6ZRy@qnaokeqyUEvYf=x<cG?
zYQ1pSpPb^P;b-HhtE<91IGsmYE9@<m>Sxc)7E4)t*}`7kycXdXy0YXFF47x8ph!XE
z*H+sNu3}JcMw-{&9&R)EB#F*zxr+a^^N~SVn2e3n3}qh^FB43jhF6SD;JXU_QZVes
zpL9il%^Xp4m;boRp_l7J4%t4Ls4(vEN{!UDP>T@W&t@dEd0U56+H-6H@#*S#bc3@{
zwgI;6ucPcsAmcg9-x-kxKL+gKs%K`>$HK7_XE}dvTNk76io`dFE*-w9j5XPeYVWBY
zTi@TD#*tv)FI1nLB@=P1hR=mo@C_Wft~FuyHJ?*XCR#M%qafP|+D`oe%pvRvQS2@6
zbKj@82M_nmWB!YTS$;-*KfbW12w|AP0lKzIF?{rU+&(*T!1>em7$<PU#2CqUv}xC3
zgkv|!ruCEZ*@E!bKw+3LxVrA=%Xs4=#qjx{_5j1JxfcpkV%_p5id6+!fQT-s48GE?
zDVL+q4|9qTI;C?SVB|bWl0~9^w}VmOzs6$&ot0{|3abAKuU<!;H^Kr@o)i$73Wz(K
z@9}k8`}G08_*0MnYzlu<Bka4t=FJ2}%!-!suG5O7CcX-?zZ%#XzQIXtBVWZRLA^*d
zf;Q{wF#guQS!+)i-A8X<*u_M)Jl(ruz?#o`MU*bVakc=$miHZ6H0*vcP3g;B!p588
zojN1kB5n!Z=h!|Esplh3ZD?zYjZt!Ff*N+n*8JK3*RsB`l=vj&g;Tye?fu|{X5~>5
z`z1&4BuJCdHpUlg5S3vYRd4-|br<?EI`^P@ZIF@#J(Np&)X`5jJugrG6K_k&;hw}V
zO!6F1W<Vxdr&;Suj<n|rikck074OIz8F=ZZXU1xaOMW3C3n*m#@H3^xPJ%7FjNG#C
zV5-OgEV4-fhQ9TKr3+0GR4$r>SO_RNO(xgIVETnzR0SP;qM3Ct4nFlNO5w|#)WAgE
zk&it1sYYIm@B2juEp^vZp1zsbViZ-gB3hE`1dI6ZTFE#P@Q*yO?_d~EBt7)mbEbg!
z+UGF`>~m8nZl=m{OKfBrhc(TTsVJCVYatIZ_+m5R6SYlT)2qRDnvW>AjS8+g<)Re2
zGLE|xgsJ9G4nQ@4HiM`Dd;Y9BF1?8PcaZi4eY1GDAA1^`y9kGw=)BwH7Hg@roEvj`
z)t8(5l=)xzq;)W(NN!7xAb34zIXi3PUqO1W6hB!l@Z`Y3-~PoPZ?Mk;Gv5-J$Y~*r
zGnsFUB9Jqkimk4tC0du1{UbvQO2zQd7tK>@Pa78Ecn{1rc_$ZRS@R5QJa2FB#Z~r$
z@dd?{ZeKwWQn!E0!fb3gvNT+ST7<H)2vaDV$x%rkn=M4b`nKbGVyDPyD~rcHma7Ci
zuLjKCsO1N<4M(#_me;x9^I+@!vye=}@5+_u#Jm>Uun*bX7OEO^`H`*)^wS)`t`uy<
zpXHUYSsa^ujekC;&{>F3uD1le3py3xEl|LfbDZV!9W|<<IGd-H4!Rrt>~Ky4TDG1N
zSn2dt5IcXxti)W`Qy@`0SMNJI60woZvjE5HvRZaFMAa<#9#*xXq|G=tF?~T`SVh?S
zp(tH!Jtx|{9kynkSU|3oj^kG?ySZtfRM*zl=fhXR)<ZmWVoF3nH0zL}CYm3tb2cyR
za&aXe#H>MRy%Fb&m#czp;l$tPU}j?p@MO_l6Q^YApr#^f_xH-nson@+EIy+jjMU}a
z0801GCZO*Ty?oGD{m6DpS=sp-_xK;U%R5Lt?6ww>rVfvph<SI@65Ky0;P2C*9Kt+s
z0>g<!4qX1nWkAt8=IK1zabMR{GADWA8up#cv{W}nt#rboPX5+4&$dIdM&P?RX+z_R
z?Q?J6IQ<9b^`J!4(<sBL(4?3GwhR@X@@C~5u=j}f2{m-<yRfxmDKblskz@zSc6NTo
zJXZ<b7>{4aJ2EnG`cCRb-8NG!&EKHZ@~v2k7G5IXP;LvQRwBq|lMd0Kpy|Ur_@481
z?KKucD|YP(VfKzr)d&ZBKL}TpQ><-P6%KopjFtE0&VEM+7!ORRp9o^?<~$PY_hIWM
zO54Xr0m%Nk-=={NO;1CYYA}CVdrp<EQ}ButWlJOu+}ApfeG<}jU&iFDu*e>d`X!~n
zvH;Iaf31s<O2s9#Li?F-*w5r*Bz^BmkW#Q=-0z==%lfK$jW~66OVoQ6r9QgJ^ImNh
z;=g~V{<MD3O8(G>;onAW)&BdSF}wjU#Sq<5Nrk(J__Y4*pqKorsebI^NMehZ9ih)^
zSv94l1<h^eiyOp?DUfiCsL5oc>j@q3?Y!Trc<K?qlV=J{ZBP$Db9B~25b9|om(f?-
zD=V<x<|;uv)p)y77e~AC2}bl-D84G@abeujR^3DH{j$P4viGuw1bdEaR&BRW-X)5x
z{J>aoLlh_Iaa%ZW;^Kx(=BzoQSUE$iro?HeJVA{|Y~!3^wJ6&4UbhS&{ArD2j4!n9
zsh$oL46+HXVeVQyLZn=jiA-bMt1s{3?|mP_0m5y_PEIGuMP5Z3d}Pd0{XWQrn^E;w
zZ(>&G?F`k~lYB5Gjx5)N4E704Zut|{njP-Y;!zh){>rhFr0EWmiJe8zJ7xk!cG)=x
zPcUL}a#5KKsj`9JjrR!`C03H641Otu9htDIPY%J+$p%qsM0gwy4=9x?*P+RnU^ys4
zTnbyvuGL=nZWD(uW_}&(;&J`)oh_ECK3sEK;P7V88!Q*nu5Bl(7!cKu$8O+h0&`fm
zm*W+r75xTHvA(6zbcD>U#+?k;HsaTLZ*miEF@T-BWNW|*hK>Me2@@KFFCn>wi{Q$V
zU1LLpUP^eija5rObA&*cAA0sjK0FHVHe34VTni7Y!rmhxodgkQAx}e^WZ-GtTvrB<
zc@rT(o3;aEwR--%gQ#nA{oel-${-m0cLdSwYk#zK1=rf=%!oeJYz%#qJwZN-i=nz@
zU43?S%j@U$|4XB<(B3zWW=0T{K#ixf<T|?SqcWu}h$ulqq@4133*~<A=10O1@^1p2
zX#3US-hc_?-Biw225jbBq0GR<#LCr&>{#gr%X?{Yv1Hi85HxD+nA^l_XCMq_u3WCm
z^{a}H;(PAFSu-3B=!M<90iTwowbdg3JC4Fo8)L|9HLsx14Y4*ZeHcC}Ccl6S_sCFx
z6TX?6a7mapp6xm$E`is=CgvK`ksMQ!txtsJRCJqWyY|m+#=h9N|7m;A?RtCPlHk}y
zSb2$CB<QaA{icj7ZYAR+PswqDQ_~Mal+bz(D2B1ZHL0U3?DY@9T_pDDXi^3awG8Ac
zbEJ0igiJuAAQXx!1^I_$t2On9YI`!$jT-jWPap}UR+33WLYM*ZQJ|s97E1b$9u;=P
zl%KbqM$CgF*KR0`mkLeW`ZK(9_BpCAHs$9@tlz^>h2jOLxc#3ID&5{HGpZ(7R~FYS
z!7ksj3g6lCbf<G`6=~dH*Y(^g7bT$jbFj;|Eu`<@9mNYZyFw#mqu@N}mmJ0se$JT3
z<-2~Z_tXd2GpUk-$X)fnt_U8~->k?lb~!9|JhNPvl+4?Qd+G~Ij0$GsDV#E&nSUSm
zfI{@g&R`}?pO3UQ%R9=|Vrsn)5#8?X()^S!mL|nVCme-Eo;M@!AN1CDVDW*bf%g=!
z>O4R~p_N8?J@MM3|CLhOp%fQugbIKS3J~!Xv;Ybf@F8%Rwkwd!lb@WN?NxzW-S4@V
z|6%r`Vq$qK{6@<0w)#raO*B5)l%ax{x*4+h!jbAE1>O%S3@5k#>F>V_SCqdL6HOD9
z8dY4(-|@mRzhjTuYXtp~ZNCpEPOG7g+`~eUtuM~rWi(_hVtYM~8*hk5(ifQ<l&Oqm
zww3np^8=o=F%~_A9XqnW@c$t2#?GLy5X#|;-;&HFna>}|c32VwTxMsWtuXRc)7#E#
z^bO-sW-ACHyCqcpAkp3Qd72^Gg!*Rl2|b@U5wYzLEr%$6)GENA_Pjx~YzS#xuULtt
zGWVid>SmhOX8dSWXuI#Ww%YvFJS#q)QWo=H<A||i^4#2IWwt-0V7N!hFl$H_XM0m@
zO<yIquFLD&zg04))6dhwi`KKe7|O9T@MgczD&A`zDLNvtIXSP=kIBnilj2j_nG4}f
z`Nv*i?n=oQH4#crH4ua;L^wMq+UX}OoqGE|wA%fAfsq9~PMA=Fgmu4thic$||A>!N
z^l5Pz5MN51s2Q7}0Do)VH1sJ;7ZHA&YFt#TX^0BL|EW{RYt0wqJz6Nf1;5(jq<(=6
z>Rhg<D--j`pn<G8hKV~YqDELS`6;(g22SKuChegyWJq|-o)6WA^IeDOx07j{K_DUF
zS657L@GXv%c1_D$ftsIPhM!f!e6Jx2hSSJHt9-_rYAPq1`*8Nb^x@_9e(g3`0?cK5
zN=^H@pN#`8z_g&r+9E&VH>sU+Ez&85(jC`HerEkdbbd;5g6(B?N$Sr&${y@{vD+2&
zfSZQA-Q90pW9yHcFCLf=Eg6nsF?91tn0?kMydBp0(uY3j6~f7wk=voMougYKDBn>G
z;zYQLA#z1H42iBb2|9}UkF@Yj`b?5yBXzLkZ(r*%wh^FHya<Qqy!*~K?P7lqE`6lB
z%!jXbw#n^!e<r!Lj9I%%=tDadK<`~nuFhrH_%j>HKE81{d`BMH2BJhWJIx!dk^Wb|
zOqw48O*^kK)}N=1R{So>-Y%K?HMlfyg=N$=6>JBo8DnW#=AXDO=v-eE?z$5_=J-(^
znN>I7OBCZYcz}gz+L^IjFa78v+mm8_MB=X?(||*ppxf7*yi_bRS~bs&`PN#FG5k~#
ztj|I_0tqxU6no05w03}gr;uXsr^s(z!4!!Nl!?Y-eWk%CplY)}Tjt=nmax-~EB?Hv
zD_%Bs@A5i~xhVrddU9T}p8V_c<ytCn@FqCblaO%uf_qI66HDvOpY61W(T<rYvV#w}
z-wq>hZ+>hZj7Jj#vB+;>Ghb&@gw{w0x4EC!-fb%m&K)oKZ09((Kf^wv9WO4C;my$$
z^lKBomtslzK?qCp0BT!~!n;t+k1cs8)nY2Ic;wB_N>(|rr;b3@p9fnOSC)6nhNwzb
z<A4pnlyuYu$uaF)%9HtFya(Xcz-^XT$G3IpYN;~d?d|OWO$80zLdt!E^Ob=CA_=sa
zSgw_d-AAauSVYlqBI{E}qM6`4p8Q!Qzs9@4^z=Ww>gqUNl*ruYh0VPctnw})Hjkc=
z$`ytcxKg9}J#(xU!aJH<Sm5j<3i~5V$aJg$)Nr?8LYh{&gS`<OaDTF!Jk&21YbVWD
zcF(c8hQOV#XzPLZCYX<cZIASB%Juld=^+Vkl_n`XXvftM=5Er)+C_&;#ah-A;)+6)
zmA4w4<1;l~{pZW}iwota@UF{X^S0o~dxl})*=2Twllyouep>h&MdX2{d#PRfAc?q9
zdUF?Up18jL(E@QhEKUOArM1Y_HfWJVL*7Vt9kAD+?r~p-4tPe@H>w)Y3`Jos$|HGL
zBF`|G#m2~(cPL3Zw|~1K0ZOtmno>o_e55$a%B((|2jewwX`96HZ?fVe#)CW?Th;GO
z^l+6+Cz&q);g=pUQ#raOLLyj>oygyrytw;ifi)Ws0a_&*Qq(wRW~lZ3UqR9)BpCkU
z4C1y#6SZ-7<Jos8nvG0NdE6%r2au?Zrlpc{sc2#hBB%pXk!gD;gXW@y)bfZVv7uas
zE`vi33-64+U<(bU7vAU1PGhb+kd8X};%<=u1=yblae^wtv^Uvj;f^(<_n+jeYcHm7
zWeo}fHhwAOYY3J9G{>weWi8xM84CS|6DLV7Ki5vjG)U4{$X*naTpz7-DWe#!Ye%w=
zNJONUk!7@D4>WIS8=GI_z^ciy@dDwvTK7?s2|pUI9@9WerXA`!n_>26<Ocj(Cl%}F
zcWV?oc<2ZEmS;fF|0C+0;^S(dcHhRfZ5z$Rw$T`k)!4Ren++N_)<lhyiEZ1B&%EDv
ze&^iH&D`y^*Iw&+p8v9%Vg=PKuuCejkKPQczSJ9wBB*??5TmMwUeYzq%XuHY%~dr_
zToA<Nv(rc|1YspZ--bj=b;etgS|VPHkNL87UQfY-x-^Y{C$Yi6g>C^5*5ewnmVrbW
zy_f3g8AqA>;vm<&u%T4MjKlVqm(YtiA%`yixPsSq%cNV+=@MRc1OHPECHRrKiW{Ok
z`|{i_i;|($RX*5Gp3z~-$4B?&<>fvsUIdvrzS?JKn_6B914>z&c|zjd=Kj*$P~OKs
zW*j7-pUg8C4`fwE#i3FLoS9}sU7l%q-LlI?tBJ0ilr^Mx=PMO`z*0Q*^dMAvNv?6e
zFsk{e@8ar65TTlItZXv*ZMSS%EW*H6!#A9x>AanW@dS3fN6!P)o+-ez^>7iCrJZbR
zC$?ukA7D@teN1?6IC4*T0w6!9rjI3=E9jHXv*~-+^FSyuQLkI5N)<S$9xj!hXhQkx
z2>k4#-W{8seXS0wuII}(x2lM_5^wk8v%-U*<#g>~Bo0%nafuD5=GQ|hBL`6@fw)fN
z4~8VW)F<3n5}S3+_P!2mek=J9$}2<UI$GT(HveYk**yKRBcxO+(Zb7;%P;?J^~fsu
zX2F5T*<%TvdPb=awJ5RP(LPFQbub6MhwFzRir|2a8$>m9=miwE<|{(Axei7~ypj((
zwkW95)RsPCO(hnRaS`ttrLm^+G9<eJ>qAkSIjqLQQ%I)fzcvZ=KW$PCFLt3&e1$yO
z4r)!iiagLzMVZ>@ZI{#c5!aARjxxHKD!-gQizNHnRzZ})%&rPsP+?6P;tif0ouNpx
z;ON>anA0?Sfv-rar2iI#0rBrcp4HjKG>4zi!}GR5y3O|nLRYLiO9hLX(4(%&>xd;p
zRmlSBL_g-h1s6{VX1&WTMzdP9)yO1Hsufz&Ix`b~y1)m@lE_@>mu3ZAXr!1{3R<DS
zNRn*};T&S_b?Vf1?FgQH4W*{mQX+aTf2WdV&*Y<<?2bT`W)YDkQOZvRox@|gYhhq!
zV!ey<_ze9ioIZ<f(cpJ+TkIFn8hA|VWhe@rmd#c)C3U_KAjQvZ4S}~&M*Zi1<{C_+
z(ZQ*OrqorE*L)ND>fg1pJP^*RzxEKhxCEzE%rF$K{8Wi;y_qUZxs$jda#}5Is{gWi
zr^ng@jJpv3<6-Y|38ED6enk4^$mD91DmlfdRd=(RCk~?%<w_mbPB3mePNc-)?yx~_
z&z$`9bW<j^NSVv2y>vA3BH38;L(O7eQpcQrPsf@CB;<Ic(YBlf%LpYU!=Yt@tbiad
z4wrAJL#BkKhr1-RMl7{tt8E#z4*`5~go&0_(oDB1_`%2)Z4$JYE}?aIm3TacCgac`
zwQFug)n?>%m1o!UP)j;%^Jk^Yr*BW$@W<U@;^i#)RgYhevhObbuD0ez=F{bizzQ$V
z<nH8DfcbK8(!GRJUQ3HSQb!J!`S6bFsMUN06SJ(YX)^ZDtMj>^?re*VtpO2KE2hTQ
zak_#-J|kvaKOB5g21>*w^h>yV`%0Ak;aGPLFV1oej7rzmOp^Oon(T7($Xo|a69l8V
zjR(V9unTHWZF1bM5CWzd3cd_#9BHdn^XLQ|t2&ZIQ4^w`xJ^4AB#n6O9gsQ!mdkta
zia%78Iwb3SWQb_3h}nryss)oKIN*&HvogPax@P?Lh+l;yvMn}`r?>g1v71JQBM%oq
z%!ts{3dd?)8qex~LN4_BToDE8vyG{EwXXIdawViJ6Y<nBcH<uT4Lrqf)t6D>1`Q*?
zN^c``5vtT75=>^69Y90nYZ<;k{<gbx^0{Ln^}eUjV#%jWkJr89_)WLc^U9*vdIcgS
z3|~Zd%Vx?A$ZYaTcw1!Ftwoo9N3w--o=PG7TeFwdlt3=0%s$FcgA%i4$#^yI#2HAz
zb>|7C<n2t_m{N|U?Pu}KMA4ATuW$N$?yyge<H|~_x?fH_YyqF6R!jRrB}u$0dD%PG
z10bL~RzHPst+|~iCLA~`@n45EPT10_u=g*T+H{AZ3i4&ZOpSH?4Mn7C_V0dLc*-(h
z$4mF3-o2a)dQ(1CL&i%#q=%O85lzl<q7l6a0>h91V{yvJ!rwT`_%rf@j<Pj=p~W1h
zdgxLW!|AzU1i;6){kC$4E!$Yggg6``#A9!0DUeH6Ow;?!-#dUw>K$$9QP9Nb(oCSJ
zI~fUpxMGfpVZ48dIIiqRWc5W-Pd9%5z7Rzh#($VO(GeUUKpMz)&+)}D-TbY(&Oh$s
z$HYi}K(ve`m?kiTJ|HVvLO-wwjp&unNs<6I9{Cgt*Sf)>Ru^F7)JDjz>Ymkid9K<9
zkR~^q+~$8(3)PUxOKzs{)g+@q;dS7Bijr=-y^*%!R;zi!`qPK`m}Y~oDWxd?n-gd!
zsWcK14liNFhNf27)yJr<8xUFg43PXInsv3CThZd0weXB%O+`;RIgpTBvbMH2+k!0x
zrm!G+u^WFDj(a#<uoGZ&6{kA9_g^*oKGTQ4leml{mQLcR3QBQ98OF)#DJNR$(QWKR
zSsMNj7{x!rUkrcxWzJnA9#u~csiHRNUjY}^uO(sKzvBqpdab=;BSLzfKvXW&kXStx
z!BRhjbk#+j&2Pn$9G>+As&(?tIT2YW0}$qp;D7A<wm1#w?$RNAzJG}o)uG$;yr7A3
zvGskgcG!yr?!$;q<T!whHKqm9q=E_A%zZf6<l11JHQJWj?y61>Qv<OlEYp1e=*YFR
zU5LvUmN!Ts+Pz5@kCA&f(Sn<WmCm!}g-nf<=S{2^t0U;~wbsPl{mS`a!{hjow(DhB
z?=KZaZ3UBu%>?%v6Z5x7(*zglmS(BeoY@W1#(GDXGN>tKcG}putd5c$-UNPk;t$VD
ztPxYXKOT?$KIs`5%PR<Fc%dSL=<kpTSy}+5VFW3#!(=B=syZ$;NE{cZpg=p4ML?q@
z0I2bTaF)6aY<KxDBsalo^Y{%p7^_rlly6Mhvw*UG&}rCG$R|nov7Oes#a0_NENlBc
z)$zk&;B>p~&Q!bB*D-M_Hz0Qk2;QnI9LEV_>fH`{TZS+2&FcG~b4W|W>H~c8DW=)W
z`_g$=3bl;gg5vnzhOd0n7^elKdOOF%4hSx!QZ(-pwSBIGhsr|jg{7t8S-Y;2NC#}<
zuvo4KLfd%8^?B#f%2~trnE!+FoIia8iVE*xP8*S@qk~-|ZFI#MM_f!J+Jd9_mZk)g
z-V4b5LpI!Y$wTl#PZ)Y>a4n0=u%Qe8zhnhmwYAVx5zXn+$YjwB(WI||h`7(d459vY
zeZ(@eOfAg||I+9{<({_`PRID6X$A#y6C41w=l)oP*V};b$Wp))4;?LaLu0A6Shu02
zmseI=OOq%1-TYnx`{k1gP`!Gg-dy(EIoSCGD$%Os)|H_K&#^hJTn`enHs7@MeG>><
z2lj91<N(XlC$2a;y87Y_L9Qbx2nzxPM?{-hI=^gQCgx;>CAgEddH&7Q=36>EMLGn$
z@(VotGsnR>JYiy<A+>L?^5w?D9&0S!x|-9^EMWeQ!WcI{VFhe1a_G<lQ64yjIt!aL
zR>`Nph}o1Xsn85PLD|SU9%i`+-29@=wEY&@T<Dq)`Ywr2Zqd|kmgWcG{eU3vYiCm`
zzEvzuwSv+_B=J7Cz7%+N+Wfc_9!aAtJx42p&$9rkA02wV>~QBQJWpVvHgtMO6bs15
z!Koh-J;Epu`5uD^5GVQMe7|Otu6K1RCR?38q;C9Yj<6L$#va=D;pft0XOw4@AfTvL
z9fPf}cKw+<z#EbtOog#6RxU{&mHhpXdXejUNQ9ARZYQi@&r95u%kvnujf)Fa9{1Qe
zGe}dP_Ya9W&iSMCL>-G-whBRPb1{zfwT;vd9F$x&2|l29LL;iwh#Z~!qx=6X`5mp2
zSR^qz?xaF;^?3ex{R2JQBp3s7f+7VjGWidq*PFb^+D@zl--(9wkC3FL@l$GIG|>_B
zpd;n^?ubLNx?XPO*H>2eCOx(eK+kUZ{^#t;fZ`No$FaulZp9O#43chz05y}m5>;hh
zf#q%B%H6#b<fiWtjie46QM~p1DiUmXlWhW7sz`}tydEIp3yB-hI>@oi(sJRsYJS)H
z8-Yw&q9=GN%`7-0OD&$wGCf0yz1yzTIRmV3nJ!<c$zwLpFE@&)+5IAOoFOt^SZ5z5
z<{!<QBBFhA*i49hML3h9BQUP}U0GrU5iP+d^wI#`bA51>dB*kY+b{CGHdvoe18wH=
zH*qnvVu@vxyO}iHwznOlZB+j&k9t#a3Nl4d;DY=(Q(eJGG=Z{FyTayGaTs~+LVY<E
zse8ks^UdstETKy!zNR>5n0wRIEy&>a7Rr2>HZ9R5_QWJ}eX}FZueV<4YG$=ta8RyH
zLOv)Joulf`g)|_1V}oStW6{I&B}$%Y!Xr~J|ERKeGMbtzWQA?hYUWB^Pd8H)%R>7?
z>GuYLhctbdcc8Z4|K?cV59Yn_`9yQc^A$UkgMwph5q&RRSlM`%YLL3b<OrpB74+JM
zQjZPFE-PSm2Y0ebzBxfWlqOPHhp>J|8CB2`)djp(m#)rOx5&DRH7%m7!&W=Cz}PsY
z8Gi=K3IQ{yH`654)MHE!*W3&}KL)TQN8fx?C`g+Enw2XyDB?nP{z%4S&>OwI6%-4$
zhCYNx?xm6TWapDg0dm=pRB96uRL#gbY;_UknSR&QcQh+xB_smx_V<)kV`;%*>#f3i
zlnY<&?tSKgov(}eLmIU!BeSsI-HQnRSPd}IFeG0>_Q7BCU32K;18@PMa_AHlemw&R
z%KAkMa*5cyT8w(!z%c47xR{Z!%1E2g5O%Vv7)@mFR9ZKtf#E1RyPFfEH*C4Rw)c=N
zDOu(2*<#|Mn&v~<H5V+YhSl7?+f^NJ*tPo+L4D2bv0xdv!<0yQMEWz*ZUqOmM!t-h
zX@tjcahqCd4JOqp90K7t2-44q0#jrEu=RuaNXuDm&kvi#gJ!Iq0Y0W9$g97dj`pU+
zPX6r!y4_O;X<puvqB}&48LX=Dv_Iea-EtABS$!k_J;F%qJ0H<iKS<6w5vV~4<<-G_
zAyhZ16^*QCI2SzzBq75{Caih1MoFe<NUCtxc(N{UY*a8ItA#bddqyQY{WIBN@K38E
z>i*S9ar3?1R<j&EdEp^K^jqJN!fkL~i1^`A&OD);O$#3`tHJ(ynY7pA^O4hYrhAEO
zCpJ=-D~bbi-V}u8R83<?<DeObF&Dw?#U$oXcG$Ze-(>1XpNbMk=#%fsctCE;10uk9
z0Q6j7QeO^Z+)*ySoghD|rlDglolLc`3>!1Ao}Wb>%@uYN%ONix!8GH^Rh)~3^&BcA
zjTByDQySbe5|sH3HIJ%9p59U?9(HFQ9+$j!usv5s779leQ+9a3lE8PzMVWo+7rR|#
zGKkD9Q->xEYQo9Ojc?Jo!%HMgbT4wx=lm-xgq;YrozT<xCwvOV@1B31T(14T)k=!g
zT(-m-B|#1BrO1Oo0?B}g;$zCYnHns4Ny_MxW_6**x4Hc=W2YXeB-E9&i|{M#r%K5m
zF%V(9hy}0e%|}x|{GLs<FS5Q>Dl{pgy#-zqYWnbZi^wY48fQH(aOUdh0~KKCti5h;
zP>rQ{S^XY(M&l^|eD6Cz9+*Xue<XS&slka{5k$#RDA)G$O}Oe+-JK^<c6+;O{U66Z
zV?y2atp|!zxdaU|6bU`mFFF7#Y%{x+gXIA(T<<!2f!WxkQ@>YbY1Q=v%FNA(Bs3V=
z!nUNdul%`UWZ6RORR2t3%V-G<oCgIaM5cgU`jH;cQ|U50j*{`*IAMJX!&XZ)@f0>r
z(aFJ+7oDkLqMlK=K|Ic$^Lt?l0l}>P2N{vYSUb*Lu1|OoC!fN+4u93lZZ}e;h~QQh
zdF{7WidE~JZh+ul9YG$!G9zX%G|*hB105=@qzOJ)l0j%iYcqJ@%<06GF@oAhzBo3^
z(L3Z2NSxVy698O;KyQUOww5hL)UA?7mfvSq)2?JmucA2{{}m6<*nHj5;iy5aRpq?V
zec}`$LLSvp-cUct`Y~2;Nca$1a-qk>^qVZVVgflTMmB{e?22+7%7JDO@H1UC9fcX{
z*;8zC-kO>nv4!AQ;MK+oEo(K%K@zV3DZ%r+UHR^r_1$T=IldXKdImqNuKWay<}-d=
z)X0THBpI`bsKkku@RozBd*oQO$+jfz?(^94Vg-k{LK|RADVis4dZI?0xQ;pg4jkiI
z5~mcKsS(Jz)W~bSGncC#LxLJo;k9`wHGG(Y0yLbT#|TKhcA~apQ#H!O{fad>4>pPJ
zs>`=gN1n8z*)WVxCcK1RMp<t+9NuZ(Hz=uiFD&BdqOHCWaxE<;sYWvkvP?&s4SM=U
zx}w-J#p%k`@lf2xHz$GYx#{`uxCrRiI`ybj#h7K6h4I+%(gmKcF@2vCJ~{Xni4qRl
zp7J)DrW;>KmB9+G4&t7ioP*dBY0-IYNo2F8xUFK-N*~!tyubr58hCL_%09|5)+oZj
zO?X+Qaa^8b!7V6M!|sei`#wR}tfy$xvRc!y=MT=tAcM2|LYoqQOR?oNWcXq|#~d2Y
zIWV-)WUf{JHdUo)3l8xEiwfv7F}3~P?n2SpeyP)9yST`|jTJl>T)(H6wBEp+jave>
z3xUO_@T#C;Gsx2MzC${sef&*r3f7wx@+iCH(@gJFjHvJ5+!3NFm{OO5HV6Avr=*2G
zWP7gDjKrkfLJ2<!>-o%fEi6T#Y97^V9uY``+#Fx0P~yDD#Brybj~=-llj-E}2u9oB
zDCKhzo5qLt1_~`tlz%+9lG&T13vkj-KmSs9{CxT^gZjT<BlA7D9a|NI$bpji_0T4a
z;tgVVL%+D8GII%n5Z46+QEqufakuB{EGgRVPt1aAOjv8(sX4WVde`)6{{t|t#9-i%
zGS>y*(8sr{IZ-NH)Y%y|^8+zqCS7~54OHLTGz5Ahiqi~zz+vt?gig81{0V?S)j^pS
zxMk}7$oSxB(Zt*leX+us9k}_5!0pHT`e~edoZx{HrQ&V15bUyz<V4g+0y8q`)u^Am
z=(}y;Nq3t|F?}`45{fdLVt2GFp<}(8=JM%VZ9)`O3rj4G&_y=toNb31L3qE}HSAmw
zCu5R`bMme*lAjof2^AdR3Hd$h#LU9v+v?zSd4?_hteomxcKZV&m?>&h-su%5kfTBn
z;{pxTHX~4tA36WIPJ@#!@C0XS1qWTWw<=mNl0i^2Pk~@rDa$sTtc);r1J^Z7(7zH<
zu&?zo@Jaumgm@vkhDEp{H@yJBB(+GFlrzpNsq~j77m>#vvl4QI<=1V<w6THkC76{l
z3B}b$&B~l%I8?Qcm)`Lk1mw(2-}+2x7%NznJ&l<Ds7aN<{{j>yBbKg;=ishk1!E7z
zW#8V~zXjZ82oJ2DG7K-PSS*rr2)Cc>y)6^xZ9)#-JQp~n@1{=sJx_0cpI8cbH4=4~
zpz7O)*0nlS%`4sP!;7lszV_<(+`l$QsbpjO!-R+0!IfOLZ6+&PMLAL0hh<(F=4HjD
z_1t3g$>^HC=qQA?!+Rr8mEFmmCkcyEByMS9wF#ePel)PDr(>Q<+&?<Y=jtv`9@Pv(
ze{z&Axu+`AIGcD=Ic?S$s1^-MO+Co;CWwH>_>NnH^<7h5F4eUuS-abiRuSXnsn+-}
zj0+rp#NQygen>mvZ3JHVPCu!w_pdQA<C(g#8PvtuvEWoUalVdY?stJ3_@)HqPrK-a
zB#psw+~<4N4?g#!h_=x};9nwRfo2Q|hn$-!=F-PDvE5Wk@*&*tkPGB=e&$oVu$y3+
zz;VQA)nT(la)DUd<O&l1eBq~z4+PojsM|u{mXy1#poQGO=`X+bSg5SClLsFXH=3cH
zu<ASd*dN)ZiVQ<Q;^*&3)z9!XhhU1zPtd5x6WQg>|1<q``W$@;6@6xt)zKa3pyRFP
zDvoWx*;vx?I)-2*^@1tmiWb4jlj`Mk{BB(JF`tC}U4yl1-u_`W%NUKK#(9<x2yUD=
zM3WN99qbz+#55`7x>n)ce&2rg%2#1cPGMw6TGSF-XnGx!5Z`V-R0}-*!=!3~@l+Pd
z0^7>L@rJNxK4riCxi{K?=K#@?ZjsEuy?%LM*W{j(uw{i#9;Q^L4Xy?Dhn}sYXE7bh
z8veBmHiga+)38n~7#e3yz1`K<F_V`PC#{Wn2ytT5Jm~r{$u-6|6&)%L-p|*)Z=lnu
zUpzx55oW8R=cYOVFKw^b0_@|;XjzM!rO0Q*ehj9HW8C9P5~xZB<g0+t)r!ovdHT3a
zaoeob!n-qD0K=7$X)TCMlHt4sPNpseCH;Xi=WHumhX)?^?6iVsF3dJmuDW+Jydp1x
z?%t;US1%|=RrI=(G+%U#>6&t77(r5T()rKO{_n!?(^gZ^#1YP!5{NUZSpRTE?2*h{
zCWOvgkzqEK3W!Aw(^IFCG*eEf(SFrWjkj+L>@TWYm=-UGohP}ZmK{_I_Ys7A0RbqI
zF<6;UjNBx{FO9|j>B$G6Zhq7hf6vqYgSTcN0CR-&@MjNK=#vSTRF^V(S7)gb<ejdQ
zMw@$+@`Z=FS8-fV-I-ggP(|V)ll2o+aBzN#Q0ITgQ(2L0Q7fv0mFgv`aaEX-=?fqy
zBq<In%*3N$wq9EMX+gN&;S>i@n_y)wN|mZ+<41+fn5yMi$~u~U8|np?N#|F}F{csg
zOHMRh8Yrcw&pREe{<T?12xF)Y_W)>7m_djYx=AqCqMv|`VBu=XHLkI7g+&73cWM`2
zIGz8u>M@Pjki-MP6bMjyPlufkNhWIop5#|H);b-Cx4H%U#&iG5=cY;g-Zu6YK%vX2
zzXK6Y2>S-D0$t#Ui!6?Gx4Wb#>vhAKrm(83`v*8{9!I7OzbEAxw<OWG49O2Bm06>n
zkpAQxg`w+>IJv0zxKKP*Q@N_aNWxg+gTn%fUL7Y^^K9-k(B@NAwV-K7{}dCNvZ9F|
zQW{Ld6&f;n8~>C)b&&@!ffO57KlvMU5gZjF7$cY|a3bO<RM3lcZ)g~6Sh|Yt*|J4w
z;a9xR1j=2(V#;sb0KDN0xUM!HA5QMauhI-cyYHv}^)E6oLxJD7ZP&=t@@Bw*Wm$ZY
zTN_Kv^EQNh$^VM5Su?{*z!g4KOcOnty`NHuhA02pqyllWntqX-(@~)dSB*TiD%~g~
zk-o<%-S~=SFKVaKXRv<hz|kk8*|4OgdBPL0TfZhveBRjFIdb@jBL}~GiJs$O-_+JP
z?8)_m?L2ucm{E)5NPOBqSlRzX3#`iRT-^L5NT}8=d9S(9V?gJtE=P0@&Aj<bR{8_+
z<VNlLU6+I;6Hd{>hF&ba6UuP=K$i9B^iFKy-w01a?RGV&MRDv~Rc7j;?{7%-&|^WG
zsK8pxz+s;l(({>mu)%m+Vp~4zGIBDbN7=6-&d>-MCS9a(QXVrG)xa&PH~~@{{R@YL
zSJf%icNzM~G%&m;mP<3tjYn+Z#|OImql0tim9v{e-(i1ZT=!R($<U^8I}~qB`$w03
z;{-ie$jeJ23W_03qnqxtSAjqJ^*os~;~a8&jW5S@`rd~nZ?~);6c=UGT{qj<e@+AD
zzFSa$C;-)}67<-MYdU{sy`moaPdGP&oVD=-^_}&n81n|aB6x6Cl-XzTwyM73TuFQG
z;iGn@u{+WUBz<YlZnS5#<UIE5$d?fg{O+hbGUy#2wIK~inCC6{n2-izzDHz|0Yscd
z^y6=B%?!bm5F^Bi#^G7HX}kUcA-Ei^J3TxE%tALSb&enB_<tW&JsSlE+93**>sRZ2
z?iegM%4hRwG2r98b2xYadjLG5A6wpe(j=W?CA?rYgDMB90eJH-g>-jfWyJF%OnI*i
zs8!}v*uhC|`!i2`BgG91`?G9JNYj_@in?lS)AI-j;Xx|I8f{ukM-gvY44@z+crjI8
z|B_{S-|?EuUfxj=d;>7`Dk4xjuh0>P1%|Ku5F^uXTT(5x4Ep;Xjg282`xI{F8bdqj
zmj6RR){#ZuVWQu`FR#fm9W6m9T!r95+|eRd^c|q7?8ah7CO^TZmv`)5x7oUghiwty
zxkBY&P;}fQpYPx}&u_L42AsFZgDO>x3~33GR<4i}rTNdV&Knz=!a!)GvVZKiO#Ba+
zRw}1|xQZOgk#yeTDGPOu-5GtLRx-0hG-qkGjp=B9YS#RI#Pq>#3x<jg%D~qQLnb6c
z-`p|pvpGXv!~z`QV{gk<{p{%@49mK|G0&7?s=yIJh^HiW1cuQd<6G<wJrz{)gQZ5a
z|B)Jjh^a0rD1b9}FNLQhxW`y&c-a%;0Yd)ux78T^=QQ4Z5a@hH!BSH3Q^o6SE+c>l
zpQys40=*X(<Ckgb?>17k-90>Pqd`)8$}~xG*|m$AY$9qdoE)aSPKmlT=$$AG+4MLW
z*9ofelLwR&R39WMgJkbK`2?ZqfsBHRhWcPPNf~mjK~*OpTiY1hgq-U4M=?z%cdg0N
z#&2d?GXy%wTarXqAvwvG-4Qi4d@|hyih?}m_=}AbW6%?g2qJ*4$4ofiH^VEvahbF0
z6cXg#5wUe?xPB`Pq&9G}$uy((>LYUXNH{3FK>Xh|OKeLKXJxb*l(UqI+xKt@1YVcp
zW`r#M)IP?zc#K``k_w#$E*d4$eh0haG<bBQ>f!chzEb#@2O3;fk2W7R5u2!ZsEsbN
zg=~9^HYY9<4}l_KeAVg-0lx!!A?E{nzVY$>--G|^s~7bWRFbV@yshmHJP|0UWkV%o
z#KOTn;Z0oI0jES%x*veaNB;d_*%SA$C{0r>USZqAE8+=yk(b6Rh<k1l%iaPK$e%=n
znpX#R{ypl&Us4uZuD&?_Hn@e;#dkMFS+%Yrtp`g^M@~KMFLhuw40VhTa^Lx{gwO1;
zdlbg9$kgFq1V^mGkR`+vJjRZ$eU8f|OwPmIBsm@qP9N)Dx3huW_NaNKKke-uXZR_i
z#mk?pJ`+LO!z+ohLWPS~w?()@BP(~xMakc$6vWacG&}~>c}Iydk}dGm&9kat3({4(
zp5V^<b#m!doFK25<g7uuazwNe3owb~Y*T{f3@VGs(}$MO2We%$HbwsiDmEcUUmJG8
z+Coi*-(H6y9zAc4*RA))=aD(%m_I!y2E-^kBf_ss7A9eNp4Sq4_r4>iA09bXSMR*t
zs@JcQnznSpRv!pzUr>AgQstAByg{lq$lXn*r9-_Wn}@rgzR{0xuLy~=9ULcSF>F)l
zT|@L8z-s&{Q)Sv%Bl3uPjGAzqfEr2Q#v<@y%_8_>om|F=m$7#sTZ6+1dr#ZoU3KJ`
zARb}LHjfZhS)Yw*3B|r4vEyjS#-^g-`pc(&HE-~AZxU%I{2!+FB;>&Vjadn&s9VPF
z#PMwY@bMuG<g}c+-cL~u|4ttOl|)A&Y_VZ?tD>QTnG6wikHZW_jbba%S=TeQ{z7o{
zxjZdOQ2bRdiL*lev?AP??{pV1Cj9Yfj9V~>7&hbdod8j;-aeZIchEZTSt$bLdU0IX
zY@O8QW@m(rJK`}q+@7;p9EInqv8i{Q^c84n&2>iPiYjWAt~qtLUeOz@!iRQhN9L^@
zRlMny8OIC9A+Lv+$)E}sggvMDm2BPVkRr2(;b(}B^%Uh_CamG2>ae*fov}tFD4T${
zO@)>t7S|bez3&{%utI1RQ;4Ocs;qykTI8Blw5foERK_$JoHKq$%<dT(yiW}k<nYwv
z!1Rg5N*V6)c}0F;xzrbiAJygadiXGBRe)j~ThJKqU%%yMG;7If`gT`~q!<#g*^JoT
z2U=Obqfbn+IJCIL_I#B+<ig-vhsX2nv|sW<dno?^lEmQ4DMIi4U4f*H;<RTy1ao;z
zKJ=z)ZE_VCbucC_V}CP55rZ^%lX}#z(N&w6CXlWYjzo1X5?;sI0KNE_Sd6)-#Vc>P
z&xul?87#p_W3GJy32{LV;$ynR`6dyFDl($Rs8N(iL0vv#I?jEl76&9#*%lw=MtdVx
z<tkfIc;*@D>acd{Pp~{GSH$KNXR~XLJan8)t?BKzY799hE*1p)Oq={ah;G)2hCNAi
zgr}_l<*Ic}U0s8POJz$J<i~T_4q7SCl)Hc(U<oNr6*t5aNbC0~SILQJAtOuPfN54O
z#NKkzfEGSsK(N3iP>w2-W)sROSEYI(swC6B6ATEVY-M8*kxmB#3*_`Ph=IYgo*0f$
zqy*?yN13(UViS&%a<V0cy$<gXbQQmzDf4fIxA9S|A@B*JBq=ku9hH*1Zz-iLm#qDt
zT{69+v64aArZqStW$pCHK`9+VK?_QBx$|h$m2_fS1t#5yamZ>I3j|!?C5=GWQ#WbS
z1ILu`8z4>t{WUW^*HSHhxu<iIANKX_4OdvQB&%`DjSK@PgY@8J`pvp4IxCe-Q5FCA
zw5Gkj=Dou7p`l-J__Rv@C_z3Bj1RK^HsJV^8XV2^!)|^Dk6}LQ+;5y0?n1+-r}hGo
zKE5^y0nzc#^n4$m4nA+v&ZoI9Z*jdhJpVcao~}S=50}ra?d`lDr$+D9Z<D>xJfAbd
zU3n7vZS%!mjFf*6uDBcOv8@nuAF#yG69ktG-&ks0Y?<`^A@J$DvH~CAYDFYC9rQFZ
zs8u+{Qujk8G!WC%=8fTgqZJw26PxR(r9%VtRM3_1a_{Hocu4{db=-M)wA&ohU)~7Z
z)rx5$6{wasJI9+){W8<!Gh`^jy1g(Hepo*f8Pa&#0+rgw*c6a4#V9N`zM)7&;)rpn
z6UQU;l}jjoUwCvInnoy{MDVPF>+7_(kV(X^te%0lHc-;xlhfPn5|;}B?l3eyu#cQ&
z%3Lw*KE2Cf&>v2JQ^_$+P3V)hG8W?%jjET>vC=<W{@0IcUBlAodqbpjW=dRw&VfSj
z>mW|c99%EP27u74+SZpy){rWCql7NDQTZF4EUQD2Cn<%lsZc-0X=JoLH!a+6yVMGJ
zn&IWk<5Zx528$en$&(@A9Ia2ZG^tKSo|R#TOW>2EzZ*8=o{CZv&R%(*caRVI{jUL#
zhf&&y#<90XT6~4O0%Oz@!kJ*wQ>jTu{~ta`jjkq}rr^6D<2b`FGd{a#03$Ul9MkB;
zyhQ2-M?Bdfc&8wqaxwR*%6dlEKYw<FBz^(R&PvNKh%aAq&U<drTnb*WOoXZkegIE%
zKnfnS(WkZ{Ny0Q!T4e<6#M=CE9~ET;Q7U|?{#5~UPO-m@EZEK|@!6KFk{9)2!^Q25
z0`eAG5SO#bG}*JF=ttr=;{5Elx`#48=1DER-icBOugFUa4Q(mi2q2!Km-fq*>4L>c
z00Il=zpO1ATo&l^Fq1l37qo*-G5&DxoWtnhc96N{7PinRs;8z-FuP~Qc5)%bN{PqP
zCvo6S3rr7W0gUOopNbl+<nG<8yj#IEbJB?~yU!Z8^O7Sat$%*$KvT`r!6}vY?c%2-
zDj~RkZ1WG|PA3c9APGg$jUgEeib?-FB|)@QnS#1)yB5(_JSc@Qn6$@2SCb49+9BTb
z1_$8j5K}n|DQob4l$GSOTpu@0H}yJ2ybRq*+~GQ5ii6KYD3PsnPI96OkG@f&rr1Pv
zx<MxVFn%VF{Yh>Y-%LFf(dj`X5$dM%wVD|YtiVrliT$dw6yoyXO}GsbGtPpsm;mU`
zoh#TwQ>rt!d!U?+q?T1QL!-nb^JWt$ktB@6FGQtiUsM-32RQ$)*p{$A|K8I3xb!dL
z^P2Rw^<!~sciiZMqq@5K%lY~F^;+-y+S^0##}a*XaPNET=OgLqtKW{Y(5O0?YMb+U
z5xKFniTrD8TO>k>^NQXebKk)Jq3y$Gi(E%rq_(y&w-1njgHYtE5}4ts-a_=saKH?f
z&E<&QBPsBZX^47S2R=5z=xHA!w@##RNI7+9f7zR7WnramA@ZoxzcL)cOU7MmCFOi=
zJDG3CfjgJzcX`w}TIn)#sUh74!(LK5+|`z(!>P~!pXt7J>vD3Wl05f?<*HaUr+g)e
z7X^7C1>`QvY%D<ezsMC)#Dbj$7T0I07Mm_nbA1IJe2sX~7|bp3*$-@<NK?@9``@_&
zpGZVp(xX#hLQtnS@ySo__R>?=J0UaK$of~vOaCYCT_wzRfr$HHpf^U5=!i^h;?!<*
z)!Ix1S8p-_K7Dh%tXnMl%vrqGRc+__#RW+46eNt79LM?7T1WeuOSpX)<f*`DUrgCg
zd57$s1Ir{qyn7iq@Gki6@pb+XSKF|PbF1cVYQq2pG2|&af>qzl1R3a6^I+_+>h?4c
zh9n?zBv{shJ#9jOOmeoY#mB!G8bNy3l!{iw|7UlQe<w?Z&^^EWp--zu%00EbMkJ`}
z90?S|hEcvdPmeFxxJjzq-^+SK5L9~NJS7)J0PNiI%nIYQWfJtGM&gj%WhmrW#Mn>|
z=VhsmxPJB|`YT9HtQuFHRMh;N(J-FYQ7xHDbQQiz)zsWAt%l!hnHY55Y+VQDxx*_m
zwd8rx9ArMkR5#=Dp;0T`V(b;I=nl5q_|GS}4Nh`IsT)iIT;_TQv=P3XCFHfI7}yP`
zVk^^jJ%P`sw5$%w0(XG|OKttYTMB7CYOkzWK>@O9u;r|ZOeY<D3yKk&qu)G=QSqN?
z$SWC^3_iuDMHBmY;%0*HrIwcGCFqFvt#`}p{-skQwdiX^A8wi_0#ouQtIsFC*-F~R
zQ*dv0LN`lyb<M=>X3=&~xZStaB*kcD3~bJ;*=8wLvt3DrPzo$iEEeURq*3Wi1a34g
z$VWU$JPv7L5=XYE8**lY4OyKo<o5f?#pKxqyU*2g{q!UQSB_dp;Jp;_GCF>xn8jwt
z)49u{u1{_6jbrPYM$a=tN(!dL9Vb9HnXX3b_Z}rt)C>v@zWxA`2ThZSfw#;_Nona)
z>@SBJb$SQ2U-oKsOV;kXgVw;ED3XEz8`gHQgub|>{2#tnsUK2^()Ug|#I9};ADSUa
zN%T?`Qg8!A3m1y3ss${(G>`cr)PDDVuX&1#cCeBiGMvjJv7>K2+Iex7L&`R|*}_hS
ztJtc&tmPG}?&fzz7a{~H#i=xte7s_r1}3qa@60*d%$o=>8+E9iLQ^Bcq|g5#vE6!~
ze0oQ~mH(MN?f=+x_4)~mLbP4?vCupA@&5Wex%D<#&|c|tnguK33V2*mR^8g{Us>&k
zJ5xRR(K4K>O1HB^mvGv`U(<a911t1E#7C#|Cuxs@sM;sZNgZt!fh>3-7$biW{p+Mu
z#zB}l4lb^?e%p9$XY2CvYT=3v#*klZdM-c5WQb)s9hRV&a0C_xk7dwA_u)6yR`KNf
zzJ-5f0){N5Yd@P>8f2@tEPXH8o|EuXqd5f2e_klz4qirGe6;YFkMaRRbUIhfyUiCm
zOgEqSe{|x$2*seABfT#dinaSNXvl4(Kqm$J(9V7pS2R}WSeun)CzX}|d_I~|Rn8{z
zJZ0akUPxRZj_o>vwC>-dUqwum1L<m;-R;8NDd&rO@uoWUJ)q?$?6t$yo|9C#CzM^o
zL6FkAUYpCbgARu`_-mMJ)pWE?jxL#3&(<U!z@!T4VXw#h%>Db(-a*e%W3KaU-empR
z61*4JYVh#`=3PTnnb8CN+al>Q(Dtc!o_lJ&gMIwlGFF3}2>d1g->)IXC@P13Yc+Aq
zA5=ilMI5t|sofLJY_<k{Y}b>@*1*u{MoK|4dK<hoYd<Q~@O`7;pCP!!um8dx{3iFg
z)Y1ywf>Q~$nk19=eE$h*%0?#)kQ8OP{tUKv;PSjm=pE%X#^d+EK@WZGmWiN}y%wA%
zC_+V`zHHcSOYr^jbg3HJjH4J2b+nnu7i|LK0s^Qz-kIgBW4~qBYwylP?KOL3OSIrT
z&&;f@7k^(HF5TW2*Iz0#{H@}k2b=U<awL;>AqWqnL}T&P{Z2JLgBQf{tznI2Lr|Wx
z?7kpTPG%3_^JvF_CbsR>vth*KGleRKLd-wByAadNPrPeb;M)`ENTx)QOw})%ER}vy
z-SP+8Fk5g7>;#(jZP4y>(@%{S1}`CZ+fCsr!e8YJ_8#5w+q<D<1ZqNJV#p+uws#04
zj$ep%>)ry3Sbd`n0r5a_ev<gJAXgCvAERnf)uR>Q0wldN7a=+$A|n&?rn%D&3J<P*
zeNT&M$pV$!%MT(FA+CuMOCiM~INsJkMk<agm0JT1)3oL_F<OIbjNBbG9#W7)^UCIO
zG06jdKEE_u2WqW%;iV7CheVD4XUo4iQ@58MZM#dK(f)o#q|nL+$uP0G7&7>C8Rd7g
znh$tDaNGiYI2JkgZ$_CpAJ0zIrvHj^oaFN}K<Gn6(x7?)Y!`e~e*ef9^f`9r6TUk`
zP76PARY!n*bUIE+Xn2X(B8$6ck<o`^?ET>cDPp4S_dqN+mx4v=3s=S!e}5Uo=G2E>
zDyPSk;fns-oa}omf0xGt!jirIL77(1$nF?%G1Z}A8w_lQ{X8wwEf=ApFKuBeA<f?-
zQN(7Dgw8H<IHWH=+7SA}0d!RyF~!C8-}aGnUVYnj1@2Fh<?ojq1jy(42#W4wts4kU
zYtSmIe->d0zJ++$G?wfOx|lWahqPvYN28Z%E*oVucaSPeKF4MXTSNadgXb6+o3im^
zW%5(i)yYCi-qpjz*{34YYo91B!-ztWX-x(%DDm>QRxuh*U5CpF$McOUFeO4Jb=U<$
z9$R{wk7XEF0@$Gd_?>`sLE=!u<9XKYq0{M=aklkB&hKGZfvY9!<fxCkg^MJaQW`a_
za@?eQ!M?w+!p|k<)MnOy&`@5+?N>%Go6+THeUFKQ`)Aq815N7sUK}-%LZo7!m^6xv
z<jZGR^PU#sTs<m#VSOl^DO5aL>2P5zjcT^GZSuVPox9#YB$mjP4bv~=2F|)WrR)rG
zGx#h6II#$2bLU#rzbUW7%%}L4lY(Gns1t1hCEu=nYF@uv%rQv&9n-xUs+U>UX|_E-
zWB~HU*?rIHKu=d^-s=O&k+EOf8FzDx(_p^XQWD3S1}r9#0A*GeP{t4JzAf@-8yQvT
zuNA3R-|mOZVFA&$p5Qki;%V_o@Wrw--ThKzTG4tah~VK7$S>Suq-tebm&|`Ur_VM-
zDHM!tn{BA(A<z7bZC}zNcHMDZ6l-bAV4}Mc>dYDlZC3Wn*4>#8I*&Z_PK_2v!j9ne
zI;XQ#&NH^LNEeSlxn29*4JcFFC|$c?<KnPET&CJ&KHXjZ(4mi+#Xq*8$1aKWLk+r0
znm&r!Mb(-|Mn;~ykcupU&GOuFn$P&er1`c?QY*5JVJz814TE9zam^eos9vh<4^f2l
z`<@+?eiiT72hSm4@@@B?N+#n)Y9g)r%IV?Ir|W*ep{d?G#vi8|)}uN?nic^rL?F&P
zMdT0<VdUl(IyPOVOjY7uF?5d)D-YrwP~PYpA?K3Q1MhjiK4<lF=B5-j(m;@-v6}?z
zCIqU^<0cS;9EMoW&!}cmM{L?#AVk~lXDc+hs#ybH=aSEAGPN4=uW#wZ)ffINH>@(&
zLkXobc*b?viHl`lD4*d-_7d)Hre*&t&yBlj;pwdxy?~g2JBJTsOiawH*UhOZlz^kF
zH{$$&kLQBthfvdB13;H7?`^ES91pr)A55N-dpUYpSQoRR+C`n(fjrWWOiWJBNc<?g
zyk^nCK>Xvh_DdEw0%_1$seZR}oL0aaWzS1q-Rwy-+mkUG%)J+E2n{)$K<MnxpLV+w
zp|RdxrLH#vUrPFuaA4K#H0U$Nn|+8G9uO!wBj|BPm}d?CZY;r6t5SN9-oJjCUoC8W
z9bo(MJLf46=I8EA=Y>%UH1gU@3+h^Tj*JQsK9HT>2_$K2{4Uk*k=)z?j>b`zUEtxD
zG=+li!oAMd3a>j81?9-b#^x?d2Rx%q5_()Liy$d+Ga&jC6hW7-n7?nxPbkM@z~fq<
zBeTZ(ImRab6(QsF%}fmT?;Z$a%xHTZhGy}9@0_>czkhMlU!yEkVIVgMBXphVhb!TQ
zHHX4oI)$Pz3s3<%2_Tqyv4yL^gm$yE+S*+%c5szsS~T@<W_b-qj6@<^?Kld5Q^IfP
zj4-+WK@b88Urxk^+eYAL(O5;Jmn?2xU>#F>8b)ePa(5AfVff>%RKy;Nc?U4WO}p>g
zZ!|OEa1@f+F(Vd_I}@o&TL7eJk}@kIGrU_pdQ#OoCK3;q4JFXor7ck+Zw$HF`#<RA
zy84F46aV?*3}v01jOaq9AL`mfW}i04jc=-i#8R&K0=ltk@u^{EaXuQh!^yaEyn9q!
z{I+P-OjxRlFgkEvKsfv$f<pwJJ^={UP`@FazV#4~C(I^L{`uBw)qHk7&p?#VA9wZw
zi4F-9mW_Pw{(0`deimCPI!hz*=oh8Ai0QKJ@0>8P_oKAG66<7ju1f^%-&b+*f&Dva
zfZHuoeu1gy&t#+bm`_U9o`2ZsJd1caxe#{jZWrzmKjvxZOcgZehdL)=d_J{XytVLx
z`#xuUVkr|*@TRFh|Kb-{FP#{5g;{clYEr>#dW@48QyG>FJTlgC3h_i_yCjyVKaLSZ
zp+kh4BDTkp+Yrgl2Ulk}T%YO*ZwvcfNv>|tg!}CNfeWJS6D^XEz-yCXURmDk%VS_u
z%N3DBWko_K&`*+v@nVQ09y{Qu^^eILMj6WD#EMogwIZy(!n`9Mw22bU^pV{dKJ9Na
zTKAbke!{+4nYP^1*1NW>E5LS>%8XSZql-^<bd^iJpBWzhVtyI30Kku*dkLtTFz72r
zCu+u6c3F=Ih!_5U-?g8+MsJAzH@E>wf<BLqkLN2-#-}VsJrlpwD?D$8a03pP2xpd)
z$nK~u3If6xrI0=R)^>x>lm7nUhL$nCjurPXPP=H`G@f>mY?gC!i*(A{h;t#r9cds{
ztZ;l91{-ACfJ;5^6c_e{Gi|@krYDe8>!fq3;EYA+1xS(Tn|z*{8XCq<hY@-R+4#8P
z)mEda_*k5B>l+$IhadSi=pc63_=F<PA|7dGA;Sn`OvmCOI?w-ZlsXnX<=CyPCC#&3
z_!U~@bKAkhc8E^e|9A=_TvrigAtOya&Uk$^ODxCt)4lqW1<R)kS@^K+r~N7SYDx;N
zG?0W-Z$}kLHnL2K3Sw+UThmK4mb!coaZfO~K&!1&?RCa{x*_6IgiDLKHBfNh5E!AJ
z(3o~(JsgXJUUXPLTqO;NlWRh19R2H>KPnjEq-Liqm><a1k53{+IQbmn<LTj*!=p96
zKS1O&Y$1agcGU6jw=_R!ouzfTaz64E9l$b|!Wn8Pl4(p0Iu~^bRTQMb)FJs|@YVYl
z#X|Z@s?E%>(77BThPjMJ$2bY6&W|M+7%|ciF)@I<nVgb|vyeX^*oL_<&GbpD`lp6f
zV^Xb1B@&M%nJm-PxEnE}Ptl~GNnyaFn}jC4Y)}%*Zv~mAr3-v<)z^7@2girAt$qD}
zN7qm1{cido|JX6T9Yq+!@?p^mc#1O{K)hwI0B-Eai*r`al7r*(Sm)bzONRE=W+|y8
zl4B$ZOPW;9469o8{A`1TuP3T;;y~KhXMK2SZ%_LI`(Amkis9i!nmoe{x<!C}<k9QO
zo$c4RVQpERl1n!O=#>J+#n~jt^jxxqM-rJPz141q*w{p>C~{5upd-^}zfB;UG|Pl)
zEq7E3-t?p@Q{44&%5+KS{%Xs(h7Go}u~q-6Y$zJWNW-WIpN_Qpi2-o1&tC>1lZe#Z
z#r9SFqTw%J+A#YukZ#A)u&{R%xVAZ*PLk-j)g3_-A@1stcD%53Bg`L~B^X%kOt;iR
zsr~b4W%N3j51UCJe5W!e<KL;rXQXCbSlmLeAm`j*H;EGwWm()pPu=X`#1t_&;*xbR
zY2MXNRJMzyi%)d23m|b?8$Nxw;O27*O=9sGLEZfaO_XBN+%#^3rM9AaZXfbySfi9S
z7Tqjb^GC!xMOJ^Ew0iV1>2<^#GQBx<%D(2&<*Q_`MDv0;y}aiv{-f=apv>17<){TY
z3xfev((3<TsCRdF`dyw$4*qvJT^}!Joyjx_f{%EgU{0W^&|TKFZ{bSMkR&1hsS;Fw
zzRqERcj_r=0Oo=+=VHMoaz5W&(2d&AYG=k@QsIgFoxQ(gmYp;pf-ZUM&)h-y1?Rkl
zG&jT78`P7>)wbWz>{C~3<0mP!ZEu&<tF?>vwk=zVstC737(Pc&4bAm`!{i;<^@+#}
zShr=4n{}C(SnC@b<ZQ@raT4=KlxpQWG1>W2AtC9<ARL>Rc|z2QpQg{RdPy`};T3VH
zF5lPfRfyjN37-Rz+tg&dyIG9!&<<~jcwHmM(>R>6*{@S6zW*X#F!zX>4d;?oUD3*x
zMTIlYQtl+1MQtdpuPZJSwYsK}&horQ%Dd8<Ng8YRIUgNC*W@+mz|bnwus5U@;8J!5
z0LvoLQ8@T_bQSqz=auOk?3+6N4Bgt+%_?cq$A<qan*RF{GC|@Uoww#Nx3MD8!e40t
zyml0+LN*;J5%68!kzQK*7t-p$W~6Tnz}@BzcP<$}Sg>3C^Ji9f`yd8x$+R0(4m#)H
z(G>v%_smk^y9p8!T;|`WHsJByVJ?Z=ikcM`h!y`FvDb2882Di|VXsQ-{)7ke$E{10
z4*xpvN>b=-j$Ny3#f>*`x}n-%7&4P!@`C@tpakP{ZsPb%r(r8)9y-rUWc?TM_y=65
zc&R!DY_J5Vtz&FshqUq2Wg1OEImB^BO1EG2WHtD^o)5eAIc+*fx&tG0{82ZqTI=_1
zH>rODY0o3%>c-(^uP~mEoH5;!`!ysidipY*gK6ye#2{1ARZ;2`A55@!_ZaAwBIiJ{
z^la}aW<E#I=7A-}R5CPVfq34z)z-AJcuMJrDOjKHTgSM8`)>FC?-rVcF&rbY#L7@I
z;jU_ThpkTRvOkMr$#kbFh_{J0kJH3dPe1#X{{ogAz+V2#ZT@8MlK%7i{F=gMr+zmB
zjJ$K2(647W%Pq_=Z1QGDQ5pnhyM>2O?Dy+?8lT5!naih<8?@Iz5o-2=a*eI_!I_o4
zSQ5by(_5>ZVArdd1s3a66=oi7z0Lgv9;5ON2OUj7__JSAF0WVWI4;Ya^{RF)(tQU^
z8Ff1lt?BMn$54A8ctFajhDQm0Y~>^?77j0O4bpkEBIC$17y}9G4U{vlsA5sQWFXnI
zLi$+3FE_O}+)w|}H>`?Y4cpwPn>>&?p@tqes^%o#`v^K0DYIl8T^EI{(>Y+4!Q@#q
z=+bsRI;Z0r3#vSj8CF<_b3U*QzeYy$frr%qlZ4OL@exw(3MIMGBjHakq0i^(fElBA
z%yC`gFT-dl$%qi;Bwm4|VbFgzd%(uyI5O;TvR4tHYTN3}nthYg4MhZrH^z4&sYG}#
z3r!}JWW83L<CfP{Mnb0F%!#b~jf7Uw(|(tnT{pvo8ia1egN2+bmSvy=V45Bgz%|vx
zI&<ZjH+0YO?obi@Ayv6hRY?a#6_2JI*oQ0-X`A;p@W0+&T3J51J6BhBJl+=aX&fa<
z{c7%=h(&f6=t`8UutJa$cO6T<@QrSB4t-o{FMiOhe2kBOQkLk}IKHVu8T`j-jC}-!
zX@$Gb%Xw|RcJb^=1=4X)%hzU(QdI>4$w!Su3nQ}#E9qNRnQS=rK7Wy)SYFeV0&bCD
zf9OLs&$)2<oWi1r_GxF%ww1WPoqD@}-u}9{1m)xXl^pI!3RL_ccW}-UuGJ!8bJh1#
zTOK?DVsX#tS#clU_w_pG*9&=9M#e#Yl8xptjO-oVgtH}wo55cL3i06l&`zejQdp~l
zy~@gJkzQSj6uS~0MtuA>EP1}`$vxlObwcl$Mf)UfzE`Sr{ACl98aX;h4xi4H-<@G?
z4=Tu!7=qFxL`t|R=47q?OV3`QV-0GNtLkqePiE~oc2l>sw;9XBLG*ax$Im`g#PgNT
z2#5Degvna@MOl-KM|()cY@3cek`1K<vqUtLv#;Bpf&Yi6vkYqUi@JS<V#PH;@Zc0L
zn&Jh5TY+N5-GaMIaW4+Vp?Gn3DNb;AcXzn?zwf>GQ)V)gnIv<bbM{$#?ceffw;Rmz
ziUfkCO!eRU)p9u`=0EiJVeKcRZ&IstK18$T$YoKt-w|ES;X(c&2ckdui}cIHVZ8ZO
zZYXK5Gg3xU?{+WAdfuwuCYdfyQ7SII=EfV9*uln_KY0v0aOAbW;|)@dozBM}TwIJu
zkiZs=%k#2GBrRWV5-FSRV8ugzj**glI-hMDy%sPgb2|eUDr8U~>y&DQtdLf%a<URO
zH>GQ8P#PvK7zyZ2_$7iTib>B=SjYBy@X8(G)*O$o>VO?P(rEfHsc+KtlO{$+{>8;a
zWNvp5VE^_-{>Pf8&+u4o#_BW@AR<sQ4nsobJ(f$INkM_&71_sZVY9qo43vAe9-j;S
z3zt)Pd;QPx-<(iC@cdpFfIOhovFL|u+L!2;cjnM33Bk8rwdW%f{sKIN`}GEOC6>%q
zOO#n-^ZqukfUL)*Lf6emo+2)iE9H_Ax(}+(1}jZ1!!H4J^obJl)vO<8*ZU&1tgLO6
zD%2*F$XN-M0i?87UiJ-E{iJQr8!j3QA2U4;lrYHCC2?Z&H<vZaD$nQoEYPEjyZoPh
zL#tNWq!*Xg+gQ~E^2#HpmFG@K`WBY(1HilTSzC>TncZXfDXA$tE(`cG>ZftEb^dJh
zSOmXv5=E{JRq?VDMe|VqYJAex@1G_2%lp@Vf456b?vodVlOZ2czv`W4n&ox#{^^&9
zaUm`P4H`b*)sZ1r(#)PMPlM3D#-4G(hXS<7MW(AmCwtE7J5dulbUz6GvUgfZ(yFoZ
z16GC_iMnX9hLOoB2#!jyJwLYf36G5Y^ziIhpMTOE=UsZg`Vw7HSyS^5H~P!))D>@e
z*6v;k5}zwLTeqUBwaOx&2$sORAKT-fUrVa%z1`p9V(52V2EUN+@Qr2({i^ynsWbfe
z#)tEF{o(SY-&~^7oUwv=Pz=EnbKYkM^<wdlDI@1DtZ3d?Y#bb2`*6a>6S&(D*Yn}B
zeK4u*s$&1@ai&n-+|k)z0hT}?_H)zq)+no>_=6@%DXnU%=bi`Ep?y4kqWOiCIJ@3M
zET`iv52PgWHj+LN8q+|5n>TUwYj%%|1n+hXSBkE?3RyPCtVTe)(KV&Gm?6#tsUI5w
z_p7uAtt(uP=OApmrCvyM+Q>kd#&a!DFt-U-b$cpYedV7%MpXb`P{hY?MkHQ<Z#pPb
zpS==FKF!qsQA3g;fpLj`^?x`(3}1a>TvR70w*IE!aRUDZvA7MH0sjI;gQ`gRB{ea%
zW}%<tX$4Hb4%VIH=0g(e=xij{n^|D7;tS)^pOPP5PJD&e9x(1>!AMa+TI8^!jK%IM
zQ=i1})jiz)e_q16i^Yc$*dy}MHWiY8@St%x{U}k?5IA8Z*t`8T`}S^*P;h7(9=>2u
z5)lrs4mIxm5yZke@DbxXZUknLB_xyG;R%&pA{dbzZxYf(Zif=3W(|MKwl~&wZNRL(
zuO2Xe#tVu9o;o(GB*_d+ONI{)7@z5Bz6D#<ij~dxUF-B+s8E0`OG;WCj&&@qiLWIB
zY^wi_@6l7M;7TC@H(_QH{~a1FSGo0WyPa^8VakuS*pN~Cv$b^6S+uajqsH3YvEd;V
z-efe^4|+9!Q;k{75@|%KQ&Q4thPSPc<qG@~@k1K5$tXc(YT;EA>=w1CxbYr#;G}FA
zFQ(8HLL=K)iO>UCn%1H?^j)gel1*<pQct{y2#?-8emJ?VFR!KU@xwS=?U4+Z^*(^k
z5X}CDQ5|~e31zh60+qqJ`>Pr>#ouB)Zd--a=u&knMrwynD_stFm*G!<g>*hp#iRdj
z<Hrkn+jE(1RaKi@dm$&g8FF48ArX9SY-%xo`sTKA?jS1{z5dV*1V7SW(%}-<GHyq1
zGpo_{n1^_;n(Ehf$q2P{Zj%gI*?zm*njE`a3qXPb--BM~H9j55!rgb(&~(^>ru#Bs
z3GREyL2h6G+o?jI_u8U{*8yJnj~p(+-a_y5vCD1NXO=R0s%aG3@M&qwxGja~X}}>_
z9Ee;{4b%By)6+_5*{9<Ej{4Jq&1%~SozVS1o$o=fFyb1a#HK=YYOKKO$zj23in59}
z|AEpVEbjGOyeP|N05l(BXqZuzTX-_z?>9Y7gPNT^yV#k`_ilD`FZ=!7hmh5$_GbbJ
z^PEP7VnwaeAfnhR6j>)}D$nY(rssa|>1AR?d*G=+AiQEhBO)U+v9gLNKF6k&kld@|
zXD+q2v(yg{n*P=lYG7_0>vNm`%6cskB(cet>3+cOKfVx-3}t0NPFW}HSu1!GOIrkL
za~T|!U|0YI&d{&3aq|U`2LhbhSE9T=m3=DcyZz0@)psu`D3Y(6tGS%Lc&fQBmR~&)
zvC#TG9z+?nx$OK`9v>9~^tuboqJ%gXuk`^IGg5&jSJ|9bKfKX%nECi9B5J}`Gcy9<
zWP|HtYm;d`1H=9p7lL1a3+0DaL<xm8S3$$V(Y0W&%;)PLl7)iTWsI%%_|xc}_*=Av
z=MI}@liA-7&!11*Mu%8iqc$7wM1--VFM7d222sDc{Din5Q;?!^E5*j<&e1^{st?OS
zE)93KmX2wk|6C*byu?Sx3^fu+(lGxTNwlx8e4hk*56i`arc)(;NF*_|lF`)~bPw#F
zIM+vwGVHvc|5oE+H9a<|iv3XO;`$Sb%qzhp_iN#d32&53zc=KM)Ze@q#;+CR6NmlC
zZ#89<<KyDSpOM3wHvFE`?-B_$UZPp!)E4fD@W&EayfSh~1v0TD)|@|wSF}Wt1V?;%
z!m{vuw>lEAsmGC?=m?JjX(}UQ3P03th9)`}hMqnG;?LIZIb{_T>>h8>bHrmANrY0%
z5?-HwJX1D4Ar~qD3pI$1$I@lY4N`4bb5^5^PORd4P$21w<*FKsW*0Ea)5^w4kXo3n
z&nZc1Sw&^t16aXp0|kBnK$0(+y-~C(8_N;)kKB_^GH<HUah0i{N6A_wVfQ%sFs$Z?
zd5=u%8TMW$MQQA>v9+UzwR_iEfd<Hi6v&t9{9a`F_)Idh`!^2=<zsJtKfjb`TncKB
z&g~6^tZa5>6d_eD0AyT~SH;%nFKd20QF0K0kEw9^V;hXHs-?qE78o1jF)4NY<|C&+
zCUpJuwnaC*!Q~a~$2{r&)MfiL)-Dgcw<`TyH0w@p{5`Xc`kzdQ2B@?a_hm%gr{~mX
z^V^d!PB2R&TmRV^+h(qzY;;mGkSJ$0s-abF^T8ldpoF9bSI+X7<#h+sJ6ggp#N@yA
z{y^zJ9;N(RqmhA}#vvD{_Y*Fj-Ch7R<E66J&s!VxN8OPQ9}S$i|ENo^n~Olr{2ijf
zGL5>0%8Mqp6#+`WXo}&#YKgEU-p{x8IR%by3RU)LAbsH_e+6%CzN)nTVDAgV&RHk>
zP7I8G#uOC3?$dEs|8->EqtPPA45Fd53kpB5{KI3*`Fb~r+}}xvq0Enlj_I(wATY&f
ztci<`8bpPJF$#UU64|*Og`RJClvb2yt9&h9Gwp!9M=iN3MbDJABZc4<)+}QGBq*^I
z=-Ef7O7sI&M-4syBm503y69iMQvgAv(|bH0CKc8nv!`Qmeoa$lFjGn)MWA<ylH{o(
zr%;iGD@@F{=SLV>8T`tQWoH60y+Irn;8=><{%Br?ter(25|x`7?sj}l{<o@j<V?|A
zU04O#{O!yy?VszJsXTLI@#3w~2H|BEyc!*U6r~XQpvipDpQFphe|VD05Pmv2()3ay
z@pg4zeZFp%5s!ApE+2}iT)Uj8qFnK~CDp7&J>Mc*()L@w=QD{u)wi+1RA*VrYqiJS
z5Da?$AihRLFKQsQ7?xKIea748%Y%w8$(g^$pOEPFg57;_T%xr;8b5LGDVF3h*Rl!-
z4NW}+t-4_+YS>*?_?)1i>Psa|Y=2>}GoFEs%{fW?TCyTnB{Y=x9)gOqZ7II_Q*YuP
z{R^s_$gE=dOp)JfP5B`hFJC1%zuTc7Gj}79N7pCqK|bWk&R6NTtc9djROMSWO~ndj
z#In=HP2F3u(<(<D2?esR--J3hW_3H&u|sDXM6t|==7#>hiGrSKvNKh*MG=u2I6AZH
zYg9sX`H^=s1H`97=QVV4m#r(6bo(*EPaUStr_2gpo&mNk=Y~zKa|DikYtjOY9)}k?
zFIlf^S?qj*!{bt+Hto^7k8ms<Rv>}8A@kkS%<{LHL8k}E-$%55P3ZQ{H?`T>B5sTS
zqSnK+eEd8GYi3yt-=DDPW70GSO|z_N=Ro0EV1n$C*zXcS$>IS?|MJZO5?JoF-}XaG
z4Mp){{|KsA=omRVVhAB8CntAYdSZ|Zic~96h8ZlC8YUF%m)HC&M^sCV1KltY?!aq%
zEnqgOq*w$t$~!;j_q610$t$(ZjujQu*z{Z&sbplES7+z+;;DNF=vOLVHvGd`g;A5T
z*RcsI+N5*;e)VXP$d1|V$o>NS!r9+8&j-hwj`l4e=B2LZ7$s3=5pPVz0M3H?oY0;D
znq>fS!YWNl`BPYJ9jBRf^VY)8<t|*k!td)s-iOgYap7Sko;#D#YmfKLYD|wyjWB1!
z{17f1f)bHEylkAjl5x>k#**P*8|)l)S+G3>=|!c?g~(|&MoDB$xiEGVJ|erT4u+{}
zl#2Fqjzjnz3N3M5^Ebr0^jBAx1~4FW#O8vo`{IxHBlX<>^tr_Tr_Tka$@$zo3R#BO
zR35d=Ej-6$pygFj7%4GA*Y+igOOMvL9yFqP#r~9O2gJYq!?V&nTCADqSsZBRuy)^|
z8)7k%>ag16mRKD6JUo1~eKe0mj~CqF@QBeL4Uda-3bJpMjeCnr<D8c>nZP^3XisS+
zD5@7h1^HZz-Av{GYu#{#N0dvrdU;E0^^w0ZT><OP^MEk3`nRI8jO19QhNcF1LV=i9
z>zR#^`ip@vfSAnK$`dC|<iyg?RL`7_(?p$-24YQ^qrnJ0+JnHv&mW?FI&PeU3|7R7
zr*u!5>yF*W&=anzasNp=JXFuoNiJRH^#DFLeL|2G^!#;W@?j#Z67QR}!r*rfyoE~m
z*M^i0-t{2EZm~0ECNRx-9`5eJBrdk=7^nHX>_hw2h4Vb%&U!+hj=goH{i)f#&)nm&
zqSuBbHF?odO?j=fQ4A}%PIx%E_>2}gve;r|DrW=u1+mY8GZNh!^1vhnIcv)xfz^8)
zT}p~Gu=i%Q0!|ZwmXR^xk@+{MLbEJU2?>(3_+hgk?A*LCwK1C#gWw}tsR>1ZFWXsY
zYkybxX0mP^2kwsR$rXE?>iJCcYIfr?&9Ih0d8hf}3i-^XTEBDj=Nq?3p*(LQUppEM
zN@`4@9xJ+n*H(2%vlsYCga4m~(VM}EJF_M&bhRHo@hS1}?(%$qh!zTpeidW6U4!>7
zL1G!dc|x2xwnsr9)brnu$HP}nnedXMgpwXF%SY;bkewCQaX$OdHnG;UNpMj0xS64q
z{-Od}EG(~hP-5vc36IQC`Cx>2^<iapoVF9Ho$0j$p{12%msozp@m18YRELkVSJ0B?
zolDhHB8zH|<u6-Qxju7?>1P%{JD4$WVk>JYHkF4%jx%}7CCfC8?G1#Y44>Vtt4!Dy
z@|DZh73Aiyms!}mn#l4Pda5Wx(WN1%_pp!^!}q++&hpKJaZ;2J$zUR%6DV=XYKP@*
z^<go`2tSi7Gx`+E4`T&m5}$^^b5wfPo^yWH1D&o7ta=+=6N?Q1=j0_ET^J?mogh$u
zYQJ%Zt)`w-t$}WyWBrtb!4C;lAsG_y6Rx9#VHL*2uqS?2lQ2Hh|4YgvTs^s!#@Z^y
zvWtUAmh})+2>+1iLFr)v`(5o4O;_@8es`Y*CiQz9^!#-ciT=W4l0p84v-m1a4J1HH
zOdm>6Hf0p~mmp~s-Pdz?F|W?BO-fRNWO)3FSOqV?pH~0ZFM{{=e+<`fU(FHRGy;R+
zyzY@d=VD&O@umPfe)pP3RArj-qLK4<^b~ukdV7wH)&{<%7BLNbmKfJr)_dHMlej&t
zzj)IFk7Sgz`n1$@?pfY)%aY}_&Xx_#SqwFw;=j|4s;W|YP)ds&`JCe&HWBkt+e%7a
z3h!4a!--+zn<@^X>W_#~$nS5^<|ZTsvkDiFm2J6Qw<THG!CZri#U1eqf~Z}cXv~3u
zZy1~x%WuP}Fu<(1p9b*lW#x}+c^ZgapO;DMW0nJNj561LCg@hkv?+EtH~<CQ!@Q?<
zYvFK{TFqKC66=)j?`8cqd7JBKg!a|%^?$LiHJ>~ti)w20zYqFd=@JK9=hIZT(Mv!W
z)2r!0wIIfdGxLa#6mUrvaK@69?vrx6-H0y!v?{_YI5fMuy*K$dL8VYTy*Xn-+J8L|
z9gz!T$+*CebH;=L<)O&yNPFJlHSzu`m1upk+#P^`181djM9SDv1pkhAy>DkP=`o^l
zFwiHNY$q2q_3ty1kfpn2(eUpQ3(0GKd&OrQMYKjI_)P)AjfX0~W882N=KEK*^q*(z
z11UKu^*x>u;gpHL@>)5N_=JHTRemqpFh10_u}M?;@x$B3vx5)(0@mSKGSlTHepv6u
zoI(w>j?m9s$p4Yayk$UUH5!YLh86SdwLss-YVN2-kbY|*5RBYK1{<5{{xEpXa7h8%
z5i7ViZxZL?Yi-NbSYTh11bTNo9RZKc!WF=mAUcKgoe>@urDgMX!J?zRP+?*z6(-h+
zQ>H*$h4p4r6a;CexT%vN;ofD6RhM?m%yqRAxkhDp+@DX90raPC#ZcBOd7H4{?(rZR
zK>v~LCUh$kL9~F-Of?KoP(r{pYHVr>FRHS4bZ`N+)jPU71d|gC9)=gMBx#ljE`vQ#
z*^<fJ`Ga*NH^xL%=oDR;VO+ylvEgG1kaV7O(RUy)M5Nd?DfA<jE4iP1MwJa=KzHb~
z4}bVHS?5GIp8PS5_a(pIIfcXU2R}=Xk-fyv#IqF~4trRd{CIL9iS_Mr$D`kXLTiJy
zE_cJM=^!pcM}bs{PhGdD5s{JO(Zui?0_*xS{h23{q=u+cY3X<0;U&U|Ho}rbK#zq0
z-pp<9$fKcC?etpyY}?S;89*!_tRQ%AM^M5EP8<iUw<wehx1IQIZQ@!cvHj5Vw0MF0
zz4SrNV<~{1gFsMiRny{O923EL`n&$9^xW0c%ZHNHLd`NGQjJzi+v^{kYTf_IXO!eV
zzjIEW?k|VO#xAwmueF4(9|>7?pM<|=NK@xgm!`TFoCYPhMg=f^VEkcl@<fzcpCmQ(
ztWzo#CXO0JjUdlza=G9X<#X`?ObM;GiEX+3vf_6CuG(~L;)rW*LD$sLF;Jp5)J`nZ
z^Wb}g^3$Tvt4Bmsq_R8{fF`!w@a=%cOnBONLUbDnOD;i0?0PybYuDA@8O>Z+;6ye~
zq(uDn2s(>yqcCNXsk=#PcgRllhKaBKfINbUor>G71P@{$EFw}tFED%AVx1`T%+qem
zYm>wr*O9wh-A1}l#ni|OEh6CXLpm??mwa=#fbFZVH=b6<WqD-cns?LH3HlXwPbjaP
zopxya@%h^71u4^LDk5cWVbp}qcac+Zz}Q=TMP8{CjT==V{l^*7o{w316cpOcLkA=k
zm1h?vcw=7v@)eS*%MuB}#otmv680NaNJMcMtM?b`duXR(z4aX<)K4)ri?kPF={?=}
zHo;CqTWEP7<;fyC@~}Aooq$poiU5kjvu{~TSHOuK+L0g<s}bGxe2THBVT5S1EJ^)*
z>8Uq_Z}a+S$=CTCH1%pg>}rKCds5hx2m;T^=`jS=G#qR!b;x9Th0MO(`i0+i*QVGI
zMF4l68-CR_i(hP7E?F{cWLf2<CYkq=MaITXZuPY9+SoU;pnNbInztzHcYqL|^LWPt
z^g;Eq0Y?kOcZBb10yg;@I7Gx#Gj7Xqy~@AF47H!Oa*^OW4>FaOS#vb+f+R6W^Cw(*
zeO{Pb<<rJp26jb(<{a<s1S}kZgbbiiW#0YA<Gv#Y6Z{h-gvBi9x2maewCpOI)Lg`$
z{?rl+xgX<;vX0faKXYU~Usbziw)&K9h|Glk+hQzr{Y&FTA8&WPfFFpJ2N6v*>+;MD
zA$R{7)GJV=3O(x(>7ocZMY?+QI=E&etY`^1B}`69-CYk6>b!TrRHnx}2S>Y?NjFzD
zG8u9L?vsyR8dmGoET3aFI?YMfdFx!+`>MwNCb+^n@iAz5hUchv=b9R&G!>wn`*VP?
zzcbiq*VG5<D~30m?SUbBMsTUm`zl1+Ym<c<Gd+LByt1`+aBq&SEQ6rIIP<Q@0Z?}5
z^7`UJM6~!jeTKbQ+#LgAg{fIwMW7aN0;5p-SnMJ|p?r}QI{lS6l7J*Ft>tlL5w9Dp
z!Z`{2!ewY!S?3LhKo60H)x=2kkAdaOn4*;YV?$9QxEwV<bxnyN2)gsA#p{>-)yM$R
z^At=j>x)O~pFvw=aOp-TPaGT0Zd~@S<0H}yIE)i+KWd>FTuUlVd8J4Dz=P_L+!uWI
zle|OSTVD#jec<{O+)Tatl=5SYSOv+|#*FF3Y7Uf7+Uh1EWn&*!j2lvV!K*h8TF7h!
zAu5ohe*8GTuV}@p-1LZXe?S}!T<-5X;K*m|&;KaY7aw=Az+GmNAxb$md<2F|<v$F^
zGM|A$ZvGSJ>uBS)xG}Y^#OW6!k7Vh$DW@-?NQ1v86SM(r?1I5+sg^wDRjn%PNJ6fW
z^39e30fTQQ&1s1(-{=Qv)NUH2zX=N8+&+5-Lel@*U2=b%wHdektY|KpN=u)3f~uw|
zD<3w45-wA(p{Z$L=@HHwisBC}lo#q)&!vN?WXz+^lk}RUAtTZu%sL6htgQLi^oCNN
ziHfos`M$mh@3hMUUA_0O(htS}1`3-r*k(*?sGY&4gKriX<*SEey;AlqlK?j5(6u>s
zO%fO6E}+cpN-wT|pGk;K^N3>Az}zsI_a@IPT`(;syYn7V*E!@34*p~kgSpjLljGic
zei48*df_tG`UhXHB6pqe?@Gnj@`-;W^?Ccx+FFx)6dT=A)XKWTS%3VVIWe|k;3z~6
zXPN{oZnyxIXFD6nuvz>zRUEY+G%qhAjoVz@LM?b`*@N3=u4=Qh3e3u4qbhU1W!PVj
ze)4uTj10F#PgRm;Aio$0*i9DXP!NLbUvDlO>D5q<+{`(Fq7IFY8sRW_Q4KXvKr1Q9
z_8Kc#<b0eAJnZG}g=h_0ka`D~M)JHIQZg)0WAhjW&6s@Fg0{kPefvk{Q%oE}MD>%O
zAEPQ=PQg)U`?u6&K2J)tbp1a!1(k&A4DysU5sUm(mmT~8YgUM^Ul@A5sGD!;-UW!{
zqhub!KZF)V{ar^3jtMSVYHc3q4L?10=CiIO?eP_X=)aw%@?gz8RT&@zkf-Un8YP@p
znQNA@M7rb4N>{IRxV-OTgBv|3tho8S@3^u&)*Zl}FJQW8a+)g|_0{H%aGxjgyXVzc
z_0kHK1>7!Sd+I|pG*n-|+(2d2eO}MG{KPJ(y{r667w8BDS<?L%7TylD@RrT-f(;UE
zJ!M8^q&8(P%JuJf@h*~ErPhVW=QK^$iuH1;6T+>YM1&&7(*DTk7o)1iR6~M~8uV_`
z<K{B0{kom4diPDgJej^xP6BD1RcYN{Tq=j8W+639pa>$#<D`L;DCJo)F}aB4NG-{!
zan%EQ({W~54<3O`qD6$fi@@K7Yb6_BX9p01RE6*9E`+KVcD=~_qg8&YDA&c%>EcBL
z-tBh3N0b@jR+EH8kmK<>Jy1pgzli;*WY^KhOkHqCeVzR|O}@LC747-tcM{zG=<>?)
zx<>vK06($eE|#a!)2Gp~CwF<)J|{4Ec0^+GRx?6bT6<PRrgQ3jA;YMGCZ>db?-v?G
zNeyk~mReEAP4p#~woo%xsFh7zV_w93@kOX4S=Xg<`Hxm97$;Yq#gt3k{L+%1nVCQ9
z%gm46fprzMXf$II4k>vP_2@7(TpAAT<VIGz<gDSazNmkwK|tJi!-&Y}@#DG%w?kC8
zPnIdE!q}ksR@bn8ByHq6va-o|0oi?75Pqr`5wZ&i4D}qLiD1FE)PIA6aUaG|15O{M
zBv0NkX_XNuO~LDToPK9u{1ARJU^Sj0iLM#(?0w7V@9nmsrbdOsgR!O-dGXO}jsmh~
zI;y|erMnFP1^~ZPf}P1bIg8T|`7vgJ<H(X&>T03oZIF{%^0hi9Ft|18+=!&gE$P{G
zlqxUB^Y)AC*!cW@Rcq_$#scwQuEAaJV;(@3kh~&Lq!-)VDb2e+0CI9%(rZ7DrYAMt
z6fi+rBR|ehGg4l%*>vTBE%C8MH1uiZ`-KB1=p17@pCsW>&f9gp?vv5q^Ab{*RJ0@`
z0RdEV_HT&9$1HGRbcB9A`@IJ*)GjsW)V@}Gdn6G(T~=t&Nhc65ueesuqE55M9edg&
z$vGho&@ELpaCZmqRflqBjdRI`8Z3?|2Yeaj79+V-aXt?C)PHV8nGkBuFLi;m!*{)J
zN*}QWjh*LRa2_zAJ`d`_#b`>ayx0*aQhPreyN1YHuF$NRy=)c76x?#t$F$|(7Wl6k
z{tx1r%V+TSue3U#MMo&o@5Gi1H>ve#n7&V$G9u+PRMNc-6g$Qtx;db02KLPvZMc!?
z?>TI2vHN#yFGR}J@PG1t+WGpYw&_`?OtDW|cusmSO6lZZ;vJCJF6KQF=3gK0%x<LH
zDi@QujAJMMW5f}Y$jC_ii1FqomwMaP^_w!Smg~*gVs##K9-aoT!*@w6ZPFF$otS7D
zx{Cy#`@+66B}Qhdv@WtQc|{3P%-k1#YR6rdCLg<gnbOFRg1oF(yL$91^&pqpmWTI=
ztiKuU!&&!O)jOJA{aRDOCXlu%YZ|NF_^%rCi}2cY1N7J$R-RWopVwIZZ)S&9Ol;Eg
zoaf~fZjy?Mjm0UJ=1dc)47kp6{nhJ>@3}e)w~LwZs714J3yRmzZuOOgb6I>U*KWEO
zd>65N>Y>}F$xFYU6BNRY(JG%Mk^6P3w#wb{6wMn@313cw2UJ}X$SJ)RZa$?pIM+T@
zY9TbVeY@kRFs>I{od4%Pqc&~7asOi9m=$R~t=lT&ZG}8`>D(_9koGV0olYJ$Z!R^^
z`-0L&wGC-G0Z&X`3g?Za1AGIks}~0F|3=82+yv{cnB*eHPvYj8MqLCXt<aMH`U#^H
zjwessh<@2S^t`xNU>D#VtF=6h`{TijCyU6I^sCiztYOgr<lIX3nvgzcQ47cd>~{Eh
zm6nzMt!opncx6js{job$OnoQfE9a2%UZ?Ljbv1whC*kCBWm@LSy|vUmB=jFcruPHG
z$x`dr`dLHoRj>#aRxn>2C|h}6-@_x`Mu;+rJeP}0y0%U%bpgr9*f{augm2ZlvWZuY
zGnB;r_S9UWMBe8i<nc{Oi)&irKkTlKdXIGLzrAh{(fu`mAKAa^UBAiFm$h$uOn8Vt
zq=3{|HeKrz`lF8Z^dDHvlC$8^c8j<Ph1$WieUNB!{r&#a4MZB45ipPz7<Jcb+csEC
zM{a859(VCBuayxx_0<wdm2lnn42*b*os0exCz(tXkB$GW!zp`g9|g`^U8$vt*pG!o
zD?a`+J5*&EpH-K(Ys2eZkDf#>^nFF+!#o570&e(;VyWbCb=$AVg!cNm8|b^Iws?K^
zi!WQ8E(xyXqa|<TRZu<sDd5Ax<2KG_$5T=p2{UH|2CjjJJdB`FDhQWAiE)_NI4l16
z!igy!T}r=IUzBPX=vzwBPKZ|d<~p*JV;J_}jU6y+y*q^WrAtNo`62pun$entTn+-v
z=`6oiPRrw2j+@*kf_hR#>aB=s?#)WH>meM(m$Eu`v4CSr{OuwrMzb+EGcKLaCDT(w
z&))tQtIzf-r{H|9lI?_foeS>78?8iiZ<mjs?)H0^o`(P;ZiT~DRfd6WV=s+vaQ3NX
z+Q%(_oFvxOwPFRZc|B~eRFcPzpfSlYFM+=7XHI#q)t_V=nBMv*`a>&5t$U(#vrM+e
z@8$hBaZIPn@K^r{&-2+=R`6YiTQKTZ6VsdDY-$-o-$3RDPFRyyZQ^Hi`Er6`@)?>9
z!2s6vdzPpHkTwU&=NR6u-p~F<_7PvQT$7K59$2naaL{Nv>K8j=+5gFIqKg+{SLM|d
z4yb5_B?e2#MaQOp>A-CBNfBimmeLgdeb#?_5#n<tCwRYPnw-HmaD8Zd?5tnCJlGeE
zQTIts0te^2h)~DnOO02sT63=}?`Y?Hhf<b`s<uX9R`&>*BwPE7tG7(w{X|iihNuzh
ze8<=VxYcf5Pam4EngTRFr6i~Cq!bEv9yizTUArHpWi=C)OeRPhhm;-5HF<qy;gK^p
z^S7{OT7*YcUhs1o2A?ah!#59iLgyF|d7IrUyjSJ9wHaMn6rS_b=Vk1b?{zEdDPHHX
z@)g$f5J^asncDf$Q+qam8Dx<G_r6lF;iwuUgQ4lgiJvWOwMC2v2OAab9;$HDeGZWC
z$5Y8j-j);G>30;b6Q8!aB&v*oM}z7s99-5B@3WCKv@}dK&7;Ut@+V5|uIAVYeqlJ6
z>VLp3M2;sK$i<EdflQW`yhSEtue^Elvh_6RjFbb?H*!tqwQz8KrJUiJoiDbgI@egG
zul9LuJGx6==yT#GgZVE9Vj)mQ3*LaTPvK3935(4}zCVeKT3VJ7(Q-Jq225`mKD63I
zOIM6s!vpG<fN()Jo!1S6GQD(&8WNWbf<of2$ijy-*ot~BiQ+V{`)3lKy36ZzA6U9o
zJdnLsA3RNBE~fJCU`Uh`u-~}dbL*`&+!&MIr^Ng!Dr5JlbYsA6PE0aGPz<j|{%v|?
zW=crN!BPV>BtRkE_;S$6!ZKVo@T<i~?C#6p^XJVPfp1Y{e0?)8)I*n5ickD#+Y;xO
zpRgFZuGyjQ7JerR4`7*tRfQ0_cyDj!6jbbs1IYdk*VvgMmhnd%@yQp~K?=W1l*9y0
zqtaoHhy+hc1qpv-Vl3VACHmN_J;3_qlGm`jCok+;K1<QPluMJ@s+5z1OPWa<8f2W`
zyBr8t1dPuaqX?C?m^gi|5Z)opURo8hL!@vZK&^e;e8)cE5Wfj3T-mC<KWlt5b`8I>
z7OgHH2SON>ygP>PIjvaB6x;1Ss-xH)A*-4DF@280<}D-p2xq@^_Gf~I(Kk;^+3M^<
zCyK^q78q3wsh`e}R(CkFUTC<J>XiVZ?LWog5qQrW`sD-B-d7y(+fu_f%N!219DNFA
zcUv%L3=q4|z}_2{<xm;n=UMghbekO%5gq;4_T<g>1!AYtzeuK8*QepPOR)?Q3C*OW
z)@WT&j$gV3F0_BoE2P<_C#+^Pbaj5u%)KT0TjdmP>$!Ufs@f?o%<Q^+pQHKyEi-FH
zTpo&u9VY6V@qOn#No>Dc+zeEQ_VYVQq~EKq=a{_Q)r1vT3-a*x!5_ImYSHl5gqS|8
z%rA2KZXZ?}WYb$$(^=;p&%80f0#Ss0$RMczMPgPoH?Gw!0m5>-gP?xb+mYraotDAk
zb)>nb>S5PEe9lI35)Ns(lF?NSo`H)W(~?1d7ecy>sup@d#M$VBa~R<RrI*fr&!<`h
zYZEd!62Y{P9Dd1LQUhztjEo0H<&e?}_4#(6tX`W~gNXOAuyi5@c%Pb&%<GE5J`+<~
z#n#DgnJQ5_@?r*cGdCX8iqPNiz4SzR|8t}B;}`A1j*SgsI(cv))`?eDlLF8qFI4pK
zs!ii~Avul~#NHDto=+~A=<|Q?<tFVf-9C4i_c!(Jk*AMO$8hAq{q&FLZ{E8u-XZO`
zw=ZN-n`b<)JO5qQD<1l2PUO7;iWLMyudJ|sA>z^EA|w#is$=hhF}yCIHtK)GT~63q
zY?J%51p8OSI05R$!(S7L$VsFP4b#;s9lcUUM~Oy-oKr7M0uJkKp&NVp+=NNVDOrVU
zGUaV9q1%a{Bb0fGGPO;<3tL)`xcpngZ}#D7#(&mBDRu1GQ~>+$KA_<Tn}0Se=<ENt
z`w@Uo9bvM4LV7bc@;lvE$J2LIvmbUOZ6(D3uZzRnc|~I|bv(zmVcSDj9;EKAz+-1z
z>(b0|XEzc<yuxUqj%=}z^Ol&)-7E^o=V$LMwUS<WxA>NdGJ|_BE0RiQI&*hh@mr;}
zvcaT^%h4X0BexAMSk+jhB?5zqdO`Q!8#JZ0%G>kFQtktgWqnF2(MA%vr*FIiCtx<S
zBXI)RkVfeyD*CcCc%=7pQzG?n#urg*rh>orNulqN%$U^xD?DL`7B;%=(#<V_6(7Xf
zgqujUeUP!#fLwBHL?$%HU-N2F3*sq4`q*h8$wToy*j*!=52^PdWWofd$f#nVR;LTG
z&AUc$5rY2OXxZ1O(<~75{o?0wL>jri;lFf3ZyTa&my6BiC_dziO8s-X-&vg9zxBBC
z%(5?(v9OvgRw>WHmnGvJ>hVMI<XM~V&!Un>Yt_w?s2;hwAsDaq_3=_b6%MnCu6>Z|
zD-YOq7tGu}{rfK8uyeYvAvs$KL+BDgeNX@t{tYM^IDS<dkmgsSV_c-uF5l8Pl2@dp
z;Ths($TzBh<hlKT8Zrv*nUPPdUwwFs2@iD4m?rh?r{<a3o}OJbvG+_Z%yRe3-xj#@
zlPvL2$eRQNIeK}()KkR`as9?`hnwLgt28CA(*ZA?LZ+T+b?=pIrcKG<)f89xR6|rj
z-FK5`%U8eoc+9mCQvV5aF8`jl{QygH1bBKPLj1x+#>~-C!oIT>a|JUyUu~8a&*{Yj
zNk!}!eyOTlg=Dtk@5Oz^5}u#AKfT+dXTsQ!;peL30_n-h_qDl+T&daGWd>`uyi=0d
zy`$Bq%Apo_zlSX!@h4a{vVwk0!yxr@s740;*XBL;mlQg)lOwE^w(xqM{^CMR4tG}s
z?5nbnOtqAx%uXVb9Xktn&ye=I>@wZ$0x*4zF>ABBvo#@hw4`HV@Cu~);c$KM;g=dW
z8{-I0YZW_5_2c&umMzN1rvMGFRh+r<G58Y@v5Oe7C1GB_C>t3SKZeh4PTp@ebGow8
z@dT>U@*UQ4VcX&CJ2J>%#-^qTQ_)5UQe5-!mr1J7q}<n*YF-v5S+t&CS?P<h-<b;l
z^1I@VjhuJfo0ggww<}an7bxo%?%h&4Y{D-Jqr!Iq&fAh_y#H^fU_5wvS$%zc8uGbk
zbz2W2biJPbkInn|`Q_rr%g}%GDo@ouH)D73w`z5~er1_oKWaaY@+f`?Ih&(2_{6Ru
zT^pa_9JDE%9(VPSP+-b;*JwIn+kvY<R-ouO@keeCV?Ihe&KD-tF|vK0PMVoMbcwY3
zLhy1PZ&kVkVn(a3{kC&e6>r5{28*V?89Aj?cl9e;G;lHw@D}QF%2>W#b7z6XogfY_
zXGZ~+#HlQmK4Osa4}`||7g^Oy8IBw+p()XAPlZm_XbB)2&UxQD#bL|2ZeYXMqPwa{
zP?6V<qJAP${K!2nLQAat&f;);<76<sdvnaF63-c-AZrj&pj%O0t<sk$<ESM^=DkJf
zJm_+No@QIs&>)o#y7I90J`Wn}vV-2f1%Ng?0j7G97a~qbJ_piNHm&Cz5R9((oLAx!
zK3IseP!1(IKgefDIva=J;F1caO2a#!^-k&txYoUon+i12^U3{0&ofMTw$0tnma-`$
zbSHQcn_Dw<x{eTEGbC^%;Fxe=9|VeG7>$DdJ;UQafN3-eFW7G0soi2w#)$YH^aIMq
zmj(ZP6ASJh$O*TjX%r20uJ*I*0|+@pgmPj|mN`#^7I1xrEPNXV@=7n+tN}VV775MJ
znfgWrg%em~*03xWjT11wQ~iCN#!>yVpNAS>z+BGL=zPds9Qyso%GS;jv5HE(9-2~#
zMEzee%ib90+v?XNh|Q<vwTEUN23A`Q>OJo?hcUrm`W@$64RdP<38Pn*xc>`sUj@v`
zt`+zTABDosXoqHPYd}H1>ef%Cl#4}A6GcxdvD2x_o=};}x$e8y@!>Ej@du^dj+gH+
zwM6L&8X}_gfvNz{+r^5r8xq@#j@;b_k7HUDhiYC}Ws7@AKbf2VWTCu@E!SLetyKt`
zUWdt|(^g&XtE1w${`Zpio}|8q4?u7OrXg2bR<0ubIv=}vs@=vGk$Xm9Imgnxwkdd}
zz{wi#bdA9uiLCGA?QhauU6p}3DqU$LAmk64fPhwwfLx6lyy%w1ft_#>krHjT-m-8C
z2#hSPczND>uqv^rn>jrksSaLk3PS$;d|+?&?&<2Hy1aZu{wPnYmVu&UN>y}wbtvem
zv@1b6(|+C`fhHqu8Z`mcjyUu;<QU@ev!k&|kGpyJ+4@NIvb{^>RouT%*Ka3-n9cON
zPoy^n6f^gpI0%@nvYC?dbM<*a9c47r(DI&91e#EGY&{C(8@bsaeg+x5K5^-vf7q}I
zm`=3L(e*zJV^#~;<O78|*}48G-JvtA-5Oti4=;-Y2i3?)iA;8BL$y!}<<s*g7sPY+
zU+$U{>;(`^g&_58;$gN`Cdh$yr@a;^R7^%Nt`GRtIx8mYH~3VO+cz0-<cmi#Agc=?
z*1Dos_&12|*Z$`;@Y6tms&o1ICeLbJ6@TQui<alQO4e7j5lvK<08yRes8d0h#xqp#
zrE#F2yr}3AmekgB`n+;_&eVRvq}^yA<ghy$TfhKs{VhGOksHxa^%hSzT$b7Xoqgsy
zc@cAss`|ZqWaP^vQrH6_WRT$H5bAPV2g)z>0xUBZD|pT%!}65EOevzM^Cs{@B!oZT
zFRl2V&T^+DXX>4gQbAeQ<D~fDwJ-^ZmX&Q1pZ5({w9f`nTvEGC^8_UXCxQ)uFDo>Z
z0WCN`|8tz$YNj3txu2pYF>^enQlt9&?nVA>@}ff^g<=HSZV`>O$?a%WUBTXGT#IMg
zRu|puR`>9yDoGTKblU4LdH)2f@xCKb##bq#=HfV}+U=e=9h9MtIESyV6SZB!cmPBx
z!l)wkgqMA3r#ZDdJ7ogki?e{=%AyJ!U%rT>_ta<!rr;84V#<5YxHHOeC41b^l7Iaz
z1%WPn{-yg!xp~4lFwcNv9^+qHR9!Ftko%(y&UQ0U2zAJJaOMBYWe-}G4-sWoNVNbO
zn)PEx&B$U#8pQEur7@pn9MFozi>dxfQhX+3$^(9J3>+yVKDLOQj}Mu@UF&Af`?9e>
z(09&G4_u`NroJ~UV%5nf7vmgI0*0U4qyLl%v>8yfK`s6OT`+l^K@Y<XfqF*%7}Sp=
zi07=fDiv!_QIz55f=@w=BpTByr0Z`DK-{xKEK7Y$unl;yb1Ug;K{OBpmem%KlF5SA
zlQ<^y%L^H}I|3|ADCPc%pzT(Y%RROAvKlCO`RdOl@etQF%xj*2!{Y!zDehogF!e~M
z4;nm~d`l#phM9F1<OsJ?K{MoFT)hsTX3Qw)HV`TEb3^7U5K2}(NaJJf&4@iIr4S;v
zK>aO<UwJx0X}(Yb7!qBrY%|o8%9R%$8OS`ZfUI7d{Z=;C9Q+)E3(H?6==xLAGEE9w
zL~0j)PoAuKU@(50QDzY^4nJxwx*|Inj+q~Yyn*xdv~CvjWs>-uuh6K)R1o~4Pk-&z
zQ=1gr+FveM*UT{AD3k3m+3yqVEO3iB1Z;Jh>OW9)B^>><ISb0zty%mu!>z`U#FMhf
zIVOfF*d>OEn9l2zZmGU1mK=D0|6QA>!-2G^m2BAtz9fm_rYak1;HIq<RdbAV{>TpL
ze6%KIBVscpyoqBc2s}PI_a*z5m#WvAyU6Rh#I=n|(0(v(8cBnjVInhNoE-}5%kO~)
zs1++5cSDGY3}*r~N&7xx<=0)ti0g3~(2uDu?65}h+`~qR4LxWz+G?eiRJ3DIqJiMf
zhXwA%MFHR!m*9S(bLM|Ts|8x8@wC{|7Ejt8cY($YtSr_zJewX>niiIpG=au(24>tA
z+HH=JjVbZ2B3%P^vF>>6%`TbOwSl@{vSH?wKOV)>^JTUae8RQ{aJd`=VY~o_i4<BI
z={N>=G2-0G<VBYXWy_>u7h*Kb<IP1@uf&O1hCsZUN4RCr?L$7=tnH_>z3$yubMeuD
z{ger2wPrxS#$GU#t?wbNAe2@vB69j5H~jy%D8ibir@xVT-|4>@P3IHCLik_sqU;A!
zBRUOg2sa~h%M&7!w|Ro2t+4Nyu*X04-!aLmoHGkX(z9neA9Q~z5}Q+;TtmNvXk+2H
zG{sneDeWBiQm<(*`FL~F8hXvk9Z%w>PH}EMo*I^S2PUYmn!UX5B9nOSFL*K2-F$lh
zGiqvZ<t9&&xInW6W;2WYE=0Z`EBNi^X09*oK`%08@oZXNv4N{=Iuf}nYNC{Oy%>No
zu0g&^QG6p-+}c@H_KP&pvM$B^yy)#<gK4H%GhYPOjCCg(h?|4Hh`VOdT1_o!uGp&j
zTlsI*sQ+}f2@B`@#2@#@;g<i-R3HXHrtFBb*LFID`L*8))d!`mmI@_o;qoWpQik<-
zmidp>c2Jj@YHim|8exj-3&RFVl466i>Mn+0g)n`0woSN7fkNlor8n8f5Y~gQpeV7G
z+t8e!T;K5WK?A$<W@>y`sPeb`+DyZh-+zc7x=bh~Sw1Iwh1^UWODwtF%k@Pc89odU
z#a)4qk3UP$^+CHRiKh`E?NRE3>rtvHEXR~Wf{}RdrVrm?z^vk(7Z4hAkNDi8K&a3s
zWS()yN&a5tx0E_C-hv#%nMVeXI`zG8&Wg*ZjYvWw2T7$uM|#oKGRz#by0sHVPSyc8
z@)<VyXda;1w)5BQdZq22tVePYY32NV8h_9wCWyq1zdS%*gqN)NuRO+k<v^Il?Fyi>
ze@)lrpqrVHaNlp&jUL~$=%2@CaR(Ga8j^w}pKbAJ@Kl=yLMejvMt@@ao)tX&)TShh
z6OhpGPC9pU*r_>P%liBzs08S<&QQl)XFkQK`sbT`7B9rQRkmW*)ZY0QU8d{1SzPT#
z$JnwF>o{JTGo3($*;@b!sidKXuCAoyf_sBIMkdolaqZ3i4HZ@}b%?@wM<ZR)H%7d0
z46~#RxX(;ic1mWiCY6_?i_<OPR|AZKU+b)I_UtAzI9?$JuRq2-da7ZM>q8>UiVN-v
z-cvnao4<_sh3e>mV!pKbQ0s!C9jzev2o9H#am|!$Ue4Am;ox=#=F%i)pxU-R6DYfa
zuR7dP%V;y;hWXsinHd5oP^I_;ZoVp>J^(>PtLNcN%u-Hs9B$_#&t0dB?n&`SPx~(C
zav(xEJE-BV%EafIdvbyb7K!C;M#N6M=iCvG6^!LlW}J%k(^FVEdVaSLoW#<(r&R7p
zTJ#A%W81P4!RNlkRzP^(r~YGot=JIk`tX)Y&&w?`w3xxZxvEJN3z<KP7VS%4zoamI
zPC2RE?BOx9n!?4}{14$?vbG<TXp10dl0o)QF=3qE(pZ?ZyKLzXs%7oGMV(HcJ60@@
zZZq=?4%wY8KWt<}_8K?|D6#f~^Pe=Ugp#5Z_O_?Q)xzn6OL)~GLm@74!v-R<B%-z>
z#hkrXTv#(2gIYN5;PSJ%*ZUAI!MBpopLhS6l&JnQDakVgovy0Tx#%PrRNaZ_ZafbS
zpc#G`(PFY{dBF$~dPFmEWv*t$X{PuW+t6#oqYvWBD;Eo^3SQ{bOuHI#x==CXEf3%I
zs3ArV(~nw@)n93?87?Z}CD`GG&qYO@X*4>7FHT!?&l2#uo&gNIaa$hy7V-K&g3r#n
zJzuJStl=sMDKwK@9z^SI_Y%JP-n)fVO$bKqDl=}g&jk`AytRbSQL=pE1ak%Bk)eDE
zmJH{m%AY=fnLF*(8}g6P=E_Ga;$s1%NbqcQDA?bPf<x`-*!x5bJX$s3<2(qZrOZkd
z>hz~poI*Y^Tm*ME!o^z`Fd0_tes)d|lewiOgWsCvf8&?FVAWDa8NxvGq!UZ5dVF`)
zhfU-tmKd8F<$k^=7mU{*c?Q0CVW48P^1h>H&Q<Nwz;U2b8+QE-g!p$p?0$dk9hE0F
zwMQF+(tmiHc1=8~pV^oVZL5QZrz-+6niEVhsk`F2)K3);TSYu}xm7%(**UoS`tlF&
zsQGr}N!QOlW^H{O6C72`bK90YJ?pD~0T^OzKB_e2V?Wsqz&L6;&qZ7GehzxrVnv;H
zCj*lcmZUHs(y-%kvznh;VR@6Zj!k%~6C2)fh@AB{Fm}CBA^$`9-)~qAZRWiQ=qVd&
zYh;_}3$#{Dj84fSaHLY`fMZo>7d(oPp1UfC8e3h-Fh@cMs6qp5kOc&l2G1EjzkFFS
zmXv1lI?q<G59Q$yrTmenyIOVXhgX*08VwdQB-@(5UBbn>h4X2H*ikl-B;5{FqB+=k
zBDfh?W(cIr0yd>DsB_bm6@O@McaeMJ`tT|^9kS$Zaxyc2vftQa2bMKH&1lIEpR5yR
zK7yh)?Y3bb0-<L0sewW7sxL2*5k}O|g(@GEayN5_M@QG1FeHfV8^JE8EPtz2@@FH9
zRGkO55n$M=mY;~r5K9gZxwPbnr#ZxMKj36+6d+QH$yy*Ai#zV3rwUkZk-Zvk7@+9q
zM0_tZo3*U5{5+b$+yAdSB-T7SC~da!ex*52DZYW4-qsrT_7UF1{Zq@hdcShJ?}U?g
z$X4${K4KE@Eo1;Sf19tI1KJ+VXl{8LgwrP;om-+>x2P5)G26A#P%_`{&Y&)<uNXb4
zXLd)(WA&q~QG9Obs;sXck!=dS@<5(_sO8WbZo;sd(H{f2vfK%pw+4~8x>5@cS(5Iw
z5Jc}LHPqT-U}n`W)@zhnDe1QM`iap2$WwO)EeQwYK7jAes6AgeeWK{H?muXf_F4+L
zXAu5)^4i(;>XR@8j^(PNR_|)yB2ZeE95S<%v4Oe#@ZK_z21`0Lh>lx)ny_~MnHr{B
z-C1N6dnw{d%(`rH2Xi^E$-WpMlxp=d+t>VMEs;JKIbHH`io|$OAnhnkNE!#?$(pBn
z<+`5__e6i`-7%RLtT2xwZCN2N$;~1txz381CP%)xDlGdCq4o8@(Si*2U{>$*@MhUk
z@Ak1DuZKPvqS<F{S&xJ2Z5N!ax&~-0Yp6*SLA4*%0fG%7#nq<RexOk>q8N{HO=GAX
z>4Fy<G{TW)WdIIO=GmPzpV?Ps#G|9NyX1RVG!A|O=LP%k#@v$LG-GdT4+_Y2W;Tjl
z%~asJkZ3nQu;|+SJP(EZNkm+IeQ<F-TX#UijAr}X8XYp4fmBvJ^G8-z-r?>{e^6kt
z<#{IU=*VjF)Wy8k9Sw83IDu!+$pt|6iBjiZhL(%cLS;*nMA?d-sYhgd$oo95B4jYD
z@W;q%N0fta>5_snV2*HAFT5z@;e3V+=kUew`Z-417|cuXbr0tjD(>)c*7j(bmUgpy
z%JBuF0z74@?P^;{+|uJ6v-%)>odU<+8+wTm68=igOopQkVvaIt5qTyggSy``VI3fs
zD4*mjHbXPIRv=sFUdWqzvV;MYc3dhWsO66462CW)XJZ5NDI*f@^o47cB&Qi=BNVW9
z?78adQ_vAWso1QNy282zjsbx()^gq(Z06f?j;i`z>pJOrflFtgWs5WcA=mfk!U5`i
z!9*$P+jqtBQ#lwT67jKB5cnxCVOZFe&$F7X^C>s(XARF+{ZsabTJ#N@5Ie3AyARVr
z#x+V-<?4If?pJ9gz;yK3D9QuNFkLA|3hkEg;?F2qp1)@9(8|L??U=b~zo{Oe+pw08
zudH-oI)LwbHUsaW5~hW7+8U<2Gx+Yc#y5g@Xtntt0-dE)y5(C%>X#)SxS<yL^WPab
zIW)Vjy|8{a*7rA4JY<RD+*Oy3)ax3);*Q}2zU76r<#C&pT16hZxn?S`9;g6GHts6C
ze`!guPc6YatXZa>FiAR{uHWLR5_=%ODT~%decwZAvNYB>CYMG$45dtf>Ha9GbT7w<
zNst&4JO_-g`{sZ^hUZuEia<ARw4jIv^Tg?m+d5|!wK`z^dRBM07JjPxRp-*vB+qlo
z#PxBd$!!!xZ#G0k%Hdp7*VU`>Iy(xrsf(ii3#wl+HgIGmVREuaJ47fm9(K-b6Z+bD
z2^II1!WE9cX{t%m&4%;#&i7xL2G7|SC~Bz4gBv>j8Pb{YXiX90AJl%b&mOA!KL8>@
z-M*v8a9x+ptu^Y6CcfukfOe}%t6rfvRHP92Jb3mTNur6Oh#S{$v9!L<orMKjo$h@y
z5Aacya{v(#06q>SEm(NxGU1UK@(&e|zDI1@7!-N8%l@sKxI;Ze$3WD#ow>0QE7F+v
zp<aa@4U>~l%we``;DwMFLa8YV%}iWpOQd{1nq_+KD9z1lq`f-VuD*kp8)0-}n#xYh
zD_{FA=N~@9$j~U0(+7F@?0H0*Qfbr}pPb?F;X~}~Z1dJzZ&Imj((iX@wfc+<)v1=(
z==atc9xZ_Da{JZ-sfMMs4XTYM)$Jvo+^_JR6#5d}0yru8TdxpQ%G96vb?OsG@zXXD
z0a~SqELPuKy1{07jY3%DsmGrqm*30W{%P*qy3O^g7unuf<%uVr=0AM-kF!_}+m$kn
zb{)qJvRJKIT3BLcd>TI&U>HCq{_Di*kMr@*M?U^QMcq%@gZ=ui|M~~{^*^y<19_QI
z{(-pD=g#ZzFuHFq&P*PWC_Ld%BxN?~;r1PrhNw{|>GuhnErw<$kRJGXpMJGP7l+*F
z5TPH^>-X_<0Y{Hb6Zd+w>J2_S4*Nzzeo%I3PaI@<eG#oYMEy3cUXxBIX7BhoSKhk7
zwRhfSeD5TC4;&<yFX4LuV<jKo6BP3~PMti-(W8GJAz*!Djc<M9kBIv%q>MpDRMx6o
zdFL9(4xVLmxk_bYiJ-m6SZ^E^jp9lNH!LBP!EqF-e22#C|B3L)KTn(sF($<*jZrC;
zoibb7n<P=pr=I@|2j=&YYM0rWJ>0%|ojuc2JovyV@`VzP6xJtH*&7Kd>9jkvI~|fF
zrCi>j)oL>u1QypN?P+?@!P&i$J_4MEGL}7*Sp$-Gdu(00LUDGSWZXxj1{cAIQJjx^
zgt14e42@cqcCW|S_$bp;lL$wW&j&Pi8hAJijgBzn1=M<dN`)Lx?wh97@3UF2^XYNH
z+-Q!MD-LP#B(06>m>@>?SBblmB$Y13>HXY!N3n46J&My~49)Mw$rW*2pW%`Rjz;+o
zlSlUP?D5YdrF9i}`@7#E>Ba~&C`E6xO!>k^M&=)-x!R>)UBHcQQHZAL_eSxgOD-J7
zQ3l^J^egKu{NTHcKKFV0g`CB3r3RfOY;0{)+peObh=-r~1SbyeBT_Eg)01r8U7<L>
zhq*IH@xnZz=R*{sl_ed3AZo@$?LIn5Y43FDb|Z?OmzfR>Gy(^hx{TrV5fCMA{{9~!
zdJ#hpoyK!Pr3z14@V=A+<GQ$>OWf*^1_A!$G-rSDX%4kk$b||p2xW+yO`M?|CNW53
zi24zsl=u@9L~EOz9+H@Ni(qb+?ael#)1uZ~;Pv%&ilq_<_aA0_WR$ztZqwI2PM<%|
zXFmN|PMkhR+&6sT^E1p$&QPgU*)u!G`Sa)L_xoJCc9l}8#OTN<+uIww_UbEi+Z9|#
z6ZgBUE-c`>MT+4Fjm<l(FI?l02bDWzx}$l*LJ^(B$iPEKDJG8TwOa&t-^AH?2xd>C
zF=(S09UbL?2hOpvzQjU5qMs&|%VjQJxy*KX8-TNC&M`JNNg;4?g|#i~_1o;!w<+&b
zEyH=ILbunW)oy2cUia_gN5^a)tUfSgg-B2cRJYH@kKX3U!8xM61;Rw&<FGgFu%}Z)
zxXJ>BYE|e(&}hQ!1b9Hfg>nOoM2wB#_&%l|Bk}=LM-%8OsQ37bdBwpJe7(7c=EPx^
zHWpB(L(=O}j~cW)iv1&_Tz~Tdx8J(J$ox3dhYpj^4HJ4HlO;jmLMb0|`osy29{me8
zR&TEG?Qi@6NxzOG+ejs7taiBe_AQR@f0*rs4$X}P#+sKoIO<tnQyj!_2@(S`#o4@o
z`%;4c3;zrKVh&>zBGrgArdr-+tGq!P#k}zB3mlr;Poe{6XJ)x`^BS|$6FhYO82S9L
zjn&$aDu`2q5iXr}n|8OG#p(*JW}D$KKqJu7v3|op5vxB6Ox?%%_~#=Zf1sl7gSbWf
zEX!z#bPQT*6cV8vYPZ+fxp<X*V`FHU$8`h~MC6JADoJq=1i2Ctm$=cPDJ8|}2^=@X
z^Aox|J9q+$g#t+&<G3EaBk_V9-Cjb!-Qb~8%-n$@e{?m%o8HG-x668EonE_&))4BL
z@pheXXqZN?$Lh5O`o71|)HEwA8%#~yAP7T*5L{1s<cm3^gri3eQQfJ~YPDHi-C}Ed
zhy907;VQ%C!rN5VS2$A$Y202TF8Tyv9%(c(P0%K#*KSi^yh|ZBMDEegAu!hOPp330
zJ2ZB7c;wt8oI806N5|Z`bCb7Tf0ei1c!@A5@IU;@FLQTs0Y_TUcakPdOw7>j#k~FY
z8!Rp@66Oj#_4tznL1^tv?i<Ge?~_9KC{BPzIfy`@r9vAAp*%WE8#J#kGW~b~EqpvF
zI7rGwF`yr%1PF@x60UG)*4h~9FgZ0z;Dr>zn4Rry0wLH_EYgi)e9vR3I7A^Vu+xjF
zwVOOyOgS>=^QBt}%j3ssM+u$Ht0cYmFdc#G$An24S(u^|Mbs8o(Vj~%HbHBBljU2t
zaf1Ma!+Y1>!Sh`ZhVcXQw0F8h?H=vTP1-vZO7o8(j6-+p4f?C^F+My-V{Mss=u^t&
zaZxxbA&nH(R)eKGw~2CjMxTDcUL*ymlzOE~f2YE+)2Ddi^idqu<Gq{9+<fygo0ng~
z74Z1y{u)~=>o|S@X@pWK6T_p#J;l<+8#Fg><M;*kKlCu3@7iFH;oAlcoI%@Pn~!FX
z<DDHY{o#M)aMC6`w4cZ~NI$phSL3=aUM@$kUZ)>*@S}je!;&F4u<T!@kO-o78((X}
zVxB1N;R#99kMLcWk?~P3US1}r6V6XLtW1t#@<-WOUSz$oN;7VArPTl@V0d_h*a+Tz
z^Brzny2#YjG~?3~OdpshYDc)P3rUm3yVvRWBbwE1hD&*xjWR(1M~@z0eZ{9<Z?L&h
zBaSpvV>86t%d9V4B~qGx4DK+XW70ID-x2t}Ys+S>kka5qJ(63Oa3>FemqXwXcsW#@
z@aikyMGA++xE3HCJ4oYD+1cjeJMS_)G{WXqnOGTa-d$vQWsPnUqf(uT?tb)z+kb9p
zJS3VFMS_PzKzCt_{#%zRO^l<(2)+ZeX#=O|BI4{V=7)hURJ$N+wy<*x5GdF#0|J@P
zK^!Ao7n~tXpfG)n+wSprvCiHp%~#rj+~hv)^dnYkYqTpZjD}j$<8W;YZ)k{4r^BTy
z%QQou+@4uhmdi{{-XzRo^^LSmz8E4M&9Nf~sFtg=+HF=>%WQ5}m_KxmAl6h@U!t+G
z!YR)X-)xbd4RHJrObSU1Iz@L{xC<}h7bl3G`JWKP7$ra_G^^XR%R4-N`Y|3laRQ+`
zEZn}qg&)7lJ8yiKoL}TG{PGvLv$TLCCED20+~oLSy1kUQ-+7J2r3J#=5YIgJ1VQ-z
zSS{c891mjkM}VpOI3NF9<l~P_eApPnH@@)=e*3q7n=gO)%RkfaNCciM5h#SQ-fnT<
zu>9&}#%K3XIChxvZj=2+<K>D7=_8dDVhr*M9Yx?e80F#wISBgriA%duBR4UI@EoEz
z#t(c_1EW)uY<BDPx_#y*hWKkw&hg6aGGDu`2&VV3-Dy#2R4B(Woit^#*Jp3&GZh4A
zF+{&wWwjM?H}j2j10O#O@WKG;2+|}aNo@DSF@mXZ4C)P<>kF(_w;9Uii90Rw69u%2
zQK^ebV}$WY`zb1l=yw`8*M7k0iHC>_<5q@2r(hJ*6XTpZc9K@DM!VNyV{HZBHJmwh
z6e)Z**4EkH+{AGldc7`}E?vfTa#U*-(p0guw1hE|#~yx!PPap?Stp8P8z}51qEfYs
z!Rq(=|D`SbFD@Y@j+6+jh@?@VB18Gz8;nj(5FD81U=;B*P{<b%(#6S~4~7bgu_(s4
z5|o4RLQK%36npIKR2ZF_MEDNPIHDAUqy`R8O>v=Dr_$|nXkv)J^~gcKerJvEtR^_K
zr|34P=ml$Z`zyq~3c1!g#ljHza1z5Ps@kB@?o*Q%Um`sZ=?6HzZ!NlFgGvNOTaa|l
zn}lASq<oF`+D-i22&%tDI66(5Cd5gEN@ASArxW*yqdv`cli=!`95{21R&j#VD6~qD
zCS_!z#L>h1+1ja4>$O;0+d;aL@q;IjQnIzRLv5#q$Q9`~YOG$ngK=}T>s_=`G}o_U
zqKNqi&RV1BPLDKBF-BXhr`;f7@6)K){m^?Hn*_H~;m%jT#^m!)Q#kVguIJ+mD^lx7
z2hVZP!a>9-W6e#*WeX_<iP5;Ol?`x0kEqf>D@9PuBY|40!Gt@6)G#?*AnJCQkTLmL
zk8dq5Aok2*3I*;gTtoC5Xrt&vEvoenqd~}lp<x=G7UhLITzKVGV%NnR8fI*4lA%Ho
z&$AkD%ZrO7X@bE}tyJmv6-Gl~pcMC!l|@#|o2+zt9G{q^*laL5Tq2ITkjvo+LDcTz
z6>>!V2q#tqVUDoVAgQgx_&i2w(zwgc<|bQfi!>TdlqfPeb%@FFeRzI=6pBi<#!jV1
zoM;dbB^EdLf97(K_(&X^?omjRR8afDRYqoK(PxiypuWXSzYDn`kU2<GPzmG{==Uw2
zE(swN5cVN<VW*0i8V6rO9E0y$`H|^iOgBPwV&<k|{_1(3*Os>U533%*<bKwBEy~Ru
zR#i&Fq|8_99Bs5Y?n!z>!_>B1%C$aAj$@&GL4X_jxH$`pNMl8kN{ljiMl$D4;qUCA
z)^4!gtTGzrP@Nvd5{NXm?W;5b?Lr!1qKKr^CU@yY+%r$pD9u?BUL7NJ!k)2l&Kx~W
zt5&DgtFyVbO5h}%J#_#na%`-vQ{LRfaa{VnE*CFd!F7sMYBkbSv$(X1F)oik^f2vi
zhp2TgRyzjJnJr)DJkaZZ_*ngMKK}X0#~-?=TUc1&U;V3p#qa*^?@}(8dEtc@e&*lx
zMM@|fI@$_#Vl)~}QfYJJ<=1#<s6eT;j!8NwPa>QMqqHp!Fz7@<KPBzQ$byITeT*MK
z+9zta5cvSdg}B?o%NH=(r8F{0x7Nf&2@aZ19~oyT6#UyOYeasY@qGtqx0{q3HI}28
zwKyU#I}BGF7EqlF7#*2le%}Fdqa(CyO-K!-8YfM070_wocI(8gI@`@Ax4IF-`G6_u
zGp-^gM)IhnhX_LY?GAq66UUiGnu;hSF=}%W4v%LnLfE@^j#_h@n>Q{KD@Cd_s~c;q
ztgVu2#qiKD8{4aFZEsmIUt`$V*kEI7g+j5wzP<Cf&pb=F+oRE_Tc%$>=H|_tT)lcN
zbLGV9O(Cr8#z%1ab7YJseIo7JQ;yv^RHMVvTkmqZ^d!@r1}2F}eTj5ZL~5<3j4|jW
zg|4B~i*bv#z%qV}pY~`r>xg1V;0oHE2Dw5J-?+?;OtMpJp!+exDE{K1J$ShtzIyX6
zYG{~n>O4t%jb3$)R=q(dYU8p^;Dtzk6mlW?(j-GOhjB{7B<&_fyBH;KQ;k#t6Sa_0
z1=X(5Ew8cB>QE|$4ChD5Q(>%-BTjn=e5%bBr7)l!^)oPZ#AKp~*OuYL_$~`}cF!JG
z>lLouy-Q-OWms*!Omk};Z4^Oai28PeX1Rt40=l83xxGVSy+Kg&8JXRK<9>mp-=W)x
zz-<xtQmS|Fuyf^AlvWx2-zH9pK<N)Zw$a848|zA&<!}9vJ;o<^@D!#vL=;5`;o!L*
zzT=RKn&d<uCl4k~afC(&Ax46i3yE7zdbKJJAoDp&g%a&%heAHb?A~FFDpIf1m{B#p
zcygF)oj$L%ddNM8sU<0^)h(LUCQ)L?<0M^XyE#fSU{C0w9FJIl>L=9lc^ucp4SZ6i
zaD<1J4go1Xg0x*nS68Vun%r)8*-<G6hl=F8P3A`obHj?L*TIa8;z$P<fr}(cVn__i
z7`&)Tyz~~j;GxAZIZyKZv(K`&w!v4w^*yxc(lu*rH0pG^ExJ*Yb~hnS45|5V-H-dz
zSB^df6b)Q70#Mdesor7b`#<9NaE`s@_mJHx!UvoPQk8uG1F^Pa)m{Qa4g>-C0VI8B
zHJ}vOY}9E$p@`5fdZdV{bdbG>kc8*wJ;s8F|LsnLzB|F>+<w}fCY!A~*W!fRDrP8#
z$xehP%S6S1(cww<%^$!UFVd~H(UA{mOd6w<#BWt8w71c%3hS*N7gED$&gXz}n8h$X
z?4S|_zJgW<0^fEOwSp81y&k=_t8j9!)kZUdy?ggjYi)4z<~u~nN(rrQtg*DQW@GiR
zWMg}Yt*tE_AxVv4b7O<e$z=+~0{i#O<3Ifj?QWNPqlR=O-G0K&n>Tsy>UFfxyK+22
zSa%xp5l9n#oR5D#^6{UksEeYAFMs*V{Pu7EHb4B~4^c|-PygvZ<*)pezrxVa(9irj
z12Ml;8x#%-jS=7(N$>6k>cYE>>@A@JjgAr=EikUMpwY~tC2Dq1u_iZc?Q5F#20`Ev
zrztX)C?QCCeWVxQ2uYaF({FWfJeSlMTGbW@e22eue4IbHw$6>R!0|&4OwM2&NvqqT
z(r(g9V%&a;pC*Vj#Y2&<yXbb8B9fteNFrdP-lo-S6Gah~R1%AXNzY{#7)k1!*<Zk?
z3k23A*>O-&LaI{YUZ0$HC<Y;>Rzc{*`jlwswz|Cg=8vhBw+IU(_+gPst-;#HI_1h1
zxm=#&GbMDA;uwYR2qwoy5GJD4=%bBhq%=%EEYfavSYO*@vr=Vwd71LgPBzlHhnoMW
zSaU*HEqARoT4W-r5**_Yt#1=vz0IM?VN|H-^<#W3(5@uRhmZ*qx9csESd$yc5e6P>
z^%^6APdiC*ql8!pI^7<gAK(dxkuXo)?GX4LePP(FH#zLNJa=%GA1&RW+z63wfuWgm
zAOvx*ORv2`9M^GrF|r##w~jFqz2#tfZ9L)Phk29`bQ?QF-7ZP9OsiKw=gJg^OANV1
z!romD?Arr$aYO-MIuu-&xF6G1DXmVAAtW<lh^p3cR18WY6ttUNRxVzqS1aS@Mv;Dz
zZoNULzD}>TitC2>;b$RD@kE!AFrqLr0wnYs5yoi3;t*a~AZfSR-dLki-KMd0ljg!z
zn*AcN$>X{Xo;Ku_-VGifW@#fcF;;`y=&}6$*YWc{lV?tYE9lb&t#Cra=uQozJM>jZ
zRI8)<J!BYCni|6k1Hz#Ky{#&ZwM|MBV-&(1wN4)+CHbKd;wWa!%dxS#f^0Vbf9(Bt
zlx5d---&<DIXAxiDtB(6yU~pd5P=cQ0!WGyMUe`&6>Q0}mI6IW9$O=svOP{(<DVo;
z;#d|%QHi30N1_6im;-Xo4WM&(<*Jw8jVJv6xCMZer5VBmC<+T_t?pG-@6~;;-nn=0
zefRzbZkU_n+THv3{E0Pgzfcm-FVWNU^v44x*@%s@WV4vyi3xQjiI7CnBas#_3|MT>
zvirb(lJOWjoDd6%@SvAwoZ0FzC<;<-@I1jD-{T75IJ%&?d_h1tq19O;3?rnHcv6D3
zctVgBQ`)A6C@g0AEM=>O*mWbyf*(elUfX0V(^#k3N>`s%D|nXU@Kv>Td@(2ijY1=_
zK#0PgT&I2e9YhNbVh!1je2FI^4tG4B1PrECao24@7*}|^(1%h(TIq|C77%C<9<r{m
zBLzW-a}K>-a18jteTMg+S>c}bh#-tOGBb}=icvCQJ)N*!l!V3+Oj7(}oh?_8t$C<X
z%8U`TW1mz?&W$sM>4>5<EZcxgY4%FNCEBptj<{hr)SkXRqy(iPD{)T4G$$@Wh6=U0
zj=ChrIuDE>8znq+*BuNtHwbGpcyWWR{*;xqRW^GY#8HjrO>JCJ;@g5y!NN=v$&~SE
zg0+@TqfISrk_=PUmLF%i*XQiH^K5QyW8IF;!ZUmGHyGLfuk7(l`S?YVkDpUfckjLT
z@_`S0fDeE8!yG<*m~a2~Z|B1w{xEy@?&X$SZh1}x9Xu-tYMrY<pa`4*D+ICh*xQi!
zp2m#Fn5=pZs%jIf$Vd7fUL(dhgA9C}t4x>Eq9pJ<L}{_Y;dC`JaK@s1k5;!sUKDI^
z4>@yYlepPre#Yb5j?HrWnI3m<jo9jqv7Sf0S!dzk0Z@wJctk!;@NjJR2aJk>xHPns
zfT?2DSe{IBawqW=1fF59lw8{kxu_{QG#As9DZ+(BQG`;G&<~KN^2kfa6C@r;4Vxz+
zJ&IM`ow3c7d^+arsV69nPgt90c5Xj2t=$w^Mz6QcXFl~Qio!COBp7G8{qFm?@8O5Z
zi;|)=m7t+{*8JPpFP`6jVJ3OLa*9BLsBj~8$KJ+?O6ep-aart9_YK8xM3GrE0;fO;
zfggo<zDL-sQ#y_3dst`CIJS#|mggY~Ln*;(gHI(c8~PsmTWyAA!RhgU2i8_-MKxxd
zF*of$$YU!fIXmi;W=-%s{4k`p_b4FAr+rK^K-Su1<32?)!JAC*>or`xMpC5oS63-T
z0~>op@rZ^9*wvh6z6SH{y~HxdbAm<~5hzK^4^XDk<Jz7~@$o>GuyFx5IY+5xaD|~L
zCfIDuXzd=X?V^H3!p45WXg8)v$Olhx;qE_0r;g$2F;gk<Z~H4`Jw)q*Vr(f(LoUXc
zJVDx$yztOPNs*Vh_5^|q1LOrvlO27lFLSb2S_l;cNIW4KZS{HNFF(a~GcDTt4kC=i
za~^HqW5H<Rti;$Ib8aEgqbWmes4vYC1QA}X4t7GawME>WA#T(v_Lwnvp^p=S*@am)
z&tD*$6htJv;ouB6?CtW0kDcX{7seD}#9Ze9!jufNguKW}(v$%PS>#BdUTTKcF`kUb
zjG<F#no3YA!MVI(y>NI!&<uSVR`bsNHD12hgkp%vOTxHLJ8lws0a6K^wFqNqH`}aY
zD0VEN;=(dGcanJS7*00X+Uj%a+y%@tAjfC^tJp+eZ517rBZB}eVkhBOH)t9`U3%<_
zbA)GMFoaZqla)3G0wTYXFKNXfG=yG-PZNMHD~pxFf^<+;7E6>C;VWb}fU+RpvTUBw
z`0ap$GnW7TfZ$`xDSt7@SV@MIzDJ|gU}1hQPALYHG36vd8N+6OL|>P5O3Oays1M<!
zf(J`Sg2I=A$Sb%Sc$4qAqE>Q67iJw`9AQLN(0Ku9hgAYI&MN<AnW0u6BuK9&_x2);
zMHfTL>6qn{XUNPs{MrIDvqzX~9Hh*0`r89O_sP$a7mnUEM`O9|o(H+-k;ll&lA@@z
z=(N@7az|(Ki?CMD5}x3BjMXpY;}=CfeojT*%U}L--t(UKaK{~YaP;WW=klw6<BF7`
z5~nSYqB@N{rvQ(cSxGyx=;08ZRmROi2yE%dg(E0qL==H9sMYF~9+I|TEIKdnr3cpG
z1wNTcu)07AAGARV2<s8M8a290b2ufj!Z4nUc$Lu{Y-AjtNbXu6GVbN1y&g6Qi0Vz+
z%?^GL;Ds?A5$v!!GX_OLImw7+pRg#Pw9EvO?>N@r(nd&`jxZ(%;WN{orxw%*qo|tE
z3y73YnvH2jF@E7N>4^N?L)d*UM{1wlOY_`t<7@fcU5}DxC2f(?a2ksxZq)G8A%oF~
ztgw}}qI0ZnZGYK&^yRL=Q`Ml)X{4-Dg0u>pVzwh#@C|9dN1kTLoz!83W7KyvYjtET
zMg)TH&K`>$;msl|X~aIlT7n>;uN_616N<1J>q+R;LauJ^X5Y*_R!It>SsstU?qRMu
zWn-#1v$;l|3Cw5}>w9?7B4O(wB8ZS)M9>K<>m`*_=mIky<97vKmf_NXSP5Qv)dB~a
z9i~MeZKnuo?62=)E~wFsW3UxQrR@idiwTQiNG*pvPZ*wj0)60W{376zr7jn)x{h;q
z+=edt1hWBNyi{4Lg)wD3LpB+p#|GE*nQW9KgD0`l-~^Sax0Ij?V8<=N`50YRK_Mxj
zD65UpRO9gHAtPuU##IDcjRPl`tPVKynLD`bYa6KfT^OloDM7m(BCW(1J~C)R-Qsj*
zDdgz@=ljSgg0jGKj_KAm{%i~3DRh$Jd>=22h-{NaX9iJP!XThT6BGq+*wbYu6nySX
zk5gL%CP3uJw3>C)T(^>Vk=2M)mYSO9V}yq+ErIV-dYaS<!ts>WaDdATd~12<<x9M}
z1C#zHQV6=8F3o0(X0wG*3aJ!Umv}<p`##amC}%pI5H-8d+r*65pc#`-Gm5O(8C(7J
zBJQiNpaZBO5GW+}nLb^B2L!Y`Da|l}ehO(;6{bQ|W3YjPIEGpb0nm&gP*510wvZZ#
zy~+hE48UZdGmsG!8X+yzJ))+^-rW|V040egId3W9vH<QV1Rom|^tOA9dVR{k$FDc3
zHQLB9#0zTd5j7N!Eyk=DIfJAm8V`u_0$Dnie8-RPD|xvOHUXOhJ~Y}?I*U;NMuSiY
z<&$Sr#a;@2xJ7X4b8yku;gxmvF3ocN4R7F6cbz853)-?EMq`8_Y(}Weu{|D<<&}`H
zb&d<${lClY1zyU>FP41#oQk?Pz3EMS-~%6EV`GDF|MqX^mRoN5Uv`rKh&+vT{*DzC
zJl$Y)(z17H1~Z*dYF)Mdl>%o`uEOHvS&Ax3sC%ebfvqg3l;@F^nkcMV(w>ljpv-eb
z=pmJ)&?Ul3f<_a+-onY94o^_09yjSWx41Eixw<#ct)~WzMw-ENOq%zw#xf9+JPwe)
zq$mY7C&6f>D>*16)|9koYrN~&JagHUVSfY9Q?wdY66ka~_+f-nmF`W2IjTVAR#mo3
zo^ec0pCVeg94FdnW4P<#M_3t7EC0xJol|R@)p%3xfZxquzlZy~U%t{<=k45qIH2lS
zIpsJoGebHV;g{7OE2R+5qAWOFGD@ceMF9<;=FT2_QWAI`gR;a#4y+@R3Wp%gGF0s2
zE5)QNkw(&N*4R7OK}$iYHQk`WQmx6U{uX=o#yqfRH+P=hCLIaVw2vuQvE#E8evE0<
z5T3vog>T{=p!9&IDR8EwH5+r?v17z8VRQNzzN#}5EwK={IN0h^i>eWX%?qUPsCyn#
zSd6Ux*4Q}KPcPG+zZfmr<l1uL_`|Gktx`^M{N8EIiTgn*TwK`?nZCpp4RWtQ7t`uP
zeg+n&tFhP4#Hdo0DxTepa_8Ukyqpslrx4PBb0`$Huq@yE6s=l=BX4;<eh{!LN$D6U
zRfy*#w#X5#GP9N?=>7yfg1B2lhas-i1c5-OIU?{X3#=rCnnD1Lc8jDxB1uzLR<>AP
z?o)3!xNJV);N@K&xzOj)ZNu3#rQe^RcdWrg5LUQC<uQ^4cpgefCP~72Szz;w(3M=}
z`?yeY)3IG#=9i4lk7zG;>2w><FxsS$l%*!B)kvo)o^Mbxq}}N<SY2Z}9MWue=~%&q
zQ;*QTVlTbzEz&&atLsYre_u6nG(MH1X7#^AVlnJr2$AFA&JBaBx?0Xv=V4h?b}3~E
zO$qfnn95cR&x10p@OA766eA(as{&8TYBXykqFJxJjFo`W;QJX%(Ufju%GIId+DVO1
zoQLtmvYjSOk{$-w6oOGL0AG@qinaq;YCKA=b$~XsI*RY#FS&Rm7>_GY;${PCO;GI+
z!l3dzEj1{G@DxT!oU+yTrNupU9JTu<tP3fP<<>_Z=gc@mTgPg?%@dn<?2I>enCG@C
z=${Yx!}GOX%EvE`eEgh>y1)F(zvTS+^L+4wALRSK@B2v8lyCW#ZvpsUD(<YTOvaza
z|5b8Y5>FUHS)hU#QI9G-#m-}@s5E(;5NI5}6O?I+w2mMOh#PeV>zgR0D9aLyLn=j?
zX82JZq`;F3otJ1*RNKQi@GU6ML&Y^PU8Z@)b$eP|+zwdoPtnX0NRND&@X+~n_TBt-
zFh9!!Pd&n1qr?2hI+F(;q^BKA2X}K-$3t&z;=G9Y*+tsz#xtIiQ5aM1ScH_e*f|j-
zQ5aHYIa!|3h#J^YAF=^1ZqV<Kxc{L?DR+Xwj&?%z!I|elLG}e&R#vzz=bnC#6;4i}
z!R3TOglxnZ=g3@T)#T@z+!&0Iq&VtMkfjOII^r;(+iY@beGAX?Nb($mBlHy0B&S|$
zB9tVOiYzZlMMCI#loDc7Sy=_taayyH<y=4CWq%{$u|WcI9$!e940wF`3>Urrt=PGF
zj-Pszdb~)xzrygzdq}d9-JOG6G}~aDtRq#8z0G6nX|yZr9~_-HCb5pt^O&X?by@9k
zD-4-t1^p~zSKLC4M~HNYt5vw03lHAQL^POqaWy0DjKE45)uEO`Lsf}z@a41Jp06c@
zKnZ~-1zNkxI>|Z85|;0N3{?-g^qR|1+sovX2qmlQ**T=|A(dZ^S0@QBs1)17T8;5&
zgeOEroY?_mn^F@AQDOL`q}HtC1T5~JWp>v*D)h0+kftf27jSGor3~uawKC%I%>tzr
zNjgEvDV{b=9kh-dqA}NDb=>FT#Y;G(3#Jd;M`8_!F4@O!F+n-U?!&XxV;?CzlC(<j
zNtvO%5a%pqk>dLj<ziZ`4ujP-lIav99E~`_ZLZ+Vl&}`y?T9r0eJSgVsD#7~RZ@m(
zBpZ-Z7Y0HJjqur+Bqd~~VhSn&!T<rJ)9N`<0L?nAjX-&o2Q02)$R-8ERAR+aRv109
zst9y|0pTf#V<Mx`x#l&CaB0J^F>vIfMubN(G~B&0C3@S<WJ`10b?On8YHcn%yNrL|
z<E$cC+SlOKp+lXYKmfC|g0L--)?%%w3Qp^)_vnfp;F&}Ps_HW4B}5X}@8PHGl<_Qs
z{)GD;euBdN4YB%*)V{yEcK@Y({Nl*R&zUH(Z{I%N``-8R-uJ$jkA3W8eDH%G#PdAf
z@s4-!?svbNcfb4HyzFH!`_c~9=ui$u2w@STx;;?V5~z~S+zicKyYL$UX@5wmGn(B7
z$wnWUmjsO%>Jd%}l&J2SO=$??8tu7Rj49DYjtZ+;e>q93E`sM{!w4URDRPp{F;TOL
zmOi=Bn9|@%i&7Sy3~BlidmA2p5Fr~a#Gzfx9ofyoYp<hw`Kw5O??Vi;jNS7sCdn$7
z1tAS%P)f6Sct4O62Oe5eX|Tl+Ns(9T(Kw1CBM1Ye5Y+1phU*)olPQhX46!vNTgwEy
z7Fk<aXJfOs69>LZ(bRv#l2IFzPC*!isI=aM1z!~Gn(4B%XNkBT&>xN{OwMAb&F1<x
zDlKTYYRE<nD+T@zgGg&bqgLnO{4BXHNXwj1SJ=C0nt%n*3-F?tI*P2w+1y;A(`=9l
zpP|tdg~k>R-?Jps5p@-CAogelA+phia1YJ>7jeaFZeZ!E8~B9}MNHD1rMUoqy2YFy
z&~Q1PC^)eD5ZY>H0!3QFq%4@N2W;gz%{T@ulRTpphWJ7-7dJRN*`zm_u&Xo6434d>
zO}t%mtgUS^UF#8SJEU&nC~$UMnT3}=i~(6xi1n|ct(Q1_N<5+1K|~>iz~WGeW##q<
zxI7dzI|}V<#MB^cW&Nf+4<Q9!6ygLF${{?35EkPcCNGG?2xqEmH!Cu9k)mXXCndSj
zNKeu1cEAs@JC#x`ENInR42M0WbsXJ)lr!r+`r{PT9L+c;FmPdWlf|2F;iA`l4Ilf&
z$5<bY*gM~(KV4>TyFrwugkgvIxi)2<pk##;%Th~grp+Yj(`n6D?7tle0wDxp7*Uj(
zqCcRuzCpXwrsId8!3%46zW3F%h5w(cXm^o=0Z;dr1OgvRq)OtsdDz>5hKKHr!Awyz
zF|73wNePXh!qa&Y*fG8@4b<we&_WbNC1+y7s;EnMI?i4IQ4FrEZg4iXpyh)Mz-mNc
zak)b(hnP;7Rf>AeK@ebD0q#glYtn+(U5<D4EsTHnH`y*Sy4{F)e1glofKK7?WJ%|!
zj}jW`S7YNm6PQ|rDoR8gR#s<u20;Kq5Y#=yS^?t>T86-sj5p3v>^;od@&+55{T+Pd
z)3N&LcKTOwDgRPFeo^G(FU$DxYrf`d`0d~RZB|xRc+Y#@!=L~8pL4?vH+;#zzyIa0
z#wvj)4H8#Tw1LH43miRmEpcrTDFQB>S|J-vF{LJGH&CrQMp)unM6KPzZ`SeRu$skt
zKJ~dSPJ&Vbt*g#}vpZuR<>NK#ltA2Up`@g@wn~~z@WYULvq4mg$<rL+Dy)Zf8dD_H
zY(X8vY%O5p&QJ2>=l_Cwq^K*w`l-`|rerSE%(Z;>?A}GY(<by|qzv$bh{+fltr^D2
z2<iD1=)cUdJF*hBTBFjX(Ha#fYMx}cdJbbt*49=?vW#a>oUeWy4#1HsudDV#=qgyq
z>aJ$*{5)TC^ipQ(vv?xlp|j^0j3!L;f_l3_)M=oFrCAGUHyZ@?`ZG3E*7I4KpTS88
zr6ezRx<A%pwZZdz;#QLhFw<-i2+5PHYiuPG8d1nnyG6GiF-Z%gQz!$*YSLm#S7$6Z
z&E7cVi92rNj?aFYC{p;I;KJEeLOW*8v+RxQT)ucQds|&veuO6i8bQP)Gt4#SSx!bf
zrp<<=$kDbccjjtMN~0-^CI~!ce98LyDy1pd+*qfWObGOdsxLKx)K+6Hw_On$3MzU3
zucD=0HByJFsqwUfMTx}<ftOesE=MVY8&1&UDdjjJ8;nT%1I#4BXbpiRZpI)5QYn;F
zlzCCv<`|2V3ZWE5l2v1?op_;>3_Z;dJDOdxEKxh7o_g5C1tG;`i@goOMKd8sx*>Zy
zAxjs}bICPFQN<R4&e*eOFM)Hco<57n3+g6ep<82Np-m9_&q&WqMmZNQY!cTZ^1MJP
z^^CQT)+GY)JRj+KWLe5|yGOFsr;(?a;Wo8)hi<#IbJc#e0Ve+D)qV9FC_Rr5i^T5W
zU!l7+OY7+M5FNmYnCZzOW|&vw)=nL|HINqS0d$+tZa^)p#+E^Vm~Vr!I3=p_rYEbD
zNSmstYc;EGU8`1|V=Gfg(rP^0szW_Orw*(@8dqVW@`6S=B{mr|5tN_(V^%)>2Q(vS
z3rGK{Q#49g^fEflg7%&|UblwyWM#Y(c#I|ieyfSes_{J5RPWDK^cJrcU`w#tAp%Jx
z4av$Wtj$?lUtyBvJcEPyq7my?&$#ukub1-iiz6RDr=o8E{{3ukKMVHz#b5kIMxznk
zZkPAG=RMqi|NY#3_uXIe@%Fv(7TT8_tBl1FC<IEvWydZ?W*K@iWwN%y_Sy<QlF3$|
zTD?ZxYJjt>pFP89WwioO3sIdc0`PpS3fS1}BdKQlK-DIe0^uqAC?qu{joC#yGYh1{
zK1qKQYc)Z$PP;Q#y(3}C^Mq+WAxkE>(TJIH%3}XKYoGW%CTC8vwRVp2Q%|xKN%k)+
z(CM}bLV+<FDJ5B!bLy!z^1L!^w$@bIHCkg#b=NC|KuSrHrc9?(rsD~rkc76vp^dXC
z#(eb#guiWFcGK%<?>mfCl_|0m5-A-wUUU#IvlP>U&GmK8Y@Q}^itX(_-CCViyM=R>
zM^B$(eQlL8ODp`flvOvz4{%<<6a5Vo5?@wk$AvMVst!;ijv4BbU2}_EG&|4saKQTZ
zCdQVu+f5F1W{^mPuuRK@tvq3nOtHfe7nc(bZm)9u<A2KJ>}is%b7U7DV$rK{{p=!p
znjIn!a;@=%U{qw>zjB^ZXC$S?Sc5MmS*a<Et@c+4d?gtb1Gdu<o8u8pAqic@E1c$I
zY%#&T5Wu$=aFsjE<|+wP-3f`nIe`_DT0_x_Eah;7o@O{*BAiATgGn=rB%vHlD5nWd
zm)KGxgrwGLP+Ieh0k+YaAdV>}3D#PKQuskY=!dv0W4gHt$q;KwN>f%cA;w_R1Xo&w
zQFua-rbB#NP%AR#rD1OQDeis$e`4eQ&w|NVKYfaNnKRdF(CN++1U{aph@zN5ug|Fy
z=h?k$4()mf=>Y;$Y8(!2O*Nr&RO~*f$n%_{ERd5DyZjh+3T8X=Uj`oeizO|-6082!
zcj6tp427utK@linx%luNRFWXZQ{2i0@|7N*b-0Z&qF#ee6{|UY9?q>+lxrcX(dyIJ
zu^&-v8l2l1w*nYj3I2KtqOj5&oa@4D1Ns^C24D+l)$lv@>hlnmqIBei!6Z4}U_>YF
zabWdv>_>kEbMi5IYo~CJpXQ)fFf$)f>w1J?jx;4gg2^naCrYwHp{&Ljc7%rw&(0+U
zNJ)}HIW@S+6yHItOTaM5CYa~RkM%3LUdqQWj(q%_#;+GHT=*i7x&Pz;_&<2t+urs?
zQYK&W=oL_V_HpcOU&mb+mJwTHgz%_`(2i1cxsLN-``j6tafq-MJ51R&=P}k&j3!uH
zU<PAMS`c=+D6axu3Muiz5a$HraYARVjg+cN1X6;8pca!3Q_8X+Zguf}kHPjjNx4Nh
zGfS({BA1FJ9V2v!^gN6$D07%*mMk9=_a@Y$C8TsL&9^yxbT_^q;R}WE3`JV7vbD+P
z`Ub;+&$Tb#%W$$rEouV}UFv77KZK|x5wtN(21BOnTQoaO>dg>iB-)9;MsWV>zq<!e
z2M==X>)*s@e{T(j6A(UeU}<^<S<%ON@YM1tx<RXw1DGb9EzgoV%XBoM)CGBOOtFyD
zo|(n>{mP+2N}@2v3Bh`jurF%kD^+#Gq(n$Tvr(hho01ecbFEowp2wN34K~)dXwS~F
zx7A^!Jl3ZpFd9!Oa#J$4j<w7(E~W(IAyKrzL>YFq=Xv9ygT!7;O?W6zGfH#rI=9K$
z!8wv7;BA**%E`$pGhrK?BP~lx_pDf`v6jLa)<;9umN(cl(_&`#EV))_r?5_91UOgl
zWowOmg;qkP!z6@<M}{XVgMg{7@?fJ8VX~C6C=jCZ0@hfh0Ot(3$-$N=&&MlroY4@+
zgtZ#=b`w_^aF!xZ@f(dwav{(0>n&8nXRL=ro{#4W`Wx#w<rB}$62>up)WuB{qh60V
zR7{I0X+Fi9R={^D6-`kxGwqR13&ie=k<xJR$U$az%_Hz>C`YL*!`>#7?Ewv6vVVV<
z_3;M27grs+yhI9FS^XG;^nB`#2FY+tZ*!B#_X&N4KKmdOuSQs_X6QnQzb0c)t!L}K
z9mx9sBxLuMUD?6IEWYF2oI8IWF`OWfw0y%{umxowe6VL$X~ZC`h2FHn$vI$<Rvomh
z6p|cfW-E!4r?ub07~@8iV?z+D3WH}WkIYJYCWgTTN(rrY)y>-;K`}ziwV~0dK9(#&
z6%OH3b=Go&%Yhsgw^QcpBbH;u(ws;8$UMUH0T28F(t>g`Wv~X5Asl#xpq%#b!l3%Z
z%aR>_2#CtBwX{(7HR19$vRk7*7cw;gCGwg2@9#^ThL`g3izOd_S>xBgDHuJc7xx8?
zqsQ3)x;NrNfl$zGN6gIC@VXvFHlmnJm|vX53XLC1^mIZ#9Dp+f(#I7US-(d)o?wfd
zvdF<WJTD+~aBh84QLjbC$a2o1ghB>BUKAso!@3etkLfJ#Av93-x1lTvYIUMo3*W22
zvc3=)laVJ=ib+npJIkIpU{?^+pIzXLiOH=)*pkv!f6rvl!)2N)F5Lr4Q<Rn<j3|o|
zQ<jx(kEl)}s}0UNe9uFhf;36+rxVH|!}9{1kY6ye`TMbCBxoKv#GdP32|fr9+O3HF
za}A;yk8w6&IGu82b_p#F^-z*d#*F&|oG~<|2W3I8*JC;wSJo|Mj<Xie575r%iT<QY
zsG<rKIEODJG7JdgI?6e;EASdME?(M0%fh6$jVKD5^*XbS7Bv+htfMBNFgcSfW16MR
zx8_)i0v3Wg+wHwPtZNL7Lz<i^5~(EH(`{^NdHsQdl+H3K9nC0WS`_3(iKf!5ltSQ)
z!C6V5A~IXDJ)KfZ(uxhKLdJfBkuPzq0F%G#?fa5esS0A>0G6k3UIkRiPgp|faFobq
zh&SKFbz`Evv-pc`jI?;ZqCVfoYuAx+gz^KdHdvu(FD>CWDt4qUiYf^xNty3});&ek
zs8bk&s<oJ1+DD)~rd#Wz{cWr%snwgz%*<B}7U2-3rb7H_$|9#fPLWwo+h!azC8E*d
zF|WmmsiQ1XvOLA)f-tOOtY+WEyD-94U0UBG&vVR9f|641`0NTi&m-_WOre<!#~1@0
z*(UH7$a4LR;II?|2`DLvqku*nGT&*jXJM9oyXKgkZ4!nG&l8o`w}7v73~-^tU3oRl
zH@y|@SGcg5rXrkEh<S(0d#G_qv>1T2h{(ZY3jG~ogaC2}+hZ7wA<wI#&sc;PkVwnQ
zw#EqoRLn}8s~EAN4^e#wh@MyO6HDC+im)|66eXfjhgu!w34{R!putSRjG!?Sa&Z6`
z%ZO9EcX2{?F-{<Cx+7U6DSHl;z`>&-N|ltQLxo<2_bYUjErkF9PCGDFZHm>9q{O2>
zWuD-9Ax=IWtDh&`ps&DsDIdQ$^6~TL{rbPM1OgZO9DMT|Svma#={=8d_~>qG-2hV*
zq+31Wb_*|xFt#M}L-Je`#toG85KdA6`E-mB5?{>Wq6p(0NX5hkY$s)fy^$3P#9E7D
zht(6s*s^-u76vJOqQ!aK);7u3HnrIqg1APlGeduC9b>^$iYS&eXJ<K|2V7WQW%iOI
ztkxRbhd^(xaBR-Q&*~VVC^d8!I&>C0oZnoov~V00MS-I7#dOY9V=L!SLNXi-34@SO
zg?QsN^!g)2f&Wa~`tQ^7a2Vg`*iElu?ZO#~2Os0`m51ndLy972YkQlSdYdqgDNISn
zk0>%jyA~2k59tKMbi^o`Ah*_u#R9}Jg|i6hF)#uBNwtScN|X?lNUi4~r6i7Plx0-_
zm_j3^M|Wu+yV+xNYn%DmSsHPjnf46ldg~MpLgmqlBug`Uxu89sTDw5&qN7a0HlKEo
zY%bHB_oyWSg_IPgU|*-pp6(2{ZLQD<L#(xo@`AuQd|7Ex2_Z1n;R%n;WSf>3(3Ao7
z$tK0-Nut2Viy{JD#}6%OGQoa9XZ)*mnb|o?Q=+tiodU&|RdN;9fHQ=hIT~{<lvJF1
z>I8e|JCr%Zb8YfzN)UQPjT%mQxE%~#C8q#EqrrA>jLS>v2)rP`=<2Ro`U*b|q15Er
zm>_P_UEE7P?KABUDKdlVwutL>$|yiP%TNqJ8=TVUBBLxForNW~rXDISxai<TY-r1U
zMNZLM=I~OBx^W00F~X-gJ40_Wsib+R<O`Ij29p*+sY_&7M}`4NjVTr5e!|i|2!tUU
zY%}f;kOE>)(yrCH{F<w{>gua`<D1`1EsmJ!cJMrpFp8KYQ%;>a#oFo`ciwdmk3RYs
zk3aSpSzchC$tN!?0AB(+@@akTTiH1M2;B!B<>JE&s5yztGuW6w$HUZWNK+vE5ON@v
zAiavgO#)K`n?1l*`gPWV3YZeGQF<KM8EwK&w+^cB<Jyiqih#m^^kGj2*2l0ug!vZ4
zb%JieWHYboK@X%7#PdEYy2J9y1`C&6!c(n%d=gO0k8pS<M^0jpmRv)$BnXxQCcQC1
z2*TN_sB><Ipi~iMgmsmxSsy&#p`^nfo+Vtnm6~5eNW~Xu4gI}XFXiJGM?QYeMcuFb
z%CFF9Gyu?A^J~BMYs}5febICN`9J^XU)sT(#VSEKzr=wzzk#iD>xk9@4_|O}l!3gU
zGdqisK0-TcGjr7E7pe=^S&T06vpTA`K|UTbm4cu<iwFY(Pf?FTHnx(=AXf;alHi{0
zFbdzR=*z|;aJbTdfcoM*Vl*ThjfrOFXtlc-T@u&p_*BN^D)Lx9agJdLt>G5?_FqC@
zMEu30_Ytc(4$XL!MMhp)F5119{&Y-XG?A)dv_>jffl;Nb)PQzc`o2;WWkEWbVp=T%
z9k6loG@B#w4A_3>!+Wk<_jj_Kfzp#S7k0Dfl{Ye6-6XBg^Ows7^HP)L1qZsb6e>XB
zn4h0#VNvb;vPG8#^|VQ}y}@8SVnaw;vvY_jpyqjWqLB4rALlCde@`iJV0XlpMBt;O
zKxm7^;YwXeU@XofM+17p5i_&%>}hq$%aXZTjTng*9#QOZ`x8%*<bv90o4xxFk*JVQ
zp1O;cH_t0(Jd!daw*|)*j&W)_W?~IXQju#-pcEceQCBr@1lm~w>5-Kw!)#1ZZ_?HU
zt7lgjO?J*{KO~4^%A%k!>6h>+i)ZI`&*%Yu<yPzH70fux@R5gb=xRJ}Jrow}Ec*@}
zV1DrugiIL^HrTkZ&f-D`LPb2+Liirj(U5#=gIcSDU#o%g5N>B=9fU;n2Ft5wX)P|G
zl!w+89>$YV#nP+Am~jDCV?Bjxw&?^l(m|hMyN_rX{CbVJ-KCyR$)*!(HJ@4|Bvy)`
z(PFT+#$<e+`M_}Cz-0`a;twBxfUk3w%Xc>!jruee_ELHg<Dy3pHSxn1QhAi3>Ig^+
zE;q=A!;^|6vka$_T2vJnQbBp*E;_<;<2BdtjqiFFH{Ws#2M-;h(QH;-BizpUQthq4
zaMe}U;;i7Gd{@DEG~l++e~v%>(~t5efA$w__4*j|0`I$Zl%613+(+weH#0oHiEi!X
z<g(A4=+n$W&NUIr!<H5~Hv<cm)>jn-O2}%k)rVmLVhD2rT!2#nVdUY4nrHrSC9AKW
zZf6Ewg>9=eBr5i-5U{IVb^eA^gcsqp>(t8<QLjYL6+X76Sbk*0zzTNvRycI%T2}lv
zzxU8Re7BS=btc$ClW9S7PfVGnl-3b>R9HH-Ds2Z@00961Nkl<ZLpOgqhYNvljzCp^
zRyhIH(x^OR_{1sJi!Rz|{=UWPm-6w8BOgDfqHh2G{k;GE?|<g|rKKf)^EZF<i=HEd
zcn(FKKsbf;0@`~HaM9P?#KPPh$?`Je^(_u8T#UcaAzeF1Eeya{ASAeI%wv_r3w&C$
zizHsabkqX@UcJRuJ|v%P<BUNVkkYGIOtRu`Sz$p)q^w3`Li&({GaBO{s5KDeSX~f?
zA+7c-CeLxc#CeMTa0ruC_Ut=GnK|ZWy1e;a-^v|F_VL(Xe1v#3rtJ4wyyALXyT<l~
zC-K7ueppAT0B{w{Mg`T0jihuAAv738GRQECP?HhOxl1_v*zG$Y&Yjk<dwRsR^SM{D
zs#L5=@z*doUwGnuWlug~K{^lRN3@rg*!{|vGe6sAZKcPB?LODcT}-slWU#)>LKGl;
z4=E(ZR(oueBnkrd%`OrtpS5ux1jLONYr4<YWD8xCNCUp~5WWwx5(hP+!sRIGVSJ5{
z0aAc5B?XRFt$}nIM(1?Hh`p^DiaaOs6~<Sb8xI(z=U7;L8Aaw;>Na@$H@ux&FPY<k
zzx){Sa6~fQ=Ex<l#MIiHx^RLhsL=>(gkFHy0nSSmRI1`AG6zBw2<tIOQpy=YBxBm$
zeQZ5(8)=%O{1|Bko@(HCI$$iG&?v70j|<^&gbaE~)y(kg;Dyx(o{zzNWt~LMl_fv$
zUEj-nzxK-{_uf-UoJfHuBq|CxxMx3Rk|I>X#_5xIIFjB7p+Fi-K2DfyZ(_>=Js1;p
zX9!yD3Xh{CN_YhIn8d(^)gF77Iuv<{DGIDHm7Z1L<JTKltr19EVK71w&&?4gDTBd)
z&{~34hwjV*BdG{$bvzVK`Q&*<zIB1du7l{rGTW+i^Edw^&cE?BJn^Uh5BHyV2qqZ^
z-~HX3DI%h470xM&z$5e=<8eid(b8c;jY(Gt&YUC56~lDryz~?k=kOJL&p-J#<}SY;
zJjv<vPjU9b8M@tBI_);Cb{nY_olci1it#)j5C~-nRLE?&$lJgE8+g;3-^}-Y|M&4*
zANnvK_~5TI8joMFJ(gDBY=H11qJ2l`-12%_Gc%ND&#<|%O*nrs;jRW|<qRtF5Km*~
ztgQm9Co4IS*;!B{7*9bCpkAXWGSoOlY6V(?^1%<P4j-yw&!VcLF3{llkh%&thk#nG
z(vZ_RA`bD}%}Rd6S2$lW9cgOAgnbw7Ax|ZpnGV<ggYV+rqkFjf4}X)_jW<y}O?b^^
z<n5B_h0{c#j|@CeK8~HgLCVVc0>9$_N)Q@~9z=^CQba^^*K+EvhcMPX6RV#M#Z`Q+
zR2H9(#$WLG7scwogR$*P`S?YVkDpUfcjnBQ=k>*%b4Vm8feS(|zT$GaWllNWMtQT$
zFSPI@N$4wTwHRZ}j!qOvA#vD>xZxmbw{U3E@fc?W>%9bPM4f|o+)=xQ8#PvATaD4!
zR>Q`&F|lpiXxucmjfp2V8{5{M@7}e(`!CG;%{lMc@7~Xb)|?(D6+p#;=B9GgELLA8
z>he8=Mu6=OzyFI8z=AClHR@`ns~c@ishxa^@K{&hE!wi%@Y+7^>0~Cm{pyls*Zq;*
zlMKRy0k_nG8fw@<v!|P*VgxW%12tu(y$yKv8KiS`Ohm@X#H+j^E`}U3<{9#o0Mctw
z)L^oe_THE0TebRML0Oj=^TxdFdItxGiNa}ArdDGdGDoqJF~f7RO>fa}kUcl!no6`e
zG>+Y%q4YDe9#-%^E%C(ZSQPP?Ksyj9-_<V(Q;*=4m!e5Q_Y(r8Jr^5$^1&l$;YysX
zJjXJJV(BaB5w0KYRL%R$N+OU99n!)VlCb}gC9i;ms$cU25UZ^9?Wq=*OLH#EsJu?p
z(dU%^UOSn-rYl8is$?E2KT|u`;eFMCas=VyX5ced4{W({n_Ie>ZZGiOlRSa6W|B&!
zW6LzIXSJWDasEcb(4QhNJHVkaFCaqrk}}0|?!YSGAEnTshzTbuG@NavoNr)S+7N-6
z*2ys9B9m4eBMIKfe3jH0_LWOK;I`O7B%<sVrKjh(9+KjDm~t>$UNiT4Ar#z3lqo7j
z+vhYxH#C#sJe#rStW)2$n)L9LO*4ao4ZLHWgD-CipsWPqmRt#hUx`T$)8M6@ZK)b9
zL90orU&J(*TwYRco_KvL8MfBhaTbmDN!p3Z(XI-^M93}C+@J2RT{&glsz2+#A+?}E
zFk|?!@4Qj5DZdu~6if0!`Km+YLjGJ7PYm3qP#d|g;T_HkCc0>SL{|qrunePd8FxPq
zf9=rirM=CtgrV~B>r(VLXFlx0`-&>myEmm{)ndWpb%wyyaVM)?g?)<{jcGQ;-1B)j
z5}mZs@o-gd<FJ<`7stlMWkAJOjwY&P)7bj&X8dWC<TJhV28;Rd&~FE)ZFj08?^s1B
zKi%htZ;B?!s6WD2L$6PsmyD1$#S&vus&%BmC$#cXI)I<ZdoWILW6D*Y1`+``RY@T;
zj#7ayr2qjp&p;;4<ClA=nXno>E(_mA4=J>~t2`Kd38fl@9`WaE`=C{DGI7^^$_kSQ
zv(Q)(Jze+ZoRyo$c#99}BNbl0_ZG+z;GA~7^g%J@06n3bTu;i;FchG@Ng+S{=a^c4
zi9Vsb5!Wum9Lu&ATbTY%7<X(^)i$>ypdTG<up;*BlC&uuczYvM5zbSu^GcFm!B@nM
zonwwz@ORJ6|G(S%GuQYer^0+W^&!3(oy!tV=Jp5+dLL8K9BE1@j)q7C*Z3(&{TyX~
z89uTBMzRbVK$MG<G73kBWu~QsNZ&<E?_zduOr>p+l36IvU&>3q0I9eKT~~S`M@~k+
z?*bYygW1DpzDvw2oN!8-2`QAeYRPtaJS)J%jETx&lZ!CN=6M)fx@|M~?7`C<IM}s#
zK090l=*QM#U$=7^{IsFot79+w<L#Yk_j;y<&`K#3>irMe>hF;v*7qxT8?sEs%skJG
z=A_R3SciWLTwP%V)jfH>u&6BTTykBT!rLRaG321p`WQs|!?ycHo#(qUp>f|wff2Ww
z_{D?E;>w1p9ePx&AHOw`tk1qXr!j5H>6Lk?I2+v)lEq2l78bORZ$MOdsTh}3W+<46
zvR5Y`!tA<0P9>${{oJY<)D0D*4XsV81dmHWrhktz1L_i?F`>acH5`|aLh-=Eg1H>h
zr9tI@OHbQWo`LHO8LjE6$$D=qoYQ*4BHGcY9P_TB`=yQE{P&Mnypte45<ZZgyMl1&
zlX5TprensRbLWe*@8^c7?Cenja)RUr!-DtFT06<6QGbZ1|Ar#vn!9|!<BRm3>OV47
z`w!W)y=0}=Ib0I2o?}n8hM}eU$m)ZIqrRJA1our}`j#yKLHznX%X{q4D4%Q>L+S2*
zg06a}xPG224ZZRGt*6QL$J3rT&9(mWFN*kuxW8a$n;-&{IQWn&{|tsV$b^;o_%K&q
zY=-`prfK~dJkLJj)$sjJZLqq8DPFJ*!q`d*Mf@zFs!s-1pQ_@Y+n$1w;MB<&(uQnM
ztAY{7tT1DYOd~ZtTTbw`hj0VljO~(E=Mjqz0aWu6XhwH+Tw<v^h=Qv#>ZTlax{RcB
z1Vh{+aZxws0ZGmqr}b<s8V{L02gA~ir)l4^ols9ysF|M<xoZL9`LB_yZvef@1IaR|
z=Oz6!&8FI$D(_M#c`BN%Z0XYZU_zDuWjWX9X3k%{j2*O}u0MWi;Lon1?<r6~qLv{@
z1;;<c=->mDZUDf$d@uJ0=KI`%XkCYLKAc8v6`5@q^M3Iy)hW_j&rFA1&Dtb*qRzv)
zKj;_JqKoLdmV9JE!-=Z1QSz2Io;X284kr^!SVfAp&AkJ#=odYHm$l9t(fnlFAmb7w
z#t3=|>0L8YHTk<WKKA!1THp$k36vR!WM*XL+Pl^gQt5f;5~u+5Xz*V7JjwQaOZSfq
zW8k@e;>~~Iu<Q|2Li9Y>(U);f-@bm;upA9|)(h+EkgF!~LwSDUor7+c&4><=T&Vj-
zb7`hO+<A}eu=k_J6M1RNk;ogB_%|Q;-=I8nLM}&p8^oblq51y2_B3H3I4<xg(1}rT
zz9m!X_kTK-^Y;_Kk3;_tIj_xhK|fboPzBO5<5P*4d}*;Ki4dlx+O|%t1St&K)4~S0
znrpZ>Za9V=8{wxtqX=Um1I9_9v7F6;_K)HOI8a$5ey%)j3^sRDUGH;!T=vR<wu3Bv
zTEwl6%Q~k9Q>>QlW2Au8{GWnk!kh}r<*C}Hrl3pD`G3?++Q}MQP86%yw5Yvx9^0h1
z;PR}5pT}5YVWXTK%-K?;@LA|p3c(+a{auQ7ffy4Li%&piDCU+sYT{(wSv4}>od#>4
z!D`QsYXW@*UywvdQ^p@Glh7{(31b7dKK-rN6TtS%UI_i`o24tLtxqv9hB`>-Wkl*}
zW9ykHj~9?M%TL78%{F#`;)WYQl?RxQ5i`RJW-CGuz|IPDwIut_Mb@)wVnUd?`1{8y
zZskm?D0>r3toBH;{IT>XPj+q3f+QKbU+=v2%}fMh5fAaPJC6;yu6tlrUx@lyL5pPL
zT?BzS4WT{ugjd{(da!Mhdf1$-aUBCHXnQ<BVYwKK%L9xUH`g_t??<?7u>c9WU9N#9
zq?6MyT<V6_3kDXBYOrUr%IkP|hbYRH(BQOl1N<FR`>@hvN=MjWCIi%t0qfA>a(3Mp
z$Oek;04`-*r9m<bXk;>US#+-`S~z+vKZsXsY>y$VrK_&32N1XC1tiJ)ZT!uQ(3`8m
z<}Gb?fcz`>R=CdyBBKXyro8kl-MzW|>&<xETXs*Tx6qpo$C6~Dg+)Yr>XasH-ZxJC
z5z^v$7^4Q6>65~g@Ego2{HQ_+9{i+-7F-9rZ#v;{6bKqh3f>vyz_f)XT3?c~?~MGm
zp(XPvYLK<~ZK{N{4HOAybh_naj91pIe%r|>F{6!A%H|UjF4wX9VM{%M*DM7KR@=6L
zQO0rI7q0wQdtzth^h~@yA**NK3|?a$`rCretI%<XRtI8hTGMDzz^DAhejrGwS4H<Q
zg|@AXR*RT#LR~V;;a8fqNN>mHvz@Njp0%wN9u?H$QaX?44YUO)?;uX+^7mFvC#B6~
z1dI0J-Lb0AfiuZ&)OgKKmBaJbYVVcU`BT?F@Zddl2afx-f&YUAx7Qu6&~1Z|RN5iv
zRFJ+z*v1D+`2sGSLcBQ&iDN38+_gJ65=ykDzsE|$Y64Hs^)e<<H+W8|Mt+Op7}x5E
z(><f68@ddz3%|0Bkhj{JSGEqWAB@(zNOdv!+1T9A<eQbwNNiF1PT4;fp^i+=qluc*
z5BYBQb6Z3J5}m8{y2_!{5l`PxMAMw9Jz{YPc13_6NUCF!FfftG1C6qPV(BL4H4x&R
zw=Bvl<GyIJ6P58?sA7r7ZRtJzNUp9He!U6vF|*VS;s03z-4yYg$hcT|<h%N-zA$c+
z8@^K<Sn;`uO5Ykp_4z~vSKYyXJW=?M{3lj^vllssIedX+U2)jGeG2x}6@oo=<$qpR
z^50iJL5KV}^0{lC-K;35Nc|XpNQE=L;7IfLuw8S8nzuHKXx8ks=wJ=`{-y6z><DH@
z5Y=Vt(~?3PJmLrPiqIqUBE-<rz6Lp;hI<+|#`|nZOq^dP9b%tO%`JCGcQ|9!(>_mZ
zxI2%!q|zTuO<OPYF*{D}yb=%N-Y^uc*cPS4m-$%dYK45mJxo~KF3jlC(2MTeRXxkH
zn#pbHymqbGTRn}hYi@=>+4LE&Sm|^VXZzXq{vd^mnZ%NrVAZFt=QvF1b1~1>+?F0l
zT&U<9LK>5iZSZe9P>g};!U$j#`5fx`)(Zszi$YwumM1|=Qb1bmsSDUw)Ac&T1zVgx
z6SkgWx3=F&o=C?lPWkjJ4FU_Nc=)OQg7AWGpaU<-ss_E1gw=CX*_hjEJ7p`B%NDID
zL`k@2j_L=q5C~}c(b#hhD~pb$ip>JCqU_h9X=ru(T;#5L7yT|q4*?zStUYCNX*|Vx
z1#LEY>NH`Zx7T3N_E-JnX*>Oy-4B3B93ZXz(j<nOQC3?yTi7dP%)-qrt{IB)??nQR
zxUSC`zNGv5sA!8s)#>e6cJ;w90Se&ljd~Sf)A9C4u@1ad7T}PG=Ex>+3Ya3vRt-<w
z1RtN<7Epdp;Y{^#W<dI$q3)j81^dtdHUM9gXklCMdE`+#1<5_0nK#b}U!YNt3vJ(s
z;W!_ozJ2-J&hC!fdfC|8b@#u22a^_IR91VBV77LzP5UV`XKfeA5VU-r$5%T1;zp%D
zk{0_J=rK@*m}zg4%qv3geQLk<=u3I1!Ldp1R*Lb&|K6aAv6lZzVA_Qo)H<|`P`ZnB
zOgf4vs)f}7hk7V$oA_H*Bq{W+A%h>Pf}!rx@`>lZmo$f>a;a7(Ztg1;{Zf=E@32<z
zH$h{zAyi27v;$%6loiY#Y^N2Q%|R3Yt8B_clHEcv;L`{V@xfW>PyNq>ia%VK-3Dd4
zTmJ&R6>qNI6$`$`U!@<$J8Sf8yJ&e@OKL(Y8kLxuKQi$;v#<NT9=5&at$9Apz;d_v
z#EqIxc<`IEO^!J0!86yWS-;&&3RmcvI6KEmk@zO0cb%#E9JctQzM|@j7j01G%V$`*
zdw?E4zs$+>gWoFYqc?A!?PC?_?)!GSK?YcBD!|rW(jcN{>uMu2D126zwN&6KWd)^3
z-eO*Kp=`*q=2agvLrG{gY`<)dykW)A%h=L&+g?*8iJ<IhA+H2=W}a3-heKMeGf%So
zZQ|PHzp&$XUCvhxk1AaB>fi=$keH1~o=Ah1ub;|u2s<J)LSxBIY}G=B=Zcy~zDHEo
z25(gx-sW5TY3spow3g1AxTeRgPL0K_nH5%r17BZS`J4<!DYn9J(3&W&gHGu=CxL4C
zM}-hM+-FqV^9CWmW5`gbF9XAcv7aBYP$e}2EDB-4TBg_;ns1?3@4klKWSv~RAV$UK
zE39Acep9S$$}@d$vG0G!5`}!h_s-6vx$$$B`^p8}GW71>tbAf@Asz*B$2EnbBn#H0
zvoVESlpwbLOr~Q~NYB8!W)%Ps9foY*CF<}QjxFRa(=@J8o>jZHQxLdu>}PJ^Of-(e
z*>>t2&&ntl6X<LqHHbVOTO%;m&9x<dS$<2n<wA(1sN}Fv+YU|leGSbd&xjpCNL+U4
zPfy_Ap+*bYkejxnWi`T{`*^EZ&g<--AHYkWoc&a+>eiPpo@sN(E<7HlLIWQmd>39Z
z_kJRq*;g4Op7k8^FT}^D(W9m?7D$OS&#<3+d{2EOFE|s-7i4fgd3nM_LLX?K`=5^_
zmK@r;uKP{vcUfFcGZHwvp2P~YztHMFw<X?|<KGcC?@GC@{XfD#2WbTF@f!Vhu<Bx6
zJ!T7j1~yZDSDNr<Y{rCkb`|%}CcT4SW9+0ft6#84DRntGIEoER*vDK<!V5-=vqxjg
zRfrToRo;6NWgAs#j-r#G<L+MAM`;*YM^3MAT09Kw$fIQB80J<_4-awe`@u#0Z^Wji
z(BC>5WY4Y7sL(L|*Tgk4A!rsqh5)KTnl52X3~tEH0c!WPu?QP-G;u)V>RLxAQfQsL
z()~GFs5Lru0ohHjeT3tu*3p#H<V@FIFrE8`Ytn+>@WUTqR?hTzS&CK2W|mSM$zOgO
zv4P304=A5c_YV_n3kx`bOT4AP@Tr`guFvyej#K37t}_;{Ee|5#!$qG)>RT{ol5ceM
z+cK(;kb}n@!L#D$htTy(&kOXxTj<tqA?@<&rt#w}xdI)((7)c(&0dd`ZCGL)>L0@v
z8ay05*@l2PpUzMV#GFb^XT^q{N+sNizF*DCv6))bYilXV=u`2mQ|Pad$K^yBX<u|y
zieZ9h2rZU$_{|RRDwsI0GRk3MI61GwY;-xf_6;p^736;RM~EK&5#I46mQcPJRBb6L
zr$*;~(Ww}rrH<j7Z;xgm>`s{p&Ntu8o;`m-^2&Z&$ln~8u$6W<%0skR($v~4BtX1L
zE%Pw6Np{v&g3SA~`8bc;$Vxdr{&txEY!zkOZ^|nw8@Py4#HdYxpf|epYmPC{F^FRn
z?+&4v06rOj)e;sA0j4@sc&@E==c!lRl$|%Obw1MgCC&T>I($4h6rHE%Gn`Ga<}k8W
z_rj2Z5jx#{I#&hm_&6`qR-~oywn$`?hJM(?7}C)v4Na+eWFlYY&hXMIZZYRXm6ud$
z1r{}`_e~C(BC=~ty!<9q^SgVcbz;zS__`$KWm_*|<v2D!wDUaf&^BQvm7jrs++tW#
z8Nz1NOs<YK?cQ<8u#z{wa~R{Z<MpGixSxH$iTBO;(-UXQd@bbns=!oP?KISc2pT*&
zW#$>IO$V&$m1!Nz+$8QUqFwxoI~Qm3VKd|R#B3}$fbV|p_(0!!8~`6SL%yJ4>+L6u
z$7=&H_-6W#d+W(qI$+Mm{-7`#<$TbZ1Z#fT`^wnw4Coj0p9tGq$LG_W&l|WR#vjII
z#NfYS8%4jH<hO0q4^sDa01&55`U_IHnEw4=jcd7{JGph2xt|8J*WZzuNbKZtm=fAa
z69l#4>o=+E=n}v!5^y$TkYLZzQ7oJdIymd}uRLHbt#KY25VDv>8wD>qsyV=suM&g(
z^!8$}ZXxhBTak6lt8*uJiG`1ZjYx~r8;{-laT?Bdb|zy-HyNm2O|E_E45*dpQm0tL
z8_trQ*~}WF7MFmt$k6rprsiP$s2g(&Y3@cu3J7zw2`wo(%|KdeeSO%i)Lq^ivx%#%
zcgidK*ydE=Z<ADwZM}WksI-gopk!WVOPjXz$;9|0yL)zXYQNp_dF`#4vlV>fTiI|_
zFK#ory2`6a=$~Uw|8K86{OO72GlQVUAKBq^)g5p(T_-Ch?I8{rbY0RPVD7rac@p~E
z^*`Vex-%NP32iF=-b8js)&et}@~5zv8WWjpq~(>PO2jj&E=W>9RQa1c`?`9*n?R|M
z%(m5I;Os0t$0RrTTvjG6d8vYfvF3QGX|ksvUrxamdJN}OqP(Qf<dVK@(J6Z__hL)g
z#JLXTgRf)}UAxY1`9d^Y%h%5Dxb?qkRMY5I@pBkw7M1X$kwu}y&Jk%?j0Po~y_Hnh
zgn9<&txCIo92}(M#<zGJRnO_R)(!=ZqdNMS#e&|CgDZ1aB->aY^V)+)7r&qGqT4hB
zb!Yo%gcm3Xsl1uHA}>FhFHJxH^~aUwill$FLw-L;_Oc*0!(7tjxn;?JIV)?uHe0%m
zb?~D2&vCRi^Kg^;Z&)t=y8H2+AAAtKbM!oKZA$5zY96CW`Xq+*-nIa#-bH%qmjK_$
zFpa7PK2EF6C8e#qeo)V0cesI+Y{WB)!%3i170==%Py<p@yRsP(Ll*ZWB<F3Dmw)nS
zZ@)%m<x|vpunilI0OOkOJulYzCA%#GeSu7(H4iJB7p$=-47xqx!gLqihk!*cKHW^A
zp*Pz&6@4nhD*lvTmQ^$Dl=%vdR&fTXn%-9^|5!>FRzl&(QFMKw+&8ZeCmi_uD(pRw
z{RH}N#nLp?rAq}KF0##<>3MuXK%S1Ku}nxOq&Nh97DoA!6&PGqaF>;q8#Ja{M53|O
z(10%lbM?!~aaIE@k(W{eHGw=0p0>`*J48i=jn6LM*3gH5H=WbZBIUzy|EKWQ!;e>w
zrx{?+nLza$i$Ej2u*H*A&@^eufL#@ti#n^W?{D-V+P}(2!(Z@!;L<FRkP@a;<M#~v
z{YZ<2@jFX~i=;2b?2>zKIA6NH+T^|8#E6lPU-BDgJ-~)ZE&#qXm$+`jRS74dmN8t~
zU#;9;mNv9cTpZ=e&P`EbQOv0!i<<O$xbj!ffdrr@Io&|4IV5wd&a5<{QZ~M!dL7jk
zyo64VRTf$TOS@6&M27qQ*#37y#5;S}Wb96%u*5Iy>Me}<uC*$nwhS}0G`|)apJ&s%
zlKfwYgvMsiAVLm}a{Z{)YlMg@)`-id)<=Ot@2~3~8>gPHYaif;gkBttZ^#||zl_uB
zuBgus#Lra!$9(^+t*7Zrr8H(+vqx}}0<gDx_-p=s9rRAPbt|xSD}?3p<=id%KfP7^
z7Jdg7-Fx6y29(Nnj$~F)Z|dGIx*x+eq8?s84op8U*b<aTDSl2pN=fmJ_3wcy%2#fN
zZW7~0Dn+?}+Q@1)Xi{tPRsGHoWu>=B?L_7%ui_%vD;MO!exjONrbd^7Ol=2Wh6f4#
zuOOW_P|U5;$}S9c=RLR@Hh%tV7gfUhrn5U;LNzW|`S;7zig~=FcZ6rA2(OfLsaz@6
z$|9Ra&%q?&atEwXHQ?<@b=~zbc&`oW+;YD3i9Y`7=mt*#J5Bs70ennILqVqCI1V)}
zUF}8gu0L*X<n%a(?Kj=pJ~n5{Q|q=m1}>zUrO-s$pif?4+t}Ewd%aq;J=<(f@pXUb
zWaR4%LSO@JD~5VLxjwVhJ4XFqOCFPPU_^gWkAohaB!e+Eb^fg+wa*oMiW0&Z+l|JO
zjXI~e4gz8sGN&&Zk7HF=V|hh3ELKSj0UVtH>$opgqdrxVN;b2Erbx(xoZHv7X@Cla
zTSH-X%UTP!yjvYSM*rmCjt9VgDy$59yf^WRp<nalzrufeMK(Fo<|{%(EE<n!a85;a
z3aVL2fNEAh=w6qKc+tL8wWM=5K3vM8b1Wv$=L)Nf6H(kcpBsKfLB~!f88#H=98xY4
z^V5moJ6QOEu;K%sVw&0l3GdVt;m5z=J1|mp{e78&AzuI#+JyDe1)-HUy~Yo;G{Gx?
z>?9?xvb(uHN51HMW!k)3UAnE}>dDmGxM0tJRRRAZPv!Z)SFCR;KF=-i{)B{`|4Ygl
zwY=)Qyd71g@f@(x5647s9N{Lvn3zS6Qx`A?xTQ%Ru^eO1rb2gzAHH&%)Pe(dYpl<l
zY7j<*xY3+=%2jcA@s>+IrA0`I1eKSxKsp@)?>389a_yS#EPdvkU4Ks!t#W1Ay5=-M
zK$6C$3*LqRSFPs>Zalxzp^7hThkJ+Yckx4R!ePxQUG~h<R?#8?rE2@%OX_tvSa9VT
zZBMve{J#MQWLX1b6$3e>f97W6@PrJHxVSquXoxE73lW{jdevYmTtTwG_=A(`FKY1m
z+Isd+ne`WWqWbH*IVm#*hYvOd2X+$JEfj9_wyrmH-OglVSP2EZz>z1159H6QweBJ(
zhG(`kM|{%=XPEA(2wQ|*vib|swK}&o04k}yFk6+60S#YJpDWBeKyBWW?-n%wD<qPe
z64yqBpM|w7nVzSmQTY8PW@K&1)*UC<PjXps@~tnJu-%Wiib`>F$)aGbpOMZUiw8Ac
zU%~x`1;F@4rLkEPsy&3k8IE}vR+Hpg)#DCtZ1iu=H8X2S{Om)Ip(a7t2mL6LCC|4f
znTz&U<@$B8lsI+Y*p=J-o$gq(3-_cQl-)TQkqg4i6xw~~V^<Bp+n~w%>oieLhhr8Z
zfyMHL9eA6op{{<#FQSWT<?_l`_oBM~cHpo_WbkM=0XpxG9J@8`Uf2`*m{**(=6de&
zKZatq{a+cF_U0Na0~7mPiTJ&g`M)^$z8z#c9Q3@Uf8J1$fX<gg;4r^4M~H=T!dUAX
zi-}~S0IVc`#iQTDwVXSuY%lzaLdO}x+)QDRn{F>;MKl7}poRcq$%$C!GK=JWn>(Qh
zy@tr}H?&*~8a6HgZg$bZ;Y^1Qe8k(^1r+N~VE5XFPW4WpH?c*|&(?zlDV}!=x{9^O
zxk^QNNS0Ogw4BAAO35LBt5XRb8>p^X*eZI<lC=&wYYyVLs%Izz1z?dVw}vRr&RG9@
z^nhb`6WsBYx{tx1$@lH>yH^XECf&^HFY+|eTb-WAj<*kzHx#o>@>F5C(1tjb^b|8Q
zQ@hHrD)`O6zvrsvPVbkOb$7a7FShx5cD>s^UNq3ef0TxjMOnXH3cZ+Q%-+adeqzbK
z_q&`CRy(OWM*W_jr-<kOd951!04<Yd*pQ=M6lM@Vt%FhqtxRG5S1t<>JYzXVD<?A^
zQhG^N#}E%}G@jIq&$O9HHWxhoU}EYvBDtSJ@~kOdB4Nk7=kYmWd0c;o9_{n(JZT>w
z^tv7Lc`0y~zX=(y5$YdRG|Ryk92Mm+#ZnIRtBQAo45y9>l7-~VPNFdnCp!4UrDJ(q
zlng=O{z+CsqT&E@@^p^$=$1R?;^va|IB9;x%l8dH6zF3?b<~nQqmPmlPK(2R6%%%R
zmGVyw7rGXygeia)op+;F$19bdX&YF9&XIh9QTFC~HU*=Y5?wF6pG3M(!2-{cz9)>V
zY$brBFoWbXk)&`Q2J8pvk<L@7eMWR{5&Bv7MY<e~ySPc3|E->M1N^25%v0zy4=E&P
zSSA+>Bg?z0Rac^vF+=^j?J13?Os#5Le*K=XoD$<az(F@-6h+L{X5t*UcZJ^UMG(We
z^fY|g3Lzer0#UBU!P{`B0o!;IRi5eO#1@=$Mq!_0UQO_iF0{<25|APxlAu6)6Y|yB
z!zlJTk|wk3LoI~uchljQ#e5M(@5ZQbx)%&vD`ioohy*OI0kLcf>SR`5dZoYksPCYq
z!StR=3`^tzBiXcEDj5~(Io8Y9OCeB;->%TV)EW<DxKRRwPf2_5{qr=E|8|^y@1^LD
z8hRTs8zZbV8i|A<Zs~Z8O^H{CE<zu~sNvbl`S0&fb(>!9(9X(`VEW@O*@)H)Lv6M-
zwPmYp=RRbHF3&^{bK*9$m6LexF3x|wi4|?<&9BS;nt=D8h@A)LIM-LXWqA8x^adT-
z7@r_y!i^6*ltmy~zygXJT-_8xr*>D;nY64a+h99V$&coRf<s;g1@(;gL1vbRhh6t{
z4_9vp|AKLrllmlV{)bUTfY&|zMRODx%Hd!5AmyM5?y=dRspr_OPhY=JRR8qCoWl7y
z^<3AKRi%`;LTD34`-(xGPzA06t~Cpj(;la-_xwY?o_7IV|M!Zk-LF=x<fX$sZ;2!#
z>1hUDpEcc%bua&wg>?)JfRFDz?}mB61Mu+e!%sHIZc_ROmwY%IE_WB*C|G^S_K{yH
z6_`bbT;gzrk36QLQRvq%<^=&JLe+mO!JLKnTFd4ZDA}3Zi2>dGa(wuBJwEYWcZ@Wy
z?ijgT7u*R}+>r4db_AWrJN?_C>u>|Y(P;G~pm~PnvBXfL%dkm~ebKY7{bs=-kw&^Z
z2LAMff6^S0_ub-igMha>rn|`4ovsrP@MEfJk=V9)qbrt)K$XA|SpL*f%Anj1uTa;t
z#;2gYQ$uHyj>5}mSv+C#ycwp-5W9S=ItGN1S@`$!kd@~Ua%Q}b{?DO~t^4!N`ziye
zstqzcB&!9ekD>SvTAVbErxB9R>p8K!LsuCxC6(W~1<K~o$Iu0cqUhVdA{u`U!k4h1
zjEN?{g#B5er2Af)nu@UXI2s*u&QI|RVXH6soslh>r2@0xtCA`#7B<f`%#(@dPirIJ
zc)lRZKliEiaDx^wg$;1w%V1<x5Z4^Ec@i=mEnRC{Ca5!tX#kT-7-DB!MYe$P&nv+3
z_c+pznUqrDKWJFc;>yxItENn@@?5WXpfp=85i8lR9}>1d`^@MovFUAx$YyHC=NDvF
zN<sK}!G>@uFQ(@5qtFEirn=<k43^Pz_`&fH#<S(3-1%AM$xBNzs}v`Q-rOZWgl+><
z*d0E3yZ6DVZ0x5!$+<X_kA?U*+S5a>=OWa1{1QJSmotSNe&zEP6*!B8KSVa_sAO1X
zi5i1&cB8oi`!GWuF-ZtKTOwW%$BrS(eL1O;6qQtWP&sJQ77@8co3{0wW13M!?=$3O
z_4xZo93ldRP1E~t&uVTYw3bs0fT2X+5J2!8ka6H6r3?0urS0EaE?u;aB03Dln<1>E
z#rGPiTt$F7MF^#JBb^4wB#BjW@v0xc14XYtR7a@qN4T~JB2mh?qjg$?$+#tf8?%e2
zEn5|2yGOwv=>_E~XKwlt=YgA0*I5pc(<B3M8Fi_?gX7S?1{(0k=6kif|BA#0d)+D?
z?+WkE6P-9K%v!+j4hoja?=d-sWsdCAl^O<o%8~HiUlDY)BIv##uJQOmB(mTVIkQzb
zgP>6BQgnw;Q61`g8qy8kJNGo=Ym~ApoR=(sIh~&z9qxR^x69m!iRpm8mKe#~Wm70R
zn9|DP^J%3$gc|*tO{p{tBv{P-5EkZIzktJ*QLk}A<x}x4R~KynU#Ws)y8zIRdKlE~
z@w*XeAEf{a^%QCDW+g7lv9@hyEz#6T8s3^lfv=Z4P~_~(E{p`zNGa9(dD6;r4tY=X
z4)cdp_c1|Fw$A1YN<a9va**!8zGC*@lhW$09VYo0MZ~E1e-TB%`{8PY0vvU?+UdEz
z4kz}C(CcvX9R();@&n9T_8(J=I#ArC=z7feLni-BkX4B(P0@<Gf()t+yn+n587YGg
z`%Cmar({BP)+*@G^GO9wPUgsSFSM8gqDi_IU_R1CesE<L*0SXjL8DOFU?o*72F%2;
zIn(Zh+(MN}rqemb#h@(S&Gqy=9+`zU>TORF{u&~Je9k~FqhuCD$GB<@9aW3n;M#5q
z(jECHN?@gegxoxyK*LG2s-ShDh@0{EiWRgR_Xf(g%JqH#hV|_5Tts)*4;NsF`5DP)
z;ie!xCks9MzPuA(`W$l2*nyPHRS(P$pIR>8x%dHZfIN?zQRb^8AHTGIDc(pns&ki+
z!*8Z}Y)(nlOT#I6c`B3xtX4_nZwcD#;?HEr`|MfrwCG@qW__AWVnZ(Vd`ck`^bhpe
z1VNed=|u`Sn63p@8n_kr<?A0cd?Oe%Ek`SK1LY?7cf5$V>F8b%TD(RHKkNQdjx%_c
z%g9M(DWWSF>4{Kwgc#+{YMqH=q{5WAK2^`2ThEhEDLT7x@lJ4AB-jU)PAY}l#&Hba
zW!R$PnYohd{NDJEO+w?OtBw98_G@sd!5xb>c?c9q2+_s%HsREGo%IWnqo2wP%Xjo_
zJa&Pfhr)qq5dK#h`G&sy2l<;fUIzrol?@CW*>8K^{yFUNo1jb}v2ns4e~&Q^3ogp^
zF#lZYm+-{jP!irok!(ch%p7#hQ#%Vb9)N&cc~&!#?sl2=V>EGyXTz9Y_02kkBn!N}
zva8Mn?fKtYTaxsV^<;+)JP(^58P65i=vKiW51cyh$S6btgG4PtgS)_RLVlM7dwO#=
zJm-tMjFmQ?ygSiFSa1Md<G8ViNQYfSYcC`-e7r$+NMCxwgMfXK&IV@6bKcwnFNQ<L
zO7m<YfoBwBjnQ7nKWfFMdhg5^wIKSN?m~ed+0vZMa+|f!`Wi9CkY~)yTR-zV;pKDQ
zXDoe9dltAY4a?xcMh1tpO8ht|a|iF2AH*y?nlOL791R?m5lbistWek=2tT;7e#-LU
z+B^d*u*|OFNSsO+@<kDTcD9J+=ZW&@YFi2@k{ZXLzymn>Gmo3~Uyw$^J&{HJS%bw6
zX*W)y0<5hmmDxPa#$8>gNJ<(xfL1SbJnZVa3-=spvsF>kK6xL7AWiykcj2@$etR3~
zEUV^Z26CwWAv1<?6@~l9nC2m($&a^tZe1gi4fb!o85Y3vh20DPmnrRORjE9y*<)O)
zscWGB3z^qD<L2{}L&;j^fT_uUK5F;Nce*_O5AJxO`#0eKo4qgW9J~oWUkpBNHgD2e
z&?IJX1ky_Bgzi;^hSs(|9+;m!syVo!s56a$0^*Zosp93bFJ$AT(A}c=Lybbko9Dkz
znq-?+xyLB0Yy&GP*uxqp`mAOf{~YYbx|L`$3W$mx91s{vO`PWqdBAuda3G#@vM+7+
zilJ7sEb0B`BS@rFy=Z@gSe#<6MZ>M-nbMxEENtRCy?PIrTK^V!X(BHpNXWLNv|K2D
z3TeXdJpwoOM@it2dU~YR-Y=oOsoOFmkce@Yq?LP6kLQb%zhL#eMZ8f=(N^}=;Po6g
z4V@g&k!6ps=eFahc{L^UU?g$iZ3Ok*;q8{>#VG0R`FwA=a!?#HxA^E9u7(W0nDP!&
z`%u0<Wf%n!(8XnFWFPihxs2nh1yQn95Yvx-t*?vIb39XcVG<}`u?soX;bW9NJX{V=
z<fSvV1cH$`yv>nw&7+|-3<i8rYw9Z2N){qB%~gs0pSTHZ>`unW^{A}aMr5g<;SP}s
z{x*~~2ycMPUXo&m&rm6Jm5rf-hYH>TC_`s?nf^}Jnx$*zh2*UJAhG(h`tkH)q6je2
zR#H=xjF}574+Y;H3(T!FtZDS(jN%^U`uqE4Bhr5TS^8m>Q#IV0&qy?&+>hFITE)<e
z&rJw?xZL&jzXLu!3W34!b40;6hkz=F%4D2p_J<4<M9oUK`4o^Q-0UWHK@*%J>(6gs
z8!+F(;C+U15}KP%hL^A)VPWj>EkC~+mX4%Nzb0Ba<W(&Ur3)Ubu}8~}BaN{^(eOJe
zD$mDN7j$e8e^+m0cv<IzYdTpJ_A$4aD;wo4wLA@(L@AvOkEe>>+#=73QH%p)f5sTs
za?0{KvP(s%+7CCPjueh`hMIV-vxx$UTyKyA+J&~MuSxEvWKFw`f1nF#K6NU;-n82I
zwykUPU9%*u0cpv0i{Ita(`fVu<7Dl13-P6#Ug11<QNkR&Q!9#cXivX$t8Ks}TMDxX
z{W5XCEiC@M^_?{lC0|udE_!}7e&z_0ExeytfRi*HT6V_DGfKbhoHnoYP8ng2j)lKe
zp}r>;>IT@W+7W$`Ln`xvMDoN<f9S9skWQWv&~zl9tWd#Jc`0UB9d}5NkZbZ4>X$8A
z+i(Vw2_ah)9<>?<pnwi{1-~q;!N<ur`K(ZHo|ALryqNqE_w{pC;(2ZEb?YvfIc8z{
za-SdI17m?@cFh_V%nEifZB5Eo#7-egp^)rI`5g&#f;H9<tB<!hqyN8`^WR4E_fS8!
zQCq=ISn&SwO(c$30HODK7wEb5x*5(Imj4u)|CAJ8*0NOB6ydWxK=7E@^tT}Z0(7ga
zQdlg$XktB8AQzhn=B%KVQ|jZA-hSWK*&VsbVaWW|3eT7jN^0qPZPtGw8I{$O!sZ=i
z)jd=JHlqbvtn_$Cs+Udfh=qe`LRDa?>LuEnrw8ox#`yLiXPjwlJL+g7oR%t{2eNg~
zJ6Qj=iZ{Y;o47N|%Q-&Xlq0H>LXv3CLils#M@xFF-MEE<{dDWom##|F`*nm8Y!Dy8
zHDY=Z0`{(Q-!)Jn(ISgse5U<BdyU6w+eUP_?j3^ez^$h$)aw-;2XB3Yo^^c*F6$U6
z-+$L%7LJs7p}s=y@5soIVu{RIpiAke*v11n3c+X=Wy~Ej9tkZad|>TF24+!ItsxA!
z_i3IFn%qoqQJh_hCpFh7A<pC8hN)zSr&=@1q<L-oM66@9(cHHKXyEe?z}e#UgiPX4
z8972*b+BHmSSl65W@z9XsOQ{4%j)47MJIa#h+%R<sG!Rkxb47fGfqI^S=|`G@Etcf
zT!cd<rp~682k}Qv;Zx25r~+Og8Pe&-)I>@VybjRim7NUoZrRQH;(bq-*w<`HVLEs1
z9CrMDj?^lB?}|XL-qN!n$EKnb)j22(*klaU*(zyN1(#QqzLNC=QafA(dvoyLZ?02{
zXyy8~bJ<2~C6mqQ@>IZdxZ9W+OW{N+mn<^KzQ%y2D&z^T>=F~bUweSLpHj>ED~8#u
z*#tcAn-NmPJVQg_ZpyXt4r)dvQU{dUIyaq^;|MX%A%5~5e9?7O%_FqqNt!Yf9s%@D
zjb(6Ij7ku@R#B<MtKu;!JKKQ*{7*RfFc$E|{p=SothTDMXitJFaL|y43?=BAI_Ru1
z$+mVe*8-t4k2f7XpJbc&aSqQn2JW|z&DcgSr^_?c;t0!D?B0SZQGZ8lh693C223H1
zG-FMRzT9-av<24(ke5@}7XHa{k8K`E7eHCiuyUz6yuO3R=$fd|_k{-I>I9-W+KEs|
zZlqnN9v*nz-=1@GHg(PI&C?dWWK>Zcn@u}QWE!IKXD;C8-;hdbBN&GE;0rHE<)+Of
z3y9d@G7qLn`p)EI_l+j<bw?7@&n%QuD)YA}-Lb&^8`|IvX2HRe$>At_2Sx?^$FaK~
z^Q0FJMCUHSW?zl}k(kfHB46IplK+3r4%j$2;OGC^6PWNP==TQne~V@iwQt+U2E)o8
z>sni%s7Nl)Hu85Z%R(E%Mzr#NbYP{*w1Bg;g9xyJ@!yfVNk~)0b)u~YIiT{{CKJTQ
ze|bQECskxeIK{QI2+5T=(4jaOf*GD8AUm&}F)(rUjn?RTG21iiKc1s`nkW+jMbb2n
zV2BjWY>y<`&~jp3?@^vB#8ZY&PTH}sbW9uj*~8;OV`__~?V6xZ8ATzfrle?7gbJVL
zGN~vX{&GM(I1Ravpt<NKupyKk!O?_jWy9peFP+Q5pqWxigIV2!gZ@0r_aQODoU`S8
zwgB%l);%9L_J(fVVV{oP-2oC6k?dL3?4VUXva3|7qw+C0y#Pd?pjot62w6HOn~1C*
z4`G6HRzdi^X?JyOrrM^axSZ5BF8I;9n0#*g1jQ>@9-hII_dlm<P+Rv?(^WHKq(J)H
z4?36o<z>xy3c?G<*eJR|7x1P|S#G|;M#N4=tqAJ%^DC1xD^w|$tc=cY98}E#Xerb{
zJTdbh3bK{e!gEc(`wl#a>d<zET$>#}F_~K66a}|p>zK9i=!Qd%3VO?-ZAgajCw+N$
z59{8d9bT!kl%>n`4@o|f{l^6IpE;;AlZ4`LvqxR}E$54f8P(H7M7OoVX(&TkBlyW8
zu*`UdO6iQY#f~U^^EoCXc)<Vw$NKt56+dY1F*cf?a*+ENFoFe+w8vK=Mdn1VJ(GXZ
zu9mHZ`RLn@hxN`L5f2QW5x%yW`MD*oJmY(^?)T9MPOUDcWJ=;>>h!YYQc}(#g&39!
zh!t)2#)O$lG#B|?Dp%)<5Vb>|JDq^JwrPra>2Nf)izbglu%0(}JV^{ibOP$JJD~yf
zQU!_~8{|l0qRfLCvcpD6@*_7$RjxynoKb!bQHS1OyR=aM4-NfSvmUq0#rV2qtKCX6
z7AnY|{xW!&y!fk$IJe&wgO-R;`bv?=d^%X(1*kG#-_i4E(dxgSsTejMYv!Og;%A>J
zURdz9!t<;E)%`OU?jdS?CR6wJBj(ET_ZsrsV=>~(Qj#OQ8v3Zh?FXwHGYJGGTbksv
z@LmY83kn+Kl=5wUi{*=^{~@O#9RP=y%Tgp{w5%zcZxWj!gLvans76&L<SDiX4~YkO
z!yhGr=W&b$h$>*_B6kKO5&7zPCCHyPFMUvreibd)UV}c4!Nlo}N5Pu^6|w%q3tV}<
z?+t`{1N$5C-u?11XcV?{c1IE^TzZ5RcI5QN^Pd?IJ06tQIs)!|a#U$Gwt?v6N*RRw
zC0C)sn!;xuU3Lm14VYxmvTCF+EcUY``BZ2@Mm6GLN$XDxs6V1_UERX7`Q0A4YkY>r
zpPzu+aXp#^b4W)MlQ(to#D~owdAK3_P9<`LAsc-FmQ9O56zAh~-f?ZSQyPcNF~{lI
zW|uIUIHh_r7K}YdF_Cmsr+6w)b0emw!fzy5O-Lb0?)XGC2Ex^GqWlgFgxDI))y^x~
zoduSOKy*{9(Kcs#X}a16_6hd`&!vuTcsh#Brxz~Ar)e@`@3*I0H+_TFM|*Q@94jU_
zW&MD<sEjc_C9kAe49ya_5^)$v8YG#tx~8@v?w@F)XtWZOx0&TXR<wP;g@R}^e+B4n
z7?~Mz&mwc{BDKA*Tj_i5s`=cgW~~Wz_xrFL0=g#+-WT96+U+7RiHd0=5c3J_`NzsT
z6q1&KnO2OB?e0;H+=_g>y;!#$mS0)3n@ux{|BR++F&k#3dYHvE$I0mbl2?Q%I3drG
zBKJM<eyH;~1@;BmG8;B?d6HGNfBRc!GtdZsgBMa8vzR9;-TJOD?%>wp^T@R>`kR0A
zX?4l*X{^HYt&eZC3`ByVo=$a2aLO<>S@iLF_U>700ZiSB^kuuP?s-3#Tc{xV(~n|I
z(Oc5y$~9`QjBh7^zd|uOJCZdsA-<d@;Ixd65myVm@gPhiA}-&+m$|T)**J+kJ5X?>
z`4R*T0ST=XDtbP?mcI2`<qnen1Q=imfBPb$u575Wt~eO&b$#;g*YkSO<~ha1u9#PZ
zDcar&iR(~c8zik7WBYsk3-GoMx1M%bJ5b`cE~<tyF+$zfkbVGUlxQ$5E|p14vqzl4
zVrz0*CW}$3*u=&hr5toP>LiQ*E98nwZfFrK_t~?HySwK{=%MEUAM^CN86kgj{!!<!
z5mbRwvs*i+l}Lx4iM<;FJ+x<DatW(K9$G8$?go>c&1eH*0W^jxS<!vH$g?`X@DFM&
z&xh~bzcAI4^-v=mSfd7Q$ZqeCc^tx$lFS=M&n5`L&Fkh}p-dcroZ7Z75&vb1sKWZR
z%mne>n2RZ<s34)&yZrT9UnzrkHr43YYZOr6Y2bA)HQLiQ{Ro5l1=hG*pSBt{;bsJ>
z%(&2eqFF_A?a+f|cvu}$djM)3mTn}aI|vRN$n5<5hU&2eyETyB^9%eqX_qZ~X3Nc#
z>Atbx<9}a}@OjvIA|;=E6=GGazy7XzfNJ&gJEsOR014yg5ySTo0u@cUr2W<|CPo@y
zDMB{Q9^RgnEUBt)uRHYOldX=#;e)TwVI)pM<}g%LzH>)cl7N97v`l52lNnTb?{W?6
zQr=<B+rpOcSU|kbw%EGc9vqEgkJp@!D^|QitjC*bbi0kby!_SG0P1HYGxN&&zf@+Q
ztPa6z8)(rwKeOb~=6($nS?yU$2@5g)<LRx9l8*MF+O;BS=MctJ?l{qBV!w`HG!y`T
zp*rfwToA-pzUARP)%hSbUOvS+z8^zHJuVStc|1e(Ld&N2Clxu~RKxVz+=3GmO&(h&
zEp^T}-)H2uwl6~UW+%Xc%O5VE2S&zOTep$flsW?%dFQ=aMLij>cQS=MiWm_bo;-^`
zFlb)HYel#}mJZ&A=v~}j8;M3+Sz;VSn;yA#n{u*kw$h&&w6q=X$9VG&Z!8R+Fs0&%
zcM4#Mjhu{=P{AWi4zKS^>7=^6L-w4aZ!(|ImEYIdox8E`uLMkt$K;hpxWK*VfGiiy
zW7oFUwmvWa^x~0(W?BuXl<>nRjuRqilp<UCNr!}ex<n>iB#n_<E8Yuy485V33Oiyy
z5Lw>yhnBuz+vJi4v(wHQoRu}L2Fs7Ne>>rEfcv}e!E+$TPnY~ZEJ(?W)*p~GrKk<d
zmhAEvZld9fv=qzd?2J@6?ynwJVK&HCi96iep+Q-v4aBiC$#yU4QweAG+drsBf#`hH
zqV)D&W%;C)yU!4d_;0mgDku?<6{;@lrmcQ>CH#V~Pw8#DM-ZF-TN@KDtDTPnJrl<`
zdAq>mJ3lvss+HyS-c5i;<|CEx3GPBqOdUKgBU*FprMITeVv!tY2Y0uWiFN7Os)+Jp
zIXwq8GmGW0=a>h_ASTb)_ozX`9DIyqvW7kh>|EG7{bbd2P{I6n@RtRB+vGVs6Ts|y
zY(@OJemmI3JGshmZ=Q_$He1x`vfsclV26Yf+y}RWF(I$0iYO2!qtM(XQJb;tNz6HZ
z%xFLr1GMB4_8`tZ{;Jgy4HGu&Xl57dfg60qA{+y;+1Zp_gb73;k;KWG@tCl$k<#9C
zXgkLUBU4RgXJ9xpv381*?24QTJ3l{nDmB0~)e<RqJb71U!QXFX;=es}cUbO@okU&~
ze3fGO*jH4N&hJYli*<`}^r`%IL^%-os(EbKj{YnwUYBcx5zVo`QHz2WaMt>#i=(4f
zzck5SPL7*Uj8g=+dt~QcX!{J&x${*F_Gyuw*rHp=@ZKrinK?A%nJ~^FBM0@l*Sl0E
z&w$G38r16jfoWxri^Mqjb%LqwPbsXHY^ASDA^a;Jbbr#Llo?{bfB_LV%nw*mjoDa)
zSzlG%u2xZZ_d$l~wbvKV>x1r9$NL+U?!%jM<`4Anc;cO%*R9FDK}wRBt8nh;lL{^e
zv+^ili#Y#Bro_giFvSW_$HYpFPcp21#r4>2S!Rwy!8%$pIC}|6NR&v$x~Od(%c%;C
zD-5B1JZH`5G~~u*SzYPF!e@4&tIw%YPLAF{C%K@lBrZxBSGi>ai}@Iw?t&9LqbPO`
zNy<uxM~NA3Yc%+W>u#ojlZOiZnKjN!Z(7)8I=Z@o;fDk3XAH@?S~2J`MS8PdusXPR
z9)pIyxfR)b^YxgSX2n)P-?CT&v~lbCiP>*-boreVvs|Y|8+6QxuWvOku^>~Dr0EP0
zv6~s(|NIzKo?1QO6`9<9^!wa$+S^t(KR9?EkrM28cpvxIYF226*DS!*)!SKb=>hcz
ze%<1|C!fPe=12i@$FE$rB=VAv84I(kUOKML+O?`DVe!C(b<%2*plT7=Q?NAh$9Lk+
zM3=6tZtgb+yEYQ<(xrD;SLE=&VsW-#nitF!U%27I2*RRp95woWEy@{ywP?NR$GedO
zIfC}gP@1(nYkj`~6A3Lz_AI^AbO#PvrPj+$qJwO4!L!B0?V8R+G2Qo{$5QCpI39Tg
z$6s?4C5!c=KgtgQ6VQv|l+Os(EXZEjTu3+Ge45x`nj_P=<Q0ba?{C{)QB!vuV}S8J
zJDc7ZbVwKTKFk<*b)`ID&vTK`?vxm8=4s<oZ@6Hu!{_a%MH`kI29N)Pn&8I=s!3x%
z`rpd9ydgNn?T&!R(Zb?NKXeapt&=ocQm+(NGk@p_w}+AN#a;#jqg<Y5ARQ7QG1)Av
zDVkt??qb@$n!U+0>unoWGlzdHG`;#J)c;eD_+|TB61Ad0_oT-=scaZ#_x7fDeBIkW
zn767+=c|YfzPjuNZm!Y0Gr~o<LQCi(bbWo&h8Vd+W|rEh^6P}o(h-C|O>3>QXcq7?
zzCnEzr>3H0r>?U*h9Fyq&c75P+ao&^J}0XFVW(|i?^K}wt%F~<(9?hrgWPfFj7>lm
z3EL0Wq4fzuL`1|85$QBDO&i>JshD9gNt1Mr@#%ISl3Z;!cUArlz(6x&g1ANc%G*B;
z9uG>3=e8fKh~Ee-O>@|GWPYKWnEW)Nsz1iqTolzG?`Zg%tpgr}9roujm4eAbQ#GB~
z>U;`~MB>rn{t2JLd(a)s9|&y^{=>w$K}*zdJd+kFgPK`(dK;k}oh!=}e7|gSez;|j
zZ>_A1yMoHd!}7@G#g!wDmEk2WoE|aCcc~hmQV+>4HPuo`qvj?rx)5bl%K7QkT(&dD
zKHgYKF4mxcSK~)af8lUe@#sX^L`rP4sLpw19kuoOzO}zR%UXJP_`HwmpRx6l{&`aR
zeDa-m3;+6w2+OsfyLh$s<Hj>=F5L4-56I(cpPOvKbamt_#{q>(h1qm$wJsTf1*wx!
z@3-74yazKD8hOll9!b3m7&PQ4hCAu>v6}GuD}pbdMHUp$OV7;@fN{T7)z#+{?exZ(
zi^ZNJ3MY@7p<g@G_ej%zKmwX$@Y<#Cw8rnkmC~$DfF>xaObMxClE`GekN7S2&(xR%
z{KNR-h#>4D=R#a^4F84nLc|(%V1vtl)xxX-Z~lj<b8wIQi@JT|iP_jrV>X!BZ18LB
ziEX2?oyMM^X>8lJZL=}&yzhOUd;f)*?>^`3wb%YEt7RY+?O^9XH0kL2p0z3R<|Gel
zauHPT^w6U(b#1vW<b3vAc^OTKxjnwe;`!d#)7h`l^=t|;EWBi{I}dFr{P<DR+Un;V
zU9RT@S`qS3_~<wK=#B7x9z6J&)8-NIxaAkV(%pxh$l#yh1y)aaVc-g3w~2AGQQ!W?
zyKGfcHDkg_QiqE^t)E1eb;3Ey;iLhoB|d<w&&u*9vTm<^|H)!^S+XW9*0Dlq(}v?1
zskdcxa4SE5o7!BW$F|_06@@?!eRu}-s4%`mIO3I2ogRCwD3XPz3v~ZISJE}s_Rtm8
zK7AfLbmvW#=ono#D)Aaq-E+Ne$-#ixz+*z1u-*?(Va&8S?Uqvo(9zfG=FspE>}w=Y
z8~>u`0k-jm8=v9^nB!faVgsnvi4=0ws0|f<aT69&o=7)0D6;6*`&>rQJqvVwI9~EQ
z2EOc~=xppv)ZCB?SeG5^<MIDXTThIE$B~6KBFQGQ$=T!8_k8Mc#JGU}CO6|43IGOQ
z8R|*0&Ge2TD`ey|p}dG!qk;Vbj~Lu2mkIEQGf*^L3=Ypn2+@6>QfuTKF4t-J)ri`!
zJIJbi34CZQlmMHb`E7>g-4D)N+rxMKNipzk8$YdF!%9aifppZ6A5&1u=E}I?aP~LJ
zqK13RYk!0)J|K=u7L#X4BeVi&KEbej`oK0i&>ERPG6Ho&dqnkHb=`GogP7L|u=U92
za}HQE@p+l^mlE;<snCUxqrec}!&jI^DYc1%XAF6XXRjVhm#j_OIE$vWsS4?J%C#j^
zG~xG%@UEb9hXLoMQPKRiv75pZ3;(E;a@G%K1gxF{zWD6d_gj0KK4=-qp>j=9%mL8%
znRt`1zSr)H%g>|$XMXB_<B}tkL7@aVw6i17<#Ze9(P_TY3}#;Kh*$>}r9=@}hT^>j
zR{<vj-jddZToaUZk|Y>%YJ;(-vygh|!#QpXyBVmpa%#NHGH1};_`EzK6KW(Z71)(9
z9F-;l6f{{B=CuUcq*xaLT>-<Yp*#a`uqy5o8g)lyTd{A9z70`V-t=XLS6{@m_xFpL
z0MWr4dg8VnV`GPWop1ek&qV$=_Rd!k-gbYhDtB)xFypp%e)Eq8hNK%Ej1s!Hzq-Yq
z+vg*c!5T*!SYz=B;CPBw+NLM%2+)63Mu$;IRK<bQ#*2RSy(U@R!BOpYp)+N4W4KM`
zim?2lfKigsjYph)!I{+Fz4#)3E|o%!{WR#rh0nggG0rky6AL-tp8}-gLlhbXa0Y7^
z^uPl6)dC9XCld2}6lWTh)h>n^5Z0<fQjGNeO`1=;{I+J}ZPwQxb?voKvjf3UU=sza
zs*RBE#kjC<A#tDl8<%KDli47Ubri4n84^egaC*<LrlvM;#jSf_977Rl4BrO@wvfm<
z!Utpa8ku{G;J-ljSN%s3wq7atlhla$U9OAyBG1A;ktar1P5llw+ghB9`z+d{sj#(;
z!jzmtfnyQKL9SmZudj*2p0#BXsHr_gRdLE6U-1uWmoo;F0*~m7Uj$H9v*)HjXl}I2
zN?f)Y*D~JSJLg;)Jd@R~D{MU57JZ3ioOr!G5ri#)`aA^(ZPB-c5UwH?@{!w;W*q#v
z1R}x_!nqXY7^4O+uWzz0z1v>kj9narQ$%|2P-H53<l@?MJ!@L6w(OSuwu>g7NpZ_e
z{W8RK;U&a*VJY;c@j=)zlXeh=cdVy1xY900kwnJg&=9uQ+MFZryTfCphTN==%CIr3
zVuGo)P8TJ|Zd%JO#uNV9#_Ja|CH%ol=6wZLa9ceWa7dPCHE3qw=^QM@>zqoT@{3&i
zm^CQ5)}7@FqU=*;9n%P}Z7Q31T*b8PcD?o?i#DpSlLrzSMMA4t5HHb4&s@+ow~Fi<
zVkuHM^DHiDof4bY)(rOaVC%k}$G;De(E=f>tJ~GQL~I)j5QL?}{W(wDmOz-Ee+Fm=
z`8?D482z<5wt&Fxm`ec1ve&R(yec>Ew2f<*U7&+hZ4Py@aWQRmeNd}BK`RCTUAtcW
zyYOx=n0`X`cTfZ+5)%YoadxKHDPb5=gbDw48s?WyeqDtScHhv@(V1!At6!N2#>qL_
zc_z&{!i0s&Rtn3`CP@ZkDD$YWIypISw57N;i$-?1ClT0`->p1~go(d)U+EC?hiIeD
zSdrfryqW$GJ^tmCM=F*jqGE^~b3r%Uy}Q(n$<|b)B@fF3lZl^(5spLLWw<qNSv)K6
zlKL5~uS+$lms~R+$V^KlNNWa6&P=2wq{DzEm|tTd9P=m{GAula?M|1!DS%qjNNKT^
zjc4rP4KV@sOHlRy`LPl3lZuxLPGhOn;y`5{I?+V|*3_#zb3M0}>^kCqzWwZ3jVfFF
z-!y`o&=h(4=;;|5h3aK{^5|vED#UYS_XPTn3-AFZ=H3dL|DMK=K7-MPA24Z)&3Ij|
zyA}T}`~-Ho;ccb~v;Oa`|Ew?OX!BL0NxHwZSPQx3eV9@knP-v5q6iKG8c2qFi>z&F
zslhGP%?gbQA|F#tj6qH0rH#N&q&|wbm(dNIS)RVS;@}pL#a|?db1s)FT|fzg856Rm
z$B+Rz{to*Zp9#y5efzkV#p~KqO<dF2B3qTJW<qL2p59ibloPEYIeB})4dA?FKxrJV
zkxDwxe$nDavmO(411GQBizRUK5HfE9v$Yupz;PX6{vSGBAx^|fUAIVG&zYZDTd(h1
z-eV?{<bbq{49G5Wf>Me2aTw&-s#+4aEjMC|9@^u`Ulk*?e(9jl=!ffkJfHKq+w)K8
z3yZtHu;20^MN_8j01ub<-Cq9E|GR=03gq$srGTuzsR^Q3SoeRzfIR4MOj<>VqMUSl
zi>1R2)B5k4zMs&sgMpNz*sB-pd!k`Rf3--joP4|zPrS14jY3lp6%Cwot?`XzF`MC`
zPe8E!JOe6d*ncP8G7|P#N*XB_AsoGECzpL}*u=G3TG^B!W!+`{LDVM1I^qixP4LJq
zzAf{bCVZv<HP2x4KMb0}KE=VLlw$nPW#0t3nNx+^?y3*M&%4|=17$g<(>I!r`>Qvi
zkL}OLOjE2EqHSIlHDrW8ikBsu`rXzFyB<)D^fF;>fD^+IX2~3e(G0@~nwXdJD*c_c
zsjhB=w`l}kk8AuNHwT%<d)5+r;d2JB-#*5<I3EQAx6-957|Vb1{<;MgUd10F$^oFK
zsF<<f%ZO}Ps%ExMjFCH1GpTCC$ce+;n8~M%tpbd}XEprM&Avzq>|&+A(#`M|UTGP(
zx_+4npI+O##2<?#eQqOVS8(^<@u4Z$)UYj)Pk<mWOC#-hD+&Q{`0oW<ma4*jl^!vE
zLmv5Sy@AmK=3pI=Rn_+jRircceRYmx(!K^3p6{?R@l2$<av^M0tUfA!8nMu#f+&1G
z``7w>i5w+nyo9%A&VCf#j{mi|?j`q5Yl^Fz@G5z-^|<OK?@<=^8?z5m_Uor(Q^?d3
z{Wu(dxP=Y#(7lvePYqOaUiM!ezzOtW?v-Ue9VYID4uJ=cG1C_1o&2^>E>4TV!PZKI
zon3b5V#ZqtWiRwqWJ?v*45|@r28o%E8{^jj@XIjZgJ;Q67Ns<c9FnC|a3x9&truz=
zvuDZpaQLb3DA5iTUHJRurONe3M>61l1Tk>1!O3PNXf+qEmg~hVM<NyE#WEmySvaYX
zp8HZItG2(^JYEZbEOv?TMQRxEP1t`N3h!;bzvgb2y~mf1@Zlce9V;j(KU(2vos=CM
zej}Xlu+cv%gM$nq8M6+&r-TNoK2{C>$E4x&dSY(3+^IaUr$a;91umG>jA)FkfvVG~
zUb(2Xua&emjs%X}SaSjxPHZm>8SHJ8ksW=}clZyqE%F%(!p+)Zu3DaQf^abZ;yYDW
z9;1J?T$|onH96d7C$tUY`bzC6W~EvV6(SK)ds^=T?L?)izh&y`+iy38QDiFkM}*)o
zgC(k|ZG58Mf(7s*Dd%4oX01o+F-0YT1ifRF#ja%~Q(-)JgAa+ZCvt#^Gnr;htqg&C
z?Hf_U3cxhon*4mK`f%k>!Uj-y?9YUG-XL5ZZhiJ#bwLKXlom@w=b1VtmY-0?#l;TE
z;wh#wr80?W)G;Hq1?Q~aU|09=USNPz*Z1~5A?QBWRK&X4tr0xxnrDe$s|7H!F72GR
zBr@MG7zSmf+Js_~BjwL@@U5>&!Xmt~$pOkaK^t|B9E3H?7BddIp{0FhO!39iJVpcq
zXtIjVp5{qyk7OXnf<f}%=MwA*6Vt1^6pkTMav?_{_>6pYbz=vY@tec(CI^d=C0uj)
zL!Iwv2y2SPbVdU>ci!TK6*nKRLgSUS^FKv)2KHNMn$TdrCfe$EM>83Zh`WCu-k@#4
zOX7?$MVJN{Mj{rx{F)r6?s~W}02OFS13b=i1UzmM!k*8sK5+L)s)cKbN`9p{S}2Yy
z+q2GQX%+Bni18;<as%?GJ$3y&O~KYiOvT4ZXiX}UP@T{FNpGjGTVo00(f}*hSzZ?B
z8#`ziPy_<vk?=b~B$wGkjzD<!FY+19>>XTEO1#3oG^WUDmEOfc_!;s7eQSs7Yl>i;
z@D3$4t;ECu_V=4FG&I_>Ed42CAVt<GnJ}u}2!3^(aW~cCUgUEkgJSm^OZdIOUNczA
zk_DM`K8ym1?B4wYtNes$a2lhv_~Ig{pr#00PKo%h<9S{`MbAFlBy60Z2hP@4YB8*4
z?yLwcsZttt(l!;mV0x0a8o<HLvN-y`j-#9c`a5(uK!D-jjhzSVAp=IKD&$HggZyT@
zb24;BF}ZpcnYm!z2!hy9j~s*t*O4XfPaIFJ#r+J!;99na7}agn?U_CbvhL=!67C%;
zI%pH!yi=nUcpI>eewXS2QubBAQL=1ai<AWW-wNfNo`^B?RifT(BsVHqQlhsQSjRtT
zBnYZ%mA6``A(GlL@q|&mtrQW=%PXTR>%+%rM*lr&f=lIlJ~E8_?8f)&2PPky)1;$$
zDEZYLhQQK%CtA>APVOS4c+oN2y2lu~TQp732i|ErEbtLu;Xx$&{>{f3Rdz2>{(Ds)
z@E4_MssK{II|_ZD8zYD<eQQ<2{Z#Q`>H=iBW!=d?<y81zzpO%{lu7AeTJ3rW4?%~1
z?A3wKyN<zbs?Z%HadT;BD@4f?1Nuoz>RG|o@R$rL*99)b{5;Ju`YR}#@YRAzGK*X=
zwR)i&87BFRyij7;sSOII^9XdxjIs^koa(Cn$Y)<BQEU?w?bswIjROc^&s*DCWvhFl
zZ0(yAlPhr%wZX9DZW)`pD&Qg$4`Dp<MJ1(_s&~)-7gibpFZRu?O$UXAh12t>&GH2<
z3Bml&m4OKpb?lL`Q!1Cr9a8GX_$*0mjTprkAHS*>eg0!)XlJP8ww&f_QyiplgG^O^
z$eF*No@i{|24|^8nt^4@7R(Abd)6WQCTAS@DEKi_Pn~gR%`&B<l;M)G7@^cWDDa3?
zS$-iG7tChn=2iBn8}zFp`WBJB5FspSgPRsOc5xxQ%R~pWOW$*&SG>xc(NJvz+mZU;
zH)edq3vz0Eo-t-<DEhG0JxE?6iZibD+&F@(+aQ%0Ep2)g!nlG9d7VWzdn4cbf6k86
zN9Vs$S!`oQutIK`I#W5<Ce#0NYg)KyT5-k8IeAw1EWUh0egumA;o_{TQ#j8A&57YZ
zSzIKUqa`d*hmc=9ig3q=^w0++&du9|)@<rS+eH)(x~v&_eX4AJ4hQK*jfJph@%uwb
z10){yIHeGu8oqR`^prIq(q+^i64Glx62xm&Q2DMCMwyAQv;&j|0pBU(22J#1>W(I>
zSI^k%ntz>ndSAHNDW$KY$eYppUE3Q!>iVUmhV~R_-g(cIJrZBNVzx{)tBy5AEmM|m
z!OiEgkAWnakGyLUJ;}j@OQ#IznzSiVszfft8jtaA7C^L2#FoTV<0LH7-IWNeO22iY
zU4yno32{)SPj+QfH#AANZ)~64n_Xak&!<^RQUB9&W;mP>MADv7(A6*^VTLH!1KIBP
zwZbCeQHCa->O(Bn8q)<f^*qq$V?=U|IM(;v!{C}dV#7o2@roH#!D&7$$M;hF=~#@R
zo8nYoRHwhw0a#m)L<R0R3>&iLZP-DGqsqx)D)o!>rIr4Z+zBj2{p*EuKA4f1@ygR&
z)g9c_cU*R50mGpOoWyk70+3)M{PPGtMDHat?(OGgGm?IQNx)S~c0VrDazf7)KtivV
zH(-QzUPlBQ3y~HsGl@2E!*BdF1X43uKvYLbv565rP5HFH;^uPG$ACho@HW1nf4*-B
zX9^w+&S5_D`I6DfygA4%(it*5<x^5dE2KsE`SH+&+<+7Ce%z~7eDkb>p<@C%BsBJq
zb!RI*@4wN~jy^Q|XfRHNRFGTUgo%0JGuiEeIP_MZGqBL5+QcH8DwrmMIVfOf{zPMW
zWz)nCqFMic39p7*qSz+#>^ZXEto*_zdFUSQsZ+XRL9Epjy1_}WFy0Ww6DO&keQze5
zOE!UTzjRaJ-uy8|5Oq<4YgIAOD~MXDT19M!7qFEdANt#F{A7;EMdQ^(E7l2F8+G7L
zp~s|5sf1g!0(aL-NKB@H%h=jBhMsT^kq$t}a<M0Ep%as#O>Z6#jlrVRGNv?bREIYs
z&n@NEzXzb8aVPc16YYa+I(sKBk$Kf6XX9SByxOIpk$%xhVI+=~YQG(2wBOd$?{g)F
zwDK`hLF#3a;F)e$NIF*Eqi-*p#3iaw1-Gg)zb`XBs&iwL+w%Hq-o_<)5z|`NI(cc$
zzU0c6nH%Ck+JmAER7jSW6I6D$c_(MU9kIjLoU{DkFChFlDlf<JNsE5+2$c}>5hRy<
zmBrZy=>Ik-6~=AAQf-Tcsno}gx~senBZWH(!n^6nL^d`LvWlTDNLJhfoe{|`)e}Cp
zv53NjmtN%5I+o+l8$aPKQ+@AwvNQ60kJ&f}F%*<Q(fxlpM5IIHsm&w2+e}m<E2fmA
zDT4#49yiAlMzFJdZpTpCURr&@$}>*Id~UBK#G8-WHJAE0jvsG)=R{d9iM8XUP8rPW
zd9gay7%tF(8DXkulEscdyrD9RLTDIS;GjE=6!J913At(YnyIaIrjj=_#eSUcWK6t{
zydAR?Z~sQPTAY8>14y2DU3luCEED)Ih?7|7++n|wKzD&&2hEcOE*?1u3D`Rw1{^we
zLay7onmL0AIay2-X*KMR?l>KtMN5Qx6|BzN!?uQcx0DpJ$k7Y<nN-nHE=;pY_ypDM
zJo^V~YZ3`Pu1@@@eIE?DE9ngU*oyTG%bJ^pT%@y_o&EdS(g}94bq>GA?`5SxS%~On
zxXm`OuZ8k<kEi2b84?(V2gQZ%SN!8dKhwT7{Ti(8Xn}Ngz0)O2YevtfNUhaXuN6xj
zgJL;H{SbM1+T7ec)2kyE0|^mD8s?T#{0PE*#^E12r*Z6>d3`UB{w3-ejW`v8FItt!
zruMiZe;DLehU1j;&D%NR_Dft0wnDN>w6RShg<6E7j|44-Iu(q#wUvET>jlLr3pGKs
zn#0#cceYp~>mjFS-AM~7iYcVRdWA*9e7?dB`ukipwIbF8tfc+uMW;l-+Y81<;Bk&G
z?=#NIIYl+86d{ICDq{0&=t8F{<M;j9FvsBJM1eTr=q+X7dxyPB`y>rv9t+LmrZ2jD
znEt;7@Xj^ei`CJ^i<RSq#+sl3D!9|syKL-mk^kbM|Npk<+`Bgd-%hctM`~bI^QTR;
zdNsegAn1uz!OKKb)yzVo)f|P3zvCw!LFlM6Hs384i0<aH7Jj=XLq`lc!|_5_LFS+{
zy@<eFpfnz0UuO{;fptf4YGt(6ih^ZJN|EK5f<WPhTY?hgwHrh>k2y-gs46v+W&xy>
zHP)Si#G*@@*G431Z{Bx3*&mN9IUF9~t?0Kz8dQOPA!~`wSg~Ka5~SB+`6biwidDDN
z$=0Oj6&8BDzf@GqP@@YzVff)E+s$jMr&KxJj1GMNd8R4{+!E(QnsRp3MJS~n#D|o3
z<r%c{F~`c1D{(!;=x&v|vdKYEYiFaFpv@j1%Z;J~P76Ceg>t1Bl4Aa=OI@;a5DQ7k
zIy^IRL_9nzr_KFPIf{4mzpU0H&Mm82P5ZpL<RqO`$HE?MNj<)8k+uVNK}U=&#N`Rj
zS>TS<RQyU7@?IL=_n~szV%PvvJ7}~e@nSLbQq>0pd6z(VRX&uc<li0&-$~iWL8o%;
z3!763gRtMezJ+}5U$K2(*l9a5Fufn;$S#$2LVmB8CgmD_@vWs4-^cKV0bbNLbjr2Q
zP@#TZ3Mvb(Con2?-r8`-@_^)iL$lb~)YLeIj?34;icB^%_?E(o)&;EDQ5K1{ppAY7
zFg%>jd(s;Od;!T}Uchm?Bp61=RSON!)5V0A+m+{^x}J{GEDR`?sh0$hsKyeI*^;ji
zIG(O>`s{+;IcF-t=L|r0%866hq@<?_TuBv-R??8=vMRGNT{-iqh=?7GknTv1FglEk
zHOTq8Ce#MgqcAqTwOJt47JdXfZ?n>-@-@#?)IZJHsAZ_^Xlj<IuwHCrM{8o%Co@7R
zs#;$f(8C+Cva%s!HPO4v6=MR3x;lvMx_e}MPO}$&r@pC{U-rWjH4hYJHvGnjym(0%
zgB55}m00bW^p((>xhhLwZ0jk`Kgjp>apt5RAH^%n=Y;w}8cqFe+Lzn(_RVm*Zwu`j
zSf#+=u%KVTt(ant<JsaG#%5u=#j*WXAN>pG4~0WH%@`BN(h`-@M}wmL8!|E@YN_be
zA$1V!E%f(zUS_hYOVDm$k5jbjwn`pxzynSr_*g;KBrk8c#uKegoGJ7|2-ub1+-jXJ
z28l>bPY*I9BF1Z<W|k~c&HJY+63eXxJsTE*8;&T>jywK-^a*qSjEkr5m2<iJb6pB1
zU?1)*QNc6XThFWUjR?UrGRzd@9puBw;+@yINrMzri-s=rgt2+%bALPflSQ{<IFmV1
zj^RcSuZX(xssM<PKMH@l<lBI;-`c%u-RX5zUlc+av|};8x{4HIc>DhbZ`m7u4e&Iu
z`LI6#VpCL;#*1w_#A$lur&o!#eP-W{oV`=MjvdY$gT|Se$cs#?J0>PyX!kQfbg|QE
z=^}W1_<B70gNfEbwSknR9O19!tXB>V_8zVc<Z~&$Nqj@-4QaVK#S0d*SN3~e+{s6v
zVDdX3x5g#e0*cr@_L-al(olLtKV_=5je?G9^sYHG(<4N6*7n}FT_N3}5>}d=Bg2BU
zbxkp_tVBRqVVl>7qHeRK%)pJnWwX93sS%&&sVPt6_s7_gPmeG0s@uF+(!XtWk<{Y9
z5(MBS_ox60RRIhMV&tYRt{^5&Yc_SZXk!+=&fb6`FxQ!DBA0JcFN!^xEeciycr?kb
z(<%`Mq)%ew^x03T^UlJ!L81<iso`}t4>JG?`{7_=N2c0)`#^id^BtpgJ4pJ{#Lg?S
zei%a=Tv3FPQaJ6Q&OwTPOberAZ(oO{#bSG4VVj;d%Ev#q<6s>b6hoUDx;DQ~m~)l{
zr3TTsH~HIc;zeGV7K+pU!9XTb2(F=RZR1KmW-Pc{ss+tJp4r%8XjsiGArtJPM49#~
z5)73@Nn_k5>;sddoaggcE}C!gBlS(lmq2%70Vm(HuiHg+N6Q!bhJ_qVNh1CC!Kl{k
zQefyRg>bxs4(9qn=h3+;fQnsExYOCHBk^DY{F$-|dqa1;6ilUV$LPnixProX{DZsf
zgih}EizR)(^^an#$O^l0H}`SfvAyM4CQp9aIBo%=Gnt{Gim+%fAbJ2fkVAOLd2H=#
z$$TcWZprz(wzU*Qhp87Ro<<r<s;4U%6Dn~GMzPMLo>l4RKT1zsYVk0`PAF7p{2FHW
z@1)-FZ-b1oJw<XKFDb|6qf8?H)<4T1T#)2Y`{@#VTb0bT-ZaF<#inIFzu%ytEAUKC
zcf~@)Aflg2rOkftuBM-{|5mSszMy-oGaf6YiB`Q<*TUi%sa6U*G1|2?h2N4igCui#
zepDbYOWCt`PMDn9ZGH2}J0lenk~qvCqdzSerqy#}nrV*J9YQM)@fUi>uAGFO#)pjz
zy&+QRFii_|dPE=&i|jP9iG`m1@Lg?L1dP>Rzu7L0_L435_SW(Js?U-Lc-Mzpyt_=j
zMkN=-BX=ONjhma>x6ro@u#_?XDcj3Q$gKPwe{^tHcX%KZQFLpb^^KfxW@T@y4&Dp4
zp+^{kVULvkRXh?=G)M>uea&>Wtp=Pp?Zyfi9F?Kc{pc8ed05nWwbIz-g{Fe!>B+iE
zO(4(6o|J|B=P;pxELmCgY!y|*7Ouvk*U}r{3stc?aRbrU{Wys}pCgB^Y-F)PfDn1T
zKrF(1YVN9t_{a+V+%Y05KZj1H!{-|WH0Rd<2CE6ekv~I=tAv#*<4RN07Tk_!6wO#c
zlyQ*ohvW9{dtKeF0afw&jPqkmR(+D$O1^bIn)KYVxg3Dyq5zK6TzW>IBaRAC>RW^Z
zNwUsiu!7PgmWOL0y?o)382K){-GumVZCza-BdbrYH>nS+UrDSAnwECn5J3(`VehV1
zjXUL?Vs$(YLGDYPFwOT2@K7uV_rgvoU5?>$++$Amch5JbLM<sANeLB+x|M!RG|P;{
zBR<*0<gev*(ZDgkW4-9Fg)KXE22WU>YJ2%%q`T=yIi}v`N1--0F;WcZ21te9tV=@v
z2K5}A8+PB?#mSp5%@~hYwV#QS>ZX|&&lsC8AIuLDqT;wKPO>f8sFW>rLxxI&mMxO0
zhkU48jLz#eB$z`kigC_ozP+te0mUYrbB2|qwN;x&Hb`iS<0{Tyb!Pq$dsGIMjb4<4
zDlGZ^NHp++0|8Jfa?DJruVHI~iw1sZM?0?$86;eJ$WbIiytC2*45IqtbX3u?{yq);
zfe~EbO^)2xSq*&au^E`xv~-)Bjw``F2eABN2w8;z>+fDvK@4axXCyBr5%&yg3~ueX
zC!~Q$F|5cRR*j_Zaa=ze5JN5--V-u-f7@%Wo&0d`3QJ%)*7rCFwR!VutcSP%ag2{&
z#KM}bXa1rmUoW7xZyC;x!DHPIX7Bs>rClew{AUMNs8>aX0dvr}h<XfJYOi*wk{fC1
z#=;eLbVkAvJ~ry?nL3DqXReU;=u|xy4}y%=BXV|-b2_&OIvL>!A;nyTZm*7z-1QWC
z{~ZNDxj3Z0*N*CDMkXh!a4diVM1g$<XVQ*x-&xT8u0gOdXfY$QQbAC$zk>m-gWKl^
z6YtdZ+{e2mI+C35iCi!!V%qg1mOtvJLui!L>O)kr;tN;qzh-C32INObhY=&dAe{=y
z7@gcZvDh5?xHkZ$t|PE15FWh4ahGRUMouw)(yy2D|Gl6h^sX3iB3}T2C|J9ld)%uV
z11wC+<tE(Q*iI}+bLcg*f(C;n7FG>8vqpHDv54YC8G@@Y>^xC>g2S#)O~byLipupm
zlIb{O(&n|*^`#$mO_8Ac0=niX-Srew$OqvlQfhv6-{MLtd_hi_eA>@_XNXuJ7H|t5
zIl=uQc*7hTbjR%D>KT3@?c4C`Wsu$3Qnd(3T)$0c1;$<hC;HF@``vTAQ!XxJ{-=%Z
zs)_k*AJ}t$xMRfl7M8(gh;-Ylsv%VS>q>9=ux*T;MEd5;UBTJLfT&5GYKmJctIqy}
z+>W^$Kj_R8aKcG=7ty+wM%{;Q9b5q)TNR!ZylE(+ZGrpv!@<KleC>9aF0-7AHex+~
zosd=+ofUvFYeb1MavRMxb^QikDu-z8Dw0<uUcl*O4RGjoB(SpuDGsg<ehact;N=dx
z_L15k8D?e)Okj5I;dot4XVFr*_U`(8g%D%o=1!7L@mLI6#8ld!%c6O3B$X2TQQ&<G
z=#9Au6H53X9_C^ONtHz)gpF3-;wOhTO6{Ait*-vvp)l%<zhA|rkEan=cc~|pP(jHk
zj|C<f_s#GgDV6mO5QG*-BHZ7>IX8cbI;hpdHtPBK0ZBoFTE&N+LO7_|JgaG`3yI)d
zH5&e}L~>ieW(dUc@N(xgF?+|Q6GUP%A<6^eM#za@72D5t^ito4K|vC`V1-hbyuk!M
zS%}C^9_hGfR%kvyO-|shSCJ__MyK*gVmsD1{(i~H&WD;K-?<jFocEthN^E=!C8Kfu
zf@!6T7O~$|=&#j~O8ndBt@kAeMz;wdK+>yR^L#pXw?vp$kZ0rk;l=gLD@KPXnG+xL
zYBB>m!V7lb=I1=-8aggtA7tW!kTmLPW2H2$2`mx(*J;Z2dJy^Ptw#_#h#GIsB4QZO
z>+r*twVjx?^sXFT^{VaNVyo3bKtc#VD>j*nR^ZPx>p$oJ7!DJA!aa!D2~kyMdx`@$
zF7(PLZ#tZ!Dh0PpMl13IfAL0z*I~{W4qLhRa~^Rx!3^Od+X(|l&iA2nj+BoHMYpks
z5M#^YE5$ozOn}F8mOr+6h14~jjAY6OFVJfir?=MgIQd5CCze+>cBbbkAx83ri;E5)
zlAxNR?q^R|0*+)#K1MFfl$y0IQ;;)q1v3#5k&V6mm22A5v+gf=LK$uR>QG&Aoj{K3
z`L7=BjA>s}x?0H4rsO5S@aX~cLk1aQT)<%VgHiy2NEd=Y8%B}xzr4Bb7wAM-0sOoM
zBgd|72@R^uYEJChl`La^<?+&R*{_u1=oA~b%2GL^Iba`2+R;@HT;$olWgNzQ45(Sh
zi0@6+v_`I=AbsK;B?0^`s+G$<+>|e}ctO@XbQLj$t=n7gJvmi57RqG^x~)#XcbN&l
z;M+a~dMWr%n}3VMj~T>~6@@g$SGHwDV0i;ugR3%T{syKhyS-7U*fWfw@0Vp|9&dwh
zo>4<rfBq!(=9hByA}L^!k6@@enlkRR{Kl!+>G2%NheJ1_u>=;Y+q1Q>Ma5$!_q!9j
zs7!CIP+qzv;LK;*T2-L+x)Vf)N(5nco^8v7oKX`C@DGJN@3p_$zo(9q2?a7^xj`Pn
zlkQz17y8A_<np7zdq7qPV3?F<kW@Hm`*+k40bk%@k2m649Wfq-sQk25<BU=^6YF)>
zFE2q8Z$2sQpaH`Q5hU>so;SO!B3(Tlkf*ON>yM)f0}~r96BjPR=`+_ANG9PsVP_`B
z5T-0)f;od^PD;eWq4dL1#ekz8kab%__!jfB<A}8L`hfMOB=`9oRTDq>8jxi?HRwu>
zsQOJE0aHVRD>NP^DsnGLsP~)K9)<G+ZqrrfmQD^X7jddCp6=?pVaeJ+1DtaOk6N}Y
z@z&mH-qfI>$e1IdRitQ+gO8E$#yj#`NY|Tt+C~%XV0DYv@3JYuxK({46+oz?k`c2?
z4z>qcogeB-Ii_ZhMrCJzu|vAB18eJ?1TRYfdQuFol#=p(Xs8Z&F|DZ02`!*UE=DmU
zeDK>ls!C;p<Xf819B%zWxeZet=Q-v)5q=W;<^6CZ2f@4ah-%El;C`QwXPY!3cX(TN
z?M4dY2sjI<$R-)dpn6*#hAUWJx`zO1jj`|BxU}bZ?=54yj#8K1-`GkzZ{Zd{AT#{<
z7=x)47MTi)ra%e2<d%)Zm=Cl{Dy^)XA-Mnz$OMy&I0Q^>A?{u_mbXL)fXTRkCz;tj
zZHd|^AZ|NC!+`5N3iuD7fxf;c2~F}e;Utg|Ps78mrQ4OJ>4$d8!*@#_aiFjSk)kqW
z8dAuu(uXgeVg6*4WRsQO`-Osfk7?(0627OG-3b^=AnENGyB4<<qLFLze9Gaz@k;N$
z<%~Y2!NS=HS9z3&jViVcm=>lpGG(^0cU@UINia8cc;fS&kE6Irqh#>@c1)PaF2mae
zx3wJ5tL@hUg@7xeg-b~&+7Y3dCG}@kW3O2mr@8}b?furpq73bV2bLu|lntNi!KcAR
zEtVJ;=y}(%{`&aQq(H_OI!`rdZeVyfF~#0-Ly5|a;izRWameR^#p-}jfng16o$5yL
zgv(wR=Wz&dABA@z)}BBsT_Ybu0{#SRA#25A6N9GlqkqdV+0(I8jH<F!RXMt<o!iEJ
zP#fskmBdL2dc9u=LcDFx*ohtri7^bf_(^p8{LE|cot;?Et_CExH4uNP_f<>gl>dRF
z(DTaF#;*`a)vCXFLLS9`{vZbr{EoMt4xw&}RI%Y}OPPV_zus&Dn7^!kxrkaP@($kU
zCdbQA$qo_>S_HOA&*j2^l<)b~WfFO%w0p+ih^4G+fzn4pJQlO9SE83bTiyvqFh<;U
z(y<)vN0^iT?@>&`H*`$!XgRmnzPUSjaSm`C@*b|f!{3!`tM4woJ>N>rd3jRW&+KET
zEhR{gnj6$xYH?u4Cs@x(Y?tsg^6Qq|wztr#s<`jn=pzH^2|0=HS2j2MRnYFMLmNW5
z6ba&nBvfEElM}rUaP?nNwjMRLSBGZ6(wbUYhn?P69Y!i8h0NH|$N?j8A+Gr5DGmB^
z1u(@K8R^Ahd=0gezZYGGF!PUolK>p>qtzg><y_DYA^!lj5Ykh<@*Nlex8Bq&Q^^hH
zGNnb27fl-RCf)f!ARBZXEu4pumtXgsdiQx8De*Wt%RQQ$ru$E*g50E~Nhil*0xM_p
z^51?7GSkK<Y3FTF32#c=)3w3dn2GE-?|0Y+{(Ohyd20owtG&_mIZf4QI`(upY08kG
zj^e}3{@_*Z0t<!2{)4;^#KuP4NRKd|^?t-m>UY#&mm4dnMNO9s)k09DN3clVdhLrh
zp)`t@Uc1Q7yf6Ph<Z<HVq02#B<i~Yy@rQR6V`~B(c`N2pxF3koA@pM_r&rtFLirlt
zVH7dTNG>C9-_(l=QfxKsaTbL=&&Z7!)JBb^bc-gf=Djbqm<7xxD$y)ll8L*Xq`e6^
zEtKv)#b{G=Cwln{|9Lnwvot--*eNMH=O>w-BkKZ}hAwF=&&2j%g>uR;<dgwDb#zmh
zitfj@cgw(kyy$Nw{$y;r9S#Yl&+jho-T<hkEx#<foHyFFF*bqkV+&<x`->?hQ2t7C
zWBB4ONw4>vdKnfzC=8Lk-&B!<VD&c>Z8Qg{_Q%~t7}2FY^PIyAZn?^pe<922Bek?5
zZ<=UjHP0}83dfFVOa7|^RMdFJj{6pijdWafMuEsYgo_s?v=t$mK>y=A)|$Z!HF)Tg
z;9sZ6@Ryy}X)^LRoJ=Vu`yP!cOa3dt{<<x`AyuKio-TpzVyehLA`5Fc-gKPixQD#2
zqYT~g0*T)(`(BGRnO;gXi<yxWaU@Z~1RLEK9p`hDJ;e=}r<c^MA-!m7Q4t)3TQEL)
z%kjABJ%t+83VG9V`|JO&o_tzc4wFni;+jsk5H~~(N=2()mQzu^4ocGScruD>C88`R
zCyf1~(^v=<y|{k^Sh_!EVkxlI`W)d;)UF7@!m@RF9wH2=qJnf&vGS~bsS!ou5U-cn
z(Zk--+PLrwg+7UG-TN)J0z=e<bFvG$3;sZcu-60h?zAJYNtB+Z%I{+A&HtmUD{jlm
z;kxqGv5yw)7+sw*;2Q-y!5?Io`!J7_IYvYd*V{eujr-awn&BT;aw3kTTvjyVTdjd9
z@sQ~SYsopLmdTQn{X?brFIDIw*vw3#@~R5Xre=@6B^+K{qcPeoZGplTn$fU*DzYD%
z{MhU%@%DNx!_sjG<INOE%jb`eFVeUG<d&J}In4Y$XAeg*f<Ruynpl%Y_5fCD)vJbG
zj=)f2=gR{K>RPXgO}I<M9e6l>$HWb>?0b_QjO?Ht>HEY|$raQQ8}LW_q}r)opWx++
zU$&)kmjv}V+zJ@%f4=xTsWim5>^<&;Du${AH-xsp$p1-VuH8^<e-65vG8jy9B_3-3
zIh1~;X}zgl@W!pcRe*m*X@+j>5~(Duxm6^i>Y~1?*~+5oSqNs4WFdHPAH%Nc?6H&!
z2Ay2kajq{H3(Y;JpTC$zM;qBNS@PjYAqEUH=Q!XGfEHVcm?l`JucT~TvGtpq<;izu
z+EUYWvS5)|as|W8sik8{2JQt-wn61Ir;ibBZM>rtHj9qF10KL)VF;5Xf9}rX=|L-J
zhQj@A>XQQ9?`{dvih?q3nfD#*Y|9tR^8=37HWYjD+9#JcGClyobNnPGZHTphz3~&R
zQG=DEvvuV2%>UUPqf-2=n_}~`$Ka6B;GI&I<8aw_Xt}FCE<1+1mI=m^;Gkgq`zu_Z
zYj&;iJ1U!KK!IU2CAUxz@e4@CDVL8p){C$5*IZTM`h9%A>zisJs+_?@!_<;Ua&np5
zYY7yT7JQ`ym-a8%im_88sohLK=O^qLt55c7{+gVDp!b(9A;~BCTth)^@WVKrC1TXD
zb3EE#G}(fujU;jSCKB^j^%9D<a={N&zt5{r=Klmtc5=d0jKWlKsMwe;b^5JCqTOG(
zO6XczSK**;nU!d^0D~eK2fVR)^jNztGRNK~OFj)9hh^D<2F#G;$y9?(owM>P8@9U7
zLErybcx1Wy59nvEaSBv4Cgwnnz&(#;qMfC;I{(L|ra16m7Z4Bk{^tj;7M%+)5@=f8
z7PUaYGeW<cJP5%G1Vw3XiJKF|GOBb28CG4><LYTnR@pn*Sw{tgnnSV;QVuV!jQvjN
z>(tbMPnXiy;<-WSzVYU{o{6}KIoMeIq}n(rFYHHBi2fH;S3CJbq<ckGzCm>UukKHq
zbixp1$lGuK`hGq4P08&x3$E9P3-E#!8<Os9L0)x=N>An^Ws4|^DY~sKikO^kRr!<c
zuX3Gu6M`Vfz<7*(@MAhKP$cg^&-0gm22aG5c-1w&popd~l^XFnl16YLxY5{Zi$-;*
z+D4+_;^Ww{-V6<}5O$L5YV40h8r{tcW7k+_j$J0xO2L7ipjV3P;aQxmR4ze03SuP2
z;ToKfgMkTd{Z;J}YcAIJIDz@N&9U8CKQr2lH|aQ_2-TmPO82hRDW5BXLtD{}Quc*w
znS#I`Hsdm)8~M+=Fwg6mA>QW+*_@a8u7eiHa_KtB_1NdL|Jdxs&`2^h*wFYBOCJ<D
zZb0!J<Ti)Gxww7MUeS@63ype*6%me<F-Tr0;)PVA(g;7-7YLNccWpD-3!h;#0ut&~
z*_$AthY|9kRr%dV$(*Seth=jKu{WjkCz}f5U;v=t%N+nggsXCmufCI2G0&GPLW(rc
zWmm$J`(g-SBQAW%DIzm%XK15!u3>%5Qib{G4hV|Z_J;0SG0f`kcNA+|`PLwsprZ5l
z=6W6rvM?d;@^Hl}S!q}Ev<-VdAkgyE!RmCdmV5B4njVA+cr!~Xq~~p!H3xnl=^%T>
zXC690|9t%vz{~Xt)z4Au@(oGvYjVEIb~j|WKBGY9tjj`zX&qb3Lru9Qc#{PA+2xdt
zjeWl(rC^}aHKqm;H&O)<Y7y(&l^7O75%tv`x)Pf6Vsi)IR$D@M2;smJ1%-<!EN=JP
ztPt@`bc2~>1EJ57EAj{7taV^m+(4~a2cyEzTRnPl>~)`LhsCs>+cVPFR<6Zc?|8ib
z`ZYYewWRq_B{qg=3HI~{kxO^LU2k@z9#`Nj|F_Jnmx`^uaWel1#UHm2KudJnWqVS$
z22O#Y;}+$Fn+O(fzjhL6#W#xjiD=+UIwsB_WnuR{wV$8}%!!?7^3k0Xrlr0ck;gky
z!k1(x6}cP8-%h%OntpM)dCmRv=IQWO38Lkw&yxT}T<qYYX`0~H+TfTOEo9_5hb`bj
zCz!+3)v7xi6~LQPNIe$QB`U}Su4>6-Y<2czF(IB#Tlq3-6365}Ndls1X_YwidFw{-
z^o54fW1v7(j<%D#-Pbt+n-k!;&fI%Qnxe-Cv9x_x)Zsq0iaA-(j;GmKO8YRTSZymO
z^@79k^o2_?F0aoOjF0-KHz6g(ogfS3Lq=;q^Lt=s`c=(E$n7&VK)$lUjGzZUoC~6i
zt_j@OjHg%P^S*>`W%J0Hvf!eO1=TOwHwW8PFstciFe{-eVN2=+8@-2C4e2dQd$}ST
zp{(LWdl3YH0L0?>X7jr8!m7dfhhmRRuNlm`6q-a2TIC_-cbw3|+mj*h<IF6H@My9b
z`C@p|3lLKcRZ*%5(AVxaYt9?#*xK5DdK??({rJJh+T{@#y(6lQQN!b<*Hg&pNKoO2
z;<iH5%KGNMM~75$2`_-<&W*uoBwISR&fW~uaZ23h{ZS~S<n_reDryQi(XWTj$;S!P
zyBj7n`dgX4c85d#n;SFi%ysD1Mw~I!9(>_yepkzA^A?rhKVq%&n;=@tc{<X&2Q%ML
zUc!8XrZ^L<py}<U%=0_<;UNMeu$x*{O?M)G5Tu)uRRkn6%Q%ZqXauV|`n62q8IWKU
zJlzXuE5Ji%e+B%Q$;s$P1;%4d-9REzhp67Vi@uKU1w$!y$fgvhiE(ms#v~`U4?C)c
zNtnTIpep&@&t*L={HU&}kxH?O-Ehmw+o`+KyW0DGV9KNa&g0w}WnfL|9HK-S$plq3
zkw7(s7(|Fb<|)kTb6Q7yQqO)9&$#tT>h62}9<IB`L!Fe(v&Ty@fi&;`XTLEw;L`3f
zbPLSu^TpSZB!YqlRL7b&ViXvSn81G0BlTmZl89%C`=6`gv3c4KK7m(tk`WAEO&9#D
zdETK&V#=8KUl^Y5Q7F&Ef43`XQ&|_cw{BACxxbR1J+fb~5IrA3WG1h-mMq!AeLHJc
zB|O{%gXbhDOG_ug@GkH&jjaqdKZ*H$0w4CF9yT5^43hlEqmy>LBNwiOCdS1d>->$N
z7`ZCEP`Ca~tJb!y4N=NK-8f^p;<5qUyr~L!_s(0!j(a;54fHOJzgIn<dPG$(w(nT~
z-MyJO$`1dZj!xiZ#^B&?doC@8g^yvcMxK$$!V@#;9!Hvinlv^ZHpk#_3q^`6=oe5n
zEhAlg5#~Q;5pOhYYu%;39`Vz_X!fI}SD%|_X83<$E1$z{G6o(w4d-0FC)J&qHFbj9
z3v*SqPA5Nm)q~!{?5KXIDL6;=91;|F+$y(?I8WL|f06ZfCSstOC`H6YXTHvlH4>QF
z=P@-dh^WT0Mha$`9-JnUIzB66-qc?S%KZ(TT3LOAn5Tj<V2zY<2k(4q%MHzs27`J&
zhBo^SZkmSM1fvy)cYoy-l|ECQiU#7Ki9Hp?{EVWP^>k%3>-k4e%s^Ls=x|11M^$)D
zax13htOwLBoQaO`bOR5WE7|6mAEo(i+t2zRaP8fB1Kd+A@M{*<4%W;koo{Yw_$&D1
zHr)dM3}f6gz-wrXY!IX3cuq13us1FmI@e8QKQg?PA{HN-HJ8-&o9Qdp5s0=0eCs|&
zG39QYrf$<Q<b=kaMP<^Oj3YDbK<uAM^8gxqU$`=|igoMb1$5YZf=f%z5T5RF#blC+
zl<E5n9h?VBpuD6wV|o*y4P<O~5^3S)nfL0qe>X&{zz+!oMilqnz(lUcA5j7Qp{+*)
zb5H0C=#i-5VyPH7V#tC`w4{ts{OV%&yi9wHp?I3^2F2BR+*jV_uWp~K{6j$Kow;CX
zrR(0wB;u}D{Wt9x@vCRHIN_VGeg`5%Al3TC{&cevGOAT(Xx73-JLe^E9NXK+9I#b$
zZ(lC%6}MRP>Z;bsG#vuua3mfvX%nws1v68p!!g6!FOZ&^A4qA7#teP;<8azt`o!80
ze!@OqL?Zx6$qtv}b4vEwe5r&sg<}cK08uE)c*8F>VOXbGg0)GxpaI16T+O7*7@0?!
zEn}x%gLr;lQXxwJ9T<V{`n(SR+1Fu}>B=glaj^Ym)RKoKyWje|CeVeBe<FPC9|{Wt
z|KBeZ)t-atB&bqs`dd<)jELs^OvD*`!Z(Xr6S{776eAMhTu2JBTm(`?t>F6e^ysI2
z@iSrc+jc|ie<cP+uSNzJ2?uZj-!;9E{(8Lq+y1{*k}2r;eE^}Y{qBU-FqA;*RYVS)
zK5Zcvjnn*%>IYmU5;?;#f^RYzr7zx0Wq$jYA?In?P)3q#*p`SBbWo?GRce)ij3o(w
zFXTW@9m@5{s>=ElQc<1$57Rk@kX7&#%YEG$FV6o&p1AX!N9dNZOR8(BsH-iCNkm(+
zl5^0qcT`CMx$mSTXP@zhZzeAh$A+b|(LW|Ww5tyPSX;<Nyu!Jzg^Vu^qkBNk91MVD
z5G)m3vf>2Rmfk*0qCxt}??+Px)Qa})OWnJ+Lkuq<7QemBW$K>gFERitL5M|BCLj?M
zxjvxQP}}f>tNU!60tT-{uMgSksZ&0|B)GTdB~g>E9G&vx&L-8~AO;cl@!w_J=0Mbz
zXBOv?aNLR&h6lra5t?<0{=tsG<@N$y=g8~~Xus=&ebU~pwy|NL9p3b-r^@sM7Ec?Z
z{_I}S6>vap%O`Re4N@>I^MdF+d!5+m)u((}0`3G&KdA^G<7Y)JBF?xH;Fb}?DDk$J
zou;_9URw#y1qFn0whbQNbp*B1SR}DHH&=`fJEd13Y~JE~Utd8)!XR;R!eqAnRhmc?
zH}%@ioElO@=mT~GHItAbGxzS^bsT6}E_{k)3H%sYHMUTrrl>vMaT0+@TvG&8R+lk1
zNp(Y%^Ju5ek6o=!AE)0F%hO1xb#zQu?+udmHTzoQMe=xjZfhhS6=;s$0V*7Y7qY?e
zCR$?2+Jl*yJ{M}a#@rDnm*r}iIeQ2ilbjH|NwVyp3j@kY?iqU85iDFY^OmPm8fci#
z>i!2Q`R-13;VM-c`_#h@^ldvE=L61MQ98syo}R&at57qTdgvdxI9d$X>yEJ(*8Ale
zc6CZG+o`U!5u6yLu+Y}h)O*o%lk`H~p3qmv?4*9zzCIr(tk(k#SDZ?yNt7V}iM^tu
zA5O0wMO8vBueLr<8bYE~O%r)b;`><9VcZOkDbAjT=};{G{}4N`@7%*{p~CoWl?+<s
zD7R-`t$}3C?OqQ|bc7Upf8xq?iD7a=NTKVNDE9GU&N^G$CIlZhCZ3&Bhi8dWX@3Gg
z26m+WSE(omAFq27C8bWWdq9Sm1tJ7-+7r8B8Ptq_;8N(FAbwwJ!rjF)6pZm7jhMa0
zu++2zX1L6hHd{_ko>5cn?Mm}4L7*#N*0r_NXaBdl&m816B7Ew6i=8oo=*oOh+?4-m
z()<~>&|SWPC|<_q1Fa6k|CDvlhF>^i^BCp=9HEO<;=~ePObIKTiO*iI?c_YH+VGDL
z{T4qrN8D-~%G8A7<$EgnMGgxG9(_zFxzSVl#wSdBVsAXX4*0bFUazfKsjJtr^T;iy
zBG8ySII1$6s^2rmH~aFWk3Gd{A{ltpT1bGy5Bg3pgJSJM(sz->91P2xZ;z#5H6jsk
zq0`ya-ixQCZw7QG<3tCJI{f?QfA{%uwR~P-6vdyP)l)iU1daz<UZ+p_hLGD_GhTU<
zIqEx@nuf%1R!wJQN`r5r&0L$sW7r=zd<7CBmQOAd2SA|&cVTO9qc_;~zeRtzX{H(E
zrQ+UsSvQNqSRqL-B4%TLtFxfVvlCT8YMmhJ{QX^_qw*TqGg&rOdp%vX?M^G3I<;{(
zwKC5amOO|J9YX`-B|bnucaG*12&8tvUR7*%@eGStNt^sHCUmLWS#K~%=*A;_M0K_&
zP02hkeJ$2Vn?f?<_dS9sHU)JHiL;S!f0y?QP$xZZNz2&GA1~Wmx23hO6z|8)G$;;|
z<mwys41>Vya82KLgK+_S>;s5}pkX$sX?~zQDroT-_@*E6c;Sp-(9P7hvZIp$l+7gY
zJ;tvx&3U6O2z!#p<1hF<A+H>|l?fzDD$GSy!)LyOl~05c2b|MR>}Cz6Jt6Om*jNXQ
zVD4)x)3U`@15%F(?qeD8x%?`h)j2eos04BdY6?zWDlGtVS7Q46;PTLsfj$w)jLct|
zKAZe+wr)-@X+wT;r45=3d#Vh@qQ?2(##zk62tDP;89VhkSO<Ny%yGdW7JSLB6JQ(n
zLi2yk+d<MPuM!cuWFIwbNA!OTE}#h;SD2`FS%a2E3)+k?z5@L&wBzDt@_luilKco8
zc>LOsF;-?ic{2_3IAiy{*~j!EbgL<HsEc=9=6{zs{F?{J397G;+eW`;;2HD~4vaQ-
zmlDC^R^{UlG4S8~QJjD383^zFk+w=&!>)iKJ1E%1%qGnDf7pAgzqY!lYqZ7Py}{j`
z28ZD8QlPj)3&Gtf?(S|y3dM_Skm6pPV8x}lojl+Be&<g(S7%=2Ccm9zueCDwTw{$n
z#_0YZ9+U32fseWvPv>t`Z`Fa%2@ktHFJHtS7S!g{EHK2}|0qYu**>@`x9(rmnr|xS
z^<3Ts#7RN0T`$w-YPzmi@WBaDe?*4uYZmy(Bd~iMu&A}V8@n;OG8g|LK*6ebKKElo
zNNA9|Dnx3bgoagNzANtX{S-`Ep>B|k49q&i1YmLwjHZnsvYUtJ$nEL>Glo{{5#vz}
z9ysq3T`M)AbE9a3&)K81F&-j|_^L*2?TUcJs9x$F%Wm+S&q1W)r_{-*i;M5bWIggq
z<WoZa*6k$VK;Xw!b4&)%zs&S@JK-|nb?J>X@ZVFxuCK}c=i*MaBla1N{mOIUZwFI&
ze2j@i3=GH5p*=SNXCeWK6>0@mFKMpI45Zim1#9+bhSM&7V$SEdO>AD)HkhJE+by>m
zl?1gPyv86%AyjU5i=c0piIxt?CU>Lu?t@;^zTAU#ocLKdq&99jNaRN!REN8|Utl6~
zY-OooaSV-&@mdZlGs16p&<Vm$A7m%<E1RCmkJL$UQyO8KTKY=CH1UWXv{K2WWT2UU
zbn(a?pDTx?>S~O)TWA@%ycLt_tqcLIw&c0Rq8~#~_?VP~kG<Rar809HMjytmkNr+}
zB73e(URrZ~E)fInV15-7<D#Id%r62D8*@`!6rSEkql^#BKLX*Z(UH2ylWmo8^>K(0
zBiW$aQY!HPyNj0%o*|C{)9*4xMo0})s^y;&L;BOnaRAhr1lcQ&Bf)<aba!HlN9yjr
zb&7QdY9Kve&QjKLv-r$!yiqkO1V8PJ>$A3tyjz(E!mW{Iav!4XhHN2^CnBS2btLaG
z^{M!*xsFtbYS=XWny6UfD_FCeYd?oRXS_WH{<e8L?b*NUxv6lw4hspwa{VoIHv`7v
zUR{nTAYxOse7*_15KBJ;j$Bn~opk$Um-VqsprW_DL;It7<-o|MY44H#0}q2b*MW;d
z>PI<BSnoGD#tTK0JbHv<;oqhH$-iP&mz0&1>e!diMxKjv?Ku2z*d~5K9$X!2ei)T4
z*v;83Z5_pRc@icBa<|QRJ;z7>axg@7(BrhY&Q<k|E`w0)8~uPB9wDvgD=pUl=HoeH
z8S;w$DlQix+U>(_u4ikVHsFt3-sW$I>!8_v-S6sms`$H6RK{TwHQ)vs7*&z~EcAAy
z78tQkhXnXjJYeaZsOp?JI?0xTuOe{`OLN4nqWjx-oiUe3WK(`;u6}MN5x}?^%_J+_
zMeGLK#mDKpoOuR+7wY=DE%Vfitm@<E;ThV@OA-iq(vaYd+h9N%*vlh(!Vj~`)gDjQ
zfGX&Oc$YgnKYx20nR0(s%|Fzq^6>6s)dGAhyPC=?o1yrUabtUTbaS#n*YNHOWaaQY
z+3_p)S%;;nVuwJlmTB0yDFJx#)4_@sYSn59!w;9IKL53LXTZ-`wS1ot;vZ-AEoa*R
z|4cbpg(-6PPs?vF;f;B{CV?lWX-?4IHNx5c>4NM_XErj)r1%T`YrW|1;Q-`sU5)D*
zy2%z)Ddx~pnT9^pF;bJ7Z2S+C1(V^3Ph<-oJ7P&6nOaS`pW~a@>~<#jOHd?!zx*6e
z$Y|BepgAcuUP#nCS@u*n5*t=&{;$trX*lyoD#bHu1@Mfe=8wI&w>`)HHLqLDYon;i
zmqN4X(?j<WCpU8!);|=oG5^RVBzHdYK-c;*LyErbvB!<Db&uJ3k%lMHh0A?KSl<qs
z2%sddm^X?j6)#@M0xkJJ672$IN!0TPMh0VLaZLiiF-|bo41CYa74atvvA21>mFRC?
zEwU=Vap>KBK9CYm8g0My{G%3o&3I)o{2*LY&?DW_5aR2WY?fYpO(0$ae!?wKK4@z<
z|N1q%p^0G3g@5AVE&tbVDQK>xpjF8Sd^y|<j@*jP)v|)WlvNtHG2C^5sUI3Hhtpyb
zZQ5!-{W<qdUH_V~UVZFn{XVoW(f@rDp{1B3rs;CMpNmu@0|GLlh`_mvZPB+y3uA9^
zFQrp2aSGXMf8KL*!V7<Xz>&lA{=q9@&ZpR3n`FfO!Ow2<>&h)enJ4oS`+moUCLopC
z%U<7eNls4JqsD^mj|_ArmSN6-X$v<hrl97c0mTv+ofki@f37ptF2Gk9>V!@3VjJrM
z<7y0_c!df_e~wr4TvlwkKvPi)YJo!??4c6KgDt@%WQu*^%~!GsgFe%G*;-sv$98^C
zS^=2$2xokp6CDNm6BeBV8z*ODQ`0}nM+?>f**aqw7xv{T@cWXs!0WY)71Xd`Y`9e{
zEzZe5AUSRmSHmXO%-ucBx-kl;C(|39a4#07<?^$wEi(10!sm+>RGd;<p!=rE(eGGm
z<$m$t^v^vxt~=rXjgJ`xs<V39{~OOB&8T0|{~Lxuit{@E`^Flac5z+t|HKYaA>RM{
z0j=-SDyu91C)yAzr9aF3PYkh3&>yk<Z+v~vqM@hpKLJKp|Np<@|BHcg-iX5HAFluP
zOR;f98s$>e2sE_XYcTE%v;>23nFEDhOjhovkpEjoEFU%>|63#89RdMz((6Y@EJ{jB
z=Yw*3&lpBsK4~vccW>H1svMsWV*c;K`|pZpx(_|H#dRx=a2LXFS0?|x%=tZKRY))q
zl^?X<y@vYjT44X|YLpSpwRn4lM=9b$ifyJEyjdA+4ouU8kvbQtYKM;iwdi9ettc=e
z6SQ4SlLcYPm=O#>zNG!GhZ4*!D5Fu>EUSpj3~P~T;zmJ;cL8_3+v|}3WSoLf!)^Oz
z+5mo!ZNwmkpUh&cK!F}-*0{&Rfqj^yo(PAEApYvo2zkmqNwYWxt|aD)tZCRl#d!D7
zsIlEA{aoa7N*=;miaLnvcxF&+9@2C5@%8;vI#yaBv8g-w<+_HS;8)=Y+727NidHm8
zD>H;`Dhe7F_?YuH=kEK$Ch*Ew@XTj&?`Z-ZOWBkmiP)b(v`*l3S$+CyvrLJ9&u&d*
zodmQ+FYja*=fO_`)`ceHnekyEGN_H-y}|_6H6;#_5sb+p8mS52dXvyLubtPs<kmxm
zic{%0HIw=1m&oV=rEXjXS|o8ecvGmh-}pPTV>I#T>9I`Fr3`F;7Wcx1FX~&swtfA!
z^+XUtCMz7!mLL|9QvUJPr+2B&sGULIj%Uz}MC9;nvPGt53{w8iVu6$KYy|amIWGEB
z-a`ERf@x9+NTs1)_#)cROWqPk+uN5M9noqUNF3@@ryT3}y)^=|il4JMFzg7f8Q%`x
zpqMN7)-ch?x1r6a&5NBz!(8#e_tYYQB}O$_j1rZ{6`9lvXPLZSJL=jPJS-3y(QU^w
z@!O4wjdgjPCYl9?Oi{zI5oMsJka!asb=n|Idcq}9lctR2B~p`VkERSx!a+sD(i9P-
zh=3OJ9^>7xZOwa&v4(2Phlc|Wqb)6%fU=tZ9qA6R+wShKY+DidnXzF1`t7XR$R8%&
z>uw9kx49jpY3(`(mh@#(+4Ay=hxYa@i<-pUuM9U3Z#}uMwF<|(D<*e70B_e5WpQ>E
zI=jGZ8-9S*E^y7J7a)|L^>(;{kZBvoiyCvu{~BC_zQad9$x~NnwEIx;%6K*NuoMT)
za{&aNQ1S>0D?Jcra6@Z{8#pjzGcNb7W1(*+qUZbPuh6&D(1dR68tb-}{y(FvomYDq
zw(i-|qa4Sq)%se6sWPbV^<!I>HpdOLUQdTj!$OJBwvoK*B5>dw5QimlW20udWs=Uj
zCX<P?>BdK592z?Q*@O#tsY)C^dBsbaxim35r}l#h7Rh$ltO*jLAf&Px(C&9(B^QA3
z^Hk`j&K}SMfUU<Ykw{FWC12BQDf2$4_3jB2)IzF3#y!UCqoG>PUv#kLv*&w1CdnM(
zgM;(P5rIv9kc9oV<TXLtCI%FnKc-Nuxmv4u6<j?&)@LIMPDEq2tTIVt(lP#V@T*O+
z7!a-vQ(~l(8Qy%Uczb62KTnjU++*zSz%=-XyX)t1_NI96i|r2Yf1<72!oo3x+&K=b
zd=8#eX)ywDDi6~qD{u~j8*#+)Rf!~Ah~@HJ2A|(V%8dLrA=l45ho>)V8XR@U)MdQ}
z^|~gW%kA%Vydp)@%xbC0?*5)S)^{{VLwmbl(D@7-NpwmUa~W#*`&eQ*n65}13b~Lh
zF#EGbA&8nmiAI68(KJ#Hafxs>(rDo>MsVW2cksRGZ&S?<4f<c?V`ACZ+nlXQe+zWG
z)W;7iDyc{YifkQ~HLeBhsD`nSSMkF+D9M?|$~weh@fqhQ(Q$uDPdY%1A8Xi;IK63H
zzTZVZc1yo);1&#~7-n1ZM@FX9w=^$D9OGGzw&OQI`KxKGAZdrZmm2zqBr2%Ov|4>J
zNOdMl;wV?tKfc<CteMOV=S(o+%Gf(zO%wwRQ(>8rwQzi2jPX1M2FJ>nat|lR4xgO4
zk)AnH8}-e;ei9l&82d8)bxPey09O}xHyaqkgeDKm0AzrYh8B!>@o@C)nrNv6gLcYh
zU%bh6cjSqut}8ztHQ7r(INpmpLO~9Xc4N+|+(Or%O-ZNRap+L+w>tQXOXZwi(3BwY
z=4H~Ea1IFj*>Sb9A3`lK>NrNObBm{KzcBH*YB~DxdYxOGZ%^XiUEOf{p~b==asV?^
zO)(Xc87^pN?46#EDV?H17#q;z1gCMXk&jGFGabY>{<CW|K?ddK&!Af8K5xLG%~E(?
z4Ag7SQcb=l+{lJRD+pJ<sJ@JmCHz-|K4zx5)0Lj;>{-{-*Rx+x)ydCoBb5F}Y7fIH
zYKi54RXR+>(Lb6)wK%_Uc{@i<=uH+!a6g@O#DrqA@d&Phfa!g@|16-(`f;wb(6&^8
z-c=IF>`()dq+pV2Nu+VBQlKyTHcEQBb=EXKQ>#LUs7lw2gnoE|Lvkw_1TWQR9XN?M
zm=I@?;Zvg8L1emN3n0!8(2a*~ubc;WMf-jM7*cMcN9}y}InMYb<g-iu0n+ntTbTJ*
zN7z_#E>iUY=g-T2Lbva@f2=lV0|XWqgJ)X>`)M>z{+fi%21#!hqN`9NFcCwrHO5%a
zkzog+^6$+M;=R87v{FGErm9)*xZCr(Dl0xi@X8-j40iA@#jxcQR-T-k&B}r7rz;UJ
z!5GiwBS`;#9siSsuVKa^0^k-JDQ&g2X`0KlXIj(_$p`DeIlwe=vQ20&I=eC*nVLZI
zlA8Jmh{4bug>YsnPQGw1VEk|^+_pZ}Sx;!xsBo`dTYd~aO8>873W~iEQS9l}XBv{w
zG%S!oDS}*j|722`iEDVT7Ds%E;^mP-n2+!ODG&I&@3XXGq3qG~jO;Y|&%@P5z}EiF
z*QXxET=`l3lE&D=zgi?X!={dcy`?xL#)b{}sl1{H^3-*d?~^pQfv|t}SyZdbmI=Og
zqyx=)_I#thLn4qVr?J~3s(B%X?lMS(ypc<I=!@Keoxq3PPm`txI(2n5_<P?%k<TpO
zUWHM~@Cjt8`1CP!iBM?OeH(RyWzG<Z957`xOl$jGMS3Nv5_g&%+8NYKlb39|he=|S
zGySAJkKy9qtIM;QCZYh$1h0=2Htvc{$~GL~;X;S99je3e!Uk!a>+khTp@lyIVKfSj
zFw%RKld+_Y@ba469j$J+fBtJ4KgDyLr<P?1P4luy#Rkd<K`E@yt}m%feQ8No2NA^~
zsXx^t#i(F{x$2_p-R7OfcOwnLkSoC?bG4CG%{B|A9FjPT&tVgvJ{elOe+d^MOSB?&
zon!whyk_>D9H)Xg=L2zj(~R6fwlEeY^kVkQeH^MeCJl0EUnKxJn09{S+6r7Itr#qj
z%{R`qsCoUlzN}I6XH(jSZyoLyOH_s82O9%3K-8$q<LWrm`igHraTnDgwRGR&p4RSp
z0HtJCFq1^j#;|V4v_2Za_gqSrC^n43q*37lDcUgeU;ZI8RFESSGt9BH%qFTO(sl~!
zB;eu@q1YJ9GY*5z(TSnKZj#(N9g0X{Zf;uN*&lmFLESMO+Hh$jVzByaK`LHspML$~
z`>_T1B0yoTBzo0YgN4#e<O0E`Z`?#*#_v4+5r2I6p~$Sd7LfZkApYode=jFKM)^ES
zdA)W$m16Z#tnO9H_3Ahy&=$Zc?jFDK-*6u%6kfXw{rv+}C5!}>JS2e~1HAy(9>d&a
z;(;Y#jJ?761`5z_P2G>5zgj>76p0L|u5&8!<Y!##XpysFan0yP8z&ZM19@^^p%<^K
z7i6cD6p$;%SSX+W9N4I{57;76NBAe{u%@ubzZn40EhNqOxb%pkQSbkk;DPp>wsL86
z1DK9i<g@|oep^c5=Jy5s9Rix=!S}c!=g}(p9klj@LrDSHr3XAD(g=8daz&-3TtpUp
z(X)wvf^gw1#uadf&sJ8<l_hiO2JJva{Ps>embV@Wzc-d~mwE^^g?4EM^fStaLfP2K
z2c6td`V*Vu#44sdR@UMwBlwAK^R^Dnk@w_{fv(Yw8LCN`9bxf9K2)_J`wUQeigk>P
zp+X)1umPY7<>ehSqDFG?$c(O5=9QU_W;$!CZlZ&VZt7I!F_X5Ldhqz*iCNPxYu0z+
zghZiMw}<O(n~!8Ow(rnVwazZ=45PiY+j|RZsTd(7abN->WY^Cno&2yvY^fj)gb~1a
zBwNg~v1hwW!}#@7*O)2SRzB2|n|lUS1dc6RTVR`D{~-z~&MmJLEwWB+>t3fR>v!)u
zZoG0~l{MDu7`9y6R8yaft>Jg9{k*rUv`HWDfg&O)>ZYHWW9sr<wq$$fi322WFPWEp
zkQD3by%$~qe3@-L5%$WJk~?FlQKJHc6QD7Vt8taPS$jAMz-<B1m_Gv`?>kyCMjfJ`
zn-@lwT6kmM$*~lh|F&%k(WH<ql?S;IyeA5Wxv|E5^^CCg{f?muG6|e=FdSS=KTc7U
zS0{^+-XkkBHavY^N!nwZooJZgFi^ug2cFc#>I-{kx$*7X&%pmyX}(~(XriAQXSc2C
zO<JPZbwOMpEL%sXhq-MFxu#0hxAK|%Ap(@C`x@G2sp1%<8s6!q8kX7$WaxSC?T*J5
zQ|Ds`Y;lXOntvysb!0jUMqst|;s`$GiPX%0dA$~S>sLAv_e!Ld`^s43uAUE9rXB<;
zylbIt@>c15%8xl_0!<AW+&x}1{3yeb4+d?GXg;IAe+xW%{io1)+<lH9Zqq*ADBiB8
zYgJnbGTVF)XC$MEr?B$VGQ8tU#Fys1v%kd9!@p7S;K66~>tfpB8O~vHQD_vHwPq-}
z6Q+!qI;cV?Q8O6-r~0SMl;7>%hHttz_$(!hs!!#e%o7*RyU3dF2ki#|QxEK$=XZK2
zl_Qy<=7U~s!P?Bm7&yh@Q&<utbo|4}o+ST#=0CU=(Xe8ePJ_Y(G}|HhIK@=t8y`wt
z#JQ^;D;f;aU}cqBW5CTZ#xJZ=*`kq?$lw3KlS%5d3*4+qglAoMd<h~xoDMd6^3|vd
z_J|iGT47Y<%P!a;cWjWbN%Tdh|EH{*{*2%P>@S8Qv!7z!ffCCA8_bpwuEj0fExHuD
z1U*=r`g+HbEuXzSIlc@x;}VB@RddyX!Tfi2mIWdW04^jtLHrIt`LxX2`S=Hh{y4o5
zZF*nITovDc61y#Mjyi{2=3Bp=Mb@@{O~#Zn>OSCOMOUH07&3u_DE=KFT73eGD7kb~
zLg2>FhCgovK;4l_#56GM=*{iQhRmxK8|$ic2&n8BaHHZ$Vk0^)cE&sZU3m)np8<$D
zjA&|G)RW-olVl5Fa4Zw$G02n|1-PiFsbGH34R%D(wo<Mj;h}J((!q0P`*2_58ws81
zBmdS#aU<I?*np@E)LO&t<wm;G*j^Scz4FQlDv%oLm`}%4vLVa$Iiay@?8`YfwsCD$
zif7PBX1dz?Qd!!u5)R1sJnL1O9ya@|F-wESFo)_*VbEm!I1o)M@txIb>r4{Zr0qq(
zYBPB?g0D7X-!#hG=a$>hHDgxITfF)aW4LI1v6`h^MzVTwON;UiI3@bT_p#1+d#uBW
zhr-?2+qTwlk2@^7njkoOPX*<K;ZuA=`X%2$3>>u|iqLpSBAB9{<d~bo@9XE@WdovO
zq~?9Qnk*WB!gIOE(x{8AhWU)i>0F1o^HV4oNTXi#NSSu`rBVU42E0WVBJl=pbU8`T
z7oRIsl@DJeBp?)tUW*%vK?;0j{L#=QIQ%O%=i|Ba_6{>$VHm_E^*u1&U%{ylC}Ksg
zCR_A7a3l%!9X=>xE2=l5q0X`0=Y81XD9p6$?WpzjV=1Nc-~2hg?_X+~PtMP&%jl{;
zM&irmCy*X|{(ckLr%qd1HrJ(6;hM9$WCGoNfYB(oE25GSd!g$ZIep%g-6@*#SrVj8
z-ku1P>aYAvWn(R%%4bw=)^T5?7yqlZdXK3o*Xi~FEo@Op!P|@z>;|cW*>dVTS_i^o
zu-`E>2`sQxVxF2N^rHqhZwG<MaEgR3i!xH>Kg&cD7}608p;+0kJueat3)AMZFB@;}
z=f|)zNuaB3BQAz3*5Fg~V(Tbr!A%u4OZ*L!cW9a&)Ckf2?pQsfz0<orq9SRH=GYD~
zkoFSs{CIX{^Y-ZJY~6qLz|WFKv}pJ%4@*&^ByF}%Vc1%cF(<h^`cj`d%xZdN=QLD!
z?bj^_tctz7FG)x|q)()9;KkYA;-4&{B^*2aZqP!otAT{RJItk#tj}7CqfaSdfF`+3
zp5=aWgT;h4^(TDFWqN%lD}wSMC-BO5H^ne$fCwHb&wnzvCk%(eBy3!rV%1_yl&GJk
zMq)($v(Q*<fPfOpkm9jfO|zSjua^l<6&g?>KVj6dV1zi=cIq-pO?)CnO}sw}U#ZkC
zEHS$iC8DlDrVXXD|LG@!Hj=v{VqG)~ys(25{dxT<lw3DR9j<d=mvlffA;2cX_$}zt
zHYl3*D9-y^C@yEuvk0g78yYbuo6F{&R<5`!Cv-Fte84hPKgL(<>dRN8X|fv}Aud_l
zjG&1V6?u^FpI$*37(#5~6ZQ7e8W?ll`ENFmv@cIP0_Y(Vo3(wvsz>6Ka3Nr0<DcVu
zh6;=iW4wdA69p1^tx?B|8o=CH#7H1^09u(|1ziIDOHts)+Sb(TmS_D?X4-WHEGLm1
z9C~2E#pf{BW}9m70xY#y2#36iar(Iq9k2HLzI^@c8?kVGcePxCopLCnj^*IW<gK<S
zSS*iQg;HBhy_|aV0&MwKr9mJ6AOSef0E&5^9UT5^@j5DkJS-NxiaqRN@dECIk#ofY
zQK`}V%&@^m`X(FX=MZj6`uMJ)Eb7TP-2Orxcv!#Tb31Zntw17++06;_hW8VPY!Env
zBqAA$S#-8~;7Mf!rHTNH6h?tcEoNeJl+0ISRx~bdKuq}=lXMuV4Pl}~Nw_2m+dBF4
zsrw)HD=v$Dz4so{fLVk*%RVtt0|nAL>UvQHMFm^A&aj12t^K9(oPj>HVF=x*^BKjn
zu5%Q9(>uwo0TEQJjYdBpmW&6*gvs3SVh{$}B>g-=_K8d@$CI&?@$yUy(pSj69t0&Z
z^<AyuX|R3jd6Z-Oowvhj{3=lJaQqfj+7gZIrd`rpKN!dAwCx{@lO1yLv~9yr9(+ad
z*S&Hg&CmDz6&>ZMroK|XqzsOFB+E3<>g9HeD>UGsYt?UOE+Oei@leVao2K5tRQCQ(
zp3;D<sJLt<3FHu${GEX3H;XaEn~$spvrNrtv;toUxx??JI`B7)sS|o_LnPRm&P|T3
zZy*FS#4FUSejgGV8ajPWgqt$+{yx|KyT(tI^f*#Xfi`ceFg&eE6Tfe|$8^SNqR;9`
zz34cK)VXX!1b_gicNS7l%<<6PyCjXua$ME8HOp98Q5wSQrMJnq!Pf3uC$C2fldZk;
zXXe!pPx9M+g1_=OjPH5S;Rj<cC8^@@{wx^lBOsNqf#f1iPq7<Cu!_t}XEU`!NK(AU
zof{K8w`JDb8b)`a8I?yYV}G30DbkNmw>Q(qu~#D$?RNB2q&3=bN+qfiH5{s<#s<_C
zF?kp~1t$q%@=~{Vx84t&-J|=ffgslh3de5c^th}X!*Y!u)0K4ZYR<7Eo6M=x{t4qd
zlY+k|Ub_;kM2<Sgtm-j(C0%cICDa20(W(h1ePVCCwl9h?*24Wvgr}F)8}bMcEmCGG
zOKqznb(Cr7@Pb?O&GKhR|3zv44!m~gnanV`Co}Q6o%6dJa^hMJ<V9fE;695GC1OEv
zH*bN^XnVSQU-&xIlhn9UOFfeslJK!sRO%3qUZW22$)N^obAJ*T-u0b#%pm07t5QOT
z9VM!16qraHBa}(eYlpcO;Ph4{BNKD&bc*gy+Ld7$*2(o5tsLQQ<sl-JXcV+=`sADS
z4A{<7k2BCI#I+Au!5|xfbJv$K!r=i78<8~q#%6GgY-1c=@<JIBdNrf`9MZsc<m99h
z$|cUtEg6U<ECLwqb(>}5buuI8{ji{SKe72rZtV9Ue!g{v1JTr?L*pDC+~8h2Bp<lc
z?QQs8Ttj<BL7QOUuz?KM;j)4o-^`#Iz!eXb0?2KKLaC9F-$On)y`a{n`7&Kum8Vdj
zKEQqwOPoChsJ=m*W;S{#x_FbE#!oeuWzgtDN&+diK)b&?1`V{&m(nWWj80cE7v4fj
zbO=>GwI4movySc=8G!zNuGrM**$en8_qO140htO+;}pHK+wgkeBsu9K+}OU`UiPPq
zPfHE9x%UUAQ=-O)!_0)Y@+5_n*z9*T=YXOSOhZ8xoV6NHjUP#~aX}eguIVVGUR5;C
zT3<-H%-CDFV3t7X2fH$oW>S*<>7cH)%!~MCI7d}B{M6{mr5rYP_Un@aG02~iso8ZS
zMnyc91tetOoPS8Pk0s1rQ-8~lykQ!Fh6dLvoysE&NUebqsq>XBs;=9?Wc?MLkNejr
zXPc6SG1`S`5<XEErg42Nh93IB-FELnJji@;>Z>K5!4Cl&dkj^dFg(7L)bK?>_{4+B
z>yY?fhWHKQ>qNmDcS7K^Yg4+~j(?=9{i#4dYG+cU*P<ZGxKIQq1_&rk5IoJq*lRb)
z7J83+Y_l40gI7yaih^71igb<{xdPE#Yik)O6sO5jd!P7I#IzIHMv6cfpEfmz3tAMe
z*~}@Z+elPC@aMvndhV*tT=Bp(Cv=ix{l#DCNthtKWR;wkOzo3FuDI^HtNeDRHNL>2
zr}z0Ot1<)Z*@d=XAEHU-AmWYHTz>Glpr_yCpELE`AE;8@73+z|EM*xW|A(VJpiyZ!
z?Yrm!JD#M5SE;raef(#6I4%A1Iou()!GhPPK&iVe7U2p`XddP%!)F}kDNDX$gKAz1
z>~VWdbsdfGU&4sCM1~P*TAAR36UzA+lXyTQB;<A{O}%=Kqk3@EbjYn(hj)zH2U-Zh
znJ3Ps<g-~)^d&d9TtM;3K4*^S&bu%PUBO<evcW=-j;oNht4$0{Pu|#Ru)Z~B9MStN
z$d9L}q25^^w|TOtvqGGCd3p8V%y(&&!6MlwyfmGvc{qW6!@%w2Q_n3{N9)IjXX2ib
ztA5nnRFW?Z98DJel926=!!c1dG^MZTfROEZyebspsGlk*lc+H91zW)Fy=V8cjt_Ec
zTfjab{01e7#*&y>D|$5Yp<N))7`YUpuf@_#Z@@}}d;c-q^>^~?O5^h<c4~t0S_qEV
zu&M{{3J%?13FcAx)KBrkatswdB?HZr=tC~axnPuV&V@-l=Hh`(;R_r1mZrs#pO4V4
z3T-XG6JUGicZG%VbBuU7(<AVl<hGR*>uFE{pP9h5naI=WbA@q=$jdxE(E+P=HR1jN
zQKhZ}OfwP{^`n6j*$t8Yc-qIF;k77apy%c2^eero5$n*Ln#d@Ag4i(Y<_ky9LxKBV
z-C#rX%q0c{3B1gce6qP#v6-{IT$0+mjYEkCQB;h^gM5Vzwr9+#YKF1RNMT3lxX34`
z1aqFE_WB^*UxNv9+lOi{0=JJfJ-&H<LJW4=^7hOunwe1p!V%FTtv|XaCQntw-oiP5
zHr0nT<Gn}%(|Js*2UU%CfFQ11A(}F960jIP4M-6$M=;0TbG`RtR<vy>uJM?FC@goo
z4a7}~1n?9t^m8`7e+T$*fMPd-D|#TsX?Dpmkk$lGI)qZxpQqgBNqqjPc)gD24<*#U
zFk7C$Sg~^!r50DSD6AiTV?xMA?jN!akxnUor1}!(!t~W|-83?XYKIyQ4z7=zrGh=N
z(S&v#GDg+Yt1Ck!_?14I=7~)Xs1+H8x&DV@$%=7yfrD;bLqy*yS(4Bk8GwPCB{jzS
zQ1R#EO&ecN_$oF7d$*OZtsC2pUUiaX>=LMW<Zw4|as#YrhFlzg#f<Z6m-*RD*=ALv
zs%z}PhnU}-#=ASscaf#N9dPI0fi_Xqn76msFkaw0)=2k5Iv=+3`rXM8&4|-wXXAUR
zndt-r@zvXEoI@s=DiUUGW*v@19wsrwki1&zoE%}c$*Euz<RCDG4LnI+Gs30crw!ub
z-Gx-n;=5oaqF0HqiwGhlRj;3;LBbz5x6e);<_%Ts7g!dw!TZW7j1>NnycI+wSjfOu
zHCRpY7of=On1s*VSCSwq-?L-`1Qa!!I|XLE7Rt2>7tYj}#pK>}oL}@`2M%)vT=vMO
zrlS`-HVPe&DZ9z8wUFTRkWlx3be)8Du{$(w3zw$}(d6(<FaU+OZkwI72!#D^Sn-ET
z-oq&npeCbS<Ub72e`|&4BcsIKAo41HC%`+HFf&Ap*S4qhNx+rQqj8VbrN(y)M`b^_
zVjT9h@MsmP`Q18RIG2<Y(huxx*0;bOVv8WuJ7$U%?5sow>Bv$g&X!mvb{PI8#p+mV
z^T?~6_?#@Om@RGlxv1MGUn_&s%FHV1ue0=$q{e)qu^9pCRey2m?hbs$-*^UPIg6(B
zf7l+RcSy<+<jFqx83!jD>)S(4A0PHiUiniaK|1YpXT*zmX|;N_{Q3iQ<5~NHAVHku
zexl%y(1eina7B6}f)v!re{KG!{K^djJuiyGmV9ua7%7&}<al4b3~#e(2-O8o|FQ_6
z#`v3uh5}n__Zl>T&G~J-R(FrjbrjW{O>DK;?qe2ly!XdqJZ7I^bIVvBEa4Uy=F0;3
z8SLRG3AFSM033`V87CkaDfA^qi0CHtqM~vjkfvswFL;_*I6bkrUSIFZ$mXj>H5F_8
z_z7GpHjV>!c!XGg<`x%^NS{7;m!Cts@yRROV^NjSw#STZhV?U-r^Nd4?4F*#GB+E9
zr#K|Qy5n34+U#n*_eUAMpOYfIADZ>X$JWQ+J|TK;##p?CI(M$`4N}Lmw_g%0?qCnG
zUf@H4n)>etG%|9Q>KfY{q-K>B=3q#NC0B?AkOBgmuhdi{AFeGg9@vVqbgmz-T(S`Y
zQdNC){xo?ujK$eCz4POx*+eEPvUrf&Pdwh$B6InC2psRe?+?sZOMteG*H-^9i#`p4
zUM2=vL}?=`0y*;ZcZ5U5hI=R>UZiZU)<6$R?)WZ9HuI!%sJ%91glTv}dv!y@q^iNi
zKDfgdc?(g~&YV5631lC)_?^ukdGDtlMMa2A#2p1SV#89|;RrN$&_GH<KQ+SIA~+_4
zv(dvkt>&W{&jh#3q^U*3s#E9hvwe;<-8?}|aNxTa?&dXoyC`sctBH5sgKV5)$v0+^
z+A_i;S66L=FXb=qSyc1|eM1rge+`_!RBsMG!lknPWx`>V*_N7ijHRBzsR~m}-ukko
zFQ-^+CC@TIdcI7VFp9|MpcD4hb;@6@1giC%*ORYa`!Q%%Q8D-?r1k9U<)}xFyq7a#
zQvCScP;l#>H<2_AtjYL|2k4G1-qb%)P3xmc8!js^=}@OAFt6{FIL4brWYqq1bnj`B
zN+$MNQk1@PAp5y~>O}DevEHk%x=Q=(-ASZ)DAR*<abad`Tfkn2QOBdu2I;p@EY=uC
z{j7qDt9$t06AO`ZnR{t51mUy_{2+zfTaPT&3O6Bg7O-NOrkR~LE;ZdJF)P<=4!dH~
zn+_l!v}VeI^Gz^8>=IC2aVC@58+uVU@BC15l(}dc3wffk{AN5E@b0{v1TCzl5Q@e_
z#si24tqPoL?n-&I{@D6WP)fC!Om_=|`m|XHQwd^~WwUiP_RLOl(XU+3>-(YF^^_7a
zBW+hjcV3tIWT72lx*u@`#^m=MmZOktaQ|o#Ab$9G+ve<zDuV~7cAm7MnYks5ur)5)
zGD#gPiRacWC!k<8%z~gv+8OL8vU`-qy4D^5-)X@kZ_Ca$otwkh{Y<jp5y}*0%b7Ps
zBA0DDT@U7>D^MkM6OvX)t2D6sYQ}N#?<)3eA4ySf<dIwiS7zp5U9aHauZ`I7pTA0x
zpiu0B1;Kh6%K$~i?_<29<ks0q#gR2k>;!;C%q$#g%eW>|T;>Ve<lv0^&d>?ue*!<Y
zClI8`+Dd*&b#?a|(bH>0`}5MP982?9jxQQ0o7HcahC^a_DwEKP0tMuP2%^Vr^0#;k
z+6Jq%m_NP_bnKqrUf8lICT5V<-xh`sVt&i?1C@qMgxv@<U8S;A9|nKm#g~$7QxBh(
z<o6aCCB%z@L`m%4qBwHz@89VCy%Df$`400e!!0fxas6}Z9md+`d%%*`1R|r4{RXdE
zn+K9ZUH{6d?O77Zr`qQGB;b0D4$Mi)GV5(|_d9S1{LL6pr|59$IQk}xXCKB9y7iYa
ze=ACka<5F!c<*E2fv4AHnjX_DFUlFwq#o(l^i)Dhv*K8<y6BvApl&`5=W<)ac(^a|
z3=H;=zuR_>h0V&&(ZYYTLVgaIlLWb%1<U&8A6{9^?`KP=W!_%#4C~KILkVFU!cj)8
zZ+Tc62GrjdGfi@_MVe_UI+~-Rzgxa=Y&cx`^$K$a=2YYdeA99u#%;$^)FIx;xZ4I^
zDo3C9flYB<Q2>;t*YDTks_%D5;&mc?{PQ2|_fR0C7!V3Aq~1kxNW=YzbEm77{G)T2
zE7Ada=Zz&Yt#<RfAD6l~^xPGkd0XorTN{+s@o?}s`FRxK@*=MSJUHmIH140rjwL=5
zLu_R?wqwijBsCpe9$t}wfe$W~z5ErC>z^O<M6W&?n|V#sQd$8VzZN%}^YEjfAEcAf
zrLn{f$y|*ln)6u(+E5+<z`0yK9%Lytbf?~>Opz%9*rHFLx*y_xfDiUp@0TTZ;SY4Z
zwe4iA^DOuXwauf_neiP)bZnb_5ES{VlAvpeH#8Moh;jDxUegRphoz&>f8Zl44Up;u
z7wsQk{BI%n<--l7tg(#nXbuuVn%PXzpc|ZE=k8LCtpD<ggH86Dms4aiZZZioMC58v
z@V1o!1Ia>HwaM$Q_Dfx7>!jBe@Wvj_D$GfV*F?|ykMpjI0Pm7Jgrff>@nVCjYmk5z
zhGEN`t@*^51xB#J@;<S?FgT>KLX9V{X?d-4d~y-1OO=wDzE(HSqeG@vw`9eUM_hc$
z3up!;<2iDZ8jrz=a&q!y`uTZrvL=a6r^N&NezLWv@5f8663&pk8FsC5Z#w&qX^S9~
z^Bt@t7(X^nHx|MxogV}%63_#a>f2o6^Wte#?J>?h-i5}C>SQ@zZajhIo8mK-8RqR(
zQt`?nlKXs2d|ty_!f_BmVJ>eGpV2Gq|1HVMlhVq>5{C?pS7doG5R{-U3=!oW3!fNm
zn(_UTZ_G~DS)vmBC~(g^ajT!YCdDc;D^%m6o<7bz{-nFhRKXODd*1m%vD)gy_4kWe
zV{U0fZ~qmX5@QC_eh$FV30&{mTlf(qFVn13u~4a@qD7X5Henqz(_|`=HBFAyP>xdS
zv`tFxIK9bkn+H@&)h2Pf_5N!(<@)zT#wZsDb-am*YjKB*V3ArKWuoXj%`;zwGe~ni
zwwF{A5Ba><pbSJX<dwwJ>X#i_2V&&^83%7l0;G0{a=lwiNK0mlhox}74_}PhA>lX#
zGQpnIBdAyuG73bmE80gCoo{BZoNxCq06SO<#spYi(rfeH6ZNB>ZrCbhQWu+sYf9zH
zPP{>yH(ItBNV0EU^XY8eBUQgr9Vc`olwR3E**1}{=r4ogBBj{6rbtta9n}1x8#zzY
zVVDCrIn@+ym#x;Ya&pSJxpChoB3hQ7b!^Tu)bIBmu;gJ-HNQx2-SmP7Xjdyx6$&c_
zwvD1sy)la2j+#)b?uXdE^OgU0yJs_FaQQDq?P8f0`+Ao3#}((>!ui!~^^X_j(2k)w
zOr#G|rLl=TTV8*=d6lt;cj@ChK6)&RezOo`iyzfcwTPy6f_3ZHt}bQWwl>V=mU2%$
zU%|0fq>mEkxV5w%o&dNG<8d`L9Tw~a9Oh?Jt(z6AZvp_q7AW&dQx~QwuU}!?7pvtJ
z=#Z1>)KHH<HiyFWEEhY75sJLC?0ePOnK|?X1}G@-=WPRvx;kYmXf+h#9IxMzP^C#!
zDCM$)YZ9Q#6V8C=zn$Xo&@BJS2*~Aa+<P0!tbtXGbKXw(lTL8jn0a2ChE>zL+3PF|
zBS-CQWsi&}@IB=jYZ+nE?6xaHLbCcc0bbddZre{U?I3coCToTISsWP^m`#bK>gP7)
zGJm|!3R$<F%j<uvpo<I&f2_85nTd3X?IxEl7kN%@|K5JXV9@$i!i~I+6{o1&rAz;X
zPhCTMwyCh8N%=`y`R>d;dNgympN8_Sed=QB?Ll~ijrKB{FB)`gY;RtBL4t-n`Ry!;
zAejzpYDI{6{z`~uyr@2Y+fv!s5=FeM(LE;IJ0cHDUX7C`5-h9HM!mG@5C4T-Mxacy
zYW9@URQZr*uz_(F7nIsBA$Ku-E%%jB+#WW6zwS1=V)(&r7glJKbJ;0=XV;Vi2IDbP
zn_Cl0Ep6@=DKG03Y0XF5KQQ+mKIw`<$4Aehq5e*$2WfIbL$%dT4IS?IP8G_-qPgya
zGjIbNVtI~X%sen8RhdzWh$thf1i^{-%JKK?80rpim_&(yG1o+Nv}vT>Wlywjvs1D{
zeCJG-5{huGgt(yC|7skO(S73{a7i2x?&C{ZYf~V8A=`;wx6NCZmx^a|N-e0F3Wu7&
zs8E74y~n(u2v1LwsyM?^rjf&%mX6=_Ggn`1$U;p7_qS0==BTBnE-+*-W>H8R?S!rJ
z9}}Vm-L%loK;8N=vX`OE2uYP@X_K}={CI)iI~(>^lU3_CQ8`clgwQVT1+6`br+SDY
z1@rGjxH7npv`l15K`R!<0h&9_a^7vOrwmWRt3cj9<wviBX`kRx=U!)_!F?^+rw)QS
z+?ye*rg+j-jFf~d6C)Z;GzapeMG@z(;cK5Ze9i?rK05J`V~~=s4f)Dgoe>CqPsVh*
zzE#>Ry$gpq8JS?!K@H6FWYkS_2Yh30tXh7+a5i*p_`l=+AHR91XlO;-r>a-a<IsSr
z5a{t)ng#rI;5DRJ)SBI+5=3utsx%LQ)Z>k_f$ZkGg<^E>CAg{IGDxBmK-<swb#w~%
zJ+Ihrv}$74IH8ir8G<ZI*ZS0z-xKRqYNhPM8yGGkuR=QK;3)yOHdv&O(+{HB!g<)w
zC7&d~xlVjqLloVrJo=jY!rQlTty~ZH^*Yt1<_@kYtmmIC+^xc;3AA^}lwzutD=dW8
z{b%W{r~qlv)?WkS`wFM}`ra?t#{o5sdVJuaX#4vHOj%pe%sIzSmuJUHA{Wax>Oy@P
z9?Ej5Ot_Uq{kTKfHu*luauhy^dM1H+o3CA)2W5IDFIfnl9ZPpg0A%P=ki~dcf!8}5
z^5Y-p&dydjXO9$lFWd{eFF>&X-5zYnmE6{H-|6vt$c8F6)Hn*-ez)z_)+&`0XUE1-
z;`V#mn&2TUd`PBAGq|yg|E|E0fj7PN$A<O6FeXDNb;-*O9>n3Wa?I@;)a03F;(`7<
zSh|LA7ztoVZ15HV15~u96#<5%f4t8Jz#;=+<x*DIhIcm8rd|tPLykplk0idN00Ht?
z(svm(u7vUh?#HAum2NCHfz#{XYNjgS(=JQQU4i-88d}V_yX@RnBO}fhv-h{XmJZXq
z?QWa?S>+XsU>~~JUS=5bA(O7BM(q|}DHf54r-kx8jwsF^o>0$@t$<4L_@Adhy^N}q
zUStYchwv%^Lz~d9OrL|EfPC|Q67r<$&TN;qAtss9xV#zZIRtjciq{AUe$UsxUslBn
zJ6cDXTz7|yONa5pr=qYv2VZ5$fx}qr;9B^;>EgP{rWZBGUNMJWzB}i+G7saXm3<M-
z_K3V=&((>&b-HRMHnuir_ZFa6m~}DG0slKgM@<^aAYiuXE8Zj?f3Cmb#XQRQJEB=%
z&%y9PZHTO#Sorkmi|f%msw|Yc)d>*f?uO0t)o9by+lqpIe1^N9HWh>l)~^-DIFwfZ
zi6YK={Y_UjbJ>*HsIT_;oW_r+bWB50TSq%NvVXTJ8#CE5Ubk^*asz4o?EQo+>pSke
zk`eUTQaE5V4P$|-ZOfmUx*C_`VHKZsY6Te_;VG|Q5ve+46D;0~Hausuww%cP)#+YN
zN*&2rsNkx$88Xg(ullTuft@5VOR!xHsWe=gEU=)V{Ss@=PyR<^e6zQLs!k1Cfw`Gg
znmkiykexBYyF(z8Qt;2aP;DxBZtlw3V(d4cTwlvHPoK}x;8<PU(><aa{4eRMO<Aht
z|CW4B$v=r(9N$u)$M84JS`D9R(8~?9apiv*7S{&Rh@$AV)aSD5G-4fyWU(%Hg5jzA
zuXTiQN=o>KiAbBR;mKR8TK-_%3Xr(pR?e@=pDLsfF-%oySL-7TlObshxODXIT;ch)
zeocuJt}Kne6MV45J~x#93wUUnvYamy?E#h2DbRt%nvmk$W3PJ?W)wA5@M^UC8b^g{
zLVtTV;go7f%b+z89rE=*F?W1e8o8z}IYXKxK*0^2ZX(wEWN1*zf`B4tOyUBSpTi+&
zx9|o0p$&#ap`4GH+&m7t7VNH66buNDTdtMVlY%>B$<T}Y!RjcT?><Fi(Ny8onRaeU
zlAs$*963g`iPy{ugKeJ%PI@u|&rUFak9UQcr<9J@SLh~;v!rQ?V_rNv9h?4X!pAw%
z;BaJ0zewHhD{K045O@1Dp_I``k%B@(H|*QWMHapsmD1Fj%}+Mo;G}EfNtp0=OfxLO
ze=^Fncr(p#&@(CR@(XDsrWo=Kcjm}diJ+Y2{UBA|RH~_Pe-Fw3PxX6iVQFK_<aOg5
zsoC*Wk9aN5*DI~HgFgNS1LDf|W0E^h*&C`#Tx0QPe5JDs@(k+gBzez`-w1#-uvHt0
zqd^$9NYE!K2f!h`4^Pf(n_c5=jdEq+FeuHGLbXl}HjqM<oz;~vJtJeK$DMHthzaCi
z=-?W-EQMIk7m4;fmw<B_mUn>hFR%@g&^Ksahu-Kl^}yu%`M~Ahy)ryV{>~2)h@MI3
zz$iL#lmqGTKEc!H#}7}A>nkOiwO7jz$dU>yr(_izHT4}@EtpDNWENPAT|wfxA!e{k
z617pO6Im*~!d*~ElPWXkGR}scxB>==v>d;VK(w?8cu)Q@MZ?SH!T({ksbtSLofYoQ
z#!Wd1Vp1R>mLdp|cOWAi+TVz*jbl&eFrunzZkU5(Bmlc3QX}oAh$+;{G&l6Vo*}in
zE2Br?NhTJi6{ioPgp6L;^2it}<qq1?&uFUShJgA@N3wts-0|kgw9?@VbJ#c)m}KZS
z8bT}`DCwcK!fR`5=DVW(O3AP@%IHx`P-*eB#RYGd#8$SZCVzZ_#go~w4cYwZ8Mf`D
z&hk2*JTC9ihmC+IvO5zowOreL?=ul$&kl@7!KJ#|>p`uIqz0yT9xLCUe25s&1Rq#T
zqpzr!2L>o4?muOTlx(ij>XJ_A=2%1dWW>yP4u5vE5BU>MQhyF(1)LrM8c@bmwh=jy
zvt-?{k*JlF_zgRJmCwH>sXWcIke!oH$cCmcswU=8ylmst(raLNo~>E#?RFKtXd&gA
z>E=PQxd<(W31KX4nB07r=Iw9H&8>_6@N<^&6Fr79^o{~|!AK}UX)h6jc5Eq|cldX@
zD6Lmtu79P^?z{_I5uevzaKWPHYZ~(ASL_34zO91Z0a?JLT%Tt;XdZ?457WCvtcQUL
zEM70>fn14W|B;X`$OokwnU$~CX2a)~;?wA1$QC@`Ua&L`SzN!c)S`I^F;KdZeEpJW
z)b@^}pEzF89V=W`#%ixTF%nqJJ2DMtS+lZC>EckW#C?jd*S{N7$)o0KMyQ1`GAvp;
z?^V_2e}Ls+!$H}du_zde8ps)c3a-&nSNd~BgAZr&eXyS*<CGZ_1vT2#$X+C}hF})x
zmb#Tz?3}-rjYr9|JfiPQQi@KxPn!`SvG@K-&ljOoVNCq-CQuqMgn>+6KKFi-P?}JM
zjP>VlE-;P~`TP(FhuT?{_vi**P&^82eZgjb@@LKAqttPx#2C@f=Q9w`2VPhGu4}W@
zh1w724b2S-0QB^9qC=0C<&{&@xPyiW3g}D(<$xORZA<GY1inc%%YiM--k`K{!C@~}
zUn;vi*I0PMi;qUpd@@&b{PmT?*H;+h(PD>gh>#X`#B^apDKSbGX$m3u&J*lJB4(Qz
zfG3OQGt0N%o*+8Gt0vx84NFNd%1A)Ofr;PWAC@4QAG{29vdX{E6bPLC_OL2AQh@;{
zbSZcc-;bW7ltP5cp987lRRm40A-&)qq0xXe-55pIA#40-$%ySmhD(?RDlrsYu2~>~
zXa+YvPMVHBzDz<UIxdvDwZ)fUPE0`u8BlS}N2F#NV-{r*P8=<Ke7qT9;Q&HBLV<{F
zWl|DDc96H0@w>1gGTs|+3)@}+MY4<IRg)h8f(pXsd)?@3-cBih2f<GdtMj@CQ90bK
z#JUQ`-Qg3~+7-Po$sibLg)UdjPx`H2FcI%FlIylqMw%<CMyjS0=MLIXn`>{Iuw(^)
z@gm$r^PJ^*7jg;)F0<^wXAD6GY_SkU!`!J)j%5T`j$eU|^Sk*e*U}3DD@@9AV~pfJ
ztH+04@lA^yL(%cB2b19A$73c^loUlKZ2~ce8v2ch4yKPpj7%y;?q07PUgbi^)5cVJ
zgW-Z1TOyOHNd`eb0Q>ebVlqIb(tZ*y>rRki(5#%1;qe8C0_itz>@)Dcgz&PG;e_*-
z1tm%sbx8l}_QP?fKR9N|xJ!}VPE(;8!$<4u<{HBkTbee7Y6Ps;A?t3*QJ2zLf5N+`
z@329j+*!HY-v|2nC@!`AR%Aox;dWIok)?damWv#d)Dt@rik8M?maHP>?U|eu=P`Q4
zEyGQ1x5zBw_CyPFlOat&CJ67#jD`7XYip}lm1=*-^jQ}>@4RM65Pev=+QGEYvc_%0
zL9^gBs<L<4myi1%?;Y(;8B19KW!9_vTE7&Ya?L%g*6@nLcmK&y4q%}Fdt!q8gE_^Y
z9>prIu+@iwmYW*>C$ps+C(1&o0Wi0iL^$cByM1!~I%C=UfJn%`Fqv#5c4Jt~d%sc-
zRHi9RY1HM<w4#6RY>Xpk^`wJ>hL41e26THS0-23)Jp^@jsQr<znqZeG>%XaXqDm{(
z4Gc2}QXK;dp)BCkWE=ca%}fPZPi@m{-11C3{v#3u=xb2l<d^8tcQ#KbHYDx<kdWJ&
zG$ZDfMuO*te+DG~)vd}D4Jl{%UR_j{m%V&@`deZgu}mcigxeS~Jnha*f5BG7>g7=Y
zS|rtKo;&4y?)nv5x|Fm;7Dcklm~+B|&|rMFL(}4g%{h~KK4f##WbeNkVrvvqF!(2T
zrd~={nQOQ(ndQ4C{V^fXDXPqR7qco=8bp(4@rSgxDX8>Y+E9`Fj{3vFfOh1{Atz@%
zT-&EYVPR|nnn=3bZ5WOj3n2~d9xzmC)|D(2Z1hN0^M9UW<IF3>(<4!XtaZL(eKO0k
zj%SLml6F68_VB&{<?jtDgkSQNh?8k347AT5-;gBmvTQWAiA<g5=;<4`CumhrV&pEa
zJh(Yh);1Ch@u3Kpf+DwcNdHEvi9E4U20mJJUd)a^UvECE7~eCiRkq!p`n&PLDg$yQ
zkIlA$^=sl4?*jpruZ44D4^$-`4Qlysr9c6z>f2DT$QeZ<pXUDqEkV-0YVCQbR=Ixd
zO{5f5YY$;Y1F}vB9oOEE<8e-J*u~@{j0rflz_>SLuzQ<)(m%xUxZ@Gh9>Z~;x8He(
z$DaCB`P@Cm6_{eSPnu~lA-4qrC~`wy7(AX>e-Mr{Hg4`R89~%6aqN%d*!<;i9OOYC
z363v+c9~ik=(>E0Y+6<xI?tJ_Z*$_|3z)ozne>U9DRXB-Fh3Ysa$}G}QmZCN843eh
zDx4I-Fxon-_e7B{UvQ6PR?Q3U6j9nDSy*DypD^6tMT~}Ay6-`pEf@}lM2#lt!XhRc
zlIIhmw2HTusG2fgkI`wJD5{Wm2Wa7`wVO0o&l6YMD4pWHqrPwo?;?tPgqie68f_kV
z=wZI|jX%a=a8{P{;LG+f1!+_zDV3>FTuzb#vdkiE@FA4~A_4VU%ISMfP;GdMZiW(R
z5IM{ahjO~jVV%i42K_$9gee!&SfmV-SX5cXc83#zLMQLXSwGXD&KQFr3%zL>Ncsnz
zeZKXr?{ej>9^UK_UR%Y^c87!gKFwyG+0p0AKqRzL<dLPp@$4NA+p{y0v{s6CyU9Vm
zOZQ-pdbLKcH=sWnv$)jc<e3Y!k{Y*fU!~XUP^r`qQjr%0)9Hj#uEpN=4wLOW2&LHX
z4%yu9flBaJP_4%}XQ-!@FdAhtq1PL5_wF60*)&wpa1?nS=KD!i@=T8GMe-~U7oL*k
z%M+Cr6(P%b{x4qOfB9$sl7tQ)e<;G_JH$!g#5p_r&ViWfaCqy;vkXd4zc61h?1d1a
zEuxP3dC8B!rQCUgvqh*lXpM+rGT34L<@e}b?}6DX4{*t7zfZqAX08=350Az9U^45<
zJq`w(ge<l+DxC>ugpfGG_DCtIHfoq`pV7fCX|2Yj*Cm^dsn4x(&xs{YR!;NI`dxPW
z1LA6fSirO}OtTCxJcWY6_BOrk+aMK_VV7ZV2V{(Q8lARrc1ludU`>I^hvefOwl;1u
zm`tda2hVUa#aW1TaGcjtk{6z=FpS1i@;patg)d3hVHThR=lRArU*T*2+y4)xZ*uDX
zk7Lbklu8hiFaTSG;bfk&YHh+O9=t<I@JeCJs#IQzZluzP($x8{($XYJ5`O*f|2ITR
zGaPgnPx{!rz)T8MT18aq<OH{P=fjw^rjANW2~FM`G8y-Y=bPB9pjIzi^`$4B_X3G8
zJt)h<2D6syQ9viksy8y6EtP7Mw30F!jwuXenPGWph0p!m&r?Zj?C<xeR}Gi%e}a{j
zHMTbHu)cANgU$icONN~>I@Lrv7_LT1%mWWU!W(bB!TS0gA}#3;hon(VvsT6ChHiJ8
z-JLtEox6;Z3Tu4%zL&xfJl2NMN?OZs(Yo+?y%%^FlCDoa_Z*Ku@(AC0{(1hzZ~rzg
zz4p@)RD|U77>*OqWU#}Xm#@&h)&;X$b_f)s{T}_!gm&|X;aGeRj&)g$2;GF^IIhfa
zyuU+QtupQR$R;D|b8B2YvCN6eY2I01XRkjbI)>vc%ka{Ja`d-1DYkM@ipj9Yu)Bkh
zG2W>o9H$MeEpa^AVSDoy{SwD{iDPde(z?X4A}?~L-;3k&`^|7Hyyu&L@d|(bzy5#G
z(M?X?{|Rhy3#F5A*8T__hcQ;_u)6#{9E*<*$3kkPiUEl?1z4!gt#J0)2bsS-kI*Lw
zeIQGHsZD#K#zA)&#75<uZ9S{Z`p|~<p7PywSVPH&R(S8q`R$?7HSP_G(q-yPt4xPO
zDsyepYMsIUI`7=w<nH)751&29%3O<V*d_OZN~B3EY3MLnN$cbq?UN~3L$%qYF~0<M
z3esYxlOSpsjDqz_S|h2=bMo{#TGa}p=@_XL^-2|~eMw;nbI64Z<KgPn3Tc`I3m)f7
zv4jf)0i@t_pM91u{Nw))<CeMePJxO~V~b!^EoBren0!dWH3AtabQFV_8C@HcS3$@X
z1z*Op^moV^{&CjNG^lf~z!VB3D4~%=3`Tukd+iz#4Kx)X@CU}@jP30%=gyr5WZ9#h
zMZNDAO3cdNA~?Q=%Z-`0okvQ|%E~H(gWFhFu(!X%-Md@lSw<sGX*K3pJ#mWFwew_o
z2G&!pG!c=eKj`2~<;wQf2HV@W+1lBld+;8kfn}VBR;x1^Iu7$FqePG*!NJDHEvDI+
zxRT-vk2S8Ws)z4{4;39}f)A1P4r?3^K`hnb;Wy4F9CW)p^7x}%efLdXeCrOUT9&2x
z@I3R}p>-6@Eu9EKg7svR2?CFn0xQ&EB|}+<lM>;=C1|WUR91bM<z0buAqy2ELVC?~
z*k|X>8z@mj$^;?8>{K(&>Ftg2r-L1cD=#v4Y>#KeWaQBv2mWMqsI>Feqok&_a)M!J
z3ug-kd)xH3?%;|Z)yjEJG^;#*c7;dpImuv><D8&fYk`j0?eAlaBg=DcZCvKY)+Rf<
zJKP)%Fw+6XjFBRRqKAkLzF#193@GrH&dxQa*_3Lnip?EGZV@6}X4YARcSr}WFc@RV
zouw!oRtRcJlSUc`925n2_PeY;{v>->3_G`9p<35eYE6(1YX?Z#!a0HTo*WMbiY!A(
zh_wvQ+y!2tE)-6L_svOxb>_z=-+mP9*=Ih+<BvR!F*$=_hv{S#+LVzZA|<X@G2Rjh
z#lRLsQj($&-s6PFOh=eepTq?PyV7A)n2;*)?|f-q3rK$yz#XsFA3q}`0;N;d&R^!M
zfBt3M)FITA`GqE*`sAlrJGsVOdx_TEGR_(Ly*?+`F0r<Dne%&_y#D6%Y~H;~cdv)g
zit%VbV_}Y7zmJX+s?{312YYloU3NA*RMLd;cuKX=WIUd**Vzw=r&D-o@vc<5s}RIR
zkv`<2ITw`g!Av21kD2l}iZstY^&}5Ja6jL9`DK3l_y2%b-gx694eq`l#}U)v4m)pL
z|3Dn~_eXI0#1Fx-`d}REV>osJ$L*C9jNXUibzIS>nns*xReAF43XfkrMQ@zpt)N|R
z;dRVz?+C{?H}2);))qT^+uR!TFw-8!OahKg4;dMz?g+=;6&!3I;n?N=5XZ5QVVcPg
z#j!fX@qS)#YxjVaM;>GUYR1m3m#Nhw(pnRw{gF6Mv_!h76kPrEa2&*JI@~|L2$ibF
zA}1cb7wd;OCLt&>mZ*}@Uali<tOG$ADuqYNP?_@H1;w)sTu50`j6C2(Uh>8jLWR%Y
z*;uE(a0;a&nky%nJGF+4stm_Dce_KreRYEu-oDLTEkP>J{;-2AGFIBls7R3}G54H4
z#l=%=te#lqLhC`IN`lr^oHGbr!Q~Rsit)x_uplK1CofW|R2dJ)$V3t+q2=mv!N6Qf
ztQ4qN6GsuT)I^a&IT#v)&4VS+LKJb&iA6TP`zp)#KE~>W%M6E#$sh+OgJHchC>2bj
zV<8}dC`~b5Fg1?E6*v|6^MzGN7p6{m@6jq^Hplu$t)FR7=bgupqYy|Dyz0CIV+)ih
zKqZs~d6bk)#*<P|5Y#lzg<p3#J>r;h^mtb8ohVz87Vvmy8I7i)T_K>?>(cFXFvc-I
zzld|Nv$KZ~nrgK{k|b2yOH^wu8uP26P-hEHu3cn08ZsF5c<|91Jpba0eCNei=?=#1
z?oUW-F$)U<pA~2=nV)aqv@XY?n6SyBbT*nfl)IyrYow!)wYo6V);WhIkWamb#~*)+
z&wS=Hy!6rw{Ih@hKeOMzO{)$nl2{u!h0|%q>>v#)w2<K|rGmQlP<D4@fb0)PbA1FC
zz7G{8L9k2_X@tolT#+GFf>1GtFyvb(K{m~>&XE<CqA)loP(tHodt7EM>X^bgR4yd`
zcv;_hvT=qp!7gOn?J^w%@1kaFfx-&j+V10tSg6%$)+)3YSE*H7oLE^4ZEKtJ_=O(5
zZkNq|pO<bu!VBMhn_F*vo#|-Iq&GmNZIWh!H&cX~lGImGQ3|fW+Z>D`5<c*1X453B
zr7(d*lqNAk$#994msuu)ZGsn)`yP9QU-<kd_~!F3@s<D2|4cS~ojB2flj9856_|W_
z$W@tf%;G2_(z+ah=MP(tvp~R?zDGZ9_V-~|k|g~7zx($|<CxK~&v-OoI2eHUByk)r
z_f{pe9GoY2hNe!4JRr*U`gqE;(<3n+S*f7n3aIE9$H<gb0&y739VugH!U)P%W?7Y%
zl8cw_XLV_X?)DuT)tHALewYXDdl0PWo8R~vqug`PrHfp;aF)&O4K7}KlrwXSB=rWH
zcdqizTd(l$mDiygv*gZVMZ|u0K%yM|&WNq80}c*ytj)RiVw3hl!hZLFEFU9vgfqSr
zC>^Qy1A!+La0K9lz)N4!B1EX(m1lwyq0L!IQ=Wb5NglobK3@6mcln)v`|o-6%{M5D
zAFplmL)VYMF~@N%1=C5uab8&R62~8OR{vlei&E)a;@B2Y2qwKQ)4?9jSsLvl96PUB
ztT$*ktIRK+pjv5h^2F&w96xz6;CQRw=fzu(@%(FV9pZS}AD}92lEx8^)7nW?lz_|e
zHpd_0So}a7qhx8Y{QWq-|MAE8#m|3|Z@v(4JRQDHlt`qilsGon55%!1iZr(P>El=k
zkSZ<P<OSXq2<=f(g0oX7!aE0HR9dB)BBdlrg1N9#U|qoA?0>WBtDX7BiqPivULt+)
zmy}9|2_HHJ@2SnrQ)w<ERYbqDM=U*$oNOaS!EV1tr#Iwacb(Cw%lh3Lfr}|)zWvTT
z?YTKlwdZ*5o^yQa=|{NdvFE7NYgF1b%xKKCZ}3rqw3YB4<r&BXbd1muxtBzgo^`^5
zQpBZTaFPX^mzjb$(wc<8jTSgwd;WR;-GBHKKKJR5^GpBZ|D8(xGT8*Y4U;mE5jw7x
z?|Ru~be=rVDT)FeMTcFo*>eMsrBlX_TZ{eKjtL{Yl*p*$jYMFBmz={OMF{+o)bq*D
zJjA7Y&xCd)gz+RomE=N)kpA$1IDU|afF=MyYjqePjYm_qw+>J$B8n0sU8PowSX^Aj
zdqI0{m3pJWXgp-l+hfr0Fu!<`MyrYUo@_Fr-#uV&e-AGsl1hU!Yo{3xyY#1H{_N}T
zv9U8DN+N0rj48;KWa-ozwMq?V@&Hupu*fjBOMslI-joXM2VaiIap3?jY%uvWg{9q`
zLq&qLT7{%ScUzE66p=T<z}R`l<4FL#m(0eDC?$gespRenj-=+thn;^kBcy`gjt~(D
zg@_~kScSvLTc`lmqY}^QCoi*nZjrnwP#%nNNTE<VDmf&f6%|?$@^Fs>Pf`P(IF3sH
zLdeGxhMmnYaX{9QYK|zW(P%H>oZ#%j3a!>Wo#B9;?hc!~J6u>j#meFW;Ft`1+}+;e
z%5E2-G_^*9ldC5g4IXDQ8L|G#*BM^DgS>wp9ZL)zCk^$v1?rV5CeO&H8QwSwAF_KX
zK$Q&VNQS8$O6x$vF8`Wj8KW#`QuUl^H&Ic7N>fPY$@U{^%_^QfoXx=xDYC&~0PaG+
zBPg`ALYP_k-V;ZGe^wL%o14Rm)nCQ>*wY{5*{7ey8pC8VWHK61WE1MO1|TS;qf)P-
z5DZ<oyi%dUXU7veM_~=4!HB9gG~yH$g^_Ji2!yktlH@J38L$ES@_k7jWiwYu6+Eew
zVr6-m#~yirE3XeYb?z)rKK3Nb3riHna?gDa@bas#^Si(E+x)_3Kh47r+|S<52J_1&
zm|t9Detw08`4#f4;JxdwbNlK$IF*u_oH%lHI{RePf+P_{iDZ7E#_mp^{ew^?k}6;U
zD>$BfQ3js@KQoSesr;U`DqToYh2+SZNPrUX+|y6<(0%vuomXDww|@Wkc<b7&54nRs
z9LHjY<JhtG<h?AN4LEilibEXhlDF&7N*=Zpe-MtXvv>#jXu_zo8SGS44Jq41agAo{
z2*-2n1v<k4+npWm?(A^S$x|#Z%ma?;sL$Q)J>J{xfYem$bxxi*$$0ogiQ}&^yfVXa
z5Is0$sn0J^uhasLvl))di9iDM55Tec9vqLWp0#HE18`hBhT{Q6{y{ia@53=7aMKQP
zoPCry76=jHeUA4wd@dY9L<ptv8K4AqD!KKoH`%(nLmI~v5+bdzK5&Oa^;uGwP^}A-
z5K47A4txmmr+HD%W{0F+qrPwg=Q5lb<6NI2o6<elMk>$sopla6yDT(XtS+2nZS^Eq
z?%ZN)XBUJdt+zR|y2Aadi{K2SL67yj>(rWWQmxiG_skPCPPcJ4rWCnHCl#di^!GLx
zjECsBLXsv3Es3Q&wBRvih*VgMEi5JvbSPa~YE^YYJ&mc21b4c9OfkhyCXfwDmK?g5
z<BAGWNf0qnT0Qd16s4Nlgz;cLiVkUF0XvpB3TpG2cHqZZKhvO23WKp4bl|u|vWZ{|
zOKHmM1=UpY3%~FPpZV-lL}^5x7sRu+A+rNJd<`kffyYkb%c_(=lyg!(NQqDhmFgS^
zohi<Ysn=>$t5s^X8cU0d2;jx%zsb$JTim&OhX?Mxhf8PApe0o633*;{cl|mWch(t<
zGA725wi+z0onbf~aP!tJ8jTv=et|U_-4x`7VQQv4v33v6IErjqjxsBOGz+4XKdLek
zGB)o8PIwAu@YZ9DCDsuuE33!|)=r(IJ=bLGx+Hg=M5|E7H;(CKOi^U$NQQ$#c$5w^
zaJ38yGe_WZSqlZOG_4Io&_C=Fq5zvf6k07RY9O4Za0Z1%*uZhT@Z4oCf9w(}mQ3<I
zwB1BdBZWD(!K*27>O#xP`qKClai|OnmxDs9gnGS3KHkDRgHBEp#Vz8rPJ4bHyyuU;
z`7$@Q4%phh&D!~UxOnOWsc_7<D-?Og%U7=P`key?lLF}qYOM-Ot4kbAC+yt1L)=&(
z>+j(UjjRSIcI*0_IC+-IBxhPo$nv7BZp6&yA4b{-&W#FzcI3edff164E3gKpxuqrq
z4=v9VCyLFptE9~~qw5u{vqW+p;d?j(`7~ov6lkAd%j6)61zKAYUHac4(IQj?rMFlU
zT;R3*=^p_}lJF1z-mjBH36t@V>14=c)DHo=4jdg@6i|&Rlw?{IBvOU<6A*-VHFq96
z7~;kg(&Y+T$Dl$R)#DK*7jGt`oGHbRi+@C^7?Mf{A4Z``stulb_TwxxW7bZcVr_Yi
zc4Ho5MYXj^t=8boi6!cllwRk6xy2={nS{cWhlQmze(tlsOlxkAKl!si;_mi3Q|oB7
zsvLBNG@J9BSYD<-+@U|%V>C4k1`}Gf777tAOzZI>MmwCBIZLvRSGHyq^}~u@sdN`=
zMj-S!XA#a3MR00C@r9=vymjrv?V;euz_FB^fA%u>J$nfis~>>lAP`eh9pc#eV`ud-
z97~}PDy7z_F&%A!bLeD^C~gx~>a^zPz<a*@tyj3VxzG0Y9ZsFQz@@blq|!3iN-6T3
zSC8RXVW~A!PMlt$lTX>cb%(gINY>xQ7Yb30aW===jFr_-9O5|G$CPx2636~Hj=e{V
zP*MF}95<!l@zq6CECY@kZAQ~d@XnKS2(NG^bSXXv$Ew7!C~+JKi804;>^~YDd*a|D
zWF2^iH9b5AB_a$8FCC_ET>aV`y#CcUv6-ckq~ODSJ)834iqJ{2*5QRkDqAKYfsr>8
z4hSf#nvOxgM_O+pq@u_UD6&5JctB@w4}{>CKl>c_o?hnU$+O(P@*Z#9+F;@o`v+s9
zG~wy{@8^jNOSGyn#w&*1F?VlnFrQRdz2`hxzfaPPkx_vkS(J*=s=?;^bt=`EN>n9{
zL$%rY99;(H!bnVMHzkFj@DAlAk&{$)gw{}r6@BY@VSNXaJJzn<;pe+O>K80h3XIJW
zLLs#zsnn>(F~xX_F1f&Y?lDF&EdtHN;wfA~WHnAWl(h&^V161w-M{!3|AN2$xBoUt
zlKki&{xR0ve3jj;36O*gR3<1H1-E+~QDpeUr|;u4pM8Qxvsz|-Ili2MH*2GOP^%@h
zCT67QnffyXF-TZFeTljGMTY$j{n3QXI#yPfSSSa(M@ht`d+(*w+2{1B)1_!3rzmp9
zgFZKIUS~X<BBDC~$Ctm%{SQA#ue-;o<ra-vokl$(j=+1%!oo7W{sB6w@zhh#V2UD)
z%_$X3Qn<r|`bedF^nD<3X}k~qU6aX#|M(|=Mt?BonI|43QW5QT8+62Qnp3OG;MQ&-
z%QD8}G0kdv80jlj&LH*f@XrdythOX(Yn->i1o1<zq!E+#43GxylTJw*5q^>*rVcr0
zS$p;#E`RJE8nqPbJ-N;Cgn6Y}>cb089lOke5<FBOXSNk#VvRs~fmg72@*GM1KJv+3
zifjkxJdMRmtjsk?BS{fyF5I(5x3|m1)n!JxVUioNe9GQ_mltmBG8*O7qLjb*+E-Y*
z^dQ52kJi#WX}v*IjZrm$S7Xw~{bZ9(lx}nXBTq4&W{k55)`sV)We}nS4oO*!EqO&k
z2qF{)4^H5u;9xlApZ(Fd84iYg@~Qh#Dk5#PA&M!o0XnUQ>Z`Zp`Gn4R%3`yAxW{p%
zN#ZC>ksusM&X-R8))Gfid71w7ed<2;=o38o;3H(&h`g9E9(KvHDYaTH@Pe$tIjA+7
zA*=VEs#1rClU5FaLE;#W2AtBGD2c=U?*%fH-mp@mP$d^M1av4BNZf>2&jCuu=qNf=
zqbZ>|d*&STRgLhTYNZ;e7ikmWGfuCbpr#em$$+GikW?yIQ_wrup_pdj!WWu{AN>Tq
z(Ufby_wU&s<alAxIwH?9cK0@@R2Ab%MwWX{EH85D-pe8REZb==$Qw(c*DMfq@0*U}
zd@yB>M0BX~pH&qE0_zQ!h{BGzdG*`8^TxktJlg!wdzdtT3>-gmf%~4lNTXiCcu(<T
z;@CNd{~<VDI(e3~ejoW{gCgHO!ts2QB$7;}<ox+l931R$adnAdW*BD$c|K+Dpv$*!
z?lKx?RHB4${MnaTzVsl&L7&#r0+m{Ws1l=U60fFI8t2F+TPWS;fk&VD0353i#<40p
zL@108XSI{i8BY0s{ikm;9E|wXGxrmzh_u-*aom4Dj<X5f@sx#QIF2Gk8b?sh3b#K+
z98bDA?YRb?aj1No3D)EYU6T2Y=GHgf;`Oh-MQ<=h3y5`!5F(5)vj*?{q0Pr(ASWpI
zUIu4}vaNn7VzjV*=Ni+&I&pP{IBBA!q__PZn|ocRxufD7SKqtE;^HdrtZ&mwBWg)Z
zl2lk)T;S7}R%tX6M)`!#f9Z>Cz4Hc*)*PL!T_%Gb`8dap0jHVtH%O|>*rLyCFMOMN
zC8d_s&`MF59OKO4J*>)vFib@W-zmpZ2mysdT3BdQ_`m(lpQG7o@a1p3N?~#)y#pvR
z;v~ULf|<CGiX=^#uT|LT^@GPCkmZI+R+RH+Jx+(tN|>0V$yAOM{+L?&hp(R~sQZoI
z_zk}J#V@kFy!?~=vLCp{Lx*)a5tY)(JwcKnydaL?si)5J^S}5ktyTl2Wbiq1WuSAc
z9dW!5{r$@yR;EM7C~!(}&*ewC@6k{3)vx>^#k3$+mNZSMR;w&6EwQ|M3NK<FyzgGd
z{T@nqDwPVhD3}cUSZg_d@iLWK1E~Z_l5q2urkW@k^%@HcSGjT9;k_Y>6885p+U*4v
z7gmufc-kCh+oeC8a}Hl}5{||nl?UO>l0ivgUE|}=KEduzhi`rB+dO#RCFbYnXfMvw
zA8xW>JkAAgsG=z7_xsf6=gP_;r4@uK50kPi?g_!NQe}(esABkGmYI|)O=>}1r4y7%
z@jZcB6r6Zujr%`-KdpL=NDB(*DO}0O5asBs*~R7DvCmufq?|({m|1&CDbWb57M#2H
zA?`fzG2VIM-{Fc;XxpR-?Rvt4YjZqwVHGD+K6(E+c6K`?%ClIDFh#~eZ@|>S!{=9-
zuUApx=W$8O&GjvkL{qJ%?6lep_O1aHR1}l#Zc}f>EH5q-X^5;wDIb!La*u<{x<N}p
z5{c5Rv$S?74?Y!Z>Z-<tr|x5YzsH|^{Y5TaK2N39B5AKuOx`AK#ARUXF-6W!f5hC%
z0)+|W_(&@%kp{001x3PvQ;JAO80%5?quJvhwzLrZ%43fb<Cz*qk>v~rL!>Ncww8|b
zQYmz#aY~U2NQ5lu2ZE{!rnSaO3TH_g^|E!R@b8;c0F-JpDJ7`VMxd<F2`}(MpwlXm
z4l_$bKoR3~gLZqKY&;A+rnrKLs>saJXv|YlDV^RXB8rhhvUT?s*RH+CWH2U9sw}Ub
zq;QT#d!9!hd62Ju^F<E!Ml3Bis8wTh9A4LEt4+-~(sZ8X)ze7i``V|@=A{TMj9zli
zm!{;wvrY+#wRvd&sVF3=v!w(oN)VHRYwvuWH@^G3*lY)De!4i8i-MJhPI3PyAE4C;
zIF`<1J|Y~Sz4u{m-*=67U;OvDe25e+qBLQyR^s^lDqfX1-s_S`&r&U>C^GiD1E$XN
z(D@TA)N3g5d0bNA_WBk{s;O3EcG_+F`^Ry-Nv)w-T3q_jIJUCN7oWV$o&6qv{MGMp
z>GFBfW|O3~LNR&w`*6HH7|~u?2o^6utb=fBhU2`%u~&*HiZJe@!*Nvg2$>=n;%tE{
zrZ_(V<*-xF#w)k@?pI#rpf{v2jz|h38F<dYINFz7_hUs0%FeI{Waz%l^moo$3LL6{
zYj3>7h5NtG1E2gAR9wOOh<rL>e=z0Mn|r+S-W{YH^Oe`GaN@)wQ!CgRWJGbw(tL}G
z6ukW2I(vg5mOnyHCtNstis__aG@5eaU`T5@0z5X)NKZ5v^m=UVY?2qASQSB-XboP3
zfKw}lRN<@)@2kU>ow8s<WU;wnXh;0pZ@t0U<u-rg7eCG0Z@e3v7CJk`m2*r6GchFS
zC}Oo$=U{UuSiIoLjbTz0Aj~1nLdXzE=0#4V1XfAhkEw9}iGsQh4w^p1@-iqhogqGS
z$WDQrv#axb?sHGkYSqhwK_R6LRt_@>x0n$`L%S7!WM6Wma(ll54iVhUleESc{`%h~
z&mC`k>rY6vq1|Yb=!ANuNm8p~t-@7msD%}V{e7m{2wxO<R}e`}C61YER(a})N6GVo
zxw$%dHfFE0N4wb|jRW9q9z1J);q$*jv$crzhKQpPT%k(l53}t<wH}nCXFQbTZ>2Pi
z`9&6%mswd{qR2)_;d$V`2g%1%zWFDANNye4d4v-<Zy65vvGx?Ev`RUu28O?U$=M5l
zEPc<U_-JSvUI(6}D@s#Uhfpa(WHe8=x%BL18jS>5ie<F2$WpDWlmub%xKfc6GJ9(b
z%*eu}@v=O=$K}L{C~feGFa0Woaom3Ok5ST7tu<&yG0TlP(q@a?C|2q%F0G!Tx4%Vy
z+`~<BT#*wA!CVq?em>>o)0gN?3szbQ{YlR5a7d*QS}ICa0%6?G^RZ9<SIo~Xlevrv
z@nOatK7UDUpHTxmI1w^iYs36qDJ8LvIK9+nwOZxU@*KT9CwGqJOXn%ZQ|^B4dT6nE
z2|iR{cLyELyEO`L4n-vhkYbkg*)l*Vu+|_%7%}wIrqo?ZVix;7%w&RYwwVlvOs7*S
zm9!KD1U0%f1yn1ck|Ab@4y0v`CHG*wN2UqYY8y#tA0rjBsVqW;>a+BO?<ULN!IvTf
z!aJj59i~(eL>bn5sLio^_dQw_fh`KW$uao^I~ib|VLa{wjzlL+`dx0{x<Q^>>g`3|
zc;^aVfAM8L|Ao&pzqm}T67%pw_wd^5x7ppFGMbE;U#OB+BJw<EI2f_Ga+xTu0cEnO
zL$$?PjVTNeDjFV(b$L+nN1=6TY!Lz_p^;_F*?JeW70z?}#<zLrwSNuS0isASCj0Ok
z`9pBrT5EIfbNBsV9Lv8vj-`-P(gvUWrC*~ko;%<DW0ZtywLvqES!o`_ajnU{C)Yj@
z$Mb2z`T3O9XYQpxHN4YG8BAw5UO-33aO~#!*vJ1C3v(-E_5*P217<%E$1YUc#UYO8
zI9aW7-^x7wJjZy?%DopblPPz<cH{eSyw~q=-kqlK#bKdBz7NNtYFqp?a4ZdIk0@fC
zw}8aC34(;~-2>kG^VjM3Cph7-zCc<XRKN(Vao|<x?g)p#frMD;u-AnF>yK4tXOW%9
zdP_d(@s;2E7ercc?&&XKonk!f(akcpy1PvKT@r2B8gzMQ{U)u&HCpvKL^VN13AuOd
zO*5|Sj=A%-H)vO59y%D)s-~<o8%+8gY;MCW^(etH;_F}eSM2U>kg9|cS^3@t6J^Rm
zie=U|4&IhZh(lVEb0O$@Z|8vZ;gHuic4@^48o}n9?-HdkXFjtQ#9u;y2;Hxh)*@Hr
zO(J<j$#LG}yewk@g_43uM`#`6ya=L-A9Is`;zs;Onb}{yZ<&|AeSk9sGN}crzL2CM
z=d>Kq#Gt)Jh$CfBa3^<Vpf*!dxp3$lenvQ?(jEJ}LWE3R5+wDl_5y$NZ~a|Xni2aO
zS6N*?f$%=OfIcEls!S(CyvuR6KnjP;Gpw;F1ZFx$yC!iIQHfKO6l`p*Q{)EWCCUqg
zmmG9@JoT|>`Gqh35>b>QT!xqKebvQ703VCSLvG*?twq8`7F3kC?`%@5RyccRjoY_w
z;GH1SF%tvR9BRs8UEtA;$35~qWcM@Ydt+?yvSTJe_wNhx{4sLthgpl~8f?GW!P^`W
zH$od(OCp`qP+clmB5#qxLK)~PDN2X(a*qj|2D~m`ddgM}M^%_(?dmXYTM}zcZGMq2
z{p#Q3>(zwL)-C22mJ!ZVWQN4Wq)DCJ_YrQ2H&bF9b~2$T99jsn@sx_G(nw<FD-}`!
z-?_6#R+Oz@5<oH?cDVQPPw@F){A<K%6=@+>rRvKE;<Iz1gszVb-*zGx!xuJK^k|_-
zq^CgeN^inyMf2FjGraKbRlElkM;L1`CPK>$@9TIRok5pD7Az;6bxiYuX<;bL(O%97
z%VJg){)qPY2Q3Xe9VK)&?qKo}F4qhO9l#OC5%>@Q1Wkj(I}6GZi*OF$K|~%Q9Jvej
z4U4O5=vosSm80TR@b{ur(S??Ln3R(BExSA0^!hzak>iD@S#2@D*kWm=O|4!%e0Pc>
zLnw*0hJ*cWdV9M}CmG}!t>qPTtdR(8KILG4hio!sZS5SjW}Ek}-v$-&jc>icy%$cC
zHd4|-$}kJUruIULwFYAa<AZ|UaKil$K8Z&Kg_{6MKQpBy-Ur3Ib*3~N4cx6TC?}}*
zo%MJX_Rd*{^OnLI_SUcP%6EPrm+zwyWX|C}0%wzjvki9MI(Q$Bot8u@r=fdPrTf7+
zmLCC*7Z&-_ul-HFUQ6g~-DG}o3E?~@Gx!q6neQWuDb8fX!ha8rCsb^OMiR47sUG3D
z3`T?~aXjj9>9LRV`7iu+k`l*JQC4)z=YJf>WsUd<$L9NS{Bm#1Y9-={OK17cyH`Vi
z9><}jXF$sgN*o^yx(xE*%3w<zPu_=P_fy01&ebl9YnKo@fvi{Z(Jh5b>6<nqFUf?)
zd4(V^d2_zp=Mb<XN~;|az5@W(9sc92N_%+Ud5^)7n}V&~ZT{dl|95_2l(BH?B6~Yq
zpafARCL3l<vH^)~;ba9FrQ}|st5xDSP*>7gm9ZMpn@-sqkC-^m$IqW2jTF92DiD@R
zbDp=q^B4T#zxe~ElOaxnHl`e37I^nc3Zw`Wf<xa{AN=Ja9Y&QJ51A80-tn=y8tdbN
z=ij=<Cmy;#Jh>uA$01?!P9cOrAX#Y5(~Ki<AeAJIW2#Y1q{GykkgzF&${1u6x)MhZ
z^oOsXIH>#l=Rf}=Ck?#(^2<Ng?==}m2ww><qjC5>D{jgu3DM3r+poON!r~GljX^qs
z!dSh}Svhz(1un|*YTU6O_`bky)&|g0F~4|<&wTOM81G)gW~0Dg@dm6TZp2LcQ@T4F
z2n{y(2yKZ)j4KR#+nX%TwWu{BthZ<ZHp`ig$4rJJ@G!~Xk$WEF@BaFKPo+^uc#uAN
zU$!!P!4IAA%YXJ9Rql?Iyn-}IsWvJhIB<qI4vyCb$JBWynIx@vETPn$XQg`USfw>6
zM339Rh+|~znM(G<^EC2t1hyLXI8xvR$t0uR^l)&O-dopLYBv!{4ALUKKip$aNZ<v5
z;2)kq*xutAy*RW)@P|AjEhSnhDq7Pzd4|t_@e6b}u8~cq6vpCQjxZLPMik>Q+jnmh
z3rm)HgfY~Fg2Hiazt4S(HI^$8xieH`P%CBAoMJozudzmO>cRv3>fifsskR$v589^K
zXx82<?MzCJNmLFP&&%&9L~vmD&LhgPG;OU|NHrA!zT}SxFR&gaVoD>QL!}uWh4&HT
z$&~&id*3^i8xyQ;{E>Zxl@eKUH~dGm$3JMDjS^OUgg+QF-R*!*n2tvzam;a{lXIc{
zYMm!i5`jIG?<d0Ioxr$=q^@b5Igd({Kn$0GhZU;wfyPjHhsy=Guis|<<~mzD8?0}y
z)9>~$CZkrVu`s{D^3oz_&YtGp`!BMxvcR-+mwF|lU2V|W*<(B$(4K44Xe}a;2pfS2
zqLhAjKpG{?HycEih);jwSvI$}dFAz2S-*J=Atk+j!E_pk;>P6Uc#>43G$fT8r_Ww0
zDHj%3gi0ebE^uiMes~Vd&IgXf9|(tY7GpxIFSCYmzr*V<{vmNxXa4MG*;xNJ)@9^H
z??dmUnZ{si@56CA$*8qF9IVrQ>l%yg#$N%BzesoED%m8XFrkVmj6<dob~0h(?rkDr
z$fgbv8LB0YukZJ{f2qb&iQ`m!0FDb-J#(2~{d@l%wRRKD5sqgzQtlXzb&2ERhv2xS
z74ye%TuEX$j^knumE^eh;W+;ZaqK@597ok0)Bqyzc}Ng*oiZ%IFP~=qk<08H^zf4m
zoW_aJB8+vI+3KAK?@(Ifg*a-<mb1%)KH`Wo?!Cjhz*Q@ZWn2_=*RSy(|KRtTKl1=-
zH6_vtDPp`<*h!AA)HreG0=bV!DiT-3486f?#l6dIUf&+DJ;{k>#Ke`H#55+Z)j@c4
z1lO*;&cFO$|0#omE>=KcC+MQ&H*++iQb`%+q^povNWvXyoDGKOA{c6Sjb$%0jGaeJ
zaz@iBUU-xc#7d&$96J>V5qu8o)e7xu!lBKW5X4%C5vS$-Z=FRsf%EdP*fN_g@uSvH
z9MpZ`3tymKum7aK>;u>OoqZ4n<%6TLw1$(S;FQvcBId>mZ?Jss49ibGf>IhqP#4W2
z=R>~PEL%TRy_IH`WvjLPf6IRl8KG1Pts|V&G#1vVCz8~S>F#ecoet4LGwkhg_xgLJ
zN>OhwqKT+in!yXrL$`Cl)$3Qd_rCi{DmCz)Y&vB+nNm$+=I5$B_~a+}jsO0Cpw^tn
zmww!MiIlw0YQfFCeEc!+5|pXoM+DXy9G)nOgR46pse)+4Ig4=?<6tuNG!k%LU|rw<
zOvWRu3GzB!j#-=e(aox!v(F;SAn&87$@X_fpiDV;7er>LM|lWR#tA#Gzsua(DvJ-C
zMQM#fpoPGQnP?>JUleIXS&j9C3`fXNrdEgc700qHDTt!bA~z~xdG!p7k)h=$+}PNo
zGZ+wwZKiu$y!7rZTGDf})u2K`yD^891Ek+&yEo=fuW$2<6-A?b?<cvTm}Ep+k=82Q
z^T=oUyZ`9_OuZHMIC$wvbjhJmB4mJ>F@a26URp|qYJ{DAcEVVl!Z}hU(ISw>BNc2c
zT<Kz-Sx+$;6UBjG?wzA3ES>%UYYYNN<A_K~ys$XJpXX&D{u;vHwLIdRe$-1N_*mSe
zAr)pg<lycNbWKv&f?6$wGGX!FqqSx-9%83iDQF0T;FW+>dWI%gd+2J7`r;D83D5#r
z`pij!`S8y69@8x2&h;(s+`P_LzV^qwef4dIlbNkgXbHwq!eV=YGbhh*?!tMVeDpyc
zx~D>Wu}Y=YBCb@~zH^u6+yZ#m-P}gWh_h$!q1)T1f6yhVB*YRTDOg;b$K$!sJB4=z
zced`bwY`TBj&?gCjy2QCglOVvwi{e{;9*uztsx|(5<Uk+iAz!9e@2uJGn$<V)CpNq
z3%oBdRsfoMqegG9&-(RO*|_sMU;Neo2@!eD?L5Nu_g-e}?tkKNQ0QNB^|prr$CRR~
z+)#^h>N6aF_Z{ZePOy0YUmnL&l!M;+!8l$yd4{F9C~>?`cQ_0<-rwTocW=_bbE@5-
zB9dVFB!?)D;rIbfqY(r;liW~@1CEnw%7ur2j=%Tc{$uLxg|bzAgyXctF+gri8OV98
z09Cpd*z&U@W;nJ~<oj`~zX!)=JRyo~;5W~3-0OcIj-9Y&C+mZ7tUeMP&z)?7jY=gs
z_((H!DT6$Rc*=<<A7TI2UAFtzutkpXpuH!e6k?TonwR{r*@f(^E$5quM8ucfq3|3+
z;*uAJah?&Gf{i=Z*~%iiqRPcHE6lfA6n>R-=_JoSd@nsOxP8!L>?d@(dpO&|ae|ZW
z8eh8i6#s5iFc=K*KsAYoYbk{(m<&g}_wt|f|NejeKj`o8B89{vz&i>HlROVoFa#u8
zhuOpkln<X%;!&l*!+U?YH{;UA#v#b93(8~_f<mnzsw(olykn&xiZu)E7CO>+!k=3>
zi}V3E)`fG~I#1+9;PKhnwCf+ee&V3+pZt@5@{tK8Ji*A)JBuIgaK;{RQiDYygeTh{
za_vvQPGe<}+PU+f6td)&851g0M`D9xnaXSsa_Lkqg*;SVNhuIgA(X<aaCr!s;%lcE
z_SP9s4V`X>VXw=<)(&a4LSu1>@o>aJw~yN2VRvVjjjc_T62#pO8+UHAu)NB6G+;37
zlG&Wgm+#}@$DZZ6U;I_-bBlO;<Q3;}rB#VPv|*7-AcYGhVvZE4o>}(l$fd{2aFLbP
zjvxfahxUsx1<t`Zb2u|=`8k{s<k=8!Z6G9<oTWf|@svYJaV0-PN`Z2q)kksThl|I1
ziyv%Lvj^0w;Dii><L-!!=U=3{&_thHK`D(cxeHlg0x<L}i_(YOmf1YaP<2(Mk~Cb3
zu@VTSkdel!Fn&&`3e4TZ_TEhnvK_AM4%pZqa(8z?tDbObsmXqC%FTX;Jm_(2f57_o
z0U}at_7&f`zQ?^wIlF@?y~%{a7OY%+i1Uv;%`boHOEec3!+>tD2w_xc`#9%`v<R16
z6j4Zp_4cqz;k?B95cC=E0x4Ol5LlN_4*`V>Z5MAWvY%q2F}_g&EZ&ds#xovvamI#h
z8BZhtt!65iGT01ztQ9y5`Xd+x_d)BNPB|?lLOc4y4tsZQFu%Hhih}ip_YQ|8QZW)3
zbPiByOq5g*aty)~35Nr6C-A~StwOclL~4Ohfr~PoPTAeOi+7GUUc18b(kb43?@j*j
zkNyqUZ(YOkegTzrFv+H@Z{Owa&IYf)`xcuw-{buM>93(qw29)FL|53~dX=4xT`oWP
z5T`F*=GLv7y!6_Oj0Z!e(-F=W2;orx9vamuCr_TFKj<>N{tlg9pD4D>&nL{yCG;at
ze>A1uKE<#7&3{N5$G9TL78Yl0X^Ro8guD;j$8a_fg1GWsI;0{2(rYU9HcoqXHg9w7
zov-3e#;ebNn@ZL5^I!Tqyz=6gDSlipgde&}96O62Y*9Cz55VzD)E1iP)xR8$)o~mP
z5d;@M0LNRqH#x|5xw6w|V}HQiogwvF#M(-W{oa%t{VDpO$F2P#ceVnK@Af1wT;Jv1
z<&52aiDOf+eDOifJ^UoU{7b(?YjNoZ;yBiUx)UA8F&wsvoeNIU);ol9SgA{1=MTnl
z5sye4l`@DN;SEeiy${53UfO(pFOKa;hGPd9kP6W_$sk%}IPs;grXX5qaQf39XSmzN
z+zFOkhG6}XL{XR^Q4R+1hwax;z4N}DkmO7$NU&Jv$xSfHcX@#}2Is+>5n}%xPOe;H
zF+IusC)%tnwK%)lq(8{Gz13y0UFFhT1+z5Ig{3CvPA%aJ!|6H4C+}P3&CLoYS{3H1
zRU#d+`Oa%xed$~L?(hE|o$W1zmVhl2?!c>+LX!Ie>pex`h@3@>AW(7Mql7Ez1)(B~
z6R4Tum!qpRF2{XJtq~FEBtl3RD*iG;YN$2lkx`hzZ4eY*fU!_wT?m140_&i#9%;e;
zm@xi{g1V1dDJQ%@IS0jnQ*MWQbsvdf*LakW6y7nsvCYjt`zDuvwS`<<JPhJyS?laC
z%IeuH7&@xd9kp9$_9k9Pq>jUwAwtE>$1y5Ci$B;xDvLsrHk-jJBuc0))LC3RPp{hz
z7fvFXOhzO+W;)5}^$zF{dkjWhl4_On7oXt5r6;J(FJoOETDoQVK8!C5FUQOg{!qzV
zwjgB5eswPVjJ2h_S_i|_VAOdupEdkU;B!ohQbFoHzWo19Q5?$5XZ8p}aLD2FxRRS4
zvZ$ly`k{9jX&rvrqvm&+kNXHA7)?DwhOt?byE|;Y@LejOe-<^@3N{rov}%-6hm}Qv
zJQR#L%6%0_mw{Fx<3|a+(&#8uN%X8Lmd3>CJ^0NHv@)Rr+o-TMSA{fTp<3tcr8evP
z1EiMdNYKr4QmyGu3^zI>cE(fs(-D<w%0rhQ<jKnq(q393a|OnS89t%^RsaD207*na
zRF~fR!%<HV_;RwWMnp=Ih@f~>LV}Q3ALcoF>xt2zBh+DOK9uo+aClftVI{@L;KM|E
zyw9=DG0k$UnK|0KL%vUVCh|~q7C4lp!tkfie$@z`(lt^M#_eDX-8(l(@4tku)DSjg
z!%|A3D5iI?$2gnNYBZ=+Q>Jo&@&c`VxCDhl;Yb@zl6sXWP7tLt`?T93n;h`Mx1Z;M
z`<~<0jl2B$SO1vx&0EwOb+Rm@usM}V%JSk8+uOTLrc+8C3E6l|qv5Ee0*46`ZH(p2
zrHe>3MdoqBarWd{-W?5C-`K|#1r1SQFdC55>tH?8$&m4A$maGIcQ<w^vYbXkvAo=(
z-|u3K<ADbr=5PFs-=NiKlTF6uD6{ul7ll!Q4qs;f;f3<dD$z~|Tqz={*XKymD!T_8
ztY3ePz0J2tWs~iVZ*zL>e!lvrzstt$>txwSF*Rg{<32UN!+bnI2;Pt5^=&p@c%6Gb
zS3}J;e<Y3(;L8^L_u^Q}U>owmIF5+Zi>z;LB4t4=71c({+4&lxk{;st_Ff+^1yVyd
z%c;z8d~k&0w3hPVy$|rz<p-HtUS{fw;J5YxIL@slkqRXQk(4Adw8nxE<Os*vF&yjf
z!?8cYv5^!*gI^B!LBMgC0`$Q+mZ11P94je4G8{W4AZasHtL)`1@Sb`aC{9W$s%KAf
z`cqFb{*OcKe&_H$iL^T8Yr8UGEBmdX>RbkL!Sl@q1*vo48QU-wXPO(V_Za7q(=OTc
zZO*No#=3~yS%&>F*LQ}v(U{#~j~j0vaDKUsaAB_Oe3~#R4D;29ON$Lo)GM?rbtc0J
zufFvf-+JR^_BS_4WfUxSN)D&DA(#S)%DJ@0ISOmZT{*Yd2JWJkkZ1*3ka+{9C@pLP
z|J{^zNFfD6E2QwmNhl`GicLysDwQ<Mtrkan=gU6_LYG2BGigA0{^rM=@boiPze*I5
zA9K=ea4yQJ2uY4adW#l<0*5u8ombwVb#j%{Kldp}Lv_xUVhL3q6tha*Od@}DP#*KJ
zlcn-XC`9R1HM=mab4V2`dU0B1uHB{$y#3N^jQeBGUbq)iacHlo=_W%v<X~f+?Y$j-
z?&m*4tzKondjLYSxO|$&Kk-F`(t%KKDOEw`OXNH_&uqd2aMVUFvzu8EYQ@Z-3WPg$
z_=j)Cj669j>3akd3%PM<WLd4V6h#rLBmU^Q8RHHEyBWXAS$o(H{78a2FOYsjHQ%Js
zD9}P-kqB>)C>$1Np!@FI>?|#@_RK?&gjPpx3l7JcNp+av&nzsYI0`(al;!>fzKNF_
z6-QKJO<Lw6nGcf|BCTlDCFf5paMr^&pMRH~Ue5jJ&oEz+D4}TU3OjbpU~7lF-F3e3
zspnW}CEOWKKt!BeS>~5M{TZaz^o$29&_ae_MT9w>-nyWyET5+mGE@=E$2#L~<ksPQ
zAl}N7({Vg~3e|vc8F=S#vdYxvSldI2Hk56REYA)(d|C+-t(X?B^jZ>VABK?|V-Ug~
z@<u-Dbxx(6Q88M1jPXQLvcIuMJIhIv#%ww(l}Z(BEd725DI~SDPExCpvxmZ?gr%?^
zRVrHRt1BeUHZn@lO5#kx?VC5*-?~k&Goaravb(#9wIfzn7a30{oH=`%ix)3YsWjN$
z+#;V&NTQO{IL&CJnqU6nr$}R6wmB`%8Y(hkY!oN&dyxLl9-X_ls7H0qo;pK+)TNeI
z7>`Ef=t9r#&JH^}+uXT*n|I&2jgo>^Eh1Kuo$V>hD{K79-}rS-te#~u868&XwTeJ$
zL|MHIHY*ly19{xppqSHIhxw5raOf(vIp!CZnVN#$-ZlqYZy`mD(g~CCCbw_D!!YYJ
z>W#@h0`cn%$CZ4OW+O)nc?`!g3}!dbefKT4mzG$2`awvdQ2qEJICkHMV^LB{el(7?
zRx}%u3n%6|=LFw){vCFEIrm*S!(v5}NX0@_WzS9+ZSAtrUFR1+`Rw~}d}?KhU;fNz
z5IUk~d@xa-;h2Eq8IKMqgSaq<`}=Tg98Ucx9CL(YRV8yd*7cCIe;|%yt;h;jKBEu9
zvHVDI>~T2JU=*FEJBlfAXm9Z{OzBZ*q=vZ%A7^!Qo15Q!89xmPWgN$c{6gz2p5jpW
zB84x-JY_q!v@r_2F^8}QlNU^~f@z@MkxGFvfos?6S>C(5%QXp~`^>|fXeYdW<vKfK
z$JAPO^ASs`PUGYXXXa1w@BZX3*z8U?*Q|msxV^c_zxty;#G8V)l6d7qfK-ZJgbI94
zRXRTS@-EF?DN^)dcDi$pL@A^IEd$ZUTKm4X=qyB%N|eG2N2Cb!88|dUPm(13cM&}P
zGGmaD#JUjp8v%tep+#G^e}5`L-HjVJSXx^8Nq)%>UCkuo%=rfOTN7fd@Lm%O8>(qi
z;H)PXig@g~_RSY)tgSMC>3(FS@Ma3uBeVh$9ag`D3lBMnt3r9x35ifD7+TH-fCnZr
z&N_;speS;Lx1b`NFgWR{rY-J!<S}+O?lOPo6xF#^?6g2n2bf5(wY$z^=N@JG{P}QE
z+`NfLu(-5_mkHjQFyqw+QuvXhy+j<UUmd}iaK-~HgSGNBm=Pyc$+goZKg*ZeBvF13
z??QGDP*}mlKve~0oOKyR;V5ho%uA&~;c${ev4s2{0&b}0efSJNi4=3EYN(Ad%17WM
zv@n4=tpvXCSfOzf$K6+7qqefZ(%Chf)<=7cM#wm{)V&P~JOp8+Z@A>FR7kB*ktCI2
ztc{ieQ69*XykJ@sNN15zgcVb)^2}A|`NR|V^ZNBo*3PZbZl7Q}%}KHmCYEgM?{I$Y
zK`!2Nl4hcL{?0z$!|9b3tc>x_A*{u?;PyUKDTe*jSYN76g+-`fTcOH5o)x~el6bGM
z-V=FGDrOUG%Xfg8>aoo7J=mgwu|9CteHG^m@;s38JsvF!V&O27lA{$yGg%d?b4J;)
zzwa|T`|wKxPsLS)44Gvx(e+HaQ_N_JTnz7a94BZU(eHN|kA_t1bxf?NG%Jj7c;(PI
z3Q-O@j|8h1?x9j`5(yA{*Z9Vt|0(Zo?ogduU~_Aa)29YFG3E5xMNXVN$)is`L8~>-
zSHALPu3o*u^=mijb;n^4yAq^{=DBC?;o<wvGZ~M`r#ZG#go<jCuvly3x+9vEI>|j3
zxc26&yz|Dp6t>{fefMCjqu1-u{r@xfr$Le)>3t{obAQjcBl6D5touHI!hJT-jc)cq
zicPXbQA3j`iKE@IrO{?)zl`n77nx-Gpp{H!E3^49JEPUwN;@K{nbv5?7DaY9`vCgH
zjY1VrcU4wZ-bX~n9q}ISz90OJ$O0PZ1_;n*-K@I03RzK=;qmwP@#py;jXEpq>uhgt
zar@R1oo*lDN{n=9cN4z!rLXa&FMXAXnHox2$NjNT$~Z{J`dK=zK%#B8L+i0h-NyRa
z(WL~W<5R9qAsy&+nry9Jp|!Kb{)O`hR}!s9^jb}lcu1!+u;SoPxO~r{HlM?#hr|1E
zEHSCFIPPoKE?lrUKCoc3u0H_B!YKGa9J@HK18#<6*Wy^(82}JL@4lGDahf7^itG6(
zF+e#`E7f@R(KB4Sz0QF{2dGqM84M!axKHd$*6N!aTsY0KWAju4k5`r(7+`T`j#LK7
z55lnoVI)GoAII`O9PdrAtQvL?$H7OyaT=HG#m5n297pfNaUe7*!?F1*#BrQx)>rx@
zizO5hf`}B~?wv9SPvDxAv>-X~<P&t)w%A;{Lz<*WFNbg>h6JM`Yl$LVaC{tNgDM0D
zVHA!^QAvtQH3OYs(v(yiw6vqgyzi0oJhbDI#u1Jw;3`3*wZ->-a)C!r9$@LlZJN7l
z<Z=N?Sj3MM2WIBEb>T8suHK_LjCt<x3<9{jwu-A_w2&A_;t6BvN|~LC(RN(yI3vEQ
zgGsGeL1i4eQ9&9j%Yt@<z;h)-lr7e>_fTWTVkVIl;f|+p9n0IZ&60p1aPiy#fgsgB
zk#X>{)CyrFiPFe4#Z?MV>+I9>SKw3k%fI}~{OCtNVq#(ffZzZ9-{&9y<A2Q5)D!?)
zTU#7DbclYx|B+trmwxRGXAeKl<v-}Lb?Let#oC6}IM6xK8Yx|p^%l3k_cFo61kU^n
zjuxca;0g!fxLK&T#H4%GH9}-jGY(R^HbWFauQQ}^bCaFzO&X1DcAHH)-402buz$+o
z3(uY+pC1x<d88+)RjTy2TI8H6y4xq(t<&AwrB<C{a%PG!45@E#(T@hi#$|Tl2x$_7
zR_`&|9N%1|WB;vL7fok;r;*wf7`5W|R8XV;&BjLKgF@~pX6#p(j5h>9F7q`qS?hZg
z3JrS&=6$bjJA_%QnEUfG|MKhH=MQn<;3;nWNtf1*+eoc4$AgR#ohf|;9%`q<%Bz>~
zt3~8=1!)AaG6cc}*GIYzQY&<9GEs^BpgYn<dJeAZ6S_XBfL625%_}QxZ*S4uZPKoH
z8T1FJL@_a8C_X*URKHKjFXDQFqqPF9wK@}C6~Eb`v9rzUPMch9g6XL|wPK!YD_b<;
z6k{BY95~2Or3h_M%5pM9cF5ZT**=7|F(gW5yvI>@c5jbOwr6`gVc>n|vm=C6)!8pb
z3&T(<d*Fq6%U9GIl`4`nVdQOP$5qzA*LL;oD8!Mr+h8(h`cHEmb^{h1YiS{jA@8`h
zb3BOAy{;WA3x_bw5v4JmZVTcR$ABApRQJu$mX~maLK@2lbRCUX@z^&LQd)V9{Gh|F
z*M7>|-+P74$wl1C4j4h4B!qdv=YQ#W&OPu5Nech#|N6(g_QqRmZSK%+_0bv-8d4|*
zE?@fGlT1uZvMVL+PM68bK0MDu<Xxgchu%(uv>nrn2lQG^X67dFi$x9|J%Tm~8=Gsa
zZmiPi^g&AIW~S)GlEuTPIREH*j-NQg-1L6HAy$2aw!#oSKB$Gg7+}YJI@=EGUALDB
zoEnYDs2V|`j29N@CVd*)%iOv7LsZgbdvlrC3|QEIhU=Z5&}wzE&w}KWUeM0HaF7EB
zPjd4cJz6)Geip}i569_to7EAHrz$uf2gk01<NAc2PbvkS=73w**4WwJWVg|x-RLst
z^-*bt<0t2s&T#B`l4I2Z&9xn>UX`HLVRw6*wVgJ(+9Wemc_xYluCHvf8z&g;aOB`Y
zhDsqaugH&&W5FI|-;3k-*I7S<V+=#7N#o)FU*Pz{pWmcWzsKp{E>S&w94$Re!dMt!
zq(Dj+LyS9H<-qfg5$(3oo4Y8X@m;~tq>x7FM4_D&>1X3bVXcNzm7<fBRK*OE7*`uA
z6M0UaJHYJRJ|?Fos7{s16$`kY!_JPN^Mh433kBLyND^zV)Z0uRn5B_4h-zgfXXn{J
zTccOsW@EEUqtoT|!FeVF4=EMbZ{H&4IA|%y=FpiK*LruNP-vveRH(x8bw}^0$sW3y
zeO<OjAqek%-<5$>>G%A$>wp_b)W8aFq>y-?JJtj!V?8)iZE&T`WMX!zs1>Bjdewf!
zK#7l6)Lp%L^*weX|MuVhTmGB>=D!&ib;cNm!{JAMy=R|052DA3FMomN)(+0*4o29F
zN(bvvq>Mrd4@Y@4uH0hvl~*|Qh0mgc5*i7Pi*T$7XQmZJj1No)*Ygmb#1R4QR-b!I
z_qcOsi91WTxpDgj>znJeJ8hzPh|!V<kInM%xfu#YkI>KA(MQh1_k9NKEuwypBpHCK
z@k@EUTu8s&WpjOl&Tz=&;!*O|Ih2YRpU^OVj*p9}_l>g0-aT0dXT()dnL>3O737%r
z{28#f0)1cPmokQ7Y*687{F?FVMs@;HjQ<^DPbj0Xf!+2dm}18KzE83~_t^(&CT$Kq
z`vmXq?jjpIn2Z=bl5`uLqP+mAB%P%@Y`k4%|I-hmyd26n;Ccwp$MIZ*wi+{$+C7%S
zBM5w)(7~4>+x0HjZr@_{?h5s#yX@S)L3eG7sM#b=n#efM!l`M_p8lImmK<tf56_Y0
ziXO$#r{7qm)9w%t20%(K=aLUS>h%_v*P3*rgaeC<<R@l`RbqqK_Zq9f=wZOvY;&}<
z$q2a!Yky-#ZGO8qiM9dQM(!2a_w#8+omeF!a1+u&f*%Ns8Dx8G6_P-fs*vTrYK_ex
zOwc+(N{#PX&Zgsh@;x4{Pzat03OG1;!r*&`*ciG-&`l$ve&2c{Is&CM{cf9KtBr9C
z(uI7!NHkreI^|KW6e$%8l*(0#HJ_>33KQNo?$R2Vl+NlMW)2+Sc6~snyGE_Dh|-#w
zxk;XQ;u!|Rgde>09e((OmuR-ybh<;d7GU*m_>RkTwaVF3rzn<dXf%zTdS;L0Qpgn%
zv>6OqL}EyL&>~I}<_{iV_TT{u<uW_Vcc?d83}VIX;xTgh8mAxpEWfq5NM&jca4{pr
zoL0ET8ROTcM_{}7B70(FyKU(ZHcJ%WCrvF;T>_39P%PCTO}T#cDi>e>KI_X@3B3|Y
zG9-%Y+_?H0(I90wNbv$eFZF<relCB|`rKz9q?NWwo_UhXja}q!{ljsb`m}E^vH5nD
z#it(l1vplKvxnoH#qmy~$F-$ftln9ne&-%LOE>7OZV@$_B<U_9&a-fGnzJW=i^;M>
zb%f)R#qprAO1Ir684SRr<U*Hx&ZE(2^X~dC-6&!I!UBbf8DbTGA{=XDM`Lkj&yi+$
z{)ISpERNM6#k>c{>SuARMHVOd0XXK<!SOf0`6`E}58-|9B_5iZL{CnS=>QsyG}d}Z
zrW)nORE{5K-{U*neen&Pq>n2UB8f3^gi2D3u%e&HbwDVzNf<^$(xD<&LuRImoI7!f
zgGUcDedHihM^8|mTOh2|aQ!?;#mbF!F1+(cY&TOHgAR$(^xFxY{*e6C0ZNrBe!c+0
zWzbLP_7f(js+`|HNjdOYx_gK2_8Ng}b8T&b`u@EnGrrFLqaYix%JJxZRFDT5H`H?^
zp5rihpQv1u6?HPhw6Ph+UQQAZQxI0@=r}H}W6Lj<J=>MAidRQ!JmKRAAFU-(tPsAo
z2F4#@eWDcL58~K;obFD!RH4^TDI7Y$kr$p|<&VEf8tE}fTnH=MR%8WnVp#dfo0Mj!
zDL(uNN@*PF?TI3=1jbZZm5`J!!nMYIjjbN<UVeuw*Dmwo55C8p<y%BaJic2Az`>;-
z4GDuh?N%G#2`J?%1o^zp<W!m{d+-L80AB_~aYW<Z5{=dlk#>0O!6$4xR%={o{p0Md
zJgd^#>$I7XJetgh&RE&=C=tMJwi$cMjsB&!mJ_1^Z1(>xCr%36q>L0K+I~nz`~{_w
z>=mq~ijy5kBTcSxvVt!wMn8?)-(0Oke<volFvHwq4^n^WMbboPZDxUxE+fsa5LQ33
zbMX@S$r|D5BP2@WIhKFmAaETQ;|n5T%`P3+M}~Hf*VkIS{_X`<uixhCw|~UWt(O_Z
z4N@E&*Twe*<!G8T@hOKsD~$nVugOBWO0iNxD}fVhI$GO6cIlEAF0En0ouy6IyFG*v
zeCCn!qzKR{dB;a0h{-r3LfY}Gofg@m8c<jwwJ`!m7*b<UMw2K_tSxsWHTpf{oUFPZ
z?=gUoE;+4m(ioLSmXw@z50tU$GW$oR<>LsexO1}N!ULodNVUgd`n1+rH)Nj-@JPw~
zmP*izpsxjKqDYg-%1<>!!#@31lXw{6doD9BOh#KAn2|j3Tc05)IQYIT0)%VdPe&?5
zY}K%(ykz6@4Vq~V<>oPYg(Ee^LXk$J&#haxSX#Qv;^G2>K_3b}abi91JYi`V-R^+(
z%@$`)y98klAsx21H#l+R3{v_C;SvM|;v~T>dsNFcs?{pK8?b)w9?SP`(;M^&i!(g_
z{BKYwRnc0Iq%lgx=uEhT-NCWH=h#~_v(3=TWM6A}XO>ct6#&M7n_=957Z#D8&re_b
zDgW%B{@=NG`&~YJzQ)o0`xvH8z$Zz2^!pJ~O8i`&Cm;O^FJAs%KGELJPm~$#B;*%n
zn0@R)>My;7PCo?4nGS6G?Th3lYlNphB8~$GDI7M|+r0kH1=emZarN6jV*AEV=qC+Q
zV@I=N9LFA|&|{@OpyZBlTtue=nP@snF-SE<;gXjQ?P0=?Z*Q{E?ST<|=Fx}VgJXf^
zasLdC`B@zE9vqvDRBt(D9}mY7X&RwT31d`-<Mh2amIB|!hRMGN#~_uRt$!LgUfyZ4
zFxjVnZ;iE|yw3EOzJlX9R;`P%5tay~vAi=}n0@SV`dbYKm#*Qtl2{F_0@fH3og!jG
zU>u~>q)Ch#rWmbQ*k9$Dho5Be#0gG({tHYVI)~@w_TDNW76aU%Lu(Loa3)VHnPj<H
zr@Pxix(4k_DvnFulXQ%trxn$l$FY-(ES8HPG~fN{EB0z7T>3im|I3u~n6aD#M`TuB
z0$ian(oPzbu$6%$1-|2qf4I_NbPcm>G-WejMxA$??jwxA@x-39gk!tXjvQJ0Xi{ZJ
zjKG(YR9PDwJM1$!P9~mkKN^qX6BTuzbP2~=c?i#C=KL9Y*Ke_X;W}<+u_2`-9gPiy
zwq9quJ#PKzRZh*#A?75aHf`-W449E$nd^?NOmHNFUdrWnuJDsrf6RaS&Y#iibujn&
zHi_&;=3vky1c8Szn)S^MoVk#imq$t$KMXJ$92wyH0gYCJ?d@&4gC6x}o$B0Cip5DQ
zZypD7Td-+k<?sT)+~=yT%~-0OT^JadHjcV;T95b~qmEuS#W#$L-@T$vP^^}5OiG;e
z(OP64I@Y0n<h>-MRmVz6mSd+Ghb}AbOPly)%W(v5W-(Gdb%5c$BebvG#?lQ&T}G8P
zDKtV0OuNt8g^TQ;t`P|rX)-mCv4yM=5*b(`s+(~ohs_>uy>pS<ue`<EfBs!|cXt>J
zo46=EG=)+L=_=fyhYqULat_J}-d=4Ih>(Lp3CHm$Oyt2BLKzSa`rO>@a(lN=JBn$y
zTP)1&XR<OuY(NX=J={2D#CQ+Is8G&GuCXCgNZDt!#QuJy6p7MfBCL^F7bhD7W%YX|
z^0EGkp)rJ0`*FNJDj9&ujc=6NSc0__PS$DAxDKRRp|f0J*Y!wKW21dOg*^t_rF+cJ
zBY?7RKuUUvAyN*#RHLOq^oQU}j4(vK7X7G0L3cTb;Aqk%*EJY1KvaAiOfM3ok)W(c
zrO6`h(q5ljk}x5BKL6YcT)TUh{vbk{gfxli4SN(z6)Ke~&1M6wEos~L1VTHOtE*x<
zk>TI|(I0c@zyTH)rm0L+xpd(jk~n2?ei6@gEp5XK@w|vA8q#XE*j(SD(b{3B-NMh$
z@c3uGOs+hGCLzhNEdhia7u+J_+IXIa<2WRVy<m+leB+K?wnmf+SCWZ|?8G9-m+1G0
zD5W`h`Xtw{y~9=`CFgD6ctacTu3#`266Smk9)E(b{jL9!fA&9qvZ8K`W7nZ}>L9~=
zhv{6uofQ)wj^p*W-ez&SMkJjN#j&^_$E_Z3Ub@Kb*WTjoKl?6?#twr)6IUo)G=<^>
z(oJ#wZiZu*M8n0^CZP;D5R`CSk3uPrFc8Wf;c&ps-7ZVJJz8;0r`2M9Za-6%2_j>x
zaE`1~gYD(skK?^d_Iq*cjB%|0ayZt8aP|O>YjJE${)2HW1&+|TdW2(@;n@FFaja8F
zQiDkq^@}&j9l1v3%t_nfHYUsG&BT00Vx&e?i!46*ICnO82u*6cO@I*wfktSB9@_UP
zO%+BPa<w8)J@PompL~)BzVcPVQVl6{kiG9lYkSebL28diKgE-VM-NUg(0%S~w^2z<
zuieE6L0=W==L?*gEOUISL@fxYRI4nnF0r+{O@Odm>})?^6hmmM04;10W$OTepHaYK
zY-2<U>xCLP4xa1aN{J^WLT5}jF>dhLv2NCY(aIL-(ats+AIB>rgwE`$EHBeu@Dxf}
zflaEcU5XSgsWQY#g6G+YKq<Tr*tMTsAKTL8{hhc^!n+!q151$k68k>)9G&%D+_kOj
zV6+Gt=U5tpXB_%VD{Q^~Ci9;yqE(8kLyYT;3n|BOakAE?PT}^o+g!c&4uAF+e@eIC
z1vIWJnV*{@8VraN>p6u|cs>|X_AMM_eSL+M^<{iFpjMvT3m$h|;y7V_eS`I_HTqEm
z!l5>~i0g;8XwxIdIm>?-^?gU`$dTIX{w!9lwe_JBBC{gN%o?-r-*ScZm=|YXopAv?
zD)53_Kt3oj+#cFaU7BWgCVG4g62caBBO~Y0zr#BGXCKl}p;PBbZ1j-?5#*S7>|qAm
z9mLkAz0nAJSa9DSyV@aITc!EVb=)U&=rkcz9?EwyawI)>39}RuJ&1Yd+7fHmZgS-t
ze@4B%4YEyjHqZ3@QJhrUE~F<QO>im|w9!0tc%BRQws>u2n|zpKeqxR^QFyM4?|O72
z&CTr&>-AmYVFZrL?DQO=7Z7IL#WXui9brhZoj7|K?<q?MsmeqHkZRDz7F<dj5@kqK
zI#zR=Y?PzT9>3j;_RFy&Q*siM^YOzfs<VVvj@@6avIE6rqaJ}!8Q;c0lEodlu7gTP
zQEZ>i`1Q0GG9f&2jy1P#C5nC`7#QfQn6mUyogu!`_9maEOb%MicGjsm4h2=RF7VQJ
zG#uq3wLnENUeP0oBTHgVQ`9hG5XIyQ61SYEQ1Dq@SwW=<aqQzrmxG6n0dVosTSRe0
z9N9w4l@7|lFwwZ4=B>-O`M>_R|Cs;$@BS7~K7O9s+zj9S$@f{Do#WKu<9MD=v(u*8
z+$ByTgd-Tn1Nw<(-{A*2cIqi|g(7J@w1vD<7^BC<{+Q;Fy-!lg@#QN?Z48*>yf+VX
zFXcsMv4$FhMEZGrFGqcMmzkMa{?GrX|AJP1iz}~w4_`sG=9A00B(eRhg}D+>fBtV#
znfc`6HhDje!;s0xA7QxNL2PgRTpSNaIDX=O96K0kKi!_=5c;0wm_;$~T)V^CjhkHl
z<3D4ky$!NObuQ1;>>(UGsYUqG;<##Y{P2-^F5KPXwUr$TL7utEIaI3gT#vx>>9u>@
z+Uc-fZxRiMAYJCB=LkLw$NIfEc1LIRhvC@#WpFGZFCpjSg%xz?4n{lUb9z)5d?1c9
zA<d_a<Iz+H1!gc{`$sR6o2`<YozIH)?7ffzv~64@QW5Sy$i%~ENq^YKaRY`<jEqy9
zECMe{QwDKjd5SrohtHj0|EbfQ{gtngE7ricBtu1K>mI@l5W=&LUyg_BS(k)YuHEB_
zqx*RF&^*0ikF}i@9M2(1W1={sR4a3Gw!;3I34&Y>RLa`sD#Lil0EsI{#g0H}f#-}w
z34pQIF#tKwwMk9JQbrwR_uP>J-<9}|JMN%Mkr{>~#x0@|b;_u$`JxOe8WKfKoN^u^
zjpdY%E!OPk!d?XAnA7QbE{-Efk`&jo&(YD?{uix}ZE13UOxR#B;HzK#>RwFMeSEu*
zvV;IcMh;E{!Tdh<Kl?b1Z+r_iOf$96s2wgcb6kO!DpoICB!BWKvqw)NgPiS}NE-|-
zrTza&oY2^Ava-C!Phb5J{XrMkwHvbkzyfDapW@Pmi*!4k@u+jyPf-f;#S#nq4{&$s
z4sTt2ol3DvK3~9*(Cc(qy0^sg`U<ImTv$MQ9y_~zy1j^6IR^$(lpU*NN4o9YkQo(@
zICS^F9|CPGU^1WJ@z;rrm^uEn*=zpVTc%_%=o5wkVJ^?HBS*MY-zH8iKSL{JNvfGN
zo2*NiCYcf`Nl8_jJ%p?hd5<dkmtP~osg?$z!JDoz_3$}*->;)1m3ir90og)egs~oR
zDy4q?GU1W^Bu67$iWuQoPDkd;<VZ<0OxWJoVr6xcdq4dt-DHJAEyNFfss|6V@5r<4
z+<b%n&IYR6L$wnsm4H+mCMq=^J?ik|tE>EPZ(L)4ahXakz;g`Stv;7lwpiWVLTiDS
zFCtuzwR)F&yU${+U?1>ClVnyhCfnmt=T&8+8S}pPLu)}~V3=BCNu~D;az_sdjTt`-
z#CVS(9rp1Gd1M$+IdGo(J1c0_$t(l(xahw>)@T^T45OGNiiwla9=jM^J`G2c0C+NV
zDY=5Y=MXE+Kp7(A(n@0-Dai>*udz!y9OAi(YJZ2?V26qu;}SAxG>HTR6J<Mcbqr~z
zk4jU#at@UwmRF!`@4l2TBGZV_&*ORyz2=a9cR*0`&~ZYVB%C~T2G<Mm1D|LZA+X|u
zILg9Ml_E*u-ue#z{J;NKJp1&c{Pu7CI%gk#l$U?-BJaF&iE_DYyGTOfcpk#BKc9c{
zD;zp{jsyp#6GX;o9q(hNWCsu?z0X_B?xpLv<K!4?Hk{r^#~AIuEdV^>VT>TvF7;-I
zee(-6I}P4``we=7A#p}$NRyZWhCx3jmk&8`^a)Nq{4^^0<SKUOJvbJ7IDYXZbfoui
ztTm4FJ{(_rm+<i7eK>Z<ICe4|$HRo}^=+0{Hdy}gPw6Jh6siG!&ZlzVAPYyIW9R0Z
z40hH@dwo<prcw=1+AvwE@#rzhkFTup@87t}fzr?7cx7V?rO9yYv)<^k+ZnJ>&5v;$
zYmN7pz;QIjvHs{d9`x|?IYi)7-G83?yQ`=)!?9Kwj`h#rIErGD;a?4oZATItPvsio
zjXGPeyg~6RU%?3+`y9)XKNQ*IEK#Ix$n?2~8Q!>qcQqiUMW9ogfTWEzC02?c=QFoB
z&!K}09D4dQgt-Zu-D@Pxdw1Bp{wg!4o*+L{!*M;NC-CYW90zuqb$)tP@xp@#`JHFZ
zv(!4wZa*SXg4t?;M-T1e;Nm=vgt*(rBr%IqWhScy?$x^(4I!Dg;OL1gEGH7kl8MAv
zUuv=rtrVKj69gXkPG(;tY<h+=1_y~08B;fVEuH;YFJB@V4p32_-KEP+KeUMB`u0-i
zIM#GK<2B2yZpj5cVHgnPe0<-}o{b4YB&hojk)N|ZR#Ep4|KUG;&#!;|*MI$IzUDW7
z^EW^8>)B{3z!+juazVh%W9OM(T4M8!%LpUU2vT7PvUf~43RfEjTV2-Qew!nYJdG+=
z#^Vh+lA9yBabt-=cgS{qoja>b<O{j1pqt?<U-=TRyz&Z-R+A_(=uBi$ulITDof{lF
za-3qR!oK~B2*=_2_3LbIY_hxCKt&OL7*ecO$Ok@waM>O7`Qe}bIqN$^o`3c!jvPK<
zJB?Z4I<hdy#7khC(+J1G$&w~8*$B^AB@Eo}BW6b&#*rAK5MorES--!0nB#%7XL<hF
z=LvEJX6I)4SG`?Umv50KR<>=W@0stNGNe&V8b!8Cmu5cz>0qx0pMs@{jYhErU2GJ2
z&m%l@gx!@DnwPKHzf7lSVZUD^05`$WhPctCe)THJgO3nZ!9n9n7srzXj>I*B>sOa(
zH(PWX4VtUhDAy(so}zYSfwQ0e4Q{@1k$!!dc*mokc2Gi7*&T58>J6ScbCl`oBo7~r
zx#-k+<N7-F%^mv9CTSGogdt(AiXVh{(xo@(@uP42CpNab{OWU$^VsnNR!JqKy@6$i
zf6B=HTaaK!RzaqAHo!28ZyVY<J-*{u0<8flU6d;`nS2&HW~|SlANtIlI>yN-{yY4z
z!2H5I-~QLjv^OtW2l}xMjmZiHO}{sw-y4u5R&}Pb8?)m9?)`=ffBE&;zIhHxO(C@f
zYClO)#-*>JuN70iM-EBU?ULJVFg>g@PmELW7|J3#iXe#?^g5(bMDf6ab*fhiGmLRl
zNCBNBBq|||BXkmxlagMyM`{G+YJqOl#M{+$IvsYJLmqhGK@J=^iqa`JuipY~-;+eC
z@mMX*47qzf!*{>`GMBI3;y?Qvzs^HXJk7@HU83QDRHaC5FN)F$IDYOK4xf0)?u$_f
z+-zJrik7l_KcaP*?3mS~PO7wLl`9=oHijQ%;EwiNYmFA*2nW~kkzS6wo4dUBy`S*K
z&ppS&;sk4J_qcKWI$kD-(Mk|y7D|48n#Z2|P4cB8TWi~&c#l62$7eGfU%9@AWBI{2
zZd|?gb8&p*+8x@>CY?ru-L<QfYBi*%sUF(L126n_ZoGMke&b$-<F>_dW5Bf=w|MgG
zF{Z0iJbWzSqO-#**Vbul?J#JyERKVKaH2NGaevt5NB{AgZ0_{<>T^%<@X<wUV<RN;
zFM;DI>+JcCJ0?s!zYxbqIq~>^hZp9VU);xE{Ofykw%$Rf`U7!1=nWY32E_N{Sb*b_
zr6YU_<5z8)3PwVROD~Styl{=lLx-7o;4J7AVQdE-vcFRKC?N?ZD@;A`0RHPgC2ZLA
z=0qD*qKQ&XA`EBF9j7`uL1k)+iG||~yB$!1-IZnD`m_I?)6YCkzBYrO-%|+}Ci9#-
zbCe%mxWd-%7O!0mcr=~mxd#p~H9t$CR>KVg)L_71r%pT^5bK0OuE6KN_Eq*x7x`bm
z`$Jy2x@0@5#&%D&mR9^AvQDZoco=k=5@qVuY$$>w?av59q|%X|38fR9)bb1snPUf_
zjUpPvtX;aw<sZF4-0!h+Z<(+Dm-AHSP9R;|sTE{F34rSg^2Hp5Vu8T-5P^^5`}UHM
zfJsN}zhAsQQBn8lEQ3mICrL^P<swI(e}eYiRfbzF?*B%t6e9BwQij&eWrocrxw(D#
zBD)z2fk_qpVVBF7F0gO*2)#}hS3;>+WNu-a-~HXc$6x%#w^?7`WH?O6b9UdA3}emz
z@c;S`+`YTOQ%^j~%<LqtpXd0whdF#Qq1oJJYkkFL@1-emG9ZZscUO0K=?Aa!?zJVh
zHh1{kbD!m*hfZXwq;yOGy??YKGpYQj-=>t6w2!!SMk|EWS<-}+KC4t(za4X5!6m>C
z0;<(2XV0F+^Yc_DrugD7{{~<G`u|LCIJCDWrR|(QBlJd5L>$LswN;v?*1Xq%QHCT<
zK84-2-*=)^=8-26#emtz&e31nAZ@hvn6ZE`4iXn+f;5`Jy;XYc7R7`638a(pT<i_8
z)$Q@l+n1P`nIUQSa2!D}RirdO#pizKKj*FQ{g`fjjd&+PwFRCTQd*oNRUZHGfBAD7
zTXmj#>|v&-Cz&YadE)#L+QS2EH#)3sHo*0=?A!<y37RYG-1*jjV(Xnd>~7Xs{p>S*
z@v&35o^>G~i5QhJ*5zH;j!dfcI2STfcC4N#ZQa$=6yexVZ0SfMZ5YPV@>0y+r(Ym&
zeTRHC&!J;S34$UEi}Rd#{x^90n?vF_xldSQy9WIzqTL(N9}Gy9B2H4`*!nRg+F;U;
zr&j;4_1nMlYYe~t7odi8;*?kmQUg5|5jqZe=_7>UP~PRh#$5s@Lbw8Dd<sDs0!7ld
zO2w$vBHu0JhB?wSCXHj#G(i|hhC}N;2*i%hWJPh~<{gZmGBr0vr!yq<Tso~5Z(p@z
z>4zVBg6-`!0^g(88`6(r>-Q+_11#0{K-L-GW_e|c|Hps-ulU?&9_I_sKS3IG8AeS+
zB1m*fJ9Y`n2k?UclcppJjImDfV>Nm<c9qh4P8mEf5+PFSkvAIOWL}0+YaD4m6Xrg(
zyU8*jy|Bo{+<s1;Jx3u|;O)2HL~@eL@4m~R*JUD<2w_N4O&lwNAY^X;gPc736e=||
zn>(L)k3Sg4p~dm~=Fi2kFq;0o75eS=hv2x+r3;ssnVGgY7J^{9L}7M<7yi!Q=j|7N
zN^fV4csoIL1fCjF+CNLGJpSYl|BS|Foo60@gsGVcW~wEgexyKac!;ffoAr%4cq1Ig
zs94Zk*<|UP-(c&~Z5o>mR-S*FUwZ5mjw?S9$IdUnaWclS#&;d0BZ)N(V(~LLW`yHf
zfx{ylFU)iI`LFTzH?#Bj{Wy*z+Wi5&!H`s?7RO0q1LqYPjz77>w8(x~E~z#6H|+uI
zKYoM4;y&ET+F0lTV1z=EAhd^;g2LfL6e?BP@ohR<?LF^2hg!M9BTqiY#`-#bxrEXX
zHCuEx>U{Hm`j<TZD_@{EwSZsraa?QX5f6sAzTiLmFa8FT^V7We<5$?)Xt3UB^3qQ`
zRP#5OpRTgFFoQ^9d*-?UPGFc=oa5LFUuORF2`1;JIr!c0@uxqzgh~_w2W>`HsUl0U
zu)dnE6u4NCTj*K~6hBkDI-Z9~9ir5xGX+wR4-8J$P`dA(A&FAD^$yoBUm+T{xp!}c
zqu>4yJo>f2V^!YLa!5zK@L~{BEfy#h^0=;p@A>$?XUFl59lxjMqcvIn4;4y`#*s?x
zd9?`y;l5e+J^L`7|MUYy%9v9oj19W87a~JkZ_`>|W8(OkJ(~_-W#uk~&@dS6V9X+c
z>v8<ZA^d{F-~Ic4k0chXuhnVpc2UY6z`o-Ux-O|n*=%+A`k#KE|M2H8;`<I^=#vXW
zDwPTo6BEo#R+*|*>34c8t*&!pX_KXu75a(d?(!Y}=<ENMW~)xC^;w>M_6dT}#(jy5
z`!*`lWEK-;i!=~vf=c3Xj29W#O&epQP=v$8^c;Q=V$}WqclNbLhNZp-rSb$1J@yRs
z<}QuK4*j9ElTgZbK&+zH7LQ4i5XbR&6r@vyLJ~)x{DE(GnXxsCHCz?=(-mruoMZUX
zPa(?4r49%qZ32g9ue@oyLvw43eVLc#NTGCV=`Lk2WjJVI^c0TgGjZk|ZaC!m-}xPM
z>eJk8FxYOO`XRW2-1IDdZW3*Jw08%5|GO{q(u*(ScrJbr5QLI^F;AhGBPdPbm#2w(
z3614->bKX~S-wc5puhAcZ~x&g{q`=6W{bc6!eiueuI1v0EVd_eD0hVbT#`(^h*HFH
zid6RIWsI>Sa&0YSA{>rSRwy|3;29TnLS#u5hBQ@#J}gwqoPYdr+N}<~-7b|l9v1{j
zsz{WL;z^ZaFpL-sV}!7`)YPP;B=kRpqV9pmpW@=nKcUy`5i5fd4ujMX8N-wl;5ven
zr>Z>j(0&SDWCaCE;zeK*+Z}fO9HrVU?cE)s{(wU0BT%R`!o-Oc7o>@e_LGwSV8H%j
zz-!$uD;pbV?U1j|GCwmxYkP|$hZgaDmpgaYSzlYB+vzim6O<ao_+(Lg*lvtAd*R|r
zIQ+-I_$k+}-{s8l158a%P|W*m?QHYvg;ff*(;PW`fI{el5LvfRja8HOGn!E*u+$ML
zB}Qkaq(YAE6ST1-NMj5lwHzL$6-uONG>+@z=gaK2dnf~?LXJwM#N9jhxO)9ECW*;U
zXk;MJ+S+BLggp6~uac`&>2&JUcQ!xq9)Bo~XR6d5KFjc>mmvZ^0LLQ3@$S|p^Xe?p
z74N}u#aBeb7Di7ZJ&)R%vtt|^70}#hFxbg(ECu203_)%x!|{;s{>4xD{`Y=@>$!MA
zKo~mYOL>Z=JV9w1zdX&bm(W;Qr+#aVoqHD;rZBku1{ePAF8y|cMzh7&UU-Z!bV#+q
zWTg3zhhxjBlTtFo;n>9F&)`^o0FL)nEsi^_4!vfNN)nGp*-5IbGG1AaxA)*!j&K~)
z9}YhG7H9lVOVl56z6{ZNgZe9Pk^9o;?3dYjBStEO<Dn9RLE_aaOwR9P#eEZ%n#`~n
zPM$c*ZtN1~Y6!VSn#R;`T_qhz{^DQ$YlIUqv2TjpL=EX%BmE>A5)TJBBA`;u@z;Ls
zOZ@V$evvdviH8w`K}5IRqT6aRY&D2l+vEy9lMDNpIl7<811AW4kMQ6r9{*c^pF(Yd
zsfh{x@Xvole~{v5ace5oBzn({z!>nc>Q%Hu;CpD-A(JHHyYASQCM4zB&K5nTtO#;M
zQONLNV-Q`2F)Z)4c>Tveq*AL;oG9T3)`Z)U63=z50K;)`9EV((qmT<R*wM8sTs(YY
z_0i~UAFHT);e{7|A$eX1!FRv=-H-Ny##nRUQIK#J9E{^}@Zs|`uH9zv&RsCV-VPDO
zLQ)a}SGcI5rg8ly`=9&_!g270ol}=96>eR5lim6@ov9XuLXks<7dd(MF#8u4dHd~m
zSzB2}r;1X+&x8yH6gGK58$6ldI+7?(>2@O!9adI%h|_CCDibYax3v2~3s7%0_{KN?
zBT14HCoy0A;<NaEh)NY?UPIDxvay{5S%|mL1}POnrbx%27<pNNGLng@eUz&c)_~G}
z>c-<9$B~u`hvj4huFtU}CsCh``RU6)VtsRqO1WgSWR0>97ANZ}$c#gfrE7rwev;^u
z9~34VMF_A05@7*g9EZ76N9f(YN8|dv%sRk^9;Z0u07rP3p{9BJF45zCgzHik)<38=
zQRdwXZ_?SV6Hm|M<_c8yFS76CJd2C-yz|y|y35xg9#S}1Km;y=A|!^it4P`!*H4k2
zLmDT<{gf_H7hP05B#qxi$6eCa4&A;-oO<+=uH_n<n5ex*@|{0LC6*8MJ70N<LKqTd
zYDP~A+YQRZD;R+i;MzBYSV~;ic88MGFf@=V!Li9H=5wJPd<%OhYOrIWjHGO^mIoE-
z@Y$0`QE|d6FTX)^r$@dLU=%20L)0Bd*b9%eA{(U)THm+FgO6qu{_|E?svvWDqST5>
z66MlWDWT(%a|Nf4Re0>gBwnvY(g&16I3a1^BJzQ~0Y~;SQ>jjq`VLwNh`~gPWDwz&
z3-tQ~@<oqm5F!1Lq#H9`%p-PS*y~WJ&Qi=*=rrpb+;;%aHS~H7!hC^#w8u#qiCC<|
zy^X43BwR@hMq1kQ+c)lU@y0!T$0G<`JDN!hufFyw=T4vI;fGHnP)GyH`do|!5{}6D
zz<Mt^Mr3ZZ_CjYF(l%)(YjTaw1|>n+PED>*p;VhBimeA>shH!?;yjn$eT&6~1@6A{
z4t@Z}I1C2~S_>8roaXfTr%828d$&Qo@hNobJ`~52xzk5894}kJ!-wMd9?=sc99tD(
ztybpUi*M1Y?=U+xkC!V_-M`5E$$9qg-^V)_ZqQl2h8gxL9LOU=?>-!-ByEKo?BO`d
zaLh(0!|^3_+(otObb2n4av3Cj3=YyH4BKm@-~LlnY)DeY?|kJ63PDJuY)XSC`EVS&
zk~o#Pj>U1DW+@R$aCCBtefiu6;@Ena?cum89X@yR7-^jF>Ilc>UyNg85-aGykbVj{
zmI9O>`AtY9junt-#p=bYl#U<4J9F3y8?}SR;2K2?C=($=hl!I%34d6?q$$lPCDDq;
zPK)!8KSgzFmi5gQ(m|WuyZ2amcZEAk%RKz#S;`BC5uQhXXMjoV<p^KkhAyt>pi&<d
zD_q|tD29}bVTLK3Ai(nl!gFwZTO3JYu;yVJgrroR;oL9(Ca&+`c`pC*k6)rQ7!o=H
zj372vs35U8jWZr(q6DN0-!&wu?Y#SrwX*SC&GqdqR(t)7pEt^Uuv|;qY1c?x9O=<e
zDc`zu1=sia%x9lrVx~l|86g~OEu*b~%vg`+Ec#7Y5mENK!l)14xc9A(Rn$H7&_f^k
z-#VQRfB1)gNc1s+1}%Tf-U`S%b<$uhH%cWAedclM%R4x&AvUwsDwCveaIid*^>?pw
zX177UR6@8p@O)gaM`O1`u2AI0^~=mm9b|59fd?KqPoYre*3Da-Jb8kZm1VZKw@EUg
zfW$_7=rkox)tJc4XjEVnNvcVW7@PW<EC%WR(Srfn-8SF;-k+1p`2>N_7e4=4E7BlE
z%1qiWBwE-T!07NFSsYla02d=YVr3{-YD`Q{GnT<-Yc%F@vQS{9Qj#<wi4%&Y5+{zI
zCQTGCzxqSAw(1n~9!jP5{m!@n<1AR6j#aQqDdJBdG}3^*fe68vYHX|$(G&uvAANxC
zMjg?Kuo6{`)?&<4656nH^A^c&gUP8X()SUbOVJPLwubomJnOe_;8qqWPVM9D17|r<
znc^?*{E*twhv_v}8MM~XkwOh)dx?-fI_i@S+Bl}TXR<7$4W5qLC{x6c#}qtrg<0e<
zp|jqy+#<&%L9@HL%&qTy9j_1&cpiV}%TE)!j_t4rEYV$vEKAq!u`<>-4os$~@{x{D
zH`Oeb%N#Bi5MpnSM<WqO?&V{qN)g8~gCu1x%<<W?CrDDo>#x2^d#g{FcdSBggsIFM
z$#WfaRvc(-hdhqsPrS#e7vhx5b{u3J`r42h&9rctDI1P2dN}dGb|a1LR*??TiNJ9k
zyj*CzK;1rRoFLph#?dIOa$I$Lc)pJw#E85g9VWc{?gl|oFnOR%b2p+>-@z}}=yV%s
zZRoTbtZr`b=7l%8e*M;7c4zi^bfrLI3%Wh^xfN}hOu;Cm?XVBhgh4m4=aSL9{?;2j
z{>W3DIDUjMXZ_!$FqyBT*&|zLH}*X%g*{;eVsx_?d%RnnDdmMND4Z~)P@6_N9=Xs#
zr7>|7v9`8CtJxxI4{@Zzm5{`SG%?_YJofZg$X9A~JG*RcZ88{q@^0*h<5*-kevs})
z1JRAJt-7DZasAd!i{nX)V?38)7|?AE2nu=Dm#*WM7bs5e<J^N8j+cH&?dU`FcGnoR
zH_%av9>xgaAY_1!1{TNq18^L-QQG2I7g}Q+ch}npp)ih1Vib+dd))la*KrF0q37^B
zUwY=F;W)E>avXureK;=td>lK{dS<)_$E6%EoI6RHDqer}E!x|C!kqgvIF@n`$4SP4
zPL(416mTrCVn=1Nv69KunZ_Wx12%s8I;DkabSVT<gqApt!6a4?VWeUD)M>n_X|}d^
zQBed*N^3ZvT+0*nx-1+&P10JXv$@P$ufN4obCX{^{w3<SE>fI6Oujmg6L`p6fKAv)
zk*<dmh6ovB)Se(oNQo2rIKqlX92`uVV61a6AjcxZLUoE0U-%kHf5<RS`Iq1L9)m<7
zkticFAxvfzE;6%WA@DO(=TvJ%5~0(Gq9?i6Y;&{I8C%<oFk;VDI}rnLT$dmS$onp9
z?E&9>>k_W(@XT{hQ=QE-=qo(Wvw}ilmCRac;v^*(c({)Bt#n-{>k9wk^|3TF|KeZ#
zi=X-N&;R*9=g^@;eB&G6U}0f_Klp<`_&6_#BdzbvXyj;&vF%OiQ8{#osWS&qPS&o%
zfRUgxr*9$XudK6m{W>xmZHC0$yYOQ+m#*VDf_L718_#nI{WAStOl!A?@8y`7s4+EN
zBULfcFeb?YlO4ywkV(|FEh3c`*{`wVB|YkpiScHPzaPh;SS%uhq}guqgCD=h+n3(q
zwbx#MPw_QQh%op2(e0TO3n7uxLko{wp~Cdcf-N}p{kQ);#+4nTrX)#1nxqT{eYDni
zZjKYj&+yRs$1z&aXf&+iRqOkR-FyG8Q9*aV#lk09#(<w0s*WFU7#tx%xD*zqDIHzJ
zxG6>%5C)?K7>N}2#ujaDuzu?nuCSkt2~xJ+ewE#Ow-658yLy4pb8-DVjn04@8%-S7
zBVU~)EY{F*ous>t9`+HAK==WA*dgupP>DgOc66l^gH8>q-$x~Fv>qazDI6ylkLp~{
zqg<P$FfnC|hs|Z~z4;=qzk8KG|LH}f0WafyDa`n=P{y|S6P*#Jv%}qWeUu28$d@@;
zuHXnUb~!iM?LMQ5+8cVJNt2W)j;Z&Cqy`GU&u7k^;_QQGKtXS(Yk4=R%4~6Lk>X{l
zbv@o=^Qr8y@C1ITj4=+0b{Hy6AYJBX@;vkWITj8slAo>NRPz|mlI|UJ_TY`sal%N!
z9S<XtejlUk*w~Q*6~`pQh*G7JsVd>ju^K1m%2axN_D@f<Z(@q>#uj()txzskxVLnN
z0|yqFnVKZ<TzfubXT9%9T+g{LL|%^hOBTzcg6q9edB$iou(`d?rFY-r?%idK9u<Wm
zE9}TFOlJINZ4qG5W^|VqAINzYSy5=v;5d{kQ-rySof8>DG#oG-^r<&?XzbRxbom_u
zSK~Mk4^y<3OwJ$X%tOzT>V#&a!A@g~Qn~nv_xR`G*a(-x;xxsh3qKdf!_5sgZrvW^
zc#@Ruw_l@q?>52_+`IC2hT{T_&XAj%EgZ)sSD7R%*U-@pNp}PFejIoEAB^K(AC+`I
z1jqh+aNOOv$GtaS<h6IN@y%B*B6&ZKc@K`$_u<%ay)lkYR4TYX7suv(IF9ba@oCOH
zcp8&HZ>M9YJ!%if)}L+<$LXhuV<Cvrbo|*J(Ryq|T}t=vI_+zBk*SNwx?HvkqpZ5y
zXmaxlESx$;nu$oZjmzBQ&rv&k0-eI)vyU=}Qj)04R5{1VeKoX7>9^_(T01yF5wBdr
z$$1#1?AahCuJ72fy$jOOIGz=IxxRzUROwo!XqArN3u6o_i5a#UpcTd1G>4!163;&T
z0Kf74V`w2rGETcA#JG5OumzpnfBStXZAj7>X(H;~HkWrBq*-&q$=Gxwp@hMj47;9#
z?|Fnlz;rQ(C!ib0eD|%3y#2-ngeNG@mhnB;#*;bDm}j44P9mcs-jVM;i+<kvL_+94
z_=7*-<jIr#vw!x__(%WfA93Z%6@L4-fBPf<d6h<4$hnp5+p|&PIUa%Q;pa;%K6DOm
zB5!?MvRh(`Jt!O@@CPv)SFS>8@Px$OyhZZn_jvRB-)DP$op!Iq=GGcQyR;f@8qGRS
zKl?nt@$0`uwK_#@qK4~BT+ikXrrK7+%82pXGg3Q=j4xw<&w{{Y|9!Ny)~Gb0RxTlh
zWNl-ex8He-H{ZI%?WKF!fo|phMwzj{m^SwNLwZQZL%M<W>~UNsCZ{mM(Cu_cdOeIv
z$HZklR?b;2LYmr^ey`U=Wl0pS7jWv-13Y;CF@*FOMlors(Bt@<tYdKB_h8=3?)^)z
zI8AU|iIeTI{kyE+Pw2S#K}hY)37krP&;CUi6pkG^Is$i?uyx}WB2g3sgsZn%dF2Nz
zeecicY;MsT42T+A2;;C@Z?m?u$*CtE=8M1nWy+Na!pZ?;Adz7n;fJJ2Oyq~8i3h^9
zgkEJ`(`^WK-j2kR7$E`xo8`;s(xLFxkcrtvge=isy2j>>cX;v5E4+1c*&Ytqy-~*c
zKWSq~jZKadu7m5kxNeAWeT3_9yjnyE!Ok#d5T~}_#`3)sHZF*acx?ALVW-n1Hk!m3
z!ocU5^AB+R!BYrF(Cfzx;)FD{q-)HmS77(sj2;9Z?VR|zYdGi;)TW5B=@Hr(P9B)z
zYrpye=Z>GFHnX4dzD0tG8t!xjQwm5?RvoT2>A<EhXy1WiKu{^*cn&62xY8v}5>lPw
zhdxM&m&-9ZRU=gib234+n@mj{AntX!vvi%EdYxXY%hu`|iw6!;E){TveMq~G+*6bb
zF@DJF?EN0m8Ac!f`>{QG&?&9CcH;&&ZrmDQ#71B-#u}R1ec!W}(V1tN9)(+HUu%E=
zzDrv6<oCQBrSf#f$Fl~donDva)q8C1Y;x_^4K}wnDTLB0*%L6X$C-zpp*Ax|uiIvO
zdxJQRn4O*d#C!Zva6ED51hSGtryq{v&FgzO4%U`fe&q+;{@#D0v$;uk&?nv<;kd)<
z_9iEvc!Xd2wJ%dDRS7HmkpUTw1JWcW_CwOd1>t-Mj^%rCoQc22I5tBjW)?BBM0@Ef
zn>XI!d#_*S?VBs_!?Ae}j@A2cY|m?pV<`mnC}#MPalD)181kXdvkyJM@dr<W6m$nM
zgXq0De(zcR>ESqV9n!=m56Mw%ozX-pVW1MC;eegD-$6BdNY~Bi3j#+dTTClWlJnVj
z<`l+r=xIZA-~@-B{~{;8@GDg2=O`7+1e0?hQzmDs9GqL=oge;$KmDWsop!Uqpt*q_
zD)b;iM=_4;B8-jB5-2i@07hP}CR3w|`^1AIObb9r7vTo<nhlIjsVp93{`ABA@-q+d
z_{oDP_Pk>qX^T6Vb?Kxr7?q-um?R#c;|S7}tIamrz*OiH23|&|upK)`WL>@NtZ+RS
zNBYcECh(-h1NwuA@4xy6SKhb?%1|nV<O2`awUj8t{dQ+&&(rQBuD?`K_sws9lZPI9
zh~NMH-{*IK_jkE*;|9O~`@cWFL44Guw8nJ;JU6r*xU5MnrES@JK}c?ThM7}GFkV(*
zp-Hqg<dhgZZD=noL9a)#dxz5PH_10vdF=QB)>rOP3Uge1>s34{*<RhE({55I<eA&I
zpWpr6{|oc;2RV8A6t&4pMg-QG8qTIb*mrhsi~^&>J^Loi`)^5B7M%+NrY0xQN^|Af
z6}EQj{P2fArPu9S6Tyr#qtlV%4G<1EF4}Q1PJqJ2@pAZnjwDW5x_*<TE7urwTBtN3
zO%s%|2Buo8%&9&_>x4UZZ*ceST{gEiiIbQh$Z_WEd5#`GZEYQrBs(OmNI@wz?%27m
zJMLn9(iJC$zzYez5G!v1u$)LoN<u#%n5a-bbQl8rz%yu+wqGqVxZ2QNUBPMhnQPu-
z_R2*n?Ntt)I6-Iq7D1Tf?v)F8LeSdi(eHLC7mG~H&GUEu?%$?7eT0c4&k;^q6V^n<
z3}s3h>#Un#4=|(cAtoFL>4pf;J_xkd4EkOBnN4lxqErJ4l^m4=3z)=EzkZ3%Zk<1U
z=`D7<gN#!_<}hzbxEf<e?v6w1IvCd@!X<EgijK=rYhJo`kMCW&OQRo?7{gF&hT1U9
z3`eCSGV{@t%eQXuqid_Yz1pH3rIhj^k3IY#v&Z%`Oi~8JI4kz;9;ewJyN<-k3@|^9
z!W~DDo0}qW90msFDn4I$e363a5O;Uc!w%bbR=IlNDqiSQn4iEY2H*)IosbTP#8Dq7
z=TMlb;N*Scc!*9DTZ=l9RK*w+N+)(4<A<agDq_I%2g*31j~jR-{TA0QU!q#9a^=m-
z<fKcX5K=1S2m@~%qMas5rn)zzI^)5Zy&}%q9=x|(?Rg$y7-EcJZGD-$EBEO4d${iC
z>|*cm6<JWXwjC|V3dZa#Q6`J-(~2ZXGCW#eGHE2GLJil;qg4tR;$%pDXPcFkd+ao}
zxpeUz9IYtiB+<~2;!vC0&#4C<K|0Xb*=BoZlZlB5O6Afg-eQa6fZzjg?2d3eS*3F5
z2>9YdaIERBF5`3t%(YgSyL^#ybA?04PS9DujUVK=bLDM3A!%=R>GwKR%0;I4?c;a;
z?%$>~eT0d_&k;^e0fI!u48}N?9}UO7_u)7lP^jjp?%#)r4LjE^(rGsM#!GL}><<6Z
zIR5^XyL{*3eK<D8D%(@_vpBwegCAdC<I-A_cAQeq=Xm_#2bnvzpJALb7$zTrW5<r=
zJ|!IczC)A@86<sdrDUfJsZL3>W)KY+ZZ~M&ypD);_IFBa+hgn#IhB&q!NaI(fwhT4
zte=0J*2I2p-MmG4;V{v#$I<go5NShoUx`Ag!fvz8=JGxM<^S`4q_ugAog1$+*xt6Z
z2FJD0WDW>v3%Zft$OvmA#W-M-r0%=4fk3zcQGY;ZdlTt8%$$6X^29X1_2s9TDCX_7
zP40E)93e9fn?@>ys}wR#2t~>;jp?K*p(FXq14sFlM~^UB2=Ig?Ck3Gp<gz`{%Gd-M
z>2m($Sw4SyKacFMQ3+g{-9A5l`8C$AE|Cjd3b`QT#oI2Q=Q>&8uf}H6A9a1KC-<wb
zzRC+PyujDK_BCF3;RSBpy2U^E2mgRVq405DX1tt1jO%!IER+$8?>FysT?Z`%Gw07z
z+&_&*Wusbq%Xft3CB>V|MAu%yee+w0?p;nSPP3S7V|E%m^VlPJUdq+WZ<7oRt<7DU
zJ6mjQtWz#e@w<QT_c(U+43%m*E09KW>%A8AcpQ^;3h(F2nD?W}*js<0kYj3kf{Dp0
z{eFk*H{JjbLGZqz)ok<5r7M}C;C+0X(bzTP*%@Jd%d`;;`a@c+Hqjtr-@biJ?VBSA
z0=70cSYEzsy`zM|_kE-z$3llrr$ZDC>G%7@u_dABas{R)=TXXx3&N3=$!O#X?5SWs
z#s}%cme-5Oxt<MFX4C<(H%muIJjcaILHW#a!nr9lsr@V{g+wESMmU1FzCv{QHF7`w
zA#S(Mqq7T4CR>P&+Z;Ui075FZZ@-Hg7<xMcwzk%|v$~2b)%eP9{T(LvpCl+&QE^1t
z*kai4VUoRl9dW2EfjL3z9!B@jIzj7{G<#ni*JU_}Y=&r|M1El(`NcVsc!$Q_Yiu{$
z{OGNl_GXV6=f5hQwfQC3JEufv_A#w)pVe-kc9ihQp?OXno~7u!T;6W*+VVP1=I7|T
zE<%d2$#=Wcp)-hRwEJ|Ul*AaS#UhiFGuGZBiS0N(O^M>faz{sdoNb0SpZsay_yttX
zLraI*D*UxCpJM-k5^gym(Oaxudxy(!yv0Pd2pFP%A2)EpcL>UPVkgB2jh)EbhjYK#
zA{h)xjWyOKNm0WgDvmHC!*U5KQ9M1DC+O@TojfW@*xX#_&b>Pfx;<L;I*&Z^1V>Mt
zAfFF#U678jpF1*RS!D^;Qb+_=iEiF^SrfuOJbd3nYeTQ!q1oDHePahlhBnJFd&rJ1
zMONi5k&eUI#z>5~K8%&gI#WWXDz&LKu1~I5%Q(FTrBgbc4(qEcZ0~Gx^X5&qHnu5;
z0;v^oBGInLsRtita`pgG6tTU%PLjmT%*<Gd%O_kG$3EdN!tv?j1aniE55%#&563Uz
zc6WGmZXdOD3$byNgXhix2|G*glJ*V#odMgMYb>p<;#a5m%5VKnhU41L$FaB{$Nl%=
z*mm6CgX4UO{Ne(+g;|n#huyo^*xqgNqYEDo$DQ7QjoyH6obuS=1x_ECrQo|<-EQ#4
z^14+-3zL05#Q0g$Zg*()25dKbw1*jvOC=_!MmUa0(u6cs@5Qkg_3S<!9LL3weiAdx
zGSOw$yfD~qbF5ONGBn@5%wTN`feXg+;x(DaoiYT|hX~F;NB?tQAt}%D^6DBp!3<_<
zf!i0}1nn|&{2|<epg2=tZr=fp9XLSL*x--<KmP;UOYbmjH;}%M8~WpCu+f={zydVu
zs!N39+MlJ9aV_^cTf$aJB>gVE%?<in+xYnjCXb$FZmP=XpEzwtuVPP`>`7}vtPQqH
zh(;QPBcM{sv9Fk;6uA84#yu|GSf?AMJTN=SQwJBbh&eI7q>U&q)oPVmG0)sYkxI_P
zlY&OO!`9Xsp6d|!A)e>rc`kwP;d|bAqA;4Kebg-WkJVax>Zzy5<#PPq@BJQ!4<F_a
z|L_k#@az8T|N6iFNUzsjxlMkeY^&L<Ej=pk><z$08i7-*a`fp(xK{rj^ixs{IUz{^
z9L<rbkbQ$C!q5RFsLf9D<n|WpyD&XFMSAWyKYZ~G@?nW6Ed!eT#0-D(Cx6ZhpZfy)
z=Js>t%7x6DAp6Iej9@OUF|gJn;8@;^EqvaiS$NOtbUI8<O>*ea0alk+SY5lzVE8DP
zE?wo!nPb-H3WXkVYy?>}jJ@>>2}%!n_wq#sgF)6!gQ@8mKJ%Gpxpn)7wRJeKpCAYb
z!yHMHP_OT>wY5Qin2^g)P%77G?Cy}JwqRXeUSYZxFi|OGQDnALFyd_3U#FSP$;V3w
z`JlCZ=O&BORl>lxdz@{u<#|a$>EIZFuI8CPf0mUW{se{!g~1mR#|VU*aI{w8Tz`{N
z;DUzzQ`5Y#HO2OfOL=ON^yD$_zIp*SD3K<UAT(hm;{W@tpK#{sC#lX(v%LiApg|n<
z(b@skM>rrnZ#2ucdmu_+1_)8W=mg9jOr$Z0M1U6(4|mChvs4bAqO);_&dw?-?(@^P
zZ}7xJhslQl3Qa2Za(_nvB_m*`X~N4F-{ptH%*H?&W~L|ko6kPP>(_74><xHqVU|+h
z69fUx-4@+;kIv36Nu1KDY$7UEu3oE?B#I=_?5u24EW@x=Kx@07QJiGily;9(m02Zx
zM3?fPx6&j=3xPK=!6T3C=kVb&j?i?QJ({=g(CZF3a&UoM=n;2%OzhuBx4w%bAgmT3
z@8e=f;t1dO5Jr*q29Ro`w#Bwmns_iEC>D{zL6c(A6e)eGzQdC<4x7;g&29(NZ?n97
zlljFXEZx4t7k}kfn4Fy9{P~Bta_tInGNj#Xv$DLNjlkY(4+QTWyN-U}cO8mhz@XPh
zX+^u!VP$2Fvu94)TedNov%kf#lnzLR&<d5>j+-(9GOC1)0Nmpe8)4l5^2Hj`&yyxL
zLM)CWwzoD}Ub(~8&KB=pzJ^hnVqTzPjlyAadX8hK9wo??Xg2HA>)TW+6~Zw5R6NXf
z?%rZyrb-xiKZ9c_B;E+ewLJ3=on`e0KZVHR*c;*4NjXvtIM>^t7`O-x2c~BD!qzrB
zGcJ{xX;P!P_u5;yeu*@h0AUEq5&!yIKjG{%Pf?ls5FAT^@ZXQ);`?yCSKPh_$5E49
zZieci({$FCXm77D8$ZZTFI?yG^M}cY{zt;`k1kx|#SF(r!_3Si|JgGS@!Iv9?DqOR
zvTx>RaNKR|lE#XDZ3nkf<>rkhabg8|^|c)eMe|-9CrM^+vWMgJQ^N7=u~W3)y~1D^
z*{pFyHeR)&k%3A{g+VrYti1Xr2fz4v2(8h7Y+EjvRC1#+;O){PoneY`U}<xcKbxCp
zrebJrY~hNl9C_wR(x{E+ZDX_`YK45}YhU2P_g`dpZG+k41~&+h8P{6~Y3WKyVlQCQ
zrmYy8Cx*!!9YzZC@v^s9{Cti6V3Wq(8&vikp)z}zHUCXM`@~s(^xAd$!yyhf8Z5JY
z&gO=8nx%}vb3H!)#N&K97uaT@BUs<q;@^D#6`nbJhRI@qS60?(4I_rpkO^;sQnf^B
zdJ^AvDDKRW?AAGbdJ^Av@jc1R{#iP0LD=kA(U|8E`rdmiPCVEC5Fbq^@$rhf0|yQu
zgy4_=_>VvIYlILVxv0DR%|E3wGlv)WxUL7-@r2eXj^hE=-C8Av=~E|}SzcxLm3L5?
z=sFh)&OUL9hn_t~KC~gdMu4jYei(43vdg>gUgpemPx96qSGm1(jePk$t4sHoRhr|+
z_R(tBIey{<OLvcQYiSu1rwEe;Pooizhe^|{s{sfXF^bv(ezxx|Fh&w33DwFZan$AL
z!6U3~tg*AZfy85NeVc;^4`erv)DDs}4v!Xl+?^l@dHji|34I@<G_6h-&&|=OH`v@<
zWpZMgco17lW+@o-2h;~$uHLwe=jOP3XN6OzA4CWz>+(pFBqm8*(j-OS=Q%YpPfb%p
z5-Z}^p!7$(8T`EU`itMCGBZ!;`?!w7h<}ut6kqxvBw8ADVwgC(Kxh3Z-FL5`g$)PK
z`I@KC9p_6ApP(8@94u!kCnRB*<7m0bwQILH{L}+%ymOV^)jLduMRu1WD*FwS2Pf(F
z8XP}%g3Y_jtlfTvwCGUI)$sEj5Glg-F>wOgpp`(#5NSNrFa|U@dwcAsDY?`^mmQQ!
z3CnY6)uTFplHTqHz4kU*g)FVqIelaiJG!?I1TCzou#zK19TamRf9<&^nQ$d0Rjl=f
zgwkWV)nRFUjcU0@CrWXJRf)&_gx+94{mx|^FJyOhoBhX6A#iX#&vq!1gd~x~vE^4P
zm4(V@d#scV?oLy~M^wrG+|}(iky7&5um3Wl+u_Y?%N(DGiB|3+9FN-k9G>qY(iqiE
z8MNBCQj!KPj_*<^7fFXPW*FPsc$yLo26)0nDUCxGC8Z6n>sbYFlHf>>LRdg2DbF6R
z^7_ghb|&_5XZ<F_euu`@%N*K&h>Nek$$$0V{@<8eILP<E|9$@5zxy}1uFGz-MQ_mB
zE9mYcI=_EO0o6){YPCYA+eIr)l*BZfyC|(}lFg{wo{>Ww#};r(CAQX3YV3t)3%k9d
zPGo8Ft{V~-s+q5)L1{&=+hu)qh0X0vZZF+oXM2~x6$FkURRRKsGY>w+#MC~}F<TpJ
zq-o0B+?>U9^2traU;qC1ERMb4Bj9+E&e{=rm#?CwLKp}=&68)3@RdhSQTrJjha4@#
z^=r2{{KPpn-?>74b&2U*fyV8K>H*Ef!3o-(Iwy{uWc%(4YfG=b7soPFO~(mnG92fS
z#zzkmZ2J1%<Y5oTE~$zM%Ci{Nr@HS{hT|=a3b?bf!>J<+_H!}9an_l8KaPL(`KOq4
zB|25Cb_axx$8xL7()ucua*ghLaGcT`^x0W@7sn0Qxwp>Z(No!3ZE>7NLlWP8FOIW|
z|6c`;pZVpl@{>W2K|CZ;DP&zIS9-)MvSaj&uDE^e4wd~^DV~22N87?{kVt-b;R=89
z^1IZB_9-n5^o?RL9B{2O$=P$Kx%q=P*!R!_OdmSV=AgsW$&lS8!Ora+D&-k2Uw((V
zBZrtidJ>^Q3q`6V#yF59h(LlP5RL{t#7GC}dI+I28kb!LFj6Ch!u0}%?JcBNq_J{^
z;^ZR5>KxJ;aQxs5m#*I(b8KA<QfDHFtf12dI{g7(|MrXQqzdV|94O@&sDy*nGRGGW
zvfJ9B9C)PKj!T20M=@Vu=J-RXBx3r=e(qkqh;`9OKx^DEhd)dS0uRS^aWaoU*c-8X
zp2Twn>LbqBKUPt9=gysv^X0#AHD0;M?U`>f^TN|e&sN9MShdk;bYZPF6mibu$O}($
zv)&-Ob&JY@3QvCKK^Bh8VO(VoNn?K&!cdx+;H-kXSMSl=Imbg!oMn5X#nqcPICcCe
z8+Yz9w|_qcUvl}~i`-jY#VFfiC2Pe(rczZI&raKsiO$HlV>^Ou39NO*lTST{=lYZj
z6*}#hnaN4=`2t482%}lPeT##Ob0{ZZ%-7S|C|8=XG~W`WQ=({qBLw--#rIq;Ub@6^
z&|@&j(e1X;#`^nhZEs<W;_S)u9653n&kK3zp{Gbv&2Tuh@k!eo<b91%djxIUov^o_
zQRf2qRCpTOm#*^W)QikNbr$K^I@UFU6f?HzkW!+QAqpH8o;=G+t4XqYlbN|PUw!yI
zrx)iiuFh^uHfK@-3zL&PuVLfPD*gH~79Tjq?q*DV`8HFB4$-<(r*fc%>uRoFy-aue
zF1{9&eSvg)xPBQBHiSLRJfAeAgNSsuj_D=H@+77t33GXpI6}C2Twk;Qk@Lvn96_N>
zGFT&Dp2W}PLD}?!TQ}}-^58Uy8`x31HYjPb++W*?6@cs57`$C=9m-3Ekdo{2tv9YR
z91a-xKHI%MX-1swZ8y*grVl;9)S)SSH{kgvpCw8RgTatiyUW(rCSY?@Q@v-+BSp5y
z+3$T?E9v!En4ZD)9EQCHTfHH5p(y4GlqaW1;+S~QA@4XO$&jsEw{egtp%`>~)DG;&
z@jPoQl@VIC(&S3HjDO@1CFxjcZZf)p@A-J4kMJc$VR-&vh3oI`66R}cHMa?+q_ep~
z-4Xn|fA#<7um7FD&)L(DaQfuieB(d<JKF6|W+9Sw-$z`r_jl>C6q|fLL^^_UsYt7B
zDG{B{kTgjN0w<$X*wG(Sf^m={E2yRIET$PhMvz+nv$1*>gpv4R3F+risXf%=IA&*O
zi+jsU?9{ipcI_rAHRK9ds)vBdnSC5R@i0N?(eL(X>})YRJ43ltx~~xUWNYiv72cYD
zk-1-lV{LJ~_~bd3TTRlnTTIQB_|=Ecb8i1U#x*~S<B2JrSFnC}g<<o!#qm~3eR+wg
zg9q7Ns#85!!}AQ+uU@9JeHUMu`*2+1gK(?{5oxo5?#9US1g7K=gnKx4JjLQe=WxpN
z1cfqjf1N^Q5--ewO2ELan|C>JaE8S7_i)UI<G86cz7$l7Im({Px8JzPa5%vC1Gf72
z<G7C2f|<h)j&c0iC!Zxs1;c1atJPy`YZI+I7RSbp>HjKmJoo5>JQ1~6ed$efYOgq{
zFnGvueydgnH<W}!#oZsg#i801-u@G8uDAI5E3a{3X$6^D-kTHxBdo{Y^7a<rabWT2
zA#xL!x%=j2CiewQA3DL-?VC&-o1t;LPLfE@Jopghxg!|ilB%pDtSt3PxQ?By2xBSe
zo{Qu8Na>)pf{b^Fw46?zYVN-FDkf<#>^AZ9`{_1!NCpFh>w^@WIDL$FZ?EI0QRbVQ
zWx)bs#AU?h`+CMOAtX_%*=_e}B`Lr5^yB1m1)^xkY@vWFtPZv?JB@S<8#i8O^VTMD
zJmlLy_zr<<$mMd>swL*<7m%`y<GHxf8BYe4()gaW5gOGZAGO8ue`ulv!y#9`^G#I2
zAv|$}SW67n;>5{#D8g|dPDutsO0)Yo{FUd?ufS7JAEi)q(W*=6mhDX@NkC~N;O6pp
z&JIs6=K0_L;@kKqPqV&!ov=`0b90-sCr;Z#B#t<+f1X_7mx%fuR+iTp4kKLG8UGtx
z&m~FHG0$uq>?KA;oycz7f+I%`bL#X7>RTPsShH_&j;NP%=G0lbogIWRY+ZPnq^b!|
zJw_VlQ8;7uoe<7_e1JVJPm(H72^-tn+`f60SPi*(v%$>t9FtSi7^7)6o9ygt<I21A
zJ8d3)>=AahZ}aMFuXE?l9pWTmIOy^#zw~9O6vv+uoyF3m*+XX(Ri-|oFV4?hNyLp;
zULwgscxaA+$Y|5TWD;&WUNfmCi4&@mH5Q+Jl)^jkn-3pjvJjxnkkI68c4ro)CNYHh
z0>0eh^V2!L{=@I%9y?BV`(2!1n(p=v2aX)4aj#Ai#}xNXaJuj;I_lF}ev5S2#ql%0
zEFq9VfEvbDiQFHeI=jfQf-JfSSCT7~=(bx3XM)LdCzv|-DE&qsooGrk)1-aH+@a&d
z-4>1!+<D_IIu*sl1LukI1ri)$ZMzfF+4DAX1ZZiU&3mcBq$zjS+FZGDlhnkl-P_>$
z;wm#U(_joyXP16`ht9Gg?H}ae1E;xh_Zn9(Txa{<Ix0y?hg}|g{s}}eH{N5VwIzFJ
zye%{KMf~Isg<O<U6%P4QnR*y<>d-8OoClt!x3z-MnyKko`t3I9Ai^=W>){$r(rhs-
zl?iG!gy&|8Hc6peKng(|r+9@tNw<R^y7oeqrsy~!SI85ol-<=`YLz1UDh?BVpJ6aV
zKk74#`iMc7QmfAOOK<Uw|L|{k@-tuL$e|<r>aYF{UVH6TmhRl5Pz>q!yL8$;w2~PX
z4FZXjP%f8|Qj^OUkP;>)s|*G`!oagOBmF*smj|-gP6r!yp7Hgx(l!=rWCx-Zf)u1P
zPLnWzkRD;NnyE|eCAHV<ac}t^YnyA_TV7#jr-2lPKxQH**X6{SN2pFtA%#PuzJoEE
z*_oM<koLnlA|GcZ5!YUMkvIprgR}HM49AHkiBqamH5NbfD1}S#Hy%0u0XVh-F^M58
z6bKqyd|^7r*I)WB-mw!Ij;F>ruHUJX#tFrJ6P(IFgNk~z@4ZDD^>F+e(z6ff`*E!L
zL(+C5!?A~O1^Hr$UZ;g{YD_(Ff|>J=)2nAVo}EVZ74wHq5ce#Om)>}bZdFk`_YhGo
zPmC>GKMcodYEvMLO1Zn<;oTcINp#HGy>)KvUt@ZD3NS?7UHbI~U13negB&<}lIwS`
zbM?YCw(qT5+o=!1vHmN<F<y>ypZ#TmKJ;&`k#{sAwG)$cWOJo8coN_DXl(6p_q#u)
z`-tIxe)(OtyAh7aF`y69#$QRtPAB@^4w0JU&o;XJy)S*4_}W!Ua}S_~bxO0dEM33M
z^wAkS->0*@&86OpJQS9x9z2GV&!HomEMv4m4<nq=w^_`ti<GYA<1013Fu2kO6SH&k
zI-BplL20Ih@FlJF4cvlc<>osC#TqG^<Ig-zx#3WM>s5kYmq4h@;v`F*$e#Dg_FQnY
zqzK`<99^vP(3vy%(r5pX{hV3a&&K8^2tin$q%b{6Ss9$%ZI<4;%;}@&c;Jb1%pKj2
z>v_1jGOvI82bgw`EQNM#HtmgUPBPbxu@B=fSRbpX`;}k$75?at{)kGY0>D50r~j1S
z`@P?zTCD<5uh)6*x#zfi`SM44y+ouqySv=__P1DYzleMA2%clCA5GS&!<C>@;-rrh
zRgUeiGFx~MC+%a54YMCMTlm!?X_Vp_%R!MwQ!H0OwfV|I!1s4In3|p9&aFGdgM^DB
z<)P!JSYO{{Yipa651i%6XPzT@>AQ4WJ(48B2t$zbnVOoxV7PmCm83JY9YqPkF$isM
zCE6xG({4Apbn#sl=Z_HtifSdtZqjAn!Xm4yia|SJci3j|<`3wu+#rAAae_l<NqrA#
zvhP`r3`#TkyHXgHpwom~x3ALd)M<A+R4Q|P=f!VQtxh3@pg)Lt<+TfFt+{^V7LPsp
zFsPK!lg!Re5hpR-j!%~0gVt6Co~9X5JF)HSQQX+4QPicRINe?D{`5yoy3Zl^&5%gz
z=QZ->F^&XdV{km7d1}7IgQW-YbY#h@Ncz1Vg<_sG)i}n`jUybbnJw2aNt3Tk7x+$N
zgYtBl`n|iT;ecgQjvqP3#^x^F#xDC#ALYQqkFt~wh<7(Jj*mg%_#R<-5saaE_cFs6
z+Ch%uM2wefL0lqqLuPU_NG*udE?YOQQJgx8<7@m<4wH14p4-Rz8VuVByGf7s8*i|&
ze2?i<=cpb#%^>t}jdR}(7W>py*v@H^k|Z&gu3w|m>oV;2$(1U+^x}(DDi!)bGU%}M
zy+5Hf&@AlV$I)~D0MZ!gOA2$7Hb$qN!$?R4sQdOfQ7TJ5U`*84A2Cnx=dCG6;4~W~
z3^{b<04ALzYHcuV?qJf0+T;XI5Kt}@$X9E`oept#NHS=mlZ0dt;RHjRzz4?`RgUA5
z4u=>tMu3STWHE<PRyd%O7&q|I32bjQxpixsnf*16RU9tgUFFce1#Yb@u@c2pS}i7>
zfVW?KneEL6r_Mdd!oK}H^<O>DrAu$|@+&_gNqcOrZ&I)CG8n`NBM8HgXP<qX<0p=D
z{pu|WVV<?s+qhmxA<Wqn6s1YiB(o-w*7y`K(u^~eRhrmB)F^~d7_8NW)>dsUq=P{a
z6v}u(9;IwnbR5TQZ*Otu?k!sFCbw=bp_IZG)}F-(L3L(^qo>Z{`aV%Kq~6$Oc6Nq*
zKF`Q_^<%gpeCSF^a5{T97W30T2geDnFg!J1=E355{0zt3hvOtwxW>>;BBa*Ll&YAt
z%~z)Le0O()@>G%f@;%frVnry9A34U_dVLSa4?W7AWJt8T0Zw2OE_|0<c@d)x&AV4n
zF|_?0#afJ;Ya!w?p_|Ka?2x2=Hm_f!IC%ueHw2{|QPN?0ejn>=hG8dRH}2AY?G4uN
z-DBqTIVwj^F$_Fheh!Y)lq62Lbp0xwZf6h2FTF&yTBD~@l75?|@BT6Efnwi*eH=OS
z4<L;Yo}e&0Ng96;jtyz5vx4r^$FXZXL}h}bFMN^h^xNo_4LoJLo=Ix6!$l@4a$TR)
z2!62K<c-#=#4@yo@<LN673p<)c%DNl1Z6)&ONaGl1Kb+_`$nI?R-2`@*<||AA)GMa
zka4(t{R)eVGnA$#S-yCc+pqqZ<DQQ{F=<m)j6}x~{niRX3Tpe0Ap;L-G%5j|I3Tl8
zyDPv5{CtJ^a}Uzpx`v7+QLB&eD{L;`W8r`xSMiw$N>rYGmW3ln*|_)yyVouf4ii#R
zlrh!|Twuot5^x=)=i>PRkDPsga<NRVT%g}>^2M)wh246cqU+)3@*H~d^Ej@<%&AST
zzkLO~5EC1^tJ}DNOHk<}wZQ{iX>*6|xYf0}@xmY^NJk>nUx85KTi^QDdlDo5$v^og
z{LR1lH^)U?nx<U4cI_j-o-qPxB*}W6rI%je&|j;OE(8SlW4|cJU>&Ve%|vH|8NG||
zD)2&t(ny01{`DMkK}g)|L#pwFz|G}oH{0yb!8zT=Eza^z?>0&EHt}%4tKA{Tjvr&#
zAM&GbzR0oDCpmxaJc%)^F0U{(Jx!rhp|RVb+ubDy9C|%N7zRvE)&Na3h!H~4Y_>^L
zMWr&y!{?txB@)-|5r&E+Zc;21k$5CgmmL+;j8bY_OC&p+=)pGbxzD2A97b5>&1fW>
z3Cu9s(Cv0vUb)LoeT&WQI!fjE_y6IK=(PL%+OK`p+9+iFsU)$6l^DxIa$Orm;yA8t
zr5kNIKG}~nQDc+Zu_XLcY3*AVe$}Y6_1fD^eePM}DVI>V)??2deZK;s3`hDM&X^$u
z2XF&~ww>2#5Ro_zdEY0BBGOb*l#(EyXT9CvP$)P%=y9bm$42uGT3<j%eXa~+4jef^
zzaMeyrPr7|c7WMq=Sb3r=x(i2o0`N4%d{FTl3^3Cl*6qym@UndpPWSsfg1Dy2WhrQ
zdtHLuapq4yND_rOeuNun`e~D5seq6!X&h76F>Aw=`K@IJ^&M2dPVV7nNd1z{haK&4
z_7;n*d(~*QSy^4DyVIm!--da~m2dtV2I~z@{rcYmDM^)PkO-12&lhcwnN5jKK{y_%
z(j@nZB9vC+&&Ej24L>Ug_{*$9=t5r;s)VGqi4$m|))uPUWO8x>Tz5P=kYPaHm$*Sl
zdwYNy_E2$*NfR5t<a?y4qSNZ(C`ECyMl|SSBsi{T1x0C!PEy<;AXh4L;`}j&QA%TX
zhleL~yuI9JWp|f@Q*+$jT;oc&&3P0vwHa<*zDDYJJp0s>EX>bv<j8MQt4;DB|MWjF
zKexc_>>QUay-O4&I1Y5X9aiqG<B+FZ$uaDgDCTRlI&EqbH9HcMqcn<4F`H%ksf@0{
zh&yNOd|66R%3@k)3}NZ!3G%h<>5*a#^m;v3R_?OBy}{<j7F%086mlU_C%CRvlO8|w
z5VgsD2%yoZqe)p<n8k5q776#sANJYKA&e%i@38smg<p*0L7OvrNYMqv9*%n%jte6k
z$0<eEB`g$J@9c6Ul&tr=TrJMAzPp4rZ=&NKmj^Kij~<}c9dh%f*O)qXfSF@wNz?Om
zw^pf4OyJ}ybee6_sEJ?66Zad;7U#%M%v!Z+-{M&4O;o>&A0B1?)Pp2(faAsQ!?A~o
z66!i;HA<P^9MId@A?el0Km0VQTS5!_0R0&pYfYowVtI9)UcE_g=e;;S{TqMVo)b#Z
zPi$1*2*-}=;CPPZRNjx{u?MNevH2^&u@KsJUZ-Z5|ICwYee*j6TPXvQu_>;VFj~_2
zjeM22Wu9&-FlLC65-lLe=kS!G7exeNi08X>QpK>>WwAQJjR@WjOFYotWTAWvIdKlh
zFYsVK<o2~&luCI{f8{wgFW+MQ>f3mpqBM7q-JKoM-Y%8NDqdJ4?nU^;5>c;9zrKYr
z4sJdKp~+87Aw7?%yUXgu*YSb@Dh^49633mO-R`iUW85%cMg|<nxvcCvLSc3muQ)^R
zmG9vt5keb#F5PcqCagQhRJF>XgZnAXPEeeggB8sKzy6<-n<(+lpZqTvW00<c;|Sc$
zok9pusX`?R=?IL9EpJYaL=o;-93iE&yuh)D>!a=C$13XH|H1jhA0#6)ZU(e(EVKUd
zn;iY}bC}9R7GEUBqJ?VQVY;7C4GpQAWztD&I4K>6;dYb!%mj|>;`tt(R*Pboqf)JK
z;o?1NuHr<P<Lse>IKpH3)(vhpw%Ax*;_%@UOjM@0@X8w$Ckvc^^ih8K%fHGqpZOf!
ze#}pP{1g84PyaXe?{}D}4RJh|QYmkl71|I40jn#k+_`&?N_he)JQ4;}YDFB8(&_Z+
zb$SfDF}-%1nlm6&1<T+`TNG}*MPdY;dm7<RV~k_nsWVb8S**<R@?DzEI@>#2?B9Qw
zG#2=7K&j;N;)^d5$1!vBb8KvEW(gBk)tY85;iEQv9LJ<d!u!)B#^Ys^$?ly`ZiRqx
ztkyt4|K1ACx2}=<%!5#H7{n16Yw{>0Oe8U<V#S;X%8?`(Y@V$_x{}^b6K8q?&v7UP
z0UOO-X2U#-<w^c*{RTOqI2!iZ-gguco~FHWi`~I1ba!qtvv7iHd4lzeSMjSk7S9~x
z=r6s%uRi}In~~-RKYo)}zWoi1Ig0Pk;FhZRVcF8TwZZX1yr4||-B$=F4`56VOiHer
zLz0s8d$d{|;*KKebqSq51y!_yo213ujqCIbn1>$2b*E5(mno_3J`0o(+_-y>Zo5l=
zdyDGAJVptm%o9%LSbFK3s3c<Y#DjFNUqy;^%z@l9lFkCL<CJ)`$NIf{Ji@Irw%GV&
zE2k6`ixkT>ba#h%a|tgRlFR3Cb2(Hxa(TCM@zj6_JaXkKNxzF3^dT*wr9dkOVnx&+
z;0g!F_vyEqc2wc`*3v}_e8)vaDKhVot5opQ1UyZ;wa%jxlC6yvsW2Sbx5(Y~RW5e>
zoYdRQ%*?TI^BzBnAW5K5E;BJv<8S_rzs-+-@*?-{-T?&Va+%M4_Lr!XYs68CLz7a;
zA?jC{nOtCZw@$S@Nf3DWUO3M7v}04O?FwR?@0|5>5V!~&j7dp>AS^Rcu8}L05su7M
zvREbi&JHWfciGrj=gO69Xroy?aDe*iJzQxhPtJ1W<arz?Ac_X;wsu)un5I<BkIkj`
zY?(gcdOwaYTqpOL2Oyu}m^~cVl9<zxVm`xhd>@X5E9uoOj$Oy06ozcJn#=?__LnF5
z^Yt6#rQ(R{v%PQ}5qPv$Z?oIaaJ+beYN^KBh06pJ1@@mg!Qo%}9AEwH(`-hHmwxyL
zFMs<RsOdwv&Lm!`il46l8fq5Deo(PEo;rjPc`ymNT8866k9NC5+*QQAE+wZ=AuZyB
zAd)8Y*RRnt0{zgV_}(O%`*1AxaD4ObJ-V$fy{&C3`(`mI2oVxY%yaw2KS!kz6UQE)
zd-K`{<Je?4PEz93jvepEvA)lG;nTvgCFd)Yg#6)y)Sfs;@4Igxx(S8A7E4-KKgS&}
z;9}^o7K5WSenvqOfKC;K=?T)!9a5!e4~NXxCRrc!SZeO@*B6e{(=OMCJ!Y4$g3?qD
zoMN&#PkH7f>vu0xZ|!j4iH8{ObXmIaE;pC|jIV#|Mdm|?FMaV@p8NIRMy}r??yZvy
z`^5b=Agxk99#Eb;%*4^t44SJ5(?tlML8Fht!7r4kRx9{r7r$6ykQV5;ZBiq^n_%_q
z0~965SAImAY$M0-H<r7vG>(DOM~_mfmMPB7v%9>Ca6y}hsGD-^g)iawA??+>R2TMx
zup?W?ad2EYF3^OOc)o}4`QyYXM>>1<JlW1@rTM7$v;RXmb%`;!*g)GLh1Iv-p>lYh
znFk&~1d`Zj44TkQskJu{u0wYSCXSJ=#H&`o5x6;@Ub{)JUZ*%Ug`3YY+}ft>=OHx|
zJcqpRaL|p|&ZWHB7R)amVX$_UxZmQ+<qLSZGLuu&H0mwB|L5P~;YS~&QmHaCdw@qC
zd<e9rvD;v3dX}R{kFve9#T&1`#?H<b!(pFJXF#Rgq`tk*pua<<RzkXxTQ_c_wBg>$
zD!a`VJ6o%qJ{lmUK}Fzs0a|FByBDF^z<KP;#MOOBHALaq^oTU2*X?lg&J8xV*U5)@
zo_*>$a)l`t5ANgEjT>CKe2J~?b)I_a8NUAS{}`<eNvhedZ`1CyEfGEAM%-Dt%k0z)
zei&M>udIVJD(dd{tox)ZHQKs#OAjxF-D_7VEKDO$9zYleWBM2jfuose*KwswI|3C)
zxQ-+&l`&G{hXIXN6SZ4sYI2G&m!q|{#lAp5qN#d0@&fWQr5_|*>}v8;M@j3ip`$u$
zOBZm0D!Ixu{YIBtKYRsozQqqFDpaTUaq`S*cB3BcR-5X?G-pqpptV!y;%jfxZET~G
zh;%SRVc`VB-F4!AkFb_QxPr#gJ+x75tkvna1`OLP9Nrfq9YZwK6uppCKzQdiX17iF
z^z(Ek=Wta_ghQGtOsc3iTim>}Om}M&Kg{v)Q%@4+C#mh5VPk2Hog42k*uBTX!_TpM
zcL|{#($uiiY|-xyk-lTe^GdU`yiK)M!u5R+He~(%-iX?mkIugRIZMWU5Q%kn*F(}?
zi=;oW2Ez_WV-<lTXLOxvM4||T98r)bjfSW=!8np2F|>z$?yNL9b#gyx93hOs4Rhc+
z7$B9_*J;>o;ix<~0f8G*%$12N38yA{Yz<vrZnQDJ<nZDmH`Z3T*6T61w#v++q_wrh
z_rLvJ&OdUV`Nc)fojcEolPCH5zyCvS-MT@!Ji*lDK3pe2iWc+pvvfP#<bpC2wF%Cg
zJc&qQs+z}jT<fSmQdbJ=+hvgUL74f(S$do8%@r#Z%9Sc%SVT(KcGb+DiE6LcV|{&v
zey`2U)D#od8p;?d<ubd=8tHl*J@o*!sd+Rh-EJEx4F?t%afC}EU}Vnlsf=IWkK?OX
zERIj?M;He_2*<HNM-i?o@k%AMkOW~squE3?8X1m5np+X`ejXA{(G8iGg1k)W`7xJz
zhWykK()#P@$gqC9j}ue~t26W(9d7>MW#l7g`O)kYm5F)IJaCR?GN9FLQ=ORN?8)Oa
zcj{bv?M*t3ZB!f)_ogT;9A~h*M%?QW)<SRvyLXn+TCu)Xr{A<VKD5u@!?7EZ3PZ4T
z6SLbQeCByNwOL%15FtrZJ;w2krDeKXTX;cmAC9NlxU)|E#yj*I_gHxNd3KlY`~n>B
zP_346eeY**tUqNOyP04}>4e0E$p_BUy|;;e_clsu3ZjTW5GjX^>NKfNDEkH{$dM>T
z(rP1(AsP&stWDxfPO-hS#h~A(HouS6r~|%3J55>930>n;i~}AiMD(_<;!Zz`45!$C
z;tA5>E`!Dfh3N{XzVtct%Wv_*+9tbOE4=;s>(snFwOW-%eUrB@y-RI&hDSg14CU!L
z96!V>ROvT%$WJa36if8$eN3X6Ja!Br92WM^P^jg}RS$A&IYKKzn1fEz$2EfC@l!Ba
zCHnriDDA9~IzAF*twlVS=~|W3#}8AU-$y(g@XCuXG3aekU*2Hfu|;N1JxqS`Aj_|P
zkMqCv_Ye|XS28m(L7|jGMG7|r=}Kk}&$3%@lEg7+YlSq9X}izw(vqe=!ajbSP5+=<
z-zT{R#s>11LWTz7-jG}WKkEK0$g=Ca?*)Hrtv#RloSZXju0j<CVjw^e5C>3{C{dOu
zU&(5<d|Q?k?ucu<JHl2tI@)?^KUm@D7kAk0Yr8F5y0RVDmTg%wWmBdkP!dImBnW`S
zJXE3P$~jM-`RsYE)ek$X08o@fTtE`NBI+UYWbCXw>-_fq);IkBPk)2X@paTngeWww
zP}C_|D-xVkp$%l?5k;C3Or|tf7LYD@kw-e15~Bz^O#&|<NmIPg=iFV(<VnJGH0Jwz
zaJ(Jx)PZ8QaEhDT*D+>Jmd=<a6Xx@r^<Ixxo_n6xzwrXS^+ncCo+6AIc&<;k;o-T0
zdK7Zc-FI{D+>aszZ0~IG+rRni<mnVCbI=)wTLWJC#_Qa5>I3ZF9B{aE1Alg%Zqh?%
z9UQ4}f(Alcl+x&fD~zB1Zv>D3D0vuQZCTmg=#s<!0ekzq6j{#qeB%2!dF(i)aadbf
z<?^LBSY2M?p$G0Esx?qjR6e1)3ha=Q`E1T?KBp*34i5TEr!zXfU*!*E_3(Mi*!(E4
zdmno0w5_Ijt9}b8<AlAJUZK9$qv&~v(&8X!Tg`)cLXf+Z+A<yvm?a6_UWetSMTC+>
zk<Yk4#zoQWv}w9N!z86q3wiwbGSf6=cQoQtJ;6pN<o6F1W&LiZJ8xk99FuO6$BPt)
zG4<sWT=|CP+KaDJU+&U8wvGzxD95AKa6uWgr&v6DoR6RX0i={{Zr$SRzw-r(IIl7<
zT9fP^aPa(#w2ys=$!1JCxQ3IxNiAC@O***JAq-k5ZSj;PJ>2HnQ@=y!i60=Xg;i62
zS?MqB?Cmq$AE2|0hd=gRTsU={Ogro^FS2{}7WLjL8~2^V4I%>9gDk5Y5soIrvPn)h
zOEE>stREA{8Fkk^atOPvEq1#X=V;>keQ2=G@(GKPi!+ZYr-vXc^+t`^;UPj)hc9a?
zEiWmm?7P((BfzWInU4=CvJ~aI6w<P@H|5qu@JP2oKAj<yiyK6ss=gu#pcGO;a=1s(
z>X8|6eV=+5a$=>;4|N@`J#&q%$%se_?mc~mH?Le}GtZdq-eSWT+Kn@O?aNPd`uth6
z%~@Pp=H9#SVe|SWumWr5xFK{pb$k!@cS_Ek{s4%QZfBjt&3)E8F-|rlixv>2#mdUb
zz(OT2b0@fi+*BpgrKKKGR7Z&F;jNFdGxq2pRq0J0u(!7j!tkLFJ<1~wf0+Jo#FxJK
z1vXz7G&&tl-*q3#^~mys>1@o2V;eN;4XR1@x9Q)#4_y=$$KstFzxXoswJt^Xog6>(
zHjYQb0rMoG+v^^2EQo48<NgR2dBky0aU6ww^u!9&BxQFz;`<jQ8^-6ULzkj{H?!^6
zu|bATZ;>TS6#Y5%m6N>r+{;|~#>>=KyR?q2qk<Z)^60dDlqVtdSUP`-C+_+|q*82c
z-sHJ2JV_DffCjC}_WJC9<9XW09%XVXCLLbKNiR{)R!Gw}u5bw3EtCdNS+f2X*T3|;
zbUyz5%xfXqia*S8n()DoeHVA1K0&5k`l~DKUEikOTVws6v$#=>(0Loj<q^lzw{d(h
zCyrC<uJfnPv66x#S+?pRRf|}C<bzCZ?c*P$;5w98woyFeN5p=CE1|QvL_G}YAN1MX
z*~ONYc$Uyw>QbN?4~C5MgiaJOoz3~(D_gXpkkw9?U+=?>VafL{6u9HpXs+B1Wg9>2
z5kx)QX2A4to4Y^q2<IPrgz4r1gKO6@$&5nh42CIXZXnLsdG!*%^~!%lYt6|Mr+NG{
ze-SV8s^Lq=Ay`{t<&lRN-1-{rQ+2}LS<3JN6MvU59FW_RBp#p@WLgp|o#6EM{SbrS
z`Za>-93>>el}OKJV{MJz>LMx(dHTuU<>gCn;;WM7%}v(M9%tI$<mO9nkk4mmqmPt4
z1fE;%U(oXKn<3J3X|6P=)kCJkqLM|jhRP1Eiq&t6B7eIkPw%??`qMxC)A)X6yOO3U
zf8%fbjXRDKZ|}GJwytL+Qdx{NXe2@_hL^YLzxW!*yIqWoaFk$yjIi#bixP#v4;x4v
z^6{LBl+>3OP)>mB6~z5Mu7G;2Mv><z$Hh|;*AI!4f^;_KgN>YxlXZUmwG4OpI9qYd
z_Te_kY=AKkn4I2hM9^$8P7*d>zeei0G?$jSb!&&Eg^Tz>fDnQ=iiD^hA|;$VdzwN2
zkm+>B=H?!I`+ZKFxd%sEcCUS%!|hw#R|^<l-XgDi_(6oU23ZyuTawIXjILj$7B&e#
z^l1duI5nnZeAs8)@AJt0k8uCp_n=J9l`EHc?&+`a+}FNH;Me$(AN~wiuU|t+g*BF<
zELd1rAx?6>{`IeM{l;~Iu*OF|{5ZZJRPqXME63g5V{_Z9@_oqGS*;Mhz(|d;3ZY%*
zH+Go3ah>H4*D%7vm4Z{0EYy9{ti(r9j~X~aF&WOVQn9$Si0=l}f}H;TKE4pF*6YMs
zj_0~G>rHAwjee3doK1MVUUGKB<5%7+xY;|;EGwArzDAL}ik%A_FDEDtkhLY|S;p}C
zCdPH}y9-RWce(lIWgOoJp?LN6Z{T_k2utteF=qXkY?d(F-DTE4q;c#KgjK|QUnAXo
zm0qjMc<Uy!z@rgHIB1kEC^OAqI_Ace%Vc3h=cC_Sy-6C-C8NUuY5$PB?!AY{?mdUn
z39nwd$)#ssX7}YU;RyKfr~f*8x3*E94`qhdB@3+%d7`=T;zcIAmr-7g6CZpS*K?}i
zoOk*|-Y&x%`~K{)X9`T3k?5Fwd`MvmJV)WV4)JJAP>)d30V;bgiV|TA+GN%HYC*>_
zD3`%x#*G_8&YxXCSxZ)GT;E4|E>cx~uu=$0X$hMh_I7Tf>TQJMlFp_WV+n&gt#-(F
zpDp-}*Y_D64k?7?++C-+apgKP%TYSV6*(u@7umUfi(8vJ-1~vEL{Y@Oci&5vCpeCw
z(ex<GjF(@!geN1G78bd7^(xbGhR}zs#UAO!7jYl_D1}mBD?wg^z)Dj|n2M?&Da#7h
zR@VrFfO#BKW+jbU10z&5<$lCslEm!oZZe&Y2!eoSqlM>1oVjo>n>VlU^{)s{ox4DH
zVHKk_@qCU_isQ#Os$M!{0dprqXWxe=daD(}`zDTW?lO7f2Fs7uFv7=`f)kYVYCcI;
z;3H{-4IC7c;S?)n#jz7m^E3MU`vg+HlVjDS7DODRIiuNxk2DNtHU$6qQqHZ#yO}2$
z^S#$7(wDI_f#Vef`2n)N$m~YS@anCK<K80EoqcY;d8Hb_LGqe@9@nol_Ijs|GdqmQ
zrwOyYU1s}z8pl3_v<~t9v!u6Pq1Wm#-nzjoaE~}PC|#g4&2V;`<B--zzNbR#9UKoh
zckkUicHcQ1lknQ5OI&>BWp-cseH<b9@TdP5_HJz>eIIm=(IxG6hdk5Vc<~aG-77d=
zjpGkK_zsTwQ{-4Wx|+;35-AKg1&z}iv_Ei`?Wbl)o01#L6&&`bDYd{Q#WI<XDQ!u+
z*FqT0{_Y-GkuXaG7FX9ujb&>xAqsr5JjawHrpb(@$fwij@VgU_i#KzA>O#PYgEweK
zb?Ut{xV?37p|n1UzlfgoX`Bh@o<D`k3)1~vnj5Q}y8CX*GzAO~-2Yt^SpimZ@a8od
zOEvsPM00JO#>x_tThC&prndSK!nN}pOlNdXoTiZ!OyW7YP?U~`BOH3oHfPS?O*o%3
z_?`b&9nc)G%B8!sK)AHX>o2^>m%i}|g{<+wy{EbNqmQum@b^&^DPQ_8|CF^;$1r6z
z=`NMS;otx-kevR|eYDo^qRggTd*)e+yu?w?I}&HN4~F8lGtIYIVt-db-CzByfAt+-
z|Kv~p<RAK(pZckv`nInp5x9;-Pzh8TE67vJ&9A&nZ*7&@U1#XU6HZuz8`crhLu!pN
zC4QtaSyq)~w1e_P@Ke0PVRkqqT<9WPMV9AyzDH@Hv$V)=JR*rxHWr%v^~X2((&YpG
z`(=&4e4PFHl*92M2YJrCELlua)&q|v-^Yk1>0rR-G~=qQ+$0>|!wY=ez(*=US(Fq-
zb!t*puoQG*G-k4Wjm^P6%`hULPl*<47@ea_2V3R{>r$pAI?G7sW7O+U(Yf$J@>;Lb
zNH8T>&GJHzd+xfM>2Sy_nX<EW3(v9Kch5Pb@YvbfW`B1Vr4&gL^U_N%<EW6q@Q||9
z+_-TAYbB39{0Q@S&Tuj!%k#IgAJ$e)oJYMIX_CI1kH@!MLR1Fog7PRs13I%Dyl|1u
z(gOa;Wlm)o9|dYr9U&dmk@G>brs-xm){&qU!VR!~LL+zS_YdhTEh0R{B+qF00i}gA
zOG`YTj5v%_PA@e1e>`%E&tKW%3!4RM?H=OE5{bV>oZcc&4hg5*)N4(mU=d3PJs2~d
zr3_`&?<rju>G>${sGlph=t5wPKw1dhMMy>z2d^{Rx`Y?D(diyRXPL4r$cqeJ<S5@`
zo~Ptl%49adf9+XL-giIK`a-2ar3<7jX)iQ5cjg3p{X>Syl&!5k(vkE|T|i37-d>+!
ze~1X{q~jr*uV2ABA+u48(VEHjYuGH~*aIIR&2q|lLV08?F007lj)aXhmMn?i^B!B1
zV=0;MUdPcTVWWnfrx1C_(Ht>Pndb$RW#td8EuzRM^AzR#C?P4*oLhq(X447Nso}#9
zZji?bj_cwEK0;OY>Q)G>sLG1UuM_wo$!tt>VF~F4jE4ho40$%?<EJ`M!qaaaaJ@fa
zlp5~6aDn;Ykp5uEIDd^ud917+$8WCi%5$%R=h135kjkQzVf)4{Zrr#@5Z3tk$3Drc
zFTX}MFG=D<&ULG4?eQDW(plU<uAW7UqasO_^gv31wSpv>@$$<rkR&l@&zxgEnzD8E
zI;T#broOaLJ)Do4;fI4hhldB0x@2{A86_3<W*4bEqI!!)Z;`Wi-H-1D<oTTWe9Fla
zCulV)k5_}KCIE#6QTgS**CmjZYsNb`HuPV3gU(V9|HLvU^Nf!JwORuq6-rlqV9kcE
z#HDX@>;>2$p;5RT9Q5D8ar21dvr9`nACEYkC!AYo^8bAFG{5oYCcm>;qSo$ZK3*bq
zw~5o6<Y}L1wneShBJwMa&Ec5QG-V(v?Lp~!NZ&(w{!!nRK^GFEE18PWTZVK*e(*Z8
z%}aP;>yO~rB`z}ZG-W)S;Jx|`C+_{g9geLokfx;7t8v%a(;OTQ875OUw-1nxs5lmq
zz5PC;{ty|~NGC%!-?)NtLuSJmQ<h9NU&iJ+#~!$kG|wsHlrk^jh-3Rr9Op^$p7+>T
zjnO5R+d6O-6-bsJ{Q!qM`=nR4x$Fmg0l{oG!I~1SDlOUZMBxNJvNQ;(F?oVaA|_tW
z(#kT2J3H81)A9nwNkS;#jqM#CJ=<fc9`gF$i2v7DuJB`bt@8sX73J<5)EYga_A0en
z1J4hc*Bi{@n0Y!!+5vL0L$?*Oa1`g7Ji}xuWtL*K#qlM|vsL0qK$7oZqb}a!-87a@
zkjGP0)FT-UaUBoOD{)*OEpqB%fFFjmn;r6pKF*crpQAC|#}N{#JdAW19!z-kwTmn-
zEwT``IB|RxT^bH9zk<=4wbf<1OADAh!)tap@qzQmBBs%6ARU+Xv3n58a{aj%2z(zm
z3UT~Oc-p|*%CD7Nkj%claP3_Mb-(mWzx03Lqw+^DPm}~gF*ou^^wVH46o)e|{{C0_
zV6#SJx{WR7XjdXshSjFJsbDdMhP0$ib7aj$dLGvEp-jnUQ$*yW9LVD-ZdAh>hemsm
zcsRjk1q#Eb&h}^qf`9+Y7MT~(JARVcY{J2K$jvNgE6<4JoYr8B<0x$C(`heo?D$E-
zPMg_q0;Pr0pvn?Q1M>+^JR+Zt*q=;zGtOv5K1<T0r!yAX5xPhaen2{#<M|$Wex#A6
zGipVS-n|ZIdf21J>-F^whLe3RU3{5bYf5d{+}YyR)+VLVw3;n;_BYwv->XDGtz~Ct
zhn=Nc)ao^kuOGwt*vE<Egz<QUQj$2$xpe6guf6tq^&p^9wig@;N4_s9{Tx?%WX7qo
zj@8bg$8&Bx_X77e9%p$z#ugc+Cy}Z|l=`Uu$YP2TV#_?qaq3loWxX7)OqfhYh<bqU
z2xjv!VXcm59X8sF^oL_inj@^{$4;-}hJAkH(p7Y`MX>ZB#cYdYu*GyVW}eM(*vEGR
zq}PEkAZjeqTsebkv?yj1tZ}ef;FJcb1vZ-^vqSXkkoaJWo#~uLBcK&^h&ZHM3(3m_
zfyZDnr4jhdvh-+rEn~6J<Xbo4LhnwAY;ARw&CwyRU%g6UE6cFq_5qW<ZH(3UwI-we
zF_VKK!uN>-$z;DzZF`L0@My2CqMRS1Naw`k44f%>QZl%Dh5jpFK^uJ}$KzFdj6j?B
zUSfCb6g^Jygurdqacd#-@dQ^nXyt%5h_b|F1xcD$9*0U+9?m+)N=;@8`uz#k`YbKD
z_(D=xfgAWJKe(gOWUWOh=g5=N&}g<89PXj>3>8Ee2Z}f)@O(U_d1%FBvA)7fm-l(G
zKj!kKl!c8AddC;pxpf^k8<Qsqa&pMZ;u=#cn2+Yw?-Ni;mt)5+P<QHVUh7lj6Q+lK
zR=b)<y;h}ZH5)R0^=mYiHz*w+P<M>Qg%B8{@%@1NA9#qDUib!=-@L>__dmqO@nZ-n
zF<MuH7N!yj-`m?_G#VlVw3<yUl3s5KEnsnZjZ^3Dp}nwFW%rXxI^x8!4Fpx0^mgW5
z>+<ORde2*oQs0{6ax~}0b1!ml<1todZ|7Jcv__cQ_v#W7%RDV`YEGpO>G|*AI5^@s
zIO2G{v&djL#w00NhaW$)h8OnvjW@5L+a03iCGzPl;=wkP;fPr}!(|W83y|&+$IT_0
zD`#-(E%Mo<YG)~d!c_O{Y=+DRn0deAcrv42_i05P!Xu80GC^36!DK=+@R(*P<=Z*l
zz6s}hM}2OvwzkUVWWej!u2LT9LyorljBfx3LHNFNTyHVjA2U7}BK&~ZlZ<x{h`M9^
zX2r2oKSYttn2i#wnvy5F;`p_%R8Bxtx`s#A^Of-KdtBQ17_0C_d6eaq2rF^A4bFVz
zQC=8-k*#J2+yE!5)Dx^Du}2O4(sw{gqR^w9=9nUbEG2UzTCE;-8k3gLXmm(s6QnG-
zIoM^f*G5GFRyzFBm0i-*@Ru%}Ld6r}!7KR9MWWUc?Rt$yRHM*cGQEt+Gnl8$C;N=H
zFH?>O_{}y#DWn71m6d3H(8BsXWPOoZyT@XqMlw%`%a~XN^rv$?H^QbF3+*=JB&8Wf
zDCOX&fL+(YKYj-B%05y^lz^KzH#yN+;{LlX;7W<31f65cwAYRyr6QY*=&p8Xuk<iw
zRbgzetuR0Al8!Qlw{Btd=WxRYjc$Y}TKJJ-;rs;_FFZokuymAVw^cTFI-A1(`~UMN
z?&F{9pu}x)-O^Yz3VozW<yuK{bqD?Y3$)i8Xy0J60%Zi&k$9nxl~tNDo6OO<A#4S>
zj>BX$#`j(FvP9+*Ed)i9BHbz@5JVB_bdGWzN^6-8rkwN?KYqT)ufD#`#RGxz0!}Wj
zV3lMV&pDh;m=-xsTH=)jqAYRIl-mww7E?#ki~<S)JEIxXWJZ={9F~$?6fC+9Ye2gg
zao>p=9x>n}j|3WZR#57aJV^<SqV5OS;UU5lm1MkucpCG<v(GR**dwU5@q#*s!!cVs
z+Z-J35rz@<k2NqwiLx3`2^PC;gw2?aQ;apV8!e)s&TKMgdux~7!vQyM-sGU)KhlwU
zt3UEc{`7s=;|h3V3`VG<9zIZ3k!|l2ympz>i!F4ZNzxq82#g~Mq5zIgD6`R&A~%HX
zkid7@8Vzat9<!pv$qI5In8yjO=i>@RJBS!Lb9~Pw6_(x6gfp(g6Q|br@{Nlej02=o
zqq%ZF2tl62B(q!O`3RNd$k>N?gtZd0r?ANkS2%b<gcgE$+$W1;ipc@fq=pF(sJ9w4
zojO5sm6OL;ffyxfcv4Yw9P%_L)+N(<LKDeK5TFMml+Hm*gofE9=GKcZlMD}V!Vc1_
z6OYEsN82RRO&lk{3%(!95?92ugN$0c4HTr~3~LQRy@?ytC}uPEceWT0_8H%}#N^s*
zOwu~Jjc^==Yb=pAcY^u%vXbEej;th{N~4iUMi$TTnk@)?Dw73ES!7t{Fppz=&#4X$
zc>zeYsZ8JFB<HaQPg3d<FAR~Pk2tbTGPgx~Jr9h=U=Xzkf8<ll6NDe(cmaN{5z?W^
zGV025cBRE#$2U0n%&UCv#(=H8EyiJk)#G;&O!sgH2c-MkTtC<a8L$|6)WQhiMrbFb
zEHnCf!Yod3vKi-3wE6Vf2CIR<SOw(NXFG`b5Vv^(t0cxb$}HL{YbK?o-D>f@pZq@F
zxO9o9zWOvD|JX-qwtC#!yh*3iMM{Uk;DFuT9cHr$-ChSF6uw_0@FT1+tgftc{KP3N
zmMo2l<2h$eo}y8&qf0}fD;JU55*Y80D0&a;kL7r45C7H6oLX$5Lrszvcm}Mi@I$}i
zSR8Sj8NybVz<1ahjo!g=cEquB#BnEx7&$Y7z#|o~KbdmQcX{H}I)8BeBK>iQbQ`qR
z9>iKGk{QYD7Fjl_I8FjcMwPyqRG4^%E8od+9>)~peI`jA6Yf)M)oD5ng7_*YPpkly
zcdVxrk#fks6~~8Tl+F>l(h{3Z5;k9ag?M;?8+MRhjd(a_HrlE<b^|>B)1Wo3NN9%{
zwN4j7NjlEKTB2459W~H#!ol7ilfgct>u)f*_A29~PGKXIqwtI+e4jY325K$Jkw{7e
z5)70=;~!rmeduo9+#TS=Rp6E>izqbNJgtId3-xXbFAyZ7m^3Rn*6A@wGn$@98F={C
zQm;jnc}i|e2Jwt;vq6$(1V*shZt;3v@Z_~!K6&~yOPvy3#*Fr^Agsl!w~3-2z8heC
zAMMv53@Q_sl>!kpDA43djI@?otBoJFaO!nxUV+jz;xuDgmL&cPS2l-OB8*MQwPse9
z_+dcgI5cY!Bca(!5}d^@TKQOIFpj0!jyU<D4-$2kDdIy+Hl^0<(pWe_*gQs=&&ksN
zif|mX@W@NcY;cIRE@9gPp_mT`II}TM6tVn)Rhqp;I>+y#v3eY(%$<q(s+j`($>iJG
z$9EOf?d<IEPygvZ<sbZme*nNoKl)MP_^tAdQi{)h_Oq<7uYX(D6Sq&Ywi^GhBmlq&
zK`0$I>Jrx}(bFk9t$qeom2FkvB3%ck9-^&AdM?&#w6G+3f#*1g!c^K(rA8vEQDoPl
z*=dvIIfsX1u3tYOY&2NxI{f5?9$&pa;Mx9!{$Pr69BPdkOD9i&QjDh)vU!Yy<#0G+
zlIMhlrl|zX6+7B+DbC2O#8Kcmnhhy=sNr*}Avv=U(vS(l`UF9MQWD?wkh=1$O{Oy>
z4#*M?u0V1gqdHh^st?J0%8jcRDYT0pEz(;!PPe&6o~8^2hy4EU{T_L47|mj|HGK8i
zZ}5$mUna{7^5V$&S(kT&g|&HSKOJui;@%I1I;hGvRZo#Q^0++`lU3lOxF>X|xte@D
zAx{k@0;@oYDof+I4t}FXVM`p>!B|a+<uK1_Iu0V&6tWtlcM<4P<GT*Wn=Qsg&b8@?
z7k0O41`)kR$j6SK<hAW9+?WhW(grvVuJ03VoChS?e29)m$f!j+9g@d0oY@Sw7GY}<
zah@>T*`W}1Oz04V6Y9cawb5fSg2mPbq0Dfspza5FO44*al&%yk4`*{+oJu5fe-k^q
zNuj#fT$ATBOgd$<`y9q}P~I|r{WyNGhR$QM(VJ{O`#H+QGQKutE(Pvaej8cyDa)LE
zYA6a#CZ^~tMw)^wcPR6MJS(uR8F(og$a0v+N3ulad)wnO&!HY*vYcX^l3R;%T&(X^
zwmnCwB2#7vO8g@YFy+WQ4kl?KFeTi3VVTHtF~;CT0oL^p0t^mD6h}td4p@OM3ZxQ5
zwI)e2!<GfLAYz{9BzeLtnG$*ti#4A#aJcXOlLQ}l5dW?Z@Gt+v|IW2*uhXhGX|1di
z4@O)Y4#>=mKtMw{)SYUGt1yNf=!8Bee2>-P0qv{_ZWCcWyn2m58j_tWl)Vk4bSw7=
z-jb&POP<CUr8s@+6emxd1z~vZ`DfYM+UC2z`;!z!&erA)4iEbXDd=`OXl;q2X62S3
zB#R5nv|C+f$(*uuky3JOV*_1kiae({@_E%%t^-2KO0V)guD3Y0-;(2dLx=ipj+3`I
z?sVG~$DV`VjQ$YEgFL74EjTXkaNK5;=UknPcwu*oW)RYA)Og~=NnYK$%#G=QG--fS
zmFc7IJzxddbbwBVh_FpEIV7LYaAtGddWel`#6`kzXPcY`HgX7}F*Q6^>pd1C!E(Fe
zSlN_@AK)rU%d0rHLNIt6$9De~c6N(gbt;atDP=rlvh&P0alBe_JbaVQXFi8cE#qqw
zVkx-w)!$~b7NE<Fd}hcqO@73&FgaP~P!{jxxTNBE_C9fJkKB?ZAUJYCu>z?fMe|}@
zFwGtOP6t0!IKm)G7}x<tl2K*_*?5L89Ll^z3WbYc(GM7v1^5o00qHvASRB_O%?qxL
z26ViDyV_kow7$+bnXo@jdGgv84{WS)y0u8y?jVYSd_HBi{~7{=-{|5*O>CwK+Y9)9
zL=n0axrR8W?l^S24O~%R3n)^bk=1NZ=j_FXoxzZ~iO?ctl2t+7IM2|IL|DsWvq3?L
z&@N*26ryn&Pxq0YsshDYm-ev_A>|yM?T}8dlE$}CUPRDc=j2EKBFdN4drchaaQu-c
zSidjDaeb8U;y4PayvpXxagU~4k9z7%l_fob4<$i;o8j8K3hMs(KmX?phr>Hxzxd*d
z{I$RK*XVY;0DS)QpXVR_qkqId{D=SW+qzz$6jh2hM~W(}bF2a!x;;rNFqGplWtt&u
z^~ajRk_n4fgoq#jR}e)tgd?Fepbce~<4OmN#qnHH9b?KIC0r;qQh;9zSgl8NRu-z0
zwa`pw6TVBAoUEt3F_S#AH)cA>NCpFp=MmHzv>I((&%^OU+QK`U21^;`ImIj`ltcVH
zhr-bHBtLbb&fRsNBAK9d2EwJ=S|su!{2-`$<^%-FB}u0=f)F>i=ww26^Cir&4<k#L
zwUtFa^ytTV>X}zb(t?&qsM`{QA*@HZ$(Ye(LYf=21!FBc{lj;&y?-jpb5vF6vn5g*
z1R7}+SVgZbSaLPVa6p!%M}-7|7M97-(uiuvC`5RI&e0x=va(&u(}H^FB8<WFJcgwu
zFEV`LW6KgHp&j`=&{*SGcM+o`xhUD1PQk3R(3rD7S6n~XB})Z*vV(CQoM0Khc@p6T
zNGHH+`v@UHW#na!o=$N)0w+zeiASgekKez<iAI}wKBP2r1QEw;t1NgCoiGGb^^CS$
zk7+(*+4qSu$g-I6m5Y=o9>C2#?po=vdH=)QeEKVt`4F$?;e;!dAgCWwgk93vh;pj2
z4VT$|K|H#Iks2#NmsR-#l><jpQ`W5<4FKW-@}vH$sy*j@iQWU(#dlp&m80W=#zGHA
zDoPAaP?hnEtU%jJjwtXYMwb}ZukKb-GMtqJuE%n(hp-9}Dzvc3qCg9c!a*v(^6ilZ
zEvpBPDKt(!!cHfMtbi~;xeoK;lp;%rf)HC2_^w8@n|#;De*pi)18hwPY;A5beDx}?
z-n>O31%dD5YlF6$u_+ko>IT^kLr&FO)TUF~`IIwDON3H@(FCE3>s6LTTz!M+!~^8*
z)uU-vdt`fNKv%(QQDz8PqP5`VS6^T<8M3x^oG1u6=<hL|j?t#1*=*o>K1G=m*8HPh
zNJ*>JrPb=Nw|5H@hMYZfj9M5`<T+`Y-f@Ac^a%|@3i2#}&wK3N;W)1zK5yf=+lD32
zkPL@pN%Bn`LnDfgI99Zy;2j*NSwT&_ljEYu@PtoVmN<7fUgcPC0V5TKEZLk+uzH<^
zM$Fz^a-+XTJ`(8h7RGaM{8jw+8Kf5=oB*%mBM=qGI!8}#bDUulk3hm>4=i!2(Iw7@
z6lV4o$8Puzj@ypQ<eNC|U%5zm;{J-`)ec(^KFV)D^EJwRgume9Mr#$vK}1pOl1@jI
z(-PZoneG?N-@J$sx^nv|{sfMbw{fiBKaTUV!0|Lz`dFi}I2c(1CH;YACsRbd1%fET
z2#c~7$AxB3Fr7~^^B9|H2J-}4R;5|tD5Ah)8W$KL$g-TSEAiZb%%p^YLt$VzOL%T_
zi)#nFNNw<p!--ZKl+XS&Vy$jD-Rt4CBcjeSBF`z40m*0wn<tneAtS908YaSbd}OUg
zAr~oqA3Q;d<f%&o-ki;-`#oIQ0Gn`aw#|Mv=P-_s1e7hKBBvcR*-jE>fgnh0EM52j
z?&W96{gl^k^*Qm(Q+(!;58`+M*{DuFt2|%JWQLh11obXXz0#ajQqo=B_$K9<YTCJK
zz<sAIjga!rp5D>4<4<8w;&Y$-9RK8>{F8Tl?{EL@zs>2>rvZ5EvB&s3f9LOf`$3&k
zVyt^)1qHVYepVVbR=Vi<jG`>7dc9I$E73n&mSkyy$_uDDs8E5a6quFckQOC@Z?RY$
zQJp)AEJOGXQc3c%Kp2TvZ{XINSb3D;@oLn<2AxKqM}v?D28(?8>WImtWHg_WWCM&g
zjD#c$J)|qi3qfQh=n`oQP6~<91+8AhU%If!LON$W+{bYg&3cnYqe;8n#`S~B_feM!
zQ7K7-#V8O`;b_b3+Es$3d$FQLsWs2M{0iISIi<GrlRd8P9#mx!dDJwk-#+aAov~b1
zVyyF)_$*b%$|}o=ZkJ>>!7ZviR!SkPK^d@R!6cdE<vG+{8b^EVNQvh-jEVvsSTL60
z$Yv=?Q&i~UD#ffQkXq7cL~JayDW#w&OFCYim8ikhVW0JlkQdh1_}Yy_l8GQmhUj7k
zGrd9XhUj{Pa0FT_TpdE0ff>=zIaU`mdm#^BxIk!Q4(6}nsv6y3g{82`$!3Qrs9FL{
zmLr8j&2f;T3P_Yvgxa!qZHw08X-d%|D-D<5c$vNa4#h0P9b7|Sejb#<h8oHoJ(QTd
zPUhs4`Mi3Q@3hmDRi6GRF;l5LzU_ft9Q}L$H<{8HgzMl%A-U0&kE;epDrVD3m64Nz
zGAk%#CBA93#t$Nr!m`!h<AF14B&I-`96R>OwMU^{5FsuFemI93_|ip52kXos(+JnY
zt3_n<m>>+WQjw=Qq3@%FphyxzDd^=T7r*p7v_BG(q$MZMo#!zR&J4EM!sYrbVN+|Y
zbU7WsvA}^+l9Zap<bc)C<-U^(tS^K(7#v67ISM5#y2z2`A?5xJY;&1P%tTaATx&48
z1YJ1-?CflE>CH>zI;GX>vd~+g$Tfq*KBLhPtxJ60$5_kZV8CEF#t%Apo{R7MEcBMR
zdif%?z~R`&`kkICWAq*WW~fT0)jj6D-x0h&hU1gn9_e&~TNDT?+c2y}8H+7*risP9
z!*R1ZY~SWMFVOxwIgZl|`G+`eQCsLxNGJ<U+pDu0HMlZ7<kQDOzCK;$nHvMriL5xz
zw=mO!%#F~sN{7xE7pL^!#&Lnw1@%tABlp}*&Bh$eU&WC%y1_EbQHvAJ4z)1Ea}*{k
zDvn(jc_h-Slp@fU-D{h)7Ee)%7P&s+_+XoImf;*;MPK>`C<T#*BBvZmOkN{%i?=w2
zBbNUJj^BHX{w-H-Ts*;1mbnTxgtZ7QxqPrk*ld%#Zlx8eHC70;6iC;j88s+;A3G_S
z#$yng(irk0r|UOy(ikfwvqU4TB??1yTu_vnM#IG@#bi8WHaWyK7T0wcWK*{0WA5rL
z@!4y09&2v#(NjI5h9ZnQIKeW)T_Z3hYMx<BgLE~eqd~b~0wxlK<1-n>eD&frrcRBO
z#wki&aFF!bnGd-#I%Jl_WMzp8JhD<VEeecLC@nC$Byj}mcb#Oga|s8H=Sa@qcORWz
z2jwbkyT#J+lT{zOt~~373ql&Ct*k`lACiB2dldIAq~ZRUAOGXly9(;AU%$?&Q>X5H
z{i#oV>W-WV!1?p%xpwW^w|zY$t0vOh{rHs(kHirgU*;$;MAU-H^X}+1l~=Y_RtQQg
zuB~p`(pbE}BdpgL?H!<$q9_Us7O50PlHvw6kfO>WWCbM&s^y?9xCWHtpuz}rhk2Ir
z@OqQeEsw3?oRS`%bjZdrFWuVX*!O(~7JIz#<|{1J+br(yF?-=f2Bl@?<Qn(49m@Uz
z)(KebEz@e%t4Sqm3H*@a$g-<2`B7jd34EU-%}BG9dQitqhLDc1VV&V{!t*b^N@1vg
zvm71KfN#?i?$2WxS@nBVCglKCpPZruo8fx_vL2$XCAFZG#LY@Ft<ge~V5wO_np7Tn
zq3_XYG`PCg$8j9uEJI`Q9mPD(s6`Egk_1wbW(A3e@g0XkLa4!YT|8>ovSdHacx17|
z@p`~(qZs5Ou8=Sr@%q+vPJQABFbj*karISd;WDk^Hsedr5vK)f?US79)tSb7NENZs
zxWIb7MX4<YOFImSjm38y=1EFTR(ss^edcM-Filwvo2cmok&LlXo&Ipn=8IovChE+b
zu&S-JCzVt9qy|v}bp=HNqxU|U{C+LhsrutI0_nI_7DiQ8S@Y4Dcsi?sc3EZL@;t@!
ze9|~3h$6-rNYBS>DCV=AGF{`^u*-hAMC`Rm%96U5QzvCT9B^`BmqsH)ssQUtECp^o
zA|K5Vw(6rTvD93TKq|B@8BfO;U672X-1vhhiJ~^OctSkfW%YQQ#vJZH*2VLn6)9S-
zCMa{Vv>+}sq_RXxV4MO^DI6&f+MpzuGAGM4<Jk<Zze(fx{Rj$-3;~U>24O8(p7G}8
zH@SHE5~V4shU%=~p$9!i<3sxWeGdBj6h+3hYh&`Fpxs%f$Ss>&o2;*|p_D_nw*(j#
z7rHd-4a%~tyg9AG7)_p6_Y}u*KnV2vAXN7!aqN0Nq84JTCB~wp#Le{`j)gTx9M6#^
zJmR>{mA!*+!Estm!oP##1mAHeC4^Q~0j{WHOU+)I@!{nKPBkK48^u_;fQy9Lkk_|v
zaPo;yV|t5Ry!I;fXocoro5`DBC(cS%S|>QO&}16#fsELwUtq1?s-8<&I$=a?48G$K
zCmHISI38sgD`68kogm-F@vRr0t2lPw#<7(!i6Feqarzy_u{4!=yb$(IX25EL#sL|U
z=P7m;BZVMBpp7B-6uxqBtid9g2}=;w(K05@G6v%*cde~ZmIa|7AcY_+3+lCy*?f#O
zhD2+kz$3IZva-N65X*{}%uYZl*20jdb`H5bNqL}K2eU_khDMY7*EhKQg<nS=9B|*c
z`<TT0jC;pu-SaSY-(mCS4K5B7+Rata2{$I!xjC6seZWF;7|*FoK~@yhq5y3TR%=FS
zN<H%FgaKRg5xUvI);fs!gymj?lMg(ID-F_CZX)1Rjm566#2l@y?%lUbs((i7T?KVT
zQE+f@z{bV~0KfKYzxIyThr{8!Ys#&+DMk~7Fx3Z1fHDS86|@(+G*(w}>mJE)Oi`vZ
zI(6dxAu=oQ>LJtutP&_u>9XlU<Ao8eg&tZLlzE2otKPd}mQ<%l$Hn*oE()Dz#0OJ?
zMuSqiWV%Eb8e15YGL-R{h8wU^cW}J`S#KiFtg>)!jirx2Oy}P3BKc4Mj&Yi@w%BAA
z?{JUjQ`Z`$N|w(a2Qos>p)9F1*usD~&nh<ujvjEnhZKTZt<HFFpJX<t-s}<@P2AtY
zTU};%dyoBt!EN!Hqc`w7DJS=>mke6yWDY_jM5Xm647eg^wcBB3eTA^*F&t0Gb;feL
z#lhYoD#>XzBV;|oNP&Aa(OZ_9dQ{`&VvnrMNs0`=ta|U}Nel)Y$HNIiYADh?<KSSM
zcB4)zT*kU2&r3{haUDZEpHNdCCqjp&=OgPa@YiV^KgE3?{SYhne~AC}-v!K)jFkls
zciv~g^{CqnN93GXJ40zox}G8_U{>VxY99S8qY;LHVV0#deIHi{7Q#9=W(N#rGgjL@
zx>)-C1Dw?bc6a;CcL#Wf7Fjm1<XAHWEunB>3dp>=aeN1D{a|yGZnJ^wcqqp~xJS)s
zwi3QJ+Eo6(w!m6JnWZ>VQDhmLJ6mk-%((l^DIDu_>u`go_t)7C8r=8c2k{(`siD#E
z@g1KpKK%kq7egN1*yY5+6r}`30X{Cu@v7>ql+@}C{2-t#a<Vid%`|C|plLClCk$_G
zBKNjfb#hjgnvhierg}D|oEUOpNt^^n3IeCfT38j(ZZ`@1kkF4n3UpbLmnGA9&S-c@
zZM=&v6G|yTXH^-?XlC;<FTeaE*KS^8FdDGm?{n`xA0kdu@-kyQ8L+dv&6{stqSI+{
z{nk~!{Pb7(_{YAFlPB-Fql;;+rB<)eTU_MiiIbSpQkMChN!Q!P?ziPpF!p^ZzlxxR
z`6D=9?zCB1TOq3X3?~x`owL+#zJp^?505xjZ{xUr#BuQ!$Ijb1c5#A$8j2)OnI25v
z;<zlx3XRS!zGI1J<F|1fAZjf{utD?q89wmw53~G%5AnbLyZ^>4%~)IXag#&NyDoK;
z<I0R<tEbVnr0Y6l8pcIVx8`z~rPRX^fN6HrUnc|$VWZ-BI%T=lrHf^7c!1+Bu(Nac
zHjbNbacmUiK1`++$CYH>cg!;J5Y;&b*1_7ca<x#B!gy@=2W;QmWIjvjwOY82&+cHv
z^4bz*<TD=4Fw^SbT^NlL0>@EI=M%2%9pF1YH7_7@Qi`Ia9@Rk#iXx{>W2Tfuen@Mj
z$zVJousI#a!}lGOFUd+lt<hnsGcG3u<1E2WM%eT^qiffA{>d-1^YuUAXaD+NrV#|(
z+D*ypE4-Q~OoT&WJzltV1tmTDSxMkpnx4<WEM+MOn8q<6Xa*sZwCW4qkH?&DuCUN-
zu{}885J7Kgg>q|`^^GO!i;GwVh^kZ~q^e|uFjSi>|6KO>T?KXL&!6WjU-=3TJ@nAG
zeEq4Xp5m^%?)tW`cl^U2;rbVT6Gv)FTb=d|mgUtY&R=*4-?bPqXY=Ydt+r285Aa%b
zOq!#up&kX*be#p0mq$&!irPYl<e-mIm5RM^oIAZO()DrbiZ~llYc$dG0fXHg!gd$G
z*&+xkDT2vhOymYs_O2y5kExpgiGzte_P_SK91JJa0!3K}_O4!|s|yx<Mcwg;T8ntD
zB%M!@Qs8=$*<L|!xyy96&tkn%H47IR#%L4}MRf+{4!SH+o+7f6-JP3smp{Pn?ly6n
z@>U)GTMsyVr>%1zc$kaNeF-H>G%i94tcH!nMLu=@ZtAY25E(DsxXDVhMm-Fvwd&yK
zl-AITd~iK10#oWcd%UvPrN6h2FC|$~Ae^cMXiAM2`Gm~|`{^M|jV8r>$fccKy6p~~
zRtqnTaO8}w!GxacSEBN!B+2J&ls>LkN!MI_`YT)?%?ScURtPq4?9et-7Mu<XPK{n`
zgHTEa^C^xH)IG`OIOhEFBG)Ge+}CKJjV8`BN>ep}FGLM4>m71k;&~2TD|vNqhoz-6
z9PICr&t`b#9Eh;YAh}(KIegEJ(0^KM|JqIZx<HrUD2GP7MRT!>-)Z1P0e&8m%x6qz
zbBZh>P!gNW5GcmODeDUhv>e6F{yDbS9z~tB1d~1&cV9p!mbz~9_5B3L5u84=%*u%i
z{NC?=k&ot^oLd?psXRBO?_s6IQ7-j*i^%sdd5-h~JPTzmXhsX@WXy6r;_jAUquZkq
z)tSx{q#xqQCFZjc!c_BwrmS=^UDqY7HE^Q<-}P}-H4Kpz1$mw`8V?YnB=96kmuM@G
zd^$Cw(U7k^_Z2p8UFYDiPhJ#kte@rDwVQnH+23b#Ynu-~_z<tX_8K>CTt~^%eB{GV
zkQF8U{*Yh!m4C;Xv*&s6!H4N|I%H|a@r`ww^(IB3?-;!sQz_8{cP3wj)Ro}#dtB%4
z`yiLT{?s4hczt1k@4Ijh4OddioR@FjWT{c376sIrb*!ILSi?dTg6sYvj#n3Ze+<VF
zVbi(8@$PKE#hqPxM;vz}g)3)lzl~#KO45AJdf?+)g^7IL_}W*wK1m3pBaUzG(lS$e
zPKQM&VxhS~q$G#)DXtVaQgUk?bAEZ98<PX>Z8XrPJmR>-g*zN)y2L-?__f_#mKM)&
zu)q6F98Zrp&Vl}SEXQv={}n#6_J2kj2Zg}u%0=YH{vMxu`2{A0!8JY~JadAbam;+4
zG9OK`X^txu2lE8-teQl1ROKFnW@~q!rL|R3Z3#RNTWB1qP>v#>C&Z(Ym6b(y$3xbe
zi`=`h!NH)<U^pb7Yl>Er;|pyLiYlwAq)X^2Za68WD3G(1`#%0j&a`X9^OC?doH)JC
zQ26W@1$#x#{%D)I)+{vI6vk3!np5q7SGV^#-CpJS$pobwYF>aV6*;gsn{Zwo!&fd@
zR?-)m``3@Lv-<`st7~{s1i}H4BXEv#Edp@4lX&@aD5AZqpzbGr;wSiffA8<{v5$R>
z`|rR19k0Fk;*0$KzyJ67yMOoZe%seO_Q~&Oc=I~LS6@MDQ5{^Q<(><tk!ebqmBh(5
zhr8P>w9hc>4_Q36hW0#?(TKep*9jX<!d4sMR^qVMSa4j7^4LGvV`Y8qj>eVNx+<Fp
zg&X)JWkJ2SjB*1;!vnO=skM7}jT$YpKsir97_uxT6X3@lVNqpeocTU4{mxZpoip@{
zg7%x2xM$hp_|j74#peoiQ6i-zO;fgRjj<v?>$1}4k`i4OXl<&t;v?ySI7yh#=gg-w
zd|%=psl!jx`TH`(_Gi8B`PdWe?rk%={svMX^%Y3Xqo+>dq=q~%>CgALesGPGjR!b9
z9CB`B9qm^_vaepd%0jb6r`<ui9^z;M57)yw9v6q3+_!Q9b>y#^YfYde3Q0W-85cRL
z3(L5k&(%Snyv$ka_Gq;loG>NvJi!XXyoi}0sK-9NLUY<zgtIxn_k~-`+NX$3j@^8T
z<wl!Fddn2Xf&*Dz;t0VcPr2Mb#1V?PFleK3r6es1a&4;p6#`dDCi#fNq{hK?Lc^E%
zRse>1HpS#K?04#2|BqNF&YvSvf_$Dc>JRAe?$f`yMQ?4H?(sE(AfQZQbe@sKa|$JK
zP=szs6x8Xpx=d$Xo=G<t)93DU53(W7F&$0WySdHxeCiwrdvm`0yPLf6+8#w&@X!zb
z1zz}#U!xn%S!ilB7F`&;K;j66=Xh0PzgMGPmyC1EH?G`d<LoLcy&gvAJaqaj=mMP;
zL<&4FU^*VtsMWxhRdzkeke)}k+e0V^Hwv&tS=~52m+^c`I-Mbsl*Nr>c<p6`HCPRp
zk|NJ|?%A($@!~5?rxTW!*17M24<coQfB6glKW^UGVmf(}jg1YSdG706y>f;9-F@24
zCKoQ;!$%(f1VPy1*{?stum0*Uv%YqmdM#kFTSr)qvUK{+V8-nX-EHG{Yb{y&-i2vz
zb3DFr@f{q0`1C2<tm61E-sRf<b&l8Xe~aUw;`p`eSLrowbL`&X*g77@@whbH;@*|x
z@8noYlz5BdwZ#=&FW}msPo@jjI$c_=2FFcFa>Vhhh?yd&#Xh}KbH-Cd!#ThE+c%kZ
zP7~`4bL&O!YP5N@w@hISIFJ+tju1@qjLV0INGpg7L-8#+mKDdvh(S`PKb_L_CBB6_
z92Ya}x9QdSow8o|?ay=1y${em{Q-=!RZE|dJay?!@=)UW4vY0Q&aRx`>f}`xmb$!g
z{U$0&X!jPl|Lk4tj|L2e15A-&tgQmOB1aS%_a58eg{>QRv@t!$Au$H64bwQG<9XcL
z-6YQv&MhnxI6k-f`wWL;u1SZmUPD+(!>iG>mT4{Hz*St#H1lSIANY%>X&qjorhMe7
z`+02~^9TFecv3<Mq35BEq2-5MSn8rliCl%F915)o9S>irN;A?LN^NL+O%|dCi>)T(
zRFg@IKv56sNat<A8}R26!F`MMu7bLs`I(>LbD#Sh4?p}cpZe6NIDh^;02eP_<jE(W
z<nhNJ=VyNAXTI(0sr6%A_zR!mnawRke~J(eH6L0*LRsu#9XPysokr+aC+Bg(p}tks
z@+UKl$<d=JI?3_d9h6fILI^2wd>?BC(`ih5p@ozRV@ws`N${eOWSmeGIbpMd>pF}M
z_lS!=ez!-n-XxQXIGG~K0_iwtlT&0cPYr1{B^=C%f)%8+tSq)Tdwvbq4RD1*IGQ}k
z+3p{3u(!{6<nqvmHyF=$iGmhjDU0%sy^9b6YYe5<%tmA8dwm-12DOHd){@eSZxS5&
z&dtufeW{ZtdFT`0!<YWkE{ta&Ttd&#aB|Xoh;`u2t*dmrCPEv^dCZOC28lJyCo_sN
zCmT%3moi%29<J+F4i!=o_#svZ_TrdhK?_$YN~7UuG`vxd7!2mbdB#GsN8~tM@9(p}
z*QeF%vC(WZQ4V|a3Fs0>DKcF!HJ07fFwN(9(=kD?#7t>cTZ?@1%t=BgBoYqFDVZc0
z&)huV#^@$-?C}@wy_+ku9lCxCtR*Q53TrA=dvVl<Vl{h{F}qv)taqDq*Lq~7qO=NQ
zC0c;BIp48I`fpl=HfTrT*E|-zMVjp<li`?y-CgGM8LOvGk>-Xh*C;E<lAOXSWG$q-
zw1mayjnO&wyDsDFbGrF1%Uwa~hIAI|%;G&ZPIviq`z~&7Yi5H9V-lvz=ea&iX$M#F
zY95|+Dq->?+oV*0@)Z5UK3{)*lew&uni)pah*yqtc{1TlP12~<2x}hkG{!k98%yO;
zriSrs!r8OODYBd}ijLI$0SQ7$rlTR{G@-uOq}g2}ZLClV2V)DYDcRrI;-wd#V{vhn
zPdxr{PMkVNQ139A7kv8DpT_fCu3o*ulTSX$<;z!?#|iU!!o^FM_~=I-qb%m^jVC<x
zzz2Eg!H0S2mFKzq=Bu1|;xWpy%D~+Y-n2gIH@3DKOgZ{~k-zs$opb7K98Y2p4xwkL
zI|XSv#Jce2<~7=09jPtlaK??|2C1>kCR2(sCml>E;*@4@0oQR*QX-`y@Iy4PAIGe}
zons{tl6P>t*y>Sp9IhSivtLvkA8WRmNX1?}eJ97MWjC=*iy8iOf*&j~a|~;(1-|>t
zNg^kDi{m)s+08?)4{tJ$9sa^S_i$~tOUG~g(HtAi&Sb>S_8#k<Cf&6KGJV9csiuFe
zE%+aW<MQ%lp83uH#;1OI9oluQDTqtKr1S`b9wLVeC(bd*6O58Ly(XtzmrzLRofe+s
z;V-oDb|XdydlW^1Hm1s+rZEawsYTpOMijcll@heU8q0J#qqnd^%@5fd?lUV2&MvR9
z>Mb+e-{;!+khUpVtF=LSl)4~vJX%Vy9!Xk$ovhU+Np`3gid(}&e)ElM%*uq=7L<iU
zs}}N+lgpg#ZBVb3{MrlGINRLdFdc&sM4pFt<Wp{}MM{Y;73=LT8^_nV*&ouswU6%<
zxWc||7PT72{7zZ#N|n-eU7mdMNq+9<eva+!ZGQgef1aQJ`JZQNYm1-zxu4_7C!fT1
z-EaSwyMp@p3mkvqlUQFM6m(hv-GvCJ<B+El^4W~V<sL?qxW1&E&&b9juo_Rg*gPd2
z4k)HGOrBBX8EA{+c%&9??#(JG6j4oNS!+>3Aw3r-2ocs|Y=NkSw3pZMH59``C<?r&
zMi4b|ok~i<6#}hOvUpBD%V>3atcM<}UdXVw#C08#8H+Flg{}Uc*=T@GOYXaS9aKqP
z7(73q$P08)R62S`6Gx`hSYvS=hf?Pxaf~~kQRFF(<6(uoZR7l|Hu=wU$w1IJcZT(c
z9|yO}R=1h~#~11ZU59BpVmzO7uD3!dH8o$7%%)6-Bdpdmqyt6HU@%}lnO0V6MTRv7
z$Mq<!%f;agDI}#(M<pnWD<snQ@WUF)T1uPa)FV!>tkX0w8yq6?oJOriuihk59>Q26
z0lChYr780?VX?WuO5m~N)i`Wz@NyY3))uKV=18QH9L^6hh2aw?PEuIItgtkKfO(#i
z<pm{`hO`s{t2Ne0JQa|dg2VZoC{9SS43y8*tut{YmTh46=e>R3)(RR;oT!EgJgo0f
z+MIej;`o`9lv%;etJk@;y~}KxFdOE~Q_Vch*dNX@LK0^Uw&gJzOChFI<ax?+&yen?
zTz+jIOP5#QxXj%jT<5~Q%bdQj!RmUCdmcZ}p;%y^xCrfFoN9i)&;<pWv?$o#+2Y2n
ztE?<WoVsuhiO1^t2@nDM!zrpvh>HxPB(uqY!NCDHZ*24Oi?`UkHRZ&qGfa~yT1tW_
zq|6H>pgfPFD40(U2_lDXr(I<r8VXZpm1Il_X`b@XLl5$akA8v+XYQgNxFoYN!@(Y&
z?{nWh_YwFWTU%SSI~_m}g#q97*kd#sO?J0$@cGaG3jhA!|4V-73;&&(AM(h958^6^
zq9`bi%D-8bRfAz~KO`RA?`}`SzLzBfN%P!U)<5_-IQ9<5Cl+c1J%?E~W;CC2uD41l
zOKQF(olcnyhnU+O=M~5Cc!JK~;@EXOG!@6{4{@xJzK0h^Z*km+IJ2@&vpnLsD5%%#
zcQ}^D5(z=Bb7pD6Jc(IobXg637My^?<_52oH74e596N&ktdA))pE!P!Qdp*iq3(x&
zEXTeINcC-w<0Fnerf%bi<1PM2;n*aa@vE<K^XV@l3IQby#|c;WXCwy&jj&6%(c<-k
zT^y^JrG~KCrnRtuQi`c55x&pD;sUjH8|4QCZh(-AqR3Gg?(4PC#^6|e<k})Aj7Fy!
zX*{D80!R21lNpybZ!mHMjm0Ib=QEj1*`JP>mnlXFX4c|4KFu(q9!Bhq`)tJ-`(@1H
zXp8&ZIc*7A2Gp^b!m!j_LV7O0b@dt!ik7FCqzQH3C-A+h6VBQysB;uT3NoD%DvzZ`
zgIO)AWRl7pY07Fy0+)9gO8u#>cl|%{JddCK*`MWw7hWKaW8ygG#TQ@XXMgr*@qM3H
zUU}u)zLr2>eV3Ep`$^jOo+mOXXU{JXbv$%lkn{(HttL(ope-oZ$5@FU)(LAZys(9G
zL-P5Q<nRC+$JnAo=NU-F%y=Bc<}FiP#9KE#DXXB)t<;@OZV}QYSYE`jlDJ<z^t~`5
zYIiHw3sXs}1fkEu#s+15fz7QQdi4m?UEv$lxq2`oGZ`usU=&3OouxMI<u-%ysM4mh
zRJLH&qNKvw+wun%B_!j~nAv2C%?-|UmvZkF0?&P4l;QuZmxD#SE*C!bUDQvV!^>jM
zo?fKW@yYX?{^23rS_?l6$ql$}K#rmn)mf~!Xhtm@*Jm2f==b+1=QB)RkmngfI*hbO
ze^%|WlCqMpIMV8pQWAy{!t+5WbZ!vRp|i3`h~l7sh|CJ=VU2F9OKvT>t@^^6fy>#I
z4We3?o!w1pK}gwL<BPV=^9NHVHp7br!bozRbFAHAz0>9LWK7-nF~%^-a*VOKvMMzR
zA<)L+2#16Cklpzdof&Gg1M-6_1YXq`?u9jKjb<g|^FP3vO{eHWBczK8LZVg^D?lpC
z#>NVyF|Nr$Yx2@k&NWFQN!>0|8sgHWeR_qB^#zXK)g^HRuOGU!7F<S$IluOwpXD3R
zj@aExx&Pr+LI;#7xPI$^G&s&AEhxyaQej1vv5M0fo7-0zPY&sHo1FgS_u?-t(%Lx1
z{h#_?I;)FZI@l#K64$tt^PGdNZAQC?Ob>Gw7Z<qe-erc9ZH$3tyG4<u<naulBsfkr
zy=M*eR*QPjB$uFViY{VIk#MlP!%Htb%l_6T7caiT3$MJu#Y?aA?AN}`Z-4&3@ZbOI
zuduy&je9Pf;Ya@B57BBg5m@d%e~t_1&rued!QqgVl@+2$@x~j^a_RNwXw?F&DeuUa
zRR8oHQE}joF}N{?o10tjd5;||+H<-4;~$}Z>dX<x3v@b_#bLi+aU2|R>^MFtl4e+A
zvDTy+HgO!ENj&4AzmGoRxX3evau}5^2e&yE68TLWM`5jU5;@{nDwp=^B7u<{9Q1K+
zbKLBZSwn6s*_4Lwa(4L`wZ;OwJDW74Iz?-p-?MdIIG8fE8Ez~PMv?1+jdq7)-5!@G
zBkG=ywf2wXSP6%NcyNc~Xx67VxJux86~{r1sNVP=hT{TIS;0%c{Unn&Ung^Gyz%BP
zhy4`K_gPw6q(2>!%wv2hDDs3*XvXoBz3GTiJftjAnzbg&%d0qzCenBD`~a&9x+-I-
zUQLIT#$YVCN`a^bd*b;Fo2Ot)9A9B{&R}zgR2Y_*m(YR7bUvj&>JvzXF9fC5jMJPq
zwr(&R@3OvrjLCe0H@L+QFy%ulEm|@luUV|p_)2nPFyMOsfD7#|lPpCDLEHC{QcxU~
z*sDyPptPEm`XX!f7Rp!5n=5$B7fA9Pdo-M6&5_9XcgT8QY{I6~DZl*7zsxWE!Y}ak
zuYdi{jpa{V)?k#tUtHnD_kK71n|p}n5-)FB+Db!~)9&@q(nXY(sJlRIajDYjvj$xj
zxM>YF*e9EgnM;A!=^=a%&r#F@pZ$JZ8P*GdRMj-fZ5IXMI@Qfj8w3_xXb@0aUPMgB
zq?0K@cY$WBgDwlgS`8P45egMJY+b&|xPa!k&#~ioF%$v6`RX?a)dFX_4n>}l6^2u5
z8w}@Da$ORr2wj#)C6QDqUW=`4kc`DuioD24W^;72iC22;U%AG?L{ue-|A8i7p)&q1
z97$tojrGSLWxR7hQd{J=wsIC_Nt)%H==8{yhr+VBxX9A-(M`}A$|9$hGzbp&8BHhb
z3rVxLfCxMy$Dtkg?2U(5YjK1)(#QgHBz`D7w~`?!4HAnjO00n8<wfLV#9%z3+goJ4
z*&!<m7NUp{NhutH(BZ2W-z3fi(d3Yg<7bGK&+lG+hNiQ~V_k>1NXbmjg{2Ezollt=
z&5BfHWm!pIP(2K+tsb<cHF(k?EfU7*l)ToURp#v6*k&?2`b@h%UJz2`Il2CJ`k};I
zI!E@;zBC^->iB_olo>LJs7~a#B$ElHj#*e;WdC3v&jGhtV?2&|V=Q>|M1bo#6uL>&
zY;tvLAB5(yCyrCjrgX;#6pwUxdGmn5Zorej{u-2mmE%pWU+Uxg0h>2(a$(WOm;$LJ
zrP0ij8GC!%7-Q-77Kj#B=sx-h;?Mma>le;)=KKffEv@jYS6(5_=G5Afoy(idW(lk7
z%PcIdl8B6h(Kdlsr_){lVTs2RFb3sS{kyaAkkIk*Tp!_hOp^qqEus(zld`{egRgw)
zNzx?Y+}#hccI*Uco|Bg)N|xw+PA&BLkstZXJpJr*+_-Ur(D(V+M?ON}2ORYK49BA?
zK$Z@Ri_4t9>nx5^7*pP{@vE|O0<Ei3mpK}mFvgJQIlud*ufFF!zQghI8tadLnDOo*
zNo|QIw+j|TNt$IG?{vwPhs3h9w8-+xx8S(H&uBVjUkV!C1w`Nzxe@I!U~jTpaV$mk
zu>WR`kM7m-wkNcJm6at9CL<1SbKL5Z6(tL`w{z_B^u;%sXAqA2Y#ckoT)KSW>N7N)
zB|h9a;#g-~SUS&@c*aa?Ry;*k7T=O%Z4Ler$CD$Dn`OrKwM{0mtvL34{2-((a`JEY
znfMNl|7@330=G0wwr+Csi(jI>u}n1E<1@!S_KZsy23+qS;FcL0!QS33yW2;9mykG~
z&n%j-u(V32)ndh1MhAP4X2f~HAW2#4G^+ccsrnTWm8=NXV)JSM(mE22MEI3+%+}Tp
z%bhL@oi>LDJKP-g2|SPe*_7BC97>XT&J*{ZqaHxqS?0#Ii_}fV4_^2v<8p_m`UN#N
zBv6W%cdl|*y+y-un4~eG>yT+fX$6HbmHU*f_B%5Lp%c;x8~C#^BRAmKxf8T&7OC7j
z)juKCpJ=NozInb8-ZotE{;l^E)P3noU*Z>j;TQO|U;8!I*4FsZAN^7O*+2Vd-~I)x
zKv;!zJX-4~IQ6NIv9z#2ytT!2ug{64)3{4*lHHp`z6Y)XA;B87sm84x&!yR0CU!jL
zlK}{DYEAmtm~3{4)f%Czn<7r7)NPO$VL(WvtWxSix{!g@CE9`))e&TvGROCQnynr>
z%doD*I*Q?V46_~9k6oZhEeqWa-}{$-f~U_N<F()XEa7BIF&wgd-y_&o#Np;8T)&R%
z*HFp>Y-Pf%yeiN{Qdo-+C1^!FO3}*@$$-YfUEFx>t4Bi>N2OZ(&d-B*+f=MQ`cU3}
z<(+}xpY6>;7?9RMxdE+}71kdAFpIqwyW0ac4~IOsaGGGL!Dw%brNBeD4pLNJpxS^|
zlECvg)>|e}F1ynq2ng#<cFQ6CS-&cYX>g^3a9xlJEhJi0k{U`nXtzX24=F(Ff*eaT
zsv~WRE;BlQz(%u6o@E5CLc5Ba(-D*8CQHj7Bu_0XojO1CKm8D2zH5OOe(OoX@q~DO
z$ho^7M@KELZeAwvBI<sG?|2Ao<%O!e%H;z^YC(t`VH`$rLeUijGNjcxM*rehNRkZY
zhDalDR2{e725oReiE_-5=8?tXGaAGg1GqJC{OS$Qrl9|9okZL)Km}FXg|QaO5_2@}
zF`mze!jOgKB@QMrPraHjNa6f~&+(3r$y|)c@FP&u7SrJ#3+oXJofYD}7T?`y;68hm
zr@ydA=(=2eWgnrSe=wod>@ol}F~X?>1z8f)KinrxGkU!R!m!Tjhrfr$`WfzhJf*eL
z!w>6pl}i|e?Cftd-rr+1j5&4JUDO&i`r{$<Y+ebGw|fXBNau61B%$5uBHgN|Cr?tO
z6)5G9Y0b<PwA}`jC6<yf46)W>csSte&pgN8{(#b24n}=e)>iq@qaR|ax51!)$XCAd
zRnj!&-V5isaNz>iZr$Ku|DfvSEeet(q1Wls>-9LYbAKy{77AlDMe$bF&X^J~T)%OP
zmtJ}8J@4^5I9^?4^|41;TxhekHQ>fz$b$=~@E045_P75Sj(v_TtPnWf9gc%qlRf<{
zICj1T$4XUBBGN-<wvv<iRvfq9$?>h(h;ed*#ifTRQp0Mu&JX^@ALL8tdVJ%_-yxh#
ziD!MzoO={gYjbt;3V|C^JK|VkD2=$ou||>H=GZz6k2nrxNUM8-{-rMyCppRqkVc?X
zlUln2+Mq;<<5U)FQc!U`OfcGlR|D6p-tcq^dhnk2?cFRZ!HAr)NEmM2V(W`vVsC%I
z@#71eURdDS{YxafV>+wH*-bKRR$#2DDkMsxN==fbT;Ja$DKpku3t%Ccj#zU#EHxH!
zl#5YNm=crcU@F&lWhEF>4KrFmXH|91aq)#<Jey)fh;jo4Ny3fcK3)(aa*Zl9Cz~DG
z^^@%Fynq)<-t3S0)nEQkeCoA59{b@RCT_K<Hy610+S91nnDZwtkYsaa>6}z&7^<Nh
zQ(7De4w?{wwLVsqbZZfA-;(H<VgCw?=f4Z(R)z@n$dlR{bM!11ck~>qAxMK%@+eNP
zbStYDEBCDTd%f!`?(XjLumAPG<`;kQ7nw{Z{P>UmI9ZnQ+0TBKd+xdCU0ussi$sDF
zSkLG5efQESGK%>j%IUGV)Wi)WzN?6$5Un*zRzZmr5=-?Vz(UY!Vkt?cQ>+&34PuNI
zl_9a#NXJ2W9##n~23ac1kxsX@0)++F!8jVNON2JKK}c2>C?kl1h`21t$`UIC+k4w|
z+AeDk-^-=#IiBm%ZnXH}AOEZTexCCOzw}Qy)?Q@gi67ym{ha36PY@|^Op45B$YhR9
z#+Z0OX)Hoo&?RxInaw1jCvkiSjVA7IvUlqi5{s(@k?Z0`A?<F5g~dfAkft$dk}!{B
z29pVs>71%9<#rJF=abpf1{4bG2(0UJ`mPO*YfUnn6Q~Y%FEx>_rs+9!Yc+DM@nofE
z?nsHzl0xTTq0??+C^?+YFeO}!XO%OYwPdBnaUDG0$2ip<JEbB=6AOVgE*=))I%r3e
z>jI$-wJ>5*6nHKyhjn(#l5tUDgy4<-EoxDN4?J`?uiQ+ic`n_k#!vj<kK&~{zxB&M
z&uVLl4}R<id1+76Irk_v0g=s6MS_gy=yZym4N0xU3WLxEvqCe^6g5W>`Ywgl%!hsE
zx31D|3<*4e?}j+tCbh+7y6t78Fl0qeS{9^n!ep|A8}6e^!*poKldLk2|I?D2|6^8>
zX5cE6QYazNR*<I|>3ojPbNcf+^;$$BeO{RiS+7|>u-?FP1X5&3J4KJCEG_le+N@I~
z&^*y*aeW!j4Ol&Kn#Vsp<n@<sp)<*?t5X)6O`d=521{~+;}+<$M2M2Z;Q@ou0MGLX
zf{5DU8tq4)0H?vdANn9}t%-3R^2wOZt5=u}_j&M>Kfsx_lO(VF9-I4HlvFxeQBX%~
zO`a6wX^i7I1hqOy$>`vacsylcd4+njNhUR-*~W?hVN0%T-s0!~_5aRSU%E!1BpWL}
zMe0%*O+1Tv;Tz8*te_S)`JFGm%B9QKaFpV)4?jw~-DWbIkY*X4g)~hu+OWF3Tm?!;
z!w}ZmqbXN&r@u~XLs@7n)oWk){V$Vd*?Zn&^GzI|Ie&~}r6!rp2vnPUmRl&#ka#ZL
zTKFe&tSJ~w=jakH&c7Xw4T-Q-KdP`;*F`(!x8PW?9M&q1%ZlUI4{lP68r=8L-Mo6l
zaj#bACx7_=#INQhzxf~j6|1d99)9A7dG)~1JO3CpVX5m3l_$t#j!q^O$4G=QpbO@O
zW}Ye{M-ckXTO4m*rPCM?cpW_DquLQsZ;8&^IL>rQ8Yj%A+qmH#rMAq5hCKF;?%D5s
z&))5lLQtX+g+=5!7hk=`HOC=pHc0d#FK!Ppxni6Z9M0zWPIZ{M?JyzLZD%CG;b4Tq
zW1-bRxo~g2fvW^cNdl+p2hGzISz3gZl&;1DA`22thD7ECPNDFXhjk>=ct+dx$fRXE
zp0eh;v|Sfz4fijfWHwKE{^^%^@N~p1%Q=`$`B(q(-|&C&UH+&4>(8(?+~=#m^&fd~
zMRD-T?=iK4sS444gswGk!ZxD0KtNeVxQ<K1Rn$Y6m6|v!aD@QoKpDTHzOYQ=*a@^&
z<l2%KIb}L$J~-sCzlTNOdm*705Y!?X%`Q>1fm9$Js>Bpfg?#V-dRIZ+XFl^8zVxLp
z@q<73gZ$%v{Ezwcr$3ESihuDh{>A^HUw*r%466}H>S3L_kTlNUhgZyTWXkB)>p0Tl
z`97Mrl}?W&K&+5hlcUQ7FRW2!IYm4{9yMo1zKe7mj3^OGfsvGDN|7px*kN7<B&EO+
z8D3aYb0vYO(!V$kHq9X`aeNo)D2g;Ap3KPegjTCTXMLT9mov#TiZmwZE_31WPw>>g
z`R9y`=HT@keCeq@tP|l!0Z)AI12nrnuU)@N&Do&Wa#>&TxGTGgy?B`<Q4Hq_<+=Eg
zPihM0*I%ZkB%l7|_wbRAeUuM9@(9OHo}jzXqgJaUuw+?Eo@JzI%I5YhzWn7c^TjWI
ziEn)4MGgl;%2NNigmuPRT;(852`HAmh}9xv<@5%%GQp80n>#o0vDAH^641t0#_gig
zj51PIlgC9&Bdn2SIn!hY){qt{4cEnS9jvgGJ&Tc~MZ!E0#B;?!drY;&BgYFgi>{z8
z6a@+#hce0VvXU_Ha2<z!kujc5nH4#msK?sMDtj_#KQBo0j7F`=BM*Ow|NblgigB6q
zwbyR(%-3e1eB8j}vF|=h&~mwXV~3WvPPeOAUvUxH7Us>1q`6C+DuS9vGx8bRlF6-$
z)IE<Up7={V^4Q0D;K2`aY-62Pt4Sk_5CXEYB+GNMJZEcvo3B0lWxo3KS9s-_ml^IK
zkY|S^BTf8gywRJtc8DVv@2D*4sLJoK)@*<+DGm<Vy0y>5DB4Q!fu2X`gC{(+ftoiW
zz4|<z@BIwcgURrK&25S6`CPv8Ix8#dtgo;0#1EY2^_Oq(`fHatI9y|Vc#E^eHC&Nm
zw9h=7vVX8gnU~aSO_bx){?I3p%>~L#qm3mDYhWeCcpo<%6Lx!CZJg$}Ud{N_BR`CK
z=?nOy9RkO#BtmqFJUTo!j%0O;B4_*7P3H52jam)sI!xjbxw}SbB!>rk{EPqX-|(ep
zpC{LrxG2#FhqM=tQIA5rv4>6zbgr3AW?X#ZP2xDl^L?-|9L*^T!+ahynM^5-B~LT3
z+zIGz_XZnnF~(Aq1zKy2eoM;a*5)=Zy!_s?bWjCc-^}rv%vd>djHpO(Wy!6boA?;&
ze()!7oX`yGWLd#1o?&lu?7YRXbQJc8<2Vt-v0|t_rlr6WIc}g?bSjSX+Z@Lko-B?y
zR`ipMNjkg3@%qXddm>{m`zDV6^S@$T<UIZQCeJ)Q1;@t=JU;x1vjlCITbsMI{bO{O
z6dNlJGTDMRFOuXg^Hky2T$)k9<cQ-)Ieg?JKg=VKeS(KR_%O$gt<!F`D-T=&Sy_-~
z6~~+VTRi=>Kj150{W7mT`wHX!A$c|+8UNXH>?lc5mRRWkn(cJXrsuQu#x-KEMy?%B
z96L>lV3H&_0+g(V_l!NNN*IIkTs*hNXg<XcJQ}qoLRbRFtNLGy0*}h>(@I0=*QnJj
zwVI}@GFpxy!zWK1_6y6vYCJR|FL0xPEObedjIt<cMgg5xgQg#GI6vgp_D$Aoz^d;P
z`XSTVh_8P2DSqUy{w+GL;^PmT<YP~?pv=gUjHiC{b&|uJlP5PAq}S;0l-%4dFwsfG
z-S-js1@%yqOfyE)gtl;LpSq8^fv3LsBwu;zDV}-eYh1r}oxyO#WIn?Hf#VYT0YTt#
zd}V`=fBXrac;dUc?}PW#>~&CH)x!rzb+kL)ukSs~dsji-=Rf~>{?_06Tl~$x`8PRt
z?%aF+<(0#OQyH)ua0`o=%}KH$t}iH6j%!K`qH^Kz+{!4pv{e)Bk>;D#2J({37`!N;
zC}L`!qF<IQG^5J;#BnK1Ng2-=O_sR2zr?MghV8DiJDAf9CE`|}wmswnD^r$RLqf+A
z)Fa9`18XZ;6szg?w{abZC~Oc#4L<Vyk5ZZ(<p5ZgPn=}&<Vlv&IfH!6I6h?Uo<#%_
zuMyB}*BOpyeErQnSV6ZJ@zWm!Co>FY4zo<sQ9eWAvZGzZ>LNe=-5;j5ehNPd37R2B
z2oMUX9302R59@c{)?+75@Yshx%K!cU_A^|*a+zQM>}UCp|KUII=H)9C+T8BM;m;y_
zXQ}jvbY9?SOQbC-O&Jv_bq`8WQa6@T2$TTfR(q_qB`r%ROG>RVy2R!sbCXk#d}c*L
zTS{)4nB&bB3W;z%=B6Z@OxT<(@XD~m4IN-ROY}!GnxRA4pU^BPe5g0$Y$K=VSbEJm
z=`5){DV4-v*cxBQl^$)sK{KrLV?Xo&(>x{et3ILi^<%6ZJ5F~}k`yz_ywAd!vj`+M
z3TSpC4klN5<=QR+L9^}gm+#lu+%QWOX{w;%u_GPc&>q1`i_d)b57Sv*Bk(=)(64m!
zghVKX?<fM#ho}Yujx8<m*nJQ3*MIVV;mXZTe&h4M!EgMRf5VmMUd85+mM+$HAschY
zMDkB*t*kE6@ZIWx6(`JQF=d=%Yym>iXgA5zlzdon%JEnTB#x`_e1{~>sJS`y!^`YE
zf05HC+r0k#fES+KVrwg9Z-0PL&$G1Vu&^3(Yjcm;wx!WDoX}%B;v$i!&?aYpe~;;C
zj^}!Weu(X^(tYS-kmu~a`7Esy=keSCV@kGfT%slgr_P?|zkdA^H*OjZ4oW`#JwL`>
z>1FKI=OLS;Y$X%odmd4}0SFFuwwaA4wA*b=QczB2%#|cuJ3~BA_?6H8I$wG21&R`&
z5>y|~6Tb5NGc0ynv>FQO2wr~iC9Yn*$==}*iQ%4er?J{{>*gl?!yR6K{WXT8F&8e}
zP0bJSTpuGOU@GxmtueZ4GPkA*@QR|m6ZHM=Q-463zCSip6~_vbmpJB#<Kz!>tiPG#
z(vWI>hhwWtOkOZIhDPKwFA`efh~riZCEnsVole-A^muL9<!a%h+lw5I=QIO_-kZ|4
z6F$_PbEc8gbuGQt+c{Qm<JhI;HE2e4e(W>f#WYU|k2u~~KhE05ae7ltl1(V`KE3rb
zU?nE>X?AMtkFN0Y^&JF)R@>tz?$?kTX4!3yefFip>)It+Zt^3a{9(GQ>x90K4Lvf)
zMK}(Qqi~fX@O@elAmI4&GT-%q5ArvE>TmMq&6|AgbN`k9{$Ktr*Pnj_Q$SXFSl5AU
z!k;$B09jEYeV0PQ+!2&fNY{;+<=`5JsgO9<qOHZ|l~}ZLB$4BQ<Kid>={U&1r70zo
zSwh2$7@CsP%H!H3L0e6h8*0)eXtg=sI^e;D1I~FVYOu><r@<sCQEP1mqQ#dF8eE+C
z^mD=fXhPJ8u(BGYQ7BGyPLS(@VKQbkpYU)0<L}T4BI2Y31UGJ6WAplT4q7ge-@)~l
zIk@pU2!os@j1H%)b}sP2bH}h4MuUw1?O6rBq!E^cfkuo|R!zbEPENcv;Fmx5%N%Uo
zAW2h1oDq~Ilm+FH!%RZTa5e?t%GNet{Q5IQ|NQ^U-KS3RV?X+r_>uqlCpmuZ46c+&
zX{+`rtKOIPx_1@S{lOpn0l)Z*zsSQ6Kg`{C-_2k7D}RNb_=%tRKm5xrM@^^NVys3u
zqH03c9@djMu2a?7^OEuIK7K92Z8gDnVP2w(vZ`ZCf$)56Ji{7`=Y^D6hOvTX;DHfn
zky8u>H}}r+!sHm)u~RG@?~r8%V{cL4Sf<s!ADI{Y{<F_=*L;Wj7OvB5*1)YMRbAIX
zYfYA>G#V|Ga&eSPbkujFtwkD*4MH~Ve+c>X@3Go#(pfZo@`=ZJ=#dNT?(Tzy@BRLd
zaPZKCc%JapufD=4m-z9Fv=sDfg6lfwsjFL<#s;HuOnLRQcz(db;tD5EoMCx+o#mw!
z7MB)ry-Lmc#K|*sT0Mkv5emY{;l2;t&)pX;@K=8Ff9Bu*+keZy`o&*jYiIAM=Kp69
z)LDz8MD;$SQI5cK9i)*MN8-s!pRdRYHn(<Y)@wxV7Q%P2vy!qXFuLlg5x$R}&(Oxw
za08Me#|T9yY=9NyHfP=!ywE?+bJ-$xZI#9K4rywUW}C+9BF%OPdC3=Ec$KU1Ay0I+
z*=RPfUWT@Yn(Lx<Ns<>dgBG52@Rduiw?NB6DXJ-D#|t=p;Xx)(e~qP1lc?kJ_{SdL
z(GQ*BjavtlM)2L=_aI3>CP`Ah_Vi_jrJ&5F6k0Nk1h1Ny*RF4(qhqADz})yvq#M%g
zEVFUqI7^E?)>e8f_u4p)B(VQ~>i#>-vg5kXh2Pq>)5)Pv=jq7-0}OHo2m&OS2`0(3
zOkGM03RZBC6>RC#wJqC{>FXfNvM5rCq9jTrMUk>dF^D7x0z@370S1G~(-S)P$!W)`
zx__L;U_g)}APs!|t><Yx{Y>vZ-PqOd*{fEq_4^6-&dxBF&!0Dk_8phYc5G+wPyaOU
zdH1{d#Hat9Pk-$9IDT}2fD+Js#n<@UiR0wdNNUw+O&n=dYVbXqT*1R&k<Df4bt5uT
zjEwD7jN{paaf&N7o3cyP&+KP<`}eUr%(4H$vkU`?Cnhks4Av7IquD%DrD6KyyM1Q*
z$Jw-5VLKL$#wu&etDp>i-o*%*d&9f1Gg+F;^E4lOjM>X>Kv)jKq%0nIoNZ+fF}s`A
zv316FRajl^a`ywL$-d)y#*0%p#~&nZEnq1dFY6&4pD-A(wzM>|$Z{MUM-oRv^ya-d
z`56uzc$zQWdmmw<c+M1kBr>3hm142pWVH#Nv~W<Yb(V>=A)ocwv1K!byhjrD$Yy-B
z9#z|}zwQQZxaL|4#o|b?*D&hQROSVhx#xx>j4>QNagryWditWR@rYxKVWiPUA*H~z
zH#n9St_57kaUoBxTtfIR^aIk3qv7w#u@=O{3>t!`>zmmh)KJsYOl}?{jtrf%OO!V_
z)?v(74;*60FvsncC1$?^$HO?L=oj&%#FGxS+Bjtc>N$=*pY40EqW|!ICaPs}WtZFT
zxR&d#+s2V|>nJU_;|*6cSnm@ILmu6Kl6Gp7M17LfqQAlM;WG=U>}Em}qRxE<%l0Xh
zCz#nhcb?;^T4mHqWJxZWnWmb{oj-ufxGq=h+{wOw{xiJyAH9b^`m;~+nUDWICyp-@
zQfA~4_{xS5ptQz!EG%K8QcaYk7)v2iL#z!*3&(faym^jxug9R-#Dp=TU}L)@4N<g4
zhaoaw#C9#ZaX{7`4bU{A0Am?kJn|E8-NZbv%B>(5&arfA9p4vJwr{3+ZiRMpfXsNj
z{f4VKnxEir>T{fJcOdq$g+)J(v87GU%h3+jiF8Vs#;mWc&=LY^G=`Ba;jyP5qwwli
zvr=!;Z$#Yvm4|rXfuro(Hid0zzVi8pDNGf~<}=)K>m^j(6vr>&NKIj&xzfsW#m=qB
z{wXq67Hix0p_71KbD4AV^Q^3_b9Ql+rPU^3k}}XK&zw6)z2C!Z=o=2Ah{ul{;lPJJ
z#P9vlAMq3Ke;@Dtu^(q*b{a&A5Eo}k`eh|`H{Em-H{Em-zwsNt!KXj{X+HL`kMXlV
z`?F}R`P$dM#@yT-g+k#azn<Bc)YCdenH1p&@?%wo9mq`NP_1R8=OBxDL=X};J49iK
zSFI5A`w+z_tx2^)VDMd!WH7+C9Rwl5a_EPKAPG>ZU~#p^qrqmzUUe;t3+wFv*0T(f
z1f}5YnJ$^U=7v}AVeWOe(L8>RBkP7e;wZUd7M&&=UXRZSd~V1FI3AWI>9*F8*$ky>
zf{+f|F1dmu4}6|tcO9ot;4eP)brzSF*mva?qA;RduCR62W;(4t5AT1L)mBW@?DMRM
zIF=~-Et{ZUr{7<~D1~oZyypWy%s=^=UtoN47T0w~(=-B(9C@1m`rm$ye7?vPmtW0o
zuX-)pwr$6@ZH$me*JtzAZT$St{{na3c_;tj*FMDE_uNaGzVbUir46=aNL5U#6QnB{
zw~N#zxV0Rj)xh-}yiyT+7|^e`=mr7hT8(D617Sq0N4tF?3>DX7Fc^@rZA2X52#bCg
zP>%=1iQq`R!ozBk+Vy*>ue3O@{}@4>qLg8Iv4fWrTy?_^cD?!vTBqvVw`RCi%rjBU
zq0<CmZRjSM9x#Sw2`p(Ngrw12Lpcs5zlgwM*Uo)>{ef@bbn1Ay3Sar$ewLQknBF_d
zAPAVKjB(YTO+@Vh`}dz@xfRmrHd(RyG@}7Y;FH$-B)z9cUYDNY4L|hL{M7s2&*WGY
z&$iDSK|g)^H2>x!e@G#h=juy$@`f9(<&sUaIC3O>X$gzD$w_|xr+<pq-1%BQ^s67@
zzOUX*Cwk@Acy8M^azGkHSeC$ceQ<16&Yq>Sx`wa~nM{Uk#%7`vFjcFNu}4EsLfE82
zQ7NbFOb>GM9tZ!@8@T<>?X=q=r{~ux)Qb2Sn;m;5$tl5OM;@ZI_9)w@PhmS2oz8%T
z#RZZe!Olutp~+uzCABMWA&Emyf8$G3Cg;c%$5ECfZZ}vz^e7n&Yo<ykUgyWZ@6AMM
z%wW(%+b*Ady2mAxw{q>&65{AX6f;GXuvlAJCJ91N8l9vVWr&4`Roz5mnDFIqe3N>o
z_g#d-M~OtLAl(>xLI@0oK@jn^Z{EkYZJXJ>V<!{iHLktt8p@?Io;Om$kCJ$#lV({L
zNGX`}N3t+_PmQ98Z`}7S;)`Gp_FTM(V^30ZiZq7cjpv|wA;&}d>rJ}DA?5KJjcy0R
zm`E!krLY84T#x>D<akJ&z|ls92h$10Z@85DYMUn?K1vWLXl+<pXyauCSKP3jo##2$
z+-jX>yzm^yY;fFX#Ic2CJ;(7n+O?nOc*j2Oe&8OQZXGXQ=1ZS_n5ES<rgo3h3j!u9
zV_d&?FL8Is<By+axgF5xHdwU=G~)qD=#w@E#NC5nz;+Yv{J(yRfAanhFf}%Y>&T5w
znBeI%r}?)Z`9li10@q%)n>XHYExTr>urJ_v^VAgo>Yx8pUi<pj@u7eJYutbLy>$OK
za4an`8Z?w1j^&a@DQiiKi9AFrJs6)L4rAh0n<ST`Fj=FN&Cyz2rB<#$#>bZyzGb6~
zpcw@e3Ppx#g7$o3lalu(6J8$Afy*YAd5yEg%D0|mVy4E8w_i@ybC6Q7dFvSAK=asr
zM>+D<M;Y5Q#T%w9?lwDF?es?;?s_DNA}wd*Z&Yu1AS=deQ&@3AC+uL1;_#6tdCQ%5
zu~MES>^FGV58ckD*=ZIJjtU9O^)()N`YF6}o||sk#imM@TDi=O&Qfz++;W6nn!zjX
z1j{0c;Y%O?UH;Q={U%FmYedP&rWD|s9Xt7rfBDbo^twEJ;1GXx{{tL8yFjWn0*%s|
zQ;UoI>;LCJ^W`spg@6AaeuW!e`&w-G;>^MRoklNmxf~z-;0O8O2S3QsqeuCJKllUw
z(|`I;{NgYEBCmh_>-qfWKmU?mM=J$5q&gXi`K2+g<&a82*jy*0Vk~J9O7OBCf)qK7
zpxMRC`h<gk(cCDQR1+jAZpJ4~Q+z*z>sj<t7(@f&?gaO*&agaY(_TMLH*E5ncU{HU
zxW}NYS#O3^H_ejw9D;6-POZSROIP5pw%G30$z`)dX*lv&G^5E?r8Gz%VHEv^W5`?q
zJ-Z!ayHuyQFf<CUzQokjwe&kHJo?ZPmKN5z{Hk3nEuG`|=N{z5u@wrk%Ogh~Wo5m~
zGf_-kX*`wUSP`xqX(L^B?e+Y~`~NAE<FiP|8QGX5qjK-c$|^^X9Hvn8IeF@7zW$BR
zbHjD7;;y%TKU;TPg0yUeu<<iK*Is=UAN}zE<Ujn!|HPmC=_g4p;v4^WiX>Hd4n#Ue
zsRS7(WQ8Oz1W}_!K`W#!=}E}vd?X3(AjH&LlyVtb{UM-7lVXXc8K)F7K4F?t^?fqr
zX!aE=;efC|&V8LKt+JrIc9vnk%j@5|mwd%x&^N3$2Gr&z$v6((?trwC<@oY8yn35k
z+zwN@97z(8YK74Wsa7OPqg@YyqOouS<@s!@%_1#}@$o4JCS}lEp*B86uUF^k{qrm=
z)Y)>`7LF{f@U6doij~t1r0MbasfX#;ms#y~S<y{u!!iyoGUsDR*tGXHe&oIHWAo(1
zh8C70O%$nCv>IzPPamS2$+C3%AotvN4>w$ME#G(NYq@08EVeazC^%Ab!`@5y@P~hs
zU;WVk!=HTQ!$fh2whTgv7fAYCtgw6^=~&p=47Mc^_NX6qQ)VYqGm|vyYc%TXSTbSz
z)(X>;6%rMrgdkJ!=(Ph<W0<Wb==wLAfA|@a;ti<M7AB<6R3VF{W9s)G<jkq37*C#L
z_gsqSq}1ze+N};z8k2K;N~H|di7o7Y`v<@nS|<(@o;=R%>+Zt19zrV`$DU%abP}(;
zjYBJ4hJ%Ftk37v=-*GdoriV#mmKN7J`t%}aE}!LXulfhXXCFn?S6N#=H`1p{!I3u7
z0_DWS4xBnM&!bNtMj0+>ak%hd{KETg;JkE&HimkqN8_nOJahO2n`WljyJt6B=eCl|
z<ycx>K`kkY#UhiFlX$lC0vCx3MTbw!pXT_fQx|oOzZ1tfA&G<_tY5&f{T#<B&LG6B
zKhLqiU<uHPq7m!!99Lezai6e1&b{p_t+J%IevUz}%Ny_7OTKK=>uXjUL&oPO$a)@~
zP9HUv<M`@!y!AFWJKb;NIDVdE?|F{3@3VDm#PP($h~s`^g|YDoy4?m(K6rxK*>$$=
z+rrV6RUY`#(=45;BTb*j<{xHIUty)!VMR5m4J$Y}$ef2EW%iO+@nb*q0k%v{ZYUM#
z(PXmGbeeUVr=O+cXIMIYkh|}_n;Wmaj_<$owd|gm86{Q-TxoOjWtZ}ikA9ec|7#!O
zPk-klL`gWZ_SIrzuW)fWc5I0NeCZ;ki*j9tr1Xp?Gf|^!*{D=eOrh5gP&iP?U`>wE
z?hmOJ^JqqcQkE+j7Y?1_kc{op7zPZB7Bycmn<<bftn!BHI<=OfIJJX(vCLon`2l*(
zgi=v5GoEK<evy6{QK=S?>pk|)93h;n^3ABk(x8v)WO1a85Eiy2aixWfLX41%Z`q3s
z15T}<MJdI~$|42f(4UxPePxMqX@VQ?*vsm+DNa6lg3D%Rc<n#9o=v+a(N>Y0cil$5
zI>ux!L(z2!?F^|-M;eKi;OrCM;x|6>KbT)y!f1mfEL_LNwj|Zs1ht)4QVs&<c3sO`
z-gp<^dgy*W@x?E4`0QB}qsm>BrrdYnX@2PaKf;Ip{lDY;KJWqT?_l2W+{XT;#@~CM
zcF&$Y{OAAtpZU;-KExNk@C82hv5&pv?=?n&B`_w%QZcSch<aU|jD-`YBq}A2V}v%i
znJn>eh+ir&?1%UwgUmW4?KTFD<9MTTagdP7W|7%E8Nb0mOKjtEYA{Z_Ql`4o!FNlH
zRVHb-hIqMkCN|Bn`>HJ{mEp)!3)CuQ)YKTqpIqToZ-%kTI+>hAAV|{$tqf_R38Mrp
z9Rv`rH}M)B)HMpr7z))2VJZphi(L7t*Rf^OEKVlNkt4_1|L9Xhk%82Z&kk^wp5f5q
zX->wPq1a$miczz%0BO7IyZS1QojAf1Pd&=Hh0`pqoTJff(C_zISXf|rX%T5#WU@Ym
zN}ZLJ6&`r#KHl*9yLiXje}M7Y<fx$)!&tG%ul~xvrdTTQk&pf!Q5++LH9E?^$YE3p
zmO$$STPNg9N)QdnxPnZoiIid(MM!PPWU_?AKBaPrW-p+W&*A1A2CX(4L*90Yl%f?R
zjAygB*#d=jNJl|NyBrS3=oNDmwz?GDJhNltoU4ZvoprWO&vE(XTM0s&$DUbZe5}Oe
zq~zqG6;2Gt*gR-c%h?D7kxJ1fB~b~zC`CCw0E2o9iG<5XBkYw*8HGi+vBd5hZ>2C@
z0ncag<T;KVI6@MEQ3l)VW33;ge)=g|?IE4KMb<c2wm^HD)E-IuRIa#&2Tz>h(Py7w
z@yt0ER+eZr+l0}O=K5K>Yl{f!V!J*;cAc*`)_CO6{oMKLJNQR$cpcM~%K7^~RVwgn
z|Mr(D6bpRpBmbSY9Fo@+%rK(gZC}(iwxx&bIcSxTgehqnfeA1gEYGJpyP4AT6unj#
z9VDYE^aO+f>AU2sWjc)(X%aC#?vu}VSY7=*gOvh<w1i$17^4|;I_w!Ylq;HcJ7i&D
zi8u_9!lGI#FgZCzwNjz{x_6QppJmW%GXKTTv1$8uoK15GEg+6K^w0x@ts!#f6hgSn
z%$0chz%eelYzE(VDCF|I{*ABZ^MCnG9)D_?Qc>_`-=%ft9NlI>G8DMN!jUf7LY`xT
zluoC^eGeU^+3jDDy1jg!jDaAEI6i-x`E%!}mP*Xc&M-YYi!qurr%!X~rI&EkzRRhW
zs{kZPaz0UylZ5*oeDI>UNW9?p`VJh2(U9zg9LEus8V#d_!vW<BInHiyES~4M9gK7#
z&vTp_GTP-xI7YXSCAZn8=w_L&jB&OeP_TP!pWV#mS4=PrB##|jV{*L0#JJ?7+TeKD
zp_U(=)STy7zmVf$y@@3x?0b&m3JQx}bBWzI+)93`its!ZPn_e(<IfVu0+Z?sI6gqD
zbpglHpnZ+9ErgMbU3DFg%%A4KvBNB!Im@}_C7R6^VHnU@KSO7AVT0ogL2jL|)>nCG
z|0BHNj@R&yUjI5KOQrLFL(}CVANu$IhC;ExAN|g6(zb`>RRt4Ve2y*YfC9^MNu^8Q
zvhn>qgHSVcEIcX6c|J-ChQmI7zCf$dqE^n)>h+2HF@>=z0*hW8k#Qu|tjk)zPZYQ5
zlqD6Z8H$L@Giy}31A3<tI$@uM`YPKevviUY$IqPP@yC|P=M*<zJ<ZZ;1AAGq=bBAi
zl3(Lgp~6}}K^uc79V|=YNSmT(W5)p)$WM)<ObSno4!9eQIucE(wwY(YbvM0*IwlO*
zymvd7-+UF0XRvGw6&6sF*D|r`QYu-Sj2;cbl4_)}kUGUE#o815dDX6+Tz}b>Y}v7$
zt#g~H)oS>eEaQ{gD9>!csFWlclI>6Ow#gZ8zwLIu@Rcv|hoApE%dHkL+V(BBI{fp0
z_5Y*S>+(ZC`A=|NAElEGM~jzlIB?NKQlyl;^PTVHo$q|-Oa7kcdO-vd8O3BgNvdLG
zFF|{PB<Ns*5I>he<a|uSU@U>1-58noGsO0QFdUMmBUuq4H2rP|FOww>x>U<GlE9&E
zZ(+wxJIGhYNLwk5R+BR;b++!fjE5dRz}6kRxZ&0-x&2L7ap1AT^uj(@-MWYQF9j?&
zo}`c);JTnvJ+kuC2BTACCNoO*f&eRu&^kt@8Yf?*uk#cJZGs0rL60(-@+5mp6Zq94
zp<7^m5c2pp?_p{21ji!7(3p|jiV$ea8mYvq6hz7={`}K?=5wDSNdhb>5JF-$rrwf7
z6DJ0xQ>gc`mj&fYj%KsT+WHbtJn=B^`{94gb=Td5>t`9n3g6H2FMsamQHkcGzxVqj
z=_^yQGmI=(gps&OO3t^5f&eE_q_$uf4$)ymF`vcB`J{D&kpe%TVUQ*iGI_e)0o^bl
zNQM*~8w<^1uSYSH(o1{P^5X=d!>TjQ_Uq@!m8(R}gl4NpzuRKxj-5RA=uw7S=eXtC
zo$PqscE0)aS^CkC%dXqS$$LUhb&fHS=~46<4dG}9VFZR0&vy|BR5&0Ss*P!FL#|LH
zwH(r6o3%&(lEm@ImBzTNFhyp(LgM5(+Z}TJ!6)gRIlw|w(1~137?5!!ao!~iy7Zh5
zAb9ADALHQNe~8u~q(F%5Mgx3^iUZQfVDu17isi)kr5U3BEPwgGR(ay!Q+)8<@8;E)
z??YM^S{d?whF|*GpCL(OKK{GEMHKYW(Z%<n?APnq#RAxdpw%YqbkHinbu)N{BG@+4
zwkcJ{h}}MMuR|IoxIqqtz|MHo#wY2uS_}sPc}G*)oIzVDscIm!1<pv=D@iqtMxEu=
zRV<)d$x<m-nV1;E&z8{DNiut`AWlLSzI8YGsLS}Z??5|Q(1K2Pm4jdTD&ww4=iE`Y
z@7u}TZI^T6c%4TcILIxxUk{eewKwf$*Pc0k>wo+?lU6|Y^gPGsSLn4I;z+Z7E<<Ii
zM(Q}|%{TMdsa3xAz=P*2W)~yURI%1*vfgYWpFNDUBx#ax{Ny~3KY4)bueqANyY^5l
z6wkXy96xcA<EQ2?+VAjOoaY#clceN)hcF0m0!?Z&;yBpgIQwFb^VvMz-U~Sv8yuIj
z>IEFHIMeL9VJq26g{TozZ}u3jx7fXF7mq!9lyLiq<K1uC$vp?oFh~L}yMDy+$@VeE
zGhIsFa~#_UEifqD4UW@rKsr>w3pq{(E!H3VOSI#YtJJu>I7K#JA(lDLbO#)N;9h!X
zo?@XP=teF&3dlN=B<B+bUAh|_r(gWT=Q*}45E+aCoo;X(Ym6EK&v86UG??YHpIhOv
zr=H?R|KYp2b>F4uIWG7ae);ErmL!e&<3IRq;&6ZtFaG<rCB(=AO(<g5rXM8u*&IpM
z#ZDlX$zf$YI=unOAizOjDNWEFkhdMWgC3<)fodj4*0Jfv3C0NKN+ntYo1#}{t-DG#
zn=q3saQIut={MHNjC)Mf%4Da;ICSU`H^1s~GRNk*=b^(KI(wKmU%!LAm*c?whuHI$
z40|e<u{tQx4#AZcSfjkavK*o?!U%)yXYun62#FDbL8pVQ1~~Z&lefMXE9}v4E%QWs
zjX|%DoA%jWE^z4`ucNT(a>fgSj0FN4r8HKQVtX!*lLceA<=yY&+P8fGDZNq8w@F8e
z9JY(}9Bj|Rk}hDOVolg?;TLMW_r0^+eBBNF=EpzAH=aCz(hAoSbcaKJ`B(oVjxG7Y
zpZIB{Gpa!TUE*aWbszlT2fybvLJ0og5B}gKz20+HHcBh}xP$1eBhm;PAMF-LJ9*m%
zIU1W*i6U_nc0NCv@<L#Qjb*#2ct{u~NGY+5pxIc*qyzGH4Wk?u2ZA$CK1n5=<NDo_
zT4j<?J=Et9k3T}Ls(9?7L!3A`V0tElFczx|LwsA&6pC)&XE^MQb}&f_+QO)mI2>YS
zMz&wXF>=r+jfdEF9%DP$PLZ&;#)Q3!YMdZBrRk@Fy0SRqPx5Ra5K)`sLBLS3u_cgb
zqY2^*9=-<F*BS`1ArWAK#d9V2jizbD`D%O|Lm2dFHHM6h4~e6M&j0)`{DbfRA>RDv
zw~;NC8MZqVb1uK|b3ezz;u8Ps^Pm5oir9abu!KNv*sbJ~5I5)}(il&;80leYMuQVl
zq6{Q)O5iB`LV;mCntaxl#Bm&=IHVDUcv9jRIM%8Ysv%=iBDBrQP_cI47*)pEy;ZWe
zGR9vY=<=!4PvGP<Pwzj%<IfJ6n#hrAL36o}XZsA%Gy;eEaDXFSk~9TfKx;*i_Q~X2
zgn={(k^PiJ#R%ykT^HZ?NV@CP0>favMe9sLT?-PG&~~;_AAkv$SdN#dXFU8)g@MTt
z(4fT%wgI;TmgNwt9tnMfjIlBamL7vtW6~6bLuzLT1`=T$vMNWETV+`4lAD<2RGRYZ
ze-Lo`uD9_6cickOb%<0-DU;zBfA;5CJ-5zZ{prU^(u=QS7q#jPq5*z>Brl+RcQol}
zB$jPqx%OyyM~rL%2c3jYuTNwYqVMAsi+C9yw_L!^dZf{iWEg=GSRzGBi#U!+jYa`s
z)TL6+P$|}^Rw`t(S!~Op6Aq|ebrS<4Fbng<&pyfAj%`?5_FxH%NTr;7@LRM`oM7AT
zGI48x!=L>KrK|5`*HyQ3Xnx2S|N2{GU5{+8NIvJGQ_YxZa_;0wmX`%VB$+6?%x#|{
zJ67aGE8s}8$7eqO724e)Bl`m`>e1Lp&smW&QtwxZ;^h1S3yUi}G4~YLUUenAw(rFE
zeC~blp^ILng9|v;ib4|NhIK@m;Nd*avAw~ufFyY#$L2YXZHF)pXhtK;Cda^$=K6~{
ze)7n9jxXKd_>=RGW9KwaJ#v^wpBXYSo<V7g=E?xi_BJ?nsSgG?)^i*eH#qK*&Aymp
z%kq$pv%ztl+R(7nZ_zxRu&xBLPUtw>Xbc4=Tw*0&q@MBdJ7t0mj+<tA#Bmpda0qp0
zgX0J*lVa&ABqJTt)Ho<R!*F0BOlE`Q6^5k_xrt4jN>cvI?+rNp_IL2^+i%`*J4h*K
zv;6YU{{pM$*7?Gx{^S+mSZRY12FH?goh(tNfVN#i5`+rSmLN(}RH~2$9M>TeI|Q9J
zInNm#uj`Z~iSfM*wTw-eDiQ&gk57`d^BhsD7@4psCz-nG3hHK=XHO`eYOL|@8!P<i
z?psNE7T3S}DsH_d&(g{OTWD%?1;RKWkHvT{W+Go8))9{FfR!LEn~Ys#(2p^I?YgK!
z2FLc$IvRNmhdtc#IBU6SHcjuNwp%hK4URNaqce<W48k9$mMf7}8ZA>SB5d2m&Se0B
z5u*)GerAfCwlG?tQ-LPk5QYaS4HgPZj@mCY3a?xx+b_~zYp~mO_;>#z&(2T(8GrOQ
z_YlMhp6$>c4Eeu*^*=H*H^ZIpdiUSC!R}=xbszuu$Jw%F3s+x#^+oB+v9J-w!cr;T
zU>%V*5K%xRY_eXA3j0{nBJK`I!U$ozxLzJ%y9j9$4TdO2$CJWnk|-b;j6^~mOOUaw
zkw9W>ktmm}FIYR>;c<~+=cS6f-mrtCCstY4U4o9`^obRwC->nv9>H)xJ?Np_Awh4W
ze=RYD!4RQ9n*`T#@!Tw_(TL#?nG8{hCe<kdiI>e#pHG;rc}z}}uvCB|Pb2X-x^77I
zfW=mywi>-*s`EBrFTS_hnCJgl@g1&tVQaQlhIFWi7J39jg|s}r@P*INYBhP+yWUNu
zR3!-q)W&N3yMOa596fTFqsNY+E^=Ncaj=YqR0*Y^i*1HjQ9##NRO}Ke39+SQ*y#}r
z1BBy|%@mN1M_xK~`+X7wp6w#FrXNQ%f+1Nc$q7MLc;qdIVIoPjjq3@T%RNruF}Kt3
z`rEg1^kki;9?%&WmgiTQ7{3(Hb_s(%?QnpW31K*z^a>I~5DgGUqD_o%Rq(tViP7kO
zil4+J+Ta+CZAtQ3pL2tVnS7b?LKVyONh#4uJ=VJ^N;T;>I&{MvzK%)FG8zXXB1|w6
zXI3dFQd)5nm0Ad4zyYB<8+~vFV@GH?*vJ(G{ltN8f>;l!HdD%5VuWz{+*j{rwb$cE
z-*^Y(`8?WaCJF`q?|<`4oIdm<r=NYA_At7r(x$^k8`&F>^nFZKQ5>(~IL@fLsT5%x
z5f28$X*!xn)RIG|J9Od!L6i_-BP>bQbt%~v<+4w)Y~wo~z6-Wg*tU-?Ee1i4a<NLS
zQX}JL(85KlfO@x!+O&god=pl`M|b}>nXP28w_S?KW=M@i+O6~Or~e!af$svwVOM3y
z+LK?QwQz<TU-vGK6$_l2Ut?)ujiMEi4bJhzH<#!xHRwAQiEC4t^cmJBIMeR3)Ee^T
zN4~_;$||-1gf)6NUc~$Rq7Vj*Ac|9tpFGRL!ZKSn&oVtV&54WLOZQyZ=Q&O(1wCvt
zc%EZtgJXMx;~)UrBa<m$J03ad(0h(!$3>bE$Ms-HR@yJ-*gVH^pOeBR-U+XJ^;V9Y
ztkY6`k^wB8T4iE<FTUdu4*PVX7jhgXh9G{PW6!GK`WqY%)QDqcaBPh&?G27YX7VL!
zg(_0_MjR`jMpu!hb^7%-o%0;8Y;YW6LfBB)LXgsmTj+FTR*wso>LQc{7_^Mh_F!W}
zF6bvNqL(1nBPz|5bW4n|Tt4%aud&kW@*{7&gIYE>;<#Ak-~7ribN1N-oIY}p&M>~H
z94Ev$mV=Tmt!#-z<S~AZvQRXV7Pjk?Dn+hRB%8}Z8Zrzzq;W_;O(+%%I0m95L<x(6
z?=zXp(N7gUZ5Zg3o=(V0g>5B7fuXjk#2!N2!|;W#9_R8s6O73sZb7hV_bkgRPZA^%
znOu=dwSbknWHJ`M>yUK>IX@4=AcRM#8bm=CM_9;0hN0`@NI`-{q7=RUkS%$iv%Mi}
zojS_OW7DA`118EAwp}J)Et7LJsnSS1Mhj()jI3a^XEPvex#8lpamfoNR}C2JyzHEX
z04Z=?4<|Q)leIBA<h}2FJ2Ry+zxltu#QLC*B_ylu4!`(I|C&$ly`0UvFWDG`dwKD)
zlDeP#$)DttpZp~A^YeV*10UeM?|m=TYV~FR(C5M=3Zp;`2k2fCku_M-Lduc!Ll_Q5
zlS4`}Xm!X{%E(-H<RYOA$uK~tDMA}Ur3jJ;OiB>Mqkgq~Y2+Yb*?4%=#`B=N)Mmy|
zcFb%oWujD~diC`@{nQe*$^?}$pW$%8{>Klna&CeNSAmW&S`&l;!$E-22FGzJ*QRhC
z2c;6CPKR1vqT_@#jZsP?U7v0*B~dV0tk74JoN9SIa&8`Gw~|R$Nc8Af1NGd4;9@)u
zo_`Ic46SCLqetd3#v(}!opy)!z3+omiY1abW!J8q{Or&CbN<=S{0oNB#oYnwFhTi(
zbQqBKy4ZOO*Rt?zNt#CV!+^Xiv2B~xW}Asx6*r$FQ3|6C{lSnlNfAcV){15tVT_^^
zN1$NBFQKJCDT$3uxt62Pknx!+mYXs=rYIMRY`O9(o;bL|bfrSEoM8|SIDGI3Ybzyk
zQe&Bs6hb>p=to_QQFt=PY<Ys5BQYu=?Dp8~`ItCCyQ6kj>9`C7NVH*Fwn|-3vDo)G
zx&AC-d^=9kM#YA-7ZdB0kN^las=K57Ja}%+((w7^i(w;sA4R(H@$JhTL)<PfaON?3
zbo_H)V(4^-{LJ^inaN@vgW;0RbNu6<`+0u#XMU1i5MR_v((Gk>sg{cvSB|F96NQeG
zjn5O(I6<kDG)@Sjh{O`CMuv7j!4neKN${+Mf>e+ute&%&pI1a$B8-RYXnaqT@ob6}
zkMUZLc0VGJ5tbAv9nlP8wqO5Rk|4r5b&zt{Bf`Snz6XUv6b39lzMoSMA7s1;uI*Cv
zGxS}F@251|CwcTQ{(#&KZ^BLLWP0<o7S1t0?D7PLzNNu36nu}P#^caRz*KdbZF_6H
z=k{IPxZ322Cm!Y4v(K`)zD|^m49j1U^v<0B4+g<7N;rD_G?DqQZ2$oP07*naRL4%9
zLA?Tfu>dMeH#i=W_Bt=*IN9LXk=V9mwK3w@+u#^&7z~0JaNJ7c7jtY0lqIpXq%@YJ
zZ-$J`R53DPuBIsG3v9XSN*;S^nW;*JQYFhU9B}yHG1iugWTi&Dfa4y<q<AvNrqTrY
z4UVHu_q%XxNVH*lwnAM`veeITYW)z5Z^KPDIPS;9DkWt2{8~ic&T;U9ul?;D4@O^B
z5V!LT+*4@X0b_A*3Z2e~<B38JjbZPW&HVV!{XD<+FMf)CkX%%brSBjy*tr}|u|nD#
zO)H0KitYFmGC8{O$cAoue2kT~HKJ}0TiRd=3WXewYhx^ns+5e`9s-++=VPQ`IcOu4
z#<o*Zk<wc4Q?zodpQ|HMmoux9R=3Nk{rkD)`Z*RG4O)#pxlDm8uGz=*rZ#;gF~T49
z(PlCf`~p}OQ5tckwM;)&2<hUN%bW^Ba9w0L8URV-kX+Wr6@p{^Iuq6+bA=(hCoYB6
z6U4TMlp~LLAuZ&$NsOE~+!W_Od;SaGURHa-mq)K-FxVdCC&#g*4;WtWK)IabSO564
ztoC|HAvidHn*aG<{|o=|e|`iz>%Z)R-OEbqKK$Vi^XtF<>wNBWpW|a6`xw9UOTWZh
z-}+X5<VSvlH@x8uFMCQOZLFfRK2fKM-D_cGN0kK>*!D>CMp_nDE<@HB(j-Oo`#7G5
zB`s{v#fl<Snxd5<j3UBee`GzD3anV*7ArW`kePgd9N$Y49%m5Mx%D;I;@X<cGgB1H
zC3f!EL6Ah$8*A+U%Hy0{U7}L<@LCBOD?sUlPQ6VSCiu3)SY@0-ZHm+=h(dH6Vp}<M
z7?Nfq;#4CYpCE)zcffRQ6Kf64y(ibmZ{I>QOjtg<M3Rb&K8<)`7)HCuUN`2*v!{r{
zlrW4))0iK5|Bq3~mk<cv`ObIp$DjC9?t9?Di~94E#xiL>ORw7^(`zI1WfG%6fh%on
z%fhuJZoy#^Llh@8dOh;KKeCkZee5tI-q=}pl8}Df9StmKK~E|2xiYQ@7|*2m)jgaF
zPty*%-1@4$<ZZ*gsTyOIJXh|RAxu+N8%@4)_hHuSO`N=g-AnNaNp#BDPKz+@;#n?p
zg$br>(<CVDC_u#txlD#Mj)-+iYz($jA&w;Npu?WZEJs_KXV0$SZQV&Y7&170j5N$M
z9P~)dEAzlLp(I%rgfgb7o<YS4Pec*Fp2Yn8KYBZ3g*=vkcf93I{KcQ%!paxFcu{}8
zDC9_U?nqiARTwKBnXg+mw(lYl=yas-Xbt)dyIm%WHsdwPcutV99K0-KGGn;HCW-<E
z!yes!OxWww3tW1Urrt_fX(&#d>?17jJVCi2D3lD_Z@v?eFB2^<APzrGyVbzE<_1cY
zSwf{yoi<N?<})N=LS@>J&lCxiP0yO7L60>j;B*ji>PvsYxG!n6yDSV91EWC*5}^=4
zaTsCu8XP|xP@Ar^Wp08UJ5p}F^-g~BpZ^o0VZcKVJ;2}m%~$x={SR_(VG;G*#_bh}
zjp<nc{mQ+Mq|IeiKJ&M8EG2Fs$E4v!9J{`Y3^(kdl%bsj^y4lz5~BrurJv_GtMJFR
za4I@OC+za7+xC*PG?&iQs8tGFwR4t8Db|}UzI^vFR@c^X@)En3k|83|31>QO!lZ*|
zxoj#-Fg-qXp5yepa%`XH_*hGG=<G7ymK}saNdNTF5y$;kmg9H{gp6pYXHZGPV`0dD
z+u*pG&tVC8=Ud;xXaDqOR=@n^i~953$vI?Wm14QZuvA6$nm9s&El5;KqDNYbQMbo3
z;JGffQi)y=j4JX$gvunij*BY=RVPEiuaKG)Az>(?*J}`&94iBxVC^`&y<L?35_f;~
zDf(LS)VUSb8eR4uIg2kPQ?(lJx@#ZP)04!#l(ZxX@NlKeM72uB^@()EneH+Rs|(mj
z(o7cVx%7ua9M?xn2iFp$@et34YPQ6o#%V4alU%iVmgw{ee9xh<bvw4Fk@Di`#l0xR
z$QaD_z^`Vg$$d!I;g(E>f9rVs%EvxQqu0Y2!zcgpulT_qem~dWe*4S*F)w>i!t*@d
z^Pczcp7*?mGiT25@sEF;pZckvB27~c9Xdolpa0%3vJqz<{T6%5MT}PXM)Q0wERY+`
z)s|(Ug#p(=CJM)Kv7|++H8P)};5sBhh}F>ygpDg*q97#JhHfLoD^D_B>@qGE=nZaw
zpv+oph2_;cx7>OSQ8HwH{si}a^L|botrHCe`Fx&TyRIT$mt^`UDV3WH16W^gfP%4!
z;z$R_b&1my!UzYAFa{Nes5l`>(~*~6x=}@PbDV0{VLXSI^EvwHqiAK2=j*z!R2U2b
zmBL^UapJ^jk|ZWo5dy)Ff8Zza{0xOcfgk^|ALEfnA0>(}UK8%2C-<|jSiq>1icuSh
z|LDV!Hl8hUEQ?ftaUI-5k#St4ltfzL7qV>f90r3SvJG|1L0cY!I3z@|)?cBVpJZFc
zFdeNCg!{0gJnO6LoLg=4#+xrARw0MZEb!0+k8}KZgH9i0Hp9f0-K0%{HJqoETcZ<s
zEOeGZTTGQ_*fTbTbZiDmgcAg0G@}Q+iZDtM>J(#*OrVX?WZgXLy;a17kDbZUc<3HX
zq!G4tvDF7&5Fr6dd88eIkwdgTgjOj}q#?hF<`>@kZVJ9fIhW%HKk%bGcJDV)!vWDn
zI%_wnLPQGNwy<1(<aUCCFbNoi)_@sVxMe*blLP<CF`OM&vv%$Th2|1Fb5M5kAQYmn
zkxs(Icoo|?q)AK^Dh9)lc0Xbe3X;eoR50ur28qw5S6#>Y@(PDO@oBu)IVPsZ*?q+g
zgbJc0pn2#q79M|`Qr56#(=2Yb#{I|EdE#u7)JZAKOd{<R)ok(fT0~o=XtS|yMj<Wm
zEQ{2D)P}g-W3k;MXw=DM^L*uRzQD6bp61POc^CiS2Y!(6|A#-o;<>YY^uxc!pMC06
ztks(%H<$m-)=w8JMjYocDy94#IQB2(*mH2>6yI^Nr6kshY%#~2=h7bxvD${la~vlj
zp&4;p%1^K@W0;E8h{AoyFh_l@&ca%UH{Nt9v5Gi+Zjpx{e1c=g)@k>^%4V3@x|_5i
zki*lIaw~KqpT!N1Cri^@GB$~HoDs)icmc=p2FL0Gjulxq&suL4#y#v@p2j2J#DxEQ
zIM&K1ZCe;?KubS^Rw<9Gi2sS<=imGN<UNl{KF51L_@g{_-+iRPfcPSDeB}+V=Jfvi
z5w=6qk{ea#k!C`yQo<x5)(YDIQB3UG4C4r;6t0jA+dWQ)5rupnH71xa36=`Udl@pq
zV!BYE*<0qspv<B-O(ydMX|~5?d4es|RsQPUC-G#4E4FRnw%6RuWtUA+Eeh0Nz{<%(
zIO7JD-9f8ALB&&SsZ4_fi_HcH*A`GwOy2cLGFgV!Xb{t}JuE5lr43qPIU38-45A)W
zg(+wZNd|r9Zg@Syk!Y2I{qki*UTTb9SL1jfN*TtsU5d0_ZniW05`tg(-9Mrqgfx3S
zKK45w<*J)*!tp#_j!W#{Yx>o)tWopxhF;Rkjm3NKV`6%i?4}7U+rb!#G%-%bN4OFd
zfh{GD5C{#34e!<UK2aDTgh0ws6TO!y(CVGR6@pGP!SXC3<*?XEnH<kC<E``TnIqhM
z(~W%nYw++_mM9lo%C#}R_QfMScyAw1%wT&KB9^@BrYl%^@DQnbj8e7C%F-EvVL-W%
zV`gRo*Yz3dlrWAl-3}$!B90Pd5}^}K97VWNA%(_uG6-YS)sl%Fn|SzJk3gK_NJ;v4
zxNk2%%t#qW3Bn-Z?AawOOQ7|=jF-px{_lS`mgV!>J6^-i9XojT@R5s_XD&ScDAmbH
z3X>J2?V_|K!^q3mvMmx@prypKETlV{yc&if_4PWvV2BU`S4y(BLp58VzI+UcLwDF8
z4alTrmfDJ46_=fHhlkG2bMp;X@Qr&?9{Tzc;{}(=@e*JC>IwEg*ufHIEL)I?9j?D&
z52yB@WT+2QElx1sJVg+nqvUU4-}p2IH^aJ4=nMm7f54dSG6)mwBqG&@eiWbA-tuGx
zVQhw$LvD7C#Rt9)8m!U;s<F;1Y4I~;bnKc}CETcr5Ss4FafAgAJ)-%ON`)VJ=NoaQ
z&1+x%Dt2CeC5NASglKSa8tS7YMx`E>V}Y@7l)_jsu}~<Z5!xCV!fQbcgwkfy^|x@x
zPy7<b2u?o!5J&&~qj<e_90R55I9e5nqZrrr$w(V74k>BFtTr3Hr-s-lOd8?szK+my
zsXzP(Q%;Y`U0bN$d<Uj7LoXa)1|8<_{sJlt*t4a~^w<nfEW|wC4w$%NGfEha96QJA
z@)C_$(^4ACwh=1DAb}0BUm=LmBX397<%-L%;730AW4!iFuOmuRPM$o$(@!2C7!0}o
z#+%r@bvys=*M5z+z3ndk+kg9S+<otT^oP;^UnX|vHaN~tRIqL5MI6Uckhzd!%O(sX
z7V8bVL4YMjLa|xfW-MEzv2q;CICKXWa;(`^aoH7jc<9_|Zn^$)?*2x~gI`~yR&bdd
zFY?u|9p{nzo6mC`I^6KaJ<LBk&p;ocTAXCQInOXYOUc{9zVR6fZuZ+ab{K>yb`p|m
z!yt+-;J7;CSUBY8=2(32?vcz#X#(A-^Gb6(G#ebpRjj_IyL=o3JovEU6V(bIc>5c0
zrQ~(5el@!;zk<V0?<c-U9B;ejHdHvIyVM{}Vx&++LJ$WbI#oC+$W#$&jfz0%luX8F
zVse5~HjCB}E5#s<S?{$NpWTTTnsPpaPEw>Du(@b+q~7DOSK;<;JNf4QkMY*G*xd29
zEv$7TmR8!l{q2|X>er2<OcRwDf?mSk{I7%Tdet8C@=EHRI=gB+$$DklL6ZY3i*(~5
zlhUJ{&7z70*3+1L#zDCmi~tfr?Y<?*dmaXfHj*Iip+!PYr1VamX7j7=1lK`D0haGy
zlzaK@ao)3@(NVAv8snrm1((w74vaS3afjr`8Vx@DssBZsrhMre-{9wt9cBAvS0JsI
z{-*J=le#F1&L6+tfB*g5b=O_|)^Gh5Z+g?4uwJgF%Gx@Izx;VFd)r&^g=FRI8r71-
zlQz;aL}7%Ad$^WOr`4jlv`(+tCXEt=8CAxxCDL>8N@aww34%V^oKF&^bkUe@9i5Dk
z^_M6&9-{s17&pCoD-Y~{mcvi2v2^YzCk_X=X%X9l@yQ~ux_K9kL+2PC{3iRRTST#@
z+v}3cd(6(wP$-m%qlkf*BNHW9y$(5P(eCuHlLVnPDot=*P1YY-z|{Kza<at9`Li@z
z9ZI&vAk_?B!6TCkcN?S6M-nn|lydIu3a;z$S6}{1w(r=^b=TZLwNmA^uXzp69yxN+
zs)xE==g0&1a_Q~2(Y0)jEH1Gr?@+WHgwgbZfF$XXwQLsKP3G5DS?RP0(-_O7U<61P
z*Y?Qg$_Q-w@qnzCCr%=UDy7?7BuQ$Non<Ob$;#0Z*Wa?0{ZAa{nL~B%S(@k6@jjN$
zA|1ilM3L*S+sU~jOLULy=c-zVVXEnc>twxYE~#x}CSRf#g)}lY#eR(6>QJ*R=35=y
zD8@0GD2=gga5I9e<FPUvpsFQ0r;ib~R&n!HlEILM9-<^KXsx*z;b(HVo{upaFOx+|
z8$^mx5h^IqTRDT2E`R;iud#Q>W^TK3ALHc`x4!mu969hfFK*KNyG8xnBAtwfB{M|A
z(2phE)L}3T=*Nmqze}n^@<KCPmQ0RSK#$D4g_mRZEw?c-S>?co|1;a^q0$Wr4UuBm
zHXvv=Rsj%}L>NI7MMT38A#CREx`gJ*)11BM5v*X9^|6$zesC``h)7gI|LOg#9(an$
zs?WB$O*~s4a9C!UxO6ASo;kwe=|z%M5h3X4k(Q%Qp1U`rMbid?Wb}C~Qu5_8Z+*u*
zxZ%cIC>3jXg)E!5ZYND5mKPU!_L+n1*twfRvBWK}el3?>eg(hz;s3#}|M%ad*9$gm
z`o3M9`2XC{t#uCH|IH1K9gZ!pFq^X}N(X_aABCj79yw{V*lsbuw#rhwL6Afvju8my
z;W{3LYz2%QaqQ-a(~zOkbbAXV$=HbFhRy1+GB@11jYl3o#-T%XzPWUYlgIj4I)`No
z#%cv_xM3FyN0(_I-p`e_HbJWCg=-reZ)LXdVvffw$(i<uW7inM7jW#d8VpfmB|2x0
z6E>G|^EHyekh&hcVjMf3hcN~}lifJ}OVKJK4GZ*EP9trH&;QL=*}Hv?Td%l`@k)uC
zUwbEqFCxcQb&AXHd<#$g%~vVugiGh9IongTRRUNP{S1y*qMs&g9vfrl^el7LF*1&i
zv8<6To=OPgkZ!*LLQ(Jv48j4P<B_v$%7tkhca3H|M{^=&=CUUDf8!Xp-Lj7>ubSkI
zbGzAnSqWwJQHi0x+Ts5D4zXq59F@y&<I(jovbLa<tuYAu99%uknQoJ!mtkTM(<%B4
z$11ET#g1x)*mH@6AY*%^sURM}APJet<&f(wD%*CDsn&?X7{_<WzCuT@X2ZfySYks0
zW%PixEE_vxD9&y{r7{2To$ut}(RuE9{7F`uE$;pL*V%sA<-D9x*q4>m{nStW6rcLk
zr<j|Y<AWdkAfNo?C#ltHFYABq!cbT=jxTWh>-Tcat}KV2JVJG<h;%aOD5RG4vDJ_`
z3^2x!&6gP4xdSQ0sE>}3RZp+mrQc}a2!)$*NaKX{eu4RBpVyWpMh%eCVXkn7Ge<wm
zTI02BtClIbRr-F89oqv+C6D1CVb}I))*e3u$L`^>spHJlip(#Z!1Z8qVuEUU0<8su
zVT&kN#-tiG=+j+U;oMq-Y#d?5F;NsEg(BlaHj`mtFre=hnLjp9*4QYaXra;HW!m-c
z8>7)`>%659u<?haQ3`@#!t&Aze%9sBKKTi@Z`r~4<P>+_c_+W~`@c_ok&a-EkaW*2
za^k@!nBJV{zGt7NR;iMgE`uaslUKr0LxynxD8_STcFpa=F&1%}kfKS_l$Ak`<z5|u
zrs%pPX`lA6#(KTQt}Qu~8sJKY%d$%xI{tN5n>R9DE|7D|^nH)boAZ=Q9^HPzwk_kF
zJ#dWV^nNa{o?%O|$btGHY};aUX$w0_6C?`O25o|J4wI(Dy&>ys3p`n0XFQ6KQAB4r
zM5z?l)f60`lVP98Ez)`RNo-xg=pkXUi0sxt)%lK312Gz$8qHW)2qjQP!S{H0NCA#3
zDc5R9+rcP}F&gw><lz__&?$q)BGPsFlh6DWmu}s{WVy)gueqI%|MsuL@Zy-g|Ly~)
zSW7{PJa#TmZEO?9c5yt9z56a>_pTj$<1?S-_^HPTgOnZR6?P>q(exH}UvU-NcV9tv
zW*f7we-~?C`D3PB7uOk0m>De?bXyDuZG10>(h;dvSg`?-V%Ij2csY6p@4>VB?73ne
z#n;?LoGTFoJ!myq`o>>kf{^XoHgPsedDbs*Zmq|WN1h-Y2IR0%QW7e2UN+>r#dB&U
zjaG}_{@vf=Tle41%{Si4O}D;^%dWhFYIU5=TemZedVJ!~exD!sfgh$?8K+zw<6r&4
zFEKqi%`gA!e@mm;*?3<cO-26i@L~S@gb^0qbBmmO_(`TW<+$h2lT1`=6l|A%95Cw^
zkt(1c4>2ZXJXd1-rk%K0MCr&)AyFyIgDxxGHH0u591m!R<J22Xc5S|p<3*l1_BGa8
zH!@Q#l95Fm+hNnD9OaTrKTvGnI>FhePY|DagngCMY%i5~a{U=>+hVS`h3%yY5-nc9
zanc{sSYPCc^)+fyL>PrH;5hGOI2rVaywZqcQ^lCUOXJwOfMfmAIJN}Xj-*l>e*wql
zIgX=|6w07J;`on0^H=QMx|xYmkvm><2Y>XD-+<ua%)knxC~w`uC3oD!;(VVsU40!Z
z>kT^bfMO=c*=Rs&=O}nN1d_FWi({vc5M%Ic8%Ic7+hMj)X0|+yF&6D8!nJ)WnL3k|
z3WxhKwk!~-m;?PSTz%O)n6BN+x%Cc#Y7r+fXHO4kv_eWb$<Y(bY`x)9yj{2RWMe1C
zTP<#!+((ia=38eu+FHW34Ac2BvaK${mbk?{6H}YGacY`%KSS#Klyf<xw25Nq#zQX2
z7VyIs<=y*8Oo+fCBRqta055iRfbZ_abN*=bJLFwvW2ng(NaGltrX&Aj0SFs}z;Xqy
z<&huXjEZA^;{89!!4oH0ZMXUA*T2pW{p3$0eeWf&<I76wKKjv*vU&4nwr}6gz4zYB
zz4zYx9k2V`=RWt6UQcM7NNHA{IZ3a!jq9%7!?Pz=ICSh3H(fbL!FQ2DlC9Nn@)?X6
zDSK<J2nH!4Ns$gzY9&gQ0)xSjZofzB=XmDyfR#vrrO{E0@O^wYWn1Me&e~scW_de7
z>oRKjBDJzj8U(1tI;~TWk&XAWb0VZ%a_9{@*tVcnEipAdg=H)T!;pU1Ln;@QBrG03
z$LS}Jv3KWoP=+81i3TA$2=Q%DXzE#?g`?-_MJdwFVv?AEV8i6>yHjro&L44KsA}@J
zg)Ie{jECoWSXdZsNYj)!8ChSMk#Q~UcF5Xl9n3-QzV~bV!*~85m+#$2tv1He(#lI(
z!|xf&A<>5V$$2`JZQOOq6@2r|5|7WH;<nvenebg)VKL^_$mO%6$6KmMwPwAaVB!=<
z!j@u@*?fVuXuxvNMY#o@I@X~d$0W{Z+EThcMLXe&{4$5@cQe1XiDqXzxk8al$sr9Q
z%u0jBGY82g2f4hIGM0B(4!R(GrgA=)S0-_^#Yz;i7WMJ8PmrX{pE$>nBad+PrmIO(
zMQbpi)9w=v0vs8mjA1pCr+4BAv1K8$C5T;y)*8moAO_+)uAS>iGDUYJD4j^$wx+HJ
zq~k!7jVag{h6JW2^JFHbadQ;}7Dj1wl8}a=wU5yXEJ@fI;$+u2HE;Ohx9;ctZ+$bD
zU2+MznQ_*e7sn+0^No;Al{s#^=?%Q+eecDS@YOGWfitI%@Z>X(@$fSb)2cVno@CRG
zBI}DSymrW*UckfmK7fujTet2dzx7hidU-0{Rg5Lk#*id2L4QcbD^jhFGZF|;#L*DT
zw#i<1Go3_%6LaO|J1JjtD{Gr)N7aKQVEw6Yv3mF<#j3}$bUEo}c<SIh^JiD^w4@*{
zQe|kSnzs4gv#N%T`dXqndv2Mv^;I4}u%AzU=1<tNbt`wi?v1?R4R7UBpZ)|#jyyxZ
zKj6oI^q*3$RPi!--uJ#AMJf2@fBkQ1cRH`=kkmyD<&tQ_`tehAO3!or=&2Ljwr3mT
zzJn($#=NoTIaZ2PY1VpuNK#x0+lwXU@<mpoKFdK5<>onXtV=JBNF0rh6RZu6ugov;
z?AqPTuWX{#*-o)gBwzJNg9N?YVBz#LWRiniTS}<qU6zI&5I&PRkISo*xJI%vc%I{M
zgX3opKg89u&vV?pfMab8E15jKlZS~B2)_i;OXRrVA|zOejoW!?98XT>$WBaca4ayX
zL8mckXkN^5Zk>~-3}5`#1HA7oZ(`rxOUO^xs9z+GkrA$!V{-2vD$Qg3#la_e<BrYT
zdikY%{m4mH!k9>F)}u9gojw+TBXOi;s#wIh4xus}Z#6mHZ8Pl`*pw?HqdwPfEl|uj
zIH`r@jf9&!iOXZ%&0I14Hg?UNrdl~lr_-ia53sW?V$%d$cU;TRy^e=lF00*`xm*Q|
zVyV~QiTWb3P|V~rOlOKjR!GKmxnj>|uDIe_9y$9gjxDL?i<GJbGWi^_5=bGK84f7z
z+=c7fn7D;-Gq{Xw1d*fDfeXXHh7kNlU#tb#0vj=Ue1F#hpCOJEVW&qr9AMBO1=4nq
zo{zL`gp{BSnM#&qILV&fSMq(Ye;ptFi_dc4nP+IPtuwiK(@S2)mzC80@DKm+MQtD)
zb;bxxtXVv}%-$P!vwc$$)m_3D1H5d3OrcDgq$J9KZGmSarB5#7qr-@FL;Gsb>40Zb
z%8X$NsFfqu1BWz;h{AwUr9u!Xd{3}rqQiK2fc2Gww1$Q-7I@mQ!|yXz)J#n0Ft%VA
z_V9ccKh97sl}WW>y}3-Y+ojcu7%vaW1wB^Io@H)xjnuLj#)^2*CkX~5gBHGRSWglb
zUB$xUDp(FW<I@iZMCQ4>@SP*;``q>Bx6*7jIC0_xtLy6o@!!5H6GD*7<ro_qBb&|Q
zNC%@e!(l+T*QGz~lOzdRD}=CV)VpLd9$)z4|Kg5YUqiV%#@4M{E?QEjpA&UR6^qMf
zc>NXoxOBWsf3SdUg%mSo%7t-~I3Y+i#*ql$#kFmwvKi7aBurAmG-ajJK}(03Vhz_4
zJXs0oH1b3$p_2q`DOae6ijpI^q10tte1x;B$5=n38ECMTX3h(k%)pjPmPA53?o;#}
z`bn0p#WJCR6Rl-dhFu!NkS$YPTEig=i)YzV+d(2Etste}YBT8dNrr74%dn^tnkuDt
z`V5#cOese)-=Jq=M6!;BdeKTduE%((#%sUtAJFa(SeQRW`|JYt+F4R7{Ej<6+i`GQ
zhuriovSZUYZWhaOMu~eoBn~>n!!Bt!M8^Y!l_%)5u)Qo_`s!DC)2%mAEti<xyo1xv
zTpUfa0L{tH0+pqw`RgzL1rI%R564fRAWBEcuG&y^a&cK|4Vf<IC{Jx?*Pbg-;gI(F
z3dzhIxB^|9AyuQ}coiq8G=U^xYIZZ$hCsJ4DZcN6pC`>#nZEqh%w2sm?Fa55+_D4P
zaq*<Xpgzy(`|cw(g7s{bMV;a4M~||y+y&1k#3oAmG?b!iUiv}ksIF$vN)QB^H0?9!
zN7U<eo_+QZzxzACO&A7j+C0I-`@c@HSl|Qi|4BU8$Mt;P|G^)j+v)P}{=<JDh+bLm
z<%<<B<oGPFzv413t5xU^7ICbIO145NUnPnYhKUA8g6raKaIAumFil1rce`M_Y$}f9
zN;psn={5?)I;ML8$7Kg@EOywI^f|k7l=ahwP#f&ju*n-TwZXBHf>zvnp5sD^7?^L5
zIIa&vHcxeF4g=1eJHyu6PLy<Lg(>}3he5wj5_E7Z!_spc&z}LtFy#@){qMnXtuV$N
z?|c{S!H~t%^K{QGBG=D-caH7nInGsQu)PeHaKJprarea>x0~4R2FJJDM73OIX3I`a
zAG$aWB9_>wIoXoUo3`?W$DZQAYLjANGqKLm>^Eum>bR~;&XtHb#e^|olCanxVrMgC
z9EXhKFjFog(umWoWnu+4bbQKQfG2&*ZkAKcIujY2W^A~xRpz*R1>0x$F}cGcZ$m09
z&LlY&2ZEE$K6Yvt^GkHXn1b8qWV?w6>>As|Y`)AyHpBcvky<{-)~(w(vATd%3DvAa
zajZhVSi;R!NR-89=~0t5#Z7YvJ%s)ccdCqMSs+|ql+bNBUj@N{`@Zyrk#*mW9h9b~
zk=eqiQfFTHZO?7iwAKX82E)cONi;y~7%U6N%i(6qc)1cX<73%2#%MC-F_Lh|_rK$v
z{M9$UNxj)(asJdRm(+dy;~#%Hmw(SlHYQ#JBRN;^6GVzaHpA}iGbklkU0$UarA%$v
zJet8tN9N|*7%b^v=N)VWaU2n@t`Y_TVGt6A30pl)=-BiJnog(1*hG~~CdXiqU}Qu#
ztJ&roW{oCAU}=LRU5tQ4hgg<9x;{t<B8O88Lsr(-Xtxv6*d^mzOkS2F2?xkDVd2a&
zWMVE)H6~3kS`ltc;Vx;L<%Kq_VMO8;AWax(g?dhW{yW?oDJ3`Ed<(z*JHLapZH^p1
z%tt=*+x*!l|LkuauJC=Iv9WPxre`RZ%VfL^X_C@zw~?#MXf%VtfFw!5KolrC?H*^v
zDIR_7A>R1rx3g{Ab{>52k&Cj%%4k3m8=Hkzmv*F>&1ATFYBRACJhQsYbfmay+g6N$
zBu$A`iqZyYJ7fwj83esBpi^I?69)8?A;To3R2neomgz-`g+YrwwJNn-j+I`FvO)@;
z;z~!;!w?~GH29W>77&;STiR4)j(QXmCJxW84_Rrg(d{Rwp@;8V?1~G7aRNees=k1e
zsgsEllpSNVB2EI7G1M`1*BXQ-Ks!?~42gPc4625L7`*7dSW;3fSGnd5Z{fdx=vR@_
z;@Oj@`Pd)*37`JMkAV5kN29jwFgCrJY;_a4(ll<qitYNCG$rb_5SEP*W|ZXV6k`l&
zXo!X#R+i54;FAY<`)h7z%a)xyz}GIy8Utvp*#Fcc?0;(XcQfi3vyphX!p5<DeBt2;
ziy#T8PR+5twnk&+EL(PNCrQ(h0C%K_;~3W-Rcw<eB3H;F^FB5f!yp{>S=uf*9>wi@
zs9mxb6%RRf_8h%Vhw<rI6pGb{9%cD-9WgV(ipg{Ci4&Y|_mP<bULnt5y-uVRUBpHk
z#LMk(Gbjx@f;3JE!xXDCz;;}uHR?Z}i(|g>wa+p&HObrG_HGao&-3}QAOBGfJ#&yx
z{`p^!zJh(V7b{-KafeQ%n9cayJT*tG1W&CjGE+`R90O^Z5-Ek!8Yvz8qDuzoM<MO{
zI_)rEkPHbnIPQ1K^g_kLu*DuzWo(0EEhCCn%2kedF~=4Jdc<+n9&sF{E{7Wd%dK^~
zy#zJz@O;6J-Fd<|Mo61e^#$zA8rMY$N{%Mk<9LWNhII_x^$R!-LZa>({ju-Du_Y;N
zaQr|1%ZIQni-RXl^M`-*$9(3GeuwD0aqPHMr#F$SZX#En#?4i+-3&TSi2E&swX(sn
zei6sZi)VQ7zyaQN$E(@8c?S<(B#tE#jKLKtTgHl1e4DewfNw6I!Ig&VX18#zvCN~#
zk0FG>v29%2#ufsVMg;AV3THl(p*k^5C6lG#<jA@n?V+TRM3j8VmVAw8#5!yJKC?v+
zEA?q>oBgq&h#(^jv4B8_SXz*^d{UKCPkT)GF1{Ue`9zLQW7Cw2E_OB`j10$~>5=hV
z41(QLTS$jR9`qe#>LaAY^?alym=ucARE><E!3yi(xOjF3I|X&YT)>5Odlkn#{jGa=
z|3CgoQl+_i?=F7&AO8e*{a^1xNbfs#ycuJNx?Kj%1qQ7JqTV`56ks_vez8Wjv<ZQQ
zCoP2UBLvu4iI=Z1vv~)1+;{_@`J2Dt^r=%^bNj1ba#jCjC3WBTec$&zuMtA<+0TCV
zCB2?D7RpEl8gwgVt=VL@TBKMRBMB3{>^Q=*Xs<Ugj1nC`o5As1gfkL6BvqutA%#*2
z&vo!z8%qq4X^$;ANv9`?N}A648sl4bk}ugrksu60A{}8gNGzlnxj<O9Pof12D=n5*
z`n1}bVKg!n_oZgC;4rh>VRF1kVOr2hBRXqs9OH1!EjKWf;0qUR4~Qd8WF%|0OLuvR
zM1$kIgx${QC|MXVFKoK75F^uTKf@1v|GOzyD+nRD;il_(`l%cF)TclBU5k4bf>JKe
z?Bon{n>I5xHij*2;yA{#Bw-j3Mgei0j(iD)K<R{br;F=q?)m1|x&4ki*)lixlGg8~
zt+B)yhd6~+S8$|R=gLZnO{FmgVN5Ak!?Gl&SJu%+<JmTqTn^v&k*<x|SmUtYXC_}H
zZ#(oNo3%h-={9l2&<||Ftl~^#jeVPTGE>RYjVv0&fT4*|%HU$*2!&-yTsMO@7RMSb
z&NO?p`<gJe0F5gZwT#Q`CYvp#0<|hE>Xc5qhh?(dcI9hnNkdNhq;5cx7!qUCl^No-
zQ|P`y%(<xMDv47daRU;n7#fVcCSi;U!(5m0c$K%j`7KlmIfQ_lFW<}4H(bwO|MBlI
zXn^^){$~jkwN0@yPOdaVVSJ8kt%zk?q>)AliAsXehnt|4L1=?cJ4C%Y(%#Iy_utPO
zZo8H3Test9b3{Qf>czXrv7y6G$*xj%H{-jNQN2zNs8#|NP9I_6>~Y%NHb3#teugwn
z(Q(B5vEytmsF59y0U-^(=hGiXoLg99xxLEt+zi#~45f+9RBw1SmgR8r)LEX~{{U0d
z6O3<~qq(@m*S>TQB0ol4tZ@3!DVCc9WT8ZgMZdmIltzStQIh!g7+5esY78pg7)Xi|
zETIX)m_#KQVff5v|BT(c_i@SIs{li(T;X5-%U|G;M<3_-srmE0H~*)GHWGtFoV=Lh
zVVF?Ljbm9hXI9ry#^6bdaxO>q+c*yT1EzB$j=Pb~dLR(GMO@YN0+%4GINey|@=ZIL
zEobR`8^_Lb919EA%>a^PjTWbyUD~}7$Hr(JsTj*R%+A_uDUUNgCRo%d?RNh>$8BlI
zS>E#;W7Cs9@#_2wId%)9YhJ{0nPI+1rB>n1Z+;8aLLMRDRaaihp_^~yi=X(tzlCE-
zsX9TfG(&!TGug2MmMu}SqQ7dBCgBAfk6a9N+9m2WHaPwkuYc7oY}<D6IF>R+hy>q(
zv9e@eZGy*|ZBDn=$onqs{t8N`Y@eM*3PGe4jb5KYFw*`L%79~&D;8MjwdtiXMK58R
zB1YD5y9uRSOm-;QF)_`PXAX0=nNaaH<M}Lk%R^d%2t{ISv>7Ga!r1JcsIjwFVytM9
z@kZ};v9eenCY(K+a-y}$YTdBKa;Q{`7!mQcNAF=w33@8Pj<UFp#37}WhIo@(v21}3
zBJ9#QmQE1bWi+4ve6=n$DN@@Edp-X0bDyE!>47oa^XLJtyzGA7@}3_WUH%<HVHk88
z3|eRDuFNx7Yok+z^dx2`MLJn*H;WxP8-^4H%g&G~dZ<M5hS$A;FWmhNPM?{_ybMwC
zmzC6&N~M=``S*-S3j(EywI*u>ORG!dT+LXiN*bpq3~3Nzxel3P9yjBUjBYJ!GzABs
z!E-%yu80OaArQ7h<VqA8(@>O36>_;e!Jtp3Tp^b&ku4V?HW-}{>y#i=thTzGIl0c_
z(ttP+*tW*^5@xD~v7*n|Seda(9$`sZ&2_r-3sf(?k%E^}-7-#Y(+mS)VY?1Gafm`m
zJ9TKqhGuVo5fW?EIe}%lNTFV20B#UC7R=4ga{C>3U|Ab~TuI8&BZvRiY+Xu8Azz?a
zEKw;}n4FlzaU8ncF8zKV&+~9>2g{NpQQ|Z~2u&Dhy4{e&M-Ef3uQM?*alVi5?;A=(
zk}A?v<G7MT&1EXCVRxxY5GTYyFbuIhhp|$TLN<fzIEalQ56ghIpyX#r3Wd>;y#S<5
z&UP?p8oh*>Y>g>DM`O@uyi#Q<=QCa@Lu^ngW?&LJkz&5v;rQ8APOlG$Ly2?^o|`aH
zN*K%Lna-D)%jdDA#hLCZ>nlr4Uw#c4sTiM_pfWqjnt_b(p_E4w+YF5&Oa)P=4Kk0F
zYP1wct47wuXya~Ff<|GHrB<@3ZQ8+YH(bke9Sy10JbUsKwh)N;#gBLk@UlK$p^Td?
zk*O3YPUex0Mc9jo`&q26hb5hju5Ez|2MA|KG8_=QS&pAL#!|b@_{=np>u%WJU3?!&
z07XJ0T|`OP>P@k!wjJN`!C5Bb4XL*lxbn(t*uHHiy>^>($B&aVS8yveq%G0HAQOjn
zX!xUt4$%nVnrnA3baO0=5_i4j2f-~fY&N<3i+{n9#~)<xdw&c|3+}(?UY_Z6*>%Nb
z9DM2^Q72&N_+&kww9_O_Q^MzF6fROSW!N}&F(WT#t)bNnalLgc%i>Qy@w@!eFMf!8
zzKCT>cJAECPyf_U@bka;%Or`qXuroR5ijJ}I?wU0Qk5|75(C5G5ZiSaD;Ftdv$)Qv
zU&yiq!U)*lI4u@2eKv-KY_hh4F`7m%WjZ^?q@Sfe7%*O`FrCjZQ7u6Xsfq|pLO0f&
z?sYkKZk_qH0a0in9fRv7j1^;Q**sH)DqHe-q?DZLt<hLnV(Rj1$XF@0iE*kk6RaD^
zWPFs~;8-hy)FSG(LFTYhjgbONjNzN`J8+yEaXhz!SAQGFhfmIrq`1Bp$C)yjN`cZ;
z9%)O$eni~M5`H_!K_B4^MjU$?jvqh9a=XpM%nXimaX7Xua9xS*SXi;)>hU=mvBlx`
zIvLl&MUc(qP*KcUvrT6(AdVA^8a>Cf0HL5643X{DXk=W2Q3-`KK?sZSqQgReot2=^
zM83dkz0N{!$V%M;4{Yg^wH(TxWM{F=rCX=iHC4dNC{!$nV~@4A#oBPlLZ`{8ZWpB#
z+e=eS%+9g&_`?j9WNmqw=Gi5t5DbF=DMK7DC2uLR4)}!<n0RCJC@^|dsTG)SJ4)3=
zaY|$1G++PLgXf=#jwQKb&+buA{I`|qp1VfHF;TxmG-xqwbZ9U3NYj*T(Z?$eP~`xv
zMkhW3gXQ`dW#i;*yr{t5z5Ccysj|Mbg4X(#snq@9AO7LXx%_)ZU^GKLGO2exNoZo0
z8>{4V1+)OqbIBEpSdNWxB+_xPgaulY3<JU-V9@Uqg%MT)Nu)NW{>;WPq{iqW`BDKt
zmnRHD%rGRiZG>l&TAIaXz>y=XG?zNWae{PGig`(GJf&L9Q7%@<W-~ahhY%LSXvq5V
zGQ;73a;AcxcPQ=IPRS_b*fa(?N;-tYdg9RV%dEABv?GPKT!O(6VOd<d=W6`C%fpY}
z&xQWmhLJ-1<@@$Aw`KFlMh-)--{<h*!!Jq#FAT@F$>nn7^LdKJBKdqC$8m_`7|-*_
zWHMMA$*@uiOMuoR{kncHp}yW=ZGDwesf=Y=e<$t1Z;wbDl2m~J+qD_!h=Z+FCbI<+
z0cGD~s$9WxZEz&6<06Fstr-l4w1a>~zfU&|kf|VypEHy;lw2Q~GYB0pk<U`f7HEed
zY7mUp*msC6&9Tmq{ijx0UF#6(7|Tw{W+k(gh{;Tesq7>bFGJ38L0GIu0Vi8abfYG-
zxf129ptN-x<67faYiMm@Bv`^`m^uut34(Tuco<=F<D~r|Lc;bdu3&rh5)R$}HF`~T
zetm}-NztycbN61h%}$=bSMA}DlP6DLzQg3PWl3x)u?s~UKZlde<ClH>tc#R3X{@ka
z2iwgeED0LIV1VaXXeB^ph`JHoR+HuRI^|LYAzr?(;L8nzrWKu|yS7X{*~#AN8*#K?
z`?jn2rC<FOFowqZDr?JU`NpUIl-HDVI9ig%AxWB$Mw-t*ev%`?<*xU?gT~ntoO$pe
zwp@7w+3mY<JeTEjt2}o9SIK96sxzBuHyV8H-usx^xs|0er)ji@Si&V0Hp5Pn#6V&>
z7^4tIVlK8u#d9tX#v)BDR#$qsp5)QT9^jsDewnwu<=rS0mIZHn+g*J8Pd>qu2M%83
z-{Im$`~r?0`X=J()+&>^3pg&7v0Mk-3pq9nhav4Cpx*D%84j_PMHHu~RAU=M#r1It
z24R{^=Cf3?MOx2s?ATbILo5x)yCIL9T480aLzKn{J0+8qOqU}jvPGsdlT<yQoa2FI
zQI7*o)EDVSb!M^!N?Ad1%T~sf!XF!d0ms3G91mX1@s8>xJk$UBci?!O9lQ6kZPV2G
zd({pCPM$daT{)H)a-7SNDf{?YZ^Ut;kluIUI0ckX)QRZink>~fIDWZOV&5fP%ffL+
zPCaRqkhgrU9h+gf-@!4^3IZ0Jb##;<l9V_e^~_>f<Z@Xm*&GwaB9&~8)G>IjMB0K>
z2%2F)iol2xOT8|QL6<Q<gDhmRlN24qc*<gLDZ_1hW|^I?;@d7daR~+~3oV!BL6?QW
z5{+;`FOD!;G3n>Hq_&B&SK(~GO>1E2oIJ^jlnmEbu`Gp11x6?2?3i3yCtEDwWgJ8@
zz|Q3{Msf0)$4R@2eK)-Z*W1X_3|PjHrV&REJ<aj6i|6lEC6{H-C6{8^-+4m90BIbN
zggxSZmtiv?7=)xM#ddAdFeZ%#Xsy5k>9}YOmLo8Z#w&T0$7*bvoTS;RzpN_k-%EV>
z?;D{pBp7^4B2mO~$UvqvyG=4~7O>E^g|Q4WNf~qk+T9*W65)9+g;JhkE>9;)SXx{`
zO0ZD4jsRQW7Cq)FMQq<CND@>U((4#f9kbjHIe2!!;#x$}G0fEr<E0Fxl7nT(lyc*E
zehwjR;xwW^955X83Bv%#wV5o8QLYxSBg54n_?MI?Dja*}Y3hsT@Nzjca4Jqoi$w<Q
zCVg#?mSLEL*g~+@SR-8vk(TvB83kdmu(<Bp>+mxE`Q88G!a3&8&VScz-LfoP&qdl2
z*LAQR8%tVft;ywbIL>HHJ5?!$Q39xRI@+ZeK^l(|$Cc$pa+wmY?GRjqq%P4KEd<h$
zSVEDi4$U~?M7K%V9vy?GwuP`XZme1F3|Q>786*ig*JG+!U?!JGMG=SA&tO{`V<b6g
zgAJvkORcg6&-Q60F;Nn7uC0kp!l^;P(`y5k*CKr1Fj;|ZxeQ|&7fVKL^5-b~c`Rwu
zOCuJ80nMOKFA2y=$yjL<+w&D<B)IE6?`LwX%p)g{vUGNV@>rF`KqqRG_<7<^hd7N9
z2?TBzS2?V&ED@}>i87M&`*jPV!Vp7_EB9TB?>gsqhV!fItS%lSb|VZvknr4DyCnsF
zA&*lkj2;Y*jdTRU5*s>oF4B`AM*VFFaQ!SQ4bj>|7>~4{5Dj}Qtu7;TIdak=Qt3rf
zx%%5fnUr(QBP_O#ld*gz_Pm8_k6(fnr}XQqJo2@#GTmNeZpSoA8&a(a2LrxwaG7r`
zba?MC{4;L4{Wdy_=lIlPPtjgn<kXW7k*`c}|J`4w-3z$=&F?@s9$&rt9_&npVY5SH
zsZOfE5h<Zg$VrQ_>FvB~-#)(Z<<E^oxxY(TLSRd4)W`dGI@uZt6(_M^eYH=q*ygYQ
z`m@}8<Et2})kxxmTB*c4-u5=0dit5s2J8ROh%ew+2t}H9Xl-y@k{cX55+OBSVrX;&
z7W-{_VMNw*nJyNY$>uRp!oih!ERkZ2O-|Yf8}fOZ%S)T_9FOJ(#|v#uY*OZjL!Mq6
zu(&$nc%p3Bn#)qlx(F*|Q)Z5epTm+i{WNAN9MI?w=*B~G(xO_LV@JMvp5v)nnf)h^
zv2<>c%2<`yKs#!aczNR92FEFcP7g=fG}c!M>TQy5<2Wt^K#nUfyNt{Q9G_lYV`b_1
z2FD&F9l7T@&b)wQ=S3Xb7jlfj^)sk6!WbWnPuh=(20fNnSCIJ}*^9)nYgtGv(j-M2
z8^_L3^KAB%YqXOgmeAOV!YGT_aTzb=*j%ho^nC_V%wng*xqcVRbGdrcR*bPgSoD>`
zq|ogoJkeYrN+QNw58utOrE1{`hh4cW*Ul8!JmC{W7RxQciT;4KK^vodP7mj420gS&
zDA*oznF6(3fwG&UpDNaaMUM5FC|~mzzyF&otu?uI_od8j+e&l2jw_(zW^sZRZea}9
z78nJtokJNzZEONLp2egoc%$=F0SGNI#5{K30K+gsh|y`<j+rT@H}6DXjmBax^n*2{
zQ?PUxpp_a8e8n204VJKoh7oC`MoFVJnzT)m7)!bcp%9kA&I<fokxet3NPB(K<fRW(
z{ePO&MMfiqAY*|m1R2*Qhyzxd>ujEwX4q*F*I}(0vQ}@A%h>GPwS~#)X)Mdot=DO+
ztU;pKdC7M2<CW30oL1+}v=gO~Mvw+Qx<QAPg$Cz3nxji8u`<k76ZUK^GF7Wl%opf%
z>Nsu|>3V1($haPj`U>H&hi$u5N@L_QnUQUfpCxXsQ<|Q|E0<Voe2b-1hnbq0#j!m)
zmWAuu=wU#j4H>NnMT(0>bLr%Fd7K!8CE2rg?}k?#APk3|eTLQg+D7Nki^nT0OJLg;
zwj9~2r8*@}V|>>~sgbRZB_+bLu!O*}<Y)@kfH4|v1ifLv=`&~9Jhy{FE=L$e{Cy3{
zmDX5T*dq%;JXs)22ORCLaaCo8^-hz;W`o7<khAR-a=zr|ZI`fhau&;ih58yN)>aWJ
z<&N!_GFclN?b4M(8AB=zVXAPA#bD5-8Fo2Rud_HboL)~*%1{|g*fCyaTXBpn`2weV
zb+S$t*YSvfG0)>zdxcie!?!%<vSUnoKAsem{2aYji;3D8`C@@n^#Q93D_pX53%=u!
zngs2aFv&XUFv4*H5^EVN9iy?hpGG!BWzrWcnQ~j;OtE+8_VY(J00)nspw&D@U&a`!
zz>xC|l*XX7Li#SkvM^$FZ&Rgk?a}lrLX6rH#OUbSwrvJO1;!?g1G2s*8g@B*?i`z@
zwo<8#6Lnfd`r_zZy_C>KF;D|O@t1$efy0M*?PZr>7SC|~n9CJAC$I#lG-ep}dHCom
zcb{ld+d9XMH{HtE_%vspIZUTHBw1?nUw`^1x#1n}q8-Ng#WL4icMF4l$m5SahDa3M
z#dU&6gJq38T#ccd7{s8@lLsEfvE>Ua@x*hBB*6DG-1(Y2dE@I|$45W<5l);sNs^|2
zr_MWrVK`7U8$(uBmU-ym2YK6F?*vPtbi$k7_<BC_JHNx~TK!_(=8GN|a_k6jWq}~=
zbF{P0m6d54ofa$YCX3yGGp%KO&*G-7d$@FJ7TbcwMxEnpE9f-l_HBEa8Xp@8j4M5w
zkQRm@RXE0CIOxy_yBurOSscRY^_V0vRH_L(#!GA~ZE)PJlXWwAjz=N{RoCZOdzDtu
z8+kWps!aMmo@G(?b97rRCdX^!iUm&Ahpe7k=8`R&FW|U{X~d+%2-gXS#S$VNqp|of
zjqJ$R^g@nPw~28kxpddJb9|Ck<K*AYvF9S>a~zu&a?AxBOB>sE7z~FS91qF*iugRo
zTd0)Ah}%sf{r5b6l@enFiGsu!qBJ1u<k?r5=F#<Yj912(9jh@{%CK{6k|0#vd*TRB
zEuJOSuz6yNo2O@yj>}h%Kg-g}I=<s!<mj1W8$s5Q#6pwHR~XNfxol#b%{hymQyz{D
zji$%Zl_tkqYn<wK*ioKh5Dy9C0LK{WVZh#*S;n#@A{}vhu+DNA5Gq46=usZ4vAnWM
zuhU`AWPzEjb3Ak6Bte?sAdq2#HJ%&U*GU&wXGuiF<mStfT5RBdfoQ+BF*@eyXGenk
zX5*QB&E88W)i!PTLnuav>d$|lQ3e%_1}23ONXr`i%rr(=qbX;jQbsn43Z14nzK^sm
zgcucLyj+oOTeop$;W)!V7va1#w~PNzlRBj}8R3w#KnjJB7FoA|uq5m44rf;TJaw{*
z0^WShHg<0wBU{K4_S>v4FXL#*L~WdWv4E4yfMcPNB*PS=H7ZGQTnEQ<Fk}(g0u9yY
z>2p0wS({78ET$%NlqwktxiYC%l*cBqZOJegU`Y$%*hptoVy#vu_<z~^^I*%4`@R$W
zWaioK`tIA;+JUOVUf4H+6h(2dD9Mt>mef|a?GBIIGd+%Q%N|FEyL)Vp?Vb)t&)Ds?
zC)#e?a;Po0)!HRe+!p}EP7o**P^i7??c3e&J!@vpALqVT1&Sg-7HCj4=ffzX@bK=v
z=bk*j%y0g-47>_b7?6+(fd~h<We<tV?(Qycee0Xt^Wc;CjS5CNByolwBzQs)3yWvP
zsFLiDz(SNI{N``}7OBx3yX_d0lMSAI`e}Nd@Vn*!SqpKJFdPmU4u{lgH5}m($1!P|
z5=BvdqC}FACdoVA#aKa{8h-DypTl>`JoL~*y!gt??CkE|)EHZ%QOd)Y0;%$zOC{C7
z;Bvm(Wu?{T%J~hn81V7?ALPOLIi{<1c6&|U++3%M;;xBlW@~lIl`?n=jl&>LG1?HP
zDOKOgbKGnN5r#BPn+qEqN)?}jHOX9~%50^?Vx_^rWE>u!rR*xUhy7d#yAG~WNZ_{m
z6b-+ED@=X}H3Eh)<x(Dg@9nhNc;gkEhwsCwHPC?0EOrp$NFVLQ1Zf!^^)Wj9!{KrZ
zW~k}6SNQ$UeUX8-96Pwk^hBMPUV4$<W|O1|(0<3uq*Ayg4}`$zge)AA^lRv<Mhlte
zpe7kQiLlm?B{A7B!g1XF=G~c9q+y3IedhDHvcl4%PteyH%}a0H)EIw1q*=zRZ@kIt
zXWr!gxk*lq9YTsS+Q#TK<HDsC*0xd(jl0z6COPxQX^u<7g|p|dvBql4>iRb03rj4&
z@dgh)_9W#>omXFdh9pgBudWir89if(#OU-H1hLg@?5ymcu<}j>O2JsI&Y$|}Kh59%
zAO0Sb6BFEa{5XI8ul+U7o<BG8o_y~ZW9hU9gki)hFTcQtKk}ngN@e0<#L**1c;JBt
z`0VFCpa1d?p~LRm?N~@fNsXhe!-Z~_)pmzV=hn$YpO2q<hzIBAn5m7?>bE$vu}0ZS
z?iinBt~N%gQU=e#0R1FC3_ea$D!xZK@G-WE3?p`TTAW|+P^ttRn1Z>n3bWM`OXYEf
zCgaG&9A#IrHSFUkiR0a9$J31O*|CHWfe3RumV0)5=9{?huwzDc?B#ZxmC@lJZO1m$
z3_2@(_KRO8LgscnUgPDLUV3*smLl&*7Y<1VHFVX>?KsUa*^PD_UAN<S&yF3Eu)~)=
z`vqKC;ouV=W?(hVOJ{FtjD=8SCL=Z$qaCc#4AT~hlsm>79Gn>A_>n0bsrbEby~#IL
zHrVL&m>iqp)Zzj9!t%!MCaaw`!(Ja7XGoyxRZxz?_gwmM%pi=B^f2Wz(=+4Tx!7Ph
zh`G{%)mD$y-jHkkT`J1u!p<uFBxAl>C(AThl9FZ`3%u3Zpc(f_tiozb*;;HRU@(j@
zR#TsyWbuJVNuPOvtlLDHH1{-D0y1lGv?P%lC7oQz3G&>}>vsf{#Sg0d^rwHE5{9#v
zuCle;;fW`opgdVasQl=!Z+kTMj{;PVk6SL|*F1*37-coC=j4$VKaWN_et=Vcr(ly9
z+`#42Kl5plw9n<|zRvU=_q?ZZ{6IyW0ZPiDEDV^8G_zEckIdkuOFO)<64G!KpT1*?
z+ZM-g%VpA3lY}XY2M*zu1B@}ao`;Y={ceY0a|aV8_@y$XdKJf4gi%Dh+a=C43rpi1
z+3DfdJ<^F9tJMjHZi&<y#$69D2oTa?yi_7d6XG<%^Q*)vrc|x~0xcaPose2XfCG+8
zZ+i!864p1@=ne;zCT9_zPrtbfsl`Rmw|OpD@-9t;lPu%k{pzpsg)e-bMx#!-T;l4r
zYh>9U{VJmkS(Xw-A>D3=;c!U5-zSYz(ljNGW3nvG6IzYISo4nL%gh*bY*}5~#E~wp
z>ktlyH$BGIg6lf?QiDm!3{+%*(t_94u5tBJn^Mr=@e_x+cY2C)rOF`5=nYdIT{=R!
z9H5OM@O)4n8|^k*yW5yBVXRbQvQa1S9d^P2m)jlMspi1c6x-chWZ6O10(Ps@7_WlF
zFzfn^l}ZTda9g=dFNtX-2{o@m=)_D^DhMG+giA}u^sQkGg_Md;Z-{Rqy3H%J!VcpT
zQ{cIztu}-PrAj1mADJqWEJRz4HKPZ8Z|E$j7BoeV&-~ke#j{U;ol>JgsifF?`wcql
z)x7IC;kVsMO*$B&qlhexNCq8};W*JCA#*i4G3YoQCEglrrZHJWqBY^LkF}P*cJK#H
zO66-fK?7wiX}5jTV|+7c3oozl@)!T0#ghjn`Ncb?xNEG!+*E_51(!?xHs>of{^c+K
z3ghEtZZFsP>4z5RYLC<1gst@rrYFXD{K+TKT66j8CED#4CQWH!=v#{~EE#zY@7~h<
zhtn7YQouy5#xMTjpXI;(n}3V(sR;l+_KA=4SN`fR@mKz<zd{y%`?$Op1lB?nChTmt
zSXo`>^y{zk<P$%F=XzKPAO6r&eD3pK00ia-P>p=ESPPEx9qiZ$UR$}wm5XhHV2sC)
zALi8TG?i*@$DMx6Ba4TrR7z;03A_NQd^S66Hg~tt!<eyBiHSy?lJBq^4!P8B(@r%@
zQ`2m9c2O0T+i`gk>sN7vVaD;P-(bi6B&I#G<DnBVRmtb-M7nfLOb<id<aXTc5AjV%
zclQ#lutQ_dj@zAmI}YPMGIhSI9m^){iXOlBum4w``P#^i%MM#-U;plQJPgtC4R$;j
zCmJSb$DmV97Ja)N$0Rx<9Q3hfWXHoM!R5=iegk1OX{UYDV=M&u=HDog9*%;(3DL^(
z(840O9jXzg312<8%!}(g^m{$3zC)uL@buLS40=P9l7w-L$wKDGCzzj^Lsd$&hXZz+
z9fo1Xbg9DalM_t1iYuF2%rzS5ev0n)nC|Z)qCVOx*3*dgaDX)#XEz7be4hzFpquol
zWC7d#K49@o!i4gv3rSmOjMg-2Rbp9YZ2Ab3<p#Sun~=tM6IJj8+Gcnvw>>E&qqyoj
zd(8nzf#-Sr^q>0+{OBh>N!V^P97N0>y&Wg8d0Fyz$W2zRLU}&QE#cM1@v6I&DhZCS
zP_D!&c}V5w9$Bf7%E|x8+~enY4vwpsJ+wrYMi>K5rT(7A@dFigQaIF9?w&MRfFwak
z#kK8-S5{-Rh6m>)hvz+nmBitIa;3_|{7jx4swBSSptFq5?lxhwi6o;^8>3urAbgi@
zzsIoK$9EkZsc@Z$+m6k0Y3>A%{@MQ<b4QM1vxwHEx7dB*>)7U1#snl$I_hYjCpK0~
zWvG^L47_#uB7H|tJ9IClu^CQ?5<y*1uZ^Ld3Q3ybIRTYQiJg@YX$=;GwE`&&>3i6T
zhe^!x`Z_Bc8@bE=T`FIUwZw5kzu%?ZZnE2K;QM8KHzEv&MDdU`iOJHKG)qQ@qThhE
zIIfH51$^RTALHJ8?%}gv_#D6WyT5l6BMd@#1Wq;@;|h`pDHUtoA<L^8Ly3b6iW3t)
z!YbN>9<#MN#~0^D;bKA6aY=Q?m90&7n@wDsF<T#FvN3_<dn|W5Y<9cU9hac2=%x`T
z=9g$L&GQpK^XHjaJOIY9wYJRiE3cw<R#^~&UYL?u%P>s|Jck*-jI5Ojtl*2wr`ZxN
za|iE5H);sSCGcUQTF1z7veZy@%ap4#G*>SmFj%RvQ4LXDqq8aUJ^GM8G_Pw5+G<ue
zSJ>^f2`W{LNf@--Bw60t_S;EfjZFrmy$<PEn|N3yY&LKL566|H!<00RFj<VrV$!6I
z7HzDl;0I-NmNM)#sn(}(DpQ<z;=?SSIKrD>{4DFg@^5ZxjNcDpZTRw)ZJu4;<>A>H
zKX+o9gMLQ0Yx(BcZ(^0hK*wy`7UxV0>G(YOz@w<yaSj|lN@IMI?x4rY%2m=hqEfB0
z8z!J51csq~mm@ocrQ~}2_$PmyzwzJwcT7x8?4JoX?00znxo3#qrSbP4O}<*_Ot8J(
zVtad+mtK5{M<00-H}EmqaOa(O;dwq$^!^;(_I@LTN6F3Y*xq2r>%Ae@R&qN&H1BY7
z(g!2=?fAsf{I}V$=E~Lv+s$2EtC^{fF*P=h6ZkCmI&8GN)EtMPEa|3_lk-dLEY9&0
zKlA6AT{;NHu(`U-$}8VQHP@LJg2Bj+`y)G^^(%;4g}}m>mfv6#hxtSIq9^J|$HnvZ
z?0AASHB{U(r9C?q1}iN#{$A{OnP#s=P_1E1%%JsdcHAfJc1Y`cc07&~c%*JdI!wvF
zlO5aY$c{6H?Om$%DctHbCmw%_#gj)l^M%i_{=fb5%?+}R3DX)@ScKchQo@*wrLhV}
z7fVQGxwNc#V=Dq_n5foS@3c8{bvZxJ)EIhcit8wj%`GxJJwqZKHrg$=Tic|=n8S^6
z7A6{W)0AjnXn2M<R+gE$c9~!QmH#hm*RFz8%pW?+p~oLbPc5-x4Yi7ki=gItgmKDR
zIKXy#45j7Mhwo+qSh@HL&Tb2%6XF<ldTo^M;QIMF!-HOzsiirrPzW4^)EFJ$d3o@Z
z8`9$7`rm~Wlq)rYqsK9s$y1n|Jn;IR&SDTEKjP3Wm+>kS_+gLoRFf=>$r1;z;^m=M
z*GIZOj$c8^03i%!R5&V?!Xy!C)Z@*seTl25&vNjQC*IRAexRZ*5EfTLszGOx)GGR+
z<@|a=J99YU!OVC7stVEtt%<@A$8%9aB8?;(Bp9O!Djqd|7K}kvDp(<j!VqZ%({r;J
ztx3`po%-0IM*Y4Ax${F$B1=_-)yy0{M(>_SxcHSnVCCswXCZL#l!x#8^tv5{E09UV
z+Z#Q;`sOAF9=(^(Kl>uL4J7yfnLauml7=a6xt@ELf`Hk%dCtA{W*#?p6*{?|5%*r$
z>ty|2$+C>$a6qr$VP|IxM>$kVWvn&Cam-*aU^pBC7NsOw=M&G~XZXIyfdh*iJ9d=S
z)fMi$_uiWpbxK;45@eacq=L*k3=+$FJ0{WrlQqqBO`)7JNC&j0IT%v%Jfx5)9J+%T
zZ4BepfW^`rLL0nl6)hB<;SdGfx3EB}HT^hdIt`FPjfs06;naf<<2ZSd$iV{#xbvP<
zyztDkoc_vZxU=L_Qa<&d#7euBSHdL`&+T-0=E5ZooO*&UzV<A)rG0+%r+$Vc3K^t^
zpj@FO92%7}6H|*^J9idKjPWEUvEWFgaIkx;z#lP$+FFLg0Z|m@BKuv&+FHoMUD81f
zr&J^AY$BwKR}QeoqT`q>9*~871Onyer?JH0kZ!w)7bu*HPh)PL>4S^3Hn*9-=iZwd
z-}jqX8=hI~@Zv_7V`Bj)4$e`YnZR`<y;j89+9r1&80U!x?jpMHUfxPm>g768NiLp!
zldY`{NHcV$MyuJSgds(s-zBCgfcx(|#sA~){9Ptzrbk&0mL!S!SO4-~@Y}!hJ81ph
zJ!pGpF7MfvCEu9!x&yYiw>W>{EZbY_%uUZ?jpXQ|BaBt6KeVE5&yIB_(5XQ0*>Ni(
z(ms<lm+88~aj)BPYcQnjdwD^Cb?6O}JjJCNuuz&uXpL8`p{1fT7$Obax41yIZ^tPz
zs4;o!Q66~s(S18UbnqZ|-Fu1`o_>ZmzV-*)QSzz%Av+Fvez(ih7cX$|)RTPS^>1){
zHsqr}^)qB~^bR{7uaudXTx40C#o7epitpKu2g3njoQIO%lO0F9q{I5ijyDj}!7G>Y
zX2VHLmJI)}9mipA$G*a;`ZVU}nLe~cYioyvd+xjGF-~oMWS--IbTvw(R6W7cM2WzS
z*%=5nyCG@X!|D{2!(_F}AdWFclZlLa)x(jBW;bFbT*gR;q}M?i?7d}In@`mC-2%l4
zTHGB<aS2}BEl}Lu-JL>_65OFU6e#YbxVyW%yE{Djuj{&>Z*af6_uDgTWo0tw%<S3w
z_#F!Q+9GnD!y1*Fd;l(C!|Ce@qp(c8_A!)kA-QClz>TY2xt6=BYk}L#7?*dtccm5~
zFk)@EJ8<Ws2M456XRz1JVe+zY8rJkOn(`14mbF#3e~Iw^rr{Wj^;rW0;Df3#WFOFR
z_QWoA)<i@>Y?>DrB)%Ryc5@12_iiBdE+YH$X68UD;D)hn9w7gTro}n4ZYD>dEqs<?
z&@mhXYZn3Nk5-xJv+kO_;ps(M=uBKzU0@_`W0n>b3GD&OHrc(gIkzGfR#wr7BQK{A
zjXx4l`il(+hi67J%q7h)*r?~2_k^xTzjQ=3JF8Nf^(4~gpP>syw@}6Lq^9OK9jHsr
zvGmX`pR9D<<YNjKi~;(8*Af*iZ!p^kfaznmv^2f0N-*2~f|OTtd-oWT5=RLWw}Q%4
zP`sG%%*)bn3PzaH2IRXe=;|(|y+inMH~_O<e|JK!$We~#DHy}(D@r5^Ts`utd06$J
zOoKYN2y#denfMl%Eg2}IEE{jeX2eN;0X40pMHOSh?RQd^jiZtAPd-_LqyEMa1o}~4
zuE0IVld7lLbZCQ299fk<B3tJkt~$(m8h7U>U}Oaj)0FjnT{08$qBLCknj2WINXPS+
zNla5Z${gdKfmcWh7gq)YdQ5er=9qQzrHJ!(4T-^2+YLl<_`O;Gyl!l7pS5S$u~|Ye
z<liw!N4lN1#*C*Q@H1aFF9FAZ@5V31Jn@phvi*l{ONv`jP^aVGqwvQ2ZKIwVq~|2!
z8TuQ&yD8=f&CyA}WtJq=HC*fgCLx}mmDC-HubC@yiET#jsc}cDl(;P^%mGHO;OJ+~
zb(%<H*LY~XNUz{3{8^z!ZR)u0N|$wgV}`n`x=x4xhTEf9o~NzS=~W~Fe^{qHwsAQn
z%}Nc}!unBWAV|BTyOD(w?~h7Yw4-5a_42uLLa0OA3k)8M9t9z7*=>V#y~wY7b1Rt6
zTq>zzo;z<R?J1_+#(-m<hyT2prfuujUj^YFvVuWaCRcEt<)dv)wR7=*LZK!bcX#j>
z>%IwxMdlcyQ}|>iW%_159${pCl$rsi{euHgO)g2lADL3@ay9+sZ7GCA2|`owbJQJy
zqQ|mgtTY`#CA;Gi*kaMIV&7j&#jjXL3mX!rW|sZmNZ-IZa0v4#xd1mddN>PndPvB;
ze-u48>k7OwO4(O!XYEgcIoq$7zEnA~lp3Sjfz~|FI9{M8QxwBye;k@o=!&O^LJiSe
z5K=t1<)aGffAR;!j`iAj*8JK&xVP_Lyf^L-UWqP3KU=Q8Bymm5*=~uef(|c7&9?lM
zO(QE`qsjMSet6d<g?~iLY8Hg+=M!hxg*h+?8kUJuiMqv2CJ*)D;)v;|Sh1H8CXLrR
zMO56=6zUx>51wx$-pu5V1qIrFvCS~?k}Efj7(RVu=d!(V4+l%~0|17Fx%+cX+d<Ee
z_eUR1Jw2lu?cJ+8cq7i8TNb~xm{|+pWrxY{@hg~wfr6Q>#FCnBmX!c>_eWF&c|0kZ
zv-g4Qjq}%lH#YAHi|do)f^MN(>rWKar5YMd?d!eSspM<A`R)hg5v1Q6N`<d3=KR{`
zsX&GKjzdW;N3%$da&9%$y-2>q?Y}UvGHFvedoQo*$acZhX`fSP4C@Qn*c42hk~cSA
ze4bX&=T!HtD7RxZ!|QTp8&h*7Rm>-+<>zqt%TXyf___gw5y&+CRB|T&%5l}{7U(HZ
zO#WhHgGsbIo=Ev*9CahENZ<f{$f&rPTFmygPDVZ%6`^?V{p~WR53eI6z`l_kYILC|
zME}A42{GZWVHP2=OvG^p6=#9Efdh9@<{PTFnf|N|=oxJPCZ&Cd<9#p1>yXDkCd=k=
zvE4ne`K%?f!(-AC7Z0ywzVrq*1vq?#A<{?Y$GxCn95}M`iC$Dzmqj?J`-2Pzl`Pi3
zS0Sf(W3XR;D2QB{(INOMZTwBB0<Q|<;v#|dp&}(tTf+QP8ZIhj!o5Sy_oG6vwbmqY
z0iux>sXKmlG3(xgxJ=KiI(Dh5Z88QeTdhd~F<%{AkQ7$`LtDWf1aUJ~OXL^LFjJ*N
zxr?_?@Sn-wZ$-^&(W}E<K{S8*ZC-Au(3R1KR^AU1Ue4rL*ey6171IX)N^y=H+=bfm
zg!w+;k?nQ%lkShq(|X?V1pL$xdgd>%Pa%75e^CGs3yb7YBN1%NAzb`TpnhX4A(;4u
zJ@PF$rR!N&?Qf7fHD)(Boehy8Hmj>-Pb&COUou_%CPM+r`tWBlo;W|hC*n*C6C)rF
zYiz$kkUmz?()0~rx$^7NT(Cr2I6`HR6H#*uXZ-wwpDc%uTfuILLXnfD^#XODMbFqH
z|0gSk6|8g((bYj4(VVu>doRny`WsI`u)$MlK+>|kaGyr|mBd?E`@09^-@mI0(P1||
z>siKK=fHCxY8iPTB2s=T_3)RKs~#q~N}inQ<BPdPp2rDnvccu?2j(HC_0hrWpv}`o
zw(~j@r(lxCCkDh}=LIu?;t5CJJkC8?7X+KB7tbDgV|UvhVhDJcG{TKgQ`c)3{y8fC
zhP2(AJH8?hZcFP?;YYgW#F(}2lP${H1ULV?NraLonzMit2LJQjl(g2#XbH(mVbfCH
zjS}H&R&>EPIEyy!!P@r6FD(x~?P<%G@3#)l7Y{PV4B>;8#K1s}49eSiX$9#oidDkW
z{D4n++29jRN<oCmy*NbHRxncpL7jHbE5)mSPtN?hQW{%Y7=O)nt-%xO(9_MHgFkFm
zA3!A5gTawlu8=7%EAq1*A8g9nHx4bgPDX%6zt!2zq9T>7e#c}9LC6(Gg%lT}ro*}8
z;S`rG{w9KulkGt-(wsT4Yl9*VAi&z(0n2eMFY`e3K8pJ4Qd#AH%4nlw=YM<XFcAsg
z@=!m1U6kz^&!v)^eY<$pn^-oiSw5oMI)W$sj@EWrU&MP`^4<Ri?;VEhX<9fY1NgG`
z{?ORuyT=*S@X(bD@r`O)?u;}h-rbUwfl%A;v8J8>!rEw}yk9yN*ypvPQw|2{ZF-XK
zUc%e+b~xdHR#P)9SouC?I*k|k!<37S45+@^`&Q+hO^Afzsds9EGODMCigiT3Yg>zm
z3bNYT@z2=&h`Rm!H;JtP<EEM9?=IG>Irs|p5A?Mi8;^SeXc1G%0zls}nk+@T=r^H`
zk`E_Dc)-R*GehbM1+4?Ztfp-hQlAqB?h3t~t6mghet*3B2&B*`iGQt;BWhP9%^30Z
ztEAPT-C;4)slWFJTG20jp4L@AW74c|Psy=3U%aU@jg1Fasg*+OqP^r+_1Yh(chJxR
zl<~h<*88A#r(WYR=W+z^Ki)k1b+MwO`xeDk)Z^OZ=$E=Q_#|&SW`IhZr{}lFSg=?e
zsK6o_V1IrJN7Nt_`XsEOib4*xMA>w?(QBr;qbAX+wc{#f=ISLG#Fs$+{S@c{gd70%
zK@myPCH^jn92S!5o_MkHlk06`M6T@QB$igF3$=urxyW9)myYaT_I2sJ=h>8mko&%%
zyh1c|&#JDk?TwI4Ga}q3Z<?`>D~#ul$ue~E{^&F==DWN8N}F}I(d@?D{y@C#Avn@=
z!-*JaaHUcaA|H37UIm)0!9exl@jGuLLBZ!Cp95r;bHQhFUqUpbMWe`O!ox_q?0t5A
zazX6^pF+pxXxS0b{0sZ+EmB-#mH#HM-ygq-7=jBOn2Pw30`?cLUtUNQva~^|Jrlb6
zdm||0wUEgIV?)DxfHV6!q`c~ryUthtYv`izC57ceRrpo)B?Ag*<8|3vDgJgCvlj7K
zF?@2v#8F$T#%NZMkB9*KFB4F%p755V^qa5@>V+j~Bnw1^I6op!vQif(U+v|A-dJ(l
zU^Yg+#+ID~vf(GNS<*d3Q#djef<y=FUBlu*GLBC7#F4Xv(HLbO%OfdrWeq02<J&tY
zFfNgiJwaml{lIne{#E+TN4aC)yQq|=hE@f_XGFmQ=c3`)#`XRpGC#-l=zyE(=Cmwg
zDTTSkwN>*L5UsokF$MIBHNNmib99!eku~X=e{AJT^Lp3llBZ(QFpGfHc12j)@4-_b
z{U>@2<NH(1sXArk07v)VnlWc%@ZI`h4_!Pmp#=UVk^%##K!vRr^2g4`MJG#g*miA;
z#>{0WWr!EQ;#{&*k<yQN`a@-+>u@gdG*jRWLBS8nURGjlMaA_#=OPJfCb=jib$D2S
z;zN&9&vmJWR=YbIi~My3TyUa1IxBZKogvf3!$pLI|6Z0Vaav}N3ws&j!FOtw)?3Q<
zw_6KduM2!KA^-UCwm0Mq9ez_CqLESJtx?j%*1#FUXeID=(~tSFOd~TFg3BS4mx%Y5
z@mS#JEz^NhB_WaPTaCrFg%42SCk3AjMi++s2zk|mXwG_Y%GtAsW-w0Z)dRIYWaMU&
zKQ2oQ7yW_}?Zuj8w+TY0Gvud~Q_n37^Br+ZKjpC6Y2^yWrZnUH&;^=lc$_3;Z`j^@
zux2QLjNxHZyczP((9()~fU*>qjv>?+zBNGOqU9o7xB``)og1batel~bOD+@I+y6)b
zq#z4w)TXA%O?3U$Kl~mcg;Hm9Z*!$(bQYM3!BOg6XNoC}<87lyNsSD%rA#DknVivV
z^)ml-_EWE}%v)&{c-t2ds58q+3KC~%`?Y`q<5EPSnbHYMsh5|8U9x3^7<^(xV$D3_
z=P5nQWHWsq21-*J8rwY3GcbPS;E)&g-225-n|L&OP0!cfJYegR_-58E_AP`{qaq%q
z*Q^&|iA!&ihh)c8|K$K1ublx>?|r?UMg|A9R%DSHwJEPjrsP%K^#pVaCQ%W+ylO|!
z+aM8HX=X=3!eaa7wWwPu7-y*6Yq6*9kq*C`2X;#}EN8w5xIfE;YW*mdpVHH(a#Q7u
zEo5*w4P9_fDlZ3ktQB3*^{4&L+{^0w*f0gIQ9$hbTZ{@;R8}|PP?%YQ&%j8<8tsTL
zc-ib~ybp2&k`v!7Y8gP@#Nx;jtxw+!9u;AFZed?!1;>{iVcE?N+}GbFju9dnF1lvX
z+HO7GDPn{kKEF+8KQeS_v?*e^H~p(m3Ug)Br1<zM=+#rtbyBzFdwXc{(C0GF0iGrz
zgbCQ^aZWO}H%~6?g=J!5VxPnB#uvkvJh64iO$@G%y=_i?!Jw4?6OkVqRnGPA6(GXU
zVVX?;vN=)&J%U&IU0O}46WLL1P4rjV$oj`vW8~hQYGM}i1q1g^*J8cqO)&ab*to)U
zkQ_^Q{rD?@q*()}p3(z6jY}_M89LMns^IJ=uO)|U6O4ASqXmIGf^GehfXkhLMoJ}{
zdgaA|9UZ<oLk+itg<0l|!ayLCf@dv(@i13IBmc^&S7H$zbSSs7`mmNuXj$2ItD$v}
z>2uruw}I$@533Mp0^?Bm)nJ{)x0$5dV`SVVI53AhSwn%3RFvS~>Y0MUbbY07L`k95
zN(oXTJ+>NFUd1qZ#dM@>zWFU%+HA;TtaZtVo{5`!e0aLS*6rx`skhiVesKw}&K`0F
zY1?0MzmUyH(}?!>&kZ`(S8^VQ-p6_bJyhKkjcV-}Zvz(xorK)SGod~;M5FF8MeBSD
zi%yf7c75n1-3kK49Zr5{nH5y>LnkY6m9%WZQFaX)bDo9SZ2OF&MLjvIG%nxxn6(*0
zKyW5*%E(k5)f*=~A$|dvpskcI4+!h4yjQ2WdN!zoRVh3yiYCO{Q>lajO4-jrzy75q
zcqF_^Crr0rYss^8T9SpA%u)X3iim^Jped;C`2JCxz^XC$N=|xR0-U6xAGo~>L9B4?
zz%HOIm_~U#f`Y!C8UFY`(8mbgP-h1`UHe{c@eoK7aB)F3pL;i@MUd^R=7*Pkq5in-
z#_Q^scVMP_f@0~<k#E`(f3e}6dXDU+NxBy;vNt2g{exK#*=F&#r{=ss#90Dd!_ntX
zXD_^GObDo<db+#hUlTD<t4!xLR>>IQe$ay#7FT!Ie9rf_dWU;1N5{=k8neldQJY?7
zD~<9IO<>6>=^K%J(WySk0tP^MN=a-gvH=UdHY1d<cot<~JScrWg-Mnk3E@~yl%#r+
z<Vt5h!`FnDT2pgtviWTPc3*fL5}Dm;tAW&#Iurzd?9eva?Xl9itP~Ih+iy?xdQ;TY
zRNwn_$L&Ix_`N_&k|SclTGCS`V|rY8xYocYIrkJ(e$1)R3)8mCkB-c6)TPOB6Y3K+
z;y58Gi;zoGp(<u>*iTa9%4uSG6-=?_BX)5<roZ(XH9mnB@NOTI94v?X6V=@#t?ism
zdw=}BW6bYW1e*@YAmB_w-~X12G?kt$z%sf6*^!2wBm87lma2_oPCH&Lp<sQrpAN1m
zS}tFxZ{<V)0iD#?-91?{KZq?~Vov+%_2a2M5uLNN{Eu&jvD8|!mGgz`XY3qYa<#L0
z!Uyj;>>T##z1wtyd}0+R%TAxerQA>S8@Fvu`;8<W7B&q{a1Z&P-A_&zt6i(JsM#?y
zT4UGSs)ji<#9DrOWdUyyD~0d9*4iGlM$~_Hpp+0>R$cBbKX|MTE*7=-O-<$(K;_5G
zb`I6m4*kqlxV!Ex8OxW)EWW`O+5XYFhg(WxPW4_$CMIXRk1}(5y^kmwj{QTZVVS<c
zB{b2hVy1>gBsJoJHZDf@LdXI>`6N*3S{njCAnd<^;=P3{zuq61%4n}DwNT`^JrS{g
zo+-i*2}~5L`nXvoi?3BsZ>ielgjRh<HOzc_1Ux)r<Wn5bhTSSw!YTsNWKR@?FJKj8
zq1>_y2lR82@oQ7)Ywdku@y|rae&VEy@q=q*M5<*B&Low)UmKmSZ0lPkdk1Q)3&lJi
zw|8Dbyl{&wG+mg|M)>9AA3t7i(||UwHo}Cvqylyr{;uJx_+*%<(#1_(jZU-FCR&~~
zZM#pYXUm!q+kwh(cIbxXCg3S>=H>Lf_J!M(+U8%H9s0F{iPVY}6{;?mBIVco`L_iB
z1q&&wSdd6;k1eM)c?X?ES`?cS40$FF<F*$>#Dkp;2>^OiOjLw-j$c0iL)@e7!{{=G
z6Q}K?#r+z(5WPDx#NGbf2c=VxtlaG+oSf+HzQ00K;}mypHfh<sXqF<je_(*1&~-+;
zNFiu1pX?RLH(lq^u+Zk2wFD09B#$yLn$F*p@xNNY@tZX9E0hvAsm{>mc1L(AQ4K5p
zi^|K@ySZE>1Cnc8&BcPtp%yW}sPauwcOj$CI#*`)`OWW<g+L}C(W<|_bam9+exb)I
ztI}o{adgDCHSMr@h-2mFTN%1f5YN=*qPGRdo;mUBX-3v8J$my&oP<I|?jRKrTL&x7
zh(9?+UH$PStrYdoxye&d|AF1tYn%6jc8k*1=uyVR9Zp3vTz;lF$?yL6_wCZkf3Ggl
z{lZyCn;QNN?Liug$H#IfJtb@BmJ3G$pU;HGCQujc8l<~Ew!-nXk_h>4QJW0q&=HP|
zk4{H(ES<Wo<RF{<r23NhTZpmRk|xJkB#s4MhsW=WL%OWQrq_?vMg#M=+?Z`1+HehK
z--#4ARn+`ci#g-pQOrSsG6ow+kysFoa|AQ{=xt@@Pb2b5?EL)QH#eLd92{z;-|Ob6
zW!U*<#ZK|)K6*`S+eqhyu%2;%#h9w<TYC*zA98VAW0#$7?`Zy(N<}DT_i`XjJ;CXJ
zUa#%Z3fbrDiIDq)y`&POT_QhUzrxKG8V~d3lOT4!i5Mk(IH0AaVd%WwKW%MCP=4*5
zJ~-3uty}oT$Cg{8`j0%co9m=W9xLhbbp4gMLTAfFNdV%|PqaS<%1)C=I9+xcyn*aj
zEJut4W@d8BEJ)E#oZ4MiW7ApE2WQ!`3iC_R&DH7HG!-%?;#B-qc7G~{tGn*D=-;jB
zaWlF<8HnzujnHsdi#AK147g^04YoicZAuY`;A3L;#RJvhqLqFl`U>}{2w{t4p5-V}
zvHf1P&a0QOE*}0s7wh@wmmMOhf%cBls!MqF9YKS^3*qdb#M`yO)3Uq*3x$NFt;fKW
zTL@3bm#~*EmKP=&tjjd$#$0pkyfZ<8tF+dM=fU4ZZ+z&Ovu0=sSJ?uQmzhn?B<rr6
znT#%Eo7=cw_s-BOSJr7RD=TLu1%j0n6C_~=I7@!?$m+)VjNjkHe|YE=H(AGi_A}Xr
z`U<59Ik9|M_f2nF?kbUQExLdBCZSi<sK9f?{=V=E3ER7Etdgw$N9D-qRn=9>2yX<A
z78u$#({C1_zNVt}c(T2t6i;*<KfNANkguw5>%wC`zncBtJyk-s+5S}dZsWhL@rKrZ
zS-75y9ft4A2xj}vC%c+$_E#v<;dA8lXa7x>7hsqY@#d9|g;v0^U0*l%_AW1YR<pE(
z7a;8Y&Lr9=&|VF58v8T@3>J}9yShk<M8k<bR|Uu(aQ2*?fD{l?Ym}ylrY=774N6Nz
zj-n64odJL4hv7k`_sn`NeNRV>x7B<mw?_)?_lgdYinx=W8vxsQtV|yw8{%}kpQ$pK
z_`knl7Ey4Qa_D`zzB*#go!=T!j9JwuiWe%1&fM`dNk-8b_I<qiJf=UK>h+-dQg84i
zd~MVIMAUM#7y!MC$M8YGA2u&xS_|DD$C&Ah!|kJEd?^%u#_`Ak_PAJ#*Iv$jqRk6|
z)F@eVrHvNz3z0$Y>9x&Unwq1}8l8?KO&KMR8MCgcgUIP2Geb1qMlqH?cm^KQL!Y7D
z)(fIY&DC8KiLpkawdNBve$%~dK4w?GWLf$QG=<htX$f@n^K@1Q0;7gDpDU+2h_WBv
zSdPNNNARCI98<|mXkwebJ?SZVyuLvTw0I35W|hiWf(-m3ZuNXM=1m|#CLW{T$=_Kn
zJx4Y3zBuvGnxe1+Uk<D2K%T06>*kOreCyq_zuvF%<fie^95V{xyoZ|M^3g>z-?D*Z
zM;;E^%(mG0p9Vx@&K+r$w#dVWv)NAm_wr@LB>}T-`Ni(AhUc+$*Nj;9PUY*A!OlDs
zFub=r{&IZmPCN?Y^+X-dpef#W7pS)Ar$fJ7SwtyocMG~Sat$W2u^!r|DvCBT?Th5M
zInXEau#J%KHT@@>10(;TFkY?|^oMNM{p1iJV`kca4Ix3d4Md(ML*~Z?fhy)wdA)wi
znxx-7&J);HZY1?z^sVbyo;9;^aE}+L<~^=7Vzfy)_OBwQu?yN)6G3>sL~#abNH&o)
z@WGNA{8h>N>-g@Pd1(FV5PDC$WwvH%@9cbrDD3YEEePv2=)V%bkGDUgKlTdWXJ0(@
z2gErmi!BO?HTULtB~$BD9i1CfpV;3&AiTf0wA}Sc2sz)AFs=H<n?t&e>g|XRHxR=r
z6dzZPhv~PkXHG<ZOC}Yxrrn={DMd5M#qd#0mR^PGorAAVIIC49(8T2HhjLOVOKAj*
zs-O<ES1uj)@kIIb>oNU`bV$u9l^Az?^2CBNc4y>H?F!awI^nOdm~K7>-hPja+zM7;
z1?P>+Fvx`cw82`zqPxuB<U?ywiC%oA$&-QuPzH-0=@eEv!LH+ar%W`NW=DGJe)}!5
zK^yLsvyO^^z#!dQPxj8DMOzotl(k*{3-ZS>wn;-upg0w<N8BVd;*v1u@oC<mgM{ef
z0{fmrQ)j<-{oPLpvM}%rkdcT-t@cNXG;f6^X_xe(6n2>S_hYy_d|hwtKYDsEDm>6W
zn_9l4$1ho}D=v+pValyQkwNa@_{9~n6DB^VprCMl5sZEvWzjBMSa9IlK(7G<9u-qw
zm#b_Mywk6(uMVw`cP<?1DZ#dNhx&Rfqw+CFU~{OSZ85NR6)uxGuPtuu@=zH!DnLDx
zUM46##<BycH+j~4)8|t>wAL>iqXkE($ew|f0{2V?P;L&zXhASL3b?m6(E@!t5%2P^
ziKiJltFG@E*ITz{j0t*n4&-0w&vE}`VJE|nBxvCU1sONWe?$6e)|3tg`D#F-IuXe~
z{5>*@=8c8w>!eqRSm`}5jl`{rONcsqcCd0aMZ&`k1p@s^Xh`4(8N1ne3@(|j;8VNb
z?z3R&3XG37RFco$F>ERYBl+1BITyZpf8---F==VTYs;R0$zl5@u;l%oBbqk1*L?^f
z1@YpWNKb{(M@Y7^a)fqM7y%1}4-yCxvWaY%K*)7t(zPK#Fh1aklV3sK^YJIq)xAIK
zN+L;R<=OJ(PDz5TtG4(uKPn%fgsMd2pC@JGwtm3CCyRuX@epzSF!tdUcgrCDeTOj4
zJ)SXv?&B}bW#!Gle$uUKvS+x;)*AtXm->Kj99C3P?z-uMA*%XarE{0ELWteH`v%*>
zoKGta{_&Yfw=2<jcrN_v@%B@_PO|t+Hm@OVgktK$^QTRXW7EfM+1lEaPh#$<+X-^w
z(2g0l8k5VnC8rWy8seMw{S?`fR6fT<*!zIg`uXhM%-`fk4&<vJGSSxA^w&{*Dj776
zd%7mpSB&nj4NL}QcUFk&mtudR9-#I`$ipmRWb}>@rttqEmR;KKaE)kGpYwVh;?<5A
zFG64M4@Thv$Xj(qvRo{`pQZ4+@3#tEUcTdC;TI60-Vk=5ecBrH+7j^x1=q)?LGvU!
z;W(!j_R*WoVB+Tq?3cbr7grMzEgRkK+Sb}=i_c+4-)k~S+~G;1=bGA-O4xgYli=sW
zJ9c_O<@DxUawYC!vUp>lEZ)E*p-onb3+yr+xGpKHxYE3g3HLc|@L257W0piFTj(8h
zbMZgCCidLvU(jEpoc#_Tj216w@!kxst|IFgX{o8{3W7y=GYPQmydS)}VwaPXTc5W1
z#rC!>!PeI()Qi9Vgk$1+d~I*?5){s7f@))9Gh<z!p-9j4&}md&IDcGSRmIAeaMr2{
zz9d8c#nyJq(DyJ`guktmeb3iN7GC^?{(r+W;Bk%GPvn226P@Pg|NpG%6TP_J|9SBB
z8^saw|AzfXWEBnN{|QGF<?w$${x@_2MQ1+a{7<aF;N(^0{7-D)=3uye`rl~$@c+Kz
z|JFcl_+pS*gN^L$qO{9HF}8x6LVnx_)+g&%_IGu`fczbv+VQuAN&lNV(1}s5^xs@M
zcLNJ!s5CY-^g(BwP<RaUgOH}azPXjvp^vu$TVFs6{r{Oc#$X4(HM{>ltx-Dr0m`I(
zY`++Ozw&tKiyZ%<Jfbw}bY3Fnd!mY7Q`@Xi$T>VUB@&{H58J1WVI06yeXmrA1CkBl
zk}j?0lQ;&y%zqUP+<x+`h)G25Tsx+Y-X?ETxJ9O5rKK5WubY!TB>&liAIuFO)S<#$
zftTZp$8c>EAIJq?j;3uxzWzzerdrZO{*9vqPIT$jrg6>~{L$v)N6}4svj3=Ccnmar
z!>oM6l_EA}vjtke6uES4_PyQ?10)4BX`;!AyVQr-NBSNY4Ibo}NP)itSUQeu->=KK
zO!=tf*t@7{p>6E)329(z1*GoqfCu6C$k)p!|GtNSH>vmVfV&v~D3_EqafX=q^qr1Q
zcywVWa3o&EWkSG}s+zUg6m|YGNmXm}@ZKD)&hoNtnFBQWk*J)fir#)9{rvK+DUI!I
z{cW^;4_)XQ^*`WXn8fuaG^4|&v*7jbSMh$OwHa0*{daEk_m+_`vO3>FPdX$ikfb&L
zQ%2A%R1_NrjE>v}(idVFk1Ku>1%^-R=9JS!Kk4$m^OX*#ATkP>HIB^nvGWmaJp(GJ
zE?=><(p4C})EH@ol8|GUi`n827YTkir=j4DT@MYjxSYM1<<aMZm{rRerkJ`lf1W=H
z+~sEPY-d05R>6db9OGmzzS`7q(+p#|{*8zaXBz&7n7F+G5sC&S9&^CWIx!pk>*+BK
z=2_<(6@Kl1$1b<QYjWxAEw!GSiKk6Y3--9uc%N(UC{2bqU&3fq{NC@0*iIb&>}o!#
z9kcgofIW&kooTut?a%b#g~8kI`_-Smhi>5~lb5vj6S~I;vUjJlpDY2VR$1?Tyxi)I
zX~YJm?zuz5>)u&R5wTr)1>f=p>Uy;!V{zrq>^gpM8y1SnqGRMAk4x7gmZQ!6m?Y3j
z9@hXNCdNzSh|;fgn3=mr#Jn9qv4%xLU32!?mBKE`w^1ZZnYb!ASW;}k_ITshscD&F
zj3On(8(u<A#2kXW-7`0+vPes-0FWlG`JQsYOPIl9c9ZiXT1dRK*K16*bHE#?`UWp)
z);UGuQoyfj+<Y#4xyt6jxjE#7StLclH*A3is(^?A_si7}QppzrF@2uU7T~Gn@r(1f
zO9MJCHYot<e^il^6MLwhLY5aX&LH@hUbla6zGd<{*^Hna%BJ&EHSs;#qyuu+_HY>Z
zL?{NNL_DEN0md&%vi*O62>xXzx;Z6A4Qn{IkhCoKPZV-{H^)FsF{*nBo7*KR;NU+U
z!ChtX$2RVdf3OSoZxfsHPao-d<oJls;T>IjLNaerE|H@Qomcc7c?zNUg4m*nV6Flg
zajag=%t}a~b{UQ|g%fdoKxGI91Q2UoH>bl2W3x!MOmtq}o8u;$Ow*G_E2mljlBS#g
z<ntuPfH&_{tT66m?fcX5nxF<hZ({01yMdqiUf?Z(TFDQb1v330nGeo2LjBC-VHK5a
z-ec9p`U?tru;Y|TW+x7P2@ZH`Z#l*fAkVW`0=M3cngK|)wq|*bEJ?05x;?HJup6QG
z4Z)u&bdLFFhCIlVBDrDEG;NfTF$_of5oC-zOjDV<EbHIBWFZf2G5e2@DuUM)=rVFl
zvAY&%%>>U{0I#-3A^)~KM;ICmI9*lHUbX=On3zXZOUDh>(wyynfgteey9p|U#%QK}
zhGilPFCy451s^B~^!OilD!D|FtFRcZ5P%o3zQP}_!t|;<zT%dmIVH#T5TuNi2QGGM
z0Olq&tr)%>p#Ebr;2pnA@Ssc-$B*KIv2oe7l5fK}__ke$q=FMqjbxR+S&U<|#;d<R
zj*kw1(GzyIeYE6N$1OkVO)SI*#oUvb`A+5Q0`MKcus=!|I^*)C?teDdCsWHD+<M!d
zpHx90okv5+U%m|6+S^Yax<_nz8hpd^eP0>De_yH;lB%5*XO+Zh^gJZe0D71w2|we}
z@xDHp%_TPi-TGU>F{8!<taHY79ZSAq`Q@XhWB4H-2k2xD>52P03BDRaCik=L?1U(&
z;Mk)ywFJj!Ba37bVvP(~U7ltBD6L4g&+i>2$%SZ1o0J7{h00U+u;+u2ioe^4JBg<M
z2?+)Si;&T9PGf&(W=O?%sAA<23uqEEB#})b99_lbUp{_l3rhrOiV>sen%&cRg2j?~
z3lx9tHy=#yDN09x09Zli&uU-3pZ&YGFXQA~UY;Rx<F2F!8PR_U;)3~#Drpmfp~m=K
zR^(d=j&kX2v6{0=${(K$<%CGS%y+>-P8qEA8+>RF<DwMM(=yb#cf-d-;Btn9SfHh+
z^8H97EsNB`+#bxsqGr_yGNhg>{6-JkRdYXRd>$Zb^b;;2QrYC2O=La5@Iz!QJnzzt
ztW_K+hdBFb0mr3QZl;eAE(tz4(A|RQ5GX%Bw)N>F6fi^&E>`a^o<CNz`M$+TGYs->
z#jrt>`Ags+zl8um0FW44*ua($n->*5PDlxN)+eE8X6}r^WuPSM1`JGkQ%M4Xj>kSx
z>{xbF4uB=KGsgG3Bp9-D)IMfz{Ufc|#_;?uyNfJ}_>%KO4J)F@O}DHbx)nH2`B{V{
zV(kK@r6?qr1&9k{L8~|TWU0bqGc)*+C~`N<oLYRWqcJZBRHob41<|ytLcx_D<Sn65
zT!%m5lBnsWM46lsekC}>Cnn-&pm?ZU$Mp+@e85Rf?pbmR!I3`RIVLsSkp|18y)(lw
z;E+R!65p{BdyylGhJb$+TXB`+^!}Pf(nmSf(_rwKRy8XWt<ox(gv>}E?uyDICB`CB
ztM6n}h4c<gt{9_6Z*O_MN!w~O5f*-yma(QxgxNibzz7;cmHg3RXx8UhACgdk*%WK_
zNl|hZ^TxDqwsw8z&PQ0hk}rxQD^k~rshVvBUrfTrxBSxW7ukcjig@~;rjkgnnBz+X
z%A>o7;B~(QktPM)=bpt_i?%236y~z4{kJVTL@F6#Yu9*;8DDnTE`0E4ln!#x=D&VL
zL(Hxp9~gT4k-`(9mZDC29@J}BNajk)dIoeAJ}6~e(!_}hi<<3SPfKQbRdR&rP~p$c
zor%kg_4g_2!sci0;Zh=0)$ZJ2kD5AHIuW<uBZwjJ{DhW9aMnARI6M5~x=pE*_A}7E
zVPmK7ONGJknxbU1dlc&xGkt{2GiM{q=E>L4$Ak#+4a8wc<M>5xjQ<Z=JMg4Q7?;nQ
zD3Yi#CQ#FtY061i23ZudcTgebvht<0$f@OHJYg}Z$EqJc+0%uLxjC)$6kv1@*X8;e
zkz#Tw@08!*90-`TPU`KIC}<IW_os#4pf4`28u|Ey&EC2tYd~pH>xLGMGL6e7_@O@+
zo@08JEwo*I`0*2r2x84BHam828Y?c3hL2Ib{h0Ot4aF_vI+k^%qS|$a`=Zd4^$`Zo
z652?i{KKHCydP-(y30@O{fOR~nwT_f;dML`>B?6c5-gOUigMGepzV~+3BaTd=?17Q
z<g?9v8^TgelBtTfw+M_!rl48o|GLyFbX=NZ+2}5v0BsL|IBrs6CzvHtA%#(7s>giM
zV8)Q7NYX~jAY^rSup1W~>N5n3AO;;b(K{O%+Mbn+)X^w#lm)$!8{4*YJ!l3G4k9r(
zF}Jejswa7ch>9hfxi*s=sV10e{UiH0u+H6Xan~1>pJ=M(f1U_hVY(OO><=pZM69t<
z9C#UlT*1QYY?~fN%0G5<gR|~^{k5S%hn_kOgWe32GP{t**&syckYb=WhMT|~XOT4@
z?Z(ZaXk~pxhAA4Gwb>I))qh4E0d>`+9%I@cceq+Qj4Ug&!|KJ%+Q$Hx<k5HD`a<c2
zGVvRpfI{*dc6GW7*r7t?wxXuf1=K8yLCTm;R&#If-_K<JzsG&|-gD~fTUZi*E?}oZ
z(LUEwlhNEOR1|bHU;()2`!Q_Xc@JJM-?*OygF}jv<?KQEz9GaC-XDVCt>UzQP!-P{
z*ZGjJ+o?#Nv+O}Ecn!Ai&Po$`vQZlSg|@nQJgnS6)MjIuI{A<{;v&-Kp-d%#pOZqf
zpdXjIHAcUd!QW)clJHB7ZccWOQ8g)?Yw;U4xndi2Mm|R@uF?`^E#EVpSD%Y}_J7~?
z=rfiXq(?9z)?!k{f-$1Bm6-m8?o!P!v*27pa(3CFnJ`6eQmVB(r9J{ZkSG`@4)r6J
zVo0N!zo#u<PiZk(*If?P*zOqd<R7cim#|XC%xMiDK)4_c1E}g^d8%bHPGt#FR0Vx;
z7{IF5`k_t^7Rt<L8N7WAI^r@;>JBr8L+QaxQPwZcc+twpbP6g?vcFrVgCC_XgMRVB
z9mogDSKr9QQ}7cp&olFQ-6l02x;L-5Y%%xsAJcP2SMRv^3QatZJGHe+MTMlRz%XR`
zp=TX?Kl2|5=C|0B557GkIrCK1ho0V=js5lya%_^@Kw_i`dt7v)r5#o)Wo|pC3@IR1
zqxT`5%<0ux_fD=YGBCfKKmw4pW`N9yH|YzMrPBg_u+ieh&z%_rZ_^iybjvSH0s@bJ
zz5L?=>EgM;KzXU=uurew6R%h)B|)O-VSy493ro61iz|iP+!5a>L)HSuTR%-N`8s_U
z{V}CU%{Y5LY>d&Le0us(TP)tpcl&r>%V2eH{2ebl4^%N})qvTTL(Jo0q#fCrBe^y$
zc-KF&C#k`lqII*EmYun~)_gv4Ua4$})@?K}HwMtC{0fzjepgqYUszenTeQavRl)4@
z%Z_8b50%?A4(hpanqnLm^os^gY1;5U-*E}uY(6tU%sS$2ID@(1U9+@1q<?=n$EO;+
zoj?RnQAx*XPp#=~Pd(009L54qHtA(H7|KSAHNGVN>5IrC@Lt8u9Ai!beJ&$Evq%Nf
zMugow+o=tj;Y-@upF5r&A<NKKNvbgTG)0SAgi=Rb$UzHCsG)Z8p3;S&u5?s}E40BZ
z1tc5E&0GN{I*R+d4{5W3<Y3CI_bUx6E`E6>HQpvpITKrO354L^sM#M5lt=A|>U95r
z3?}{$?08sO$#Lck<}5wD=~VVDN<Kd}&DjXSgOY17cv3S?t>H2LFZzEhz5pw_BbL!G
z=ZI1UY}0*MKS_U4{s8(xJlaWV;9fWPOws!aksNqEH^$U7=_D+0P0Qkka)E_qlq-zu
z%<Wu~hKIlCYAsV-3`1=g<INdxhjYl-@KFf>p&zJaz}!gu8OK1xXi~wxuF2(M4LP8w
zp=!i=bb=S=>DUKEL|HV{`74X5p%-nEw6LJhUeb=Pt0291faG~0USKRtDRmKPOHBI>
zo<HDl{9;KWOpO~K@DVPnXo9cT1x`H>hdt&fvbCyClCi#>bSEx(g|r}JoyBL{^wDq1
zzW#FM=-^p%s!Sq)wXd*NL2<3R6VLx0wH5_!NNUf@EzHt4aYmY^^Uve`Y5-4CBq_Be
zI^=|$8<jSK%{z(d_;mT<)-^{Z_4}^PZxV=ING$y>hw!{1#ewp73V$uEnNr*2BiH6%
zXWu2#pHk7fdu97zcbVn*OjcA~@n{!Mkh!&yf*r$0*`bA=K{vrzVc-0!atfLF*TfG!
zw{BQ>Q?(;*6qaB6G(#o}=E%!>v9^IilGM^etoU2fj;y*~*BQjX`hl?|E<`2cbHZ+T
zFKTBUM15FNY~|0S(m!JXbF#nhn@9t4@)V_iZYL@yR?%Z|kjvQk-bPkdLxjXLQom><
z^Lj5+Vkvj!Vdl5Mv#wMDb<ol}*{^nF@Cs)RV)(H>k&Ba#eZ0EJC-|vrcSh7OR|A1`
zRlXjze_JALB;)^9zVgjx^y~}(l6|p88|R@Oj5V72qw98fOA<0uw={2AU*U8*Wy3@a
z32Q#{!k^EEW11_N_qd+hTbE%dxRGiW2;!Q*e!hyeO>VyZXd;RJL?8Ce2AU_z#hWsb
zJhP>jo00x|c4~Wq1SmDu2E%OuO?@mP^BY%@Of^gIuQ8lsq?zBr7?znRI;Px9Mk_7F
z!*u-U3}2HeWcUpI7R}5cgF{!0R`E{jnQXg~5j{KI;hz|;^^&-#I{n0i;Px`%lixY+
zkCI@rtDm6fLKFeVb-(BPYGM2d>-q(|2F>M#jin*v;V-4oDq{7rEtD;?x%e!^*0t62
zG1}3Ezkz~=(}~mhXMU5fQ-*u9l-D0rMDV03?YJO2U-uCco?w3LR8-y=;`Nw>TwoG>
z-OkqHbSh^LVeokhq5!%PyShe&v#Ecm5lpwp4{;F6QGv{_#2?heSodKChtQg=TOM^E
z3=z>+ZXWm^EmAc68+c$=j<$<qq6mb+#Dr^Q2(Inis1_%a8qzqZkEaY;n)G#fwGfTQ
zwW01(X;SQ<6TYL!W8Jhd2q4V=>aps@Lp%C#v_BWY5#@lUE(fkh4fBf!`iFI_>Zsib
zxI$7C4N>WnB8{Y0TgtGZdJu}Vs94j7Jb9+i*K)J3#r+Wp2=qxXPbyG-@JsU<3ZI)G
z`qUCpOJ)8e@Ezs?kWx0tW25SV067mr=MZxGFn=Ers&{h7z3MM+SJpFg!W6wdZOaWU
z%{G|{w1daEk-90e$Tomp#dXW))sWiW&IgS8>R|C|^CF0txj^5M8`RQfHkALoDe~6F
z;@jSMp&BDP7tEZw7tkTmX5X8%V&51{A!f)PSx*>5?g$%ADl~HF&^i{>-4FD)3>ppU
z*G~3vEgb3%z<Agp)p!mQ5OL?k(?PNN6E~HY_>D0V<H>Ka*j!7p<Yb%Aj97&#SI+wR
z<P6StlY^zUK6%p4?qcUuYGG0+^n)2yqV)9vcK#)N$45vA28yORn|5A8ExsSAUFd;|
zzbA9}&D%<8!O*9&&-wL(<b~rwX}nwYMThgb{wElG^i-j+d->sIWzu`TOQ+$3-%R?#
zPExU4WWpn!CJZd}8Il3=wHeG8d%Gl41%g;dkF|7nS;97xTE9p~dFRq!>xz66_P|$y
zQkZaTPr@1e5(R2%i62#|knN?4KMFR7VEccXCE7q<Sk>O=PKv!%$LPi~-XyJK$)(z0
ztxMYdgrkHML1lVo57M%ig8nsXGgi4O6~R=W5k*A9xzm$uT3;iMP9Qb{kDID+ib!{&
zfG+l_v1jzqI<I2Z^-MmXeO63<0pVx%6R(z;TWaaUz&Xo|$V_TvE3QVbC!6g3-ITSB
zfldYhORso0>GU)-58#qbLWa7!W&oqRFqt=VkhpV+f%J>y(P60GHoCmR-y*`P8U?!+
z7GTs*X%=Bb_pw_*h#X6HX~zki-~MobGB><PL5zv6E)Ik+bu@#<@WFtJtA_}Khf;z2
z+x4wYDd9mLgD3V3n`UXFPW+{EZgtIq8i*y9aljG|Xe+k)s(b>GoxKn<x|wXP4Z_9^
z<=s-rY7&Du$QAlMHy3MlA$2I!z7N%Qe~S9!)T!zvEKc@k4~Z(jBZhp=4>@kp-05rG
zT!ZuRlgc>ny7C?J?Gp0dL2Nre-?zL)fNk(_*VHkvFmp@#GGF1ed-Q#7j>poC@8^OK
zx~9>!0NWwu&ml2u?(VoVOAwB&!yUxkE_m_KsqC|52Fj<h&I8W-g^!Q!?|r(9trD)4
zsr1=Camm!PwiO0cj@U=_$k^MTZIf$!cac@fAjbzH^69zTw9(FK_!%8etCKM2Acpx1
z-55vw_yWw5F-@)Y0;fwtjf#qRLe1raIHaFf90DbSb-78sD_-NLK%jUxg;n|<rG&1M
zIS;K80)%bTJ3y6^NftOmTZEteoAQ@&mhXvCL*rarSErM1-UwD7Oh0yh6jOnfERMwU
zCynTQ<$RcG0p!^B#=)=D?HN2P-5Da<5rcPovcd7p5OijABrQ|~KoketVou`}!qumx
zxO`LhUGn>$WqJ*(0NZLWkuJ1p1yW}>)*)r0jNP70^vy@Kp;=$l=2hn1A$+&WAJy^h
z79F1XJvSpTnTjYTluyAViBy+zq%wp?b=@Z;oU>wbYBmY@+rghv!(h8u`@Rbq;bWma
zgOxaet_SoL$51zQWv?D7dN$;My^?S6k%YVcjBHn1TGH~Nb3QIuEjLrwzHisr`)fXm
zP%7Nv``FA0Te=@>MDQ_k2u|W~p)b|fiZC`ecu6^StLm;j6T1`3>5Q<2Pv<d3ipy|c
z*<W(NcuDx?fZW54LX$Gx(fOUujtNMEjGt|g;yL30VJyX|KB5-683{$S@y@0I)Zcxn
zEUz}wbn%1TmGZNG;*z$Dn21p*J2DMPh~uT)w*2_GD*JHD(;f}rt6~)^hm3@!_cV(5
zpCLt1A&5(D^g}t$rPjcnF{lJdmf6hWH!C5Qq|L~KpZS^a5D!c8!rEeBT!OeP2KTtm
z3@tju-=_ezL*-*u!oK&uVlJxdnS2eh?rt1185OM1wRQ$V^!VkFb3*^V#pP0pQ<6j&
zMfVzGg|oC9yc68<?h8a1QIcLedV}Vxr<$cMbsq=)e1OvSLuaT#d5;P+$P;mMhnJTd
zw%SAM*Dzk=fg8dD_QwwY1<4&}UJrxQ&s1P%nLB5>JL4T6$k)e>G(VH&N>f=X?NNpF
za+0Oy<~sLAX;raiK~q+6YF2tvXR}3Y7e+6I&6nlhKF0z^z#Q!_D$LLHyEz<k@!tyV
zA+DR>)hqP2smR`W`v^HSsp3E;C%3Y^Y{+T4l`u;goSax9>DmQqnl$zpOh1@r#U}$f
z2``|^iLXneD7GZPATIAj%d9O%IUv4rg&G0i`8k)!<|-!^guIE!nmnSbRJ?{Q&7;)T
ztkOk*bnkm9@Onh3ty3`4<mvU3|BN?OF01bE&kp>Edao<HH*TuSw*41^T?K6$$ulp9
zM71)><`oZSYacJx^BY@D-*q#8_w>g_0@8o)YHibF?0NKm7LTAcjp3y^K8=(Vi1RM;
zF%ePSl3i%G5yjn|N>hcn42wacYiKpYT=It?WsF{e{WbmFwdd%qzgRw`Ncp2^+Z1<V
z$)W?gRk4x6Hn#THkNn1V`i7ju4OVE4ysi(3x3aLAR@@CbGuGQoqyl{-3pknfkt3GY
zN$|C`2qkm<5pJ>@DNc!$QB-+Q^?7^$O&4;Kb#TNb<15g((Isn6QJU%oi4f4I{p0U=
zi7I;B9|kKZ1Gjtj?Tyl)pyu`OgNlOBTc!f3q9{_?=dub1P-!AVitKacXc<KVr1%SO
zPWiyK0!M;RF*_RV_%>GlKuG<gi+y!Y<-S3ygF6$d6|RZvk7@$hC;DCEnp71bq$o)!
zg#5a!{48(+0xO1iQhZ@Ee$8^?HXi=~fbnOEQM8OmN?%r7i!>nYzVxe`6n{-+Rb+TH
zbN?p@-yQC}Xo8PcJl7Y#jTqTm*@TY?x;laa7q-@7tsx^jp23elgVjnu3ygPa$2>lI
zX^&#$ekue-c;^r0ib+8Qn%yE~Y_ieo!|k}nIMF0OC2He}c*)Lv+uTDsoQQSjjaX0e
z-SkI9(?MuV75PYFkt$f@m~e&cS;q;q`IogP``5eI*@Rw!4v38?%2%&ieNlr|+B^Cl
z-zAZ<<-;sLQ_W_Pr|1f{ceax8i#Mi(=|vHYLc02gmPN*b{!&R~eX#+H4`>zqmK^%n
zm;S0&{4$<}H$^!C$#MHDPYOxF=J+UQB3SV|hHa^)b+u@f1mrLi{(bFpJ`;AjewrlO
zS=NfV<L%dx-q>V10-{pN#2#r0pmbW9wySF#f05j8vtPHPxZ#}blGXcai#i&2bJ~!1
zU&cejPfZm)_)#;MCpKV#@*eaK6b}EbNo+*Xxl>BBMyV|b&^{zvyKD8UEo;zM4@`|Q
zD!@27o&~089t)3*2Xhs!n$@}!k*es4u+8~TbnX|%D+Fl<bOxL>nMqoSp_VOG@`a^?
z+P1ddnI6wJ6fSd|hQVtm@LawO2N}2(Yqz!L;6)_v`E0gzwKfr%ipgog<Z(z#y;84$
zPrV3QO#|Q{pwD@2v^DcK0cG0PyHoY7>j~XVlx1_L;WUuW_}%bU6;u2J0=Mbi-5U?@
z`0ADgt#6U}6ri{4!;vh$Kb3rs_Az94(L3lztm{P9SaS*T2WwhprI{^zEH(0_9^D^3
zP-VZ16Jz#LNZvQOUwTcsMppVUm;w(dVw);DJU-f_rKbN0unXr7be!N`T^`;))wXTO
zZvVKG8+uPs1o+~#`k}-g<IGg^c;_)j((&wt-HBMmc^Js!X<~l^{MMGA!cAtA>s<L|
zr<YY0lFukpP4?@DK#@drOFe{kz=bxdKA#(huza*Q{nVS%wGfo6O|m_Zn=h@jYpjCU
zwHfPP+uOHCn^LL2+f^8;+uAVFw+G)`KKE|$OHc!;;Kr#Sva9oNB5auX0rxfQ>{KpW
zG7FU3V4tQ>mp9qewvFRVy(O6-kUycDMPew0+2r9(*<%^68Vwg`WGCNTuP}JOu&_X>
zfNi}SSp5X2dL1c0!kkwSwvUiFgw~QHDpRDHnVS$RQrX?N$2WyyDL?8sbuS~|HMB%H
zYaNk;O@$$^pfGjBi3ud;b46vSi2Nfse!k{*zT1BG8DZy>ELCJJIWd&Clz3#OD@9R8
zdvX1%dB(CX$f(!~j*6vlHWhV>+B*zmEg|wbk{Z6>sj3hBYw*hYEknXaE?vTuwpTN%
z1_J;pNb&hKMU5xeuf%jYB)1<P|Ix<^8#EY)_^-{YmDIbLRD1ndPf{!fqba)oq>BRU
zl#MvFXOdPr4Aj~w?-n#xI4nOy%a9}hBH@q^^~{Q4gRj)>t4zLlWlX@nmbSsb{ZrHj
ztYQuNq8c(FEfy8QnN=AJ456U5>RVXgY%?$dB4RSe=px!%>bO9;kW;8%Z~Q;%u1Sh*
z{R0>gJ6&FVDsZv{k5<(#n;yE0Eq)izKawlCvObV46t%#olsfgPVH*Cl-RKo{Ctg?&
zZt3D7`(C~CSlMRK=Q0eG`DXo8z}FM^2#@xWUxy-Zz29knum%t>6*nX`_+is?zyzKx
zZkMPUABe)Oo^uAr8E#hAnx$kBm!E6^4-UqYcYNrvBQCM!&`(g#G_)-n3JONs!`*j8
zUzfC4$g64nsZoN}>_%JO^`*P&{{sX;`@U>RZa+yw#MGK?V#g)xw-|Oi^!r`>z@u6%
z6E&Ba5R+`nfH-Tj*R~88Rqp0%MD8HU)1-38j>PIwMuAianST#s^G#0_hdAT2h)7cj
zuaTLUr=~JG&B3VRIieiEpuJ06QPdru#o0Ml&VLIv^Au&H7&H>Hx7KkdXYfazFMxf=
z(!242in@F5p5@Ag0Uhb&p$jP(8!r?09kk7m$`VE?=g+Tm;^aZ3@bYRJo4cWPmh6XZ
z%zoHov;zdBFy!;-^_?q1<O3m6(s00~^RIL9!fC9@==WO8PR%hpF+u76I~gCZ-~}Et
zv$v6^20w5RO0u)FO%z46S}jUmK&ez>dS(hCByke4zP?J}I}EyQ3=YG%%WL2KD)&9`
zIP;6g5t)He60l`dI5!e@jssa{u{K2tiL`R>7qBKGBSpv(j#omuefs@D9=1)|q)DIc
z&GVQn0x9quiSJpewQ)+N5=sgNSxgX=Q8HgxjWOs<lUain7Nfuz3Ccx!rJEk(rKK9X
zn|UZM7w!pKrzj^Fjgw^MYKu;%PcW5xaQ3`$`B09=&|0L;a!&+#{-NXS<w~ki-rn_m
ztm~5>z+NkYJdZHn^%}I*XuIe7W4}lye})XMJ8GxlsF667)N3{Bl`2b%i=4Xm-u*=?
zj$_)LJYVnfm8)F3a+zz(E39v9u+wZ3hT$z+pz-+QhdKS~Hf`x3g$E3|<L)_X)e4nr
z!0y(ND;HX<E^o8Cx=FQKW-q66lw7r6EsD)6(p=Zgy&$kZp8KMY+*_xlVt!$f<x4wc
zS<LS47HjJ(3<n`&W8(;+*xcHoRIcMF!S>EJO06?KF+**vj?EIbH&?lK`7-O9P0n4t
zig6WEz+}y%R;e;p_gLT1WIDz716n%?hYz1%VtfXvz!iBF<U1GDJg0XL%R4<%nB1G=
zc^*PJ*9$02UJaUMc{`ynwPbo!JY=pHN0|L8SAnA>ncfpdc~!mL_fDJrwof<9bI)Do
z$sf6mhS3<CpfjJ>zIloIxDN={R$4srp(EUW`~YK(GRaPO{rj7}=)lPPF<LB*8TAqq
zfRuT}V*j4y9x2xes8kx%y>V8qU8C3E!tp#Fx&0Az7O}InNhV=&@i0QTw3@r4{&smt
zU)v00q1kQHs7(@hK9e&wT;EbFm$BARZ#cw*j9yohX+@T~wA)?uuuYn2a9o6Pu~OlP
z>tSOv+5}!NjsY!l7lI=dLRdWC<H?5~V0LDT({EnDVCeVym?WhZ_;i#(N>DOiwCnXc
zz1=3R<6Iv%Yq54Ow|LKMG-{tR5_j(h%1JqKvOyM=8FriaVTDjM@ymje>k<tGLc~}k
z7tdbd!2LhL=K58HNPzh~eP<#=+?#>}A{Ee;$Xx>)09CBG4@ncqJzRV9vNd_Gv~cSz
z-f;&9c$;u%2Qvt$bb3^4b?&|ABu*`$YEt6agA4-~Jrp?BWy$YS^<r9^9qJ}wqE;gs
zuhaIcjJpca-bE^i!#=`VqBJB+FEdkLVr2`o)L4}ZIwA7{A3NG`=qw|}$v0n8qNN5q
z@+!*!g^Mx?Ivo)92h8XY$EjfYO$07U^BguxalKsVSEL}01|&|*#@Y&xJn|5mktF-l
zm+46m^A+5W+{?H-MMVkB$g9tL^vHgoqHgJ6oy+I?DCr{ckaEbq4;*H9Bc#_GGCMcM
z+iz}i{@e<~@P4Z0>xra#)XX=pCe0uC8<U6kLYRB(m($w}X(d@Y;H@`aX8G!y^t)Tc
zaZI&3$<o{$hYl<dg=4I*ud=nVMW@w8%G{OPXpB**RH)bMOiWBL?1%IReFm*Ina<GU
z;gZeG7HPcA(!nKszs$yVlb4=<hM)aYw}a~;h2C#v0(!J?Da5E^YR{l`rmyb|rNEke
z0+7<d^8#Gg#|WRP*~4tL6rNw9zuE#a9M4jgKDQme14G`(H4H;4mHH@&+UC8jtQ}4M
zPToULq=-?&#+&7#hY#}A&s{;uyjPQ!5!JC0$v~3t+622h1J+izn4Oy1U+k^T^3`w@
zUeCQ1*(h8sM|qRZsL*7uXH#xyj^pi5yjp9tN%Gf?QhDvD50<hDqhngFEe8D#j^og7
zH%a4&ED0%<D;zm`oUy4TT)zsV_5t=gc#g{D5<%cIH#^INr%nN*9H=;EcXxNxc4cK$
zps~j4`Wnq<lkgVE*%K-wi4;l(ND&}~pi)=-*vB6s4Ey|xfBIEyrs;Q6E}UQC$k7Gd
zksy~!zQbFSdmXOZtK73&V9$kT%_tlz6s4eol9DV->2%ukdjmW#U}|y(qb-w@a}1-1
zx6hwJ=!A({6-Q{w)qp4*aPiUwc6VBMl?H3;+dT5vBW$c)!x4&w*;ythF5!6s*Rf<t
zg6~)Osh|3JD%EkcO_1iD&%qw=XeuuD@&qMP2r`@HeX;g>kL^`Sj>MbM8UqYdNObO<
z&cj$4Ns>2KH&!EMhQy4*oZRqC=Aj^^^7Qp@Z?69RkR}q>GiVcGY>bc%n%kPs{@!y4
zsrcBZ?!faT^NSN4JTy<cvsc|XnwyOQDRYlGViX0)z2p{d@X*<Jc=W6V*D2uzb;ic0
z8OAY{N{L#z#MXA3jn%8H?W}U)?Cbb`Ko|{ZceV-pT>{t78>we0eo&%P8^d=z7Uz~Y
zb??2LJaL>xt-{?W=a`uG$dZ&e%PTEw(&5VaE5uoh=a+CjchpZUyPhs!#C{$$jGE&!
z>av#^PzFy*`u&hJ+vKUIp5*m2Z*y&VnW?wlX72dCc#gzKo3}C%ifVNl<*XvC-0#t-
zl#`2%y?pAuDotbd=jiui4)N+8s?%lKD+#U};K(vkcvQz+9=`t=r(eCs?0kc8sM%a?
za^isv9Iu4VTEG}WRFMZYQtJFBI1)Q1;64z2i|iV#EhFVopoO*JNaS}wC0JEre&Gm*
z^;<;!HTvBqrX@LY_ylTflBts?aeG}l8*7xp0aCe?0v9npLD&db5oN-uar`vI?GIrP
zGT$GN8I51A(7v+9+Ikx)CC5+RM*G?|9Jx)cvdC_?oA+@YH9F6G5jt=~s~}@6O60<C
zfAe9%N{bNANZ`j@THfOB@i9_2CZ0Tk*SSX7H}nTrP)^>M+bNakB@sys;b1^xdV=-K
z=kloCU<gLR@kylTAXMJ0`@I$}exRc6+}o?@%tJ~q54jr4*tq0JKlUhEYhHfgGG|`f
z;OgadcDFl}mdg9D-D||S7lPRHbbWh1*tZKit1Qweu3mVP?Tsr~9T5&Y7;W&}N&1~#
zTJeA^jcGNviNZdUQ!{w3kJWh#gz4EiJkKM`G_L1!WPX7tO*nu49G9+Krqvy=y4EE#
zf{E#ra>-|Rr^PGJJ<TUR{wJw6W^kOn>o-7_?+P&cxpq>@yy576Xkt&0Sp2}v1zi$Q
zF4abjEdv}CP^*mNs4_x?I7%}&v&g9j9>DPf9M{|T7^zVun&UX5{FOaXX;BJXH}{0y
zEN9=mN)o%dYo1(e-E-d|u3hZX-0kBBk|<0#e}0|2?mmnXUhb6`o!(*f^<F~UkW2E8
z{(Ymqg?nNk_uTD;W+j=<_PuZdg)x>ajktLBb*@}KLzYEYli*0p-0Up#b90O}=5alb
zFkE5n;uYe|XL|7@Q`1Wb>E_j0CNI>;^YHR!&U@Fj*Y==N2`HBqSz27+v4<ZXd6NZM
zmeA?7*xcG?`PwQME??&S#fz-0uF`6^|5%B=t&KLh4JuseQ5~x?*6=YVp;U1=_0SSK
zYaP1Xl#3Tu_>mud2;Y<Y!e_6-FBfz-6uSt4-OH;3*sFRUwJ@^Au)Vd*=H_l5ink7>
zppN5YOiWAwf^R(i71p+!oIZ1!Pkrp8D5Dvl94Ad;`t25jL7#(%k1;+u#oX)^j#A7|
z)NnmXtJ|Vpt>HW1DoMND<*AQ+l=~lgF!vJYN)xy)=x(e;#paD+bB~G|g+p!bd3SDD
zsD-&9JZaFQ=PK5MD6{xtlxJ&`ywPx$j^>HI=l!}T&{&N{=T(t;AEjIv-y{o569}2N
zR!L1v*d4IB-e!G$i&CjVwN}Ckg<p~!yKRxT-n^1u=V)_%y?`k%oUm4m`jpvyF$8-R
zv3W01EAma}+S(dRN2c)o661|YmX;Rq1E0=dK!4aH2_w)Mt>%C<jq{EG;eb%NdDW`x
z<M}>;=Th=r;$e@K)fF0zDf;b@qX!y{%~tSz%W#<Gn}pc#`fIP?dM>UTpp?p;2sb44
z%TaFR-uJk1{IM3TGg_?;UR_zHK2{}64Ax4ntz2PjYMi5Y-G%QeBAbWtq!5(LHGI#@
zpQ%xFK#iK_?up&KjgOQvKfyuR8<JGMf7Um*=nuANHC-H4!E>spq(pDn;^^J89JuQ+
zW4`2bzx5J>zGmm@8%!MhG`rgX4%Q0nmP8*^NH1^QWzI|NC4m{{^PaU5Deop%eWSi>
zrinGD5uVG8w~kNDFbN3<9V(SNNz}vNTt!BG!Z@Pc+(dQ+Q<Jj{6G=G<nQS!hOOqs4
z;#<olCOLnt&dK>rWd9r+8=Lfp2`ejoDpij|vBqk}+Ugai4ouRN4n`*=`_HNasF8z3
z?$1+u9X0m1PbSZb62_8dnly`8Sy|zZxhd#wfvO>tk5jJUxtmBC;rcGlWP`WQZla80
zyS>YSnOQcj$m@=(OB;m#8O+^}B87vJ?Ds}|H$G5N_uxaf(cZp7ujAthm&$mR+4%;G
zi*pQyeXd<@f;P09F_$l`F*7@nHww+Orx8k_gxKE=?+ZF1^Sl+iXJmO?JU_8PlbV#P
zm)=I$7%X^hfLm6Se3vv!7^~Ot{c*<XV{GlL(QY@X*T!(15~X0AG)W2k5@1Nvgd_^7
zk2RQ|o#*Q1D|Fg@{F1;kmSMDmlx1r5GD+Me9JZ-7X7W5RN0OJlv!9@EN2UIb<6uR8
zJlLL4&{(9EIG#t~IrLnC3S5khSzkL#l=SJeR?s>b<@6|KXBL^9oyQdlD<Dcz(oCZq
zFBc&&DmutTh}=tP5(4Gk{Nr)x@GPrWwva;MsDM-(#-}8w9=L~{tsZe?`05wVa^c)E
zVfY}GvY&@u0C8hw<PAA2dpU`Eln^cuSb05!Ev4B{ZWYo&T8XtO*5*gh=rrcsx!3s*
zzxuD3n((;u#8GBvXUNhNVH1Rj@dHVzR3`9CxT4MZix;?j=`{`>zK3JS9wMktFp3W6
z3s>G%i@n=9ax-GB7$vyo3pQf!csieIb&<md4)esrj{y*dA)R)Mt?li+?aI}wT)K9R
ztII2FZg11>b_v5F=KYw#-$$PM=v{oif016-M+uisGv>t?-eh5JiOUz(=(H@pUqaZJ
zwbfmE-8^AEpDfLuGnVhPIlJ!t%1@vWf>B$6{9bx~h3%aIS)wVID!87@?EDnd(^DuV
znVOkm>F8}7zx^096XR3@mvX6$$ue}7poF4UDN`#4)T>Kq17l-VwzoG)k}a}0PacmV
zn3_ArPyXyL;simS-eDAAu2)?{o&#uZY^{`s$o9m)HTlUgVpQ?JmlY>O9{$gBh4TWh
z%s`lc>y7fztl1C2!rt>fsyfVzA(8BDdh$MDc^==*KQPaJ?JNri?x$K0$;fa6$$<ll
zoO$aK*1!wTo@Z*l%*^a0M~*Jwd6GCW`!>3Nl;b<RaHH`1evh&|r&s`D|93ULeupGZ
zP|9U|e2VdjNt7dLw{{5vm#N7bJ3CuQ0n=0SOij(wYHreNcR?z~CT3VTaFDTDg^K6k
z89~%(a`n<BjvP5me-IHTxx-+0w@G7cf-DJn``lamxomslbk9qlkGI;-XA}E(Ss+FQ
zTsAWplX7rrmMBiSw!F&0Lr3x)Qe!D8LCLqIkwuNtPF&B!4}x5f+tFzo*5+0JcCQsy
z9*(mXw6O@8=RDplT4(I^+8n#{1m|9B;&v+ps>I|txSn9Qy@aE;=(J;$5L|d`l@Hyq
zNw;$lo#n^<*$$ZJ6x=V64T~59lgj)gBl1xD^>l<hy~4^Fgn10d-^LEV4ocxDmrA9E
z?|V3oK<kY1<T$=k)Ef;p*RRs;?ouvSP_Ce>6|s$Ihc#R)DR~id2YfE=PIF==LauBO
zh7r}eLdlHn-DNyqQmqEqxP{}+=SNX$gU<4zON+T7moE2)CP|X(j#2*ho-@G+u$d-G
zHN$R~p;|;FEvg%DGK`yqdJQ#dUE?T)BOT&lf^sE;{*dybLnZJT4Ep)pU~L}$=Q1UA
z426UG<B(Igz1yQcUSfMgAce=EoAK6L7rEoa5j@}Fk;jj)wc4iFO}Kb*g$Euwf%LBD
z6#(eHou18$2^>ct6?q=Y4Lc}cWEhqtiRpHB@Kj8FtU<kYfHVmSdu_JYH#m9pFw(e~
z)X*r8;|UMPbr4oy5|DvQyV=V-PX`Vg+dD+78^lqUN-1A(>a`jRb8{qFhi<>0C(U}W
zynKb(h2wc{*j{eQ$QzZP9cf3NG?lyYjU2T=5u^NxJp7npu)^fuRSJ*-CG!N~Jeik#
z(enb30zEoIF5kW9i;5jJLeBp^Zx^!X77<ds=fuqKMt!`@%v_mHGr+-NV!X=y;w1M!
za3_O7k6-=eXGtQ%^0h6RyFE%lejevuW#rz@B~MJgo)DfNn?&AUkGx7uj_ysq@aOl?
z3YCkwWJs^ura$Phy}8MYFMgf#Z?B*aPjh&&$A*q+j7{QsE>WBiCmEifGK@mPWI(N2
zW!Mk7c;;nRE??mIU5_xkcpQP7=jcg|Fr#qcC}CcrM!Cg#u9Vh$IYoKSRUVErqqaK|
z&vO~8=fye)78iN=zI(A+6UQ;FPMgaYPV@4MzskmTo0Tn1Hwutmm0o{97=3U}mB03t
zvkZp$;^HVD6FXeJyv?gGU*zjwI!_XT9Q7M(w}xzPw3(Zo#oE0hp!~nxSONZy95hQF
zMike>9s-GwKIQ5po>xK;(j5%(wa<}KiLv?^Q&UsSFC9WS0U}LN2GV3ml7`qgAxmRi
zDbZ1cV=|P;H&WL1&}m9M7!nPKC<Tj4$M_5X#b0N9a&f<fgeCWa?TeEeMd>>>84A7M
zXjLGQqoazrzJVEeDEGfVdB`P<EoE1X3VZZ^+R62M0BR%%g)#dZKq0P&2H%gIx`!V*
zntxX<&^o4Gv7CDF2+Qjic;dr%q1=>8)yImIx%mm||3B*f^V!nmzV8G-nfaLGulM(s
zckR2Y4Z6{2G(Z3Z2+@gBki*q#Br#$Y`(igXVqfk41sfBwJFzY%w6ilYF(Wrx%1{(&
zM1X{Cv`5>j?(*h(bG&^_X6}o8o^x(Bxe}<Z4TKWWL|3tJ>zv1A{(j%z*G!k`_6OwZ
z^sH~hDFy8q-Ll4H`=Bdib1Mls<}67T7G{}Wm?4g1q|^wJ(C=&N%_HI@VmRva$i?%V
zIdzWiC}wkapVUbD-4S7-z}YkBS>0Hr6u{TM{3Skq?;~dCm)Je(&}_G9He=RT6rL9l
z#}Tvh0lPa5n#}{3=8=tD-Eyu07FTc{FHT)-Scce&tWY<GQ5-QGMF{0FiXy<`OGO}D
zQP+>12diuLB!yBH=_?R&a=zujg<<@6jJd#{p&I?CE>HSQ2XhpPKGo?WjUyjll-b|g
z<LU<=@%Xa~Bz>PRedQc~{+%ndTRjH*w<woZ==FV9xD6#@U=Sl5_kG2{mw>kwy6q52
z_q-nyz!`xMWkeE_h(04_nVz0vW~NFKjp#HE7z`pFIeQx4S0t%netL<~Xn^PYNG&OA
zL#jPilMZb>R8jI+X^){YL{W=!C3ME*wG!)VEA$6P42MI)KvDDu)O<LMat?OZR4VB*
zj1?KDG3WP!FoMi}Va$nDQsE&;hY3kPAdm_qC{@JR5kh2<Y8XNh6ZipIXp*5zun2sg
zp4IN$u$H8gaIklq6HCh=ipTEzf8;)X?kC}MtnYvK4!&0cA@Nj5GLU@o@fM$)+v9`x
zZ&D0v6bmJ~?Gg9y?h{9vVqp@X8^`gOcX3>ToU2LoF=_wguyEs!;iyA<a7d+EU}kz2
zYf8NF$ql3ms8y?s`aK4%20>7y=!XoW5lTtoVMNqFB#H+l$&j#CA+gZv^f}nyVle2E
zBq?bcad2>e@)L^10>x65cB{jupI+hVXTO?lbQWV$jLtusBGu7&(>CcbQ%0q*<L?5g
z5sU^SZrt4As0FLD0jV8j#D5rT0)+cTjt0!k%w(Qo`5}@OkeQ0=xUIm$DC++3Uw=#?
ztT;;m5zy~yZrr%d6OW(A52Q2DMAK=-+`Y5I;^ORhbefMJK6}uQ6?EEQr5!8$tPt*H
z6HdL98mG(~3|eg6y~g3e4&_>n&@T`u2ttq1FeQ#PON%vTW~P~$nZfe|tc6~GfafWK
zKp~a$JQ|D!?C)Ra=;%J@E<DHCM_-^)ok3=N$nhb6TqS1>X%c4~7?&DiO^VhsBj0Av
ztyBbIi064(j;A6`5>S$w9+D34vRE52SClN*EX{6(SHJncq&C0IVf|=qro6Ym&*4#>
zR;NQ0$NzsX_CCGR#`6MP#E#=-)|+)+d+j3z15h@EtSCzpI67>O`J8E*f|bq^1Q61B
z-U*a5;?4`ZOnH~tW4la91&==Y1+HHCfKRS|ND@U9Jwv%vCJaKV<td7l3fREy1;)^6
z^pI$D662vTiKc$I%iMH@pf=?Q`>`b+jTnt0mgi?FJzM2B|L*^qmGzU3%#Fw_lqUF0
zZfi0r<lN`YRk6+`KO4>JT*b{;-<>Na&9ST`i)EC!cio97EZREJhAY-4CgS;icU-}4
zvvJqM$dq{UnOREZJcl~L7?_=%<;9nt<jnaggqLD)T<6+Uk)`Dsc6Rn~ZUfE-%x3!0
zd|Z573G2ScM5(LOG()QdI*nLdtl@cr!Dv9g->270IXpPR_XA%2>Ng0`JoEH3eCN-;
zOSM+v=*S}oLMl@;%+1epu(M6Kb%;$<8qF3f!<cqwfIv}ioFj@YaRSQsiQ+zg^vC~_
ze!rQ~K^}a6#(;vJ@SO#d(W}jxYHKaS;SigQu!wQegpwYf@{kfr1%(_=?!Q#v2gOWT
zIZ;(xI}r@IpFP<Z$K1`&xdwv)*FWrY&)><6mIFe+Lcj0v=1;G2dgCG=e|U?)t5B^~
z>Gxu8d~}_se)Tyzw?cHP5ZdBTCy1$x8e<J&#-d9K+j3S_O6hTtloxjvfw5%<ks<2c
z=Vb9L(=)T^sVVN>y3Hsuv>$tfD2^#~j_|#ZQcz?t8aOA5;egp;%qSY*r$bbw$Ph)l
z*XQtHhi<opHkxEOWN&v5>BSU_1*T_f^t(N(vO_y4pmpLJFX17*{JhmD6^!>N=eFRc
zFUF+fhRzQPC~1+tAR47~Ivy+23WKvxLxArok~Brx5yfIiyWgYJ-beaRQ>)cDJURfJ
zC0SJO^WGaHroOVv^z;dS-YB?VXi=A@DS`-N&G;~AH0!+n`iI=O+Qy~_xJ#!v*l*IP
zcc|4W<Q<C53Xyz+JI=+)H&@5CH^)XDj^{ExU84~n(rO%0DV1rryL9>k<`-(Lo;br)
zSmyryo3z_aip4TQD&izzG#a|h9|K4G2MqT&5lT^S_1N8Sg9@=)P^ttNlTs-ZNz=rQ
z*4kaRw>PtV!UAAO;@D-eN*9lE++HM(V$w8q92CJghRJ4K`I|p|i~q+z`e%eR`Qqa~
zX?#EsI!>I?`FB>CErjEz#Zd&Ar=L<83J8cS&rW5&OK!yQ`45A^P?ILk%gXaYlD_21
zhnrk}>WJa6$D@y*VCz<k!^1kaZf)`OGv}Qu(Cu7*X~cfq%yo>~9aBd-Ym5ZveH77d
z>~r(l`|R)CM@mH!516ja5-2!xW|qeveT4bxGQHk_QpxcRr>Cc>6bke@9s0dKn|HSW
zgOCt-FjK2C7<Re#$q%SEcX{E9-=sEu(#6i0{J?z>B@vm$!SN_KF+RlHXw71>OkRal
z60K!6b{6z{4eIp|fgz~|PeL3w`1I<}_|~ug7c4H!^Vp?J=rkcs5~3)g-D$ITu*de^
z9y`1H?Cu|N)NIo44~UcGp%r`Dj*!|3P?YeI_zdHO{ewE8tf8cjtH@GTR;O8An<q_E
zq|Et2lP1F<I(eiY<7<h?d8H1$kvQSd?7|wq`P=^x@8AB<Y+ikra;RCDnMGQGNetF_
z_=Pe_uf?c!h)EMn44y9t{UXvrvtH*@n{w{_1;S#5VXsTC-yzZor_MdXnG4Tx;gRQD
zV~!qEB=Rnu>~7|Wx6F#W@f7~pYc!^V93P41C{VUXsqC2#3W3XoOkGB7I_4gl$={p3
zCt7P~&y%FC5x_NxLi#)mOA{$Wr!YsMZGs_Tsxr${Po5#BN1EATU@gjn+1aT~p`Io9
z2s1{itr3%?giPh?5X<a+u=!;%SQ9ZEG?79hWI(lCB=i(F4)1bwbVRS$<DEC(LdOH%
z`|v|XsUc1cqu~&#Jp8bT*4kN3wc7mfCvV^>$=-gQVxhv)!V-gMK&&A#hMAcue)5Cw
z^V*MoNSY)rsQs9A$#K+k?5!x}*ho7s?3DCS7WaE1Q?FW!%J!|is4MycPfjA<q>}ia
zH#y6)VlGbyfsCr7-NA3HL?Ri7oqvvHttK@cly;OABaHL^R6fmigEwA(pWD~lq@w_Z
zLEv$?-^KPf@w_5Bj<D8)FhKYgX{_@pM1ahi7<+`#C4?v<rAu;>!n(podDy8{g61}j
z`aZ=%fkY@;-5x<$=0m;0_HdDN{yo-<hxp|Z1H0&&=4gN~Kn99It3`X~9$pyG>h#&&
zZ!t^+2#>&b+{jW`MC+6^iD<XGi13J7=`5YT<Fk6ckIamwb(Tq;znei2;02Fh&5)Ek
zne~aq^BsqCc7B$>_>0#Vv^II^F^ey|C>7y(6N-h=DTQK4yVoO%Q+xqo5R8lW{P{{!
z$?l*|x$2?h;?H#-f1yR4F$rmsL6pKn;?wPS_~3(E_*C&I0urGVgF(dpev4D5HUODH
z-}&D5AQuOb5v(&_f|9Q5&G{-)dMqui&~4mDn}nnK0o&Vq#8E`GP++P$!^-j+E9<9;
z;s~^*RIDL<k50FV)tV$4vA?&&{{DUT4t8iYuG8;o2C*Bpnl!=-ENPnHhk^nU4D9UO
zV-yVtiUn+9$Mbu~FUmrZZHCerjka#aZx94hWqcybARf|awRrNGr?`3T3h!LqWNk{b
zFzc>o92-2(C(#Zn*4BWwM8hEhizfwIsBwX$w82O@hWAsQP6P^{V_6fUQ{}v^GIedU
z)#R<$ukhi|cBxdVBm<2#9{29-6UT-yJnq|P#v=%~DNMTB?5rmzu?is^<gYAPLH%%(
zD<8d0zk5Kpdq5f+2K{N~r)Ict_6*DORfIIurmK{SOZ58#d@JbodI(QB>HDxxwOV3l
zX9sIF<w}9D;9+bDoepR<?s4VAA94QT%PcOP#YjO;V{+!v&NMQVp+lII#2Pu?py4=*
zJjwU9)}m605ulCW;NZIR=*kNnYw?xk_N|YpA04o=b|SkjE(1C6eM-e5iwiScI)9eo
zaL8a3GZ;p6yFD817WHO}?cF`zegDJHb&rkKBsw1RWL)zh0F5;%BBMhrESC7qzxfrG
zmllZQ7(Z~TWqAzY8joC$T^GoUrU^>o8aIdVt!|v=cmK}s@&4;;q`gCymsYdGHGx!v
zO=<1kr(ZvyIz5Xp3eWRtG@2aM8)%adgdTf4J50^a)9y6r^cs|^HJ<(Aw^=*+D5+4G
zWb^>_11pjEZXd#_*xjwSSsgbS+2||^X_9f79n!+_qeP}s%?}PC1hwfJzJR#fAk~bc
zm9(A^_OTY5r06I{Cn+XPF=-~S5spNkB=P6Ezkn-{f*?y#&={M#%wK$@^+0Flt;XZ#
z^*!$2+jA2Ek>}B-#||rpa=FMuLq>jvoRX4VLt`yyxOU|oo_+3%tgM})R1UEiMxznK
z(U5w*PBiGDWQsW0C$XABX^Jo?AwnPJ1teA@f&wGc#-tI#Y7z_Ut0jU$NTF0fC`&Q4
zT-*AT@BG0(CmoG2*@)F~V<%{skTOGF^FnLlV&JAD(vhsR9#epFfrGo{nQ1YWk_Uk&
zFd5}VDGxvJ-8+FqTj}EGj1z>p;xa?$vY&~V6#1WTS!WZbG)a;oka#K}kY!rY0q=cq
z1J9NTgc{Q!;z-cCcaP%YN#bD>Yg2@O>H&1v^dv-JO%G|WW6cOe$?*V%kCJDQVg{Q^
z%GEOzMpsGGm{#kEL4QOX#Yo|yM<w<?Dsd;SQk}hkR(qr*44MZ_msSX;%LqGQsc@2&
z6D@*K8=_kzqc+XfkRbFBLQ^Wc=O-+<_sx@vc59!~$_Cxctwbu<oaKNK8U05niBLYC
zSOZl=A&BiwbebZSN9YC2&sC^|=eczgKD>RztUqL9Ns)L4P$Y>Z8pRY!6@=9cqlnQk
zCJX{nV_mXRCfG={<m&D{`g6Nnl;?l0yZ8$&>a4M(#3(YjQ+QcNnxI4i%6TAxK}pGQ
zFdQol-3ICYcgKmhpCwr5ouPZ#T(GD$ntp#ooVYC3cDqHZ*(6O3v$OLU0|y632;n(3
zQW#R2UZ7N-qB^_cfI6M9x_*vPzel&<=CP;l@a8-3@b)|J)9Ma5tPd%a1Lo!gHcIe3
z$?R+u<9QgJfKFW`if}`T@d)Br4xaD%h|GI0PsY$_rwX<3%rnpN>Z`Bv?z?aCkN$7}
z7wVn+OjSVn60IHYY&3ExSuQBZyJw|T#!H*<Gsdw!Mn5!2-)KXU2{-b>z0>XR`kS9X
zTx8U9f$dUA4)&d2U9DCc-^1g|GLwC-u?oo8WBek>$6KyAHr)K=BXrawjYo|7U9{Fr
zO-(cGw+Ma3RHe$n;XeDLh+1vR{ah(2RVuVwt!&KGqgJVM`qW8=qZn&7herprnr)iR
z4rk7vWOwTtjl%|C{nqbNt<58i$nKHBicA7;eB7axU+;<g{N%Zv6m{VKXrKzDqY?GP
z+bCnf1V|F3H3UF?(Bj6mE3BV5?Z&vqH8dvc<dt!SmbET3SqMcK1kB7#bLxcV(F^B3
zcTwjG*u-TY3WbzDXo0pGBQ(Mas%4K~`}G&Obnz5OiPkZG&Rvlcl>M{MT~3%sPB?ZV
zLoX${x1-PE%1OTT_20nuc2GJ&rzxXofUrmOckZyeb(iIZ1qS^d&0e4M5RMM_3Ckf$
z6$tzyahwuG1A2o27RBuRI_qbjbO>FPPKvMb$jxCS?lb2;hxrIDBTyUT8Z&LO;%@TY
za^-$@34uu-st^L9Bx~#I-2J#olBVcHyLS^8V+5d6Lo|wrM$uSAktB&5$6Adx4z>Nf
zON=K4&RWHo1XPBy;yL;TaPwxJfAz2ag2SUGK@gC{4&}(N%~)_S-mmk9)@BX>*|jk)
zf=e39!OmU&=->Pg{O#ZQdx-EX{lw7fc8PV&C>=12`dAwg_yN}QC<ukI66J?Z87c(T
z=@||hEz*%e;RMKj7*Z^k@q&;zG2Hm{L;hd?&HssZ<A9tO>-Ln%{jyl6$hU4ZJbqs7
zBo<Cd;qr?0#Qr2NTJy#(f5&}gDSCoYHm#8&%uI`Yw{KXPAwl_P$PsJh5YN<tvd3cf
zKHp08fU@vObbtIV2da$3CrO70;bT<-8JAcHMg8E2GfyUTn{&vZj67qo(j!M*Wyzoe
z1?*$Y4z5Uc(;)z3_Yh(g2<Vv*M$Q(c*Xz*j4oH)PBmlc$Svsq!7@y{7he5wiT(8sF
z-=(7sLFj~d_X<@CwHl?;X_ilXf%9kXapKxr+`e<4PB&q1KW275M0timZ$PP3qEHU$
z>H$_35K4}7oZSSF@$^p!flgy0^J6?!B~N8i*1GYl=druD&+N=R>nG1JSACqffB5gI
zmNe5<Kw8qo&}}s+O)X=rW;p0D7)Dep<*e{Gh};rM8}uE+T|Z`U`s&Yh4}Y;SVSqI$
zNFRh89|qQ@7*glIWGrF9@}*ZF=lq2eZXAhI`#4^W6DA00QO6F4yrXe1r<;X)o*EZO
zgZ_y9y#`A8_`VA<E|-1g=NGY7Fg>$ErCOyw=+SK-(d{&uonLjPrPdM+`*d0jj_OBP
z=@S;KoLJvr&}-2d4fy_#ud{P7#1DPSAq+@}m1JRUopP~^NnK}cwL#;g=?<lK54uvy
zf$COeUAfyl5^YK1gf!7i*Je<@pinA7SfsTth=vN^rq00FSO$Zk>mscjk9?fnhRiHE
z+_B<i=s?cNIgXfoz9ld=L87uTn~PB!4EhwLhY~@CQYiYph{OFh>+7py<TCF{-9Hl!
z({bM9aYRRE8*P#dIo#i*TyQ*wuuw!vNg?piX-sD@K&LU?cAe2^fK8o+iB3mg5~{Tt
zl}d$Lt;VoFBux`4wGvV&2E#skdpnd$0co5fjAi@oHQssqhkWfDe;b5{5Gls0$#@3K
z@#bUPy=R;^m+vdVxY>R#3~}9XhhBFJPdTMCY1Y{T3d@b_SNPIbf0?jU!iB0jcXX@M
zY@{v-0)eq2D>ekA$%Mc4c~?Fn@}1DYmo?XHk$^Gyp61J6eS)W+K2H#OZqOEIo?96w
zY~nCKLFq)sOUppI2cN2A#h|Y|j0`AGFQa`QFKN-(y-%y%pskNM+TCY$W1V1n8qY`;
zPo6`^DZRxNc6PS8b>jxN_V4h`U;7ni7iOrp8XyD<OY0asNMh`<=Pc)h$qX4jUW7Vs
zkV~%7*{JonWs5ZyD>;VPedap}L6XFrI<di-^XK{EOE2<oh8>!XeUdaoe#}Hfl%z?<
zztqG@OdL5>&Y2WDG$)M)59KyXCDBQWz>?a&8@J<SNCs?g_xN{z_!^sA`$)i30f{gV
z3L}n>jhS%k+%Fg-(b+?gIr|$v{rDaJ@o>aT-}qaE)kSs>c0sx^nNo&Pua6%VsLn1U
zWq=A@Zfa~s2v4!Nw8ChZuyy|)Iv#;k6pBT{FyQXyHd|YF`SU;c$21Q1o$^Fv_vTpP
z?FwaAz&R!JgwJL(WW{|aSCAT;Tcx|fvbA=MHs?*YC!rWX5>1jQoI=vWA%S@ToO7P@
z3#BkdWV8+Ym)?*6)Rm?Yo(dS*ZiXr<7YQdr5;Me*WKBhYwJ8gWCBFET%XChc@cRYC
zyhfZ5=(-QuBBaG!H<*2}whiOdb|8!UG^P!75JJ%(oW%5EdhJ63KST&cr80#o*BDk6
zgLKIKYoB0k%G}~CCr&*}VJe}u`!T~&lOPN!RxYx3X^L5|0HR6K^I2NE$kSDi`#UXO
zf9F0Utyy1D1X^NDghe4|idf-%PK8oVO+Kb^xxI4?u;?B-9gJ1)o>HhVq);pnB_mq>
z2EGj`O|1~i0?p2dl5Y{7CQU7RFhJCtGT!KvQDiBX-QH-80hPZ3j(aSxNm4uYbKS#V
zXi+CcN~%5Z9J$?>H3WO;9L~0aQbF>KZ#>DXUwsb0;1kCQK|VGlFXG(Sa2Hs(@m7XP
z*<+_d4t)z1QY_8TXpS)HfJ(VcsZ^p|F0(K{j{x3z^EK{n@3FbP$)gv}a{lBAJPD;r
zNE|0@Z{22RbBk_2Vwk2Brm8HgpP)DDa`)ans?{>BPJ&K7yqX|RQby^B%j;(`sUe9*
zS(KS@-A`Ubsbf4Wa^k&kV-%BOwIxk8f#<Wdw1V_ueQlNLnHqbyC9$!Do<driaZ3gy
zNrdOSjt^l`p3A`Xq;R8N1WF3Dh6Iy2SUc(U=UsvnqZRn>0GGngMo(^pBV<k$cSfbH
zR>vJ$*+14kXm^p1rpBF9#%IYD&}8>G)hUBvn_;iV%Gw6C+6jjJ9y%GZySv5d^;Ig>
zd1}>Zws-GvRNrG}dKRTDLTG|8bg2eX5rze<kTly}7v3(N;&#yOGaL>m6sruA9v^-1
zW1fHUt4vL=B2taelkr=QDgh6S;oOMbAurBkG{6&H_8c1$ozm?bktCYMwU;@n-(+oV
zky}?k05L)#*x%cw-EOi{seyITP}*9T*C?EfL;-&0f#wv&iZt^_`@G8!Qldd_*y%)1
zO88ivpp@b9Czkp0SDvR>EIE4u>kg;f&o2uC%fJlBH~dS_k<29$lQFeZs41*+k#mR;
zueyScN~GQq)*6aok%jej_Vx}DyA5XNmPxv<TOOndR{0<;&piJ$)2l0_Nz7n4B*ABK
z^$ck;dZ3`opXxE*YICwWXq$PgWwolv+Niv!F~(%JN|W>wVI8Lu3j#?oEb`P7kMrAq
z>$j=aW>{EupXm?&;QvFr-y=z5H>ohrxOdFi6M0Q!CkYPOf$%Vpx+GPQy{lFm5<SQw
z9i5+7w+DanowvDpV~aG=Ste=r`HrDXjPK!umpCqVKhy9C>xv_XKqh>8^?i<7eF}3Y
zNeymvFMJ3>!tylBr!V4Hrs%bf&>|&@Mihz#s?{=KP-NreDJ;-D+NV;TqE?;5vy$yQ
zpK#^FAG3YqO4fX;apBIfy=Gi=<^_5_D$a`f<4HDF3AC_GN<zksc^*gQMt_20DCoyd
zy`TG_IU;^OwspOl6b`ww#ub+ukkWp}`sZ`rXWgSz^cW0W<KTJD12t5$#AXvs;%sN6
zV16#()n9*!y+_V-URv7E#Wa+}?nulni9JZMY70Axux1+}dI)K<_dIWe?Bw|%q1T8&
zPjghii?jo#r>9Y#q+BV}$D_OuaIs&&w=GO^n-fb_KAv6WljJF)t)EbtI>&{}>uk)`
zDR%qRw_3EsEZ*5iI9u!!X+@>9M}H83C}Ty;C`m9_!mvi6oZ@+5=0%q>RyZ5$ifA(t
zWE!gx)}Tbjk@S2#KcLlWQa?Pv7bVt~H>l3cP(R#ec}`FWJfg%kV#ShcN=%y2PluFD
zh}Dksmq=%G;{<=8jAiF;_vgBYztEy?X12uP-Vg{g*g8Z>-$DBrd_U#oFJI!-uU@8F
zD`6}oX^hR(a$`dBXZ~@Uhs{*r*-J^*Az47e%EozSX6Na3nsoX@B5hb&SzvB%4y6=+
z=yU$U1)9w|8*3YxXdxy^Vg}s~ckbS1&>JEA3V-t5@AAkKkI`-&v9>rxwOpZE@d<oK
zKAoFeq}^%Y`DLDa?ga-Yrw*ZUko&}(^RKjYAUhzeLvMz|A%FC}@6+iHc;WJ6_{wK`
zdKx^R-YBMAkyry-Lli{}1_NrP!WhDL(s^(EVu`G9%#$rV&vSgRhj8Fe&xF(u5)h6v
z<f|$qD&ys3;myWUELko;f02`?R@`AFCL{LT?jm0@vt_Zzu9?V3Ixt43q%sV9ZQ^K#
z`ME_Zg#z#Y>}O~Vr<P{v)cYi<WVX71C!8WJjbqC5^J6c<I2w`~%gNJcKuSLN@O^IG
zxJj$s=ic20QYbE7tWpT8>>lj%-p}6PYv1@=NPILIlIZ%AW1FwzTINCB;LM1P(cm5P
z^D+>pDT6^5-z)GNfA{~$Yd`o`JahTW{0|5J2b%ReNMJM?aCmgU`iXPy;U@V@hzNk^
z8v?9Tct&xCD1H6|zcy83zu!ZMB2p=&43NU{VHf9WeEsXsQ>zsnUSR`-&|?y^`-}jS
zU*~4IGULU%v=B9J;+!F?6O;(gM`VusQc92t>4BAspj4zbJL}MR6>_w9!2HP#iqk77
zBZxXp?tJng3k%E4&d=jZ#pd2Vy=X*bew{*X0j*<%%_9gWoG1a#M=?iSk0ExB0G?S=
zq)y#FL7$xUflv;G$t{fXdliNumNHXQQv^YgQe}pxp8X=jM6<iINweL;R|=!S*!0-m
z<_cd=DH1c`-aRY>>rQ*f(#kB>j$A=$A`B5$7zV?XAAJ7`AAfX>G>u(Kgpj!lM>d{x
z@oNtF%HI3zdW@HS@Yl!3wW-#KbjZQxbwsPjh=AFJ70TrrepsL~GsE1%BAc6=6pIyR
zYZcU@WpQbRf>$7kQU-$_#bSu5%`i1JOEE0bINas6KmV8X+xr9(@C%H6O$AtyHJmwG
zY$v1Q@vvAt__@U~y36?YazxrZn3w^A&~w5kPrBw>3dd!0LJ2e8Kk`I_2jdVYDK`&e
z8QJ8~FA6^RNkHQ;a!8%$6M9n&^d8nEXdMxj5}teI38Kf(aZm{88N;$pXtV_8Jutm}
ztUbiWn^@Z$kFFn7?@kQlZO&U9?VzVFF*CbKzuV@h)uC7nnF|A=odRa$BiGkyk7g*<
zZ}G7dR2F&^4yFhW%ao@U`NHfjX{y<5cNyGzosDNM<KI4_v`}Miy2ipniD)|_ic(fr
z=V^7CsBnsrn8Ej3`2GSy)X?VMga;}n6d8N$org?>XdIV&acO}sti6aA1l+mx34=6Z
z?c_RF5B6yfG=69hNEdd{Z!tYH!>Bu=owhhpSjF=M{l3N-?JK0T8D-KUIzQjL{0lAW
z20eo|F+RntI|-452rCGDc<#BAeEpYSWNNC4QqHqSn@QJnoIiIQ+JA5z3wIa7DJhlU
z?8T?J^wcZ-@X!C4WRwsnO`#A{DwSARSYUBw4J!g3yL5p;r;QSpVzG!$5{A7FT5C?9
zyGXHI9fvXBz2{L171c_axw)I%xo@yG#rH$%M-kK0bIi}JI1;9r@QKL$;T(*gI{sO1
z>yZ_9DvQAqN)mWwzWCy04iB2V{`wm{cIiB`volQ3&(i7bGMDCEg^Q}{bUIXKXEJ0k
z4<}aR4GPG{oHBDxx9(wj2;E;2S&WSVC(~E5fRbJ&1vf55q^9`hx1Zy=FFZzA42ToQ
zx3bn*S4gEY_3LCT{Flb;!j0jzapNh^3s_rS=T85CPP@fab((%}z(J$M%*+A{ORGfP
zF5BDp@cj~%phB7n8=^r<;QMr1J;FkPLI_6(`;@9R=4a=*qXIge0TxT7Q(_&Vbb<O|
zi?@FALtcI5TLi`GcvCkSDcqR++Dnqujb7cDUARDN>o}%4&&cyaj2<!?4M;~l-u(Wb
z(`z;fRY+JUB0cAGhtV_}t?XdW|DHTj$h`yBDzx=c2m;|yHJzIAbM|>wuWd7nmIO~q
zgp>8)(dF!vV4>vWTa6Ho)Ge4;aKK~)r<6i4LF%07!WFr>pkkaBBQi5;FTnQ$REFGX
zqwxg<Vc;4AYbf{yrWY3Rl}CH$h`?6pH=FbuM?`}@su<!GLRv=+jt&nQY0Gm@y^x{&
zPMPoC1wA>Om^_ET7#VYaE*o_lNUggX51=VRk)-)Z?pUCG_UCeBPXOg&nM)TQBlH5^
zd;cAFcJEUv6bQV@-<coiW8^YUj;Z!6597SA)I{cN?~&%5TRQ>*ZtpaB>+P$IVv9wB
zNF8FV?O61aZ_V!B9gBrKwr?L-1&@UV)^UT$`Dtk%+25jCU7_d;rfSm|C77O@<^G+!
zq=O#)f#Kl(E&R}D|LBmp`2|+i)>%KX!qtzO3<e`+r@@+pR&$qrbK5<4o`+N79vgY0
z$HPrERy99pW0HR_b?o_co{IVZ(oX)d7MVRS##(y)0pJn(erD%$Jn}H-+`1w;|9f&y
zbgmp0LO(Pm`*7IfX#bFS<Y7#VrGV#o__BnegtalAXSn>_DQaJSmRrRl>7JxNC5ZeK
zZwGV&*51dM7S{I0``MUBJVxmzie76nbA27rixkz_3k;hd5)JzVGIje!+h_M9!;Vo%
zdK710Ak>SbyMm~$@catpxqxCI`To{Px(zt>;u0*Bm~Px4E%}6%GONqC**j>EBr%>B
z&^U@wMTsmq^I<8Q2+?w9RMv>N2GbkAoB1<ptwmesEMZfFj)uJZ-dn7#E%5YH&$HFt
z!t-aCUR<En-({-eC?>{$P9hfPSGnKa#2*e(N@kIVY2N9&t2AM${9Nl7%c--{DKkw<
zpcDdw6a^sW<jO3+^h?h%HC4%8B88MXbL+OpHU~p)2AOkvICkhC=P&@-2v8Ch%6#)z
zf0sBmeEj<ND0nH;)f%A}QYqF5%O$i{m|__<w?wZ~XB73ZNrE*AzVs*t0W-A{&s~0s
zI8K<EsSrm4j+#eI*Qyi(LFfx~9HB&kZ+z`Hs7=kIZHmvaz;B!LiYA8zxO=E=mQ-N#
z@uNp|cAmM#MV96lNTNPcSRTFf81Z1lYv220Vr}q@%O*5d)9cmIdW|%*N_mhQ<TlgE
zh)9FTl~$ks7ALU+V=y*#g}V%p6u=IV!Z1A}_#40aJTH9VNlK-n^WsVqJmongh8&pz
zXDcEyq-5e*C$h1`q&qfPs|ox<7K4^@RNrTB_kc7>C<Z>uD;rFotx>B@vAes8@&bDO
zK1rIA42LuZ`)HeDg`qq<O{5L+-XS{aQwS7s6jLa;LvA>1VRb;ITA+EfL$_ULu2{p%
ztO>MEu{IfxZ6>4R?7r*VSH@!Phy;z01xl47GU_uL4KcdM^(#NZ+7@rUR%J9AIR1^7
z;z{UrI|$E%Y;$4qxM!+5iNZtUNaDzJJkI{StKdrpMqo^W3W^9U9xy93#r=C6eRPG{
zsc9&<LQ6tcyqNJgW0J~{*}VvnQ7DcH6ON<tLV{9AFN;vJ(Cc(LI5=Q`Z=bZg!((S>
zDa1npuYeGiLa{)(SYXsUWSDk{{SmS(sfAUB!y!BEU5?s^c!gP}rk9+iMW`{FDul_-
z<s2EYW8Rw0u4P`H>5K!FYJ=4p<OGdRw5Cv<!w(B%%7D$N5pMe646~Co#-^HLsm$54
z7YKrocYo$_^ZKWFN|Pi>h6GNc^h9=TAIP442rEhqMp=aPF{$GqSu_Ldb5vYHPtTGb
z)-jRD_`8DKME&@3`|~k`#u^uGW^ysbF=~R&#1lFliw=x2=zg1Y6j7U9CLUQjgFgLs
zlT&9-;TJr%Ht*n9C4QhNhd#xU&(g{gqwbLXJ9iny3I54bcv4V5*hWbKN;w`?zW3TZ
znhht)$&|_CiBCTD$<7xgRfZP3|IU{~+cJxzW8BevAMSU0#Ksa#&EqMDv^s#pc^yiT
zo5j0?ALV-pPa!k0m^CS5x&1>}TBi(?A%$RpuIZ4b5uqQTq{NpM5?yCuy2QojAK^}=
z;4Dk(5&F8K`<$RuFbqc$>?MR40x|GW_KrUSwuQ0R$IXTV6EdXQI?vJ3sDt&VIraFL
zIlBG^fgKQ7i!2QpL{}O04iUnldR@$>q)<6SrF0P)EHUhtxvSbJderL`TAg`<J3cFC
zo~E)f;6&7AeRYXXuWjK;OFSG=2nv|$BGIUgM3OL>UWoBchCHS+Un<+9a`m=0hGw(L
z;qDIGJ6m|3j|qJ)KY5wi!Zd&OZ~rxeIHsiB**xksSw8nHn|HVAwOyD&r7};W*>b`w
zgPTZ*Oo{t*x>5W>i@J-K7J2uL1}06AVcB)?LQ)Vh8?sA{6whjMl?=FIW5Qj?`4<N4
z<f~-O^7w_D{Qg8Wr=((PdXB&Ow||$VnooV_CM%1}2y4+v3f3nGON@p+tcfu?K?;M3
zBXp`!2-492&(sKfpJGrzNx{zE7D=2UtVCIXu#!fz&2wLPk#BtK+xUI~VIr)YD8OJ+
z^~+-$f&>2Tc*B-RtORWg_cwPbmx`P`vCjQ_cd$m_djZ1~MlqC?K^w=T8w}bm-b64~
z$fc=vUUu2xG*P1E=t2Jf`+V!f$r{%_ZedM=RQ|Yd7bMI~mH3r!zrYt>e2Q|pgv`!Q
z9lH)miesK%LXdMO9VgGJnxzoqk(jjt<*BTAG8`T@Xtz3)$|aIS(`Yt{MlsX1X-dTk
z)AI{FJ-fj9ewSgdi?D)Lr^Wuk7X5CIUVp&u?k4uYqfniqH;kz_4vFI-#u^HR3a3u2
z(;qhIb~{KZ@O?wC+hAe&jN6ozKzTUi!Z<In+$cLAp**PCmJY2I7UdOCen5XXbR(Fk
z&M0Ye^}`<#M*~3O`-0H7m^5K-evzZYJ#5yw+iX1J5H5>V&JPMCsi#5tSn>Ibx<{Tk
z$xpxE#>N)1Xt|0Ra>7%nq2Sg}-)DM$p5<qsgp%*LLsp=LAW744@#_*cCJ5tkrD@h_
zAd~_n9eO5x;$F)3-WIoS-sI-Zo7}#4hrRs+mZt=ZfB&0=1xetA_|nHKhe#_?R-$Eq
z@C2$9(C&3OIM}DzZBuV`IP>UZ_(9Qa=*IC`T`T|{LE^sIgbHy?pX(4ql^^(-S}{{*
zkGuJtM=2z#P-Sj@IYXD7Wd`K52<I!A>J*uz42Ppk<P)&Ac8Waj^Tv&j7>yji&W?c~
z;iwBt3cH7uHt@;Eo18j(3K=MDhWL*}z_x*CZkiJ>UZ!z$NHmPF*2%qeCLWMNWTKHw
zRDfneq%!2{*gQM=xEG(W8k<I>NsP{rJBdJN`(hd=M4c|(WJGFW<`x&J-``@`+(AoA
zSgK(!JjUJIw`uG(81#G0Ei6(l6i`WzqutwheqpSx{;SqR*g$qhkrns?i~D~0_+QGb
zSu>L{X}*i*$ez{)Kl3EB+Hn8gJ<dKgL*Th6C`UQ)u$f`4l;DK{e!+3Qq^D3`_TCy?
zQj+=n?=iSWWb@u#9)Em>PD?O~M-UY8l!qq+jDXqXh_HP`lAgf$KBBC#4MS(gXKpEh
z5ExYuRI;;J;BsC!01v^9>{`I_GbRV4j5YKIU81DK+2?<m+2I}3^c3ZaASkSnM0JL}
z28CjT@CtxH8-WThqJ^TXJk*+iJ5`4L5yRmPqN~d+Eml}7cd3;s6nqbpY8uTpL3xgD
zae-*q#*+o)F>d7X`)}<eSwv(X-Nl4S+|DYI%42n9kyLB0etLtA(`N}~k*VqwA}G=9
zjwsGJ^m;TLuvjfoo1UfH?qZB#sy2`3-5n>=xMni9&-wY@<6me|Hyro~TZBY7S685D
zMyyHj5BAxA?<3~s7Z8O2q(N9WzRI%?f$_nJJvJ^p{&I>3U)JNOfF~8R^J~2Nt>0pB
zcncl%v+gVfZ3wCXqt1xd!4AR$9Xs_(AOcL1a<sq8{LB>Ps*koBPk@eMMuP#vULPzB
zBY5)c)BNs#`41^pD+mkH`s2eKCIwv{d_CFxCdge*G8RHm2n&>|MLZ?Jqy$+2bc$hQ
zEW=1rC|WemGcArH;wXBc%W!1*2kkT_<0JRy9tMh2tIUDj#>56$@*I+Y#3Sb7Vu2?g
zIY}6}3_vM7tjf{Y$-O&1Vt-)dYOPa&=G?#WODK^tAS_hb+i&B0eM-d=o-9x-3g+h*
z84O13>>fJT`BIhDjTI`D8KmdYZ0<4ab!j#BxpwtK?rvV^U~ilKqYeknJ}M}XCKk_2
zxpw^qm8zmrDKk^5((euV=*owjI`<65$Z-y&>!vf#WX6l}eLu?}omj}^4U8W+rLgZ8
zD264h2$`B&C2B-2Q9#Ea4Zbf4eM_NOr8-sP@Nl2hI>bk70|tCwAw7xG8ky?JHB#!q
zcjZ5EZQbjE^^h5v*w-m1lx9wOU|_i4;pX>#M4?h<_OVAWVTj1C!T8{pxnG^i9zSQs
z>~>Ez9=9SDN(H1AcJJ+S{mKolT>XF#u6)3~t$TF)T`+>DF02s-A+2T`X(dzTdCG+f
z0!a`S!3rc1QLoLx{vr2wwit{?cwq^lO2jFYiyp#4YPGwUhK%T)`zB_VD*3^kjcHS@
zU89;8Dl)S?L4XWcT-?C(ebV&UftK9|W2{4q;{+)Z27@lOsi{o442w(aJoo&|7@Lr4
zOOhs-tYFL&I`RTCNmAlCA&%3(bkF{}YjvZ96)`pw5?LuZAY?mDu`*`usdF?Ao812H
z4=|mkE7G#2+w(khrt)_Cnrj$sHigIwIhgS1oKV713ry;oq10HEr|^OhzYrob?@Ld)
z6o^5mgH@8Lm02R&p)@m1Wp19z>?}4(h=)T8l@d$K%T&SwcdordGH4;Ck0HevJytW?
zOr&9j%noihxlrSt+l|k26fURCIAkuLg17?DnjB>SaFKz%{(#qh`Ubad-QbfepYYp-
z3QJ2Hc)mg^Z`|MrA@O{lFbtjPZD!`2pA*tkc*2rCB<>_YE!;rqWt4~!=@3$jP$9le
znG%|rK}6%HZ?Ukp&fSd@h|MCpFVXds<TS_`L%QQ(trO%4v4ph^;0o4mKXC5-RUFEE
zq#5;_d~(;aa_Q?7W}#d5@WYef-9qaF$A`Y~C@S4RB;Zj%lQKv{z7X%xTq|?`hKCRv
z#EFaeT$Ce5ae}axII%>gf||O(sN2O@OCn>0{>xnbe6j%9xp-{+?#?B*ziVUYwOfwD
z<OOuQUAj?(l@c2Sbb5ltX=gWNQ;YHiwYeGM$WbPOAk9X-lSxz_O=gar-~XZY3oYul
zHtQf#ly#(9sZ&-(!kY3BNx+@AK4$UM2^ODy(kTORS)ZRh;9c>d#t(?dNckcgwPwGY
z{o1{pQaJ-sBR#5f>r_HX!3=2C_Zf|Pc*3LCK4Sa!bqY#RnV!ePr&6prcPb05W`moz
zZ*bw#BZS2=SW7e-F&d62g#oiOB_4bBW&XiG{NGZp&Eg#CELi7#bDU>mwDX@MKQM$F
zL>clis0%eVSbRS~c%JJTmE&?4qe+c+A>boRH3VY?+BmiBaL`Amu0!^+n6=!GE=N7{
z&ypL4K7{Ej*FN3FT8Y$*(bUjJ%u7SHzDcsPgEv2i2>i)C`m8e2X6hE)p?AE4a{rP=
zCS|`TELb5aSLaC+=`2YSIaWC@E-q0Bi_Fcf5)_N1NzC5C4we3Z#g&sx&8{&sYgt}B
z&HBa#jt;iDd;dD$|KShV-n&MtGsF)h<$|PCEmNul#7TsVMs(UOwzuyS#SxzJv!YIq
zN36yOTq=RdIAnm$%~+j%t+Wt?A*Eu0VG>d-Rgl7`TApJN51E}?;_Amg!ikA2VaaFX
z#A&2*A=9I1NV!~d;{jtyaeiO=GvxU(Kkaj_X0s2{B8<iK>zvee);$;+H=;<bVYpf6
z)}Q@=uvnya{v0+-Pl?mmH98sc=I*1LJ?IR5BBLEh&%;w8gMPw|PjB$S2k-HNpZq!3
zZe1se;|#F{4@GKRJf8%%cX!bXlE5!gE)+?V6m3#QqdxaGZ*%SDCk*3+a=AieQhxB-
z+idQSc;c}~ID6&{o+lh)gwBv`#`}(xG82bLw8)}^NWmC{a0a~yg_R!03g+gQD3(fN
zC1hUQINpvLBdkf9;0GzY`*$%`U{Fj=)$n~6YQMC&j?oEeR@CMGbXps<wxp>cjx}*)
z(E9Vg$+NQ+FcK}mE0j4<Rqm%*{;3gorI3?fx=i=zh`qPo#rFH;-kLdLICkc@dQ#(X
zTts5QxHvbf4ceqwogf3j-0U2S%gbzRth2nd%+kUFhdqx!{lUk?qlh>iVT~jn#tcUh
zrP?BepoW(gu>(st>f#B-`ua&`r|0oWxO4jo2$fCG^cXpEs9ZK+x5w1D$HueHekgT?
zp7O?HvRt%ajBy6ULb?4<I@HZr!(cez>L*wE<dcsX4tuofhfK}P@aw<!x2YA2c*;k4
z66L8e&niDF@<ap35u^+)dPrk-fX%yY<j`biPlAb4r1B|S%esQ1tZ;N=mlNOpAv3@A
zZQA7}jFuQr(r$QorxMa#kxeHeFgsX2aLs_VkPD2+`Jgr%!!o|+CUwT;N$-GrAAd?(
zd4lj_kIGtwcyt-%FEYJ4%j)Bj?fV|O6Jth({?Rqs+baY!DZjCJl|Ql<xp%`ur}xq6
zGJcXm5U{#9MQbQv<^s)bA8T|r?QwLjytp*hWHcA`0Kg)ILv8cIE~iE*<r61~t2?x2
z(CN@0jVO4EO0CNF{w~AFfh9?jGHljaSv$%78`p_rO|>jtjyZDtcjZ)opLhLY&4+O#
zHET4scfbkVVATVSLI_J#?{VudeoS>~p7N>Fjzg1i%hEI*$4NP{!7)@R>u`>%=f)NY
zGE>bdh0P9CAq&{@8ol-wgHcMe)uh*M(bzknP%2WLUtrMd(`a>2^#cwM4%yk;b^c$i
zCOez=nOj_8(C^aiwTN`g#fz7C;^`N8>6d?#%FMhw)HAO*iy3Ex<~A%+38cuo*lbjs
z!TaFQo>*FJmI>(CWA5onGhRxXx=hAFWN=iAOg)wg;;84iJ;oq~V<BeMlj8&kFv4cV
zra&3+)aO4848rBU8cmuSlulS7V%77I9n1Z<K49tONrF-Vq(;~Yx!PTaoQIMr3ON=T
zmThWdvP*W|oEQckdGZC`d*{dOY~G~P8<J?lxpOC&pPwU56TbIne?+}Iq|vPN^kWw}
zcj^>a&HT~|QV7y0LhBUe7dUb9alZX)bIdNj!H-{ii=#$|?d=}LvY=A+D3ufz!Dx_h
z_s&g*!y)C04_cGN5m=e=WE4iHXp`r2PR6fpr^rT*%Dn?9g9sFc5u5ivW|Z{kcaN~f
z8G|Y>WodSq%g?=lP(FSTG8#qHYEB)V+ghYq)1ZaLNH9_$l#lX*&wY<AxL6@QWG&s{
z5wD35k|+gFN@5J!2s*d!bMyN@;ZmhcX>k>!Btko5UWG&`522hLkJS=sEO|nO=lKXv
zG3uv$^uZP0eEUuQ-Jkp+jrQUAb4V!&e1$J1LBSa!udSb^ST3@?xlN<qpj;>uB{8jb
zovqFLZ0+w7l#7(6$^=qzP;c=s{^g&txcnBcy!2(BeEcb%z5F=K%X3I+Fh-0;L`u1F
zKSE-qLgJ0NQpXjoQu=5sDHY00%`7;2K$7wEq{#R)S$vQ&C{H<hgfZ-GZBYnkDVA$v
zB(Brx(rR^yf|!|V)s6LY%M_8{bB)eNvkzejEJkQN6>u1y;i{QqXe@J{v)Ay2Ku#4n
z`;Au_^?KAl_z+SzRt=P&T@!au+swWQ>E$dLoPU{d<5-iTb&Mwjvn#Wlxpamzr_XWz
z%mq%LI?epdJmqqUD|Zff{hcia1H&jC5v37ve+bf`1t^mswWKv_QVNUA%uX{?n_*#g
zj^^$SMuRSssJ>j#A+shi=gj3Cy9d4CthFdfX5Hct$Kz-tR^v(I_M*?A;Ypg}`wm?*
zR@3h`dGDRqD0m*LYa2|}YG{*Z<~nnIr=Z5sFpO*FJ<pqf8V~D615wh-_Pr3SB26O-
zqaJ5v#I*291(?**`rtb2wG!s5-=JApz@!2_QUsM4Z$^{UHO6dV?WZ8({KAnp>N!F?
z9=Y3$YNQP!FYW<5x<ydvQ%cUGo~j_Hs)+C$7yTV>Uf)1#$?9^Ka-i`FOFYMYs;wqr
z=yUPh7Q^r)!>vcDl=mqPhR9-(FTVUTd$!7-y>Sz(Q|C>X6>!Iu#MXkwpcCg*;h>Hj
z49P$m;S#Zg6)2p5#fr?XDoGH7J~H$umx_#1=nf23&ovV5W&?ToMdao-L)oWPEF)Ba
zvCiYumArWj+<yj0ov;FB3?yCF%mEj?4idp(YEeRxSVQm5K6k(W8W(?a3OPSNF5-0N
zm_I=qCQj_uiepIKgO2t27)N>m%Jav>VjBdg;3T%Ohg2Gcq)@9ls}Mh=JXc|U{WR@X
z3xvfFCBtE#&<hw1Biii-onD)6zeQLoar)e4&YZtYd3F(PVmH#wDsC5F<_>jZOr~rV
z<02F$wjKF~ueD1i!1G+hjF1_zT8w{=J5XSlSP~-;))9@h)-K0Te<toG8-W3Cqt)4l
zgneizj=RUwAt${pPNjz|`k?W!#<G9)F54e`$f>zkKzU;l{c$UX$%uWTpvq}h`H1$j
zVi+L=GqWqa{MEn7KmRBHEpc>6;AskBNUiFOh0kBQ$nqfOFaF|tNa3-wy+g5BKrhS_
zCkcmpo7DFYiM6FRGe@a1#fgm*JoR{k!*-jWynd5WS5T@dMxzAfNoHm%L~)lW9#EDQ
zY+|w~IK!9|n87{nzNF~XiAX%(pBy4UWXXc#afO4hNIarZ8%yF)DWzChIl=nLQ}}*>
zRGt%77>!aHnL0xnQsX=wWu7K4joY|C=aN?9nTVNWhtoliP$-fdVHN^oEU}P;iDdud
z8x*FeIrH^bVQ$(LLN+_#rI&4-S&{1$9SG%NJ(q=h{pu~=e)~=S<sbe&gMm}lm5N1t
z&-r>`HA$S3#3{xKils7s=p&S3^Zq8+KD|k+-J;!YgE1_wEN4#iF^BbppS*p8c5guA
z`X_8{Z*uF-b#`~Rc;WfWJoWTN0^iRpceJz3k&dUEf6wFZm)mn(ptm-X>A6|bG-mVe
z7HX6-yRwKX7Be3?H<oh;wRAB?L5Q)I{-8&_(cta3cW5-~E;CzLPOmRIlVOt=bQ$X_
z=kR5R<-@QwF~XpwPuE}MmY$(uJOVUEC{k+(Q1}=MOEa8%^<_r&2ECn4H}3Zo!jp(p
zyKLl)TPU(;#39#NQ<|h`9Z{(kID6p?7cV`^7r*cdkDR~6Ol{iP-3WmurQnBjdM&C`
zC9Ig|sJ%yjIK*g891THO3?Z^mW_o#vjkR^ADpeGg+x2adcz_iiNQ<&r?sj%?TEy6b
zMc^`9F&dpW-MGehERax&)J#TnzEbF@i;gFWJ}?<EicA!P5DelW*Kd46lBRs&g)g$a
zxJa)*KxfunWc!`X(A`vP0(IQpM~$5g9>z*_A8S)QJfunS(;lbIE@%CSz7ce_dq3kS
z#lG<&CuXMUpZqfMK?UnU0f|_G=x_&Pu4C<p+!p4uBlqJR%G{gKT8+t4VRCh}-oJ-Z
zn%PG-Fnx)K<z8=vPQL_$n9aiuQM$v@#WLYcjoH;QzKuBw0v0c{Y0MhVdJTef$YC0C
zw_*6;{xzaeAF0A@)GfxvX5L&Y>6(^FQ!?34Vd5<)q$SPJUN=p0KPUHzEERl$ARx7Z
z&d5@(0M;<<ciAY{*ey&E3rWE%;;E20NgkwSeMTSfA6may^I;Ue9k6P4IOWF_g(OBI
zt;Q3A1cOd32k(8%)anWwUwH)z4xQ84CYFyta-_~?*ywSB_4v@24*9Y|A-ur3k!Kf5
z8<WMi2!cY1ndxb!fluE3fI(-#$uk#7iw4gs%3h70?$OxUV*luXuYCPg%9T>)MDH=b
zxWO|oe+!{JqQn(TGIL!Q*+>jyALt0i9alDw8^>ggqf|hcW1EBfX5@XY$;N9I!B9h-
z8a!my{WzsuPDsxGt~AS?&6QnwbePUI+lNuqNhHcd6yqaKi54flAqK%w>RyOs7_<4>
zyUd?C#ngohZY++(U>;D<NvX)>h~w)uxp>F^RpeqAmzweP^RLkDwE4kz{y9STsa2|k
zLCD<PJo8Jd#9DCX<Vl9zHfcN}2m_KhX3%YO^VUsTjW&t#`RLF7f{nAMICbg-E9+}S
zCZ$v^vbEQxS~X0S1*2g~xm09&dX_>MI2#77$5C)(o0-Tsc1D{l5kRA?!OGmaChNwH
z<zVN4AO84#>IW&OmqUz95T<|-@wl)pghhn!LFzjHQJfH`hN36OPWKKG&ivklMM;b2
z*$LtLb1Y>gv)+*Nx=9TWtRfID7+YkF=)`D*^zr+ao!390I8|cxi!We{jxQ#pmm$!g
zgmfGffkWq&hg2Sin@3!`euaPcr+-L)&_xJKxl-cg7r(&uPd{aU|BysGu4uC{;`;3^
zPMkbTty*Jx+M<FIGt27?M+t-Bh*qOcG)man+U2beKIPqyZgbdb=K~?SgC0M6{U;pN
z>kLMH`hz}Sdih1CsK%*bonljDq3tdP%nAX@Ls*3+a7d)D2>c>xs`0Eq_#R3sc6YWo
zY8>(S<Bt;-0!pPK2Zu4cyW4#4FaDHftH<K}8ZW%?3d@VjJoWg~NToQ~+rcIsw~u5f
zjVs`cYZUZE(f+WCx{fJx$1GCU20=`MLKx?xC6pi#mZ6H7KEKYHUwWDA-}x>aHqf{@
z#Kh=K$*P@dGJEFTI4(t}307;0Ri7uGew-JceV$+b+OM!QxBS5SZfs_4WqouC&GsQ{
ztLrGwvb7)49Srb%FDrO_78X`GfA%6XQ&S|-2pf;sSed1LYLz>8chQL@2SXehHUR`I
z4+>peqpaPzbIf*%V`q`D1``c9+}}p)#O+<#pEEYk@YP7+C_u{d81#GGx_twS<@x7c
zWaH#%`a_L6E+UdS1w|u7Fy6->XlNh8iqambcBpJ9BG%0wD}LfKrKLioF-j_i){qQR
zir@PYm%d%(%88dSen8w-oSu#OK<;7EUgoGUu`haH<j%1Xw{ggquuk!uCJ8$Gb4o~t
zM;*f55$V<xN3HuvPji0lJ`1aY>uQRwou!u!xo>?+sn7K8F0=JI_oiNDAiz$0_+Cge
ziuuV0pCXJV^c5&)1LZzT=1DkC?wEW=t#t;LO$2V9%76weMp}e&Txc>eP?E$n2OSpY
zW~f#x2&HHZG_ysELeg#?@v<GTyL^t`utUxF2m%;JW(;PyI)I-Wjs1ldb+ypv#OW%P
zdqV=Pu-0>jfe>yip)Ijc1Ov;h*WRMKzQXMJN07e4rXv?a;wcdR82u7P=Rz=Yg0&Gc
zi~SIpYEHS1&t?Z}l8_`Z!a7Bu5h+GmN`)ydJ^3^TJKM~jSfezvf*vJ!qb{j0*gM?f
z=~GX!c=|M<uh4hzViC+QtYc;9U|s#$%t%}pAkDF^z@c{zUCfHL?0L#Zkt*Zbc^P*s
z_dv=~ADeM;01_=2rchG2>^KvVB!)!0XecRD6vmC=az$C*l}MS5news3=UV}iY8bJc
z9x(4gfFed@BxgZ_WwhJm?)QJfV@r#O+1YFg>6DS<jZ(P6eY_cDMY_ADS(3;xo@I`F
z`d+}Rzx2xlrpMiD@3K6<h$lP(8B!=!&_+`BL-h14(Xd0TBNyA!>k`Ka8>h~K3i<Gp
zPm#jsji0{3(p;HJxk{xR5_)|EK3=F8MiHHUohMf>5qbsEIKrk`_jc_5i(_HzN~!FS
zGLVHr3+Z98#5(4OKl~B@?DzkOsz1QLTmf@L5J;5Dy^`{c(ji;MT7oci*F-Ak+n3)-
z>3qzjwJwE52z<}~EAQK1XRWVR`QoE5(fm=1;a(GE4JdG~a}tF>TSFoQ9tzu!xckG`
z3CcyPmoJkj4=<OGA~QjO8$Eg1cz85OxOM#=Z@v9G&2}B%^N`B&l~-S;*K1R6HHi}U
z{gssT`U(H_|NWow+H3DKJv+@Xim10*wAwxTgCYI?h$v2p6GOAzqdScFOO63Tkm!^v
zpMHerfl@w3r@Z==FQKFyckX7)hRlU93M;ds1+Xb3amwM*5rcl8Zm)x}mf2v1J9lpr
zjYg!g#>+T`gTa8&Xu!h!GGBQ9Jk`oHmoLA_!Ql}fy#E%r@7&;Ef17Xp(l5bO1&7W#
zv+*R>Xl>nolk-s@!f5WGKH|sMKA`;43f@!!BP3Qk+L#j*fG0e#5otlNeEBhY2W_^0
z@&*i?&!^RnYiO;>vNaRu5P*SHBT|FsSss1jJYRhAi~QO*e~n7n`8}qoqYuPMN}3u_
ziZlTs6^)|<l2Odb(<hl<n5WfkBBi8So@Qoljtghbp$v=$J%&lbT(QJAf9p4R?BW{l
zzV|cU`Pt8CHXUjwP-D(ij*>j6#!cPz^n48N+;PHa#|4(SKeK<dMKl_s)Nu+;u8Ou9
zGVfvVDHaQih6A*TxOeX+LV)r;PMo+%Aq?<x+aAeyG@J`##`L3yxU`9}I;I>}xNvTc
z>HD_{5-`T2C|#O_l>)6TLll13Ab<E8XMeNCjYpnzbc>os%@s*^4P<ss?!Cw;Pnj4B
zEYh0Hx+eD-P2D|^0wok);8FC8SfLmUn*@~^RHH(_{Wdq4qd9en*49mW_pc#CA0tX!
zKXsPsV#wZOPq6#CqSxI-tQC+#(Ca5in}C!UsZg20w6PYg0b$04y*po(Yo3m!h~)2-
z)@Hm1O9WbElLeguZp7Y;dI%*c6bdL$F-#4;L{RlC!+sw<YOr`|m9!orR79aLPovXC
z8nQ`TJ_Y#CK>hmlUw@RxPCw7pf7<8Zqw8)ID|2XDWF5XiN{?i>&F#N<gK%~hwX%pZ
zg49?%p&S=SN`#fzG@Hw31ZycAe@;pd<@p!{{qBgy&3%pz_h~c^X|`H)dtH(=VPjtL
zwU;kZDvb#J64IB<OieR5Xj4?vnBIV>S*Lf<WM+Dvxy5-3g#z`%0|wEM*m^9jogqzP
zFnY|Hf{ZdHGMf&Yt6cR1q|vcFA>)aSl|uRd%_e90Hk}8+PyU=m8>hMriy6#q^?1~m
zb3?`j=CPmd5JHhXNDn7^<_ljs!G-xX?*6MClGX^2_*vmDge0+=#0VN!u5<9S4_SNV
zWeB`68hd<nE;2)6_ery2PUK>ktY8uxE9w-!^6+a5y!h3>!KHI284WrP7EfY~Q3PQb
zZ6J_7(XfleVxt&i5`3kYs!dUyn&R*L&flWb?&9eYo-!1|keQi~&25b_DRXo4v^x!y
zU*!4cUnCxl&?aHx`tD=3^VySPGM@9ZVt&FYF&N?aRdJkh;^b-m&fonz{OJ4N<=rdy
zS({3koAr<$q^U!fEsj%@Bfqs;Z9?MDezyDf%mqCwnuV3fhk(?5>$jff+?gr1%bNH9
z?H?nX9V{8a8xTT~kddH;1}o7=J#PN^4IY`9g^iP#`9*vcU_FJE?tUueBcx)~A9A>V
z#O~e>pWgn2u;4<^zxeWH*47sJ@sDpX8o5!86j1aPsnzT>JN(&C-pxK`#&<IA9f=do
zP>V4FgP7#=f<vt<l?sEwfHX<@^u{&j<`(dkN2OZh`DY)`2sZ*NOxCMI{@yqrH@6|D
zI^u&5-Xn<<eBTG*v9i3uQN6+8;VyHt^9%=r%z{QZpFJH@2#UnR0l{>Q!-Lzr@y6>M
z930^J9<`|wRziwX)~2Z*+Z7$>^LZIv<MXcX{QkQfHY8`_5|_UIDrvERu!ht+dtRN5
zD1GZx_!K-&z48Kaw@2rlchG)Bnp053_(ot-gGp1jPog<-YJ>CV9^qHM@i(ZJr?6Hq
z97eqJ-utYst+BYYNU;>6l%QPk2|Yy`Cp7CvNMLn+jVI4P&g{%A)6>)VVaVbB0fSx-
zfrs)^PHt@Q{Ii$2{N$sYJ$r`LwPoIV`+croze8@kZcMH`%n%ym&?A`{pj#^u*3T$4
zU`dfWMM|h2?Kz91jLUs|S?t*S-S>R7_9&HVbXskMa3Yee%{xc|zA6v~9=`7#<AlpG
zI+qunN-0O~dq~Te1Suj;JT-+~I?3Fcp!3srP?2`Y7umBeEkq{Ef!*s+`_mt>K0VE$
zx4~}W5iHLW?|$OWL9MYuBa{=IxNKbqsyINxnU@QrsZEzzSvkRcc?o3{!+3~}G@~S@
z(5VwHE-^e@B2KELv8H)%io)3qhRtmV%gE;*Lrr<KZGjOrqTUX*<tnpbpTPk3y8Vpm
zl4YQaO!6=ucWSJdC|&b&W^(R+<`6!%gmTTaP2Hc-X-fZ~i?|>ugdv`%7)A-bp`}=c
zI2tk<ba`a0!j|%pooxzXF;mOu=F&N>;^(`6ztE!Y<yW2p(dWW9zeele2(^EN6%tR#
zahI>HMhhRMeHx$KV*AZEIrX(yF=5qtZYeK=JjYtmObH|b$5Rr%M2V12d%*oW_qlub
z4tMX|=Em(C?C$T<>2`=>r=EK9+!D_`zDT*^QwWMKvZd(e;ls`W(V$O~48b!5)e?TO
zz@XD(e|L}WXvEz5SxVE(XdPv{SGJB3`SX#~@im5^bDl}|!fG=iUK?XTJqTNa;}7QH
z>_RwCs8MXu1vxGhwAN$U`vl>d6hOvmtW6QZ;3-Lp9rOD>&-&JHyoj&*h`qa9`?DXS
z;?z;&U_53_kfa04tslS1)R~i1)=#38BGJhMBZlKh-2+8khUATlIe`<%C|B?s=~0<k
zq3mg7)MC`{p-qfPQ@pB&?8fvvb%c$b>M|Z-wZ%gsl8B-wC@w8BGc!Y>=rbI4Icgj-
zH&Y=99C21DkNVMwQn|#bQ|HG5g-H~go9##|Gh2vUDF=S$y{NN$oS1|pO-PcIrKJ^q
z?;rfHc=GWl_<#P_{}uJleQFhpG?Fy6cqkGL!bmh0Qq7>>qgttr`J(yK#>J_P(O;!p
zT-@hek3W6^ZTc)c{RF3*2kigZ4@pNF4>A>(MU#>C1t@C?tYoyg&;6gi&dG08F*-#G
zhX6_GA(YJ0HY9g%-DTJv(P$ho9Cn$Xt#Ri334Z5y{tN!;KmBJM9yUo+M_DNN3MC}5
zPD!#@C6${zXT*1h_-M2}hVF7KbR9R!jH3|?Gt;!&9pWhAqbnb>zH*Yc-+Gsg)g=}e
zr?Yo95BzqbBUfBcQUQXnz_ZVNfoiFYl7c8s2>b#cef%E7VUI?`(rVT5d?zSqcRF0Z
zevNjk!&I%x|NejXzd>2cbhU;Mf>9J<EzZ6q^FGYStoGRZJ}y`v!us(2Ys}89viXxY
znVOkr>ZRup((N?{r&{&L4XHDRm-7|Qy!-;692{V`H<9&2Qjxg**|@|Eqf>MmBZQz-
zEpX!W8ZUj}6>61f(o|!u<n=e+#u&-W)C`4CjVUv;b0vQDTVJH#ud}(i&1lr+Xm^ug
z>ySd|Q!Eylo|#7~n4X(qeR+`=o;k}C7tc}fB+{?&ORs#L)unZ2XXlunewVl3`iMA*
zobujEw04Duu-XZDvZ5?+m~yoyIz>psa4?|NICRgH%B(3Y<7PScs04xU`^1UQ`sx{!
zv|tlj%{EDru(!9(z5BO7SUk@k7jwC#N&Xo;&m#!DpBu&bd6vZxPPJ+H9Xhux(Jx)*
zL^Q&_^*){ugi<lGhERGWCMEU)3f3pNcfi_r{)FV$e}(4MI&xHYAVofQCqL&h{woHp
zV@zr>IwmYC9=Y^5C#E-O95%Rp?G|@8@6l*B7!HSMYnTcHe);c)G?y+CCr>d-Zjkm8
z`kz!$-lG)ePU6oO(bXQUVM3xq{PG#*&qvJL7L8#_z1782e#V8(DO=fhAD_e7xZA;U
zKn9*118HQY;8)q?&o%5G$`8<y14xWDbh;g^wFE&xxm={*AJQ9Hs-Zw8F|Fni>n7oR
zMRC|uR4WB?TNi0_p5Mf$^*^?LvE$e36#YR;`P2!{eDwvk{@_nZBc~V=*`b}{rqe<I
ziDm1jKcl*|K;@asj;kX5i6{aEo*>oE7*a~-zAc2JanR?=)sOk)+Esq=qrc$p<}IQm
z9v9mJP<RZY5rwcsr`;h?A=Tm(VX5RyNVOr#h^paKNFYO^IHGa?4vqE^kx@MN<O@#Q
zWelElqodrqK!^;vfXM?+Hj&R}#zNQ$In<cU`oQso@=ux{PB{`lRvfs983$5E8WQ6U
za;@_K@Fe>Q&Qe7Q$4|*Mxh}3!AhKdK&*1z#Yhiw#{y?(vr7y61^A_oiI~d`-y%Z8@
zA+Z{1eMZ|2HeY*}^M7L=TdP27oD#}Yj&p_M-&kWax&4C%k(2ZRAj<`{nfL<b6wYa?
z8I9_6du@b)PV0b!t@}u$nVMe$<AIg<GQg&qR^y0;*%|y$B7=~n`2}|NVp44=gatwm
z3Z6j6F&LlcF2Bm`>=G96tmlxrW8BC5c^G%z*#__2;{~x!(Wxa(H05HEi{~%V?RW4C
zA%q2*hl16i#4ET^daYANBj<qb_<E4W2})$@WuYFpM&lwGx8ZmX0jVn$OZ10BM6tl?
z3(t`5?sM?&r+CsS8kKMx)VR+T7AX{gPB^@Jlj6A(tXy~;T`Z6BE~h>ejD|g~eew}&
zi>DFBvbMa))NGmG`@O$Ur!`_{>wwW9qFi*g60XH{&5_oIIMrE%m>hHEtQDkY^8UHa
zcT&)~WwBc0`-+W?Rd)CG==C~$dhJ7&=D*JCKYfdD|LQN{gb?mOBAX6aYcl`D+&;+_
zw!=~1`L5}NPUDE1*RC*(dK~N@5*7+nt5ejQ4MuUw^_%xxY@BxcP5#leo`@f!C?ZXA
zg1q|<xtG-gqr->5sT&OYOigP{yUpDnzQwt<b*iV9NVDngm>MDkSkOWcqbZzNX8pw%
z+5Bz~seCjM%B1c(n2DgHHdSM3ZHbk|RnDKfz;H0c7*|YQzjl`wU-|;&a_H_Qf>Ejo
zl;C%M^Q(O6rOVXoEljE@6^mGt`8Nh0qbNd3n46noWqyWo!FN-jZ2400=!HvEs#QE0
z5cobn_~B0+-x!r0rfy1*bDfW8I2+@5RuVU!Yu4*{%42D5iP(c)y@?)0SxSV5@A*h6
zNYVu5xnf{(X^CF%fWe?MZnA>FBTa`?D$_K(y>S8OJnEbR+xG=QC<y}Ta-<*Tsx;WV
z-=h#z>7p3?@>iIzA7buo5lBUpy4bW~ni63NvO{ff{XVC@|3l=rzlAAH(JO_RVaHLO
zu-G(l)Ek)r38{wE;sus3KKUY3r3HTc+K+hqy*KIgJ7Z2AxapmZeRl81OrMBJZISxk
zlMF^z5D92u@%H+J9~Cegibgbzs99FOG~!aRj^FBXZF|6A7$K!{4T(t@djd{yxXF}C
zfDmqg?q~+CVFSnWJ1*AD(bHU@CNe}iiblj~N)Uup%VmzJlUl(rfgmxo+AU1>JP0W$
z6+>Li9+^&DOlpLmWBp>)ugY0@2;XDzsYmHwzs2GE*YPro4JoCA^1<=WtiU$=-1^B|
zTv}d6EK9`9g0ttaWchf)dAMa(CMe0UpK|r%Px$FuuklCU`(yh3F3#0F`_V$6Bq%I~
zgfR3G*08(3hgvQ$<Cl=qBPfJe14@Q?K}e(B;PCK}-mp)-RcCtnES1Wf>-aN&y!_!X
z*3I0V4q}ofVU2ZuT)2lxiu~X?&gXD`IeAy%x-yZWz>@>Z6?KBjbPZ)w;$(m^g4DP}
zN1JT?D)J7}S(xbTXUgpGGn!**6Q6H6A3cGZo#E80&vSQYi+GT_7Xl|BG1ia@ftM(D
z-u{T`v+K+~|1=UsYH%a=Jb%+@hp6!&Z;~-5F?#_rg|qMblb6-^NxhS#^*y@1E~9Rj
z#{MB;vB28$3dvx=FiKcjStmCPZX6x5U#qZj@-#|VJOODO6Aed1{Sj6Rw2{=CZJv1Q
z8Gh?;{VqXRCeb52^7{qWifmMq4XdmeZ{XvC)Jlii2}_bBe0=RDVNl@o={45ZR+yTq
zad1ZxrxIng^Dm1v!{Nw13t4e5r5JbfdH-%Ru9?Y%7{USD4`qpr0)PNnsj>FeFEeZo
z=w83=o-1sYYir5xhVW86W3j!M-S@BX$Sc20IJZEMkE{f2Z{4RDX!^Ya(zz8X<qGG|
zo#N6HXIWZU;NI<bN#cl^=_-T4V61xeRhEsJq@)@*X34KZ{#?q;11fKpto=+;=ak2#
zQW4KrT)cRW?X7Kg_qS+u8l*`|qtRt{cGigntn>A<GUrg)V+wpC)2-0}0RR9=L_t(%
z7i5jb>I9o6+`D&^X8VwOtHJd2D*x?&_m9}ytMlrsU(OWr0(7Q!Oj6P`AxRQ^-*Z+N
z#&Y}iT^8mRDU?eXY2DabO!k%>t$P?k32X9E8N`Q&-23kLx$w8YMLIWw5E6~R7ZYzw
zX*1QcQY=4tiNTFK?B2YJ962GA$$p+T8Y|(_<Bwux%GAs(K~SLC?sN0n9sZ~P=^yhu
zzw_HHFVExq?&l}k5Tz-e@CZYJTAF5UVcH#jle3t>_b4cvMQ>$?c7_%M<NZ}g$;Rpl
zzWU17NRk+B;731t4Wpg7M~duNJr+eMtbsIkXsmECe5p=};~}2s5mrhR0*_I*Ppwp?
zQYvw9aOikw!y&cmG?j80DFlULiK(e+78d5IJX&FXet}A*%GC5MS3bSV-eDb~lvBdu
z_K^Jjo`~xn?z(>CCi82vNE>5zyX;*)%b9O}neM;%4#8m`B?O+3SRqLAs7?W1BpChd
z3TNkMiC+F9gVhDL8wU;*!->S~G5a2?ojBsr(@#+K=lHjO^ap%+^&?UpbNu{~()qSp
zVHoxIXxw{&i;wN&SEty#R$>(2B#H*4$u8#Z3naT6lx7n)zOcs^;s&$bKEvLCMkjJ6
z)?%_ZXAlcgI2u7tf;v9=ks{*(+WhGPF)r$IUsv}ZFUTDQoPAjojfkQMKk%v5YDg)W
zU7BauI-;Cf+N~x@8Y8^|LO`h)fX&5%`SJFji=|1<hms&mRaU?93f<i%-p+v|?_(#M
ztpI5)zEuqFY;o|;&scf2g3&3SDL5`pR!BJ$K=;*53b(J_=F@8*^T*%)4!uDSFnFG1
zWqFxsI3!LIgm6q_KXCkr)wPrC?ryQQyGh`M%+%(bDWFn#$|H^wc6ayKJ=kFoMIaP2
zbL)6P;TW0(WOMNhn!}3R-1osCaB!Q9&0?`^rew>bhO&`=rnZ}SCgsmq?sq4IBrJxM
z!YZS~QI@5eW_BiKJoa@+h#hwq`R8yPERGTIAHqG(4!4=}XXrlp2)nO;gqM-xg^<L!
zYh{Fh-jL0oyg_Yc5wWu2M(omifH!Gy2PB!xzCdC##bi#dbx55=dTyNW;B1Q*1X$%$
zC<YXihc<$#sT!(KBvmm|DX!nWN&Vn}{rx?Z65P9cn^L*V)SUAei-#jd{Q+qbBPGO%
z;?iR;@!S84-(z-eAuF(aw3wJ!=KP%8NIAdmpS@0AR-lfw;ozvw&dvcxN4redLS|-W
zu*zeQB$P5yjW#ZBCP`zwpfLWsoGGqEe8w7PY+++2pXp)TV*wZmG8K4>i=6t}m#DXz
zj1HOz;o2pW-7SZyJCiS^A=)~kd$>>S<hk*_C50d?1Z>{<gl7Ga_Dlmm@LAtD!R5=B
znX1ik|Nb_YFF(!J)_p#{@-b25o(UO9N2Uj~b%?NgpzY+j9)Inu=>F`N%3iF)qeD)g
zK8q4?=ENxujt*%w_wi+sJ2!7J^V}3#d6^&~yRKM-pR4W3cm_CLkxq%C5x4K&<fwVb
z-u@xc$miO%n?zB{pZ@8;CruL`d-PGRe)1_UVM4QictESuLP$lw*C$O<nu8|&ejmS3
z81D=DO4F1$ju{Sz)EkY@eUI~NBn%jtFuZk(oi~2Q%r{;|hH89gJ)AEgfYpjr3Pder
z<>@E5{`&Wc8g)`LA)EVQfFG9GIC+81ooi@oxWBzer!nAv{2%`brLxD``XZ%Dfh08~
zNoGnd!S@B8k|Z{8(;2ekHhca9-=!@;W;22LOQM2auSZxY5QZVAPo3e($DSsNBaRLa
z`S|LMoZy`ma=F2|&CI18{|*bn^C*=H%+AhGC=?;p=s022>G9|@mzbNKWp8hvD2_;y
zgfJ{HH8saGPd<zCEQg1edGoCwF&y;x`07Ut27QX93dQOo!c(YYh&wPw?((zbVWj^Y
zU(qNUtMhfej!j{7`Ygq7evSAK{~WK~Mq->xh1QtdjtWpC=)Lw1tF;Qfg;lnd=MXWi
z$LEhu6SPjVv3r?w8<%+ZowxY#%KM~7BZMUgeO4Bi=yp2UzLvUtSbK;`M|||U;OwJA
z%8$-)<?1Y>Q6J$!n%pDxHMNs7Jlox;+N{%wMyQ}jlnhCdm^AmU#f`SC5Lq;ryXD6d
zpFB^EDJvGx>4X!Vk6`ooJ|QH&AL3lc-CL2SDgD8KQmH_#Rs{j2N(Gz5L`jcUv&m>Q
zzz;%{k`zNXL5bryQ;ECJp_KgWZTc(MFSw`!go}D25riwNY`pwDjX(VN=uztW-OPO&
zAu@AZfuCx&-~W)(#j`A(y@(8p7@-hSWvV&%?@64{IBc=Cxx?#k{e;1=hv&JQvvFdL
z#~!)FNAG_~uiG7uI!A*Pt)Wz@vbJ%8dw1^g-iPlnRhgz#Dx)OyyIt<wzr*J47O90|
zp^Wr>j+z5{{fL=b5iF!=jMn5On9UB5oVznt(u}W%z*qssDcQ_1PaKhr8^^!4<F>&a
zlR`2a4k#2t3dIuV&Ya<+`XO<~w=r6i>U8W~MTQclNs7+?Nwb&C8JK!Z`tHwjkF)Qt
zKm;WgU-|-#n>%>>N93y81T#4?tfF&si><eR#_Bic(P7bz(T|xfOAA8h#;VfgC3+Gm
zeU$PrLSRzMs2vl_5lT8VNqLgM1K;;izNI`fOJ#PBD{sAvP7SBeU1VkX1oLytlnMn@
zAn5nI6beP^M@Jw%qS1hEzr`qtIdftIgvX=Ly~?lr#_!;VMJzT`mrkO5jw|S_eXyW&
z%o1l*DabcykEMkbo__XOcD4`rH~;qEapBxaYPBk1sYG|!W?E;7AsV76Vl*03E{3jX
zk)S*m(W7!iO1RBQ2r(wrPyCiX=aTO4#6oOS3PDKW(mBq4`30{3(GQ3QuInBX2&@Dn
z@VtzQ(2r=`y3O*F&m#(oP%2O%rOGtTcAs*!#+57YbN0-WRQ*|!7#j61>uV=zSDWk~
z>@qzwMZ>n8YE3C83W9OgxvQ%tJW|V~ptW{<)*OFwK}d?_3UjlIblZEZt!*$jyMXTr
zJONwpz0T89hH&jXrdUD~pfe(U9>h+L2G|MhqFz6u(KzJjaG%qs&rvEY^3qE$BZbG_
z?iRg%n-`wD%#G_e+&ii*LitoGQ|#{Tu(P|1G4R6WXDO9Rj7CF-qY=a5kbbvIy?(_0
z;UW762h{6z>h<R5zQuVzC;(r2#G@fQufN0m;xfz6J&G0P0g{2mI8QfY(F#f@mnhFK
z(A~V_%KL2ZTUlAhswxW$C)hvSqSx<n{rVlQet3i3{e3P!`xNDBmBBC~ik*+27x-9_
z`FDAWz>{vmk<(nV`|QbM4To`YUJu-1At{zB95$L%tC~Vs;L%Huar@p4zVymV>>o9$
z*E<MlGt+w}6cB5&W<5LjRe{GIeUd_<fHs=ImmD4K6AcEeu5Hlk_Xta&Yi^98TCL*y
zAy=<`$o>0w=nuNQ|NfiQYE@2boaDI|&a<?%#_R9AN4+~3cO6PeloDB-Se7mQkj;ky
z(!`lp8?DiuA-170WuM^_k5W2l(Ek2U5V64nQe)7{;nP|`ASJrpVej>~DF4c@;1#PR
z$&s5bm|Pg)yb1lVz=`Ekq><r+kAH@-8qZT$t2uq@Bxg^Z<lT4PrPFa|v^CH<Y~i2i
zb8vKp_=6{S`m*AeUtVFmu|O}HMVpYBwH{BH_Yik?X-kh{sRB}Qc72t4yHBGt0O6)U
zDudx9S!5>JaWqa<I<`+*SIE2RUY-Md3?8;lCGSk59W_E5O@Gj5s#Zn_OSx1c3=6DY
zdW4%Fyi2#+VKf*}E>B^xc%CE-%f!jZHNL)=S=;=4=kYJNsLNh7Yl%rIh9Qg3J;lPE
zJM6!EHA7_tsj!4O<f$~Cu?!D-?0)b8XD+{lu1t{|mdebvUm>}1;|{~#h{O6Wcen3Q
zDi^bYZjo<&>z8@+%{OVZTh0n06Is;j1K#`i2B*%Pr&66_bz>c&6xXj`XMb;>X0w5g
zB7#DJ%JdYaFd!5j&EbF_f9FrwJsR=lmtW+}=@SlV%nH|WW+aSJ0GGC)P+8IhmN9ud
z&9Xw(xU+PE2&*L4YJ|v(Gw1hLDinF*vB&uG%U>oemRVX};=k)R+1|WGnrMe?;p|fY
zjJBjvoTVl>Zi7vo{e@IyA%~xTJ|~2AQCWaZE#dMq>tB49#=rY6CNaot<F}arL|Ka$
zTeg4pF*6s=BhUKC*#*}XTelITd{P{L%y~UIE~T(Ot!AISom~zO59qdf^!r_+Xn-~b
zKk%5Ep5nxUpn84|V>RXCG#8(In!Wq?@rxx|?IB5TfM-G~UX@`oV1M_BASlx7x9N7<
zbb3wj1Z!s=;qpu0=Hjz2BK$B*4RLYrmSg-ik;Qdsm&=)35{$dS?C>|%c^bMRRI#&H
zr%){N(u>bA9Q1hf(Z?C~`+V@)cS%$0EG0A~NkX^Vr8GVDKzHeRD$AX8G=3#rtTx7=
zrF*uN{Q0`S|KL5&PMkdIPD+XO0~Viol;+MZgEu}xrdflMAty1C0s>!nnAEWQ;Z-hv
ztwph1LU<v*=W*lH581nW6VH>p|KZO#b@EZH^=LQRSQ+s2<>#25nq_76B>(fj_{XfQ
zZg6;T$kBfN*vN5WRxL!{PyxuXxvV|bAmvLb$(O(UGJ#*fmjx=Tix{2q_@j^0Y8{}p
zrgyN%@TY%DtM&%|!YN9RKZ}@ML0U(`&c`(ll$Opy!Dw#Zxk0^gi0=iw^uo)`%&u|n
z!dc#a>val&<m8D}g0Pr11qPjJW~Sz;*IOLyAMoxw?_g0pdhwD&@VZ@Y-MYm`AAL-t
z)uPuQ5XHIK`0;Z3b1sV$ut<g05w_mq_7C5nw!TbZWzI1n@;@pxNadbW1gchI_S7j3
zKK>b=)g)+)agO?rUA~02hLx3dl4Qj0?lvEN^dap|oztgJ^6tCuaq8r0&R@7d7%GGk
zjN&*mk?~Q|?Prb(oT(J8!I+F{Fg^_b3O_d?h~t>e`z=;iSEy8KJpSl2G~f6>&pr1v
z-}~NAz?zH?JyDqFadvrQ6h|Z8dHYQUqamp^6nvj39&q;DC2F-9;y9&Tu8lqGmY0|C
z{Q_|s(;f8b+>cmTJj;tOeu4G14bntYEEJI<H)J;%I_C;HrQEZTH;E6^rw$y_YqaBI
zr-L3Uj+m_o-Y%8VUwehMR-NA4SMaRINGHS?e0Sm@l)*R9y?a3E!=Ev|c!IcF$KuW|
zqg_K~uuRQPAwr){zeT&(qFgFru`DdkaQ5t@G@JM6kA_4E`L4>|ojp#?Uw7HgM{n@K
z+plu^LXjueZbJJuqgDfPD?$|uc+)k4FeDU;)~L_nc9&mz;fq|kb)9Q>Z)Z)P#TrMO
z5F%sDWp#-Z)&VoZOqvDUxu&E$#r0Sa>8!;h66J-YvCbx6n%(w3=P#V)xy#RT?!pV0
z&=9v8q@y1Dd-v#dI?T>3qLje*HRV#7-k^aBe7rEo6zo5r{nTHpsQbO&`@R3@pAkav
z-~adj{a@p!T6F4=Bq=4-DxCT93v}*nGdgH<e6!8RHcASVwzO|<GHSIbF0T?~Tp@(O
zrkcU1$JLMCXLace{caCWLbXz1d2NBe|M&kM-~H~NvAerJwn6|1Jjp0F{5Svaf6cvn
zd%XC<vn(#n;RPkmKmH7-FDA5FO%8UqP*S=PM=~Ud1^2d(_~DP<;>xu<9PA(Ql~-Qn
z>8CGbkZGEq+vIAyvE9TNt#kGY_Bg6zG$MPMoze_+>UdEPsGWq%3Z0&w=CQ{f!w*VK
z&Cc_UU-?ab|M&kf{n5x7UTQNA|4uTe>o|_byqh$2VZz#i&BDST*0RPB+muqFz|wP1
zQ~UTTts9%j%xG9!i!YpstA|H=)Moq5w>Y)1fYB+w4KvP=D^jGV+`%aXMhS%V*u7Wh
z#?5PNZr|s_kKW_{)*V{iHc8^F6g=foE>(H%(RJ4U!QY@%E)n`Ayuf2-W`-!%ES)}w
zLelRwNN-s_y>=BVVQFKXVlklEZh`PvSX$xbum2W>!ZfK+=yZfJDLD~8=iW?4p#mv<
z(u_lxJ*o(s5pZ+l&g2|VMJ{2Po1Niar->&$Dy1reSP`X~5R0$^Z8Y6(lLfs(k|yK7
z(^DSSn>;5rZzha`_l;%D)A?NYmYl<tTT^6)w52j<UwM(<-XYQTU1auM4TjW!ivV*!
zqjl>phc|ChTU|v8K`=NXx&9V!{`)^=^|8nC0*_BW`H;sReVl{sL+}*kVvT;k&)2{H
zD}-U0*M9m#+Raw>JRk3@1<K3f2(>{QKq#y^rVc%JCJH1Q8|$pCt<Y)=F^OSuZI0Ga
zpT(sW;v}NiOxV#0%~8UP=^>jt*v>ZoQ?HP2TtL~HJ0x-{Mw(*Nly<AdovnMcS`EJR
z(pT76S!EP!b~f+u?%O{>TETDp>bJRd>lR8W;yB`yPd){Fy4?W_^GjU1cnK*Ly<Qh7
zC5QDUS3kYR^;>t)f2j|9cnHqeA|)5zV2Sn)x%>S$DgWlzC{7jK-ev_FL4aWdq&7kN
zip9q+a^o*5#G?^5897dxoTgGKk&Ggqc;YfYd+$f|(k6wX$Lw5{dgB0VHH!-?jD`_m
zSay3@stH4P$SEwT$a%)cteV955f~Tt`S-{vEkZa!jSz|(H*T}KzQ*#>I^|-8b7wbr
z`>itFP8Y;v^eRmL-^YxfwX@zT6iSZIt31L|g$oy-APh^KJ#&tF{g6hpj_>(Yt5u$Q
z>Pf2A3WxOu|JDEb|B4oZPPfnY_7-6flBN}u5Mx5V@B5ho_CaF4E5IJEP-4slHtxOl
zt)ZaZ^%~SFsIPyC>S2@k&JJitZLuO#yIVydHGX30U%kWP$~p}T@i7jYkWeU;S(;v^
z5QLb-;7iHWbPX@`060O%zIgumuTrmXb97WEiY62#UkRebu=U;_GPib`TBU+XUT5dp
z8lA%$)wx;bDgmcX?z2$=sWnmDM<<HIMxEDQ|0y4Ra)U3u^kr6_UE-bVA28@gfJfe6
zAGeFi&3|QfO;n!bp|a8USS9V=a{+{bz>^{~@pt>}aL~sK1FF?32h9ctdwX>12R!$}
zmw4wL*ss@F**NJy0qX=mp67CYgCIc48rGf${>^{%-u<<Tx@PnLH;k{f=6m1!9#Iti
z+(lhJj&usBiBDT$b&1uNpP~ClKSE?3J_3YQpf$>K8%kX7(B9o)_WYw0n+{-W>mKET
zWjH*-+I2$D=lq#d1ZBm4`FnqlBo^%M)M+()=<Goal%n8yq&8*0-R1Yc^L_sHpZtIz
zP!tLQ#X^CpsVQb>XIY$^W`1UxLATGH?Okr%+2_vI7K21{Z}Tqy{P+I}t#+Mu`&C|k
z`31s)c!2Pj7in%plNV_q(gdBvlY_?E@wE}gpoL;~!NnR`ecb=f{jA)OGVq~Vo8{@}
zzDT{*q|rEHFd8^wvNoC3fgV4!Ns<u9@#JAnH5y4Ajs7~1$A91+AG7xm5;0Tb>{q_P
zN81NP{TOSUa%qhDcqzWo9DI6<nai6jpMQjcEuv+{o3s|0smXFq)X`3ZPp)3&jdx$;
zdq4gQy1mXA{?3;GTJ1LTRm&(D((899g=Lgf6iNk@DPS5c2HhTmPKRN$$H|kYm|31-
zYH<mnQ<69aA(&s>pio@`VbDpYfIH?F=(@iQ6*XqOF}qJJ0B6r)tjK(>oLD8vqHzj^
z0>Ur2PBD!MLr1>VSccj%jKB{-8^^O83>qX!><VN(;bA)QOTJMtMh}&-DD!YqLx3?4
z)CrIr8i|k;7Z*AG$_um)zlUrMvpKwlYy_)vK4@>q&c|0c`|>M<7{dEkuphp`W9!RY
z-MY_QrNYhYAF;N)NV#01wcjN4io{0oCx80qJbU?hHr7tFwRMZT_qWHTjEgTe?mA~G
zQs?S##`(L)(_S3+N{}QG*RS8=)QLy%J%b-Wnsz7_il7yvew&I-@KYZGiA56*>!j~}
z7j^$8-V?77O|K%-KE|pnNg<(ebV#dy#F>rLJp06xNE316>J>I`e@d<FQz=iedG9`(
zo0~}K5vN0L-?_ug)C{Fkfw`#~wI`oqFo-xjtn>EUKjZf17VS=V^6Wi8xP4eFw|~h}
zNpP`sLMQB8xx&nu6BJ*16rEj1X(gTnYqe_<vCLmQM|I-_-J>I{(ToV#c;;0edHzfE
z4)!>6>MS38^g7st+DsMFdES2Wr?fk5e(l$OgSnY0+RcEO*(v;j@6uObT&(3g$hdhX
z_uS0C!e=jt<2%pK&T-|FkFmnz*1g-TEG{!Uy+lgH#)&mPx%vsxnn_k|K9&>5xT!fR
zD;3i75Xxs`V~zRwWy)b0t^0iYo4-z)ju?$b2%ECJIENI1QB-n5j*J#CCg3ZH=lhxB
zb;8}6_;AXxwaG&->Wpg`!J>6Ss8fWIxbzQ2ytK%fufItDUw@bWUX#=soK=ZLr`8Er
zq#<o|neOyZ#R4{s#y@{*W`>}+jENx?LyEOUE}T6}Z2O#l^c7mIK3HhC`mR9}5??~W
z^Dt-{y&lQtH9VOim;Ybt{xew8EWPi<e&?Lmr8n1`s$1SyS9?2y>A_f-0RfVjB@_#i
zp@Fmt3K5Kj6bgtNz$bkW7or%k400t#NF*3sE|yD>7!U*oV|%8@wC^tO?e!*a`pnmz
zbM)cOyj48|h~1cm;fbiKd#keYX6Ad|bDr~m{>G8S>mF_8v*yjSeRz`g!5XzRWo2!h
zoA<W3dH(_3VN83l%$NVuf8Y}z`vfmP@f>enzQ}g7gOnqBlrirs>a4XWMwBHdGej4U
z@EE1cyzHzoC@j6s4&ku>KKjzIKNw}OKz*{vcH}W!?s4tfRi1nP6Kpjy&TQWIv!br-
zBR0okONry0M9fcoplkP`in{;sAO7Sj@ylQSGXLNo`~!U7=Qn@zH$V6j+RWM@bBNkA
zF9GGU^u#%Kuis&K=|0%ZenBFLg`^||j&RVSVfV%@_C5D8gi?4S3o$QOD%`&M4$a0k
z-N`nELXiUp7dd|VAo~^;Ie-2#Yb&dmR8uN=54jPVEP26LT$$h~Nff8_dJ%{&E2}%i
z>2;zJ-_vMAqD^Lp@W|V0w0HUaulz@nBqdH_KJ%HE@Vo$>Y7m*qRI02%5mGVsUlYb6
zrAEjUsWK&w8T~A6CH2X9%GLS^X@GGsvJi|Gg%Kg$Xne=x$f4utPsIG-wePaNxkaU1
z8aY1bY~fI&4PR!9?l@foviFlj!w)k({*-IX-h|H__Os(q`S23eQwQn3eT^SMV}(RW
zFzt}F3l}*2>`!0{<*ePRRHk$l0wEPf8wUN5o7e7g>B@Wj(bxVH!z9bn9CKhS>=9t4
zwP>4?4~ld<eFC>drBK5y1n5{Z2zLmRKB<YZq{zU-%jb|5Ha55DM?;3GV0Piq$oEYn
z(ky3mY!#Ee-|WqsM-bIL(TIGgj2;&mS<IWph%6TqDKoYA=<~`{$%e#gVrvN_MZrzc
zWL6eoI3$X*=&*5dul3jp$6`mhgyVvKT<CoyT^nn&Hqj{ZX?$2yLXs*;{pb;<&)#P1
zwM#gO04EE*PEaV3;2=@j(!6yG+3k|=-XgbpfrGOZw(f6meYeLae&JJGymFcEyz+Hk
zc;OR7@qn=3WpZ&psUEPgxx%4Chq!t@Q)5aoK14lygG#EAJ&4&u$b}#?M73i(4voOz
zxjCME{9`0B6pIOkyx?Z<KJ8|UUVA{R(O|kBp`?qBG@k2Xti&@Z;o4<(b~h=!@YhJI
zlL!-mwqP`!W&<6DyzrARl7t~^YY*7oUPTDQ^vn!WNOl?<blPoPY4IGv%w!Fthb*sd
zvaz|%+WID2jRu`=k5s2WN~ij0=~dYfnFWOb7!`G9G{5Cw$b%oe#ngd0)O?-nBrY<l
zhs@Od)*58B%;M?Ote=0E*c#I5qll+I$-c8EY29!c^oKn6{7>+`@4w3Q!ZapM>Gqqr
zuHs+(fB!kwq#QeXily`r`w#AaXvZ@?P}uBVkV1?Q;B0MvScv`bWEb>%eb(1DXtx_w
zYYVJxY|!bp(N>ZgIPm!MT;3R>n`^ixdp~N_dNsz5qOHUTVkb*3QVyQ)bL7x*CaP1+
z&(5;Gd5d5Cn}369t;DOZzDl`VAx$-&JNBD%_bf}sof9DhbF;H-ZEun$Np^6w0*Uh@
zh28i7`B4s?Xgqw9WJt5IN>rO?(j>GTL0>95$BvQzWRL!rzm6SRw6rK;h*9K_<OPIU
z;9tJQ)DuszHtdn;tSi%Qwm7%{7&4A%>})exoae;xQzUAKR27+;&2i_(J>oE?kaPDu
z{-l8<)kKLwgeid|vV37ch!BB5n=MxETV5ZfcceP&3_MCh=!XNo{+(}dcIzxJKKlZ1
zU%$k?^$nC`vwhTLi5c06Tcd3z-Y~+Vr2skd(<ECzvcoAPVV7283vJRpM+1Pc8!~97
z)N&pr-^Fu};S^W7e(Nf6(&zaXKhDP14ioh%SVg%MWOOX8vksf=qS8NyqV9*s_rCW%
ze*M>fop;`OhhP2GU**?-{nx40Y9ILjXZ)>)#;h61e8g8Ml{oORXJ{<%pxR;9iV~L8
zWIQTG2J^39zQ(EMF8NXk;pD*eaNIt-%`Uk@k()QJFg>}%?Cb((&z_@DD0BPPZH^y5
z#>&bv+uPeDBcXs~<nEuQ#OVkj8&v~YD@~K6hSZ8ZQ(rq~h(EHjTA<VG@YQer5xJa)
z?|XdeCqI#iG{{`Xh0JU>%&1NG=;;9LXftxK(#-<pE455niFn`7%5H42cWTf&&5Uc~
zgkq`0v7;wR6U}RHyvo*AgJRxA>+~UkTRcQ>Qj_kXJX&kw=;6ug&pIqwaCK%jGeU8Y
z2tVN9$DXBmXNAFL>k%|oB8<RM0+m|2w^xbUEpjunxT6Y0%4~xcNH*6uXg51-ZEf({
zo8Kc!!_jA<$rtjRJ9CDW<rUiP4v5Tq%<(MBvE=*`8=D*W3k530DsCW<d5I<_))8*m
zp`t1v1v{HtY;A4Qi$W?hOOz)UNs};ZpPC;+O-F5FA)V2`WyO4k&MNlQ&>~YvYHa|$
z7jcDh9trZ!suOJudX$f84D`d4N^T^UFqyAi6o#Xp$7n8O{FR4d2)oBEV@&k^2r7W=
zK9+m)c>K4HQYfo1`4WqdpQU?mnRKO_DV41zHFm^>%>*;?<}$-8-@_F<2r*>e{yAi?
zMPoO?!*KSoqujr>#-)qza`uVGNunYBUY{dN2k?Bwl`9w6+TNv9@G(giP2`Tag;EJh
z<qARIv%axS(jDSBl43DWkaLkJN~IFp+YQ?7E;BRpBZMHr3w$=WHZT@e);DOjnhcxw
zI5=Cw#sjR7xPgb%DK<8Q-46XbH;Dz*Kl!)N(#a4VlTs|?IezR2jm8GI@3t9)LvG%>
z$>pn8Y3w!#{D9}5eSv<bLmWjUsp0MSF7Vph@6zmah@xZ^XZGh&x&7%wzzFFaJKSe;
zJv9a;4E>D_R^PnJfzQ2&%*m0tu)sxQtc$jW*p*D2IEu}=^vpcN1Lx=lHNJiEKA%0j
zkCn}3_ASmcJGY-kV~6_O1Yt5j*_bDvIKyjiyn^d04jnp%qvS(`&SnBLoupY>fy^Si
zu$nA2Ks}tOj9QCWhQp9Q{KG${Q1nrrV0~>HsT3D4ou^itLpctmgZn6drUt9`7~Hu?
zy7K@Pr)ZI~cbXrq=ZG;{D5%#bnVOnqdU}>_w@p6h6DJ|-8$HgPIfLi<y!P5_yzu;U
zNFi`t$;{L=)p8Z1Q-WN8=Q`|PTI9}yJG2|khaRVotTlTviXRc;{^%Hc-x-q<Zf_7B
zo+WRxOw&H#xE|{431*hp*?s+TX6qv*!dOz1A!`nTaI3*o8nNo;(1Rf)LsG5T*<592
za++v3V7I?ZwOGTeS9s-n7n!c*DHVLqojc3Tn>QE^ha_><X_C?bkwwNS#ui8qOLnad
ziOwJ*V@8^LDIVGvWgi}Kl5pY5d6a^upMDM`++BSz;@Xb7Ya>yjwRZ1-C`aqEHJPcp
zu+gYSuxxFv6UUh=!g#`EovMUkx6M>}4v`4le3jb5KDJixaQFTS$k%!Di6?37wsEkO
z3t53bE-kE$FxwyOy8W4E!j&sm_`(;yz{@Yc%&}v~xPJXQ|L`CF!w*%^jYqb5v9}qF
z{j{ufsT?@K<f;8=HI_m@(zeO+hZwA^vvuP}<|QK~0WlBW`yQKjZ=jUm(&h8Gu0!CJ
z8T4b?%|4!+qh7BuIaMRoF;N(kB<A6%2TP(YiLoR9TQT|`Vx5@+?{VXPh@?7>7ZO6!
z>a_XJ_g>-r#Y?>T=3DP8z8=jJd|$pG;}%Dz<}|`3SEw*Gz3?zj>JgiaY=}`CH6=+B
z(j;X#9AJ#Wb#fd#dWy%-J%crZ-QCP%OOKJW(fb&`u5m&4Xrc1s4mCp7%!qH5ooXeO
zg$3r%9mjWE#0ZUrtn70}H)+!B(%#rW*f9@rROF<ZXwc`4AN+v!ZkL_LHtXB#l!^rc
z-=kJ5@>8GxG=-v17!C--kl|p+FpSY@hHgwv&rv9sS=(GAP7*vfhvyc^=gJg|HLBG~
zlyqqBws^2|mu6>&NK2l0=HpoDV)kOd9#KE>NQ-myD<emXy~Rb05aqEA&DiH?JeT&0
z!qI^udmJONg~k{NlMKb(OAJXN8gwvehBP@NW7F|96(h87FUex`|2_(W^*EhE$jq#D
zEJRS!#Zd|`2=MC@ESx)oD0-tJ7qrRLb4pmGvMg0x7LpYLS3_)Yq{8=H&QuLs@1AFA
z|4|N|+Q;1D3|DVlC)O5`O7330Mr&uC14|41tuOvp96x@X+1Uxo#RB=jBbRg8x3rIg
zhYwM$SMdV}sRWfug+qrAv$S-G#l<BiCuf<Qn8tNnwl>y?hPxE<nf1>*Z@h~yN^CvY
zW$pS^s=XarH*XLP`&gZT)kv)|QHY5`hJzkFxQN?$0N5;c$imd*3~@N*t=GT9l?(53
z?b2no*PD1UkAtGM)8v(Jev7x>e4AdFkQ&SV^-b2cb{K~7i2wT+ThIa7J@?4EH@fho
zIw4LIlCaO#jq7xmw{Z*zYY<2rnF+!S87H`~Z~$lDNz!wl#gq<mV|kO?-Xz<B$E)9e
zg9{gL^4tp_XL@d$iP=f^A3j39P~p_c$2hS6Ab0NE;>L{|*bz^5jCAjzRwISC9NmMC
z8k?3sq=IC_aOciF^0@+1Wd+RU<~DJXu(7^Qzt<%2ToxRkLS+Vd^l1v8`bBalUq*X*
zM4GMDW+c8bBf7;n>TPNNL2A`WT-W98ci!ZyU;AVJ*+2Uq_~SqRLoQvp!u6YXxp?Io
zsn(gVq!L(@5{3hc#muWN2m-dYwlF$VgpN6NV>^?;56I{8RLT`5Cni7gOCE(0*sPe3
zdTk<Y$fudDNKXiEOGQws;(YuW+}R?UY;8hVQms&uY#omi3RiDXpO_kToh(DANZ8sT
z)G@_^$DKQu5z@sM0f!GCqdYTBrBb6>tx&BNiQ|Yk&cqvzqcTprw#3?I1)LEiMv&;t
zvpyYBjk2lxqe2qT^T_3Lq+0Xd#rL>!<pR&0Jjd+BWX261C4cPg#qUcmV$^aUS&r!#
zX%m`_4Tgi>-s2vnwxp@1J4%*NPC_YgIDO&>K`sZva{s|?e(>7&@mz;?r;83_0<SpQ
zF93Sz=!31*9}1~kUS8%mfAcr_*Z=xo^Rqwuvs}1vfs-dsewe3t-_U845y++ca5In+
zSE-C>m@l#T_!&C4?vrk32#mFs6p0jBAiNsJY+Sv@@lSplS4f=An<VdilXt%TF7sy}
zqtkD*xwS^ER;9h$K{x>~yz~=Tt@$^<^RKAYYiy?ttj=;Uqd03nYMXg3!T4my=;DrA
zH}+3<qR%$igj%^oyWL@JW1aJt-ldSsGc!BG;Y0f#l2AuHs__J|LV$Ef-hDD#thAv%
zF^LtHUbl;DCBpSU$x;7*>@AlqDruUcos@pRMYUcBt8iSO6DQ8n@AY}-?N<q-m^2-u
zIFOw!#rxjF*mvv49Y!Y)^YzAJ0GqWiQrBbt>Bm{Sd>6CPBx5cZVk;<&m64&Lb^9*n
z*_TMAgRA5oZ?c$^#8DH>JUUJ(6$1R6OQDkImwxGsJXl`l#?7mAdVS(Og)?+J37y^m
zKNm1NzrdZFx43@u3QG$I$pr-<6YZp_VP|KTt?f+`lhWyTm|i?Vu~;XynNnCuXLOST
z<5ubDz5lRf$heSFdv+J>A*FlH<ks$Sg&x{qW)B^w16TxuBs*Y?xqDewY?N_pMyE?D
z$Rgjy4B%0wC9r20{84b~w2p9Hg{%DR(3LeM_LP272#gTaP98<KdThRR8MGk9Qb6Ka
z!_1*NpZw{^DFgzWWbW!1L#Z~w8R@ZkWt(@t@-6l+Eb!JFui^W7*4Nj`9Xi77#5DJA
z-o{A5{zHfO^d~>dSHJd$lulLXw7R_bv5zyqcz`RHuW|mpw>Wn6H0AOH2M!#<4+6Hg
zws9RrwN@nzhg`jUo?_lZ%9tb?aOJ`+%H<h+>9TtJ4ca^F%sC1t?_$yzqa<k<g69yA
z(NY~zln(LMI%4wR2sVJm))sHS_I=jw-y}|5{QM+y3&)tK7eFQ?CZf|FFpS2&et!uC
z+@CNU<zux*>kuLJ2yHePlkry5D8degEWiC82kS2&s(Bb4&|~vZ)!%lD^cJ2b>|3k@
zot+l9+CyI3b}7%zap~d>PMkf&(Icn0eEtI7{0#YGfwwPSLnnrXxdmotrw~f*0cB%5
zDgd48J?kDhx)-y1hw)dVgS%jH-y+*Pn><*#Mw<G>iNbXiVH9FSikB;(r9!%yK+XD7
z^p8D>S16NQ`aZUKAIBtE;bdr?;~*S|O0~+t#e+;w&2Z`JdtAJF5u=9`3pLK1eS+>F
z<PX2{@2FJD=#lxcqa2#8ZK{<br%s*XnWvv23j3V@pDy7zE;G|J)M_<qwHmoxo{9Pd
zrBab{sZ6O{Bux|kZ~xt|f27wuib8~|07cyIv7TB^D1#4#Qn4;1Tf(I}JJ0EtpXcuH
ze*@h$IIbcwU{JV9V~oTsuQPl6IIE5#jV1E%366#LxcvI}m_L4u_0?5YXID59?C0sn
z5719EEB7Tk%^tP-6zy)8=58lzXUd0bVrs@l;*VG+Nr{kS+b<YTf=BnDB#H5yJe6{Z
zZnwuf@4Z8%QsZOip6A=|zQ*Qm^I_bi9i1M;nDwNABuf*KDFR3}^txRf<&Z1ovP|44
zvvQJBU`$G*xtp!$CZ?3{GC4KR%=8i~%Qq3i(Ae4J```Z#Pe1(vaTHN5)!6L}K`0Or
zmevQlZXc?sJAV8)LI{5C*M5zcUV4d}H*a$D=FJ~@+|T{o&wZfh)5hR=Ib0|Bq0JyO
z^OSx-Zfct86NlJrT?4>kNKD4)kVurVbnf1TexIW35Zrl({O&4GAKlN}D-S3IIWD~W
z2A}@)=h<G}LIoZRixU)!RsQbZ{cre3|IdF&ePW8u^-cQS%=*GI;yGC{vLzW8KiL>Z
z#AGk%hrPZsrCToWah*ENW|OPeuX5zz30{5m2h2`S5CkfV9TOum{Rr+wC{S4|!CD8M
zYLu(+yc|iKaQDV7+|)96a6ewLxW}8yL<l;|OHERIFX7((o1{h&1O>{KGQOYV)ai4y
zJ3FLl#+@^xRnd;(dd3LT7_Iv#hBhgl=b}`AwQ^Lri!3fjgPfXX>g++*b{e21$V^@?
zMwS!8!20E@oanUi3VDQbkxEgnR=IoaUA8usnVC9BA)n*)iDM*${BQpDe~qJZeC=zm
zu(7^F9GN|Zv!f*KcF4c`@+<7$e~5+o85Z{M<JQfaeEoahAn<%9YE?Yxu-({ZZDWmL
z6lFY_AWwB-0msXsOlIG)*TZIX4H=~u*oQwe+nU8AUt7k?*u8>&Zys4jfhyygZ^Tu5
z1l<K9Er?QsH;R?hV;{ozp}vpu%EZVVbHu@vA~OsA2sm|CTbv-kb-mGCHlssY_WqS}
zGUaO!aPWm^88%u(w^lMyhhsT(Y?hCI?pbm*2|9cKI!bVaz$=v)wsv^wNS)rDyEHsS
zt{AX#?*Tb~m;SKD?8H3#_V1(D?eq4xU&jr6&Kx_*{L%rI4j$&&=RZamTAn)h5+{$J
zVYk^}>A)dQpFT~e-QwQ;yBt_Lj5V4@bCY_tK)czb)9$jhv4P{1$)oArKF`|b8q@g_
ze%zs63NUGi2z=t<kTmd#lZ0d#5*S6!Rp{+yh#MFJNK?v1pL1s(<Hn6Uv<C@Z+M#dn
z(Czo=4qEhwAxUCK$B(}n1s#A?f>ejZNxZi~KAiVi_d*Eh-&y9t+hz9s#M9tp?YBM_
z2N*OHzWdF~Z1uA!N2x$77<4*Z*$w#ksS`Z7a)}$~Z{rsgCyyWJ`psK-#ejXsj<It4
zHl6N}^XK2CSgtX@FiVj0iQ_C<&uB|yXB%S-bF)*p<$RWE5v9bjMmm{&NwJs-6Rusm
z%v*1Nhg!WtyPYr?T8hObDz!3&a*j%^PPbnmu^psiNNt2Pl6d+c-p8ll(s$wBy9k*>
z3Z2CnDwq1?G?UZQ?6ewu@69(DCU9WsFi$@FG0r~zJW)ENGx#IwQ<G?|@q-+vPM-vs
zP%f6~cH3-iZ;=ap4lW%SA#y%vPoL(QXP(0MJRBu6Bz|=Nb~>Gp{F)m;mWs$&6GsDD
zQHXL~d<&rvXsIy5(#aPH_AfDc;t;#<-5r(AS!YWLi$dU}DZ%D4s#u^~oI^h~&G37d
znOmHtv9(S)2w1*zi}|@pyw!Eea}x}^A)1syXqcLwXL@Fq&Gl^ty#b@1&&bPF!03)K
zjF34V7IEX2^$#Dg(=?`DoxoL!Znw+3=igy&YJtZN9pk&h_lT2>q9Uw7phmZ2#-p}l
zJ%CA3u0xVUnV=-_dHiF~vDs^|e(5ghU^o)GD74Aaj--%G)qRHjfTx~#f_zCc>~?7E
zWFC`OuU_EU=U-wwHFSqt2%!-oLWmE>asN<7T^MF+$lv;{-{QA^>$m>I$7B-m4^|MU
zVT@BK<BnWN$D6^p<s_xRNWt{E)AUyx#LL?#A+nehX|o9<GcHba{dFAKKqO6$El#n1
z^$u_C^m*|oKf^ocFL3S3c@7^wN_(?OA=skV8!|aL$KU-s{|CPD&94v+hjdz<M>6yt
z`djT0sz>UivAQaI)DI1tsSyi>9Iodfq@~~Ma^vPDCaP5~UA)RuPn{l}GP8-9@ySMf
zJ|p*@(^x}12x+%FC@GnrpU0*e-}l+v+@L?`v#_{;5LsBTRB}(~(Cv1Kq7Y*QK`wJl
z&*chCPRyZoM4Bd})C&Cwss15c{gB`H;|&p|IJpw8<1sGUGmkKZbU+3LW>1}@ap4;2
z#sEiV#l5u>B{GDmdw+%I-8)nl=0SM~*F!0f&8=Mur829l_X%=&7Wd6^?9@Sy96H3M
zOE=h9+oInKN2qG1#tR&W#2WtSn{TkQ)8MB+|C2oV#2Mz74&fKd+`V&~SKoY%VYiDH
z_~Z(CJU>Uy@kpd(Yqw9YA5kp@I0jNX;$GUU9Xg&7W5|DxD{03qAc<5D3pr!Vo+0%3
z`>|)=_}Rv4xba!kVo5a&6M^H}%%;OeByqBrKq5y-oYoI5bTTkQ@4ZGTGDrR&cgX$#
zSuBoPz9UmYVKF8}%FKGkN=RajGd;nv&%Vf==o_%U%JF9o@yrV+@Nx;REnrMa6oxnm
zQk{ZHsnjNEZZ`PoLp8p#eiu8vz@1wg$bN&U(_?FOotxJ$F*~<NwKhS&Kj8BDD_pyD
zlfy@Duzc$l)u~CO<Irz!;Rs7I*yZY_cWJkp^t(O!od!Z!2E#rZYnyCuwTZ(7&sd7R
z9eNwfTx{%6%==83l(|YkA)f`QOCbq74-;pWHp5;YC0r01)m|muzJ{qDN2LzcN)fyw
z51KK<IKlQp>{jF9I<i+}|J5nz0CXXT3?tHX6r{Abab|v&**vzmCS~RAcPY+Hkw0}1
zi9~9_)>@aZ{oZ$Itw&fTkgfxxPDj7h<hCQJ<;r~QZ~Qg3uifUE6E86Ax0#ul;l|DD
zXz6hB+&OkPws`O2C8TtC>EoY7I)Z$mKoUl5ZDj4QT&~3A#017buiN9n{biykMkvW}
z*k@s3o~5Pz6pA?}CZ`F9hB%B^T-b-B6brL6)T%Wq(}(!h#cizekgg!@^sq`~{&%Gr
zCO-c2^rS&7-@pnlJ0VF$xjKQ9&+(0K{2|@pkjmr?i%UmIjO3m7E;1PQ*tfXE($W&f
zq_|RWU}=d)vq8T<U}I;8#&=)cvo1M&<Pb~qbDTJSj8Y+wc;DO@A@4sT0{I9=s+l}+
zgx&QAm`;!F^>xhAGI?$2WRXA^DH$T^<Z{eE{}f4QNO)%*E3(4fla@q)LQuJXm+7DX
zX*Q>Fq@@8DkL}~B%ePoxYw`3aKg-Q)SGj%V3MY@B#o6ug)Z=HF+MeazH@=6XTmnBJ
zO*Gwh_kDCN_U578E9UqB`?S^?-}5q4_rRsu+TooGuk-x#ALI0aL!7^T=izfQqN<G6
zYAMZ_Py<#_tyV|&U@2GMxrlTXX{wo?nPaLx$<13g>2zA8qkAVWBl=01eT#E!H0tC$
z&7hyq?+@A7Sf|_GWwEx%_HuxbC8Rur3}!#jW&2P?-AA#>KRS9VcgWYv8CPI$Gsud&
zjFaOatw7bP9Dd<Rt~b63gOn6YPDl~}WjHh$Fh9JTAz)fio0;Ib?Jd@uFf}ttdgdsv
zzVbHtphT3G0Ykn%&6mIYM||?-Pcc8ckE>VTBh?Q{=8QH*HQG$G{M9jbGLF{*`=_kE
z-ENnO$q5b|*w5<n3ae}P8HP`B@!~a3ojQW&`(!lBQ6j8hMC2Uh*GkYP<nomZ42Q!J
zng)|o(|qh>FLC?MO$Ngu`}Xg{_kDsON0KBo8ar%lZ7>KEa``%?a*f^Q4r!W_rkdsD
z6{c!F^-5{4s7sRMeL>-4ZeARxfAX9A5F^@Jr%<clxN3w>WzyYo+fRBPUZuj!siUkl
zE@F`wtBI{di6pZE>4&UcyUu|ZU&NJ?K=-)x&Ua{UEQ5dx7vJXTCtqfAVvZ9hPT~6n
zZr!>G)>6y`c#cAw5wSUiSSN%U-nw*~^VjZDDd#wFV3Ct2j&kzEF<yB2b8M~OCyirL
zoscF&q?YV<dc5)<{(x4`@|jQl1k;n%3~AMN9NNp+U^=6GjtZ+pzfT2gWTwd-BOK$n
zuJL2U-m*PX)seMJ$AzCjkXi_1$UEccIE_v-Z$X1bN{tekcT|>#nayRR)ZU@%(b%vb
zXY5=*&thd{)a#5*TP((ElyLX>O<HS0lTtZ)gu_3h@vpwkQ%~)~^TDPmN!Y`!<PnL$
zCYpFSBye4{@8g&9Y_9EcrX*NV1C|#TDU~N#y?KrG-UeY3aPxW-;rJ-eCrT}~T#g6V
zWA0zOiZPl*3li7GEtDwdDkC0Kwywo#Oro=vqOTO5a>;iZn8qr%c01e&2jm>Zv^Fdi
zQyyP%aN{<fSJ~rkNo#SH!*I}{7zDV9APGZyH!tI!Tf|VLKaBX!>u)kh(+pb}x9zjL
zVefkXRT~pW&(ghqnK%habh@|IN$H@+Ii1Gh`VKU9xbfBRa<o`PF6`s}g;jp=>IH@!
z5K3TG#(iTP?U*De;c{uMPkt`PiHQjsTMZ5!J50Y{rkJa6bNM#wn`<0Ce24?b5Afj5
zU0#3t4Txj%r95B%>R0Hrc6j`W$N0o2KEushR~QbvbUSULFk)h20zb$R#WB0P+Z2ig
zlv_Z^5>e8@&H0>t>I}>4H<?+QWM=6wJN^{819kk{%cOCXDa{=hsa$fU3dQLeeDe#e
zZ~u?DX_$!>q(d0T+`GRF(&6#P9;cWuuy1~mBZm%AEEI|3m?VwygMh?JB5m2+X(CmQ
z>FFA^N}1`Y39981Gt<+QibZ_ac_>hMUjg^$SJWkO%)z(bqx8~KJh*n7?&>;?!^bG4
znhi%GM<7hR-D6XFR7(Zqr(Z&S`RmxtHi?Bm2s{K@8nkCwu116vhcuC_&rS1YYlp{L
zDUOhwef9}n|HiAx*x~4@L!@!U!u%pTi^p)3rBW%gy7~a+s7H#mtgRuH%2v%0A}2(4
zFOB(>KU~#~yKy^>1_uu8XKrqWz<1eLUuJ82ox?LrtTuMpZMU+n(H<cNGP?*(rn&}e
zB&ScF;ly!+CLwV1L}AFAZ@tdk>>MvV^CD~O>+J6C&}nrj<~{nICIP#wH~NUw$D}#t
z<|>RMB)h#XlamSg+%Xb;6s3IZq3#E|ZhxtX5?8+R2UMnKaeWWRaoO7pOo~!2ke%>#
zVwpN|oayCNny+7CB&;VF2u?qHg2!JvLOvj~Y8BuZffx9ksx-NL`3k2#{v7YVeT_SJ
zuahsIWA*L>X0+kxk$Kvk21kz_<L>>#+`hYvjZ=ip{H+8Ap<Ha5sYSDp2PaecJfiY@
z|Hx3LC`qVRCWxaRhnEhqwz0-ea|4OX+WIz2OZ&5m(54t9V+xxQ4^bZ94|wL;7YIBL
zs}1dL57)`D+t_7ubCrqu6!9?5ESaTXI2h6x_PBQQ3a*pm{=F4WoO}!+6poW+rzS~E
zk~pMEx`*nFZ-NYxwj{A8j_o5(*?;zN>pNfLsfBrb*Uv;T<M%lt6gwk+%f#suv~I1U
z*LMLiYUfxK7HKS<J9iPG#xJHg%U4L>{stG`xy97U)2ytm@YE9q*D28NMReOkqG7_h
zb7#4I`xbZa-ofh3<XQ=dwURW|L@L1;$#x@Uw|R|s-@AtEI@Iem9y@)E6GslBtfUu0
zuNQLt_5)u1{uNd?8-#<1pZ@&knVK$UEleRmK8!#5;V8KG6U;NQN9KBO-*+7t*~?gy
zO~P?Y7m*>jSwcm2@Cjo`bD4p)8Tk*UNhSgjVx-cNWVUc4;_o;`!U&T+?MJW|T>SUH
zM=4iE?3+QjnKE3O%<@9141pF(A(Di2I6zj)96db4eElS@jWOEdDvz+&MhHtTpCgR>
z_`c7e*C+3K<h+2bMocX)d8RNRDkZ$O`GA?FqeL4%8ym}5VM%O{P&y3yF=^mY3|#Vl
z0pANiSQ6uB0*ffY1_7mF0jCr&Y{v+x@T?^nc3IolV!gRdGYlCbs5_Ep@*Yo5xg4&i
zD3eeqPG`>@_!-YeC<52dtUsj1N<rTB+1TF1japbg&u*i|-MbG)Ngwe3_2aK*K?iX9
z3qQ+cQrKu<<E+)CM+nULKxni^x(eU*Xx+cZ?f>)&@#&{|>z!qWeT#HF5|Kh;aDsps
zOV0C2fL3FR$@y7c+Z<4sn#VNPIWG5+pP9uk7kOr?PN&=D&b`~n)Nt<EComygIe(r%
z{OTWY?b3OCU$DA$pRL9g^=h35_wI7#>J@?@;K-3<>^67k_4-Ut&GHj3e-@zv(iHNA
zBFBy&q19L=K{J2o8LG4MY(@zxHaM=2G+9)e6=~LiaU481Pp2}AojwR_SF<*qfQ^kc
z4jwqc<4-(~Bf*h^0|)jow=mE2<P51v3Db~zy-cx?!&i!@pE!s2^i$+>S*vv{F8h;*
z_|Y%_9Dq7&!7Q(G=G_9rQzz*+H`$I8rt=;Mi!ce9X>XKeuBMvS!XgtNdxrRrzm4uE
znWA55mXFNxozI<Pdosul_A-kX8|I4iW|~aiyF~x-Pg0+q=fRyjOw3mahdsh5;>dvo
z24TY82e&wM=rCK`4O-m+FrtrGgBU5yg|#EgFK5Jm9}E6|6n!y9kQkj6Ogd%%{34A;
zlkLql7A6lcU#ZgU?vB=LIpWieC@j|Qc@(B;${TO|fMFP794H0_l(eKe;_RtYR3}Rm
z>m^!^2F)lW@FcxnhpRU(VXeS(i_|Cf5hi_vm1qt9zM)bZkSi@>`!2Go(fdCbdjFw{
zy5Il(-~T|5`=Fe<-PbR0XZkBlfAR&Sn{8&&W(KQcnRaHip^0-Yhd%inw;H=dw{KI~
zU*WlrJ;uVJS*)Y8s41)WKJ-$(&S?$zuRWl@bB4#CJ<aw;n`^gja^mP=HtyYLcHce<
zp5)5q3p`j}#cG|Im5w$<7?ab@UjJV2eQ);=w($@Ml#)F6{L{FON4Zd;+liT;m>{1o
zV0DbJhUGiASz4S$D<4LP%{T}gY4_ymV}fNA4N*dn4;(z#;ljm>gu_0=VUAv}gRz;v
z-_~~KzJB`nISw5<jOzwG{`m7GsUZwQy4@~2+Z*IPgVhfwe71OuY$no784G+Q9iX*W
zUgN;Avv?<uqFgol+$_eXDB+BZ8$njf9D4q7ZvNr75m8p`cnahw?>Ah3K)QLCeCIO$
z%1w&hZ4OV>xV^bfF7UZ={W8aopP;qVMmRa1dG-bF-@n7g<|+t?qRix2dD)zblFYWo
zlJU#NCe%q9(|NGY^}FkM&i4phmqcr#SQDi&#t5$7xWd2woqvr^H9z|^pJi&Q1|$#L
z)^;p#5F^hyg?faftF1s7(AuK)L$S=BZ`~duH=b8yMIc#GCxyUs9TGB?wl<d3X!Ph;
z_HeOBA9z&pIwtPQJ&FI1f05%G4|wn2{tkzJ_Os-wMMAAHD6C}9Xd5R>K}ph>(xAaq
zvV)^EMkpLRA|yMCaJvbU7B2`$grK{*MX?%ywVXXZNfHjx@sO9Rnx_}0_}+sCT1oN?
zN7)G@)^<1P5Bnq*w&I9p5)<?W<Rs*kLK#USV4$Gl=a`+DB~_BJ(WIJcP?q&1;&QV`
zgd$LiGVn8V9-rA)#SYupL{ltPn5xwY`~amCCQbGZPQK@ngb{=O0Oc1@aY)qML{3hz
zzOqiU(}O)c_rHvhJAIhP{4a9*%U>ak4X)z?dXMXg$xO>}T->05X|}m}XPdR=HMH_G
z4=)o_n5Z*ov>ApWULlWBnM<iKhF-J7v19YxO)LXZ=E&v-o$eJDkDg#=YCm4FPi<nF
zH0smqw($da?&B|V>ZvDLzW;#rdk+Zv16<!B1*BhL-+?1cPEAuRmKeq<`NAYhdaSIj
zGcz|uy<VeOD$pHnqy0RkqfatFHA7?2!vrN-{SHC}2$_;Emq?Nn&+`yUkw{5532`Q;
zFzYu^nfEUB=}De>@ng*GJA|iF(zs2xvq7`Fj^nvZ)Tem%x##dbk@>6%*gre{zRz1c
z^824h!S+Fh1=do-(Hjr=gj?jbr=Or9iX25FM-2*r!E7{W?W<$0WdV(T;uL*S3jgU_
zgl@!@r;qT)Gp87q9F)-{R-v>em6Faxo$FDENC(VZzDV`ZNnB;v+}t9UFEA@5Nq@ln
z;UnmM`+4W>cL|3>9F-Y`%SXA%#*9dI*}rAR{f{VMc@z^DOia{x?6K3N5!9*y@vzI8
z(`R|Ge1~B=V75BJt!A5~x0|WC<;Wv-@7{fQ7}u%Aa|5)nM6tn`glc_?N}-H0aO~t6
z-uUjT^t)X|rNn~=515}`<k+zjcwWHuySKUg-n&%GReaB-Tq#n>`IJgk!YPAT0*W8(
zx_zjk?&p90=im4JqiT{z|Ly~m?8A_&U;hfapa_l~A~q5`3J@G;W(uVsPDzF#rI~pS
ze*Pygufy{%9HvloFuF(Jl+h+dCkbeS1e{zRSMBh^VxHgq+E?+8pJaXc20@|1=H@o1
zkDX-w{t96nv47tjxx%N323=N`*9pTYlTD9agX20RN&1lU^ayWiwBzoz6%QRc$cdB3
zXl!*!W5fL7EKxt@)QQvdx;qGK*?R9alByv%@ib|WN29U=3lKs*;sZ4jNg7R(YS0NA
z+uPi^b&Xhu+`6^P^wca9lT%o0Xti4G>}=!6`wY4ro_hL8cDC>E#+z?(@7_J)Bq1F3
z_~}o74l2dH#lno#V_Azf-HR&wNW>D&HWz;HcR5x3B95Q6=A;EZLXO67QftH1=~HZ8
zz0Kg-ePWd<BONSALqT`R-Tppq&LCC7(Zwn5_wMoLttR`ApXToED;zm=ke&4qw-B(n
z?<hx(oS@zAvc9oNl5Ubl7Ud|?H2bVm4MJvow>|NT&1^$bBuNz08H8CG`miq#!0r3D
z_+NhScL2*5zVJCFC(EO_q3lDIkDRoH5F^F(m_wI6pSD@y_vnfIp@72dskbZ-Tct{7
zsBF1XMe39|Y43T}X{*sDAw#?~BV-AqG!sF#2&D*NP->KQ_z|>UjZN6TdYOobxtCwU
z%*^9D24^&vg^(x<28o}<O!vFg-5BF2jAWD%n&RYqqE4Is_BN%-DWvC;^8@;WA-O^U
zp#)B~!p7P<@i3txcKPDTIbK>yc<s(MKUnGG)h9`v1%^qF{-8^DIAoZn3{uNxnj)|`
zR^SNmwP9^_g}yNaiJ|PfB>8}$lgBFxyeLMbnol39@|pR7sI`UbC?+N*$>s9o^92mx
zI1bS;+Pjv8Vu|kNHql@}eM(Ys9PY2&p;kY}_U<Ne^3lcb{Y4F_eCD5c3N<urzy3CE
zr-frAHXR|&qYYJ%<qo%s(`>5=Qlk($C6y1U3Bh!Y!A=vMrle8K#Pl>foffIk_&Grf
z%W~=yG~uP`Dc0{?CBOd!)ygE;Ji<&cQ5jOH<k;->=(SqxE0j5O`YeGbaiWN*)27+o
z-AmfASdJY(hVOX<L5_SekK<TkYX}{eq_RkLYMQgPD(!yAj)3h!OuI9{aXoZjXP$h1
zX44ZC@}!PI8-<u!Lijh3^g$_xq2b<x`y4)YoMPb8Yxih|A^9MWjZ++BDYy=vBi^@0
z{8b#g+Jd9V-@45?B39?>bWD+>QNlGhThGmKNR*~HuE#{TN3&Yt_6yH*B<k^=>~iJQ
z9I+cis)?NfP8x!?s3gTG#jsl8dKfcUUuW`mf)pv8omFggnw=m=G4L5&xx(DiKAw8&
zDZ+5T%7eRXtZ!p&W^p2979uJ}beGIv{2`rXZ?74x!%{h%JNE=LvvnGqeM+SQLlqN~
zGQWS3{-8ymeGZ)3&yD6J@x2GQX);n&tB1U+G3CZsq{=8`Lb(W)<Jjp_NHpzcgV{ov
ziTV_s-T<LoR#(?(Z11qMxy|9jhX`sXID2jq)4M|yM)aF~R?JPDgGaCjC1S@QihmAB
z-4Ao>N~IF#&!6Yep+g_$Y4%1UQdG0a?XUhZ3(jY7mJZ=6i*z1w-ghLJlsFk6MU^A_
zs>~D~L#2a^0FNSUwehM&(kR8X5XTYH8j9sA=nkJ>@cCAAgUOjm?%lpeJWRMCQXW5g
zg7x)Hwzjr8e)craz4&pGZ-0YcyHAoNSYh#V9+Q*PSS<JNuab1b3{jM1PV^@0VVP0Z
zWT(^O;)TmB&K<$`HPuRvX3}GRVUg8U&9IZu3_A?p`3}96o8*r@gMZ*Osplf?=snAE
zHetpyv|3|zf=Lr@-?>Js+o03yQmM@H^;f<^wK|Cug26E6^*7(c7{iU5w|V-hCqSnJ
zu4HCrk~oR!b-k>m4Vfsz8q(B~rkXS{U?0YfeH2AqY!jNdZgBgX-(=?FFOUX#tfxS-
z=eZ;ug&r8vg3sYkKFh709n4M>?Lj`LIB;T#$6q+YL`mZ+hjf@wov!oj#s>H56%IW2
zG(x_~`FCIE+_|S{Z#5|wcd%CQw|?nwv%I=Ot@eQB2P=ewkkllj7O4OfduS;Dg4BQ+
zwZf##ik~0$?bFtB>+Ws7`pvIWDCGFU7d}O?P#O`*v(ykVD(oD`MURTLB*}d8#u>a;
zYaGR%PoNMAI|}O-;t>@q8G76w^t11kqFSxcS?!|r!x*+KKX~MIC<Rhv5px2Aum<HY
z3U0~Tx$+~}OcD&9HgvAtWPo7tCqIc^ba8|l33sw^dskX&;V!;P7z{OOzmG{`yh@p%
zlqYq-&-;XpF71^y%2Sh+N+psw!3u$2ERjYrle4pI+`mPdLfjiL=R5q(vvVAs4)}u`
zE4;QoAdy8%Za_6x!1H~iD>0+{Dozs|1SQW0<x|B)N=0AyS?+Y{bz02E5k)kQADQG6
zHA%eDr7~4#a;iog#aJUrQjPQ#)*9kyu$M(z$QNm*>sc~@umrBcNiz|}ARc?o!JkO>
z`71gcB&q8#{p>U3g=F))-@}a}B9akbGR2GzMtwJ6J71w0yU0|dMs^~OQl$Nm-0UR9
z2}!rnKn#X}K{x`FYA#&fASe{@^Lf70kzCtu@bUe1a=iv&(qXPrr&_3!FV&C|YB+>$
ziAFl06^3X7u9f&@i}GFSr804xV4?&k@X+y)#0px*;5Y%fsYQy_DGruArj$=N)oe=1
zdR*Ymdl89SBaT<G))K{WHiwKQUn~-*N%o$LGX(QzkXz@`PRh=1i{|zU2Pae(`&8Y}
z$$j%EDKp{7p7+=z;J{ze5n4zE7!%`PyTM6cF#V}dQm;*O*Kvp}n8A=VO;Mgh(&-SU
z0c};}E6*KcyT6GvF$fPQOweiofrA?j(O8sDh+G$2EU-SOup3P#i_;`qF*|Dy2;+oy
zr9!D(rFH%y)AO@D`S=SXb7{k{A7YHgU?|qg)T(ucVaUCEs~Jb0%mXuXbjmtK+GZYu
z_wL;wN%m7K&*AwVK~Q09Yd~STM!hsmcfCbfw<#X-QFEv0-Q6bL+Q5y{EDld+?gzp^
zmbvJR5=9inVugLRS$e}R!zjV`ip)>%qnOW+RQ4V2+_{IfhNqw0;o#HHQFTgGr#mD%
zrQh#o&uvrW&_m7%Wca}j;D4!{I%~6xz+}C_-EY6nfuE_7F8KI(?8RSM&`4}*sCPG*
zHcdQ7gBu{UL0X81A&#ra`2lf%0I9(h0w<TF)9SD<2WLzNr#Qo<{vDFm9pZ4v8@-St
zM~@H=LcaTzS2%L=IOoosBe9m%<rOBUrYMvu>^67l^_uv;qTja!fzQN54KPH*7$GFB
zk#e?Dnc#_YFQF5O<MasvO%k^#7K%t*lBmayj%h_HwXM4(JDZr{HqMz(V4NIQJPZ%E
zBg>JDHel)XdMvNpXQ#2n=5_<EbNsu1|3B014EVX9`@$aLWvnGhbY{M2?Op_pHU^~}
zQmxT8<G_p~Y0^aR@!a;v^M9HO<U<UDz)LinZ@h_^s8Bq64%hLsD4%iMmy(dA#9>73
z=n>{$dWxNIeuKlO=Q(}$2-6Ew;H3y{&~cn)B3eNqA8@wXW_e|qQ>PB|NB{9v9#odm
z%-{z;t?rOpw;u4<zVNsB*0;Yxzu#lIyRv6WX+}Ir2_u2dSSdQnV;zZQe)wZhO5yuH
zQ5f>hg}0fVou)of<FlXnNgPLJt!0(96fIf%6)Q)l$Lw<*yYoM4A0uVvDVHW=cGYOp
zA01u-FJNhLk)M3|6NE{^)MTCSZLN`{53l3cK^;ad2RrsLd>Bscx(-@1GD`iiFENnC
zqIn6lF5O|(FLC5EFQF!;z(}moSYaqQ2~(|YgyWK?hN#sh84d{B9rW}BLA{D}0^EE`
z-0jleY2a0>2uBjf5grc0QwX43pP{$2OU`rAjv(yznK72XwlvA1e4BTdTRa#V`l(<b
zELvpoKf-fS(qo_v4P!A1gw<#bbA=*@tY%uI*g$b;-z>+11lOcY9IR3-c$w*S4CQ)(
zUeu?W&t-?8B-xukN=XL8A+6me`TQ1zQh`DtK<g|)=3y!KXSu%~Z6f`}4S_*OOPW_q
zKX;b2*JSkv?;`r0%*tlOXO4`dnVTZjl0r_R@_7)FxYZ^eMz{v_K;zXa<b@^dbQuf>
z<o$qte}K`Ja5!Lkb_PFRWF;;0+ZQ)^e73^Nhl@nCnCP{s<%`rxbt;ZW*-@nU3`>@w
z(e!N{Wg$o{?Lm)~wH0g_QJR`0auwY;C9hnLPSu$TJObsA8p}>%c&i_AaaVJzp-E*x
z80|6$2ZX~6u}4UTX@ch$hzCOwX>bFN(&6Jot=mYt<N}Y6J^d7A&&xPpN`P`lpX5VY
z#a~)q^+SxNPKXTTtiaKbUcAYgaoG2{PqP%b+)ysg^d#+8hj6-zFamcN(s9?=wS9~$
zFd_hLNTfrq)n-^L69xh9V2FrQVx>r{73Ml!wkCaY!(H|tJi>dgpC@Wvqc}Z}o``VE
z1^Ra$;26o%r=DP<QYJC5w!DJ!Tq@NXNtz<NyX5jNNt%+&=cw1KSgVPKF(7EQIv8Up
z7OI>*{T%&nNIsw9D#g}%mvopSgDh6DK8PttnvNflJ6R;DpCq|*4V8r1eJsF^T{5iM
zqsmmOWfU<RTN`xyA*JFB-+twfh-1Sq{^H-HUaxcW)@`hX&5bRNm;3Z}h1pylI~?G7
zKEuI~QniPw=8(d0#3cO34{*)?Qj0o9zGaSu&dp`kUwencpZhqrQYVhH3osTfRO2pF
zgM?~eNgYLs1*tN$PAS6e7WwHq%5iW#mu|aFG00J=R(S8i18R=uSdioNfhCl1S-yRf
zTf19qtls6|!DG}blf3u(+Y~1XoO|jie(G~S%Znd-ncg7g```OMfA9yt%f5YzdM!k`
z4y95aOU5PfeV^5pRqow?K&4zq3YUZ-m0A%cQo7v%{cfMI7t`-_sHq`=E@Wz*v`yjW
zyChb?nHLbw6xOOtZ6gFmW=6yaSYEzQtJPq8XN!IN4wA+K&+#dh9A0_l72-H%c5ar9
zjm=TQM5by@)ASMEfhCUPy-n!R^oYIovbCAF&NyD@#~PUcL-k@-zWWXbrsnYv2Kd6;
zLz7S$8k@ueq+93kv!^&$iCCES@dAa`U0huln-XDCgTzv(R5>vfbMNLYj{aA_NNwT*
zYa92eOqRHR{U-ZPoZzkVS9s>B7n!Kf0PlAihjHY~*jzeWWXD2-KW!LekV4@5KH)Ir
z^|xMQZgzpmdW~nFeG-WR%){2a$b9#V$<WtvvV&9^;+OF>GhUIEF4g)BuAh59Iw&9r
z0+yDR_}piI3X38r<Y@0~aO>tJbfUo+j7_mt@1>PI8caQofO8yYB*6G+h8?$&r6iXM
zR?lB1H&bE%^UolQHFT2VqNtgK2>>cV)xHI05FvyjD3tMoB05Pik<1Q4L5^@R#Lwr5
zdqd)02OGF31%qyfT&avBC4MePlFu=SLYkXB?%dv_Sg%r<a(Q~zXTM)yqm#0wecsz@
z(~KN^EwQ~APel~1VPGtWo_~(%!%J**8q7@3ax@;W{ey2(D_5Cv1i5^`%v6;m8sNyx
zW2)V3GdYu|(`#p(-!vJudPY3Tky1X5W0FCip-u4$HG0ty-_2*-9BuwHpXv{f5!v_Q
zr~Xl0z7O`(MTTu#V~O()2S5HCz0D@g8#l1A!4-M1l5Xm<QLUkq1f>LiwSt?^lU6G9
zwi-n3E;0!bi9>OE3Zo>$FvLw$TqQ||102s~cXN~ZLlt~ak>qo{wbA3sc85<NTi^?a
zCUD{bjg@;y$EOtJDdh8bu1irmOiPbcK<YxQB#ZkGkQ0K)cUeUs63whq%=ix07`iqg
zLGtAXA>Uqa<K(MI*JU{9vE6JijE88QB3uWXnk<pVjQrPxV3@=>^$DV&iU|AE>s5{%
zIf#%C%Xjf3{K>!i1s%Y^8bTvRrsWEyp?C2X>$wu;&%T7LO%fbjB<c6)D8qyl?1mk7
zRYWJ3wc;HtR;i44={n@LnhYnaXvZNAd}{3utsr1!a+0U7tWYK)t~#7Kcn~QZR_@;A
zeq)1tp-gpl0oN~a^}WmFssa0#4w0$?x2|0v(J7@;8LZENr6Yu4Rzp;(Wh@%2v%PwK
zZJm4f@8fxSbZS7RnRlUZ84g1hPs|cG6T-m&C5D)!h;a=<cNiR+!Y!Xhyn7$hT0<C}
znS@(|9E*jNWqxX!I2zDwHJF{-PdF453T2dx`07``MyJzeer}G9%`LjU9!1rp?Nzc4
z%IF}jjV7OTn6g!J#<KJQ-ADdXi@L;Srj(9w@KacQ_Y#$Zb4;H-i|{3}HCPORlTzz!
zARI+61RKXlN8(m1paf3Nqu*)KZ!{=QPU7Tqgj?H`y*#9rf~&}TiX|svJD2iKM=-Z|
zh~e5b;z66MSKh<Tm6@2FVz<%eTYvO*o_OjpDwQhJGy8e+vB$v}cAL9QPR(%m@L{%h
zws`xkH`&?QA`Ay~yF)7F7LDz7hJzg{wGz^i+`f4SZ7dH~R%y1{>};)a^01GT79D}>
z`WRtQ_b)(e7xnb#h^zBR9imZKonq6Jey_`|dpFtKUMC;qdFlC&lPgTJxHQl0n>V?7
z<sw_#>pcJbi~O(u?)NaplB9<1#x|X92j#k8A=R3DckeSZIgJ+tAlXCr#s$Emo^?O&
z&<0#GURAob!}@FQP@b+qrB0G0nHXk-#+uYn8Z?*@9V+{)phxJ371%+9%y~q^FeBDm
zi&rdCFSR+74|(r9uW|a^aW>W(+`fB{V}}p0a^nV%oj8m!5r+>S=E&V6+`Yew&CGB{
zC{~tyFcvsu<A*$s#>&*aVM*eM$$E{JG3@T{a_PzirYEL2c<2DrQ?;xhGZrfz%IqH&
zSeaLxlz>ubkwqw3p{P$RP_0j)O%@*gDC#C<hW1G*@%;kTi5Wiesh?tPYn?DjM)5X}
z;6Mg+oRY?Ic1_bX;|nHf*5<Y9BWeA_*5U#lk|Cgn33p$6kIBVZoD&6v@$e+fIhuUE
ziV`l40h1<RG)BfqD@b;Fm|VtJ8ipZ$E+C4N3^^sUJ+a1MQ&5hgR4${fp*H1kwpzx?
z`v@n?bS#&K%y!$<iX}cH4smt4k<CQ~)8!&1sd%uv%ZXq7Ii{X?k>CAyzeBUvqlcwB
zQDGq`sCXXb$qCXZi%W{*n7g-kDOaluqgG~`4jEO!W{|p+SsPKOSrB&^hB(qgb+(8%
zE}&Evk@hF?kuy%7eVBwG*h@<I(-eGvQNtU}lL(Cu<Z5M(JpVNBZSG(;LQ=<&*p#+#
z>BI(=bHQrDeupF;;`kmx$;A#VIvPOa5h<5^twyxlL}&}bLAeelNf08XxxG$lvQC_8
zgpl-%!>bRP^drURk4<xIx{hZxNt&|OY?2IPOp@S8N#OZtDHs?7CMK1J*jNZHHAmqU
ze0udsY~W)I{M*YL{GbcP+6+mY&}#3}?RM$)x?l`xs&Rd9Pt-IFL!5k$La|5^_KB2>
zUz^2jZ!xzx&t$DaCNCcq`wwqq{!1z70Ag#1u*6oN07oct8rI&v!0b$g<Ig?A4Zj3a
zRR*0Vb<c3OwTTp#VyqBRj8&GTQo%Zkq!3W(b|~)b(3zSdEfnw?yO5np8Tt+bZP{>R
zX3OAiMa=IzOtgB9c-Ut1-T>v7n4FwuxD&Fz`Witw&(hp}tdk-f#fvY!j3WZ#xXZie
zUq@>}AzwjjL#x%GP%KleR9IWN$MK_wsnyE}N3goGK^R8lQ;)6g7Q5RU%q>We1}B!J
z(nE=qWWh&NkCD9P<FDMsC?APNsI2{3C>E)eU3RvXag;}I7}M-*@(X|cXXtkgcW&S0
z=B=xI>Qg_-@BaHQ6AmME+~e?Ki{Wx-Bp%Gf9@{r>F<+acUo0^EfFR3{Rn+}SnQ9+q
z#1`@+S9=4*{gB(Qy+dvP0%|soNHnfc6iJy+hB%=@8;II1k}xFbbSckHB3<zE9^r16
z0g7C$OyC8X!Yb!;^w<pXa7ee+=5rIUzZ&rBwxTzAkOyn`Fs4rw_UH~f^!qUj6BAr|
z>s_wC^BxlmQ!E@fNG@N(bA9S156=}8^EpnPJjv0ce;pxUZGDxme&x%=VHYW5&=HN*
zT`s+Im176bvhiS-#`=By-aYEW33OD$kqRd$A#^7D*4wvfzxE#qUij<8xd3ZZtTtGk
z(%9K%b8CYniuu&bpXI>5{iMcWes+#qH?A=^GtCp{PLVH^va|qe&^jyZkdl6{Pp{u6
zNm91Acj$I|)O<f{^T}-S8E-Osh22NMsf(<|7vQ6a4QyPw!Q`ofOrI?v7!m1BMx87t
zJ!X0xro05LEU8pDI%`)%y)J<_^1F*OA3cG=^L>s@TW;LG$+kJn`h%Oe`4T&gUG~k+
zvAed;z9U60UA)4|$~tM1?4>AVe;TB6Gv1;Fz2_A;b|U9T_fnhvmdgd?^EuYm*12)(
zIwy{v<~!f{0e|fapFydts4}BCB|G|D$6kk0$}BlUW<G9O8z!fwsn#YzWU*~x+=Lxb
z5#&f-XGRLwa=FG6Pru0O#yX95o5dtTr<nrQn9P4L3d4*aI&S&utQ9I1h_NgEk2z8U
zLJ3l7&`5-F=&rSR@SV3gQ?A0)6chaxGs6%mGzg1Hl1#BKz-mJ@2r<GE6st(bL1T%H
zK?#9O4Mqrbq7l|0h02O1lOi2Op-`sadY}}-Sah0__&I#XVGu=BoIWoeD%05QlFt`$
zibag)u{2Yqw|b4#4=CmwZ0gc_=PD;`#J;J;k?C`mmeCsYSYKHq8am9KDYMhrqF$ZN
ztbvlu&(S<GQCz9$cRB>FOWyYhQpL{QWmdP7jCW}nd1+<W)^i;S`5g6XnVIP+rl+TI
zavqJwF0E#poko-0Mw8)inAza${iMH`cC0fJIw&JX$ctv`<bI}?&vNgZZG;}sN3pCW
z{bmzGf>sjgxDY8s>}IJqMHj~^Y?5G;7^^a_r0IufEJ3kA*y*5cW(b@_5x(nEuI1Q&
za0a1MmUlw_&DCAb?yGTb(qqPRsMZU(A~T04ib#?WUqU`fQN|q^U~49)@)*}=h~h$$
z5_My~cWaXes)RSG*xlY@x6#J&oJ`F+J|J0eU5_}*IFbE+mwL5?n+u2&O<!0J>^sD6
z;~LYGGx(mn_t}kGa{ncc6ieGC)QpQiq6P^8gefcEdX1T>I>l3`=^Kx}aM|)4?k5RJ
z7$Y4IzthI)_t85|y0cRxd7rS7C%@UGlx#AXnqa$Hqq?<2vr^&Gp(#X~Qr+2Qvk;R%
zTINs#7H5xeZ}l#Ns1Ji4D@m7jFQJ|<GGix^u7~40%+DM^gacGy@qEcdeHtkx2M!!&
zdG$7l?XYiYnx~(7jGfI*Tqi{*1LCgX;@cOgmS)(vze~5(BHX%3rBbETuM&CzHWxr@
zNL`EL_9#9%i%<r=wv7}nD6-_CVgYLt`n?{d*(I*uStahbs8*`figoJM3NOC+B1=pA
zsZ}eiudUPB-e7uZACBwO>vkE2AxV<bYxkI2UFGp3M@T;4Vc<g*b+uaUeMMZa*W-yN
zp7;?1)7{<O5A=N2jNJ9dn}Jp|Z>_O&{tEl+b&Sm8D8VEVxuTCwQWOF|S3=?txBGOY
zq&PE)aspg0VX(7<E1*~?ki;>{aq*PI^>Yk{31P3zlcku&14X`kCBmKA&+1^nT4Rl2
zZx>@AFfkLo7D2f}dpKm}>Rm$5r93^u^71;<lh^Ts03ig|k|FtG4k_X2kwfh6H0XAF
ztgLLZxwXU6;Zr!;vT^rq8f(j(DFn1{tr8bK{2-6C2AL)pn=<V6Xx+O*Ay+2%%;yo<
z%!|~dv>Q9LcXoLCv8Q?L<SCSixqbTvZ@u<?-g^BN0>8j7{M;9~bMG!nDy+$jXeTFU
z84P3Ie*1Oq-M>eWEAY~DFW~z@My~%M6zb7--A94e86yx5XlY3i#JwSRUVn$$f&Do3
zfHWS0F!)YFt-FEeI2hw1CAhB2(iKK8BW|?GPt=G^j7`&wH(4yxZ}<7!{t{nqZ!@<r
z!=2mrP{OBHpP*nBcWzwez|s<~|1oxVwpd@^CQUU;KD4dTX$oW!Z3u<1BbBoq=YIUq
z+#M-Y^Z7h;^E2#hH@JE0DigIDD=X{l-#3G`0xL{*4drMfmI@pfVKS$Eq$NoV)rm=}
z_37+27{j9<{g~(HD1{JNQKxl+gW}-Q5nlepXL$A1Z?fBJlJh<E$g@rx{l3JIF#?+*
zeju#KOqqYoVXZ<bvVx5aoz+ILb9I@U<sWeJ<rgUK+$Yw3oDq0o0XOhLiaj$?Eez6g
zv3B&j!<d{OU{Z_YyJ!P(7~;A)gp!%qB#g842I-(pii{Itosc9NlI%4}EiyrhwCJcu
z!FO<#pp0eb+S^<^{|24%epUtpmb$wvo|wTemGK=P;aQ@=fY#O~x*v1;#1c_DL<kot
zC2<^QdA(|!b1O00Wd4h>X3*)Q>r>=pj^e}-9=!8yj5WAYWM;;*)4cexm-wll{uxf4
zI!V1&rC2Hx<Z_u8oYur~N~_i4-rYOA_Jber)vtYxciy`|r`P`|Xs;h=2%$-}#@g(_
zBZa{$xa@!N46W@Q+E;GSoLHb;E<!4i#>KiSJD^xlo`V+@kj4^jbqM<dD&-1Ml42bP
z>$@mGx(-%mt_MbII_(aQRD@x|=E^p`-3~UE6!JwP1UDKUe%j-c2PWBH79`3L<Z{@+
zr!OQm3A=ah(%xt=U#@d?`3fuL8o}wa1jcgX-Wt2IM6WX-T-Wr5F^;1$FR^x)f}f)o
zwUAQbdI3UaQBA|)0M~VKl!GS~q1Gfy0fK^;`OH1?>HWoAq(6sYfv!!l#=|OwRoTH*
zS%=<spWe4#BR4rgcY1+9IlLn*?V^WDQ%qt>ygXVsxPu|(jRw8BNm4IIl#j`^J6NL$
z%T)&DQs%W<EaS#8y}DpxYm=_&abf=ib2j8aGQ)O1VZF0UbEiq18f+9&8g`hil+n2o
zcXnD_x^f=3USo0J0cIxWsZ>f}1P2z5aOTwGcy1YEdaM*m^m_x^tv**)Z&ICJq~=%X
zY~JDi-OJ250)1tL>A=vKm;g_qjYftcLMN2BACQ|avpK9ox1V{8N`+%AgPjJYQk~Uy
zOm~nlJA0U!N}UVm-{#iMTfF>Jzrbc!QL9u~TV12GxkX=l6lM>S?7qXDJGW@Ix;%E~
zER)qL(&*71^`9Rfs;K+N|M(w&q)Hu$z;z@7jj&nPY8+Y~eE%X7^K%rA9cE(CVaXcY
zToEBXq|Q$0eqLjuD03y(4$9AgAL1nry+)JVWF6ruqBzF$JyHv`=_xh_Erx@T#mO@N
z%?pcs=hilVcuV8Y>}RXrrO|G%9mn+3l&N9JLf|p&`xsFs+}&lR8*yhuo_Bl?FYs{#
zKg&i;Qj#PSI4CQa4(iZq(^<RA%I+5BT%Ne!B|llf=op<k*fd61mo!Y#QAF5pqprS6
z?ZlJBg^3K4HYr%m%;W^8j-902Y|<NcSzleob1Y|09YqR{_0=`DHa1X7F&qxKaN#13
z%CXyQkfxgZ_wQq^<k_d5ra$P@?6ipDIP2j9So;XlV+e=gk9~~=3daFN7DO+!L*v#O
z?VGo7pQ|BK4<#LDY|4DpN5vk);ee#y$MIZpbJJKQadQE~%?84A@Nzkf@6&6y$@>At
zYMH_ME~j0K_NREO6%sbD66=UFCmustmzx(aaq7v(dF=5gh~w93?zE6f;W*i5>^RN{
zL9u&gs(+Ff;rLw{E7)x{DOajYPUJCZ%G&yU;^=AKefvD~bDu&;H6q22L@y(vwUug*
zbZt_NbS>pdnN-INdLe=I5Uz`r4_kC&I6h5Mk~GCh65=?boU3x;_-VS`4)2|Ri#YG&
z2ryQUkf!YQYpsdm2st*HHU^t$k~sdcud##0mx2T}LZA!=gA-e}-n-6`a*paufN)Zh
zZh}nBh>vPeo{LZlnRChIT!bqTX@-)<QB2^C%t}Yb$!Qqkc)3h0BORox2-_`0*ue>W
zteZo+9yyg2<lR;q*9{QTlB5x#(Rf1_8^s6(j_y%xbQro5xM_qxStpoWM2i%gM#RQJ
z;E*Jm>ST#Ry~u-|HFCLp#;uYPK_<d5BN{<!Er^6PQJ7dGl)zZcVDmnW)fI}q&vVZ_
z%`gA*FY}2{eTw<TMU-+M&bJZzmhspmxxA%Zt+8)ui5Fjbi7$QWOT6*s8~lI%#c%V~
zuYH65VDMvK!~fjS2x7x13mXs;g+SE{oc!dAycWiE-3fdXka}Q7w9+JuGOnjF2&GU+
zY@)$t#a_<y3DX#>Kp2a1Btoi@fwM#*Ns^SP+oQ8-G813{gRo1d*P>dRWSAygx_zGG
z`z9#{9s<K|f4~6C$;CM?{Hy<o=H<8f{O7(vQAt*l4RUgVjW8zkJchk3c3XYCppYT;
z)>0}L*j{f_DKF5`jZCSp-mjD|<Z{^QDF$JWUZ+pLKSU{mw07?*{+HxM_@|AoO*3Je
z5GXk^gh!BQ=w4f+_O<WR{@I^o+xNL2$22_;Cto180-2@=n<5G+GEFh5i_LrBheX9Z
zZl?=EVe&rGk`*t$PbegvB&58xMc*B8<?JkztpW2FH*u*@Dpgo-Y}1Q-TpJG9h+}5m
z4zoFrnx7}>N}9KBvv%)3i81)f!_Q@Vwd;FWlV%3{Ix`Y3do|oNC0V}7*4jOKag5Ff
zq-4=wZW5A25_BGEG(x8o!VqQKsQqUNuRp*=BZOiYXa1+|>^!uF9DDL94pxeIy-g18
zpQl{(s5qK>)gc%7AS73=T;|N_)A*$tT0^(n<NEa*luKpKo;poqXNT4Y%=i6JMcps|
z@-Oe5NI#0<iIhMn`s2J)n{5V3qsR4c|9~gU1xnpDY|=-&5}_ijHkm^{7L#ZQQ_?U-
z7F?v~VLczxkf_&1<b9L_@t}*FFJO&BsXE1=*}+B$3d3iPOi&I4|K`#vk(Xy;{{ecv
z4%_V}52BdWI3h3mRCe1qj>6`AYSl^h?LR=SR;AbMKx!d1s5HgVK)-`CXc2c?Y;`)^
z97I&|KGV`;LPtzi^XO!V@B_kLAJ6xQ;}~l+Mn@Eq7`<^14o@)7nq62}q}kcx#`TND
zT9azS%K9qHt1F~NQz=(i-&$dFYcnHRTFd(SI_uNR6p97*FYLql_)jnx3~9GpnL;m&
zxpCtLSFT*$v+xiyYy3$uHmCh@uW>>kR0bW}tVWG`3HM)pkEx|4?BpzguuKt?&%30<
z7)J<1n$mB#Nt2k$!VE@PWKIw^cJb!saPm3god#u3;GlT0)}vSuobY>`URomR57@fD
z%o~G<qeqTXDwKKs>)++E$Ir60zRE<c&i2+8iB3?iLq1nvI0)Ho^?t;QaGZ5Hj_49X
zaNxjx3i%u}({ps&+Z<Tj&)UX1jrJyL*`?8JGf}Vc{wTQYv$bS-sF0;DILbu`k2Fqr
zaO)1970ex2!Y>s^i4%4tv@mF+Nt1+Br3?lG3Z*i>U*PPSr)V~t+`N97V!_Q6wR;6&
z=8hu}d&u4RFXJfsv9GZ!HJ(%qlgx`wTlS2Hy~Hx>Bq*uStu7|gWCkl5tjXN@JrC=F
zms6lKPMM#}(QUN$&@f{y%5g};%+f_F7gZ<__Xd<IRXUv=c3VxHe4bpjiWB4syb@Ij
zNpFB5MMp!F>twVB1N~vbus@_wDB$LE1f>PmqXAzZSRNbpIapKp+9fucFtIEioTu6E
zW=R>2gOtvQ?<Ww-B2$A)mWVSL8d^O;A?M)deIlC@tzP4W$IkG7{>@+F+|!Sfn1se|
zgJ!cyE*Ic>F4|}cg%ag*m0YfX6c&rYiVUT7eMx<Cf}ec(6FmLoQ~dic{~rJSKl;bq
zf3Tb#&V>BP;~W7Jqlk>4?2I;b3&KK*MlIB_$4}C^*F-1}C+}hvSv-yu=qSc&jqZn#
z7)+cZ9T(|HRL(;V60C(ZjS&J|KOl(*SYwcmgX4RII>HW<>|i4WDis(JA$u+IP5`B8
zu5X1@N|GoVlJ>gDIOgJ7gV5xNjH1`=;?`?SA3n%#K2M_))9-c(+g;pTW(4nf9tRH9
ziNhYbpq9-QYjIs?FUoSvYlY0yQuzV7q9pE?D0(^KWRNX%N{+?r|C#QqecvB@L+t(b
zmpWQTBZb8$!NJ93*5XJkskWr=UgtnL;QUjk(L3w7QX@QtLLjux936~7Bq=KH;dcd5
z$tM?uMEL^VppQ`&>${LfU=-R5=vC_EI|Hl+r_-mIQ}mx%;F;?yaIcG~*O2Zs?e;Fa
zy)MnckXzx9lIT+0ZeU%Bs+0+eC8igaaB=~IVM;DG2%}Ltt082+iR?ESwp-lq^td0V
z9G|RH=!bYE7b6UQ8WP10g-(ZIsfZKDB#uP&LiD`G?oPqh1|S&5AyM@hXDT_Cd<SbZ
z>2RCd%lFuBG>}TtY;7YXC@FCqmz`#lJ2!4{;J_j3l`2m>_5|~@^C(BLyt<4Z1Z;0N
zKG600P(|J1;v#?N@BAIU^rbIx^ytw)%lvrX82bUH#-dSZ3|3@5L&H1k=y%_vx==#<
z1`{PHBe0Id%lR2ZF9cDikB$wwN|rm>X|?fvmpDz4u|x|&G7OQfj}np~pC|10QLaO3
zE#2KN2Ry|Wk5BNsSJ$|{El^&-fvH)nl5_`s8odtPB*qC-yfi_iDK46H&B62r6p@tk
z0f~V1R*&wmM-)XgQb{Zlrd)@4pqjKevsA!i0QgzRyi({WA=N4IFeGObML)nc8(9pC
z$q0vo?tu5+e1qopCPAT!7ZhnU+pMmyvEA4tm&;T9cnOoFD68?5V5(k4*obaB#8^YM
zR3RS}>2><7t!}W<*yX{42W;=`jKbdEFSh@a+KwM~jKTyMq+%4aW-(a$%MI?o_BIDT
z^>Jo<4fg32r&tCYNE57^N4Yt|UYFi>lfrZjM|q@iNN;0{!t^ALbO^&3=_*bgnIH{A
z_IW8Q1;Z701|6oXY~1ByXP2de$C;?l^5!ew=k&2tc%>pYu3e{Es}aUAgJGW}3Auj#
zCfi%>OjRv2)t}6=yU<nxl70K;`Rr#u#p=oy7>TG1sFf!0gFNA&M^Z>wTiK*uod996
znK4}EKZwavCXfgTsUC9m>J?U2tISN#GBZ1iO|p}5zt<%-21hxW!qVt0^&&~|{giHR
zhx)_>Lb?<RWuAEa8Fm{xY^>cTNs>o6d|82|9EFJ=s$i|P#G@kVqZoQDc)?&17$GT^
zCAGXDZMV_GDDx#`r1aGZVXP5Jh$92yp+#to7kK1LMS?;OV^VYykt7Z{3Zo6A8X0&v
zd51WRi7ZrR7V!H$I=kESw>B}Q3U0BIc?Y^4!$A*6IG{6eRvc&ETDeM<fh_Z2d4=*p
zLD?_U!QqdVukv|a<M0%8yDjn)v&ix!&Bi)Gb%CHfi3;-I=a7DY^*o080Qx%+TZTi2
z_COI7e2nk2W?iD$DGtpZ;?4KI!z<tWBc`TiIez>sCr+NCT&c3LwZi}F|NTE;t)*Np
zv2Wi29y@!ElPAwnpP0>P6*f~-TP!)x<(Izri<~-jivR0>_Zz(a##?_b1>K{=kpf8;
zw<jz}1HCBX!OjpV17x{`bhFndNAY%X6k($X84IE)n{V1=Vv@kgwjiZc=AEfEZq6r-
z23T!KO^OwgpjIc5eQcuf6nG<lOX<5fZk}NhAzX_U3hCtuotQK=46R`1rO$Ha#h1yq
zmMP`}%%P)P-b&cn-NooM%S;{&n4Bpyx3GYpkGOuh!BlyMR%Z*T6w-CaN3#r#R$262
zzEB|FFOcelq%@5?dzMs79!-gXhoY+=9U}yP^tISqbN}iHD!Sts!;uf+UIw}lXkl<7
zFt1(27F_yOiR)<WAT#>bu0s;|SU_6vyM2Tw(X}FBp+IifAoX0FD9$WF(iF!eMAF4b
zNxNR7yxXGhx#ZH6Qh&hCM2SVIu{U?v={l5xfa>f5snPU?1Dau<-6+OMA&6t<!iYZ5
zPg9&>76(`=1Vq}f-t5p1dkiAYt^p%b76YHspikLJXzi~u)o-y`%#pfI#w|%TUJ{eW
znm#U}Yfw`~?2ZF66UP~8c=y$BWt`A_1?BpzZS1hQy}@p?!OYAo&pr17IoH9H7FP)h
zIhQ!@5{3o^OixY|#|fM3TQu8k9;~l_pzHOain_n|_x>Kg{oB9IZ~fM9@#2dw@})0*
zi7$Tfi{$h94?9l)_RLw0wWHI!$P`~fkdqFJMVS#9yIpjcJqDGvZB^hRT?eO_LtFij
ziz6(<IKguqL}D@pVXCt@5p7VeL%CKZienm$HuvsrlPi^&syqDGCnos8y<Ohi>9Dig
z#W)UyQi16MOQ00(ZilEpz`@dJw&=t$xkOV|0{V(|ZMZRrh^)j>;5nK_DS4vgbFd^i
zJei{;hY0Hv1OZA(eAh$jj4(dz_K-LrQ`o)@!{ZoL%NDn|Pt@;n|IT$1?c(RBn3&v8
zy*y7GhV1S(`1ZHHMI0Miy#d-9e(>fyymRp)QJfGbIx`s7=|hf{`B7EpkAC%Ij*;3c
zQ*$b`%?fu%!`j<d*talE!5yNB5T43T@Fs(Lf_w$f&0{4P?cf9jR1gt&JNTYQA)hCS
zV{&ee>VzQ)eVW_be5NMw+8wSeF41}r(jT<BdE*kR<#py}7rB1<I>%3);!9up-*fKC
zXV~1@<#&Jgf92b+{4q(KASL8-0i{xbB*|K$S)A9X)U}$67cWvR&tkP<c6O3MH|Esw
z({#H#C;@9XFLUPTJSM1tNi+54h)O>`kYt{5KC^T4bXqOme)}!@{UP%U3%vB=i`>3^
zk^W$i@m7_<b+e4*RBN<OxpU_x#(I>>b;{)uQ`2)i`PB2gI%pAwF-fZT(n26R6()=m
zM9f3f?#CQ*#Pt(oCwhUwv034|Fjt1KPa4OW(po|)!4+BYCv!e{4tNgAb+Ld>G}85{
z&CM`qWUVi&O~$txwa1mpHkBZsBkXsGQ;A<FGnw=0w_51dfFw!rOSMdSDa+)_J=(2Z
z3<5t_ASgPxg#uTuZPDq+lzV*+96!sUkNqUCe|L^6@B9G0)<E|{=KjX7aBtufVQ3kT
zs8$Bg;rx5|>20SBheLvr%TLW!NJGnRPti>!+8BO!b%&K;miz53nr~cUVc$N!^eg{I
zo`2zGD%A>7WG%Y2%@uCkxQgR|wJGN>yv85@@t3LBr+EC_(|q!ipXb!cvv|IXo6D1=
z7855tdG2xk*+2Ow{Qdv-|HP}```-Vb0^-j;v@uy5eAG^~WE(mnWqUhjV0~0A$V4ti
zXE=@&2;~s?665)pZj2XLjEqR)ltH&gWoiOr44&tc3W-iroO}@^7@eR8L)=0@uiK+G
zF@pl4)-K7QPdd=(I3ZW}v&~dXd^blRBuSbQ`W|UwX;YvxJ5PZg%1E|$+T6Ulj7bwh
zG%}WWg*=tXX++-T>h*iL(g!q2n&A6BD7okJI4;&=o#5pH^5r7^;gHx#m^!q?!jyxf
zoNVbEsiGNuX~}}*_2{0IRxqk-GZg=aAU(mO&+(U2<aLZ8#bU<6_9HY#2#1&iEhIt<
znofdNE>5S1h%z#PtH4ytSl>na9w`=A7=#oU&&9O{r8S9kut{bW<wOxCpP|8V;M1v;
z38ILjo4Y*mputL|#BSZ8erA%TI}Psbw%F|ntm{xN7MP!1!U{>N+ac)>FakEWcj!hj
zg*4l*buZhmZ}lT$B(8$MfmtazR&XenBx=$jq=QS2ctq2QTn~~AeRRVpt4VYZQORQq
zN0G@YLMkHTp_7RFH!hJFpP)F!#KeB4YWql&g!PqGzTWyO!!TtKrI<M8Yp=YD<2l4h
zLL85r3Px@Psn!qUJO4A|Llt$u@f*LvZ~Vq@@X9N%@Gt(wzu;GY^;h}(fB*0E#V>x5
zU-^|^;koCY`%sf57^(S$jgi8PHUkAX)F&jBz>v1vq+v7)OBR^K5($f!<Pbpst{|T;
zARL*=@U$U~VqEEfu{fSfs0WyIM5s<RQh;9wm@DR~%}!#K#0X8d*WpE-a-bM;ttWY7
zvrTt5Vz|4D@jQY;iAt%8>v=eSj;io7oumk9#W6`QBqy8raSVx}?n(aki6SS9KFP3y
z))5GodS!~dm&Xr+QHB!)%4HaKDFr#)*rJCWq6Zf+`<_FlF7vZfJoD^N@ah|v8HNcJ
zF{Efy42E1Wk2`GB>U0QWgSMHm?)uILgKPghhd0{rEg7e7T(~0ysuxjMeL!tr73&$&
zNaMsQRtl8kU=xdpQv{GQ02i$5;`u&<RtvvgM|xQ{t5+!ExIX<sgzhx?nS}!R^)|gi
zjeC)yJ7^L|1A4;_$}4i^{6#vA9r~RvM^8M?FaG>5QOcKDUAxEH+B(g4bEJOLL}7?A
zmb-WFGZ+jJ!eL?V2syt(lC&ulHSy4}u(-(TYDzpv=v==@STW?E{v>IohPH}vToLwR
zzOPb>sp%>9Ei7>E%wse=9g+mvoi6w8-6o$efVFs@PcD~Zr;)jh=W=;co!Mq|+8tci
zA(zXsu&~6TL&vj}i6q@Ku*SaMM0t$XeH7JuPGu&`)})NBb*&HxnNrRPgfhrl9^pAi
z*TKvCgu88IoZ@*3Z=wu=pUsm*XVeNn_&&n*S%2_=xy1#fkVvIQh~+3TBNw2OF!PJD
z2}UXk^-0p)kl}8V{`NNcS`DXApj4Yc8%?fI!jTSEI+@C>xy<6y35-?vj^fDi<D7i#
z9N+!7|B7$_lm9yhW*6DqPWZ|nULmysIFir(-1AIMIow@s@ZfF}$8qs}!E+T4IUEpb
zSS}bIBr(l9yKHT(5cP&s3I+Zz|Lt$`*-!s0!f}YAAuG%GxOC|P|MB<#12?bT-Yd$S
zKv5{<*w|X<?!6m)=hd(A^wS^XZ~Ve9arD>;3Y98Rw~xSZ?8s66(SQH{;8*^4|649y
zxq|+P>$+j85x&7n2Qw<zGd_i6v(d-R7eNTDwwVA(TT*F}!ofttj7OPTtdvOCL0L;Y
z9MIg{#>?ekBt#(1TAM;COcIk0G=AWsOiFWOnM!?%;^Yi*uS>t(BncyScee;iWnASW
zyi8Chgo9F+Lg0`u6d9zN%~ViuVY=1ebWU?eyQE2g<3(7T;yRMnR-4{#mwglSgu?+!
zIY?<qtw0++#_F;-Dg#Pr3Y7wloox~m(QVvd-xHtMqrfEAVpBuA-Qn8RYiw+8v9YyH
zuh(ZV=p(e{%<0qY+rN+b`Dy0nX7NU*#bcULR@m{E2Hb@hh9pd?T!#4D6eTn+hSUhG
zl7vM;&(XN;Au<fJ^+pIpYRR=MQ7(tf1z1-y%m-K}^J>hBqnI>sz#3B5#W&!@A*tho
zF=!>wuE*s=^W0vX1!0lOQ0jFlSk0NrfUSYU+Gd;PZi8^QiSb=>g%Y_^73I3PL5`~Q
zQC47NL??+by@;aILWdD1(G(Osd8$mSk|W=5Q%#4omCKD%jiK_1a=|D?0tQ~dd@{^L
zK7)j~G@$s{BJ#4qu*t2vYn(j#aoQK}(&-H;(<fhvKtQ2XVi5IccH2bJ2-!21)$N^+
z<a+&?EKOd1`DI>y`DOm;KmDit&hPvV|MFk{OI~>41x}tk$(O$LC4TMKe(eK2PvE2&
z>tgl}Gj@bz7#3&i=zfnR%{Fy8@`zA2^Q(%&Au3Lw;Gl8}j2g9=9Ks|e@Uson5fTt2
zQH1avq>{vGf-n-VSi&uou~MK@jqepG<Vw^^J3JfYIJY~+_wKalbW&RVF2iUSqYW(~
ziE<v&mBfi4ZzbpyX%h|ziO~s_i9CPf#1xZZpLTN#$5E7vWlE(o)oK-Y6a;Q{<{9QX
zqZ~a2Vr1pfyL*RV`ZQKlNVVpTi<eky_er&7XSm6ojqMEakt4;b-pe5PD>L$r!J-hx
z>=o|HDlCe8$zk7g9xGx@yN^x4DhKI1xD^kpvW7?0>0!c%V783P=Ly3QnP`$h4^arP
z!eVWTQG#;4M%ddW>Gt@UBXbnJHeWGi?)XKPw;vFP4NS6wu9V5^g#YEA|8w@8IK=5E
zpQe;^*uU=pFMa$|>|5N&^&8iC_0?BcUAd1CFgr8Hox68gnBULB{2^McHdFN)zMHVV
zyi2)MW=jS1;~{JDklvjO3_4qs9)B5U;Utk`F^?o^K*n`5Ml&4tkv8*DD;G-KxOJUD
zzeA~5BpO6}@j}DlkWQz?4_^N+zF#CsBnJ;3qu28?gH)lYlqN{iK53e~&mT`~JxV-K
z53_wgf{nsc3EFy$DVTz+?d}Op5I7X3XUNscG*`Ds`+ahy9A3GAO*GnS(r}2-8s&Oe
znYFrwu?Xd$`~ZVwx7A~MzK#`Y?>#A@FbdpU9-F3_2qn!jI;HDVo||RZ?-KTU<lKOK
zP=M+bI*E}IQeoKG-lS5N9C-FIu5Dx)mdRj1P?_S?3orBi-~MMfGG+VDRhpd!N1uI)
z`MGK8vlVK!5}*C4j}wL|l}dsC`FFlaD^_&7J<jw0qwde6EW57zOz=DB-0{sX=Eymb
znV3fc1Sf!`NJ_K@%9`A=Y)htFQdL`RcgrQwZLg|o+e^09w#vP_Y<Jn!Qi;}JTM{K(
zB&~^q#Tg($%picAr-;ZHUcC9;@eKXPeVG6yQU(N2wD(%MA|oU7MdUs2-m}l%`}h5F
zo|u(P`+~wuQS<8TL+^SIcieFwr=Nd;FF*b$pZ@$OIdkSTz1|+f;g}>XDN8*c*as;m
z1finUj+ss}%JMU8U)$#GZ+{o>e)oG>=qzE16gy82{`Locn!o-x{wCw;>(^EXD2-1|
z2-Nke6X2}FIFG{m<kK<5B&m`f>&Rvq(PA4ypCZ%f;RKiFI4r{Vs`rR1Ne3fpo`;Nl
zv~crk*u45{DjZM+qi#|ZGlsjHq|GkP_9AhpDEb46JjD!$MD+%xEfGqh0Y9kW2Ohf2
zC`S9l?HX4vUFMLlxXW4E>o@V(){Kb}M4nIJ!R4(@I&lL}$hp&t!#Z05%55>fUo3$T
zC_JLb;kioiq=cfje1xTghscd(dvA~5|M<uF{HH#{mmYnTYnz*7c?MKo;nn}O5@5t}
z$U?WvgKv5>?|RRBc-ObQlVgVu;d^qP*nRT^7lm__svz`ys*`mknDInalhDF3>oig6
z48JI`IIL3h7Bq;mAe3N>5@Lx7E7FCPiXx7P2V)Aq62=$=R(XU;f^7vDrSOeGl$N5=
zB=6Sn=KRhqXfUpKSZ?i-D367cs~kEvARlRx>4aIo55_S;F!L2&Eu<(Nz5ySwrr?ke
zNT<nal0)|##My#3^>&CAJlAM)rrBcLY>|Z#+VfDls6<1a4`G0(bELJTKwVv-=$=H5
z_Ho9L4|{A5r(}hum+o?Lr#J7=RPQNUwQqR62PR$<r>@ax@MAyrWBk~U{TRRVJHNw^
z{K${+fe(D(6(@C8+Nv%0vdut9i6;!9EKosQZQ9PE=U-Drg*$LURHa-eDAN>a9YGWj
zH|h*`_E4(wowPWlQj}?iAJst$JgLxmi6%w0J&Xh2g7Q36Tm#c(mS^05u*FU7fX%@S
z%_4#H$VUm!T;1XDx4jRR7kTRJb1XGFEbs0xed=lY+Oc-*0C#shbZ-ymMJz9_(r#Bn
z@6I`*Fs7Vac9pi6CuWi;3@NjmEYD~}4eWRT*$@{u7!1Zd`OH@+?VO-iJyoj<Xa4J$
zz4)3T4aTXLRSxFdCjkf7>-ajs^h=B`=M}vYR~={w8a1Tnp=x23uy`IqIJ7kczE4?d
zR8b<NAe~H!+Z_l3YRv`m@f1DFc;}H8D>a{AeR>;~k8pW^n`tp)kW5(GE?8)G*?c<V
z%+oKhaO5yI-Fym{FJsD#qlXV6AO02^%@*6+8+`6_pJFnZu)BSYUT=d&vqlsJ)k%Cf
zA(<u=sU}ZSBqeD9o&Hs_&wd2|;R12&9s;YdQrz(O71u$t&SH!PYuFo&c=0PwQJ9Rq
zy=_M0!HpzvI2>Y~Aqc8N!Q+oV%J%jSX_g^`APggBvoW`wI>ch9RVkXySKEkP2N_qc
z6R&-VwGs$j@#n60iKKPZ>WWjh-%F!i!`LZTFI;AAtwS=N&|O(Z=LM6UUG_IN2>g)R
zLKnZ$LaGX0DV2v(KDiU@?#@^}uvoF(n7ew+ub=dNtgH?WLOR48*sX)AWnU8k+7u{1
zq8`^ttR*W7N?<x2(QY?rudcDyjqz(yHU1!B`N&aLZoQ4p-Zi?%4)Mr&pEulf2M@gE
zc6xiej3*PC-5Q5G3q14GIkHKgM?U>L8yAOsDJkh`OW?ql5+fx}N~|aNU;fvh<;kZW
zW52&gQI_O+Ns(*v(xHu*gRieoa9N5%lgy@!hB;}fu~smfC7eEehQIiOKg3N(kK>F%
z3c>r{_dY)QJHNxPf9SVf*TbDOl?X$;thj3g6vkt3bCdDL7CO}&xb-G*5qVNn!ic;?
zNlD>kMJgyQl$B6NNCCP8U9i4(gw3lP6nW)>Q*j*UPB&?WAJjnkNaH9+V}@BuxU@jH
zu!x%u(bE~lY)U#B&|X+Z;>fd<GPUHBF~^U0SZ~(Z92H!=_&m*G!nfRZFC{)tZb}*p
z3(U5!z-X6+l_TWS9>uheGa5B_KEey|bW)AWcpmk-V4-0dO*Q!}Clns>;X8Qx+<AWG
z|MxHX`0xEb`@21gB1h|L>;+fB$KBk4#9D(jj!Bs^9gnzr^(r6zPrt*FqeuB`@Bb10
z!uNeI$BrJkZYuOLO~p$>u5WZ0U~_Gq6SE1Cspe&c08dn17v~INmLjR+njuaJJYz8P
zu?dk^3fSu{a-CO5kfK&&)^5@2?c<jvX&mF}k}{5odwUdhja3Tm`^YpW9F0k(hx7y`
zl1zD&ag6IMQcqKQ9m(L$hq!-Z4=ESG3br-0uWT`DuT$Rh0O9UM{2-<tHE4hC6BJ3w
zX19rVY>D+rk6Lbcd})>amd`ARsAnl@Tq7t7X6+Up`5bcvgYF_XU%5bQGA3<yD0D$%
zMKc*M;6;|nxJRKYVc1vg*Wcih>uXBtk|g0be&aXz;0Hg*=RWs2PMkQw2R`tDSN;Nx
z4&`u+5Y<Tt0m@neRnl2npt-(|-v~$tBTAjo>^4Yt2gtl6Xv9#Da7v&=#bq<4A&6_V
zmliRmL>D<KtY+`ZX$sDvJRciI_$W+~lk821noYFy$&JR823J~?vgl+)(~mgR@bH5O
z*=Qk7th03L04s05pY9!RCH;+m%P7k@u-syrT;uj2q+tw7X;x1j0dnHNLu<%#O&CWc
zMP9i%;3$fWAPkT~P_H)_?d+0HXEa(1#MY4XHVM{O+1}b=cdtLcsb3nb{jWW2Tu0CH
z8w#-NiMy1Vg@#{M+vb&8B`V;qH4dF52<K3>7!i7eVTcSptS-^kVznVq9wOHm>9D27
zlp2X7uGh)NQ^KH9YTdOG@fYsw@{3PxB32GCPRHyFdptWyS<XsMB{NoQbw(F2@yNA3
z&cAq>Q#YTav(Ulw1D2N-xZ}1vIeO$MOA8$)qkWXjDi2#JdFr|6ke<iY%}s{G3B&z8
z+MSHZ`PD-=?-4!qX<XPOJ$!4W3Uk+4cYx&vWZG)YxpOZt9Pd+_g8ls=R}Ng_;K4&U
z2fe)>r7l=oS>x8*ZYPRr+;h*HD~4NHGMP@;-PytvqYBY7A9B`O&y}>*rlPfZX`^|K
z;VrIGUb}`A27y8fffbIEr%uui>X>QDWV}f-8WOB8lI;xWr#X3+QH&?}66|b3>Z)1F
zpdMHDc-A76M@q;r&taa7DyeR$3Y~D^$?A@fq5?@vhs9J3K2zd{5qXm0OM~ZoNKc`4
z#{OUrDWP6#6UPl6eCs=rURBAI#$bG(<9FT5<YQ;3B}2SM#BcukC;8%+p5nnb-$uLJ
z<ox*^zWl{M<no0b*5x)|IrkJV%u0q-?X$r5DU7BlD~0E8e)z)(TfXF8Vg=WQ%GJUB
zdaeDdWlGjUniO2UyvuZwF+4nFG#atDdyT*JgFnoD_uP-MnkbC;@gMtf{@`ODXE++a
zZchl)-^E|<V4bRFLoGNZxVCwXa~EHrlpcO_fwjYj*toh!)NU}?-o*<88p}%@yyX;m
zkumG_(8=sZ_1Ia5&U1<^V{N(3rO7tZ>G?so>YzvsWjX^@bu*M-wffVYJ#@Q8qtzxC
zf^s}!JnW-9pKLavv<Bfylu*=H4l|qX;MWzq!x4Y*>%YPu{_ZDu=a2sl?md1hnK11A
z$6w;srpBIyz!b<NC(AU;t95+OqO~HKS}w)M@NT=AdTeRMn#njPD07a80=2!%<5yp#
z8!hs~fBC<o)S7;Oo6XIuY;0_xq|edA$JyTA<if?X?Dcj?lay)<se;XC$&_<v&+?D{
z@jv0iAN~k`_y6&?`S$mIE1vh7w(Hjv&#x?TTYQ+c=U&7KgQA+zmBt~hp$G)&;u@2M
z21W`FZ*H=^+y<?QcLx|xk}k9fgvAB{!h$L_rBV!5SE=v!kyaBF1!=s10K6<Gt2Zg)
z8gXgK9n|)Bu(6K}Y6O6%4YH`DS<<8!o%8HsgSa%9FlMy0%;M1nR&<LmJ%5?EomizC
z2PBOHh?@?fL{2S;2(mGk)|S|f6{XM^B+9~(S;4076DAd(P+EhpH9D@*uQ%9}9`R^G
z)SfYDHaIZ3N*13aj&?||4ZkV-^`Gip@`W#afnWT^U*tm{`Vd-ce&7dwfS>u9pW*Fq
zfBUO?og;608<#%u`*_lzomXuJI9AtJxaHRS2z`eYGd3=4(e8xQ8WBOefz1kxvovZE
zL}B$nEc9G(p{Or)N%wjvCD6J=c$LRrHEbQ?H55rcq~2^{W_|kG*N8g{gsnDFB#}}w
z?vJSX5tY5G#uNz+8zJ$qwSe6({%7_EW9pGYmx7%O7g;bROQE9S1=QNh1imDjO^{OH
z2a@Sd$>Qn)lj$zYjV9I^beUJAb)Z&j(AU>6TB8C*%}KVeU14GMF1EL~NV1HVhAY0d
zz~R5PSh?dKcE0#Xwc!g@CGO4=3QgcUqD}+lOH5I<orJc?(&C&U%Ts)<5siSj-U8-c
zt<F(q1wqZL)>Np<57t%5kr(*1x*gIyC!1=fqlA^f=lf1|_~5e}c(TLE)kEw~hU`qn
z?2m@5B{P<zI<00KqdcA-4nZjvmX_GKdW8dr*64Iw96fTF;h;g9CM44-XTI`0SFdby
z+sV7Rc6q>bv_rCYo|}&xK?;LO9ljqTOoDml6NE;?v3szpQqzH$w`8l$eKMVJ@$z{F
zqkXn^b~tt79zOIR{yj$yAE(u7F`gzo_0*F@LBQ23mwDg&zJvb$6;7Wy&u}<FsQ|4D
zP98gm(^Y%(CDxsqqa!p$VJHhjS-$q-|1ID8?Y!_GKZHLTp>1U+q<lDbut`3>2EJji
zd4;8g7S@({9u(sdx+w6JhY}tJLp~khoJH9Mh-0);cq*h32kh)kz|Ln8JrAVBBEeSe
zSS@rl^Q23hGf1cKjYK1H$|LXtTxk%Vx-pkG84roVxcV8Cj|yY7)#x;(X-kCiICSz3
z9{atNV1J7P2X4a@J+O*Tef%+M?HWblkj~Mm57|5Y1gED3eOsZK1TdXHm!P_9i?sYB
zckU~@cO4bTyDVY9Hz7^VF`0}>(kcJy7yc#x-4Fj?xc9#MkRsrX_dmdc4?f7pKK_Z<
zytdcl+^0Utn~ofzY}E0@d_vYpUO0UbQEL)dkIu?bf_f7zc8OOPkWokwhWO0}28kDk
z)C8n`&{=`94qcWQQ_=|~om#*pUQ&umDKYcUO_8I#YFxmHN(rrmq8yAE&vNRWIwtVR
zvXs%FhX^Xcl`@j`wKYbm&$-85;Aq#w2|-%soPF*w;*Wlazwi(LITv@Yv9uQR-amH&
zWmBxyeByT=V>~Q*=bP_j{ZJbxVAN0eH%~ZRtmrlhTGkOukGqep(@C~iix+9#_dV=x
zU*l7M_y>IIQ-8q5#s<^L1Y-=h-*zwG{S!ZcK$7PrJG)za;gQeq_~Tz-=h{`YEvq8O
z+?wp!(`WdJzxk8=oe%sJ@Bfj%LcLbQ-axAS*~RGmSw8*nLwxJOgM??#Rh2raDkZrP
zT)O#IdP^rkCDbMZ?mBms?QR3#T4rmD)P@tRa!i{|gzqCti_{tgl+7kHX9)YFxxcps
z-$P|NuF=MY9;R+lvy5r0!(#6mbvY$nSR$>(h%mw%mc-M7Y22i6mS#3(7<i=lj8PoY
z%4fXoP(%<3+Rd1DxrZ%tMp4NqP^9$^n=4C1qahX#?Md1xSeis-j>4iUCji1cBXupZ
zFlJh-(-@5bOB(u!(H3!h2dch8uAhFHz2D2%>l?dYzpA9}pZ?Q-$_GFALC%~x!vhaI
zz|Z~M&+$V)^g}e8%|F#QKm4xuGPrVy!B?I`nrbs}(sBE(HzBhOookYGi~a2_mO3Yx
z_695;K7a`V(&3Pu%a@3oE#h_u;RgsgXRCQWRt4<t?XY(6z>T4FV+@`Y)eKt{lIoJi
z;ws9I7!LL@rl8(gBxu%Y+a>g@ay!WLgj_(F1jMCgAy9a;U7q>m1*Y8-^vaUX*)!a}
z8gOJ~rFw|_qN*H8NtR`7ULD~?gfSXxP4&RiRjc7T`%FqnlBUdNGiH-1VJHc0Rmz!U
zv#&Cu|8FIZzvtWOpF4xu*`;(9gUnh-t6S&Po%hk{gor>h-RYriM%;)g_s8g2iVS@U
zXYie-NE4Kc@q7<I2*^g0ir*##p7iI!fDB(nIIV9~WM*l`-u@UB#vE+G`)^(3BQNfA
zZU~*I!OG#I^o9dAW)rr@BkHpW{&1Jog%yq+KZO#SOD~+}`DaffeMuPjRW-;ttS)h)
z&WTgELzc0#d6xaH4Q_2IvJ0mvZa#|gL%cWuD+tpm`BR@jC?9wDPNXfccwgl)d*#YS
z#^V87*EWc2Ep~T$eC%VNAWaIs@B6<O&-dx=53$aX=LO&WzVF1ADZ~CXws&{P@|@{p
z!l9M-K(}?{kSOLV<f>{{;UtW4bEEFp7OQW42kBs+bHDd{h}0saA`ToiKf`N#q{A^~
zn$TKV!(j0|k0LLL>aEIEMR}BElYBCuoKBb+L)2XWKVY1U$tL?)t+Cow$yeaxlt55o
zi;Oa}jMEx><2sW>kXb?O8P=K^t$0i<G+wQaP9|9A@I9}x5-Uq;alH~<cvZE0aP=&z
zUSnzHAWr#oRu7UpLAraDd*1vGZocU#jaI~{KOn$Sl!o45z>_aN$ED#uJ!j{m?emJ(
z%iUUdEtsI{UtYJdo6QQYUcGjs+WNo$;=kf2|K|U|t*34yh(i9{d*90^KKaQTD$Caz
z+n@hKuAV%_(QkPdHmqz2lhQHH>nH(ID%xuY+1VW+@hNhP*KFYhK~)*1#Pe!|tuD?v
zX2T((DDb4j*%DDE)PhPB01QF%zI2_(C`5IL!wJx~0%}VM_PWiQ!;=C;NirSdONB-=
z%VyM6lPK^}%F^vDG8rXopS!@)N<fUn3u;ha<Lr}<(cjr-m>Jd%+(f&v%!S9FVl*7H
zw=tkUnDYEH=jrXWu*Pxn<Utk#O%XMzTgA<}WG!;EPj*>YIEWtw%!-14^Q-@wU;2gr
z9b4o;RVZjRTO2)hl-uvToiwRjPmUfv$vt=9&wl?e^VCz1@gG0@Yh1p30i!KemxLl<
znoRkJ|M+Lw+}h-S`n!LJS}lHE*7VgxKAADR_#%(qem8HMP6@X62t=h~&jZO`w~IfX
zV!eXJ?F~ehQR|P<kx#zb!U{=!G$h>HW!70BZ#JvS>zv$T;~J40u`rx+xznW#L%bwG
zX^rtiidsyRmjnoU2anP0_h{~3BWZU?TOB5gi<NZ8Ivkp{GR4j+!Qd>2ct*i-yTiBk
zL#~;l^!t6*2Yoh=FK~KomAJ^UuEcl>p)HG(gg2er=f#7Esm)TXaLBo3rKfe(f|Y_I
zj=AINCc}Oo6UAJPV~&?ICbI<^&H6V*+w!WCx}W>GpW}yr_=ow>hd#ueci#D$zqvYi
zm|MT&eSB$S6VaO>ghxGuc9f#a9h?XIS1!?vLxi>1QOdr#im{erJjL1qGn`=3g0R!A
zfS^iNb9rHibArhvp|jLR%4#@VNC^^xT1+}hDa(Sm)y4NchWk4tWsh)SkyfKcE)_{S
zLFjTW)h{S>m}QnMpAh$_)S@+{bgV76IeE(gd_Te$3gH=wv|y{Z$KK8^qoL1zZ#u+i
zx=k%=1C=`QI#)^vfwLBE4AbF=*-npUr%AmTVvIyP!5`fo{#%IX<SpFvuJ`cbZ~Z1J
zs}gr3EO*>-3kz`vk(W4|aq-M07MB||+ch!|f<}z=JY=(mofFzkSt6AnZZ;Tf?;)i`
zYmJ9PDv8buR2)|GlxYQ;T|Tr*dv%FIYqTo}2m5^Y@tDtEP58=W!YrGz*zK~qw#@!u
z$e=$!&NAxk(F=T3qmE5W7FvtMQG?Ptio)Q^0%<h9DY0q7sJF$H(SUjwVkeR^)RcKf
zBqfr{3J&40*^KcQ{uA~0EMl8YECy#Rn3C~u$k~f$+1cC1S<4&ldw^snIdbF}dwctQ
z?sK1~EDQX=C(lcawHRH|`Rlbhi;GKSc}@^i97hWk*UWKao~|s*8}@MUC-JjhdDw1;
zWAFYp`kU9-{PLHPPO`RIXYoKAq$BAMiQ8>d6rpuNy}Lkj*{6~hTZ}1@IwTGwvS`9|
zv`5z8!;f2R_a+!4P(oI!X(bV!0wc(?lEI+M3)ebqM@P7pNe<m~0Er<VO?mq2MUK}d
z+}gd!V%tCvRDQ-%U|wd0SGgTTIA=(<uHY*Vee?ZD-%;yy7#BXZVvpiWzs~NUWIN4}
zd5Osqwu2^54^y&fpQ~jxi^n(c@O2%j;T)8O<=SR%&bYPw(l39IpZ>|e&C2RJZ+pvI
zX*3&*Mw8d)8}IFL{ttearQ=7b-EuS1WQ?YoK`%j%3fc!&aJG7ODWT>l5RYs&qAUum
z(*!|E+*%}BSi(s~G3+BsO=%s87T^V-1kM^LHO^MTDk<z7jiixI;+#N~3WTgo)J1?P
z3%q)r!fAA&vD)ByA)W3L)(IxlA<1OO7oK_otVEX%ClzVGPw)BX$&cU6*~?oDPHDdM
z_|rW0$X8fjU#HQGc=F3Hvb(cOqZ#wJe)=zQ^zb6B^)?R$5x2)4!gEMbqcjz==rf=A
z13vxPkFl`4z@Y<2xa;=&xc8nnaQvp5IC0`6D=SMVr7A{iVHpjlEHAHc^zcdE_`uuv
z<bVEMKK#+&WM_Me)M{$&CX-~!FZ|;FL8DRQr~cO8A_#&%(}bba7F(3;ZCvKe{daTc
zr@n|BWz{G|Eb;PPyjmUMGz(jsTx~T_S%I2PS;#W{ydaoP5muu{Gm_<HidGw|6jn)4
z9%-XVdt;Z*BxM>`uxDj7M#eY^d80-&8RB`0$-)Xn;L{uphzm>6=`iRn5++mpq@X`{
zM)FJnsiomGGM~U;pD<X(*9F%OEOYjz!x$MbsJB2Fv``$ne31vQ^@wzhi-#hX_OCH$
zuOh5P&3{R2t1f_4G^}A|XOF1}m~|TTf{Y;9B#5JzHP!y?_Ul)b)cw|P{TAc#_^T4S
zU;gD^=I{NzzsGDg<9oj6d-xas;$Ku;pH~pl7c_3Ul_T$X7tjBvk05gBwj&mnYIt3b
z*<PPwI%Rov6)QA;C@E)CilRUu2&9iGQnEoGseD8|#wdyB2Y6mU=AgebrMuk03RPv-
z&Y^^d4Zw@yDp_<Tq8`)n4wCQgQx5h~okfCL9V-OB*T>k(Yt-nBJS!>EoOX8+FAgZ2
zWU#oxB@=T*Sprc~I)zX$9rkgV=FZ!WfzlMErPGcmixOLw6-Q6ZZJ@Nl*{YJDO+lI@
z__HZlk>Pm(PRKdz;lGW94nWjG4!rGQt~~ZQ$(i#Yed>W@wFBAs8Y=Xe?QSyK+hu*F
z#iT!CVSSZkJi!$uQ5@i+2xlB7D{#i)g%R!L1)R3%vcPvyB|6D6f?BhhtwK=FGNcy~
z)*1v#pvw}^i||xTx7O#5spKPPx9RQeg6Gj%USehSB;_o{&1UTE?vid>nh1iu9lTmh
z76j}k1-3{DO-@$Ape)$dmToQNc+enD_UYDJIFn<whtnm(`k1`(_Z{t@$1JQ<f7_o2
zp{toXtJ&}Evb(p#Y&zk=``^kN?tTD6KzFgt(@#CglTSRx<trC?`@;|MsZW2lnn%**
zS8?Ym@k2F3XPkZ69Gyj%7NZ@`3aoiaK=E2b3Ww}=xcz(nJj3B0>GKyja^fa(p-9FF
zX)>d|v_kF#ei$O72r(aow{uZ|(*+Je(5%z)9FzS%#bks)QkDgcIII+rQlhOTpH8^E
zeVi{3kI=mPW}1x-m!CV!#{L8^6db$#2(8<Wlk5z5VWY)~@==yo{P~PqfzcXkEV?wL
znZ|el3PCwb5k*l|kBq@@G_k5nnQXB%zJ!}hst3F0QFdBf8RsZju$L5Mrc!NcUi+Nl
z%j1Spdd@byj*YPvw07)l4+!Fbr=R~azyFEf<%fRYhdFxqD94T+<LRfLeO-PAYe4Vz
zc=5v@ruA3e&-BPL=g(aw8RYnJM7X%bY%r`K!$HKf-($L09aw~rNadsbge*^~bry+Q
z?aJXKo3Yky&}em$N}@c8F&a}82;&eaoFCx(iby%4s6coM<s6yOq_z^1U<_%Rq5=;}
zgEa;zJvyBwiXx}fIfX8m>}`W__^R5FrPZ8&?nzcpJwVo2L{D;l@GpNGci(+GkALY5
z$t+`OLGk7{-@?5QoIo{KdE>or=g2~h+kHhW1ceZk*5Wy^Mw3qy-v5_>oZD`_lbep8
zq87!3VL%*5grP?i$Qwx=!qM$U<hf=vN{M2hDE=Of-*k-M_>F(dSDt%@@t{u-)tOAE
z{ICD=pL5$Cck;gP_>QVd`DYZ>{o=6Wlq<&!-FH4r=l4I3LQn*fX(vK=6!pC!xhWYh
zE+V9*^acK8Oi)z!cd2}2Hlx-9kQDV8r6k%9D8iUhDNbJ9=gW5-B0{3d60E~`9z-F!
zjKP}9mb2Yr7{-Kq`vil1vi1Uby@8OD>2QRvH3o;&8DU;hOG{o{XtU@$qNqW$=5uQi
za3Qb^g~9g})>kwqW2CWs`j$0hnWL=0Dn*bNm6aV9Wi+J#>j#8XqRp^OaZ|__dj!!j
zI{pe$d7w4_)h2a6`?EjGJKy<EzU#Zb3xH>zeU=~n(I4gg?|(nH-F6#4_j5nTfddEl
z`+xuMzoOR@2wdoM>^<*d>*8tBFF(i0TMkg`2AHBC?e&S<Exahg*pkQ($#YE@H&D_;
zI7tEIvq@#4BbIPcgmDg}Vrm2SlX9-Ks?=oGR<vdzC0-O`%j)a4Fi7bWtuEtw`y{=6
zYKsd5agADMfkAHvW5H93D3&xA7rB}TY;0a*@#a%pt2KB6f!^EV)+G-=t7C+s)X-h&
z&{^$pb#Jp`(>W-L0)?AX;^&@^&Y^^4G#n8IA)yNKCfn$p=ZFIT&wP&J^$00|c8i<8
z^E(;r`~qrF&~1ja8Un9bQ05bQS1!?Q)e**$?G2duK7JUIrWw~R?9*IaBJOmNp^vi_
z(Y&_MA?xj<1e8S%p2Q{^Y}KOk`~VM0Iho*j66-5M6n<RCUlMr2aAa9>aHY#fp4(x2
zkTTlXqv!1-;~LFwhwe&)Mrm=gDJ~mfrX#MEnzOT<e9q}|NIYd(_7#VeqBiSuU?E~@
zsYzL+h$tqTObNrFQe*3qJf9Js`y$~zZ=<YrD`l`XOosc6_xHK)_WO9-8{b3}O2)Gc
zX9|*J%*xUNM-H#DwtkS$efA4g<xM+!{T|cV6yNtNNGy=$Ii)VE>vcV;GuJT+bLQjx
z&tGc@z$iierjy+C=f0K6=p(rHK|X(`pjOT}(m95|x<bBll`!(Kb2O0<RL%s_p+%(x
zG)00F9#NxCIvG$4Blfjnxup<J;Y3L`Hk`Y9is$`Xso#H+GcR7@+U_}|l=wl!$-CA$
zaBPV(%Q!zC^5Wz$()NfO=U7~*Axv4B)0=s_#%fd?0Roq0i0KqvCMZ?M2}v9`*uR#u
z)~U0y)WMgKvci|H_Au=Qyle+OSGK<%!jrGMwgQx@#39$)k(E=(OY4^9hMjBsC@J~K
zfBKI+^rp9Q<k$&ry6O1qmee^RP}<Ue`f0ALe~Q*$_%8DOUG6@jm^q0Nn%(|B^;%3;
z6u6?Sgf4Rj5h*M}`jmObbZ-~eT_$Wb(SfAdj6fI)1M`~!XACF^f{;dQ%%SB8N8<s9
zYK9^!5PnGRcQ`wU+32^~Pa5cShLi2ebwz8Gr|^S-G)pQ`hAD7H(^)(~IZGJ!ui?s^
z?MtUQAw%@~5q5T-C9K6PEZ6zg?|c*c+kKkNkj_#IFKjVeyn}m|*Er@&Ld3jcS`hd?
zf$xFu@k2lG0~qJcf!qS5Ln=|Z>!=C`Qu+D<IHUy6_lRl@8Vw&m0An<N?XUg!eD3pq
z$VWf=VX~sY4?Tv%A^+^3{l9t3gAcN{vi!QN?JEhZEvD4iz@YaBJl%n}zU4j|pZYTU
z%N>fgkI;r-Iwh~wAdc{?A@@CGVJM<HTKQNf38W%S#?*H$W@#B)uj4z5_5#vIi^b=A
zh`S_K$tsD`l?v8MNg3BrWmVJ>r3LAeudQH*Lu$J{vV{&=TxYzp#+_SN7+Fi<d1L~v
zbUW;MEgn91g~j8?adC(HgvV0v8lUetGTXqZkhY2#wFDb09S-z&(7uPTj<6`uP+?Ze
zf)toW5u2XR8~OvjjNsbR3R$+p(e(y0J4(G)qj-g$t*=+Ss-*6zr=H?xe&%OxeEw^{
z_G`T9O>g4ofBxqIIDGgpfA8=8y;q#n;Vf1O!sRuNzUN!$UD-jjR(N*9(NPBSg3jV1
zM*0Zts4Xl}UtX!^DV)XV5<jb>`n%+l5i=<Wx{HV~An+9RC}g*n%smlBg-UQQwVZ_S
zRU4eK2wY_}Eug--j2Mr|#uK82C0gw+Mwi6(Iz9?36e{xAJa>gr39V6&!$)prAR>PM
zD^C!sB~C1Olto5fT8<w$#9%g|Fq%kJM&MFOB(9R~Q|)lpI((%l%7Sz{!?aoiI$-zw
zMfS!5ZRd7V)x~=?OZUwS=P;gR@eTK}{PaCcKmIvR-g1EUY8`|l*_mLqW^rW^D-2Pf
znC<S87bULL2w|DbGHk91yIrI=pTqEcjPw|e_gM&|3Rdb|g?0!+RI@=g7!w3O)7?$N
zcAG*)<fg<~s6|a?I_H)}kN4lV!kOKKt^SPOaK_Gf#O2AD-6$Z4W8$!l>ntG!WNA*U
z1_VWAOmCg#L?F1c71Qw@hq@65>lQoN!D}3T$qKG0DUHS&gEA!zqj9^J;aH~vGh0m<
z1RPph<NkZ@M_9vbw9o0&=Xv^>r+MnjU!WE@`H7$S2`*f^RC$g%hc=cp%Sp2gDJ5x|
zQ5Kqu7cOyNb)6s#fC>R}Lvc7qfhbi~uV0Tfc0!`On9lK&ocOkPu)eg2+}VIZ#>(n4
zzNhe{BZ{L6E<8^rt5)J1U11GqOQAD@z(e^Vkuj9U5%~eymK0^dwVg$t)<<aEeTwIw
zdyc#AJi<GVAI9^13I~3@OK+pkq4h2&Z#u-zr74%M9$>NmB4IGY_g&StHUg&&WsxGI
zxUz;TN@PAmmj)GToRz33Vl>ehCs=897?d%Op2^q_n`Ah4H+SZ!q1QJA5Ck5zI3o1J
z`2i5}A}5<;GO7gGBbk*9hhu~>eD-sn;s*~M;n>k*uj3CR5YAF&DZ9@+&F1P7gG-kQ
zmsSWDJDl4%L$Wg<I&w2vQdA7VIkv|*3s$0p!1EMR2Bgyw()aMA7KJWo`Z1&F3~Q^l
zt1mr*M!?$A1`jN6;4WQe@BCTBEjQ8I*rCh~nJ&5SjrVZ;;2KX1Zew#l#^iZb9kC86
z6>83eth%>3Rw!zRPGXW7YV#cEoZ;>cp|pfchj{vtKjh4XOC*zo)zuY*6!gbA8`~Lo
zJ@f{S9X!QxA0jaqtO;mD*IS{LJz5lpbFbGq@LN;`jF;+aFR!pw$)0nSRcxafas0$V
z>WzqcBcRi1v9hwxum1aAVU|pgQu6rYkMnPS>6iK2Km9+<$M3)K6}xJNxN-xMQ%f=3
z<;rdMaQx&xwPwH@_I!r<ge=u0%?3pj6DA2sr^{?<r2-K<i!dc+-b9Z=qQO4(-X58A
z5Jza=CuzilTH_gm6A~+`izDVXWzMUFF3wn_)fGy~S+Z^y7bnD%DX9t=*V~+4Tp{;;
zGA}?_fvyFtU+pv7@6%peAg|Tf%mrq5$jvFdu<lWMf`fU<lk4lmx}ctC%<6TN(O9P{
zCakQ~@c>3DE_)ucqGWA6rPpjRHW3G|ohMz0F>ZdyoBu!lEaFuqb%Vix)z#G-pMUz(
zpXS@Y{o8MR{>C@Hk*im)zVa7z0^t<W3uqra%JFY~n3bg^lFdyfJ3WrB+=Rc<A>F=0
zEet9$tB~L<#)4I{QaUcKl6V2LaUTSD^%lK+L_Vz`fWm;3-dCyStgs*?Qdah=Li&({
zGaBO{s5KDeSX~f?A+7czCeLxc#CeLrXav)196Wq0W#(8~=<=TL|BF0&>M+m${%;eH
zCzOK$t9L$tYuDJ{ID;QH@WVPv1yt=511gx~86>512%*6!l3|8fg_?|LF5S%K=O3Ai
zD(0=)>%ivggYYj?#k%=G{PnMVbt3p@oD4cZD1us@gAcuxtrsrfcbC~3OIpmx`aPP9
z3#hn;(K*3<9!|_#T+)HfQqulDLizY{g!B}`lXyWiXM7o=pu&e3Yb$?B=~rAkC5c6w
z!QLKmcafmc!j+mb&2VLkGm^zxz#Yp0w;pKW{1`pX`OLFt3GaW1ktg{4GfxtlfP8<5
zC{qqPL0Xi|q~QL8E#7y$#*(uX=@e@-Y$=EuZQ>}z^Qx8~+G3m`o6HD|q85a>{yyTU
zLs*GJv9@-A!-o%W?(B0|A#uQHGNBgx+;!(ISR)zs2joeDnsergqTu-P6Xb>A^y%lB
zOsAC%QC>2cOzDJ+IH|D0;^q&S8{lE)Q192g#sX;-NP+M}j^1)Ji#R+F>W9{7HWZ`H
z7xAUQ_x%b?Y7EYTb(JdJxeClI!T{k6*)$~#V)EG-p?pTEp_$DnrcEB%J;eEzp|<%X
zL06KKpc&KbtaIDR<1DT&GuU0@`Df0vwb$p)`%m!H=7hb0$5JbyUiT=A5@7|_mK8)c
zsO$#yXhbA6HrMDVM;n6|#LN=MU^HWO@eo&fITy2pgC`F2+0Q<KF&6(D0EdmDfW^fH
zn$0HlxPk8l<ay3$JYqZ@k)~;-7@yCUj)r4ArTNU~KE-!@=bvYNZSBSk-fIrwutI||
z=yb}9k9>)~@(_NLGTUL?n-JDo<l0tkP~44*aP{x0HfJFbQAi*J`7FU#5mP5wRj@TI
zal#^;BJ4;`9z4gL@*Me-r&(I;@W6W?B#L})xpkFv3ZMJfb3FTdU&3Fiaq12Gh?RG<
zbuAz(vZ~ABc@=rUS45$YFgXrEwAdwEXjdv|A;`ufT4ByKSR{Dx2U%`{!(hCM#!$~E
z+_uo5v3!h^aZKpI3X7z&O_PX995Z);csV<+a@2YS?+xK9FJ0*fBf)`#i$qaCv)Lk!
zoA^P%FaP@wGMi2*OU=Lg^?%R%|LTvhy0VP;GkT_%MTvD5<0^&uwPc?|?|mo3tN>@Z
z^J<UFtrgC#wup*pb$xsfrxf6jhRSZu_n5X9$%2^n{tmU?4taZlMo@BmZwFs!TGN7w
z^gvd7zH|a)Rn@mn%@Ia^Wkpvyly%r3V%qeH@{GJyr@yd5JedFz=?Ow>xud^H)bOeq
z&Oy#+FYS^|4R`2(`&6GR4Ua+CVY_>r<Fm^M6*BfCOc2hY%ND0Xd~LuATo@5}kP5+_
zqk=2V8NKyYHk!+Hyr7zwM$F$M*Vp>1KJe0>UH$&6_iT8zc;Z);)UB?ra`ECt4j(>D
zQ4~D!#1s63fA9}(e0?&R;QRh7em&>rrq4>?f{>f;yn}9;Q_l8L-XhB@E&NCl`ifdD
z#u$T=bGiF`zRZ})Mj>jqaA?xW1ZM<0{RC?SSkT5)5ZfTYsmfGcYK6s8Pz{qS9QYpA
zt9&(uvG`F;u1k~^)S?=R*5q2_gkWoDi*Co~!2Ne{W@|>^`*fOZzWe|5SNU{N@VO8E
za}IZwS$oI#^2}~Q>*PDBDe!EDET+hGhD%4-q>r`^AsrY^k{PB`NgPPLutGf~y$yD*
zUPV$F|JHn;pcd0v=(4oDj0CbQAxl$cNkV@-W;~fy>hK#6^lwgO&skS(_<^Fnw8rsw
zKg2?-&FPDmsFfMZvc~FSmr`4>hWbJWzup3Cu%*Eknsj0)rejnP;;99+^YBFFgX)xI
zyEo=|J4VdKFqOv+gteG#HX&-X=m^PRe+MTVwN{5%wwUhCDkl;n5mr+(IrVmvtR3^v
z!o93~%X{z--^$YO{5F{`Ik>yUwNHPBYq`T)Xmj_W1rGQH#omOznBplx97Qy1H9Fk|
zBELo$h1G13t6J^JbV8_ngw(j*^N9RTtZadH^!Iu^_Q)3~(u9TO1Jv6cHa54|xVD8K
z2CS{EvAK1Hs~gvl%40U0@x+r)5l1cd`vZ1&cX{U7XXtd7`0nrcPJ$q~VFy;>lCBTg
zztoa_JvMyj98wAp7KEY^*72r88uc2nP0+;z-{^{^mn6tAKn5Ys6O_XVCe7w(3xl<q
zP<hBOL_13uhD5Q))Jm)|>`oWC7PdHWXPcF!6Ig91(-EEaF&=&7aX$a~O>VpWW-vKL
zVj1p~?3X*t)RZg@D2o}TugQ`G3`kGX=`L4_*s`EZX83I%n`h`UCo42U`4pu<8-em`
zTpi6=yyY;b&s?FHrg%z`y^7hyKMExJTA`$*)oio6vd+rND(z+)DHOxu5FsR5SL#V^
zTqXVxhBUQ|$C3*dF0iw+O}EpnRQ9jK+~jqSZk#|E&1gJi6oiale2&s@plwX+@M2}B
zWi(D#_!uDto|2&ERWm7Yx<tlxoQ1qh@uL_|fsls43-LXV18ck7DK9d4{8{d}`wn*Z
z$9(w<8vuj~E<Sr6tt5B>M>~t0TFJ5Q64S{oWP2q}1$as!lt8G;X^6aXv`|rmA4Mn?
zVf-03O`sg$*ALTc97eXA;8(uAzDqdX9dmf8#Zsq5M^)a#p7g%vLcS&#u&DTzN`mlU
zaj8iV9K-j53dy4!|K{KPBGx5bxNw0le({TZ_xrx<PyhX1WfXZauc}uaBeOr^>YWd;
z<9p2ZcbE-ZTv@C$Zb!sZOS_mNsrEG|Ds+=Tfioy;u|ddad70*Zk6<uBMRi*J87>Z3
z7)%&1x3He7IF!=O=YxF$N|f*_Hmy<!;Uk?vl?9~|Xaq9XWI>3FEm4-BoM0hKS+^<I
z78W=!>l{;xz*oGuxlMo4<hu*WJN207mu@58iOI}S3aLqF5*f8AixJKTVP*&^2(_h9
zk~QgZBD2&JSj|c<NP(L6kTV}?44!aAp2U+LG7Rwi2&t&LHcH}2L6YZ8(}c1xFY!zN
zlzaB8O6tDjJHCUz^LPFZ|Ky+i6Mp-*f16sZ#={RkeB<lSJo5}EPMmnfujl3)9nK(-
zG~xygA!*)nCqX&GlNrOSFW^Z>5QaCHMZgUNh!YZL3ye+);yOAnD3dW#D$25;7Wzof
z!wQX1s!~4bj51S{iN{Pwq*~yKoFLZJeMuxFhKh&fvI26A7y3v~QD!;GcuG;EwA)R(
z2M^K=3dTiFnI%LEtK9mQck<a^`rjB@!`=&*`OIf`a9)kD7V(bv+(T<2<oQb%sC$Q4
zZ2KHs3%EJIfIEGTG*t{{3KjT-wUEq~%q~4kTS@-hyWY*)9)5^7KJWmCj~!)UX_0!p
zfxwaH8AYCxWf>b=SNZ(sKhLK=^%<Uc;%WB#g9=3ZY8Lx-ifWA|2#520Zo2(8T4_Ra
zQYYIPuzcVY@luoI(wXWs9fm};TGi4s8e5jgTF9(-jlAC@@H|wri2+a^{aiAd7Tkou
z3CVTWY)^UE7-TlZlm$_v%_5Fre-G&dezQ%xvq(Ok;Dn_#8QK=q@)`|Uhe$Da<_k2V
z7ID4KWWSHMvcT%itDF!4O#x&>?99+;)d*@0>h&1rC-qjdjZ+m9&FK=5L|%Y48Bq}9
z6%Lb)$gVw)UAYTk6>%VtR&(x!r!g;_rq)^^u6IbMITtTpVDst*t#`K3MS(vzhaC3%
zY+k)Wtx;!nWsO>Vo?dT{$#g=N=j6IXDuw5H%+I$~OYFMC#Otu(JLjrauGT0EzIFt;
zrc8T;v4<Z=6wcv{rRa~VIo4*4u-!%%IX27Z-3f~{kQFJu?;*W_IBpP+`s8@HB4k$_
z;?{THPS|W?wnt#$bDzJ!mmk@{ua_ia%dB^SH@*3`%2mQl*c+4xFUOc1XAMbOk|dQ~
zQ@hcjwX#l$#ic1u7nP8z$T3BUE=qi95Y7>JG0rM<98+4y#j9I1JjqmBvi|})M2IR;
z6M(g@n*03|qF%4FzIK4ahmNqmwoV*ZV;kT|k|~qP2q`NoFzYIaFGesOYx==}ot<km
zo2{z#|9T~L#(?rL(ouR2S}kf}hj9+d_>`W9?@Qu#2c2b@G(~8GYS&R=gql~=5dxE?
z1dV1T=&=r6WV9p6IIQD0M?AE;#nDl~<j{SHu}|+}kB!TU-tGp;bOJWv_^qp?NzS=5
zmuYkb2k&grTe^cJty7i)<tb2rlqfGkCs~z<g#o7C!1E(a=_9bDNy5^Bke%U_o!$tI
z!ZUl^xl(7rFX;`&taY0eAbmczQ(c&^`tYZUKX$Nk*Y`K5*CUP`JAgKZcR&2CY;JDy
zdmsB9vLxjrANeTX`R(71=e-Wq_E!+P)Zm!Ub1JMUm@3CO2$)7quHAV*MWmq|)9hVg
z5cnA7RjtbFYI`9tN}<q5QzDIJ*6z~S-$Q28YL6DCs^%*NQsRUHZ3)T@KlO>SfQAln
zS|j|NGL#HG(0E8;D%??;W2H~&`^1GHC<{*X_t>n3Jllxbo(Wc#>Kr}j@$B=GX}`zw
z%_Wv&%aY6qolm|W)2Ov*mnrdLfs>02u3EHhlFAZB0qIR8o>9~j#Viwe!h=8|`#UU+
zMl80Q6(Cu9I3$JEWO+$Z<VdLqqX^$qpz3tGZ9Hkn(u6cCZg`Hr+I{;~C3QdZGe5)k
zec$)-j(5C+Mx((m{n9TH1i_83|KczHBH#9H-}cI1(7DRE{rYB5I>dBFnh)?pfmQ{+
z)!6ym0fCRx2B#gRF)(MoIb)&F<kk|@BFZwM9w>TRv(&0p_BNhRX*D{TG90gPVRwbA
zWgWM0knR4ARxA-$dvx4BcdboWZ4Zb&N7SgHlgb)IN`-ZXUT+KE^Qgs5YPBYBd+&p2
zTU5Yv9IHo<v3%?pE7^>GF=CYLbKv%61d^Z`(dsl9jHWz(wg*nIuvp{$ZvZd1^rs%v
zT+&q`1L1Sc_=xpo-v2FcqJHo=VJ#+V#nr5VP)OxfX5w+<#$`Qx^eAtB(?k4Uf8r-O
zcm5n7{_Wr9*MIHTID78=4cYxGs~5Z;;ZW+jH?E=<)Mz3IPu_}KStODfrWoNVL9G^7
zGeWlF3ppwlkRL^~7uT5fc9`~i)RaQH7~hWxqnK<kuG|KMydeNk08-*dwd#H1N>sB!
zO9+bnAzDak^(G#I>1>F}GlZv@&BioJLnrca{XL$!e1UT16wR<s@A)$<=NT&<K`cB}
zXN`6%CYcS1d`Ve4(zL7`38X?hho=;U&GG%JTHz@*Z}%%?p_yJdP0&4wuyvLfyWD%<
zTln-Bp2C<C5gJxjmht>LSvsTN-{-L}ehDdkiY&vEf=;U%6Ub9Tr_tcu?|283Swfbj
zY;0|imzw_mK1ErwytqiG-6jqrq?AbU`c!e;{NB)I#S+68sAdBjIO5edX0nS41H5hv
zk)`N7qnsvK4RNc^EKMoX1ca*$)P*F^5`3?Ewv_3JZf6OT2V6;F&Oh}yt^5FQzRjby
zc#ykqU1xtdrED0=tROE6CdrV_Qpn}ce3seh0IQUQqN)TAdIR!Y)2TODUS7kCVhUYi
zl8lDu(AuKP3|l&iJSXrhQBckG?M`N}(&EAs&l4E1(lNwVAmp#pCJ8)`hu-=&I;}SS
zexGv}FR-<<{dGkuQV5!Hou!2(R+d&+TU{fLW2Vz7)9JKI(3PsxqsAa5I3sXYptWO`
zrfh6prPE%a7Dg0B@tW`S>!S_It6E~lg4P92=D4K5X#WQMn_5V`*oN_pd^9GAVu(W|
z5GWtvD~uBrds2C%voU#ArzR~<Io9fXERIKPJ~LoAzFIx&MNBeFxbL33xN!DG&OG1a
zsYlOo`#mQ)d18%?OMBdN`x@;=O1<7-IxVZNk+ew9LwEw6CzZ8Tt%mk$s6gOlekIgN
zQ#vgVPr%-|Pshu+|G4601k)?q-0{{o5&H7ARpS1HcsZ-G`UoIUo}=CfSU<Rq&J92K
z1Am#*XJ6#SGtcwrV~;Wz3|U@Y{D%BYe>BRxtUPPAL3pxaN2UqZtKk<JAz4LmkQtd$
zU`mVt7y1=c-dH?sz#5cuU>q_pP*RXL>!`_ytF4fe=?Kr37!_3bu`<J-7OajBaA>DX
zt<<>s1U+5{H6+eMhH=Km`ix#<L>?*fT7*}Yn9xUi9zj0g$Yj8|T8(KCvDZ)egL8d~
zS)Cy4P^2?X4>b409{sw&CMm;wM$k%;P7&D*X=*f7#B3513Pp>8yf<a0Y9whwQV13b
z!J)#@j20P&EpB~qk-=zKnImYyU^roaI7I6LUFMWUI={NrelDaS2z*+d1s0Vg8IKrE
zCYV=|+`X!#ZfR+W&wS=Hj7B3GjYf4ke_4F!Lm#5iXndW2{l~*Ge^?k-fh9eG3Ic=)
zaDl}0J+Ky2Xhz$+g!LMJy9J>Sv&t6A&D(7v2yw|2XB|Nhqw^f=1g)q7zZy|c4g^<r
zPV&_FF!|x*EFI~R=N9X((m1qAyK^_PDEahbk8$(t8h0*TqSdN{U*Md>_dSd;<XJ|u
z*{)QU%BR*@L@0rAm1T;HVh-JXAM(*ZWPPDUciHl;cf5uB9=Mh5?Oky2p7*|uz5B)_
zvy?|3d5&Qr36m+A7WC?ZOD5s77dElYLk#r@ec`tW!ic5iHI5xU!RqQkR#(<oURlNu
zs#d_!V<+gg7ZJ)wD2Qtwciwe3x7~UxKm1qz62JV*zs%47A0K4%+78bB*(7z&RbW!1
z%c|O;9b!7g*@RZBhEP7^tDA&M5JnMt-Wrrrf>4!W(-=HIpuV)icyEh*&?7Jlq@<X1
z47$MCYF_8MLimOlAPg~ORl&2`5X22c7n1Q1B|Q9EOl%Fl+{Zab=t-<88H_XXaz=fS
zG87#)+%_v47g=Agadfo;Ov@;s%nCdeAO!e<V!y9hUhgoRjcK;JIA<w~oG=U#QV>Nk
zMKUAJGMZ5XGuVf0gbQNQyx<E@KE>6+3}+p^^eX+Npw(=kLLU`Mu3WuLQ8=b0SS(-o
z(&M~v_8f67rWVK4<2q3k5d;Cfy?rumxODjv=gyu-dWw22rrT-L?RHq~bf~|I`}I!_
zfwLOxG(rjLjRuJ=$Y&EmT_BZUmKHSP7*podIxUh}Mqn*rqlPZBs@h!S_@0OFd*ri(
z!Ke>9q3%hn5oE?E^kHXn$RB(;;Fh~19=@-|Y^FH)@Iy={mZQhkk*egQAN?3JlM^>m
zqCHDJ+#}5l`@KFsinaAFje48H8S<jUPKLC6k7Sz6Q3`W0mS?HOP>W)=vYb6_*xl=6
zWk_x{8DGB&E`;Fr+wb6)e&ttLSX>}YQ~uqr{t7?!fe(D$q|Q@{R;xw3-KNv&&}y}A
zw9<nhAP54=xqoUH1QbPqu~ns0SivmIdFq*GiGn6?fBW0{!lRF}x8Hjm*0}P~RwyVe
zNv}`baumH3*;z;D1;S2oQ3$2OZ`6<q(#ec^t%etbb4a+s^L(N>CQTB`JR>Ws#9D;~
z%k~OqK6RQr8KV|syddOl58T3uoA;;(4c`8qH}c$bOKfaTKpC1#&{|$1)-B?i!}EQ7
zUlP@8I0eE=#*+b=&XLL|Xf(+ZiwFWdKg3fKT_)6GL8o2kTJI|B3n6zbcG-O9G7rAt
zcD%^14lLp|wpd?&HQh@ejhT-UiR+%wFB5n;N4FDEtS=*!;(Oos{rrFaw@d8o?Q!MG
z)jzYO&XfjgDsh-}uKK}CLtqQiFh&*{oQDp50wMA8lJ>Q2@>-3u)xw1V%u-ZYRlZRQ
z7e&Zqh5_;*rtb=TAz|(@>*<_e5^&4TDUR<onJx4fbT<)E05-yQ3bgCtPc?UJ9cEPT
zaC-SFgGL?H2;m%g5a5j+EnTAL;tLTA`j>WbS`dUGnexbzoJ+1nvz#HKlzMBCT<owt
z-XTg4Bm4qW_Zf~xgno=)F4D{_zB3dKdUe5YGUnKs%jo7|dc_E*Q<mEiwOR|MLabGc
zrYT91P?m-?8FA^-RYv0x##FR|%#`G5h93lUx?K)*x(xgKj3?7qv`&92o3M7f{Wbr5
zVPWA_z0P$D4Qn;VX@oCmEH03YAYQ3s1{KFlwHk;tqv(&wvy5<|OFErGUSO<6TZ6z6
z1tH~ZhUfVRGKBP*W{$K>F<P*>v&3WR5f&f1pUsUuzVzAWnUy8Rz?I8m;)dnHx8K6y
z_dHDh>{VXebKD|MQ*YI<x~!n3+RW#a=G-VhL`ubYu#2k2v=>T*@;QF%-MskukI)+L
z;WwN7=C6H%&8;o&yyqx+mec8UId<v@qrsF%zw|sigMz$2<#~~FrZh|k9_h5lbh?Ez
zhRBor&|muz{>D#!faTSL1c5*29Knk(KEwa=&;AjOMvJ@ex{rq+dKbr!--PFRIH6EM
z#F1mi`RSkj0PlYHyZM=a^s{{O4?m68-*|_L>nBfZ4AyDzKzTlHR$u@XXaZ|7fRv(Y
zo6hH}U_Lu31g0!OSeo5M#=RaT6E=q_rl@#hPCKOX5S|Z0Rwhlt&V4~WgmwsBsUE$!
zM#I5mI3$*mAg&WNTa1VM^Ytyr3qzJ<H2i=i&#*a~u(H0)soM^rghE(=5^2Hdf;646
zyOWTnio?h2><@Z$7M8KrP?kB?YCND`Z;%Z}WYZ~)Ru|7?=t&Q`ut<M@$Wzb0K<TKM
zPI<|6Yh&KJtN3KX0q#0j*s(hpvO5^gueP|+wh@TA?cVik?;saUXBj(tgUa*Z`n=O?
zj4C;CXsfZNL=_pzLF7ARnqZ8k$a6x^!>`v!M?>%=UZX~74W1u@^pT#axRZH`HHOF!
z@U$S53g1)Y*2k4$#j%trx8HW0LkBx#YRHK<y@AfkGD~U3V^2KJm%sQ7qd~!`<99<Z
zz`y)4+DAS5gFac7va+<m%F;5{7^bBr$V%jNj4X1tuk8@zIie`Y(hPJ-<X3{K{d&a4
z<Cn;+K-MCX@rc|(HB$e_BlH8l^E<wi)dOoNsfeP0G@bG%<prMS5ycT<7!rmd%ByBY
z!!Vp5cB(nC>+TCL|GpMMRv3Qw_dkv<96=b6q{(Yu<La;^P|`!H0Oto}rbKE-xUfVj
zJt72g7&6H+!YC#gj;Yl{av><DDXJS0L@~-<e}+n0oo@Aft1qzEyN2{jj?_zz-}eSa
z{w9yUXh`~7^rur+4zytwqQaD8C)e5S4XbNm>Kr)WFxo?@7$GHb91zzk`0F$oGuqig
zS`a}5UO;JcWEdg*kSOqQ#^Cz~r3|J_INGW)JhRFA`Xa661*|dEHGehn@`d@D>+P@(
z^1LKXGYVZ)kR&Mye4i+&)>}%2bDHi#gCxy)_@Q_4d!P7yo_pp=E?m68{rBJd`iNov
z3vr!;S&8*30f_P-+}>l_X;O<Cq*pW4O%4-BmAz(`6ZS`hX-dAZKs1{nvK(bCp0zj}
zc@W|!Gqe{_7CA#dARZUQWr{LzWOtF9_7(k8&r$SO@GicAQU*xpq4s@@KOtLxo>X3=
zONR&h9iDXC3|bMw7@U*@2z=*=OqEb7PhpLr)2LzMCC0;Dtj^gjLT)dHltG=-VHY<(
zz`?}@YRx+93oBgRIZL)PB+q9oM{AHbXiQR0tDMuNVY2U$jrW*LH*v-gC|EstlB-*%
znT#qwZaW_lI&ff}cYe!#+<eO+dc7%U&YWj^yH|;yoT1E9_I5R49J9E%Osm~tV`JmL
zxK7_ZhhUh`^Z006;&g%31%dR@N|5#Uh)sc1l1xEZs}O%G$)P_ctVLwAlu9=WPFvDa
z6T}f(YoaJ72qhB@vwTJ|Ug2{)2e`KAG2A=LIP3E*-+M2M%OSI|Wv`#HaOfb7&?g;F
z7%eq<e(P?codGumJ?ga@xy~@QS`=$dW!t49gfmPx&Y<c|?7^FGUckcoQ4(hedt0on
z-OqHi&0~+e$kxUlcinr6t*xt^{SRN@+?j2fYRng3e2neAG0)`%J!1)t#+NyPl33@s
z<Ni1DqwoKltS%o!`99WJ^xPd_dwYk|FTOyt6><K;GkoGde}o4gc!=-%&hO{giCZhC
zpQtv9`|rD#fBpac7yQgW{(tc+|MuTgmgP5gjprM_bsB3+q%Wy2E|4foxDex#{mTBR
zFidv$h`Q~HA80C%QfG?lHLSzsIYvr?I3iERI9FBj0#&&UNKYZ>7E%rYHUMXl!eG4;
zgonxuwk%PtI&}}y>69o~qS;xj+#13dX)VGpsdqZ;Q^!p-_nf+et-&s*3&BmHqvHuo
z7~=$NUD;#M8_@4d?!NyNli3cQ43JV(Geh$PRfwumSL!Nh^#YH2CfK-oo=)oon;To~
z_4@P8;JO0JR(`F2tV{Z}9^!63FZcCU!+q87v;}QnJ6FwG3xaBWO^%NyOJ)eKas<e;
z5#$9yqXwa0sdlx-YcvqLs-RjSP+o{|jv`MnreK(jz(|_SHdYFRG^{rz;p!>Q7Z<p2
z=>oUj`yjX6@+R^k;l&r9=U0C5x7ocC6R9QQMvDjUxsBb+U!YYxjg*4%XiTFSbKu|s
zf*@u*osz{ZtT7bB5f`63jgo%#d6tsR5==6o=EJ@&xSHl{?+<WZi18#-t1;q_@7WT8
z6ON^Bm+yG*x8u#b8d;X}+;h)<-C0H{C1F?%`76It>6fM?^h3(BydivW&b{OgG8ep5
zYhS%zYPAOUJa`WeKlD~U`{*P5#&7=CYg%I|6*y2bz^Q;zMFjOWW#-VzBlJVOC?YL#
z^fW<qx}Yt3ni6;(y2!wkU_AoA63G~2D6OU4Xkmm;stulJ31vZ=JJt_&d3P=6?6Yk?
z{K-w0mb*-zd4@agTIa<xXW6_wAqZ-`>){{5`sdkIA(;!1!lSdaKoojR@|2yuZHn0#
zU-^hAL<@xw5-*4mfkFiVrBzrYMVTR_A=Ej=bjBTTyqC-vLMf>K2`bjFmsE+boz30&
z`)Qi7zu#v%oseV+7E2feblM$Uy+ss7)y$lLAX0QYEsT@A=RNP`%<1R3eEG_2TGv+-
z6~J82nW`Y1z~qvk)xy+6X4cU>QZi8mWf;-Orp#J(gvlw5MegpCb=yQ~2FBoo!x=|Z
z7|JlhH=0?!&eg7=do7`!C8(^ydT)sx7t!{P;A$be-Lp*UV_HFt4sE9C9*zwb7F$Sv
z5xLdjfz1YAlxOL;8u+##@jNtwrJ}$uOPX22z)6gZIK0|mZ)bsWvc<+IVbOThr@Nf@
zSBV>4%DhgtDH-c+qF8ct`3@%AAvGtcT{z9z%_+|nC7Y$evn2sJfvVO$URWVjK9g*M
zan(KVDTNS%*=)x1UwMJ0<t9M@_uPAso0CIaxw6ldOIIk04Cgd?nzGj$Qm;j?Xq~>P
zNu6_*Wsh?jX$k^Yl269?vBWPlrO}mpgmnaQjUq{i+D(#aMkHcX&8Hj=aaepm#Ar*N
zmc+Fhs@5Qm_L*6Q=K?OwmKk<CES&U-f;NlYRfdCvu)fF2p~KvK?@^42dGYBDmbx9x
z+9GG4+~&gM0E^u{;(9g1qIHQi4qaNZyu>OW0c3l9!u=8E4F>5P&4n(RR%E?R?s@1v
z96fXpKd$lOi)Z=LV^5Rk4z#0Do8fOg$5%ElalWu5*KsyVRG>ov$_u#jzI!=y?nRz>
z`Z2C<Tw-(kD*OF?rqd}K8yj5P+C+JhxE9gu_SoLu=JStyhHv}U@8ZvY_YbnXw8|Xo
z$zrR;Km7auGp%-$fAz0_p1dd!q9Xji?rUr{hzM;;oGFl66RH61`{*)58I2bPsCEM}
zOUO2MQOjMtRui4)6^+|g^L0Whbe`ge0YyHe9tsA7oKB;OoDT~NPl1zIlY#~VXj!SQ
zeBoh(0P7VPTOz!WFsjsewYWyvUO+)rQF0ZF@Tt=~Y$h42?En>Zcov1fzfHWLU=|~U
zB%e-kx#8|Rj#6(auJtA?c2+CLhq8QGo2_bj&ytkMWI~!Igr0;f!MT#X-VSM=^OCIn
z>z%$|uT%N!^E%F&`A;lRc}eVpH7P++U~2)wI)tatMnjfk4AjCJ(hHE%kt8FmHefB8
z(Tvj>MV26ul&Qu}OIG6+u^h7~u40ONQN<$LSGW1{XU<ZKLOkE&lOH|D_SrQ&(LlC5
z9(wD|jJ|ROdF2sqJ>|34-$MD8)s+J@>s`{qFwF)?eHc@eoO$*<=`82gn~zdxM>?BQ
zWC{6ngtdk(tJvKfG0IA898*pvSWzW!I1gUtB;gPQp5)eBZsxW-@0eqY9K+#|^XJZg
z-6TLtQO%$VN2xVwno<-65=oxt%w{vPEW6>4`f>&Hb$m$>MAT|^?z`_EM#B+z+;-b*
zUSlEE4PcT9eA2SO_ZsMCh)5lN)ItP4dOU?WW+m`E^lS>AM9PX&ri%=V#1BGhaRXy5
zxz;4QsH*6y#Ciq&{VA3PXU<(_XK#j=cR2mIGwhwa%%fj?g2)RwdEyk`^}hFz4)<^k
zLAL4`%@(NDB+JW-Xe>M1Lq?;0d?zb_d#laV3WC_jcmYC5f*_#Kk}MNsMb2ugj@qAb
z^PRWiLJ#KzZK3e4v+4fm`+xq?+UojL)wT*_@wMXmdTFg04*P8HY|`uX7*9tCDd}|D
zIA>APLn(m}VI?9G0j+kEqLe)Nz?=DZodvdcw*O43bl3l*T2pyfI|oi-ynx(O_*nt2
zBvBqJn-N-r)fN%?ltGM@4u6^>@{+O<;|~UiBFA_>RtbVEqX+`ZsLjZh9NJAlctpBI
z&3LH8CH=jWLS3TS30YrVXDAGp&!1zG>~YKSW5~snesP{gd?$<h2e`R+nI{{9QhI19
zF$h-jl2%sGv<Vl4Kw871zlVwydM24p6D}o9?zSV=4_Sh4n{GE?Hdc%VJtljSM%ckN
zCM=X&T)1+MucRpx6voXDbE-m|IN<JEZei)zQ6lAGvJ`<OihRmECrJ`^wzlc*?hph4
z^;Sr$-Q?)8CI=4O#pO#qHnujg+99kZe}(hU-{hpu&D|6fDldpbg*F9hQes0vnT~L2
zMpTax^$53L4bFSDn8Fx>D5mgcWLbjN6;?zD%XB;<jB6C>n1#*~W$M%Oj&kD7CunpR
z(F4u?pwH#)9>-4H!6T18#jz8oc<^oa@Qy!sFHe2>1t!^)d*5~o7k@Y9+WwO?>obA?
zjJDOY`v!?Vu9CepMdmry7AS4;8!e``L31`F{rs;op+nqR<+k<;(L#$XXtFoUc>Gg;
z$kygL&g727xeB);1XfD4Hq}NbG{5>Azs`sL!>>`6DN+f9P&n%-bxB!Tiqc`MhTatK
znxNCI)9?4$-P_`cCm!Xm{>XpN0}s5JAgVDd3{h0$Z~xR!V@k`v{`p^^)Zdtjotq0D
z1faAgaGGK=A&3<+Nl;efgd-Q0s1~A}rsxkU%ci2hr9}lV6#}IM((}++j_*Z8Hpe>2
za??Xv37()ZB{rRs=PhPQhuKV02#NG^y77d1Gb8dUKBBb-FN`TniNdGZXp_yR7$>m4
zU^bgFzIKWF!eL}ua_pvq^hLlY{_ttq!7>Ny5v6tHrDJh@fu;3DHu~34UVtA|OfjcR
zoTyY^QVOiDth>_jl<BoSnhR|jtq7c8R^-^PtD^koAZcD%%v<c%5fx*^XbYE(DUCw;
zU0ji(q{K`!(&2>0;v#-6n$JsWbWv0-iHbl}&FyGIQ5KZi(H~|63x>u@$&qZAGndcv
z&|BWXXG)(Jp4wu0<s4To4tefN6GD6zmK6`a<!16ro7ks6%pFIzDkmwOv(#;}vbc=Z
zCFALUz5N;SLP0GZQ%w6b8<J;EpXc^m;{uHkkQas=k6kZdxU+?`f;f(F`3xz1lzgeB
z^71R<3&}lq-A%LItO$|Lap}q>u3o!>6~Nb-Yjw_%=LO^Om~OYrY&IiLbJ8TCC~}N0
zFuJG?n9jazHmmxL=Sdb8Iyh&!aPb12#aA~L^CyK6Py}^Mt&0mA80C<~2!loB1wj-c
zg+vNVktevaz#2i^YNF;9bstO3s{}DB4De%*>7<WxU~S0OL{RLU<5YNxgKI~5>d`N9
z@#6D1?PxYyyz9;H=HYkTLEO%n&c?iO=|$X<#6R)_j5iaOmJbre5xe_Cc6N3_2x^|9
z2z^|$&CEKw?G`2saGr<f2b5ZoWH8Gz*8PCCaI7C#0hd$5qmDvQA$$JlS^*SV^E<!)
zy97bN((*Dl-*kd@tBv{-`s|bGl&$S8u54UkcXyXGOY!|GVTz&}jYf;QjS;wNWW@=B
zP||7#tgo-Lyu8YAICx!LVP0t%ZBTPXwRJhhj<Ayed>=)Mj02R*QAt86C8ic4bWAuL
zkhSX6@{%HoDWe#_D5zyArBWEBs3p^Ck0>>PXP~q!%$nrKdlWLk1tXT*Ey~^!<Bc9g
zP|($@>|X2f`Hd%7KeoUdZn}?R^)lmOi<|o^Y__hjy-))oQBGD$R8x|L0-PZaEWXN7
z%_eFXP&&(neZ^5f=g{OdyBD70ybht+Aw^(nEo3~R<!QFhJjuDVpl=ioUERr^B)5*z
zX{>`s9(|P6^QXzv31Cs41gdI#o)>5%Da#V)rig8**Y{aiYSM1kIdN){!-to7=GhC3
z#}kmRY!CRRCUw{0l*SsOVuYCNA#{#Mgbi91v&Zv5RUlSV8cN^bH5!zqu7G`_LLZnS
zA<O2<Lnr9(@8R@}hPQ+>KASVa<tLw{s}J+Wn-xplRetS}DIYxhMd}NNFF*1X&OJM0
z{XmRxlAVo&$TReXVLXjUk_l1<m8{!JoY7S^A+Bt{C<@eUiY^j7uYvP?JikRY*=5Ds
z!R()-ykMDXLC;7oN2@%a3Pe8SY?_h?z)3)1oh2<^))sTHx4VxJu9{_#)rsIbOW3$-
zmgf4yy(l0{rwsNJ7MBx>ykzt*evTjb{vYPszx{pG+5pBtIlmo};fPi}-~&JPQ*3N*
z@!KEy2(Ml|{mR$bIq>p~C{&cXK+O!kvgBoo&N7q}_)(0`GvX*hkEc~tc9w#3XlE%h
zL*!RPaq0QUB1b3>V-(sNj4ja9i0%Gyp5I?)S9OTi8nhQzId|qV?NLhS@-=Q*8gjHd
zCh#=Q^Dt?KFAah3Q7A=L76>8O+}dEdBk109f~|f@y&m)M_x(ja(F%C_GasWBNlZGU
zFtB>)2!muoTI4kAEp(}|Qs4<HHrM>xx_K*EDv8zwSu!QknwhTPyI<WP{cj_LsFbug
zO_=l$#SpKo)Wl?tJguzSyjn;(D#^=&GO+}Hfb>1;i(R~WOqu1VsbDHbC>byrPq0c*
zSf7pkgab>fG~*4L*FML!FL}K2{*yd=UbB0}kq(;NcH0664z$=fz02s4XNY>g#~rJE
z>X9TZQ{phD+wPE-n#)(Oad|UkGA+4d)u)@KWTTYcg?(;+%fo0-5XumvW)!Joq9yx<
zWiU;!QdPJC;MUvjA#R2|_V^<dFFzLvheUDbU3Vjtyiw(P=9#DH55`n3Uw?d!t+kY;
zCQWCI$3uq0HohOCq##QYiZUn7Qj*z>>&kLt>`R9MQQ3+~CHUCKKSo(-7M5PSos@GH
z>jXl0WbqQj>k!lkF(gGySZ`J>xyYy4ZZho;&}oXNJfu<taa7@b1O(Ef9>s*UMg`sV
zq1#$xBcCxTTJ$+gSbq`|?6H2}0QbJ>B;(NpDSh7Z@U6V>`|l)m8<@f}7)`jmafO>6
z{2sPuw;)AHtJPvU$=KT1Bu%F@>UH95$|UfxjV9gobq=jAG4%sP;1k6WA_&OJYTsPW
z3J%_Vh)_CQniJGL!iskLHLs1emS>)MiofyK|2mWLm}axZkN)_N^FRNO{}a;t+Vwi)
zER%G`U^Jk&-{ac$HL@(D)oRgbG^*D0yvi@<P9oJfrKMK)=yuzzuB<VcP0_D|g8h|-
zDKtu{`Dl!(VErORn5;sK(8kt7tnhH2Mj3<keS$0}Z#5~sxT<FP0oD(yE?b%7msI$X
z>6BWLqGaWu>{)@I945<4{Pr$+V}a4`2(fhsQD~{RBkntNFY3_ekhC#Tj_EIugg(>h
zVUG09FbScMiVQ_lYuZUd<U%|fQ4|_&4AJ@$Zo7eYGq!caID=V{u;7omBP}R*9HkbF
zM92&ND(9zDirIucbcI!O6<|64BXcrYl5zFwHsGp#TU2dppyKQydFdQ4D6*2#XrD%7
z!pd60`uZAgfBPFacXpFAr_a8kHTtF|b&|@zRT@n=+e7GmM4plhk6Kt@vMEwYit&sx
z%Mo5c5H=8AfKZh>&QP@{g|n1-N;*sMJP%(AVkrqdkGznSCZHAxb}x;1T*REb-SA!C
zc7oIAcG$CH(vjoRxoy^0@5J{*(qu+2onV55bTXe`D;-&yAPiVn63BossG*%jBnhfa
zFr`IXjX)9BVtN-#4laePuC$RR#n52C3^~2$(00b=V9L-4l&q55uh(|_s!D+2Yy6$B
zP6w=U=){n3Oh^-hk|7`c=!X~#`h4&A{sp@21<Gv3(&7^T;Q#seIsM`boIZ1=vR-^0
zLrD*Tz&b-%^bzBIgtK_Ej&0O1Wr{0GqDBo_k1+#B7>ATeQTe#W0p(~|k>Qn3k>~iK
zkFka@4Doyz=Z<MUL+Fsry)_;iA7trG_pua2jP}P|?oZLRh=rSPCid^+iRWHm=kjUp
zI=D?&6~v7iCN02I2qn=)Wzm(UQ`(&dt0#}Mb7_Q-idL<`5B=!h;1gE!<j4ObZDTn3
zZQqM7r|g}33eO9wg9sH}{v{S;g{pAY;w!~uIHDHC1YU$3?qjEWgkD^6nT-9K?}7hX
zLpW8b2(k<&`^d;uiR|19RUq(WHSZV#HZSpoLP|+q<dkKB(*_TTi|Xw3XN1mTjKr0e
zi8FXfA7>WveaF$(dA2U~*}Qx=wbl}ACy%qV;*k$0=%=o*(tCkr^#Ui>Eh@AWrX-GH
zvcj<4*POq&!FZ_YGy>judyS(<1I9Wd9S&JKu*BM}$LIl`UkNj2WyrqQWN(-;$_<8q
zGD-22WH=tNv!7J==T}A5Rfysm8&|F%g&@xpKK02@k|xF1y_VKGiXvw^9TN;f!YINy
zh=PE!EUKzqmZRoalQJ*9s=6x>DCHB!HO`+s&vZKDCimLanpT*KHyH+)<`TB7kUeOJ
z=f?y=NReyGX)>P+4Dp4eC<|<<Nd{v~Rua`>ikeU948oK|^^g!jrw(~TFxqfjiI%wa
z$SFSa=|_3@do1_7b&Z|v+XzF)Tfg-Xnc7C@aN+DOPrR_n;rBkoe(heqa(2j(1IH;#
z!S>D$<7uDJmo%aUgntboJq{dO=h&@BdEvqZq=LBCAgtF=^(LtnEc*e6qksdeO9-7q
zUeIiH@aN*Cm%MpidLCy*!EgNLhuGcT!CK2~nlc_IU(*%pdS$QFn(=tdXf$Fp8Zntn
zD9f^H1s4Trn&NpLV!kg(p(?CbWvoD>R^!m&BlP>%m?Sg2ZxCwbkA_B2_&yXfjH~Pq
z3(-f&9Fe7j<&>G<#bJ=v5$p{yfrk!jlzugXEUYDT)i_P*$0+9r@(e$lpnadr!6jc|
zt)Nf^UJ((7(BE0d*9z6SKxzza+od!)F7IH6r-%=pVLFVFMxb+t^&|^p2$piB)JUz7
z2D&=qVkB{)htyz7Mc69ILmw{}4APXTbJRx}+lw`HBO=u$LN-}VLbB}{xy{)~N;H)#
zmH9f)YzLU@_osdNpI(ozuCf}yl#bybCrt*VGs~%)L*DqtTUcCZ^V!dR;lH>>-`w_V
zS($2^WQLve5w)s<p_D-96S8bZ;47q2%myRs-43c=s|L@FqfAn))(Gp!j3F&^aGEqN
zs(H0WyBeC89w8x1%MGw&mJTdpyn=&Enw55&h5O#fGf!`^)Lo&w7?C71zV!H0Y+qer
zB`{!foVD{4a*DGK-w)_4tr7S>#+2lv5lan)ElRX5Fvg;Si19>I8dz<0nVMBD3_`wm
z^&%WRMy$6f=LYqd8?~I*<I3LnoiUC<f6D0>FXCLqQX36N{MEnuf2P}NQx=+2r%v)e
z{N(?g|JzUgEt34}Z#aLdpQLm+BM?PFG~PpqG2|JU3y7lwDC6+Lpz1`)3UFE&jn{}N
zjjdYjlPP(T;Hj$gpQkBtt&R$7gi)VaS<O1mhD&^EdzI08n=&7>`NH#b;x<WP8SE!)
zZ9Y$HLGs48+|I=_O{Q00<gM#3(CyT~D{*tKoO6c#{c8k)PrcTr)m`SMg_8`1eXPz<
z@fs%{d<Vb$E5E?a>xa;X?_p;<<M2D*Kr`DX?rq{`eT+_V29QD3R-B*8JSnkOSN^`u
zCVsxh;QW_Rs$P|0|67Zqz(z6ocz~J=5K)YA7C}Ywuk4yC!4FE|*F0>Y>0jPt+}kIg
z&d^5Vq(lTEty)Z;Olfs|wg(ZHhEq;<11K|`7gGyMjxS!LJKkrf7n59zaAitp$1Hj&
zt1FgPGpwo{o`=mm&Yj=q%C(f)%uo+3cO9#9<X8)A!S2;-wC;G2<;8#n*&$e1MF<}!
zEEdJ3+%jr*na(CmtRukCr^Gjs-K(d+_8seBmS_Ck|MBm#zP^r>j>%-qg-e(Jgcf@h
zT5FOdp;6|{W@GBLm?&HVxbc6QCP{^1`5LdKq@otb7^6|jV={T|hT<X!kiy|N+L%@s
zw_l);ptvE}DsA;#*3%RjdX)2-zrgeH5Lm65&8Fm5bL{W|N?~a)_*esxDmb{<;llQa
zX?B?5^4mD{ww%Ylc#b#R7IMoyF~@H1P^t;1&rC7eaCtE1@H^i@fBC(fxfBsa9uCER
zf5OiG4#M-OM-5u-h`c9=YB5Jn9%pZFN+*obk&j=-_(4ouZ{kVG0clt{*rpo=cv+5Y
zH1S>K^iWCOU%Ednm8b1aZ;y|E;^Xs;9fC06jyvz1E0w>_8kg0)c#&7Xtg&<N=CooD
z3IW!lwMH9F6nb;TX~k@f0*`k-{5H~T!XuA7!tJ-*`kL4Hm4s&;_4R`!SEtq7U!d^E
z2{aX@lAsJhG9uCisjskFQyWg0EG-~>A7QE)X;qZS(%^7J))MIgoJP)aoRN%cO_X$`
zL5b{pw2o@Dm@u7<FbZ-tV!V>D{grbp){ZjS-$!N%p_DAx0+B^%Z-&YsC=D?MN>}es
z{D7<VF4mJc;Z-vQOKoJ-!`CTBLMmfAdkN`smq!~lx~5KdSFpErh2>)>8D6-6u@&_D
z8?kRYAkPigw)$jQPElyq4;*@BYxFw&B!BAA`#acLL_X@{O$NwXqoP$C@H`(SB}z(U
zJ*MUyy40BI6h91+O5%k9GS4wuV~r!rbFySwS&wOfECfNTi!T!nG*Z;^ZIszrW_gdd
zearm>p5@4aHCmlECr_LpEpvMNyL{>QALr`M7Tr!rI4Fr_im@f5-jFOSi9Dah?lR4#
zHMBF3XV@aEs!v&luH_WkqWp+7gVA`#`qCkG_bs13zf0q$qx7?qYge`?_3Px#_;tbo
z#=vA;@Z$3q$TLlr<!D{-qwoK5nvFIB!S{U6_wdWV`fvHn=l{zmfmikYCqrDlj-HMw
z$9-g5{{Pwg?_kNY`@R$WoOAt6pZWZ&s#m_fLpRXfu-yO%I?%{a(nuUpB1LJbS;a~-
z5u=@mS%rd{8EIBCv#XsMjx?d!m7xSuLr)k2AlUF|@4I}}t9st0&%FMevwz&zjRpxc
z8tiT~MB*D!Q7<Ygv$F2FdC%|s%;#I8G;D)Nx+K#Gi6I|Hs4Pb*Q<heJ51lKF));Al
zw8^pr=P*Mt9ghgAHFO?PuQU<4$9l9tvd|%1E*OjtsL%V{eA6sPj`^gK$F<cFL8Ha7
zBP|w=%`m>Yz{Qsow>dB1hvlTKW!V%(L7Jx2s`W#Q9-rIqxgVucY^QusA31%N_R=C+
zXIy>ZJYRcif?a8{u-xW-@BRUf)HTk*9?ob3wRZ)f6cWg_!iXHBETTb-nll81kX6_`
z*rQOzo9Ea0T~fMz11~s5kxX$_4pfn-l9y;(C>tDG;z^4ni|B4_v%R`TGL7&Z7uR<Q
zf`B~B$j2klIe}x58p(6LoZVERWI>Upgh7?G(AbV-rk&Al<tQ~IHyWgcZ9Ifg7-L}^
z$(0*DuC0xjMw-CUoLO=?c4UT5tHxwJWqW&v%DhKV^JqQrUXHAsB&{?U^n19K3R(y@
zwZRHq)P6*+4Yn{yQI^nNOWGLLHa6JY-Yy3WUZ*Xa){y4~wqxPAHnwYlC72{J>)Tsg
zzkZ!4PQGO?mBEl@8Mobb8#ivOzbSxeZ~iFs4sdHtA~(SB^5IXH!K1PfhnzL73W}t_
zavj=>bA-)0wp)6!7HNt}V}wuyfzPNP;Rk}2YjIr$2<cJehV{{L%*vl(asDw5`unV(
zyULY~HT=-SYt@*&?HK;)53;4-$(7XvOW3%=B}roj!vVf!addf!h1r1Fw#T)RMda7X
zh0EfRS=3lFk`g5;?Z7+>SLDnUDb2YhgoJ4n6E?%cekr?j?|jv)ga8s%aN*)5u3foW
z{+SRo8x78!IfIag4mYnh>88x;WLZv;D~vK&!osyn&sy8I|IoC2$($5?osVZp2tAtz
z?!ONqHE|RZ_^)p=@{KAV^towso0pCrBj4G<cU{DE!fv&KjlmQJ#<IyR8)ZrIdKF_?
zIBAMo6l9)@whrOIs=&_k^8cA-G@}UxlA;jQGq@7g7`g*QWrL}`Nqy!xI&s+;Z_%9$
ziPnaUdIPSRYxvS3@M<hJDvU+OkxquwHE3mUiku^p5q7SaNQX=H4pBYCbxS+03<V8=
zY*(<e1lyAAS6VbSC*0B-vC&-M<Up`-eV5v)V+_-T-p)Y@{C<n~?xHaCdl4@_e~CeV
z{JLKEwgz=Kp7;_sHR~9y@J(ra`|>Z#wo5~IDbd1!>tN*y$8pPkB(1Tk0X5elj}zoj
zGZi+j<&wq;nKq2N30`xa+4_iCvB7xyE{L1#^>^6bJ>cE%c?W48vATMNFMR&XT)B8a
z8VRb^Dre5zO}1|lOs>*s_K0HG-|vEg&Rm_@nFXZlk`?7dpMypigUS+AmXj9+QV5Dd
zpp=E}R=CzXfY}OO#pmJ^PoR~-67nssw3TBDs(|S<<;s=o<atJ+QUrp(`PY8|&kv~8
zYW&Tg|9KvJ;tA61%^RB7e&I=ux9TLL9&R*5G#X{#UKAj6@+`ykY@{|!dqb+NHhx%^
zv|VYDj3;OWw&S66L7JsVA+WLO9_&&iV`iF7RO+%PmpOX>5t=jekXVc+BhoA-&BoL^
zHt#!goZ-Oc#-&}F^^n~>r5~N9nr(1uWrS<nC{x0YwNi(t;<A6was#CB$dw}03a{B@
z_O{z;Ub#rn@R(#<bdJn&WA}htqBCT!!(-1*sk@8Za^fb=&J{R=%gEI$Os1OY7%HCl
z`qZvpXKn2T^kK&KKc>L$(j%Yd&fw4B1~qKU!5E1!3fn1dnhI4g9t==X!uHxaD%Z@;
z&ru1(61Yi;K6x3*k*k6%&#^s+#@;UbiA$blOr|3`b8V#K6Ga7DCn&8ELLsz9D2?R;
zj>lFv=ed{mm<$VQq2QKd9!KVCw3-csWic4{8TJlnHml6d)Uc)ncmMgnMP<Ij7k=;K
zY+O6f%<Mdl<5M^eMj6yJ!xMs7Xe?cZHQ#Q9GzP7{Gi;0ol(5KhO}97T#@Z%>{)nO|
zUYUITRw^g-3$CrMa_ZCxW@g%~t$lm8)wk2B4}6&QFaH5~5)mtf5Cw&BDB={8Dl7+-
zw8&IJSoK*rw#4l60&W<PtD@}t%L{6)CYB}1a)XI;T+ib~J>qn$#!LG-xE|zF;>_XN
z{sQCn53;y2qP=o}=SpyDWU|85bcyW)Lw_>C6#*DQl1!-v7E7}(Cst-@G$l!qb7d{#
z!upiJ^=Zu4Np;FIUwx9gg0x7mR0ZD`)TE;A2%5DD(kL9)MGDy8-C*1sbMwu&^D?SV
z94@>vhG(C9mS{44C6l(Ya)OoRm9nq>n@r4>-peJ(+4DStAi!}P6lF_ErBWdb!&j^@
z<*T6TP-1Tknly>{>f?{{)vrFvyY9XB4X^R*QJUve9)FtKe&8X_tsFsCY=n^IO)F%{
z(6lTH+b$u$5^!yFrYP*vHq0^#RSlVV4o;j>j0Ax_W@vlV<AhU6)9S}egBDk31v~bD
z>TbwDPOzc{=7NOv?N!h=^K&b>+NMUEPW1>2^NSP=m=5>xVvDZVqm$X(+#5gvS7t+Q
zbmlOYg&Vj~6l4t#Bv^rq^jsVvC<2$fkmRu;cYFqu2vv7z&bD~sGhab#gDWLMy_MT{
zxpx<ZVLVQ_cJ=k)Yu?tNZu1LYWNvYp@W>p}b}+`mG8s<bBU}rWmP{8%h%)mn4!u|R
zC!|SSItEx4j^p43HTvTlxI!@O<w#GGDu>Ns!TfB6C2yZ|H!kqt1NZa$pM^(%Z;NKl
zr8(2#v!A%YSHCd96HC~hL}V84ec(=ZzWNe{dXh%F$<Edd;wYwBtFW{*hwJ)8x**9i
z%xFl%m1Jp-m8a-jm-a%I!V((S2@u9+q%G!79pTY0JqB5UV_Dy&CRbCUQI#NxbJo_j
z%1KQ91!kMG{JB5(Q%LFa!G|8^^r=&vJAdI#yHRgE@;MfcpThA1Dpm<FeI;-|Tsucf
zM8zghs?21myzJj86j_v%C3YyAZ^_A$1Z5RS48DNCwkUGRUZUA}{skIYk9$r9)cpou
zeC7(Sdn?aA{}OlKcPH<8_#Wows_fk8v2tvQwR1gI`^(Hsdd$y+@Cy6ukog4&+eZrM
z?QP)H8#LOB2+QWwZFh5U>1i6Cq&c_94}9p|oIHD!>2!iE4PM~#^m8|O`gy_f(I!88
zPYsbl6k80Y7LI4*SA235<JA{A*jPsjgDnJ>5agQD4vw#by2~lbH)(O+B<(-?1gn*h
z<#*mgEG*WpZqleq<TPi{8zQ3&nT|0<&cgg0l}6(gaBh-nL}4(Lt|-&-=#{cmd=PWG
zlA&zDU~H(nL+Z6XI!i0mTo;W`UK9v@*pn&@TRSl?Jh#FA{)kE_x$T5zdA?1((Ly*5
zNg6Zm4Jh)Q_UsJ4>k@=j^rVMZtKhg54i5TUdhS_nz4Ki-UIk^@MB@=;8W+J>Yita~
zTeuE+Ua)p!18quR_Uly;xXjHoS-Y`Dr`5#u99r!LtLtwpu$#O8U7!U!&pwZvr-b!@
zM0wa=f#rHsTMcwr!Knny%+Hf$G25G4#BqXUJ77dP;AC02zDLy$3B!>1V8qd5O=`km
zOApTvm`=y|j-e<F+Xn%A(#Llf2ptb}L8=XLnj#TM6OyZ(xlYKjBOMlJN;iq9ki4`W
zv$Z~98f*MolR^~4+UH=iN4&L7MP+3Vw3pFvQabsFxmG~avyi&%wJi$8&dvsbS0~R3
zEdT4am>7vRDbGLuEav6I7=k<RyqlR$yA1NaN&WTZV9oP9>h(H75KygF@l*+$cHI(a
zFD<)-J<GBQ|J6#9l<n;uo_p?jW@cx2_Sxs&(E7e!<v=%4oc;0>@X?QOJ$LZz8nP&P
zjhSPkTpL&ASQ=8tp@bJ`WR}zD5AmZ2VKmCJC>)zOtZ^*cB`*YLMv9KnT$L7`jVY_k
zO)gdUd3bb&&G`$MxJ7k%7c0#^V>!l+JZkY>cq||yn}dxM-MK^^g`3Ch&(}FI?b4h^
zWS-5HnHG^B&>2rj-6f0_kyZ^t8~m_JZufAcgQ|GAjzMJ-v|%J27EZ75(#7lKafWYM
zOtg7T@YI_lbJ&*$N%Hy*hi_|8$KF2YKlMA@_G3SSFD!P}_GmXOJj=$ChBQf0*%(*a
z4Eue0Tl?h$In5EK%#2}MSe}d5Xd;A798U-<K6zR&LSse;=)6PdZ_(_2jlsDN54`_4
zUwQl-=bzhSYyBcu&d0b#9ovK1`8w}?@C@CT)`?#DJhw0QNi)r8Jfc$dSU$Q$t=1$<
zQ>I>pAkC5EAr(t97>>(y@}a$u>uN$D0@vkW5@XFbxVpMVuRo+=OQwZp`fd7a-iTgp
zXH))MB4o0(V0~=|*LC>qPyHq*Po3nxcf5;stIY==ewcF?F1%@XYJPp4tB-t&6CeBl
z%68Gl!Z8_^=N)=7mY%2@jqn^~l0%Xq0uLM;H}px;1f3ToDrX!`5W2w11(udn!z$9U
z(1lG=`1m%A4r2C(u+o{qHZGt1trw}ZH7`E<4A(A3+<faC!@+><ort?0ILqoMM(ht?
zq+Uz0q@Yj=tqnz?NRp!TAk>P<##LlkW%1}4uw52b&azrWv<AC)wOKy-yI<ue{``CB
z%+|;<!v{a|Aa}px4sKlA<i!`Ru-z*N<COl`Ws<>+<)J-~ewM>%hkI^1&Dne2#p2=u
zGwl{e3Zf)tI*nMpc8#l7FSEP5$0Ul$3jOs4+;7-8zWWiM9YtJu<O`T0<4j$1<%O$=
zz-F(Pb7H>A>4heKt4R>}2*)9({NCD*Pgn_|P#7U8R6*YDk;N%lnvf<LbsclO9x#k7
z!sRLyns_?JZ#1x_gB|(^ZP@NjdGyH}47U3;D}uYugftq4PICdzuj05q`+J*2(*cg{
z(`>h?1U{B+p@hIp#yFve$u)5_=EmA8!)QV{vxwt)4EGYipbniQq=2{ZtBcn9J4RL5
zw#~^C%Q%){>KV4SR#{v+N)*N4{>9(!^E7Il{m6$=+K^ig?|Jt-*toWj=P0Bfuse-0
zPL(9fxq9g`MVf&WWeZJUBODuPKx(jaO;$u?lhXV3%3e+-wD5h0YOTS+ZkH&J2`dhk
z@k>TuCMfb8A)pMYH9C)&X;irVmPH!%($9DlNp|-V);D|f`eQuL<>>Ka<Ohn4X-@yj
z2E)8SA8g{-<p7U`=A>^^i6&GlRccb8jHD_Yau4pk>)mDl;5QjYun{zBb?&(PE}|$V
zOBD}2{4kZO_x1DZACi<M8?O>pY1EsvS}n3HK^gG<0MGUD{gALy#j@<u0#TN4Bt%I)
zD3m5iQUX8Z%$YM>eCga9UgOsz1B^6;)07jR`F*7I!>rb7EbZSQu@v2A6=h4pqJZ%f
zVcYmo%#ocvj%^<>8^y%hkQ>8D3KGvjRjb6>u#)6x$6_0s#Z+-^(!-n15IIvWIhQ!Q
zcbdJ%OJw09t#lbLiShH}sJw<{8H^pH=AS|YQ<g8^L|1->QQc>8cZiYj)O?He&K$;Q
zW+x*GUow`b=r;CoqFsFH(vzh>cP2e54Vyv;>;kGHL)dk$U*EvBCANk{t8d}<{Kh4v
zEZsW}2_aBUpuZlytwEha(!IRF<=_7T?>G~3{+SE37wT9}fKC%;LLXa2WJ!!MhOpY8
zbNUpP5M>`7WlDcM8Zqg1afHGR9EvPwe^O($H{pX#3!|o3mc!B74Q^cgb@sX+<V3qk
z!)-J1E1WtR(`a}^)0{IW7ukF2CAjnmw=G;|X{OHV#uZ!-=I7>UH|Nk=Fpc`8l_sXp
zsOf~!&JOE)UBWCyW*KRkU<pOwLl^{XOs7n|I;)pf35|^sihdc&@K$WZ;$^gflwU97
z8AI6v196nIy|sfMy8Qc({W2$4PBA;bz=uBcA^yd`{Fh{JQVCWo#f?Xwq&7E$XAEnX
zH&|Y%5c)ni0^>@I0oQU78jMx;QYxj&DdNDTX0=Hs6Y^n-(1tjQ5Qlw}VZ}uOLKy0{
zOR&^nGv8+?tx)Xu_=`V!lILIAWW9*kxU9H#ew(>w6HC+??N3OZA^SjYXtUUg2t&Kf
zj1+(>E$qahJyX^n;g6yclxv_dJ4Yr2>A^OO3nzH#g$;h?mp{XYf9PHom)kVzO>VmV
z1SiiN<%RR>jFd|@oe&GnURAQIQZC)t!JRpQy?BDVmX9*Gaudr(kFk907<01=_`VMq
z@+_ryu+R0?YdrPTlYH_spXRGydz|rjde|)g^6v57@$`MQlyHcLIhVii1WWHf&I9+~
z#`Vn}y}c{UHQUr%0d5##4$DZ6ZBytnaJDQ9tKx$(B*PK)W)oEnDYCTeksVKvj>PU4
zRO&TaGaaHhM%(4^NvsN9yx!%>7Z0df1-G2>IkM0}=>*r42*;){>d=`*fibjOGk9L5
z1pHbM1D@rBaOm&tQma-`c1WHpd^e!oYO!|Z3PKm)I;gZjNP*PmEqkc{anx$Hxb>FX
z_{x{RNR;G^qb@f#d&F__hTq#Vib|)-%^!X*dtck&+{NphJvPUw;~l>A)OEVO2}M3E
ziIl3q5kR3RvY2TUl}+-dT<_NL6Zm0>ANYi!%Z_jel9*=Q(2YZyvvYJeuQMIwSXPGX
zI7r(j7oc;EB@D}*8aLg3l!f_#NC|d^hONB`I|oCeD8?v7v(}(BJ4X~{?DeL^I^%_Z
z^FOkGeUn@7yo;q%x3D^$VvJ@ELlNi5*$_ucA|t^_EMZDq58s$|GYAld!1MV>zxa>X
z-P<Egas-l-CyyWEN6J_IVF>}Q>r$=ONShtz;_|?dq#4q(2!enx41scJ#yhlxeDxt+
z6$)bnKl&p-ij?pgHZpIz##({2FoTGhPkfH?d(Kdu4mrKH#<`=5$W+l$8Ru&uGtqz`
zNsxZPrK3wc<#-sSkXj*(CYZ)F21CLmV%4)~SejaHWAECe7_7+{W*u&XBjo4;D?4Y&
zIvF|~V>vd8Ev$-#bTv*R!p}o0YYPmMM|u8iomM)gW@2_56)w!oqpT1&ozTc5dL|)i
zo<+zV?!I&dp^osV<KfyiICF$Gmq=^sY06^V<Js%0SgFRc1QSt^8F;Inx8*(JyaFwU
z5{IQ<$1=aYoH}ijD$UM|R~gTo;J$lq;@p)TUb=LR2ktsb&3CbcML08qQw=aeU<!@a
zig;Qe@&e0&)=Yy&tHyL1F`A4i{0c8#pR$uGkQ$w42;awd3r@7waQ1$a8`~#|`?oPu
ztuxcKDdHHlc|iZ#lZ4shoSsW)HXO#&A+{};X*XDyT|gSiG)kBxV=U#O@|?}f>s){4
z61SW_3Ca+s3F$OJ#|gd-3e7?2vvF~qaav%xAtuj=i78FazV%8;@%8?yKg2ivma;8D
z5R|EWi7cHBin33s(51x?LNFL4?Cl<ad4bP;;j{dOpZv?*am($@%yih=+Ie#p**L_*
zj7xv;RTdstX6^DC<57+FY*NC6wW8))_?}0aBqZaQcswP~b8OqfuU7D@A)f1ES6g(u
zJzU!-nMPy=dRx0VwH9?}%tF|Q*3BeQPPHbv^Nu4BZ*%Xv=6V0=qkQFySIDO>Q5sQX
z8f8+p_A?46BAN=)B*pU}&+<bSp(0Bo>>wz4fJs80WhG!*DZF}}aqi-Gx47ku%gjUf
zpp3(_Pn=_a{|Lqib~d~0?@f?u%*E*r(;EY>sFdwoqqK{>K0>XY2g{-sK2B$Dj@j7-
z78jOSURvR{+wbI-TW_P;YBRTVjQORb-2d+P@iRaBb3FR!*Z6n;?%(n8PkfreV1)WE
zLP)<sT2dH;HHz81vB{bHPBY(fkjDZ{I%Vy~gl1=!xnoDk^PGG<!S%e7UFJF<1+vii
zu1DboBw3Eq5-A<RY85GRK;Z=*X`xU>#&9rT5|4TE(tu05n)!OcU1#eo%`FlJerdPl
zxg=?Vv}}|x_(7FC&T+#KAtl2kVsA7;E}X@wwmD_Fw3g=Z-2lt4QmE3hY3ax^FTC&^
zu@p3BI_#`(lj)bUCf_@?@P5C~Z~o4w-n0YRH%KZRX$TkEEUg&6^89sl0h!lE_-!hI
zpg-OrjpIY0JSE1)Iy|6FGce#aY9vWat~6;@kmVj;waVs5FjF;5rvnP32r5<LL7yVY
z$#jI0hQM*S^~@Y6Zknaps1i*jSJyS`-IV^IPa01U+7S97)n)_7aT)Xmq*=~%GR2;m
zW#>zmN&5p9XWq~9>6^H|dXX%dB8wbL3apAtYBau)6cU_QLb-1?##;I@&$QaK8_n`E
zy(+Wv%>%mfS#DX9zz>kZA_yvkL5(Omz&)H6Ptz2~Dm{;-l;4mfG#C(?XgcB4nNw84
zfL{0DO<nHS(w<2XrUu#}J{WQ0Yv=j=2T#+pC3juf!QCJ8wFi!q+7{z>oAofnggzYh
zz1l@VFqvSKBC5Get8?rZ1&#5TG>OrU%lddo5ERs+m|><sI;_?w4CM3Nyxn2ZStPGo
zjO+<Ppdf7_*Jr7xH|X*uE*y8!LStnq!-~fX%@(7ehLz<6X-cdMj2RFt6x@7igXZp#
zr&d<*rZc!Q$F180BM&{9Fqg$xew%}>eKIM@oRB=7P>B3n0>puD48*^K{o8dc!qCTY
z9E28!?8=<15ESa$f4AS}OkH{6F_~s_ZJS%}Kg-D@b<}7JUrh1B8bPf|kr(93fGxqZ
zu`HiT;G>iD&|9-$Ivj#$(+E090W-~%{n(+%Q_>`+(P|N=3f~i)nj12kJj?#h3-lvH
zk_kL*IOR|1)HQR96^t#2k};m|;%5QvMw3Dt_Ium(MkD&;l-XuPB_6Z0w#Lz8GZa!1
zWr}P%A&;lz(>}g!*w1q|UB$-cE=UI*_)L;1sd@P-d~=0Jt@x$@07#WdL_t)(&qsgw
zN9pytT)A?E-Ti&yEPc&uDukd?snF?k2*VJ^axhvGMKPoCh)Fa#gjy<uu<0I*2m+7a
z{lsta(0d-H+3s-s_{y6W)M*2`hUs>f?w-$Gcbww+i(7o{spq+MrOB!JCI{&VUF0R!
z0vmzE<ONZh(Oug@xHgq$ov<@QcNpUf!>E^I1r~8`u|F}imQxnu9S%ldk~&-cUEcW~
z!STD^g&TPIzR#%>r%*zZ=P~PR+x*i1^l>KLQ_Oh<swhq5vn(!sOAlFuLK$p7EK6-^
zh7(8RS&Zf62rnR`P7&>pKmLDWYAej1xr_SUN7x@_>>NxnC>(oABfh~?FFeUwnzF4m
zQ>_V&!a7Kh7=#cELyr$V^uzr0U;jC(^%h2J(llm&e~-=0bzINo)Tx_rU5{$5!G}Nm
zL%iob5AipidXj(oPyZ>O``jN8CFys)0DH(%r@&@?cTAipYGD=s=n6${@GCw-O1fLS
zIG#hLQO6B@#Ni+mN--J^8T9&)8sb4hG8rR8f#cg$X6nqIIEw4|m@FpO8Jin>ytu2`
z9!XBN3eFt&=(H;YeuZ2s{Gfuc9U4JMUgX44isd*|{Gj9^ra4cXzXHdPasK&>%nTGa
z|Im+f{lZI3M?)+xWHOD=TCu#mOtaG<?)Hh&l)}K&fcUOo>DN|Kym24jZ>5(_!i3N)
z&o^*v$vDoqbZrYnj^)_YT1`}z;<_%<wn>wOEKbnbq1B%)N%M?h_kgfk!xRdW=Ojsn
zDl7`+G9HW=^@jvOfCzmYV{nY%=uDLd?p&bOuCO_@c;Vud!Jr^Z1*7pUMUjx_Df0_Q
z@xwYrkuvP>5+x~lp%6MJ_bOyjhNht1nc<Fiypw0X@;UNojLdWVS{)ogj-e_f1rU@w
z^B>ubzaC_jL7#ZlAcOcOW!gXVYJ@PQr|TirA<q;-mo6l;?G7i79i?8Yz7o)Vy#!Z&
zkTfYX*)NyrTCL_AU*y-R7=w{ubP2(i-6@^TJzhMuNSIDJ%K)p0NE>aUN(H-6m|P<q
ziLq^pFklk+$TT~gfGzMwL()pXPQAh~*PLujkcpua#q4D<lZKC*D-J4(N8MfKqCMt%
z7WG8o<`(D+#9_p9GY-4e5L;xl;)rgo!iBuf_0|maG-Y{kz|0_`J<Z7KL;6R0RGuAj
z`P3qv>5zPO9-j(!y@Dk*RZ(zODXxsiOhysf3sAPjG*T2$4tRZw>s0YWW{)f&aziwk
z&>s#+zm3<rl!98VM!Vg{_kA2`AHv9E`ojTn+W)re^R@<c`QgNCY%JCfCd8?t76zO>
zxr9=J-R)iKX~DwEu@b0TsIqKh4AOG2t0n&^%TkivU6ME^i4&3}=eVay9Gl5hGaUBm
z%(V%E3e#zhu~NcNbHX<)8%;r3N^z``mzL`UDedy_$2oE8aBU-EXK#<eAg9P&0$(zJ
zTZKHCVih?XH?|?jxTDaeHI!B)hf}y)+GcxWKtD>!y&4obQ>|Vx0{><=g=JYh@Zh`o
zrGN1+uxy(P=g;%M{cpd-zyH|3f6d_v-}mWsW?5QXqS<T`cmYLTFc=K5cDK=Jrqd~T
zUVwo#RtyJY*2FcQc=Buf;1B;eCr+H?t6zQWO<Q9j1X@XUwtL)ta*k6+n@qbq99yW<
z>C`c%f+GZ0=phgYWyrIfBu>yukwh`wgD#`~HLAWt5=T^PK6#Qe(U5ohoLO37*4<%$
z>pbtc`vE@p`x#F>a+UenS)yb@oWwl+<kO5s8AV=jV{MDXNa~#lUN6Tr8K9Yr6QXH~
zIV=q}n{)WKht>s5CP)>ba)nkoAo0BlQ7>c3-6y_qm7S*^q43&NmQQih>AM)#D?EGk
z1#YZf;=&}RXUc|tD1Ci}q_hCihKD}#BmDKB`P&>>IgaHYmRt_6<N1rv@z4I*|C7_F
zZswyO{j;1obu*T269gd-zURIC+dJ>zWB=}#`5*tM|2J1}ti7rv`_=~@$`}liv6iI0
zn4N=tmOFJE-^Z8~BLqd3Q>j&P>y;8kBrQx)(BD5GOS5tyN?L61?$GQsXw5gMRw`JY
zhp{9@q46AtB6EmU!O)7>8XmAR=dyI9K`(3*`yIMM(6K$LzKxATmPBZyaUF+L7hoGm
z6TWtFm*=K6W=)l8Z=LSd8~oQl@nht+;Uhou^W1so-GqUM5}LWWIef3gvkV8ra%kr*
zzpCF`vJZQ5YeM6+eLBq=-AO_*=wf>=SrnJ0)%jVHNklXnlP4*rP^Fcb5cr-?qh80a
z1_;N-b_JGaBc)BA2)g?PIxk&gEZe2gY~i~hcb#f-+mcOgY@S<<xUtctHyBW<G{{qp
zEhR$Qq<O^4ZAZP{K1n(SBazDB8-em2C<<z`GZ<r03j!)Z$Ux;JNkVgR0fa;%kphen
zSTEyCzD=C=?E|@2tE|sRqwqY3x!D<xA3sLDUd0$qwOYlptk--8*K_ebkLRC#p5rHu
z@hyRPZ-K^IAG#cuxRyvebXE^=kIi#qrcKw%u$AQW))qJB7T7zoT$bL8(qTdx(2hs$
z`s9F@#00xLcuC5MB<5&bFw-(z-?A{Hg5LfPqgIvF4+x?H;Uw(W4ZBr?Z!{JLr6svl
zK?sARB9!gWuhi&78RQ;!UF|WmJHZ($Oy*EjB={PVKBf}#;Q8n1%QpSH-h(I%S+`(}
zi=X5ag=DZdBvKmX*%(z28;ubqweVZ*-?Q`c>~8NMq@dkwaio5X>o-<k;Te82`F=pF
z*<o>EiAJM=@0O^ie!mY=(CvSZ2X)#=l(Cp<(EWnFUXSH=oqDT7p5%DpEJ8{K`&|s>
zh4RAy$8!;mUDhTGMG-~R8Vx+x!E<e-h_H$=D;103*dlFchWmTWuAHXYut`%v5+|fO
z#b~gQSfX@+khV{*1sgkkwst1;2bw4q7^Cqm&3w&a>8!*2Y@OPoU|6IK_Xaq|;T`XO
z7m)?NaMAXZEY+mOV$XINZEullaD10!G%QQW!ho@Ob+%3-(85sl1ODQl`zf037D5Q#
z^}suM{<-`4m0$hXw@mIy1dU3S<@qI!9yvy*)4{gv(hyu)BuPw?#$;K6J}e(8oiiAY
za9zzKpZ|S6@X&`?IePR>TVrJe2Fb)o)(;}0IHw+Z+;XBtHtEw98FR}eRIw-&MwZvb
z@@<5Ym^j4`9gZAZBpOZG>m6YEHd&U_pH{gsiny<yqLm`>_h@yWV6=Xed+tBUV~@Sa
z`4?W~$jVWE_cxyAxu-U9tvX@oQf(aN^wA3OK6sNW)WdzU+@sswBh7LG+oRK(p|P-B
zE>WJ5jz`QjTy$2Ds{&&rcHq+;<{X>zSemP&vnfeBWODF4FYIjaR6ipm;aW6h*Bn~8
znZq)TW0z*;S&r*CIF*1Wo_>srSI#4Cf#q6Ms#RL84y)I$asJ#(?Ch@b+zVghLl1wH
zkN)US(&@~Bw5T>({Or&D4Q{{V4*tPE{6#+h#V@13g%#DCtwLL5S`e!O*GRVZwsEbD
zrP;;O{lIew8+C+RmX>s`*xf%Mnog;=>Qw4g`dd4U_IsQ<dy?9G8wF%ZN}lDYD8utz
zG=_1K5i7&dBP}N5n4M|^9{MRB{GlIXW^n<LM|7`TV&m~I<6VE2xr&c#`{=x&$Z|yB
zFp3I3d47YH4?oO3Kkxx&mrwBc$NmEcm#;GxnirmUf`yaESz1^?pmD4+A2XaLs9e1b
zU5jtOEU9p%qGCyOAXr&CMzXfea5BPn94rJ%6%6)A=qRPgOQ#EANotKIK^S6LWmd^=
z_*5!2awQp0V$yMse40?C5ekJJRPjQG<)d?)=?IRup_@o9TsQ3Ori{jY94VL%yA(<x
zu>pf4Ez)Fy8@eP>gfWJ|5_IY<^uY}#dm9ucCK+nRlL59aT|uHG;$VND<7du-un>8H
zbS++G8u6CM^ITR|mJcn!zA-3!tsqewgAf)s-*gk_&R?M48`5Z2-@;G&jhbprq!o^F
z&<LgqyzPWj+k0GIXfbTfVW&A=uT9}NH1|svAuTPEFd*|i%;ElRl*W$}2JH@BJmq}s
z(6b^Mdc>KAMD+!8p`yRJ#qRM_Od0`hCh_A0*QB5{1r|ya<utMrAZ(ixdjpoYhSd5B
zm59<Xz}CbSoBh)kYYTPyCZxTaqH>8UY@WX7G#%HWH`=1m9)<=<zQaHWvf&i8!FFuY
z$*9D*2#qn4S0C}pg1o_Z9YW71jS{pmMB|84r%vHHE|L1?J+1->J)hamEQ|At%*@Q-
zIxcx$;5cO~$rY}B+x2-{gSu1;Vx`HnCNzSr-7PAvrqgIsWCaRC5hqC3A*fez10Tz=
z5mF*#Sr9;j=X&T$9SwLwAZ&-!waQ7guA<RsQK?jkrxSu^i%QraY}O$&7@d>pf;ds^
z_D9^fy3gj;lq?q5w#N5zmfD6+-KW!O(rHx@(xTtnXSBLO`_}uZc?Ip2St>`Cm<oyQ
zI_TUXO)Lh5Lq9Y0##4;2kcZ1IrHduhADM772pkDVmzVj#Ll3?3WiRrAix<wnCcDhC
zENayn^?HL=v&H<}JdWcq8jYAtCZ%<c?UY#(KovPcXp&en8YP^+aGryMedgxo-ZV~H
zZX}r&OqE8C6uaFn$7Wm1&M%QA1xlxk2P0;Wm4Thn8f8jmS?KxbBtG2SQ)I<sAzWfS
z8WC%FVQtF#RP#RDh9XATK1;QA(sRE~cKRncd3uhkFC~2X*S|>r06g76iqgEh-Dz^)
zJ8t6Y?|qG;c#_V1m0o|Jes6?jB(t*(=4ThNg0N%`ssxj#*up`l2}O~jl)-X*##2RJ
zz<i^{RJD+KmCgMDyW?^K`s((GJ!KHmAe1<Cme5SouczV*o$)Kb_G^6dQ@@SlNPOSL
z^L%W_!EtP+)0oj{!0wJmr_*46|6lUl3s3Tw|I*KK&pRH(k`}h>bN~Gh@Xvqgf8ihg
z;(x<`__g1lDBk*<Qeq5|(r9CFJ&TN-z5W5!uy)wwE74M*vXq^*9k#a)m|vV@ZgGx0
ziP$`U8QZp5IeikpUPB2(G#Zoi2iTTPP^%zpLBH20%L{zh#c?c7o;}Km5BxYE{9At?
z;|G93r~;?Hgtv5xbD#0x%2RYG-D^DGN68w4e#CZK@RJWbz^T)>6BwW0|M$PmiMG$o
zLK}0iMf=`6K?_uok*6_2NKCG%2uXxKM9sX#yZ-l(d?5%eyF~3|iqH;No?ReGV=N#|
zGtwx+6q-V7oX}&o)u2+XG8&GV3`SHsP3DguWi;q=_4&(4Q9g&P!zo#5Ns=_fQHIcQ
zs1+e6Ix}n><XqpHl8=T2$(U3b5|tB2F}g_5!o_tx!oa7<3Z_YnnZ^jm;_jPo#@^oL
z=yIE0e3r|2gtZyGN=P1$ah(!2zPYi<JxNNa4WnL<`Q<q>VbBJwx5$F))su~{M=w8*
zCzFKL>sy>Wb&TEpE@_%F98BJl&-l$$7{gc>)Nn|ZAyox|5Zt@Fi*a)<&MqJeRXSh9
z2}nuYsF8(-%tuRNq%2#nP^7+(U8}#+KfYr-bUm9?=G?Szh{6h^T7z&rX4;t{3TniS
z8WaYK0*8zsgZaUj<*Qw0*T=Y1fh`TTS0Edje%)iI)8u-)M&dav4tm@-yv~LC5;~Oh
zk2D$7Ye-cgn<faE5Q~5;5lpp3e=@-ci6sOQjgU5uB`C782>bO1fxwc`tkp=1?3F%e
zAtbG4ldX;QZ+Ts%5QKg}C9Ke>H<+89!}opSI3`JA!h`U2t<T#U)Wt><>C&X$^(+#T
zvEALJTB)H$d6KEr>qy7OI2M-Ul#@tWlSeU095bCvNRt$qL!K&BW<Z<6GNi`nh-#yT
zU#XJB2_{M?Y#ZU(6jHO<i@9)Nm)_QpEX%Q+f_l|rX11VRuh6Ww2*Uu!^$<c5r4jqv
z+eFcnX3)Z~Iy7!NNy8|t&LW1=v>M^CpF4E@CVPX3L8{QwC7wnIDY^Bgd+@6+k3R8b
zUfl{~0N1j)<M!J*x^nE5>6h_j!uj*(UlViS*fy0)g=)3>iZQg~IAmG&3gl8A2E(Nl
zP3bj>5=_Q92m4+2_IGJCn#jWm)Hh>|r>cAq2m!m@T^gaFS#6+IOc1od2>QGGNULlL
zwq55ic(<@BRf3>`ou%}yuTtbWw&&7vY$}y0Vq#I~gd$0Cs}(#aXQjHzm6v{vpm85Z
z?!S*tZGp|rK1LTrNzN^IuF!NGE`Rbd@(Yh}c76wgPp>z?n1Y#hlcnW3Jg-u+2=g3~
zPVhw;Z0al~&oZ(?A#Im95e&vtmS&gP8B4x!Wz0rAX8!iu**kZhTpN@IX;~Ef(Y_~f
zxbR7wlEx|0GT3&>XtZPr@H9rCGz^9jdwWCLvsFf;DK~Dc^5_1{U*XUE<X^`31F$4V
zmzVg*|L7OVvW(yS?caILbpM;7#26A|aIErymt`qaQg-`$%rs|6#$$H3_Id8=0VdD6
z@Ajk2&Ni41dkni>=9U-nDj|;NVWdlcXNRb_PvC`w%@$VRGaL=^JfB+KrBDi^6v9)i
z@2u0?U&E`_S--JLtI?*_Xye!GoO<^U@#QBU=UvT$(66CYfy`|-55^P<8r3F^)(pi;
z%8`@Hyr&j(b$`OmOLvgCHWx2n;%3ifI2lq20)&*f+7JjRUWFI<(-7AMLf{L5(gsz4
z=XlftNvF{!N~Z`5Y|kZVcq}e=sMV@$ZtSqNxsBucES^1y7Y1x^Z8O~OLY!kc8oyRS
z*fv54;%SUy*)+m{(3Wg$47kv&(2gS7m!IOs6JMgcy@TDDrFQ%@-t2LDeu%IfOreP5
z7(bMl9B_ReI6Kb+i?DwXaq_m?iBBx^pQ@a|t8n9`XV~A~EKB^-W^-ec-u?i;-NKa?
zmJn3M;gsu}tnaK;`;Cg@m?xfkp2g*3jHim(xg+fEY~VPrU9NpIRRDWh;R!>cb=gxZ
zB@4>%;e&0ul^RnaP*O7P)X|O&wuN#Xj6AeTOmo60B8Xz<rxSD`$TNeN7*J5S9_t+&
zTV$lZi}ETMH1>Fc@_cYhU)*+<bLR2^ORHU~lMG~rsz~x>&3?sUtJ$Vk37L8U%C_<H
zjFp2uj_>y=ynrmy^iCc}h?pX?$O?hmXp#;#*>tL8xr3!_`biGLB90O)1D+Rh{L~Sy
zT)aq{eSM!S8&L#-=P`|^aL6i@QsB7`ulckP0^6~1U56kD2!hfS;N?E~dcFR-*5_>v
z>JnqfG5FHLLXl+&Q>&mi>JhjhAWMHXX|VEw=`dz68k6TKp660)RH;|04AY#g%^fTY
zBnsCNU<=&3$I(_D+joic991NYhlWCDY!4D%SevrBmr{2OM`sMPjethOLE0IO$}FB=
zL0C3fkur&<MAHdL663fw^R*7mb`3i<-1FD}OPX^nF1`3X2b=47l?ocTmK7BBI@3Xq
zi8ff$5akKB5bSmLDE1O8DJfoEPZb7<<i30F!wdXF*sNi5W1ZEt)o+<SlTzY(E|zWK
zx(>EuBM-|Rl}ZK2aR3g75y}8i=>lT}#t4e63><g1H>m^-T-zaj6N0)#8&U(d6gZwm
ztWtKm`!s6JlC}@%qCnafPT=9XF1AzlFCk@FqRtgj6yaDld7RK63`pXF<Gx|)SR`>l
z6i;Z)bTFBu8Ypg?StsoE*!s*WS$&21_AGcIn>z`E$1l?xUZyg5np=+znD4Z?vAK%Y
zDeZQHBS(%<tv1NAl*kJR(hLSY>Yh!W<XBk(nWiW*Y*CJx8|`3wRjyv|vYtkqx&LmS
ze(n<cTYF@xESczU(a8StioU$IsyH;<5c+USF=e@)a@tY|MHKZJ4kAXQ5&z$R_5Wdg
z<2pb8bAN|stA!DQX06Kq@K63*CesO@`RwPut>5SyQEUuG7(6L)5rm#kp%aGF365p4
zwlm?2=X$gPm!Eq07Fz9k+0&^7M^Bu@xHgVm-k+mx7t)kYtBv1kBOSNgtUaF~3`+-z
zJTF<5I)e3!{Ko(CkICwD%q^Xu9{8NR=`^$RN2$-wVir%)?_b1|DN1W}k#TjuMB-d}
z<{2)oUgN^W7m<}Zk6+*9*j$UHip6wxmNU05k|YK3eh*(~<bjJ7C$t<3lNTgfA`E1I
z3Oek3A&>|Xt;vKWRRy6Ju+W-iV_)OX);Mvb#x09AHU}|Jom;1SFrwP*(4Lv0KOV5X
zwt>oXbUdX}sZpJoMOJ*G(THh(h>f7xX(5FmN)l|TxORDqjjw)+&;0X$LN**Au)y)D
z&2>0>-@7^a13%4;C?in?(sfX#fTW-n+I;l3Wp*EZ0U$8n&y4l2JWVn@;FcfyK|b;e
zKToZIkN?m6TwPsdZ@-H$l3u@0p-N|xX1fW%A4UGY{feRp%Tg@PHSj!-PQ&NK(GE%S
z#<4ZuiMI2C1-CR%(^4Q1WN`s*%<a1yJUh2UURdS9T$X%CXB5$#Cd`aS7y+Ydz(KW2
zYFXU6y2(O+%*YjtEW^~6*uKryu?j;sz!9K}9B(?M@H|>c%<-)Wl^Zc`tO%Q$(J_au
zPMd+7G4VXcm06VSfRu#sm~b?}kEhIKDKE^lh$;c2PD<Rh`O@PXRF716=K~r?d%V<#
zeM>Szuq<s-%0YK^n0?EOjKRV9aFFHIRjo3+i!QU*FNg0^S_o-)EeMijSr|;&gL>Gn
zsI<a$U0SUcTU%SNYi-`%@~hEULJ&xBg&=TU;w)yTx6iS;MWSJ!>;U$933~^9DuK=E
zGb_v=8f1?S4(RUeL9RG`^GT|+EmV<%R_MbFSFSXc5ft&5Q9NX4qs#hGb8)L6Q-<Ys
z&P~VbEX>SMuhtk24{+QN%k|Jg5V#)QgB_A+jBUHL8XYP@fF;56L$dxpjm2fWW`oV{
zm)N>?o`t1l9NS|kC9Z3uqnKP90<B0yfs3TKb@f|3FbqOk+;qz=hfzKt3@@E~k==v6
z!&~BycJ)Xpux*KLm9}byF35@u-}O;1TV7cfLdw$mN?PR<tSO;}+6cx`%=H^<96Nf7
zTBSmgrmrw^{<unv!C-Ktjcv+6Xp$KEqXCYz@m;5M7_cPR@)c{imrE1EfRW&PKCRhV
z{Lmu^T>8BcQVls7S`5a5es4f$woOoJGnr&aKjqZIn7QdwZ1*2!u<4Q&AuG1Tl~bCP
zA&aML*p9<!JjCdXX2WNBafw>BO`$Z?Xh<A3(3wI`MuZr8!y$H2AhjmXQe0c(SQ^jq
z5yoLJf`!{pa^w0Io2%>PP>-<Sb%G<meixJj2w(q$SN|-~TGH=kMAI%&l<}$G`%R{k
zDgT$h^?ze=VTn91Xw;kh)BpaTFc|!89)0|Aq>$e+(DwBx*9r%VP=GBJ2tzff;94%{
z)&~5+r6CPz@qwG$1X57s1)lHIsCAIGg|r;BGUQo?UvaVBW_hpL4r!KwW8wKeATheY
zB=G!=Z9ac#la0ZYV0oQ;f9$8Y{f@iYxO##0^>v!<1=1|TR9eK_8tGf)N;6FocKQiN
z9L?TK&-2tEB2UJcB&XSpdH?Yxw4-r+56^Ka5=A&3aMVN`pPylWYnQwzzc<Q!>&dr2
zX>z4SVGNOmLYHe;=qbi1mS>vWxzgaSV-=p>9rM_8Yed5_wj+@itY2TnBsoee3RO^R
zHmS|cqlL}rpi4CDV+%>O*<dsnqP51hZKP#EI$<fO;YgoCdQ@sn;v^!Sj_AJd1ocXT
zV-NonH=-PDNTUc_JJg##OP<ZAzW6nsdE!w{rG4J@(TB*T$8M5v^xpfp_=PXg9rgf;
zK7<Q5Yax~`d3nr8@O4YDSM&ywcdtH_|8X*Qolb{`-~XOJ^7=f_zST?K5=Gij7(rnS
zo)ox%Z7c#Id2l?WABJ43G-&PaP@gJh_QouYhO8f(<I3?PjH^C&60@+r!HK;CcAIUU
ze&BADZzJ;@ap=}y8I4p1trh;Lpw%65d`r>ZPMG!tRX@jD5j=BjmYqf$gCdNFBw+})
zRrZZbi}wCDT395m&vVULQd^MuA=-nyn=swmr&g(O#@=DjA7F%Lh-4agq*07KbimQ3
zAj43ogw5?6-}KrTL5b{1zrH4(=aZ|9>=nzvZ`{L$0BK2tFc_oARZfy6_<{clYu0g{
z*R?ipFVaM70^v}RU<rk=B%xbFNQ?c!kQ=)bp1V3i0YChX6P!KPA*_WYlL7nN+c?@{
zZf2Hhy@pc>z>#Py@~9|#v+@GRb#OcfLx>1#bk&6C*T*zMn_D}Qh4~7NRzR)Nq|l0H
zXCB+Oh~jB^ksTZBFty!o&*8fjq|hKC7XlGaaY7e~L$}-I{IgGU$Gz{tYgW*fO_mm@
zX@)BVsW7-kytWazfuzXz&ENhla;-RZ`V^f`ldpW~ON@u{;eGHe!a)ORmJvk}Q4~?D
z)v!g$Jj(NO!poS_>^sl0Z+I7N1Zl4M_@_RNXNTN(-+es()RP=^yKiWX3$1X4O(?+=
z;5!ac8ne^iV|jKNQ<Q&4SO!;t(Td#U6fb{9Q$DcbC_-lntu#hDRGT$y3ug0zt**g!
zD#U{!L9I=t<`GQ`vLdG%2u=hJ3W3h2NLe6liO@}QlM*EfuIu1?0oB$F&3X%+7sN%%
zBpxEwNsP`(hGVv_uF@S%sHPcKo>SyGj;#m-2t1EroMFw@$+hO$moDK``XGL%-k;xE
z=2br_jDa{-Y^{&bh2(MXYxrJ;pZRNllZAyvro92jj~wG4{)7JqKlk&0i_O=0()&)O
zjlr{QDnes)i65#Mi}5()(JLb+iROJrZRXk@mhF)xDHwxgTjhEdDOnPuvXpq(BM2%q
zj~<~=1(q#veHVm8X~S+W;a4Afk*{4lKv!JCMx9<i;bd>jjq5jf=E)~9w&bQ;?qWC`
z&^y@YxGo(UWTEJeCM;B4-hHykrJ_sP>r-jh$t{Oz;<KJB=A_FciW!dAxqS6Hk3V;g
zbC=gy9VJYY>`zS~_i9-}kQ&9r2#m0CY(bo67#r?CIm0`bJ)Z7mJbH1265#kANfgl^
z9bmLVs{&&*p6fv&*x%X12!WbRun$X9<6&tLXIU0#jqN!UqJlYg6F>0Nzko9xPzfE@
zH`aOfsi#Skgd5L3&W9iVL3Z6Hqj-evxM)+bJTr@J!QR0Z{evA=U$U6?27G2Sq3Kzu
z43rWyn+>MDK3-5kX$4Xc8BGO=G-YUK#2bp^{)mh*f2@G+jg)QQY;{#Z*%4`QjfF1_
zNvarlJ{dM%5_40Ov#>Uxw>HJA`<%b;BwO=MVn@*4+oH4EW$gGobK(TkT9v|gkdB3|
zG<KFDltOAv>U$Kfi=ztCR!(PagXth=wkep*Iy^t;(5tv4UIj0YIhZ>NXe!ewQRt&=
z2jw^@Y13=XkvSd)g|ICgoe+Bg(y|!rZ?nI-$@bbcGUjkxAK`e!QA*WFMxsoPSQzrx
zsC;N`=#PdRUs*vJ&3HUPS~io>6#dFp@QtrGrtBwA(v;zF$Z$BMR;wLiO`Jnc?dyB;
zyuH(}Kw#N~LW3^I3q!^7kOe$(?SRKNVw$$akKQuF>E#AaSW@1zIA{6j3Qp*wwZ?H>
zg!Gt<hD5ysbduo(Awj*0?ODW0!eBHaEfh;fTAVl-<J4X9c8$$yo5%^sji%+exW12&
zHmx8a%QMnE!}Y49R!R_70D+P=iOR^0#>WQRVZ48U(HYyjTa2P9L1!M}dQ5s<$PEsH
ziBVX_kbRr}>Z~aE_aFNhpZ)A-Xg2GFVZe>GHHz0To775E6gf!}Ga3zvqKL_4LZ0U2
zc}|+96h;1u{g5{0?=Rmdh1RIlu(`F1Ez9QoIEvncH8%LtrYba2ln>3ot&$Z5(`dkC
zFd&X#oI;)^xVE6#uG47M32Q!q=U_{Tg<vr3GZ{~iXe`GNgdy{D2Cb&ln|11q3P}_Z
zwpuhQK1rMqkH?VaNTsl?0%1zL3Qme^*@&V_k}ISYBAg0qyFDi3F<G9G6$SIN8A+6K
z{^<)$uWoVsO=rr$J59)<1fdna3t`}Mt=}hY&T{T6&ms$hYY9e$dP}X~0P@_hvpv9e
zZcwQ_inJs@^D{rs{LBJ*GUeU(KgfUa_x^MK{(tp<Cw~iAW(Gl3mX=AT@Q~zaV{>U|
z!e$~l?pn;YLi!U3>Ke~-OGx*4LY}6OC73iL9*(JXW@xM&E3L#VfpjWJ;gacy7p@QZ
zor`_Eg$FsBt#hvX0^MGLkIl1BKgZ&cWxn`@&+~x~ei)+_!-G8rH&$uPdmzfvM@32=
zyrqdIV>^{B&bF9Zl2I0OeK2M1xm7x=JvM*ki}d;<qC=SOpLPihQ&m2@gq34(&|0%H
z7jXN$O{6ue1I=(WMHeZy>);WhltQa=lG_V{vf13LQgtlGdwZpgm~D|K<v~seg){=`
z!<{RO+<R=1<wk(M`2;V1`ZxK#-~LrPi$^(e>I_#ey-1!Uu(w8Qsl!xxWfj74IWt!U
z499P|g}`z+d-hg(>)Xt>PT`XA=;l7vgMFISI^#y2dac1^IxSh1wk%slg#3E@1K-@o
z^^bKezlVaeKd#0`l{SR}Jj=pIktPKTnc+*GgPmv=SNC~gGhyC$n4E00(+MeD4?oE;
zh2-k#(`es8Dox?}2;op2^a*<hm`ss{0a2rlvIUhmra74+h2-4XIqpa|2s=L4>QzEn
zWmZ@4ij095q8tw+Z4N3G{3I=d!D@$@C}u105Jq502VZ44swgc-kyK{dTv%O2j1REu
zvxuU!z?;Sy4iXp5$Y|`>NUXeaeWEFA*RM0v>CmhO2np+J>zLPo73W1k5+{tuBf8xK
z>h(H~<Ko&*$>V*URhG9ksFT8`W~p+cEqo*yLRzftCp@*8q7>Y_AUVG1B8()Br-YR%
z?Zvqg_Gn2w+eQ@y!|pzDuZN_dQfm;_n+VTgG#L|(CU}mGEiD{7;q<9_t}Wce{eSg8
zW8uU}Op(yPcAoBIU&Qoo&=8O%IocTVyufyCszC_V09(WP)vHWwL2czug2o&-i2%PY
zsMi`OyF!*1xVBHF5^%5)BTd;oW{f}zP5wIeh&W4G-`-|pXQwnj|27+oHik6Km`p|t
z20gmHCY~4KISFwbk)#oMmQv&?d6AW;(pSqEY{$WMeLnJ`4{_%mckrptewyF<-H*S4
z6$WsqS_V&mR)#_;Dz=9!94>DTIk!1sbw6Row|V&XqntR>rajj|Iu==)Fx>Al84Q^Y
zMpSBT=8h~94SVP;B`Z>lt?A4;_+fzMg{U;aBq`c;@WPTqrHY&+&&bk(I8TWpIM^L>
zaBx5rrKG7KD?phcd9E=^AdHReYP=IR&18tkMjTt|Fv&B5G$)Uy6j6kZ#soGf4E=hA
z?Hk)1bceXYAw@Aj^HxJJ3{VuBt&IaL>k>xW7%ljPU-&!J!T=!*|H)tYi~QcFKEdZ6
zc?9()e-tTf0!yQ`peQUdZ5T{)c6$QXcpRzcU<_AQcc^!Knqd>)^~jS1HI0a;V@#o$
zTU?@X;wZUmA#;TzeJm}A)Rafg?eMv)86W;DKgVOwe3egKeTq1RmfgUqv`~@ctDpZo
z?|c6T_^H4AR~h$4>|DCSqK;?=Ews_(qzJ#lL6Y$7PR5sd30DU(QJNEH1%=Xp;_~>r
zN9X^KMcNcZN+PAfatbU2fonK5?@*OFFT{f0aLUS3ohw&Iq($ld8dR%TwgYa-8w~3;
zgeBRza0!wW+i_7zj>=O60^71VGGF1RAH0Lxmm2uFB2F^$nF?!-3MWsU<jBdJ$fgM!
zH!kB>LOcpwU!bA_*LJD;K1&^+twM72zIXCxKJr0qJ7l0z7LJAd)^lAF<zPl*)~;+)
zs|I+pZ6fJm#~E%VpqDFf?Q%Fu2pkRp#BVSj`;!pR0VGC~8L)&+MJh0YJoT|nh9d-X
zdokyCGAsnAX9S;_u@D9L@r0;SW!ji8Ctxi}YTMXFLH%G4zu!ku5Z9VSjV8vmX^cmN
z<0-N0lG!$yBe{Bdo=Ml`#ft|ddW0iwTJ<`!Cugyby9|WDi^`rhH_J&vpFyFZ5|C?n
z_qEG3Ey<1gDR%pPl*pJIj2JpLGeI5Ov6!R@N?3?&g5(f1Yy=3!Yvfme@>Utg5tGgM
z73Z06yN7E-oWu-=eS&UCwOYe*JRI91j;0i^v&8<k26etLI1+LNs*vQyViFrJZ)Xe&
zo10viYx!VRK{}umNgS8ckwPN1B$;Mqy}ja6^X5TotV#tVBuN}2jbL_RzC3~DIV$%s
zevSHF_j2od--8UQ2&0%gd5ZBJ@8aqszsJUxew`)X#<g5L&tp6qA{>Ft5-#qH`TV(E
zj^2MKpZUt;oSsVV`O6bj8j;62PFP22jqm%+FD!ED{5dcR?O3Sn>m6~gN8d@x6-uIf
z6ipdVh8!I1VOutpAjB9=nx;&rQ=%vWN}C#`%Jlim|Ayze96h?ssZ%G}+}z-<JMVnM
zL7gWckOpN06dr;UDJ}N;317Z4Vs)&zt>*H<TRNOQ(ZQ?LP)gI?-(@-);ae{CN|hr^
zOL)}^BJ|N}0i9*&qCla^ivpDtBuPY(Mij~*Z9#51jB~@re$39+h<ucg=4F7VjE#$h
zv<%Y9u!JFWEP|?}RrRQ~9O|_y?G>Lvo)QldE}mZ_m|3EpYjl=SWC`hXh$A)INz9ht
z;Ku4E3Y)A_rI<{J$@z9xaBo}&Z3KA^8*96zBZo0mYZd<bfBLgHwnwYg;y?d;f0wU4
z_88;o8_&SLS{0$Ogd|0fDT_&(kxpXBLKZDUyQ=686qZ%6v$sRAQbiP!>2yTJ_wX$T
zw^75bSJ93Ig~D+hj284JV?O@OK40B+S-JHNKK85siqAjy1i2B^<Q#!KLd89T$m(=z
z9Ugx8!x*Dkzj%(vKl5pBao`WGZSzuZ%1e7at_>3Q<D58u>&&?SBx-6Tg%P+I97_{A
zhI&23t7P;9?2ddY6_1J+va>NF%MM8L1l#io>NVUT#H|DjMnm>4TtN;ZY|lX@8HJ9}
z#!zvpJapF#@11ejJNGHJ#I2mD+)Bd%<q1xmUg5DX!+YQRKK8bE_}phdNvk!>?DA1A
zXoDrKvVW>pWj`wT+U0FliW**hjwDWLSAE8XCNl*lDbR(&h;o{96x3N4AvgBB%sKN2
ztMuIz@>Th{KN<1Zo1(~QB9r6cAcdeXkf%1&L~+E1xR*0AE{nD%_8naO5@Dkum`+i)
zUE)oI#EWt)RghGDM!`HNjS2#=9sDFg8`z(p17mRW94pV!ewCXSS`60?Ncsl|VX@yG
zurnNUdUBMTkA+-rIAo4b>ibkCV{AD=rx8n|oZ7j4qML5Uxp;|Ys;Px7^1L9NJ4%)e
zD8>gQI8?(Bolcpeuw9!XRk)%YqIo^fkMDF3&$68Hc#PNY;(9I!s8y<m#EREfWqDhJ
zx{3s8GzpSi*-X=%VLT<7Iy8Z$dWt0l&%Lz9JKyP2aRP)9MB@oy5PA8Z)Jl=Z6CB5-
zxwM2j?87p)Bn<0VLZY-P|Fc@)sfe8ypW^!SkI_1Mobh<X-24%|N`r+59_HfYfP)vm
zM8`6iJjaqY1U8dc@qa#f4RQP?T&GGPB^UqTv;5Eh*Pq9$S6R93Zk(XTWEvxkVQF~@
zzgA(?8`9`>*xT3zDG+ZF6jp0Pnq^p)q$pD2WQy;(q)AE~#~2K@ZDUy$#bLkND>s|~
z%Mw_YU~g}io$VdwmzLh}XR2BTOBhlOx{zerB1+-HR>F-)(6TIU=@^#gTriR%%fNA|
zH(SgtEMP2&E>wy4@B#`HX`Z3791v(_kkTP&gk-j05RW<7+Gdy;uI|KajwCwQ%!Dx4
zFwE3#{Ln{O1)){oy8)it#IY?L+rjo+Y+F)j&3M%3U~eDq7D+Pb(`Yxj>wWh#vB1Gb
zT7o1k$c$iH_-wBC>5e0`TS2%EW0g~yx_z5~?oCPL8I(3;Ic%)&l)k#sFh4WTkN@aT
zVx;DQ2Oi*o2kz&ypL^tu{$t;u&?+;<8HQL%`V-LO5(eAxQW^nt4<wDcO&X7hqA_u9
zsMr=+l2C8WU{@<xo=XvDSi+`I85;)&{Mw`I95im{?uYK-Q=j{Fp1u4MnX##f4)c{;
znQa_rp|gPP89x7|M_4_02Dj#M?ebMtuU{uqnxrUri!0!N;x$!emdQcjTbf2i;#3{9
zNZ3qmdJ{#nW=Il?rG;hUXmoh6XfjD+3JC|h+a!Y#W-!IF%0aDy3@nfN(BscNxWp-Y
z!ted+zu^ZT`VoHc_&tQaWKc}lpNyDUp5do|?mr{+T^{*^e~s<hv=$bqtej+UxsNas
zAuJ|J@yN4#Skh;{U156^;l&xU@HpDB`N?DRe0FDxBlq7+7+M@%UgXT3ceA*##6oD}
zI`*sLx9^N4_oqb)m`KBt0Y_+JEy%UQU<3;?=HgV)P(H07Vj9|f;Dt3Fy#ubdT4*8h
zC({zuR49Z}7$I=u2`J5QZXRvhIBAN+5I5`OmR$~~34u0}R4Zb8%=_=Zi^re6LVLEw
zWH_Wh>~dpc7r}9ko`nrtp_D-*$*Tcgo-=o0o6EY!$;KSBMT&RvD)+Z%Fs%h-txakf
zit!GXACd`0yVIh-y~E7hEL-clXau$t<Zr>ynKtFrtWtS7R5%&pdoD(mz3{I`Z*Qi~
zRV3ObGmt1je`3&)!XjkZRdi|=2mJye3x?5<pj{!bU5Y$q)bC?dhDkDtIHlTd(^y%;
zcrGf-upNoz*$8a3DIMBWVYs@p$CdqkMhYJPr~iQ3*?U+xdWyMDhkNgT7hd48a^FKd
z`S_Qqli=C`f$gG1!DtwBWs>mcKlE<yd+<SC_=C^$<$v|BdEz%d&e$>h+5h4wx7~Uh
zzUQDQJtOP&Dihx!EeZ-_nEYW@;hQK86uKNh5mMr~r2~fsvR*-wWsD{h(kw@7^9=))
z0)b_f7?e+b>JvDw&+fq+8<GGBZH+R597&-prm0~vnlKnySVnT9s+gatkm~>;ozgPO
zwh0^c(lJCx+(3dcOh;2j{VsVt#?}(IQo*Z)xUPc{plt_&DtqaeCoWCUxHO!MrA5t=
z#X7A<n~EPWnRKx%MK$c;*gjBBGHRtLvYg>$!l2tFil$UVjfQWtuvDR%X@Z4$hEfvR
zE=mgW*pO(+z7sIoS|?E&<4dB^ki?W$aNo*G2Edk*+0G2($(S@tQA&SD8KjrT>0j}N
z)P_7)Y;5e{`7XcqTffTD#bexa&pQbGfS>=lzrmNj{1uY)&6}2UJVTOE29>6u+b?J$
zX<8YbAU@>R*+>Hwzluo;lHrj0+yZW^iq#5`wo7gbbfy_)J+7?}_?0KONmlOT@BNqm
z760r1>mT#frI*N!AQTN6?gCmEc9TooSpN$B$u84$$n(2zWk&Q*ud+<il;8;PrDV2c
z5jlp577SyTNhYb-IhoE8mO~KM@a!7XLg9D;<K6+qID$0Ca&2;<Av1V(m77nr`RE;W
z=BMlItabT?zxj80=jmIqg=T7cBxZ|39AE`gc6*!r+Q0ocS6;lp$vbZ6=-qem;vi#`
zX>0_c<s-A0w(D`nyvLmZ+|&rDcn-c}(X6}N(&+G0$3KW8KlB{~fWNO&1{mLR6T#*d
zI*N#~F$nfY2G`LTZ82+GG<-o`7`%z*=%mMfqsG9kQV3}Fde~WnDHQHFB6nQ&PoE%b
zHE^;VV?pA1Xv;!qt90QJn)ayA(&i@aG-ah)XZZuSVl`(m_&o8_HCC@)U~~U~Tb(+s
z;|>SDPg1EPY=KA<;!#98x5RaEgrqaig1b$1dz<0r2IKx7!HJtWe&iUd{R~$YbebW1
zttv%PV1!{@fc-~CqTjshnkhq?7Z?meg6q2_%4aluUGMSs26ZdPh^}mcHsq$@pkGio
zf~qvs{EWJ<+1t0N*91G;8&oR|q!3I-L#koGq?clOKC{cqR2LV>J)0s;5rK`gDhM1D
zF>#Xd?4><E_2LyS?T(mgiDPn>Mw)luaz76|@IH3duCjXd3a8HAMy=f@YA!PxUuCXh
z$cr2yBwO7vse<Fj&(iEHa^lW+@NxAkR7FDIxm<hVagN<{H_MBQCDx%D;<*kQS$ekt
z85~Yky*1KC&>xJ6;*`r*uAvHz>v~Kk6XG~VDfLa$!NwSjhI*~eh0z6~XrCacyy4H|
z&b5elhUD6iXy{Ks#08cLSd^N%T0xrTyl`Qi2ks9Eoq#+^iLD8a?_=8@d0wEjBG)ON
zCGdP7$FJapRiy8b<vGKCpEya#3&reGgVyGNMlIvWOh~2bQm^{d!!}4wt<k}@K^2C=
zXo8@^aBqw0bcj0istiJpTHU8HQ^gUI_Ip1<BN-v)j)Jj~w#1e$aqiJoKK&%8H<|zz
zVGuBlCm3OYP=CZq=4DAKIeYeI{`Y_WFL>#t7x{O;{L4K3>@!T_*H<2uQeasUA>?6@
z{`GQtmMPZPcBq6QANvo#%+X^<nVp&A{qKDrr%s&Y+=UBo+PR`}=N<H(IbSwBlQNw;
zG{T%}7*Ve&uxuPVBZ)1vQ54fDPT<ok6dft)ZuQyij_9;1G(4XRm#_1$FC4J#ukf=U
z{&9Z)bD!e3zW7NJqX9`|229f)I{AJ8_ubO?rSo(?@J_Up*jUu-hH0W1Wfi&;7o!!*
zlx)d3n?m3a)M||Ob|}U%x*ubT6r%*L?_ptKH7wqH^HKinnJVYL_!~U6zt7M9rC;FW
z!Vw&D#^!*%=_L-vH`wiMF&akPe%IY}x8Bdn(s7R5d=u~e=tsFRE_m<CGRLYO$D37Z
zu0_*x@NM}j@76yQxXJGxtoKi<ZnUc0eb<LcKlxciKSj%&{>V^u!NX8>6*X7W(+afD
ziSdByM!_e~+(8h>NTY}ypL#N8G(SggVS#+6NuUahEitZx7BvJ4N5-_e35(BOLtS2H
zXOg3YWIr4*e|x~{?u1Trjw836B}!s?``cXEjW}y*dS@0el<qx_20zQ#IvCNfZ=kaR
z<u5Q(j0rs(FCJq{!@}Knv+?C;$r6L(cL=H#y6gK0Q}#%k?+{V-oyY*=NzB&vE?Zl>
zJn`f+Sh93Jc|CetgSt~c_z^CQeh0nVBR|{>>On!>jc8X=Y|BF8Vhf4ydK78G!DvFw
z_sNrl&g?w0;$wwA%C%8S;W#cfE>aa_Ou4!<<oBN0XS8sFBk#J0^x0?F&MqMACOiF%
z?!kzQ7cR1S{USlV%8}z|$kP;E>oD58#Ee_QdUc+2d3ORxhI<D{DXDc@EFE9u;il#w
zQRF72)v6&($<q?5oLCu*qgPF?ycT}&jVsS{_V>Q?44V+*6()}7d1P5Zx7&NepXbO!
z4|C-=KZT4ka+A^R7u1miQqc4??V81YPvg3ZQ8dJFc=(=Ak><pc2{s@syX4miMNp|A
ztvYrXVuT<~QYPaGj^p4uE?J&o+k#t9b$Bs%G0VrOR;o0M7%dWNNc^w~D2#}a!a_PW
zh0f6`r(SF0I6j_Zfl>Hjg<>@3uAlsC_;r_aPd-L(bDf}CMIpJ78U|sLo!*o|TA(eL
zG)X`REZfJk6ls!w!!>!?-#7Gq{=%RCDc<?u1H9|q5AtI_`4jv%zxa#%@BhcY{I*|U
z-?ItBp!8BLG(}O67X`&(P^uM7$0^&JTezOX$3O8q{PbV?t27!79(eEp&Rx9lre%w7
z`r#kqa&Fjp{yZuzuyu$q4z;?%4_$JV69zCDLy_5tKyrB}XDqs87k1g7B;35vA#e;&
zp1;U%tWG#Lu@K=d|J}d%MGks<WW~$ZP@~Y_dsNsz;kxqa-=n$Ep>xwoY|BO45dw$F
z*kL*;5E|0l#}6|sY2$_=DovSjY<7|{^=5@;*k(FT85a?AOUL-3)3f~GF?i&+{(pSz
zUw@2W`lWxvnS~=*LNPG|hS@Gf5m6BxR%%PM78jUnuJF!dAHk6pj^p4sPRWw{ezxtq
zu6!^;{n#C-EJuFw5sEk^*D#tmG_91no6+!dYyk#`>&7BChRSqIr`uz#-NKuuY#cp~
z^ld~nz<&8q>iH<+f{Lk{oU<1PRG;6Z>)70U$A{Rvex6BXNIrzktv+}Ez=K$>i)#tY
zkr|%s?IInUygR~O(ImD)3rQ_bv0_cNUSJ%}D9P~_k5iL<%+UtNclw-nYh;;&s5a2?
z0l9)W%}BBW>#&dGt=z+nF-*o&CgbTJ>-BGIP&aq;?Gz97c;R<Gg&t>U<pC+xT0*r}
z`b9ccMxF{B%ZH>O9}Tg*0O>i1Mu_G5*shIAG8}1xE68=sa2)gbbA2AUmQp`{k_Ya4
z51;wcXZZcAPZK2tH7jDi8uG$3FL28pxAB=zf0`3#&+wri_z{MK9>abgS%5SaxyqQP
z37g%RI+_b#`7*!%m9G+KV`iESuI*hW$L2kE&vW3pT)KRTTSK2|G{UjUvQvOyilI21
zg!x{RK#;@<2i-1r-g!G$udb4&ZybMd<el$gdVQUZ$DSciQ=-VF5hxl?LOo0g9gAUZ
z6Z(Sf?G0+xCbo7cl7wtJB`H#*GGs}D5R%5i9IfSL3SDMUmDbdn4IJAkEyU6k&v8&E
zLuT&(K_2)|{x-sKne=v8f9wkkzw~LUBEq&kDyBk`#-w?U?^j7CF|B3?%W^>%2nkw9
z8252Q7dvRMeK2Hu^#ZdCi&(;DU1>zUN^)?(L<3TiL=moSao26{#0xCG^wrN(yeg?=
z5IB}#d0~N{`0<~>wk?E!`Gq;|yz@5Xx7nX;+oo1)&~CTzJs-3oO;aY5F_Y<(Buh)k
zun-({r_@`0KKIC{`M`T0;^e6_{J@7l%)j}!zx<}*O{xn=IP;;07>@ebS2n4b0%;u5
zT%+t1of@1_Afk*s@+ew@&#w--mL+(W&CE=VC%1=OJ3nM^nh|S&0wyVZ_be^Huh-CE
zZIz3k{T%OVchF01Qe&Y+MjSyF+xVWvbdu9-JJ?=9HqN+d$)Rdm+;P0k(N=}cY0L}P
zdn^aX`H`Duc>k;<-@S~hEdJ4d{onGgGk0PO0a#Q;hpN{`_;(?Yz^ekc+S_*>zOS!q
z-0!jz_2|t8yyL-pNxtw^Ob%)0VpS4?YC+&x__m=ZQ|2e|MAO6X4v6y{J1a;V6-4M#
z9K!xZoRe8L(gv%Tve27y^7%f+PQk(*?_xi?#uqL<#m?p~c@8bdrc=#XefkL^Be3Ej
zc5@Efc4>A_Ao>>&1z<^Jk<&~QWCpGAfVxcCw<CNLQESc6Si8<vr9pAJPL^lPRYMLY
zA)_>=Rc|shT|yy9jX(>&C--r9n}a%ITQqMyL;J=RMvuQlr6{npOBfowK#^nxp>K#H
zr)0Q9W1LEb5rTzw10f7rK%!GjW>6x+3=J=B?DM-X?hrL^;q-gn!@<D~{)hkdzvtS)
zE*eQ7>UeS%twL1n^2y))4Jwt8hadg{RG}Cg?6H6KGPOAujYew4!Ej8|mHgPL875-P
z_Qser8q>`UJKZUN{_Yj3R)L(E;rN+ZvfMBl45|1&SQ5X`xR`G{3jKD;^=!h><<h0g
z+<Wic)ay0Y);Hhq`&vPTmG?f#&b4cpjeU%=F-B0WCNwG~x6Tp<TL@g+Cd*=C19=it
zuY`<y8A=<PtqzT)MP#c+mgnUFi!E?MA1N)+5-Gqj2B8Ikr`fxHnKYkLX*Fn8R%p)u
zMGkJhgUi4EZ|H1a#?l2yL6&9MjziV2A_5<@Mj4w-=O_$LR$v7l!dM*a?{nd~XSn^I
z2k~nyGOKjOD599m7{VfFC~^vAarx>6l+Lkii{e$5Ylf20f5&aNaptBoyu4QxdCvKB
zFTLiplmP108VmD_%+1YFtyZz6O*Ea-?e*B%-NuXuq-h!0<vHx_c5z*s-~PmJ@^}8`
z-{bB(?xNjpvAehbhSwJ$GdXF^W#!@f5!RFBSFWS9&FY>c3p_k~N?K&x(30O@8?$j^
z$fc1Y(gsZmf6No<_x0VAgN1c4Ve^Ym;k^G|@`VmUDwbyrYFbWT7nzOD1#T#4x-K6&
z)nvu?80mtmn?3H(iid7K!cFr{j)wwI8pM%1xa+Uo`AWc6{w}bP-=8-99#=H%vo$=x
z)fwvD_t6=3iJ!iNS163~8Tu0KYwB5stU+^Zb398(3qw??(G(V2^$>x^!r-VBks74b
zSVPU?+8)(sw@KQ^Is1VJiFWqbdh{F{qdo>nC>q$zqH-TOb(wE#Vp(Bh5@WkA^*isv
ze|F=r1g}xr;KVVln!~XZ6_gjVy*pr>MqG#s+Wi>oz9o9ekYsX%NiV_)Ys7lY{7jo`
z_A1o^A~f;$=spf_b5J)N?=wn9EIx1#I?0(le;%VPHV*`LMI$h{Rf}nyq7s*|74qC_
zMBycL4kql56GW16$C*VI+HLl>H~8#om(Oo%j@<nafA{bIeSYJY{}sRSx!+-!#%Me$
zG?)oaQupSWsn6pXi{;r0f9dCbgyT1zVD<b3w$ERr6%S~I3uvP-Mp3hEK6uYOLZ&>r
zvjKyc`b>-IG$#pMJj=oO7PjXR`T;7BF$Z0$`4k;`WaH?NWL}UT!iT?C?hD(tIC*@D
zYQ^W;)it&^H>o%3H0sqi{NB@K$kZ3iKX50b&pb*nNwIOLRuy6BkfZSIggmv&p@pPi
zcYBScxjCZon9l4xB5<)nAFEL#M}rbrfrGRHBo5l7XrWLDvP`kv>oJJO6yt=)Kk=(9
z+<p(jayfbWR%(l<S^m+#!qs2<XRM?nTw&pQ9@Fs<*LRS%K;UrU%4KrbVrJ!58Z*lX
zTzXsEXq~dYaf5L*#c#}@Y@0aF&{2kqAd_GrP{wf3-}&ZiEHn~_d*5*%ei$4c_zZ)=
zfOF^0y{46yB_-7`WUe#Ek)<Ovn@unf#|f6@kYy=_%27&@WjPqZbdoU|P5IK7|A3$R
ziNC<i>@3Gtj=gC?-JrL_c-*J9bc9n6c{~?QSYO}aT3Rr<nlPeZltcRcs_B2UYPJ?=
z*P3)MUSyJtIr}3YA!vjwHgXmgC0T0Iunn_S&1|PZ<5ZKSP~Zsxc*vW}{@iaYlfKDT
z4d2&yRe3ZbORsUZ^B_l50`IwxOxP69yogp7`MxCdi!uXOlaQv&j4hrzTH)l4h^m~h
z+@0bKBV?4YdwP+z`B_%BH&|Hha_&ZkIdYm^e~;wq^Stor6AaP_g-eAxo!}%jcY&G4
zJa%r;X}LVP_84oMTby{{y#yVFl~0Jh`XOLg69pb?_bwppj9a#@;|?>tnI=;yn5G_n
zuF*+G+z*+g5mmR&Y*3-uUq{wM(*BsXXVJ?Fl9IjrJ%>hmTY|cmGZEYCH!yilv(e$`
zgYV?ZV8CR3m)^i3s~Tpj85Ju<yB57+jtT~RZZ)DW6myo%UB^0{J-UR^DUUw;6o0T5
z^W?yyzI-zu`Qe}7D_?$u|Kqp*HT@(8Y#O3T+h3vSuh4GJ(QefM#TUQ$7*1o#JKuc|
z)2(eD|HLQw!R01#8l$8|mPHsZz@XS3!qBS`wmotoFpgj~O8DLN9sa_plj!Lb+nJEX
zIhAz65i8@w+zfm9Hd5t;5+<r-AAPSF4$aKW^3HqT$ydMfc`AX$aJWZU@zJKtF8y(c
z`Y*p`bA6ROjj5hI!I680?0xM?5+&Ig31pa|GL7(DhGRoD&2WZeq!1*+#cBjh0td@-
zkVX(qA`D<#Hc~l=GzXawo1D$Pl*i7m^W@qlyQ7%AuxWY)vzNCy`R<1~e)bM7Joh3G
zyzhNzKft>616+CJS6K`k$TeAJNToqW2~S?z=jki^-15G=`O1ZLjyYGDA5ZXV0gmI7
zih%B5NZ3ljvKaOTnA~6+7?`|t4dFEc7*m#DS5BN@G#Mif8*g7a_Y&7`TtmMGf5WzH
zs+B5X7!m{l^;#XvvJhfG7zR|r3ZCm>Sq?&$wl-;6Fz652wk2Qu@*{lYhklIZ<t3hY
z;;BEfUVof5{_<D9#`@+a`~4BS`}_0<1Jai>lj(Q-w%^xVq&Kq0XV0AF)TxtLQnN5W
z&!Q4Ev_mzP*foQ!O}^gScDTzGlac}VeZBP>jYhy5TjTFkMVuiVhXqkE?GNz&|Lxs5
zZxcZPNAc%fo5b-tK7tj=2a1qr=qM>DBJs(QsJJ8u1xM$V5l1YHW8-z!yL^OTgyLlM
ze*I{Q*ZiB&G}<Ij9<y3!EZ%?O=d8vohGFl5B-(IUuM(eU^jmXCFZi(9&|*q7=<t2i
z=e!D!&fah^Uhv{u37r#;9<*71`N7+lubHlv6htH>boT}v-Wk$9Jf_=iQLDzBy?RHz
z6Y=!)l*auNX3wATtCNsrE26kcoG<yhUned%^cEY++5yQ?gLS1O+K)-IOS+#^T6cRa
z_SO_sSpJ@p*SZXfB@gc%^D)b*<tv&o<avdN3fVQ!x~qCs${qZHlcp)7(dfFb)jt}K
z#|($Vt<Ko^cCm5zY#crthwrM!I~jFFQ81ZIXf~U^TDrl?vSdD=)9?3*qG+qLHohHf
z96lR|&&J{V+r~Q?b&@1W`oFfD^pYe=x{(}pk|aq1aMVeXBn7}xCrOeN07so9Nm2kD
kb&@1W0dUkwk|YJ-4>_vttZ-5N0{{R307*qoM6N<$f~*s|-2eap

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_021.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_021.png
new file mode 100644
index 0000000000000000000000000000000000000000..4e1f189bbe24e74a6fb3248c11b9ab426d189310
GIT binary patch
literal 127886
zcmb??Ra6^Y7%fGE6fahaI}~?!2_7i!?(VLoxFt{|IHeRXP~6?!T?56TxLy9c*0uNh
zkYv`ZGqW-?XZE+hz4wV$Q;~a*MuG+h2lrk<UK#`k2k-f|9ST5x`~G3C<o5O>;31>q
zq2X-h;ce<>38!r8;o{)z;b3b<;brOOZtLvC#mdLZ#X@1@;o;&g$j0XQzXz~7yIHf1
zi%T86odnfIUe_HC4$Jhv&%2*uCAM(y;NTRbB{Y4qkDqO;Ep@#%qMmQot(&jfoXI$i
zRUkZ;U$nlc&c(}`|LvC#i;y3rQ;w};VI&3dKgpy>Z2^|ZpCq;ba^HU!xeVfL!)1iX
zha&>Q-~i?a`ICloUku&9jJxk=Z9q%7j@zD1bKTEI{yOCWzr<t0H&GdBYFV`hz;fzu
zjO?Emf-7jt>(k^e<;jMqeQ6SRlZ8{{E}iQK$!`$;+xowDC$c4mHUB?*I^ZrDe)|8f
zgZ>+Ok52LbZ_s+wy<$JnpN59gIXeJl(B#BFc~Q!mc}^mX#l=PO;Y5NCvFZB`=%qX?
z_<lEe{Pl8(8v44IidA-n^<mc0+1dSacS!Yki)<*jYPH2X5szLapy<rE=qk8n@BfUJ
z6MP+^4^A5w(^S(gn}fS^ekFh9c|F;9<#AeTdWSo_jy5LS%l0xh;c`Fxq5b@RqSkXC
zAIMbxK?e~KI@keM1ioUUOd_H|F4Op}WL_Zyhy}w>V4b&N7Yify-J%zPb?iC%3&6cs
zfnbn4qx?)9i*n68*`q%)93?*RL4HH(0gaYc5&%Xr|Ey<UH(z`9LDEFY>-e$D3<t1E
zg&CWKOG`_EE!hl9Mx*@z0{&*9jyA6wU+r+@CW@`eq{(rE^_R6?JyaF$Akg7;Ner;7
z5Z>Q++<5D52!>;yTI)hhuVC!5=1Q`ev`Qr$t_4Gv)(ic4@%<z{7J_>nFK4e^O2OnZ
zFR%brO2V9=nBXU|r;hvZjaTH?wb!fM8h4M6#$Uc!#NyH0KkZH!?M@dnKmAh;j_li$
zDN@tP>M)_~NNfWXCCPJllPIN`)ww=kBQwr%_^WSNvgx&Wd#@Trgm5blO|vXieP4n+
zu1B3%TGh^*RnU%b_kUltqLtK9dO5{`uU4b!?&~O}k;K2DIv0+l|B!QuEtyorXe$g|
zqLK}-x6s2=jWP(jO>A;KG&WCpSIBWpt}A`W9ekbsB0_?$sPseQW#{#u*sJ@(Jr=MC
zJHk}H&B!{?5mFo}qvGJ;wm(bEs|q#owT-r`2Od1icM^Fgz^t`QDkQ8K`-vj3<t-)n
zp$d$vxwhdNY2s*+mm&oeRDiII1PHEtF}7r0e3yxWtoh5q?Th`5Q%9WimbPA;EOCqX
z>C~a#vXsQ)@}}5R;;TlQb{dCn*id%h;%Wk#zWiqzTiHySUe{6XgbWqBUWM*g*g{ae
z(Cz7S>DgblnDKXE6nW}pHNRVx%I3Cbhf=cz0@G%!Rch;6rB5bioB9s$E3P@9E#{2H
z($>0_?&$I*?y5LnaYD)AFsB~!#y=B?%mo~Md`&8IekEu>@4~|QJ-+sF;qVX7Nq(oa
z@=M$6=4<DYJFG@fqNLe{g6POW84xOq9gGE~w;gtVJw3bZ*LQyVD~CG=Pr*-7uOS_;
z!LJ`!4Vz`{1~i{Nhl@ju?}}lIYb$#f0~1%XktJ%wDCh+DB@gbDh-k6W=AK{YiTf9y
zpy8JR0c>vVdV^1o<To%7+7rd#bJJC>Zl+0G%TD5>dlJK5p)mQ@$kWCLd32Mh<s{3i
zmS?H^tJNq%?@$n$YLn3NJ&wo~OgBv+I7dt)v;TO#qxXpImFaL-G?lTg<?DLmc;d_K
z#_k3B#Gbd<A{7C}Elex57}XUQZy=L<#%3o#{08UwJ__wG8d7ByoBP8*I&bn->5s}y
zW&*yy`j@WC9H5R#{V}o?39v^Zg!!<cjt(`03wuUd_#~o;x$i8lDi~H0&@;I^BYrNT
z&NhpQN7g%aKxhpyNKcqeyYXMOZ}mRLAoV%*6T7(2=>}MmB2ELCW?38i=Z;f$3>sPo
zxIc$6V)W`INp6*5Ph0^I2>ETG`7;>iF(Fe)mW;Y-9K`4kQ2V>T*!i8>U)@X&F%lZR
zu0M^7zR={rgu?5PQ&F<#Np4#OsTpbGN;ni!8WHk%Y)?C5-n+cIQ`{sXj3_deYEr2{
z1-c#{a34=r#APeq>6c^)du}1xf20-0%8W+_`HgFpx0odwlXM=3gv;n6(Dm3%wE4%P
z{{>S4sDwX^I!FOHYSXcT$QY59@o%<Co4}_VvwGCnr?{8Yt(|gYwS;vUOX%$W3*%%7
zh88)GRfB9O<{0xjF>L`LA0MNTGhf7Zv6fyXt>N4rlZu=r#SJAQ4>p_!1l;$bzoqj{
zs{#}Ky{l?nFY$+oUvpc=#S#GGzGQR2NUgG=hrdiEDo(C9qqb@3X*$M+#Y+y@i0^>E
zblvtN$%b#Oy8cv~)a4u&^XK*xIc&m)GKB^}jWnZX-*!0Q+RwtTt!_22BkqD+MngGT
zVma+6=QoL^h!D$jd_u=6&^3SYfa&kzgD*GOGcXwYQY1Q*8{7w*u3!q=sC4elJig~u
z;Kz&%4kbD~Yaw8M0Qm$KU^;HTdpnnS_Iahq-38VRV$~@7qN{JXH(IKZE#R41+=ry%
z;8fjMEx!49e}AnuErldHtc@jbj<vql=gZws-J!0=gmn;s_|8?hn<({2n=%r8HH?A+
zP;gf2?)aV<`NVqm*ZaQ}GX?=$umdk|-<u}#E%KL+7sZrA${FcNvg;;q7~O>ED`UCJ
zRH>|ex=XkVY9WZoUVVm9gE9%&+&hb<Fby8VqkkREJKZ@5aP&-$CCuIJkJqQEq5H<E
zweUB4Rd2gV)zQ_1zW`LJ#ssO>G8y%LbiYm5Wl<oQr>N*Uu-LDPaJkJpi@FN}=Wh9`
z9r%&~OV9jdV<i}m!hqsKo0QLJ`PpJv@I_VF)A#RfyY_k2?_k><dL^~SihHMF+X4~T
zzL&=QRFNeC!fX`h!U1BpSiVWRG`M;#j{*v2r*YnDZdk&KKUg4>)1Pj9Zcns*9+qH?
z+P|8_8g#f&km5WqBvNRUb|P#^#darRiS~OAo}`dGBU_*M4adb?V-`02QoDYRxwB=%
z6vcu<=O7<sQ>E5}&j_qlb4&1LEF?|J;f|8<i#zp(0%pp^<&))_PNf7=hhCNV3$=4R
zlDV(oWXQPC?B5eu3uh<~#UU|ccNZwjruF^vcl3*D`S<8JKd+^ut3PmM?_&!6xhgIn
z*(n7KpUOaIzf>|v60%B0`MA3oTYn&o{Q(q)qhpIBFRV&c<j-sVc^VH;rq<qVhk#ao
zP85IY=HY=9ihLA0b{-;5RB0F(MSd6<xmT}T&I35gPU1jz#K^-T5H`lvyRGdsb2Fdd
zh)OJ2S5>N;4|s$y3pWOLrV@_396tEV+un5(vR!Tyj?dWD5B^JfD<LlX|8#|m*=2XF
zhp38VfYtkrZM%e=zHkt0dAM))w57sBEOxjz-`g5HoGQGQ{_qs>%6z=d!C<XbO+|fP
z$2oa+-1h0+-xowzw3n{}s@Y-7eu=nTeT0ydy`K;JV|61;UiTq-kSzl-iykl!bjYE&
z%V|kcs+2lZpux5LJkOwJ5>c?SW;NLX4aW9nf)`%`->wHj2}Q4{875xcr!VCQ-yxdH
zSAa}SO+&}xJ$dKn`D>cGmbk3>n|#CkDu9|yrgfCc8U{BkR$auqUCkv{T}guo+{r8L
z;C{*<^Je8;fY=E0;DvEq6#IRZ;VWT8GNc%IUT?C~`-%?-q{^Kkyo{xDcfX`POZ#Ju
zA86Nu_OO<g1yzf+iv78%0Nm>|`zFrmvi8Z(&mwio;(kP_rmkHj?5Dc*?0Gag;q0Q*
z;o^Sf6%{@D&M)k<H#IVhGK?`g20A9@f${Ozo^R#<d~vBZds$GSLaR{A^FS)|2T@^S
z{yv?U7J)7Woq!v!MqHA7>FOi5FMFbdd}oZVBm+8BKee!K)0Z`)XCD9zx>ej0N3n+%
zri0$W$5HYy?tx<2S+2IzgQBZmvWzW#U1Rs^-}^<01c>E>QBOD8LqTLgO-lZ+Z>6BR
zzj=2~Fk;GVs@v>KkcLhu_oias;f(cppqLm+Zfomh#ZkF)d38-w|FyoMio@s2TCppn
zKPrx!NM9eZPNz;t)vI!3vq<xML~1op*Yx~Z-Vyc*xo<|+``xG%Yu<l4!L8P{V;=OR
zwD#kbMIdnG0+rQhQALh1s}UEfCmKD5JVX6iSa@|ebP6?fC4b_o7f1dJND>8w*4LoR
zUbVpOOH7TX!#qC~equU!5l*CJsVtxuad>f%6j({l=_QGshiCpkm*eK+fo@*{IHGiU
z5l%^io@A}I24%oRpM9geNa7(mWy{3z?C_?MSFl<i`Ub2D_OPXy7v0O`u;$<w`zlW5
zmEKfrDC4@%32)2o%2RyDvy9flM|m91Pw}>M9|a5?hvFod(l2!Uk8wXVD)MVr;+rH0
zm~LR52a+?cZ4r#(WN!--%5I9b%IZYYe-4+(T7#}A^^5pH5on0tkYckJk6E6P$lKY?
zWpck-$jtT{5jNf~UeO*?lNayasYS$stl-MH3)sf!_XMDY(0>3x(l;?`JHrd4na{!g
zbNxEyUb)2^f#QNETf}0o3an~Z#hX99dr$NibOl^-Z6nditj>TGMJ$pkPn{-_GQMP>
zVsS|`JZ!P-JSYC|m#@!r&aSX|vYrc3US=+i!1nv=velNho{a~8W7$FMI4WpAxDYqP
zgYXA00qxbJ9vUG&;Y}E&m@~lkD#Fx>jkQqhhQiE=OC`qb=R5l(AAsqB*X`m;TaSlu
z-pYE3!6kp7UAUF`2Xl=Fl=P)Fu>wAoMd5oQ>p!fZUo4FyOW$<OLJ=)t+L6+u$1+D?
z7Uj#$t^pD`!X*3~v&+k7876^R0ecKBwCc>lo|Po!i#WLi94eeUV0D9gZd)6QL$=@X
zghWqpGXo=PcamN4=<p?O*3rjo(fYo_3XSrr%4s{EQmwMP9KI3p9(I~HR*W!D*Y|xE
z>3DR6Rn|IgK+b0vRFz)CPzM1sc>>juXoWqJ-+e^y4w(e(Os*6E9&fe7{<?)R=lS3f
zDv|laLa^tm?P+`%0j~T}NH&Q$945Egaxd6%b<_dmZBtECQEN{;^N%nE7QH_S7Xj{Z
zsxxle@KZ2BE-c^%;I$lCB>gZY*eQzpuSf-}o0{%>G!CM$-}$twR_M77t_-8S&4{fU
zcWYfe9ScWCh*nR2#U`%U#w}_OyWBSA$Zvx)6540;!7M95;=<bReiae0$D(E2h%ZcY
zBspk<NfvZHnNQ`4C-8zCSo8aXDO3Q-&e5MDylh>z(ERxERAf0wQa3^}3Z<+Gi)Ln-
z($`NfePS=pL<u*H{-QbL>Xm7zQK+6sE8dmt#PPaoize3oANTx!QiyY4>3AAxX%2ha
ze4q8nX=zY#gqpn}YFk?y`M2?)o!p?Pi#n+WdTel`muFPh)JVVsde&37@uPY{wd~`;
zw;6hsd|??YiV*~LMsfMWlnaAbD!DxdCX%K{SL=}p907qV*y%*f;wqAQLcB?hK<5r>
z;;7QnT8m;y+knMl@rKa$?mil9;<I`wSYaE%3$wDeWuxDZmQ3swo_WL?M*`Im$p$X`
z1IdtUu1T_)5z_23<cr~IzSU+@J&Ac;tc+Vy$<G%Z`(!;ch^62$w(5h={6K-)bY2my
zCj4*4s`g7<Rgp(w_2RFSCQ7F>aI}!C1xd3|3b~^;Zkb^)zsmQRsHxF2dUD2}Tw2(w
zEJ<d>Tp4kT?Pu<{URf)ASU%T~Tb*B3HJ)Y&w4axWz^t<{6KCcj1cCH;#?>=o(vD(w
zI59T4Pf@n9Ec#g<{Tc94E&!Gbm(IBn=kv`Mp12%rq_pd~4_XA;KiR&NY7UzO!hc{t
zO+3it#d2yLsZcztX?FcO1IsO*((xXIj)v1~AiAFJ5st&{TS`n^t(Mifhx<0p24MUq
zAW=oi)9_mpk_V-KctFNu#vq1|9^<UsXdJd~zuURvge$f_8zs&g4`wC8*+h>_A|{YY
z6R^1pPrAuXuUiSMG;WDavHYPC{(*5mjwrr=we9`w*`Fq_YgVk&6StSy9Tv8FH6mgA
zAx;CjRQ+gp?|i+eb;eHlQ1PbQ&qzQdF8_Czd2Ah%kuq5kOXT84DtvLvj;zLh(6%*w
zd9&f<?cQZ3Ma9@zLMHHtr3W3LNVQ|deDTO-2ddrpF#pjZM|7gJc0#R`B@8T2YxRV|
zK>Dqv2u<-?7>iBrdo}A-@SqVIu=|2D7uyF^up|0>>IjD0YFeFEqU|T4!B*{GZTC!#
zp9v0<fP!M&N~P^hL#+zC=A)HWpo=D!mKl9ou~xOR#RhaS!x0kyu)B@2q1{8|C0sVC
zDTj!T?{TC5B+;3Bd7n-$z1;?mj~`=6;}M@<zO6kYl&i7K^6Ba<=e0KiOK0I|y73rr
zD2kv0o-ADWMvO?lNcjK&>2MMXeuiuoHyazHund`vu4IR_>Ma1svwbMK_RLIkfzZ4b
zsFb!DB^p#-3O3;+Gf$qi=Q&+#ks^YeC%87Cu~vywx?`rCF&DX6X4Kd71c*^6r=Css
z<dy&^&jWM)+QzrL;j@m894f1R|7q@i_=|>KL09Z1Uk{=N0DXpc!act2yoyPgR?v<?
znfTW;IEddk)~&NKB{BK$=H*@`scGq2Zx``k{vYbbKeP-jeV=Lemf?Y40WxIY_ADz6
z&RD0sZkutNjz*eV2N?O5Z8#0{d~j)d%yTz-Zl%fr=K+Z~FG55t&K)&lCY)VZ2JAYZ
z6bllr3>x3=ZLXZS(m4_rmY;Zfjdr0fVC)I^pXTcC^o;diLy<fSr-&*<P*lD}a4R1E
zF?TW$*uIHJ2zoSbeIsK>!ngmpG^wyG-{;iWoJ{h#4DmK@B8u<j#U@XGiL+v8e8qnM
z<1WzqwrRRfBQje(u6$^*$v@U!dn0c{h~k`+Ad4qQxVF37EF4&?bf7IC2FU*m?csf1
z7i9}RF;=f+ad~*+DtF56xn%~>8U&b1fS%%f^NcmM1H=+VBD$BGpaeHNIrm$*br5WM
z1zhiMB`6Gs`Oo-qi3HJ#J`pDVk7Oq|l=OCEt(R%(p^3Cv<20XhumOjA-`XigL|esP
z<t3m=xD(8jGH~)BE}EjQ73Wpc$=vFWEVsk9qt?5o*gIf;kql}m2j-9Zd9z4px<-7z
z8TR}rW(x?HF;I9y%<FDXID{iX{Ck;MS|65NV9;4ZQ2g4-NVb@F3f@P18e7SR8Dw*8
zQ*rFHN!@OnX}QQWH(E2!3~YMCN`nVvZ>P4g=$VQCE|aAy2c8Ps)@XUO%x~V3YFI=5
z!#Ko#e-BREi-$oSoAuMv5eW?oiDF9Jq312b7u6kw+_^|3zcp6v^`H^Tu+Gy{UWIJ6
zUtidl$KfM)3c!-+glFjDMHwStFtf%JJB>28LW+8Yj}Mc0G$59iOUgrQe^0Eb`&sCd
zY#}ON<;M09{RBni&IqH06B{nGlKPR?yHwY2X`SOf5Rk$^)i7Hed=oRKc(QMJErGW!
zQ9{Da-AV8Zvl6|(@WmoA0U=@Z4r5dlipggtjEitw*W+W_Uw0AVNf$odHopg2oY%K*
z>iGm*Z>V5o!D)W(?_;rei}P^ar~{+$VSa0?5`%?N^X7LA_jhqpc`_;_$l+Z&-<2LP
zqCWi^E0Lc;E%`&Xvp;uBgz40DJflbEu+d}pD8hYtK?8s<U?^B}_+8W9HIz!PtEXq-
zeuY@Z%*_4QQ_^@#(Q5afZ;!@B23b($b{NY)?v_Cg8&I4}U97}e$juaM3ziebF}NgG
zc%<7bwlh(F8{hJV*sBEtnfmn4#{MwGdFP(e-rm}V#o6ueKi^@9rMRiW<*2^d6p{T!
z(Wv{^rX4nqf_mg+4I$2WY^H^*ek#T<BoTW|T3|RNT|!3;-D!j@N{EDSS_+%4p<(Bh
zWZ2KJGZc&{1icE{kR>7Zd8+q~lPZmL8Ba!~b4Su|eK5hS0wZcbu3p7>_cpwJnZ1Lf
zW6hsG!+(dFn3<Vpe|-oz^qCT5&crp4K_Hds-rv2!a<d@+G0_Nu;vj<)$8c}FoH-B3
zT+VJRc^CAAJug_UWf2OIE-Zk8&=lno53JLx&C@UIzCwk1h9<Tr_=;up0!^M|rN0_A
zh+v^vosm~E?61`tV&h%l?t*%mkCWZ|(t#+#+?nWzb4b6&X6nw`WH+nWfeC5_5t^A4
zZZ{elX%>RgKa(JH@87CLIs?-fzrPG;3AmTko~74DyE!EceL}b=ODylVe`X(8@$-`c
zs_}euZ6IjH+It0-_`0d=b!tg^y0#l@+&EMsdWC^22bpeCwbqLB9L(wx^h@SBy%3q;
z!C;Cmc{2C&?rH-u$;oNNdHN;KPm#w>t+x2520wp%{SNwm(R=ScYp48Uhht?0Yjx>g
zfW?@3rc%aVoUuNV8K0PJrA4vE{_#^wy|MSa6QU#LiDHVHMu-+ca;YoEQ=LBmPm^C@
z7Ap6dkljzvo_9$1uyI@SZv}Nnr+Avt{nJ+%15w$*F9Pd+*JW{E<?e7FddSY&9|`nJ
z$mM*|D)7q3nwCL%?r@{<dh+E-^5~Mqm6S7&E?diTXn#4A^8>I{?<j<E>1?vxj$~JG
zIxF@b;zj&=$P~w9q{ma?z4?gUQaX@Kxgy4@3&dggIcI>>O7}X3T42#XE8X@QG8<@z
zv|LP%ifrrke#2d%*aJ!I81B&sf|sLYZBy?XcP@4#76S~ChQ@$Odp)?Nb(D)q^J)RM
zc${ln`*UKZ!s|QTeQY(Yml0*f`nQ+pJsR*)`x&Ifzuij^cQcIk0ofp`ig4)rkkC;Y
zB|3!dsU*VSZRZ&sovvQu$H#@Hp!>k0?7WBp7j_fDcw6rH9y@pNWPT(;De3K`9u_mF
zc#1D_zCjf9o)Tgl*(qRMdhrBDD;%fR8;2UdKkV;ciQ)KJyK)^?l7h3B7sD_}C%OB_
z?8@6{F&2H_kbP6cf=x1pbkA)MxAZbXoXMK4(TN&r8HX%l^s-NWF|D|<znWqBu*NG$
zT+>(wwD%oKUV9GVo{=wRNT=!#f0x5OZsZZoNPOXuZl7-8bu-H)26W%TklzEM$O^*P
z-28FSICj7>twTL#k>Xk3M~m))HU$HU`C1M*(zq3g68#Qn_6T3nU*)|1RJq^GBE>y_
zIo)WA#U=(aHA7i#e{^)1i56vxxTn?ETM~9c`G@}gC4tyM{MugBN-x6=k4?y!ykmta
zaxM1+qW+v4*1{1E^tjTphS%Xv*f_2GM2(~yQOB-iuj-iE=0~z?)Fmi{TzDwSDiwaA
z9K+q#<RndCw@qK+^ZMpRKzw~x+x&v~F-vtbiZYM%ET)OtoAysrs;8kpNd=eigh*{!
zdlMazIpQ~5fK2^*k*1-v56|BU9+Gv&xXFLZEf^?#Ng!+p6`B7r#++UJ*5U=)t=FwB
zHzAJXcnW5OI71Z3J?@A;B5%oI+J>751DRaEV?&YTnvR}#VQF>sO}+@9o8uvDoI85s
zc>x=_hiyKrw1x=U65&4I_CuhwhGzYc8o<_`9~*z{{YC=;Rw51YMZb6Z_-S}qgi-&J
z27PJThyGrgA^Zwc?h`yP?X@4VId`FJF8Aq_;<CpwpTOdZ1?RH4@1F1oa|!lvRyf)~
zPW9w{cI}~uQ-!NvwXZcC>avB+9!VCMaLVR09z=Uky7Rtv!p-?saAbtO>FFhR)7RX?
zO2=kiz}GDNrzR!K7j_-LXVH-nNbAm&WTkPb0lV|D!5NV8&!0b+Ut1D`E5#U)!_|=|
z(AB%;5ID@AY@xRs>oWF=nRcwkZDW0e<2$j~roUUMkP@KBYFgg1Ah)0ol1&svHen2z
z)TC6pcOKey_@LyyPkH@BO1}<0Ztnk*zp_?!yoeIAWf!>D4S3+!{QfQ5WbnhFf<h!w
z)0$zTkW?nl520He^f^lB6Gyz<zZO2?HvexbLfOHW=$iIfI+R^@?3J8!jKKcNnv_=6
zZGJ!AczU1Z`|Pl<`DjM@95lMaKC=TFI2}(N&mEV;e8I63!7s3-AeX4;HC|97=xpmH
zw9MoIz6=%+cuOX}b*kV^YoWuOWQ5Gr7a*VT^aeZCm>-g-<p45PW^rR`^IbEN059){
z<I_4_o4V|Rp?eRVB3_%Ln4$iIY-(EZ(x?l1^*W-SxQ>+9j+%RZKw(CLpNC-7v6I+l
zfBlHZvefnzcwnOPcT>A`Jt*I(st`<dP7`hvHD%MgwsO1ZTOALRs4$|83n=(GbaO+d
z&RDduw!fDil;xgy;<gpZmJg2Nfj2dTfm{z_zSUD`OpM4a<0#Bzvr(SZj39VeW%T3y
z_81;N5vs##&emC@h^9xB2*Jh^R!uF^NTH$oHXJvDQAELv*`QWCZ(sI_eBhgzMUOD`
z!&WpR>EB;PiTJqx$p$sNa@di(Hx*KX(06-~%eX2Ecc;k7O0H#Ej11-sR1S0X@nQhM
zzKzRNjYGLxaVU#9_pytIN9XBHpp!v|WC!2gOB*k2q7qcKPzoNIJ*xCy!J8BjwBROl
z-TA95ctZUo4l08%rXv*m?7Im!9Og+3f7T~`x_CFyc8-@V@Jb3#?jCogn-?{|tSc}m
zlGV418Z)Yn7{y7(yz@~ajhJqRgD$q|`h~Xn*-(WwSK??^KtE%fQCXO{>1%u<V=yfE
zqCw!bmoY#Lj&wna^rJr96)fx7XuZ|cYAAjtzvbx?IgPl)mF&C;WIZKux~@#3>a$n8
z0XyGBJ}Qd^kI)mnM?O9~3kQQqJ{P!&n{YBU_x==v0|i(v>KM~0fCNui!@i&RpSd4B
zSds<*mC1P=P2BPSVom<%fXchIrQqW{Qdwm$*j8>@Ha4T47`0nIf)hYvs%B-oM`Jo`
z!b!aI7`Ek+cwqx1(!!?d*T&trN(Szr=Te|Wr}k1t0x8RpOq1se93)tBJC5Mz42+Fo
zFVBzv6Jv6OA6Z_@I<6F1D!J9Bqe$lnfMXudUDL!!zANK$lK4~+!=*U)xnq@7k4f!a
z7jf_CddQ5Q!-FL<NnIayk%1%$37EuK)DdTRU;aw(0UR~hiF?-JC_#3R5G!*CjlzM)
z1I^4E8-67?>v;ZTbhoPQa~w^m##v-zz?@0Ra{rx%@#=R;p<Pj_`(58B^Pxjis-9(x
zvi$n$;r@9fFQ?!vqg!mwk*$}{)npJih>>#$JqWq&k0Np*6GSh*um^L5S)h-8;u3O6
z8nv<<H9P~SuV12LXlUv0|A@EHWWNl$3qn<eBxT5D#&a-9EmGXtrtr19r{+GDvK}!r
zhqUD4AXUv$#9+FY;f%^ff0Bbas(8Dct<yR;<@47KlSGUYVXl5^3w*_tLwWq?kILtC
z?CMt7H|*xfyDrZZJx2`f5luaeuurd=u_fQS53{OOY1;m;M|xey4EpKoau)c9d=-`v
zD>i&Mz)BTyb!e65z@O919;uwETYUz+B(CiLv3B<nN%C%7P)U?i(QlDuXHDDqhySGM
zqG*AhJ+3Te4j<h)%cIb3^v-u+y-b?mX}f&@K46P7Flst0pjBU@@WBX=v^_VVE)O&M
z?2+LE|2RBeoyxhTz^i+gk4>ZR4w<TmD>AK2PI4pXcb<El1X53?J6lQq(`A;ALsM~7
z?8%%HKOo(3x`JCaXJp<^hO$Uxd!P)vC2Pz2?)iNZdwYA27K8vXH+G*2S?tI=pjj>0
zst{P=meBSb7{Er6qe2|-=h0%gJ86E^OR{#uTGjRx#gfb4xBuW9#Xzz%nYt~0er}UW
z%!f?)WSc5zK8V`zmU8%q8vgCVC$65u+qv~CM6_+~3+6*585?3mzDx;m&e@EC!H=2K
zZGSR?L65K4@M6wK^yEXF<ZGO{pZ@jWgu0_8B$XF^{Fvt$Gmi>HV&#&U6wSWeTxZQH
zW0t|f?YFSm$qOL-w@5sBhMC&}KzmW13g9}@OY!p3wNFm&m%1Ta-Ckci0yA0K&4uXE
zT6nf~bE%5Tw$10QeA<4*(M_AO;RbXkb*BGI`F?$EnHm2^aOHfMEQHCpNE2)uRxD&|
ztJO+o+6NliFZ|0L9Mr2SD0^$hMHG#Ay$Dg}FO*@lqTW9}jhcI#1^gue!fevRvq+v+
z-FpokvcFIUEqcQr8!qDr3kJhfO2JIz`u@;Uvi4h@gz?594a_Asb3(KYL&U{X-Mw=E
zTYl}{s{lD!Lh{`VVB8pcGedBb+xL0m1!se8tk-kn+TS{|7J)$8^!$cQ<-heLwka;R
zH=lpA>@DxvlYSyA5Vqm|)f-J0&C}B2uX)6yrA>JF59_|4{FUHNH23joLhx6F=sNyZ
zW*uqaGUX;xw}-Gk%D`crP7!y6DI9QG#~@$RaeAp;`rtEr!b*|uR}e`tsrQeKC0Yd#
z7z_qll8hlcVP+QCl)HrneH{$mAi2Uet8Q<<sPTM!E~-D6IX$lZ)BYx{HoDb+KU~9E
zh531RxZK9#U|K?t)-Gfd-9z?mDvn54<KzL_P9wj;-s^$!>BcX}^n97}B2*Ysyz_je
z3rCor$uXDK*6EaSS=+obBs%1<fxqTB73_!-&Ai?os)13_+|Madzs|F#{i<{A7sEvQ
zoF{`yZ*OnHrbI~O4!-<elKeuosX?{KW9un0$Ve6+5zx_pxi1jr78xF1#-MJQLi&za
z;yd24=jixanZ*M=>Qj<c;gALCglwseBV)6ivK6^gLieEvhLj3a@EQU2Z}n*oW)EjO
z6J>EOA9^pt#Dxe_3d$|=YO`<dHN?{UmgGdRzFuH|CRWq`WjrA<@OOWKwWuL=F?0=q
zM3E$Maie`~tz{v47-vj2qd1bp{^(0iz!N?M2vsSGf*}>z{yjZlpuxZoe*wSEKmt8a
zOf0PAX=_1GZ;yr*TCPl5<0+?&ZB;=Lf(2e%9^Taqz7*o9d3*{@j`gBEBa0-IHQjx|
zlEkYMVYc2Y?u<V=LVFe3!BL}ezI(;rdwN6y*u+aY_HdQtL;^NSmJNzl4tE~gue_Ie
z%<dA{<FTy>)M7~MpF>^#&T9eV9?5e4SxalMxnyLDhs+XDH~RTzy<sS2@lRoY-QBCc
z@^J=aCpU}-f@MbiP6svS?&lUgK*TH{)pR;y?c}3@-<c52#do**m{M*)(e5F=irZx`
z3A25y=CHmJ9je5x$sP)p!Ke0(vmFJQCIc(;KnSQ*GI+9Np(Q5i>tt$D7q`m6`^3A{
zs#na8r-s*Cc^nGP$jN&-S1<}%PzYFeJuivhDDOjFY#be(RQUPD$QMnx!V4ltZYpBa
za~~JyMD>C-`{Wb`>|R+=6&rTsyHXYPcKC_+%3h>A-btX;WZifi0%qE7^mF)F;P3xw
z*hVC24vQX@eYx=#SHOyvOx;AnE$hQ5jm8_BV_p%yee&gt`gx$|kwhme)VM*w&tp>8
z6(9O8neg4}1at+Q<g{o_koe~8<1VmvAQAmYMZ0!`>M?FFl{E1CTS4pRXu~LIa_A6?
zzu9JTHhL%0C=+3RjR;l6hb*qH=4rCJ`}hoQZBa@vBx_Mu4ES)G&NlViR|P@XUXM52
zitm-0d@X#<n69S7li)3YnRaz5q?9WJg#;{gSt-e(fN#U^g}HmS1`YErDD_kuJ+Hp>
zr{?c?GfB`O>iN!+R3uh(0F~KEF?9{~5tt3+_^Du7yEe8eA`Gig>;y4;h}$XysrT?L
zjP^TxH2-)a&%9tuJ#yLSXQQn2j_WU#21}HpHG|RdmYHM7#e?tR6H4&H7z*?lU8}`R
zTx*7=L9szqszkChLlI85?@WuZ>C6Q)Q>=&;=6A}1&p$di<m}9r($|2@E_Gd5nHPLR
ztl6!MUYM*L)v?PWWm53YIqq_3%D?=q*AiF`B0OZY(2`6}&cFdSv9vF&Q%2*x3Blj7
zcvNP<ZW*L~o`j7hH1iB~NLh(ZQv!dd9C*e4ICbO(9MZF<(Ba7z;*5+i6J@#Q%i#3x
z%Qq(D^Mn+vsH29e4__3!Yx-$<ja&x>5M*#<W-Gwdgv<}TFnvzmJiO;+QAu>(`7|5e
zkJ9Q9XOBws=3xv)T8uS92Q@V{Xh$P&YE&472}$B%T?<%IlrCD~2aZB!l7;!3)))+t
zb%%__*E;=^=2%>54nKrv;IPkT{D<lOZ*li$4hM2P-8mv8l3=V@`A-X$0qhU!&Jupb
zK+`(cjdvxUsu>uj!305t&*jhAhpsb;UBOT2@ZmBqwEVdeC@d>Gsx3N5#TUaauYw$s
z`uzJ6f-6d_0nX(M&<%xrrQvD4x^c|$-{PR^*d2EWlGBHp*0OYj_m%4v#D~T`B@iT#
zE8|d%td_jZ8)(17M%(b-t+AXnDnt(g)GX9ub>AvY%@%4I8os1bC&C?IG%fArRv=yv
z7SE)^Kn^z-4UBiI@1L5YTs!rA5^ofGyMFE;8|~wK#t();FI6uhMHp~9`z;^S7l*P@
z&TNpK4q7-$uMQK$QMyeWgj)*3f68v6{%}Q;R|*K>$Q0r;Hs^-!4}}%cAex&43OV;c
z(sm&5IkvCwvYEnFO0XR#XSS*wS_ENU*n!mHHMN9**aKeT)}t>h6Mzl6T*RXnsHqRR
zq+zzV9Ig*(@y4?8emKQtX?}O*iYVK{IQr*1^_R{pfiUZ_n$Y|KA56g+ni-siHxDRl
z<6bc8vEhK!vDW<+6RqdIw#4^BBRAiu)U`3V_vPi?&eXRqlefSRH_3Y~EiHcEOPY9D
zyz;E|jZgmu<0zTZi7^I_$^ZC1u6bV|<SGrk;oF;Jw$Xou-+l(`3EAU7$#ci882M*#
zX=nYdXb#htQ_y+t79LC}Xwd^WmwfseqlFtfR7<0XEnc`}VQM=E`B*`zjCYIC=PrMU
zNd#Xi9C698TmT`*jvyZgh}zoz0Y0I;4kcbamGEA)3jf4=ms8@i-^9hjs(pdbw=4^x
z_}Azfn6+i|WO!4c?8N4)cpvcd=Y2@q!F@XGF)ek@zs83bZ=c&S=V{T_H#Qx3<4uN*
zL`In7ll|5s5eri;t*n&<E?RMua*$v$%HUkfLf;G%m26>&dc7VS#S}V^2a%+(Jk5dy
z@|w1m3IFsrFZy$Gk-5?1jUvY5iLlSmM}BNy)TLdn@7Zm;zcN>1d7Tu$j$cw{5bQ<z
z1RYG$Fsn;^LmMXA^P%jgVZaK5Bhc3RsrivyKUl(MaU(c{8(EbpKy~j^lMF1cGj$VH
zjfjRMdd!!)`~3!HkM2UQebDc_cJ#}b&habiA~ZGm-xl33FQ4AJ!?^KPTETdpR0Uv_
zkO@Djn~O`Qg<$cFXi=a~XWj<AqS<7oRqgmTC|0&srcK?NJHy3gcJIWSt*{kaVeoVu
zGx%JnLxLm`dl(9%0Ib%^zBApWhV1w$9R9)$!ruLqU5O!}$XZT2m&uZAm;0UMIa<yN
z-!+AEtYo|;{zK+ewG?L#1$d-?<}l*W=NmmzPNwzt)z_M+5tn(r`313uoCz%EAZ&Fe
z{nNFMcX}{d^GIB?9}ALaz$9mQH)<*{wvR7Y7WrL-sd$oiXfx~jJ&gHZr9}uXTB@=D
zf9qQzdE-7z464(o35T~O9?O+}zdJf6EQEgqx`>wq_IhyADbtx?#LGhZ_K*k&(rrF}
z=1eF#CinDtT)Vw!1sFpb{g044?k8SwE*cXx-FSht4FAadn$}~bdk+nLRnE9@8Q=@?
z=q~W71BM}jLDwv8(NtV1G+aul_>SD|XUV<9Lbt3OHVx{ba<sq9rXx%{!s%Vlyl;%M
z&Ds71waz|2T}_PP=Bt8Umf6H6(u4O=hb=eWUqHS@;5$GK_h*-6!1TICjOrC~t@r((
zy}S~vy7Y65U<e*C(YEt7nN5Sct{n(Vlv445Qfj~XTU@5Q#koTgc0hJgh_c$@k7vro
zDtwQ~7WB6k9FF_w|42$k%1Net;3fEW!`h6@EiEG}Dn65p(ZR!JgxWAAfp+o8Jr@ye
z;k?f6*8~#yCXT}mZ}rM)tuX>dSXiC#_^*jgeVE5-{Pdvw{BZf+9h&Ef#sJ*Q%u0qW
zuy^A^2nUN?3xO72FIm(P)fpsweA?0f?)dyY{cc_pSu=m{@__<^0-}WQ9i_e2?o#Ux
zj~(4c50C++<#ECyfy!6ZV(!V$YpnHp(2e_aXSYVN&GofFt#oaxy&<ZZ$gpravWHBG
z3G$zP=#14=Qo2K@&mUJ?9|#0LagvawD4A<%HGGe2?xj}dgu1v?57=;<Sb8EA3wKY_
z$I`P@RF3x3zeO9c)8ylStR4MsXi;tn>{Yk!T|Wi0UoHyk(n0->8@zo_^o?PohxFC=
zet79T#1shnv#VDy5v?dT(X1o5@vSdn;^sfKJ~Dim`o#uvE}#5F#!l>cc>5raYP5NA
zr$MN@A{<RA<13mv6ag$AcRh`%C#XZ@d;VrtD67tSZ#uXQ3P+#F*}-}p>*%(?0Y=E<
zq%4y$AHB<c8MW2^I+?QZT;bj5eH-%J#yaO_-mlE!V0A;Bc+rrsGIDXbNLo4raBT9t
z{zSIg5;HG_93EJmqF3LY=N~eKI^@xPGagtz$HG=c5kV1;FzqusJscy27}<F%)eV9u
z;JVM+)^;@XsJHZ-2IZ;<6&fY-*L<;F?3h^Za$FC-;%UE2?SR|9LvrkFo^Iu!+fYgg
zMy{|Xfq=}wE(voZE&Xw-^~>VFaLeoyxs*#7jNjTtOTRT7P!r~yYVeJW-B@Dc>>-QU
z*}=ENCDe*Y?C`!?-rO>9;U$(jZ~T!(L*?U?az9&tgFN316TJfAlg<L-4v>7tHCL#C
zb&A}x6ihv@MX)3|x4U|XCw#w)=aWOBz**o%%m%RDi(x<;5{_gKyYT2X;WXvUNU`hT
zo?mVCji`@tb9{50t+-eBueFW0OM)M&-2OiGnKutLxz86=jDKX@FE*pzhxrD*@Ys^z
zWpa8ut?rON|2<2^@bW!vrGk>;Mj}T@*vo4umaLdgioQAOl0;wc+H+9??>pyp2hLkj
zX-7CCT*WjP`-U$$i$T#-HveiopCmF8cqG)Ug3pqsOgb-zagiH%dLT@RSQ($&N5Wj8
zv6A@AA7O*#**v|YO`(|AZji4Qg1e^`BOZPzZyL0gHi3YrU+CENhaA5L1$g6>USjp8
z2trw%qp&b`j87Nqf^AszQEyTFX{FSnuSa~CgPhUi*SdN?xyV#~9<S4z0(Sy)TJL-b
z^F}1kc-bDa&gLI$Oc(mr%;G7Pe}ENvyw1qa6JLhTrdBq#?gwb;L}|{sBDjgD6*Q!(
zHVSJy#<m}{$c~Yf=w^sN4Uq0E68N$4ei+RDRnu54{g&-pa|$jHd_9g;y{H!pNS_kV
zvp}VltPZ09-e*(^ziEWj1cdrer_ODCQxfU=e18OLYJ$qm1IHajRVuK_8_Upx^e80(
z<+BXvQ7zt0e)papye7h;W}GCot!`n5MXJ0f$-U;wa&rIt9JREx&%vtL4;F_JeS4M#
zgYO7)&hQXv1CG}3$Ja4WZg8Nd-%n0(+Rtj*iSuWY-hzupzK4lE+ZWc!OyAW#Nhdmp
zvK_9}xvnN$R!bB!f?1J5$xV{qEVF<rY0awIX0NoiLLnk3$APzdZl%xDY-;wZ;QM=r
zIj07xiuqmtM~_1vuga56sdQtplm%yQI|eZ+Ky-;@0lSH1Mc0g~SpOR+&m5opJE%8L
z>2<k()>kT-Y5kESwAV^4TWVG;mz+%I9yBiqrW;XDAC>9T#dUwa6f(h`rjgK*C<akt
z2*X&EWi>B#3>-ieG7iogrkunnG#cV>fgS6wL<6X%_!o=jFyHOyl5arEhj5M=8mR|t
zFiYDOX3p7U-^v18G7X+I<_JMXAU+aeJ}i^!^0>Wm60Nb7i0SXmEYt8_%7~RBZ4o}{
zr}h1tG4I8Vw(hMi%yvRb7a>4y)1B?#_{`YFVheWC{g2cM5MuYR1~3_@sa`T@6ZWVq
zc7J@@6D?)0SAiz^L;i59G|7(E%(i&k+Bo3rMDT^JlcAV};5^2IhLb^yTqYe?j=|LV
z7BUlaOz936H#OTB;SFa41fv3vZilj=vMs8b*fx7pAraUlP1HDa1tJ2{CeA@<!JNm(
ztCoVq$;BM=Hgu_|oNQ3#)Cgexu!gG0Sku5bHAF6$oo)vCQ2!|<e`SL*4$)S4Yrp=M
zT~LfZM~)~mB43s09<iM-3$?;+R{Yn!4%R@uyL8Y77DKwhJtGSQo4@;P{pQWu@{QWc
zODvHDtAR(p<OIIVfSLEsO2Am3om>5KH=ebfT@Kqud~o3>a39hpnUhE)I4EW}hw;86
z@<{}(PkjmaL8UnHJB~$hO>KjPdWIuaA>CekeRIFw`@s9=knCipT+8?LkbY6HMz&<;
zhZs?c<=4U<YT592I)uGiOipPgnw8bQm-zI5RCbWwwV<HTWXRsbAKJ>MBG9=;MQXjL
z#C0+V9y!6TK|&wQ<C4l0eE3$AWa;-Om~Sk1NkCtl^H2YY6R)iM^|Gg#`C;*g^~UCb
z@HNH@ZfZsD{>I?44S{Ww*LNU>Cw2-9$!Ts+iF1b=NY<Bqf3%I%Q$>Y<i5_jTdc6}A
zt+2q8mGnmc+JUo|dHtuCj(d^=@MO>78sVLrX!z*inu$Z&kx0i*#!;(|<-sIz$*mbA
ze;!qV(+t+7>}2F0W#pe&`0V%>!bF%Glu}#5d*CJH(IQg!=Z!i{IW}<U;UT`;bIY)+
z0AkCQS30^p91HxdAUJU%q>f;2y8rre9@2Fe0x1^EDY@TQ><AzGA<!j8e_}6{xl7B>
zY$dp`d@{r8;MrcF&P8^^p~Xd_Iw?=b?eo0+j_)*q_&s#J!Zma|&wWt8D)>`fv1aM?
zPd<f|?<|TJnxY{F0KV!KZN!|%O9+h`SvyRlYR=~-BR02j919au;C2yt?^o9?yku2E
z6;*4eo0&Z%_qC5LPgtP=uTW!1(KtBOt0b$+f?3OYzu0RctLiZ2s-=1GImRwKhtdag
zZ`t3492#dp&yeEQ6qiDfRp`5yGBKS}sPi^`P1%n*iLutOFQ<}NMv_3ubWX<u^FyCm
zjRJBY5T0#*+8JavY<Xk%#U?q;-#%#TK|7Y1x2GCly>JSbTMvBfuP>t|go72sfs+^c
zQ*>4Bh@o-<Oj9}rKjsaJ*V_9kj?P>fgTuqy9uXyo8ASpn-iTSKUHvhzx}GGwD}+YD
zFf90ZO@IuVs7cm-mhw*Q?j40EEZ4yLOhUF}WuT`goy?YiE7L>Qz~OhR)(=hGIFzug
z1#jYnud|jsurIq6<DllgK>`b-EbZxisJK&akHdDIrB}343Encw%B=mHGY|n%G~t=H
zd&{6$+B-&C)l2NA1D<i)FQdcYQtj}`iu1aCh+u16?|;mordJtAFeUcMhI?e1)%{RU
zwBaf9SeB!}SKm!73Y_+lCYrkn8QzugxsB_qoY4J4i|99RiX;nI_*I_0WzKoKU);Bd
zC=;Oos!K4W&vBa!sA!L)?fHQmPpJaJPUtMJ&-&vGmv5k2H+YZU4FsbMziuJE<+UQ-
z%8c*JjP+umOJopVZ!MXmpVAMhZ6A3e-*P8S)Bq{mp87H@p9^&Jf`IM-(V4Em{#K!%
zcU38s(z2qby-SMYE9|TYXSaz&*I2gPWls0r3=;MpoR-!wQER>hM(ra+qv%eoLl4Yv
zVIeF-fYzkWSfcko-WI=A=vhvlxh|fDgzRa>dPm`5r$t&tTDA3db&&ybK||K;{BWt1
z;1R0GLDXKY-6!b174F0pJ83N;<^IjKFU4oAzfElxsNGDzJ;LfWuq$T-G2n;iWU4O)
z<~Ja>-iO{S8&CT!t>+mfHnT_apD&4HWl(;OrIz|CECtS+bFEX*tC4^7G!?nLrufcc
z`FQie$<RMaq9eYh`p@tGqJ(;LAzRfGZ~1CQZ7QHTF(YZSMuoPnfk8JGyvKe&3Qrip
zm>q;gL11wWwtwx97+R$6IBf?$yunOnDw_LsQQ;g2g^6CJvamQb7sk<z{A=Apj>IR<
z@EWpADJl|l|CYhy!~bz?`aAbX<N2t}OWEt+*M67DorTvD<Gr_oA%06BX8pLyqG-xF
zgf~(;gx;SgoJ)bh(T2Gx|Dbc&FN!LT8kg+rX)bCgh9UX4UJ%RY{6G5Sdj@S0ZWQ^k
zk^Hu-<LGHI(b@fURYAD$@<YPtohKEni@0!Vel%35D_}m&Q^k)?OQ<pwvAno8IPoZb
zWAl7dbY2gr4k;lE4(p#$<@$ux<oSpUuU!Av%aU$}vd}cQ81ex~+wf69#f+(*WXunN
zk|x~iEsbzW4JrQetGc~eDcB`MM%`Jn$@2=$x#Rf_?Os#Rxd##A%D)%VCriTK3xDNS
zP?*L?Y+9#07gpLQsH!TjyfS_we5EfKgxv<8dEcEFzw!*PMbs1MKvIgovShhD_R~&V
z#9Rs&{M-%tt)j-%-B78XM<livgGaD&o1HrpHCk2961*+g;dYP7*6NasK2dR5QKfPb
z5yeItwt7Tf-3qcc<uG$)v^ZM_`+Ex>pL|j;ldUvwIIv5bhR`?$vF%11cEz(LRduUG
zFr<`%pW7Z6EtH79*6GFSWNlF|t@w9JR`gFlaPIw-JV?K4s!Eu#<B}5im7>I>Rr0oo
zYMXoi=41shQ9_CsVYDfzJR(|e7w;ozY9<>Inuw3OHcnQSm8j0W(8RDoe0f0$k0ftC
zXfSP!AqyC&_<#%cn9CQZ3dAzkC?BiN=sIAOv?)R?M%)8^DibeC_(Eb$;<%>A4|V?e
z+c%rW<w)Wan|eVjxp-l(*fV`qfizN)%ut9=xzBx8dgI+cuGOY`IYy?7#rxFO$9+dH
z@8!D0uiPK6b!RKDMoYf$XIm~w38zq@iL#!iKr94JIZrr<<N4ihIMN9IS}1hKmuVkh
zm1OHVO$jYqb7yXrCRvrrWYhWpQ&c#=LOJ^F9&A4nGt{LhZO6Ujslv{ALCpzsTD3BT
zI8)?}n0a$^g5jo#S$xjtqwWbCbVZ&t_J$r~fVwd^*o_OSE->^1ryMI{Y2>J6npM$q
z&!&$y@B8FwIF8!!>Hr6z1bfxIo$nF{`FPfA`;bt?+WCZU-Pxm9wn(0TOW;dO$EzH8
zSk(5|hX?r8!~bjg>uw;K*dyYaQ+j>9j7ql7&VIb(>H3nTa13KQuolLa<`LdP^n0{m
zwpGb=EL-U~jPCMIBBe;<Yay$`^wx3H0N{x(bt<GRy^1O_O2@dM?J>kQd&?i^8ieOa
z+Dwp$8$n#CEK4s!@l#rfYm7$a@LK5K3RV(Ygc{4|&p0w^FD98){l6|u9?AS4h03dP
z$5P>*4R2ERZ$K2yeK0kU;JOxTG?L?I+X97p*l;jVf`C$bAMnCHaw&?+>DiG^5wSZA
zF|0Id=_$z#7G!H`z7{7Cks44|;pZC&6C|{;Hes3n92j(mLWFA5eDbN%z%P93AjHU<
zu$U^v3L4&qqBJ1#_F_<&2cNxhlvH4oWHG2)JBE7<E|HyO=P<6PQ0l?^lp_NJRf<oQ
zT^VoKQoHhBU@oviHqFk<`^@!mp674;7Ymhm`RxebnN+DzTBr9q7fnj*l(cnpJya<~
zj&Bl9tX+No@Gz6TzWy7a;6|-QxS3o~9S&FP_OGp^4^0zMerP9btU%Sm^P_|p5bCI$
z)kX9e-S_s*8R+jvPG>cKSQLE8J3jFw`hyTrO?piKG)3Ol|92UF?&mfMfX3!>AHLAN
z(q7d5f*&aNmB&dlHuk%9FFZC`x6W^e?)am#DW@}bC@BXihOVK|(9{;ItuR$aLMIJ|
ztO)Wmb><jSHv>(KFbzAj7g0jGf`FYYaZyOvBGc}F0C7Q%zLIpn@^X#tU_?|Z5l>^R
zfpRq>jLPiq9`MS`U*oaIpC=5<fFVs>DPQG#Nd*EWq`M2q=|4HjBxFV=1U8uxPbbvt
z4VIQxX|)<W{p>})`n9jJy?cj7<PlXhe&7+uhC4UjWADLD*3O(KER_kukkNQZYoR$W
z!<u4?$(&1P$lO3`DS7(o3;6zTexiF||K)GI%&Xt*;`;$cx_NbSTIKCG?r`SR4KN`G
z`+c7O)Oo(qzRYmY<-K=qaQ4h5o+oBaFojQtg5R7E@Kp*4v(GjYkvC5O!V~Nr+-GTN
znf0|bq!1kJJmCJF+eD{Nla5OqwY!cw6V`~+1R(|KR5R*~kji6__An;Huay{26NclM
z_R&3(r0*yJUWt|EMMjf0NfOg))sRB*(o28D*~dSHQX$gjL4H<@&=Rw)CPj%Jj#0;?
z%$q_CNtR*#&~eB8I;r(2D-WSUq*VwpbaO|wVKVBFB`LlN@cn>B!_lc+$N2o+DP?D!
z=f~$3ei->kL0z|F@Vz7tCft6ses940@80FD@7!iI@bHA=6O4u_yE{jmIkPc)sG>#q
zgEE^DbBsgvun`D6PhrV0dP28zKsh>r=lN{!AK?cTRyVh(dI7g@U1dBTQmM9(aX#xM
zn=<Sik?DldxI=H)r?I%qXfkF0pv}SI5z_O?w5H$hadfo9@^T&D_s}|JFzB(fbDt-l
zE@0Ap@2e?Fp9)MiGu|vs(Wx6d;`#nVTq3Mx)E)3o|LJe@oo`*?g|mxPD`SS^PCf$0
z%<e%q)!IS4QcA*LJ`eEtEFfX*9Q7%<7<~8|{3k8O7{(KrOtP7i@;o0I3l0yvy!rYK
zUjD{4tPVh1g!O124LLd*u(Z^gb69|4Nb&r#W5#W=vOq)Uhps41hoMiFP4Faijt)3^
z;uM3-vU|{BZEb_~6Q{7#gxj~Tl4TR>l{(|8`?u0~Ogf3l^pt-8h)I@EZM5i*ChYAW
z(CrS$tn1=CIM~P9n8l?ULEw0gz220AgFTj(xANuHuF(gBn@=T_<05^)UJBT~K(?H2
z;c)wafBZlGCJ$~r;4@F0M2G=#(#I{XHxC6(pGbpT{%;KXns0vNU7mjOEY(sOgDVf~
zyld3z468M>Xu9Cf<aTbve4uZk8%u`3!=y1+-un*Vs1?>|qOwBeyOS4MO@hE@d1;Ma
z?~rcqh|u@Zi6M&(HdKtpBPQb!Ug$B+QhI|C2Z#IY+`mnh8j>`o+v{R&LTjNyJe@Ea
zCX5CHzV*$o^7O^e;|D%g7nimB@U|%Fn}Sp6qwtR1OK!}8o{UM8gi|NBKnnW94$oe=
zz*qn94`?6ku&Ui)5QM}S%e~vz=^PzUU1$-6(JWZ-JP(<7j%HapE9<z`gf?$d5l9JJ
zTU(!ig1=jgO>w~_7^C>+H?Hv6&pyfX7oT!vr-B5ZkFl`2*t9d(yR~lCo>0tmKNr`f
zF0JBV=N_H*E=od{CWO91<p^0A1S~JsF=m6&aLE4t13ama225<3j#4}?q}@3p&0<s(
zFdR=fYIk^W|0d(n1f6A!$3w<LL!%inolF41cr@kJSH8}V{lqVF?uln+C<w+nM7-eq
z<p`5fiqKOtput+Vo>6C0MuQ#;OB+1>#M5Xpo_+RNp1SZ9-`LsVaG0=J1*zO_<6e88
zciwoJwbPGNE?0=ch~aSP9JO@guClBs@6Ts9Dk(6^bs=w^Ji+4PBD=f0pP0_a(=q*?
zMTi8f18M?4VmwK?{Qez!M=5W<x<?Qw(uu`tkN4lZ&6mIOB0=Ct{j=>kc79JEr(Vtm
z4BL5<l4qd^VQIFS+`2qsJPsIiLr$DH#kKy3Zht~hu5n`P9Hn5F?d?1G!XqfV8IK6V
zsNcaFNYXLw&OTnLgz$Viy&-#h`=r_>!w_@h#w{u}MWs?AjC{K7F4wPLVleDet+Wsl
zj7c$>!Od6AxvsX*i$NzT+88{gP`2m>ciU7#9r6CVSNWfQ<C|3MfQ#ps=nZ?gagNz6
za37DlE~qQJ2CM>gc02B!zb?3zX!DQ{_kIvaoiS6i78r~hz=y{;p71B%e49yMAa%u!
zoE08M3vS=q;d7t6i0``&BDemi7(O^hlnOnIhX$EKQFss*d_Q1gW0UPWJ>uz@!Jx<9
z-XUq4P_H#uT3X}e+DT5Gd4m0e9Rv{Nwe0?=OPVBPNy6d5E_d(UWYX(!cyPqd&Vc=1
zN~5WeMp2FeM9Fspf+rJ_1V)1i*REaW$G&V(QerYonz&5P@zz8>RbP;18R@KCBQV14
zee-6ho8j`6tGx8Bm$`p;NEWB8*P&6#P`*V;w^veI&jl;S$Jh)Vr;LYv<jN{a3A9kN
zgCvE?3EVETG}g=v@rNNo05-|z%Xt_yO87`5-F}akUV4jhU!tr)S~s9}I?>#_yU*k2
zPm^zarI-S5X1fuKd4H}H#B3u$(THR7Z6%g+xx(7o8b=5BNs<Yj&LOvM-3APmQpED|
z2FnYpY@L0APUjF|p;E4+jbS+Ik)|ohbjt3|J$82QG4333uy@4H?vTz%Q*TsIC~D;h
zp(2tvLC6Wi{*>Wh#Et7$dGf+%u>y2zNs=`0AOtzH7f92LBuQtac_Rd#m=7z~CgZj5
ze23TGe2ZRZgw^+1DkU^(7NHEFFb1^o9WrJ=e#w32jIGDpZ`|R=&3jyU`b>@v=RFcd
z7F(dxLLzt=DLmfYKKs3Vp4oIXqJ6MK6iNa=q*iYdcpmX|NVk2&skLQ<4ww$71o*Tn
zi?jKR@o0*k9x)pAm`?k22VKgw8ogeh2Rn!C9gK*|6@+$9S(|`CZ-Vk62tsO=kXtw3
zWi;&2Y%YS;DB%#%*_oLk?b1~Yiv}L!Mvxk<F=^t~7VR9NJfFqI3R@?)Sl?Xd^8P-(
zu_lP}<V+j7NBcaucZVky7G^7dCX)%RRx59uf#QxR#@iIOO-|t7+}L=C#Q#Z&vFRLf
z6ENso{`e1H=iIqX%H?wL2uR4AN#>)etT=X$=kM*IU6whzVWms5kM{S;rbCoagkhN=
zP)IEMyAKGIq#CR;jT5p-MyuQ+$;(p7G-1*k5XWuecu22zfOebZIXG;y^WcbSDiJb3
z7$}z_CgaqB0oGElRp<_ey!F<%IQRIIU|h$x6&wrP=9_zpW)GBdUG}zUq%lm!V@Bf%
zx9{FyePe}2qsfzxKgpY~e}}{FJ`1s=l&_36S;qCN@9<);hZh8RUchiPAj`6Rv`Jo)
zwDV_hcj){bMx_!bPn`HfgF0(dtc~-LV+w@D7ZC!%ja&EFc`$OLVJ0352VU&&cG=(U
z(rh%DcZ}u>kC;=CicayOlsK~kn}5zcuv=MKqqVTeWIP~FChY9oXK#NG&x3MQVr~5t
zkDa~1{=qJBJVqKxxl&^iPZ>{!=rm(I>T~aIK&O95zkSHTPMe+m5vlbE0uL2PYK;<}
zFPV(TNa-=?O-Sty_wV23iKkm&Axm9=CiBEwVBH{K3p&k+(=<odjq~ZAhf;3q(PTQ|
zYhV95S8r{j%!nkuP0b%tuX_k-^7(d(B(s>*ZPJnkCZi#qkSJk6$yo!2HZIl>E+@<f
z?}+cG%7^4526cwiBq(q}o$xUl4h}m6qDrJ9urg2HJofiGZmvue91AiWccGB^P|A5g
zH=B`@XR@vyuZY5OK58W4sJ+kLzMI^B`b3pVsfri)AU)1Kc9Bx4g6~C?t4+r7kTgl~
zrC>4|aCET8VA$ct?dyE|wKs_0zfQM3Kx@Hx+#`rA&1Q(_fgc3a>JjbEF2<yIN`TS%
z6nMwXaKzhr`?ip7n6j~MXQWid@eZsd%`z(08vogU_Mda@+9m$rcm5^!c5V@#R76;e
zlxPicoG&@HMdxI$FePW%kIkWZDI`j|0NCc)bU{-7;fO@yljsT7<j;tQ6h2vII5_MO
zsX9so&SC{%6??l!WSPP93v#_BC*jQamF_E_^;Wdj=Ug?LqevJRC<cCsFbs&}2?qze
z9JLP_kESd%TSQSA<$F|XEtXfe2>p;CtP+*03@3eTW{?=h!#+m`yYvTbE?>XG%dft{
z`&Vz%>x{@U!EmGrLT4}9jESOvN)~W<xDELbYJ1#Xtj+fT&&#n=3XGL#V?k%GjXe({
zk}PFo^CbV}|M~yITW`F^SO3j#vvbg=6eyGrAaZ-j@?hY@9sEB_e*#b*42CJMefuhB
z&aP1k=Yx%{nIju854w-94s(Ce&c?YE7nBu-%%n_4Lxh2)#TH5}V6v3^w{Ow!PdK-+
ziZYgLGQ=87xm-nCi61yhfs&Th<z*)02r_p*d<1%8NMg%qG{6`b4*K|!q}lSxvJ@$W
zD|rMVX*_iZdzR!#l*r4;<O8hd0j|p-q?=PGaKVI8931TP#^qa>%;1ZRb7xL)_Usut
zM~58l-Jwy<cR*XgWHRKz{VSY#@=5&AjWigIMy{F67S4RGzvIDMQGQumTqF!bCX>l0
zpy#=0J4H%^)gg!&r3A0PewXjOew8o$$OR<9?Ahj&h<Q`Y@vg^01g;T*B3v|Jr6nwt
z7!P_>su8D7ZUKgPJmkjpn^a0w&Yf6CWSVr+!}kJer4pIdgh7b4Db;dFt6ruY__RAm
z7@-M;LS%wLe~J)-GzBk+3CoH`T`?MrsZ`2rtSz(i;3hV2;Ile)_l6Y|i?|BZwD=4F
zLdiUVmRM;pCgJ*xYrK2y4)4789*fN~7oNVz6Bl0N>f4(A&LJn3i&C!T`jywYck5lw
zK5>y+qrv{c4&82-#l^)LGUzH4cJ?ma{MLEJ<x`*fG_St)+9#&-G#ep>Pm+zmTBPUW
zsR&P&8IHQd6OTxNq@2G9pWz_p^5xr{J-3zLTP8>6<U@8*1oFpeA<E7t78F#aMq`Ci
zsgBO33<h0>!vWLjgz+fO?|FsveU_FtsaERven1%238E6iaWCI2s2TRVv=4Xb_KtY}
z>Lp%&^-ZqaxXZ9VrPoOa%Q0bOX*68vyIwER8%?-;>2;ob`q^C9Me<poiuv7TOb*~k
ztah{OtVmrWpEV@$gs2ko<uCs<FTC&quYUWRyz~dZ%ZQ99ut*QAm1L>r=wO#&uT7=3
zNEk-=o=2Lx5l+~7a-pQc$hqG?zIo@*l8>0yi@bACIbM?VfPBX#84ikdMD<cxaB|D7
z6){@7-Fce)`-<%Va|1p;mj^2y9Wrl|cO}%6{k=olM?FfV2q97?;~3u$SXx?TI!!n@
z?9*y35ogfe-XV%2*1QNmEW5gS-tpFG8I)h*oL43&*J!pDxwn0vyAL{Cy>&!$AtZ`Z
zywo9t3#}H-W(!|QOqPOibGZs~yNm7ShU``XsToRPj3rAmWR{U-nkX!D;li^#{`h(B
z-nqlq?%riI5_rO5WlEM>S4PnD@+HT*Yjdo4_P(S`Cjs)6BlC;*has*@IKZZ3j7^c`
z-MPZcAz3%82(LgPGV0Bcwe^J=%IG$bh}>SFfcfY4a%_ed60AfBgHrBuXQ^d(cc1aF
zPZ&iQJt0jp!Z2iEVVU7zjMhG_)*_>E!v5%xdcDQ63JJq1_y%iIO2*P?YP_JrlU|vy
zTw`f*iS2_uuHD$@{@#$4)sXQdCJGdR7qPgsK%>!g=ues=@=+gi&gijqVN&AD0IBl1
zjn?r}GV4lwS*oek>YO}zg3ZlMu3o*umDdhf8w<*1gSHuIY8j0t`P!e4AJi=_moeIK
z?)(y0FYWW@8#nmkmo9MnRB3)rrRz?f1*)??w>;R+Nz?9PG_x|S$df)la0u@1&OO55
z42?#UTBFH3Z@<f6kWgtZ@x<wKOh!Wv4tJ^5n>1>RG@Hv<2}nt`T*Df}cr?OfDTB$7
z`#bk}@6r{nUBBrr?Q8o?1{&#EN`WEMF@2L#ZTWb@BbiP~;yBlfLW*O{=H(o%gLwk#
zmX%9{^(Z%6{N2Cx9E1LduYT<tNagd~voBC9SNW}9|8;tk5sMX9dQamC2YYwvb=!EJ
zkMDc4pzgR+EY5Smu~W*uQ)>katp=4!<r5F;qH@5+Po3ua<pUb6h~<?!@4dCntFK(*
z!i6(5>T_37Ot>%jlQW$Q$4L|Uu=yg<D&Af3Udxp-{ZWt6=zwzhEX_s>Ur5^B0htY0
z-q>Qfwm_$Iz+g0>UTd+?T%p=nKwFKJifW}smS&8GeXP-R20d=yyUkngyw9C`_vsH~
zE?qmon;4X*@uelp#vC0@Xe~d1PBOGfNs}Z84dy#(oqc`x*sa3uQ5Fa|f9@n-`qCG;
ze(g4Izw-{YT7w_`{Etx&Ykc#Mf1ibhCXyD6U^wWqw{wS$Q|Af75D<*VW17vT8>M24
zkstsnw~4v#v0^q(<P+NiYx8?L9%4yd<wW{!2Py#Wo)#+SO^TF-#S%+PZW~Ib?d+ab
z(x+(Z`QT_1_u1Ttjv@do&Qy|Q?Cu_rC4IuE<T#F*C5lQk8w>P%L$b`H*;=GGoH8Eo
zv#_wj!qPfnRC5&rol>c~4t6i7av`V?RqCv)Tw?d|fJ;|)*>6u-TMHSFr<5X}pd7KZ
zxInd1b-p}J^C89a6_W)FS%4>j3XrodXjh7N8z|j9-E1y${`})CEiZBD{rB0q+GTAb
z2%@|LT*GKM{;m-(#V&Ju42UT@i;J{Ee}tXhA943<z8%8%u!wyAn3oeeb^b7tYE|&%
zpZyG<|B+{~){tgyo|{1BoFb9)#K<Q&7hgH=#)>?pw|TdyL8$=GFR{1R1NxMrGPP<0
zDW)thuMh?ywQ7^Yc8_kaP2fSJ9I>4xEU#}61`+*1hu+a4N9`ko3W;h}w$41ppg&|V
zJ|eMj`9_<=qcM$o37w|IM+4Gqz=h|38k-r?G{$6Z(XK(xxFa&(yQ-9%3@ve4wu9_l
zbWf5ZfOBWgl4J?@x9^|=MJhZ-6GN?x3kI!0CmECRn4)vYSo_ehS}DlKn}F|mE|{|h
zEy$6(ALg1c=>oWXfHa=l?W<%2Lb)**HY;2Ki%TKD@QW|<?8S3=^Gs2S@Q#gT$p<{=
zWj>A(it`!&Ak-{)3W5>{p9c@xDCzbwR!U8@(JU-1QYw`xMOAio59kfMR6>bOC+Kv_
z(&{>%lnnbF+K2meyFI+9LZ#8*%wvz!X?K|F0n5WCx9$%)Je*M7iWrU(#?u~YCivN3
zIE_wH(sYW+TsCWkL?d#ZMQ$&`P3<pGnPP)Lnq_!Nim!a0JpUxqH0I!F59xbMjbs`t
zqR1em!sv`RPMF5i51q~ZY4X>9`B|bUq;oXo{dacR-R|-2m*34dZxqcp4tbRMTBYNq
zR^gx<r?u~a77HQq{eW7d#$;e=xA*aVkLh^AcsxN0ixP_F;tE@*&vPb8iJ}Ug3aHeU
zkiO3}p5iNoPN!7IBMuJsP^qQSTH);3lq{Xl8xQF8W3Juo(CupKTP3nA#w#f{HaAdG
z5l^Qx^0`uyVosSgmMotw`B0!Nif%MFhN4obVvHaRBR>D)r#U*<CkR}(Zcwh!?hj~`
zoy}*N<>+9SZs&-_<#qfZWHcP)_lI!1jd25>9mmTBc*;7bE|8Q<C03S~I6OT3*m^!L
zlgWVkg641iy-!i8M%=mC=DoLfdE>R)oPYcczWn9S&p15-un#pe<pWvsW)AfM-OmGh
zcj8-%%QP337_{%OzqgGNg6T8?ER#{6c$(5Wx5kOnkCCPsrKp1EMO11_SS5(#n85Q$
z<1w{bi~apw!qBq3vdJ?~L*QF_;}PwC%-silx}jiwJw)q-pyCsid`HL8DOr}fbC`2k
zXMCni=eujwEcq%{G9ti}f>IO_Cx+FvRX!K_93JdZDc4wA-#|n)_S$VuE-5UA@nFb}
ztM72~+%r_`O)8ZNolb|%&CT4^iJX7yYTRyTuB&P}n}AcNw%FX*;O@P9pNP(rbcDqD
zW1VS~=Octipc+hc$E|g=33wJ^1ogVdSAP0M9y`B99LFI1{Jg7qk}Yh0W@KJ%D+2zx
zUl+LdX^o{)U8GiD=H0h%AyJe{WonfcX_n@5W+Tc`mA$<~I=wcLZ!yV)c-&`kb)CTX
z8TLDL4)$qx+IV4^YNN&DPd`Hvr(|}_sWU4)`1T!kcgLJL6)_mbj3y~@0zWSOWB0Vq
z{kq#I;?AXhh$1$>r!Cr8*T6X%QL8WW^b=2!=#)mY$@2Od*Dt+CZzx#wjpJ7)DZ_q;
z`**LiaqcO+z%@&xY3u+Ak$1lfP*Na#&mrB`qVtN}hmnsI)K$tb=z{fdlhIWbB?2-#
zan8h8%4N&X{OqUr(vQDD7)7KyMQ0hxlR59Jh~T|>7KFg)Imx{+f&v8)Qh|&Jq6S(E
z(j>*2m^8I4uQXX*UBOz*H(&Z=dgFxG-+GfL&z<FoGp9MZwLx=f0i_iC`#W5_e3{WG
z;nss4y5k|Mn;WdOYOJnqFp75x13{Wee7{7g?voiy5Y@Qw^hGA)u?y;qTf8oXKhH)~
zL9}M}@@3vR<vImj_e-ZY;OMBwsnaJ}Tx_wqwn0!T(Vz5Kl>(&ON2!fvH0sahVmYKv
z;CXpz+Q`|eGAlvmvnGsDGv1T>kVf!Hi|-3gpJ{OQ>K@|}Sl_|uzN#TJHRtbGORMSe
z*Z$fUc;VAeB0UeCrD#v0l*qXa1u{3c7l#<ne_fEZ3z0~Iu!2w(jFH6C81R`UDW^_t
zve0S~Pp5qIrQcy{B(J~m9e(7qpJ8KVk+Wwlt))eDCb<9L4!5q~U^LNOy>%O-1k3BI
zEH;}g4wpGNN(nr7?MCGaQYeg&)a$F9+B(Z*JR(h#JkxVX<?M2kHb-RgTuit!i?Ob}
zZ;c?+n!Up|!{L}yr#D$yUSe%^1?h(j;wcSPI3Xa@&R#x&pzh4M6RfN(@cX~}8alIJ
zB;S7dDxdrO1vb`~XDH3_@}SKT#-d?INa^O;<zn*kv#}1Pl+xqzC!S^0xl58v>Gh77
zP7~sIiq(=JjF`qT!_gkTUt%;i#PNtmV??91L=bqS$&|^k$NtU(jyhfNd{kJXwYJGQ
zEH`dkW^HxM-TPgPHiTiBa@k`ro>DGV@I2`_@-tK+pM8j(4b+8VA$?KepVL8v(&#jy
z-RThqK9x$9L9a_WDj_`&zf$Jl2-Xvc6>er-r+vg|(5G6f6L>z|B*y9t6$Gx|Hs^QE
z_ZhqQSG;fE_gP<G=lx5UKenEai#D3fOnLg5bCk*f*RSm7ovQGSufEH3&p*!6VskcQ
zR9ZP7Ta)X4hG-ryzaQ5<j_>LRC7ycv1>SybpERA&?(8!jO-Yg%ow_Ex@nk};HzEki
z3`d$c-l5qV(P%E<2Oi0E!erQEZ|4D>UJvU9gq0dgCr)9c<>t+6Y^)F2+Z&K3nsTLz
zH4}!DguoAw(ju*#BmXnq7u6R#Yg9wJU~4vt3yDz~$#lv%F+@>7wOV7)>ryFK2%`wU
zQs!u{%}Pt6(PT-+!QL+YZkH%35e6Zh&LP%lJf&RsGuH7|j|IcWbziGi+4uy5Iw3V%
zr|aCleaN6|!50W633F<UCZ9cNEY*tO=YQd|eE###xD|K#o@%AU2mHEx_?~{qugeek
zbx9Ca3Cjy;Es2u^YZXaqSX*17*=mrc319o_?=wsc-+AM8p1<%EXHIN#`cy_^v4z00
zyZeCaSFbW2r`))EpHUjKva!x`v%<>qI{lF#^4+F}AS~loQ?!MJa+A}i&N3N|NV6pO
z>+`0ixnD2h6o7p2u(crE(DE$HXm|T`I(<%`Imz1U3R@>n@a-_7HyN_9SPUx$o2GQy
z`xGPIq;%^Sl_&CM@ne47f)-$n8*$`)$hos0DX1$~BYNFFLPW0CEq$a25oUt4mLN>|
zv7fxi7r*p8m1=|#UXIE*!eK#l|9%Kk0ZR&*4=pTEFeFw<E<FDQ-h1a&4)?d|4JMSz
zmXlFLtJR`btMSZp7t#2vuCB7!s1ho-4=&3xrlSG7JKJOyPM?2*?RK9a%_xT@F2DaC
zTbpatt0l^1$#CpCyUOJ%yZeW<S|?ds+nn*#zE^3pD8qiRRFfA}Kw4r0@4Wv$x9;xn
z{^d(N{p2air7~eDWZ1SOnIz0BtHwfqFd$9cUOi*x<Xu;)5wr4)%ZBny*}CtYAtpZr
zSy`*|Gk@g;Zr!?$B}PzkD_3NI5&_1<xEWNIpZ$eT@xrH{AS$`3+$PD~ffsUSFY`-*
z*=2E@w;@C^A)eV#Vk-ll|IClEeg8T~hljL#L+Z6M)oP7Kqd~P)<1?Rsk<2PqSC-gX
zUvmRjLpKyQn@-r-*+vB+k8Pde>h1f~YBjPr<>uafR+bm2*N+H7H;_~*g{MM}Is+a*
zf0BB=kq@Z+?jWluOIbUgfi&mvxVcA2!6Zp}^YWV<9Co;Jt-<-Tn^ejbyeMEW8Ixra
zPq~ZJ3QND&&pAvVPxcSDSz500%yS#;-0kwlD|b2A9q{hkH(6U<nvFA<t!pb-^s|iI
zaj2{iZu|-}9Dc@8u|iUBtN=br#f(NHMk6;EHV6WAW{KmB)zwq@LBRd(d&HwISv)44
z_F39E0S0#W9&qjIWrh>Yt8cuG_5{sV6W_)xG#6M~9aF6e;v{1lCoD88+_=#t_;d@;
zQ+ZH#+_s#Cgnc;ibf1SDpe&L#hnUw}O?TFH=E}&`8d515%?7uxSq39b%?D!*-Odq{
z;h1DP!uNcP)(&|J0)&{~8wIbY2<(byQ%bS9vGK8W{6UdYFp0-(@7(3wxl_FGxsx32
z4S4&t`|RBB^5*MT`Kd2|7T=e%VE%);FT~{Ua;|&l5X~{&FDz{$M2HY6qtTGzU<$Zt
z0h4J$Jk3~JKTQyZ+`E5=@t{LG9g$4BEU#~Zg}sAqu3Wv$aIAUtt+z3rq_xn(w^J4x
zElzBVxPJ2<ags0@$JFa(_HXvVxUoszjJk4M_vU-+KL69A65y&ahJ^*!t#5TktJR`W
zZ=igirR61d?%ty})KooZFGmNvjCy_IMx7`KF<LVjkEzva50xrGVAb5G6wl^)9_P-U
z;gxTH`xBwpl*)>~@=KrMSN?|``rS6*=Fce=qGW{83BnpYFXl@>d65^t@C@a0fXJ&Y
z))<86&5*f2+po_9XUDUaSSfk#(?7=To0r+!zDK7&rc#aw0-su~My*!kxlg^|{QAlY
zYfCMpHAJO|G)XaO#{S+e#=_ahpWxm>2Q3wm7jX5`Wj5EBX*DWTtIn@$ZD=$q>>V7j
zuux}laVe*~m}9m)f9KBcK1>rUnx_PA0z+a9Z@%*m4|Wc?di^TrPOs5wEKsVJ=y!%B
zsUQ?ZLy6_^V4oyO9=i8k3D%*TAN1><msduA5F<?n9gDRYT4(sumDrSwFj++4d3@$G
zXZeYr`W)426;DYJ7G)I1X2%ZCEKq-#P<iY{&Y&jdft{6t&8_o%>C3;&um9731o{Zi
zSQiw9A<d>6eys60v9U=q>|;#^LL)4s$poV{i%ZL_tgP^JKmAh-`hB|HE@hdqytKe5
z9?@uoBw0e1r38LN`yk^-f8;sBu#_Wa1?21$;0Ud5{nH0aH8Uf~4L%I&{M=Wb#c24g
z-~L_p_uI5uEuMSfB5(f5w-`-F1fFtE8kmenOs5maU&zY@MPr2Z=I`(L$Ov)lqWWQp
zl#*LFc9>2)tQlf+9*h(|QbY)#9Qpj%k6++(pT9^HmGC^5{W!F0?j(xfzGw_F$1XFO
z8<XYuV9UIeB^6IU`yvmvAMkJf^=}~c5M?aYYK1TiSzKIXb)!UPB&RkviHAc383<Q0
zPNx&Fu)4O%^2##5{O|lC<I#xy{XP1P60KH)gU%t<s!tqSvNWL-l<6MzdHk_Yxj~WI
z0Dk^C<m@Sp){w*cMF1r8_lC%ukd&faUgY0>?s*3N5x@B_f17S^$okqE&p!7ImtTE_
zILVNK>kh`iU^pPnl8>#ww_d-=sAsu$WtTX19o|Wzc>B%keD1|(SXyjQaHw2>ogdi`
zX6my5#buiy+=A*E?<S%@n9}PU0t#Uj^=fO@36HhB`JHd`-nAQCzH*6AUwoF6Yb$J>
zIzfGL5n%=U`+KC?aOT`&Of$Eiw_dAqaB!c^wGA?xv9?-cXSd6AJSJFb(5Nr)r7!+8
zo?k+n1ZlpvGdfOKpI!FF`LlDvx2L?B<C4CI7lbG!s8y?IB%P6=+;m*CL9fkl&?AnQ
zW&>QO(<zNcgZU71_wzl}vGc9-E^y1%$rB%2#~&1}4PX1?w-DNA)VI9%&TX_&1Yt<O
zm-6+mzR5Gso@aA&+0h{Kgln#Qj`62Hp!>XhEaf~ojmnFp+GqE{Ap(W9ie`O*BuN>K
zhiIK~`MtMz^W96lcjYpVoj%D^=gy(igi5_m6qe}qdZb!&>g*Y0ROayTfVI^X9^Ajl
z`pP;%6majsHdn6gGUyFhU8+;9RCwyi=bY|sg8V*7(D&2*yt1T8k}*om(cYj`qTXy#
zib6IwH+bufci8J{%F6;_413#mIoQ2VYk83<Dq$^jyImTMhNBi!IC6t0ycwq}L(lx>
zi=X`r|LoWQ`6n{3Za5mUv%Nzy@v&BuWeK70;mLqNHPG1<Pb!{xa+S}2;Uc9{nPSns
zZ#~j1!H0+L-$IV_#=e(dCx4Gf!Rf~?@N<9t-{+tF%72gTKR~4Ub|7oq&i3^U3_e>Y
zPm&D#`M5B*7hK>5Lf7gImX?<I>wo2E>2`aJheOJqVQHa>m6nZ-IyY}0FdB_GvAIOQ
zvrqZ46@1UnsUpYyI*<b?*6P__;NyVJalamTWq$4}KTZ$?{Qv&oYutOV$LXyzeD=l9
z^2VRMOm}$1N>jPgt~T7ebB%tliytba^hk1Er{^iGCyx2`xh+a*zHiX{LHTIq*NS4)
zlUSP~RfG`Qbwy!Vt0}(txeL^5AyOEy3Zz7;44oQ>xJa<*oPSpkC+CE51he&FvkxF#
zTeQS2t^e$g{1jNrOTYKeFmac~W|KhpgsMa-svu=Z7HDL>#iZY(*WW>Ans_?FlM<O}
zN}i<Bs?%yVSX^49wNPU;84`wNYUPkYZ-}xEUO#o}9KZ1Me*@2t&?b2(Xmf#mJ|~Ui
z&dg=2I`8n2LLro-RIZR}7mU34xlhyUwiyqPxbXM|F1`LP!+1=ik|B*iOG}cC8IL+B
zKX_<NiO83ZgTY#zI~ce4UWmdmSbj*=SEQ7zthOkHg6U9TZ01S~LL&$f3^}t><;71u
zNh$IW7_b3Cx@{W9nEdi^Q@5>kQ~p8Bf;y>+cKP|?a*~f&K;VTwU-|jJi9+z^OTUA!
zQd-ppfe46HiKtY?S|1%&P>m(}?LEfRE+Xex_);OYrW{Br%Pq3S7R`kw-Sz?fai3CD
zp&I!Jn_<#~H1&Dmh0pSZFZ@(KvLZb;e^JcJ=*|hMhipk>zlXh4iDS3x@<+e;BEvzC
zh1L>JpMRS7zw;KOWJt+(`#}wsIO#DR4?ed3-h1~3!@l8-SN5>lZ50;6=hn?#?%djA
zp|wChR&>VuA@5NBV3JYj#03DdpxR-rrCe=Mt}b!)$~7=hsZ{Yi&EisnuoPmgAx<Vd
z_uR8ot7W#<*H~=SDVNKnN$j>AYfWomk(HHYR#%ot)08-l>2556uyn@*)|M97+v#An
zrQPZ96F>D8&YU}o@_eMr7QPGFF6KQx^!vHaq>xw)iAl*!>Y4*^6$a1qh@uEjd6df~
zgi`b-8A~;bXJI-X(d!%#Pfz0e0YXZ~lL1Ps<lL&_SuFDvQiZbzkgTjM69fTC^6>-s
z<C3Lt>C#Q2pvKGJxPwkSBnBybcDH+6e*X?@tE;nw-FfH&RTQ&#_OR~rghn`SnCCgP
zA`ELREN=47TW_LML%CEUj0}}(L>PrgDHx9WJpJ@jRO&Ta)iSHAtJG>$vNR!+3X^5j
z>J3&`SJ~X$AWag|G-Yk2g%FlZ7}l1Txp`w3Vd3zo%?qD;o{KL$kM9L23s#x$(*0}%
zOM#R^K0rPX=&__SAvGzGR~3ZHR{}-^en2S-@q>VVJYk$#LJL_u<>+XKc={x&tnhrF
z;i&KYB{D}ibKU1<zM@gE*b&=mwWw9Av^$+oMCY}7jn+cM(Z0g!1Q7)YVUW1_s?AXI
zqn~-2rG+X=7=%>_=^>PMGe$*mri<Or?wrs4x(o1&6=KKHDJz-1)2E*KEPwldZTW-W
z{1v9X9TuAn0^t+Lh;mTI^Gl?GK~x)L;|YiDdnj#4(kV&`q%lNF;Fl}Z>oq3Rn0lka
za55z598#~8i9EONZaN;(YAy3O|Hj`T2&-f!#yb`Y**vIo_t<y)bqC@&yOs)~sEkFC
zW`^f3UZ8!r&+R)mc>4SWZoYS&{!xd;Mus$?Eu?zFXmCiwU%@i>>+^9s_94IScGAx2
zU4NE*q@eEE=g#s+JBOGgLj>-EQy9v^vX&kq_U_}IJB11jh@ABM5Nb3(Ec014%wE_`
zzQ@Ch9DWSEkT3rDS7?+32lp<qu+Rb^O%wbSO0^0}Iv|UOWa$V`39@8@upq5KX9;>b
zB2*Qu^iZV|X}LnDbA(PcS*$S{rio>eTK@h2=zquZ+8V}cq_G%j9&%i>S#dMyJpWjE
z+~sDC8KiLKwtjzz?|W1$W%}JC2O%3b%49e+WKD&(24fupI2iV+H5LHZZIfkgVtX-^
zw|Iua@QR7vz>Ki`=Pg=mo_y*Qk3GJ@tKU3w_L2{6QdlZvERS~Ja1U>N1sMdmvWK%5
zt&f-UiizYf8-*hZF5UN#8N%y2^lD*)pZzPp#7Z-ye{h$T<yDN)&R(>{FO^A#J#;+6
zm>4B2Su%Ac7%NDpV{AMnh$1}Up~@BF(SU=a19Ym<sX=Ejo}{#v8~ktn!~ckCy@ocK
zYi2oCu9wH`W@az8SOQhp5`e)<R}$#<`jko`^;(Ty=g8?F$PkPsmR4D!ZLy)nm9jpr
z+`0dN_K_ydJUmKR1X3vCso=f$ZgAn5$B+uw`DR^5yL<2U7>YC3kwYnGgV@~XM-YTH
z&OG)U*RQ-s5|8QjhxkFj!eWCc3Td{QEG@6%MOB`;@GQx&pIbl*nZ^pJRVuXGU3_bZ
z$`P_sAv2az6tKIujkFS9dI&2SPb{mOYkcMB{yHiOU0P|pS;yWiV4Ia=a&A)|T!HKM
zE#@J5LXj^{b4jZ(&Y_peghA+nJ5Mo=HN)6a^DH_}=?@M_(!{+dsTdCXfJO>$R@yA?
zNFiK9nTI6HFU)46Nwrd;{Rszk63O=79`P&hGMNf|S)~|=YIVdnzWyf9Jo^}H>&v;b
zDRl2HJ0XjEko@?{oG<E-x?<L^6au^w&wuu(c>Aq4n2rt^45#!b2|~emGNxQDbMd*0
zsIbhl&%Z!A8ep>wB@H%BuwKAot4Y7xCmRiEEHCDFjG|Vluz$EqpnMjaO+wFSJT}Nc
z^0_bmI6=jY=26z8m<cy4d)wpVl<W+FoHg?a=hTrN#%83+n6T8unv5`rXtr7`w3cYp
z8!Rj|xqolU;n1>DwumI-?yXBa`Ru3hJfBijrrkawj>nXvD(2W$k>a_e1kV#_<IYZ_
zQRkTpPxJMc{^VopJ55uD<3oP_7r)3qxW7ZMH^zh(&kylqi6=Z-X^-KhH?dEiK`sa6
zM5<XL?Ra%+cC?GcbVgI)d*9T>{UCB$mhwD4^_d@GxmD%bTi-z9w$)6hBm5x3FSQUR
zBc1fo>6pNiq{$d1JR(10kPI1i+bk_E;wc|LjF8P1qtOsgC`v(y)q*5}!}fr${OrHO
z>2qgYfNCsO&8afSD_|d-zi1z-G8(tyF-sYZMi^^dXZol#JKL@0CGOt2!dk3Q(qJ+{
zr+vWw-hHZ#MWRxPG)Wnc$L{Rw;|b=xlUSRF_bdNt@{xkN>sQ)fBZM(v6c$OKr>sds
zBZj-LyvfPa=V+WijSURa&dHN=<>yhlox_-j`L7>}<d1QyiZ4$p{BnznFa8AC`O_rh
zetrlOM3y1Y=*fVCyEh2SC5H|Off62p^4Z>hz+NTd?Aga$h9Lwd%b1MDOvhugB*9uJ
zRaW^sfA>G(>5HEMAy8K38Og)J)bWnL|37j66=d0wt!ILtyYIb^KB09+Y7HbrLIZ_D
z0VPnk#JzRdmug7%knHZUv9Yl+n=zYt>6dw(2immtbk7*cX?8c&H18#EU0J$N0w{=r
zN@OCDIwK=9bi~nm@9pmM;Cr7F2^7hy1}@9Z0*O?S=N$L#Yps9%>;F?-z2r~b*Lo5N
zve_&_aRMnIjs_G9#e{UKiG^h_5X4%dF=#8@q3-p`7xfUK8YRV}pAQEz7_}?+f7bu0
z*XBlzfAZh`fa-3~otIc3T)&ut>2XjAvdsZ^-guAM#RKH>S*$ckb(*$4Y%<P2YS$0p
zd@1hZT#9`O^{_>qBp06Y@W$tP;&Wd?H`fT0){eCi#4!Rx*sN09S|eL5U}A~z1ilDR
z2=;ci7%LT6T0RI8_Yg-By>^>^r-RWhxJ}O;=kNcQ|0fm>9s`nqFWotCg`d6eybPzP
zv>slF!$c?{$>(!<ikUo03I<Y8EEMpRA~c#vK({Z6jdV6r332_ikh<s(LQK>kBTIP7
znMhkSnLsi%Rd#L)v2qu<Rrg)$DLnwIhAr3hH}4~UC#{!<AAg>WjSb#<;}r&j7RqYk
zC?bv`^4S8V@)**~Vq$?Rk6}a2?(QAX5ylEU<zcm^S=%G;&!T)6DssMaD=|?-;7NQ>
zkS~n$5B`h)GxLi_hsl{0PJShQT`s`8>2YIBA_|-Y{3PMi+WExFWHR4q*SP?JO3<22
zCPyxpC7a7q?}jvbkSTzQV`|kMbQ}{aN0RUNL!5)a{r4b3lX^kBAR&mNh;pgG#Kib#
zhSsFZlNkm&q*|-uixQrU-TvW<^zm_*48>85LuQfzpdmWv^tD_ECYewl9_}QBi^LID
zNM;ug^ObM>9sc4^{yDwM7M?Z4gNS~=ho?No$0i7Jd5l(wLJ4Ih)x9-h-Nl3<N-8`l
zY1Ar=mvZECc{G-+@8JuHjvP%WpH(z_@YS#Vb)J0cOHPW*8iyz`@~%x2P)~UnnTYhr
zbno3+PVaYN1OiJO4anx)Fo;x&Y$i*-P$cjJCML$YcW;YEPf!XhIS(qkoAmn~as@YZ
z5=Aj#6jI2!WY?zWPC)AG>;_X5*cjjQICA9BXOh$b{P7>Y&cXR}#9@FmF*X7r0;G<Z
zvmx`sWADS8?3}#H;b)$}X8Z)n<5$Y73_r`^nUEj5xA!5S`@WQvC@;(5)90C+EMa<8
zbQBTm2y}W^N7UEv5Sy4F$Y3-mUy%!PblWX9*6va+70DO<A;KE<`wV)0qQL+mAc{Rc
z|Khj!+VA~sgzqP23H}hVg;7$szlA)~%Vo&lPTQlNQm%#WPClhliDJIszK<vA#n2BG
zS!K{jqA;T0?>lD<L>eLPi|fF!KuZ7JUw8dxHevNHNR2XvIM#S^!~qeqpcKjm)bCcg
z_4-?!n;pjsW>LaT=6PylwUWMADTQm%CQIytx5W_hOoOw#K&4Vh&&NnbrZ9zPHIw6Q
z_BQX*>9!CSn)Mwv?yitkl8oo0jUXuGP@W<RH9FGVzI}@z6EL@U03khe7`k9+Er*t7
zn4b6f+yC(Y$eDAGkzQ)nj;hR|6NYtFo7SmSl})BuT)QMa5V47!u&hih#cZN=BNFt@
zL>O!7hlWAuCf~HQXbs_@PuTBbwMI#Sr$#&*m)P9@l}aF3V-w^1|MW{ou)RK@AN3I`
zMk0x0Ll6%*AY*2g4<=ye>OJnh`yR(9U%(ri7^-)>%fjGzp~+O@Fj$NvGo%=CTv9P&
z*Ro13LggWqhY^bG*gQrFCdOLSb~fpBTUc!fx9_rk=N1#?5_S-vHRN+8BH<psupe^$
z#x=Z*&*bzh#y~t6(ChWx1$$ryzdXm^|KI;#SXe%i09Iy5u9ivCnIdy$2(>%Pfs&4E
zxBsz?O@zie??jojIdO=d)|yzm^U@s{q6mx?XzR|)p#NFumq{!U+6=HN2RL8Cr=K~?
zBafUWj-7|IkPb;0Ud;CmeWYAFGkli!C$AF6lg$;!`P#Ss9+Pti`Na?a6x-b+@B2i(
z0VdMe$P(moL;|8g8ymI>d__1I;K_h|kS895tlhnX?+Iq+=TTmOZ#<;6Xl+=UpF=$=
zc;vaSaN^9_q$SSqHM(|3QmH3s9qIeom(&R%5*^62<>;<tbPS%WJd{*qb2)O!gkz~#
zpjvIx>RU>IATJHoy&VRFE}j>ll*eGuMjM0gJI4`6!cVn4Q=|X~P~rwjmKGPeb$jJA
z*z0|}E)~8l5ZNwD4A5qPr+lo@%*+;f;R}zG$prNK1N<N<*i(d4%KLsyTUeu(?EW@(
zBHy6`9(nwEisRG#>_^{W^ZNVbl&05e)9Z9GB?B2rCR?E2>tVt!zH$yNQNK?{Whmte
zY;A6U2|0Z97}-JrDSWa)28k<-o_P2)GF#@$FTaeR$+$@}oRG8_(g>1{3i~M?Bfw$A
z^>y#X6HcMPgniPkf)t8uF3<S*1ciK&xw#oU?*_d{)96~fqM%-@u(!QNsXQ|*6dH{t
zQ{z*3A|1+b_HfC#Ab@jnQWC8N51%`S=Xp-F`ctmzUXv_c*9w&p#1Y2xDchKXo*=KX
z3_6BuKY5GdM2V?K&LJ|MyQ_u7;N(42dJa4Z`&?~YaWabQ>?_=560d}mNaZENmw5`)
zhf#wvVQZImtAVc+gF%;OwaV_=DoY0rVh2uEELY0Y@!D8z+1=T}3d@O;r^uJa5XN~f
z$5G6}%oLqYkB6Rmkw5rb{{ZFZVCY@!KI7q3e*!AFL@v3j6FDD?90Cm18X_GNn;0qF
zbgjiD_sa7@Kqi|ZHiAYwCMdh-xZCa0ZFT6j8V<F#HkpE!#MWSRB1mu2AKmBV{hIOo
zH*=H#r1j7y#A|PHP;YY}u=I4omx~Qs?|;P9(mcyAJcllphVW__$3q<>_rWPhh`<h}
z9ieff$;UKE3y=z-JgihnE5t_?=V`Zh=xaf<*`(iYQQ6!fUn(-OxIom8sMQ+mwHs9S
z_So9mAdW-Ir98Dt1wY8(2N~-1T{`_1K`zHr&%D5qGv_(}@RMj`M`OmMS$mRAC!9}D
zd9Jm(&(PfQr_#&VfHA~2NlHno!(=SlaphvPHiR0&$l@o#&=~7l!eQTyxm%0V`%83{
ztOX+y>!ndrCr#o!@!QGiaF_W2i$qx0PR_LMam4O1>kGQZJ<z%zarYPRGO;+%>|>8&
zvN@!67iz3yeCu@T(xd>pPs=b#{8b`*GP)=d{)SWt&z%=nU1gD_MOyVWdXb^tsL?W2
zb~m?}nVX|HJx#wmpw?-!)2y?*z0J<fCJ4hsro`^{Hc~oK&U#~)ZrGtXKEaDGeU&qh
zJ<sCN6N#IJyZF=dk|z6DJLH8Lk&v%NdO5iFLvkM6grJ9%uGNw1I=S<r(K^Ct%OJKy
zG08-!#%hQ^gO;2pvS=G2M1(aODdFgmDPDZ>JQHK%L?$ANVuFn4w7lf~6TJH?-gLi)
zWb^dT2r+Eql*$u)=_}vh@X{RDKX{E?06`{8uiv4NFJfXtHlL@vw@z()ja;FGvKexj
zBA)Q^l~1El<D-u*b1dvLIXy=d#`ODbA`?=YoZ{%o^E~#s-yxGJVszy6Hq#tq*e^`N
z^S;8(z1;h!)YA7RoccIA4pG_k@>OKBS@MMfQb@|>60kw2q0_enzMxjE(CxO#<Vpx>
z8H5poFe0DLp_Ds9`r~N9a3Deal}*3`WYh#<+$FXlQV+<;0vTWM%+seidUy%n3y5`u
z(x5zb-+i5aE+6Nl*|cmC`zMKAF{V7lnFr3Zd|;W&Z~uhq#tq7)aeBQT{XrW)$dW7g
z1VP01ox5nO3H%&F_>_uy`r~~z*4J6Rvr4zu;qZ}T1i1on)JN+O87PjQImh9nk1{uR
z7$d+E?`xqk@+Ea2aDx6Pl1js5RNOZRXEhjNqCqzq$aAgPOeVv`<Rtlgfr*I;^7$-{
zW}8+Y@_|Kkx~$z>Vd=<OH!<&d?Cw=La_BHp;&dxxvG4UI9UnUqUq85ffXRt*c6L9#
zR$&-7oE(kHfRRLTpRwL1hgF{f9)sk$Xm3`y_T8WG@c0;|W5<!6H|(rLQH085MoE2=
zyt(C)x)DgD?n~-~bZ$~sIS#%6M0NtL0{ZPPQMZn<JD6Tbb$5@0$By7-bF^9=+TA|u
zd)sVnZcuI3@qNXh*JX2ko#|Okzu%?XZ4lX*Lr0JE;KR@I=<{D8lPh4fNoXO~xx}!K
z6Ph50F1bsnWW&2H8F;hKGR_E-#9x6(d~c=a;rTvN3WC7LAZhh9Qw8T0+wXPgcY6%_
zZI`%POE?%HNc8EDj(jK6Ze<eJ|0<F?X(h^PR9NSLsB%&^$Oy%Tb!6(;8f;gy`t!FK
zo113h;fImF=jx8+ff;S+Xi#E*Qnw$H9aeoQGUEmA!%8m)A&|i$c6${+6Hy4Vn0bkk
zie5h=$Q3y{y$Dt?HL=a~)Fidqo)eyo46SC1AlIkTsGv>E!Q&5b?BwIjE*(n@o(4mr
zKBN!7gPTp_%?1oFwox0|O++~q&pnh1HEf|^668YeZ|7JdBj|>flENJt?MReyl=A3C
zwo=ya4={#!Um%tHUe|ADSu7f5B}&KmagD?5am?#rQ1p$(lak0<bgjqjAHPO%dWz!l
z<3s+EiOs&intWbK5#WOC{b%L=^8!eP^fGvA9|b^UKrYhUy+@FdWR(v~?mP`5Lr@&!
z#QZ_RFk)(Af|;2q8jUJCiitu^quBsysWodL;KaEnIDF(0CgzrjjRre#mqd!9VqEe-
zdR|8QZYctkR<g#}#AePq^iGW~zqD<YA|N`k7YntZ9~px5lF@p^Y55c@qsHj!ZJe2P
zNO>a9^Itf}?A!#NFA*jniU-5J{>0wfM0hsgBiRHw6=GjfM~XDN3)J%poO}H9EFW0H
z3u~BYK!4C@FleJ%l4f<2dp9prDiqND9<@fBRz63iT4Qr(8*L&cGG$tg78Z%}6s=Aj
zAq)qPKgh$6e~DaqhET@{9S>V5LdZlE*BkaztS}@4o5{Jm|80&qHx5;gqrp(Jz+L1S
z3dIsZkYQ}Bj1r(ckG?g8k)_w|)9tn?Pr8InYfZP?V{&{v(eN8dhNKo^?%y4qpT)(w
zPrc8dSORk16g^W#+aAV-7&G9=iA5fJ{508IfRvtVYa8RhNj}~h+PD95gyij~f}}})
z+L7_3WP0`h&wcs#Y46-b#vP=!^!hzAev8(gre4`*Yh#s#g(dp!7R_##AeU$5_H8Ql
z8kww*jx~EbyA;Ml+TA9DutO%B<)Nov;^2{ov3>@V2)(8+$6D!JPOxsnG0DCsC|%0q
z9qvI|)Vm*b9Ae@D!Y^P=?55TWMM|YH^0^%4a*<ZE1-?%w(gePxR;w`>^vLClNGWNz
zyTq|3lL-(>ViCu&qd6oTJ16Ap(8GK_$Nc=<XOh$z8)7kdR-#NuK5lS~Z4P-cEhC7u
zLoY*P>E7LC<wvh_x>zE+xG+TABkd;nhe-Xtq|OydpVY7$(rg4M>7l#~-o9Z99ZV8T
z^a$H~WU~QyIWmQ!>qclxd2*J;gJn9cHdB+6?CotkZmO~Ldjp!SI^BMUcCUq>2{`k}
z^Bg+#5V`RwVjUvP$m`gli2L<6M|X9K)<YUk1v)|TgmLJeRO$Oli!PZI1S6VICxTEX
zeLt-U!vRqgC*T(QP*?YCNGZYCL@i*H-2FQ1H;VewDl+sq#1=>WF23|!@Pq`8g@J`=
zug<NX{DS`aW+J-g_!MTi`K~Rd_6M&0^a~+|8N84ZM=J3Uo{viGb(C@lj4uqHl6XOe
z+{8FbM~`By;KCd4(%7l7w^5<KS!dAJ^jk6Y?Iyqc#k;h-eP-rn2r>cMN|efQ?DS(S
z96bYp@3wfKrAhi)_iLx6uTv?7@*KA){Tj8U(#l^9EADjb1SEc?8jB@H&<)MLD&5&-
zMNyPAV{CF@_CFi~spJp=-T%=1){;6Qu*O4KNsczh+2TMDqJ+Qiyg1@_WCh*38?5~7
zb)vm$f@+DRwQYvyMM)2-(sum)3Bu^Sh-8Y|p}Nv{6_fA#?(@JBNH;m_WwHQ6LA<_{
zW~Ny<e27-P#fR^HKy$Cb?s|phb~8CI5xZ-9yzz^-k;-FwcG_v1;o<uQPCfVxQ_F`j
zN;)?Ho3>h$Ga%IH;oi45&InqeWO`oQ1lj(?XQUpUW)tqcrokQ{vVvZiPR6++BK{5R
zWd422S0zW<w*sj`9)0Wx=N>wV?`QCo!1G4N)2_00Hi4GteqM<E_fva6epZ!Ipz=&C
z9H%gSgqiteI^8a{MwRWIRqkB9NMvFPQxlNQP@I`U<?{I1fMdsxad7z{89zs@+C;}O
zNDI=(QyHe`4wD<7!wQG2*@W~ilKqj&$5TF@=lobxl<DIfm;H%WnplHHM`2>y=TKMQ
z4=9w%WU^T%#>eqJkMdZVe0hRi2%S!sW~1h$ZIXTMbh<-C*Jy)|VyrfpNE3%4I&`RM
zIvF=VH;eac>Q?=_D>hv+UV)tF<YKS}&+qf-`IAgejiZ$3{IGp*I7}df0ihE?@%R(_
z4@n}l?<9;0vwiz7J-qxB69>*RHhYw@i5aTZJ@#tbY;WG=-mPn7%O$)*4q3=DJ~xN;
zJf>!+Ieq3d^K%Qtk)>Lxp`(yYHbda&@TE^_Y!;c#xz}o4A~)K952buO72tWkL*woJ
zeCgr->(s(<G>m3U%%InH?PTltS>^ILa|??UiX{#oK8Tc($(b3nm!mg;dn>nS)G8Qb
z34(xbw@a<jaL)xU=v!?Zfj`pdL5K-M9C6+i4v(LI^fTG>Uev^Dfv*B`qQ?<-nD#AJ
zcmzrzEO=7Ux0-<#?0xVNx4!=hQKjYv+Zb_!hjCKL`&ya%?`lXB=A`=~5`<2v;a#ni
zL}4i83$D}RXP8`^XZFA{mE9^Iy?2TFZjIfwUAomCaS!6IWp{0l*I#=Jp*$ugC&}h=
zD3u}b^Bg|$FvXb}jB-?$;eduxu7f5=sB`*EN-0NowHn>kN{K`qKj|VGjWN;ieR;|w
zm&*|Z8FINCekPz33R<D-!i4=E_3CyaitqFXyWNhXP9^azrKmamQ#C{of1ULkA$8*g
zpOXg{P#@hO2rbqs0t`|Jv=j&}2oV?^P`SCzoj2a(^w?Jsg`%504I@NOK;j!KL^8;2
zH73#V5;z3T3MWh`mD9tM%5w)kjv2(7K{#L#4qV?Hg|>aP)#QUL^UH@(o==b~F?H}H
zfyxqxU2>au>4q&Hd-7q9KKKCT`9(S_H!(Iwd0ECL<}g-b;}EnS^5RCxS;D1B49cz0
zLry@Nm8WIv2t|`f<7~tBr&uGzB2iqoP8OgSNurE(vL+^U6IM}7tTjT!ND(82pJea}
zJTD|tIONdCR3O<+&Hc8PvX<GY90wO>QCDvxW9Ji`5iwQ>j1&-CLL~WEv3c<tV+WTw
z^xX4^OcrUt#yar|b;+Q>BCL~fVfgU5vtknDQz|#LB|RTuEZW3`am-*aAdJ%UA~3{g
z424XQqh}wWG(JvgdY18p!${*Mt1@f1X?1HnxA+uGCyr5`o@Q@n7b`Tm!Wh}YxElu0
z5ok9F<v2M2a;UXtj6*JMYN3-r>`6-MJ})`qY)GBXm@&jbQiXnN69HH2CNR5^Ag8Pw
zmp6mY#8Dzp1-!V3BxYfDoW~wJMJD690V(Apq(ylnM}d?tWfN}5ei?vIA~eIFA=5%m
zN_Tk+K;@97MFwGm*sId(_sMx4Rw>3OCTKM~_}L=mLWzu*qond|ZEaBBu5;nS`<y&?
zn%U)pU}BWIgAkHTChMF!w4*|#aaxEJWp|UwX{*!PeMw!Cj#v_^MDq8C0bdB>IC8QB
zLgDP}6uGP;?&k}6vbik%DCFRwW8AyiqT3x%-P<9KoFoRq(&@I{>(UlI2;IPz*3Nz_
z{hW=(3TMkUF)=|dpQF|K^wZA!HVCDBXA?ygB_zj>E^_9~L6mg=KGH|Z7^xh>n2IJt
zG8LL|6@S&<-&eE?kZENrofC@ygkQ!Mvqb$ogG!ZNuSZ5If?|<up-8LUBghme7si+>
zOw;fRtZ%Hbv$w;07vAHGU;8q12M^J%RZ&VHgu?d&j1ky4<xnw1P=;;5Wd9Lqt2;s1
zlM=vUohpGz3+}XCYl*`FBehA3=XvDw1+v*3V`F7JPZ4U(!hvOWZr`NYY|*IgF*Umc
z!ci)^-456pY@~??;V7w3Wvqn2nh+}-ig#e?vuVEzKpD&IOu^mN*H#e2yIQ*fJ=wPq
zNiLMEUARJNd67fUJO`PKQ-z4NqsIszC)Z?=Xt0J<^edG)Gg7%BIH~Yt!auVX9mRxE
z%wW(bj6#Ib#QW~*e5T0p2hOp(xlU<zj<NYeNG*uM4w<zQt!|Ca9e9SNV@D}XO|rec
z1;UUoPLj!uVT^=0Ou8p>)G1I-Ba;+;w<akH-62gQw}7FgqqP=gT|w<xH*95vI{~RY
z8!$xwwH0(jNj7Uy8e-i?8$%q2?oJ^>35^lXS;Rdjju&qd?)9%rr}~YOy084+gPc1(
z!PP(Nard=%@RG?r$NB**#5Ts_;SVeuZ(e3%X_@gS&LLHR(H4WN%Gif2nb=K%U`0Zf
zblI~~K1%sS#!%g9u(wfRcW;YIb(earMyu0e5JfDEE57lihsfo7WHWgXma%e)uxW^^
zEp*(VQ>oCX?O;O7*yJ?%LV<p_&tA34KwHLU594Q?{uO3quCDe4Cz-UltZ)yt*?0Mi
zG^0yyjQeFzNI#+EWQ=S`uq3uH?&0YV3}IG~@f}iUtsMppDcRS~7+pM$?KtYar)}-u
z))K(fRKQ>Rdrvc&d591HWs~iT*T@(>(i{XQ3T-vm0NK{u`Pn;^<`>AHJ&qERIJOw!
z+UY`h!yusJbSh8mLtET=k;*5GB5GR|b~pFf+uNjC-J@Qw)9SX-vF7-J9526cj%=<+
z#>*lEjF-oV8a;HT&7j+*U8z#9ZsEy*@_3m-zCg3lq}r_0i!FzaoWvrWA-EZtJiC^S
z%T`p{W)dlyrb*db?zoJQ+fiF*cn{qFnI_fS5cNpfeZ7Gt3IsmMdHEF%7Qg9IQldm2
z*kjVRDcJ}aM+ko;0ihDl*Zn+MAW?nTukDn`9}KO8M%-B?Nl<u|JogC-Ei>p`4prSo
zMpcUAV@%F3&~0^i|Ft(cdh!fvwt!X&QygdfP(Um#k1EZ=k%MH4B^rA>7;A}zPoXqP
z9QMHeYBv}+k(ah#z*>>eCtT%BB1e=e&kqwi>w<k84KO-Fcpg~y`Q`I@^7#TrYsSaQ
zL|Sw1p@%7D1Fm0wkL{gpqW*v&=b@CK)oKyNF(w&8i6Upi6h+Yx5!JCKw(hn1o}^qV
zeWppBHhrWInK*;5N)T7^+dGI>12Hv$^qe$}K;U~mVHBdZ9`U55Owg*tBX-~ZyEAeB
z{<}d0nIJVInG6vG=zIZH-$m#KrSUS83yZYtO)kFq4o6R(L8u%-E(hb&%oPiWtVg!s
zGdVqpm(8O^9p(FI;r1$udM+v6cPEdeBS!o0o=52h+k{9a@XdYb-7ulEU@`Q2E%H$x
zp|W5#z8^3%J<H_yB$*&!c4itY1m_=rf;;6RH?Lmc#*M2SSU!oD_3(Vf#^xr=iwA~u
zn{@xvXDpSqNcVqne)coj^Dyp^&Bwg_`%iLs@o}#Gkzn<W_Yk4R^T8J`aw#k5n~0Li
z;5B1zegAbbxgzDq9>9}6;UFfHclWXIMm*^IWBZ6xC*>%%*S6~HZ0xbSvq7c0N4;L7
z)oByOF-wz*uYK_$f=mWq1xQaaF+NUMYZH47Y!uVp+oN5r;Hw<DLXN=q=ym&a`h6lR
z7%R^tXJFtE^?kh0PsDaBom@{k%xK*}qq`%LBl593#F#jlfX}#OPbu>Gf|EANWXR{U
zblQDdT|?0m1X5G2?hr>Iky0q-(dl$?iWvfFQ&}T-Ak(=1S0X<A)H4qvygtWYe1Xcw
zCeaO7342nZ6D~+>G_ePs^{~x8w}19356+dz9z2Rw0am1!xHl9gln5sRO5#%CD}<*I
z)@OIS#+}=%++AJa*6kbIxP61|-5uKPHenos6&zcb;F-slD3=3rnLL3Ckbxpw_K5op
z2K^3E+yfQk7c*qb1*|dbtZ&k2wdfmo=z%AQ3<yH@o5iQgaRv4#Yc61?54&ZQXc=SB
zW;ErP_UO|!L<(f8dt{PiEFG2r&u2J+6UW+3!YKE5BqK;$2;qc>(MV~BdR8BA9sQ?V
z%CkK8`O`dn{uH&!Iw!yU61DAJ<n|s`NIW4?h(wKD6HD-o$6%|@%FlkznW-XjdLcP0
z9w-l?9I}fLiSVuCm?-HZeTnoutoGQryUps#9qz2IaO>6$Zrr&^rBb2O?GVKg!YZD5
z=nzkxKR~|VkqdIJ4Vv}vJdb{7mwvB9Z2Jh`5R`IcN;#szfbGpKTHP)Ld6rK+Osw^&
zUF;tFWMlUcs?GiT;Lt2PtbX^Omr>9#_Y)(Mzn&6hjdAF#6_!B+Q4A_b{_baB6Q-nx
zFBKU9x#}isufM@mu|#lS0qZ+{s7>0)Q5@rWUh=k6WAyujy7(0do#Wai0YiEyKfv3U
zxMCv;g(6{n4=psAVu5V7#1mimJk4qyk<C(?m`1ld)N4)p%^rS`M|lbpMpUbH20ErV
zIZL)QjWIF8GDHy^dYWRT2^u#H_+p5jr&wu<;2AU8qjZn8j)=oPftSNr7l5<bJmaMa
zmY0@z=pUYAd~A}VCr(kR?C?+i-~SDjdW%l4jS9*LB&~Lv!C-(&b@Hqsj^q0ThEoAr
z9cz>|OiqlmzOnhK_j~mH4a)OT(kGA}fv|X6cUgV)7fj>|1P2x{zQRLdrE4?ApV<G@
z#_-qQe=8KqyMO<MjVTt#7&L0c7)r$g*<6{&zmTU@sUb6N|4q9?t=1xJ4)FXuLV)fM
zXf&FHT2q{y!Ox95_kmxz{~{S(`w#5DOi{*9?!N_e9J>9_XNPxvAzxx@VunM@2YKY7
zhZ!HA;^>J})T+Du<NyBu$<}U#cDK!#UqT2;y-}w>=nr|?u9$anC+YK_(k!%&@jj!@
zyCve>ql-NC%sD!PJ&u3*B`P~RMAui4|AE*ZSK8cs<)@sRE}{+{#i#%)<bOQ2dkW#F
zvAxdf?K`aAUE$WPo7}j4{gbhMeu5_-U8Gd<$p$$B6`-;{ISEm(M!(-B(p_X=@QWET
zr93*;)ONRMbvg{J;P9zOh_rFMx%A@w6tO)u0vHZ5N!M0&vDifd&+~>}t-kNO<UfW%
z1j@Gz27PoKqZ7J})=olaf4e%3?WD2Y`zuK5%B3;7J%c|t!|^XY$Ib06Y`gCsa@+$P
zTb;;px>i#A?hdOzeVxO<JAs&(MkGj`kR+`tXN?gj>eNDcxVe6x^*g&<y>W#PE`7ky
zU;7DLyX(X{1t$~e7)3t{DHKYynr#A=p<I|CmnmR;jl?6ClF+#3?nHhZV>Mgbo78Gm
zYOOj4k3P)A#0=W#q_6D(iGg!+00ay({N#TzB+>|0>_cO0;`wQZ7r+RPVfc4#(xOOx
z5<TT5vp9mjhOB{DyYCg77-M4+ZU5x6!=Wb98Kf(~N`W-Vfi(I)Q|<qhYkacE=f3zb
zxlDnk49Fio$k8u6$@-supLk##`f1&CS&Ta`Dp9DfTv=iL^<Ogor7t0h9uhAJ6sbIj
z!gvC$U4kn-4?Hlj&&u_?++4ZNFW>nkufO#g)y7^TLE)A=FCGIOk;^*dEl?RsxiK=i
zJQ%P>5aIm%JUk{!V`yzyTf0ZKUZvh{apv(alFb&-Mx%sE?4FD(0#m*QeC)iqiRcue
z6DdpbzVl*6?Nh;!hh)=VNXF(JufuT!HG+YLAU!Yk({t*i3dk7EqO|07mOJlWBcCgC
z_AAfh&(9&HbN4_h209)FcE=;UpO=!hq0=_!eQnu<Vj+??hf)%w4V_MhX0u7F(nE%Q
z%7q;LUWcq-Ad|~7SsFtQVxn4wcCA9xZ4+cYOtt`0Qm<5K)a!Kn5yws(bMz4Fte2!z
z2`b|RzEzrxi;<pYDsPgultCqfN@<}#nub#-P=YufkQD}P9qJzV8OF-vjE_w)J2S(9
zr9)&hd5#=^fWP)P|2{wa`H#4_xz34P25SX_L62^)%UH2I;*KV~opfL-9jbEZjb>qP
zhD(>PeCqv9m$QnB@caTnS1>Crd7WeBoofUt$JwtuPi}b;RMr(%(h??|U&n<SIc|*h
zU);C<z<v9lkdLv+{$rhOTC3Tlwl_exyA*R7I^8z;phz~C#Vd~?2NAkaqgB}@?zPDl
z0^&>#BL$6GjaH+{AdHw_UPzP%3@SnO-JWWUO(_lxk;akmbJBh_1`?i~*4DLg@7sR~
zN)Uz}GPx2d6(O6=F*P;A<ir#clVuJcIzr$D965HDzxj9nOMdqAAG5Ny%Bjq81d49I
zN4-&JW@^SIKCwV~5*fQ^B97x>5)nm_Yj=JIOG?Z6XO2-G8z-!E2<B!v_0<=-*{Bik
z)(M27Z=D`tKh}g;0^#nt&b>Wu|KO(_{@pxkei5Usi|yk6*zUySli2PFz-M=(!R3!G
zbM@vGzW<}|vbnuZtm99__IjrS(xct#;>jvwg>iD30@l|^d}1XDjiW|Q=EpJCu(PvG
zwN|CsX)?QTl!?h%Cs!jaN;+dRlg4&_MQq3AE-7liKeoF9+}t1AElKi~0yYxm`FI(}
z(=6um!`R-9A)}$)?hwWyp2>i)M6pif-;(qvHVNIxCM{io=ly!|{5MMKgmM>+P?GWE
zN0@*7AvS*gE=o9Eb;po%mPPKr;6;M%OV`OCKE(9-M~RE$Wb%cPPLyz3J<=ala7amY
zyT#Qj*ZA?PKj5dYzsex&1EYiNszT>bU`32lkjoX=+}=da`DFb93Xfc_2v#7ihw`(u
zdu_J2x2QMkw0mt@-98U59!~<*$Ur+O(vrk}9|{A*)~wl=tH~q???+qGAe|NmCb{q-
zooJNqLXj>=d!AxuZkm2=2W_0Jfw3l$v2a4KBAJ|&fRTVs`E&03Ia|?GUgbC0-Op|n
ziv?!pCR4f$!uObc_$=M)x7fXKJ(-$wHF9k9s6B1L>s#)=eUaRuLljS(#EefN+<Yh2
z4#$v?N)3YrLbATL&Gl>7`LiE<oA)nWaH5s$OX!5t08_%EO-L>q5QVV0wTYU~knsx$
zA;}f;SZk4FP=1zXr$ME<O{?9c-fj?T!OZO8gb$XIg50vh|D?5E^6SZY(Pp2|aYDN=
zA3HCOr?am;l)(PU9944GT&pUZ%`iS*V$j;7U<C30r0`QLh0hFzMdcFesHeI6`uk*r
z;n>SBB4_8_#21bz94FkWVf*wG({IE7bLfne*knS?lRg+ptJ3Dy#s+s*?{M$#JvMi?
zs8lMPT+Z|RUwhgCwnfj`!xa=(M-1WyanL7<Ljn<y@iT;jklnpK>g_7sUXSwFoFgBL
z6xuMPR!qjG1=eUuDM8==;Kau5Xlz76AK53UDx{l24kPmW&YSOhWOF%uKcL-dJLEo&
znVFvF+2>xQR37J}%O7xWW0P{JNE8`*y)NVBu~AD%I;DMUN0X3gatCl=ap6<%_5Nk7
zChWJ#<>nbjv7j7jJ?v}lzHtfBZE^aWFH<~p2<Zn%93Lr+qJ&$wk37B~YEJgwTE{1~
za>SQb(nlibHwWC?++^+ET~_bjWo=`f-Mu}g@`k_l`_FJd8dO>0DIZnzF;PT6s-rU@
zv5xSC&$ucRYt81?I?ZmAPPc>a<?*Dzuz&wuGNa9C|A!?=y0=5nT@WW6EN$KQkqMVN
z-T&AKI-MScvKfA_OeVwF*aW#;fqK0`xG+EppYi2Mp8MRFn4X#E{rBExb#0x=$uZ){
zQmIy$ot_!CrlpklzDHzZ(kbcv`~O+2SjS`w0a8f<KR|dM<0p=A@Wp4i^XIP+v||Q<
zCtPfg)4(V|MUuS_Zz8AO<j_kmFz~YEi=||{N3mV_F2Qrfo21`~xqj^i?|<+vfA;-9
zr`2g;?MQRa74Q;?p*!f&Xf-*uyvXME7G?^9sREu1$Yo28a)XZ_<me8%?Cfq)tySp`
zy3|`O7EjJOi%M&qiUy+&oUuvrx$on&!{p7C-4ZY&y?N6De3bY*Ovez11A;<I36jZ>
zL>|xgD3?nJA()z*rC+ZQ$A(t3Nf`FY<_ib`gMR4Hy1*fuPT?Yf#*)~cit+!))^C>7
zC1{D2Ac8D&&pu6SZI90NJH*;TBARFuK7zpc9qLZR=9}-3pPQyMIfJm{NEB9$V0%yb
z$V6B-?i+4gy~X<<{F0x(_7eti2*MfOF3in2=Z82(2!}{zG6HQPW@hIRO0u@OhVNyW
znV3y@PQvM(A*k$CSld{m*=`ZWntXYZ^7tG<;A4$(7k#pUZqh8x@J7|#$0y?)xiX!a
zGy8-K_a%a6ByW)(9$*+Cq*Pb|lQYw-*J}=SNiK5NQXAqm2@;ap$&6eqq?8yvswCxS
zJRSGzE}8t>(j?-k5@f!_;uoH$wOPm8+{PG-N*QKJA|tT)Rxqfx*?j#i4vmkYXBUyS
z2q;i0*-x2FsyXDU*9^IR<2FD4#Vfpb@m+K>B^YFU=4WPUHJiig9AohVuxJ(!9AsnT
z9&4NT2>dLQ<tZmYBUED2ZrRz{Wo><pcDGAx4U=;R$rUH@eTmgcVj!G8-AIuO(gI<4
zUi9b^aLB3)9M(8NTdSRGNGkW>^y)@(5otS1DV3aO2~sdMGs9-1hBh*3(|`J;F4x=P
ztRLZ9Ntm3|em`dY^$!r%a_q}5l3QFP<<^B^2)SGmj1(eGpp16!1y}i!Ca1!2YNY4W
z>GWB>d5gQZSGjWiLq59p5$l`lG}}#dY^a?+z?Z-9Ab#B<tfY`Fk<AvtLg44{x*ool
zArnWmdu^)KT`IK-l|~KWWeBn*{J>A-h=kLXvPsI3pl%v>w^<ybN8(iAb}{*~E&+E!
z;~JI77#K9x2IPz5E@&j#yiy*yd>%i@5QYO{9TE<D5F6%a7RWuGWnyfC_b<G|_D%&~
zTB`LLb2D=zj7vDN@wBCWU*ZppkClh$(r2?QIvya~+Z<P2ri7%E2%z@kn7vCYT+`p<
z^w+;c>BMm-jx8kCINqO2>=i<II9`eC**K!SN<Nxo`cNvA$`ES9`pR8auHWJ2t*d-^
z<r1rR?^1142xHB{Se`F_;WVY9kFbJLevDjDK=_ux&yxuQd@mpyNA$uz8(SM}Z*Nm?
zHyMN)KU2UD0<Z>~XnGF$lOzS#zWsN@5~Bc`HsD8vnJb)>O$hTCVl5aOA4e#O(5@9H
zr6ga-lgVW1_Sy^vJ;x#I^_iTU<>7PZ87qzR{)M;M+FD0i!|q;%6USoqPs_VWdOPG+
zes+7E?05nkh(tv}NI-dvojb?$dWGH_7f>cfVv!<2rxNW?Wi2uk?7er1vBf!pLJ7>A
zn;=f^DR9q1It*d8;P&-vtlYWDUwr@1XttYxK`O!Y)Fi(2=ycoev(*uqKoCYf%41{5
zz-M!P9VIg?%q$=hDJV~Q#LA=DY_YzvLA6n%ABK2Amhs62ydXfOT*x$zrjrfHucW2|
z06Nyg;@BADe1Fq>JfT$$_dz&{jlOSR91l);t|<0AXQNmgD`UbwVbmcS@+Ol6F^Pjh
z2*f@lIL#y6bN=fk^S^OYhj0jyh(L}{a^Uk%Q`_1ltT#!g#*E#^8BzitL2qM=wTqV+
zKYjvXK&V7|BFRFfa>1bA>#}ukkL}HEe(~0;L^^aX3TO@=TIS5D6MXRg2UIF;*OoAr
zelZ}9E!kXw`Nbu)He9}XiBhgexio?23!=Vz;BVczMYGe!^Rs001wtd)+G%jSR3`A9
zzFcgz^I$^^;r0|`vt~4$e?QUI_0zR>U&Ev??bow!vSuI1EQBN+3;;{u2h7aQ(Qh>A
zRyJIa)&_0jeWt%|^Q|$&QJnPUQ#~U$rf=;?Cg!)YM8f|{%2p#RM1o*(k>$@lPW8`z
zfbB&I3T4sB=VWo`B>-x-?vS~0jj0nS5q1iYSm6x^24s??_q!cdZ?Chyw#GXb-ojWX
zabq!@J#&iL=_%fS_X4d}J3&sN-;2>kkS~^4T3+VvomJky^ez*n2?~W0o`Qb2%iZ<6
ztgNok9fSng9GP5>ZV0=TCI=R$P|~5PPJ}W^=u8?Y9LkpRVD`5az@d&#?+7PkYDdjh
zF$^MZCj7pSRtoyP9!e>E-)DLGAkFP922l&G<IjXscTjJz6vVWRB+?osBz<d$2ZpV;
zKfqeS@t0pBJ3EUKPAxYMBYZCd!UyG8k5(iTkvQRKCu8KHJP#=Xsyj`tUcSOdA6?>&
zw|~LaTh|yw(Xfb8D7Gs#gbdJ;Wn*iDxv2%bu^fT&@x45mz<`n1ewS9W#pd=Vd$kHe
z`S_lqT5D0Bm?0M=lnTp8u{}lB6GT)b#h;Y!c~4t)T04}@ZGhud**FFvh^?SJIZmNa
za+j`980#z%i^U?jY@WfO$DrRw;L&Nf$mVjCO5>b8`v^fM%X{zslGT;#Y;JCI{P3|M
z5#J`=j5L|w-x4&&FgZDn?|TdepPdKnAg(ec6pOM=K`Qz>MhQt|4c%B#zp}!m&YyDf
zYhPjN+*y<i2xLI7(<S2vSP>-rSJ!s6I!ZDl$FUm8_yi34hTAt*xPJ92Z@m2)?_Ye6
zN~4zCm5yw(SMPA;#y!p+J<8_x7QIrBnaL%J*#ZKeY&O6ef$aC`^m^24Rkrtb=!X&6
ze2KNYcUU<10OgXC<B6?yyeA=tNs5I;o<$5h0BHvWL{eBM_FzJ|LQKc57!v}#9Hnv@
znZz?~9a(;2Vv4D$Y1-{3m1>n-F3;ZXHl@-exon;j$IntOkMqIB_xSMQdtAS9le4E!
zQ7V*yQ6ZN~Z}`4HMC?+X#ZSA8G2Fg=lV_iJ7C%Ai1QOvFSo-4gg!MM9i<gnspb!j<
zMoSL`#L|+J3SDn=>!+{c&&@EoxQwq#XdxXsmn5bx0W&n4Ep~UR+_-s-cBeH&Br}<S
z#~yoxy{&DoUaK*PG};)Vf#BB4Cgyp|?A#)rrxK3+Z6+sX$OHjcO}$>@*2*n5b~Z6W
zlFt`0LUDWLE|ar|C>MON!p+g{lM+f}d-{Qopp$){kqcrDtTCUo>9km)um@(OHRpQ)
zrDB;(CQA@_<g!@+PCsyttM9%=y;h^$Y)~GbLI?=Mkanj-p^$g&+*qSy=YN}ORYt$2
zu>5b7)FBeI&cMLN1ir`QnNuu0wMyreH!zXo)5OAkPzr&SF`m%KSg?2T8jWYym|Hr9
z4jdpZlgTF~6~vZXS8md6_qn%qhuzu^#bObyBMu!o$X8$fGOxe!I@M-_L6o2wK)V(3
z!R1?=J@X)AV`I!NEP#|;yLyey58tO=s}c5kAO(eTiCnozMrIIFbMfX1E8E}ZbI(4@
zlaD__KA%A*Ne0pdRmY!W*h?~246FHcb7Pw(DDID?gjcum3Y+jc5@Qwj>sAmbpV{dd
z{`TMe>qzNy=;$%7eRPrk-#`DS^rJW#6?gI}4mku98{)o`Jut>#BPV@ehhBjaFz7h?
zjkfo*T~5PF2>0Pd#AGuWW*>WmnUxi`-u(!vlS@-LV;tuk662wv2V3u5;_%bYFgbk)
zE8KYzDrwsYN!*XPef1`LI}L8EUZd0RB(1Cgk34#o#~yo>pa1;lG&>!l*adGX1lwCJ
zuHCxF!9zzWmBv{-pg~HmU%Srs_706kowz?hdWzy$ne2FhtjG|Hm^a_O$a=-+bI(4-
zBM(1-@5y2Ej<nd+a>(wh{?lM#C8?l7Y*Xg|fMmd6*d9<=C2;N`qn4^blJPPuFD>z3
z{TKfLAv}(qIKeBEWq$abKVcBYPPXAwt{Gp0@CYObG$NV2>ml*Gu>S6c$c)eN7hl9q
zkK+kN*pJBMl3=MkS72C$h+{}};godKYeFiT)h-`iyv#fAzr%Nb_?J{0_Y0{iDNxeV
z4|{CyZF2naC#g5<+}pfI6j^2`X3*NuA9NY?J8W)ma{0<dHurW2as`SRpGLpKzx(!|
za_ZDM9z1)R<HwFVnG2i)O!_>z&wRg&gh+;TlIl2)hA6z<w{HfCpDi*rIh%-zk6Jr!
zSRzLzlOc{Z%~p#b$Wg8CQ68T}DuwT7IezRk*-ReKSJay|y1hQW@?8ttj@mr?6T9@Q
zTq;s56h70Wt`OHal*{2KNw^=IM2KC|v4*@3XszyX{f9r{wDKsQI)Tjw*nuI64S}DK
z<y9g<BxQnPON8+7RDkfnYRUC0*SK->I^X__Z*%p=M~P{<TMjt^<-wh`Rf3$yVriMp
zjdgncRhH+MDdi-Ca6lM#Xf+#LzjcjkH?N|Fq?j)ul;O>H-(_Vp;_T^@Jb3mDV`BwG
zB9vjd-}z$y5QpRWIR_cy7#TbH8Jlo`Q~s<*F+0787r5kdID8`nzVDOIJ6cDx*&?2c
zX*TP0I&F&OB0?ypr{{U{$>#`y43{qcl1i;XE>j@Y>U4aqk(<Q+Gw@k0Z8RjVL{cQ^
zkhJ8craAc1vutdx6YaK;LSZn(!s6l78l;NxjHR_+W8=e%96I+TI#+aDFd-10Os452
z>-V;(>{h5Z_PBcc3ZAdXWV1-6`Mqy_op!6urQJ>X12-6`v3OGP!3Ve4tv4u7OfxYx
zgSM8HTPytX!*{6c?a^tqF>y?g%TgRKll3!XrDV6;=1>0oyKL=t`P{Qla&UPuiS6J-
z_EMCL<h~Y(A&^Rx3sNN0C3em*NDOHQkaQlT8(QOIfQ=_TiunRhKJhp&e*Ps2<uRtF
zCi%yWU22t0XDXl*H!9)Iqcs*CMnwIA+lM$QC@rJH@;|bEv!u?wP1oCxV`Pv)=6sf(
zIZyTGD!aGWA*p_yw1Gfcg{Km~*lL^ljTPd@w0pP{<WJx<getq++`f5(vEmH1N|iz`
zAd}CqeE10e`G5X*`ObH~!_M|DgMQ+r3Is|LY4~UV{4c2NHu=H}&oVzhgUS>+_0V|^
zo$OPu*V(vtX9$ajQ5TG&+3s@j;x%sFdy{)>>+J5-dH!=xGc!{L;vr&?ZtlpsEFsjo
z7e91_7;>YuMo2Rx4jUkh1sXX-2a<iq^E`^hB1exNWomMk()bh;6O&xK`VqByg-}QM
z+QaJ6L~R^v27|$H+&$&nxYre7<%m%GTUkRsm|L-qM-#wfMUH;{DVnPr)OQ*nM#!hI
z0)@ht5@jr%jVi6XYec8d;}5N<((|HOy>pALjZNZ#q`J4q<U|Q1JomY$`TEzt#vlFB
zAF;PvBZ^Go+X%jrv^ydH=>PSvxOZ=dr=C2|<kUF%u_+!n|0GBIJ!+K-n`?JnixUW=
zE>`<AYF&Q$&IkPXwTo<SZnC@6;JIfXqgcv<jY$dj!!|1nMXSxeiP!tji%ZCbvO|kA
zV<C#%dGUtiPWSzSz^7a;bN1|6^0_h-GjmC`eT#NKA~K(T&vSq=lDu+$!^Y*S^h``q
zW$@dEwVz+WlOBsNK1pVLh6sVS4vBFHeB#vLJD#T%jz?#eB#s5QuHE6o%a{1`@BcZ?
zRs#T~B-w<U<@*YwBO+~B-`YV+#nSR&^7%Y>SJ(LH$|YiL2&0f{WtU2&0-j`SYLX!E
zkSN-<Cjah_f57<EYdrV#^PGS9JdZwdmgVIIS8R~zr6texP}o~}SfQ}#D2g-)B2r&x
zAp(cO8W^9PCI|vY-7q%c<GLgv2(oVS&06OzjiKJyrrDYx2*%ud_7sat2YE_<o{Jy6
zL9ahRX2tL>*)MjR{@k^0Vtg-PdTNqNwf3p^JY9k4G6`TaNDE_1va2;gG8EI3hB2e)
z+}P%$?)NzMwGLAcpFygeYj@^TE|nW@5<(<;k**N7$_+)d_Bw2>@9>kKzrvO4S1`$d
zPvCn5zK`cAgfQ4xbN%K$Hnw+o@a)4BN+q^7Hu&J93v}CE22n((+orm^OSMs>JUKzJ
zT*Q|OV-0t1U+3F@@pDd`IL-6Vyubrz&T#(x0~Cr`q%;^aO2UTwpPqq~AYvdE#_^h@
z1U!%)u?1fR6pKaDxaV4A?t3~%kqLqzL!;TH-Dy+E3pO@ya=<%|wLU^ZK40Se`KQTd
z0~*ae=BA<{iDN&H?!Ghe85D2ny1jB6un9`7T#FeO?+V8ca`xpHxcu!O;<bD9V&i}d
z0n(~O&|Tw2hQ0T$Qh$DxV0wl?rYv9_LKa2bx_K38GT7LHG0e?O&>i&n*6;r=Cr_T_
z5C8BF>2?O_1gVw@x9-B_JN(!G_y3M>ef>oqc<?l(Qi1t{N10q$rqk`Px3fcaZyVq9
z(MA(RU3BbGZ&Z2Zr?2tB<!h|0tn$L=zQFnOrwKA15GOUX*f+^OIs?Xz?sA>9)YG9h
zYfU0P?Cy=oKr}GKR+Dm|l~Uw#dCH|RvRRL_XV2mLImV}E`D=gU@9{@}_|Mte+hzX1
zL3fvEL%ZE(dVKOeNh7z{$&l1vO;VR2Macw6dXW9eah9Hcl-gbeRcY;GpSoVH5&|U*
z#FF~$yM%+D8<y}Av0*GXuU#ge6Byk=ifMA0EXNKXBwGylhyT0(f_lBj_EwErwT+Hl
zVE2?H>-%UlyR8m?_!mFs+uwbKT;6HJ<nwtZCdQeXoMLunf{Ah&X$+g&EpFZ2<I0WO
z+}qqEj5R-e<@;1?6&j5SFTMDA4jx=UN`uz-S8+q06wc5XfGBdR@+ptbj(Xujx)x2o
zI7gv4hTWgkkqFTW;anq(#q%;OE+6GfU;RCP`m-NVsa44YbHfR^{iJ%WHDMSg{K6C|
z)EFCMtR;#*z2Wo!nw5@x4{gw5J)9oGLUwVE#b+O;_1)Ky5yRmCfy8R#9xM^Vz*1XT
z#l$f($PR-<97Wt)xk0y61=B;xBIQDnLkAX^S)As7`w#yCmp-`4?oORXvpc-F15c3k
zJVG0@zE|g;{>yLkKmVIQ8=jX!p}@rC1k+Q~%ubClQ7#en2dr&1xV5^=M>lS<z1tu%
zhClg>Kc-f%Qm^mv#g{(M^z@ix4;eeG^oM*&;m|rOl&iYqc;9(}v{16vf^lTzvB?E8
zx!fn9{?y<-iXzBB5ac<0{0uLD<3Hn_ci*IuBuJlXc@hz6l!T0s46G%PicrUdDnJPj
zF|e$@_7=wUS$^>a%;XF@HtzR?O?V;7aj86!=m@$tOugP>b8DTS{QSo>TXn!)k`F)p
z0D<zja^(vBZkI?Kd?~qlZJqX@Luqn~so8my@3S{C&c?<jgSN#ll^FLFW8>o>B>h2;
zol3|rFWjcl?y}ds!P@#gZrr@W%F0tbcK#8bc=BQLc^{({)@p>bZn{jWgc~P@b2d(v
z21lh35|jkvwCGCZ2^Vl<?KpagV!biodjXkD7Aa&>ghh0^JuY6n!1VM%l;^X!xPa$*
zOiau&F+NAL)h6SqVX>CFR~Tc7V@nhn!qA{&0oG$~W|nI=ZhiVaPgZ_*l6+90yW0X|
z$$A16JH0*+K~G2I{T!YZbk{3f`?K$HIvP-V<WaOzV4PHu5C&-_2oIz)IhQI$@Dy8H
z_qcoSHt%124`UraI+xG#?9)$icjXQn8;*Vw8N=E}gKM|$aq84*hpaCva>W9hn>$o%
zRR+S~_lqn~Pf{$F2&0feq`A4W!;L#zRO(gs-hPAYH?HyQQ_pbk-cy`E{}5-+9>r6x
z=s-BJcF%J{*fwcPS?vlmGkgXUITjRX7@wFyc>zWnm#id(UE1O;6v_md0BbGncAI=I
z&&tY8nynuB{1o{@j@j8+d_Ume!INCPut&GoCFA>}pso|-O1lZ%D6~Y;XNT0)>a}5F
zBZN%km6AeO`HY`E&g{K4x<7v(NCy6d6YRBG<NGeY_xGCY-MYo>x$|zIGD(!$?Iz92
z7FyS_wuJBbEY8l;*EP<ZJH<D@`Mdn!`#+&pZPV*Tka(zOJ&CW}nfu`S3ODZDB@=k$
z@&WmLj$*OM<m3c1(=$v@lo&7P=(hUY-KcQ$&JI^@-lE+LX{_GjU;oj+q*C3ZviCV&
z_`-7(OPL|fVSij7+Lz%5HnrBH_?|3dtrMF!&I(WCGdVen^1PwdqYwhm_whWRZZDY-
zms#?KGLJs-9QAgK+qbUM>a@uE0orK#{r-LNJ(V;X>J$#e-v7hvH%scQ)5mg2o6|;f
zskmh3)M=I<x<~8P_Yo#Rq%dfl9#Me86OzH&F5R7Na-~U>Gz6Y=m+*YW?dzA>-nz%c
z*b=idvxLm@#M6(nw7AH<)pyyrw?Q1n<a1s+36Iba#+pzYJQd?BL3a>St#=TjieQac
zYX)&5nSx*t#SEeZ?o6txP)EG;!8`PNJ;E^J%U^zh6DO7vl*P3^oy>s3ii8kuojigx
z7LlMX(k4@@#&N}r7UT*OOwBDIq@R2*F&u|>F~UoNw?+c9lXE=w*fYdhbL-|cvbhEr
zXH(;3Oyo#ANsI<1_K#$1ZJglhZ(+HLCdQL~@;N!;m=)TIJSv~%M<1bjW0lUu71D;7
z#Sr794}3g4YiZwJW6-GLW%Bq+;v<kjKpcnMxp|Fhwa@(ANe&)7NWRhI%fItg#)?Ji
zwJJMXyVzJ$$UEU;30Q53v_Z#Ctvk|&PCrD{+6eZ5yF_tJXh_bwpdUqqaZ0jv=cO0+
z`N=PSLa*0jFbMgb-+7Vw`6;vzm?TL^73~pD#>XPuIk7lv866|lXfif7t_@wDm|=Ws
z28(n758<RrL@Gk*C0r1VCnN_J4)XL<pT}CbyZY$^{Me}@46fcrB8CW!Cw=-RrjH?$
zB$lElSbzNjS}K-a_&hx$C>G1^xk%6&*P>B)Uf?)rk!EXskF~YCY;3J15<-e%InS}9
zhxzEE59x;yaqMy`X$3!f?GpLyzvcIT_bW_JPS79pX|yA>Umz&u$f^=HXwwNbm7OXV
zE??*3)w}G}>dD}f+pjm@f15k^?y_5P!xUfm{Bz9A%%IJ_N$KH)`hJ}%k|f<5A*ndx
zvqiJFyGMO}hw<@o%5yWxz=~^W1(|H(XrSr!`*@zmAPU`(gXhz1HmO!CRI4?vUcbp?
z*=M{|91^k}>gQ+*&Nk2_-tO)-j*pFf>iyoo{HaOKe)+3h`p);U8{2q7kX25r%vT-*
z6Vb;~aLrZ1oh~=N`x6ethRKJ{(etwenG9H0^twSQ=TzY)c_AJcD!X-V-nzjc>?Kk-
zmWBBlCMHW<z1Aj-4N0C%w{Q5T|Kp$0?HZnb;t>kP0-au;zV)y{ncP?r<Cp2S+tk~d
z&9w?|ynlr&x7O$k`pKY!VYgo4yRUqowe@xG+`Gez_2)VN=))9?MU0JHJ3i$;56{3p
zZnTuHXt5HFQ~1vnazycft+g$pMw{`OX$lkL#IXy;p6^j86!CqZUayzP|0s-!nVOme
z0hP)gl}d%Jtu3xxxlYc90}FHa*^!9^`P4c^i$1$OZ#J8RQHV_LDT%wkt<X-}FPG=|
zb5GH_w?+TPMnXYIJYi8}N$%^|QoVeQ!53=Csabp_@q`oN-MHIgYi$)N%jC0joH}`m
zdUu5{eep#yzE8K^rn*-{U?~;?9J#^qUn6aawCjX~v7yx)AS-QP7sQG~u1x~A7)vjV
ziM7S-yRVy_7JvD}zogk}((U*7;uoJ|c6Nd|7OtohB*SNpmoB8^raKua5H^YL%2CU7
zqI0NY!Nl|&<*^C00V!OA-dg7@;`=_JoPj`~Q3#G4I*$GPm&s=GZ13zbH8qYglK!Bd
zxDF-W;7%87nAnHECe7kEO6nYWR5_LPwBer+yJbGl@)M7+adVAm-FbK#>E66kW|t@=
zy;_6aD_5C0b`sCaf}eA40V+$V1)Xk}dSj1FCd=VNM|k+*N3dG3dgl(y%Lmxm*`n2I
zxhghHCxu))G>l@f`{f2k@M&aRV01pKjt`KcB<X*nIOg)zO9X+3@&wuJS2%QV8LcDS
z#9PuH6E0CrTQDhl2~Ktir7%(w8}Kt(X6BdhJm1NUFcPMfQsO0w(pqaLDKH2T2<B(!
zdE&8W(6Q#u>P_-NKrW-)!Ai&T#fWE-wqW;bS?S+e`!%@|EeyRhag&hlbqiFv%;9Gr
zXYcMVs?h_hu+sV8f^_mDKx2K2`pO-~<`z+&kM*5c@F22e@@4LRaGmL?L*(;CPR^g=
z?AZs|+}dMzXPcu(53{kpPOsM^ieus^9%^W5ZHS`S4V9#?JK2{gHiXs*Ke}nh(NyF9
z^D+oS-hAhEG8qrg6MXZVzr*Cj1X{akZRePfSl$W1IB4D)H|UZkX?PwOkJwuBg)-AK
z3$6{K-Lr*AIbS02!A;#9G*KAhNsoiehcNnCy4~)l-oBGx|1zukFNts6B@k*T6pc`X
z36E6@i!z`(nvFN#C&*@*di*isnOU-4(V-ptiPBPmfo847?%ob-+xO5mrdY^hZN%UG
zJAc5=&JNvP&pC-C?PgyJI{k=0`QB^%<h8f(eGiQt(h#hHFp3Gaa{(KKF=4C|uHwEU
zi9^pfcQ*L*@BIndMzq^KzVf9PSX`P11lA_rr6RFLveNN(r35cXE(}MiGjW9PWwFNH
zYo3=BW>Lh->TL>z0@Kq|OiWBrE|=Ne+h${9okp|4+WICJE?gv=b*NG<n`3NjoTa5D
z0^wS#g3*+#^VuB@NWiESH$VH?I18j}uhC%Q+=DzkmgDOG@^7)5d*r2}85=Unr=w%~
zR+D!p6VFKWc9YxRdj%24h(|1?@o@r?#dxDLAY_u@X-l(Sqtolq892Yod_ITrEr0WG
zev3<&KEzs=yjW{dNe^#hx5<C|PyY@7>|g#{cLuCQ8%-R$1Sbrm`_F(jL&>QT&k4dP
z;=<(%++ABio0yG_9bW#*7nqzH$6D>&VKDp7Kyo`M4}@gsJdx6D@DNs!2{J^o>j>^)
z$jzJAnVFqoY<!Hxg+=PMDvf59l{>eo)~al7?=mqyPmswGh9QMQk%@^(PM>*@)Wg}@
zkwsYgWz#eG84kg1Y;F)n1M)#`NEsOPyAH6B5-AjFW`<K=d7ftFJBVsG=~Nja<50_t
zg1!;#T)xKcN0&MB{4~<@z{}z(4{baqXBPP7oA2_-Lz9%tQ%o<od#hS&5r%zEo;=RR
z#yZ_@mpFE|VG<|b6h@jbvUuu~@jcdt&`3-g14;L8wD-nZI=wDGdG*JHVV}XE$2Y(6
zJIu^X5$nhy%hJiUnFOVEG@>NF@8ewxe2lb226BZG6O+@V0>&iD6jDlZxg5D%4k;DA
zo^!kEv|5zQlN>*G3f~WS<BeC@+FB!*31~K(L(peP2}w|UXBGNa#HmXO-TMi<Qb3F%
zvoO!{V`sRr`zk~NtV=$Ok$BQsZs{my?cEEUdg(=iq8rK2GT^<Revi`BI93?0UHg#Z
zM;}6FCkY46Q8NrR%gakhsi{<U2?ws#l}SV|AtKUl6QxWOCs%F9u^}FIgp-zUGDVag
zj%?Pcfk$D;rOO{MJ2OKj$ntx?`&Fi<Cz1rvBvXp+p}J3BjFIxZP#(fKLjT<S0Wz5k
zCTXF-h;x%JtsO_s7(*|SYf+wLc6N#Lk3L1O*JtI{RdPAUci4Zqqy*}eU-$7A?Qdm?
zbXbE#I9Mc8!|6pW(Na)8a+tXX53~Kohsous+=FBUN;r;L*zU0Q-USXl{y0LG@ub9y
zTD<#{?{V<>G5UipYwLF?muAT4Ji46$!unVvSYBSD-*0pKc9kf!U?7`uwoEqGgh@3g
zQxno;ay5#bKWxe%aF@pj2^B(+%jJloh(R3l=G$*DJv~Dvo8$Ms`Bn1yj0+H$#IMt0
z<-YS`(;70BE$}f20xw{8{y+lb-+x|UlupL6YrAW$>2$lqu|`Ud<pW1~^pPh%_4XC!
z=Q#fLFLL+Ke#qd~CcXrL!54x^YqU4=!}NrQZS}eL)-TcJ0u##z$z_XJWst&0dJffA
zQqXC4xpnmhQGY<M+heR$Ad~fZ?(@%c_JId^`|bakR<lbQH+@gyD@9~uI(?H6rxU-_
zk8xi58i#bczL#rx?>qm#?~}=7=ytm_I~{)b%J-9GpwBnH_A*OLv!hF#6qQZv=!9@d
zYBDH+1g!^b@9Y3D==JChy4b8w$&=i?d7bgGapq^|$Yt|nvswDRKGkZQ%U3TkKX(96
zNuGZ4Ddy%6pyQZsr_1Wy723@@%uWnt-iCswI^p)%k<*N|*2zF5&K{rH3Ove(4|3w0
zU*gW6{s@1ghLRw$_)-w*7@K%;TP!|4rrKoX_17_DMJ5g$B;#j+#AeY`!{Me{qspCI
zt3+W)uh(U2qKxz<-}v2^nVFeqZS9A2I|HnBK{D_po)CmKrrV8(O@I{18AzU@G_d!z
zlLyjCBJ}Yy5Cj39=h5r-s5I()_XmH8MbPhc`PR3-!Q@2IA-l$6Q{p-#XTTwjPT|4b
zZ&4JowY`nens6|nGw9-(JOPI5*RK+WeI7dZ2)SIroxM(%X1mVyo7Y%cI7B`d@ci>H
zF;<?$T0^7W;Lhr8y1gz;k2^9ysZ3U!P;ktsDEthUOeTJ1$^%cnzcmIW1==c#hYqpu
z&@t9tdmpq1N+p9F&U;)U2s=I2-hH3LPd$Ywlu=3%=nn7v<a;a~JwzDwxwCqcsmTN6
za}#vCAy@}^EH5t+hFxyntP+N<c*+Kjwqc@}D0b&<C|;e!lPF5iG1B;cKX-lqeI3Rz
zZ@>2zGc(g<GC6+lcfU%pl*j4hC9!;8?vOrzuB)E>yUGLO24UvX``V;bt<k?pazb^i
z3Bv)#Xu91lI@U<3IIwht#~us$#V`77+*@IKvY4Q-BOmHiCJN$TdpPDdO6s(Z5|wjx
zAN;G7hqMag=b1fog3ZfUiEr;Btp~y(btpGk6If{8UZt_KLvijPKAQN}`wXtV!=*0x
z<KtZZ=r$*hKZ1=Fdpi~KWy|M3|00cgl|TQ>9g5{5)k*__LV1cXPRG(yP79Ed)kQj5
zZ-)KzPkuI*C<+-XmuVYI7)8AE{@W}r9^l8X{ETn?-q*=wl<UEpq>{I(j7jP`p^!?D
zPA$b)CMKrH<Z^^zgwTnUp_0Q*8|vaD$eOj5UcXHg4G=PiKruhR#AA;=MZ49Y-EI(t
z5hyX_1MEls((k)p`rx;*MA~M=IPD{moR`G$*i#<LN=&Z6;v;9N-n@g}>Lw(6&_*JZ
zL)EedcCTEg*KCj-pC>SV+8@3~>&kn)v%Q72G4H+q4ljTC_t9aGt<7zMY>6*_`FFT`
zZ-w=ZHM03EQPdmqdUPDRhs`Dv+N7|p81bfte2e|E({8zoB%)j@(rC2k^m@Gc%h#Ei
zp5fJ3U+2qT{sL0E;9#)B$*2_3O?A3tg5DVfXv^gEEXwl<!w{Ja7q~#;+W9J#*f8ms
zyQSN0p>>ENKuV8;%ZESx{wYQ7;31a3@)GfPe@18JF4k%S0c~xFtieZO1Q@9agrT$B
z=I+nmVD{8W#>N*RV8pEjbQrU@v(DPR6{aT+Fj=19z``8Gu{^*3t>0sNt3jt7G3ZAG
zzQPZjV+O`i7<3GSFv5DtbgJ73H;JlE(r!s!f5PP%AsZ0KF`2+;dSaYrvqiJj=8a#z
zP9ay~JKy~g-}s#`GB-Do&ZHz?N!tylxs{$1AqAaon-4y?j<uG+&tR-%X-N=dvJ841
z;(U&Ft3xKpk<I4lcDuCNEsh;N$*Gg4@sv;CSt`5t`0<Z_NV8R^R@-H1evV@Y=L!74
zB^Fi?$C@~D7DsN1bflq`%Vf#tb2OT*PrdH|jb?-1Af%8DD4smUk>3kg{o!l$Z?0gB
zCM%>v>!c=+L<@nIF&Sa$uGP5p-5*h&pQG@|^UmfZg*vR8-fOigtgc*VcKRU4mXBj}
zi%h}ir7wJez1=oOE8<ua_>v%SmR>^RxTdk8HwX##pMiaayVj}ZTuRZu{|rQ7h^Hho
zQ{%K+Z7S6oKl%BOiJy6qZ~w`6_|i+yao|9j+#77-1Aq`nRKlTlQ^`UK`hzYPF23u)
zg&>PIg2*U5Kfv0EW}`;4(WG1&XKZYoFpBAR`W!!eijyZ#;d=ol>QJu@_|cEPPo=s?
zvsqzrZk7WJQ)Dt3z+e=eWmHss7sVe%Y3Y*g66t0b1f)y4ySuwYItOW_OGH2fq!}89
zA*H)xKstx!z3(U1;>*mscm8qC+55Mvz49{W8SiVNmKVn@W*h|m&5+mg@c|-HyzSq!
z-h8FyDfb5Ebuu_$@BH0e!SI6|^aHv`rs<S~la$V`p6fu;1>=^{?pP5}Wv!#`p!3T5
z4Dd%;`E+)v<|1+JmLfvDdjRU~QHIfR(_iz~MF_aQB_y8IHR+;Ja!b)q5mCs&k`*Wh
z`+Z}zgv%?;gF0#WU$DH@U?<tTy7RgZ4ta_%07_Dfj>?-mA(QXPxXUZKjNUp)n_@lg
z&UvEY@>b&<Tb6`=lIw>ombyp!0+Y3)RvuP-`W$1D76C((b~<~wqE>aUOHYNHgNX8p
z+wUAkX2+;kTp$UVZ<*7+;5#4>n7unE;D1GB6>-;wTYU64{n;WRDFhO!bYddClMAU?
zRrLgQJ?p}GV=lc-A1YNGcc{TpwRhqVm99KXzbx*>1CE$GSX%rar6WiVot>r*g!GLJ
zc4N!)5eKHvDBr&$g#B**>-V6x+En~6Rw{_zqZL;nB3R$2?wAm@QB_ST8fDkE{4^K>
z1cOD!p8#-DSlda+=U8tIIwrVfuGE!T*~@WQ0N77I0sF%pKtR}Jwb|Ky?_$!b+-A`I
zO}OHZ@xA;BA<2yFl#}+ta;s9cLHU$pcrvB1Tf_xKLd8GmTQ@CzlJ^(8M1xK?i%iw!
z?`s!LwEzra^*?8eG}30lbQR*c+_?4>Oo`MI;kK^j3I;nGf=8F0(l*)%rr%%Od^4X0
zIKyJdmRch3_EeJmGVYE!_p(&mx{t5Li(N)mXGy*u*ZZy0a&<na<Od!r0qvLCAir2A
zx^R3(9aXr$q%b@(SU&gN2y*{HAAI(~o+>OUyC6?W0X}xjNn`vN<2a|=9I_>-LlOO8
z&jLC`|FftaHrdNgQ~C0=&JNU0D<_!AK#AS#bIWjS)FRtCqui{GVL%93<Z(Y#E`K+p
z(%3XtF+*{QDjxA*%JcW>>`{}TLfiwE<Z-z}$~vOb>Qm<iAEm|L!H*R$tYLh?tylT`
zX;uuo*hGp|?F)Ox!5kOE$7mZzEUwRGWq=t2czG;iJbVIgKJM9ls7)?08rul;h><Vb
z1Zt7xOrNQ=3Jz@gYz<0pWavyAJta0es2@OC)n$K%4&j`b_{GbXbdQ?`r>d62KB4i-
zY^S#Mt<>35sB;5;56^<YFs{&jYm<Ntnf%aWeboYxq_uB!NT+m(x}~ix%DASBo)p_J
z)6VXz%DI~jWGCznx|G!{<9$Ls4<;m(At5DA&R3JboY0etgLtL<`DgxHk&RSUkVzVx
zc|RSP|BE}$=zb_w{vcJ%C(Y6{@ay>|hkSSecFHA>on%jR`YT0z^x^em>dKl9)y_H8
z?O^0_B84vcLyW9vIPK>hlxTnhiU$D)L{-2QK-PFVn(W(eDTX)hlXM6jZKV-6f8lRq
zQu)ksr(eZ3`Lt?x{=?SeFc<!y01qpZ@o2#o6EkeTf1=Q04}3VZb^YtmXiU0oX={eQ
z)$=<Y#3M3_$PWz5?0gUlI>X{R>$w<r|JTPue#QQ1k{JN)+1-T}{ZU{ZblVzu<6_h~
zhTDNOc3ElvlJsr5(ot{9(x!DJ3`L2SI8SoTz3C5CT--shtyyR0O}TM+_vet3*cl$F
z+gI5yUq&iuWtW|vruB?l4UUi0dmUp7o-jG9=JfeS&%W0AP%U}%?+vJutx0fj*)Oxj
z=QS7RzPEv&qTAK^>r$P=D(8tKU;&GEl^_asai7*tc*Wbgx_i-B6G;h00EKCf%FA!d
zG;2qpj2(svKE3uAQ>m{y)T&JoE6BBt)jJ$l=<OJHhghY2d?FB$OjXf^6?L{U^xr%2
z^L~2Kq?We2p(|W)X+v9}X^<LH@k3oSPqj|^RiYTsaa>#>kz&W;pxKCJDJTI6s?<o_
zq@S&*H-tdj9)VldN3x9y><$EkD8t-u!Y?09LqpUfxgs^XmiJ29)2$W}k$dVbzq4!2
z>$@r!bQwHg@9xw%m~!}cvEp<+)^fT1mhRCdC!kEFe70D-=2Nwf-+o!<?TYKu8`V4z
ztY@=by5&-y4pb&_=1wiO*i%;`xE_4+zT)5cEhQQ*CtZZ~)iLJGA`Z}YC^o2WLlY0{
zkXv>AW$;_d{F5{r(30{&eA$mxh}+?nLcd$#%rQK$qcT^{#p5R&G^7w+!KtkDonKtF
z2n=MwCnn(-;o=kFk*;bW%zn(;mNY0k9RDYw^uA=xN@fLT;u@8{?uI=aaeFM35+|OH
z3v^aWh^B1S)Q$4{F?~&xH`lhuuzr|CFCIY#SNsTF9wHV%lsVX}$jUv{f+-lz*+zep
zq$9iG8__G~6`ec-X!qL9!7~Lw{&NKSj2RUb7%bE6YT1Djli9x@p+3(Y7n`@#z<9X7
zU*u0Lun{o?Km}bpo&=#L61QN5VXz1v4xSgHKiEH(r7Ak<Jz4kiCC7pkwfXxao7B9O
z_beZrm(%LYCv^(TOlsUuR>gtDlnO2O0tNb-2wIY{<>1Xj`?06v%t5t`uRZ!4{r8Ko
zS-)89M45q%BV2HCtvjxTZ>{-?+syPO_EOFgc7XHswIJJeJjZ+L?Zbt7W!16Y4K1Va
zR2HCLh^9WoONNau?JF6eSXAmY>yN7O^<n}~`ln-~D~&`)bVi@YIKeAwm>Ok~)<}Je
zcV-(>XMTBkCma_{uNw6=!MmqGP-Wmseoh&EWAGR#mz<}L`QKiC<98IRT5cUGT(fp|
zPLI#z{CmugQG&iU?gq#dJN+(<LvB*%{18wqJ!iI^c;y7O<WW(jTJAVIU(teOK83aM
z5QE6H6k7DG{BL)_F2qUYh>D@6sTBVy<^0SNbc1GU`LO;{L-?bO*`*u(G`SFe#@BWI
z`8L1Aqe>bJhCCv^<7Ezx_qEWLM1m+WU)N8pM0&XreQ-h+U}9Y~<KTI7YEB(EvLA_q
zic%bL`Kz>EeR|fPvYZpQm+yn<$3LcCkEym@21;a60H!M%LTP2;k&p*f`DN7BXHqu6
zWfBUb>zOsH-U1Ly>Z`||7z8hUQMtVWoDGgp!%gCSEFlduOP7IEjg52uCq2OAuuYFP
zb+lHu;^H5L>qT$D(-$CBl>E`8K%;2z5P0=YNwkA_jw%fXP}jQ|_x|g=u<<S&Zew@K
z+TbJJ;w2d2iv?wz9e)L*FrqUpMS%X-f)YPcAQ69+dM0=fCG@b}fm3C3-;2ca<)r5f
zF0e4#xSV9P#l&xWR|)bu7dA9}cX;IT*X7dwaN17Gw?g|JkUBo~kk|Q$D}r3#q67z_
z58L5!M?UudE~z^?euDJ*MjW{8oQvRaa^c@lJ04V4twP~|Oxf1j+Ij85yVJF8s+w|T
zLDf#309I%@JPqJd&Uogm$s+)43^#h9-iVu?6uK^Vr21gBq0qwOqLrhgBdb{B@82WK
z%LE&=<O9ew8VT~OyQ`SAPIV2^BPWIl)^W+NKF?^`jKuk`Q^!4T_AhP-Z>`=fhm$f+
zM<nno`^=*2tT8V+(~I}VO7jJEY`c2|eSy)0W|j3Tm^?1L%X;CotZ(ZYm^w4l_JI4`
z``j09JMOu+?tb^UdWs@1^te?5XTx?~fT`RO4`OxEQYH_J_{Z<;$o#=+algP+|Iy-i
z<{|ZM&9u8Lk(?~45*-%U=C?22agisn6)o{V?P}@=nOS*n#VHM>|K|CID3D#jcX){~
zWlB>n?}Hs0*xiZf(k+~-lsPlYChUOCaL-#B6-A{}g6!XXDZ}`_G!5+6dqjxDQ<gV8
zv~{LZyXCg`#4JZ3;;&zbMNYmJ;P9wVX2o*w_`;wj5jWDN$e1U7@tXFv|64wjwy@BX
zm)Vj44sT~<Qn-NYb*_otb}S1ub53ph?Hw04%6z*g1e&M9Y1+}#`7{?gh&3J>X<`0F
zI4LF0-peH#hxmRhK|aJa=I>*m$CgQx5@)7$KtM`Mi+w9p&{xM*y{4(z-_lyREQ8tX
zI7sw8UxXLDQ!8TQvry<QR>#GP1VXOnAsH)^t5&~eo!Dd9*c4d)tJ0WCf(QJy9Dk3l
zqcmRSpYZU2qfP>+53a_|&}1wcP<xPMf$K!D&Ib+$dI*F`Flo)EmW+W%Ck6kd8tbJ^
z;t`Sy?>dkZLnDbp94xL@Nx>u5*_Gq^m|(#p2Nsi=Z_;}A28U93<Z>E&!6^e?fsuw_
zZZA7Ak5t+ILMQm^{W*qL`uz$0{X4-?gE<py-WLWGs+3{_e)b8KO<=}plvRHu?~x9!
zBV&)SMyCJ<7w`KV^EE1IDX!RB=Hi(jzoGS#CE5&vU*65toI;**p3$B+O&_ryl%r@<
zJ##-gF052e?)#jAxN2WcNiTeXLV@Y7@>scYD;|TDMW}~uwY5*c9pAiXQx9wDtlheQ
zh?G4Kx|+JEcUt}09cH97pB2+bS;9rFV`+Q`8CgtT4GW*8s;ATP%b$UJRt7ZbMIeFk
z4}p}zK@stptFr#zf^6bUV85Xay&JtO**Q7oTI_s65=w%k!hOq1WjM>6a=WQ9qsc!y
zxR}k4Sn$Lw3s?4qSINI@$w7D}_+W`kx(n6L9{ZE;bzIhZ$uVT<EQ~Y?CJml~ZMz(*
zVq;OM*wdYXQ>a=sTCsX8^nUG`;pt4`5#|*jPLhcu982z~Z@6X^9E_sI20#l$eSZAG
z`*M}|rWJa8icNch%(&&iy5^4aJ5YOipIAv#?4+EGG59Jw_$OvbDg6OjmU2}1IC$)4
z;x&v^wlI;2!8p2?G{Y?L_49Skt2}l^Mb%u;ppqsvq7KkIdu2qA<*i?tK1AG^q#m^y
zI<2$~ACHkdpPN2%5uIy-E-p#Yd+q#Y!|2Eh0z!r(XgZ(AR%ud@TyohMwX^HDafB=o
zS|+2`2kLOiLB|^C{&%Hn>&`voX&@8Q`)7Ac08()6cC6s+ktQfwXzh_nXjW(cTh_b-
z@O#KSJa*A!Cw^Pt_S&#>tz4r50FW5S5n=fQGm&bYc!TPH&FfY_W^xpA<B$sEOtH}0
zcgg%MxBqbcy<DQYhVe?L%;06rRUcy#4n!!1=fhAqz-O#EHSOT-Ha`+8R8<L|wDJN`
z)!F#7PhR&Z=h>imrv7f1NXhE%mS|Ns#J}+tzgzLKT=3{ndL;*ygd|Z@OQ3JMG#akw
z-KD)?6MI&NwBPdYBFS4-Y%K?ACl0_bqNkwUUE@bv<({E>WfNbkksX;W@^YizaK!=-
z*%EMsg8=mBPMq;FES6(Q7u%FHeI4wpi0ukK7l-mSH9^I{N6|)|Av5q_f7;UGEPLsq
zU7AMkY1(^duSxQC*Z00or72qOy$TQ*m^=~C&+?C}Iw@6`9q1Vw<46+N=)#jlEujj#
zYO3o=u9Uc;2Cg$1`OgGzC8BFUnFcHfriP^29%FRRa#eNHmC~oN=#P34#5a0%GW`tW
z;4X%81p3SH=@dzR+5$E=heg|xM#U(;Tri+@4@SNeDzEw7?)6zLymEvRtXOYsZfl!k
zDF@N4bgk>qf7_wV;$p9ELN8E~8b-d+OwZsSuEuFuHsy?EnMOouAlh|3%XzR=nF~6j
z&E#BLzkzSD?vy%29Vybummg@TS_5tMZZsMxJ+jMtt)Je-dVlS_n?3KD^AUV|Z$9<h
zu0)y3x2<e7%!MjKR~a)Y5c9P_H`D(vT#d^zRW-G|)Og31Ov3luRPFAhX}eVC5dq=*
zWtla;ICr&_uPKsN%~$oNZtGt!M#NA5lgCw4yt{G+2OJ1wBvF2!az5XXBNg_zAxboW
zBwhuMjE$aMxjqSOa&Lz|Af6NC+lJfq9eKzSzBkLgKi7&0_*hLA@a{Q~)5mgI%Lw>V
zU6TZgeW07J@T%&Xse{>ChlEPf)!hn~@sfILp^!Yrow2h%g>QO7jR`SKvwb+-6Hx(B
z#kQ&KW2p9dq`jb!&&}DpH91TOKraO389_k_vLyf%rfkMx%m3*&Gz&*6$ZLnZKPa{(
z(jRv86|?9A<H^Yuin^v|YC*s_GPFb0aiR^kCAo*8t-UL|rn4_P^dY#XAx_$^O7g$p
z%0ULWBFjLl_dbK*s(-Ym=`E%dN%9aw$-UA2PGws809*R`!?vx&y61R@awngerE?+x
zkF05b@Q%Bsf4+u1C%bSJ&z(%#Bgwq(dp|ycv%mh~16bq~O#bkiRbR3yZ|8QOa$uAV
zdwWSvDdv%-(<WElI=XyAf71Dg1KW)soaC=%!H<gQO?O*d?hg}Cts86|CaDy749RqG
zhA39|Tpv^OO-0{MKtiPI7tDQqGj%FeT8xUPjwTMS0P^|n{6tB_SHh-i0-=nnL=|9(
zmXGI3`VF0Vq)Y7uD;(H1|7!xW9t$4m)Krs-zVE_O^^1Sw(zkRbsMz7Td>nd5B=XQ=
z+uVpWR<2fIq_t%Nf9to@40@`BaA(J!rOC}75waF&tY7qTEJNMHr^+F2N9XTXgYVzx
zKP;y%x4GxENaIVI>FGne_fu^}g8vbX>f{96ue!0wOiJSYXCUVN2H~|H)ua&&Riku*
z8LuuZ?ex{A-YB}+?b9{{r04w;9q3xT69nofOU*f!t!YlmaTg$M#?0M5yR0m{FBYm>
zzrd8RF+1d}3@V&*U`XiG99yiO+OKQu7?RVKco~#IVruE)0&k&Y0zJDl^|wsPW#{C#
zk6X^Fbg%@(bq@R=-8MthIn>i*u80^16w1?VTf&MJtPDNcx9$VBPiJT`0O7@?If@v6
z3vC#5+X8TD>pv6I*-1NLcWCrUcHQ*7vTgHTy*nkheXg(J@x75|W`^l6ty-J=d}*8e
z<#9Lf{*Yi+vsgHDOaQ~TpOe9qB~`bmHdpz!vGkQ|7xq)=g<r?!=~ro@3+r9P_x*-S
zpu|cseAh&VExOn7z0S~lqw7WV?<_KgJcU0f_A|fP(-bv0NY2uGXlQJGe4@I>HLJDf
z7Y)jd*e6cB;no+w@`(EnkB-#Ylg`!`<UX+jyPhVcNI+CbvGdFU0AO3_B81+rFBFm0
zsU(Z$-OMtdJT}$-i8QGmO_X>9-=6cI@A1BZd!@Vk*h~5S+SC{`)grB@40r;y)(x-E
zlDik`i-P*6xIo>9MebFAFb38d+iFzQb-DctX7I+v)Np6{0s<6}hBuN?D+b{Tsj%ga
z{%{t^x`K4uRI!OGb1pwpR6D?=>g`YOoR2<zdhu^{WD6poo-NpN5Dv{DwND%Y@|7Z<
zci6sj$Aw!jawrVFmZ;8mYG0|xtYE7sp03(r?cT-}pJ<(mi@=wcDHoK(W@eW7$$`P?
z_kABR)^Xz-pwf`7A6p)wSChB}t)F$FKzPLhKN~>DByL%r3)C7y^AQXYk_UM47w%W*
zOc~T*ImeowA;(O-A_W72vA@f#GPLi%e!ZL%Y&kEX;q!iA%{I#Jy5w$JEekxNKXGzW
zXWzY9-)!R;;Vg<6pM`qcWSIMW9|1T5&rQz<p`#Mdq|bzHZ8^ZyFTG(V(Y)a}LJe8O
zD5!mwDNX(T{&w&s`-<e6E+CpDOZi+;!S*dtAjs7HI4t_pWQ*-`&?9NCb4_tZQi?G+
z>HQeqyVzvvNq3c~zG`zm`R2KX`Zk%SIbh%QG`oFOJpi<v(AYADfRDoBZoiBqy%|0U
zDX^&=9djY3W6E<uvvi*nQy(dmDAcIn77;&UBq30p2>Dgznc;SntBe=)^YQ7X71}za
z36YRZ#<05*-K&DIFbx68nGeE<j<*-xz9F*D$JDcm33ccw(0K-geSiUhH);?NfJ<c=
zouLnYj<~>ac5&xk-dYNM(A-rm_`dq{v^%oN@9<%(YR$mb!N2RF$NW#jgYNUpb67M@
zNP@1(E#1`sCryYOyxXby-iSha{gySm*jt!LCWf&&v?`U?C?hiRwJKmg!Xeh=o67R*
ziY;OJq0=;I^STlcsU9pnr|ej=oZ{wxP~kEUr>{lRs4R_4{)_9xtH+dEEJYOAHvzD0
z6U+ZESEdy6i5ps_ehI%@6bE#>7}wW7gS>7TRs#^h)FIpIqEMX25E{3h%|)`~U)8mL
zyt?-3zQ-y$SYE%aY2dcF$Rus1e4tQ$=sqFb1}V?s5H9{%sSKi7!`=HxM+k`S&nXIQ
zchpQJ`6=e>wwF<&{r^Q?SPlGl1S%3Qi-|M~a><W=bGXuhi3T}4BrVuonUgT{w1ogP
z{#lo5?s{&ekh%x~I&@b*({gd4!<Y~=&D)fZO%r5C7Fe6QMhh8mxB?03E*VWh(aNHY
zy?avZ8HS+H39nGgh?r~Vcvu-zGMHYSi(98HOu)@uLg5fI@=M7p`TpfY>;Yl4Y>GL~
zVsH6M?ee*Ulu=>AhyW;#&iQG}1qUZ3Eg_Nc^L#xkadTt!>>opwpKZt(2BFxT;}&up
z-KT)q{~I4G404o0uUT@(4C*|mE)!6<?)usMb^S2S=_s)WArSY&9WFO$<8Z|-B?Rml
z7sS@JN3G!}{7n1X{ZPJhhqqi{xuvx`lBRvAf>5s6e@hUGt{6$lXSr2n(%HSC$r4*a
zZMlQ-oq0&>`d?oy^+G$kP*D!5b)KG6OSM{gvFuDz3bmT})hk@EUj~KlTvirY?>~}$
zsjsOp)lY+WUOD&_C(pUzmRtmo(ITqy(CL3)OO#D8|MoBz$Dmd++%@*f(S&^HoU`EG
z`(!{<V#H}#9T?-f+72JN<cIo1SL{<N%1%2{v_FoC0r}DF>Eb~crv7ID|K=yO(?gBL
zZffnm`i&x^(K~2IKib+;Q|?pv8iVd&p?3ma(z=+FFMI6YsBYTQKh0eenVnt#XhL&f
z)FzuFy>jSv1SM$l@2Hykzx+kr_qU3BO>KJJ?0GBgg7Bmisp}fJL*+?1J4e}b8M!+m
z&VVTJhS6uf<L1-;u3<0CPcOnXD~fWQhVs4PEsm!AwIlTl^q_HIP8skO*eH&dmo3bT
zl)YWUv-kDAC{xA*v;eGO$f>6)=!sI@9Q0LDek@*I9o6?jct0|WJ0T_^m20o@n^8+p
z824dqOT*aV2nY*H=`-LosjsW+>6;fe2mCvJB+3PON7}1^HbxB^Dw)i137Y%m*0%q5
zQ{!t17f8voNpY)$7_?=cuDHlWVA?Ow00$axzMOxXJt6&ayYQxKvZapx+gOehjNYMR
z&U!+o)u)|tKOjQ0!Wo$>Uvs+9Z);mVmf3luvKspQUPs4;mr|7(IxoPxPZaZUGt6p`
zS%y@?zg8)UQlw8BSA1}B#rO`IB^{k<uO$k|c2cU3+Op7R7sVS-j4Uk>%#Xt-#Yy>$
z_kUA=EWeb+OjG&dPw#1--a{rs76l+tG%K}PcjH0le|td}(@LNV;QHddh1%q5QnK>P
z<GfGZB)H)NRL3^hG8Gu@c^uBO{W5{}EJ6Vjo)+NvLE+%w0Mgye*VH+HP!FhTv}(CD
zz-ZE_o^%qlZ!|bO(gt@#7C5|hM*l*YO{;_zKFg!{+E0prT@s*9s6zSIA0?hRzn9_5
z7Dq4%Vup2R{t7hnt*mOl6Ga;bD-PhoT~}->nqUYus%-=Xl5E&U&$JDGI4z8Ifpu7|
zjcygS>DTvf<xLgF4Rq}>=b<w6@tG-eT*LW~9M|Nxh@;Ji$f?;J>1^Z5f2p<~08z%c
zIFdgfQSaU~_V9q6xX3&}uA1kc%sP#YZL~OV0si&3B^aXyq;!2&env8e7l>((WO^%Q
zR{D`OawY`|r9u?`bvo@r!98mk7axz+EC(E&<c0?n==7>o5>gePwt-7j?z<t{?b9O5
zs(T8fwrhOf<6mF2gLdLg8PQ+fJyM5rTXB<d?;-d97&x6Ag?taBCXXe>@L3Bk5U$aC
z+Qhn<bms^z<HkTPtv7oEIZJkL>LH-M!dbJ^;~WrjQksM28)^-j`>YBO%4CxNx$sk=
z*5LW_3rHw(2sb<6S_IQy(%ClAqzN8wJ67A&INDr`?wFOA$B;S7EuvF8VTb!7;g6oC
zBGZ=#A7T0J#*f&yHOHLE36RYHi&KC?pxl_oZN2NIy2SQ$ET5ChujDsGOHHmk&5zW8
zeVE`b`xTzYpa3sbz7jzI#$xjZr*eWoMqV`zQpeT2%*Ol5r}*P~U$2mmkoy0V`}DK&
zSPHZw!ciR^fy3Q9Rkdx@#5(!{L2c&q(Dx~2r-STS*PM`qfN;fdXUH@pTG})iLtH>Q
zLiVl%dzL*G4O{d*W<SVsI6mk&XWV1{fG#MUDC!4=I!D4p6TT0gq=mzQ^-EOd&6X0v
zhyLoN>F}*-$DHkYrp6{Awv4!hop_^;Q*gXo(rG$ZXe{eL_uEMsb}?<C$LtR?EkPtz
zg$u{63K_4WNTol2e1r0t=EjeDxwYZT2{FdYf{F_F-)_4jDA)nwn?IE97p=dk6F(mk
zTH)mNz1q4gQ)Q0Gbt`hj8Z3pm{uUkhmIYmL*3ot4jS2ktb^R{!T^ej?`k%m^`C!9=
z(CumOy9cAEFKhm{@wuWQS$Wk<PE{-avBhlR3?Yqo%)%qQGZI>HSxPC%vmIt18&R2F
zy{rI9@fCVrVa|mnTR%l2V&b=6-HeIy?0^C=?Xk_9-s})ul;e0_A|+s9?Xql$-$iWz
zsOW155xr%2EImgQZ6<>YzyH~daA?RCL3VHS51nH`zX<!cC{8WxO-_M(vlouVK_f9>
zdJklsL!RtDuQQJt5RM~ru@l`yWep6;M|~`isqLiM1fFs+3Ps7~Q}(=ja(*TsDzrXE
z%Yb&=&$rp|QiS8|jBl7m2S~Ef^7~9dQD^6uR+~TR2<;?fLc6q!M0>tvHMAk;SM^z@
zQ8x8FcJ8c`|Jn)eBhiA~&Ui(W{2vpAfNaf!8qf22bDfjw-%#jz1(NF&1vFm1cEXfN
ze{9i&DIf^1<pPbS>%{MkQBU5<+wZ-VRAs*VLek@qe&xHiaats@J;ap-cNiPX{88hg
zDI}7A_fX${u^QNZUUG8waACT+THWSl_+D|!)2Q9+t4q`HR|Xcp(=ar9{-r;GhP@vg
zQXN*vrT_i+)nMw)S$%0cY3bl^Y4z3M6Y-2)2t!(#-U#2`N7`4GI|@ISs0>hN19+xF
zB82p$jn)|KboV7r$+EMzvX<vRnXkgDg$aTyXDn(u<l0u&i_oaI)+c!tt?e7xwb*m8
z@2EP?#7u4fJ*%x2fS0}*A+zUGvrpO)8za8?KV9DZz>qHD#!Zv|OjKc`<WNC85r@-o
z=oTSy)A_sC=aRIo*TvUM&sqO{Y_*F90b9y{F=4L@F@H#n_U1+N6$D?mBgJgLOSB_Q
zh}eEc5`6eCMR&TomDv3_(epaTb$6&C<ly!y&`Zo&wuA?H;+ksv`Xz*aRD>F*#pj5;
zl|v-h0c#C6OWAh0*&XmHx%`gV`k9!soBs9&e)Pyig8pz1yFM2Db$~DHO$n>!NHQWV
zdiCq8o`l6ttz1i8mdCT>8cbDS$Qxcw$A@W3oMy0TH`+ZcvA(BSu=y2>YPwqt^tdN-
z-c;l6C`cQ0Y3T)S?N;LI?3s9a0g<@HcY*rmr6*#%<vcNNgUb~J=97YN^&dWXfI{%8
z%30RRg6$wbXx#b^A#5*krSi4@QZoi?$OCW4n7o0+@g0IyNwR#&9GvfYHw|k6&MplR
zBQ=h3W~O(_>vl1BPk6{4&fOkMzZK6k)ld6_YWZQ}`jBBp^L$Rc+4I3jIeQ`cm(Toi
z(QOZQC!FO%O@14<Y91H6&K-|=q$1Qwe$S7NpsVS6bfY*az|R3&_fzhitDLQ##YTm-
zUyO+z9-r_Tj>K*iR2%+XHu#f9w2;!a_Jm`Q_9s8^8^1&G174tqiz2N14NBcYLQqSW
zV2;t#W&NdUVAV8##!Gu9X84&@0pqZnvNdt+@8*UW7r`y19O|{8D{XQpb6}wBmf~=A
zpl@@~G3D??#Ksv<IhXO~$rB<pbL}4b92w|)OfYU~)OkKEgWf}1%qoz&U&oH)o?@pI
zzugzPQ);_lH{IfW&zzL-d%zrEXl?bP0OIKBeLk{E{Six?j;rXZwYD(d9sAs?3%z~K
zA9eq9bN{T=(pH1qlei1t`4-+IUGZUj9(g4Vy%es{p}RX~{rs>`LCAY9_1)KNkH^Sc
zw9x0yLU|T!oD*)!<r)t(0-JbufkEHN-H`dPcW-vb31^<)pB1HuhWGPj^d}lFSIAX`
zp2h!2lA~4qY58+A0YhCyPDDnLE5B~i`%J~U-dyn$YDOxLd6Pm0seGX_Ul5(Uu8zX7
z%SHNy%2U5xCSQ)N|BjsN<j+KfDG}0f6B|#<cz`WDdiL~LE#y!W8%?gN#@P)d{C)a0
z(9Tw^i6;78-`1PA)TW1$f3hL`(wMyJ*T+A^g0z;tzFjbhJNt7j&C1*DpR%o=8yemm
z1A-QNMHd^4@q{M037EO#inD>VBjzw)mK^P|JEeKYiht2eC6-gkWeo~<64^L@6w8%B
zGKQX4yPiXy;m?CDo}_+<wXm4f+<m8{|8_6P!uqs1feL%_o5PpOkG<!Aqm&QWi{7!<
zbqH7JfQU4oOI2<<BYxmeu$Lg_R!ghq_JTM4F~?!z8ywn40UywLjS04~bo##>D}Ijt
z<-)9EF%>Z~a&~~P-%({?Dic;inQKz8{Y2dv&^jhm|F+r&A|~WIb)<Tu4%(`fqb^@M
zJReZ1aXX%50zbmX3AlE1MS`QkAA=)^yzayej&NojTvl2~SC4rV(Lrl6ia)O`3OTyT
zl_<H$ux&v1mzEsgJ%iSm5**Ok^i9_`-CX5e6kQmV#&BJD@5kuQr>z|d9zd#pqJED_
zudPQ9zUu8&7zPbDdm~WmcOn9pkS##~M+GkNNRvHoneV^*wwTpI)HDV0dJM#@!1pSp
z>tf$KG+?5A%fH#=GmmCM1v72Bms>=iyu!h>{#eL^y^C6Bbe?vsmOt;pK@nSN7Z(<e
z>8;?MRLn5hz9Oe7SVhIwO`@Og^|<&wnJK^$y?~N-j%v~bezOp>22q|JZe_|Vg@35m
z`)*|)cQ6>2b5%&rsb*wcTIKfY8dfbr_klIA_=1>anQ+uicoz<UVS<k+Wxz!`SY*R*
zhx;jYTS^sVK{@LDRB2bN)m=igByB%VsJaJa`Ya22d<5}D2lUW}{!L~uVGJ+i+RSfP
zz^EC#IKnrmPoQU*Xktw*p(kLuXtl{7GxpQ}(CPdS@_a?^A2aogSBj_pozIu3{XMvh
zDOUWD$H*29V{lpS=p9}1Ws@)?dEUBNFkSv$|5iifm1M-hE#n!%e{@7iTEvcZ6E=@$
zENpUoK7h=&v5Nh#zdZ)kPq;T5zFhQS5CZAP7)+$g{3n1aU$jX!ji~((W!+?GE&?2o
ze)2HkXeHBQ36bcAIttm>T>9<C4yWC|TtjA$WymY+*jOF)dVf+!=KZ9%qS10xWf!Px
z?Nh4!phfsiKcAO;%<m3P>Xs*}o;GDSEwb3D@ReBajkQklC~0XPYJ{o)|5uj56(M}&
zukM|4JYAD6)EA8-wK{pU+}PInshk#Ld7{JFi2`W~BF&+YrMcBzyJt>+177l=xIv_y
zAo(E;<>{4c2v*7KD_H;~F%^U0sK3o+#m-V?Au8Kgx9b8P%z}h)Rda^K79yW4_~l;H
z4@O&L|4C{9^-+02(P^bPg)m#LFqtEK_zP`bUH2DNi7sMo0fl1L45co-SKwQV*ap-J
zrD9sChF6Be3>|;5{1#G|wO;uCS}PQz;w!-RKR<)uv1jUbTo`joTniUGZQHV5V9#L*
z&o4>0;6T&Ci8ff(SpSzwO>DekV(K`dFRN}&LX9I-Ih>fl2g1yPw$E6y#+jRc$>Bwr
zy+1f8C!-$IH|T9xNO-!=DGfw^vZY}Pdh%?pL(BD(x<H`vs$L)RtIV&q414jI+lCgz
zBMP<wWcN+y*D|JphixqWZ|`Nd-j>aBq^9-L%hKBmEnNmYMiVq@7-E*Fvxwfsb=tAU
zqeWJP`Uc7W(3VN|@gVEC2D`ZCRg>nw9<TP3k_i*7&|=G*N|bZjd)+b3I9sUdhW(x2
zYU*|V*7|E|nXg(=^DZ+<mEPZ$*Qn3WRH>-iuvJZIsSf#CJqzG=QqNBSsx~C8cjbjx
zQ<FbX6aDZOqol&QxnTh-Kxa0@0AuHz>HUkTeM*@3X?pK$h03>4Q5lol<udz$(VV7=
znSVk1Se>%#O?n(r{5da_)a$ZFev{W(OJ;C<72s}RHVYH?qqqtuxT-Z}G8T^y_vB4P
zBfoc(f}z8Mz?QDAIUA=j{0eMJz~&8HBNxp1zttIWPOwb5&OS6xMpXwK-Jzt%#yRW8
zveE>t6u>C8!ihUS$YB_sJ^Uu9>QmvrUkj${Y?HmY)Q&JL;lQGV$-Mn9R+r?G6lJ!Z
zNl{;;+FfgAvPm%|(bg|1{J?*Yw@jX|n6|5xQ?g#Rk#D;s=N+X$j@$0b_(x;ZGw}`~
zYJe}-9qF0be7ozNR&%or%ZJqnkx}E`9wIvBi#BpX&so|ZuOad&L@;mkFP}Bbo6x17
zUm5)4rdInu?__31pyQ#xCb^7cO>_F-Mf`qM9WCD%zl`K^B`d~S83o#b`TLr6;iz*<
z)9-e_3flc%+P^~)psQ?*5@dqGP*7pfl+?#4U-rGa^Td#E6Pc~_$A%Q2H;Qa()A}JS
zU7;FNi06jnBx$k+Jd2O|<dj!ij7EyF6j86rn}qMPA$7<+8uN^Wg6z+HsFb9fzD5BS
zUrt(y+@**<^8dqE^CfA2DA8@XbnF}>qzOzo$_YF?qM3A>)%OI_oiC^zZO7}fdic~N
zdYwdVZ11U~(ZUJ;R)!~|uKDWg>r1!91tNX2R4voerVr@M_bmS~fwDmq*s@-W#$$QV
z5wFI=wMz49b5QoA{pduCHBSNl`1ffjSjI#LUPvQpex|L^_$asqZJU<3D){uviB`oY
zC7I&y>-e)62>1eHw)I`#6n=CYGP!a6>o94H-v;(9$uiWJVoqH+sW+&_3`F?}7ll%8
z(ow*E)?tK%N9fnu=F=GG(*6C&ZoGup&V9Ya)Uz^2Ce{#dFP4-;1J1C2!fd+C6pDCe
z8jcH86>%oljh<xcPnsic7Ar|hONiI(1|P2w<Ke`ye>-L8vv7#pQYW;%d*FHjP%e6}
z$MEFtmfHg}uNuH(R%0H#mXFK&e|XwY;cBUzq`)*G|1+Z7Q@a7D_;bUIx1zBJ@)&1W
zxK}+cfT0IzToq<O#FIUerHTM9Dw9XQc*zrEiqToKmA_NINdo>mVxwvGtyn9McN8DB
z3RPV%$l@H(Wm?_iB&t23%9KR%&S;hWMzzMaPB~q+Enx5h?QXq*$%*khf`L<zUkV)u
zEwI#HwcZ5rfgJ};PB}`B)!dE9sk)|q(bQTr=VE{Yz|9>SUqeudnfJTNZ{EteU@xLq
zX`2a*6`LWKBtFg<M9}yX<&yiGtlwkGp(kUO5AKga^t?0WUuRcbW3n*~xerEZE}Nji
zuDHwSR6K*``muuAWn-3ePL+o(Yt~rs=-_Agau;nuw)|F(TSOEGhDP@jVtxV;H4xIh
zJs^C+YKXMs#+pQ7ES{@l(9pF*DucK=UkUG3F1SAOq>V^l+n5?h&dK$pwxa`%0AJ|J
zZxW1$2Z);q=Ib1*2s6f~(JR5Re)~((MrVm}ZHPu~R*Y*j&bj}fy(4hv6~6iqqRvZ6
z&6k%X3|f@?Z)!jH=)O&vnXf968~c-DzFoundLWcZc}5D{jq=u#JwZ+1tnAKHucUxd
z{d$Al`}?F42Gc2D?-<ggtYrb_zm5N!IHT7gyo6C5T;8Cpu3=u)PfS9Lnl|tmQCMo_
z4k~V3g<XMi|Jw58Ou3<ec08Y-(&WVZSy1|^?0<+~yEoQA;ZJUv;Z{Qy;G(IeF3*>%
z0Y~p=n;LIs)^5DP*FCn}vpbtDEPne#ghp*of{S|Oi6j@d36J4G-Zw)j9zQvL2kED6
zCgkyorMGwblOmNfe_&vAgiH9n?@4{u`|JeNj{=|Qi}?xU^p|(=7v6QMo$i%yENiAE
zMRV`Me!Mo6OgveJj{5Oh35)Iyr322e3E1tSZD&`*!Q9{0^_@)9T?U$4bg{Pk!;@oh
zU%IlbW`Y3AwJe5yK2Nr)n*0`Ser#|vE%n1);q^Kxp|+!5!JmPMG~tW1w$MA?#4ApR
z1AaXPW?pVoSeTDs89qY3sEZ7Q!F$T)6^jk(vl<KHF7B9>ZsDBOCRS52Oha!Dbsh4|
ziy7w+4)JAQXujidF%3-IFw3C_J}+U<M~F2(WiIrFoU{hiqZ@J|<o{4%3-@gHIAeH#
z7d+rc%Za6t1F_&+X8^6+nR$yxQDN~23_Ov(;fO%r6Lr^!s_F*Tp_HQ|V`}KMHR1_a
z?ITX0!db|tn>17qox7|rTgP)vO#TBXU<=cLy$^o59{$fkpJcf;I2`UZXu*TXM&y^z
ziNFxxdw&3{mo+xfel|c6v`I*VxDr~m;mJuUpmnjSUl`xsSJ)tkd;@eWSmXYg{fR>3
zUg)a+kn5->o|Kw#81yWltdW9gXBf+(ME$&HZE`b7I8!Y$uF5U{SKkm(*Fe;U!S%c`
zw{W+|hu!E)20^l3Fq?m}tnVCpg0|_+nR`2?uH99yZ5poK3MO4J6GZP=sN%^u<%xtR
z$MsvpX}n~5tzJE_P;VPk{Cg5bgxb1DpMyZQ1plP-S<rUexv2fKutG~P0%vqlw9)G=
z&1y-dT%X)q{m%Fr$~1vb+h&aY9^L2bq8GO@)t%S4e<xyt3upK#yj7i@C!MC7eICUJ
zE->mB62y=$wV`hElwAABJEN9!vQY#1buAR@fi&zWG)l@h!XR?q{4p}Maz^_*Jmn|X
zKGV9N7SCn{^zhRX(P|wCd@qZC^v6t3G#dT~1_$mq`mvzvtydbg$n%`E7Umbg-oxR4
zmkPPqlN$@4AZkr%wHq(%^!|pR#6>(&*F1I$`R!&yPXASMeH|gdQHcxuM?eAP^sO<L
zzN4-zS3N6uJnHP7>pb&zF$q1{5Jf)_XdKIZ{p#ZO)wph{J(HD0;5COip4LIv)t)!j
zY6ybLWx094Av903d`=RBWH{on7v*?msZ8D+yl5?*Bs{NXk8JZ(q)QyId8>gD*PpwS
zh}6gNoqtFS{d`TqQmodyz-KEAG)#<bfHu>T@~-ChdN$va)m<~Uk12gszpC7KCQF>E
zycQN()ar~<N|kWXMDFEKaB$HQ$~_0B*zCVO8vWJct7KE<wAtJi_l04(JfHc-e8c~@
z+3!HErL}H3uKjVJEHs~v9FsctEKJRupZ*drp6R0ZMZ*<%2$>uv*RC8>s<byV-x3Oe
zOoKWWAvY|1{S`2YF(?uU5#IwhWm=+!qqIW3&tI+HO&I2idS;m3kFKp}Bfv?sIBKA4
zo{sy?rsiM-`J&#pQNgMGEa)Ru6bvTMrl-HKzGpXRC%k>vfCQ$)RaF9Pilc7)vVVe6
z<v0~6`6$#_4KU56F&W$wWvGlw^peo2@3F5(2-AjS@(10h<;DN-9MT-t>5au^D*Xaq
zGa|+%MxDUB*GGm{`a^9e9lf68;=|QTzHce1>gwC%@)6pOB<NoX>$r1A$==9a21282
zZqayTB@}{wlydfNzIcIY`TERms5;YR1Ae8g6_{Q(f@d9jE<n-CmLCh3`6!rUKJ}`6
zqOfsMhF#&UyjMyJ^_y57-P02C)W;V`E>vvqxq>eFMU^UYn4TBdG9~7ru)ABi>@|TA
z5{4A(5jVFOL)YTvn8zdWr_Sf2=lGN0Eeg=}NWGJB^XESz<${fN-uIyUWjS?4cf=Ug
zH4<4d=XiMRT;JL{7U&9%BI4=D*<#llBW=GLSPI&-EU-kH^*Q~HR!(Xrb159A)c^c*
zj$3CPrg=X1OlG+2cD7|kfPEZ0`<*s_ye0Tq5G_3T?g(6PyG^=Rom)J2aMr79YJ)g7
zPT7f23p#w85zlnr)|OVrr0RK%gM^<aXgJn!*w)6hcwY!YRcUra$S`CLob0K=54b&;
zq3-59J;?^~(Ope((>-Rk1%WjkW9h0C+P-Y@ugu?s>DCES7N^n8a4vX7je0(~<osCc
zR!sg%Pc<$NVFF^Gv|PD6I^{a2`Y$ZNk7rXcb?fdd-w*Ylyg+6LlY}EK?W%P@44hzc
z_dk9l9n7^nu6jSUi9HKs(m;o1Z1cZ&c1SXKeXYy^cBkPjwV()%OryYO^7wx2HP05`
zXPj!CoV+%nFp;8@E!OU*)fdQP7f+iD_F<VJzh@k0V$Y#Bl|O$I#}uNXx9`8WOJ9S5
zN@{Z;I?f<a73c3(;DYDjdO{r?jsldC;}<lKM3VF0n<jUnC4#b#+L}HU1FHtyp9nai
zM!u(&SvTr=a0?gyy;Xa>q{LEP$G-tz^kaO?Rs_?PNwuU$+a#@DmV;mPxLs~Sl3Tu4
z>4Wd!m7uG$%5}1;Um!Xab@~@O_cx=cklk+{{0OnNfUi&eQ_G#9sZKmB?>tapD*4Ul
zWbynPc{ClW(1p$OU4MsPO4Clf<Vm+fy3gbkh6tc(NYQ=?l_*m`m~cVP&x2&NDtYo$
zMMOcRCXnB=lLs{(C)eKp1c2E=DUQ%ui)^6*k>{wzpW%rWic5xmbMThb3hYW+Ia~KU
zuI_`KN=A7^_)->yL|^dch1F2EMQkyun#2|MNyinreu^g9Fs~{Zb<$fZ2pJ9<HYABB
z(%km`v+#S?s*QuSkgvrY3h|pP`7ZavIGWp8TQA14MNQT>&|bJ3EhOu!(~+ckxPfTo
z*f5KP?-$g<xUo$G2-WT42G8N+AUGjS@GFMO7v!(AimOBgB|p!__<59k#FY(t7F#O`
z^~uZ2%Y<htO@g-U{%*r7)p+w&{VsNN?R^7`!~-LBhzJCh5|ZQ|c3mF@-#G$SonwIx
z<cjN*OZB9K&%&~(G~O%a?}Wf5=bw5^ZN}3OIr7Aa1@q1Kgxc)ktESZw{7d=r*Y#Ue
zNhfO-CA)6Pqr$qlZb3Af+vsvSI>syqw4&{R2va(@vOrcTvLEoPR(hXaA8gyJ7QN$K
zJ^f*TRx-hpy2T^<eqm{q39k4@;9;X6D5q)=9(PV!hr=HBg{~pJw*XT7=QmkG^VNAJ
zsAE#PL6<F#$ejO|>pYpN8+?#NvccAt&81SIdhIn+auGevr~;aFPe?(bj%|o=@j~J!
z_T4EOvBdYhoO?{q`%&AD(PB<U?L&vxOMUTda0~d^`hQKcAN&P;j#sxrZ(N@!I_?{4
zJg>5C=Y>5{bsZW78~3WK&Mx`J=_9s*C@o4VD)kz<!^7qBHPfBDKv(*{lrGnG)-09p
z?e}B%LH};w<0O4ohWNN&;3K|h57Eo%g?TY>ybR?jzL}uWu?56uU8O9xK^U!OB0281
z2+nA^z^d!nn`suXTxGp#&DvIUP5a1op4j$&?*BKKC>R53bo92&Ow@|RGRyb{A*FtI
zP~-aS=+Ykps<Z>M(tdP>nekirrgT6jrrQT4B{~i%ES$AY*gmpmBeX7(8=oYaI)hl9
zfQacEk^dxn0xpExU(wT(Sm<(s58~AL;Xm`XM8y4eE29Iyn~_ZRYf2hw4-=+`aER0S
zy$~t~g?K5t@Ha|zBF`ZZgUN2yM>@payF=qZh}hg}_mwocBSlr-Vvpwr@FZS-2?thx
ze%pu28S}EWk|megmPK%OlmI2rI08zV`j&vNP8xMthn`-D_xVn}m&Yla%WN-U{a<N^
z34ZVSb?>Q#LZS?NT}(dYUeZ-@W~;!kAvo@ly`&Ns{(`yz+sxK}e_(kzC$*j@)SXZ1
zH6FL;)_{Y@y70W#lL|6bKJIz3`z8tSPR27S>M*RzsDEalu{ox>40$Z<?c=I?3?MMT
zDkVsp{Dc|H`g*Ab3Y%ph{R9!hTa?4h?BLW_M`eNYs4{^$&vW}%Uo;N>sJU_rF3@Q0
z>*aZQT7q6gv}oWkiW;(xKb~3IYnkte$yx1tyW)$+5m>&Eqe?c4%D9ujk;IlFK6`*~
zdshj_p0rm(w}_X2tJ-?TH&m<qV}KynD!>R}LXi2(hzVU<M<;-vLg099K{UH+(rK-w
zcgYnBYpJdr!Ldz1Ya8)hLj<YuE*aLfb|jtnH*0)Q-|$V@0B^ZI;pBDRAz7EiOBE_R
z(mCH!)9?V81j7>`Hzp{<g@Im`5cPg(x%r7C9-h{DRq&*;n1?f_&|f<JjW$NubC>=E
z_@Syj0x>_%U-(GvrVTkr`BZJX{tnPv?-U%|0vqucgZl35@lXJYO4e~Bo*(!Gf`|u`
zR9F}h(nm^XF#SNY@GvBZp!-?T^3(Rxk$Nqk#?$A3vIgp(cW2gB+&1LKU8-K`{&e^{
zBR)+i^pu=E?Q-$m!=mXUrkcbfB}t}<XSzxK=wn`>@UL-}{Lp-E>=sXipo|GH&h<~R
z>ONi#UJZD(u9{`EA80se$}amhs%gu6r@i>!f9EF(A0@RE(2(<O&z$-u+(Iu2Mbcyx
zD;VJ;n{6)EGIl@Nz}jiQLutmw3nwc71foWv!GfOqU2C%v;0@thg7ytU`bW9481VI*
z<T;bT#P;9B?F$Cqhi}S5Oe7A-5@A6XG*7-k`zKGv4^xPJZyk9xBhJz)c0wuiAg)3a
zS8hsN3|(E3i+2TO&{<pL!P;gmOvi5<u8Cgz(H>W>Wm_g_bITsz?u!9%X6H{6jx~xu
zSq1c^X4afnXRejHd-b)i?CkAj$r8Cc9AFo7x8MI;vP++;YPbaeqk~%zi7+>*)@lJ&
z=iJTRYDc?weowau`v!x<!$WTBH!WWK>UA8u>t>lkCV{c^YVf+;oTlU2L2O6Ea`-?X
zrE6y|eWJ*KH8Y(8>CI$aE*a%QHEM2EwO-Ni>t%mZ9WJ`Na(vclIPJQlL{RL;1X!j5
z@26}%6mIkUGVH*Ue;&~=XUf18;-o9A&k<TQ4yuhRo4Hx+PJj9SjV$J}M&13Lw`isD
zSn$3l>&%DsDIf7~itA;s7q#SJ8Wb7#(P_g>qoc&_1~K}jKm${hs{+~jN?kq$@v)jY
z31K#r(*5*8cR6cVt?FVlk4>QjRy$`A8YEwrRdSJbYmpg3=GHo8%`)D6^<*fVBsYIh
zAF|7zG}->_BT(q;=GHYZK=L@_J*%Uu2PiTRD;fx|b8I|<Mh<GA-3BMvaAAgQHIe39
zYTcY~D~=3;oOw(h!9*nIFZV^W`m^9d&u>>*G;-gKe^$YlHA|Ew6@dpO{YimHTu@f#
zk;oD;)?>(_AzK3P$qSxG5Po+IYG<=KCmvh7Dn7o*AI#@mp2%Zk99(49=q@^)K1U{_
zyVU#-amrG{bC#3VV0o3is^$;p5x1_yw;;r-@0!`p$lL8%wlr?1tX!A!s0na>x*kg(
zE_s@QRK^a6Et`6M*d@wxF(Vb^(@=#UsP$!it2)(=Ll2`Z5BPfr<(Y4R5uyzV3pyf+
zpdf7y3>vfom;Q~i&u|NBhusycvN&jnGL^u;1U8*q$j+Be*-st$*zuDq%P|z&$U{j8
zSa=jsM{GPvE?&Fm*0qLE*Gwxnt?jI?s(c%wqakxrD!%wbU}u=fs^q1gJo23b0%0||
zBdrR)I?-gub>!$tvpV|%Kl`jE{w%_oCjxm-;N$D#j~u1p6<yqQdj>ZKQOn71!;|42
zEoQYw^rw29gaF)r0O<bv?e5#&E|-Jf6n~ipH;>2S>O}?resho2b{n6n&VNuyL_4bB
zD}|4nu^f6PkZ)iZJ%-uw*C{qdwkhoWn<V09ml#q()~$%33jHyx9&RaAHdTEB8X5*&
z9eXbWVj@j{)(Sh}7sHH9#v^)HHV?Wq#ygN2nlp=if*DQ^gk5I8hmL*4B-7Ovl}@|p
zs6i?R=RD1Lv;U)@`gAeiYTWnvF-M`yV(BoN%!)0)_4do>1+M{%T1F+$Mpe^+QQjzD
zQ^ZrZxL|McCGEhk=so&g_OL(Xag^cKU{U7$HCgDj+#m4clAUVbgKT}S${-0=$klM?
zm`!LRud&;~-TPnAd-Bw792ks{a)VacI`0*C&ZvAUWmT`m(-aK=ZW%i{J#>9mS=NcU
zWOw?^qFqKzObi>tmEKKdTi_iy@e#g=RCE9NT6d1}6**JzBrx^CJ(Yq4Q8Pq-D1RGv
zu-TY1vdh=QxZ5&Fhxu=tlxqXCf8)^^bqj@Aa&X;^RKfxOhEC>1&Eq-%vn-kW9{>kI
z_`dl1@BJQ|r%%y4y+JxpKuR1@<IatD(b_N?4lCW-bbxXo&r_6iIXs#%pDnSaqVF_O
z`#bFK-e$0fxzt;uJ{!>p1M)0}z()y5GGE}=qly_>f|Qmpos!;v8y<NUQwp#J{Z5NB
zTc`Np-X6<LQEx@GT}7GYlvqju%fgTqhF_AR++P^l81BCGeO@U_p8i{Z2Y0Q93LHvZ
z5(@9c`4^PqRF|vaG3jDSz13p#Q=gz0YCh5`5fn%nP$Vf<lsJumVv!&b6iG(td6;G$
zcex;##XKDtR_b+*&tJfEJw`iw9NfFk#P=8+?hw@4cwUH9F7LeYF7Lno7T@~)f5oY@
zXX%|@N2etQ@PdH*x8DPyn2ajdmAB$4jsw<e9LHsFIAJzhpmW1&(8W1CV7Pai{qdAD
zE9=x|Q)<3TS!5XF<H(R?zM$4>kYyQ40~)++Mt1uRM7xR63TsL_tvct=p5tKWK9fa8
zQ1j_{E``<<fL4OI)TD*s-_~LOd56Vfg#jy1?AA0@{m!9S=G_1O+iaXWN8|EktTe3Z
zoRxM6+QG95>y&sw13#?>n-<fQTCa@^BAh~#&!=Rn@=L93&MaQ7j?N1jy%k1B`$V%L
z|H<V|&bK7L|K0$9{S5cVV-6=r<k~RFQo`{eezKqyHn6_We6rw2+k3p}zel^>qTB1z
zY%~}TjtLvrXf$dRWrmU-v-zC)d``Sf==M+JS;2DuCii!*Ba4zkyOf?&o$*qDDUrIw
zn1VcxneXo+%<r<bd4@7*RPV6Sltsqq@PMZudy2;|KZ?*9gTXFuzy1S;hua(-Zj&XC
z;7X^`r3%4fv0yfv6V+NINy?qOcX;=`Yn(lMo)>=mc^t<f%d#J1T29!rKS_M3q|U+M
zOMyjJ);Y#jZHApOx4-ogmwx?A^mM|?WQ;Ega?iuD6(>&MB7KGLStK51QX)N%+)=om
zPdpmq^;%d}xiI*?kCD*owuvVLN-glyl+T@Q@kqbHKYsN#c`cydd4$2_kkMkot)yT(
z8PlClSo1yVUPM2vSH15*F*(Q>?C+5k1=<L-R46n;xj2rGkD#s%-t3Tgc)-r!fJ8ZL
zDwohssfP`8nN_3TNrLMriaf^}O;#k-26s@$_t2{sfyzHJ3<IuQzDTdr;^wXEEYl^E
z=@ckAckT@5&YeLB&9(R6BTI7-iaYnVdHLlZvblK{-wWt;x_Is``}=#`+rCGm(PR<F
zjE2K%gOTFL0@@E)o8&(s*XCy(g33SCL*q#(t*meZwi4rxuI;k@(yN^Q{0sEv$MolO
zT&2;jcQXENaD<Nv6d?j(J<7C1cpkV;mDr6Zs8$W(NU}7>_XBbRy_FujvoS>uQBm;M
z9_jP=YQg{KmD}h}m$mL1N0WUf@tpS;Id{Z}{$j$K?-P0<L0CshfnG|6am>Ck<VAs|
zn*3B4gz|7558pzgEOE!XOb2^xkB$*R4L`RuoRZK3U1pS-pv<yLBY3i2$|6SYypK0O
zBB`%psq|INMxCpVK7!{e-oO4Hak}JqG{RBv=%bIIl*4R3=HPIz(o!OL=iT=?clveO
zogTGX{il9@<CkCOjlAIOuYQ4AEg&s&T*n7lRfDCHIAMqqkR)^JMMx)<Y(BC^ST3t7
zaZ#DSC!-0qPK&?~$dd%u^~mxJPdaqFJ#O5+N8}hjzp2Q>lJ{f9#@S1V;SNW`eX=;g
ziX&7MF<xYZ%4MxlXPG6uv$u`@u4R#Hbl?-#+cd&Djw2uX4Hvo~FAZgpSN_w=C$NSf
z9%IIb+&La{J;~|TLV{#Uzi#O_AWdRSvt3R5SrDeGi0ZrqH-LPG+I<JTQpeQKA&ug~
z`7=EE^e6e<Z@j{6xrA}dd)s^Hhm)BPczb`zlMX*+7+{oZrmwxr8_EBh^)J3a^Xw@W
zWl3o(V0hJv7x=Y0e&7=?X1KnhUu$vckqv6eF;Y;i1LcrS7r13fPzxz+P83BfW;1GG
zNUPOmG959yv&ZvKxX2ZU?;U0IH_qb3Eq0IYG8+z<jgAqXOKKg;Otadk5oRgZj`q<%
z6vVk9R|>b@q!~7GU4`dXVCK9mD2*Y@3(zG}xdhUsmMxIuT?WU;yft59JddD=@kGXY
zuTDH0)2h`l7(D4=^AhDKbg2=gAWvolhu2AWB251Z!Zz7jTjfh%{wm-2{#)D|ZNoU>
z`u<_%e(=Bx`j_nS8>(uZk`_-?7LEdgw3gBul3~oXKll!h{B@6Rze8s-B-A;D=b(f-
z8SHaVehpUzDCuEhgHR6EbMb<JWIDxbH!yDX^{$cNNTOz)JdQ}_3H~DCxz&gZtuuV>
z`XOIWps{(LWwvB69dJJ_*h{Ch<}+4(ho&3Sj~bX#qeaPF7fg>1Nwr27no>H|&mkq!
z@$scYs3rb#f*u`lG}z@R$++BGfyEGC)zM{&E*0rwf$upLI>%^oQxXmDVQ;>IbM@2Y
zQevzo3Ii@(y2RGTI@|Z}u}Bxp7IUo4*<4>^Wu;52S>wiyYm{Y)<9O^I9P*7n`6iED
z{Rm39tgUXay}ire=#cksT;tT%7Fk{}nM|0^=aqMCUi>?sBYfDTP8xivSQx76f)H3?
zkQxr(zDaLuh4ZU5Ofe^SfU=}Kv2&5CdXPn&V&_Y2;80s>k~<F4(I`_;CJFd1!jz<O
zfnRG<8cVacMm(C}oqUL=>hL$7-ax%~pKolBY1DeGwK|NKGsg27gCgg!NbyBVG@D`^
zN&UcQb7h^=mo70HC%DqV_g!4&lINCWI>62jNtR3QEfel%8JTi<x)o5L$E>e5hyoX*
z3mm6Tk*1h1pePDVS>hT)!|^I3`_)SoY1}Fr^){VWi?`o=r2?iZr`kZ?eTUh6hU>Xp
zee`i!tu9uVC}|0N2g;myKEih;k6gUW*;8k#5$B?0JRUQe&iLUEf5^?7w;!;GA9~<D
z^g;Xug|Hwkxboy+FYLqb(F%jiH1~e+7OSg08eT?8j8p<qRO>}NKvif%7H8P`5^65B
zZiDG^f#-RcJjWC{76+ZB<XGGwq9{G;tsdkBjsSz^k*?xzKC{N}yt~a01|_{#pUrlk
z<K>j;e9nGeu$RXKA|X^$%y6HU7jSCxG;3$iFdQrioDkplQNoaChIqJ#SR651q#VRC
z+j)WGc|09BH0BF7H#<mCUFTg#kuGB#*R5Vmt*iASB*pO|G}bB+GKzk0jazr#<IOjI
zfOLG2lI@)x-oJj0WfBub5l=qx6c$-g`h_J7T#9r_x>STG$;?j^MV7Gl>MJbL8JnN_
zIA(o=)H?WqgSC}V-gP}3&m+xJ#9~gjTv8JSrKqMe9qE9wh^1tCaE#w?<9Zc6eY7)X
zC2At9;OhA<MH-`Y%5U^E&-7egofO!thp--l*@SsACr&3Mg&`FZH(%l|#?(|mEAZ%H
z$)%xON<^6>f{>tALxn#1Qd4(bJm14+G0W+c!QnQ^e8m0OG7}{$j^Gk-E;2lO+QToW
z1dTq9qY$#1*hgzZ*W<WY5^9CaEMhh#e(y&(AL(Lj3u^>-_xG5kDVe3p;LL-K_=Abu
zFWHIy=fVPWtE#=``ZbP*`&{_qXK*fGAo1%&Vertd#`gpKT12{x$(At<71OW@LPBmW
zu9Bb>GVsVoQ_4IiY&8%B*knn(hzXR-YB!+nts#q?=lzn$FND0Y$a&31i2iwwCS%6S
z8Sy+OK4Er=GR9ksse1u+H=yrGGUZT8fh`I|R6|)!tyw2Y3#4&Set=V^q|21y!8XIu
z9*4`EBMq(+Tvdv`&Up5GL`zNquSy7%#8DDTO)UsH(mA$F5pj+i<c!~V1^dwmTf2-x
zvPcug%O%;vD*Oj8*uQ*9y#-QZ5X#~T4R~l<A%r+_X_?<W;P$t^%c&QhK`thwu0`mw
zQhQm8RH~8y<pud7#wvl=tK&yC(lo<!Azv<$^~yFfo=yncZ5-F7(ODs$ZR1Ia)rM|h
z`Hja`@co?MxwcPiBQ9*5BhO+6^C{CLX0$9&BBwT+BOQtBx%BI8E?l}y;3<kS!}mM_
z&qHfVx?GY^4lv6pyUUcFB;m-wg=R!OOK7=@PPavxE@^}jXiXL;;8fiT*BU|u=3pD=
zQU+eFLP_~GE}p;0d+)qaS%*3<S)TLuJMXZ6bbvO7t*tE{xqKA`_^zVR1-2|$&W06f
zVSR)Cmw$ykFY#TE>14(tO}TmFCNI7G1M;Hyi|x<;g7}c_*D6(b;J2nN)o8B-H#V5-
z*XcAb<GPkS%}}MlNbqYRLaAz#)P`gkSG*x9af5(lvA|Ik{-MHQXfU9Q0%0tE7*O~=
zxz;R38Kc1zjo?={+SIXp=iUKFTH?9^7uMF1p35X&Fq+Mn#VMjJakXJO8ZdurL6MjA
z5wxTv_Z0Wzn7w&QqBWVeXeDSV$&;Q#CtmX8y5h-oO&O1I+NTjh;3*eVXo|8VTgJFr
z;yD64I>Z(;Aix^KY&_zXm%h)P8}Hz_5pLK*xe>N-NS6r*d;8S98XKD>X}kney-0bH
zaWL2=&kFLqATLUiG-J73FqzIco=%v}=9K0C8o%Zjym3@xos`C6P?VU8OfDp9nj&vr
zr*mZ;Ttl7|I9gzoz^_LLN8vaUN<*3?cuoLP;P@VSk|LCXtU#9<qjhB=rmFQ*Z#I}6
zkI=?4JeV;ZE%98%Z=LPa;+U8B_Lv}v{D{@9Qy3)}FBVK@GiGUqEH!~P3=j7>o(!>h
z$qFDGMd3>B%oZHQIjPp<28@EfBe~!x+UcC<&bVCaB<M6pb~-4faD=Qji9DlNrnorx
z%90=NA<kUFYK7IB!NCD9edjwIAMBv~I)2y!8Gv?K%;pRZj%YVJL{XhQO-`6ykY-C}
z2UFtZPY)%{tf6Ks(>LECUd-6|@~^<DO<ZaSV<_^XvNUsCWU{0dk7$Ynx59yeBf)Vn
zTI1D2mhl{YbVRM&Lk2#LsKInLqupxLk``g>%qCOPbj&$NbLHF$cY5%xTZS95#i+f)
zU^!zrn~)U=dwIb@R^o~!ffNKn;UTb6kT`;d>#?!2PS|d7d~irZ8|qk$w7fr=u)8eC
zttE5?E7tOS)8q47HP*uvTgKpow8AF7<0G94GbFU8UaKRNBr^rME~u#{gPmQ(nH^-i
zhAs^|JNsz;z~SP9N!%~jA^z`*16`84K9Mz;g8?`H;E(8k{<Ab5dkoii@RVD136z89
zRPqNeTT)*n1fjtQi9jP=7b7fgEudU3NvC6Esc~8jn#~sbgL{O%KJ`|euoh5c8RNk*
zwQR<(^jdtZAMxFLOWr2npw;1Ev1GWMvP@&fMaHzys5r$F5GqA@!X`|V#c@6Q^(O0E
zTa0E?)FPo~451^~Ph+l)=1ets%2IbFr+}}X3c1`8xXdXF=rmgR(#1cSqLm07A!#(5
z939R`%AA%E_=}Xuz4r(o-M|tsn=D9^6k$)?M}UV`K)<AsKP328{_XnYdp8KIuz1EL
zy>Xw~g;g3YNvU(h(qaWDufk0kS&<h^QC7a?PNl)s;)JnoKxQR{)aar_=Os!w2q9?q
zx+H0ery}<6ACo2}S3{4#bE(5OZ;!Y(m{BN~jYboHeZ5MYlO@SwiOx&HG9_Wkog43y
z0(EWa0{9MNqq#kvGc7c^)?`S02QIjZjY6{(<$U67K#(6H>-~z@;kwvDBOQlixkRKI
z)lh`#f_%ISE9bD<Qf3)<ZrtEc|KyLb#-O|=e$*hzEVdLJA03j%DXpl*ayGAqt^^oU
zGMOHe4XQIf%S*C6CyryLvl+wbjPYbbmS!Jn_xMmr-2+Zl#V{8Se;FY_NzQL}P%5Wf
zEHH6ZMX_jfQ4luj2+u|O9!h~z_pwT#ZPnhAu7^Tkb;Y?Wik#pBb|3X-gVJb<JY$*U
z)H-bfzvR=WA~xNU@9xChjuWQGOG+h)YBesOIfd{XiX_9;8k1|LNk*9)Qd81YmLwPK
z#yJaXafBfB6g5-u$&Sx6-H7#CN!?A+MMk63$M<|(*R2fdjX_F5yjb8n9zqzzbcb?w
z7ag4fEeL(drSs>x@$T!4#z%NT$;M`vm2MwvtDooEwfDL9{tYHcL2d-w`$zoG|LR{;
z+6p{q9@v%Ge;cdD{y4AuB{(}CjC(7oPFQLW)(h2!Aq0Wva(2Cel?7%x!={=-sOoFW
z62D$Ucn&J`ke)!*L#(Xm@oAnBg%R?^iA15$Wr=rEJuao7*=nM(WNAWNXc~<gT|eiq
zp09JtEBV1e&fR3eB%Py`q}FJ%e(@Ye3CbkHD>Y@InZzj+n%tJy66Oi)#ThvQR|-N$
z(zOMjTd8ri<+D~V31mu<3%cDNp68>K!jn!VS+It5xujOFAxsX#+pxHRaaTZFS`D8I
zXU}r)*88NBn5b^q+`LS`)54g7{d}9(fAk}ak&Kd((hA<Yd6zr)b|_ACl|OA5D-c#9
zH59k@xDkJiE5H6Ff=lNRt{_Qb1mJiM0R{f-fI>RV^O9mV!<GeJt4Y{yAUsLftdk5T
zOn3HZ^n28!sDi~}3F=LXqM)@BP#8lt8{;hIJl^l~=t{(`Va6Z6zs>g-8MT!)q^r<*
z%6K^^%`)<`WNtOaYMjC%bAjWO4D3GH(E)juBP`TiMeaJpM&gD(QK``-Jhkre+m~v1
zX-bwPgkeOd)1gtT<2VkE<6zPpr4&*~+U*XL;egp>f*aK6I1WeK?-SNfQ)HHS9`oau
z@Z=Qx4>0bRDN9|)R|17XOcJ(#|C_W9CtUc_Cvm!c(lo*I0)imK3q0KUF-bmVsYA-e
z61|v%QiQ!WK|Lbq))B`g@o<Ek&uR2}^w(C&ixS@tZ~}qy0=kaR@L-#GHmBG2_!}SD
z;8)X<S9d1-%j*kN5TbfLq!g7USDH}dIkTeRSZZWpQ3d#UiC31y@sg$1_*uzmCGdTx
zVzPN2Eg?{OiLjEdJl^1`mL#6dDJ-;FE$UH7=m$t4F{08bE3N7GS2;e~V>%oX1~nRu
z4wHMg&}$d5t&24L5YKTD#2-lN9>g!fCjA%0k*Sgmgu*#V!UO_|BF9ilNh@%vg%+(d
z%xn&&KqJV6#cGWk1_;-|ZPd_|2&aNoTLE#JQi~!)X+ToBK9nWGb#Rn}7y1=+7fY6<
z<kqF(dt9wqHp3P_ygT9bQNixyn8mDeBU)>BsGZuZ>^RdLr_|(yCXN$KQPPkaPYY(H
zVOW+}A@QB+5_m?He11LRLS3>_FHtr@8<#K$@jMSH1uj+bsVGZKX|Pt{rbDv3uVNb^
zMl?{e<jSRsTswV+*M9U0nkB7{V14}}zTZGt!{OmR-}>e^$qP%8SM1q8{?>~qB`Hm1
zB490*UiGJJ&wr?-E^<U=nrzAneu~9GB8@|>4XaH_IhtV00&76ZYC|iNgaVL$fCwDi
z(7_E{jNqiTVaSS%An=h$r0WuA3$)IWR-v#IR^xgO>t{CUZLL)gQCi|S#*rSqV8)n`
zuiZN0=Cr_077WHSjB>G#gX8<SwHi@x1y=<Kp*b=si?qNKV;XUc&P!Zj`RbKEe`UkP
zEE4i8#vrNJn)I3-LO(<~3XBFT2wjgXU*gwlxVfNoQj+Vh!Q)+ouG5G@&Yr)*!QA0!
zILG%#EbSeFAfhOj6vB|1YTC1Mr>exWi!}RiPx6;&pRmH>sFKhPu^5D{z{=8U9IL=O
z^!kd1TcSrXrYKHY#?^W$k{D?%(htCw1T_!kIo01Q6}B|wafa)<C?T<qOERD0)gy$l
zNGv4=*Y`Mc<s558i4}s<mPB!kL~*8-vYWen?e-CSOHDbPayT4glnahqAs3>E#!3fQ
z`IMqyuS{5^1tBB6BtdJaE5YA<VvXn9f@~I3<}tVq?PiO1)FcX`s)a!XCk<VX{d_@F
zggAvDoli*bz6Iw$gDM@mofhk-F0wrl#7RcD9I!MMthOi@6vC2fLtd0T@K7kVX7Mw0
z53yFAY$_H{NRpi+uKn@1dDM|KPoF{f3L&cZF>;|hO(8znxyzC=Nh#u#B95u8_mN(J
z6V(v$g2mw>VW)@VyI2HSp5pl)MPcc!uCcdVf)VIAK?I7cJ%`$p>ul~E@wJ<K>`xS>
zP((`6ZME^Dp!&O{B+D~&SyEFDj&v!d3&JIEU@i(q%M6Ccw3ka<U9#DW`0a~*#CSp?
zQo5Z=WDSBK2*`>Or{R*sDN0wCb?tVS+3_*U$&^l~PpjQx`_6UL))Q#8iWKJLUHpsO
zAO6FM*l1!4Z6g7R6%ZAc!4Ka;xPl8`_%wd4O@TmKgBK}Q#szhqVpTQiZA(qQSW?6Z
z<w}>J)4~aAnA}pv3+9JM1f34rQ{-idBPG&PxFVptvd-cC9Vj$<nbHUZpE=j05tMxE
z-hfxOx0z@k;Rrg8OVDcJMLxo*j<jW3O<dPq50ppENIciaI-27mV>XZR2ZuDW0wWAx
zzSQTLdc^GL2)C|S?XFY<L4I^{XSOKEp^Ot;YjBiDr`=^S95Ww}X*AocdOp|Qeu@6`
zTL`KO`hQ5FG5C(ga{`JoM`DqpL}GA_i&c_Sn-MBqp!3SA?!mAYSc=6GN{tKxtS6{7
zf~sn7ZIu+~1*NKd-W};sDvvylQ5~=1by|Z|0ypru@{uc~d5&-tS&<PgBR*~<Pi_j{
z85ex}{)kDYNe&0h4v#U)LAV}n7~(|{-9`^rxx{3Q^Ne|tQZplLmZP<0v+nZuo?PXk
zTe6(ZF=>JHJi47OHMd3>Mh_>orLlNQGS*8R(?M!Uv3HMhZ5`Wt21moz)+V*i7I%g}
z0AlLdKJ&t2&=h4#DIv{D^5VoR9!pvj@RP~pK2&_Dq|Q?X#!5b5jfoXtq+zYsgghZH
zD-MyA5@QV#kVca09I-4BrNVP+l(|HCURB~00$mnBP>szii8Yoo&v8_M6#`S1n6g@M
zUev%lY0U*rL@kJj0*6Z$9y|9G-?@IoV5%98CoJ+5imFUINtUR!F8hafaD>Z>3h<Xx
z9(7dZX5qPf;p!?ErDd@@1iz$JZ&0h%h@yyQqlIwY>Sxm$R|tetm5wX(1VJDj4tMtm
zFHW&eKv`;DeC1VkMq_eqSjH)%?fZXb&;Ml;&Hl6@1o*zil?A!3n1&C2NGrf9&BjWb
za=D-|6@0gH)3B9epH^gLhAaz2sqnoz<PvqFjWV`E<hZVjIni>MvcwbxQdCP?7nO#^
zbwd1FgyXsBtRQeBLO-N29`K0k^Vqo!zISuLc&0fXk4eiMljj`A3*x*6os8|#n7R|t
zwvu|f;FRN{T}jaJ_|>Z`tfU!}y)iPhv|DYWFd~W~8ns5{*Q<3k4J{;|=T&$R2}ns`
zCBwZPqOCbrHi^@mZ@%<0!#F`3ST14)i`&)bdh+?&hyVANvv3MqwPGB=w+aW#(d}(+
z{OR|3^4CJdrjMf)jtXcfh$4@&DDaer-{=z5Ew(5tJG)fkh8{A=Da#z!Ksp(tT6J(0
z*n)JC;n(VD1O2siCP&BA9fve6xwmu3e3|iN%j4{m8@#(W<n2MqekPes7tCjse$zM#
z#}BYtFt--#I2ZxKXiNb$YpL0S=Oj3hPvnPu<?;$<aEKd(tgWq4tJTOe%`!<aKwg|6
zi*!z|3!Eq-sx=70h-AJ%Ta6Kxwvx<-d$^k`M76N8O0>UJ_pA>c8J62RqpJd337!@B
zLbAw9ZhiMvtRuMcnHR7@AB9iNZ{P$jD)-5AjpszTof@Grn7n|(!Q>J*s#BCD!jy=#
zAQ=vktqAET^5p{QxVV0Vh-&oK*GVQbx{W&H@q(lM5l&X}m9-`x?FPKPJLdgK&LC6F
zlNl2M3W0GQ9529>Fg6A}w>n6*!DJ;(DXE*BCw-UFm2CCue6brbKRBYj(xKjVXw;g_
zW=ocFitGC%Nk+X{BhC|)68N6S>iQPT$&~qILa~(-L_W32F~xL`_DY9S8><|Qrc^88
zgYEcV&KO&b<7uSGt6@m1Q34zx$S_!`srephyu>?^302b3%J13gk}OY<c@CjMdex-3
za$TZkgXM5Up6B?!kJbh^@QH_Wyn2BU3L$0X?Q6lQ1%!<d<+=n}PAzCq3nONe5mz>R
zK7QpiFW)+5GS>_yW3sZqq;rm!bK-gp(~h}6n$U0pnpV<I7o2w-3QrP-KEL_IDy>Dr
z=y-yP1f5o!D5}o+S`^_ZMS0@lB&@BZ0=t$>7g%eE3c<mx_XwNkv7%0v6})%*HVa)@
zcFreL_Q%sdbFce(4M}~dq^?vFM^|`?ij7uz@HtlF%L2z2xb0@eCM(NI7+sX;d5n-2
zd<TuG{2}8Q3wVwqY}6Sqj_^E><#LH5s_I6OWVo#eg~E{zwk#>;b6h1T3s9x0OoFAN
z5w%$4DS=t?m8%`@?T_HfX>dFu;c@r&Jx1#nc;P?)d;IZ>-z39uvDRSpuYZ^MV1j72
z`N-)G-t36d!0LJ*FA&u0O&0S7j^m-V#ub%*q|`aW69^#*gMcE=h?AH`P)Fo3W;sN(
zdMuX-Z@>K><%!P5KA2F2D#Q6Df6@PaqSoqAJ9Qd;uvht}^1xtSAQ8B>#3je8HE?{7
zN^YicJP(l-l#3YYfOQo{Sga{2;uxX`$Mf)N5!rN!>pG-qh98AsEJd2*M;^*?@RUQD
z6==Ias2X{hB1?n7;wy({t;0NzS#dLd{c4-t!4&JC!*K(=0^YlI7yqN5<<jTB%-3H0
z2G()8>ig{d%YV*nmQmkW;qjFaCIe84_02W>KoSHI^ZA@8tdnLL&3XjZVouD*UB|_9
zUGim2k|qRopK>;We2(#JjD{1gUB87fC+xvL^Y5RJ*uJoq+*A|Hr81xePU*1!#x2^7
z2A6*&L^Qhi!mz0ntwsYQTpVrDc?L#fr9>7bOjF`{OxOx3^Bl_RGE^ifLOTdsX;K`|
z2P1F;pH{a+R^;S)L7XJCS68UFUFzkMhSTOkeL*T~yg!b4XP6KdE-Ei6^O#6lWNBCk
z!Nrd~i>il=i<rw>XE?i@a{SVFv2MU4r&dvB&dT{MJda9PofjZ{tPm`+CB}*qbL|Qt
zRL6pJ@O+=6-F*i82lO=bf`By15z?i<x{8#NQXBq5ZM{B-J)Ke70Ia~37A*}@3q%5U
zzVjOGsK%)mJ_c@$8kUW~ryjJi!bKIjS_{G;q(Bq~Gh7fWOW3M2PcxJUlqDv~Fixc@
zwc4PRi?9&ZBc!#Yc}AWlq<O*0)*5~tYUzTmYj|Rsk~npKbU5YaBuAGXBF!nY1q~`T
z;-YYQ;<GO>Nf*d^z}5BB)Nj7a^xAubjW*|2>KGEvUOk79C6VuuW*G(#RLDG?Q;XU>
zsIlZ(N&t9%fK;ktD(>#n?)6xALx#(kptr@^`YOD0<K)x)<zEY4yF<9Og+4q$f8gpU
zPb4#r)d-#9)qA+U3j&NWIIdgin->YfaWJmJ%8JvPFAZKK@O&RH3d!dQ^?Hpg%kjbh
zw4q2dys-Y`eLKyoeOpH8BCGaoVF?@`TW_<-5;lXJUw^F4-e3m)c^ogm%`NZVx=Zwl
zFLM4fpXci@egj;OM?H_dfAP<mEixLLt32Khu)_gXIc%KTzz+nzA2FXV2*Vm_n&Q`^
zs(5okknnvU&vi)`OX4KPueT{jN084jUdU)T;qJX1tbK6C0}ucE^Bt1<P)VJ;vO&JL
z528ef%8XrVgAxTRTPw6SFAy|bOp>vE=N6~eyX0|2pf=i2@6;)rs=X@9iZ*YABx-jk
zvXr7oDf1L#>ea+aUQ}df$EOT6A}f()&ir7Tu-U{0bxLC}d5OtN&{>6}(9o-gNNaFw
zF0$RkZ9mFlWs6IXJ;E^8eCtoYiC(KwJG0K&vpw3{VzZRy=?$XL#gqxIQk1%6krZ^-
zYRuy$y+$2tHF=ichn3c%R%<YwACfJWw7Zt3k{s>ap;kN3@zIdogQJr~PVmb(`Si06
zuhr)2SN}TK{^>uWI36Q}rgF@yO4+VsSv$Q&Yik2Ha!AHg=7R$^HhRS487=_;VXsM{
z1k!O4g|3X-9fxMW&s3L`MP5Az8lbB3zN{&PACRR5zVgVk1?j<vu+_%6(L*eS&T@n;
zpb$vu(+ORCSK!n<OfBMx{*!R|6puagEccIxy!q}s*!32z^BY|970sf=>VnS2Gq{dL
z*aC6l>$iwYHbbAqcuG^sDk;qp{3t?7RSh}J7S*(1t3#OOERMGEHu~&s?=c+zVsH1Q
z&=~C@Jc|?>B_vWPvZUnx_ue5;E~h^K2|6Yt7#(BDDP4Kuk*SeNfvkKEi^a04$hB((
z^%}Fq9Az!i3RG#)!lH{3X)Q>JE)2?%Naf)-BOYJrk?Ru2_o)>PIxS7NH6@D|eB#nN
zN2SBf?ITJd36!7{)lhlC-SLE{|Hg08eDo>)=nsFN!E%AW)}c4J!&coRN)v*p%gRcJ
z`E-2Z&sgW?jbqj~&#+t$s7K9*Of6#!R%?XoR)TO~HX2cL1N_J%tR1l|a!zfY!t<(O
zp8wEeVhc(P!YFVm+=X+Zv0BV>uK)25PzF{${uJGGiaVZQWQlevNsToX{zfW=&P$3#
ziU?ikR65J~;Sowymve!@NJWvP_@S*5<HF#24i;G1TBUbtozfbV=TXa>blWBU*%+Ir
zJb&dew&w+Vg9#;qKtQk2pj<AwpO$>)Z~sm5_9|cd+8?sChSf)}P?It1o*>9lqGq3N
zx5a!i!WcmmHMo0wpDR~4m?ih9N7cJ+g{a`~rb1!)VTjh6>F}69D4e>7d$>*3yTp}?
zm-*vA{Q>$XL{a?3@$)+`Y*^#+SN=M8{+HiD9~@R1@dx}=qmfe6*;uEwbs8s9<kJO5
z``c`;cS$C5Tv2Jk!fu_?k=4Fk8jRK$rD(0JGP8%2Wwmd6)WB8fEJsEH;fECYiM3z4
zBt07A*BclwqR<9YROh^~1tte2eY%l{t1NEag?f!A`*paq#?_~uVt+E_wYT5IthH%g
z+~jhgsHZv78k*-$6ZjHLj-ayITqXr72$+vXG!bA8S(f0P?Awici)1n-p3kW_+tl)s
z`Q9ypQ(cBfLuT{k2hMqZZqMz{hx~Vc$&cgm7rqREk1GsPR2jS!f>yV~laD`5<o1w8
zakz89?0ACC3WTEwRyq*42x$o;A1@4Wq5$Q47$w2?sCPS6>sBd@)(BN0Eo|jUgIlXn
zSc_Y05Y?KD2LtBg0a8NPh=^JZtOZ?Ilu%e(vdk7Zwjj(B+O8nly~(%!+y4vWd+*Zk
zG+6BHk{=F;vIX5xu(Hx*V{-$mEspYWT_2-eiYx%BD0Gex0%J5qkv|+-h{A}nEGp8v
zBMCgk@^~L>b4KIJ4Wf!aKLz(6XL4s<k5eyvj<YX(7V8K+X&<f^Aq1PJx43fgaXit6
z5_WIzkj`VuG{dV$c-<z(RY+ACA-kcE69q`$Lo1000_vSMd68GbT4SrEuB@hCQ4W5+
zPO3FgtBdOej1CW3&c`^OB5KwMn{{-d!6<~W7^{iXIgZY#7dc%Q^xHq+TmSukO))*B
z->!3f;|69lp;5;4>JDqG9ah&?u-c-8OAypZ5^$Xkah5*hx*1)qmj|`BD2&LntZGF&
zl28d2#|IcwaCCe?n&tnNmGSdgFDX@Pr4ma^B$aiTMpI^nd*6S9yWjjS;n6YVDJdm3
z&MI7ysoLSvRn>bQ2H3y_B@vFo2&&-)Auy#uS%r`;>0*JE1|=2Nnrfrh7RUARY7N4$
zP9vyO4<g!)7R_dtWs%XgQ=U89;A%(l=xUv<wK}U8*SPxhqZH$N$mN(`uTA7Aj<@$Q
z(;0P>acX^)Rx79~{jP`BhP!um7!Ib?8YM-MKg8--pfvhKPI98nGF74%r%Up*Bp8n{
zJ3r#oYM)N4{qOn7SDwkgELyxHT`;FS5zAKw=8mug0<_hn^Nj0%^gZr;{o6RB8AX|s
zS&fM^iZlajP>w>m4qh$9sRdOPM^tukR!Z_L2PC$r)&jyIozJk!;wrcDL(VdUws>xc
z7uJcwI*qVSGpf;R_lO!@;%vdX7xV0y8jrLkkF3{OTZ!0sWQ)fiKgam?Yt&4_*7_=r
z(M+aG(&I6qE!f&vrC#^2!r*vb1z*0i&1||rxiQ8VN?i@DNu|ik0$mnJA!#<-U?Iyg
z=JA|-kyAUkkJ$bJ&p!SrYrX!5@Zc7I?)N|6wJ?}XX+8HmSAYGtkWq*$G_J4}W(24;
zJDflJ2%e~PWuv1Zqk|zj%~7F`-*01Fi4c}B@NrMhdFi_tB@t0bz2BwK8YN|Qne%<h
zEW>DnPzt}<AUB4f*(L}ZjE|03jE_-D5H)Mm+f6V6Q-Xs)2*V<op~{pfPw9IS`Q9sh
z=l}VC#7y>CY1J6*?vsz__{p4J-KE#9v$nBDb&quM{Sa#$$~>ef4W%h6?w%0jd4|yi
z0zs?Or6`ID{4b#$_{{FTk0k?2tbH&J_=`Q~KUAe|?fK6#zI%t=@BS&OsPqdG%jHX(
zxMfOSBv>Zw+`PfcN{`9GF`d;OUgTqffZ_f=##$P^K3=nlkm|>}GRGw=E%W0ETdfKM
zp|K~{G!h}J)=XMtgsl#J&u2W?XQ2<Ncl$U&NT<6(JeeYdCC_5AJY^Zr5rxL_TsCaM
z?B1IUkB&GFI+V9=@RUi}=xosM^hxp*U1s>cPmw1a9gVniYs{6aSD4PmxK34aF3S>Y
zE32GpT1c>5CV0~s&GD3=<`a1?SW8?~-mU*}rf+|72q_V*9%p{#E9~C9&g9*9k)^1-
zAEn{Z%cn5$j6zuQ?2yBq`&_zsj>-Oj{;3U|FvO%8NB8cb9FKaxk6Ww515vINSjS@?
zXV|PD>h-J3Kwe<5NGVXhi{rZ_Wlp2FhT{fIMn_m(Q0w&YqKI~9g<`qH3PYAIFj(T)
zM--aCbJ#Q~qiZh_H_vgPJlyx*<>S6#v)iU#Ym(}eGS3Kt%7I{aXUx5OGoJm}lT4?F
zv>F|(wG`#a3xAT-T4272nN6mIjV3kE#kbXqu*ed${l&Iwmc}5J#zSLp(N>=fMu`ef
zk-+%1cX9gxQBzXH2|6(dW66cVHfku(L%0qy@(2P4oT`%Nx*p4Xj;maR5J+V!FzIBB
z6NLyNQI3PiHQ8)|uR%l}Sr%i8lH62KQCny*8K96B9APPw88uJgI+B`_EZ=*Dm%slV
z=KV7)h0pBP4W1T;PP2g<)d?JdbOcswj`p^h4X2#i+`@{C(pZF4RnlR>V9ARdH?;Uc
zfa^FcrZa>#_-ihWV#e&<mr2h)kK;Y~BtLj#uYO-$Za~Uv@6figUwhE5`NeuR|Jf*j
znbE9BgGY(Q!%!lH1R<-nkiz)Y_i%b4QA<%I%L+PcEV-}*jR@s?2**K%6;NC#2^L(>
zV>y{qO7~=t3fv$h84q!z5Fr)Pb#RN4Y&OT&K+Pk~Vr)^88$(%WC^W+46(h`mE1`&I
zM6SYfBsD3C-+%RFEu3a4Lnb$G@U+l$n)PZeD1mY$!Wi!F?JyfoIJL0}B0pIRk|N7b
ze2^i}v+B<UA+B4k1y0sNvzRe|_hr)aFW~y(<eC22wNOo#{+nyTm<Q?{d-BiEZ#`Im
zu+G}2zQFp8YwY~#w~%=y!w^_5Uf9Ib8TldwF=glG4SL-U(}QDL56<~8V05sLHkM|u
zkJo4*l=HCQXkCw7!}NH<)(PBO1Lj1sA*8?y1JXPrXtr5#Jw}5A7RQIwJ3Uko(eCyi
zp7TYXkmbq0an8^0zr}EH#31NU+`i3IWz0r<gMN4A&z$pRmM|EMxP5cX<wq_vn@=8s
z%geIB*s6*w9=JWlG46E6d^{nj1=L&ztR*RPwE5*Y=RZ_ZXZwAw{OVUZzIBW2{ys>B
zrv$xPNm(2rJ(v0ZHjXW+M<F)NnC%{5J%@a;#Fjb2+G>I@*SNJ>Ri1YgLOF!>8i)4}
z$kOZtpQ5S{RaV+HFA7Pf2}M!$;8)g9F*!aUJsRTmx&%Rl)0QOh3~4M%8>FyR)oz)S
z>lCq=5~LpW%?7C_xO(Xk*843i4!uSLEi`48F+4ot=wL#(y-sHZ4v**5yPFtoD2uY%
zh^Z<W)@TeB+%P#F;}kXO>mABcQX+pT9`w&M)XC(Rd*&RE{OVWv;r1S4Gy~}pIneN8
zv^fSP8SUMp5x6*x#H1;cd%I{QDCSG7EzxGqA}a`2R&j#h;d+sdOQY4~?zQ{$&uD~@
z2vParDk%|w7lmZ=F-7SSHaj@3VtRZ;G8p1_yVPrSa-~?tGo&^rFRnrBf_R>imoe^a
zLPVRW(<Q9?Ja+X0?OKSi4mGENHkx=bV{mZ9(f*vXr_bR83yQeJ^8$?4=(5Dv>gFee
zK$m4T;y;~{hGW83jj-oXYKeY;egS?_e6wR+a9M>IP8z@`7Z-#gLt}(b<T}&?OHrig
z*|L%h+3IVvxxw{(gzI5ki|0kvRGgATjXLw;h%C#f)oRrQT`j^ILlGwg^(IpIWVt3(
zE@hUmoE+dqHB_yIANWXHY21=Yg0<i%fi?xpbcy3S)bawM1T}KD^pw4w+e}YgVP5yx
z=$xm&u}Wzm5Rkhd77Oy_l4i~4!i6;s7egG^#TrYV=Qu*(I8J5sEd;LT6ZjrksaejJ
zB=se&PK&l`Fc2;BtZKI@ancGAf-np@clHctPH)ku*H~X$V|8s6qXgHkU1xAKV0Z73
z-Q9f_^BF~<KcK?;#Yr~a$rCOHA-NHRj;SO!fGb=UwxlUN+!B;5DNTu4B-p|{{GO9(
zi68g~$HTf*gKvdIIu7-2hh#cqxm?oicFFS`w_Za?MLL@kwYxaJkJc7XD5Nf!?H!^#
z7q{NViy}gx$a0EVL7ryRLWK~Pc{)crf?8fx^BdA|LB-s=`!31pCm5`>2}+m#<|#^J
z2`uEE#Wg-A&xu^g*)u(sSyok*wI(wbM@k&mB~23QEkjtV(`YtXj;Ac9b7o#ZRI9NW
z)JUBQjRRHkrcR!s=ewNV+T!%-E!xcnTbrBo`hDEc=kC3G93C99y}irs?g7*3lswPT
zf9CQ3<*bM0V!_~Gm-Zu%@x<Tw+bo77j$eNZM^rc;W#(kb0bb1~nH;0al)(3}dBI}m
zfJNw%E@DiUBC*VtDN%m~x7n<aK&ooVM(q}pqaowt37wTDrBswB&t6J}Uysm}j3SG1
z!#b;*XPA!<Nsf;3+MS9t(e05Yb3Ca?@)Bz`#umiOg2Lvw*@F7Kq<5ytAaYq>KgY(I
zHMCW9>rIR*k-+xeF2_eBe5X!-&E;S;pjlg~D)bhNu@q&FBkK6R4@$C}&6pn!kj*BY
z6^|rNY4uiV)FY;G#xHvE{ljtU6sXguxcFDU!t4LzKSC}vjnHFlvq{hmNtO$y$H%O#
ztsq@d9R_7VIgh~@l=P4~r&!LBUPbE0_s~}3Ne|&F4rT?ldBSR^hM@v~VykId>$x~V
zgtP`@3#{+bTHhi+8dd(WZGjU61hR=gUy!5;%2rlQMV67o1+`j(JZ$ku)a2&IY2K0o
zmB5;tQAmOHT*@*hnayZLAy*#TWH28hrH67I@+_+~0-}O03c!>Gqbt&WX>yW8;l>Nn
zWQL=>$|{Ab{r3M4VF4oy!WGmXd6ctX{37rC%ilw$mR>iYzt+HQdn`v2;>nbiwN;Q7
z$Cv0lqfDymh$B6OPRJKCq~}-DR<46a<5>^uO70F)w$cnwDy-1ev=UYFBz+H+g$QfV
zwm?J?owY5J!H|4BB53z;gMbk5ya`zzqb!tVLXj1uNk*&F$3%6UdYk(jhSv(0CsWIc
zO3AH6dM-tpkj>_-_gb94&|_!3kLQIIs5Z%}RvkQi&rP8#dt521mVT^I^9gB@;JE<^
z2|Vy#{Q1Y+R(KRE6zb%Ls)WTA0;@qwL9^*Vw#1|vw$vzfQbjceTS6)<qR0?|LS>So
z%yGjAFAQn-yNnM8G-RFSa!I2WB3zd;$q+RVZh+fVq|*gqt%3ACCX*rMe2(90QL8ub
zYY{zX{p7~35mrN<qfL%B3SC&zB12{wEib}BbFZ_)TfSm4DS2ELsK8>f0MD;sghR7i
z!?=PZ%Lu}NvM4G`ED2IL6s5)(i*P-ptB}%?#4sN(Slto?CdEvL$g-LmY=r^OKJ^h^
z_}pjt!k53qlTSWHtzH8xN;ycWu(n#WWuaNd2|GLY`QZ<Kz&F1BHNN$y-(@(OoD5z3
zViLQlHCRq)kV2gNvUpOk)L^8<55ch+<tW3JC%7wN5!PZ$OI`|WkwYmFzMwP(P7va`
z4(+uS4)1PbshE-~^ig3zv6xru%JY#;kL7rVqZF;)8q>*;<>4-Fqd~3F!VN>(-YU!S
z1dK)rgUvE@SfEvoDJAo=K*tF!$HSKz8`im}>SW`RCq%(opfP!X;|3@%!mSzHn$O~R
zf^-yRSs<lz(uP)OZ7T4)>)`}GLTPlam>kdPZ)hT^5#vJ~U9E-4b$I0B1wQrJ&+=El
z`W2pf<{7%3F5(2%E0qH$Ye8#ElB66Q?DNjsZ}E+<eT_f(qpxvrbVQc_<yk)(W5|jE
z=~$eVJ{N!WtIQ4$kozNiB&!<@qJE8BXUvX|>Ge8zp@S|{1dz|?*s??^7snckWJxkn
z2&E8VfHo4>^-!+Myf7Re&FQXnetgcKXfTcA;Diz6Io1|fS5jNqKo2J@CljQ#IQ0g8
ztx1?<OlMP+wHRHZ^OQK%c+#cW?Q$imA=XcEgPi&jPPqk+w3MRix+Sw29)hb+o+8O&
zw3fJ@N1o@{vcL#154tS2+PhWgQ<ep3p5nwY<#d4K*74ln&t1rWG4bJ&I>1Q5<|m(L
z@6P+gUwe_$XIE*iM(DC69ga~-5i~3CtLu1ZBT$}0;J64JOqo+IM<nB8lz_0^!59at
z1Xg)WOOIJ*SrZlJ0;<xiR2BHx2?9(}R1n)jBanEjJs6FNkB5ZaE>0~X>U5bbrf6Fk
zkOu(-%{J7#{P2zY^t~GGc85b;UWq5Xz?^Qofs+aZE`_mNxU@+r4a+2<+2~LfCF10!
zs+0n25wgN{I8rg6Oz~Y8UwQb+n8D7g_|7Uyoj`bhiCQavnepJK6xKrEarR4JWcSwl
zB(J>9r7Nch+a5a4NsosF^%`yzQCdyt29$-x3p^a<AaE!%mm(e$kB)Ezc+Do-D6CQ#
z<uS?~k^+49B=NTu;8#jjt(P#wl)9>r<t0)%1ZyjBJS09IQtNebgOI4xB|fN77Ov|N
zNJ*{RW$8A#@$L?5L5*6c%YErmPLB9wlv0oCh#ahSFj8>((kT{sLQ$5~Ye7ZaKY@5k
zDNlG;;-Ok*IvNvtKA!Y(=era~Z{xd8)!yP~$DRAIacnd~Y8>e!&?swALgN6bwHRS(
z*22osLYGKKVw43(p`?RNbEM<o)FZU8;3^cA=20s}*zSUWd=cZw%Inz)1M=yNBG2(^
zZpD=hOVTXGuhr;>HR9=rWHw=$q<D=sP8ibacFD4YT3)c6%}~aGtv2_fFx1;!7Pdh#
znX$FC#ep{5u#(Z<4W3`G(f3`9(v((GZ}nLuakYu7s`V#M+Kkqrs{EG*S4pgtEEB^t
z7DNq;a0IGMF?U|&bf?LuKlerc-hcUD^3*d=(`<K8j`~0wN1p5lprisv;J5-ms?q87
zc>IYc_*;MLZ*uF_E&kc>{B!>4@BB~PyT4s&M}O{Xp=3->Lk38pa4B(w#uu=#kfL!^
zNRj2_WrlL9Bp4--u8S%v4{WzltESnNz;P5-8%n^9YOI_+MKPP<xr%Ht$8WaC<}<Q*
zN#y$or3jj}%0D*r=&Ws#FPAK4L*i_J8`g2^4O*Qpd0uhy7S62FWqJW+nNVg0jvLTl
z+2XCe0b3U@a0F}%m&xdW#}+d#cid_iB=FFtfyR;-1#ajf9EFlDR#d|h!W7sz#cTOQ
z^*Tp8mU(K)N<&c;)P04${R7U268Gt+`TPHy|1DqsD}ROl`f8QXp9~sSMes^HCxik8
z$`yEiNVDDM@|DYc<*Q%e|8ResZ+`1r{Imb*pY!7PULwmrB**k82xATVM+ZDDp2kQ)
z_pvKn`mMjl@*n*Z{5+;}s)dkHPGfYM(pXu;N(Vz#fun?|UaPQD#$w_L`FsM#(&+cW
z_bIGJDGw3U+1{S>j1Wjw{rtKtk&aXC+rE!^aAs<%spw9Nn(q>iXXw1d?^GR>U^yi%
zN<7!c^L<*8i{EH*=l+6pcbo31Qw*u|iZK+&dtB)I1gQtgM<Z$WYt-5e?jCJdc7hKp
z^wl}9%d*<Dl|$eM#ET^(3!ZYR1qtPy?@>ShDf+!0xAzWy;phB^YrooxujRM8oca7`
z*}rxdzt`cN14X?|3DXj<QAY#;MV!#?bZ~;8@`$wtp$qcD;4A{-;Q`5L2tuM7O$t}w
zNI^aHDDn)bf55?lhaZ%1D=wVSC!)ksV=)B1E+SkI&*wz0PZ-we_E+(gBJ_N$7Rb<J
zcQoVhXhe6TPpeU<bM`Xty!{&U{Ud(U^J%qUv6$h74eFf^dxHVOQAi=sx`HiQDwyJf
zp>Jahu2STMCSAmo&78W?3~z69RI_AxR(=2C$Go=x*uz??7u%8e{dLZM`SToaA3|e=
z*Y6uzO5>yjjdmM650j<T+Fcqe4p6NZs|$*}jx+a(kB^9lgK9&q)hHc>t0a+MwZ%B$
z2TCEZ28^&E#R*rin#PkDY_1W)60NNuW>b>coS^GdueT_>Sw+WJ4%SFi=yK=YF{AN}
z)p*8gvx#0m$Lsg+pvP0baMq<(Gt3tY8toNS7%)88N4b@r(r8`j3+)3fU4_T6*5WFM
zJk3d$F{aVL%^e1}?{hRRD9tY}p#zwh0^#JiQevzETP2~6tUg<drbuFh8{v1`xV3;J
zNvOB#Sgpy1Q>-?Gp@(RM2;T>7tLbIHqHr4xiX>+;9ntfuVVFV+$~?!b*RdydaaI?U
z*%CLXQ}6Zgnsti7gmN;$<{4hSg&#!J+JYciuuL*sHy{Wj{DzBHZ*cwggx%SYQ=#C@
znMZI=UErHLw<vd~JilJz+k*PWX_{9*&i&zi-1a6xt&QUa;CcvE9qLxXI8sDMyZ9<3
zNoUMbMXQupCCId(5jy<u|9Aft(W#5v-?_=Pn{OheORLdgYio;EyUk=iMoP)X#wpg<
zHi@DLrB2%J(nU#0z2)%a6HjpQ{CWQRZ~is@`+xWk`5*q#KPEYus{MIg5>snAunCtP
zbt3q$s&L-PyH=vf;*@2wq_f_^sRbwYPy#;+NRB7Cg+^it+YM~!BZR^hmCn=(aO!oe
zF&rQ7;wp!*t#Cb$;-o7gU8kysX+svz5F)~>H|R7%vgsT>T~Op1veB;E^U9&wFE}3T
zla(cbAK-dEPOU*KUGDDO;9|p}-(REg#M8X<zTjWKw}&y3^Q)Cue)D6WV<t;1K7ON%
z8?_PYq@w2dSWht(hEOE51_O#x5$BR+Ubzow$3>PgfBo0KMELmgBwF+P|N4I>%TfZr
z!Pe#$E32y*H1oxbwY5#wR@Z5_dU(DI7_3kT$0tw@7cXDt^r=n0`fI<&AN<iD@qhY1
z|6jRv`_9kwSzBuv3<p(3N(e+4vGLJQaCr9)gCD)gy?MyNAfX=!v{HDDM)kf05shA#
z#>vGMTS-pbv_?8QCY_C0V)46uFdl*H(Cb9J_5R?b3U;z@f50(GI!-13&<0cLYCla&
zoFJmH=`$aWDCRLiqd|XlgXL^SSZ^Q=SmlBvINCcvw%e@C$DBR?7(*lZ-FIH)x5XNl
z`Vy@R@)EkM8~;D*{yWaH>$>kmzk8p2bLG@IqtVD2AVE+hMNuXxij+i4wk6A!EZdST
z=aD~teq+D)?3rhe!`SmIT9#y4VJvwhE6}2tKoS5!<j`m|8lAgxy!oDd_TKZysRkiR
zqzMr%`K`~VZgo{x-MaVeb#_?m7Z}7LSyABn9!eD@^h=hmNJ^Jo6KPpEj*XI%XcUq*
zV`@5}SleV!+W_c57r!La=^NPTKgo^VTiVo#U6Xn#9jX(v?0@6|=Iaeiuglu%7W*cq
z7@KR64B7~xkiH96Da)W$DOrFt_?0TIwTD4_leo7-fW~-LHn!S0MC7T3)Yoqy8<1#9
zn-+vbO3*?fqyrh4!k~=7a(${*iz3f(s{xJX1nFo5NR(4rlC3Xa;rNt`vRvFuaQi#%
z<*sl4Hoo$kzsAdFpXFw%^hX){?s<fExJIZo)u4vsR!Z9|G^R8#-Cf`$xiJWlBa}th
zOHtDXM|!j-Pq2Jxb+=?Lc11m|U(Db4i4qbkz6z3kL;mD?{Ex45^hOxos6<F(BW;()
z{C*C-_hDwHr%2Y;SY6xUz|=nc`7uUYtJs!6xHjxo+thBp2+L5ZHgPS7etVO+yN$0E
ziJP;svPNKK<Y|Gdco@4hQ#L}Ng@r~Tq_sOq0m4Cw(sNH4gBw&4CdCvPp6AkNPLRhj
zSQ6!w4h6l<YaE-MCesq%*gW*X_jB_8yLj=pevQ{wFL0pgl3M}$Z@rU^or~ncqUKdf
z0o7fbD0BTbGXjl8ZWKZkn8ISzi<z3W@MJ)J>IBzb{z7@L?EbdltN$(^3?*#2{Qg%b
zk2lNZd}|X*j6h9{lZPWLE#HKr5`aSL0?W0jP4A~ZQNv^fm(D)Vv0Dz3tDMU8IPoYX
zR~d{7tSrZJy|S%r3>Z_|0#>RN%Fyc!nBG5&v}}|rTTqtcf@5LVYUn7$XoV6IDIJ2D
zX^J=^YIh0ajLP^Jeo$p>a)!~MhhM4TNDH&uVo7&asJTsw6oNG4{&&BV_k8$+oO}H9
zJoSq|%|sG$==~q#%kDu2`C+ii)B1^<_R$~4oO$^gDpBM`#`tu=dukE5RR)PgSV%ev
zJhwCCN-)hT0qrNhL=^uXlT&ltfB(bWdFQ<xIB<Yky^fTUPyG9T!P8HF1!)PItriFN
zAL7nC?&Y@I?xE3~z!DZ9F_wkxSlo2vF#oUr=l_Gd?z)qI`LF&BuV1+EEsgJ6jdl>A
zt3Em{N>INxI1=n#oid{{+~y>;xxLtd#m2>JBypdGgL7ClA6rQBBqdKGOrB!9e(C(M
zd&rftlp*qJHQK8iRK~}!J&!z2D6$OCFZJN41lU?3jNZji3A|bpQ>l`4y2R}ccC~`n
zY*8PZq)-K}<KTNiX?y2+Y;SH-&qtsndXO;Q3i!$Y`5)6+TjGnK_}6TodzGr+#Jc$*
zUb`Hk=Rhf&M#E>%{wdDAw!+3@k1UQ6j>WrfnP#uKN@`rfw5;M?>ZNqIVwUSiSiQQ!
z;LOirusD9~X1@LXAL7W7qfAUr;`=_Aubk%-Kl86qs=)PZrY2{({r0=K>(2X`o7+dF
zQpcE*m?9jDX1&VCKl%}l96ijpz4rre`&h?uToz5VL}23tb>?orjr!Oam1cw9<%_6c
z!NUAam|B%$v_!31E!X*7OtUd%N6&H`>f_Vcwo5o@Lw6gy)?^?OI-Bd*qM%3>Qdp&k
zzh$8*TN{QlFAS7bUs3wcmS6x$tvN;<B?zO+YosXWO-q6!+1lQsVY>L^dnsbYRI9;z
zKKgw;b8J7C9{m(vcZduJv<{uXpWn;k(pjV%;5gOtFh|+(!4l~54WUH|@e#t1M-cWy
zrsw9U`*qfwNjjZA*XQBu^_#EG+t)vt-N!#{?!Ny2-`f2BmMe7v2^JXJX3w#sOkzMs
zR43}p&5coM2AHIe6h6urP^IW<sXi^i6lkTuvZzl^FxcKA>-C{x*jVr2NQWXXf#sIt
zARPx|m2cK6Y?M%B0*sMZ7_e=OrBOP^D2?O!yCSEW!1Xa+PMT$C+os>!!I*^o2N#Hs
z9VhA}c*1b(<~!Lpw}+qo>p#rkmCGEz|J__V_D)QuX!;t@XzXHym5d+`(Qyx=E`=!R
z+eQ^ciDonc&$h5!8zl@$cZ2n7R}n_zN(gKl+jXcl8;nm(P_0x*;)pCwh@yyL9MkO&
zD2nSouU}1q|9=ZX2{M&JmgDET96WWBu{@&~wJ~~(>6s>e)gyCleA}i_3TyYIQ%aS9
z*`=2Y*DdwwcJoE~ik+<iuH}$rdD+Tx9i(GpBnYFiv_nBoAuKQwOMr1~j45gBLTMbo
zLR#cV4HYjS%nQ;Y#|W2=%?*5CGkbDBYyKRLUBwrIdmng&{WG)t!e9SkwpTYe^`Sq_
z%awx!`|hRbfva<@B1Fa`h)3wS3t^8;*i?l^6^ba)3`T;nIyjz9VREAFWwusU%1Ve3
z__mGXd(@i^CMPFx9OsRE38RQ%6ww=uc6|e{=gGI8F9p!L>p}c07l;>^fTC<mSvHn3
z<Ot^X?59~<z@#CA&IZH&fVelp@@?FvhcGbM++x`4;CViku`!%l18G?$WWknrUO;2K
z&P%VXGr4yLEW5-M<$1}T6$0CFP(h_+!BaZRD5J4lpT^#KvT#H)7~!=X8ueC*cghQF
zOOi`X6vkAXguMq3Fk0A$F%Hr81_utGqWQs(bMcv{SbXYH)WwUuE(GV!Y$5z6Ud3ki
z(0**+XMMZF#@YxeB(+VOo9}X|2}52;dKJO>B4y|58pC49c4tU79J2r5KK_?~@~=5|
z)2&#JySoi9?Er_vkc$`3<9QaAEqU$SOMKxApJsA$hCA-Kmv=w%er`H(3$`P%J)b;R
z*xKcLKl)J)A3VZ8{73(or=NM|kF6T8gvI>3zMbtSA4jHrBn6HXRBTCD6cnPu#Kb;q
zRl_7hM*Tj~uurodP@ia(RkmEwU0o-QA}X~8{@56HrBZGnq`-DcS*xv$9bUh<#;KEg
zN#nE>rp|Mu=arR`N(G(g<yxdmjFSN>6BGC-;=z!#H=xpLVolF68Vpbxgaw6Bq*=tl
zDZ%Wq1y=f+Xnm7nG~(dNyQuxmf5^Z4vF~Rh=`ecrG|#-0(&^>cex3bCW|%p+2d`S;
z{FO5#F?fzkk;Z&<evBZfpsbK(51wAxB1{T)2FrAJdXOc2_+uaC@Bh^QOk-?vmoQb{
zv*Y7abUIx+?G2<fT)A?d7he1dzxr#x$n4BK?|kQb`L;*i&(!n`mg8f5h{K4(`w!lb
zvA%TWQYm30Ep(A1Jr{p+n(=CddN{<NY!DPS;}bP(-y-u|9A|g#mZfBqkmcd6&;`O6
zyha`Az_8th+z?qFqhW?;+4P4IP2VXCIF`f+gVX}4T{4lOgpJUpMVpZh+Ne@a#uz-m
zLYCz?zE8bcCmapQiUNeB+wU+j?lOP!0GB%o9VhtG;zQr}LtNN<fZzIue}g%rIrV4%
z49|sC>iZw0<q9m7BJ&Y44l(f%9d|I{kkn{AOM>MP#v?|Npc>~G33fI>Z7d@5;tjm1
zZ#y`FPovdha&i(OAdMr^-MPCTju?!h(rxr@oV($Vu&J|TX&(ZTYEY+^r_>J~!(Ldx
zl_^Pg8Ivbe0v};nyF`bt$~jn4mJskFN9%%WbBs}Yiy|77+H`47y-@+HlsgfY1zM69
z5k+E=$2O7nNDF~QisLIPHh9u10hgAI))`h(U{y*(?l?=x(wKI8n|h;4rCDd}=9@YD
z+-r0@I|w1DOwP0a_Io+Mbdlvwk8>|x<HDJ3ylRu6?(+T*-AiW2Tv}dcqI#5xIh#Fm
zHv8JIL3*Azw&;Zxt)`7%@yO7G%V)8}gu8FMo%<fRp9dd&fFnnbGCey>z0tsR9kMhb
zO%l>Hq1W&6;!7{`r7u0kQ%^s`($X4fmSOZCC$ar)2=UeSkdzW(6atHSP^WH1)K8qk
zPDeOa%3%E>Qs)Fg9dkWju0M^qj+0a9B13CKwb3H#?oa?m2rvbe%C5(tEz6u!kmoUZ
zVv#2{k@Cq3jbmjvzM*1)D<le8+DRoTGA*!t7t6MYi-as1G8lAeH0xByTFe}{iPfuX
z4Et?_?J#-ZIAaHo(LehfSJt<9a-(2lWkgUPqcvIKecyQ(+eL@9l`W>4Cm7#vb7;Y0
z`toy_)DVUiy+~4TI=EGjevvV{a+)fJhws0i2OoTp`|i7s1BVVWIWtYYUdNIaX&S$g
zFWc=Mo_XeJzVz5*JooI2tZ#0UXZg3BE7};cJ^Q)i?t6LhSAK!$>J^NzaD>1{f@3i|
zIY~C`BdnCc#u}DZ=p-lW^(lrU(ljL>g*cW#kA_5r!dKu`D`1zKY**Ttpu*}v(+e{y
zwvAD$BqHsK7)#4SyAD#7;=eehQ-#qQV=ScW6X25P8J6QyuQo`ckTgrtwoR|!Bd9p|
zfn@BC+c^KqWk$m;MrYWyChz#rNBOTWJWeg#VR~vGuZom+yzds~W+!QmH*jo=@A~L_
zNK;L%=JPwB{0dh`un)n8o%2eTv(kaMoe)M_Sd^`!54``oS(rIU8l}Wpm$R?E%qy?F
z#OvqI@x&94vAw-r>RJkc<5^T|0UMj^T)y%;k3Ifb?z-~<KJr~3=lJoHR9a)C!w_BM
z+;`92{D)6`g1`J%{whyB{S->;8$LJm(v;~(-p5$2&a0pNEu3BtAq<YN5GWdBEh?5I
z9SxAykp9*>wOWH>l+YWlB7BE<FvR2;u7s@D!6=PCW^nvct5n*hCZP!`yne33%{Q0o
z2SAw*yVpwEHd+Xz)~KQ=NpyuN<s_uUuQtfy6jS85u1lldBo0IJJTC=i!y&bX&)8&x
z;q)YfI3o$WAP+II{|I+H@?Ms{@H^Dni!^FeXs^b-@4lD0nF$=%=kT%p{LuV7X`11A
z4!``%kJ3Fb%YcGwcFId}PIo8?SJ%js97_tuCMWss5B~tJ6%Y-Fbb33Se&q#TdhrEb
zef1U2ymW>%ODPJ4v;{%Mqf+%*U0vk$^RMvv&wr8!?tg@jeAo9eF*!xm@feNb8#31V
zjrrGn)#_6^H2fNMA(_ABPMmy%EmOKnXR(FB^}KRTzYdWS0<jC7(ng{41Zg=0wFb$c
zi*^mlXdD5a@0KNBTOx!Fg(Axn^4KO%9Y)GS6$-~maXmxD64+QMp-bJ%VTLUgwo|Gi
zBxy=K3`vuSX0uLndWz;)jbWG&4STc}4zmB&yRpWn8T2-}u(-zOPZtdO1(kY>JqIUv
z;9a+~ytqNXH)OJTGn2CpN9GM<XP!W%Fp6yYiD1@pumYE~+hKJ1CB~~2j^A<{?|jEQ
zxa+RF*n40<6O)tF>UG2}5hzVlk|btrV~r=De3Hi=dz=?vdWCkoM_%M_3p(qqZ|baF
ze3Ak?H`sB)aJb2MV3UgkvItiQY-u4JyG)Xz6mK+I?M~VnsT4($mJ00^pCSsW1vWe3
zkXGGA3JV-Tu5wg3q(7SG>eduXNkCqoWwRGi^T3X~j9Wc!o*6RJ%n2-wU$3BsF-Dhw
zd8A;_+aXOO8jWcxL4yM~ouWCum$A7fN@obS&TS7p#D!1&Dw8sYs7<Jcc&#zC3vSKF
z$^=)gZSeedg2ZBOp~j!RNux7EcVsb4Et=9LvR$?d2R%N^_rLQYnn!M-Qm<2~SEw~x
z1l1aTP{FosT(3gCS(aWGP~3d#R(|07{~TLe+dT2a6a30A|1wW}`O6H$h(7_G^dC8{
z-{97+FRL^HCpR!05vDzwz9g3^w#s04vu|7*qcrBt`J&6uqEO^%g6)<!QjsP2o<)`<
z)B*=fI0(xk%M$XipgWl3@>YvgRi&uSu-zNd^gxDfX1yLa&y5(b<@l81)kT=@vaia8
zq|@EN7)8ae6Zmxw-uEDT5BHdynM9i$>H6II;DbDW;YAwxkRmqZp2Vvgs7U-q1$Pwk
z%4^q1LW3o3Zaz8AcQp$JnV}onM7aRlqc0p*w1qc2&5ytL<FpPOC#Y4a)T&flO@c}d
z*RNnni5FDMwiSjw`ww!*9e48=|Kg9ay0XgWKKD6(=@);IuRQzww>*9qR#*7=N4}4n
zl9U&I<>$#Zc5nelN-Dm<WNno8&=e&74pS4&UGY{y5rybH$F^-O;bJ7@(GaC_L_UsN
zucMSiNeC)csuiD|&X9vsHIyp7fXiGLCFDX{NMm7yjj#kp8nh|O4{dN=r^H`H8Ft{~
zxGrgtk)$ERL5Coy;3~tR!$+8$Jwju;MxKpuyc!4Zy`RRh69l6^M`jy5JEl>I;^@(R
zq*+L>*C+5@_8eGXbEAh8hSO(WWpTaBQ=^n~i6T>_H>zt(uq2JiDW+y;_~Ms7$1AVC
z%rnnE$;Q??VH6QX5mA(oq$PV&n^KQV$`Hr1Xw-dr-4TPqkmaRi9(>?kyz^b}=D@y#
z=)6D`83zv>;6Hrg6a0lA`C*=U?)k5gyLkI!I2!Wgg^PUuM}GkA1)TZKUqf_8RHPsh
z60H-4y;bTBm%NDaD5&^0SRl2g2uBE&W7!T82ctrgew!josI|tht5tGU;7On9*-6x^
z9d^1yn!B<xs9n~vCDGC#v4B$47)zpTjix}U0(8l;PSONRf$O;_&m&6;;&@0=WK?Sn
z@|8s%{NeAV*G;MT7AhSg{c#@t^FPXeJpDAyw95mx-o+=M*}-I*z58Yu4tosx16r*X
zd&g&4y0(Sq3O@hrMe_BOr^Aeki2@BS63?|zLeQF-X1lY^r#|)DJoo(5Jp0^Jboy=L
zI3^4u;y5Qsi&D>21ML)P4e<T4?c41PSX*1?+_`f+@W4BH=p7GJYc+4kSTC(Ck>z<g
z4{BA~`dKa}Nin04FzwRx1i4Icr6?g0MuX>Klm=yX--p5&U2eveAxjf1$H5Lf>@21d
zxMXpR(F(_^V8rf0Jk010=ee>u#&TYvsL!y|8_^6vj@rz)Jx<P!n5g&hEeIM_RFs14
zVmmfuFc|EBFnC@CIU4cc2S37~7ZL;#qf`9m1n>OV_www||66LEC92*SJ9dFrYcf(P
zPT<oTZ}RvTPZNzaQrMh2Il+fp9?A|F#?dZ*(xO)w+IEB1Y=fVA|Hm0WcmgNzsMM->
z)e5d3;J6-^W#LuoZ!}r=9XQ1O_dUc<{P^EsX=#Zko_K=)^H+X_=bn3>FiF3qgW6l#
z)IrJAQAMc>gJqDmgJqQr7|*Yw`lXq7)Egk8kf7PZxDHlf$fAt077p67kgh`-h4`*V
zCK76%U{f3FPKl#aMM^dbSlT?y%e{TX^LtphcNVQJb}n9J=D-5gnolu^c>INDIWpSe
z)Z}Fv%>dz*4iA=9Qt`uZge5I(%f_)?Y%gGD-vrXhQ7XeRFm>!W-uMK*$vAZQAZz_?
z{=#4SKAK}yHrBRqs#QMl!FRE=xJgkc&YZr$P+6ELB*`th6~Tr{dH(7uX8a(1<B&Y|
zDDul+qEe|dH@m>h%mUNXb1W<@Fg9MY;^yWS7;8;nSvE+pU6@<g!}ost<9z3Leg|KE
z{Bi!(zy3Eo@s+3EL}~vYe22Fy(kX;MsS*PtZLn+?ky{ucas6HUmrOC*>0kvOexrf$
z9Au^_qOA0T6as7;sch0X!gmAmB%|g#j0Oq*t{7^OC8SZ4i|Yq@H9COaH_P5*bA+Ru
z-qH%Qhxg*S4)Jz}M_+k?Q_XFTPb^cd)xpWoMq}G$do~P19NWdRY#iIcsn(ee#*uab
zDhJ17&+T{cg6C5civ5R=;CBr_@FO3^c3^w6!~Fa#-}SNguy}19T^O!hyhc1OiAE`r
zlI&CjYiY=1mscqod+F<h{4>9Uv|Z};Ci8Q9nVz0!W@e80`FWbn2Daz0Fu#|2y@iw|
zF3I&A4jekdkNoHXzT-RI&%q<dzvbauymkfUxJ*9yFt_F@FaF|xLhSUw_K?Eh1rodE
zFxnnsw5HaqqX5UU(YeC)t9U_W*E6{wjfP~y9^uv|s>rcwby7={#R*e&L$4cR&9u<c
zAndY*ZV*W2V2maqrBDKq<$yvM3&%(V7D|+CITL4)E9_lOtUQWI)0kSdifw?=h{?St
z5e~#rgjdP1D@`8!Qy=5<Fa8T^+n1P{Kg<_D_Z%<0@G5uTcapjJCW+tV>8Bs(nXjCq
zW^3H-^L%-Ai_5tnHUbNS?O4QGQRJG@_BQ|er~ePaVTU*#q0y)Uv?}E)^4yeFVgtsM
zehuaK6>%8S?W7C`8LBXh!V#~XJIjYY_z~{9^DbP+p~w^m4jkm4{*!;gpZUQbVs(Av
zTORLk6vh}{f9(wByT6xx-|;~TG_U^WU&R{62-`)e46kBiH!Q-n7}s_1D#05}a-Au>
zN&_zluq}r|HA%ug*|5(rO$o*)kku-gWl(9%WX;g-giOqjmmqbk^nkVyNadnYBxERM
zka+?|BaDq>EEFX(-f?|Ql7Ve-90x0tq-j`6lL!kDMOa17^qzeP#}J1@>TZMP`~mKK
z-v{Y^=9d|dSMlmqe&;tH<H;|-z&-b$WO}Aa9QXPBXJ6pCXD%>Zg{{T2Txf?Z7Lvl0
z-o}<K36-WWhO?J0@&Ebf|Aby=lO#<rZ_0<FFysYj1V+nUWc6-7WI3Z@L8qNC9Ap%Q
zqTSx+l~+#lp7*@}hKzNRB;`S47u!}Uis@Zeu8YYmw2-*IO};DfG2H23`7VC5iE$ld
zF3F>;ls6K9Wg~5yEROKpk||va9Qq>0Gk~xtvXm^Ub9sF~uY?CE_Rg{Q*c?f$=v`Z4
z=Fnbz-z91Hc=YAxxjEe7#P~I8jRrUwMr$0~K_G~t2wU1%!opGpzuv;DP9g20)XH^z
z4&Hk|&;FnPfriL9c;qm8HQ@(;?0fKho6XG~X6I-4@JHXn;<XL5(wskcg>cd(jdSAb
zb9XZ0u}drH#$I|RCHwSmAuWemt;zh{9%iQJnVy+rVSb+GSPR>+nV;W7voVIWN`Yg)
z;&Sx(2@W4U%EvzX5kC8w&+xDQ_kYWaFP|nazFIYXEAf^#b=OO?CD6RUSWum45sCtT
zqJl`erC6lz;Tc2P>oXk1I5QI%4M}f+018#INo*;w9gj3lN_LXYiBe6T31Sta6T#Bv
z1TSR=n0oh}Y^?Wr@{4B~#0f^jl}jDGKymwnN0@%sLu_1JL39l_ne)`gJWO5^6pSfv
zkT+D_EgLKc6!f}ntXc!tYf))TGQV#>;l?HE4s3S&{PHjSF7N%o9aO3|aiZ9Fc!oRf
zJxQ-S;`EuTv_r#iFye}pa$YO?L!0F4fZ;38lV=4kin|}YkDvIve~Tl>Zl+qRVOthL
zK$6D%fB)J4Ot;hH#EF}E$3ySt`0<-?-BOZXNSj(?jQ73&{oHZqo&3@-{R;o;-~3xP
zx3>R>C()L0z+F(NE~_fIlIr9HQQuIRX`uR>Se}bh^>OopwB2Vgitwf<$m5KB*GsQZ
z3Iw>)LQ0z~EwCL2S3^I{DKbOu#^})D_0<V321l8B&n+xoSmX1*_Y%=A=I-*vHr1x#
z?suGI?41vD^}-cs7aU~?zvg3#0%4VTu8IQ3c1tR|?INU2k!FlWeS)z`9J|WI+<ux9
zEk@lH#*RM7W6xdWKmYQlc+a=rhBi5=7MwV>pL-v=m8+Llc;(C$I*G+{6m!u^xTH0s
zzC&_;hrw_aohv-c@SPv|FhBk`{tkQhAEpviN)0#x?e-@B^q>Aiq;$CD*4ugL!ACfF
z;1ITLl{g*Sr8PeBwv6BU#u`Z)VOKp`x82T7-xKl5Z~i)s^&YK>7XG*gN>PL(+`z}F
zR!Q=VYNbMBYPMvHQeq-yp>cd49n=U1UE*FFH%qYmI;(3tR9wSGJ1i?sQk0MjVWU-w
zN^{cCp*L!<GN>?!Av0j1n5#$3He+gDh!Z&26-icPNMYl-W!vTrMxE9agC4$ZG3ae$
zB-9(@2(QMWyWYX8|JNtr+N<3Dz<msQA%0L}eQAfw=P#2QiD!9?PuR%rWtLYKxtJ;<
zQv$bz6hul>=n@B8sGJKIE+fS5O<MA0cAY7%Ptjk68JIW8*BAp?s@PcXG8#qfY!B#m
z`mC?7^X^C9&3Aq9yQx;{7^AuGzWey8zxVg}yFc~!Np66|#hc^Yg>#HX1N?f0nFk(1
zYs0I*`$-zBZTxx_t6`UZuW5|ydpMN}NtWVP0&0`fyIP%KbcRp{$M?{cI?<p{+}*)R
zB9vcaV|APHn#IaWmm>?~WnNpd%m<xPBpF%c(jPTh8C2;-n#>3+3TEpebIpXB7vTmL
zQut&l$8s!u&)>cHr5CBmGt%`{s+0Q}^>@&N4wXs+;nz6zZ6D^-kNp<K#b>$yj_>E`
zT;Ww}EM4B<{Mm~{g~63BGjjp5eV*0rO|E2$*pzyd$`VAmqS!sy7E!{v*DrmQCMQ7X
zl1O&FdH!aA&F{|#Ly{D%uCy5pLbf)$Y;SkCbm`&^8RswxiQ@?0aY`|9qd*vP3#wBS
z#14296^J_6l`1+Oy|LDhMiKVZI7UG-7$J-)5A?!d3xRZ8vNS8jyLCa7Yof3~`@3uX
z#uzW9d)WK!_pp3zi!XlqCBi(%Xt;c-O;Ckf?>kKEp$A!8-9&XQj_C{3nhs{S!e6Q%
zX!NeifX*|nulwEh4tBLk#i=nqcYwJgM;UHiV`lC_9$nqwSAOwRJoxYlYITn!Q5-tH
zz}*j=qO&vLm9vX%#2&pY<(j>_)(;%AiyeA{byQxi^>;mRAAkMt|7{K(Il0Sfmk0q-
zG~%EAZ~vI#V8qFrZ{_Yg@8`scllXq=b1p51=J+Jv{ju-ifd?MsKmF1#@v}eo3oNg$
zzU8(4kF=>%TA@{rRb*I+q~>{q-8OZZVq1d9k)^L|9Aab88*SXF334IPd5*xKv?fgo
z9M=Nh#YW>O4UW&KQ0O=!?alMp_B_#^D*df%^t#);=OcIE2ZDYtW4oO&b8vyV=?2Lt
zVpGnt_{x2FTLX@IYt*Y13R8k=uY0)}16tS!X<?-yo9CXRF)<Adg>VB}b9?Drd4Y-=
zQEyl*Uc1JB`IRnr-gAT_#}07*{3;h-f0_%gts?Vnp8UcqY-|iUljLmV3RmaYB&GZV
zhUVNnf9<dSJ?^;mJ}loY7cFDZMx(Xi;)RR!hdV4Tzs6%ve2SY++{TAK_&uCDbtjIu
zYoG1-OifPmmw)U>xas6g{O!N@KXLKm<+nD{f6Q^c^p2ranlCb>Qh1hwwr!%$4uMXw
zZAt1_NZ-K}DOOgXH#@jvP1LTxCKybiNs5voriH;3KCUjvrN^kq5FrfOb9|{gLv6mo
z*5X;RY{bWY=sx0f#GqR+h&1zu_b@eHrN7f<M;_$D%K~q6zyY^U;QQpe#0<Urw+&z{
zx8$Ej9k#EWVPbv{PM~qCWB6l}jIN!f>1zUC^V+Mgv$eUy-S-}0&%Qmp^wK3>J9CZ2
z%Uig`Dxd$%8LsSvoJ|xRrSWx+B*u|t%kjXeTln!G`#YRCdMnbiOC93fjY*c}oIn3M
zmSebl^)#RR{BLpR9ryF0KlL#V9zKC(TO}FkZHRWaLlj4h`AtONvG9(E7$yVG{>Fd9
zt5jGTNJNz2X&1|{AUvNuj*+|eU<k@8E3#yfVX!=(`s5^0PmqQYDpiPGNg$zB=?`mU
z*<Q{Ql1S6(&+zh6i~iJplG+$@(q?wH$)G!=v$MtNrB#krwm3GvNxfl%RkrGEYgaSR
zm=aH=3i44{0@Md>I_rx}&+Wr?t5m0E7~6Z4?dP9lynlg5?m30qs({u+gI2jPI03Jp
z+vd`1=UB;NLgem3T+U_J4T9d_uo?45hD&I`D8Xo$b1hk=*Xy%)?>wWh$Hj~1`Kv$j
z*BGCe#I`Md;7@-)zy2HlmB+sP<+owfzkXc2dWlZA!=Bm#cB97LZ+kbXD>(D3zk$N$
zT3<kz5y&dIG13pn;|OC}Z!lerRv2M{DKH3ZFQ7K%kn|<-aD+~BJl7!{MdXP>OId1N
z8lXr?qk`UGnwOWx*lF!SHD}3NhCO@78FYs9I&DrbEpsT?;l#u$&AJ1&#%P0Mmn>LA
zS&b^Plz45K+HD3q30Xe>G80qt1a6hF{Rf#nc{ht+_#`vk*LdXKyK$Q(sbkn1B8mbW
zpG%i7aPHh?Hj9+F6ifmPh59Ok@r``AE<pZ9eDlen4VVH3{gfzNVP~hq+}zCDF|2<e
z2EzfvQ3*Q&N?kq-MVeq4LEWz~Y`1Ax8LsORI~LM+(MgC6sP!GZsR;@xQE3Lkpp_=g
z3mn@*crKQ)2(-cSD(r+Q;}uOlnC9vBG}}`?$=Vfmdh5LFgEuoa8PM-0Y;MI&?we;}
zu0@`tY*q!AUwsJH6OLKeXw?G>R0$6gyRtG$k%i^v($Ohf=bok3YEi9EVpm(tp16bM
zt1r^Z+l-C7Ts(h)mF0Erxce~2ZaT#I^DA7r^bD_^S%$2`<6nA(_WFodl8m)nVXGW_
zckVWTKQ_(}|KMNc)QLN>0}rfHr$B3s?Rac#Y_hU)mCM)8@Y&D(21k$H%!mHe$GH9W
z`|yGq0t?&qSy-6oZ~pba#zPN2$lv<A{}ZQQIrEmr@U3C$bO92y$q_2W(>W>(!G?-u
zAkInh452gy65H{xT}fI*s5rx|d4!!1m1WqDgK#8Sn&3DN(hCTz3Z55}S{6;?a<)Ik
zP9-2{jL37z%<LX^+8yThryRfSFt_;gB&pBES6A3GH%+q^@ZvB=4)!zNTcujF5k^y}
z0=28ko@E7*5x7P&QU!yTP9yGlFItuK<+14*mg1aRf0_I4ypz-zJg>@Yubtz<<t2==
z8TCilA|>dZ=at0^TuwCu%9~F@d5!FHq#t_sBiOFX>6cz)V{?_&jb+w0R~hvCwA*c-
zdFmNJ;rkx7X290=CNI7GJon!75a0K`KghuY$B<H>W$D59$b%2?^FQ;`{KQZGEnYcu
zcDG#rKbEkuz@YLJV^m4));UsVxH>0~`q+WQ&GJ&yEYC580?$WC3$N)D_d={(VOKq(
zZWp6+Y^Ma&rg4ntc}TB9fX4CqWKvSqKCkvBi6=dz?BL2elhX@qY^^c9kaNq?!?;0>
zeowG?VU>M*CK<M4UhFBV#a_n7x&##$p$m$l*j0?zWNASnEF1%aG+}i5G(7knrE$Gu
zQyZIL<F%N{qRsvH-iomWT3ehwbDkGZU&O+p+wRfw6#2@lJhOg{#auAj9j}62)gRl!
zvK=1z_V*IT5idOZ6ziL7tgJ7xv9-={Fl2pwjk9OY;W&~?C7{;aW@BTWr=Nb3N8a^5
zzWXEJPbFx)En_$ujaXe@V|=WIu`H}=ojnh}lR<BQ6-IQ;G~xCY=4&<P_Uu8p2HEc*
zEP>LyUa`9$SXRksS4HW$SgC`7WYnV;1oQ_v6{lSOY|ACr1zBWR*_h?o?W5?U`<R@c
zqH}eFjiq(2t!~kp3OIV_2~>QH?e$HH?H4#uyvq2PgKb$Dqu=mwOp=r&Nf~TyFoVGI
zs`OrX0sX-HFm;7xyG)$8i}PRlBEi|m=#CzvGe~i@Kopw7Xx6-OUTqI3;x1d-fK*6T
zw$0ugZ_iy=-qj_mS<~JgkR%zS(TFgL_~-xff8{Uz$Y18b{)5!3HU9SB`kOa=Q<o$O
z7cO69-`+hXiOmmKxaWR`{XX&XDm^(**tx=#>$0$L5>ayzy$-x7AEek-;<hZ5C{S9V
ziWJAMVi2T*E*00KBQzD)?BcCl3Zuv(&DQ1=Us*X!e0V>5k1TNY$_6XTn=EhaP#^O-
zeA5AHHyxq3vCX;dSJ;=m$mE2?u`G-zTftIDk|ZHX61r<EENDoB8iUiPQTM$E)0{y_
zi`m;A;HBUFB*E)XF?{7YHoGy77D#2#Mb1`boR>QhD(<qWG{yp}0QEJ_A2%W&I_(ij
zoV_j2{0IJ$Hsnc!R<&~7F$!riTvIx72x+Lg5;4k%RgO@KLK|GKf{+@c5{f9rt@?y%
zX&GkOHkNIXW(l@!V|zY<Q=w8RNTk5Q;mT-|H9No^kBAb7xw$=bx<hIc9gdvZ&k1)L
zrR!XJZIzjcNyeL1&Lry;!~L{6OH}JN)~>{kRvJ}kvaDc~6*yXwn1bP}ufeGY%1W(c
zGqq<A=MqJ2=PGyK`Y?%>1VNowUp>p>@;Z42@hHLQgjRH!*RQ?Kg`}Vl6uZh4QVKGw
z>`L7A-~-f}4PHBYhPCyTa;@K3q2KS*>vehZ$uE=VDZcMgYX)p@Z*k_#X>PymZoc<>
zevsqGPL(=%ySK?bx8KIk{mj4PZ~wiY;>%xo8dZFA`lWA~C;_G7n#>fQkY&qZG{UR7
zq(xM=V}ylW_0dT}o`iUf8gZ86+5wIil$Kpdgk#y*wndUeW#7SZamc7Tf_PXZ96!uG
z51gbnF;2Xbv9YyDn8Zv?9_HoKXE|}|6nETvh@;2nIs4KjQeANG!$&y(NyA##rPUl!
zuexOkco$79v@R<NvSiw&-7YeYP^Q3E7T(kXouQ(B^<|pPA$fC-sx!^IZkb0JNhCZr
zcXl}Y<fpm1^g6G{nvp4ubc6)UwaM~=JSzah=N^BI)33fvzuzV=(p^Rn6h%RjB;*C;
zd5$py<Qg=Z6~-s}wA-7!^2+mk?4y5%@BF|=s8lOx*CR_uoVxiG|L*_zH~iRN`3cUx
zcJ8eV;U8NRMYgM_wwo_$j^{b#MTi(?xE`eG5EVuEj*C_G5rdLg<JPNWMUE4AIIc$=
zmKs;CZDZS#C>l|v+(cI$jVlcOehYc%IQQRmkZNO$(N@ID`Z`%I85=vmm%s8X@BP60
zxZ}Y+96PbVOV7TBS97@Kfx~QkDP^PmGPO#G?~76^O#!L|X-luvSkqZS(rzR346Tb2
zu01x-FjNF*A7`w305djAr8&j^TlWy<n$)VWxR`V0+0U`EvCPGU(mxf0B?Z#4NsEFa
z%@KLQuYdA)_`T2mE`vb_r3-{CaV>eClcpJ2ZYYWzOdrdUw8pA*$9rsTuk!qhPx0Xo
ze*EnjKUHvT=^D4)d<sjJtr)*P&cdCy<Hr%LBqR!JOw=4IwF+6%$M;+u$3?wKMD)4=
zx{v~GKopuZ%dkBcTLlDZis!gko<bBFZ3*&0#Mah6ULGD|`aSosczJ^-A3Iaxm0Snu
z`pg_y;NbCj`t2T9k|CECPm=XBZn9rza<YX~x@2A>$+Mgy%ZWN$AS7;3BkFcYSC^4v
zQ}{Aq?zVdvL^k`=n0+#4+|7`tK=?Mf(`309A!JHB$;mX>woR^bZXC9?yl&b68lo^^
zdAY>$syyf4|Ljlm<3IY>*t>5Zci(l_4SEf4dNs!InMXgvdmniZu2Uh2MpT+p?78c1
z(uFxDq+mQBV8#8?RlriXu7hLQn6D9*HSCfea*$e(CK1B6@ai?xXv9P#Ac;q0!w9$K
zlSdJ8cZ@S@3)Jtw1zXsB`SYhq&^V5yFc#JEarPda#j`Cg5%K)iO>D0@+*W*rg#{N&
zN_;QCbv(i_B#9Hmwb#kQ5U*0ls@B<h`Uw`MX7Rih2Os=4#MlAi-Xe#Z4t0Nk+_kK<
zyc%gzAW0Z%O|AvDWl@NNlId_mM&ONnFbd-M#<bg$G+}Xhk*SG^U3UiuTR7-Kp&b|9
zZ{svPQk_yH3APbfzKgVNhMODI8Z}07>6h%+>x7+*B#E)5Ro*jsiYf~HYK^K>U>i-5
z2qHPjo!@>dmAx$pJv!|jHa0dnSvkn&W|!_@nGb#J1DrTD&(ftGPCviNgYP-Q3!ikk
z+`deszQp8MwUk|wHc6hM^PGIprYQ1~RD&SAe2M14o3Lb+*<0?Rn^$ODJHuF%5Y-n5
z#-@4aO_Nw|l|g10<(ikDew6i<b6klP?OpM81W3mzWpauFZ494(;t7`4mKhE@gwYV&
zmb;sAMUtfCdFfE1wMMK$P<5G@?6SSR!Ambc%_9%LpC9=Czrf7w98M4r<wFi1IK<EY
zho9je{KJ35fBE%a|E64J-ZD`FO34E^AK}JZh$6#L29?&4mcZntFe#P<Daeu(cd}mU
zjtL2tU79;*iY!ZzQsOuc!(NXBg)35gp$W5qi&;wd>F1dD4{&5MXMb&m-+J^C7ti&O
ze#n=<^c1(<zRkY<HL^HkV>MxRVv@Gh^r9A{a0$<$G*mB2)r&lhu{;|sBqCSHG(`^w
zWO<A&0=)5Qq7eLSM0)vgdhL{cAt|&)YI$r=9Aq_B<lR*+B{4%vixac!UpGqf-*>GR
z7gtzWUM_W)W!dw;z78m5P)6<!RJr|$qmWKJVr;xZ6vh0)fBI>zT)D!Z|DnIc)XXfd
z5oDu~TW>zazxZeWoWJ;^e~Fc~)j!HG-muVGlVw?nW7+-M&j#3Oh{z&bW22mEIRR>2
zwxzAIvYBQn!PpowEkIfrX(22ZoyVkUf|M4HZP9J-5b1)Nt#FN@oz{7++a~(r4#%gD
zFjpHgD<*mLbLV;Kg>{6R@bYt?<sA>7q!N^Zmn&C>wCdyR2#fBpLKF?KEx+`1R06ai
zO+uW&FXeKw3>n9$Xb44$G#0_cEW=^J;i$uK=>mgpLDyL1+9GkPY>XdZV>BY`ZgM$G
ziLtmYHmS5`l)fonl;QmOi`R8L%UkFTLF()IvMYuOh(r^G1KK+w;}cclIAL?+?E%dj
zW4N|>t(1{KV$ev}!KpPFt5g`v8!Svt;pqs+&d9H|aV!f<iV{Dy%jDC>>}n0};?Hac
zsTFAy<JqOAWUKCTc72Ozlt9g<2y48sGe&q|A3N)-bfZ<i>(AUnwV|;spG0Ytta0tq
z5(~3a96P)~cPr)Gxr6u{S8=?Mdc`SC#ErFk|7T=_ZIl7Wa<FZS=;}r6V<&KJhvv*Y
z)yWx_x7N6MUzJLwiWHjIO!Crdm(=x{9CumU>b}v^y%BL;3h0e0Cm=*A#+@dLjn!=|
zyR=UH*<bhsKlu}Xhrp}dkk|6HvADcQp68{ll>uQ(WTi@LYKE3BsJB|!*$Ce)2)C}_
zI(9ibIu1rFP^N@bm@>Z<lrR|AC5=NIA#oj_iphvGMYo$^Rby0UbFnkUVr`m+F)S{<
z${mmF=hi#+kQX^gW+?0uQW01IhY!w?#1<FU`&74Q@v9q*H-k4?^13kPf<Nk#Wf6W5
z;MbZ&moK2=0fiM{**<$u-NL!wJjcw~7}MjENRgt&I2XE_<zd8u0|$8F<yXM)$C?jv
zHy>_97-LA%^bM$$0Ys7Fhnq;(Ac_>i8q#;m`OvZu(gLe=gDGm@)B=<?2+INKfGlN7
z3YFv7Hbs)KHR@9&BgX3ugtl3W0xrD%0`7}b+;e!GePb2QTp97JSI=Nq64K1@>Wja_
zJ03npx4lE2x~Nf1=xDmd6utiIRDFq33Y{B-7O36Bw{aYV){y1U>yqUW72#o3YanYB
zNr(Ez8M>#x%pd_}Y;tV+tp%3s261PV#nFgOcHtzHI($*~`^UW7?eW};FPGMkyX(C$
z<y^0|rqFVi?Y_I#$0>u}HudI!;V>i&f0M<fMgHOs|2U^^xt*ZVBJA~<o}A+U{g3|<
z>+2gl_QVta-7$Pi031u$NF*4I8+Q=HRm^aixW9r*b`V91<2n>YLEP?BWChalkdBKJ
z_@&{!QV7e&b<1{Jo@GR%A+Bd*vW&WC;}}V<B?=dE*x3wehdKBTciu9GjV3D;d1SG8
zVG~O=sQYt_x`Nf^A(7~iWDa?fljkW}nqYE`HU^z0c(pOK(&#K95Q01$po^SB8~mU`
zQ3!@%#(2$R|3Zbs^O9qGJr<^G^!wW^uV3Z$?JjK%QX0e`;`Z={Mzk>$N|5J}rVuBF
zB!xIFjjjt-PIB4^v=J0aki>?~^^mKVcUWHD;F)Ki<flLJZ#aM999|Gmtv9fw#RK=>
z&p-SJ|A1=Xe@zqUPd1EFkmX2Wmsm7i;Kw_d!7^sJLOfW3w1+IR(hev|NqR$yLSs2T
z(k}Bw2ti>=hPCVZfFMm`k~l1}b!kSU;$mq*sx6FkC}Pd##*j`_P#Lds+wp01uE>=k
z9l+JsH>fz11lBa89xN`kNu42aEXlH*yvTMvO-tDflV!NIW{Iy$A_8H^qX<=G6dEe^
zI$@+pvz)1Bz=6F1hvzI#?Dd!)udunj!qUbPuXl!w5J;(a>z7|=Qa1eld`S&SR(|A4
zpmfQYG;hk6C{(Pi4!L$^o3*vgw{093mlpw1PTYV6*tUmlY>H7xKI~xPA$d5!6=iEG
z9gfJOu*^rdJTqodOd2PoSw>tG6rvQ>O>>3kSEx7==~vkv#w2k}yB}~duCdr(q|;wz
zd}@-*m)GbIS250zxqVe0dG8V4_u<<ZpRcep=yK|gInt^9ERPP-?hJ|J1Y4GvR8ts|
zD5Mw-(W*dL78S2bws8%UWaOs6@oJ18JjixCW0Vw(RqNQ^Bp13aXFEBKxoK9H)-gp1
z<G4{Fgut<F9NWUNN-Mmwm0dcmWCg4)x7pg>=BXE+<iCCDw+X}H4SPLrj>}gsv$D2|
zW!p$;VPWw7DjEk-K-k-XG$b4Juq_)`+GJr&Ht2y;2**NtE_srX#R*xOk`y_q2HUGp
zXb`qT#kJToKgs3I0dW!&CyMof&vG<iY3(&^U()FoT)ldgf(~{N^UjA4@!k)e<hJ_{
zkcxy_Q*iv@o4JtP&b2j3zdythHbGEDXh|G}WW5gI>J<<cllzYluU(^9Ss}|KG!8dC
z@@*u_;o4?|?>aOp6GU>HmzE-|iE;Xa5jsw>Zt(1D?b=h>R(YVcc73gHR9wA$5z7Er
zC6gS56>XsgD-^>O;{FPjPH-#-q#*A0Dbf@}={Dtf9**ag71~`Uvg3Mar3gnu3>rsD
zJYjHzMxsiAf8jFdhOBMJ><mLr-MXJWV?J3@fRU^(4p_T3pb^Yr(`0)irI+>?SQWZm
zNa6%3z-Wz7f;^8gQH)Mwj1bfsW5g?$QE@_{GHkcX%#ow4ZbfCz&D1!u*Wt*#&9OZJ
zRV!z0?Fv`6wph<JNP+x1&Q7~#KWHV$3PF}ZlCZnhYtp<tlxux`tp}ycpX=<TT)o`p
z%B3~VoOy*${QLi%-~R1iC&>~TW8>JC&HU^f|Kgwh^KWYW-dg)rl`on!Mf7*DMA?F}
zJ%K2Oq+yOmV+ldh9pF_0YBSTN|0M<F8AX<%cPGZ&K!HIQ1yPb<3ztg8DXpTUgN_|0
z4>j0a>~OU^j%x1kj$5i6Ul6#*Px8bUFVP6biSq%DBhmH>R~D;Swm_*6qcnM*F&Yff
z+F%Jm;Q2IKQ%F^ig#&D@%63{_kO_@t*(f9F4pSB;CTK}Rq${k29%nXJp)t**(QNcX
zjNWZ4y@}KI#<lvIugo9%)z>cU=8Z?CDS^^gH_G;<v<ZJU<WGP6&vD1?cY@Fe1>gIz
z@8JuNewN?(t>32jA4JHmNK<4SqAIPj(v-x=K?mt;Y(RKmvM$LmenW*k-0o0m)bVRI
zN+TG7N>Wsom#jJjnK5VqS(c(zjAy$DHz+H5!a+w4Q~NCX+g&bhSGjq1z{4jUZa%Sq
zo^SHhV~b3V*U5B9n#O2*mF25dYPLcZahWe!N;nG3cD)qTf;zQo3sgbc@8U@y%h7p8
zZhP3SR~`VvltXiqNHSv6<Vx3YroBybdX`3>u{MY)`Tg*<km)yv`2&xC@PGfXpH|Ay
z>n3FB`nNWCZ;Hj`MY1BtvTTeh5W8(AqYa*PP)UIa9Gt38G1x`}MVMeZDTyVpCt3($
zk;e(rl!oDjDxgqEN$CrXA+UVvwSsP{7$;yWYcO$aj+y)Bse5zuciLRLy2OD)hd6h3
zo!9^C%N#qt%ArFuY^`pR#xBuT#CEz#%}mhiDjJmpDNU(!s7f5LR~W1~BzGNz<3YcJ
z2}5$nqrhVJ=y6{C-7m7aGh}XJhE!HKv$n>}@q?_bt`ZJH9KZ#Vx9uu_HKc^Vcko<y
z*OE|B6a~3b<i&LYk0#GyYqN)xf?xafUuNIJ-W%|0{$NB=#JTh5IJj>=mTkjs)iX01
z$M#T#AvcCfwFaIf-&sc`2{Ovbw|kgI1+U&fsT>`rAO*Q9FomMf8pm-lXnfyijDqtk
zLkeSIOu)!%aN9$NnclYtnYmoLbcMAwMQePFFFo;Po_K7NQ>P9iOiCImwpJr-ld&l^
zJN+@{nu~a@Ush2xWQC$>40?MDsSWPLI7OOLtS!QxgGkS1>d;}_<^=0o8!U_sY3`ll
z@=A}PV{v3*ikCj~Bp9d)h|JsHh!=okOM<||v7OSLN1GD5lV{|`jltD*HrB9<f<l5&
z0uzVGG=z$cYk_cb^r%gmMYxv0Bn87!L}Pje$M?y#!a|i!PK7F2ut>69D8?`zA&sOO
zR1mg>uq<3-VQZi2bjC)S&{@rS=>Bnz2L<`UEv!Zn*Iti676><GkgU)Rd&IFoTNy=C
zl=3htE8$G46i=>Kn>gB#8-<EfY$wB{8F`SQjYV~0lHQ;o)rRS@NtDhB)dXkOI$RxP
z?BBP*@b!>1$`Mkc{*bkv-}jy0KX-ra%Kx>$W!PoBCz_4*9(i6Mg(OXLHa9o<f$#sp
zlJJ@4?A!az-REzu{W|PIr9w6up!z$=mOu-G&>CHo0CU^2uss)FDRiNT^R(=93t5U1
z?lM0OnmA7x4a+K-BOO$dGB!0wW)z;4vTv+LWlsxBZZJL7;FeSSiAI|oz2zihduF-k
zo;$Ey3#AiYJA0nr`P9o0&ogFQIC+M&N-w)XKSXIw%dap#HHXt^mYQb6K0)B2lLVco
z6joMhq6v>)e@K!fRK{jm?>IbqZVhMO3`RPvEw7UmrYxnZ8)v?*wIPWWSFfy*Wtud}
z`H#QwA22`o!`y!B9T=mjR;&EvPyQs2e(5oGI^7%gTHZ>;y)K(CJjdR5w$PqO*6-qs
zx`?2P7UeZFMX3^RyB2oU!7B=iLJ?)@?&(mxkuSy=3N%rc5RQfjt+24sdB)`I7@01p
zxe<q(vpD-S8(yD-2c|i4c!s3E&P_L+pgA+iJKu3HmLo{BkmsI#iU0arPt&U0jGtw=
z6o6(pj0i_D#z4K&V0>mCQSs4PLNe%6tN5rSM&v1q5=)}AL${Zpr*fRiI2Sh@p18Qo
z*vb9$w|netb|@6PoBtaZN*ki+?P~*V?`$(14ygq}*>Xc6Y@5hZ%uY{`bio;~Ac_HA
zt%i!LRBVf=)4?{nWWA4%;a3Bqeji=r*p|R@tP&#_rv#NMLB*$8>0zXyNFihjXU=8o
z>JZ1(O#1~FcMSje7hghpr5EUzmoDSJ*JpBkmUfiVk9%a;gtlTh)TOhA$S8^eqYbVb
zPzx&LSp>3%tkw|e4yv<_34Ai)F}{C48i(E>Lt!C|Paar!tp>}_oyGyMBw82$oh$d&
zLkOtXs<c`yf*`<mf)cb|6m<Jt;v^zX(-QA#1aVZ*Y4;JP;MafSKfmp-=j+8wue`(u
z-uM1D#A&5vp&X0UmegnGQE7-jT_<a=;M8jv&&J9!vM@wNF)GdpS`896CyhdcDR3;i
zgzBUv97nhwjhe$)P-?klu}v^POFJ1MFLh`Y5r+bg-+N+(=bl<8Xxd~^kM7#*yyM|h
zG^!Pn<T8Qp5Zao48Za7WR4M|MtI}ppSp>B@gacF-<5ubzFQ8bxf{gCL2pi)E%-wXH
z*FU#KKP<4Mg*G;=ebcP2?vTeRo@Ei<pu@E7STvgr8jS{lU%|1;E?AnTw7VVR<i>33
z7SFv#=hf3RPu)#{A{`DW;t)}7lyHhXFTH352*<*!S7Bhtiku>&Jdl`O=mlj<H`kg`
z5|czBmH-NYEj<DR1vxX7KH1bMl3Ir-G8{WLM}5M=8Jp+G&BwTRYz*6Sk(MHihO~d@
zJfHr(C)hK8E9?|lLZY-{G#Zko1+K8D*IG2jCMi^k$ORgMgFy`kh)NYDT<TNPB$;G$
zXT)T^g|f$aaanVEeZ>CzZspnMULhJKWo7L~uJwQ^44w9fS6{tI7-ocFM3TgO|Mz`A
z6JwL#G#T<OZR$49zJ%9mki{daX^f;&{vkuDm8A@hDOLHciVty$ND5?NBV37H4QN_Z
zBw<7nj&NeTG<#NtC{0kArhRpVn@*h|us3O?TlnGxgRL#%(FS+ie~<wFKmos-ZhHVD
zHCdi>b@3vvy?&MD#dUUe`ixEQ#|%B(HAB@}!Wh`zZWG5TuIn%}HAA)DLO2dt8WVOp
z)M^!Unv|Q;EFmv)gmj2QjTSbR-{8#JGJ3K`bGpR~k3Ngecg+NE7$~+3185CtVpv|@
zAkTBMJmY77?i2jwAN>hVo;Zcpira6$oezKLL;T18^outPw*6MZL=hK$?~~LQ_7EIA
zLJ<$K^Atf)o=r>%G}pF`L&=7-s{ty`O1rMW0$bwNt27;(WE7HwBW$soFL_3)6nUZ;
zEUt6>_#tY}Dz&7CZ{5VXjSgBZbMM2)m^gV5;TpmyWbx{&oIQJ)jkRssogw>=o+8<{
z@H+)HZ-czBY42>2Wr}(wV0L;IuiC%}L6N2;gFa&wADzY&c}`)Hl7xdz7{v(NL)kS>
zuP#zQIEzAa?fm7^=4}^;bYnxH-o^!HFdVScX;a&+1Q`rSL#7IZ@o3f>47=O3oE&L6
zXkmzr!uBg@lVM4lEJ|=1rMi0D?~x=amJrysg_MF(zek}prWln@G@?eHD6T9m60LcR
z-Le<o(Hy^do<lv4#eRpvgZ8RnZEcIONe{Q)XZh7<P}wBP66A%8QY9HH&lQfeX*MSC
z9G5&AB85}h9|=Ley^5Her?4|>V`DU$E=gLTi;`$Dy)e&}#brimO4G84RKZ)>(*1re
z<kt_|wyD>eOi#_yYK_sTHNZfx*T=SPI=yzu7|ycN16LRH`y+(RdHVUM{wTxyM~bJP
zdzv&W7j<Kbauep*4AY27+rt=5Z)=-|1QS#lgaay-0F$J66_0$FQp71<y@D|X@h~OM
zGb|yoEDNOyy4&02!cegjNHeNVL~pgtViF_TA@8`Og;x!D=)pq_x7xJjh+&u$j$)RU
zRylHFmg#*FFMj1Eq}?L2TZ9p$Nk*}&A6FADwML74bDb<1AuR{x`l$8>MSBx0AC2J1
zy|;7m(I@Ek6uQWns0VCTYCL=T6@(@6DA~Q=>|M8O2J=n*yt6ApJ~um0qu!trR8fVZ
z-EM;rw7WOv5Wlgq!s-9=E8Nj&;LI(c^zKFuQ#KpS^_H-MjR4oeu2;};hDZx!V1aE>
z8E;}&e3DU!85vy5p^z4NnvyF+7RB^a!I8uJm<`UcwR4Kf<bJMQcnz(uaOYomA2q*<
z^aO)Jm#bH<aQ5tFHa9wSyCaU@c8X{Ntjxj=1|*3^x7$N2Ls0h_ADhAk@+3qW3*ov*
z6{5O3nARA%WoXSWpo~qYpJFT*B?B(EuQ7f606RO|bhkQKhO#JSZrrBsI>Wih1-)*}
z>*>Xk(33{&Y;E!rf8}prTkbdYOmAsZcj?j3GC8+^Tdxv0HU=BZ<XD!2<v6A74RROM
zDY1ROBuEsBxIZLUd1=yZSu|>;M_;b`7-bO-Q#?<yJ(BD#1k-iJo_K@p-jI_=Pw@QH
z7kKKk=c#<i!*V3AymX0Q{gp>?%p4W3j_Xx9b=w@{({)xa|2h*>nlMS&+38{-n3-%b
zH8n+HG#WvaC5T>|APC6fA)*lEB&amUb2N#TxVDeVUD`@Ab8<h=Kk-~?ZDv_m-z2^G
zTZ{6fql|`Bu(94Q_1{dv&-~m^^LPK+PcgGFi|2U!;Gg-k{Q7VFSB9g}pZLbkAm}b{
za_O_5;`oQY2d~z^w=6UcjwukXTe=2V0&QC*W=L9J#a-q(=_sTy3W>mRT|C)hI2fX_
z(YauhWcZd~r6)KjCDlgC%+?BtImyAfBb<Nz3Qv9RHST)HAv{0eQ=fc-$3FiWu9%_X
z*YUgt_rK!^i5xPz_J1+4-=jC^((MlLq|MyiIAg6z5>sL=vn;}mhe+Ecj|bR=MXnP}
zp5r+NjE!S^gsDpu_)N`B^5o~gf>El(9~EzZb>|Pe@^5FPY0BE#8v7UaeD#@y29=>>
zg()&BzJ(cvWZEEAX}s-?Rp_s;QmfWbxgswLoJxf>j1gLs=Ov~E!!B?-!_pa^)JOwW
znd0jSn8<kU!a9D%<^Bij+*HrG_1<GlojODsN%riYM5%;FAAN#r8%xYh1c+@Li;%p~
z^m=`C0YTGcyfuMPY1w6PD%1lHQAncoC8{UxCC_u*T8;7P8vT_Br3{^9fHywQ*4fwa
zj6_LAyh~{MM%UW0EqvES%CaJvWZ4_!maiAJYMrUcY3An^n4FlTQmLX<>2RWTL7r#C
zaf%d*qR0S^F@|0zMK-^=$^Pw#o%RmH;edLj0tk#S2%#}qj3Gvl6F3qR79>jJ7=f@Y
zf>x7JXNOv|PMBl}BXEK$X`E8z33;B9=OCrPu`SX##Ip>RR@hcSwJ<bmO?J9No_JL=
zHWl;m1AD0i8LhkS!keDJvun)E)ft6D&ive$>BgH(j|I?^SZ+$3z;GBMjiAvCsMp6S
z!aivdqP3vesG`ywy}gDrxu2}aX-rSiXxb=k$P2^J#3-xAaJz>uBq@rN(gorh!76;$
z!F5V8^EAszi~Ji_UZj+?nk{CgXIYrvL#x@sbA7U`R8UB>8vy_)LmR=)l_joz{?i<I
z|A+AF4Lm8)#>O%E?)+_`rNG#Bi6561b|t`)k|N7VhY8A*QZ}yZ;y1?V_WBrXRFV+p
z1)dNrcLfJ*i)KAxqP>Lnjxs-fh{f~Qc>2-TIC<9|DwQg~`YXT3i_a|MiD?4Af#)~4
z|J^sUovopkzRcKOpKfo5es4%%JIu|G(`<~Bn1Z~}XcH3TF<3T5GC&rRT&182T-$(l
za4eU0<WRI4*o_*mKKe@OTW(p%{Euh;))<Ty#F1fnX&r;6P-Tbi&wl)ezNu$<OPjj<
z$`a>4`$>*`$NR}0pFGtFcIm<v2+tx5V`^4_HkznEq`%pw*KU)=DMFVQ$g2kU^#+mB
zbb1k<E?p$-yr7dQI{g6?{xswMdDhN0*#F)S(>=MD^Jgz};OHgNRPwn`Ucya}U|Swa
z*&I4D$)WvwxcFP2WkNqot5s!baRa4tCMW95%`f2EHl2PKPYXz5+#*FN&1f{lLy+e=
zs>rAW0^@*dd$i+(i31B<zPiL_yGzTm8EQ><1Gi4s8Bf>$4P_}?Db4m)7sqk2u=x33
z`dR+^Py7w+y2Y)x-pa|7H}TA~&)u-s_05JxU~5VLrB}E*)?nX#_mJ5hMWT>8$CW<V
zU|d0wW>j1cr4&(jKyP!0ey2;GW(cjZ9J|byW{WV<bbAq2swgs0Qn3*xjKY{owaJuw
zn%7@yam#z&OLr7=`uSyM_c*LBMZEZU4>v!EWqD|0aqFG?sJj8}v!7!wIEUla=&bZX
zXr`wpn3<kM3K+!+)j~rUQWcsa%_*XY2D?%;y37|3fg@eEh9UDu4)DU6a}1-HmSwZ8
z6gS3x`^JVbhAUSu^1xkpmLkYfV98y13L!|N2-ipop(v6N3WZ&(AOaU5BuSRy)aw+5
zE?Hxih3)uQLXf91d6tr@2+OjmIDRRfECqEp$BQZ~+}0ur3+86WIe2^ubcF*ak5g^b
z$>W%3o_UtjFTKY4YKJ37Z=q65ps$_7uf)Wuq2C|kNWs*^7}da|AH~@HJ_Id-N(DNK
z{_+xj7Nac(;dxBVwTQR2z`#i7bccO<gAr1?<QQ@i%s1H7Ng?>oZ~p*)?gxL6ot-vk
z&%Vm1Kl>RjUAgpimEolj1ddCy(PUz5g1Ol_DwPUpnv&-kgF(L(&-H!6IAT{K3u8(<
zsp0U(E%09(!_knhJo9C~<K6GYNRYOT>v^PFQ#cN$-^Z%jWGGBglzz%mVl}Fmoo$jl
z#j95-v_VK0YzIe3jFjYQLK-I+qme>V_kvQ^$Jgxj`>gg3a?{;SRBkan)nID>7{#p3
z{7r|cR;t9K5syFd6`p_g71~=P4jnsz6E;!#8pbM!Q&o0tq@>=c;#wAAzfHQefoV-q
znV5#nWjbeGWctV*B?M#de&!EP)494%nkKl8pxf(^D}@n01sX*14bM|jNIv>qALU~o
z`xu*BTfBPqRX+9UPjhYQ+SmO(VZauWT2P}=YcMlCgYWyLaB@-5>)kl0O;>AFA!tAK
z3`!^tKKu?sw~C(>2$SRP`Z){VB8yY1o{uqxs5hXuu|u!jAxl!EF-UuNy>B!qv`wcM
zVdW*v!Vv{q!x3BU0sF=#+2cLK`Dbiyd*}O5((sipU8ON*=(Y<kyp-T%`;eAPUfA4n
z(;T&+%IdXqOgNXZohE}p2tqS9(PVsV3QHJ9gC3R92)ux5CBO_j3|5yIJ9;-oQBWBl
zV{F1l#SyA7WIAPG{{pXEzD$^<j9E6FLUUsp-DUit>%BIlY0k>>MrkLgo*+rTxmWR)
zHg!@fR$n-cS8s4+qTrdQPcyft#l+MkYO_PsAJT9n{UW2+?huY5Z0Qns9znyyXpPc}
zIE?AHyBOP|RrATygz0*XjZMX{$f@KJM&wleguTgW7C$dJbkDm`Omp#-j7#Ts$ipgQ
zbxBZjx#RxB)Q#lYfBO=(oyWQBc!MzQ5XS==jSBPg3skC2@-$%-$BY??3?tk;rMtC5
zTpgoX<m9_zn1K(*w{UEaE4>~`eS*c;uTqf~3dK<CUGc#i6W8hM*9RCtP`VuCxnXmC
zhahlx?&YWX!skEFcYff51eJjIyyrbU``q(?;zGp2fFmTNDOVqV8sBsAtc)k0ewic3
z=cqPv@;s*qW2#cq&NDhY9nv_(u|0g(rRw=_0PDlSknVN|T$@_pqAf$US!HQ6E49ng
z5L2(vswB*Io?-3tnp^LBkPB-GFFe-c+Qk7Pt1{lOX-ov%ara?FKVtRw9%U+ffm@G{
zv9Ys8QN&D)HOd2kS0RlfMrp=G0U{mZ7CAc`n>eEpK~azw8Cqpjd<cYt#$u&NXludt
z(gsyalA3~%`9^2>Z#}Lv;jT{(*X0P_j<~XTrPRfv^nJFZi_TIkD?{r6VhE<{m0EV5
zOOfZ8L|}x45H_~$A#8(6N|}UGS>GZxhE(ke62}QThi0`!PQg^L%3=FfD)s%GS-Zf+
z!4k)BndaDuTS?Q9D_1V@Ge7-HY%ci(avIN@;r4s?Gg%1MxJYZF!JxN^QUw!ZH6|u!
zkj9WCF{0ZgXw4y|MR#k9)loug*rlkBlW9S9VjM?m;w&YsH0fO2B2xkxcw|XRrp?!Y
zpMQV&e!%-a@Bu#j(eFk|i7ImB--Uns|GWHk?@NKe^E`adCkO(n)hdlfgD8rqR;x6d
zEw;C}%CQ!ujapvlk_D|x&)7e%P)c#`^>e)M;YUij8);!#9x6?+><p!cn1O(*hi&<2
z*C9``^1TbY%n8>k!ShL4Rxsb36GmZFL6H`iLf|(f1vv}VYa9tq;Wy`a_520;y)Gy3
zoag9?TS?N8<>hPq%)k3(c2;Xt>=|5ljuQ_Zq&CdiI(wB`qeXjX6@#KRUZ*uSg)oXZ
z9O1OvxIqicacOUEvbM58^L;&(SEG<V&Dm*sS62v=n6%Plu(&~vMEM?B6qB28z-9-6
zO2GTR^E>(Q$39wmVHY|1U;Qim+<*LqH*CVbHiQ7jb;>H2?^CH%Xfzs(Mk8vq8e?PQ
zH)w>RFa`^WNm3S{d=}RWIMh%){nShBJupFQVuC`oNd`mej-Z!kblV-GFvhlA0@tJB
z`53KGS`&>zdOIDAW7DYm6lqMo5wNyd5UBzu4$<RHnw6OO?z1d^q2Tzv4<qbpUVS#?
z^0_u5X)xZfsMKBVxbHAQZn*ZzFVG5~;r8R>48k3<JfhXCF*i4l=hw)SgkcymUTB1h
zaI~VcwM{<0LoLtC&Y=bMnnkWGgs@mmMvSU18`oB;N}C)-Xl^VT`WvnHS)y5A*~0g1
zo_yx9Z|eEp8v7N8G}2sp{5j^|b&!SGNiJVn<*64}nW}l*ynmdiNO3H~)Wih)4jsUe
z4q9oFI3mwWJct1DB%|N&(H{-49Gk6S!LwVETpDZ*Lgj$NOtXz>Kf&cM>`<LJ&Yn9?
zarD3e!Z^Va@cN6(3}0GixbrgO(Q_O>k|PC#qan4L#r(_+t@;FcE=ZCAX__Goq)Ei~
z<t5fmzs8Y0hsg4rG|z~`6h~-+z#`6bqDGyI=dX|^1y~Nozz9t#@bC>L)9Z(MQEqa+
z*=?4T;JOx$ZG+|wV#)PxuQ8MxyNxzM)#o=p`D@&9$DJJ5f0&2g@lO6X_g|8JlRNqw
z7Y0j$G7e!k;o@V@u<*|POpP@-dv=wL?iCJ9*ElxUBFsWO$6{fA4!>5#mNu#=h@uEx
z<ZtNuWl=)6*JYT5;Mgqn3ZB`7iK;~81yUCvZ1#+;BDaQI{p<$gdv0d><UvlI^cY1E
zCR1EGy+-@#^RV?C^ZGPLj+8DVX&TaKy3EhcQS+N5nITHM6nTI|6OTr$p1VYMagm!I
zcz2mEX+j#MxY7{#F5NhxXf!$Z+-n#mNTfqfN`#_R4|@Bj21^QDw-oj?yVhpW+X$e(
zwzNc+Wu=RUlo%m!)rfR!1v`s~3qc^qkUB*Pfr&GQojyUWiW>walYN)rs0)QwrSNj5
z6=4#SCouvGm1?#xEpq6@F>3xgvxC<N`v;jB+t2w|SNY;^zm84~wU)=}=hoT2vVdLG
zQLfFzo(hNN=Xm|MKg*<8p;Z@btPgNJn~8}@Jg-JtB&1n{*xA9|pVQtRaQ@6ijvqNn
zF&dJ{fHX~UJP+S8^s<P4wZ+b0fH4BARwar?<nVRxX(3>u+2Zay?%eh36-X)YJ*R}B
z=s$eE5JF&C63Z$zJ8j$E{hr0x*cjW}+qk>eKk$Rn6y6kXY~J%5q*?a$;>>HW5=9Y}
zN(Ct`gk@ukglKaKH!pL-qlqO7ln{_)47+{2S`DvK<^+W*S?y}qi&a2wG+~@jq%p!s
zip;Qm@hba|9-$trGB<jGVf#-pJAROh=T`arZ@fnM?wGNe3TI#3p|i9PTh%Fy&BTny
z;k|n}{a-%KxL%`KcUWH!u^o%n*f_3JA<I*WB4N06jq&;Ybi#sHUb)Ck$B$FAw@Jqx
zk~}9EpTxHe-FQe`Y0{5F&;qMkC5}Q0Qxe;y`Tcje*nMbLsvOyW04c9KtVrhPXUjJ8
z*KO;G-D+%!?UcLz&yM3TF)_jB*4rmt^tDl-$%O)3vQYE-m!4+g!Tl`EPjl_c2CFZw
z(sXU^I6B3s8{^uBnaN3lBMod}m+O5}R^lZBVaCbRjDCN>FdAanHtne3#T`jx3?^5Q
zCm?)gTiaOe$GH0F2F;1%%-wo3#}4dgl%)8M#o6bU=)ACu(>X(<|7A`cEtL}DD5O@C
z%uG$vtd5c8f;1bEr4b}4nJww8Y;onKvz(YaK%S*!svye+LB+vDf+Wkx8#NX$E|O&m
zAzTcCSY%*s(0VVxb;|p`bW=3DUAbacY}$}zinX;Jr1i~7=e$LmFhQ;jVi>b>d6{p!
z{}{DemH4?c%#8=k&bE-!!l(jim93N%g|G#7y@ClT_2863IZ&Nw5{%U;@`7N>=65bH
z(y}2d637dr=VFXuVJxC%pJr+25}n_Bfk;;{(m`s$T;3%xOUyO5nLS*^U<lI?-}mwC
z2`0uSQAI%*w;7H`<b{ncayr`s&b;(GcOO1TvpGo=#YA~RHXI=Hl!`CdDsnbZY%Oh)
z+YZ97k_>y4THfCP4<Q82dac}mWf^InQ+yNE6U!1b>NRS$8jkHCj3`L~X-1qTWO-Ir
zJB(pCOxfOQBMkiR@BI!x`se=&M-Lxna%z&*m9@7rtZ#pa5*wqHpfC_`_t{+C<en1;
zs5d)AFTBQ{={nOhEf8ZEt*~qhB?N^jkgkPeRVcy;trSWXXjjslo}j8)Xst-AhV3gW
zOiD<z1WSXK21mfb$v%z2^DHgBPUp%zv2l^og+eo5w5f58g^50s6ZMh;U!(+{k6&-l
zs<+6}m{HPaFpAJYl_Jepy|%&C%PZV-{0Ko%CyGMiG$I}Jv2~8;`)mlq@?c1Rt54y2
z7|SCL6Y}f*vab`C6g2A%lq$%woHWn>yBI7e<qjH+I@M|g+j0;F;y58nw%)d(TwPru
z8V>N>8Kf{sA#u|oLZ#S7AsvCq1{7(57T{DoY*mm(3Cfahh>MGgPknNnG)c&kgrM-q
zEC*?8!Z0Dn5T=IZcE(KysvIbmdF}j{Iq{Bfr&+J@+zXd^@u@2WRl!%ju#QYUe81rM
zZL{2bbPv}aeTK2#<J^42A&nAjJEvB!GTxei0>ZQ?D<oSx6sh9sN{7Z+6Jr#kP8aPq
zN{d6!rBH@W64QwUX+I*d1X&c9c-HFzlwUK+O9Yk$voq5y?Ah~1)if^(x}EMfR5Fv6
z#0bzvQJ8{U<zy;*V{=!jRB&7u%d*I_j65%X->LqOTgvF%h1c0!U**uD<48+lJ1#;i
zOp#!rk(Pmch*D+M$E$caMw3M`+Aed#cASzpCnib_!z7`i3o<9h(uQ~xlb906c40N*
zu9M>&5!ZP2{G%MY_dA$>_pLnfl?#03i<jAdxWZ$fSw?0qu9I`%$QbwBb(AY#dWlB&
zIc_@QP^1RO)&!nMv))2z2vY-E5nZ~-*j?{neWdZKHKfTHEG;1`6C_23H$6vzp*tEd
zstH7#6AMEc$AI2dOfZ<Q$q9i(NJ9|#)M~XijhVprymI6Jb;#;M3i2W&Nn_GHCCyWu
z+`)Ei9LJ$vtKYEkO^qQI8et?#Yji*1;;UEq;CoInG1ee|@f=mxVsf&MV~t~Wv2$9K
z>pix>saG%&=q#sD8tq7`;|(ffb+SBX+;(|%X^|+0Q5sVu30^(GXjm8zX@#d*ZeL{k
zGtUs2CIvPT7PHv^->fiQU1Q;B3E3D%Lu%C;7Miiv1WGC5q|0a&l0^l@sLxPVdHM8Z
zPVU=BUAiRUh{Wj9;>?lcmZF=d^xP`l^&N6)Bdb-i!C-g({`%{;0P2-0mR0KRrddw$
zO^|1#5Y+1x>h*eQm1-nKQIKURagvZ^Wp_gx7!DHF*LS|D@q0^~y5jmmV=UG-dt_OG
z?<#IusG#(a<;9H2xmhNrXDE~^ty3`Np+g&N*DICg4TQr18@o;jS}T$y=cY-UL0^%i
zIbqaiexgYjL9Pw8M$S$30ZH1W8zv+<NDZ|KL$&52O@l(`6eg!$Z=i=moNAqo?uhmE
zHrw5tXrv)<*%Qa4;WqPARaUoG=&f%vn<gY_gpLR0iQP6V<8p0zgHcjIV~k8IhPo)d
z*}hKUTnNGB#3VoSiGL5km%jKVKJ&THaN+W$Z%Bz2LQn~+jE_w)IW<M#2iU?UiXukg
zh^?JXL?TF%q-5?HhkiGr5_mlNg-`RL4}6%(**W&?S@;v*)JXx>uCAN_l$NY*bWum=
zsrztpUj-F*SXmh{v#^)?*chr%xb|+o0G81>u3zemVd%AY=<IBl_yVmEQO41hjZ+ws
zG$Tp|Oia%bhJwP<jE&_q8eQUKhi;sZ=K{+$)Ek0I;39N`tVl4zP^;C+mBO!9S={Wh
zzR{-L(?n50eWA*15|T%4=B8_0SiH<!XFwxO%F}4nBQFY~JZ1xrtJfAu3xnON6Ap)r
zb~ggT{NYVIAp}PcALA4M?x$H>TjkM5zrg3d@Htji*U;a9T`h#5TCFoNKE=eu1it4Z
zg{0RXu(tK~ZR!TY0o~Ol#+y@UX(6;ANiwv~z%EHprYP;qENNl84m2eyPSI(O9Rvsj
zIR;!CR|I&ji*LIuuW#dM$*5a^RT6A2EN^mj{s^AC&A#k8E<LJf+<Ytd+;e~^8qi%G
za`f;qf~vu{T?XqTmaaTb(D^JUW{21k;v%6I)EIA$VcQPf!4~T~Z8moj?Bpu9Jv5*;
z?vq{4dHKvaZhNGQH?;?Zz!{$=my$$T3`Zk$R$y5ko-NRpL>3bDMwWgx3<6t7=4NMU
zG#hW==kmOuKj?o$5WSQFrE-!qA&%qHP~G!Dl<m$EyIEq>t{>{&++zRZhFfG=#^S}-
zIkN9Ch4GNaBF+^>8bifHXBj3h33Rq)m14z>8kUoyk{rwTu!JPn1==nNQjX(Lb$!-1
z+n7QU4iu?2*p}qt>NY12G^kiZ4tg(f{z-#<^cEhv_aN)-P1Y~R9Nc>p>#$~evcbxQ
zZI&N>5xM>;PE4#*^IiI*F12bM-*E{1fMM8Uy}ilCR?0|Tqc!N_H*7M`;iWU@IClI#
znukv!gutDgMY|rOTrx-#bdn*3pz1jgfu(HJA5z89fQz76uQM?@xyyPjJ1?U!<QpCI
zgcKC2AWIX%C}c1gl+|*lbWJF3F!4$Vu`4}hkl1YPj2Mg(DvsvpY(SEXSY29XY-*gT
zxdpT_*wUhqWx2{2P2ji~RX#_WB&;nhVy<f~W<8EiO0I5LL{Ux{4j7*rhuk4G7Je<~
zNM*>rY(Otch%y5OR9l*Q-Nh0$3av=9n0lj57{%BvmtL5$w!Y40J0=<mQgWsTF@+m3
zHSRMA+jK8o<4~HCssfY6WN8ArV7Y+htsdP`jCQMFTMSiBE=td+uQysj5b(eJ)Blwt
zM~?8s<B#*XM?cT$GiONBZ;W;IJdb8`jOpoVDnSL;a!J#aQ8;3IXA7d;^<HU^HsSD_
zvxa+Xn>qj?8EMIAP_WYtnQ8hg?A=cqM@VT?q#0ROAY6+mNx_&m41J~Vmofrbh8h?g
zX^~|q(I_MeQ%(eMW#G{sC_H<I@rf~lU<^S@nx<$Ct|ORg`K3IqDG<h>l>{flwQZt7
z!Rkhv?M}=vHjKgooeL_iWM;x<|NdzzwqkRbuza;mJ&+t&*iV+{SV)x6B)K4yl8wS<
zxZEKUAhjlsLb6?2xA;0E?ASK%d*8S7z7Ko=*LC^8hd#)6Jo+7c@ArKl!_nyL4}@hi
zH95un`~tHxvs5Y-Z0Qh15o>E}Z&c|@qxk|MBypV69}IEiD$hLo6+ZakkFkH>KAw5{
zPnIkp5D4+rd>Lwsm7SQuD4|}d;_lr?mKI1$kVG*-qluP+G)_wA0^7!NY-Ghn2*~0D
zW57a^WhtXkNE)TwR(Dw1vFQwQ9D9?oiE(^CAWcg50uTfqxPDOb({#bEOdit9a47ZV
zwwJfr=_L%KK2fAFg`^fp_D%a7Iy{ACGSYs+^2II2C&p+@mWLLLQnM^orH$N1s0lX*
zB$kCpQ?e|117Q30p!L4%@$o<Xr+MJvcOa$YBj5Wme&OeTj-ULSf2)+v`8wfy9y3!j
z%rDF{JvB`b1V}6z-7YV@{KDHdmT{8s@|V8A{_z>)%mPK86DJ9p1npE1yB0>a?V$Cp
zE)}IUl<Rw#Let;eAsP(GqnM&7khTS`kCZN>uur||W3bri<rtMxL<8(vjmhc&X}-kD
zbBiSQ0`BZ#4$O^{Mj1ta$Z%zYR<uG(J<sty;0BT?jj7dZXf(nNID2J-D_1)V1}=eX
zamyYH&lDKpacyzH%}4jqtax;FLDjAhPE4b;O_aIxRERPLtqm$KFanMpIf>;7UORV&
zHxu9u2BQs)S`F7L*{H@CvMixD>b~g+_y;q=7(<$7j0OWboetG%71wofEN54jt~3sp
zQXnL|I(dI`9x_U6e&@G;le_2kP(OMKAuVDnARI>2t8O{|#$dZn*;+G{)hxSoMzDk=
z>J1olx@1vAk>%yPaa@8*gU$6-+=@-jb?J0-k|JZYy-lz%OC!iRDW75E+&R`>y9v8C
z!|ahm<h~{y#>8jVX>Be2-|hW(ux{CT-Ut5H+IjCUpL}l4eQ!r}BS$up2m%CyESj>d
z;!LSLDvn1plwBj0?QxYoV|%JRf2gL!v1MDbM3J^=QYL8<L?n?LKsP$~?b|n>dvgBb
z&S9<jW1nsWq7#Wr5a4;LzJG3=y5~FlU3<Onde{3r4}SSPC%fCsbpzro#Pi(x`)I|=
z_K+L5ci7nlog17uq47sU#*GFW13|}aGG}V~SxD2HCOvi<e`=16<qeAI3MB=G(i8_*
zr-2vn;rD%jKmGD&$g=P5&}b?pn2bim@ypb@`U<u))kRnWGI&KN<hxLhQ6vn81A^^s
zY}>BqE)Nd6mTld=_yYS?#Zj}Kqs<ibV}r{#w|Lj31s0E-Ac+TPWrNCzClPL^Nt)N)
za6=vx@{WU$wS8t<6xfzco+e~TLO97dWu=&Yz*Zjudx-0;Gk^3nf|4vNDRqTuN~T&i
zU18R`h`K~;14V)38<erd=Ej(v{g7d#2*ZLb1xp(AT9OM#0(!-S9lOC`q$zhsTsYRG
z$Pz5uMKxN8phdDhW}+b7i^)|@jHtqhQs|d8<A2}=E`?xWet}0n@?p-MJIA9R{RqGM
zfB6sirC<76Joe33ml_g6&}=rDot|T1Zh>~Yjbpo%Maj<2&dW|>Xr;-D93dom{>El!
z-&9dosoK6PQWbV)u(r8D$D7BpOdREcDRIq!;od&dka$6W<9k@HgE(Ne2@OW6@caNd
zH%p$z1PzaH5)uzWrUO9|3mQinM8gqQ(822jxY{R8lA2XhR)EGZ4Uh(DWwEw7;Lgep
z!*NPc)I$!ArI`r?3o`+Wb3Sg<rc4x-R+Rfgl&QG%mJ2i%rr8mimgy6!k}}t1Zj;?;
z%p{623=7XTiDQL6n5EO0@Lh^k5(!e#X*Br7pZ$4U*RJPcq(PR1h*zru>)7EtweHva
z%sh(=i?muTRHfM8-zQ08k|bm@86!lFrlP6}kQJj*h-1OmANvX){^(D0WO4C(*fc*1
z2oB}~lu1@;BBd}R#m3GiXBKBL9T#+l;o4{o`&-+XmWAI4a6AvwvEg97NTo5df}q*P
zG>v+_G(E!61ZS8qnh26q(r}wh`h8jp$MF3oZsn6EDMgkcON9^`!!gm)B+VqttNX01
zjtHZSqO1#1&(<uo1xIE*=H>!yS01n(q3rju8aDU7^&X6tM<xw?(`BzLDGE(!b=mLl
z5vMtp+r*Th1h1?Y^?mC_AP`csFf+p^e&*w~&LiO17I8Q}uo8Q<=L>=B(rtHGm|b9D
zp=L{#Wl355aOWwN)?E4OmpSixoc-j_Q!23B02$W$d<GSUX(23=vdj;R(_Qi^VZ6D^
z#?lJmaD*=F&b=@tin>rTBs4q=U6t&P;qtD*d<)VrBJdh0A?P*>I?fG_OzyC~`fcJ{
z9x|WcsF<#qF*{w-YkMdqP`V`W8Vn{0n_D?M{T+sb9M>>7cLYwI>d|mgWR+u<1$JQ2
zJ=()??ArR!cL`eaL`IX5vKgf*i2^MlO){(ldo0T^F|FG2@H;`#R2qaZc=?u13P{t0
zL4U+|8f$+xu0>grrYYmen0|jB-}f;L3)`}Yqe(6Qf~G2R%CbPnnjj&B_@R<w-VBy+
z-{#r>^|!e9-~M;;j-SGCd<-qnS|SlhX(CL6$`S|!u5D8RQGdkl@;W=4+f-SGAr9un
zOi5KL6pEH>0g@;!xG{3sjw9lvtUZ@3AKxiCJ+sGr{4F-eHj{5TsG`6uBBscg?j+1k
zyI6)#q4HV^q|)r|hOF;o?DsQN0kbX1*#`q=I}UodiPc`fur$5d9#+FeYA_rVPqZ;Q
zi)>%tAuA<=tN>{cM=3S}!!)oA7t=7wM8S7x&UJsbEcF4INgqs<YX<#3WnS?LmHGQn
zRTX)b6GkEZ{ysq?z;#`kjTS|oktXq7jjz^<LJO2Kh*VV{TDSKZ9dl{<O<b=*Q53jA
zmn2S!q8Qir@B$y(a}kySM!lzrqQo|Bdb6`+Nkovw1dW(zFr+LsYkLwy`jA&7Nrc^O
z<F-ArG$Tt>s=Podjp+z5Y!nXtVal!RJM8QxwXUQRm<BW)&55Hn$LAU}rvqe}lV>3q
zE@?91%-N%yxbF;`s-SIjNzIrtgSZsz85ZOH4U~c)2#81f7{Wm7s@?>?%T40tING+!
z```H<PMti7G$aBr&6?f+TEa(`G-!GOGgC7x&Mz`OJ&h28IF9Q^R~!;W6QU>q%6jhf
z4Ju^5siH2?nz*Vcw8FOpVHUBvz0Ubl=g9Jsmv^&ftBKk0F&qoSFb__cvK}O@DvBZ}
zN>Z{orz-10$qgD9!r&+{iE^6lHm#OV7DafS4$?I6J1tsDP^yytIAwKx!t&-e+dE^@
zOyC)UrXy(gB;A(9RLiH)s!z^kRkFRl#ds8PI*e&{Y>prA(CqZkjW)KdoiI%8(oap)
zOq;FUAyKK(mO-ASC?V+e=84M*;UxN=h-tu<1}9IP;BD`G`z!Mkm8y9D`R6HKhrAF3
z0iAZ6cDqfd-J#RzP*oLanqu4a$=`8Yh@yjGN(GfBO%#)H&aGRw8I6WaPtClUklqz&
zAXWNcy_8fM4AY=L8Q||}P98l`7w4ibhTGjPX5eDj-?d&ueE?8ZMV{B|B~NlxsnNp3
zYc(;1#W4k1C|aE^wqsHz33j7_X<IZr2eTApDrawyu(3X7Wpj(2y@;}q1SWJmL8n=-
zmu@qt*NbhEW;vUy8^pQfoPtiz!9Cg`=uVTm4J;|KEEAN=ZfYRyCL62EWLlsQ6j_E6
zlEwK&b_bi}S@CCHaMv`s|Ni?qf6s+i{w>WiUby_iYbI+U1kFa1cB@UR)uPks&}cNs
zvW)&+5F}j>6K?$J7cni9*>}AcV|tN%yo99%T9<%EOHGz%<bxp;CEIK3jP{2#8ZAzo
zK2;Nci~}(cRZ(U+Q5+K{A-FD~?Xt79jbB0($F%YUzu6+s70QOGUQVx@Q&k~_R%jtH
zw2MJSRe|LS;#~0J)ipLYbD|V%Q?uBEGbdUsOt;9&g7uYEI@fN}J+Z)%vkTa618I9u
zR8&G?w5HfJ9Wv8FWd)%It5S?8C^NWn?b%vO>ieUqq0|+ZU%J9S{~!Mu-ENmg(_?dE
zot5P^w0iX|UMWqSB#cG_Oxwh?OhDkdE@)MYj^`O=Sz?)H?P0B|A6Y>MuvwH`d+{3i
zZ~RZ(_jA99bK)pzbA~F*5mF#@iBcufG6?qv<dtH)-)C!OjjGVhEzA-4K`oUdG$@6x
zN|HDsOk(mP$7^+{_G1QPSRX}n8wPWWK9x48gv4)DoN2@;6;V~195AGfscWI{vMTXg
zF3W5C+*}&6yRT6QT5iSB1)FoHX7B=+{oP##pZ@|U|NZ}fv+uYMwRs1b<v6BIo~Ib5
zjd$^W*1rBUQ-+H+4OFR-RnA0}=y1gU{^{TSfmiGr&CXz-U;iingtxxqZM0fVsw(I9
zty`3@A)%l(q*=~zIKYsi*1FfwZYS8LB)>}*VJ~B=DmiZ|at&!!VF*D|XqGnDIJJ0;
zUb9V>7gSo~*bdF6kMx}ShQ%;yX0#9pr74sqFAAb0r${SQB`_?9=2Q>E$kDpQ^*q|W
zY05Oiatd(lS_-8RU?@#o=IrbwtS<LiSzn<)%#c_#Y{gs)x^080rc1lkAZYp+rerh<
z+1lK|Qctt+3%|t4^9wX}ifKDo!bDmUS2r=*i|l4wWTinr$q=%hT}5k<CMOn7u(p1O
z>bt_Lspn`t+vdW>d$8?VgHUTt9LHp@bM-0>1JCnbVY@e*O<dPyJRUQdOmG|*$98aC
zmn5r4uI`GXM5gOREv3M)Bn2hoWI}&9#1;m&Vd1u$h=UWkY1IFHndU^332B_5(Ns!d
zVNxZEa1^4e0z`o{1db2F6t$tdVWFak%802XD8pbkOnKpEpS6`fR-O{r2J>w}qp9fy
z2JJ=@&u`SjyM{qkR!pLZ{$LxCD(2=o_+>(V>Ku*Zr*RRCoq#gUC>@Usi<^mJs?lNZ
z_8KXYQVCR?ARUWGKkyM=y!sT^Ze9JJcBrP-6?)+Q`<b1edu4tiNfIu<c$w-|m~T+q
zs@S%T?KpUzhwuA%p7#pF%ynIC+pd{|2Qy|ufKmdj1W{PGYzKor?RMu!-l<a>QZ$KD
zq=0P*Ok0wwjD9$z7xn;!ZCQAY01N};;G?xBk2AvYm^{tu^r^wn21QsBhZ9gGhyq~<
zf(9T=r0rnYCdnkk(2`O@B?X(qgy*j9v9mG6Qw4!*aD4h;z4#{WW*gV_F|7KkATJ6A
zlQDz+9SkivGTX(~Il;vTnSc0gSi)o=EJT(eJ;_8_EEkfl?b9D6<ia2?Dy%9;+Ag2`
z_%HGYpZZ_e+1us&^{Q>d;GOSyCqW|s>J4Gg?{o9U&DTuWhGAmc4vy>K`~EBIrK&1i
z*ZrZ#b6Y87s#yQxmlzKBIQy=*Gt4zhPrk(bLIY!zG8&ALNse9B>_tEDS$yzbEW<)o
zC25vXqzNwzYAah}N{@D<{=OngxHN4MPAZ0Bj?>?v*$Hqxmn^L)ijqpz9amI=P$tq)
zNZX*o<=V=S>o*6asUmO{C+6YAY=hb99=7WcXE9qV>tHIp;TG;g599tjf1eZQ&#}6_
z&EDn)w&!970r~ValRUxFbywfk3WTCUz^jT-)O)~lS1$Aa_)q^+g1`q+Z}7u$$m{Xp
zin1h(B5c>jvTaP$pw(<+2w5i>YQn^5GO1lme#Ejx8Jdk!aeR4$7k>M<IPtNMGTZ2J
z=lWArMb7MOn{0c?{^kzTu17ZBLuk#h<40+<JBWj#D~TgiSpWhhHNvy;T@z1h1d0ca
z<ZKTOc84Y7c7!|HWp42(QB+Y;P?RZJOSCAEs-F1}wnkbGVQTRB6B}%86$k{~X364|
z#nIVmI^8x%$===$!`(j3P762cGxd%~nL2(7*DzS!-68MqGjrlJy>l1Xu?@Ot9Lr!3
zWmrN|35`^WR~@9+5M?R<^nd*Cnfl$|!Lm*AB4ZLpRMl%2<w|RkBxN`n)(r>Kq*5h;
z7my}*SDSFG6q(kfN>Ko&1T7_D5wW$ugJ-)ap)p+t%XJZktVgV*L|2L|P6)>ns<Na~
zwE(-RO(x@vJc)o3AuFUQ=yo)gEl|?Lk&q2X*nU7|2r{j?vlH^dwLK=gBLb-iJjuDE
z2F-?I%6Dis+t`kWv@8Uowo%*f6ZQwVp2f7I2^ucl_k5VyRs$WhC`wIH<_KZXw|%yx
z3k8g_l1f<QafYqIZ?yP@PyAc_Prv!=OrrR^v?lAa&a-SDc;J5cuAIBMwZ(Wme${6y
z0K<^4kY1dFy#L+uHO*$TzJ5KJpA}N@M%@YDbf>ODV@pxnfmMn)OWE#kvoN!Wr9c(+
z)nFBtjQ2x^VS;5Fbf>#CdL0a{8Eoycy1YW8*=1^R7NP2SIAsc?Z370Pf+UMb^N46r
z5GElntwdbiR!}NVOi50jY0_&qFwF`?Myol4Z3RffB+U{=gPnsx!-__uiPtgl8!fES
zh}n1j6i46r9`;sl@$A>Xgkyr+YLf|rRiid(t|~>Q!7Ot^T_O_2AAR=s$cp68`acSV
z6c!IX^l&}9K^>6o?Jcg|x=Q&fL+?TeEZfAkElk@)m;xl^WzBxG?V16s;owwMp9s<{
zMXUNSp_FDc8MCv!g<}}lrb(9PZ<tp7!-mq3D@l6b@oP#0OKB)eqNHGFu*+0?8eM9t
zyrdF2Wm+)YpD>A2T;HKP-K90%g`#3_ZIkWw9j0bxnOdB|&{ZwrVM(NA)l9@nkz^53
zIw3A|2IC1Y+zGk0Cvd1ZGiP#au0^NaK$BwG(DY|&F<jGPG8qyM`(#;)R+?_Nhh+$w
zogQ+p&*}I5O=j-B$ekCS<<66jF*P+s*F8ok47NNQFVD%9qEZ6Lh*d?A#Qdv&{m)6#
z@XwGQw9uH+<}D9D{7Q!#G`DWwWOZ}pRbwZmz_Cp%+rYF;gdtEU@*=M{2>XZcE)c1D
zTv{g?8;?IrK8|_IS(iLYdG@Itw#F4R4Tp!%&*1qj{Fx~%&!H5MQ6X&;H}G(ZMop9w
zkmm(^cb2Hik|c^r;(`ZSC3g~+&3%n)MC@&>Gkg3rUfai)K6##_mBzwA$&xe|+*<2%
z<=PD<lbn_>IeD~XVYW-R)5i9Elva$wF{5#xM$5zT9hyao3KPzJ_!Ib3Z3e5KWBv9u
zI;}R{xg*S<JjL#dmnpN7wv>#uLaI7UulG(Vtr<+l3@797+2~&nTG!8{EDJ14QdI?>
z@6ZTZ?CtMUsEWP8kSIy(*~K4iO$Wekr9cac-F;sAqc0HOdyI2!kCp4oEUoOYvKI1|
z3o}@rM{{--*L4t%U3Xd~n2t*jcu*Y3vq;Hs&}Xo@S<4zl2{I{o&x~Y!-(qvGK&Tyj
z*TL`2gLF|ks7V7Am_nc}&1j-{?)o}6ZtbG9V767U&@(tXKSQI@#Bdz)B4;?<BTUB3
z%ye-ZKGymw-M8P*!Xxh^jVEkh{$GjrSDBtW!tCM^x-)YmX@RQ+wtzx{S-n=7{~hFI
z$=2TPn_3e9Wu-}z6irnhEL1_e+rqgEhqv!WtTdUf9WxLbQ%XcpQ56;AXvE&6Ps?wk
zaz$n2s6talIeYtK#$iOK*QPt&q0zNbNyg^#D!ZF~7LOf8_5w`Q!ECLGv`naK)r>Ta
zh@u#gMeGhHJbf!<ZDi1pCFhS=EY5iJx=pftg6CN@{25HsMj95w(Js+sM4r_%*i*AJ
zNU3QzeDrRgdq4KebdE1_<=fw4^|>cFcH$I%dx|({vgbLpvz!btl%~=tu~KB?0srhb
z{yUN^`7?3TLStJl$4{O-_!oiJ6?c|zGm6It<FkLZLoW}ci=+T))~+m-t|-a^$8qYM
zk47G_-#y2rC<}@=Sk=9$qAphzCK6u=Y+29$2w5_WBU;IbA_-Yt-e74rVK|OC*7LdV
zfirY^Eh<$q+1z1oYa2sK&YnF@b7l%$i>fH84qj|^z>c*|gDeunam3A=o4m9y*bgOb
zyW-SI#jzs+z4ijlARvrJSau7g1ZkepXgK6~N*s*{{5G!J#B&@hsWIGuYBC~l4NTib
zTMgE3-K5)aIdS$rgk^*86NeG9s3?^}3XK*TtrT(c>J_*`uz2JMRaIh|CQ@jwUcJKl
z`qpdO#J)Ulp;XOkO40<+_i7{ivaB_14t};QYUBF@%iV*{rzVV3uHCqaAso(~Im@-1
zH^_@0p1J#*MWqFWuCNgJQea8&T$id!>4zf%&mj%R3<f#3cN5Y$;q*d>OZS~2XnGW7
z&i<Ws277zBj>Dz<F5vY#Xv?I^3#zh2D@9q=bb3<=l8K@ij<|7ildJoZabhwPq?|ee
zM;DrO+jBU!MUqUgEJ>*ps>o?LE=e4crBS_JoCcm_V@fc*2IY8w7uXo4OR6o_Z(hgN
zXP8?!g)%IRMnE_kV^o?{Q!~5{<_g1j^n)Le2E($Lo|!>a6^3B|if5mFn*ILpHIlZd
zB`|b-wO^DqYtnT+$|Aojthi8>WGI^2B1;r&&t7Lr%<=a7Pcs_CsLMC#cnaS)37Sm|
z%cfEa$Fix6dOA>53h6kQQr6e=#fZT7sfq&IlsI-y9)+Cf!R7#@5fHcyCc^=(={YbY
zZr~$?ATLT@ygB5`3+ohNN~;Mc&sua_g4ym-9KV6I943<i;iykpmNdM8cC(FT8g++d
zcNIgFNW(>#E=$+0;<*8DTR6e&(WAuAJd4gM0xjvk++}|g(joPz)SLVHqS7Q;Nmdk0
zqL6FXuQMEuNz;s-o%(?KqwdgsH*S>)rEXwqhyCp#H%1}7cOT{5?|Ojky^wtJ60WT9
z8dfb8Vi+i;a2>zaG(6}+7_NnF)bbf-<kAQlsH(uROj0)^h+;$qgOT9mLJLYwJRTDC
zrZ5c)({+$QoE1EId7ax=chGsxd`B`jC+W6arn?JRj)ycY2E!eq@d&Lnz4jE%17eEk
z@4@DEg!(Y8&Md-kdG3kFnO!`^@$>g__R=M8T)T?Epd%#fRYkpXzcIYoH34=1FD(?4
zNldeqvESckXLpzW{;TXOKO_tf2s%n@ECi0IFGxJsrzkVF2Yc9tBnd|h`USUlW3o8q
z^kTh-`%RCc$k@HT%3yB~&vm)~p-Xt(HrOU*UZBddCLtXV|11gd81m7Wn>Y8lx-S?f
z7V|If;fqZ=tvOuVBuQ;dOORIurG}>ClBEenp3v~yxNg0N8&YE!Ce_X=rYJD%0PH4P
zca{h&i%xHznWM+ZrKC)ATp(5oArvLLmOP4oaL(6gvZAPopH+ovNHm(Qtu3<b<)+FH
zj<?o|qAW;~gd|OflbBAsLs1r(rit(QSf=&LJ5s5aPnEB~e%@43r?keE27wTmqBcfu
z*&dc@vp-I_xfSu`?J<smkH7si_n)1{vK`VWWP589$FMkl@+6k)VA>YKu_&sFG>)mV
zyuQMg0@wFy6Lo2z{RUYxp}(@j@oAHja{<kcN2A%m^BSZ@f#2*N$O7c`NywB~j)PK?
zcB_ZyHflzeR1}6m5hrvV6IE%Hf;%@~V#l?4%TN6brI1L+W!N8M74<b|j>gi0<h3d@
z4M~yl|N5`~OTPBl*SU1>MNXVN#^*lsX@-NzYs5mMwW6vjq9|fAnb7HUNa6&EBu!JI
zD5|>@2SuJbc;Q~&;9pJ#|Kc}(lO)QSTbQTWY_i|)znSp<&>-r*F_qQ?(jpKFNj+QV
zy8*Uov9TX=Wo^uJ%Mmk<$<Mv#94C*ov226MXu#giHVr4>)Y&tbwvFjHU>S806NR;y
ztx}k#h3UEVOCl{qtHaoe39~Wh77UKfwg}obexr?J`($NC(3%1?Rh1#7iIM`xb&1B3
zPJ0T+_Uic<smX+(oJ_EM2VH9Nyx_)_=dnzKa}U0YTnL0^Gu|KLDZxn8f43<F>FbCM
z_a|fi!~gIf@b34%mx~uKu(&wK7ysl>h~w8P#_N~7EDORgBn(5^?KVjgqpNzZ?T3vF
z%|H~)V1w3(N^<+!D)*i4(QX=CI1*5#6LvQDF!uMDKYEhJREIpvDdP;oFtA+*>DXWh
zM5%CH4`X(oER9Iw7{f{Nd<WYpF%5<1yI5WWAt4<Mv0RT*DEeW-w_e(1X?e_SP;t+B
zm-*Qal}hjf2g~zO2yD|Ni6)dqMyoYN;CFB=6JhAu?rVD!l}%7t!OZL|d;L9b-oDOT
z9{m_I3yUBGrBc*-pZb8<qF(cG_7{c_L{Y-h?G?(ZB2BWtQbpYNW27rWQPR^604mko
zxU#~9#Wvlh!KLF(RF$y4x{2+XES@}z(`b+-_3*o8TUf4xw5@s$FE0p!#w*^hNgQD~
z8NTmf*$TIx<M=*K(4s60is2Y(Srh_xh9O^lZkwI$m}5P~nKK@<Gad3Wq}6O-I4+g0
z)d`}>2vBs|)A)W1+Z14k+EHk28C_*mnaj-V96Nj4EZw=u(TfkTaP$~gUwVm3fGPAV
z`3n6;dUbz(Xka+3hvti-s0)U_jAfzgE~GC^0-+JF?BPC+WwX(raAkGOvj=<lFTD3W
zCl=dSwj>-5*xlKp={7ih?kuKlVLA@NHpvdWy>(eTPFR+S<yk1LDGd|Z>M*tvqI}Hx
z1<A3w7D3yo_i)E2D^h}H@1UWPV@MMv!Et@!$(T-O3fuP3S|JQw7i{ARmTRFaD9VZ(
zFTH?enVfmx?UaUD@8SKI_i#mZKygc7C+F{vhy2<<_%+`Bp7(I^;&~Pp=lRl~evu?D
zUbWhP*-)S;zJCu-P^Ee$=a*GkYh4z_D^f;p1aGdWL*N?{PinNPuP0FoQ7WEa9`W^+
zn4W3!=*13arkvW_4@p^SjvhUM={r<fVHgIc=@3m~#{F%IaDrpH_?;GB(4^Fw!C*+7
zBq*ggwa_9QOqc{dt5fqd=Z?{|O#H0Rg5}coU4-diP}VXMd4UinuIb|hjat}K8jPcm
zN_z<3!3a7;gAtTDQ4&+Cvc4iUEu?9ZO(INP-?!8O)~}@}T|XbKdEv#&EG;eJx;D}f
zgkePS8qyR{YgwgfLKsd6!wHi~h?E9WLz2XlWr5apBCo{=zZ2lPxPz?F2x&4m*QML(
zyy+9+4->UE9UvruA#u>Cs-C4Zv`Lv&JaKcMr?zuWdKMqL)McSzBBXtws%>!M<Qb%E
z)rFg7VF-)KaLlB?gNkDuCm`syu>+@`gBy%UvkWaXmriuala#F2VtM)~txbJ(RFh%f
zzlx-E$LK~tI%I%=bV*7|$AnQMB_=K14MV!SK{_T~A`Q~b=#F>Kd*0vqJ^O3hxvuNn
z_u0Pp6`x8&4bhxbkmqcbs6^0VqnAL=fXu@#D#QH^uQ*rb^VoUG`x)zPOc!2;Oq<r%
zMOa|5IQP+Ev7=&4OEa?75o*0Dy~8W6>NnXDCD9oa&i2einobWma|L~0^E^ur4%Bor
z%Vxn$7`#bRed+rFst&!W6T;ekt3yQnI#H=w%>QjuqZeE}f|HYmRXt>PsK@TTZNa?Z
zYR~)BG}l+OpfrrUtAxp1x_2%E3FJv+#jjlbkdEiJ?~S7b1X1bAkUJu}i1~Tilgqaj
zS44q~NAc+IT#?vn|FuyqaNyiHo~f6IsYd|A0KCqVKW{t3Um_j9mzh6Q!VxvZ5&p~4
z-935b(D`rE6@KaZVH)K1C|Vg6RWl*BS!@zk5iHVk+4tXggu$x<x^bv0^_6g-ww*(w
zD54giLeG|H>ecpdFy*h;j(7w>+%E)3ox?#?wfD(N^gfRu1VmV$KG$(Q)EbLXpTn-*
z?-~E|j8I46qmG9=xGGnS`S);Sy(YN<F9lv#{oEej&Hfg_W8bDi(LWf(*t*CX6PO}!
zeS#9hRD$O;#kZ$2TH96=jX<A^)*;GZ`!S475|Y0tHdpuZTJnM?qUd<5q$qpMm-dz^
zM&eGY<2DLDH$zfMUmeU;DG>cSueI%fL6v{%A{1)sEg0!0TS-D!9o4G!**ZX=>?9H^
zJpAo2QAY>iG_T5vTWf#)8w{d~ucQ}Y8%4aYLY3I_%}jq~o?kbyvseIz%OD+yPaOTG
zA`H<)iQm!3RLzJf61pr(+CNxav4P<|%IG78G}PlIkdETtbwR*)HSI5yTlE~o5upMU
za*RrbjBi!bERm2je>=?rPmA-xdxQ=NTrDei>Sse231LZDU|7p6-|O<Y6y_M;^9&%8
z#nuc9Obj@+d@#1ZmW(jyVYA?)m_{Hb?FCOCUla&v%IM6rW*0hl<&(MDAS#L7G*e&K
z!t#oh&|M#M?Xme_i`?vqn?A}eu0vg?KD7c;q9mgZZ5@)5u_zzx)<qa-u4_1CCqNqw
z-%=};zsy+o7bdzIg2PLH{^pO)7E^kvj5+N#xxIhgPSZMs*`%xYbVsHSy*e1lC#Yrn
z6!Jj_tC!Eo7`<00u5q-YKx_$c$oKqDY(A2N^$0EN3p%|Lph?!qaxO&ipP&<U#?xOR
z?sP&OBD@L5-Kzraa5EAA&nfx$G=FAo!`ZOXf*7x^+%W@K4E%o{uWZT6!TvIY72HWV
zpir*9!SDG;MEfif@#Y*L=^P1tdiT7Q(ZYL|BbCRFKi5MStMf>$gsvD(%F^Uv&W&K_
zWV&`W^&1+f7e|V~FHsD%vRYrnBG0#TwPnfG6yP#|*JIjvRWmXRLLj97nyM2u)ha|W
z;?=BxezQ~Nyvak`uSc6_A+<XHAq%;(CcUS#ow?>aVu+WD&bBT41f>eR2LjI58rC)P
z+vB*;&Gi^pyt3#=9WGi1^M&1!yGMwWIdO%`O?*)5W(wmCD*QvKC4I_a-n$x;TS~4n
zz~i+s-o*K~$2LHiqPD3s5FyMKPH3$2agfT$(~&UYn-l{N8-7s1=;fk!W}3&I&?^Aq
zCrTc@y4F%i=%R!9d!Z?TI1FtKJ_=Rxm=FbaXY7oqBkCcBeh;pI!We#{e~n0&Gx~<Y
zNY;GWp@lPvaR*%$%?wC;CO@2Of$PmG&et{DP88`W`+EiU-{-G}F`YPRr_U!EWnorE
zU-Wy#{9vQSxm0ohv4b@tr(@W?{?i%RkDl1HG}N&y$?<^&4t-r7F~R=J%PfNq=|6Q9
zp~X}Lj5_o|m5wCHtL|?w^v=xk*b}5$t>X@78o=~fa^Da;kjBu~=cS3Fz>!B@Fjvau
zjPVeIs2&;3(H>~&s=M{mCl-sTNsqt%Y!H#fVNvg%++~3Sx10`dV^0kF-)Dfu$FbHf
zUiR96VUJ{|m@KomoS+zafaJlXz`_z_PfFsRHQSnxf>+oh@(mfn3}O+*G@O>5wYI+?
zG}e71@Ai4I8QSs*04U`FA#3>!Sp}}k6VJ}=QM{4jm<Mx&79m{x_pQ5Z*|o1|cDev@
z46ch`Q{dvEnHRb!)HU{e5m^!WzqS5T5ocF1Cppg>Rg<ad(w{{quQ-Nde8Vn_ItN@S
z&5=VV4b*cQ%}OnxO*uEW*N;R!W(8)NI4@o7x-DSnuiY6BIR5~r)~8bg;Dxlo(6LMt
z>dgn?#xOe^s}hVnd?=r}hvSa7iyr`<X!VqpY0}Ocnv-%`82;Xb&7OVmEnA|Z0cBr{
zHHgthdv0VtfRQ?aFuVTG4NN&{i%BE$=zVI4sxC>QRhXG{+hQMEhQjcR*)p#9S|q<Q
z_(+5p{d#7oHmwF_02Vw8W&o1qa4oQoLmCCgw`;tcAJLzwi&UcQ=9`n01u)}aI95Gp
zbel2R^QKq+kY6_=R*k|Tzs^&jsAWf|BNj2vn|&0p_6N>FtI553ko)`c)h|8OOc-x;
zk{yqw+7*#Zp0iAmaB^Cv{aVgy)$=4$F!e5>%-?UKN~c&dwL(xKQIBKn4up&M9#Xaq
z>u&^4(fZ?+7m_OA1ZhO5zAaW#EX3B<uKi_9@HPVyZQijHyD5Z&)3_JU%3cUE6~zq+
z${+cZ)FXhpT!NcOIo&XsECaJbn#MYohJw-?zPrNm>@#FPn~I`2G@&B*2B{Zgt1;7}
zLuXCnwHNcy?!m6^coyyOLEG2S{%m~Qj%qhONiU05+-O%H7KA67eRfw(3MUVoZx@jP
zAL{}d>86W9{QPwy9q{PV6Y9F-iM?-sw;ggy_ep&L^4Os$K*tQAotc87dy3lmqO6ov
z;NN&{c8-FyK4uT0za`d1dZ}5VoHD6sM`>St3#=)~r8TWld)p-x<M9Q23yX@7^Sz}v
zID=?sY~8~*NEZY2=^XUc*%&S~i0yBz%k&uU+zLOXh`41`?fEI=(1@9o%gps^w%^76
zW69tYzRpY}Z`GQN4g+>ZnX=)pX2Xf&EuRC!Z25-5o1e^HPZ_H}r!%C@YJau%j2trH
z0Z(4cIvSaAh4We-(ejt<Pv6sb1X6hogc=3^VM&`xOfu);ZmDVQu1RBolSVU~Eud8h
zF85Y*)q_lt79!e_A&h9=zUf&k{1E;~kLwoS4K&vA%sjzGG-6FkZOYO>3eMRkc6$_c
ze_VQldFx6>ec&;X;5W<K-h9p1Q>K62Pk~f9s32g8qfpS#c{{B_%;;H53!APwj{$&j
z{mofQXBXY)hleVCBkwzQpybb)thJIN*rtu&HU`2M;jP~aNhsj#CqyE0as>m;WyuKW
z?e7MbQY<5S9#mpVEGwRHrYJP~!_AKcs>AprH)sZEkqMdZW&RvW>xw@%yvajk3zWlu
z6@z)ia8zc>na6yxf3XdoXIkg^A9$e)>0i=x{Cj;E%Si1Udk!{0mPch<*HsC+{@qBY
zse8TDu=A+^GRAbRO&j!^hyA7pk|Dwx<8rXimF!UIa^*|A4W6d1l(=Uq8e8f8OmmN=
zpUfvLHXLUGpZj7cL;f!nccZ+Eq>7vJnXG>}U}NJH0}h$L0){idYsv%PUM)CD*}27@
zUs4_aZEwjPdr?Ay8gE3QLKs}i7eDvYczy&Ob+qeWL0N`TD>C%6P9!;sTPc?_u=Yc7
zYe9w_k+#M}T2>yfuyWiMTSYZW)s;AEWME5I`IN|eX0JU^CGZM0phXs~m!4HUyNKEq
zl=U9|2mP6)Ec=lwT7z!5^r{#eIHh@7lrr10)tDygjRy&&9`NWGs6yiJ{1D`yb8HN9
z9OrS;$FH8if*B=;G@%!0dWA=YQl!oFNS~Zo^%du;_k+$~PR2zc_Hbg*cQbM(MQMGV
zx#%cRTE1+wW`pO!@WdaQ`H+{wj>~jd7E`&_<7bR)`}z8IU{r&g1_t;~Hk=-TqDYIO
zKRHFHkkQw@-^r>f&lRSvhwL{DgsjpN1~7BkzIu}uDB`~MDk6<G6`*Xu1yB}I{F-_c
zoNRD6(U~mP)P>#f36FpIFfZhs>9V~fYcy+f*AYr_>{xx-XOjkT1KU{Xk$1>szJ1zl
zv0Be#9gNBRM;>n&Bi@aSv3umJD^ln2ej<dcXKxG^b*jQ_R8%i_)!g)JxYDz5U2Hz2
zC8uik94L@Jc;%F|xrNXpk1aKrQgz9Qi9Nyqb;e0d+QbI2x;k9F?5aJrrgfcx2j5bE
zw?<g~ZB*1rtZ5Lpv~m1B@_;=JffNI+w(eAi1U5!mv;+zl|1E6ltC1EJ2d1`l$d?f3
zF)>#NutvL0>UV20&jFt>@SdWwH~RV`=5oyucM<lT7l7>*4lS>sHE+FW8J%RyFDkkp
zi7mQqDjh7c+dmX?MRRkX+;l#1b(}f+W%nfeTp*srSDv4;8KPLV6Ej8J!qv+9)&5Pc
zA{ZyMf44QGJb0Ryam;^a4&nKFB^<7q5fp50*Bl6L6ELFH=itFsBHm=~y)+hTygJQz
z838Bb4Kjy-f)$}%CDW|CqizKLG8!DUY*HkGqC;<h^%Ft*YDT)8ob{fLI?h(_Hr?;B
zuc63Hx2LtHHPQIfeem5oiHL>fetW*Y696Y1t$>0QRiDBlpWCFRrngowl;gF{SoX%y
zj$|*y<cYF?UG#bv{dO<LHy~ryBwt&!Xgq7<_XE)Izz$pJ4x{2z6^P+-BMhq;@TTg8
z7MnRUdwsWQuB`whS8uw~$EPTFCUL2`rFTSHsm{uRN=Q506DXB}3W%MrWy(=7r#!-4
zwt%ypy(-`sL7|}a?~8r)hApv+L3ZU^95S=M)A?e}qo%28`0#;udnbBA!a1E$^1P|#
zcJVopUpCT>hhs~K(voB6Hc4U?SJ`iW-Pg4&b!~zBc^G<{ecJMT+Z0=YMtjp@vOc`(
z6Z!OWe2@qkBy;v!xU%zOJYw?bYR+)EkHF35Y1DOah3(5qZ$xfaEzTLRKod22D}fsS
zol#X~aQ+DU(E71a=lf_D_Sb2oo0Mq0i6zqPDWZlvtFa$NH(Th`LO<%HL?4;GQINAZ
z<CNPGl_lSJFTro4ZZ_^}mt6R#0n`A>Z75bhOv;sv_S(L}YPt%sw5=5@<CJ*#GPZiH
zd-YccNrVr@NY~=b?98Sz+2zf0m~?q{7@|({Rzd6d0uJ!nS-hQqONK=gI{4np_B2rc
zMU-1)1IOACzKQqmzI@a$Z+m%O$1foEa&IL_;N75r_RO+3Nbu8FsgnJ&EJUjJ+o@Vp
zLfONs&`8!v|C8Nojr`z`;RfyZ>YX>TD#L3%!QL24XlAY+NpL9suP-tE=huXtC!Ee^
zpI(FoNMXF!#Ndn;L1Kl2rkZ5n4W|!v^F7Hd$+~@c^s=y@wV51hNI2JeEL*M-<&3jR
zT^mV<9_aoynW)f+Mg%i3Ftg&3(igs2uV5Z)KbHYW`r-wSa?v2CN@qaI23~ta74-J}
z!ZQ3oAQm?<mjRC`8IFl~4i3HD9rp`(I^z(gfx@~+Ns=9xteq22;yN!KA!~;e96JA+
zGE^EC1zKMsmCwcTDsRe_mK(Ha{u~s15@bhtS-?R+&Ln9mAC|(a8y;857EAn{9#tpj
zH-G;}c)3GWt%wIP<8a#HaZQc*+V;LgWsc`&+<Cp2YOZtk(UxFq1#VWJJdZsUNQ?<2
zUEe9s8|{Qb+qo#ExQn=&G$8Wi4z5J{6Th^(W0HKNeTLf7im5USgm9#_I7ueRON^Y=
znXGT8JmC~E>HR5vfUVNKg=4c)2<Jtr|2n5%Ip57Qx*yo^dFp(&$?xSQAW`=@K3{tF
z$E)TMj+r27cgVP%*N1ssc@349+Zg+;YhIfGg8!B$qsj-%mIx*h+j9NL!!l$kKPq>&
zd}32&r8~Z+h|;R$KH?{j^4#g&NtNV_D%DJOSQ$tVDjLh|cH+$6_@tX|JHa}?3U35Z
zwsRMP|DM-52iS;dYeLSNr8+!=njROR-FqGqO3Zd|25*(|@Y}aE8XbV}(eVf6XiK73
z>=UwK2RHe0it!|zF+>lcV`u!&W2^Mb)Mtys&p4BJ5@s4F4=Ei=j_z^iJ7algfbnxt
z$Ud>U#5pnOi~0Ut`WdT(F-G@AFW%HI-vk#CmYlJ7fqE0^A2i5aRk{;d@sqo4jV7|v
zcz#G*i`C7A<derJ=C)Q3574v3?-8-u!t|IvleB7*M~9giV)#x(&7^aF{#E^JY^#AQ
zL_e&xA;9Zc8TfFFS6A04<KvzkoW}|pr|p$ECl!Jr1qB%|9VX;9(Kw>!sE$5Vjov>E
z$nOa(JdAC}COHDfW~%%m*s8t$6-f;FUK{nrQZfq`sloDR9AXb~kRXE6Y}W0pOTb-3
z?o>mSuZHfWna0=J(w|e|Y^u94a@ef&RyvPqna>8ETC&Cm>!=~)s~<RI71f}6|84?4
z*!r;M?Yup!?}--z0nO}evaa?>11IsP@ox%c3Q@BX>gsI@<FZ)NI0%#&jNb&l2+C!$
z@`ycSRbSmMYZ9J&15sM+nnJDn_S<=S3mHaYU$JC<J4SfDwAk%#yhKDc_st=#aYJL1
z*LL)Y$pRYh^!1zn@t;?<MJHtHapgyl+-Zg?tS>hMvz{jvj>xsQw5@^EceC`5&Puw(
zmvj%t4pnxhd|`UV^QItBFlU**_zO_FX7ZNZ=Y8q*`<DqwHZXi-WaJd@c};R%@=pGF
z()e*j5@V`C^s@Wf#TWRqwK4pI#%{o)loYz67Du{++@cP)9OZB|InK3{IIZI`d9M`h
zYF(UF*qKUtL`1!{&Cj)jWYj6XxW=$QFAa!^IV`|$H>cH?Hy7>Nu-5@cdY()p<cupT
zS!=s-8$p5KWW>_;$h`OIB6#YUA#Ya*rW2onDsA`oo?b(jPZFBw?t!}s`Xp#`I$)>L
zkiPrGJv>|KwAzbjjv>V?I4bvD0Q+vk?1%u%!#8iX%ZZ0&7|UU)*%%MY@XGk$&z1pi
zN!L#y$=ZB>?N>63sn&xcil+m<EU88vE#v&dL~8F`Z#KQh*1zF`G6`d(qqD^W_eAMl
zH*Um#<b52)(WacV_pntP{ceS?%Hr7(dho4E;gQ%u@pHkF*XiPtn6P|$NKgW1mNi+`
zwB=K`ht?*E$J1t$C`l?gGMAwZDIgTh*GzA@&8vS80p2{x-(EOABRH$!;x=USG~lU1
zCMBXFk$!g-gQyqus`+T2{{*{oiCq$<^gJ(p>Z$<ROyYc!Hgn=JAyZlcWjC|UEv)8z
zZJ1xe@mepS4=n%Z*#1n@swJUQKz<#H{0*lvk2OHSdm?NV6kr6wvExRIP?Q=Ot8FZ4
z16J3c3+>TQ+{*Qzom)Q{-A@&zzK7s;iuG4_2C{=63djGXDU)~Jlo{O~J^O3z;dTf(
zXK;PwD2Y#{fSX#xPach|jb*P-g-Q+x`)xqCQFfx~US&M&zp;$a`k05Q3SYmeu|V2`
z)O@o`2#FQUFaXRcEC&~y=HVyI8YHMRdG7YeH{XS3_s+$zM`b$alSc}2r-$fF)RY<Y
zYKRf*_YCcG8H}>%SRu)7N7e~WJh}YB^4_JP!eXjsioc0NY|O}M2t$>anDoaWvmUn_
zI}R=oNv2YceD=La577mrRQ@*N!lP+^ai!;!sy^L#tYfF>=1+>m=wygAGNe*B@3nrW
z22*^<Y5}icuOak~2>6`Dns$zZVFY`+pmn~>Li7o7PLRlxlI_(m+NV=7<$V*Xe7$Mx
zb4aN$oq5@y+SqjEd6l+P!VwG>%Lm<sulHv}J!Ww1kNIro86kJ~4po1ZR?<{KRyVpm
z+MoU{xS=|lt{dPc^N=4NX%6tVTca>zTa(3Tin1i_{7JMZr~#9ea86+M*8W?3cXWPh
z=R4+ivhnNQr+<3&hHr5OoNrs*kUhCdTjyhwwm4h)-OA}~M79(tj}AuD@OQ8iXs8}L
zb5PL3U}<a1UkeHdPaqqIeDev-%ldvGX2H!h#yh0Rrc`ejP)Z)fIQ8qBY{tGrr)ERD
zf)di__9_>D{R2v6?($b&SWqO~VtZ#B(If2Z%hr{ew$?NL#HoJyEV#GwJs?x07KF}>
zBbG}>oLcztCocgq$?j8ndxdH_N|f;Io0B$hqJ@SiX=^DDfnH;1Id*;YoFYGsw_i+E
zB^-EmiVVwfa6G%vGBGzt<eZ~rh|w<^db<`1x16r?t8j5T9rk%mF%_JNee}k%Dl;%@
z8fs~v*8Hjn1qT9Ef7AsbZ^^}>wTs98Z*RbPMBPXIXl|@nI*r3ic%p1O%u99WFYOe+
zZa7aQW-r3koG8BS>K@BfPxm6<f8rlqbr1s=F?hHV&jmD?bV+a0M(r}v<hfYG!g9Hg
zT-bfFZtlw=jH8kV^B)<CbT10#T0pvnHOi*{<zw?J3&sz~<s?v0(wU5sH+l|S-7GkG
z$IC5HxCMDSq~JDc4La}!KzG3pgBB)*84f%|BE(kF#5|E$BbqheHgFM7UFht9h$i6O
zm-Kf&6md9d*M<O0K(ZVok_0j~2(dw$gwH(yg+1h53#7ixaNLL9d^x01cxOwTEVTcs
zxxBcZ$w}P4cYW+C^_?+QwYa9wnNO(x=iG*m?5cIM{g*~OOpR7K4Rm3Q`OK*%dVTfS
zk#Y$S|5{6DKo*O*Iy&_E`6Eh3EWZ!NPfDIwp;77L10qBEsfL^WO-th!BMcSusMf2e
zQeW;Kp<|2?uy#^j0|OU{)fe8mCHjnyB2t0gtEU)Ls}EQ<2BM#Z(0^^j$&)KIyIeAl
zg}>2*sWvS$IT=NA=Hc`B1ojn6;`wRU3I%^ZM!|F88)<%$&o1J?QbN_Ro#G{0aF=3b
zo0ubirNRWWNa*K%TuV(;%V5GzOV5(O!S!r#DBw`CpYG|B(<KSUXldu)K36}kp3K|;
zcV5`tnB0@5Ab=sZSnD;GjvhUiWwQi{oXZ_U_e(Fv?PBe^_4JFC03(^0xgus>vGh0@
z?U!zcavv;8N5ixrKCa#C?veu#oXjq%BrF{g4C4duV?0(T9x+4u<GXxIjYMp;)v`b`
zIJCz1L6yX9S<6QBxf}1~;t7#+6@pRzjZWgXZT*YszIv=$P0i$gX}G6XxXWK+l>CLK
zrM+>J!LzwhieXj;iG|WhL=KM8$mfyVF9F`>i{IX-M%k;6Q;4&rqcZIRDfn-sMj3d0
z0m*#toj&e+RR<9F3dp|i51%YnhkAtSqDm_!a<B~rY_;QmmT})9eZHW5CepoM6aa@N
zBy~S!;n4hYf+aIz1a~9pD=x<!%0YwV<L+UoHyD~vU$ZAr19-40)CY`O5ro4Yy6oZc
zleRZ`x6%d{Z4yVy*Dl*Ai0xo?UtD8iTQGgSVqV&Wtx|oWp0!$>fi2k7F0j5+-Olf2
zeqZ|6?+51kq}o^--|JPdu!ex^{YeDShle1YqmxDt-yb~Kx9xDr(TXU+V$N^61KPbC
zw?*0A=isLsE$GO9WQe64<k(l_#|YX?J{;{Nq8-sPZ(eZDyx*GWQ^xZdZlG;f;)WTH
z){twLH;o?#NLi>i)lgY^gjSz!R145MfKS04kckla_`qalk~D%rMUPs^Q|%@TAR#u6
zvROR5)PfnMe_(EIdb1q;8XWtP9)_?mpuzChj^|Wmk9bw~*F2}MtM<@fOy>I{R_s)M
zE_Yl6&sN6lXD@5F7%2lfSm(g~)As6GJK~v6%uac!*)Qo$3XSx8+^4jAy}LWrLVLWZ
z4{!VFh9;C;&P!Ntsbx+cj;IoyN2K@ZuR?t;*77_DWAABhUnC>qikLZ6QjY{Z|MJ-;
zlZYGgadNnPHCgXHn4>h*kt$YFc}Z6tM;vBO$Z@F4#Z~i*q6Ww&vX-uLD_Y3RHWjIo
zFR<u#%FR>kqMFR|lcbTIZ3?g_AO+IaH}}FVv}3IIPktrSyRw3;d&BkHDBcCXm)AY0
zep}CX*`{eiVh0D7UV{%0ywNMqJSLC5lK)%*(-8Y09s+H>DW;{RHU7@q3>BjoD&<*_
zpct^RQMwmjrT=~|IUlm-=VCnRaK*F!gV841X{+C`p3`5Or{t@neuat2(?0xVS=m8z
z1<j7c4gJ(wtewd%8gVR%(A4;805QBmJd2u@9-v5RtVmL@%O+j_Uauq6VUD~p!Yr-0
zylVojhjf#VSzW8c*Y0=G7mT#mU>e`ecFv0BZM-ix3}Q%_zA??2y2mxNDEo$Vxr8+P
zc0(O<*hirTknCmx#9?M|ex7U8xkvwuYK7~M>vf3COQ;;601?#k2=>;M2@k#<4Q-Q<
znOz?n;%lj_HqOvx?BWHh=)^Mo=<9wl8Z)o_wC%c|-zDi4G~cu`d{U;q&dgjWRKrEq
zsS#huT!~5$&ZLoO6;jXiD9tOm_>q#ZfF%6PR897iB~fiuz_^LY+QD1Yx|Ef18As8c
znFIB6D(FqsoT_;j9X=O8G?+vDw@=4K%Q#Y=)<5p^u=4?P;L%sN%4qvB>RYUC?hhm-
zJp87m2|#bxyJb1zWRrmrSkW$luqDsjZ~55xZ+-rJlWj`G;@4`?>`|?o4Yi5VfX)&0
z`ZGyXNCcS5sNFs59_h)80TgO-Z$6vaq8+>@kY4waC?KwXmzCG#5^*wECQ$mSgbVn2
z!}AgtfuO`@R81$b)Q?>AJ}g9)j~=DDS>zx_wKB-gFU`07qkh{<hB)zA-P-J>-Rww)
zH^Cm!Rz05-2do~&C~E_m3nzxmYqQUzs(jL>ZPbvyD9-|*c)o>2Gw2>mg@uFx8@4eo
zK>kMhGy?A(A831t0*B)02%mj$;5ZfUolx4F*A}bWIn#LfTkML+^>T>od+J7XMj!Sc
z+Tfz=gb-hy01Ya)Hfj2LDN?ZyiaD>}K1TY1T>6LwfEk)?iAolRYDXJof<3XH5|R4L
zN+p*2`?=`ehxs~UgIP=ZS-Er}tYB_IVIjGPGB|8wLG4wC?7y}jej`IuSi+7=$b<Z_
z$2Ey_c{Ou74C|d1arRgUS`xR00lyl*TC>pkDa*axTcp2*g*G!)-j6pribPSIEOY{}
z4Gk+^dE-z`*-Rkcu&o`f7TG3Nc&X4iH}sXo!;Vz6b+M9<Jtrn9CDL%m>?OFLX{KD2
zn1sc6xXT$O{uO0(7_-2sNdE7LJ74t!Hw`AtNz)HD?$M_1$-g+`lKWh#@dej%mDbeX
zRh`_AFPXkAKPdtl;S4O`&VFf?Yn!%$T(U%ciRnD8@mdG=W=W2Do}(6mX^pLk*-DL0
zKMd#U`NfTz!rDi<La-bV(YI9Mv?YX@wO3r%fKa|M5zVdFZ;-3c)?Hhby7J!wmNV*O
z{*~p#l?U;dqa%j2rVvQd;Sd(A5`SR;?{%S8{1&Re4Iw-}m+i*}e%ZXG@H>(B{5a?|
zLj_|RY0?o`BySrVW&EDcs2z&5j6C6?X{;O9l8<dbuOn3R0^a&48x?wOvK*D<s><q_
za?fEyn(a0D6)AzRT}Ox5OOKkJodaJZ_PbtQ_#!R>xMFSkF(?nyS%iTTmCBo5(!&fE
zqM$K~#(FHOoLU$ef}#y0s87#tyT8$lWs4%Ldd_?*T|>^d8-v`v)#`}uh|E!CcUo_(
z95Zja8UxDDx_TkOM`V_9UA(F+EyG0b9Guk+CS#OY@_taPu5(>y4R9(~F6nPwJL2dM
zee>B&dBS^mV1*_>??9jMZr8O;=ykB1K;bmg-4y9p0@Fuc04}}Mv)irCd)_gl_oGuY
zc89A5(he0an=4+#BuXi%7%RRA&Wgs~nF@)7sF$N%-LAfOia)Ax6N^#s*qQy0c=ls>
z8yk0pR2bNunOqBgnv|xW!c8h>XWXdOvQp)-x`&3)77a55(s5d|NLP_Na46HN;*a!z
zWm<=tY;}>dWBCbv#fu)ndOp2XRQ?HtoEqb>dcN?vfou=GWu2v^)io9ETI#L*jUOA+
zILG49cs{`SNEVJCVmsB}H?s4w>{<9&5*qz{BY7_fYj;Zn#4hB4M|emwbM{ntmtno7
z#6>Y~H%TUs8TEyRob8^$m&2b^tIQX^5EZ;ltUOUozyk_|Zm_|Y7g!fnKyFlQW<>o$
zg=Tr!UwMsX*2TE*1Ht7-;05CpL*;-<W>5-A-oBXKR@KzE+X&NgC4mM+0li=n;U<X7
zPCJ?TF=wDJZ_Bbdd2^mDxy1<-7hJ+_v{)fF)XOVN$Pgay$y2zz$LnznwCa_!LY=DP
zu|%P#d&|&qJ0+2AOgY)Ulb7lIxmm&!)~;f{iK;H~MC=G%%>&*ZIU*zeM*j0?B_LdP
zQm7ONUpSUe#*aO#a}<xhknU)HU{H&j%*?X3R~EVtFWU%0uZveC%&qSpyP`47!<^Oi
zP}H1LI}7aVlY$<zGM7S~ZB0LuWD?LM7av6qDe+J$Taz-ciLk0-+tY(YNu1-)|3I2N
zH^D`q_c%bftPV-S+6$e0ZACeW=v47<na6tx`O<-;7_)NwXa<{>`rQceju9=31ZBs2
z*~2pRI^uV{9MqN~n;9nFktR0<(1`zC)`@wA<N598eO$v}imW4y>cOYf=m~4wUhmhk
zXZEdJ?ogWMbvB?z({vUbZMI2taphx)WW`r(6lx^baeBN$GfU==d_(-{tEcbaDJ?BB
zuTg*NLS8KLx9DTHhlN?Jaht3TMY;VBK)jF#e-1|~)7mme-eO_r&S1cuuvUy+kCxeT
z<42u*s}T^A>DoNrgs?wvn+?q$<$q{@zK8ahSM^HX$afA!RfrK|U>{BL?xbO<O}iBB
zHWc!{ZN7d0bUt`UhAmz+KlZ#S?{G>>fb})JPAa`Vh3%Z|%Ba>W+*}P{xx`MVIV4!e
z7IG)WYlrkWt6=wb)q&-l!`nKQ9Sa8=mVFl*^Vv99_Ti4SP#M|fVUIhcPPiDDx4h5$
zd?M3#@4H^6jLnxYShzlCq$aa%J8lX1>fW0QPho(~m3W$8<E9iE7y%`Q!$h+i8koNr
z9i0b91tgUY-g?vidqj9ic0NfySah6hrYI)%bmedrrJkuxxI|AKj7@;iZ8<vbX5a)*
zu^!KV78M&kyw<0oeQnn1#QM?>29{nb9IxH^O@)W_%Zwtw5GiZtCo<{u?_3XLu8_Mh
zsW|76i}ACsW5Oc|RT3Dr#Los_be8`>B{rb>!S^(SiP+OBYT4=0r+MC<#q<e0^(33E
zE72aDI^#Im$r!we7;)BCs%h(lul(inN}oDXfgBn;ZfrX#G&jmf*8aXStxO__%`r|`
z%QxrGJn7FMnB546&$2E_uCBp0&POge9hQFeXxUkGH0$$KTYDZDu67rkligH3okGVF
zgiU_FPyxOl;1ov?b6t~9YiZR}(Rfju?x{aw9*K^YKv*$m|9U>~XRjp`wK9Q}tjIUL
zR6q+Db>?>QHH^NQ>@hvxe`#LS1xPL}Z2TSrM)6383KA6kkjoT!!&fx;wJ1uB^|caF
znZb0$$47-q^JBS@dmSlFtJEH***}~&OGw9~wR8$Evk4uR<TPFmlD52XT`2m|`->n{
zD@8!jTn~c^;evehzcCe}_oQCBrnde`hKU?3h5z1++n@I@5=orCg~lXBEj}Dn`G!A<
z^pMfiMbT{g4LCm5JtvDqcr7ljMs0c8SC&`%owGF3V@3ohr6gJn*7Sqy)vJXGT{FO~
zx>82EDnsE7W_>g8syZidM%L;rrA9pR=pujK1Qzg&hXS2P+<I($rGYHC7uQRblCa0G
z_4ORE76ZFAg{V0R8SWIjSDM$a1}GsA_@3#{hyanQ>Uu>ZYu6ZB`LQvczp+WojmfDc
zQW{ZdyYcG2c7bpY4!^zUi!$Kp9}6GQ5t^CiL~KCIHt}qB=b&SxAN540I1~?Lz_&z8
z7z0-sidT%&DtX0>eJsi!0oF|w&HR$NA(u{=gc_&mW9I#P19}JGSLdqZdd5j$qsL~f
zeRyZWv5&syFyLe19Z_*aeb+{1>8w)3KU#v<5tHF1=V}x9=-=B%%IV5TDTfLhRFi}s
zM0uHWt@k@CI@xqRLcrWKCO?I@oF2DHV8+y^MH$R(axv)RtK)IEs7S!!=sCkKLwWOi
znDSz3;Q>va*}I7&wMmD&M603B+bnf0eHrc*D9jd;VAJtLPAm48MTIaC6raTF`I3aB
zcY>+j_o3vwifuiLw^hF*7kDD`U}||$)~imTE755{bGL=`K7(4DudqNM4Xwb6Yv#N#
z`(nBKU_ia|WqD<#pGVENn_J#{<lBxo9@>Ei)?Ensd8c1?Z|d{qJz(IWYIWz*oV(ud
zvDTQqe_#M@)S8Ii5Bpu(bGo$Ms3o77voitvY9_qfYy=8){!dSl$XgT1|8%5Xu^n0c
zq~Y;H2C$AFD$+-f>-?j0p*e#z>Hj|b@Ao&#J!b#i=9K7rZ|?E`YmaQg`XW?z`9FRC
zZy5C#xDl=YHDvY+0IA!5jVSq2gmeG@H>dY|b@Gf}E8}~2&oGJq0{K#uRr^>eZ5r@D
D5rYL?

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_022.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_022.png
new file mode 100644
index 0000000000000000000000000000000000000000..50fe7411c911745aab05d496686ed02d98e06e3d
GIT binary patch
literal 117059
zcmb@tWl$Vl^zIu-f+q|TEV#QngS*4Pi~FF#-7O66Ft}R?!GpWILl_9I0|a-6!@usi
zx9XnH=fkwk-rZGQd#}Bo-+DSyO-1$-8WGyNcke#Q%SnOWy+d?=I}S!cdfWc6RdRXT
z3Ajn?xM?_Ax_O$qSiDm<b#t<FbhEQDBloayakX)D;9})t<zgncc5`!b6=Y+x|Nk1W
zI=Wb~O^QkGzug4YNlw@G-8)Rv|M?;O6fL!Rhw$#5yp*`6SN74f71TmIbshO-QeXl2
zYUGtBl9=%hyYU_FXuevf?&`mPv~6Wr^6K2SvHdiWRFlZD$^4f#sr8l<7z`lnfz3<`
z2BL<{Z;e*I`gm(pi6T@~Tz>tR@K)DJKmOZ}2qo%P1jG{BHrMLWkDBC^7oEO0_l~#g
zzUa1<yPYd(k1ob1WBx5{x4WHB`Yadm4+#J7>Hq$3DD6>Y_5W$`zps$JZ#n(5`v2?6
zk??=55#Y4XFPS4OHulr6c}9r{vMa9DG*!8ZhASW8)U>q3i=z9}+o`8@m6kww;Kj~M
z#_LG(qty$Ys>o4BAbqI=7#vb08y{eX>1q73kfPt@;Jh=O7=Oj}KYbkj&pt0TT(4>>
zh7EcjH@Y5|1H)fWy6zbqR@#xU71n^UKEaMJJ97r@Zjn;~a4upYZ(^D+Ff?XFh}cfx
zyJ9GEAVkb8fuKTVgfq4QkAfM51{CeSm4IFVVgilGOZxSqIq*uTb5Dz)5(?s(mgY85
z4L8%bgzq3~^aoV_m<LQyC<=udUQB~^<&2(cfEG$H$KQfH<Y;Mfr)Yj=2?Z!LNDSp5
zSC$c~DaSDKJBB)LM$olu*_s^>%p*_ie7Q8%{=77mzilW()#EV6u4Hw$N%pA7Od=Es
zI63_-*ax37jTuaj;Z}?&Oy{35zpt|ePs@N~O>w24a$cuIv$NM_6VIofE?>{d&M!qT
zAD;#t{o(`Kqv*z<eHOSBFum{J$<O>AI!N<g#+?H$0x!S(AaaL-P$(jcaBamH*<STt
zyep*XBcW)j-d>}wdq)h#WYfoSclWQmd-r8|&v|mQ5TAx|Yi>#vo&vgas0o3sgerNl
z?=50Y5a7G!_F5!|9D-@j-cBUWpI;guOoMj^2<_lAt1DoqZQ-@G^}TRBi2!qk{0-e-
zDM$?mNY1I8!a}l~|G4(uGmLd8ndf!78b+6P73~~+aDPePs48F9Ej(6K+n?`N2a<g8
zc^Y+mBI+xEW9t2-{K(8aet9Mt?*tF#8dsW{KE3|Cw6--k@pYxuJzA?=I*B%2a?BV4
zsHk8*AK1$+S~#4xX(E)tN@F!FSSAao;Cg+1eh%zje{6fr_GE33O%RcG%q^2iWD9sE
zcI<Eub_}=-*A-}tfRe&m+w8HDh2ASo>M3~A6}#WA?BnLx#BiW+bao+hYCDPc5wNXT
z5g#_3-a|qGQ<oRMB_p%cufk?7Q<Wbns6k%z>yk(+r^Ns{+giMO*!}I2e7|jFZaZ>m
z-gcX1w;n{3o(3Tlgjhn|38FtB=T_E8;-?(rbW1g@moZ@<OrpjT%3AQc+<ndPY94i>
zUtF4p%)9z+X9Wg-1EXcVcF-!;-q-2cveQz?YZTJc%74;*IT|FY=g3<WeieG1dgXgv
z?|Q9UY4?9$oWDN-Do66$9ugTyDcAOfA6W7G*nKR+g^fxsPEy!;G_w{D*RE1l_GCWg
zm|!(~p7@H4qSN`f-0clnl|fxjbr>MH=DISO%P0)gsckieUi7LI{-K&^gNqIrGPdIs
zc637?zQ~Q+?EQk-WTKcYDhf0tD)ObOx~YlO99p*ZyaeSv1Tizz^5e%}tDQG1jirVj
zV<;CL^}`Nujb9T*hk8eytz?^VmMb|3vJ!uws%IxqXO0t_v~dvGG8fTroxX&`d%|y%
zG)h&@=U$J<E{`2M`U4-E0-~0M7S1(;3o}hkYsWF4_i+rGog%9PuE)mN#;G86*C@fm
zr?bNL9k}$i0GNr}cbKKwVM-Fcqut^BquukJk#DYq_nH*hEVpzi1I%Ur@hDhi_#GjA
z@t$2r-#e?rx}zFHPoD~RkNrK<3M?>l!t6ak!sDlw%3J83ol0gB?RhYnDJOe=ENvXJ
zf%mPxLBK8kjC}YeipLGhpipo8OGa1!pB1w>FjDz71%^|jh`fei&ga+Nz~`oy>HmpW
zt9J>hsc?R0v(Sd?L+c`j6c*WdZ2@=7U~uWb;@U!^%FTNb0X(ydJThb0d6(irOie#|
z#h^K=-r>`!t4hu({#nPF|DYMhk*8=;Eg?t`h$&WB96pi5<4I2%Y}}3c9<mouii>DU
z%x|vC>&!nG7^9BUM^Pd@EF?Om!%^{(0XNqrj1D8asLVY~g)4?oC?<?q@hw2js7>1&
z_GtT}eC*_eA}W?QO{I)Ey8b%~@yb1gPG+Lg3)JwL3sN!AVicuQ%MbsKQL>$7?ojLz
z^@;L-;u3vjGlG}kS5rVFA`)USY~37oaQDb^td3=(f0F%fY#pn~t8LfmzeTJcN$`@(
z_zHW<iOm}`<i(u2Oq?~;Ie0YjUu(W2?Ye&KlFX(BK5gg<9^<X9p`d;kr5amr<Hb|;
z;otwrBwgiuV)sF<3QxfKiC}(|d&>VM`fb{~ZXP$axIgnARk*aY41o|jow<w6apG9W
zivAL*v3~G!mKo6QkCX}(`La%h@*XX>TPBM6ygAXr&(=Qvc-c@bDw?mV&f^eH_HP;i
zVmjRWgiRNpx3hDe{6So71G#?LzIgaOdU?(oq4U{Eaq+^j>%k~ZDjxj&ef-4NEDV{M
zArV&!W*NoeD-LVu=vfFfxfDMmhfR&q_m8bR^n$i$c`-xejwJk~<z48+srNsOxy_9U
zbj)|p)54;}ABbQ4@j^iPP)sfI5<xfo)z{k&0NWW>ZuUla`;ahMH0)r9RdEk<<=i{F
z4n-$iFLw_SYyI*JI3++Vk5Ghp(gK^zhy=%fO{)t2+n>ClDB1FExox4v{gP9a#oy*4
z2K-*MWlt;l2ZJ+(C2nbPS>I=SZX1X6&tFReSGkOJGItlxEgbCEq?TjMMqGAuVv|ev
z1bhl*h!npy5RG2edYZN+faB>s;V0cUwOC!~%N>ZF_~a*pH^g)&0Wx47Q1HcyV<%r@
zxK~d{t>Jp?u};hl`wgHm8nfHumakBAqh|Hi#cp9zm&Y@=rSFEcW76GkUWy*2xP(AQ
z@ELER+F+x<tPaO+h^WF_v=KiUN*^bm1XUt4Y2&%U>ACTZO!^LL=|10PVkXP5ay>=L
zFZ;8x7%8|%D%d-#K&UF-al+nB5L^rED3!Fj4bZQ~k|}Bk&)5$USC9tmJdlgDS8C5D
zF$CjcGjvFLiswV?uE7|0?2Sf;SMO~9ARoThB8QARkNtgcMjYB{TsR2v;1w9#cgOZQ
zj{Xo3uDSOBAZpC~T4B`HpSWF0OF%v@K5bLsx>Lt2iE>>q`6Mae4)#8b<(bSDnzVK7
z6nC87Ij5UC3uVi-G7ZS>v5rSSUEBB*$%Tvlk&}X5tC?P#AJC}f5>y0ccOv8i^L(bg
zc%>!&$)u`6nfYE?;9a2YCp=FI&a5QDtS+}`UP4{Gxnu5wBm0?SUR{H1s{m)zsp_pz
z9<&lc3C0hkAflK_ViQ8HNaoc|_`5Jy(i*%Rp+X4-c0?u?#K=8)2sN@r^wo!6Mrd`-
zQV=2&5GjP`=9Z0O+C;d9kQ9K@_cp~ZTP7FM^$mLTLTXsQTEs+8i&HUsb&}Wb%s>Wo
z4vE>t{1E-w@z?(LS(Al^14su439=|@h=q-|7i+!Oe8K1x)RoT&<qHb}-dpqvt<fQB
z;goCdoip^@ol+%CLM29n8s=F7j++8Qt}B=Se6z6J`&Y5rdcYad8J!AokApO_63L<z
zk9T<tPp6dAkn;OSfyxH+o=3X6&%v_Zj}vJ+ZO-UzT@(6QS^cxym$Kh9xfDCpr*fu=
z)UyG16gFF|zo6L6cR>+69f{^s$?chz6{i@}plMaEwKl&m7=b?}k*BSdoXzi1T<ct+
zV-}^V68Y|*QEalAR>b@H@oYb7vl`M8UQcWN?rd@04|qOQY%$nuVH4^3cRUue1+Bwm
zxwxOo+uJFlc?e0!;%nshzTgf%E&4PW=@B2HIoT~lCr6y!HqZLrO!-A%uzQEnsAT5@
zX1dfrWFIcn-xV@hZd=oG6s`sw);UR1R5C>r4SG4>I^vQ(0*qde3BJazpv2Z0fn`TG
zqsDr{hBu~LxEr0!C8xhHT<rQ+6mR`8i>PzY*|m}_up!h3$8~kjr0o5AsIU!rbYA#z
zlhI}(U2X}eT&N=PrJr9H+|3W{Y0VtGJN->4XIg_;e0O*IwA!>F;0~vE?7ZZCHjS1g
z6`I%+xg~dXt;3L2W;HU{?XT6qMGQZ#QcDzqD>U0F2f(-gVZb(LkWJg3)D|5?-YxUI
zjl$E${iD<A!v-ClbPJCFtX=K1UdB)}mLZSO1f+7V8h*(P(Y7G@#`Dj1QzYsX)QiN+
z(zzvxC#*!DP1{Oq1{1yoe<AJLw9lL<M=9@N?B38l3*B65KnKn}W7nZ83N$L(k+!eI
z9wUJ1xSMDsOsA9WV00`?1V+BQXiW=^-e;T1?Y`&s^|AL?R+E>1r}Dr)K4?1;huRDL
z?pp}Jl8aI-0^0<Y-O~ch?!|j9LgfAZNdtyZ&Fybv!6Nh}gQedad<rqkoFB+9f<T8c
z+$1x{+&0G!oWvfwy1M3;4ye2TouTvef+jv<(L??$0`-5!I)7<?N)hxYIE2h?Ge=)i
z2P=}vnQ}-3?S0ouKKyk>-HQgTW?u@zB`DbUXteRHqN@1Fghk-@LBZUh`3y-l5bpge
z)~1GtLKxyaMH3^V6RgR4_so8gnB5w9ufc{>qP(RZF^OB<^?MX{Oe=c6Cpy&&B4Lmy
z>`5roTP36gkHm;GCftkbCmS0f^u~exw8J=iRahHOk?%TBF&apWJNrs9h+kdbHCjXd
z^O~JK<oUZ~V8MMfZF5XH^7c|tWG5%6N=z0zCr=!bJ8pvhhEipVTvN|4Ar5}$?L}xG
z#*BR5M-6db>jbpiX1Au%e?du+B}8%=sEbjoM^tUmwvQS3cB|WZ>^}Q5#OmqE(IxF~
z+V|&4k#fE&&cGe|S+vriWY66g<r9}JWF)4FF}E^YIXSy6hlL?r26B{9Dd(>Y!1`Dn
zy^PJ3zuUn3arG+WwIV(A18(qdr5Yb_GJl-95rXY;aldc?-QY@2>QtV(9JEx9a>BN{
zbMV&afmDlNGFxDRo(Ox05Tt;(PSOq{O(2CX7mLXm@Z#)!pjWMJukY{e<><V7dK=+e
zxcIE?y<_YC8NKXVvGa#g8E`L#4cGY~1$B0w{|c64)flOxd~l;LPqJC$Hj|0=eMHn5
z4+f){IJYj0#;3&~H1i%Hq?f+_TcG6+%k*zb>SdbQwZqKnxNixq4RIp%NRGN0>n<p<
zh{=`vwy~w7R=%9;L5f%~sOxt9iE;L8p%jTNRbhdtZ`(+IAk?J<%`DoSyAZ@km0(RV
z%_}Jv4Z^ei^S0FS`-gJq96rS*$hINm?iB&?KkaGpEX39ZPmL8w><wzRJtAQs8`i7*
zU0yZu1;-XTGJN^sqgk-d2pCFI!qpM+3Cqa%0>zl-WOGahV&lO!)dQMn1Vm5|Fx;_C
zD+extb|*Tz$%xpH)9)zdiP%7G+U{>`m}`EmIV?yW%W_2tgOY%v7;0#0WhGA|pQCzS
zhBJ+cJsrPUs5nYUx3C6zwln_8=0%L#ARR~lMp^*eeXByASimmu@S4YiSMO<GRb=c*
z;uDQT;bw$PL)^Q1gcxzTa4cTWNr;mwmCJY-N2AcVhwh(uQ@3xiV}Qbr%r;gg|BU+T
zQQ4z$f#6e{+w!9vrwXyYrG+V8*@vP@M&C!4%p8O{SEg*>IYG)l;uGhAe(~SE@V)G;
z)4oLq9^K6E?^GGYcIiyK$r{mvFS1$VYDp;e-FZg}Sx4h)S6g;+ak<-A94Za19Cn9u
z-V};UcFtt&YG1R^4ug3T0&1CZcHRB^gc<TptCzh~*X@6wpJ#XlS+J})*!*r8n;ac3
zPoI&EX_>=N4h1miQ%62bZt~(Wj7;$~U9*=J?o=;j&~P<glXjBr5Vi<i_oN8Eg&8d>
z^}F8Llarbb_hjqElGT+0fSgLM&gRa{N9N467n(-T8=G5js+Nk&#T^dxR~<ptVMwh`
zfnm#YLS<SD8|M!b3!5ka;?!-n25t7F7%6TXS82F?2k&Iw(Cmuw0Iei&d)ozyWJ8hg
zeKp6Omk(w=KG|X(R4XvSG^Ao$qetF>xmV>+T{@{F4*V?^Wif2cZL-}swe&4{{;;k6
zX{&7<h9}@*PDqneT6VSbE#w^&;UnxsVx!Y?_){H#es}NlsyL%ZQYJ2*{j6kSR5aj2
zYV-Q`#lV7Ym;c8iSzN@i*<Ro!*X18eCg>PVjYu#Wfd_KEa`Lt9snd3cr>>63NF-N0
zsG6;1Uj;syOh{IdA%zt_*9RIU9`2Agzj#TSDd4V}_F{aOGi$RhmDGwdz?ZILIt-Ar
zOLe|pvdDdT!auior=Yq<$Zs^g7({@zfgYMbh$o%ea#bDeB9d#pS%^vFpN0Va4lcF4
zgXejN4QfWZwDb>7P~<!>(;sjV4_Mn@xSjKZTxVJ*>WGn%^x-p!!(;ZO<>gsVYfXV^
zFZq%spq-VLWWU=sHWLh}0x9iSf<*ptbNzwB)~v=p^Mr;Jd2tD->>|S|Z|{DuD+V`J
z+xVW9t?P^^mU(w+;&#873JrADH^j&-C1LSP1pr-K{kw6Op^M~9Dr_4fB~hYu^=E-_
zrBgnTpiX{t>^Z_A*Xx1)Gl|&7qvqI`Hw2b@x?AQUnM5sB>v}Z!x2KJy(M+6jgn%Z~
zgYRW=&ripr38%HgKdHaurP_Ge;?N`SpatpLCo=lf&egAWZ0iZACvM5jiEI;$?rLb`
zV3$c;jfJ2$8rjF!xfgzY)m^g;<P;UVLXH0qU1n97)z~Nbc;aiZb6V(M8x=C4-lL}q
zS70S&oJ_^K9{7!|eN8jA{1p(RVMSId(u7DGE|MXn0)p#y0Tua1J`+2g{xN-@Cq#z)
z2g7q}hz86OTNG!qU%wzGdnb{f;-O2(j4h8RYtZ77SQigWyu8HF%aTvsaw;1D08k<?
zIExi(J$9VnR34&hiZNgMvH0!Cu%sm$VZsj84LbU(+t=ExoQAp^!xxka-y6e}qC;>i
zYrsg@CV;o(12IMra(wZo7+mb|k-HXBWLeJhry9l_Yg?R2iew?d7S0ZFe-8w{%~?e4
zV(VZUe~fu}FZg2s9($|P=9kz~0)FFrxqALDfW~BS+qO<*wk%&~x2IW~3z<~R`!8t}
zE+v*}3|~Mdp*<Q1y~S=qO>fWH_^MPmZT?yIcg!JQ)p>ti-cXtdeOgZ4{UbXufh83p
zumGR{b~fOy<k`aMzVVeWo!guBE%*8*0pujdW_}+QLeZr4u?d!0vAzahO3}|Y5SZwF
zB32-Pax_Q%Ta0ZRgHfrORbZkCq5fNsN5EN4Uwk+}btmMg!B~FFyZmA+V-Hcxq6*Bt
zor7C4Y`$?>D3oB!C?X!r+1~E|U~w?;N?~&B8n~!M<_~zlVw`VzS_|nIgQ?n$8J3t#
zUDm+!F?`Y;C<h<dt5{La{)<q4a9iS2@1+zrXyi{KaEn9WSg1DM?HS6IkSCMMtrXso
zjmg{W8<utX2Z4GSVKE?zC1r|GxUcy<({`Lh<Wxd$*E&Y)@&cIYu=Jq-ZwORY-YlE5
z!NfrDKOL?XDTS604`8`-m=Y;JyG)x*&{IB5oa!8z+kJ!a6GI~}s|)-}J&u_zsTlQG
zYfc(!Dh9xt(d{+)BORIwYS|dRmCsT<_0P)UTiY6rl;wAZW5My_-hTuf6d&zkhr}5a
z0Yf!6gX?%dOB1FUeo}}5<02DRiT4)PQja94_-2{ye#pJjP0#%?=iqu!$!GWn#m?S7
zp&xbN8)4BgMTjf80z*9@-T9t;!hiv^YrYxJfOPnqfwMbVL>>tiM~^ER)<?nSe@c|L
zraz!J#Nu^yj!bKSzphPgaKGyOd3Ptf#Eqh34Bmo{Bb(~E<{jR|+Wsb=a{X^mGbj98
zb3{tMhj%BHAa!jN)V((o@|IUskhlL8W{k49dVF^J`WTo(_6W~|Pb?UjUET=duY1P}
zX70oO^LX(81v*z6*em-zv?7sX)??)gPGhG}%82R4WY3WCCjG%X03XG_t8{F-8&n_B
zTG?wWf<E_y)JXfd$3FFir@53VwHORu3UekW{$MDYNigb?B$ka+OrbV^m}E9A09*|B
zVO}g9ItFfiCJBNfmG~9ci^bF$RxmhU>P34z;^DKcz;{OfBlS7sR?r37PRp3ozjuq4
z-tCN-Krn=xFa!4p1caT`A3BJbSA;z?oc8E{WtS&8ogZs^9bP+fh-!l^Owkz7(z(HZ
z9Fvt&Q0oBT4@K4y_;=7~i`nMB!NHF(hmO9LUinFpD~#B`G>Fv^SfpmXu?^mq8~NV=
zVViI14tzs}?CDY^C_LMT4p&CjO|MDa2(@e!%7~B=b#Q%SfYhf0bEr)E!B8UcQ8BFH
zFUg=8GZDVE9yO<s(^ESpVb7~LCAHTt_MV$@jG8~qXE1x84lG|G!ZbFcy|U8DZHCBt
zyaChy!I8AObkX7_w8~(Rz)A9+ca?dzl_cgb0)fW8G`TI)S_~xNfJg0Dn&m8Qes`D9
zloIdX+y!$E0(O{wPc;!Rjy`oVLa{PUaw6$|n{cKyqYc=<HfA#42yiKxrN;@%=jFOv
z(>7e}O$Nj-aA+Jy<KA8-ve7o2-zhGPn{((b$YUIMc+f-R3yP<n+(I1M#^ycO#dliP
za&nYssY`SiOkG$PNcIzj20TrCUpxaqivesSoXXf%kpD1C(B#zLIW?|9z3)Sr>?%DE
zJY)=7+;T4Cv&H43p;g3X%0^X|+4k5v7zDognBHao$%leO6k@$yx75x)Eb(`mO~NGe
zYhcq3r{`m$kE;8g3j(*@sBpMt<aPvnt!m68j3PjdS{d|E#A+(9&Kw#egVkr9=J~cL
zTrYhk7PzM`ofJ_3`w*g>)Lg63T_{ytBq$L7N+9{54*u28U(WxM31|_~QN+F8zrCU7
zn>70kEn*sKY4C{+IE@n0w$)v4j-3f($1tpeatCE^U_Aw8w<?*_ulM1Ipu|J4mVT@>
z!?F|js*RR^%$>3?r;8JC%xF+$hinj&0(d1C$GwNaPNS$l8BaN}%nbA{X-f#e!qY@x
z_euSc2$}b5e%p^vlc!H2T(OjjuELFW5=9(iPb<s=AXHJ59A{iLX1quwf5<#k5kZiN
z++N~`zCqWMVOXF2r0C5@IDCqtIJ^NBHrnoRf{~hlW;*Csz_Dkgw6<U_*oT?jlLV7w
zIMAU%;!I{Ix{g5h&wZZFD-{urkL2?3sB*d9`dCa{0Fmulpo%j47n|J-Wy;Xt9kOs<
zl_k|`=2U_cLKN6Ddi<~%e9F%>6FkYo?>eWRK(SiBY3mR3o?e^!SqZwu+}S{$DXP6G
zrI@XBYRv7V^0Z{V%YBp(z#DS!zFah1-=60)qK=atALqWCP3!ERdOl(cMwr5h2}7NA
zZFpi<TsfT>bC(cnP~RS|{c>}g`Lv_@A`ozt6_{Ch%>%1i-FO&?hW<`BKPD<uLG7a)
z`kh`)oM+}yr364KmRIua8Y!E@oR$}aM*Ww7#^nmJja9v53k0VGWM3RT;dXhJ_wCAo
znK9m~Z1I$c^biVh0u%Na8LC-s@bn?JG?qH9X{ZS<UGkq79Xx#egBzTsj(srMX@?a0
z*%M+EA^@;VQ83|kU2<4lMypJX&$bR%rFqX-GD<zip|WC~zVgxZQ@R~u0!V&gfOh`F
zIA)=N<5s@)cEmm{o8xeHUqis7py*}SlPg+4*bp)Ap*o5pD`!(GcY~LB5C@(N5rsyt
z8mJ7H5Tlh^J^VUL*`ul2`C@z<cyZ@?w+Tdvt&Qv>VD1$}Y{H#Q8SXA2HE|);O*N~z
z%T2RNgC1U`R#hZtKxcjq<uAa;i$oP_4T|*2es}psuRoQ%qEcu;)#~+G3i*HNl@-pl
zLo!^iD?XNF*Hu9Jts@%{3;fh_nN~+sDXFT}gTslu$Z<KS(0I4uv9J%z)zv4{ByJgH
zs{ZVat|-fdp0RPYhAM)xLyN3C6eKl_Lmpf`H~d5wlgn(Em5%;t2R^}}e~H?73yktC
zJuDyIExUgAD^MdR!T|6VX<p0-N@YTnJasxAD2rxTx1OvrGW_2ybX1;pdk6&FT4JXt
z-DM+gHXDHL@UbJqlfZUUQ$V5w3}oQ-N?r*sPqaMejYqZX^Qbf=My(e-eTpN^(U94}
z*Qkg<?^otRCqbElFx9uOin1&!R>5vg*@y1eUJx#h#Z_#=t-ZE9K|(lI!gPpHBr<O{
zpQ7O^2pu_1sExA|f`uE@s^_9htvMb_I~)$hDzh%SH?En?LVNPZ6xm)#*I!__0s-b>
zCkE)rpE5FXOF8%yTR=UR??IK~3JL?1MFE&JF?H9iL6j7+p3M(yO_8!~z^i>%rHO1A
zxfCn7h0*i-V%z?j)4FSw`-aF7CEgK{c>4~5!w?mM&%VXZZxx=!K!v8ZsjU@`?^|qj
z9@|I~v|gYYn<n{LD2TzW-IrE?67X50oeZK8?ry=Kk+#K&)5ePX6^==<xODWp;jekC
z$Zkaaj-Mbi=<8KMqhoM1ulm19jBr3WM#cmx?Bw-pR#W;{W$yjyZ^yL_wUZ4@_%=L`
z{UH%`rFc5JnI$0Bwm6m&6rg<doEbYoO`}ir#Pug2=C1}$)1Fc_>zdkH<pP3Pd&je0
z&wz|MOgrb75w>PlQ7YmCcMQ#Nl-cki-{g4QL^Qb^Q?&~xe^NKg$So{gqS_XpWVk~X
z75W(Uc?3BMzzniylk_KL<GVzpX=bn+73qo)QM`^*HikX3Ogd^r6i4tTII5a@fZUwi
zA4o_E%zHy$8I^38C(DQFN^Nm}0G(I6bU*(qw?3q-=*v;6x=Knjbls5fqGuDkBZc{_
z=o#*9m3T7XN#;KGmiGA6XW5VLCO9;OTzoVoNE$@WiBf=X(;yC9jB?7Gdv8TfUF_cv
zHM?)4t-s!QF6tV~{zG`4{|Z$e^=O$)yb;OEEw2TkW@O~h#D*UoAE~iG&!R|MF$2`Z
zSn_r)?g6j=_^N8X%u_dSM7rLz*IBRDAU%~QTtqNqBE<t&g57xwB4^ViboJ=J#ckng
zwdZdjNb$kf_pzdK)hDm+Rz4Do7ujHH)iuj(x<8SW?pI4g<J4L}mP^4=Gj}s~7}r06
zo>hXFkpu$3caB@<mA}7SN?+eiW4cKzh&nRL&CnJttXtV`^Rji^`5JV#%T(fCbdOm4
zO5u!sz^JRMyAuy0L$Y(QODJK`&8)3!nQ&T^JdDZ6K%1w|Uq9s{c5msp`CK#;giQ8A
zT!%hn2wRT;3bC_(YOO~7-2If#WfY=P!#jN1QqE2Qt{t=xmN`%0j?>ZMm$&yJU03+c
zO^G52ky9H5#3D(lL&|NVvek0)nxIuH<=sT&?&g7IoW&OgjR!c*d$87gTcx^b9J^aO
ztKW6=V5_zPw@T-3|0M68WA?^n$wk(|Fl&R^PMkN4^4<p1)E9U(`|Y%@u3;qBHqEB#
zp+6*jjt;J?mtJ7Ns8RW!`e>haQb6g16aU7|lHhWdKK*WuFp7U~%_I!=In+$9x^7^M
z?MaYM)&6q2C{Gx2Og!6x8FK6`Z=?so=NG9)C}MF2QjVE5F2Jl$8+3DxDI=*6Ma3de
zc5P<=C^Mg-OUpdpg4;A_YwzmCcS*85d6WE}xVq9J<nOP#VTuLpv=ep1xsR+Ax@|40
zu3eHAB`(9T==OU_s=-g9Eh2IZmomX2G7%vd)|0WK#h8}kgL9dtUD9<eL$E>nD#I43
zQWEDZS}PScHNy&Z#uNj-s=ZAH`O=mt6ZVb4`L9REC!5IM^%|6F_ciF1>S{aX%eLye
zJd;nK)=2lr9uYBtcetRK>&^U(zSW+mjL#NVhRM=l^@RD(+bzv~Fslqx+)pz$;}eQ$
zgwlqVW3Kz$1GoN=h;t}yY?yh*AbNCh<RIF`4?;k%<O8B8P&4P@{SFoU*uv=bb3*TX
z8S9*lu~{jaF#o=5+a|C56?v%JI~zNZ_q2KV=9k0f#2wM?xxx;rWS0zsqlZfqG$00y
zgc7QoCP(MXL21nmb;i`6P(L{SD6zh-xwCVkM_c$QKKr0rr4}k0MQ73?bBTJj8cUv~
zuNCq-dPGq~PyY6x+?B9a9P_M$!}%zB*ThQVb_H|Ji0#Rs`xPQ==L1~wcrsbbGW)VL
zmP<}+Br>F^Sd{>m;B&an*E)~mFBMk@b}Ti^s}r9XAO+H;gt@14=@~`ycH6$Bc*FHf
zQ+M}IEQ9us$u}=;nE;WnY+t9$37!xrLMH~nCl>s*j!3p|_4>jP6oi4lT$jMRjdS@@
zqfAn_7*eM=^R-PVpWpC1X!w<%PYz^$KCAwb1#WrhxBb=F%CT)=BzcH~7QPgFI6lb&
zkbM)PYi_9C2)Rjjo#yw$SfqaY8rdSwc*z4-OKpCqw`(-V9yZV~_^j0+7-8Q5A)3BE
zPy^fVpIqo)nTS2~*+k1l`#QHP`qi0tU21(okGfUL!oAhcD`U)gIvcWqPu?%*uV<nc
zu<Hd~MetR+>Q(jTA8&=tM^bg%nZs*<L}3Fh(I;^Qhst$Je&Wv4y*D?~<8!=gabUAf
ziq#xqS{aN(kIu^|*jc7vRq2gXY%Pv*Zp#n^f1L>8S@g>#fSI_w7loI{2>FdfbDeY%
z#yEQ#Dnt$nKxXm>;Dd}E;`qrnel`qv{-Y_B^4XlkQr!Qh58UW+(YU_{fhbG@r@_~Q
z$`7lgA>#1|8Az)}=oMyQ|E$43xJi)Z^3&-x$BPECXLRTo*GcZ4vf(yuY#&MUwyy~9
znF8AHaf`bO>~KWFU|dpl?%w?fKI13DGl|Ju!9Fcl7BVXY@i*HD&(@z(tG{|(vDijN
zJ}#$&mwvL~5l-wZm6)|n=DndW!G~2s-8U%q*g)NEzab}+7H@D)9VW5s-?DbiU$GtN
z%R?nhfZm<O6EBIK)d}R&VP`~(+~d*T5wJ-?TfXXiMCmfMLW8S}ET+2O_1#ZTz*UuX
z)5qKbP}xQ}LT3^|j(WBBPpInewz}w2#3cT=RdVLo4eOa=jOnsof4TkqqAA@_aBy~u
zVaS4u)|2rx_87J(RI{4?Cna}i(cy!F&s{+9rZuF2qsg&7I7MVNN8rZiQO=#N@v&1C
z?33zI&eh|Q18y6xHQa&boTC7)QOzEaqnX?v0@A@jz55uI8c^-@eZ|s@8V}cGFq2}O
zhP8o=@?krwyp;}@gE`P?XOPP@)QQmEKfF@3#+XPk6XWk{bI;O5Iiv(J#YL`xxTZ!2
zzLJ5%O{C$rB#hV<1k2RxPyvK}$SuyI9iN&~tNw3sd0En%;!T1F1j^OTQ>t*m;Xka8
zt1q`a19I+Dww3s_*n7VxBA0k`dFZVAf1Ud6K1X5@wiqi#<S7{awAM;nGh-dmfTg|m
z{5d0Xx{sERk8Mv^9tFMXj{!4Y8Ly=^pU_=rJcmPLB^1$^zTytS{2$(4G)5a~?v;A)
z^G|&Kn<BBGKLk|Ge$F9mfhDT7YA&Re&Kq$fE5tv}7r#r2vDJ5DF>aD8xp!%!2PY7j
zaw=ukG&slJuKA}>lY)L}vhNB8LxFxL^s@2HoL&5qEp9P0^c3{X4$0To?)5OYHbY<;
zeTo7z8;crq>cLH`n_wE&EWihwGhz#(5}0=yaHxG(7x+?5B^@-3E3ro=9itdXxrc1w
z9U_s0ssAOvk_xb#j`68?i0wZ$1dK4r`k4sxpdRouj5kE8ZGn;|R1EM{zLHm&0z=L{
z>f*~z2rdiEo!C@fS;mq_MyCRb!q$z>%s3)4dcLLs#~Z6KI-v{=rxpWn5;u5lPcWtT
z7wscMa<?w0gcVvqPC1j|sHHF!PA-r)%C55DWX0KeFQOJ>W0PZDt$#F#W8<Kd%<9+L
zMnl?qK8T@i{cUmAEi}sdF-A%sO5T?SKq4(w740^DUVm|agCp{5{+BeY2JMq2O;sBf
z9IahbH+p%AVNRBj5?vEkmj08Zd8VCsYL%(<sc12+BpBaoKtD(<e$oZT=#Ry75J>s$
z9_DS9S&{AT1;X_gSM@7A!umakC$dC$K6M&HoLElvY#%~IjF#!-&3bFZDz&zc@r_Mf
zT`$3Tt6f!hXt)L2^Qa$nVn{BPsZHcGdUh50?ypM8XwM8#br1c3K_1&w<C_uLz)>E=
z91E20Vc;^?Afnh$zm1<fS8cuqEq8*CZCQVQ5NSsJN1cf?RO)qh^d9JTL)|AH|NM-X
z@@?QFuuTYsBBC&2cR-+j3%I)WZudS$#@}JJHBvJ*pnB{|GY4KX0fXz0cX&tb<v!BN
zpg=@6A~T4mNfs6?@amXpCa|@65i90RJ}&)0F~#boLQ#O2cAKzwFG0O(;;=~Z*NjSW
zVgb#GDxqo#QmHO25-lJac`h{5h=!_=MapTIVE7OL=XTf(<p+?rS3nSmEKge?58Rey
zbn-_RDEjK>ZPm%~ok^S5`{bS?w8ksw*fQq$fz?jbnwsKKvc?caDYd4qWpp>%F@c&Y
zYM1keCY4%}tCg0HQLdFkKqjb6id*3f9f7C^4yV<#y7KjhYC@x=@$mN;a24O2r-C`Q
z<lF_Pl1TE|vxN0t=Z{@I@vl395P%4$D;0f3?d_n4{d8HMhv`SN@5iWwL==AQW*Ou`
zn@;<ghO?iM;$ve;!fHJbVxW*ta*ia0262K9=7Q4RXtZ1p%uD<y?%%FR7UM3?{h}{s
z<Ne{cxD*U2eyMn4NA`-q1FG|3sU5rLW@Z)^dCL|zo1D@-Z-8;=<UMyBqq#dAQRwrP
z>FhvWa1AR)eX@#I8pR)PHJV~_#W!T%-d}RN)}0A8E0L%)C?!S$UYv2fRhUejO;aUw
z$fK~on*~!(H2Xi|*<{WhUZeQunHBVpj$((JWmvlAJwGenb#~(r+0YcJOI5q)q#50@
zbe<Ol651I2m{qdtm>5Ee%}Qg|h4Xji?s5K#)39XcrKNh01;9J>QkW*YGHmi+lFEW)
z@fj77eR7sokQYZW!)+MkHr+@zkIbr0!)$HT0olq=gxB$qjg<MGC}N(qb+u-yT6>02
zk@0&%?e^<h)=Yg^&8HVCDvxLF+;xqZISM9;_})+M*A`I_8#7G7iM>v+UH9{ImXj!A
zy3^D3%hQ_OHZ6QsN!)%M)Jy)D6z<K?pyU6|t{gaR>S&cGZWED2YV+&Oflvf!Z9MSl
zn5+?fgIY=4G?Rh^U<Rndm4p2>k0fuShmL>`RB5RF_T;PGn$hmvNO+Az{$<Rlxrf?I
zZi-8tIq`gtv=$DSmTPZAo=-sEYdgagaX0kNDA9CQCZW(h=E5LEzGK?N$D^m?y{+6F
z#H@B}ROrU!eOM1ddri=!m>iT0011tEX}^U&ukw!0;YgGxCF$61Q4iKt**R~Shai)Q
zYzle;<ZRA>@!UciL?{Qr7m2xfUtJ=vC7@)p_U<;vCphsk#JaiLb5nj%WrNJ1yQdb>
zTYPn+;fDuaKJ9AP^NYhfZSC5OVomy*`ugGZyY*Kx@10t!=BqV*zuV=miEAc4dMk!3
z->f^aLNv`E_=KNs?#BXORJnXEm&d(0F1hL;DcCTLAX%Mn?2HS9B2U1s%ec+^-Jso~
zPky311Nf|GCtn?>S6lk4IGVAV$^1sB5t%+)zxRw5OZb7>VH7@I%<UF3aDiRj`9gHs
zal7_7TJ>r>dTA^;p<jJ8XTtzJ!vfXRb4qeL-7ceJR|MN?dmh}hDb1JGz*b!orwTZi
z2uY&5;lNRje8nk&r;E5jYy=6o1-oQ*!<g@Bdp~@TrzX{*bmW|hj=!;8s-swqw0?=j
z(#i_Ut7^GLi$WQ@ev$wKslO;4R4Ye-bZ3e}@SWWRR~i>FZ+`wO3L80(ztP>K<Jn#E
zlzRCDMXmRb=@=Zv7$FZ~)g((i(D7&?xat!glNo2?cX~Bud_XXH6wdE)O64uNSDP&6
z(VAvXqn{rrD6S8@fiyF;RP(HgW>8PHo4HJk_X^F--)m1q?l(foM7<uz_f*ew)>~6t
z*Zd&RY{5Du-iyC$N3-b+oJY1W$K$zgp1Rp=q$X`7j&{vomr1T=Y!{ea9qsVtsMvW;
zpW$Ds4196|9N)6U)|`aFjBtMcCxS88Pe>u8kG{%b16UxRVw>@&CuUQfK8+}!%1I}7
zYDKOjZYr2zMFXya^ut&ED<7)H##<f1-``JY4La{W>wmQ+7WRM%O!aOu-1gr1y=2x|
z2rg)gnG@KH{~}%%N4fng%+`Gy6_bQP0jif*8!ar(*ZRrD!*%MS152`G82G*&8Ygr@
z(%9BBDzjN*r~o|;q<rJiWiW}Cf5yJKaEEcn=h2eLH|LDk-3-o=BTaqgdrM2nrONcL
z3yu5N$V&5)w<<2bpfe6Q--V7=<OLx%$>x(3J}SBE(t5*N5nt2R;iJ0?gjYB0gIk_s
z<HFnnl2&BoBCT?z%VwUwX`r&BfUe2(yu1<}H}rQzu}FPA3EV0pZVJ%GTEt#ZV(f&W
zU=G4peP3oW-5-B#FP50r_qg&NO)agoYb4f6Fa@WwxoR&Z_}jVu23IyE`HAF&?|r=Y
zMvF-tChxB!EmI-SYbt|EdwqAYPu24$qUblDb_3v^G{5S-zRPYs)ON(d6o%51OQYb8
z{x`7~u+QPqxTIOm6R?ZFX5v^nRM|j+2Zz%!*j41X@x%&^eSU569p2wLE9s8Hu%OX@
z1CY-bS^M=<`CYQu1LJJOU~X$1Om2B@Gf&T~p8TPwUZ8?Eq(BufKHV^#4h8YHz(0?i
zIf+QQZqCxzw$J;l5@A^%%d`i)YaPjbv1s!q?{2#WMsW6O6qJ8FwN*&N6_jC|0qfyh
zn%h%w!Jtd-Uu5|lnvCvHM)v=*3bgV{SJy4Lt>=#260zI6w~r*_YnxqgB{QvvWT0a_
zpR*)vS*#<!Hf*#e0NJLH%Hs111OH_p4<{+N=H!k|Z?$e$OkCK%51SIcMJWNzB$_#=
z>M=;CQr`<Dk^><ZBpgKe&rED$<Zq5zyb*Q_k8fcL+tgN5xqSo!Rw9o;#jzo0s*h1}
z4BqD{j$A^9Lzh2vbp=MbR0C5}sATU~jbA?oJf;N%zub_{^KXs5@aFQ2PqI=f{wKJ)
z41Aquo4x(;;v@xT?Pf|{?XD~^WhYV(yql{adBk}ACnWSUDqJnfh^=t-@1AGCVH;q5
zN6Rx$lh|#4RS{8Do%O$cOyuIf9T-)LtC&ds#n^WR@tWWH)&ru^PmP|a-1pxBrvR}S
z3J^Jon(rxK_f$B3i-$$D24*(Aa>1eQjRxA*Nf+_*-0LP^TF~X0ayG)n1_^U(gFV-2
zYj`EU_2hRAEjtFpYhB`n^n9LSFXU(ls$tj#1+o43yR?+g^A87;3Ifn%q;V501XW^x
zKtM-F?^oX{JnKm)o}-aB7-?T6oIP$%I!Mm*iRtM0TscX5+If#=d^a0!5*j}qZnf-3
zQ1J-dj++Rnl}66Oc2UW%McIYZ>*X6%QT!)n?-Qrso@(is<uqNe&=v-A@H!qTD%I67
z7~EOr*y8`8Mo+aXeg<dPq&&MI^x5%T?Yn(`%Q(sx_#$M`7*;d%(MzH>=Zr;$PE^7b
zN_rh%Mum<}uiw737<w8vSVO&|s1c@HBjG0AL81NVaFcKD`;5d_)e$}UfPt)>L~rU5
z5NbV5LD-xU<?Gvq5iSge@2SmFq4@JAMP#)6m0TnRKxjb8GbbK@|E?9!f19=Yw743H
zf-a)SB%P#FyRZb0P3s)0EyCQ{o$OnzdL!4t&2)`p?MFYU-?Yu3TGbylqhKHB%y0gw
zlP2*q`I6Pkl8k4F0|#|`x;lD;2?fA-YwkjWUps_(BhnOelN8wWo@=A?bvtxH$G1r`
zMWZ-*f<x<%%Ux2bVR8q8Cb142!lifPR%AaGNGL73t6zHc-)iwzwnfq%K0yfKLf%8!
z{e7CjeYblK`tbmrPkhxKV@ng!`xUBPP0_H_fdTC2Nm5ud#dendS99IB2n+<t&GM5g
zhmB88sU}Hvk^%5W`*lYH%13T)kS3n-mfH+(ZZ#NF5{ib(F&<M-CRx@3u0M7@#OAp*
z8pU0UONxjsTby7{8h_ok<Wvz!b2ai;-SS4r;MHVuFJ@%t^CY4Yi;h{X{OY#kUyLYv
z=KWM=m7bXa#=QoBu!pb)%usDHWs3b4lOErS^#86IM;$kprX|rhO#*5s`!Geji{f<y
z4y=p|=G>Csil5CrRC{E@k*_<uh*m51BFL8QOzxaO7zfDpl-qHa2eG*4TeHu(Jv*vR
z5_v_w0V@s?se_)sOu;?}*PPfeG&VNtZB~*?B}(*wxryFJGE9P)j{65Av9Zxxk#6pw
z{@}&|i-vGEJm31@itpSn&CsCZfmEK85~aNAi#`J}ju`3L{)L0;lq-|kU<nGJ+Pr(;
zoq{^!mq<4W9cIcBlR(E5_zQ330YDaMor#<PE9@hw_YuRlYsQWMG6tHCfV_i@7I~<z
zPGYf}{X%X6N(KdI2Il^=*YA$iZP+pE(l>~j+(AgHHTUQ@Ev-?Y-^&LDq6DNAD$tsc
zRH`K;c-WJzMJeV>RSbCwB@sqTgGY$VV!lycAy%&mfP>(;uBN#=_G|gm+MDY@mdqP!
zX&e5w%Px}%5pU|-%Y7y;sTp2>tbRE?a&!cd4t~g?)qeE%(Yx#(MN7Q+`3qX@ccUv&
zEp`fL(Ml5GipDgRj%vhHMWz3*h5wbm>wyWMfo<Xp|8=xWvU=T|f81-gB$`;4)%9SI
zak)%Jej&x3jR7AWc-yjf_iv$dH1OK^RlB~e!#BV<)j~>LssMKjnUrq<*6iVa9luiW
zzS-ZF%+|aUttyB+Dc|I7bb)Ly+Sot6X}wvw2`6!oL%X`d`QHu)P?dzB?feciOLh%R
zswOgZ$eynn=Anl$OBu6s5y}^VWbcNZEG5bJZaKv3Pw6_TiJJuR^}hQ}WD#lX1Gj%x
zk_DCo+P8)W>SYeTuHnBey1b3N-p37Ty6a1Yq6&*PaX%L%!-_;;y=y0=)d8=|em8Z-
z;vZ>%j!%T#3Cc0|cU+7mD^Wj%<klz4(Ea3Vb!ZzHN<bfJgbO#{718L)?*G~Hrhz!o
z+|uRwg#Eq$h@?)o>+L2;^6$D#C?=1XPYGqH@4TG@Sw4UgsZ*<b!<(cy!HY3J-wzJA
z&GtQ`#>{JaT4L)GZ~6j`-5uS{0$xoseDAhc#jG*h4bi~-uFw>|kfJxFj3wxK^cnmZ
zJBAOSrBDPx-1{0N*IhsHX=%{Y|0uOZ$@tyi9er>h0HMbYi=rnY>AgvpDH=~K*}i&s
z-<reQ@e;xBzIcL<vD(AidCH$<%g35nLiI#G<T2%3CKUkpX`fYTonZ>|HEBaAGx*Cl
zi)vaBg!|p{lJ*$(Ir0Eo7D@FBK<a~daQKv{c!O4n=ss0(u_ysMk3YdO4ays>+P;{H
ztIs2edA<{m07VUIeq_*>D|qwhV4#n){ki)yhnakkp7%nx4qjBnqBdypOU9v#-(VQH
zBZ`r&(m6!RAcp_9XI^=`>~nZU5zAR+9DuB9^P68>Ub+?lKw@-kpT&(Py>y*BI(DpU
zsU5t!A_nQA4T8CeLDhV6bus)Z=^3_8X3mML&%!sixBE9?#d=i!ifY{TZCQ{4a$DJz
z<_p$`!*#djw0IrYMk7^5aCbjUztZq;0P!M9lY8%_zrp~b;3P`4G=$yG!9YrXx*r3X
zqf=Wf>P-oagH?gbWguD=!-S>7+p>ky%bV{{D}ex&)ePjcMWd<x#x^gnyIHCUGo&J3
zF0M~-JVo5-@209w&;RhA#q%;Nb~dbOF-tFhhmB)u${*f^EsnI}dhV-;Gl$U=&8)2E
zYhBZ)q0#f8Yf)8N`h#P^(U;te?q~st(`;Kzc@h8<P;74z0hI6t+{<39RgNP~#cI?F
z(In-%7PNty)&DSJVxRvx{p}<~4Hr#Beu*X)84{}Zd5xB>vCWE@X_w$=)yaTX5+fEs
zPc7liO2m=jdpT=n_q{Mk-r(TNCa=?7ZlPY5D~W8GI5;Bt4}z7-xYW*S*!?$J)60li
z1t{|9>z&G~<I|``%(<bobgN4?E-B3N4s>xfxA!v@kuOK7na76pm;1RzYVl$eVwsa5
zL+$-Onku8mQX9sDFZVAExdW^ZJc5;YoG9|!kQsVYpC2(Q$bT3=#*z6Zg{$Q4t~MF%
z4LJJ8yYUxh%W#KL-3YcBoW#pPw(e$%Fl12wOF0jG>3Uk<x9wX#Z67tc#x7MMrQ@Wq
z2@KkkmpO>Ae)yo>g!B{fX`t(A`wIEZeRu4#xuyK$pEPgv$_D&|mCByuX6>_Kk=C~8
zbNnmL%ftHJ*Y<}CbmUl3&&UP`2h*h1kGrc5ZyHOVOt9l6Bz3Rz4$uMr@~}$Tc~+K+
zRzaLIh>3gck^z7D=vLT+!rI}PH6^-GYQP9j+!04^6loV_-kVoy{In>E)$y@nETLb1
z6kchIi(VDlrAEK)8zn)&PH>%3`QJ}XctoN1(lG;!Pisb3$xeglnlbl6J~@UQC6SXV
z3hC&S2u*x=c7;-#&B%&7>gNIwalNLgEA9+;ULlzK*vu_P^{>S$Wa3kMFF{j{?~$B$
zZa6mQ`t!k=lD?W2d5Y==Q@sixxL0B9XO}8*Y($d(o{1b^BPEJoYCJHA4CT+;ClVMk
z>LP=b#)7JoE8|%)kQ6xs8a#x5IO!Um|Hcv9^fQf0qkH>sJ}-y&eb(S|x7>^Vy!`lm
z{!r*D9Be1L_+hKYu#0S&g3UA*`Bo6C`3HnQ&idF<z0wOB^|3e;VY|lP_L5f?ZdEq_
z4e~!Dc!MdP-HD4@7>FhwIYUm1BJpjWNFBKKFQr_rB>mU?u$in`^?&GpY%TcOH0jX_
zRFiJxwz0%@#*#6cn{tc?-rI?COD0M#s>B`7y1Hp(nbIxhQqt$o;exRt>~a?8!qP2*
zS>U!J=YE@1aNDvZhIiK)g<TVfe7N=*hyT)d#2VN2`htb@@eeAm<B;2Ol{tOM5Irep
zR4a7sBVH;aB$$S%CuQs_@b)PPRGY31ALmcWK4tPe#Txu7%dEAGI^|$?=v*@Tk}qA|
zpjf0{2x8bqp0r$IHg&zkqf1Ns`RCV%+_`W|*DMJ@elmt}=!bwraZySP$u}qLb3i=Y
z*FNEUOBur&p_B>B+%;#cPvq*VJ>-f3&+nnn-!7HB9aM!0vru&<t_vG^@tnFe$iQ2e
z4gwG)3ayVGgrpE1wylHhqG~7vBiVTo3lc}MkwkX?F)va#U8nMo02s-Pxdwwoe*)d)
zM3O9@u0#U~X-3Ie6p;EQ2hi7j<A+EC-gs6c16~x+2~L07UONi2Ooj*K$#e{Q<>$j3
z1g;m$Cw?CVNd^9#+pj^{#1)uBE&ADK&AzcyhMb;Ll}LlSz-nB$9M3PpCv6oN-eUqO
zl^_~m*kbO{NOFVDIDX>Lnp4gi7Ar_G$F*|vHPrJhmpkyn)`(^Jca6Ae6mx-yhxhI3
zfuzc~Ymh2lC?G2V)^SuD?$B=@gS?_Ekvi)fC8g4?%wi6+r~w_Dp5?d{Q-Fxb;Z97~
z%@g_z(j|Hpz~Y(XHaBCw=Bgg&QEy@_e{72kerGDhl@bkl-9em>F_Gz2P=Pw`kSOdk
zVYQ_H1xrD+zD&BvoXq0mky#7EIdE8kRUXpw5K0n_hICr%T)OlwGt*0sPY}WO+BWss
zX$HEBkcwmTv!rRn-tG=>eD?*mH+D!&44J0V*=IN!aCJ0fGtDSfDx5j8OmBOiMz_lk
zef2pWee_|z^X+dj=(W%~;!ROQw_>u*b&gaDB;79QoKNJJ81^&%$v^x>+O0OVn#Ut&
z7TCG+3NO6)E%ryr2VN{sDgLkj#oy#V`Op3{g3#mg<x8Y_`I&=)(;cFXC5kheH`Ykg
zkI#$xq~w)XuHzSdgtD%Ka|o}z^d9FQI>WTB@%X2XvwFG7^())lUfZGD8Bi)Zj?cJ@
z(M+-#9AvWdtnzq0*I^O)uE;1JG8lBol8l+TIi{w{B+-a&vq`%%pnU2CLXQ}YGCV6;
zm|0=mp_)Y&9f3(QdaXXia*=k_W4O~KP7|WuHeo0TgMjI&ITjbE+1kE|v6f1uNVnHw
zdwY|kC#En4gv@z$I`4SZfY1kAAbHRsi5TB=#|=z@43j%=%JB-7kB|j~7y+p>a6;hu
zA=+eapmx6QZ#-xgd%!uG@De9JvOs-8<=3D4nb+^Y_GIMo%1f7d_`!#{wYEzkEOPSH
z5#D=yjh*cl?dE_+t?CAfd#)oe)!D(BXBw+O<(p&RgRwo;38SdXXfWXTiBn8ZEs~`P
zt5>hm8$=wPZP07(G8ja7Dx_X)kYp)RN|GpJ&>lH#&)P^|VU!?>GrIjg?fo@`)PST>
zZ&0fh==Bqn@|l~j&>!?zy?LF7&d(iSyLmshkGpy~jP2q8+jGc6n=!Vl3AU342x&na
zG_=UOvAtYjJivC0?NZ8ob(PIWM2)ds;C})E`@N*O-{*h$yT3=V<dY=|p+KpCZqM?<
zcQ5eJnTP0hyUffsICgBAE0=C_du^M3FQQy3O>BLZWfMC%#sF|Q`GK}R-_(+(5rbiy
zINsy*=`#dDgLpLH^2JLS1E&^eX>M;4ja)~5t<)gXZjEFzOc~fA#%S8v9)2+-Hil80
z((QF<@2?@1MF_>r%nYTHq}z+|gOKu4mF?{<Ha2f_^jOs`5fOQ(dp6!(EzsG)NUb|<
zy>wNm+}0=7VSJWoQtjah*CB0Xh!hzDJ+TTRe<#*j0^g@xF6UWV7!M)-feasRw+z0Q
z<bQ#Wl!|Vr$Ax#+c=Od&20af?7>tn&`YBtR`<y(vJW<98H{>oEmu!SRsKej!eL}eM
z4jD#|Xt#DL6^`I}J{vpx_(7S4m19)AfYlq984mlDD^tkW{e3!%=(qOCbi!cRqSNnE
zpP6GYirCq0vb(pB^n5a{>2^Ep?{6|USHt%`v`*>uI&5y<=D~;4JT0E@eKq6Krvh_O
zqRG+}ox0X^&v&B>>^L)FErWKCpZ@7z=GE_C;>lApRLVp8!&W{5#!UDrf^nJ3HIszF
z!8|~DND0Jb^no{NJb@2Cf%*>;V-mCxD50=oJaxVRq)h2HoeOW%v>ElKd4hTkN)?3X
zyUtwcCeCA=(Uv&2v2^KIu3UPTWY|MW!LR+=FYzPa_(}G6x9B!|?C)O1>Vz~-NVR4d
zMVJgiRiM#m@ZQz;h+BJL6Pk$&pO)trnQByN-`?b`PWbZK8D^FWeEXmNx9sn=+{(J?
zS(>#)nyr0S&OgaJ@4P`h8F9YiBhEa?TQ}Y&-QMF1Uwndw>e2n~FSD_Jn_9_Zqa{A@
zJuNOS@e}{<ze^DMWLe6kOPAc-E~nA&`+k?4l61TMd)|pZN_q9gJp#V~c_|JXuy%8o
zYnS%<rFULIry*%#k$ALQLw0vMEG^AVf{F1sn=znUdoa@EUVuKy^yDBmiaHE?J?ga*
zwdq-^#S-tmbAe1trl)6+TGQQMN1!Mb%JhbPtP~i}qqpB;IO?NKOlmSpjRu2J#O{8R
zovmBMiD48)wA*dEon5N+kaD?<R6aY~EpFbt&hnA7uBtF;?jkqbicIHI?yVz(>)aW4
zj@X0@g|U#u8Sh=V&c*9(7Apaw2yvQX?ZNjszWo99#P<Ul^#<e7D0c>L2ZsIdyZj_J
zN_5%|+YL+_qPpEam)^V0J8!Hp>MN8mSR)t?Qnt6696!D|!EXK#Y{%Ms;Q0Ho-S-rT
z3~eGhojr<$WqjqcwcErGO3W>-P*wqJH?J}p4JlXZ$XL5>%PeNlZn`0_qYm9ckJ{7>
zgD7Tizs25u3#oi`meK2VX*Rc+nRRRVw8`lAyKHT5aOPZ&?M9=s6l+9I7f_IMM8{(+
z^kfg{0k#|KvN?m!fS>!hU+2x2ukz^e=^WcTcz0p@SntPoBM1XPydT@`cyow{+sxyB
z!4InZYK&z#vW!L=DP1<wb-_!T`)%HO<2o;X_bNIG@+~aNVZd&Og@qXoRzEU%XYR8e
zocAOU;%-~-oZ2Ap(I!HY(Q58+WaSu<5p3<Yn4Xzq@yKyhs##sVOp=VKmg@{-R~5+O
zA^j*J%Obk{CWClHZEA+zV8r(B9?fP4BV0q${{9}jyAd<9HG;q=j$^vr0o&V~EG<8f
zm%=qV%fOEBopeRIe9TRjrs&K$ltRfvw%#BNw04^O(|`60Tz_wkr_U~<ynen%-R3%T
zCqLfg+qL|_&y$_5D|qbWe;D~_yI)O&)&he;3WdUB7?1ew_ugXE6-ZrnV+)0c(Qdi>
zr$7CeTNi~J6W}_YEfYS-xIqHvt%}K8aRA_knS$>JEHAIHakE1l4e9kdY;W(8rU|ub
zo!Qw%jxHYM_{nqZ>~11}LQq0VkM5vNnj~aN!rtx{Yqzd5>a^I~-Dh*N$4)1u-cU%R
zC=~)k(RTxaM<bF12E7qiu3Y5#ue$XxCbOi88_VE$Yxy2P1ZkF$CMl_QJ4%fJPYAq8
zWM;T{=`!E>{tMh*>yyPPOEsvMGn8*p(!&^wwr)iZ#>dzU9j6TYUF7@%N(r=3lM5k*
z$w|JBZ0cqO>GwrK|2ksFK%`uDO-dgn3n<#mM159Yc#FZcTg)9@X7<<$wWCKV&(A|8
zz^H&sTheGqsZ`3#F5^;-zzYI)wuY3eWnOscH59+dlaD<{u~0%JApDTIsWOGoC(;?4
z>l?gx=^~rk>#T3@B88h8HC3szFgr~$>eIS$lSj%S|G^g?r!iM0i8?e_-=##E$Im@U
z6U)@`b3FCs=Q(=xC_n$V{~xUW=Fd`0BWB7z(=!DgdiXetY_a}x|CEAEkfj+$&D$>P
zcn=Xm@a$)vW%-B;U<QLBm#<yQN6!2~_vzn^a_ik<OoG-DV>6_1WyayC&&w~r&Fvd~
z(x`wWLlTfAhSk*#&OdV6%_tn7vK<5VffF)LX-sCTjh#E;#)k+T*?f9>iq3wIPP<9D
zRH57H(HoAa*QQulKE~qI919D#+1g$w42vA`kD#??I2w?phS0CD-|VxuxsKP`q1zv^
zv)5y<9Z{`S@uWw&6atP0kftf4?vUYN$jzHKc;>SjCG*NclH`EyhJ8!n-iy(iBu>b(
z6d@%(OcughYmr#iR@eCN|NcMZ_RT9ieQuW7s>Iq3fuH{#0|&FwB*r=?F^Wd6>B-gn
z4#@f1-E+5I&m4@9_yok77^592TndH6W0Z_|@x^zDdJ<{e{guK)Yr(CX+dTEudFPOk
zZ}>IjbIC+Ld(L%;z7N|ObFCb<d!EnY!V;UeyToxszu#kbcaJnpsa8B@rWaUQSmEf2
z2iV)&MgXB-L`sj|utS=pWNAWkZ=2ieHyL$1H20fqZ}-^kX4GmDVHL&Dr&RRO+7d+(
zX<`}nM_j*lh0i_ju-#^sG|9)WD1|`e&Zr>O4%^3bLMPZR+<s<b4OgyS;l-C<X8pFq
z_QfhxOBtT;=qqmQkC$H%hwUcG820<f*?Bj|bAs(4wf%u%`v)sSim?gS`dEyk+@R?8
zyL|gQZ!+vilwo4)lf<yLy3NB6KY$Q!<j2_7+q`T!uEvea*YC6ShxDOPD6+J;%<lFL
zk~pH>ZgKPG8pdRl3n6oJ%Ph<+a{Sae_M5v13#CE@1M~(RvLq!*B6hacS>IS=*xF}z
zXP?cjKKp$)%TQWHr4%A$=xn(hG3X`qdjqasz06~u`hrWcq`9rn6H7t}*OF)<%`%cS
zA=BDXG5`;WCuQ#FXT1K#o4okyYc%%;*leA-QbxUM5y~*`lGfJGUzCqE8EF&~4Ti}1
zoU@rX8T~*qb;cM{lc2ykb;8GJ*xhRphzbQ&Am8GP5FR_*E#f#OECh!X(RrR@nM9t4
z4q4=7hr^v;0!1M#;RimMPS|hmu)VX*a1e3gNQH8-f*1H8Jx)LH7{y{4-z!k6G#JKx
z(j>u`g3+MI{_Zxtev50XS9$4`*NHD&rQPhIwO}~x5ELwpMu_Ku9|Tlu1)8lbj7c4-
z-e{LqbnH&2mqt&9Zwl##DI4qN*+}Jj`(5jLM!8buKmX7F4_vwO9>4V~zrd}{8-zy{
z1uRBNw1zl#EzB;eA9Iu3uw5x-KIX(u2<S?>OJVcU!FWLG#}+FwnSpUB1-Fy2rW6&*
z*r6raXtTZ3WarW~s*Cf?ojgwC=sdyV61G?%3kz5c)(20FO9~!wlwy;JG#;@qH^c2)
zS6I7xnZ+YZ6sk39l?p-V5p{cPZ*S1=cj$H7bdr?BSgM79qtmnaSbBTg6cf$UwK6~Y
z>?szGOcCe|6%C0@Od3ov+})%)JIy1Xev&iipTHn_{>T3`H@^ETj0QUt(=LykKS^VF
zgRNiv8IIR2@5K!w<umLI@IJst3w)oiKL2G`Mz@xooo#O3TEqUE>9hTzV)8w-Lbx+m
z!Y9o%@4UN;N10G~xK2=kz-Qz3K51eJgYj&)u~X+F@57XU@mfHM`yQp7BQTD9J>@eu
zyTGusi8d+w&22U|c8EtY#o2&*wZY8%A`8n$ITJ-lq3~otsZysmXc5N|(ptLh7PoF)
zr`6fz%C&23Zwwd?G;!QRYl~!nu_@)mCkWjtJZlWMZe7Dzof~*}m&VOV<H5?pGAYSw
zS6jjYfl`w_rb(8vv$xMvPd`PW=yQGJ8m-+C#}^e=rwHksi!_z_XIB{OW_x8xJlT|@
zCqs}yX@ku<YQ_pT-1QT3>>ReIE;v1e?fcC(zNk=8AtHA`1s=P*ZIUD-41=63?wq;z
zVY_%gwvYFKk29V^D8fR?b&u+VR%@5t-5o}wi1JFAQnBpDcu2*`Q|BoZN_Z-ySg8}m
zL((M1Q<7-Vr@6mFzu#tc?FO&D@fPtV*JY@+U^MLEhn9LHaM<qqR4N5p?H#Pi5Xwy{
z7-PH5`JOiK=(fh~_!K6$O__1nuJXB|I-^vs^4I_R|Bh?duJHT6`77MszR9T-=ja&e
z9GxV|9IxEuw2VBU8P7t5F}6EfqO8Cm@51&UOm^R$nNOOHCX@+@(}$5Z?CrG(Wfdht
zV0^9~+gtl&smAv`7jPKIufF>!KgiaPyXGzA-0C7>5RjxJ_V%{e+uLO{jA+yv6bmJk
z?@=r_I5T^MLRi2LN)$^q22mHCYLu`H`(0Q0>9x3g?J_UE@*2^_8?>8!vP>`-bO}Od
z>oc7a1Ob&w$ll%t`LaW-aq|VoXX3tQCV`f7xW|hJvYhMac!f!pGB>}#-~5|@legb~
zgWvtNUu1n}lR}{IeWznuLmcIc8OD7Dxoe@Ma{rQN27e%#x-oHA3WXG&BZnffR%6L<
z>{0<}iY39(WAlLH*JzWWHTlntw@KKE_kPE`8tazk3SlQiT%D!tZ12(B?@%lj5F%wX
zjPd<|+1Ul6C}DT6%hc2?aR$weO$vnqi(UagETL40#ATi8QwHT1Iqj7QN>v(DGu+y^
z&Dwg4%QyCEOotTW6fbod!s)3g8jUG8wLD85i5EL*hH$>!L1&!rxz5h<!5d@A(hQkp
zWSOQAmN<X@QO=${!`jW8{QlY+gMq*k7AsS-)G{25Cey{ooUAcHd$N;ed<pYCfpD-|
z<dZ7t38C{|rF`%rn$X4^5pHZlW-P`?R3@<9lx}yO-r72anJUu{oT7H-BwnfL$~xm6
zmD2Hx-ucdN^3qE$QJHG+%wy--+uz~pySJ&=Y79nQx_ZQLG^Dw^gGfh2!x8-?CA9%3
z7ni97A^p}a+9VW&<TJ+>`QqssD<w%yr^JIEoqmt7SY>U$$5dEA_G=WS8>ebKpSja#
zSUho(-sVLLt8cP2vb0|QElx~J2Ej45vxKef9W;XY1Lo2-sx_W@<{1PolD=`{D!V(a
zPvrgoQHV95Nl;=uLc`s2ty84Pu-<r1mqAK54L!?Jj0xT3`OE@2m+CGm#bIC)Ie!Ro
zFt1AMgkEnzlB8H5%eu5%Ewt85P0eDhU}tw9kQ9qmf<RHK)+yKLDb{Dn(gaTmX6Fa2
z9DRUjG~oPWJ3RQv1-|#am$-awm2N*}YddCYra-1Gc9h^%JQf$H5z?(VvKj84=b5Gn
zkM2;nyXSj3#lYd7F$SG?$r)o=US8p^{k8v?e!s)t|L^}UuYKz`nXMbD5|D0K^KckV
zxUduFO}UL8Lb?vyga39g`f1ErJY3@Xcp_Y87KIcFAwC$}3q`?^BXc;OT#oI=1^QMh
z+)igEBSOZXiyYgXqo40LwXQK>cW0kgt4pC!K#G*nFv0f%W~S$eqlEotmqufTIE7|=
zn?e{e?-lUFB2xKCYoTbV)n`#&ffHT{zf_^#nBw-<2J4$`uC6wzPdRMYDfnJMW2!-;
zUU%iu9NUF(<#TKD=C4E8?&W;o@vKl|ESb#^Sw@y>!m!AL51!}r=?A!ddyVg{-)1-f
zPk@yfS!Njy<NVC?`-XcLo^lfc^kHll!c7jqOul|EI>#p`0;K|yWP~v(<L+Ch53U?(
zOa`8Di5D_zHJ{}pvlB`}mQxm_^xV74ar5wKj`91%y2d{t4jkGvvutne&}nxm6pC0A
z5y!6iX=Y}QL4SlXJ*K8+8Ad6)`%Nm98uJSUg0SN7$z~Ldr9Nfwf--0Q5@ET*%*-t7
z+nZdwvB%AgF0=C?qft!YOM+_1%*-_Pdc%!_OcS^6ET>b9M=`;{SR3E>k=|ijZ^&#)
zmT9t7Qz_S2dSZ!V$BuFJ%4OdA-WJ25pj@&TGP2Y#7>o|^B|l$>0iN5w@ezZzkGT6a
z|4_b%MYuUu!o%Q}?t<gRR4Rh6{;8+=%;z4#TH|JQ<y~YV-(BZ;V&s#W$A8HGn<E=z
zZQd<vP%6Oli)?RqfG)*C$<2Qe5p#3%gh5EP(qOOIq1|Z`cu+4D*vJy*mX-;F0^MGV
z&i)?z&3%LlDO4*QJNW>;ZlB(8pTxq&Yfbj{htz9Dbea<H_eirI=b!jGHZ!DYjL9;W
zUPihJ^g<xTnET)+K+AmGiLtJQKhqkL6akz*d5R=UxV>=`6(~~SF&G)DCD#gWtU)Ik
zqu~&n&q^}Z=99MFp}Ub-uA9jByu8!UpmQ)Cuao(>QYt~aH9~0R2Dl1Am6SOp@uVO#
znvoC`@bENlB6yS%ukW$mYO}k&iB~H#ckCf#7GMJKyaH;}WPZ3yMM25)dHk`5>5P^b
z3<spqKEvidgZ(yGNNr4}1y+@5G-_1JWqSQKgXR{CzTjA`$dP)5$7bMYQBY7YYPE@y
zm}sO4=N8$!vPxk_P%2f}-Pz!jju0xqE7n*z`T*N&@51%h5&afN=RHc5Wxjnm!Ynsv
zzjFg?A-)%9NWzH|$2f87l>7f!UVrU1hJ)y1zxzK*F-E&NheEnxwE0M0Ycqt%vkNlg
z?1M{u@yk!SMg^Vaomwg%L=84?7Et*duyJ`d4^v!uA@At3#(?q}3?jC+_VGL)-wO%+
zGNqEo{M-UqL1SuxYHf-*iRg9rS-*9Ke!s`m^o+~cL=nS&mz|w$;zXkb)EhILJa&|@
zXt}l3<-0GgGRg)V_dPr*u>A-jG^ZbOqZzU^!K7}vuO=V%D<EIp?miFP8N69YvlNn)
zBuOb3D;P9`VINfp8EVB~00l1trO7gjwlQ%M=jE5N^T1@U^P!66gfp9E860v14~)Ub
zI&qg-_MDNGBFM4b9AG<=az*gPuRO^!&ptS*JbWOw%Q3DW1nh48+5xuP3APv5-RS^5
z3Sp5-rG}6ZbF*^<K}e-sr`hV#?zRa$OSKqcbjs}fB0&(+>v!n1_GvYn2rnS4R5)_{
zB>moi{-{Z6rnr2gMYB1gRxO~jjHJ~gO?y1_=$EnDl4W;cyAgT$)vXO1PxqHdj696(
z29gvh;N-CrBsyhdYYpXlq$q}wrBc#JrLe}3rWw&_IHAwj{9TOU`d-+cWe)SDaAbUU
zmwpf;Q1}P7-jlhlFCcIf0mtnGX$4bLKHvD!XL#~cXUVdRG|P9t3eTN0K<>bg$MPJ`
z;E!$nc-w|G7T+%-WWeS|3n{vkN@dE$I%$wHJ2Ok6P^46>v9+^Dr`x6!3QRIUrxA0D
z%lOKp-)+*`+ojcRp@I^nYK;@8&(Usm7$zO&M^ju~>$0~uqFOF69Hm5Ymn7@)#jl<~
zrzu$)V=|rVgPJT9kV1d3tvAlrr%8$@6=xqfOOnO3yZflXBesfBBq<gQLOBQCBq>pp
z>jUn+PxQg$+fDMus1G9_$*C)spw|ZLflvZpRZt=zvm>lEU@WDQ<%j;%r})AbpCl|4
zNVOvpDleZ?CB`IQ?*Nn#IT<<U-HeID;}oq_APWSAI$8_TB*mJTG_}mlH&|Gh$6Cwx
zzVo|uh6%5|`34W3KE=6{Cpda+na1oiN-1`BHo0=~B7;G~jrC31!#)cu%gj$zSy)(R
z5N{C%f;5x(evx9$5&nZhmGcij#%MTnPMtBx@#OU}@l)AJ<Tbu<Zn?1qs~u-X8(N(n
z`}-Y^pE%0Q%oHn&%LK(D-BE`HcWbeE*UVth#qG{@fG?$R^Qr83I=ee?D?w_&NQ_Yv
zev<kGE`G68p|c-hbpob@v~WVC1Qt9iNDw4eqh%h+2}?mcUPv<1bT8jx@Azeoed;;9
zu!OG^N`Sub0>zE%JQ!rG@9xpsY;xj}vz$G%!|(pquM>?7%cU9xzev}l_+A+$1-<Sb
z@!mGaq~YAO&sQEk&haDj2%FJrHfeSDsa7f!rH4s}RHtinM<dS6g@k8LlKEBUmKLxk
zMOlsU0;U$1xzI^jSh7?rC5nZROT8kUf}p#-Phg-w9zJ_0TPG3tO7hUT2dOpOY_yB-
zz01%2%b)$gaVGyZh>|HL@mzXA3OrdsiU3W9!max91U&M{GC%$k&ojR;<08{6L&`&C
zHz9IxA7_)Ln|~)@vI*XbvCT&aiSmksg*wevgw-P|<uc`RnPRcX{Jd+%c<=2uxqM@d
z)!S>VEX?xA*|VTCO634iD6_Y>$@<zV&HWCe%s}8%ovL%{p@$fzJ-S1;UM@{Dlve~I
z;!LA0oH}uicsL?W^K42!8*W@SM%ja^0{P|;>8KW|)!58YE<CB2nVuym26zE1E-WDY
zfPNG)Gvm5eiQS&e;h^gRx$%rQnI~tkc*ga}3EU8Koi`mB?Xny~eKPmFR21|(7VEk7
zWxgsSvjA;JE_G`x#iHdaUwNF*e*RH{us~)~bf({r?cz>sw|Wfgd3p9OY?m@52<v1<
zx)DfLk!C$+=jzPPxeU-tFMNmYDB;bw-{#Dz6P!7AjFlrP_30_3Qta++asBcY2E&xK
zjV(H(0dva>%r&aa&n+>Cb_jh*rk&GJC}e1BsfQKLojp%99Fk_qA#C>!uw7av{)ub%
za-FXeY<E2IcDK)dv&*q#D@;#Kv9z#2P%P4my3FN+b&Yi!2L^-QA#8VNt~`-*^JJds
zJdEuEqvQnJrTlOP{-cVg1jkR-xO#P)LEqH@g_Hzd1)2F6(uu9F)g(Xq&1ZS$+4CsR
zce~EDb{jbEu=R&Cws+fld6%u1$e@Jut7s#LlLU)LoM?_5Szvl*hBQt1-QWH-MjBpy
z?NvVg<P#iSoaf{z!_?di)@U|1Z*lGN6^0|tmDM$3H1kV~%+{;S&MwgG7((AYPgpGD
zsThM`t}?^vQ)d|shGY|c@Z65#J8b=dev_Yvu@;*T)7DzE+icV6^f__z2y?TuEG;ac
zf{<P`Vp_UGHwJ8`8T5OTvbD?Hg7R?F)2y@gzWM<;b)`yyPP>Z`h5R!5ND(5;2x%=r
znDYFWALH|1c!F}JfDmq2qp=pnLHEc9Wst@xNRe~5CLMD~tdg96;<LQ__RH+;Y|!bA
zD3&Zo3k9a8rl?k{Jo5NsXnYnH7MQ752$kFMnq?W$pvTtc2APEuXU?(F>=L9IrLf4w
z3-5AlWszE?NU0?051rzcN)@(t_L!PF%Hrb6gr|0|a>uv~JN|4F1#m|W(h?hZ`@#io
ztZj1P;(I*&;BiXD5@9i<-?Su|B+M(T#zME(BTX~cFk%jdo(kcjoylr7>&ha=7(B})
zRs2ar@WWm{Q!FtU9~w5|xCetQk75CBG?}tE%CzI1C<z(hrGn<=>kMw+qTZNBN=1;3
zDE2lvH4}0s^Lcfn%j)lco7HzNv3TwTk3ISn?e>Vp#v+4GmvVE5hV3z1(3Ix_s$!M~
z5f$HKF4RPw9hAsZ{bD@%J0%DMN|hS7u3lkzWr3q#{xWeEGBdYC5=Rt$$Kj}~oTSw@
zbQ=-NQ#HJx&W)Qbic3qp{lX1|Fm&&ju;s)e@P*|3!{<T3x4-rK{Fnd5|B2V$c>80&
z_diORovX5OyAOo!ELDh-Ax6gt41$!?XJ`55k3Yxq@+?yMB)JnLrE;r<@>2WV^olV7
zT;zjQ4;-ZYd!%9c=ouPQ(+v9^dczU1Hk>}W!ra^(zVDNnj0eu1VP|ib<;5keu_Q@M
znkGmDo15#buHB+OwZQAIy~19n#j%seFma!$TAgxnNEnJdBc^CH>I_CBrl*%Vd-g$+
zI3dmQE`RAEWo~%KhMWUz8b80Zj#Fl_WY+TH>u+%3;x(Rn;t^Js8cfw2l&WRA-2s_a
z2-h&`$lbjjna(CW*)b`boNpww#t7l)%35HwcKl5ahMs?1DV0LHogPp?Dwk<hBEXmk
zVJ$(B^0_aZ=h@FbLaAIp2!%}#I$kH({s9i=WZs+18!7I>_6Hw%nu`}+V}EzkVSCXK
zhOX1SQmOFJ`G?WMXKrqu=~|gkDT>7+I?IT4pPlV3w1s1*PP5hO;b$48pv0An7g<@F
zr&cLYEJ_9=hwa5;nVr3T8ucSAEG%K{AAs$7C*!1HA?H{+aZX=q;N6RtxOsbv%U3RQ
z?#xk&g%Uv_V9;{dUT_IYj1~0z{mHhO_hCDEGsHcx-CF&LVf(#gdbYw3fAcB+!QZ<<
zzc)mPB2r1D3{WBfogp!lN}8{I<8hw;^urX2Zp)BKGv}D)HeTkL+p!$}_X$T#*vZ|W
z;DHo8@zfVsyYU|TyF0WyLn_rG<#L5aqfWJ4<x|f*jTS!hbF-{0%pr{?3=3pwN|FrO
z+}uEVKBv!}<?8KCDnSXMEmtpHW_fXzTD3(OLY6ll1wnzmW{2a)mzkQHo;0K!mhtnC
z{glJl=El!Br{7A5vy8XjdW)^CJuY3o$k{VTDV56j#gN``;8p>W^N)c}x8ugZaKM%K
zJkMp%t<Ih0A3&o-uVt||Lu)s9RtgU#3z)1x;CVd#^eMjdl}}TtRPdBLFqIR4$C1S$
zr~DxGe3(xMOu`Wkz|%^>%CR$i;j4d^pZV#340<2WSmzXlA&rI`eys60vb;jl?_x~`
zLL)4s$q1u0GqZEd&(HIXuYZMJw@bU-rX(}wW~Uj%1M2mVBy(MDfnT7xoAKP|9w!Wo
zc~s@+&nz9_3Xu~$-5RjFm3W-Ba3e`-eB*16Vl@2XFaIh#J58phrg;3x$9Utr-)9gF
z2t4H$2?L|yfG8T}C71L-d4$Az2k-B0L-R>Sag8ZKzvVe{@TkMdY=fGz420|cQo=<j
zj&bAuX=FyI3@Rtf#v4s;zWgdPCr(nRPE*~zOd)A;>ck4ett~!#Y?g)ngw5_vTCc5g
z?&t|#y0}I#YVzoGnRAa-IM!HT7!7GR+ZY*B2pa^R!YYYxG)4=OXn;*pyfR4TQz%r3
zwI<1iOw~%PoPGc+YDA+p*6$IPW)N6PvvcTFvfGN7Ef*-1%k=$}X0MNqQ>+lg0)#bp
zuvwgWQY<a4@KgWnpYgZ<*8hulFJ1b`9Eg9@WYEhnCc)ZbKHefgi2}w(C?Pp=w9JqF
z=`V2f*dl)52Bs>jTp3P?iQzjun&IAt=wN8~xO?0v#lw$1%f<KJ;iVUTpLCR>L`Dz<
zlu9M0r>B{jU%<+c2hW~m&}m_HMy2BRB*w!&y-tV4r4<&JkK>0$N4F4y_1o8&o1UXL
z>{F|GwAu-2n&SIaHm>Zjv^Yn(Tyf>ce5b2CNYS|2dRgYk>p8$qoI5GqHWwvOBH+}C
zX%-gddGY%%((f%$Dwdd<X|U62GD-|eiOKx(et$q5$Jm^&H=gz7Fy_v&5Z2~p*gTv6
zNg4NE$6~GHOZhV2Tp<cb8xwdQPds^o=f3bM%H{G0V>@?YJJ!t~<N(_z{bh&V9a6A-
z^faIS;*aqQKl@W)nt0acdwWCb^*XZ)i)ibyvbaP%=s5zraAfu@cGx~OGsE269AEqL
z7wC0+bh{mjGGk`C!7v$8s|OQo5B(|H`<l-_d!8^X<k;@z&nNDU?fD?!$>+HP+bewa
zOAlkT<=1}wx7poo(Wp0g<gxR-{o>1v;^CdxJ{l57QC?b2U4WcB#-4Iigz+H$JN4`b
z<IIgkO3BUDErtVyv2JMdxGLpFSs4nU&vRdRh|higNs7fHp641xoIaSlY~X=D+`apC
z+*)(^Yf=b|RGfL}NuK)Lzr(-!g@1uCT|8r`l*@!c$n^9K^UFoFQ5;)dChqre>0sAv
zmBu5ivCJ<lGdDlWPki%77z_sN?(Nd5_{>Z-*zY!}RDA|RO`68ks&!g>EzUgfDLl_Z
z8#`G&=i1HPlWOCV27j=vSCnc~{KR9AGaSVH{4e}6t!9^nxp^LY{88S0@dctdK@}9%
zI8NnY&?iX}jByo@AoTObGEn9ar}sn1M=QUU6@!k%+7zh@2%(V#2n>r=#pgeLo?10T
z3IkT<g!l}d=25xC5iPA7205O;V)L{sf_#n8p=_QI*c|Ys@_6QRUjb|R&aeFoOx$Ls
z(I5~$p(;`=l#wzd3pBDe#i-k%)7?a7nm8KaNr}ufMNd+msxj55Gc!BG)O3}>s81M{
zsFp%{oj%HDBvHii<EQzNAN~nEzkoK$9WT~ZMUCr{;!s9HII!0FFk&ebLP?6HGO2c{
z`)5D>6rFaHVSk_VXV3HAYwysHht$g%(g?J)B-xPRpoQ{-J6hR=a*QPM6*|_sjzw!7
z$=Bh{LCN6b$hmV5vVP-5gwRx@Ef(bfC47`UXlWnVGzY~z87%~!RCq?aNIXus`QocQ
z@Y&BZH(iI@FO#CF)}|QlHCd`DX3IrVE$IgavH?Hz+zD!-q7)W#e$$ZM)g8*SMQYQF
z6#N45U`V^)LP*c8qqD|!5m`Z4Dx;;M)!wIAEKn{~5h?&<xc%n$IP=*b!sraC!7moE
zLegr-B+G)-ganNm8*g907@wid$m~7JslsBg{G)&TkGOL63L87yjyw5pxtL5Lj6mxI
z&-Vd?6d@8tqa5({uYH;mCzo@MjLSMHrAafB8zz8p7Dr4tVJ7bkcM-ZXiigWG<b$h=
z<vKt5lYfOMF<gD)dlY?5y;>#oLMo*yg>o5f6uMAmerkqvG@{$zMQTfuL<GKDNMEaz
zc<9_I()2i;ZkO5VGD=DIyG`oVBBNmhHg#omugcfI`s4XM&+r)!^c538H{-21V+Vzt
z`vi`IBpmOmRBf<y^aNnZq7g|FF_@j>D_{Bwzr5FG5baS8-9`=s#L0kY)N|ePI-3~3
zv6JKsqbtog4A7IWPpViSQ%WVtuq&}9Mfe4T$eba^FkewT`>6-1{z2F-J^<Uf8{5VE
zuwCTX{^T>yW3A<d-}-rM(qXz*&#}EgAuJ<h;IO^gAnJAL^tO?iVHA&CSAAw2wil|@
z>otbMX&Q|RgVBI6C{ihe^!)(}O&muYJ$jO_ef66tKSb#C565;6u-zWQc9f)0ER$&$
z1U&WB<McW$MuR2~o_>hSZ@x!A9#Ji2E;yE!G#fG+wvl6OH?FxxDVuASy*IW`8Vr9h
z&YhH!x%nwdMZu^iFgC?kK7tG>LZHjBnF61A>S0Pnrys(4AT3gX)+W!;Su7dG<|TK;
zfnKXI1%tcv0e6Rx$iU}|U;8m6@XGJ~5{i-22N_UMMGA!q;G@GbvNl7%y+?nzk4Q7p
zc;o^Tttt76^6V7TGgC}WPtj^N=nZ=m3S}zA0Fx%5Q_?u$k@HXR{1^Td(hIPYCdC7N
zu-2gNY3uDhY<-nDF?hbu=Rfyp`rQsYjcFb}d!7rgzR4g-DTPiSjD$GpG8%O-*gI`K
z0;h4#)_)lJNKW0OkDuaqHuo?|h6u(pTMQ*(Sxom3+qdyfA4i2IPpjTt`?r%4hPCFd
zsOQkP-9ujmE;<gpkk5bdYt)N^-COT5Jv{|LnkM)u6su*Dv_}^A$<hIy5@g8;VI8qk
zX9+qQ5UMg(dZ=QNv{a_m+DE6FEY=teQDPaTmcRNR{x#+n7co{Njm5}=9gyy@?Ic|~
z`P%q1<PnNC1}R+UZMWOU_dP1*65aMbX_8`$8)ee(8?uH%Tj!V<qv`d#RO{1#o67B$
zjN3`~)%Y34N_Kdey7~mlG8f*t?LLgjSe9Kp;h}`Z6FI*_N^C~XC&+g^8i_>XBk_bI
zlV&@6>|eUbgMN=fyG<IqGRXAuGX2dBI_)l7dojy16+$l~jRp)0Wkhuv5h^ZU+@L-;
z!`$K&p%;)wBW$J#gAi$)!6<|!hEYsVaPtWhW9j!@Q%w*Q5X!?#B3iG#hI;Y~B-sex
z6!2t#l_61-;;VpW9MK`A-|3SSD)a{(T)(G>D=Hu8^C*e=)_1={W<RMe#6NmDb8d#$
zU)sm&6d4wf!jXZM(5!d^mP-=f8pp2#Kuj8V98%|mUz_s~?IArJN9c!)eEuuOaEX$N
z+1W+@y+8k#`2Bi7Yx_D&^Ghz1ml`mNVxh{&ju0lrm=q~2HjU9~iq-DfL>dzo3KW9^
zVGyFEWqWrEYeha<2duVqIz4{u$NvnEJ@GUs5E`t!#~tTTxj6qERvZLx!u8i9rQ-JH
zHooU^{P<CJwl-MPn$RzR^5~5$Gc|_?&LG6}do2p(8ek{)TuPPS<%1@QgMUBx`X^!F
z&p&dKZ*9MZNwvdvBnpEkEDLE9vAcmfc@!1c3ARtNMi!9yxDxJx?T6)quw5#GAmB5f
z`y#c1V1NBGQ&SDVI&9CtuarpA0Xi9wWy2iXBZL)*9NV*eCZ>`es!$*;m1wn_=uDF(
z4%_3z5~uJN{?cDzc46L?Ud5qdxz@RI<W8v+E)$ji?@nyD2iV^0IczVNO7uG|H-y{f
z*gi1my31sZvAy5#QK?UXIE?Lb!mG-a?*QBFLAm>r!1jBI))^0+Il_Yv9pk&(xvei1
z^HRB?Bn<PT9oXH(JF<uj0<0L9;00MW&Mb}}fBa76ow{)7-XE6nXH`m3395YUPkocw
zs!x0S1`G2`SfiajXhBdalk~e}aUY{cC;?gQ^g&wTGPv=GLZOH!JyfYgJS?);+(V}t
zooKX%VWery6#1+F;eSM}QJ-X=t@$XnKJSLMZtRdThEAtPAq=Qit90AV$u=!51pT35
zx+2lWtryffqu=Y{2O;(l27t&rZa=(yB&Y7`r6$;dn<Q-%7D=EZ7Nwyc!`h2)aP-7!
z>Ss=11B0{&M9FbY_|7!^fwy~@qjH#kb5Ngme>|!1OH(}d?3c*SoFEx?bFN#0$lU4;
zJ?gQ$c8#!9#99~qE8!6+pN*Y$w#x-ho%;Wi_h-SD9rt-J_{+?-)}Hs7=W}YFF&E~E
zBtTN4WXqB*U)c`(>fY<2JG!HzBl@jhy8E@GJE9|Q^g~C-jc7Zr?Q6>#Eo-nykpuw{
zAVG`;)O@N=Rh@a<^RR}@?1x-y?^6X(A`Kwo%1lKGRm8!mv)9V}@;`k4?>ptDON78A
z3B692ez!{!MOX`k(hUFjxBvHCc=Ra{0%bGr()(ZXKp0dEb@kG(8fp%uQ&jnUo}fI2
z6p+Lr<x-g-2#B>N5tc9%BwC^|XzP6Q`n^8I@*zU)ASfRGc`%Scmfef{p7rNhoo<3c
zFdA($mkSYEAUumF71~;qaBWwewqCt-YRn3ak_IIN(t?a5nm2wz{&<8JD1`CRq7Mir
z=NHJAM_Juk24W&B5J7=lP$U-wtZ(g-D+J6ej#DU<2)iwW5R}G7Dc~dr3>FjQk+~ek
z2>MY(JL=Qxbto05ke=gx^mIa8tD~C@m=3WbaP31CfKhld$5z;9eRrR}^5~=KS<OIL
z?eMZ`4rTu4SLY5d-y1>LJSMha6bOZHL&k++qSs{S{SP^^bOe#lfiy@mqP)1T&FUlo
zTx0(5xjxLJJG=(;0x9v7Vs_y;-~6@TqP4Y*j(T97f1=eEFJHim`n2|ToM*CzIEj!}
zl2bl=`+H1Jj8n?bIJ$_ictF3~CF+NS-4LUp8%myk{008@-}^1RoR6>uC4<4aeh}0%
ztFFVXqyfmMtwqoC$d@X3N}=Nr&-2phU`-+gao>_?<-~^5$+=FaOOP+Q>feC7i<8y(
z3?AR%3+9X9)ZM;W1Cv8w!8kp^K!;3ALp8Fjz56i>M~>s4ID!q-eb{b{8RAVE_#Ccm
zJ&5z!v%DYMU0~-G$9VY3mq<<?A?mhV!fO*qr)+h<#s1m~LE+QbzO}o-ZZXfX<HrZs
zZWB$v+av7u(6Phzd~u4u{}2BeXCHbZ#das#@o8+g_hGyFDcBBzd_G4|8XaJJu~;PV
zGi--262ynFT^mC`>{HO%O*@iJI@?2g$+cU>0NaNF&lggj{fxD@+v5NG|NaY_wYIzK
zHDFv@x~TgsN+@<>?!Ns#vx`UYj^-T(SaIKe$oOWT$w57Y15ym-P445TNr|UCyy`Sh
zzW7aaW0k1eauqBaL+tMQxP8FE#y#?-5=Kj;0#EpOLb1QIO|?>F>BuoqjuUN?gkHNt
zzthDe3C6(K#8LjcfBt`D>Et<Jc-J3J(huW-E1nE7z%ndexnjPcP%Ki(myr^po}^SN
z;d>sj(Zs^xPGa1CFzH}Lzu!Z7p4$&<cuoct&M%|=TCaCOYLqc<!Yhec6cKYup=?0?
z-T}AX`Z148kKzT>Dc{QZy*&tmPNQf@{1akGv=xK$rc6hn-9wW~Aw3@>6}i#`p4E(x
zw%OafN2l9DSZLOF*toYsUP*GEk2Zp!R6u!(IMV1?bNlu!f?UAN{31en=qRGsbC<@^
zg()Uyeg4ru{XcO2ktZQ7f2QZmaep!(weCXKIxW*%mo;0m(S7$45|f~fbuA02AkDaJ
zmZ2G!(e)!k7&&PHEiGC@6!wYwU98rqRLf|vxwyXtBZPZS%DU)YeGw&kWARMLlB|=r
z1~=m>$mk_9-S~!#2Wim=5-so~2x*a4P!yJPV;+9eBT6(Tbk!AKKomxZMDxtWqZA4S
zn(a3Go9omY4T55Uv56|P)hb@0NYd?Kr69-$&P-doztaGPqEF!G&{0f3?$Pel20SmT
zeDZjNMx)}0xY<U;30n4v;*eZm!M6zM(Xf)n+CDZ{z$Os}<pu<#`5P-J{PV2qS2wU2
zl(kqb+@zZcIi>`8Q)2z&73NP~;n>(Sh`>WK96Hb`iGL7$2!Y_xIhXfO#0~!U0jRt0
z<9j|@`i##WXRPECm>!LTeZsIuPPJ)nN9?ZOrM<V!`1CxU%Haojl#=)=Ku0mlw{DOt
z<*AH};Rz3&B=kBx;(niE-lI^+@k_t*oBaKM^xNbLrSwd>@gL6{wpg-cZ3wPIfHjV9
z;DBDc7OC(1<Z^k}zSEK*$T_zelc2HmLWmNVp&JBRJ5&28iZI$q$tb13I>(gk`M8Tl
zq<Wvi2_)O$)cxmKjYbcY!*-o$lunoz5wj__H`Z&cyz?FxCP(ms$pN-2rO?I<o7Ga{
zaX+?a)_~%(uwA5>dI;Nt;y50fv1*6??KQgHHo{n1ja{}@?~zl2!1FO$5R~%HYczJI
z>UZz1;OBg%=jI34-tYCm!qVI%6VpDw`A_~iXD&RBD-Rp8`sZJb?WBR<a5@!8+omRU
z(8#dewGki0_9%AvVWdTCOB9C0{jPiG_hGwBlpn-*DP6b88kbCe8A~Uyy%7>8Jx~eC
z3KFdebi}-jnNc3n2JGJ4;O;x`a%Q56H#Uw!qSK0Nh5?Y-nH{RuXZt{kfy7b9Ps{d%
z$o7NsFw!GmoyJJZ*hrg$?R7fc7HC6lZ=Ll!w-_lEv0i}IP#h_tjYk|g_m7)5Z{X*A
zCZ=YwRuG3Fy<QJbDvm8pA@XDV_W$<(#F67?-Op!D_T&dYi#ulwFu>_B%2Y$wy(xzW
zjy|CongSf&Us`Jt?e;@2G(?d*L)yB#KJ4`feBbSdT;Sp+?I<7S=Q!s-pQ8jIt%o)d
zUVDopdYi?-($gtlE-`F<aD|D5S(d)~BD!1|5J02EaYZt~mqGz*m<ch=+OiBTt3qW0
zj6!)>sobCOQRP|M?H&4B&}=s8w_EIQ?oh0h8JnLY?#I+>4ffg%_V@PK+S(vVBC3@l
zwf%kkAcr61sMmMt^jidl0?$7GRgRy($f?JkaYW>7q!=@Jaf3_>@RaAu?;MV%1tzoq
zar5pZHr=SW=_ZkmK)e2Vg4TveLlj$ldx%35Ym%t%DjqI%PumsQD3y~DIULkUlR8g)
z0S8|qm@OIR%X#8{gcnDH_SIoMqCr(dxF(u05+RM#^%5{KQ|9bLi>~b@p%-`QwAxe)
zMRZ~?z7w3*Qc)fmB_HJIHkw!|C{;$l&yj>(_nE9F>Vz0$k;d^YRiT8hBz~^o^yCsj
zuh%6b>GrGup&&{qdJ@rZqvHszBuN-iE=o+I=v&Qd98vG|>E?<MM$FAEAw0|G_6AWD
zfBuL1zw%{p6X@uKxZh+ns&T|i@P&s!pxY5EKX{w*`6)&&oCnWCTJWSIPA|EEEfA8!
z=bXdI+W^F63Z8D@0HrX}b9-79F~vptt#$h07L8_&u-BnmZ_=pOSUP$X6Gv=rZjvuo
zh~tFYcUI{2I^-h7om<NsJ$@1)eY))?oqh)|$npF$Pck}toUi_^{{}x-Mi^ISvkye=
zQlvuxBK58&Z7B<ZbG>nVF>5V}jED`Xy_6&?-RYEbqjikYhA^>2kwqHN%AmC%iDGo<
zk~aZq(j*vs8!6Jki1c!okkPNaFW{gXn_+v5*WF<$*=63h^iv5M6I*sZy3Y9AGz%|2
zi!PN@Y!?pOt-%|}RAm=AX;ns~LC*u&E;4L)`VCg5m?eB<X@*W`*I|3BNx#$LV0)Kh
zsl@2q0!crn-fE&dO%C?=+1c45(Git$kw)zR-_PL(0rln{ov@9c&-2vNFLLC}W1M*S
z3ADvwJMs{@|Gxy=(;#f{&K<UAtvKf$tced{yRinV!Ni&*4%2qEYo`lv$+#5`+tVLL
zWWlen@C9M}XD^F2k-LwrA~A+SYmK9Nhgo0HGcNHmp=R|*A2PNu%j~mHq4Q4eN=iwr
z6MSp&2JhMJhrxj1Ab=b0iTmz)mF)+G6d+^{S)Qle+@P0O8jTtqTVro?o3Y6WD$`Sh
z-H3X#$!@E`-p&r&JDXUWFj}p!x3hzkj%QnM?9vU}6suLf@|ACL_VE{3I&%SS%xCTg
z7?4WvJue-c878E5pu3k=^Mu6+HP{c1+djPOoiMW{Ni1>VeB0eVA@z)p++MII<xgf+
z!1USt4D06#>ZFw@t5H#%MRC9x*&rtr8x|D8@rt^dyRZM4k(o)x9(xSwdl=F(?4b-j
z6CHjqsC%#sn`Qe*iSz<I-)G>lC<HQ?$8O)n&&8C2JZ4s+q@vf42?}LSOwNN9OpI+a
zIWbPHw&#3cV?(RiA}I9PZ|tK@!jV%Caq`SlOfQ^FIaDDQGk9rESzHfZV8(06Hs-L^
zAe|nO(aw@cN)+iOC*5HcseMjX1+c~v8$ma+R1_{t*N!lm#2Jq+wI)fm?1q*yX5dSD
zSmNM|Xv^fi9KZanmzeP%V{`ci?dvy5>RqHU2;ulO4h&tj&Ip0BkXV6_AV6b%$uqA!
zO0M7&hY?AyPr>(TY0dt@0aGIrNYBHh*{z9_wDMK}C0q-@_DNa?bh~cSEoV|=cm#Mp
zMFgr)KuG13*&*dC8lo_v=y_-*q2I$d0u%KiPKc5oCW*-BET*8STfzEnjkbV7py)>M
z#G}vf{>MN1g$U>Xux{ciNkYb>4bF-^Mm^YxQ{D`g{&t<^@BWZS#>bJ1i)p)U=x6)j
zId{utwTBJqhrsSG3#5ni0%Y3G6GEW;5llXZsqK;IKCy{tw%g1dJxab(CDE`nK0&A3
zXJhp)<&iQnfP4_3q{sf=KDlC^y@M^nxI<}doP~w+TzKqxa+OhZ5|J6r+w}ZqQ>Vu0
zL2Fs1ivNd)J8Vh-*IFa3^L@;OmTfvE2aa>9wMH95Y@i=o3IM6~lTI90C~Nb%X}9}Z
zoX%5o$XWWVisygnk_d<Gaf5kL<D_hm6CPXE1rr*B?Io=K@Lj4?6O2Cm&;Z*VEg=ou
zvgh}+U^~Lipk&zYd55q)K>D+=vx*-il!6>~+Ev5){g@zM;^@g)u!8ZiT_z^RsMilt
zY`3)AE&RMstx?0+gvFERIDYa8rsj^5U=cdP3Nr|t|0-;E;<yZw5}RSWaagQy?^}wj
zYLnu;b%O)OI&AO97M1>8+PL9}Phq>ruwA%km|4-m;NK(N-SOot=X4;gKpBmSYaC^d
zQ>u*+icmYngBXKpgsgn;hg7B}sh&NPzJIW>xv%Oe_^bEB;P+Cdd4NKCIXq7d_JaxX
z2yY&{bB`bg1uuYkM{5WpjbEyAYG#Qf(M*huF*P+#y?#IvC&ZDV*=it_q}DjV3d_ld
zE^_?D6HLq=CAJ!2L(Cy=ps+)VgVBfggBofU4(@t8sOUOdJa2f{+o5LX;A~q<sNH_Z
zlkNxYWJ}@)_JbJ26yj&T&wtLKPLZR>EL$A+6MXBr1Bv5ZMFK(#`Mo-~zW+lWo}8w5
zWWhZEoW7FG<<jXG7s=i~=6vX3$)QJ_6%wR}@O+f#<E2F^Z4ACJcuL|0ISON=RHvue
zT)D@GZ@<UchaV-(#dvsxp{CzTsBbs9eC0BaJ^eT*&Yq$dcF|U%RF0G9o@DOWd9<~d
zC<dEV5z+%~GLFrVyEXu_2VOKrr!83b*|W_#AP}d&G8qFmD<tSfhGJd}fkQhV&XjtN
zQgTok%Qm1z;#u#$vF9&}Yjf_QDIWjE6PQ+$(y4j6XHQVOahHwjw}|%-u!+XU1#&X0
zGYbcJ)>?c7M^4Xi{LCEEN{nYwzDE?sRK~_wy>XjDP#_3=C#0$*#AzuziAlOK=r&Q*
zBGww;^T`JRIlqANeMHW~YDqs%pcmogv;8H~>WBra@p2xCfnK*s;7d%{11*VdLK24r
z5{z$HOC+t$1_FuI8kFJ(Kl~nz&i?0qmVZ;0M51iSXuQekpvxo)Lik9W;B<_uE;X*-
zX7!D?Ir%H!!W*j&IERj7k^#3upay$0%W@C%>#|=atK_KwKk!|aE(F?Ir1U{ba)lCO
zBO_F&rdeOP%k@jwIeO|W<JEB{3*(fI1nlf=vAMa%`o=n8H|F&DF|>*4h7nTwJoe11
z96$RcM)*Xrn|`&zwanZHNCkQ_e$$}<Dyy1gXJ2cFU&ewkLjccKcJ?6+c&#P2l5V7(
z_-T61Gbw|tvYaJ~nTA=Wd71H(Nn74u)HxSc5zuGGZga%fDYh%bVQfc`+i$S)=8ri)
zF-384Ztxk0*e?GXY|p4RS@4!&yNB|8R9cyI*zWMo3j*?^BUGlQ*uJyM$M1f~sdEn#
zHWNId2t!TSPH615xOVL-4?pn;M^Bv~j8bfuIgX!x0A4|%zMPpOr_mN{6sJYh6x*#B
z5Z&*?_Uz|;0Nb^3)``O2kL?5aOP@+M)rtk3VtXN%O4*q75>H|v6Pz7#%)k{eEfS?3
zz;?bo@nH~PJd`!$X>-JGvKSaVJTwwv6O@#MHX%aLTi@aKYd_?|*aW$mslol`0%Mil
z;ScPGEP+e^zQgyZ780R6r03yz9xA=-G1?V$rNGPOD36U%ot|dp@^x;0a)Z<79wuzW
zNGl2ZhJG_)Z~cI4H?Hy0D=#rMHAS=4z@YHFB4;i<%jonXiLuzoou>gW&Po{gR}c3?
z7N`yPa@vN^_^*fe4#T@X-4KK156j@L4}8~7H`bm1B#F}@nam<}hzCFGJO8<Yy3vx)
znIm(kD>n%ui?xaX7nN%%5ZXBm7#*;`yw07sf5f?wZy`!$H+dSxh=PFBH&zH-d(LWS
zSe!CiY<k$Olur0icu3{BN<fkjCYmsE_Bz5CGzx9|Xsam(d1jZ6p*)|UP+{W883L6j
ziMkXv@6nA~Jo(IHoOt*ls<ZQSR+cd~L3w#b#%3^9Vw1?3blbzj8>AViP07m!D^mTU
zH0u<HUf{5bCXvQX&K-u{Vfof&@eWi3y;u_GTw76_2xAlCI3dv*Co|xr2F##BX45Ru
zWfMkbBx`5Y^Lf_`Up+^zRH7G!sF6`h=Zln=jxu)kIIACDVe95?yuCKcSgg@lk;6*u
zTC>JdC@Y?P=@in&cqTx29yTvA%5%Db(^Gi4fL^=hOw9cpQronpB*+JOr0-$9DnbPa
zPhf-u<0M<+C_yO?kyxa58L&t#r;K)Oj=0~)&zJC$n5fgHkk6C!I|Ra2^Lm{=GKolR
z%vR6RYIQMqMBP52iD@O>!TJ0}Sb<ecRt$5x&X|-~o5K@!;GTenz7P~6xc8HfsVprr
z|I|~i!e#{~j*<C-%UFenD~4DE#x=51E)D4!N2Qf8;UPSQ^xQ?Nq;p}xBq(bMayd|a
ztS6|J#_>fCG>lA)qpDN%B1<nxFu6SaD5BMB^6=A7FnjbEwVh26pj3{r$t8@+VYT!6
zbQ~!=<b7FJsmq*PvKC5KUCM$y$n3l<%DQ2l2hTa_W$!#OQ@#~~ZX_ssE)divA<@oS
z%%pl-G95t4s1@KaMJY9?<VoqoOTUP7K3eiQwKR*mehV*hvL!wSi9icikdFal0&2H6
zSpCUQI9>fZqUf;QB#8@-{#tBzLl}p#T__J>!6Yt;jiQJs3c)^r?S78w#Un`HC&-r>
zUp$4MVf*$P-KfpuPd?1i^XI6}%(8vwHr6IcFHdE33S$)L4BNA0J^f`<Y_~&f7e6z$
zi$k^OL)dQ9y8?&p{a6wQsqeIjFvj`CCR&5gsMLPyer!+Ml2Rh2xv!%4r7Qu;TBgSG
z99@`2UAcuy6j((dv!Bm>X@tNgKHHaXQC)hE6JPxr<O@h6oDgwdAcUXV$Js;s!IIh$
zV1>j;$4^5_Jn12opYDf*I7tY@kSGcf)}V#N5TmUoA5=Jc<{ZUxnezA~WAn#Rc*Iec
z{MsGby*kfNKf{r;XQ<9hb5J|LqRAI36sr>$BO!^>pzN@)bb8kh0)6ccK)N4fwh<2P
zhr!^Iv8e1T(hlx$KcuoqSd10)B1y?hX&5@h7)>0<B*tJxg2Y)aX8VDk@fP~Ig1TS&
z^@n-n+!)t>x5vFV-^EK!i6suWu#ngUi-#XtHhy%Ov4tf@pMC_X0*tm8BuaX4h)ivg
z$UOloQl|>#RLzx-Qa-UU9PBjM+t_D!Z;SndUFx+Otxk(Dj+q-({OxZ%N}<>zpDTi}
zj8rQ`O+#|fLMIJ6`};I%JDA8aGCoPMRHEPQvv+Vns4XMY$MADS;wVh3Y|d|09c~$<
ztp+P>I+Zod=0)}v(^T#LsX4MIl)mU;_KjsQA!r5kLqn7o<b2oqvDT($=`K@|hl0%E
zbWY0081&!ep?~-J*6CBnP$B|hky>I^Kz_Oem4K5|qf8!K<mS7VsNGp3jv!}KK~M>i
zvONCGQ6{GIbQ?{2{e)aTk5_QSaS<g1`5acd){==eLB8NBWLjX1B@%+1A0Prp#ZGL3
zjT1;THc%wmA_7UiQo@9NB4ZHgG?7U_7>h^}iq#2p)F)~-$QSZ-+8tz)5cd0o%{oyS
z(vbn%{TAI=qlHgTN6ww&3we@%bJn95r+My?bFBPPllIC6(ioIXTSGXqc5Dnr1keb%
z_1cf9&d%c<T|f$njy1-*APwaW{yUpiwKA@iaD*8U3gJ2MH(Jo#Y13@fXw(mAHk-~5
zSSO4GJsv+hK`C+EG!^*FEiTZmwV15TQ7RS4BT2j_rDB<jPhDhuY7*fG)EiA=V+cxP
z6f2|XM7zt~nDodFtMU>D=rV1;y4h}q{IJ1ia35z-btGjYWPeBjU1pJk9a0tg5yW}$
zl{2)r*7>L=iAE}u%E)AwOg0$h{%&{43tZvjXBQ>?tJW*OdY%iXN4fbQd#wKG1N_YA
za)9m5=h-TU?VImjWpr+V>Jt}`DnK8`cI6E&$osKfBE=BfrSeFOp}E&&fAfI-{cR3v
z`!pJLTAemgk}x|W`Q@)ZM4`|pA4GtqTCNbaHA$_FPMUOU2Q=z?*chs#;}nZU!fv1a
zdX3N+s*^|Yg91r>2;0-JBAdpPWQjGda*|>FXJGrFEW@xmldX&ccbFO@+~cDk8REPk
z@Lclg-npZQ$ivoC_hGx+Gx9!cH~0UgU*HlJMn`>q<DWdk%t*xLf7@d7{VVu7y?=1`
zy|j=R4Jtr&4ENr6pON`_N{^gD3We4#aI;b&)3&-50+T>CI3c8TJaeg#GC+AAQLL$N
zAF#Wz$KKu+`?WoH*LT|_Ny72j9KZ6)qvQ)ca()pZETh#CqGlhRZxi+Uv=0ubA8aA<
zsEm~<m5Q{REe`5+dfGB~^b|rU;y85lhqQt&KI^WxHtjkkhj`-*D{+V3l}$H<ec;9z
zwodJRU)u`}aZcc?;dE^r$LY_LcHzu}`{A!&KUYxq?DG#JygsL1`6~Mxo5VNoAcP#q
zTSzHMj3)8GvmUnD=k^cY;Nh7n`6DN=D!>Zm%KFkvL0t?s38XF+zCw5kVSRSDYuveg
zmwR_txOMv`H*eo$dv}L+yG@itU<D`V#(4gz1*+A6Las=l0%V}bS3Qz`gRtKjO!N5V
z9QkSqYYaQ<n>1Q2`UW0-=xJgD0{1<$mvQ*TxbvL77(oWm*4mUy0oe$0HWiiqS;^rv
zSoVH0QBIe23m_Yo0MBP2Hku?(m{KYRnZn^-yE9SONNMrX`2KKE`FU2kQX-6EjB+gz
zl^Q(b2jnKFnJktN<pOtp?9o_WLx%~Tl_+Z%n=Ekdu_JiOqm(PqZPw^^55QV{rSO70
z(t}>Ni-#cW_wfrkK;Z>GwpyXz?qT{J0&*xnz-o{hjPGNGha$nm8tIp@IZ4i15}P<z
z2MtD8^0|PZS|twqs8$_iEX_`bAf0+ndTqLiCd`enRqqg52|6UPhQHzF5dTH%SO4yF
zjEq6P_f6jYoj*YBH(ZrS2<IFkoMx60;A=(i&Mvoq@CK)T{hN5zakTK#bDmaVqzm8-
zPCp8qKd4e3()W-uAc+lI+gq%z++%xVo3)L5Y;JF`zrRnr+ab}0qthkk{>9&>SPIAo
z1(Z<ufzRYv5ijh~s_oOSH|TeoSOi|NNDu_nc6K>v)M>{FXU{!>K|<U=r2jOG(45_m
zyR-)_<MgsO#yFqVL+#YV_EN@}`|5PrgnE{|>69mDg{2=`qFA7m9dP%OI3d$`lW7a*
zaJ=R?UrvNJs}O!s$=g#;UjVPq@t0rZU~8M?);)v}cv7Nq10&jMw8*eMV&#W#b787N
zZt)ma`G>Gw{T0|Q5uO6;v$s=c_0B!k*6y)#=Qg+Q++ufc*9Gx$gte06v!guy_&k-W
zPd-;jv0aj{D3Y*A*zb}gJy0=TDIixVVU1;XbBkuXP2UI}I(L!eL2N(t*$-no4`4f=
zf$hdVfbHVYq^})*PY{E54u&{!*sfAHAZ;`{aU3g|2AEPD#`a-SA*8DS@daZ0XD#Vj
zo_p~WPd<H~dVQ1A-+YCGtsUh0_Ta8}L8^2uLm>#PN3`4G_G_<mVX}fZy?_xephGzf
zAbrMeKV(y{zVi+CJgo89T;1jF%3bc>z0LCNWtLZNvb(oOr_&+U5yC2-I6KdCPcBg^
zdlZ5qN-9*|qo5RFcb|T*Lt=X1Y5Yo_T%|}7C2Vi+(CoI+D#y{YPY`Q^z>qnI9DbTY
zxYoLT;C$V8_kYG+Z!!$H&&?}qr8`_O#-`jsLzqCEfbxfe!$L?WX_8e7r9@<d8$V<6
z_HzYw)yfFnp244);?y@@WO;iF+wS9~<#mK4u^MNj;rIaUdpq2H?JbV|oiW7NBqGh`
zgmf=hNQL!;tGJ~7bT|6tvwmlnYd5d*(Z?V0`kUWpYj@qXFtUdaC=~rTqExETYPJbf
zj%sO)Lav1IH4=|RN}^N?XskGjv6`*zO=`6RYOOj)PCUlg*c95hcB)D>i*%aBXB!fG
zfA!5E(>a0k01hI1H@Rv|rZHm=zu!R%MjW~%J>{OFMbOufH;|<5#Ka^{tI6e->9b3-
zK<RYa6kw%5niP<X{z6LhLJj#(u}SbmF8lB#QXvZ^DvL`TH<s4y9Q)U9qxWn0iQ(eY
zC!IiOtWlmql>*{g2SQCZ>?5RyUn%0Id=zmQlY|jLz63bA0)Zguchk0)LP!HSr=3P*
zX-ATzukpPD$XQCIQ83z7H)D&JSB#8Jp>laTQ4cTbQQzGq%#YCT_UQFGSZj#-&?x5F
zuQdpb#WNOP3VQtFMRf}c^Yp_8W9QFs<b{iDefu@g22UswYd|U@9XoGd;d+4EA6;f-
zeuAkNUc?kD>AOlV`Lrr0JXd`cNTl-J1us3it%&Q_ZnAp!9zS{iM_jt{0eiK5qSy_l
zSZk3eWITZtmR74l-Y-zhIWI*WYhv9Y5*;GHPaH&)r^*C{fM%n??(QDVc8luxEQ`m_
zk{F8=#<hk&YuO=D*`@)XCF4xF(rq@~co@*FbuF6vz7?eXJ@<WGFlNvO!UO`F4MYvU
zm*=Gckq`*%AP_dzq!Z<7Ldane;?K9rl?uJSCYYM!#8;nXWp^9X4pHgQmynXsIBZXw
z(sb7LSbg&ymVRXvF*bofq&%ua*p4L0bl8-_rr5r<y2s7s8+>y06W;pK8*J}wrrct;
zGHegykYce!tJ%Sq4XVXa^0^|$H%Js(2;y{4jTWjHYuMS@rq-xYZ#P&v_AsMklc`8B
ze+{+|n_R}a7Miq9T_AhE!^``S?Ump@Y|nUNRyb;mr<B8X;jlet9kv@|NsO~*gY29*
zY&X{4kL@Dkd^zpMFRT4psTTRlH=m$TDAQ~P6qXh_`O0(L`-4A0_ce-a7(@}1;1rMC
zy;Zxh%Gw)0X8v2>L{`R-I3H+)Kp{YR7@Xj)5Y9yePhhmq-Q{&|-n!1aAH2((Z@<od
zZHMGACleeuK8Yz7$~5XNe3_$K93_`8Vgn0S5+jI=MOnpIWfZM7>l>@o>Ic+XEzUmn
zGR0yQZ8e_A_=;)Eo%H8&Z*o|f8{GB6RTxF8jhZHO*<LVdk|?rjE@0EMix8<Wxf`ZR
zH8ALF2(r808ia|{<ZU=SCey!5$S<M&DwIGffl!jsQ^%Qo>QOdce-9-xJtc{Ye{K5z
zouT{o$2TY*JIdt6CrHYp<cg)VO_361g+%(pNjRkBV7tY&tJnGN8{g)&x85L(`Y=#!
zat~igP*|);Pznl#5}Vtb$Qhr!UqazgD3rkpr1emKo_4Rz_VyO_W}SAgO{?4IvH4Rt
z19Ho-{EY+ewDf*Bmd{$V=1{EWhMf)_{Op%CS#p^6)3X+iYxh_jU@9f?JjK+^B>mb>
zsx+N$oYc@4n-Wj6gD(M{@#j(!uLhxqdRm`%4LU~|$rB4gr{nb=N(Dr@2qwlkIycA6
zi4*j9H<A0B96L2fF6U#Fn-KJJK80c#?M2AIcNbX{<NJXN<}JkCK0&VRvUw@+i-iG~
z0VNTB4wD!fEy?O`mCeoso)6V<g_)@XV%?;eQ^aA2(VAkuLa|&$TS?OEU^^|g?{0DA
z!qY@ih_cW{Q|o!u3l)-NlWtm}5yl?calbfAc?uZ|$mdvm{t0@^H)-BnMd4a(R!Cxz
z+~@a8+i>U2pO9Nvq;UKMHXcJ2QZ7v<C!^fPwZdKSMhTK0T>j_=m#=@wAOFRlu)2N^
zWA2|amck9;BsQT`C=rL2z5N|#CKjlc#!!JKpDUp{4nI^rAWZty>kW2y_GopR>^JLN
zeDrGsUO?h%hDfQ%L}nSZT^`KL1=^-G2_!;g*k%XTP8q0Y=iF5QQ?1JM@5!{QJf-lY
zq);x<*Ii0Rq<VD*ZC&zat)EWOxeMJG$Fs{SbV}hoMAQBGg(Pn>ZSN_;$cdv&J^m2e
zZ+(bIin>Ce2Po+9#fxF*>N3TnON>AI2)bBB6p9aEyG443v0YH#ZFBSbO@8p!_xa&Z
z-b%6keqT_hUuT_&xq?Ez#P;qsa>^&?7m;`r3Z?WJdni9gr{7^`cZWu^L8sTI-R<-6
z%yBaR-cMtDhQ)^}whv;vWl*)@zKc8!<kNn;u};nMQ`j!t)U>A*6Vns)>$?~|tnxa)
zU5yLshCDwhaV^XTv0V(MPri&*EEbran{+LHB@mus@{zN2ZY;C&?iD=g^cAHPXfv!v
z19+ij?fs7_9y?0){5kZ<IDS5lOp6%lCUg5iNQ82C|MuD**RNma&;Ig{dFO+7iIeEQ
z{or_or1%q)%XwfVo7-Ec*&I2)gcLH}4<PYTex6pR$=<;}jaHLpr%6B3OwXQh(n%bm
zO=Rbm;eN=({<A&c_Ji9D(x#&fX+k!r+NO8C`+C>jckTJ%esJeNk;?^)kB<;G_b|p)
zbd7bB&e?EHCaWT(!@n=ZCQJzTfUN`(<e7QlIa+IbbZ*=s(H5t0<;bC=S*B+}cVaew
z^e)AjNh;%02s?^IVU?GDUxBB5WGcRx^bI$!-Qt6f-r==3zfYJrPM7B?=4NJ`^Fxw2
zu8Gm)asq8)rlw~RO0u@OhVSK>8k<h5>B5oe5$x~nv$nBDv)v*}G{x#T)zKM(z)yWc
z(*ul6xhxq94143l2tJF(4c4l(RWtNxysva?%zbtr*}qc*c?X3RFg`WOdcEc%_rwWm
z8a>R$-DZ@yHq8Al7E(%#9;*6FbzdF$zi5e+D3p0lBAmpFl1_g}DTP81TM*?k<*_O=
z=Z~{kyNM@SpcKk;KCp?+5p^OWBJ!RGLZh_6YKzRd@1>*InRP2aZBI*7E|0UGlIX;6
zcPHSTyR+=)W|%vDmc6Y$<IO7Fjqh+`F-9TjwmVp0Vr&Y8lUOmG9&2~jP`Nxi%`T-x
zQ}#-<$+H?ud>}|PhM}4Em(%3<U-xnelS>}Fkr5VOd6}C#e};e1bRyb^h}_`zq;DYF
zuCxBuI~=Q&Nh+g=N*Q4#CT-b}UT`Ws_gKm8n>Sd#b&db<hrh$l{;tbkg>z*PfafWc
zl1OXF6$ILbEX^NfZGDx^olSC!MN0V!p338U1@ZwHBd9f+Y;AATY&U6k+DJdg$k?n4
zSX0n-d^DR@+3&w9Gr-R--)!0R)^+XJ;W-~-fxZ9h(>7ivVWgCYRFawbId<=^kZ3`Y
zo%8#${fyu9X%~wJr(M4Yd5cVb#!A4?G5yR_v^NguEZ=iOP-(5w7=x4euy|IWJ26}D
zen5U|obuQt!j60fwtEiSjU@>!w{NU)>67>Q;g8>N*p9{X6*E&)guUJ&Y&Qfcwogt?
zBa~!gdjsFgGchtrHpu4oljLB(#>VCb%}$#*(G)AAR7R%pAHa4o!1hB_m=xO$2Adpu
zpT=N?lh!bw!gewIn#0(h{xh4FmeM`fv55&bn{^UnNwjwD*K~4n_|7E=eP4Son|?HA
zi0!W8`^7Jre(wPnRD#TxnE&cav^EZCt!@HokaMUVDX{of5Y@YEy!~T>@lo{D9MV=0
z$ouw#@_cvS8AI67EMLFHTR(c8AHVlw664;zANb78Ow(?*((~l>Ui|=!O<0&e!q(P0
zYn!VCex8ZyWI8GC&X~2Zv%Aac`Wo$Shu9cKCl@GICjN@MK7D`o{^5r#katuki%o;?
z`x3fCcm0EZ&VDOX9CSk=f{E#A)*CevBL^5T(6{7p(3y72e#Vyj&mGhuT!l!)AV<ep
z{K|9Gwswi?O{~pC7ws@>RZg$0x3R_ArH>gsbsAwvqjZ&y2+LHLO409i*;?IWdvlu~
z{`d`I9XX|KG{=rEasKRSKKkGz_V?Rq%fiwx2PBClUnns<zkt?;%hx`pQYcfcjN$o$
zxZh`EYlB;NZqe+t@%%jbVu{E|wssnvs#KlSg@wdwjCRLAJ4jiL%}q09FDtV|a#^`C
zpdJ5u2=p0WBW;!3-!5^VJqkm>68HgA(=+rNO}hIVj!UNv+6+NI<9%3ToQisq41zzG
z=_#x|r2F-Gml%Q@DOnUDkkUg6iINgeNVHA}tsy4^D!yS%IjIPtHF3K|EG5b>q;@U|
z&Ha4}xjZH@0|G9-!t@;<&Pyd7EEeG@tP%8M!|l6e-s_xY;hQh8FdDGCyUW(yCe7Xi
zId7Jzf0s}RTCE;I-e+=roMzuq1iH-@>nrONOq0o6i9VWqu0V*-ng`SU2o)MUB$2hL
z(D4`W4F86$Y~L6OLMsZ#PcZY`MVjAv0~w}r5Lxgd(h9c}1V&Q3dY!`Y1*!{6DTl@h
z2@k6lE_q9mg!*oawY%$l@3n8Ud$8lOU;&kCg=e37lI5E>+27lDvOe1FccU$($|y&U
zd)&Eoo8^@oEKV<y%M~#uA`U}p^#g9*zQOv|24RwrFO&#^GJdX#9|Q>Prf{{@WHG8R
z>E$dLgmzL&(8-~54vwf!dgku8354|AAF}q!o~M+OIF8Xe!C1@a_&CkoUE-vZYE+p)
z;$-Y$j2TKHB(W3ul~Sf2p~K0@&;QO-o{#{MfE*cP;f1HDZ|@Q{+5_2-`>`E-1pUn&
zHZEUf^!UkC=HoE7yEZH+!m!8A#y&gSJN)?Fx9`LDrKJVVo;txNA7AERzcZk)^!q+q
zL#|L{dTx$HYp&kBOeJ5YQXWaMJ!ETVi<NsTG&@dQI$tOf8OhFGljD`jr?6e7*iKe;
zA;tE=#h!^}yNBZZIWufe?^lue-IC&ZitSko7D*hrpw;(%Ca0$8x0>`0wn((bBpTzn
z_Ny`8A#8V|^E#EdFgoK@9me)Asa7XQ1-ISrjj#|4{Fxb+UVet!AO1O}8xDqvGS6?f
z9|Qq{=E^-buU%#0^cjSmN(0qQ-o)J<0zt3W=ibUDo12@w^WNJStMNRyAI_dV#nRj?
zAH4q|jYc=EibKEati1}Q3iC@#+`Ds^k3M;y(b6d8auwx4*zK{lvBs^tx9Rlz1i3u9
zLXln!JG)JeEKTD}<t*FM@0;EA#@Mv$mg2z!`0olSA`R}{^!adD^3YwM4qqI;>!lF%
z`+cO8_`c8l!XnL`9m1aTCN|oj#Zb>O3z&^IPMA3P5-N3MrC|f1O$dCC@$+Yyd-g7!
z@4bzQ-9?-T7X&E;Rwj5tBNM^irRy|aSYu}4EIKeKQjt>S1cON|x2`VJZTDH-y2Ea5
zhjO`$)-gwqAK_cS{7v3^`z;Qd4Z=7rcLMEJ%tx1RapC;KjEsyhJvRqZa{bzMHb41*
zdaXv(>wy%Msuc>=GC7$;NX@0?6;`(Yke6O~foGn2oMJKOtgq4!D#?24>H84^V+Bap
zM#;cCNdsD)p6SDe%R(rubeW#Z1=IGZ6$Hv>dUA??^bh_Ur1Uv@;w0CvT;kvT&;Jws
zIKd<uvJv|<n=pwX=|?0{<Svz1W3_e?D=N#}4LbRv#BxO{ikX3)YrSTGZBz5(sMDq7
z!FX~&Q5d@W4LbV=SPP|z88lt;#T;HPKuSfg-KAJ8(d%_7mkLOwoUwN1hp2Gv4Wa03
z&9!^uTnnCJ^vn`bE@EP=Otmz}?943R{oPCSn|XAs>2@M|-9GcPW8`xlo6R2MK(p3F
zBq4F4m|R+*320%7g-4vr({8j8hM|wGl!(LT+P@$xWkeu=L~CS_WB$1(*}Zd@{*~pF
zli);%vt-9eO&}7YE^NMciP^^=Wn$(SRykP=>8YWsfR*fS?y$3Y!20$Yw^wgDO)91E
ze9O0f`BipOibSs;rxXNBx07(t=zxNe(Qy`!Y3|&*&3AwJJvSJG#YPePwHn>1Pf#dO
zER^sA!Nx(8?|k=lo`3c!mKLW`&acT$N6W0C>MC|3<-fSLtU{@whh>r$;j|TwXLIO+
zmADMtrX&>5DW*Guc_GO0$U_%6dgK^SJb97HYMJ-n{sB>vI2uJpipLF0SYt79jE*8q
zl3)|(PGE3+q)a-*ak4(|N~@L@B9jn!9;0VYG5gfR+<yHXOq{j=g*y#S-p0vr$V9Mz
z`6i8LHkg_}h4wRSceDv56;{K_job9vA#2;K>^Jr(l}c!xaOCI$ufF;!Kl#Z!)SC^W
zWQgsZcETrDS2%O#Jk@HI=~>r}xp`xm?W>n))a%44wihd9^3@VK86bq_^6h)v-T70V
zd-iEAKJf^}!VueqbX_Grh3)p!*v^C4o<=+z!uFJMG*BWy;47c0iAjF*@BRi-dMqA2
z#`2BJ{K3Eb*Mw1m)vo<&g~@un?gH;eByooACdGDKJ5;(vH_>0B5PDj@7U|Cvk&q7p
zrXGKU>6I1MetZd`9Y`dD{QyXfCp4lj+5X@Yjy?Yz6Vu1B!tDp4Tt`YtNfK&qU0bGp
z(Bjs;n>4!(DwQ&E9P;oZ=Xn14=Xv9e*QvMKL~$k;B-r2UaOK8bjvYTuxje$$;sQb`
zu3x*(_VzaQdYz=#M|g_zNR|9ZnY;`zBH`UjS6JKsBVK&|IW9i&FhQW4Lx&)xH@L&C
z)3<Ad#Gnz<!;?dPU6Qs{-SEx*NtF-^tK=aJNbh<CftO=pZjOKY-~YD=;c?>h8Qz+k
z=1>0UcL?LeP2(#8>0R&kgC^;RByo(1zl@*`fHaa!5@e7=7JL?-zsSMzU3PD;Gib}m
z^aERkmwE~&2W{#%S4f`Jcq$lF{saP>81{F!xxIXok@6I^{R2vcfLt-h(y`<G_HX|a
z{_M~GjGgUW!hY(d3Is|LYxw{D&ws-HZj-Nl^#x{Ur%<^vXCJ-D(KCJO^*S4?cO1b~
zN}{+6M$v3{xpe6|w^o0|>e@QHJ9S=u={csRsz7pIxpOE8I#hUfJOKud5Cg7-)(AN?
zxi&zQ2xs1XsBJ~sW-6D<oH%iUiScPFqZ5pcjdT6l6>9Z;A|2yv4||A6og|tt3`tV!
zwXAgOey$iR-Om!oU&O(8ZPu*Sd-!=kS{jy4d^ild6ndI*S)*60)85*_CWg|;B&w&K
z97!qXTI^ATNi@Fi(QU^hah$4kyRv-dr)cm<;)s>iaemmoKw*BA+n4X~!Z#m5Cz{pO
zZHRq3>|k)*oLa4hLNGNsLlT9=VS))`qHcrwUWdn?IEfq`!$t`vkJHXm9$`Or?^8l#
zoR!Nj?y{-NNkY-j<4ulp{FUcuZfw)7b<^O*#dk8TID1N<5@_AqqP}tmefDXjl0%RY
zpoB-a-R6@IKOygz*xXzvPWp_GRk1?z5B}lbrPJwf<;pIdZb+hCuqCCS*E9U9fBh$n
zR3~}jv4^NsMmcu!G==IY+uPf$udlPcy-BH5VrpU*q(NCl6vzDJ-4EIM%S&u-Z1M67
zFY?GE=h7*)tObh$QJZ;gN~B6#t!^}7aORCMBxw-l$zeNGNJVS}iTRB7wC6KDJ<YSv
zKFi3+IHTiJtUdl1o9p-J^b-=D;2AG%3k`|aVLu$m9Oy)+75UT|)TTen7jn+s=g*uD
zq{<-!kNKw`rFMIjy*ry^&8iHurIpT(&=_>BL-Y1slE)5Vds+psiQ!;>m%F!bQz=c*
zsMRRu19F9cg(FA#2mkQ5_|A8}!|u)=VW<b#?kh=>z`yyOzhHmA#mg@~!_>?qDkyXM
zp~pCKs!yZQU~}zW$~}`raSx25)$Vfn%1u_*-(`JsgT37*FFf}Y6O$D(Y|jeu*4jZ8
zNlA%F-?I?T5#m0cv^JlK?J4o-zRHg0d6Y^ejvYJ3*ytqXk#R;x$GLgq293r6k&f|F
zY&R*s+r$t>Lu}99`KPh{iwNXDV+#x_@FJ~9>(rZ8Au1(~edQS%ch;$G?hik&aMf9f
z3x<`kbhqm?@2rwM^bDT#hiyKbJ=*%}9X3{15jM}k-X5bPWuzy0_}Pc~dw=im@_WDk
zd+hG+6GwW$Z}dG$x3Bph|3CkM_4R#Tc>XCSC&$T^$GGss)12t{IXKv7bM<a|S3?+g
zL3`92ecpZVWA?xQ0qYwZ?C#Wg>BXlR87aE`fF0a%uIe`Qp%qz?$2zZHYldFX)+Bh+
z4)%iv;@A~_oEz0}L-@W=rBdPCxeHXvBaBSU;3>uO@(o(Om`KM2+8gXOtqpM$4et6l
zj((<h_;Utz!_odhW^66~_$Ui6KS6D8AGO~~>(jUgVVo3#5(W}U{q{Ygu;+#)Jdu{K
zEz8$0Q!EIK?jXe^g<PJK$BvLM2mI6j%|D}F@3FmAqju0nCk|9RCCU3f8qIF2!+-db
z@A8L#{yhrC9K~XuVzJ2B*eDa@6HHHyF;=Z2jbU@U#jSgLT)lal)y+MkMDv~R{Y9#g
zw9l)re1#)N=8)2$^-xppa8PGSMQ{N!&$f(fVHxl)gcKxJP%O_-Dvxmgn6wzQDB?KA
z8jI)Um|r@<H@@}jy!L~?WWRPmE|@vgo-v1Bx+aREl!KCGlNw`Pi|30t_!(F_rE@uW
zwKg64P|Eq-IiXn1Xupo8MXz2b$QQ^Ls@T3J?6gs1B~Y&PY^9_hbxGn7-wRyeDAO{1
z##yow63ymDnU(TaXttZI-rC{ozj=`$(0F;n!G52(>n4(X(?^?#+CiPs@e-x-2%Y``
z{WzhgdxXt4M#1RRG+h<YI@qOL9m5ET@I30ZJ&J{Zzy>s%-E?REVlHcO=G0jbAtdD!
zN0@u+A(nseE~0A(d)*C{NIXxXM1l+j_2oOnI-)=!<rcZt#OC@cChpSj*2$F%m^fi^
zeilQ*Z~nu-$C=aT`IrClU$VV@KoX@@009Nh!(jQ~$_oGI|LgyQrKMRG7v~upA7x~8
zgz@olW|qz{w{(Ie3JH55^~M3UX2kVdYkYEjm4jxN-Ti$I4(jahAMnBpPm#;X``+?{
zXWg}z95*#dVsv`hW^GoTl9z>W#^+WR3GyW}zt919R;kq79CCh+ASf`uc#?1Y>TmFq
zpS(@0+rbaK^qi*`dhFa};y4~8ZUcUmHdwy6a}MrYo3sV&KA*rF9cAwMN2%{0AP?eU
zwf;eD*OJDaHR7mG&Nyt(cwx70U89f}7}G(?2?{}uV@DRqmjeFrfA>#lHu~&t*QwVZ
zz;@3=ld#u{`J=!10e||JZ;&qp6bd<t#R8)vql}G@Gc_^FXr+uamaW}3ch>f~e)|p^
zTL(mm=KHUI&uN`C4|w_I7g$`J0co%)wu_&J?G{L4bQ}X2wx^a*7U#!np;(@xP^x|k
z+vR=O4xX1|Zs90j`^K;G<_}-vpk5>J{hxyEhB*EVY{$_7k}s?a^S~akDMvfi<<L5#
z1GwldH#5cji;vUzlOG_XIHlv@vO6tMo<O1@g4)Vmk|ZW)3hu5?3m|uwucPBO*7opZ
zfl9H+v84sZrbhT5{`o)S=8e1T?lfrByBL%5^*py9Vw<pW(BNPG&VS<H{{Ej1?)p-x
z$k^BzlM|CnO^h-&Qb9*C8=DPo-P_~J&0B2l)QPp`KmFN%q*gniQQzn5uYQH;+40Z3
z>jyTa>3*=*rgwcjY=3|1u9s9sXUG-H_wj7q_Z{vB3<N=*BPY)C%YXYf`O%Nwpw_4}
zF*=IYsYmQ*-1R@p`uT!7>qfNQ&F-|(T<|TKI(v?#M^|aR@d3i56)6lFryw4n@Ps5>
z+oij+O`$T5l7_%@Z3NF(+`e&{?X6YDMi!W!nkHhJr=NR@h530_@4m;z>IO-aP%L<)
z_XI*ilxQMt@Kl1Y1l=&<px!}<0|aX%S`#LzWD0^XP6*?y;o@FSq+{Ow=v{if9#ItY
z&2N5{)2Elx3`wViWjB0ag_~dzR=J?ZScEYcqmee9S~b@FQws{EF(zi_5YkV-ml#N$
zIOTjV1xt+trpIS^^2z5(wC2|Gb@GLV<B6t|uHvv=i5Lz_92Pdz+K@!OpX>Gh)OW5F
zQU!oaiQE=z6QuOr^GT<e@=2F`cMl^If*_BdD`WaGM#qrz@hSxn;0FO?Qxg=5dFu5h
z^_>Qbg#ts9XO&ihVh#>`F6l?uBB0*v^W3W^dGz8Wqa!6g_~C6ZId=ATDUwk3w&_Pb
zv`Ltrnn4?ny}d5?)?<z)7SU`lJyV9I)2!|9(yQ;WaN-;?2=IN6gE*pCsUX65@Em_J
zSIQ^KTtd=zp!5UgpLv{vJL`0=-c9L44(zRQXUoG~%B_1F^qY0^r7AkL@)HOeja^!`
z4Gs=A7%5LPH95ucXr40{PVx0uUg6z$KVf@wpKiO4l)~v5ruvJqiHWp9G$i}2JA8C?
z#bH49^CHL>3QUZSQpn}luQgcTsnhKC2BhfS{T+VqkAH_&t4^oW=IdX7iDKD9Xps(h
zIU=)2Ih~$VZtBx!O*PU=vQDKg3?{LzU+v|XnOPvn7qRC4GwJAAGOg-sJms;pw7`om
zy^6N5y|+uDkVETiqE6q}+SSH^`aqk~;>28yaiY4P`;Di*l~}OO;lPzFC6lL5v+&S5
z?YA!>Q<M1twj=Oy!&2eK9=*LC^5rp<H29v=BlSGRo#kunY;Q19onvxxiijzmc<OQH
zXJ^@1{fNzttpT>X*1vVwt_`s^c%dXwhF+LZZ*&o&1~?g%P^Xe9LK1xjw#Pc=gHJx7
z7xswan6G{9MNXbvKz|mtlVQ8N#HFiZIBZw<W4qOYd})-i$ytQ-u~s8QMyk)m;8SeJ
zVf*CxG><>_G)bblvvQMsq3QHA%^?+offbb<*w9&<r0>ZE-a6`k?sxu~`@mYXPOyHE
z-t}g1*E@QF$HHTeP`i1D&igkJIG-42DCdG89}mw$XJwtRwoi~R<0<3ziSJ=-!phBS
zY;D$AS{z|<VG*y`<kfF{o$;|z8ubs^-q^+@no`ks?+38j5+|C(f+q!WBI)#FWUT}2
zBg86kqKWkIuJ1<)QF8x&=*J;%{OEOhy&ho{@vU#Y!s6l-Qb>$V5CT-z>K1CSAB-_b
z16YGG2~yfm-Sy>>Nya8;0H@=ZJySs{8ZJDnHO|FKNakl2dFJVtFxs-deh1I@&|0KP
z+o8K&GDsBvy7hAfbr#oRPy+_J`y~*_Vv(h%A7^8Ejd<M&%o+*WSd4LHZH1&)Yp{Fu
z8dE3F;CXrQ3(o6J<%zVQ)9q4k?2*gmId=3ok3IG{RtxUlxx>=ZB0D=<v|26alNKd|
z!)%NpiW9Iy@5SMCs3A7OiDzn!MKBy{9R@;il5qLj#{_|g@&x((FLCt9l2d#~r!P8<
z+J&n!W_-xuWE!9pMoJO`elE|{>;j(W2*2*<O*uHJ#DP*?nh|NDFhU@hou1|CC!a?r
znmc!wDFy+BoO0koI@KzM)<K6ec_&z$q#5StTNaQ?rq#4`#5&sxSrBI3V8dv?L0&X5
zex4v#MMR0?NtFsn4?HDPB7(p#R}j93jxBqQT?)+}Q==tUt@9mECG1+>-8{z5>>?}o
zHu?2`@Cc7Sw@7iM%A4Q5O#8sFvid{1ZN*IFQ>`R4n{7(vA|vHd8qFQ9+-z}dX^K)g
zAYUmk{g3_!s!x55JHPvHsPArY{<*JHDUVPc9btDTN3+wV7rAmgKgYz;e@)9;>y-Zm
zIEG<1jR(caX-+)%1bgdy$ab7w?}kJu3<6(xcnG5HI*pYTsxymDLO}Tl<&!T~S-W$S
zR<lVMcd1k=96h$kv(G()jy3zcTl9NfO2r(7ypPt2BM2Ld)&i{|iXyDK|D1!XZqzz4
z_qOZldHVG0vshZ44uAgL|C|cs8h+(hUL_ag(Iy6rt2Ri9K{-1st6bugRc}OA$PgIg
zrBWY?sp%yrpQ8Z|^K3*aS>z8ZE=fY(&vWL~d32(A=baz2vAs<(mnWb1UGgTB<2T97
zMM4fLI*_(XlZ508y5L1xksr2>)61w(VBzA!Y^|&lZ`K`|Kf`vLVY{GTZ?Su0nTcbk
zP+lIqyc3^RIbtp7cDpni2LwTmBTGkl_~A#eMzFeikHv*Kb^}Gb-6lyiNvsiei0$!b
zV!P`Y#U}Mx)$vcqcAaqb#uWnJaU}EkS6Nz|M{99EwsQ#EtvQ74_7Ju!w6XX>p2_KX
zJkKAH?Xv=oo5DPV?M4$v5z-2#r)IeL_|xblVRdbVLJ*M4DJP3!AH;S^l4O!rDYon6
zi)*=m);<82%u6-E{J~u>1fpEw=<`ppb9V=^-vI<hxFDZY8Nn0K+S=t{`4*$|ODNAD
z-1Ui;s9cfFt*cB=9ivbvaeDF$4?X+{^?HY`%?*wpKg!0&I^Ay9ef~IcL8XwGL=z{;
zU}$8xAKX-NWaK~w?5FLAFphcm!*>XB?ymo}zx^vrPfd`dvN2LR={AFN!oX&&>7hQM
zkOGB=k&4(rzF1~*dfurxXa+yq;A@q0Q_<QD1IA&55{kuzC7yZu1>V*PTU%=sasgUb
zTpQeGLc~A&K-AAYHl@-M*<o1^DZmOvZf=&PC(ok;lScUNzcI#{^(rB-al+brAELXh
z)DldRC*=Lt{xdd?u)=Wt`X>Z_K&d!R6dH6Qi6S>4H9I>+6!q!%BI4K?E4z_)OROzX
zVu>{*Mi6WF`zSGP+9(}Mmf}Hjqo7bI5Cj2<F?@XaBR;tF0pI=Z>+J9EJK@~4buthc
zh5O2*SsoySNLBQ6NWZ|$>=L<LF8#Wp{K4TDuC?RH8Dr?BaxDnS^z;H3pLmv;=|%ST
z4`_GNsW5BP_GflssKL*3zdXX{7tT#|h?G9Q7r4L2TGS!O4Hwjb=Ue7=NDf6VSA{5X
z-i~R2CWImCb&10Wofz;vtmot9eQFWZdoHU{p6{F^BFpXdNjAqGq1o>9Yrl1x(nOyh
zy?ulK<$rn$k23SeOZX#DEsap=u2Lvk`eB#pi3yTK^Wl|!s`(sKMNQP{GV;)4ocqmx
z2IX<?-nzl|-Y&jUj89B3HZjiR%s643VB$Ef$vA-f{K7DQ^Hydkf4GGp5oo0tIeUuv
zhfX4-Mp&DIW|q7;q0y+*W$pcskVzlqdEn>qJ&$`gZvcW`*yG;4TX<f;)Z_xwQ*-2U
zWwy3<dH(rlnO~SCNg|>sCW;b_v2Ie*TB5`dC9Z-K8%wO+pTk%aC;Cvk<I`sMzVBnP
zbb4*R``Wkp=#xu)|NC!ajKNcGSmUsAnP7-@hviK&@l@&MLHJBe%`!4FdLJ$hWl2)0
z3Y!L`*3j$q&{~tr<v4r#0#7{tG+3X#-TeU;9OkWN$r}vMd6KqfjsE=S+yyOZ%h?=m
z8>QGjGsD7TXVLx;+g;UC+=uO(A6`QD+JlOFo`_3teixfKY`?jDg}@Ie7DkDqA-2yi
z%riYRNfd|l`zf{y($8#(jV02S*c`%k<FH--HQ1id=ka|XZ48&MUFOo2OZ?ymZ*g!?
zyRR*NAGX_}sO%t^A;b0n>E)T4Ss)04^lM$vmC>)V%TH@f7)2Os==OTVNdgF_rsjD3
zv8R}xn&)7@Mz_<!=uEqAc$Z{9n>DT_cUT+o%UYR4j^l|93l`RzRMO0&eDnx&kDMe?
zi5me;KeD^)9nV&G`>eh90XFKV#UMpacX;=Ee~z+-IO%iu-ZGw_qgWWD*NwsY=)|(L
zw7}%ln7iu-`@uaAgC$Pv;C>(8^_D2M59|kde{dxP#bOc9^N5m!ci;appIrTf?|kPq
z+U*XW=aE@@-KRE@{W4}qIZ5w&WPqP5F*Cb_=Xry6>EZL6EuCn>Fv1v3x7$T08YvVD
z3&(i!;&Y6SOtQDTPq*7m_k$hC<$NhYo!0SSg5_`qtdvKfe0;yi^!d}2m!?S2SSvs{
zutzv<OkkmT`!0>09cS}{Cb{(i;q`ZU>o5P3IPCNB<x7s+kSO+c_Ng@*eB~>z@ceTx
zlFR2QSIZ!++bEtIOiE?hQbv9?Dfh!o4DqR8YPc+M98oS6@jag?PI&i&pKxzumG6G<
z2Sj1)G8Svy#8OIZwV6c_fKW)~BURvhZlz*ue3D$DKoq$NHmmi3cVW_@4fkAKWzAYk
zuiqw)LyQ4|Vs>_cC!c(lsp&b|?GAC|Mw?A4n{c@DcUad5_(Fm@tKEou5P0~03SioS
zjxmNX46(*g@H8U@MXox5h+@Y_vzDY65p}zmD1;<&|DAHgb58~s<r5JQ_CuGME2r1j
zs3%<47wFi8$DWzMsspZExyOI{y(@h3!5w6-$8PgBrQ#G%96icH?ix}w$mcC1)e)}T
z-lo@q(<h76_V&^HyUZScn9|53L6E1>ZqqledQvPFSe%<Bj3N>ZIV(^Yq_qS>kr8}<
zQI<6Zsa)Crum+n_5>NVg#R~IJoI{S~U65l)&v~lLC0rYJ=kitBwS9z;_-OprIx9c=
z4y!k=0-6s#e3yJaj}$q!Hh0<IYjE<!X{yyKrBWWLAda2nfK-{pjT2?nHl0AVP6X9h
zM^?WtlZH<Lx|EVirGmABTC>3qfAj`dZ(QdmKY0%!hdPw^S%789TgoHN0v4%!3?8M*
zIHTj!K$hs~`_5ukMTiq;R&R`<+wG+T5a0(z&YpdUM;>_si=feH5GP4`&W9RVGJPLe
zP?vqqFXEhQoeZW`VOZf*(!*Cio?l?<><P*X6KI@Pr4<%!Qomdw@GUg&tkK-vMP}H(
za*1&HL*9AqHIk^$m1|er=g^A%-2>{)CNI43GEYDCEV*2sa=8S;Rf&{R19_JLwx?}$
zlTK{>EZCmJ@c`T7IN`%fAF#Hy#_O;DkSKZp+f9b;_I_-aUW)BLMkq$dCkXNd_YR}j
zVf!HXx2cRv+A2-4z2EPUBw_k&71PsmJoeaAOis<v?)He|1f2}UY|}6j4&51t_4B@r
z<px15TF2>`b9cQ>pSP!c0_79ri_ATCj{M9Bi8X0O%<YHFr#5F{@A@*`dJRu_1U969
z<!!px-{;3acpYOB-v97@azTKJB|BR?^umy@f8&>V?6Hd!izN!Bg4+*DAw4w^1$L9j
z*<Q`=difdq;nTry97j|u6_ir+`XN91$=lppU*)wQyouJCHK1$vW&@Bmog=WqmFcDO
zP@cQ%t@N0foW=`+!G0Kc45tBjCUmaN@UHK6TO_H!xDbk^rDI&Y_#D;IG1{GOx*t+`
z6O)lY{`&QE26YKidiY)etRl0(kithviJ!}nAD?33!Wm57U4~W|;_Ltli*JEeo7T!I
zK8BI{D*m+}Gu=MG_Ivc&O?*$XeB%>@7VK?o)9W{xUs&SlXJ6)T{jI;x#N-S|jxAEI
z7C}0{Sqo{q#bw0eP;?E$$U0jOiUS&>(NW0k^aQ@|(P%dL@Y4HiZ0zygdmmvEH}zso
z3NVRI`9?U8JI}{@K6rVI3_$vn$`z70VP*Li%O78+eXx%;uIiw*Nz3X(VAe*{Y#p$<
zb&uLXoldU}LbAAcoQqFAM|E_PFpNnOCzW6iwSKcKCY!>^zUSv%UKk>bMoNVj2#j{s
z5K=x*zt=%(!z6~0f{zU&;(m{$AEDbFl2(T}>>+#)uUtmv1B3+=$4Cpx3#b+YqA+%o
zhn^ygEH}1~(qCMpX7<?HZD92bE6Wjz9AE#HM>%<6j4L0lu{d#@?)6I?AJ-&N%=pMS
z+qFI`D_tHxT_$XIsg4vVwHpW<68A#_DG2gqx?w^Tx$*aj(NVhH5M>QjB^mc6c?pr#
znVtJD%+fkc+Xc>t_wYFnlt&QcsV*L2=G+M)sa=~=*ugoM5^_eMcMoW-Zr}@#Lfj>K
z=R1sTuJF$5Z)4(!db7^K-WEC(H1-eZ#ciH?=6QbWw|<*bCm&{VdWOnqks$A(JdZe)
z?#SAmP6OEvE`Bk{%I{BfvSqDH-h|amj*TD@tgfwb?dDb9|KKwB?ru1V6ANhTKs`O@
zPU^w~FF<$!!Yg1^4wPbad=z66wzsx8Slb}!cCe{%xpM^=`~ZUtzT2zO>DGuMH~AU_
z1s;0naUOZ>Nu<gVhOr~$8^DUe#d?^)ovkn8oV$sA4^QQ>Hci^aA#4u<a-$Q>ojZlj
zxunbrLo#^w#$kKAP5bUTK89*z9slNgOm%A5exF{ajjtrPm#-oc!T#1Ry?%??xdonl
z@_Bygm;Mf86H_c6nWtDO{MFdbP-^I}#dfRFam4iGIKJo6Y_|CLlS^!F?eoz`pL`nI
zQ{E#yi0wIyaPBvSQW>2j+`WB=+gGm9snu}q5Si_pJ`{Ajz@XJWV0&kcTCG7h#rFC6
zBRu-V(^N;s38R=qKY;Dl3|qPK5VqSdC#b_%Ot0_mdgbGLdH391Bi^|PNr|7!Q=FP*
z_WVh-@BSTwyB>tF1P1iM0o}E2d=WFUw}QI*BaYaZcCChvLZW_`jkOhY&#=3>MHF|L
zUpT_6uYQZKeeGLJrgwdzlm{UO`$1cm$dR^Hf7)H21$c+|hS3_UW9FtOQBtzMzt1OE
zKIZoARjyvSfpvCHE}+XWz&>!-s{m9UEu4ErzEC6zLvCKV#>(ZZ^qP%9Qh0bj2!!J-
zTEjtoo2_lPAA0>RLV6rOahfNde3nA7LKry=AU*p(D--DF3hG1@<Eb3J=hXh~;6S_V
z)$=@%lFIQTOq@A}mC4{iT4V4?`BXY$fB7=GJ0Bpf`~{=5I5}NrL3U{E?DO#HlPoWP
zLciOlzTc)(YtgFjb8xWFtFQhtf9KbJi?e4QVQzjl<rtX33q+)Kl|!u$d+3SS3=kf8
zLRrPDSS)bl$O1xG?%rGBpti$@AAZciLEU9@#-^G3p^AWTgi3e0IkK(5bK-;=CdbAp
zSE~pi=y$tpZmpwrj4>H0GaZlDhEBJ`-Me?#+TLP&duK3xI5)e*_}G+FB0n@4<Ah6x
zQ*ncd9sB9e_+PTxcW)AJtznw`<l@MQ4C5pgOe&aYku1b9e!GwD^og1s`i%x*uSY-W
zV~aj&q>S`EY@$i}eUd0b$L?27g)X~Q3M4f5eAaUp*o$_VottHSv&|p<?q&MA#|z(>
z<Jqqs<#+$>UsB&3p%5!Z`|ndOIhg@ekxQTKacZH!NF_k`Lq?`1FpWCJQjRd}W3=Yj
zsdKcuAz>0a&X_giy&REohRnG%NcqJI=iD-hMX5Z2mm?F~g;Wt;2t40M=JHIRJxzXQ
z9FaPLX6>t#iGqy!)NkD)*WRRb^9PJ}w^%GI9zQwF-P^aB86DxH58okBlI^u!I`t-2
zCkP?==l|k=$C=X)^T;EQF*iSpv4{LR4*{Gw1mr_X0t0ON^f`B+H!@OXe0-ERj=6mK
zW7?e-Z@&39{eCz&--l(2vTtmqOwGzYj8!O=$5VNX7WA4Owl_9u*Bj_Cq}ywg=!iIR
z=O&ZQ&{|WgA8_^RWmaylu(h>`Hikl>#F;Y>F+DTycx7oupH13kKZjE4P@Cw}pYgwB
zi73HS!GJqEoNP`jubu}=Qa-ZC_^Bh9`>@^OrC;ppnA*xUg1a9hu78gzZH`Trn3G-F
zy9b;<d5o2ntAxD{je`!oMw@n{#zC#d%P+smuYBw8a{AN-W@o3J<HCd3o=T-;|MxRv
zJAz^{&*IV?LR!|=?og}m@$scA)M|~-#P%#vco5scb5>s(CPv37Rmun<2)jMDcQ(<+
zVY}mADiAJc?sj{uuCB7Pv(4`A-T>QYrxzF<nMmbJlEL4fVLR5|H=+AQVf$yTxZfk%
zT1UhNKfCL*{eV4m*ZWwd7(aWO>ajTtj^mdAQ$ei0@xVL`NP!A_9NfB0e)%1^{#_=6
zgp-pw#^X9sv%`hc$GQB;dl(&au-&9vZ_#blIjGh7jo<h!zWI$`<LsG-n3|eMsYSzJ
zRt*06)9-qI%Cqv6qFgSrusDY?n&n&9Xm^_Y_{Z<kX?NW|`V0(k`$0Nhy;RO4Sa|5f
zQ4yvmCMlFlV4>S;va_>=u`$}XDQ4xVK_RBuXmIb|J+`(s+1uO07=uzC3-iaQRK`+)
z^Z35%{Liv}u0e@-XB#;>;;eQCFM0?nQc64vrdZ_Ia~HY3xr?p!202wAAQX_3hVjlC
z!QG#b7ZJu&oIW~5*y<8{qg;6EF*@BAZ~WkUeC6e@vvFq~B^8OR^X+fH#>+2#g+{%>
zwQC;_0vxvvj}B))m&@F<!#*gje#VOqqmV)&hsCnAG|&FtJ~x-IGC4NQrAwE2>7{3|
zDbL6brUkOwWhh0jP0Yt1e@Og(OyCFjevWdv!Xq=2>>up$?z`{s$}6u>DC8)W%19;Y
z_xsdp2PBC`mK7r-RS>Qfqjf^R7ZHY<em}$$ZfZ+wJ=Cz$E=zQJHnA(+egPT$yZ_<e
zGdq6-HC{sJ6{b3dS1FR@4KkO<A}B;1{Q4gKB*HZN=&+B)V3j2|ULiMLL0U`HZetUd
zhH8^wG_2iTM;Jk=m_y<bg(26rs$629BS%kRBg>s@2OK>*%HR8EFL3n80{`aU{4v)*
z)LeXQ8uQ6J92={lO^oLkxqPibIk22OK1!{=%h*_zLZRx2*CI!GbcE%P-{<71vvhl1
zd?B$mq0?<LF*C)=`Zm54?1pxbb^f9)(Q4sUt9XNRZm_69;-FF<rJNk+=(CS;_uFql
z#Nb>7!V*bA&ct+AK0<z2LE1eEp62Z07#o-FvR|3y{0q<W+7I63qYr+<xpNP5u(?M*
zS73L)juZj^?LYn({P|z}N47RM9EU7bRCkk_?u@(BJ#4c+c;S84IgjI*QmH_t>RNrx
zR*gG%m&pftZr;4jx$`Go;F#i;#irIk(p9mtY6+U{tqnSz7C|nD@B56;&rlv4rPXTi
z(I+4C<dYW(d_|#9#7hU7q9_^+NL4Bogpz~AtW+E!=NH}Aq}A|@o{_a&ACx-yA}V$9
z-ZrJtDxN3P7VP~&vBP$ODHb^L^kXb<?_=t{0fpWd5DCZ$!)RxNVC{W!BEou#lS`9C
z?H;i=!nr3NrrT@t_FHf8{PSO7>)s|(DzvQg{qMiY^Uu9Nv(ez@%`1cUbNVJ-K%QDO
z4Wu-F25h%L6ousT4%-(NW;xhD;P&n7OpH!(`SLZMd+y@Tgzd(-B;ccu5`5ps_i~g=
zWzJ7ea8NtogAd;0<(FS3pU+V$mQYd>_CxB8y2ExUsip~jv)Oc<yM9CzIV#Hl+l?lU
z)AyZWyH1GX#5oClN$uCL*<ktGf661j`45p(Q+U2N<dr_KACv(xTIIy^PjF-Rd(er~
zc*^1nK`fvkG-Hk11gnCah%xyb=N89l9d!7pJk8Ok9;4BCjkn%-ooAnWiH$pJ2qEc2
zTYUR3zRyc9y~;uDfH?jr-T!v*Qr%#W+Xq>4mko3c_QOy4oV{L`LSdd_ser)JY&N;Q
za)VO7%=POxdGyf>>3$emMrM1!BD3s*IO*~J`#%A|_X7ei$H?d?k1o!!y0Xl*>(_bp
z)vr^?2b9ZYoRdb6TD^ug&Ou{zbQFZ8UavU~+P<dWj|syNeE$%K_Dcxrmf!dePs~ga
zcsT+ua8FNIS8=iuA#xxc(RO5Uk@=^eVDo!#k?2&R--67+xhGHX+_R_1dkG$jZa<`$
zD>7eE9O>>8)*3wW=xOd;d!0*{-{<_fbF5sw#=@!7#7Ucb_wI7^$T7~HJIChMKBn(J
ztOTSmj%aTTQX3NEbfz59H8tw}44}8h@p0<S79*7^b5k<}!e{qji!kYP?fM-qKKVHL
zLLN(IFXR;DNnd=HEqI>KnX~7SLK5}+)ap%=M9}R=+_-Vg$*#1!RLUcR$<=PR%?BU7
z%hvW5o>$_;@pDuvBj}WSXN;xaYce<WEP3BaqYU)2tg|>t5=|5uqS&OS(VyQwsP)k`
zT34@Q0zq6XlbfDr>huvNP9LQ(K8BP&mBtn_=@ECj5JpIAu)-h|6iYdxZkJZAfxuOB
zl@u6liKCFc{W`)5ss#^?q1(-IsWZ#e#R;a%B_{kFv(I)pdZNOSBQw1A{nvQ=FAtcU
zm}ksJjCJ3kFqvSaV&z7iuoLt26Vt?TM34&@85_fR9_@S)TODU>{T8qN`5$uT;xnAT
za2}~GVc4g>yTi<QiK)>s*7x^+VFNmVovR;Gni|9R13cv&YH_$KhVT0*Auv)h_t1IT
zcQ$Bza(m!5;9H3nhLMRJFP)uaMAY%EqNih~C&qYwu14!1Vx(N;+_|H?{e#Pt%Ogxi
zV`v>RGPB4J-uy9NdHGeQr{)<SALrh^4G!V9GrRXZ$DlRNE-*EJN=tQMKI@!YL97#^
zI6`T~$>YaZUt8nu>MhQmeuzsSU*qJdV<{n9r;ZuYaY1c14wXp?3EJ%z+uNIj{g7@q
z#Pf5!`pPTZyLX5Ey&bykE`?%|An-BP&};X2@BMe!J2;?F80F;2HI(NMsxon$u)TGU
ziP4J$${Qqa*@c!#o<x!R+9c6@5pxo^-+qlprzi2f9D(wO*lu--?Jlq}#!{VMVD{o8
zY`y*yr_BfWDYl<^>=;iybt1)fk6yn^At*3gRxEZ82<t7*KXj6NH{Rm%)sHxH<_veQ
z-(>E@3F4%~+WI|~7LRiJ^l7$t4lv<iY_}-iOR=4w8QU=ejiu3QQ7u=PnViBGK6~{Y
zqNLBw<$FB#_#@=<xxW_MmFIKn^jV~E_GI;Di$n{0{g~zD8~94l>+~p>DoEu#J-fX&
zA7B1}?VTMwugI~ZXDF4c173B8?K2Zkk^5=bt{t|=vB8)x2HQV##W6R&`yDFtGaP;O
zTLgX{Kidyjj7^Xzw;$38=;>3ZsXw;K{*OK)vW60lUm27nk3M&fCm&fL@HNV~0JoSc
zabn!(?*2M5euN7To#ea!=_g!Y{)A&kk8<bgHD*toqSLFfxwXNG<0skMyT|(Gp6jZl
zd%@lHJ~nMxC&ps&QVhrj3_c43TwA}_kMKRk()>I-DqwSGm50th&L@|bIe-2f%9q3a
zfC7;nwBddTf*cnfdIX6e?DsjSHAoUcv)$#^@=bgt2)lih@KbT@h_Kz`?H|9v{z095
zzRJlH=g1dIgZ-ejLB}1AFHSq0HA9}|&$525pziK>-(YU;7?q2UAOaV?Dg!$0`)4Af
z+GvuT&+JnV5$|r%{O~$f#Ec%P^2|#YIeLB$D<aT_#29>~NP>XTu`w=m>-_kW>s<KF
zUuEXlCyb4cvbMd;-1Hdtu3zKa$y4|_!QMfGR=Z1*#HsW`dIr-{rAd+@H-W5|p0!c#
z|6Bs_e8sWjM=2Mp%+5^GYaej-%o%R2+#(D+=)Puucb~b%MR$3a1S5?L2ElQBvY$gq
z%H=Y?Quw2zOixb{_(cwC`)qBkQ>|3!_qr7FMTCGPPT1Sqqf#E>;R}zWRKSVjXE}c2
zJpFzj&-3W@d+hCO;ECR#I(c}?&syzTHahjKO+WkRUCI~;6I2LsJz{gC&epxVoZQ)A
z;p7QQQ{#-*>I88_-03;KoU)*tAzG_Zqp`W|wER5JRVX~S?cxNK=V9YMWhLpweYW<e
zVSa{(E?gkKd4+g-i3<;pGk*REx9{HL_x|li<m3_$pPqnAf575c4TNQFy~EB<hiA@@
zP%H-28hcERPor{qY6w2sYvJ9z!ngnZzva!h-UdI%$jlVCZrx^jae>_KE~B1hr`7z0
z4(I@`{K=m%HMfMEoyPM$5GD<35+oi%NQ`j)r*dq9r5B$j-rXbKKEPn`@|N>wkMR6A
z9;Y(yVUrjuoe+=j<)~Kj966}5dGjuhKl3E-zjKXRW0z93M7+L@)|Mw9KaWoO96NTD
z>o=xdYhDc7r&t5R=`mug8^9A8jX~d+S>LxT<nnp67EDi0B1t%N@(k-+>vXzJ*4Ja|
zwKikp;|MImI*TSN(uz>pf))Zi&*$9vhbZRrC<MJQM92W66INE1DHd|{`+agj9w}Y&
z)@rp_-&`li<>~jjg#8YSizi4DO)eKWU)X99qn*I){Zp_i<ppZ@D@q3zzJLpU^|c={
zJA0J!W9JdR`>e`7fbFTIiq3gVKYpHgZ->^$H!&iix>(`jbB}T4%nVj$*sk%E<KvBt
zj&Qcu<h`pmIrnSdWa`LOM#o0j+}&elVwCmeo18v=96yj8)M~WbZgAi}Y{w2NSBH6L
zKMS@WJ-S4xP+@v{oL;-m>60f}xqF8&?4m=>!QKJ0^Yf5myZtM$y;LeW{lt+`CZ{Ix
z{Q~v+0Xy59R4Qfqy)OA&0f8V%6885GD3?Y!_wpks8F1{_DUKdJLl}lA&!gY(vA?&4
zNB<DEXS60kf*$a%jrroR{j-(<dd)7&fBbDqBhv(Dk0XL~&@toqN!R)p(gm$aF~`!2
zPmt{G(!X&RtrF&rkMrDDo?`CA1eOS!XtV`S2=c`O6XT<tX*alf^)?TE^94To;5vn3
zk?s8&lVcS&mT&OT=@SIX&~7)WH`*@f`HZ_>Ywhlx!~4Pf^!<>_`J6m?lyYH&(UBr1
z>2vPvIquxMO&s^x*@|d2TU5s;Tu7MR^&&Na8%Vu?=P4?cGQRTg<QOwE)8v8@>l^nt
zsO>U7Hcq$OA)ha}7JWbDpmxCI_zX`y`2qru^XDIBZtfUS6d|Rg+v~Breb05Y5<Q%V
z{>(w3Pp_XVs8h8zAOG<maV~Gk&rH%Y8jSCF7wO2o@En0T3PXycBb@x@uW_Tc%h}li
zPdtB?VkN-n9#VLzJ2KcLandH$D(5uZxwgTF-+qI$r;oFI^ESO0Dy0HbBct5Cag$?b
z&yd7DMyg{>j8)m+bER#i4l9E@O*e4BuhZWPG9vY`l&qx3v(G-w!w+BJ=Cw7(#zwHZ
zMY&Yw%*oS~@;=>W%*F?=b7bTdl93s-vsWGPI9x?mq+`^eh0|7wjlt@ezU^@3laJ{4
zT7+ST%K8Y^Y87i;J7{@%nW@Q1dhHGqlLeHFsn<66$-D2;YPHzk-@}*iORs!2vy&VA
z{D)gKhg;%bNLxl2{FIZD6CTP~Iy)`yym^V;Yb%^Ny~txH$B8>#$4ALYys<LMJs!FT
zd#ETvC(grDBSA@&kZ36xt>n0Mr$J%_I*iy=bBrCI<l5V}A<>Nb2aFyWWqPE@l{ao<
zyOWGf`N-uP9G6$f`HGF*CO2*!aOOyXnYjsSja@3GQSyZ{wW!Uq@%i2xZ}P^Q@3ObK
z%jkH8SzluFIjR$*B;6im6Z0S5zCkzq5<M0FhOF?$J+A!GA9DOReg)67B*q|(^S6|T
z^p>o3gCwP6^PKqRi>&<Nf2LB>JpGl2Id%3ZBIic<Ju5MB;@C*OM`>h~XO8Ci{i|1K
z9$Dnc=O1QoyT#QT*LdW@1@<?#88?QJa+QxSz0ca(CK7{kT|uXNmvM_smbIn2ZRvzL
z47oOgmXVMwF3$7POV6{hy2tG77zcZ+%+Jg_9+0xc{gCZD*O(Z+NbKcYfx=h@qD6LS
zDJ3KrozUsDkT?%k-_zV#xka;4LnlS*wSCH^64qLxDB{+wTTDz$aPI7R^0^`l3r9F|
z^fc=m8}xcTc6WE_wwm~p6*s7HUu$vLj<v@9Z$D24|5q*5XmjPy|CG}?OMYtHVY~KI
zjl2|Bl|%wz7*ZUqaO~BWxK-ch^h};do<2>n>|;zHX%$u*7X&2QO+r?xoY8{S8(Vz*
z{kJ)N;s`6Z@6t=4T+B039bs+xHcO{Yk|aGU)ltSrD;yj&(1~^}|HC$o4-A-P@B2Y)
zm(t_OC!gTLg>&4xvCino2v)Z#mCBqveu84&qt{B<yz~}})tAxLDUzRx?R*;BO+vrh
z;rf+N==a-%QI~SLN~KZ(EO8vOvT~cr@d^5!F5}~Qlt^gQxA@?L4{5hs9Mlf*B)t0a
z3!j1QP8~rf`aW#`vZh~+6<Fh8xAwU7hrh?U-}(*ml_H5w5JC;^dMPDRDTJ|vVMuvq
znv>u98n@ejz~kqpdHCW9azziL`>x8B-t|Hoq*CO|6`ol1xO4eBH{LFB^5i0G_tuH5
zWVCV~B3N6##gWsePzp-LGLz#~_V*hYZ3cHefRf4u`ai>d@I0SaU;PU6^Yh%gvB}KL
z2>W|$OpH%pO+qQ}QQK*={o(6OKKl|;d3?~0&jzw!NcRv@V2mb;dPK%xbwaP*;>yP#
z5`|s1w);#>OmpPOF~GHbZ{NPn_~<xMugBQv7{Z3^?cU?P_uiw^>9D)E%V?#@t1rKJ
zKVS6AsMJLy#H)9?`lo-!(Qmzqo|qy=BmJSVwN1Sq!%oYIz%I@6(yzV1F|UX44N@55
zZcMo}Lf^*NFm`6v7`$9SXkuPGF~*<Vx=lSeMZLC3F<)VOdy{gd%t*1!<qtn#{`hfz
z>6_mmj(c4G_!9d&2W~>hbCY<5Vjiq3gW6<Z(d+0D0zf(9wYB(xkLSUyTPyfpnIQ0)
zm>2~M=g&Vx*z1wRUE0?#(~J-BFFZ}|=ow5fN?+=9O3aBBx(b17*I4UbvNZ{NyW8Bo
zyG)q$h-1Nxn^&ohRH@Z#^rMLDH<tP2ldHTgf5@-=^4B?g>KHp)x7k>`2~v<G9hC25
ztj5}4&<1kkQA?7_oESZnJNP2N-XOpVJgG<!U?fT66#mV2lk$8M5#=!b2op$33lkV?
z$rti;w)RN6A->8XgFJy3V6nt;9}|TL9a71c=th=SVyMIgw$vDxuil`)TjKo1IbQt6
zc@~b&uzF{c_ug3L*wQ2;QJ2ZupEEJ5>4Y&KU$3z+;dAE1H0@pkU-?w4ql6f)9UO4A
z-y&$Yh{6uX$45B1Fv>*HXX4l?{6dZ!S8wpIFR!xE{`|`P|1~UY67K!v$LLDH%(G7s
z`5qb#N(`$uN;<#ac8C1LIMWv&N7XmEIFn;+s(_R|NIY_Wi7@UVjYWqM0`SXayn2ln
z7IXZ`_g=#+&a-jv8YAN~Y;A4x@VSTB+uEet?{WO-5=%$_KEMCF|BkzN@6l?s4-E<k
zWZ<XMi)qV>EP%^c^=S(YOAzFE^2x{1+K|tc$>;Noj+Vi~Ll+*VkPB$ELiXPNE}?Ex
zxbPH7v4X}=d4(XPHw5%Fu(N5Pg^s8<nq0f`F>%yoqfx_J!NkPGptZWPa)<S`b?)9?
z;o`-|SzMUn+SPaX%kO>{tu@_#hf1Ny>0`5C{gfkQ4^7MurA{(FUFInBc~(q9vcAfV
z?|hr3SHFrLpCCaa{6qFx%2ivVZiiB-$jHJh&wc9|j-=Qwgdyq0l!{gQHX(^(L@E#B
z1wN69c;@&B-(I;x!#_^FzC|HlVP|KXQmIU}P~zIhA2WCK2(Nzat0ZxsYgaCFuvbG#
zh4Lg;IBXFJTJ0Xu1K5rLc?jDB-$!|Zm6bbqUWp*^7#|zKTFcop=Lq|KlB7##`5LXb
zhJWrN!O|&=UnRK@+Z|sIA%@s)9JcT8A8>DNg(&F}CxYeMH>gxA)EW)KIOgW^3RmMB
zy!G~v`NpeX<<yCz?C#uUbMuy)=uA4uR4&Te)Yli%!3jH%usSSd^JNW6NFh+tBeD_Q
z<r`f8&;OZ)S6(I_86&`lRG3=`f%PQHTB1%1Q7JQZ>Ig@E{UweRB82A*?UQauu~eaN
zB1EjQae~F-=krAMI$t|A#vk9jjDP$D`}?bmR;O59U*r7Qv*fJi@`oR=bm}C({&)Tk
zNfL7N`gQ924LsjNDT$R($QLoj(rC6ZNjkg<>3(powAmRc6bgi4kJWqY_+Fk;xrk0G
zNU1n`_B=u;k~pSy={*{)ZE_DigSU7J<CTe}b`$o}wdS*AOQh97s}rnFSYO{@bL$?F
z4v7=VwHuf51CM&6K`)H>_>-&XIN^tHz0Gg^=C5<+)G^lAm)YFBm5LKL(I(}0TIKlZ
zR*)oLf}@1N5*SPS<}FsrMP^@p6*U%+Po)I1Ngp8uq0xA8pNVdRsd1lNs^4T1PzE~n
z8inyHVH6{Qq#u$K5`Su%-knu`^>~#(K3M0-{88S0{}Lmm8vTBYlSfZ8Kexd34?kgV
zbC>6zev$p%9cE|d*xTJB(J>=qV~kBqAO)<gud#A_9cv)S1$dr7Cz<b`)4Yn~koEOl
zo_XpeqQ0h7$`Pp&{Wi?Z%&~Io4v9%v4tqRNTP6JQ9^G44D82A1a(oGCJoI5<K#R-t
z(o9Y#3AgVo)9lo_b8m&?#~$H(uYI5Y<=_81jvYI}g$ozx_97b176NFsx(Fn`@|m2N
zB(Vn1bG!y^&Cr@eyF;JEI_*Ze6my`L^aX%D)>*|!QrlzeoH-<d0H4{35=pa(4M2_+
zQGtgq6>+amV{aGV%TuUU@p6F^D77&rj_G$AblYt(nq09=EEU~Onaxn5N(rTz7?E#N
zYc>!Y5pRC?3N{KjbF7H>{(s`+`~gP7hnJg7l_d|Kokk1jwj0b$FXH7qZmn-|H!e|}
zn<cun$Eegiwos)yQl^-5UbFY^t?_^RfBui$-K&58XZ2sniqQo9nDsZ`#1s_8M;^zM
zIRZ}&oDHm!C<$TQLztLRPjl|c<CLQ&iAfM5Anx{&If2Z3#I3GN_L2x+KxK5CMcd>X
z)0Vf17@Zhn{q7yAV_n`$x?DW>2nWpupS<@z=N@^OufFmP{`^n>h;p$=uiM2~O;E@&
zJu^cP<k;BQWPN=PtUyRhYQ_Mh`gO+XY<~Rl$DBHOo<d=SLLtESMo9XSW5-XhwYg2C
zyKJ?Z#GkxDbL|GDGf$B}{TN9uhcYS~2TdpITtCKXS30e-d#fw#)pt0k?=e2U#D^b$
z!1d+p_<jz6duv;?+8qRz$;nB+_OeGG%fkFD+FIJ}mXke5*+bR{j7i&Eu|_8rH<3IH
z>b{5zUIRH}=`62sw^U^2D_<cR@rT%+;+qhX$Rv1ipRrz($uW<d$zhDfCLlF*8+G!d
z6{0xB_Atfvi7EQ4>wNP_l|Qd-u{gKH2OnK=*dDeyzIc+k**R{0a+QPaJ)XJv47I&o
zrl)5(*xM)3G1bvAM#skx0yZ}{xO->AVSC_$H=`Y&c!2Fm$kygA7oT{ZxUVS|10q!<
zY{S&_EO%G#k?4e#sL!K~4WjoB=&oF+_{__Qu|<^eFj>!!L+{)etZ^E2ckiyy?lic&
zc9)|^F7U>iZ}5A+|N9&{a*VTQ&(P~fG+S*1&~A4T2z=!+F*cE6yW_1UDYn~87R5Mh
zH)dFM`a-e&vsPM>GFZICQoHmCx?E)H#TSYE9J!R@kWsBj@A@F_Fr_<8Oa%lu*9VM%
zs88QC$W_XOQ3qiyCW^=@g*P!p|K1wEdbGl~JNrzI&v5hlZN|oG^m{GNpFYpj_%t8C
z^FE8mj`7lSFVkw&SX^9WXLAcHU~FQ7k+Ctru(Gnk+UgDhL6GxNN}$ub5nNTf*Xy#q
zUE|3oUZB(Jkjn=YizP5HJ3G(D>Nb(lEcZK{-MdZehdapGlN4Wk6;+yb)}Yyb$S9l^
zCxfFkaU5~$_6?ey26tBPaQfsE{J|gnKL7Gx{VN`N=n<x-CTX?1^!g!62)exxfh6Y#
zOioOptwBmf5adX#(@GsAbU!0`_UH3{MS%q`g8dJ!;ZMx6@WMq1(g}??Mx~W1r6b0>
z2h8a<IjQMcgAxKyDNJGr8*Kv5BghBDanEf6-<9}FV^!|0?(@u88Qt~ST{=O3cawva
zd(;o=Ox~N~?CEoCukW&UdzFh1Jw|DKjGgUWo_Y3p$|GaE{`#AI_`y3g8ZA8EBd0tT
z7U!u}%g(MvNc#Pd<>lLyN@c$O%CC^mjnHoIQ?2Idh=g{%iPaExV!~brSppMj^0cvA
zpP=4JP)~o0{`65iyfk8W#!A@;x)G9AtI5X3D*O98^m~1h#ISPvE~{&692_+0bUJA5
zCZp_OVNsiw-VceOW|lFBv$n${*I8-&3v-mPZfe;|>!La=Mp&$bTuD*Q8H@+{sd3it
zZ8)#QB%#-A;7Lhkd<<F4BVtXG>JniMffpc3WpqEJS_&w5kn}zJeV=2Gj?rmqw%2QX
z_m4i{*vWBr?+KoMcpl#Q1I~@FkW)E6zSG3UO&)!GmeGkxnyor<tO-gL65r>YdmYNZ
z^bKD4%2&C0bBANmHqCC6vnQred5=4*tNdU7@Bc4u++O+I&*{Ie<@}vxv(Mch{Dh+;
z)A(nO<D<w#ucdXth1QyKugOf(qD&v-OMDxEv3Nq#JE&8go<IbFtLFA2@<D)CF0;S8
z$FXvrT2!MqHp!jaJ2cns68HPO-HAAH;y5FPDsO-14UU~Y$uk$9p|2CxSJ#=Gnqqu%
znzi+H8udLa1}QDdQ%p@yP%P#M`yobK!tnpk-hT&6nx*M|*zcV4+4Py$-&@|bo9>?O
zndxyoZoPwD0U!VphN20P2!(=FFrZNA4~E1DMMxqTB105x5F(ZoVMr{1#e%z-o!PN%
zroAm+x9+`lZTig59-}|Ltm>ZSEHI7j+1Y{S?Z~LEh`Lpo`JMBe_q^}(K2O15Fd_^C
zKKkURID7gbj*s?fwIb4)W;o5L#dV5QqsyEVp-Bo$ZLo`e@tC|E;eF(b=pcUI&71`h
z`+j=|@;u|t_N~P<?H%HHm2ZCI+uXQ$lgpQ{@a(gn1>WQG-><F<#KL#7;;-Fz4|qSV
zqT288ackZ`?Q6bIse^ynGj|TW5(ckb$M3GP@yRDEY+t+wb)V;D44rhynw=9$L$<(n
zwZQglQenFv2Iw-aIwcEiuXUR2?hbjnTgRk6CtGL9PYxMv@9^qy%+l@>mo8r6c)!oy
zojsm-<RjEN9gdGr_}IrjL8H~-Ti<?;H(r04$#{n6RoaQ0o9i@Nb;_bZAj$KbTer5U
z*Xw-#*`Fo~Tg+zzn$3_z80O;%7)PEI6j_3-fh`RoGdOx1^|D4i{gdR&r|}k=h|USj
z!jH3}_h@FbDf<U|42CCUS%xu=?K`{d?Je}^l79iVTk}4FYqzM(Rju{+8@k_bwcdBj
z_^`13dqpMXA;2kc20wEgJoh@m(mK({9)qCqz_sZ9S?}wTrP&GV!5mLHGGp-;G!3L8
z8_g<nf*`=?44oJFVTkh;QM<w6PM=S#FEPp+6lX3lKiH#x=MLW<4q0DW<Kl%YY`=RO
z6nyIGPZG6S93P+XxzB$Qq|aBr@)h2A{Uzq}v|0~g!0FRl)ansMUf`)poACDSyDTp)
z^HV?Zix{m41CM4SCZEr$&uk!>W)#^B5y@&j*fDPJ6}%TL<`X|n(cP?)ru#!%i|VbE
z3M({&L7$`J1CEaOvBoi<rQEr7m*bN@^LfhF)@Jo{3p+l&P_KaMthc;Zk<~^c-!}pO
z{Z`~ZSf%d1znv0<xn=iTFVWgu#(U@?aFr~FGnILFt(elDXH~!3Ieg(zafG%Of=a+}
zb~2#8zKkCPcyY*VGNtK91S;h2-7yabicd$1+tN~h^ci-pzRrGskArl~Y&xR5v_arS
zyz$B#tZ!}cbARZU_{3*Ehjl&|&wYgFu6%^6*WScfN3+>xI-Rh8u*>oB5mE@6%^G14
zQ5Kr*?QK?9S6N<ZV65TZy+dNZ$=y45nN8-52FKLuphOLAC7yEl^CR-_{2I~2&%(JU
z(O!VC_CYze8q71?zH^J=s84@z!j&rz^YFuu@%hhxiL2MHVob^N&p*!-Pdvfy-ah+#
zdz58~)|&A`z%UF$k|ZTBaz^7ZVHm#enY`ym_ucCA-z`x3uNsAO)b5XtyUM`Z34s%m
zPA4YtAYNagECnVj$VXE|nS--5msjx{aaBn)7GW*ulA<Wcj3I6`sJ9vf``g4)Fm!^N
z>e7jOY`=a;F<)gKr)+IBd1zCReDm+~aN}*-HOaxWWM{w6GY@xYb(cU%!YC&88+e}L
z-Tgk>QIkLag)gzXb(T;6^iT2j-~Sa>tfLY5+&nnu|M=hjZEoGZ`+NS5{&frI1kP2;
z@a|~L&exu&vDCt?Z#=NH6AP8}u$-}+%;`2XN(u_9Ny+<qRcOQfctE_|t!}uUBrgg)
z1WPMxeB~Q&bNXDDIfBEpX9@7Qed8+I!y``iw>fp@0&A;VTz~5tgoY<S{&9ZtOFzS>
zKl^!{@Ok5nxB2y7`&Dor>ucxH+7g5wQh_c@e9s5qwbx!p3WcX427?p2OLaV7P!v_7
zbaL1ynP$w!1J3nILL1=(NIR!-<25p8@gDmu)YouARn~DrV1-y{R|)nH?lBk~b9i*X
z>gp<;PLEoxNw>Sics%Cs{@uSvv(*G4nNB8@Wm&alb@l6g`i3!vq9`c+;JwO>eb43K
z{vqHk297?AeO{Sm2vCBsu<X6?8qM`3yvvthA-Uvi#aoVxIjwnyw@}IVg+oOlrL%~j
z5<s8#htyV=@PYsphs>ul8eT{sLw0v3TnQx4gp#|`QG57lcCWw9(O{pWbi#Z#q0?I<
z@Iu~x{cTn^Hu=e)`x%~o=F>RmbMDNeym;{uZrpeWYoXC>F`Z2~Jlv;$a*Py$z^@U8
zhm?h8XJ?0%<z<$Z8yIKV-#;Sq8|>WOWj;+A5BtP5@Q5*1EwKHV{N>*uy7DPF^Vs*m
z_Uc*Oy?cl8c)(!L=hCIiT)F%ZpZUzQ+_-TaV>K_m^b(Ig`Y8MRhaBwhFD8~Xlj#I&
zEkPKPq$x$7F_}z=g75)v>;47mzAJ&X?nA})?;RDbr*dhk+Km3JVEgMY&|2@|ojnIu
zlFJON#Z!WIKB1N5Nb$b)5J!~8AR-?*c1%zDG}e|;z7K&%mgfXc;>(D;+hd+?M?Bkz
zIXr=_N1x%=)mJz<-eETxGae7=Ep332y!xG2Id}0Qzwk%?Fi$-F3}q>}eBp7v`qh8V
z-n~0$<7l<o3`Ya*-P>U}7$Bvf-45{m1JX2O`_3I&?Ix>hEws_>?A+trsY~45-eo>b
zm<*4p_ac-C(N++60&lcK`P@ID_K`1OPCbGVUZwKDeZF3mh~K<*ozb|@WIX1HCqKr<
z`onzw3twXA?hf<$l$T$AnNNJ;<Gl6uRYrpW#lq@uJedHFFbJ3lL6&DsCsVqe&imHG
z51>+);|PEVg>`~tFy`iS-{JA~Ra9pct#gFdsL-&{9~0PwLe$776I_<!HJj8|d)26S
zBr$15cGRb~x<Xj1Gul6<(`z85<1-&Wjma|_cFw<X%IBZ%ZL@y<VWxL1li7fDp3px&
z#Ba1vevSQ|d;H2@`|Et`%U|Zvr=Di(!Ud!&IJL1wySL2KANx2;DQ@4s&HnyAjYgfl
zdpoSIuJg_tuOLN2=ok13=F=H(y!Ixa{KOa7I~XuIzK2YYxEOC>bb%~EHR@O=2$M0z
ztG|KM1zdavY>K72-C|0T>6BY{Z*g$E&vIv(&wct!EUlbkX=#ZUUwWS1J9kLuBR=_w
zr<l*??C<TNwIj(2?j7!v7dhHgesQx|!p$4ESn4jL0uL*}N{7?;z4_dS+L8fqR^kN;
zjm1z|GfAhAPEZRhK_e#aEwlIX8`Pr+DJ&MKcb5rTEh=MjheczZLq##6@1w27Nkvg)
z_!I<A68Ju=!eg_0p4Fc@%~5|wIvvx=9mxxSkB8dVSZYFw;PrQRxN-)TS34LizLE&f
zBMg0{)jWUmnC_#Gv%YzjgOej(yZSaesUuX57k7sIKmPH5$F0HSL-~&Wb&JwDe83lq
z2*KgaZFXONopWDo5k-nZq&QO|Ld)`Wgm33$BF3gUm=eF$#B0=W9>~Zen@o_tB5XDZ
z8x00K`z*IQq{W<#UY#{1=|z^lNO>n~uyOhV>F&Ff*_7*7-{j8r9lA>^2<P+lFMo}_
zJKJC^r_WvH)Y>|q{q(b3yl@HW1w8fi)9A9~-K+2L)|;>6oZ;l;h?SL9#=`+8$2(lS
zc!4O6&_?t6n{V>Sl_%KV-D5HyF&iDRb-Gs#Mr4K{3^B^$ZNG&XjZh!|S&Hr{q)Bnk
zt9VsLlV>T{ZeHW4e}E|spMK_9E?xR48=D)v^2$q`93AnsuYZ{@e*Uw(^z!Rmn@lLR
zA<GMP@9kEu5?+Y4j@c~X&YinF_Q)fZZJw)K5$yXCy!YSSKa2xDr+Vfw65|BvaLn!R
zyvn0%%c%AWS{GmpDm1JQ#sn^<n8y^;DK5+Knhk2p-D)B-l;|WQI~fo!FA>Hy#)o~{
z-8xb^o_T5un`hK)!XMl8_|5$t*3MpHwquyhMx;r~pnrs4Z=(E|!@Yg}{@?!p@ZxWO
zjfb9ig3WX1P_ATieUo;##}iLJg;I*`?QISZ4`?*%+}qz}d1aNWZ@-QdDM3);d5}zJ
zy#3ZYeC(;uvVS;Yda{qqjyWH%VRTtxd%cE35~dT1H@;F~`}vQ3H@4f7bT;Gm?j4T$
zhx9r<KK=1$=`C;4?e%!&)t9(;_b%yt%*UU8lCOXLIS%&s(Z-VHB?m_b6lIRFmLe~i
z&l7Inx<j|Kgz`PCct5sR&)wJ$7u&ygI4qXT-9PI+3LGc<hunVg6&_ydp_)tRqDtz*
zK(o;w;0aA`1M=|%SL6iU4z;B&NP&nI<#0@PEQywT)EZ4r?(9<UE#Z4}o_*#F=$wai
z!XI6O-#XrB>*8ZfZoE!5n=+rxI5|GRt2OY#2HV$f^0)rW|B}mB9^#Rwo?>-ti&`kz
zSleKI^At}!`BAL3T)TFi(P&5<hupn;hi<pWog42Wm=T9Hd?h$OI_Bn$TWqXeWHg^J
zI@qRer<@JKiWBNUhIK#?PLI(q{nJW9<@94<5-ioUJGx{%9CByx4u?nkoZ2|eXP)^S
zwdOKQD?MI(@i}(x+{T)WPkihtw(smP8Vu0dF-uYoj`k_ag0j>UMagtL<@W7uT2HnS
zN>$x4@%>G?f1sq!7y%v_=_rLm7{TGY+nj#)D&eQw2wTJ9v4mqqXrfYM$dHS2Vsdzd
zEizi0tN3w<nFQpMifa&eIs{=zVKlX76VF%ZJmTc=kn<|#-+N+{U%P(Hyt#xF5l5pV
z%5sV>a*}z$N-gHx@(TB^zQgUSZ{deA&9x2IPHhqRaP6Jf0l~q+0cX!$pxy4Wvb;pE
z(`IwEvuL4MP7V%u?X@@AT)jv>(#%Ht9Nm6{$2%~-Gp9UvipcZvBuK3Rhe>7(zW%E;
z{W@;*0>b4Bn<33)Jfwegh%Y>T;&WeQb7hIbX7rElar?$Q>~3%KOTX}oT)upnZ+zo<
z0JJgqUcl+o=ecp?Iw$>Orn5OpR%XFvS%N1YWbWV9ig};I#D`I$mk1(ZNUW$PR~PpN
zCk(A-jn>jR>^x!CKc=<0icpSvZ<(mu#X5srR57e`7$L~=axn>I$cvQWY((TMtaA8b
zNUz+X|8|F!vrDv+ImuhEA(B^lthYlWmZ;F<)mQe}=twr!>R2K0JVjm<Rck8tNs5A<
zQNmWe!*n#_Z~WE2%AI$vGPeaglR0}|e}`fD1JVTe*C5IT(^ys=30!Ho{n8t(UO9(<
z`5}ar2qWky*eWxcvcRQ+VmhUq&7eP^xwcH)ZR5v2<ph$WJ`zb7g_W49b3~1Z#~wSy
zd_2L<X8h4D#idg4^+Uyc<s5hSwlOZH$mb+w!g!Lg(e1J)9sk2$|LZim4NhIUNUPPu
z4<mZrRj@fq3hHshANj+7jOFDurqdCxz4kKSdH$<(I&GZIpa^*L^*2dICHWX;`HZ7G
zZ*e-F5>GVR>);86Uyl&lp*)K@xyA5X|Crimf0?2d<D99uRk~z49&vcM&vZQD6Hk4D
z#~=9!$`4psS|SWRoGJOz=Re02A9;+o-h3ATqbuH&P#$KXFrOwVT5AsW54d#UB7Ph#
zWE3Pqymxu~og&1)*DzKf7TB%@NFzDEewXp}8-$-|gNwm=^qgZ!7{WqhT#mGkQX7(^
z6Ks*wTwlhI0*v<_wnwctK@d=AP26apyb9X~$48u18Gqrib$;V!pQOITArU9z6UuUi
zDGHLLU^$LB(_3c$##Od&yo(n^G*{PH+1eoR9XGGN1qhCg4moxD9IbYTrCyhAyT$rS
z`@vvC|L~AE-+YJlmGcy1&3t^w$@bek(t*j{g!1$zq3`1<5LzQViAm>-zWq;W_;uX+
z*$UeiR$|l1h~ddG9*WO@>RHy8mndz)@c4l3+gG`_v&+x@^v`hN;zgc&?mN}@Y$|^A
z*49~W-@ZkEaKdz!;HmdvyOQtOfQk3K%#AS*_Ra$l-XBB=hq2XUr<N9tKxo0<Yu7mS
z@J0M5TL>E>@K|<+WohyA0%JfrM_H82?j2x@jOO|>K@?(?Pcj%)k|K=;%Jb0HQfsyF
z!+>m_kewX!c+GMCBOCncJIBm>o7@|Z=}(SH^@K7_$dY@kH)@<+Ugg#suX6Q`R|w)7
zosBJ4w>D|iJl=ilRg5tl9UXJ&(nG}c1}n=;bXpB=onBuEsvGWYZ}a*aZ*z9@VP;2j
zM*V%pci-mpl4QI)XMJmf&<pUTfU-nLK{-5P_?5p$=V$Ba-Wu}$S#LGN!7+p5V_I>O
zpZwexS!y@Qij?DnU2a@^li^99Kk>(ZnYFcbe6R8b*2WNqF=x(P<i?F_^ap)r^Az9n
z&<hg*wKzZdJ}++{kW*KYqlL7K@qR@yE!h6%Ypk3(gWNiW5|#~H(CgN*CM0kY=f#9k
z6P>2ydBJ>=QEN4kVT505vf%)wJR0p*MLHG&zt%wchBz&l4Nqvy2mG6luk+2jLw@rn
z#2Xj5b8^UN+NU&@ah`H=a*y>{gKn*b_B4ahjO|-@c`a_y?RMz(x-^?j#)AR1R!pNF
zlcWhsMocDCX44r-lCipeo<JMMJ8!dn?*=X9Q5;)bTVjf=5~Fkmp*2RAq?0l8lOxLK
z{t;`x`0t=qP)QD0O`gmcogDMoPkn()XU^lwl)c?O-hTZ>21h#_?%&1fPvMEG-6e!z
zKA$n0&k+R7=PB2&UggHE+dTErr}@HXp9KMVmc2J#{X5j<-+TSuH@+`4o=^+ZVOP=4
zg}@hv)926P+mKoR4)w4yT=t`osN1PHPC_iKP7K;;k|afFjk5(>mzcanYD+DyV{p`?
zA*Zt&>|7hte+T^Xh>yg3Y^<js1VI$=){QB$%sG9bNvpj?SZkn^Uy;a-C2BS)3(aJf
z5bf^qxBlzD#;td+(r(A}Cu6S7^#9-M(EZE88i@$1N`V1{Rm=~k>^=7??TvM)_i#$m
zGdb;s!WIf$I6N;x)g?ueGCdyRJdZGr2%<W9K4ErnKx=i4POpb?jv%U0Dn;9mm}dnh
zOZa3bVk?gL)wgr}jkD~{6Yll*NRt7^z|fT}Cu16|HagFFbMJ(ivxL1R&YnHbo!dJ!
zn@yZ^<jI7S;~k<nq}>j<bny!F`JDND&fV>INej4e@nO=@G52=f!H!1+vBxAVDeEC#
z5LK3EWq~z<WHx1T=Qg_5qW;OBuaX4|!Ww3i5&eUGE}y;3XFm2xBIP(b-s6RDeVbRm
z^G$BvxW!-m^Zz#IPM^i|6aqo1OO*0Cd-@WC{)lV0-{s)wgbU{{a`D2&2RdNy$r=3M
zE^igK3#2p^#lk6yS;5ZpZ?b&q6#m8*NW+>f>2_kQ2_9g(A2u*qMv)iHr#ZD|0~tj4
zp&=WNP|~B`YE}s_0za;!d_$Dw%*O-j^C5rou~nYm8S|Ap5XWb^+dpPJ9iWY6Qe^Z8
z2dpJ^I&l;2l?*3ycD8qUGiuW5wCQ#`G#U*i!x3>aqF#$hvjinVrqc>fvm|3>^(=ug
zO!ls_b8w5M^e9g(t|c*Lj*|jm7uaqb*<?a8I7WZx*IE9VKZjA?;+bpmWX`yM!Y4oe
zS<at63p!<gZ=ZMHdWGTf9!Cc|Sp9Jn;(gdYXFg9C*q(9Y#tm-Y+2-+&JjrK1`6+~`
zBx4@P@+_>X?pNvTVo2h932T2CA@1t~IUl7A2)kGh9_b+E?l)dxX=4+XRtbb<lah9?
zfz>s91b8vQ;;b(c&3sx=tJRQUieGAy{s~eln%y>TLE-RZjk+H)PcyRVm|i~OfBp0-
zU%qk7Z<ijm`bBOZ-D5f*VXbADB{YtASq&RB>n)0;<X|}E=JngWT5CTz>l@7`lhKfR
zvqr5Rk<2GZ5iyxen9b&7Y0l=>WrCt$u>Bf$_wTS8M&t(tZY9JN?^_RLiNP@+4oJpB
z#CQHF-7o!VjF1(#)*6!Om~=AXr@r)aoLb+&<}+^Jyu-V1zRd7skAwXq6cXXwgTbTe
zWJ;1`;Cv>NId8r7CObR(eE##F=hEfNRid9|06!R;Fb_*8Mb&FxEVj}*=DR~~ef2pm
z{nAgeWKvc}Lqbtf1Oc9_ZYKgC87jiSAqmi_Mg{?er|^T2WH2OHYU5Pp!VrcbRzkbi
zAsHQ`jUdc2zHqk9L(5J6yO(ZJ)FYO=4{<y?WH1|XGcDOkhxDc+*1~{BP-8i6R2w8x
zj1F>+_wSLHCB_PjR45EW`FLJPh@fFC!Q_ynf57hX0jcuXR6emA(}<fGonxe=OjCSM
zQ5FTx8uBuweta8svW;222w1Gs#Bs!x%NJSdwz+Zh8uM(<Xgmfq=gyts+_^IdVYvG4
zJLFjbLUDV0hv%Pvk<HDsgh52N+avJzs%fvCZJNy%vm{~A?^gnr^1dC)eFk^+yowKj
zKrI$tOv2qCw6DZp8}*3I<rVU5$aHW(qt(I)Nl>d*f5=t(X4YDYEGJ7dd?}E=z*>b;
zP&z|iNVF3e2SEgD?Yl(T4pQgTe1{4h+G>JY#NMD}Z~q=2e`14Hw@DDU@O>YxOM)<<
z%nDpsXMZ?knj6}CcTroX2tCbIm)sp^KThbrkB~y*E0AI#>gfb!={S7-HfyKfWc5=|
z)6fNHi=0jnV&W#gt2Eo3HAM9it2NR^2od6f0IwENP9`M75vo<kImn8fAPgysf~B=p
z4)5F`nWdaxZSfy`_B3C)dc;4wtqIr9ax|N8GVU`dOI|aY<-A}soUk1FEX7UoG+}u2
zHV3!vVw^>JK4Bc;M<IUb<E*AAY&AXV904^Np2O_#Fx<P%;cUX`W|K0R5&O#+onxiM
z=@Q{7bfGbMPBtGS-~BqhM?OZ;SO!<!(uFg`VZg^e`ZPK(*q`*dcW{I$=A1jdLAPBe
z&r)`_w=ty#DcIe+$6If`O{38xnWs#qGp=8|4o>jQ$3DhrG$Pj(=LSF_7DE(^ijB2c
zXUVhQyA+RyBakexy-FOlbtL!3-1)|LxbU+zx+Y_3G9(b1(hpFMN-s|M$WRjcF@gY{
zX=LD2<_bRuNQNVVUJEBBMgAVPx4IqDaUWwOF3b7s=@yrk8vOlN?@-htmfDv%nH)2m
zkGP$d>}4Z5voWhdK;4h&#dW|zsu&kJCkF=<3v9PaAu$N)<9PuAg1T`8(<73>5%>B>
zWKyvn`b2g@J!)W0jx8l+mf?E}T@*O0$;*t|$u?g9E_UTS6}B6qDB|LU^Yl7xZr{F1
zk|&JEV+5Krr%!X{%qfI4+_-U#JTE{<ws-gV&Py+|v3{B`2<f!D`2IeJhX?HJ?a*j6
zNs@%Ya7d9Ce3z(#yFYFT<@aeB^+7{W2?To>0tp7OI94pi!RW>T+uwYVGe7k#E6I?R
z$pl|&v>)JEhcOaQ1jtAcJArTkWmcW_+V}9ZM=}`Vb(^3hd6p7J5v76N@)CRF0lI`*
zQSb|wmw0rw#ee&qTj)-k)!r&cqXS0s3D=T>JIR3MS#{ROL5(PGRJw7AWRRsCSVK{2
zw39d~P{M)o@l`+wG_}DW-D7%kkDdN8CWwhO)Kx((1fz3wA<%h-QZc$HuvSx)3C;a$
z_~U&{dlQGjSxuu}=i!I0P>VvYU3-^#Hm5%vqNL@@<x4#L&=rz&#<gpMN`V8xJJ+uB
z(sy2FX?cx$txmVyV}JjMot?Yf-P<7wBc}5?gMOcRntor;@Pj3F(h^F=%q||b5I9j;
zwj92BgQcw%&ac+7<&?q$v_>z=hEl02>r67-bdHNW>MJb@&qI0!WlMCLLg*u`CQC}f
zdJAnGt)(@R!H7UgtTUWyz@PZ&2I`%=e0gU`v%bVyyUTDsVK|*|ToxRb8KKCiO~zPH
z(l`j&Tv_MzrArKkDZZ@uSHAKo3P(CV#!U`M=X18_DR=XnT={&o9nqL3tgkkyMLyP)
zcwU1t%dl}&d9P}XZ!JwPpgg>XSiOXE0^t<RMu%>@&6}@(2k8Z%JdRIJc;oH2m`o@5
zfzKlkKSsOV!x@c|jwtj%7bMdGp)Yyp;$_aBI)k+iU22BIA%pRR7himl8#iv<*Q;FE
z<-GUnhY{jF5!`90=-~)wSzBMlXO0t7ioBq?+M{q5;VX>MRGFZnOcGpPBCST79IFjw
znNwsbX`V9~XB<x8(WjR1O@OCzI(1Oapq-;m(+7qC0LY|CL_t)vP;a!!gyO|F-sbTO
z9lA?hq#qJS4NQ^Z#}Q>#;Dr%oI%RK=P?v(U&4PFD-9makuU@^!G%Y@)Z{=U_2t-LN
zB$-o{qX4ebsY?ylzxg7y^(EFqjT{|fJc02Gqywi-1*)dfqcWwXm?wmdkQorgb+ROd
zP(k2<E765P1rg4M#EmX4*O(+l1s*^D&<gEX@ZZ0+M;<g-K6{?&Y|Qa!$ZlS;Ulhc1
zNk>eHy(0==(d;g<zPiEM=`|*^InFptZb4VPO;?ntWQ<HkWRnp)<0;#D$+@Ktoy@Qt
zc(m(2M(5xMq|+(B?@{Dgb+GA-dRbr&Zo#D`?#mMQddocV#N+Jk-r{y?D6L^M8gXm;
z4oQ|0MG=$Pka;pA&kKZ*OlC8-Z{K0-l%?D1@Yq9-(rs6~w7q+K)arF6(>ZUv@j5#@
zdzGX=_XW=Hd*4=~(7)GE!V*Zy+)B)YE22ee$I&~t=xr=>b|uD@b4m};2IE}Srk9o2
zaFOP?S%Ql^YD-NDFMSW&vkVb>U=4XPC#W|t#?kDpkPOEUuzj-*f8>cZ)YUzHdv8di
z)?=mBVVq1E&8D1`1;=GZATnyx3D%c14nx+KSJ^s$p3ykPmp(!0<4d2ibfl9L-1L|<
zNw_;t*~<%Z&*SlCL}Qk)w$dPue2ghkQJpf&F;Ru>&RBeFX?Ov}@d2z{SYW%P*=W&e
zw|M)lSCA?IPtou9dHd?SOlMPk-{;EZhiSDsIAc%}qM*X|bT%aLB$qE-;Plq1O17fZ
zj7DQdlL;@s{0g^k-}zr+d-a?5MyP)PAsqP95J<&DSCT;YxpuO!?7Z|AE9=X&>IQA+
zl%7Cnjb03L2vx0zJjrm=8LsZr=rzgbb9_IbEDCI%L*QY{oWfcBxLyr8beGXdhA#xx
z87_4cf9k1q{{GuLeDk1BueHo(XNmrN%xF4gzbv^|q(mYkR%2X$pH>jDwXwy<xpVYS
z`h;FY7>0Plkr$S9xDV4~X7hxDdCE>+KoIa$&8IP$v3aJ8r@%RluM}C5;D-T4Q8}dq
z*3y)U^7sJSTZ<x!VyU-66nMPx+6zc81f|%!caOKPzRN6`6Gssr`N$J^UWjzndWa(*
zU8JP55pm%2_@j?;`O+nDf;`Pgl9ZFdkZ*tM+wAP_eP7S-1GZlmqKMXEkr-Er(@F_`
zVzJk*(QRGE_Z>x+qqM+E2<tIIsjBT@Ea^N!NdZ#gM-l04hNmR1)Qg+10Sl%q5!MmL
z5oH)s7{hFkGdLb&5d8E;hX#(XZ69!CB)%VUVQn25_>7VngUN(Rk|DIlH<t0>nCTlc
z%0jb@pe-duptzeP+?!^k#*iC_QG&LTJRW#-lR1yCD;{4r=wyi3IbC_aDIZ%JO0CJ~
z3BHj8p1=(bapeSvaMm&z4*1S<-{$u9xA6QLe%wa+HC*YD%~KBU?NblxY;0D_<k+fh
zS(G^k$9oicNl}y(r6$dC=JOe&@r0A{h{<HSFggF-Qn=sy#`l!Mp)gKYewPxIgN>CC
z;V0y2T8Sci9-b)x5JB(0a8w75^n8k<B%jT3WkFdclvcxiITjqbHfXCknt81BhqN|(
zSQJ873g;t~q1kQX`2qjzx396(fQ|K4ob(85Az3~}D1j**N_mx%xYZmFQ$Bxrg<f6p
zcCL8g+8&eS$HKzjZ&Xrj0&Of7fmuvEE5IGi7=G&oI**^l3mm!CsJW(;4lf9hz6YMd
zl!jt9M>Zn7ARve%ifmlTT;v5f1EsDiw5sxwZM55r4vwpvJC<xR=Sm~wPdvQD-+${K
zH;yD;SmXTa21W>`^BIHLgk4kOXBt&XO1)2DHCZ-gJf2XuQ1^V|TF5jlIUbKG^C{EJ
z(YF>WG@DUGH%nO(C6Au#V2U|X)bKs8a+578BlR-Ph@GMt#n{0BVKt(fR-(*GZoK<0
zcW+;(R%;Q&O=d~P-R&KY`-kLt&c{FY45v<=K{<mbES+YZZmUL-O>oW;20mv`ouMd7
zMuRci+q>L5I^yQdTMULH-0#@u@2RK#-uL<azpKLb>X}=y!1mN)Z``2MynyFhio8IT
z0;?dX#g(&wSGfzkhwTDC3?5)R<)V#jD`x?-!1^$XD8r1>80Pbw;mH((;EQW*>Nvi&
zbHs@ecz(p$l~rWmF-{VO(+SfwM`Rjb8zzGj=INZW)byO7tpug7*iBRRXE~WM6vm+?
zv^>e9zDGMvcyvwj=&HdaBfR!jMJ@3>T&XFwA)hCBM&f${hR3*a3WN)6AMxsoFS32}
zU6dc=N3Hj;J;^v49})*OHr6$Hk}R+t$};EhWS^p_s<1_A$g-SyGG{!QG8m7U&Sw8Y
zZ2!RyOYnfdr59(tUP$yQAyCr-u3x3|=qaRcDUt$ZDpicAQA2pD8eG$rE?l4@oRs)c
zK%OQ@PvMHP5)9X-x|l51L#^3hk`-uc86M3T^=C*=@=K?eXu1==y?2j+lSE;SOPgCb
zPcod(7>&nFvK*-mp)m}O4j4>^*u_~dlqB^PccwEAl7dWYa$704bUn#M&!dyhc<Qvr
zLrXb2El}MyQc8TOaN3X;1;sqUcOIdFvcHc!KgXEL9jkwI#Pi?$ItROV@PY<G+(xJf
zv_~?TakPIxyV0iI>XD~uB^w3gdBVNJX?4yQ)p|(Nl-X>~a5APp95J2EDoe!gBR)`4
zcV7dox}Ay#FDnEn$@$GLN)_nY44V`<yy}i$mc-2lBJh!6fKuQ!LYxv9XDg9PKd9OZ
zrqaGC%Yta(w<v_5(Q2Ztp)7LdX+gc)Aqq9mo~p6wYreUga4Sg}pUlxpQmfaweC8A)
z@F>$9-xzFR7^gY9uw+)#QjW9`>?H*==kSCe4it4;^675KQ@tANbxp(1FlA1&yG$5_
z_`Z*?JcPAKDM)5BLN7oFix}^sCwDNlQ(y#fD7ke0JlEfTmBH|cAku7X_E_mH<6QN5
zu3mkYtM6WCl$I1$u(N-}|M*XSy%M7|b|EF>{smSc?)?b=p+>9>cHxsEuqY(HbaYz+
zFY*~34DnhWgdY(F5-$i4QY@^{EkPVmxDsJ~l$1C$^SlxjF3L);sSp+~nc$l?C=cf(
zwKybf#W*E+>CIh^4o~>pM_c$|$V3Dj$QH+TMk}3iCXzI~suh&z5<$V!=T`ZRodI9I
ze!%$03Ec;Z`#NaWfkk4SSWM%Abc&5oQ1>+y89K=-TOnbgZjp@|LP@+htcI!L04qQ!
zNv2Ce&qEq;1(le#u?Sl|J1_8QciSXc%F)q)?K>xg^*XC7A%E`C6~1|^&x`#b!$gAT
z(P}kWUpt2ris5+7Y%;+Uj`3u|{&0jTHCsS8P$VPCwJhf-GnhqtIrbc<l+R<4$GMK+
z%8EzLjqzjy-}ezh68Qlbi^Y-6<|yZZ)QIETknUr=E*MLmP8c2^ap(Hmlv*JCHr?JS
zTD3KbtmJUM#~ZJ{#;xnO*gfb|Xve|P3BUFAZ&283y1ghgrLH6|ilU^{`hhb-72mzN
zKg_LuAKNRWxzDUxV7ru@-RK}yiJr}IX@M3VID;)pqGkgE4;lC<Rl}<X3$lM@ll6XV
z7x%H93fmEaMzgWN_M*b}PMa{)d}=dh-7k55FXc{}G3{5_9@pwzJhfS2dsg6UgDp*!
z3>22!YTN?b_tX2>4pE?}*^+14A&+-s*6N0upJU6MMt6z8e-GPnl_g#}pA#s*N^>WB
z=;>WdyooS^I1-#cdxo1=-(WbJ6GVpfjV?=_9?ltbo^#{+b#7k2$tcwnR<L(?!aw<+
zzf6g(gpwa%yZv2?Z@(M1|KO53hr@dyoANu>gA|0m$N7ymLKT?F6r1S>XT8&!s8O$E
z%i;j#3sfTpB{0^K7X@(?Bdx<~hZ2<@oQz@NTtKVcLOVy1r6h%>(P+~N3x4WcgHwU#
zx&53w>74P=6s;unW{b58XK_lQ)11H<bZHnR8Ll)GPU93zGr?YxlL5Yz#Ga(<3O=`7
z<FQV}dczQ_jIuOzyK4kNfaiJD^I!1O)7hL_Ek?Q$(Z7wIUBm=yprKLsxp?**x39fT
zI!uV_md(w}tn|7#rzwk^*I#=LXC=c_Q(D3GJ9pXH+oyEzUk~>U1pd{Y-v>(SYF^b+
z6Sfk?6I6Cl(t6Z8u-cO7!KgCuwjgDdS?V;U1Y{T?A`d_I@FO29s!9;Xl9xGA6e5vG
z-zUju7*imfLg6T#!4Evv&up@^wO0A=Nk@_-crsuqnlL2hH*X$sV_f1!GmeK7tnzW5
zhZlzU^*Xhs6?_#TgyG0$%(9X|3~43_rqKAp@rzfM`I!wLJ4-3@1dF6mZ?V+s5{EI$
zQ(z4^LF@<Q#hkES$1eoh%Sf-i0+00&ra`k7bN2id4yGPQ{V8EMVD4@cMKwx2rxcdl
zI?AFzI{=HZF#Di9nIF&|S4v}?KnQ~;DkI}s7_z+4BrhF_>au?39NF+5VGtmkrkrG$
zQWFFrqEW*dgRu^r$0SRcrE{c#bX0<coR~mJbeXff)}nM0yHM@l-#zAq*LHb$OW?OU
z?DjUe@RNU(Gg}uqmyG!*|J7e1Fk>z>LeP#f&A7B4u`@6E+wUAROn&U;`avU53LzE3
zm5a8ILph0b9$w9IdcBTGrr0#al?E%S?>Ms>VGS~h5P^r+2=SX0pgZSq+Mo>to?o?+
zynsBJLerpxpwt@WD^%pOd3J;L$`UH{u~IXhj``e~2B%`l)w$x;lUY@r>7QW3fLg1~
z*7^p%@1wkcp75~-B64z<ljIgL8q@GoL|&qe;Bv$9XRox`3Vd`v!sr6LkZx<4IE)Fx
zpgPW+B~%_sGNBpQ@Ct`3rewQsV7ESoG%-uvCO`QTKg}y|yvt#~Pot^n`95JB(`dCw
z%Q2(jl+k2HQaUU!O;X;zar*~&Jb(WY`w~!Kiv_mZir{ZO>TOtVO7w7q)4IwtMAekj
zSw<-;Y`1}`usy7>U8#zDU8Ff-7$Ol!KOoKL82jDW?gbvJr`G9itS;n^9BG=Oq)#uL
zG9u<Hw~x6!Epg*H{m~SwJe=pDf)Kw}BknHas}La!$2Mo4m-u2tBT2BO!Nc)07nk^n
zH6J_AD2f#4B=vfeUZX=CM(@LRh`b8hgIXQGl;|QOz4<yk)B)R|5l5UpbBUvw$MIlB
z7z~&@^*(HO6wXo>m6`nmY|sB-VEga(PbsB|{TN_z)>pM#&nblS==3D*$e;%!Y^fJ5
z;p(g}lLX})B8(8BBCPv(p||*Z6;4}<B(L@<61<RXJ|b+!2q#cDG!{SfId$nQD_Ua#
zgQ1=zNa3;7N;oJ(e&g05J2QhBj5+ELvC0R}M@11qy+(6+9p4MFlpL6hX;u<3z)w?*
zwZsa3`O!6=?MaGhLRls_r)W1@^y+PDQN6n8I1APi2LX98qfrm=3PC=fkllJ6E`0%6
z26Wp^Hcww-cO;q3Q=<6^bF1+E0BvSy>Bx<xC=3gyTC{<g`EZL4A1J8{REc$Rk$hC)
z1XyWVTWUd(Qs`;~SITNETLRKb3R58F8le?|S4S5T6$JRcj}ro;OCZ8pi&WJ_i7pB}
z6;+ZLT4S{#@FId*lfaXpEs<BF9@VHt9+w<EdhSWScI}Acv7tX4F)K1CQ-<lBQ94Jh
z_1HhWjVF9oR75x*^RTBX#q+@DOOLE_Q95ROeF!z}Mw5EIPOVm>)odete_<$X@P$Bl
zl^3Lw;G`gu9*2APh%b(DUWC?$Z+zz^b_YWWW0@xzgPpt8?z!(a@&^Icf0rnv!Y`|2
zx3CT?1?#IdbP)3LwFwtiyI7MWbb)b#`Oz`9C@a6cR*Rrr#|ec)VoXty*ri9IOQZ}q
zoO_JY4C6`$gCl}a61A2PMlc(X`PNI@v^+~=WtHNopX1rT@b4po4)fU*)rdL!%x4*V
z<$vN*y@hclkqXJ9I{)c6Z?XU5gzjG{YM#Y8wQxIFOv5eyZ*{prEHvdfe=x-*SctNg
z4qfDk(m<&2+A+daJ&s@@q+6CHp7M}RAeDzHHM%TN_$Z~w7JRy})*x(kD!VLW2>cq2
zTAOzLn9oEpul5Um>&Ae|9Fk<tEInavkI94}^F2i9A%%ys5@igT)r6*C%R*{O9^Pv6
z$1bhVnDrTt2ZWwSz22hTZqsSC@xvG)1x`Dpg3u4JuBrej;bFC9eB&C?@<Uh=AQ0Tx
z+2h9kfO%SQl$@}AGD1lO(qgT~Xos~AJllca*Yo;c7!|gQMe?yA<O_jSmX%%;ij2}|
z6d+_}Qzii6B&8`~ZV;s;@ZZCB&-ZZxjMjuvgtcyg?G{s1(j5qa)m0lVP$5CwAoL^{
zOX$Va!kAj<a~^o;%;S9f<`E|o%V0ESUSyzCMrp!0o1<1VhsQ%ae3rbBU_RxtClS7c
zK=8RME1Xx3`ThVG8d~)R^;(TMj%hYq2+ylJQKq_N2v2=CwhNEr{e9wd(+Al8{Hw2V
zZ#bebmL$y??(P1_VEgw9At8($z9=cJ_uhKo{#kF>TxnyHIi)ex1>pW!zfj%C7kafq
zE5e|TDI_ZJ34#ENLu*YCRCpkj!fK5vG!h>Wn6j)UW4(y5-bDF6rYMR08nrNHI6UTx
z=kw_K4Zd;hh{42gG8~fX0$0rGr!!`G4VDUa4~EoKNEeS<I%UiAu)ZXyd;IJptE`tL
z<9h={Z0R)H)M_<qwHl2^_3yfcqlb`^KuR1AE5RvAjN)W(m)hwmPIk!hlCM7h0>|??
z+EjXzC$rn%)j_Y;gWw0b^FL5hr<KGrcCp{9HztL^bB0isc%i`Ww2(@mwZ;#EiZ?S&
z5Yj>DVXy?4ia-wpo+55G7|xCef`IvajwfnJA*!}%yM{vHNe`zr<#dX#1iA#JEeMA%
z70p_kS&<RhIlu5om+k!lTsaM1KurYPxwXw;{Q{r)vwwl#_{LYrv0SV-8T|S`WO_V8
zv^qR-x=S!QLR(l}UnYnIjYf;vbcW{z7-R57fRYMr3Pey15yVkMndBsCLNjV0iUd3F
zBic*M=P7T#`3~ARpjs3UcDnn)KHLu@S{EK-es2#unSj;A&e7?uaqT1^w9r{!!xKq0
zcApo-VTfusC^Ai{Dvpjel?gcrJfSFb<xUbeLf*UwNJWz7pe-YHLPHC}AmaL+dmK$P
z4_&U~fANp=`M>m^5XLcjK0^nBv=lf$MmSfMtc7MfDfmBq`wq7zKW@4HtA>*rgTlP8
z^CM(ct!;!3XEI!<Knk?0ByNNSWgLYqaTsJNQCgAD4Srn5^E|vTU_PJFYPBee0wu+w
z+LI$|0SNG<q?~6c5g?R@P?A{rgs~#<6v9?~lTV*t;qr2m+lM1k(}mLDvXYzoeTLmN
z9{-u2;_I(GPh78aF>Eq@>3O`&=fpXlIKRb}Sd;D^VB?5hcbRUxRT;ns0YOl`7o`-o
zEbxWIDfNKlo$G?SY~scvNQY2wlO{Ree(`1Id9j!nbPEZB#gyV7@U5<;Kw0-5whM=F
z@5A;0{J2pG#~Y33`_-_-0^4PvFv5W|lxd0sd`}TK8VqO0bbX(BlHiFLDJ5l=<F#U>
zQ}|M0wLvfVcUrpzwu2`X&A7>|%!%xrpSsdz_iz9gx4;XC@wvOb!*KN+&-}?h!&jdB
zI=Qo)uQwU~=C3h386%o49y`^+pPrztV`*)PAQUv}O=j~s$_p^s;0FOxN{lTKB0xw<
z6ou$Cud0S&4N<1pWB{!mNs{sQJ6C^XvHg2R+-TA|a~^YV4`Zqkoclu>_s@D?6a=*v
zepIihHpW!zp;(;t9^fmiu;4UBlHgOr_kE&zjrm}L@B5@_TJh~2lzC2653BXCxa^tP
z9I2ovbEK|B>;vV|taq3d8C{w3OOJHeIU2#k=TLq~Xe3u}-iF6M$z#9phxm=J{d26R
zcqp!M@Q?lhlW{`4w#-M?>&VeDPD<7`R|z9Q95<Lwr$kYWA}?svYL#zvSs)gAs9_LN
z&NI?9C2mI8=?JnhHf}N+PI%|t>zIYt?0bq7)$iY14?oy-@qv;$e`SN>-aZIjd6x;0
z#-c>Y%GL_)%?m_LADiav+`h@_^&Ul%A*{g~OQYMM^s4r%)D~wfR!C}{E@hrkmKnOp
zu(ko7!WJdM38WXIV}r;wQWs1Qc8FUoT-2y!GYXB(HJBVJ1;)TqBSt!lU-yxn7Jlbp
zW-D7<dh{Xsh2g7T{|aWUPW{X}XU{ItF%Fk!v`%kOi+!w4@s&aw%`7eHt<{+(bC#M7
zoHG=8N*G5-QI+q<(?jyvoKDZtQj(+H+tlmlIXUXHcW|`W?DyxFesnhb(|_X6GQ54A
z?blx8=*~?TCbYNCv)1d<SZWe?JCsNLD!FYp8IJ}uy@;q?eK*2^&I^PmiK3V^FFEP&
z(hMb&(lIVFR=t3evE`$eSJ~KFV>lc!7|nU*cFOV+1kZelpZE{|G9vWYy?qT#L8I9t
zN#-2y?{O{;u~=@6C;T72b%X1p?{D<`$ML<x)|%*9Rvqk%B`dANcZPb+BkpynwHqh_
zv&ooxyMfIM(!&9sCkcCP3TqI7w;;urDBq*m>ypn?z>;SvL9>oj3Y`^%by3~Ol}8+T
zWYY=Krl>e3*N|I{)fQm{D)cGl13G@d#ibBWg_J=^z1`!n!GOW)c|P^${sOBPF7f)i
zZ*lR=CbP*ldOYTtb`QUt<9m+Ig*80S5PAwJL;8au>l<xmNlL5N1)wZ)!Z1PzLA_Qd
z9gaz7bE0~i2unUVK!ux(h7<1G*?C|dUcK(u<Q6TQKfn<zucHqRA7Hx>Ae}{`S=v~p
zwQ-KH;bXI$d)v3!I^TVO?ao;0od%^R-otilu~rhd+Y4;ZF-3;8^@?v-=nC8Y038`*
zUV<r^9qtmfn%J;jVS8yXg$A2fK9~l&^#!)qJY>6x-?_|eX_E_AE-}ar-+b=t*wq^K
zQ>&al-KA|EF3)LhtyKyJI>TQir}MO=yHaD8B=nkfoU@heN_793HyO{4$mes~-6fh*
za(r)_dhINO;{p4JCqJ^-{=LF$w0Ptf|2S9whhL#QI<DI6LM<@B;;E9=Q=7CmPvJK_
zit&`A{av=!d!*wT(nVP3s4q1sRW&dtw8d$Q^%RYzC5$f7WnP`3u34S+d5Ne?yda_s
zG`<SZdBXf)K-}u!!WyM9i?hB&*c@C{>Fw1*0_E`PKCao|v89io&z<6tC!Sz$IOe5S
zU&5_7X`Nr^VjyUj23wYNFPtXuMYXm7V=c3!Wc_r^cr>OV1y1Ygeb#D7DXG^Rq{A`U
zd`_#=qh1)MhucJ_*EzU%$Z>!8z9bHQ80+Et^PO14<uCmdL?OPYrg-FHw7%Wz^7vyP
zrRFaot>SR^fXT@SlUEu~(MlI0A0ZuaEhLB|yjp|`1FVt|1~huzs&%UrcHx6)&BDxB
zDg1i9vM;JPsnuHykB^xSkC77MW{q0AiE~twkV4^HMacJDNt~y2d_isR24DSa|96Jl
zZ?oKOGTYswI6S77&*;U1m6aA7n;SUe@KlKJhgjoN<`GCmX{wCfT0>bD4+zY)xK@ej
zT3aPAfnt8Lk8=fsVdVx<#r;+m@B959><AxvlvAJo5|96>e}jjA`H!&t%tu*!=n5<A
zYiyo5gDZ2Kkce7<Y}AON8uQs4>nT(a;EW}xMKpRHr1a1RR=Z7F^%|GXt@EkVb(V$0
zm;m7i<hs)Qd;9t>lVl2KFY#0VyZ;m~^7z`X|0=IP{~YhU^%hB*G9L9A-?_nhE8@;H
z<v;$dYg`*7KkTpUM?Th{c$$*xvQ~k5r4;W+l1D%KQI^&oCT=a0%`H1OcJZX3w3^yV
zx0;9yJ(?>W{B8%o*+6;<gFw~lbk>$AiX15%T9-IaRuga*lo#SPo8->X=(NeF6UGPk
z(6%IqV`}v}f$x*gXJ8~kOMD?H(+NEV4N(xbn#6j-?rYzsS&#4q^bZawEu39#upH(r
zw_~=>o~6}pQHvV*VGSoFS>fZwF=>%jEk>&;iwx_k3B7u~iP4%e&+&al)A#6aU&q>#
zlj9@$qcQL8@%QiTKiCN!z=dbOfQy0!wz~(|-s*OE^x?;ey&lp^j`t3k^vCaE`%(uN
zdLSK99O6eIUR+_jlM;e}dbd;YC6&S&TVcDc+&GX5zgD9J{91#!-ei2zXEy31l^|}!
z)LIRk6Btc}?beXwbClD>c}B|<)b?-ljlcb0G1|S#Qm4u6-aU%rKDA;_Hxeu_H(6g_
z!&--@0z5y&+V948gv2edy>fIA#Bp2+s4uWR3=~QK5bFwt<36+GKHmK3Vf*(A=XtDu
z@>wo?{!2Jt;mb<WsNz`));8C9<nl)dRR^JA@AfXqc!tgkyjn!C)WQX-I+tq^{^G2c
zp^sG(5k)k*9rCiMlohN6DXLD6bD$JKqfTKAajQcR))*WeG8+%@d`YcUr`Bp>i~}Q)
z4y@D6vl*T(sFyii54id&U;Uf^IoZh$%iSh}o3}9iA&oMjSNB+3?Xb4CiZd1^1H!0A
zkqM+|lb7X#VbJ?$eYJ-Laa^M)ifXvUlf=GccCwGP1;-}`q*>1Q#(>|y_3(j`y0uSz
zp0$sC0_poGSBYUt9G5R`;_HlJHX$GM*}ZX{W-VrPaDp<9APR}1i2nXQ2iI;eKRm(|
z3(c^_1;=?lxpwqVMo3YaLOT{}=~7l>(oslO=0xo-%j>5wY0B*6kTOs4qJVC11<wl*
z!cpW2d66(rrWE4|-h9G_ONqB%Xa2?uoE+?N_sv&m+l-BFla-YfYPATXD^jm>j-#Ui
zZ@zwuZm-37GOWzO1q+q;Mcb(ohGRZYnT#h)hhuyVHQ|GEBxU8@`lGqUK?feD7ST9&
zo|Pw_<l!%Vmd&kAq9{h>b5s-)M3s4SNoC|L94Hll0~H2D?IziH#`e267~I{)vxZiu
z!}R!&Gyaq_Jw-Eg)LTsoTcYZ9R?ly8cB{c-zxYeE*LuA2&EG;;1rBcByh)j7jE|0K
z%?G@7G~_S+vsbwK<Am-*iHkq|^VH6q!IKh&s8rAeoH=)zwN?)^E3nfUcduW=my*fJ
zkVdmf)M*m78q8-C=KVf4Nvk~)_?4}llE^qFEj4+r5y<MGx3-${lvNc@`W}UY?)qth
zAZBuWKsFsAlp|_4X|}pJD^N;OlqqSMkmm_@I-_SZ&P)!u{tx~J+u!(gq$?TUzDYDZ
zVKq>k*;u2~YY>GBT~>OEMqBRQIl+?+k~Dqb#&h4bK~zI|_?}0WWfWyep-X%vQDq6P
zWRgxQ9jhPFSpLLkSb6ekgzur8sD>JFTs*&yuQSSIN--L6@AfU4am@IrkFpLw42YtL
z;o$*CH*b*~pL}m@H;#UPf)o-Z7k;+y!}cHw$#g;3>aeuBiAgi&Cr9W!!;1nsoh6j=
z5yDZH2}O~Tq%(@?6hE1==2D`aw@BW8nf}o|cHen}w#`^?H&|L)CXPdlE(n7P+fR;%
zyz|yJoo<84bcE;m0F>{?_Qi*oCn>Y(l-cM$wtEN%Nm-zc{*lG@?-fD_M61iGpZGa?
zkA4KH)B*zpN;)3Cd<wD|#e7P7a>T**9hy<h@ZK?=69jQY7z7;c>~MJfCdqLhTP(QE
zi?h=C0rRY6G@L!a0Owrgk|C>3o)-qBx}e@$rMI#{Ih!&+IYt*bel4QgT}H}<{8c_D
zFA|buMme4krc<`;9Chsl=I^}9;r<SLZ@$Ko(rk9yEG;b&#Xe=3FSwqD{k>z}erubh
z<u>E-kYKS+?mM~&A-TUIX0tid(S-SEOzaeqtMprEMOMjb{gC4WwO`%xGFN`?7dg3k
zll<;JNJXFoOLdJdkC1`SbbklWX&SW{m*q_M4sd}-F`MIbfpCslRuUHmzg|ZwiS!gg
zdBlx6hj$OivwSgPudvQ8_&k!J7L$%s$}%U28?3CKVsvsqcGM?W>JddXypAMICP?d0
zrn+@Hr<u<S3X>sbW1=jevDqXG1dm*Li1p<*4v(c~6C<io^#0)yM+YN%oprh^aCkDM
z(c8oti!OCF%?Xt-uQiqA!F*17GQ=zEG}gOlEz!sy=}j+%!$?P{Vw7K}v$TRr64K!z
zjm<TrtST9y^6=t_Y&u7UHBgluN?|pldxz}bxI?HEr%!EQ1Ih78pJbL%k9}H=8X^#Q
zwGhvXn9pXc_n@_WiKD$+{QBSds|@EkkAC!1{G)&V&nT_Il$PB$-)8%`&)<6W1}8rj
z7XHv8IDdtQfBu(v`M>#Fh)LD<@+F+xsF6?ikU>B>I%0Bi%=ruF$p#b13&M7jd^{zY
zj4)c$==O-0mr+r)*v|r^e7rE`aOaTb^6~>I4Qp*hoW~K?YGkt!GEj6@w+M#EOos!6
zHU#Y+%~qE{c;rQbEep_^tVmGCVp9p^w6hWEL5ST7DaR+A-srNlvV<2lXiEpeQE${a
zIXvLt-Z9gO&nG^+#b9zwyV=7*B`p3v4^HbGSy3=JI-wCaX?0q(dL6Ke!hmrfhGN_g
zCtPodOF#KD3~t}1*gXId;449|uF?7g8F)+&_VAphUW;&f&h*~l0^1XuDG}B(%QW$_
z!LQYksuG<SvcmSG-6Qh6Kz$Ev_u~jtrs%T3kLxV2Z!qp3k(~?(x?RF3#%l}GbcQev
zX)RJz11R&Pq%bLRJ|@b2>gx@1A1<H2%*s*|heuB~Dr_%u2FE8HAC2g=*XS$@j`}l}
zyX(~eht>})xGHR~#HF)&LN*wo$~yJ64%#gA8$UFe_dgi1n1M!Yo#LUN|3zNd-hrbb
zNS|0*+EIqq$KWf5_wG^)6k*_D@|4ltJ*+1wr*oVwtFu1W#LFxAwHiW-#aZvuXg9fY
zug{9s)ujCb13V=WRQiY6bcogkQKQX@rx=|ak{wrPeXZ8Omx?5rR1`4h5W-?@$z+yO
z=mc*zB6clm%K=`^=h25AWvN*MeCo=gogtggI5{}taDT$q<{6?!%Cw&o1S?o;&{|g*
zAb3w$T|hFKk{u7Jwd>S69@@z7u^xUnXz+n5by7ia`V<#`_7@P15GftaM!?!;i>Mc4
zM8^2!gtfI5q%V*@U`zCL0;WXC01rVqpOTM9m^8(fC1qLQIf?KU2a}TiG(|{-rP@?D
zQSq|^A1|s^f}pO%g+A@|En+XE91d`4iWdY#jTT{8tE#!Kvim6W8R;aYUT@Oww0Wr3
zqP}sOH)Ml*DOjCT>H;GqS{I~~3GG_UqmOPfnof{1K=~d;o>vw<!r{b%Iio9$8Ko53
z7Nlv0pUlwd1W#3j?+2vGA2HY$4xAG>C9z83;S)K9tTiF<ky?{y8Qx+-*^eT^ZkwVk
zvDz@Wz03Z)w^-_RS>0O2_>$3hj58%mYhBt)9io=t>?0SdmYW0ZAZ|8k*CV{0>wNit
z_}kpQv%_>U<@&YjeD>KdG1x!gH-GCZ{MRpC|8YY1VMpY#^wh^#{luqnsuJUle2;eB
z60{v6oHIJuWwqDF4;9LDm~_r~Z<lHR7_=pHmTWX28;&qZhR$<zS>T}v;(%EW<7rau
z^F>QdAn>GzibIqifYDeq@lu!e@)~lQp+^IxHH6IuVXZ+ReS~vFN+G2~=PA>1iqeKw
zAm}ACG!D6OnxN4|_#R;pAZsB-o-jE&#-=4tK7NM6CTQ#M!w6GU>$jrFRjo+381@On
z08{3qvk8$dDDF>kzTd9zM{J)<&k<~Ga{ecN8c`3C($cK?tgJT)yD^5GX@9`#>N3(7
z6&c=X^lS>YL`fe{2y`+dACECvhAlOvF7ccMPjWQX45ukV$oH_FipC>+9~H((s&<+S
zd|IoUL|#BS8sgFn6$FI!CSe%AhwY|7=X0`YPOaXg)oydS)}Xew#oMyRK?d$Vwj&o_
z-4ohz%)<|@GoDV7(ntAzh3yZX`NFut7_4z9RZaP3S%#O)(diVW0)$jQvd{c?3yVei
zg8Icvocqb2#zhJx9ldtM%6gNa7ol~+Y&c|XeHoO)3k1#-)miUMb=EsaF&#fx52h?B
zbwS`NwTZj^oN1aZzQd|GB3xxtD8c|QsD|Jy8W%@&)=m*1DFy?`bG#@bs<#M&7-t=x
z16^j6*^G3O(Q5bTHJdzCZ=*I(^QwtB${bFo6lIB%9%Ye{O{T1Nn_RxU%3w6X^CP6E
zDAKIr{V$%YaA3>I7(qxuX$#Uc#hXsh`JBLuK+5;6haXOSu%r&KQn2~yPqF%`j}y6)
zb7$MMR%;lo$@)W-Qbes*#i8*6ob*vaOs&x&uD9`PZ4iQVctSC%9*!+aTv_5&z*q-N
z@`|8+-}KflN}5tiyeI<C2cfW~!&ynNx<nMjBqvqN$MZvK-5#iFBwHw-C<^JVuh2Pt
zo*R27WRpIfPMgDciSHyZ98WRMAv1x%r?if(3!7+ZnWrgH9HUG9z;jP21<oO4HR<L_
z#dI`gIv(RnLzoW9c3&d&97?IhwD6B$I1o-%yb34q9e9m8L8DDN8e)<JohDf4nC>4l
zA5Y2VDbm5o_8x~f?yzxslX|;GDPTMq<10mXsfQni7-w0(aE_>6r_h=-%V;+m#8FI|
z&-luB-sI-?5z|S+OV59eCqMcmk3IP$zww*D!9V=YYfOtD%a(lDA%(!jAy<C>=cr#g
zOU)K+uE(sM@1a7UY?_c~C9S1yB`hozMq8}5gh7Pw*AObg76s|3PdS@FSz?M3VJud8
z+?yH(N!9)oVxeQVIOK(t_;C%?LgGL>f?AW>+6pGkDf%Onwggd3)b3zhRaGm?5^D=o
z;In-8664fys@dbzN1o#4pvMcN5|hr!iX4{;gzzZJf>USLXfHJwO(sNP3~phnZmS#m
zeeSHY7OZ17nUc)s=-lB=$C$w_VlS>J2S1`A5f~|0f9B(?eEcawr#W-FMQbI-YE3>I
zA(bR-HZg92?ZQX-F>$?3Tx;RSEda9dfMPU+qQDj<R+pdxrrKv(RATb)Vf!MnlL9Zi
zkL_}S?Sf#rM-;@QgCQ=-@cn?e-9@NsI$J7_APnfNF45jR%dLBTvdNHkr_FKP<JCkk
zn#{4zp>hFuXy@2Cw~m%>Vc`~|%kqI?x>EAN6tHuZ5bta}VLqASN=ulH$nU*IsD8w;
z{d>gyR75y%fzO#QeSzMSj}h6N3zwIvbwf;1koEfnafDxQpdBbb!U+#Qs#9yWiR*2=
zS{r9A>F}6xG(;2yHZQA0Sq2OXpQPly_23+K(E*cE;>9t}^8tw~EKVw-wG{&CGe7Ag
z(i}gCsC9b_7Zi}5PZUQiZ?02YKg0FA``Bbcv(sfa>hbc#FqtQnj&1=s4+EUKbc(so
z$#p>-MifP{aP(N%;K76Etb|}P95J6vp|JR~1ImNf3H_?#_`_ZgAFTcA97aOe?s4Xe
zpJ)H-9m1t9ZyhKaIwQ_BL9>B~BFZGC)9qGCm=G0D!ITuG#hXPW{R7fLAB03TTa>=Q
zlY&MZP!>7Rc%OsCd%!CEN*7L;g@%?kI4sdp4-wBurc-KuNE|olEw2(NMI3}UBam^x
z-eAJv(SY8@GVNxA?%B({_2w%~_mB7^K}fp|v)P0oZqn#>xp#bw@G8+=WAwrm!6C){
zj9g%?#aF7bdzmHZRzbrW`nPsCsyp%`uipO;*t72k60QPajH#xWJx4u?5voKepJJ3O
zsu&ixkOlBPkI8YLgPV6)KeL4wRvQ$ck!J*9K$hebvxLnH=Ln((N=nA#F|CC@&~%>g
zD=%#G=0s2k$Ez>C#D&Y3`HTPVf589r5C0+m`QQ9+n14_a&mYJ47tUb>)Xtpc+|T?Z
z{oVhbm8}(~CcsWg(!qeR7UI<#<V8-i(ZcKai|U3zSVviuh>1rw8IqnHS8F7S(7r?|
zA1|npml`Rn7M+yxfn$mg0vQB2Q2{-+(y9}lPj_{V(V$N{8WAroQ)_l;%%?=PI-%!-
zbr6Njtj~?zTbyqMv{pALWs7&RDQY<2W2aULl%Xs%R`@J!uF;>3G0x$8UM07p7Ca(V
z@nxi_+C!mNRqK=Kj7F_SXeJ!K^E_$uQZ-PqNCbY=!r4lHD{OT+_3URjxN(=D*WsN*
zMcpLCnZ|F_5n+f<GurJAUKlR0-GMDBON**Sr2QkZ;h<`_HJWHoproJ{`IHO3os?qn
z%)bY=-`7Xf7KbJ5br9*CWI7{WV0)*xg8u;9K}G@lqbbKHL%M4{T8%pGGZ%RG?KhYm
z9`o}-K)WTF&u92ilX|Da{>jM$Y`4aMtgP^aSjdw=HCQ1PWoby~3A)v!ZX*VF_Bg3o
z^7{jGKZ@A?yF_)~T8t-&daIoM$uDwr`yR5r%Imw(LgS|;^>!PA5R;|UJ6)P9!F!h=
zQ&JWNFAhoiha~+IoRr9Vozkyt#A=~mo%M^!Qt>-5fQ2TkSjd@B8l14yR+o{p3F&l7
z?1wZOZOWyLD2xb|hjkJ!_SxI*b21pSl1y1{wQ!r~dG&=|isKW0`dmPx>6p%@)Z0r0
z^*TF;2dD+N(rANF3%}jP`>H;h#aA9hmXppCTCFC54miGbm&3$=z`^@ru7?j)sS^v6
zdMSPCOY5Bd;-@+F)H68M;@;t$(p#Z^_EB;t5yBv&V9}NUi^IsOMG@8NbXHH{1rh1!
z1T!5`%;)UgzKc&nnQNq#Naca&0adwP3vi<1V>nMBA|DY{dRfLfR1j4YYGpy-DO&9%
z1l7+QUs5{D{?08PS_v>JKoo+HKKl%R?mzmES^CVEc=dR|$#}|mR&eGMU!-qgZY@FB
zzz=FDKdj7(g$F9YS#U^7=MbVq7{z>)W3<DUA)VES+1VeH7ga@~qFOHg6Ym*wSFigw
zfd|T9i^p^S^}`A097@W{GRj%78Z#M?jr*0;fbZieiHZVhofa|*DbkFccW<z^v4$Ti
zqDF(T-9{*nX17aGIMPYV*15BU%{Ev|sY}8zBC6MrLh@^`ZF3Wajy#I8hBE79<0bwt
z|JnbQ|Lp(vR~S!cKeX@UU-uAH`%HNey(^Ef@`<Nuoqv>f4g;QhXUNdC@K#Py2#K+l
zu+hM8Ht?HGWE?~6q3SWsr8U~!RZLzm@1J1Olp;+@r&DUat~_AvqTQkt@7WGIoDxW1
zfhTajL<%a+OJC7g>EU?_ljTG~NOx%!ClsC^B7zt%2-&@Ji@GhqC{zhw_{E>&+$Wyl
z#niFgpOCqPe*c8k(-$#9GM-KF0w1{$pp{6RlL+Bss64JM6~|OL#e6(RYfC_lT5yWp
z!!c==FVro5=RW?I-Q30Z`R9IcAsh(1;J@`&IsNQ2Y(DWcs3r$T38lA8?exRsP67rQ
zemAyTvB37YMtfxg6@;Xt0cJX(NapNs@8XkE-j_1@Utqg<58Ks!Y?qb2lygWwqE>HU
zb%F0mTCHA%?M`643fm8M?{Im^$9e&x6g=^%kMXDfga453CqKs<CnNfkDU*4@si!~7
zzz9-n34%I)5F@<+p%hkl2<3wXheA7tkQ%Hao#a>zz6xotTw?cVLh&Pv?Y~<H04tDA
zAyhzf{S@ba>RC1(dkRgRz5N-kev0_?Llm|W=N5qvN>)YzxChpualJ)n^%O#Rq=RG3
zbV5F#v3u(-k(X2EW+6oGBRn4`-aG5Xf_v$Fi3oj!uRv7m!Ha4{QH(7M0^g_AUcyyv
z1FR=W%aW7*TRgO0qr^iQ#b<u@C-}GiqyIhfv5)cQ@qm-zn9O=?J^eY3ijo3B5Z3X-
z8q)VcDxCBY(nGk)W2LkfAxfN9B%{ia#Fr7R)rZ;MJwY3bSgZ$rNbBJPSLy^3R5kj1
z>dIx7abOebOHDR5y42b+E*m37gmDgxe(-@D5}dAXMM}|JUSe{5L@^pe&2g|lz?VL{
ztjLMV_mRF&HC*CQ+QSG#A;39_!hz@El*QN*XDoggp^E~vB@QB7Sdte7*7F#TPH-;c
z?D;LyD-V$jGD6|F^4Qay**L}D_}~6TCa>S*p-+8@8&^JuD-7+Z;s<$phRSA;&avqT
zk|A11gm5^cNixf92BD`=frk-}Y<R%_o!ba!2_(dxhZp!X+AVrZOVn#MvNWN{Gm<1>
znx+iL6SThX^ICoGA9hHkuw{w0C92HPMTX20jI)?@gu&v4A*L*mh2hTC8?0@v6E|a2
z=;PPwSOlIIFdEI64yLT1*&ymIQ6vdbJ!YO}wAyW)fERA=^VQukdu7IGT4$wxmb(80
zue|j<?;gL1`2()`^&=mya)CifR1kCV(;s75D*VG;Ml;3w${L-GB}RvLkvhdmz*(HV
zFTSd9wnRFJ+dc3k)4>sP0H)sJ#?4)nbL6=O3(EHqzK@fY*DTh1ASjJ+cpg{`)|OZ_
zXhT$MlIs$QqaHU&Eog1fo?<$kFrS_9#FZ`PXEt%U$MRB}Kl|_gZ`k?lr}?%2^Iv8*
z9#UVwNarX1D5L&8!nFgUQWNSDTc(sILzF2}8@$q@fYKVY)1;Y1alcjUKq_|b-oqM$
z65t7mt+G{=@{m$gs`Ql*0@_xuKS;~C!gf*d0zHq-ix=qO5H_XOYp}l7A#O#GPY@!i
z_MqXrvE9B8+n0Jw`^S`%3B;D8!y&%(F{P=TB|PtUV7tNyOHth@kbroA?HXrt{2)Y^
zB~n|WAjFm>MY+KC@c>80sdF2o7cY?wa{?R}A9;eU^-ccUzx<b&zIBJoAAgqHmp+Xv
z46V=+U{E@#usut#*%+6MD1{^v7H2eRW|_|ffu~Tu!r;h9M;vZ%Kfw0LQ+U2lz15`K
z>rsni@+_q&a*`xvo~4W?(?4+7{=0>94!Jn%@dM63`UqX4vFQk{JFKmDh#Mh#;t_}#
zZS8|(LM{??XEoLugzwW@USfQFL^&E`V#Cqlh(P(2MOpD910UggI8|{ImG;m=SA&L5
z;0bWPhxH7`YJ{=)QB0;wlySsij4O2|N8<Y&og5MbhEorn;<$SbX#+w5mp<_<XV*9R
zn}6xgGuYeb(VzR{yxDt%;LIm!hmOE$R5?dwGh8~ur9<3&L?L_{!eWeJo>`_dNvjTC
z;A5O2Ioadz?hX=%Ck3&m@WOy*r_IvRGGQ1zIP2$0!Zb;$^<aPxc|CmKq)tf*&H$2n
z+@w+FG|ykb+u9<KIoWUrS7y}WsJeYy`=Es<gg{A&0h}(eR#R_xne~s*$)wUR%1fH9
z8mLO}P$&gfQtAYqDazC%u_3t@D02MBQ1c+XKUkr>idU0qRL#TlJklbg$W!|LW16ix
zwRV&4V~_FHi*GX=oFIgtw!F#Nk3PfIyVuznjCkkeJ6wC?n6Ta^ZU+49A9{wuOS!SL
z!&3b+OB)`iHayPs--i4uX{s2_6`i(6R0}Dv%y-_xn`eCd$&d2MPk)NfeCE?!x_p_{
z^>v!97D3=s<QaLEk>?qs@rajSd6}<%^|$%<bI)`4?mhCNz}X+(R9yx3EP%Bj_9QmV
zAU(pAB|1+~vbvXeQnIssi$=3Sd$|K~h?EL1jxdF0Hk&garEFX{Lu;i+Ivf%BKGs34
zRzr9m`=c4Zc5R07H`zJfX4&83)auhr$^q9-Uj1=G_xljm+KS^M99Bv?y%l=a5Gv@_
zLh3b1p(ofPrCzV&s8r=2&?Q`TQ@75db%`H^G+JHK@rcq2iiMV>uOv>2N;JjuKzKNt
zR(50BkfX3h;<*wjEuK&)D=|)j=V6KxFVh53i1vKu=^W!6$Hzy+HJ|2ko5mxLaCqmK
z{^0?aKN@rPBcC8VeVI4D{be4%^l|>)>jRGN6fA9Aq1TCd;-j1Rwj>|UXa|B&%$T_;
z4DKQe!+Zv#nV=pk!dgfv9J8Hw7~S0=@;z3Tm-)!U4|D0t70zF}$mug@Xti3*XA>sl
z5yRn-gX1Hfd*KD%dg~pAqX}BuAL<^5`x0Ri;S4xQEpE`zIgPWI@HRH^Wllc6S7Cb;
zA(e+)c%<H!d{Ihb%J;Cn*6cDL9Ana1HQiX2G+H&p!XB)`c0p+}bSBZ6A~hkI5qK1M
zks($PNr6G(l)~jDF4w5I!uG7pDT<8Y@R(+^PTXwLefSaHdF3jjQ6DKJwWSSCKmIh=
zcW<&c9CPi}yWDuEPf%+SH++8L=RQWEGH&nf(W_seyY8{M=5cCp6<6FKO(o+*(QbMK
zwV=ZGy{mXh%2OYCoR59t<2>`sGn~J8ftA%&8qH>f?L|(JW#oCzcsk~_*I(nC-}ojk
zzW5S5dk2-?*$?@-f3FZi;EY8|wK(e)jarLlp3uDfDBjsK_%dg9bQ56;;${Q4kR}p~
zpTB?BTW4@O$680D-DTE4MlYmIuqCxxSScl_s#=c`lx0GhDay<vF%gA!c(PcW^@5Ov
zs)XlZvjU06i$av=k?M>h%^8jQv|3H-oi^U1k8<bgHj~i_Qh9VvU1a6TV+>yWDt8V~
zc>d~5c5jb~n_XJHh@bhzk27>*c6JU}X+1>mw9mOM#m3#2Aa^8*$0(7+4WF<ckl2#>
zo!5y>!DpX&hEIR`Gkoe(pRCry+8WJP6Hh7f_nr0qVV@UXc!95e^|yK9h3~Mpe?(an
z*biYbeDI`BDKHjkz$+bInlYUo(u+MxkwK9V2!SUR!uKj7thQ)lan>xRi7e6>be1E0
zAHNo%lR1sp<77Ui(+rS8fiEaciJ4Cs&sMp8w8GsirfjZrI7(=Q;H5))YQ$q}Q`Xug
zv9d(X8fKc}>|$X`!DMtoo+q?gtJLBa=N|b8?cQm+8|}&#CTQ~HXFtoefBOGrS(cFW
zncFF0r;80BXhf(&aO=(i-#N~ZC^ohl{F{$hY~dKr6w_SMmH~+uaIAf7Z=FB=xzEzR
z^f<L<lUlPzqun8{H;CdIp63ySHJa_J47-Klv5!2#pZU}O7Dq?NeEsWR=U4vD-{I@u
z_y*H?@}WHN4;uG*gT`t^SrC*dBDsszL!1{@T6D@o`98yY`?x%3>C`6H_mEN{<LUqy
zjQZqBPH%0ArBfSBPfjptM%d|)8BG{RXbZpj&XC&qCwb+!{t=;CX07)qsp#|E&0nRo
zKjzT=euSeU2V*qS3Vf?j=^SAV%PULZ=H$bIKtdQ)RW_?N*4pplNI6$YPvl903L<3S
zq4k_F@|aB$mYbc5Sga&YIBYg2&wM769tXpi<D$kWw*(@m73FNSbJm+PB2Qz5APggH
zRzOgBNIEH)Cqt&QA>D3=dcDp1lTVS^cW5tF-oPTL@v+Z+fv>#!HKqr58EeI@+jI6#
zQldEE)TQ%endUeC*;^EIOB8xsd}x_Jyxt`*VU{Y!sYhH7@ZylEv5*e=#h>^B|L%YC
zB_4nJDb`PI5XCiuAgJ0b&fyl95rM#3O_nC??(XpQuYZ$&_^bbjZ-4vSOeWKX*X9rB
znJcBL!`wPN?NDjXY<5I9_Rzc!+m%H63TwrC*zW9m*lulQ{pS1lwGf>q)FO{TGNWDh
zD~AG4Q0fws=1gYG+&x)hH;XBos~nF~>LGa9h>jfd$m)#c=D5Q4MvR$du<!^+!gSmx
zFH)M#6>3p~GY>sNv%5uSy@{~}1Pvbh<R`iDoBtzAQsc4#iJcL)J6K;3)FV_WxO4ZA
zSNb^;$=XJpKXs+T_Hm+^<%(82AoT+J+QW5M_~lQ3lGgdhh?{lddW}ZAO<b$6T`3Pg
ztkGyzE)yGA9)9>S{#U>Jr#Lz8^X+eci@*1G|1Qrx_Z-uC`Ug^;S}BRs2IasjEn=Q9
zPy2Ktg_b!==lE2~8@^XnX^pFNS?{0q)*!W|%rjKr<JAImo)JeLMVip8`6yZGSQcfD
znHLPF8{9hVva9Rp<{C%C8LbH9yw6%N=CSn|OU;tVbHvSBwI1AJJqQ?%j;c#V+#`;g
zT>AJkte?Kj($*5z6-Ynk$<IE^*IxNL9i3p)K80_HTHs<y)T$HZId5IPOENV`;c?|k
zhhJVZObf?orkEEJRK!#U>}l{;yZrk<@h9n=dxW@Ir`D*`=vHTa7}Zcp5kw7Io$9+i
zb@n_@KKV3%@h|*4+`G5Wm%sdFe&tvG4llm&9p+j7drI^_P*MjKiPUI~a|V<{dOk{d
zSR`Ro$BZ4W)Fh(`BAFAnJGj6{X-AP1V5*5Ss}wTu$&)!z5K@SYMkqM6j;6oh!Ref0
z7IXLTBCn3lkZzu0>zQ?|Rh(SE&Dyyw>Wv6JN%-1JFK}sg!bg^G(rU+uuyS}%N})tW
zkX2IQDG%QZ@WPn2GfPOn#Fzr#!OE3~2zyIJuHeGO^V}OB^Kbuqf0A~$&cVGS{Cb_A
z|D`W*cl!{n4R5@5jj2(%WKLEphBd)~%lXdjJzVcRW9O6o_Lq_0_$swpla2K)*4DOI
zUEN@7Ym08Li}Dm38(VZcODN@m1TTP%tyBDoKlP{hg<tqNzVWrM@mKzf|0iGn*0&e@
zJox=8-B%NURY}}0bLd}3<)a1h%2m}>3CUo<X#arK(_0wdLwZ5QxAT4G!*L}@S{AHa
zJVTx)%m*i|Y@J36@M0f4c<JVV?q~iGzxI#+0a*s;nwK%M;M=$UXR`9g9J=4Pa1P~F
z>i;;4?<}E{=;;VkDpX*RMS;>5lb1+OqJjt<RePz-v8ECfu1N1trgOr$MmZVLjwF+z
zr5h_0K1LeSq+~XXc=zBu*NY{Zm$z6seU|>UZO)xpV>B6Y^ZIR;_wVxf){xC+AHQBh
z7`@o%74L2`86up+_d=o|roFt%C%?2s7(1*jkwS9z>8H?9jZ@8-lkQnY`H(OD;V<F)
ziq1+4r(;f^Tj$n|BXTFWapQ=ShQlTWqglWdPy)W%pD}K2GF{oERF0ijf1BrD`#P;w
zi{+(tHaE|3{``4XSJ!BFyF^h95GdtSZ?<{p;SQHCKgyr}Gk=CxUiuDy<FEa7{_(H>
zGsctI`z~<5Pg$OG2%)fMA-*eJMH-hC@jQrYm`Nq+kdCK_Bq3_IAn;Kd%4CkakL^++
z{eU8!69xf=$Z3RvW9w-69s-57IptWfdvu;R##^N8n`}J2hO?6X&AY6e-Jo6zDW@sl
zeC=f}%=<jjyF;@TAp-pX+og~s^EsYWRU+jDs32nHRPO<{dk$7EULx%F2wlOsbLTji
z9P?-Y{2!&=sd0F4f?u!kQ$PC|c6U!OrRJ@-ZdTZyCS;{z6blZ?d1d<^bj~uyC;#Tx
zkk5USIBu}Ey20w|1}n>JY;0`O>2{F5Vr_MUPN!EDLxjSUK5HAB{E<KQ$N1@={wbdO
z*0=b7{lEVz&%N;C?>n~tu1jI@5>jGHjdcd8KzTk+dRQR|q8g>CY}{u30V)cJ+HG9m
zBmaNw{dc%s*LmKH{>GSn_1z0LDp3e_QX<tTtI?L^DmNWFj_ufqTjG*9iIaPhJm=nX
zlbqPGo%+O%oj5MB?AR%?71^qkNHz8X5CGA!0qnl_s<Y2I#<+jX1(32tinK}D$>sMv
zBmodu&%zwve8)G&_r331NQyWIU5SBctGbEL;|R|QDAJrpU^D1JO<EviWg9kVa(-<u
z$D)0dyBFAfc!4xgbT3_I{=ja0-zDAY^U0?k<K}3a02@K%z8fbm(P*{6%`1&NyRs#W
zq7d7%v80V_xi~?Cxjhp|r^J*6!VTDU+il2NjhZOfcf&!HpYnY_@*Q}-&Bn$yv-7ij
z^S8c}<xA^mr8#}-0`a^>6sE+bWTz%rD<VF7VFlgVP2UvcAOA4Y`WTHyo5h7C=4Ka}
zom*hn;v(&d4z^>nxVS{SHBrr!3UEA^J^S|a{Xg&neA~Bv3m^OFNBI~3>fi9hQ%_Tr
z`gtqz(&J?nb=T<UT9+tOVl1dnb%;cXKUG7dy=u0~_wbA%>kk-(3C`RUMnl>kB7jmU
z826rS$0JKp9M{9v1#zY+azUaJbSk*KF~yVlK4#u=7wc;SKKH2;43iY2;ljBtUZA++
z%?Fu%+gn&adlk_$95ttDPI#EHV!SaG8}Hg+qO|NvE<r)Rw}aJa;d&ivt!Wnb>?K-1
zN7I3g-hlu7>mTO6_ufgZZj+>nJvYp8=RG&k?}a>l{31J%VKfZ6U}c=viowVxy*Ola
z>@kYG#6|J?H@|@&|GPiU!NWIGZ#1wi3n3uQ5`OmQ{x5pnJ~!TYGxxsb9UM7wv@#S2
zgtTe2CivQSzneSnx{KfXt>5Nf{Of<i#^%;njaZ|#0^wqez{!U=I<6*zjKQ&N3X>Cc
zdxV!)n4XzI+78$iYsvOJ@;IT)bK>5B*@aoKZ3b6XXti2M&qZs6<M{L=#X~C=-NDm5
zdi)q|dk<Q`vGtFTtLQ6#(bwV?3p4hz)Vcz6WrN*rpv@Fk+s7uQ7;T_|GD;|Ry11<-
zVyc5F71?kE+Kg|?1;TaFae>Gcw(Z~<Lz0(BZIhdVGRheBW_V=%5ZU3In3$bqV`ZCX
zKJzqD6tlYCWoov;(YN2j>bWHzIDd{eq$k*KQY_y=mkQ}fK#-+r)dm(4X;(w<whTy%
z5GPCV9Yt$)k-1&FNJg7X?hV);u8@QoH@)Eyn;Sb|4Db8S`*2;4Jk9yVU-&&nT2dw<
zp;L0nHEb3o<w}qIa>g*u5M|2j!W>`s?r-3ZJKw<K;u6hf6VG$Wi=5y1&0nToZ}Gaj
z-oOnv9KrW|z#wc%t={6z_r8@oZoh+X|J0}W*?<1?Jp9-bDD~C5zquy3TdjGCu?)3|
zHZdjMWPr%}Se}dJyLiTs4@QJhf-^G-8q(qDdDvcYC@jY#ODh=?TNfm`q9~}aJr-Qu
zoa9)shv|E7qu3ns@MlgDrWsnprHehhKymBc`<Z#m8`wC%g6JELaE^N0#T4p#t`=p5
z?E(<inCPGkgI*V_*2M7?L2HWnrQJju7ic(!&Hjkr`<;*Rj{9ySs97YbVrl;@x4rI0
z2K|WRCzt8Q5Dp_QSvlvFVi;Ov%ORr^Pf_M2j!@ir&+GZ2pZM$Szu_qLdcA643qh78
z{HuTc&*=3B96o#_Z@lMT4jw%6C$PQE+u!*vZoB<9{>$(EKEL$uf0fOx?LU7K?U;dD
zNsLtafFr3-O%hw0+FTPg+QRZY?3#~PDze^?QIz1$PNA}rd@ur@+v^=mB5j8vE3h3G
zPiuxrK~aEG2|BhpyEers`v&ITbvu{OZ*c#|o+3_jj1ina+ojghyzVVGGW|7gW%=wS
z=$0H2mk633x-2WrSgk3`64$AuW`ylj(q=`;DD2|3r|_Hxo!KR(=4J?cD|8keE?mCI
zZ~xXGao@Y|KuATF!J(UWbN8EX<?7`Po;r4cZfda-MVzxU&ML)d=#ZY-VK`bt7bUJV
zyzA@V&ENW)e~&%;4^ykv#`ysKL6?8}PyZoG!7aDk&b@DWI|mLNtO`*;S`O{WX}<N_
zzMVI|@s0fQulxu8{jdH<`u)MnF7f|VMV-<LtqQC%$4Vs)&m-#X(3BarC5Rna`MM?%
zHU_=EgF7=tA*#6wfkA0amX<iK1-^@o#!(s^pRiQuBqr-G^1#+2@lu_^<|X>QE#CFM
zJMja-pr5m~lQOq|7Ynm3(lBO2&a-^%4S1VF4tc9I>orPKR5LzWkLmYR3@2$}Wf2>v
z9;G!k3oV6k13C-4>0Wr8nhI&QES4`_;`e^L$6a?H<ly0boIbt6nP(s3%*hpGvBl@^
zKgRm{h~sI&dZBQ2flZ1ntFh?z!XiKM*MEjPZ+iol?}8MSprX-eZ8&@8EQ8TD%U4eF
zz-Ru5qc`5pH+|!GaLX-s;dpjsqvH6?Oi%NpKl0Z&dec$<_Rsu1&YnI0GMD;K7uU4I
zv{AU>4wfk~0ze{N8<oV2R<~%>YSl~b`V?h})2gA%l3}lp$})VfI>`;YeFEE|)|o(+
z3OfkU!s27kTw&-h@zKwG0$~Dd>2Y!Nc?z9;RsNo@#dGo8;jMblrO|k{i>x&%$&qD_
z>pG}LKp-VzN0E<4SXyH@>ZmM7A}CZzQC6NRf$LMGF^=bBtCSLlC|Bq##zY++THnRy
z{C?^a9Zo#{6!Wvw%r4dmM>cV65S~vk;c)BM93efrhvmm-u(m$L(xSn!tcsX`K`H&5
zPSlvWCi9fl<x_Z#20?ua+ix<xw2$GjPg6H3hLZpCfsb(P_(|UU)|;r;8+7|4vZT-P
zr_YneeNH~J#m416$I_gOd5KXL<yt~fDl8%JJ&zyw-XGy_{M6sZ4H{#47gK58<q0QG
z9cN>6jgNie4>+*@Fz<TjyLtWV-$)QNL0SmmQg5|+&wJj@Ew|jlKmUba<kx@eKa-|k
zosk2hz8KqOSz?h>_dMc3mxjo&r66&t4vR`-9AnU%J>2ODEFsWEc^%tTrf{6f*IS}-
zlmXi(DitP8$OiL#ZfBlkvBq%wGK1bWZ~unda09`ppR?V|nAx+y{7i-ITVjqYCtruR
zIpTo3PNNo3no2%GkJsE7gON5uSjaMF>&%n1rlu=wcYNA&yXakdikgaOHY~1QzRdsl
z{XVzfd4T-~_i*m)8t2bG#<|mL$YPsE9z4Ou=7^JN!Dgv(^mvY2R(B_@xp{v02Y-@V
zZ+<<N=Tw@dV{1#L4d*YMXEf?^b>%dl`~1f^a`+bB^X_lq#v5<L@d7Xs+i{tmn&Jn(
z|NA&{_z3^-|N4J({=&tt%9_7mE{yj}0Vq?V%N$wcxR#4?U83$bb(3P-lGKsNnu{qa
z!Ee3M#hL6-m`b?$+Ve=W64$lSuE7-&M;9pR6BRjDTo88m^68y<+#?OPuRKeV40+%8
z+)bephW(P>P_cN!60=ivM!f-Bau4U8miX&K_BmbZK|rB%jIN9Tw56+ju-uA`nhw}J
z{R9*9yYOlnJ7`gxnj<`YnWkM(3k;`Eond=(o7dfQh&}uFaO~JcPMuih!kKm4a*g{x
zdYt90h-cE0ol@iJ0*eG&R*c`pLr3_DAN?6_I(jG4wLyxCKsL4;JALLfP;lYWF+Oqs
zA8^O*ck@l(_#GTNbTiiY2HW<078e)zdw=Jrx#{Rp{{BDsN35=`y{z?nc{6potSIcJ
zK&T8)7pN!#8)}w;q#!MFgwm8qY{$cPC0QAxk{q|;5p^R}o?|->!ja@zisLv)FCegL
zcwRzgS+tGIiNOrpwSb@%QWTQ8`6ae@x-9I?ICA?9-0ClqW<F=1xysVQEbT_X6Qcw<
z+{<Ktg?htA7)`0l@u^pk=OwWbxJDAHlHrq2BksPhYBgDciP<?WCk2hc72a^yU1Y}K
zd38>nJjI#wmodsF9E8{+Bj}&z*zy_9XPP0EM3#i=8QImF|CV>W9ouzz`pG9)-&kQ~
z{R*oaD+~t%c6N4n_zMpM3g7o=v;#J`HhA)>$GGS2xA6Y&{66;YJB*Y9Ei2K(x4-#K
z{MxVl5<mV^Kh3e@Cooj3wO30U>$(qOS)xaMq+NnBxQ<I4My#A#rsKMJ6D{m|gCtG}
znhk^$?5u5L+ZJJO$lSi&C}9v%SA(^-jTRb5x|~@X@KmqGmE<Dpg8{ziu#uldnb(Ac
z|2(3)orO{t7?oj{38F}GErU)%Y+aBVMP8OTflp~kYV8JTFG6@eUf>ZADsy|!wlSq5
zNfKPoLkMtv3)@QxjlnMzSGp5C6}J)j26AN$Z^mKoEp_(ncc}%_3`0d}wHR!U2pq*m
zf1BeQV9J}R*B)bjZUU`R%CT1~NQ$B)GZG6)UWRNw_IVZ$-dY_}ESt%>1<q$F6Y&;@
z4(}#OG92Oai4Q-_(*E6acZMuqSVI+tMSFvXANd>?qMR#Akr;!kOMJ`1^&FHoyz%XC
z=j-16EhM9qFpP<k5j)*2!Z4)Q>+;A$k5Q_adfj7V;|fnb{V+Ejy_0YG=J#{gU2np%
zU7(sToS&WM|MQRjA#-!H{QST8S446As(0j!E)c>)n-ZaOT$4YC?Gi{8S(zhBg>qeN
z$HjIe*)Tz+IZoXt?)6Ybj&0irTasrPj_V*DAHN>pdI`C#u>I6<ik-m6Z-x{_li9gN
zcDg<0c4r*C<p4MNHPX!E!l_jj=VoX%0*;MR<Y+gY!5X!Ch3#tGe#Z!kToG$<kVLv<
zbo?2_U2j7x1D2pOJ;T+spgvsTb$8rOt_`kN<Mgv<ICt?XMp;C|2utJygL9m?a-NI1
zW`wwo?LtCfRbsSnc<a4buFHvIPqVSL#@glz>l<qfM?-quE{{L*IH2%+mqsgKdwZK>
z$DiV^JKw<jzU8~vx91>2f~u5Tc*{L^^Kbs;FYq(}uYbUalc%Z{=&OkBFOUi7vYJst
zP>n3899QR*$q?yDT&*yLA}jJrC)W27mV;LdNcs`vB~HU5?)K2CsP=kWkR>sm?;+iQ
zKo}fvNG2^>+T&z@nsCCW=Js%DGCQ}6jm>ptmvU~t;UI2MCmdN^IJLr_r5U0i;fa1u
zknN$<*``)^5T>M5)oiaSHF>5;k+??C%QHq#J&kztd(p-qZHv~_G?(L)nRJsk+<hzB
z5(wdP>cknIe&#%uap?C3G~5!idXmSkUf@yzBgVuc0Y-qaE79S%f8Bj#Ma~nCKFr$2
zDp%K*t9%#^+1Oa;_*2KREJ0B7X|{bfHrILNkuUJpd*8*kf7|!4xVWd{k4S+dCExP(
z?_pxH%}@QUzr%$~%P$!NUN%tzDpektD?A}9YJM2vHC(bXri?;dVPV&ObedA65nii7
zk{7skfa3)eSw@n^IF^lVTcl}RJtfD*A*b#Ll2M&#@&@jH(@iv{CP}t));BkZ(uA4m
z8+hvJ6Wn;qE!=s}0S+Bm<iwNb$aKj)UvrStA2O`=TsrNLX5B?8jTuiM3#}234bsLa
zMb_&flNe=6Y-Qoi?4mnTG%r3yyFH?4FHm=8dE2dvC?ko5$Hw+HCqDNPE?$0?XA@00
zmR}GOEY~J4N{YM$3?F~+0iJ&5DF%Zbin8*w)dtG4Bu!I_vQoo0W{6ycR=dXJ)PS9x
z4UQdqjBo$8@8Rp;`#x&*8rt>9vyfYEzJ-7HZ~rwv@?$^FiIb;Z&JzA9LLiJpCoyK!
zLrgjdqezmJbLXxyVGG(blemor;b?^Ix>%me&c+V5=Mk>2Gqr00yI!L#OKedI$s)jU
zJ+dt3k53J8=l1g0#~z?Gf|2P~au%<}pG%l=Q6~UdR_}R{rD&bftl1P{LKa832rSRT
z4Qgmz;I``IX@*nxu~UaE9HGh#$99pzkR>sVW{aR6;5*y6o{h?CtcqP6eA^+Kdm6ZA
zk}ORyBot-Ac7K(_H{HnG;v&yJy+YW_*}HESgP!8@qakyBhgPS+b7Xb1s12CP-J*~J
zTUf-}kX>EH%CgFD)Chv<dHP{VedQ!?JbWK&VhLgUtgLzL^hVU{4hIg+u(N!UjT4V>
zaxh|3Lm>>3iXc*IMLHhw2=kN={N8V{v$M(7l`D**A!(M7WjUj8L>#5popJn{!^BjJ
zZnwwrXP)4#_uj`}`mP^f*U}ys7Yv?b^W#7EqxgZ(KmMmbOBBb~1(08rH6Q!dK8Nj<
z<)Mipyt+$XMwC&6D=qArkIpiRG{$SzN%9ic_i;R*BF{*(1p6AcCvkP!bsQXWYOWv-
zYs3@#x$EvD)F&rMc5*hiwurKnsi^}Tf96>Z-*h9l-nEw-4$X7w*hO+(a@U*pbM_;K
z^}b8H715|UV{9J}))-S^yQl~sS-+1+6Z9C{-RVX8BTaMpIIUJl(VnB`%<z_*=21vu
zJhr+!Jp1q`SiX9WGpQzeKDIkHd3nuM$nfdUe~xEP9cMJ`QIy$Oq_)ELEX$sU?Uj~W
zvlTEgIiTC!<iyFR_~vi?PQLcOZy=}z6}IOQH{E!YpZ~d^<!}A%zss|y&%7#Y{gt9D
ziZT0LjPs$ycWm-fA@T?>u*mWeDvI!I2dfq!M<Gg;c&!F$S>V=u9M2;TGs+~vv27fu
zntv3gk}j&*3g3d!u!Y`zgg4xCL&cEYOjuc8C&>*{lLz?1Lyz&+cfO6=->{oQH}2x8
zN1w&=Ebh4XAS)j)SnEDXvk_4XBuXhN0=EI95S}}BC@INyc940B)+IJzPwiq5Dr(Pu
zj)}%9dSZcEdxoXkmPkuOY6V=msyX}6$5>mv$eGNL7=soXTL>)2CM(KHLQM!h_{V?5
z{SSPC;jl}oit%1=D2kFat@5EPtG(W`1)YgH6O(;5w^w=M$%pvv@A?7W@%FE)x*)EL
ziem14(_8pA|LR}zV?X|rT)uqeC9U7fCQ3jxf8iD(ZnB9ea~x$*SrcgqOi^OF_H{4B
zJk4;Yn-#rTNU&^#?P798o>%iPj^i-u_eoK>BEuJ&C=WQBXY?L=ltq6Z2d4}6Hs<)?
zC(m*AR3GU_eD>2{;PyMV*t54mp5&~rq|8rEvm-VAxI-9S#&f7lnoF=H6j_4hRl}a5
zP{=GpkA~z$f-M5P$ywqE{5&K({~-OHj6o?WwMAxmY)$QFB~ui=70#szqw4%N_O8=}
z=@;KFEiYf?>Xj=<N%fBq*DG$N49du{HA)4kNgUDLiJ6$J5yuI?{u{r<g$oz>E8qXa
z%*@T>8bKaL+;;OV{KC)ubN=cN{V-QoSN=51_>w{x3spi9ji?tHXqPz6c=pT&=`dws
zUxW3v6<VEXgkw=_)EM@L<XM5pBZ5v7uhYb6LE0OT>})f2<6)$<u#C&6PcLKdeH|B`
zd5mH97n23{|F2<;z|K<w9fL}+^Av6SXi=czn83D?zJqXW(*B5As|{I!krK=CKqeKV
zt&DLT2O$jItu3@Iu~ke>YO<)whBHB`2RLgDWVuSIC6AuEz+HDA=E}xJ9(}aW8{cvp
zyA~%H4n~|lx5}X-OMK?|kj-I(`C%8k9aL)WW!3dS7ucSImR2QN97X7GNM5G+!o#24
z#b5-xyO&w{pZ|)@P~z1m@taf3w>y-x`?);o6J33pi=#fPT2{K7V~L-*sEpsQ<?AOO
zc#zM2?m;Z8^2nCrIk$we)F>?}Dp4xLFk@#YqSf}9oa!+g{ywMAoZ$Pv_lJ4Qo8N}l
zXi=yEwo?4HAN;GtQOeK#!Y`726&zHwF*rqpn{FdihEsyd8x=#jEUUKCSdyg3a@=}j
z?6M<4+6c?VXhoi9V{EqxhXYchaLMp6#6`e`Jg5KI6U=*iIWSeQyFSZ@AH2wg(|x3$
z@X+TT<Ca^uSz2mPq&XXF8FLd;bfsYsw~3;wcus&)<29F1W(k&SqvdnhZpLeF<Ks<C
z5y#-=G1<k>Gw9|Fm7vr%x#hAmv6t0cQ4ZF)m?ngO65F#c#`Y^$R#{zHLCWd|PYS`c
zgwXm+u{}*<`n`lsCm@M4KJdH0#-&S_`0nrier9Io&_+;1F*o0Mlz;qx|KI$?PyNl8
z-Fo}8hB1aL&#46gLR4-^UOvK328cYt)fURBU;9g91s>Ry14ou+1QT^+USWWdHqvq_
zbV8P=SW@CTHvOF)5~HYDC9cs7vIeKhKG6e@a>L{eEHxu`*|U7){*xSkY6IydJpS+}
zdDEMZ(r#DJc;!;WRA-89A?b&8;;4&b`(q3c2x-XE2*>v-?aRDCWGOltQ4}c_4#Cts
zVW_C5eZs4sWY8<<X-TPVQYT<zazATPOwnKCe3lYZ38Z5rrO}MOXs<V%J9hy~m?{_k
z#9m*H_j)tV2Z11tM(k`yOitBF;*4MV<zM8&g$sQ5cYi<a&IH;B$|U99H@}&G_D}u^
zf8!^9l3u_671!`(6?K-dD?vh|ag#1$w1O#9k^%xbgKbG1*P$#6@|^)DFOimqbX*+E
zV01-y7M6|c`V^U_$TNy8#q~UrQ9{$Rag3zU5`_yz$@U;(A`r|^JKS~a0vE3KNlQf$
zyDXpCVE@u0P2b^M?-DEVh*)$<bB7`=D2j|cOEHB;yAqwHc#Vmw@1CawLQq6Qtg@iA
zeEgtBSqer`&Sb-*=37wY7~#<^8=UKIaiuq)KZ@AVAf>^mstNl1xc)O^C{<-^c5U&+
zSca7A|GoAn1X_Vg3>#|^QCLvqf`=dZ97&S$e|-1%a^%oa9ATl#oHxDkjr^m3@DKRu
zzw^_?N&2dEXr$6;m7}Eq*Jb(qD(f2?JbP}38|PehwsNNDW>8Aen(k0UDch?%%q?^X
z`zg(tO8GtNbs4U%v2f^M#h4S4)vYZa+>*TYov-KL{I_5IbCGO$Ext5_sD`~Y8Ydq>
z{{mWPn9`=SXDfQX>tRsjX@<&+YJ9<Ua4Z{RsC+r?YF1Wd32~g^I1Y{_>4ih;F1QGy
z(qgA9**UvP?zp_}NRHRu%i~X4-2d^@G^PycD5Myj=FYqJk;XZ3-{NZ0qpLPaiY7_2
zQk~Mv5?vUiEy&^sI|wR?kfOxHphrW>GD8a+uhF2$3@R(w-Ox-}87f~!_pT7`_&jy`
z6vw(F$YL&+3N1hyQE6eKDfP9kMHRZxBn3bP>Y}nFp!yZYGk?`guRx+o!&aIy9OQ(f
z9Hj*R@z?%6Tbo_p`wibjqun7LhPW2|*pK`u7cO4ncYgPGuM-+x>9C}Y6k}{pdx&rq
zQ)Z+k_;RYkcE>@L1x0U&DN2NNz8KrJLRc0KKBb8%^1Q-!*Ch#48lH`VpwQwvw(krh
zI=*0b(&qM~b1bh8NXwEkcDQnGi@l2rc)r7h{$*B^5sB=PB{oG?P!twLmQ@yojzs4v
zUcFtJu;(d(6cka2EDK8M;rTVn62drVqV5q~$99LFs&k>Y%}RgBa1_%uV`1r60oymT
zm^dma3c=$~JWQHqeCxNrpMwVuV+#wda_)KE>-oDs{kQoC|L7l*q^}OPf4MM9L7pLN
zO+bTI_{lb;>zE=XQ5JzSi!2Kq*QLyI%5J5#XxSc?<KS90Mrlf|5w?RX3`H7~r748~
z*K>%5F|EMC(vn<T7~7>xV>Z`AYIe!=QotQY=QzL8qp&pDNOJM?I&XT@EeP9T)IU!&
z?2$P`lF+8e3(7*0rzw;M<tR*+<25>Ht<h;hO;jvxr_3p(L#@#uP83O6Fw^$wG;DNH
zg7nzP>pZi)!j=AzewZ*oAcaO7UN9ewHWWOU56{nlFP9G}P?080M?_(PR+0xE`~=}B
z;{D(KJ?vfDM=5QzF8TWRyqCZ6qd&$!{%8NsS6suF*M3!?m4GZm47RaEi3Ql6K$Ii0
zs6eB!gdpt=@oE8$x!FpKiV}*PGSAUv1#pFd5`!*F;xxq;F14BiQevcoP8_BWwAfhg
za<MmwYHxGzt#yv<61Ych;xnH*M=O{lDTX+XMB7)nuw2Ko1xi&E`=ZDRha<E$SV9nZ
zKCR9SQkCS<5L+u`k)w-}TxcxIMj1(Ol(B1SijFkIy2fhcaeQM1TC+?W&H5m!#tx*o
zKBWCc&+1S7%zW9O{mEC=T>rPy5RM8iU0koOz@<&}A0oc%|M*_+yyGqq8lm7jzx_M7
z|C1l%zkTozu5UE2Wa$1T+H`On4`q8?K6eGj^*A_Vu{hIWYGxj3N&IFVjN#c6=h(G0
zN1kM~r>C(jm!v-+-dLm6Y++BdP+F6hIluSFMGn8~NBP5#{UN>NHHPb7l#r$p{FY^j
zGu*;f309d`*_j(uX)s2ii;{F0;srjn_7sloVJr=KL0J@N1D55YtrC<b4kIkv!V@;y
zv#_uzjg87&I<s);+!`ALpW9v^^1bgn%<1zR?05yIpB}Ke(PeeDMXS{y%_G9dVC<4C
z%Fwx{EHmQK2%`+zF*pVqtyyHLDAN(H8j~`LlH3K)@hP>W8|ECCYg4ZmXe~%-v$35K
zW{SzS&*>{8a$RY0mAd?rXZ75h{JeKhKmY&y$=?YE#Wex>BvGudhUA&Sb$tH#qaS3a
zyT$vz^Doh9btuw=iB5-q`1k)$o__KPPMtpe>S)>NF}CM9V%SAq!*&;h3d!OUEh=o!
z`a|3xpfNi$#&$Kv_F|0fJcsS7B+W7`;SdBq2#FC6I<cAD+hFTzpXGiB)!yNaH`h40
z2=2ineBrYfX!;YR#R%ILXls?_%Qb8Zl#Z*Sp(uz(AzB$M7ktO3(V0f5l02!1JJ+$@
zeGc3EVaCGbBrR!3b--HWaq{XKqBYHw(QFRm>)5VdEo=`9t}Jhi&FpQW-;DY8Z+kzt
z-g-L}8X@3YzWJMZ@WD^<p%4GjtAg!cE>w{tq6poXuE-~8f*9@~y~;5{cp&mVX_O+y
zqX^+nk6NRF-)sONP?$7F<;9qQBgl=Z?DmQrous&y3!Y!Ktc8P$ZDwz1(p~FvZo7ds
z)8lQoc-(Yo0kLP2&wc75)14NX8d0hOW3RBh+`zLnN=0a;$%~w56k@bNNJ-tV(`ZeA
zD#=DYTwx&3D<=|1V>=#2Dd>k8hjvY2iIPOOxHJ$P+u5K#HBY-NSRKY-#8~R(rRT$o
z{jM*{2L*$E%7t=ue4w@%j)r{iU;aT39XtfN!4o!r^#}jzD__(x>OrkW9u851ZDgnN
zX%t$c%d(o8vMPHFUnz8{NQ$h|j1jUDB^)z94Vt9L2uDM7SyUD>X~x9N0=ZFmR>q!*
z2DPOQmRx6crp2wd>?IC2ICSexOf1cF_uY45xfV*NoIG)wKm4PoAX#L>ws49ZX+e@C
z3<nWPYdU_7$(aS5R+}<U$wmW$z(c1gy2vQ493un>kN#jpnx@ny=2`1HeDc&P&Yn4p
zbXdKzMqZlg)I6r}zZ#*nAx#t)FRYU1nk+5&Prv@FnD6@m?zrtvjM3EVb$;rneu__i
z`T@4Py;uA@oPXfsEbQNh^c_5{c;g#xB@RRKp^NKzAPrulhVX38oVtvK#PeLFEwMd^
zbT}m1TBpnknuixb8$@9^e&Pzpdu{H?QvT=tfAorf@7Ll_A0|sNL7g&=F~c2Xpb)Z>
zj8H{RkrqfxVA~#~DS4Vvo1UPgI<2ZKr_A!|#so!}XC!fmL_=mk30jkL6liQ)&~fWD
z_DtY<0}k%rN3-UDwax9Xn_>30Lu|Y=W^!VZpe_i*kf%@lF{|f0Og0QcrAP~kykIzt
zP)gxj9_@)~YLipwA|o3O@f>>$=sBg#!IEff`hyha8nkO2Hu4UiKhtG7kC{GrgyqK`
zqfi>7t6PUxE6P&S-5GM~<avt1kfkM26!Qb${{yrdEy`T6fB$}d@+W?rpZr@tjmlqn
z0+AUFAgED<BlK_w*}8`9<~eMa65Dn0N`+R6q{zqEE+}>7ilI$qg_aaKQ8*f7yN${+
zI#Y8LV{BjQ)CqRAvE&9b(@kzVx|<~0<c6a+(pjA2uDfo-awJM;oIZVyKmNoqNEhkY
z61&W?u3`JI!uFQeWMX;_r`bZ~8ToKT;JN55Ll*h-u-#=a49T*LT6=~|J%<O*uH!7t
zpoPQw)eVZT7PhB~<xA@nV{HHJ5Bvsy>ASv%TW`J{Z4~udoge+tAK}5zevX}9|5f`g
z&yRFCWbLtsxZxd>n4m^J>|>3(2+v20>X{g&P)g%C659>%%Mzs&NnY*sQt(B4y)v3O
z&xpbys1jS)s61n8zD;gQYEHs|_6*)`3)}9pXZJKm4lfc9*SPV>VLJ0OyyZ=IW4VwQ
zF;6^xod5m@kI-<A;F|(hmk6N=M=7IGf;NUmt--|XBC=jbXDR7mK)vRp(iBl-XpvQ3
zzs8~0Pbp^$+{P3aHXR-~bA{HCC89yX*4j3O8vCBVilH^cQNg)0SI9F>niTxXul_rJ
z=&$||2M-)ZD@~(bdr6=5vWmKm6HnrGTI5McJxh?(#(Fkj+ZHH;V=8rbtL8(JA<_~V
z*a%l**8|$t3~3aTMj=jOV+%>43~`pCa?Q@gs~o-MMgn_-PPU0JZe+N*NfNGe=NtEP
z^!7Jlq$V#4E-s(t<g*vKvb@Ij_JE1my_k`QyK1OgmoWymwsuI84A*s-o0+5D>>wP6
zJWGhWT^fxVI!i0y%2SH6KuCuq(r960`7MsGUO`VcXwP<d{F9HM3k|~J)i!T5REwRZ
zhAUUrDaKxh|KUIWGC%r5Kh8}z-h$SOJMOrHZ~mrl;y?YyZ@%K!^yfeR30B(;{LU0d
zcG)cMYT)}efi+86q)5j`xDG4JTby0#^47ceBdltcCJ7_LtxeL=h^eLBC@Cr9l+E5Y
zzxLEcR_s0e{IC8jVf>n~@L!}zRxVL%*2zX)+))o1_$UlQS92m#N*p1uU7K2`MVV%#
zgAm8}uss)1^Qlc#QWUzV=D^CbAWss6uuz#HDz;eKTPLt~SPU|}+5tAph)b6*@U8E^
zmu6!x&B+eF?=#sc+3fW=d*Lz<eEK0a`Y{u8O}w6@rn^8%uNM-IVi1z4&Lo|wIi&AX
z<|#?PPrc@&(*#}QC@ZUsx24A@jL7nW&g2x&T<UQljyZ7mEqwl=V<f|vQp;Dp^;MnL
z1^r&a=~L&)vw|#3QDx3w`zt?8z19Taec$>%{?l*%CSQ2?(O3LCY(9I8-JK?R8dA#>
zqzJ}dzF=Dx*x-!C#;;?0Rw4rnVO!WWkJj82Srn7T5l(7d$Mz)8D04&a@+yaJJdAH|
z)6TZ>#UY;E-Xe`Qx$X6PIeg1K7@;Z3g3DLV^X!>rR<5qo-5D`4y$2n7xNC;Gyoxr^
z-|3O2mAuaE^epve8(~-0i*dh4y%C`E6kQgm=VN<p&{#;X#mV&*%w(O`bepFid;(o+
z%vTZH^VG0%b%UZPDT<um{H_1Q4}9MbapU2mXsx*AmRotxd)~tbe&=_;IxW~N<J?F7
zh|ba;num{4Bq5C=MdDSKkwRcwHpW&sf+`>C0s4AARLp0mS)=V(q)|i~hB$JZ52Yqn
z8AYn;Us~nH8xK-<R%zv3eCu{jZfv8974CWKA?9v*Gr}>%Ny3#&Cwb=NdDd1o+1Vbk
zwEsr(9S471&~R5NON(xAn>;J2*F5HC<_Q`tkg_804Ejvh19X~H_Lfv^d4xlpWC+Uz
zr^(Y-FA?mSLAo~QPo1d<mS3fOPzsVnbLr9wd0tSI8Nd9?|CS&BvA@ZlrF}1%1^u#$
zx^tiW7}E>8aGP}k$Hrh|nF7mlupEc7RA}Vb`%Pl|enpTdD;4rW72~9^Xf;~oNk*Xt
z6_tFH;dzp+P_nxe%r+HE$vRv85jP#Wk;fi7!xuhwn%XycSdQe_ljr!|-~J?yS)k@M
zalJaX+`ho%Y?GDq|COm3O_Zi=Z}+ee%uRQgnVF$98jT>%Q$&A<APC5l5uy|nq^PXG
zb2O=zxVDcfTy~UX?xwvw_L)a3YctEjDqp=*+;dS~<tn2g6RfZ8V2nnal3)3czr^4D
ziNDX>u6aDi<NLnnFY{mj+ka;ihOg+$@5c#Il+nM^!|FR6nz{u?2yiV-SyBB*TYdic
zq0_wa$O5KNgy{g+aVgRiofb??%@VZRNFhk`gfBdGfu}YzF6YN6^lP#uf6?OX10Sb3
zJ&lkywXrv=Fp6t#2*#?WN;IH_KsYwO<zU$sN@;SfumhjE<Din1GRugigCk3#q_V(S
z+l-lBGBj&_W_H#{^E+_$4C|M!a^}=k-u%`(8TEUdUSH?G|Ne*BTv6Dxn4E00f8i#Y
za*FWEaazp{^3q~sV;fayCMWAG%r9U&4n|1gJf-Mv(Q4FDS&S$RDo-hk0^1T4g+$su
z2A>Oqn1x#oaPjOiJL_8*Q5lK9`n#z?8v)8N8l{{&w@j%@3YFnG9zXDx{~C_((r&f+
z!N2l@JpTBT<mH#Pg8wXW@qte<J-2|{tmE4@7#nFy<k$kREEP&f?6I$>?R%BauTmt#
zkW!Un|5uA<y-A+r6l#dk7Ezetx)$4^WKjub8YPR#2HS&>8xI`j$;Zy|$fwQ`cplOg
zoH%}w|M7bd;Fx&=uYu#%IeP0H6VpvrFa80O)0#NV*y;ALj9_-E&GgjNbJ$*_FxbKO
z1BxU>C`nP|=)Ayl45^YhwudTRx>_=GWH(QK;mJQgY|jOo8{G=qQT*m_{~AC3Lw}Rm
zg*jZu<-5M~{ruq{euyxNUi|}xfx+4~=RW#jj@BA@^(MY$qm6@Y6w+~#j*Yf*v?Y;@
zg}#HJ@^vptvM8d|B|-=s+revgh(;00Sm;6$#wordS?f#o3rVY$GrzGynBT&|-G@1G
z{2ZVE*eTxlPKE2a-2d@M`RIqAz-6AA-@x^n-1DaW7{8!*@%Ne9Z8MBUba(nVQnD~N
z!PMk5nJFnX<avx8^{aMkGQ^Y?g-NgH1KP&1J+j<o<a*5Pp63gn{CqVxYFWtQtC$a3
zlck2$l}(J%C|&LKKm9j;1|0DfKLuW9QCD2J%;}GPh=cEWH-%FX9RvlIV}k|4v&f@_
zh83WVCLW9!Z0ykA*&$Cd#CVX_s|Wba7O~QF`!SxbTqNwGq?;?cgCSG?ER+6eR!_9p
zd*3(HyJ<J4Pn>7pp>t%J<l`SYhnpS5wmg)wIdE{A1ACV^`@xSfr5~ZwsdIUG9i<AU
zr<yD*?!vWgx`Q5`7LX>mWrk3iFdX3_D2f7A=F|d#alo}bc9N8-eY-e+@iH4bJvx@n
zNNZlr*Xx>)=CvO~#Y$63v$fg7aa=4ce(ksZ13&rWe-pcDaocUTannsldH9i6_7ty$
zO`!x97Au=0u5N5IRTngyElLb3j(PB@WrWtuwS9~R&$4l(g;I*TU!&IPARL=f9CP*D
z3cq)@%cV5CHd*vq`~{1h$DU+m*Amlr-iC2Jw6U;Ff#dpEzD;2YRA~qt2Wbq-2#P47
z+wYS`F<L1s+rn{OT;C@*Hr+vjQ)&i%MaviLB$95F)1GcH=U(9K`4imx<~Q-tk7^$J
z=qeK($>j7LAOGMvuAG>~rU|Yf^(0^O_S@Nd_!M^bIc9fv=<Tc$MG^Iy!>+{z8udE;
zIK(pooh4Xth-2AA;gFV*WLZv`Wz>9#M%64<l7lxrMQSuB&s?G=C7BRpuW<47!ch(6
zD`iNNjLXYcu`J2`4}608=>^{V^<R%I0>1XX`&e4q#l=g@ujt!TEMMj9r#`}g``$_E
zc;vap6$O@KUB~t!PVp=sZ4AjMWVqR--|bPPY2^oQTeyLb-)xd7O@EN!YE79-?4qQX
z+w_M+CjA*E{Ijf|YO?h9_tHPIi*wIjWY2+%<hkI1k6geh4qyq7Qd=ClVTygb7rF4^
zPcf;Vpw+Iia%B^x3Z|x-%*`+0*cQEEwdSNLZkZ#rCW=B_Dwz&d76krt*xpT2Cig6G
z@$w2=y*}+fXKdfz?qfR+63K7>m*3<^e(1-r8*OgB`DSjs@d%GS{=}>Hy{?5pU};Hs
z{3PcmKf?aEzJ=WJC^CgG1(x){0$fR%=F~kOtu=9fNPm5sLAOVdW{9yCoL3L<+ijvu
z)9c6Bg`&&^swmk^GDcxUvo*ntf1ItSI^1;MyXlP*jz6)?u6-Vx>lqJyauY8<fV4c6
zvAFrRT}(FHT>QlUWX?T~A2e8ByMjS6J3GPL+yb@~Y=;STWiW9}&1iI9l0*?5At;I*
zqpQ7M<^sp^Sc@YT4({dH$usoBh>m5mqZG+kJ0G;6EEVhP+c>U!Js-aFJKq12KJ#T2
zby6!<9)B9I+2Y_-$-@sl&B9WLshMfiMwfUnqUA^iWln#mOBBY~(k1Xbf|iHT8l@FU
zlrY%oVQh;|!>7nnW}6MxHx#3?pjN~fQBe0&cBfCX{0Yf{yWfUlmb1rl&Yjw(i0Vu<
zB|*dG&Ntpb(?~A;&rj3XevsE4X%S^zl4MA$Rbz2+7qxnuB1;LAgb5>&QH)z;^ftFi
z>Jzlff<h})nGyJ4d<)0+xX|yDHm6vA_98WDp-_yprhFCa5ivduUsuMbV$*4DC<?>I
z+BQMp@aR(yasMYi!Pmd{8wqLw?|Rp}c;wN?UQw}1p)IsAlvpIG#q!lF4C<P>nR$HQ
z<LXwIM^*;B@74)i-@_6Xwry1}4GQCUh?>X7c9*Ez;{(r}<Dp^7SL5vSTKwt49)+C#
z_(yT-9*&Tle|DLfDW68mAhjlmV`Q3AN{i8O$j<r}VRwiss^NNcHCJUUfpk6WS`AxT
z7!*^}b&_Gm$mEo952c!PYAduh9%l22CU1J%4SeCTh!c-@xp3wTo2v#}yLds#t^+gN
zbJq=QKYfwWlMit7To20<^!r^r-)3>w921kX7*#TiBI>$&&jVeMM=?<t(a;)Y6h;@+
z0)h4guH|vI*QLH|hNnLF7-)?x1%pzNy^8O-l=4deU0~1}lDOc?@*1A&@cV!8yBs`t
zkfTR#Vs>VRx7_;{E-t@xG18axU6f+&sb}ySEe=eUJpR}T<`-K`OiiJ-`lN#qO-C{;
z3kKaDag<<72jBGw8qM*of+kH8hP^(<wrJI>_V;wdXLCytssg`AFqG8%lwIjFT=}$Q
z-(7D(F~fxuITz0CP{egQbxF{0xb=1WX&AwkkA9YV_w(Fws6m|dNzxI`MveLT1%g_O
zGS7(8gpL--B*rZ>23tEMwGPd){1ezNacr06!GN?e$(6I0s0j-t6rt9y4iRM>FNW=9
zVc1&l68H{J9)FAnA9#STyYJlufzQ``&DZew6Hk6sgoqt1$P1P~_XvIvu-7a2{1=|&
zz@a&s?VO@4$-|h2W#|?Ky`3GBIK{C%eAlDq`Dm?BS`m*z`rBP_Y??I>m8aBOHLh+H
zWV*!7B8+I#3I;58A7%C9nw#&w7srcv>T^Rbo*7`$Iuk96=7i6k_uPOT<y`&nCz(th
z=FTG>w)z|7Ma*QU$>QP?p5Gu#62dH_Py(5SxMj}H#s=&TsTT!>Do|y0u(K>1A#5&Z
z5d&AUc439Ow8)JjH2PKDtHgMm<%@m|jZy|O!}`i5wZP$1pZx?k>^u09KJ(?VUvbD{
z&AA63W$|tM*)>1S`Ex6L;fWPy8Xh<Aog^+Z9Lq2>HN~C-`*5U#R+==4DT>NINB~8e
zGZ^$4ghMRHW^+{X$fl%_23vzr1>i8(-a+hqhV%DtQ=dA*(w(<(Xx~1fB*hZ&>=RcQ
zJ$Z%E_ESv8k8<Q-ffNvhBN`2h#kn~;%_)jPkfuYjEJqs1(wMFDmsx%KBnOuckQW76
zk&{Fjj?e^wMN$;RttMwrUm#0MupEqm5KVPld6DJ5*0vEab_;q*Atku3g=5>Gxjt-q
zts`IzRbjWjLs0kmZy)+E+<E6+?Av<-UvuwU`FZyj$+G<AEbEKLHLg#h48n3KrKS*y
zmCaQ;tqz|$y~C6xXxBVaRZ#P66jmibsub>|&&B0+uC8qJP`}52zLM}-!}TwAC}ELo
z3^@6bPx0p0&#|+%&Du(aU28#6Fk5%2NrlQ|j5f5|Z5HO{v87eri0Dd>tw=M%{*b|-
zk8c|YETTklwqK*4Wwf*bQlzxl*FMkk*^sr(x3On-ipAMQ0&9-N`G|JgV{1KR@4_rA
zUw9IE;X#hho@Uo<i`C65*iOmJ<P0;@3)PHFIv~nYK%mluVld#+@sqUc6HFEbnJUQA
z1kW*O-^L*6+JZ|PYYevsxWXo-q;E{6zxFDrCkg?M<BYvFt6wvwdeOBKM5C1TwGAB4
z;RC<(Tl~!5{y*tVPIBKn-^B-h=XYOGZ&-~@mKZL4{t4#q-OJ+a6c;b7@yOHbOx0bE
z?w%l4IgT_;Pfp_RYmBj7k)|nSks}2NGeMCR3<m>-@d(Sd*^Wvc-<FiZ;Ajvk2V7=a
zT}1B-Tzs%gZSo+Cx8BHsy}OB%3`@Y-r&kD%uMqAWr;|L%!F^+Fk3#Bo$^7gr?ZzZU
zDaf)BSzhfgd77|&afP)Lr#Y~=ufq1SAdNE|G`??<6eUTs!G*JzDr}c77>HhNY<H}&
z@qD%Bm8BSCO+zRO&E{s0pyu(RKmH%wdfRR6UE0r^-}ol}t^4m@QEdP6VX!QNHV*M1
z=j;Oyv+Ha2F+bJe)Tve0dzaWfRpZF63F0EgwIvJlbNI~$wy@Dk5yvsAEXI4iAx|>;
z{XWAu#CB}14oV){fMy_&rAFuifz9H?8glE?T=>`~orRm2zx5zDA8`re7)ryLr>@d_
z_#9^IG3JY>IC#)d7Me5;X*F#Y=jUnG+GGX9S)U@8SZI<sV)gU|cFvvS#yj6s<wKs5
zMj4JYc!A9zO(|L}PCa=BT|#O(C=3xAL3JQ~!T3|=gCzy7Tjhh%fEiotjbCeHAWf?q
zJI{Cdo&WlqFX?k%rcIcj&;~I|xO)By?|kE78jU*1qsLj8449wqAf<&-CDO7mQjnn#
zw!m)IFcIjosO)?ks86*CCYqE*Nibvchv%2+*pQbg6eZGgF~+cKBBo(K&E@TLbpQBq
zVqL>X2dM=MMUTK-W}&^s{0(&shA4~heIL)BVrpU<RhC4_4x=!nC~b6E(A^qx{K;o|
z{SEtRx2K8Ygt$n_M?++hQS$|xWx+a%&C44UwuA8Nq@zC7==TdRofLw0vr)aIc}`Xo
zFNr;8S%OxxL8H;Yu^oh|u8FfOC&^OsB1h}WOKvpE*xKAd82IoX{~<s0SAL8`H{8JV
z%rq-kS6|MuzG#%jppc|m(ewmLmyDvAX1&d&jS*)zV(y*ySlf(A%qCMaleGLgc@$Bb
zo?^Y9@bI%MtlBpJ{&bK0wT|mw_z>DAQyO(~jg2|M9k(6i;-xJfIewADyJwi0tYaBL
z;7;Pye6%G&>1rm((iBmO(HhGUv}f8hCYyv|NRp=HuE$f)ZqxQla$Q#QDwc!q795zm
zinsHJEN|~&XX6OenoKz^`bwW*txI(7^E8S_IWSi;(eW9@1Ehp@v(Dn&0<QE(l9V_J
zF@;1bLpO|g`oSldcPysclO$<MQf8#X1X&i;0!cSd83dDDe)KH4uqeC+WjZ3Gnj?I%
zu%vj-cT~SL9~2=V@Lg)P0Nb*`fKrMgFUZRhrPbKs!D4$mq}JHx@#7Ek@MB-#Yv2A}
z?!4^|TJ09SZvT~7?kcpP)CL)*tXy2-ZLhz9T0J0r;uP}}K6A5egtRdF%V2vWfY_kM
zEJ$rbZL)>mX;79WzOnh^i&tp5kgE(zRn<@i7CH%a`x&lw&(r(FQzWK_5jN5YW{W<7
zxyoE?hq-+<j4(u5jPH4P_9T;?DO6b!r(MD@qP&LfJ0ngWJHs9O_tI)jk|Zffk&=fa
zL|I|`wkp{`v2}He(ssbBk%xV%lIg{<z13)7j2RQxUOKin8+Gc9I*xS>+Y9nMC&^Qa
zqNuPPh{BBRtxCz@V;}zr-~X3>kOTYoGc`5I>gxI{vfM8eLR7`4R)SJPwliYw@)~ct
zWe<&Zi}<l;*frB&ZoUIz0+d49HcEg(A)JaNA)*+iN=l{Cj-)v=Nlmp;szkR0Tj#IP
zl!h!%kz)@w$H1PcA<g0ATwXa%@A3k%_7KuVD8pjerOqXGP4t-E-K5YZNs;3FKCOD4
ziS{IUmJy}{hGB%Z+mw0E+T|_Ip1#E0M{b~2YZJ#YaUPS82H2_~sQIiL&DA)jx7MR@
z9gGu@gxUB4;)V0U5`uQCg;t6@&&i6inv?tr;n<Q^t4Y0HA79{z>d$8xNmh}1s=Xct
z{h00T9{EdTy}zuYuDpJnv{>EfljkMAt2nx=hSDRhEayxw%riARN2wIHWl>^4R8j}F
z>tU3_7>Gtg*2hi=S}W4D;OMl?aG*%Df+!xaIMpVKq0oj#tKevJNSgKNMJZ_kQbS|P
zP;a<M)1uS`r737OTj<dUr`}|}7qYgt!&a{#4mAWWOG!c&ZLv60XJzXu{k1LTvy?Q8
z(aDf3%W!PN6<jV|Stm?OXibo7#YmSZQC*RKxuq6DFg-QRul(}A2jJ76`ZOQ?_(wT&
z{@mD?^2>)1f?7~#a$<_<nHd5<z!o-f91}(%o7)?RRFI}A+GwP481!OlfyXEB{|Mjo
z-fw1leu1T>U9V_Sr;H$1pwLKRs5vgaZ{bOolUE0L*v!^l`u!L?2r(u@`z?IOWhcqG
z|EVinP?8UvyiE3L_BCFMS1eT3JjF7SmE|>Fe`uD4xf(~eB!M+Vs~XMeDTJ`di?Z@2
zb{#A!@#=NlT7W954L(g1qTvAPyR_;)R#r0Il*qDTDdu@Xd$LZNDO^vpe|E@hev+-h
zNqXxgNe)sOW`dZhj>UAdfw3S?hPa-KuYFpL2Euk&+u34!&}BF%NrcT(K4Q4jC2$Pw
z=^EWAr;(<_S%ivum^{U{3@b)*aeIg1Aj0?!3JI|)$AJFA@2u~7eAjn;CvlQ+;=~Cq
zU%f(@#9y&R=DH4@&IA(^6S%I6Eo_n`A<Htl{Vr*iQjUu|p)7W`hq$iG2mj!|@`k(K
z!0hY{iwg^{Xi;|!+l{eU-yBrf?r07#)=)a)>eVYu&CN0~Gm9$6*e<U98H4S*7*!oO
z<8Z|0`WlRHqOvUK@TARXSYdk{51F565hoS4*P8{0>m#y!KtImNii&C7m^9RCE|Mmt
zDJWGzqtRp(f>Up>*$Y|U=&{`|NFxn_!(y6}N88L#*I3(LVX(f#T$YjM2_^~2vmDzt
zT*cw?$_7zZLbF4u6(g;_BDPOWPVp~)@n3*}&wlo^-2dtOId}2GSH$+fuhHpDGCeho
z@B3K7B1saWC}O+2MIx&8H(Fz)&2W$q1Rf85=99elzV|RSH_zhY;;UcOA+Us`K!Xq{
zEm>Xfp^h%m@H98=u0hu4>gAZ(T}!kkC(%kFEekcirZUFhxL##xXAFa0kDbj8z@U{z
zB{@eXZDgiNvz#~{GPAHq6bVYBX?F^4ZuLph9r{U1mV;#s&8DDMa}lORp)xd@Mzcw&
zH9@V$mCZhD>s_{cB}uICryUm4m?GL@akjzvD;JpE?lY04q-jDC4=9R)q=YTuaq-d$
zX{oU$n<Q~eIL-(8BKfd)?>>I@mw$!roo(*F|9(F9iI1~<<?@T#`3WJY*BeYuPFMNh
zRrwG`A*0cVt?niykmottXtcKJ_d*=UdP!^e@`^eDAqll49F}bNBIepYyLRs-OJbz7
zDYKkBFA=UqoMvDQmThC%7Si{TLQv#6YG`nzMV@EGVMH8d+!(-xp~ucp;n~|vPE8O5
z69_W0EJJH>9l=b;#~=}=L>Pls5}X{@wuy%&E9*OKbrVL3A&g3NA*i{Mxhb2yduOTH
zij7gqm5Vzx1IfN!d&!Fe3yBh%v=HP{vR>MZu5^h7NUbU2h#Z3%U$cI>aBQ2e{o1eN
zYv21`T-W8j-}H^V=acW@JKz6)M&S!fci5KA%=8S4i@TVgo2OQ*VM~WNj#*t@y?#rl
zl%_08gpedj!C*MTkt;m>$mjXSZ~k`n?%Bh`5500^iNXjHtw9PrQ5EQpkc3ge^2jjd
z3xqHfxrJv70^7$V38!~rF7&o|cqiqPy%E_f*eZH0Ufn27B^a6-i`9O_D9Wk14u|&d
zCeKpBVM3H#WzV4_xXnsEz9`E|qfUe6`MAD78$()_M5B<hEQ#ZUJT19%!s7C}WE3m9
z-7O|(Ch<I<Bq`8VhUaPa%m@}{1PY1N220j40tyvlSq?$aAkGR#p&`u+o;-P(UN0lh
zTs#Zrmzq!t%49?~N;tN*#*w3M#mzHxS%NBvbA{y?HjU5L#fyX*r0q~-36as2iv(Z(
zoNh_U9e3Wz&;G(M03Jc%zR+kkxq9^q|K)dnhyUm2ewHLjUhvsUDVgX@GB-QV+}u3%
zS{++fhqtY*EtF9(ggA~XmkuL|!kp1?#IvWL;pC~~+;z_z*|TR4Cr-R_zRwCULK0~~
zG%DHY$IP}p7IrO-vE4ys1w~#W9E&8Yuw7omcF!AQdx0`w3yY%6h{Kp9&N&porIAZ-
zr10z=CMG-hK^s9%mgi^#t^-qTufld+RwNuP5o2smh9zsuU3U5@qeK%$CAyRZu4HD?
zW@+~{fu-0CGgg*+G<?aP#ogp(h3$pVWF_QMvRT@UR{A7kY|oRgjP15%^Y(YVgSX%J
zE?n2;zW00`?|SfE{FNX4LBbcsc8lq$X%^-en4O)a7SzUTp0K{YPEi(B+f8YdQV1bQ
z)12XGgd<mZ?1@Ks_j|vIrKKevd+gOEO9-l*uV%(2p^=o`g#IX{9n@Lcy_Y=Ak(MNj
z6Uw&6NJ*BaU@CuH%eJv<0XBj>P0_|+3t8Qwg%N3-aa+UUQtZ$jmN@nXlhc#<L5(cQ
z$nqRx!1rL*3(yFpE-IZ&Es$=GBOIbZ%H^xuZ1+<}$$&UiXeFro7Q1E~_8*$Tp&%Qk
zTspVKM7u#}ZjnNjSVCZoB2|JyTdXG#uJuWzMCJuqmQ)?N7h%79uFJQ5`*(2fJKl!0
zEWY{MzLnqmoe%KCKl-C&`3twnJ=bM+dX8O-yO^1sp%&Dzgw1F)Vs&+eqRcDDD{atP
zA&^8-hBq3$q&0k5MV-=>eOIIvZf0?HZIw<ikK@>Q+6P<W+jaWeJ4j0s)a!V)0LS+b
zc6DVh3|OTmsMV2kv*c+^y%7+OLgHS?bX}0dg2t`}QGb9_?+|qA_@+jhB;;v^9D68Q
zc4f4fDu=6UJuY3|q(4k43Ij&qIfj|KU}2`t;#>{C=~5<@7-G58N86g4?mR+cVVX^0
zXxTMFT~g+T%x|(44H-q_O~kc{V~rWlTNrG{WiCKvpC&b(MuTtrruX6dE<y+_X^|x%
z;)RfcDsO6@N4wEterBG<g+*Ge7Fua`c6LaTm?Q}qjfMzO5fGFrKq>}<5YK_ne(nM8
z`?~kAYjN?57&U*^NVFz5hCm1`p((UN8;6x%L0Py=PvqoEVqxJ}3VXaUeBn}`PxnJk
zXJzG?`C9xh6Qz-8EFxvFLcz+$7KfLn2%1eC*Tu3O9N#7C4^c)Vg}@7Z+`vaVHrQ4L
zz)^w}lF6wl@+`q~ZAOC-OJ~HM&(1(FRaXqRw`ecy#d2-3Op+uiMVVrhKp2UXB^Zh2
z1Ss2Lb+gCH>X2?<5hqof$g>pfn#Hc&g1N;Bnn45G$;fwxZ1-dCJbDAW4j*7!D{3;J
zCkn~}Qm4jdni6$KXv@O21VxcmKGi~F%oopBNh(p^n(y-+@B20;XQq%saPZJU-ul)z
z^9#Rd^TH(^_^!*$#5B8i?O}dyo_eDW1_py+b+ehpBuPr{l$4d6Gi6!O>kn{j&8Hu{
zpIh&^i``30uk4y@NogeU7~5Aj)@cW`IJS-J`IWST9WdDGA}x!c7U0$b9MAn?Y*(5f
zs3B)($kUiwJs=uI#Dj>bnjlF8jfFbVU`V^x#&6f~OpPo}D6)JkA7QXv3nMK`Wpj07
zz~!sk42LO2`5d-S)dllY9`my`{DwoBYBU<v8=!5?k=qVYpPOV`7@Bs#rYR_L$h`*J
z(TFHckd{N>TA1XyS-Ka(_I9JrH@^Gp@w^J#rLf4-s1ntFfe=98dbH|I=BDSEpPQ%I
zY@(H>+wGF32}u$$8jVOrjzObTi4Y~jVTkL%Ll1wRx4+{)7U$<*QEdP6Q3)@X#wOPg
zDUFC>Wn-1v(kzzi;u?^igFw(--^6wtf<_Im7F0TwwuKP_X$)3gQg5`dEm`gL&46$)
zBI+mf13{b$ntqc}w@YiWLCtUA>l$g2Qlu$TX)FwuXJagzJh!;8vcu&o1BPKnQB>M_
zfn%6!3wF&kn4hoX1{S5t(HP2Z536Q#^Xre`b{eEoQnzdLR6$V~BD=$Oze}8C*lr!y
z7UNm+Dj)O<<iqU5B=7r%Z^E{%Dj!^%I388r(=U<_HP5Hh>M%b$&%(k2^?Dtp6jBP3
zBqmNmhQkp;6qQkfDZrEr`Y(-%|FVj@#28|wD2%4&2*NDl%Ek(Z_a7q5%c>y{G+RyV
zMh(kzuq+!X1wu-s0j(58krO2;S)8NFs!;Ol4J=`?1dO7bX1h(RRU?Zcf=&l%Thux&
zT3S%*lI}3&%F2l4wGB2lhoqSxumnv{(4Le`v>c{dH5#o3uIEy!l8u#hhJ%O$VNA2*
zvS&|+W@i%9XyeMt3Bxu%-PA_Uv{~Qk6O{(zSmarX7J|vidE#<JIEucgf;WIGE%xo*
z%iV9h=Q_JdX~nT)$0)wSfNQ;8r_*lJZnx>QJ9Ijoar>(BP6Y5gAEF4x<O!4^O*Ern
z&iV5f7!3MMPtClXCH;z#Vn7(`Qeqi}0LoIZvjfw(%m+o~7pN?fB&ShRT<L2*x*2jo
z7q5wy{4W@}fmCTSrKwB7#?B@aHO0d00%q)knx#2PYl2n-uU^A;9i(L;tg%j|HW=H+
z4{B(w&_dt~iF6ziCuF{m#JOT}Y6i=4$kPP3*`QwYsdZY29JJ0UOi7w)Hv2;^pWWf|
z)e&hdu%yBFN~YVIsfNqsWQ*ykCV}q}MkCg)uCox8II3iJ)@N$(0;1VMN{R0Tpe({v
zQnZ?E_6CG`iLiXo6}U?2BKgw%H3)183v+Y4>s|L@jhSmkYo2}f6lM9s@3{c#L5*gk
zNxRi%YHEshyM295b};DUI&L*ji>zD?v;wJ#BTeY196R<D;b_3z+}ta=!Dkp^U157*
z38E}!Wpj-i_8veLB}P`Qxkj^zT@S|CZdKSWWObo2o(qhVlq`7;+r4@XOIR#Y`K~rv
zZCXvAJdW_&ZKN#;+D)2cZ10USuC9h$+1O-jdqkRnZ%G;+v?eV&4V#H(jYhMM>pPT6
zv9-3za2T;KjH$O=mX_Ky+Z{~3jUy^-w~S9ewNTS7Hh22OrNP)^|5zcJn3y6dL&EUO
zVY_7S?%mvV_v@aA?V1xOPJG2#r+U3ktJR{}s@A>LZjB50499g40zA(njH9ube>F>$
zrFv`?d5Ph0z~toAE3(!v76wR-Asz4a$`~wL(hK{nZE5!J-cOO`)$ART_CyD(9$>kZ
zPNfvj+w04cJRk4%NscZJ8k?Zi#1b~Uji68^&4~$uz@dm^yk-;2aj5$)wrZ2>f}MWK
z^2(6QYpZN-jmQg0%@TA1=*(D5G+idH@AVE@p0j>sjVRF^$`$Pi7kjBgV`7@3*2EDK
z$9EChXFIVFUW>KMS4oV(AS-=UDVd*Jq&wUs9cybo{~7tV&C#Prx#`whuIEFM7aTu!
ztP(YTfe=E_Y&L0M+v__WTCEmYmJvk}j#CxOzV8#p&)e&PBz|e#u9sERWyX*ggOmcx
zkrb2+lM&s1A6HnomV@7JA}kAG%W4Q1DASy1G$M@?3>Cw}!A2#Ta1dft0iv=V@@gP#
zffM)$%RxsG%8F4Ev}LgsrW`-l<?7`wPM%VCE#}*TM$<4^w`ez-1hodX<032zRVqeN
zM7OtrNHueF9cpDlzW)%7JqPd+481yKno)WI84l+X%~Yeq_JymYNJ=fxaf0+5-to4t
z<;1g(bNc+V<oR=pX8>DBZoBnXX6NUs*#ZnnlJLxlXU2@cFCUh)aNWvwD+mH=wHiSX
z5XUjKS`FX#aa|X!)OeWIAOvVFFj^3WC7YX@^m<*|?anI-_Ng}5Mp6@ic2P<*D08w*
zFzqRtzM?D*x|C!|Wh3;FjjLRIt)Tl~HWFiqltBxF?O13_v)0|D-JZZ0jpKL(^*WYq
zV;l?1b&ygb5tveuCMiiAGaL>vg+?kUGmS2)jiAx2BZXlwOb8kc>WzRTig9cgWlMzX
zP*|F^e#*0_H`u;9AWaK|RZweKEKQV5HasReQ`BoUJkQ6nZL*?ZtGmTuql@jj*hNfz
z_cYVPqXd%+6tyOnHfS4+2)LR`{Q3lI=g$)xfku+Y2}W4F{cFCK<(0FXJ%8ql+@h`>
z-i{u*ksA&jy#Ai&MFr?Z@xrHRAp}7XP;ZPk`DXJvpu4V15Ck}mgJav33At944Y&eb
zLO3j0TU)0;=rhrod}WI|V@OR!1h8x<C>f;%{lNfdjO|{lfsog*y_)YS)0{YrNR#9`
zQ{NU9wnxJVqYI!!Sl|T$VO7G#SPnXl(N=<zpe)Hwm~rZ2kCiKZoFb*>TFf>D^+lMd
z30jQ~fnUeEj_nm`tJmK`W}4ZVHi1gX_ufEbX)m4-jJz5u%P4)H9ES_3X0qO5=h7-E
zvSOYlDbjVg_bu<>*)vaZ?&2Bp;<*j<8n)kj(@o6G&R$>hG);K+)TtMpr<2mcaU49)
zBk%))T0js4BymD52=F~0$8pB%J*L5mij-)yAc;!0x3}r{d#^aQf4R6;)Frwq>})A;
z97%3+deMN%(GY>bv2FZj6Jg29LCUhgK#^ud!^&dQ7=tmU@^}svaX6~>dRZc*pjHQM
zB$n%8ISy$U;Ru^jfU*SZgM?#ec35BO<Cj_0R-LN$`l*^tyE%dH*RUK9VM)rOq(2(b
z+u1-WSe%<A&>8i^w=#3f9XLn^R%IPm*fx=IxtI&4yaqeHm>i3u)YvLVdM@Aap7-(5
zPy7#7H&*zQueB>J?z#JJTJ6@Dxo#K?2An>98uKFabe2`^_1>S@>y=X1e{9*-m=&$L
z<|J#3peiB0^4h=bn>vNTm7=l(Q<^wS+30StFtdoGK^IkPP=zJKosfQ*;J6kO(-Sl%
zJ6OihTi@c!@?{#$38ofj5xSb0(6&IjE?^-PBw0k7M?}4XFbO$zIpWz34W(x9lw{ux
zO(xq7Y+HfIXf<bWojTI8Nwb7OZ?hs6C`F^uB<R@G8ZDf`fY~>_ouxOvh3(7ddFr#D
z!n46|waJ9V6$|PDl+t7d>@p`bB_h#$=p+A~tVqy*(%f5Uq;R<7j=Qj32SYV5y0Nj&
z>GRKy=ajxQgb+Bcjq5ttu8pv(L7u$Kv27dIb;nFNP{!c7E&@oi6f?#Etqp_Gkj;&C
zJj=qh?P@OJ<t*+CL}@fl%fSKiQWE6_WnLhS&vY%raTK<th%L>qulVib>wI`4d6h&U
zUW>0rWZF;wjucqJP!t)dZP@B=(P?&&2C!{P8ysQKVM;$v84ZVEG(p{`S+6tEY_Ykr
z$<9U>OKKd;;MfvlgWq<kPc|@)MUrO}d4e7Y3YD=j$T)r_;__NT-7+jqYWDB)nVe`K
zt&~=Mo*-x;EgOX)O(T*)k0gmH%AD!x4nebl7lq8-^j2>F?!QjFdYO|~ub_$$w^k!I
zl5?eEzaNmKC7BVRHJQpmTCA<Fvc0{HWr;7g02Ky_<gPpKqSmNgf6x29E+?No`GOvi
zmAbxN6>W}#?bz6kJwBjSr}t}mRasFWWo5^tRDm&-o?VtHdi_4z+ncl+&C0Fk74oD_
zj6q{?rG<k*YDJReZ1%R8o1Vu}V{8`%DpHKPA;T!ac5OOSZR(vimN5)Aw^_Mzm3p(o
z<ia$l0@tz8mO#1=*RVa!W3oIZDhlE>;`G&svzrD}YW7S@_8e$1(X3<H3PeG(F@@{Y
zu4DVKzeSd%Xr-yw8w72epx(q844Jv-UKU^f26nDq<oH9M#k0U`w#bFWiY2H$58Ex3
z6PXf`YCiIb50MvXHP`q>i>DDd+<NQnSdM*dn`LuzgR_^;qFw;)gb+B+7~35SVF?hB
z7de(?;kZskNfOnq-?ffRmSxu^PL!&!eQRq2SISou+rL~G1G$zYMx!w}Qeqp8C>2qf
zv)S8ZdSVu%Koupr)RbvKe`mxnPVobmiRlSiQyoO9*;!p@^~welQ`1Z>&LDM(<Jg$-
z;SFOUlp#qY;w&U83kIVR&n$<W+XCAt4$Ru@nQt-CZh*+JZB4T=i(~s(wnG>V8FjbF
z(iE*V?al<AC26!Lkvm=X-23g!-+mico_&T(4}XSstBF(FPlRO6v#G0sLK{j0LS@9d
zL?xOJ{Qj?#WYL$rV@7B!>2T*=uR~bln|fndzI>5OS1-N*tpcPJxUPfa*w}8B4@MZW
zBEueIgE3}IP|^6VM_vfhteRKUFLBXdR#BI0g^i>p1g-?nbr7<oA4jy30YwsWWqFm0
zTM7MP%<jn=x7>CEoyk@ug|W8D_WA~vlpH*Gfac5;_)f)?E6VD$GG@oRwnY{R;yB{m
zxiwDh2zEkA+g0q}r`f%$&SZOmX1z`r4RG8RS_{%VqtWol^OQImP_MP|{U(9u;Yfq!
z*HNPZb>G5veT>uK>iKg_G<^0RyanO7P^%G#5wcK}x?+7AVK7<~C&L$fJ{rN|u3acq
zV%s)S7@mFhBr7ZHFKQiEfq69)S(YV9nh?|ij4_mDSuOVX_m)LTSys=;!1(6V5XLE|
z&z{2)9*1r?$mw%u$%{X`BAmfcw=~)aGG#M}GDK#m%aTq#2Vvp*0>5qe)VAh>gAv7R
zS9JeNM{0D{0+13<Xad_M%SWvDw(tWV3m9(Z48xL*?E$0Fh^1XKEbd;Q)~r!x3EhoN
zy6c<N{5rS2{x+<dgUYjNCJoTSP?iR%4SCcj8Km?!2VCjroLVVKOIYX>+_<;N{OlB+
zW{bgK2g~+wf*M#Bw&M{Gd-S%~k-{SI0;VTtaD>5cv=M_oPQ6BBemA2e=iDbg%0x3@
zYGF6nE~CIDO%p_^$hAT0j6^7`(s2CLV=n;B1|cnOy7^|1mF6Y_E-hc+;+2cfn;-qs
z32WQ2upJ9&2{f9#%<+wnloF*>b=x5&mSv3#Iz?V^eN)j|(;tmkSy`n_6@lwrH%|Yv
zMqxA-l0bqZt6^>-6oVw9nT{w-#LCq*mbWtoqnKS20Y`7%PrK7Z8AZ6U#m?pymXI7c
zu#d*nB=|O}P^hx3gy)q41J3iX{o;iU&U6LcNYZk1_U_f}TBtM8nxkH;5l2HDS5(+u
z<TPB5BF{+TA+?}|?==WK4_g{6zlI7&)O-ue@i0z<)r%KsH++^397Wg;1U^X^V-+Pz
z>#B7o3|ecF^o0k}jexoNg|S<TjR4M`J;U1C#*1RRWmS2gwWd@hNtO}>mE)2s6^`w2
zO*W?-GsY`ie-7KDB;(xq3rOj4VBdbuUA*w3gZO{85NIPPw89aBK!Pj5_dS$O>4qa3
zwSX)d(LbMaVKX6)686nCx%t+EG+K2^mD9VtN_T4;*KxV=rbE=GCcw2(d4Vd`nDkeT
z^EgtHj10r!khAC3cy>q7Pb{W`jQzU|ON$LATJr?HOB|1ITtTTdy2xqxK4}t>CSkSL
zdkq5L!xjds)<6yW1a$|?@yYSHdhQHjX%`E-4}$F?8+F3L2umqSEieMKsNS_W9=z~V
zvMlE2=BmBkvOpM~KK2w_+uau_zQlN3quT3>vZ(g@AgJ7|l&UoA;QHQQ-2#iT9>Gg{
z&tFzir;Wjv7Ih)8g`zZumK)&ME<3}7bL$b0T^QnN_=bB9aO=To9M>a_LN?ad@GOTt
z`}S4F->!r3915jK;~14yj29^del5TVK`AXvtwGjI=w9Ar&$P|HxjM~GK%?0p2pXhC
zL9ID~lmex45CUUM9M3~*NxL;k;5RC5K&dG#iy}^#@NBfI)Y~teJH@8&a_8H>o>EAp
z=hN>Faf+(VnPYH_AbGL3*FaKa{D1$+Kjt%^`z$xze3ZTWcJt9c{v-Om5ig9prcz^9
ziYTfEW;>k@Nt_^&q-jbNMdW#YeWCRj$ghEUp$th@@}GX=w@9L#xrJ&Ncc<HZIm`P3
zA@L*>#*ksLHBd}Bl6sI*YY4Q_WXkY`RP(_>@mfLmzd<OYafL%&fFmmnsk+}lU@;6w
ztb|>jIXB=^H{s~+Dc*k5E;@6QxUOJxb(v_hkLP;q*}Ioovw>ar(2iZr3FSqVzoo+Q
z930O^n+8@RWBKA$p1z#X3M`Jy!s3EYd%~kt@1TM9#0;MA5GN72QUnbbX-SkSnVg!T
z7PPS~3n>)ZcF4n!xu##e=f-B`!a4fB!_m9$MN11Os4?u1unUb5hEfPz1i5~pvo>HL
zD+@mV&_m4bT4e9uy-ZBDc=VBn+1l(qZ-f8hPY~nkPbo!`Cd6?}nx@ohHBe9#1w~O@
z2l6ycuebHCwcJVrS+4ltAASf`3f}R~ck-FfeU@&o_lm5!Cv55nY*Be;HUl5qa_J2d
zE^Nj;d1-`a3}1KGK5p7Sh3&eeam4oKI<9nB+Or$SyN2yHg;Hd3jLHj)F-R%#0{=N|
z4;o~RjNa8PmZog>%my^tReL({>tsbi(CAdyu5&<OEP?HL7%gZuJNSO1dPdi<JxS=e
zm6)qmaQVU+YQD?uZ~5Bm*goiuunRyyfx$L{^ebX}&d>ki&+~;x9^%Fu53_ssF7E%t
z#~Ad(7sYn1O@-}oL>Ps%+ij9GL7FkP#}QecJ#Q1&Uxe+YHY9n;Z~xBkktR7ab8|H6
z4YK@|!1gZ<K~+L141us|2!o`gG*I_yxVFoBH{{uskSCW%OxPCR{N_X4uzLd6wTXs(
zwl>$PyEXRRZ~)tLaa<4K*yLqF8bz3*D(WoTuJQpxY1z<h6Z#3mw8z0Yi#>Bq8f}ky
zqmA#^$jXd*YZ4)#R2foK1RdA+F+D-2J&EVlFvcLHA(xUe3~>V&Q&!en=bn8ARixZ_
z_uD9hKsp|SogrQ+P{L4RunlBiAs+^D$lv{YKf_z!`c{tKbc9_?i+tjvA0rIY7wM3W
zjReY)FpLPpkWQyV9LJ!!zSm#VEc|l$@KWOC6?F(|mL!k{qpK@yv__PgW6J|RdpTy(
zwt2_V4mV6yqJbbOOT*IAUhJBOGP*j6+8)s;X4u`J2uFB!H6R_-o0P`T>-C9~1g$mu
z7g~h95u<vID^v3{=XTR{Y-(AT1;?jd^AWa(MM<8g<avP*HojfMt2e6ZTUrdG5M=^H
z&BLm9h<XDkbD|`s)MeEowH>5wlZ_&5BPwA~%vaKsuAYxE96#|47cXAK_g$nV2*Ze?
zc<~lVQIw=<LKuz+!x5uVh?Ex6K$66iWq~o8ax6$#F=C$6h%$ys+ezBY%}p@T>F^>*
z^RmKbtjcW!h0$cgN-t=-EwO8cFe`XqOY{4y`n7`Yf0NM0;Mo>+VGyRIP=>m;2^@z@
zYa>2;e#lm+dGlg}uQ@VDd%le&Y=*rdE)G*i4<pBI3CFLY5JdexX}<@h!SQ{9W*zBB
z;xuItMq~(<_RX-~TV`g?!<%fd?KM$O9orHFLgEKL2uZ8$p;Sp$<k+r{qym1+wlP9d
z0GXh=5wIOpnj^GgZS4xRz^5<<WmyQvWz>(b%j(80GYZRyujomftCC;&55LNXKm0$L
znVn{0vd!}HGD-Zxw<p&&?IO>K#(<utDZ}9qTi9e-CEIkZs58b~Cs15_*OfAaNzSq3
zC&{xCVTl)&PkA|EjG$&qd<jO0Y6e(KB${U~4|#Ycp<~&+^>CYg6COf3fJBv=#l_uN
zH4mk)VY^ElCJcL9l+g&+_Nld-`1Km4F%0?x;&H2N?|g%(KO(FLtW3<&nAt_cwdiJj
z=4_u<%|lo&mQWOVTFGSz3-@{0ZiK}s9%FmpB5Q5p{t#47oW_*q%V0Y%AKOnpdy3`d
zWjx<SN<kEVWo$3Yk~B++qL47GNElLD2m@)FP?lxoB%sPMu74@E=VNT2nd#7Jx2u-@
zE3)ceK9sHqL3OLTsW+-3I7(xqbH!s920XHwvE<vl^XMeIS`I=wXd^I2uy^kPMBt*e
zt~yrIWi%Ku>TaT<5YMYq@3iph9)&9D_68(rhQV;-o(^dkkx#T(o?4_evx|le^|a5T
zTPLWwV0l<nLcMujVA&p?UB|80tCqEp459&A2rMVS3R(=ewsE8(j6(WRSiR%615#3?
z1r8Ez%InYeD-AW2O7ZC9k8$eMDLmgrNEi-ByvXSMSg^X<>qny@VHh$Rh1E5%HY906
zSr%X_2DH}ddOp0IS<08^{VHtgQiBnw(gTwrrOj$D<LQl(LQ4*}Bui5kUQp50wWnxK
zw2@=^5yz@5QsOuw9(Hk*rrv22v?j5ApInvn`#o$anP{~sR6&tv>^ZbV{-*cx&hP&T
z>XVZgl@qRCX8E)Cv+~qunMosLQJ^h>RvOE(@O=-<wUMsN`HPo0*V|(L_BXN?YxcPf
zI+GoQDDnIjagpI#Hua!^@40l>H}El(R9Ex{^OAbnrPlPr5xlT$(HDg=)jUNMg^a=h
z+uK{z>rIRXL3MkV7a4h)kY_2SE-F^s7k>)1ppFn0-}nvRz~aI@d-m?;KmPi!zvMpr
z3q?a#X`UM>Nin8|E-QVo!V^3+%K3wp*9yA-ox<SIk{X+9?|F_8l0j7P=!J;oLCNbo
z4tMQ$2?B>a&Jcm5Inl=PT#T$n3T(?J&vN=3TV%r?wuD-9l6t2NuFY^bBpilVmWAiK
zOf?NR?4Rc1^nPymmhWcq&|##?=v_ET_xKYyWsksip~%syD%y;Ix?jU-)DgDDU^wK$
z+GUzc2a)Yr_P6FSMi4Xtq!W<n>Ndi4{i+R^R(6B>8gQa{DZ&9!nsDyYB`#iB#+b^G
z{l#8^YCcL4MIpnmPp{vl*=!+%g;5nDAum!?nWI%Dw5!G}$m@<3#^AahzF*_(?z@+p
zZ#v5V_|S*=^n(xnnP;rT^RZo3*e;~aS}*6sR)y^cJA%b23oi)B(iF$jG&-&4VY^j{
z4<|A4s8?ZoyM^DJ!1gL^9}N0fQqpO)D9VB&&sn-*k>c+Ac<XonFtv#ZjLL~OuX5#~
z&#-pvA*QkzSr!;swZ$yQ!uLF+V_~^27niSap})=CEw5)iG3<70bUG~%C9c;ZDKi|)
zrXJMsT$g^liHAWw58GdgE_$gogD8aPr9IPX%{6fp5r#v$-EHc%%4tZL%1*0z9=2D7
z-Iro};Mb8-@}95zTIS~GSlYdd-}>zjyzDi8sZkh#BP{Bc!f1gmJp?&I*j(C7IKHKj
z2ySc(7N%@mHy}+S{Gdi_vICYxO8ff2e-wr!!yX<6cWRQLIf>=Dq(x4@Kg4q!0?#Ka
za{|X^*Wo?1zVR>f=5PCc{8kHH#0*z1vi#|fv2yAWW|Nq#%s~pYHrTF1;JXOd#<Fal
zJ$;7tGG+Rvd)SIKhn*VD&Ln9bVtXy(+~7GLjd}yik__VrM`#oVDFn*smoyVzC`E5H
zgqJ2CSLH*g2*VMzQ8gP|t2HVd(<N~nk{20Sa&50KpJyNVD&5r8gvOPUTtT62($XT%
z4Hs8)28GL<E7;X>D5;T_hFajGvYaG}aeWWla>-QHVz6vzPc*U0CepFM56JU^EX!!O
zTDYE9wIuQ!?YPXib%udMFwsN=KEgP(_Z{Y7Yo4{;H*(>Fzr=pq;5+}n_Rc&;vg^9f
z-+g=SRb9Q$($hUX-Lr9qo4ARhOe$hbQMP0mFf7=Hjo1MUCs7<2kdXz5VHkmzAa)RE
zfn#JtU_+2CgBEFVF&A+kQZvILhqF(w(|dRA_4apf{&-Dli9`)O9GW8Q1Kyui{ZQxj
z@1FPWIlp7!Bo5tn2Q6bk>@i>O@W{z!29NCH&6B5DTW)a2&;BB2DZ|MZUSP-AUS2!>
z8ir|7E|(~j%Ph^zqMI5B6B)<v?uUApD5(g-5VzZ<)o!r4wN4yIScX9qMRYnHT-Qy1
zH!Yn-Ddo3oZG-^RGT66oFWIca%<K%4lapAMg@28+0&PtaCjwn+#45nRMpuHhmc#LS
zi&vw#m(YF4um#v!YCsw#I#HsNU~qY*!-6L$=@z4TjY6S>W;&n*e%HmeZ2%0-LLu<|
z2q_bC#VqBVgQg3NToI`huInNc3=I#ZwPRsO5<84#N=$z0r@7^4ei2L?R19~1oaU89
zE<W~knkSxMzy!mxF-()d_t9(v6?^!J#q%dG;#KnKuH^LNPc!-Rzs68br*oyj(8w+>
zFU;cD4y96=T%o|`!WFVY5vLE!=<i@Q`n^ac`QCo1cOpq-dc@Lcv$eHBHj_<jW_67)
zOyAVG-EMjuBY4{?;XB{fWiuIe?A*cr=_y>-<(^O6&GRq5KotM5cTMk1R>;&0K1n_6
z{8+;a74vH$&Ddhlj2S5##41COC^EJ~5{CG0s_nXN5%>{F9HSXfEalM>&`bki+eBeZ
z7zPxJ1uV-ZiX)6L!Zt1Xtt{<~Nv2#tb8Jv1#od#P7lv5dv5)yj{s+4aiDOvUQsH(w
zXc-e?pT&mD<FBvKe`qh~UO&smYLi>P@I}mGhS!fBV|4c(-Z=9HhG9@F7Ri^2tjxz~
zh6chw#t8=R=rMauD5VI)5ZCL{Znvn_H@-{TyWK9Hm%g#nbPbh!yM7Yh(srGRiE%O+
zhxxgACdS9HZ3{mL-uwGSA!STX)6s<>1d>F<mpZ~rm|KhKCKh|_gpodrAhrpW!q#je
z&&M}(Oxp%s5V`?UCg?g;swG6M&}|E4*@Qtz6oyp#`Y=tCFbpvw6KUCuI7O7L<5Y?W
z%f@gFDwBt>OFLNDHO1^Bf692y#?%ZPsc1Kw2**G~KIfKQ9y`6s*lmY+^~G24eYp3R
zXQ>eI+RHDpW6xgJY8zOlNwwNfCZA(X#%Pw2c9_Z}EtR-Vdzw@-{Xh4*G+T8xx7LV4
zA4As(f?#``qL<d!={oWEtcR=N-6eH8Xi5^H5lWr5Cy_~j=jxOai{V^EA*XZp+zJP8
z5M-P|Y|}zTG2I}5I6}y6xB6}y8AX%_hsX~N5(Vi<3@ew}_GwB5Rw)P)g{^$NvnM$7
z!V~0%MyWR%?3tLNl&er2-$(VnpJwsV|I02Sv^YT%8dffYA4{G*y~Jgw#K7c!hU)9I
z=APq^|JkoGv~Ln;aF~6!+|DZ}PvCnV`FxIn!G131GFa%?vBax2Fg5K5j*odSNa6&9
z!m<pa$Vb!D_Gi!Yh?4}%wlGWsKMJ><roVkBigaF|uEV9v7wL9f{3z|7<r<SZmC*^p
z)R0m`Cnpk=5}b&Ajzz%_J+|oaqeos-=tAO4jYyhw!wAv#*sOyQJB((0Myd`ME^o4T
zpT$sL9|D7JyPf*gAxNS)9p^10NJ6A;kss(Ml1Xa&VdxZcWi(AgL~&a4Dl{@OVE*_^
zjNkn!3PU6KzDF)oBvTn>=Y5~!Ow>ZZc$|JiL&OOxQCRs5NfPti`8A$elN>!b&FbtK
zg2h$-@{j(219#lc!om`jfe||A&JiaurDBnCUxgV%L$@sKMA2z?u`~_wPBnGcNLn`+
zkt7lAPBZO<(~P$TT`84ra!HKbeo}ucp%VgxrpV<oG@DIUS63OE7{|74qS(Er`}$68
zkCaZ=SIESDTiXW;G5M^{g-h#9TN<Nw|F*U#bi?+xwx>r|UblmaV~YI)<Oc_~r%Xu9
zYzEUXwntHck`N{mTls99KgIcDPmvuQq0wxzdu)<ou0&yMFO_?Kf~CLxDm#b}u|z}1
z%%;iZH_t9J>lCR@>|?M|p}qJLfBNO$U|{bA_P`)}kKW4h*Iy?Hd~*35{R36zvfJ7o
zNxVk$x@-GezGbFm5Jdr+E>dmxQ*Ae`RNI3n+CFyw9&Hz(=`b@pgX?<uafqg0pVtgR
zPc4eFQo>v7Au;H90iDPr^lU0pXDHXjG6ip*Tjr*m$v~-sZP~;@gzx!?C`4>`^Ld@L
zE52`dluSNP?1xAsb|HtMr%fk1CMfXZglyumaQt~LOdmwc=IC@?_D&ukW9R7Ge}qQu
zL6%>5h!IDjf*3<HF>(%Gq<G}i0vm%n7~MU=PRr)vvoG;`zw`?X?H#8wzL!0d)66})
zL>LBC`pWe6_c7;K<Z=b{dYho@rW$aa)`L<I#|c6M)6@w=AKj=Rg!Xpng=Jcprb!V0
z|LfsucvmaKwiR9^2vm9;S9eKWyrsjCB^b_YbVCS3f$KNuCKbw721|g~Y!L=7Dh!D_
zE{>gLU~(L5pptfJYHHi9emi4U5GRsa%jM+E0+%-%1So>JMMiJFo2kRMvUz2NBRAhj
zu25xU>IN=8(MP|(O3}_?<+CKN1m9<_)?&vUxACb5e-hR8xcusybgrzkG`EDcXE%rL
zzLy&g96%CdIXQ+02Pl+t_?vC~Fecb;Ex%qUAXcd}exlU&5qm<cBz_#x?snOzHE6ZF
z$nO(mO=xH$^^JP+si!bZo9XHO<eWU6?zM^i-hcRMywy$C=V!4xE;0x>?shpAb$SWi
zhZM(@BuXOyBBjxE6XGBwi5v!n!C)?JoT=3UhW!?iDv+~s=$fF_OuOJMO(SxB0@ov7
z?W2EUh(rif6k}Q%nq{D2rpI1}!U%OXI}KKvby^#BzVaV`2V-Op!pJbSZ<?Dw_A%_7
z&BUD#@Y?B9<dugZb!;a?qNN?-uU)B=-9N_BAHScE`Jd!3e*X{n)>r?6XTSDWln))?
zXaDKH;Egw5CkQ;8Oo5S+A&S)^bR$#{6G(O4p3q_Y|6D3!DKSg~-AFzAvY9+vTQ$%$
zln{79KpZDXslK~3R*3CRdC8-Xe}i1Mz(%cxP}()$n{6rlB-Qn|wOx}a=`dizP)?-U
z9)RaJag!1y)4>!1uii>6X~KxO>tWjt{o{ME`pe(ecGE&Pbd*X>m6JqJZ+pBke}&nt
zRNKRaC5CUjlgR@&Qd?T%;L#h%<tq$N9%AO1C#VMN6s#;-K11TA+P=_eGkWVS+;jf}
z;QGv-I7@eVljVgKtg)ROxZ@KXoSs6e7}L%&*xygSw5{z?OsGI!XKhcmwOva2wzfw~
z5=1c_*JZ2GpxtqQfVK;SmU_T_^SNg+O^eCNDRNGZI@jJH^SwBw5<89@q5Il#+(=7}
z$fcm(1wBv*Winz*2J<?NZrY5}?$#-mQUiYrg;#H+>mdk<yB^tGfq|(B^m2h93=u|J
zMxg2GdPtNaMq|C+;l%6$^R*VSevv0<=Gc43eT+{YqPDck(U0ANZRgl^<Tjpt>QSn~
zC2Qr;Y(eZM#9fcMrpryAKFX(m;-_e@ta0(>Q#5C;uyARP{+nw2<S!lJz`lLND#FR<
z=&w{LRPuCN0Z|g+zh8TOlJ50r5UUs+fft6f-7XvTI<2nzc1`WO<y|Fp#RJncUcZPY
zB{3njMueRe<b)z?2lQnnp`T!C7QWkNb$yA_M3H*4fuReshE1RZ<voLx2Kv#;1p*l(
zq=x30=;T01Vth90K7V^+hG%D%+3bYGmX4_+233RIhQrj<0ah0-Ge0-W{zGLf$H5t!
zX7$uOB|AeDx}X$W?SQSe$9Vq;mFfsupCmgvj(25?v2u|xiFo71=Qw)nO_-X25Q^fq
zv1-@#)2Al{@pX1yQ$V}jWpQDd+1UkxAVk;F@c4c@;QJi|EHtv&9PLhrIBqd8FvOR?
z{G0s7Z~n$L-PhRNpFlO6l$K{8OnBmjmw4v%>8q+c>hYtAJy6D5YatUwsA9HS5==1B
z8O$UM<^-;%5E!)Eo2)wyyN7cGZkHqsFwtq$8e}Y+k+E^|yGHR1g$zPOPDj`B2%!_>
zBg2IGwKh+lzR2;Jbv9kFwUFU`4etE>E<W?XXIYuQ#LUHWj7{t(U+p6voM3C|1iMNl
zloE*8zz;&!S{?RJ>|<i$07Ma5-zdTAMG_gZF}ujQH{RgxyY9p=bTmV!S}tQdHnp_|
zZWNR7L%JZ}1C&S&Z#!KN*IVMs(h9m}62~zc8yhs6&9r!)@6+p~(uRgyHqX-X3Sqdw
z?YG~_efNHfhaP$8n(iw<y^q$LGiXu~tB6J`!gd5%p~%`1m5d|`5{zwaU*A}vxTip~
z)kHT08EF$Ljq>h(3jI~|QvPji7pC!D+P>Kc_~xlOUYuWLs~eJ-I;M>1S50>6HhcF@
zv3_Na#f5pM_7^d28++G2)=n={upFYu11YI>LTYWFvFZ?&$`D#bkr^4oTdA|VR3J<u
z-aPgahi^QBsi)dr$Y#;>bT0Y(sqLMP%hJ*c^Yd2-!w5}FAHv^ij`_jbE)b;J-tBgY
z<2Kc5gD-vQOMLlXe)&DWS7Diq{n(vcZrACYnM0F`NQG?GBdWR}uSxQDNI4^EbtI-{
z(r(q*a5CgZ$7$8;*tVHkgg6G(;Sq|115nHo#1XooBP<(T<xnCf!e@Ei<<aApczI!+
znj0c51BZw~t;vn~3cJU~X>KetH#@_==^M!83nT-(scoDlYbJzo1WK`5cWHYeLqj_$
z^$k<Vm7y|%Ri9zJQX*+Lxb*sK+;Gc{7($Q;DCTo$x=y{FcAY8&(RF-l3+Z}jG}?qw
z%%zL7gkgkXqz2S)b-iDUP0_nb>h^y6e$KUg+B1vkY{tMMHUqMzM>QAWSP(@zItGqu
zV+x)6<_59r(pShMh0fsUP7>Q7$=E0*5mI4U4w^JUc{JjPSI=(o`1u->_dUp6Qwb+t
ze1^4T22JEy-3q8}wV0oqXXgAFFm?9qJwOtMs9cG9H$s)Rdst-5{6>dPBxrAJ;CI?Y
zB<wjb#YAh4o#h-C)>|AL9ikcd5XC4b2AR-ob@9XGZO6asT{fna!gYOI*MHZa6+)14
z9FioVR4meJcWAX*3=a)o^L=T#0xhqFIXuiuub$*@&YdN`j*jL%{{Go@@4d`C{3Lp_
zOH4r25tIu$c|D|@2`M-UT~EW*EKDuI_u7P!Pq)*goXZn+e0J@b0LR3s6p3ve;t0#K
z(S?a7A=W)=jesXkt#G0qgT5QH=+L*Lj!Fzx>OS>ai!*1=&{$t0Unw%Ya~zc<*dt@C
zoqL(lY}z!Wsf5MN4oxqCNKz_RK!ct84=_+)Wop+TS2`|ct43-3NZQp-jAiJ!o{t;E
z$Rvh$oZ|W3Af;q;vretn<SomKt3}f^3i&)j6KUcn1hcbq{MK*%_BAJUd+xoLOKqRd
z+%j>pt?fpt?Uh`NZ9(Mf7(&N3Z49B)*xDlYTq=b;QfTxKk0Nay=~$p7LV{)42x)@!
zXeBWxF4TDXVx94O?&r3>39rBM0_$=PjU4Oskb14n;=&?xm(GEqvwM6h)%I+WW;cey
z_U|PVu55L1W9V#b;&(g5NOn(6vZuYkXfem-%{Egb1GGXP6~-tl29?liyZB*(cptUh
z^L#w-@22g}wzd}v1=^j|_G)lo@H*V-_Hl1tAJd=xEXTM09B*xlI0^Aw8<|VUTQ1dn
zh-qpF40NH9u`*~9w$@hBk_0>BU|J59zDl~+J0>J)XSr!*(4>J<UABUd7f)~U^n9Be
zANV47T-oHrE6>m{Rv^i<ywPQSy~gs=G8fLAq*(1^Y<!x)^N_hRorMIIktj4sA}+7D
z38hATd4<q*@k5V22d3DW>you?=9?Y|b`H@E()EC)5K6LDZxcrE-+GWz((Sr*zv~q3
zz3F0ycbU{x#wVG&?~|O^`WkYxfxt|QDsut(B48L8W<(Hc#4%(H8+UD;N?#v|gv>w@
zm9a4`3k^_SjA@v}hD63~W>#7}a;nMMw$8y@?k3}8`RdbO<<!hMl31s#hYS?5oH}ue
z@yQ7ud-PFu?iu5AUr1;+Yt%N@DJCGKhLjOe95KJy!BG&*UFN_2;eTSWwno9S*^}sO
z%`J0udWxbg=m?#&=g+b4@Ig9Wi;U&q3dkbxNlgjgE1Qt_8iAnG>CkGoIC%IV^K*+t
zQQB#Ct%Q+Bty$yP@z?pw#~vemhcQS!J{;`4?RJt{i%U;EN9=ZiEK!tD$oiCu37V;)
zg%Xk&9fPcyV{`EeS=+*sEwY6&dNzZWwUG+6AWrL+Oam1w+F_ShXF5Fk#yV<XH$VT+
zf0?JB{00v_a-NOF8aZ93FXwRK++}|3{?G9E!(T(QO>X|!U3i@iT|YnzB)T$FE21c5
zVY5w?2wGPbc;f4S&3dO!CkYtt&(Nr?Qp{G^H(cfH>I#)yiR#V~0^cP`6iO;|wQccv
z9gC#DlfTO(2!Wv8?hr*0hYlZPW@aASwyD1dK6T~Y<4k_y<D9BJjI4zS45W<7<^uBh
zwARWnV!~J>iNP^!+>H&&l?q9saH=I##=<bo@78ub)%N+d4v(E~a-pL$ee@18PKJk`
zdx+EX7fE8Bq8`$p&v5$GX~rhTc<RX~*|BRE4}MnCZZ)WHZcs=RLTN}D6D2W=TU{(A
z2o`7gpa1nwSgLPOuq<{ZI<<ur4o^)|v@~3ybK%lO_8yp~>$S;PHa?K~5z+S48KKSo
z1Jf)nED=R9t=11V!M`RHm{ykR{!u3H{RF2TdITATpbSt3`GQNnprJG!(+G*WCQ52J
z!o=O&WN>g8g}~}Bpd1^^G$E0Q2z1jR5-~D#IeVqaBd@oZjVx}y{XW7l<gbo>gA0rE
zsKlVp2&v{Xynf;&mA)e1c;pdo`}oI^Nlc@$MPqZ59pACn`%%c^rb_{+ojJ{a_}$-U
zqrF9+m4Qw~yk6(dqc>ogf~FsG@$w~hj*sC5T^!v)2$xLS<NJYL4_Cvx^QqGW**!ZL
zy!8-`XTL?pi!qdeWjbW@F}@$;n9xPyd6GB|&`gb5BpDtopy^uLF-A-r#Ar$o1#N=m
zh{rE1@ccr=;K6&i@unS|yKsX4{n!784bKI1a-xW>RgrN9f-Fxz_9*@RRX+3J=RgWt
zwJjFUoaUZ<1|<|464wm~yInqa(++|Xoc3BY*4OBwv$|R1_Pow`pF`IY9GILUMB=*v
zt*sh;l?tnKOJp^TRwB9f*$mf@eBQybbk3c>z~O@j$mKGutgdtB?CER%?q+SB=U#f5
zzkcdjqPUmPeH1XV1$NwaE2|50)Ze^}l_+#&lFh{A^BR!^%Ls529oLJglq8AQ#jaGy
zRmxad1Ko7M&`~lW@FH~6L<T+!>svhj+9qec3YGC2sa7)l&cFNDyf%9QKatpE$%qP4
z=BTgN`P2XW@42$H%zdB!adb_iS>NEwrSt4J)A^_(iSeR{rOggU`m>bUE4=vJi(Fn>
zpqO!3-i#O_=BEw~6KM%IO&z7yZj$#yn(aE(Vudc7l!U%L22p^J?@MC$ZqfBLp}9PB
znccf~F*-8D+S(>7%S+e%-AWTUyGI$gX+O;uU&HZJZ8r^vOfJR`Vr*03BJq4l5(fxF
zqaI6!`U~ll45?5g#6gTuf;j9Dv|^sVbcL6%L<~&d#SKSBxOnLlfBucHu;r)Po)vj)
zQANfMgc+WD@=2<Fef)#_KZ8=xuGd&Pf0ny)859alNjxtgbX^|2VT7P4IO}z2ZfxKR
zjkT=?x8^j)$~LYEQxlU!Bm`bayH=-CF0;0<LPpbQf55dpo3pVjor{+)abS9yY|deA
zb(0Gh&R(BCC<K{ImRJai(-ZXHI7R!|o8+PxT^ZQ6gOiQ%f|!h>5wsPa9}|Wix?{5u
zMC=$GOlOlTNw~e&3n2DmI`x2u&&+Xr-C*pd`#929<m8*L@WN}yXhb1GndJ2nrl_J4
z8^57=?x`o)JvPSu_diG&L^L-ySi5wdyN2v^WS2^4bzF=j;tRKo(kaEPbec5R*6Avp
z<&6gS4rUoG=(I8>NB14ZjY51kpwnnjsZ?0psFM|fj#5AJrA6;5scW@2Xmo1y9X-Hg
z7_;%hafE`kx`tU)I9h_Kn^@V1pewKqgLdff%A2d~_Y-PvOuN}-AZv2d;R%d5=FHq8
ze|5RRY|zK2f999?{7?P_zx`WZ=BeY~#Fr9qD2P7#b9*UR{q&cHv6Wz?n&oGI`2qGG
zoM86cdA2UkqOC5n<IpgY1gT;S1a}=8q*$^!w_2mUT&Iw;F`O(luY)iRqy}~_gQjWZ
zESqMng}%B<C2!Ex(rF46O;C$d?;6DiOYa@WW_)ZIO$ZhimRVa}Az#Q-DHnP1#pkd2
z`=9vcH+u=)M+`q`)6zml?z$1T+eOW;qANkyGYP9cO2*iZjbTTGz95Psj6xsJpIPMa
zeudVs@cb<dQ!q48A*<^wEzNUsw!_yiH^}V0g)jc2U*TKFzR6eq<}0i;wg}O2$Wyj<
zlGXZf^a^oTr;_jEr@r)a-0{gz(yBLUt!`33f0iBN1(cA0B=I8l588~44{)i~<n+Qh
zR2Y)zWoBAUri&RIQ$p4zlh0GAlxQ_wg4K13r7TSWC85Zo=*X)b#`FH7X@cRQGR0z%
z#f2+W`}!~~1Kkk3eC&m5{{D7*i<Voba(IgIFlOu6DTIQJhK7|_*t*2fP0UP8=z?V!
zbRvfnXV=;1N$Orqr`4sJH94|>4@MGmeqo76W}D216+Zd-FY?(39^m(X@89y=YcFhT
zyG>qHsAk8>SyigVL2M-$u4ehd&wYwL`^Q<haFMOKd9<}9M)nV;+D@wNw@vp`DB4_H
zZ_rt7P{>;7PKLVI0VCCRE0;-!$D0{ijW+uFI^~>!r)k&%%9^15LDhEKW^C6Gx)3a0
zSxqN3<a3nD1zvvT<?He{g%Ah<_2wEIev^^gZ$btM?UScbQfGNnVCEHuP?)xcW`#tq
zi8E@kT7xsQYfQN@>)i;q)n!K|%MJU-QEr!0^Gkf~Vv`MRgrEMEf5IKN+|0lFH^0uS
z=T8zU5M(Io{pntB_tRe*!blW*MvMHyuYR8X(SBymoMq+gdHTEt)$9m~O80t4)A{7l
zAu?HbeWi}OvP!X>1KpwJT4;t%qQAA*>tt<*`o<>a>MG^3%~=6`hDIroG`2k_KWOVA
zlX0%<2>+^*y0?n<mKW!dVL++a$MDTJ;%?P&&&^T~4B9J_T|*&xEktDuS{=nk$K}{;
zmsi#kUfG!AV717B-Ght`Ww~-?mNyq`Jh2?HkQA6YbcB08eh+{ACx6Hj-+Bg5N_1>W
zT9s;MjH11hzRCcVQVvDJv(FwwWE<S|W4Gb0ZScz9e4RT73PgcNXqW_{i!TILA;(5H
z<Vxb8_Z2Wjf=`T>Bs{*^Vq&?$j*gGhb+H12$ZfMrkHIk5QSBpa*U82Su1GrpC5q?+
z6zYlb3HDD<a_pt&$z^p~tqq)vMIvKn=NEx@XI%IeZ++%#-+27(mi!(+I$TT4mzU;o
zJ5BPHDwCi5IOp1r;<dVLH8s3!YHwl20g0_qYr*buhf{M+POiqBZOr0!eFh2{4(}R(
zFk)?^#<9gFk8MT_?!1{_{l~vfx3kWlJ@jX+cIrrU914_;Q7X<jmEsW9Qk8t(VPR&L
z%?+2Or6pD_UFO^i&#=Qturx&{j)^0WNH^FuRwarff-dyu^9UjF6sWwzn{mQTz9gp<
zai@c6nE0&*rRElc8Jm0|j~;lG;+Qx}yVK-VOc{7T85|ts@WE*=T|AF2LKYV;Gd!|`
zzP<{_UU>z$wwSLjFCxQ$LZQOYksI)9b=-^dG((;Cy1}l2kc<{09i3KJu+{bW)?A0<
zo04yBF0j8+V0u?Sy9Tl>EzNOmrOwmKG0U>R#QuZab^FKp@Biz+@bt;&@nwRpY>Ijx
z6=xTDdz4D4N_kt`UwGkL;54}H?pyFTH#q+IW8BuCCk}lA-5?BH0>I4Y*m6Ub5(}-8
z!yv&|37$-NYOBN8a+A@HhvT}Kx<Tl6*rCNJLuaI~LeOsD#0i#A#40sT4w3JfY2CY-
z0`UDlJ2p1P)Z_%OzWNf`jG)!t!gfp&8ME*nX#4jgk&?~Y7PXB<0?(yf>Sxy-w{odo
zr#rtw%{B4Y6(g09906nu8l8l-PKT#1bve_Dd1a-;;h{1Ib`LSyZ!t4-j@K5O{OwB2
zrryWVk9~|AZ#c^D{?YI7(%BP)C=4=`^#S@a;}q>(RI63Wg)Fi3dF;_=$n{6;KfI6b
z$_g(({4k%{UBUA`q+t+*E`e^KWi{410jnZ|UCk0`3ST8OqnL+RH`%>9M7iT4yIlkj
zbz1B(Vqz!9P`QlsTI7_(m7tVHe3eGz?-vIT9^&}%<3Bv+>MB@VT>Q?&wRipNi0$s~
z4i|nFcgEf&T>qK;-qqrt`|n-t{^|bySE8Y~*WV8z@m?o^%cxqN9_CK}EGJ(xB?!1i
z(9PX_``@+e=Ea?<H+gyV&ti9=R@*AW|1S!+8tz}UWSx7zoNd*Y7Z(>FXP=$DUFkh=
z^BqLw<99#(m9Jsk9|9M=UxC$ur0rO~Oxgw2=iblHxEEp0V`m3E+Wy$@{r|OIGbxXm
P0SG)@{an^LB{Ts5|3{@3

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_023.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_023.png
new file mode 100644
index 0000000000000000000000000000000000000000..f4a6644800a477189218cdeccdbbdc4ec474e9d3
GIT binary patch
literal 123801
zcmY&<WmMcy+ijsZ6!+rpu7eeq;!@nbID@+ucP|4JFIL=&yASR<*g$cnSh36dt#7^e
z-XF<HR-T-5lAM$5XYZXDbyfKfXryTG-o5*vs34>H?j5}6+jb}l(%ZVsLD~JS5cZJO
z^U!j!@$fcxw|=K$?&0d_;^Am-LFHxb4zzc1=H(FN;ANw-^YCy50ysIH{u{vI;%>_c
z{v>_$Rtc)Bf<ExwJ1q16EO0+1%I)95y?du9BdP5JI(<G)b6WM?ig{kDi{dQ+!NAg+
zhB`k{dAXCZ<<%{!*jK-{ucJln*9~~dGd-zX2SZ#9$=&5Wg6l+z2EL?irW$?~pw}<4
zN<o0brPKRVs6L+;1uEw~xBu``cPpOLs*b2#$ng`~mx$*5r>#j*=&5auch>++y!LX-
zKflKIWYw?bzeR9u)|2-$(tr2=_o!NTzOd)?Qsa%InBo5x!I(?D<>&wJxwa?MeSU`v
z|Bd*cz-M8-Y0&TgzYz)KlN_zP|6A360-<JNz(4<gvoB<9&l*9a!Xrydgk2_B)Prh|
zVq%7B_P!T6$_yZhU<zi5{MQ!@Sg^!0@8s)ljcxYpkl{4fYh0^1%OT}h#roRX?p*0-
zwX-QP!{C+^$Uu})r_Yt;Gx#m~+2nr$^UYs~T*L;RIJf?}We3xh|5`qQyTp2J4DOY9
z>UpL;-|G2*yXKb%p*je@r3?&h-S&;D4S1L^>h?mUC)h$4g^&2OVERsTe4dj!S%?nP
zg*Cq3=K`mnw)B%ccZ#aI60Kd3DOKs9e-e<QlnxN9u0&n>qY|U6@^Y)-K^Qp%8;zdU
zsw5>vN-+i-{-Dn{shQB>Y}Y#yTgVptB<>>$FG)NI!O<)#K|qiiy$mXmQjAQ;wZyR2
zFsnfgUe|Y}fU;Ys0AFugefLCO#adPvClDS-5`CNdYoD8Y2HGc<v1vLpTaiqAu%K*U
z5*DpgD59)A_l<um*gS7YL_QPx6#jbsTHkxISAFRH()W7sIv#vg@Cxvz+(3dO?*f~J
zK6C^eu$mrTFHWeT!a@$<VTX5-G76zEiVIOqMePJ(DYJw=9fJFisM=ugdntvIc<dCp
z%K1A2E^i@9n&msoEm4Ocj6-bkv`iW^4uoL?7A?s4XKl;#uAwKU$=83|kd0ETkqr^)
z_TCNiKVmmQldrJXKd%olT(5?~j}ixlJ<lo>P7(-k?9IxxM6kG?e=OeT8wb1vj~LMD
z0%;{g{==dmuc@qb+=_mZd8NV#dDM64;vmaX{FEg+H!1vb84V9VhKzn$9F$P_Xd0!`
z6`U~dggqecml{ctq%_<EF`r8*s*<8q94&TrogvBtVegRsaO`}}ru9x1{w-lph%_}Z
z;PXCt#yfU|FfO|he{2dg=4|GvcVouP52}aRDrIfO>`x*WR6S4hU$U|uio7wZr4M**
zMv{_sLOGNuwjU{60xwKctw->!Bb;k22LTeV6Tx}SQ|O<<eS2s3+`tw_iu&5=)~Y7i
zX4~<lO#~v3k9wBOv0<-=f-hr?{ZGMnt*_$2mmRf@U4cm(@iRQ7%jd8)TbH`lj*&%H
zT|E>1-JvW_*M~Xb?~MC6KB5r2d(z;3ie6eo^;auPq7rO{(pgQ~Qt@U2%@FVX)%2?)
z;tD!NkZVA|;O#A84V|@+UAUmf8hJ^oU8$l0vlP>KQNP~%PbT<wHJ4yboo^40n4hZV
zi&$8wcIR%06b!rSTwEdZd^xPs{S-cGLQzGlt+>o|OTwKu=%@WbZt<$OtH1Ba6K^RR
zpYJ~yq*(GztgZ|1`c$_)%n{k&Ec@O7O|7k+@~+0XaYKO|AYhnS!?3z`W<zD24#BCC
zHIs_0K*3ePGtuk8+r{_%i+c4A+Ut3CqBs+MCt6`txc>P*qvx8%`(kU4x8MOIi^Eik
zMb_+V%q0f+iJe;YtV0tGQ2}zfv{PN>REL#)Vz%DA)K{kUsz~h$-s_TA44ngXmZkeR
z)DQ1oF_nH@Kkv`P#bxG`Vzt6QU;mei1ib|BR*P)<qjU=Le=#qR>ctQK7aD;A?S2{Y
zqhu1zw-B%h_)Bc-u6)4sIgUH5&y}Ep&Jb&B=S(mNPm3%K>goXxtDK&G9XGYyd!h&3
z2_b_?JJG!hJPDyJu|@lky^+aAv^~c*2~gEtWHA%^OYZjl>yzH3M5@-Z<6j_jsm<>O
z9*bhT5AeBTuic|`<R<r{QMdRH7W{A%N^)5tB8vE`;2yU+LL}4ie@Y%GE^cLm28KN*
zu78S{=;y|EUE!EY`^$Vx|7oCv!jPE74BjiGFF*<xgLRcu$Vz1t<7eE>%_7hf3%KDf
z8%@E?AmGj8Ey}|_)SgWILJ}?Vi}^f*n=7E2e`S2fXpajJU%@q;JcYQocNe_?CnZ@z
zcD4c(72fai-ro5K<1F&TkI(mRTZ8-+4{@w6bj+B9g*C8o4}$>s&BgA7JxPt<w}pd)
zo>3<aB_i6bW-f|v)QEuM2*8UEs=gahW*jAQ^2{unD9rhT8zR+E)Xpo43vDhmsotbs
z(G|2vO>P5)RE6RaBLrl~1#)L~MOcc7S<t=sWRlNCo?u%LOtvX;U`TK8H1E-M(y48~
zfZ?#~K!w9-#mfJJ8*B(L9%_az-7ie&`dpR5^zeWWbJn{;k_wU7JVL*f7W;IZ_Df}|
zLUR5RLddrw@0d%a(2|vKWl;@=)XX>{%m`(DN(r~)I`@Qkez!K&NE)J=P?eb6e^3b3
zh}D$e@J7Vyg8o<^dcWpd@)vY+Ye#zm{d2slm`8nu8E#hgzK!B*6b@jkG3=A#Sn7g|
z)4kUOuSDH__<JiI?nH~11l&OlDN(#myEUUqnsgBN$8FjrMm@w*P8_d9>;8eq);9Mu
z1|`;=zaYVH8@?am2YAd(@DX)%G}7s7tdvV9cai?lWxF_YG}d>6U$!hG(qv+6eVd2=
z{0TWYz+6I`Yf4})Ty$m!>%&&uFri}bs^kHpxLDnovDPS4k=S<Rg(Y$d<Q&WYpfE?J
z6qT8|;$}L{FlSbCOQ%oen?>`^h0Viju4Cws`lWY-q2(D2GH!#P{>}i3SkP4EMtnZf
zmKJ5SWDBcl5Ry@zU2Y`sM<j<-g$sgLaxR0!FdfK!Lchc^%_-=M(12+Xgc#KU{d-K;
zsP?!TQ<Qiyma!YVpFiV<djnMyz4Q#aM(12}N+H;W1We3V;Hci_7Vafd5B;JzcuSi2
z=A#B09&W1q-V4UQ083!P$ueg{a8rUC{aNNte<X!&t&{6bJHfc}+{Il?5fj2`_ibsI
zOKTm!N(e)#4DM-%Ai>h=S|8RlUo0*SdDJJv2s0`=`-S6G)6Q?bdw&qVp`)hxO(WXQ
z(ng@6hA2n}7WjW^+qw6(?dy>3g`oOdUv+1o2q-9}e?b|fsd#&XP^GY-3H)lYv=k|p
zu-kKttzjG1yWS{j`myPpHbsFW&zrt%CYG0J`(3Tw`_A~6_#M2#iR`7<pQhF1eP-6-
zG-1Fv)5KStCfzbcw!kLfOA6?>P2j~1@<spzZTsQ;MA=3@;9sNQ!`gJu<nTX0<7o(G
zr*GDYuFIT~Ry?*4=bxjgLVO=iH8C>;sc*sFo`)=L?{9iyTqT*<PkvtheXr>I^{C6+
z2OfJ^mlzseNJMdhF^nHCFCUNTMVyqhvSDN!2qOdhDNGZ$_s9MuO<Rgi;gQ$n=1I2Z
zIBE9t&)&XRi<eSF`VNB0w}c<fGNYGUpgRS2Y?DHo$W}~zpTe9e3Rj<dOn7vYa6vmN
z!o<6+!0_6{pvbpRn$HEcxD|XJP!IYp?(zq%<nDlj<lAKa`7bc&T0C2@`|%PV_V6~d
zACAq>Wk<sTu@Awzp6QIq{g$A<rR)N-Q!Wz9`ucY8+OTlRdC?w)ViggWH=5+=C4Y9*
z#Ktnw6qe<=a!e~>ey`D|v*-tGTUr7XPIZOOH0|qfF5U9%3Chs+YR(L-fekU^$xXZd
z#gk>9i={v^UX#bau&1@)OY2|a-ig)3ZYN#x7|4NZXB=Quql66XANV^1{(3?)yYYoL
znAD<Xa5}b{^p)VErLlJ`rI$*a&dJz|NZ}jTNXNbZ%<j4Ab+{R)877Arm2E5yPFRmT
z@}^3LQsY;1$ha*BMa9e*A26UrrDji^w8L79kCTf&{uiVD&X#5$*r7ioWJF}kb`VQ_
zKMzi}LZcTyjwS%_+lLI)Y{>nc`zIBV4Dg_|s&aR|A`w|A7fgphZU52d$JYB(kIA_|
zey%}ZCnFop$ib))xIY!Jm-}LYE!{HwG!g0r#6V`TkGT^YzcW$|`y%5`0v>ElthtEY
z-Q6dzuDCV*Z05rn_?ise&JcVNsP|%EY;$Bc`kq_T!1hN#T9}d92M!?&FjqkWNdcL9
zCdrn6tW+{a7Rg_cl8PTpN{9NYWcCPXS_BHmV>XuKJms6JDNW2O@iW}u)=y|2>FTS@
z-PPGs(55w~cwRD!Q@8V{2oskWzchAu5vVGwfVuPn7!LO*E8O{lszPaW5dtfll<fWx
zNZ%_YJ+-n0*l*bMmyFyQxM;Yxc}`>Y7o}t=4k7%GFhbrwp+#WD6Y7BczT}^t?j7<k
z!EXPk30?42c&mSWc~s}+u1@z;29ds@X>qk&X9ut#Zc%5=KGu^51iZi&ioj@#?ESJt
z!V!bt?WOv0-6Ys_39cX&U0ztvz*xl7^KjXaMax1(QNvoHB3j#ZG!J9i8bHo13@{S~
zSXq0<WHjHbwRlI*Iis-5z%hPo@GE55`<hF#;pt{mfQGg7jY?Q3>*mFSu#irQML9!D
zhgTh5^5-~Hq57{PK0-mU-Ti%WsfCu0Y7tba`Hy$bycFd1LFX1f5fne6b$a@+_M%1n
zwIjzHX$dO$r5Zb0ruvZL`m86iNBLUplBO$<uOG%AEAhPNe<}*tb$O~Ou-@Bv$BcWu
z3MzehM~$%L8xI^!R;<Btu{@RZ>0tsYz^U@-lgr`{$!OsBI@9#Z`a&1H_^5x6&~!?D
zp0;uy1mB7^JVei4iQ^93R1vk`_X9N14jH{ywk39VtcL##%jS2y==(i?03n~qv;{;A
z>?KFb?k|<3NdI23P$r-#(x&<oA1Q=b9Ag=YfAkL<0KF0j$h;b}tPF_o(s)*a%q5oW
zxO$9sb+qdB_@Ev;ToO&Qq<4^}kq%*+etcKK@}uN0eY9Fo-|OD%?fU)!j%?;jI+qJg
zJJxU~++#Rb`LL1Fw1@lPImx7K!b2o<IYlWsZ^Qg`Cwd@_Wa5OVE$F7v<I%xq5HA~<
zh+cFXtAArLdxDY6<<^YL?!%$dA2T8l*7Q5D*{}D9MZ-pyknE1+SzOKC$ftLcAEHwm
znEq4`FdDudnszH%AL9GUxU4|HHDW<01<yFaXD&e*i#HSoW!n0Z8z0RG^pKYxNMWuM
zzLd(z7!-3kn({srD(wuh7EMcitK|Y_@bN!6wQzXgol_Lugg8e#ZAI5?FChMH577^Q
zi3ubx<{0E1j5Ylg?tBWSDlJ5-j(%I8|I>K_R(Z_Y5#?85C+hi>iBnN-+4)emaD+21
z?cm@WXfElG!DstLs&i7CME5E_`BOap!!SM5{J>^N1Ti%xgJtXz)3cFAqD2ZQn=5FJ
zI07tAt^&UZ2;zvJO6%Ijrzh{wEo3`fL~BuwYigzpq7JW=hdo@p=C-b^=a{R+dgCNf
zkX?n}I(5G|n3hNwS+{Kzf%L~aX>t!d*3SMFRfIczf1E8qk_wak$_PriOjG&7fFOw(
zN|&2NLQc687+Jv+U*BXv9?x9IV8t5eJM7^XR^hI!qn<%xFfbGc)iXS%iuf_AmJQ|^
zZUiye&pLC$;pl(i$@{{@Z0G5xhb}8_8|)p|XNDS9THDCi=k|?(6&F5f=SUw^Rp1>6
zQ6jmO!I$4btoahfQ-D*E&8Kf$=3bJ`O-@-@LREW+0+(k(>8eo5l_7b;bpQ5T5t-l!
zlVU|vP=vWI=+&u%!!R}<mnB{)x2a+=WGVK#cF$TVfpvn1C&4$oM!!2_Q~O#tvA!uV
z*Mlt`<i9Au-xS@(Y3BtxxQdHGsi~F$m8^~lmc(J^%!ph1*A0UP;T33|wF+s`s-$~e
zW&m+3hZBypS#Es+`UJ~g#$A)Z<;=GMnnda55imJxrxa`-{yEFEaM={y*TQt3)Kw^K
zpiD+O3(ebJE0nve-&dYYHpFD5X!xq7thb$I14}6sC9(UOx>t0qCyrT8#)SR0lWGH+
zwi<zachTSm;ZgdtwuC@Q+Zj@g_lfkZNZmsnO<(^<QOVON+~bJ*9}p3K4l;`<u)%S7
zsW=+>F0swqw!q~Tl3RT%*dA~ts^nJ<WLkIZHY@2W>0WU1>5aG$rfc9xVkTJ3dbAfl
zT)_5md^Q`vZSzaRpGUpuel8ac`bz<}DXE8U3<FMBPyXt5cmSL(aq44pl59b3v=rMf
z;@;bSX$i39Hb0mSQaF&BG_@JF{~djE5oehBqDWELp=O}?Z+Rj#&3c6QK*W5O1<8;b
z(4JI+M&z~c8j1E1D@O2}Ve9q*y2K+)h6l--Xd0vUEqCD%!ofq0Oh!uTCyr{S{{C7r
z(T)Vey?AxxDsRXC^Li%n4j7CCyW4uvYT1f<Fi7xSv#0$!**F%TB4Y6f&fN%pM6NFA
z8P51vJ4fmKjVGC|^ERp1Xd%0bGO|6Hx`tEAW)3~-3#s|kBi9}Op@0=bFx%lx=uJ>)
zdpeR@3^!qw&Q$P&)r5Z%Gr9e4Oj@iAx>swNqXb3$^L@lc;2q&amnS%H{nO{9n|$>%
zOq1*#zL^hO)zXkFnsnY=-ePTS5f5C(u(1Xui-<v#q&uYJ_v@&f>R`+9_`Kr4^DY>m
zvFov9ZT3RWNW8Vr%pP_7!mXO7j2B%e5k#RymK41L5iY<L>z0||<Tgz>LQfzFQ}T&$
zAp^9cFs$)dEyin3mhApg*DIZsN~LA$iaH>C5xYc&zsnrT3Jz($b4!wl462R#SD=<6
zVB5yGZx$3EEhet3y;PLYhdvP=b?5WYXeu*-LM@%%-PV*>#Y;?aU`fOI>F3VAy)n>R
zLio@?R_{boFTzy_TA>u0%Dr=cUa(>5X`eTdrO8l?Q{~V#bZ`{dmSvST)K8mhrZCPJ
zKD&nc^)`nn)P_s^jRE2+8x)TF0$+--6@^j5izxqywYus5DM#(tbDfRzgFY#yqL$)X
zY~N>)n}?<v*@$u(uX7I!Fs!)i5v^|Aq^Z3~RtnE<5#S!dkEAni)Q^aTVk}a2W9yO7
z=zeHnZ1r)#7#jpoB(uMM%BB$+rebm&6ZNLT$NiX}hArb3(Ht79o&%!$kmCd;AFl#v
z{hS48bBAdW<r`W)N{j&j*|W~*R+`jSiuz!ojX0lx*X&u1VanH8!=~;n==Bf`<+xt&
zs~?y8Mf;k67LB%TlbvIS!R2<iR{2-MZBYQheHtv;8ZinxS;lQtRSsTjf1?A2nMBzo
zFD?O>)!5lI=<X7uU`uCT!1NVwESe^lS6mo!IqTGNv0-C>CYtrDYt*a_M`-aTkRxy<
zn0j`nz=uMUIeq3Q0B_?`?tR|7cy<SQ^=}unlJ~!59E4VS;=E;lAj)8xGI!QBQ$pH;
zUnvTnb82x%*h^aYWk)%VM8U^^T$0EgHG6FOB|!BDeQ;8P_!1afv&a*j+2WICJ2UG0
z?z{54oOs?_FP1Go7@_IEoaN|rDa9mQc<5lo2-}mreGNk-+65tIs%GfQFHa8!X$CcN
zV%zu3om(2Ibr0o@+^@R@&$M1GTZf0htTvyZv<4PoVR3jUwxTkqsjkc_GeIM!d{c3(
zA-+fylGhrP^yu9)GE`OTdUMEH?PLU}iM>c;{V`RBnoq-kTF3n>%J+0UmDNS|cnUu2
zXmyHapK*8akyX~0I~D^uM>bAgfG{NQB2dsODkLwDeY~h-@npww>S1>*;b(e(EJF`M
z5f3D}q8!*4tUVcg$ZW>(pDr&6sg`FiU%c^p7dcbIPj5ItLSBbJ&Ft46AvkkV;vLL;
z`=1!a#xn)cukQYg+}gSZH$;U`*_3hZKVt^E`^chT4bDhHLq-5q?pmH<{zD}h$7(iS
zfH)*ikaTh=X~ZZ*%Dl#IBS~xWV7|{tNwJpqlN@|G`)Kd6@5v(;Gx`Xm`v6>n9Pp=^
zNFiMYQYW=*RCnY#+bOpZ-s-u!7+Ho|s$o!AGR&m1gbj+^&sV3@tUG0|N6(o@VPaF3
zi&_~I-+p*~CLBTbZudGM%RTvsf?AwL+u2MP0t~d$uOBD&7*fWMPh!N^i4<v@+L<+6
ziQ}ZM-JmlNi;^sg_@P)Xp!(N|dX*@Zwcc#H?<8`tJb5QNf^IQ!mHtu-I|rGz(3>E*
z0V@b~Hl-Sq!;G>=bpfEZfhciGyQWpv=<PO)UHbhwJg{;+xyew-e~m8!TOAkx0-}^k
zm94h}E-^@o%0nNn4U(>YngLWCpw)wxzOKPUp(Ue;QIE(*-EDnVq64d`3dp%T@AnWn
zPn7+nwOtx6RY*+6U4P~F)}Z=<idlGRJ}7;**fpEzv<2iP+x2Z4y0`~44gA=q{!>d8
z^5CI_JJjSKINn*!YeX1jE|8mME1cpb+UXzF@RzN|Xn!Me0zVQSr?|O)4IYZ|o!AcT
zJv=qd58+^VI!IBV>9@&w*Qx$)Z@qZVF&H<1;)j*ecO?%VYjOX0YzlTnUk!7oNIJgv
zk;>k7qWzJ|J+8Qgqz=1%<DWKar)L<Abr7vl&*-GkB4NOYBjDqvF^lEuI(O$CSgia0
zAxm)$buRK<RsER?gR!fD(ugpfxa)#d)6OLGqHB@#JRH1gV$v8CZ&wd*L?e%#R|whH
z<XS%XQbe@vj)4fw4s^CmD2mQW7+I+tM2m9E6ta)%H4p)83dowMuW{d<2SuI9yEM8%
z7>H`SM*sbzFL4a@PrYv3DOMtZKL%*AJXO3u=o(no=3q|y2c8qU;I<)S5-i)fw?ogX
zo>lc@^VA&gzuL-mp^B77?VFv2)6btj|Lb;`o<_`~jtl+jT!>a`Y)f_NgP<As7x+@|
z^$iD!z`iN^lbsSzb#0sfBa<seCv@q|SZ*GtYJeaDFYa+P=6kGxHyN4>_Ev;5KTdQe
z04={VwcI2y`t)y@JhT((V|3()7@=vKfV&%)>)O|w9&78B-Ho*fa#J3%Uk%+YN>!T-
zv3T(x<T@6=rc9TixEwcGRx74-53X3AcRqeNgc+LzojFw3^n@gnUQArh1Tw)fOHPe_
zoxAj0&=-MU(vRntaia?UT&tn<t!BTsEtM@-eX=aznCOKt{J_)5d}Y03sHQbIB%MUb
zt}G4J*My7s+rV79fYjmW@+JuFg5QMrlSPyCPh4BFbLMpGHxX#>I64D@_J^_8h(*U#
z^I~8F`MOP;)T;NRi`%gJq;0^(doEKzoNxMq5ud_5*2vR&$2L{&#dXDU_we{X82{n9
z;v}_<i)T6^lTdlKu3mhj3`;t3i8_<Mh>7@VOlP*nTlda6pBevVE@78i2VW4jZau(h
z9x@dMi5~I@HJiNr<!aSq*&CJP^(g<biKx-s=KUDrRTbF_+kuG73KWum=o{-*3{BRl
zsX(SMwcJXtrq^UGVv=GS>cmulKTnmi3kqf#t==h>xy+!OVd-Sfp*#or)J=i&F0P}t
z{}FzEPG2-fK4$&$MNm)QSi0F@Qbx*ue~f*D+0bK;Ip)%14NYW>sVenhpL(bUXCK+)
zLG)oxmg6T1B-arygz+lOU(Qv8(1WgxFkneIAC@Q$(N7RfTEgvK_g2P|{PpG156wBz
zPJcJczgml1V$Hhew1NNkCwzL8by<1uOX4{u8-d&uBKl2}u5!XF^D}((WP>7Wrk)1{
zvub%cR8_0Ca`jmMyF7A`lTtBG`<PFsXYcqp%a-u2P`~!KBDQQbdL0`Df45N@7Jcu_
zMQGt6haz<$CP(hCLs><1_;bx~suH%Nak0*|YE^l=@#Y>LlaAu$<Er;Bcir;U^fA}>
zLNtn6|GLE8@b}o#WmP6e(xAfn0+!+hyt(<w!>J9~+_{dt03>F?<6FUFk2UQwD;|pa
zrn$<`VJf#83z(I92*;l`eeL_{S*=hkzNSG%i9c$87@u=}I3zmq$gGGJE}q5gSQm-g
zsbQPg866>IMx7#(+Cfy{lnPJ0wfhLNIhGt>%hBy-s10hc=GOiKF>aG5qhHc_1CNM_
z_1>aB9)plE9HKdubXaJVoGS;p?R2}m*^jbc3N?Qm1Ez9oj`2PH5#D+%n)7WiN}bpt
zyfPFwyK?6?Pw-)w@`*a7i4Dtu>{P5Q!$Z$lpC;Bam2U_zAe#FlV_28aNVbE((kaY?
zepuDl<5k3J6|O%xW$`ALDPt!PF_n=?6Z@PTf~t*1T0XH?jq2@}GkcDyIOlA)bor*~
zpmCJ5@HQN$>x#ef|JrzBgU#YBf!l(iV*5UcsmZn^s{x5dRrVuc-z`nF;;j_OWb%9M
zCCzH&fBtMalRzxT<P4{aj>ncVH+mBp4$}S=qIk88oH|n0i|V0)-hXu_ps*)jTAyJ)
zaNOZUg;NsC3C1mF-=#JH+<Y~D=8I_DZ^RiHK|Xa9JG}LG0S54uW*>+K2Zp0VzZ`o3
z{yqk#YYrOlkdX7{PWiZAP_HMo&XD>&V=9uSlu5U=cqtm9?xnGJ-#qq=0R_GOgKwpo
z$pyMwA<eP_ja5Kgep|cQ=6yCzI30x=*43te=%DuwqP{8Sm7p{KJK;46$nLQ3I1O#0
zWyI*rn&uZ_Ykll;r^5`?yg*M~MyWz=r;zFaKoXv3)_IZTW!lF_M($<3tYEG~kvFic
z(th%sE%FD<lRx;hAH~H<nC%T9_v1fI$y6R0ncR-@?b)Ul5BIGi=Xx#PepUIkc@NJS
zc<7)|di8=2BwTI3eCWi=V&^{u>qpR}lXjt92;qjlKkPM7)3EIJeuXQ3#vu{_YZ1sM
z_;&YpGNBhW$=<qgn~qio%n{sHdXlDzR2OKe;0vjUI8m4G9J!8PK~N{Uf{dNfIi29_
z9qf@zy?;OojzlxzpZY^34#rsE1TFg#LtrfU-O$oL@{8;HrS(&1r~1~wsM#@mL*l0o
zeRFfTUCu-q3}pTz3Z&Qf(Tm3sXJ<|nWaKk*VRSKy7AhT^=H7S7O$Q{D5>_vHs7Ah#
zXSt@m#qGZk0wMPPqcv0$_*;JN2Qzb53ccvPA8c{JJPTo=@jgK5`eqC6bJ2r>AS9DN
zmzomLvK-j(Q-pS@kwJpc!snyw=qIXvy_iJNUq7><B<<LCIHWD`hZLN2nxE-#w)El=
z&7F3Bd^OIpf4lGc6rc~1C_n4^gx?^2Lyt&NB|;7UFO#lM9dDz==C7X<>7QZ;o(2`(
zcw|noI65#1Xy7Zyyqr}3Zgt@#aIFTxQmMt8k~eSis05c?>u?^`I<v81L0+D_hEMWy
zC@Z?$Ja^+mL|2zsY*A*!Wnv}EETF8i>}a&1@i}DnK6+{@{w?w4UgDZQWaAeoFEqNJ
zzcv8AMkFrGXjB;eq*IW(on7u+<*D^KIWyewcMY-mRFyWbN;}?wl0yg~N~IhoUUNzy
zk{lo7pDjvflKM6gGegasXLzY@{M;k)kg7*pM-Z(CVRqGK((D((0g+9u`UUhJu6-px
zkC^+7$06in``Z(-1?;{^QpYf{x_|1R6H?i*?<=Q6ijgXm7~Ry1PqOFO-?v49b-sCc
zX%@l$IrH&5Yt|RLRs{R+*o@rvq`U>HV-3O`9gnAuXQ?d)j*iZ$C8%y2(94}{q0~v+
zsLrc@7Y)A7SwyCfXd3Ce#yi*i3n7Onk>hcu0RyYQA#<54JjL0p)mL<Gu`n;UazWAQ
z*)0nHvA{XL;dUN`I>_$<J!uX6qmsyaOzM+KE0v05z-XI3#*(hkx>e)B$gX<OB-t+v
z15Fk=v=3H#hPYz_Q)xMB<lfE_`&yhv=s9S%((k2e+7G^xmocCj6MHfe^5Cnml(;FE
zRHIfXLEdRulbwt2zihT@)R%33op+M2cJXKnFi)GBQc+c9kkN<*CYsdBgBWC#c2$O4
zXG)iS6F0lWXvnqpg&S9+)6($vwSkab21a?2r>oYWsOC7`HrI2TQI7UcbKkJ*5sRP|
zld`}gu|Ax-U;Th^-G)SUL?t;oxA6e+k;gFP)>mRqS~-!Ar%8A3+f6>D`Z!w8Xbieq
zC+rpD(bQbk9nHl2;bT9K?}I2Bx2bi*9~ulP6mqp;4L+UML<RhLpp5bzP<v0kftNi&
z3gkq95A~-!Y#A4N$<CI40~&$ZCyZQ_)aAS!MjJc(c2O4rC*FD@!I4uy!OQ4g&g=3y
zB*>cK-zUU`w(lRQu?s&ge{4!d(RxGcG>I11Y-TS<;DR0NL92`B?C=_R-3CFU!+pEI
z7bw8UB}pl>Oc9&a_|ewvDRdSsoDrO%4cJj1=5no(3|h<p614f)y}6wJR5qq;3vNlc
zsnH`KZ>r|N(t`Rv)+-&p37D+L+<JbsvxFr?n&fbdH+JnF$4WHa*wD~Fb|pk?e{o#S
zOn_}4)BJ7GJZsS;uIo7vA$80n(L3?VZt&^~B~M01Bj)j(a*h%NO?YAOrX8!Lk8Q!m
zpwUwljlvATEDd!N8J%G|an(x>V;yfSq}%>!u;I%t_x7Q9xfNXfci1sn+Z6_*=h>F^
z?K^0*vP#UBC9SaBMesOS6&Bs;xTC!KT<|KYh#z_S3zaYJ#|TQFR^?e4Mgvtsz$l~f
zmLR<`K~oQQx(;~hR$K@39!1v|6H9eP?}9gHQh!?yb4qmqYy<>Pt7BjCBUdr=&l-KF
zs+RV7Pcio;?wGEasmuySOCb@^>jNm;Cg(c6YYRA_=~9jBqk=Q!CdQc6h}TZ0V>aG~
zvpsGV(RLf#dV%wWWZNL3FkQnlBMz7+1qeXS&swABC=@qrMM>QtxacO;HRK)hrxvc7
zpbv@VM3}bpOC3KvLWV0t7Og}fx0@nli8d>DTZMT{naECui}<!;KFJz(v06MdPOTLz
z(mMKcf1A=jf@qWKoI_I_zw_k7Z^3<xN$1wlDcsvZ(=#m!oUhy)<DCxoU+=u0^E6QG
zw4<vBbHJ~aRh;ozGWMcjLvm24>vUu_tUMzbh7IFV6PcD2>{qr|Exe)B3i7S<#d$1W
zba4SSa73J-k@nQ0pF-log`3V0S)VUD`ZG?9M%^BX1L2mSo9uC(f@xwk2I3PTZ|8)E
z`($s-*XWmVfbQEPQ%8$JN3zqfb^u(yPwHLn1E`-YezBUjEE(sv|M)qfd_h|WE}iUb
zvp+jthSlA0@t3llBd<<x!{!NZ-V_Q57-<iviiEES)fz{vvm1Q)+45{D?3sJQerTDH
ziK7%@y6C1ri6+C8BHY*E;&K(&!&jQoVjp34g$b%i2s}gOY|zCtM}%0Em|HBZ;%H>q
zvGtP}=`=KY=C*D<;f-6UcB~J1A`NO@&!K0vdAOzBOc4OiCam-oL2`OKS9>FR4kSq<
zb`^H5>mZ3%$*M(jQ4(I#uDIkxna^CwMD)eS)Z6!)+q*6=HLrM@gDAz=EiX((=cmQ9
zN8Fy_QY1naCLuSTCAwyHR5V3iL<9ql;#db^?%$+g_YjW8IgYRY?Ae$=CEKP9Di$n;
zg*1qObRJRh(bk;`P@hj69t9`8)X33Fu@cN}=egI@xogO=nnW-=t}**_G7=QwqO)s=
zGk|VV`r%;2Ts5J+hI{*wJG<8D*GwSg;d&CKWwo`Y<<`TeN%ok;{N5+5px4%W(<jbX
z4ykXyh8f&V@9Cv}$Q)%g_J^|LyyZnP@Y2cSF}ngRxRW*#@ecu^?RsWgXY47%5}3fp
zfoypS0-XY4SVjinDxXK0Fe7sft;&x|HhmKuMp7uk-TCQtDd&`8suw$kaaMdyA4wz1
zS|^6}2dGR!dqp$GK+g%{;SWSRm(em>%vNXE4tU)P#w?p(XegG6Z0u~((|dS`wkeqv
z<dW9FMO<;EgP6u$0h}z;>RkA1$SxqTB#B}(Phx0@b)}b%KRIHlPN1rq^L%x&Yx)n1
zSr9i!&8{0)5e!#bmM`HMwwKeFmNJc(D3Yk7tEmukMRO^LXr#;*rO{t`dbu@ge8{*i
z^lD!cAKg^^9W0zfmYk8vl-jh5=~fiuKDM;Pv1hM(a~vxai_L=41klZD4BSkn5fw*y
zqAn&Tpl5m>oA#|OW&%dyLs;!RP!C~bvSF3<Y`3>^8$R6@fLzP!s?Yt@wCm9oyQt*9
zRUX=|PWLOMiR`?CW<%K`2G?Wv(r7VnTS)RiX4U1nP4+{mP;f2J(VZRzrhW6e-uoIa
zf9l;tZ=qVdRwNWQR#vExD8-eJf6`eVd+Y`kQ_gW%0}E3n_ua)yNS4Vni?6V$DDuTq
z1>fg9Ycv*ROehAm7AN5cHt-icNS~Y!b<Df~XV+gI@s5c-kqd4BNvQDf?jM`DvR+n&
zIf0Zfo-|ahf8gb(IlI5(+A;+rY(&n5%Dkd3Un|A48t$*+Z(^>D0;DNUP3*GUgM$lZ
zePo&%@WWQWdxbmjG<tS_=EE#E-Z{@3_@4UU7+dSRpd=h$Gh<LxYVFZB)g|J>O<pkb
zwe<(jO2bb7aLCJ#scWUApjvqy{aqnBmcExQ3g48pDP9ZmLXSbSmC{YcqnpeeTdXVM
zRE52SVitDBG)t3L@Wq>A%93}E+m9Smr|xdtYrHMijCaAV8AD}xK-O>28Qy;&R49qS
z;iI1+bdua1@pbTJ7AM0gzk)92$90&HeQ}e4LNygVODqwfeONdit7S9xdu-v~5df6Z
z!S-o`b7N=w3FhAygTpPQ<lO0s%A2$jBx>QyZZ6O)>Qgs8dzB5#J&PQn<Pku!%@rgR
zuuD^F5wPO#OcR<_cD2)<_|T6EV)gO4Am4iEYt4zjs~hG4DC+q^2;Ro^_)vm%StAaz
zf7Ij}FnvKg!@|-sfe&>KHN5YQ`%=~UlCC3bE++(4Tz6u?Y9;OvHWcwUYV$y~M1P}g
zJzFR@_r{ymAd$UjzAWl`SxfKIaYOeQs23n=9fWy&kcKSWQp*U;jZ`YV9PDF4*foM=
zWHX{tlufuKz{=ON_}4|0rc~9UF<E8gD0`?k;eHQBbe+a7-LueT(9XZrJ9CbqM{bu_
z;>oi|E@Y^b{N?=Mlj}wD!y}j4*U6}GIM^zv@@JUqNtAie1Fnts<!x?lBdqbR&Ep|5
zq&-7YbHE?_*R2G-(o#SjRxg?TW&d8Jff*Rt42)$iZsu$KOlt;K`v98`zG4Z!_rKH7
zq(TJ`Ejr|88>K-Jb!)Mhd8TidxQnmSn-`l7Df5J(9A*?ewJK`I_Dg-c$=6Myv|q(j
zyfZDlCc_Q4p<8)^V~31B&^I;!ST%tmFbzE3;PDl(1?2C>OEarjvwUNf_ESo=Q)ykN
zpLVZo=m(6jP5)GilXc_yP916H08M6HE}&I1z)#vPtnKa{*h3fX=GqgCVwRZKblkE3
zJi@mmG+)$^PmVl|Gg#I20rL}`sBXG+8DZr9?K$_V2iQuqtWrw7IE}Eq2D_gsgVA1g
zFy+Vk#d}GPK3{ozmbnrgwRSiDLx#(mbh&R%(UqKyOseAChs*hfdtI$bQVg;(3Ke?B
zA9q#y9j%UUX)61H?xKZk=xcdelQs4HBTPc96sEQ=-`N!GGSO6cFD7^v>8Hnp6gtSr
z9kuXzZRUl+ifg+0mi9uus~qnbi<$SX;+YcNP>{mK-N=edN*2HV2i`Z5aazjYM`hs=
zcwdOwyt)pLS%Lrk3#nvV-CW%rh5^4UYthuWF0q4tCNrZ#4Zjfs`|g~^fz&=9)*=+*
zKD~(y0s`QaZJ&HcfaOV_G+J)qrFL26YYV+19w5(T?_WfsY0XviP<pnU2D`eX6d@UD
z{gxWI;&I3{C{*I*kYhe(fFv{;AIe#sIx5~Fg16=*s{_Eil|Q|p<R!yYHEpdwi^H`+
z885*OH8jjJ{j8O0%Uk`S!1#w<Gk^V(0!D}7-J7vbqQ<F1GGAt+^95d)!(?Y;E&0K*
z-w{yA!cwyElgZN8ULlx^idhqo#{pX<W2TeMuMe(6=Ap4~W?~(nn^=($GPia@Ogeyc
z8H%=f$EkH)BU<5CQZ4lTBr1pRS&-(l+I)zGH6I!q2YCr>00EWFI{Z?yGWH+ITQ8Lm
zMUFlwMPaB59?wv6u#?KNXr+`nl`x))QGWkp#Qu}R`Y1s%Y%XB`jZy{u7HOr>nB-R~
zQAi1iQk}kDG!0DX%ER;P67RWDGliPD^ba8WJ66|lj;BFfqk85QBtXx8MN@%&HE$^<
z7Zc~IuSc%PMG%ppN{$K^6D;f|VZy2rFE*!ilK1w`J7<ca+NHWUok5XCIi>!0y((E0
zCL_7*pkmy(exe-X>A9nuhsP-7Dt<E$M7L<vwoo?C@Bg3FMld+bLD#gGm0&m*MWfu|
zLxP!g8odiolC6a05mIwR3@SIFe5=T(iuBSKwC6qm<9uMlD(c=WF|*{zcc{b-X)X~5
zVt#nyXhXbedGPP6?*o?fKd=#o0%|JeHwrrOy`f8%v8VXJ=_s@ASM%(G19AbG;{y0Y
z80yK11U!%&XM|nJR`})9WHBSxE4VeNfHkXNWP<9ay`vpG_RA!3fo!QF8_k!kq;~k4
zkCTu5hgE-(9OB<BT|X?RrqsHI|Fy$}*pwU^*j~A_Vd9e#?e)QzX06~g85Ep7u@dqT
zXwWUv{8c)UY3Fcj(1pN=yjDvqroM>*&81D4&`}Yvt$SrUGfw6{%9RB0V>Ui@hiiYH
z1nazOEYEPecvN0_B#Nm*&%{K!E%uFTDPs{5$NaAIG56!3hH~mKSuSYnnn#P;9+iY+
zK&H__l2D66VhwVCpZz>Q(pXVkjZx)5A<Q}I1U|T-OjeGHt}h-%*fv>`VV7k?aDr55
zx6Pna|HPK@l+i^EO%9Fm_%Q92sdYL%7K=)PL{DpQ9MaUGEGkY@ptiE^k6&VJG4_D2
zd;c;(j&zJ&$|svL@9Zr_WQ9lUhb&-M5y?On##_9gB@B&1X4IBFl9HD}!FNMj{`HHu
zfcSR}^+iB%E)D0Ammx@`6$znye)*tSj#JIGvWhrD>jYn6Agy#(5U;sJai?3nGZ;&2
zYOIkNUb1SFR#V*y*p&DN=*<D-$g0&#+7Dz};3F)gbg<?=N#iO;A<dj>mXgoLIY-8a
z?|JJTydHFVhK8j~2=i&uTTfdbHv(SXJj2x=)9lFD@$Sue)a6vq45plF$@LuArG(S9
zG5$&03|LbrsdO~TxY#LV<BpN|YOsL@7Mnf&!r}C}WziI^mG2)v%@o0WbVXbfzbEbz
zLLcoxzz9ompb(x0A$$#@;=~)GG_nz3RoIP?cv&L9^-A)6ItpRc<w!1rcz%?l1BpRF
z5LL9w${(XPLfZ<9LPbJ-Yw*?kW*5}0+s|)44T{ENIN!|rrcSU`Ky+1^o|c=t`=C`o
zznKt*iX86YJWieoQAAOt6m#e13B+m=GRFO8DVH2UWh@45ZemcI)!{>w`)484Z|;4y
zHuoD}Ko_;<NpfTrZ+yZqo9aS~&XNibwRDA5n|yrjB~AYlFCXBHA{@dp!AeBJ;Y3v=
zuo|G8Bk^e*t(2aK!`M6!NQ+>}aw$5`=h1>jJ-fB23RJnqL1PwGg8x@+`Jb0b-;b!+
z#DS0Ms1@=0+hmt>p9Otx_@`g*)n48N>rh}~Yp`pUu1U8>TJQT6O-1bA`A)>~A(ay9
z#kyWfoz;k+^WGg_5;}K1BekVgvf-G_9D68OkfZ$GN^ap}WtEPoURnD+{pSyoRFi_1
zO6F9|o~tB*?3ioCL<S5>qEa3&#~9CuHTF3?p32!*Zr;fQzX9{kxl2~xW>B&Ad`81Z
z%Bp2&*0*P7+kFWsNO7-vA$BH3vuT~(-7vKY9`ba#Lm?m4k?t&$B0`f#Gpdl9u50or
z`Yf3=<`J9z#s-XEkFU|>dj7#*-Yil=;K<@bp)F8ae>9tsJ!a(IzQ>;2eyGnDG_F;{
zf-Q{<KJ7lhl(&+6G2AA^LLZ*jT{JR;HRklO)4Q-rcYDdJphIZwTo~u?xt3FkHkFti
z6+r^o+XwfqQM=(>!8gFa|HR4~Ni0Gi{w4Jpf%=}_C|u|AxH)76>KM<k4DKoivHT7<
z5^~?Imo3_1M&T{r^NpavgpXP8@1{E<sUz0(P1EaCWo>hV-%AndqC~(ZVI)J2;4VWK
ziwS0(eCo?VfNOFkPcci_uwKtHQbFU*Jw+AkN0eDwMSPw7`Lkge#V{Cr1)iLH9=$M~
zn2>q?0lrR5R{j@VuDpx6lMh!MHUdGP0`_#Xw_x`cQ4d0ql9Z-OgUtwTv_?i)tV5!U
zC7yD!KG@*GF2?KphZi`TRCBcOM#-Ok?hQlt`yrj6WLr#nTB>+_hni~MKp1zpM@Vw&
zs;H%q=#+t=agkJ|I+|2bB%NyN4`TJeYv9%y!)Y8w=%)%U)%xv=hVFW05Aog23rcl@
zZxdO%in~9OY@M?GLQqQm%u<?D&=JGeDqt#3RhizE!N13R$jK?b{krWO)Xrp@YcfEs
z@ru36eC`Syo)^ZV*rJW$&jOoTSzE`b*2-$&XB&*$@r-eqBzvrlDx@@tFcHv3(8MAD
z!se_>oD{$G2Z&qDP@1v4?<Qd<zCBsxvEwD6Ua9|$Q}WUTg1nzTB+&rPeQm|YkC*d*
z6FYPL(qlu4z+u1A=umxgLh;kBahSez`(`#Q&Cx1TT~pMmg_o4Ofu+Ck3b9$AEBM7i
z>IO0#pd=V;PqTvC)*hPb8Hvbu-aZG5&_6i#{BEXbqG%9A=f0l+r*GeFVab2%sH=#`
z{G)&bTm9fY|2CHQF3fRZbcuwvn`^N`nHZ18ped3(vbv^DInt<St^!*+MRaWK7WiY~
z;4Ap&1MBnct_hEP<n>=-8f>EzcqM4_j<+oryao9P<*F>$4VW_YuHz3xnz;4}^)}Ru
znmYO`cXOfFgEjb2M8zuE9to-m%$#XeVwKw*RVZKoA`7ljyH~hsZEJG=cwDL~qh0nl
zxAZ@JS2%ti>Yh4kdKyW%Pgf_F`r)~p`IdL}<^eo%#W5viJcx~_5g8fO4!KSS?0o?=
zZ9s>pV@D-ZuM_my#pXYL*bR&*aRAES(&7DfDk0V7s0#~7Nwm|viMOpscYJwXv6@vo
z>~n?`E@C^^hk$|e5WcS|GVC-2DZ|{q8kqQM+6Dp)lIc*NUqVEueIE0?la(5Ih}93w
z-#z{)Xy7|_5AZ;dN-4v!<}YKhp|kH{(ER<KQ_fi2ZuFJEM-|1<_{(Joo&_}nQ^`Bq
z{<JI#=x2Q9Ul`UxGTFn}_BvG-9|I2Rgkl?ta44Ev;5D#NHL1}EJs+ffV&)=pd<tAQ
zeIQZXoDvAIb@WMpCVDkqT?yGq<W0D-r2fuCp^uZ>k5rTX67jwo32$nh+g#0q3IEDX
z;KF2k$)UC%HDM{8-FPB|Mh*#Qu`xRitFO}^2?!;ZB~hx9pa9-1aP_zcrtO7;9{4P0
zj(1@w)EI@*g7o&NyDM5YmzOVau!lDcWNaGI{H?^%rb+<3SwXve$*Lw2E&Zi2TZL%?
zGz;N4!w2N0dx^-!aJlUr*%c%lP}`HH5q(pTmob=E!gI7vwo5IM&L<E;7l_5S9tN($
z!F>DNdq#|Pu|6abbOl8nTSa&#zU+wNYi(>Yi%+=8U-Uo4ab3BNK{6!2Mhb@Y4_qh>
zv7#)=94X@?So=Z|hsGkV#xT-YjdIM~v|D6iP;^PES#fCM@8=AP(!QrLgy*W0pljm%
z=C0Bj(Tg=8WfJ8r)v3w5ALI3ea<T)PdPe$8B5={zX+c139XV2245V0A-7KNtpu8)g
zVjg;plJF+eGY^5RTMJb?U}hlIOvm@~CB7*>%x*V=nkJRhL7#%%PITT-9j1z!zy2Tz
z??isqBTqjcU4#?^>OS<6sI)yi<pcfom3>I@2<Y^DmJ(ysMymu}|EIH{l>GNqN2Mx)
z<yAq2`F&;jV-Gd7?_DQEk{v`e9;!>w9-Jvc&4yr~-o7!M`C0MR0E#-DSen-CQ4U9W
zs_qdz)6frVrNenP_C{wiG)BZ4k@9|_)%<V-q6CwI;G6b<z|niFaSu$P(0cmDhiHgq
zzbkX;`ut>h*x8w7V$#T)OtdRxq2YyB_{eM3`o~r$360lRa(=Jans}WaR6gmBkod5V
z0%7sX!&Lee>dgcvANA<?>(|_$zlz0#ZVS8jRubtzSeR;6<P46ib97O*q{}CbebE3{
ztP=$L>KtWvkY|%2PM32WTkiXOp;W!x6lmxkW9hFeB(k`!j!^(h0xm^`gD0Hf2&1iC
z7JSmok2GT4$t%^J_`Yn6(lqrNjnpqyx+DCBqo#`s6+7DB9L?+UXyQp=Dv2WTw2u`>
z3(x;vPu7`c;f^(U%CcFzZ;)NC);`y|zMgZ`Gz}%>X5GJ>+^wvN@4hL6t}~`q04Wim
zj=Y3fysrhy{c2*84Zl?FaHiEmw4=b@+RevmjyrDau`};&bJ3~xhL*u2*Q8Z8>8c2G
zhnW6!I|9CpvJoJtHNJ}0s=7`omEhZz)#O#5SP4eT;GAnrAC<KzvpT@WvUu7Gu>0na
z?7Tis0JvjHN4`&@D;cX1j8n~NJdj~V>~_X{Cwb84!<A9kKq{}kc&8JVj>%-F+Vj&|
zs}pOWs-w#!4-qFKrxNm*I77YRy6~N2lqpsmvy-wdXz%q|e6o=;{9#uA%@p3QFN$aR
zq%{-}&^GdyamYyjQxMhR{Bd^2oy*nSL{)RVx>&6MUm#*XS6RNB&p0(|TV(iJx`Fak
z)Wd9c6#Wp%WOLd2$&12do=9A~4@>rI)%KGlia})Qyebt%-Lp@@-T-sd$hsW~@O*>x
z?&7qu$GhK-uYCE8?KaQTe8$uAF7Jc!fD+O_DXP4Jp$*GQ7Ap}IMSbh^hxE8Cj(H`+
z>`IH^M{bz!*h8uKfn$FFfOd}g*(3VR_&>5g4?jsvN0ue;L(i0FE-)=5MCRskNnG<D
z%gbKKOvsTtC$ez$S@NT<vflYAkN<(FR3PGOqKq?Lw1Xlsbvb*j%r|I>`L5{ui*RYE
zm<fw=U>~K592r?&awTSCyCne6?CmE;G92@ToZ15u1Pivy4CI3ugW2N<TM|)o#I2$d
zj4q^ZVdw%>K_cTAKKO3}GBKV{zmnXF66|t|8oIQcLqX~K@&s7s#1a<%&YOGq#)PNR
zsrO)Q8=dm!_Yo?6M`fML|Mg6d@=6MVzfdq>{~d14azcEB!Y*o^pE6ZmGK-j=dEf7D
z4#d;SB-ScS@6E`Zw8n2Fi|8dCSX*N`^BuMmEgNO{>ql--oDQjzPo8Fv*^5vx+-;jW
z+K*@wtLFS#U#B>g;(Vwag`*3~haGO<*zunban!jvC%8QKy>YXWWUc!8rk0No{F4SQ
zqUM*}C4*B7b&(Rq-$GC&W;R0og1bn^lw`>rl2>gW&)3s}fUL@B?j>{f{hXc6H#hk6
zAAW{nGFt*7a4oMmhp=$In}^Wh(?hS0bBRvnI0!=ZjZW?-7~8`tE|yLuysZwFE}Q6v
zStAO2h~5j2a&&a8Yi~(x{FXOj7(c?BX~16gi)O+-VrG!0j7zf8#z-}~h|v0AbO<Cm
ztF5PWl5H)_7`3jjC>5RURV7`!T*MSc%TG3(ZasFpB>iQncJAE6{(k#|E#}vDJct_$
zUtGjH9V(e4Y%Gj4Qc8p+K)lu`Z3oohBc{l2epCok8$cA>vQ6jZ;`uNrt1&SNk=$Hf
z>Ced6LwPYM<pe2+OoYaGiRd%ce4;#F(B(67z@6GTcdlKNs}N(33aeKOR=d~PlLbuB
z`u_m%Ko7sl_pe5aMVyJ`pZGZ;&|2X!IGb^DyiYs5h}IEXdxu1Ehm8xD>FSt!ci&|`
zpVH|rA&c;O*5pi&59=u8n9+31U}cS2mb15i#Qwn{Qb!nT8IMOC9&WR?)+dTI)>cd=
zBeu66@#0HW9YNJ|UtLoMQ{c>kOJ}MITLlRUUF*~|84J8;b~54DfBiT4lW)GmXRoi&
z?aZ0Zk83#)x3~-iO$i`0lO*w}HeBNnd0!8D7GZ4ASq-D7aqv%${garBCo!K-H0mto
zTZde``hLva5B=RwhWQb3+-Bp#WjZ?M{!fJYQH{B&G2dZrt@i}xJKI}4|3W?BS@<x?
z#MNa)>i1S-?y3q`V=kg7BrSYM=;A%|@r?iOzx$u~lW)GuXRj}R0OkwO>x(3eS^9C9
zE2R;lX*zg1m`^4kRnVM@;{?^}QWRr;@a7|4|C4*<bHTawKJBE-d|tA@cg%(JYYQ#9
z@SM)b1k@a=`r|FK5Fs(*YZ3igTpCLp3w#CHY>Xm|8QMQMLCA=~@(M%==HWf6`HXhA
zPh}0hDk-X*Y%;;ximaG&G8vO}x{Rke`-dkS9*xjCqAV+Bvnk`rA^m<Etp)S>6k`;d
z?gTx2kxCcnuOvJlYhvZt^>9!sWNKJ-mNoB4j71xZP>MpzAaURvZ8Xcv%PcQ1^UZJn
z5qI9Y%jd2S>9liZ*`y}ki^Y_b=DCTEKxs*`pf3bv`-Yk?y$@xWT8A=<BD7Xd1@p>4
znKj*hfux=16}vk}O!j7Qv(55~0u*IonT+QwErsR<DMF)DLrG{#t*4!^HRd&+t|`O%
zdUvQ=czomBd0MTM+3`NulF=l`JH>Eml}^&)`1p`KFX%5Vk<Dl1#SCvM=6Oz;&8W%(
zQ_gAidI)Fmro_9Dag;*hOv!jMW@V*=z|(GbId|?Nli8H@Gj^!4!Gh)@ZXCVftU4R2
zPD%ywVFcmA^^&Tn*x5ed(e{jv(sVlmN@KxU+VPo8zpJRKoJQcbIorXugkW${(Du#9
z$sncVsi$&$!gcrdh)Aax>%mJ*<;b#vdv|s@+P}x>*kCJ1ReBJHgZ(i@QPJ(TIc*p@
z4Q3lqgVYy+D&T{7uaZHCzpM;MMcPVfwGu|cFdHAy>$HfJW;B_R6&0P{fQ_Xk_IDp~
za&kng-9vdpRh5*Lp%4W{8Rq|vXCu19ewe>{GG=%0i2h&=oept!%5*ZK+tak#Eu42`
zS;o=9K6y4HNzzafwpHlFuH{1-uDmm$G`zjF%Xl=V-${r>C?P8qreG9B$?yN(w|VF7
zTReB6&!D9kPiGiYpw*K>_9^u&*!CtE7hdAksbo!K@0)CL6U={<_3>uEy1X7w<N{e2
z&AiC?#vgo-Y%GwrQxEahI4ih!cRP%aib7i1dr>!<G`x#5a{%E~bE!54zK*MehbW3U
zfBphn_eT`@oXKRw?(P9qRnqSbSY6%Z(&i<uTzi4N{cQx0#%-k3jAtiQWr-<E4)%9=
z@bC`V=$M23L$<dk?2Rf0LxpsTb{ZpEQ4k2uGRhKWlZ^M?dz-)X%PvIB##5Chlw6b!
zpG5@yvdS2$vZAskjQS9ug+MQ|ZjQI#d6(b+=C^qCV2UX!&h=r?F{sFgVZY8}y$*~G
z7va3Y78Ud97`eU?GGIb2E_f+i-M|qp2<HRO@BJCDpJX|gVJ%M)bM=%k@3XwR$@$I8
zT)g@M2M60oAc@)trJ2;2+ZywS3(Sw$-X7POcYh@2w{E@7&;G)Hh>{9#p)vt;L@1*R
zA22t@V9#K#1gP3+^)=@2zV}|h{NV&slplaO0dw!cYrHqu()>HY99PwoKrALye5ojS
z_trf|qdC4xI69b;oa=xUJbrY*Yp=aPGY``HP(pIy?n9}DJ1cOa(Si~o!$p00Nr^EA
zDFrLb%bXm}D2sxT88Mm8kV>)CAE4GY`KL>pbXRZke04zOVImUbRIsF{n0!W_WhBRY
z#M7Md{t3GWCmbBjIGh-UOA6@~vDPT9amF#PGL8<8(Qy}la)&-Y!yz!#@N=<61Am|-
z2Qj>7p^xBY<B6&&%<lA#7GQKtsUliS9nN1k&&#iVn*ZT9euKM@=SXYV?0LFvi;x}_
zMZ_(O^BS)uLV9#Gz!^t2pRuyMQVaV^oD8mzwe^$+5p?Q;krz(|^Oco0TMv(dEJJXm
zv8LkT-7N;rV@G35yN!-Qmg4Ae%%#hl3t6WCJPSvzX^0ARy4q>2GZsy%u5ksF(hP?~
zCZm$g&GTgKl)deTq-l#tOWLgtD=V9<Zd~H{_z-LrGM3IzlqFSFl0-4n$r$epE}K!z
za*9b#J|{T+R-mOKO<GjOGtWwHKK~ga9dmeaz>S;1fpfOz_Cjz*Ynv6GzYX2Ut|^t(
zWzkwlJuh+|J$}rit;Zak6zrELh;mGxR}?NscU7?ME=XFGg~fS;w~ivuLImi=A}i;l
zL)g$!8eXq)h<`XS@{=rMGK3G534x-@9aZIc>+QQZ6XU8BEQ?vS`+Fy3^9r{l>*gF+
zJDH(uxVX>j_ZufExes024j($vy+;c{nzp%c;R>!AW6I#%PR3IrosjffY;IoQ^3@w$
zy7>~vyW5mij`xz%2<Gz{&Kjz+U_KkMyYrY?Hs*LVVsCH8_CZFLx7a*Cq?2m2j=@`s
zyuezR&q}r)JtWI!^ae}d9HunjgRG3ywRk*$b4;gG4vvq=^8({6?aGiu0#pcOc6N68
z-uHjN<DFxQ!m->kBys}2pwnq%tw;L=Uzg8?18+iOPp8#J$q0|cx!|lTAwyH12<81I
z=>MSP$Oo;D>C`zYSB9FtP=O<2VgKNmSaeBMx`<qa(CqCVQxp|R8lSmv0tvKP3y<{7
z<=zl$8wrPywpXM{n<$Pjw&d_=kKMgp=Chou7rS&?U346Q)I4|NHCn9>kxprMhs=v9
zRap{AK{lIkxWCI}dd%&6xA>z!{!@w{+~VYDg0+JAd_<gjhQkD{A&O&q{gk8Q9h|Gs
zN`SK=!z9813mF-^5ZMiL`$Rn(&LUONqw(HT8AGSr<M03d|1<Bs_ZGkXTmO`Y+jmJW
zDN;O6O00#VuytpdAQby+z-n*4;V1>tszk})3<XIEw-7`Aq{}~nxkkwdSA87Jn``^&
zVt(afk9Mnzjx>TUSD$;0H142vLaWmwFXvQcD8=TpDaVJqOlQa3U0{BjlcQ;k`MAbB
zeiHMe9lWy$6~uMlk2xQLIVN<&HN(LxYv^?Q{Jp>T_jvEUcle!u{?BU6<r&O9MNvyb
zxOz_Rzbnjrkf{*DGo6;iZO_4EiWU-l1j3OQCDsH&`_bWus<K2;RErS%#l_uR_-APW
zO^J&$m}?cIbrgh9b?8{0XH%@5F&a(KGG%3T19$EULJmp2=X*R)cF@H>DzNW;1GV-!
zSW78)OX8cb{OV)uc#r4r+@|%tH<{gf$idzWk08q?M6qYMl%N!ND!QGR{gZvH4^v<U
zu|n2k?K1Ca<2ffhmzy!fDj5GJ8zGHBYezIph?0nFm(KAY{U?8ux8C?Zx4-*sw)Y=#
z<(x$8(9K#D)#98Zz=)`h;sfSF$S`=xd%O%>K&Ry5vS~P9&VBMresr7%WgO}flTZj}
z5Oaf;F_WEL`u#q7n1GbbrgN;Z$XJKWtZ%Sx@IHh2Sw~(7sLP5?=1vG1gr&7cDotm&
z#ITqngd&a-BCUhGP874cyoSl<=r|>cTNrDwrebY%%E|E|N<xxG9Go0+|G{G>vz+7O
zBWB}*qr(}ctyqfsTz~FUtZ!~&Do>tI2xAumN3ZE77Bx?HG5Fn|cJMbP$FteSD2nLz
z`h5NCzsmi4xB14u{T+7p_bH7-7>6+`bbOmY-!bO2lRji*n&;#fj-8Mm=})tDO;*f5
z1;@@ihw&y%5)lexLhwDF&Cqx>3W;t$A)}KS**u(krNo(YUhl~>7&kMC7lDTdV*~Xe
zbZ<w*NlKcwC`D*Io96|Hn3a`PR#wk3T-!tn#mdH2P@2lXJon7zQ_3O_w$>IrH#wrr
zN9^r9;;pye#lCfy<4MkBREDM`?dfzSaV+Wv9LX#jQB^t4R?T3z)BiYE+g~4&8B%fK
z;zj23Iir&!vb+jfy|vv<(-wd2ul;pK(}G*?-eTw8J9LC4jvUqo%1%`U+un%9FB(c}
zo9NC2LZSLsXF**pbK=}bKIb2kj3IE&)B=?nAvK|{Z1EUQ75pr%w&2p`bpV_VBC^&}
z|6X`_&EV>VYB!u7dL%;l1>xSBioM+fjt)n(S}8(QWb=Y3idkLVAkRzo55_Dltx^~`
z+S(>fQ#N%<l(bPQK@!sHgC&QGT0E!Q#O)r#r4=4-J>tRRW8S@c$Z$C!Eh=;svPH{F
zOALogi|GV7i^MOQA;LX>dhjA0g-HlaX|Co|Nn<d^k|u3le)(0NfBq&9?%(I%K6t=v
zCeXs;Wre9c^I3+irwBCBS<_t`^dC-j-9ooi$dcHGsQ&?F?w@Q4yfgSllvHTMPlP#m
z+{eazcY%594CWDQtLs#G$>HIc!C-}=bQ~XTlcou4Dke%=C>e(sG8OcP4jHw$b|YZE
z^dXo>XE0w{T0$!sI=B1}V}9D{Dv&xNR_avR#Ccq0aJpJx{;8K<;l|CIJbch#PK~))
zU~U$e*CkjP#u14>OU!*E+2gSW=F^gslN>1o5iL>`s0w+W(O+tF?)+-N+y?>8FqB@S
zMEDdmrS|5(X}C*G_5rj`h@&>P%GuvP;BfyCkHQ*BRXSSznBj1VwY=n2RWN$j(~mW2
z&r=LMw=dtI*v;|#Yeb2|9j2IGLUM6{Z1?%f@_^OVcewp%i@p63kGDq*h7m~?0=2<X
z&@+_%9(!A!&W(hLvOKVkt6uOb@L4>&c)En_Q>G)@HRjR<^?z+GYtfKyC#64ZF_@-Y
zxps|WHsSl<+-9CZC-pdE$g_eh%dyfe&H#00Q`95CM0Lv3lv1G;7{GeB*e6e8Wx2r<
zE_m-1(nA+dPlVD?KAAJx*=4nzlC)EnmU@fdHO7Q9E<%<eEOF+XpJwgC=q=gg_7>-8
zHl34aB`OY?Aun6>2a-;^Lzd;7h!fJZ!)!9f6gizvk6vFBw^~F|LZ0Pl9g*i5I!;(#
z-azWO7WNIAq(Ve=x@}&1<uz7THW5P6Zb!6Q?RrX!3#IYJCCehX5`;`tq$KK0M$!xF
zkz^szi*(G!#yQ&U7H_}(7I$vHM^S<@P?SN_&XyHs(JddQ?noiS3{z@xUus8B1o7@5
zp>%rJJT2_K59L55B}v*s5IR;Z6{_h36plC%tga7;;!rIxCQvK{VOD#vgFa+6>yBq*
z0=G6yp9<V_rDEc^g>!JQANaVY%7Ki@B*QtyV6cSuf_z#KxB3VK#cYlPlC+1?OO&Rh
zC`#tD8BxchqL>YtkhFRXhRba49b#pd2ip^*v#8wR9Vn^jblUWLJ+xK`XBRVfUCn**
z&V^|c9)$EnQN*A>q}S`=oMm);OjTK=w#e9`Rm_D8mw9oxhK}3p?QSz49fKEC6<7-<
z4`bmP-aYkxT5!voV7w`_`*4n&Hy_)5|CqC1Lstt?*nCPPq3m{sFfN3o+ZFusU;YAL
z{^~1u@2E`Ba}%gV+iu{AHN^1d>(tmcqdG#js6(k39ktlq9RXunX`5bmfD}1vYwIL&
zLa#gI;Aq6j=!jUupq;X1O4iPuCy7(WlVe7Q2OJ(9B2+@!>vH+p4JP9$lldW~hqrGZ
zad0?i&~Js&CWjL$GvVdWd_5>rS4B|E_X3Apl#^slW({&D9$CwsI3JYYjkQo#2;jME
z*C|cOqpka>SWyYhEOYeQ1ZU7YY-z~m^Kj^V=#Ub@9s*(3rmG{oNb7L1dxsU&S)8Xd
zGOBh;)fvpihhqLkzWmjfpDO0jkH9>N6S|!thle9ZlVcL?>2+FISFyTwo;XU$=EsbV
z4>&nGgc|b;@5lVk5eJ7EgMNyw48`Gu(oA^iGhYwUtSxcIz>mS)2W902UaoEukqE9_
zx=iINws#)Ym@8(Pr`Hahe$HELY00uv%>5b6#h)eS_4pj2L)Y>=s|eAy!8Nmcy+Fu{
zuYUCvKKHp7f}yWWt?8uBT<Enhw@~$EetPLQd^?1%g~$a0jTA9EyCbU75l1QK&z&QV
z5{82nPTD>0Btw$;n5Fq6qCt!Dl@80TA=}?s!M!fwFCTEFyUlCGgncBB21B;>Ri2y9
z*&iKJ`2i2NMjRd&bUFz|VVKPf4<77s{rVTUaMAIS%2_EKu@ABhYcfp!7>at%ZC!Gk
zBhk(gIgi3ojK_@k4mdiV@IU>Fe}&BpHr6lWm7+wE<(@RQ2o;99mqmq98l)se)Pvgv
zC=niuXofKeBCQru8<ucE>1J%)Ctkrr@j@V~Iek}#l7CY#a-P}lA;Xm}uYUduT)lQ7
zBobUGK?&t+e&=bR7QR%?zY_KPtk1n{(61$Fb<B7gMlHorOwwMV$i}4YZpips%j1U+
zDNV^>&?PUX%y#E=`U9-793Ad4pN^@jlA_ERtgO)P^k}zR=m>_(JvPr>q&Hk)ZR0%F
zS?03>YYR5dzW|;fkn3E%N97Yb$qj|T({%RMU8_DgH(~avV0C$g(v(c5BZL$f4@Chn
zVWNunVf<TFmi5E}6PzPae-SuTP0Ci)O$XlDFi=9CDl9x5%t2~FzuzTl#S|1cUxk2O
z1j(j$s`&J0uko4B-Jsv^)|qJws5(e%?D%OFA(Yz{Rpr5(5ag){a3o191aV1+qoabu
zgIyx6=yy8^;TiOMByoa{V#ebU<C9~Au=Ls~smmd5gSRN3ksoa{Jw8Im5|O5KSJqe_
z&u|hdm+)wJ%4Aj%M*{DlGC7AwIp;3D8U_Mb6NH{UbyX}-rbJCQl0wpMwQD7UoJdD_
zA*d>ciVTs5Uatc}5yufKQaGt8jUcY~c&I*k%A&xU3OLQoh7qWKAp;}py(WafnjnB3
z9DDJRG5=UjU8fC`6Yv^@5=5$t5;4YSA+mR#cH8r-fBAEK?dLyBlBQI)!Wx6p4cK)@
zl+>pOA3>lg)SL@W)Aiw!S0Gd3bbz&js;uy?psGA;>q9m+*74r+2fzPs7|lz*^WE2Z
z@ww}~aP2CWE}v((x{Okay}fPTd;4u>vy!`yw>g<l*|>0?^`#yg8|Rr7J0!87GLk52
z(dtJS=ZVuEFTeB}*?b<Hy1M;5h_luNc2#Sv(Ks!2=d1UGLDv#z9mk^yhleAsT)o7~
z$`Thg&l9&=jI$9NQq=Mnp<8k`8wX<Uf=nA)^YP6ThG6u)1Ze>$aZc4FU(Q6$Kly5R
zU~>E+nDarHx2SBLqJL_bORBQOyOOFJv$j5Dy~g|xzY#G1)9=2{3pM5!E}v(x6foaE
z*yc`+`MnzR^$X|OSn9FaU>->-<7&(e#(4&5kC$G0jWWwHrU=orR0t6cH6Mbxe;?-F
zc}~Vtj*iA$zI>kH(tz^|%*Sj5LSl`%V>X+FNd*mePKrfYM*kUMp0)(jk-%#O>NI$>
z-rx!2a$2e4%U^ksFMQ!sv^xRwK#rED4xC^Si3S3A;8!*!H$gM&x6Vchg>ey<4tW+7
z+`TtUXBo>YOAJ?*v7-?$eE*vq=Z5jU+w_jE^Evw*hnLT@w-Pg7gID7Ry!wOpC?+|@
z!4at`SYGbaPgB;{Hpt3-k|=bXrD@PYTV5TIwg)_S;{`4!CH>M6SQFt=oja(cK-Nn0
z3t<i+u~K54C-o2|ip`5_T)n=EbUk)w_jvf|0UPV<$T(qIOc~Z^psEaIRnan_rKIPB
z^Dd;Iw1P<22A+DpZ#{L`xlo3+N)c)OR2=y@5>zvfprXs1uCf#;#<mP2-n=Kmajv7d
zyxb*?;?pTBbw;_-rIX?`8`lJE3x}>g`%UI9%pX<gqzfpVwHWDerKQtfraM?hNXf$o
z_ZiP}ws*EUx4FUQ+6HTDtMqz(OjR+R9I-c>lIIl<9`9kqEe0z?oGIycyNF~!CDv%Q
z2k5AcR%HlQb;9Q6d8(=kPMxo{C+iCx!D%DQ;yu^FRoxNk>K&Zr1+&?l;jl-i(`L}`
zBO^^#SyUVxypY6q<ar((V!s#!RtIka<?EnHgvtfp<76nus`@d?r{L4ITHq@|Z_uIN
zO)<8htYV;|(d$Yg&u2gT9H0Nf3#@G{BkQtjQ!Nmp(Hw2M{loV{f<6j1&<DrG6S}+O
zBtfZ|yl_lrIgcOj({3eXS;@x6GQD1q#Cq?IKV|pugg4%JlfiJv%COJ#FWg}D%2ia9
zl1)$8zJG_~;|WKjBc?@0+HJGCJY;SCJSXFvSV@!+C>4>mK}P}|rChyw4J{;9Q7$^Q
zLnW#<up&saSZk?_B}rnEIHsMZL6g%|cppYfjwcfag8}WnVs&MOPOrmsPho32T^CX*
zifl0tJ!G;&W0X?0T#=P%CF{H0gAY7AuL7r2eZ(muAIqt0cT+|uV}wY95T=ZfB0;ze
z={<2$@t1z_HGcMMpP|!D>x@otO4K49|KZt$P}UM5v%_h$CkdS)FMsAMy!pnrIoR7`
zG|6bSJ(tpyrKKf$y&kW8`ZX*O8yg#}47wzB+psZ)d^TZcdkf>?>dhC}IvNvKhIZ28
z?H|0!<qMniyDi#n$#h=fy`$akva@%<($Xb1H!mzWOCMC`YD#j=XInUXKsbe-!oeFq
z_yKnxZ1aP+-{Peguh4F_Nm>chBTs20NnK@g9>$Z&qGh}}NTKsh;=MQ>v*kni(>aF@
zmA_LC#wT2zPRi)`LonxqFjpa)<sAz7Q^4GDDNPwJEz$1vc<IxxVf~Ysx7uy2H5A!|
zz3nZmhs!sfUtr$(K+FTKr<h-6WAj3Au-=DxQxN|sm>UN_c;{X2KG@=&_uk>f=RXYd
zz&n{vCPChVQw&utkck$EKLgC$t(eJZ5=Ml{TI5)0oU4#{bX4)PUwesP{H4#+?j$HF
zv9$v$bjZqynoHF<>}UB_u5n%(4i2HT7{@IpvpH99e42OPdV|r)K6zHM<Qz$ou&X5B
zh-2i%%dF91eYnBu@_^Q`!>uHy>Su`S1+R>b$*pI<y~@;&Ag@?S+C04bF1=QVZaXDT
z!ZR#N&v2<vSy~1wed6^^G@zV2<puhnr6{4k$bp&@Iv0ZX5VW=6rc@Xmv)&%^mCyeb
z?myh)xvMWAW67;Izl)4wW-~BG5Q$Kps)`aVpxteOuO~DBCnfZ(B#BRrw69B*N-C80
zXjf;)_~e);QAFkodZJ*U4cLg%#YkTw@hD}vdgCIO&#jQ{@6jF(5v`=oGKJSaZGt-0
zq6_#oWz3p8d)7$`=_k%zyvD)a7W3JdqA=(vG(B}XZQ7j<o$Ec25&eFjVlqVwg)1%j
zG{ZZ~$<c^>USTX)FX(o=wBnT85AV}%x9F{|Q^^^58RV7XC}BLEF>JRP42BDwX(+&i
z{|-+5DeXZBffpX-MH7Dp=LXo`+2!!~gmg#I?F0=)r4@zoBnm3yku^O!&oe4h(F|{I
z0-(mW2dY!xS}N~xPU2Lp_eiZ{_X*e1atkdw#Bm2DJ)Q_@W1PzXL8L7&zjBdJfA%@n
zH->>HeWrYP%9*U4b-^N7LwM3whv{^TF%`zSFbP{Jk~k)cV_tmuGwkm_qN+;9Cu6J)
z)R%UrP25V^ymXPJ%?qq=Ucg&}DKh$l0r`ARw<DOH9OJxWX?>k(S<;JRJf6LsJwElq
z3nWoRtEHJ|0nj*3D9Vy{r%S&-B&yXSn&*#bk_@2<#5u=oJ|{0rqD<57r06InO5$2+
zPO!DT&!epaq?Ek;$}3c*A&w&|uP{zvtq2KGFUYe3YwY3~Hs#mQIW2J3200t&09Wh#
zHX5INWZV9+oVv-e$9sddhDgeqIG^H7N~|?s_`-F5;g^1fZnuk8p-W3Sh4Y3+ar{g&
z=Cs6gCS?!JCpg_WqE0_vxO|hZ{qo=BzyJ0B8tft3`EWrd3BzF!!nJ5Fp1(jj4V_R3
zi||mD8O~Z(R@YcxU+1rU{g;@G$DEv;(3Xa^)n#VIjKLtGG(iU`j#7^H4L|qQPm|QM
z7$K_Pv!v-%3UlD>DbbZD-@B}xnk4S?SAOMHoP~e(oBy1>y(5;EmiY8%U*q*}e3Mx|
zBi1Ttp+RWd&ojcX-|C4&<n`(EYaEW#R_Le0w*VN|m^)_zVZXuLro`%}kNNq_H~G0=
z{+s;cU;nSc9zB8ipg&}7t&Q=T^XD#5&T7nMI5bs};jLw9b&b`H4Zi;MUt&5QGyWmW
zV@gwCOi3IqakOvw*{{C#1m@xW$wi3@7vwj*tpERqU>@rZzxeYn@beD-zyIR5IXF0G
z`A1+r&*tPsj&rqajtE_O5`mWr;e(LtPZRUWICNSCk?260mm)^uh?VB^pS#Lm{i|PN
zWqAoDYcjVA`g4npTp>dly}r_$CaLD%7SHR<XJ~c0B-xm?i&y#7mwuIh`_KOeq??ck
zMYq$1rDgVe7bxGm#$zdnR&KDlGG}!tco2E2mL)!C=}sc#)oVOn-Jt*Ka}19T>1QGP
z1d_>oPPZG87Y<VymY3Gp-!HiKg%?ptm}y#xC}gPWQnMA%b`C8p^IE>dJ4Y%6Gbz~v
zo|BR{kl07}_}*{-7N#`3`toNuH#o=AaEY5QzQFxAUT0opbXDNdRDqq&@=_b8U47BR
z0hZ&k;<Ap>Qr6DoS$<t@3sj?l`N@_yf{mqwcseEWf(nfm8cBrGic6PPxOVe0aonO9
z&6(}((7SLErxY?e%L!`Aah@R1?iSsz&{P$}*={&^UP%T^Yh1he8sGiSA7Y9eVFSfQ
zM=@y{)9&@ry3Me^#N_Azr8H@;Pn>qKWgg0~&e7?1=?n)b9dWq3!+2$xIBAm%SJ)ec
z+0V{aDAgsKIi9<AfhbDq3|;bM@Fv8mGb+O6+<d>`2ZoZFB27|Wys(M0j_<$uCX>mW
zR;$gq^XItz_B)hSPGuyzjunbBr>a7B+ZaPbsgXk9)u{lk@MoPT7ZerX>0oX(tzLrj
zftVjMfiadCEd@89JI7}}`vU9h%YoA@!3$Yuw$8kP#=b+qyv|%kal&A@%E7@t)YEvB
z3X}mIMO?W29H0BruP~cCguRP5Gumkjts`{1O*(8bPzjfwi!t*VRUWD)Mk^u#O2BYw
zfJ}RAoLfWU+1c6yESE1{Lhn9idDte)!_<mcM;z?uJb&XVN+!f{5@fS{9f;J<Y2XtF
zVd8$j!%zs0PezobL2FH~*Jb7*iX(1ZU*Y<-tK7PEKQxZ%h$~mGu>IgZ*fNZlbisy{
zwZ39imJ7c8Vw9dZ!#zFoOvCrw;tu^N>*JMQJBryz;(diwDMDBzF#^YCPw}%q^D_Nj
zf)ozC0x6LyXx%iUFFcJNRn1#iaEH#Oy3b^Ggs7wBFxdEuU;QQUp5Ol$zkw@GSQ!q9
zMMR=nw9*bzCYabF`%7fw5u@=ovTh&NQX-9|r6rxEK1;&^E2}FkE%%sZQ<9`jubnU%
zO;O%Z<~dicJjZ|VYkvbBrL_|L>5~nWhq_aSGvt2}lGh85TquN+wAvl2x<&m<Kl6D;
zCr8YuhrIm!%e?iS?=vmt3_1qs1Xg-VGiN?KMn&-xk`+SLHGsveO>oZOedt&W8^H4v
zf;l=J#e5_~mL~eM#~h{ke2w`x&R`yih(xwX;|@~Bn8+bJL-N^((R2rCEm;{b7sf3x
z9}EU8)tJv0n5Q-7c7gd{{<Xh>ic)ap6Q?Qgftxt3iBLZnbEG6q+t|9y{Q1v+n#pAJ
zA((@ep32P0=Od(!7MOb%x*x+Rfgg?e({S$EZAm_@oirICm>`J|%-HNJzW$d!%i3BG
z)zpGS(6uwBZi0~zG7yaecRC9g*+_DPi2gLA2T7~LV7Sio=#cBLd>Na29=!3}47v%c
z!y)rl2Wk7ncM=#C_~8;;QO@4a&51US5s5&XoJ1+Sl$`WC?1(O|Kjcy|<L5gG<?b#?
zlG06MOp(!PMReP3%(-*Sf9_Y=ic(fxMNv0&G}#*g<bXgeCL>T^WM8Z^wv~%GRFX$;
zzKzf&CM(z*Kj6~y7io1mT)%pgNAJGJtjuYr7JL{fXl=>H`5VknMuAW19DOf9=#ZVm
z(afX_*xR~H3rdQoa=}0M!YVDj!totJesGMW2qj27y+m^T@+NUy&l)y@>Gly(yNy~I
zphVgj!*DJnVP81H`)&A>;#AV8(f4#fl^cqAfuKdZz08Am-=)aM%%&3tNsIOV03{TO
zj?k$hS=}HX&#`%lGjsCE31*go_4GO&`u!e4MP#!HD{JdGksy;ckqUA{c!#MfmX=oe
z)TcgO>y6n(a27bNO_M+)>A~VF$o=@bp&@)PiW6%2I<8#4$icxLaU8R;x`B>TirIw9
zTBH*Q?J=gHEapT>n4l4Yp0f_+sf0!IerkEgP#t&*ps%j9iDb$&D-j|h=v_t#>sU$?
zUb=aar9n!XG*y(~SXz8N+ZTj5z5hakRn3xAA>-BQ4-ir@J~;$ui0Vc^B{W(`eCp-T
zp@ic5fBbJb*}2d1phqmD5KO2RTD3{qEz<T7Tg<4Yhm;PgY>bu?1!yfvJ1J?aMcVGO
zG8j<WlH=)!PAjFIL^x+~ret+>mCt_R%M6xQNn5Q2eW3B>o2;EchUq-@UDWAyFveiA
zk~D6Si5xG&Gzu*>SFc_oNjprlf-4uV@J6?fnM^UxAVX;aYYVC>!`R@|1wGGT&jar~
z*!M6>tr=GqnhP5z@1v}b<<!0U>FfOK?E_qC5OJtKD;#a%*{lu_yN}S%T|p%-7;bpq
zCF5!247@*;=~!?Tn)Ly1UJy!;u}=8epZ^sGEy4c7w^&|Y0-&l&q6%8Q4rMjL6jMw!
zLo0zPvp_ob0&7ZaJ|j^byws>xi>lq>`1lZ8SxjMZ7V^@QRi6Lt|KUGnZF4h-aypNb
zr+lYo6Erdti|;j`q0UTM=a3?lZN}p%QKad1+Kf*QsmdydT1v@u>M%ov^$u5$PMS=|
z^ajg7&`C1J)WqB;-$Ub2I(Mp7_A~@4U;XrRAB;Jk&Fb){ia7{`QlvWJD_{F%1}!*v
z`1UF0wj?qh-A!>O!xl4)oi8w#3(N~_J|k8gln9ucc88M%<_2d4dFjY1_}~2R|BtM$
ztp^n*;qfvwyzmoXj)%!4VBT)GnNCI@g1KK{KATPG^uuRtx`@3{p-Jh-VE#0mx|=VY
z<B#@7xC%rRx*8>pwshQRmvr(mk+G=6;cNa(xMYHHl=ZXlK4h~NcU4`YSy(4=ws066
z`*gTWFOJzhJYf38FY)Pf39Dwz`uaHzkjNzmpYr(Me+yP$!oO-@qreJ}Ec1F0hGSq$
z(jq5swISBr_6g_H7Uzo+TUn~YVyt4Yv_?h5wZHlwvm?5^X*BnwU{yO_#0SY4A4*_7
z;I0r{@}9IoT?kaY9}?&I9F856vvI!5&X(uy<1Jo#tq9WteCRKo<&Nd9tebGanu;4Y
zFLB?`!WMFt%|zgpB$f_ksabW6)P=y*`?1N^ebV*$FTBXDckeK`a2{J2TsB8KPa<Go
zX2|InVU{2c^+{h;j2>?>OccZkUaER>PCc(SL~HfM-*D~1NV7(~JI_ic;|bI0gz0q3
zXgX$IRgk3Q(<Al{4!D2+0WVxXPpi`+PU>kS-XOKZnTjGSaJH;3z97P_yd$5@0_j}G
zw7V;e=Q*|to!QoSDhogJ<*(4~_ZIusi-s@uWF>YHY}Q;lhhI=KoUi}gd-A+MNr;mK
zA;Oda=R7D$UO9{ndqw(??aJ~DsiXSu)Q+7vlaZ~@cE}pl&F@V-@f6ParE>$OQ$<l!
zb!|VK-2_-NGt|ieQNDpN6-06PnZ|6_c8<DnFE}7I4TaMxjFggIe?X)pRWT0|RJE-T
zLXgBMpL*ppbYjo(_AP`jX(hE%h4rEH*E*!IU~`lbC|gkmVNZNn;mZ<RM))}8!lg^Z
z2o4VS5ME$S5R5k#N-O!rU;P!9*Ed*RT@6`Lck00Vy3QGFyF$uST~MV|-Mw1NM|qF8
zV2ndb8?*|2P;78@b~>w|6MKa52nEhriaf_!v*1lTYcW+t7#49Vt8!Z1Y8+1gk-4)U
z%c;Bd&Jp+&;T$-HM-tndP3agE@Zb+$=hD^Z7~H&yj~&v>1+ccR7cbKA{<Q4U)UwZh
zg`a5#DW866sfgN3y!NGEz}&n_IUm<tw-R9t0*lQi>_50o(r)3stOa2;v5MH*d(3Vp
z<@)s-q02~wIc?c&PClPw$`bFP)!E>G{dfK@FTM7K+HkAz^E~NHoz~<}HP#w<79zVY
za7@xPCGIXE1(>R!+vyU=F_pC#;VBA%u@Z~J`XHr|XF2U|(=5Rm2kG?xFB&p9EoYs5
z)}LhEdgt&*VD8Uge(eJ=cb^DzUt_L>M5ldT`NA(y-MogGkHcYfC0K*UVe&D@kM9tj
z#T-qfA`bT+)9IvKt}z#0;6DWO|JC3AyS(sngSkKw(U4|Cvsf^|{*1HgJ`{6El7zT3
zTwvaAwa;MA2Vm~Vi-Ode)4}KUEwd2o{V8L9|IQSAf^-I45XOvM$@O;3g`~~y8}D%O
z%2j&pR%jNf>p@|Zl|?NiE9&x9{r&<0xhcV(V(z@JHMt~(QpCe$9wT_j3GJ7^#7euM
zaF!KkQ8?U&fNnvu^q3dooLg0bL7^;6OL$-$Hzvn?sgv?M7p_ovuvL(f$mcW4JjXlF
z`uaI~D_2<g`Tv+l7d}O%JU*8cNRAcU)*kIWPI|O-tSQeff`Rb-CFf`(=m1`L=6JeN
zP&!9`JmCiq4l$)AFEieL=QitUN@Xpj_fQCo3r@PRV60(v*k`u7bSh6HAXc8l)uKW5
znR3oA&Q3EH?rCu9#v_9uMYjyYix=2GIYgBu%hJ=baPnY_-tsc-Rtu>@@K#OcOtyB=
z?GCElp=p3=vT~v>UAsEXAe5{Hxh1nn#>vSsN5{u(Z9ih3&x5Zo70a6!84TLEvY@gC
zr6g6JQ?+t@QId~Of<UZLIAKCFPb6`fVLF=-#SukSQJI2ncZib_RWZVuQ1XhBKA->M
zSJ^m!1!`&owNodh$9uWBbfFHEe2|f;y&39FT~EXjXdMx?+u$A6mLzc!MrrwQ!5a(K
zNh<HqUSOT0s!B|K!Jjsdgv=3jKq;g^*3v&h3Zx6?>?tH?dFS<q#OX3<jf@m{K}1DY
zczTs#^xi#+%}uI)AC<=FnlBe7TbyPKgCj4R80U#nVv~_=_xkuWA?)V*X9#;*$1N^g
zxyiZpC93%-XynNtn`Nq;!c2*ZDgS@o{`*O@D^KslKIfckUT@Ny%JK%b;Tvtx-P1jO
z$RVZGii(xAly+CEU!?y?At{0iEg36@xZ;vSayZjwdfLMS1(dJKs;u;{ch{YB^uxLL
z%`7y!2N<B|Mifw}LS?;qpJTq}`96=8f-KFPMsq3<Rx%k5@v{uST47;niCVKs-0zdc
z32B_5(+q&H+~!-~{bMd%eU8QDRXjfgTR2?VlTN~%9whNRXZckK$a|i%<cmd;pJfuI
z+*Gec5QHIKlsJB~5G0u(vjSmU2Spo8lBBtawVvODCp0P-5X)SAxaa~2m`@acmgn=?
zg!QK&HOe^QL@!G@BNEn?LfL@+!!CD!`U{?4Y2gJcDB;N8Uar|zJjJ9G_GFr4K1Gvr
zGENpmZaE@FuJ-I>q#~>?;#ti?Ys~T9LnhM^!oskB$nL}Yl$0d&e6$e+)iTOcq=`nS
zntS)|5`+P(8)p#GLnjHd*$k}Z+~zV%Yd-(gfBS#s>E~a;U72Q{xt$ga#JcF)k(-Om
zIak=WCzn=cGPFzSkW$STlNEXO&7>G>i4sGcI6WOLEm}hoM<me{t2IhGqUpSSNwJAe
zen{zp5t|#Af6?cQVXguH?}fROvpU5glnTGnLJ3W!F`;|#km+oUurMATvH$2kp_2HX
zk1>LvT0%*0j`@Rock#l2m5nn9@#&cVv;X7&g=^2gL{V~ct%MwNA(1M-q>P?Vt$}sF
z&S`F)K#L5Wx0u~pF~0$G>oB*j8l$x(jbpMTcEnlf%$yMxBmQA9AC6|AG|IZQtxU$I
zNVrrk6L=Lydp#b${T|mAo2YgRg+N<!KS<YBbbh%xUj#J|Mn8!+`5}fB#QF61wuQbI
zve@=GiVV$ilRLwJKZuW6jE5YbOW=RG#y7{ij4~gQ1ylsc*u%~|#*-=2$%Hf85BPfE
z@srjPah5?6gE7=A0k6FH6r#F9@a#Y1uyU3-iy-h2`H^U~KpRQcF|eRWgyvD`1b;i$
z@^4Wzl;EN#$h76Iw7ez+=Kx_&m3b_vG{S}<+GJ!F($tcq28ol5$uff=<ou<tFzNL{
zN)8jnim+JE6~_wSQP+YKbId*U<?tz^I7LX;>Is^4YD+E5<S`|w5LOWNXB<6vNTX55
ztK}Mc8lq00N%NTcx{nRQxgx+^x<)V$*d5(Ml0+H3ZlA~dJ3QWdj5gXOZ_>jGi5FH;
z!V;EajGj6kRqAo@=pH&wsg)~~MHQ($RN&*Q08b4NDE4-Dna-w^Dplevqt;kpFpUtv
zLaRpI_`LM`8?3Bta?&2Kr{c6?GF#3qor*x~w0>A|85^0O19?l-6fv1_E-xmJbhHL9
zERjv8j%uMHO;fToal+M7<ee3PmHFopP)tKhgcF~ZPBq}mz<ki1QYr~*?K)a{D4Wr+
z87(0Rtxwz?vw!0bm39+VE)g^WtOTUfW)!&>soT>6ggU9(=XYK4L!k1utE;d&9=Z!`
z(=<i;Wqh$jP^yshkMOKzI_aa+l>YG{)k>ML(g5j!@-af8Geh9}jQTy!oIj6}iZo5o
zX+oUu_2q>crAn1ozx6%NTzHCw<rRD{=SWj@4(3{6E~v|`EroId%0-1%c_5YV_`^BO
zZV0sad49I*)M<7a+@OR8PZfH#DOuvAO$5y4X`~cI&U-2_F6M}ez5hE1>I@*Qhc*e`
zc%QTSfHQ$*rgQF8X4rq{BNjK;*!s>l(Y5+K8c*`lrRRBh3&Hsa&Z}IURO3!Yq>FD>
zfK&+OVP&Dw<D+V8jK_yWS}+_AiN+&3dxuo&HQF2Nq)|$*H{f_YpmTi8{{Ajmme8nI
z=yf{yL5LrO^!rClq7gy4%<EtO4i}z&naeMH6>VG<!diKv&oz(g-I3ra&$U)hWgs9h
z`R2DdSvbq>WF!)c$*rAi&YQ~6+K^~SQj2fhpEJg~_Fxj_Z7d76Fh_SKe~g%0iWC>T
zG`S1Gmypi=Lt^d@eKDVoLpolX&2{bsKC-;RWO_ub1*74B*?35Q?~rn>Mss7GEJ_*l
z2K2@Qy2r;HoWgvRW9|hZgTa%SpS$)F7oK|sZQZ48?Ma2?1ap~Vu8>B43g#!(Mx?<C
zPGRm2=c3Z5b%xP8#~geER_ZzCI*sR=cOtjr5qY~;{*E!X)*!8iu?9IkVnZKs-ZOZ!
z8PpQ2kQ{$}hxXbsYhQf{Q>x5cfoWc0RK@-k%3Y8!$9(RWES#`vEG3zcPMb|BW>JC>
zhge@x+V65WIbx&JrnHjr=iNg-NK-Zrk9gc{@m8~qN>VmP1GeIb+WsE4b3|Vn7F3Iu
zj*sbT%N?DuKI)Q4!|LW4o?SW1E{nW%&td|JJ!^pkVFR29a2}Yo6;=vBViJYGMc-LR
za1>gO8U-<2@T&k8eK3;~e@6j`VxJpLrZq|C0=qyut1F!ujLz^)kn532GT~NNB-~Iz
z33pz|f>R}3H8}9SF9&lRRu&+evAJ4-VU0m=fEIu;jCS@Jv|21*zK97!JPS58jCc1a
z)vJW%%E>+#dGaQ5o2{Hfi4hEYUAnzv9`Ej8tn(O_zDKoMLuVSW5C-UI2yw(Tnouc~
z==VB+ptZPwkRG!rCQcJZ!vWoXm;P`-lFmr7lyWs9ELWXIs7VRJfb-|Cu&{cDrHw6@
zWasiac?{-ooYO399^BdabgRWVE{KBqA_`ZDd?S)i{ffe$fI0m@Tbu*NWQI1*@=a$N
zoh6P|A&QE-8_p>PB8n<e5p@bUxf8$OV~io4M9h?=VqzA>m}L?S22WVfhW@=rbQjwM
zm5OuysFd=3UT|jJS?%ODRQ`RqD%L!x%g=V{`Cv0Q^n)cc8l7fH&%-NKusS3f8)m5`
zIzA?uj+yj(gw+aR&_JQlaY8hU*x%pd`1qJCOIci6MjA^x93hjG$*{*{Hb!`gD_5VU
zy>Wr{3r|z3)Cj^-P7AU59zUsk7a&;Sh9#8mBXjRoVcb-;RNg$;GxJI1lUAua+nwH_
z)29V#3UvZUr%96-t2K_U<jxr-1$r)Dl?Qg@g$)Z>@tODe=L+hil_;xGNuM*K%T+lb
z6pCF73PENKcB<L_n_tjeU84QM3rOE{IzDa%oCIs92-K%9_|xsUxusAM`FjDL?^D>$
zSRs(X2KHbZKTN3xCCr*L0G~xELAl1or46ux#r6S9iwpF6$Ib^fHH?NMf^tM>&_SDw
zvzMRc($!a4*}R0C?}>3XE;(@45tWM#QE=9ZpyZ^b=oHCa6)$%MC<IxhJJ^aKrfAt(
zV@Zu*npo-zuuk4UYn_*zPg=c2<UVipS>rf1Fz0#|pX-Z0|B#p$x78eTA(7H|_CcpG
zSN<CI@F8BPsRkjexdSAgrTFC<XD@DGt)SgHU~zGQe(#tp$9ysz;g=(NeTVs(OV4oO
z@+&N_UqE9KI>8E)SKVyxjc7lCxyW1h>WR#%@H^GoiBEgNwM-o5Mdj+GO>c}PwSsA4
zek11gbbH@{ny<x^_RyD(ISQ#YG8?mE54a?Iw4|gPr=){1ep_K@9=pGKkJ@6J##2`j
zg-B{%L3Tl1p)Ri|WEf6kj+7+%k75sHR&c-j4$051V`qj@ZwToW-B@AL{tCCP<%6IM
z8(T~~McqPNYjfjZfd|_&*3Z}3T3P1l@c>_I;yC4*VV_kp%F_{Eqse}DmY*MI?CpM>
zN7OCEg*KFdN`$2{ui*$;luRU~2D=9g!2W86Eg5KMh;P*pWf-S|A4dssGMc|ji@<7k
z(G^xdv4L4?QNG3Mbbi6hyn-Pm)P>_%3Lp{^BOx>=7rXmx0#A7^8T!JQ3#7G3Goz^o
zER-$w%o>yN7}C5=l~_7=x2ZR3R5mtT+Y?DNnlSEks4OhHuOlT^s8i>mTL)TeW|JAy
z>5RSo$4<3BPvjbnCacR!bdHaSq8MWZ)~}I81BCRcRB9}*D<<PH`v-?qD|PC%CY8X)
z=#=Sfgz04@Nrtfk6{HvyllA*ZU$L~Z!Q%23)#d`)0;F?l0QO}60`q^%eBh(#xVQk%
zf-o4JbFYM$a|ureh($o3X6}MFLg57^!g7T)na%~3vn+4F<`#oWh!ehA9(?FbQw(Oz
zCt<$?%mJj<=>7pM;7nChRWPv{ffJq<iRJM6EgJ0>URZLviRBUnCvHB4UEG2BYO*3P
zV&sD{^J=Z|&^paq<{`p!eybo*K}daZ9g-;lQ_^yUpj1K%iLnyvmnb!Bh&W}nRAzbO
z3_4CJ+_SPcMqn5W2WXwrT-l_xe2(_&7Gb%D=lSHJIY@U-*?cN@z7~o$xbzUpLrTwe
zYotMHk#j`zs+kq@KQ({mlq?s>C)!ooold;_y(~?08ihHj=u;H1o@}$uxzp0ARDPD{
za|U&akQr<2bHUH>ty4mmWGDoQfY?ImxX;}m{DkM0R;ZlaEM65xYh)O@%q`F0Pq$@@
z?`1yfJHOzid&!=U@_f9!NTrRz7Y0vBydb39ZqZm-VekG!-u>BITzl>-#9@kuM;vRS
zQAYn@$n}q|^TKN{a`D<_X7LnlB}#={dg>L{&p+)*>xm<ein&CC0CT$ceeyY_5cBG7
z5!7LH-hw6XlNTJ5V(H}F8Iw;qV##n9e5K@qHWw6Tq-jb)pj5dh<Y~T#MdDd+uKDJE
z{ma?-pC9wpzZd57fbJCLLOQ!7<@u;wB|#g5CyY~g4-}<li~8aU2ip&M{}*p@<r#;$
zLd`KB<d{GI+KX&my+WK!=9pi&@-nOEuKqgafKWECmdPjE-O}dmh8%-pryHHO9Sh9!
z>clC`K|F!Ewgu*3i`MQZFfR_w;^48;>G=KbFc<jDShh!O`7;{E@t3d(*<{L03F082
z)En~X7jJR7y+CPg#a#kLMM)L5a!#$++CtLg6yHh+vb>7ps&<N0$0Wb}7<)gT1Qh}$
zD@5rkGpl$|D>FW-@xvX<+0`Rne|($C($mD&qE-yG)ZmR{9_=17Jw8Ca{5*Hto2({%
z7G+4O+~i;%9`8KBMiHJ0!Sk{IGf>UoX#rO=tn?iI0vF8f%&^xr?Av^*Ghg@vpswL{
z&$Xsfp0sll`9$C;?qzBP)6`P-OirfP&K@q!4}Rc|5!4gOMc;yi#DJf>2^eFMg@&2;
z<+uw71Re<sc0^6a_`*XjEm2+VGkSbTCfxayO{N^*xz9$mhFYj2g)^O>9CS#}Op%Qx
zkOGw_VB*xdH|#l&;PG_CES?egPVcMMsB!tqMUIaSU3K4zoM&dpGE09n<=nY|rB(~O
zx5xe6y9`HD9zNP7i&Nrw#$+;LI-79RJ?8A$v(y_^4tircj|a3D7dd<RWg1IsZaO`4
zt#8-nw0Sb(_PnvVk8{3<i_dSt@mhsNVx1qfald9|emE5Y11D9xQW$dO$|crT*Ezq@
zBpx1f`{u`FIj>4<4c52>Hm??0lMB=uOFsCRSNu+@IQ)7z@%KVq+L&eCp*_-sewCn7
zVum|kQZyL?J({xj;SEZyHlFX}ONH=Uds_NlPEB$9y=Y_4&+kH$aNc>tO?oPY^qrQU
zPBW(C3FS)F*>z<a5qOAll^`t9+1X)w^q8}kt}yD&FjkT!DM|*!vxNKi@3VGxgT;Ey
z@vyxB&(B!exJ0$J;sUy$L?O50rtb4@o6T#m#Rtz5xcT`lrIR)y59H>TfWu0gA|TAW
zF!>+_`ASH&z{BzBjCH@4X6YyBS|_xwB4}6E4d}U{nnDltcM#OIsy<iGuA@G>O^{fu
zRRlQg94!Sx3laoI2XyW{=E2W?&Qr}lK~!sQ3Nc9$WdXTwtPmhEU^T{V6_@Sgm%7a-
zyQFYN&jK)6Mx1HlBqmN`gfVCo+D2%rsRSj~w$7tGpP*c4@$6LsRU%8Kl=mJoO-8)(
z)fc$<+_N;+HkjPMgRvRPE75GPx{M%8Ko>z-@q;`QaoLl*aB^+BytR-A+`#E7nnW5`
zxjB8^&IbcvK3Qq4V3tbK(6!jKNia4eO*1mB5h6p13@Lo4I8Bi)x>0i_B}Ntk$%Zc@
znQK*juAN;+en!l<fA({(HUAh<tvk%q)NP!?EBH5Ho<~dhaD+Ql3PED4HA$Q~Tcsq%
zc@rv(jUDD*iIp?wQNB+Q)@Yr%j8q{ynNd1uF-u0g{Q3)=d-hrCE9=a*@6IuAv{ziK
zJWD|5ZO%OFc7dIllV=M~oXGu~3Pg)o-C=I#n1lE`Fn2$XQjufMuVGG(xxk-0i#QI?
z?+){N#plAt65*YjG|~)d6&?m9Aj4vZV`gEQnrhL1c);$v?{eXdH=tZWN;h~UtUxFa
zk;`&O<ER6+$YyZWo05=D_sv>%<00eO4r!}Ixm-b%6eZ!I!!nsYW+oO`^~xM9c*H;Q
zd3bT1g%|78Z)CiGcOMb&;(Jgz{|v8e#$3I)&h1Orxbs$(=0+c>1d$2p4Pu0zV5Nuk
z7TsB-Bkb^p@beb-(KT$afOyuznnX-1@Qch<=<ErG^Fq(Z5%Nk6yCOJH0#jg6Tmbpc
zi#22rL@6X8&e&bY7?Y7@g3LGq_5IAykb=#`66BNRQz@w_gEBI=oN^q>FC(Zc`I3!|
z2Jr)po*0A*@O?{tbCq~BA?uIG5LoLmIUdk`@R0TLRb<IW;o5j{e~4ZHDM8l06DaA;
zTbD&OBAP|?dp*Li$}EmizK0)p)ao^!dG=WzKi*|Bo)XU-$12NGWFBn2fBiN>C@x>T
zL|AQd=HexU7ZB81WQk_jJ7zNK(b<1Yy<KH-Z5_W_<M8GqZr<BwW$i3mSB8YuWx_zX
zb(n+V0?yoy4~3LAx4Jte4`gv?0C`nB$5rNC2Zx22FNl*$s^j=BEG$s3wy8Brv}+Z*
zy*_c4I*w)LY^tQSKyEpfSMlZ(_w$N;?qFdIr#(dfzE3rjT-|6>**m0)g~@~<@bWv@
z<*X@Sjn81O%i)dNoT)lXEY+-H5Qw0JMS}1hZ3g2yS^@+v5H)$Q<IaAGP`-0!$zPd3
zB9+H%Hbwa%gcXDz;Y%N_HLK^(<DJ<gC^gW2mFX-Z5Q=KKjJLDR)3t!w(juzS#Eg1K
zUn0{Im0A-a6iRuVDx+}Xc9KigxNvN~E{c;|P}s*<gw5~xJb^Dx_5z+_9YHZj=e}-g
zz!w&w1=^$->!w|`0jb=yv#_{-3)d2sMHf;^q%re5=d((beXgMHkN)X%JpWXi8~>jf
z5C8T}JgxJ<Ow13t%w||T{MfSl^Xs%Xw`jfgJW>T1Z811PEZ&J_l*tnJ!mY?%3lvf*
zl=4x^CpCud;eg}a4oAoPbh=0Mdp$;z5pkNb-ctPIKllpe%8XK2f!seX85*+g2%Qa>
zbUF-rhnU3DTv(z~trATmj=NoAZE3EY#}6x{Nt{>N?7SR4ZwI@R87o{V+}M*zR8d59
zc{%k-*%O!-YF*8D9+6|2+lE*HQEW&`g3z~)OlmFJskV$f6=a?WWx7M?B$}6}e(#Gt
zfAmkE<GH8W-29A~KmPfLv^LM2V4gjR`8>-xg}LL-<=LBb8b+xxjE+Yf?{(-L9nk3>
z)9;_cysh}N-~S4w@{E#S0W9@;opfZ#`V)j6G3|62^$)R$rP*GhQmGQpBD&oUv9{Eg
z&fx`RvI28XV@`f>tGU>(u_sf=^Kva>PHN8Mv?qUijyWfolVhGQmsxB{N)Y<|8s>9J
z2t-cv`rTvx%6BjD!lees|FJ{&?f2c}bB;NH(1uC3OWO}AQReX7+cZ|!s6KmzK$WmY
zxb~a$uu3r%0;vU|alsze@gx)Jkt#u?huj?Ra4s`=YQi^HuaZj3Xf|dxnlKp;nME1L
z*Y9#}@j33dQnIg?$UMc3QkiSd1=N}qdX>xAXv8;0LoTL1<yMP(S;+F)8P5&}STE%9
zw8mtx2gXB|o3Qa6#7kA|zZ*k3z%s-<?6~*2{R-lhRmAfOHhdVRIXR|i*f^;I!jsFv
z5nu|?i)wz16pj3s-6c0mENSNG3(gAAV6)5#m|BHB6g*xNjE!Z%*=!<n9;G8sOSe{)
zI<?UL{?Aj-Z?L>@h6kq0!HqjE@j~F0O4K)&i6=9%B*j_}mBMg)pL(-N<HCv?WYQ1~
z#u#JpMNxfq^nrfAho=OPA*ICkJX*~be&BKK*=HCJX59Js0grcf==S=|W>b<hMp#Ru
zB6<4C8XtUooBP|luH9@j(kPmXtCWKh%4B3&Oqxiv4e9h_dXKt1+B;;D#O!za+}plS
zsx4b*)~QrV^Y8nF>Lcc_(FNX)yKE_{^*O%zy<qc%S{sY!mGJ!_cl>Z)V_v1l(Kajs
zV<e?&ldba?!4H^4GlC$@l_5?M_4Ci*TC`@)sdI8J%oDgT`n>e|d7i$wN&I2J?9l`C
zU_d&XB1*Y?kO7Us3Lh_n&aFq(7ni66l9I1b)e4Ry>&|YWJjWQ<8Cj}H<Crwd98cI<
z7yL<yl9rOEn2nDa4~O)+UHaWFgTav0SXw1`;hA-+jWV@XgHq@d1Qm!iI-6l9Gl(r&
zFe4jBgkcF8grE(x$%I5({OSTKC}G8u(j*S^f>I&o7`q*9P71D~sK4=K1~;$j<*j-{
zF48{#-Fa6BaIO-0g;M98zBEnqN+$ER|LH<P@gc;i0MCk(-~GFu&lS|Y{`Kb&Uc}|^
zeuvKP9_j4|2;pQ#bk6I^j3)EIvmSOBaqmYz=DF1drLz~YD##Ce<+9^^+{5uX0CL9y
zUm-k&us%lzJs#ZK=Hd2z?%uo2?R&R5I67oJ9+PAVSiz<BHeY{rlSU(;999TafD9C+
zhDR0+h@%N<HUpL6*Fs8-D%KbdAMY_3jff0<<=NLr4G0nnd!W<3?Sdn5vN-aCJ?B5T
z3Qv9@7)OdOIB<@Obu#r?Z2Ure*Llebz|4vBp6?cm)24I$E9VyFdh2uEh7eAK7LAk^
z59z8Yrxx)Sd0u<{8KfV7DVQTYkS7aNoWfjs%9ZM^N9Sn3!v_y}xc%Ts%qL@#G{<~n
zfv>%?Nv#o33MvFLKm`if@G!}cD4LS#DK^JEY*aDU(BIo*IChvn^YklZ1_ZfIllU~u
zi;CLGyL5+s(VlUDIv+>o`hmupCwy4v?Q(H)(967Slg=@>zlOP|PA-UfQ%B@2n%@!T
z-Sa!tS})N2+N;bC4@tHkgAn*al3Mq9vLs<V7~p$8a+L7s7jJQ?UBy3h4(pe&(gWpT
zlt-4w?nZ+fHjx<VVT7PFO}Nd+TuQ6Vc1!F&>T&nphupq<o4td5hNB@#5>pNpFI-vY
zzyHcQFVzBme8Gp<T_m?X31=$=(FF2@E6hrmYKch@^Rhfd#W7=5W_y~sbyxW|VpU*w
zZ(*A+=ihmXwKK30h-biCf+^Sx7#C#qLZ&6J?mS&WJmn!4PDsxV@S;#B5MqpwD{f6l
zH$@$1mL$cs;qrF85RP7e5HJwLtAaR{2x+hi0+WF?2x(EuO=2qVv-v^(UC-I`YiKiJ
z>GCCdM~BcK<MyVavamp7X~gmFVJ?F#Axb&Ad!I@xKsMV5;gO9avM3@b*ImV3fXpi;
zqhXh57E!H2tyTw_Qmxe)_eOmD(G7m_%b)Rh_c6oq$SLld3}FQAdWp@AI+cp!H+eF1
zBy$bJ$wQJPB8^j$G+~-Vc!5V)sWO?&7|mv+7B*HldAz?*lDKKpbLY;`XjJFHdC?Yx
zld6lDTg2re>LOPxpa{bA<)C~@m8uJD@-t-qHN@%VpwR|`pvub1I;vD=7R3bN%<-*k
zagl#Q#NKKOd&>OdV71N%O1{{0<*75&>UFlh`5JeBJR?aHvUrL}B(hXOiCoS@xQ>dM
zrR-n-m~*WPsqzti;Hq9K=btH`EJ=xGDe){KiXx&YA+y@0pjI*(P8fFkgr4T`aG%}%
zU5-1)^ap*S=?twc3-yqN<{K=>P5g=?EH$ypVw9ch&(=x}jERvU_>~YLARhFYMl)tf
z#^%`z7%K?F&`Ijd1F_SA(TQV-ou8liS#OJ(R)@J0{dU6O1&soVv!1y^ANKUw4i-tK
zK`EC2%-x}shdO0gkau(QR=*p(6v8PhxRd3i68v{PpDU<q)SFCa27h&#%YX1q?i}o6
z#}QuMMngz4t8sFP?i(6EJY@TaKjr*CX(QT8h`eeeq|5w;RM=b=K;#PqCBXVTesIK%
z+aL4Z2k-GWfBOUWj~<h0J^$7Opio3<LbY0FG#nGCkVdslIjmxQjl?69lEk=1U%S%6
zSk3;y9=%?d-l)&ni!ab_FXyB6u6+PbK-NMY(F3P0O>5oZ=DMH)A@U5)x<6-ff#4^U
z!cP@>iZ)*{Vd>>rjzth@C>hAK`&yaFFgC-DCW!pqxpKUaPQOfn<64<K*fjb}Y0FgW
zHImr=7R=S(i8(m`P$$gprs1q79FtT?P|i3W<FWs!!_C__9p*pz0sBY0t_^wWQ7ED$
zrc$jl$uX~2+LZh%Mrow*$b=xxF>mIWA0O<|&oMu9;RRalWwg~ufh*Bln@g3DCmRkk
zQ*3tY)Z~SAiHH#vnP=`!IQSEopDNYaTx*Kcm`gWJX%Ix(I?Rpx45u;A|CqyEJNs}e
zt^02&5MMIphw*i+c$#2wfpgz@neEOI(KL3!r3HjUTSF4XO#1`MOKrw`9rk|l4rdxI
z)Ixx<5-mNn%<Fs7BSTR(V1*=Akd7jT)BCJvb&l`Gy!ZZ_{P?Fo;?eFlvuNh_dX7h{
zQehG`*$+al4o7sF9v@a3WMjvbu}2oWVp%C$P#zCNo%313pYK0LcaKS|<$*p=zfwWG
zcmZ*zkAC=5$6GocAgUpD;Hs`xv=QP0R24jL!Ji<$Vq8M9oo`|#5MM33fX>fbP6kBb
zh3;TS2*DVSHf};tDM{dam^3BI#Y;17(8ld$tuq_}+mQ7nx1?dEX%I#;4Gbk=P!bew
z^3GH0OPPKxg(adW!E4l5x^kZ32R9(iTnp3-XsoX?7>pp!4d?|#y%F6z4_J7%j18;k
zC?)C-sV*!zt-`$RB~*a1ip*%@EJn!yn<>&H<I$sgeE&z^=k9~MjHhE)#mM)NvXWH$
z_<q225}|~m99Hm^kCkp8q9Ie3R9j@|ky%ST95R_rnMO0(OPiD{H8P!X)Hy^c=VyHW
z{8?(XD#evqoW11mS%IV3Am$D-#Y8K(Dyfi?a;26N{qyHkFptb9fAgS18-q4bsnlpJ
zE>o*D7*D1oNlXy9>Y<PzrFCN3BA;^3xqSJD%H0OOcu*IXLn_q@rNCq3g=g969Aj;U
znax~~=7l5{S1H2XS<&&3!|OL$uGY}Xr`lY=D^&nN7Heiv#AH0BH|UcjiECFXcT@L|
z`t0uQvh(O6gTWr1?g5kO1f3Or*3P9TF%qv7GM!BkS?olng~AGr@g${I4PON$aYQAk
zV2mLi4;hU|3`Rrh%V#N7YFHuhgYeXMe5!wWQi96^I_BKhqUCxL=;ip^li<4GN?Ym9
z{UV<;^CThHYgCd{Lx78Mj5TCB%`1ChUdeZhMhffhbLsALlXEDGF3e};)O{`nj!*)r
z+y`x4zQEe6U%^U+Dp~?U;^?hTh5;;IDmeJy7KiVDh#n2d;v~<6$PX_C!bAEYDhQFj
z4=SL0JmSX3H~H%y{3ZYT-~MYJ?LT&7)Wv5J!l_FuixnA4LAhLI?_iJdWQ;WuMM$OG
zq*`uLt+uE)7V!NN{oaV(y*-Y49Y)iL)s4$YspkBY`Mo$#yv0fT6y~j2V{qf?c@l5t
zziv)`%=_!)T#b3X1xJ2&)i<Rio~Kw|TOrZ;*Uc+F#ukKHa`I_DG$wJ&$hu0O=c(Ka
zvZyqE9&~>L=Ea!xZ^T^WnDbjPpQo+X9ia#h`~czm^C60UXUxryZ#m5W=YP%4{^KVC
zI)NMgR?=ZE6_iR<4iEO2OvjGPCPK<#gG#AMrP`v_Sitj34ErPYPGP>XcEPoJi;1IC
z-*aK8T^wb_Vt4{`dxE*mD|B<rIl+AHiCLg73VD;l%vrI#woIZ0+K72JXKZf#oL9zj
z%3+SF<9{Fp;&*^~Rn>69Z<6M@O_ra#hLCQWH<#o?=Z9K0nGg>K__3yQ`vIMsx6u=K
zIVVyQ=P9Vjhn*4+GLv9u6MEe}I=4IYKOFOyfAydEAO7|Ko9nl)(;p3p)5KM~b83S{
zBNg<m<|oSMm7^mzb?kJ}C?kGVMMrH88VmfevA{h%CY>(H-Y)%qk0`YaMher-Ft^6o
zxKor4VIhIFKFB!7spU%LFIuQsgvbz^0(ligAP@nFRe^X$Av_n*2`>-i6>bvv6zXy6
z+N<u&QdG)iR@PUDGH3FxwaNFnL7Nn<A?tvB2kwn6U2Q=de4|N#iIk+OIHx|l()@3I
zgxgz4NqK33`qsJ|QXx-3YBj1W3r;IBU-L=}hPy{h_Vyu;&`HAN@W^qJg=@=s0*?0|
zyRM9q^ap+7D5KXMa`V<r-uv(aj(c6I^(xhBIk!#AP44pmM@Z)k(H{<(MX}>OTZQsU
zD6fR?mk7f$O8G=l%y2YfGMx~oDUHSwzUSxqo^aGT<fwDR!-tQFA{TTNiIBA?p_?`N
zd0xnu%oB3qDwRS~F4yL!>&BQ<H-tFZh_>K8Ls%}6WR4eCuGCptUZqqjGo4OJ)AZ!1
z$@e;^1!!ensj>EnGvy0GF2M7hY*cMymDc5RDBpLc@==5+XaV{EZ*iFSw|5zC@1RE`
zqQL;6HEEJEozCbFhaB~KB--iN3Eu;$7!D@vKHlT~58vZQKm7rF2fOt8eTJhEX_n>H
zP3e*pDe!$oP_8lP_Q;}`=`<pWG)9D!YD;*PCZ$>nFQ^dBQU-$|{o#P&cuJ+dj5Y@4
z`A?kf#c5s??QHQMMc`HpLe7J{;^!tGUoH6bC-2breT?OacZ3i~B~en+Y&OU&6qa~d
z-o?{eKiL^7FrRk;1S)r5DPoe}iTJR{FL5hD1SM9#@ij&}$4qWLAk!{R71&$@#$Ae@
z1wBdG`}vzxR+p$REF)|SiNdNvyinjNe{QvrMTXlq?()uizvPF1`vc-E0b%hx#ro>1
za{<UQgmBe~Fmxou<&`yrlI-m5;Cm&O+bem;M4*%-fOd{M?CkC^9FIscO{K9wqqRy9
z_^z}%Z?)wXQKyI!MFf7L09{n=4A!bV+c5JM+UZu9b3k}9FoZ(-A+KHuSXf@-alhxP
zbXhj%Rh)#EZt=rOc2l@mNGUNo&v=A*GK2hkJ^~lmJ&AdT$*udJz}!u6{U*$@Ed+(;
zS00p0Z~i`YV!3<kKJUEuD}MNsAG+2$7SEHct*#QyW@Ksl1m?!1EH1BsC)wNE!B-^~
z8p}CcIbj<?@3_Ox<I|Y8sWn&dea}?|^35wau?RYixssqw!JjPfbW?AFe6lg$s8*a*
zW!-u@g?YZNRIa-xa~Tkac|Vu7IEDGCKRYLM3*n})^8`mo312eiY8+5*H<5!OBq2*L
zy+Cij$Lzsl)Vz&h=Z9G`98r-1A<7)S`ythp6{_u}la7as-Fi<lX2}P9l<iaNtT8&A
z;?J7A`S#EGi@*B!%;K4QKb~ZHagjKR@_i10F9-s`qOkmJ*2I<+Uq9R@u9rFds0HmY
zgg#dIl(Gqz4-a|#=n;d-7$YRD`Z|N&ZER<ZjVB1<0|6F|y?+&P={*W<HOmb|)`9Jp
z5EsrOMA>bG%>-5q_E;d!3WRW@K7=_g8IEF5s=;tL$4$;43<8#y7unp{pw}O=xw*>w
z{XTIDS?VyCISF1^3xf>vSdna4K3;=wVhA!fU7bmVG=-&=IN@V`kteS_2_ZoF)XtwH
z8BR$KyN)A?pt8_p)*oSq5l+kmL~IytKcrTvQL0u+`a{xbL{M!Z1ej45Gw5*CSK#{`
z9drp*!p_bP+k1CF`D|`%lB5ycu8+~0>2!v`P0&gYtkqO&b>vLa9}ZB;r&_L21W$#O
z#&j~JKNt|jF`3r*VTG_<M@T{04@lw!V>E|H`&23w9zELO%H<1=v~QX3{lbQEKB!P+
zQ+fB!7()=0P@bDUmoTrK6~VxKYIYtdLRbo&O__E=;gw32rKMH+y$(r|=2MQRxUtsV
zE!voTVm|*&Md#>Cc$}st2rr=Z)HR}`KC?%=L`e)ujF3*Kw-}y~2#7WPTX(6no21P)
z*<?n7WSVLE!{LdXiIVufPdrQM9`@LNbe|7D{(v~1F&JusFkq#z$YeYw2s}zbh|!u-
zpb*khZ7(3ql;guaYV8$_DiKPJ@&Y$qou$l16T00FgYl5bY=-npNWTO~7gOYp5GS^p
zqIk6wQ`t7Jq@I?DLgvFSxqX_GmpUD~xgVbIk;Vx{++Va(RX+LKXf^2_ck)C)ptU8-
zGzghjg6DB1#USR~rHY)hm;bfDv!D(EM!GhGY%Or++h3!%e?-zB&dZm^TIQ4lCGZi<
zcK6wN{{vc=uQ*d;B|+pos(d6`2t_oTvj6CqgS`WO@{1pn>cnMPXwIM8;^}Kwc<-I}
z=yb;J5;T^m7La9@Qn||7#wJ=DuHX28dbviU-p2C<X%w-$zsubRcNtE`cz%gerAlHX
z`-cNA*Bb=B)0fMvb~;3e`IK2vV=I6tz%N7#-BglO3fG#Gps(QK<q`kfDB8VIgpee0
z3|Im`V0mSgXfR~j*>$`sZJa=<KH=yU*;<xnE(pj20g-3N*2?+f`l3(4<#L#d9P`)d
z?H_(F%p1m4dJEmBLV#8XNgT~Mc--Y+{{(ZToD9af)0mG(zlM2Os<N`WiOCE%Zhk<u
zT%(?2o<$M+`@7uxHOv)<M?)^u>I6PdU@n;_WrgH#ejpT><60rYZFp^SAyExh6vwwH
ze*1~w{{(a4Ft--p_gVQ&n7a>_nGB<Ux0sViRJlaB0NG%KY&6+;?FEKMUDDA6%?WSN
z3I*CQ9Sy0L1hc(;_HW*#dEo+{wZs_0S&GmDti8=}_z)ZQ8O9mY?gH7M$6x==UlJuV
zlyrj5n;RQkx^R{oAKjogojSf^BALxnln9XTG<mD;(Yn~-kM=)ed%Z^ga+R^RR1d~%
zKD^KV{kx23GlEi?uu><EEnDRZw{?fuBtU{B1kr@?1bZ(5Q*j5IjiGuKswG?#UIH<8
zrlBSfS1s})NNW)xh!DuYC;^^8_@`Y{Kzi<S^L?L2y-u20&R@91@c5YC{x(UL<~$wY
zbj#3)Ws98|&NMZz1qp#lv6<7il0u*<F#mG22@64Vp%iF=C|77*JkM-6#mr)?l6a*G
zm4znBBt}_W3seZw;fTThF4d(4h!fJ`7`Y+|vxuyJNF__?XC7Im84iZT;|>q*-(!-F
zs5I*|nhjRhuAtMHAAJ9>vDPq~MkJXfNt|qi@&oFD^GV%1*dz1;N~H=?I%kIda6qTq
zA<6?0&-c+*FdEM&l}bn~jarrQ)U`W1og+fO%;Dh?>#Iw^DVjxb0N^T@&LPM+KB2W(
zVetYVV+_OL5K_ycs-VKkNy4As#6@y1l_Ur~beu7nIpqdx9lvmCd6mhyN2X7TA|}hZ
zbeYCvISnCCR=<=zj$50?Xsl4qrKDD8@!4yPI$ctmp-qO#Mb-;f7%K&ykR-i6o$ZH|
z*H)Qy2k0<l7{{(kl}l5FVTkW}96avO>-PEZ<M)|NrlfI1r6f4BwaLc%3b$|GCXN%v
zlY}(UC;}$a8KpR-T=EG89PK{B^FyRpA{h^;1Oek=pZ$Y9Mw2nAcAAGavkV6lmP!?}
z+@tlhSo7&>>bx3vDp{Yum?wfco~+j9gi0ZVbEn9Y0GWT0G`A-6lt-h{pxJ7YO-3ls
z&-v0R;HVa6UIEl|zMjsUFuet^In4<^b0Gh@g1R}k$3Sc|0^eid>1(XNzRl$OKf|QC
zfM~9Ut`Jz6;R%h*1jp~+Wbln0RyVJq1LFpUr9>zv7))ll`|%y7<A_K54>;-_QmfU_
zI_2Djv;4^){~<s9*-z;X2gK>g=<s-y^4|5kJoEH(G@DIU*4IHwZr-@b-iPnd@AXJ#
zGmwI6qfWU|Ba|URYTmzdpZf>@hBv<P4Ziy7i&QG111JARWnPLk#flKQbAVL2Kb-?I
z%Yz%8W3Nsfv_iNFm9YlrO_BHItsqc7D@)7#SO1&;HB$PVyLgG4AHC1N{1^W>qBO%~
zPRGa!omURQWQHtC$dbfe`l)u68=>dJ6kyO<`nll!bj+Qcg2O!hdt<(|dChslk}HfW
z=jm!S+`D;~**M|R-UE($N7SlSw9YtpVVytzqd(+lKR=DRd*9<x%KIPQ<?7YvXx5u7
zt*wKS+`e&>{SV)v*Xxn~R?Pn`-}w61c;%(%DVH7QuFc^NR`Njp1aorC?Wbcd$ZHq*
zYm>?ugrCOTx**yA1m<TiT;%4*@AI$z#sB@+FfX+5j5cIxf=v@w_5B@U?s%KRd0Ey<
zcvDO3(nVHY+~eWT-oa?sE6{lms{o@D^r(lBZMq-bX7uWdG*&NCRgfuxNJ6x}$-D1p
z#?qrMef-#R>;A_abq}eO%V?c(W^0pgf9pHE_ujAQk4GfgNlRojN_c!{hvw>4j(@tq
zkLoGC_LK|vZeR}&(Y-Fy@t9izYRv}aW{pt!%+eVjy!#=GtIzPlb5C(&`!>Vw6l)U%
z9gOKBh6=<&-Xc#SdK;1a8Ek`9Q&_Oj6o_wt&`y@8rVQg)VG~EAFjC=96moKM{0bqE
z3r^ai-E8rX{`61qgEGy9CC;p`^Kbsgf5B)nL#G-cRL-L`h`<tl$MD>0O2ek8d|kM<
zjX+A{wDhdOnCy#Vu8pIh2!*x=UwVYA%QVkz(z){hYaHjpYt~^jL-)qHtBJ;HnC%`g
zJGVuxwaj$TVP$xPOuCe0O1l~`9;bY8?-4;z<>vM+hT|T}SCDDittyu<U!Zf`VP$O-
zi-)DpXgDI8W$f-9v$ncMty-ngXdtBE;NXza!7h`@m^5)~J}8y&N+Ajb!mxMTr8D|~
zOBc^`<>Gm&)qsYQjK*U|qan${jKjktTFp9P=)1(++7o>}3swkpP8lhNM4XGE3*0MV
zg+wGS=*tqOvngQ^Vy#69MZH$z<(FQhQEyVN*64IP-1zW1gV7k1X=LEJJ1&cI>PCi6
z^B+17=!&X5j=K#OZNE6D4sa9?AhjChc~rMHSh#kc_=B6+)C~z}Yh6+gSZVO2K^wtn
zdx!4UI$p5AG@g(k@ylh9&hxt6Zu8;0AJXfL*gx9ixZ6Q0Ns^}2o6EfV>I>YtbBl7V
zg77?!j*gg&r;LY|U%mYyuYIFRtGz&dxkXqmv-kKh4|i_T>re3zlq)4ll`?+d<0~KK
z!S>-Xo$=dTzH)`7^-Y?Yp;{>u1VPSAJQbXgSLtx~xyS=O0U+}E@1jkMlUb8e5rid{
zDpizMp;E08MG>7&caC|h-r!qre1j;8Ie*~-`@1{*_y<3rR%;Ljp|c3fC&OKZP-Bu5
zYcwX!u&F*7#t_yG8QPq0`x&0k7u4miLg&$>3_@htXY=bX)4j9J(Y?ozXYq4Vyj6HA
z_YUoj>EFIj_KL<+!MyS(5ZKJnIXd9po!c~P%k(;3s^x&N60&vv0{{G<|L^&WzxWFd
z4~~eV+)EV*aw6jY<6rzcI!8l(?>paMZEYD9*0}bSmpOMeqTlbc`{;oaOO%o%oq|ye
z$5Y;a|0Z`I{hUWTk2yN*^Q||&#`1Cl$ew6_6{`J33$_?jM=*DUm~*PMM#vL0djlky
zKqE1ns@w?hJh!PYUcAWS!V2})BJK79H*b7IuiqijDZciw`edRu%QSHuyR6<Ar*))t
z{<%`Dbe|<nzl5MJf3!J;c_qjE?lwnvKO5%e<qRdtr!WVr4c*Q$5ANKhQCp@z$2?%`
z{6+rj|N4K&fA|llFwfsxARXra&;QrI<M?REcfS1%R#umhevQk|zs%XI5yRhx`3E<-
z`{)-u+Ih_J;ec;^<24o+>pAB6XXB*vB8&0!;$*{B=L#;3vAJleFeq^fbL}nx!-@L?
zfd2{1tBp48_5!zV+@RM#egbo2T;<M~EcXoj-D56900bB^M6HS^Bsz{*d+}+8+mAVX
zbU;BOc34V0s}ONQIvt~XA>#)-Y&?^Zrmo74e9Ux|5tVCftzF>g%_D@=9337|E(Zi*
zz{Z)g{NsQ0&-vAx|AE7!W0K@VLPl6P?oRmr|Mx$0_Qfen<*$%lZ*cu9i=^u{wwA|S
znI1Dd+;<bEQjn$*CKGfzV}A1Ux9JRi&L4dDk9hf&m)Qv)uy@>tOaN`52lz<w5|+0?
zci`i{!rniRsKr>%N1RzhY@LH82)vLhWyw2<#>5J11tsP4d~L#w@`R*+@;<Eq&+{pl
z%bYuRmc_+2W}`8cmtW>!f0xI*4;jy<)KodI(rQG*<LPD16%xEGCCl_2b0GwoFj#T{
z+w6;TtV}dTRznb{+J~N^d47}WahGXt<f_=dPjh|6`KhL<qvK#<nh+g#;jAG@$Ee`}
zo^sWbR<*?7xQiK$S-gA(DKtk1dz?Lgo<_aGD=)vm^^b3J^X6R^+N(^)F`e!a0!yYX
zzj*T_CX<M#u3o0qsG+<Ps~cM^W*L*ogkG;R|IC?Au|^QZn%&()I{gO>1_PpL%&RZI
zNVOJHt5uoJX7mSLO2G!l#~n5{R_CqUVrW8=2YE2(QfZ^<b&fa|<U<!8LIpIIeP;2D
zgZ+Ir&upSRMOX?cl}aqM+bk|F<A)WRtwqwrFpDNcafD|)vdoE4r)dV#kR-_*>$yNN
z<mU`%o#oY@FZ$RyhtU{=uO!y@S$gUs<AY<m`^R`*PMHyfVu3^ni!_#5e@y4zHtnrv
zNTxA_AxcTYa=^;c3OkP;vGec|iZY$!Ba{F?@MvzW^0lvhjkn)^CohKj#PNh6ETd$A
zFovJ}>SK~j@bt4!Ga9s6UTV@<Im5bF!|(J-qZ##DgL<t(rXz%vOlJ`fclJ5x+++9P
zh%=iPx%%`~Ha6F&R%?_iVb1$6g1b+4P;v@YPAze**@9Y>@3FkK9Qb9*wK`r<LU{pk
zk}(;Nky219l~`#k;dwsGD{HK-tkF3>W;&Ztu2z{$Mx<$)SL!r6O-b@<tkY;b+2=NQ
zzp&PlX1|l5t{{cGK7VnU`wJ~Lzx5Km;|{7bI*Epb>(weDP{Kea>EC-u63+-LRfKn<
z6LaV0bt+|n(G#Rtq8ygEbp9-*TEKt%fBx_2_h%gJ_vm%U=*$HVo|2S&AC2Z{G~s{z
zcYn>l`48WxTnVXEN>nNp+U*vL3yZ8Qw`n&TNMqPL7;*REF(2Q)$D_Stl1%fTzyF`;
z_Bsp(9lrP7Z*%tSI#N0w!)a@lIj;_ZDI_b>6rC1=#qN7QA^Qq4E2z{~sn(h&mtnDx
za>eap>dW&&HnuME2Y>QU`QeZLicYUf7_83iqfQTDXszc5z?@%aHO4ys&7uta#UGnr
zHaX@7hxr>X(L4HVn5W4Mzfz%4<FVFq@6HV>C4n{*q*$gDmbi3&i&8D%zxnU~I|ipQ
z*9GQK^1M0b|N7tmHUGc=_<hRdkaD#|wNjzgYO%Pu#OiXJcBB3z=6CM#Xs_ci|AE7N
zIPCK6Z+(+9TdO(dIWe`cz!4&^go)hm56Dti6+1P&qpj1r6IP%RRH}<qs`WfToD(DF
z+@%xD1J*Xq@%w-BPx#R%FwZhQ_Xz%n!`uTAV72pk4O>m3j>c;?SbgOgI=w!kKY8Lz
z6voLCW#chAtT24Ii_WG@{2FPJVZ{!%XIU3#7(URX@fa1j_RRUMO~Q)LKmTw3PaGXh
zIXdbv=ug}n0|HM%=qV1mJw~%B`-jJrOaFne6d-<5VqB^4LAlBKGfh@oHJa5j*(~AF
z?vNYz_W9uEUG@%pq}uR5{ri8z@$n(Q_r`Z%t;F8rJvT5ReBcOc{|cMq2z~7IJ;c;y
z)N;><+1C+`tC&Ed1<262;cXLgO=-qI{V{{X&O9-gyUyg2MroR2GEH~5%l_dJ-~EF>
z<8S`<2lR#mloU>MI?D)FV8b`4)XszWBuVlmBwTx*FHeYeu=|ka*)12!%(+G&%OQ;m
zo5Ztwh{+gZEJ3|ab#<9|XD`oyEXoLyevjrtiLKQ(fpW=ImSyz2$CSmCsu^I@gtgTr
zb{^Icq<r<&SLq$~8TJPZ27Th#Qm;2yT3p~TIV4F_#!<|B*Kc#<_B~2rfbV<wzDKE4
zrc|j=tCT61La>(MFy^>7X8)+eU^pQ&hFiC962~*7glC_5inY}x6oT1o;$)8cBi7bd
z@jS<SG6KL^+F7j;HkX+Zj`vh=G+YZ@Yi4nZCj*4?=^Y=qKs3pa!V%@IwajKS9zNX0
zO3&$H8pH1X0crLtn)L>=Xo?h?G`WoMy}9^smZoI7sOCVbo$aJC2A$d84(MligeOn#
z6l=h`VG>oVvG~jd27@6&XQ+~oh#eIL>(1|ZHpU9UWPcx>#ZIbMKp6V0t*!C+;SQrw
zmna$$D2cBmtIKVIAmH_HyvlGi;ox{clBD#yL&oC~#%KcN;dvfG;PaDTe!wr@{P0B2
z4G2pm8jS{v?Is&*tE?<6BBfw7%XqZA$KFwo=`3aU;AfmYv&VQcB8r~j?D?~3Q=?X^
zI$gzLL_N>k=aicKZt|3eQr^5`W}QGF*j&;iDBN-crAmz!f;7`)y&e|FoAaa(2pY{6
zUwP>@KECl0NtRKrRO$CSB>7pOv&`92TGzUDZbEKrX=AbCM7#8}JfAP9v(CN1UEoe9
zi+lqtUwewJuRLP(<984yuSj9gxer``!V{8s=ZNXy0p<DvN*V&sISqKe;@++69PB@$
z-P~kld4+@(Ui;dsY;J7uX!|X8AMKJQ8I`hESOy|AB$+1B22W-9N-&Kxy8Q`4bP?>3
zX-%9J2F-#v&4|-tOx=}U5}oqqdv7wE%}A1zKm5b*aOKKYUI}rnPbpoLUX)>lRak3~
z#v<~=RoZ-NH5XhoT2QXGSzKL5NI(BtVlI0jBq-_SeSM7tRu)!y<(03KY0cd`Hz}0|
zV3-S0p4hJd;$%=_PV~f8lCos>x#0a&$Xzq}*D)88<)=Om=7)s!HqsM#0)+DLJjMOn
zA9Jw3L%ZQHPg&vB*I#9GeS@8cZ?U_xN0y$#oKu(^gBPD*E(V{(ydbVhk{t8XGir&>
z=9tGx%J;te2A3{xBBgZI65$m4g;Lnu;ey;UOuFm_BQ-J?B{md87lN?dU}0(HQ!pnd
z(w<;0tYERd%u6qR6`g7B-aWzG=0c;+EBGG<b0HE8QO^G~$dXT~U1t_gSh#qB^=o_F
zfAeERt`=``<1#{kms!$DM6}l-?)GV(tupd_e%xDS@X-U-R$>_Um`z7uORO$0gV*D0
zZ+wlV<%>Lg@I!WY_Q_IBDO7XJ(U4?@>0rVvit&}ZWMy7KK)_F}MdJ+Rjj>FlggDcj
zzWyfBlpp=<Z<tJ{{QmEKkF2@OQ8&R5b7J-+oTE^Yr5CHlCUE_is0(K?r6%LZ@{N+9
z?MXaO^57T0;O(Ei#p%wN5WqknHh|nBFViWB_W0$S*Ln4|D&PG2cer)?V}{cSCQY$f
zMpYRGnRBNoFfWohYfhB!9S1Y{!k7=X9}@&2gt*FpFbFpWLTz=KY%pfn?~_`KQIg8)
zB6d1O4vqm!So7Iftg*IILwNbX2cV?F_dPnpK7KS}cHCj>?0FvV-{<O6PvHk4y>6e!
z_aD%zIeCyIm$)$2wf?lRj&GKNHZgV*Ii9cWqm3oa#eoG*1}f7}?sJWyH|%li?p+#<
zCV{Uwv$;-xFvOUQB!MK3@q@sf?VPmbL@voAZ7vor1wj}fgd!A@db>=kwM?9(OeYgO
zPm$$*uP*5l%%X^DrNY|!SsKknq?BB{_6&yy`!pIgSAEYSib;+6XhohGXl>@UV)H5c
zFClpgDrIbx&Vv<lqCNszn`c-)-{H}%`}jeCv>Kaa7@UB7AQj{346}Dc((j?#D_}IO
zMvX8m0fwCi_Zbd)EG|@N)T+d3g{Pi=nzgkxI^B;*;xxC(%bdx*v1B$S%kR%4DRNg4
zTg21|;nW8*OFl_np<ZuNtJf))OD6*^0`5J$N0KE@GRPXvUpSZZ7^`?*F{n^<Q5>bj
zv22`-jy2W=^b&0D^{Wk_LR1jAn^F0c!ZK1wl4ipG?v8t%<UCcNR4Q@)`~{@+*gx3C
z^CiAtB93A*?E*DpvfLZ}WKg2`w}l+@_%q@9xq>>2;|i%_yn22F3S_0i)@v`aduNCA
zao2H@ausD`1WHO2l38!S(Z@GfzH}AOD}i6mg$qk0S}>VT=?{(x!xHDuUEqZmUc_p_
z_Jap(ZJpupaG%j=M3!Y_Nk$<PVT>V3Gq9)S9g68tLuv#%mj<zjQ|jL*mtvY_T)*)F
zLExc0L8<gdoIAUP)+rW^Qm(RQgq#cL&Pnz8g)JqZ6h=xi1AbUyd2JKV^XJv3(_WpP
za^jp?YmC(-ae_dwwz9@+uY4VyX&!9fp%Mg?L*)X;oX<f{l2%I1OSZ){Y?fu{FQffx
znM;2>fw>jP@@L2V@Z%dST)cu(0eB^Z@)0T|(SqqI%+H;>!1G^u0j;gW{LChYhw2Mp
zZgTuilVnexOlz*+{E#5<@H{A$|A4JCn}{59Szyj7%yZK6lg8f)VI`RbKL}V_UPF1l
zD*~OuTnK^kp2R#&5+s7<r4?R&`KxH1@nHKdl^~!LDvZ_6d-Wd(bJ5_$>MB3vRkcj1
zR%5CS>(4*M@xy&M7&^{Lk>?qSC&11UrrjQ$yANoc+2rHn8_a(AWAwct2SQNJs$dO+
zWJIIhptXFNa_c<rU%$o9qlc`muHq@lWHND=S*oAZ$ICP)_HxXt{aPCmBQQ3dOH6zk
z=5d<vt9O5i?<>CZ#&=L+#BRqC#>rdh?hbcf!RCJ(0%Um$niY)AxSs^{jO3+K03Y4r
z?H~RPHot7=M~e`MSmaiIl1yi=?Gi{dhBtrpE8ckHH7;Me%AGqmNn>Y|<&`ydZbRqR
zkCV@E7t~y&I!!U3k!$l?F^^(8_wV7aw(ttc9GkyoKcs!`3`dXmNJmqQ)~HfI<J>0E
z-7&3-q_(n%AHryVkJ{oAQVEd0Q{_-T_#ri4G9K*Hiz6;yzR2>@0+Z2*!=q!`tu}+v
z0MA#<GN*|c2EM!Wb3(bXg1PE%zW-8fNKG-|;D)tMId&n0WOa3gmE|S+{Q(bmwvi%b
zp}oXndyz2kNE6Gnvqz)iVS-97*CXZ=kMn^ynYUz(a`Yc5sFllPSwgpSOqQi&S;BZS
zq7;^M84~Dqy9Az37AJ(BN3~X^R4TFa=mFQS-zQ5G8nuwi=QpvZc-nJud24gpg`Kx!
zv(MOy`)yB9t$>;4Lj*SG@49!X!jPpa7wGICVdI!mtxi0eVzrYYQ&M4#ARbNVJh;!s
z!ZJcBN`Y%}#!2QJ9W2ad6T(uNR<p_0*>f1<F&GTkSYPA#_?Q8PFbs(z$4w*;<lXwq
zKU&V`hr9(1q$VRuhRl+TG`Dyxl|qV<2x~1ndpk}8Ryi%$GiT3Wt)Wq`<x}Hw&Re$m
zz2f`9{7eyoT;;$A4blq;eOI9?D)0jE{eXI{PPI~{(>WqdGq--T42-2xE_43e1+bRA
zgU7CIFAPZ%I}Rusq4MsKl5?8P{A|~n>@x=PpL=Xd73SOXsS&5aV}&AIUt{Z)r_n)i
zDC8g**TPalVAG79x86lhN4X`Kq(scyKm3o51TG9WZ+=MN2UIHyB(Xtfk|fb=ZEdo)
zwoH;lM6-l6)d)oH?dAk1lgugfXGW0fT&+KInNDGTUYO?=j}U@#*^wzTWBB0ud%W}h
zJN)%u{|%i^C%=SMzMxLpr>9$?Ma@qL;S{!Ii1f>>u5CG0>%5XzytC8e$yz&(oH2%3
zF4uyPtgLME@=LF?x^jl&PM7gy;)IQDuAP(ree=(DifeURh5U;?La9&3Tz+=UAHVf3
zdODs@y_Sf1=STku(%4~s>pFgpd74;^R=)vrXWyedmyi5r%=3)h<Yvg9z#Lo#<@>(F
z{G<1H=lys2;SYa8x7+=Vm=}|6#njR~gOpB(O?n|qD;oqs;HH_1Y%BkNi}!OH^Jq59
z4{5>D;u<f$@Y)>n$z+mWX#a4Sn@}=Pe*XRDS&I;ajT&C1O1Qkt+A|l?p2K_|JQ=5H
zC^Jnq8nOSuM?9P)^rBs&1I^=Lk-fWpMsbX`8QtCirAnEV<qfW!TVr8qfi%t7++1gQ
zd4V)bNaB<%J<)ixIDzIggVf}h8;5zSnHy)jR@tXx?s*<bmhsECf5AsLud`fVpxta?
zbBT)L5^*n}*aSJZ(pu;Xh@7NOO~h1syZq=c|2?yu{BE&%#ie+4t}0((u1!WM@R6P&
ziX(pZ^IsCjW6qqpKpbaGCR18Mv*jDasY$y+&~#3L$^Ah;>r~ut#T-LA8qt05fNV0R
zFeZ167>VC#(A-)BWyz8lfgo(uS-Y^o#^xfv2bhfBgKbC?_q`i~G6|3G+{g#cGQ7A)
zFd5?$G9C><YevHnjaG|=R+Gd1U92{o-8#q8@)Fgm^AkomSrubJdLF{_(AHJvOpbZM
zf1L~d;;K32a*6R|0!R*z_BrkyaqHGy!my0zImwQsdqli_8`C>NWhtUCU_TXLInJGP
zLz2>?R%@W7WHuc$==aF7lqibn_j^ocW0J%L=tsvL9zNRPz4zbe=;#nnxf*DrUc>Vw
zgF%PsY&=gEi^t}iF1N2wE_YX$_~QFq_&!0cf|bSDk|&;4V5KBzw^_P)hFLU5C`nK$
zxro77rlUU68f>Q7`QSrr5~GA79#24L-1+cb0woDU$(>u*Q9|PTRc5mYDLhOjXtr9k
z+AXro;=I9gu6iD@YwPqdGn_7HW+1h$N>{W>jkQc?Q^u3Avl&Y>gcNwbk5sN2_h|PK
z4<0_?<Bx7~cyPdUHe)y%&+i72SLq-R=%k#NdMvo#lRij~uvE<}dIF{V`PmFEu&dVU
zcwRu1%Q3}qGS^xymrI;Ke~E?GGEtN`Iv0{Or_;=3sd8G^oU@%*=RfP&{<(rWty3pJ
zEY%Z>PbrT;`S^Z?m8Y*z+gc_=V{=zvV~SP@1Qv$(wiz5AqJ+XnlihuX_~x7Z^soMk
zG>-V-`umQzmMM-8JM;zvzWwd*^7XHMldx2x)~JEBu4?G1`4p@>kHwtBWQt5V2kDc+
z)X8H>(}Y^Jg75hxX~vuH{ECOWkNE5F|A-`ZgkY@2YU2p(c`I}-Iw=%V`A8Kw&r_*r
zFDwz3%OpvPPLh0z$<DzlE}SCsveq(-#-wSCF(6Q^t!?tkE3dP>vd(xsAx%<rZXsoh
zOu87?K2@>>zLX1I+Z3Vrbj+oXAAEMqhld9!;o+mv58fxe^)^5K>%S(ABR;(TK2i#7
zrsy1X7z~Ge``h2;^{;*76y`2?Ql6T>cc+#1n=rTerTXdDUYe%Vs#Sc?BS|%HzVj;{
z?LOj1Kl<BW$J~4x=FW(^_+f>l)oK%lCDJTKCn;8Er!cp3zuzY?k0+#AoWEbi^71+_
zz4Y1%=DFmB{RhHa)z&!hDo!H6p2Rg?Sf<%tAS_o|x_qAM`U08NSmSs|*5(2r)?!9u
zCfko#=pJ&MOc|7FY`pwBTj`L;H*e7IAG3FFo5)0XvW(O*>n+K*-uMm|FF%dv`&6ps
zygT4%4<-+4wJq9Xr!ddo$sF_F5bC69O1)OaQy#N8=4Ze7DTl{LEHA7Q1YU8BQ^>;z
zal(s32yE2F{^z%_zr2aP{t;`(2mH-{{0l~%zLOZj2^tsvoI+UQXn--<e2~Rji_lOm
zRnb}#MG1fVlb=y48CF*}m`tWTMaq)S5L(YYBTwU<oyz1tts41xA8j=XOWGSS+}%ef
z85*}<J*n{gklNZZrFM<fYJ}F5^ps}V!wXyjB#j_G>M=e(##0_12HSZ|vUi_{w{GM4
zJ_oxGx!4NuW-}y`gM&lXmKLeis%UL#w-@Lh9nm}PaAsqRa;40|;sRml60mBmN~_&w
zVR3<4qnZbNuGpYF<=hB`od@McXY_jmn)L-rC09M$KiH=?=<#@WpERf846+O}9TGpj
zNA&P6CLUw4&VXJPKU_d(tfX9aYbeW7dfg6jGNae;Vy)oO&LjTzCx6TBJ9j_`qBvnX
zi|Gvp%w|)t&Rt}6WsS?1ud%qejMk0_@3_-8=g!RwrgQAS%zZ8p_~jB_80LRR9=J+{
zCp<haq`kFCSTB=i34XbP3KTk;VrFATgFZqVIuEv)4*Ccyh=(Jl4{uQ$ciG>5h@Otv
zxqpkvWCTWXxPQQOHs$GOp69~(%P8fuw6a8{T5)3RzW1r?*>FmMg2HcA&UH@5<1uj(
z)9(*lkgqji5IA0^*4*E|N4MAE#`PNv1_P2LVKf?_+z%E!<s`POF}dHbfWpHXi6537
z6(^S!a;;z|F;!^u`hGyAQklz_#PQ5==JM`QtybmY#mh8X3uM|71|dR9($ryJB$UNF
z`1_pgpEIb-kkZ5V$~aP`i<YJEQBvZEA*F?7HlMkQDY+7*6^3+v=vsUWjK+-aKf=e*
z>_5W4@e5YQU2HUCHXh=8k~_CPL}<bB?g6uC$j0Utuf6^){?R}BFIik#<?Q)0)Ee&M
z$&5a^;H}In$@Ym&F!$5TpC^QYt<&q#NzB^HBEIi27!G;&{kPfOJ?5>q-g6mvu32LY
z*vwhASX{*l&&PT`cqNPsK>Ac`b+R<${++wr`QSR^ZU<|MGP)zdYi;I$X`>mAy6o*g
zq}S~;nT<h6&YZcx%ddWoMr(;Uc9U?}lMP=~AeB<{iQCV!2!7k+di~0^$_vazj=56!
z!Dq*OeE$(1n&xm9@7AwA1@r4qV7|U_hSy&ECV%>;|3!}Z<`bCnTQRrui}}+qci)>v
zr!i~Gi};?JW4^c7;oWyXoCi|#Zj}BM%#{ybh!H+Q`BbV^vMl4_{cRrHxW%O3bEE3U
z&OJU0Tnfxbqb`SsyA1jRX3+$MV14~8FTe6S&DIiWoaQnopTu1L8s=Yy+Wf_hb<ll`
z@?0A#50a$x@XKZDi;I++3#?o|hY97pN$C)AqP%uICF^uJ+I__IZi(@^Z_xE0QS0p?
zlbC+@m@x3^9UP)#!|wJjmW(rJ&$7C@Lb=kU)mmiZ%qrzt2_ao8-Qh^?=U8-%P8zL(
z-;TL88k41NpDT~!ZilzudkZnmSg*H`vIN3+BGMwyN(%2?Yr*y)ej9H5CCyvEq(0o|
z@cv_jfIs}hKj44yfB4_B`0{zCGpC29JV~`(!}DCgn5HRdnvlgQUw`#$eC?ZG;rzKe
zy>5s9^q>BMW;?_Ws{}fsG{?L+cPt?0`i`z@Cw^l9`*|K?EC_{<$LyfzCRvT1lfi|M
z1m!A?wG~hhW-*m)lFxR#SLF+dND`*|N60j#nv9SK+f-*0$WpQ_CMX5$?`*RcfSAq*
z!hlw_&Pr>E#pP8dlPPC5H_76J!`*$B7nj)B++=lqgBM<Wl^0)pl}3GmI8KS<#7XaX
z9*t&$g@rb)R*QDKMYH9G8T=q%>&!XIVTGl|Hsx}OEHmspe#CGv;lce!=p=DkW0^r_
zF=BKG`}dLY6z$<8C8R<K&uO|zMOZ3fjA3tgj|aE!Fzg<aBr#eWy4^lUhez!1?{jo?
zG{2bRctU^BWipx0b?TbUHk)V85tb@$NTyIy$P>vza-<m4bJNDTNAMSX1Ry-0Ft^mQ
z#oEF{3Rmq4YfTn6*T_tU^n6MIm}rETrH*HqX=D`9eXxzPV7mJxciv^iBs7$u;YsSg
zNAGwKiEslk!ww4zOPo7@nXkO?3JVL1oI7`hmDLq|-^UAl*BTY#WNV*FE{n%nOW=F?
z;E-mVJK2mzW2|v4>qf1L6prHaXlI+jV8ZtOhfHQul2kL9#+dw!bA<~_6vfP@Q`b5b
z5{q*0CU+$n4abbTL$WwwHk~jU4CX@V#ux%Wq*|$>l*eQ`VleCzMJY)VlV&NkdYuaw
zFH@^Gk<z17syfFBZRY#jowp~$nNJ*Qf5-FrY{HThPlfoN?<ygAmRPg~JkJ9usb4tD
z;??t5nay9mHO6r@guvGcojcbFAH0M3=s(e7#HE!Qn{vYFu)}j#E^+71heXpc{mz(4
zZ^WqIq1)~7z3=@o|K!jA8P~2p&-%t%UXsoa`+TJe$V=s7&ao=eow>sICl33f`cbKr
zIeT^!AuQVu@6+ua^6tAI(CzkJ)yCMo6l&*}vk($16b6MAIjJ5Gt=(YVQhR|~qk#~D
zXgXzY|1nyp7~|CP=h7$IFqux+-hROT!9E8Ehx5_q^|dV)+RJ(QJ)4(h3ywu0xl$~G
zUl+ao`#mBxC=uYRQ<&%3jTGW@Vcxy_5#gf`;KqNZ$%qRpH8$mx(P5Y8u3X~oComrm
zy7YQ|e((4GkU#&&|K+D*&KJVmeIJ!-fjM)`-+%8qz3ve53R~d_QdH?a4RcT!Eif7u
z8Z9cd8g5c-#^K?f!`$jm#r)Buha4Rpb9{U}$9#2l(|HnWeZnjHJ1{q20_L;D4fgcU
zF)}-8yXW7tP#)z*9aSk&IkQgl{3cfBNtJO?r-wxtjUM(84<C_UeF1yvm`=RM$k(*v
z9@E}{jina*54V|(#`KQ+=-Al>U3>ajHqLJF%(KsM`SLYZRsd{4lfIVofcOM+@g(q&
zbB~Inf8GN9l=q>eq*kqP_Ur~=xpVg>{XvJ&`cxZjz(3*Zok|AXWOH5@wic;kcK3D|
zkNfloJ%04|&$;pNI!QdERQCDG^UvaYKDA~YrQFwy<CufPeRg(sh@yl`=bq)o7hlHn
z61v?nqwy~7=923rXXz6ZkJDNENh|itp;(ZbJW=yd8b-SZq{GoEEiL!=l$RE0E!7FK
z5z1uFi`ZIx&$-S>VVHFeu#fLz_dcd7Q<ht0s(MO1oUpaJ#@%}l@TL=5vneW0s8kzd
zQOvo`P0nqd!veKN6>H#d|B%UG%(=7YIDg?H>+5HD?%5Z(a`|b_pS#SJ%h!17sb^VT
z*&qxn$h<WghGkdL*M{+UOcYP3H%hcxO~>Kw_c++!<H7w$^twHiW75qpcoC1W`}grj
zJ*2hsWY#S>iP5<=V3vWf%%T~!N)@RTE2}FkE-g~6R@vLzXQADuTyY(n<Ktucg8|)c
z*I8L9k7lz?7?z1<(UV-NT+V1-jV;bM`1J_q@9_u?<WssqIrpADDNm!6t58-~mMJgR
zk?9DPO$b9p=u0XxBTG{1fluebP0ap1^!>N6qeB+Ggyv+Q5`wS4{36ld2$RH^*f8w$
z7!Ui51|y!j_AD>I_*GU{H`v-bL$kR+7?kjXAg3@C-{_}*!-dE(KpaP;Y2pNq(-Z*&
zLBL|W1z3(dN5sjLorgON`U8?QVK$5AiU9?+#>+b|2q!$BB*|P(##$UztCUwajb%6-
zk|Z&46uG}w3W6X+DUT?QIXXIKI2bb=PUg-?)oPt;wc#p`N+G3kKJWSj|0fDjPrQrY
z{(P=MiS+OQ*=o8ncX4sT69<iF!Bi@o|Juvk+&jYdW)vSM5D*IprD0*RL$LiTN+Q8{
ziYw=qiAPgXuf;R3zQANU;>SPwKHvV<@3Z^hF-j^j+2=3+@`rrujc+sP54dsTgA-Ne
z9C#OCx@>(u)q<E`>O!0-Nq_Q*lbCWj#A4an+MsjX;r5-6S!%EF{`=Q?<BhMnjK<m%
zGHziDB~EFWnUoJcc$f6;l)w-0{g7I%&hx8Fbh^j9`Q|VA?svaSxg1ih){sgPMG?JT
zmn_rBnxffkfDjA@1GLVFW(jeuiJ}-!xQSe?wUbDgSAZ;8mgUue>`Te!(xXGF%?8T<
zJ20>Q-I%u~JNOUZBoqnOQ(XF7n174mV3cF7^Qr}>l8WDid0x)`4f`REW6Grx7L#Ls
zJjeX{hadB`ufGgJo;YUw2F#7k_~@e#csEI%dw>^GtJZjC`5C&sF7LeiHgCM~2BlK)
z>6lj~wR#O9$%jJR(9EY}p3Q&$yTDwItBe{e2)~LDy}adH1i_$!kZQF~zglJU#b?+)
z>cep6CSWc20#ZQC#vG0IiPAeTJ4V=y(Zw}_8;>|&Z*ca?d9Zi+<<EY~>tFvSd)qr~
zNe>a$IJ<d;vu9TM;De92d*{|~z#NKd@Duf_d<lFS<QAAk5#_S8g4o(xr*qul=B?{|
z?bUCy+^900bg`O~uLmayf^aDt7^>}Mp1Jf2YwdNW<BSZLiHMztL(XlT<;ta}nMN_A
zBTf0La~qh=W|YE^X0t&M`0VZ90Asj(=_`bxq1;qtGsneI;)FYwWha_tPP0zuxMyF=
z1s{)xlq(^=QYaxwW)c1UL;Ok=Kdewp`BV1T+FW8fI6$Ww9miNwgzqEDB?1W*JwzVg
zLwFKV3RzoeGwBXdl^P3Y*O^R)Jl?*;g$q|PW=Q|=F<IE;gSXyiYwHZ7;fU>nyNo9h
z+Db-)5g)(*KArtNmRDA&HX3+J(rh&G{D8&f6%a5S3>l3^D9>X&8WIK}z3wsACSVhS
zz$42NH=P<<?r(3ivAoXi!+X52{uIVVYDnM50xFHj4jy2&hQ<oygLb=*chFgi$uf5K
zc9>2lY;B&U)oOEi+~eZKOVnyrtj;(-+GlBLfnImuIMoPz-)B4<lckwkkF$hmmT-LB
zp;9V4!DC}`PH@hv%C%unhNHd&YR#|@m5?}|5tMvvlDW}RT*cp0xtw|_VDs`8!;fyF
zt;Vla5OIlxbb|IJ&3c)N9WuK6CJW6fT6(lAnoHdool(N8U;8SP(SUcZzt3|oe~rmt
z$R692>x;bi-iN&S;){$Y6K>r22<7>d$`xkOY`&}=7g@}c@Y8ot(S8lXfHX~+MN@n~
zWHOl&1b$AbfOfk<5-05M?{MwPbL>8POrues|G%CmG1hUu$vYrrzk=49!EnfMIB@OM
z*^I$pi06BpIkQO=MeObF^XhA_qm(2`Vv;052uPDzPO&j~zME{FPG|UmB}*OcVmh4?
z_&#ZxVf6{VMMvlKV9aOjynU{q?#_?@GcT<!6L=wk7vwBp>ncuGB18z%MYqi}XV`e{
zCHB7mQ%7b6JPR^}r(U_p*IvI&$;<FiOrw}eSYe~CI6LhS_Xa%wl`A~B@i)AG{cWCp
z>M8Dje1px)S4gul4<BxG?(BJ<dg>|m_B)u!McNXVDJtKQ)TK6L#=4TZaPus4Kfg~r
z*0@Z+KOE7lH&|a@B@jMG-F@OL;>OJfy!^_Gl*%P68e>7Y)D)-we^Mp#JfExAo_1r-
zQADplB+CTTDB;$v8%}~_Jf&W1&Ziis(=qS7_a^%X`*>cJix-}vUT>mv?wv7~Xf|Yh
z`E^RZlgT(C;X6As7YG=V)a04u7q<_(|Kq>nD=SNQpN2W7FxQsm)^EoA>MIv{^%Ul_
zD5m@wG0*gvhmW>7yX7$7-|u4M)0kV7?|**Gt(#=+4~IGCD+I!uV}A4YHZQ#NJf%_y
ze+TADd0f794Jiak9MK<)$c$tbCEUGx6JH6U>5OW%j#BaqVs3KGC0Uk!>6kzD=DR%n
z{cj_D<FuBFAjulAIghXu;FU_$)~l>M{}e~Rc!yLQN`P;{25|V)3cJ@9G0_p|h?0O&
z=yAq3tk@W`jOFzfkMA9E{rY=6^~}@Uzj1?w&9j&+Ly9VuO2Ea77dbfWVG}*aTnLNw
z{M;PT<a$m{ddE4h{ASEEqZv<TRLUjJtgTa$0sBWgBw0*m6kfGLHXJcm(L5C`7E^Ip
zOE&DVz5N#Z`w8Wi&qBFP+XLmLTsU_IUkchwZFV}3u*wZZj79@K`0!oE;|b+bohw(K
zLHa(ENz7yd2Zs|p(Pe9Gi7@92Yh6_8oCRT)S&~FMab^t^)9_#5Il6m`M!kkqo|En|
zmT7lD)b1kJ0z3i`lF(WfT1`r4)*0P>geBMQic=Jtjce<iy|{wNG&aphvW$`nXjgnj
zvk_#Pa~IF>;aj(Oyt~cX>M~UyIw_Pr#bg}OYAv$RZgD(#$aplP?D;gyWroK`jQSnN
z=Pg$Vt5qtZ#<1VTS{P3z)awm`utdF9rCO~}uQ|C1V=et&kI`Vn+Ttc?LHB5nnzU5o
zfJtYC`a&zOfGWqzS73BX^!P5m^r6~73RqrRMp(<R-*c5TEm>VzV>%h(3BzPKWHKBv
z84o#k_AJ$Eg*V^)6?gAHpxfyX2JT(<`+brmB2H4WG-h?N<%s^C<iwiL$?IrsNmE0X
z6@lItZ@C|S^d9FL-*n>UaZFGyp=Smya(gWykrGN3pN)+*!YtwNtq(BTQff5N;~90K
zS-H5x`RBLjji+Qbp%sMKC}ttFEa@JTUY}>4zR2yH54m^u2Ak*3a<sF{;)Wp0rtIzR
zu-IPU%$ZGY+_;DD2b4->$1BV-H%Xm$h~^c`{O4JwQA!cy6RT<F65i=7rryv@rV-^*
ziB`SARL&TU2N<jA9QBCfgi6H)OJS)-Dfa?TGo2qRmf38IQj(SBRiu;*2YrlzG}BBc
zQ|{coMI0qOdh`%K@ac4p*x%p7TEo5jw-}E{_+G%`;tE2zYFrqU5mFIH6P|zO3e|Fn
zEOTy6#XdJioM4_hDHW|h8|I%YsN4SQAG5xGp8CtrBLWvOD+4;$(ULitSQ||i`mDYB
zEa}lc!*_3DMM~>zgRj2vGUuLN$BG2BAu|SFDY77-)o%05w9hX-yvZ~Ft3PA){D-s`
zTI?JgvA)vg;msR7b?GvGC^+s87>%c7SxP=RKG(Eyo_W~`2Z{XGi}uYY-i-h}Uvd7z
zIck*#YpYAl#$B#my~^GDcZuT(I?{BGI;@{L<Jzkx!${+TLBOKsZEvY0wOS2dDg0K8
zm6c@zze2ayVSoQIje4DEHl<vuAcQmYK0ZFCUTgB)GcTf4z{LyKxNz}lqA0@iJZ901
z<HKD%F`HK>=hHG$Vr`zi>)f|C|K7jo^YE{K%-Z^SYNs)`pN_c>ejW4X*#@t?@iJSV
zz}!;`F9>L~+dMVx^Q*rDb1#$}cRP&6Q_?J@m?V{Xu<_}b|8`!X=X;#Lc$QlEH)GyC
z?y$D8=`gn$R{UnnE7c0VQh3cKOG``mewjhP$HBp#!#tWfsRJQC74r)hu5#|&RpKN@
zDMb`TbdUG(M05&s^7k%&XPDccz0C`&=MWb+5T0;$P!<X;E2R`jiL8_mDq!)s%VgbS
zrZ?_kMM}J0<Hjq`uybXVhMW?_3SliO^oc7W(NdGr5X|E}&c5^#*WbHCtzP5d&Mq5k
zOHA+FVs&K|U%9>EIH=it+ZGjBTL^(#J0Ft5+WA1mr#+=o$fb*Csa2Y+EVYQI1D<{6
zS?=7uMVihqiJ&AiCZ2~i3A7iX*P{sNPM5`w_9q-4+$XgfeQZ%-nW)ocxwVKCiYT5^
zYgD1*)?$`s?Ck7f3|zka6dS7>Y@S=?<LzJa+|yM$gDLGwiBfXK*>!0%7o8@@-CFYO
z-ORPKe(gZ}dpx~c4;Zx<scfye7|{sVD%(AzQg0wy&JCulQ~YX3<H9*alo0J4paMaq
z;<0&tgXMF}c%i^d5)2v}Cm<xvc8fJJ<>A3GtIt18YxM{zJi5aPi$bto5!BDN*_}y_
zY)G6*X0yb}mY9?z(v)mQ5SGz4WjvX2-07qI5@8rPpHbf@i6T_s;rog#P4T2prZu%%
zl`NI4ZJfnKQ%V-@A03gk8cgGaac_ue){zK=Hh}Xz&60#9jxhW8sGNTmfH3e;0<v<6
zMzz8-SDvO;Dsiy)kR;0(4F_%_bJV9)Dih7zzQ20q8tbd;TsVJ$g~c`Yb|2H}9&_C3
zQYi;4EiPfLAxlq0?15957dmx$FYpUKNAG=1tGR@CZi6(5P=U(*X<gFdmlW-=&Lk4p
zl27~cS!PEEL=Se6Wsk)xOI%!8Lai@RU93as?mcZazEosVaH(D8?xS7SUU`m{^(KQ@
z(jQI;e4qZ|J}aA>cuLUkb;uIuiF%^4kZZ~2zUoECrC9M{5a#X<K3SGhE|rPm2xBx-
zD%!0kwQ7S_tAWiTR?eNL)9sKXGlY@Evl&?<!}mh6G^G@V&c4!fLzu<1x>OQB@Z3;?
zut?>E<D=P(QV_7%UZ7d8;|BrJ;xe=8jL~Sw`r0P-YK1J*tgddbw!VS2ZkR$UpFy{S
zBq8vn>$D^ha8ixbMWxm{PVHws+do%Or+Q;P__zNPPn9gC)g@*|gYg}oIM>b+o<dkB
z=2&Spx%9`s$F1HG*Vf9s^!000>H$X2kiv6mrv;m3PTHi=;3*9cZtU{zU;dbDS1xep
z_C01P)T?Edn=Q6)-RAtYt7PenW~0qwyFur;pGWMcT_2pL8{~%TW=`XF|D3n#oDl5m
zufN7~&ppHK8#}byO{^YKtJb)B=?c}7&vclw`_A8Rw)tJM<|^7*ZO-{x$P@mJ6rhE(
zeMybM>XgV%_~^s;h-M?=ctZVglSZR~wQeHs&Ye3fFD)?}Pgq<kqhw0Ix67~Ie3Q{=
zM5l9%FX4~A`yEciu-)fBY0+d)l%~IwNyy%q>!&b}j0XMbn8%;M{K{IH=RbkD)lQ(<
z<d}zz23JmDe)Y-)?%ux7EVaLexlW@T^LmcCK`HT@F)t)Q=Knl72$fX4{`#vt_w1)*
zUMcy^hADgR{gg9}@1Pq?Wc((~c@lGLusS22PPy^%hjYwpwK}z0{WoIniHzakfH&WK
zlhJs5g8BD84RiZD!(8^p#NYqV1bh=36*40b+G*D*=_dE2Qb^^Y$|dANo2_qsmHUGZ
z-IWp_ym*0eJ)l+-gi5*7$N*`E^?W9+8j+b&?;JDx=wp`8ZqPm2MGKEo=+j<o(c9mr
z*=W*gwkTI>v>H`<y%Aa)lyvQOaGV-99Fi;T|2E8($G5-r4X#|d%&m_evAo>o_;5_4
z-r(BhtHdP_IgZ&{Epw;zG}+EQWH<i`xug}@36IT9sN<*!tuo?!e@gZ7$2>C(8IJ~}
zX~r;#=^Tw1ji(r6x&Pn+3#~Sj(FC<-X_V_++kA=b+m9Jf1>2ngq5mO&`Uk7III5io
z^G2W8I^{j+++6$%Lynw99A5u`72o2Qt3<|Pgd`nBjCKzw)#`*gCX8cH7Qa$u`KfCh
zPA050JT}iQ69z7Lx9JHlOCZSN1TPF&X;_YTyX@V*$MVVo#|Otu#yzT~60MS;o{V^I
zagn#SyDT*toLODtxZ7bCM+~(gieoA|qap>q@Q})PQWsW}#Zk_CvU$SgYN|?kT)23F
zGwbVg_d7I&U^wlQ;?rs^QuTerEMm5Q53y3gR@)dWk;plZRclQ!9U!9-rc?*vk|}F5
zs+9mKG`DVk#8LNvI7uj#Yb-1-vbD9z{{A8N?>}HVnIVK?I+<{|zl+Wc4<6iMe}9+l
z?QN=+GHXi>tj$Q$lqibwvrNog223$PVJ%S<ePPViG~(f}et|Durm?h0GMNxI8hC*s
z3_KQ=noOqyrZYp9W(3tL>n}ae<I#Y%g^-QQYv|dC;b@A`HQKEj(QJlJ4N^$_N{v=F
zqA@z={(B#B>D(DU`0ytCkNT{izepe?``ZuMIDZa-#a98PKr)^t_`VyK$i+QSxfW)z
zrE9e+)oPXL)Fr^iy8EPDD&r}|<x3a1dVY(^IHTQaqGXwJ&>##-)G8s#EF;<5MzzXF
zf0-njVgiZqO75sNhUskPrq-=+;?$m(^XXmSez3dCbTTAMBMuMBEG{k6>m9SRv(1AC
z_o-DXl+K=UiN8r04SPI#xZ~Qyvxp>~@yxZWC<$qn5{7|`hw^j0;7wx-?abdn`!yjW
zeYDNT|NbvH_b1;&FD{d!k-j`3s4IyF#FG)#T7~A;25<cNH#qOj5WYbQLpn{VRhvXM
z#l|UUgT>&50kKK>=EXMu?(RMM!DafrJu0O-2M2r9>NT2`8rR=_hm8vt_@h7k1JZQH
z^$*^sbJ%s0fS$q%OSw`4D;SQa*i6q?gD4c*aivgeEq>tRd2sjceSEJ*5cn)Ew!p&E
zPe1D-`gF?p=5>Z?7yp^p2+v)`1T7+&8*B@KP`PlVaI71fU+&gq93LI9y?uu`n~|o1
zTem-^(QMG`_lS~&o44-p;fEjdH}WU^=^y_-*DjyuaQ_~=JGVg!vTTC#eT>yu8@Q?j
zGVj4VTaseapO!oLqR%HW&kD=~IUg?hB<78+4Zi;8-(c$#m{)Vm^(o9%7?7BhuU~BQ
zAMV~`7+j{`+daX&R-<tm^FKI^c@L%JDa^}YIfc1<3R#Lm(cfa-l$HAJn2*valUvsr
zr(OJ~U%}tHf(aUA;xy*@gjP;qc9^><*>3lUhY#<bV1DN&^~Pz;@ABb?H~1U%Q~vai
zzQ?u8=Q%pK&+g+pPhxIPW6l$p+us4^(X@}<(>(a`k6C{0CCuU?86MKlnRQZPJ*U~G
zO-7|r$Alik@4du(UWBPgT1$0GD!0<p86r-RTH|?=xLV;-Z^*b6U<ZdR)!V$++anAr
z^tx4MzNA^Ha^r*dXs#^tjjw&3$!NgsTes+U2S_DRN@4|hYg{m%%rNnZIA_6W7pUBU
z!4d>MX&Q0&&OLmuMpz11TxtXG-1A>yJc;OK6HM<hSGFcZ7uI+fH_>`PNpIl|A2ZYG
zDQyMe0=;S(@%nc#<wc%vUtzrSF5~G4Z<a9_4_RGbX8-Xnqw$yr4<2$my~~?#z0H5|
zPyUP-zw#{8Nx<>p0Z4(?Q!*Lkm08~fC{{Q=Rk6?Y$)T^c`COB6zv;1LWP>hyAAL;y
z$`wqxh87@<VshN4vCyGxXHGkCrty4_uvX^6(`Q*!83>KZ40;w5RLY1zIvPPSx`;)a
zgbRxyA06+ZS2mbMBTBwvI2cf`Hz<WZqoZA(T4~T;yvQsb1D1F?BgqmHoiR0rL?{#n
z9Zj5*kd#!*RT>K|;y6K~kU}9aNaYh|G2I8ZAeBtCVrWu~@VK^h27ek+@)Wvv$n3_u
zRL)+8g*8OANoF#z28AV13O$=Zxq;K~0pIf}hXD^C++q9CLo%B&9!HEO_gGz9XEGi$
zo6XqSdCY_D9frdZFTeB^l+g5!A2S(ssFXd<pIdh>Hd#g#B}8$=-u^zlUeCEFWX?9s
zSThgmz6_6NjVA3JvGe9zTzvIK<l-XnWJ;y1sMX2{304Y(G3Ypgphjh}#pRc;&<aw#
zK!O)ikuWxy(~6u<P)JfC@ukEUmaUql*WE|1pCR-$Z9k>Azt7^v8s)%a`|fR4x3+lw
z^{<lZn1>G^A%!6H730y6G|TWj4<RH`lAyKgvcyR|w_NmnM-!=(tGSr6aV{mL27zv%
ze23YYGiLxzG?`(J_pu+1&|6O-n+q5;1Q9YUlc9*C8QN;Ff|;G+`QBV!>iF=$iG^n|
zwOWHDP0=Rh=xCqeXu!Su_prur_uf4&T{ug-)u7Wk;@-WxEG;jSW+_n|k){!N?p{Ap
z#c-4hYn+Re>%@Hav;A`gbp}gdEaTgExnHZW_Py_++5shzS4xzcPmIMz<E0Uc(*es1
zKB39Y$um#}CjB1eg$8kwx&%0i35CR8USam&5r1}}!N2W3=Iq8f-hBIgn$;drG~&{^
zD{QQ9a`W8}Io><s>#u#2&e0)jtLq#e9h2#lX1h&$aS<tC=kX5r?>)vE2*Lo*6X?vf
zE`%t!3Neo#AMw>!-yn%J)oMthszhU0U0vt?-3MeQ<4!!|rQRdrUmP>N`w`V|d=I&>
zg)|=ew8*LD#Nx?h=q%&jgF6f-eI7i#&xP~P^Zg(GfdAoN{Y%cDzsNJsJi~OBFc^*y
zz-Tl@An}#Y(&7@CHFJ?nom=OesEa$&rPEt=<-?+UUHI00k>?4_>lN0%n`2IHTlFcJ
zx26LYKLzt?zej1ILGme>M-LwH#}^v>#iw9?@!S>G*EhNG-iLHP2j*BYcLymj*P`Gm
zB%cZMd-04H`j1F{b&R}sgYxUYhiGr2jE5=u{#dZ$q>5)UjMhAOaPJAsfAGT}@Gt+>
zzvBG)3pwU7!_oNjV{R;2qVsb6myh}6xI>@>(+AuCfA0S4*|PM^)5D(6EVunW-lVOp
z%yK{(c%ab@pvSU%CS)1GkaR(Y6ui`p{tZRwo}_LVk{BdI98sL%ICf7D8r=jMrhxJ$
z{mJ9c_Ivry=wj{6N>mrRQH?@3pf4gLGom80PVQ%~wZ8Se@B93Cqy<~w_y%<^AR<6W
zL9Nu2_@u&Hx{2m>YW{o}5X1r9PK43|j77}n_+A@d)=2MZRTUGh`EYB48+VVn-yYJP
zp0KsH&b#m5C7WmLm<r$O5O+IFZ{J~Sdxvj+{hKUi6E-(DnT#eFThkj3*w|c0Ny)*Z
zM?AQHg7ZyixKhnzv9l}3m20X8C&wec{?(Vsvyx6bLDHd09b4P`EZ@IFZ4HzK#qx+N
zddx;N<Zk#9#n~CzjT-Te|L7EOH8~<)y8&C5i7G>M^oXPRnA7t!ZaniUZ@=~D93P+X
z;iXGlxpD=BVv#Oz)-s=`PfpqfgJF|~fWR3`TI57=W8^n>6_xemI{!#O{EK=AKkf1Y
zq=WqIf>MWsH*R9%uo;Xo;3rd}j>JnrX)4-zPGT&*wjvY)=T-9}WeruHleC(FI$M@F
zXVHL)Th!@<=LZq*E+-6oL+;;yh;nm|74&**gki+#?FYpDF5iCsS^n+(#}r|gA}^3y
z6L(t3AfT!%CZiGAXomC-XDb%-aT71+ng^Q~kQz%~rtDq1M(HeVsR*KowvntS3DOp{
zLdSW&<YuX;9=wMekC4wkkB__f%Asfyb$pS5t)6_L$~@=f@Q}CP{u!g`l-VL>ed7|h
z@7(53|MXAkbUM87!b>dXOXl+>i!@zLWqQi0!nsvafc2QN<mmWhCDSd)iyUh{%6fhL
z%dCRmtLj%zlyC^7^JGWI9KRE>`^7JSTD<@#;VL7w5x19BNzcc$%>va`2xrk!5U2o4
z&cKySJS1_<Y&=6*gNtIqI3dhtJl__)adyZv*KY8GAN&}pYbF<-DC)AlzRt<R1FEdx
z3$J{!iK8>kBxn)`NUdmhnmD!bWXxomA*7_&>roaJh}8zjKoo}*dBJ!*W9yk`h=qr2
zhV%id-D7=Ylkxcl&U?;H&DDI0dH=_lq(}1NON5;Pwk*lK#|S}D)I@QM#eBi}**Q_5
zN#cl;<0EE^8E5Aw+`RD;ufP5}|N7tjYm!!*D_5>@czi-plw8`|XMZJ7_YR_{MI5&X
z?5ZbybuY!KCOh`2mvz6o&b|96Htd&KpDz2ACPqhBFnawxg7qEtzwreK8?&;hYt-tb
zO4*9FY|O4*5=l$$9ZCqaQaIx%XG=n@3FENoWib{4iBOXEu*buL5nma0uvx(Q<!30)
zj~U;4$aFkq<KZUH-F%+Y!*d?pJ>Uy3y+V6!$l2*RU-|0S>GX$u_q*TYr?3A7v)Q6~
zt5ll(OMCQsohFt^NQ$E5_U*g0+a3Picm4z7sLygaqSuQ_g<(0J<1LhFO_`_2Hn_?X
zv&5ggh58GF`tm=f*t&{FH_e+KEPfRsB5^{pSj;&(I$$(9qsR-4aooH6fP+Vm7>{S9
zX^OSB>5Y3zEXw0vipRoR_t<S`d0IgFbdtbh#l>frDA6zvDj5CbJ%Y6z_P_B(hyt8-
z*pD$EWMj5g%uDZ@R0*Z<PcaYT5My$@v4|D(*09H;gArdEcCh&;G2h_U%@;U5Jm=sk
z=BKCUeCbPHq1zwwz3=@wKmF+sSIira-X}A^_o<mjEb*srp?+eJU-$=<8<)|j#=h>G
zd`TfC(lyMN%PGgl2aHDND^-E#-ra{BJUU=Jo_=P`#jk<6FG|w03tAg%%-_9BFxX`G
zi=V>>P5D6?gIXPhiK^yNx#TBNNwn6Y8#KxmuO;4CR9fJq#0LR&Rns$;rBYN{k#2T*
z_TdF`a}7U|oUd(R<`W*h_b#2S9opRv;pQgU2lr4~^V;WLrZpJy@WBD|rJ*-S@ZR$1
z;311eiq;yb6#JKU==VERWsQ=GyePPH=RTcwkALv@|0z+>BaKFMdokXYWYZ<CvgFH(
zB3lq>54GbBk<rOhszE~C-(vHpb1vj;6&D9h6T8R1e*~9GE)^;BlSfRaW1=vk)9zs{
z%;!rU+<!=xrnFlL%VkPoEoIf%<{wY8TEAj#3|X4d?Q{Wtp81nhk&k+BJ`;7(YLZ)7
z&*IS;b<!icx{If=0dy0M7B#M}5p_lEENxw&G#K4Ppf!~QkF9HbRW~`Vw&{D+Qa6F@
zNz7=p;M$;vFBS8(4eIHXbTVnQQ3UK=x<a;C;OY^-|NIq3F5qyIvU}w!VXMu<gGZcS
zoKd#h-~|dnzt=@*jj1YH9ztt4xfoHYfLk}7Bh(2x$QUM)H4jN;c($*ZmX>sp5q1Nl
z7las8dWw7q>e4Mz0s<Kz@Q7l8t>$>u0xVhy!ay?~UvM!RkrkG~;0oEYW;xGTE|yef
z&B4JV(yYRH!F0ah!#nrLtBRxJW32NOWwGjyZu-L-oyqD!{xo0lRA1R=&^+rs;04ma
z;?5!c!5YICuA*d2K3fn*32OD-@Kr^U&WPm_D;#)>cA%AjC?LO>(~$xZ#*|@9HJxC+
zAcdy)H?Mgn3dl3dwX4_3XEPQ>feBie+ORX+=HTF%yC2-;+Wu8Kg8|dojOSi>flj~A
z`|p3q-8&y(ogfIhBuR@-r%e;AV^;%ehO#sSD&)&w{2HMU#Kn}gL61-?rn3owH5lWm
z%L3UFU_8=gsA3Ga4f5H~W8xmwa*2}xPK5xJMcJf8RDyIqWi&o#na-)In!G4EK0e{-
z_=q5kNYlK@kS}W%i$zmP_YF^D4Q3@E_p6t5Dddx$No$(}0rymr%oFeIujQm})r>5Z
zASxY?{`|-E_BPPZJqNx~G<fUKO46zp^wR<*70!DC;ZaG9aSlR_l9J`cnD)*FK^UTw
zh~<1iH;4&U#KVU(o(~mYj}`Z%qy6GnIC%dJjz@=_<TI9wDZ}+$LLKww>u<8Nzt2DZ
zr~ia6f8%#?KH$c+&++|RpX2=x-oiRhx7%m2m~(vci1YI^q!4twEut`{t_%+zJYZ{U
zi;c|=)>)2@PDz3;5AQ!@IZv66&uO(mi5A95wDJV$8Rd`u1@ZIWgln&0bck?nwYgi>
z!ky*b{ku%2BSzy3ZrysG=bwLx-}}Aq@cswyV{Oe3e((cce)(k{9UgOhct~B>7-N{t
zXE@*F;$>Oh7_(=yCtjY0Nps=<+=E_x)MxgquE)&%Co$K5nRz!Y&?||f{siXf#hBL4
z2092)Nklqd&<<h(6>;!z&aF`K)mU*)`W5pxpD<r6W~{C65(XbJ|A+tN5BcI(zK$p0
znd>j{{H@RN!G~`@VLo5XI6gk);^Hah(J^&Z|2%X5%Q4sK8P!kz74dW5fXkm-F}EKv
zU-dvec=(w4`HJ}q{NC?<`zhvs!OJgyp2Ndqj*gCgA@iy#nLW*1z~eNLMo#+KGq=uD
zr8%f6L9fs8Pu`-twMPHOb)?WwWUX3B;&R4=Yz|#PCt95vh5_JL?FbGvU0~LF_#j{w
z2JEF72P)tc&5P$t?kAd{)1j>dy(`ai{NN6!2OrYz_8DX;I_Z+=gp2VhdzU3&c<rm4
zEiTwv7pz~dIsEV@j~-0$9+FnT>DfaTi<-HcqogDXJ4E3LRaJ54_8oS2x7ph6VV&W@
z!$<U68{E5lpZRRT^x~A>Kyi*HGJ;VcNX3}px+j%exO)=>5-&pV1`!D8#so+@hYvWK
zTwtn-mtXlJ*Phwut6%vVVGxj}OAZbm@`cx4<EL-F&4c?7D9Z|G{Sz-Ct6t<L)3|C*
zT}30CeS9j{NUR@69)8r*`KvA^<f=E%5?IId-b2<p34XVQ3_XFgxV%7EOK5UpEE0jW
zAgw~gF&2l6baN7wX6PtrjQ&DUR28uf2vo%A;)F{{$aMu5IP~@<W}{QQ%|J<Ji&K(L
zk51U(VtmdZZt?1uUWT<DW|d%Tu*2Q^cR0N`!wW^yZey$^%~J9-M@o<Pma4Q+6fCDR
z+O3*yC+2c5pqHlvTC*6PQ)Cr&S<zc>6Y2m&ljW((1?up9WY9%zUqxDv7d3Wz4ud_g
zl0ZwWhw-B1c(h<=XNRQIW@~pFCk;iJa__-?wzqd-GG{VvdcU)>Vwo<RK5*SkE?Z;B
z^Bk=j*)#Cu6!z)Vn8y$K&w{zLja}1ANmLo8Z@<g>T1>FN54I$o&LC<kGO^7F78Rl`
zk<Ov8t2YaY)SAjT;@N`Enj((^WUo&?nIKoLz<c+Oxqfwn=fi{tX-coV!#V{sEhwsj
zN@xbX9#IrAJ|7Y2m~VXTo9tY=L{%Fu?LEWiUVN1Y5AS1)V|{&#$>f6Be9U4##nd&n
z<v2b*rmhUfN5|~0x7p~1lvU06VoW>fa(?eaiUMZib98@zGxG5OJ+heK@4Qagzl3WI
zs9jFd?$QneOs_|v70wvmdFx%AGn8e??(QyIn_F1>+$x$(ad2?R3(r5#>Df71nv&-^
zNt`eqPjJ>ESN4Odsz}q6L4N>()yb@VvYCH8B=v;(FUj>&ZR$!qArPZ*UXYDv-1+{G
z`25ZmYOsYdCBhg~WZ50f2wg@cT9orSzR1ztF72(g<{uSHY+g{Djc9Fc61Cb)kIxyb
zb&%5YjW1osmIWP`@*iCa_#Y1+uyg%+7WW<V<(NFp7@eOIbb6?u#qq%r|Ih#IKj+W?
z=#O~uwJ)&$%ri(|acOs-{@Mm#_~MsPN^$SrJ&uo$>2%s09v!f?wZpq_zK#?bQBV;m
zNEb`q{MlQ4^~>Ml@MO&V{0Nz!aU<EqnhIHi>a=lQ5M?u}pZ*)Xso=&}z~wlC=7wx*
zvc-bC5ASkve$2*TgWvu7cUa%N#QOR=Km73zcy#|hc{=5*Uw(~rnR0x1h%uh5s5m-3
zrmRYgwUlMWa+z`G_FdM88>mp@Bsl5u=CNttKRMC)EZ*M#TIR!_XTI~8xolC*R?K@{
zTAOQ#6?2mplxHL274!M=IlZ+GQo>ihbP2y={vTe_{HwzU>|B42#eMsP`S@eZ|EvG}
zzu^0S^vAsTh1b}-ejP>qg!!wly@r&M2M-={a&k<k)8g>(7cifHGV^DWU7V>9wLo=R
zcrOU^8TA|g6TYtCnJ<D}F?S8~JY914;k_r!f9Gr8W_@#ywe<~t^fBgtyv#T{I>Ojb
zX1-Wv+`fI6wc!RT3~=Hj=8vbkKWpX%jt-#Gg3>861P^}pCeLrJq5BcWRtRU2p+ipR
zoXLvNYr@40UFNuUn{s`CR}z#*78z>dv4a8AIOc`(F^~EQI??>#*$u9Wn!qjDx}4xn
zXKe3XW&Y?xrsENd`J5nb6Slip6lGR1ySU)l=bq#8_UB<O;Ob6AFJ90c3|PB4qiwD;
zI_onUm8fvR$;E{2oo(Lw*-ub1B?>CE^329F-u&4+{LWYZ9*0L0CdUs5i&I)^n@(NQ
z^@4HOL3qK`F3_DN8_(=9x_5{RzJ>BXL43i&OJC=*b1Wa-XR(-ZW&bKKzVb!Fw&E*a
z{wmHH?%%nK^fj-&@;UC`f5-!%C~M}+l%vxV@*;0?LuU(2RrB!SBi7e8L9Hg^>gJTp
z)Q!vP@h@u|lm4%|JO+nC(F_UHS;57F0|vtZ>%%o%T|iYLl%-!6C|_X-FlE{FokcNW
zzm1MVtaP}##^n|-1WE@KX@SrYLRzj}8lV(JRJ=A&ym>yR-Cw6%N=j2wm3ia((g9%@
z;p&p(w|+!ta}V1d((bIWr3Fd9M-sQ#+`mkvHS=uA>BTuolu*v+M9wokIb>7TbR);D
zofaKa^6rNZxN_|pX;$M*MN1iW+s#W;Q$5gtAT%YMALF!yP9J4!s@W9P8UhhzUhvlY
z?{aj0Mi3?J?p<pJLiR6FI?HS_rqi3TyT8veFM;NTtTu-EVve!(lagv#)?8eSX|>u-
zW@&Ra_;c6v;}^N+&%ZKq%2hO6AOx|qWb+xv?|eYmZljYfnKOj#2&E0fDnrO6)`d`H
z$ZUbp61~2Lvxas{;<F4ngUxP_b0bg|L>OR2$(3t;Y`LUk3%;?g`03~zoAfD1V=^3z
zuuW;9bOs%@F=fTi{^AEbynUPP%a_@{w1+J-2AvLZtHtMEc^M^DlZn1u;=QBDOWu3;
zUGCj}kM(}Y{ziw6mZZxW5AHwUb1%KhY&>Ipc8n{QY&_G&mKCO~h}t2_dwe!0Klwi8
z_6_vzv-G17Z3|SlL!cG&akG&$wq&h8;Mp55&>yU^ySGcX*CShI2=96J=CgeN`#&UG
zrYIdCwPrG(QJI=NE3nS7Sf(5u9@9$Ncp<2arL66f&6|Hb`14p@_Dgd8RGT_$1!%C+
zQwxu<g46dNaQVIWiN4-PxE3DGI-X5oh--tTK(5M0+36XsD(LNP5hM|I9#YPmo}y$h
zAdDg^YiM=5=s;o0n2Xa>uB)8?^yNMN#fRsl-F2jhIh&qQ*9%NllBE@!t%PeEn;gCW
zF8AJln;=T)Ztt>vX`e8F58izP5S*NxaP`_V^!r0LH`ZAj^x4}Qtd6xE7bhqD>}PMW
zw{?SZYFJK>IlK2JpC7>NeoB4q60r`@5~MMJ$7V~$fBN6k4chp<XAr&w>+#kwpG_E@
zoe~Jm@Bi+%+1p&Fas{KaBktXPmq!mC@K65W54m~sCg1zs4*)RM66lc2m#=gC_J>@I
z&RHx|lx#dUbzOs&PfBf1=YT$)Ht?Bbzy2cT@4rv<wLZeNpD<q&hNw2!tYKaoOLlUG
zD@(e2n@^Y*^X4Ix3<d-%=1I4M3M95nn4F$)UFG~AU*6+ieRxjV-FS-m5>r<V^CWJV
zfAB8%KKRSbZ@>2z2*Jt8DOavur`I2_xv|08aQMm0xBoKp`Gb`D>Lo%QqNP9>gIF=2
z{rUeuD`?|)uOWQZFn5mSe8zai{5!w<ZFaUcsBOjgW6W<o&D?xC=KMODJAqK(g`-GQ
z42p{nA2PpjpYXLl(rWOE0q|$koE15m=5$tAM>w9~OijHx#K$3akrB?9)E??~i?LD!
z*0M}mxPd`?%U32Nq{{ip&25Gc3wFC3oXVKdY=p_qap{~qw`{f&^n8uCFV1-TN8hK_
z@6%b|Vq<rQBo>&9O>&tqxj15XXMjpH&prDB*LR-h^6qA1ne&d5M~D3M^*6b?{~YC{
zVm3PD?Ec$)p%1~`C6{hq=EESuNbq$H-qOl5UR$2>(-&@1jZaw)zYH%#h;WmQ*_gYd
zbB3J(&wcSLoOe2y<pOhY!o}Gm7PB*c|J&c<>g6k_RWD{<6l7`6`f!WoB4a*Z;H)7C
zHF=(sW+^hzESF2tG{YH-T}4UNRlV{SHbUcHnK`TYDdC&;5f97r3B}ndVy^=tLB=s1
zQP3&{p;9#Q&dtf8$y2N_bb<j0kMsibIPcm)7)Dg4CQM@DRtdi8D_d7J-`sEW_UV#Z
zgvcOZUM%spz$!&q7$ON_9Fw0PVn;`eOH0&W<LYpY^<E2Gjw$qj{b#RpX@7^X)uJl}
z$FC<G_G5e)ve}K1(-HR$j@aJXCzc7`)XXnVx!eoLN9VY$4YUp#58~tUm#J91|1*Nt
zfMDwq*b-l*pcRY7f(Q2>AiUt_GdJmWIyi4gmor8er%WdkUU=a}Ha9o9d-nm*oVY5f
z81&b;7+sL2ONz2uX{lDJDXYw7Se4kH>Z$rj_wg$u$2qaev^5AE%E0vCgpJ)tbe_F|
zD?FyG8FW1@?a;Nyni5@=XltAAw{xgr50!+tyu?iB1kSSG?Q<Lma$^aS4jw@%1?6(d
znyvZPrFGspol$zItD4E|glX+CNsD^Eq?}LK>UNo*9rN(+2Xxw9dh6>X{Vu{4+<xaR
zR1mOOE*T8g7!1~-BXCtoCvbEEK_nfnt~h*n#Mb&Q)uJX_oO68teXjKa=65vRoo%G6
zQE?lSgOpIu$5f*+gN;4HU>jRy6!RH{f$_x|O2DAqWB0~BI*71!L0x+4B4c!M!fUU5
zo^G#)F-_lOx6>hsTWoFa@UVEud_F@eO%z2~>nO{z$&_6&R*wS?yyvl-^l>ZBFS9;n
zPn|?b5YnxjP>O0%@!-#X#^#kP$o)$w;n;N*Yr{6qMucACbwU((F?mi|R;2TSR<DbU
zVuIRGOvWgs>GXR|c)Ab-tqv-%BzeVhazQ5@^MCmK4uAG=!oR%($?h}Uzc^*O7*Sit
ztjxK%IAUknVc6<nv|&75^5E`6ewOqY4hO8Q4e55f%*JC{y@XCXA<HwAjG50DEEh|%
ztYB;BI-#-54&LFx(QSH4Q=L0}Ut+7G>8bG+VGP#P<ntNn#ToVY|G#Yi;r|JvLcH?`
zXDG8J(~EO{=WE~M*(=xab<W|VL*9Aghm6k-I6Z!dH(x^w*~Cl=L7FaErYVAuG|l<o
z{r9<j_a3jk`USrAJKqEWWl?-&+;$(uNBw-P&}VUS^9z`N?`LdY{)qXOs~D`c@g^cz
zF%MVFt6#ueD^0uKYbIe=%ynRi@{)9NK|3AuAANp@YY!*<(a$j-ub3BHTpY8rY=5@Q
zk8jhHpTyiCjKrFnVm>3kIHUgIKWF0~{U;a|G)?T@QRE*nzrJFA{512EhPiso96_3<
zpTPV+uf6)(-#&A%kP?Kn2;-?18CJmIkKSf&cNf32hY*fCqUODBKx|{Y)Oe-HV}-11
zbm`H{3X{b6AfT!(YC1(q#e3}zTFORPzA@metT-u4UNloOH{<o^HrZ@>)cu?=y3C`=
z1+#QUWh~Mex~5=f*(OT56r&mEcMo_wY0>Fzu(7_z+S-s#r^EDOM7x*J?i#W*Mah`y
zWX@tSCr=A@c5e{Y6|?(4<<9Xv23k{ISmaPJs0x%2cw2*QWVF{83!XbYB8WTu=__AF
zY+S;}0<QzSGx)mZ)+?{@Mz4>ak9qFsn4kUhFE~25%cFxsUitjXU~RK$Kv3j)<Dm><
zvNYq-!-qULc*OJ1y}<Qr*D3OXd|pwMW#ia<YSP(0b+iAgynMgx5^)Er48h}RT8S-c
zruQBZy>x}R)kZ2$Z;_)5McC;eYKM~=?JP=bDsQQsAc!Jdl_Q<QRdq8Z9%<^bYLXhF
z7;NJ*t8!0ZOTK)$$KBDAw@>G&q|2lz$nvH{X}v|7oPHUR1dU9%8eK42E?IPDs30WR
zs2N_lhI2Uw5ALzOImCyajqP<PO3L{H1ngbAhB7s7dBJEpYEl$RMU^!J9o`A7@!&Mx
z0@hND&Y0i+DcftCgm|*W6y0yJoXxm6JLl@9YitbHz}8HsQ%(;L$QDx;vkBO%jr(pj
zS5sCMWmzMHrYdXB&(63QjoIJ7#EomuAZU6hpX}P6cD6lsGrC_sGU~5f4y2X@zL`qL
zYHV$|`0#-4)(*PAMqO6~lLg@*!0R5`ShRDfFhYcFFg3n(=%j~jd7Smsvjve1*xeei
z$Vw(zjcoVvI-)inyO<Ld3%+t?$i~Tnw@)qY?f{vk9L#1c3=}~?Ih!#ELRv|iEMAbE
zpHb=nA!DK>CTg_^qL`%9q0>odg^Fdiph$D7vScZmeA@nCi@<s2M|U|HpAbn&bv`4H
zELBI7L@iudLnyJ%P)sMpQX(F_f!N%pC@Yqe86}X-C#2H}U;4sVNm?z=&QCczJHvZH
zk<J*MpRhjYQx_#wkr8N3S(M~?ju4tWFE|{Xkry>rFJIx>^H<69oU*P-lBChI7~6PX
zS5vR<Nt?|thxJo!>NJi>3Oo{A)96<_Px@%W-9Pz0&;FCY$GXkgoKA>DO%;Y{-#Ehr
z0WwlVp+^#8a)S&*D&2&_XX6RsdLOTv6ox2@a1#1!1G4Ek#tNdM;9FPwJh##1|Mla$
zRPC6J;d7i%PZ=+#+{tSW@(F8;Dcey<Cv35ibeanwR!mPy&X12M>l*6>R;niRPX%Zl
z5h3VUM>szv8=dgz{DfR-_EbRPXLOP-)|6N&sq>sbE9$DkJ4;#Tw9oIME*@aFZU7GF
z4M`Gn>*fvChkb6}`GB-Yna*Z_;o7w;T)TD!AuR8|_wH(vM{)1L0YCV`582zhN)*Nn
zhiinvA;-r@JUDnjx7%ZxWsFB7%Cc-Gvw)T4=83neGQawS)Gxz)Xmi$9%<Er*d94*;
z7?G}+^AvL<p}RI9Uoj7hf^S~!b91B1fBoaTRPBU~;jO2bR~+OuYl|7%QSeJLcL=2k
zbVT@Z<`<{rQnMF*#N3+__Y`yU6!YZ#9%_7m-M9{ToVP#6{N9}p$%^GCGk<@@TuSae
zeE2DtSB<yn*Th^1M<@jX)i?#MWigwxIJn^8FMi6^@3hG3j2|s#*s8(>O%^7`gOymX
zFiIf^5w(LbLgxjw^917wuG>b-#(m|30O<o>9}XDI#&n!xT~?f3Z4(X?uD<mM(`nQ1
z4H?ZZm}Lt-$VwjNB^&xN^Mtmo@g6d7m`uiK=ctUudVy2Tb&hI!x?&;e)`n>Eh{eS*
z_r~YsD&Q&sZJ*O`b?MgyFDS)0%Qy-&x~j;m#nlD9<q@IG$@&Tik%7h~F;_0_6MD<l
zcizNTB}a>SlN7MI$=cdFMUnBr2k%o9B?!U6(GfrT@lV*^-lr%l9FoN{<Gl|)V0~kQ
zR;xvp=S-)wX3$59j~?bvF|R)Z=4+QOVIJOYCYGQ2Ur<=&G(*(TGCA^Mgh?|*tJNe9
zL<y!U(Lq3{H6jSH-Xp`XImL-WOi|)+2qiG4LMkxU5hn?CnIlAit!i#<_1Nr0{Q2!i
z6h5Z4zD-u9EYc<C`I2dA=>!##@Mt7GF+i0HFFTaSk9cD=rij|i^NQo!cadj@^kYRg
zN{CcII|{Ii877@G&zDqIlGum<i`I?I+gVR-YLX<z)iqj4>bgWOE|B>Y-5yY93skqp
z+L~U!M=OcBxHzXW)#}vLGF%(7wl)CaIXgRTHflmLo=kc0@FD&F5Un*ytBsn@n9pZS
zrW2wlqN*#hG{ac)RQR3$x!)dpSAUfyT}9#r+E}EOPpppRMatQmAF%PIm+9J)IGs_|
znqb&NIY(dx-a2FuA#{ZI5-NubBZAl<jKR#8sD6iTOA%wZC=09-=#33#d5ZIrs;IcU
z*`gOqUcW!0mJw^)+ni@h9xYPNO2I{G2<LM;!V~KN8OG>V8{h3=7fYIbccb>nt;OMp
zbU;f8*4rT+1(B&J@{~!QQUhz=qVgF*?eWDDp#nr*fY(r0#DT_{3Uhj&<?&t0WF6-#
zA_UuOeHQz>XeBs4e8kDw36t4`ByJH$5w~tW16Ynv4x8#r7;yjKA%F41AK|?s%W{@!
z#$-Cf7(=_0AcbHun^ELBX<i^!e66uhhT(uG8MS}i`gEH*>4>Ca>6+tugun}jv~c>?
zZPxcUxxUrL)e9;O7=!V?ae4q^#<l<e;`&KMK~yQ#l##Lmzew@1roGvt(i*8P%GH=W
zhbTZeLy^@)?H<N?dh6R{<0+w(IB&Vsf&cK;UDUe|`J;mg-S#@${UMWd&SWv?yskK{
z3nEd_n$K`r(m9FP+uY&uv(GY~<OEU^MFD{ds47oBJIBvY$<vevY0ks4q*MW~_G3DW
zjGe76tvJBh8m&9jMS)A=rq{(70_W)JkoxopvGpw83xrp6I|GLOK5xD8Bcu*NY0fV$
zc=MgNnJ?x9VZaN|zeK;khPMVKJ#nPLRAh@WQ6PEl#!aqXx`J~aQyV6e3FFzEAO7%%
z+`fJHv6uEqGv%jgs{AaNU)gHoe;MY9roGWH7f&%4Q2@?RW;J2E*9<wVm<KE7dmZ@S
zyt<2i_aT3DF#d>nwrH4FC8u>sBuZNI`HFe?3z+i@m{(Lv^GfgMnOj?<R?KaZP}eoy
zSOVv0>yY~V7_o7EB|QD~%(Z|x41Rg$ckVs0e~Hye-(&8dg^5-QM<67H6Byt8{iZ50
zlR4+_-J!p;&YeL_S*C=(#+NIl3m%keGD6b=nOBHTgzYA{JV&m2z(ieP0!7;zytOQn
z7T&?T!*#Aq7t}&9hy`V&c;=;bdhZ|c=cfzSx_!2K1E$%6S-RvwUGuOmNJK$<HpA#6
zI#JBd)-G3X-DEPGqpTo`B0}k@Do;8)!%dG_rYRq01rN&#qcyK}L%Pe1D?4pku_m>K
z50pYw6|$~rn;KbHRMzn$9n#NEna;PMzYX5wk~VQ1^43p(1ZYs2`C`eP`wy5drg$9X
zwUTakKvCo<0fAO5=CkHH(~7<Aou+@yc}!I^tLMxXOWu6*O&%UR`pefjd^8UEE3WST
z9;TQxIXdRaH%fSh-8Sjb8EG$J(D5wJPwBQ}%ITaWiBW+@C`G<pVrxSXMaWi6)Jd>P
zBAh~MMUgjQ=j~3rNdc10d1c)0Qe`<&5MZofqZjeDn|u85{vjWZmqfjFcGtI<XDt@9
z8L4xeR|Qx@*VO3cj3f-{x7zeNJ&MH?sS;$GvaLPCBtqdZ%MsPgGRtztWl3f%oeH{1
zi%1KiFu*$tLSi20X|JZ6U0orKqZpl|S_8Cbes&dv#8Je_@c~E;LXxISj!#Y~$^xkr
z+uJ)NajVIc6%c3%&QcW_T1mFnH(2Wr8ga6<<ath(7d$+8$mnA9)Ybgaa75$7{FRa8
z(v(<*%>=d+2pk>*UNL!aOfu{->}s6JC`?na323x62!X3=s7+Hf2qb#g!Q~RELrhU1
zZ4<2+O3%8s2(6i<HF0l%PjiAOrpC~1Yd(LahdrM1=43_~cX@Wu<vh=s)SdtlLGHfX
zNJ>B%M^G;@>4Na$g!NX7jqNR#iv`{*TCENnodlaz=qjgGXLwU{VLazm4N9;dDp=-3
zu|h{0Q5Pry%36eT;5~s-%*HdKc8}`xn0mKOx>zER7#R?1#oasaF<)e)dBJ#e!Duw1
zHkKd^+1}m;BpSgq%P{R0WtOs7rYx3AA{}sb{}NJ4&d<*A9_DGv?GJBrdNIP&`0BDO
zqpGS;&Gl23Uz=~0bsk4E`QvDUwS&yzKKOuP?<PUusfrS11WrQKP7q3=wL(}&o@Py7
ziIfC!Ouk&AmBiQ0v9n*vrEFazoF__R>L{YJmgTr)d_Kb>_y@ZKI(WYO;Dj?P34)ks
zws(+Wz%*Mjp3j+Q1;Q8t>zIwtS-iQVt_&Lp`chJbiicUo(V`%?meP8x67-ei^Py&#
zrF?!z@%bH#$tLK*WrPrfD!|p2+8D|-Be0TC3;g&LU(bOU?;P{Vm>+%rf9BqYZ=-`2
zLDEMBEqtvh(wvi{W7=Vx-91B*rK^iyP}L<T=Z92fO;y!YwIMG`(sapmHs@kCWj=3$
z-+ysT&}UJ8{i~Uq4?dvZyNO;g*DL1X$Cwvs)|9h8#+-VUh~S#u8&}uhJmHGDSuvlU
z&#*}T{_cPd@ZATeoLSK@-`ZX=PaEcWjwlR)wSP79Wl3%fmGzhv^UsBvVYcM+JNgN8
zHMoovjg!f*8r91*MO#Vmn7K(o#(3{od@ANylUXbHtC`Q|jHa`O`Dfsw`|FuE8O{!Y
z^#Tjjd4V`va&Y@LKl}VOtgysYjk6Lj!9@`S3UmNSWKrV0z$=A|Bf85a3xJ}e?=-?S
z9xf3yCiSy!mu_0JX$*r&&HiMG_JV8sT@pX#`-kUDQM7}Q^{s8ZR?M=L>2k?MQ9@}E
zHe+^i%G=q3qBd+e&rk~LQ1Kwkd9W<WjG=N(LPk$XUecQFBIPUlikG)6CY_?zFQKI1
zTnC8S@`7_*UuM)+ve1G-dWsUCLxhHo6zt7sym9v)Z@m5kmboK{JH$y3XB1^#GMSDD
zb<D=5rN|qFg%nWNB^To}>Z(Z%sEws4o1pK-a>;bDV40>(uhp+}#chVyt}~s?a8=Q~
zpa^J19#N#2^G7Hb6JU^d+%hN63v?1S<v?c<wV^C)LMgCGh!jm0r%<Yyd@m}ZFl_o7
zSN<m*hS=JGloZR1$*4xCfLC`mNW_RclQYsNAqo@Po0~Xq8PBH7vn8eT*wx0u)g|Tl
zf~AYFojz(gr_)c!vx-^aa5!q`C~PCr?G~0^5OIA+Fwh0kN#sfqql5(OaI#6mkhUqq
z+lzDL`VK*R4QD-BmT~alJ}1YANEIQ27^wohlhjpBQOpT-i0iqgqO&TKTWeS>XRAb*
z#yeTpHC0uS7X@kF49$EtN~&MHe37Huio&lvh|Xc9Ln=@;*y9s|cnhIC6cuhVN4Ek{
zjST_6ddasI5vpbqHVm=LB{I<X+F<ciX$BtpErrxDRf46rC{u%Rl=A{zSzcN1pxuHu
zjxShv=*9`1Bw=YaY1y<PRL)a5M{NytRdT$TW2|Fc2-<;W87NX$VSOWOmSKbuk^tCt
z8gELrI-YA?M_A`X)(~q)5=wkoQk$Cj`331@ifngqj~+m44QlYhG8>P0`<<UQj>Mv-
zs61Iw;GJilPD!(h#cV;ZH)J}UaB(rlTg!UC&uqCQ&2m5x$MKV7iDj0tNK>*rZw8^B
zh{gYA_xVq?seA0h^0<$ta}YQoKuNCe4N<DXESI>f!lUt6Y+aLdJBTnqMj=Xp?nHPc
zu-;MCH8KcM2)u2aXLVf>i>BNlgrL*wVw|O}O47WdJsc26hHqYKu@@Nr>`}(uEN6C+
zVw9xSZgca>B}AyH^OC??TxFT%C8lx|&d^hyyb>H{6-)2YLXd=twyXL2Fyghf7CUW2
zCn&IWNq4wG6gC-yfzk-)kW!E>mqa>52#1(G!p!euTbIBJl1TFG_3M22&QBRn&In_}
z?%o=kYa4jq{CnPi|2^J+??a|}P2~g!$7lR^|NLKKo`iWrlY#e(oK*g!ecfj>0QpNZ
zzqU6-stUVY;<Jhx4Om=V6F<dVDo~vWl)(9>@m~fZ8i89ew{_LrSXa#3E9Pd!yfqxq
zjtt+tl&}}npTNAs%`5v2^P(a!reR)`nA%a;Phc(uu~M}C$C;Pd+Rzzp5QY(Y#hexM
za=9dUin(1pz_l-d6U3jAdF|og_>6z{Fa8zvlis*hIt8CiQ}m}<N=lSm<={P5Y6yYz
zkf$k=Za_swG#x`xA%#ZZaAASzv>?!+6+RB|p#~`sQV^M@+`4d{)HHv0R2H~)r)db6
zk~@PQ1Lx>gHQjkZ*6y;Nz#m*~aWyb}?=a_nQIi!ZF3@z_ZFa6*!D~s8<%Gr}D$Bg6
znm!e087NOy3LX_T3-8fVk|;sq4Zpt@@;jReJBcAtCAPMNYwJwHn7dJgl?t3=Aq0DA
z%7N5a?^w)E*t>U^Xc&>xMudXh%a=(1eZj?efsPGZTLXe9A&NUFHRsOVyWF|+fJJFA
z9!}23{F^`d-pWb%l#}dJOuBtmODcix_32%?#^SvX@eYDOk+ef}D`2f3QkP3&C7~>v
zsl3W!$`T!g2(260ur+w$@vDbYT~|apL`q3sSi;y6X@S(@iH9prl4j5csEj3wBBD_9
z{Qi(NJ>~W^XI4RBBsEQIz~*2;(r)2u3w2FZ7`(7Z9gujiLC8t%sVj>T4n>nf&@K#X
z(sF4d=E~jx=Tqw16US`?tCSciSGQC_nK$uXQUduHvpfSGVylcW&}^)4aCmq?mM>6&
zrPte_-D<A}m<;D<XPlp%lUEMw1+)2*H{X62OVi5IRMA&Ga(~OvTdVuBD=p3l{A$?A
z_$DQ4bFGaCn*3g296AbcxxqPuij!taSO|QiA=HoG1p$wI{NL&(2}J>c6!bz#&{Bjp
zru3d7&&ZY;Nw-U<1z(C`GlCDthSS<J%VtRH=*obNVGpAuv$94yhwy^jIz(O37oK$u
zR@97KP3=4_r3j=T^qyPofE%5Ver!=@Nz21duf<vnK~UpG*r;r@Y)<Z^#?%FFc7j-(
z5VW@Oc#^;|Xhqz9|AIwYfr<#DE@fSkWfjxugwc48lmUyfpr}d~Y09Ic6Y5oO9KJbi
zcfY1<{Znn~T3R$7A7`2`KTU6gbei@6wt5mXp5jf7cOYf+reX4&8jw+ph&4f?3F1c2
zpp^paDC?3qj+%k`ARsH3SX&{zLgA^sB?vV;SN2%n-)=HhW%IVAWypFwXF|fiy>rIx
zS&g4AIiJjND!^-vjv|6~o7Va!fr=5ra^?z_MNKFsbh8Xw83N(?!&@8t!`%S4%&E!@
zhosZ)vECb!L`mZoa!q_l5`>ghO4M!>R05+5@(*5zm(~!rL${T1_4+MN7Miosf+!l3
z`g_E2i`t~r!cjU;T~!zl;IIysMfvGS{+ezGe;xCUp2U7i=4!>fu1iE5BN50TAb*Ov
zTrt<7X7kD(gZ*vHig}Uc$QASPius-SW9BoQ3P1;_C?aULXsvA#sOTx?X2m>hN=DN2
zhd2K+^LD$(pf@C5F&C>FNxWhnJjJ}Y{W^ZdyxmH;eEk+D%Yd`dk|-RL`uoIjvSRK(
z#@yrl-+=kAX|=RKBEdH^krIdU66qDho(E=$KCaQWLI_8r+Xm?^QVG0`@LFO+g^x5s
z3ecL+c?MZY8b#FBBZ7c-StD(Qce;7;moU+qi=7Q#Z+F=ckcNWKl_}ShVk-`K-NwB6
zaKxiUNmbOO<0<G68HSW$NYHB0?XR<@W2A7L*^FgTqo-3k=@M5PH1NN?wa)))Pvhqq
zMUmqIiRyOwa4=*Xwa5aEHx7Z|Oo!yQWU4jO8e&`V#p-RUVuy>c#ls|~x4FlospouZ
z2*Vk<dqfn)7@JcI-%S5j_0KWS{|4oiU&jiSz<Wt&cZb$=!s7lhNiQN!B7(@1P~o#2
z5hpn3QI93K&LNyfIssB40}HOgR=RPZ8B1*{ypU8zV!T3|8o3Jd6t3}cDXr*luF)Ra
zCyuVLu8}*Mb~I&S6nB<27e;~;%tjNcWscNA6T;qZ(QWmZj|;+1mpT|QHXh~^w(BKQ
zDT3N_W3$J0ZRsX0gs&;enog%j62_}cV+r2iafCX=*a|5<Hmj(E6iy!@lOA?<PCE`*
z+t_ANN|vcXtCY%1v<|VZY>JtF70eDiW?p^<J&>Qp>O`{9VHv|JP!Ht=!a)=|x{=4v
z7YGk3kZ55M#^TEoYZ{FWp+dyEIjMh~ukL*#pN=BJFhc48GoPcCM0$w`Eix?eQAFI1
zXbpSKl&1h`6jwGR!=xZp2|qnsaJ)2)_9ahQRF2G85DFQFgi*rgV8~hwxvMEELsS%m
zwh^Q&t@+|s!fRcR%PVSA699X|HXH2@LMdF4BijjDNOYjlT2gzD&pkpEl=t5uxE2w}
zPLqk-A27-t)6Ah|L1bYxo{?rLrKzb+jrSGCHAf-NH|-QZckcRYd{h5ao4QcdI4@T>
zrDk~nPCB;Ndr;+6E8D!3vN`?)q?1&(LZk*^6rpZoDv1h1f*`;Pfi*P{<D6?eJ<emQ
z3aw(i5IAFS#uDn7u+=5h5{x6(E!uI5R;+o}!;9Bm;kzH4aXzz*CR3JG0d>wKPnqT^
zYI}|2(|c$Uu&H9AbjI^qHE9-Mz;|BQ;)e7r4@VFg`kgNAcAHkKMX%dO1dUMBSxX=U
zLhIEj19&NjrRMbTh~&l$uVaj{eD6m;=FxaUWgTf&Fg|#=YKU5WhJV|y<)32xRm`{6
zdx)o)i;pp{Z3TY?^V-xzA7gH-3RK*@<r#xD7PVp?t(Zr;MG_~pV$F3A&s}?!KmXwL
z=b4*q#e9j{HXNUhpJG1ec`ct}zReBkR?IEE74x{&qWdxCT4<DBy@jrrM^bZsctrdZ
z^FRO5k2xHznCE4~oL}34>SwyNG^?^JtKL127hr{>JLqyHPdV2XRHJAR4isRlKnG1p
z-GvIP<0iYu2bdtBB|P1_CXC}IX&{nh9z<E;gGR!d7(*-sqbR{`wK<i|!~X|WoBK(d
zZ$#(pBdDL<;nw?SRCC8<I%ipx_^e<y_e{e8wK?QuHbZ*NRuG}H1<y;_+^_@9@4mRr
zSCnP(V2o%v!hV;s)1r)H?)CZ@t?^PKe2o{HxmGL!MHh&iBS*4+alxJC1!}N~vX=LM
z@;b+pDYbLtMaAso`0o<qr&?Mxn<){-^e$cE{9=wuLPiC&S^-s25DGvx@oP~SqEyoK
z^sRD$oe%^%#+nM7WvDQ0R1DS;L=j5K)ugV+TL)S+(P-8-EeK(Rj>F~@!&m}Q6DUm-
zC^qVf?Tvtk3(Kr<j1*YsPzW;XDC!!MFE}|qV#BP{Yxfw02}NB~X9brAZC<+EAzaLv
zPc3vM##!2(F6~yEFbtpc`91C_l(P9wiq#;RFpgp}p}n7xO(!f>%$)}hm}MEpc}izU
zCzIv_{VllAwWzQzz&MZBfOk($4tw1QQRY}{QK3eN0O8F_uj25wLX^#5lnR;?Q=uC9
zxyR#bgA4+sRHz`tNr`kIltM;P(_`yuTD>9j7L?XgmyK|{7sm90IbI4b?_TB2`{zuT
zHTg25(4L}zNm(;1Yu5XkYCd5;a<l?XjAdOZgb*ZA#OHTAT#}x8yhMhcy<v|;8X~O`
zQH)*<_Gsao<RhtA<T<&iX$gTuusAs(9G(*<P4(i9x8LVtHp7XC%6U#tPB=Ng_>625
zzxw)An>wQ;+BV=HH)Ts9(B2Zs8XXCOK@X_}#u$PygxX*i8A5u9G!{pQrOq-u5Nbuz
z?J!xM5r!dYnxaLk>Bq=Rf_@8yLQ9P|hI+9ePy$mojgtsZAQjzKpJi1LyOck8VaS8y
zG2FTgI;15+?%#dDc;^|u@xT9%`8VJDca%78w7ZP|<^PAp`4rI`@bcv$;rtBaU~6ZC
zFcx$=J(i1Q^Y&#efe2AjH3HU9AcQ1|W9qCV%QL!h2T^6XbcE=ylcqUuz4b1}dO)s@
z*&k0beKw8lzs~hnF^~Te^Jdz|*9N!9KzihgIiJ9Mwmc&UL$Z%C*Zmf9#oVo!>!+9t
z;n7mjZuQBlf|gtI4__E^cs$16x{S~v$y3a~_P_g&`J+E;m_M^(zC5486Xxd_=h@uZ
zAczI+PLFKaFmE<;AyO*rQ_SNy#%9f^dglrAF`~atmgc<m)_WND)Xx3Mtm<z*b1TIv
zwSp$wbj4iZSeKRuZH--PH9oSc!UqA$I67%g>I^PYNMUhUL{;Ge4N7v->F~MPDQ{@a
zFv}SUffEv$7r3Z}*P2W!bZyvMrkrSv_J-OCBA^aq9`)CFp-8!^N`Cg*HZMN9V4^QE
zkeXD4Ts(NhL4S|-KmFelKKPJXzF=p41O3PUitPRo%TAA5*Vb8^pHnzT>(Ux^OK_*%
z<=Qgm5dmH2x$qvZ6w(&d+EWY3B2KtIn=?o=4mvf57|Pi>8=b2}ndM)<{r)F1=eL3R
ziZenHwA%D8?V`gFj3Li*{C1CMV-sgfs0&oDjd0+K3NM?7wR4UzXijFWHMpuo3&<ZU
zS!{{55mGlVNT{Z@1KDIm+p<K(5eDBhtqYHmk}!xVg(NYC=QkoId5PKB#ac(ALN3M=
z^8N<xt2bENx`XKs*y;3X3rmshbM)u|S4B=Mo3mV`?B2ME5INr2=K2m|VC8HRvI%yl
ziN{h};Y>wc8k97kD@0Wh4f~8v$2>Yb#y-hl_Ixxt+zgtmf<it4|I>ZYe>dL^rv+X}
za2nleVd@%TUE`XR1|oq75=5k%l6o(0?gI;`ts!WKSZhFdd{s45=23ubC75)KkP2H>
z$S`PPvdR*lv=CAf1R<uZ5ZRJpw?kf35H)omr6Mu~sfWSNCcDEHXQLTm*uu=G1V-@a
z<N~{MnOnc}d%W}B+tl8&7RRVZkGMD;VUw6!I|IVW8LlW8taa%1B3enzY&=5frra8Y
zVY3OXN<>RwyrZfOrl_!Gg^XGVYq0qo8?I4RhDQg7*s4N=VKXgUR=+VDI{1{Ex?ppc
z>gX7Rd6E?<twV{L&HYXKd(RMe16*EmaPJP6ch;z~0^u6pN@v)i)=ha;n&u782}x@(
zq%I5Uy1-Ng&UHX5Tva2yK<WsSSVU=%rebk&K+@~s;||6-TxD>j0b3$v6J4|3Ns!(V
zv;*XzM=*Gv<>o%mzW5xY%JL_F`tPvYZQ56MxO#P+f%W*Zpm%wfRubS$PM{RV8kTv@
z+IE{oma^XMtRiS~q9jHNNxR);wm79+rVQ3RJtaANbdPrXIu~an4o}Wj8@b0#2fsOA
zfWL12Rm|7-H|gy?L)Z=QdCAeeJM8TYsj>p=EZ#cWkC}%7vi^v9@-gQA7ciHl0aLL&
zIUrs!ubsoMnEMrTXJOb$5G&?t#eA``&-Le?<+ZZryMOlYaocS=Pce68LHF{`&oQ^w
zu*_=uTW!)TW$-cPLP%QeF3ZI!<#I`HZG(=ITs*o<yM2xE*@(lF(^dcD$A?&c^VzH6
z2KIc65KYM)AsXfw4!4Gk_pf350YP4J>(N8rzA~Vlm*5PwchubkttBoB00G7Vn%Q8$
z_m@jlRj^){ES<#%8c`IDnMDT>$IPh+P0fpEhurISsG~OC+NMXcwixf2SjThLu`@_m
z7)RNTu)`rQ?e*w|YyA7O1z)-O0`>dvP_^51uJ5zI86puZ^Ma$x+iZ73mUTv|B<sdP
zVY$23<?m%F|9Wj5;Vf~LQ-uLuz_istjVBD2DGzqwPzs(oeZ;km7pcb?CuiqRUFT1)
z-+bn}-NjC)ARGB@XKjdbj&`wNYir2H-abh$qMS{bp6AVgL~W3+>B9<JF~+-QBV--Y
zDNNbKL~1RW&lljD99SP}a2}}?LdnLLRx7+W<ntLiicvbm34z5oq4vfQNQp6)cA%Th
zSEwNjSx>fDgfWAyHC{{--XAX^Y14{5H=p0;)#tDA)*t;#7RIyn{PUE<ZItsUJT}b<
z2W_fsNjr(a6wUN25+NYeA$4g;T}c>qNVFuIT~H-U&W?_m&E_clDxy!K!$vBrbu%?@
zZIk6&TT4~fPr1ha?!Ob!`W`H1C=5gqagtEYXDBS~^&au|J|<R3W2q)*1j9bcTBK6g
z*&NY{krKSt$WkG^M+$}P_3_5yt%172)f%l;lN%iSW+NGdD}9c`j?M^@1TVto`R!^V
zE$~_pM*$t%A`3D??NPm$pp`J#>as}s>|MWtOd{TT`wiNgYiO6TzB529m)N?bx4%gu
z3{-ie+z|qyV{|LQ){R<7I!Bq#(Y-Z{Z`uf4Rb%s<pg$nin%U?Kne4HcE||?0xT-`3
z3Es<JS~c>wwLaCR?&f#?KJWd%{~Ln3YU<yjX)Nup4f*^_uhI$va4E-+PMBXzv6nUx
zS`lv!$)<CJ^dzkabl6DsOml{V4C$;5NzX=&e9tyVxvpv~$|48}+HJC8N~_(arC@x1
zPT0w54cCdfEo7icM=7@U1fjzFlC&(?XvIWj-h@pb-sVsKv;Ql__BA$!U6zjyk*DXh
zhMKi*2ifbfwz<jl;vB6af<RGKp1O=dDr#FbrGZrg@Z+I{R?=c*YOHezEs3Qfy*Q?I
zt71H9k^!3a@eBFy{=)w2){SrfeLnc_{xb94+K?AtdX?m3%(p+r9O;Qy%ze<@)SU-p
zNPBHa_9>W8&(G1FoMgC8uwtICnCno1FUiY-c*R@<0ru!NfA)X>zftV`JoENYGgvVn
zY;H1NG1q~nsvOnFnEMrT=bNVABxx}<jru@nNhB57#WBfa=4rP21V8B^|C_yLU;Q#$
z|Mrhjrb6PIb_yY=`aOPn^94ek5V@RdXBP}-3;JfAC{mn6eS-M{FCA4JA;KmMTR0C=
zKpb<hHe_%yAybN=Hk4W+G7HAy;~3j%6BHGv?JhER{K45NA9j1ZGZ=CjCQPJcceJEs
zEweD>hAX*JW!!DWJg9OmbU=T3%<p9d1lQRZcHrz3|L6$9l7nG~pxfh}&2`qMQ!Yfn
zTx&YD=WJfnyAtB_2@53&o#myv;``14f}<qC7>BAIwX;m5XLB;b+lqWXAuWp4C-~dP
zeB;?$oW1cns;&_Nq98;X!}{(XFTH%7Bmis2%*PW<S+Bw}1-hM37X?B#8zAq&34^x^
zowTU(6dmg3-Y7sx(PRR<)zrRF2<<4cnm~j&Cm5Zl1Yt$oO9*v<qOk!uFPe;HYcSMA
zQH(bgArgCXz{BYnxqpMLOIMjaIAS?p60Wb&>G#;Zc9rVvkVR23InP*33ak$)O2-$z
zwn=J2gb9h2Ky(JkjV{tG2~<E?daMU46;T{8OXt+{GqULgopy)a{mbm`?J*b*X}4R1
zanxj<+mGVh%Ch9-?3BYtk2pO&BTe(BH2rtTeD9Snar%Sr;?oq9IJ$#1(z2iv2W(w?
z2B~6{s3<2BB1@ARqXnYZ#-<CrcgQeAMhR9mQ^+WR2qS`FpZa`^2m*vu7y-Vn(9R*F
z2+`?KXE_8BN^A1@1g|tve@N6y5K@p&$4nm`pgIG>P*G+}TBat}0-*!^{FK8#{g;&e
z4c5AC>}W)O?>@R6GhFN9Ttw1o<EQ5YLK21{)&+QPkwRkY5~VeDUgN78V=Bx_NAy_m
zjS>VxQ_rV3SF%j!6m<pGHo|fAlU~B#l=Z1Lb=zP2Jtp_>arkF{imE{&ka%uByGLLO
zsyfFr<<adA+1y-bdUC;VYn`wa;o_Ll@iESOy6YQ+y&giUkNV1VKxsUSiz)m4AxbGM
zO>BmgO{j1XM-+8Q(jT%BMoi9+S=v)NYa8e|VYs$QHk~1arz$hbsvyl4h}xpVfL&iR
zfA9vQvokK@A?D79yy6OWhr4VHH^{33Yf7RhqONkz&c@ukGvU??x0uZ*1UhIA9*n{J
z$3eS*5Tt2NIGfX(%$nxUFaYn#>L$nYH-G!~$!p{5-(>pW9!DQz{_L}R=oNFAa(w$X
z>znJ$PcG<htrN5&;y7lqV&3@$%&Qgi-Cw{we2V#67%@FRCVh%|Z*7x&Is@UUs*I{C
zD6$2zZc4%1zGm_84aVn>ncwCWSFk<YWiZ^JTrm%$2vg-;oQ=4DXUelL++;DIp#K`?
z!EDZQG9zk7w8H@L{|C$;er~|t@k@vw{}^eSz7mNiy}CtZ3*y`o`JCrZkGUB1*`Ch1
zyS<JJ1591BKe}LZDtT+ThV68l?44C9SG8L7{e<h|8Q*KQQL;G^5!T|B#4Cwywa{hB
z@o<CxF$nqHi&G*~@uRg(Y%Ad>Z1K6#2qiq!IYE`Nmjy)5vj_tYFnsguh@<fd7o9EA
z2M>8Josn&=@!r-ZpSLwFTXPVF3=6~N*%<rbly`4D!{_n|kEFs0LE9L5uIVR4fER*1
zFW6<ta6BdNwXsQragOq-m%QJ`)w%i%#bnI<-W`+!<(e48K}(|RjJh_sdd6aO!C=s%
zSmq?11QlpptyxT`c<Tt`gdk{gkcDi*-i>Jl+snnA;pSS?${;Pq8D#CK>V#?5Vp@P~
z4_LUIdOSl-41-R<Mn52oEv_oip`ch~7#vbHP5sssv`vOv<mBgV^xh4kK(ZaT=q8$W
zr_0{eE8IIfpsfYtvQZx`bH(o7Hl?Wu!idw8IoTq|6a`xg&x_lV+QA|<=&GVu)PSP$
z8p4<x&ppdGa)Xx|6^4XiL=;5?s}muDkm#{o4#(E^4xfAJ^Q6m^gGUFv_uhM)o}5zu
z+7(5=fcedv<fjMBe)t0<fYh{leRg+xMBY*@X4G!M{LvxTuJ2P`ObCZvtOgN;RHG4f
zS<&6!L&g!-JL<AT=>Q=l)(gtnl5VF<5GRy*L6K#|L4XWHf_4j6RftXpEd}NHoN78k
z^oImdOuyS>KAB*A(=(Yb7sN`Us{k!PQc<7YW-%Ev3)<9&kJy@x84k7y!+@g7@kQG7
zf7+USnR9kNWqoIZA|0W%B3tJ8x>^nXG^!ovJw;ig(~M-9p<+!SCDvH7vcMRFmICWM
zUhy0E{`RRhb$(-mTmSeExwvzO^5HQ^MW_VpZG)-LkYT{$_yFw<omPS`O6G?r_)t?V
zQ@p7V-m@%flFAaaoA6?(6+&r}PMgz*CzM6GQfVlhcL>?^F@>#!e3nz!C1KKGbLSG%
zixY~o5#jn8aoj==BzZPRI^UE5gvWbBnpRY{KrCm(MM!6_OA!iQc=kDVHu`up>)kF^
zG-LXs(=*OarmPKi7;eJp#e&Y-UK7((8|<ovE=2Rz>F^Y3N`5gx*KImGLyVCttJ81s
z8m{r|KmG$ockWO<#eD5C^C)0>bb$6XtyT<0$>J9<Z&eokG3Ifn&Dp~fs$al7XeHQL
z-o(`;9oBa)F@K7A^fYtZFc-q(ydg^~YFi+Gjyc?V_Bl3J%v&qwi?U>LddB(5l;L29
z!6sZ>ELeMrx%&j>Wtvi4Oi*>3_KLavOU!?}z5%=YM@)YA3$z~`P#m8%eO5})N(_M;
z6Sg<FJejesB=740ljg{ib1o~vUYb+ZB^QArvYyI;O<K*yNH-3<vtE~5kIvYrYQ~Xn
zHu}~!P4`mb+6l7EnpTV6kbe^d+&VvI<MfQt#s;Txi^=*1ucb3;=a9xR#Btx0yjqrg
z)s<w+DJ=)Rcz_Gw!xx@mvcHCv0Y8ueAuU_u2``_WbN*<|gU%-LCC#<TnD=^H2<Hi^
znwD>j?p_FD+nlWIXE~db8He66iw6vf8s`jJ1y6m7zg^6IyUq3sFOpuI;};o7L7*g^
z$Ybmh9VzmQ3DO#r_n769qOKv-)Om%g3UEy@y>GP;TH#%V)(R;!?N-EDWf~QQX!_4g
z;VE37^P)$U#6;@?rDK99WBu|jRgqHX6%Qs$)>{z+Jp$33_+i~je`HxMQz}y;X;O!j
zR_L`M`|B~CUJs!oHZNV__=7hIokJ^yDLr5N>Z|Nu8*p}hLATQ<ywYZTF{a83%F0mt
zn5wF&t)NgfN6Q7HQ$f0QmAum2K6=2L@B9qw1WD4S)oPQp+H^Y|;-s0FZ?)QlVf0b!
zijs8uJznU&$g?+YaPaUUumAL?oSdEgt@mY#Zo<~*KhNn0A0Spk5<wU;=qjr0947_U
z_ySdBXzB1}hB?cqY>m$=gf+P3k^+J3_nRKnAV7q!ag@d}^I}9?lxVF9lNL3CvM5Qy
z0F@*NTZ65UVN5jKz%OUm=@i%NLl_ZvI^<b_tZKA(^m{$xFrcgrrmhI;g0NnoT9TqF
z*uA>XV5>)tBJK|fg(WO9<|kva)Y0n==?*1F_m{NV1FEc`6-r8H00}~_qW3&?QR3zc
zl#L1F4$cZ{V{x@<S|?Yn7E;K+Rc?RX`gEQ;1?uu8Zv3M^;HUr9{{y+Sbd!+ny&mye
zLY^*}U0krey@?E($tFyVS!CcGN`^>VQKyUM*j7r2LaeuhGDHN5lX*>hk+U^y<B&~=
zt#77jeHftQ7ScPMuklepe`lZSY~19>4g@-miDi#yu_P~Ylm}}{>arxyYTE5CRnq6V
zR*&1emw8hPR1VugNi79F3^1l5pU>&H5^lY?$N6G}lp!k6RAt$e5JmF_F92s8&N`G*
z7*~<!iXdB3<a4wN5lZ8!_-(t<0|?~_F75v%=9_yxqP2t~U9!BmU~_vD6$sofU~bX{
z<tH#FP@K+dT8o^`U%*^_jCt=V=5ob6l0Cx3lA>&wJ5yrIiaf7qwY&6_KDSyu?(AOX
z?G^LQpu|Xl3j?gFDCTo|t%PS@++(sBub2l<n0wLq0H0<qSImoCp`T*@cyjKyaUHhT
zx%$itZ2az&5C8dpkIFnr6mYQBL2M6cDaY3?#(X&FVr-37Kvhsk!9D3QIwCJKg0dhK
zk^%u?0N&z7*lYwAmT;27?hqk0_{QJFin(jG@TNw1gKf2VckeRt^qh^03r1@jR7t{*
z<$z0B&PJLt_i)42By~lV)y%>cKI$`wJGiYq(3VavxDzyr3c`@CG1SW?E{eGQ(k|c0
zml&^@gdwUd@y4(Q1Oh4#dX>K#YRP_Ck!Kl;Y>7%oh%!fN`I#j={B18GJ!&vu{nqo$
zMrR1)>9kt(HV4F89qN2eRTT6FJ#?V3dC|0>L;=n=&N}H0Wl>_iCu+5j5~`{qkQ%8a
z<J7XqYIG7(*HD!qW8J59X&aYUoF1K1)e{hs$uvbt$67ZaSQ{`eEEq>W7^A`%pDj^9
zX$^IiQ`e4mw@cV+6GsW>J<0tlN9BgKR*lpFt!|(4^dZ7&obxPZQ_jw7ya3}2-Cn}_
z`W3P?<%9Q+DU4uIl*};{QZuV6#3&<QJ|ZhxoL)?@wd18%USl}iAPAeDFd+oi7=HYd
zA5c{l>+75B?(Wm?4H{(%&5L0e$2@c62HQJ3y#3DG{OpalD4vw?fBnmAkL>li{MD~9
zefv$M@+6U_zq^JI1;uEDoiFGOJD@zMCd=35Ia&pHr7>P077HS!U^N8c4G7=Z0cM3o
zX9fLU0<v)#7Bxhs5l)LJMoNM79v5iz+9rN7!_KGpRtKH5h{G1W>6}qjQXmPe!<&*a
zuL+|TEtRmPG{?gY#;usR7R1^jL`bCqs>+b1IjykG#^oW)bPO~@Isvws+%`4JH*FT)
zxF)+@g7*ej)+lo7DsP;eqKQcprUo50imu<F^{KX)5^z$m_w}!_^|db%`<iQ4`}DS2
zSYs$g6O>ZKy&e{U(ji_3s4$_`8IZL51g$;@K|Z;lnvNThWmV(r8m~fTCS+dv)k}Nx
z<2|lfODWND+!*VH!qpz{B;nRNahQ-@jBr_w4kB8^HBcJsL8yQ@j#%5-WN`U9w+}BU
z<|77!KBvh#KgwV-TQr%4g+LHcJJ0?zdl>0R^PD(IFtvHo6seSIJWH~<iEF7?OlK@+
zGkk4{@(IPG9~0?jR9cANuru0Ua!CYENw&ZK)hEoau9!PxC`S{dQiPA02Olv{K7o1p
z6mwVCphD&*WbrAOJ0}TN%!?KCAc%;EYalgl#XN{Z2HTr-FJI@wr<k85>%5-9bj4g1
z0#?j-pV`AmPnKsyNrI_s%!;{GO;h;O%$L&{*=&xlEkQn^TrrpGQ--zt2CX`Y*xKG@
z@%68<_R8mod`)k^%XA|qk)E~D1y(o?!Vqg6br5pdLfZ=FLCivh7@csrs(JPNgm^T9
zqCl24!c<M~qKJ^W!+MWjG3Rj^N-A8^f*=9}qP7U3aXZ_jVZzO`3*u#l2_jC1YY>El
z)^j2x_kw`?8$;f{bd~We!%Z$&>vfRbE!NABjak}=ZYywJQ#*L)$~J4UWHU>#am+1a
z*s5x*hfK<*k4rRfz*<Vm#e!S&C5zm0oKKM_?-Bs^lgAu?JC<(}HoDg?)4se<U~ATU
zA>BcXD$lU<IZ8=Xs|8_z2*YMtw6w%gOccf_6#-DD3(EPtc{r{<4yiN+A&bgU8$(^i
zJjyzxVZf|dGS4Plxw*-+FKu)6#yaa;9rkbRv327T!?i9lQOx}&%d(4D^<g<y6RJv)
zblN2SHO}W5rp#%_n$zWy2YF4F7dWX&qApc!5VmG>r_X3Q;f*(LGoLk^!qIra`|myA
zCqMcSSLd8OI^@Bu<e>K4n=UyWFIi49SWKzT?~{ZPufFnSc6Tn3v|4Dbky5Qr<pk5&
zl*x3&-FxrxgTMGLKY9K8j7H~AB}E8?QncGGzVPa+{Qh^oMX%fa+e2O(yoB|aUZnHv
zGlasiwb>!w=#cdLDAP<c$HP9(nI<yLdQ=!AlQtr5<D)Ld3w)j;@*H6uSl@U{Jrq_^
zn8#hFO&^1m3WREsD3k^jAaww)NvV*-0V+yRd5Lm@!b^fq8?6I!D@c7)<>_v$5f3*x
z8ZW7{B^z67q}_GyRFc_ziSQ0#6r~7BYv^wE0jSD~D2$sl67L#q79L+Mn+b3lDSe*i
z6ln(5q3V?S;&!8T@r{cQQ-Udf<2H4luKemfRzlQY<I1;xkK_066Ri(<`$W+(1xaBD
zyB$OvQ)fAY;jnoi3(@o=*_x_$=w(bcIw2pAKuA=#M;!>X6m*i1x+;OrQ)vb~=~ofK
zYJ$Z!*#;Pk#}TiuA(AE8VnHj2NRkd~8(V}*k%ST63S<&;IG%HQHfC*igMPQe@aj$8
ze(QA>$7lR6!iat!mdiO|(xo#Ta&&%<&<ZJ<F?}T)Er}G5OKbw?9D!0)wWU~Qm|jK4
zSw?pcIBR>#s%(D#-^RCi?{QKP_6JXx$LmAhKT#y6pj8-x6?2p2^ge-k{e<}`#dw4e
z64~ux0)di(_TP|sv|^qw79>GLoOBp$Y@wARev0|Yc+S~V%==ev^3K~0^FLTIPnUCo
zq(^Hw<ZQ)UJjGn#o3d;36XAc3d5Z1zh@WO&|7M7U|0XMzAzPh32iH=r{m$1py?01=
zFyLF~3Re_dq2#O+p@pEWB=2>59IiLF67L+|)+p2F!#HGpazSr&j+cVE(?vR{gkThh
z6h(=#O~OOtgljDOtB_2*kHK1mv-sK|@GQ4BI9)EdG@J3E4tO)}u)Vdz17mnSPUv|_
z6$R`s3et-yL6k78I@tDrGgr_(8Z&VNs!c(X&p8Pa4%gTDUoU0|8imI5&T*mybKtgA
ztYjZ3=g~@2d5~F7zf&`Fj_lxsalcKP<)3(YKmJZXj^=$Vx3Sj!#`NKNXYo=XqlDp&
ztJFtF$S~k!YKXldDhvbxLMe>*B&`lX6k@BoX<BbviZmz6!Q?5f$SCmWRtpzs!YCn$
zLn_yZtY=k=JO~JMNnJzIZc&(ucE`{k_Sx_PQ%R=dIsJYIV=Z@&&k(xHdUr|E={HRd
zLJ)*4gw~{K&Lpkbt7`%&XxElWWq7YBcxh`$+zm;y44rhieDx;d#e%--Fd0pm&vJwp
z)MW|2CMXNeizSDJBlk_}O<7hXi6s!QoXi+}?iSf7XFM8_=NZdvL7FZo%Zj2XI5|22
z99k=aP?KdTr>7_E?q1^Ao6pnj^j4+)Mi+JU(k0@*_Z|M-zyB_i$@JIG+*nVb6*_9O
z_QLa+5AKt!cNrHgB3q%lU4pVgwEJXvO45$d{Wdyk%E`hRLLDON4ndvb${bNgh@hEz
zZ#pgnrmn#^PD3db(lp#ups{t0R0@ee2ewhF0BW#?uM3KKO6>yTZo+VF4H>nls)}VL
z7`9rZ%Z%xC$#xRa4K(%I2Jc;5FuIuVJM9)>ztQIuI--hVOgevZlqG4Li{5!uASm;K
zI;a{qpOkoOsfvoEZIE8FI2|zyJyl(;Wb05?70&+_4@!ttNWGK+?e!h5e*5d}b$j^f
zgrmbVuB>k|+#OKNM+jk&aR{pEpK#VU4b##Sx7vj267$h1`SgMWhmYEvoQ(*`s7uvk
z@hJ^j0}_YC3D@MyNpQj<WB?U-<C_zgFs9vBn7SftC-nO3l*=U`u|X3_c6@M`=QcyE
z3JEL0tKa+zU;Kx^&!7LZ|C*n^{TIAMo7uAB%9p>*C_N%|j&{-^2wNzrn$Tk>n*2Kl
zp2j{cM2)bDbXsD!Jb{cDY&_4w?L*3{F?P$<!=d>wf9^LL<)c_V6rR?fJYK)%?8jfU
zyg)dG2m)H`J6!$tH`wiV@zV)shi6<`-=w!Ypqh{17clqI6Sdj|YM;f5dAwqNel{jt
zF&Doib5Vg?F_&RX(pK2IB4{UcKE~X6$H|JhQz2A>m%s5vUj6-V@q_>2zvZoWe#8s3
zSS%~{zx*vG=@Hpe%$qW*!#ROa0zQFxwqmXyGrx04Rn-8i4Z%mgGO>!$c`Ua1%g^+`
zdw&(S@v>!|h`}?@vH$fip^_Gtvjv0shzA=RJRBrMXUDieP)0#>ZLdyOmC^|Bsasu+
zrDAhBVmLZu>>W00@#=Wa6<Kj!%SL9uN&sOMlrEt8vsO(BWc7W*7^HQ~;uZ^`5mm`N
z3iy6+jn6ISgh*}$iUR>Jo*b}<9di*<l!}Mdg47S$fA$-EYk8YH<2k)a&e8K<=JPE_
zP&E#%M@g5lPy|@^oyDTiqPg#RJVF|*k=U$af9ScbLgJ0<9Njx@vYP>|G)WTE>2zte
zTEuZPRsFQR&N)w6mZWJ)y3CsJ?z-mh?mGA2B|<8qxWneN*Xb*bO=rxPC4(>`?yljo
z1v2m}v9Lu&VIv>cs*!RB0fIJFo?^?adEmv4c~%foAbkg0g_vZ>&b1w)ZWmiYnrB#V
zXtf5MPsXgRud{Wj$6&q1c(ec=aQT^S#`mYRYfU%sD;X|W0m3<E(+OJ}m+>-y8g_TL
zxb)@U<>25UM<)kp5iniW?C<u;vl9a6Xt#Q7cGvJojF%L7#^~ri^CDxWLNf1gtB`wB
zIgImIFH!9c-hBTqZ@>3Gd6DC+d5Ue5-RWF2^yPik41I-~D6YuzjI;AoZrys3E0?b|
zl^oebLT+ts@%!KTHh=WTf5LP+{q-V8qoWWn6}sJH?d2C(Pg=P7jKyd{FWDx(v4yH~
zgsHJAX!^&LYBViXg^<2EIc+8Q*%+5DkwPNFm~=8Hpu!XqY$K;OE781lP31$b1k^%F
zluU3%+4T1b@Ii=C1ql|kB9uLnZU-I4%%&rjMaty-jO~^n7;IpyWb^t}I+w0+@b=Gm
zc>0hlfkK6Cp84kQv%GkKU#3JVK!pMwbtvnCAW#G<APC!}H5lib9h;Ms%NjQjNEOm)
z?{RT7rY>uA6e5K}si+wQ`VCs2`lL=EK{algOSf*afd`k-UhlHIJEYZ5@Wl)%VyyRI
z4PvE4@H7!Djd)Vg-B@RSenvH&LCbS;JSLDq)0FNVDhQB4&`g-%QAT5hr4r!1MBzbe
zymDAqugv8!rmDa=k}$?cb+en+n%VRM?+dP8-zUHI9ND-a5}sQxeSs^xm-sLK&;JSY
zH}3G<*S^E;Tfd91EdAIKd51PjRI!9S#pP4TCKw?R!sD$WD;&!uL|UOjjTN3^a>DWb
zdkF6cB_vv-!+=h|$J+Wj?N*B-&#20x3GmKyCbK!lJkEJ-{@s6<s|g+#_!aa0ThFq#
zV%}Qsvb8%T=_f3U8RC~<-q~1ZaehWUogrGD)8i3=46wDqdEYn;e*ts6Y~Id)0dr{_
z@rrr1V!oJ;z!zM;zE6Jh7Wud&0G@f_74~=b_%Hu2|0#<%?(pnazr*dD-^7=O-fJF-
zMVb^@ED=S9%P07Bf{}{k6PW8z;~q1=cMrTL5Rhnv4g=c#9&2msv|0&eo>5gLS(dTP
zb0)I|=1HmQ@BZW_bg~hC>mJdXWc}sOaWI{t7nZCn`QCaD6L>Di0l8@WR6+;{LqIgX
zs<LYSJ#oTlION*soM=2D=s0%HCTOYIbT#*k#U%kg&`rb`9&dH?`3nh1yl{=%Oc|sz
zh`PcAF##2>wj72Lt)#_Qi=3gA4CWIOo0APTN&A-wyFJ|4vPcg&RhQWdH+bguk9pX?
z!QIYHWLoj2zRFtLqS`I`p=3kV1hWaJ+9IQnx|dLm&L|4UY|*@Qooa<wP|q(ozjudj
zs2L3R*<4#AOcIiI(m2;>D$7x%smp?*uE~m=IF9J|d$?7g`+PoUJQ}e~(_ed-`^Q&P
zNRW-D0>|2*MaLOr<uS3Q*G&*6#h3sUG;vq1t_hO>g{EmXU1dYBOa*0@LRpfQhSSlE
zKzM|;7^|3Dfw{P#J+};EPgl3Ob2w!*DnQg69-OkdQPAmX>bhn*FX*@06hcszA>E`#
z2tj2G)<9XMIGeMv*`vL)MLwFLZN~QcCTo|j@S}hB-?AKM__Zrc*Dh0JIZhg~5~_UB
z40))8MUi5#ES$$`kF$=dhO%&I2|XdOIO=T9!NCP5C#OgWp;9QVh@zNQuT7u>s=B7G
zYwEhmAufu7vaE@sMw#)}n?GSTo$&ngFA*m(3W2K)Tbo<_-Z#I=|MbUyOqOMTExW(z
z3WESGn*k1#Vr^@KE*@txqJE23JAeR~oZ(%93KUAL#L&L+2jZL0h)ALOJ^XTk&r+=M
zEGH>p5L20^#EFzxp&HMf^JrAlgNMia=6QjZI9KCvjXFnb!cgIp7RGp#4v6Cfy}ifX
zcYek)Te5NW8PaE8pj_0r?gm>M`|RBO96$JP{!8wjo^b8UUt@ah6@2;y%JYX**^F3t
z3L{ZfP8eCdFR=uT16768KA@}}Nhpb15uzP4w-v>F(PR_H0ZK`PgkGmbuh%0C1Dr8f
zYp9LII!94fD|14#!MCp2y#G4ar{2^l*<?b3q@8r>)Fs{Px6u3hgtDZV9N_DMRuUtW
zS~b1<l?p<jq{IT=G!tX({*dM943o_pPgz;h?X^IumBvVclhh`|6pA|6WG<pK0!2v>
zTUr_-shTExrE#u86$aJPXsyYsf~w3Jjn3)z+O+yzhA+Lu+dq7V$@l^x1g(uduD<#e
z-hcQZ2jeO4{^UL%ym?O4?vr#w{?R}E3YE^eeQ>~f`zGtVnoGNyE2DRye4jj5OjE_6
zuZdd`6^``aZFE}jrB`0%t6%>b-}uJYdG_W_wsv;tuD%3SSx^=QWmzztP5H^|KjBaQ
z<lpl@fBy$Oe0W4zRZZ~ouZRC%&m51C3gI0fX(e4+bxG&?ErR`hf)#UD7j%9}=B@sa
zbj4gMNnO^od+kO<`U%WUuE|%-H6=Q>BrE22#oW6JUl??&iSa6`0#oLUM`sQ5UYEfO
zFY)eA-eoqqKng)?eV6^0zs$XZ4>_DP%s+hV;tBJA@K3%(r3-E!JYubVgW;}be<$F|
z<lVo-JX|r)4&FhhIbVGFWxn*4ukf|6eT^H>-e7%un@+cj4g+jeQ5OY8Su&f?_}R~1
z=TE=;U4HO`AM)_vn5ruIyA+gb;VGoVdWp~xXFL1+*m%tPkkZ$rF6GMdkP62*NpMoF
z)Fgt(3Vo+l(+lT4N@@-U>%^loHkY2*JLGc7tlh>ybF!;+^P*N4=u#6Enx2aYO`|S|
zBg05Z&V|O)IMRd{P+6P`m}<>GkOEodj3*a_QAm`8jPx#7wj=sM#?kvVdAZ<FwRomC
zAgZTy9$s)VN+@m2X&kZle2DAsFnTaybYv)6zUlYb_T;%B%N66)5w;YgB&6j%>ET^U
z^Zz66zk?-9&+|_3_Z@%I=gsSH`Ks>fs;;&`gMa{%AOV5^-Eb(05+zC_X_m5FNuimQ
zc2}`6u@Mt7W_L%slt!A-a7LmeN<(V2E0Q7!k^l$~G|=5>@5<NpCVd|Jx&7m0RaZ9}
zrpao~7tuGWZkNf*U!L!L=Y5~|c_hoLtAx!Keh^^%8aS1|nd3MFwK}zW0H#UDsUbu`
zmS!Z$gvm6)bsWOQ9t+(r)9IA`{e7mh*&kVZR?MgPv{g8&Ahacc$_bhgVW*DoYVtvX
zkp-UPSDdM|V~*#}<E5(bUV*4OCw@SgOu-H28dOW{`&9+rb+N`FPZYV31R|p0h4e>;
z0tZ_Nlu9@gE)fziDoUnCQQ90;P-T|-ptM5Dl36k(4n5*#i}uPYli>trGmMmY3(IuY
zPcXRp5`$UB{`Et~hg0xEd|z<>!P6uPp1U<BH%)x6j=wJGq<f$bkkTP51#u)Q&pj;V
za38mn99!Gq<jIqqI(>@8r6ropCSep+_DotKr9>%3p66_BZF1%66>i?VMUp096Lxnt
zky3L1`3I^&3Q<WrHa0eR;=S+XOJDjD>Rqxtas8l@Hkj&w90W0*mWY*g;$9D9QtW6S
zj3Q_>DoKh_s3OH`Sw$xra2->%?~TD}G|}TRCZA!AK@|qC-lSC3-mXFip+VTPN^+1c
z(gbKNa2$noHNt{HkY@(R4-gi7H>!rRtep?GSfsH8aexyBh|UVzn^(zi?6VQ9&|6(5
zzVHNBKmAGMP%&{8gS{NtH>m!Yi;tgUs5~}r?6bDgrPGaASdF27lhWzncSGXkHft+Q
zE?rsyp^0~fJg91H+;@SC58lto)29$&%wZ}?l_m&6qA0B5#w<uxvbnXzYp=e-l`Gfi
z_xlt@Ns^|_W;4>Hs?Oi*Yy6!b>gHkWjs{O!yewxr-De^8kRpdNB@hD7aS^^(J#i|7
z(sS^-N|$i7M&$*<_wj2HDotp_9!JTPPO~a);tQmf=w!-xw#@Z|C2r+0vboCsD5VjC
zmkn8PM?APPWu+~NT|?BYp{H5(A#+@C1e4JbMUm2KEmMnIoVsv{_QG+xYi+bC5kZqj
zKKMai|E>R(MMpx~CoxmPP8SnE(1>tL!L=KEJaw4k;Ig*S;AbxwOlcX;T&9If+X+a$
zfJ5bD7FPM0@BJX{a}QB#HmNmhG};~FdV?sggtb9fquH(+GaG0gy!0@C=4bvK2M33I
z;R|2jm;cqj;tOB=BGV-O-|0qgt6bK%Fi)oYbZQ=wymIS)2h5$jnAc*DVKSxD^bv1i
zKAA0Z>tKn^JVrK`IT&TsLyMn{S#U=@urg(-Es5?h&)>v+Iy#~(QrfL$YH^c|2Ogx^
zJ5GDGi8Uqo4Icc!2e|P&zgjWRZZq#-0tg#1ZYj8NbB|}{%vaaz{Fw{-o0w-_KwtTo
zg%y7KLm!}V=0V~{l^NV@w~6Cg^@Qnph#-b$3xvyx)!cvaLH@)~{|tk}K41Rw7x_2;
z`rq)SFMXM5lK$^<n0E;oxE5DAxK`ldlSL7$X`dTDIQayrHF?}1pJ(Xexu9*Oi7^JW
zscslrS5-UT#WZ3}l8|^li)Bf%+{NNzJr~y~f}-Gfy2xU`O_&G>1+!U<6)AP8ITc85
zEu|ba6%*f(*K07#5K5yQ2jd8wWP~je{Cb-(j@ZgKaF;fiswr`O7tl;h<$7<FLQH2!
zPce%m%gc+Le()q3OS0+X*%scB!i_ac-Vu(3VIr6n4)tb$^nAj^@@mk5lPx-<K1Noy
z3-x*%i;FB}6_I2ud%N2t$&4U~sMqV%>kSr{R_U#}%!Xrz{R6V1q}}PzXf)~fkJ#DW
zMg36(z}re7f>bzphO0H60K`B$zf|aQiiiSAU0`*ACj_>XxV0F+*#Il5^cbb8YP&WF
zV^O85ro}y%qDY8+4<!w8<ROFy+9QYs)1x8dq=)umqIiL#8Ign@M|)$Ui6OaVS?`o6
zlaft*f;fP2RiU?@Maq&ao1wL&7B#6ix-2fQAf!TWNvtjqz{;sp-1v$|O&Io*Db}+z
zIuW=bqTXXV8gn$s8H^J8lZ>$G^WIg^%90gOC_yYdq~kHuE~eJxu`{QL7S{=)5I+nE
zq8Kj-@os0HdI3RHy}ca_tE;O#{`h;C%@VF(zs7UVJwv~LK)-*;OD{^!pT9`2*8{5V
z-vjrbXKV8o*RJ1qSM~vG434Oz0K!;M21hAO(ucsosuao`$5@P3xL!5UDjWxl!LguC
zr8&k}C?(qWu%19B8l`I-&2n1p4g(p3YtgB}CJ8mCOIEfSC?DPEFdSw0o`*^cTs>pC
zo#DEQxaQ+WA$6LBbxRm(42HZ&2%`|fh(;6>tgcbmfT-I5SKxXcXC8ixTVMVHUXgRS
zyUX@Ka(uB#J{1H(NMIcH4~O(81;+7NJ=x=X?_XkJagELKA=AVm&IGv|ktv00^mzaG
zKR{!3o%TW#0v|7`bH1~P8`VkEl%sy1Su#W5;5dSn)pg$U_!FdQ!nJGHc=E}wa{2Yk
z<VAs0ie#2B7!FBhGxR;G&cEXkJs}*lQdq0Ot)60C*TditMs@VqVx_7QB+`Vq-KlIT
zl%-5d&;|q;*To5ZiZme#LQ0X-2nGAb()24$l~x7iEauk!S)Lu8AX`7i#`{(=#^vbp
zbyiMoP;W%2Ny_J+{u<|IM_gLGN~;|s!YbjxbzNK`NRn9<S>t;6UVs<IteoiK_!6y4
zd<#qWpC?@C5!sS6XHT;;KIAX`<)5bAt+TgtfM2iki68rZZr$8RDa{Mdz0Op-*fb%R
zF2kB&&lWs&eH*)Qnz8lCKlf>zFMgI<t;yQz1}iHYEHAIIv9UpSp^NLetgUU(?euV6
z4;=6USlc+ppZrsQiXZurkMqUPf1ZEv|NaHO@a3<}^?C3|)uywyN|7*%H!-)ahp`Sp
zTt|=1o0)gsU@p_L3h%y)d6_1JbLRC>aA+(I-$UHRe0%>a&kawIt*^2E#0u89^sn4x
z<<uIrdWf22eDRs5IXCNbVeu-Bw=j3!!rb!$+#q6Uql@qrN=rN+mhQWcU}1sCmYhCw
zhMn;N|H*&)$7y%#9PAy@sMq+|kA8^Fn+IsAdEtdCOtlMdVgB{&+i;ut=YJpPOP{9}
zH(6cXV0n3+rKMHY=dSaPCs<lurPEo!dR4_kI4(<T>-@w|{&7C`v5)bUFMW}J_>cZE
zU;Oe{=US^jx_CLCMW({Aq^k;T>IuqvodcWUBN43*-Z&wiX&UL2!WYO+7knS9Z8ZgF
zDx+o`x3VfoXUHICl;!k2!-BOacP^YZif9_~zTpO5zs0C7NKY3O+Hhn7g7uQaD29O`
zxY6a|dXLTJ>%7_y5Rt+$ma*foD;$=xw2GM%9>({OdqZx%pm5`pcySl2CS=x9)>=g6
z6xTLbiaR*#1H4{Bdm%))0SBu)EO-h(h$)j8?WRPxa;Bx=Aady25_M~j{qAvIGm@g*
z#Bp3!dP`h-=uv957Ro?bN?g~YKRD#Y7pENc56Sa{$#j741vHy&TCEOotwwL*J|>e1
z+dDgeU}<@UD300OyhWD1yH*#rED>RdDhhOwB7~`uA8^p7n&!T(b<|~np3VqbO&rg|
z!A0r<Gxu&p2($y)Y=(C`N7%^9@6Zvrj-~Fn2)E9WU0~#{(Kxw*+i_WMwwV<Lmp;%V
zs`=<F=jMwquzh)pl^~&0k19(VQMI#VS*0}uen73(B#3Ju1hsmTGOlA}ful9`UJtj{
zz{ApLb`Xi=LmzrSxf}E1wLQGf0v8``a{1M()Osy$ZXb|oft4jvJ7m6Kiej@r!q$4^
zILND)5!bE~1R<?vn`X03quHX}Zd0q(a2!Fa)h3SX)j>#9I<i)~!^4j}%7qILaO>uE
zp8m!&3<g79dg%ooym*P;!UEP<{J`h2#~x#AXNRJ=_qwgyjto{CR9T`_QFUe9%5WV&
zthz47Vv_<>=7>11Br;NA$_!nUcyUy9)_f10=1AROQYdsBf)1%S>sT+q4=h$i><v!w
z{IG$)(q-eUi!TC{84@lmQESyOQu6A{uhG#p&ebP`jUH~dOPQ5eN<uxO$V$R`9i2u5
zu8ZGj(Oe2yUhPtAwJ|t!){n7z_7uxeK)HO9$;*8{`XlcpyS2z<cS3uq%LDJfg!Vn$
zu*U0GcG#I_*zo~_$&^_TaO-f!)t3sQT8oMFNO!+VP>ZN_I&_!TX|xt9=A91pdL2gy
zI;|#?$&573sse+6Fp7BSp@+F};UYJ#U*%Ju`V6mMxs31mv|26FG~w{@kjeB9U!8wP
zhq~K34`USCDy$3jUWY^|qFxP=5Anhf9fgF}QjEq-XBqxV4{IPFO%O0ICs$*`o*z<V
zIerl08A)0gWGTpWhRFrD_IrGzJjK!vJkH+km@obQOH8sHYv9_I0b#6p^pBlm`TIY}
z-sLUC&~m}PMzb4YWi_B+twol%Z;aJ=J+2QLM#DaCqeT#QsI?YZKXH;|?+Q&H_J?Es
z)j$7DKKhBrsMS5PTyx^=3Xi?-BBSAq=U%u@Ke0?FGp@M>uNlpF>XBccFn#_hWT^<y
zy!VfN4?q8Z`a7Jv|3T`F2A=C81ms1=fAin|7Yqj@9(dqEKJ>vK;QaXu1h?ar9FIn;
z%MbnV5A)dLkMl48<uCIOe&HXoe{lFmHbi^7@oS96%$di%4w+E+bLQ?{%*(r&bBDP=
zY4u$&&oMb{@Ar7FJVoyZ9--VD^OeuO$|Ni1%twT=;*k%WVd?wc&+e5S#L#lTeT~Mq
zWG)b{UyaXe7?1k6^(J1}qSorMdi*5G{uP?O;b1u7-~7u@^3jh!O1<HeCz_LIR(bTj
z7Z?p@JonNK1}P+y<ZYNgjVvXB(7fjZ@8jqG_TS{x+54#1o4E6hvMh4`;s58qWH=mg
z|NR$v-v>U(*>mTK{1|Hut}BQdZ9e>wAL7wRALU>Fn_uA<{`-H-!5<;>fEF1J$8BZe
ztX<IE4zbX<QGjH^RE?<R6H*k;Y(`KXp?e)LR8e6Te68qNOA3r#pD4>QQOvr|NVUe6
zf~L)Bq!Aa6I#AZh&L<pfM+~1ii4J4@AY}NWBB)809vD$xin(@FuzCPJb%kNiL!hu&
z5-bZ^S9(-HIxg5F#3W*QdCD}47#|Ih#^5)*93S`Da!1UD9;)n8cY2gZf=)N&%$d`?
zXkO&tXoqC0%SOw>{w~iK8P8iqKk+Cw6Ovb7L`#K_Vg2+uy4_Q}`r0i@CFfP741m+8
zm+5puV42}?#%z`#%Y;177>@e1TV1-{9`$;I3l}bM^X4rklL^gclQU<|aP#IZX0zG5
z#yssG5Y<AI%(1FK_~6%KBnH0~V~Yfl`{>Zea|LpmQ1plRofg7zP)b#vjk?kjSsYx?
zN9GyE3XD-44ieT^R}uagx9BpRoZ{NprFH)TK|5rU84jldw6@&3alm*sq_q&CBow_1
z3@1C(CWm-linW>|E0D6{U5jh@L4@#pj0KtJC{-Yf4A&^!AfnlBvs)y%)-cHmo_Xd9
z^xDY4!}GwHf(IWuOFxrb-#H>JA<ql?j^fZ-vfM!p3i91e%-ja!`00~;?+<>M#idn(
zFsyv)0Hq|K{>&#y(~OmsRZgEe$MW(@HB15^T!OI1xpU{)*f_>3ufEK)&p*$LFFwaZ
zmmZ<h>4LQ^udHzT)G1zl?e%*P-Ek3>Fb0c38-<M{tf@gG#LSMcu7ex;;DF6DkW)<T
zftHB8K+g)wQsLGDgmB2qkXbJ9I&GY^<glL-hADo%L?}Uj6mrYyvhsoRO!_mPd1eb;
zWMss6!!eFOU}>X<+E`_GC&cb<b6g&<w7h^<uThjB3!l6wab1rviqO{JI}TDQHm~1c
z<IHJljSj*OSvj^wG~Hu$vCij{Ek66%mpIj$qRot1<`68lsdd_DZ5TZD8r!>?I4|j3
zJd2KEjs`Od9Z+l!m`*cpY~5h(<Z=F^pZ*z6oV<@#tBvou1YyYX$_h_@?epYWMz6Qb
z@^X(nm+bBDlIJ;LSXJoHoIS^1IDUe!fBkFx?(cn?$z(#K(IAdvM&mID`v(-|AN&yi
z2RhVgqcK|IsuDMMXoMljuusz|@LWOaJGfzh$r3y)X0K1M)I$oV3iZXJjiJaDe&9kB
z;9>B!0Y74<G$u<aM(cd;aGms6o$<j9M#Dos@{^Ad#e(su<glN!a(aWc<reuYW#3ul
z=JVe}crf9<aEE5ShO+XGNU^fn03#|Fd$&l~f9316ddtw#h#;o3cAUYrr>W@~&6dl}
z8#nmXUmo)K6X!U0|0!O3ZJXC${VK1&yp1Cd`O;^eXK!!H3%O)ZY62ti$nl&?^xa-t
z=l}TE|1OU`{5`l)FxLWEj5Qczx%~QN#?vEiZobUtzVKUIc;FF!{Kx(Tmo7byA6C^X
z*N<3QT;#9*mHz`5E?(e2`@4US%a^bIfua9Fm>Z)oTH>e@FL$ViA;qvyea<}h9h@*g
zX9@mY%-y?~o3}7`?l6z%%oEDdI$t<kCp}hUe0Yn|@Q4rp#6v_ihsj8C)X!KtwZ`gl
zlVX-}=&W$_g(rv(CY%a)Xw+-yw=g%>Vx6k(<EWIQ*Pfx-TY{#>@nhPn$CzAynrP1a
z){UF|+kZXc@h8r5_Wn~`zP!V0uRh7^FYn;WL%#Bv=h@qv@Io%xR|a28JW=3Pax&U$
zYy9=U@}KkYrT5_a4vt%8$yjTV#&G%aWhS#ex3*s4i(mdE=kI@rkN@z0%!LaN;|4CS
z<6(k;g<g-J`%8a;2Oc=j-}}G*eXd-&_D90}I584I-P%e!qg3TR<%cw;`wZN&O8s!b
z)MMNtMLQO8G(z{f7{^7F6$7gcEh$l+gY<n$3}fG?>xHaq!xMRqmWyQlMfN)-&TIkg
zEm1zcPeVt{GL1V6NK_y4lxDmN7aL4&o#8$E4o~^7GHAQF!cZDJpE#_deoX0NZH$eU
zSZk)}sn1Xrn4rXXbzBUwP6<WEBn#PjbwtyNQ5z+mSEtzV>0hnkt3Ji{D|~VDh-dSX
zeP!^B!lk%#okxp{G&h#n+1th%OPVH>vP5fz#bIaXGVM+c$F<a(0qyo8M@N#`tdCR~
zWmzztj#*e(qTB7Ue)>M#Th|#3`b1I8>C>mUe*MO~cCZ`Xx=E|muJ%Q$Ho9vxjvpY4
z9A8O<>mc3Ae-v2~2!WZV5Qk_fv2!a3$MaEnfwltN09}?ijv&)MGi7Ou95*psDQmdP
z9kzES48{k1{|`Qj7r^%BfGgK`X?52L+!j8H=7}D5Z;jb&zm2n6VyshDFRdkxe4;4A
zU@*e1CM>ftS$`Lqoh8&3<2Z!%7E0%MT2hOB`u#DlUpwUNV`mvoGRDIx2S;OW?T$dr
zxPEzu$s}i2Rf7|*0Vk1Fq!bXf+I;T^e~^v!lQ`b(_`_<z%~;Ff;URgEFc|G~<?1V}
ztgLhC;v=lC9jlD0tw1;~aa`k(haX{id700C;q$!s((^q2$a{(FF|Ols@qr6my?PCK
zuiNliBRqi-1}h9kmI$dTtB=_PAuGdtE5NHolzCDOyG#<Og$S?mXEOrG3(x@*Swn_S
zHB4f1T8m9Oi!pv^Nrr9q!qce5E=SpzM2<Q2#916IDQr~{jXNQWy=s7P*!4*^TNL6J
zp<fLd=J^0ALbB2)@(a8mqO=fTaE0RV#!L8do!)9dyR(D{>Ri2Y8Dmco`HJgTuJHP7
zM5kj}Iu&s^E4cZ^SJ>MgQ44ae93IhK=x{hQ92|}i#t@9gAafF}$nbdL10TlsLk{+L
zm`%s@dW)P{StE|?92^`{lr#E+ZCux<UTZU+<T#E;DXZ+*FbHWjn!Nx0A7E{5onQUc
zf6M;C0j>+(PKP*(*xB7>b|*9bovnAKsWS>3FjgXTL1-j8Nx*}eYax^5vP5Wu!ol-H
zyuhJQDLN|&8X?IbL6;?-?<0JNvdHm$A1927-5Oz-QMfK`8}QP2iKALf+?pY!!^-M0
zj`{=EP8OVh<SY+G>*Phm<rlX&wzf>W5%bJ6!<n39VYE%X;UTO+X*HiX5tX~N6$I8{
zrWKQKJcoGVqtyhDE9fq-a4VBE#+!W4<BwBVOBmL9`Q=x5{pu~O_Lz-lc%mR4y~gu5
zU*~FJm{7?Isti3aUX^$G!5{cAUJ&ryH=bc{f1B;SO?LLTnT*Hu`+c7L%9DU5ib5Lg
zn1jQ8zVYl+yzhw*^3#9v&vW|J{Wy-mIN%DwhyU0I_+S3%KjG*9*5Bdz7hakhp@?@+
zP;E^0lx7t|7X<bV=JA|)A*;5!ehYJxmH3U2^e*P!TbM_4=F)ZXZNRH@=20u5ECtIe
zYYYY>)>|d#A34K=;VO9%a_!|EHddEuH)Fmr&2c9uX^*z4&6%6Kn9EX=Y6uV{MlpN#
zIo$XBK&1h51?{C}b}~tAyv6$-f1J{C7xPW5cA1S6{5#AqzlnJ@AYd``r|=K{z=!dp
zkXN36p1p$|w)VEz+uN?D8wY*9_SLU_6Z7Yue}?xw{sDgSCx3=hr%qSxLs7N9zxVy$
z!~gh?{~>?-Kl{79_~J|RNyT@MxzKH<jwG@QYf5HTLN+7y1vB48l+X~B*{%=*<N9c~
z(prinL)QZQWCF6}&~-76!|6O{9QYhMK3D1ykA)cz_#R(ZJ`YXW><2zNYOvXds5xXo
zv^aa;B4xvHYj>A0U!dDKM|~MgX1Vohop5%73+^^wsVPiQ&be39WvbK^g9*BvfYYEk
zyUuv*V~YYCYo=LAVI5Yrq&&97eXGZb8cSTiew|ldo?x^f8NyP>gY9d)a{WbK%_O7S
z%n{Xf;a1*-?|J|C&^fch%g<gR9gp!HL%rsarjoo+?Cu<JbO2!(P-{j+aYS5i&}uI;
z9_37@LyS(*dXJ(g=t;$~<EIEipS}Hk!YJbO>C;@fa^+6-?_Mr795ULxPQ12QeeR4#
zs}wiz@LdP1EQKm@jI9P+jHM`YlHnBVI9LHP&kzEFIK~3WEUSL6_3<1JB@z;;=;|gj
zDVgL)bk1~;UX3Di*xKBo*)e39%lZ4yqE*Dv;e^wt7HQN2o_gwOR<Z(ZJ<_7e+6_V%
zE9wXYd0vuRful4=l@wdss0S+VKw}JXy;ddJlp{``c#vF#2pe$u+D#6RMugU5n&ucJ
zes;jl;E*F_Da38&RA1&kny2qOhvT{IY;Q7|P8d%{Or|69JSRyKZr->B7<}I&jC^Lv
zl<n;;PMkQ!qmMj6Z(#|Xxy_Y@^_5kA>=Pg7cR%%MUU}^$9)9={9N%GSd6|WU1@;dP
z?kRJt6~=QbV__|Ejf8RrXoRD2v_Yy8w=A$ufFA^a#*`9KYMfe(J54dV#P>ZC;~}j9
z4-6hrtBaG2k)os(HRwmHsGy1R5`=89u+ZgbR8VUgj-RR%)oUo_a<Dm|-EnDmA`Z6;
z26mm*@t8(0Kolj?I3&^oTT;g**U(fY^*A6L9@46Xq)Cb$PRX=Ge}BkgZ;Sg+o+l|{
z^0LQsPkxCTdt-FiBbnv2+5)?Oi<>&*ns8B0PL^j-CZv-AnUdtnaOu)TjD^ixn@lDn
z#*-n+DArduXm`3?zkZ{tcBAopi|4r{X@QiM`_5gWx448>iY!YB{D76^CI0-M|5^Uc
z|N1LzZEb;oD2zC9{205t`wag8wl?p0Py#AJty5}3;h>Dg&Sr#-fI_9HWJVxdyn2Mm
zb7Y<nwi;xmB=BPVFs>rI@)Y0o@I041PnieL;Zsug1=+MtvT&9sK5&snZ-MNnWba^~
zB+pn{Jj=7sy~G2TF7ep=&T!xPbzb_$6$+zx-}jy4wcoJp3<Emt8O?g-dT;gowHt%*
zJ#aj%))d1b&fFu((=OrC27{@ldHq@1?J2UoM%`cL`yX0ITZdGH>>nNR(w9ET^;@s<
zYG#<(s(dXRa08E0Dr8x;8-MTfpX0d~pJhDmBUP1eYAkmqqND;TCDu-GcA(X+vCtdS
z@9*>c^H1@|f9lWjBcJ$5YV{f>2q}vhmma*tKmJF*z+d@mKhH}qzw(ZuzmsLGnYUmo
z<^c{$+bSrz;ZuH>%vDj4=jmME_Gab*CGj2RXL!&1AE4Rmk{*@p9~_btISairJp0_s
zoV)k{m)>)Nljjz>{K6)M(LC{eXL<cMEPKO%c6&zS4)b~Q)#y2M7po2BaEQn<j8(YW
zC0JZ%Ff}x9JV(1VrR=Ox_m=tI2Uk(JWP}{{k9g_JzjKFqg27;M1XX;TQm9g5;X5!d
zkjf&Znll$~V*cVw&+s4p#GmG)AN_IawHV`TRGISN0~h&)fAHV&^MCVi^U^D?ysJ}@
zGh=ZrAgT%0sS*fdaIz92562d?i;S`3VQMZyN!-~K)vV*?IVNnNqliwCb4p3Bx-Ofp
z;C*RIBZ^5~mxiM`3LRQ$jWQ1K7ZXZ<LbKN+uo088<WLV0dy?ViHtA$aXQ9PPv_XxC
zlaGZA$EyV6m@~~GmlrHeZ7Lr@4H%0IDvvO%p;=6@<uaxy!CLn6fc4y=R*vA{X$}uw
zzyw{AQX#?!;nz65)ZyyO-(c_h3%r&|2Bu;z1h~G7RQgWZ$XCDq6br96m`q1#RaV1A
z7PM9rMTt}}RSIhpoC86l8PV?4sn>_JnmroLHTL(8$cqv&9ixn)lqGAcs~Bt9KR6_e
zLXI6f#<gqL(D!&EQW(v2e;<t@mWreOBWxsye1{-(DW)Sxhqy|jT|qXSV5K4qLzFgn
zfd^WXCNt6`!Sy`i$fwA2qA(&YT(nihLQ<9<w$<d~YCsvdES_2C=H5R0J2^!Xv$cJ|
z#_=^y-`C|m-*bxL?vU-5Zg6HT2eZg1xk=<qY1RY$u!hzKqcjRx9Wb@VRf@DfpiC09
zmKX$1t$|dQaJ)^uU1RDrsI`|^3Zb{!M&YtQ6O6CjWHvnDKw1hadnqy3!)aqE=kNV%
zSFf<MyG5R-=(&y4ZJ#<R71~h63z8!QL~%g9o|9%%cK5b;<l*;n;ekt)6^1dWQqpR4
z_|Z>%od57!zs>g6HYZM;#P>XoA3t`_4|T?9tOf@M$5M4Qw8k+K>x*j0q$m)MgZ7-N
zx>)4cz{QpdH?HAC0iwvzr9t7+SZX4yL2HL%9_Q7Fml*dAlc>w3?>U1L1PpdF21ipw
z<k8+(Wo!Rsjvo(L+vrehx{MA76cfqDnI`#;kMb;z&5%VwRIg!$k1RcG;3AE{mN{;g
zAjU^{#f-SurdIPwSJp6il$+1t%*NQLOB}|WJ=Vn)0u^|S5|7Q-UtluX=g9F<`<qCm
zNRtVo%t%T_KJWaL&%eOst5->rF;bU!zDp1UL{Wgz&}uKx=`7$lF4h_XUr=v`SPZ}S
zsbA;5`z~?e!b3z+gi?|$%V;*6{Pdsv6a31r{3=`9+gJ>)>#?@JhHxAPgTXuE`UlD~
zGS(8v89{b{P$j;$=%QKqOi6_sc-7OKBPjENV6lnoI2ch`7YHxFmYTB6tNb<JXF3{@
zqX|SoBn(LzbGa-Se)a3DN2fTqC^^|!;opDy3YT9Q;Y0~v{OnhF<k3S;oNQ2LC41XB
ztGz|~j$xE`m?gIeeb7o_6nGAm#v=4E!a+!lQxur#gi>aBA|_l|CQTqJXB1aI&!}H8
zRu0Oz6mH02?=;(mMh>^Rl4ne*3{&kqvviirH@#OkZ*H-*xmo!xJ5}Z7w%%Rq%E+<`
za<06uvNU1PPw6hyNYjje{!9OaYuB#v7ygsK%+ks#ffbaqgohuz#Q*ER{eSSE{@h<?
zYiIipg7Ldrw=IyZwfJwtT(|<>!d%|P+`fgmEOYdnc`_PNzJ>XfvSj$=Gpt7^IJ>Ag
z*;wJ%K7EZVuZ(b_lrMkoDIR`wz_H_XWKpueU9#Hi(RVE4v`d=YB)G%ez72DQC=2v#
zLaA~h5fLpcy}|t27Z~>orf<i5w9S<~XDX^Vo7>kyQoM!vtu3}Tx9(!jonK!uuUf*l
z-`^~q&6&rfY0fYI@;~LqjT`*TpZ<$1FRfs$gG>`HUAV}9^I!c}{FT4<*XLsTcV`W>
z#Gt`|D^t>J4_i#p#z8oBgaDlv6rOv>Xxdc@-KbZQU)mx(pPA>=EF>Q)O14~=tG>(o
zM<b3qmWgY)<_eGz6VqkVcS*fA=TFz^w-x;_Mcg`+EcWWO%Ewt+AJAR5bi+2Y#An8c
zNf5D8tZ*_JvK1A$Qeu>yPhaL7dTop`c-BK`No_$hy#cpR`*|){bu{;Fzd(^m(gLJ%
z$%V&GW0}{D&tTl=)#;SJ6_~ju*IG+jzPS~)v$MzkUgh*5ZZofPtT3ixe)mDV$Wz9n
zj8;1$>lO5ROPn})jD!6FSvI}%5TInq`iYa2Magh9qSb7&*z4W<stqn$!f<n!sFu<c
z8RNq-hlM4s)mUf<)GS4Kg0S8oUR*%uIYF&XUR3rJx04gfBB#hwWLXdf9-~RZ<^+sZ
zM1YYFhsu!by@s!bcu|*?&T(#x464w0Ud;Bb34^_owS^cx*`+w#;>2piEypmPxdcr~
z6gDg0FO^_TQKi4QE=n7c(S-i)0mssmQk7K(njfP|OB^cf!4;BBaC49{9waCyWLEF6
zQ!2`I!jUY=@9fnTyF$JBAUzt6n2fk>(Kr8eRGK<tEat8bia?-bNtr9^jR>m^&piDV
zMuP$Geed_sXtt`M6e(%9I(*{eKg@6bhflKDTcp`+v9WQCZ#?_#z4bVo|NATgPvtP%
z$3y|H8(>WcuE0@MR<aeQs_e-UrxsV{%1R;}2S>OV<B%zZ91aM*8g8C|BPb>_`g;?~
zqQi6f2sL?%<@z#fElbDm@Wq!WJoVCumNVwsvqR25Fu@5l<ME8?NYd-JxVD$DT=#H_
zA-X8%!#f70T}q|Vo&(k((h(f)P;6dC$;zfh7CD21Aq)8EWK1y=NM#w{+(c_l(ph3R
zlc?!{tx3kr3XJPfipm;IDMdan2w4mJ2Yn9czlpgJ3BK>(c^>^?LaRApWo3;Z2*@(S
zWKvM8`-EY@(@%eeqoY3WeeVZoHXFE(VLToa$1y+pqd&s0{_6k6a5Tbdjpw?ot*&5<
zc}GUn@90qH3J-_s4GgjYV!Dl0IhlmmS;BK2{J=*kN!cG`ONHZxIDUZdRuMpIp7$06
z5wbAIvOpF&K^T%vGn%1?Zyls@&;%eAN8^NUELdIidHkU@u5FFTl}4rkH(%f9^s#lC
zk<XRk4Yt!MsTh!#J~Ed`Sx^=^RvJv;V2Yfu(Zy(kDRW{WkjVsBNt7EA#Wj=?Op}s@
zMo2AkL6%q%GEfb!3=Y{Gju}l8`iANQukAb|;I8ZC_7AK@=_=xC{*Aj*2Hu=D00N^y
zXO{ilgk&a>rQpf0eTgi~`BOjh=Qw}g1$^P6Rmlh5|9<{&fB*0EcmDIgLz?C9%50<%
zE&>N@HU3@9a|zMgFsD?Aw_%R!;Rg|F&Ri8aei)KZGa8|XXC2gC%n!#Y-PmDy(c{re
ztK8Tg-^KiZ)5jhrihQmPH`z`nbLJj0mnd0%rtW4Q)H_&h&}C6EpH1+T#JC|*TtiB}
z1M>-^X~N(R^KU(iDs0u7oDcop;r`|yRnx3h>qVe*!~WiF=J3>0U#7@&e(I-whBIgG
z$8iNH#rvLkAODxX_kZT^{=L6TmcKj9hc=;(Ws1S%BSzUCBAd`wg3PpWLYI;2qGXPj
z7c`Z@1Tlr{q3Z!cYC^5CLCBY_;gKxiu{>vr<=HTxIh~;#*i?qI7L4_gWx%-f*%=A^
zW=^}UIU2*>o<$`w;l!oZnIPLKN7ok!>Mqk_N}5?Zk*1x?s`{!bt*R{w1TI7oL9dP%
z2*SgC9NWYi%c!tqg`p6FdKl1cxj0H;h0jPg*&a%^`UCoD!Gx;4F6_5qPQ@EQyy@P5
zx5wlg6U^EQbYbZqW@M>COTmegHO}9El3Sa59335^l}5`FV>Ih0PgkiEWy$i&3ZwCu
zqPTZabdGhafrg?aOZPd|vDCF-dpsa+x-_CXy4?a(=_|Db<5YGuzFXyJ$~?z70yppp
zMO;<&w4&2(^J@Q)z^@*BwFc|B2qQrz)PfvJ%VKWW-&06eqS62n8$!{f8Rm=vO)||%
z^_UI;zT*@4KGWG0S)?c<aa|wR56P08{hfUlS}lA8Wl;bw4i=+y^>7k|gn>tGnBsK<
zQm4-DXhuIxN%NBQcKVBb^WH81e=f7Iwu*W)cU{rne4WpKpAdMShv)ey>5!)e-z(YO
z-Qvk7H4i`X7)y)G;J7HMSXo-(10Q%Fue|g+_nkXWx7{I%B9i3ZnGrjPDyP<JoOFcI
z5@8!yH$;dALQ8NgMiy9^L8$--i@=RSq*PTG#PLA*U_7iU$kK#3s#q^}V_XY`G-%^d
z&Pt|*WFe051DCaxCgZ(6S)uVu!S3dW`!22{VvqgJLy#GnElFkpHnIq1$Yn}dl(??N
zlp4{hqvjRHX6O*=6wimicMzUOnw9iN1&y^HOQD06il~Sfj0=vk3FBExmgguBLLtaC
zlx2Z38fEM^WxU*Om*3qi(c0j+l0XAYQQES!xJ+EDVXa^`t+eE|T8o=EucDOV(MR7)
zueXTnKv5Lb>vg{Odq2c){q}F8sVcs%>#?@Fij<Nh`R+7!@2vf*!Ds<Rffyg*3WW=J
zp+Kl9MN&CIyF!o;Cxo?_#>#RP{EPxwqRJAZ=2|BU3X4&SG|%xwK&|G_qosUI=CgRF
z#s19!*M|$}_7NX?sLuHfL2&*eU-<niwBiLaIl=cGjJL(Ln{`}Ipml=PrZT3SOfkmd
z3PBu3v^q;TT2UqwJfm@BiIIv@RBxiT4#R1|Mz2T5v81NPP7?CM{x-ChS+s_|ae_4<
z9D%{`*6seSKeKQ9v$x(N?47ULz-%hHaeZ&@dE=4%V#1&PQ-6-f9(^2yL1_4sfBaAI
znNR;N|L))a8tUCJveOo;1r#M>e1!8B=H;BZBLu~8Lhv1!D+?NnowuJ|5fFb9^CtT@
zhg=(UQSBq%e<|kvW8j}V&zC=cjb^k!T2ApifpNFEezSph7jq>`lD9B#c9#)aBhzUm
z09PgUHgl)4<hjkf?O3vJ!Th$K>YFe1w^ID(AAhG`VeT;Be}j3F@>4(cGd%Rr!+^!p
z@KZneAM@GY{}lhfU;B0RyK$<MnsdrfmJ)9^qOesy;v{qsYQkcsKp0AE@J3V8W*xn}
z2=mq^0-RD}w54txjthZ<<+PG46&X_}B=rR`hPLaWGLQC&8asOlSFWTSb4uLikn<ZS
zxxRfw9@WswvUM|Hc;FJd9)ryuj+m0m8P;lgrKC2Up_Rs3@cfX$`U=io8$#H&n*DZ&
z5I(|Kgf20<YBeuL4NPq5`V9`!fS0z0jNCfyuETbEh^)ke?+$YdX`<NLI;xC(g~!Im
zGEo$>wRH<)HAhDWIG)G)#yYod+#n1B78e)U-oAGgdSm7%R%J*B18lR2?^;$m9+(NE
zqZv^g5=RkPk`ja=VWTnkEYmm$O3%j=E=m?8$qb{b;fb<L>BSy}bmo(>3eWXvv>hh>
zf|(BKxP9J##^q?7!D5SRm$zxvYAiGjp6fAmh8!MwL{We!G`=4)%L=yl4mmuUVGuMM
z9!?Np@)?n<8D%pj(<zOzM2o6Dtdt?i3qpSpn-#cWm(teR8jg@bNNhFZthjT-ws(Dn
z%9`Q_zW>8mYuVb`VsmSg$#i;W;`&=LcRhzN3W?`lm7e2L7Me88I66Ausjq*HhaY~F
zjrC&)>!PIL!HW;FdGi)|mJmiUjYi|19_mb4V%-=k6?QU2wS7FV1D?Q|3^SWzm8^8)
zNd|$BXx6KivaN<`C?&2Jpsa^B4o>Kj=NZM+5{Cgx-8xR-;nll%yDp8D2I+Xp?#!hj
z`W#>G5Vz{o7aHtc+aaol6e_`SQ`!qDlTnREE5eu=Mwh5EB`qXM1{e<l*GJS_$Wmj<
zoKQIUt^;n3R2uL@T-Rfk&RFmx>b^&ALJmeQS8ray^qMRPhs{#q2$xJ53T^NMAD!#(
zI<(JHbzzG9_SC9iZ-1A?#U+}JCY83^;J7}rRBUbEq$o-rd+Z5TSJ$wm1RPGEI>oV#
zb*^2%QFXekB?to6R#&-x<N9}1seNbd*Xa;yHOkopJwC$e2+TY-#3+SQ3eR=%!hlF?
zjM8MXz;iuZ;h^RphDxVp$YjB6I>9JeMMdQW-K8~3s|nqL6Ws>2V;x*)kL9Hn4_!J*
zI@{;Ihc41Rw#pMvJcb*%Xp{5uORw?kzx6C+>vTOAUzRv7WLd^!oS==N6V+H)TElO(
zQDsgs9TUePCeJalK)EGW2oNEo@svEzsdZP`9r%3ul^y&OD_F;8XLFZQRTFl+OK#s#
zZM1o&xqfYjvNROA<e&ZX{|Wo&|00h*{2106>h(H*>u>!nKK<FxaWoj-`xh~tVQY2d
zY=U_!bFI*-YI%5JKsaZf%K~@KT;H~sum*#nPzA|bn75YJkasbUkF{`|1C|$?Tzv2t
z`RtIhmmZ+AzRY9qxr7^d7*p`dE0_5XpL`aPZQNn*zKOY|9o1+rt>LxW=(41oPAcYQ
zj*$}UmX%ofF6NzYVji%wxkssQ$G5$EdNyM%d9L2X{Fi_EpYdn@%%9`ZrANRT>a`kw
z<LCbdzyG-}&>!A==I+EW)0$d3Bbg4-vBI?yQDg|m!7DA3z{SK7s!^YB?j=H1u|AH4
zRZ&fOht_aLD`c9G$&vtvW|_0uTj88B)V-3hp%`~2toh3f^(H%$KEiEMG>#~|J#<u(
zB?^_)NOer8V;UG7GehJ#k!xvWIcAz7mBtHWUhXa-o6C3`=V(kahCPqc7x?iyn0-8d
z9V@m-GJ~)s)O+YW;>*_u4BKtA8*p@Wn`x5I_pEo9d0A+-ZXF^^!-*4$#l<DyB%7Nz
zfs%vW9h$W|Q5=&bDcx?DgM$O4l;7!(|L$Hxk>Ugac~MY|rxZm%-FK<g8)$2=iYn@<
z=HkZ@N-A_&R-TZqi*=lNkh?^c7G)G!T16mQ1+lVtPM4(61kRLJ*CAZ&;)oHo*k^UM
ziz%j@JhnmHX>sZN2^?3Ttz!S+fR|q1r7Rnj#&eLdZ!@%3?C+;o4ZVe!Q|qgQaY&x!
zj1ER9B?yDKs&;w;D=gZ2OtJ#EH0Vswmm!z?GiXF83zPjZ+VHm5xZ^k+KYoJKXV2og
zF6ZvQk6Sly^4p*M<hSME2!}YT(Wuv{)v5?FM>uGsNM<Qgjwz+)>rZ_R-fKCwejKSZ
ze&F$*_ddoqo_>*~xJIklVt@bMwcTx&BBBUW=8z}I(5K82ytody)zmT;5&Bpapp?WE
z`JE5~t99jDYfLprCmf{Klvd*VF2c2#z$0jeG_~NN`_>5WuX8jQ6U9D`qev$y$IqUk
z(=c3maIrG=mO0xuH<{epr`|ly!pbsBMM0zCG8hyb9SxwqOyEY;TV1BL21TABXETB*
zMk_&5C<4d9^*xG0gHe=uP7p28&wO6oPsy8Ygm4)SCTQ!DN=T)~k3+J&n9HKx*)kUL
zyqvo)SPl;M0ZX&doDb}P=hfKV*`h2oWm)pR_kDo1)pc@{6NVx0d*TURdiiC-Fv4@I
zBU-&yqqnfY!Qs(&d=2mDP`CfmHwZf|%4|lx$Z)7tAASqo+)>`Q7RPmQYY}7xB3C%E
zhX@?JdQ970B2QBCWQL!4c)~$yOIqaU($K%Y#f3``5PSP{iUT6?0F#3Qve_Pwz5g^9
z9{B*)F_cnr{pMv}e)T$=H+MNY8q-}qiJgW7JC?e83v1!<uuqm11cA@W(hBuv2jTma
zMMg3h&}h^!MP40B%N(gB!tu!xgApEX)Z&GmP0V70_Hu`(Km9d~G$35C?{0=$CAx(o
zw`^|iA*H00CBOJT{}24tpZj?(K5z+RG><;|C_nMzKh8h<rGIhH+i{04KSR{{K4dn-
zdn<F-1+QW*z6Eo)9@BD{DCW$)Z(?rgU*F=sOAipvnTO&W(}P3u*#VEd?<Ds>^ggUG
zlv1&~d6idQy}{Pz9)tdrh2>M2yO<jbqdUw!mX}tjH9H8;r<4Wha6r9L!xlMONo-j;
zdsoaY`fZtiea_tZBk})zJLdo5m;NXI(qH_mT)6)t#%L}*^e{jAV?V;b_?N$Y&;R~2
zuf9h2{tLKy&bTblF?gm#m@1M>2u)@|3yX~doV0*h262S-1B~Z!&F#^h%{VsJOmL}-
zO6!y=i7pMxH+E6?onhY}az-YkR#OhwSm<sr?&OU0l!3|cJ3bA!!$Q*`wsl(d1*{u!
zaBGa&5;Vh2>c(*2q)(m|ReDUP$IWI7Pb}d*xIsMLXQDzn14-6xVO0Y+d>D9!vV<aV
z+4ma++kJGWNz`rd^7EIOP0~9$x_5gS3uS58+dV=ljZ!7;_9C5jhtYV50IuJ-&hcX>
zNz#lk3~9BR^!x8jME$MScz1_pR3j^L;zE%m2~jhyG|!HQbuFIbP!>hiY64tawIYS@
zVFHhu?@=Zh#%Sub2EtaF);zDO^|R@iT0NrOESY9UM9y)vOi3sS8a@lhPh+h{Yt7O4
zko|)J{XxRP(U@@(l20UJXbH+AWX+;gLc0d-R?OPk616zSx(;$SqsV3)Kemh)c@#$D
z8z{BH7?;HOgi4{U$KGUu+lq;59@k#GSq<6X0^e*O68b*p&YZ38b-a<4`<*@^p6e0@
z0rgs)dc95%gxLA?uvVIMHY3Y&%Cg`qU;Q#4eE)}7U0ES73YL}@>2}*_Eon6B_k{KB
zHMw<-TC;|dIlhrtBglsXYOagpH*h=`KL~IF7i&VC)Pgctr?LjKp^vmRA|Y_|lCaR=
zL?kLBEel-FMdzCOf{(Bxoarvpqhq}G(k}QV-}8MBu(GjCp__Piz~1gIyWiNMEDDNL
zvAA*qr;LyT#k6>x-E_e8pbu1DiS0%cFD}4!De??8o6>DIC`yHO1&;49n@#{hDM1^7
z^<r-H2hgrllY&=YyGrQ07~zns5<d#i`i%joKghBcib7U05Np`q-$Nj%*Be-C3Bwpi
z1Z-_>qm<+;U;aEF{NRUKSzaa23l@3{+;{FQ&wS%q7JEy$u7?nUUa$9E9qOES@?x)i
z`gcjMT_K3-#J-2c!?6O-4RC!Q>k5qLVnx-f_o9fPQ3Gqq##3gaF=bKUxh}0niy){W
zwS(0z$y8#o9L^kON)R?R$IKpfH09!b4=~<r@Ri?vjqR;n_ID5Y-0wZbfB9eh3cvBM
zH+lW3nAz?+rxqXJJ!c=Ln_Z^2WJvOqqoW}%f|bP%OG`_YKOBOz%n_qLaU4@-Q?wG4
zWsWW+p>MFpA@Cx!4CrfzmGdWg`fD#$(PplTYu~*J-5bkdv1n_Mg<)^E&y~wpxO(FX
z|MY+UCv4x^MClUG@%Zz9_Gf7}8u$DbbnR2WM}F-pepDy&J-E%>doy$1${b5Ro|25l
z$fCe?U7C#+-ZwEf9L`*lIr9y(hZ)VdaNqrmHyb?p=_~AP?{c_z#1}vF4FC22{onCh
zzp}%Xr|Zmi*E!L<kH^nENH@F8!jd7qoB2|Qg{395HE1D7WsaEi3F8=*%`jR}$^u;~
zBHv<-An+p8TbS#-K)4R>yL^bhd%czUwVPM@rGN2@>}+kLw8VEj{_LOrGc@n<Mh`If
z!e`mIc@r7LY<eC>I~b+0t_Pln_5?!AE0)5+#t}M>aI7UvQo`vBw=5Y6hi}vxe8Z2B
zMxeFBbS#m|GMEV{U^9|rc0zYzXmt~cgB~q)hIUYADQvRbIfgx0!a00^eD4IeUiP^1
z%1!VmtcWA(3l3$Pkd6mca!I4X?{*i+JVV)ru$2?m28@D$rq@GwK8|xg&iOV$02qfb
zh)~LBY#bI&ZSd-=H>(PYh2W2nxv>^0Ek}nVZr;4baJ0|jQk`a_K`9MIkujMJ2)t^O
z?{vEN?C;)t{Wa3V1BB-hc$EM>Ns}t_%JZ;}zzBgC1cddNpi#%K*T8X5MTyD^j8q`N
z^L#8GQiC;BTP}zKgc~4Shg!>L*)LI(DXUA%6qAVUn`6pSA+_P^jeY+8uRX`7e{YYk
zePzb=*IjB(kJC%*EPF?s+JMEjpj~%a>elIX+jzobJRYM<g>?;0Pf>3+(SWc{)jC(8
zfzk;n3PYh>5*%s^EpBb@;X6Kl5F);*=_YU;L9^Lpb#3*|cdfPbkB-ph+usX(->+h+
zf`Blr@<GF>iW3u#gVvBHIY&o*`hz3B{NxuI4u;55q7~eC?i>P$<M^KO2H1V!S&IE#
zyr8NGS$uS<C}&A^z0c>tgmCa&A2*62im(XeEJG(LvQQY;!*A9ITtS&iB0pv}63oU0
zgQ;PfYZ5tQcKtGkSI6|4>-2B6*m`Y-kdm^$#my_X`0fAivwZb;hFpKvp%^Z(*jlCS
z*GV>C#hQX?KW94D)clZ@Zi{-uM~K_^#~eRPh-<ZKe~cm~;~Bav2tA81)lpq5kBNrv
zv1P7Z+d`Ms)Uq&QKIrFE4~M)lB=`qe)`FCZEX&D@oW1=$iaftl5b5<6NoEDRyZh`P
z?(*d?e~$5ZjL`-qHIF{>C|WD}{i8eWZrA<pYH;tUQYWw7;<exX4bFZ1hmn3Y=pc}|
zz6UOd(4|Z=8g7iShIBk-yx(Wk?^9+)C8~5?!g@^9Y>{fiV3ZP?DpA0bib1Ivj3@M>
zWfr2>*m<eN$&da7!;8mx?WL=ny6*}_;qZIEafP5bhv$Z9?Q!PZB4<t><MO}%U3%tg
zbUJlz-P}WK$zrd`+WH28=P?)$35|d}BTxlG8)ma9Ap$8ST9wpdf%PHqLi$-w@6-lY
zuis+7KcwS&OpPIV7fUvGdyw6ZaxrsJ?%}}@-w$wc`CtC!FY-5k{%_+oT^@e;VJ=?0
zz>{D5`aOLu^4d*a|J~o<>|M<LZ^JwrPnqub8TSXsEU)x7u1ipl37ajlw`Hyvjwf`Z
zWjb@_CqD8Ch8H(@{iSQ1ID3Pl5Par0uM?CNbEQ4bpX+hv<Qi9h?K3RQnQz|3e0i;6
zJ{}JVtUwfRFi&SQ0wGYcGF7g{ybbeB_6I{cu3Itxql$ExuQ!;F@ckWJ;qtHk^*`sY
z{?)%hv(e_#rAu76aDlIV{i%EU_m<agGX9-kXX8hHlw~g<$fp>YLs+W}hkXYxFOgn=
z4=kn`#}4|O><{savYPz#TtpPI7uTp8!DL$CN{v<mS8B$E#-Ge6{Y7@eYutbI0R6LD
zn2jd8*J8ZII@5`zu6wvylkvD@G|g$9n$hV+v|fVc=y`hGI^)+~A#^NJBj)MFWzx{)
z`^th_(Ug&wp=_m@UG}FOj2uMw63+1&VF(0(A3$Ev*;wV3TU!j01V<`_v1H#?Nq6s7
z<$+cWQfO@$3?_J<ha((%-6naVk*Z)Y>=OntSg6-)_`ZKnTCj33VCS`$SiEo^>-$Kp
z$x2C07I<NR2wVgL-=EuW6ea02C7I4B(hOCWm6p=?2&0&yFpP&Ou2z&%;X8&>2$aqd
z&I0wI&tU5c-3K3Mwoqqxb3j<%X1`zX^{;GFP8<T)A*eZ=IMv|Pu@2MMuCv<fv$3+k
zXf(ue1b*P+x=v+P<yO`#Zrr4E>KtC4FjIy=8>Cx_-ctci6e3H*tW?C^4u=Ov%(8;}
zPn~C{zt538s!o4*mc>D^*y~ZRzwr>Im7?G8W4}$CP6&bPI+ZNR^YQ(l8kSIMJ~-jx
zI09`9*)*%PYQU3EJ<0cd==<;;k4~pUr!yuD!+YvaX$}tAdge)vz4tv>*QYELr79@%
z99NZiVL&;a5Y=ipMxnBj>A?~Cc!Ep|d}~RqL(z<>*Xv}3=4dCS*6M<S(QeADE%0L(
zw-D5wKK+|FSa|Fa*3T?*v|q4$`6k75z-zA-#H?c+NtVO%dYfik=e5s2P1C(aJ#KNh
zmk@?Njk-^>9^<(sMVh0vL8y|t1421iqsj7|Tst^M5(fe$ERGe-3Q1$B%l6?ulPse-
zcQVMV!4Cq`JjYmjr&jxid8e%fB^9M35D517_gLT9K)60}T%%U2Go6lcy&*!lJo)68
z`0$56f)F8#OUo=REwQz=jqAE}y9?j-i+M+fI>%_XpMH+8+2UMJ@#I&ZW9?Xn-qIp^
ze?U5((()a~s$|q3kjzp%Cm;?(;#QR-r;R2{GRFNO)^q7JB4m-X+-$J7ubC=IO{Q2O
zsYf}-^XItvDTgypd_S6HE<ay#<&`63Qm5N=h#LWqz5gst>u}>ge3r)1=XuZh7D+K6
z%O<p1HP+WRsMXuZB4?ImbghGvqy(~HcyL5k@6uKhX*60D#1YuY#rHz4jYj0n9yedT
zPR()AXr{)X-o^X75aP~%cfLn8)hDH8fA@$u_WAm=U*$8O`V>F%i60}b#eC!=AK`0X
zf9jrUzV2pzrl)xNtIx4|tV4IsJR47G`VLc7GU*RUXDO}|5QQO8tA#ZNV>Ee^G5uD|
z^=;;gILbMeKgZ^$1!vy-y=c3{wHHdRygImxxzD5DbBcy_xb>S~pmFqh9zEY8zl-_m
z`UcUQd6H$cjDwS=__APfa713~(vS*i3`&=Gn0sExjnRm_xya4eu2B;%8pG5Y-p$%<
z#T)K^Hmp)Oc!PP(=c(tP<a57&k&l1u6V&2}AN+w2^Yy2my0;NxN^<>kUnFdI5vvX#
zeCh?<rlh~RNLw4?X~IG1qsxNFMq|P(XNJqz56Gh!VGXWQoJdkG^@r6>w^=8d8M+G|
zhX;mnDLErE>PB)9TTYkPS%38b4j#CL``|L8-7dpx4qh`ri7}WNi)SXZ8X=9>BbLRh
z92~D9_O8)11M1C)YuzPYs5e;=iu(%1;lw4%5^`B^;qZvXqJ~{OO?N54t4qSrAR>oG
z;1Ug|6w5Vk-PpuYn%pQ7YpFc--<@^)8h2d>$8qK+;?+~RR+j$Zh$!-*2d!>{qr;rM
zDDWH?t*eSy6h-&sP^XPyw7E?XN3@%on^!jJE(9ngh@%>&)HI`j)GE^HjM*ebN<-lL
z#BoH_4#8TI@s!=&1Kc2_-ELrHLC5!*q!ur*EG^U#o<OLCroPF+3zlg8Bxzajr7z7m
z*qsqMKF3yFn$4IKXL^Jfvdsh99B|@zhqRb7O-6XGAdcG9>J2<=$(6<CCGld1)`dqH
z-?+kl|A^jfjNj@YjUo4;;X4T7aFA!{$mMYF5Jy|C-Mqq3mgu)Vq&X}tE#tUvG`G?$
zWjG%2b|VsZewX9mxDJjh5Kc8%XQ?!4zVBBCD0ABzVO`Q$j_*&nwRMx%UwfU0E?pvu
z0*)P9=Nr#Hk0S*79$oCkEal+k*Kr(|UX*d`_3PZf+@{rSqs9r2<I?suN7rwm@(d*_
z=L!cwEea{iDyq#L4k^bI9M55?9+2c?Ru&i7K7@loPLBbt?jrSi!Etws@r?$j&OX3!
zFyrXjnAzbNk+}F)v%c2h)B{Vj+>l$Jd5K{75@#P<!q|j3v_wV_2|?&v0-+gaGpqpL
zC~B_5_;5rTHj&dQPA9;~0^hf|jzL(#embL!8yxH&QgZ~>SaJ)cHE61b{K}+>d-5=6
zEtI9gaU7B?VKf}k>-BJ4hfb%%WHMngN{B*_>$k43{@Ml?AAAVU^*DRx95-&<U^1;z
zD8AcD?K_)(#ivLOS3duB*1!KW8>@?4y|T?$p4n!p5%S>41yWVuyOyP1j}vE3;X6J?
z8}ckg%CZ`x24r3`9spuNoxYD4&nCFO$H7$bwF3v~SUf{DPL0n>yN~F9fvcZ6qTV~t
zvBxfP->Fk1S&l2<)n_)Deq)pA(X%Y1U+4Tei6bDHO=&b-)>l^OG<!%X$nyzBQR3Lj
zC+hI(Eq0!JnRCa^P|7M9AxjE;VTfawOiI#LlgqDNqsSGwKGwnv!@QXLt>3KM?gf<_
z!gqaRa2yB%w;G$D3wDh)+_p%n%DR^Qy*_b0;@|zoukhGok8|qeS-$T>-^>3q`0ws%
z`19S&SHJ%>YpaV~zp}$u=gcphTp*t_@AZ0YoH>Pe7jtzNbCs7&$0H_pGe2~Y!r~in
zbP4z@w)=?w7rFlXeHy*{S%35b=T09dNlRQ|dF`1UrqAv$J9?IG`V{w_tD<|-IrG((
z6<TxV<(#=YXWqYhi{0m5;q1m4q*9bxQKWA$&!r@5HM#uS)rz?jV69|USq*b<F7-Dr
zM}YrL%<Y|wWNU99=KI98h+q57f6F6}Jj%)Ar}^L?`w&6!3-^?{wl3*7=hff&3@e{_
zoZ;#c_r3BW`Q<@1#dxgCMqZ$;;^}UO&67=XT&z;Kvjm%_7|$ULd=?|g`Pq!KSxV@-
z><mku?>k5=9Su`05z^&_#)!w{)2u)JAP0+0;tPU><q<}vOvVat)?%T*PGe`EZtwzT
zJ?FxBL@nFJ>-db?UA|Ipb51xspi?Ln+C;2m8LPK;SbpU)2OAI4+=zMnVeq^h-)gMy
zQfNioZ1C*Wn`F{5D=Z~0GxS_j_nnzXQB>_~QqH-3`y~VfzDpQ}1fGxMRDPFIsmdNm
zN|dr39gT?MkXj6Jw?jT0po}7!WwbjB)M_!)>GTiHTv-P#Eu(9jtl!^dxzpn4sL!pP
zgtbnKjfFaghb2j&FgUc^HP(+U5QG6zmgKV;d6E%@9;Z*AzzE2-#MnBM%&<QL<48h3
zAP78+5Y*!-<DD0{eDjcrzs&0KCGJ1bCJbF_VU6K_g4$0o**0;x#qxr!w6AG`<CHA4
zm+7__5pICdGlceV$`XuXJeskyH(+K<qRE6-t%;Nd$FKHnB@IOwa(HlvR2qa=W%U}3
zv2WjWYc`u$YjG;$Nrs~Vvsv<OAE+F0+uIQ4Gm->Bgw+<U1cC3}Ioum#u+}n}q`1Dv
zb1!{^<HwH?20>Lg(VAAX#dtco6W{oUSynrgiDdiv*XTaH#_Gljv;LHwg99Sp;n;E=
z6+6V9i)yw|^8t^@6O>twOceE6%t|xDcLYU}k)$(xCu4swWOEV|G#acf)QH+Xf*5Bq
zrR+b){^xGeSX|)5LmN0J+f0%KSt@pJ=5TF7CJ$KURTl2kG@3rU`vXim!x4(Q*CcQZ
zWicbq3Va7#4CDO)n>Y9H>Rl)bJS8c!5-AIemXu{lu@JJoy-!(cwBw?H+?u&AubdY|
zs)P4;`wk27Mn{d?_xjr(ig_xKH3l$@#{=4}CZQYCX}8$+T&yvS#|dE+@Qr7m<oJn`
zwA&rdo;}N_KK&`QRt$!Hmc5nl>h-+Cm@t7f7BS7(y1K~^zW;t2jXK%aUtnz^W_7hw
zIVo#3pNJ9^RQYmVvxZH;sA|*GKGb_{;%*Zq74eeCuV1}M$AePkAQes+V6A1No6_)}
z<JQp?2EY9@si|QdAIAvR<dE3jVy%71>e)ILOHw36QAFtV=yey-N|9uJrn4DRc^D-b
z98P%Q8?W--v!`jd7fI8MROXb^363nNMS=q**+X-1Yai+Ph^S6J9Z}q-_brwq1np*{
zI)IfWg_Nl8;v#fiL95xI(P-d%KEev5EGdeTtjH;4dFR5PP74kX`UnfZ`P;wF&;5nJ
z#(ihcvbeO!_SU^KrhO~(?|=XO)Ejm3IrHV!4#II^&fL6(Id5Se_u7PW=E0K3;kBD|
z0+4TEzS>Qxd(W|TbcNw3pLrAWRXL<)Z?V!oWM$4goih)-9^LK|N-46ePcoaKq=#0L
zVSma?-*}bB&Yq&#?vdv?xhjy;363nOMS{aS%n#<wYvj`r<(*CW?Uw-B^G#otl0r)K
zccI6sv^w?gf;n$u{>k6{P5$D~{v2mdpJ8#a$M*K_y<=`G=JH_3(Q_~I(uW_VZq9P{
zg;yzBF<<L7C_JA;m1s-l@1}A}sWH~#AqapvmWC(DT6LaR5sp?AEsxRmHZ31~RnkzJ
z1j~Ui{C+#<0rxr94zDrUJHyax(P>0<@z7$QmfYdM*<;*_SV%?$@`zR|WTFJW-|EoO
zhKI^A$Mce+5HvH%#$ieJ;uQ`y*4dgxJbKpQzGg{~=4=)jfp5r4Mc>iv?Dm-nmm&zs
zlaf;KZKqxZP>W-H&*RLQGdPY*l4Xp>Blh?AzvYeBaRiNeOtaafUaJ%NQT4e|iY(7a
zk_2llN@>!hWY8Z$&(dkGu!n<GmOL+LYE2y78=aLkfEBofWZWNc|FI=%wSdj7eF6^>
zZ8^vR?QnXv!{S1%%7QhLG*7Wc(`YrSy+DBQJUnYC^8$Pr?+%E)>H!sNz;$r^fLcq^
zNp8~6dw9ED?4X7YBcxlSrV}D(Ogm0!blpnJrDbIelGo{W7V+JHLgo~uLg6Ee0+SUS
zZSK+S_BhzR$-({}ac_yTEUGGuE=lp2U>FalSjR(qK4q@nW;~)=%4nW?`e_ah57}5>
zquFe*x4nZbzP+H{Qh9Y&#>ftF9212RjulvKNSt|<R*1?K!a`9fX43>=;l-C<qVxU^
z2M32-xq6i-j>w9fJkS5o%ze*81CFu`k4Bt7zfP-BM_t~c)rzUL>v)k%Gxkshk}Svb
z9HP)CXx5nQk8zB~ps<cd*s9?-B9tugrOVTYM+9S+qFyIt0mlnyQX;)Q+O_L=MT6oS
z%VajhSb@-nlS1K)2GqhSjYf!^LgxE~Dn;c(){ZSviyX>yz+^T<OG{u3DpPD--C|*-
zOKN?}qC}Sk*>pxWO=vbkrq<F|is8`+B?QI~kVT5bR@2+YR_7NXFj~E7w~xZA)2OsY
z-*sIFzK`oxgEP1D#;vvUUt<Y;k0^@pT(>&FJAyQs(WqB@eytXhX9Y^aY?>kvy!7I8
zy#E8=%kt84)uNqE02oh3-xb$)bf{Bz%zIpR_D7VZA__DYHfm@yW%Fjq;@T>`<rUQ2
z5m{0BR#{{5f)J}KAE{(IVQ-#<V2mcuB^MSwCSy%rNRo8Sdaq5Af;5&!OLCz(Auomu
zlgj+oG0^B)>Wu)$wopc*tfbj&VWw03dXv53jNRQnheJs^GY|(H%QA}OkoBcH+lO0>
zb`M!Ca`Ga@WD|;_!1pYh1l-u%W0osubt#QzY7|;j?a;T|?u8I6_7?f4|HJ<Xz-NE|
zv;5BQ{SL2Rz4C3NeS{Fy;yMf69*av$m7kIDNYj*AGUMQAACU|4JjYmrV|~WMlv*6}
z>Cb$UAOFNpu(-O$v11$e^icN(bC13K@g3$59IK)Al%1O;3u~)%msikQ6MP%y>2$);
zUCfJIa-sJI^DG^++-sB0nb%s9`<fGqV#qkjDI^F3jYUhX{$}R&W{YGx#j7_t7*5&S
z9dI-(?=WA>Gh}ke+ESg}!!1TThpd)4MVVo;Nj1RZSvJL-c_E?s26M%mt)kz$7JG~Q
z_y6c0VX=JSb6?<7pZ*k=uU`F*BI-D<y#wan9p=CP+27?yKlY<6tgN!Jv2jlib<$cQ
z03*n6?BXAuVB~3Z42@*S6W0n}?k#ZGUBETE%7AqPoT$nP6?1EwzR@(2gvSr|iHspr
zib*;p?0Sr*hJG$NlTP{Ca*LjUHAnDjQ?gqd5NAUkNHl&DGBOTj%`%QXd;`adKGH21
zbsH?trX*3!BeR^vgCR17S0{>@fc)4t#W0}W3TPf}lByYX)+xp{3eQ%|P01La&Fw=*
zg+T@}nJt+ajhY9vf2-AKG&ytTG|hU0bLY=7I2y8laDeB#93CBh%R^lhMzmX9R+g9P
zbh^~zI!Y<hH09uEA8ia$Nt9Ax9j4QaC<^cx5!D(LSq2D-A}0ufd%|2%O}Dx@OeR@n
zso@w_TLFjT39s*s>9*QjJh8|^GscgqAqlM%&F%uO>tKwTCnrdT{XRvWqKg8fVYTL<
zOr<58Ovlt3O{{QeG-FmviSi8EPmrF%u`>c)u+R{+8X=BoVNpm`;<y1)O5$3ZR2n9O
z0kh1Iqy;7niIpZWlGumww9jsTK=)WmAxn&|9*{~c$^3}HXhN<uPFO=4O<^@gy!~}<
zfrFz%l$5;xeeXj!f_AgR!onhl{cp|M%q2$%@LZobswQAv;Z_XGlJR6Ruhdriq7V*w
zCJFq4*Dt@qgBKp;*s<fRtSqy;yEm`;{NYzt<QU(_4`N2sq8bzn1;>^nN?no+N7!DA
z_R1ojEb&EkD9`8B%4V-aS!kR{R336kQkgMpgClbuSaO+8B|;t{@|38zgq0G}5@;o;
zc?p8+XyKCN3Sl4&Ey4j=I^=0aiDo%!<Cu)ba)a?K;Z{E(i@W57L21d!a}t#d5P{(8
z%}vt0PUn~;%reS+Ql){&lwo1mI~bA{5_}(tA-4wO`kXv<nj%+>rc;Ka5wqEh+t+@*
zR_D+C?9XDY<?`#7dGX~JIXpVT^E^(SIz=4E)t)__k)#PqX;2c^acI=*G+Rx=FvNE~
ztkp2LgbCxgI;-djU?~er8J4{I>I*#d(4)k)I^AxU(P)GtG}GziyW;xJ4s`&+VP+g=
z6UEUmVWl0har`7jmf|=bswgQ-g$P{Iyr}*Go`>tXI8lTn1hOpA6N~S-lx0CWOGuM~
z2V%H33F%KXp?AbWuS*<v5fl_ffiV#Hf~8JWNms2xSc}mP_$7hokxmrbdwmWE8Pm)%
zOB6;5YJtN_&*S9DWon*gf10y-y-zcCIJI$-Qc7GLv@qmSP&y8K%451YsKm9#Ak%~r
zTTKLgi@o0WJbvhhKE@Ay;u8cxz$bqE$N2cCKhB^0>7Qmgn|*5#uE)~i66@<5tgft5
ztJUxvpEOO`+1a^MvC-O~ROJbjWs>n^itlXm<k!B;kNw0S=j4eKJo(joH!Sf6bHQw)
z=noT?+Yu|r-e9ipVqO*n7>hG!?nY4+*HV_~33!f+nlsOma?U)s!@S+=61@%c-rF!2
zbLQDZvAfsj=xvxU_dJfDTqg1ihtr&`IrHPkP9del!9fW_CM*5Ro}4pB+{K)`!vNlv
z`A0tbQGVbfKS&S+eC&rm#t(hrWBlx&{n>BJ-1q))%)kEBSNZsl{v;=kALnadyLZD9
z8VkZDvo6IXXL;*@m)BeDojJwQaLm3E^d=c`+n{}qK&nbBB)|xPi9)OpIC+juX81zj
zYK@y`)N;xB7Tl5n(^RrD9J095V$@uqVc=p>5(1mC<}iX4TQXLX+*lkOg4nPd_?(<N
zG`5E9Omfm3<fP!16u3eV)&<4d0@n9w=aS*p5basY6X)<sgC9if>nWuYNa1p*Jf^#2
zCPI*xiagIyj=+AKN~`C(Y;0_hrYY@qhb%8xU0Y=|88e=YzNL8Hb6pm?3#_cHa%_Er
zg@s;qsMDImg9A!gR&Ao9K&h(oS>%e@EWvZ2(Oe@>ErA=5=NWOWQDqDNA#d~+2VX#G
z1w|&9rIO`VNEx-rr!HY6XuFy?Ft~wFo>vi%VI1Ll9>R0+9jCJB$TD<MAf&4Nd4$hK
z)8%mDGMMDFYtU@esn=^1c|pBhM=qb><i$tuq8MupayDYNeT%HWPfSs@wv}ZvoROp@
zlTn5nIe<%+<|rrQ*r|0U2--?8n<-oq((bL2D}yI0d4mxyd6_YD9kNl1MBvyeXZJgq
zSO_6FdHOUPCr{#eF2_!u;`F(*{PSP>rEi(;VU5LTJ#RaQ#Bq&!tx>h)EX<OGAn@mL
zQ*Y=jrPSnk!9<MM+1}##@e{n~@yEGx<tnXqi<e({^$*1x!0~*v6=VjoOf#LPEHz^q
z3ya9I#5z7oN{T|_dk!9R-%&uJiNXNffYL~uqQuda33XAF<eB7H2s<-DAwgydLf3JE
z8nKb&%Ak$F6$V#8J@QedLKYfr1j19)8a}b{nG9zf9wqcQ`;?`owzP()1+|((yIrTu
zC8NWXYd5B#6unC;_#X2IPD7GwTn`cvksi$`MK!okNQpwwY<Flk+Y~iJtJkA59`n-6
zFQK(Y2*JgR7rFm|^Ej@{0~arH@zO<p@fUxQdcDr#(lT)z;W{owo^$!i6=t&;LO`S5
zpwsElZnue|7*{wLBgwK9M>y2tm`=M*krh~|(yvO%WHRLDjcZ(d=n;Co-aPBs@NN72
zcdd7HsMDs3?@Ej&C|tI7cj&~c_`XMIBJdm{ug>VGkK;PT^*UiK#*adTSG8k>1y>p3
zS{-L)nKDnQH)3Yf8R>AwVqK7>g2qOJWHiRFcZfT6B3mQRGRnNbQI*!n^;~dVa_zIV
zJLJZ#14fe^DJ|GaOu1AStS!}9U#Sr_15{?v)}Z<$jAyv`*m)XjiyR0`%d0Un3RPN)
zsL5e6VVb14u1_3zq^ZHqwR9GbyS<DY9B>SsMuVUF@t>@m&xNR9XEVgNIt4Os<AfpY
zMvK*@Ro2(mX|-AyZRq#=WLZj<&6rLn2vK6uXf461GV($nzWAlj@zIZcoQ?JMdpaTY
zX66pNbLQS%%w4a}_%7zb-OSy)nb&YumXUc%)QCx^3F&Zphk1RYK{guW)jLFYF_&}Z
z0*f<eUd)-_ymi3%ZJ4hv)mdGs5jF#KRt;<DIr9e|J5PPJ$3R%>PL0GURLLFYvm~pS
zhi_&6W*w!N3nn|wCO`2bKUR4<xh{_9k!LC2NsAQ+A)Pt%jkR@}Z_9l826Ll8s5{KR
z^p!9010VSq>+5Uxg!S#EjEk}k%4n={xOlM7pkc`u8XR_;OtX^Qb9wJ*gbV^+s>PH+
zfC)qJU4%RL^RsZ+Xz{;!9t&B*{o^sqeu5Y!)P*7~1zA+*+;GZQ>Mb^#bq1}NlS$68
zA|q0osRa{yD91%>kN9v*dwa+*E!kIwqO@Q%VF2}|nAu{?tl`rzmeh@@WjhQVi8z0T
zq`5||9Wbo5=#2*?${`IK^wS}ew8VrVzU$&+$Z_ZVJcKZJ4tTejH*0m;jTVFPfO?~j
zHI^U>aHuMP-|CPR1_8}lgQbNfR+d*;URuTve3B%gD00$tRy~<JF2>G7*Qsu}z7J85
zAzTkPvP{Nf>h*hbsB`9Db7K(Fr9Ya|iUiZ#qMVSX2MbM)ERkRn;yA`}ohmuTa}kbP
zW!y?l;QMr!dYH0AYek_n`79%HH4xHn)vAL|?BNAIPNPoRY7n#<2vNQ7LA*rR>M+_r
z<i@jKLycxgSq+fLvPw8<dlEMkNTcbjuW~f*Q{KGJ*`STI8KGxb*jVRAsffIYS>WTk
z0g3gQsDeT%gyZ0Q4sL1Z@4`20XKyo)!;ps_yoB#pw@udpB@4uNP)s{Q-0`Xle4i)`
z5dzAx#P>X0r<yDljtgeqqDGTt8rQYlxOJV6eDtF%F7|llnQzeTcJB#ujf-_$lCs2C
zis4{@t}WvSL1i58I`~dV)*m7w4?hg>q5$hSSf|o|x)zLbh~oxs6yi97s8OTHa?)`^
zJ2NQl(5U;kvczq)aiV}SjTn#8D&5CQTwmY?4q+|A3ZLmTVSjg8*<N)W@}zpmS?q?i
zmja@mPZR}JdsS+5lHr7!bN8*nu?31i5DAZwQ7CB<jSh*Cq(zD6_;|jHXD!O{Xm=Ox
zkak@cYxG<NUhUD3J^bjM7OM~ftt6i7&}p@aqlhpFtN(8ezE?dwc#cCOuG8%-(Cu`n
z)oK`{Nz)mQ1J)P<FQD0I(f5ZKU1|J^T;hm=%a>o}fr}5(X}2p@?;!K{9Uba2Ye}_6
zT0_kj%!-7~{cY|)eIG@s<|FygY&G#3HQdn0bv+zMRKdR%jMhk5l4LnWTB22T(~Rm3
zT;XzzDuc7xZqsVjD3XM@)4}mvYMmA>BTz;$nB;73Pr13f&;G%Lyb#2$pcx9<3l80u
zPp?&@(P|Kc0ZJ?OxA&Ng6VA+1nw@|XCpt7c3)n`Rz^PIZJR32{J<L*@y~7cSvRL1x
zEOLwxEG(>&swuN+@}|m-1p>$A)X9@P@&5PS5l(8OdH(t5k>5ecUa!~bwA-}XZ944^
zolXa>HF^F<izN&rND`P2PoOP%W|&S&u3o*ycsydUxAabszGLP|k+8GB&6&3_Z`{q?
z^KeAPTv)JLqhv{%<&<d^8D?+^>vde=a16mDDX6#GG+H&vIdkUB-8u8&Bxh%P%GS4F
z-f8)C=gj>uKx@U}_8!x5%86M@y<IVHbQYl5#&=x2D8RK5gWSa|wb(lvG1CU?xmA{>
z5G*V#lgSyg>Fk@ro&wk5)QJ;3{=WC%%wt`Q*1Y`U3&`?2GjGqCx4$Lx5RuH@U=DYf
zU%z&p$z)7#Vey_G<cm3T$6A8G!7nqO*xTi+ix-H6#6}(pc+rcQ)I;1>6T&yzfYw%(
z?Yh#eO2vAqsmZGP*RRE39XcAkq-3+z;<;MHv1!V!Fd(VdczL0P)-60;vO3LK+MBU@
z*oVOkB?VGA2rOP`pyM%W)Y)#<m_$Cu9mC_JA#s}XOsAyiI4pESUTJq(?nVT@%S-i`
z)1!#7^ib^%d&4cpRv@iYX|jYvt=_`Z8CqN7TFh`bxWn9c1ii&BNtxhzUUidp9Q-gu
zN=bg(kLqn!Ew0gQHfc5+v|3GCoff|5SJDN)dJqhQpt^QMwfP&MC=AI=(uyTvv_d)?
zkY*WqUJysox2*_&H%kE0Xi5u#<1i}<cKb&J-ZFs|)O<w{I3$@S?z;H(aPD?5&{82x
zSw({71+&QnndkUHfFFlARuIG?3&KGvLsY8~hF*0j2m*Xz5a|Rt8{yVshT|E=#LUu?
zqa(>cEI~4&7YdqnOVsuV{T8)a#I=4-64kH-WJQ9PWW4g#&m(4i){dW|eeNRM+yr3>
zy%_7)DPv2L<rvRH%PJ2RA*j_Fl)9iSif_8kJt0`=b~$nK<el#uYuVY^t&%pr&GLPp
zC<>|9YXpH$tsYl48d)}<es&4MkR-1xO{_7MP~SjNXa@ZwiXua6&1^O!wzYe5s8du^
zpSTbN0n$o_(;=%X$1vJd9VtLW5i)iVQC#gaSRBV8a9msmN}Xa27%PxPHR<U`5p@JB
zRx?ct{6-V)I?RSMv>&3hz>OnnVS^|XlzB;>XpV*{li7%}v<Rheku+=I)LfiKM61&!
zTwXvrf?_%&9ggre6rH6ea@ZhjwaEM!RY**1iGn8Dk1<h$QBqJC3%-Na1}i+;ogTf#
zC5pVLBtzDarj=n?;0G)%E^^}3iMu`vmc88_9O2MvwJTd1*Twbz|Lwh5tYz7K-uGK;
z&-;w`+_9=|%|mxrk8BQ2iIhZ1Q5;E@oY+BNBnTWJFF^nY2pr1@kcS{nfH*IHuz?s(
zjL1`jM6qmHmQ0bNI5npx*%Q0F=Hb@8b?1BLJ+C1T`&4zY*+aFeY17jEfb&*&_v!sR
zYwtDu|KI15&6+HPlmwnfv)QB($21xZ;zqrY(OP4)p)7J7$EoY>TnA<Ads>#Y@%-NI
z9z~JUY_{$<Up#Dh)rXqY71ol?&eI)FQlS`U6Z*p;zHsmzkFe81I1a*<_aiP;UXn~E
z<Y`u$=^BHFi_Q$w(G;sI5Ear91ThF#;6)+A@i0k(c2cwi<2dY1bAE8W&&HiTUYXPI
z9acJmX3NrRICPpVqPU6c`3T2BE6pTH=nr-fxnX&^ORO@=)8}cPI75hF95hgQfeIoD
zJg#Sk`DT~>n;Yavlo6OTLk1o{|8u{@o0q@Ol^d5S%eU`xJ@6&ZJpD9FE6XzwmMqJ7
z<IOkTx3<cW4!-Z>2X)(W9LM+1>Ek%A8vuMCqjg<~iA4x7MqrH~nJRX7cNq-&bUNLS
z{009`=96r~a5%)BF%M?Ur7LTLM9a*_YC<t%Zj8a5F;7NQFct7N^R_E+qY&PO`N1^j
z&Fg*c&X_lShvkl-)v|OO4xLtuAgY<;I2ff#l7zuv2bmj|mODf`N1Z-Tbo?Z~a7cm%
zIxkV7A%Bqht-It_q73($|J={~B5z%JnQJ$$P*y+uAa#*E_2iQ*E-&1F&}M1MTW`Jf
zzBQDNln=+;n0GMehnNqC13KNgkC?f^vyxB>92ZKf>1GLM_YZL5fUPiK8b?^i!SU)&
zM#_qyQY@z#3uTFwqMnFK7gK;c%J8cS$CkK`g|;E`1pBd%a$MRfC(KJkrI-%`x@k%0
z&WLbxOjzXvuE2>TPP-y^;cnbu7{%=S9?Ew)EiAuS7PKdQgfgskV~VC?)Ldr2Jw|s=
zapdSp1Y<t#x!g7mYni6g=&&`;NgR(-L6PT3&trb+2$Sgn2ZMtj=`MAJq}6CK$#UW-
z!f{=EFQiP;x<ThXirA%)c%F;r`vgHq7}TqLV@%xz8~8Y`i<vD9>O4`4HDD2BsUl0K
z%*~%cS3}Y?<M!IxM-pLX979zJv=CS+NGr`KQN&(BARHQj1Ab8NB%$isGa*51CZiEW
zmVz;;vc!`ff!9C@Lpsd?RcD_@;3|pqeUOrJGGREFFvtrkDR|@V4nx{l8Uz?tk1f+}
zhV(8hG1TiE^9_xlQD=>s3R70>j;B=Zh}GjK@K2xS_Bf;TBFfeZ2iB((x9GN7C^W|P
z$=ndrl8pC;<l0i1x^Y4{lCx(YV|#a<z5N~3J3^JEkQ_a7gjTC{|Jhzu72Dfe_ZM65
zCLsh;x=6>x^<4bG$Mb!(GB~b>74?o+F-z(&ruM1B*2l9|VM&s-uD9IXXLoOpPT9eA
z+`8uS5maJPj)RZ_*R9VYk|HCUOfcGFTnD!i<M;w81xn3&!K(^W7In-bV6;U<F;%9>
zMpG~v(BRamGo>ooANjbBK$JON)8M%R;R=*;Slcb=?@jTvBJd>5&?2J>4~y#s#Jvy^
z&Liato-mYBQKn-8*C$@?GL#m*dXg||Q5s7jU2It)g`wG7VcK0GT$*RRxlLg}8H>;g
z?YKPq{0n%kfQ^lHxYwMNrns(0vk`OR_$rNN^Zq^4O0jwOE{-D!qiAMX;~=G^sw#vO
zI8x$y9`0<c`$0eu1QdBe5Cl|ZNmUdmRn-R6Lfmhb5yGJ;3dW-mS(?ynHtUTOYaiOS
z>O=3;X)L}JbsU&Bq(#n7e}^NBNAV1psy-UjQ^ojT%5a+D`3`dnb2NKh9BUbD@3Fpi
zhh}Sz`J+n+Q`bfrSJVR@2caP=67n)38B|QODVOdfT;4ID3@7I$r_Q$Mb(*-Y22s#%
zE#i9(q~nqo8Kc2&U1Y8`&1Q?J>k>EHc%u<ZKlM3QKm8f@@7&<k7r#v4LfGz52#0kC
z8lv9vEG)RHWNH;6GyLA?e}|&VFy<WtN*JW@c<#C9aec2oeX^FFoo%k%xQu#_fiHx>
z^Id%3!}VQ+D?mc2N?g~)_x<}}(c0R26%XV^UO((@oe?{ljM?4UCU6{l*QF@SkHlZ=
zy_oOycUW3HTCdP5jhQi@984KcGhE-HyD&#QV?N&AV{`2ejn*8!qf77(=61%sC=!Y?
zAstkt*_6w760W?1dG|i^l4ffGZ^pbRGA4sv${F*<jCt5><Bmr3UicZ7Kk)*CjoZBO
zov#x(5Og|}!eQMNHFGs%uHR<<`+xW^D69NkCgv6?B+oqiTx~}}1qk-`_qclV%6s}r
z2?3t>kj#z2_XBv4IS(-3-6otd&p%*i@L`n$AqGb{xE8F@<Ymd}y>0eaj}n*w$AB~i
ztvsi9FlD7EF_F*ae3xszE*f}zzt7VfTilA9Y%a{PQ{|*mk~@%iE`^nprKOjov~9_+
zClj13L2u{WJphblt}E#*`P_1RI#ft0xYX*g8AUjr$9Y}x!gNTZEU{X#ABGSLZbS|C
zql7fK++H8DxR~(R1w)!PaFwKA6x<b-Pzp*b$N;A*$c#o7hPCyZRH~SDB>wQejX>75
zq_divs;Zcqb2%97P^gkT|IzxYntAO_=DHH$fVP?}&+Z@J=vfWjJ!4wW^(f1dd*yk?
zK$0eOD~DEZ9?x~L79!tgJbg4<6VdIW_J+7tAgrdyHMyw}YC^BwB5)+C)L0Gac+8|f
zB@CMwt!Ts{jb;<0G|KhJ^OCR;5(Q4Z+u})t3=poY^Pe-#jg38S?d~%zAyXCmgN!L5
zM^2sP`DacNh$*d>W_}^US_gUjG<LXyA4QZ^L6S~LCMm<oh@&S~nO|JQ4+1u)8#H^%
zES@{dI302M(pxMoFMtpzAxOhIHK8ag@=77CAh#Nf;mymhqjdG7wUojl5F9(Uigca(
z&(+Cf%<lf~d$i2eavZ0Q%QA)}&Dh)D=b+z5Ajz|wgM$Ooyf)CSkMJy&s_wtQjkUFr
zZk}P4qR0!1u;lrVJ<m74{mn;WURV&;;8}rChAOQ%$P`W`==K)zrHhAzq97SfQJFy^
zaXp9jTpMq`PB6+3hV0+IixY%|tq5=C$2Tn$snEE+22tP<iUJWh;An~}V{6y2Ig+3a
z3$dZwfGBb?rXp7YKb#{BW8!uLnr$3Ul1>Js;}KyH(rUKRLb5X*<DGnx?(xT%4u_03
zZXqNz^a6XsF?(@?)$Sa{`W9JP;5!bbu%y~><@Plemltb?2Mc9YFijKkvW}5Fwt5`t
zNUW{L@HCmSy|<0yx_G|(KsbP*EDMCFYqMR~MLH74l?W-QRE2cv$)2?qW%bN+vfli<
z*JK3s6DG}aCX)%C<07Se-wlFy>O)QHN~3X+#6sXp2z(DA6~i>4osFonDeG&S+}g_+
zj#G~JVxD^DEZtrkZ4{HOUG}$kaHQnixihpD=OOgal|oml-j$ejO8TxtkqFW>;rjJ0
zE*%ICrjm}YIep4-{8)ot=LoGvgK09t^V=9B$jgFeGoUPU(qu#<?huA8q9DMN7AI_=
zCnFl6gX@P_ugS)Z>&!JnPM&)T;rS59q|*dhX_PTYVX>m#7EiPH?m#gJjvhOP)@tSk
zZ@GN=E$-gke&3kby7S3mjGnFZGom<}*&3+&hcNrMs!}Lb-=n&0$)b&An&w=&b{$6q
zoIiVxE7z}4egvkZKf*ll5JE9Z6I$5_l}*`P+vN6M&S;!+tQT?dnX|NeZH!S&w|42z
zn4da#hUPn%TVv`UtOf4@=C|LKd6Q0Og+`-6nvC##fiZ%jENC<XRGE_}BjUJ&AGQdC
z09RU^u!)|I3B%ge-FPu~Z(O6(j95MQRGlagMKk6KWh_$E#}5`umcRE4(+I)J(IY6W
zaioj24%e<-;qKin-lJ9?5Hl0g4`43tF|X>5s##xN-7;S@zjplwjtn?^<}6olT>nUn
zXlwf&NgzQ=i|e~cG*OmuYBHc%<~+4&$oK2U2hojq{fU$8c3Z?4&hH&?ad)587`}95
zk-Oavgg#^yIxP@dA&mxM(UD8)OP(AmdTYDf>RbAWq~(_^^%RY6lN)iDSEG<ms40WE
z&5Cqb)fF#=F|A2XGfybvHaCsrd%nlW5#&*mo>x)i6{;#J`a7h1+f4I_gVj~`stW6S
zwRx(o|DTO92m#ty@-lsI=F~Z&Qc5P%DOaytp)54lZ~TD4VEmp%^Y_QOHB?HGW(kc}
z6R9LcQ6Pk<L({E6RSNxpt-`&4@T@3_+X~l-Ilj8ee!ou~MvpwHYaKsMl`GsNMF@?+
z$6`<_#X+DdLH}UF{!rtF1@oOIXHK^8qxwiYA0<?k0#~wdWQm~N!aB3^3uVD5a;-QR
zRBQ|jg5}dJ1WO#ed6hd!N}?h>UDAsSLO($lnxZW0`avYt51D2Ky(mU0MO9>EmBkO*
z1i~fo0)*qyZbpRj5l2?KtliBSjR%Z}5=Yk)4P2M$L4uh*d!TOcsl6cI!zQFI*QMR*
zV2s9*b$!+D?hd2T_&s+}gqWEPTT5A%OeRyTg*c9}#*${~eYW+Em!2J|zTK5&Eock7
zyL%|5SYBFWnr5_Htw-VwQ0W3&3Y4xr1m`;rZp%Xpi`F^$C}n>wWtde6VOi|P96!HG
z+-{&N#dK%DaPI(tU~%;b@q7m@E$KLAG|tJ<gfTQ59#JE}Q3^F4v$-+g?i5CuL&Gy1
zT@rL!J{T9_#iVIL7)k=i68Hw+mly}Csw9ls6j}Xz^czi-71+@pQq;!C!VMYj?h-^k
zj-8o#2Y#K!S}2Ml$Mt=1-MThVSX5QrZ$ikkjG`#+dxf^UU5v3azd_6H?hgIoh|sH3
zU5u^kwIz}~FY13g^D(rxZg4Bh0?+fnTC}bcmXy{cX^JtHAPDMr^{h$8T1!^eyTAgM
zm8E62cXu9oV)vmYb;epk>Cg}YS7?;AwEYOr^*I=4T;ERk-pw(AfnWX98J<43faeF~
z$&{V#Eds~m#HmwwVSwv<h@g(6$<q{FlyxnR6ohev6#^w4Y}}-1W%Tdta$>>d)N+GX
zH=@~U5=Bk&sv>U9)ln<D1R=0)UEgAiq|@#Zg-t9Vq@i*gsx)IRaO*8*1GleVVmI{p
z*ynx)B_uKk8TQ9`Rej`DV)15CBJU{+Syk{i|Cj%nuYco3E<X7LCr=&c5C8QaFdR(Y
zXX~_vuXTN7IGIf7cDrP0Hv6IEBuPS9miJ%y_m5^~7p=4<D-{3NKm5mJNy+k&6<Vzp
z2mMD6`~DH;#?p>z=7R^BTlnQqp5fxTMO;6iNT%#>Z{a&0$7al(2bdQ>#Qa`?tUyZ#
z;wE`3r+;VXhnP27O~R;2QB{O9=H>zB&Ww4Z-6IU^E%TbWM5QTh*TLxeUf;fcnJDmi
z_UGP(xwH?*9LTGZzxB8NH@@}lZ}a#QkF$Dwl|TBzA2J+F-^UHZ-k;3AFZ1F7=0@N5
zKzf_G<{$p!e@d2?EU&E4YPRSP9-X<K)pvvdLQ@$_Ck%)KAAgcFy))waPKjs0dHyu7
zp6ZeNJ}b$TPw(uoX$1dzVS#ZRVZs3Ed#E%coMyD7Ad?1M2Ng$1Dd;(H7EoiwaC3*n
z7IlYv6f+GQyc|Y6;V5o&<}hA>)&*lBFrG`_b;zyYk6Ilr8Hp9(dJau5V5KVA9b_N{
z)o94p+D-0GOU^ufo+`=E!eKH>@QuV^DXqb`F!_Kz?m{d2g8?UwAIJAy%Ce+?u*<Dm
z8?(vxyUW<x?jg%+FS@*FGo4HbgNP(a8IMQx$$~LdRf$sShsL=wmb}z#T)Rx3m8=|D
zVcB)r+1-6)=6<Kk;_>68w{B9Y5^Y^v=@SQ*N|)T)SKQpmAanTCGxMBUT>@2ta2X5-
zgmFxFZUHX{a6Gq8$1s+%D$!-dBrRFnFBuq@-qKORIO6J^HLm9q)^-w>S}u#dn6gN*
z#<I1!P1uMyb@l>fk=GV0u8R=&c1aDD^=O0+q$zK{@g}3HKzb3Y%g5LoPHD#tlHnd@
zkuVqyskA|3Df0?fgCn4{AQ4m_FwQHb_`(;z$jw`~SYBDC*Xwfm(pzL%@xIRT_q@yQ
zbu>p&R3FnJP)bph1$xF=YlE5bw${$P>mbc?Ui{{_P?e_B=}@V9!uzNi<rP&S=m_Y}
zIYbSQAn@@WmxKO@{ZYmGP!V~8pL$}M<EwKB&m|d6*x%lvQj$(@o`r=rZV-UyQWY6_
znxV3SX5bQqE`B4dV<jzU*QbhNCI@4dI)atCkhmG(x(-@76qO-rwF!NJ;~0eNfRuQC
zK#^)1tuDjyghCmlAJp}bN`tOX7OaKs?YnrM$LVK2i8co5d!&;LsSH_`;zV)X2`vTd
z8ygJ!eS#pw4Fd*)0aaCxVP!Ob@CU!o%^Np3a`Xtj-W+ee`36~%qkYS*Teq0&&CzbR
z@jQ<#tLq+>(kQ9O^PDtI>wNjLs5@?pk~B$YJ8@I8yukN;94TiL#M-M+X~S2(@->R0
zqSNgX1OaLK-V4nK_2DLU2x3PP$-2%%YXin0l;QQY5ij0J>A5aH|3sIw^8rHofJ7<F
z>gq|{I6&Kao7r^(l1a+Aze6>h5V#?6w@uV&p{!*v7?Nfg#u!c?X)_&6m^5P6=T~Sg
zAE)KH#6_PYUPvbn5pIBkqAYXDvO)-#(2WTi&Dv@~I*gMk+D3>tz-e?z1|v`<NtU9F
zs*gzBI>5S^B)Iif&8+Xyeqc<Lv6dga`3ARc-69Nqq$8M46RPS19vD?^qcELLm`*25
zCR3zzkQTBmMX72w>#yqF%m<Ey?Ch+)s4POdEHBS7*X=%Z`}2ooZr;g!J)`G1eC~-Z
zr{@EN^zJi1vU(Ed9n8~7%A~)8nojWDkf_@xY{V#Q84reJGv=p`w3rUYOdBzGdMh-S
zS801LQ88fI4QbDqOBB_NIl{$%fI07E9(Tw`Q-rR=$}6K$)*u`Y>3J06DXx`Z>xaK%
zKR8RkSj!LIdh>^wC+YjX+^q$r^usfs4R&}t7uZ@$sZ7m$X^y$MF5Ld;nEO)VNkM7q
zmy$3NsSH=|PI+Y~XE6+T;ldmTO-Z33ElZLl;dff|m|hFl0V5=(wzT$#e0n-!S(m&V
z#H_WNRE-GVSUx)(bFs*20Ai)hbU0<;IJ~4iY}sVNb@_ByF$rTf8Vy`8V#8K=x}sMp
zq#v^F#jHm$qN)h2<YOx5aVZ(&ai=H{-3S>r=nsd;MvM8#=Pm1?fug7gocb8kSX(!9
zeZc*(*0Q~|#dt8l4?Lt4q*?ZX%(c=KWloYNBuP@+l(-&6o|ESVd6839RXw58>VB=<
zgUl-hQo1ZIbm+9(+}+y5et;L^LopYUc&?3`_84sCxJDzb#1(?wamMxif>=mSEk$&E
zS??GNNv2_LWd%R<K{((!7;2BPEFDu#Gq%P#yQNEG<rE987Ten!+<I+|X;si^`W)%R
z5VjdabGVHqUfrJHdmd*`ox}Hhiaf6u;&FsBdKQuEfDoKoYG6XaW;UUpWH^z)%O%T6
z&T2P6>5@iciKy8iKNwO~6|Qs<7CZzQ7Gd8n=xfFg*xKHr-#@^0U4*1IxP70lL~HBk
zrBW3|QQZFsF-ujd%6fs$+szU79n2j^Vk``XBaW^dK?uRt&Njp0qc7h2sb*e!EXEFz
zQ%}O{k{-j>KyhOXE$Q&|(TJ6fSDTV&1*72<;x2RTm`=mP2?DfmsU|7axKCK*b?u!W
zpgL`g8=xi=)O3t%G)H<Ns!$Yhz;3gJ6SoM2z)kziH~spUJ_u1x-FaMEh+H2jG)@>|
z9f@@WnJOq0C@;jiAy(C=YH5+<7)5Ce*7cB%OE#L|JM}SmsVaOTgmg&Klx$Kr8CBT2
zLF9pV=iSXs_V)H>8NHICs+m_+#b7ufFA4_30j}esl}78@m$0lXS(XxJG3l%!Mp!6j
zhVpgJyuM#cr_DXf5<4s10)(W~?$B&B*xTQGXs#bF`&GC!qya0?Dgav`rOW1^;I$n^
zWh58clGS;KAc`sS3=?QtbDjHE6rSUPwWMi6I_~3XL!;XvYWHx%kWwp#!vU_8%(XjI
zx}quzPMlw*{Hb5&7yim$rP1qQb;)%54r?#|F?U~mk;ObgRu#q(7;SJohcFCqd>85a
z+_-g{>w`U3p8Ygiso|90q}%Hvgdzyrq*a0MxHO_BVc^r>-XX-IF!wut-Y2*DzLYTx
zrxSR{W;AQ-x~U|YGMSFp-`}IrXkiTq>YbdjDk$@evdHTK;UAf>S>mXH5DtI-SAUhG
zM^-p-@;LwCH-Gb?zVrL(oy@H*f0(%zT$qy_n|JVM%)LO<oG};om}5zkx0%P?4&ejL
z$HM`RtUHP-U7^Z?<L6h&U-)zU^ndza(d>3GT9NGCW$RmCW$X2CGXF!&E$)oD^j#d^
z=k~4J+#Kw)^vnxvrG_J6M7P^UNsS+NNUH)*IK*L0=zH{&DPir8ULSx6i++gvr;TBh
zOc^B)e_yr!5X^NoTfzVEHKHhfoB7cdPMkQwKlshxeB@e6DI6kcuvVgVfV3s9bl4r0
zTpMaCBRCNmvX<f7MlioS1kY!4X^yZFAUuzPx>+VFOU~<p1JB_~tHXY)OX<3}Wx*##
zBMzj)|JQEQ)tb|JO7p}L*VF=UZS0T@OJu5WlO`wC0v8*WZBr8JiqexD6Aqc@G7bVV
zM<RomPp{o&xtP#D{T!FJGm?V=(+UbbVsU;IwB+0SuqJ#QQ5!qEzK62)E(zk{#<{kJ
zNtQzP@b^_4Q?J~U34>vuxY4LPA6<v4sxavLaYpsR?Va^}5QQNi`1#NM94kjwxOw{~
z|LC9m(+`99U8Twbt#iVK9xZK<dJ7>W+k=X&T%&-chGec|5yHbLjTR2WLP4k~QiZAv
zfh*`XBUCbFI2f=tt(ZDpjz9hsXU?4AYhV2m8#^0VR}#t?={k5vk08%H&oiI-b2K_J
zou1_6@eXFR$JQIKU=R9K2m8beG1eG_6!@;o(RPEV5pZpNlk0n9q7!FGghM9`2^vkp
zriULlsjMUl0w%)2D{`=PT8I^vKtN$Wcv1(m`nIBg_c&gE_rIxYt%@QiNhbJyFdN79
z?wvLYT~)IMyS^V&1$8}MyA>0KA!S~0`qW8!y&lb0lb62x5%^MChdIv@cmM&$C=4*n
zE$d0WxLb`JdMzK}`c!F3VFa{VgnmFAxp+~4Hem7$S!RT_?+#wj!1R`AFD#?e6f>_u
zR}|wRdOTo31WZmp#qp2*6n+$eH6;Cgip|?(>(^)oVcl5L?vj^^aB&g4y9Z%_3`4dz
z?lLqG9X-V@J>|S`@O=+245_R`QfibNpq&!WahMu|<9S%=pmqIDcRUxH)=4hb);Z7j
za$oP?BA}@1grFbdYBWk!<VAgC@3^()R^9Ak$jSnXLAow}5X_$Krk>cBIaQHUD#iT#
zLfrxDxpmjAtvlXbx1O4P@)OT9H#f)P(gI)p>en6$>xY`u#lqlANvWYSF1d0@OUtdz
zl2PTe><f-{Jrpsyvcyq{E=sZ_#Sa5q$EVP>Q@Z0qXRd{#T6G>@7*Uq>il)<U<In5?
zs<Om*K8t>XapV!rwGdH=upXUL7dY2mVe9zg-24~+JEvVs==p>?B$-T+kq;_kZ7}8c
zU)f~one)8)${Xx#4*B$7{wsK$h*!Sx4UU~W%d2m^itBlFyB%8HE_ZL<!to>s52MtF
zyHOuf)>w+NBux^=lOg*DyC_xRyDn8#F`Z6H)4H?Hk+p}CF+8w)1D@}4{`}dx&HvU-
z&Ye4l@B8GBm<jr$%;mc>*TfXc63v+BGv<!tQ|St=6w-zEJD68xMOhTI9$@aym@oKE
zCb3Vm-N1D{aDBYvXNlU&>>huN+rRhsIq6yg&nMLOhnTMqru^$yHd%b;0&l+hCOex0
zKK_^fGyFz_H@^LCj-EWjYj3=U>o{~e?V9<m8FMKxmBtnJL*@2<7#aKi%#-(E{@_dA
z_a0z=^Csucoy8CQN6vwbFgVgd8B1v#D&<gAmi66|apkh?3zpg*d$~{L7~b#$p0%1-
zmkhLvwGyYQ2mu>-ycN{8G^OL9r9@;Y6b0XHwkYEmM=N&IjH^M!=N3aA_ZPXj-zQND
zci{-pT#o}kK=e-XtUu;8HDIsE*sm&%r-~@>XxNGq2Pu<F+f1K2Pyek;Y~I|(ee5Zw
z(}eSJ1Gha#|MqPLFWsf^LRvvck&p12F>>V9Nd;0O)ccjb|H)--oef(@flU|;4!|1X
zD8w4ea4=vv8j+?6O4m&j@9b)JU58_<M+w7#yXzaAK6&bX8o{HeoQ?^EMtC8?;u1DZ
zNV6$x2bO*7vm9HZKww;tq_o6Fur(ZWpj@`nl5P;vZG?Ef!*nuYl%#B=g0XDTSv<<o
zW5@Zyo3F7m+{29{q=wLV1foeLn$Yaf>UD7fpY8o^+*mL_zeKfsnzvs1CXcs*dSd{W
zGA+=agYUxTAmO{~3FVPvZ0_tc9IrEf{vtPAL4R{d-0ZTQW@uG=i1i|$p)^E|n5@Wg
zDupY-(RD`9N1p6F+m$b?dM7qXC$qY|y7;{)O3JF9&`BXM)V~vJE$vo|UT==cXu$DP
zM=+*he}9jKg@sw-!K0}uA@&3c2dxE_amXseXs;ku4n1j^3muBerPSbi0XD5bT0Fnr
z)h?5|*x&QK`<|r2agd%*VKius*J>eM7pp4#W{0xfBj4Vj8M!pOZRD(zw=us+tGB@3
z;wsm^`*ltooujaZQYl(O63)+)Oot58itpbZ^7P4z+}YUS+M8EcdhBU}IO5ig>-g;@
z`u!b7Ns2Za&v(hH99<UpQH0V8DFlHZAY{#5-0LuX$a`nZtUgZHNhkLHSgp%yR8@hn
zxL(3!GD2&O>(mJ>S(@HoOoC-Po#OlcEdRHj*h@!p?C24E-)CoMhn1C;x|!-B>gqn+
zq|O0p4Hc5oI7~8w(FIxR(9u3Cjfz&o;nL+TE<P=Y!X*OF$5x7@C_q(+Ss`eiOfXeN
zcWIgC@)A{1uTK3&RA*pP*F0DYg|-AXC%g1AZ++v7G?tGt7!Emg<^r9@9PQKRng4~q
z#QNv|3CF3BN+X5DZ$#wE@TE8IawF`pc<ynQ2fK`Kf1SVkSO064&z~b)TH*Z1U*M%z
zUM9~nn#~4_OAFj+MEE!a%8>Pkcv8M+G}uQWt?I&Y-*>619O=|6wJgg}TH^;kuIG|h
zCC1c6)AwWnDZz2z`i*NO)0DibJ1Ke8N!>#-zjAqt$DS61@4;Nmm{V1>mX>MEm}_hB
z-i3LoExt|2FTKLG?|qeU@d)G5h~p>E(P?&Ro;=Ik&;3W-{laf?j0!Ph?lvMyW%%-a
z=8v&Fm}7kVMgIPO``@s1{v6TrGUuNEG~a#o74o8>(P*-;w15}K_(Bp?hHNmx6XJcN
z!5&pwnHh8MN0?Xd!u*{(b>cmkKk}r`0j#044wZ4pi;}UanK$uRXr%a_Ai2ED*$YrD
z_Sk3y#7eU`8sQkjRMhDX&txe*t$Ay%$CbG*PM+gP0o^uL5Z2`l)*yYKNILA>l#i(q
z&p!1S{i&jL`~tg0L4S9Paeu<~vmfUL^$M?r8Ka%MOdOZ88MCS__S!D%QHzVmPf>^v
zd21hM_b#UzZFH58x-pv@Tde$23pJGp&&LUUBDYI(I$@k-wH=eTKe9viQAz9in9cJX
zlq#_ps%mDzGaOMA1(i~@bN{>S&Wc&H(w#eN6h(;>xCBA)$cwjEfAAWy6jVsGacOlj
zJX^4rNhZo+U<{crnd}T$S~1L(25?YlY;N#Gg50lnxQjePD}~o+6SQ1LyKCJ2@&=_Y
z!E<ZxGYs06O!Pjb+GcqDE|b<6ui0XnOj+ArrV%yC(~SLSk!x39rsoPU7OM=U&T(bJ
z_t%E3D>%OFV{O3R_FaDWAOGJp78Z!xZI(};<(1c8Lpd=5$(3ov?Y#+(flNANy22Gd
zrU@PQI@RkOP{)yIt?qY7R#jb2<2p{AmHrOq<H>|+IwkNulDy*j%{8i0(P%YMAHt0G
zz0`ETb1dVu#E)FE(%@M`S@^_S(DHLU$8hKNKEloEG#gmaz%c^S1f6LLZD=$?Oj%*I
z!fkgDtrj{<(KaKALZlASl_HKCxSop)LZVKW@zxy<Hg4cH+T>|Uvpt9J2dJ=3dHQMg
zCL45JOE&DYnPj-W#5xW)vw%Bp#FI-a^x_81*Dmwo|NC3atsLdRi&#4I80*6WT-U)D
zf~3Dse{Y9Y5Yp;)+1=a6+WNWcx-OL{aU2Js>ihpb;Xyyd6=uie#(+Dkfi$z4NFi#g
znAtd&vbM!3wMOZBqUpF!-7I6peZSY*=T6LgUh5<OA}?qw_0V0gA8Jw;1eUB46t+I+
z9;BdhgE1k?zF?&(NlGY7L7ERq^c-D3!WWQ@#&r>FSyD|?!XRex+-dy9xw`X1+S<w5
zb+NWC5LMdHAE&%>^A0!mhZI<f+v^;C{<B<o>J#j5Z1K$V&(UbjbL7I)T>D~=g?yWK
z(7<oT=+r=-bGtw0*rz|mFaGMUW0Q;<-+hzm#x8en-^D+5f+s)w^E~~;6BrbJ*kEO8
zkyf`szCR%^6@>-;Fz0Z8LRmmrfiiWGx)oq-UGJ5vib;~N*B>&TB(;atJI;rskVsK?
zWPRz6{}|T`c<iyqO?`JXTn+c`OQMTjql_pSUGz?(Mu~~&qDA)-jL{7eMDGNnchP%~
zUWRB<qBH7<UPd>#^WA%Y>-U}i&RS>hbM{$ht+UIspT{BOUGZ=qwLxbpbBd2P;TiO8
z@y-bvAM{%8xz1Qxxq~`c|7ebMOtXl=Tz3Z%r1DuG_?M&jvFkoFs?j4>=Fw=}6q@}o
zO9GO+69HSren)}aHsjjQ&zOu!Up4!TA3lUw<vg&<U-5LszH>wlub7&dF7-=$WJcwb
z!anAK`_jG-C>ukboplH*mvU8XvvKPAnsXfqiUWqws%A$DKc!K*Z_dI;{Sk={Ur90K
ztD}}g!D3h_tkJN<075k`2?I?sckh*FoJ^@e6-BL7dP+y5VM0Rr4Gmyyt?=BSHpgs|
z{Tv<wkH=91XSXKJjuA26yC2t5^-Q}O4od<FagumHl-l`h^9>JH?$bfF8i+L@B<&s4
zp?yzgLq28hgsZ7<qLsI9Cct<7bUt>evE;;fluce?6>?YgnX*j^I>j|5ySxImK%%n%
z|G38FXpM`6KLwo`BDu$Lkf^4%aV9-#wUjA946Jye9rTyk;$h4skZ6MR)$8&}OOb$v
zBZJTC6dar}KPbIED0A*b_^%m4X;nB1Ih1Td3s_kv6tR_45=sfnMcX(}P2!LnGFrmL
zx@}e3)b)kr|2~Swu#7>(NN^e1=?SA25yc(8p+#(P7;zY*sca~qzaKTS{2MF(Ib|!B
zLR<1;5ieE7X7cq6!MIES`A7YO<yCfo=3f9Y*cURIC8JUD$2J|K<&G~OO=fn8qyO(a
zhkc4D0N}JuWqQ8Z%Y?69mrX}pMzcN|I!krI0;v=o#D`A6=|;AB48V+G0|PcPb8L@f
z@b2Sc75m|sLXpPLlM>A>gFSH{-GP+pRiDzi5*?JK8by|ve!l9hrAgRHE>@t8+_ip7
z)rcQgVX+avZm(&)WymGS;P|vs(ilQ?acoTAM5^LMijmPq46NMS@n-r(7Fm_Rh~W)L
z+WD;KK?7_Dm(32tgfWVVNQrL)>cjY{+i?DqeiQ&M8YJF{;_hP*Z@szw=!E+$VD|>(
zb~+4n@L2ruYl=j%f}0-6F^Z9L%b9v!sMG#p$mofw>Ez`}*Ec_r<5sGfM%PAC7d@^J
zHruF{+l4AHXX8sA!YN*kSk}m4s=12@_pBR7Cnh3?eQr^+z)ifCSomZC93Q=wxBN_Q
zrm*V1--_70D^vtmSuN0zygw+krvJ?w0kP!P!6Z2!sZV}a{8_@EaSL8lv4yXE=_PC{
zN%_mm&(h9ESg^->@VhS$t2cLOfnf_hIU;_uh-vYZaQyN*aT-Wua_fDD5wIs{I$pLo
zyyRmJtcwDKh5!Lpca5S_Z<O?kE%@+~f`ZiscfYZXjsf0W5}!^^TQlHg9yD=z43Ku-
zcx2!A!%qsM$WlBDzHYX$J7{U^89v>E*TwpMq)F4Hm_2rK+n>$yrRH~(@ySLUhxmaA
z)EF^RV<nuTwqP&}aUIjuV}8dFc$m;R7iQDh?w2|X6Wg35gOe*cV|JsmbTlx<3qMmt
zxb3JQXF`2oJZd}D<!qyp@3I8vmtV$B8Kvtbdh@_#+0yEsWQTcw4F{{|l;{AxMpDU=
zpD75?B)m<E)*({;>zNvFs^Pwcg)H8Pap^D=_A<ddbA6SIdR@@ZihE=kV85Vq);h7S
z(@a<(iVzO2olV@ixnD{5O8ODZis*qxJRVg)dIW5CJ;tcsYQ@ysaaL83I|LV5%4mD%
zvHa5(pxie{{2>A;Yb|F3u2CcxzcCbh;l=LKhLuou{cuL%2Qe+ve*)J{Rwl?+YOD06
zb`+v1;i%ML&=kM)VdI;O?bpTuPnIk_7ZjoQJmja<s;=|gd|5CiZDO`}qWe^YAsMf|
z(6}~{Yj!$O#5fnj)9E$%+p(f1Ek>eas`toAxD8DRl6hpWejAhWA}WgKoK$Q$lA>U5
z@agA38EG*Wjrd~8rH}2)EA#oh2wSt!je*s`Orm^F))7deh}GdZlibBFsP1$k@G7BG
zujZ#xQ)^tqdn4r(2NO~#R%bzkY*ZC-tgE)oxVIGR0gkm#Y%=x)coIL~EU9MUwaSa{
zfAG1;Vx!o9@HsgRi0GKC%<l~Lq!sGYae;q}48xL`qoPJMaq)<6{ZsV+l~-tP9@=uw
zu}|sz+Q|Nmm5bXqMofEIWss&?79^>e7NyWI>9o>0Jh=(qqrm+{VCTSOQ}n*AiC{|p
z63jn@qHt7YJjb)w>*~guaWc5{e>BF2B1(7(8)`+Kb9J4dx`vJ)%2lmarF^2@KNG<T
zxpiD$O^LL>ld1(#hQ$Ioc{Jz-?M0H@e?beP)F>FuS<A>axqnv0IZRAmK`|1=Wx5!m
z0lJ*m-YweU9@u69`(g%{DxBi|5+p{*Y5td^S>!`CV@Inb?KdPkT^fvHpVlRIuSvk|
zWA0wGpZJPW4i4T)Ui1Q83rG@nD0%rYOrFII*Tm&P7d!?&NF4w$+%l7ejYLraVI`6E
zo>(<RTbGiyF(!p9zX^OBFaM<SnX>k$sKipA_S{=Douo0o`-k|B4msnDy9mWAD5w8~
zoe(U)5hGwiLqkJP5_X012xeiKAZi%4e<zJz%y|?I*nhmKBt@HH(hjf7)CYh5ILJ%!
zxl-L&K4`K5nIAnWCh~V!nVYLCLB~HJ3F6DE_YUgSfBNymiVDjAb}Or5%{~vQy$%Q$
zVV;l9E4w}_deZ4yG|_5)tru_`gxq!s_{QW<$(8N|pgJcVtXMJIM1F7XhF~Ur1nx$b
z3{1Fc-wx;7W>z?7FN-yKdaPaIpi+4Xki7_^OnQWE<I>vRQCgI0Dm}HNE+GT=tMo~U
z7XqSJWyOKZeXGVA>Bz!eKW~nXxDYa>PvNuaTJ2-rLN+qY^>shbh7kLzFg#9yKrdv#
z7am3xVdRQFB@ZJquF4Xd7R5ZOTDGx0f`^{mmMaN-kY6hq|7dnIYp^icd?XCG1UfAn
z*&Xh2Q~)Id&ScukYGU{cl-}6sTR2he`{@)@v$y+EsRb!O&NgtywVWXpOKXcqA@Ddp
z;ZFGSw8czxYq}sUz}=DAXx6xn7dXP4;8QZ@En_2-?P|R}39$e{k-`-qqWSrrA3^!&
zKV&r4s6>K;7M;g96n>xkbF)tz1AcpOS9!I&7mbXJ5ZQmiY?jjC6o0^%?<dH>H=5+)
z)<QgwN2I@UYL>X0Njg~m+2iY(aX(Yim3_7h`!f4W8(k3xQ%54(3uE9zzSOK8k<nJY
zxjoKOTN^#e<=@wT?A0S=#<by|#-#^z{zwsxk(E6Eq$My_m~Y2(FwDH*F<>uK#QUd$
zcMCBpF@f4(!W|%O-ee8IfEb=eu9}3gk8z#_(aPO$1?-RmJy5jFRssf8cuyMs7;al;
z5+X{0r&UC4$#1(oOc<XUCFYhVO5oSvr)QB2CKyf(7zWp@wK4aq*HFWzmKWZ&)MeCR
zWME2Y_`F<9?JP;jSS=Dcr$kB+fZvD}T-B)z^IrRB0ZRJMoeXpIKRpXB%$<3!S@I+Q
zHIbr=Pr^+;Nc?;u1Ig=UyruqC$Z+n9=$4NWz+CQM+6Be3%kJ_ACFVp0e3`kv=Ne?8
zjW@2@(`bC?UTnaDk=Rt^))AIIXY$z)im(J$-@?ciiJhTNthI6<k;_Qx>K?J9%E|k%
z%m|x_)=@5)%v~VlI*b0GCqZJZH+&f*=T*lwHoaqb`1U2<9q|~6a(T}o!(Btupa_WW
zWm0m-di>D>jXF;<u>{>qMi&)td#DchRnyz`8Cdl+F})rylXOr$)w+XXo2YFObpn-S
zTLnIA&U~uGOx3fz{Tb8RK~_fa<Ip+iOs7Tsjh_Hlk@jAPRH@V9zC^p$;rxrEKn#wR
zPLWZy(CAG{@#w^u{LIzu`=qwk>7AAnB6;`-URUq}fc4~+%A`G&$kLbn4PWYSiJUkI
z4Dm+?LUa3h&y=D*J5kN_?0gxc%HGd23D%}`c8<BZpHneOmoPzJ+eP%IAs)_LP>W+D
z_S?Tb)M-8nR(GJEmrP`>lgdB(=!3Gh1klQ3_L$GES>t2yD+()aZoS$jd5(tn%bh)l
zyQRReiw6=u`L$$EiI$Gx`Z&{ECj8>w^{MY8nSpz((@vUOVTIV1Pj5zZ&)`11C{7Ce
z;%TnU;j}1_srBKdlwJ{COlTbR5MXm;8Le1w!vfJ0uFl=W-I8u^K=dqUaeVrjbh9{{
zHZhSVe|kDWq5AA$A@z~I8u2seLt@p(n)wPCKpv7Pk4EoEdY(P)x^)bUIJNFyvxsYI
zx`OfOfp;*a1OqxJaOb!#;G}Of$L9c>$vhrEjK@F#ax@&JIZcoNNDyLCdH)Zof$6CW
zNEAgtn*+_b@fPvj+HW!dx!wxr*glR3^de~nrR)#a;5cTSuM{>~2@7y~zffU;n9qGY
zN=KljEv@q^J{0>io6E_#^n(($QFd-xhguQ*G4zG}b^-y|T%pkWc7fC2C5CA3)L4c|
z@Dpc~4$W7~Y%42P7ih^}YxooXX!eH{@ky(gS(*uTeQxE+iId81#A3H?de(~U-=<b=
zj-P__1(@Ly(4<F1FLSX|5j{gOSEWy9<61G$Jk$P8ea@3O4oWNUOYqbo{R0)wgbYf=
zfZx2RLPYuUg|RC_G3EwGA}}Rp%fl%Vi^7pT;5|hMOJRRPTdbzYnp&KtRAullzUwhE
zOfUU2g@xXVQ-PqNDfrd6kE5_%2^nv7b`Qkgto`zN#ywMXiqeN^Vi~bH@m;=A$jzBe
zTZbeDU@3Vf!vWKZ(9Alpuxlc}clb(#VOBt(7w70-Qew{)**$d#?4R7`P*8o3-!s0-
zPsB4qrT2BmTi5zpA@@L*G9t%>+GfSNwyti};+2x4Mz*{Z3i(SrRu+=r7<`=T@hwjE
zd#X}i(If$xCv&58F(<FFhK$Upx01|UFI{lt6Os?_w%Z$!QsEzjGquvns2XgfN?RJt
zY8(~iE_XrN)*J|Ll#SAw@EGCk5*xZS&(AB4%}*M-Jy4xUl$vAQ@nsdUIo`<NEzUpb
zeIPPmAISxbeusyU(v@C9uev;n#S#s>esMIt{kX)KV_ul25qjs}*;Nm(O#;6Ch5h;I
zX+=Rpx<^Ntc6eiPl}E+i_#6N-Fy_At`}2+wzNHLC$X~3w2u`?7@mhIY(E0h{+1tFW
z534u0c9rw^K$aslI0ekMzHU4ww;`OHZ4CqKB<w`~4t}bbLZ8eXpIvB9ub*Q4*3Pyj
zOo7@%NkdPQQ;et6^@2oTwbZcfx?bk0*2<*wR_O7>{DJMVj>^?OT<mKm{0Dwyn!w}w
zj*(&7jzPeMz!&o9JG_dn3tai7=$c+uK@4mT#*Y3L$5imQn_(es{fLBsWa-Y|%w8(<
z=o@?IS_FiaQ*g_n=Q<bU;v9Rp<P}k5UGXgLX!>zer=EV}jwH3(WV<v9^xIy^>GSc~
zY>_zg%_6A~!2i%M<SMQGc-*8MdtD9p`zMFNW#eGXKT|)~cC^d=f&>A@Lj|4-JPO_!
zMRveVcrA4Cl}G2gu9azMkT;Mnw4(*iaLvR@^Xl5nlhOKyTup#Qb}R<Jm&mfP3OG^F
zajT<Eya|Gjo8>C#*y3yk<e0e(n&7{@1ZMiM^<SCto41}N2sdK~{h^M17=^<E)|Iz%
zFOMOnq`;@9_Lj87p=BbxKN1tz2)?g7v$4WA^a*#>&~^Fl{?GLk97m>1mn<8M$aDO&
zf&5i&sM&q!{q`~B&?4H6HMTmlB;8QOK_&iMOHwjt>;eO~F^~gzjx~qy&YFecG;qW8
zB{ht&jDGE88|Fvzd6nq&|Co?qLQOSnh45;B`poLQ!I46{$lJoPuQxu)#$zJeqMrLK
zk4G))cKspg2PesiY*_hR6>9z)#QRORE42HM<9}YN{+GG^-!iodo>olNWo&tj@IInF
zTar2QWW%ywvw{&H;{VlXr(XtX{WgsLfKRkY^TPyU<I-LEwYb$S#j2di(@Xq%l>uQ>
z#$AJL{!}7ALXcnI&!h_qQ(;3!M1j6iZ>jxq^^3acE|CkaP5!Qs$^l~9&sUh|hScqb
zU=OpaNs{CHCTQ&T{_6c;T5s@2<qFIwNFV)!NmPD5@ElQ$jpzfjx3{-r*l0HIm^V+8
zv$76VJDaG<Do3X!y@EDwMX~q1gr?Z+eUo7vM<MJ<nXJ{n!4YH_K$kA(+cQS;rf_3H
zBPm-unS3E!z23$w!8uY!BuH+c)GErPGay34(cm<LPl;NXk-My780trED)saFyTI2F
zc(!-SfQpjTQGI(;kL2!e1R}UDe^c4Jbz28$94#|nVIe|?Ni$y1Z#Afod>cq(4;uVM
zxGBR161u!q+~8CJ((|VbT?y7#xmod6ZN7gWch^?54d>z79$gYDSqOJ5`Kq2k>oNRp
z&Cff#4>?Oy4v(dHD%a=Ue>-@qX!?l3o)0F;hEB_M*yXI5l+ND`UXjyy;!;$9CMX(&
z$e>&D-#-1iFT|y2<_pPyL*B{jX{r!J)xzP~Qo-d^E-Pj`IS4wp^1r7VkHdjIf!F(8
zgOBI(54eHo`|3o7Cm{_=ao+}wj32OxvZ%3nB!mE{)W;EX<iUXOavStpFhg@ZruFCN
zR&?oCM@ZWPl5bUNr*!pcpRk6!E+=0L6})%CkbXsn<Z9!g#L!=~1EcFZ#!RL;5<GOL
zGO5nLx00fP4yLrBSS6q;mTzn0S?@fXn(dp?M?IFHs0;zQvkQQAz@;fH17La7%0Df%
z<>zvA;$0=#GO@g~TcWkeHSm3DA05S@=u9gc@F4#%anopiNBuC@bt8WQM?5Y@`@Gh;
zGz<?+#>iIiboIl4TJ_*V+<STy+p)#NZj>mheCf%{3<C#IG~m<GM4D>ng!`+Wl~voS
z^>}qP7CuupdQQmW*T<w`lLx~zP|1@i>me9W;3Tc7;&I)8%x3(7ZW*_MQE{t9>6tRD
zryG(+g;nWzCzEFZ)Jt}V1>7CTfzu;{?3z5iJ^Xts^A+zRD5Iq$z04kI5Z>E&PIkCx
z{tnl#PGnzV(sHE3>^M+|;e7t7yaLx(;}sO=q$ls48Wc6tS?ah3!4e+M%+0$(3rBrO
ztP&(ExA>Z`kg^1cDeaM%!gI4E)_=)#aUOb@vqCj+ukC4M)5c{;IZo`syMRh;gx$qo
z><{>2;$902tND}mznw*I9!BDhj8doxd=jQ4+M@MBw`Qe!dqt=vz79}&{0cu8DXUN%
zAv$!k^BZ-udCRSsKrP}QX2nA!p;9kPcNTmf-5J?>vWy|AE(PB7?gYj;?u*+JkTB<k
zJ6sbzbq-dU?`;h7?Ax&tz(_Kk$>6p7vE$(l4abkGJ2oYKa>byY94JEh+mn(lGbz0J
zr%-Qs@Lt&MvUcJrV74}Vj9K7T8?QeQzTtgQ7r)nhwHJ^$i0OZy`|>>vTDR!8mU)Sf
zozJ;7TbU6Dy=P}gqSaZni+kau>|w{`$TaJz{<BHxJC|<yuRqB~regKtwLhV|{2C)n
z){txoZhF(Uo+-D(7vA^dB-NzruO}d<Njck_0??pp=(;%=awQhyobf9rjf81&mB3eZ
z(Vd(<35y6$<8wP5C*0ifMNLRW`blQT6B@0t^*es;`ktRVHP+nARU_ZW<g9~!e^+8=
zPf#?(=}%BnlG`j}xs;8psDOo${3kFnLe3&3hv1p#24B86-qa~}e7xGcWx{`fuLyXN
zJEVbOfAHAfJRG);-jv|wZ6L=mrH4i1`uB3?PlmU+ZJ#OLi!)<t=xSFli2pMk8uvvE
zoG38%JQw79LlfL<Q)eU5X)=|}A?8L8;`5tMK3-P$b}&Fqt$yYoZ}!>frvRo}iHSL%
zB7JiNob{I{`S@jSh~@8KIorn%_g!POU02jXR<8JehkTF)Ne#hzzL2O<kLQLW=68!;
zWT){MW$=<A&r9!J(tx-eV<&BD`l5a9WUf}S>q=Zdwc$__urHrX-mtc{G3wjBpq*G4
zV+JxrEz-P2cQ<4+7JF`c9vbH+#G;?B``Tt-A}^2<=bW@$xMbqw+G^)|yxFXfTsj)=
z7m^Ss7WvWfV)OxIIpKQj(lk5wTNtT|iSxh|MtcPVXzG}?eLtJ3+9>J}#l+=(uq9Nk
zLH?1Ux#gS$DUVvfL~{=a3=;@AF5jAEIo%ESLG~rN2k9Y^G_ZXUtbw_%OQvvK^!bh%
z3!)O%RWq+3+x?RY`&bV3k5^{|Ja`nCnJyTIA#n9gCVeF@g(%28cSKTf>nt&$**n%i
zIdb>^KyDI*5i6@o_TP-<#?K%V2?vLelNAHWx4hH-5|ZYVVD;sBo5VNiw!}x)CRz4U
z1?`y*aCZ<t#H(~*Q<@=J+N!g`W&VR8a6IRwuh`nzvDrp@YL(zjkeSQ(;=dn?n5*WB
zreOsu(cq9iQnSv{-5r|iQ9F4R)xO5LE)Wz7n;TRttLV;&#b+A!WSr6@@_hH&N|IGE
zNe`C|r&_Z(q@A+2!s4Udxek`f1`Va<A(<b)l}Ic<K@rQGypO6IT;D*_Z;!pY^Hz4_
z{i}DlPd|;Jk5+u=klJZ+c>Wc$A#XJOF6i{qKO8Z}>nB24VUFC;yzZo}A`UL~*;@`r
z%CRi9oNe2oGm|9eX*&@a3&&(vn?m~Z_yIOm3CJz4`6JSx6}odfy1ev);+bVNrrjBl
zkE^})B!xlTIXKmAZ<(NMI9b}IW1!Ubi#D~odINXqKfn`MN_w700{BaR?kn9{SB~M`
z7@;=hL%-d(5&V$0-Y;wp;+DGE8xB!Cua}If&Q!w=5=7+@RzBc_B}eg7=$mXHW)4@I
z9uWAC%!UMFX<(8AJf8<a688z&9V2O^W7pSge@%M!K-qF@XIK}QAbx?}UbDhx03w=^
z=$uZn*J5w3F>9QB&WAv#)yG%KpU1Oh#vl{26V06{R{0i>h~Qx?ORA|o#)3me4p(WB
z<?uQTvN30FwMA)L)8}j#sJ$5tfF6r#@q4~8yj6=mBBc|3KB~&R?X29D7|!{R4i+_T
z-!@_Fh0b#$<x<f>LtR+?n(hHz)cze^CAhM9s?jS2xaYjp?O{h>gEowE4(bNRV&q%|
zKP)aD^%K+@M)KN$9TNUv^0n48nmjwTX@czIPZh0zyPL10hl=+8{`Wo3h`xy~tDK9A
zi?AN(5E?YO6MvJ_G4TQ=@0X*947`*B^qUyo_2~Fp5ocs%G#FPds;|F))EY7#u@+kY
zJ6Ql&^mvHB>B^9@ey{_DjXoaiV(evQc~KYu6Vu-NFsYe55|1B>OnbnO@})6m|9k#E
z?3PR_^S`gT=lZ=~c=TT;psyrAkN>g=By^W3mHr=x7zaAJeQGQJ-~WH?a<G}S_Ws}P
g{;!({s=tqT(iW+k)vP;g|6mR^Wi6#D1&g5n177ZP;{X5v

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_024.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_024.png
new file mode 100644
index 0000000000000000000000000000000000000000..d682d2cacf9a35e16a56c93fbabb08b5db951d23
GIT binary patch
literal 55715
zcmcG#bx>Sivo;C@f;$9v9YSy!+}#I4aDr>_;1b+@fZ#d-f(HohF!*4>-66O;eEi<~
zo_p?h&in7JTC--?)UI8#s&{uk{XE@~>Z)>>XryRxaB!Fk^3otUIE1gS*MTT+U-ut&
z$}X>80&X%oZkmpkZXPBs7H}#iZcg@&ZuT~&)b188t~QPi+-&@8+^p2rZf;Jlg6!;H
z{@)we99^u~Vd7GIuMa_WlGk;GgTprY=L26NR%`<YH;<zrEurO^b@+6X%%GdP8tGT*
z6u?zC5)lJVDX2>|{2+IY()q4aGJ%H}K4^^%wJ=fk6glIJ==lx(ti8u*<C<wrwIR%u
zQ|A!FvYAX|>Q{4FS+kpk#Yfxh{JyD^rU$~u_2!O<zc$27OwmO&JIlZKhGJq)cm2nn
zj5>C+PVOkz+iqf@*8iW?F)Q`_^6*yHjOV}C=6e60IzMS4=YK6{6zKmSKZHGBJSMot
zO;3M>LZLX4j{y%kvYY6nD?STHd`e16@^_ziZxTmep2+DkU$)8qdkZkid}+e^T2@w8
zUS95N!p~3o(m~;Iva-E5QHXsh^m^lih%DdBF#msV&tmhBV_!?i$jIKTi9K~ZF+CQ)
z+^lZy934>S|J+T_V!mhzctT<`Y#%ABYVEE`?$-dVw{_Bqqewp^QA3tq@DL)UHct6^
zh@)>PX@o&&ZUp%Ic#1ZNeJAKeV9yh0SRq(i4p$LSL|~rYWlpm4@J4JG4o5QIJS2vP
z3&ae-mCO?cp47~kR!}@w;vnpf1g;{-%Tc4m0)-_N!QgM=C?v;nEMP5p04Fqo5p^cR
z>5EmDS5`yO;J3xu)%EU&j(C;ddpAh>K5<5TJ^U|2tJ?!L$tbv@!}`HvCp`kuu3;lZ
z7;jM(Y&q}R^mnUa8`*_o#HYEBV$b{I{@7emt9Pp}axdRpATdcM?O>2K1kjh{(j;AE
zR`f@{w3}QZQr+8g)y~7={7CF^SF4a>=LdaClw8V==Xp~{*Vz<1UfPg+zBpQl7BfU$
z55J$GZl>|kE2dWG;B*C^^1wip%H+$@V(6O#W$YvN5R_SMGytMD;#j|6_%JIQw@%u6
z8^yDKb-=A5g`BkBrhyI<<8bv14TShP%XPg+gwhT(U4cV4%9Zk@XUwJohh^wfTkrPG
zCtrz1_7f4$c<Qw-!Qt%8hP!!vN7v^i-RIGh4y8igJS)DguF}ooyZ)Erjxl_-tnKaD
zj%#&3s&(Ix3tSks7)*{&EQ{~OmEvzy?EOHZfNGY`4(y3c$EtKj_<esr@T+B@Sux`P
zUSIDVP%z2~s}(uz#lPn4T|g>`7A{iCGnyMPv+#J)ZAm3#L2F1#3Xg)Wb|8lDPawJ0
zDxnxZ1BaJF*4|hN&H%FDr<jId&4MMV4Zi!DjW|t4v+w>HzXf@RJU(6=lA#x=TNA#s
z2Vk*N+hOu?13$IOlS9pk&Gv2(E85DWZOQVfL2<T+^L8grS3^ju*&^{X8lJk|;lsw#
zE8#Mx<unyVmv2#RGE8H-bd)T)vRnx$)}Co;t~WdsxA9V3(&gp$e2+JQKlIpe`#JUY
zmqhP;Swe|S9AX9=z3<GDy0B__)r!dV(CZd{)O)1Z9=+H4X=Yt<u?xOa($A-j1(Di)
zP{`OGOI>S)cxDB!U9N0W0GYzr?EqJ&c^bZicKu3$?+@+Cb(4}T-(+$3d}R#j6nxY8
z^p)w-$tYT*Jliyz@PL*ZSC@|%p`)?#52q0`IbQCI{}~2yt{ff~F(_z|Yv^O2qwj?b
zn5EXmH0IJf*JvugZliawkd=K*vow`zVTUd&-bu9kZoU6*U?+TPC!kWwkW=*e6w3-m
zvHKJ+TC}{py>B&kx?Lp->wXkoe&)+HY8@=aw<-Y-6Vo^uh2WPuF(a)gUGFMWo8Nk*
z`P8jXC%j}bt6p<uP0p`27HRb$U#zU7X6BudK)p_B#o)<e6s&nYY#gelY>$350>>UV
zaNT&q*kUy+RjG?4^Aac5%Td_@yG{hfHs@71ayaY3HFxNW@~En75IO1zJzDcd{#_9e
zddQ~|`Spr7NJiGp%R+~T2C|>K>`vce_I5kJn$pPbnZ-4F*z*KOR2n77V-qg=G>=YV
zko(dl>jowN{K5BDkT_5}0w;D{4_77wR2mYeh*C=ZLfYO5=~N07|JBmgl~R_AqZBT?
z<rom!SzYTI>fjN7-%GB+8mSec?qY4WD>5k*g6O1^X;OiTi?e@L31^no?~drz$xKCg
zKrx99*1Ch2q%kM=UlmF+LQk&a(9ImZ;+>i0C!xx<G<8ori?pd@gNLt!3Y-E+sb-ka
za2?og_E;mTWg$2y0B0kU7=#h0FG5r^aS2^HD-;H^Cd)yc4{Ki)*C!^EFMaI4tZvuz
zt7(*toQ|&g#q#^RB~rE4$hPqf(pklyb+t5D*C&PlzC+pl$GrZx&Z58W*6Z%xQ2c$K
zK56TrUS!x$tCs!eKAnz+VdDoOW7G8kL=c$R%||&_ZHvhy-kGiZHBnDDKGGmX2qTg(
zqGz4<j0#7Y7BEW=ISCToZG}_`%WMP(UgR-%zpuNjTWl$>S}s_%n_OCyGzEU$V3r#j
zrGw<nHg?U-ur0>2)uPJ@`2E#|LW!Ix%ke#TPrKKh<$F2ws6()22@!T#q2?xd5QKb}
zcqn|6heTpCt~KG01%dd6bzQyK2lHUY1(`dhP)y;ZpzIy21_vXv=RLGZ^wzIEL`F5$
z=bvyj4o|xa>tugo$Se3M#$|+)f>;!tBS8S#I6AkYU_-F1jmV`pT7#5)pcWoGVuOQW
z;qsix4_aK&#|X*%Y-%zdYu@8pmveyI;RK89(6@zz_H&?d>!Tr%Lq3w|TebO&NesFT
zU+JkQxxXjoiHNF>Fy5|~JT8m6ySSZWupPhXXRA%h+GO6HogeAhKe<FW$5tGYhpApr
z*m>|ZUA2DVv$AD>w)bUR)6)^(K5z&q1QbYwj#?ZG2ok-a+?8~b4|@WXKRlj9d<(*%
zK%4zKv%N18`7TGIA)3IV^}X}cg5Pua>I1;;;6u3v{S(*fb0Ky~%FJt)*#3`J&LX1s
z!#4NyzI-s$K9N~!svoW(({H`Fma*(75^#4zWsp+UpWpEOAn`5jPRa~dG6Vo#OXwm0
z4bo_zI4L%Z5<?ynCUbEVUu%_GQxN=8X>kAT;ihuKKG)??;1|c(<B0HP?N{4+mfHxO
zzDqqrQirsMS+VsUD<WMT!NEvwH7-tPde2L)Rjp-zBf|vOFIYtpL1~3OTK7*xle?Pg
z$RL64f{z${h4YQE^inL55#($dKn#RP{xG}Qs?+YYc13zKb$6f@PA9G?QC37`h@^b5
zoJmZNUNq>h#K?n^T-V2ljRe@F9^J<^Ji;+!U)8QJ;>=uWE>HJ_t-L6RSmF4jxJ}iZ
zI621e$DPR^=}17Iku@EVcRB4jqJ+b^lp_hnh^n{3JJ>Cps#}Mh(k(oEV)Z|<SJ!wa
z1F!%5B(&YOfgU7-SuAi!^vD!laz?W$W@bc963`qBp1y5w*Jg>lV5E49kRycjKAir#
z@eH|S&VR-jql=;q;`ct)?G7`!kP0MN3(qN$RzSw9L(9x|3#xx0dmL;K{ZV6<2>Rt+
zHAa1~u@4VWSk;VNfASFXvCjNg97fp<r`n~<q*Bck8_!pjkDMVaihx(x5d?o<+r)O5
za=4l|7`r{(;5G8!2hc*)CksOcT-OC~b#9+{qH8_LyIbBw%hOtf%uK+K=^i`U!`=ot
z-4fB8{1RGzw|dvJYUX!^XQf)@<Ytq|qhxb<6yzEKQa1ScSw*p7RWe0*S})zdLoJD~
z@o}DPsr`;qPEEM$)tnp+PUI-jUAsmyD8z^j(l()5kSCR;qyBIyL1=!+l5x?>Zm&&z
zU-A!;j95&TD|~B`8V(Va+KILEeyUJ;`0Dm{#f%~?s4XxPpY_fcW()`=L`ds+mbMf}
zd=7p@;PWXZ*us62&nrpT$!-r({|RUN?4pUg>>hq~^^u-NkgjfHue3P*dsbdsOSCh|
zDD<t(f)u#lbusL2o~o<xVg=K<fQGg}%2P;90(17=LrxYB9s?~7s=j_h)7LA68wbDc
z&*KH43QaZ@^|BO=n&PR|o?&NjYTcqCAAf^&9jUU)p9zhTg?mA<=k<<_p=q_d(bcEL
zKeox#o$F!O%(t}Suz55)Mx5s^fyjCRlNlsO-XgK?FztlE=w0w+q1GHhIk&I-8EKgZ
zmw&2}g?(#Ke{vHRbz(ut3}FcPQh#1*(Kp0Z7D%=JIKJs?e$iudfh_ok#6&$&HX0>_
z5fXnz!e4c=VK^64X}Fp<Z^I=d5bR{y$S(jNm-Toc;Te4TmL09iFuxGz9Fx#O=AI#e
z7TrJgg0#Hv?$jd2Wc=iOB3x|qVf6)<JDD(Uq+a(1kepmQGBye-)sEc2g9_rPmueBa
z386Tael#lw(D_Bx_$)P^YkId~QU?4T9vRJ*l8?OstPEG9oH0wBo@cYpFd@~Dpb(18
z;Pg4e1<f)uNbM{KIs=~kGm942x2v8g7=P?KN1C`^kj}FCMS=am<8BLI-Uf$6Dj>75
z{#vKSr7V?hKjk$d7B{VvDj~2UKSq}y2GUE)$x{@5jG2s*x2t1rd?Q*cyp<Bv9qjT7
zp|jn|80aNQwG7rHciV@SmTa^DKbl&)kR1au{X2qxrfU2ONnSU!t?0i~kNpm@p%fXM
zCzPaa9g%o5kRa<yGm!1$M@hr8VqyQqXniU^p!>efPI^e=eOhVi@SmTYp3$c{ijW|p
z&*gMcO3KQy_-Q;ou_ErjXZP!n?aLc(FDJ2e^#owtV}9u_khpOKfSYR8r>^%GftiGo
z=WNLB-GT`+7Aa$8bZ4Hp>mg`kv{8y0>Xcw~-@0Yh<g(e>vz6NV2{WJfYitl~$-+=s
z8B#h7l$#U2om|(Lf8f>s<<?m27s4JPIB1)AS^Uh#q@)&p6@B*$Y*YA}fjsiC+Ag|^
z9&dEmdX@~1>OS+8SBDk>2q9!}kY5vS*HstZio`AR@aRizyLaxvC-$0#MVfsqppR|f
zHeKx>J6>pizlI{NokB~KT*KlvtfHlS*;vMon5;^JMGiciTwdv*UJ175)%H4I3h3RJ
zoAG*_XD<Xpg#u6ce19|F@~`KhnGvn)9<YMo)6rMrE_*baT|dWKC0ugOf2-GTm|wPa
zax;nK*~jnal)Ri+cm<}DJTz>d_ATBPb#8qGMs!D`36mcPt8PvEAQ72<Z@!O?4-WZC
zBS~y{fKt(Y5x$G*4pLml0S|vnqVM{FT7yNX;!K{5`L^--TT6oj*B%eY{5OfoVpCyN
zsV)sJ#-3)L)AaH#-LOgAK3Bp=G#55)d3^aIb@}}ow(X66u^njr&1t1llNRkh_V?V(
zUK8uzzneU&hOKhD61N%RvUT!}$~4&0n9~%c^^z?}W{S$I8g$w^<k&e4TD&7&nqV7~
zT$-A-e{Avdcwphu(w_T7>@UK9X|Ins45r)H%q&p$#fR3e6a#zd!e*VztB$H`W=MZ*
zovyWZ2}Y+{6ve1>1^G((?s0Yt{0nK#{TpcoQ~XVk)(8P{f~@o6>j8=7?}kH+`kgoL
z4Jk!D@EG1q%z2&vO>H$0yOBj-i5d|2AT#guox-_ufM$y&r5m2{ptjaIZIPcn-&Ca@
z1Tl*JZd~GQ@0HlsCqo*`r=bNE=%1S-c781+=~v8^RsWEbWm}rLCd>EJ2g5HrB+C{C
z9$KfyvrWz!?k+d9_b*LsvQi1)(_+Fx6~#JtDWMXBZ*~#-`|yp>W-N59PHn{q%FLw!
zY*PnUr=shyD~~g-f~NLPZ98-ZUP<xNt(tTRIoxD!*TW;e2)B|k&Kc?u>Y=Y*0}#?{
z28wImZc2CZR#-v@9ik*l7B-^SiHM#}t9ntoiH&6(b8}#?`fE{?56V!=_3MP)8uI<-
z$7Pd|4)d=XHZ_n1Msn(6`35^HwBsOnMnN=+PSBn;^Y=d{=vo+I*8lbozFE1a)WiCo
zMY;RD!w1vdNcy2&hPTgk)okwXPYEGhv9x9wOnN<ZntRLWXO6rH>)OV~0kin4rrKuJ
zHd2iW`I*^D=a*|j)hyxu%WgEmpKH`<g-%{~zF7Xg@=O`k*Vw5|QYF_(pbc=3PqaD$
z1481wcYo}`&ED7G20ggm1p+52uUKZ5^#lR-?fi;=`7&J=l}lE<y+82Fvv_!O-yK?8
zs_ZZdLwD6*AO}1WLQ$VUSBg@Z?-LmX%eJkj(=MtcH)Dj|{Uc47QH7Ka{M@Ov7l4D-
zU=}Sf;oUk0o06_^8_bGml}^Q5>)@-wmmnQQN}-1Cp$VBCBCC>*@vo4qcN+zLsUv{O
z?n8w(`Uk>k+Y|^Lx)uws-qE|!9l{Vda2?{CV0y^>{Cct2^Sgk}lSjIu2wYhj`)6j_
zQ&jP-=!vb~ZJtxyRyZ=rnH45t)`*A_J-NQ|qpNBd?Iau4=u-=sn<m*08O5O*YUiZ0
zGn*mE_Yuny`9Q2J<aq2~Se{%~94i|k{YsJJoPyUSe}p5v0`3qZhfydtA-$;aXQ?A3
z)dJXxcJwA(S*PZFg$8V#e4?@1NiO8yP{!w)l@<(SqKG5A%Fwcn;!6y=igP5YkVVlE
z4NeefvJ+snd@={1ghU)cWK|uO737(e1hjkxSn=alteB<MiD>c_Cx?0zeK1eocW`<-
zeoTDq)QQI#Y~o9UHQ$XjZzeN@IpuxtHcMQD91J$T|5e}w<Ekbi3FJoFQrE2FosAW{
ztO65z7LHabPShZuOpZGx-{MCGQW-%j*NoJ0zM)4(;T+3Zw<hZVoK_~OE*pLKMPaBR
zmnc*}Yu#8ozu}QG@TP!L&TIz%F2nxG<*zh{98l~LNHzcHwD1^+^|dPC2-pAgNWaO&
zH6|fR4p#Bm8`th8Q+X_9&z;rd*uJJ^*3_mZ1E1XQxQ#6910OX!GK*<BCZX|)Le`vE
ze)!P{byjCnNJ|5o>)n)xbpu}wbkZLDUGN6cN;P|JubU^^A^B*OqIsH9(e{^eO*My{
zK-F>j_wRYL`-u<|o7BYbj4-dEX_ljwvO0tK=_{4gPud)O>$TBTzuZsW^{a_S5h1+a
zQjcBPWR7wJ;{`s@qK4Hy5ZbF|RG%qDPI5{T>1iQAl%$&_*}qT^RP#yfiwoy6t8t06
z;Ouk?{bGfHN_T_8ABJwR0<{Fe7W!`_3ktqBYWd?_JWu8qd{>YE;7VHo#jOjA`df+v
zCA##^7QV#bc12fKR>f(-@{F%XRfqnJ02%aWP|hG#CQL%qcaB*b?e}oI`~_9z6Igsz
zsUZuGwcQ49E4V||ll1u=o%c87tP=}1BjUyG5G!~k!<<^3<78VRPjlpL85tQ%XUf2e
z3^AO*-T6hE1dCMn7KM&ks{%ggxvO01ALu?N*Ua%UbRH2x7%Y)?@g(}5X^ci1P_$p2
z+w$VH=iF1zeYu17jI<Uaj)vM4gGA|mqqo#ntMZlghxVsK2Mclm!NAHtUN?ojMG;a9
zoMBZjL%IpYcVaKb&ujVi{}lxM|2qhf%^oLS@+g863lJ%7%~EImp8$NbWHGU^7ZqZ+
z2}h%R9w+v{IQYes^%HgzqZQ_63j_xY=f^$~y_TdzZ?_o^ZYo_jhveqUr5b8g1m(XQ
z17$lFsrU6JC2}h&Fq@b8F3>NSA%GNG<w+}dlN?}NkCYs)bE)JGYs=_cvV*e2<Kybu
zSqn2i@zQc#qtYZyp3Ihm)`wau*YjS1R?t8}L7z~kc#c@Z-48Jg3Mx;iI*B#v-#gBC
za&4l4Ad4SRE8^Vm#S`-dt6#=#e<$F%l?y!VX|l3atqrIP7Ob9Vmc`cQwv7>`rhJoJ
z1AwpJ0G`&TxE_X{qo|ehJBLl0d^FD4AfGviUy-Ib%e%Dr^V?)K*M~_#MohTuY!O8W
zUzfRGj-_j|rAN589|8@5AX+O;+8dJ+`eQ%&NJ1L~tBbD$C|g`aQGfKp6QUYgZ+7B!
zjrFZeyh%%3cKy@LKkMkQ*4EUkSc*hPD=Die<0;7G^W@N@+9XxCH4!dwF-Q=&x|${H
zeIe(!BF69r64YGVAD82&DQUg`SZRERQo*bAe%HwhVnrkB<s=UXw>aBMWt|xkRqZ~}
zJ!tgtd37(o`ZK|fv&FG-hlcOQSsTTU?bwt>%=yyk5^AnjLx#^QVviO7vD{{4gsM*%
zeYgHtz-qQb#N>HcBj<Sz80)yBT7fiWu4Qdb7V~>;oXNRJa!?03nV?AQT(Np#_o?5d
zy4sy1d$4UCNImzge=ulqgEvr2{ei+>GHnYs#a!8F%Qu)98BwM+hyMNh#nOrhr$Dl)
zeE-I$*E!}UfgX^o`8AIH!^EMRH6K~sf(J>_TM`L_GaezkJ!WYN2g8RLo+JD;XaWnX
zXS@vc%^k0(^`3P_Fh7_5)n+ggqO_t$g*NdV7z7Cd2W21WA0FE$G&2r%;_h<=i9$f>
zOgY{;XOWx>*;pXO*KLa1wPyhOJj_)^u7!~Oi5`_s<1aiQxx+qC8drlnUkff^jdx((
zil~%(m*U7h+_o(Y#A$XVs3!VEe!w3PqtfI0#j!%e)^S{m3^*${3k7^+k=BsbMXKs8
z61_RZIt94QhXvKKr3{?NZA{#S5p2;<7L8RL<Ze_*^ru={#WrpXl4F3R%9hP+yh3Yt
zDJ@!!tjxZqKv&o_$Y~HSAN)F#e!RJF6I~P{Q&v^w6ycR<DyW(H_6<~CI&DjAHaH1(
z0c<Zo&_~#_I^Tyiy*VMNJbIE+xvj0oGH>@He0+yXYXl{p<k_CI;#bfNpI4dFaXfDc
zI6|^#Y;?cyj!=wr=}hEPR)rO8y&mPCV(v8$^u%8zj$#*&Qd#o@U7u0NZZIok(PJ5t
zGDhg#9xyE}8PoO4L8U0(U%F!L4&0h4pcW$baT2DKG#1u#2NFA(%}JtFGnh;_I2-g^
z%SfK&!4kR46(6DfJoUCD5X&z8qJ|~oww{_kFUA9Wa3LC`M6!r>$T?=16-|Yqa^isg
zH3mZanTRp&4VryPN_n{P@y&yANOOqUxp{T196gXj3Jl`pYv1WkLbNEyh+2kG`PV6<
zi8N{EsY5&Osq=I~j0nBB%>kY192b+DzEK4+N>Yk%alv<V6~tt@J!YrrJsa|q7}DzF
zzxm$KNRFuxb~oiCh=4=iW3qVPjvDP(Gl!07s&e_}1SkA(!Gh3=a}02md)$t!50=D_
z;oA_hAcpoWP_lhkS6XjMq^v_W!TA-UPMV;z-n}Dv>JOe-r;OrByVyd~Lm~0!(?XU=
zbyt|=+jfyo!jey7tJT!Bp6~jOm~1|xtXy5w-j&g&CNLQjulAWsDv4f8k@xKp8(XcG
zBBv|&#R!Z%U;7p61Fo+#mgaRJpg3YGDr~z~)*+L!#plCYg>X<)kkiA#+@dKB(ZK??
zrXAZ0Q+Du*kJ5xr^YW3?(}G|(>QCOTN68dsyt)h@^Le-rybE?<7T3#L;$2GpovCPk
z%KycdW7m;AM!#HBfJ*`|nMsl_2^`z*NYq5G5gQyG)B&G7w?@!%eLbVE>J8mJ1eH}|
zuzQ+j^nS4xGGp>VF?w1%=7254k|>CHW}5fGj86_}$;Ff=?%w3dKMXot(zUHR&zb-k
zLYZzYeswIIy6Lo>HYMKsWP&~-Sq*i+49&}{6(`j&D%_*nj`Zb8Smu8)7JvWFJY~zo
zOi^3Sq1rOF1&uurK66`ZL)m`=n0z-HhBT8Y-1>O;Ckc$e?y^jQn2x;JWga$SOOC5&
z=c_HLh)j?ov1PLF^Z`O}Ra2ACk8^K#oR*=qAMOh&_z7S(5?f-<peM>hpVju-+P?UO
zZDwX>4Y|(sE6M#RZzS!hD`_1&Ov_cqncn3&;QO75Mw}LR5|ei#YP_D4(Gx6bro6{X
zx`4kUhQ8G39IM(e*xk(v#i(Yq@U_RFq5b`vA7}D>--?QZ*C#2mhI5D^_7_%~o|~2=
zn#=;fG#6B_-y9NkbTfB!s8RJ1N1ku?nA@nz>3&JTjbQ5FrpaqlP;#1+5u*IXy1|qH
zmom{Y6-mu@P7vDEx}k-c>*ZM716^PP*{bFW$Tk_gl5XsUh2I-iTBR30{rQP2KephV
z`LdJSCe!h7T2S;qWD#t)61zhI+U%W7<-Oq<+uwJW$wI!DD<7J_#J<DYrC(Ns-Y;M=
zsj*=a$tAYm_R+YE8IjXdQa%dMjO(p)0vC&G8f5By2`cEc`8^^wNYRJ;OAfy4^H+@@
zd@Y$SPEfcNwtMS#?asjmSzIifx8daEmTM6E>}r=*aYRuR$1yW$%R4`>Yv$mP#XPB#
zIpM&stKZUp*Z*rxAflD3=sk!QEsMkjNjcS|mRc^aumm`OP2|)tv19xgEw<5?Yq&G7
z$^FMpM6j*RfV3M^Kp}L`3`Y<Q-tfz=-q_x2lWR<9z<%1h*01r!JqoVFYK`9D#twQ~
zTWz}mZ-Dkb3Mk7ZF5;KFVkCt)1bpM<JgVI6@nEY>dE|TvPw5A!eF8)@_W2@RImDl{
z<;n#TpwSS{c(*ou#?<pUrBSSFc8`CBXQez61U6ZNjn12l8d`liO2yfe3;Jf2iCkj*
zDixp~O|^Z}epz~7(JD+^2O{Wu6*T}dZbCjW9}lvC{;8^|Nv#2i)1!tkk#~VNJU73c
zf5&Sv)M@u{JnYVR)JULl8tfwXKd*~p&2}#Q<y%$_bLITRw!AYnb_^op7Jhm})vMqG
z&8V|<`>2(r;At%${cjlG<=xQ?KNDu;2x2D4Fq+&gcJx>7U`Y6#+JH3yeBb!hH8#!2
zJBk@iWVsc)APeU8Ip8coZHfcqK*ST}s};>Mjtr-g(s22TLlTB}-J%K~-|xX+;IkUk
z*SxK(Yp{Mh=xXm!LG#VUKt@reY>`??Q#s>TTRkkhg@2F#W^QG{<V(e<CAXO=h$g#0
zFAg__872Jsp$Oo%=vQgf*89ny;nL0(nhbdhFObVHJ0X!kB9&@sKJyyLQO6}ia9B+?
z+uiQa#9qbPrIx8KdJ;LPNr9z$qMQPSWSvoXGYJtAXE%M}tPRaBk-FmNzUNY7v^4yd
zH|xGSrcdY-2^|B>tvKoLO`|1AgyM$~RoDXSa~z6g>hU&x-?FjFA9P<}+yiDvUX!86
z37g%A#-_yaU8_e}%5dxbko{St&H)Ew&v%Snk;EH$Tv|NPR)#eUS%pU&CHBwf*v-Gw
zO0UOY*~W7w=lYzyiSNn2=Yd?OpK4hczwbajC2H%CG5D9hKUIxOs39WGtN>`)$`|Px
zCb7`-=E7AjLdoK|av6eW=e-<qz+YL*A*RH&M-%$l#wMK^&6kZGk(qGP+-nGnMiluJ
zr)EJ6&WrsUPJRv4<zPn;76#U@GIJ7l9LMxjy4UJM1Tm9KnD+Q^FWFOB{!eOdEgzQH
zM1W*cON0ySp1G#1DbdunI9FXk$)8`Eqf_8g+`jGH(G)fymSSJ;lx4?Bxbc18>K|BO
z@Dr~O7y9Gy(BW52lSGx!TJMBly&?sL95rt-l~<G^M7C&H>md0u2}?HvPeGz%(bCE;
z*xf4rprP{Y!b6Pu?vgZinQsQoyN%I#o?Ady(k^1-d(9bFi_ZZtEk6a0NH2k)*!`qS
zNQl6IkSn5G@Prde`%yjRdL@}eszMYh_T2v>S84PVBla{y<>Yb12qBaqiBHC}qfua;
zY4+u8%y-R88Br+8MTR7;_Yu=GT+854xi@Cb{j4Kx&w?c6UhIIuD1Q>D2zThE>5~4K
zelSn(5gPtX-Pn=}G6eR$>^MGugk=e+)$_?(j^Mz-1Y&A{xu4#>J9@2QKO}Co4|SZ1
z&b<k|AZclNt$ETkSnT<r;)RI`Sctkw`C8pup4k_>suz>!W6t-zHqgJCBO@O`rPG61
zOi@Z&O1$3LZ_`P}PdpSRMfK2arv<O{vRiJBf7R54&0v{n761!Cl{6t}=hBCcF%rSu
z*s~3n52G7zD7!y(HNYt(ncF)Y(u6?KyW!=5ldO>1<t_nrjSvFCQ<Yx?0Gdf`c1O;C
z_*c5D7mLn+H13HV?15@Rpp|3`Rz5CQ0##dsCq|$KE1VL9MMr>-3d80La8Unz^}W{4
zjq608bjV<$r(`Za#5_lDaBgYN(xX@b0+Nl0@n@%rGJSPzVhqsl-}=fbqe;YbhK8@a
zM{UWlG=xPlB_U*dw-+oWL#TSd;`}vcu$1Jua%B_tBvQGwWa82+EigE)8$qtDrot&8
zAX7vSGhbuH6Qfs%UgD6sb15p!i^?sw$-4MHLLJ*h{7P_cpY`W%9nNzvHTh*yEjLLu
zydUr0v<Mp%ZgZf@d_NQ}uw2zFHP9+&e)fwX;S=h7@J6_CC(tvyJD5zH<K6052NR@d
z0Dwxv0S(z55H20U!D<$f;MNen=$(4Jc6?#e&|*~j0G;dX8c?oGrk}X8EOasbLY@0I
zkI05C6dS;~IRWD06^J}i;UvC9IL*#WvnKUHwNe{(vA2mbQ<C#~9PHQC6&&1}qrdXb
zcB!YFLCr6eaOC*x=NjAfS^SGVxpZX3(W^uNo}c$BAe-I4!OMunS|PiCf`|Nlw}J|e
z!z#B4?CgS9`5nBE1^9Y=hMIL{Zq7{d+3WCDc($^9;s6l<hccbJ68@Yk<Jir?{th5T
z6cJAK??i3#@BRhsxd+})vGF{Vfq0OFpe~GyHc9!vU%RRvq3EsW3;3QV1IpuGcp+I{
z15-jptSD)HM0XvL)%jkzpArSYQ4<=Wv9XM=T2UpJcCL67ysPU4cmvPuO5wLhCQn97
zFDc_s3vHF5r9=ir=X7nK_#GIZP>T@C#{yyP-MK+I>juF(AE6>jHp5j5pW^;rPozu>
zdZ8~|Q#)r?{+N#Bx~Xe1s4HlApNP3m`iaePgt^iroL0=*u-nL|FyOO9y`ugUE$O$#
zMiHiuyFJ9!*uo`VU}citw^7Pa3GDji<D+^OQQx>zPYT+fub3b;$-VT|VS{>3g_%i!
zYKl7mYG2CV*yQLmJtBmUc@iCR85!yoOaw69S4%SzuuP&b=o$be%gtgiX=|XQ>gy2g
zA_TNsaUk$8KvsT9H<Px4F-nnjglaQ8_vyPlzEK31sZVjZbgid()Zidme;n06XN>q5
z*#>265waT@8Q~BRQTl2hWfagkv%u2kkZf5swlMQf6MsrMMrFgk`*8_jS~1q8bmU$y
z_&E>RQ(IMA{bvYH{!e>(T9V2GQ~WFW{0KLBO*9RskM{CnbK@NMN>yUfc(?)g%h-E$
zt`$s<&U56K5OK{^gxsbzz&R56-+$L#QvX$V?Wun%lWc)E3RVx2pVpv--iodWIpx?+
zUmSFtksCk%tpyKRaAlgrO4y`dNML1|Uc`c^o!rIvF&_j4OnN?15BF}PmJ^!Wgn6Lq
z?xbhtnwywVo`^U;Zz1#e08RofX?E3eH<kwR%gOakUEE0_1euw325oMU4*c6YPrRgU
zSm?t_BWkfV)z#Acy2S+OTy&8IXennXo5IdsG5u6NQT|m#d_KG9irLnlTsK2Re3aU6
z6?7x87V`2ODSSrxgoIS$a&=5BEu*&k7&a6`wL1obMq>B61l1ZA(a^iXajL)w6&K)?
z<eqo3MnQGol7YqD-fzISw0u6H+I?yCpwJy1jSmb`KNZC%BIdV!gH(${>nRyf#u)N>
zIeJr2Q{%}^EWObV7}6?N%q=YutXyFU1k$D*6Fi4LDJwp$%O_P<zOp#JDc=Xq=L*%%
zt`}>I6v#dxJ+;e5hsZ10PLoDQN8|K8t`$4hI0wWI+GZyY%jxLrZ;#wvlxrQB&wtZ4
zi*E7IZgRU|OsiVmYAVh(`9*rMhab<?!J$@RWMNk;kqV(#kR^}<vm`Sc(vlna<|tA(
z5J_Qc;joMpvDPhI_W`pTbjZMCOi<|E!fcKd?JbO|L#l;{Pm3KxPb)%;BEQ?aOY#ez
zc3j-qgw#)p|CzSO^DFh(r|j!+3Stt_0CYvnGzL`rBF=!#_r?KH?0$D7wpUz|*5oF&
zAOe%YfCekw5>9jym)jmEM!YKec(Dr>BKFpv>m8O+_guy4S`ZsIFK^$gyrL%jM35R0
z8j57e_!q3T_hkL5w7$x2zoX8aHbw-rJ@k9i^Ywl@?N3FM>l>N38s+PyxOtPT2>qNE
zJjh&BR;j3NY!u$Adi({JOUwb}e7VK@R*h4{<m|LQrPkzS6|2VX2Q}M`*R0S2d{lIE
z7b#5ypT|UBjN3L0a^Xj8d2Q<tY%8@AJ$y*>l@Arz9d93(75={X5*<&=T2)Uu1Pbw)
zQ^Vhv-|(w8%%4xOAUSduy3v;G2me74EG^{02RF}Z^S=t4WAk<%0$ztG#c4kxH&L`I
zZ;+(K30%paZ|`h75QJd@apbx`w6lw98b&uBS=)@TAr#X;QLaBsH(u_`Xsc#1%~&X=
z2r|L4ERLQ(%*=jD&(!brl&jaXaCc87D*gey8qihE5e}s-)c(MoqN2f)wvUB(-v(dr
z9hrV-K^btLvIg0ad5c6fgb!5j7-0-m2?q%4*X<`tC(0ghRRwI@G`C>|+%uif{Gz7G
zGe-rO6=_k^$HjR21-+~}wvBY$b@j~u2g3bGB>v{D9cChimodXIP?K>cgUATjH`{)O
zcjZk{L5~veg0?$)S`^U9Nkw|)DxCRw&;-(7U9N|g8lSyB+;Pr-`xJ0C)_I(ljh$&x
z+Ph_rBU5i3$MzxFUC+}>Tqh))^=Px)!qPqN?#|2qVzynPWx8O(T0JGKY2Kz;AuTcz
zuT+7PkuGlN(1FA#;K)W1qIj$zcPFBB>earaD||h6iW|mW_K5T;s{d}(*bmQY9s6Z(
z(ca;h>-FWD8WZgc7keUEmPpfipM|M={i=n19v>To%%39BhV~vj-M#--lRFh~tMa@C
zEXl|wv-V_H3?)#<(IfB*5b1xFxn>WJ>>333u*jPYcK5AvOl@5<Mn?@<oO@Rvb6>s`
zg<Q2X)Yx!4_R)qtLh_3@4LtX8yBBa;M>n3qc+$-ygPUs3>pe>3`fO&(!N7OQ%HZ$W
z>7OmULS0Is-H!-kSz`Tjx4><ebe?_FBud78x`OYDknV}XuOWYgwW;LN%m~8l#|!7b
zfGV^mV8JOm0<eRQ`&nH<_qRnD0Mr0#`hYd3)26U7smvp@f{1%&hz8mz?$tf9@e5Rm
z=lc5gOic`$!39cvv#SE&B5NcbHSC{}-ON7{`3I>r1<C@&djGAhbWK+A$}|AZM5TUm
zN$SvZBJ^*3-%w%~f!*q$0?0j)K(r}cmQ7K0i1DLd0)fWdOQ#k_Dop4qB}>CZ2<$;I
z-?TD_j33^5j;Xa@tZpW~<xWoSr7b98(Hi0{Vl^?1CeJNcwVkxDpowK5xtsve#zG1h
z>Pww}Ffx5l&o;63j&bOaD%G+m*DF{)%xk5gzySI`@^oB{c7(+Z>7@BPaAVKo5z(Up
z=ZqLw-7}-XintlLpMLUkp@UFfTXeiKVhz2sDJaDrfVpiisM$sR(xcSbor-F<-j~fj
zCCtxRzJcmNr-DwNR1pvuNh@|xxCxC{`?vCu-AW=)cxni2)}=8X$Hb{=pd9KWSJk3q
z*&OJxg$N@Pd?QiH1>LlwD%LJ@e(mOo9JCP*8u4bCCS9(JZHOBQ&++T|b!LLdTvUP}
z@sV13r4~ohj)~-A)baVN=>4h4qy*9a2a6BHWvPIapI|#)*ccm!h=^K+o(x+2Fvg@#
z9k?i1lrSH6g_E;g+`<<@Zrsg<EkD=G6qi$obMTK&Wcoak(aUJZV04xL8wLcmA#}Aa
z0$j<p0e<#RTI$F6b|PLEwg`?gjcX)Im?%hW^13ncDYuhK@y^?u(MwCN@E+Z?DZQES
zLkfS*qW04V|0A!S<9koAYP<-Ayv5xW<OBZfuKoUnj_){MjY=+sJVMguK99d?b!q0|
z>HNtwc**!nz;!JW4VdHr&-ngt6_*CC{qj@H{yu;j{3J{A+J01|OP1`$p-WVO$aq92
zEYDue*oV~&*1}}x{RyHeo0(f&?6Q29tdkWg@%PoII&L~1jX4_{iCNUM`UjuL1rB{C
zLitoY715N}+Fm=X320iEY`;d(s<dfCrsrssYSs1F7_#Gflv&C>fK;?^j#Q&-B5mNx
z>m}+Yjv8nI$N8QUh}pT-)qjY@^WM6!E9Rw8E?{FRY5>b?y$0P&M$p&2qAzH12C|Em
z8%j7MU2|B6aXkOllP(+OvT`$1i4HILJ`o&rBcBpkZ1r3&2M&yJG}G{2L`Ofi1)O5v
zRlUr=q|jdh(HS}CMv}1b-dVjPuF_IR2&uh%E25m5mm=xOVIx+aS*|1)kpLlOsZ9Kh
z|BDs=v&}y9wo7bNik#O5#8_e<ZHBn@>~8plXr@-|bob4{+x7F&E9!Z^o4hm6b&c+P
z&EUDlG0Eo9u!z$w*R_wGba7TWB0b$3y;6%l|63QGWjhICr^|X%d@64$^jSlY5wG#$
zI=nm!INF<ze>+FhMo!wR*HU#PsUiI9<P2-@4>E>ckyse1AgBOB`&0Go&#-Y%mVm^9
z-8MS63~_L$>BE=b#|Fh(6>EMtT3V9$Wf9Y(5B7hCvIMW>*96f5xx@S$`txRciU2E8
zkl&pzmnY9vUWcVb;wa!x0{rI}nf8h7>=>U|{d1M@;XH}tH(q00qi86>=U*Y-6{_U@
z<};=tObUolMOLf>2AJO4C-`Y^iGrQZVV9Q^sRH-*%4-1u?A#ncrKUk?b@A;If@p&1
z@b0VZ6_N@tqD2{-RwIXS4wN&*nf|R8OtD3xNO7V}f>ADa+a)8<ir+qN7)6~>*gNAt
zleG%5BgTt7(Y_>iJala5KJLDdWbd#rDk*r2+%HqE=z#H<;a9x)__S>f53AkGNkXAk
z2)hiy<x;M8(M_Ahxb-aE#96~mFz4e#v)*avg8gLa6hQ0Xsf+QpSV=*PadTy&DRKAB
zX;VP$3|6iZ4i*t@?WOi?)5QgLy0VFk;LH(RJYq49b`bNc;{3&7>2~H{O=`sdQZ?fK
zt!k93=oBJP7P!DmeBp|V0@Eqx`zmk3*k)IKGnsaoCeDNvQL>P;laZov`5bUT{5?o1
z$z`F^7@FkkyG0nq`lSRQIbbRCwB`5;m&m+{&k@$L5!rhF^qlz>Ol3$BG>JQdx}k>z
zA=I_4N#A}Yt9v2eLm`DIr8PE+9jnME0`0>6ho-%&cMmTSa=E@a)qZbJ7JI4*g!Zn#
zxi)!PJI(e$m8f}%pgge8+1NHLrJcz6AK6{1mXN+4N&J-ug?oI4^y|qd!|&&XH}HY8
zsrIOd_^!tS6aeZ*YuR#s@L6OMjjC&WkKMa9FX@s4<Ie-Hf@?ev$>%Q#UI~;*e<su@
za49!`tszDEJqNxr8A}@)dIhc`xo|FO-=|NzXfl$=AfIo$zak4W0!(xKp5H$fz8I#w
z9Oph;{c(FsdH*EJs1mfCW#V^5$?sqwSHM{cAwaGwU!}-{FzwejUJv@PQV2!<Npk3j
ztxfHNY+g!Je9VOACy*9-444?JOUP4ro=Wh+(VR9?0|D13S8MRy0^!<d`x23Bs8*wI
z|FHu!+Ynq*UGQPy2aj?X=Q{>gZSU)AZ+S{1u-pG|$1XudxJkQmrHSmXar2mQ(46*B
zFCJ*kb64>QEBmkWX?BwXvmRfsjAYjEww-g|yYYwZyrir4f_iX1FE)q<3{@h0Rt@>d
zKDlqtRUM>V)|WSBY4s)fuTFL;gw8xTrmc-}F`H3nak0Fr%2aviRwtz`o1&mmnjp=Z
z=l34gLT?Hhd&m7pRndF$c;P3!8LyBxCl=J6Xf=-XRQ5koumD8YuKPHxfz@#2!0^`)
zghtX`#--Jbh!xwVCud7DDx8S@I!%RLTHuD`az$f9_0B&(WfReVAa=kwitSTTz-^nq
zXSOk7`^c9<mkOFz^pE1*CX?`d+<_=LwKh{D$+}LA><8ERVe$iT=85Q1uY)zAOi39*
z9PdLnse#h%+GuL$pLu%117xDan*;9W(@LM-bd4&VOdTIhP}JG*G?Yx2nJY+|Ss??E
zD%>+yvdP@0;JIsSnc=xL@s&SNm|Pu;RQlh)GWR<Q*;b#CC^%KMjNxdc5`Ef5L+hEl
z*CF!Rp|JpCB30_?b~!YR`qlD!o!yA9I2CoCz773<g{XeLn!*#GXLZQvT<5SUt>V{0
zjgGFYVs^uV;(w5y+)9#-3R-0#m~l~tupvf<e4VR29O9Exk>Mk%j?vz3R!udr*ccwR
zu?bh|a=nPe=MBpMjx62B<#0ocQZu&|y2{)3n6y)}hDb${`o<W2Fmvf1HrIw?wq~u;
z8WOiA*$7OzwxNOIUdl4gy-dY`q<U(pOR%R#j&k>Mp6I)=Xe<(<0?_Y47~}9wK0$YB
z%i1pj_L8SdPzf>Gil6)#^B?~`0Y@Ao9{Jxh8`(^*K}LT`G{0f{otI<UF0C9De>1Oi
z=2d?13>QdXNkTq8Fp)b$cqV6B`%DmAjY|-{wEOdZM$kZ96eh)vIkp@1=VOk``^%A#
zr4s>WR0Bi^^<GG57_3j}y0gMpl#8C;-%}lqh5?k~d3aaF)#3@Z`?O+maBSgDD)ice
zl61|jyi+m@P0z@%u(eIy-xv5{MI`RIu>9`HKm6}4BE&meuOdzj*^^drN{N|d*N!#<
z>fM>iDojQRcuJs?^xq523gbld_slciQc*x;Q?fB?9yy|VB44c~J{dPMZL^Z~{Odww
z^TA_;XMm&#d;0|Y%fEVrtU@mSAF2W}N_DDU%sgS0_*GRlY~Y|8)l3})1RoiG?jC#~
zb_-A&foY2T&JwdnSsenQ0th9Ur7NLEF|;2j-gW!u%CpnU$ByGF%Iz(A)x&U6!?p1!
z(n^$7xJ1Rq7Dk19Z)<OWH@%wl5R6zx1=ir-`85+88^uIk_kC8tugAFe3fKn>iYY+9
zXRo*{Pl5vxwow#%m@h6WeBZ|PL)JT*;9YgCJi<&rYY&&aW~KlMOoh!p1Sb=z8&8Z^
zvz4UwSW+6j3u7HVy?Y4)Zi(VZk;)geiB=EXl5<d59(H`Y5fBC`V5N2w<ruBuA{N-=
z=wf`XUzt)8d7IwRd%hBIDEIRW8A!2mB1dPAvik7GVoTy>rUza<^dXsMQj?XQ-RWBt
zU0jwoL5P*d)w>xbMacKaGliMss*8yhs<Ye8c+H7q`Wi?gF9-3fk~^I1J5o3%+I3#4
zX|6l;*%A8-HD3NtfRjWgkb^QoP*|x_9)J=68x!ogQWq{`a^gVSXb}%lCyDEN`s(n?
zWSB2v;!#Okn@Q_Pf^cbo5bp>ci)@z3_ap3hUN5wK|JNKFm@gntYT5k5#$gRziCPt2
zG35y%C)Vqf9a;!s5lU9<e3N^wzL-z+1%5|l$+VP=UIq;d!nuEv<hS0y2H?2@wdanJ
zd@I-_5j9pTzo~*dtXC(AheiQE#8ro22D^F%6g$jXnI*m-(Z~fIsj2P#8KdB%HFG?!
zv?%;CIhEzT;S428+goW3yLby3R#yJ7*4mp%U|*vf{RUmEL?Qg#BAlqnG6j5>eNwgM
zccMpYXwBs2bU=~RJ4jS>^D47gC6yB>_Eji3uOwK!HTxh*T*M)dyMBJ|c;dJ6mqD5!
zi#0@33V5?FcUoRFqK}7J$Sz}n%?ytfO!U@_LD$l0<+EXsN6?7RefmgO%c{f9nlPL^
zw&1QdIsIsO?l97Oc(5E$E{|Qti-+DG&WPY5rvqHkM;1B7i9dzTYzXxUb`3h~V}OZn
zkK+xcaK_~b4UZ(6?SWGMV9YhShV~o#rZ&+HOFNHHADQ`uK}L?<$;{J01_fFyG+H>5
zv2G#ww7y}OH+vU%=bj->CZp(~>S(RqLhvglQyBY5-6#ePJr%iaeX#lm6ryB2Qhj|u
zf-24?2~g&vF6_^TVoLoDs8O#&zi3^b-2Z6U*i*kb2)IJc5f<x%1kCTk)K+BlGacvd
zP^W^=-o`U33YR$=n%jDX4ie=`|0K2&!6M06`1^D%0W123LRqUt!&*k5D8o#cDA0dw
znS3-&@%#7e{QPo3O@my^BI6)eoHC35?x)-W(2*wFM+r|xXurHZR_!7*gK;^VBEu)M
z@eenrh*G__**IBsTZ@vZ2>u0AgyP3vJcDf~IPM>jr=%mQ0nyaFYf&sj#&uXjBA4B@
zBw>aC%aVvOEVT<ymoH?2L>%(uHoEFjl0xrJM4$vmAYxerw3AXerb8cbndXCNVp9~!
zM6tp#F~^rxY?um7-rsC0ykSlJoZ)D3{||M(*YC3)+x2(DCF0(=wou(M8#%4tMfnVi
z0BOPio!$PEKIN#0*CkpOrpLs~ZTj{{#~e-7%CCRf)H~J4Rr~@b`17b_bY!hw`icU7
z(~Me@ef+g^d*bhF$D3bfJ@f`3gKJjL`Qyir?>RYU4(^Hb5E4d9XrvaRxp->BjB|eO
z&xG~%uNaEqyBuq%!h`^%(xZt+r^M+7Havy|8V(q>pSmCVN}ikvr=s)yG?Bm!)$_Yg
zj##&|FB{qav0OCrzd{Y6JRO+r7=ryJouAf6!QDbG$7b^1WR(rRCS4_U+LUuV?d=9`
zn-!n3U_sMZJWsjAn&=W}j*0&q(1nhpt<(MHyKelt$>-s)x^*90X3*xG5?i4nn2lAc
z(1or_mA)}eluAMsUkd)&%OVSnI&XwxWY7n*rSAuGNnhb@pOLNz@SRjm*f)>>J^!8&
z4U(gD^G3@&H6dZ;>bsfXzRTZ#!klZ$JK`>1a>in~c@G!u*N9_^DZ+@Po=_qo7dVW=
zQ^pGOlAOhfBo8aPNP}=ALVV(7O+}I8!pS6$Z{Wa__=tq-pVia=h?Q_gMw~W=mCn+h
zIe!pBex2e(!E$pqEPz?r)P^)@S6@PjL{FGlu6Gy1QQjW95(dR$4Tx@CyRu0~vL;Q%
zcl3@0_(of{jPCp;)zriX_uHAz1alSr2t6Sl+Gg24Bi^W<U{FwOn6)gf>G9~ec~e<+
z6nNSz_`|5GmV<vRUDFEa_ZOo-K6mT3baUkO^`ea@qLu$7{|1OdU55p9qAco14^1Pr
zSM=WH<M+ROTgfY)$JxE!Dy4)ZTE&P@ol@xj7g*f!{L$8x+joDDb+&S8Ggb-P>R);D
zbyDW^zb9p4<5k8FgbG3hDDS4(oxhk-tditifYBsPTl+w_;3QbKio?sPit1wup!W|e
z?af*!MovcpS``v>XG;+xH8Lx=)KT9DDa$>Z2#tT`p8PxSXZ&L7dxG2G2%;oXL1nQM
z4qsn&jk5)U6c8rSygH1Pp%h&qxAZSPOn-*|A7a3wYI$w$bp|OTqo+B+QUmlxLitZP
zA;G&HcF*m&aXxw{mq(6Ed&3fISAl)6y(og0vn}@UPIMgF2(clYUW?E9hn;gyA2uZi
zgvS5Kh<dEGNy*2biW2DTt}iAC{_?t7>%jy_uW{MniYnnezrS#*O_Hj&_f;ptp2Wg3
z6M=PghV7<!x*4q)2{*78Q1;*+uP#Uu4XnSJ75zpU^to&dece?w_(pWsE(CKR+<2gh
zrrOJOuXJA^F55fkii`@^afwdBr?<w^)}PCuV?*Pkgh8wmN|DxOGG)av<&k!{-q#q3
z)Px_X@08g0r}p@oI-GOk#U8%*-C4BX-_LqAPj4*E%jpV=l7Y|?zqOhwmn95cptN2x
zmvS1JX7}-2H$L_7?=Q{w;?uyP6H~t<jhEW}9nUx<d|qZutM^?h^848tYWtb~*hTKk
z%75f@B}0=8^XZf0v&_#zi*=sPe3;h<$+p)}D!98}f?<ecM3w3y&RO%jIemYa03Yp4
zlIh2i(IaE9w&o$`9)+WjF}8~~{`O()Ro`HGiY?8A)o2`MRJ3@aw~#zgY0dANFoOtY
zIbMPcMF^G=b(^e=h2Rqw8QB4k9_H)C(aN{tG)jq~JfYg$bLx=j3%6)swP5sVl~8rK
z5`4}R*Y0|nhm83DVe2fT;_9|^oe-pu;O=fA5WH}AcXziSg=_HO?iPXv2p${)!KH8s
z?uA?7dh0vmobJ(m`)B>xd)Lyr=6v60y@42Y)k=|DzmNMJjuvp!s6F;^jAL=Rj~^~+
zSs-@;ud{O`YS*m63%eGfhTd>OAM;nDkvHbiOS^QwxH@(Tipt1aR+Yl=@`j--Nin+>
z^UgnQhM}XqD_ep`B+HM~DTyGaJ{RSPPH1%N(gZvj4I^c(m=N{r%iG@?#oh2PqE+3p
zWD!0oETnwnb1G}n(PR&yVZ%{!_<IBuI)|y-9BKA(_Exztw-hK|xp{PHfFB1(gZ5Vp
z0e;ETxv@Ar1HgIBsC}aQDRkY&!2!pacGown;@+2WmCb#nW0bc;^p_OCG23)m>6ge$
zSRAU(WESNpcWy{x1YuW4+8Z`J8DsD0VNM1())G<1o(miqilr@@+{k|w*CUDk?O9wb
zR8Dm0`|@>^rQ-(W#BbTXW~Fx#a|HKJu$z=e_N2#?bAHV?<FJ>-Vg>{diIUt*BPZ6X
z1Z1Ip&JNx_EIIN&pPIPeR0$L{SH~o;#*@Lv525a02Ve=#smqxf4Xiv-i_4u>uhL9g
z?UnBRDg6j9j}VZVcS3ery;d}5mqM)I;h%jmjIaglg}{$@eBm=(ClC;kKo`fNUc22R
zTO){G{G6%Gmx+HNwbps(_u?#%&u@gnX6pMn{V}d{qmu;ZN7Qj3t|;c4cRa5h0Umx#
zbl}sY9oeP=l!uu3EzW2^M&`38v0CJ)Z<y9-B1cvIx-qU|2>!mF=}qG&9Y5F{X+_B!
z77J3o&ESzQ;FX+y-5eeBJO3Q?IiQYx_!%>hoEkYtT={NJoJz{OE?(s$cE86juN&I~
zo;+owYU+Xb_`=lEMPYZ`x~Asg&L%&_aquSr&WX=Qobq&HO~jG=DFpkB-(jhzUv<Ik
z>9i*oe+r(y99%HjTX5$W_@f)m+hQ?G<c&w%vQ-cI#K=8dovvD)RC}qdqV3WsuGvhs
zcf<-O#Mm(r3V4Ks>zIg1T0Ol*PoKt^$o_ZG`^vHLk3Z2CiX=Qs6#`-8NKBa_eIEI6
zT<x61qmNrM41O$JI837ypENOotPK)m`tP9i2CHI+Ejb@ohdlLsTeo;V<soq(dAd5`
z9daa|ovX_QcJ&rPz~@s)6m<<V$Pdb3&hkORZ6+Oi7S@jz)!SPosIgjyu?b3rG-5o2
zwQXG^*9HrA9s>L2imm!I9?93^^a-(XI5J{hBi=C$W^())A~-(hi}!_W-Q9NxYpA*J
z9R4hvTP6BY^|X7nGBg(GVv8r!BX2|%WwzcUAdcgA=FR=Ct3y8;s~v8}=zG@z65|}o
z+*F_%;$ZApXVVH=R68e(Wxr!_VM;~mt^I!f>5|~vs1->S>YUgBVe%L|XLsL9FWkl4
z(T0#+@{X3akGJvhgi;;W-KQSWOqDk!mo_63EW!+q-f(PkJJ^J4DGoGPU8#g{BB0u?
zzmy&*WEuO{ZH-Gx`qhZ5$}uJVbl=^GQuy>w5KjUl`-@^=tx#kYRyTP@!oLbn7y59&
zsGC++5SV-CHddX}jK#>2bH2Ci?6Q|MsN2~L-kQ>b**p+-NBlD0Em7yv9W=N4ZNi1l
zd5nmgX?fRf{C!w~qS_@*-vMR1A2{6Hh?9bA!Ygq9r1nWH{f{!5>c4y|e_{k=JE-Eh
zW)kwZhen$Z0$*)fSZqHKVe!a>|5gaa$G3xDrZ6eJUlPJkpD?qCq^?O;6O+Z=ZhiR+
zlkL3qT5yJqRJl&$!KK)3hJ=fp?hA*G`v%^>PTX?!iS_5AWqB7<{YQc~`#6e<il2gZ
z0J6io3K@EjJMW17{EDB2WyPHQV@JI6l3A#VcefpQTU#$`e4d*Yybr={%C!}S{8OY6
z-$wkd9M&lU@C6l%k3JuArgZN+ey$2t`W3F~&Ez7SD%dWbwuWks$1Wl;xPEW&N~JDM
zD&mnnS`avP_6!Z`rQ`%!NXY6e3VweB298no7u$KiwZwOJN;Dplo8PG2+eBcdVgVNT
zQF(`?V<K9RPW!m?6=z_Qv-LjnTfj8+^6Wv^HLpmy0<X9=op&5yE11@t8cz*=^_>G#
zac!8oECW-YN${sKgl|xnms;j%RXjYNQ1dP^NPZ43!lvQW3Ek7vDu(xVnTW+?>u?G~
zS^2y@t@6x1tPD1>4iAa?fJYCjqq<opfglqw9f4>QzjV>xog$T=Gj=wFXQ3NaILACW
zj%T0GbzVoD&-2dwGnSj@Re$u(uhDxgd~Rs;3{!6J(U=fM)(`ioZxzmRF)eKM%s5G4
zRpWrGW*LwROMrujdk^&sTEWX0`(z(LA)oY<2qlhVLNggAyo6W&mGb=$w*_|It(7q9
z!i43fED~Lj^`dkGulYP{R;`Z)0w_hSAUiS#c>22mv|a*SKiB#Mou`<|rQzY>oJj=?
zNLE@r)1n?o{RG-)nb}M1qtb8~?%RWIg9e(ru0Qhd{*p!Xf%y=$lgD~Ro*_YfVC|Ti
zfNO4i0*Gn}KP2Bbe%l6#i}=0v*;;SA=%iu;IZ&!{W_vq<4fK3!u+r&OP(6-l68Ma_
zw>N;?G*qBcbll8^q9q*61y*GW9IC0qr&>+4*!VIFVBug~!d<}o)&vVxL5H(U)};`l
zoLn<|`|QySVXK-t342cJ$_vl#X)vKeGHqrDUCOY~or?*PuueO;K63q-GbP!YsW<D*
zK6Uh)Ke%bB*cDod)<bg_8aY!6A16AW_-i_DxUEZV^fL3zY`juPct3x;ewjwa!cLG%
zCKjs_m(omft4F6bV@*T(Xpo9BCm!<d{oJ!ynsam>T!(o;=P-$gPmIX(X2JP3Z26uU
zECn__aco}`e#C=WQ8&FxEbVsTQ&Lh=N3BvZ9Z%d_TrTKE?xs}k|2TIC*I_Ei*^|xt
zB?p^XSfo_y)SM^0Oa+aJyh2{zdpXx6%i12FPM@Lanskp&Px)By?O|~Y=>y&}hF>tn
zIW*V$8Ku7YNH>j8FG#Hd!seMf*dAr+eqgCrm)4Ku4i)@J;J9`Cwthq+?jM62EC~Pf
zpCDXtc9*;uwAisU>ZigbG<=y+VCC$G)HFlB*NH_HSrjIUj3hnX;u^_yfP~!y9!<S4
zZ1+#Hm)dY7V6v`A9my4Tjdc^+JA3`}iV!DV5O*zf<0+tvV~nXUH=-8SUq5$xx&AtJ
zd$+LuBAjL)W#6O-QyAFWCubcH?LkU2i790G2bZLCj=X-|A0NIy{pplK49fz&uGhCW
zOEk3p;q7|CizgEZNkw<Kg+3th`X&6vyS&tk7)PO+IX^~Nkd_U;2(b?c)(?OSa%qs+
zv17uT-N}<V!dCcDb~m^FRQYr+6Ep%|V=eJETa@3Y<XhHP$S&nOEB{KVo68G48|HwT
z?mg4dC}Cq)C-ovxjl5&a0LDkM3FrRKp1fMIjL7RG+lwcKIG8+gr)w+2qSHp%k$OQZ
zT=A<%&}PLPgRCE{4y@$fwehY;JMeQulA0o})aIG+d3*Hxip<FK2Dspzo9ajZ*GQ4Q
z^}tsZ@3T;^VaEZR1tj;EYZdRSB#m5Qp3$UmW@Y;u{f}$+JMTTcycU-oCXcT7J&>**
zk+ZHFeVNq5JRrlLv`yb1pLj0nuJ>CezwV2?L_X=f_6H4rOHq$fkbMjGlzlKT?Ccmm
zkf9hA`>3itt?<b8c|}@RHAw<FtzjEM3Y9EeCP7Cx@R{#>g<N!f&=D!PZL_s6JF@wI
z<78kaXMdno<hY<MeK=#K<>`u#c7B_}V;lx1>#e4%F0C=Z99Lg9+0WpLhqpXdsmO9|
zL)+fHxW#psn&ikUD_hq(@N<5ukFozQLV8Aei{GP=_SsK!b&00$5?1kV%&552XZagB
zB;_X+(Awiz#gD05BJ0xocGOV-$t>q@pFWLZRE&Gy7I_`qJ{H>z)I2Zg@PVx9Octr=
zsI4gJBFWuRHYL{i5ByboJn;{p_B;go<-iZ+v!$7Oz>q|>!0PYBj4lYNvwXaR)WP=$
z2>C%TTCq)r-JUQ60CN1wZ`PXF4?w^p{9!-{)7vk0ufb!FqY-k5i@KYDl7;Xxm_)OT
z#;qja%*gl<y}g>g+B0Fn<P~v3<Q4@wgX&nd+Vgl0r|GsH7|shQ<lPv1!mb!6boWX^
zSP(@#`Kek`g2j9$NJ*=%+tFZ4@%2Zr3yJuhJl{{_)={t5D;fVsm9Fvr4+O)G5lDyw
zt|wr`g)>L>Xxb84jC)nOk8XiFJy|HABs~?8roIsz2c;71-AIa06`L=eT1(@I2|_od
z>rZ<%=cmhO>>?rwP$H6}+m=9>ghL|%a>G*$Hj682ZdG=3CZW?@W2Tq%^EcLGM#f9|
zpps{G`o!m!7yL!U+0Bso9NnV(i^S@`F+%(QjS&)rp_+z6P6@l<;q<EOFSj_WjN4^A
zg9thX<!Zs(xuSAW;?wI6_1Ayt_LK{xjyNRKRuZk$KkVQ5a%C^4%3CRuiVV-Fh=ewW
zEH{6yt9MHsANPRa=LQHB?HOs`<${W{-m*o;fV%C+lK6th>D~!aNfM*Ab&FXKxLiyl
zZ0Qy?elvXWh^H{<Swc&AZ>dZq=v9y+muEpZ_eFF~k|{Tp5J|0KY-Ni>Uq&qA4SPp7
z<x{z;JVXe}VwX&f$pU7(+=AAa$L-FaIB0HX$}@%a(DBi5GN>?@M38G^&GQVHc7JJ9
zOul03_suj{moIEWH^sYeUwbShl_*;+$tcYQUI5sHUr}OEWiSELwC~eI2fod|T%!me
z!#Rt~U_oO>t5#J$4Zlg#dBvTjgOGASOl6ru$co2!k+W(BCrdVN^4kBI?OT2#PpkxN
zMwB0}hTLKv{n2rbWSv6PnceZv(T{ExRJ<is>`B4FlN{$Au`+|%>T;YsJOfWp)XmQ8
zJ0%$fkb|-UD~A9At5I09y88*Ke@JYL{mL91x<@LkwD@$)@!Uy|5|u1gIqaK|LY~Xe
zt6`LFvD<Hxn8w^2L>}~G(LeIT{pP^`#zbXImOl{&)~=YKoQtdR*AI<A6U?{2EvT=A
zf(mPLvEd4cPCIX@el38TTxhMZ*IqZh&h?@R;zu(?sN&7-=j^8Y%;IR3K7RNBqtJnq
zO5Y(qr?Q|5&4^djGc@GSdUK!-Aajts<Pt{?YY>Oa-}K01Q^BXnhXdo;!LFNIGXCcm
zNd9MjFzdGsQi;|O^bf>$nM^SWx9$xUytrK$3sn&cYRrdJ`I0Bdu?sk2<&8KBgB*4)
z=n6!H6b8crWI^%+vcm6+9M6uY4?HDg=Lvkp(Ry@ORO${0Zm(<qv_EIaxORqqd;g$M
zrMYtAzPR?+%Jqd2Hgh?k(9tiy2_<S4P}z34$fQZ5AKCVbZT$_I({yzW#ia4VSi<jd
zALqi6AY^V`QlD{3aOFREdV1c?oIPjWioAS5EeMQtjR(e8XfhZ02{Y~1TCTL#Dywe~
zIi9ce@>84WP#vldSzBgVO{&N>yp4$gRaLR1(BtRGP0y-TD#Ivp8LUQtrzn=)XNtf`
z4msHix$uKut*@s@cVpi!n+^=Sc!V~{cs#2FZy8HyV(vPok3eoD;%LZ^CX!mD|ItN~
z{Rbd4Qp+>ie?I%kk%-;^CY50P!ST9Jboc>8C|=|!dA@-q*1~x++2)$Fw?{9}NX@eM
zfJB?IzB?}RtWj_IJ3NL0NjOH$(fPD?!SOP!#y?NT<Q}`kE$hLXtQcbozB*#g6&&8+
zkEQRyw_XN3c-f+F7ZKD;SFh2Ea+k~|T6^9gJFQyV^@Cw)?2^MTKc8L9Z&s6(${KD`
z0m3!0)cXrxOEeXgCe(ju%FdwWloq<hnJ2XuCZU-6oE>2BTk2tVAmYZvj0(O!!4NeR
zg);}qB}6>V3_G2#hLYb}PJ1$@2-pKJ+QM&lW3E!_EbXM*mn{Nc024y5D5$8T09>$G
z`ulwhe%xIs!aYb<@|H?$ue!gK$u3*%hh}DWe!m=du0{nlCUdig@N27(ziPB3ideSg
zKO=AJ^w4PSal>eO3{wj0T0f0NA?Sb;otIMF3QrvVW5bi5#9!22>#|01S%hf~ib@S;
z9q;S8h5JjLF0gUK@Xa3jnr$(r|K$M!%)*)zbjeVhvEFa^`UMb!9X-E?Se@hbHQ+Z=
z2q_!+83N+y;s_};A)hK5mQo(U;0<>IFp1OE@<;@}rT^zp4kFxqs{QXMjHq~4vJFo7
zD>PU%8bC*<Rw3!*8^1dU-Wox{^xho0)Gf!%A9{C!ev;!A<)X-Sif!bZa}Q<>oGe?n
z!?GK0`cObwW$Nbv(_{BfU!t&eJy``lHoqcrHlUIUxTZdM<5})Y5TU`XDV$TH=X=KV
z`<fQrBCTmtRh!q@^=>_(sTJG(()Yi+c!ZH;plf~<n~LTB(AKRcUaIx8w`_R(fEt=f
ztK3*I>Dfbz%rO_}lE+V@7FioFq%*}gG2@T>yLcaiQ$O1^dFdH+|5^JS`nmg^xZBmf
zT)y8X-T;G*Tr!#k%blQXa#<LP$_preo}lsX>sHgd{a(m+58)`~aAs!iZYzQD=Kd58
zCGBcpWalcP!;zO=vsa{6Kz}FR#98j3fP{@#qyqt~8|>wZ*5z|YltrHk+HxnH!!r07
zdiDHYw*DYY%n*(GodxtAF7Z(KUkQ(E$1qN_fX7W|;33D5{F4oSV2gA_68_mAL%%H4
z`4HsQtcnC3KF=*iSVKL5&z{~J`yY6ur|axv=sVVYrmWgHOjG&VSiEkxnerBNV)R|X
zInpDcC;-aZC?H8L$ts)WbADf7q*r%!r#CVKk}ScNmoNSb_p9Sg%~y8>ChmD+$Jfn!
zt^aL0q!-S$-ajNvig$yv3Vax6sIq<L^t>eGv{^Xt_Z8?COYkbJ$-)U4Dm^xAmpzzg
z@4G>uw*&chf#c@z33!nNUH7nl8a{><jmB7JedC`hHCm%m;_SMJTmLrY2)!6QbN$dY
za6;ns={`blXG*Ge>yr{T^V|&-(v$NVZZj&hY2VaaXED0c?p3ht^|}6WpPnMnJGdc%
z)W+VP#j&<`*PjdnK8{G`rvu$3AC>d&yD?>@nG~lmOf+AiU!Go>S@}JDp_q?U&NgJL
z#;`-HGdnLNN`WptJ)SabV`vK{$HD6p49!XuQC{675wE1y%6pX<ctQ0FSM=2vcY(D<
zQ0j{64WGB|j0T8Q&rl2y+>{;{_kNnz%HfD7G%!P*luVW;9C3Hv8s@9x6cGv)Ri&UP
z<7bZ;^+Q3?6kgcumV(*<CfEc4;$MyBCQgk*F+^H^pTbVfr)#2siIEWwby@vte5r4*
z8_@gE^~+#EzjISce_hS~gv<`{i~S2L<y`NuvI5`ynxN-}R=-<(xf&_28YPRe&~d?q
z8l#;k&azf7D6z^cSs7$7vpDYlYF#}$AUW@-0Q)a)nE3W3f%iru&@PkDFQg7<5L5np
z6S;4l@GVz(swBU7B{qlL!D{FI<o~QU=l)Y~Hu}{XoC_-of8!5Gv<2`{ErwQV^IeR9
z42_q2&#ooz@V+9@!pTpwuXe)XM6wa;e!KpxZ4m7Vc~<^T9YKESvIBG=p=&~?>*+()
zT3bRg3Jr<>ZM(?MXiZSKb<uE%0zm|c$>%2jq=@J`7Za)x)nbX>y$QBeKnho(I@9I_
zY*VUh==?M1P{1z0JuygOvU9y)(hDIL_|vZrfQz8_d83=*y4f3Pd1liGJISP#FARFS
znWg8gI;{oVLj0ePOv==m)GFta<T(h@VAYd;!kDTLFi3(JAbXsCo){Jm(AL3_U;1J4
zLJlAG(%<t2K*Rxpqa5{PJ8rlby!B(GlRYRPgGIsU?@x5wglO2)ebYxJ!0u3WSpHtT
z#L(JC#$0h=ZH+aw=S>AylW&}SiDWjq_w8x{^fHExQ%LB~35m|ZN{4(pU9p}#u9V>Y
zET=8j*#t<8cHCrThbqv4j3UTCiKW4?^@eEPn$$qA)R2h=lh*Firo42s#M`tJz;4JX
zzH?G9__dJPiQK6!!g3OF2ctSX)j{sqUJG7hp%9fq!-6@3+q$TYINq*_sS-P5?@hZO
zeggij(;H(KKE>LpejkrdyfP10EHr?zntgv5P>*=>f4FVUL1ceurJ~-Uu-!cSWr)KI
zn#=Ldw>(sf_<R3P2B7>u48YSRlAzuOR8U`=WYzWtA}WQT8XMMkc=e53o2`AdbYuVY
zrnBy-Ix!z(zeGCV>l2f)&pp=2(l<32xpq!Ht%SY5ZVi3dIEA5r|4+4o$li9$8Fc?1
zy7F|4J(`e9#YuL$(mH-&!2k?;5p*~bXc1$)PWyt3j@K{tWtKk{>Rsse^9#kV-MjwH
zyx%@5MRV5Xj=3}-)IMr+2d}(7v=;Y!eAtGIwkuDR(C>p;$W9`LY>Y-jq%V^zbjzqN
zJKgi8QySF*zEf-@?L$TgEPNi+4KUZ8nm_Wm+bHtVTi2C2A}&+Tw{diY%K^FN!d#{w
z<HtG``JyBGX3RJF89ogp)5mjxfTO}pP^t*_L|WFXlRvw2YLA_a@6Lrf*S@HTJf@N~
zpHW!>wm&f0HRr&r<;WK8EUYY-Llq#+N)H&_8Efj4D=Ky4GHt^9AuHpPv}$D>J9n$1
zA?tRB#F6d3oG(Ot(J)NJZ!^L}?KyCyWVBg!?90ijNBSOqR(@XT)aj&cQ^?6Gq_fr3
zKO&!6d_yYJIZuaNj)~n)e!q`sI*22l&+A3-pIYcmo`L1B7$yFdR!_VfWZ}nSPQ+?q
zm8U%gT0f=ud#CgPc})d8#cZ`|tqSwIkq-%&4DVPwy$I4o5i>i+f(}Y-vVgv+*V^wa
z)t>DxfYBUvx`TafVRol(_ekq|rxSrU|JQo4dUDp;lSIQn=yxh7Ng7*2s(XehzE?5{
zcf?YPTX)yHZE0}SDBKZCK$J2#;B#;d0lw^Gb;$Pc$ATo~&9*X!E11ZSxV;}!Mg?yc
zf?~pP3O;cQ4R^-DdP2n&a!Cg+2n;KyU0idTKE3)xPvi9|B8>*jm&=ZJ&$zSBhMN@>
zEb$SrB*>slUS>>Y<7$>3Xp;f5nQ(jF^GGg^A=CBz<q34VVkD1pL1ZxUmV1-`?c|*p
zsLZ$qhe?WyZ}{nn)K$~U%Nd4ye1x%`*!OSCQ!4>)={X&pojJ7&Hlhae#VRs27$#gV
zXN%Af-!;x(AqD*6)tUM(601*&R^aE%PS2(9Ie!MFCU^730Q<%^%HlMJiIWfv_FN^T
zXV(N>>tyAbz+8Q%v4w`LpzQ&hT*0o`I%KcY6>~qooRSg_ae!hv?7f)14PAi+4XfxA
zg4ghR^Ad7Wdn{Q^J_5z>Znw#(Uz6vzO9nRo{#`0<UFiGt4L1WAaLLskrxedhA$8!i
z<(36Y1*TY?PWXC&EiPoWxZE&|)2`7gxV;lnA0c5Zqqbf!WB7aP(o()4^KgL-HnWda
zPN(eLoZ$q>WfQ^!{Xy3sjfvPe`MBk92hBQdh{WZ$^!s$W?>QKhNw*;)J_*%xJnUFF
zD*0xc_k--NxOcj4-N`e4k#4PSJ}fSwJV$3H8}nX)vfZzv!@U?B|MtqAL;tHh2SbXU
zc(j~HzS2NO^Dlc@0&IGus8X|z7?G<nh0C>c7#M)hBqrM<YxofS4}y<}d_OGM!8b|b
z5VE=+5*%8Lfk8g!&G`~7;iprQTmewBgO2m$glUzw*=O_QWu2dHTPu2s^ET%G(ODI)
zp1(*4Ez}Ob5qA({GdQjJA`L-Lk!E<raIbbVOm?Ce$YJ$9KSd>_4W}I6YVh9~oANXo
z-|E;ojKHzFs2U4;{My13YAFwl%Qq0&FL<nB&3S*8V}s8w4;U!e_nYOCSuJiEV2qE!
z$J(zyjoJrRj#-|SGqV1o$Xj#WX!}4`VL{>Y@v$S0+}(9eP+CXH&iPABqoSIg&eJjh
zwTH`w??9&6smaKB02O>V=%R?iKXR<+c#H*xwCFhRMYT;*i>A%9PEfA=Gk?PaYCvt}
z&2&3kbAriyF2EOm87R(oQdDTdq{<n;1~d75ad<1NL&H`~*MS%=#>><C`a-hEQp8X0
zdUHj@oW|T~nE;z&5ATK*Y6J&YR;1Sap=4E@kGCpnss*SWzN*Z{f1`!N%mNf@(W~(|
zHMSLFRm|#zOV{^q(keB-qZm{pF6Z2qOS+tnG5yo>Rqon+L4nz}{(?0(TmY=i&i7iD
zupMZ|LUE#d07;~MZ1?(YS>U!q%P_KU9ywehUzVPHRP+53C_Or4HHHFofbsgKX<>l>
z`u|!>jH0R9{c<K=?hHs@$DK`W_J(SQWYKx~YiNmO+nnsLnQ=ZSbiB0JxQgFmB}t&j
z%#W5Y&=cOF6<C^*kw>!j)3R<Wl$G8U9r4=sB}^_F6zP-0nl$b~H!sRQFNPxU_w;be
z(?iWOv)(*pD?H)v-xadMMeofD&Cv~nYGn-P<H;5uTzu8V6_TheK=tOsb&h14GN}xV
zf)o-dLlKZ$ygq0t#Gk=7J>nTf?w2%u(a_?Z9Nb75f*3a>^fnBw09NKD?tm>FPA-o(
z+}JY=qwqRE0|)N<fH2@EqTR^uhlRVHl!?2#C#LnMIFrp48iqx~?w-TWN=k$QcQv&z
zDE`^D|KHJ$*cFRQc`BP#Rj>URN>vhz%Nt2Cl_5ffbfQ#j+v*G9a!Uc{x~>FvqUv8I
zbNZ0QDWX#P0T>qq)&H@5{SUtk?x5i9io!AXO-nP3D8Ilub>bdHJyE^90IUZN9v`D^
zhF4+9Vsd!1wqjcCSK3V}Q$H%fCegl;e<XOEJ2s0ZT3+yBe|x3PBPA<K?mLirnhrM3
zgXcql@%tDdTS1d=_*E~`a9*rvD;6p#1w+NeNtsgLeE7%-oe_N5nfX^6A9<Z!z!JT+
zTn(ZP1+_3kJr@~q9#2=dc)&HH3BOP{h^P&B+jTJ)iB2BN%ACIwT}=;5DxbhaXd<^F
zxr$|KA05``TpM$5V||9|Co?>1>jqBF{1d>d%c+yLL?Vsj|G%mZa^44?Y4y#^xn_=8
zzeBWVF3m0YRia+VtOUTenfm)B%Q&U=;A=7}jG*zwZPMvjEsLB%j+53D5t)uEYH#mg
zy=~Mvj}(I8bp~=g0G@Tdvg0%fA%^mlB@~z)3~^+LJcXS!eKZcZP&jsUs&8=#JE?Yx
z9nZ|nHn+D%WZGDJ_FL(2r8uKx?YI3{JFr<T1E`%(u|mE*bH12Htw}*=Y0{6<oFjBM
zHQNO6)~xpe|1=d(H28VRt*w=MJ!{z41ia(x3~?v+Qze{CDfmUUMpVVV(&7=P-P}_x
zz{}^eCD--(hvNM8;Vdu#he7#b(IR2$sEo_N34ZePEv76r9=iZ%%y28x9VQ3)FNS&^
z$E&RVcGV~XeAk~z9-@FtHvXH+BY3zduf!633Z{-`#**`Ci;g10pXxsWhu8f2uqS+3
zK6U@QVz&_-b-}KQd%^ZPcLq9|e~(X)aK43F<92_DpiC+-9GA0@nVl<Y_f{DZ`$%jg
zF;&udySSh0bTR<?*mANgI|K1eHmmB<Cpz|oBF^c@wHTT?8e(UZ-ru)7CtAtvdzHh>
zhp3{I&<)&u)1o&!*cX(^NZ9JIakk0WTO-$(8Uv4qoOj%8Irv{fZ(-APXC<q_v8o#M
ze4|Btdp33K>9cuXRt}4bnw8^eQg~8iv=haX-(zq2#^PWOVy&*a$hu0ykSO)VKq~I#
zP9=Ul#VtonsF12q)E8WZ`BS%B2Ir}Ew|BdO%Kq$WiU0fPJO7A>088KtP#98Y>sj76
zK`7%xc$Cr$-7vKOrXH-Z$~Z<QH(qs0L``~)6Uj^;V>&1q#t+vEgLS{sip$w+Y_GJo
z|54EJaAY%g*u@R4=Js@6E38{Ky#bR#z_~WtFZj2gS9V(m3rcx}`u7I#**c$?ucMd(
zpSBCmH<3ij{yG-XD9PEuAgb)b#taG)+iN|Pn@de3v-Tc=yGQHAhS}d{{+juBDC!XM
z$KkJuPmjw!7+JnIv}c;>sR;^sXcg<c%R$>qD4P13o8xpLfnAD|?AhLbcxdG5$wPF7
zM7Z7J1Ctq^ty*|B%QY{Pr6>C}p9OD+npS-5Q;B>K!S~vLpfkHrqzc@9zp9LO5HC!_
zGfrM35d1=^ZWkw2uG&wt6F5?C3pCce>wtm<47>cZPEN=ko>L+DKH-5uzrm;YP3qwd
z39!1SFRRWTx~ASi3?Z6jQelS=#@apW)B~MpVlkCc;#E8CD)HjCI?ZZDRWg#I^@Q>T
zo4d4{SA2t<SMF8gNEvcIsaDNitd245%0JJ3l|Tnup5@rNM?vVZg8%MWr*eDisDN{K
z|5GpN>Z;J%%bH4g`UouT3xzfZ4cvfve*XRb!_;jT_e0v(3}&M0P9tq?YfR<>5gfpf
zE2)KnP`z@&$ShFZA-0Kdvs@wOv}Q*Buu315LmG0>UZ^x9EI<)30)EUO9Q{or#<=pQ
zd!6VXhY}X}zm+O6v%B)A;>hw*+iN*z>|I<j!d~e9j0?B(g=xjvzq0nxRn<2;7_iB;
zm@INbb&HCN8#)!7T+<ojr_!+Vv~%jp6|3Y|T4#3rtGt`4TA0Y12938)cN6rMTPJ+G
z<;qq~?e_q6tf0+MR2Eh?XioQ|K*!xuhO6+_RTP8w(J_u;*Cq12&%gk^ZL>jD^|}R2
zTg3!#VW9;>M6htLbOq$VEX3B3LtPs^qt$QNajuYgGX84TxYdt8xK+PLyJA-hs(GdC
ztcsNSzjy#IxTFWa8k#WJS|ilW<TQV-U08(Gm~$FzCO|Jw+y1*BOdywOqq;dTqrNkA
zRzt=0?1k9j_6){09w5xFrUysPGj|#n&??z-<-ii;?|6{Y0+D#1Lm19SwXP`~aToaR
zwlLKx4Jk9)1l1>Pd1M=6>wl^;rwt{CSvN>A#Se3YzDbWKHS_Ud7d(Rj9Jl)Hz<EN0
zMG3-)-^ZIAG6*YYTRa{SaZ+Z6e*0tiN`W^^By->cj^IwGRTw=y2?23+aCD@2h+Lk{
zjA`~OT^>j!T(>&Buq$73qbjhM4UQZ?J0@F2s2f^4Ix<W0ci4DZ2QyHZz<%X=wnrt0
zfNTmj(sYc=gw>>?W8H9}&adMXb%C$K(3u$b6xOq@pt#SVoW_MsS&#fQkjnIO>$pQn
z*nx3@f|q`-a9x=Dq;i|)z~I*{LI<?f6C-!hZHu_MWXAc$o;Mj*wq}{f?u5KnzG-|N
zO|WBxqV*>pSc(q)cFjLQ|4>w};~FT#ihAy1@z3xwgR%Yo<AU2)&De+v=0JfUbgt4b
ztovp#5J1#YpoA%uv0`Z~GM?FeKV_JI&{pFIPRb<2spS+%Ge<EgI0pST1w|e{k(SuU
z4J_fO%y<-Eb=>r#@DQnb8NJ+0eRW>#9^V><Hahd^=NZ8~e=r4w<h84&rq#(Q^Xo~_
zg^D-SZ+*MYroiHUmZ?~U&(|_5>Oe2w?*$=MeWk-ErKySg@S6Bm=Zr(7d&q`BSTvfv
zgpp2t(#fmke2SstX-lWIyGy~r0q^hnj{7cPzIX^tvoiu3qFy8Su<3O{A3nnte6;;l
zT)xlxY6<25Z0&iNCvHe-@V}p{;pFs!k#>pJe9oLsSKFkzmp66^3QCp_>OEhD2fTel
zQ|IEileu8G;v^Tl^9%hnY3~`EWnrENA}jlL3`^-j)vws1Z8QuU-aL0v^8%S%)v3bz
zN{JR~R*RPps7k)CADLB~d03=9EZ(VddD^>$WSIEJ8<!2=H$eD&x8PxwH4YFb=a{nA
z;_{NUg+tm=GXo#!fz)D~Uz3(8lKP!qkC7k2#V`ibJzAtFvJ(IP7QCRkTF0sh48cy1
z2Lss}-tB+H$4^kLOwfo{FW#>M@{Sj&2#<132o7FOqns#b(8$EQT<*wKHFdnMwR3qj
zCkgMa9a^2f?(i3My}spjUZLd9KQ$F?@JXtg8L&czoA3^Z=*NMmwm07!$k4>!v;O$*
zeG*>8$Do_yj-<%Sx)p9(+ZxyO4yNs@hG!kMbv`v;a*}KnxAV^h9=KYn?#|D05r(oO
zukj`Ck^eJ>VE2O{7cy=in9H66$%XAX|8c%=`oYN3*8c^S<##}^t({ImPN)ruQ)t#}
zb<div!~y8^7iINuGNA2_nG#V#amAf2a5B&dO|lt&q}Env7SD;jNgI~iQjeM?{$XLU
zFKd>}u}nges~PaJsiE+6G54{}f>Tz_x}i=ky@}mUye5Z3r`pBbGk$Kgz!iEa^<6!@
zE}Vvi)QCQQW5kByUC8ydM|=UWzpoD$)WE5zqFl=*lBF}u_iJI{e!oaXH(Nw$`jb1c
zgX3Dq@066ujyXD58KFLY61$lF%is<Fan}V>tQ7atG*g($Z{H4X7+LYy@iQj!IB(Yz
zclw^S?CO`iajnnBkFkEu&QO8w>*ZWNP(;Dgj?<giCP{v4t6kcKpGH=GHZh)MG0FTG
zJ~rQW+1^_Abk~{aYUPero<wVNAc@qp1g2-5yWxaLHv~5kff}GyRk^g&?DJtQu2tL<
z!q;3)ZfONCS4teqkgGa8{8D?C@$t(zwxF~#*o_>p<NUWdTh*GlT{9mEYTB(+sU?wn
zimZ6S@0!xYG3k`B92wEL!ENT*`<B|aPzAE<VZQV|73jQ55Exd$4vyuLNgy}#f%P6m
zP<mQgN}pPkwMB-(0vqQ(&s&x5&hfo@Ay$8vgqyRP6uJO;S<Cjp2Qa=tr|_I_ELfdd
z&n(a_zuz}*#xrI;&_27oE}A`iI9y7T>AU97WPJ%k=ECZ-?gpvpL(GK_l4;m$#f>_=
zM8{3yT%&xBL85$|W5_VyH5xJGcXU{%LdM?~|D0qT%aL8YY&Zvrn4ghk|HM^+=TI0?
zA@|=zg|~bj9EliKwgsU&#V4uDBF|r2Ed~=uWz>3I`$`m5=3IVB9@81?Fs2QDdKQ*T
zo_2a)9-ly^VD)|M)8euelTv`WAlErwNV)!NP{ZwdPr4PvgW+!3k0R&<%4T9+QZZJ=
zlJJ9CvF9DUD4e@_;`dqmWDd?tu_$7D>1aYJuR%#K6-9-^bbWJK`;jE-eoU!oF)UiK
zUO$XfUcEG7s1NAmVtI@LqSb@>CaH=!QE?T|wQ|dPuk$vy^Jd8==gfd&gtyP9-@R5w
zKxl*E)L>U*0nvYhx=-daWMqx!&#-V)UF=A1XXktEGN-*{)}<Lzv@#g6CvI?y6$A>{
zgFyD(`D4*x2OIViZN+pOMk01{l}zPQOTC`ULnb&ULZhJUN)m#7<^=nNEnjR6ytua&
zoPcWfHhPK#*;~(Yk;j}W|5p>~$~QJ-z4jxHBQDMvyuwe~U{aCrO3=J>74&M`WWyA~
zznrt+@?a)Nl@!?nvE#SOE(O3%b3t06kXDxi4h-RQOy_{@98pm{T(*Y`z+;-axJiJZ
z#mVT!7hX2<>kdo|+Nz#|KaF0*Hm3q-Z7~p7na*dYjt#VT#k5r(M`@Z}I^IdDFUm;e
zx>Vcg6Gxjac)dt)5Eha$I1lvO@|0e`{#5XAJjm3*)O$!zSx~w&djknfIa~tu#f*j-
zsd5Pdi>9=K$C-buqA}5BfP*4%9BlS~lG7aKZyh!XCDVRiwVOHdBuwyZH7rq_$#u+4
zQQ^!FIwV3zM?^!E81YvuV?6-M7&uHjjv2FCY1+$M)Y*-j+a%${d?0nz=z@ezki<`@
zojS2|huo}?nzhGV<>*HM#M&21h*g*nk?!PBiKGE|$?^1=*h`&K9YDlgs$vy9;=MGP
z&mVY-rIt|IyZW0{grvDEkUQVV;B0>{hNHonU5^Gk<8bCpxC_tRSO0_(L7M3yyxbK~
zm<*FE*zeaxYDG0oC;P^<jL2vfg*Or^y^S4^5|>+R68<f*1=ih=Di$KT+!ru62cTRq
zoFdE;>0L-Txia_Zu^FK`-PPWs<sYL)tT3q_1C=Xa^WtVkw|mqNygHy){OTqT(hL;=
zH2diO%l&X-+NW=1v@8QB118MsXwCATyu6G@#1;tB%(!vR@sRK^z|sJAZ1eC<b4y*j
z)K9_rRFz}t#%*-z3__*~DCxS!tX6nKL_OH<#8~>osGp?(k!9X)3%C{EP^Oo81k*r(
zils(exZt=LRLCJa!`lEe;OjAGs+0!9>E+y!|0&4C)aeA?43nLgS7oE7p|!PV?nvjr
zy-l$pHuO1Qq+U|a6?je|?LrZMkv7jOamL#>Ga)3g@fg>O5fB6qy95PIVvhjdmSNb8
zbOuoj^V5O}B^vc-!*x@OXJ(zBMo5K>BhgSuft2j%a`)rM$Zk~_{~AF$)JR_~C%jeI
zW+I<Ho~CL>7QR<*b`9@oRYCEP)aJd19%bkk{t0%Es5!bkUCiZ>U&nPu4)bgxqZbE>
zH@(bS)1~t7mk$!JNtOkl`#1@4M~Gn_^{xaq+RSKA;qNKRE`L*Dire-o;ADI>{dJ&6
zU~tI`iWv7UIrm2Q>(6C9CpmZHZ%&-QulFFe*r_h5R#CCywb_oJdNIUf%g#fkk)pmv
z>Ho-*q;!@37S;!lOVnYpddJZ{-WA<?H8qjEMyP6VYc5D2D}KOeOE1rc_n&c8^Y3xQ
zTal8k%3O-CxNSF?dikmLc?SVMp4HX9HN2WZlYCsS4iwj&qCTD^uDwCe#CmD_(W62T
z<A5l&dddgAFrhY^x)RyIsL{inKzj&10$$xbZ&Q~f@Hm*J+~(&+;n6$(Y6NXx5ZW|*
z=y2w0n|V7cKuU~*+)PN$?}Me2UD`4@chfX+TOr}o)kT_$7T@I*>*o%o{8B?D(9blw
zl1ra&1XKBzRjgjJ)D#Z(f;CD;xQjshdhN!6=Ja?1RcB|)^ofmJQJ_@Ws?E=Z?cu@h
zuUBxYF!-^b;Kh|jEQ7_<Z^%!9v%-a(L87?ic*ZMH9m|w=;wEmVS9-8~LVOdtkpp7r
zTcjd(le=Dm7MIQG7={z2sN$NZ*WKSWuM6v+y*=4*Wu5bBv4NmHPh)8Ig6eJC2C1|6
z$sbv!uf3S-DX{0|RsbY0RQPk}V+U!Zi{9n?>@gj-l`bE+t*_qJHnsqyEj;?CmzeWB
zDK<4{;>BT~J$)fgKX2}#fjX0s+{Sn=7(VWhJY?h1^#`mxFr_?J6v}i&vcT8X{m1PC
zy={}NhmLGpz!UI0k7{~0@wk8U_%$lQ;^L~gewwV)aBQJFWq9AM#C)64gItE9!{sLT
znu|dUMe8GQa0%V%eQ;`QQ%Q4etSZXHE#tIn1NJ=J2r(VPG}I+u1~a0#-QrN3rnpam
zx)@WpXvzaYT$L|o4sW&kJ?n9?qsYDB^cNg_OhSf9{H6M-G9MyOCu{zK1OHN}feC~Y
z?5qTKzPxK95|8!Jr5y;CqNKaa>KDEx1+6yRbrmtor{e1(n|=69+^Yew7@k3+liF6m
zk=mJI%1>_54+Q}oeuq>zgn5LyN|Hrr4tx?~<Xa>&!|XVcE#QeyFj+Mpl=T<cx=GWr
z+~6E5C5;`7ZGbD}?y8GL_skMxFZsz81MpQ>M~HTSU9ne8$r~{ZjK#i6-Rb&m!K%>e
ztmV4?L(Jw~$BPqy3Xqym{k~~F)h@!BHkiQ$X}QH0O%mupgn@pyWh3%u7gEp*#d&Ip
zyu>j)Q7h&%AjxSA!Izoml1dM0lBFs?Os-z__)%Z_!~TbU3IQE#X9E&9@Dg*7t4y*H
z{|uz*aNmSzJu7vsUmQaT{w}Ur)h!gQA5GggMHMCfz6HP)P+ld)A&Y?ZqggD1{35U8
zjtskf;{t=?mANK%QFU6+uQ+G>0Y)~t{C;c@<yD#FKEO6+sVJ#qkzlYhtyMPnDWsQ1
zRXE(?5dYBecYp&N146=S=?a0jVHKK+PB!#D)8!MDc6*fvbvy&IW<*`n@#Z3mpkls;
zHDARqO4`8bE!$=V5j%^@>;vyH7w1qbE1%IQ@*0upGF^)El{SmO3K;=(<nM`;@tE^w
zhr~K0xwuMdJ#BRej2=p)*P`-axP%Tcjv5}~hA6Q_79J3F02C%RWGLL98$>PwQrZyQ
zl6od`z<AEj4!L0?`)7%HNju9wc$%NRkA)hfKy$KZJ2&iR?u26qOtb;2$r=tIMy<j!
z4M*+Ga~DEIBOI@#ufJs?-pU6rpu5VO=9c_$D5m&^37okzne&jM%_hMTpFH)PoLAMM
zvooH`KEesl<|R|mg>o0PCAl(<5}lOT(#h$61e1$I00a*^`gjoyh@^d@ezJtmxu?<8
zz#r-L&+-XUeSIf2I^Pk<y3=zXHkV#{LXg>3%-JkU7<SjSp2?;klN#d4Pj>|uR{M48
z{nZ@`Js+YpW@IvFnexTcOoH2|MXIk{GV^<S*(QChdx7ABpmgYj@WiK@imHu=J@oDu
zgEI+%81|b<^WK0oy`3|H&6p*BpXANJ*V{j5kMFDBBO$$m1rp;X6m8})e~`bZW2?ws
z{@_D2Scf6jDP8~oi4ReFxdvu-FN9XYR$dT(EmzTIGPk{1xh+*wk2S%khNJSS_n7!9
z_m(rrd9CCyRnNp9fC6S%qk$29sE~0JHJ=bgLz*uPDg+bG(F$GZ<^Tr=8cnD4%^{4j
zlo{mQ@y6|gQ};gUY4#z83gh0j6Kgl`1uYKbhEF~Yk|fZrbYZ7>Codruhf?MJ2=kO|
zU$!$>L|-l{=ME(N;tGIY;``A??~Y#vZ_N9)BGGvAj#P{L-aDE}9v&r9ngNg54)i$1
z$C@W%TTgB>Ya$VuszjRWM-;MW16wejD%Zj8fewfM1NYU#_ex5K*eYLC0hKv!d_Y0k
zY2U_&+Lg5ZP%7mGAgVeKK_N7kID4aK=ud*s)6gKwc`BR*(@`=bN|J2xSxTBlY(=7S
zq020I1BtwHa6uQ?Ha)AK8V@5iSpF90vKAdfVm);+)7*~G@I<nxx^&HtB#5}H#vks*
z?uy3jt+hZcl~STabgxI#u=C%%dlA6D&!-Xg%+^RF!k25GvnmvjAst{;3Fh8_ucJ`a
zW-_Y`H$iUHNffyStZ*iLa^ja8i_-9_<-S<_TYwpXRD8;zma2M`s;?`5R9+cPFFa??
zq8M{&UG5W(c%SnaR})A`gxd6?-z<4TJ5%3_LEBYJswk~^8G|bYQ**qv(yLR<!6`fa
zmeQw+nwys_a?7|{|GN^NQ9`e{FA%Wwg_BVwDU976;$qb>ij^JP(8&Xd2Xi*-K@BF}
z!va5a+`#18zXe5P_W+`X4caCqLI~3wV}EL#RU{X8VPc;y_{x9H!#9i14S%m~Z2Vd_
z8Qu)x<wW65-mDVaz(FvZdTKy()=J8~g}8BE9gj|Sp`{5GKuuBjp1;N=d>|d6Z3t_A
z6WCatN$gmtck0{NmV7C?MTJFrIH`{*9OW#xs(|q=|DO%P`F?>jZylSzz%|gpEl?w+
zO>|C|0RyQ-q8>rlPET)NJWazcj-RW)5lsEojxMbNvqIWdYNKf~O%ma4*V%U1?|tI0
z*&)61TU8P|c=^#T;-*al_7WRX@N$vAQ@(vn%9xDCP9EBh4Q*HHicSz|{&PB1D?-Ro
zjZzLm^cnf2qV@sm_}OFtM_7TGT2W@@pkiQG_}u5wXf!uH{@_74EZewUqH>N}4!bRn
zy!ST8j}y@Z5VkNo4#LjGm7KQl3kbct0zjl}o$huBy8Gfn-=@_S>-F?pJH&nMLk(#q
z!hhNM{wsY6uE{FVj`CSvfOH^y{t?OlP{CyHXM@*<4{<d6{R98#bXmHR93E3&jEsL1
zNVeiUpx|>^BzHEWyzETslsigMvKN@m<*+^eJjuq>MfnRWkz+M1BkVa&zcg1hHbHZP
zq06MOCy3njJ+DzW^Qd7(qvJjvPrv_*df(W@M=V5SW6a-XBJMu%V1?hCl3^tW-C+Ji
zwncS-y-|@k#rkaz*XzJM-W+YeR%w(`13rXUhnfmuo>;M>2db-QTgRKb(RDa0DNqZ>
zF3;X5$q38+r|UApvcRvF`%k-lu}~4=-1XUF<quId*>J6K1#u3tE)lBp-i5!<yprU<
zH4R|Em2+DsbC*boV}#QER<$8Q2M9{7*l}i_V#2?*%ifBDXOaZvPAjM~3Lb1WmgXy-
zr-?3nYed*QFZrHwm8p%8q;%iazSq*`^Tulh(O8rk?x$hfNYCL($L8QG_1Wu=3}apH
zXAhwFy7Q;MWYlc|r`=BRrX8WbYCGCwF_D)}SZ9WDCs02PY{a~w!&}n)DoPnf8&7>$
zMLWvLuSTL~wdhVUWk9I}%J;t{YIObL2<|r_TwO)&S9@8On1MGbxud_zSpw#JUlL7o
zl`TVEW&$j;EZ#|_+syjN;Ytnti9VUnEX}O<u(deX{E~_%6Hi*gNv44p-}CZSL*3;{
zYtZL-=?%REzsz1{&_uzOIX@M4v{&cBiT^{!ixAWx1L*UJ%%Kzf^P3`{t0lUTO2rcl
z&#|e1@M~a?czd{XY}7l%4SmyxQH3hCI6qx>ik5iRL#cA^`%edc1(0k>`45ylD_PPd
zkW!+P-j-nG;jhGe86z$4{K#4K6iAr*Hs6dtTZb<E5)d`t^8UE9_Ax4>Z(I?=nb&QY
zdGM4_1G{Q;4ES2K`&3Hr@iL9|BTfmEn2HvVIPhI^e0qw`*P$85!{5=^?fFvbvq4_j
z2uc{Yu7%x1Q$^?AqLh6MC;kUoxuEzqmhz+T7sYi>=eWOqU@22}Vv><aRwT8V*L&o(
zEcZ(}H8w5ETPVCQm6Q)=ubqrf0&g|@TXx5+cc&aotASSn;-5#(r&Rpm8>YW|_`_Ur
z`1>T0A5hzx$*7_Ns=ya%;w0l-HCZNh?8EB#{7$)?npVn$ykxa=ho7<?HIy|Tpz%MG
zk5^h0D?Oq+^8)6ZJ<<Kz?&NI>HjmEJPGaN6oo>8t<@B)*9_Z53B#~v<mFldNY(A4y
z(c&v;?Ue&X8T)A`Wc5Fmw!izXJM(#Pwyv<*Jp}efG|n}5>-<7}(~0{`&zM8gF_N_>
zYw(H(QqiNzv@z;Ft?={aSLTl{fw4@?v~6;vF68=lmK&Q2$Om7Cy%hxxw>tI~)qJSH
z{Ek|o5VLAA9R!bWwi8}aejwfLY;^-Qw0zolaet?*)6^mDuM6>j<gZ5rye%iQc9x{r
z%J%6uVnQOK`l$Kh090FM)RSF{m5X;Ga0T9Wq0j&7L$`nWFb3QW%5;6n+j%NQ8OjTI
z6re)62$v%NoF-4TKR)5QE`*VIhbmjS*j3l>?fjL=A*}3c`X1#a)-)D_5@J*a7ZH`V
z9N^s&6f~X`udX{|nXEg5+bqrWq#d)w_oN+JJZDwgK5AJTlZxquoorl|H8+n=hYPlw
zc8iACZ*i2Jp(bCW`d>j_*h4E-8h{ICz7~1*ei_BN92BVO>20KX0wJ8}D&cp5U%7=n
zh=Mg|OZltPY|4)g`w2s7l_%Q<bYsqG)xOEz^0@xoGM#Mcqqwo}^8EJd95DI%@XK~h
zWjgh_ha9X`+-@KlEKbUQ<Y1ISD+fh}lpmouD5VhBd`OJ`5nDSjcq<`@w-syQbOKON
zFxs;aQ&+C=^vwEM^UpWAoY)CJfL=2#x2Xe6Ff;h_D~Cszz;>ysnM<3bw4La#3}W(`
zXr4sJr^$`&ch%^`k?Z!s>OhCO0rDG*Q1xhe2Uz>KES_l*ZqQ+gkzz5JHH_lj_bGrj
zm=yIm*GTFEEKu$6qPl+v+zQG%2cAlu$i3~V#SCA$|9Sh#Yh8jZD=iCb!yn|sK66hY
zlh<7j>oA%Ae?l8=@<7xN>(|7^XGMnF4=k>^LJbi$s6P%eFHH+b2=dEP`p}whR0psm
zsnl@mx?6|mK?JhRUZ1q){0}z2&-)0Mcpjg<Px{VeumxlZrlITAxKoo=%aH8k!`DeL
zyUaaU<MGJamU;_<DtH@z!HoYg##kB4m7^Lr;C+xu!^+V006f4cuyJ<Cb5Ic$pTH;0
zI;w4KluVW+G{TOc0`qx#qI+LnD2$DJ5yc_h8C%SjHx2OlWn#et#IpBdWi04JvAQB8
z?>0#FR%hLxoC9=UJvVJbTpHToA+~Z9CVwnHnF|u9l=6j?u7=xgy}v5F{{eZ79elS3
zX&1b|qk}!(<tO_zR~xMyb$WU_aQbYL6Ja6WJGHagV|q|T#sAthzJ-*HOe1cf_-|PK
zNcj&mD$;zf#qRH4+8iJ9hC#4@iRA%dJj513!ODkYVxQl#4Y8(+?(C#0@dQr1j!2`M
zjzqirMOH&tYr9+JaAn}tG~BEkn`9W`^}7EbS6>wt$F{8-2o4MP;1b;3-7UBUhXBFd
z-QC?ixVtp&?!mo-J2ZYf>+E~>ckgSzRM)IoHOCmgxd%AbDRB_)X9YHvt;BO<LLLe~
z3)0p4GSJz1@)a(5j1f%!9i!33fDgPsF7G+R(YKdKKrWZp6KtIJ{mt!}t)Zyo6;2Ua
zNJ=dWOXuz&B=4_TE7#ejIfHG5rd>TDi2N0nkqcP;Jm4boJKn1XVmZaUD%>GjtFE=d
zdLjL|m@agP3pQz1dMZi8m=uqT9{Qk7szw~k#(LrEZz`>Hsz~&%9*UV^>C<n+WFb}>
zx0VuS2!qbzYEVrpyqEzyHEcjFy|djbtv}Ap{rMA#Z|8)rEuoW-ni;x*=iTt<PaIYC
ze{=vAP}ISL+Cf9g#WYV_pK5LA^mD|lA|$6_p0k3BSh_iM<`W@2+av@tRf=gI>>){l
zp~OuVD>%km#f_pigZ|M-B|G8-uC}#=Fd{B1X6G+i7Oh3j_y43TjqS_^xC&FS&+U9F
zkUE176qbO>6^|TvHzAOfS<7M6$1QA({`v7qiTAm3Bwj>b59m)SZCtDpMr4f*HgJ=T
zY@nF2(%@wXz7hTsZ5vB_%1*hxDFVb(gnRhPBhnf2i+!4VtvbEWcFN*~H%GlreM(e!
zpAJIBzcIK{is$j-a)5zxcjXAhV$-hZ?cc_M&ma7#EL`5D0`P#f{6MZ|RVpl~R>J+e
zM_=MXLb8&g1!OJlQ}n43TxI69brUl)_+!LklvR8Iyr6Pr)?Tr1Wf<dgs7b)pj?x-B
zek!ycW2^6tLStwuMhR2((HPQmrqv$Gf#QaGF^E2%l~NJi@t*en6HZ+uBtPBY2cUq$
zJj@-%jAaM=*&3B%wcGpO=Sy4;ICJA|I34RcikGn@rJHI-M@&3rCO6$`S+b)OScRX6
z)U}Oc`Vhh}riwRCe(-6~gtTCJ=<k@TItvb~LWp{VeSTjdo|wGCVz=4<RL-A_F#8)<
zC?^pJy|x@C<%(UVfKgA;ix9RHT0yVe6P64yFc$1}p<jeEa<I?x9KWKYsioJDFo4mZ
za5`#}E!T1PETB`3{M44uSca=th`hfsz)bqn?K<#N8gMb_voTX~t0yMTvz8_~n!EBm
zs1g46AS_>$1X-}LUv?vhZ7j9GihJYl4qGIaFtHQoz%b#lPm=}uRg%vnK7Hfk=tY|X
z(bVBbCK}5f&g&2=ycS{EPg{>W#C)N5y*)^8R;9NY(*`_~&hH}<z-=CI>7nT8>y*UX
zE8DkrO8nz#R}-T>Gl|O6ISnIv8S~|HvWoa7BTT>zWcHo1i2VEoMyn@C1|^t;_)Pg9
zlzjSn^_WVlYy{k6e}`k#LYB?XbuAeMvNYF)aQwu<{maPalzlgl%BV@_=O&E=Gbl1g
zM<O1$u%b`m%A>C*=x^(a4n{t4#$8H)Z4*BIohUpcNsMTxQBjhs^6LFR=T;V$GPl`a
z7DRJtP1d=g3%3=HiocV)3JYFrdbDKQ8tT#K$S($DcbPUi*Q|>tFLn@nb&G1n1Vhhh
za8VF7tP1PMQXrMxo{G&Bl~0S!k}H|ivC_9l3}4=Vo}z+AMGdn}m@}`2kXrVkpWldv
z-187mz3%MzhF(v|lrhPY;~e~Nq>qlSUs1CTO}$>kMda(c{2L*4`A`w~iJ2mMDgc~4
zHY&2g{xZk7r=Q9ozw2>{w3x|o@RJZ<1fYUs;@XzTEgzLdqLYZm5t-o%ZE;E{hTcCF
zacmw2gO^ym=Evc#rqjVaiiw#?<dY-P{lB9aS(DORA<rDYGM4>UzJaGHLI7`A0_wvN
z&wER2oyR-EZa;yN4b55y2_w5mSxh+D34NIm*9y|KD0@9kX=w$+7CYJcHkr^WEU9u1
zV>8oLaXumJqB?vNVElJ3PEbn7Hb?^A=}&vrKge=T%yHpmrb+M-#ONb7M>uKl$z;f)
z*=VEOhHr+G57Q90QrO2XYm+(L6RJvr{iaS~LRi6fp3FJ1@O#MT?L3^WZfLop2mvhk
zlwUd_Y4o{m(X_0tQwl5o8DbZ6qgz|9^E>9{W02BL0Z%}$w_~-Jh+)E9;3hhZfreT7
zo<Zv@@(L2UKMV==^Y!nbJp*8~2Y90DTYev3V|EUFUQSJg+e2P(o_v+fb5^_dT5-iC
zICh$`p{m!+E1gztaYM$9d#;+gulET2hOC@28HVPtA*mMX&0EmDN*=x*Q;Rxv33nWT
zuFwVDr6qM8(t`+|M^$t8CMi63mRnVfrN0^y8$3}&(SfQt1DG7f?P7_uix?3eU`)Cw
z`Y)UC4l}3jBNwgErfX4&P}<II?D{UJR5J<837Mpx^8c+RB>vVCh&g^zf1lLwlfwC8
z#K7K?%5xJoSPIj!;&j%xvZt07bDq4WTj;W9D@%pePg3+L8S1w+AwG==LSeCl-xIDr
zY68k5a}ud~W`pk&z961Ud4dL63T!T{G(9>)Z^T!a5f4bTr*Po_>mMhy;TFxj=^bQO
zn44RWfk(H)X^Ybh9&hNtRLE!q)Rdh7uI^N!4>rHF5U#EqDL0hWFABH{bg_UVn=@VM
zYD&0ukVxQ-Sg;EBZUBNkHbLSVPw+1^$pmD%Xjc$xF~FN@g-a?q1qP}C{<8oQjE%lQ
z&hdQJje*v2l$fW!`<qO^rop4;g%rUS5b^PKsl<HH=*2&u-m1nX<_}ZeKpLqeUxQHu
zGXm+)pFn%*Y-lxqEFw{wWa2$oc#+w!&8a?gZRZ?zChX<l|K)D|=pVw>lLNQ{|6SOv
zI~{RKX8qoFN^l38_A6%yH06TG-_Z>Xsbq8#sZkkpM;`4&0!{WUe00R4@HnpUIPW3k
z<$=HWCSF5DlS2^iG_6*(gysh;1@Xl$G}_x?k+?(U@qh}a?iu<wbS!8%BkG)Dmsg^n
zb`ewOm}MOVUVoai-Cw>~FjSG1#MvY4>gVbe{(Yt&56lZsd8RoZyUhEqaeP3oZdI!I
z{9X^Dkqhn#GGuE>tAc?9awm#zT+QhCy}E>GUx#l_()RdPNYoPM4guE}$V>K65hhl*
zpA4S==$Pb4JPFP}Iw15GG=-ss%KAH@hkTwHpoyyInAUn_7S+VEkRuRpt1GRdtpc*(
zU+jMGi%^vsyX2N~{f3G#Lior(cuhV6;_`c8V)1<?0}}c2;as@0)v(@?$^D_v%`J?i
zIiTSr7UdL1j6g*;XV*5$P$a#>q`hM9S4L`C?Q+E1r4I;Ry}g3WbFH0t?T4(UC>Uj{
zXdszj{pvAo$S!$`b0Wm~_(p}owD=1HZjPYdBiFEIr37PyLPiu-Qrtl_(~RKkl8rEC
zPF<nKs>;b)uScGRiKV{)GOjC7MR%=d2rmW|nL$k44kBTaG)YjLA)Gf@KT1~G784zh
z`Wad9_IF$&A>Wk2`xn2ohga4apIsC8YGk^RwD2wnSWTveR)A9N!ia?eYUrTIVsfQO
zet-Ph>aaiS-7a&wCh-reC&j|1ewrXvh2_MXx358!oS=DgE2lhP0%1-`nYigaNa+xk
zQf)MZ-?)l8>u0GvQs`1z+hq%vg;(Xc@y_Jj4bi40V{Ix^YK3z^xsksuo1YiBWm9UH
zHqYM9%-+5$R+NFC{*<V(QJ|ORthgEAc1ImRPXn{!;39%9HT834?d#V=`=grT!B2`S
zT9uafN9h-r`Hc?yFDlE<9q`o9kija3l+Dw#A%)WdtY1`q1{4m)Aw|IMB0{W;53C`m
zTd3{rdJuVllhAlxUOJsS7!viv+PyAz%bED6b_`KSp~m*`sfgSefVBcM_d%pH-cJZK
z0`+pMCunYvJ&&*ka)*cC+kA6EmbI~C-3j>i9K1g)sb`Xakwz7N9{CF?iu)7w{ScTK
z-0v%icEB0m>0DjA_t`vNsfSDIyqk<7JE|$rX3F1Ep$F#TnS^f41fS>Pt}|wQW35&=
zsZk~pUS;rapcUYgnc=v7y<=8{UU|13T3T;rgHYx!r_e%4jWE%I(bU!1@W3?qwY`3)
zncHI}zt9<<hsIyd;+}7o6(+aAGSbK1q2<CMFtj74zA2~X7e~*fM~20}w-tDtR&1ca
z9!K4ew+Bg)tb`61J(L&pQUqE9?ciH$aL3$bhYivj=RP_Ua8tDwkl3`T2LQxygQt{p
zfd$#UnE0o?G$@uFDqPwc87+$;VWbpy1QE2?Md(bd)J(R|wLc-PbRBw&N0keTr-$7&
z7IJlmEj9;#fFNYYarW!=SC2Gh(MHe~_@%xmJc4MZ2|F<eo6@yJF2WY!>`QqV`4@l7
z4gI?*`2KE+mj7&u*W!yh=2kGL9GQtJUA`qFPT;4qa0(h+02NmMV<wxU&D`)_RbR%}
zO-=kDM%m!YeJV6wIm;0uuj61}Dq#fZo_tP@uQ|+_j7}cr5mplNo7!ex>2t_^6R?w&
zq>okdTE|O+<LEBvlN5d|_qa5;D38<=(Y9_G6c>&+HD1S4+rt|PM8g{4`2&?(Aj#a9
zoR7i6D$;eTW{m!K0#5&Y!UEd95;|RPhgRW2ZpWr6r46(C6cc83d07c17*qY;nuq$^
z%GY)-K*CnWi!?7`qM~T$3YBeh#2(blEnI)FVlC+@b*R@xcizYQ%o_zCU#IsTM&6Fl
z`#jTigpzNfOs3HXR5VCMxHaXfl&d>}_+u<@XUfhKabhA$!NHj{pmvbTUyPR6hW@-=
zu<U|=>ST_*>t*xO*m>8XqD*jVX|4VMFB>5}6`3bQ2L3qUFir4@U-t_Q_Q0#kpiiw}
z7=oxZ615C(#*X7yp=RR+jsATo7g=(@8eVH&drtc@SrnZ>jrX&UnnTv+G4lQxQPgea
zR7+>GH1LWKUQri;;rEu^^`dLo=LM+|CwV!o;}KAjjiNXYRhrwei|aR;;5+`=SYJNF
zHeJ@0OMiyzB<x|{@7lXk2y_h7xZ(_`D!VihSD9`9XwyKF=8yksB<+`wM}D~Gt)zyG
zhCpUxH3*mTDo_7AX%&KF%lQ~v`CnK}WPio9X_)8pIH>?}2|s-z^G_p5WKFAvUEG9{
z!EQmuE9Et1Gb~(0$SRf5-m*!RLXu*1=<{!NHiU#9%z~0>Y76OB>t!C!?%CewxKak4
zk<F3aK&D2=8|rG;U?OwxSGhdkq;s4X4%7VkP<Wk#ucmB4Fo{ALb@^3t*d?^P(d76w
zbg_Yne8{EYajaADEuwhM-orxKD(?A>-%F)m=+^hB>KH=s<mmCWSGV2O<xBNHO8gH5
z7)AElvdVNY5rz{f-q$z0w(pqaKvnNEu~8-8WQt>Z(|c+<Iw@^^L2!VM6gXl`%p(%a
zK4e`n5T>c+X+YT-oilk|&3kL~-hS-PQUWv3Uz7ZvXF!8xLQc4?pZ&UGtun*ZaoxL~
zV(fOzeV^^Fhg-W+G|If%&b@Kz4tjB?4BJMIj$kCez`VeMt@j%nJBOB+d3AM9O`o8w
zkP`A{F|hV_VA>Ly(V&`*GfUu#aLXuby(Tu@dq%GJC+>hp<;{m>AH+gqS0snqZ)Z$S
zFXt2Dnt4>&2BzR;fP>Z16qkuh*^h#2C>5ecp~M11zGfm_>BhC9DCWk)`!q@yIY*-H
zqjM|Wsr#G3&N}f1+GTzZm?|4n7{Jyx`zIKbJ3ayfZK`lueo@|@u$EX?3T07LcT$O5
z^sfvXyNW4n*i=X&xB*nC>I$4|LiAA8y}Yl#Br^DtZ{l;~z99pVhTDD^3Rv%_h0Oh1
zM)-igCmIWX_f6ctaBBRtRrN4;jOwh;pCkl@rg)M_2z^}R!uVO9@!jw-_7MZ^UTx}7
z6R7+c2!dQpcVFoY9FR!3xmClwie})(!K$dejaPtFMG}4Y8)*xtl=lg_63aAVc8DAf
zGMvf4XFKkJ8{O^E<y$V)&m(#l{5kEYnp^=B(>x~?zV$(}hXFTzz4mo+;1;=GP{u8x
zd7nzUm2vjk$cuR|&G)_41L#FWVd-%YvI{!KBJw`pM1oZ%^4ep4DK}xwb(BfTZVjCq
zENbKbaRb6Br&I9XU9xj06uSR-#+%-7K&f{9qC^wyUpV=+{l5D#xMMh)<F+-BwFPno
z2ggQ{z~S@A#JRe~y7^o_-t35>Tt@*w#7|ToXL*g?jV~j#?TT<(diqC~uXjq<NkZ5A
zf6WG8>$JOhoV3Im_rP|lN2+<x)HD1qn76k+@9*0$AE+TL(;PlzIg0J}?=<g#kK2#Y
zG{H9yT1?-axnq#Hw#1C(G>wE6e@T*TRdUzuCdxvVRU-f}Gd<0ypATj!-Fol%#MQA8
zEZnC_Q|VUaaLdi>3UiC8n<Bx!IImacdlv(Vw<3yaPH{9Prg^F^S@4kDf4Ik1hg=#z
zov|R7iV?kakI*=UxP(UGxE&9pm0j`<zqV`MmMF<>c=p-Qtt2Qb>LK3D@4*KULnV8J
z;7Bbh=0`!Iy36iDanqoppPc*fpN2mR$dX<ae)=sNj(s%xz@~F6@IApKdsru?-;2fR
zti<?lqYgm(!3$`17TkYbK6pjpC}&$XREU|!s)%nIh<C12l1g=#jR*G~Y*}UQ+-J*z
zee;mb@v9w@JIu=}*!V)<P@&zmyiNMz(tid`v5oKI6R+!$^DP@UIfd*Rrn$Xd1&P-i
zvRAEcGHLO+B$H3*MOqHGd_O$duO;8WbK=~{Uyz-xA)VLpO4m?GQYOc$kpJy_gy@Wu
zT404LBL6XJL-73>JtAZNm=<ga)$jKn_nF9Z>hG3Fyald!T>PH7*;qIo^-NRp0Xej^
zeC1bOv|bABU3SNGZ;+X|Ak^Nsdz%7`1U&+fv7i-PW1O_i>zA)zoo3iF&m8Wb7rf6f
z00Vk%J(0w(0YX#VkAlI(o^TZ7aauax4{AS}0h4-e+e6!LUVeGNJb*jfVs|08*Ae^2
zi_q)Gbt?Gqqp&1Jk+=-bub~BFOA}p>t-dklxjxZ_qq91706=&%a+t7eL@odGn?J0<
zJ@5O&Ht6N`yzHG9OmbgrH#Faw<)%GsJE7sMWT||Dt-pP|1-#V)CJEh$4`cc4ik1*O
zp2y_8pAd1<qowYiAqp~B$0R3-+yB`e*+Fs<G<(TlV*c21GwH^q>lzSrv`9Qg{_G#*
z?~v(WX&yt@y;5&)8*x~qr`V#Y&0jzKaOjeXP3HOLPuA#t(-_B^B6pC3P262Si~?iQ
z{|-}b``mm<r-quJU_te(5q;N7;Wd+!dV9p}o6Y$?G<uY<i?O`+u1o~AQ)zn%;WTL!
zLWxyXIm`0S!k>liI!9mle@`<zFm7nakTb&#;NStYr}I{RS%=glmzS}ylFtl;jc|-#
z!E^~>#*P`h73&gKIYCO&ggTGEBN<Y|%Dp7_mQlwB^~52W7^k3y@|39DeeYE*qI^h9
zZ{sVep`@!ESzQ;}8I4GEs^y*XnZuXByM9gL;&wI7(C0gvNFv}7?0dD((O4Jbj^e~O
z$;_}-w`8S)%QpSVzAr^2)^(JlWg-H9l&aa5y3!=I;;|`)qRcCW+v5IwZ0>%s$B})V
z9$V^ONI{efysn&asBemf@Jg?bhC1a4NVfBSS-%SOA0c{T)m=ZWvePp+@yfZj7<_*h
z%zK#xeiFkBY4T#>>s;r*`GezeZnZU<_f#)5Q=a5TQqn8?P4udTt1H*<(di>G@A)`y
z_nIoVFd-HcJFE>ys{gwAa@iS{_a^iKU)%9wTWnFI%3U{Xe|+QgdQ$rinfI|*h8Kr2
zI-h+^T}CatL*{o|`$36~eRTZbSyfH2|L4S}s-|i;1i}t<(Ry78xS^Xu%jL9Kr=YXZ
zO4r~hn*egti8UO?19KUEuWpD?$24TF0CtJSSt&6gd0tr;x{mH{5a8x8jTz~N<M<^V
zK6EX}$LQ5kc+SX?ZP+10(1>8gMndLfYDTly`B>^No%Y9h&A~?d0&XE{664P?QD^62
zRaHbLiAoNeG+r{3A#>;`cMs*JGPuu!=TE~lUG(Lrfh2Gai+@zaP*I!t(z~E&;C0|T
zemYY`m;9pBkx)PnQ}Ya@A-Vb-OjBP213$gYB*(w}w`3lW2ex+nr(kw2gkq;@X<!W&
zEvz0;C<{LKS26$NEK7o^PuGtP?UI1*x_+RpurmZQlGjT;kO-~i=T}n3)HYc@h6SlL
zd5Z{ICQRoYanu5`9{Ty8^oN$GIV}?!o7^*{!0DuXm)&Ww23~yTGu79N<T%vTN6mIT
z15b~<oG=h?Tgi%ay|__qVJo%_{g1+)+$PkppG|X(vWM9s?T@KpF~op3MR%SUCtGMb
z@2qy+cK_^CWWfF2THh~MdSgth6_{vjH#;Ffj*8ERjH_5)Oo1UF27yN0&&H(iVjjv}
zR8QwSO10dTF&8rjY<5{3KX-HoKZT&ZysY!TgK(bah@TvKE(L_-=Vj|eq}3JMA9KB)
zaD-lFKvsVHW$xgNHrvqtw0|J`1i+sv0O+^h^d@k+Acj;CJsFr(V&V(*0rb_rJ^PuO
z?BgfN4EMdKUsDfc{@4N25FP;^ZV`oWGq>@D+A$z~>Dt!bhxc%qO@fjq$k*Nhb&fS$
z80p+8hOK6E?YrGvuAZ!vAqjNN!r&0kAm@%Cp$!s3y*)!?__SEE7zw_hQMU2EPv-Q}
zwJaUYOm}E2{Gfm_)A3;vLnWd^@Op|FvD40ZZ^!kd=0nGMzr*MH+*n=+x^dx58<2(@
zU-0+_y0aK|_z6E}#Cv<@XnfmO<&eUXkWfAfHj>uES(Gd)mcks8o++YrO0><Yp#3I;
zSc`>rV5aIwH+E$sFEEB?lqD@~YJ>lRhO7`2^L|w9llAW>2?soGsjmDFxiyCx!Z3&Y
zR~gh3vZy9#*k~MC1^#=ra)6tKB<-)rM8o=^M07d|YLs2*w()Q35{&|<FY=*c2uK0c
z5y6zz#<{h!FnMo~R&z4b^vl^6wO3z=J9=+tkP?)~sf3>K^zVBaN0ge%Zz8Mj1D8T1
z9!kv8?=wnQ;Cz^MOgP-pPgtcwh%}_dZQ{)$a-z#m2$TkcZzoJ?#Fvt9_D1k7(xRKh
zbtTG0U2zoa_3=;ocyS`L#|q$!_^J$e7*1*}D=Rha?DwBtii<tquZ??)SdjM*qI3sH
zMjMlZ-lE%|z~%~1PX?g0?;Xw&01(bQtsGq?l;Sr9)e8Vn|NYSq6bAx)qyWZaV(x{I
zV*D()f!wZl)E^^4&r%zHGXjfd3bPkGLoDmMy05Q(ARfW@h;0)utE8M9w4k8q4oCJf
z&nX{S?Sf#lit!^kiG|uKXe%xjjjli2>X0QK1$tQzxu3xQ*!_|>&imDk8=p4yfe9lM
zT?wCmrlAQH-Nn<CCiqtF+ZWoj{jcQW+GI_`2oVTA;6+*e+GAj)psp@oMl)~*^)gOY
zZMr{9v}|T=_Yb(im^6vgr?AB90Yyk<AglO-2p?5yg+^54o~$u7)AQUnaE6=uKSL@v
zbau)FlO2CeHrpb=q9xviCdm>kFGPiu;2=b6U5imJ*_H>lO6d(0vlvZTsEZIH4S{bJ
zn^_H2`kc9caV2ms_CE|(hH1N>DrLc8)(?>IH=hSh3Cj>~@QvGp%P4$zid+<$h5FU;
z%SDTccj&?zk!_GfJ)lIy^>j`uiWw0DnJMxPoJCQ#Wq#7JKz*}%kfd>{H*}c;7QDpT
z9}mZiemSAvIdNLg=53Lrx@l)_>-Xd&#el_3QPNvnbO9fKZEMD#BN!0-9|)GUtkqEw
zMRH=5J0vTI*_EyOD5}%#!YC2~JNhHyKW`rO_c5-(l7}WG`yi_DW?OPr0q~i6oRDoY
z9i_tjW(kT25O(OM)nfKrI1C*}9)8I3R+rQD0R=y3Zn}16-rMsIHhy<rh4lZ41W#m`
zKL&(Au&<LJx8|cadwhwgl>QX4AmaD!5C1TOM`l^k*_Z^Q-ltooTe^?qH22pCCWrlJ
zY_6J1mK-&$iAyVDD*_mZ=gRecD_!X>(EDj^A6fBEZ~xOKM0a7Kz{-BWA?Ga-oTfi9
zng`K9m8ccIe)T|lrYA3H0;DdsmI?g2kF_>Ck!W)XE#e6nTeFHBr;<a~0tk{aQRXaI
zS*SW4wj(ZP6e7@(5uj|kL6POXoo9BRBk1JA|BC5#OH}-`TM?PRK$@e7D!K<v>Wdn1
zletv)^Gu`|QyMFuF6K8eX^c%Qql=7<<IS9{D)v$oTyU5JwoFpQkEk<BC;o+}zsJq?
z^M9CXP9wE2RcPN*!{>-(EiiXNAPcCHEn|iXgY$8t;gixI##u#Wq3}--^whOB2fG)M
z*^w(bw3T-v{3)(QWV7$;+h_L9^;&9)>^+#{j?b@4Uv#qhXyK+er>F!S18n(~^y995
zBJOSNkAET_!bi(E46CeQEnGuc5z4~tF&C~Ih~g`x#9mj4EMv7StMwNqhdCtn6(9WR
zhx9ofweFY1zZrbsbxzZEPOVmsY>YuG8hY=!c)Gvb;PQUF*IVlb8JHOHmV?buDtTa%
zh>ig=PnYj!N;f;7Q3C4VJ_|h*`DLQ`{ds#$+XmJCx%BT#Nem<g^*mmg_&qD_dl9Jw
z+zAjp-QYYa0dAtc_c#V1f=%QOZ?Gk7Sst=B=?s<-006?CryTa`Y719EN5rxh!asNJ
z#(rU%;O1GUk|F`G8SDGj?-A|@6~UC|HXZU)^cma=Yx#AHjyK9ZX5-`)ty8elXz>lf
zoPJndId^A%$wni#1hYj>PuG@@MagdH`lxkN%EdRR<Ow^ky<?;~fcz1R=$@<aQ`&cb
zDz*-cwwbbL74^_p73m9bEWCEZ!0N{h#vw_T@-hqY<}bjFw!G)Xr5|!fVY-VD4QNC@
zWH)YtH49P5^ep<p?LJ2EuWklhTl7o=h`@~Xf2Qp&KrXEQ6a|^FEJXsQ28JPGT2V{*
z)f~Ah?nc6<NE%;mI;_2ogT=orm=<zJ4J58mE{x$;T6rIvL?*VEkgH9jRwV8_*(93y
zBySjdlu!6P=8|ODIz9AsX7II)Tj}4D(0;?!y}DBcK22?VVBT;X@rM{_N<(|(vJsWw
zS{+Nl-^xi{bqYAEe^ZnkbO>iau+}S755a&GAGK#Mphie+SLa0W7QCB;dB4WWzWd$Q
zBU7T+$nkShfn#^7S4S%03!}kWeg(am^6jPHyTShZtHEBt3-KV2T;<YUncs8S)2x)x
z@K)ah(_O6;U>gW-gttpmnwU#rNT24@ghpmQ`e&T(Ps|wgyGEa57$W-#Y{vks%wgyH
zmJ`GG3fROtl)73+Vl!tq(_YJMPJ*VatH#_F!mjqB`3fP`a@N}DQly9O(KQIK`f*P~
z<k5%&BD|*j*@;X?(ucb%SPbgSql|2N+dE#q2kILxkA&2CMzmr!h%j_c*5BL5#cKHi
zwmZhBYYtnQ)f-|+3#iMpMLbA@Hb1LTlt&>QKQ(ot3x#KOf1zR--Xs6z9K(rk&Y;ZK
zCr6geX;Cc#Uk5`4HE%gSC-{T6a7j00EZ!>bc~H)7<L{IN%rjnId0H_7lZ^ZRGI3h+
zJcH}DOUBEZ#tCHRq15@V_@Re)`NGr?QpDra!bQnVP|&2iG~|^Sl#B5sc!y>L_({uS
z$mtN}&3*&CgE}v|lUCPD4K7e4`0y^z;+0z6uXa^8?kiUIi@_9iMgyPRvb^re`^v1d
z^O$s9!GNq_s4uap9idKjot;X#X;1+(<MRmBip0AwV0N3Qy>Mp^yKb{iQDvac3fys0
z8Q5+}ZSgQ7=5%$4v-?oc;CU8U-wa(~QGpC_wD)iRh8=sl()#N(v!VvX_3>bEz49@x
z<nWmDXk5J;086wJ=6ePE^a!{j)}JfFgLo|`ekRqwuYj#v20IOct%iWxui)TxI!z|0
z>wdGAmWOgVw>E|kLjU;HwW~W5rJGo{uGn+=+reR?$C}MhvpD|N@D`=&)2}VyRAAP!
zO_@<-<X__iyDv>0J{qY%OOulMnSO6>7H#Y1m(-xi8z}iHknmE0QVia0u(5^4Y^Y-G
zAN9((JzpuDTB}+8oMV7Y-;YiX8BXX6yqr^d1*HgVi((G@55EldNb-p6$nYENA%-A$
z7iwhvO4*Sh6_qtLO9&Z$BBDi3klp<Pn=9;rpvQHpM<bHT;~^-KhT%&5LB=KcT#&X=
z{9g@It)>2t>}}fY?d0E6At>E>XY<g3za-AQxcQ4Q+-ZfM1Qp?{%Iphi3{NG{MED?p
zqf!{918C~ndo&WQo0M?qKgk^=6BK_E#BP?GRzMq#*iYK-hGJX@dQlaf+6dmh-&Sp}
zntC~<#rf<+y)u0))Lv~F^h`S$L}%F6I5=J__MG0mO&-oP)|tvgfwMp%5c<ZA^GZre
zhPe<#G_b;V1sw=f4>6*#)zkxMtVFl>swl3=<O%ESDCmHSS0yt-OUe8y2#5OZDKDBk
z_T15pizQ}YFbZ+TU}rA{@pS|6mR4oZ8Ow>4dk4A|ddF~HJJ*~{RQZ&nn3eZ>o!6U0
zw5A3QAjt81yL`t=tNFgu50;;-ZIm%G4F4Wbc)09kNnnF3nUxsB5&HOhN$`a`kl|k8
z%qa>kna!52tf15dNn=fL&lA+4m*K0}=$RPm8&;)w`nI<Gn?|*?nkEl%Ks5`sdxLv9
z@|-K~7f~#iZ-=MgmDsPnsz>pSIG-Y1wzIy;K;a+x<C*0(y%j+i+dgHB=89B~cBV*)
zyng3Y^ect}4!z@?F;_*+Y|NRPj98(UC(tzzM<QHIiTac#8y!4$P=%O>&%O=!$s_rh
zpx@E!booIzA2cRsyYSz)8ZjNrD|hf!<#V|&vC8)Q@Z!0>xGZ_d`eY#zb0hGLL{uW>
zs7XEACDoeUsHuC#-Y<`q39W^c#(GFRhcTm`0Eta>y=}s)`6!>2AbOZyd}2}x_}jf=
zBEsz*HK@~W|5X)-$KwIgjOMA=iwQN^{7NmHg{U=8```g%#QmLXYbNlG|IIhbO0$)@
z^oIi&v~4be${p#m5VnBE<OyIz8)nT>ghN|&iJ8Nap|=-=(jWEJud~0Q#$f^r^!<bl
zJmjk=eW3Z8Q4Ow(JPX2WU64)aJNZ6)3kes1+yEdu(7MpQ+ZAu*%OxPM^Rg#u#^)We
zcGYFm>2Y(z@8b=(TBc0({ml=w^1aJBz)ExF_a>fUv=56e@bq#w<z_|JfnG;S*Teb}
zrOo@0w@iu1Oo+*%+`l)B0v%URd4;3rn>>{&SfIH`Svtz1$U4oc;RQAr<&#xTJtLG~
z-6nQik5HYw^o`wQ%Go(X0YuTK@ZW=odalt*hhY8oJ%U|>0}fwRRj0R{GWX?dTm?Z!
zM@6@|7>>+qM&Iz3OEcGK8mda^k#c(aY|p;Q4<}p3vcXVci4d@~GAp^61r%?KK(sR|
zSGUpBQWLgRl*bf4h<)chczxMc^4Udb6H*g>Suy-Opfu+N+`GIlfgkp%y|1c(n9jg+
z8WH)KGf!jw(}p?m{%E}*qFhw^`R9hY;2I{p<H+52h-3DFBnSswbT_f=W?ZY$h3vWa
zPe7XI^Y9LOt&67LYn<}V3M%H@$_tSQwsc?%HdxA5z)>}Bs>l0vB;w4Oav8%BZ(<wa
z)y_cKG51)?_A~W)cigQWu%q}8!b7lhT|rBiPj>!9ISqP&-+WreG!t!+GAKvjW|=@h
zPgOT)j?w{&Q3a$wTCy8o*x0*d&Xgn_(#Kqn{f5A(7L3Z&qyNf=Z~olj3bvI-8xb^c
zziadc1rZ;-Y~6kyl$qq)x<(fK2x00WBf}8A0)Lf*xFe|>vr&AmK6~%h&(t?I4ox2k
zx@@+&%E4b$a{2>0=qKnP&JB(KiEyM4b-2%vBk%ztqz%LJFQQluB~m~Z$(?ja%1yZQ
zmu_tqZRQS?;w>YTdPouuQB)IISxM2{a||!Abl_i~INf?*w}Zp!TfHy52L=fAa)1dl
zdK7sGH_E+Xe)}-y7di)vzs(dG#I1dHy0cN^62vi(BHLu~tmIDCs%ogmD}DyLadF~W
ziD60%Iz%TTkW#MUySBQmZgh!M=FA$T!!j9u*G<$n`O*OBoH&X>8!>pJt?@f)tnxc%
z@f_Z1PrG0j`rl&Q4+31AftxYL7py$i#_u2yj+WOcO>7OL<8?o8-qQ`Q+<Cm0;$+e>
zx&corHa55Kiww9TizyjpJLwQ5LlxFji<T6VTwoc0T+Of$1dx}P3^bC+{{yPJdq8{1
z+t~|yv|z~IUk!S6d^nJq?-GW0f0$Uw;CIX!+bWtgH`v+8z^5U}QM72qvZABA<3n6*
z&ZizuHVbk7hK<*WQ(J$%x{%G|%BQZUx{T$avxg{0KzGye_0-k<{grZDH0Wo8PH7kq
z?-fJr4=AS92a=6efHngqR6zeDe_a=@eGQ7=I8P>opH*HP*0sUb3mrdbM=YD)A+xpX
zc<Ob3=fhosKRR$0$YAHmhP%3|16=XD6=EYo$>NRz)0w4zJM${h6vs~u$ck2TsG9-K
zcXS}yYJURkHo^L<&{$QtT`^^c-S#K~A-CwlT=XHL+`0oxc!_{Z*zobI19By&MG5+?
zBn(*zZ4LQx7PJZQ{Z&;bYt^uUZ&*r?h+km0y%I1q4BKjS4e{JFGjPcr;xHRYF0Smb
zKUbD6`0ii}b|UjAfWquo11U1o5aw;8bqUVVw+aN;9J_ShJ^Q>eTkKWXgM5WCsiY~(
zZ4?AE;}uF5zlyn&8n}BVbz6@8T-m=XFEWM2$o&BA|C!w5CoVs+ThiLb{;!;r$q-z(
zff>}P@ED8Bq+fZs81;dj#63Z-of7XMRNSn@jU8)JvLS5v<(P^HvPk_H?m19tflDz?
zP1gm@v*fv_`Bn;D2iI9Rweyep^U#a<crf6R2XETk1JVB47BuqKBPDCrPH|jQAo;!j
zs@}e|UE0zD13_$)d_-aRtgfMvZT)YR>BkKDvc<K5lgyw^7A5{0`dUGsNPW8m`iZfV
zz@My=%ak${t8*N>xUMxes$PO*Kb7S}*;6C8ySTvIe3!MU-tX$kE-~+%p4}>ewq**P
zo5PRZ!5E7!;FUix6H(Vx_=g*qE-&lq_JD@$`{Dx_GkA*f3u-J=YHVu)SK~2R1^4Uy
zdwXQ&wn7T(n6D2nTN5$Wr)L*uY&prR1D>8LvXDX~$mIRZy^o)>?H4<K@aC}Q%Cj7y
z_@2?w!21vj!J+nA@V?<=4A+JfL8^RKE(@VEd|BhvHIJ<;veHTDU8X!Vjh;{`w)=CB
z!|ic`Y4Ci*#(&j;Mi*ssBFoC3K5NOh<`5UsF44d*<}T>LQ-amf_TfCWJPo_rbwl05
zJ#BA=Oh;&<n=~4|zlA;0Nz)xWEd4uqndYAqMtcO;iA;<##aVRx?Z<c(V<<QzSsH$(
zUMotAUY?-%#I|)x>t@vP0iA(93b8c2+KxwCZ0t4fKaDd+UwN9X?#lH;p<n+t=QGZ}
zr>Fm`6}&@?xy1akE><Cn0Ug>I`xTdO_jxIDh`f@XzBWImPZc5NGsHA+7m4+BDq(RF
zi*6Y5@iKm7;Zco5-1X?UWDZMCW19l0Oc5KxbK5Ds92N;4!+2~=4r#=5XsJp<5<MIn
ztENh@lfdd26ky57irLJ`DQ{CmfwLJEmSDnZ+&Qhe1pBm;3)l9dK?fVtEOcrFEW^L#
zsll*6e8yf76?2On$N#E9oM;BvwPy-8bTCTDd(r&B`9%CSD)rQj@=7$L*L7uF@xT>`
z8c1QwBGlTuH;?Xsxep&lKapOl>iW=v7LremP}L|6WWzV5a2O2}eC>P%_fO$NOXLK6
zu6I~?XK{pllC>*lR<|(3&sUe=S_AJNHhXm7&tO#Wz?5K!o~dTmwpk_->YUvpp|P_w
zYB|<2(|6+BQ;B9eS8(9bcjob?rM3GXg4ZR$^~&nz_Dg6Uh88<>J$^u+l8*ey=bG=^
z#`n^BxeRkUg}3mWGgl1OHfM<!TRWpFagH4W#fR5-Lo^2*GJIy1w!I1Dw@se6i|Ptx
z3kBm&{JT!h<r4{q@W*#(rIxBFEDr_Sk>R2tsNbSK8NlmA%sd;4D=#-y8`~%+pt#m2
z#3ds@R1pZsR&Gkbp~~(ii;kA!2ZRZK5&MPt!nb~AV~eKZ2&~6;F#JV3b@YJx{gXwL
zllx6c+{1tC{{Qj7;41%&npmD9KavS{T9#$n`(r~zeTzgR;H<5zj*mv&PzVM5mK%11
z$r>3i`b`t-#~E0HRw3J9DVv(Qfj3a%F#5iV>ghGfG>JCi1SdP*FsK9P-KQedPl~2U
zo;`pM?<=w6VQMn<ye!xW%agRhqQKFDO%Y&f*0(<xVTL|aCqUEY>yf8O&*O0irLI>L
zk84EpQ-qRIcg@7{&0pTXDV0~Q+`es~DAf*X`__G>vnR~08z=xJ((488+}t0-wx;6V
zGK2<9dEbu2spG&QdCSYO4IK@#i4^YhR1>V6oSgdKyT6#s>5vD5S@n|p@IBvwPI({f
zLRVr!;}QEK;DX5Do#X_ADwf=T#`k@}`m&0RNNMyIpOn8?%{^BNH~7xWY4RRNOIOz_
z*v@L6p7Kh}KCqCw{GMxDQZX^u@7eENb?agCQkdu=&aLNE?{y9AQL6qHBLsv}n^rcZ
z`D$qbEpJOu*W46tj-|-;Wd}h<eTANoZ*2I5A;~+>R3)B(qw6TvtVVy+%zXRi%y!5r
zjmxr?pSRx7vd7r6Kw@5<Td&}!j1c^vW++8>IC}Je7Hk?)Bur%j71Phi4-c!vh}a<q
z4{e65)GXEh8l7N!5?n+wq^qEOni5qjc{64mRha-{7_}71G%Up139U-_eW;nZRs3S>
z7>>4ZVOjPa0Fp<-|GzK5uyza0X_AiEH(Fb3+u1nHYoEAT$sHN8+97*8;&KXyN{RGA
zkaGA{rV0M8u%%pfQ#{>-nuKH}Sh_Ah2BV6t8_18*6xSjXM*gai*S1aO2rrSVibbjN
zrMG||T16Q*?@Jc=g;_V^X8dIZtBCp>TapTXPokPrf`4J;-+pD#y+o<cJ-mROnG<Xn
zjkR~~4lZB%Gp;O2=(@m?#4bjY$9*-*8pRh~@8C`#`2H!gL3V!LuJtK__|>S8S{N=M
zF_JZptmpSaz`zZ?k#y~*nTusgGr5&>r(RY4+2v&t5YI!<j1`YuJY<0VD1M$9Ujg0+
zbWkQZ%m)B&66ooA{boxHMkNqvkW47^zELGMPmINjL2jL@9)OQPMc79C)4~Uq0w3RZ
z(9_eCq?q#KlUEt?X4c_^TIl&Lh%p3|jFy3NAxXmv<B6a8h>wB*FM<z}pX+XB-Mab^
z`MEI#Qz`C75%Z4OO}GdbFl_oA<&S?p>-Vo>(bqECKL(jUO#<^oh>=(>aNkDjLA_){
z#l|LdCQeLd$tl*+P+!)edxVO&W4JxMbJt_<uGRioHf;J#(XTqFER8ur0##E3UU={s
zM4#2L>Uf7C!MakgX^A!rFOrP<6<*eJltdWrD)3X;KQ78DKX?XqiKex5a)?T6`rM@r
zlr`|UI9q?oNtpf9UoHu{DZzRESDOrmseS$fQ<ux=T@YljEKVna0xdHfeY|q`6mv?d
z*a?0(xO5P9xdtNmB>z0gfu~78u3`b7+nG4Y+97jOpYWU4s0+AZ<mV$5Yo%JVzLnBA
z0OIa2WdL$32u(`_(TFv*N^KqOQxDVWg_wh~v+&T}9%J8E%nPz+FIwRHKiROapXx?S
z(X=YD@xJ-*kGwcP-?q8N33DPp<wkK+^Pj22Rd14(yK);Sr$B4K9RW4R?t+gg-ZJgK
zz(T1w+zi6&-(Fc6Oj@ulUxD`HR6ig@rm8ifd~2m@|C&uKTt2c4f5U_pLH&dpidIZR
z0Hba~AYq#kO$Q$H4O+09Iv5@QpnhwqZ)+R1U_Y2IiIN}2lCVW4*JpqXKt&BDC4md{
z*QgxY*dRDzuH#ft_3Q>`B!lBrr`%X8D)u5qh=DJ@S7FMsq`^DQD1g``Dn({bNBR4E
z%EaKJ^$RpFn=|LM4r+T9mTJ0|2g$0t=VFOaAXyhR3ANStvfVe(wP23Ew7^)K=M&hh
zG3$_=Zo@+=Z9Ps?enj<w3NOfE%3==}g-);)FlM4;SJ$y|2uh%gB{*l^j=}e%I_}lV
zuE(^LZRLu?z`H%qwxug(ySz%mjVk;_-&{RgXyLtO9e#(o{Fw^6r!3w<cvf+C0jY>X
z5upx}39ilgkk(Q=KRJInbdm?128}5Q`d{Mi|663p0!MSiWCLy}IWiptcMZ8IP$Pw(
z#JiD2X@aS))o$-Qeo0edMAF;Ib5;+H3>>5$TST&PgyWY(8A0}AfE<Fq334j1&<;q~
z$=35RxbO8taMcj*XhK&D*RNqT@x?x+7Oq^NQrYke2~&s6h00`t{UjtY!RBd#5!2=1
z4e0agN(QxG7s4p(jH|lmMA}N<^m3-xArc_28!+W)`sencwI`}97fj7(5|mT-p>!RF
z$c^9d+xDTat?{>_m$>Bzj+NyOuG$fq7ZsDiL4vWVSd&Bw@N}nvo?HgKkQ)436^p_r
z4uz|VW=M1BW3|S{#sNcn;o22H35~1nErOsv2rzN8ST%;@hS|TRUCD}JStqcUDv~Ub
zLWE{1y+;GffdC_vR0K7@Sky1+fqD3^8+am+r&<Fsz|g}b<mIwEt}NFx@#d7*bw^H-
zKww(H%@>E*<D(Y*nreh#N<Tc5MaSkIkQkIv2qQ)@%HGsTJoc<O?e(k}r6{Qo)iunU
z^Jdrec*H!lW35$VN1an*M~kdrd2pI>XE2YlC&5R%NqS!lL)@F7wG!q5x7cp!o5_n=
zs1Z4tP{BTwm~*G_f%(H0zhF)51x25lynNK;Sa6YW<0Rprl+hPXoWBKZ|Dm!Uwf$m_
zbZjB7zEoTUvF|M!(|arM3%b4eKyf2G3&hp;|CgVai+BV9m<kyrCS&7)=Rr2H^wqn%
zbuLb(hNYR;@5#IDBPf_oQBF<37iu7`^||BQ3CCIv5B72@T1!MyeXcMJghB`Nftjpw
zNpVo7rQl(P70ryT?rbUsVq#Jm-{j+(doG0X{5d@uh@tp~dOUm1uneA?Gt{+HY~xXS
zP49Is{BQ*jV}b%m=(fV~xJx1?y(2k!pbpvV2u%HCLSOP(NmD*Y@*wr`{i0TNnP=H|
zQ{z{i^zLo%XE3boyWf6Ju)eOF4FJ(UXC>pKi-J8fM0WO}iTbTnhUoObOmdM~QyAV}
zZYlS44z*<HY0hQXWnRf77QS`eU|BSGA~W@FYpeFzTlf@kYCzJQX>_wfIquw`H+l$!
z2&!?Fz#YC@Om9lzJl5zD#Klu2$<_5|6xB65knsGp<NIDCIL*NEWYH28lIhB0Cjj;-
z#o%d%<Zk6Gdr{|h@^;4YcwAjvd#|o3Cm+h#ga+!n>NkXt`q;i{UR{{2yOu}$J&>iX
z_k(KN%&X=RHcmvL|97jeNj~$|(wGu=N2xN29)+%H%im?q8xR@At%mw6##3Sf%knml
zjuAUI+R55q+@mOLi@t{-Qs`F>WoT}~<65heHW$s~Q+=Ly=(|Tu(*ip;zMt?*!j<l4
zPu*(%q7wrUF6dblL86#ZP6wveb$5zs97$t<F#{^-cAKxpxgN&@{|dGL-;W2a)m1!&
zD5^BvXJS&00s4}xxL>z4E48-v9$e<!(``0_yIU_x3wf8uRX-Zq2ey_uWo4(c+jFMp
zA;kqtm(z8)=mH0-+Iv-#%Sk3w7{9L`Wcc*GtIZ=Kl_~1j@~s*7xCJX#4zUyN!s=CF
zVXDh?5~`>}Yx5SS+=Uk+JK*lgg+=P5kA-U#Uj@PeTiYG3A)U^{G0ZO{KoAu0zu6_0
zOX1@&k*Nr#bsD^171l?{wD>$RP1tjRq#^si&GUZv_?+sd##!_G6D)Iv_}r3FDH>-x
zfeqh%K8AW~tYVfGzI2nYnh-8j?Y!RNd?PzJJ7&uUT)lGrBiM}x${opAD8oaBIXkyu
zokEDV8OP^1j?Hfti9Bk}{<M8Nr?a0e1}CC_U{Re_M6nT-E42+b-5(R90Cvy`U1-g-
zy_m6xB@l>wqfJ`InMD%i&{?4n4Rq!<XPz$jDaNTlDg$Q3#+iZ#QUjI&J})SH&}cFg
z8H$okI(lR!7P=z($(V=1>|oR%tT@i1R}z8a?WQ~#pR>KUa8cFSROmQ0VYK7U)zP$$
zE8S>k-kr>#EL6;@TPm3^)qZltQjtx*<XFOs%pAKA{Bp{*y1n(o!7OC@hOB#9ITY3;
z*ZY-hMJR3V@@uwE<na`=8g8g|q#9Ad6+8un=r=GKu04%NSm1w$SpV1ig!hynL9Hpc
z7X#*;V3TC{5%OV#2d71)$K6uiV)q5iJ$lU19z+q1bRK_L{#tOu<9kEli@66aNOo5q
zQ!vUZuT7Xqz`35R2-l?6R;=)QjHj(-;94RsiH?EyE#0+DfuNP<B(s<;3~Y>3@il%f
z31k%BoT%@X%l(sUQ%&b*$*;!j_E|9RN8RzTpY2Q>0;T8tWmc4IAbhtE7hoGJL`%?v
zaj^Y!&^3OA63lje23$ob`6PB;wnt%5+3i1WJ@O&JV)Q)GM1b{kETQZQ8S-Dj0nj(}
zQe-rdf$R$96(>V|01uK598Ixihl7%%?T>z@#4;>8DUD-i-ZNr?4W#aE)N{o!?${@w
z&<j&|l7H9YVwkTu(`ye0jE_5}S_+nBrV1vsy1Ega9GD666=R+qY#zWz@TtehiW`?m
z9@PE#h8%Fy=6TPnByf*J{PIU??DR%AQ~jrTM2<BGeP=0lt`8Sa(t=b(HrMNjrlDb2
z2kd+wMr4+{PZxqS45fh{3h%MYlA(AtfC{E^pmX5m()OYdNT{QY*sQt67D@SZR7caT
z<E=b$9FAdqpO=)Y{e6CeDUoYpvyi8Ti`uLUDo|owIW#?`S661mC1CFGj!0P|r;r)_
zjFL0x<r}yC-cKp$2(Rztm9|jHprb5WEjQdgj1ic6HCj^WivPsQJ@sGY`G2%Q8^y>X
zg%tJ9-0K9oh76**L;&8sq!rd9fw<P~CE%TCcsvBj+?vDLDts3OIJ&`N7$;GYbr^Pi
zaRI&nC*iJhuo`)d2PTMAJJ+Q8a4~^#p)Xj5ZuhyLHYa8M4~R&{qZdu6FijkeNxY!E
zBCz7;d_6f6sdcjhWR+U<0?qJHIE8rgb8!&bpGfwZSq&~Js@%9c{&AT|VCGU}UxEGh
zxMq(fG0=UloT}aXiB%bsB*l<dq5Ra&Mp|bjvv4#+kxIS{<7*O)3T<%^|4e2ynfY9Y
ztt-!lW=m_92=VwDY4OPTIDylf;P&3JqmzBjli+E3F~;~a^rQWwv3#71!lXoFpj8dq
z4BOWvhonHk3|5}5*1}Qn6>^4DS5gf?PX-47a1Uw{Q$$T$s{PC$Px}Q|xGj{5H~<JG
zp+UySaOKkkHt&w^sKDO}BUfXUDg{eq3o5d0E5n*X>y}7w(1WjLB{3veL{N>c>^n?<
zQflUqX91FAq}Y^K>hH1LLh;ICoqtsy8nPGuqCNU6n@;!UC;bvl;}WoaF{`Iluw=X2
zk1hny-7qU=PU)eHP92d9QHCS7<!c+e?ek;wB@geawCUg-gRo8^=36vbDk_~!IrMf%
zna^Jl7>Y&SzW@gGfB4l$KLH4FsZBdjl3_ct#doW?$^0$V<<$qVH3zGKWYHMoWFw)W
z)8A1%EaZ0RP_oG<R2r@!dg_~=1@<oZ^)Z$4rtHD$(!_S#xV6Pa^|9;5kze{BL_IG|
zuM9f3x7?~Mn;pYTO1|m$QbJ=Y31p}El~BuC$+h8ys95YYLtx-(x8E>rNmF3*#}6L7
zrsJ7MG*Jvmy@n2oZ$#B?3Yahi);1Wsv#X^+`=IB8Y&$3$hJj<K9uM@{m0lj^<-}6e
zuM22iUY$)kgxNM5zZTd_D&#82%4iXrssRC?2xK-y;f7GWKOP%hO;h%%Z1=a{UuXPu
zS6-{Z<B6-cuBTnDy88ApzvpHH@(7Gl%V&;{Vcw_~4)qMK740<n1^C#n`wrarb%53h
zQ&uvCG6iq2#)EN$dkU}YNkOj?I@=y>Ii^jYNT0(B;FhAZ#h{M+j+ppM%FZaZZQDE^
z8RU4xop|@V<BHz=D3#migrcP)^L@W^KxE2z0(1Z644NUawI_+I`?%AxmPqQC|J{y1
z%Ezk^Iq2+i{&)~9hN$(9o<ou{IXXJBm|;StWVpG8P%j^ogoT54nD0k~8ny_G?T-@s
zI)LV)4(>iju(8155=v(>MC;4+u5zis*B@lYko>SVmUxJ9Yi}Z|i9)zwZBL&K``^>R
zZ{Fp9IQC!@o`^aq6&1W+)lvmm+J|jj|EIF^3~H+T*0^F&dar^4f=E%SVCYB>2uKk_
z=pcmNi*!U06bL0?Fp$ujG!c-OE}{1#g49SCLXloV<evA={onifp4lJv?DOT!oLRf9
z{jA?(B>Cx@8Cm;SYm#8@C_(@s;Ruv0gN)*e$#hNf>$HMLGYL8VH#S7$>_2@fx~~@k
z)ZB3xPss#4vm-?n=5&FN->u8Xfz`m-3w;XU?^ccnUlke-xh_*pu>AYi?l7kq6!n6y
zVEh6kR_l4DZO{JJC?549-MwgnPuqypr!5)U%16QoEcjnG`1^I@AyG-5q)b4G<%dz?
zdR)W(XakVmlSXE&=4dG_&Gan(F#-MUpoCCFqxZp4zX?M5P*fxonyN$b=UP$?+z=_i
zh!V3|8vG~?daSy~L;xaXlLZ*DkBGq50Ja)_aOM;ldf|062bFN<zBtScIF0C=0ItU{
zH<1cX#Px)G>R;}AkTuzCGXJ(v9WCEX+cgP|o^u7hsU?DE<Ny-?z6QR#rk8yIrf)q;
zbhi%2CWZZaU0yj<J)_X)rLwOWDYD^f(F9RoR~vJ!_tAYi!HTl_w!w1_xBwD)`@-0Z
zStv9Wr*e^}{?FMrPMJ1>Paj9^0(uO{vk!4Pzv`LgvCd4i&4NP8xAgZE=?evBOieZ4
zmAZzpramJ_zsV=1rKi|d9XB)^jd_N$z>IIku%cDvV7Tz_gEg7JVWEn~Nj~xn5S{A%
zr>HA)LO6NYn7_2#KGmXJxKCPh1XTXjRsKPj^oXbLNq96ECU|(wvH~)|<e&JNje%tY
zWHEJ!dhT3x-vq@vX&!AhH=zznbC!p?aOa8w%5o{g$jxiDW|JH=v02R|x{wOF#^T{G
zf96&CkMN|sp7~m?uUS%aSWTP<VzHL%_7e^pgP3xfKn|NTkP5N>ub?P=gyBJz?{SlL
z&pOYUlqUAOXFU{Vy<#sasLJT77a`G!+Rywz6C!6@dv%e5EU8f}nJUeAB^V+5mY3#F
zSW;dk-8G#y-@6DwlBMuJ-h090E?P#9%mbP`+aus`)Sm6Pc3U3dE;}^wxh?%!8j~WP
zpF+KM(zyZ<8*e-euwnpNGjo*LoZB7_@m}E*s0;9ye?Vrq<N$@47tHGlz0WeXFvxC-
zHzX}YuBx}4`njt_zWPVRTrT{^-~*q%GY{p975T_xnH~z5Boo9u9ct*);(x!DsJ)S;
z_<C3LSk}vXbsxxb5({PtD~2k#)H2oDKNnY6Pbms1nnAg4PbzZ#Pwg~h=j{HeNjwZ+
zGK*!3b-T%ED59nUR8<m}!?AL0eVg{ZuEr)w>igyPk!dMreCf|}T*=@r!jNrkZ@>6o
z_}Kr`A^Znu8I#LjMPA5kN_U;}H_qOpQ<oR&p*K$TNjI5Q-B?;%C5{rbk}vt}BcN3b
zK{Lis>cI^d?+?jLTZ}K>xY~c<ZD4!L*+Gs2uIRgQll|mv8n|C{$~^S9{+!Ql*>qEN
zABkQ`wXEh_io(wF8U0`l)P0QB0<;Z6nhQ}?awo{0|8WcMIk{vI8)`U>Db85<X-Ddp
zZ|_^4J1DC6!&?Q}je(5ff2`h>C$V)cyvZ9+LwJ=z_juUoMhyU*a8nRjxwJD}*ndbs
ze_Dw{3j477#xcQ6-B_pK>6NhG9X#C!>Oid((rm&B?IgZ2Qwa#S00+&nfWDR!d@M`i
zVgU&-IB~lLv*<>$Y|v6zJP2tN^kpN^7(6_2wDil~VGVtOn*cw2YI>LaHl~oSOmCn>
zD96BWr;Q*M4PK<qKBqZKCmbqfzwk&2y68<(XYwJPv!PUo!4eD_xAF>JOUk&Ud|9sY
zU2h;#s&(xgv);bk&o^aL-dEUlJQ&Cpm|M7`;T=*v^17Y}9#M+V*rTFoxv~s-^323s
z-OYm)Pe0w_{gE_(?IJMx-&_da$L!+3WHU~4$pR5_7jjV#$TN!aBhDnZZZ_Q;JEI*!
z3EbM&mlDk<=X$Ok0M3LHz?tmZ0bsoU!)+550uYPKBm5PbCO#77l+2Iyz^R{KC|8@t
zw4<(S%G$ZXF$>#P*Yo+XOaj(A8z2&L;P3-m3W=`p2Zd-#N>Uy<E>HO9Etar91E|x}
zLfYdZR{5u=l}^*Nayy!HK0Q|MR!zM>L{SX~Smnnoj2Ljwq{H)#7|DPiTl~9*y2tnA
zirJDR>b*g7UXV$H6#UP1jW#_;SM2HGy_Y}AcsfzH)xeDvZ2bp1HPK%xeJBnGn9=9@
zi&1?7tLJsJVm7T>=4L|Eqb4j}@h&TP7PVA?2ZG;)VM*pN*y1j!rELUHB1WFvwVG;g
zkD8yi3Nr-owW?vuG~8~PWa303F+qzT$RtukRih<n{-kN6i>h)~=y}j!?$kRYkN{P^
zp)rqo2Hc`(5|Jk?sz-%H$)>MAN{~Xc)FKv!m#UG_tMSxjmY9_-LtQppa%2Bf-D<f9
z?{5Et<fW>f^gil$%;WvFviIFGY>l$gi*|atw<IPh$Pw8EKM5Y#c56P!oi&HW{dxBk
z3li^|yj}s?n~~~C%^2Ge`#zR2O8;3`B7?@|=-hp~+~ftbV#-t2w=tc@8h6cCjo6}J
zmv<X6n(I9vr#@L28hfM&+9mx!O&|bb-oLc9LI96~(*7Tqv8!>=xfQ;nW;wdAkTB8s
zE&b|@IOHNL#!R0@)k>7*b_&lDb(gIdJH~M^Xa<@KwN`jbsq}<<DN7To&7h$}cs;<m
zi56F^86l4Fc&|0JWj~s;JKZl<>HpgBhQ;ISu^!!Vk7bK*y54&BbsL}kL#vEWxVjLU
z@taeAU=VcZ$Gq@&n8fsaVTM78WL7hQ&XdPx5!%+}r{J(-ZDD81_SDU?p#7%Sa$25L
zGsCM3-9V_bJ;#@F-hua-f6E!7!S5xyM+t$PhD8ZI4rb5IZ1w~6?=gqr^JhIA68I2#
z?0Wp+{9s@;F0I-h?)Q`uI;M}Wk}Huy;9RtjEalyDwCREmsCR5Ex~Xpk&DG!%?cTQ#
z=(ILC@Q|XW(1N;OGYe@kXP|YptU#J#8M`uVEW40H@XDArLtSC*?bjqq3*y}|95;5r
z+$iYJDA$56M&~bJ;@NXDCpY4@W0|vi_VP~qb(!>1Nu2$1vUda)K+)CK$HqD;qJb?A
zHd!Mg?jAZ-$tcI%*~MEqxI$>)%Hym3#R2=W!>#K|E0ELFS)ZbV=XAiY^Zz$QfB)|t
z@tOf<5wRN~YO`<|u(iPODya@_R#q@iu1G&pC0wWB7T1#O9va(l7Mu|CvNS$k{eoAO
zOl&IgV=lZp1XnmD*+%*q2N!(b)JD+nG{RLytUM3i8n5A~_+9qC&;aLP6Omv7_gp`T
zI0=#q+y_N&YSZM8UA@tyaB(@?RF4PuOCy%?Cz$2o+szA!m`=nVUwbD6bhS>HJjGZ*
z><Vj12m`9(T;-6ibla*FCQ`S#k!M`>zLVTkC(b03X-swe4FkH<pk%p2o~vrA$$Hl(
z5Wa@7{>Sn#FS+*}i|np;h<H#f;P+ex4_pYwKLM#h(x@nKQH4i>o;34a!4W4*RYInE
z1s<{QdUL)&S6V}h`oPge5*aNId*);5Mz<y)KgB$*-X`Ozu47%$pG#eLV*1yY_(%(L
zEtc+cEtHo+P4<!7ta4z%oxt6Nq0>$d;FVL&J6j^P&4QE~ul=hVP%@?udNuPo1Y($P
zFrE6%7w&zBKPlG5tls(~x~8eQBmdraXux#FAAlwAr)P4bZ+0eAnZSA}L{?>FAC=Lp
z#r|`;7hEaE{&GgoXQ#ujHmuz~7ztys&43HvRK^GI&(h#eBqEg$7}q~gD+YLd=$m&b
zaBvVtmay~EerBS)lOL<_P_?)bGX7;+sAAF4xE6Ol35t-u@?9XEZf`l+Y~bxrHO4#F
z6^nR?HycV_blu8kTi-Me2Ve;DyO;7)A5Y+Gk7}cHQ^~yS(-y^VbslmX-|0-ycOU4|
zGcOUs)!D0#ZM=D7$9^9o^R(Cfri;5$6+MMZfWql3!-{8);Xa9|1yY7h^8qE7I#Dj{
z(5TW9vkq?ye-$}wox>h2Dh2kX7vYYn?5Tc)x9veF8Y&;r2t96r<$CWao|#5NJ9SvL
z15rR4Qu-w6ix<a*0cHmF?nioA*k?;VIsv2{`**g%txP9(SF*PGLhI60F7CORQA^!H
zi-^lsrP8!`53WJ0#sp(mw|VZip|0tJS>><0K{`cDLWLQbO$i=c0>~Zx)5&^7-{ke)
z$K}22z7wLX7lEkiQc9;|eTSLozAQa%9c5_c=SJg#uV%R2iH6X7ihh?u^JejZP4OCx
z@fnu#j~AEPN}h`{P-9<in(v=+_bh~63%M=A-IWsRt%<g#D0Uh-z<F8Q?rpNt@r0*s
zB1jh(n1SZQot|>>3M02?d)Knf>eatZzTX;8?sZ4I=6p+iV}~-h2V$vY&CjC{_pmb;
zb8=5*;`^{Jaj~lYk~v>U*n7V7YgKmT!7Q@!;Ifbq7V5u7`glf5GI{QmRw`4rvXg0N
zBVKTkBpt`0%f%$hyxdPQ@cDB~l&9XW48cU~Kq+<E)(QWo?c{E!c)pRPzv?C#cD-F)
zWL1Mx(BEq>FPiYn)5UG=h7pJAe*@R7!V0gTYJ02s8&I_^uWwyz^0sf^Y3}h;c^Tbu
zqX?SOq~REGwE<>(_)Cn1QG`yCQPJN!V=$^?&NW@j=e7TMyT8&Wj37?Rh$-s@nebi6
ztX&PP2WsUg;W(bkq>amAG3}@bCYP-L%Q%$iZ57+MXQjgjwH7mmwP=Z8>?JZ6JKvpr
z4Zm=gO<abinN=B8RrpO=<(EJuLmQ%YgS>E@bKFa9>P*tkkT-8Wq`p92@7@%#n}zeD
z__oNwrGR;%ghyjvT-YT4)<b8@293Z@`^%SzkjZZvIZF($+Og%a%};Zr+Tzw}_j{*!
z9A3Q4Lj66x9#J}k<?A|kqAY&y)FNZZzmPNMa&Z2xZ7GN!d%0&Ay!r!T9dTxu)!>Wv
z+8L6PmejIEhxz}dS`AbU;e`xKf6Ns3d*K?F#yRdG>@KbbO_iW4@u!)T6}WSK7Qq?*
zM*{GPH*qzS+~j=XKumJ<2O@4N0GNj>rFTaLtIA4={!5<KpWkc-aQ3eJJo;uglP<E|
zKrY)TnCSMt8lz<VMLSEP^pQjVknBBnSQacE{LFeqRk>r(sv}|d5NdAzOD(s5wkv8?
zXy~v#6xjJ1MUZ0IgNcXu4WXD<9Z`cw!vv(yaWWUTy*Dr9N#`DbPq9vny{)8HY`Qi<
zsM`CH`|xqS`bh2X<C#-6+hSyzg_7AFc79Yh6QeR8Mrf(xNZzFDX1+d8{g3V)G-&&}
z18^VE=d7$<RsD)xtCk#8t-;7Vr=@aYf;8D`gY(?OM<x}(pH`uKm}OI_%%L+OCTr67
ztS@=8BLt24?0)83iz5$dWJMfrhy7-}`1T?&aHj}65ST$BK&B_8Sjte;hxJD=a&_k=
z=X}qZm-rpS7g5upTitDuPP=59D#quD>(UW0e@C|eKvN!5zz_1NO5}f{!66}?w)WG;
zdR2#`+SGEXvl_{*^4IGa5kWpH8Ar=iz392!I-27CNh+~qz@j)Qg-8+)-m$>1Z^gpp
z!WY9Qhtpp+Tdw|sBKG#IItyiM2Hcv>N|!~MRa(1K`UH*;ILKpuDrb>)0HHD_%jwT6
zBi&bdgN_+3I|NJ5Yg7|aAafV@14OaFjCvA@Aq)~99-PXvP{kN}SstB|A=Glcl_Y^t
z=ih_7&AHNT2XBWjhPkp)4ROmOUPtN1Eb*G-8;9`LTlw?WOW>k0X9+2(=(I;)uhtG<
z#6#c_>ti4B!TxJOSi;anBU)bn`$M2@9|t--Xf1L<11->ETp#n_T^P^k|GRfeFJQj}
UVch3rzXCiEO#_WeHQVU_0;CWRI{*Lx

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_025.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_025.png
new file mode 100644
index 0000000000000000000000000000000000000000..34196fc83eb5add9fc1c6dfdf5fafe6698475aaa
GIT binary patch
literal 55869
zcmagFWmH^E_B|X35Zv7xcXti$PVnFw2<{rRk;Xkh;}C*{00Dx#Hx}F*X<UQbpZU$q
zGw-+e!<AaMs#dSAdv5Ks_dYS2>IxXBq^NJ+yunaXl+}Lo2HxxSJ`@@0^<3tt>i&8W
z_LMX5)N!%(^s#Wad820G>FVU->EvKZ?QP@k;o#!T%OS|Y%SLVQ>FMer!pZshpAH-@
z?slAz_cD90lc2aN8hX5WgJtor1MZt-xx<?`i{F)GrF4CBkDlz3tPQ=_QJzoQHGW@D
zCMPEwr43EbepvlbO~(HI4FZa$1=D*=AE94X4IgBty+6-6Nmd=#u+Q59OQxa*NXV9w
z&~OpD=urx$5L&0c|Nf4s99N~BN5U+v(Ci%ojeR;3qp7D>q^tajQ1n#dCm*5E{19{#
zlJL-?@R;Zd5oX{?j#0Iy!OI?!Ww;nL+73BE{BQGLPqxEH`zmr?pZ%ws@Eabtga0;O
z$6$U@&Heq)1U$)asqxC4|2^CP4SzHA|1;b-Bunz=+|S>?l^c!Zkk5NfXJjSl+#pkj
zS3;h9*xGCPFB094iRG$XFH@-~yORDq-H!?y&p)zW3UhzY&VI<q%#>{CSKF4XUis}6
z*9*j2pSNPjdO6Sdr|)ZPZ#V?`QZh|PzgiFO_T9_V`WDsWsu}ct<U^9?#jgDy+Ta7b
z{N0}ZxWJ=gs-2<pVx|gL^$l!l1k1p;N!n}eD4~5cbwa7djordPY)ocEFx5r?K-mxZ
znvyd~8=Bi|1k}czxY1Gulu}Rdi%hsdRxf=B-vIEH?-kc&?osLKrIFL#9+D3bgSa0O
z38c-dJ7DJ<uE+@e)R@2|TzYzHY-z&7WK?<rZRLhJYlm3w-@eX{T_RPy)($-*vzdMm
zjUh2gO~mmC#_Xe9`MZNJ+h!K$Lv9gFrlTbBWVO9U;f-HS?C6w~zmoj*JsvOu@e4VU
z(4&N$Qax=2J)S5qWNC2an?Dza-@V+fKT#~TE`JlVfi+$p1-@Xg;9^I~j4~&WokeQ+
z%;VgK&6E=ngYaut7B-L09lU{2!9HE8!?3A$iCg=*ae}93JBJiCb{G<u<<Drr7mUHV
zotH0<ofkca>lW7R*jfY_Wg~+X_q?j}uxOPqE3CBJ?9N9&bum{ES!RtZY*o<jaQ%@*
z@^V8=0<&1hf0eTO`O7EvjcNI~+G-g37B+$Y!pC*!Rwz_MECBW4$nGprJPI}NcQr@n
zIo1n@BYemClaIZ!VpVY|1>tS6fDtqLTS%@52-o%ST4XRWzd)OL=J+r7!1?Z-52Lq4
z-Q_xM2vfN;R#$(&VFHCQbT`kD*pVrnVpO3rS4ey!8PDrNSF3jJERf2vTlmM;;3?eC
zoAxwZI7PM|k}3$-0LZLX6MjAMtlgCg`%Ni681@7{E5?RD<kJ=r!+Kv-0wfDJW=c}n
zP`I`Dauy3qlcshmWHLIm>^*HUto$*TFn^4F@ibpwP7MjgN7ab*1uGEa$2x!pwn-gt
z@W+b066U^6X-<CrIH13bjM(ur4PK$Lsdq=k#REF_))#C>oLs+97bapCcSWj3XosGf
z#x^=>R2&bO7oIb^X6Skvj96g&XKf!>a7d?N(}JDd$NCUFSS+<S8y`QKjZ%s)@*y!D
zT)aaZC!0<%1V;7IdUKMk?iBDujKktyovh9k^sPG;%b~;HEs`kY!R%~8|Mcwd7FDqk
zs^8{p<^n0ketkUIr-pf&`6gkb${#ackEtP#m?8HQPxN5l1Ba3!-_vv}&nOCYJ{Ds7
zp+`6ulNH-whp!E$RtJB`MOg*U0+@G_88j2V!?=HBXp<*fe49`$#N?dn^v{X)K&IU%
z2I3UGxlYU~aTpk1&gs)qwEC1$TH1Sbax{I6aXU@)g5UxU@v(&AiN1)tn+@MKtu+tt
z*DF-$epj^dJBpckywwmNd|V-qmyfAFUM9}lVvW?M)V@>P2b8{N)>K}rrw&cRjp+K_
zGsCXCOUY@%oW_`}{l%5~I!zG}AscOZhZ*iAykukV=%$W;W2BaW<<L0S>~`n1=6kp9
zcDZ!3*T})`v<{%~Ml;Sp1%%gGinz+HhPw6Nso}p%g9o+qMtKnD6<*?7xmV3uor=F>
zlh*#SLq3hiWQHRPCxyKcDKU~rjBa#vWFl+crR0i<g&eVSQQEZcW#rqDsEb>HKXREo
z@^(Ld{%G?KU_lIhpaho>AQfXP_5ml~@pcD>-633tC$=;sY+fxpb+}WCtL}?pvGW;%
zxsG-eOBLNw6rl^fm(oM*I^e#B00XOJ>5WY-Oe-46B2%#C?jBpnrcUC;df+p<&N7kT
zR-}u&(>`6uL+A6d*yV<>v_?`hWkq}Z-;?)3+q`D?xT2!$&&T?~7$wC5{Noogu!a3Z
zY<2VGG-ZHA%UErDFpxC|JN*;i?yjMUDYRIj45wq~-Z>0He)_RxsM?WLwlrz!FMew^
zVMh{rwcMpbYnjue#t={MTe}IGQfn?^Tm^bc|Fcc3#}zzL$t>?@VC<jc;g{1)Q-)z5
zB?W*yt7*s%(qWU&U3Pu|_^w|>JAy78Fm6NvKo}5dXjDYh){byJUf?C9&Se##NP<5S
zr+x_ja{8cX7{Q8o1Pr=)&rCi~<8;7<Tz*zkJilxcbc|GN0aSNHi`ng~>m`>E1*U~r
zJdhe2w>mYw8ylnA-Hu`WHn&@c2_g`3bnAis=+{foz>#i-(orMKkv2f$)Y4G<z?Ouu
zN4dRqMAHPm-xoePQb6$Mr$4U)G(>#os5(=mgw9IM%j<jN5B0M|+d!W1c1O2&Omo!q
z{U<1oUC;;_KT7SL_wu_u*i?_>!T(xLl-+k!mvdLFCieqq%@Ay6gamSCjPOteI(JgA
zJPa^>c_uclSgtxg!A*>r9SfgQjkyWZa7GU<jR1+e7o+4R{{1f2yj?G3wv&z~`q(OH
zRwD4U_4vAe+es*keDl~e4EuIs*V|j<B7tF?f#-1>tPz@j4s}$=A<3++(=?@hO%e*W
ztaO6j$ccH|wNIxnP4?(!FS8+$kKOE;RHA*Q#+71eBC7ot8;kEWwPfjQxJPi^-}ScV
zJ<}4GF5>}<wJS6$_57~V_4lvt$0n$XOkDyuVBm$j?uN?;_k$|D654m$LQx?3w|XuQ
zD{Xb7@s7rAkKvbNe+Y%#Z?Tm{K#ffGuwfPK$hXFCT(G4Jm`>{PDci8&OCRgtSHw|{
zf;c037Xl?2vEUEc;1!lm-S&G9S6YHd9OY$0t=xhWTK0uM@e+*e;0ooxM?`OZyX2iI
zs=vN@9{|gi^vty?rz$ixlU0mN+8X~AP+;6=uzG?)p~8GFR=c3=qp!72^|gbsj};w^
z+O_Su#H#3Cl7hT$sUzp+ohDS>;(mpzA^E}_Nb-f}Appr7MAFG&fU|)XnM4Ydc8udj
zD*f}+CHn8Mq|zFz4L+_!*K-a<G?R@s{{l0PdLl-W&}zBW<Q7U1lM|M)vdxexkN<_a
z&zIxltmE(}I!!tX&d6w*P3m4)IjtfBFd@0?w(FX=9(>H&=K4API5_rNJg{lBXPQrE
z=0gIdq;d(mlrT*!O;t*?6i|Lq!57|wO);^Umw=_CSVUAsmzg$iYak$Vq2<ST>8JcE
zy`C!Pvq>o6Ogw^#hUUsB+p|?cu&JQGgZ84={Wt=<&rP<)ghkuQAxc{UBAi0EsDbw!
z49&Glqhl)F^KP~OTrvBp%B3o3!%OYx(R}GHNZ3nSS&UOY_D$0EFAc@KCR4M*d5XFg
z@a5VatJuR>vD}cJNE)^QD{AeUa4s<s1GI~joN)8#If{iod3x~jF+l8--%VIGGwQoR
zRBA0D*dEsJ$GO@U+`~#Dp&<9{ULfJn1<2cxs7fvZPKwydbHMzMTx5c_uq7X9vh8P%
zjvETSn$G@BE0~b?J$e15LMXCqg2$GaDE=S{Kocze{fwj8J8;FEQ`J$ch^d(keJCD{
zg#t+-17p8IFq(DntYqTx4}P9^PMk}-VJ`KTtE&$``Dkl0>IRO9Jl0s1nL*eKb%xO<
z;n<3$NmM;KPydQeyfG|^-BlmX4YXlJ7BM_c7wZV^nG+NwB}KkJV}J15P33jJ*KsvP
za7|VT>>Zt??BSR5{&KFGG+ws8{(};>$k0-kB?jy&5W|{4*(Ku+Il;5Lt9!TWw{@>p
z<J_oprUWCi&vV^}1|Z<7=LcXj6AUu}3|od`l%r*!?*Q45?3yT5eXC!;<`-5C2QlQ)
z^=r3EtV(C-?FDX!VG`Yy>mkWob`wRp1-!)O4P_>%D7<phWSao42e(u#GX%}*3<HtC
zX!+vdl_T;FqNx+lX0Jm+ALsS5*-r!xqi$;R4}FGF;c&Cch!a@ZUq4;=JVVy+HV(fo
z+=c#M{5DK7t@q+dASZ%w26jw3f5+AL)6|-*7R;P-&1A_N)OFgjcJXkadQ#`g3&rCF
zthM!Aw$q$)%r1;_;uj2URrd6HQxb=J^{Y9m@sOVFlC5u%t=|gF#Ao#{QXX;?Pi0v+
zdq*|v_Fzdj)KOC>Sfxl&ZKCgJf0@FWwy~dqE|q2yV+#7-PHAw7`Va!db4Oj(EkEM$
zd<g4#Xxei9X4ScY(`67YKq|>}aX;<HL?3qmkG5nBk{6$r%D0nz^7vD}`=i3v;}TV$
zqb)9tHtlTh#{U%_w_9|d%)X62-|qpWzekSrpj5YiUnCx<Nup<b1Wd}vHkbHURCf=G
zjnmD1;<P~;$4z@XJf{v>VxrzNPJq1U^h_%fvDpcqvW9>BL?VfaJx61UP%M>LEh^}+
z&76KG?EZOJfn=%a7MzG6My}I{JCQFAu^L2LN#oCWIpWkb6q2o6-Ca26{WGGrR4w3`
zxn{X`q#bJsf($F57|t<jQVy1VYP-9p?>NutOjup*^Y*DA2XFn!o{6ZxwC)9*w%t%l
z#sQUh<^-Ge8sF)P=MORCa?@mlCERT`d%~YZ^rjX>ygwq?KvUYj(^avtCB*<i_OO8V
zA3Z})V!VER)4)4xnFvmh3K=@;A@6(|uMOE~{3ptV3>&ajr~vp&Wd`M190;*V+xJG9
zSB)Rb4%CQYkW#ju+-U<4uQ~f1DS{jks&f0f89@O1KtE>dIm256L1)2VQPz`%vrD!?
zSD5;oW(AGQj1+Xa!rT+nQ1o>7Y@_NZ?bxo=zas2N;c$X3E0)I)7Vx`Sqin8=m?EUg
zN-|kD#isU=8?#3qMvdyrGrauw%vg4aJ=xI4epAz6@?SpACU}~;4Iq9+#BD<6I6R0c
z@w^e&QgR8cC-{fARpe^501$?iTuK{8{XQB8Y&Ucf!KQ*zgxhNpj=$#y!;&u4ts{aM
zfV>R2PY$#`k<UG0$A0-YdJf*;rXscX$EAYP)`_PdcaNE!PEyZ+CKd>CiFtY_XD{(d
z-@c~Kx*S?DBTNM5s#9*n6Oe*QCX5o4V0@X3RCIYEPDW%IRdDeo`X+h>ax_(73K0AP
z|6O;pnHSlL_{4kPv~%o4@R?vxT50bAd7oSH!8#nLu|gzYv`O}*6fJs`XyNFH3sw4g
z;$?%%{crN1@6$5pMSEt4+=fb#P+8-QMSw?KAQw%2(kCDz0%VcOC=K(yA8Ic3C|?Ja
z1qQKJ$#UQX?I_QMN#f^kT^{+~;n2%<Vj-^!{MOdiKjmNTt0ixH7`!1;ui@=+kaXLT
zs12v>+l6e!6KJ?3lW4K-etf`ADHcFZ9|tdc*ah-4tm3u`CTz-|Alx#!F|Bm3YqETi
zF7b>6*5VUR-bLrd8&{chenAyec8gISwsY<nOD;tuBqX$N3W$U=RB1G#kCqOkcM~bs
zs?}uAx3==YTSrq&jp#0rX<e1S7SF)soLu_|OFs?CTkKUZG!pxm`@X9Gx0~V|A>4F;
zabY#P`+9=3A!wXiZ%cP-kv8JJQa)ESF(yzb)BW;*RcL?4FyF4SH8O=#`aQ9TYR!3I
z{=i?h>Q#Xb*yfwheA38)rAgBG78~b|y$>zn63wcw>F+CzP+%qDnKFh-14x?2^^4Np
z@OaehYV1ymPu^8uc7qNjP2z1c^kcm(a!1)`PKwoxjDT>HV%ncxCpjo%`^z>mKQi@1
z0;=`U=n<6twnZ2xv;~Z5e_D|guX5OIT6m3^Z!xfP28VBwak{N+519}>e)OA)w^tzX
z<yL^b4Ti?Zm^iF%bPr07-1l+=6LQN;vbQQ=lEa;kuovDPruQ)6V{f2ot4x*HBj4NV
zK{bG+7>+hJ5?txOf~R}z?`wX8w}~F!{l&dSwL-V&)E6{e!qi4J?@lm9GsHW-Gpg@*
zP{%0b21Qs8*|GV__kgV@Sa%t254+j<!#>uJ-bxAnt5xLtjR@IwYrx#$ho0+x4@>@?
zQ@#)jL97FX5SkuQNdN>+4&vWwKZV?d-?3+UYluJL81Y^){0C`{T~)2=PWxqTN~w%;
zI_fVI#-tP=grh#tH?$?vVWgB6Bu^r>ZS_Ue=foUTxO@y%lB>{#PC~`pOv;uIr}s^y
z6;<|>CX-4?@}4m{+s_D|$0nXIN|3RQ0Ash>vQwLHElnf~1`*>8wG=iGh~M>Wn#2x#
zIXhP&(0voWiEV{JLuP*7?BN(Lr5CXnnI;=b-I!wZIhF~1NG;)lz!!12sJHNQa_U62
zyl6N9CL?_h+8I1>MB>)L`M-GE<B;zy87?#MLK_=#0QeP<IBwVG%e!Sc=pMK{S%tx@
zp|Rx$*RBT?iak>X_+@M`%Q16o9!#z-nlGGSU+XBxle$$a7n>H(o=j4BqB_^G!RB^E
z|LVO6vvLOIjXd%+b`_`nGDr(6RgQ?Z)ZI?CL77v&ptDOW$x^lQ7{BPDx%z~{>r`*R
zxoH;^RV}bO$KfztijQP1&r5uzG!6RH5=QdtEXf`=V2#tSaMP(F#$1s$BI_2NHhGG*
zhl%@KKp013PtTmms>|wSTW04Vkn`uztAv$hNaUD93Lh^5XN^ftRJ?3$A%rlh7yu#Z
zc+4>?6rSDQnK#?Isc36;$^I`qmw4XS3A|iiW8yGEKxP*9@+J}gYuGwt@kzjXQgEfM
zrQf-7v@jCaxpuJ6sA|ffzKIL&CpCPUwx#M63bd*(R@^xUioY|7%=$xMXD&X@B=E+@
ze^m`6;JE}CuZIZ*L5YE|@;xjOCzlg0CJe1<+6jWo%X(RhFEGE?pB(k{`tPRh2v?k#
zSJSej@2wh@Is&U|Tz)pRfdrUO^FqQKmj#9VAEk1!i2*l;-zhTMf9sXXCgWwk!E`+F
z{$)_{3dM1E5ar%wCK<45nns`Yqt^?Bky~k}>EoBo6(i~JUiMQ8setYJ+-w5BwvJ|f
z#!=h>kel-%$6UM(<N8hI`ybMTHnydYn=c~;TlS}`d=f59iBectu7dpHqLE|0ZJ|il
z(Y+=cjuOn+5)=@rdC!-lCW{Y%S8LIxVPa4H`!0)5MwA#`Ue;|)w_XYV)>GsoN3~bW
z4-rd05xd2SilvXAW>K#$&_yOlFxa#SaOfhak!g4Ysb}B5f)t_e)ZBc%xJ#GAYOy2@
z6;vTv&b2ba96GV^fdDC@N2^k8MSUUc0wzz5mXJ;z#>kB+Al0T${v98d<Ri60BFCj;
z=d;VL!pn>#!|RTra1U0%@7HDd@O_qa_EY4i(lFVSvc)Wr-iaWtB5NF$hmYU^!9Bbk
zKUoeLb$>@kkmf(HLe0k(HIcuXG}5uy<Y|>ru#PjPoWswr^O)T=3NEZ2xgG8@T-JL$
zY^n|vT5klQw)!liv7c~Dp2j4>GxCin&AFKEER_m!`{hf8<gh7a=ScS0Pcl#t|FrI#
zn?nkhN^H3kPP;&sp~_3h%Fd@n4ucj2<x_)(t1X&lY^{RQS63;#jou2(;q?qzIkd=v
zeVtBfE=O)mH*V|dg-VmQemH8$R&*1P3)ohn2Qa1w)Yk10^Wax4!6B)XYAaMk+kXCP
z6qMi{5;s+#)-)rhd<LJ$izPAAGtHAhzH<kXrybcLy#b@=d6?4$xM&2RugKLMywe!{
z&xWvJ#Suwe9~G+OtG<asY(vXaEzBc0Xj$DuTd!8*Lce0LD1CFE8}OGdzJe6^$}Tpw
zj8~@<lq9jSQ<NRt`6g?auvk16DB6+8)L?{`pMgRx^1sd7{WjU%{Z%rfCLC-ZVKRbZ
zE!${Pw(hd|u4qr*GgcF6+aWeBLER^lZ#<?LTSFuekX(xE_dQvenx0(r>dpA4h<d^a
zL#MBkLVSaeB@qtbd*u5bD2meV^OxJqFAG?AosUf~?2_l@A^s~Z4oVjwv**mtk!!u2
z0NuMFEH-&H=8+wqbmd5D<Y*Ocb<rZGA`=e02smkEv58(0;KCuq6VnU@I=NN*^k^m@
zlk~Ea|3uSndir(?<CT#4<{fON{;}Ou?yeeYT_{aM!qub+D+H>a4;3e@-|E*+E~?ed
zxo@r><_6sUY1YKXhOeMQuBCp9aBW*jK*7C*7rjgd)52#ZbkmCCb=~P%Hs`%ylM#|?
zF=0R#b47yjX&rHkuf8A3ot-7HwcW*`VTcuzF9y7lWG39frBhX|e<uMVPZD|RInJkq
zN@d9s2zA9XX1$kjGehyLr^z<7VLxJZBa=nY;r#?pPgW!b<N1-AB;or)ndHCSDr>Ry
z(=pJ3tV-~YoYu{K?83*q74hOu<&fngEc#j$@{9Cprw?zEwO>!!U%{)A=PUe0ZVJh`
zzS1-$^7b*>H-<dI|H<o>#@<#af~@w)>mvC^Q3A(3OQnQs@E&l`fFrpngsBd{MT36*
zT51p#|H`ZOJ<fxV(AJ^a#cgMVpM|V7%@L<(7Xb{sLf|=rOZ@Xi?H!Y&u$Oyszf90t
zyYNO&6!e%Z%-N-{PNq_Imem&wASwOxy&gd-nm>v<nkoGdloHYH`kuJI53DduT(pJ7
zrHU@1g6)^z%cz0lpJd_YhtvxwRVFzk>idNm<4Opi+xaG)k;tXjS^JhCR&n<rPjsuD
zgGrpycLRXQ$^Nr`X5a~~`*+c%6h1{QSb@Bst0wzWEKzJCE*gE~j&jbhVff~7G7Vo>
z8OKf|-_o&sFuhWL`10|8Z^S85IhBF+RzlhG33PK)CDEiDUH)@fIoTLTTC`R755{J1
zo4n@&4K=-3HP&zCINFv~{lZCBKhiz%ag2a?J|yL6JMVz>O7cg})!LQ*X(~9%<6n3c
z?V9WTQ)$sW_*lGXBpLo%1@7P>!U+g6i9rj-e&pju0UC|+=t$f)p>ib`SvB*+spO96
zS4Oh+T7`|RZ>4eMBn;Q+HQq&NRcq>64947A#s{Fb&X2H1b~X1DP?$%)2jGmzD8lIx
ziE>iR*)Y*K=|z0PL9}W*W&Yl1NKTq*jGn<aD#b)zL~KSQ%N?gdP`xWBFGZaz>sRoI
z^=vEoU*6S^4MDhoeOuZ*y##+?Ok;e@5cNXOtWft0lhP;N8FD1m4Bow~{nKA=LRGZ}
z5?*0z-oU&@_T1r6@(P5OC^iprNw+k2d|G%F-HNF!u<dykw9d3=S>L<Gt6uK3opQ%q
zv6121T!Epoy{=<lseAYATyC$uObC~GQ)4%Eo$dii{gudZrOVHt{SRM})yYMBCk?~^
z$a$O5p$i22JDTs!5Ys?1Q4;xgJrl0aJ(VVH{-JEGgKnfVW^L#9j(cp3$ZxU3?(PsF
z;J`1JE@pHOLEl$$J+`wMM%_JY+?GNkyU<>NwY8yMVDwvM<!LtB3EnjF%3Rjp7G#a2
z9+zk<Yfh;SPlf#6s{f{g+Q}X-Bm+q3N2n>3kxFFxZ#o|+6D2$o(hte{w8koCM%yk2
zFy`jH7gzUZzReHNMW9-t`+-k=uI#JPD08^AHrwBN-IQqh#|V%Vm-O0mZQRD_jH%8}
z8V<~=5ba;%B|rrBkDPdjL1(&(w9Y99=o(5F$z|)LmU)Far!;RPwNkXc?_d9J_~k-v
zGqpMG#8)v-`@0=~2XSm-e(y@uq}4m!Baz_F6+kg(P(_l6RR_Nn3+Oz1PO49XXS8-x
zmP!(x#PqBgjtBM+UW%De(v%`<U-#5muGk*u$}r278c_fn5h24*568{L_O%HXwK@Yg
zPY<C!|Avv0MOg~cNoZ1^*cqB`eg8nCE9Ao)1p7#2ruPXDdeU*Di^6B);7LnKO5Rpt
zgqtQhSH`m_u>>PJy65t5_u@Qpc)laKW9q)cB~keH2}MImGi7TBb<_jHdxE}db>Ge7
zbKKQ92di<Ifyjtfj(?*}{VwkZcCG3U-!ew~rf>x21r`>ZbgnI<<~iywMc3(qH!Pt2
z^gor~Vks#?h7BwqKHHYD?m+}%JZ~*jzk+Uen>q)$PLpAsDuQswY%cAW0y&3d<PyI?
z@39BDvGN15%pT%7iPp|eX$MGuveByP$QR@ETHsjxoZ}(KR7#0p8%14C#*Hu73-&_Z
zXVRCpwS(CUl<#>iovPRb{|`khvB#a-Y!SVFTBnZx`rhQ_wA~$GAc<pX7*wUB-wPi)
zijqqP+lN>LSd6c;TOOUqou3B#P*MWU#3E&^#-qI*<Fq}ACoT<FSGuuA#yBYop@=KZ
z!7-aZG+L%*%lZ#V@+Mu53)BY87m8+1we!I?T*?b8OE!)H@AmKPt@gY{kB+LiMU8uJ
z#6iRRS0Yu_tGl5<_lP9w0T1Yr={Mcb*@qUNJXltnz0P%Dz8k)OC!xjU$@ysKqmAj)
znjBb&Tdh>AMA9kVc;Gm=I?TMbx}8Y^BfRxdDnNcKSH{4niNVU`8|q`>VC;X8g96#5
zA~J|2w!M*Cw}(ezmuS*KZ-d{|sdiZg6RB^9$AWnv5k3NDQSx3f^z<POioC;@Jy}94
z)46W9aN-d#GbkigJ0sN`F)&?MWN$*2Vw%sU<hDfZ$~Mt+@hDayJXU=rAgJf<;DhHa
z$TmO6o}91A*=Uf<>|tNnH#W@|zwhv<9AhfDwL31Lg4W9SK=D(XAkW&K+mdCv!~qzY
zE%KBj>RHu|HM(64Xl*^*o~B-TkR(J*GVYB}Cop|}#xfA`Z;pm4_7^9L#Zn=FrX_{w
zMTpV&JViW3C`NWJBt8z9b5W0?eoEm}i)AN-XQo3|z*hd=_$8H5L%a4B8Lr)T39YmK
zpN110Qg8M-j2k{q4mGmJDl5Tvztx5N0JVgf7LHl03TLIcV}N%;EcTT3`T)Di#-!GD
z{w?AFW&O%T?-o<f9lnL)I+l>9eavh+rUOsO{NazfIyJD}pYL9_wuA<z2D^Lb^-j*!
zfG<o;fKmmYNrnj?&y2eEM##AFJh7LNlgsgsUvpOVsX8u9B_Gm?-rzhS7H-(fW^@+_
zLN9jE0xvuW>FM=|EWF?z_zAU<>B3ALFW49b5#OZ{Y@mcTot7GNn6&l9tv^UKxm-^<
zKrAs1Y~n8{l>H43xgf>F?NF<{ffC`~U?O3`L~VH5KOMM=7K4cS)Yjah6U3JR>Pj!%
zJpbJ*T2ctjghqP8yIg(Gv8AvJQ*o5O7Pu!y0LJg+WDAC7Kx}Gp*=b<ud(-5ypJf{r
z%L{a)+V@Xz#Zt-{z1Lk<4g*tEUIt?Rq!s`NEX5hXT$rU;>6@5Yf3Z?(*br!NDpIDQ
zmDyPI?Jr~6-!g;O@9Wppn|*n$oQ>T7?vblVL)G|`j4n<NGhb!V&pnGaNVba9L2d7`
zn<eUZdaE*3vLHzoWy5H?aM*2<{SM699>>A?Ay;Ib1DV11jB)o8j>uc6kf-j|Z<xeo
z%z43?SL0r}o(-*^#rAp0F&!G0JRii}^R`SjCt3Elu>Tx#&$`DaO1>x+eH+VboW<hT
z{ukfi0K3`$W9Xh!$RwY~v}uLm^tlOHHd0Xz#P~=PF}N!q$2X2ryaAR&*>ENr1ZVo!
z!Xu;Bki%A<9o`p2_uvL^6srwDP8eQO74&x~gP8hEg9L3|@Alm;{uKAQk*voteIP}j
zk?_md9JP!Xhs+_h2gg+k)|`pfwyEEr5G=KdgK*G+6k$VC_7T36w^HvL3G0@P2A9Y+
zP=1sNMMk>?G`+POm%u4bLcA;T!mP&7UJRq`5_Yd<*8l!E-g(#59C#-l{J`2dP~CVK
zF&c{uOvkmNR>K%rI46r*bxAtcwGc{xmQa_<qQ;~|cH>v?dcW}9O4<CwGGKKXvh-+I
z+>P;#((QkVH)bWSL<vh;ltDuX1qwuIV;2oJP(y^BQb62n0~QW;*kFYgy?!)lQ={QM
zkzQK<{QYbsEfX6q@?;n$eJo&K>61MMZVGvcm+pvW4T2(G(qV4bE2fYb|D&2neWt|Q
zy>yRh+)9!|F}}Nl@zn@Nmo_6&V{01I*hkO6_kd)YPeDj&TQtzY@QhC|qnxP`h{!Kj
zp*lA_^dhdVQ^Rleu%v&tq#uk1+%jIjCp3Fn(kDrfe_3mvm9KAXN9;-7PZNsOlIA8&
z*u%PS5M`fcqo1fh2mYqxc=+SUPq>S9dBi)h`6WF)hl&^$U}Fh;Hv>7X@6j`_mPd`m
zr<C^p(kC1)Llh_XCaRgvK!Qh~DUqAFk~`{r>w2L)E^guQenAH{m+f~-gbcGVX3MUV
zvr~QR3?P9vVu%DoCs5Md))q`Rys&gK?F32swD9pGZ7dC88?q!@p>A{ua^zeN;m%<0
zUuA?+miS%1COsFAOZ7OFc^`Y#NLh!3Od`hE3folq;+R@GJ(#(gl%6mLn9XyOqbce1
z4$=5pDA2K521Z)JZ$wt~qpufBvL~Cjyrt!5Pm{}<H5ayT&!%+(grc__nb-C!$Ms)l
zK}_kN3g2{ylM94kVRb&so08=J(F~*pSI9yN0O-KoLxfO;+H&7hNK$d!Os?4PX#7~M
z%1B+*>J#rHxb1L&9*o#(a$vG*wr(Khkep=R?{t=cw2&TEB~8T(>4|*HJ<xIgub8v<
z=+OaU)Gtt~#5C$&BU1|j;h{0xSo*p3``UT>37wx>i90e122Q&Bj1>Ou5+9GX;;o`~
z4E36j#Jn!JX5#9zxM(ty`uTgfq(w$?c;+&RnuGm<R-}x}Sj2HNt&*%LLOO+D3u}wJ
zwa+~hMDw{eC4L=={Zw=i;izJ`>I~`$oxlkNAt|ukt0{v?ZRo^p4?DxZSh-Yt*CYQ4
z7wuBz$kS-0BM2tbeP8nm45?prLP#)0z0w0EW+<)&&R01i86X^hE<JX`!z}P?qJ$47
zjID*^)(qi0aV~40P_1fV-_a`Zi#k5h!Af&$Uy<$?*r1b>=NAq)-6Eq3TDhL1BXSFD
z?P|V-wd2W?BV1oiPPwWB&|<$#gp_R4d{KS<IOGIVI#PBPRUsk2X#RMH>?BFJ+0`ra
zI1g}T1RD03&ZBHOIX~OVA9PRwmm5^aP4eQ3>X{3JXY?8rM7Nv8mqdVWHCg8wJIEW`
z=Xlm4Th|AuESlRrR$M=IRq#i)P?x?;xki&+?i#?bQ-XM8nVC-T5W7q3<lZIXfW)7$
z9IXHRM=BKCl@<wnXG*_&R$-&3C(!J_g_&nk_e<3BQ+xb0bBBA{cLJz?`ulr>&r0$#
zo?rluFOx~leuJnd(#KqUr)&255ymFwP@+(?l9DRCswQO<3UOJNU36f+kb?MM5)f|M
z7qr?{TY2T*%eD1>R|F99@#tREJX2uz4>e{EZb^@5cN@wbH2LaR-$3@)+RMFEEZ430
z`E$zPtQKw1I6;w7?>(H7PoLX_ofPwK;+V(bGz$nRjlh#1{Vsf_VorGJX-%keNP=QG
zL=oiX`CD9GNgD(tZE7OKX9z&Nllxt5gpr+>W9wj%oqs#6fg_8PY4*xETwlK%xC`Q`
zyL|il$=~^tt}^Kue)*m$w3dYlr(hKs#%7>Uv3=V^l)sIa?;pk$`%2VKk?st;fejMc
zwOrPViIA&E0<Nqyyk7!f*xI2W<RXy)<O{aJ#WXRWF$jNFetxg;ppV%jL!QNeqf1Bk
zTSg7n>s#)OMDikT-bn4(@o|0*v*s@IB)2CB>9IxrU8&3%f90zg!7`}^4=Kz<B~X%~
z=Fa{QxrhZz@-I{I<_}GHWAy9vo=)TKZv$dtN6gpFIs?M4W#)4N8G7Y|5OrYjU`96B
z`DcS`G<Cv$g2<Q(+|RvWo;hw}PiEWW7I<Igvo<%T<&mc;2K6w$FX;xAv4Zd3$NYb@
z#TIHtHpJZ9Z3QNN9cR<_ff$`jF>%pww*;+LmElAwJE!9#E`ydx#<xY(eVM2*^Z}z+
z#I2OFA%+;L0cIG}dkfCyZJQ<7#?3|TvKJ7Usq)p(d^jpMTwmLsKA@;ES+M2P?5m@z
zK_Cygru&fK3TwS=3%)uO3$j&g(YCSj0Dcfa?oHbLLoUJZEfY_GEl;1JL0-tLE#E<N
z79>qyNS-1(Y;PSMn6KSBSi#G<*wzv=>XF8!`(-o4-aM=I;VKwfGAEru^7H7#GtBZd
z&R%tDlgYp!w@mNzS$)XGM18X>GJe|`)Muz(vYBK{Ql7y(nCtklBlov9ebLMbjch~`
zDsxZ?GLCnIX9D%u-L@U&&YgysdJ8Tf0$AFN%+FE6_FE<RrJtZsL(BK&M=c$0j7|!W
z>V+7#9-MtV{z0T2$8ETdT<(=^{XWYAqYS_gpLl!}o4G<QWU=jWs8&J}pPSd4!c54M
z7(H;#q@GXf$wE5?!-l5e8<kZn==QNZa_iHS30@^q2Z2mO1B0ar$Cu#3$@U+|YmLXO
z3Wb=|6cv$8>p_`^g7dyNJpp2WyR=?aEH2FbQxP6^l)+?!!-D0*WY}*eE()JMrc!Ka
zXl}D;Y(D|O-H{#4O4&*z@?7vg#FrjhT25(ocb8aL(>7F}9~k=uK{8e=J&P^BA6;Iz
zC+m*SW^#Y(<9cCZ{irL~dxN;YbQZ#89YPb62F`2EQuge5dig)OblbC%&7_4ArEops
z1`TfO^0(tFjZHe+x`|%?h$&DBjp=s=M$RQ@xa{2?=yRT*74wQ+q~`BlbPoINKx@zT
zvPIlXisz4i{PqWOp;zdW=|YJ$445*99+P&D(1Y8fV0X9bMeRvi@V$LWhiQ0?41)3*
z#aQI8eoW3u?AY)4M@BNwn^&}L8YjmHs~TGHV8*kc?nfxPhX-2V_IcvP;=~iN1wIdH
z{O$dG=QAKfAOAJ({{{7S{#+Sh!uSU41DTV1WQLI|Y2hcf_RBTS_9rA{O0}7rfe|u>
zkzW3}j2yXFb-f+68>ge4Zy##*#bY;0%=d>JB+gqTdpjR@UJ%DH!Cw6gQGj@PsSNuw
zp+NhXs`Y+Dj2>$8?L}lq^cmjzvGMWq(ut4^(<YVH)iOH%dR`s$X}BwY#WLOo2+Cvh
zD*O0gegKFIG)VbN(zG6-Nw8~SgJ~eD?G)h`)kyjl;V!A!eVxJ&q~NE~myGf8k6*_b
z62;G%T-yGMBMpD)(lOA<eWTC|r^M4^6gPL*VW|XgTU&i<y-|ZYt3UT9|H~J`m_-#h
z7LKPN{1BPnlVSBJ)?J&Cw8S~!LE>5gIFm*<kD&Yhfmx7rEXS&(IAR4fe;msIJg#Oh
zBc(S<N!ceS@K@c5t?<X#FFj&_olymY)2k))*W3FS2P<;}H}FR?4K&U;gU=m>e4<mn
zyFZOkjM$7G>1cSsDp?zrIeB?k&DlwQwyrXwo?_=$X|krHl=RASXA@~_rzn@A3I(n+
zUcyJwDtwhG(<wW|5k36~yQO7ZY-#U?%yBTuEut>n5QIPUj<#d=fz=eXKwrM*<yhYB
z0o1`l{hY6hz`vu6xZ~G$cQM-baMDV3Z}ji*ex9zi=`Wyl<52jH3_wqL6KYfV-S;^g
zSn)xo<i`fSVvH_nDv^~z1Oku*b{N<8&_OfgmKndY6`LF&8iOYnP^U{t<#$rQLC*O+
zS8G1=>I6XwJwy4RN}is~Bl`}?(2?iBd(U$|X(@?;R*X4$2|FOgy<Wav%g*2S>#lBk
z_csmVjBgpz{&~UO^=WkE;{GX;=UbBDFa_eAcC2lhSR=m8UU9^tA|`Hrs|#snd}$@9
zcJ>Os2Vs8DF0;t-$EYU?N{LIZ)Ano3Cr`!ps$@O$651J=sBpp&thM<$wzL9Gyf_cJ
z!R$s-7`pUPorS1GMAg{8clQtkF3G!=2^|r)*j3|9;vLdZ4y4f>^kmiuH${C<co92p
zrP#WSejK-UWqS<2loa$Rx^M|Yfj<2VRg#epRH?$YtQoN|#Q=$>9eAzck=EKyxgfux
z!HzTTR}~yuvZKR`p`M?;`4}+Y-1u|g+8ao=Oi^<sU37~tYFz7n;2V>j&2V^_<K)y-
z@#j}ocAgv*IXv3i>g!aV3ZL?aG#Iv&Ebe_ti>#tb8D3gGcQu<Mk-qPJwnyO8p(M46
zHb`^hZ;0=VL{juq+E*%6Lc%g(uLFd^1j?<>$H*596@?7A&|hC)&AiW@fA7PwR>yPy
zs}fTB9U-h6EDzX{;@~5upsFYd5ng=m8~EIos-RVfs0Bju>&)CA*z6Bmx6x3!5pXsO
zrW(}Mj9_`oxqYYcM0{oN!|PDItnBNXZNI2|858!@9dXa4S0!AGesU3c^Swyu6bS<f
z_IDn2|8GJJoaqjY-drtG<w%ay!{o&)=kw2_FR_b}h*i9bhd-t}CUVwb86-_Yyis;5
z8(|)pR_xmNc_ORMRO>h2(o{T`MYj!2@iA1CI$fqPBHaiNWAS8a*GrA^;h9868wM6>
z0s~G-ukk0BS_1_RWS9!T|F!xAK5sIf_p(TtbVt#%A@av`S$Nh_mUIas@v3qyQIF#2
zg>eyzlTWR!b?=e_>6=K8*_fn;6S;bRdivtf)6<G87(=w&5C814dDNrrb?oG2c8nHU
z4cb$hZv3tA5IdW+BF;HmYt6g`hP>o=onikhBL15|-A7tq&-===$QMV%S7Shj%2Qwn
z7OH-qzN&`|=@J9Pu{fj8#Hs8&l<Q<g;+1}jx!Xs@Usg7#P-2ODCp4J`Q2fMF$GN0R
z>9M5-A*cX}guD*u6cQ<6wa<@cTPrOJ+$~bfiTY*%k?XhdoG+TtsDINmX6mcyajhFP
z5sl^qm`dijge9fmo^h(S?u)wAKN?iMh=lvsu}o;!`!peJ18A@y@L6zjDfZ1{{PNAk
zjWbyHwv-`J;=$qXzLj<FiI5-|&>(NlRZM9tSc{usnVTx-YwU|KV&`><NwyP>$EXgV
zr$*3oln)0CyGBEJa>JX&Cd?jLU-+@a@A;5J-M`-ktaaW62;YhC2n6J*Ze`fKv5SLF
zF~uUYsFl3QE7u(%2&1MJL}Dj*V40M+{6WlEH;@FvRZee*DxYu{wMw26xh8m>I@dgh
zm)rfj!XfyKyL{;lGYSiT+dW?fX2%RNb29kdfW;+x-VWUpz=E0KS~?=dzWBz@`Xmz1
zQj+f59G9P#!6^_<z8pb%N&p}z8dS9WQ9v)sQ;_b?JTj32qvdZ;D}tR-+if+g?v?{4
zl>_uHEEPy)x8)rv^;7EPoxxi-_A))r)KZdO>!Fi5Qk_*yQ*0bPN02--4=5=(x;*UM
z8k8naOC3vd(a>*c0`<UxT#Ss1(lrWWvUC1y{h9DbuE7BdhGkrWFI%^L8zYk|928|W
zsD4TG++(&IGYIb5l|yjgB0lqZSz9k(b$-#D<ErxziuMi;vFcfpNkXz=HzE_u^DwYx
zV~KQ!JI-k}c{;)JwXO*z8Y_(G5;+?n=M2z(!xEcxV$j!|DvuhofE7d-!f0cGpv_zY
zvnr)w)7TR3)h(X5UQ$EqrR<SvY9BrrKa^F*^_Lhx4RZ%w;QQHfwQMdmRw7t*Jr?#N
z?_2x?P1z$r+&mpJvJvz1A0j)u!xP2$*XDXyeeetCf_9zW%_S%yq6v1qi6~ld!$SDX
zKk%C-ke#48?8{V4U6Zp%l?Pom{Wu#WEE&^^S_SWCNml!}k}e&e>~gfO^^jEUg4`}c
z;63&y=YJNK2(`o=JF58}y-cvHV5)+a;{%F}((n2Nk0|@^Q7Sz5Ib^!KwytxVX+3E0
z4JzHlfpD525mYEoo@wF~SMlNx<Q&?q+~Y<-9e8Iu&AfxV*JsTLc7H4|T-ik2XlZNb
zEgW>g1pb?R&&8O4lq=bppluV8avS`<RL{nz6$8SpjQ9JPy)0cpv!<}L=N@kQ2Dzk<
zI372{mfx1^D8(M*4#@`RV#qVy<<xuO(@iwJ=k{$lc__u<z_aOUy$fPvO)JdDGPk4k
zX00_5SEwZ7)2y9?U&1S!>Uy#{czY>%lPT<RFXfbC(I9Vpn;vpk`M9eeTsc`m74~o4
z<w-5=P01yfK<{C#N1v0ckSl0RsI!Lsu(osb?=<{qzCWc-1<k-mo@@@=3P01Sd$vgU
zKObweTGt-Hf}OBNWx#uuhznQuK$<><4}^O+f=Zx*<>QpU0Y_DzwZH4xQyS4}xEobt
zE<a(4E-zQr))slmR~r>I)`wF?+)637w)a!-<MeS)1Rv>;hny*?i(jJ;h(sX5=m<rT
zypBF3I!cP~5EL4mq`V93`dz1g%4QJfZAC8+{jqgU<(&J<)LIRrTl;kr?d0|w=|?2h
zBiJ1`lg<(rIvSomn<F|)o=ON1irRjE#>PDR7sk@ylDy3#E01$CkWT5>5ki+P?;9vA
zVu4@`)#nPGY3h9Vc}(ah*}F^Pr>B~DmPRA^RZOLeG>p3ICMaqZK!d#J^|0LoRf$u;
zgd5WvRZaI$bT<vLi&w5!-DCC~cK>@MCVxdt&zxRnW$}C9%BL(kj9l{K{3(eVgdL32
zo@^nz<N`>SMZ%GRosIIBz_w7Ncu1js37>?`o0csc8R_S7KlAd`Y6Kk!DHJhZ>|rhr
z2pIqF%39w6A3xDWbfr2UmEdFcnASju^=0j<mO~=aHr|Ep>c+I(A)#Gfiai`%KF)St
z%xCABQBx!g-M@XDgaxd>P&~W5B)*7-)7@=sJj4Biu$36KAO$}Kfc+eH)<8`rJqjj-
zCBv1;9~V7X-zFv|1CO{%PAy1)c#+L$HO<BbezmwOWItCQ776r$f8oW}6%s;DBi<tc
zUCCH)E!2ghf1u&g6BN@iMziK-2_vX;t*)5c`p4=ywq}uK8oyR#N#*{SS?o3#Pu9Wm
zxEPxmW6$e*6NW#yW{-sp*&HYV%`H8V-+-ZO7@{gi9rI=pp9#L-iZU;dk+Ti>1z4d&
zMBG`4Z4<u3=k{h5Wog3#{Cq?qH$*Ff51Zz8Q#TPwJK~4;|4X3mB(9~oP+6>5sFmfd
z7r!XKk{9>L>%H5gHmC_v)TnMs3B&GTKM{)EL}WJ3!jVhy9lmOR=KFRJ*YwLdMJR?U
zL5Kk9n;lg-k9fbmsFjUJI3w5$YBfQzwHgQWJ4^NudrtPSemi(KUt6@7Tps1XYM-{G
zJ7V){67{2;W+lg$1aVi$i^N?LaG)p8iiA>yYnHU;q>UsY5L~1tY6Tx)?+mcr+JI8o
zuz8y5$L}(O<8<=ilq5?fIn0T0m0y$gL+;g>t)b-JF}3-9Pw(N&2=S3Pv3WWr=^DVq
zB|Izrw^^&QGDqD?`izW(GrQnj&R696Jd>tOe=(I*wXVR_Z+zO;`*QB`#I83#cRk}F
ziT)pRj^_bBfR@kz)?9m@r~f5f9x#MZf?0Gdpx=3i3$xv2qA+9;zz*d{B8gJ0z?KFK
zu#E@&%;X`I#;@PHPJbWv7aN>-5!hCz<5<;CM5v+EMos4|@NIm&&ijeMCGhkE1RRjY
zH+Fues3S8y(;)aeH{?bvF=*ROJy*P+iNZ2hwjwDNOROJS_l_H4QJ-7X?<Z!UkNhKm
zebC2aRhqE8uN{&mPT%#_ZRl6?wRrz3Cj-TLF8dd6A-*sRTV^@Lp-9|)XuQRvMuqg|
zHF`TTmWmkc_!9n@xeIs+_g{5A;VyU;kMNKc_j7g%JPSeY7)m?tZxeb)7ut6y-f7Tx
za1P~(W-L}->fnSVX>3_~gcLrN7$~k3c8al-dKb0F*axIE7t@`{b;X!OzgyGzPZg^5
zyLIhB>4Q`OS=20U&y62AIz4iF<2(?w&!?L>U8h$!HS3A*Pn`6QQdztgm10Ku4dOLp
ze=PP5a4Y*7&?`2baDDiYmM96yn@%l(aHrX<1wDv^{<;nnub*u3Q|;X1OwO<Ge)_au
z^aEnfqehLGT-((Rle}YohK8J9rj2qtl2xu4cJ?pbQ;4i$wu*|j-9mld$lyonl5&^2
zt5%Ej^HTYNC2#CPz*~b5YI<Vv9&2WBN}Jm=#q+?+$xB!3>Bp}4)K*Ehe>_^GHu<~?
zgj%Pj3`67)W_UJjWDq_w?GUyB-Vcu*Ou>$uAj)qcC|Wi6c`Swr(TE_}7z!LN8pJ%=
zihx7Rn~_d^geHwctmGorsd=tsDan@EA%mu>zkrLQTS$gE+Q2{}>LM_#dE{?0F9M6S
zoeZ=w2fdzBL;RAM5hAYoc@sPQQ2&Q|aa**%2jwB1x}Ev;`;}G#FL+Imo*M+(mo+Ls
zaz3)i+1z~SrBcE^09rvCUKO0O6hnq==9;IeTuc&to)qX6Q3E?Ra3dEcOVdUs637{>
zw^*7M;P)%_j)+V5TRM%tOaCXk?iX62k4)47WRGd?u*{X9Ukc0)F_PX{D?=@mF7R_&
zEVp&Yxtx5oXueD;7{`tBOA*5%<s=egq*@ZU@o&@Y=my2zoT&01QPLQkEgE$8?ptY4
zqniZQ_s(4-c?@P6i2UmNCRwsf<<_PLZ7{a@4M<Wi=Jp+tVMdY0MxdnM9EzaVrfg)S
zWG(Viw#+SUY>}eBuDh}cK6~=n_vHEBD54Rxjn6niZpEe85_m2^wB9~m;Fd9%P&QU=
zwsTBp_c7eAtA1m@X!oGhB`{9%r9LE7AwLAnSd<vX#&S^A_*xb9d>?W}wC~-d1*VN1
z5&OGg_4r5QnS?U<R_T$8Qqb9;1aXUeJ$S$MQ847;rEiB-|HjR&llAJsVx$j(RhGA&
zsdE#VFEL1q#n*CHN$cEkiwfqpq8^kh#B7)m5_|kpcmH(Bmzm)ZxQ$bS`eVQ%nQ1@N
z$EPZK*03%>9h6;7pv|0iq4i2TXRMB6E?0^=eXgyqZ<TAEEfU;I-W86+=+}5&DDEr8
z(K3MF{5JpH`eWSMv4tnF8-kfP#~|=pU}4LYK1$acalHCbd+XZ-MK6}hwxP)NNZg&%
z!P*&j(Hj+f_My4eV-uVJ(v1PLmG+PPH)-Rs(1dnZzM8WWsSO0W3i>#N!c$aSS=gk=
zpi#Y9Q(bwQPwxdG)0}2Qu8Kre(*G)P0;r6XSxPRrO~1szBei}kD=T+XLi<e4(v;;H
zp~`QB`WWfpj)nw0!-o);X!~|ZACRY76FUqyhKdzJnV~`qgcK5a)fYy!14qRgKP_?s
zpXNEHDd>faT|*-~J>m!RMdcbz>|Figi;MMzC7R?b@RRLGe0^s8HILFgN$oI}Sb1Mf
zmS-vi*r=J9_O+`;M&Efg3H;l=UaPYPU?Go6&=hG{R4*{o_oPK(zv2GT=jGH4+QJwN
zTJvc#Xq(w8;`fYD(2m`|a`!CT_iVa-c<FtK8}W?Mv4d#U5Yn{=G+LEw<d-#yj)r4_
zx9(sP<BRoJxCn3$BTzO5jfJ(P-TY@L=Y`M9v!uC|$DvI3B^K>JIbA)KDDVtdJ!n_w
zsH!H}tRoa#&iE$@EVz;19!m_i`5R@v<B0+?J|CTs+g0&Kzrm0I+>S423J!}QAE=t^
zkoHC04E!Zvw9@>mXFUDTq}@Av^wnKb*QnIF&NxFN)IHqKKo#I0W5iC4se-`r9x3+Q
zZei-HR8Kumt4Z34i#t0a<ha;YFJC@STX(kCO9g16W&-1#_^$`76Wtz%vNDMso(HWU
z4j9FeT#g<+Rc)_A3d94ikxAngJ>VBuYY-;NnX34Zsxjt7N?0gq7eyUF@k5_bk(_`)
z*~?(t11y?<5aD|D^0{V@3^DOh1NZ9Zlw>-PR5>%luwt&%A7x!KjNH?$2AMeub+TLm
zb3OqL#mh7%%)J{Cdse1MTuIk7T~JLK`IUDPnV|0z8DWrp3{3Kp3&O(j|A;!PusFA+
zTPK9z!GgQHySuvu_r~2_g1b93ZXtMp;1Jv$8g~y)<9d4iYwvyT`swTW)vP&cjJJ%s
zChK?QQ^(M}E&u#&Hrml>oamcd{IQV3v(8u~4Cr|ILp_D?%h;A9lORI5m{$>XfTs}0
z<F`&{YwzwHV$~tbW)(Oi;EnnR_gGb}gBk)Qz1E`l6~1GkYg{Am9OZyU_4t;Blq}rr
z&k%CXOg<(2d=7lF)E(j%ly^bMB#nWAft9=c&+M#8Sy|4@a{lW-|IFAaVaF#d%^}7#
zWQ#WU#W4HjyiaXDKX*K=U!+(F+x70=kNvk4XjOu6;~Cyw7|H`qy=&~McAwxkc;=%M
ztt?qiAG4F+CKNThZVlpJN2rQ;LvVcW#IUHLNHHhBTB$oN!R-g3Z~2|CE$#H~-+O`2
zxV^Ur9DjAW!`hFppOcFe%3Q~~w93`4Qmc?Qxuozg0Q^Jx#pPu!VvL;{=Kt{ZJq_Yj
zgP5ufbI7TQikE={OJpxUOJFf5LTaYJ_YwtWG&>|=sHz8{4#@s#-EsC%pwXVkiPh@3
zA7ItYHcf3nEjnt$)l#K^Awyijm@2Iv<Wp4bq8s$p>fnVtMx8ADK>c|(K5Z`BS!)Is
zf`9mr{E7>Cok5Xe-~=u34q8zVc_(8WOB$)trJ?PsIZx=!98;&#-7#{@0H$@6P%}nm
zi3V|>DkNk@nHM$cX)sYaw<WyNjx7@K^>V5@oPs*|608%JmX`a8o3hmGg?KjsT^m%$
zpT*+p#N%Sc*-`+&V*Q%Bl+%JZszqEr!urm7CD(c>Nx~lZr8d7LH_r$&izHjE+WPwC
zed5=IZZRCMmz(@2iV(w5NB_sq9?j?!(F3Ij-_nfb&B+3DD)LM#h_3gdD1WD#kvBu5
z1jY|mNYgIR^{27jN~KgBD6nddfsJGR3(@ha2CVuH4zQRzE}`B}u>+H^jr^iEjJ;4y
zMg#f&*BgYQ%TS`q&GNVd!^(-xZ^N`yds&GJf+UTdci(%)KI1zY6}g?#uqz7%)o{Sc
z@0)^guoV%aZZ3X{kW;qf3o><hCe;QWR{EW^Z<0=+Z!w{jph8iKZaaw;WAr;`HN|o6
zbtvy;sCxA~e+G*1q3lcFNc(effvi1OAwPWjLWhuZ#pEQ>uGhfRh|_C4NWRP`(}pyQ
z8zI#g7^MjS<$(8t(+$y&C((5byRn_9lFg}!=%o~S<(awpq}KMPgYH<M<sz9y948xn
zCtaGLwC~Gf&Dc|xNN^2{@&3X`O8*9Tq_JF8kudI|P%1^9p%f*W{Gc<wz?p^0`kGCm
zX@`s>QcHfwLoX1ha>X%s(%c>()gH3S{El>-AVtGmvOcDr$(bmQD*3PgCwxBr<JOz!
z3MUIQIs6a#Knw>ZikNlXAX$m@EPhca!{4~Zk0(DCEROD_k)-{DRWm9tI{O|oHunn?
zgeP7`%2}kq2lC9@jV7y4D%ph$gpGv15Ojn|$E1M*zlYfm<rEeKk(s?1E;_~>NUSuG
zxIUS%%XtsZ$kaE@>}0$h3j@U8Z1WzEq`lu-4|%+wFjP@@XhpYR&2+^0@RCYs&m3|V
z4Smxb{ZCf?-k-bsCiE&xbTsE({hc1(c3JY5c%3d!U*3Rxs(asaq0uD9h3{2;nwxR;
z+k&`|qb)3f?f}Q7ZC;iTVADmnm~}w--+6{!&O{pE_1(wY)C7XBSI7M>kG<av1HqmT
zpqNdH1zfm@Lt3oEL&P$xYP6zD<31H*yUEivC^1=w+V#PEGK(D7&zWKcsSzS|4V`^w
z3#$+V5Su;CLfH?_19&K6OGXf;F&2U<2LJhh*8TN?Z{c*iVb?%w@RfId!eHion0#*2
zK-XHaj7sbqO`g^^G$eoSRBkX|i&Gu*=yY&?Eb>7zsNAG>-9O=0S1yV(xbwG5#d&z-
zJK#Td8exP?#MpW-v5)CJKcr)e0}o;PnsNvXtki7!;e;XiVD#sn0RPzy_w4!<Y5`&s
z#`S8&(3?wnP9QEx+LW;dMm0I~B+ljM0By2Mj4~TIgYbres!#OJlUQh2097gMft|t#
zG~{YdB6M761?4H{=R5ZorGzu^1!}xnwmLh{A<lW#uj|1J7D~)9^)Ejy6Rwh6$L7i8
z7jQ?5?^R^;%PnA*R1rPPCA}3@tJ9ahe0Cn0mLVjKH_KG3Or>QO4uu3`={Ezz7B_rS
z0<RO^EjvgG^efo}1r;ka$Z)@m{7T(+k_qG}SvW4d&3phw`B%NWj{IBIo7E#|1mu!J
z>=c|IQiw~Nx+Fp8JA+~2pjLlJ5N%L`abw%$S@TAj^IwkMU={SxHvUU3O#7J;tKp|=
zdoB={*D>rSDYs5fTkQ7Ij5#FPm(6Vyn34+_pgfxr%2JBhN@HZ`@HB?9rh~CJ%J8&)
ze2Zaa_fD>B>l$6a*E&Xi@(Pg9pi9vid9sL5U%(48x}34HtSf`d)#dLhWBqWSS9QmQ
zk1%L$8*@(&XsM*|ZS@@_<T}s0JX>m?n#FK}4)}ArCLmir=49D4+=(Ak?-Te?OTU5U
z1S)bhpc&QHiT1D$bM<`@xeq!2&z!}L75npJ-Nz(b2cx#wY7pvr>V{InhY5l&vkJS(
zp>mz^8oQ0@tXmAR!d`aSv#Yc%y*q;#Yzz%`JQJg5gr>q`2(o#WKLm)eml0#j5zqN;
z>oHPMj|dtE!4yio3dtCw$Ffq^V%cT=K?&)>S;yArYdX6Jw~W9p3H>ZG_h8RW@3@S&
z3FZc4seck5l^0#(@Fub_wx<sEiBf1L711dOqM{5Zu|1WG^)<07M5QV_-zc<MvZQ!L
z=k+1cw4T>3PPUWj)&X;Jk7sg+YWsNA%@nbr^C#$?Jp9Xi%11uP#MI`Wi7DyeMrYW(
zHPj~HZ%+e*P<!td{$}67=fh~MfihZXB^v>gMBTbqkR;#Kx<$Y-163W(phVzL5*!Ks
znl8^U6X9*p)0f}6IU01Bu_L@bFK|XrJBN?w_g!NS&EnbEg$l4YzPyfk-J{|tdBujc
z?9DB3YA#p3lAsOmT#rggF1#R&RNHSmx?w8eaMoTYqwY11FSbH&Ypw#|n_URz!v%@x
zm(@7a^%>w4q91=$+a75*p!Ko)Fo+rGadBxtilKy1)qtT7+I9N*6{7b$SfuAag9EW>
z{p0g^Mg}}L|EYBE63HiI11t{BOiFmhO_J2Od}9a$!(S#dizt2d{ot{R!Oy|}byR0(
z@^3r!8rx4Ad*(uFAGWVobg}dSFNaZ&G6-j_L)_XXK1qFvLR{)|TBHPq1_mez2^Pr{
zR6mHSJXQ3Q6XFp&_R6AfmYx@@EwE<6YYeYjqCcjA$=3^E0T{B{e=h>CPN0W}kd_`W
zL~*(C^F#m4G%aumx)DHT%->ox>KY7*VAJ37dL7u@UiUP#v^itT=jl}F#A5l;*m8T_
z?7g*yt42HWOQ^M6<TDLz5)N>Jw#AaM3zg?^S3{~)Q_4Gdt*nuoxvs-0ljRbJY8mI)
z_uA95`_ebirTeh?UB?V!ikqFhn5rmVf(IZ;*_xq}0rE=|1V_%(Fn`xTKr%#lI%e6W
z>RvViMh?x&&)4oZ>SZ`}0z$NK&`Nq9zqwjG_StSn8LD#)Mx8~p5p;w+fZf+xZ|Fvk
zqgqRAOJs^SZvq`6vM!(AAMyjTa;us0-{xHz7{-fw5iE)QlZ!3USI)R9n;o{QB;{n&
zQF`dwUP&E0-IFStyQZpFJ(pJOz)=Z-smsfF1+9lxVXo&)zP}k{uwqQr?r=-)!RLb{
zCF{h!wm}MrNa*d&!hqsyfdskI-l(tCnR(g8E16}>z@42iPwOVnVTE`{0buKDP;nEf
zqWX=OMhubf-9VcQkO7o@NZbqi^6#gP0x?X%rG|AtZaBwI5g#zJH4r?_&K^2h-u`z^
z#}O^V<^dfh7t5kDGkwcf2y*MQOT8K}q>s)cId{fi3Ht4)l+nJb4vMLKTG|wMd|8>~
z)vs~(HY~BIhpRO{Rrc_}MV9VYS8owv`|G~6zEPrDz5NTx(sdYM6GwY+SeB@o6@PQQ
z*4Be&bh(zqE9d-uOpGoK&z7G4Mvu_~Uwv((OI^j_N3f-0OOb<3ZKtT5tOJsas0J-0
zADX41K+NB(FC>o>8aw;qU(qUT_>WF&@sH^>PEcNNp7-?QxiY8rQ~%e--<~S`ou=SH
zLl(`VP!V`U-PtaW6b3@ffwVlF3<ka$eF0@!ENym?Xxtr^oxOT;iL~!V4=3;RD`#l<
zCycu7{;^CLL+Z)H_pCU-eLe;3j~N{ry{|f=W{Am^MygdvG%OqY*krdZQ%&f8v1YWz
z7p&P>JLnvP%#nX9mVGgL9_$7*H;F^IG?v0(*?0mS3Aq?nPhP9@Z%`3`&S#Vkd8szZ
zc^GQ^nPHhrMEFH3*pyq<HkrHD=7P!e84=R0PcAzW#mA^0>X~0THlOYs8|GFhkq+*q
zW7qT=tz${YemM*FZjAIA=MM5o=LO<0UjOHvc9f|nf;&`w@%nuN(ua`wIaIkFIUr{l
zRiP?!Td(U-AA3aEV8%|OY8Hcm0=Oh$2^_dvkvZ(pakOxsN~qChrIDWJJ7aZe?%WY^
zup6<3PbftB5Fw^%s94I?HPU+A(kyEUjHE-_J`j_(0mrv&AjJgE`CXDpgt|PKqRHB!
zS(iZ~ct}3F%!o03CIVSHp82z9O|{E`ey)o?8E)K<Z+~S}OSC2oCpKHs74*c5X9{#{
zN&O`8l+g$Sl5eihHq~u!hvXs72h_8Fx-G3bLTnOjeY$}pDK_7`kLkpnk{A57<fls^
z1Ez^Qq;M);V#ymDrY)_a`nh7V@>`g8^Le!`9>)UsC(Tx%jLe)|azK27-;H%TGT{zr
zw6bbLSJE<D?UXcTSh!cD>`PX?m)tIW56i;oZ&<kcm@Tdq95q3Qda1;NYz}74qe07G
z<9i*~2lS9I<1dCoaQ-_+xiWJ?=RTPRow+LePN;-jm=nSw!z^DYqZ+IvUZGVW*Q7}t
z2@`i#9t43C_@Zyx&8aJ+GK+tB`UO2RC;PCXTqv9?yi$t^eSomecqo@ua%h}qi0K?l
z^Bmg;YEavatBv`;5!vFXR!(Dw&4a()&j>a7rq#-|`StZHrD${-sH*w7+n_jph9Pqx
zOsQXojRJ_+*e!fuJEJo67KSc9=J;bCo^mD&!@OTvrL*YNcS5jF+?<0>yi&NANhGtM
z9bZvrC$bW{f+A@VPHq`RACono#0M#n0x>^7Q|0>h)v^?+H5zwF!t`o5i57Xloq)6p
zdKW1V8+FmfuBMsy{hZE{DAyd|i<n|tWRo?&Cqs=25SH>8z<>rI<sn7xMv7L%VlPfp
z_2CNE)UVKcJpuN%Ag~J&h*g!>9T`=vaL@2i$@wdsSw;V~G?C@)sQop6`xb#_a6Pi^
zXE`~HWYx+(kS}(opf}7^%%22En(SxfI2<H<cQ(mog5!L8<V@sy>I_7yT5IXKzBb|F
zUOc|_v8tZQvZzg4(v?rQ#lgnxU9an6O<zM8Z%JEHtrNud_3+$vspbvbwh#Uz?j20?
z1SOro08ohU!yr)lb$OIaqeKb1Zdc0tuBJ^pgZ08<>R|#u1&%3WKKF=~rM*fOqpiJM
z$t9I%hYk*F{Va4nKg6=)1lUI?B_TonR%jL5ZY*&{AXo94cxHGX3&I1FmYDzl3F_*q
z{!8^Hg@KVI+-Fb{?vT=EANUXzq?#XrPmg~SK~C2cx$=T!QYtZfJhQ$I`KZl(!V5IT
zWj+@@7`qh5ss@VC-oY0qG7w@dQt!2>tS5hw{@!~p{!29<1i~Y`tWZ)U@69yhFkLz)
zN&7<=q>6q5E%u}_N;Lb~&N7}NujoR+4d$ZL743bVjC}&v;}^6rQXN*=;o)U(<cZd5
z<8KSvfp4>JB2B6T;LiFD_P0OZWS~`B6$Iw`=_79JrIs5oiqGC3{-c22&x>$+R9x79
zN{0ffgvKB;2;Y)W)oq==6%MSJ%G97kW!O4&Kfzcm;U}ObNtATlRd4q0<0@%08Tv*1
zL7~Yg8&->%NoHVYU}mnbW06m%q$rLR;D(F7vvArQOtO}C5YZELh`wOApV?I@M;&0<
z@Qv6-AdTG3B!Qa-UEe)&d%spMK0YlP8CzPy?$ge<<a#9}*0t2M7tVFE2ctq23WOph
z1&qJ+vYFt-zfYcCEO(xQ1~GWuZ5fN5eYwThxSRV^lm*6K)=YkTdK{y`d`2sB-L1)S
zW{hp44lBIy0Il)sKZXqd=f>oMKy-YA;<Stg4|t_80No71V_I4xosk7<<aSC8ZhtvC
zw<Di=k#xn<ODKv%j|U{NCV@o)!VxRPVc*t5C*cKqZWI9d#qunJgStSjJz2TE2#Q>+
z)stpbX9aT$BIei`St(mj#A<fnS7^~F78+H^eBI{TwM!hQDjDw+&AxQLxrIp-d4hnF
zv}9G2kDa|eD*+V^GlimhDC+oxqVxI8Zlg#0yp5tuZV0_i(H%?`KP-=QD}j3D;t#Hq
zJt8Cn)qAaLalg_xZGI1kbm{d^&^VXXTU487p_4O$W*?1)c$1m+e#@F}tidY!V_Rfz
zgEU5HKv2ke6DD6y4o%@D>rrKGilJZF#oL`@N;$k^eqiOWtL?zu0dlocgmS<&JA~dj
z^;HTjP4{yNy%KKT%0+hlZyrmG-#k=bF`!8I-~8_#-VB`gyroL@d++^z>4Qg4m|b0O
zv5ejan@46lj9opDEfFS5I;X7*KYXXLppwJHaJ?!i;^h>_88Xv}ANf^OYz_=*-X)t7
zq@32>g&?zTM!*=hq6OO8+B+5HdFCDtNdbW(dHb1HnP`V=&q@fsQ@*vykeu9q-9|6h
z%>nqyscV=VsGo5wiyKyV&OA4aGcSxwP2Mh{ZnI$MZbb2=^cZDafS9r2w97Q2C|?z_
zN#nvein~|IE~AeiK>|w#wGpyc8>C0$@66DT1w&5LrM&#|O}sHp@g|PLb#vlyDgkJ@
zXuZV*n$x|O7=w+jV@}5u`EMtIDT7I{%MQ0c2!s?^xW`x+8u(r8?A>(?48&=IxM;oe
z$iBrBQyG8hLO_U+qVp48+C15|`hNIGL+iYfY!-e!P9_Gez-d{bF}h1g;0b>^M(nTU
z_;282u3}E+c-xni=iB#Kk!Z2wR~CxwZ1dwn_9dB*P}B_G_407rmxR+NbL|0$#PXf~
zi0{29nI{)J$!Ns^*qbT(tvATMaKE*<aVQOt85zMzmcl*CFa7x@J{Ec8L&)kCX(CwQ
zV*(6`kQ$ty^yj?Zf0lN;q(7cBIxzND!=#ilj4SRw6XiN&aBuj;EiBj@`rj&6G4YuK
z3F&jVTCPs`D*k|-^H2;!(>}`NyOc&xaxT{D?~kq?9$%bqRWV(I%Z_|l0l$xmv=^q&
z&VKz4NC7$I^!z{(b9uOQ&hvXJp2uS^v5p79tUX+EAY)#5Kll>z-4{du+wZ*33<_-O
zk|!bno4cDe<zm6hy6Z)*hjh7Vm@w`58e?dE>q2cA#xS&#V%-atjSGW!Pe*2<D(Dr^
zeI7Ay4%iU9Xz&lK*l8Ht2cB&f?`V9=B?eLQA50A5t?^D!%5*P(q{E4RpjHZKVARbK
z*OyXOh5*<34VXGTg20Z3hu7`c-e;bMBygX9E|v$n=6V{o4#sGC9K|H&>e;Tf)QQ!m
z&0at}64Ir6DPLeUcF`wsLRFwmVhc!wh>)I!r>TExuqaJ)rmt*&^~#$XU*nKRbd_Y)
zE;J#=iFi-_j|o9q2)OT6YBST@HPr7n1#=MxT|p$=dz0n<bWwv5y@lW+h*BrEl@psw
ze+chLAZpjG7(zsSNT;uN<`TfZy>_gw-*$~0k%2YrDq68ojD419&YS;wif!%b({T%`
zT!X8rb2+10S?fcm7__7yz3VHNgpFB+4J7y$_}&yJxTi*YFq}l`u-Vl&I&pF;)G~ta
z*fEi%E^W<KIWUOGKe@|auyySm`{>Z`rHqsC@WCQm9myPp6neyI7<7(2r#`1$8T0kX
zu(GipDg2YBrUMKA30!ZYcmx)jSY1oq_`!-|BP&kveP8`x=Ih$Cq4ZRq=evoUx#GX-
z;}+l?HwIluZ0+ozak4OstVxK^b+3=fCbrvyoB|{4E>9$u!%*dN@UavQ6M9JiFn{v(
zK}aWk`#C^0XP-co0ILysu%H@-7VhUa3Dj+0^u_@T+*YDld^flt(fO^gd#>eEZ4bTX
zBTIND?#9Ma|LVS4qt2d#b0#982E#~PH)}5;Q_nwo)k69HPVp(D^{riB3d;<SRhcqo
z_{1ydCLxp5BQNw|*L2lnm5S8J-IhqMFiJCTgiD$X0Fn2es3C7)kV3-A$=&E&e;!eQ
zu~mT8|GXvpIrzg;g!>hy9m9h7h`VB<u)j5SmZbCdc2=-zIO)VvMMy|6QRT+wx43#n
z$0W%5e4^*1EoFVhU21#8m4zXR++F_7H<<jF{zL6fXIM&1-QSOYwYmqWh2CWO&XBR0
zIy$>aV;2fk28&Wkw6&a{mJn00<a#dJMXl4(he(l!uCGZwetWj~?F(9DyFw0wx6s>_
zrIYV1R`;0KkX_mboE-4os$*hz`qzh#UBqe?8s26B;koPfhTi+0595|X;sg<?W0)n$
z5W_`fQw#`3M|J*$6E#?=JcKpdEH$YtoI>_WVx|WW3vpxko7eru$xS)r^{yGR|4Xd!
zzt(6fR{2z3;N$T|DPN-Iull(N<2(^-;G}c<t42y%u*~OS7BLB;-uuho50%h9ndHDe
zuwQ|Uy(xnjLI;TmUXVNTvA4|TC+iUyK^IHsS=Rr{N}e|QH0qCQ)mA$-rR*k}O`4{m
z1VsJ}eb`3_)z*1g5h?3d;j&hNT9!5_iIG{e4qhs4U-q(KjPz}cpUGiDh-BGFWa(t~
zRevsAN`7aKcF%G~P}D{RC21@AJ)lashhAL@4uF*f2Nt`Y?72KW;QPL3WqB9<kZ^dq
zfo(_fq1CaFMcN})>pY2|{NJcB?ZPjP`oI#tBk>O+!o7A&uedupsaPVEE#Yln1n*4}
zwf6I%=S-P#$sNHkDv{S;Xk!06jHL1_6{M6@HSHahZ_Z{d;wyhIhl|Twowq-4@V(=q
z2wi?y+=Ps;D(3durY3Ue2?SkYVTruHH{BB2sa-Z>P)o;LhXiL%Yd5;{sCrEX@1#CF
z6XpXMky_W!An0XnU@kSPkY4*q3Ey__0KUNE{H}y^XrD159Q9C2p#6*>dzW|KR8|8D
zSr~;py6zd1E@HUsV5vNIG}qZ%qD^Vn<}n+=o%id6miDIh5O@^G^Y&rEhLC-7LGMIH
z$^gqUpYl3J^KUzmAM2Zht$4cbuQgLe)#<1IpV$N!ddv%kq@deI%Y;$&$W0W`FQ4jW
zk9|3<OCYp>6Ci+LM!vGe486ckilON9<_(|~!T%(OF7**TC<Ph7{=y<s0mUC80w<Yg
zt!v9)<>&@6|36Sps3f$s&Fr@~N9f~w9BIegVV1gtEmzP=Y$e61LD|=}7B)0QO}bdv
zPXdE~qS_&=KVb>bPODv}$wL}pF`|5?dTl1vEA^*7q9S~;BlRQ%qt7JJ#RB9=qsD^Q
ze?)Q6^1XxmnS<PJR>(DLjLZtN)zLqSazF%aK&Wci#5(3|K#mAPFpEfniwlLE813cA
zx}jOPKkPnAZ2wE*ulE0)h|$IFzMUP`pUOhPa0pX|%cdPMsFFWWNm)Pz;`xw~64j<|
z`SYM8aM+rl;%a;-fA1(Hh9@bg-pckWO$_#9oJC4RxP8Foq$JVPvT)B!y{DBDXG>@O
zz>a=O4mR`UpAq$kLSx@%@mgVt6QDD!PtZ_H%e0L@xAMfpo3r-4=VBegA28-nJaMTS
z1gI(8_oI7knccF*=atTy?+-<IZZ#nm3h^K*nkJx>F%{P7$LlOY1x1?NXCvAdsgTxI
z9jNYp2iu6-TmJc<-;!q%dFV)?kMtWlI~AMM4%fsJiZ%X3?OO5%?1&~)^neAVMj{~*
zBLUVG;fd2+jp)LDm}<otOta`o#(Po%r2oJGFIkyqg*|ske2q$7jlV2v>H&3uJ+&G$
zCefpx#l*DZlSDs2jkEM1g&ThnJt_{Rxb!bU#UY@92Nn3CxGHP>*gQiUU=6@gi~j_H
z_^2me)F3}jPLSqE?R-XSviXAh&=}zl!AmJ}DWFcyUQ(*3A(Zk^Nu#5YmXu)7q?edA
zWw888S1miHAK_lb<&FrpHt{mmygCw-37js~txfSI9hLPH?zNGwP$NBacUIcLi$H4v
z(2xP^VrL=wu<oG4=kW>7%UwJrD!V3KyNeojoH(zoAXwf;ugp#FTSbSOYG*6wJZ1mW
zV(o321|Y-Oh>LhK#m)1lm38FRGvrQ(4E-E2+dw1B^AKOJ|DKavn_KVa#rH2L67g06
z;Ypmv{-G^nwPXTePKl+orBb;mMlT#0x_Ur^NbenhCGEFQRk);K+<B>pN+(5_ZJN6u
zKXEfvT@Pw?_bty2wYogw6%+j2&s_EiNt4gu^|(XgBUY{D%yIxNEjtu9iB|aP6d@wx
z&_q6uKRi2IyyZDLh+;jQHV|*GwMMk~ZF5an5L)^`4C0eO0Y%<xTn|@Gqrf2Y+o^&R
zaHc82Do^=EG%KBN6C6}*-CiqI^KWB8&rcivhl<InwObNx>gAf1+XQ>wlWFu?nYmde
z^TaHIUyIF@+M7nlYCFVg7Xg-*_(r|hLg0$m$EkadfwfiVhWmDu;_p8HO?iU_sYPXj
zN-~@2oV5|XC5tA3rI;B%4()eXg0+k6>LplfKFwNdx<&E9f|9A>3l9Om2+*EUA0n%8
zDoXu`?OkV8#q!GL=`ius%`GfO^-m~X4t%*EpU%*V+C=)mR$tomSGB3f*nth9P&RTV
zf(a~e70MI!et#+2&$xoX@V#(GJtO1JKUN3bxrbpsalw*1l+ij`W)J0HZWt1iR_$EP
zExh_qC@39iT~<qLGtCkI9%&TkJnX3rYJ#);!C4LwLx^{e>UU)@zh5t{Y#iizPJK-S
zr02ND69`Ln3Q20vWq+u0^$L(_;Eg+!zZM{uRj+bS9i>DKURTP<Fbc?`*Q&v!Be}Eb
zA2b?@{4#g?i4+3c|2Gh3Zgs6k-*KN~w0)Ez6qA)lIYi>$+Wk>UR)B=?hDJ+nJ_F_I
zD&Hp_umQKNS5mjv2b617+Huj(4yos1FvLk*0Uj!IJYFx@HioD->e;3i(c$CUYW&3f
zTZ(YF>6I%!lmHH!ddKk=+Id-ZtZ!)5bBpmppeAspg`z&Xi)iWmit>5abvuZDe2M?=
z6Y*3?B?O-9iy8dK(x>esGLhe<$?ObVF3PcyfHL>yK;RL1sPvR^2uu`%)Tj(8s?AF{
zmD!ce<WB6M&cq!EhMrza<~C9QTfgmvsL3Zzvt2UTT))FD%yYv#0eesW<pEoGXjy$|
zNxysW@7Wm1XI&JPt^{|O!{93&-FR<10-rT3`?&Rd(t8sOm4B8nKq!=3fSPKGrY%_&
z3|hu*3Q$Oxw9<^<HCmgGgzFe;WGVyPB|)aZ&^D?XdMdyS!}z5a-boh#Hm$Kl9gwuq
z;xA&MnBttYYU9U;HdL@+%!6xLZ+`iEV%xzQ8Hqn<*Tk)T-0TEG```-6xWxW|rj*d{
zNvnLApe?yxJ}oI3EVhl|qX`qGA+8mY*Ul&BqAic`Pxk<|wsubD8j4}0X*;^(P@&D;
z(m^uuy12)*ekUn^f%<Yjd;pI(R9s|I0!<vBSFZLm<U=BFJOD$VnMkQFC0$00l61jx
zCVd$u4GQ41>`yd9kqSOZh&O$|GQ%5O-JF%3y(4e*0JUHfM}p*iUOaVulcG!#<AnL=
z84uc}`f<SMagi7f9F<F~z~9YWKDh%|RUG(8w@6QkCG@PL!A^n>aMZ<Q*b#$Mv%S0y
zJvND0rp=$mGV|9A`Nd+@99eEtq7x78mW+-NCCoo`G|2eYe}&iG<r_gB5q~{u>!`}y
zER#$`osf4hy@QUu=@oI-#3*!^bhx~{VrCMMbXINOI8wo#!%<$I1m&>VI&o|3>4X)`
z@&^npr3Z9|`#8vREwDeC%b4f=v}+bssE<6NkyU~M2FK`k`+TzJ)S5+8Zgfa`GtSvw
z@T3*K*V&tAoAt@`9Y%4+FzjyPsbtl4JsZSq3*q8+Fg$#ymwp&*?ivw($L{X02T*$*
z6-MC1v+=DHK^@(O!%=iyXJzGy^6O#X^G7(}oiCvS5I#d@XIO-MPR;6r@k4QbsQ*(G
z^woLXgBnQ?%lwTHM&%Xb1OFC7H&LBciP2tsHCxqADNf=s@3C^@Bb#6k{vkzMQ{?#s
zXFm*1y&l)EcN|2AMw<+3gph{0K=1<lB+(Cdaxt|A_6=0oRmfaRhV6<DNDCsrFrFEt
zFy%_|Ae#_Kk02Wg^owz|tQC}1t2JrT<Wp@}_vNCto&F5|<yv=02&h#Kq=C0MgwJMY
z_=Z<iWl*R*DI;+sjl3U5Bix?=2vvDd4V6(B6PB7H92zGWfQMHi2?KD*L8u#au5<DN
zul?#IjNA^JCibow_}mUIXDeR<4wT<{C-eNjU!H72S@Hz9Wm)F?!$j=Face;`V<XqV
zfdO#>10lIqeI_ZFT%j-DS+I&HTC{~Wsf!4ag+phYAeUGoBrelnq`LKI%=qvY@nH^;
zLGYdeS_q`;mRCHOSIvg2-e-WN`2Ie%5h$sXzJ8dy<(LDg>E1T*QP0}{q9Vep9ZUg-
zXI)Z8(?lU>Uy=ussn(f|y_vSv6jqq(bE_<GqNphh|FvFJVq%+AB!R|3qW|q9P0aPT
zV|H@#zMPYbdfy9g9&WN<B2Z?(5EK=z&vjLkaM5t@fv>smf}4*y7vzcI!#b&`R;6{V
zl*cz$O@k=@zZiNCU<9I6l1C+SbLiXs?24jsee8$@B|ADNH*Rwp?^hHyf!k}#V7nRz
z*OL{5g!Olpv=R8<T+NS5L-{2unOn=2`G>cQhDCG7g@TF&9P-jw6^oUS1v41QJW4Gt
z`*>+`X~a+C31~21o?Z?9-X;bOA2+x2d>SDkkAfLSSei$j9kTC;^P!a%eDmH}-nDYu
z#y&#e^r#n(XQ1db%xG%D^zdwsEY1{Y*6;p3K3c7JdghLU1QW(mp337pJnkHpSh$QW
z+4!u>BWZl`zIJ9a*PI|4$YY+xdH@(7%G`X-)QffJp_fKShv`0We#weRk_T%a1Arz5
zPpP`%HZC%MF<%`Y>rhp{P~&U-B?a{_QIILK^+w6Q=%$_BexWmgFQkwa=%=7nZI&z`
z=o_VYzpEzPu;b@4U9vGEZ$@kSIaP3B1x;C2ZdssP7B+w=r!c|jw3+me7oUZ_9S)9*
zi;E{<2P5u)LI<m%!!Nf<ulWZ^fNp_`Ys`B$uDK$kJWJ4rxZH-_&L|Ku5Km%2`jJ%3
zo2Sr8m%OQH{=1xa@XAnNG{+Az{Tb(d0_c6ZiDTTZ+rNk$WmGMPr=1Lh-4V}UwKr{9
zV+qCNT`Z-uZR6!i<(kyWRxi{lQh7Yj2@vI_Rn%O!J0Ur1a$-`NFxVD8OQ+ewRKCQv
zrF^*8En9*^PXKUJr_7<`k|C8uhM~FZk?wZD`li)&E1K2Nl0jE?)H-31L`pWy9a))2
zgO;Kh%P%fD`E8OY?a>;N{>XrOS|{RU?$?4)w>@|+=<|>;7fVZBHixuZAE-{+EA1o*
zpE|j|sv|z|?Nu*sm<Cu!O}+0aUp}bk|KUrRqbAx4^CjBaDQ#v2saA3x-L_Rh;N8By
zh_uj6x><ivMrMKg`{ovwmRWffrd1~P_9@9)y9aSOH;|nSzN@M91rk|na*ZxtA&kjd
z0lAki>`pr-E@$Lbga$~+`lP{LuY;VrX9gbk9xaD?V4!fIaQ@4lWsBD_*`OK3IUgf%
zGcCTRYhdZ`p2jdX3<fGShq$Mcz~La1Cvz08AEQF|*k@vV6=^m?dY(w{ZeF>|t^VKH
zqW`wmlZxr3v{hnq{^Fc{I9aH#f>V>y5L0l#kyl=kA$H9mX{BiH5_~;QAtv`LAY$AM
z`j)PA8P_Exg$n*_8s4j6go)#KSwQ1mOM2?VPL40df;Y(LV&?2(n^qJRbotZ$z+z0J
z7hMxET`B*dYlz7Fc#l>y1F{nXHudM+n(Ku)O(`7RqK=VXxiA4_#Z@o9qhF5arca#j
z)aKaiEZH0+3Z%*i=I_3xggA+VYluXjlB+iu@rJN3p@Lk4*RVWH)Mz!_gO6c_;siZ}
zd!@{vd61xfa8qYy=K^XQ#<cRD`4$e4)Y+x8GIjR$j(x8*qTU|{j1JlJACZoSo7L$h
z)61*Y+C;MlIH;)-M~hd_NYnbk2-}-b;J)P5E}UxAFz=VM-`asoobljx_Q{LLc}fib
zFQf=3(0!fAT1H-Bu!oka6FS!dqI<k_EDWa}muHC@sj_Vb`UHixoOBH-GL$3YQBkHl
z)dr&t?9lrNXSep*Dc?%m1a4=FGbze!U|*KWiH!6OEZ+9KeYhFc--u?6htKb*44WYL
za#d~BRc*jFF?fPGFzcDfKekl8JdtUCxEDQc#G2@NSQ-LmcztpQEiP7Q(pS|~v+(mz
z`8sxnV~9gB52&*o-FlzIH9tP==9{_$63*}N*j9GjtUsod4<l(b1-l0uFOmAcHamMG
zlncXf_3mI+sHL!;-nMvv@$gqq6u8SD`&05wY!2Ym)zv#*{>H4eIfqzHAM7Rwj}{0U
zLk^sPC(>?e(}a662Prp1MmoNa(C>7KtBA5rt*u9|a%=IU03EG;Jb7n%vK!1yU?m2;
z4SfQC7b7DZga87)i+YZ!h2)kpMQEZxA1mTnqD(v2j8{-DE2HbwdJ$p}V(>c@UG?o$
z=_r7?Vm#7p1gM%~u@m*0mlIlAJb@O;>jk6C>wnd=_$CJjxQ|gh;pxBqjI*XH$GUVI
z*0cq<g}h&d;QMv6?;Tuo(c5C{bh*czaM#S8xoaC4nOpkWme^83NPM17$dkFuDJ4Q#
zL2}ijAt=J%U1{aP@(;*d;MB^bZT-?%74%(~5+{m%s_U1z$ge1XCRioX>Q7m_(5aQ2
zO2opF&FedZ)#iE6phcd*w_SnfHLVsm$k(~2-|CU#a=Y5dXYipBD^)yyp!Ksu3HMd-
zcOKb_Z(~4SSA;WJCoJnPvnyi4apG#t3GWg0gU$h-gO7wsd;n-Gmf0lu2$tXp#4&<S
zr=+bBnBH_}t1_qYNFd7?YyYsSjdM$Dm?GeBOoLbHhq+tXHh*Xp8?LL6@AG@9ww#US
z!zqcm6N*YdBfFob$+`o)b42YCXvKc7=$Yvx2q@wpt67XvDk)6Wd0YwB82Q06)2Qa?
z8J`cMcfn?C_@>;K9^ro-S5RhMwyWb=B~YBw_Tb|3f|Pt6prGRH<Z<)ja@!jg3Rj^k
zK*VDQuJCRe-7N1DGW%)6{aSbV!8fOva#W^|yU|ixN-+@Ep+Mo|DYekV&NxOz-155a
z?o--*OH;?VvfuU)ukRhhl<+&*hgMdvmBXVM$h5;H3nF0pynd9kscJIQfzc%@0XW`A
zw;Loq_0`AX0fYwqVjtJzTT@AZpH5_;63CVPLsv$2X{N;K6>-8P>M6L>YiedHII?b^
zLYc@}X}RR0)%0~EOEfLZ5_MC)kX_Rhs?`|Bk#fQhC2%e=W;}pcjHi%weD5jAKJ8aP
zE<>r(mN>$2>+z5TAsI;ds}+7{P=%!G6t34?ry$HwW>?!=nDt52QWSaTR#`NuWD0m$
zJat+v-hdY#lj(xF6>|qBGiT@cvs<PiWdd?M3W^aVaXuVNcys4Z6F*&thXPZ*4tR92
z4C}BU@CR~1s1gkzPe3&O826E+IFn&pr0cgl-@6k?VH&x)K_Y&C81mjv)54WvMZOe3
zspgFH`9vGN>o&RbNEByLyd-X0|2-AT^7Hxcs<jAS0``9kRb6=vsxro}x`jWqvxcMB
z+#ALRORRgdXsLA!OQ|DK2q5SOr5(2@lBy(W2wsQnXfXAJ(+qXToY&bFrIceqs*`02
z6RK8U60!y$b=L7rh4AzWn+SM_&#mWoOqYtK<P5M(@o>x-*XEP$pPa-4ztVl(=WCr@
zw!PPZn0UD5+^MLt6_k`zDi@AU&s1wNerU=e6_d(^!%j5znT(&l$iA#i>v(7EH_^|h
zQLX=O3GXj)CRI2>BAa6MmYP3fSz0FZ-4<rK-+iZSy><qX+9|P}&OvZokeV@&1)z4;
z>Hk~(ZNndIe)n4w^0<r|%NIe6BH%1VhVg$2`3RNYL$WhF5Ci!JbES#BW0eJ8k>F%@
z9z-f+mc(O{R_A_$0&tYw{^_w2Si7HDxk(sWsk3}e1=JJPU=6kPzo-=nqL3q)XclAm
z?#pt$(9OMaic#0sBpUzG?Xbc9MxS&OAw!zzlWM}2E?Oz?Z<<_tX1>;l%dJ&2Q>Bq9
z;1N1EG{x4;G8HDDk;%(8^ocqiOdR;C^`U3-4tWYfS1B%gl*Nlf?+l3RTxXKauJ7=3
zo#$|Lf{AUy0UY1@`aM<}?E#2ip{?y`(%5EOB0vZj(3lAYMm<r;vs(Rau|T2t_Xo@R
z6nU?#Ug*yW_?TkOu&)m3y490AJE}0U3bmWkwj(3zL$qljs&g-$9d|1ylDKiII?X@K
zi#4{J+iGFUyZcV&aP%IM_%<T$g{X-A;&xvgV!`}m?s)PfpON7~+g0QZ@jca8i(Pk6
zmH|5vnQUn6XIMeQ$U_@(l3S#oB)9k}guGM9{qr^G1%!=T7JKt;wloS?2%tY|iUEF&
z)O9L4K#LN9D|Ww3DlR@O)*6ymhbkxo0qvb5hTRhjGyqB))@56E><5ikwz*Q9LjJFG
zEBNrMC$<zi;q_4D3)@TW$Z)6{aRV0)Y-{UVkZ@YkECXFa@`(!L0*iNWQZhsIWh0rq
zI*^k>N5cG>&&Cp$-50M;J42l=Y|ZtKL;dTQ8vV|G5~AsIx9`BO)bY+c{j8xZ`W=@<
zvcO;!eIvu|0c;}+=PRBdVm4ORsEg`fU0s9Mqihha@|)A}#yOL0azP1>0c!E&B-)*I
z-X?s<zX#?W5Zx+aTvFgOWvwLeN}<yfENoLEoY{f@%9|LyhPcegeay`4+AEKmM@i#t
z#;albwclF*hU6n3lFUwM&m9%jkIR8dgD2`qCvbHJWFG~bsDIv^<>V~fClHrb>wOYz
zCcjUp(fbZ0Gtv4e<Q?NsI|Tw^nNm2KDAr@jrY2*0szRd!dU!>*J~J>*)Hk_)-TyiN
zm-Cm|cl5BTa5XGO&hilSjee+&*MzSG{1LbK{UB3`_(pUyQ3b>c4pg<^d73jG_Z_4h
zy;UH;Zyv+g<7v5Y-)=%+Y*Q2Gh8Ng<wDf~Ae51y^jX6R3yxHmL`F9MY2!E(4Y4|8~
z$%LF#1X2Pq8aq=ba@n;{XN{_~cOg~Iz|CCe0Lv03U8ss9y=YBUQm0<BFLqu2VHKsg
zN|Ty(Itdz|yp_)N3!mBK=q>wj7V)>l6wS+&esZeer%ntP!$F~#xbB!psZ^xsN&DNx
zZyVwpVJU}7|Gb<_U?|vVAIZ?tou^YD=cIzDe~mr{zZ+5~7^qNL_Z?s>DHr0t4`g}A
z=kfL#()XrhlUzc^x!HV0lUz0JvK$brDEbeD)FJY?L>tH97rQu@BNN4N(6?VagXz7!
zm5ftYRi$PAza}{QPjP`?;mUt}%TeIK>v%n&m+_}jc?7jzOrN=5`6WP_QsrXSD0W^s
zzFraWJhsy_G_PdW@+p^>Vb&`5JsqXWg{O6>wAg-h$LAN5s{ggw&sM#~h`{$=>^j#u
zZ2$!6iy48X*{khRpj)y1eR32pBj7B@jyct9Z^YHBP$EN64M|Q$0@b(yDdliaHab;k
zv!hKq@QUEq9xHGCFpKN>1fxoF_GI1EYl6V2Qu&yp|FL|>)ym1~d3>l<b)Tzb8U5dZ
zFkm=t+D|F9=Se6ihJylutTJ402j0LpiVWs8GDYz_kJWJ{kk>^gWNoe+#u@F;L!yGW
zW91a?gJBxDcm^7=ZDrf&@LT^+-k}mSUwFzav3Z?ekmzj}(J%)pQts;-onozU(4K@2
z&v_@tZ;;BB^2{JkOB(m_iG0)JVWLjM1deYq@M^J<1Q4vDu6zjwIq6E01#k{_0!-`%
zDKzJhLc`D<Kidm3Ny34}3QFDj!gx1JN%hzAN<bQq`iH);v?iMvZKEa}APPo26@!G*
z5C{GC*^1Ix2YGV4>Z2Z<9>NKn9H|a!BA$34vx#GSi0EY>qT)}UJcAA6F$@q>`N087
zmDydJL0cbA(xKPj?DA^ay=FecM3&TaV)DG!8UZpQJuW~dwvR?sT$bpD^K0Knlu3Fb
zAopG$R+~2lDRkF$T^8w>YjO|jl#CyvlW$H-6=Eol?eXtBul+Db=htg~tJ+vCn}QBD
zrn&x*$hanVx`42-0iYMp2iqjEL<@NQ)yk5Qt~3QD3~D48|CL-O{@i~Qpcj$It9xBy
zDIJU0M;tKgfhYfgMbTi0@NHm$$L6QCm1Pkb5f_yw_Q5TekgurFAHCcAAW5mTo(aJX
z!?7j(&3(X>-`*cKxdJJu!O!qf?FLMajmuSx*TV^Z!GYkI?*6?!0ZfbqIhnuc@tdw#
zn4oD@speIS=5@*usTKXfBB@sgP<DvA=1&@PB5kRw0F-aGu7-YBn7^BwE7DIw3n}B-
zo<fAQYK_coY>3Vx7=)hSf2}}GB7{yjKgUC1=%jpV3WVi}I(+nkaO``)qifH9itU;`
zek|G$=<q>y=80hlx>-&FdzOH8YE(*JT5qkC7%s=D1NQvWRW&7IOGc1oR=XNPKzOz>
z@GGy@!zf*lu9y3tuD8v90z+CgWb?gF?0Yg$A1G3h+F_I_)Sh(QFv2F+MAv(t7?hw^
zQ29{_KDMDmC;Y|_1IiM?00>!Q6-F~Z8V0ecCh3W%Qy~>W_QIsF40|8QWJxj~;x~uu
z<4d`D!0gGX)oAeep2GDmnyCg`B5`rL6@>~k^>dZi5UA2=6PDfNiX)1SXld9DrhNUF
zpp>3fyQk5KlR`O9RR;g^VG4YiDy1C*S&!MRT;LyC7dgpzef~w^5e`R`HWA3{oji8Y
zHEC#zC_JvC4K%ZG!8?oww|ZScVKeKO5t^51*8S=lsC^mQB=M9ql413E>r}{J-*&&;
zMv{tyyp4ngUlPy`hJ^2tz{<k1dw<MNWWGoR*F{|8@kAseq?xVx?Av>6v?0qt9nN&&
zf^)iGzDx>xq*4pLelVewAjhGi)ybZ9X!2^#V0(Pvt<m?1n85di6FV0SB--aqpUm4|
zA6w*c{TuLe0j9|YzFLrsn~d!|aU8C?eh^EmbTuD|k)fr_b(0{`G8})gwxOlv`v`e<
z6CO6RL2nC?<cA95x2)_eDY=LN(^3(>s1)==ist5WJpk^_^FSP*l9f9j4mq#q8v__D
zkMtzitA<-G2>f(rT~=P+>T*VA^!C@W&^>I9Tg&@#!4Xi!%i&4v0eKM!$E#6>vNYdn
z;D&yv#stOsR4QzjT#7F8OkPa#bd#%TfuKUgmOAQY{uCz3R90}9{AL=tUuhBO(@FdV
z7a<Tm`~f#KQe`)d!|=#NvT!w|o<c_q2S$cs7K}sXA1j-rROn<jx%HU+2#8T&oWLZq
z`^S4QGiynAI$PAAQRCoTnzFs(|69+X>e@-SIOIdETA=`ENV!-ji`{Z&0B}17t7#iG
z<-n`=YWH3NOpZwOn>dL|oFUn!bW+1VSWQLDT5@Kbl*U^~mXG9z#-*BQ4WLNAw&yhk
zPQ6r<$@yhZ(C&squU#<MI3cKA2`7Ez`eRld1t_VMwlzcy8so{?#Zo3<*zx+=_CK{i
z_`ovn(Zmf9>}HwEp$sn=xCbndHkU>Ok(L&?+J-e{7!B=j@ds%nBFd<FoebETnGDb6
z`-Dcz%ybR0;Qj)!uVJXKanEdpS#B0-FoUN*B@~JcC8-Y$o5L$F@_0aiWe<ZHtxnry
zO!7%Z!|^(Q_@YK!LTT1eOE8uMm``)0iaO5km&i#8)nu}7@sNj}?x@=N+t&i8K0)*C
zZMag?W(WX31c{w16G!*9$&w&bDS-ucXG`yg!32=c!(F6Fx0v7w@=EeH^m(KlU0@6E
zeHFD;tk&Fh&3{y7OPRBpbiUquCHJL|Vh!g}@_LTLrpf42l~N+$7(p~ink{rs;=@!+
zHpj~HyNo*gx2KRUnbg7sjm-4K8|r;eOK+{<D?zi)!?u`KGtHHRnw3*aA%o#1(|TwV
z%acpJ*!3n-f>X%*7?%871KpC)JG<yx<X6edm9?!C)NcL0MRXVW9}r23ZC}N8u*aX8
z=wMl1&(0I6y_)NlBb=cK^NW@Ar};IClbpqwUVu~>bO`cc6Wx5Rf5Oxn|CsghQSG@{
z;@l6=4-LB1G4sDoFumYtS9Ta>xKN_bV_7FPRJn<Rzj&o{gh1pG*FJD<$7ppYRt3aW
z#p+AXjW{i3_adH5_|)}(&4J1PJz}}}6C_PsVhGVFwhCA`b>*2vki*x!wmE-+P~(g@
zjfDtr-Kf=Oe-e=jzVX-`GP?O@jE$`dtg`F>ZN}-khN-mm^?Op!7x$@iGL1TJl?-;t
zg`6OY?rZw@^!>ETP!sr44z(3k>u<4gwdcm@O*|6ZE`OGU4Uv5|9gKpAQ6=F)Iu7f)
z#hTU5cjNwET{}Gy=hC!M^`>qcrX?uNC_qBaQ{L3ne-I4g6H}z$Mce!8=K7GomQj~^
zy%ZA<-HWxYBZr6*+cK284EOg#`<`(P;_&4n2cXuqw0sbOSdNHkdh3SqbbOCGv*RRU
zO5N51%C(5a9^E0jttD_1tzs-q0zby}nKLi~1L_PZXnZ>8k1Z%k4nvu(3q>dX_zT*s
zXaTFXAJVc@<R`ysnuVW@PkIksQ^qVJF>Gp)WD_ei#VS@6&fZF(6Nu`V7c|NU7eS5L
z<Ps8q_jTo?QHy8r{;gokA&pi3ugg|v`>f7Kb%EQ%ced`=7cJR3V9or}bDU9W^Sac0
zj)0pRRCl|igNa7~o-l?+rTbH-knS@M;4(BJ`RBCw^=B~|ITIN!1so}H8xzbV<oXZ;
z2|zUDsd?7CNzRR&IIuS<^tWcIk9d#sABh47+)nIzXzP-axR2{FAIZEm?K!Jk7rqgf
zb86ToY9*NXn?zMD89#9{kv4yX{XPZgmFnb?bjKYCVVnAEDrT+Nr8P4f?RPaKpr%)9
z;02FL#Ob;w+cZsTbzSSdM7%#Z$|jSx^Pd%EC}6I5k-ur0&D}*{;uz!)1jUC_NnMjS
zuj|~)8}yk#R!t#`A;c5&kL>j$B?vr`q^N%Z(#ttIY<3t#7AYJs^y=&lhp4M$E<hyV
z0Z(1umH>q&O-h=Ktoyzs$y6qWGATTOnk?Nuge9F@V3^Z~D3GcK@drE&BKTI@-(->h
zn7JGXm|E)s4xEuSK8Id})co(;&Cww-(1T4PmzXJ_i<69s%__ykiz$b*CE}9PZVgB)
z2B(zt;h1~@Y5)BCjm@A0VD8{NXc|_s1|*R;{RM$L0II5kHJ3QVd4lFYdXJDK6J(X&
zaB0#koV3*W6qq`s5Gb%^V+B{(z#lUh4v2G|cujzBEYo>hH+enO`xf(_=`3o*_Ll^P
z5|n2+kTd${6PQ$(yq5%e)uY$;Zi-_^S`j+8zgOsD9wJ$zqos<KBx=xc@^vaEzkNR_
zDXyW=YpzkR0Ge6aW-P>2qg|TbTuGz6Kzi7JZ|w6SecHGtULB$!7C2S2@G8h2R+ENU
zFUN=lPi9iUCpB{5AAIvfV2T;kk8c&Z0&?Lca{|%)@6Nw}Zi>m!e6uxs<g%O;{Qebf
z4iTK%%V+x!!_@cXsrH@U!;@=y_4vDbDijJMX0Tt$qj?OxB62y@cB0otp23>Yv%l9U
z%F@!xz#cFAoVvjEG=PC=@-k_Nh)B@|E1*X0dkO+jL_4GoMj_q_>(Hg-Qx#zPVlFdt
zYcFd~MjZ){gi_7oyM*mGBxe>pkHlyn`{=Moz=J5f<!2otSKEEQ&4B&S*Nx&^zE-6y
zT=b(2H4n|3v|{b={K7|gLyzmYhn{Zcz}N9Z5~RyeNdyEFSyC$nrOU0?1eLq8M9q3{
z+mLf5{1h{|YB)<Ok{i@nRXW&*Qd9$E)u4oL=ur88_au`3enYTD5|gc>Rd7^AsQrIT
zy#-fY+t#$30Kr0VcUic*LvVK|1cGaDcXxMLxCfUNAV6?;2^QSlUGLoQKIeS*7tk7W
zwK4jrUiFkJnG<r2YN1~F%+hMf{F0<B&7a|90?6hNjT242J(GF-^%PTKvN(sJh{p}0
z0f5w#vVQnlLTz<zSCwdMtETuUp9;gy?t+btW8a^z;lKAmdn<vGOj^QhT7|Yd-t<7k
zKSj>;QlLjWYBF(B3(0lPfyuN+qf83t=!N#0q%msaPWOLukk)Vyp4()4M@RIKjDlq9
zr$x^@wwILe93j`RikPM%TG#ExP=}?lGwIoVJYu1uxxUHk%TgYXPPsiFMTy(q@{I;X
z_T(oNS*I+{j~fhjLU4YIpbuQP`q)F6f1=_ZRN1YZJaU0EQAF&-njd?e|D81wxc6}V
z2wgAGbH_pAbL<eyNVF3s^hEpI*%NER7fgq=Uz`Z9f`o>NCmO_wH&>*cL|&B-X%a)1
zxc?n??GvmjLF?I}&Wnz)nyZH;s!XSG-qtNOZ@93az4OPeqw`kN&}rv^ShfN>tKkKS
zf-rIBZ>iD$??a9XUdZeUUGhv{DI;S-MQj~sNVEx1ow&)|0xe8CY!JtY+kWI>o@W+P
zBr(_d`Nd}Ekgi>9DupNmKJr#YFHnoITWJhtNc~Ua<l8j#d@OxjRC5GF<G_OligcQS
zjs}@j@%>42BS+XOpJ4l<p?DAx8}6+%|JUohk-#V##*U%&s<eeg?Va7fHB+^$A2B%1
z%*<%gfWBDZ`7_S@6L@hwjakdf&mA~_abJQ~(T4DF)2U#d(ICO10d^bxil8(x9rty@
zp2~C2r!N1e@%ew=RmP)dzN7i}h6~Kb#MjfjsN<+LcD0u$cgp3#Dw=SSyB<e`{k*}Y
zyq8l_$~*Yu*hyKe@8f(Puht*nsd@cQ-%m4`ve<0Ep%*g(0S29xzW#o$-z4a%-U2nv
zOwmK{14?0jW4j|ZI=frFS)*6Yibz_oA^G_6?u8eGGmK2CUar5OT<r(NT>^!|k>%ww
z^7<y7d8C|8NhL3Fe@zv@Lk?VBW@uRi^dK3wM_!j&^QI)ZVoVWK(&8d@x<Ux7`Qj2C
z{L1be-Hck2Iz3i$a-nfN+<>pWjcNfzGG_dyhty?@r-Og%Ikp;!TX0M;_st>jq0b0k
z7SWqB{m1t9BDUnu1Doo+uZj@8K%U~(UtMz$Ab?RF-w3QRkzf9}6l7yBl6H$DY{TYr
z6kCNDCMWa`?aRm*W=jl;9#1l-y%S5Nyf3C>zml4qL$f<cL<tKPiSTrSDh5sSCD%!B
zAf^iA*3KM}KN5A%5o;u1JO+A^@C-aB6qd=xl$YbtLFiPwPj|7cM+oG(sMFH}Xim{!
zuqx}ebz<wDl$7h;p?lC!V8!_vFJ0q)T(h=-q+UN8yz5x|ue=}Uk#Nsec!dQ=f*eMK
z+;TI0)HwfhNRk{f9Z?XV-`t(}QO>fUYxLntJ5ryu9yd)w8)J)Js#vdNDxFHk$d<ui
zFJ|6Aie{j0i(0z{GJNYAMKjFOPX-Ysmuk0N_rh1g5p>OA5`t0!4ChKBLu#|KWci69
zqeERfSZ{ZYyRTj(+fKQUhC)vnN)Frm96C9PS>Ce^nhPwv)=@cv+=n=rVLGhVQN+Ai
z)bLpeeb7I-G=~~0N!glA=PDuF{omvhsuD#v;DnF|`sZMoJ1!Yh(`wlG*(D&FcnO>?
zEaw6Ivv3TcFOku?Brq_B2s<7mxL5xX(fC{tXQwE@uT#wDj~x#ZCTq&{52K882-I@o
zK|<NtM7z6Y@v9brG!uhfVzt<mK1DKsvAXCCrwjca*ZjOV?i`a~C0dr*$hANfm*xw<
z?3En&EWd|sda+=h;@o1CIl=Y>%hyRE09i~_hkE}e0w`bV+(ytzME%4L#C$EwyILHx
z=|4(&$RVo3IO&U0eHaJk`{c5}#VSk=6FyNkcdiN3QXZKcGPV7fvPXr)U$by9bY*EB
zv(P0F?G~tGe_8BMB=;v=1|{aSC;h)k2nfsb<CFZMxL6lOaC_?hpt<@>8#jqQLYyL=
z2C=FcQ|y$mD*<n~!HI%8Sd<$%E6F@m?QVHwTrnDx_{*0s%esq3UT1FU!k%E=!lBIs
ziO?wxz7zO&RfLoiJu`^KFgu^OXhq5im6^hT;3rikhP<KzE4N^mcwuO5>`lz4>Z;@O
z(;e%|%9HqASS1hd>y3ejgo#&{+?P>mURN7H24&r?Z9D+?>N=HaP@9NU%Ez*L<fECt
zqmK_S5T_%H(>=EJkRN5j?mf+GQd(!g&5C3DjQG*!<oo4t#oA_zop;EU`I%w;YTjRK
zqGazX)@Bty5=BDyE?Ey6WQIzyr$x9>?|+A0ymIJDOa`F`80&dvmgng!LO~%ppPxV{
z$L0#QEN!<ZQBcnez?My}L4bsPHoi*su9ccy5<#Hhq^{N6O75i1RXm+hAapwO5@nOt
zXj9o|={3{2pi0JrLHBb-ywyc=bEQ>KETA_X#z<;>B8rBt66{9p_Qd#+)b`x3Q1XH}
zwhNQx-o|Egg(*mt#p%p|>TWXUa5zfd(uh<L?)e=UHxfX!(}z4^@laqfIf<GTpok1=
z2Sc3kr(L^jo;eECRqKuSGilBLBu+6E&qUeSZ8ph7e6R1CxqdNi1UQ<i(Wl-#@z`60
z{W8jv0vbr~adk{uJINBjf_Lv)|7J;Rbn%7Hq*rj;0xj~dQxk7c!cPDRuH^y2)ve<C
z3C-hlFj^vB<pR3Iit@D=M`*}g=myj2iXGUfS5d|qFP&glaNGHh#H%_yTf}V1DF|J5
zlE0mD>u(>*W+@3>Sqj}9knAQ+2mrx8qN7V_r|XxYRUbB27E?YxC3bQh7IMEpQGbtI
z8!}wvrTZ7eKe|^x$v$zDn<GzLp{mC3A!L22cpkTV^;$alS|W4{vou3Pi8A~{+HIX2
z4d%UU`4mT$I6_2Hh*;R(N<QA~*?)JpU-8Cyql5EUhw7C45^tio;Y+C8&Bav4jAb2S
z4JydH<85%K!0DgE3|Z^88M;l1G9>QALR$&My;?~@KQyVYV000GQ06iy_#b|coDZWo
z?559D457WQ8T#0;Wp9P9a>#ysL*w<nST$wnK*zh7=HdM~yvdNcOU=mc9C6ae84fEA
zs@U9}>?VW4rr&+J=g|0{_U~Hni}C+NAk6)2Li>eU#6cxBWE8Cv`NCn7zp+g#k%_nY
z8sC<q3JG1Yrnk``l*z;#Fc@9(Tfp@-_|H|kVLQ;KT|mm9i?jN}a`>?6_t@%~MzESc
zZ~P~RZo?3=ChVZkN}3iqf)SDdtUNnpdMb&@x~zii6XopYc(=xxYDs`7H{1QiPJeUH
z_jK>!dCxC+CHw2opSsP4`bhfPa%x?E7Y)K)xi1I8B897YGvDhVEnOiW(-ySX@>Z-8
zB+;ZriuDWCX)y-<BbnXr^PL1Pu61%e4rPRHr=)1oprgl7Bu;G!MAv`%wiQsHaY6hX
z78<m|G1i0}D$XHdh6xnhuzQgk2@DelcTal1y)!B{`397-ce}mpE|=ui00a2hfL2Il
zgmRyAWit8+PR<X}Na&*mpGE9)f#7SFxlJG3oq6DVPI2&+DpHJ=|6Y7uCO_u%Jfl>v
zU9q*qIH%df!^PRtE1J~Y8ifjr_m~G-ukFX8ke^0NrN2LfUs+KoxJ9O9U@EJvgG3F)
zGpI<@C&&iP`S+67bqB^vBWU6VUz2`F!-@Gdj8d+S>5;k#IDrY&Krce1tonwA^}iis
zoY;jVAMVT76U>GBrF^#=fQsrD^qP(<^gFfu871D^`_e_CO4iy`+49J}G!<1WJ_${K
zy6GcinO#Cr6Zk^hMvT$Qo}~9<2UnGPI+QX|W(K@g&5T@EKDjijuf6;DE$ua?em?de
zE`&$_cMCuKUQSIxg;z-3ac0&tdI{mK&FE{{l&t>B>TzffUolvxJ;CyF&WnE~g&Nsy
z6jeYi&+{2CSRLGQ$8gm2llPZ>;9`W1=1*I);?2!I0?>h-!&*y&{2ce<*e#2sQDvd-
zmHRzFdwq`Is2$jS#LsHt;^l)3zEGd_%mgyXXn!%-8IGxN22c-a7^(EjZo427pIs%#
z$7>Y<gt?4@#zUJ8H!F#RV!WGiEW8#YD4wR>yyNl^{`CMeXZ<H`7cv_(`hsTHsJ6H+
z1SvDspRPY2cDW5qe2kMyb37tbM$$-S5Cmgp5rf5fLY%@fREZw7X$P;60tD@IQ48*Q
zQE6VfWLExFBe&$U^Q0mCRlJT9IwcO@!qiKw7#QT0ZIkUC;AlmookIF2EYz3Cx^}dU
zKhs*xs5v(^N&h9H;pdvWJ1>&T+0iyv$qBm_07KcA=@|!;VXcG3H$Gc~>v6L-qJe(1
zOOB9b)k`tGcz|=IY0Y;+1dihMb&S~0aYDK?g4$@_dmZ5UX&>x&x%O=Cyw*0{LM_yj
z;w2Oy_bFx)ohNGisjL}g>Ky^ZaDf~Loh&k<%+TiGj%*Tf$2!`+7hnh4=}GZKG3OU?
zF5Q)J9Rhcw3T;|MA(<sQ&|G&QfmWrMCa`uFLO=c#=lEYa8hZCwOWlEZVNB3g@Ci!B
zF5D$-N4tet772c9u&A(+V<G{L1Vr-(k<yo891px<pQ#=|;>kdiVQy?1TPpO5;JXvm
zA(}`{N;on?U*sT5Sb8s_ecZ$uOc^sB$b(-C@HA^yb^^ReOn&i3&AnHdrHC=kjG-O=
z!v~mvOaGoFlpyTXjxO6kg8>ryT#Pskp|;v_tFlp1fL0^wFQbKPA5(QQJ+gAP$3)hq
zwO>+_MxwDLns)RZd|@Oo{-k4UKe@RVSC$I!u=!ji%Af2DG@_~4gOe?kw(KqTly&3u
zP<q|;*hw(|8M|T}d<TjYhhFFnRB8W|nIS-u?())7&rL;_b-PSicT~fbt!8mOc&KWM
zdfzck)N++hr{@7mk<c-=E|-jCl|F27<n5>8PT6{IX=3_%Y4*nVGEB-K_x$`yvUmxo
zWk4jLr5jf<jhV}&iIGVPUs?HXb%VGg858kzwwIl@Rt#whLJhpUCLbVqz{RoCH#4mn
z^IH4eRowIwf$ozmXD-RhP+Y=cI6XbRyv8^iB#*Z~kN4*j69^}QT*52Qf{VkDm$Z*f
z+XGH8{9(T*?)-u}cO6JvZ;viEjE+wLA@d8LIF1OJ&2&(NrQ4GC6U6QcQsa|t*M3gU
zZC9O8WI2^}@zUk@uL^BE`mtb(uscxBn5*Lfs@-~I`=`Y?^Xye$*&aM$pq_t~JE7O6
z!?2vxlt_!cQ75^AuX4_?l%Qb<Jir%dH^;q(J$CuxeUdeiSj6gK229_uEW_~ulJOwQ
z<m-jiW#DjkFnPtjkqr_UId^_n9WUs>Q&Lhgs36%oQ$licfb{76H|yS4S$i(3dt{C`
zGJ0P6H<ZJEro;L|tuQbDuPKs5{`Z)a(?+#0!O4BUqvwRkI11Oyt3-GfRu)B?k*Z*<
z406PYtR{GAHp6=YqI~MzNGok5JDfl+mt8C*fO>(u=cafl8QMPWKTgO8P)E5k7%bV)
z#Q1t~3Y4MlB1;q<<!l+o@xipSWyP%94uR(aZI04HgvyMTuCRicoFufE?t!Bw=^}B8
zeyK-a*q5W_OduXyf5m>h-!JPcvT}OUi))dTl+o$d#hZz;pTjoZRGs{O77r3TywMph
z=A%CK@$cVOp~A+;%_4usM-hJENc(;NW~yt(vTXXz3uX+V62+HOyg1OBo6FslU$Q`3
z`^P2DDP&=G>X6<uqJYdNbjh6mt~TlFxkX7pVQVx9FnkR|*#1Z+sTRxfOfy^tu|Xu!
zYUOZG{np7|XN2!?7MGTGTlR9c3|;_C>hjgZ3;|M|AI#bPe+>=(Ewr*$<JwDUH4PWT
zS@6~@im|Z+<PL?2YAS&vf&<+_d+9#szT#B@$j?3c6q2dB<3>(l80g88R$bfbDOY#d
zViqH#qYMUquN5{nRmA&#SVbR@&^I`!!f1$VSvo#L^dtt!lh~t?g39DL_SwTjaT&(U
zVjNq<D<EMmp<@twc|~SJREqaoQ%kvbxG6)oA1&c@%BJq_*qz~YU7C)1bZQIPGBNvd
zFgxr(5uzf~(RJG`hUUgm<&G*{4BOv8s>gB}n8II>M4P*<Qm4c#8E>Z+dv>2u=+(FZ
zLlpD)nR~D#e8i9wm_3hB)hnbADUr1D8y|Ao4$|MRz4o8Wr+OcV>D-EXFy$|tjZ;J)
ziHGbz@52?5m*B&5RTb`Hxws`!7rmf!b1PkWKH*MnqNNIP#X%yoN9Hs5>%@!k6WziD
zo(|=}2WRJ<*BXa0=;Sn*`6;S2WVlF=H06`BTb3k5$?5!DA=hltG8crS%TBT193v--
z!F)c6ka4cJ2ZI-O4k|jE%I_N*bk7S)-jP`&S=%ieRoqNd?1<cisOFKHI<*_<`YpdB
zvjk)YJbDe@*e)MxRG8+q(+!MlZ6g*JKcuFnrl*5HTlRH#qhpMfci$-m-WnPnzze^y
z3-{psKKID+yVZH!>v%(PKh0Xh|8uqBH{m$;AF;%Cw6z^swA3W)uinEY_uwzqoWyPa
zzJLac&t2iuI&=eOysrnP2$>kT<@VaqV=8k!vT$?jm)-b?CL-#=P}z0P-l2Jwma|8M
z!j8c}q{#kZiIS_SY5(r2$OBOk)l(?QFmCbS!y$y~_oZ~=`y!|fLMqpA>CVZi!=`ct
z^^Dvvq~{|=I>TeucI<xJLYXrk;Th&)Z~wHQ-6c2w?M+?TKXL>~(o3imFbu~a-^#4+
z_hh~pM!q!D`ZhUH93JWAhj15!<l~!?X(a4St5SpzpE$CzS}Ul;I{zef?8kiZo#S)d
zsx7QM{6CxJUElvLoQrL(ouWq^5b4wuV9BY$bQGJ|eDExWpVtXxTE%KeHpUb-7<x~b
zpax^!6Y3&1;>3%y=8)ul?`NqqDJNw^<HBBE25&;HVc2s$qGN85*X|GboeOczH<nTa
z0~0qAO2=8A%1alUy+2MlE*~+eX2IL*Oik{CDdgC1DU|3w4MRhp%rX7Y9YD!UKAK@5
z<{$sIg~*A*^@Ygy!r+-N_o+^bR*OT$hR4-2R!Z2F<U})<d-d1&Y3J0@ZA%F{ldf24
zX+(|NLHh%uPUe>J{k|H>Ud@}q8@BmrORg*=QWKGkRFV8O_pQfI%DQIk;p*;cbS4)_
z33^@lkh~+~{PPGS{9GtJTo@<V-|?LG7Wwu-@+`*H^z+S);1}M1jFxc^QEQ%AaN|n$
zJM;$T_TRC-gA_$=KPh5kE2+vfLW{&&B=g)HD6Nz5!^)D`#M+mTI%;C>C?pwQHPWo{
zMI^&!c}VXm%d)U@i#U~Z4Y5<Vysd^UEUtcYv&+(KRj65`7-GYw##9@lrQ5+5QSRPB
z&cdIrF6Ltw)0oEvd3`gg_Z6*#F;Zk}ZT?)^pkO|^b=b;$)m!R^Nbg;{>l1{mjLPix
zrKEZ=ln7nW;qyY|!M6=tr<{9a|64fnC^kxpk=lZ*aV^+20yN_mW*WVPK-NcY&Eq=*
zBE@;c&V_>L@84CZey^<EqO|_&za@#B#rxvEQ7$0SmuQ8`VsWCHgrJKMn#16|Uhp12
zY-OP@H-j7lJ<Ajx5$h5(e&PKgKdyfh@1wr7F!zFm^tAr=(cdVfIQ4y-mAsXu-LTi~
z!eB~~((L+n0(LjYh>%4-2fx7*u=&GS9EObxFF~bBMe@*B*GXoTXrmx3d@?LLyxND=
z5r83{Z;R?1vD>Y3Tiq_ZcwPPnYlPeOyyK+;J5JCyl>H98zS*b2^wm8;?yQ=;;rs84
z6z;`^{R>TST>_qWR(`}(y?ebi1H&cv>(biBRG3gKqECDuMu|&E;KPnw+QGG}w!TMN
zjqshsso)z>;)U^sfFf{?tIvY|a4l+9Qu-kLs`W;Bx9Z01_jLCB72rTWZhZal-zy{L
z!S3zw?{fNS!Kx45gm$#D#DTN8Tje_g3z%32e+(so&SmJkE%l69Q)!J+4kM?*E(Cij
z{{8u-w65AY|3wnWpo%Cv=NI$SL8P2!5^Kjq5SapoeQMXZX=ZUc3(bbdd&|`6gWGdC
z&TErT1|jZhoTI<KH8jYY+l7^P5jJ2blu(y0EbEsxK;%=A5RTXv1o}+zE?q_UA{}W7
ztwzX}%W_YDmiPS~ToH-OYCa*TzQ5XV(Z2p0AjU)U0o?$nZ`J1uYgPMH?x#6kP?Jpg
zI||O0P0ycUK7J(q|5<U&shK;)fBPEBf)Fsq67=4Jh8|8u5?l;#Jd|L;{P`&nL}FBc
z*@4fpK=5rE^TY#QDUpgbXOAohol=7eo*cc{f+yFFkNr4VF8R=>UY?EHFF1Qe5IdF{
z>~+t95*pPjGq}@a=U`%(DhcUr@<OZk0uSBN0k$n3Ia|l*I)73Hr%=UpewnpxX9F+j
zi~Bu91yaiN^z1FOWR;W3^NRsu3Wr$Ivh&l{Glzpi1@P20zfYqy4TSk}_mZ-jDwQdu
z91?64=bSJ7dTlNgbL+F}e)8AGrn>%DUH8&1-HKw^%F0*$=ea0<G9>69G3a0ZX0_cu
z(t96Yls>Qvzchda98gpdkIPWDW|Rbngn>do;BXo1=*zNecUnNZ^zkhE$F}m17dy_w
zEB#`Odp;u$i`qv@23HkP3;sJ1Y*?JwkL2A93_`z61d~G9*q&YFi93gGGCiS0*aS?l
z$ObALNR}te{wMCBJu?064n0(`Cd}_{l%Q_;mA+nnSW1CpWfRNM{;@1Q_m}VEg6F8N
z=S=tt$w-SUD+LWJo7rw_WNxSy5y{B398@Hemfuxv<u@9ktfMN)Z8<nLI<YCG_YfmX
ziU`pJJ~|pCixOTpWBoJ=Ijh2Y%62Kd$1?=cJezzxz>wP45g@Cb#<E2c-SWV%!tA%I
zoPlG*wG~IbGw4}Jh*>O4{%+^jCA|Hg--Q6p`w-D6`A-4y+jAR_K7l`!5>bZWe7y4~
zQ&0&8OfLNM^eJW8y}x#Ulnv>V66jUKuflQ=V5l~l7@#QLL@3Y?>Dsm=_PZ4Np_U}h
zPYZyykv79K4rbb3`}x6s>+gwQ7HG$FzHDdDSH~}a<9sfWaQw#V<}NL{9%t!<`n8vT
zQ$|DJm4aq@jHPuMO*nd+3BO}@jiP%Aik&<jao`Kq_d{XXr78(FhdU(Z!!Il^Ze-g%
zQ_b7yuZx--@P`o`AkZ2lnepmmLdP--KDn&AKe_N_f2|k5TinB>>UJPZ+KD&C8+S>t
zR<=vd03X)2whm?4{qD|vRrAYXeqP8e(xWwg@#i?W>U=>te|v55>-4)Zcyn3xxN8^Y
z_|H&DVvHQ3s4W^{y6dn=d^d>vJ_M4|MMeYz)pkHtDtv{sF)c7izO`#FU5s2WU5{GI
zpT53KO($Vlf3C&IBgg6yNmD`{v!2&6h9GhZYN^|{l$^C{_{02IM(lG*0W2o|$P$4&
zPeH-uCA?aD#uy}BGeZ1XfA1+#C>i=edM)RO+Fi%f2jq8@k^>vJH`m8BU$pcdK1?;>
ze#U$@OwOnGT)L&C1JjE~tq2h;rzzwQPSX~n!gIvdGpiIjC9CEqGc8j|zTv6eFB3nn
zs+d)?fQ6I$crc}3zx@+AGY7xrw@Fqy-RqDDFNto`<~XDmg{niP?a&zDPzb5U_anSI
zf1)(qC@tj>eQKVKFJV=c6__ZTx5Y3%mk_042x{!z;mE(~9$^Tj=3&ovXyY5G@qIu{
zTeGI<c|YNI@<rfzkZT}TOQtY6=1=-Vxut7yv7pT~$uj7?bDNh7j*(1`Ts^N&l$<R{
z9^Ix!E>prng*+W>%Hz*+lBDO8dTT36owlu+Q09M9o>WN{S_v6W;VF>HSJ;;P?fP>6
zlH)xx(gYMLs_t_PPbK{g4<=C4KGya+=laY+Slh&JVpKnBk9)!VnwGn_1hIbJ>A2R)
z$jBOS3v3i#&wnCW_j@FHm3oZ=IIAs?pO0PDf&Xg3Z*~34kp<gUFH}u7;hd2a1yNY*
z!4#<IDw>3E+XLmf9dH{S+FF?AD1>;oUhR^UBVwvr47_v_r)d*K3^00i1+`h5Ts}PJ
zgO6Xr8<k*@>XYaCv3IOecBk-lX7`qs=*&P^^>uZUIi<7Kre-DPWq+*QeLK~iB3Ep4
zv$R%=AwS25uHqbaH{uMz{G-^qU&5bHNlFvp7O+wcn|S$2C4c#ZFhoK^G%_K#dCp(t
ze_(s%CyPo4P=8dV{@!JKSJfq;gadmws3N?!adc{afGpP@#$`qVTh8C<3w(8re+(;X
zq7(`148DzF`&FBl=`=x!U=jL=JuiTp_Ums~F|X<e>4^a~i4IZVG_mv+2FWq#WX4mn
z`l6!VkXr@xJ0*_FA_<xB@@K!tpzL~X956;<vOm$x;^8(C%iWR70t?hjGhR1O|LQud
z9Z)_-D<#$eCzv8wqHnQ9-B5BxApF1x5plv-Ll3l>l-R**K|}!7+Io`&Me>@^!});>
zeG&fcZNn+;-!*lKmfGL9Vi!5{sDTvh4B-5nVwW+wXbPF5ck>V@QLwBf?Qmh`w-@wC
zSoY0L%GdkSev4rCJ&)!-MFF;O<{ag8ft?}hob|Wcwb!QSs<)9fBiz2VG~Sl~$^JNt
z6@-*sDSIuv>dsrlABWMkLnad+q|S^lE~HRX*bqe|Ma*V@zs!Uu;F_;;k5U)$H%}qd
zt#wdI_CVFAqjU0#(p9QUJIeb-{W!v)mug=v(*424%Q~?%cGNgGS3X~wUdBPAVFlLa
z8~-3|;^Ty6?nG5mk`B%hksiwvE<!(@TC%<ZLW7qndmTf$FB5M{%XjNiw1_``qSE7;
zF4@jUvlp`E68NmuouS(=sHJ_0f?$@M6W;C&C7490hBQ^wbl$-*(R<z$!R*i`lVJO}
zDf-jg+IP%aq0|Qu-p=w4#=IzRv-~?1W~Uf@(xxDlLCq#g9#rv1h9N&oKw@LSri5{*
zkW*$-$<AWz#Tz1%8=Aq)!V$^ll@!F|?ctI9nCrL*P3th-BS}0{dT7IR=FrX=VR?K$
z8iX~@i16%mcd@Khqs**;Zu(U~$>@f$no&zx_uJ?@a|WH%>9gC~x%}Wf%;4E9cIl)f
z<X;xn@j8)Xj;e+#vc}+|KXTfcnFX{&VqeU@qn^+0(#@J@xTeO8zcZ&5c$DmV{@wl<
z+tIRiWw(Zq#ed&XH+g2d-27XmXrS2aTnQGa(s+mxcpNk)W$5lydOLUvuJL|ode|{|
zx!YcY+sOLgXXef_{pE$Y1rwkNTG~pz8oGK<Z7O@Uz^W*G<IpAvO)Zg-jo!Dtbs|*c
zD(;B*bxriRs;m}!RQz-#oabZNRGtV~iM2%NWW0Fmb`{pI9;V-O>cbj8QY>XL)RK>q
zju{L-7!-m@x$#<LrhbcB_5q21o_c4be8FhwohP6N3^g#T;;&6p*z@w*8_5H`-$@P7
z@2r#`yjqh<Y;qFK+94Ocz^mm=gd80G9?ZzGT-%2lzRW!Hs~e;*he8XJ&y3(LlZoR{
znB0jX`dk~WjF?pXL1gKyJm?PXmyfEsjR?<ZnEj{de_iSH9k;s5Oi8z`>Ep=tw6)!2
zm3_Os9yx^o%tt7nbzvFR6hYTAIX<L$+d{S1Frq%#w<c<gUEsiQGdLpSgQ_{adUss}
z-42C9^71#=lJOuELi2^S?c<<3RtJ-u0TMsV9s1}PkGt%s=P>5q$W!DPx*Mjnsv?*Y
z&iv5P5RM+UB=(Q)H?SV~N!NVIvhTTK>vTk`Rl%sFP)pgw4uons>E=EYTO(&x8$?+>
zGI5X;m+3b5#7!~f+gIBWT`5AmYJ)En3XC$~i;1^B;r^FnzkUi7!APf607$*|oaL9q
z?00f>d<+qK4#NF+c>5ankl~cH+~)g%no~NcGx0}!iaeW6y}-+b6b<uukUBht!=?SO
z8m&tf%_)ts1L+W}k&FG``CE3QQ!brPa<-q$7PhL=eh-fif1skKcE1mhik}{^j1kN4
z#%HG<PefV*S)v3`OgI7IQdZ6}pgaLN01|=}KAWjpEU(Q}=tDvlDjvl6dh_;(a*;d`
zgdz~&0BhX<jqdi;!n3!U7Q*29-tNlJ1)ZAGfslRhHWJvl&yZZp%wQ_fHe!@@=BM<n
z15uL{a)ywQir()@%^^WBaDGj5?Pm~_!JRNx7Ir}YVH@8$d!q=mzRI2-OeD6Pm7<yo
zXfANdYZSGPp#}jNVqS4ie7rsBkD@Zp#p1@cp_k&&J6`Upg(2d&11ynvRIFOnYu(y%
zGIrCAUuOFRRv3paZ$3F|Tz1YVu>l{JW2D*nfzkIN?TzT|{Opb0_h8LCA!2chgSz7W
zB|h?3gcIK1zx?7VfoEQ$W1}kNoM<^xnc_&dw`H0m3m^blYKfn;gD2`Kp3sLFgLu>i
z`5Ba5J)Qpj@|580TSIsF{Zl*PH%3sMv8ibgP>|WVLyz-7{dP@q0ifT=f;)A40PtwH
z(#y7zuT=MakIsLdXs^{IFr#apTlnKLMKunX+}1%R`!EEnn6{_zEe0ln_D14hTWaj5
z9@NcF^y6CLQ+~x;w`@NV^#b?7D}y&8-xn?42%O&+4Aw_Y@6H8CP6A}$q&Loc*p@2d
z3BqpWcvO&tXeOjWu-M9_iJS|=SMB{tU}=={p+nSB9|IGq@AuyBaTb;qNGF`?1t><?
zE(Hc~#c~MbZCU!`O&7n54-<idaBuogFzP(p0pw_bFz!04`){r5Pr7mthN<ZoFo;l<
z&gze+;Y=S*MJeYJ{c(@&Sz&9o-tg2!7D|4p?Z3<PsSH4`vSV*^{UBVae93_v;|x{h
zoxHb^+-b#``7IGxj@P7i@fIk?`L>E>f-H*5w2tn5Ge)JsDoCDq$OEnh(-R*=#_W4-
z*OvrOfjHE-Y>M5>t%RmsRYS|~vH6Z`#J3l}JGtkj4*iX3ddkP6C~&&agp{IiueftY
zIo=2<aec##C|P=zEgL(#+WrtJZV78ayByG95#yPz2Mm5xL+}Do4d@18>7(ZJ(!1kC
zdw<eaX)BeJMO)5D3WfC~aKao?8R~2*&m4Cyv8q>_yP2P7a?NjcWmP=mGjm_A`~q|#
zkNF)N|3s*Pn7~u}>pm+95|XA)@Of?jI=DM=Gv4MJY0dqB=vR9YE;=eQxC<XuG!_>6
zb6$M6Sf0OFAT=)KkqxVZXZns~NIy>Ma#7p$81vQ0+uyh0%P|ez_TUxQtXWeKl@Th6
z1xjmUD<jujs-=soy{HkS`=gu$pwo!A7&iSeeb`0R0?44<A$Q?V(1|9}cIl{l$P^sD
zxb<z`DL|RB0oeq(ye*<*nkj`S+?;GB8r&tNtvXJMKQ&g)G(51U!$+P(v`@t`c;srD
z;{v47Hm3`s#4<Tfnr)3r2G3gWuO)gJxGkDx<i^yLfj(aabl_2<3e>5CyIXu(Fan;|
z!T)@8OYoY?181viQdGx_HTdB<F4056xQJBD10OT|a{@B?j~~G<sfMLe=1yt1r%aZ~
zq?IBx63Vn_bB9PVrZs|~+K|J>amQQL9{ti0b)FFu_XLRBA>rGy($udr8l1J3;fMw;
zY4jl=mxbsHc-Ekzr@(ydW93^?&CeAbW63@dW)0nAJD;M5(sO>))iwv$aC&892VkK}
z-FpPGr22M9`P1YKUM4%rtP7mGblPIFyBI}O%-hws{4FTw5VXegHdsoDaw)U&U1+w>
z?!g_o)%-Ydr9$(%*zrsa6xzMKMOV`jd+w6~?|OZG?}Htb8}0w(=U;S$rPJrBVO|25
z7z*SgU3mNZanX{tiEG-tW6Z87eXp;S)GE@@M2(|GegOZ%KZa-XuMf*izz(>>;XD(r
zSaC!9>+jwxgICk!lfe3IRhDQVw=dedV#c!22_I)YU(LEwhECHr>ztG~*ZWacx(e8+
z^rO&uU-fAL2@eLMubXq18k;7a<F;Y0)MN@ox?V*(qubHDkCz0Xp<K})-#`)@%Cd27
zdc5bRE6jL)d%Qa4b^R>TzuI2;4fd8U@sn=oO3kQSoHu|s=ZnIn`)5jZ_@OSLlwqvy
zT5(&W#<?OLJyfap9==&eAI@-Oev%jo@BB9mLS1Ee^M@PW%4J-Br|rI*fPv7GjgviQ
zaoKp`K;;gUEl}LvI*=zUE|;r{s{8{kq@|8iJScqJ_G2m5n3Z&D@95kQ8ykF24@q8v
zO>s%eXkPq|uFErNMZs5Eqy{}-_MKrbev=?X3{IKy!>fIa(sJIRn74b*AN&)S2~IAs
zL&Pex#rn)ZGDf^*8Zak6s`LgY4R)al%UOIsLDitL*v8A?Oc%(<4E%_dgWNo4i!ooK
z{o6jfUL6-n(0ZuDZ@%-)^XeUq_eUa~T_DGmuYT$NaD#I{eD;iNa9@|3r_OXCCVVZ{
zc6~`tqS(=A@t-y9Q-g0j3(2)HXqD!O>yXt+W&<gJR6#K3>Cj7kL__G}KnkH7)tKRz
zv=Y%^2H|Z;yYRqywgEpWEl+K#+p2rON|lCu=2ThAxj-&Y?4mV<#)I;o%1*wXWV?<F
zzCESf?d74Q<YA#1U5rqy-94NGNb7R5cLWGOeixyKWApjtuA$u_$yS&gUss+9O%#Ah
z(xySf_HZ2$R1gY!nwcEjw5ok;x@$Qhc@jDp26ycs=lc-1oe5#@r2^Sj8kr?C+)vU4
z!M~W#T|~U^T=owkx@6{oJo}66Awx&+%Nr<&1dWf&KwuK-OmEzUxPQm6EZ;r~Jl$%6
zC$^|uzM<)74y~HX%PPAXHnmU9Z}*@cfiAV!jB?u8Y`y?$!OFVcsdxstMI{z<2iek@
zl9{stfiX?yy<$cgDx#%87QA)V`2~q%p@Ba&$I2Oe0|`<j>&FooNCmi*>)lOS`|UqP
z?e<)#!kC>-<wG%9_V3Cxv7(%uIIPu7GDu@vK{Aof@x5X{YgSE8V%&J0wlOVJ6HI8Q
zNV_mnRb|jj@p?RCMsR8pdU<J3_VSciTRyB&n6?V5avoy526dBD-v92N!9U)Z`6lE^
zyGO$m9P?E+CEnl_Vz4<SMY7d<G2?xip>%%t{DvcwxQbA@=JY>vxi?GE;COg<kr2C2
zv>d8MFb&6WXx(~~@6%sgLM4G0fa8f*%*hkwa@$G$C*8Yfi&Ev+J{tYDd}qwK%dCG`
z1FSj5baRj<^4Ck6koOR)7!kk0xm|3j-BkC(jETvEPt9kqPZy9veMDL<0(bpzO=72p
zD3~Q<H@CEgy$LkU3J}%q3P28Ua*F9<A(#1h7se1ZxQSF$G$ZSNx5C{503`S24X@fJ
za|sUmu}H#Z;`@}}g(;$p14|>y42#+HjC2quZEGu|Cik37VYsIl1(4i44oD1?g~TR@
ziJ{N&U7|G%O6}+yN7etV`-y7E6}p@*YPBO)))D*f&!7<j!(@)k6RoMb`h1c4G+(z0
zwf!0UV-n@=UY2yZJ(Uz)NKzn7&?MAX`YqURSC!2yf@SiYGzzmo4fb^OuCU(Ml5(;!
z<3UH0bpS_hX(X-D(KVJ$fBe#4f%F}ZTCfO<!Guxvl<H&5_gy6|dmh*!e-<Q!YfLqa
znREej)RXC>yCKJzG4|V^t!+0;AA~#~K{5RtK%jypO2+G}NvbmPhc6!eu%T61Std(j
z!C0I6Sw#ymXUIjU99DmzYB?x}*e-=X8&{|XdRECobo*wAyAkRcC}mU&KvZMXL8XoQ
z%!XdpI2-Hpep|xlTCbOWV~aDT895ndH*-tboqYeMNQ8IODPTk+1KrnwN?N`kY#~)l
zLC0|&Sf8#m-zTi-k^AR3A1Ed3=rPv*jpM<GcVV@FhfSG+b$)ZLCtwddOsJy2Xj`Xu
zSuaSnJQy|d<V0|j$K2X03Z$@q^H*L$k(HCTD{qYwNo*HKY4d~79In-vPGGHUF(1%y
zS-N7~Q16-zAdGZR$Ox<%i=5>ZQ&px5ASIeLsG+7@O<JhBn?tCjlF;adNV=>=)+`Se
zeAq0fwF8RMF`D>mclMos2<FedHB9PMgZ3QhC|QRmu@b3==aXf3pzU;MBZ*r?e=A|e
zVik5JF-%Aa`h-YkySNq^-;u@t`mf!JX-x^5uW!jLi(?GIr>9-cH2S2j9?aIxGBJBh
z*urX2Jm^5$sS|b!4W@m*fo*3Zpx807{^S65IBP8V%i1dgK$s03esq0<&u~#c!n>RM
z)hiP<M@lvRS+`7)mrjG^;BcFZ!VM3iOZ4|>L&M6Z@v*YztbSz}_m(l)J60TR=v!7_
z24cnECW{(7_T1jyAyfhvnM4>Re$Dfix_S}e@q8f>t51Si#^MP=FYQdwWH10x`?*{A
z@$KC?q{cU9?Qz-;5a?6L+h4z?q6RvacW?J%&aks5ccbfCg!YiVh~!pfD~L0MZ{Z#>
zOnW%%iVYY0GyCmuKMavP(U$Dt@Hm<ri1{@9Pi^PrDY{1Wck$Tx7)iu=WX(fr9%*5r
z-O{4bctszg4mCJf|LkvboW^FC#zi|UO%bsA;z$ufdU3qWAv251iw<)Q2kqHT#?PM>
zoIf$@Oq5g#18kW#p>mLE!^nDylv1hORnbwroOS<RKYHo4JEY(kJaF0l_DXSe^Jfh_
zKtK^5L<o9U6*U9gCX*@H8FlwLadjbzi(T<ys3EW``}gcgNijC9Me6sum}(WCCJ8A>
zE$$C|Y0O{Or}=AU;E_@yMW0BE!dPjGWf6+6H#itKJCWsOUBbip8xdQxdwt9;pZ>dx
z%TC~XM|SW<rX#Ci2A^dfUWd;U48kWbrczR!=8(FNGz9xQNg(Reo8Xd%kjkma&u_>`
zWKtnLL5BtfUr{#CeOfrdhidQHo`or%h6xM?$<+Q--pR2H<pFi?^j#+>tQX(wvLEkO
z+8)=p<4i>nzKFz*!I&DbaToLYdZ*TN+1wgwR-dWT%P2AsQID}<qEj$9P)Qe?P|ROz
z-r^$0&n13>D{2;feriSaQnL2F5AzEqd95Sqx%2ToruGY*@O!ud>MkGHuODB|kb|d3
z0pLz^bMN!Az>Xsxqag6{t*x!$Dp%9V??d9nRU}LQ_9Rub-A{8rMC3ls5DL3S;~3nZ
zXO{iLDIqcTy)$Mi7N;jkD%Lh<VG@%~&&VM?L}a0;-{@=sFSZ~&cZ!7@rGLxK!zw!Y
zDFS7UB7cJNBDWH-nuQ8wjpV=vlKaQtt(MfVol&$Fw~z9K6zX&79LN!&)8c{mK3E?w
zpQBwQi$P|nAvycR#kN|HS{Ba6_N@NbYT?X;h2k9G0xvJ;69sB$AOB8cdak5jVO#;d
z2Fj#gSx;mf{6*n|Ri(|*<~HjS+QLySs*pjF!Km)K?E823xv#5hTXSPaVDLwtU&>TS
zj>&t<s);42Fo?w<QJ&AIb~r6{WoTW0*IHjj!GG`m5Z$bubezKJ{3qfdon05QGSG?t
zEbd_K6JuscJTU{q1;dEo5^_7vq$(quf+*4wi~|e*wa6|6I^~P(4x*!NDhIJ@dFuY%
zBhnY268b<3nldI0yheWa6slsgkja#Y!f+17c#jL%f<~zF@R)(B8*FY{+o{ujlMHBT
zc!2?ViCPvlLx*a4?W-Y*D2G(<rkFpePsD;!E{2(m8)ZjqK3Penv7pL2!To6hB$eU9
zZR=36FEB!|VP=@v??P508#yX8UV+wXneItu=jbBxh{wWrukC5?hDAOX>YDqS<@AL}
z9NzHFX}h|)0iGU}+0ywVa`DKXZ}PCylLyxb0`9K_276AzmmaCuIBVSBuSEV|tfd(E
zg9D}>B8P*tA1HHt56M;m&8R0w5<yM?rks#kktwgSa9@lLUbc%pvVJ1~<#@lXB_{ot
zq=8|S9cWdAKr||yU<@;Kt3)76>$}5@1t4>Cx1i?sffxcDJY+ca9~j~?(JsieAG$rZ
zsel;F6yB|cyaM}4TJ0;br#VhJW$X7I7R8}@G9YmksERP*Chic?U_Mb85C0{3RKOqz
zi-H#|fRz_JY!-3CW3ZCcOV<m&7#TAnBpxj67Hgj4ov5|AG4)Tdh4^KIN*1xQ4$wx>
zJ*Fb4*o!91It%rxa7?wfYUh?E4<#{vHl=w>>GY!f`d5bqI01vaq5bi^a*G((N$=^i
zDwdrt8Dug^qATVPIh`TVh;2p!dWgZv(pv7@raHr<kussQRLoplmFg;@4lw(lRIOrQ
zZ#*+hSn_6V$RP#Gm1<3a%e%)e#+sU53^l+r?GL3Q3?*gRhrs?KlY&!L4Q90}tqbVf
z<!yi?IeCJ_^M^#$`jL=-u4oS2t3hyIuRK1-WWqeRkO52Z_d9G$k4Gd&M4c!LFYg4)
z+0@ymRjp7re`YM|SdzU9Y9ZJ2P_xR7G43_clTRiQeG1~3WRPT`NW)aoqK#2BS*)@5
z8XPISC@#`B%za!bLLJYQWWY@`b<miRli7kH{qI2_)DPYb7Dvd{L^qTVg@l?F_WPg^
zZXLIH*qmM+a@$?r9&wkc3-02SMs~ru?IT`zrF_VCf-^b2JBD_O1zLPKWK%2@N7uzg
zuCdKa_iv54wCw6bOi}wRzM7$$f?7mqWrk2_wX(F$X}Ix6JrD~)e?ZhWqsrF$=H^XO
zr8sIe+!kEB``506^@LT>RuHO>QBsroDxTP?#Hh8K>7U?{rIRtC*e@*@$8({rS%UMX
zD=*oFueL^Ea9D!`y=$W9FidpB38~@K8j-p2yJ?1JWiknm=T9<z`JjX2WPd+--M7p7
z`Az;m=-xP5OipYv_0Zsfq~Pt+cklf36sd(WGIhguRT>vZS3IVZM`$R8qTUm_0vuHJ
zz#bet^w~t35`LJpv6DUoe#a0FIvU>yq+PslI?(1~8ux|!?tD5uvL@XQ{s-rPiNsXd
z98F4v#RRj!jiZ<y%eNXxL^z|9lftlvxaTFqALCMZH0K~h4{$sjCAvVhcxt&-Qnqzd
zc*3^kw=gue^ZU_K-*h*ow&A`C9O;>!PDg+2#4uv-_{E2og!5*KNb*(Rz3<U4ry^xV
z-FH6Eq<QLeKsyC(iY!u?*zgX}fVBA}A!a{S`K2vio8DfNWq%JS5Sp?&nX=#zo~6G3
zpA!Ezn(GJN>u3|TZiuR*6&X73RYR$BHwi2+r<(9N0r2MfT}$AH5E&}!3b@7Qv*SW&
z2hx&}@8a98tyMiQjTSp<hy=6NGk_O=#q}fD(8w&dHgzjmsX>Rj3?DY=J%8&_u*he#
zeEn*bY?>qjIaK)qL50j>kPgs!%kYvxW$^UK{Z6W4rMra=IS2ww_CBfDaSn$jWag^L
z5saeJ$qe-|I;WGOhE+7-vlisHo!xQ^<P||Z81Ww`bNbbIW0BJK;IqDc=i=K^vGi<I
z62<IjT00^-UODMKx%K@Obes*~-umIb{^!(ZcYzIK{*dD&Z43yoSzcKTqYRrmwK@s1
z953Kg+gt$nc!h;AgP~5HMYCt|-KL@5JIU&MV0oWTFyrB#=EtHlXhcI&S=inKe-ZjZ
zk{ge+UH^|+*p4F?v_layoKyrFja;PYhY&pPnn3kTo{0U%pOKQ{#$}q;$(fmskl6VH
zmZxnXAJ{u{yxcZ?G~t)>Sxnn59=<O!Fp}mv03D7`>7bPmO~?0y+b;wZRm<h~whJGH
zNgO9|iR;OMumd)BateylVvhGqLvtpJbnZCwvM``%e1~9}Qvm6|dkG@O+#(D(qwwd6
zkh*-=$Kzc(Z))un^qJPXN@n^RMZi_?KjZ&uy20<-{=+fiet_*_!Zk4SK1DT&!NtZ7
zDS|vKDR#zuuUzwo#YtSfMnSqa<YykYQHMCsVR2H^fpg4YJSygIuq=A029-HBt4FS{
zU7Vs9ToqPQk`n_X)3>jRt@Fk_g!N5L3av9no}RIeEx``FDr|XEHk*Xbtwfm^YG5A7
z=h8Lvhh5ILNijN!Qi7zv(d?x-K&oTY-T&;*Mc}1jZZ3v|Us`P;(uyH-9&3sH-`cR%
zieVX4Y;#Aj%eu2lUjr1acQ8Y>iz!w!yfx{_r}k4<GMp_L%*4*SPCOVWO3Jx(Bjy<T
zB9>zd-349w9OL@u)}RMCz5ibSU#&jj%jm#+BLtk0nJ9Yk<Pp0J76{RgSnI9DTDXM~
zW;0xMij`U_&arZp0;!krekxhnaWJpMcQZs$lq4*ia)_&m3iNMH5OkM4&-!qR%j#dO
z<6Fv4YxbAHJNv33L;X+?*^t+~puPDzP1I6^KUSYJ`;p*?=yqTYwF@y}+^6%q5liOt
z(dimVHrwc$s`U*E@ExB=q2VXO95W34{6~>mAkrK51@L8tvoF<8#%K&ojE3^DbNIA`
zK*h_*;bH$>qYJI-lf)0<38w;{0(B;$q7}gj{&yS`j+bo<g+{K!hLwfF(ZHOHC<_A?
z^a+j9VoLIbS91DS%Lf>BXU>o5@>^g^ZUl$addL*e#adn4vS?<TuV~EnZE}Ri?M$Yd
zMO@ZnJ;ExcFWp-`l_5Nq6`3~5(7(V>G4?SravoE4Jy`T1k$sP3d$YA>^Fc_F2R4W-
z#R^%J45rti_tZA5i{&32niYs5P&s*cXY&pL9x4|U3zW~WRVili;H%c8AowUykH}Dn
z6#n|$1c_G9Ofo<1yOZmCjm@0PZupFB3nHXPDOJmc!0#bT|44)~N%&5HOc7lRQVE8{
zmLFiu5;QbE{S^=5ny>yH)e{dwvul5#Q+lz$kwZg)SYgp3zc{0W$ui9CY9}S_^>G?X
z>bm?77)9<f$17T11|NVOy3Bj%i=gy$>`GB>gXF%Wo7UMK=XZ>Wl9SPY1-&VfI~L;t
zwc~Kyi_pgM`H7hPd@!1f7@1W@_nKN-M4M?Nm^2Mht2%L({MOdi!A94~s9Lm2r$Dzt
zl};m~A=>e~(d=X<wo<6>=7;xlviLVLvW5^`!n9O*+IquY@w~czL&oLiOX4s26NS02
zP%ZI;q($xXeIt?7!<b=vFjIG;(;~@~&%m<sO>n|z%j2~5O+Pl5XV~tgAIGxE)BEn&
z60Nx(vp7V*Myg>EgfA%#0ybN?NYXRc>jkxi+|w_meWN@DpuOBT=C?$Pcz`3QRHbR`
zW_PBzPW122jzn!#DVw+Za)%&YGuj0K9`_p5adEb0>;vi9Zj>Sf!soNRo;KLUAluyQ
z<&xC5?4RG@Fltp_SB1l1$+USqGX*AE&QC|wf6N5ulZ|~SmQ8ZpNBhJlt&{}9NFJdq
zx<2<FT3(h{5mJxjU@ZC8;4E$C<V~rgtKF>@IIf>HBpD(w5kRJ&&JG|BrE(rIu_M?i
ze11*k$km;J@|C9(V#Xzg3l=;lg=u=FLwgfUX}fW*fgoapCB%06<U6}OKS(7Iwa+D;
zD%5S&J1OSRmr*8p<I8AOTEY}fyAed)kKL*qms!w>%JIKYs{hZy4X}5kXUS;jWn_Md
z9xQ3sVkdlA)jK_Z9wSTAy~^;)0_Py2PQ!sbPI(-+d43Uo1540`%)Jj<jXzzb(&>=O
z6B{1mE0&E}=GCp@>Y?e@9T^JYd(n*3FCHD96@au9&H_ali^g3aEtAE9k(IZLEafxJ
z;PVk&GV)aW5NcJYh^|?ygK6N(WRpZHpgo33@e}{HNtauplyl11`7;SPvhZM@dL1a>
zJZ%x$wO?_~*BDSE`4pi8wITJ#UU1)1cX}PfJJt%<&ZdrTK0+MeR5(Br4Vws_Ab-nj
zpS7$Kr+AYP)ZOJ}7>qHhwXGkmh5yUVUhQ?bhA_<FTGul5>5QaAHBD9{DAq!F#d*4V
zP?%YHRY)Q(v1=HM7_JYIZRL%`T_rPki&0gvi>2#wOECIi$|m6vL^XLF9^ZGXMtnSL
zlTAdT)vCY+rPZu_bGgBniXC+_afp>|k;g0!&R5Db1IYxlhU(XZp(o0=M4x#i%VPa%
zXS%SLuF|Et7Ar9~TU=NYsqk;uBL<x1JT0sXQ)3XG#+M;2K<*r^D+&LQ{F%52Uj#8m
z9Y45xn^rW?>F1y_0H?B5sUFm%!}@kf)-OVB+tAc7wA{@hDlIU<+-ClrY;`moRpzLc
zY9Isb6JnH9TqN@xzzRe)&KiW@{P6eOc3<c?uq?X}<ca;|;#C{wWZjL~sT&f#4ihQV
zBJ<R#A}DOWl__w<7dn_KS=z?}xc%DJFmr|!vmjL;bfVBAs?rtP{9==WRZ@05olN`V
zW3*{%zvzkA;tmm0M|5-5M*?mg-&gv|->(Af7w<AmwKe3KXqC`H-w)4Auk9V)5a?Nc
zXMa1#s6=6;ex@56q7vPPG0ZkhU8h$lR;NVrzUAUVaM-3Gla<Qfe|f4})?Iob_I5+a
zx?|i!BuMYdQ>s}rG&j$ZEe>rUrCIRJq%3j}&%_iY`~L#k5+&{ScLD4umnsOM_=z9?
zaTXWmId=Rg|MXw{i$BmYf3G#h{P*9ZG3H~;f4}vEj`_w`geMfa%`mPg?d7RMdj-jG
zR-lyQ+_^={VHG6=Rv4@l;EEEYRQrPxj{Q+WIU;;YtKzd%><<WlKuUc;{EQc{c>Ek;
zSmws{izr`F4$GvsmU;K$3jJP&Clxl2n3<VD5)#yEEG->DNXg#b9{pY)lN&bIHV6Wr
zERWIJ$5}~|TTl=NKAv}kr$6}xq*d(u-8$@OVB%n$p^!7N7UyIk`GUcUqI3)bp@LCw
zp}^(|##$=nGD<3(-$Y8E#pyY=y4wu<7jf1SX9a=kcK3*r6i<0bCq|jVyy(mFykc-e
zYI32l&J;R%c}kwfm@N5VfHRKsyEV@5h}4yZ#g~%QKxQ>@t{A3{>uV{U%xBIQEHyoH
z%EY;&TnaHsN;DYahXGpoBqk%zbCib0L=BbKke-6DG<xbvXw>WYK|q#eWNC`^d}jO#
z-I7OXqJ}7i2<Oo_aSvze^Q<2|&$VCuc}{9a==p>uWYFs)OFrZg%dI}Y{@NO|51r+m
z*WPAxt<4jE_AlWzO1$=sZ?JUyG_Sw?I@<GSHXGEMO;)d6M|l#2ht2al+o(HS&N`Ab
zWjGkn?X}t2-6GF3e67i{jDEk*a5yaXO-jIGEw@Ptfahz@o;^*u9CGdYHO`zlgYWyq
zAJ89ijQRIxjWHi%{)2=$Sh6g~m=x=Bip-J^2d0;fa#eEb%+UiG4(E!AQKgH<ekUZ-
z76eJ$?#eJKj}{+=iTz<V%sVX2I*AmDxy4gd8&kAeJ8Z12(jR7QZ4T(SmzkX|arX2Q
z<#L&+sVS7!ctJoECrItl>GWA&T|;S5szMO@l)^G;k`t6da_gu~E^zwHN0^;ChOq``
z#puR*n0ds$Us7aW(qzBtsia0qg_4DvMA2VpNz<5exr`72M^8USSZS~{x5&iK6;`gk
zQ)FQhfl-<~%V_Uz6L@m8X9iD=%G1u^TrpT8r9>%3mSqRiuOEg@*f`F=KkJV=&hL!W
zDUi;RAxW*G7g?-Hh=z)W@tLn=)GCViF06CkgMv~xOW+lPmV+dLJVT7ew&Ot$n`Ja-
z=cvxjk|inDS^P?=@JR#INCzk}j=;r4?|qwhzwsq1bBnavZBCrJhel<B`pL6QefCeY
z@|(ZNQ8Hw1kV4{DO2oP4%Wto8HEb|*<~(z)ExI?J<LCayUt{j<8N%6l&VJ<My!6_)
ziKB>WwZhEoG*>Gnd=x=$iT2ufQhrw}kq={;+<<WSz9!3JqynWBQ52CIgCF>4PZMY9
z$U5fmNF0T6uu^d4>SYG~A#qkr+wehe@MFx!n2$04(AEzs<_L?i8P;YvXHZfhg?GT|
zoO8@C)R~)~L<)%$BQH`ba0u+aGs4KEUO02$l<+?FI_HW~>_IQyVPBraVR1OA2&)a|
z!#d68EOFdtV)8|%n<dI+pI*O*(mq;i%9RRUxk{}uL9H>t+}s?koo#lu_lTmHZnwu^
z*g|`bI7wKTIm3MqeS(?UbKo>4PjOBZI$@(sowb(zl*qmoA8<fW3oA2!l-1J*q*O{!
zS`Y?hv=_2);yCrmSzMCi2f?V13i4c`!VupNNaG=r!Zkw6Q3lBvOkPaW=1v^+`4s~x
z`;)gH{0be%`P~`kcSP!fz!7DF#0gRdts&&G#oCZLUoc;l3{pr_!7y$!FcUQW625?_
z(<M!YxHKi}4+(=ZGiOfX&rB3k9n#_XK1%CRkQKHmd)*<gUAxKEoi+)M<i-k%k3GXZ
z4}6rJTkAaZ*dtVGQ!L!`AeX;1$#lFyJ*ePU%b1~sIOfJ)m!&5@#t;4YPvHg;S6_aI
z{;e%mZ>-{<IL7_Ye1-?_y%$T4A6A&3ouSsO5byMe)11VCG2buv3lMWB$Zav^>;zb6
z$ekt5GkSvo+k0)gy+NTc{vA13Ddd6a_Lsl%6|@&{?%a7QVU_-1_<@dXjQRIxjWHi%
z{)2?McA7Me$Xy05#}`$!(40IoMYl6#Yh%Ef({nVc4Xkij=`bVwoCDzmP-O0$kl-wW
zQ4g4LIG1CU#YoV)aB3*LW%I&^mBKT0U;nF+_Ry3{6&`xzNk02S%WSM(pjxjpKfgeh
zC2a5P(##BDwT7<OF}Y<JML1DR)eeR!{WvCyGs19!Q>P!`#MuX#o;!xa!`ZBuI&`)m
zmlQo$g=C32$R_T8hT_JoJ%5y8DTo}Runn^kwDc*JssxoPFMjP`ql8C&YLZT?gGn>2
zRTyV!)@y`e07il<rX>&is>EoT*pS)cb8d2Mp5jb~bO!mon*9HOYaHj_pY=x_=XXTv
z>i3?b{pMw)wdADiwKIaUpduXQAYrm>Nn?X2ed1w{jjdH0r|PsiZL|`UbwJ_-%@eaU
zW~PzN8i~yjRw6?WMFoT<CuX}9^E=<Z#<SPf*y*R_zQS`Ev#!lC9dge-_p)*GD$6&n
zbN>D&ei#xSKgY%!%QS)#Svmyg*zG0k_99MBFEBB+fSk0H7f%x1+U0n&PHHmV`sQ;y
z{Lx48q(%rwy;4DG&0rXlWEl=Y&Ib>o@3kDz>kU}BxyJSDH%XEdB@1}sI6hDm|E^0&
z%H;~ZexE$=GBY#B-}yU#m%sgY|Mmwut}*7{zct2ujQI}==FVDj6JcEnSUgz;1P?xP
z3~LQ9zxWRK-hT$rWX55vAUFGGbVx=76hjbFBXKyJV@*2pw-hK<l$x~GBT?02;`A_c
zClR;h_Jk1Bs&#(iC;w9}U3i_<)l1x5+avTH3v;u~PEAp(Rq?dOQy#)9`mG_uL5y_}
zcoQ7E?@>;id5A`19w8Ohf)g>)9SRAH(Q7f9Jk3o~_~B{|6n$d{C1X(#y8R_bntak0
zU-SJkjpiI%%g>{1PJ7qV>pNU-a86=@`MJfy86nT$HVZH~6df);8=Nt?k(>&nY~B7)
z#|K&CIRE~wKgu}2BT{$zhkt|%-I(6B6|y|WB|f>AQ1&9GDjA^<S*B2ELNCA*iq_6H
z`EbZ&t%?<j*~KH6Kx0Zl@q}a@z8@m32QH$WXT1F0F28+ok29bBah|%z@a=DYjZJe6
zDXMJjChYBYS-!E%wTth9r#Nx?UQGJ`Ywz58<GRW){H(S2oINu>#EBE9iQ~9V<D_j8
z3W!nyLgE4t6cq`TJ0$)C*N8vR3y?sF8@Pi?AV8oVBnnk2YG_L%+9XZY+R3Dj<4MN$
zocCTQF7}>@qY9efMs=O_NK3I4XL~a5EbG<!*7trfu2f~Mqi|J+Lx9y>s<&w?L91S;
z-)^BW%p9F(wz<Uq(GnL|n=I_x!$#BxrEz%;*09m+(2vdV82;}0`UK}3-EN<5_s+YT
zgb)<+d5kesD;1ipHqB;}sXbF$eqCPkyNB1@Yknu(P0gir<O%}183aWNg%S4dEpT9F
zoIQK1j7|jX-8Y3Y4y^^oO0+e|0HjRr*EyTKMkyr1fx{#Xo5o_TNvBW1Ah&KXVPY`y
zDh#s8xGo{m6Eu?=C`WFj!tqn5IXHKWmE|RR-3{t%b*`_ka@c0kV*xS{^t)X;QG}K`
zrVk%xZuS%-mE9-@Mj4#7Sf{bh4TrcVBfi|tm=Z{1LsH}9Tzda+Fdi)VzRly&j4?Rl
z&^AUJL3Q*XmJGtgI3<xbMv|0QMHn(SH=mSxol4)I4~_*(!j5B<iqT4=VznvKWb3BR
z@AJFI$GOk%$kdG;nq&UUU*z>0zr|i(N07l<O{o+yQUSt{Fr$dHL~AI7Il9+Y85<wR
zSSU<ZaD^P1Y&J>D?rAb%1|3?gZt>pLCVx7+!TDCe!lxdikT3G|3r}<Qy$cu}FdD>6
zmW!Ny{Va#(W_k9RXV^b;kjI}eY-}`W)K{q(5LRNXLTSaN>uvH5qNR8F*;7BJ*0@GF
zn`6cV+*n%X;Un`@a)P!9IREwq?mxajyVs<U&C?Z7M9@dzHns3?Hw1!qyG^szWa0P%
zmzHWMrP<h6-}1A)=3aBJ`S##$YX0pfKTV^t%CCR(dtz%i`N%=O@Wo@y%<iMpyN-1c
z`BD*OZ8B!bCCiyxqZkAX<2aJ(45TI6n??=d;k`7G8lBD0*!(|ER)7-<Z5@STmD&3r
z1_WA1v^LgQzrMoNi*=^<PO^K?EQfcW;7G1azF1A}|5<_5eUJ`|z}e(9&^a?4btWZP
z(i1x=2*XTbo~#3FlGqgy0%<umy^-atMVTZN))|X3F)DUcMyJW-tHfp<Lzr|B3sBZk
zsZKC6Gl$bjoP}_X^nzmA1Yw4!aVl0s{XR-5Vx@_d##+PwOuxNN{662A`~1#w>ZG7J
zGtKTtj<NpYs}y>g(1m0(d5R;NeqWQ%KnF{&XVE%BW+V+|nc7`WTHc+-qS28?I)RE>
zM9YfjE?(xP%Zl9#r#W$QnhO_S=a+x@74=>h2q=jPIXQvV1&E5g@a!|}+BLz~zWz8^
zL9=m#+POD5JyO64hs4tDM#P;CkDr_-s=}MSChM!$=n$}Sy}@Tk0uGJm>0|{*=jMsA
z^t%zw8x6+C##mW;pQ4mBjb-bZ?RSrnkv!RKz=gLja(v+^rBZ>bE32G4|K^sT?KR(?
zc+I`$AEo9Go;biSe)><^?H1Nb&i?Bi+RY7~`0i=6=`*=&KS7v;%i6&>BL*S7BE6S3
zh$=}&S_xb_Pc_(-3Fd|+S1E3W5(8;au^D!n0tf<x5QGSj3gZkQlkU?@!01?x(XlC{
zOg^8PTow@y?!f^zJ)N_uWfMs^sc`NC`Cw`7I4Ki{S%gfDlOzct4l+P;A}Irj6c(fG
zuv9w89y3}~DvdEQafId0l_Wq|2sDn^Nsi4wNVz(ORx0iKwKp3eT#^;%tVXMtD2h<&
z)PdwA&!p`nA3^xA-{;$st?u(XGIh;Xo%MEu@rREx7i;Q&e+}W_+L|O&apa{T6J*F1
z6;Vf!3qx9Qo-=Q)@Ic?t=xSOU9VUwzP9C2n)S7ckHGcQ*I*ZXbU-{1W_|`YS!H<6U
z1OD>b-{@Nl<SC1Bc9jlO&hBDYb&8x5?42m`<oCbI;e}ZiFTBl-cNdW>HKvbEVKG>z
z2@yPcY&Vr^jteUdTFYycOF6=Pkw&kL$b?u4xl#csC8cbRjYgATWreYkkdBn(B#cQx
zLtAc5VA>w!^EnP3oI(o0<;%-lTe-?ed4#di3NOF>(w3j^HTRl(%|Cwbw&ok_ZEBaV
zVNG%$uHOxLzt-jb%WIrFcb>-{dkm!&##wMOHAo^YsU`+vvl)|V7TEBv8iOuUH*8Kw
z#&O;JH-l2F)=4&85ClnHo0NnyhY$fcbxR3!@G}RmLu-u`NqmY0g7n#gzaPYt3`)V9
zV<eK$UxCps?LZ9EhKNCka<T?fGLi@rg)vD<)C~=!K{&yly~kO(@^7SEMF@#9mQraK
z4?XnArdSx4&W#<EUXw0lrDFR1h$!AP8f;T}_D%-8|2O<T-<kXTwoKjN1GHSbgpDJr
zm2sv{ouGT8LHEKUYf(t+s^z{tF(Wd@6+)VAN4?$UmBkKcRt;zBODs%OIC@|=2lo`Y
za%GXXY7L%WR$Mk^=8rwZ>Cd0$CqI9R=U@FRJ!=WbQI!)+6b@3!?Pq*!lCf$DhvCH+
zUqKYtdGt%4p?9s$nLq!L&rOz5QI9yx5XD{kLXa(&sCQznm^{IFnT#;>(e#Yrx$8}4
zm)DtY_sMs<WTOz(ZE;_qAq<(G7$<J6QPhU60OceWhuTITgV{r~JaA-=SN`!5rD8y{
zStnn}VytHIQVrn$`UT)M_nLdn?;XuMoet9nD!jAUB@`pcDVDXo@y2;pS2nOVq}kj+
z$fN|Bj8l+llfb6Q5yP3Nk{~@{u{uVn!NqV(5J<vMk<DttFf+`p8@wK!nxtGdM=qZm
z=G&Px>LV$crr)bg`n3QT5M%;^TxJ;FD_nAFxY-CLcjE@8^5*++bB=BDXAlG+1cA&D
zgpx26WWoSRh_x1_H7ZI<qedIFijq?{my8U{<fkZ>4zSj%6C3D7g87A0R7NN1_4`9}
zXtho|Zj)0?Z6cIXME!_JCo^dW2GFLFCs7=4S#!V7cj!KMw?VB|8yH%@K7p%NtKE*w
zUh^Hayyo}b@|xd!%WHn`tq;>&IQQXJ-&)J^@-pRe`IZhL&-TVSN2}Fh-@bjfltyj`
zulXkvulc=)*ZkhYYku!>yPDsYsq=U|9*@W4@oaDI%;I}I9*@W4@p!f+-qd+K9*@W4
z@$3M+sq=U|9*@W4*#US{=ka(v9*@Vf1MsHK<MDVr9*<`S;7y&!<MDVr9?uTIn>vrj
g<MDVro*lq{0N3wZ3dn&0h5!Hn07*qoM6N<$f-r(az5oCK

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_026.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_026.png
new file mode 100644
index 0000000000000000000000000000000000000000..ddc0daae21805d57dca4023bddf9c15e88bed2b7
GIT binary patch
literal 66340
zcma%iRal!rw=Pg<ai;`#ZE<&Z4OR#a#i6(rEf6#mio3hJOL2FKy9IaG!~XX^XWyQS
zZ@x@sG8gkq)|yvVn5wc28Zsd=3=9mKoUEi83=Ev>`{!SXAK%*o8%3x07oUriwu`#G
zxr@88lNpSXv5SMPy^F1t3Avk@le3lm4-RG?W{$7qmM$(1&ipJacK>ezv%QlA%ebh-
z{`(?G4zfDVFfbU#|MkEYi4<DFz{0@DNs4KBq#eIJMA7IZKtp^^P0Y`P)=i@zkd?-z
znwH!A)|S?ic>)J>e+mREa$^`^G(6(-AGm8nS$IT5YzlEe7%`eu*jCkIt;5f=iqpf2
zLr_f(;~fMt3RxBM_Bl&l)W7a&o*&H!vCxu~(x|QJwl5fP_A3>t^_qCeIOBtQX+YSV
z1b+`)@RAN}dk{x$I|>R<OT0z-!2I9t|BSs3uoVgXzd7$;-Xj18vQGc|C;u}C--tM{
z<$Z<!>AsH{`a7Z?{%`kv%on!*|7W3vR!=p=UthjeRMpf3s2y7PC=P7HMgpk3yu1W2
z2dON5?C)RhQ~WpwF1Y;e`~SCDHMnYpbNyDKx4G39fpNU?dtm*<<8Ia6^Lk3=?0{=v
z{yA6ZcjZG5&;K^Q(_EcjR2yq*;QsnO`aP@MtGubar8Ta5;&3V+6U@y^KL`4~UA0{v
zmU{1w817ZmkquZ%K*B@e@>LH$iQ25b(SvETkR;c)(&+j_)v(I(8t55p!EB$UeA~rs
zlpudFDf=yLfkqH98>wY}y=81uVnG&|thlayIDKswV^&aZT(=}5+gAI4*Lu4agB0SA
z^Pi!TxX4z$SU`|z-A7z`D?U;?)}^faem)uXd5^Q5_P`KKq^X%%=QYnOdyfPAHydxC
z*hNLhD>tL1jGeQW9>3v-fecB8hq8SVHWC@c3{2v3o1_Vw!~X5*>L<+eE*uFX?#!#q
zXT#U4)5q13!OCXu2lmp^QkY)E@p>3ssj2j3?}{{MC*jC!E5f}ig15c(XQB!{&%`fh
z?LFnXJf3%}cMB^^CVp@Ji?+dc=<`3t{iA=T_Fn6BBDO1EK2*(3a-VeTO)s$$M$!+*
z<XZ=^wFo5AOX=Whn-WAyUy}A4*@aan41E^|CP4bBAlv{dk#WA)MZbr)rb?Svzt`#9
zG=W-aO*~w%nVA)u=~qdq&_9?I{p2?PlIqy0))3lI>TXb>Bv(4I%qB+m7MTy;(k{Pi
z==tg$v2$^y8PYH;^R@Lej1(n{|8~7H+i}prS0k9|z+%DOZy-@jHk6lD7SQnOkTr{q
zR{?(mt<n<_w1`S-B=HfJ2;&$gmyB$GRXe4ZRnwOP<#S=mV{Qmm`T@WLD)xD-X=n(h
ztjsUzTwMBcBuQ~p_$?9~JsRD`DEha7Nv2RtI10(7f-bM9EA(~VTfza6*HZVhqZG2`
z+mMNdw~@C)!z#{XmwAU@_3G?zrY&DS(m^E|nrXzZdpC*-Kh=$;wItk}_>4TA8JM_R
zq~5K!!%yXgrU^E86@4ZDc=(X^(ykAJB92e-enJi}$@}`H->h#&MP%&okWmV_sW;-n
zZR%qen12z3`9Ye9RzU$*O3WCgjN$!^#mlYB+^sV)Qf^3y>P9fh*Hvek{kAe?T`HI#
zTnV>2?FEEUY<QCgNACg2JwgD#LIbXGV#Pv>zEjDY0;3IZ&ZoAQf-wclOJ}$+ik=NX
z`7dI&;ddFBxKA@FhA)M0A(<XyXUMa=5k9ZaRv8{!aEa@FC}e`M`Gb7}-#iW%YXxs-
z3GM8(b#?kaBT$5nAs^g$V(qQhD}Rjj56{@>RULv3b+d$Hfy*Bw(WgIJon?{5s|D>(
z@|!UE_gg9`q7aJODCj`wOR&v$h3eIc$osqa5}D24a02A$^s-$g5BknxBgG+E@m(p%
zHcXlf{exa1_9PVdKiGNcQm1T77TRC<!OdWH)ug|9kd?LNott@stzxLb-z7?=VNqFW
zhJNyaBa*QuWBy93l#fv2aWBxk8`0?bfBqgi`uh4Xu4@9xymM|)dP9`6<<phplM~=P
z2k6^O9TyM*juhWomDG;PQUw)=8j?`I@nTeo^K@Y>^8kDI78Uq0z6B?=AHaY-VyUs2
zP6f{@?RvEUiqwMh7Bz;P!lc-m6PEXC(<4Zk6_;TBYwG@tuszh|MJlRs=QoyE8dHA}
zYOMj(LQu$M0}-`Nq?+z`Sn!D8dF(pynwJ;+pw!jX^?;tQW#^heOxSh02_x)4O9kej
zc}cSBxf&p-WLh;6&9d+7F!47r<9hjEtT=_#YG=!{uErXM{)KcD8#{OO2yuvyjwZmA
zJ_1f1mtGe$5;mXh;33^k9!eQ*mUtD*!fs+4R-O85aUAhnjlEL9v{F<;BfJRG2F(tO
znIGHQffW~}b6EV<4_qYhIuqJ%X+N{^wOu9?Uhd#bu4=2Rb3ps30=EM)kA`#dmY0G*
zgoEE2MqaD0J4svYBE4FSI|xH7`!`$b1NmsfS*Y?u_4DWX*->+K`58m#fj4ay*GI@4
zpgaj8Q}&fbU6U%MKn!*ymTg1<C(RjPs%?;BRGv?dT-E@vo8Gi>oO5Z&CfT86{IHZ&
z$7|o9_&dG9G8c$6IDp7DIfj3x4^{04ki%2{LZ9t>M*1Zy`575^k%0R8sA2Yi|L8R3
z3lq2SvU)xybPVa>&&0h-%bq*J;pKhT9rM-kRn%4E`Wva&CtdGHj)}{9!zXx)iL^~4
zw+OQKXKodti?7>Lh=sTm!_0*me*vLD4BKRJt9U`xiyys;c>@|t3uT|r!hEr9>^@5K
zVE#eaG>JgL$`LOP#XyHzySX<kG`mKwt}>0M3-lyAL#z9c8$f29Iucfa-8uDHXMjHc
z;GI7^IAzm)fA~l*hO`;VDd=;4({s*Q$0-jAO4n%?S+asD9$`M_)DRQvnLA%;;*#co
z_p3yU;VL_Cu?<%rl$0}D$x#@>p;+R7Lts2K;C1Osj;C;t*mC0J1|G-{jauwSfkq5O
zQq9CK-%F2DUS3|t$^Y3&%y4s7U<G%188ewg3tb_mVqd+Gg-*?8uV3N#bue`FWH_4d
z&1PpJ>`5H$bQPaI{OzW_$+|SXb-QiSrw_0B8;8JVivu^Sq_E6;=oxsoH3TZ10vXrI
zE2dirJHwwL`X(f`{#~-}n{N8<YL;A52Ed~VxPxc8S^)GM4r9xEid(iQ{i|n!i<rKN
zq}Gyz*!p6W?-1je{bEbFunc1ArDBQ5G4-iq!=1}1m$Fijg{on|=#zF8D(Rc8*%_(g
z9S~k;vDkZ4-2b*yV<c;@B6U1B9JtO$qpOX{nl@!)gAIF7ukl1YpV$79m8V3s_yRFq
zeq~*{6@V4@JEiD__lTD!iuYGiEMrG>E(8wKK7m<_DQX<II5qjYxzOx53hk`DLwWH<
zt%hq%T<y|xqL13^+m%*1N00dUtMfre2z1}}2L3t%SRFlgUJxnSBq>m)9}}7jLgjIy
z@6<pecj}AdDd@AXHRdzD=xkXgvM@j#u2qX-)S6s6X`K=A`{7q^c-KYc!6mKLI}=v?
zT|&t)Be3W>X3U4H*nXj3EpHPdZ$x`1pKr>#1$x{&eZp+c>@K3-hFF|$p9w*lqZkFY
zWfj?RF%bZ`CB%3@E@iMw44$}30YW(*JYFkUAD2c7nB|J>%DRFYQYL#zeqLl*#b49d
z*jHbgn55*s(CmW~xm`?0x;q7($f`GKdHKv^|HIMd$zR=qZ1<p0xM%oKM;x9kF32%1
z%R%W(><eDslX^I&H-32Cl0Umu41%a&v8+lV$aE$zLXu_O%>Bt{=>eS^_aOE=8$PJI
zxppYOaeZSn7k>(f<+BgT)M6)y)ybj%0F;8%Th73tDcOn&g@Q9_H-J&7Fkr>;*Xjm4
zK3?8^@$Yua(|POq(2K#|v)jr{<OQa-tsj~9&TOCkS3e0RKvRc5?ni&~0n7>^8Z~zE
znAzgC)LGi_CN<QJbEI?4UZp&^D9?mbS)&2a!2dk(k%ZMIcAv`1#T_;X`@Vk%;%I|X
zSS$J!rA=Le1%SU>>FYh7W;{*W;Hlx7vNvzH7-YzL9YR7GdheHO@~RRpJQRTmxSDdn
z1SCm7-4wty_JaE!walec>>A3ExxKkKRHiRj$BterdH|SlS0Adm!&G|Jd>H>tGRaW)
z&v4z?<5RhsY@-O-*oUUQfrc03x|0{)V_fOS=1(gGr_wam^2C2^OT)$0nCIrW=_3Nn
zg-!kJLtX$#Vw;-1gi1iIIJ2Q5#wlqI8=|<WCgSkl2kLRBfNNs46Oz@nHMDu##Gl!h
z^-4ynR-)U$A-=%@rmPA-zgSv|_k-s-sf;!4%ubFYIv=8p6qA#4*{tA9Q%L$9B^v6M
zakN#$62Ko8DrvIc^Vgl@6loS*RAhl=8K#&#hTe2((6y_etM~JW*6*r3k+1&e?*URD
z_*^Psc*q2fl;uhhhtvU3-Z=UR?nR2dKsKU-$KXXO@_y3I6OaWjhEsfZ=W!B$cKPa6
ziSrBr67c!mqFN|t7~jmYD=6TEBIGk9pORLQG<dv?kzL+K8SWiZsa*EfW}4WYw0XGm
zGgU?7|CCLv%6#ro#aMt>e2mPt(EV7~ajC5yV86KN`$_3boIMs|J;g?Esrzr9HSfrY
zx64Z4J~rfRJXy~nQy4#hks(gZN~U)sO|`9mrSsF<CY7D?D%^2o_u3lie1*O#f4bT{
z*QZaPR?pzBL2VNmvP&}`b9s*MoW7DpIws`{gr^^oD)?bs?I{kM5(Q_H;wpS{sN#^|
z`E7{`v@s{u3!Ozj^fYpVQ5ydB6&LWgL^w|y28^JKTb8c3bzO|6TwYlvSPY0qn4>SF
z+3mF4`-EmU!`|LJ;L<X1J3ba?v&9NBjm-b5fHXoMsm>rW5_1eT<|I(p!fU9qG~Ica
zGeM4Ewsj{3BG+NB(J#&)OGTiK2NNoObPZ(vBS<$*%Z<M8^w3hJh+h)ko78PBt33o-
z_Pjxn24P}N6=DJvkM$t>zYTu^KYiioxg+ndw)69_^LiCV8K>AP!=hsI@cwc&L%&)*
z%g8aqHb~6Ms-pw0QWRPqSb9H#WD8%TfZ@COSYR#)a>Vo&Dc3kh<~&43f%8Uim+@yx
zfw8A2IETKdBpeg2b}_A_WD>-dIl&MvKG!;p<WJ|2XJ+{4DBTfcKBy_54U;d}Pj1`(
zM*}r}qH=;lT?VKPnew2^z6G?kks`3ILWO!K1-Qc6HIR{Whfe(!fFH@0q@Y42%-C2t
zyyaW>{jCNEckqXCbJ_yZrPz~o#+l(<zUQm%3(M(_+n|vsL*GCap9?8uQ&xVHxESKt
zB-+Sup<PlztZdFJ%9!sLa5fQLdj(H&q;1p@!&H{kGG=c^iGD9EZ+!C=ZQZphhC5R-
z<5P8L%Z-iX@EV$GTBdF9(d5LWWbnIHuPt}(j|R`10l$6$`!|1?%zJ7T!6!N8Odn#v
zFYDOs0*)H`__o&)!%1TRc5GD+FmN<5E)~b57so1vL11~FfJNljKLiVvG$RYI)Ng+0
zK2GD^^)}xu4}&j;vAZntIuH#YY##ZDL%1b#H=B8%&<}l`LRmyIdMfRf-6Cr3NX8%H
zyyCQ5x+Xq#+{nOJzI-SDc}L{%>H2cd2l08gdN4s4;DG<3PKkV9RyBnX92Gu?;T!Nh
z*3zZ(v!>)1*8GC6XkSJY%F3A>w|W^z6sV-ZG`t}Z8T`HD5*bQFDFwS5M#J91xmJYY
z4w7(StZ4J{cyTN{*1#b5bnJAQ*lNIS9WV*k`d@#RTIi?rITQD6>Nsq5YHBeie(s9a
zguoR(@|c)mK*i9j(q%zFI7um08A){8{dmGPeoT=DkCgb}83qsxE{<(({qe&kdo0tJ
zc-f7VfB99c!Q{`)1iI{F1In$1Ln$Zb)NqbV?*g1IZrs06C(aye8$|T-tar3oZl)LS
z&nuzHcetkF?pBF$U;$+!GZZdJi<%RV4at#aT&HU!f$cJS$anZ7AGHtp*xDFcRD@Ie
z<%}%wY>rFJ6)GcYs#k?Uv71d~)?zyNClo-P==_TS)dv{2bB+G{iupa;@)r^pRB3Da
z*Km%WSMQ$iG*YwiO0Rib@e4c&q=7`jkV^zI#qzJah)sW^<B<T1PXiVyX?Sd61qHB4
ztXuUWFqu45LNB+|OoatDp4OifUe7V!f?|(M$1~TqLAKh4B7Mm-g#Z;=Tr=>M3Gu)-
z+f|^C#`RYv`8J|#)4;1n&!hTIBmVYrI)i7z!U@5s=>!0g$!tJc7y#_yP2n}+=^S5s
z7N}8xXU0IJf*V8z1J^XkntB2xJHE;{e4{H%=ancT;rY?lMO68WSZ)C@TU=i@6|U6L
zRdLaR)7%0|*Zp~p126&kM7M24lY=+lHeuDT&o{G-?-yQQg?a4PaC<=WiiPUt9L9%w
z@-An4zV7H`_nWWjk<b4`5EpJnS18rpc|E-lCxa_b)<r@reiks-qPZ(fyydZxYA$41
ziucNdHbhhVG5wWI4OJrVQ#Fq3E|Jx-mM_&dB&%*<-&4=@h;cLnXd<sGTP^W<E$cQl
z<R^Q1A6+nGrUw^QNoS*dWnsbjBUo1Rl_4S~es@=6*IZ!RW_r(jM7+C@f7gj2l6^>s
zhc6LLu;wC|JcE@@O&Ux(Dx`$$8hsa-B(J?Q0F^5>CHp1jv=N||CFb={6K!v$X<-p@
zU6UWZ@|)s(!@9+z=Z0pGo{^)afn~Wm>D)`C>GC_Ki6RW4LXr{upu+_R@otG+%8H1w
zt_g|~n=7!NB1}l^$DLzmL~(Qou&{8)&*$2{2r_3c%$R5`c|&E7$k|7>N^>3p0{oXZ
ze-1ly6mtYoB~^_6CWuF2IK96#QgrgLwb(&{hIU~c5uSn4P~@=4Y4Le9o4puvl2tH5
zCB1dr?3WwtOt$l@tO0AE%~>~=#5l>NyI1pPSBZ+KtQm;r5#$N<ijby^P2$nD98V6J
z=PQ0Vn%GISsi}=4V5JXxBK^V?43C?&9L<T!!IjuaN23%CE$$Wc*!v`~(%}<f^KA_}
zHgC9jJ12b8l?Kh=9|2z#Y8!w|TwOI8*UHT8q1^9jP_N^f`P+JDKz&tBizF8#AabtR
zX_r6OvG4R2@AYuPFRC!Jxt~t>iVITRo25Dx+e5l#bxWt0zu+Zn23>-&lhx#EhvL9_
z;3v;>=m|g%^W7I(n#Xc2$g+zq;>^O-ILdA~PtppC8Dww@%0HAl@ad=#K;%4mg*POq
zi{cinooX%uQHIc%$GXoVP=wsnaIPL8i?%7<Wf=ulI12c|HY@_1i(?P6A6i>UWC#W}
z%`@!15}=cEe9viKgN;q8)g2wb!&F;EaVTk}mk@`gG+^}p7Kx&&(g#Y5$LCJ7lj_WA
zCF<yqia~hi*`r9;$B`;NQm<%zCagp?_@t<5Qm0YFfG5UBiEgrw`Y`<+$tF_P$i=NM
zPq2k@K9xo_(cd25T8T3)uI2{%Mc${xpW9P;IOUL8_E=Ou&x>f<Madt1UtH^w09zUe
za?kKK;3c^i_8yDF#t;d7p^m@~xpU2)&)}M36UJ0^e}~Zq%fgBX1Z#;@2nQLj9HwV=
z-wO*0ORz<i*u~7a65!tee;Y0F7)D~c36XVSegW$IcjCCG;%GE=fbsFN`|vs*(y7dM
zOwd~$o3pCIMWpfL!m1t8A&kjx@<P*biX9DcSiUmC?=Nnn-*4vxK!(WvZSHCsGu&Hc
z2r}1Vl-Lz`R)r5`x+Be#R6)^dDAB|Bla(hMrUP>?oZD-nB%mEw#W1bUNt-?r*>n#s
z?5h}UH+!w##}9ODpcrKdTm}e0?ZE9)(Fr|}8oQKPBvkQF!BA<v#-gPuqQ8xWt$|;n
z`<gF;I*4`~X(Tp)!y?h+xSQ`C1P|Qal1i~r2K^l#M&2!oNougsG1N6TJFu!)ZxS<B
zc;~|c14$Yh8>gNUP@*zSy^aXiUrR+cu5+WDH=k>_T!YCq4MV3iiNRGqCIq;`b@v;g
z1(maGEm@2*@6pSotoeDbF|5T1x}@{tbWIV&r>cpy&Tk8>^1bA2Y;0gCqoKU;$j1e<
zZ5ihwO(x#m4dr1teXU5J^hpq;#dzFE8b{|%I(24UEwod=Icm$dT7j9pnSBN3rcQJ^
zkXPRQ+&1>ddZK<3to-=rLo<PY<ffP4pyS@Lm5aaUQIT#`xs5e+s$AoofXopjnl5(v
zR2Oj3^xad<@v@gc)-lY+#{0-Wcw-Nfgw|`tYzj{cGJ<5L<M?=OW#UrzBN&5;?U9+x
zLPP5%Z5*4bBr(`g&N@pW12#u1ppacR;q@dbksgNGev{8TCUaoJl>&9#v9WV#uVCh8
zO1d!5{1J`HQkB7wnGGy+a6p1|{b`<?xa-smX1c-spxbYsx8Ve%(asRwt-Et~{VGTW
zO83%%I2X8sLMyFK5CJQ_O%sPL7rU>E`wLiH4N?bukcauEso4wA(b4JRVzQfYaTcy@
z*Gn#mJLLQ%k|YK{beC}Lk4Y_*=Nu|mK<O7t4Wu+HDBen@SJa1dohlqKWeh+@KeJBx
z`P-h*<h%$Fy0)1jt@b<3lwU}8y~kZmp~tlnO8{ZfGt`5u<!qHB-`#!e3~4r9fWy4S
zrn==Xy=O5~yhj23BxTIq(wA@dk36>oQM4vW;=)$SL7z?8v(rOxR1JR3=3isO4ym8Z
zXV=Lf$XF6p=zGSJbv~1%{2c!7nOs#TUCc%1_ezORCNz-jOsA6~*l#<h`WspB{PpQ<
zyWPUw9lsPUb|^?@Jkw0@ouai*Oi!Oz<yX}~L*H}l8GohqyD03O@H<y|rK)2Dj%UTp
zU}-2%$r+QMsA1bsHB$k9;cJkYM)m|U($kJ4VWStQd!HBW`TOs_xK}z)dhjuUI2iF<
zC(9}_j5OlwI{I)wmLb|m=fvIPSOxAdAmD|8a&5zys0tkKoIKAwQM=?wl~x_L`_bIo
z-IFJvt19{v0#mc#$b%l?4Sd8&lG^}`C{kP0QaYSSvH<GG!pZEr4gq_S2$LiZ3uROP
zmYdme^^-FWdqTGYj=g>(INog};bFn3EX26a!{QtjLDgT%WGG@j4ga2G%o9gSrB<Ts
zk}6}UE-<PQq}aMx+wxWjRYpFJ!@k(w`3i+Mk=?9{i@*~<7EPbV_C4uUmm|`oOG{1r
z(-nx-veHuYxGoi?lraz1oN?wapbRmdInqGREoMbrb1Tybh(g)S>GS?a5jqk@f^aMu
zc=9T?`VK;P?VdJs%5zUA=1W^Ot<acI12sqp`0N?b#OaoeaPQ5T>9j+bX~Y*qky_yt
zX>q(a9n9p?!)ZS6wcd3XgiH>lc`vwA3Vr(`6{@y?t=^Hs?v$O+F68z*tW4{V0Z^>F
z0w^JniL=x&gXeO^@J>Vi6z<{Tn(}f%^`OhDsDM&#?+OiDLbz5h(4x3rq|p&V4dcQi
zrupXOvu{?r>b>AxJ1>tW&Sl;XV-#I9bL#ezp)6Yk2;bJ%H+v7l1AQiTM|c)n+DE9M
zv<F*IoV}JjG1V-k0r{PkEC8pNS^Sye?{F9^M6NS~pYh-Pu6d46*8U1ijJx&qUuEtr
z7G`Wr)zU49ENsq2DxuS=LmKMJH;)%*ASKu~Og1()ig$%|k+Qb4KSD4mbBBW3-7Bkz
z3iO)Y0#Wc;w!Fup3|~pVt+)1WtN22Sa19QA<j4u?bwsksUy{cxlX@GD;nv|$WpVW<
z(0?fsfCpFOfk`;*8h~{khNflI6qS*4e2#~<IX^zwEwhrcxFf~scyne=_9IhmlF&yo
z$PZh)uq58?|JhdR_1z}SDsN-0{j`&2YGDzwXgepDTxT~Um-<z8_zrCey|XJ2MNV+=
zLM&sa`{l#OAHVweT;50&Nh73>#j4yC6imz17UtA>RfvT8twCph5z_v3+hW#c@%M&K
z8M*WlN*UvVm6p`x+kDH7I2jpiyP62yvy@FImrTl(DQ4jyvgEQi_(+rao@DGr5kU(f
z<PxY1y*bp4G$jao%$cJZ5tuO3U`|ck2?LLUIxCj(nhahgsK8<oaZ$J>78YIw9oT5M
zy_IJW0fEe+yk&xSYty(%ZB2fm^NHl0f!VA*k%4U)rzPq0-#r_3P|80l&v95|a~Y)p
z8JE!n1ptT{?<+W@pprD7O6ld})M7Xa^I~Pp=`aL%Wqd-6t-H@1LK-x7yYrK}(|^(n
ziC+RFpuG%uKjm5c?FsdkvqB}$raRvCGPH`%s)Ye`9x}Q-a^9VNU%>rtU@Z5VD-$Mh
zeB$V*n~jv`(WI8fHIJyD)@;^6T6mJ$pbPJtM#G){dOIyGt%sMlTd$Rt>VeiodxzWW
zWHA7Po3yoT>RzTY*4rdl9fTF%BRN)ox3*J0-w|bh$G6Za#Zk%Ue9LD3X?gZjhPm+e
z-q;Z_@$q|tB7jtDX#Og-NaObkW~X74*Bnpc_GoZxX!bop*40h5T@&7TtmpeT=(M>N
z<o2Vv+BMStZ8!6B2WJ9DiEIWxWs!!FJH|y`;LPQoBwv1%O8$^0<L~7qVLxO+TWIKT
ziJ;-mo?T$U+}#4dyelCADtt=`W!9sfG_mcn<u<jlir?hMF@>*$4Wx|GoCmLZOb*qB
zBq|xO#+2s8nRBOJoC+i^F0JGw&nw<x@GgiL38$*hquJr)jb^rq2^DQWJixoipvhs1
z#Q8tNhOF<-=Ht*1<@JzGmwH80g)e*tc|tC~Mxh{oIPsRA{KXezFNM0VF(q|99X=o-
zuui8nk|ywkIOngGQIN5vH^TrGfI5W}xnh46N;=7fi0dka>Qu@*rMPfg{!uP&uO*S+
zobgoS8(=^c=a8G^iFmt(=*YX9PT3=cVMEwHL@j&(T$o(O>!B*`8sfC(Bhr81-z77b
z7VW#%|Cky-JjnGzq_Gc^+8|w5tM)4dqeNE_=qlr9xV_LI;@$8<FPi8#0nHIept2jo
z8i>A<t(V`R&We;;4%h%m^2?4qE6r##03xN4WC<WzmMCV7(nH@xmm7+|sL009hS(V~
zTMiBU<Je-8KP>|WevcT<v4FMYWb=T_SZI>e-hs*-KA@g~Jm)tYl_FI{Dl+6s#pwq2
zmq6ig?S+6SHqgx~Q_XlD_Km;BH-*oFQ^V8|wN1YvEdp-08KSF8OUMOHk!plCjM6PT
zqs?5xuNM~gvk`yZPT!vF+#0OK5!)_v8y|M)GUSG`2WgEt>6d9>p}s$Rl$fEP&A8k)
zyFX0z5LLFpi$zz5hv(<hluT0+vi+V~aJt;#mv7W-zC<rnPM0cX%l#AIKyAj5D!W?M
zQ9ZgHCD=XwDl8zde0<tI(&o4iH|+S`;*GHZ6XSZCNTk5ZJno-7HVc;|b=~YPsCVIj
z5aBFLe>st!lUWte^PS!Nl&ItaP>z?{$Xs3TnNEAGI`IKalAQ$9q!NDEf@<sQ8r%Dk
zm2-JLJ*Dw_wp=-dAR5?2a3hy;$_vEd{Q37pq7QdK;l|7|Tqohyzsg6KPL1t)n$}a#
ze8Yl!_AO@fpATG1)P}15wfO7keC3BvGC@Kgw;haojVo`&cIUzSNMfjZkeKTVq<sJ;
z*!$BIk^{S3yrB#yX+Tfw=QBdspT+;8*-`+J)D*Gf63BiY_~-B-y0Xadn(F>ES%}9N
zGByYOI*^;Rr2_Mz*X^qINt_^QDEANP?17$K+FyfU*%z-~^z$2@Zu#$`bVM144kREN
zDl}l~Ea_LUsk4C!@s4DSoI^az_d5VEW?TR!Ume}(<D-1`^^4)vyX<kk=*CMpxhwvm
zJeZB(U$sq7HOy^|MD1Tl;aQeqi8xo$hL>vYjBj+wQtJenMgt>_Pha~TZ&0%Uar&TV
z6zGOO;Gz1J`*uL`EO6IVSrKd{_J^M{B9I8->mE8cvRoiiGe^wbG<vn&g%Z<Us}|>x
zxFRUEna&AA@+nvQ`$^qlIo|a{q5Dzq`pl@uE%-{$Z@s#;J)%#|QsKxN-F*374j(^o
z#_2chb5XgNBbt9tkQSOOg}r~HQKREFB7Ec{Q(a3+Bm^e_rXl!o0kTHInbodW$faKr
z@`n&jmBp_2_dVZ+%}52@gKu`Zr2CRe0P0BrKpkvQR2MsQV$$inhGyU4A>liu{AS>T
zXZMwEe$Eog)a5m&C1YF0;_PLE6KBlHXt|`n^yqJ0mqKeQ8pE;!Xmt~1+%=*HM-@+_
zExMo*nn_ATZ0w94$%di@V`XFgB&A_I0dtj=WzcSgmfQYxoe5x<#9eF5Qrpe=^N+S!
z5E+lFKY)RBv|A(`X`f+7Ov*4?dhA`=B_<}8{Uxc-QF)XH#2<W^20@1#F&-8^Nr*qM
zK*0We{(XaL#O|@rU*+mZo$0S1$#n#vlAQlMFZfL_MUk8xoO3d9^vNHS;DdG-(9rR+
ziZPNVC_g>5LZoT7fk@RDbe8G5`eA8OsG=xW=E}GvS|*jqE5DffOk^oh8W{7{B<#~w
ze%$E?EuaWt(z@M=4a*&?y7{5|V)H74Q!b~$nD!IBdpl(SswE<0c<iSn8qLs%7zPQ^
zWQj5E#kg^9w}*scxPR+!lzi`yqSQ`gDfs(GLDuv2Rt>O0F3YNth!+kTCcEVnz1d{H
ziEVPYh@>j0)K&y0_E#)^x?Frcip0QlAHZtajRI`3>3()Gm?E5(gB>ToV=O9VMNklb
z7@lSlUG;NTPFSF}ZVkmBMHr74L3V?g!!_H^@-P`!X0)Ii=BJ8ezWf_@YD>}gv3^op
zH#*fEQh6-2gW8b4%FfgrKYN=BlP_3aBRac^XeP086%LCxD?1)c4W94mJfSB?5E;`9
z&s$27=QWxN{Z$hi=ggb>f66NTUhDd4fP<7v^v?ZF_O&$G1MeQ&j-KNUf4a<{!RmBc
zj!HeLmF>f)uW17+_Eoi>fq{Vpa~fq)JVp9}Tq$mQ3!OqVk`q5#!OUoDmge=Q?|@>t
zkvY3wgUhbWzN&%S#J#eAA*R;i+v4*2W_jb<PHbiZF=9SdSIBIIY;=26jvN(!iG^Za
zf`Kz%+qAR;mfn$NI}LdZT6Os(KMWVaP>}|RESHnf&_G9ysIv}d_)dYW)u8t4HyJ!B
z$M@^oAXcm|jyzg|BY6;YP%?y%LUrD`V1~iVt92C0k!F)TX8w+o_pbQt)aDY7yh9a{
ztPG!v?+T^3FV37$5fiA$jv(<ZwK>YWSNtA!+_cH%j&h&H36vm|Sh08$?RO~u9L87u
z;Odo^Ql%rf^@)ht>a;1r!PE3{e|#!ArvnDw6)Vedabwq@jfM5)zE)(sfgN*nGYX|f
znmWppM2H;{@UbCBbQw2P)g`-qe}-AW=#amkDg0P|v^uTHfxQ*#Dr$G`F|kh4;XwB6
zZjL=0O`&0ZP7Dq&_E8%|)kUU(djH*lmwIGUn;I#0fqXmUO6JLaCfRkp)FZ+mJ8qBq
zPq@3qfOEdIn>H275$J9~zh{oOs1an#>jt20Rpm&nNh9Rcl02!H|KwOk(>A4tbb(Y(
zr(;^9J)pf)#0!(&N={$8Y`=vZt$qN?K$Np>!DS){9d<wvRsfbKR{l9KoI!6o<;wT(
z0vB3WW7E&yr8au%Go$&Pb{wfgPq|f6zv|_9g*4muq8&RwI~|Hw<jeu6#w8Jdk4~;e
zZ2PKY{djWAs!~dEKmZZBx!>MTVX$1ioat>DO_*qs!;f|Oj?|Z?3O?*mt$r8O8=6*%
zw59%}anOAg*A0@8ObN1I?-=@iO+WWpuen+d$n1FAZTfXNM!~+b-=$?@`toB-!o<~&
z@z)cjG<)8pHFFEU#QNAmzNE2cN9TytM((MGzn&APr|7Av-}rof#^+RdtiOxGJ^xD2
zF1iPF#YP!#KB6X*@?h%mbm=W`=`tuaN==<`cz)ntSlc#yj#%Fo@Oxm)^gN^<QBc1v
zY>ZVDt=vwCEjGoi%w_Z%E3&2UbH;C3e3zo=&y5(V@TXO99tR2e<r#_x<waceIZ~At
z*VoTPS8TZz;Cj)7d1Ry7zsn7T*EZq%#Yy}5JiWBH1|#+IL#gw!y?SN{P9;$Fq569<
zzGSMy+deD=Mlr%O0W0E<X-_0N#N2_c06~u^S|f}={x4xWbA-sHRa*JC;u(Q2LM9Hl
z(m>6jzXTC%2|4sJcg3t)I63_Z)953g3rg8fO8nE$&^(5ViL9H!07T&WuRbG~*N=@T
zWdQdd+GN(EG~BN5Xd5|~TAGp8%oGo;-Bz66Hxu+LKk$zo(_#nwM?8LYcm3Qe)Wgf8
z(GStZ0`HM9(@w-Y{qvd8Mmedk(Wcz$83DT66$ho}4j)OEWll^qKXB^qaKsI`mQ9ab
zk`vdqOwWL#W9z}4YJAx8!F91Xk`24Ae@37M#m13>m?*Xuf5>D*p5f0IIk+R?#W>>I
z_?*ahy9+m0SAE?TL5H1Vv35?m&HHVyoD*Chr`Zg1`1^1<JxWRn_7P;oikvU$ug%<(
zW4oY2csKXNHF{yNH3lL;O|TN$X6bnNVBU&+k~rzFm9T-AxYjX@-+50dMPW9cgCFZ^
z@%c$FJTb1rMRARhffFm6UY^u*2kzV2Vs?hg;_~w5<Be{e-Mfi{@ck;cz118c__}ko
z(&UMU02xLC>AGPw<csH9HMw39Ca(KLERNg#SY6pU{3rC|xUQ_6$c0t$UGx4*2)bH?
z6#NQoC#dXvMeckgn(kz4l3nJ-&07gp++_qoEmITb1hOmf)Z`XC&40aT?MRWgk<P95
zk*B7X-Dk!!CO$<hwk7#cCJ?MfHo0U-rEbAf>pq7CoAhREw>Jc<Ts+Y*ukKm_cj+Eh
z+e;gfX(-cV^74#raA$pG<-4<U>~f=hZENeuyBNH@ynK9i!bEtvsr-nA;3P5z@cq~G
z=u1(}+pOp}=rouzX!eX<Gkd_6#o>^=+4{DPS`6FL6Y`maIN@(zm8wp6Q=ZnPjv2*(
zP8>G@YflloCECo3en9Ee)XZ<0xcvOD$YnP>{VSLHcuTmD#DRk=uf5U_<?m{{G6s{p
zLNVq!#fE8bqtXDYWqGVv+#P>f+dA0zW}}4|&O9X2a>Ar)vcT$|;Nk2v{sX2QM2&%f
zrWfB8?u2@`l42?;rR_@)UpNJ|xTH1~m=&3Da`P2`>YZkhwR<umPB%=f%+{Mf_C{@T
z`rIAB&W6R1OD-iQ4`4NuSJm-1=#FKCN9q9`aH<i$OSw~)K{hGq*TLi<x0!=?-sJ2!
zO=T$*6eftyaGlaQ;IJNs)h1UAC;Ll5(T$7oFGJiE3W-;MbBt~pj5*|&xp2HP_EmG}
zQ6rj?nn#hog7H@{GyVIj-^KsHq!ainW+oZ&>BAygN{WffFw7wj1)H_HH0HD2As)in
zS+dGL&ld%3Z0?mW$5|);A^6k6pBuN^f3rO6s=&qg35PAV#)ZznqJfU-CpchMr3#0j
zm$l?wP5X6)ma5Cs<h#v36Ue6`*fKipL7@9Ni%CV~igNn`8%)BcV_*R~o8d+SYmK7c
zF0^^3x4$g6c^;s>J}$54zfKX^+MjH;oA9q@6HCp_e<=g!eAUq*sdjxLq^Q%T3hnsj
zH$gMqfwQgwl1ui5@It3(^GdR0kBAjUzB~`SUA$$?8@{o<o!-6auC+gM|C8zo-Ldm1
zIud-Kp^tUXNOMT0(^u}`1VQo!QG$MaXj`xD>|{Zj0=X;3sv&c*E;Mgm*z%ilr%yL9
zDe1o_)EM^Wp0Fq~ee+YgNC7H;A-rbQ7x$3)x%Ktk6*P!-=D<Q=FzUWAeecaoq-`0^
z_adx;B}@9c0?rg8v)V7na-sMRbaow~2Jk!j3d^!HcKPJNF=>T)&NNBNVb>tj@}0;P
z*uxwRt#`V`M#sWHZe0DmDXyx`A>&ha9&vxPZj7;v(;9$%I&k0<`H%etdZ+ucR--_h
zoXfXW(^an^++Z{Y(O7jDXKlVtX1P*oVY-``Svh<b<f@TOuX}CZVaNYC1T&b`pmBKu
z{z2{kCf~GrLWZISQAyhL4Md7KRj$%Pv(!qN^f7a3vm_7URS3)sw9E69bg>C(si%w)
z*pSc1jTd^#u{f^cyS}}y?8t;tVM%H8(;sl+U{fXPNJT3DI~WS~IEbP2EHi*Hz|2aF
zph;5@Sqp1w!Y-V^{jB@hGMzXuD7clf5Z7vd!7KKpg4bCZ>O&2-Je&4x9k5l9LR4nd
ziIcKI*$-+Vabf7pxR7+bsOAurH$p`}J0N_OR*Pz_+rV$zF6!GaPAb`>T&rJc@{K4T
zBkAA9lCK{v;F1D59nR|PT?MB$x-Bm7N}U)T4Qop0VnL5Xb5e{H!P5I@^^yJz{epIM
zaj|%FJj$1+@pnznuXYArUa>bgc)2Zo4O>zf3kya*K7rkq3?t~IO2^pdW-oLtdUDeZ
z%A^t%w@&2}0#=qHCA8jKIGJ0`(2U(nfwObEL>6ZU@)wTWAC7k4f3o(_xn3L-?IM4T
zJKah&;l-nTm7Qnbr=$6~esVu<_=X0Z$jmQc46pfa#hzUz|IR7pU#|?B>&|nfHvXuo
zZD(RqTM9O3UZ=P&Wh7Z<Va^h;-ibC&Ec2r$j_yJ1M!!DRW6p`5mk33-ij+DuCTzi!
z@}2mE<*THN2tt6M8);&npaDo2%EhH5Lc?ofCdD;($5yLbYcbf^NtKGATvrciL*Es3
zw7xP$3a}<#DJCS);)1L2mp(7lh%Y@LMp(e;qB>c6PfYHft#P9<yK~A0bIjwq|8=CF
zA@`DnLqX>8s6?G~QYer*XX0k&$$s-k@xAt8bkU`NS(@xCo~E6a%KMn2f!z?k+L7ut
z-v8WzA=$kDPP5oHkgN&-g`VdA6uT>K3w-#3<o`V@$ApoVt+$B(rF-R9capfQ00#wi
zLFtd%i*hP>IyW;7*#Sa)E0A%KU9>B{I6Mtsnv!h%h1|k7n?NnNU=s)S0YxW&SCLQY
z?(igtw3sF(Cd!Q4gy6d@Jw=E!^<(rBIpd+ZvIEmQt*{r*TXFRe_p%6E;kxx#18?^b
zIrpDte=O1zf(rzffHk&`yIsRg1`bdh9aY`<B}SLd?yl{s?&P?F`AF#2S4mSBj2BqS
zn7!>7l$4LmIWiKyexE*ZwOi>M1CT252dGlRS7_vD*1l>OE>c7>B(Ja)gy;CP(KG;@
zUmnq`Mm#u;%{<Y{_tJ!Yf=kM@bqt_$bLh(gx^bW9XsIfif@h#)*^DyBr>o``SD#aL
zQ|Z;&KsxVb5ShmI?wRcev`RtW)>gLW_TuXQv31xL^W0gkl4joLuT!>&F3*V%^~W79
zvN2|(q=s#Xot}NZV3RlOl;-@A;d{wIIDftt@6MW^t(g$0V`N52Bo_+1&TBryz=qDa
z>e6r%v(JF~5qEAlAPTl}zn6$i@=Nmkx<~rS2@|ZV6ouLc&Au=kI#z}ib4a$GZiRkp
z|K47KM%y@a)>UcwS}`3E_S>}O%=C#K^4wl~x4KT&y3E{Y!Wz8RG|p@=+;adZHLjLI
z{~`<%m>F6Et4n`0wLiNyU>yodQz+-xfh*qyME*JCSwAR2u(m@Kwi>6I`{tw+w0V{N
z7l&;vLfh|hE~MmQ)jf{qKq7@L2O<YCakIbDLZA)&o{z)vXWZ?w$how#$N!0y{!i4A
z-&>aBEZ*R^s#Zbyzl&9VwztC=wE-~0sFE>P&^(@hd$48s?`ULl@fJKXk$&|PQ$YJ!
zm!x5<+1|RrHJJoQn*VC!Pw%xg)%rhpq@hRInQ<giMP#V888}I;dkZRfy780)Zd2Ym
zU+<kpqE+8{6TjRNN6kSj437WF#wTZOR3-5ZD;2-Oo+!Yb(@tTzf%uq=HC-urtN@;}
z;H$QQj=3NyhmMZ4ZeSEvDlVJ}sy>$TJCAsh=53Jb#GLMi)HZU5f;1XTK0dJuc^5tu
za2A$>E&B#>%?Wbl_0qyV5mCm<?KJv9YrRkAg(bKMr)YKHO>3nU%qeU;39=D0V9^A#
zf^zpcvt{CPx!S(fDJh_G*%9&in5WlTOkkSsr24(e94ivN-af5wg9|ga9_TVRwoId{
zd=KIK0Y6rnhL0-Sy1T@Bmd^~$yjms}QKB5oO~JJmWSdzX(wdB*LWXd*?SF9il)IK-
z74PRhTu_;M(@2))6dU8zb%qOivQ>J0L0eJr)8CVHA#NUbkLR$6|EjD2J4dPh9ylI%
z<4uj>G=(Rl{Pd|+{Gvcf-bPN6&fN7j@1jV|i`-J2m3@6kJmQY0<0}2Es-;FzC*1n!
zJG(I_q2ukb&TvgunS8(L5cucS7*Dh3747?-=)h&Ts|)r$suld;|Blmoa<U0vyR}_O
z7yS`^hxPR2%*z$X56f3W8VtsgzY*fP67<HT`cT<cOpuRIMW9gkTNQvS&6p7B@<-L9
zm}D}%COYrO{3G)aue2rM)R#C{YzAZrAY!X}15N&&rMeMe$bS`eqz~@1egfv7+u$#0
z<mjomGkXbt4dfGo=~zv~<!7#TJN$jb4(}@Q*Zcl<|EQyh2H8LtRg>v?O^O+3O9OiW
zJE^S2?+LAbJ?P{^XfMfss#eHIQG^>~hSxGV(Yu6<p)K_A`j#oMvdV@q`t*>*(~+x}
z4;t6^TD=)~Y99IOya|}+2I6aW%X_w9@V<<jgT(pE(vY*#l%Yg)g0Z}-b?i&y{%$kk
zpAAqz;+)}{Ldt*(<I~dITtYOyQqScQWA0EA2XumsFcJ_oFum9-Fq{)ZS-9a?o@=$F
zJb*{8ts=?q=#Txrks{a&oYLw%H~RVm@5eEA7fWEdm7YsJAPE7aS;Q(aAgfdVd^m5o
zlh}Ey@Rcb_br?H3%w|yfGO!);+jX*lV^E%^+${H@m&FI1X!uC<W3{P!d<}9z?c~U5
zVrm)~ACD*eaDab-n<nIy5Nk7{A^ix$4^1LFe)rVi=>^ec_+)(1ZH+58Rg3i8m3eon
zgPqHmqmmUlpJoyOA7Y2t=YOo2m~K--$Mv=8vxOfQ@NqVrec@$qO-VNL$zDPmSqyh&
zDxZ_0I?s7yFM_P~syp$l(h8<mz29Z!s0P;<l<1PIpytWmqK`HB*q@OojB?xD26<N|
zntYzPX98b*O{vthNBYBX<ZMQ(VumkR^na$ly!g1eH@-vWZ6p5tAD}cGsh1?TJPZo~
z;U#~S;xzS|!QtKEcl7S}z;88!k1lHXKxi8GS?W5Ai3d!^?%Ffl^yyNlCaS!q1F5J+
z(yB|$wfti37j}Q2rK((7kM%{P1da8%(u}w@#%W3d=7(|{>N;!AavOT&2qs2b9bjYv
z{RmbAeW(Jgja}kAgQD*AuMtsH<88$fOhq~dSn_{X2gPX)YrC*g;)fVE0rK;ODtY6X
zQygH#TqzR^^eEmUyr@-?)sNI9$d4Q=)$R{0|A;PE+#CEFU{vyJv@#DLXBd6nicbHd
zjF9X<;)`T8wg#RETkl+1luK#SXa!mi1T`41DPKIPo7a9c?Y-8J5<CChl2KFJIALp`
z4w~U>u2K8G8jc$y7DE3D3fXxKl|}{Tzzs3Gpt~ZbH#Pqjm9-)n&+pjHK4KHhoOs_!
z{AE+r(l1}WR#sG1S5M`~dq9!Kanu=+fImHmY^o()MyKG1rqYxOXI;l`Ptlc@r#Z`a
zwyh*|G)I>+#bm<Tm;V4r)&!85aEqs?0*HX15va%;Zs%(xOHEtbSMe-HlK{D9(RO1E
zl96U!*cO!NRjiK`_~lLK3-4i+=k>$uk>7>ir_L}m;Adsc>JDbOWYUq`D7~!3^>|_D
z;atkPG027xol5%NXfnt1sl88N`>OuCAtu>!;6R;@w+KU9x71QE(3#1M&E%+H`&-AO
zRf%<4TL~peEp4eq$VS+GMW_+4r$B+IO^H+d`pcLKmf-Wq`qQ_Srs}@_{*RfjqY>U`
zEUz(v53u=MLB03m1)oNtwQ&Of+#<xjDAh$640l5hKGli;I_MI<X`7(1TK<@i=k?^Z
z3N5dwFe%?X{P$(nbro!wN^OcQ-nodzQ!g+egERS(7Yf}x;%pPa5L&>*Zy%3Vv;L^K
z@|Q#CYt2K}r5%3)N=${^Fk25C4cpH&^`L$JEDGmUe>s1Y)ReKp2ON{(PUKnF8My*_
zJl44!J0+dp=cH{Y?`}!11nuySxCWn$RG3WMej?I4%-Q@iBMp}5w6hT-Rl!aAoLn}1
z`fHnm^F2;@J2E$+5Wmt_1O^T!qQ<^E%^rb|6cym*UX4A|<AsVR;{R5ER}mCMFt*a8
z81MPL{d`9Of#F)npTonsrBgn_I`DGB=2L1~<!4%RoCO`D*o2W3DGZRcJrh#55n_-?
z+<BD<)46*MDLW0@XaS8?qn#3#8_W#m_0g|vxnp0-t?1>k@JLl?zbYjFA_w&8&>fLE
zlCwqS3@z;_xxtC?F_SgU*mFWjN{jD&N`+0qOfs$>pvOb7;KCi+wyJyYI^<AA1y{5L
zQ60`@0K$)#1AHGMGL2mG#Y<s;Mh)EP>jku(gp|U&mU+};ZTGIf8!c1>6N8I2T38|@
zJJ0{wi$b0sI$tXhZ@B5^=NUA`%=+JL+m`y9@=*q>Mh&e!<ENT+SF=~iy$$IWwp<EC
zrM6|2j+e7?+{$GRtjmQZv}KBn??parQ&Xc*!Ponl=a<gAsOmmMvJ^0wY19)NY8LI6
zOI^M|*Ck($I&9e=O9Nsd+DPoU=W}alOggo{41)@HMZMjI9M7xIB@pw|xBce)F?_wj
zT|kd7zZgFx>{~Jzh8aL59i^rND|AT?czmbO5tQ=M-b)SU%9R~lI0>`s-wT~Svmw{A
zr;b@Jf6S!>N*o;o$aM}%jQij_97%t~;`=^HWVMPX`E|U!>2HTwK^8~Mb+BKO;hmAF
zelWlUQrAGa{+&E1ASe11ZM>0l45FuP4C%7O()(9n#0l!+Q_<()h2kWo&eSPQ$2<4Z
z_bYnIq)k}bDlNl@*Ytm;>2QtH3^UZpE@K=9(!j8OSUfg{s#<lhq}vqp6KP<}B!@7v
zS?P$BS9%cZRbucaHaj`r-)#<HM0IFA?@!qY$~?EL{{0hmaem_S`+o@ORDKaAoTSQ!
zs=oR2R{%bWWgXLB%l=&?@8v3cG$q{`POg~);WdF{y=)WzNwvmb-1OoaMTko0Nkyz;
zAF#28Gh3sD|Ln%99ZI8S6#dkR+o0hrnXy5!qv-#gm!#zN@-OfJ*Wa4cWP1lmIgPh#
zXr_=}2G4bd`d0I{D?cY1oo|txp!z1C5eCdHeZ8MTC{#<ySZSv3LGvdNfEVe^>8=Cs
zu99`Cp6MGd()KfAl2;@7x(-oMV@J;m#q~w#cqYoUM08FKbp7??3%aS{@M_=U6|adf
z1N>rQ!&ai-36BoZmmeKZrpjpxFj;m);h+1zC~%&TR;&%wf6~r%7(>MqlL5P<`TB~Q
zC4?@S^qVdXVK>^_Z|O#r3U&3gg^ex^zFqpgbnad^I|uDgQDD1hv)=cFEv+VYN?dCD
zAX?I7C$6sggm>4paT>dO#Wq;cV5!;M-hKG78k*vY%i?J7ml$7<1EbIKFA1A2WFVTZ
zvW`;@t{#J&m?w3-DlmbQk&Q@L)2qIstA2WJZ({?75*M6bL9{f-Qsshi6wN0Nhx?+v
zASq5g`6c`Ah*}@|n;@n>_oJeL?`dERziQqo0w_E8GdI(BZuGd{NtR+Vdb069*z8Y1
zAUJ~5%ayz=@uLOLe;}eMcjR(^#4i?7*dbj4A!<x);`;NFIXiO%5?JBhm;4-9vpI@#
zitlQcF3@Zb(}~%|jjculvhwdk_h_)&-5tK2gVIOi>rEs<xLG6{(SqgAW2>7_8joLM
zb(^WVmxsm;h7<W(79ZQ+z03b4>}tDE<NsPo|9YcBD8t9mqc=+xMb4%zok*Ww)Q=Na
zU);U0t^cLh54FfrJ(ZOO;$9<{TpGn7NFol`-a}b)t&{GJS$yYlED#ExSa<b*SUTsw
zNZ-e8Z`)?uHaBgyZEW6Ln{98lZQHeZvnSh}?3&y&@9*dNVg7=d*FE=joyU2c=nk>0
zY<W=}%i!8EZ;nct9=A1u<UXsM+yrA?wW<b1bqhyu&%a6zT$3SBr0(vqi+9uS5=)(|
zD~<X1aX)nou%YWK6V;dsPPO0Jzqs0WbwD{%lEwZCx0DqEW+PWGIy~!c@9$wKH9ekV
z46@~$e4pCsHY_*A82JnG3*C(+>zK~frwPtS8Zr&W>AgSuYmt|qaalT5Ey6R~b<-qZ
zXW*Cgn~l2kk8}}Z3Ef|(ec#&eF&?KC7R80SbV1IO$fc8I^OPR2`z1|p)1N1wv2ndm
zp33kT63STEIC|z%w+B<iK9rBU&7{>MmK=q8Tfz(uwo|u>n54~H`|8EDSO9B&2$COt
z-bYP5Nr&0W2m@1spPN({qFkO~z!yL8VU&95aSj>GEu7kLQh){_wWYmZosFY^LiXSQ
zNNhYjwO(1$-JjC^Kgf1B<4qj%Sve5p4tW-MY4*w@a?Zq{-ko3G&=#pkfjp^(ccg=1
zt^g7C$b61*;2RCp_^)IL3`a$bYmckz9xM@46BdbxA0dB?TMB3qHR#A&_8p^~nKY>S
ztH_$g7g38W4B8T%55&gJ5*(!0WpOp>nD&MwW*Zzf+lapA4t5Z2$&11V5-ZjUc157*
zqWGek;9NoRldN$Ic~ErUISS+Zgl#D@vqZZ4W0LU3+ID77g^jgZuq+s+NyPN!nI9|@
zUFLpa%wGR!SFOxfh-gSulbZ`(-`*Z-V`@DcIjavj{ADlI!8<rw&QcLBpWR-rYNQ>0
zK&Vkd$x($lh`HA~PECVc7$;5wC&)*cKpnpGe0j;-v>M&rEnwb+x8+TTO{m1}d&Eei
zl(#oHmNbXVELnQz8}ZFD1Pr1z8edur9^ZKQ9EXHKM5B4v#Ta?Irrg}J$yHU|rZe-_
z{6qcf4I>08lXP)GXBh_v0pwq5;jhF85Clh1t*5nl++}U=saY&F9O6Z*{-N#O2t^cK
ziXlhVSXC>mnJbUIT%uD^?h>^{T*LlmTb`<DTm&3^5p1h9IQrEQ7Fplv8_bZlZEnhu
z!gMX7!W_d2hPjBhy|%p06vitL&MD#wn}@iNnpP2w4{i|3cwB3D^4`Ke(MVnwfbk`m
zw+~lN&MxEZi9HSxrE#i5m<ivO`787to=~Yw3u5HgaLvAY#UU_}NggTW+x-!J<lPhC
zc*r&vr1FZJg}3%<xdB*_yjj+;s3eL^L5}ubv!W82V(fGs7cRQy)J9x7gz_U~UoA!H
z>4s)af5*=6bS6v)dmvK*c4_~G6**HZ0~DVzQe9L?6{@|zGDyguonBABBj<3&;;KUB
zLSU)-=^GCP(J6~}eeam+#J~jHZFw4_V`)6ICasgo9RE;7tv?V9+hXXN&@qS8bNjah
z5^ZCRY7V6&&djK#EQ|lCfb|+v>Yq0g*tz@6i+Iq%X$!eok1~PAdkWg!ZY0|ttM{3)
z#u3$`-@5UbCf@wy{SOiVM`MW1cfO!IKe4cf0YMY8m~WP98a_+y?DVR<VQow$&>;aG
ztM>t=6(twsXeJUH%f)v^iN31gwgKnDl>XnvXHZqYo3j<=Nv6tI0Dsud*SEEe_y}5X
zX0Z~-fhqZx2Ulu248?5TAyX?E(}Xz?*in)F`-@A^AVfHC;7^G%30$lv#5y%EQ56l^
zlKRr)?}7;)Y3JwNuE}3qnxAHVf$DD3djUmerblXuAqaY}mvP*lyM#FY7il*W3<s6n
z)O5n@3r(sFkm*|(g`kyCraz7N>}&baRKJrrW4<=|#J;~1Om-(mCcg`C)@}W2Y@8gX
z09pZ=+U&0!to_5Lbld*AUBCl=RM*P??QWM*(Zn<xdVEwsKJrf9XEl5oue&)I5u9Mp
z`ydw-7Y9hX7c>B}qSzd(8$^5O3H+4*q$h2KG&-^3F>@ru>+}pJE0K-u$Ap!}V6xh3
zlF6>*XV9oD!pfkn9~X~csnf`3;`AWP<h*?>cmF{tlrny;ST;w^$tkQ?FGaT}A4cwe
zCiL26NqwdbYKCFo+sQ_hLeA@$3G-~E5oAn7E!u@f3=hCRai*OrZ*p6Dk42<c=@);Z
zY$02oFXH&gC?XO03nt<=3ahJ-rX-NKoZhTqINw!x&4$m1gU+6IkBGZQEyd9AfHdF-
z+>fxCiUtrVBiH(AU{=r&9gazvJ5}#?Mz9BFR2r8yvaPa|ZGaT3?EsKOo2U{w%w%!T
z#1VGg%l>2_ebD>=)0!u8Oj~a5pDw%0yLv*A1p;q)WRV!yKBr4YNW=KivzUO0`u0V-
zGJr}f4dz3%m!Q-TX}+S&LpNp>aA-Cv9yF2)qO=N26VaD6m8A<p`CDq{<shuTe13H`
z&w=Es@4Pn%x3sjx?RvzqUlKtp{PRSMiIZ1V6K~-G3RMtje!7(bXEe(beS%M0LMxlY
zd!8KJQg}l$h)9SJt|huGq#v_RamH0L<<9e|k2_EWiR>DaDhgb^owWVr^Nx&B^2gWz
zX$6xgO1>#lDVX8*_c|<z^fPBjvW2NZVwOr!s$}MnV({!rS)EDlFx)}Wq}hnmo^E7<
z-}MwNJTi%3naB0^TbuoM;*27}%KByz;q>ap>LARB$+S;US>7wZ1v7sP4Car74FD8t
zEo&HoSlV)Yj0VZ?8M@Z18v{=3Bu-AWjbvbXbHwYyL?cq96^1seaI`*Li?x0m-q1Pg
zaL=INZ~%e)mXY+O--wODyt8c-L5@9P41)ev0*Yf`f$oY@+|vb)qjbwvn9vAM+!8A*
z7Vja=31%x>aODi_w`_YAkw6%0iSI1pbVUfWB?{DG=2!kMUpVkc%|cozB4CQhhf{OH
z%TN?~4W$t&Lqf3tB~z$5rUd_RA&Pb6R>ZAhphQ_3k>Qbw*jOF=ZLF+mr6{c4h=-J+
zjobXl8#R@SV-Z<hS!I5M;9zBfTu#Lk#7Hx+$oe!(zFjA%AHF6f*r|VRWW%Ig&HL63
z=dzCS$ryIy1{PFKv-n?*o{2ycK3}DC4g{d&K8pO=h_ihavsgD9Z3o<O#-YG5`#g0L
zeRX{X?cN|)z1>+r7k7?V!!<L@lBGfhg=YX+tj@>*49N|a<ZCOYNz%JZlz*~ylUzqq
z1o`S#k)=T+#3scLvDyL_nJ*pK8Ztldocpo)Uyh*?R<sR^z(IdyKh;GC|FW&o{2LUG
z4qdqnX<ZcmaZfm{6meHlxENqRrcZ5|y?OShNkhocJ7cAxD~I|d_sEZ2MnZ|^Mp8#y
z%$OkF{^KQ1`;S?}>oGNpbbdDxMf%7cfsV?-e1@cGU;q5>`S?nWj*vPL-DD<~)_>Ge
z!(qY&UzaRxv{|=)T0BhR)%N&iT)iK<dwe8)SJh&|mecn>@{YYY>+E_fN_)P?@rx+S
z^$*n+7?O%%#Zv0M<^c%Zu=+&*DNTu2TxwN?7k!Lw%1PZ%M)<+)Rzmh9K0o0;Jxs4u
z$d6(Io731V#%lXDDS^?3m5@1tY21H!^k5X$kP|K@rq#sl-~*<J_p(c>pc5)dn?g7`
zX~L8G{b?xFR><>>a<3Q#r4njj35FtcI;QdYi#`H`smXNYJ*#IBo}N&6f<yRQ?vk*2
z0Su!KMoK~u9Lm;{oo)Q&Z{}!a)032eXHoQ639LWmw2?u$Nc5@F&Ma_O3?{{etrI5j
zl)9@qZGePup0`Z?2Z-0iS9dQG-c)9--rvyCqUY0x*B)UzS1h{A5UIOYL{+l9_9w-}
z!wzBT9eh%5sRutr&C!b}PlZt39>I50F0H;0qrP8E+?B3C5Oo5CUPRcr2lJ)U31K^P
zjuOR|U;xpW9O+g`5Kz5IOHl`zLb!K8I;dxpxa0{`d3fX?F^T>5L@S|drlF#NB_}6W
zU0utjty5S{Ur}iGFBPTttREavn>C~F)M=ociaPNhl0X#l=3wB=5f|Jxd?2QGJd^|h
ziTUE5XU~!qC?&;Y;{TY{C>zY?!}gfo#emxH)BJ}Jtb3z$bW%@?OC@5kinoHC5+TKv
zj=g3olCR6DDrUpY6FIS=z5&Z~%ItohmP&>MjO^?ak_aR@4Yo%q{jw3^G1xsYnO4a}
ztIc5R@yf7j8Ugk^S>?VQ(EJnt!NS=0ii9JZNLAZrQ;93MAU$<Cz<^&slZeIEq&4Qg
z&+!s@`fqvLi$|f7dlg9Rvy_oG!L!nXT{bm^$vt7As6@6|DNjT?Jk9j5aHd1JNya@M
zX(AR@QU8J+j!fSXMFqq7NjrK&KZ)b=I4FC^Db0&eB+Mw1J0<U(XeuCl!alu}rg4qh
zn-hQUzY_I%8*`u|%P!W@QW-tiyte#VRfk8DMj(Lzmn{peXtT;Z>d3xD_Pf~Rab^~B
z{M02I15$m1Vuc7^phhfFmITQRRE)I(0=Pq}iiNyLG*3j63WT(@<0={ArOAgR3E+IG
zzVq2r4xIfkO~3WstWnKjoHCNE^jDmo*|Z+LclgOY9k)Yx19AT<@PDs1(}u@>x5u$Q
z%Mfy0ggM=E`HFKAf2gpm(!#ANL?RURRAN~1$3^!uGtC9d9CJO_3w<bmM?{M!V2RJV
ze;f;|!~97ZWuwxjz&cLvup!`m<d&>vnkm_vJYlc4w{dvf(2hX;^UsBo;EVkx7H@vM
zUhh3o4IQn`qU!=&h11N457CKEK}Nr7sn2O4?Y9jS7i;6_4gI8g?{BWX?{R`-Wd<ud
zDW9O|3I7jU8(F)BF?O1usMzq5atf`FQ!`FxJ=1rDJkhMg9SVq;S!|naS?a%`HaC<7
zyc)9^UrpfGP**w{25K0xHo9D6rfRZmwtJ?w`iNcEo%}YhuilTMqG{MA7r&{S+Pa3A
zbM3S~UfYEQPbs+BZIf~&v|uFeaOVNqTUsP8D6B1V^UJB56>|7lMP|({^JNUWBI%Z^
z>oiq87t3Po)6dAaw*SS*<%%he#zf49!ymck<Iz=$q7P7QP?~M%fXM4)k+`7Yy2n0B
zM+@R$SYyX|cwanG4)%#VW+8&^K`bIdO_DkpnAH54vnNOg!A7>2($7vXJ8B=9x3od9
z-sbTnGGx_)Ad0g!4UJ43l`_dIVjR)AuBgH0RbX!X8s;xddVJn*S$#AA2x2Q5nHDu&
zh#duS2K}6o3#M}r=;RZfPXoY)Aw@;WvW>JZ&ni3TtGj-6l3ZDvseVU9KkQn`uhFdU
zJq}1=KtBhIP#DRo1q`Qn2JW}^>Hl#N%x+TdVYxulY2K}89(TwuD0~6~j@&5fR2H=U
zt*lTw5;L%1ctu)5#Z#b!Dt4TWEO@9~x`KhW8UNAoGD9<o{LAbF0bPw6E{AdTD62wh
zs<?%n6`l!W)3YA0_+U5_luO8lFEvjKJ$-Zo9X%?`QZfAdLHP(GGZKJHKev7fJ>nSP
z0l%Fk(%FqmQSy!_o#}JJkSD&%Gwt)fywc^A^7tS}OI30#DBHCS71#8ui0TsHd8YIc
zt(2>8cE6$~u7;j8893M4nT}6o%cDwRsD_*Tqx`4C-HR=A_XBd;wr3XCwpY$F_;bc(
zFa5O$gL>!Wj=<;e4(F#n6P^}^tXiSmfaT)e5#Pccqz)J4K5Z(4WR=_?>ff6W4Y$X~
zbQq5t48we+d`n*U7Z%%UTxQDxJ7-#^vvn6nx#QI@W3$IqeYcsDh1uwts=|upP`c6y
zLB}{_f953(ipI9g*s}Y`DuHNG!#6wV<s|74Jj5j29SV>gkHVgp>FNsU46^JL9E|YC
zc!ZzAwP~u6$H6=FHDA(7J25APF_)MnVNv<}Ytb5XjN}6wBP+{ac+MdOn>}a{r3<2x
z<&gy+=;vBROME)3ms7-I5IG?VaUg}_p>Z1_RrcJ87=U4u8#{c+ZG=V^9LLf}=G6@X
z@$9~%sBEb^)%Utztb>6f$T^kBE1QmhxW6{fHS&|hoj6uWn6(Hfs0;r~<``_DkxNeJ
z;0I@;U1WQL?_5T_eZGwH`Fvhqh5SCjOuQcjwMm}C&Vt`t#o_u!0$}gxHxC^p6MZ6-
zKb^KZd`uJLudk(!ihe+qFkRQUw9{3pa(q=uzRN7M$zYUAceQnm>97LRQrJ8lPJ0V7
zMn|H@rrDY8_hvQYf@yOr=eUvjk0MTAq8d#NxHxKxnX%SXH7V9wv2?LSis<BLacUrJ
zU)e`1boI4Oco8ryl<p?&KiKT`dL=jVOp-=o=Kdbvn0|^j`9DKnz1?tm?vdE#b1D`j
za+SG6+FOe?!viJSr12PRLyh}h)9h}?&xDR%zx(D1z26D#txbWb=vb97|NTmDLz0tY
zIKQ}5%ZywsK^k5}FSxd3RJ6_{l^+>4%|R7dt~ES-ykN4svb|eT%QH9G=MWqXHED*T
zG^8T<eG`i#&>WyxVU9j9O%Q)Y|G;)OH^khzJlq`cm2K5V_AQ$Bh*%OLI}PO2hgK-2
znySI;-iSuW<S7Cz&d5pdzT-{mkhhP!BJjj&dQdoseEDnFI9ldNygIqh<@d248CUUk
zJaP1s!>LSqdT72Wq;nq)HsiggID`}AShj0wnSS_JL3P=&WS%Va@fK`^4Q=C;pG|_p
z*<5qy70A67$QB3LqAyy97;~&_vliyeA@Xx$mv@HhFe{Xp%o2q9x38pQLhNA(c`3-8
z!FOcR(?D|bVU@8fF@jQZIv<zt27>gMBF$UTt*!rmv+Ops$KS(VvlDF6Me^-BRB=f^
zMx@7B(hxpAh&6{C^FEG>M(&#e@`c>0k*u|?@O@HAYKL@CL%dU>b4jlw7?;9K!Mwn=
z?)$BdT2{*lAI(bM7{|S{KN~)4xxxCh#bcZZ_4Qg)SVvm;JG+yFN?}XjmnHnf6%@3P
zMeG`vvM5?~HMc&`r+P=Fd__E}si7G2_Y4{y&@)c^ZP|8TdAhhTltk3u07m?E`r6)m
z{3IG;;DUuvu&zN_CL`4}2h})z4D-u1v`&+&Z=9Cj8@>Ov#b>I}$5EaNm-BInB-9UU
z7V$zSTu3m@r&VVtk;jCKC5Ownf5Rpo5{s=Cy^?nUW?p`@f<w~0izcp|yA!G01Pu8+
z<qS|jb}=_%@xc^1;!+bO7xPiqLHy4xZ!vk_Q{FmhOS2SCJ8Pt}>0D#KS7<>}nQm6p
zLZw?&YAU=59awxL%0iD|S*tFXL+7VUgWAiRr%pn=WMNeatq2#ML23vh2fVLs;T=Uu
z0UnPi&;h`DT~Nj!MK?-nedZLM2WeQ@Z3wr$fAiD3MYu8=b*p>Szoa_ihT_e(XhVT!
z>CPV9*<K#(C$~rF9IBDcWRF?mwf!ma3MfZTf2m2%>W>S&Kl>byMlM%o8Wm)EM?Nse
zo^cD?LzZ0)gFFWJ`&lYr8fW5cy#*Ih;%w9A<3-d_jytx64qDs?);5vqCwx^>RP|@Y
zfW)QzQYq*Ag&592Z}$MZ>m1W=P(Gj-^NUe3+Dx`^4kxsUEzEKB)G}p+35{}^L^xHP
zcBL8ugI7-9bF$&NggyJ$HO5IIh(t}NL=1#lj40er6HFOQ-Vt&+sP_X$`+R1+!E%ul
z3V`3k#Cf-SRPZwMu8;VwK;o=c<4>D!{xePwbv;uD=X9Ei*^)!99HEvrUWNtYuIPp?
zcmKff$C!KX=knW4-X~g%wXj6>k8B}t&4s=kF}fR_O4(hpi4%69vgF;?B8c`pe4fBU
zwM##~R<yy>Itf(UHFEhu9CIhHsZ`(Lm)gkqCncOm9sTgVNN0t`DG_LFAVDgb#F;;%
za)v;l0_*OA^~DNQh|Y(KmCfxN32L*Cq5<li$}M~Uc{c@ZhAxwN)xh=CJ<c9oQul$s
zYZ9l;ImraZxWYUfo^R`(y;i3lSE{a>cI{MT5I{r#77F6Rp-FT+XI{WvMg>Mb<?jz)
zW_>Re5D~K0xogo~x(3L-4MH6IKr+;*wrlsu;kTk5aczrHpeS;cYnBMKJ)M!RRZK%^
z)G{p;^v(F#w8mGL&}xH^Y-D=#>rLMrh@v#&30*(=XO)E5;?c7F&S5Eg?RShYH|Z=G
zbvREjV=1~`>rErk#r*3pHqJc4OZ#o#uZvvmeZ>B`Vx=^W2DSc$6U4-p)ZdmhRxq)4
zKv_3y6V!giUZCIXQaWXOa4|$ai$oxND4X~&23l9e@Z4inHN7%E3Hijt1m=?JGy$F4
zb;7DmtTmC`KHL(1HF{3>n5C$6tnem>2Z@y~&727C>@s^r2k*NUW+Zw4oN0y0p%Zc?
zz3i!nyTe7U2$5_T>|&6CP+w8+i<zBar%Y3~Ty!OSmwrl`3MQ?&R_)Ceih=QahJ^|Q
zYxelPLT-$D*OY}?ZMS#SnO5B$&h`tt&UW}5XD$tK@90Rbf9jP#=+(yVGHOXFd@@Zg
zJ1TluF5COy>lTcG-yxOe@I-a3%`XHAzC7o{HLR_!d+JV@*A?SI%SLewj%B;^>&bC?
zGctT}W-lwepzv?-Y6;P|`Ds??)ZOLLVPCc;lq_)PhRJgcNP(0k?2)ROrL}?5+R531
z&Z*A;=J_dxMEhL{7#p(hv?2fW&IY+DA<3v-Bt`WqQG$u$@}|+)C#Wp#5+00ZtuGBJ
z+L=9_BOC0QQW+|VM+2$i(bQlcV$14bE!7K!L#-)3H)h$6#i-4reT@cCCta&I$dtK!
z5Pv=~ZU(mSX}3=fpQ!5-w49nLxO`nD;wCu4M+&e6`5xqb6o^G{{exySzeQ4%(Bo&4
zx<(CNF7KPpxk&fR{O|2jb=}V+w(K=jZ?pSrr}xW~OBU(<^|=hMJs|KreaYyo8L<lk
zLC9mdfPJ~{ec=HzWu<Nqo;u5@x(~Y*p;sr%`{#(FD&2bgTRWJ{uK+DH@hsuWswrzw
zC6!P$;`+?Utj~CNVc3c)i<@;=M#<*ps*Y8ZxE~$j1~&_KMuDIH%<S*H!bC)7Mm#qS
zK%UPMckZWwUdraeru}3_%wY20VH(62m~DR7^n8ZCaZNb6BWVzj`sj_b(0T_`B)xqu
z{(#P9X2-inq|~&uo1ViP8j^i?e_-3wk)dBw%_h;Q=J6Tq`BjV0=ONeg(}>{M+8VwE
zAP=U-^C3m6{bk^)5lqoc*+A<wG<WoSxMs*|(Ox|F5!l~7K7586KA&oVi=h4yW5xK=
zRMM<H2R~=p(a7j$p!sIj9mkHJ#VZjBm>JB%b<m83?<X)qh_OoOxSdZqJ`em~ghbAQ
zM)JHe!F7Jfk7uFR>GZbPoi3ZbahfaiAcVi3!Pdn=)UCP4-O0OUe{^9z+}k8(M)rY5
zjCM=?|1uz?PgWpY@6UKC-SP&)X5{b~WY&;m`!W0EXkbf@SE`wNX@Eq}daJ+}QOJX5
zo^L#a?XGb))dlMZv0v?2CH1h+2{*V1iJVJNe<iF`+oZ>^c6SA}<U6M`LHR+?Q4)U0
zj0hnXJjQLeuQGkqpx1X|G!vPhhPIU>erS9I(Ot=Iodt5Z3xQ9nIkrOQ{LGWj#O&|t
zR~aKp<0HgDykp7IAyo9Gt$)54d?>2G)gZBA*@mEE7Yp(JTlZ}f%aM=1po~>Fr-`X9
z3rbXpwltxwRJEHxk52sqNtH-NEyFZ~`qy=zn#+FUUrS+K8wY+<N5}YTlf|{7@agbs
z6R!0Q>kJ1{pH#FRHb5g9tt#e^L9kvJqn25XnR?7k(@Sk_Uw2@K%S$)N12|(ALX4yL
zG;Xo#wNGT%^B_n$o}i~mh_qCk8>IVc)BC}B=}c@r4&YO#Ekv-{>H_H6cLCOJt?caW
z8aS9MsIhRsjw$~arB9r<kE>5XuG;P5+%i9JVJnWBu+O)=cj7EA_vZ)0S9%~IXIOoL
z|8vj6^R1l+-2c$?1SL{>SS!pclHt06%HDQh>TZuZ{f9cQx~HcI)_pFtdL~cis{+o-
zpMB=SnE0FwFd`LybWOBe6PJi>U&2{t_ZRWWrg9k7w16#EcMEDtVB-s(=8{rlgJY1E
z;WJuW8<&-j+%D5BNJ9(YdD?N=uLOQhIO#E|lu_SYu$I4i5;vBrrR3<`GxH78g&Bzk
zm?}4O5(ON)BsLG@K45feZ%kcp?W-MgG%4cKEzHQG-2y9<KNCbOGC##oGnfBxnir6g
z3K6T>CCJ9iim9$m>o^Hc-#`$gB`t+XHkfQRi&#@JRk4*zli`@Wx|ww^*K|VajH<{H
z41DDqVHt$qv-O9)DMt=m@t{##7HD~eInnY4>LU4F)rlM^TfX-E=MOBm=|rCq1DcP<
z$r2C4G&l~t3Plg`bxDR5!FUI+<3<?t=b^*L>W><tB>}SgEvblv6U`0H6JTIwYIZTF
zrF&7Z>mt7%HOmO&HUCIgR;KN^rnOQGM@rzTu9Pj7O%yx;CpzBFQKfdIl1rLetJ|9;
zwDdMAEbho^$(Gjjt@*UFveskWjT*+!6AMhVQi$So!e5H31m->OFqq4u!S@}$iy2BO
zLadUJ7-xD<YZiZftudO~)AnRf4xnOG;ETR;Pd)0US)FE3F6gipCX5o)^-P*p!+w0J
zQDm)^<bNMDiB6UK82Fs}jJ&wOTdvzY0xPq&A{y!?+3Df=bF;i*-n4o_FmGI+rzm(z
z^wddcU~x=LjH_ZU_gKa}0a(XcD7JI@Ah6Zt6GhVdfqiArEwRu4IY9D3&}wJj*wi$}
z)$>us|G47WjO?*}hKz*cCyv$_Hq-jV8KpDsKFt_32aki5blAkmRObrUX65Dzw`#kN
zE;hIpDa9-}WdCs~v={ezu-)M49aYJ{-jaexed_hyY+L(Uxl1NbM@k0q)5Ax6xZwJ%
zz9!L<l<HcooVmjPX~V$m<3lCKnW?9KbGRg$#a5PqwYUQ%<fmE3wy*KxEpBKC-!&$T
zIe&<I($|WED>FCdLxT2JszK74`i`Fu<egt&f64u-VehXX80z;u;S4h(%c$)0E=4)E
zXXl;}dtfyG=pL|@%2s-S=p`WzD_Fg5=UHHiE;IPM%^&|(3C0xaS0D9C*2Avj_TsN>
z389{lad^OlDip42gWC1nUqSdfh4CD<@SYNt47!-XW4Jzy=3_axEB=}8v%t-rPP#uA
zoTbih!tm{kE|VYeW6evYzmG>P`TME++%dtF-V^~B__tzaZy&dqqT$x&{!>@if?wO%
z-9NI1zH;o(e?GaO8%xTI-b9;-rA%ed@kUHXSJLK!tR&l2Czelb7FSkki6dB*d*{TI
z-TS-m4mU&YtFKL0g`&4QupS&Lbh3z=R;M64wr<Cqn{l{k<YZ0EQClOAvz9clK0}{#
ziP_=k9uFUI9rE2j7Xhp?=)Cd?W4JQt`w=^~a<pNd95(EUPf|auf;-ggW@Af|zknSH
za@bPqRM(6vv?FrO(q$M6vG$p0>;1F^XX6U<x4OL#*8H#ML(Oz>0<~($ij&<8MP$PP
z$Y@wW;nKtEtC)owUPB&M!ATa=MJ@?ja;}Hhvca>Q4cNp;**uP^)4rDs+*h7=F<N*r
z?uo?d9@P4F@n6;R<kzV{<g)9v8JqKm0>u6xF=p+$5Le*-98Du6nu?KFGGV>gd4K&o
z;63`pN3x`Vks-1$elHM_BSu@8xv^BW<$UKZ%)krUnqKxYvtE^6JDhf^g93H0{|Fl<
zi%>AWGd;^Ry3-CLvX+gBz|_P>%WVB{$t*wn&U!(Z-VFC_QeGd#716g}``G2D+xuF8
z{N9kSoAz&Gcl_9o!tcv}ehimZfu<>vYufIor}y>bfuc5ycvQkc;cG>DHIrIeWU%P8
zPWC{^hOAm!JOTfPJcPeuB8DyKED;9pq$Jlxd9K1dJ(QIe++n{ECYGj`1&hL&s*gh@
zB^+W4X+7@yi;lmnNpae5?*01wvdnRQabDlh+GomPESObrDtQ~2g>9IOxt5QBM(t)h
z*cRJbO$VRA%(|wGt>La$>qhv-*k6Y>2Pebq9u|n@#kRW#&P^EXyQE$}um`1Qz#hAA
zR5WN)=a0x-{%`PQd49;&Qp(XKJau@p9z%UD<BtHQ%HM++aYAEdLho|obsoSuDrN(p
zWCE_*hPJjaXXNF-;`-yIOU=l8<o$pwV**UUl(-R0&J$|&mp0efG(*pb#%A0q+!(*C
z5t7bu5EEJ)YopA!61Bhb{ral|AqV0FVl((S`Q?(?g6<I32y`-K7LZOqVp%yOki9ZH
zL<+dwj|dH)Hngq*&L=rs>B&_4<Kd2Eo=1_fo}(PT03j#9Zrnf}NdX$)QDxEzX%oA?
z_BS2x<mtl;fdG7Ltc@$n_0T14y0bXW=~h9G$wJ~3ooE^2b|^;tn|ogrq0}naBq{#M
zw23dJXXn2Vm!uCX=T2djExhOwnah;i$!NyPREQc@CG_~N-Dvjn7jpXhG7cx&NH_%<
z+kGH%heF~Ee>uka>_X!He@*fM9$;HOKVhI=O1{yXDVIC4W<p*cfTKYffc=BMbjCWm
zf3IZ58c=oS1T@A`C#nbGm#8W&4sfzb%18UUb`raUtK=75x=>O@lICp+XjxaYiPyjh
z);D(bS0vH2`|f?~;~CZnnGN0H*CdcM3yDLrj8T!%!>_N1`lE@DLufls*$oU>4E<Hj
z{Y}J$vm2S?;jOmzQ~<b;aUjX0-bUwuTX_fk`_vwH^D-qAtSOQVhaWWj+iGG@7d-s&
zOg!zsf0jH^Wm$u#a@XenR;SI=uBaZUW%y1u&Hq5|<q}kiOv|CL-|lhqRmYo?%YJU5
z5kh?|T`UTgV?=$95ra+AfqAE+Tm&A0cF|KrmQJ!L;eeh4>3jE^FCbe`K_m7_>hs#2
z$MZJ@Uc?MsW#P~KiOz49CCbBvLHC^e-j^zsmE=z|V+w?TAJ+!Aw4$(+sxrL|nd}zK
z*Q?V$5vw+Ub^D%v4;xC3LXMoBO4ZH&s}~Vuyx9F4XP%DcIAmCv@;;;&4|8fjW{)a+
z7{PE6+>%ap3)~FV_y>gEUH1t@3i+Y)N^A+nKr+3gGcGfG%z)7Rh0!Whg-r=VB==i7
z0<ox$6l+!QI`YS%vWyC4$wM|*S<0K83yn~sc9)SI5+TG|2mP{(d_pYRx%2tU{`oBo
zOaMDj>}dUMZMVkfG3<_<<lenDQXMj@vz}Ej|G$F{+j=gt^Km-jCA9P{NtE;Xz`u1R
zeD}%qZ=`$|qsJJwc7JHp(R@$^xNGRDU4nm(k|??@P<opwIC%u$>rw6Y3sZVzIO~+r
ziWuDEMQ2r`aiycEm@qcGD(pj=H+~!S+W4t&=E+gIkh%^86hTf}uxGAjV-OHz^SEZl
zumfk7=7vchnfk|)Xu#abvkzeFlKGXd6MYO;IaW?xeX`Vfw%fV?i_%%A$>}BCz|yOx
zIDLzerCfAFa%ny-b@ua06Qd!2j94ah(O?z@&#>8_&9*Hg;4o4&8>Yub)^>^C!@AGi
z_B-F27eMa&r?6r*J6HiVc9BEm_n_4a(CT@cP%lE8sG7(x@<?5BNl_&WH%w4*vrDZ~
z6}^F>v6H>KIV^z-VY1X2Y3qIE((Acp_P-AE7gcMclTlT6zAq@6Fq?%O&sPZKEV&n5
z2#vugD<3w=EFI#CQD>cPqUY)3@it_YG{5s*YgI%J+TCbJ@VRt0+`G@XJHH~C;hg(T
zcL1Lo=<bnT!vpOq)Jt(O-#yE7OGErT@2-0M`Dh?2z9kfp-O*a&TBDvJNjG<$(fSMZ
zg`m1b`nxYWihvT5Xw7C5G#MZkXS`1ARyR3yY93b`IGjnz9xpDMBy0(D>}0n<O5#XM
znBAufYTtZ!*)GE&A`TEJxqLe56`NLdyJG#~66W7w!gEhTa_gSUA~hWb19={6`#NB;
zt^rcw-wUg4pHR7b8u0627jh81-FDZ@LCl8!VtxFESZ7)LYf+MhQxRk>L#J-1I{#jN
zDbO^ZjnMQ0AcfS0&w;E`OR;scsbLCN5#2^7ZV|^E#{N$pvxZMVqv&w7p{<zBKW6F~
z@mt;KKkVqo((>}pOh(pR`Ep5$RLxXvZCx{4-<*_|mU|HX84V}5;|JlH1wdJyLSDNt
z>Z=)2dS|dTXmG*vs=l?j-XvTSn-ji<k(Nbc=y=eI+2@UCtEUkvhMZoBpg_Dpy%<*c
z>Y@qs*om3<!f3sI<#avX#@*CAR5g{p+EZGe$z|{gewMr<w$<tZU93D0`qlVBZ#1yD
zoS%80T&Y@`D;+Jt_5#e!jn@=|>AF#aA;~ub2!4)@WCX@)&X<!ib{kzofJ^4>*S_t>
z9-p8qAK<GGm^N38(nQ1DgEGz31nnh@GXy2F);D(rSrgg!-f$ZE4}*JJ^)60gPHbK3
zyoU<hAj}P+mA%sz@EuHcNjdy@BUq=MchHpg$yWDv%(vhgMXvz#J-8NrJY|!}+rIO7
zdyDNNV{mPDZ`oOtOwn<RfI!&7Xg0Xsl4!d8h_l|}n9K@NLFYwQIgq!B)~DqRGv~#w
zhAeyU7{6mBm}JIH9p?+<aUVbuW}u3gq%V|#o1r^TgveRKS6dOJ<7MaO>1&spJOv!>
z5Fl4-hRN%guOUM;?d}H)U4CmHP{QqjApv_hrf@^|!<@-o7X4Aa{$KV2Xmw}AvLZsT
zz$rS#ovk(7*-i8Nq6-TAcL52sG%k?Q7eX&jO8ib2Abq1GE!w58lFYu@Al>tw7DJ6F
z!G|snt`C>;jJd%Gf~Buew2|nf*QicI21<sGMwa#P3lLI5MKpSH0Z|{<MWxa47887@
zlG$ox?HmQ#W{*xyqm`rrH$MJhLYZ6CA(YhR@b=B!vi<$%;Tgf{0WR%kFfuvs?U``I
z?9a)4pXZ2S>E8OZaXD3=H>P2jL#LpUaBl{YiZ|UJm}L80;=c-c!DLXX`q>2LV}{tT
z`0Sfi%EcAfbr14-^ZS}-wAG&9-p26c`8?toc<+btd+ie1R;5>8G;pmITj})N#B3XR
zTf~vP?eOW*Z|~%^G4IdqMV#okh4K*TA{&klDG*E++$GI^n6e!;j4)qsd0*WD4%=2=
zR*-r?-Dw_|86&GJKNE&Zow7<{!$(kM@TKU#$1E&Cakn@IRcwjT4f*gM9xYMRC}mTA
z3sb#6yb*L@5!lIbXLfvXDXyjmt0wqm=>U11qcK%!j$8ZPy5))E&)E7{Q~zGMiAfG;
zo1{Gz<yPn%4O9X)j2@5tq@oL<xU!~{|Cu&!u%fuvhX|Ughu)E=>MXl116k~)sRQda
z#w~tsR$&B}q~Pj{DXx&%06N-O+ke*}G`KO19D7%fS3!Rw8|k0Ng0&?O&C=6bV<(px
ze+Pg|hyW@7mg8K1Op^pM1CXz_&lQ<)(noVfldex*$V@&80@-Vx8{z+T>dNf2AzKyg
zLK4t-HsGATza0R#Pp*BGEq}VE2!$<60gwP8sK)G~;dJtqD&XO-0Iw2+Xgn2lif80;
zG}pnQ)Q*U!d@}5xMRap($b#rT=8#E(B>=ZV3+NxRPAj6vd1a^<lRI=#Xp&gX21nUU
zatWOgXz)?ygD)rZ={K<Kl`O+-cTCG%_Th+;Ofw;pSdsY@$Z_->1&zlgtiP=g<xKW&
zzwJX-;Q-hK($P^$s~PL+=Qi|-55YqtkHfC_FPWr=UHQjISxfi?Wb)MfH;2==r;AyG
z>rKoDs`TYp>bwqnI|AeGAE61qVVADY_sc3krOj?RasD6kgz})LCZT6BzX$8yz7)AD
zljgn?+mDOhcaxq+hTd_&BXV8Wt!f*r>ELBhL=FH$$#vl`ZcvdxWJd*F90Pem6b^~e
z-4b<O?~RF~ZT2pVkVl5$bINyg4K|O6^5gLZ^*q`Gu?e~FzHyKCy5oh#wDhTIT2BkB
zYBqM8d+XaD);%8&b?n4ij7&^LG3J7K&aaQy0MGkIBU@XnWMc~pm@DwCm>8INy2Gz>
zfp76#Yb83+l~&YT2g^6*=Dv;W7jb*>JJNPZJ(h=^KCl_b9&Q0gqlM==X`zZ{)445G
z<^wy7j38a^Sf<AcqQhG&oFP>L=66{vYNaKNk`K-gOQ?QXqeA1WzsrJPHU-MO{czxk
zed>ISwY$DUhA_J`<@uCA>uwfK(K-!n{6ju+#GOGi`CYdNJUBtTJaz>c<Jv6gzggXM
zBLg)fhF9O~{-<nvl()2?DbK2U1|{z=*R+Uj${8af_yf;en)S!e3e&{!31#*SOi+S?
z|L{<0DgM(4|2iH`S1&>5IVfX1CZ%*H-UK)e*Tk-|B@wv}^f5a0u8J+t`!zSuo6M{a
zU&A2?-G%T|?UzD?(ifHW!uf?kY2+PKUnI4Z^g>hcQM%M>WozAhA@-;4_Az|>R|qB4
zo4YyVVTy(FAQd=`KuK1?hQROeziWPiyOdP!?#pjrdmw@TqtJCmQyHDDgjGmTau<A9
zBz_6~Oi<%8v2U_e&oy4{$1ay6+`~F-M-Yx62-EOmjy<}1z&IUXO}+R>N)s_ofvdv)
z>be3a?{JR6e;mc&Ls#&zAEE1gU+eGnWK*+U-bC8=?#)Q<P6&>{&S{H6zaGB^k|h?4
z1f6lR;EpJzITI6lrC=mEawICndZ%}FO|{d<y|%1|gYy}FOV-_ZRoc@z&@W=vLwu0N
zis=N6u5#xc+RV1kdD<VW(V6lATwLg9tH}eA>bf5zlaH>Hw6zV+(2dIGeS2Qt?VT@P
zepKu4RnQpzGBZQkw(^xYYSIVfeV`YKjsv~x6m+(@fmWx&^U$U}LbNJ#`!`MaJfhr|
z!LsJZp@~!iCUaT=cnmSceALnv6x|2jM7aS`u}dwqn6C3*A`59rhkGVMmJY@_8lWnZ
zbOH^$1~G07boQkwB{F}V6{q%)(xEO2>rmtsf0(-FD>94<XuIClcb;c|I}x4k8Job^
z&**0wK)rUjUyOwm_zN+5l3pbiN0gz2Swtu`cS^wK<7slKlaExpiF$25e8dvG-(6;_
zbL37Z`(pGzGo`?ES#3_%>GB`wlQa2g7dN2uKTKkCq(j2h%a7e`!AP_7rC4axo!`l1
zg%P|mgS1SHxuK|PC@Y17-csy&35SZJh)?C$+&+I$vkwePQJs<N9_3A>8LLN`%s5%T
z8DV1yOFKg>D%2E&#K_9+reS%jpu*8gvZ8Wex=wXfSgnP6&KF-dn00kTR?t&dnNTk|
ziORGgm@&Tc@in5ULyc=#a-sbLHGOftdh5$`{CLBt_IXq*SMB1#_%|Y~Xe>_WbN#n+
zX9FlC;(dqTx$T`&uCqj<dl=Pwt4;j5;#v6nAB-L>=J3vV;zJCQpy1|_u8>bk+~=!*
zUX#*Wd<@1}m-{<4aN5S<c~Q$__r!vheyFB~=<%_P8@wA=sM*a$kX<yPAe}^WKzS?X
zrU%i=!hdtH&#0uy<Nk=*^v;y)f624u^Qv(A@<j+_v$b*XRuqRmTfPI!8{YSYywtg}
zWp?gu88`C56_CxNpe?`B(LL5;ebv;_*zdz>#-Evw`FF|jY>7!Tb@}OF#2?N?HrMOc
zGuP+TWD@9H-7<N^4sJ9DX*2<XiR@UwEy}yZjZ3w!53DMnSzLqw<|J|VAS6$D*EMV3
zk1##^9H(>Q!5h<=-5|p+Xv3B_>0}TTiy__6FsoQhrmYM#;ta{7ST6GG`DDK1B;!||
z0jJQvxIpF?sH~KDQG8rcxzDQax`zs0_1X{e(>eL1?4GNY465MUs61k0YV3hSI=vLm
z(IPRmi0+BWodV^T`ej?#0*n&d_0(1?9*-~i|LgSXi^f=8{x%ZqoKIVZ@m&QZTshJL
zYjt~14&2=sAE$|G)5yq}{EQ6PBZTBO$qUDwM868hpb5009?mqOtOSk$D#kU%9~6xm
z2?qAypzHL%<&Kew2lA??bLc&;GM8lw>0-_qGfJL*o69OqRc-`vfISf$bPH7#l312<
z3#EK^CbnLAUrm_Ew)8WHz%w7t-g#*sho5mck_oe`YLh`>$}ZJGtk7l>-1QaK)MwWG
z?|i)g>o)t_w|3UN=W%epEE|+3^R*V_G^Z9fI(th0-Q(~5I93nH{xcew=bOl7;2))p
zz#wO$SXXlgu&%osyqn%#!?Ug;$lmyAcDKX{7z>?mA}CGryski!b9cm5eIiGFF8S^o
zsN{FM-}N!>ZDUi7_D(z@H^C3OZ(@vrFYRQ~>Uk~WZ+&##Skv}Fznad%>eyy%{#y6d
z_RLa@^d;4ke@9B_P4I4&*wy#n_^WkwZ7n)Qv+V?deOKR!T3Px9iNoyzZjo-Qz2UBX
z&%5p8<KxOkmqfvCTknT;PE~xIkoQ&lIK7aY9_`I6@xWPCvm}j}y@JkpNC`ex&MwDV
z>C8Xh*l%{(I1ew{0yJQ*3}L2UGyUKOSwsairnAS>QNMW;rtZ+7ES#^7lM6M<(9Crj
zypLyx3OV#dcz$vu5d6b)WEmY>nW6h%(2)t@7JUgzSikr{%i@yz*NQhkeM9Y%nNqF>
z0MF4~J8XKTf&E0O?y+_!q1Qyj;uiRSnnz*2zPEQ890v&IsV)ETet5S|{U0C&Lvq7R
zDU6PD3;dIV8QN^pB4dStcj_8$bK^_pukPB0CyZwn>yY^eO;hq;E~s#k=KN;|w2Y6w
zaNe-xL^&Z^w~2jRb--N1L!VR-dV*!4OW8~gI1v3|d^)%Pni}q8AZEePh<9@FFXu~r
zaEz5gnwd;AL=9#wBVBdh53tq?(#Hnw-{Xa?XgFqx`rp_}c-<J4(qb!oa%D^^=n!}o
zW`=vCgh>=!a4`U%@S^kTdjInS<oV{_8z|(NiT!=d-;)hD2I@3eAT^byTHoi7;WLck
zAsn9=7DUb0i)J%Vs~At42v-C_RJfJOVq~V?w*}~4(2=f9+H}3|D}B-|{BoV4nWtGs
z%NW~t|Eo^^+k>RHNBehbTfY6nG{1jGF*S-Wq?Mx{Y(eEKn|bK^r!-;ZJ`)dU`uVsN
z6579*x}J~LG2Jp68M!^3h{%1;P4e<{ryMSl-?a?Q+cpcIo^Z;E{bQAD*_VkEA+)IJ
zPHF1+hlTHBaS7Y(41hWATlsG6YnnG}XF}$#wkhlDZ5dohbv=*JLT|fq&uZ^DZ{6F^
zNV)z;aCUm$QQ&P^WS!j&h*DbESqu}IS~T<`62}Tm!KbYvA?D<#XA-fp?-DFYY0Mxw
zT;Y`(>0(-^zgDc`X-MP2mKdqpcsxDf)<im1Ypp47G-v;GGVTx&Xkr!jQupU+MPX(_
ze=!~)G0NgnHF+asNqWVox=de9@4^KCy_cItJZj;6!fSL1>pSRwbqOn)2_c*jP5<Ff
za>Z9lMxybJj<y~<B{z%f^y5t;v-AYK(&~%=4zOLCR;#_wV@*EjM!<2`lTBwNe#)VP
zT1weC!gp5d0!YRR9BbT0t4%(UwB__cA3l1=^gNjkfn+T-vdV9xK|i4dJPcrs;?)=!
z^|fOhdZxZx(EcHxP>Ix42k}H^lcZ;K_%NzGkh+WxBHigax8;)Mr}l<%)_wA=6Idfw
zRyxSe19au1pI~dPGKeNu*18mUDsh>Z*d?=Fic~f1h2ck<B@rIGtNshyQR&1?KOeX}
zy*>Ne?R_AdRM^JHl1uG53}E7Hf!W!Y4lm*&SmII)+$<2tt@iAK4YHffmj9-#=}YJo
z^Nie1^glk5s@^7QK%VCbBhQc99|0{T>ytol{`aRQey5MGb$(Z28y#Lrbw0akt6iSK
z&KA?oXGcug_S)+!zxEGgnu9=F{&8*FUWVfDV0k^(7F1zc%>1kL`cl{DheFTOvtdDo
zcWNp<e$${22_RQsdfncjA8Lf)A$oelFYQIQExqTNBDLoQ={_Rr_}5J^VHResQpABO
z+S5`A*NVW((cEF1=WT@2O_=|i)LYtToj>^Cz&`|@-*qf)3eDLRz^rfMk|>~)@2f~E
zb71{dbHM(L`=a?{YogHJSJoCCrzy(^g&fnJ-|vV*GSE13>-z9FYszKvx^T3FU;o$r
zl4M2WUQlnJbrDqpcLSCJL$&&PE1eQ;LvV$f*m;1y9xM@3pQM&RR;)mnl&HLI|7aF0
z<1fq~Pkf$lV&=^Df&+JfvMD74%2hLw{kcdV0N|B0={R}*d@#@=B0{fv<`5yI8SQ9>
z&V`KF1hJ$=pZf>oUn6c;!My$K{hghBgt`~ZuhXQL_4TzaplUMzk^km7@@%y%qZ?1$
zA)I8XuIOq`4EW7a+1`bfX@<TXXZvF(4XQ%(;klSur)_h&aSpz2q{_J)NdqaUnY4;H
zmZ+71naKZK|Ji=I6qSdPk-6}}#a^oTD{D3)|Gm-9H`iub+1$y+^7|n4ea*&!_oIMr
z3ah$#^n9xMQZX~*+@tkr*(AQw58|dZ=x8F9HTbNR_Z_N{6>>x#_X-iOxFul@7aPc|
zZQqc+10H6dhbCogbb3puh(eBGuIzL>$E_<ht7%4PdH5gSaLU~u>1s*{Q<^WLcla?2
zVSAe!#t+@Op7(~?{B#`d_S7cr4b09iCT#3&olkGAa{V8nYJ2^YKwDmE(|&vNj5#Yi
z!w*?LH*$tXp0>e4@AE=H@9lSMH@C<7kP<fVGK^sV=_#A-e13|#V4dfFOxsT#og;9z
z3c}T}KEJYW?BLxK=6P&$k03q`PW5sNFbJCQkUYnzdq=%nM!pN#HuF1jWaLypy&h!N
zqNSbM+G5tK?V30`LIU@G0Z?0hXGB}RR|2iAZrt45sB1iQ+7|ZFw7a-@AuM7`LOPQT
zc&P~}a@+r9C7ZL}Er{S1(cRLSA1?-sr8|Mg3wT*h<*I=-2oQR5^BlM}R<;T01Qp2#
z%C~;G?h5nwWB^RSXd*RUHf_m_8!9@QU>!lDKBMU3*8R9)7|HVQpZZxC{|<Y)1k71*
z#g2bAb93J$4ir&wB;y5Kx{`(bLFYqzt)YZpl$93haV3w4M<B><{`NcPj2Cf1A!(`V
zivAZLC-tymh@ZNqrAfIYmYKGy1{OG|=B0m%vIcoC{?~spU$3F7@A!q5A1Fx}-=E1r
z)AZN#9}EpZH<>&Wm%J9k`jLG2yqHM`QAIc!^k?V5lWZ@TLdiD40{*0za-&*&LyI9t
z$3h}nB{`&k<CjaZEsA9pp6$L$3Hi4qSx*t%TjGtwSKbbP)ZufYxeY$PO_F^gVP!x<
zp~d0b*S~y(*lC63(w2mMa?4mIh?rk#bF4FzEWqOz3r(S3OKjR}eW@wowbl%oVS>N9
zj5$B??){Kt7IpC?G+?$~kKBLy9$-E`1Ic7h-0mT)Ig-rdT7o@g&7(2t!3We;Qdz6$
zPJipD66Z47adf`gmF-QF5?qzc$cZ#}j{*)KGgA*;m{h?Ji7np;7`;uOJnJT+M^Ni3
zw`cBLtj9^`cgM?qC_fO%o43zyl_A9-xMcT4>Hn}n<N44+vL|M0W){=ktzT>49hb#r
zpPyoEW+ti;Ye_J|chxnJ!EV9r>V-$b51y;H(5rTMo-=!NM_JkG7&C>7&fEza&Xjg^
z2!2`r6ez;+gUWs40Hl_q^4=?9?oPCQ?;g6iP~dE{yV-F5e%+6dx1;3`-XMBetMb^s
zt+Drw!ZGwo0B!Y-FM_tqY<07p&(AMWjq@w%OXu-i)tb7S#56R5trKy6+J6bndcS!(
zj*Ar^H!wq8^sWo(9t$;CDSG|v%EsLe#3g|(sOxtDAiaFAjCGtanz1>?&VyMUg@5KA
zm5G2XWeV)M?5sje?I_ZPZ6W^X@SW|;Nf|&m2*_^=nK|hnJL`HJnx?2b3oPB}e@q}g
z18B^T+nyDFW&V+3uRmeKc~;DDiyL&md&Md_-Z-c;0lqDtwH)-|H)qRW@&4kO{67f`
z{^DV6$zi;C=O4<Zc&F>!R<JVKu~y;B_r2@`-*J;bRKh0hoLdq0D#nznExxs-GD4AM
z_SZ1s9F(53m0!{TgK*rU3Ry9$ADY1+HDgP-vEe`Qwt3++a)D4e;V#6iiE|`_&*Rmf
z<hmp{c9+d;;I;5mR%!X(R2DB}NNd$L0uw7W^`eYrOPa_49c{fcIw^`1W0hp0CtG!q
zJMba%M*=#o5v&U9=GyL5?+a%huRC;;A2jfHvUB>lv!8Pqo4nP+A20Vb?mQzc&7Zu^
zUu`+;if@{D?g+B}9{~D51;0$roTi#Av3>O>>nE1cI`Y-eU&PmQmcs%6vquYl|NVD3
z_v8!gT?*{(Z!tTVvU&45#quhqm~i#oEBqJ#*Z+)f{=pyc@y~pg3r{_Tjt!5Vzrfmw
zQ+)PwpT`)(+i$<k)vH%oUS8t;EAMjV%sGDi(s$8nN>wxz1`hW2dFlH<<V#=pHQv9r
z#qP~3=-CaP9-qg%2Hk>LULr(AHQk~4-oGWf2A=*Rgc*TC=yBgN-P_}>%WrY*=2cFu
zpW<s@`8p?0Kg!9IC;6N2e2e!kU1D~y&6mFL84mUjxcdJ4I2W1D8?IcxO5M~r@2TsC
z{rxF#zVQ|(H%?(nOVAK>B)UGvpFO?fe7cX$P=}q$BWy)xqnwXuAH;CRMMM9<uDK=&
z`-<Ueov?##+oF~3^rK>osA1?&%a_R@vuPU1g(|FcfiQ!_<Av`p#x5n)<(NnR#k(iw
z8#Q?C;^x*Rv=2a0SIhxjC^R!-Itg_}`rNeccJMi($JaAx-npLD>)unqcH0W8*)$FF
z=@i$BAoF}aXK!z>J1$!ZNLyN3>X^42PA$uF;rNF|CAO5Fs9C60)%Pzxd~DkvKHKi<
zKyB-!wH74Qn;5vdo^deUON^moKCe3mBp*lqJ>Lpg%OT0o9dmyE@U~;Nk|P)hgcxX@
zOH7^V`oTu;(F&<JwqQ&oxEWPxICpM?C10YIN2!8{mZF*vrE^5Kts$aRNRlZU6_v1w
zNV1xWhM^F0wyRxMUUpswDJJ>;#T-H{wnRsCEYLAhc#T8jY)j6UY@M{0!;2Hx>V`eP
z!*qYjcxee^3QSQ+n=-8lbI0^&0qVS0{Quee?`YZ5G_4Q(eP)}m$6bHZ+sZ6o?W($)
zwluVd1;PzhBZ15UF$+d8YiZ4zS!4bfNi)nsS_48LK){R~2!Vj6o9eFas%lr>WTm$^
zufNBJ+2(We$M@|Wac-K*+nL#2-Fe=s$hap?oQNGe_I~&KJ@4~A4;q{d7!eRoaQ&&5
zx%;W7DL%1*@IwNc9RjC?r8E{tjZARz^7;*YTQj|IhHOyaj&ho#IRY4LZju)Tt+Na#
z6LhAqP06k6*STL+{J9TZ;9op@lf`5QDN1e}+`yO_wryB0T22oK+;{3USDt>77oUED
ztQaskd!Dm*T_n%onJ2#r2(Deb#=ZAFz{bWFr%&y$y}7}KGn>S&iRad}YkcJ^U+2P^
z2Wbv0^MkA0c=2m|a1#z+T43(Gi&AH33DR01;FfdtKK;Ki$wq_=4<JGVE)bmM=y0Fi
z8`sH%=7)aQ4{+i14y~`*y>W#XpL>#5Uw)aN{PCaQ;fEjQGoSgjc(OT1rgQGT`+lB#
z?pbc_-efjkprr0kCbT>@sRtZ8iB6_pzcu4E$P#Nn(ONBYjINT9uDJS(#)i(vj3t~$
zCJxIx;rj|}d^NFF)~p>{9eH0V)fbz+)>V=z5fFtzMfc}E=|A-ktoDI;?dR==DLT&H
z`g6M{u7Q93Bgcxk+gISBBSQOr*rMnE^^IQoH|`Lb5PYvR?^ByrtX;f}wGquot*54m
zx>TlZX{yMPYg3Cg%d+0Twj2(~vYg>?NM7U=c}||^y&W2B1>O2ht_idXA^Pdhw-YD1
z{k@mltGb2&#zt<SZEfFTZJLT@HIL`IU%AArw}qW`IG;q4oyR)6>L#R{^DBMD*k-=n
zZ#ACXUc{o#T8qHrrSJKD-dLmnD<wt+l#E54Gm5$%q9qhsqLoAfPWkv84^l=Ul}M9F
z=ekc5`N^3_2#J>~lx`h?^I<g^8`l4KEPr){UB^ibeeM)Q4xsRXOlOJFlBVQ+AlF&6
zKa)D~8UtD?bn>26GFc-E(HSOw)3`%UokU3xGK%&Q6z#0#<<EYF)A!tiym%K%c+UHl
z?X3~s734t@^njwAV5^FzX;~cA45t%hQIZ);y}yT2n(@Xoa#e&N8;((#V^Fos_ir&?
z?C~c)c#hA$vd_;w4}<d$@Y1d89L#nx&U4sQ+`4szbMrA<!zor<_73K}{K6}IWiVxH
zYm@EmEhdu*hkJVrrvt{L0n2KMk|jq+Gv>27%Vo`(bN7>5$Kj=K@bZ=Cm?}+s(-Sr%
zuC3#^cW?-6an?{B9kRG}1M{o@jI%%S?_yQ74-wwdEax2Dy2<x^{QG(6p8E-=;_|DP
z`Nmg2$KH)gT)+AX!G0W_1Qvx5EEaR-iv@z5#iHVwr=RAz7hdF}ANd&H|2^Le0-Cz+
zzhcpyWbXJ0<lmxK+)4b_MW0I6(7B%OD_4bFhd!xfV()xh#fRXM<b=T3cV;kVbwNp6
zBcix(*Xq?qk!2ArB7}bXx}UxF2p>gLyNgof5ycNE*F<>xB2ab{ucD{E04sI>es77@
zwyhO#ai1MW+E(vN|K2@9!10xK|MJ~Gd@KO_b15mucRau{FK|BUkS!Mrn!08&Utrpn
zs;c@nn{AUTK=6!4BTk(<MOl_)S;llaCC_t4qtU96OJUY7h8<$pT?}uZnq9sFk|rm)
z{k@ml_jmW1DNWNf{ZLLU&J4~OtZlKj#rZhQbBycB>Rjpyz1umpw?8lb5UCLnAtRpa
zBH)F@d53ha=UQ|*(L0YP7Jklp1OaCntdSfY&I!(;L@MBdM#;#lb=m=nAbh8n6x+f=
zA$5GeZpaG6ZM?duQy2(Q+Jx9{@7l~=K3U}ti4dzHCRi;p%c8(Zjcr}x+QtRA6p}0(
zL@r>`+(QVaqcK^Y6VUwvPz12%U9bt$cnT>9$fS}UFg~z&b)OeL^{YJelRwOkt2lkI
zPazC#o}<$T5M;<gQRD$hj;$;*&uO(H%L|r!`{X+t1QnIKi=x0w*x24=d2kc!1Vvr*
z{r7J0=&1?+-!HsCJ1RM~^(Z$FuCq5k;Q7jMsoH0IcEH&pXPgf?H5kV)u2dXcYq)v!
z3XL&1FK|+!a0r#5bwPn(>^%9=b(Xu=c=hHrDy6xgG6vz0@nC|p4Ngi-Rgr0x1Oy$8
zsTkdS5q0Zj+?fXfj}Ml?pyZK<A7p21gXf-qhDE*L;P4Qz+;`tS+;`tS2;q48sV8ab
z7KGx(moM>azxFvUT)3AaFWK7KCeJQ&_39N~zVtGa$&~qW$=>cRP1B&d3nHQC`~bA?
zfG6Zh{N_i`=1R)uk+2(jabBe~It>?eEx4}T_)fHDIxdgwxp3X;G?JmCeqAC%uN7BP
z1wh5st(T~fP_~YP7NI&IkcRf@D!34m4G^J6`c^CAPA(y>xYO|-2Ow5R+p%`(8s|kM
z(+kpS9a5tup>UQpf6=gd->#+Vtc|FIbMfEavu_S!o3&|JEEY6PLsQpuk_=jF@^ZlD
z))rc8vOFixGX}#UT5GgY6lEFtKnY25&cB8e`G$U<v1=4Z{N4eC_9VBz_j3E;;UP+>
z)%10@0P&VM1mOCj&KUE$2;q8Lc)dORX1DI|faI|owg0-}F`mg0rAnC(h=;mJLmdHu
z^MR_W$+V;{G({0Trd^f>J3?Gs_IWV?G9ZG$bx2(ZsThnH+G>I|zNZWlFOF~X?IFb4
zVllXsC%Xw>L2Et1#@wCh3@H^-Xq@*5nGpaNpmnk68x4mP<p7x#NTm^(Mp%y*Z_P#I
z&WbweDWqcV!F4Nm;gJrmfBiXjE}rK8Gb6m2(Q3e2Y=R=CQW0KW)`Z!DP-;e}r?grl
zwL^J>t*U5^;Vt#jP>iNnADHf(Wx01iE+sxV?i#}%`p9|Ildtf}OZ!YlJDlCvVt;YO
z{_KdGrscY+DMZci=n$_Z<7)*MPM_oMhaO^Yzao>GqR7ZpM%xCe!<&So>r{&cFE1)y
zX&M@p@sW*^@odStGZThohIa<7$C$dt56TEVTTA9W6P;tOUqPICh!6xKC??}gwl+5S
z`d2@X)Hx{4&0Dwl+BcryXm&)FXFT@4_p`CFO>h<^17)GXwk&6R6q)4F2Os9%yY7kP
zRAbrS-)Haeh|hiQb3FIl3;nd<x(vg~h{SLI=m9353Yt{Fbf`}E?_;tZ??CiI!O$0B
z5EGX7GEKn6skm-CdqgpW*d9y@^?kd=u7)CFp?4dq7P`;MZO3lcph-#fHEr3RXt#QP
zs?J2W13kwIx(?e5ftcJO$&kbgi1)5f?%sKvu~=(**%50kM@L7IH`g>Qmlejek#i^|
zc~PMAf?_aWb9<X|P%;^hdY@XE_LWY^oEIyeUw1W8BBGZvNjr4Y&g;lz1QbI5J-nSb
z$?fmG+&-Vpk;oOQY)qWk_7))mLIkWc_}hxQ+bIrROStFCL7XNJ$6Kwhhw{bqIrK%H
zF(%SDL_8mbkOV|JcN4moiX|UBP1Dlknx@UMHpf~Skv$h3afFCDsS_sd7MPKbyjn2E
z?T$kmAq`m&VqTC)ycet&rK>Y=Rp_k>!tOQWv~(~Yd`P-=aR5R}jrX9G#)W`!9$JI-
zp0X$@%K_P72wEeu3|jk^J27|IevMzG^8t^H{%S<ev~20|&pgA{^kK3r(AEvg3cQ43
zG(ad7(FNzJ7E4r0a9LSW&F5$(2`0jt!8`Eaj6wK7F(@%bLF*jzy@tJ;hd2a3a(<IB
zfnR$08aJFI%Ss+Ndk&dr94zPT9UXDBtP$3dInUwVO=e%4V_M581RGM)=89LAORmgn
zD(7fiz$w9ol6)}NY%Ldj@SNg<=NxvqkKVi+Aq2U~@Fq$WG>avfljK?u_O27m5l|Am
z=V*VA&;RPL@Zz&iptB*_U;~v62}V;dDz06*$|xUk{(_}mF2F}6>ee(|yLp+mF|@6r
zHI}MwSS;oo93FA&@PMPE8MYVSe7!hnzy6Mkeh@(QO8*h1i!IT20;#L&ib4qTEF;qz
zoyjDt5%9s%w4SVugA{`>q$}RKivu0Ets8n!Q7fwB5=$vjXk^d`uaI8H0wzRqb1-hj
zBTT~=L53&}FCs)02-t3d%?ETh7y*fj9Pbqg@r+Ply2U$!K<YSc*t8a7Ev>OQw=z<9
zHaa;}RYg@*EEWq|W0=otlok|)Vmux*nryPMF(of@Mx$YJ{K)!&2dP9NgFCDU2;r~^
zeT)tlaUb>bAt7=0oQ=q+cPJ^NB!qQXW3a~JjAi*c?iP0*C%OH-mfLstcF9$K43#DX
z96ljet{*0dYs=7!I=|7thD-|;n-+%Fxs_bdZCl?@_9k=tKmaA_CRa_egX_)<DZqt5
z5Fn*S3P)2Lma0Z6!Dx_?X^AlwDFnvoz6I)%!^3beAmDJ$AVfs_r6lTdLM;9|%QPVn
zolRMA(McyQV7&J@)22a@6<sCIb9zY{WAWBei0FY$=Xl}c)U%Y)<JLM4)}nL<LSULE
z=378!8R`xfZFf}E^(~%cgxlS`5TGRYU)Vya7CWEgmn{Jucc3v0CSydNA&VTPK#vQ8
ziaWbC2AM1|f(sxd#<Y}THI+7=PO#o#+J;5dGTPdtEG*x9*N_XD<+HCYd0|;`cx!=G
zlHq8?!}r{U$Tg;F$ehEsj>D?Kww~HsrYcaig3HU6c?f7B803nPH+*8N;G^3^&W$YN
ztj3v!$<`@~ydcXmGNlndlGK;;Ifc#<!Xpk}#U8zc8{P#@FeoGs-G4vNe&fsR?cX3T
zE$1(6b9(y}A;iz~^wUrA^i$7rP#Ib;xODXf|KxxDSJ=?$108P(y*9$?+&`(~en-U$
zg(D@Q<5CxKI6?Kr-8<5Lb?e-n$YF_-u#unEB`u?DB<Tx*6bcujfGR|(1l4f@5xrqI
zkT@!-=u2%x(c4U{kT%xt;y_6{&O;o)2;EAXH2C17AFBwlkc$@|5RRB6W2Hh!{+T`q
zS`%9zjqQp$gK1l=v$RdrldG$mx~XZJnyRW3Z$=<xTy+nIBZk8f(~V8?B4;=nr0+we
z0g?_?>^{Sq$g+r_W!J{;KX%^^S~+!i=OK8UF<4`0n-*^@<{jYFo#giSUT!z7LFICF
zzV$7{$kmMnoj<mHT|w8O`M2{0Iwa4eVXAIGs6+5rd!4A$=OSt*Q^5lfp$mfG=!-g;
z78sI%6aghILO7H#NEyg7k=nm;IKz6`ODZW;0{2z}4IQ0heLqLdJehu%9`4V{b(f#x
zU6J+gZCy&qcb$UOANKDh`twa7L!!?lZeB>@)23-qUGXd=*1ikVuR|?BD1^7YMUX}!
zy=JrtXQmRncR;X);6ch*VcM#~0J10%r6wC_vNFR9Vhh4~8q-jgMf?J@jAb*&xfU4|
zihv1@EZ3a7=K?zy&qm)p8CWitXqmH99<k4WpL_lW&m9`V!JM1>M|hPHv_=;N*=WRY
z=QNor5yEl9*UW1}F7}x$m$=rFiNH@ha*E$`KEux|+GdGIG9FFYnQk#C1}Lq-M?&Wy
z%W2vL#b`v<3aqZFp7|2Ie;eV(OojvQz5fxe%``W5XB5RAi|`_4ImFlnMtB+@;)Nvk
zQ*a*UZ*hFNkCS5Vw`IIp>vUb7qbur^)=@N6Ym|zURUKldwSGqk)PY!>S?|CJWbi%8
z(v9?sT4KFtP}e;o<$}Y7*hUfvhUKtNq)LbgpQRNj(4zz1dwRA==ww3^kI%1N)I>ts
zdY)lo+z1FRzPIqP#bRyf`DVeTf7{fGb@AfiJlxnn!Z?r4OM-~#!F+y1UDw>Yb&IAk
z)U`!t5}gSKg8`GtnDcjCBu~?-gW-_8$YTCVc2-^Z&|emN4rG!$K5kdWdO;WZNkNiK
zO0X^tPq?UUWm<!28f??PBS_szZh!CPcJEz$>2!Yi)Pjsc!8XcnL=Hoo_D)NaZh_KG
zkUC@206^rQdgrOzCQA9#HO87)<e9h#Y>io=%`QGB@_0hKgrJKcBC;O@DkP*%N~G`z
z;nBjPRA5j_cDAQDx6SEOQwGDF*2WeqJ`RyZeMgnhyEgK0RXWc&&J&a(a_o9~3oA{)
zPM;5WE#)EQq+?UDQY18v6=$TU5vPG9L~3=@3#Mx&2;OmYu-^~I1@AdpEZ)-hdPhZF
zt_(iNbb!VP0=)E`-I+pL(OMhFZ>7YCn5?Cjw5~-gEW#>sJ;JsUmFKZ#Bm~YHpo|G#
z#cI{IEn1ZXA@DXrnYk{>hZAxw!FozPWK<3rmYRnGJbvGY`K4!WaP!cyyMMsEt-)06
zR|^iR1?ucJSFgW_78$2iNwGNOeLAW<i#+28A3MW?GBCfq3x#E4JYh5%F&qw=PBsu(
z7C)PFWI`YkQ9VcqQcz0G^~+ZnJa|aZCDuAV^Z76E>fS!B^DLG%dzW5WeaXCLv2zmd
z(Ree>wRNO!omX@F>N&g<MO}KmNP<tx<-P!mf~3Y;5R#^i7bu<Sew%$l^rBtPYI~2%
zi+F-_cmZCB^xOp22}mAgY2|(!A|O_UVV5U6<e=A}a}n7Jop)xZ!{t5R`q*X+fuqG7
zV?9DFqZXfYESFIrxT;zLfwIgPjs^@zCDZAIaxh?PYl|$4(`i}Ko6~v?nd>uox+td4
zpSsEYw)a3x2t5i)2zZ;?R^G?l=c22Fv*|+F#vu(;H&pKcQg@Qu-+Q^;dx!I;N3J3#
z&iC{KAJQ~ySGakPadBwKJ4f9%IO}MW+=+1()5M{R)<k=|h<Zoq6I_~>?WRbr!+IO@
zg^q1ipZrcicJMVa#6^!35vda*poKxnK$QzJEtrhUe)2Q~@XpbA$6yd2*ICS=&RK*4
zxmL}_0x<0q8SU6Q{mP!QD?~&u{cU+H{jNWm;}-Nq8d5~Dy!87(=sA5W&fwZYb&VVB
zoMk2YZmFy4Eq$+dRMc4|(P=_UBm%V*0v#NMH0VN*ZBCI&V67#~b1)V+TSnq^p>cR}
z9A>#B0J&BSCS&&JH^}px#bSXL!&p^U4cW#Jg+fbBuog3$ktu;SpsWWG$fRO2++g0;
zlzzdFKeoloSNGtNyFup+Mb1kvyv*LY2l(zk{%82v&-?-no(D%0_WspBWp?ubG2P@t
zcW;p&-N1S{bM6#*DHxBZ%x817&T-C>i9DK`yB3khFJ(}cm}SGVS~4leh<1rz>>@UH
zSS%{O{`DuZF3<@R_GkKWy@r!G3E((-w<_$`wCaTHSB1CpTiyx%sz9_b>WdB06;7dF
zt;VDssFM6bUKa6MCd6vcAZ1EPR96=PyrggOcn{7d?ublCUUDZ(Qx0jh4&c*6;#mId
z+pR8ucafOvqmsRK9_t*nu^8v5n;PdlZDaZ(xLQ<^B;K?%ZA;a(u_fz3DJV<LWIAML
zXPeEfEl!=@rYH*XvRt*nRp%tX!ZAYf>Fm)&Ry@CBHbosHJMCXOEw5xR=94h2caEm6
zv8|<EE~5rs+tOAwi+7-?JIU?uz1(hHbUEpANR-Nnjt^at(b|ZnHOq?D7?yQ~X`%+>
ze7U4GE!BJxX$oyay{w`Qo_9TUASpXW)Qec?#C2N|9TpykT_IX25u_zh_g9H48={RF
z1}OrWf=mlGrh}dq0rA{#P0MgNh=p99BLrAyQa<X=MiCc(qHBN0_x7p%j#%a6PP(Pz
z?)JPp*?rd?|1R!dAtXY@m1-a)arKxNU2rM2Cfrk`1#n6w@jF>*61Hv2a``SGb=m3j
zv{$Zzu>FNaI*$^D(-%*(ap3{VNrtZ)F1`3Xcc0s)UDk-yHW+V>F*-u<(k3HpAsKFN
zVVW9aYHVBM{TQ^uw+4~O%y!@qjYZm)*|kdyrc**Wj)ctC;u{ODK}vyhurnSYgC`qh
z$jvF)=KGkRzQ{w5Kgw?F_|&I=0e5!9=$>=jd+!dLE)bfU>D}iU4l=y0$dtl5%e*pd
zpB*t<F4&oj3Bl1e6~&-L3dv|R;c#}HX1-u^J1|v}8?U~|XmmffZtQaT+KqIwcZcS;
z_Q`${zj@IWX2&KQ*X~zSX0foVl9+AuXFcwm@J_W~1uJjX-~*!bcuh%4io~aJmb$L{
zgcpS}qex~5(RD$ri|9%a9lyr8xMG(QlGdJ35@d)8R&*SkC>Ba;s}d5IE(%xm&|VxS
z@YoO$IAcAnwY1i;Y-(C-SyW3}YpI$#p$#6_L<9lv5i+1N$#f&yaP91zVmKVIwKb(2
zl;lN0mL=g|I{r|=_o~sUHPd~!goISCp>Zzs93{Ga;a5a;-^oL)7j<b{tHlyyO!SO3
zCTfc{ElvCO<t<Kf`+G092W*^Xcai^A*GVYbv@Gk2`C>-h#P;cOQDKdtYAT$K$7O4h
z+(}~+k1Zm~!G~D1ti4mW)B%yElv1AQhBDTn_kN)h)*ht)8@k_yxcIUzB8Wl?vP_fZ
z8fT)HYSY$Kbw-|L(V4+Vq_0y~>iEk2ijqKYdVOp=L|<TuM23wnU=c|Sy|!Mz3hjq4
zP-y_nuNHjxI2s~^Ob9)6VtO5Vs<SwhAv`EAkt<fmo5md#br1jG@8+q0{6CPH7BAA|
zh!AXSZ}GwRe}rL{fnRX-)oUExI>6m^8le^C=`EHAM+g}h3=7bCTzy&>SLHHiyuHQZ
z#%`Q0aSov*zK!h%k>zBg5zG33;b_88!QRcA<l~Cr)(*vFh|DyL-387BvRn~D!=kA<
zH7qHb3YBH}%g^ztzxAI}pS_P$TNCE5UPfNO$#6@vJsBgXQ?^f^=HS*%w2BG5Z3Ct$
zK`M-k2IbKUukF#1;b6$FHF1PmOG>F&+`7u}k(Ryvs2EOvD6YqEp9F9c?{2Ka`Q5*z
z>_jiSL?V+VPb})9q{)hytMwfYv%gVvts@`s!V{##cdOC}tM!G5F$S%5G%ZhdU)^9p
z@Gg2UhV=v;lco<YO_q67N~$rzbwyoDrJcM$8g_{JFd=w07P|p@GjtmQE(Dr1$yyuB
zvT10oVYaMjtzp>}bv8<N$N*YOawW-g&Bn%<@p#0A3uhS)2W)IikxC(zNG*{lnb57e
z`;R9&lt%VQ$Ep_>`YTcBTFW-A#8dI?lO~?`vDMmp{922*KDJ$rp=p|^^lxqN?(ufx
zB)7l!a=UZUsi94i)XU`pV-0mvb2Oi^cd$?0L?P{ELbRK<PRD9Ql6;7ChB(3MkSGLk
z(i<Yq)-8290i+PK2?_u2IeKfgTaS2l*w#via*dPQ-PVe&PBtW@wr-RmYS+$@G|B6I
z?6eZK8HHH;zIu9VI>*-sC(`?M1tE0ZiPvk%Ms%*%$m{YVAqrfFuu2&r=CTOZcY|07
z=MTgMv0UFlZya}Y6*&9x?_>X^7rFe|PooT|^p!pQ&;>GE)0&FF0k1yyET>QJaB%Gw
zTW5C2hXuYY*}Zxd9|Dt|Q{>YrLaNmTL}xTMFuQfY#f>eLQaHj2?v#S8ET~PxU}K9@
zdBOh8tIXYX#@naR<$$g2(<~1T5hBnwOPaQ3v78}{L+2UigW>4qud;jN2Di#B?DNm^
zVPA88>pZ8nPEoZr&NdW9foUsl+}PvA=l6N!u}3&O+9%T)2zr>hzsL!}Vo{MF9x>fN
zq#P9t^9({@X`;X1xBe-962H}P`}A9vh}Jzo+cvg8WtuFNvMaV^FGT!yLrK-Q7eWx&
zWVaKw(-d0Ch$J~r)6_^wyceKVLgWHTpg6uxEJ+0chrkhnNtrWFNqQ-fUPM#x;N$y?
zZWWxk>kfhk)5eQRckg_lO*&4-IjnOmnugXGYSU7+4aQk&)8f3xrs7Zl+f&V;$k?2W
znM}rPZf>x(waIWeU^pE2d`9OCt%-ai;F#yFNdFP(BC+}oR)wq}*%XE7-5Y|(H_<aQ
z62_yju}OyMritE=-ucK&N?f{*uhd=iq!7oPD1M`HlH1>Vxqbg|4{I%p`J837WPfk3
zFX$o_A@V4lcZlS!;FB(q2z}!)$`z%5D;2$=T_*+N*3s+kzwv((**m+e(brf%Z{n|d
z;zIxTX>G}{)MQy){B}j%U{IiBV5ue68J6{eqR7dzoIKYAB8?;vzhf!Y>vA6h^xV;i
z@UhkziJR2<(~86kcHQR<A@TFN{16DlmTN@pQZmCQ4xvE!D2(jWo5%bg3rnd%W$!{!
z7fzkxkstjDZax1z%_~<yDsm;*8Ck5kfy^^zS1+N1Wjq`Z>V~7s*9f_$oi7NsMTEe-
zHVj%vHX0$7L~4aln!$L)^;fRZ)J?Cw<%92iCG+8c>afC?hI}yQ^trn@xOI*C#xD8J
zHf1?PZ%V4=5z+^g^GFd0!LnGiw5~?X4=L-M@r4O>E_m#rM>%(DgMemdGQkOlZ5np3
z-{8iz1GYEMv2_}*-<mPrzK{sZ7S}<@LdK~;0`+1+b!#7OMvTvGVQn<>{q1_vpTxID
zthZ=a4h!AmT~g_Ib#K3@lgZYoYjMP69tA)Zi7O%`L5AL8!dip(Nhh!0Mi0oITZ52&
zyQNcY7a@XhB1CBg9|G2QD|zThcFx6{cY(Hz$+I=Fz_l(=HwI(lWSw=6WgT0tjY)-)
zci7-SL}6d4V?jKfOc)JwHYX9Cn@lDQ1_ScM!RgfDJ48;dEBwRS3SXoab>cksz2Xj~
zOE>Rh+tvAgKp+-v?ii|Inj{6`d>n|d7SqW?_?2~7rZi3gyfqT_8^uX(fA8h?Tf4Vt
zo0`R9!LnL%G>d}T%c_b)4>q1lQHZ>2dxq3DU43%Dj=^=ZN8Pa0+W&5k+aG)FPcfgZ
z8m_DN?HjoAxDaw6Bqe~T+u7OcnIztw=ZY>jg$POa`L#pk)<x1g{_pM77}mCF6~}XL
z|5VZOg*yl^^q&S|S3)GDKNgZM?L$X5JRy4I?wiM*>C`Duci+W>zxT)a^8fE2BIl0D
zAm{9bDdqNnYBA^V)-BGSJ&nwwDL2+&XG`!NC3B=}F^d^8iE>gzDTG|+h)i+q$S|5!
zoY@-Tkx39SS$*Ugx*Q^d#|J|wGB(a#q`k2hJq9-gx-2Q>lwvlgsw-3gXB$k@P%RCk
z(S&xe!K1?|&z--UuStQb;B3}lq#)!O*0xkfM{EoSJo5MjZq9a*GDl^awrL{ICAEkI
zMEKcxlu}sVQdNp<Ij63U&?-kLO`zqs10;VEzqxTcvi5pfRqHD3Qi@}L<NZqC%G$R-
zBmqRZlvFtSjt7!V?IZ6^AXIEww6;a2mSCRc<eBf2qVSPB5~<Dc;t*v~;*I2>5Qj2i
z@^)@zmlYCH=sg%0WlF3KRIS50Pt!PTva_n|hNd-9O2Imss_CDPh>PZ;0G&w&MMj=!
z@?3NOJ$Eyi3>ggz@;s*~ieAp3Y3lg7@lmQo#S29wkVnJ!?mHBM)C!kThp%h3BuzY0
z5sphdG3zi@Y@^m|Ua@W4_WE9(+)L-p*+nN`qqSyGl;|LU!?*RFpXB!UUT%Nt$tRf4
z=TuEa(>DD8XadmV01&*Uh+B`_3pVTRYLfI0;#k@XtF~)MD0f%T#h;UgqY!ZrAq`(y
zA9=be(+F8G91d|VF6f%3g5Veqi?~n=4uXsx)oFvBOVg+!y5|T2-mMR;sko@?2Gl}0
z)<ubnK$#YTUCUO0j>Vi1iBr4+;OGm(HE`dDm>a$KNfsr!%cyt3CQJfe3NC!&<6L?1
zDV9I?8ScLK4AV11oVC=u`zWO-r&An(Oy@W+P`Rem83K*BEq1X-wSNmGV6eG`_nM#t
zLFF9UoTDaiR;+W1TwL`@i7rdLF(3rqI0TaX%nt1Bv%Iy-V0#-q8ZzA4=3st^3!rsI
zSwcD8gz+|?`^qbv%13N&ZgHKA&o2-7Zf0z6PSCYLkYRk_frl<&rDsu9Oeb4We=8N;
zD>HKeA)uvVc5q0MWfUr>sP?(}>K7>Vnf@a2P6)%E#P6)Q(_><Nx+?zs4sQ@q6<!mK
zr-cYn)FdUb_<j;8`?d1gb|ul$WFh9Iz=t@JF9LYs<3NLRm^26wxp^tSB_z(Zz6Z99
ziz@Q94K&u%TF=rrj0@CeRn)a@6biP+M!t_r7Xt}pCdiazI?5^YjPYndk;kI0%yiVy
zBk|+#NTpFyk!2bPD^>jvFt%Ooc#*!pu84C1gvSd-r%varOY$m~U_GHVQ68c%<}7tx
zC6q29o-R%x2A@!Z5IKKZXGkR}@&cV{@+_wul+-~0m*3uZ&Pi^6ujTgrqv&63Y;-^F
zCNGszI3FiS`(a+vC`gAKb?6u!XRK?Zik><VpRg+Vx<P?1UYASPy+`+^-TSURt^0Sh
z&x;fVSmnlUXlyhZ(zG?kv?=G9=q)RwHr(xs0kPd_`=JLZ0$!|&Hl<=4xK|qpZq3nW
z9mx~N7T;a%COeLX5NTLM(9QPB4)qP9&n*arSo@QAg;N(oz)2`Jwz=mAzK^R<zeKUK
z#S_;QV_P$*E%{`OC`-(;ie}bj38~*K5-&y1OP0IWsP=Y2NYrGC$pl&o#)BNwHo$mI
z<$QJV6ItX9i7si|sCr%QY$FD9ma`edtY9!0vwi9exl#=Bg5U&lkaKzOi0e1@*gk)X
zjmem;dmrYBuYZZz)f@c2ykKJk=JO-+!G!VF7FTZGL?q!t=WMj<Aw*BcbclKH$&{is
zj(WbtPFu#_vHQX$Zj1s=+oa?#-kq!PNxWy`H{<Y8Qldq)j3OPfKE!F(Zn#!RnOd;`
zqA%(?4W@2L!uzxWPRTYB`#r`ct*Xuh+y{accpLDI$NNYcuC2#-&(e5m=c$}yVH~Xs
zRMz1_v?Madra~yTGL(-C<V-=HE6Pl<F&Z)~ayG|fib9hYnj+KDnP{yoECeXM8axRh
zT1HtH{V_ubXsy=3ndqVslDkGM595BYt?m6|gLQ<);M%A+*CBLu9bGEAp@&{`aE&ak
zq@qaLZ-c?07kn;<1N1v^Oq}HQ_g-#4TFko<9w2rHl@er`A{d9iZKygx&od+ycConX
zZsE4#W>uer5bjvzC9Snm5w|)Q`T{Sc0jXmX{{bOEONAPiAfh8q+eGBv7=zLZtyk_l
z9ZxXr1$vI0B#9j-G&z0E@!X6#Zndpyt=N^vTCd%bu_GG`NzXMDLM6wH)%HZ+TtNrS
zg^mC$-rCN)qoPhEBV#EuMmy)Y_Xj?~g~^m~u+NptH@Ii#G+XC4sgHIM!Xe8%dbtr-
z&e7*1P!31r>Mo9UuTve|Vt^+UBd*=pC1Z(c6w*ti(x5dU@kqRgyc{XP3x|*yv=D5Z
zcvE@FXr!>FB_9o#ZtT#^=YYgzk&u1$(hEF#I>)J;ycK-pd%u%!`#nFzFaOW~Jzsv}
z*Lgo94(BcRe8&&4yST!_dq#sXSw2hz^3;o0QSmw;F(Dv?K{&<YpuwF9WU^rM)cd&f
z+~p)WAQDl$+nLAi=-3Kj9g;n^gn4ILYbS9M-z=1pC?fhK5|_{W5Ye^toL)vnWhkY4
z!~#}4G(%LHhj_k%h-j4aG{z)vRg1N8n$#Nz7JMr(Hgf0|t)X>}nK4w>v2>P&h1Nsu
zBAV#ZdkP>|pe2lR#URs6@{Fxf!6?r-Gnp{Vb2bJgS*}r;f?N<%b=v7cIp_M-cb};H
z3mLsiJ**}8&`ks;o}rCg!=`ECXKW1E*hhA)!COa_5V|(;=Z^VluC2<;vLwqg#^W(r
zmN6QQ&{-DIx}re4w_i-0<o5SwZkMtb@JMvefU%}OZ*PwXYWdryL01T#bFcj@Z(8f#
z4n7@220K#5I{&WoXNj-dw&1;p)9GoUprj(x3Y}*W0~3m-Hdq^I>Xtl-g_DX=7hD=X
z3yJO#-(AUo(kj^+c6lv|0>`4R^L*`wmAW=;e4RcYinXFN)&4<r6TPZ08au1RPD;u-
z*Y7;%{ac=Lzhk6MAVERUn!6r(m{SDsOGY~r&Y$06xG^Bqhe%Q4LI4*Dic-YUVM1)b
zpcIo+I~?7-L3?lr!@#wxdt@?;EtKF<S%%EAm~;s!t8v27rrxz8fYt=%qqd9lWJQT>
zTd<x%UJ?pJ)3$i6IXt*U2sQWKf062uM_KOG6e94*`#;7#=kMZg|408eM_+xOM?e09
zJom`&BD9W;(o+PFwsTZHhiZYZ4q&;D6_HmIoMl;i=5r{tLggAK0`>kiuD<jlB6xBM
z1Fg|{#&~1O_RbEY;gGso(l#~A<&s%dv441kwH;w|?I=HqllV;vDP(F0kdTYW<n0PM
zNC{VpzMzBDqV*BALWrIl*O}G>Nf6j$eY{_xc9Fm0EVwp$vKkkdH(gO@shnf!97_kr
zL*vuMh*WF|$YkW%<VD7y%o&vhlTpbq*NjFb10APbb*7M7VpYJ)(6MyVP((m@fs$*E
z9$k?qMH(bYOdqkjSa+zSv9XP6l1-9J;>N*xFuphXHm&Iqx*n}ctsbdWzrV|}Bro!a
z(B&EBAlg|et;zEoW8S``$w_WM$?YBB^+pBVdffh$zD@fZ{YbChiA}!>#}LIe`q_%Q
z(MLkcq(Ui4rX4;DD7nNri)k%dY3QbN(`(C)x0eX8(epae1B8eLeh4nMX2aUzi-bb@
z)UNNQVUK+Pe{35?D*gzo^SzUJiN&!5>HCS}?t{M7U3<qxol0V%AQ=rNj7`Jj{zuRk
zFOtiKdjAr^)C>nDLa9WCU5TqIDdWo88oalRHny1W-oP%8qJB|hm`sOJ&Rr12OTEO{
zCAL<WO0)C@jTI;wveGfsP)HTYw@TxEi>fVZ7|Fy{ThlfbySq1;Oh*hiCTzX`{XFrx
zZ?M023n2u<Qx~}RBj3r>uRP18y#t>7;!8a9wVM>94F;2(-}_VFNvkWKyL5@2(ZlSV
z*W7hpbI<NKpn00AQXDK4n;V*PSkU5GTzUe%sQLB}e}wP)#K-yW@BRc2J^V0d&Yfd2
zos#DnZBx_K(JcJn@PIFV>5F{oQ@_Zs{OYgq$}3lB+V(YEgp)XlcQdq9XsK`^B6ZG(
zs6#dwq!E5V#Nux_8X*WtyfoSdm3cu>fy2YYw2HMTsS!frLTu|cE?}+4*nqK}G>N1!
zf!Weh)t1_Lj;e~*d6of8fEF-Bbgt8nmm;EbQo>-AF)nk?Z*8zS9I!PSuwCX1v}Bwq
zG6AIkA!DD`rjO%8l5LP48RD>mlv(eqyONHHH=krl0=^$;Xp{6v-Bh$qL({a>b%nDI
z*F*`JV0<j@I#cw7&UGk5S(apZMCEj9vF1gA&a<AslW83Vp*xcHJAl-k<o5SoZg)iF
z?VWo6-%vrf_BY4z%hxttSDZTEpZ7||_GEOvi1vR3i9jQTW;n<h6am)+s^yY{gFQr$
z3`b*38|}@y0jR+s@6WNeH8>l!B(p4IP!4E~!@Cd%$+`t@5O^1nG%p2OB?^I*@q8!5
zMG0$KcY@9^C(={`_@vXhenZD$BIoVv7Q1g0cV5&blV<5an}DurX0z*T4>U&9&@Rb^
zKud+lbX>XHxGE3c_1a6)Ic(h^vW#q4VwVfX1I?|)jLk`o6bdqdaR#@TaX3H2i#Ja5
zN<F|#&T;)<$+!Sr?X#^8c>mcMXEzK3<tZma+^kBY{|cnw=-?Jjvt&9w!*DR=zQ;bm
z#`fK8o!`Lu7LiZ*@b`QV&-}~(j#JXWa+iglQEYDEbI8XfsujHO(ltJRvqqvge{sy8
zc+BJ5!2Vn@YZMzYXQ^{;+6=#ajz9LhzK4y6KFDw~VK^Bw-q>U?8dDC2XsyYMA(M^x
zm0iT~{ttYJ|IHu!liav*lTUy8)BK}<_z(H?XFkJhvBaNLb|>+h5bG=ANTG>JQFQJI
zNw%XyX}Z7are%FfSdm{Ahb2OY8f`9!ctc`>#CS)Y@`Evfx(yr|&oT{0%w6nS*2IgD
zWz{a~wod|OrYLj8cu+7db0))*(Xe1NEGTnLp(UA$JUhv1RV_toj|h$xZMWaL)(Uil
z?3LU>$AxR#Hi`q+HEr9*pA#DAlh|=E4jJMlN^9Mlerp{kTFatDXBmT0%nw;=v&yw1
zuG7ErX`j9$)^I1e{XLo6zf~?N$Cc>1vwA&NG9Y37xnlJ@pIV|mQ6WM|@=Q{YXE{E1
zl*Nz`1dG`MA)`c**0I~rEwDPP!=4QnJk}Yy)2w^WdKns6@rH$vNEAUPzA2*L)V|M=
zxQX3Rh9JuP2qF6+qK;oHeB2(bv$wP_?^uFf2#K?n5FDt0)R9x;krd?!cNhr9MwRj9
zg2BcnA<s}Y&@NkWk@V-4LgpFGazU9Fw4!EQ2(J6UB#T@NXB*o2fLE?R$d?c9p}KGv
z7r*ly-YafB`yyxWyU1u<Vvm;m(ieW6hvv8Vz^Uh%ZVV7bQ~_5?p@d+un4_dZD@~?z
zbTQ!UJv+$E;9N@<;PfMpQf%)~hL#5&yq_zFH~HWGxj)Rt)`)9YZjg;e{Mb)^Kd-!e
z9cvw5`^q!SoFXh2)JCyC6kH1ppMUW(Vf%g#Lq`3JpG1D<Qw)a_&Y!!;*|QfpbLKo3
zFJ5G8dkdu%=g(hcYjX#sG)T}noWFP%f9Q|=5q|u~f0WPs(l7DX{@;I{Pyh0-tXTqm
ztDKG}@jEj@A}M!;PgmH{+mm%f>0{iXgzNzz6aj$?k+{=o$vZDFXxiXuykOqOY1l;#
zRSQQoR0diLwO<u;VYOw)3OWsp%Z%x$U}rpFJjmIYjF=7zro)n9CMgu;(xXH`5qh4S
z3^5^h^1bV}YwQ1BZ?ksOjw{9bC`YnfF4N${vM=y#+a`yFxa#+<BU2i!6vd#7vLs0t
zue&L&$;;w3?+@{NAWb4iBl3W;Z{H>iPICK6Zh!aVjl|K_`ezQQo&P$7yH4VyTQ`QZ
zDC?*n&PQh%?>t@wv=V8^s6-&~9_o6*-mN{x(=p|sB+K$-LFf9yPN~SHi>I5XZ5yP}
zaY2&u+$#I6XQNcq$&8RDX{D2WoA($;H~oziojXuO<CRi$Zasbt(JzX3fm7G52)(zs
zU<nFFJDV(orQ8`J>V0%k;L3s`1e$|GX7h^d><+<0eQ<;TY*I?rQlPVfrmD&E0__dU
z#$#GR<tltFc;)&IUuy5;^bdZFYgZ5X>@PgQQB@N>yztx}#lZ2A?|z6g-~T;ad-het
ze&Dh2G?T4@U?Pb%gn(&He~?I}L1p0JV1E}io{|@v45z2KaL>Igu06*jgX{Z;{KLQZ
zFZkU*_ECl-O;tPYdGIVB{m#cZ*q`&2uf53bA}~9e^MY!4+B*(sn)<~fW?%h0rnTfa
zzWuwui$DJ#|0N!J<o%4sW3-BsGIdk&|M*Y;Pwej<@V@uGpYQ#iALP+TAL~ijLTbj-
zEq>(p{$4)%v5)Z&{=q-uul@D^g6lVK{&pL*I*H##5nDomZd^^Jc3HPVl_HvROCl+_
zCpRaXn@+7AsR%w=t6J9y3p<>L)&}a<Q#YQ&rJ-&uhf7c81T!OPY~;;#di?@WfJUHF
z%8#f~AeV5aDA*c}*cz4$%bcA-!Klm_7m6ZFoPH5eKD(Y%y+Ve>!+{lt?ifnf@0lK<
zBN3O4HPm&5F)h_%iM6X~R_7e}$W9arGL@lAMV@73I%7N@rMAxyoyB&dRvmiR=@_JK
zNc*2uG<QO;!N-02cH<<spXBy;JKk8pcXf`3v|_u>wX6S^OpD8oLo4M9E9?D}y~TNj
z_X@8hO2omYSTIR^upA!kV~jy6nOt#J2pPo1T@c~e@RHU^bX~mud`C!&-_^iG9G-Aq
zrK$RLj8Ue~6JN8_o?I1mi7yydMcrGqf&2{>b<R7yGbr1lYRR}LSnTgIkquf2mYGBq
zIlfv%y{l`x<fnHqLgI~y931ayYD<<YD04KP%z4O4=GNh>CC$MFe(~l7mUoRfyzvqT
z`#1UF-~Um{f#C3<<>qe9+50bY{>+qmzT~<*$ID;+E{YpRJX~C1G8$roSrNF?YTFAP
ziImNP>rZ};>CPFLIz&ET^ZecHz3>HwZq8(?c=@H5_#gh!J|Fvzhj{3b`*`~4%RKYc
zuky?{E+frNKKt`u<=V9wU#kt*j3f63O^uduJ+X2A0)OE@{40F)L*Ioe^C;Y!R@vSM
zo_*$74rjM``O-J|#ZUiB9(&)1`Kh1$xA?#ZK1Nn(1QL~%oIZ7mfB)b6cX{mb$N0;C
z<*)MWv(MkjS9cO8@s^`+F9NFs)uDAMu_o7v^psQprlKzOUqiA1az0>Oh{c?XKR2nU
zt6I;Z@l>tju<_JhFi%4eF7bUNf}D_CB_)}Nez8Llr7}ttm}Z)dykH~G8RQwGOf$$d
zMJCZYO}3GeuUubcN1Z4Y8T+p(>ely3yaz&*VzEgV&-AU<wtuX1eh_Od>r811T~OpX
zI?EW2N90)?fRHK<K6LHue!?$qR73~U1t!X+q?{Jwbm-fSliYri+uyBNKaXAysXLDL
zcMqod*M5K^4H2Y;q!1)qzV5kQ&Pn?}>1D=AoHWHM1yVx3)O1c0o#3}jgB88h&Kkv{
zyD7D%xwp2#AttF6)|ft{OdHv~UiXKg`^2d_T+vz9Z={y+ivICd?EvqLQ|E1T?Fk0q
z8VYZ4iv?&HssL5=S`^M>k?5>I=aR-QaaBt`E?Dd>aBYjuGDIe6o0=@kki~$38d4M$
zjZ$ocoF@)Xb89$YFr8zJ<m|b-xV5{-`Fk54{qTc)u)ILslsx<OSGnu_88*fPzBsFp
zNB6RQaGB9qBSI8Bjic6rwzVvSAP<tcvmAZtD~Rv--BG|$3AWCh<(0}XKD@+tee7d2
zAy5<}zVVGGdFJ_72u?FUoTEj<;NWS#`tmb8-*}FQ(?gQ@1$!O8<?s2yA3*0hU-{A(
zxpw_Bm#<yo%Js_}9Uii~yUVZr%C7;AvMd;H47hRgI$!$o=lRa>_#Xc7ANo_=f8QfW
zDew}M5d6S*e}ez!Z~YDa{9pV_eD!NjBv|*ir}*Vbd~1eJ_Fmm+SzH;5erijS3ZM9F
zVlkI&fdUxs2ndXe;=!%;EE`K}V-aVZr!tPZjklwPr7@m^23jGgmBa{eBA`%cBGQ%%
z$v`Sb37wl}ib4v8T5x(eV6)7bmKkM{G08N!66AtD={w?ZgeM@=V@f4i0l%8;?LM36
zuNrF{x`fcxi!@DVP4D>72`X!;5LzRJB+IgX6<!o2lkp^xu@#xtaS%hN>0!X8Nx5U^
za{^7f;S86i@1ph<ucaUT2I3^QpXB!6n&|j&#~%~D%!^nRbP`l5?4llC_qRzPJUBW5
z^bj0ASiDz+5Q`+KBw80tH#P|}Y8Ez4!+btRYt48(WH1;|6h(iox32AnrPd2P(Vcr~
zd!r;uMAW3{Sido3qM`WUx~KO}K+Z*9ZRa{pa(oRTB!u|c1FORBt;8J<O2pkPk*%W;
z66*tDKBpMxG<Jzy%*lm9k4k)9W9kLPbWGJ6ay=j`2GQ1~UPf+-R@C(}UJbIGjFwR*
zsAeM;+Yj;`pLm?{&NkJpmTNbzv#2XhpL&olf8`0@_kj=a(eHeKhabJb6JL6c##_Gg
z`yS%ye;&B9pR>6!XEMrTJE#-;^B$3Dkec8e&Hg^JTH=C5J4JE&B6~B(<i#(uu`$DJ
zoM)7s;rl;$0T(1oQE>g%EuQ%7&+y_aPw`acnTNy{lAv-;YvSlM1b*(9evz+y{mUF4
z?qY28J@g?ON7r?Yv0#iLgd^k?m~IT&-Z^A<_c~ww>gV}`f8dYv<3INM8IFedyr6C7
zeBk{b;Q#Yq{&oJn|KQK_#5bP2ldte3PU4$K=vsX#iG*l0ow|VOe$5Xu!bveH^?a|0
z&Kc)vjiWJ+#&{OBiPNl&qp_aGcxn@9lT3+AE)?CGxU)=@36;yFpiuEP%oK$Zj55VA
zQw+4EP=Z_vv<N5>F9M#lB1Yg-5!5HjM9NtWL!^Ch8%ozD;Z88wbUJ&9heL?cFCqDw
z%GBz~^CC^x#iFhl4ALa6>?i2L`Tz#U3pyc0-)8lFF}F6%vNkyJw!ZR{+<ubV-=TOT
z-W>3T1?u|4t6_-s`y^`;0_$!N&UJ;5i|5BN&A((Uw)WytV=_*cH*FpLZEcj<X<L1)
z1={gw10bY~wqRXh*QdsnJk2p=FKvl9_IXAB{B?WBcrwd=aT1X_kBB2(9OvzQ^SI+d
z35bb1H*@mp2Ew*vF5sF8QVD{ws9eXTwG^~<Lw;(4k`gZ@D2>oLp>?!v>}6+J#_ZsL
z8b>Y~O5s_w1D<Ug_J8$rTqy73p;Lx?$7lJcpL~vIpFBX83qJFyU*W?axye2Ej%llw
zYnN-z?VMs)dJdMG%ondvWJD!=8)XknQ=tkS?LUk|HVuAuL~AOv7*K4VVYz^^ozp!3
zOC0Pr99oI>ibfUO+_|62jl=9;=DE6JmU`nMc^S@Yb8KI|{PL^3dg)Ts=9CGNP2Z(+
z0T*NvJB){KwOp{byJTy7$Z}cn_x}Fh;Dr}n;LrT&Kga2_=g5PgoiF&%`#-?{^FRI1
z_;>#7pX1dlZ_gs=Bu-*IjujOEiI7hk)knxU>BU-0l1?BqV7$k9sEwnwp4xckRZG=6
z_7@F}am*Xr7jnkM!pjP9skrLSs!Tv8ppb%62}TOWS~Ad*%`#(<NyfQiYp9tNl7SXv
z8Vc!AsZET8AO%wBh}3nE3zGQ5stx;k`ynKrU_|LGF7f6%zD@MqmgHH16cU{z613Ko
zWf_sWJVRxgtSFA<<`6h`A?VPzzWpl%-g;d3Ufs@4zn4lO-$Cu#liYri+uu&SQ9<|3
z(8hJPSvu>8289qv!n_^=J_+~6@E%O-Fs8vdlcrYJ)?|riw*t;2gF(q?GNQE&9?#Ky
zhBc1aVu?#F**y6dYpu}&M36}5=)8&r0xx|}I|;$#9eC++!Xtx6_*9T~g<nzuSPeg<
z&mcer55W;UX^|<DEm>+T9|!;5H14RVQ$j~c(%{LfJ;dxX!PZm;2J$poNwPe{T0^^g
zNN6om704_nQ&FJLI)_l2JTEbg$FvQmsmY6iYF05RG+B@sFLC5xEVm99Yz+kGPGx-T
zgXekS)dOnlFw2~mpSjNccU@poW<0n55|@`VmST^(%`mmWn1;5g3C81diEnC(@m8FU
zX=?^TU=~LxYp|-M91O8mFk7^2j|+yS0@D&i!JZxS+}=$t?H_V5Td?aRbQ)Z@j#$MC
znGIOCt|gTq^<!(1HJx=(Ti+M$ixzixcXyWn#l5%|r?>}q4en0yLUDJBySuv=cYpc5
z-+Rntax<5|k~4Rov-Vz}b-|i=sEFO-=hkCl&sMp;<o90xU1Ngu+0;K#n$#~^fiBzb
zxqnD>^7C5sS@}u+^R>Ibu6*m%!w<YwnxL#GUYYk!q;S?Z)6Ez(D#nRGZBbI-ga5iz
zSzMbt3dAY<t@k~35H{5strzy{cSK3C89w#C35b#wMKZbUH)F%Eap+<kBxl(SS=dmD
zCCFmHd|uy9P7L8s2y%GxdwVoAfzT&xHR*`7gpUF&wUJewyyvRlx7I2U0%ppLFcp0K
z8A=kUvP9bS*ouQG|M;EHv}zpJVnU!@rRpzv_+1HJ8B0OwZ(AQ3{)ubK&aZ;#jID8k
zAl^?<EMv0(?IPmeVb(_U%G+2m+de?VJmQRE?&$t2!KarB%C(4!7ILLEdr7giZnhf|
zto;NHTTYrT+IIpXMC%iLO}b39{H1o+$*MP_R~Nww!#`&TA+fZ}PM(6+N!UMhrmvm1
zkd<<vE}^-Qcdu~dmiZDi;X-m|n2>T?09B81bYW0~mQ!R!lEys^PW@Zo(r2J+pgS6w
z&Ip*D!}^elsD_b@%(K0WNAJ@fflmH}%}$`p_w5L2R)iaA(xbWVE26m`o)mtHF)G7u
zFzPB!MR$8WXj^EoVLZ+PNU8#0qmwJiDovQ2Wr>~MBv%CSrfLF{XIsct^0@WaSAM9&
zruq%1NcDho8N*KSm0+eC`=stNk|C#T#*Wwg5vpJ3&P#tDf1U}ATZW|>cHrnysMSBM
zA$}Z_J}3FyQIbOKprd?DlY)2E&&nT6pHF!otjn0>?Mc9|Cfrq|5+Psw>~?J{!Eqsf
z7M=q<NNkAhb|)6f2HI?iL2bo9nLrHFK<$^505*VBZ*;`B(^)~u<gDU&4ji|#16?ce
z1X%=Tb=6WjM5$Ex23Ydl@^yZR_&{UvQ6_sk1_CLSGGLJQ{>*LKENw}(p8B7nz1e|>
zKgxejxU>_9XzG|Gs9nqFa^@m=Y29WkPyc4KScB{dAzevYTpg5*wkw;p&3Ad-p@>h3
z`7Xh00({~h+I3TD{odVS?{kZBwTzz-)!x|4@+290WmB+yvX7mgCt#q30Z*=y3Kf#f
zr8owj3~`*EZemtkUGgNM_HA=7gG2KXiVZ>-p{8X<P~X9Rc0{=}_foj?G&>-mlr4%6
zX&Ezx*|;u<ZM5^Z;Q|TTbWc9~;48`@-9d36=Y=tZ4tWIH)GijJ0C4%uG1e;-f;9=K
zsTTPKd`kmLIgoEx&^HM<GImbGD(8-@g9}xVA8ljsHya}pm-t+~M!$+WvW`8wk$i0L
z*>1SSUOvUf6*!09a&G~5tv^<}-#EVGF3(1dH^{(fXgImH{Pmo&;5U1{^Q=38ZK|}k
z2uFg2>8iRe=_%*A(Ua`y*@<qe!ne=aJ2f1n&FcWTPw51Ww>Y0LRL0<PZWfTPIZ*8~
z^j_~GYFX(is}p`)Y`s>)`=4$dQj)&VkpsUKqWfR7en^Qtml^FZZho{kG1FdeMy?HT
zXmQZMi(~fhQcf$sc;Z^=$=fiEG29&DC}QE{^2T~%|5o_rK7&tz<BGtFWoQOXH4Ebk
zhb0y!21SXF*kYZJf0+2ADm^s{AV2!!TLS%!TA?$9&fgu1>axb)@%h2hl&ZRtG4$kU
zRh)@&`WY8D)7Sq1+6(&R{fp8v*jBE=a9BDhL0uaJdNBftwV1Th^pO~mS#zD?3MsYP
z@Vu`rlzVKFZQCK-lZuw=i$>6B@J=$reS+I`;NddxB55ZuvCE~s*28S3ROR2inkt}*
zM|0YcW|}n%UxpuciJZiW;JYVqJ7qHb6R|mRY|fmx@Z%rT(}&N->S_T|G93EvGXWG4
z{%{9KGVj=&dYnpb2-%YQkY6L*RLqAGJLdW-n_;@{GJ33b7!i3&a>=t+8_!lYVnvWn
zDNu{lTANJZYSaGFpGL=1+OUwVs!oni1K_r*{0mcuMj;E-1}u>M1QA(Dvdk1h5T0_~
zo5*&iCutBcQw_qTD#q9dx3hMQnvw!Zp?q^D4q=Odq|&_hDihV$GjUr}rlcRx;$FI@
z!0Q9c=RtK==aYj!V9U+$Vum+-ebAenagiZ8bc?4;TKUi6+vxhzDUM8Ain{2gGVOoo
z>?Ekyio}>&6iNsgs?0)$R^Ry#Y_wQ*lw(%->PK)zp3i`LnB`X&1nfh>sF?zeC$rqq
z><uLc4xc&QBi)~tpF(lnmr%gVr(tQ4Ugghu<!g`65C7LS*b>o5hkI?C`x$JH{50$@
z7)IviW~9FQ!j*0(ht@H1k2ujW=ZnA1QsF)*XhzTk?(fA0%WDW_Uh*Tvab{X6;-m~x
z8S4}ZI3uGq(HV=zyd7Ca^`6iY(x#0l63?V$q|`nWw#We8vlocW#RUzZ<l35qL?(6X
zOpV9xGa(0b8#;r>a+|9Gftyx8<?c4zKd=3zXS%Y0uAQ*%1F%gOEIB^_U!@nWmri?5
z?oIJcKzD8k?JP@lwG5|4aMp_ZuaL?o*i$i$;+1^xS@zAHmd#>nfcPV)RX0AohwSD_
zY72ued@Pe>^0`CTwN=6BTY@1~Ivs%3_D9_<LJEuv&Cs1TVZ7>v;eTzA4%FZu-}@p-
zbgb+acZ8M_HiPDuYL6uKK2x7=n$Y+zgIbz7)r_vR8($w^FdAef*4%Y+(@(aG_-!nZ
zj)ba={Uo)AUQb5kUezf9FRF#m$JOi8TeV1E_1nKWB>!?T^TcYu6aJ5O|2x?_$J`(6
z<gDL(w)>7&L^yYDZa*;vONBvghl`cISV8rrJp8^dTs%ObAagxR6nA_i*oFn9A24AA
zU`Fn^y0>hVeD=tSTVj3O*4UBgd#jayA=YS@_n#Jxeu+L9)R^tng$<cuq}1+vj}O?-
zYojMo$A?}1sETdn_iNFf&%3x!`1<F!q8h<S4>EyVnKYGxMorOcIV@OrmZl_KC`Cj!
zLT#ShL0BBvpNNQ1@`y2@%}Q~Z5EhdxafS|%+2J;F)ZZ_7=yWg&&Qdy7JYo3301i6p
zJ_;O{!hDLQMDZ#Wj#CxSdJ39wGOt~BneY7s5XVRR9fhCq{VC5jNtO|Sq5x&aBc>^n
zHI0!?M2@jSj?PqvD*`GMlZXjJ>$!m($wR}uh$Kd;OSd%^8{D!jxbTyQ|E-6O*rn!e
zC^)`)XQ>m{=9H*9F}ZKL!Lu0D?ff~h9Wxw67Fbd<$`#-Ia{BoL(=b>Iq2=Hx^4$=a
zxht)mWaqU;k54)DCv{kG|FgR_+37Ih$j1Jb!Jr~uVllSc9LoXbZ5DbIU5MVvdDope
zLwg1O=smC0d^sw?ZP|PGk(SEw+`-QMSn;2f5Pk7x*P-e|Kw_BtC59E{uF9?n<#WY^
zYDr|cbztm~*7Or?lRUWq51EcuJ$16)qo(Le)moo$%<ybkP{^cIsL%ZwOzlZ`zP$Xf
zejx46D$N7tzLNOGmL$Bbi~7fQc*0d8WX}m*P1BRKJhJ4zmx_+Xe!B7SN_X7caA$L7
zuVy%Y^h!Bbt_ZG|UO}Elt@FSgWlULOMbN1m>xDsy50fpyv};UE`r>+i3~04<wNI7I
zj(1pIv*a4)Agew)MBqXl3Gq7oOETw281TPE{owhW$@|>Or;S4OJ(dvqcrtxkt@gh#
zSgT3`I!HFa>bYpdizvnec=s*=SeQ*Lo4SM!G1y#6gOHO$fGVdz9%!TlSqw?4ghvi&
zmL@k&bh;2@Dr^a%VmKrY@r)j_U~!aSeU-ysZe;2t%rI(H27k!naK#ERC1QW)+&39r
zO9#$s_YdrHSKaN6ilkQM>zyOU#Zw@pmlMHtmKG&5z-PHr=n_+S5P2Q{9J6m^U5xJ*
z``Hbys72bW<$Iw&45Tm9Q&t4o{wpP0S*-@Hu}FYz2b;a>C4%O;{kB~*;e7&=^@CVo
zMbr{e{f3J`^lC&Ksr_+oVe+?#-{0x`PkpXQ&`{{11R@cJDA=q;!WNp`5^3bz=Zb7}
zUDO8E!Eh}-F;y%vV}YDmF66O+#BE^>jr*-EcG$qbR#|UZ)QQ-ansAXQ=-1dB&xR5&
z>jk2)fG^^A-Z9LE8)M|+`<Da0-I?jWj85QXbQe2j^n@^AHYy>QYV^8L?f6lp=&qF4
zYp`YN$!D4=i()s{b3p=n#$a}VmFh7Dd-Qnq8h`VxZ_ru|9Jcgxaa+=uRvCDVSNyO_
zh5mt;!tJqSv~%E3>A5Y~MLL9^`~GvxMR~KA>6@)0Rj|_}c=UoSiOloF{~c>jue-ll
zSipSdpQ?tmLa788J2LFp5NQ4LP-8$des~~po$KFLixl^&NgqM92j3bS9eHw;0&a(+
zqnPl@&Oxc6867st(75-%_7*cQVGxD*J_|l(i{Cv?u=wCn*V~`Bx9(@&x2L#ws^Faf
z(vKa|!zIzjX9oe1|Bf;iwVCC~GAS)gz0xtwXta=(ag*o<>{{G4wlEru+60Pwkn&$s
zlE!07AB^E<1QY^LQO@?L%Am1=0+SUQOc<x|jYmQ)lM$ncXv~dOdL6%b9;(X4{|QJW
zk4h-2<sf$zaM(4=M1?%lpO3D)s~nBz+LDx^Sv~$9U#Yo}pH~7W!#o2h{JuZyc8-_w
z<Vs=iTeIP1fq%JEPITwnPt@m&)A|>Wp&w@FT@DAEhq7z*<6|CV`2YE&E$p?TK3aYX
z$|9@DJK3qtuWE?TV5{0!!X>NSVGx}iNztCqopg^Z{aQx<W5@!?)*cFum&sRU0D-tz
zXRonab0EX9*uP>)6$ylF&j_4~yGvkDXyoP$XcBqJAaWC_jnE-gP(p?WsG-nBWf#y{
zWrl$niX2%Jd_qE}E<wo{mqKy}Sdbx6>U)6xAXI3%pLIguz110se8FC4pX8V6_tuK<
zIEZcBfa=RewxsiIy$Hc9Kk7mquiV$-DW{#+%cI>^b)nC7bH4*n8ow|L@UwG`$xfmK
zved)9<tXrV*euK=w@ueXVeOB2M%%&FRY2b}I>X$0x{lv_SpRUq16ei%$E5u~3U|?o
zTTy?|%w*yE{S4~WmG5Ki@1My5rCq?tl{6Oi_IvXtu+4iW97*l3iHNC?raqr{>cj}j
zAl&IoM2U3*M5fos#F5d?s8&~xgom)04N{DC84hyB6%uUJC|>Fu{FKN$q@ta;dB_(R
zp_#}VMR(_BICNq2ZRXJA_#;(25@ltP3js)SH02MQcBF6A;4{(Zqt9I->0XyN>Hcfi
zqsMjL(I;zWNoP{^Jy@;N26hv4#x2EP94j(QnuuExbMYA4cEtV8Rw(ETT;=;oXw}&d
z9S$Q<mW+mKAE79m9H0Q(a5bYT2a5pTieNTN*+I?+;tcxrUEe$@UsCMPurfm6erH)c
zN8i^g@jG|GWNUw)PM%uwyZbaU$rn{v^ZV9zNkR;M87ifmc#M3rDYr99A`i%8zk}?v
zR%gZ@quE!I$86t1J&zyyk!fA;i0}W+!yi5zTcEY6*IQCl*!&Y;!q5<YDM`LW6)wi|
z%<VkHGcbyFJs;b9(93P$ob_zo{r3PWcN4LB3^Q2c!4{t6*9X+MiBkeQFD;5c=>({l
zWqDu@OatktuxjDfmY0udmvb7q6ftvXTtSZ%2)N!;HF#BF#j}vrdwV;+k`IE0o4WRB
zBMGIi9ppNj5F<m2#cS;$^DwMzB4ifW)gu0pq%TjD(L<EwQ@$~`YfzIpYm#zOjyPqz
zaQ?v(R)M)l@Q7E7#yNfFY^y8AHQYS}W7`dfdBUF@Sp>Rz!L`E_v>|^7DydE1u?<`A
ziM&5g5t-iO$vBH}vw2^UdKt5~T1HyGN<%|}EW~wEvAn`_D%JVGhP?LiLzn0Upv`i-
zXa7>y9oE*z_PecZwn=M+c4nr#dk7bDpPFMILDf#n^{89)B-%lQm%G?4Y`8qG8XOPJ
z9x$<qIfi|?`6`r-I$dcndT5W&08_F$N1YqI6?hGH=K*7$OmWe-jo$wF59YmdenOy%
zK9Q9w<$=**S{M5^!jVaQIKs5dA6TSV#0jD}ax72#Cvt5A6s$j|;B9X_xpN~<$#=~;
z`V*`As-d0mzE|>#$?VKfAdC(IP$g@izfJ<K@S#Jo1TwrpC|7{{SQG<QX;yi3?jjkg
zgXCq2fFc#vSx`i*Yv-a8f=zQ(3PrzMbgQ?0IG{pvxA}5u58=^isjJCR-Fl#FccD@x
z2?bk$wN3Na$KZb^FAo>*=7?Zyz!K>{SLV-Q{4;8=qJM~7Un<=iLXtQ<bKOCO3r>VS
z4S#PZ?>7g?`QSuW9^hlK`8G_(syno`qBoO49Y&0YnvfgKg0il8mDRfDysZR`w|vkz
zxNd>R`jo}s;o-YA7oele++wF;vCMC908ZXewp&L);|Pxe8r3nHT0T)+n$nzOCu5Ca
zOO<5Q#@S&EZ<HXzsckpMcl}mBu{GV%_3PyU^2w=|J7@5&`e>Qo1?P-Nm)>?a6`FL~
zQ`|8Y&6r(qS66N9VN-P}Y)HvE7KPZ&RGSI4wWhwu1b*=Apn9Ka66L8y-n*2dotPKi
zCG@vFBK#6<hngO?42A1@{1vO-#&>?L>I-L=t(t6ay!~|z4K7fYKA-UJ&4d<zi&v71
zqT&YAwZ>iWnCmtMKdV98U`G?%M!|h-Ml&Be<G-b~0RWuRdG@Zu`@|<gfbx59=@YnM
zRf8K^N1Iq8*Z?-7=;h7To%xeFq)`Z6tJE>-+ktGt<|@jc=qdk#rNg58-l_Y<duaW^
zw(By|S8Jsl!6^d|10niTrCqxL_SZ}$eNN$h-)v?p#b0)Xum(1eV*)BGVs@0_RoA(r
z+PatQEnL@a-#8If3AG_JzVEB4SX>(-OlOeIVN}AXc~x+`5~_mR8gYeRC{m*2EUi+K
zj`|{KyuxdlC#8uQeV8@{{;&`x(&R?lVPj*dNyo;3gQ>B<9`a-)pd`faJ`J43Bs%Pl
zR8o4O7weI`GBi5BE_D+_L_16u`ic<oetU@-{O<{L@W6ySH2kWAlTdQvN-zbJUmz5R
z7P*DvX*sD|;o{_-<ZgQOW1U>RCib#gon%q9Iko~#GxCQW`-v}+0Nmu``I7w$ca@Dm
zk4tzXijDB3d+5gM8L=Sg90gMCEb;}!WrIPGj41kyQUx722o2hpU5m?((6e~VoNMix
zgdTg*8xYr$C0QEr5~)k?bMptALN2xFbFT4Mrw3f^>c>;(lOV(Kwh~C-3weh3$lI{9
zjepNdOd`l*OJ1EhLCrKPwW%>ezqW!94x-qN3Nwvb4NhTgHx~STHrVw!cNw=@0k^&w
z@-Zw6J7c}ztO@EN!Ro<fLVE*p;guXxCC|`6a1fVNX3VXNOmR$U)#uJG)h3!Shln8F
zpdT#!?S-w6o193EdoLA-v!Heb*u+<sO?ew6gh<TXRXAPxRtxTWd2c2dnM?LD7-re?
zcjO1B5^m8O%I<K|=UP(ADHp$grGAN|@B48#;ozR81y5f0HEupvsdCO?Zzw2R&^va=
zF(MToJk`pmRdvV6f3(}uh?ZE+K}0e!yl?JzzT^+4YMzJ=cRsH7RllE()bz`CylXqn
zY2vn`2|PvwxRKZh4}Oc&+M@s*0se+w_z_F%Qodqh=6-32z%0y2e&3c@+C4eVu&?=X
z@HBVQ;3nW=f_#1Ix<O%7eXC}6&KUwN__F=#u8b?3pf_Sl)cn+(FHNp_oB=#{G4$}q
z(CZi?Rs6NH>Rx?Y^|EAO;0yWZK`}hHYvX?`Ek7Tt;9U8p+S))(0fEAzK>A8I<z<J=
zryL#AT@&VtCz~C`)ey$D&ihwvHIf8|7jN{^NH~@i$LHgotX$`43Jjb6DsBDzt#@xI
za@S|c3j2cBUI$VCy-Pu=D#m{TFCq0Pmb;G&=cvX5S(pq5B<O3UfTDMd1BKm>#FKjO
z4x@v>2!TPN`9{O`sZY(_*eHRc!hx|~!JBW_rwfd>D3en4KaGM?lfC+TDh-Mfk-kwz
z?F3XXp~<QmK&vcr;>E^WPE9Y$f7^q&UCBQ9$X2HmZ^VjbjnyF{N{r#bw~eLii*GBp
zI)k;|YcsNjQB!S?20w9bXBm0<_{!O7#L%#Gw(n#dKkgzBEyI(Oi1a-5ua-BUIv#{R
zW~A}+L<f-3(@)7=9BS=;(bKN$6S3<tsWO<x+oZ}X%lLo-Xo>-3<woSfS$pjM+ogV@
z<{70a%>`wL5xk>VCs+75Fob*bM_?FS?=zx8sd~rlzl#es4f*n)leX?5Xv`4dF0e&=
zSB^>vJ(eP`ptp;y+d};g#~r?{L-76w#dIK4w#ZQ6LSAbeMt9FjSAKzNjTNd9lgn@#
zXcasJotG}F`l~)`8w>jT7iuUu8CcyU`2jK6sQy3d2+2w^D$P;%PQL)d`<~_8fbw~2
zUHCu>GPoJ}8n|MUsE4X<tEbqxecP9D402n!B9db4(C9?!4Us^No=1W^rhyljvj-Oy
zIc)B&mm@d3d8^G0T%hD8a^U_*QxLmXrYf9MH{T|@Y-7^;$L831ZRly_W@BPb4sppG
zy~ma>uel)`)2POpI-EY>2n7WteZoSx8|m27PB`x0!+TAvQnf0AK*Oo_8JFvjKj1<C
ztrN9s&$553^p@(WhQSA5TlkGjsRSlohB$oT6_Fl2$3@~NZ*_CBmV{bCnj9Fk2sZXR
z1A#K?yyGZG>1M1rL9}BC(Y0neN;*g;sufs^x<PR$7`g-JShC6baN&0t^jgg3N48!@
zIl9$3Srez}KJ_hy1yu-v#V+aF_!dE)_1m-eV?xfKeO#62m>XsS;&I`>DdbJtoM=bU
zUQY%an94IPR=%}hN%0dDhU~H!H(6!a+Op-KSmPkVW@X7#;@K-@cm>|4b{1yJr&)IR
z+zTu@5UKJ81!5>5^b9tc*<{vo5_9Cnj(~{Hn3h<L_KqEUpl+#?CbGRRR?O;J74aP6
z$9|_)NsJg)UYGK=xZehZB8Mo<@NgT#H8oGwHh>AM1MH_#U!2QYE;qfOW8?fS{kH`!
zpH8*9qn>G?m%udg!$DX=LPA@6+q%})l{`t$TqtmW3lolTZvf(ailzJf%^Nre90>ed
zmNVyZladR55YZI4N>iaD(qjnFc!f+Z$4iDmRo*73r}^mHp=s}S;wDN4$1<nS)1;v3
z2}mwQL|eg>NWJOMCa4DF$}r1hIwDkpF(halt04Y7Hc@m!OA0Cg&j@kQaGbc6&rr=-
zD{m3S$RqA}m7|M5YSGUa(Is-Xv-8FN6~DP-_<dOokqEKbh#fp<G=T;>&v37o_r;mJ
zpv$N>?d%tizkt*KZ;1ciWz()iv)2-nm1R{n4R}}t*67w$fNraB#W!y4OUXOP(;B2l
zjzO8)*_All<Y;b8bJv?YH`n|&Laoz{o`U;6#9&OCf>uP!cLA1%B1!1DG)U$73aqLm
z9t1LqHX6(+?lF!!CKR|61?l&1m%p=MZA=N<=tf<u(qte8y|a_iJq?ps`0Y^-=Pcp)
z2xr6&^o_>7F4-R{kw?~5RhL4<Ql97-lJJ<M7ot0Zp`7!FHNLg7M+(YI7rTAYYi6?G
zJT=GWUG?VRs=66RJ%Y9r!z;qTl0egqN0tj2y#m1~F*{*&sI6x&`JA^38~U2ZRaD|1
zM7a<KT~pNvtJy}=B88c{rcx&N>I~V|8asRn6g*@Lw~tI~Hav=sKjnU<Lnp$fwB0ZU
zK_KH#fsdxwG|h&<6oSZm$nJ`GsX<h~&$z38b))T*qr3t;Eup9{R^-pqUqy#ATr6K%
zf0u(XG5t$UA2Iy6att>@w5$kVgjC;7e2?*8=T<X5BDGX%&)mniP_cb+AS;|<O}d`_
zdP~p82MHlz7A=?;)lnHQN3QYHuEe@>a*Mmbwd_}Y?xLse*G0tWxZJ#%NA>LpsT@oL
z7uN;46!HEaaL}!2`*0l48E{s$N)fSAIq|hEvjP&9_yS`Ia?(ojGTOfs>dnf22k#R>
zSnFz10&c?YtCkMhI+^ElL|0dV5UL3Iv1@m(+Wob?EOyQ<=N=o~=GA!041o5<6T73I
zUNQE9ZG&g!3s^5az~pW3SC#)EtsXS8Snc;*1<YMsf-5T(gKn1xJM7sHJBRRs*<o$$
zK9|p>h>2K_4f6<bW`M<#UOB+{p)`y?1uQv~v3q?O!i>yc@(YbSlwKQ;u8Un%rB=m~
z`P7EV;3>$|<Xa_fmI`gek33@^A#%*0&}vwkjAJ4?jE?k~zN7$T(Zd-+zFZa{r@KZL
z(s>)!B$sau+ULokOU3#|gx0fz{q9<!0hICb=**_smMio6TkVt2bGXAoK+GLe-e#in
zN&a*#vsRO{d7Qm-#qV(cjM_(v(~N!yt(YA3;)ld#!&k#_OhJS?s&FHoy7!oDu~i=m
zJ&y*JRt-6|d3#Vj3iJK&HjQ3Hk?eDI&J-dfPbslyKzVeyGn>2y_8O~m5-E?Mpz_~F
zl5SVU`J*-)dgikFAL7z+CC?!f<hK+VQ<>D+LRzhHpyH@{emqaTVG^i!;vis?&~u2~
zB*f+4GEwjRy+2z5zC4%KI80Nb1L6-dg{G!kzfrXpE;fPZ>u)=xvA{8Pp0Qs}OOzNN
zfD}ZNu`N#T1fgpCx^oTZyill0yHtj}0um>&rMtEr+JtA;y3!kb<%6n?v5N@0sIZYz
z#wQ5vO13R~fBQ{aZe3L%ILL<+_91T23Lx5#YlOF4^op>+j2Hf4d=$he6TA&V$#3!b
zv<yR3cvDH+#Kf2a2o(rL^9j(Gg)G|HK)rl~vu2c-fZ-%@1S&>5z8P86bsjSE#V2=z
zJ0kwXzJW6jdmVE3A<{FnNF{`x$Mcf->C|SU|L4oNv=AJvGZV;R%TdBdmIj{H49@Z(
zyBW}mZ_sV2F1h)i?F+R%?0a=Mp+9<|CsxR5Wmf_V$Rlzk)+*g!#3SP4&+iF|E5#Gn
z%j3xXPtJF@ihEVdek*5~(<{RTx@@#A#sFu|MzEy<(J(Fxv?>-eOYE<}m^Lt8(+;d%
z22qURC&TE{XD`x~CCC*f!5SrtMH09XcNUBZCVX92HIiAonb?bsYFiWpQ-sm>F`|Sd
zWX5`&dm04#zq_z>`h?;-Xc7@Iw|aXIe-*hS%UV;<ZXU1N++J)vqoemc&i|U8y&1^S
zhY#Ys{&H;{#IO`)L9We$3#P|v<gR1?o1`Xe-TQbI@-QN*?R3f=vFi76l1Wv*HdJXw
z-g6RPTV%owj}9YDJ_r$4jUNeMkZxHaFI3CFO^lE6ynjAgCP-U<o&JiZgVIMn_6%uV
zAXbhQNJatgszcnl^KlKwqTdnWqH3)h^;cclH+U?qx|9Brn&Qj#m%2r@1?=FRU)8Ji
zOlNV@HcX3c8jN8PzsA$Jl|^5P<gkR^wk3q#*XtVw*GHDTF=!zwnw8a5k1*sc+q2>%
zFj(5;p2)5MXh$-pi1W1!#~3J!nA3AlQnuX)fiz7f=2ydLGp%Cby1zH07{@qngf3&M
zL%%y^YQ#UWGww`I;s*^M%KLQLUx`1M9ZY5|R8Pg1AWDae@ZpBaEwQQqBD$Qs{?(!_
zlE`Q0-}g>=oUg6vs%{XHO5kdC>!pckW9JU789LDZhZ16I4wn%gyZsMQuZm@+6V}+|
zP_1jjHu&Jfc|<g+tG^q0f9(F4SjKUNu(bwiUuc=~ipT{_laBhTK7R9WY{I*}I|4ku
z`8`>mQy~)>1;7aO3yBEgXpsPhra%Ol8;GvIbtV5DSN4&iC@mYa58kcqRQ8!Mt3hYa
zLLaG&<|hx(hbJPy)!rk}8_g)IRfn}}R-QCtfT3s4XmrIURM5J7wr~o!5E+vz!sWn=
z)u5?vN6DQX+OAs&k*8_;x{hCl<E?(@456bfM_sSOqmVh45fpuV(w+LdnLOe>LC$<+
zaDgdNY(#Sk3=f8_J;J42k2%EbD@1I$(fCM%^Mh>Jw&cZsTX~v54OEN>9ci`-&G^ZC
zPcU+EstENJ%M4tbQt6MyN4fjoGHz!`pPVY<hYR+NFC)hL5x>ckAqJgJOm2Qjv3~0Q
z*!pzvd8yCaxqcJ{M<r}bbbn}lCI-Lb?=(s~e})sIJJ)KnWlb!vuNUcR%22r()N3<Y
ztu-k_ZxSz6va_ggr*^+SenKo19Q=EBQJ}$ee|)&uKLg80U&tbs?Hd=RP;w%jA>3@r
z2*(OIzBd-ulI%$Vj<Gc1Yy)^smOROqRN6EUrO1Y%_tn%G22>cF;ajg{h{zk>LlrIY
zr7}2kV)2pW!ulYj@O??lwYX#{uJP~@1=WJE&X~ul6d4jI7&;Mm)X-NH7pkngp8M@|
zY$m*p;9Es`@>j#rh9*buEZ(Jd(vPi-Dk~d0GzCQ~-Z`mnYvfLSYw>$e3tmthPydIe
zl7}(Ra5|mm!DZ1mc5%Th#`5;|c0!Yc%WNCO3tl8R5}qWesFw|%5Rbs1jOBA}nCuDO
zi*S3qwTt^=ywq@`8P}y$k-r&_lfGE2k4+mPp`9~AoK~gBIuL49pc<k<zU*{Wlo%p8
zhepgzVV6Ex*Uz15mLX*$rFz&9L%sw~HUxjkHH&rHOxjQUkXEPkIwd0-t_u+;g#5bJ
zQ6|&MOYSnh&7W;mVz`IQI5!@e^W=C8BVvQ6e5WFwA@@G_2NEK;qltEv&pksto{ptr
z!$k4gge_-P-ZOoWxIQn&Frbyh!p^hw=V(A05^pKeH%wc&f0v{+6$~AZ>;8VDKK_HY
zT>G<={4|!A7Kzqj6C_9Vi7`JSr(qD}*A^Qp#o-jo!m$dS8mJSaz_deL*oYs21a@`H
zqR?w~zf-jxD5I-5ix#=+VFbjRVguxk`?4<b(4JN?jgAIzjf$9DiZI5H5TiYB5!UeJ
z4YbCOI1g{}>sN%Dk|4h?VAvp&OIB8;PruiK6yc`R1L8C5Et{8D*Y_1UCS<ewv%bNR
zM<Tw8sJ?(Z<kJF2yAxj{r7_eXj%0_4ixMK@wb{C)k9U8CP$dQF%W_bUKwF`&)f=3_
z7vvRcMlp@OW+e;c+p*vOK{L{+uvQs}(HSYLC{gS~$j7t>No?PHijfHUzRJP{*38&6
z%D0e9*6zBw7JyKUB7(81I~5`|jPCJJYCi~?n?rVhPHDNWU$-vKhGgwOX<m;!jO~|n
zRa2=8XCg+-06K%#OHFo^QUE#KIg?c-yp=sQ@gw~kyN9y~cU{|xZdtxqOVU?6?Tu%R
z^Z9Bg$(P5M%wq2fK_lPY`EPQ7V;5-|=0%cL+>;zBrbBZANDv|*!;lj#h~hL-z6do$
z>38;mCw(y$T}rGQ8JzwJn%Z=e)Af%EkiPs8xTzayS~??uBDe1us77!kLDT`4KV@yu
zHB$F)`O*AeyUM(5KE5oeXb3YImrPk}-PTpg8O|PcoC3ASIbBpwd`Z<+sA*kepaSJ-
z#w;4-*g??23`xdqG!HMHLi3Ed>>1P7udi10E1CAj**y)&k3v4U=d^y&q%U*+fin+0
zJ&ohloxNju&z>y)&z#v+xH6_!a*{?(3aY6v3`@g8N6)xMyRGk8*{@eat!;PfX>lv-
zo04$ys!BSUVfxl!6TNvty~nu#HH3U-6lF&-OznMc@H5h;cGIzT2paMbE{MPWnY`J4
z!;C8vO$crmp_&`n&H#M4Tg>u1=z3s4`brt1^0u%128}zJx0maEd3gEO+5M=*w5z-;
zj6)8ctsIm)xTk;T$3>&mH7E{GYCsj=LPb6u_>CnbrQu60Hba87E5+&TftniKfEc4(
z*|6j<vhvo)Uy%_>EuOjTbYEv=f6-|L_4&$c)Tm*kyjxyDhoIWv%8F5m8{u#L6_oIp
ziM(Hk%9=}9@;=LVRhJ`=N=emsbozOrsLXwtzqYuvb>b@@aSsxx{e34(YSY;X)4^Wg
z6b^3Kc0Y@}<fN{FC`<)%>N-0oPUcGS<A!GX{aT?RbrEpJ_y_T#m`VvlPQs}$%716~
zX@@&ifyHpWt$UF<_Kgz-_W<C_!g;Y%>F3jG)mfW}IWOIOB-+qxjr*j80&<}CFBo*R
zC@v~=Bie$8p)EcBaAlvxCtb!1W8Q1V6becdWLf$V#aL7b4GnEe_{|^LmY>^S0u(&x
z45TS%kiwBYHQDgeMe<xtCKKA5S8LiTxBf3Fc0&c|Om9t$S6x<VvI1+^G3#vQJb4%d
z%EC`2Z$t(=sH?!eE%}=|u|L^^1CjW}KTK_c>M+epei2PKg@JR#)0IX*6eL1+pebji
zBg$A<{NPG>6BP=;;Gl2Z)$|>a)}<dQ>)Tv+&%1z%1%EN4&iLW5I7-v*!8`xwuDllC
zZGO1tuDsFFOA3_8hx8}c%3uJi??b(Q$D?@H+dXSxee}cxGEG*NzTb}2cN%dspv|SN
z^u`ZpQmfU}1CySae*t>!yfusQHZkmB&pwmRt<kTFas@AU<wGpcX03<d3?DN-CeNJ#
zT;OqjI2cU&(*N~=>VnOy$Cw-)BihAUrgs*MMlaB|5W|p?*~QAgO)$k>i!rD@&`Hq?
zQur3aq7KO{jy`Nomc};h3V<@^nofqAaAZ4%KS)TLC((=aAW9Pgog<m)H#RHI>7gTo
z_4NU4AeT}z<60pTts&7r_LHz!LRvuYWA9HO*hoDz5cs;^E`t1;xkU@<!64U|Jj5p&
zq|F!7xLA(*!;XVPxcW8gqW2M{VjW!4+A+6ATi<)1Z++e{Nss<zaN641{%1e}R~WGN
z^UIDfv!oX$*a=x~z{FH%3Ct6+$>jbcGy{V47|@0;k-Cf_L|ysI4(29LV#Z?1M22c-
zhZy{Le|LfDx5X&<?MkTaH8!&ov|=lsffE3b%?7QD#zp!V{fP7m5y|D5oYM)BZM~_8
zk@2}}yts)_%fX70Kl%k)r0u6bC&!D1Rb5jP%h3Xd6KBgDSXCCYr;3FtRW3i*QH*g!
znV^0B>6$9!O0A(+ynzwV0q@rx+EQPj6o^C;Ty!6$b8uaiugk!rKiQh_gF{W|d0vC$
z^OvzoZX|YBEmJ0JMWnhZ-01+415%ddB!`GM?40CTj1O1DCQve`2d&^-1^s6^ZTB#*
z$g4~Aq@nTWcEBqMFeGLMXYQ5M;wb!sEY0tb)R`C7npjQdW+q_OOXL|WYZ=))zT`hT
zxqre+v-)O!-mt!fXZt>pk8&OXpVqoSMxV2|Wq+c;EW;DdQlK9fnY605a-)YB4Yh+A
z>8%o2!j&YO;vSrr!S+7Y(D;!<$?EJ{Ql_{jF<}5*U>Bu!wSGHQm(E(Im|e45^jk1e
z5GfC-Y7QE5a?Q|jd6^I9Oe^L)01e3c!XuOB2}4m8js8w4htQw^&a}&|)`jMvr{j>i
z!eIQx#N8*Zq`IG=KvEDKI%E<fpXtdm!fwWvT+2*YFsWsu<%|UA=dpjRugA%EOO%n#
zbhRJQ!vv7^u436RkhO1xepeWLuV*-3^mIRyGHjLxtXj&I?Vqd~AZZm{OkW>VfQDt@
zppPi>SPd4asOZBQZPwSK|EFJCPG;|d(>=(7og>sTS^aiE;})EgIikVwfo(e5K?9z_
zs03%kieSVc>>iV+;lq)*v7KDwUIoM*77|*8Zbq5z9`^(B1_ZG{?AbM3zwE-rh??YG
z6MODK0l4Q`@DDs?>@|ehF|Y?$(C#Ku8xYaTSdLvh;*Jh#N1L_0MbD2`Q)WQ18bNzE
zg(CBrYGm}0T=Kb6r7~d5ETNtzNeVoP8mcH$C)c{a;pv+}qt^rsivSyDzdKyNc7bI~
zI>;1sF7avDWuu@<;mx${37Dh7L@7i&M(xF~OzSvV&jI>kZa~R3CU`qg1}i;?yp*&p
z@9j=Adp{wNX-ZXh2qgLCOvqbWzU!`+QON%+AWZCbmYT)qpSR~e)YfFBR`pGOoNvgC
zS@><7PrRkSeU|z?cgaA6k5E6e@H{swI7{t%Z)*hgvtJn#2Ybr@0{*aibNfa<ZRD~&
z{L%js^`>Wtq|bC@4E<;_+oF!Ypo_cEo>@_w`rU&Pk`dDt7ocAZ9gdv4N?-wk;eABm
z%VeG{-+5}1A@^3DQ~r3HDy$EZCiNvPbRPGq^;v3qrSh{3bHWM!n;yVP?9bAM!bf@~
z<D(*+R<#t4;AGAs=%>?rBa|RfD#g{g2+1KnUqp%HmY|G_#t?6wST~()uHVRzGA*`O
zt2pM<^vhZo2jpoR1J|>{S-P^iZpKFEy-%jp!B7O*Pq4e-tyIO*l7pSN*rr@U%&?^R
zcUG~N#yYs`y-jodK8`xqFs=$7LIfct(Kgso1iR}5a<9jWkX4eWG2l9a+XH^4F<rg3
zjpyjG3G5Q<7QC*^Uh6(z6B!&5d0?+H>Ykj)YQ~Ej;<H@Y*cf>J-I}`S2917#wAub!
z2mfyypA{9F_K`ZF+dB};S{UmO*OnqnrSPKah1cmu?czD%`FDse!6PmiM83mYYVY=N
z{WhPNi5PahEXo`Ggftk>gd}qTirAjHWt>3W!iW&lBL~_uEdR72qWI}Il##JIq-p^p
zUFoabXrz2-y5Z62Kzy>|B5U39=b$$0*{jF_PiZ~oS0YNu?AaxOCYSbs3RIEK!S-!B
zcHe)hC|xg1v))Y1-e1E^=<p){)>ti*c&)A`jAVQ9@2$p+*$T0le_nm5gqARvhVD7^
zwOr=&MQG*EzsA|2t{Yd076USI^FG>qOyC-Arx~{NR(gUauyxq7;lHjI%9D8RvsZQA
z2(fnGQCxa|(s=5qZNAbib=`0f4t$woA%}LiL~gH-dR5M`v^f%TUgPX=6kYh!zTVDX
z>=su2epC8<)$E_F@G}HXL7Fmo5e!w2VTXo(Ag-P8+k^XS+q6B)s0F+p(6Y+Jq261X
zvxe*}JB*7*3a7NV<zFOmdb!j27&UtS^LEhvlD4&QE)u!riQX}UfZOUSJ6VL;787!q
z(k>o1THGs9lp2LZ|47k>9`>7vTJq>FX}nd!WOhJ4Aek&#l5r(#%_Gxua=^N|#2TC^
zIY6<B<!LNbP9KF(TnJV@oxriQ5#)wqnwiSN{?tH8I2mfMt{me9XB}ipR|)jHx8_FS
zZ&RP>`7VV1ImCpT&=llw@WcoeQEkAP=#F0`e~!bLd?Fq!#Y$(eMc3C@<^C(3xLf6|
z@9VAgNa0Y~F5^e_AL~|`sc1}rliU<2ij{mNvUPn!!8|zMY>s!&e3F-ZD?SbQ9>n+_
z*47Hguuf>{;gz<9zH9XPbQy~nso>zV#d1)|(O$d8{iQat%G-}NG?Hhfa6?x;kN4ba
zGaWE$uDOj&!-tk(Ri&$ozn`KX#UQ5=;g`CbJB#(xT~6;W{b$(CcshF(jkcRk_K-F3
zAD8%J_;RPLFmnYljjf4>i=W2^&jMSJQX(Um9U-)29DyT$x!Vo1UB6Bke=l@*Jha_0
zbPyQ<a0N$75q-u9hy8W&s+HE`d8PCbcpUB*bm`ZBJz(^KwX}gK!+yAG-?w>z4qiAM
ztU~X|!&r~PPk2>!@P)_^xM4L3ZC9H&Hxv9Wl<6BCvch>k9M>P0DTN&GFg)yHfro?W
zaYlS&Kh%S3LuP|Kh)9@!>n+`CY<4#4M}yB4<`d4(i^K6}<RQlK<6*od>ai9sv9Ymy
z9uMp7HN{|*gK+e<j3v9>-%K&nzq&On2&oUD23~%aAJmv~rC=5oWdKsG_N?UB$XNMO
zjRvT>*+Mg0|KIZ#YONQ(n1XFF9)UbgFF2sE@}GPHb@2Qlb?KN_WIx#G(vu|`qN?-1
z-(?|dqd{auHCMS|!jWM!oTx|*fda4}@KjqF6<`XCYCUSZMscfe)0H(G-wbTkVI9Z=
z`UWT_gKtkDNd;zAF=Ua^naT6xBJc2;-<hKlyL7<NG!ND;4VTx#IFYEvus>SWSNloI
zPC%gK98ZH?a3gbN892;mWQzN4x6jU|mIuW<WV(?>iXlK7+p=2csRCE!#L`Kx$Kws$
z>k6cUtHK%trZyhD#294Tt#mey{bHgUL)T$Z)vc?bAZ5OB_?j#gtFl4$?`NXldp67p
zs<K-jV(-w@il?LpFF9Zzzf`<@(w8D6;EDq0MlSyrDaHDGF{QM2)N-nu>eLJYVK~zw
zPD+mbM!VX)-Nq+c+}rfy?ZAeAP`6tES)ccq)jV4+q;4Ungemo6uNW><^_H=XH{sC_
zM=v=%ELCI@xD@I+d`<6;oa^2NO7L7Qil=i4@NKOtO`Lz<|1I=0p}OM=Ym$LSW9(!k
zzS`%Sas6`|>g_D=8ULc~($H^<)Di4Qt@XHtRsY&6-@%J#1AJc0<^Wg9+qa4C2O~%8
zOM#@0ZIRMt-xCB<zdQ7jP-7M>Ws!F*RR3+K!v*x<4l{b=h+h{Gc@9mPFU&}R#v*wx
zT!OpTs6vxttOmV!V>>%N4)-NF{SJLs`B_vT4w}g*?dj)fT-VwI2CcVxvUqX<Dc21q
zbRT{7@Q`Wr#rU?6S@`(gQcZdp>S2!1?-YueQdnbn(l$hjrtCZjq|#MHbo5ui%thl0
zY8UoOcv-+U#)@tUg?c1O=NUyCXUF0Y2F6f{!&njG-xug|KKl(Ud&Z|lm`HPqxW~CF
z@|7>1v-7#~p_#w3=HWIXjF)mT?%o9QWR@Lx|ED`KQhVMnXtlbX;oLr>+v)9fT7r4&
z6+#;4iz@tamnHH`rS*NMhEyLbwbr+$wJ+`Vp|=8kU1uK6=b(9bIN^y#e}D2{Wz4sp
z6P0?e;AF&9jXrJUVV8|UWAfpT+lc+M)ttvp$walG$rEf|t%qTQsZImZtb@k4)ot=I
zXTUZd#TvgWU9lJXX{>`~^hsx+peN>~FVQ#<(aBQ4O^vPNlESj<{hjD-_6Yww4l$Rf
zCI^w;>8hGV&0lp55*tD@nibAm7fTu$>TB|Z96}Hh<lMMb6`tQihSt+KV;E<gKPl<+
zu~e?tA?WSh|6$9yUKcVdAzf{lxPw@Xb07EPe#-H59!2y6vm3Qz+sYwLe(OnKi<bmV
zQfya;g>ZEsXzOtX#6@^@e;N@?xTN!|#oZ?El7R7Ii$!NYXo$os1V_2kmTz^#)Y}GR
z!)jnbM=d>hTlsH-jU;Wqb+@cEZwD$)bkK@x`?QhJXK$qXQ`2Ep_KgP#ch^i?YZ-qD
z+l{eXRc3Hm93#T83cU=ZgdP%rR5djzCrO{gM}+?D!r`=1gd~IiLcuh~Ehp94szh7;
zlgRGOS}(4)mYV7N{H$6A6|}S|QYH7kQfZHR;CJ|m)5NCeHQY0ft=pz{D&KY=_=-fI
zQaV)CtqSjn8h6q!K9W$fKJ8WF3~{wviQ7(aA@j~LBx!zjXl%|<Cm}j>A95ROR59e&
z50y@Kn@r$^(rnQI9_ohcL1Y!0iUH%?CSq%NP)O`-ZMXHu#GSvu2kx)d%vOX{)gY8z
zMI6~I8Fu*k(B4h!Y5QLRuODa>f+TOWXD$Gbn<qrrVB#@*Fc@^gPK2B>k%lGJW0g)l
zP&oXF*oFCY-r=BS&rY0o&Sx_8lMf*VwaWkB{R4f)YO~C6UBwi$fsF=zs7q`Co#P}p
zlf`<i2@JRtKb;&Xv|BrNzs3O%pR^UOD54yC+xkDTll)}n6@7(~3^;pDv?ZV`rc}p&
zo}Olc4qY5yO*Hb~CAj=U>@7d`&#~v^$Dp+6)t6P>uIpLqkE0`0VC~AG^v6}+&g18u
z^0oi7=qI+{2V%F(xv&Zc`dT)D@GzC=Kv|3nmbjyz<uK{HhhOwpQvU$lJg>}+BO^(}
z_Wps#yuQ7wo&t2A$JZ&g4LacIDoH&fICBuDji^p*0zG3;p6*-1INvBD%44j`QdKSY
zB2!UOd<`{>_u`g?_gy8rxsKy!=TSW13QS6C0c-VpU})$BefJw!<Eq^3cr@$SM7ZF#
zhuZ2zZ|1(1JVtB(J`Y17M@togQkEpcnm|ql|J$DXyd7R87Fos$U8Rok)JIg7CPmCf
zR@Kbc*$(VYgIbuKZisEVo==I~&1LCo!EvHcl#9GY*+Z(d14l3v5IC<!C;v9zuL_kE
z3j;epX4LN2NOe~JNV2qJ^EV#^&s*F>N|p*!?0#I4OMV5pf_|6A2?>x=jEX<LsN#rw
z&0zkXBaE-Iy7mnk0&^PwU+>NTv<wWa$SBoeN=-H&@evt%)&34gkB<^ALtp#wxdpD(
zgBns0M4y)~lk15J01O*yYbN9##)v1f-zslmT~GT<Xg)QpYS6M*8g#z0XCbw2KXo}q
z^e4NRNFVquORRr>V@MpYFj_`SWW%>$_g?I(JZGcQfJ~+;Ssm5M1@aDRajIAI?)`af
z>%S;IN0y3d+x}flU-`hnLb_A2x6>zmrwlt^N8s)4vE14?aj193m4&=IChuf?>RXXR
z59=V@0}a4HBSpI`TOzF0S4&jYOiQK`2M~P+yls6eqVrDv4%J;A;d-s%z$IhD%@)Qr
zR(_aMsJWhZ)-i<L_&JK-?gjP_UJ&igMAYy2YtI(_8*a-P3h^0VJ%zpWd7uyOj>Q+^
z+kfUO{FF96vY3=hq^yiiQqT7gF#f!X4i**u+9~bR&^j7^*)?&VB~L_5e4}@i?rksz
z{_IWa-@8$LadYmm<#@T2hZiO$7u;x;G_h!>N!Yw>kM9%s<qKq}tfcsN?e3nJRgR#e
zk9XhkS;Vfl7nd+)Z=$y=VWV{`(;<nQia|b7HjXea?~F~I$pnW)`-U?7DY&6&Z$BUR
zUi(LwwNlDVOQ_R5B*EkQB!OX$CSKm**C`$6+&~CMc!(au{N|5~90L)cf84u~pM@v=
z!G4i+y~fbjp`ASx$Lok)jwNb{)=?UbslLw-uP)hfX%QNtR@@3yBq(iEN(fS<t`3>C
zekd$MNX@^I+W@D_Kg0VX8t}kK?^!1H>oF;rsGCd5$geoFXrg>|Xw}kkRBUyD>##4T
zT2yYi;um|?f+6!*oZ$myqirtGS+nEJrh$k1lF|#2l<=FLC9IU9q=hFj)BWN8);di@
z6IzjfrKDA+SSLbuAEJKpI5Lzam6_B3s$g;Wwc5w|+>kX+67jdp&4I|}d=0nS!T;tC
zzH^gv)U(Ff(@|&=ME&mE#E=tBSwg!Sbt%!9>Gjd75GGWOL)&^B`F>kX6WX&UYeIQ_
z+EuaKXHnTBxbxJaoN3`2+g8z_W!04WyEzjs>|0@EeXLM(E!T7_uRZfv)LA!^wHiww
ztbTg3Or57bNh8ltc)|Mg{4ZaxY6GF1JS?d!gLQz3Y3(6!K<BwxYD;Icm>D+7qPCj@
zQZ|YnWl5jFB}2S)E+bcP!F@s(d_g@<aLfOb;C8^~|G&}()+6+DPs7oU-P;@1&b3lg
z<`<|b?p}SHnwrYZOLVNqt7&7NcW2UBNx*Lz+=rahZgTX+$!e!RUp?r%)c(6}OXMn3
z|L-CbDs%q7dvX&Hil^W=`r^X?n*aT~i&&Dte|O6Gf63<jcW0jHiIDWOTgxT|k?{-o
OPgY7vvI<}v^nU=W2|v^T

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_transforms_illustrations_thumb.png b/0.16/_images/sphx_glr_plot_transforms_illustrations_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..d6d933b2a6959fa6b9606d1f2badbdb658583f94
GIT binary patch
literal 205458
zcmV);K!(4GP)<h;3K|Lk000e1NJLTq00EEy009^X1^@s6a$KN!00961Nkl<Zc%1D0
z=eMQFb>0d7BH}g2-*dfB?<!4sgDMnIC;|o01EMIYTO&2ql&I1FFrR0wUh{2Q-K#a~
zSuN8;QW8m!pa>8EVat1;-dyj^@#nlQBIZN9_h!M7RSEFJXc%i{Ro1P{d(VjzZ|uFF
z=h@E|)>=NYe&j#?ZvEZ*yY+YL@7CX~A6w$!H{Or`pYMPF`~20nzl{(QV+}$`JWnEo
zL<sTYfBYxDtNqCD@BZdz{XKj5|J?T$ANe|9?I((8?nep8gW}!4YcX15Oa|6{T`3<S
z<>Aly=qoz-Z<GJ~cYf!0ICbjO$9^4u`lo-&>#x6#5E5(M4^*B+O84uNEb*iN_;`Pw
zkM8Wp&kBD0C;Z5+SkOAdT8*_JghF}V;s5)Qzw?p(H%1c#A^-AU{!6B&raso|$g+(8
z_#gix*REZsPzcE~jkOjb1OiZ=LL3~nzu`>&toL<zR(@jHkNtiBx}Rl#^dI@xEJkNo
zYmw4(=kLeQ-beQH;B}^HN_}{kfAz2a6+=TqAM17WdOiN_zx}stY;2&Ex~KQ@-_pvm
zjH5@7@_WDcdxT;5vHs4^&JO?P-~1c8-R>uTZvR_cS(b6;%o%>~_kQp2Yd;Y{-hKBS
z-hAzCZr@yCZGD3$KXr-wA392=1(V}PkkUt6i;(V7kwPMsL|6kl!DN~H#=au(i%6*u
zLg0BGN-1}tVlWtVW&nXTnqIeq(J8f|I$^Pb9~1zIP7}1&7-O*3{gM)d1Oymuz*>ax
z2!kR);Da?7qcPf$nT&q7!v}ADmCfZ3=yzeK1@}Glc`iQiI9QD}7NZSXXC!IDAQ}+I
z32~Zm`PxlB|M|~<q5$&RtFO}B-R0GnKjgh@mpMLrgik$nno70A^wcqw=VOd;dniB%
ziINg2EJCOGC1a3E5d=j%<slG<K?npw2rLGjWrzQrrW$}^se~7VNFl*!vMkH@&ti<m
zT7ys^^T1^Dt4axvzz^_!51nbUEJf=KqczR_4c>qA4{7Xdu-^bxp5@b@`!dyP)qPLx
z_B>4!`cX_2$7ESb6i3`xSmw8X`?o)K01-m)$AA3C>}+lG=tGb32Y>KY+Jl(c(Me|J
zhq!qEeT<Eby1Q3Mq*6#Bkn&(3H6}|jS|cO`eu(ddE&vJP5+ej*Aw(%fnr2vIuvizU
zP)N_m3p|t%U@Y3WbA|wG4FJ{{jCE(&{cMHjDLkdn+K|LCSrTK6MrRqD>({w+^)K1o
z9Z(*<kNY3_9G>sH&s!U`O-SOHC`w2(jn?itdHcN&NUY$W{j-1eu>(k!W&Hbp|L=MK
z{r5lix&3c_ed<%6;&*@dcR%)@?{>TV>wo>P+1c6oTYOdjzqP*fr7!*U0YplVy)E$N
z2<75{T6uyK$4}Dh4!C&VMT+GLu{H?dGD56CDuu#e(g9j0D4__#0%1@%d}w_?%mawQ
zlA*Px)!1i$Z-ZpeX0ql}DVHggN_2V=A_$q9JVsckVT>WQDZ&_p5J+K>Qld?Q&N8I*
zkkX@2C=!N6um&Mq5}9es`pQk3OPW#)ouWV*OCS?2UN}b~2+^5F8+RF}X~H1tGZ;kl
z1`)ga?N6M+N#&75MMg*F7%HrD_S8w9dg>7}oiI0jhDvpqR2zh}d6gniN+GeJBXk-g
z1O!2WFf1aaL?8%)5Kk$LwOBO9Y6gRdLBGSG*CriwD1?&gPz6mur8+^SI)>4PG)d7m
zh5Yi95~LJplV7w#;mLqfsf-^47@YwYYYa)Mx%R;i*k4GgOF>0h_6L1tCPp}W;eJe-
zqO~T|nk-EjL<0tcK7%Nt+wHU0?0ljOPD+XI`<y?2hNqu;l5c<W6~tbHM<2b&LysL{
zVtj^~+2dF%uzBDBP)fM~VpB}&E^a9Wg|I{r6c7TG^6-?xLjG?<lBNg*##oX#LQ0QP
zrGj55;wulUHEEKdt=kV{+}=o;CsH<#EkY9b1&W0NSWA+`=rkqM8U0>|w_d)9X=rLb
zw2~G_XD2v%;tV#+Fgin<j3kcf_WShv1L7<p%QRa%yR7f*f8sq1g5XEKc7Pxpz}D_r
z5_$R{u;ORcxj%s`e4_QE$izXy^4I_QkN(uVFT>`jh(HMWkw5U`h{N4wj5+*$7>1vC
z4?k|T^8YXE$IsMH1Q2U2YpX4c(e(OV-hT5A#!3d=0kh*r7^)WlgE0bY^23m2SS-RC
zJf-kFAJ6mMNf02FB-06*1}PLtI^f!eZ?L|6nZVbSLy!5RM;M<NB^pF@n#*jg?Qq3D
z&*bbW=8j&ZHZtK3h}IY^QW|&gKLQJ7Izwg&5`k3ikXZrU&K6l_sEr;Yr3;(8Y^*M`
zzrWAi%p6*1q?A}G5mF(80zMKi)rXbVCs|3H@||ye2o{nw<I0DtjE#=~n($M@jE)Qe
z8e;_*;~p4o01HMVgheS2rM&zzF$n4QP-_W7U`@*Q<}#~G*BJD62tAmaoo2W;K^$$d
zvb;j09Z?^dV|wl!L&K8@PlB-sELedxpmmCn?jkW(lV(X?6}bQ-JOxJ6+Fv9~Vw7J(
z%7}iq!<~hjoVoA-2#LkGYExM}Kft$;S%K`wdAaz>mXwmKSC;Y3RRjTH7%)^TQ?FMT
z47x}x!t=<CK?nm@y2~A~Mq_k}wFad^gp_ELxeHS&tPn0RW-$gsx6`53*d~tJ2pv%_
z1w`dCgBZ%SDTYVp0706i?)zX}uty+_CCy?$B7{O%@N$SvDuMDsr1scvY_hv`6`dpq
z1D(buH*Z{K{=_Mir!e{TCw)O!a2KT+fYAxqd%L(lx=x-r$))p`c<JRgXzfRgjn$c$
zu5s?%NeYDm7IFA2Ng)wZqJ#mh-E$?uS0T#t+_QqK+I`>0^SvD6bnvFlt1=-Kp6?^2
zJLksQJb)N?)U9=Bfwd0FkP6S2D520=lO{1)n!3*_iCJHIi>-|-WQ=qA{HH0F${3wF
zw81#^AWl+r=Bmj-6!G@^muWPcKllFq-TLY4Cjtn7^gKKhfUtD?UEY3e87VD`i>sVG
zHA5&M6#`8j;Z24`qdbN02l#&AGHR(%%125`W(-=VL~);+*FWH^fBc`A7#2Km=`_`9
z4JjnOevd(~M-oQ}kuvD)aP|E~c6SyyfB&;gOq~EJF;;^BPx*K8b^~deV6AZhO!!D3
zNfWx=edbPliixRX+`RHCckX<XPPaq5-9{eq5IO^E-RqYEBR#YM>G}6``B`hpGELwW
zuozn17O%c|hcr!@7#ro{#p6g1QY|nR2ZLAxRwE?@L5S}Mc%FZ_he8OnF_;{3TVGq`
z>tFjr3aZ1Y6LS=bRg?$96PQ7Y6dCED!ObhTSzdUTljk1e_^C$;f-=Trm^|3~s*v*u
z0&5}5TtHFEL*yxG9QR1l0jDngD!w1^#!KJBuCKAXvrVQofs%m27z<d8DBuf4nwfkL
zKlueH1YEzeN@wdP%|=3R(B-vPZqpxR_)2o}$SJCIpCkp4hgq$}1z_Vzr1S`a0>1Ad
zgg{9N!eVp^LSRhB%Hj=H7CuC$O^Si!*zw~`PL0rQwb@!(pxJtj>d-W^N6#}dF^A^|
zkY}|b&ybYN`5>zh60Fq-sZh$nx5D>GqJ4B0Qy!WiO8cO@EHB<BNm8oi3JwNSAW{qk
zj8qtU80`|UpLd--bCTcqwO{4h<yDgYE)PF=iAxX8bLQ-MhU;T!dp8-AQsOCv#9-41
zty7c|6v8sT@3Q!Vs@y@-l1yg~PDYR<F6b326~bZ>B_ujc(OKrIXQMIJAcg%AvQ!8~
z7z7l;5TgxAG$2hAvMghFYmryJ_YP};WxEk`;nX}A9(W9lM(fPoEzy8(uTOsvlV%ys
zc8gmJOMfc?<e+jpps=JAKa(T=Xa#ooH6ac;vV#inM+&@;@@IeJ^Yc;0^P^wA%clv*
zDFBO*KLt_wDSY!stGA!Eerf;_Lg0G^5Rz2Kq=_rpuU}u|GoMNDea{6AV;tvUwad7y
z%PtHCk&}hwxmuQG>})Qw-`HYx=@y+%oB633umU9<lrZSD*xcM=cWVcv!4DM8R>c0!
zBCo#iRUUonH#vIzB3h?72tnr1gCR36JJdEo8y{m!_@2NR==a;~uCHJd&7j}K4-_Je
z0gDPkOq`&NMo16OlcXj_2#HkQ-zXgT>C5G)o?k#ppCpTDHu_j=xpRAyes4gzRK$p!
z*U=hlG9(71k5uj<c0bzYq^qlr`t5xh`&)eY{@Z-`!DY^$nm_`TN(JBdXgBxSSleK0
zV-wI63m%RA4X%Cg1KQ0!9(eRQilrJ_XL*+CE+=Ch2b5+B+NOA3NZ=QcD5AlDIO<a@
z3{$C82}@;^f@IKh)sXTqnZ;sIK}eu2nYKvJM<|bbfj+E-BuTIme7}f|GWPacyz=}V
zYBirPd|{9A(JI1#w(gfT8Q2VjB?x_&%y>Z_gk43KX@j;Jou;fT-{MdH^M9n`_qlZO
z1hvXIQVIsqfL_!ElVY+CtBY%_t$xUfvyXE6+*1?^6~_%>@jQjHE(jarvhX;K@NEE|
z%f3^c((UYX<n(748JXhh2QRa*_(NKaJ)&qpy*8A?i1{TV6-Ic(Nd{g5;^!y9x3`<T
z{l)@<fJ!A~{>T`UQ#F#f@1SLagQfHH<~XwqlO;J6804ae408avN`=-LnKq<pLc6(7
zuh&8+J&FZIxmzVQ9<||VD&<kK%#daY#%cr>r36B_D%e_sz{48t0zDS#dxXA(IAd>b
zk#=(lGD!L@K79XmPMy0%u~@`fFbF(f5Eeiii_2wx{>tsAtqbSRGe2{b7hZe~w9nLZ
zoukKRn46z*m7I_+pybb&6c&tw?WGX-L5L?k$KhIwz(GS&N;2&PBTjrHkjg^^9uo2@
z&fPBVUU1K?J%IL=<KL827_Etu7@hNby>5f$+t0JVzfXPQe&&y#b3Z4q-qJK7j#IK6
z;c2(Kyz~Bt^!ok3b}xP+fCwR&ni?W%`>3W*sq7Pq0G(y5Ebp<m-(YHT+#RGW!K5kV
zC9lrhLm;fc2mwMNrNx+x)#Ynkzw!n`K($omiAT;cHc}x7OC*^g8VuY+nI(*j3}J<2
zXJ?CEuS1$7MClIlt?x5?<P5$S<^-?BQxfGV(k#Z>3}Gb3B%}tU@K9bz97l8-%XC|-
zw7MNUPf#xU)axTC-zT#c=?9pcEKRM(2oPR?QvN4CU_uH;MoRdyf@%a9F>qO0W?5Zn
zu(R8se6;BJhb+TnsVn{6(?phLZVz<^NQAUV1^c@j+`RfGX&T{^aPj<6j?GWtheZZa
zf@-%&)08wGP^(rz2)1`N==XaF0q?*414^|q9(wfiK%S8tWae13$pC}ZhRhl;mO^O=
zgP_~lW8?OR$iS!HX%dD|EO=CEbq0eTMoNqkNFmXrXm=x_{NSDxEG}bmS!v)ET_u!8
zkQHcJ&Enz~XHOqR3WLsru+c7=ImnJ(05(_)S_{Cb;#zxK?C!3yv~Y{<txc*&M)5o!
zKM25T8hblzZEmu^yN6Vuq@~^3;nvlc>2<n1^5m}&24##*4=)RAz}n26OPl7<N)g`=
z(8e<8cWCeK;SpjogXcqHW0dD1`~aP#7;8~pfF~^?(@3fAxq|!gwQzfr?)GK&ngjZS
zF0a3QhyA@ib2D9zA2~`OC0e_kx7H$Y(4fo}m_bk=2)z9LI>_?i@@uztxqbazy6p`-
zkup0o$@uU*z3w)vi_7!}mf`UuOwXOAS|4-g&EBoz4!9{PJQs+qaaE8f5yC@TL$|f+
z_@!Ea+TLSxZJAc5MQx~#wcr^ERyYV#DJ(r#x!)si^VhA57w+S+2cP8aH*Qicmif|i
zk8u3tJSR_{Lr6b|16`1Hm7K<=F<Pg1o}ySN<Ea3t<YATSc`k`gQw>&-BoWqFN|hR6
zse~^TS(=h%dE#y~2#XL06^um)h3^Fv3yuelqlh$($<mCQSANKw=KHLN23fZJ($h~d
zRIg*wlq^e0l88>XM{f|3W-03%n_Rnj>yrl%DWFtRq}>ofz|j+<Jb3Xu>#I!`Z||_O
zyv@YKm@9d5MGqhOn5?mAcM?Gwq%o{7-{i`NFS5CQi=pZ$qvaw~W5ew1tTX6#sSQ<7
z(on8eDOajkB)vhOcE88^R+Big%*>4P;d^iM=wl0<K6}5z?DGrHRU<f(EzNRFK0!!<
zUnt{yC5ojHr0>&<VuY~F&d+e{<XH@mW+|0I$pw!jK?o12+(WMXpIW5-wX0C@c<8Ys
z+`gQk2&s=NE}WjD+ez8q?XbMO#r*t~d%ZfvWG)DkXAVY6v=C_RU<Hh&wY|yZ_g-Lm
z@iNs)og=fyxbNf<I^71{w!&&l5PJB&PpMQQGnW2fK$N7cZ|<XHhhjP8&DWpjz6(!K
zsnki6<WQC?6-a5&nRZ!@Fl1IDJs&?PA_ypzYs6WMF)4+DVD88fhK5IJcbW*L2m+rh
z%N(5}6(}D+EFwMsZzN8Cd|6<0tis8Y<J`L5LrI@f#mCEhjMXgM*&vD|lu|ATCoWSa
zFV97iA`uv=9pNerQYLI~E_3z6mx<#6wQ`Lo9yv#SsEASp27?6Ubx5-miDh(j7;P-g
zR+DbGLz-r6Zheo@>0?~H^dwngvD$!gmz}XG#%3rZ4pm!1NW4OcIEmQVxW>-rbvoS+
zK_CbMpW4s}RtQX9VQLGhNzvAVQg}fD<$dxb=EqkQC1hoTrwVABu(Gnl##)zi=Q}+6
z>?71_6;dRbar_H8*W(4sL8@Nh1PaPkbUKHQy4?;x_}(|U{Lb?{{J?Q$rl$!*pCn5e
z#63(lpx4=Bd+R1EOYd{v{hwuS{vtvujL`@s2~@#-=GMu+<0QiOeT484N-~IgXlptD
z=s!efF>4!t$li8`cB_Rt=Am?kG`Ye~DKa5ZeunV8Ur6x%@wISg6PYfeR6wO#U~FO-
zNU>Vu2Sv2iPT^>bOQelPX9-{s!rdoM(Bk0<4~yf`(llj%Z;MWA4`Cz9#ele5q@M_e
z>a)~_CeYfDngc<RtKzKHWI9C(AEQlP9SS_(L-~qcugC88HKKk4VJsUfw^&(SU}XGp
ztg%Q0$_m0FWY%IMfRC5-{p6n#P%1kdGEhZE$BKOR^Y?M|_&D8GpF6kLiQ@=zSVby_
z`wAhQloH$}m}S^3BkFfpUA#gXbto4Lc+!<qp|X_10#UC`uiYjZ#AK;K3Pqt%q+F>n
z=*M)r5k?sLahHC#$@=PTq>uz*Kv?kcJcls4jK#y^WSYiUJkKXAgoLF6aonZf*(Xb5
zJkMk9$T4Q;kD`qwPVQDw91!WsBM2Rr^GPp1DJ8X95#RGrqR8lIg<t*313dZkDF(fS
zTQ^pTqUdgKgLK5Wl&<^~IC-+pocuG5`z+r2fY$yx*aTrylBkU*3~3axx3fXJxl6y-
z!&-sw2h@j0C{?O7S{-)wJG6Rz;v{Bu={CDttN4CEp%4;=9#YwyaQ2WwptV73hnI#y
zNTF204+7FOL1#{}R4$h|f9ZZ|LnA~<LYnIQEI95#C?DVR2@aGmpRh&MTFXO^p5Xk&
z8HDvwGUWWlDZccZ7dUlpg5~8Mc6J+BgsY@XmM1gz@b4RA$g;dD&@OmfdG7_5?z~HL
zcNLG6$<aEUW`mvWZQ88{aTF2w0mWj8a<xWk4b66sTMJvPtnbrkG<o;UmoPdd2;Jo_
zgvI9k)nU((&fKLT1VLEF^GlS=!;~u{P6_4-j-EKi<jfqIb+CmWgh-|G4Aw&`A3rQO
z-sO{S;7@EQIVV&GA&8ThD9YGaZ?Uzxj}jorK&O~2&Y>&E4PxE<(?;k0RE9Vju)2JW
zg<J2kvUCfKQ(F3dfHj8Z-VPgUtF)R82PNy6#_kgDzWxms7v6K+iRT>(8>Dm>tILM_
z#BrY_O)=V%rU|{yK8@{7dhI5jr;s9bXUz}MR-mm!=DeX066twJ^~saJkF##yT<49~
z78&$4E6dw_|Jzsj!FR6F+TW*KuoQfOF9n_iNrn;{-&Yh0A%#K_&-Vy|0MAp7)5Oy4
zwt4fFzv9EUzRC9TYmjU(HX;$a$=1@_yz#;x^VTb0V|Ql}DePT-4OjL1o=@O=NGbB_
z1f-`>VSqL<S=wiG@)XtKN#a3|jkOh*-1t7e@2c;hSfW^|5f(})|Kn9}Ke5BcShhC1
zC|g9ykd>ufK79WU?N%4x3s_p*W-y4UR7w~tu{uL*uo!&LJLn!e?8VB`gvQ=F!o(bz
zonv`%nVyA_=}Ah;!!Oqy-rbGa9c-g*idKq7t3_jf58wBwR*G1wDHJ`{))tB5=&)D8
zXwdm(arlOXQ^X(@ahkBX+oUr{@P$QZsUxbx0z$gxNY4)uQX!0WJq5~DP6!8&ebQ$)
zjx+Au-o^7P_$s8=ZnC%6KnN%mLzWh{Xtz5I4-Xv%4{O|e@RULvR(BwfB=LZq%>|Um
zn4g`evAavBvBT8ZD8uC{gE+z_mb91BN?QnJX|%hvJ1wGqpF$W=s}|5EMk&eG)+(pZ
zJqY69vUQcC!%;n~u?|j?5}_0tL!6~VgD$C#^X5RGTCJXoGM(ba=JG9}B+^Kf@8O%^
zo*UVWwG<0Jg+c*e6&RnY@Y&CwrPb*1!P|>0E^aV3RtKE$<A6tSN+RL7r2~FZWUMdU
zpwrx<R4Cwkf?j8jTB$%O2<Ud(c*0UHm+<`pN(BVPB851n)9DeXn#`nRamM!gBK>Zg
zv58qcm2)3imUD+5I!!V8o+{;0C=@7^3Pe#Gtqob4;QKzuPo8FYWQ@)2O_DUDP&`Or
zTm|e0xJuabus<iBKRQ<BzVp-Eyt;-`9`%tTGAoiK8H-EnoH{*=F)rvGR>UaR4<ypV
z=SxUKnnv8Z_9l03z0Xi>gtI56NoL9vi)ErHrPu3%HFyH*!*${;Ww5bMyVaxB>e5TT
z#mtd2)T+ZyWo5}TX)rc*6<lgatVQ_+kO5J@!`}W)Vx3`aOttDWJTi*(J&ceDU*+(#
z<EBs^URcaIxL@o_Qw>7msgPdO;@XuB0^jH12iG`#dJZ8iDY+_5r&wX}eV-r<@ICKP
z7=u(~X-vQ0;=OlY<1hZ~kGcQ8NzR`;O{owvhz9KK@6hiyi3eS7UteT<`xf_IdY1X4
z=iMDA49W`-)+J3^XT*ud5a4?SB!X^tpRhE}xrfeExcMHJuYQ|$vyt~ux#!IDeZsr~
zmVW9g`HxY({S+B5Y;LxxmuvVcV0~?quYL7ZdJPX9TQ)bEw3>ZtwHlz&+8*>*xr{&v
ztdw~dw=1K<f}*wCq}L7E+1q1wdY<#=AEdFn3wgE`C6SAan=9;h8eG4%%I01NsT7eK
zP?A1L7PGRvK&#Vas5VRzB{`7}RvY3tB26=te>eMBTwLPo-+GgDf1lYAjprp!=4aid
zjWKyd7YN}w1(noT<ze$ajZeC=O!NM`>x|T<@x74Uoi1Ph+H1tU0-aWjl7`)#PTmY>
zF~%MeX9pFDwN4g&(4=NbiF-|k$|d&pnv9$n<H+1G_I9>#h_xdR<8FuD#unXZz@6nS
z_FG*LiZo6rlsw`%VtMH{gTWvV@F_CynAcfG9LE@K@N&Jk)|$88e4lT8?+SC(gi;|P
zNmCHP;U0Q9gyVS%uxM>iLODfP-iz@`mk@%t-&i7TFJKW!>vQYII&Y5LAnIo%k!E>u
zi-#UMgXbv{Bk~8xHJVCf4zoIf(F#kewM9H=GCMQN#`-$lW`oJv9I5Z)mup}JQER}U
z)hDwlv4x%89Xg#h<zj)rmk8;Tq&>E_*BKd|cIu<-fM>}Y4D+*pP|*(Jh{e@ynw=D1
zI=Ded*XIMEq;wgdP)K21!>Cb6>k=Bz_kM09=ESKf&Yhp(_O%^+ugIfM9_7f)QC|4o
zRc_r_=gB7rhfTQTB(2C@9gHz9fCvFXvvlhNZe4$q=H5C51*7#k`~8fq^;N318cK~b
zRIekXhw=i}HrCkMZL+ew$K>P)Tl-g7UbxO9k3N&1y%cbLQBnv@4uLx%N<g7d#t+IA
zibL4Aj}(I0*(pw(K1*sX+GvVlku*yob1<792*N@k=l6cG%kw>|)e3`tnO^LgN#jU!
z^Ts;QJmU}!t21=!!~sZ&RuZiZf)pVPT3fW$tSwz<XJ?t_?k3}7LyQcM0g5b3Y3%OM
z+~0FZK<Kl#*C*as<h9n<2#XbF=gyF230P24p-twXENh(r!dSvWg(OYc-CSUQ;|`7H
zK0yG!ZwV_E`kf9M<#ZatxMxkd&+VQ92sl-QHW{7@h&l<&%NrEcC>{YxoYCt>ZUeOI
z+7)?YpHm7+m)PdId?9>%FC>ZvEH5w7?Q|Kgk1{$w&-CnRJn2&#8Y8nBokZ;KY_PVz
zM4|8o{WsoYd#}wPimB@q-<M=rL=^R@RccN_;6My_=$vpNhhQ{mmQpAb`Sq{-1HSR6
z|Cz;=H<%fhq?sj-bl!H6fzkM$LTmT1D5<c@aU(zG751@~z~Pc<8l#j)+S6RVvOvff
zK1F=rvb)p5o&}@HvrBQfe^NOZ%kkJ2DSayCD(&3`mRA;OG#iYKO>pAOB~F~WAKwcp
zmTN@GfMn2PYi*gumD}v^y~WyEmsF>usliyM1ZgyOh~fdEA2=0VUiFx~`#IAZn`QXE
zB}o#3u)vYik8$gr?=w+y1iJEkoch^f(iG{3d8TZ0^3af*1aJ>q2tcPjU}bq9j7BM+
zh1=`wtS6+s0wy)At?Us;89^<?m;_@S7bZmBzhVy(0Cz^p<suttmu|Pk&dx4pPMzbz
zg-2*L_W^-Y0fVSduiIjMeTn^@Ev{Z$qS;FbLLa3J#gfnd-Yy&KtDHJ>Kgl4;-?PMO
zO%f+0NsLe)p>hFrZDp13J^vxC^?l}sQ^Fu4%PiJNtad6Ii6c@G7HthmeKZ*T{HxLE
z((Mc=6heF%QYc7Ho*C!u*A`g3v&HW2J~K1pNKc}zJm9$91+UFvFH({WSXsP^B4uo3
zgh8)G7Ak7R0*!b;((e=ZVg@S0DuJ|uEKBM4284wGLyXQe+nY-uzz<Xo*B&@ncn*g*
zE(rU6fL{o(!qDsP6D0#DQW=|IbZnCKtxZy6DM^Kp+BFq>244t*FuZG5^b1_4&rWgX
z$SlwAY*4P0xNvTUUOQuBW1ro<CX<t+m@L6&30An|&SV-b98ZFfSevlAbcGMz`2jjh
znVX*H{`-z$OiI5W5rkcWFdzs#e1C|&=01CkHvMiwSnBZho6j>he;VHl9ej()Aykdc
zT)<Ba6pD4ESHS9wDDE-nHz^hb^YinRN>$qJ7JeAGJ2Q*1QX;*Opiul+$@5Q_@nT$q
zn+OmhaD<OB4Aldk{>%l2hifEpj78viAqY@PB9LGmBqMO}hcbp%J7RHhiy$acEe|nN
z8>Z3d(ClpBD>yPAGch$w8fEl)eNbgC-2X60c;VF#xw*JQSX5+MQ!3R6J&$D2byg@+
zAq6=U<s&^;%4(evM==v)Q+)pO&vE&~cX{#p+l<vLnKtzM1GLdt+-vug!rBZ^Nv!WY
zp0xhi+U~z;2`ijb77rl;vaC(0G`<IC&Q3EiIZT|S7-LZ?1l%Pgg~~rTiI5KCw8qlz
zM69mvpkzQHC{e2n)9a<gNrsYrW{OqD#-@q;9dx4cibMG2Vak<Pc;&5IY;HS0AJ6wF
zm#auzz;+d+Mc^{bAn@^wLR*__@HLM;`UKTlh5yIRt27$>jMX7YQ{p&2<cE~!p`<_`
zC`zow#6REVC#548jmhw2NV6Lg4;qx@n3H~6ND@t&YOHaJodcE7f%40Fy19lmA$$gd
zh}D&4I-L$f)jC5XlZ;Ot$CHXmbrhilY1C(Ld!5y_CBnk1yz%b)Z0vLy#F}A~qGUjt
z_Fc20(N2vdTp$1(L2UCRCQT`pD*Si<<e%|JzyI%8z5No?V~}PJ>a>8>8YRgwahJe&
z(qN4x)%TXf{P<d2*n@b1EVfwV@%EcH+1=3W?es_n1GcxC%*{<9gu8r<6)1Zk;&5r2
z`=||Z)I^w=a=FHxTZ^2UKSC&==mi)AGK0qEHl2Qxb`-I^wn?`;IPBZ<6s)c;(d+e`
z@FL9+xN0{`HEEhTUPQR6^2Uujyz=%UBmMv-BUcScmmo+1zULn%UeePj?GiCZ#Q)++
z^!jj_;n6alDo`tzc<xuvv$@je```MI<)tmg#_O&bS2|_bLA!y~hD;lzlC)b5mTtaB
zJZR!8NfxyUm0-8oB<^=n(lRzNL8(|pg&|w}d+hABiDONTjN8{gpx<pVGB%CTF-E6k
z+TtnCL9JTjNuQulpj@dSm2`bL#!xDiICI}c!eWUyP6-0nt0OY)pxc7XxyyU3Kz!d3
z(@KT-szjFTQyx-0_r(Wz_Lm-US%TL1e&`Mb!butPW;EAS<M;&?7O9WU(;q;q-C^p&
zEF&Ysv^!01-dbe1QlwrjQLYqOTDZ;X>Lv@TTgX6Cs#KYn7~#(HHbGE8d4~D<6DT3+
z_xnhpP!g<inWvHx4~bTwjiJ-+vAorxS_v5*9w8F}v5pbenaxFmzI(l%#19<iD+SK~
z$QrEiouAYvT%IQx8!OT8y6VA`B`ggjnoE~v`SMquW^!VbG)?i8kEeWu68X=P7?U>|
z3HK2Bet|+^gx%d9N+z5+GsW<5osIQ%uHCxB#BdE?T9h&@FWhEnX_eKTeZo?}^vo0^
zi`(orJ19Ru`5x0#N6<-1lqN{&=Om^@N`Wsutiq{mjAeVb0V<$Uu23GTv$F-mR+44T
zsX%K-*!jM2N+feY?l~cVf6u)%5C#E(Cm19KDMEzsopct%efQ1r%(D;Td5Sbmk<vq{
z03j_(dSDNn2y$_UfFyzCrCogGGgPlLGCaoi_8x0n*C_{z)2B`|IWtGQz0c<67TpBS
zUA#zUQ?`HjA<b5YVo_191Ps+iNTWV!k^l;$o!;N`6zVXE(L{roVr__rA9{qBU;ZJB
zmtSQp)ufrG*Y9DpBT+q1;wgbPBIgpFZ<!J5p3dNpFKaBlZXaQtWzEv!0vpRQo~aS~
zg6*vqT01_++MJv{2*gT)26rcTl*<*i*H+ly-zQBojvYV4?Bpy)8^U6hB#nuBUDj6b
zaP!U$u3cHB)67u50cj9u_V>2v_1h>}CrzBZ6=Tt9N|L2Wsga&fk|uO|5s8}S+QK?h
zwH^jTmO8%HWEx=|Nevb)HOBLt3hr-dUB9{k(H5Q`((V~<-&iG$LXtto($W?WJa`hN
zRIW%g2m`)y?g!R+-Z1DjY3{EwF*;0hze%UL&)n1mrd%V<GO~Wc?)o;F%@}CS>gqCC
zmQpMRAT&wZbujAKR1PtU{CgiXE{enIYj?lN)?Nph_DIthKX9L+%=t<o9g!!M^ELFO
z6TjSZ1%i<n<Mi}O6e*~Hr=C2`GtWLmrBWi%X%1V7k2Je+h~ejZ<f;PY^T4Ceva-C&
zd#`<;QlY^3=r|LTvz$G5iFD9M*o3&>BN_}ia^xs8b4S_T+o9KQFf+A7wc^Ygi{U7z
zPTlVs8F5XDU=KScMNa%GAyA%2wKPJxu2@=JWMXWb$=Ra}IyWdt7z_q<du=kEabWi7
zy0HyL<i1Lty{F4hNJ+62(Cg?N&MBd+&*{?>{Q6g(<jB!!vb=lWH4Tcx59<JH&bf62
z0xt-;_|T`he*G%9K75B#A!K}PoS~sH=8vAhBoU>8N59*q(`hp~J%eAY(rPsb>!VCh
z)LB_yLrO)tG|thZrw;eaeSA3th|{G@A&|a@FMJANnXTO>jom%w=I7a7sF0*Rtg&=E
z9g-x@6>lCv;5#xNutwth&Oq~C!iU9j{NxzJr5d+y_VK(BB?@RW;Ka!hzVi2;;^f(R
zGVL6e4=gqgT%C<^XVX=2?tM+oou*WsU~PGc(Xk08CnxfzSOZ3qL|r<qJ(d=4(`XG?
zTHi!S&2YWW*hrBy?coPLL-jEx#;1t~1GF(H3Bp)Vu75%0=P6SeX_m3ExB@m~c(@Kh
zi8$+EHT1e2;v~l6s*fPZAwXdqENgO!{4XSnR(Un6Q8FOY24P}+WjJzlg2~BYlH`EU
z4qaogRQV+&9YklH4_&vLu)4a9HImtxBS;z2={xhyscOLN^byiQ7vC>Z8=0k09pm|z
zU*+A)tE48QQWnma${KXi$7<{Pj&e7Pf(+5d=8CM8M;?9*A>n`ipYG6Zw+RDFzZbjU
zYn?l@ubc$l&tX)XBtN%s?Psk<qtAlb!DvG?$oST`-asTH=*Y6Ryh|KwipAi-umCAt
zlOMQ1j*$34fRvhEr^)vA4j0c|;?l)O81(vhevwpX^t){~Hka90S>yWUCAx!zas`z1
zXg2p*TU+APnM=fj)Kv@Ck|qgp96KACz!JwX?|pEax39Gs@pdT}QY@iM6tuHZ@|40#
ziP9R+%dKTTCU5yu%oGJfwK8en9!hg3Jk9xuBGpoWvu?-@*n}f%9EP7iywd%W2kIhV
zVse%*|Mu@NHC$w*q6qzvYPn9O>Jzv3>2-EV;{<6eMJ1S=n`UNWjODdOu3o=^0*nzn
z{n;;a_Ut7`w&g^#wH8Z$NeST^*bI0=5)_J5OEo+Ujoo!d#wJ<ay+IH<+mmjuiPqM+
z)oOj%>Tj&c{d(@oLO;n;QcxeN(%f!hu}Gn)NO<JLC`V^U51nK6q4d+_FpJ3Lyt$f5
z<c_@v=8m1^EC1kkd7=FK%ubC^DwLTVpJsf*BknZmb#_VO2xVcsUgOm9V>H`sR@N4&
zT)BoXEm+BuPkw>9xf57nP|C?`UFPB{2jfmTXb3}}p^;JQ^$`|?=ID`Qd{`PLiaYqy
z((Sc~20cukZ9B5V=Dj6u?}WHFDZT*8l>!An%!zn~BxIy2_~IAt=k%F5r1D%>vlO{8
zr32Kgu^^qpfg}$i0*;=zz%yU^@A%H2{h!oIB}VEa96ff1(`POccRM6;hos*F3ujKB
z<kYEC?Cx%{yT8f!$aQx2x3ETW>f|Ggj!rmUz)Hk{1A}n^B=?xclOEODFyk|I?%ck`
z^yCZ^(?{s-d`MA2zu%`n=%S4V;X4_%%pBP&+&Q|Nr2TwrbgaU}NQKRf4oVh~BEZ^=
zqw{tC-rxT;$4|_WrD<+@>N|flDKYv0O>osn5SA#G$JpNK;VGX-ADO|ElAE_~G3a#=
zIwlMhYpY8vEUvP#+rSSzj?T|;_12ba_EnPc@mYk0ey``q_FQ=;bLFG4hRj%+tv0ER
z7#bQPtPU~Q0ndkCzfTkouoyf~;`<U~BvN>}c)~hG<KIFvx)ix7tOa3xc6K)@%YbrN
zp;#)QGeefT&UzuuU2a$itZ?ckDSZ0<gy!BBQDhjdk5a2oGc>FyS85E^N72cEt@RaV
z=N2%c!W-{=$mVvJaU&^~N?4s?(j@nNlL(t3amlaAZHclh<LS>l!|v`TFZ}uc!E8N7
z_>v@c&#1)Wdx|VqfEwkIX4#=F(ciRwstPNPEF$mRt&?&@Cp6!_O!wk>hE5#A8bNjt
zUh^zjAYI_dgO1y1;WBI?nVCP%ul&9LfvmBF&JqUQ7V%&}e}C6OE7Hd%nvK;(YPAwU
zP$C^fB>g@^)f&${_ho+ROJAW-Dq?h+m#ogl<iJ+@z^v^+L+txLVOXG4E+GUAw5AtB
z)rV9Y2E9H=*Q9k|7w<B!{PI4y%s=7Ew58ol5k_IGAjB|hQ@DMb{@Mn?nUk(x#=5do
zNDGF%;q7kVILJJu6nIi`_P&Rh9Ias6YgnC;#xZf!rnkR^k_zAY*d$|bbB&=|g-W47
zmZT*8K82vf)6e`S&wlaODV9oTog8B8CYKDG-0;r2cHl!<uOIqUt7VL}^fQZ1pi+P+
zj_42idEbq3oYH}kOymMz!97I~Yaxj>rD_3bEy5~70dw>jOZrIbF5QDFTL|ks(yX()
zm${@kkI+~IVG#1vXP#qfxJ1%i##au`5z-?u4W!YK8AO^v95FpJO}P-z={B*ML8q4U
zmmcT2U-?Z+<qBCE|A>mtiIyxzSeM0FiBdkrVuAU&Syq>C;g_oH#}>Z;aT?R_chCp@
zD!HYHi|WS7Q3dzZ$^YYPY`jXfRAghViBbWI0u!SlzwzbAIDg?pPTuQ0TXso{&Ha%i
zjzD+a)dBZE@;PqaxWV#`cNrcU;>eMs%pW^L)a{{2i27Z6oeuN!M;IC&qt)64J;O--
z3Q>|0`oo+&dC|d-xwXEvuJ=f(JXk43=qdb4o&8pys26ke=n1ZGU#FNx#8FJI+sP|H
zSM53#T!!aKv{5cW`de7_p)%ESK&z4AsUi;A(>(t82|oY%hY`XlA_G5kdnP4VcV7wN
zEMWYgLc5)@ytKoSBS#n+9p(1z1-5o~IC*4_GxwdRvA4&qTesP4v^aA72-}SY8*knq
z)ldxzjEs!a@3wR0p@&kgYAl4#moBr0Mzce)RO9%m(|m^_aTX&?Mt{&lTaC)$dy#?&
z{2VRFgYUicfmjQJD5FS5HQ8sL9tG*McV~m8AH2l5kvc)Wj?OO)-}hbA&npm-HwCHO
z&q*qU^nIikP#!x%>U;F|?y#|Pi?~0aRvjYMhJLq2v)v>b^jKe8qc&8h(cEQndYXUw
z@Mn4G=`T?%meE;yNYX3m<Yzi>oB`o3WuwUe&+}amf)MDuTeBY<lH4yR9t=={MM;aw
zp$;Pr2e7ZF#9=V}Br8ib8*43me3b5U+_V|-J=#k<Ed1alE{=~7jEta-YkKsgmm8dY
zWDkXxo7Uy9lIJ6X0%LRMNt+dV%@x*{ZxHwTl#3-&Jo-^gv(+TiG38p7AS}>o?K3+&
z!^KCR;nQFIGNn=llV#5APvk0_!?_l@UdcL}cj0*iL9S#G5^Xe5rm+ze0!WjLxZlr7
zV1W`~q{Gl<zJI<a?`aQ(fSXshNGBU$1k$D)6$#^mgq4@xX8PoDhE5(olpsTX(OQ-N
zUUBFwD-QNi5)>+2eC%1${YA34Npp9NxHrJW1EdPcjHTP|Fc@?xR|;rt>9yO`%0)i+
zx##$;fAlYyn3~08sXP05C47*87-RE2be0@`5D<n%f}o4V5@i|P2r9mt-!ka+F<FL`
z;o;*d@)<iygO!4N$YFkBS!=1)%1CQ5LJ%T3S+X1-D?9E55UzB>l3`6gcP4)r?hcd_
z%*-C=@BgEJ!R6Qfg2}N-N@1BweS)aiqSx9XiDQJe3>5;7&(E^g7~`$?yZAzo2*syA
z^Bm_dJWLR%e8dBJ^5(fVGMxO{_X5i0vg_7Q`%F&F@SY7x(ugEW>GoO(6=J2y;XZec
ztu?vL2lphy|Lc|zFf~0yNP*^l56>&12)KA*j^Ft5lT6Kwlj`I!c^1NX0UGVB|D8*?
zU}$)p^ACNF<;7J3FJO9VnxXmxQ&Tep(h&E%?Cot+u2iX|nk0s!v(sF@b{i!H6XVC3
zo}6_}gVu4jI<qJ*_uuj)o={{eV1K`XCln)-v-BG+LeJ9awA>g2AzTBZaL>QvQ#_PZ
zd>rigr@QeQsOvq>*mXu_O2ayztBuz`V0dnd`DdR&NSC4bp6^P!{Nix|B=0U&3XwPT
z6bf!O2xgqLvq3R*n)*&JVRUR1Aqx!EMrk)%^m~1zujt1E?*G&mIC<X_AS5PBKsmC_
zTGu5jl|p$37q86ayG{Zq54ts_%)62`{S=bS;eTnGVx>a*2O<Ecr#8|8S^hrm_30G^
zS)ciIk>j#UPbjj?vhmgxM(3tE`rNY!CGuu7br^(Q@ucz#&&x?n4;2;(!vG-^)=ral
zw^6=;*rOL`)a%1!Mo}tO*xu>V+Hc|&Jfbw=Q@`|Oj-7qfk-M2|60#o|kmPwDLf9Pd
zcAe;PW{|#jSb8JK%@!>>g}Qc*!%33l5^8(s(rPVaIlrO~dV}s^2h%Jgj#FehVBTyp
zry>&Vv3GlorSHASxv>%aAxAojgZE+GtPdd%8n$v*TqzKK=)Ad90n2gvYwywA-yw-2
zDwP`24~W_wjDWb;XJdUGgJqCL{L(Lfh0~XwqBb%{cHsZ!06#pb+_}W&Jv;)PX%dsU
zz@Jx^Iv?)Uk1c7@$<5;^LZ>Ohx*iBmx&D=myZ?RfXYlS?S=>kVo1g{K8ji?}+D?m&
zSKnr2W(u!Tbrv)@=-LfCa|G+b8E}DEAU(zLGZ&ei8bLQ#=(hJs1_S!d{rt05MBN^1
z(E{a45p+hs(`B?i#HSwoWxn|9|Ik?$<v|$hJm_rRx1(_vqL5BN;)eyoLP(}H-Pls}
zAk~(BuSZyvgrW110tfr)Jm|V==I7qJA6XX1xm1H<t^)I@OToiOs!Wd7+&Qx%_xR8`
z6y(Tib$}`e@V$^vKl5u;0-vP0NUc=m=-f%tiJ0#G25}rAwP9v-l>5${W@~$k{l+c|
zL%BT0v!DM>Mn)&lIz1c~chEDd^8g(9A>-rYjEs)aZEZ6#HP4;RYj{|?y*6<ixnX_Q
z1z&45mID;#W5ti39zaf=9OFU157$<}NCyu9gI>bQ58tFbGfVBvNe59#58(w!lTQs4
zjwBMEmwO;7`kjcz&OYsSo2cJn$m=pOI!&=OL}P0guQWuV>{D&FsnzzFn4fnuAHot7
zbEi>Zfn+e?aB`ngSmR@Cc4%65XoswFB05)+fTLv&I=y2p7$nX(FijJzlmxzqlGYjG
z<!UZ<(9Cyl3@{@@W$v$gRBre2jbI=IR>Gi{vhu>4RAy%wy6-G_AcTkVLqtAF(Y1xR
zz#%<n3zDUp&Y<J`t$Ixgsb*?oj6$(SYqv>Q9VIM{(rUHo^*W5s&d_Y^5EN_7pS+)-
zR3VM}xDmzbuuli{-NStV`v{9NE|U@VE(Y)BE?A-z(#-jIC5bbMQ_|*gY6rO~nO`>f
z)Wv&Q)<QY(m<t;m8SLQ;!9If2ST^3iLVa$Uxo1BM`Ro@@<@WpbzzM-6G(tIR8ZQV@
z%5|)3E0BIjZKz7x&*<#88JilW?p0|uTdb`v5|+xex($kzI`=>MMM{;C+`Tpr>_+67
zvp5V=Zmx>Uqdn~gTpb8Xj4>E(7-TMq(K;hZ;#|n#s$e4}O1Y#$XF8u8`Sag`ag@?+
zM~HO5tl45lCFn%3{ObFRADd(L(@z}AX4Rp!h{SSW@9w+z;wh9L;spi#@#Az&LaVch
zO?L2<N3l?)-;apmgt*gVbA6pbGC+DBzx9v)8D}5*G{sWo(B{Q)Jql;|kc)5-j+c}&
zmxmuXdZ!E$L!_O;FODLV3Mhmnln}Y0=Ux8DQ}SMq-p^jv1>>|ffXq_T31cM0mv7K|
zWP#E1_qoq2hjI^F8idSep2))~uGPvgpMLJkL_0UpI;FX@&Y;}}ouZLM-9DTBE|qEt
zbV@Q9aO}u&p8fUT=7C3^aAFPPg1gBBfi)I`JA*<>{2(NhiZBd_jiDP`N}ePdM8tzZ
zt~RxY>REX}V;?-sf5ZBzB>Lxn?O`sA)_D7W-=@F3hKC?^njv~C`z-y{4>>hbr8qT*
zkv^UuI!rptz)F-K;CUgPPRz~gS9#~{H@R{92D^JZWLd`F|I$NDjedr)`Z&X5!-y;)
z>Nl}T#Q5wq%JbRY+oL=-M_8zme#AFQii0U?`XggV-FrR2;f*mki!ir7TE=FXr5{6P
zjB6B>>QFA5_nZORC6ul`^An#j<q4ks;(4CFe~$P6-7f8`x1HCJ1uX=<jTVdFd6Cnj
zb&4~ySm_*DB^FXgh$-dc`2k66SXo)*$_H1taq}9RJDX%l%x^q<f$?8{l6qwXWrrb2
zNcscNnyGm=w_>l|raC^4^g_}&!dTa#Eu};vF?oh`V7q_7DY>s1K02<!L0yhm8zc~C
zmMjm#+UOki6J!E}Fi4sAwa9z(L$wwPf#R3Gc%JdLVfCB)WKryhU2BQEDJwsCgVO8_
zwR5MiT7Vb2*%?`yLk7+hN%<jzLBi7F9X8fh+1S{i(`|C`M1{{id7j$vFdnYqRT+;^
z6_`7Agxc6B8@pRnN9XXvvNNAF99CMwC^wE)=sb8F5HdM;agd0TONy=IMxqP`DR~kT
zlNpVaPR*6&(2JBVgVyFAlb9dhPoU%}#)1Y%20JJTO)2QNdMthCB`Q->R8AbvD`MX{
zIpmf$BCkZ0uMl2<9~AICpUh<76$z_lD&aVL8#}1t5Tn&ey4^mjD|he<A^V+uMrV%k
z$Y*~IghD4Vp7PybFKhBXvb)cq%cdQ!e^*)Opz1_J-?-{6O=7ZwGgLhYzE(*xDes=Y
zzXcY(+lgtJ0b$(a44afaNqc3Ng>OAiWqcfeY{YS3N}{CCK_GV3u$QAIKFSXW%2j+%
z(O>_7-R%`jYABb7NHas4X7qY(Hr7_zZ#Nm8oaSFV`&-PNy5Q~-?I0*65Bk`gddeb5
zGv}t^c>xFq3#MsGFSeuw<KXA05B31Q#~FGGXDj1>%;5V|1IWnm2qPztGXKouEdJTI
z@LCaCNOHpg_OC9m@$wtY{qhLa;W3xxp^(BEnJM3=x!32r-}yd&@vT4S)}0#+;sF4@
z5Hy}RPQTZqB?CrkV<<c_V=0Z4Nc$ZY7H`w+_BeCqDU|Q$%{VF#A`(|_S*Lvlz=Cv5
zI!=4-<grd+<VtZ4DhpqcU`Vxf5>TBYD5I3b<ksL)S$ruG*}XwnVd!)I;nPeWn_>Rh
zr?|7TO=r7{5|T`~*VMSS$ofmKG5@PyqEa72NH+jW_|6d0_XB#plviJVgRg({zw+Ug
z_i1%oXrl?F;;{?kbUPh9e1=LxSmO-xiuD5hWWeI$ZF-63%(43r>hKzK7h4VyxC>3>
z+vl?8%xGugs12@-FApv`45O673Q3d+HwPk15AC<55(w?=VN_mOC9#8-J}HJ}t>w(Q
zX`cJF$La5l6K$-r{l@!vMvw~7Lek%Cvi!a0IWb%#IB^=8JH{fE>j&|Cr027?vdfDv
zzQ~I&{eXp~1-g0v%0Kzd&+^duSwa<3s+F+mfI+{BDnRHJSzlYDmu4J4a@J*Z77jv!
z6mFa@dB&{sCOc=RqH}2AK;BNC&E?6Dg(!1ki!4piIb7xT4}>)+X~36=>;Rbf`Bvad
zp85QJ9Hqv}*T2c2+jG|Y0$R(vEPeMSPS%GA$Hzbgs35@VgFx&deIF})8v7kqmaek0
zw8ZxAHla%R_0OGS<n%CO6XQS#8kA5>&CO7#)`&LO8J<3l70x*sA#(mw9nPaT@Gx_`
z=!Sv`o9}kM&u-iT^tDR>lQh9klRWqeGHDUUB{Ej!+|l2H3QI`AR6S%OYcM(J(G`-&
zShg<TX5_`!IP#^>V?#G)+sn-*ox`<skbdAJy%1|9{a&9zFGdKD`qUJ?y$)6usMYJ#
z>um=8K0&EOvcJQT(+_g$(x=GM1fw&2A;9Q7*vi9z>0}tz!Slk6Dz?^{vknqSGE0^v
z#Bqd@+9kKjB{-Rs&e1`AJPhk+43q-tD@Go;Kxbi<=1Xsr>U>Cx1QBIyzxEE5qw|bR
zP9sIxHEb#8T@w#d-g^BV{>z{KCvGp@q*5s{Jv2tU(?($lLP=$4h-RyWv@VEvz9TS~
zmY3Pw*`hvml##Jn3clxTQ;f+~qIWa01Lc*J3XDK!CLgP5?h<Pj+)g-EP`Of};6t1k
zhn?r1Sq{4dYji%g^xnxgAqB-^0jVUz=T0;J*h4hG{W3}Fpc5-WB$}Pq-=j1?!|=p3
zs#M9LKj*BOriKsSeV;%4FTc;bSKf9bU56?}04YJ%zJor#6a)oHp;Th8xsNA`3{~nB
zN>xO+!|KW^8@pRf9y`a__^b<#`QRmM@A@2BoR|RXDhy$59z67+6w)3j+r+`hQ$e{_
zB_kqDoiM>@lM}yAPHT+Le}>9MQ1|u}&mEm)VsexX-C*W3kJDS;ptH71ddNON^TXS0
z%)CZ<d<t0~abg7Fp#qof?QJyqi$DK5fASZ9#C~fR-^)pMEJA4zg5AA6W+skMu2w->
zBAc+Ve22BI4aVlqFg`JtOE3?VN#ekV5@RiyA@{9vHtuQWEMpEl%??Z^$uaOkseloh
zG&Lkz<3<wXke_m1YEm6snudEBtNi0@_Q(jI|JBFvveU%ti>$u%7M?knDk|7{_ZsE7
zG3K6o&LuM{aQ5=jxmsI1Zd_a9>)-qfUVHr|wsyCPqKLWiVIDbGXJ%rUq0$J-La);z
zC@HKhva-B{_DYORpK#6$Rvhk!unrn`xr<9+u!b}bzz2SIuKz?HLWHq)D#|o>S{8Zm
zl^TRJC_nFE`2`c`2dh3*;4A;=Nz`r0(l@@tAaZbA9BEdcf1BdW6r&G403tvJ1<)Bn
zJDU|R2#F%i!kz0}y>gkwr3KoZ7SBF@oUeT8Nrpy;F|q6Z50uaJkvXL2lgg0UspGDj
zJkxo=a6L@MB(CZ`xaVxnN#<FbQ>Kbs#9<kvkm{5;PAC=yN{L(rY8~>z$FJOeW&p9)
zA`2y^o_>Pel{KO}>u4b{d3l}gHduY*EoL8ioMLSR>G^n`M-pW$-d^JMw_am)eTmW0
zAue7#$L87w`+E&iYuVUrp!@<8(`A+xmbrH88r^o2{l-3>K@X`Ei)-5y-?_{ak3B@8
z5I6%rTq%Zim5|7%eHkm!CLb%#0c0gG<PWEH18a^SJ<9Rpr^w<yE4MC_#tF)Iy#!Vp
zvN$1&6Vlin2&n?@A$h*Lhgk3f#vglx{iSsVAKoOh!p(G70=?g0{q;AQdE`;b!($+Q
zJl{uWhRxM&UVimuZrr&_wOZ!l#WU>h?z6PmCC)7S`&~-aA^fnw!pa?@UPQmsW@m4Q
zMz@WSlBLx>#uwH&clsFaS#*$<ww435PUIPiF|PS8%LQ#tJ>uk~Sl4h=2n(D)caBP}
z#_FxB4B{luv>lviQcV`8Wck;H^!X7S>=Q1f+~KxZN^|o}Km9n(zxXauci?yiDUfN#
z?wjv2a_J(Yvqur0pX2mCy>_1u-oMOufB0?oTe}=PK1a1wVqsy)HC!br^^sBHLBzt!
z9i)}SgFfq<YpiZ<kXcL6x5Tk#c&JEXGBTIWx%)2PL)VWa$cz)4d~_7(-EqaPimuhF
zoH=_A8+8~YnX|{w?>XlymXRj;{^0??m|ljl@jAn!LkxN;Q%^lcYjK_K(gxas5nu-?
z8!x^|X?~vi_*AY6^zZ`b9J_F1nLqr)Kj1qrej8&miecbnqn;rwhV-Kdber+vDGH?$
zs_V0IXMwG~9Zuc<Y3lW1*Ex>GnuB4$t{T%hZhrtnX9s&Ib8k9u)gIva3ePJOCC=t0
zGZ_Mbl7MhTu)y)KQrzW1f066dnQ0z-`hNE2JvxiG>Aig&B?Ot24E8##eCK(>$#E(p
z6G-KQ^bj6+UO=xK@$GMZkFWjZpRurbn>dM(2#!rWN27I?vD!G*YMnv1M^N?XcMP}g
z+@dsol-lsP+n0PorMsK#-F5(xk1)^~S(f1m|FETjJoyvKBT6-CmN~CL8=yRa)fQn4
zp6|p*KNpFPaD-`=X{twNnE%XU>~DXIv>O2~Tb0((zP(Imdz<=^6G$mZ)0EA%WtMK;
zq21c&=#fc2_xWeo+S=lscRnD|hCm5kdi@Ij&;RZBdEnAHJYSKdF@s(oD->R-!tQ>X
z@4oaRfBv2K_#giJf65oX@C^0gGT6*@PzvcBOFiGY&}JD~k~-KWR})#`f%Fd58KEEY
z*dvc}?%X9h{XT#4NB<u>{XTxtM{B1hiuwbhC?d-;S8jUF>GR%>pu2q2G^I2(&ipfv
zv%9`UwA*ro72ruhcVU(G>N?{m&LNe=TEph*G7Gns*xlV^Y_!JH&pbsG^|^NSCQ+&h
zrQrEjK4APm{weogI)kx>Zns0H*~I7!uT*A#zs+~P`)$7W+I4>SpZpGwKl%WEsL1jL
zC@X{;IqUiEA<i;V?W#6Ux<hBB=Lmg=!7w&5!f$=$E6g7|$;+?2z|CuyiK77)mvP5~
zh^W^mO>_HiWk_>lw|m*dgW}N`jg^YY#~);OX^GaWAK+{>6`m(Zb{p(|c!lFnJdIZ@
zBRxeH$K1JoiyJqt(T`d@^zeDko;k@IZ@fc)kRX6}-n+y1o_~{v9=MNo7PGav#s2ON
zX%eFfK2c}DAARlXT)oraAO5|sFg-gC%Dq?j+9Xo>crtf!)|OOTq;))r6;2UmaT#I3
z4|(P@Pct`njHRUoUU}*J#BoeeaKn|-C?V?g$#M@e;TeSV58Y{ho+ZVB($G+xnPK*s
zCuw~3FNxZHm(WPaw)fb0?RDlJdW2$i7(5T9d>T6q-hAUNUVHl`f>3h){6)4mw^?1?
zWDv#l1}USHQ*5lRvAniQuia*6cbi7Hg+X#_af`|MHdB)$q&g$Z^2S;64>(Www2@?)
zleQo59@dR|L?WqGt32`8WB8t8bLkp!oKVh3XB%x?&r=lVE*c?19zd|a*j1}nC=?3>
z!$VAc`cv#L@6g}gbM6OH(Z93C#>=lTdh8_S$`DpbJTE|}hPPgSgFpB$zt7^@9j2$o
z>GuRlJirSSrE-=1b`zaSFq&?!$IUx8Y4ry@aQsoq#Zs>1Gl$Z3w{Py2lfv1hq^a|~
zlAg``it_4BDMrV~@X{V}oRR4SYYzOv9GYNtt{M{`PyF&T&GMb3E(PH+dhr}nH}A0Y
z^5s0+l6X>K+kILKE6hIl7``9U?RHpMx=z2>VSaXohn_gW$x~<fxBvFvktG?Wf`>w|
z-DvZl|NOgr<9jbt2z)ocItVF*1)?ZnW2eDxvrVQA|I7dWzp~%h<+p$9Hy9ruL1zZ#
zISY6zRbF+(=qyDGLu%ZH4GoV|ELINT6y>Qy0{+aIi+t(Vew!QD-=ov%F)^xh@>n0*
zCji(uCQaf`e!}cQR%$UO1uLkZKF-V|=jnduwVe2O<1Xn&v=)}ppVoMONEG#1UAaNK
z(PnyTf>ZZB$HfO8<lp@t|0mHPp&0u30#>$m`G5bxpEFb~Q7HIs@Kdo!u~4Af?X$AE
z%jRwq>;lbZlmGGm@=y7~7oMhAa1+y%a&srF@LU666d|$%YYmAusG!Ku&^XEq9P}WZ
zJ&fo33{{8u!WVuMrF?ew_o-HjWLoD_>f9`5s|{A9L`mXgs-Jup4+Rl!KvPOktuXWC
z!)z_AGT3Z6UJHe9p>t=IL9<1vK8DURHrE%}T)TrWC67IFiO>GhXL#-Px7pg*#ZwYr
zN$#v}@<090|ADEAVJejprBaDPF{BWLv|Al+FK=*jah=!Sz0S_|4*%pI|0B+xJr1~M
zOL|U#VUX_JL<6jKZpOk&!paa~v3mG^JkMifc!VSK$2f8N0vbuH+hvHNo4b)^q-jbN
z#V+Wk8CJxkX>t#{_oK`4F(&ud5==aFfyvu<Xub4KuGmrdSlYLjXm9OMpF55af^NIP
z!j0=}tgSFMTH&#$K1ZX`<jUn6Xl>|q6aK@0`UacZ4Jwr)Q4$gL2S_v-9(R^DdF{Wx
z#q00h;dlP=f6tjS$57HWUn28fE<bRYbDBDf73+E?tn@*;q)Gs@GqXJW;6oHELww`w
zU!@<#)Y6nJH%^UueJ3lo7U@Bn#D@l#zwjj;TwXnKoVm|FO7kziM-pjQ6`PFRcQ4cY
z><aanBY3`ZyIo#b;jOpdq}|=)i6<UJ3V83mD<p~G=B;(M_u7<7Ay*c!vA?@VtFcF7
zQd+%)uYKb?{E9Af;rtmq&%vD9a@g!EuxOj>m<|${9L{q7XCC)oxWK0#f08@5u96x}
zmL^zRKqnb-6uIE5HOdcN;Q1I^_cH^CBl|SgW(1WA^N&A7<K`-MuZxxnB|K!J*|~g$
z^ow7lG&DlgAF#8%#m;Vv@$qr)zjTsjyGN~BWvEu9+wQprD{$p|yXVSSEOBP;@+=3h
zK}wcZm-)j#{!f%kMZWUo-(YyS>N2{63{RVU7?RUxXGT%3jWRSco_8gi!|$c7A&LhC
zUcjdw`!pEC`|rO&6b<r@@Vi~V2mK9Mns9FzjO&XK)^$j$LV?*wA7JmsGU@UbnUDw+
zN*ngC-5_c-C>|Rj>vq}Q*<ovYmx=LlE<Svm-XNhEhSY0Cy4{|08nJ>-FCvbU{HH;Z
z8RC3C%fS*t@Xq`1ATX551wQxL&k%+I2#Zw;gUYjxT*l{l7!*UJ(^RU%`3!?RgB6l2
zOGuNH`6H+J>}S5n3om|$R<q?A4|A?v3PF_JtxDW%_j}nwmqEHSqQPqFM~`siu}ds|
z`&CTldIo&q(O=u9xwy#KiPK2svA4g^=H?2;VvWx}^EC5w^ZelZFHx-&>2}?`_e5K^
z_F6Pr9S7xEOQH>#!CCv~FC$8ueD&*pim@60@}K`MCr=#Zz{kue5442LjWJc8Aj3lj
zMe3t71Yr@Qvs~Sx+@8b{b4O0|+;hLq^Dlgd{pK#2PH}xC4wgvMG?yD&vh*I3m>*ll
z<a&Q&uqH*73(P+G5L<Uv&>Kx`E{BhsJ=%-Qm<J!l^L+MpHdtG}$<R=RFMa8AoH>7i
z|M2gBAEPy;f~#I%d*>P-UcE`77*Z@aEAYS%>Gu0<>@?~01}v;@(r!2Tzy9<8k^3**
zhm{uT354=7xOv}3yO|(5GZ+yvT%RBai_Yp@fbY3Ue#efU<~P3bk9q5@muPi54o1zn
z%Or>Y5D9}`H%}mcvCA0kj32WM5qL~J{s0@-Z?XCQ9k9+dBHC}!yuApIe;OeKJDaO4
z-?@%N^Mx;bnp0=b@L&G(S4onL!1H+ZovZxc{_p=Qvoqs_Vc<N$e4k!_z}1^8TwmDa
z-4AZ@um1Tz=gCh!?2=Dx?pIjA7)ugIXq#Yj5m8vEAUzkHmGU@x<Ot_4JjjC&KhDc9
zeV=xxL#<jQNfYA8$$E`3$V?OE4ikUV`bm~4SCc5^)IcCm(qs7OQ6?Wag;h4kG9^kt
z=hh<K%}oj>q4U=FSa|PkmX`03#2KA-M71`|)1P^U@riL~^^{-G+Tgeg1kRDgVW6J6
zYsD>vpj0Zdv9-ZpeDfQ;{L1U3iL-t-+F%dxcPWrw0IEQtJWQ=VLe%S##_?gPXRReo
zGNNceyWPYO3f%v|;~Y71ntl}L53D^b-46Vn?7ddNd1D<o%v_<UP0uj>$bE?1YQQ4!
zq@;aknby)W1z~Bw|00X;zRA+^9pX4;(90;5syy}7(~ORf=Gec(Kr;*4N{o;gE1i}0
zVe^^WL#3Ps{L1xf{ON!FuYB;~l^i^EW&gp5QgC}HydstQ7=>bmsMm8c)f`LGT9YOT
zy?&c0j+mP}#={RkiSh#a{Ro}8hxOoR>%1!YdF;p!E?(#GVe!Kv(+}NGb$%>&LvXNx
zZg<%I@QQnD`|B*e_5#bx3p5)0gkFU#lU%y=05h{wXp<fCKUp4(@BVU+fa@Z5i9_H!
zSJ?0T@Vor^UwnhDtzG9EWGyzIA!BTgyQ=^%tS~$}jVHZ)sE{Sc0!h=9!JtPR4Vale
z!Xpnqh36OP^_?-Jh1|H%lBDV3#rkt2F(OxbNtI`phT8l*M;^Z)FLcS3Ksj;82bW3v
zJ(N=HtS+;-d<U%MzH^uGq|e0o7`1BQ5L(MD^ixBt8?m|5;^v)I-nnv{n~NKCdQN?r
z>WuF`|6TsiKl}q07FY2DKUd7<O17K>mP#S~64j9@s<q*K-^ly2AdVwUrnzw61KfY<
zF?#)&L4V*rGvRRlG);-40dX|=5kBe{wakIf8y0I4!di{_Pd`d|xPrFELF}oedE*XA
zuZQpgc6Ybf+*o3Aa*XG`@Jsl)>arBN@y7i)<=rc{`S$l;=j-2oiLZU*1%Cf)Kj4r4
z@@3w-vOsq*;MKQY<$w8q{cpVX#@i_Af~g8TydXfTTmh@VN`)#8F*G`j=Q*n;DJ5E4
zk~rr4x%>I_)1Rl?kLdRYfa58R)x>ekpxYttcXRmbuU|hAKrBvu<NE~!Dvzihg>r@Q
z3uh@#k6~~=J)Qv5XwbiYnexJ`Ol*Eg*xVu;blF&0V0&c~VJ)9};%R>6SAUDM=g#2;
z&bZ6@je^K8mxE?22k*GA8Dl7gKBL1!+*!QMOE15|&Fi<_z?A&*b1e)$)+=B|fJHIr
z$LwxwF=)5wcH5)}5Q?x4B3fHydwYjMp~M3ZJx;woK^!G$Z4MvSgT}CXTGnEsfwLDE
zKGr(fVpu3KdiE5RBU4xm+FEBv-)hmjd5!YYn~W{LMYy-dpwnb?b&;)=O{9=K_4KoR
z>6gFEsWYeX{d^A(#+(|96?wRK@#^rwvBprVmZ_8reDLA>yzt^HY^-g80h5`$k3oPJ
zfLFwbz&Ql8+U#v?Fz9ya^?Od4mN&^Y8@sHmF4FDwId<$c=k9wDZG^*KbH&_8Wtn$-
zWIy3H4hEPw#`g=jqv{&ksuPopUp#{>`0jzl;u}Nf_8r8=9foebjJNn9?VU}ww^!L-
z+aewe`1I4C<GC;X2B%M-p-?EG^F6!^C*Re5e)PV!)-Y195yt~wc=-ihd;LuY{Q+p3
zlje@2!4I*jfC|e<<+HP~Nw=}bAnKFqB!`hKgQ&;q`ZCQ{i~0HE-2cE6fJYK1SnV2G
zU2|f-e?Lb6K3GCqtk#ay#m(*s3uUJ6KTm0P)a`{3cmmotZxe5AQd+-C<CSl-vAn?6
z<~r@YE~R3XU-{MF<n)=d6e}e+ZSO!*Z=Amy`9&^2;zbqs&fe&y*IwaU-~KLpyZcTI
zljZZ{0hCK%!g8HbxkhVuk7Up%(<YbjI(T%ivBh8zapv?z&YgR}aZgEl*eiC>w0Y>Z
z@C#m!A{0pJ<#3T={LCpPE*!@Sn?LW8_U%Om+dFuj9oAp_AvdmHcEK+5nVC7lqmMp9
zSag&BtPr`tMv@u9Akhp4G3h~4CY;0b2Up(bKmYNcaP!teKDL>|XAY4MgDMR%G%|rs
z9XEWSUd=K$Za9u(&YZr$xw8+Fq?#mg2!WEy4Wi22Jdux+wEQH5WeoAo4!&@ehXWCu
z;aNC3!pvirSpW7*=y-sKrCbtBwr@~e>2PE?<kNM<@X!p4jV?DY-@x-Ed&4b8M`pNi
z;Q>}wR?w#Jyp!_A6(=PXIb?D7fibzJer#rvZnwvs#hbkU;T2}*CMi}bxd)cYe@QZ%
zvaz;`PBlt+7*DdfwZZ({JcUAmI8L~6>l#89X*36v%4PcfK2fhnG*m`P;j+HG4<bqQ
z;k=7ava&d0@5A>vcltEGAE2`V&XrdZ)@n>Yc8QI>7Fnx9APkj)VyttU;(}(bB6wm*
zQm)Lf-08bL4BQ@0O&;U^OAoWYwvILyFLz88f*k7=huq8EkJDvKQ&SVPTP?0#zs!}(
z_c1&&gz)@42nmc(Sg9DqT{btj(3z%MsnW|5x`Q60BO~~}&-Ts+8=ISqY&DpgYT)@k
zd;2ZwWr$0zl1r1!Nv|_anrJ>oBJ)$%-qp*ruU{pcp2;mzY<{tN44*wkva(J4gBxJn
z{OwA-NA>#msoE}2j0be2$5t`m*466-fg(u^)04+|{P9oo_3b|+YjzH;ww2(}Ak<oZ
zG?CGUFz^{2sk6Jg!MpFi$9-o`a`N=?oKtYUCsqbnk+Qk9&C2pBCNoS;O|ieW%-HA{
zg|I-9BrGj2vc1!!(}^h*OZ59OHgV&Gvoy^`2yWtMmb$(Cd9T6V<_7V`I^o2ad#!*c
zC8d!OCLX-N*3PSBQ6GiHc3Q|QFETU}aNg^3(}z~t;m);N6w5_MrsjC=xi6Dx!@KXi
z>z-dPcaHu@Lii(r&l<+Z#z>Qt*WY}d^JmZSna?~9(!&Vro?$P*SV3dINn^hSnV~*1
z!XS;Y!WsPadp%ZHmN10Whew&5IzoGI=}`QVrg!1egMo3s=#?}Y*!}?DFXj?sMX^$2
z;(_z@@2r#VHV6fz`}@RK-=ZE>nA*9H-`8xdFR^l`!hH|k&*#7J>qN2U<(FQdzt?n)
zsClk+Xo>O>xp?R+D&;aCTzQ`#{NQ;e$A=jjuA^;=MPfadK&M*MZTHCf5ruLYec%Qm
zpw(%xv~-85nWLOObAdazGNLHT1B{ag8zUXU^D%bsCjyAkn&t0*pU%nSc!2_|os3gD
zn~32Hr_sB6bl!fC`Pn*`o;<_R<8?%qPzXcLk5}jp1|&0++*w>_;r1e9lU?eQGt{az
zJU_@Wbq=AFgZ6uC#gAkRg+j>K_yo1e5ECP#U<4cM8*FXtaQt*NA889lIv67e1182N
zu$jhK!OqS$Tbrv)jE|$FVsCGcR%@T(;R(`YKqLhHUKgp!r=aCSY7E42N)l`C1wz$H
z!p(2~CFhQxz#AXJXlL*!JeRT5P94KO-eUj7*O(YBa`A~XoVsrur4xKF;NoPRUN0t{
z8t2Z^CX2Tg7@g=aG%?N4(9q$$a+x1Arv-P?<^!nYV3kTGMn=b|4-GL|A42PdrNw0~
zUO2<>#5h)K2USYf=TR<~349NeYAlMSrCUe|W5Xk8ZCG7h!CH%_3JllB>2*4e=O~1v
zN$P?+kfxd_%1F`!c;}Ptp^%cjJGZ&^)j#3HKl}&yzK7thW3pE%GyT*<=-pk^{tl<l
z&T;>frx^AI*c6VB4RNX8AsidwgWIbt-d>>J?@<|>WNdVtpinqyk<9r%lLtES6IpAq
zRHRZHVys?6*HZTOcUZWy%-sB}Yp^?**rycA^Qcv;=t7Dyf>x)+>dFGON)_Mp*xlQu
z(R639SR6vBfNrltwZI{MZ;T~P4N;V#)BJV)oNCne@)9@y;%nUZ+rN$Pd)&3#myDb{
ziQU<!_4>P1D<P*JJi_r}3okBkW~|DmIxVi&>fBykBe?7_vbRrVWRjWbIo^Hu{ln*3
z3Q1<%T$uxxh=Zk+N3~L=R0!$yn_Rzfi%S>JF*G`oLt(--zA0C!q;Y~&Zk~D+^_ZES
zLmR`b+t;vG(rOQ=)ko>~dKCSPTFK90XG4-0;y5FX|5geRcCKDw{goG}T)LQ>X=)^j
z()1+Lk6vKs`>*0@&CJ{oN3&&0tqKoLR%jT_RX^nRwVNm<D2+^T{`>>H_x`(V?le%I
zgHFw1Lj7Y!fDnw0jes#+zj2LQw;tr7hliYni2#HLO3-b$+1=S8@CulA6NDuULi+ta
z>l>?7YnvS1YdXbPr^`^$NuIOJP4`UVj4b_lu<1_(5FrG6AAG>guYHyJbDzbl3?V#$
z@*HVH;B)lzPZEw=&K@Z-JYGjiNfswyV05(3iN$?Z_a#>M*j!(sQY_L_Ta-&7r%s-r
z(`=Asu17=pJ{pZqlYBHNhnL{Y^b}{#oFnvnN(D<XsA7_s{dZpEXemRCA9YHNT<ffq
zLIfTnpS5u3))n@eyVQn;h@yza#bvfOHhA*M5k_hO3yXJHSzBRZbc7=_BYE(!ha0I6
z2>MTWz=V)=7gxFV)jwkVH=aX{jsxV*z+i-6`sqiRFT|Xkl#I=e;CUW8Nx>RMMrxdF
z?XlMM!1UQzU!hzm;;U^+g@BVMj?->5v4?x8@|p5kJ|jMV9SDvdo9E=Q(}cdDQuGP^
zGQIsBn(w^G@MBLvbpj&{ZW^P&%YDIQpxN5oVqxJrK^QVr8=~FmaOKK%>a{8t?z@0u
zz>VA2=yu!GFWiT<RhKzuZpKQMHznRv1tLHr*?8?OY_&k`smBq;64Ddkd1#$cnxE#>
z7eB?=)=loaI8C`$ahX$+QK(cnRqU{_y-w&GHa3<h77F;Dhb)8~KYoneoo%wj4faw>
zfi_NJ`jI`9QgP<YDW<39DF&8WsZ8LN*uHs{-c*6wnTuGj<f?SxrZ1Mt1<*cOl5+dj
zP1e?zsn_eoam2#H60K%~`RRFvYXLWI-ezxahttQ8V$IM;&S2)`>H4FR*Lz=SYFPf^
zOO%Gngb!bW+-e*tu|dGhXCG%w4mdaCas0j`E{TXU%9S$rm0N5NIxKk^t1C+co`<It
z!^3rkhijzKE_`IyGHr5B#jU{iIe+0i<0BIk14HN)7<8KKUU`S1M;^n5WydSz2CPH%
z8bTWe{Q*}mU&mNOeW*^e+2Z!CJDfUkit&*ejn+OJ>ubzSjbZdqK2`u%1*e)!|9t%0
z&szPxCfEPuPdVv>njCe_&G}@%(T6V*4m!+ayIgwg6!ozxHqngLhd95t%`TRmL6gN>
zcQ|rP<5kLxj104~w1Uh%+$=V4p8iN8SS)hx+<lDJ>x4e|zGD5>P0k;iM%Bk0r>Y$9
zTCWcwYE`2CfcHMUOsQC6Y;1z{^$nWMeI6K{LfV+Ml`VXs(c=}Y^YFYIKt5*G`%_d{
zf-Fnfdh>1i<pPtRc{-nuE3pD})aS@(!0f|E@lAw@6RgdUp2TD+K@f0s$mil-gB6={
zYi$#oWz5VUp;)PK?!;+?(R6!#D%C2P(JU=3GuYgwTr3a-K00+X%a!sel|~q<`UEOv
zxH?I@(PQ)WWqRo<{+UM+7oI|vNA41F#$d99wdDnF+`3Gsm+<1tuTrU0S>N2^^*7&P
zXsE`+4_u<(Z<3~cGF{1M2|$u&q>01-4wWmPbh*Y(8rQB76@4bY_#DB+L_U2`lEo2I
z)gs3pJwhStV^d8Q4^W=xY{CPdBf}x*cXnC9a&2V;GR^edJjH62`%auiSVOPdr&6nu
zW*Li%OGMjyR7yoWUy&tloT{%vD&-MsRYB-UYLzjXn>%cO_!^VlCA<rtMx1;Qp$g<q
zCm5}XqaL?zU14$Q7QKGPx4!*ts<k0*++HB+_j%}n`zaR!27_kq=j0CE!2#5n&HGCZ
zhD&|&O%zD7IA-mKFE9uc<4-(6p-^-d99kojKIhMjGd`E$2}9N!AblUB-C&)G;WC%{
zP1dYpWoL`qNk$wUr#w{W-0@SGG^W{XQL0uk#;~-sOtjOWQYyHSq={>|3W75A+8C8`
zM!j4|QebC$mBuUIAl$x<y#E=*_|d%c+*J%VP1ssr;p)}*X|#I0`Sv@M%VidpR{8LQ
z51E;s=FI66v|Bs$`Yn<y&ZqDhlEh_z#*kMe=I1Ry5JHjkdMtkXuNYO9v4<aWrin_D
zB@v^;B~JX(dFr<7%>1<Bp!}X1DRbY-F3Upk!OAMa7^b5El}edMA9#q{wcALk2nr$F
zyF09{ZBz7veB7``Nky$X%;@Mi#UNp9c$%m^VC&5vGTmFkzxXs_`i!$Sv5o`MY0S#<
zEw10ZLZ>_6)z@FA+w1bmYj4wRw0Zb}OYHBgu(P|y=y1^mj?56pDLQik5&gFmK!lRC
zmX^8wwLfG0*M5cJqetBkEMq7Jk|#cQfr)sNg0HYqip>X&9UbyHv$@61IAZDcJ`xxj
z8|C7e^JGy(x7VXsD$(h6SzX=4_Y|JzlcpJkLV<~i8AgYzREr*=Kg8bBO<F(LqJHrS
zIQ{^_cYKO8U^2IV*RFj)ub<#~KG$#D;>zVK%uP)*R4K8xdYz$K6>VlQ+L9!hOPr0f
zJ+bB<AS}R5&FZJ@y!<Zy#014h9!8KNEmWf>Q|*{gW*FrmEM(0tez}NL`DCjgWF%_x
zRINm}Q)g#&jh*dvCMJ)df+EvnQ{4Z{k8$b2$GCC*CO`PWUvVMH2nqqs<{r^tz{0`;
z<x-8&;SowDLt}S`l9$oiZ=j0<Y%4)rc@du!ec>5wZ9JFr3I=h+?S&icH+GqrJi)Eo
zi`>5bzROxmCF0nPeUwrXh6R!&C5mH|@<{Wn*_hnc`d-Li>l9|Bg2p@75n~fnKK~ig
z#6eHxq{CEmhmxOSlt4(IY`={x1W2h!wZ#iU#?uC$subwdhS^$KWqWI#iK#g}zr@`5
zEcgG)V_baT5iWmth38-RF4}<ado-JS40?U;EG$r|)EOHaL3smutu8(ZomLa=4Pa{t
zsUrj#_S6&Dpy&)2g=Bw!kDIry0Z^$7^WOU(a{JB#VGs~UF@q>3O|yLeGWz{N-ref@
z84gJNpKlnFz*vcF57>I~EyAfOO81=y1Rj>r&K8q4K}rjrhaIHoK@U|7T~nzaP|_)n
z1sToZVHUO<yuY{2*yJ3AVwK6^NuGW7OFZz%6I{P`ofn?}K1piuLrJr-Prut^X?dAi
zWtg$?5p>q1-;a?7l71W0%izvCC>tXm_yTt7lsh*-rZes=++=fmo$-ld+_=5S<ttao
zl9aGeU@(YC;)Ea!2!fDk5WCCIXtMO~{t5AuklH;h3l>SX*J0y_uM<vAQan}08bg@H
zOt*I#2@)rbPc&J#hhHjy1>X-Cml==ynntP2>YW94wl^3Vn?d?MPd)KzKK1l7M2X?M
z-~A@1PPPcbfOfk<yVYWQYnNJel+m$a2HicPenz|5CN8uv3m+nrF5)4BJ#vYIP#Hx7
zZrr>=r_(_Bb>4aXZK6RBKM3gd`$SRfq^^Y`aTF6L$>CtG16lvU`~F2QYbDl7I@g!T
z>TggQ8+8VhRx{e!Vzv^Y3ISGYvQ8H-46)Y1Nb@v#q#*fWBjf6OZ&M#1r(7N3?8!4+
zdiW8>r{;L^g&(rKvf|!rx5fV69{pZSeQ1oKp@4KSAkH${tp?fZCd3Uq9b?Wsi47bt
z6Ga2AU;n^0822>qz4s1-L7zBHX*648SxTYco|7nwQA&~IN>sOP_gH~gATtiL%$i-6
zUw(`FkvYoaQxv3TD&7Mf(~Tm8Cn=5&gEU0@P5f#FommQnkok!!qtSq1%Hu_Ul=oKF
zxY_y;5f%u761LN$zqdzeXp~DAE^+MmDT?JXTbr981n<4~2HP8JSkqx|r^n7_gTcNg
z-Mq!nu>jH`l+mb#_elGF^n+i3at*SKovlq4mhW)(<XIkm^fQc%jPcToKcG@6^444L
zlct*9AZ9R#==b{cyM0>i4#T6P9QaDO#=?7-0A@HSOT-zgZ@kCQi6e}kK7&WbMA{^1
zcj)Ewu#4lv2+tBV+#XsBLEtkxUSnv`Bb@ShL5=d>+B!F{Uxu(iAt+(HZV!X%5D#8@
zkQ1lQqCAhi-Cc||y!qxU^xI90?y$SPPrseeG93!71uE4lm?|pnBQ8Hr)azp}JdITf
z9mgyz-lpAd^5myJ%h1RyQdkxiZcr)~dEv!Zh~t!gG+;0o(CKv9-ri+;Y8LC7X|Ogs
zoKtXb%N^m&LQ(m2w;C+H^aj<bNy2KKIx*wjEs|KHqXB-cieCtcTU~69sS5;?<3rSX
zZ9;2!$s6U><qdA%yoLxv{Gf<wcgR|8!l7aAzj%ogr_Q02VsEcOreofC?Io;9$<h{^
z>)VJlz;rsC5L@`8BM2Mf@2-%%{$EKS_!46JI6@nm`+MBFeVzH4d7gaoa|{oU@%#_J
zN3B}sjkn$*Niw>F0fYX4PP@aF^q8HQ!!HCJOprP}f4=}X4lAr1vZg$mtJ^HS{05cr
zaimgA>kgIH7X2*63Mh;WVRS;$N)S#L%EV}mV!K5+<?#bG%$qA~Z13z~ynyMg4GL+-
z$lOtm&mZG6PamUF9irK25XA%DdgB$sAfwaXV`qIA-*oUA%ao7#2tAC~+9iAIFG>1+
z#EDB_EIXSUY^<;H(4~i%oIXLC#O!Xb)9nq|+T3Cg#q|4qTFn;9D(2^o;)g+Q!IR~I
z{1<Oy0!|#D5rjst{q8l&Cyp@p=p$4`%6NZ^Y!DNrDPg^eH5o}eL3kd@^OzZ{@!58V
zsx0wWE1H{k?vSb_O2sl^oDiQq$9R2&b7wCwIyO$H<Hj?*`|cY^F<@`ELuaoC{S_{#
z9n|nBa?nG4_&jmHhq(AzkRCf*8|?0G@z{fpApHtSJizLdtJiN5#VLcxRo~V?r_*C-
zxXytc`9bsLJzJUBy!G88*t@mH)|>Bf{Mj1gXr_FLM}Sq3_WLA$A1^GBHhZLL6E7?v
z1Ix(t1Zro8M`$xwsZke`ys*|_rP)BIyV%AydpE8TRH~@)N!At?8LE#^sZ|*o9p>o#
zJj%lUZi{}ai|9ox-Mq;})WB4R(e(gnir7Jy#@pXTTBtns%UBDo-EBrnRUW(lL89J1
z8_PFPHsSvBk8<Vmb$b0iT1(<M<Nf#F<K&4`fS}Xuy1s}sBZ(8TG$GB>dpl(0)7-Fh
z)|;%o{1)|z3F=le=SzGOf|6u|K546uUn*jBmvq00DwYrh#qiV^au*)9ZRU<w8C4Vf
zV70+=vx(03FpV8{uU#gn)bJ*zSYKG8Rv)4|RHs_6aqP$uHrIT%H+C7cdYE>Ht<4qA
ziXM7&3R4YGx<uUAWq<u^lnN!n6AzFjeOkLaoIQ4m<1@3|TwG*(;||l~qYMv^@k9L*
z?RF0-3#_fJv$np@k=c3ru~Wj>Op~Q4Nt_TxQSR$>Z@Y+a(+y2#*?j*tQ>U*m_xK~s
ziHy-8K;b-qvew}L%if<yOLkr7o#5}>`NkY?elPRoj06%h0TLtuk|M=Pq$tUjWXrNf
zt2}kfw#sf<w&hyg?pC?oy*w<JT&}WRWiM5g=8+mG&J#dlo{^cz@#V{xZ;m<Q&i9^k
z`j7iEF(pMH0nlx)+KY>b7a0*Z?v8WL-e;e0e;<ZvW7sa)aD?gm;Mz22+Ssniv0|6S
zn#Zg)$LBBgIX@hbWPN0Rm;U)Pc=aaMZ0W;Vqe-LPrq-x(@W26fw|CiI>yVCPlKwhT
zcY~Xn3OR2gY98b{;mRrEFsA;|AII?OgyR7f$K{>3-HD8c>|8mIEpl!?c9dtIdx7yJ
zKov&W)&1gW?tl0Fq<K!1CaAnX<|QgiX?2Ym;muhI888e%k!!Y|I?Lj*BP^dd!m_FG
zl?U3UNF!93V>m8pw~uK%2-l|OwlFM%+jN(GN2<))3w-WUm&?O3;c!5>xkjs6$C;Vq
z%HlHhW{X<0iC?cWKQqs;*QeX-kPk;}?`*O^AE8?_s3{*Ied3LajC)<0%{HPjM>y`Y
zG&9GMefwD3yu_rx!Sd1~+q->ME?#0Zj)<d-vuDq8&)xTwIys{uGAmJRMNu#sjY`Jt
zx7x%WT-Y>-M=@&;KFQ3|JV(rgY1_t>MhTM-3QW%-8xKl-q`mQI2*v#(;6QzfKe#CP
z>Uxj$(H3cMmzAfV#BWVg-FJZI^emNn9mn@*R$2u85$R}5(Cd(NwzzA~K<;x8^$Pi*
z!|2h^P;bp)F5N^LPFS8_Vy@NTg)=WAih#ohk8=L}MLL}xNnEhGv5Dt;#7WF(Js^o9
zWS%3Xq|@nA6yM%=*9{<!*z55?kjDj^4?oY`bc37Pf=VGvmQ~Zk^ej{?ONgrPAtwQb
zRVsp+hKb*7kVO$URSo8EZPK!HKC`CjCON$<X4LCZ9S*Q}cG$Ucnc2BTEZ;|~67!sA
z5*3Wdf-&2JE=G|v<q3SFK;{WWnxG4VeA1_K<rLz!_oCf8^;(T)&F9>iQ|$D*tgo$L
z2-x0SCr!)D<<f<V*p9{A%p8UgT-jJ>eRGqQORFz&4{1fWKf0kZj1Aa^CX^-Z`R?-<
zSy+6M!$+p5WGSX7C08xaLM8=LNKDU1j>ZTx0n5Zd;5C}$Va$<=!H=J4(y&S&?xv*=
zhy6a);ShUghwZgh<`$MP9e2-%CRvgrgAvhW%+9b&Lu+PRCbr0tc|xAVWO;_{?_#aL
z2+KDkfSIW_ot-V7e(EtUTwY}|>f<^#+uNIaXD+MkTUcPK(Jr&-V9d#rr#N%=EQ7(|
znjeayAjygw@}apm7KJEFf?1Ta{*9;ESGSpV6D+L}7+7`%ndPW)j__=RX;36F!gtUb
z+-eout1v6z2and6vT}asV!>+YGR{-N@fdeJ#@^Y+Ut48iajDd2(FVdaNTUQf88My=
z=}m?lG7V;22cZ&*BqldY5=OSpWA!#D7LVh*9*v5}*%wbToJ_cM=^Ro@wl=Sjr74}=
zUDj4tsd^O_=jKV`n3JbZ@#2eT>2|x<(mw-O-`u7(J#($G@kY!5+`WZQAxI`MTMs_L
zaXX=AMF_pu)ofQOCK;k?V>%8hP0^;h=EpP$SuzAac(8)2s(f}`vlC}@ih};IkJ<0z
zT)E8B;yxU|g00KzO(p^9WJrH9qN*haS|+}cqq2k|Nl2pzBMdOR8_3!;Rlh<w?(@*s
zzrxmThn=krP?}zEL==TwTDd^8(O`LLKVca1$YYOk>cunk`h#+Dr##Es?euQgXW!13
zS_=wbba{vH;m2s-GL50q5*BV(D3TOqYJ}q;qX>aSYk^y<Q6xFXO~DWDccAcivMU&8
zIU8Aq*jmTf*<xXCj%uZbhagd!!DvJnj49HD+baf36&G}dOcN3%5zz!|>mq9YAg*Pz
zwRVvUXI^Bp(`Bc-OOX~##v@RY^|e(RwK_?f@r{Qc;rzu*Z13y<dkwZl!Du}Gw!XM-
z05SLA$5IQFAxL`x#*1gD-abo~XBb-GRQ-}VB+W2f6Vr7lCK{EbXrqMWS(c4oZ;)pN
z*=WMMYZ?|b4{a3e;3BIPQmq)J3ErTKC{hf|W=d&H+hQ$<7)%1vB4^GqxvS~$jwwye
zRVA}zA<6O#a)p<Mi17}3+CybI&pz=8oo*l7o1)Y2GY)!;dLt&I5mIZ;oqrL-GD*{%
z!El2YPMu??v&(Mxl>^TA{To~SL|H_UTA+kLMj86-c^d6Q6e>YVa4KHuw=4sxFdUm)
zfXFg*VW0rp_VDU;@-!nEO?X#TBNj9dZWe57m!eW3LorEH{9z9vGlVdi1v1m%!Z>6c
zhG-=@RI$0YWpV3_!m|{nFrX-Ta4ZGRq=yXqVAY961HSg9&mfh9HY;rGY_he!iC1Y*
zWEuTlhi9IBiYzZkvy5*({1}7Ln5~^1lJu3_P4jbeH+&Xi?tvo;_Walx;jdoA-d`n6
za!kv@t@{|Jfs_T7<C2*MI?EwRky>N9K9zc%EQ%R-ha599{``J}udEfUXp_`)kfuS9
zX5_&TV<RuiUqB?_!YCj}QYxnAuDZ*;EjZ9rSe8+;ux2?$oKUqajNTRW^gbvG&c5(0
z=Pq2JQd?jc2K4&d>~3}#4@acF%acz(L7~bfxJ#Ee7zYz}I^ApZG%vUI%<K*M@0GqR
zcgAv$AjTYxX?8c!9wfQKsruLz4`G?*SWFFMQDDR+#>#e0{92VXOi0HO_fI=iwBQ>X
zinYunG91(%47JzqAPU7Cif$oU4I+eA+|{&rPt)S2mZIic*p`hhBzc_TIu>Ssm3-j<
zLKVFD!n0hwv`(Y7kHKin=K2<Ks+j~Kk3Rl5!Z0WbNs^`nQN&~tyd{XWZ*OG-u+*ij
zfUTh!46#R$rWv_y<27m+j)m|XiabUZ5-Sl%6qe)RReYi-VcK>03pX{`XLWh1XR@9{
z<~o=d1aV9h#<VHej#IY7ge70_y$cSv)Fk`b7QW||V=zrIa*5TyOg!Gk7Md4NpXS`f
zb*5&Q=?(j=UEZKSN!ji7+3s{HWC;>Ul9VL9F}j@V1`w;{MlO$0w7~NnW-GvCOp%t8
zOXWFkt%6x`u<9OJ9FgS(u4!Qy0!fbLSva;$Ap~Bd&Z2Gbv6%|{y<MK`<~%b<nG}MG
zUquLmSeR%jFtjEML%KnTl!`lNDtvO*V&2ZE`6jJe6W^;~83t*VmB^(mAzyz1={ITA
zJkmn&%*l()%%!-#$0SPFdj2FqrYMx=E8l#iWcJm%#8JO`;O(_j+)&mhz%q5Y`0X7<
zS(e3o!@>-Q6lq?v%x5Wft%~7Wn01dL&&cu|*RV=m4J5*p*tUfsplaAuOoP^3mHm|t
z&v!GP?q`e(sJR{(77->w84QqwNyH?Munf3=vCfBDnwFi@sQNVPEgaXy6b35K5n5qQ
zx+Ght5leT`YP8tuhHPxFF~4wtLL2PtY!k&>gqh^@#TCw6ymYO*`p=xWH=H?3Y*Qnz
z^|lD=HJ53-Kn_RfT%k+@WR6+!u^KLhW0mNeyudbGq%6>8QL45%7LA!U6~kb<NLZ}w
zaH^Z|csHXH<>Y3`;G75(Axyej62>vA$XTe^e0<U4?rK5RR<xQ;>a_;8Wunj&d4{DF
zdh0Ct+yN?X1!=gvaAAXHGsmy`jN^pO&IO_j)OgJ1Zud2fpKq55Q{&jG<ZcH922)KJ
zU&Rz*xx>h0jx2JlrjKcteBMPSaYfnjugn|+a2$)d1M^@SOwTqrKHcNFe$L6Am~mQ=
zC{5~A(RG)&$Z>{44!I_`Pgi;WlwjIb)M^!`8q>IrvzKg1gc7KY73BO89Lr^QB00Ib
zL!*&`AqgfElB^VIICtst+wQY(WsYZI+fbw>@uTb5v}>SciZ&E7m&Ad>ZumID-eW2+
zFbxZ-6sB3C1glk_>P&@MRWMVna?dE_x!stZNluhW#`PvetA!YiIJCLR?K3|2%{m;b
zf$v+iTGLeh8bSzc(<Ijt-QPuaE|bXymwORsw|aQ}5T#4T@x3#0#zFX&_~JT!AR5z@
zuUHT&o<ZG{D3zmaKr3V(6Gsx;^f3yn)SA}%nwpi&Qlw)e08?vP(@liy&`4`kZHHTy
z65cV)xv&!wCK;nBC74>Em}*nRF_qn24&*5dEth+zEM{zp>({7OYcv}zY`g6Luq_MF
z=)6E~UqLVR2!fmo8(p@ejNP@>l3Pg0S9J7d`L{VBe-m?B2ynbo_k%rjyz3Y=DjJaB
zl*pPwrpQ=hn_lU|JrSIukl0SCi7oB&{J@Z!X0ryaLp`rjwH<C=Ot@#5v(gEPvVwk)
zlI`0^RIQ?-2zPUx6F3~2_PL`Ws5=VJuTifxs8#CMm;fBhB-09=q!iniF=lUKeW%04
zogsrs#7=m2PwRURq6#IA$Bnz_+nW~njRtPbAsDBWOCHpGf$s~j4RA{m!#q!rxyC8L
zEKDrhCRNcM8?r_gIfh|kNQLjY2;W01AJ;Tlo(Z`#kX-6Sq=lv%B*>#TGju)VXoS0R
z0n7KfWya@_ukn2s*L7(&TX?Pql*xf<mfL=3G(ruxm{@bXcxjzpT+oX*uf@<;G-rQD
zvgOhm3rkcRb>j6MO8pE$yXoN1G?6mLK%vtN?Uellswi+gm&`WzxWh{FK-02NX^tUt
z49lTWsd8vWvCz)AcdkY#1&K6B4&O$nF~jik3e_{uGi6jb)O4u18pm}oEeG3it|eHG
z?T{rgemJ2RY_TzRIlsD2EEVyf{~hzww>QDmO}Egj_(U(fh!6&iip$*cK8&iS$YX3#
zlEs)c7XZuiuzd$v<a-GN<+*VbA;S^E@u*cQEK@LFu}PEwr$V@N6Pwcq*gkujg<8gf
zBbcgMShhwM28QLJg+|E&$FT{D0-eUF;SOtq5$l_~NUaHz_-*sXYj+?*VAkrWtW=pZ
z(X<*a)5`~N97Ph1aXbqZq*yf{v+7f7@tRbc4T|vusY+@Mtu&_RmVFN6A=)t!PRZ2i
zW+`{hI2@^)m|m4k*d)uh5G<eI#WSaAUU-h?z+k@S(eO=dTZ1-{+SqFrE`?`o*TX=N
zCNWkL67;vYd?jIReG_=|XsR1>s^c+r%WcTj%OFY`m}cE&Zh0AR+GRZGQmwkkFvjvc
ztVR_?N#0j!H%a;fq$)~|E2XhKk33W4lMyO55U!7HI@m_ci7A@{wK{gCPAY8jqj#~>
z+Q-WIGqn7a*<`|O&8O-KY)g~pdo<aiR1TC{<9QzAUXIK&jG#}@+vfa*6-JW)5Z8o&
zzLTa=%-?knb2BrHzxFi@3{1mdX>pe614jwQ9kfpH-3mG@aGG@tw><N3>UC6-P^2ZQ
zNEH&NRz+nJnM7pq7}Kv}nI)m!{;FWw@o*}2a?2w>bb?iXj*Uy_nbtk_iGr!Rk8NoT
zp-GaGKv3`Ly?B;g(v(CoHJKp?+iXk4#^(0k-H7jKu(+Pp4jks-yWh*j-~M%sPC(Un
zIdbGEUaiX3m5Vf|tGLxh$>i$U2;0WAEu4CVxHrVKZG_a=)e1(TDYA$t4zMe=lFTnm
zY1opcXHjW13GD!_t7InJGUs!s>EpN#sz@1+QcMgailWFdtWv^x((lujImMvE<_?S}
z!GBhtUEegd<-;6)&;1P7|3|Ft3Cl}UwCi;SJF5sGac7(4afF;myw((k<6u>*1Up?E
z*D1?~6~Clf$r93VjOABx9G@&5L7vj|9PF*rtnTihgP0@JKKo``6h%&Nr$d^XIJGjU
zYpt<dk0egWCjmwnFc`%oY4Wyv<28k8!?akq?|%H*DNHSK9g|~64$`j8Fdj^p^haPA
z*o_LR$V!4X+XBnPY1Aon$=`!f)~-#@M|d8V<B&%&c{IUsU2Mk#El@>_RYc4<(A+sk
zaPkp~$p}9SxpjY=1N&xZOf?b0Vm!`E>Sra9vOpFYmSv)qCJrOaJVS*ey2Bn}{PsoN
zZ)c|CaPT9aWa{|Mm|9^Of&=^Km~GDy^e2pV`zR@~>s9i+Jct#B0j7yns~{0%Sq4TK
z01eN>@I5ToLuM&?IKlQDY_|ehP~;&Z3uv1XZ{;b1laG@2I<$4lE&HcfT%4v_uaaeo
z@i;Ga1C+X^t!!B~X_}El5mpkD1Vehm!FQ;h{4Gpr{DplS`rbcJeg8fTrKtNZhxX55
z<dWgm4(TvJ7zRpc!fuy5iqHrQ#{thHNzy%*Q4N-j>HC<jk7-!s;TXdNzfwmClQaz}
z;t};CMx6R4!E+B0Y_2h9YmV-pVS2WW?Klj3AyHHyDLcgTB7;4ivq=z8L=k41G7iS2
zN+0mHG$Xu0Gi`_Y``^#vd+x`uEN0stQw_~{_X@>iOfec04Tg+2wwSE0k&P#)LLqFA
zQIH|8${Sl2mhWL#s`!2t6^Gb{NwwBQse&|_P>cp-7hgc0c^VN6n4NAhzc5d=US&KO
z6HX!wt8CUS@{-10YfU%_NO}XDtRTsA<c*T3z0D>J!PGtXu<yG*M78d+yw62uyA;s~
zQ{{xa9pc`QaD9t#Ynvh}5rX}&AeSYrk6~L_u8UKt;rmrcQhdk7@hT{plSN~S(SYi(
zgJT+~sagC=1+*j{Pw*=O&7OMWURfZEQ-W>}V>~H$J8zSu@5}AFo#@l=ypP4%IW~Xy
z)6{H>#Th{yt`qKT;QAhEr-#&vJPa`{i|WE6PQ8vW40248Nl3F<DR(B8h2=TOL{V!`
z5syd6AfQ&Qkr)P}QJ;(mQ!0um#HI`GG>uw?>Deh9*Jd*6)7{#|Fe@0AjmD=a3JhH~
z`(=4f9wpQoP256|$?tT&{7TbWQ#o>sscMCdOIPs6F-vn6GT$NCUB@*Il5P);CXeHi
z)_HCLx81@pO_X6V8i!0zRZF%{)5P)|ibPUvO_5H<sL4w{9FBYBX@tlnc@*H=A(qvo
zHa$gix`E}G^tVI0n_X({DYTITqO9BR=}xCfO3>@ms<kksAiw6o|9qYlTGKdjg5?i?
zjIFKTqTR4*wF@T0HH;#`!e_Fvh0F_do|BCxR2P@ZrdG>g5adkR4(?t(-}GIy%&EAI
z5`8uaajR8~YMs$!h|DsgVn`9kG(D4v)5L5n;?-R&%V4mxO&UbBmi%&Ovqu;ygg{9}
z9EQZ*0ro<Sq6TlOiSjKrtuPxk4t&oiSi88wLM6j=Q=;Agzvh?skYre<Lmq^LYLn{R
zJi@T(M&MKwj%Af4Y0pKYFs-uenny9V=TWIO2`57|63g@PGeM=<<e~}fB&KExWR@~L
zQ%4#SB{f>fYZTH&ni2JfxPxs-Yu988{{POj!f;*of8-<dI@{=)pf?Wi>UA8?#kI@Q
zeHMkJ!yy2sUt^MLnhLH8`kKCnOfzf*1_pVSQSoc!GRHM7v?oY+6B_f=T<~1%(HhOH
zpfS@1w>%frd5&=Gl9aY6m<&d^-Aw=?LEgTSq`AjS_V(I!0*J<__$+<!gG`+|g&6Hn
zn{i3PK8|bQ_)e)9Rw!JzN}hy7JKG4Qv8z>b;jkNwsW)F@lC)eGy+_C3*IVR4fD9wt
zN(HxCN975iN;%TRX0}pg%bVsxH0Bn|L}sPTYJ0XuqDy^>BG0a!e;^fc6p>E?bax$5
z*n67JZv}q-22E(D@4k!0_uY?v>|t8-HKJsKX@FC!V4_fk!ml*QlZdFZQ_9>n>S)8E
z6Drzmfk}Cbrt6j@Ybe}WlOim$UAI!fuQkzmhG_}BMhiVLm}=Fy<hoSGeQs%a<XKL=
z?on;o<hrEDK+0=-P%S0lB%m=KpgWfkuSbsTZ8nWUSOyF4dw``2=kP{bWTrtDhPd?_
zmTP0~K{mCNB$E-z?hee(pbdv{Zs6w%&#^#hOvgg|9;%R-P6byB@;JfvD^#0P$RxtG
z9h6%o?&s8+6)yS}%>FvJwkya&F*DypTIE<SuJ&i_9dP7HN;IC}@2*j}3Vj{9o&Q{>
z_ee`?2M%%gV;{y?zsOoIV!ASgWrJVwuxuMGC0P)Xj{3yCF50b-7df(UF)XX>Cb29`
z*F$GHh7ERAkY_oyY6BsR@<2C9iTY#OONa1pzKx4#pJe}d2Y=}h+R+$$8L}Z>YWI^_
zK{N^RCIQ{ufHZqON{p{05B*M<-upjqc8;4q{2~0!%a~aY&x*0ha4UXEUt*QumvlU0
zG73n70Ar?J!lTh>Wgv0(4owA?D5+eUjVaPBB_0i^)g0{ES&o0~d)fH%A2B?03U%-v
zC}QTSHhCJ+uq396NmE6hD^!-#+v!rPR5-MMKkex$9LFY268eK78=G5%(OXlYd2ImE
zfRy<A_Hpo|-%W7xSuPDsS}C-qW-(Ay8+8mrprj-VBa%UvxVwweGZcANW@Y-Sc#TF_
zCZ_L|XdkI@YE|+qMdp%bdzK`c;5#ltDn{Gen3WdOcizjT^@~)7m#Gvvjm7;GR)MK4
z#9lBmba^N#gdhzQ@+d}+2K4(T+3Riqe0v6<mI|v{XWs|jk6}u7Qbm%cOtohbhNjx6
zmAegDP{b+epikWE;3!2JXM0<FqXZ{V^6!|QgUSn#8mn4^G(i=LW_yM-3UM8WEKUeU
zJ^bkfrtW?h8!M-&b}wKW7L|PmNQ(g1vdcbg!$jzpbXlTt1Zhe(9MkQ;-h<1xaRO`5
zQsXqIIQZQkq_}XF&0&v)$^wp^V_Skst)%0Sd5-P4M8hs|r$cJE$TUJ1cDXPbqFk_S
z7h&Wm3WVXKEfZO2n)McmRVu*^2Ln`=((pW%Km0L<XP@W7a}QFT-A_@il81df%LPMY
zDf60|VUZTZK|(cGbVdWllM+X7?#TfQ%5$e&e&=^}0uTzYOcqX@pxUZY9c)oE19Y&1
z>Fy<6LSVTbqShb@hV*-5@+iijhPkI?na4SCW+0V=U4G;qYD-Y{YXnKiVAx^l;4*S)
ziF-fx33d-G^ZbAMW%SYe7(_OQ57uZn8iEl^HA!*lB;$$3cMY^E2)kpB9z4J;$B%RP
z=#i2TH4f=^yF7pD6lc#}U_1`^&$2_-WtY3qFx8x5x`?SNhoXD2l=|i>tGB`s<=I^}
z9$}+#9J}Pn>)Koi=IO~QrjD35yX>#^a4d<yWS9*ZPI?@C=Xc?~^PL<yaVvV<=kXWM
zaca}XpFPSQhxT*f<O)`t@PK=X{zzcDE>0oYKR?Ic{=fVU?z-n496x>w%~lJg6vIKE
zlP6E{`7eBd&wcLmZ0~m7lq;^^iLOBii<z5_GhcD=b~b1lHC!`7WC4b4?g_V)yNgP5
zifG&;+}Wkqi_sFmz;qq75a>cv<N~GaJ(>ms!!b}<fp1wDUX7&47>xSNOxeig{T#dX
z1T&xf6yN-p|92`_<OknHnnWztOfVH9p3tx)$|&_GCKF8|K@^JJE4x%ZpQVL)X6NQ;
zwpwUil#;uHK9|?l84j;D$-I4wA6ihEo#n{8-bHn1gL+Y?V#esGSI#;i5Zc7BEWCPy
zc+w~CcIgKhDwCzcn(LNz{7jK$CWSIEwZIS(!vUR3JktVK6K4s-VTZ-#DT)KjoVeo-
zsvr0uU;XcYo|?DFaQZe3nR9U3Bg|uTIAXS<@xu;LtVjY7rbRNEkc<OvI((Qz2bY<h
zpQB!@lcq7jBw%fGgNrMx^m>E0&wubTfhR$8Zh^W`R0efwMuZ%#W0|GS_tg|um?mDO
zMiPt}O(qmsj$s((nFx@El1c0k7B&t!RVTw18pBZ72$bzI3dfAbeGcAy3;B^F9KPu$
ztP{8M`6nMn7;{{lEHT}zapK4n;cgd^OgLQ6$ksL)j0DxHjgl#h3+}(~K8_tb#^Ixf
znV+9WXw76YVSQ_Z=U+I<$&+W;+S>ZgErCKHl)@1fb<3tT-GuNZ>C!THl-CFxP1D5C
zA+DkDJO`x&yE{H>(J~#{B!!}F$1LhKmK$9hvp^XT#sO&>ap3S#{P#XUW2%L9;!e)b
z&2eUXM9)7)d;S*sdB&NQ%gl@u-VvH~0+VFQVmHcJI(&e?^0Dvbwma_R*wN$6%*<dr
z&K}K7kLRC%o<IG}XZY;rKF{jfmG6`XuR5{}60*QT<G3D{c1xt%BtTXCl9E43QAJsz
zb1W0r(MVaay4B=V@eaD)9HXqj_iaW^&1CpI$K5lyUQV88s640Hnq}^$TM!1!9z4wK
zbd6eM(OQ_MGS%Y9-M4f0)C#B8PqP|jT#O5{=QlB1IiGsRdpU6M5Y2Xrt*vzi!yYG2
z+`(V^%Rk1azVG{a<eT5%-~Idlz-K=HdD7%9>I2bxB=iEXEr(XMPHS-<j|u6{1sux+
z(<rIV^;NW$#CBb>$rx3nxQ>tI+Gqo!aly{8&2B!$Bo|c8h<T^STy>0Nl?WfpvIr-A
z#^XK<H{Xu`p8KgyHQ9IMI5W52&82TV$n$501jZEiyz@Bqs>R92&oR}o_~1;3;aJlT
z9TLYS%R=rteu#TN{wZ#|{WcCAIl}DReA$mu6buGE&YnHX*T4P{pZnbBID7W|H6fw5
z+5})?P_<m9S`8|OMb^D^jg?nxy(i16&_#}AITTrn&J%ps!E|i2P>lL;WjIGi%`nNK
zYDCPNyDZmxc&<PwjcJ)o1|yOrWZxb4;NSl~8dFVX4<2Ls_-(A8I?EH?0d{Kvee4z{
z!!c*htZ{JG=KV%Me`t^?n~rHS(K+Au@ei|b^KG2C^;QlaIYztP#xgCWlI-s8@Z#wg
z`O=rZ#AiP98O~i;d3zQ?y{EFMHI@=ow4hn5V9AtBNeokfDN5>VsZe_)j;3Yc7zK`{
zsdx@rX?AxV&W{h%cNYnjjc26Px*HtxS6OO>m?#X>U=oc`GGk%?LFx~@pGK>RTc2kB
zmRnf8uuG5qw2mGinwjU+6~&cx!JV@)$09)|vPhL+EfI+6DL!)dF^=DQJ4cTmXaD|X
zs?|DvrGjAyx}6RuUpUDhe)<piy+8OgS2i}^4iCQM$~}K7rO~>;^?eL2F><@qfgcQ^
zS|v|1iX<(;F<Rg|4p36_Z*2QKQ{2V2w@929B)Md9zC~o*OwxInWAznGN09Fcad?#m
zvy1ytMM`aYhK2o$Y^@ZGT^rXksm)B`*BU(k@ROX}4tOfhSl<{Bc2la2E<btidpUaS
zI8)Qp>~yzz@WC&0{P+p(x#vDU_5FX2kAC!H{P2Tc<6r*kU*_{)_+qKb_nk1UN(7_K
zaAb<>n1~F7V?k=5f&gV$<Z+0S5(9zj*U(WycO-Z!J;>&4lj&*Aor?{u$~3m+vG$yc
zywc;&eOpK=DGG%)ZPHN>StL}am#{sHdJ?g*wn35vC}py}a*pRtKF15)fPn-yhrIi~
zIc~n`PSBdur(fXnpZ^mce&lN$yXh$3{oSAB;~)D0KKhAIa?8y(ap;#1@bCVY|F|b_
z`^N0LI}lNFM<Arc76yh{N{JR)BcrH1&+t65Bt>N<U9pxL$FlIOQVd9F=yhz)Oir-w
z9${j-gjq_{x4EJN_Vr)n_U0<CTOhP1OJkHQs8pLQAHNl|UIU>yaOYk0p85vUc7>(=
z2RVAnGI^SD?Cx8*a;3|Qo9pDU;*3_jxIH2WZBj>Z+rmC11$mZ}XDM4-SMdEBhYua&
z_${~d@sEF!pZ=+z<ad7ixB1oI_$^j0t-i@obv-Nb^p<I0*bc%hFa+p0A<J`&YL(*Z
zgjFSpnPr&RW?5d+O0u(4<8*X48{PpX2qt;L^i+j`UoaS+=2m@%s+VGDL7K*BrK$K0
z7LVM7@LdecVE?VRGCKbvQ`I^P#}6{UG|kj>gIcrA%8M&p*onBD6`a=vUpceO#0qK6
zN#>3p2g4vq5;nKiS-rf%($X?F9lM!BM{eS~Kl~B?@{j%qzw*ERAN<zu{vMr9_igat
zp5{0?ZpB9w1!fkO&8njbGEUHrOCE+Lq){mhZQwaJLgy5+V13)=+4v5+?SrIRGa4t<
zn@yH)KhBx0OQ_K|S#0)85J;9G0LQJ+oSLP`Vrs6-+~Fg1S2r*aD3v2j#l7#jmHGWc
zE<g4V@$Q&!%9Ne;K1nCRaX0vjhfZ+(=3AMYTLj>DfB)A=)0lhjeGj*uxRVcl_#?dg
zJ@4idpZFer`G5SM_`;XIOrF0@=XP2lRe^!VC=y%>3|*jAf$&`jb5szPG!|)G`Vp|b
z3Wh5gMFyv{L#)qD;WsqLm+N@7Dt>*IaFFq0H^Aw4II=iK7K%KVC}EL}`{cb17T@uH
zjB1UFx`d@8&YZf4Tk+^@_1SviQNDiZ0_VaMIm!@f!~^d;%D%<JY+Sj*sZ-DL!VAyy
zna}(_)tbj$@3@!m`S_pXf%`wicYpLF>|0vopD)bwtH1GEq&L`6{YFfx0;LKJC2`_D
zItx)UD1GL8C3wiR5s5*O1PDMXa16z`Xfc^yV*f&mD2mzXPdLzQQLP%#XmBZA;6QK%
znM;baKp7^*c!HXYs4pBQD+IHhC#h66h({H!<PH~Je3}>5*BNRsMcMp)W_6Wc_{Tp_
z6i-N!gwbe16epZMwZdbMJ<Y=pKgdu0<X`9ByWYvq|J}cb(t>~Y@BhOa?nwU@B`k%~
z5~)&L*T%ML7&60fZB%MdjK+vGN9F~2nv~s@8cfT@b~1t-Mw6T;wwG9$K8R5>X;uwH
zV&gVj%q<_IdvTtVm%ha9vnyDRAiqihZde$G$!Kd0zdcKHVHwvmsF;#dXI3$6&A$D!
zxK*DtobcS~3#@e~JRc<Fr&5YyixVeK@V9>M<J|k6_oC5+(S+aqy<cbb@;UCh^Iksm
z!H;wM9q;7!yYAumO~-imefRN8|Lp(EqmMuS7FA@*?slaWN~Tzb#;sH^iUh;9P_Bav
z0(4QJFv^A^tuQp0mW}6C=!P&D6+GVA&!xE;{AR)a=?aDRuqt(?r<!!mH+b%Bz&(o>
zF&%M@MqL>ey2u%=u2P#>qOq`yWn@%r$*C7Ekfsss=_cLXfZerK)-Rmn*-^x1t`NN#
zLkNEC$KK5+KmO;bPc_N%gy)}shJW|({|9Uv?!EUteB{HQ;<nrG;of&Yz)d$D<IX#7
z<DdS2{souU*4_#SmR{7dB)8NB<T19+F@=I62ggEqE-FqymVr9oYnU>Hf$P^0PQ)aJ
z7sLHrY`3xOkfVp2OwYD3+-WSsVcWKO`jKtkXN_poOIT!HNF*iHcf5O<_R>)*Q?rPC
z!pt;mt?v+p0fiK7Tv}&*_Hiz*Z*nQiV4NY;m^<&fi@WZ8H!7Dr_uP|gZ*B6szxNw#
zY^<{Xz&_sl-VgA}PyR5s-*z`2`}oH=d~lh6_Rs$nzxwOH@d{bix5|`~D4By{;wEGC
zqzk5^h$BqLM&=GemNk0>6nTOnN}Ppj6vSDBPGg>_RvkmZrAs4fQ!%&RHjQN~isl@v
zD^tujI}}+_ikk@onIvSpn~1v=h26y3J;%XdNNUuXSS<#F3Fj_7$(dnLvIPr`7VzxZ
zv;4w8{`(|RK$0dzaY~vM2mz;0ukgsj5Ay?m?uYrYAN?7Q9J`5s{Ez-2c_I0=-}ucE
zAo4bvOjD3xWFhwOB3i~M!-hQ(HOuo+!a!jlwLzYS6j_ERd`7Cmc0WWJ1KJInovnyr
zZ-Zk;7U^^cWKqF*9Fs;dS}JVEMdTSe36M&WDHprH#!|4!<;Z1yVsd_GhtVFCSMEXY
zH@3Ff*xG*0caLJpXuL(Y^T+J&cKNG+^=J9;2R_2z`?;TEeeDXr_vuf+p(XIT0Ys@X
zfLM8onFJU*L8c)}NzAGbTA`)Hv<=cILl*@S5SE4OR|%D*A5YPnKZd_pLllyIM{Jri
zEoPSv683VUag%c|9%Xj8!M?c^S;$hrM+>x+OwOHV_TbI9wI<rIn2}rDy6-r0)aCr3
zPq!bkd**R2o`05=I3X1pmjZ;qZ?<^ysVDi`Ltmjc=+NtRdG7h=7>@@${nXQZ_3K~c
zLm&D$-~YXTkz>b>@!=1BkY=;V|Kq>^r+oR#U%i1L^-9wkl+4g543%Lf6AYcBix8Qo
zn7)gU1*#~^CK^>D(F9<Woz%8f&`)NVEFEEb-xNA6aO;}I1(*E?mI;SB!Jy9CnWHQX
zudp-|Aqz#3D+&x`mJ_aAMBn}%oJO0XknES2skE0Vx@&B7ZMJrLY@B(JlULT*RT5Kx
zp$oi<$9i{{fBCQe-*kFA3`awDcRQ?aYygsHo_&F@Kll|s@$v8HQ{VR&*uS{QU;N=8
zrc$l(Pk!m2a^b?oH|;|xVTn=}XsvLv2odxVh9Zm0OxdV-Xjy=1m8iCGjMk+@Z=u09
za<aTex3R?h@+?jP-EPbzjX8X*j_*4N2S$~HT<*QV&2ES)Br20AL!h#pd~*wxB^cFd
zWH8}?ewe^)lZ|~kqRM#V0xw)S&1#gQ1z05!D;GA_`N#j{7m4FBSyr+VPlA|2N>*35
zId$qBk3RNI{^DQw2|oGpAK>8Oqx|A8`~p%ce*HIp>vb`qz7Z?36bg+H5_`0U2sY79
z6`^wSBrRDTs#WqdM`NOZG>MQ|LB%kcXpdfyVXW_Rcwd_&D_FmrW3h|pSZsEOh(r=h
zLb5nTOO0*2P!tf1$&&=^8Zwx$Pmh^oHruh!`A*165K!!4F^W=<o}HZzJ3F0MusB~`
z*peh>JX&Ei{NEY&NBsDY|2N!r^9laJ&;Na{tX<)&4?g&o`tH>>zI2K4rHlARh=_+^
zQSRg{6O;nm!O$fNOA5g#49Mb?YQ?7FN%l3{sJ6|W?_Q?8(4sSpxN>=u-Qfz;`H(ak
zGagT9x2ExH4P+c(rV&{hkro0e9rDUNE3u*%_Xv~x4K=E24gG$^gI{};ILe5EfKPqm
z`}rGx{jc%JV~?@1we`9`>$(A?Jlxed@em`}LSrCRj8J=oe`rkKD-YXEyKIC*QRFFU
zm{1WFcC#j3&t$(cQ4*ebw2L%1IB~0HVXi?@NCL0T#?Yo+8<Yxcg)GrB7^uMzIT>P9
zr^$^v&c#QW63>xqkCD|T$Sf{zoZ@^Gk?Qh5KS7~@haP>DN1k{TB{LAB#OrH;(wd!}
zkoAoL=gwW^^ywG)nV<ey?tRC7yz`!S@lXD{|Bm5c#8Xc{{kk3aEq<a>m_-{n$p~w_
ziWCNhDG*X3EQ10td>4U%uxw-!Ayp|09ELF!tIAehXFHKB^kaP6;F-sUq-=2XXvU!f
z(}Y3B$gHtB6tt@mGA}uxgt@0rIU14&V+^lGR@ite-(X5vXi;HJwAmUaoZY?5jw;Wy
z(3%`bqnJ<s$)AAMrS5uJ(L}3~o~74I*xu}OX=Rm_OBea+pZc2|KYBBN{`-G`C`$N8
z|M<V7_qtJB-`ZrJ@0C&&&UhUr>nNv+kmc@I$^z4IF<g&qEYYS-k!GkoD^=VuD3s6k
zC_`M_=HNn$Y?5$kb<CBOf_v{dMz0go%`*By*^HXSIY?R3n9Cd^7?Xz+46ja_DyG8I
z_);^>JWiz*&TbFbNRtxutht7;+U@kNF~PoC+bD`9<4K=>?+JR{A^rY<ANs)`<?xZC
z{Ng|Ohiq<c@ujbP<xTtV<rZ$NGg)21TAan5bP%M-J>^Xy1lqI^mW?PRm?pW>41)lf
z=J<XS*DKiHo@Up9`J0+NaP$b7<uV$@yzty<hNCSG%8)3W5KKbaQ&af$CNi2}rU`ix
zB85p93o7%|yx4PCr(hIxNsDiNc5%IBQ3GkFxO8co-~atjGns_^_5bE?aQ7YW;IID7
z&+zoK&%Q1u*&DU-+7(v*^pCjdfxAI!497y7X4#98OTzw$EKSi$B7{YfC?=zXQ5tce
z(dCetb3SU(3nZJTH+k?&=MlnTUt^XT=Se1$T})xHw6sK_H6j=Rk|c`v&R=XgRJf83
z=y&=z)bOh{3<e>OJn{s2kx_N4yyxBb^Zxfez;FJ}@4oKOx&9dlCKLtEcpW2LM_V3O
z=@`;UgyH(7enk|c6eR&&GC>su6bip$ljRP<Bw?-C<=BBXH_h2>=Qf?q2@StOr_*P5
zJS36@Q5cfNF+yrQzY2MdmBz?$f~e0?;GmO~{cXW8&RFk9Tp8u8j>cr7^oG_i@m#%}
z4OVbfVXOR00{J*6js|4OgQRi6U;2@s<ij8QDDS!N-TZ@}|9O7y@BX)J?d-ncALW{9
zjmS%1SmRx+a0_jkDAPuxNGB<#<zV?G<+GFuZJ5MKhRQOK8qc>;S=sxsGZ}N+(ODjt
zZPOKYDPrUq?Dn?kj{D>yCkg|SC@yLAYfW?#W27-M3c&P83lm?%Y};pb06UWr+ub3%
zNxtVD{Zi+L2D!HozOoolK37Q1cwBJt{4Sp<K1&>B{OFJV6z_TGef;nb{t%~5p5)*D
z&;RaCgA=074bn~rMUEJC5IRO>MTzngprpd8R7xq-y&#wr1z|XbTvMsGFb&OIt3|(R
z;w`#-=)P%;W|O$E$g+rMPM#&0L>MYzJQ<Tj3GJyS)kd2l3NZ7OA_>u^Ntg(hs#OYi
ziq+kMfzBAk3HjCN@vVl#b)HiS#-ofEp1(km|0?5Az>odtPjLMBasK9C`)fS=?6YqG
zO8xCE+}`2hpZ*cIe(%Srt5U<;aBPqo(=y1yglIS-&$1HaVLGI#Bpk);PE3v-?Q=i}
zL>yo+DhRdU;V-|y#j`!Cb)UnHD(z^SUcZZDSu8CrqXek&5L@NMVTdd;lr)HSg$+Zn
z(b*w>6Qq85uDG(c!=HZkvq+iq_x{%3=R+U-F5Y$Ty?p7bZ@!Z7Elok^iKp<*6yc58
z){?q!<H?*NO(~MROk|WKD++X>h^5VVTyR7WSsFdW#JiuBXLI_441yZ_rX3c$Pcma)
zAjvZJ@7sszScE}{IUb@-6EBYOtOD&?>=cr%tqw_EkiEeGVwwg*lu2eBrabq|N&fd=
z`;Y8j+RukS^kIJg5C5oS!gw8aojwpvF&UFj#@OQ?m<p{@v=SJ$RjR#N7Fv{oNy0QK
zfN>m?Cow`=*uJ7+OAgM@GBOetj<+~^sEyTXF&u<!Y;LhNI0IRMBN8U#F<}rh)1IO}
zy+9UCFpCToM<|(-E0ZXeEbW`+{17%QMG$S0=HFiPSYPK6Kv5_*HwS$E>kku#32B<}
ziI0Ag@A=rr`QA@_lK=Hre&uyJ@Omb*jQ;79q|3|bQLj9#M1U-?Y#X7=J`oKF%fRwn
zvOFb9GvZ)^DhzOPT5iPQnK^cJ$n=p0Cr&IOEsL$qJ}Vd3=!`Do$^zF2>Gua@Nx{M8
zefZ5b>7<;vMG{hY7Fl7CW`e|+;#4N-kPu8Fq`L90{WOzYTG{6F*%t_c5G@Ptzwdqg
z)KB~bU;4^dIQ8P08}^|v40<n~;__F&z}!7|Qj;mdG!RB9iX<$rq==V_LMcqw#;w%|
zqJTKcNurcdYO`-?z#&hQTGJ?DFmrsFPH)J`XHT)SIYMcJStG+&Wtq-4Ef$ydAzY7i
z*vHldSrn2N8F?X4%3&+_*f2A8x;x}acEdbZ8VDvSr%s+HNq!TVr~LJw`B^^n!SCX`
zKJb2i@6&&H!~T2CwX{L+*%#Po_&Dv1q?q8<D%dhdq#5CGLKZ~GEC)l7>arY~6$aa>
z<cORwA3Vm`sIhgq&i49w&YuqOy(tzOHdEn~w4HULNyee&{aCI|7$-PMOg0{%lN5uD
zs&6vZ20Ky6pgSPbH%7{77^UK*QkvcE9$)#&*O;H4<7fZof6Irz`@8x2H@@+<0TN6T
z(lBG~sTWy%$1LN1AEV(?pK0Q^nq}!K%P7)>a1vlxn$09(*v}9$Vb-~f?)TXq9HK#;
zgDPRFbA>r?6Yb|zoHkOLtgR2I9-J__aFO=@{p6!DA`OU4#hINhy(niQ42qXE{TjmH
z=#j$!Twc3U%3Zw7D!#|e^c0q96NUlfU_ub2Jo@<K{ORZai2Lq+A9J%atiMiF;A>&L
zQnLNjBc%Ih@X;lSon@mbG316}Ad`$_5|ZZ`mgnHA3bGJHK~5B>WKP8F%nnP2NmxwL
z9~2xu;q&lAr+DV^%Ve2mVeSBydj>D>(CPN@9EYW)B@EXk9`&&)D3XLcjX)_>VG&A?
zZCkP3+hN=vy_qoplv(!9_JGG8eS$1cFfsVPPyPTu{6jy;r~mkm>Gp0+j7DZTm%j1^
zjxR4TBLrF-SeA`84Md?Z1>`}@q|+ly)3Tq&awrOeB$R|nMq(x$+_%GGF=k>evbGX1
zH8<g*ub$(Dr?#jzY8+WUj<x+wY}KXLA28KwFgrg-fq@zguw_9OMr1e`Xao*pJcdrf
z_UaD#TYy-}bH$Y_yL|oY4}#YCw$B}R+{wp2`q4KPKoF2;8S7vEGL|Q(8JZ|g@S9di
z!BR+~(S)EgEaiAnSdK%S=A@}6i#4M(V{tKLE_#89zr^;s=Akd2=G60D)-HzZKe)ir
z8I!7b3`^`1rU{3a_u*G6L}`XygvdA|4pU^F;5!;&IE-zVt=<mdV2tEV;x2?BNpdb;
zT;mV__>X8crugX}{~12@z2D2{zxXAB@U5ZUHVu$z!Ro`$Fn!M)!z4uITO3(z;aUcb
zRvlxiS<bm6B?==fqsC5R(;23SJf-b#aCr9*F^v<9-39Kfft7ADzkY@Vca2=+G%GXs
zUX@;dh%<@VS-C{T*61KY#uFSzvsPqW-s%!-laSY*_0%erQs*Ok?f4V|s?`b=&qpam
zoW$f=&hAd1&wu%|+<oUexbNO~(QY-_eH&1LP%`aGVbB|A#A(j)<td6hM|d8FS9a{%
z^%|-uNTfuD6Iw<}q!r`gnD$hQ1v{h}Jx!V`gs;GaQJAvZO9>K*&O0<(K5jT+wA-b!
z{{Uthlf*G=QO4y?Ml2vu@)hH6X>o~v@h|@sbF(x2-M{zuc=WNyuKk;3S{&GakXvs(
zfn&R@tzG8A%6UeU3BBHkuYUCl+<NovEG_MO69GhPI!`>o#@rNkD}iJ{z1=90R%uEe
z#RUB^$s|M-WxZXQCP|zl3z#H=BFH!}*Jrl>7}4y9*j$Yn4Nmjy;~lcZ=kU>a=CdJn
zc@oR$5zB&O2M^%74q=+(hGXP-LKuXQrZ`qX#kH9Ti(wwp+3ujVCVvxC_o~i^1_px(
z&pmsR-~63lW#8gH?zrs^-g)mmeC`Whyy4GIu57aU`OjnBu}D7N#jjPcO#{0~k#Rya
zj>zK#nHME9E`lsh0fTW`cGe#^M$8VrLD>2*TipuZ{K92c&SX^0c@EFCXzyNR);xt!
z16<c;|Gxc5t%%YTC&`%fy2vmhw|y$E!m(|3fURzia2$~F>JruKH^V3!{DdgGZg+PE
zJo3mR%+JoUf8Tz->-`_#KmE$DFd9#8@Cs!QO}snc!j~V!Ki*(#+(%dz3pE=@mn@{N
zZDZ9c6nRFGWf*Bf)wS4}7<593AycOGS)z@CtacBhv_)OjII+LNoE6Z%^f*i2MWQHX
zer|@T_AEgZVI?Wy_BPqZHZyTd7Dv=vhz-G7Fs8H9C(&>tgb+f2ZChBTwZ|4M$x_Mc
z>IR?wli%aS@e|y4@4MKyZ;7+#&fl>AUT&q#x?nQS7!4zCJJ?_{=waCgm8mJTVSpiU
z>Q#y?C*9pa4`V8(q^cER7&1HCW?l@ah7U0sn@9=1o#VR&iIR*`n<Uj-yu5{QYz~|V
z=xuEx=Vq~@h;%$;A|<DHM)aeMu@J<s+Py8z&-07_@E;*%!N30x{|jf%oqff>*Qzy+
z9X-n2>>Npyu(q*Ar@Kp@7hJx)!56;x8Gh;~{s!~&^KV-Kk!T_%7)!-gx5wf6C8~bO
z1t@GA%Pr|iw3HZ@i{p6AG%`e%<5*5f>C~uUcrHmlpwk`G9mXV4hGz?AYc9>Yk7-B@
zmC;_Dr*YsQ%5oSME|(@{urY*5duopEu!ocd0(j>=@8Clp{2*DL;(P8ZzPInW96E52
z+i$%CAp}{Pvbnj*coL8%8E4L3WNmGgsdnpi`{Fu!AO>lWaOt5ZXx=?XCm1k#;R;9R
z>ogpVVHwowKJ{vaJdKfgP7<dy9E*|ixww;aQ<7qnGAExST^%87catP@9AB75MG5V}
zX{N)+S!`@G3R3nh&d_R3F%CzVNk+7}!{p)`^LNLnAjHxcwr{YRXIxs}B+X%rC=nZP
z(1ZZjwQ(#Pr8GrRAeCg)4|(R<XZhVv|0aL;Xa52Bzvq5F_k}OsP<cWKFfiFXy-LfD
zv0Db`&RyitT#fmrk8K%D&rOv-ViF_Mj5JBHEX~+4S=mWA<QAAJXVzLpZvH-4w-QyC
zxwT=Sf)S0&kFg}q(yWge1+Xx)gk{?F24gBwMrUgWz0qMY%?U>%Y@Lxif|c=vt(^g>
za0qnrnh<On2DW99X4$t4&#r4zt@s$26h%&wr6{S`-5v1N2fxZY?!1>fZo7vgM-KDM
zv(LX_{{bccmnc~ax~sdS3kK6wkM7Pcc4Ll4a|*+;5r!z6x{{PUOAx~`PUeu0Bce%6
zktX=fI?M75mBBjQPMb*Dh&-lk_nEE*2&+JdDy~!G@@AJ|lCyM2z+h#Ss!R#vm|)Pu
zkO`ZX%lXY7Atr$^$@D8ju;Vzq_rCj(Qu6!@FE9>XOG#d>R+yig#dj+taYBFCCyFAH
zDC5F~75?ONpXP7>ji2Y(v14yp01-g04JH!ScgNg*xW@e4EQ<2X(5+T*JP*S%5OIp@
zS4rcD*-A`SF-fBWtt5V{fm6?Eq#1FNGMwa`+lc7*BV>_NaV=)rKFzj+RYd4Kqc%5(
z;W!kzU@I|L?*-(0C0xUFki|=7miOI%KR@}?KaF7uzV`L6y&`}ZLa;Emz#X^V$^QNO
z>2|xMS;k~CA<rjFCJ|3P^CTbr$S0YbodZt2CBA#L<r*eRB1}P&hHUkB**~`r13}QA
zV44PwW0Hj-(I~?4tGK>Lb57v9K5-b(+a7RkJ0^}2<{K7sZD=2AQ}rt(Nl2Eam<SrJ
zCI`Ro$5`)m$fA%n?Xr<Ul;)(S!Z%H>?tyK~;@*4iq26rp#1oIQy0-e#?=Qgjd>YLL
z^+p|mf#>;q=OF^A1cPD7+S(ev=e=%UTpz}(O(wm7ao^+k{x&*V!!;^Y+f&${2WCmt
z?q(Hak;8aQVJdvnLYQUqUQwjfDmI7oS(@W33<nL8!oZXfO=n1}nNbLZ>(&s4&)G{m
z3?~r>ZixveyT~x2krm{_J}R0J8k#e`5rauWWZ8uIYY%xA=I8mIkA0lZZih!7f0W+K
zw7H>V!EDs3EG{lktJcYqjBdZna6F_aB;D?qhraoBKJtN&a{Sm)DwWC&X^99x3qh7E
zcD6^HIIxeyGY+09aqAV_YOO?02tlRlqY8-}4$wt{nG2>Ig-mmjG^E+~S<qeT@k5My
z7D)lFkyEc^I1WU)LhA}UeaV^L1_hd<_XZ5Mx3J@kAjyc+1et_%l;Z5pfKUk{0m*fO
z<HtYxalY^SKE*e``Az=lGk?PPRcIF5wpdzNVrgj+$8s4A`fP4*G6^PRnPh!qo6mjW
zkGc7#TRD9A$c+aO%2I*CASx8=I|0Yvu|UI~0iZApEW^Yw4Q$WHsnuaJL5;_xS;~SH
zBMs>HC(O=H;Ws?yoC&S8Pa!p?q0y#b9OZO6DZ_D25_d7QW@f&QNE620K4z;<ra>kl
zK`vQ|Q+5U+DS}AJ*HBL0eA99M`M>yA%+1g7kALwW@oT^F8?X3x$F@1J?;y9_ax;xu
zlU}dK`HSb++}<QhGe*OZ#~*)$iyuA1f#u~J_T8(k&@fhtSPSAjX0yM;f!Rf>wR%Z8
zWSAsribxbm5D`vNlr*W;Y8*VEs8(y`c~cTI&NZuBA%o5cU8GpH<j`V;R-=Vu3yM4?
zNdi1eaMQ;>fjKoprZofQaw&J{gcI^YQmr=dU5Cx>bpTx3=KT-6k7ldM`EzHva^;n(
zgSKrm)o!z}u)xB?0!k^Wl}gFE2NaoRW4+7f#s>9T?Je`&t4(Wx)X+4*HX%qN9zS!L
z-mu`d12Zh|n<k4Akczo|b6}ekNs4LPXsy{=4|)3h4ylHl4mLToFpKY+7^X>{7v#Z^
zR$~g=wQ*dR=CRxP#tXlWYg>etO&nz?rATDKaCh@sqNjl~XV3DdpZOD>c>ED|UI7FW
z2%(9Sm@o=)YzHa#dY+_0YeAeAeB+VFaBLsfab6dh<@Eyyf=HRHZcn)VaEqG`FCe9+
z&>%DjC9u3wBt;elVUi%VLZ%_NRC1EYW;7g8ZMN{6Hl`cV%159}hU-iVdXt=;-H5?B
zBg;ZO6Xs`qTv4#w>0w9%WqYVBBgt|$Q-ihcm_!Q_m6MjIGS?K%1n{wse2jnn%m0R8
zGUD(2oxjcRzv>Xuwk-DVTjs=xTbXK4F&<4gf8i`yo)d;4Sz55Rw#I`GeTfI&_c2<{
z<_(>LXrR!7M4N2%6Gr2JeM|G?X<o8r#tDw=;`lCxW21`#J4vyf31%E~tgaZ0OoCBF
zdj|Yg1Iv%8Wf@8<q|yvySiLf)+l@(5iDSWR(_`OU4PR%({UOzQ1EW%bU_@BJTA=9m
z1C)SF6-aTFF}d_%yV>L?fAS~!k-z+xc-On%#gk7z`I-RIsx>)&>}GB{b`z#)vbMHH
zk|ZQ)Mw+DrlaQyMeUeM7=k~I|8?-$gk638}37z4De&93TXrlMNw_%xOce-JqltLIL
zj#owI1+y)QPIDZ~1+8%_9&Wuto@NY&5u2ScTb+Q(IK@KJtlR8gs?w;M<T^(r6Y340
z`C~`0YBkU{o!nt<7!f4}hG9^x)=_CfmX*36rVxDK{s*}0?z_+^JlB5BZB_jW2lg#<
z{H9xI)a$ISuClwkL#NjzP0MDvORHBn{o={DbP<$V5G##k2wVe(;h4JTGnga{CK=7D
z$LwqqRiq^2fNHghGuKARqNKt%Oybdyv!^d}b~DDda&9@;W~M!j<7j-pNuH&Q`dwVd
zXKtp3U#l_o13$sjPd`SowN6)9^bjOzL6#Q;lOe^+HfX-*F*i3)k>{K`{Q_4u*I#iD
zwq@ZsHrgn;mMqIEOGT8)q0)j$nDFA6v!E2FVZ1H$CQ@l^1XW5dyWOGW;qzVYTduJ<
zQ^oacEVqX3IOJ(cGU($t9>NeLNs3pmaOjr(RKM%{IC%SA<SVDq7oNjPCrA}jt2WTS
z%lYA$MsCm&Cd1)~O0|MvJ7jT+)EbR`MU@~g3V!Q%f0xgF;d2yH5=AcwN@=Z$qKNIC
zO*Xf-@I9AFFe1%jq|DK1iUNB55f8lQR<?GxIrlpHxvm>P5)EM?83#E#gANB~=1R_b
z!$4RDZl!{0T4=41g~TwOQkVxYRK#GU@!bk2iC1sn)_me9Ww$e-(+}wMBa$$|v0$cZ
zGqtZyy=GCQ5h9*YsXDZ8K8iavg;ttzZnK`4^hOi37I?Oc(`eD@>|FcKtwx=XedHrF
zTMe=-CcBOrm+yHj&M$K0&=IDmr`g)tVt02N*YODUs4#}ZjAx&Fo_F5&0jjm?4F`}&
zX|z-z72AVdrW^ZkeGkX=Fnt%xa{xh}X5?vB3d1N1+xD250@v2m>kV`+u{{G}SLqE6
z);ER>`(q|?f`ewZ>CkReY1BLf8DSKl<B-Lhk1#boh4QP2C}t=3xDt&SMJW=Ksn!g2
z%VcZ&rAd1H=uz%{=Q{z|-r8jNDk)Jz7&Mzr78e#-US39P&3HV<_k0Xtpp_y^CEHtl
z&YwR|ZEpd*QIt{!jFm<j8iZuKzk}~Ow3;n!&n3?#@nBqP&#RnZFeVKPgtn;F8@N@d
zcr~=vOuAkA8C=|s8TZH3Ek(^&9607N-E3hPkQXUQ6d^P;JeS!Y`70!<U=l?H+Tt=k
zThWL(%TcyPvpq{O>Rt;V70>6r_rDj%by!=w%-Qp2Uvp2s@6l-1X*BB8sx@l08n$g;
z(}<7-OvV{2D=V~SmM{zx^%m$llt67kLkJuJwrTSG<t}G8V(#AG=Frj<rcq{iQwwuA
zo<|&o2q|zppF&GA(_*YNX3OKqZ6}y&R;kGx&k|@|vQ1a2bsF_HiOMNV3(N7?Twf;+
z`bgU*jdOG%QE0Lvd&LoY5XJlt|K^u@<nc#1cm6Dc{wtFgr6iMJ%wRB}QEw200co00
z<oUJk^E4~?>es))!u$fwW|RJ_st|953CpBnLaqd(STN}4RE163Gsp{x$`yr5$tD3R
z&+sZ0tV$J$rtZ0AcJF3bM~~p%dKax*-$D1$MWRRk7*8o=k@CdmnDt5%{lrtOTmO+X
zNoh{aFg-Vm_U)2F>N@>1g)Ha|hS&YxNt!Si^w`<mpiys?rN1n{mP4AR!Sr;ScDu>!
z%*>4kkWeXF78r)6GwM>WcuY0hxb-T9)J(=<*~BFTGD(^ACuC8IX_z#oYSgA}D)l;9
z923VWo8yc#8zDsyF<pU~hGu%XMyt`nFw4%;C<-x@q#_hIf9i)2ro~pCBZS2|9vg+h
zILXN*IMv$T_J0?t%Ef7Eevx<Fa}NNQFRyap;)U1Tm*;sj8V%~rI@Ma0TCIlTmIqBG
zB{!rAY;Nun>@AEpYLPDcdD~XWPP#Mfur#|5Lu)3333*nOThzT;Y1p4&J06~6)0l!;
z!@{*3k|<!OH{|?I!df?B##ij0@mXH*Xx3}gYAxb8Bn&3VqM%-Jse29we()z41QXV`
zH$WS#n;tufB+eum5@{vK;+1+8x7>US`}gk$;MD0;ubF(dW#aoDp6_8h<)Y_#9vB9~
z5G9Y26og^U+S(ekbNjK)>-Va@9+L=CrErD7Hw9^t(Hr;a4MR428B-O9!w2TD96>x9
zV<^G=@*;*~Qe@e+%sJ@yIDLMblUGU{{fQ$DmKT;VZM|2ot}^WHlBEfaMjPLCv2B~#
zcYlzlo_&&ybI*faXJlF=af(usI4c-#tiD1u&vqPy5R8UH&Yn5V%IfMX?x#FoGRwtN
zmb*zI%I3_gTG>iNR!B~sevzAx8?>8EdV~I(X4uhMW1{hdKq`|-q<G>|#Iy}JE!MG3
zf$e(K+CCaV8cfh>f>)`b4Fi=IRI62PIlhm>_k4uoKk(xiS;*kAuaQ3c5MB{uJ1(Ye
zvk^p`3j*rBfO;IGG{m{4R%?*$1rH@Oug)t9S@7r+kMrmgk6-s5vOK5T?XbSShGkkr
zVL%iH<at`c(X~bkL9N!{f%_jo3wY?^N8Y68UTdvG`)QgUy-PdfLxoic4par>Q9^Gr
z!1r8=AV6gq^_gj`MjgX4F%6UMB<0D~5l>vajJ5IadH*El*v+@nIB_?Zp7{p*(wIyc
zT*^$|`Imo`7yj^%_>+J2FF5f7Kfv5Q?_e~UU{+j`LGUusx7TM<l|g#-@+!9Nv$MTT
z5Jh`BL*<Sx%~HPh;Da=qQ`D+8g7D9l6}}D|FXUR|8sM8Cm83rz5=1#0C)e=}!NH|A
zjaHQ;iYVfgR=Y{%$br2^A&D?ekc!^c7SEmQaCtYSS(V&+WQzHjSqw{JS{9C7Vc6@C
zCJD`Yn~LjT3WKSmC)iFUXaDrKIduGX0>>kW;<9F^6v1S`&Sdb4^9jpw7>tIDM<bql
z{@K^G;~R#7Wt&J{kmcF6jHS$uOZo(*1xmwiug_zTJ&tcXIJWbKGiVLuO5+Ma#n70h
zNth&@y0p%vt(du*&mA`{Vw*O}WQ-^j^ZVv;8a1*cLCMlTD8+6Z^6W~FongxIyv<Dq
zXPIiWF)cs{49g)M_OMNx>DDZkZD0b;`6X6|=Q#KLGaSD49%9=j3(HQ{0?l|bcntz;
zb8DL~f9Z=%&$M~?k%!6hSH40R0;MH!5)r2{wqaibA;?1R$(}+k6^}jn3<sBoG#iaK
z<hyIvdfB6fVPNV4*RomJ4tRQXhnuG>+;QXxUd2HcInCJ_Jl88xXyYNKHgKv{a#@g@
zCPvkxHYZrV<7S$3Gw3A1*9E3g>Pj1?L!(u}uT%-L6wk6Tgvp@OVRdzd<(uC@CV{jl
zO9xuN2E;L%jQM~5(l7BlzxO+g#zQ*0uiU*Vtw_>@Ucbk1FhnT9Bp8#X33*-;KLSV%
zr_NkNN{hvX1$u)%uUT#1pqai$)v_pZ%_uQQb~DV(VX39)k0RRBiXu)($75ugVj6<#
z%ru7Uq6O5e9!Zp-!<f<T8f)t-%q}j|`oQ<GktOKoAHwy0CX*2lcT$X-Ze``kXIR<z
zhuDitv=$bq)EoFyb<}8#Owv~sV7|Rcsfgl`(PY4AG9-x-!Z=30Y=>r=CY7qk&dwH&
z<6@a6#aqmc<u5F7;GTCe`mH}?GU#!rkuy`1$igOzLc-C2dbLVreh#}?$FNP}DCgPD
zkduMO$)3lv+cDG07OkgWpwsDc|NB0`K-w(iivBoZ&=1+VvPt3E+%aEc$t^(yO66o}
zQF6q+!SAiKW_P#C?%uf6dkxrEO;c7G6pD7cO?#?+BLSrRz`8K7$gxd}-YDVGl`ftV
zeDAvs)1Gc1ivq9W<4sRtIWBP&l7#`T>tQ-Jah_4Q4!9QW10KsaAHi={p~$J4AcUfj
z1;R9GHm9i6>cn~3EN0pcNw3Gs`7<mWI*zhzvLJ;$U1p`>)$MQRRxa{)f9~fveB>Z2
zmo5<muLM|Xt;o`p!Enf6Fu*o#f?!fk-Xgz7L<rCuO!&&zA1X@>)yf<EnaW>6A*dQA
zwk|M)U^prG`ne&~j=_75&(dgC5JKQiP2qSR0+{ss6mdkQUM{*poPg^vzu3S$I?cZO
z-h<`Z<ed$?bb>`03~k4!;(1_OY)28E;V^4Dq*2Jm)&>V|ySL;wigVPfy7pggU;V~|
zoIZ1!dcAgCKAfT`m`uiWx*euxW^k<1hgq7E7uhxcNd@bhJIv26a^%P%);2e9sEk#h
zu?09lAuXar^4wZVkt%Lktl>E}T4}1y7N%t*WkEKc;2>}s^|DSZp;D=E<iH&558TP|
zpZGbd^*WOar|Ez7kMOoGU|JTQ?_rt-=XM8NOfwp@Lz;022vTWKtu-(#lR_($C`&f4
zHmMZnE?(f=b#}i{#(*#g+1=S>eSL;FiW!ZDByn_2Tm7n^>h%Wqz4M*qs^G%P#TzaI
zUbV*Z5_5A)^qxLL5ERri#Q|U72}OT6WG=2#D=Nq^#xxB4#W{?66;qfX3^oQKPhFny
z@Qdq=)CJzHs@!w`2RL~281~`Y7(Dwh`%J|s${9K}zUNav#1p^u+noCBpK|D9ALZ6}
z+=XGg__Ydgo|D{wsFntjtYBkflWuoNo)>gF9g_4VE-2G9k+R_7M;_tmv74EjTVQi*
z<4yVQ)n?l^QzwqI_`uz4fAvuuOX7G1GA<xf^u{rf;m~kW>S(U?#yr-u7&dR_iPh7b
zTse=7n^dhsTt3_9_=yutIvp5vvC){i;Fem>6QBD7HU|OkvIIvLr*LXDWH6@c+W3Z`
zP;a~mv)paG*7t^C(5O|2;t6vzbM$(<j7Q(T*Y3KS=&msE3@8elESJ2np3oUdzVAen
zsiu$23;bFY+w)3Znj}KSA-?NkRjQ;#h7cyLMx9%3Hi_^37`Oh=U%}KllT*(SJo-f{
z{S6G$!f|~Z$6{?b;oNA1>CDo|6$njXIGDDJSE-N{Ickq^?Rve_aUAo+Gf(rxGuP*_
z(i*}jq|@1EV{@kDtO+JWVL+Cpdn{Ue&NNMq9XrO|cizcEk38~*kyskA(fEcYEllD}
zak*O%#)9`Qdiah_k(Ko1w&S9tB2N=+G}Wo8vI{!PsMad@o{MF>EWhWIocK$B6GJE_
zD;L@P;_u>JdJfmJaeR+LXr5i`aak!GV+Hrp1(GO4xHT%(DwSH5428Cgl1<<`w^S4b
z8{6CPS`ts7%mByZA=^6})a!L>l`7-Oh$Ig8>hv!yfb;Wn+<n)bOrntfm%nlY3!uPq
zJZg?fkxNFY0lkzgvN+h(jK(ox64Q1ZqQL-_WymbUb{+hgX|!cx3PHVAB}-B!aZ1$L
zVCD2vv=1C({?N@-zWWF1{mFk~dep`CJUW9h-xxys);n2!;wjdC@&88Uz#-~Wv$R@G
zTC-D#nvV(+RI0d<7Q!^oh|`#HFu?cAv$rHouTil;xzqD~hfZ&|1P1I0V!atNt5pts
z_#+gPgu%(Pv~8GgC78lMD3fF|!Sywbb_=)J!mN0hhQV->@YH(Di&2#;d7H;a1u~Pg
z&Rizvn`A{!)ANaP11SnN`yrEI!qbmD${<U)b$OcQx`QboNkU|yP+7rCBtvh^qA20Y
zmCbLtA=8>t8L?KyaU5FB=35>>*4Eeg%vZlikw%Pb4rg}?+DJSK+$>--4k>hp<#v^d
zrFe2Hf_-=3$5(jx)RXAKWwx>pW&2!M>r;RF1-|)}FH>1R#|1|drwRQyWhIE|qy=G7
zkUevT+Q{J2bI*}v8Nv`)d&%^hEpGx?rb)G0VKC^GpqQj!Jo)xLT(9i|K~pg#3Pq5c
z><lydeUp6^7o8UjgO~-cLKa0NlL;!zP_m%fsN>cf<XYi66{@C58byS8K^*lN47RB@
z+f2XfgCwRy|4YBYOd8<%KBHjF*VZGf<2TbkbAj$Z`zJIGAEnlw!*(3z7UmFD541o=
zaS3X^fs8E6@^TA2?o+E)iQ<SPiBVFP;#hm1>3J@;T`Hg1H&DV7hM-ol(MTq_#m=x`
zG7>CT9CTi=-5+wSQKN_w!hRo}BnY5BKa1g4P(tH1st7F@^v7(ELUtd0kljj?g+n)S
z;J^{;4}6l{?k2P29bCV{m7Ok6rw+$I_<k;Y?d$wcf9L<o!oBZge*ZFl#b<7D9;;DB
zjiVc8deuOdOY$NkjK>&;K@f(dSzbzb?)k|u3=9J<tXv?9)AF3;jb2M@rjH+Gx;e$h
zGiM2+0;M$b7EF7B@g!z2=~3~0bQDv>F|OxPnVE))hb2seVbB>xJbQV_Bj+~Q4PRt`
zH{rW~=*NkMefA$X!urf%qTV)j$6;$2lDC)nzz_ZiU+ny^y!g4#v;PC{=ble}ih83)
zd#X*UBymwt+~Bm88YV%^>gr`SH+Kkvklo!~(!6*ni#7~w+veHlo}*T4a^%nvlKjoF
z2AGCPWnq?u58TUSIK&$4U^+RDQ35^ExZp~b(jca;3~EutN@v95eTQQIy*#n<JWro{
zo-A^iuiV1!n!~}vCeu^1>^%Q8%`C=nHFZmKXDjC$U;ZpRVZz<I;DPxT{`54LE?&ep
zG!@Grmzv}T7r$>c9Sy@^s$FL|>@hbpOMlR1_hmqvx5`eRJImP%=lA%eN+6NgEQv>z
z&XKw-H`)ELEW2y4#!u`qOKSv@t@sq%$!X-K<dJh=p!fQoN>h|2Ck<$&dHmB~#xTDI
zQX*A}#C$6P$5qRVg7u9Zv{E3(uUHTNEW1u4p}=iaNHd62h3-pQ!eX%sqj5oh*kyXU
ziHHiaAV5e-ZF&l;)kGNvu2t?zb_M}YoZscy%NtA{zsRxOfOmcH<Lp~rrgiKVcAq&&
zbm<vtu1jw_0OLR$zhmS#_@N*937-1&AM(g={VuIrZ{hCm|9%?HI@2>#I2Dg%6p*|O
zcKSw5X^4}I?cE(FlaM6N*xcD6Oww2EG91TcduN-oXU>9Ymz+y)z%0jM?w;G>{08AD
zgdx;X%vB4D+$5Nc7{^VjMoAKB2$TB!4EA&z0eFVZBuzQJIpK-RLtePNMW|M|L;8H+
z6IIeM;Ly>VnKTxu_cp0I7TbdeQD5dgAN~aKW|uR6{71y?CP(kMje4z0qgf|Yk|@c@
z-o(hu3d!c?4*mWJAuPf$Vlo+16tAp98-~f*bLVgzpZVGOH@*YW8h2)jgCG7l!`%Uc
z)pcqnEH)D=4iveAOjE*PpStVOSXjVrH84CE(=>?VjAz!yJU{m6+xvNJbeZuaBV6gQ
z@5L86ckV?#{;6p~!=;*Pe4TTp6B5=&oO=Egg=2EZ!D;5ICXRqCiBMXSgb7-zH(Y<d
zYD#H3ogUq8?_1|Ux<q8vYE`t>?BBo4%H@?eRmR%sbokRReIB85;+D^FyWr%uVyc>B
z6bcb(E{+B?++7Y#RnR))(Tx!Q;61d=%Y5_X<0PqK+MUDpYwV6vo;h`iNdF<-7oOpu
z8si#@D9-658CL>HzfgQ5N#NOYw2KD4wN*y_0n#!tRe_-e@-18j9ml3xsnGA2km$V7
z42Ew#3aexfc-07aRb@zOjeOahD2n`LFI6S@V(;}@l>y_=N=ouVLh;s~<-FXq)-UO(
zyn&zOwE@I5S$Ow-c<0vWt#9EX*jI~b`<gg*ka@;r*vB_4D%C1pvx(cPgKc37gD}i^
z`pTFyah26<fv5XdC}fOrX`A`Om)Y9d;QinIQ6^@ETCVYpg3W$RkQxkj2Z(xu+YZgp
ztQs`!vfN^z%7rh<Z+PND8PU7l0lnVf8uH+JlhQN{T-PPb^O9cVWtqY2F=3i4z4rm^
z_62sn_#kOwu)h`4s!QS+gtmx>BOJd%z0tyNw=pU%24!vi;#Qw;u4U}H^Q_wr&kQag
zq^7>wrSa4gRO&Tm=9d^JlKDcAsf?{bL>P8>=*wSaH=J-_p}}(7rD_SnFhCZi|AbN{
zvEvO++@g^5dc$w?s<M>TZnwxXi7><)2O{WFQn`KfFmvy_htYV9lT7d{296`kGDr?V
zkf5CkX(6dZ2~L`_-k<R3c0s>>GZ#l4p1br-Hg*QI%zcb|EzZAiiTl3mUed{!;pIyl
zFa>#*(eg|V>H%N;{oi6ViMTCOynC*JJ2yj?W#}})wJiL+1R}hdV~IcOz16DAXgpx5
zeTXd0+1R{tLk+Djx0TDQTwcF|*z2B>g@j0B3^ZtoBHhzaF|fL0a;eI{Uwne%WWJX<
z3%c0~z3wHnRy=h@^4SN=EZj1TYv-qO4^0N8gzmGSdj#`KPn1;TQc~=d>)y%?8X!$G
zR@XLAQqk!JuRDwWKkdftwPzqgFn!_#?U`A2|M-j8Dy7vBxSk|W3n*-gFruh85mleU
zwlMM>og}Pu$2_{F*|wJ13I{xU@iZIVkSTMTE0;pN$DiQtcil@c7&6+~;(!+9Sx%Ow
zENRJSe)Ioet3P0WEV*mCN^81BQ4~bM1lut15X4;n4CM9Nx0{H<5EMm5k|xa0&$4>u
zMzpF0UaP^>yYFIteU0M$Wo*mBb|t!yh+Luz#r7m8bqnnNh*s_s#W9bp4!P_cCTkwy
zo6kPP>B}qVyvl-ijL~j`!A{71?|p#L&JNku8dXbBWI6kr4vq-<=4bzaZj^B>$$9T|
zjpn{Zq-mmp3BKo031Tu`b|$|aCWK(BRVRujOiwk-gZJ?W@dkHsLck;pICK6*qE?N@
z?MLZ8@d80#usAJ2DY8+<r7&lBcEo{pl|wTX20_3VE=Nq=^>Lcj8ejS1A213MW}G9;
zH&0OUrWp29gz53<SH41i;S8NRWLZqD?a>QU;!(y}7MzY!ZW#v5kH$n{f;s5oRXqGW
zCoLd<IR*Y(WVUUyynl&cGN9Ah#<o2!u3Tnu-~Ma-Z8u`q&6<I0LMD5bE44>YzxVgc
zXht*|sq<H?{P|win*SNSD!G)`UiWsI*7~KL@c$V%z9aUU03wTm-QG6s1B)!(eKV7%
zUSPGaS)9!gMnMn?#!_(kQbfDfXQAOU+jMw#t;;w1KDU14FVfk%#2<g-i=@h6-a5`)
z;}$Gg$J9Q@j^E0Qk3B|y`2y4=L6UG_#=}iB&Tp?WF5qGqG0HNg9Sfl}dS{!OYhaq-
zNlo-NI%H<5%~Y$x*2ZP#<`*zTsa5_)EXy;Vc<Lmx?FQM>4Bg9XtnC!cs*DN>J1QX7
zyx7U8Ih!1qt}#<Jd2XZ6b8&<FKKcMFmrnA?+2=9vn6qc_{8^&JXZ1?J>9ZHuKKml>
z#WRS8L7J6JpIc!{KTz~?MKB3jSlwpn)LE3}v3=n@8ihh*yp5B~G)-E~2IFy`%a?OB
z1_=8NP3Wue!iOGvgcnYqq7*1mWI>1?Yn(0!1wy2Rsw9N0Z}xF4L69l}p!x7;$?}AF
zuc34|TVc@M#OMo(^9f&g_7BlTjw1_Ps}vhkN|Q=OqD6_elJNNNeGTsmPotG0nT!e2
z;yV1XZ=L0N&iM<M$@2^$x*!asRLt-7IlK{jyQN+KOS7UV-jEN=fb@3$+W+O*tH;!9
z0*L2%JbCgYE9)ERA|qOyLKY1wNGul&ZIDEQK_;mM7E{_n^%FJ&hup7{TzZ}>TdU0M
zTf{MZJktZWL*MN1_}PTj|M>#hWP+One8n}9N}`lUklPFeq?Tsb@c6a$gw}6=7Gb+Y
z-5sRgV(MUphLjZgx*(#r#2WP)h5=eQM2V!?YL}D=Z@^sF<>6<ZqvHD|QCYi2QS&g(
zlFD5mC~{3&2iG(?U6{yzPGC(VS}qU2@DO1fuzcVECMKp)keHW9v|?v1=j^YZghFDY
zFuP)}FTo;LW+@45X@W+L-(Q{J?0%+f@JnJ;ZJMUvqVYER&oB&>u<@(YAOuoMrrYg3
z_TJZLN-3tNr&(B7eChm22&z+4FK1o>z!!VmW4g?cTL{{MJ(z8+)cB@+Q~6#EG+tUr
zu2Ert2~>fyf4=<Smt*B?ewDfq-1t7=^_kw|N#j}>NU^lE^oABcDJAps^X%KV57RW?
zw(t61ib*M%o11&XrNI6B_hVVs8$Jj9uhFEG%*@QZ;yLlEL8-$qWIUOa)Cz)99u;w|
zW_xwA8GAsetH>L3?+K+OreVH}F8yVhlNLx-cHe4Y>^)yz$Do`u0`}M;HA<GyP{S}w
z0h#aA<b}PvEEkVftM!JUl;iQ3I7zPg;N_)<mw)<8*W7#V+G40mgvZO?_3~dytE=nF
zON-xCyLu_XULE>szk>MA_z%kax%xe>K5n&IuR9QHtr?9*<az$4$~OPCnARG@Fle<}
zuR9nhr5Fx}Z~E-yzZTQFZ0K#b+pl{}6-B{tIDC7**ndH$wZ^h6+U@o=2fk)Thha#!
z)4{f_l4e2aYlHAcI<jASaJ~16SNv!1B5&v#-|CvL7$QPo80NJ_xzw<b*C|bjtN(k2
zRVtM?6hJ1E34=kuyyjl!t>jBEjo0f({hB|%g%4l-h2QFfLm^(~yYjKt>RNDmc{cyD
zZ*|QK!=P5Hy`i1la5y9kgKG<>(i;6|03W`U>2LkuCA%8vulv%s%9ncq+%R629O%9N
zk5@N$y>c9Ct+8#JX0!Raucnlu+wGF&S?PCOK6jNe<&E6ew;zY%x)0xuxh#Z%*h^5a
zrl?ZM>s;%VK2%y$@qJpY*6R)+Qc60V4oWF3%S7w4S(G>J!#`s!eEGw-VEny0!GzL>
zhGFi_>+<`&E*~mQtyZPoZeM%FYi9KC{Lb(2o4@`W?CuO0jm9i3&9QuN8berAt4$CF
zFWWeji?b++x4C*yCrZH~%e41?Xi;h~zakS?D0wy0E=wH(uxtm@DudV6&r^F7OzUz2
z2(fp4v4`%_8X-(f)50>%a*`;8(&hV$JY#3`6yyG7vJAq+WO4Zfv-6AP`}Zc6QW7a8
zSzgj}6h%R|Kj2^g^1tDM2OfCcKia?jxBr$u{gcnps5j{Bb}>wY<pc8=rh(^r_<n_J
zHv!kmJI1x{W(_Eey?X-o79L(&$oG7&Uv@aE_TF#V&b5X9Y6(YaRd({f0&n=rDQXzT
z-fy}0uX{ep)0ok)>`WgGc9@KN)Z5E=l_?Cvz^_yh2xL(L5er%DQJg69EF*{${?o7g
zZ`^$I&9BRUMN#le|F2)-nWvs72qKhH%*-^Hnr`5FF1}aAFio`Hn*diAX8}TMw3d5g
z0t~~%yt-%$aZLzo&wYD=NtLD4LV_+)W0qwj3=7k=FpN_8=_-ZEOS`&j`_#SbmDe%%
z5(mN6MOz_@f?(36zq>*d42V;SVKtbZJ3y`8!m=&2D#y5#<v6_R!zhZuwD`4O|8*7?
z7G9SRCzA>P;1_;@?TuZotZ(9b9tRK2Q?0q!j*sV6LGLy5?fu|}VPGIYE3}qqr4d5!
z!LseW@wZ2+dTlQ2y?ZQsAIi!>fiMgl$Hg$q#a=7@vH+&`wx)W|hvM3JE5AoMkBz-V
z<?39PN)ZJEb~axi4hF;tD5JsQW4Gg1s(ZnBZ!SwkQ558|KuN`DJYsr&fnWXAU!z*B
zzAhhbY;5qie)eaHf`m(#)~Hr0967c`rQ+auH5|v^Lz{7J92$lw$C;94lDfA8H%$}M
zu=dt1W6y^#UF)USifiBBv@9&!E*E>s=dRx0H9v~0m5b~7(84f82|L#>t%sA*E<5Yb
zkfjMh1eQ0=;bXVq*zVrvtG(Y#k{1Q}9!p2BKj6r*oA|X~|MhF%`!xZiC?t*Q5)<2?
zn*`i;{1)#2uA7OXj3Wnc!?Hb!moMTN<p9$PUBt*DLkNLwyZC;|(t7pps|>y%ktnhx
zBS~V?I3NlJ801V%Phr?Kyy_IS$`o=h07xaTEewWHF1lJvq|5;WL)iF!1<&<fHb@nD
zA?bCtc<}RQ*#7^K_FqqvCRutX_H*}Fh5q8xB6VhE=@!*pt$U_t1~Y>Zz<^j1<QhR+
z+R}_PTQmLzyygWjY2*dDq|r#Dk-#o?LF@pS!RVRRqN=O9bXMy8v<e;ZDi^%)HzKp9
zYA~IN*DS0mt1|P8@NoBY&U2pUIrn=^MzDJVSGT^(Z~flyq7YbPs{>n-7X`C)N}6V*
zS;qH%{IjQSMwX>!t$TsBm37AFC1I@i{L8Np#R2PU8|+-TNOimUFF`2;0_95PgmU3D
zVHgvJUW6BfJ{Zdiit_I_$LJDWiOl9OUl`Y_4XRKosy4U+Btkq6QbK$(fP_JWQUR*^
z8O~bDvS2(o=GOav2{`WFnzDQUAv-tL>8`&>5C&YlbOR|Rd7j~%pE;0L7Vop!j32!5
z)&lDMWgaPglCr*hiEO-&qvZ01OT6^T6*}z>+uIjOYE5i);O6k06awiC&J;Lnkkus?
z24NKlBuWLoDy#xco@eCQlrqoJIztLa7zW;6Mn(idoiM5|0?=dcG`|mpc%0aPB2Xc5
z?Dx75;Hso#HXX8i@8`Vx#vUj8V@}U9TvX%bS8j0a=Bu>Z9lsB?rYuYHJZF~9$cvm_
zZ@?Sxe(+Ruaw!F6;b_L&6tfeSn{8hE{O8!(X^|vL>}+4fi2&<tb(VZ!P*Ngni7iu{
zGf1h3;~G&Ed+1s!|L?+|KPN0@Q8JrODY7Z)Y>d?@^?F3T-o%KQc4w0$X=9D?7rN76
zEmDAAR^cthokfZO6~xqQH3Ajjtnq={TGA}z?uS3%jqm-0!AXy^a~Krxwbx$d`4?Wn
z8jaBwV>M-|$+N6l!-CzteeNF~FZTK`_c*Lktah&O;dGaao9le%TVJKoku+-?Y;9hq
zv_2VLtfBG_2*MVaGR0X-pcF~1K^S;@f2BgAFsgcNehsxYWV4haPnk{zl-Y<@t4^cU
zMFb7noehE@p)7K5yM$juA*%L)aMc<*1OZ_fQ?E5pN_v+F=df0jWhL*v`EU985B`$b
zFy;6JQWf)^ufNWPOINXF>3!{tq0}X5WptYtB_G|s!yx_SjF8V9Y5R*eb}>eir6q5_
z`H1C}C8o0(7uT=TZq>>49J-jBP&?07k^&L<LmI~ks>(k6;d3~rD-D8W)H`AK!7awa
zLjs}M-d<;8V-sEW*}eNN!%@!C$|W|opQF`T1tluxjjVp2#8{0G^GZkg!wev)l*l16
zf}$9ZPtORW2pKx0E7^T;kJ)U-@=`U=Sb62L6A~*0B8YIv$JNl&Y&#40@0`&;70hNO
ziF2GDj|c)qv(_OBsw^b<B%VEINBM)C5Cq`@_L;AxF<yE#9-nh`a35@rGh;xoveG4v
zA_iwE##kW4i3FUZyLy3IqeB=*9$Mqcqcs0MjsH$nh^%^FGE7J_g|Sm`79lKA6f>QU
zDUNrk*Be~9{(^U+mHzXHAVfw9VmkA)6N2!mj;#=oXPP^=Psy?pXETltMl_OhR#!Fa
ztD8hEvFHKK>-5EcQ39cS^%g~8b(#AM(!UYsO(CZvPEU3j_K!fPBuT*1Qim{ZkWbDT
z^he~SAW1s3mp5s&mx-cUwWmMXPZCx3F2(!}`HDmZf&ZRo7GqLyhA6HR1ddsjaejKt
z#mhHQDntrfK{OIqy^xA1%Bt_<Y1U9M85i9DXhcysj*f=B`_>+3r%SA_=3LyoL|hLk
z4V5xvUX@v-aQ^2BK@h}5QM8y`=PO!2fCvGz>6n8@cj%w*VahShTFmCwI-S-MlgS}_
zdk2&{WM%ydD;rm-H#@#MlhUtIwPsd${gsEBE5rfk?~|%(EOfzab^=zD)LVq(lzcGe
z@Zb^8zxX+VAVLU(w2v!JrzA4VeKl15%wP7AN^*KU=Kh(%XvxF-XT0^sF5T6bi`$Z`
zm#(6$#n?Hl>k|qoe1a{M1SN^0gfNUAS8xg(2Auy+%RFWO(OpiC?qP~C?PkKI%NMBE
zEoY}6v%7yl7<bv&zRAkkCBitVs*cC7vcF%QPez5Ra#M>F5~U<kNgN@iF%0_;P|guX
zF=0?}+V694u*>CZH!EaAB9#Iuy?vBYs4OQAV`TnGR@&b&kx&2u#^oHJ^!STE`2byN
zKKH_N?Cf-rIJA}6D*6hi5zZnd)nbJNfiJ!0nV&Jbf>;D+CrAA0AN{x3{FuuZH>o$9
zC}|Oa!)i^SEZO9Y5BF|!|Mt(gbmKLyKmTRoxQR8I`SuEe$ZtYVo2sH_;qgl0FEir|
z)A1SI_2;>G?JMj)xXu2-TMYZ>^!t6*R@brCAe;jNLIFnlNq8#DpIW~#f3$X#rKKzl
zLOF(m8B$258yQBIDCI90J|T8;)rOe&8XhN|bFZm%<^@udQJ+Eolqj%7ssse4%rMeY
zWD~S8SSu;D0_RXcL{a9nJFCR8x2gX$v!sIYoP3PG`ed7V;w@!>3S&=kEK9Gj^``})
zd?1=%-cl(aRGzvKF3`p@8Tra8FEe&`&&g+k?d^=~*DkTT+(rcf#(Fz+4=Y(D4yk-5
z=qnH9gKt%lS!dC@WHK4?;O<B4KfHx02CT0%Nx~(Bh_ObHPA80qXG|tDAmIG`5i9GL
zSzW(GqqU5b0SH%Js=+7VEK?0(La3_pgR;y(NS4>GQ){eK=!Z=A`;10?@}j5?c=hRp
zzzTsyP-=?|L!K^ko`0JT&qEQ+vI)07I^f=&b8b90=Cv<8Pp6%r@Z(IqZH7TugTSFe
zMGyvys&syq9hIiU8HzOH?KgkK`)_}rt@Rpft6f61L=XVVVN8ZIDZSHujt)L%Y4rkE
zZ+?mO%^QS4xPS$PkOV<g1z&#=m%0RjH*<9UA}h)ny3Dz7?HdG9O!m_s@^F8T^V2ho
zF@#Zw58VFqTOkOA#8`n0o?#80Pm-<k3CJ`X^NY70p-aQ-uNQpt8=s>dMHHnUX5|S(
zehsB?sL(@DQ5aPTpd_e1w{iZ?DT<soe)2uu{@Gu0^-`CWZih&PNCj3{gfu8LhY#Q5
z;oY}c+q%y4FMpkt^{XKKd9}`=qz_c`323}DC86}mjdMOoXW0mA4Od_OZKPBT|N1Y`
z$0Lr9kHC4$8xeptem5JDV326zyvzFEdOo|WDJ7Bg1CO#I<>Y9Bl#W}sclp8>ZxRTJ
z6AmXVRvVnv2w@4r02N47pemSYj+rVfl5#en-#_L3ci-a2Kl%yJU0LDAl`Yz>7Eur|
z8IL$QI^}S04{J;6HATOFz@3jKOeZ~F`TVcbY;`f#A5f`+1<q`&#TbpY1%Zu-f*MK*
z%CcZO9gx)4S>L?DWHKX&-p0&kWO;@RLtI(<Qa%U>g8(D^<UpaG?cf3!D=ACCc#<Ke
zmJi=Mz*=b5o4jyyhkBz%EeY_Gz<lW`1LZ%LP=ujcu+PBHOexEpgT1?GUC=u_;=!E{
zSzB(iz177zN0BP54jB*6&}B)MW%R3=9>>S`38OkMyz(uU*DiRko~Ti=-m3R(wD;@5
zV$7UAD=D?bS_M`T#x0tyRZ1Z#iWyN9AXJD7qD2-PB5)!g)Rv&ozF+YSYbd3MfplRQ
z^e2?5!dd9|MqJogAqpD{mY=I2PA3FH1t=w}thxFa!mIrAbi!;l;_U30M-Ojva`1pg
z>=3Ipl=Ap}I-4?^jwy?jvY3!&DP`|~>7<X=8Cw@#q}f`odPXb`t#cM@op&a721mHS
zwTm*REHy#cqE?THlXIl(Qx-Yk5Gue~4bod`D+E$GR1o@oF2pl6rBx82R7|eNjD|C$
zFdQBXI6mpIy|sc80xPUvKcjKZR2geXpcG1p>KxBs3T;Y8gEP)f5BbI0Z?Jp+5v!f6
zL}5rRso|`p-#g>v=$JwOoIn|}Y{uc<hveCqJe_g*+N*?7;u9k&y$@8?=r_h-tj5@q
zFpLRQjF5snODVFPxVA>C)gh@jkqYu`>Mt7^_&$M<$WRd~hqgIVX@m-%A&~(mobSuZ
znA}Y1_op}`dHC>@et*Q)#u|aJXuP|)PewIDNdgrT1PZDAIN3SGELDirmf`S}lj8$^
z@cqB$@Zke4Z8d4s>(m={v@RK)9JBXmm(k#yIF|nWKYEueAM*T*zry8fFJo)~7^G5_
zlZC6+P*a)$V*|n<Aqe~zIGc^JMzFg29F2NHQvWeZ4M`_sM4-S_eLEphQA}(s#)4ED
z<S!XO8jS>@6NC&v7(k%4<?fwh#*-=SRugMWtTx!vd%Ora?<Hsi0xJwcDx`D>sTq&X
zxqIv9q}do-Ou2S>gR57zP(e(Zmqe*|^i8KDYPASN%y8IeI-3HH_uu|5GD!H^xBfB0
zR*G`x5Ylh*QkPh(kyfA!2Nt5Jj*uZ~I$`hbo1E?6Vl){LhKew>)N6HuI6+TS9BM%!
zwywl7KfKSzp`Y6*%>QYO^%dLPPWdVCg2%L2&qi|ARwj^hSJ)>S<nA#pZ!4VC&U;AJ
zXMEPc=>Gy)y_csKxBYUD^YbVgHOf+>3V~Ao@TIeo)3XudNlLv@t1j&lXR7xiCDuxu
z_1->0_;p8X$|B?RWS2qzn9=YYlaFZBLK?LI5EyO9(-dnR*~~+BQV6sz84t%yClm5Q
zgN(U$y-5@`m@`VG6d2=iX6xqtI*SWIsR*MDv&oR&$u4D)lg@@HB}gJgl*Gs&MC;0C
z#wUM%os~ic!829DN=X{^m@H2~f=~|OB;#?$?xRy)dhs$r5c=SuOHZ2@9)_`6qNT6)
zr7#E-Mme;dadL2<y+^lj*3zs?u3Xxnv$O<Ain3%j^+C5tQ=%}!V##~wq_d2Z@;>8f
zO1-tr#Y@jsh@z_q4&DyMls+Lv8@~<#1aTd09h1p9ld~hjI3b;mh$Dz1Nv+ZHC<hWZ
zC)fxtoJ9%`ML+c}gfI+{wL0TT0;Pv=g;1QGPT1W&W^;YHO1KPG7yfKk!PXdq)MZ6R
z5J+i}%As|^{=-}B@83pA#d0TM>BTE7EiF@&hCI&^qY<O=2;nr1Mic7{Cub*=rRHS+
zF8%(9&gup$%iHL(SnPczWktj&5z?WxH`;OzqNIs7mO*ci-r;?MIH4>uk{IGpQg3w_
z5Br#E4W*P=t9`X3B|-RUChDgGh?Iiu?H2u`07=N^PKRz|g*?kSI~{X$bWW?)^cR>e
zFuL>%D;M|!TMC2|7z9EZgtSb?JwANr2Ryj<F11>V%NH*5rO&@WnvN(6LlF9T+EPlw
zAVAdWl-6P#^amr(&c@Umb$<Tl_j&2nZ?LkmMOhS8)@iFO$@}{_EkIbbvjkzCAg*Dp
zAPPgQ)8zRSr7SDU9X7TufDjmGh~v7qRX3&g*p~_!MyTNFt+a&@tgm(Gc2+Q)0G6$d
zj>nWEI6UmLzTs%Ll8ObSeW0=a@~}eqz6_cI&!9ToyGuG7GdSP(jq@U7xl<!lj@e`i
zcoC?xmLTx#p)@ZTj;FqNGR!be&|STTLlVUa&F0lbqj!E}JUgz$8jtG+aSb7$%rl%b
zl=+k*AJbgkpj%%jsWm*rqJqa6vT-0CGK>i0nuoETN&yT}9I&;$%<1u%+$5~8cUfw6
zF{Pu|8`AHOSX=9O=FMW3i@*utGcc_e7n2Z{JkL2hdBpC651EY4P(skIM<l9_vn6?!
zlZL79KLv^~3_(cJqF|O49G?x4Qj(SeCu?kNU8T`z`3u)oY$ECW#b*sxJG8YR6hTz?
z$ru7_HAOkYn4Gv4u)MZHv(@$$oXim-sP=-xN$IPpidFYCk8_SB30PikG91R#l8B2J
zS5PEmX~Co26DHFc?RFb$G<H7EMLzDe3GD;1Ppkr5S#o;xfDhjN0ll+BHa2#+y0gh_
z>gUd9SwRqHzVf4TwT@$^>5MGTIXph6R<GfDBbL{H!S>EoqA;lx;@0mO>3gIYQ!#;Z
zv=D@<gH%x^D=3*vQ?$-V62ZpC28~vme(%)ZMKY+W7=y*bt15^Vq?M;x!%jOwL`}}l
zV(Rsn>(|yOGf9@^+`6^PtFK;1N`*6}4<HVNR9N8?5-FsIqJ%(5hb?j*-Fu%8-~Arr
z@sNu<*LdaTHAbUTilQV8BZ>;rltPdsHMF%A8KC4~|CDySLw4NbgZJL#*M9vH0v7X(
zq>_(&ET&=@N)Ij9S}O<{`g0uyq}d2(3zC{(d*=d;W{dOl6T&b?ASks)Tj$#W%6;ZO
z{0ui<zs^cV1Tx{$l`g;kdtc+NpWf$tfB6yj?;mpY>P~eyD>f7rWwa{M?84Pi;&jQw
z+wX9Cc%M8SQVSJjIz$P>bUb1*?9*<yP-27b$_n*HlQ2%0rYW~>f6T%05ToJZg@DtC
z_jz>x78@HE{d6u@R+q0`Pz#N*iylG{1SGXONv#D0NTHCz_hc^LxW@IHFLFNUk>>^V
zdJ7>X+FFnE2ad=IR7DGZnnwshr(I`xxk-Oq11#N@I#LQ$2!p|xcBj6eO8QKms+y>%
zx%gje;}6XF=|0nOpLBeRF(r`-Xw^dykZ0bmL+3e4dT7u0Hw3dR_xSB}MxK`p2j|r4
zZR(BYf=X!J;{bxIGIpr=_WsZrU0|(ee(5qLNkUpn9iq6lXt>Rr=tvJs$RI$4;WK5=
zQVJT)gkirR(FqqWb=h3s;CO$`pg*B^Hlnyv_&rxO&t0{-#e&A|1Bi8w>8Qu);R8nf
zL-KS&62>$mg$gA_Zpmj;QY8t)fLPSL{zQ7{ElqRA(-b8ImUE(bkLhH{>gpyz7**AV
zXA24Gi+1ZgZ{CZqLV_?xDv32Ey3D{CYPA|08=JJ+9fVRC3t?CV+p3|gROo5hp8+5I
z{cGsQ5G#>TZ`Anm*DkZRyv7f|`!V~wXPg`ls=4)ZR^?-s-;*?Ys>^RS8S(J$yY$cZ
ze4j1v5QI%6hCy#Y8pl}OAq+gVTh;=CIA)U0IPDENI2y3L;uF66w?E+ZulyQ~^%hEr
z#W^+=gl4R%Dr|?u5k-+_GDQhe`S~IUNxfd<=8G@UTIzCqa)?9{g}zsVHYgQhLqQNl
zpO6v!QqPkVT#-<(C;aMfK2NLB=3oDdpK<HMJ%+<EOG{p#Bc!bC(0vns&Sb1Qn}`4g
zr^h_J^9#nKQ>-a5#TXaVz?7Vxo-u1QFlCqZ%}viLO$0%l@bR6GIXvmJd)Q|ih1LbP
z-upS<`sQy_tG5UOiF2j@`Od#*t4oaWD2NILNnE2=ZxW~oqYI=65JJ*wHF)_8U!>M(
zljj9WMTB99at>h(QiX)5#mMF-pU+@n`O-J8g@m$3o)y&UiluHH4BWqSNc#1ZDDpPv
zR!Tgr+iNj|om)RRgo7-fa&mAVB{Vx*Tg)a?M*TB3R+d<9Eis)<aN44C%XFB6gjrUQ
z<{3Ko+00U>L0(Q#N^pGi2xpClG_1r}iFGAH_;OP_jRyjxzeodx2nBgD#%N1f%us=%
z+g+vEYID{<!#FBsyf4*+QV3&EVTcNz{@^<2IPaxcGbNpAgfR5Z#vbO=0$adxx3&1a
zZjNZp4{p`GGsd7x56zX?kRt74${cGnVHB~_=~i3QU(k_^pumr-HJ1K)k70j=pH=R8
z>BHeUVH8v96<=aY@z|NzR`6K$euT5+g=Uf#9Gy+D)-oNXWND6+5zS@`6~<MAS}Zgx
zDsU(VkpmO^x#Lg!SnJ4A51mLUi9$t^D1?Ci`IugBM4si;>WOFDELcSTy@)x4h(nu_
zbT(qtKL(SdgvIC_oF-Nh8HALk#Fm;eEfH2=oTIRY@p#H)GDVjf0*}K_Cj-Xg0Y#A$
z#U97^L090d*x@mxSoJ`(VLZ#3%u@0+!x%%HM0A%|Na|h}<eaN0#+BAWDpcratv`(w
z%;FJ(FjhoiL~a#f=vZBE&}_s!I2^LScg*E08$KCR3Y@MAMd9&uJ7-S-(lQ*J(L33t
zS+7x#B-2rkRwQV~F($709%Nopq$SRz2qnn0rpR*gyr5Q(2|}NYOs73Y!*ez^E}{a7
zH4>w3Mbz-nopILpw1o1coIqi*Wa*Hi%yFh9NfK7qHi?4BGY?fI-nE`cA*@9KVH6^s
z7R0K8YMdydFd^3hV+%Un8m(r+@j;Kn!&8=)ntlzXT&$T8)f$?LC{XEZ&QJFl4~|${
zS)t5R2E7wn(FUzr4cly^jlmR_Y+6!g7L-6wQ_4J}EOM5+ZNkuDon<;bXEGVlXm$XB
zF>*0+GzbEWbr@%`LV#4LFhE8MZA;Q=kG#wPOD(Cfyt0lGb0=l7s3d$F!Y2_y`00+v
zr~Zqzj*mY+r=7$o8FBB<5r6ovf5iEb#v0h)J7+e_sn=@@L5sBxDcwTz!4uQfB#`AJ
zbUI*V=_31&_Ni~ySzTIz@|0$h_zNr73?|2nCVg^eIXFJ&`1H(U*;*5Z5~UmuAKqm$
z9aF2-(M9nXHK{IJt<8e+iVz$hpYZ2@^%mL5jBZm<6xx5u!e<LoNTMiOh;@`wC}RkM
zfVus_(>~6@Y*rElC8okn)3oqGPNf7Wvg}Eh)6e=#z#5A+g%?E?Gs<G>vlgRKLeNa=
zm5@nPKi3O@(&>ysdv5-0mR0zx7jC7q36sf?D2l7^?+Ke4_sM`Z=ildrW;D&Q#!zGh
zWm$SrWKu&^Vo{%^`uE`nN_i?^@C-C)At28+`+GyAP^44Cy*tO8pQcO)7F`%d!;IOi
zAgR?rN}RDzw3ey^uwarotSKpqF+%vvw$uel7}2i#nH$Dh9|#>3vyAa{M1MTy==7X^
zZ-_M(l=SSQBBS3wXF8h@*BXApBd&71^&EN+W%v+Ml!l{|KKCD;(LYO(HpA$GsDeA@
zwn6hbD&Q5*Qu)izui>-Nc#n4nAQZDn$<KcBF?T;2a=M=~o0c5x_t4r9$8iM%I>7%-
zSvBqL6U}>POJ=ikmOCws3OGGF=DAB(2&Evv>mZzfVv;gS&zKcc#(BZ+{t@Hh1StZX
zF@#Y-I`j0~OIKd>{Q|8QgcxfKWm#e!1QqmGl$u+&@AI>_ACl;dB+UJ+a}gkfT3nj*
zHB`!fr_x9SUKskcO>`*)=Vvp{(g9KiOvi@r{q?*2JU-<7bjoz%*YKs6uA_pg59=JF
z5}L|+MZ!EekT^=FlQWi=E>X-Jz4IQQd;UdSkrO&mL2iO{Fk~_rFe@`gX~ypUG3hiz
z$_TAXYDvtXf5zd_KFh0HXkB7;)$cUEp<h=0X;snoi&FE(8}ISccMjQE0zr<pmLQOe
z_vmTwkJnHsi&6%OtRw}0PXz)V5-C+lsABqqF+cp_2gtHRKtS(&LjQci%E~g<i_h^*
zM9*KJzih><R6#_$-R5Yt$Nv5<=jZ2q{*^EA;w!IH7CBM`B()~9Y(|kzIoN-|qrLmw
zzjMIZNeW1u$#Bj$*v`+67>$R-aY9j)o-raE&UisnS!!eukVu7dmb3G7CPl=91H(#q
z#!?%K(xFSAaa*l%rX)}jr$XN})RI62m4NVBo5cPKww0P!YqQwDo?#^)(_<H3iulCq
zNezv)4r|=vf)}Eyun-HXsF42ud-0*CM$R=5wt@&g@eZA>0>~$6#GibG&q7q>u^FMN
zX7Rj1|M>5G()()xgFMsYtD-dZ+;gKHM@N0m&L${bCkjK-v>;8(>e72+%@7Or+MIeU
zl&z?YmUKF1I!y^xOuMtfg{=$J;u=x~z7OFn>2$)``7y!4F1_9iokKb^Xk&=2A&*lg
zlOb6)^Wrt*^+?Xx1tZW}hXRcCf~+8_BjY9dQ-v&YBn45;Vym^3l{i;bjJ05nUAy_Z
zKK(v4#xR?ughJws;G<g)3EdEB>V!&idNL$SHF2DHf`=1TKhakJLI`A4k>V^#QX@%X
zMx!46!4$0G@|EY<T;HTD3c|1kQjupfCZhp|2M@S?|1N{m8N)$I5P4!joCu11$nnV`
zd65xFg(>v{E;d?I=y{?ih@2(OW}Kgo=*?=34=1cQ#zdh-MKRX+KAzPjp$Z5BU#S{v
z38cje_w?0=--c+dDbmzKdrtDf2loi=h``m55{?i1=+aZiZN+DiN`B%#bgptdaL{bF
zh@yag@0i1*9!c2X%9Wd}bbX&WNt#&v_}^jwjNRS4eE9J#dZ#nSvyvzR>m;^rD2klp
z<9(ig@e7oNhp&`U;4G#rDarx|L_$<a?UeK0n10q~@bNvimJ7g=)LVWZ8jC3kRMPOl
zLV(nUKn2wZjZd%P-=PDsSf`OP^qhDzqoJBe8L+Y8Z5N87#At^U5lFkpIz^=kkXGUr
zw8)&Z(*e$n5GXpWE{>3>-bCQ&bl2HzHnC;K>gqNZFTcR*=5@aN{qORN4<2%OGGhZl
z94X>BLO6>pGOz}N!wCo0`Yc0wlSXX}S(dZ2bCLh*zxv<uU;kemAO7Iqk^~;kNz>e4
z@y=DvdxbUzC8Sr3x)N>XCH>Prjyx|((}L+#V{OTFGQ${4BZ*mBx<C|0=+XyZ@0leR
zm%TtDtE|de0j^*;nlb2&NTLv&Lj?&dYnO<lh<e>yGkBtaL#hT@mLWN2x86re$yx7=
z@hoRH%cwWfiV~St8?O4^=X1?@d|gSQP(hvA(lwlOjCwuJ&Qqq7k}NAxfj~Hik~~qZ
zxw&2sDfI$kdz$r|(~JG(rAX5P=Q3JR#K!s(op!_Hyw+9w&aa)4(zBNq)i9n3NI;Pq
zdc84aQLxf&P)iyFK@DU~94B<T%P6Usj)xfInC6D{amH|zF`AtCJ>x8mrbP&cGa6kM
zRUq{cg>asRi}Xj%R4xb0E9-3TT<6Y5?=u@8p^T;|3`L=_#vlVf5>-|GA}3{qYrCh%
z?qPA(t<|7nq>L!?0w;57GNIY3vAW*HI!94xK%!(=(URq&hbDxqvSNpGiqUY!uy;b5
z87}TzCJgI1Sw}?)ajnks$~s{fl1)aet?jUV=>-~{4Sx93H~8r839}-j+Y#PUB~W0>
zMd0<-qCp5xdNGxSNm-UeNWS*<Z_;cx`G5Ui|BsxVzROA*q$&NduWG{vfh17U`&UVY
z4IR4F3pnvv9&5FyR*Q&fo+DL3AQIZ`gyrS7PYx{z0+gIf7-ZG^6W}D$w~R4bFdR(K
zWzKk<v36w}71R(xgQ#93sWs@XY#@bWHXg9CafyppU#8jlAwT{3TYUVWPo_({Rub2m
zgn<N;gD&yIa$TkJmxNJV!K>a;A_0E)cYmLy<qrRk|I`10(cqY*2A0A@wAHo=L**6h
zMi409lhai{>Tf49pBX@ubX6ZOLdhD1nIfIxg%>vX{eScgmX=zSWr>o(!gNt73C?Fv
zqB_7*2!gOiX+rk*&WWNL7cX95snca|Z;w0ocj+_}UV7nqE?l|B+3_LwZr|asH$){N
z&)vMvQE$S*=>+FuN^4o_ZV(3{)7coT@M`mly%AKRF=H*-8fLSMz{RYtuCunW$s1)*
zdE>xrHlr-^>cSOO8E3)y%Lr?*Ry|<{{&bI$^F!#rlqirG3AH4m)vgo8;o|o`_N;nr
zT~1YfC>2!9s)Qh{5e7b!j>6bmOoSm}MXyszA+05jJS><bHEKx>B_&FE3^)uUq9`E<
zBIcf9pM2*2QY?BhN`*v8-5*jJ;-_nRuOTNMt9$34cacm~3B@zre^L;Ifyc6Cgg~Ga
zbh|ZPc;Nz9uWiz7B-O<stM#rbVFY+xc;!hY5HhAT0mEU2l8__~uc%Km%F&c&Jt3*r
z2!qhuo0Ntys<F1VL7L_qpHCP~O0;oE8Bwn_QBtCdydsf!uD<|+Kw+I+u&Io3v^!n4
zwlC7^t}!{gk1ocPrB8CS)&z;~JqAG~F12${w|P)`wg#}*>PrGy2Nc$pEG@_U>aV=S
zH^1>Jfl^g3EL>#H9(tK4gR*L@OTw_pEQQ_Ob6V{V8yj0#Yx(el59tpFT-;jcxtk^R
zMw8tK_u1V&;CwQr(O%-lwM*RJ9g!DPERrCqv9i2@E^-R%`=?SUr1DU>RFXh}t~}|+
z<0&#qSYO|u(dn|kCs=Y0U1-vDw$O&DDu6gEk@!kPI7e7n%RKGl%?mu_t)%Z&8B@|(
zj`*G5eT~<?^n$N2E9Rp|1_XitXGKLas(OS{5rhqTy^LYMN4wo-bJH_Q-+1%q42FGP
zxPFD_Uhv8A-CG~9yMN47Yu2{5xN)P$-A7~cyubh|tg+hNAe~LoMOod2vbql?v6NV+
z(AGnO<ME6rX>j4f4k~FfnI01-qDppWUg#~WHIxFU1VRMJ3I-08`qV!BI|7KO(gi9a
zkabEu<?8ii{@|Z}gXf;RNNFo}gi;IZ_PKNKV$r3yJqY83=U)0M@4o#O#|OLIyjf#)
zb(N&nLWD8O8p1%4PDh*^9iZYGDoB|O$Al_mbG=Kg_J}BskWzB(>I+EeF;R&^;VKr9
zXG}>UkV>LlfDv$bd_p=K)9EY`w>o5Hk3hg|nlhbDvCg8DB#Ite&<i23Vc^-Q&(ORS
zLeOf)EVUAbgBTSC+`Ljp3d_qcU*yYQzDW=%5I%T4wo94YA@~3sgx>hhM4HVDFVJkQ
z6X+q%6s)aou)4bD@%M^`n@uNZUDB*KsMYFN3)e1PL`g}S=j27n#>Pc1Ub;q-)GEf+
z<I>cBXO%630E8k8W0HEEwe@X4aP*)<;bLZ4&S*R$N@{3l2!taF=XMu9n|9XuiIvZ)
zVQ56mrVb?|l!$0I1us0m#TUMCjh&0@gi(mBst#{uL$!vYYGnHFLIxqtc9)Hv8yp=T
zVaplKW|O5(mwKZ^S!e>`5Y{l8%$Q6k6=P3PixQUFZI;?~O5+I_jb@jnrB$LR_8C5v
z+P*ReltNaOkN^>2owqG1w5Hi!LIg3Tm6WAnHk(ltIkEHhAz?+@@bIF-2CnMqJ@pb+
z-=$}U1u8)i(P_kd?d#9+E8qD7%PSqU^_KZTrIem)gsgjPGa{r<aN6BfZocwW?%jJp
zo)xUFtgyDWMx)hbZ9Jh7hX`jF4SI}5Bd;sSN{V92`pPQH-I(5RhEgHj?j{=>JIu98
zvO*3F78UrrLyGDGmJyk?^o~!M42P_)Z}8}&ghHnXEaUN*vMfFH7bwCgs1jKZ-PrKi
zdYSWvzNjo_!k~f7LmD;3ul>f${OWhUNSs6zrd*Inp4{u!I&i)>IJeSS-Mq-v=fA}F
z|LQ*w$0Zx<>#VG-Afp=NL6=sufzgFu!^xCdqfPJl2wcv_+8Rqs5ozwxq1Dw3tgNhh
zqK$JNZmj05dm*hC7s^24T*x#{x&LUNBFkCZ*y8BceU_{b(&O=%vKpQhhKeW(0EaZb
zuj;6JDSvApewxM$M1eOk^IYVRP89O(uioIf>pNcYDCSKfnET?mihbfebXYW25sK$u
z`VznS`~P#k|HuD=m3D_tdx@Q$t6aSrkWWUWlO8%v!CJOAHhBKI=jaXkoSq$Wa(+&;
zCeeAp_2<6G3opD@`O5f>{v@vNH@$z93|LxTVPmJsgM0Vb+1%p7<r@qi{ghS+d7d$u
z4A9yvAP65U9^2PHql<}O$0~@<-NIC$s)|hcs!jbeH^Spy!rbX~{sqoA<dgFr=A?!>
z1H_vLR&KO-y6|Hs(Es?)7xPB6c>J^G@5leF$3O2s@|Qf*#RoocCviwNEs#QyNXdFL
z;>Lw8TN_I>o3+J-?P;h!5YA`Ji22W#e(hSVF4v#`JUY+F#wTp7t+UiwqT5~f27JW~
zfh8>pgf(=U4VsNQlU5yL3R;aC!*M`+X@zUoUt(=-i(0+$L`6SOK3KG5gz(8hP;VfD
zlBG_E<>ggm)I>W)sV!ON4HIidKLHUuQPGIUNzBv2Fj7iF>9a>E0wNi3b+gH9uUukf
zxs6Z~;pUb(b3ZiC<PlO-R9Ue&%W=}+wXgmLli`S?yKm8LwOLtS<zk~tnHOY}KHB7D
z`IJy8E?(H-#+6H)_Re^)dzW@2VK}rj>Px)x@>f|~+aw5rg`4lfY1`qQE<6)~I7(RV
zZnCtL^YB5=#fz7Czp+e~?-B==*=$6fr=Ip(eFsv2u0*!tKVcs#sXQ)F8B2yP<i*R&
zeDmusQm@yr!cX}0!TPaBpsl0@b5WzHeAt?;Wxn%U{}Zywl*#!6mRfB#);76xr9+t)
z%!X$eTTrGc&3c`Sm#*-_&Fh?=9rI}a0ZZ+K{A5TRb@{^QzfHT-MVqo>e^vW`4qsNX
z0D&N>)!5v)OsCV~{@q*Ly!kx)cYZ-$j1U4QlL1AMEsSHmB%>NPK&7JpjQj92bRd#i
ztwA~U<!g*%%g$(?9zmWVk_br^U2!0+7r^m2dwG26`7bVxYkcchf18UN9b|Tfv|bU5
zg<?9!X^k`jVI1kO$Hq#RXk~@*s7IOQn8L7f^*aCPpZ?F--nob|mETpx0Gpp;Yc00I
z&873bg)oZQ+T7s&y#yJ?Omas}L7_FX*%;&0!a93iK?&g*OACJdGdyK#(SfF64G|^n
z*#wCpnayx{=6gp`;4_r@CVQd+nqTr#c@^@-D>o^N6p^0dOhLEZK^W~V;Hw~FiX2^(
zgvtk-S`^T%*AY^1<=RbFwytpb$_=)*E+Ca6j^j_iaRIJ^(xeKh)oK)FL9Jf*|5HZf
zMv_;GStk^>)Y!<(|9JHoDt8I_Y=_rXjsn6$=p0R9S#_FrnGu*0SsBIoK=C9tJFf&D
z`)^1o32JqAE?grC1Jc0(&CrsB5uucXp`<jDveej8<I0jCj9BWlNg_q>{FG*`#`@YO
z7q8yrrI%l0Wo5l$vU#=iV&uU5{o~J)3IZC921yMjqa_-RCQe3_Mp0@*UgYF?RtZP#
zBFHcJWQ#a1p1O}Nl_F3QUHA*ND>Q4fAwlmPbb$yHgnWGIJWj?u3@Dyp)`_Y=v$nRy
zZ~or@nBzN_S?e}&+9I96<QcZe!8)W7m^7uFjalE^!q_G5-*-4;sMXqh`K#aI^{@R3
z&1Ppo?wD61-~-8={OZvo55cxtQ8nBxMvH*70Hp+3HY3Z@1%qpDdp-}l&w7sMtfVrC
z$qWA-EDO!X<P_OEN9=4abTbZrXIM8kNt|a9EY=Wwg@5_V4gT5x;(x>O-Ji0&)WvFp
zk^!?>ADgG(KsiA^9z#B3WoZSg3+~;&jmb+8G2i@^-{D)|`E_qvtL5VTetHeTS&NW<
z{K3j{m;Py+Fiyy<Br6~a9ceZt%QLL?(DdB26;Wxdo=DE0d_L2ScRb38TufkcE|Re+
z6i07;%<;|l+4=hG1WHm?9z=7xrVk)<=7DD!$+;7xRD^Mpo1gy{y8nPI8*p~?fNU~F
z8jT1NjD_iV!pX^)Mzc<%y~JcZB9xM^e&s9t=0Ex8Jpa-c{05cP-z(;_{mSy$`Sz=G
z-d{F~6OuR~i@fJlZc3&ZbRsa;k<L<N-~~pq+5~ePeGZL03xwqy6uF@cG<9c4oTinP
zNM}fj6j7EC1)m5Qe>;Hq;Ng2Z1d5B7Z}=wMcpqmAV&M?ZAnap15tA3_qM+8O5rl$T
z6i}~wCHbZH1}>;^@zOO`R@Z&=d4BQEX_^%c)*n5G3Igi2I%Qc>tJi!sszOSmD$9ie
zMBptXj5pz%H^1irM9F93zhI1EI7kVNqh&MNPSY5tjPE{Rc4>=Rt4`2pJPsRDAVr1I
zRs(Y$zkDw#Od@u!-o(z<A?sm`W;{6a0oWLDiZa*CCNuJMh6)5l?rGGkD{EZ5vB9+$
zKF8JTFObCXqQ4;KYgg$`T=i>&uROybMB#b#7Drw>GVLfe6osbDy`zKg!#%D#q!d+S
z{n`3Ry0qk}Md9d}lx1OvdId*seZbnKZI-Ui)r!La0RR9=L_t(vBamV--q;f==8QgX
zgeirp;2GujV-T^jahauNj2`SU9`#5^Bg%B@t5gw?&N7C>A=Z^7^*Z@1B~7PX+S%bt
z-*|~{|HkjKvby2PAW~Jm3NHfoYiJkrbZfmwT@-t-HKhWqHRRe-7J^#j>8W{|;_RI1
zSXCX>8Y(NQQPNNM2mx8D$ur+1@0c0ufrH~Q55D&^Iv2O8Ew7-Yd}1=I0#wy^tM*}a
zFRBU=kSb(t=LV}wP0Icw27?nO{d3asfGA9GA|#*XoS&T%MUtygL^hi+n@+fVVViHi
z_+@_k_x~BSM$0EV)yQ3e*N)AF!QK>CpiPMi0+OVLH6>ChN>h@Sj?{q)1bJRm_C1b3
z%BqX_#6GlQLGXC;`E&rmPtQ{}<$^ijvYe0-FzbzZ@Mqtny}3r`<_$1DbMe?|Q2jW8
zkba2C91j#qp`zH&vq-uaYsjZVM&mv@D`+&@gmI00GW9kT#&U9WOe3AqAD(gZ)i3e;
zuYQx&?JJd|uVG&2`m88eV8lX5Z)RsSW}dxQ*P3y@%9s@nt$npp<~hnj=*e52#)$N+
z80oW{XJNnyV7aN-Xeq);PGB?au*6!!boYp}cR!+eX`9Z~i+<Bpb{}#<$MqClgv27j
zI!5D}SAtJ6tSM;5lBiZyE7i;o6(lGtDYT)m8dOMkZ5x34@&%%LhgQ3bltINzvWpA<
z$#-w7>_sW%EvbcEz=Nf6q=h9f3`J2OROk&Dt*$B#zuBY|sOlx2K6&=lV3E;sIqP;n
z-DQNCX1sgM`THM})a!JvUx5l5^!+cN$RIrRSjk|K;HV%(MTu{4haozUl+z<jnITNo
zw@`sMy0d<!Ni-ah8;y!1Uis43Sl+x!cV&~fR`XncR|U;#?=3pl)x~EW3s>SQox@m5
zVJx|}6xz>$D2u{3(p9iXX8eoAIz9^}+BtB>(6W85xl`)Wp$(jW{D`9;zeRniOSHP|
zr_qUOWB{KG!b<wUA*B~w$}l1fLc;I@lYB-xJw_M})={f9fq>HZp5$marhk6MB%LAx
z#dm(|54ilo7iqS;-kZ&sMFl=jIH}+#hr}35X*8;8P+Mnl!l8|(C@qw(nir%g^2{eO
z&IerQBuYB$JP22I^iQ=9g}>vRHnhq!UKGc~5;7w>dix{R08Bx%zAkQX<+pzW6^CeR
z{U9>s1IRp~UhKb$X(@ux--jx|1{X+YW2UoXC_If>#t9BZp-aLrWHcOca(u{X|CAs}
z_=A7;Uvl-u*9fB6!%b2$A8_XPN`MoUgT#D<xOT{}@>8>5@s81XX~`_qlWGmK%$q?9
z=iLEZ#b~XPnBX&#>CZM(j02uy#D!U(1VMqo0R2bDJo>92a&vVRS@~f}6`*9e&?g`(
zMS7L7i0a@fr95i~iHIV)-7dwnpnuY5y|u+sx5Icmqj!2nRBs>xe|f(6`ERmy<z>n&
z^^LVafEC#J;gym=J?;}2W3W^T<_g1~tF4VOq*{=cu9`ep5DG;Q$j7<?BvvVe)+qUm
z0YpGEk}M^HMvx=(jC_*gj3GPeF}VAXuu;cd+^N3!O4Bi)HaNEh5fX3Y;K+&`txL+>
zm*`4@Nk9_#NoRmm5mGt|lj4j<sF-$l3FjQ^JC|v8H*iMJ(`9hF!i%4*d_Y^Tbd{B$
zNJzP;a!TXKwWH8}TA9+?TkJp4NBYF8I=D|8L6j71bz^pfrqwTS#!y;EHlEVE{fNd=
zm$2O?UR#DJMpn}U!HXkBHPc{T9hb(DPiB-wL6J?d`G`(f(uyL2NaD12Cyo<Cl1wT_
ziu68ewH8}fU!b$RMG(eNO>M)<DhRvkbHcpwV~n@!e%x>JYwSf{jzU9GdbmLA0;@}i
zmEUtxVwFQ$iL4sT&r-!nDOqm_E~h;<bxMw4A|Rh-9KH1st*s5VzWG%M6v9bV5c&Y(
z?Biiw^6@$f6(XgdoLFjuAt0$IG;4K+r+u_hw3k+KL=5`_M&mwVl8}#jw3jz|;k9p*
z)Z6GXulkGrlB>!!PYJJrCsH6vjYTZHwGh6VwZ?00@&dGRUaL_QKJgQNxRY`Mly|D|
z32jw^JOwY`Q>YbN4a?>6G3%5}tt6L@EG^ml?i;jrwpsnc=RNEsBr5QX!MUlT?;%z9
zVGvZww3i|n;}Aha(rnUhbvU}eM_vk+y6dd0!Rhe{+8FBX7RAW{D;t-2{<Uu-Rp?zh
zr1JJX3nviqcpBy$1=SAaWL4?;F#%8suPsasbgDHhi^5NnWZu*9Yp<mA)t&k@n(&!k
zSl{~23w)sgkN({$Ce6G|ABQ%Alb?S~bLS1VzVi)Kl7I{mVS;qpi}I)-tRO_lkm)$%
z-mMQfI^5^@^q4fA@cHX4mS2C7R=0~HAc|^WycNgxg$u0j>~L_nOW0VV*;&Vwesj#N
z;iXj7C|A9}y9JN2T!DDpU!~&ZTR$4JG>%!}1zx%=a6*yPLIUY8p1FJNe2C4noLnIw
zN+RN#cUcvt04=JF#<P>Gu~@BnqHFMpQosSa)aa_LpU!6JvOuRP)|4bFz`1DAtaWpo
zx4Nqz&nce|0CL`=?C-!bpZvLh@8_HUUib8s#V_+X2VofU;tShc6cZ*NoHOpF7%ec`
zGCuCnyZwM@sYSOI6D4(!0V0SH&iI6^VzL<nlj)RRe@L&_W6<w0n~bSPhF70kp}Dh8
z9M^FL1YtrHl_a$W1P&4UIT&%RPNTg-5G7TSQH}nU)worB#y#&Fcs$%MW)+&BSTl#Q
zj>15${rKNfm(_4IyEy#PQf1EbF{w|thSplz?Syas&I<(ZXN*6(<7YY`(M~WsA9L`d
zx2bP#(7bWQ2VE8UOUbKj{pSn<L=d8lV0b#<?D%6&j*mF+pVJ6SzV+G;-K8$etE-qo
zpj1LAbCS4DtzDzmYLOop+N+ldlZH1ubzWa21L3Lv*5H^oZ~xX=7F95qudwh5m35Gn
zkZV^>AS^MqL`ttosw}*G6{o~wO~zBLp<`pS%|HB|*O~myh|#SF^ZE=eV0JR$;a~iS
z+QtUa`Wn_Lgz{`rXNu}Q_%)PKjMkQuhsPWq9B_Pez*)b?Qr+?ozJ87F@(SxaTi8+&
zh6y1Nt!A60&1Ga5lN-y*`Xz)6&_#)J6)UQGS!-}qV=m@%hVVf};CsSUNr4n%L0p)X
zP#A|X1w~n4tVSuj=wCYL*H5|6uHZfuK$Mhx?n^g$;ld{A-hEDg`ZmgXoU{bWY02Jq
ze?qdo%JS#G05bBWty3TZaFQU3uugINqlbL=d*9_}Z~mCw{YOmG84~zk{j*=^<>xnv
zR7Aa5!<I8<;~}Vm`cjL2@0^pfGgdFXLJ-%yv#OZm!=i$v>;h{RwrXHmSIrSHk1yk@
zs-R**Sm!GcT|r$=mk1&c8#&MB5z-KZ0-+z92|mpeMghP2yI<z!&L!^uhXJ$mGxk5a
zhjxyvH{fhC^De7#fKZA?cZDE%tQyzaVhr56eTR46c^~H-AAa;c)7gYh-3QOtUb)Uc
z{q3(&j|8oH3x^|0LL7$C*pKF|wc9MMUBEd@m^3Tfj4GR}GzQ=>^M<YU4KuSa(X;bb
z;G!>Mk!TT6XiHX<h_GDDK?nkW;mwH+&UsquoZa@zKF&FkddSOPe~uM$9!<_DvXp#0
z!#FU7VZ3`pSdWR8Jv}>VdH6{XKy`49gI;gMhqrF=&O7gL|G_<ay>p5(XJ@@bGx{2<
zojQ#;@ff=)fC4`&N-~;GA#T!KT1Tnq6O1eWk-Yma?r|SMxN0buH5DW}|L7GAC4FPs
zIzeFsWo0{ItYJ=@l@jkvt3cuVeNP=goTIbU;LE@96^82}ch1kqk9q{vC7_*P^zfLy
z@BN6&-8NCHTTP;@j4_SF&x8+|jtf5c;1)mn@eg?agLgSTJz+YVa^u1VS2q7K>&qQl
zNt-}QY&oJRhQ##{kw}h@k0?aIg`Ml(m7%IO5VFc`9#59Ecu}fw6*M((h%eTV1(YWx
z<i@{@F%;U*^qHHwN=YR$wFGlo^;169Sk^bYeBoEW!fu#y`{abQHzu5me5GP=dzan6
z{u?g--tUmKyNI9)AX0+y5T*(v2In*0_}LqL_xpd%M<0JgZ_p#lGhV)Sfmg5pGgg;7
zwBj~GO7ht$S<$1mRL2U(qel-AwGL}L*DJCGb7qA<Tc)B5IDxTv5|HzEp<akItEck0
znnTxl35D^heLoUouEP=4BoBj{YjZxM68<!VB_IqVlI{}Qzxowsdk2gjo?wLZ7mbu;
zM|}>y`y*-_n{+y>{*o06DI;I%3dw_ycKL7q)qlgEfA`Or&c>C{&=Ch-|6q;dtarxF
z+9g_@20}SXVc0v|W$$nw7bd)T;W+~31!1<bRH%yB>X1~K27vUils2aF5vpuI=9zs3
zuOzhu0`2Wgv<Br>xKau)n9&9yeIscBX+Py7grL)H(Op~Oc)5)!GgL)a#5&3}qZm)g
z`$J?HV0Cp!gaRpipC~ICjmDguo`7>4pByonPRN@P)>`_*RkET)iISokNLA>P!jz<W
zL0065Qe&LJ*~flYpYd*)pUy?|_wV>jLI|QLre1HLf{?1s2Fz#B&l~a;1@kH3g1`L)
zfnsfEjkYw5Ubw+zG+~zQV@sdGmXjINM~C!3`k31CGWE_1BB_L*fmf)H2P1y*&U^gm
z$3NtQk3S%tO~^9eY^}%M534BsTFJ^BR7gQ-G`+!qah9>P-l5T2Ud*I-Puym6rTaYF
zbpoVtPaN9E!E|0>&Zp9;D1jhE8%H^}qZf15nga~*I1_y8ZRj9Sidv(Adg(b<zxWcf
z@4i7g$*IQWU`s>qosUU&)~T;<BHLYrP@t;8a4KLr&UoWzZ}YGJhyRYBzxO6(ku6+j
z)7&%s#*;D91}t~h5D92kGBX(m$9o)|9<zSsWxC7jRWIyuzz}Z!RvpIp%BuqUd{8xJ
zVGg*cn}r1FdtPy^MrjOXY0zd~HHpWRQ(=$_<Q&d^hDTLQ&hAS$S^M%UO#bp`sM1jg
zFv8&q%jr+wrFQ88jqVz<EuVm_yn*3(kn-KX`ab{SU;lr&^Wfvfd&F9_%L&4O;c$p*
zMXYw$iL^$$lGK$v+Iz%l|BP#|ev@{qTOkXcJme4v=g}eS9J;DnrSt>VO8wZE4L2tl
zJ$7mc!vIyUk!$~ZOI>*On)H_NLVAlRuj^A4D)Y(bGkYLD!a3^OJ8XXQtMtzPgkq9c
zU!wQ;8Gm$_!#Ce#?aB>gqlH_HuMOy(^!ecre!zEs@Rv+yW3FAhNFXHp`-c^NY*^b^
zXD}FYfA2o^uz}V&M@I+Td2p9$p0TuX4T!15G1_Rfc9ma`TQp@CW{wW9a<S>=)IB%1
zM}X?!25f9?(g+|g4P~hjfvzZO-g?>SsyQqK&VBa4#=q|oLXb4;G*_2c`TR>n$7`JJ
z9aE+mBFnMk8MA`}&hiOCy~ft(zeErvsJ8cH>JNH6xch*QZ{Ozr!;e7-N}C}A$8a*n
zl$w+C9uE#q2&AMQ9pi8e2R-sU=cGR(%L`gdGdA-E)>@u_J|Ri!k0CE1Kq}8zvsDIT
z$|tPdS-@>lN*Z;~czXWD7fEU{fdqjisU?2)g3;*G`|*@{UYX$u++yg}FMojNrFWD>
zgt4T3^)j84QzpX^<@vDkFf)|Hk!NLHxk6{>BC^#%1|il-@-*e};Dq<ydzU-+ZZk{A
zB(;!I3-a76bEj!Rt_@kK$?^fF)J(=>di^ub`sYkCO&N@+uS}^omI;!_{#x_*=_=fC
z9(=t=*xUnco*A-$7HUbt#S1&s<ABg(R8_z+)qb`(<2~Ap_9Df(w&m$JqW@DWNm%{T
z7nz(4n7;i1WodBE2TRhNlb`;a?(;WU-r7OP(9ghD0cByhd-oB4{uh77&)<0y6~eXW
zuhTyt&>swN&X6Ptos|{(XMKjF2}($2lQFvoyX>DFq8)ThgcKoh<eealuBfKW(FCdl
zM0lSkE2;r(b4scoS3IW|OQGnrTeMp(^1%stsr_7{MIGqjT3wWAV~{eae%doURSB-*
znDwuHmBH>Qvk&fiw`cG)=<$rhAO4h;=bvY3eFId0klsPK$PIUH-RIx`$scm(!N;^(
z4K82W;ppIq@pytYhFYV}^7=Z*`^OB2BLW#P8uhvJ;4UX;ClExGE+Gs;q;Ql^1_i+r
z`^;exSSzu{Rjjruk#QBNcdiqYf>x_ZySd5iY@edgpI~tLp?wBj6yDcO@fezY;`!_V
zA}X97!P;xDvT^?b(;vS>Y5iPXDFq@;Ir+tVOn?11X{>H|!CDwFJRkD$haYq6;}6g_
z=Zjx_iA$HZ`T3i_Ak7M-5WMr=18&{f<I<%q&inh^-Fryy?2OsWk5vqkh@brY9qt@j
z{<DAb2VA_kL7)T%v&ifavVtsp0MW){%L^qnRmJ4IAzc(keC_qGu(7$#aM<T>{^F0x
z@{BMEu+De`w^>G6aq>and(+RBMb8_^DvD5*kNyHxb6~Z><T;`;kn^lhv0#!Fh4=W$
za_^=)_X>K_tPe+(d3Xh>zsQXbxbqCo`gM6M27L1Kd=gq(K!W1apL@PZgCHbM60qgs
zBJiKded_g}+#nG^ph#9%=v=$RxHn)nnc-6Z{kWneJ3eKye?a!!koLwlLMn7#FdPgx
z*xTp$<bXJlT<UINogXYUo6VTcavts;b75<PQd=hDG2`)=*<?bR&Pa=b(WGFvXE+-x
z93k6V8`NsaqR+&lSzpZ*uV$$%^h8p4D8q*uq?B~pZC-i#6^i0H@-$^MI`_8Q+V>E3
z>2(ohmZP-+sW4TA`AkbFJRFp4tg-d=*BG3fl0G{29&$h+1;x=R$M1c>rPsesSoh6d
zotGT$9&qQ@EgtRNV`ZhytDk>?EKAwnJD?~HXT2#u{OJd*u5BO%9G@T2J3nVM=u?&j
zLP;j$l>h#J_+yUF4Zr@Y-=foLfYcRBZBF}EsKC#JH`bwzhgW^o;JvtP#WWHKR=QpO
z@$dgWYa84A>?hwx7b&_ZP$3J)=ps!$yUiF>U_Gq;j9X9$?*)}?th4hgU*Y`Vh_pYh
zpkF~CVftW?laD@P>*X&HNxz29SRUTF&#jL><lyKLTbrxA{?#vXa(v8WI-}5*v$HWj
zee**$H+N9Am<M}%oS&aD?Da6t5Y!^h`&0hKzxp@y28M5Z{Y%7g2u?5V0#PM{etx_|
zJ8v}vUJK_CKFM}Ixt2(_H@EmN{`o&=Wp$lD|C4`9T4=2HOi^nrMV?d4{JI)ranUD5
z+w<qMQ*7az%*Gh%ttLC){xT<z_UY}O`U}P>LMfRXo-;l?V*TdJNGX|4M(jQOi1U*p
zk~rWSUw@77{DWWP&;G+-a(X&IDoLy)Z+&o&|K0!o|ID>3JG9#kqBtf{iX;li(iso-
zPI%*;+oZ+M*xx_opa19oQ(k@b=E5exvz8=5wYkco@IofzaeIUyi0gz=f~&HUFbLS$
z-sa_(zJQGDOw)|<Y(|_!7_BLbl047J(hRLDcU5hxVDb#loEM%%gsnQAi<{`8BtJeU
zour@*c9vnCq!^6}>UCTYqQcNS#!iMb>j}@_yn(fjyZ7!gnNG+m$Kn0M6TbW7w+NIX
zj=Zl{md+@(;e0ru&?U9nkk0;e@m-F`6SlWDX*Iq1q6|Ev#w`Yb*@{^g`T!Vsn|q}K
zg3$krwbfNmg{@p@gFs=d^G!^n0f#P2%EIHRc~SZ~4bPDNS?g^S2o;T;O}c|2!=qEm
zatOvEtc9##c6v_QA9z-abBsrQ&Q1=PO-3wtTWnukA&3Hg@r(B;iV_@5rZax<(|6fB
zJYj9QO%Oh&&{}7iOi~U`dJM(|&GyF}pPuvWZ+@Md&t0QlOBPP6o)}ThZqX&qnCh}~
zNEr}>39|AtlTy-XG`Vu+3iU>tQfuzq`3NN;uc*9!{)?r^i-p%%o@EPL@TXow=Yx|F
zg3h&zY<}q_hJz7qm;nNbL}|<9{vO$QN@HmiXDw%^`|LlwLz&IEu)W5&ZhnE)^-ccu
zfB$dDvVu5JWO>2A{5OBj&)#~EmDMGZBq56ZEcH-I2BQgg?jP{M-F@DA?=GjOr~KaU
z{1#iAs|(vVsU)FPSUlyl%yMj5`o_9&$RH$4T1XjG6jMRH)?j71%Z=w=W_4|gAOHBT
z8BNDD8?|Z;3yLBq%W|(H(V9F<{RQ|8P?mKb?v}vv3)lS`{?*U?v__=}ghJ*S!;kM%
zW;sctgSCd!qdj&XeoUEVT)DKxSHJZ-Dv0>$Pkv5WTH-)4OEdoK|MwsA-5<Th`uZ|S
zEg=j;BBcnF;P~u__ijDnqq_&Z^Zp(Fi<5uOKm4s<qtmHVI#<C>l0ZaQAu-xeR)Roh
zjn_1)h%jk`6gchIu-$5NVP}U6m!BhPH2L|PKW01`6GxscD;(rm?#04YPb$x{PodI(
zCr0Cw&=N;T>%tBfU;iAl<L`Nx2T3dyA}#5Ed>8$-ZxJPR%5ucv!6T-VDVHx@<i;zn
z6DKVOgAqv_V3h#tNQ;t>?(K8$;lYAAgjA>?Mr*@pnl6MnfA}B%kSw3^zy2@(m%RGw
zOTM9_K*#_m0*u9vj`VF9v;h^>=&WoK%^UD|-!A7IMVWK;(hdIM@BEMW(U1O`-eADS
z+G+)xcoEZlJfw3OWtl(qhJ!f^$`Leb#8$I(;}RIlz&i51BV18pbP1(ztjj2&j3QJV
zG8qpU56@}0YHV!3Oj&A<jt`kmC!|?HSrpuTu+Q%Pv7exq%#9g6BsH4OD6K)EQ0hI5
zF(|2+%`)2U7M)HT2dL0%9_)PTpw(C)3?ft*EiBH1AR>uloV9Fk?Xb1EO<tNRFlh$8
z1M<9h3}KZeMeZ}`yeP2MtcPb<#l<UXHR)W~X8Gy``qLc7Gr+qS<flE-({oIbA;XZ#
zXvo>g5k)@Z^2Hrm-4?SfB?<$S@(jDOG#s3qGa656G-5ydT6z51SVx{03@0gRRuF|f
zA9yNg?ZxMBP^;IFfdmw`N@PUE6hwI1YnU`i>TRS7Ft%L4s8R&1EU%E%TLfXu;lU&F
zqO6*jwi@MHDT1vfPrU}{*%*Po%CuG!#1U(+yg={aA?Lri?JFh(fl%Z}=S=nwSh;uw
zp*$vkbhL+cf-ijjRbKqu4L<z%KBhF(5+4w)ahwiDjHVNUN_>kHSUmMR%S&dthd}N;
zyu<(Zzx$V1qxmQQ`1jb@T=)GQC2-Q~Ph4pbvP3G-*8lfLdzDtZix7_kh?E|##=)yE
ze-W)Uw?29YLF&m1Pn<!mHMS_oW_}>r(=MU&yi(8Za~74vT=?p1oZa0cyMKa}{`>?;
zX1m8sk56c?Z6buC-|um9yh{={`NlWD!dJfjI)D5}e*wnPsE1%Z%ye`<qCfn=uSeA<
znm6Movz)B(N$h*K-sAuF|NOtBb-^F}<KLm(YT=yrURN?eqHvVRvOp?<cK+3BuhHr(
z1HQB7dqk3=%n798YhU{g&ceNWAF<T(1JXng;DjfH=!&wRr_)dMH9s9ds{cPXH*?OQ
z!kEo3yu#@Rw@GgwV4Zimlty#<-UsA^KArVVMu!hMdi0RP(+QU@-e7rYgYh)u>Xl2J
zogQ#{()XGL;j;~+D$1VqmeQpuDtKTHmw3$dufG4+beEQBHe1}hc@^hO#j&okROc$X
zTnJ%}?&>DBdK04`H{PWXXsyYzl(Hzfe&spJyx_0?@{bu!MimWiZWty!hHnf;dy3}M
zK2B>~slA=JQZ!dra1M&o0n_CUs>mUofzg<=6SU9-jF4f7J3ApAoD-^LWUEOa1s8TM
z(rC1pO*2w4LkLG{92U)dXr{FuO38~7;~Z81!f*V&{e6TGT)ePF6a=UsV0mq=`jc}P
z-~gfU;zi>jnJi1mvaD*xIkGIJC~}folUl8b(It7(Bv1lt95}OpC;a9s7x2R`|2$rA
zQQ4KKAfUdw%*yr4<cDYES?;qbYbp8zvf~r<a7gG3)8jqP_8&4%4IA4VEO(cQ(h1w!
z+YI`BbWu{4zWF+wrcAR8Aw2Y?tJl1r;eb*b?mfCs^0Symqs~&N!=+2xBth&a=FN#3
zQhFh6APAEhQPM=p(0A9J@09pa0~(_ZwYbjZOV@pMI6Osb`@|*Z*I$>Vr<=Q{zx;rO
zDdD`2QeRzV^K&mS+dagdO|VEJsYr(-dbe+L`D<T?bj-<}582;+hz#0XyMBc@YS3!5
zxpwU`{lPieO#5K#3}rPo72z<2p)>`iYJAV(ydVhJKiK1s{`8MoUheW+zxnH|t}OXc
z4EWxXagR;=Y?u(&TQu9tNS^R~s}`Xw3bJ%c6ve#q@)wy+XY4(?MV=Sf#+;BKszB`5
z@L5S6bM}$fo=KWpTU_|^E8IKyD^GxNl28UrPEP6F{+ON5zlKex9NzwrqoaM2c#WGc
zT|tVNPN&PIOIv*S;VrE86^zve<0`n$`4IqlHuJF46B?}`2)O(39{=sX{#Pt_yZq*F
zeur8u^yh!xcBoiuQb{B=qI!$&%DRWfE2Ke4Q4NYK$g-5x^^3gz<!_RvDTn)aD2jq8
z3@c+=g~<zFVHf4!PvdpWM<Pbd6_ue3XsxcZ`Nfxbw11AvGzLK^p*T2Z^k|nW>mB&{
zpRxb`FF4jQt@<)&M@wv6y1?)J?jJBqQ{H;(Ez(&|t_x3rT$q2&sd+vsnCgpm*3xP=
z(52=FKmGw%FJEP4Wr>w;8>1~wXg|1028b{Q9Z;_~h@ynv(J@J_j*8<2EmdnnnoT)5
zJpdVU@zQlZ|HZE|8tjtirN02HFH1_fFcy32$AK$LTt&H4K}d6b4WSgpX^+uL2YLhS
zcnn2Jd2&Edj;Wa%GF0f3A=CaDjb@!VjIlD{(xodbb;sPl_n;D=X=YjKoo_34L|JJ@
zwDE^mDZL1=_a5Ho?EH+Eo`0TZtw~&q>8`AJ2xeZ^%b*f)DqIzevMi%0^2Og%6gjim
zm`1Zpv)MuGf;^j4!YGTgR9-~x@krjM4j}m6l`#tpS4c_HXwbR1&2Z-tb~MHfCcY0e
zol%~hqmTCpri${>M~rswG8A<h%?3*=UAjxFy!7HLOvY34yre%ER*Wu(u)aTKoqx?m
zv9@YBdd8&V;Bb$(-}wbw>zgdKn*^)N2sLNHfal#-^KpV0YayLXD_Xl*^sdUHU_2R-
z)LS%~OKfiJc)R~{jMje4@FI(5K3n}6mT<vJ3O%1+O5(Ui_wpr{u3um{yzOD%Kp@MK
z@!bb-enQf_NAcksobUIbw!-T2GNVC<>o;y9WWafEz@tZxm`yY90wX+a7B^>E%`2bB
zJvJo;wK!()aF6f);0J7NZt?Z6et|fNu^6oIYbYw)b!qW4y3>?dF(IzkeSmS*SvH2z
zc))q@j8?15i!Xi-U8dET0pp=hKG}DCmdCmItfL5HG*Wt7m9-ada`N{36!%Wi7D6SE
zS;64eZRB)F?cSRVe)?A&jWlsRXJ>nz@o>zyzWp7dxJGX<q<=mj%L<R8IDvHYdV4OY
zoF~)Oovb8D5_A9IJ^u7B{(`NoExz={7Z(m57K;_WdJ$5h0zs#{%w#miW+nBdPIX2-
zBvqCL{lOXa)-v@*i`QQJI&YYqEH9|VH4oA8q|MT&7xA;3=%pqM69k@mE0rXSB0ASD
zvvl(zlXvc5oo8c}(=q0wpVDZZaMks=p>kZZ#L3}(T8*e;7+l~tfAgQ<oaCoJ`4QUM
zh0^gcDR=&bJ*IzJYY79z`r0zXL7$($^(Hs3U*#L$_!1~@lkJ4U6ZaLSOgTFrFdom)
zvz+d-W86C>sU?fq=99^oa{m#CgwE0uTU(c~dP<gM2!uD(n>VKB^wekBc$IK*{#l$)
zY5n>>p$8CEvt25=__YucB?MIBKU)PrSBV@yX;%434=w*s7QdcXlB)965{qn<Cp+e!
z{Tb%0pug1|_Z7zD_vN$Z|J088<41s>Pu?G(mOB1(Id97zx7sXSxq#`9$Yv>Mjj|=t
z<dpdEBih}Ft@MN)nWII_;PjLviCN#?Vx_yz&F5bx%?c*d4@k4Z3r;^{2|htJ5P~F*
z5W;bOcFIS$KH~DF9g<p&TC-7U(FINhI3>uljC45mzHLqrL|$J}mN^1Pn$F0}lC03I
zuKKJ_$`FI{0YuCR5pDrzJ#`aZfHMYR96>NQ&6Fq^khI%$pTADAe}*|Y!#PWc!=3FD
z-uVej`jo3;!d+z<NW<aYBkJ{#dZWpui`V(~ulxi4;?MtxJGbvvP_uYk<vg*L&OSlV
zu$E4{iM5XV5AX7eci!dl#Z4|<x`6jpbs*+bK^0jxqde=Oi;_qNgn35k=G10MnHHR!
z9-&>x#)c*c1D3k$SesE6CB}G`rK_NdIZ@@QAMK&wFo-FN83M2Nt}QLG`Q=wRJpUnf
zH1pJd=O`ZCC4B!cSsFj$<%Z$m)RD@P2Y2oe*Ze4eS6_XdSz7StfA;Tq_+W2=AUs~W
zC)Uk<at&8jy5vR9hqvD6ji0^A<qI1uuP#5HeJ&JEMmPc{qY-&|g2_v2QABDAN@Kj!
zvQ(6XUqdJAtgUa7Bn`H=ucGyovKkMrOTD;beyM>{&KS-=d=K&Riv(eaa~9(a(kWD}
z&h}S6$NtGV`uK!qthn~vIxjAzgrg%?+6}(89Ff$QxSKid-?>8&3677?2<uB+zI=_h
zfAJP>IAd;cEb#+(9I~SLe$vs@;b=9RG-?Uw{S!XEeUIl~xWV%3s)wCKb$EsM-O93*
z@o0z?A!aDq-+#c(=UyO*6ULJ<kM{1Pf;xl2jC#FAUgX$fOt%rCBvp2A$#YGfX^j3f
zb?xtatS&kJ=}%bcogpr5A>xoIi4j85+*xN;zlb_IK<-O|DB#AbrQXa*C<v@T%al>g
zF`o3e|8a>F(CDtvsyFF&yM$55v|<Y_=Fo#5>p7=<I!6?G_fl(}?|rnIoShtV+d+3}
zi3=O+$SA@^b^qsOW&2V1lGSKUf6%8W3MP{wAlch{#L@8q<8exNd5ANbqDTpqBnj#$
z>F3UuT9fC7JTr^R?w5XqKxY|)+qYR;Uh>UyS(WioM16Ce;_`qXoe>OsEG;EmSdUqi
z6B?yrD>PhhhU8jukWD#1Iq}Wc&Js)QCA!PYgkf0G)nTC%uV~9po)e$7C$!pa8nu`(
z^w{>%!4d1LYb3R#>LFD=X2N@LS?#N~QXBfC9#IrimN~{~`u#qo(WGgC5Z)f7%%|Wy
znL{bBxHrb4wMCcqsWWIHB*l2d<k17_+Z&ZSxnkxhMPqA&wa?w;Y&b$>CCkeVUcB63
zCEX*DmgiRLoTrZa%?|fYhdjFX5QHOWwAtL)VRdbtk8j`gf1jXotC;@?S=oSp@;eQq
zm`<yK(VG4JN8G=^%i7vHQ7!hye?n9Vp{MbVM<a~3)N2i9gEOqNtgf!2l;q^}gmgD$
zZEehSnj@rSJnFO5kThbix$?vYOHpWw+&py<7D_TbJz}zZpQu%5mI|yjC=q~^bgo^&
ze*T=ZAO0NSENkmcKL31&<>^BL3@@*@I8CP<Y;W?5_wSP=5k-+BYAr5ayvoYz8h7tK
zs+PxVTeP*6P{J+daLga;94b(3b=S$$oO=)Ma`)aN+TCtN7LqvOUqUJ@jzPaik!w0j
zUC#C&(x^9B>MVhP2M_Nvp3d0bK4Ww9GD5(3JYb_6Q;R}>7aHr~-l8PW&8LcxPX`cd
z+5OAE=IGK6>gvuy87h^>$!iz3x$tXWVf?-C^YRO8y!hqowAwXXWd?V3z01kbIm7NH
z`hy|6k9OHHnw9O4C=7_>7$rbePO!Er>*ud0pVgIRp1*mMMpC1mNTgE?`#nw{JYZ$1
zgOke70H~TmVHB~kv5s^;D1H3#N1UD<u(Rzwr;d+~$nuoy*EU&SS)wQ%!~U31YL8{g
z$JlpSKGvQ*?c<yyJ3C|6YLcvXy}wZ$`wK9MNxCh<v`#p!p~8SxBB+Iy2#a!phR}4B
zBQ0~%%u|zVtqwv;;wVC!1YLTwwYdQuJf=Y6<}hUCz-cW>6p=(BWu7w`jYxt3lTH1J
z50fWOji{RAB}G{>9uFzXf^;@R2+4RnqTe6T>5RMuh&A2|D^M5{gR40F#w}h?oBD|=
zfSf&e#QNG2Zhe`cUh~xLKoPVWv@dSbrX)TM*x78bePx-YmO@%bHwn4eiWrVH{me1!
zpHr3v%Xv;vOQ_dswAyXbnV%sMRCCDv=+h_Q4j~A`kXEb3^71N6%{uiYB2<#z$uXnj
zWxDkMk|r*QJ^x=xqA2vqNKrD*QjU)gD#ujCXf$STZ_kfQwvJY_MNt^aBBN0Y(6tDh
z^1)+%c=a;`5Gl!eeIEV!pK<kf{{b>m6{O`ca1lo=zxq64Fd#X8$SYsG%+;56sD)rl
z%TlYsjoOrn$vJB!^!g_Rk0k3`o79(=Y1A7;QGh9ZM(dadU-uYttM;Xml8YBEu(fr8
zMokh5i4>0Gy?rLvF0j0P0k10Zx)>=Xt#*@ktAR6?(P+&6{zI=hX}2kglD*wM;v{B!
zQxHhUaL{Ku8Pbm17(Hjg`Mp>c#wUVLyM{_J-9O~sAO0a*zxNNkI@{R=N?-%U<~Lr$
z_0CvxBfj|cn`~cQ_t(F1EOpwvP@gcFPFP-RaddKk3M6Y=4wN8CYD8f`nQNqQ9-^}r
zYy5n5`vgJ|1_94K_Z;i1>on`iTZ>@Xy>pjWZd@ejEc-hY1C_U5r_*M+(+210_ebpQ
z-eYrfhfcfAY&vE4(H`p?o2+&_blVM1&(0~c6kLqa9&!aisSQQ(*@@t1cwq^q$A{eh
zcmI~v-}_A(m#z?ovVwqo1Mt!pUSPGGv(<7$vA~rYs|{8g8m$J;56(C~-N&RO4*Nao
ziNB!gwS=#}{yL`kKGW%pAPA^8>P)98=lyZj_>xHLsg$jDi<OlPnzfjEVrkb`Ks!!<
z_E%grCy47`gryCb+sIc|CRUf6pPh2+)-M>0MvTT2f*|1GqkTU3;C+M@+`N93yC2_X
z?{J@;%}uWCtoutu`$Ow9K)(^6ws~o_ruW_lES;XyaGH2|iSF|^kzqs<3F3CZYHfwp
zOCC~ch5kS}?ad8W0?P|=i5Sl~>icPDN7D&f1}xQTw37t4yuyuZ*D1<!!O|FxMifO!
zR=H|68!a~0H?h{Txz?rKY@o`5ynjYsjLF^@5-zWk-29S%<MlQ&ZZOVgWV0!g>6DLd
zy~kuS<?!ePfYZ|xdgtfFVMJWbmoJN)B#!B}6O@vag=cQ~4906wo{Byxd?GWrcc0@{
zjm8&VrFmfo)u<drouj$ZVzqOL<#EVTBB<3Pq_to??AZ+^J8s6Ib@ZoW&ZlD}Q17&9
z#1WTwcA%o(&Rqfq!!c=IpsJ$N7)!lYqucGWzPe7knNUw6T5*%ScgEz-dlYs^xN`$K
zm4Ix1K?^AI8NIV3_V*u>W;xCY`u#p1-@aS*1QOO(Ru~M2<aq|&E_$iS!p^4jBo3p^
z(}VazU`x%xkAFne4r#ysWs>&N!X4HcP1J7i(yxA=jb4lOwVEeFXoIaR!nZpKy^|h$
zMa15y&rw;DgdwFaxq9gmM>lWq;NAnAb;LDaVGjml@={k;^xj${3_~_HHdtL<r;#|?
zjTTW*V|={F;EfsWmtTeL8&F?bFu9}^7+s+AjQ!m^Ja}*iCnTrmXB-_K@csuMA%N$v
z-yofwarf>=6h%h2^%B-LvCdJJ)+d6-euAC#G#k;DCA&ZPKBg|{eEn+#&6XbyrzE<_
zNY+;P%I|-T?R1~cQVmg-SShhZi3|cRb!(g*9r92o?42AT@dNw9bjB+$zQnM9#={4X
zaL!U|)R97Qa@xnt72rZ(j3bI7*49>8>2A<&T53^)Bx=%oc$eY#Ms#2PGH&|@L=7(x
zty~gxk)w-@ySLwEclSO=C&wHd9&mJW!cTtsCaWteyl~?>r^omB;G+-Gx@2Sdb)0Kr
zjq_)@G#E42()@iatSS?j-rFVn>mPFQ_dCQpwd&6af_%!Q`joZpCg=i}my~5eSc`E+
z69fUPt%R4;F^4t9Pmg-sdgm=RE?%bH?$QiueC?}WB?x2IH+MKZKH}}S-=r{(R;x)E
zD$2a%?7UAsX|vMZqS*jvW;7Zd&W?MW&ko7nJ|;ffL%seF5gXT;0};XrjM41v-sk?q
z+Z1KMzyISuC5~g#ENAcVg#Cj<tkHy_s)S*}TTmN3_E&o5)ic|1IMA~h<z#}+QdC~6
zmbU771d>>VG;0YU2#o<}FcniX5Kxnjz-nyiRq?sY&?+E^LJ$1~0YQ@Z33_P;K$hiz
zs0PA75C%k%w?2uZm^hA+#@lFHU0|mpT&+<FjOK6Ae{O5Mx_364GM!Es4u=4Y#}lU0
z8AX|6Oo`F{64pw3&8GNU-H>Mjo_(+>1|vpyA5hfl$T%U0V{jmCNt>LFnxPZciKN$r
zxU67q7MBEq)xdFC6%1_2gK5f9R#0oV5K&AkY4H4w=jg7k5!Y&*pZ9pQyGxpR;bvLn
z%qBC!D5llyu(Y&HEtbAnDirBtKzcO5hyh_SMO=Ovv9be6W6|##kB1x{?b09gI6EIO
z7>?<kpL2M8LbG0DG#XLn8OD@oT~d?<W$CST=1}TfkNHdpzJOvp<>W_iQZ!m@e(4Li
zTHUX%wKVb}&vh)#`Vv@+nPt^X5Wnvm^%_@_lzuVgP&jsXA5o+!t6SUD8ZB;KzRu3(
zHtnTl!Z_jH{kwep@yEnjL8IxLu<0zLC^W6+Dyu8&L}5xjY9L+0=wy%K{R4{WF5&ZE
zM!xcGTyqsMXRSHMXxQh@-H$jp+ULAC;qKjkORL#p|LBD2c+6nXr_*T>1faF>O>|>1
z+ENsAvFl?Y?$bU_2=p}N_y<2jFD<k6r7tjdzIDd3G&|#F#}YJJUN<&LQE}u^0Ar}P
z8eAQuoMjn%LUM5X7PIk~_04U<q|S@ip6BY7t8|yw2%?1d-v0%&{1#0Ft#%!&4U_Q<
z>jX>fEml|8h{74Msv%uMZ*su!{t?aoBh(kagM9Iu2oZZ(LS;fZ8ua-1&IcTyo^Wz9
z<fB{vl8ucGPR<65##3gq8O>&cB#9{s;b*8<RYzH9TqWaLIIchOd}b5fB9Klp`QQPE
zn{QF?cAy3d!)kg)OXQx{9)-v##Ehp1V^C2<F_|KRh-O*vokl>pxyDDQBW}I-ChhJz
zQLRPNYVg|YFY)ECf1BQ5#M;Uxjb@uDj@jFNNSaQ0@7;GuqL8E}P@(1Eu!qfS3{DP7
zmUCnoqfQ<Yy#1$`m%oF(@PZeY0prn-Tem+%2*H=W^cBLW!P{@YK_CU|Yinq&8I8sZ
zmG^KsVltgltJg8cErfSZHmIL|=YiFh*+~zhE#+v0j1y}4lt!5n*$iCasg78L^K%EN
z5YT!QuvQDW(eSX!gQ8$kmYk(i&-#cXl6oD4pjOR$4pvsN&QTNv#u~yPBntd|f-=t;
z8$)OcB5Np%5ZWyWL)^h#zv0&}LkG&)ge;w*i=0-&!w5<2XO1LsL=s0><H+&?qfO;6
z<>!7YrM#%H)D&e&ny1giW^;fkHDzy1cIN?McZsOoBCf^MO-{QQ(a<xLEkRXw^QHCy
z%oqYC=_Daniju4bwy^9@hMY~tm>?oZ5?bvR-A0p6cbP`gq)~4nlp+j6vUJ91G$fs+
zM3L`(=(41kYD$~Y&O9dXq7)%KjAM<#u3Yfe6M4pH*kds0(QGs+OPGwkoqVlUBM1Vr
zET<?-Or@Ji(~P_*5i+P?peH=Vo{5EppBUMjaQvgUh*!EqHxrC5so9*>@iB2xqOC^N
z6Od4h#zc(<z6=yZN^xDzFiQbHG|SvMKIY`~fL3P>71!9lbb+sa>8q@5UF7zi+r0R~
z=V^DA$n%`D(_?zQb8g?dMI0*{&46J)Wj54EJtmo+5!MA_JS4dH7T6N|(ywCMtDe=T
zHIMclvVXYC_U0B>t}J1l;L(G-Y;SIGXdSaOXF8qY9F(PIFc{EX@|BkP1o>pn3j5_B
zPowdTY5FM#-}@<z%{8K{HH<E4+Kjc~IWjV6XHktBC_z3Ll62ah;_V~~$*W>UwpQmS
z)OfJ_kkgX`TAfv7lyLFt6~6ZRH(1@gz~=e|UVZs>);BgdJ3ph>JLUA`n0t5b5(dz0
zg`6IbiG)R`BU-aQv5OG13DL(t!RP|_(yw5XCc;{b(L8u~pM!&4Zrpg88#iibYdP3^
z#B<MG<M{BH*(~)v6ajf&a(>pM)9!ewg`N9pxzD@{KXagzwiS)F(CoeW9;?q^WBK}X
zG?imZPjHi*bXJ+tZLT6CMcyA0HJcz5wYuW!<uykA9`&_?<*>tdA7mUH9*{Xpq+(_V
z`#gC2=QP$gh&o*^Y;VzSb=cY5rrxM={pwZj-@Qedk2yUYFdgJ%$9GWYcc}+WY<nF^
z4SBZ5<cI$T_mzKy+PaF<nv=tQ`n_|$_QkJo{l<%oheOVe9&zp375>AY{-uYgkQ8Of
z@zF80q)wXV#PtR!iPYXmalUZRaCH6Lhpb?H>jBYHo8>DPkV!(zO<9u@Vx1#Q>KR9Y
zpIL5;(l@_xM2#4MWf_fe4t+S|IMduY9WyBnje4DKyF;znqSNgHg6_sPQN6x!BhJzp
z{nHcw@8b#MG$$yB#AQyi9V6<?FzaLTk+-|oB?v(}IAb`T;GAQn(;<|Sv(pp6k<Vg^
z%qxB;lWFzOF-ej$7>|7bDGGF{F=a`XXOy~pCJlo_;7Y^v;EdqjBa-eC@%9=WlhP_?
zC^Plq#V9}qs`?D`c`1@cEn?eh0xWd_kr3SKjX50j$(*E~B<RVMwT%sG?KZ2O4(+8C
z>dmJ2%9>3%KRaeL=%ceKv+<N{G$Sqxnres&YKUARCw<I=_t8b}J+PKHDY7Z)bOJ_m
zVfzx@PKgo%YYIBECh5!vi!95TOlOS8V}z5;(hQ66R{lkaE=$rh{iLJnFTH+#4W$qa
z9vu<9`7W*XRg!vxby2XKkI}hC&t|AvOl_%!Hky1eM1&zi3YJ&e)MG(=FlIRn`5U>+
z&-RXa@aR4Tg44Z+3?JNMb!UgD)1|$(!KIxY1d``(+{9STJ6oH~rX$WzPdGpBAq#;S
zoN_Z95;Rw!-bIx;<?T0_j>d#v{T)QoBAbjj*nP-yyTj|Be~nzj!Tv+)A#^$&{{0{Q
zsjo7epm*M9FdWipb|^|i7+0>sum~d05`|4gO_oy7e|W_H8*kIz+(bIZj>u@HWAfBc
zPN#(JCP}A>D+}`8z&kw%$>w&KbRhYb9J3rX`7^W1`$xx|^zUGVqPM@x@ctcEFJ2<(
zEU~(Efh(6TRV0lsF`JEe^Nk;2b-~%m5l4IH)Md)-WS1}2$H>(!h?kIgO8LR}nWY)w
zYrlbz0r_;oy?eLWT-)FqUwWO>!Gw=K=&-%^d4yE_(b1n!l!inodgncc!!c{itG<_7
zO;i+U-`uyKUcb+*KxDOG4j3N}+5hevw3j+uET^nRAps%OV@#f5rUsQnIN>9&YSs}5
zl1`h#8rI7*{<DisF4s2r|J@CF?|j54o$&Br%Jl4%W>Q18yWD%{eL5?v)SFGBMxD}>
zU?a{C`nc&7+dE_T^pI<=z^xl}(S)*wahl<SJA_d}d+9&Nr~qB0T;AGYV|khTx8LRA
z-FFF<p;@nkbDZ~j^v_2Wx#jll+g!hPgQ6_i-QD%s%zQ?=){Mp@o=E~+J#%XAV}pIU
z5K8{Xo_RCZR4?m1SG?lxR<DYIRb^NH&$va<TWD;mmymu++Cuv=e<`&P!d1X4kDX`d
zA7sG|eu73U7F1z5AK57-wz9K!)qj81Do$3gWSNx=A0AO_Hd#$1T~gXk5Q^Bp_qkPy
z5ZJ=|MhPhi;}BGxRc8sEWu+dn6gu8KcHADO^oIkA*@XV#KDAndxV=QPyG*^^A&4UH
zJe{WKSxJ$NNXH}6;gBMqB3lO0TEdhGHVY|cV@gwc6TPI4Ee(+rw2~UG%oz4g84pj1
z6zCveT$YT-6UO6|@o>asIw4RYXXoc=t(i_|n9|cxr_<@8T7J4m5PB<BbYVGt_a3WT
zn_PSO8apPV9VJKyVLkFcD|g5+q8yBnS|j5K87gXOiBgxmsB(60G}x|f@bB*RcsQPL
zIvh|=hMYclKwNL2I?LQzS)<kMk~ErxNlZS?$wnn+(!-5LoSvN!%ti!D0yhh>MGaR(
z6nW0*<ZoEoyhL#0HH<E3BoS+!3*^%gx9{EM{P+>gW(y?*g)Z4YIA(KWhtY7v@zF6?
zFJI;C{G8EbOg5W&k)E}jpPf;b<x{Vr00Jd(N}w~x;hVSEzJ86JE1O(GQ;(Yn>j>%*
zCfArkAmf^+peG3;4pB}~TMnq_DPK)WF22~|j}pOO?DaXBPB`qJq9+5+9y}mww7An<
z;e)kJTAdC_vw;jlOevU+G-f&=o}?Tc-6oig2$v;g+Q2muC?nGGfZ_R{u)1{#HlM>3
zDUCQ}=kir1{Znqe|2~t^370QkV>C%Aih|wUL!Pg-I5|G$==hkIUwnzvvr|T+F}j+?
zWQ^hH<fy_&9zUNEKpc^DC_&{<V>o;7F0;)J-AfIG&2Ylc_7y=5GtIpnLajzw41G|M
z0woklqd_sAkx2OVW(Z}7a548L0cn|YpbNtB05uwr^p4nE-z11@NbQKikXfNA#zUr)
z5$CfB8)3lZj-VM^oGH<Hj@1HJ<ix#0#B7LbEYqkp7!OYQ-rxKMgUO7$_wIs#hxhN2
zXDL~hv-{v4^+uiNZ`?#k$-D2p$NdKnxpn)tXCKUU7}k%&eA-6|P|i_IbF3Rvesl{J
zhNzaL-c<zHPfP*?aC+`sTVc?VB4{R9=Xib1G0Zig%b1pukuEvvoiH8tI6giCpmnfM
z5Jf~%p`;|sbH>vt08yC|n3AQsWm#FeHbI&Oq?Ypk&)$DVTY8=Oo!D>hH=cNN-CH@J
zs*u43!O-j`*<_0y)Jf{rC|Me-N3p-y(kzcgYoz71EL$3nWP8ohENRB}8hJcw<v>Z(
zlA2ATnPU?m03sED%60R(C!aU&<PZDYi%2%AK(^+CAFPYI@tpJSv)}#f{XGA$w+YrF
z7uS(djGjzUr6$iaUU>NxvOHtAcYt$_@pQ&?R^XgtFc^BMs;WT<uHCo|0?NXVv72jT
zl}a7yn7!K&9LhR)i{)Z4p}Mn0u#%8;9f9xylHPwrc(QjO=V%~vM4}61KqH9}#*!L`
z>ICR21S7dKbCh+>Kq-PeCzy<Bj{3A)E#fF44MJ+InG}UzjLL$BsaTFQD?vb0YA_l*
zZ&W%5L3bC*3BpBG`Ha2o+uTu>gW-t&a73jls@gIhPZ2`0v%N)KmlS15UexsZea4ds
z<H;1Gy;5{lR&*8@=cY(^Glq_EPI9ye)HtL6^z&SbE1DgLvl^ra$Oz%k)7k?S;t*Tc
zAiYeulo4sOO;KAmV!=<YM^r_`7e&He5t184t`%}LAe@Zpjt*E^Sp_610wW}oSw@vj
zndVbMYj~g`Inj!Ur0>V+s-`Xr!ZIT~*n~3=5y}8*Eib?D49W=h2LrZuw&--a#Bmd&
z6utdjc6T=Ev|5<eRkpXcxqkB|SFhio*E{gN00d=O^8OEe;I0yg10;?}xT85%98B5z
z;<vc;@drtR+5B`OiGmnuz>EuIEAoqeU4pbe!NduT_5ymEaj7Hu<+CwHhkSj%#i&jg
z>XK2B6XzLnaKQ3#pOw{P=-QA55k?5I*$g!s)1P|fjQ2G{j&}nD3ae{WQJ{*F*wo0q
zJJ@6A34)L`ig^CHZ)2R~*5(%bd%MIz#4Ph_8#~)~SXu0H{Mb54oU*mG#jCGf<M|g~
zWH1~a^%e7?<l~?C<eSFm9h>NK-Uu~w!yHjt+6VijtIO0yg>aB`S_En0?Jy&+jbp4O
zs5MT26<z|R*=bW3CB3}~7t?@0ayI1iH>O;f3bIB69ZJtUcQBcZv2MbeHPpiK+AL>O
zR5S&AXgT3WI)eKa4DCb`g)t$JR~1eflDdGw4Q$*1UGwUT-{rNdH(6XcM{OOu`?uKL
z-lu;s!r=J&x4(t8c)mhiljS+HETdNXO{39hH125!Jr`-gD2*s8+(91~1PIq4Y&Jb{
z@GvpCx1Jt6pEr_{API5KaiSHn7&`7W0`d^HGR0<5P&vqrLjqOdk!`IoCX|v&DP~29
z0~RO=oS~5fbVEVQ38XWiEmqfHEg~BuvJtkdFxpWTIeR<X<VC?`=1W?$tRycg!Z7qU
zqI202tu<NU6N^{7Ef9E=*5cCAy^Z1gPb^?-jhIY{`+Eo%;R1n?5|P9{TN50e^jz&1
zNAU(b);Nj)&8b#|pg}u!yxbRT6^4;=<l0bI1vrDPOJW(KfU>R`l_gdymLtoVRB&Hg
za=IJRNkal5!Fc475CuU#!c7kl$s+k|#`PQ5+20>hC_`B*$||Q;j=WNyIEO!_YHg@%
zMNyTErxWt5n8PT23Fg5EAG~W%3xBBeN~{K-jpMwaGdVzrkh;>yFeK?T5rOxQ7jfjh
z0gXje+FKjPk|>I>jg+FOFj>iut*5j^$+veZ?o@(ICKwrbAxNX?4+ezJ5#i`-#m#Ak
z)s~BG$@{wz7h8tYi_nN8qA)_teR2Zhh$g#K(_IiP^2vy&pZXSA0nPR@vsupW_70AS
zEX%oh`!<8o(DOZ%x3<m8l047RZ|S`mWAFY1GG9Cu-qu?PK{Oh(FzI1Jjjk<05|MV>
zh%f|Uka6ghdbLHCY7WN@NzxRd4a0Gc85jKQ$p%a0d~>H_PekZu10@9*zbV__-y_wA
zmUL`Yid(Y`tqm7i0YAQ)a=EQI-Z8{sOe8~`HI#Wl8U;k-O^R6$OvL1%#}}Xc3^HD%
z-95%+Fy*<uXHeD)VqU#*ot?d1KYq$klm%IqljZr_ihge`rM?};3&g?UXAYd8(Mnit
zgBefe!gzwpX+qsZwiCieh$_4Ss|;gg7@&1c5Qa$M?Oiv{9LI$LD|OC#E9SL+#n<-=
zcBUE~2x<!o0ddUOIZRQI8q1k9<j2;Ve00gt3N+1To8^TS8jaLzwcvN0p{^-!zliB>
z&~7%!Ow5(d0Shxhk|s>Eg4yoCyKe#GSFgS633Rt1gdlFGB<&_*R1ip5NCi<SNTtE*
z3TFkv`f^c_&Xd$3vj8(e%)ZT7tTsoFC4}(&gDB)fXF{}d4D*@?Rx+l_(k~1Gs9KBI
zAf6ir34*F1oRl~OXBsJs5uE78wBvv@PDrBI%abT?#>2Fa>jjJ_g1uVO9}Ic+<<}_6
z3N;t6tCS}R9t_6>M>~EmS8l95>Kz?UG-EC8PUr5&5P!NjtXqeHBn<^^sR?R}5%^M>
zGXxFq=Ez(U+*;=sz<KXk0Ab__YTYCzX{W4oQ%*D|yf!MiHP!S>No_%gF*Z&h41H3l
z3>^%M&435H5f|IAl1i34O}fnn$()Phj6tXhnNLu|Em#aF%bK0unCrK9FwQ56qR1#J
zjWPy{;9xie=9quy<{D!6z&wA4kU|iM!x3~y8L-$6X;Nde+OynM4OqfPO0eh!BAs;@
zWr#(9a{{fDch-}VX1ht$?XoHz3!RiRoiQ(u6)*QI#+5}$&_ZI;6ek3kQP`^Dm=v5^
zN%`1n$oYn)m4tM=3v}Dvxxl0YXFbni+(+%-AXs|<Cn7fcIlcabcDKMf!)!LCtTeT@
z<h`1mUf*Bg4`9sR8li9mQp{yNEK-K7ucpK*Ll<7P*47nROVCURmzv1XCsI?-t*5QQ
z=$Zfs13`Crf!5jzPFj{b4IWq?@l0Rw-Ti_>ffj<w2uu=D$cUcR)Vg3rz`4~HKeG{W
zCRU_zOm|_0h0X#&AdyJ03W-D<Y*X)FBV4(kq}Ac(K}l6_vA9&Bbirsmq^vC3!HqlH
z_dY&v`;KrBSoFMpE`Vn2Xte@FCq}x+L;h-wvlXEc$ZBpT6?qMO=e$6eol97|N)a_<
zWEv3#ZBDN_*19DRo-o`QloT4Kc};fmKC1N-^m}`BZ(WCh=JZm^`Iev=d#rhqq%<4N
zBVY75@}3~7s$tm2&id@^mb`TRHZyIR9rW&f_PmD>b1&<|QcNrzk#q!jnPm{zxsFC4
znClt(1=BCe_K2+y-d!$`lJzLS2|+8A1Wqxo49Q$@aMDOnVZuxsG=kXGv;vEBg5znx
zYART2MkG<_-N@uzq;Bpe=v0X-r_@DAS%>6#PH#A-tX@|n*7KV3-Q23LOT)a&a1I%U
z#GMwdwg};fBT1_n(d;D1#9JB*0ajRqbV%ttOF<N2g9@wtV(b-iJx?ew4j~0$7!yYU
zX&_i>CtS)Eqtc>9NY&_)b(Wb$P0X;5IM}6;jXBl`INp*hC6Y7>X*3%|aY7&ia9-=l
zLt%5wXrF=>{lS2({UQ6KDQ3<mGIl<G^8wV~ZyerZ2njX-=2pdFAn3M3(snW@&Rc{r
zSQJ_W2pb|zK$yhnS?&`Vp4TM9z@G@MBT1!KYF=0(4FXQCDjr&`8RiCU1tTlr@}uk<
zi|HQ_U3-<53%R%yv7SP+5fDZZt!A4zPB`2<C2>SPEpTOl8}3rJPH_Fs7Tf(P)4X6<
z6h|=gqg?SrNE&Yn35M4bdLSK(ZAsV+k%S(sR@G=%`%PgTAOcSq)>(m>lPYyx6U8Z5
z2!!>2FC=ntnMM?He#7vglQsRB!dTeT0nHD6jBERQw8sP7D=*WD8r;7eu^d^_W|KII
z>2wzf!wB$%<0OtzO5ySX*WadWo#gqKud+AJu+A~a@BBcA=j|sD8PGU!8U`aIhEPh5
zZ>+F!?lM|WC?<V6D;->35jGOSP7{YjIze-JiSnS2v;NrTEQF0FRyj;Qqs+$yjSf+i
zkQW(?kt10ONSke3+9F>)L2u(SFW-n+s{5Q?wk)@rBuPLh4OMATHbi5IoG%;1tu_)t
zU6v%OV0^I6)^<tnVDJMSsJjUv1dWwd8r>zLY>Oxq#8F5T1-@qQoEJ`;vt<txif>R$
zFN|iLXTH9nPvXGqX*u7cO2UAqfo^Q)i>E_{7KB7OK_dxS?liCnnu#L}4y(Fz`4{&V
zsHoPS*Ham)N>f+$J7T~6#^UvF{-Ez+ha*T5I%m%zws(;#r`?QMUh0sxS{Re#1m08L
zT99tu^9Tvz2or^vs-Eu}A&ezb1Yjzh^9sJ9fKD?d4xtl4-Z028MKw~!jhLM%rPMK9
z+hk(_mKwfZkFj2MXRa#k@b8C|5~DS`%yHR-Obi*0MoeZis(Tfbd@o@wVWUa1vP^w2
zMhZcz5wmgP1g*7Y@<|^ly`vztA!s*zNm>Yx<SR=|R(Q2oYY`$qrcGQ?pk@<X5Fz4(
zC=6+tnzP-I6YY?=*`bP?)Mt+Inn@W(j*YUz$w<&`C4|zGBI_EgkXScYEdb&;VNxrc
zR=DYa$)Ly9*3O*YbVyJ6At#JCgtL}-VTs5<*&iVZ=`<70oWDrtc!&MXo3s~GqITl@
z9NiW&iopreP6u0+=)6RPAy#RUR!q>a*s`P^_Hc1Z7^mJYwk|o{3D{^#+Dj`;g9i4}
zNgB(Gls8}Hc(urhG@{i^2&F|}$#REw&X3`o5Zg!_40}B!1~b}Y+Bx9nt=q3l9Nil+
z>wCsKCXkaK|4A0V`z(VOUL%b|?mN3qY#qI=EsUyYrXfKbk_`t$RYAJAj0;18M#D=x
z*NQ0ilS3v6GF1p8v1N{$jY+zTG}~RuETgK%fT81{zWNM<tJlbbE~m;V7f!5@G+WfF
zWVnC8c%)f6)}8CzXq?eB0xv)}9ZyJ>776N_aW<X%H~hy55k;&$@gYvEtuvT>hS=6D
zEw)IKkboLxD*_>ogx;tML>MBXz}LmKp;DSiCM2Qv`j7~u^>)C{Il56mve2TcYmU!(
zigf~7I&SNT%n8<Hi%VhbcacSo(={Rp=ZGrlS^Gz#k6^P5rwx+<^h(KaJfSFSYW;gW
zmevubDa#K&ghDVX@37cSSzc@r%78Mf2?9$Ld6zU-d-I_%2oQn9$p9k+RvAJmKnBRj
z6DoyrVEy$3LK4c5SV)Yhi9-W6!xe^6S>trcax3Cgx<Dfisq%~>E4=%q^CdfDG;-cw
z@NR-RVU|;iDS4hzD|K%q!1p?Y){-u-vc9^(-dDatFeq7FYO%4pfGsN~`+b`2h%{Zm
zV5o-!(#1u>rbqqABq1M836pu5SO|hff-}Z1sAZ1C5vQr|d#EY8E@_04j??sB`X*Jp
zMm3tU-UvB;Y!Mj-<kJbm-UMV!)bxv=(T+fRvCpzBDYF@2Sy2@^lWd0hy-vRKaMsW{
zeTEYcJ<9dZd<Iz=Hr5wdTj-GYhm3ajN#qLga+hkBQS4<jmzRl}Ef0fE8Wg=Tjc(}I
z?LZPV8#rf)T}VC~Bb=qtYJr1kHm0l!8ncMDoO1KYf6laX2GiT;bSL4&+A21H*=R^_
zcR~=iNR1ccG;^rDkby_-O=ra2Wy&&Vn&&^XF?vT7R{OsDIK95kom<;P(y_McD9Ua0
zY(&~_Qw@gbIRUEb4=|-7U0xyxBb1C74=1dzwmq9Pkc5pCQ{*%k7O1lfHJOsMTF6$L
z@nnEiiqcI`Wl2XuyOAJ{trK*cG?IX7STH#l(C#inpg=Ue=-m9q)XEF`>CuF!s^`+$
z_YS>%uOWmWSze~GevJ0|8j&h|a**CT!k1LUkqkR?`Dr_c(`s#~jq{Qkhozq&^fKZ0
zP;k-`qOn1M4*lC`ocG`e!+?ko88~bd9+e@y+n^VMbHQAAjn_5EY~^_jx)SI*z^Ixa
zkOVF`LQ#2jUVdP8`tLc0VMzD%8AKQ{oXqGhgfyEPQw{w|!aCGUdyfNaka0{@YocZo
zYoRuRy3(|g$d7vv5X2E;&hs(Kn>EEjh>l}QRpL~Qs};hPL_!mV2|=TUOeB%Cl#>ZY
zE5ay7$U_}Bi;&)|OqDg-mNYvVHY~kH*zbMr@ZlUXP1*SHkD>YldcX55Ybz0Hno~>$
zNK+xSpxEnq)mmMnvYh7HF``xzWXPnnB-YSrhJf(p!8k-(Niz(oW*NdMnvIs%ODl48
zRiMifQ7KwUNFF6b%g1PT5}dOPc6QKZNoys>d)0U#OB5&G7pSVp`Xf5a1*jVPR%L_V
z%digFY;xi!KgH<w9rmAkp5v<#VLE2Azv)FA%aY0F3|%YG3Ny=Tu5S>v8ptrBpJ`SS
zLmGK*S(ql+x<&x)#bxTr1m{Xx?KWjyQe-o9ouic|tqlvolyR$xSY9G&MObIpn@n&@
z(O#8a$#_mXK>BJ%T~=iM0qN2)T&?CJsqlk7eZco}Bn%{LAN(Ne&p(IRyb7(9*>I1v
z(<V$);xHh#7U2YCHl-XK`1;$z0-BIvX*mY?@{A*hBb-ff#uBA1gpjDBAZoU0Ei9wT
z+^eIeEy{snF^=eU*ZIzHn-497*jm$QrL-1e3a7EgAK?ydR&m~ULCrGE?sat0^n8Z@
zNFjoN?!}9&T)f1|-Zml|A#HD7w+{$KL?8uW;B}pJ>4{<{2yp_c%8-|aFm~P!)j4Do
z%-?^{x|Y@;wdZvl`g!Qevx?*H3N}cQA!KQY69Q>KSV5RX$Ux$xz=&XOdFmVXs+@Tn
z+B~M#A*L*d<H+|TB0p8UOp13MJVd$tpcBdAqcx5lqt$6MJlLb*W|&Y=nSwBih@yaa
z?lNh$!Wxa9&A?fd5!BjJ)Rso#6E;#Hq8KTS=g&BWFuv3pMKNikg|Ob!Tq^u2y1PQv
zOgV5Bfu7-HfQ(~g>^B4M5avDi!qBRwE-RAh6eUO0#muwSf1smtcO%)@IL^sW`~>~p
z9*f61)Xtz)P10@>CJ`YJ#(*&t`GoSI2W}Naz*Gq|mTpTP)hWXy#ncKZ0@9{H6*(47
zv(=%}H4+GekT6W4(6pB~cufi#;{#4MW2`ZBmzogI)duFKU`McdWymHITBAOyu{kgA
z{znXJNjKIx^D{ri!QL)Q#}}EF71mgiZksR)32=y}#HgBLHlpn9AvlH;ArmcW)Rtx(
z`MxHCppjy0jTDlk)1}U4*s7-8U8Jl_f-odXQf9Tp6*aBX=Xka%2!;brv{I~gEFW7y
z35_I1*KZ)|I)SMbv+<O4*rRS-doMJ$-!<OZk<dEA&H`tC;^Sm5FLSFuVb!%+UWpO9
zpt;bU*Xk8vvq3c)QV#nVZ73!MXoawXAPhZePD*5w;FRBy1!)83rKfIpi873d!jP&c
zP}z`ZX`M$u@FBA4kk`KbITj7H?tec@Pf3j;5COu}1RRpxELE;hMTMv=dwYF~xjxkI
z9zXcH;Ok@Q)itqZeroU#-S7K`zNOhLj=p_lEpznW_j=vVTCDKqPR`o7q=NO`4`=5J
z%hH!#-au;<Z+QAc-w&-d%DnFPahO1~)<5)xF=)1EFE6wH(1SF_djw<%J0)~A@f=zz
zz?0yL0$r7uSw^LyEDhS&*U6Iz91`OWRbS@_H)DvAkOpY2s8kKckfaff<yD&N#}TGv
zaN{L(yogRiL>SOcW9;Gk5Lis9sB24^*R-mVdOF5cHQi2!&cY&%G$jl}ob$;yFLGvC
zPEnM8_xQsI2grb>2Oi|yPyPgq`)m(~tTtANLqi-%8gp{3QHmgrsiq^UK_3++)I~uo
z5<iB4H})co{3%-ddpQgQx>CeJNOxfsr)mf!b$^JeYL;6qmOt=e_P+Z~+}0I>wNub+
zV(S^H^m9-cUobE@oUs&>0#PgS!GytZ^rj7pI0T3Oa7OH%Lr9kHe~|M(^+~E~%I&>9
zPA;85xRRuqkTx3LeX~}CaZE8DQ1%a~g9KEG)gfWn1Q}q3L`Jb+KWh!n5q7$mq9O@m
z7MG9tu?+%dy-lFQ@ud~Ik37cKH$I2$-5@@30XhvtJ|Ynj0)ub{VZE=4)rM@Gv#uSJ
z{*3-;^lmSdyB*nglrq7A5R%TuaW(=+^I(S#8I(OjJ-3s^9<h?+Rtswt#h}mdphU<J
z5d_F20%1|PBQFKYdhSLXO2Po7)5HY0q`|bv7>*9uII)O*@B!9OY;fQEf0^HS^#%I9
z1Ny;(l(pv4sRg>JqZ|xaO*I>gDGJMIA~+*MOr22=Cp`MV13Ym5CC;2X$LiV|&KhRZ
z3EO+SeD{SHxpCtb!{PX@2khO2Kn5XU(qQHE8Dz0UXBHx@B`_J0n$AmDbHgsJaN3}X
z0_QBo`{tmuWL76w2a{p}<$|NKNJ2qdOmLzmisnSbIr(e-{5fbX5(NRxMgyc^>!oXq
z>jToob%L~sw1&uNgdHQxs{oXBiB>hD*=4caVsT}aKl<?dy_wiT7lgO^HQKPdyXRG8
z>zXT9u27aG)5(lON6Ysb4nGIuR-2V`XGsncT3Mf_O_1dfDJ?-19=(0S7%+yq%&BIc
ztQSNvLIxNDY86rGh)Vh1T!12`BoY;2Xut*De_SgJmZh~t($gncy8IZCfNQ(kjEtp@
zR}gWF_2nj^s-T?FvN?gx$ty=z2vQA2S<{GPE?>I9vGwB|TVJQuZsDw@C<=Q01Gcw!
z+1lQv-|v&nW>j_kgFDR$zX=cy!nDDu2Op$!ut#&&qbYOjY@aZ6%o`k@P?#jdrbRUy
zGc6}nS;;~(L52|`4CcCL0fkC%HY65`B!D22U@eU(A-9G}KBFwhoV))~^vM(a$cKN7
z;mOl{>%aRO1PkZbv1@eNA?G(b*kT6xgyStoqj#0@%o4d6Ck5qbz;e6IPk#KPT)1$a
z<0nqg>2#^9lIe8Jja#>Q@uio!b^8vp*$=1A?;KvwDor{2$RjKc_DLsgnqi3^Z4t(i
z2T07<!Z40W+il9}gi-Fx@<Mcd(g-D{R+NRHFcHEC;y@5fg^WEMz8N=|=^5kc7+lWz
z$DY8htnj0s{C(V5=<w9v`)^2&UE+4xWwjG>@pPARI3iIQqHWOQ=NM0+(+CkE<Oh4K
zv^re)@$=k&`4Sr^Ptt5QD6$!&@rWz0y~fKgzrxPWzK8SQWxQh#L;%u3Xa!9tSZFm!
zrKOsd1j6?<ob`}IYb{18qznk8C6b0PNr>VYV;uY2F*j$Y87J#ZYeze*SgbcW-`ZiR
zUE{3REh@?>c|PUj#RpjW$VX|c9V6{@IP>@ix$^J+HP>$+U>DD_vV4iLuzc&87fD0O
zPj6@@6Gy*_xkEr%JMMq@GEe-%LtMCYk+WwnaQyg5y4@~u><wM^_xE`E>1X)&zwy8D
zo1ghDw)ghm`%HUF2pJGWF`d<81Vw{HOnlj?h9EEI5+^=^h#){%hpoH}yCnb@VyvJx
zA=W{^Y*X6UH_;JvWJDTO2mzr~M-&n#9E+=~gv;xQFhn*wNGZv0PuTBi+O1QBttHSk
z6ggoP(NH%5C@QFHja*1*wOTaWUCuuKqqN#>j;*iHRcaMlYp%Td3OhTy%w{vTx3+u}
zkl7Dt0rYe55CWY>39Uwxjwpy!g{y@ZpL8C*bL80Q4)JB05JwbxgRE(hFsPR4nH8oc
z#o%bjk|o#YSk$KzWgu!C#=8baF{{Uq)4p(lg>&aXDbiybl=*;xN)YX1BnxYlS<YlS
zBq@iiW!vP1WK;=8rKKN9R!$z{^vSbaxOhLOPo1H=;5}W!z>_fh{Q+06UFF-~{x;uv
z`f0A+xaEcIet=D5AqCPoQWww;LplpxC`+`_$oZ+_@U-a;`($B&bOlN1Xrv)B2*_rF
z?a?Yb_9R2s!C1v&FlEEt;Z&zjBhrAy37F1C1X6JF!e!F;e}rbIMRea~7EYh$*8Yea
zZkbb;&a-rEgPlA3+}zpWc*pXAB4juX(2W3FImT(sCqMa7PF}j7`z~JQ%$aj6EiIEI
z32~fImIb$N-QufX{VM<NH-3Y!edC*qC(|F;*a80j1TcOKTT;?lUc^qfU@o|dl&?F7
zYK=4A_$eeOl8#myBSJwoaoj#Q#(r>$k!vAbO~(y6C2nzYaZDtvKZOa$Xna5;ZE*JC
zN8!Tz2&0e__dh~weVy%{eXi*yJamDTV<&iF^EGZ?cRYSB<k8I0p9Yj^$Q1>u)!;w*
zL%+zWi<dcn;XY2EKFi9=3TdN>5MFrp+SP0P_UAv(zx_A=hHpRp%-aZ<-!I;|2f`fP
z;tWxeAZZXvgAOESI>lBsqTQm-3rtlXK}*9RA_+^36KJj2?k(_=euDk>2@30&)dh>~
z7Avcl827I5=<GR`Ti&Rst~|stPMWN2oWvy&)@qL5{{Y)xh**i*=#?&O>m5QV`P9#S
zgqL2p#Vfb3Qx=NvxSHR--Xq(r2;x0{YV9}+3rjQ`Evz;C_UC^ShvTuwp5XMEvz$41
zA3yb}-^VXK`8ocF|M&mMlVA8E`W`ayJIDNnw3@&vLScxb0BHp%d`Zh38t?eVq1DEV
zzX^f00osO))gqJPGyv1a1}02two~utP)?|_9xc7YlDduqv~gGgNs^GHDMS%ryMvU1
zMjFzH40T;oiUOl+idji9oYA|}M*xFz#<VI}N!Dou9jeH3PD-+D#&@2%?HlgKV2oz%
z*aoLgKS5pBT)KE4dwcu*yZ`W;D5W?!=rb6+7Z|meS5p1>1QOyTA|fRq#0UXOd8v;;
zqAP{1m3OE_%-QD-V+Eru;y^61?M^TYm#D%B8HF@sfeVh~id$%Phi*9ZCO+0{>Zk24
zjaG{!iE&{_XKj^uKSNCnbq!M+Akr?XX&DR#^anYEamjX8B5wC_3yw1npXUAdKg6Z`
zA7o)^i7cCN>-J5;Fkp3MjSCmfarWF<E?v6BgZJOh*S`J@zWvlw^bYzzxU!Z5V>Mv_
zaif96BD7~po9PS_$G9j$&$YLF5|%_^NEleGu}r3dtD}=#OCDw#w<#)v0jjkN<mv*&
z_*w32UMEew7^tc%f+(WV>auif18Wq`?h?n&oMtq>Nh?U{bQ&aS$oWgB=q@dC>xC=4
zvOD2rrMRXw)ps{3<pB>rn)2`mKg{atF_I)@Z-0m9o_m_*<y9Ve<VSe;kw<v=kw^LI
zpZXO4`qQ80AN`Ym&I>QTcvp+y?;64@EytlGN*e@`K$IcIIqJa(*=b^}KxGA1YtO4f
zkVYZGX#|eR$noOfJh!^<V<KW^S%r|Wv3?(;>^#NjDeha?A_zPry{<}{jW+Ft6`Ci{
zU{pnCWsTL7$0=^l=!iBaPcGAFL_G1~huAoEf}NM1=h`6WWoy`)WR%Ztk=Z>iKM0o|
zeUx^`Z=!GAy3PwPJ;&us4|4hPLww-FAK?QZ`Y?asmwu6d@DF~SU;jt{n9Z$sM=pG4
zId^L@PGe0$BCHpst1ZMKGEJ~~jvM==p-R7Fk`BTsA#PZPCG^LVZx7C}x!$6g=A1v>
zK?KW$jYSr^9rmxadFI(M9}Zrk*^+4E{SU10s;d2KFVa|CrM<F27z;Xa#q)Qz(Yj>i
z*dnuu;?65q$X<DlSNA5|s1#ycA?yJ^{?U)|^FQ+|96x)8$#lf^8&~;PpZ=%ZxOs*9
z?z_mxKKdy>@yVa#?3oMv(l7rq58Qv5|NDRYclh;x{7)&%ciTJp{^1j_o!fmvAOmFJ
zrD5#6)@$c#zgE9NJ{|yxb&{!Ru&b85IZ?O-X-d#tAdEuHtVvM>2r@dx@7b;MCy+3T
zh?5win6S}6N<kO}1R<2pqKrdp$5c6nh32(UjsW({oMBn;P-!@3Aaa5@j;QLAz5N{y
zJ<^TQT)6K77cX8wYs2OHF0pxMlW+abx2Y;Gul)lVLu+tWBdsP7))O=5Cty2Q-gR^H
zo<q?$;Xuj&DIJB842uri={mi}IxdNc8v)IBlV-Dx1^SyIdRP*<iI?&>25W^!R@F7O
z$dF+~Yhj62a);DRQRSFo6cR=WRW)U=w?lvDfStjVy{baZDmqh3a(u#UR<N_P%l6JT
zw{BnO+2@|dIm_9z=XmhJN4Rw90Umz%K~9}K&hg_L#Bs#8zWo$CJG(!CBr@j+7)@Ym
z?`2{E;e2`7Dr`}JF&I^&s>+vCk;G{e8A)c9qo=z(?@o|+B9^*v?pO;EwP|)1(7EB(
zcNB6s;NqIsOH<l-o{S8sCqwjbkJe)!AzfS|v@eo`6}N8OVK$o(Mlt(41A4E#z%w_l
z@zSVZWI-NeEG;c_>Ea_?yzn5_I&R;&$;&Ulz`yx7{~T>99(?E_KJkg4;Uho#NiJNx
z#Gn6je~$YuT;T8iKmT8R^=n`Mfk64a#E`|<5@{>qI6}D6R}dNr<T-XU!NrO1!D%%o
zmWD)W6BjzBrDiW%;i=#RHmX=(k(^s@Vq}|z#U*TgoGagOga>`@J6U>Qjn-)EKt>b?
z+Z5w~jfXx=Yk8IM);89ZeEWA^<kZ<!!XRby#xBDv&++XWclhqCAR86P?FoyEi~Q;T
z>{ERFqn{$L3;KftzVVH(@=yNh-)CX5%j56=AfNhuzsTc{|0oYU{1|`jzxpqE_`wJG
zKmXVN4X?g-^>-bgcT6B|?oA%43RzANu0~fgtky)WhL`=u+a{73tSaZt@{k~qM1f;i
ztx}yi&&iX=P+5%)EDI}+^B331$BJP$VpE=HYj}+dX@=1bqb$mKDT?Ce4cr6oCu+3O
zB4Kg(GLJ1@MwKO#>wD}^O7^aPlNWAXWm{_y5DDKf-W?D5Z~wdh2b1ZLX*OXpp0K^+
z?QCCp<t9%(^&P(Q)i3gg{?MP`V?X{0F5Gt?fAz2YWuheEfBgIZ6Lnp`=ZS~)jB{Zs
zgsJA-1ADYPIa0!fqrSl*E8IFSA#v#Sj&SoPx&vo$P8_kBmBn*!{8RcemN9-OG`IZ4
zNrbMEx*{kiSY^;kk<W69VoGWJLZz)2-ph-^i}_j0?b~;#RmH)<;3$DutJ&P#CJbXh
zu-IKf8O_G}I!bFM(<!~)f%jFvhn=AClMrh)7)4NL1l0^tXSjNP^F|@B3zaym^Kuc)
ziH|}^tc&T3W%9-n;X(r$2wGji!b(D?+d^fA-u@D`S)!!_5@#{WqGOA(peAGLVV|h8
zNRTv%O+lyJBM=s~+2=qBb`A#Iy8aTk_I4SR1+{ZXX>du+jXQVv+>@WhIYVA#^alg>
z_V=*PvarzR?AdSe_~Reo$A0YNT)OWvKlYIyrPXW_MiF25(w8}ykKw(>A@NRHO;}A4
zsst)bSt4YBNIb6tDUnf#&8Fb2-;5v#Lx*)SJ+Xn<SZB4<#TZ3pD^?aetgp8@*qNfc
ztK8mOV>LS<5(cX+N*JsN&{;-3>>(bph~^S*+UHc~H8Q)5+Kd^wgx=mBH*Y-6mAyVA
zYY8Ntn5Pt9{^r+s_QhwI&c@8L8N=a-Uhe?wG_SsTo#&o=j_03$mS6b!Kg8vWm-)me
zK0y>Ggh9yXzx3tzR#gywEws*p*2Gl~D#uZxiyRq8K#I#NY#<4ufV!9=gqLtcgv7Bx
z>y+KbeJox$Ni0KzsEJyJlP6m&EiSQnqes1PjB9%vtS?>l_l>e>ZLmRz9`q4LBOY`F
zoi%iSn};@zQN$~h+avZ>$llGX+`Re}FAc{OKrH<_njm=cOTWz*zy5hf;{nszj9#zD
zLBCHJ3Z8%AWxn>cukibS_LumBf8dXD`s8VT=@)*HBu)80{-6FYTzTb{?>j{Am_UT^
zHRd$iM~v^_WDIDebAIu15H(YeFbX0}Rp3mGvL$6!(2Ba;QBCd)GEU1W%_QXcZx5N)
z8RyQGT)upqd}<hrnB7UhdUuAd6i#U`G6=}&3^f@emXD)li*WNRbZ>tbWkO1^#8w&c
z>dq~;wed3N&N2s;@||bD%Xgmru78|40`Um5G*<L`IosO@+`M^<-u^y6_p^V1lc&z|
zU;L%N#KHan|MJtHo-;S^F&tP`;fkDSx=mQkaK+G<WORX>OQuUHk3=Z7QIwVUSg8r9
zZ9+kVY0yFnM<bMk;as`{OX(yNouG;UIWit;g(Zv~HZQQV86*~));K9ivL3DRZOrZs
zq_)f|!^?X)gRJ1yxt5T24ufTO<yFG)ItalRzVa2Ub5!%)oS5HK-~8rNEG%?bT3X_d
z|FJ*L!om`N_D}z3L{ZGY_*egm>14`iJZ3VP-18WsHBy&^`92~W;H(1WOPwGPL5Pq6
zXoIc1aHG{VS{a;+Km<%o#H4h{s3M6&Mm^1JW@&dVX_PRo9I}KZZRpAxUDwd?x>2sG
zFtZslZh(xS%8C2eiNkHCCT6P$*cxWsI@qCC=9JpwvV;{3CsWGTu5s(m9gLmZ`_4_a
zth4OzkJ#NCu)V#{{{8_!@$pacBac1K<Bva1o|jZ*#czM%3ryZ~O<y2YK{Va+i(eFj
zsplINhlpZC?A3~`46t>Dvj$Zg0%H*}qOVd$<D6Cyp=!&O?FlZL^5`QHVFj78OsbgC
zRMSaHOl=T>g)qb^g`JGhbp=TWm1T4%-zIe-g>BF?Z8i_4Tp#W+G6supCgWf*n(&1$
zeHrH335N;MI<Mw18dV(hChYF^=<WCT!+-cs^2kGv^0AM8l&Y-h9rStj*=OG)Ovb~&
zLzNRxZs1gfO`ABQkbtpnZmgFew86;$UFQgAscTJa6~d<6uA$x>a^G5qong%jFAj*4
z13vJ9%Y-5%SC*n~GaMC2Sz~L9GujI@mNhP)pj3%ytx)%ebS9rAs2q`Kamy}pd%w?X
z(;>O@8>(vVU)~!G`1Ei5dw(iAL|Edr^1PyU&HnC$JGZyFefu`Q`ltQ^4_thZU--G7
zC(BFz@?ZWh>3yGa^gH)JoWxZ%a<q-C4sbz?b6%*fo)v`6rpJS4nU^6D0Yx>%C=cIF
zq8RFgBn~+!r#y6Wl~0~pVHBUnnUu5@u+zKEVA97}MLx?Y@`6wXG`maKx+JJekI)dH
ziBQUM`uH+i6Uo8UF`SOr%PZaqW$)hb?>qVM)XH(tD|qHRFJbN9XEd7dhkxadaPHhW
z{`3F+U*x5iUgU+Bue{qC_V(v@j(R+z8ukgR9H%OrQS;sL9MWq&#MjN6zjiKa=!`)L
zgOdh{BNPHDA&vurIPkj@>(EMJRgKX#Mprl?iDT&n2(1C*rN*7s-kZZ1oUzneQ`MTX
zHsnQ4QB|lz)z>*VSk3jW095UJ5!(2#{h`S*9#7}CtA-mlZnC<X(Z19s3<BD%7AKA$
z_otD`<emg6oz<xE2$x!9F+f&B|LV+~IY0Ng5JBJ}rb42frK(EmN+D&6Od$*+5RxDg
ztgbhaaextC0lFy1O!Fa8nNjMB(B)X;6H62)NLAyC3|Du($y03*WkH-6_KK3x)G!=n
zjEbB}o1?kt<_Wi|6{GR%*Yrk^WL?)(l_Jkd%2H95HDy`x!6!b*V~;*c|6stitJiq-
z>eV-Ck=@-Og3}H$8xdstK!q_e&iKum)t0C+@2S}goTN66s_;G<Vc5XPh`_~YYnWQa
z#j|Zrg-s?VrqK=<4@T_vwrQ)3x~$1nj;<B$#YK{&iK_}^m7}W?f|#O$l~m9SVtQlC
zK{lp08ZxXER&aEJJRHYbz0(P74k<T|=}d9s+Ag1SPf{zxAN-|1#D_odVLtxxk8$JX
z4K_FL&^zeg+xRhm&kV@vfFRpLU~n$OS%oeukBDrwaPxZ-4##i?1{!e#D?%dKWF~4x
zPV?x4P3}(?nbZ-B%ME(_JvMi5v!M&>vO=2)Mj4vTHc^yP4+aFKcL)_x!mO~IS!&Yl
zq`c5~46B0O@tCQ8yJ^-N_rdy2Kk{D~>zK_nH?QsUoB!c6lx5BT=8yd<Kl;HB@dtnL
zm-z0p&+&Ku-rqY~58fJYhw)lVvA;t-IYE+*5D|E+i}QZmhdTiQGKz>ADU~V7%4%-l
zA`x-La+-5)b)7vs<=B18+<*TDCJNcQv&Z%8HyLJIq*jwMV>BL97B#0&tkYUrqneBf
zt-_WCMpY;yndKHO7kD8z><GhXxQo$;OWxgwb1)p{JoEI66nRBe=KRS&_NRE_iTCp>
zzw*nx^nd+Z_v8Wx;IyXy;*0b~J>uqsKp3p3{AASHPqtbgH7%94)XFj%XBe#!aX^s7
zM2(uo$V+m!mYawqVrOSUtsG<%?Cb!Y_o%ZzP6Tw@9U86F!-C5k=NzWU<{$-w){=hZ
zm{pE`uILR%OtnV81@)ueMeY6IU_`$^Aq@MBM*oPDCr|Rf{S$wJG)+0Sw!+W;^iT1b
z&wh@bot?WMKp`+y&Fs!?RD6O)k%6h^>H_mcn)%J4G-l4^*Vd7h6;)Xi29hY&EJjc!
z0j{00w9%%s)S#>kz2213c#5t@Bx-`w8KufeqkvYcLm0(4T|<>)RfRDIr9qP*qZYl~
zaZpZ}l_j;Y@3@h9+q`(o*Ty(z(~=uEHmPe(k>>~jPrUC59)0v--uKvJ?C$ST_U=A`
z2q7_5NpY}8s~UkSA#&ifuP`K$_fVMw7GxX}BoQiiO!FLFX`)!tjw{xiFk4ziwj3uO
zIL5}wWqPBWo$Uj9<4sgOrfFx`dcwFK6Nr$N)nlGll1&M;qAqicvy_#<B`ISTxoHi3
zH6hPSjQQTYrr#~f%5dYx4qAPlvM6xI@S`94aen{L{w$yW;urbN-}>x5J+2TEHJeiQ
z_lYM1gsE{sfU|}`N<<??RkfdobHxp<6?JBq<T<7^M5$nDCF6WIK`t%RScZ!qzQDrj
zBE7+wJGXb(9$dqi87(uxsEmXDkY?Io?f7Yb5YNWMSn49jI7e216DeDDm)Fqriy>9<
z-XP#*ZMbn`m-;hLlFc%(hL3#sqx`3T@=x-a&wY-UuDtVx`t1qCShl|UMY0piG-U{`
zCW<38Ahbo=kaChSJs6@??VadEjM9QK)65D>os^ty@3LNxnMNnL_S%?MH{r`)c$Jr)
z?a}JAxqSWsg6*#on?3r20ZZLB>+8obQlcj#BCEYYN?9PCAruigDdUD==lTumEPu}&
zK<8jGDR}wi*Z7zApOZuhzxtp48Giov|2)6`5C1VYZr!@~2S0QH6kZk8i<mGM;QQV?
z^24gW!2h8Wf#=;w=_$`r;)DYsIi&abk2jL7#OqFlho8UcWxd{ix!17PA%uMsgf<9*
zqYpVR)7?`Dfi6q>FTX(6v@}$Sk;dEB&Uu(dX{w?`mo-{@M3ylQRa<760$I^XG8P&K
zRCa@LDHtA<^bT_Nc4zc<XS9-#j#4ZIQvx+cplLK)wA*dJc_>RvTwtpTt$csQxR8mB
z=+`-;ET_!NBh}h_yjaSjW@mesr{{{jp$s^C<~;9z;{81R%rorI&AjgFQKh6D_St*-
zn=D^k^W;&2xo3sHkB!n4ld0eJYV9c@1k|M?&uVH>5VU8kHFlV|OKeX<_Id@=YLCx<
z=4A%GnqwPFob6av#ywO$q$)~|t*+2nSfbV%vI#=h)J2I=o~dq$84}aCGkU#&SJHlu
zg{G?1+_|&Mm%s8=aF9d^Kk~jO_<f)JNxuA*uaez^9+Z%h>E>;Azxfr;o#}uTgt0`V
zF{Y}K^GeP1VBizi{2qXax^R@Un(3^hy)xs3zD_32vvp_6?X3~T)qOtu8!s}O3eI2J
z;GxwfOXDvy%|;ldICt(0aid9I=E!UcX-r*K7*!HVfpZCyD5b9k9PI4REqmWvsM@l%
zwa-_+@>Opkku-Vep-1?sPyN1kpFo6w(MvCL>x*9`JQ0#*eOe100vwSlsq=zzQlN?o
zXEY9nMxY9X5R!2zFj++^Myw6KM6vu+Z1);`?X$PI^@?I4+2Gvq1s3<OvLc=*kR!q<
z;MA$pAOu-alg=g-y&h^hLsk`Wq;YA)7}%do7$1yKKz)xH*F(4E(VWd^C0Ad&$-n)L
zPjmjvc|QE%ALmDZ^uza@K%6y<uHEDy&uAVKq!H*)BL##4BI$+xobirxPW$qbb`tF*
zLTfZL;&PAm@>Lvyax+H-G2x&91F;&BsvT0-Cy9Jj6Cp58FdL2uN(Id}GHN2`_MaNV
zFt6z4HIuR?n`>?zMjXWiLEz;Xq{JA*EX&@scm;D=`6!AA!vKMxC@Olr1HSf+uM-9#
zYb(b%c5Iz?vq2n11arxecm4k0fdJ#{H`z}mvQv{wO(+D-SQ0sd(*~oohwo~G*2=r+
zIViQHc9yuxS<ySJR4vR13cJEgNs3WToEf@$#!9_K$88hKf@Zr(qtOHeRZ(I_LjvOn
zJFDorA`li6I{Hd;Fv*zY6_v4=dzgko2=Cz_{IXqDn*F^4p8M|etgfsQ$dFSfPH=4f
z7_VOa-O(`b@C2AFXX`6pW{vlO8ZrqXZp1{^5Y#nAmQ&42Uy88?D+GmBSmUToNWZeI
zR0SLM8tdal%J%!X@zRj#v}SNSVqtNK;}W`)S6PWSsj`fPg$`?LD=2GmMM*FkP)x_T
zvc#H-RtiQ044q~FU_g--)DG?yDj?plLDJf=yK}%dzVQtfy9=zZZSa97p5Wxk6TJG`
z)w|!JONp7zxcTIhbbjJ}=vj|eBP2=_B2^>ml6;y|&q_?KaPDvnHP$+67jcjqOkJ`P
zyuhTtNZxsb7oX~}x1FP>3!GkG;H0oLcfQ5(_8!Aw#>Vk=mKK+o6$P!b!VUY3<AhFC
z<4i@jB~db<rwrS7cF{^xaPKMDYx4OGt~Hju-2vbD#@AV2Tj#$x`B(V($3D*g=l}aZ
zGo8)e_P6ioNT`)y^XV(B%XLCsa^=bv51n0Qu@RESG3^s=goC;)Fsi01D&p8NEMu;2
z<vdg=0yL}9E$sHcgz#bV)=65+pr-?tu78ttd4;9!gz*g4P97(Y6Z*qGNmVhvwMBV*
zhZB`TjmHG4L?(jkMb7Q5JxXFGl-~90jjvn?;wVDZ>U)+H!a&k$HAv#bGi$2~r8Rk0
z^XhBY`1HU1=REkp!+huiALO6?i+^!X$3kZ<)0?*$Oeb{hB55Ka^^o2)4r!+m)>w?0
z3leG#+Mta<8HvDPZBD1`u}~%$C)nJwR4yeb4OnQ2K5JqV!2}t-j_fivpcqeZSwYY>
zgzZIy5E!kn!ZIpr_GcxTQk2G0zhUPh=ZPUr6QU@ho;x)HN8=!+APfVNG$x5+@7bN_
z7_E8w>8Ans$cI0|{TDCMY&MAFc)lyUi<sBK^WKny>vw3MYN7*8fFWol$k@9K22n^L
z1%WnDS+o{JVThH6vIbpO#BRuny#-mps9Xgd6W5N=RV=F^9k)dwhv)#Lj43NeJ<f^8
zGphXokrJdU8Ff`-T}^H@J4H!9E1A`r(($Ib7DW-uOG|Xx9jdD2pg*9j%C{t8N!swf
zHc=R&l%gt2l+x_&?epw&&#<zx%%cxK&g$|q?N;lq=ivMuWq7B@=F>0HIoD*nKOza>
z%vy&uk~GuApSF}jS0!30qAVv4TI^Jcy>ZPYt!T)c<JIrrCOh<+muZ9xEG#Oba+|eV
zuW~HDg-mi%nX<aL<{{#f3C(HF__bT4ksv4vs%%CQTE^0Gb1>pyl#z*m>}_{4L}7@O
zg1T1Ub74%9h-RZfAOn=Dsp=Z79D93xzV^+papBy3eBxuD;{MC`zbi=UJBA1tZuYtJ
z)Qc=!X!7#4Z5CTG=hwT00W2&n5U#GE>l#&+=(;A)a~cV3mWth?K<b)iTyQG;0&aMl
ze)|%ws7<v{5f5*&cJLDG%^j>$EHAa#SX*b7=S&E3qX~O2U1jycd4fEnoK9)PFu-zk
zzsEFJlrp5SZ%3O&VTb^K5`4#NgGQRrY&5(jMy;so3S%t2{SjaK%IA6FeNXW4LyvRv
z#0jpv`s&*r;+=UQLXc-Q2M1F=_`n*mo)Jq;ys$*lX!sjO2%7B%MrrDPAFB$&09I0q
zscOn9r?a@iao4Ale~D2qqSAsy)^uAnQ38c>AX@ZCj@PblQdXLakCx0vBTVjZP(2x=
z^BE%?FYiy7l$tz{WcAyRZ1-Qj%wPC(f1cf)9sb3?`j=e4dGpQVB?L>0OPo1#nuYEH
zWnQtpw@rU=KvmUD#syD3^At}#^>xmiK24|7p)Btn((4=)%92-xyeJWbB<+Mi3gRS2
zCJBKM#B*~Tq(sz;<}AZ#fprqC9gU!-9nG)~HuOy@7m#URGLg2R)?=_0!T>l#Tw|?3
zDgy*q<DtG17*>Y8OmR>s`bAA`EJjG2@H$)0IWAqg#Mv`vSX^49(`nN?IN;y^`~N^y
z&HFdbv9z$n;=%$eE30(79jdD0#?2cTqZkfH7;D+t+2O{`>y%|lyX_n6_Yzg9m<%)Y
zT141%#DOD}f~46XjN`eAp+Btzd5#SP${CVKQ`XvhBnU-Q4vF22PPs{@0<^XSt|AQz
z5EaTQjFwC*Pm&DCXicV6<0*06gj!P-Ii;!@lr`H^#aJ0~pyJItziy|)2cGx<mo8mo
zZ*QNkef?`}@9Z3LI)o6Uamw2AD#wl;BaIWL(<ul2KI6%Vs?yxLeTOSoUS{Li30mzI
z?N0mNfCzvw(BGdRFLXG2tWA~Ah@ybz@**<yO3esJjHW6Ia-$Gd(Tr+R;i&ThP;{G;
zSiHj0>>86Q1nX$VC9Sr?3dghzsC9!ITT>>Jl6YmA<D(G=JG(5eA0x?gs_77=Yi?=B
zjoyrz6U>16Eh!?3BL3(f{v$m8BR|5IzVs!&@TD&?olW2NJI6PUbK=Aa;vi-?7_hUy
z!*n*IDh+q;Z1bfrKgolaALP=dOLskXZyG}j2YX|Z2N$_~vP)g$1W7<^amkBn3PJ1|
z#HK}=<;+wGM$(QfEg;J>Ha3=7Xd9CJ1y(07Gpj>{vow>EPRF8!V_HNgBYElC0i*tm
z?u7<oFksX>pn3W%VwU+aG>Y4W;?CZPGLD(#>TRgeD2n;hf9g;1@Iw#to1ghjp8Wi8
zzpb~K#tElRo?>Htoj`>2`+auycD&=`d<;MH+27#u#Rs`?--UOdK(rHNHsJQ&jQ8EQ
z$f=XZ(b}MuhCmU;F=3JrMBer@$qUMSip+8@w=I)Ql1(xeR#r(DJIF>!XAWP~#xly_
z_MM#GAR{jnk(4ZVW6rL(X-UmNe@L3N2-_WuwG`U2RZ4abrYHo;DBqLuEq#9>f#m0Y
z=J)fz`-^{(*=)+KJ2&5!K$0lp^vTma^3Y=}FE2A3jIhpY6KQR!YsK#FE?@cD7x}rL
z`%ma}7buFodzy5WGE|kJDr@4<5d{sxP<ki8I7WscLFh|6+8A6QN#hVB{ec8!h^3|#
z72a1xPyOO(a*Xv5!MyUMrNg-sR@yVM>p4$H8!t&MV5SX&$}p}qlUjQtn!^M#$6uW~
zd6LH-eT?<v>#VFSbMw|MKKt1xQT6M$c)Q(Zb#;wXCr`1mvO=C`4Eh6R*_4C9K3Z!I
z4i4De+d=7?Mx$}}Cy>K2v@k6#!&$*fr^!Ow3ldtu2|*ae^D*?0)SxU0!;pnEKq^CE
z9B4xnd*$P}u4&bps!|kX%`_{?bB~5{&XUF<jWi?;AuzQsbw&|H5!%6|7W4{Bf2t_8
z!Ac21Fo(A|qR=mDKmF4`&4)hxVP1Uk1#aEGLGR$;D1ii0((QCOb@DXl&Yh!?r0nnS
zW1XX{3i2XnGMRGw&TVeqyh<1d8Vzswa(Aa;e`i!q($5Wh2O}Okzf7m;h5YJT69`G1
zB!p>%kP<5bqGU>0<gCO7l|Y^Q)6v4(0y0fVS`{nW_e}E2us5pd^>b!3Zw1gyLYBJ`
z%Zm}R%FsnYyto8GgjI%t65N`VjHVUZIh3_t-16vMhchQn@n`<bpXC#u`~>IDo#n-s
zUwl&n5dymH4wo)m=Kjm~BW1v?TQ?}{%A?<^lG!xp#VarJ!b{I_Y<2m_$?$GM0M-eV
z3)q_(c6(zUKD|iYNC8P%)`Uq+oW=xkjF19@Af7g;tCH17Q6`e<q(+wo$>JI!iAY)%
zYqrKZOJ2e5P_fs`7)=VSg?1WpY$;`JIi{SBsmq#pY1uR4m1d{|ZuN6!d5tEZo|n%2
z@cC=JaP};J?$7@Pp7`JsoH(_?*T4DoH_5yRV0m$w`!7Af`E%#N!HpX?D9VDWu28jN
zI?Z|ZyWipJwU;@4`qVp)(K{xP!a8zeC@RCwV2=wMCx{y<VG<KIQ=&A%xp^ZqFHqVc
zL_)LG;@Fy`nL@k0fYX{dO_6b%;kaaTYsj!aWs(;NV_0lQT-fN+?Rb~1d^*MEQ&vu|
zvk;yj=q?Z#%TOh3Rw4cA7;7cXMw7+$6WrRod9-@2EH3ehPke$P446#D?C-sO;cui3
z*4NfKed;vJ%gY2ZptIQ~3L}qLvX<#YbLHjN_{mS^w3^Mkns9Fn-<6m%fvxp5VFPG`
z(1s9kBzSj-d-n^izaATV&k(Ks;#e7n(qQq6z6=7Q$QwZ&HqgSrFO)J(5lI6VCzv^6
zPCNhcCI`ar5%v5SUY6zCbeY7Qzj9c6exqaJVf)|C<=*cV#dauVQd>u1G_xY3P!cC&
zf;dEm5mrdFwOH*>AW9jG20=ip5fcR=LWG2o*D2DrW>#t@Q_r!mTBD634h4<)P+lKG
z>4-xQZ%yyJ1S{+4D5bKJz75!~U|cG+u|#o-i!5q3#W`rCDJPE~=i!GR=Je^)T)%$R
zs}{dutivGiNqA+I;~U3G;uNh7y@MV>5aAq@rRLyZ$kyf-QQ9VsqI>GISRgYFX)OK8
zkYOIM(q1G=62Ad##7G%pjKx-ps;F>A5``%X9fxp=ByJGL2~I1bMv6kRJDSqln=zf`
zsJf;Zhn!tc>9i8kMu1fnTFr>0;MC)fv2fodOc+wSgl(0wJsXl$-lt-9Wu0kVG931g
z03?q-{4ftb^dJt)-u^DxTWar8NEW+YHr6*db?Ow(ImY85jkMwY>7Avn4g0%8uD*7a
z3l|?EKD2(g%P_#ySYiwN!yc#CBHC$-Fpde+gs7P!q*qzirJ^iqumMrnU~$QTt!Xw}
z1VMzbhDNJF9!Bi#P1xU^F`i_UWkoBNoZo1$*iC6RBkFvL&PF70z?mO;nD&`-s4%3q
zF<YwPrAoBJ3C)S)XW1Y2$nx3z?tu?H{y3K|T?AlrbCc>VQ1BoKSYBG<<i<(Po<2iW
z)eHv%8c8}Q;aciSvvp^m8#ivSxUl$+<MWPcyt(z7N>KpA*@*tM!Lg-PtPqTbGpv~q
z#Q}kUI?E|0IS`VjEgG#B%f~|+NrN)WIOvbL+AF!aU$77wPOc_wtVS$$7wNPYsZ~Wb
zn^IIYjYdo-6rB3GKTaJ;uHCssk~Z0r2|L=6)tX9ssS2Ugy!7H<_wkM6JoxZK0NlKF
zgX=f1zv;S!5Tt2Bv)v$RB*aNfqtPIYB0q7Z#2JCou)BA_a4?{iHtuQCIUJ>RWX3Qn
zb3n2`@f?jPj7c(0GmU6A{DNV$_w8O@T|RmbKvI-7`@;#&!QP}s8OLfnA_xUZ8nKe3
zUhZSgR*o7mGHwt*@;-v)V~nec%~?)>WLF99RDw~Z$t!QvW3V_0fe3Kg(Qda{TU%u`
z88I3U+1uUo&1q)|!w73F&1RG3r6rb^mswj|V>(Gm(v+&MQOcmTVQ>F{Ab5?eD2e7K
zS$Dhpb3pCz6IpArGG|hRR8AmkM;v=!cPYGCmab}~3`iPH;@G<|iXb2g1M;$Fm?;LC
zqAa};Su+;2ngOk5N*wuef|pJ)1ZhOt?Gl}Tm}!(So=nKBWT*p%%2CwHA38!eN3;RX
zkt8wgPKP8(7*9s*>})d_4BsS4B7=Y=NoY1xn#~4DoRX#qK^Pv{78`AujB~cO_c*pb
z7u*zg=WiqMgx?a&QUJ!*9E=AfQ9?5anbie)I3)@L;=~jF^2rQmA#JvittRbN?;juF
z7!QZ+->SK`Uy&6hCl_N*9gAr;61tr(VHlC;GwP~D2uG)p(q37k^YNeO)}0$<gFa&y
zuq`8orJ+`aN*HWaptL%g^?{V!fB*fguB|bf&A9T)%Z#UQCZh&HK$@ncjg%<%^72WX
zc%!yMw@qUy3d82sHfPRUq>-ja&v7@QoTI?f3Idca*&Xh4YV9PQG$pST)i@&v0-{jj
zj3OIO(Um2N5|T8gxgbcQ22NN;`}^$8G%s&^^xmnZh*QTx+N~Cg-4((xVwR05iWvgO
zVk4$~?mpVT{HM5f?G@B$$WX-G5+TD<Q7es-j!4z`?ShXujCkyQ@1wi8Kv@=i_q)$h
zRc|iEM^Qwx*&uDEL}`qW((C8R*Efa62&U7Tt*tE<S2lpRpMu}<1me|7p&=bYz%ZMz
zGbp&VTXJft$^GZok&&R7WdK+~r@jbdId*)FXrbff$gTCYRioL@XFUB{kKJL(@fFGa
z=hs+UT17^46iS#-&L-esZFPe<3J}1;u{Cbb6j#3cG!MM*LzGEMkriMp##yp_LO&b7
z={1*C#nVqcg;F(N`|4NN+v~mQ=Rydi5L9(VQ5Ll8_MDVw=kgpD>%cnL@Avua=bl6Y
zX_~$#e6=+WWh~R8MoMTVf-s1%HaMclA1*>i23d1&^jce-tZ@iLyMsg$bra8V5QYeE
z>ua^299BxKkp3hSglH7CwFJy1Pn^Kmxq7Ye<v;g^NimzvI5;?9FdCqgqCe=r?G$mS
zg=LJMdxPjX5z_m(TL(&MilRg~K@!L4dpL<44tXpg5RS60*cptdR7M~|))rf=Ew_mx
z3C0r2fV7)JfR~Y(KaQ$u2BRT&dQ);ESXyduY@x+MGbWOjFbL+5;EW)M2*V~}yGz_`
z5gk9x!DPzb!8MeT6xLB_i!z4F7%HXExT7&L)-oQCdFiE>NYaRxUV4eaaBw6?a(H61
z&QNPbRr{o8^jyi+S?^5gAj=DOcJ~N_1Zylo{5^C}zfV{k0nn5l`KOHL&Ox93VGmO~
zE}mTC)QM$mU8AdtR<l8Kalx0GmGUP|DXH>|gZ&ZD-{_-(vnQLJJARCI8WRK{gAlDe
z60VuHX*3$hI3@`t?Ts^RKl>f>;U0A$DJ<yHmv5A{%tiy==q%%a>o;!j<Y#}2T2(yv
z+;`u$dzR9VVNn$nWl1Olv@sZK=2~X{UapPh%B$Ciq880Y)2j>IB^uB;M+mf~BMK#@
zs@Xa?;P#8#sM_(!xmAuIU-nLnMM1mSpnY}~i$#~7yA%Wg`DDbc+daPX+K5JExp=0*
z`IGA;X@U$u$Pl9pWs%cuE&A7+BsAiP=E?iG@%hgYNP|vNGV8Eqjd2ELEnE9rZ+YQ@
z+jlnk&Qnj3=QEyr{`t41&Oi#R)f83X_5Un>?3_F5omvOVI)3Nr?{NC$ITkvdw~f&|
zoIo5B=t$|k1Z2q0LB&_E_Bq#$IQP&7?XEYWSy){mP7**c8xAo=L9^Y4K$4dgBniuF
zErN5WIs1VRkft%JzeA_a2!*58H9?rrZgz;$ltDft2qM;!7P`O3ojbQV|Hu=#AVSTm
zH?g+g8n4~B&R_b^{}RWJtuh=9Dc;ijcFt0i1^q#vgZ=@52pEq?ltqqGHAX96N&?2y
zf|stm#_{!amKPWAdfy@tL{bnVQA(q&Wq&fnILq0!E|-o6kcM-EA?NuG^U()<Q#}Y!
zK|~xzh#=&|gC9b)mso7Z1YrohZS2h#(M3T$o!|uQ*@#?8TB8}EjmfHtaZwORNvSPb
z1}JAy)=NgzZ&}#C^0lw>^wUqxld2aMtgCvyEB1+EG8wbGze5-W7^BJZoGhDBmW3CG
zw3f1}7)@rZuN`AyVS)XA|DGU*0gxaGoMl=Vvax1b6kI&kp%Dw>C?tv_f;pVnS&OMF
zg!D<&IEM=)Q6r*ze2w<;^Q@jd%R(%PCVhhOK0#fAD7`mI5K<aPWh_l6S&$(`RWq7Q
zNL!0&>yx8Tln!$wpW|aToAJVnFEIU|ri5Y0-rgS5=^G^Oox>PKo@I<jBl?3r?Pi-<
zKBKC<7mJ;nkJU;u%L)*Jg>Lt4@724FR7z3+1hs-Iw?`a|G#@y($l7wtn{r8Lb~{8-
zgsLja$%G&bk&Oo07;5WCI}MhPFOpn(h!YP!KwRgD!7ho)aaN;EL6S6SwmVeDFxHwy
z8L=!Rv+;y}e?V*H1j^XCv78rUeJfhhSj%Uhe3C1#Tp^4?uHU%vHZ=)j471sc{=or*
zV*}zKWR}gS>&l;)ttUQP3;n^E{$R-2GiQjS_^!v#0Szfhg+Mrg)sin>?a>?7{K8`!
zY-}ug>Gr5avb;<fMU+KJG3*nC5pk#O6|k+Nwb)|yq-FhyCs;ds4l^6Uc%O!J7^_e!
zr`1}dv%E}J7G&145jP1C^bhtyI?{#X=-T3H<E7u<VJ)mw&Hwxl{{c___UDM=kZU(?
zyzO_C){G}(4h{}jT3SX}N0!Y{b>-K?!%dNMT)(kNGhN`+$&*Lc)`#(qFy2%MA^}QA
z<fZ2M%^ITw53M&y;s{+U8toQQl6VP@A}6$l=F%b}PN<cl(P$7yAu>v+fAr^h=#Tz+
zgwsr}y-M%5KTYG#%S2I3lBS42@Z!#Z*NOu4Cdt+&WnK~_9pX5p(`{2b>zkJ5?U292
zNwvGb&+h&^dwVzsd69E)u+P@kmUmCgGV*-(rY13XKEy*0J-}lRKT42z8_v6p!(PK&
z`2%eo#=If<y@|}{<|F@xeuD6x=|Ld9k9U$H8V$nO`!l$?5Blp0zVj1>R}!2P4#k}C
z=XbTg+bXV<qU3cM_3zcgaEIbz@Q#$oL(L-3bqMYOLFD|2ObR4WDnUNAOiKgK5s10O
z%i*<K?*;Fy=i10HK!zd0`DS~kYq4Pqclk1l_diOj)k5_5sINVb>1`sdM(BufW!Sfp
zNp6U4-eINr62q#(I6;)A9{Cjp-cR1W-4ni6ioITs!ElI>^Ip&!%2y80Rmh8+(P+qU
zIKXJdESpl61zKx=lAB+@3<A<LK?uR<Eqt7J69Pdh9L@zOZP^`ZZXZ}S8j|%+Oenl<
zLmW4dVc>0w%aSCC2pdf-evDhKmIo^=uF?I)Kg#J3eH=R*GkW&h%$|CZ=4?O|#ke$K
zQY&6M$moY5E7$K3-MQgUs)V%LU1SuYC0HqrdVg<?@pQuTFTM0VugfQp3B7||mUfnD
zHXCHK2~}ClCH-HYf30?l%a<<VfW7|UuEx*-fe<vM!x)LO0h@izt9zFBuO+N6HL=P-
z6q2-Bb8vz}7a7fFlc?23S&IxjIxh($g0q)6`)B_O>z5v)JlLi8jW06!)^E~QmMD&q
zHfE4le0#IcC{4Kk+ATix=nf)CP*$_JxB#I<OFy^2*X7Kzj90!7e@H3Ccr;{ZcZ+7T
zNfJfOW)sRH_u`RnNKR|Vj`7it{urBE+f2*iO^@?-A|Z~3v}i076|ghZl+%#2Z9$gR
z^d}R}v|3cNjI7_sRuuwAdv%qd*+5%I+HN9+WICQQtH$(Re2V&OXIa`f!-<n;X?^S$
z=>0GI9Gmq>n+^5{L%w#PIq~BkW9#Yf@b~}9|ACc99%XTDgD8$Uwy{Cj^#m_9%ehwy
zx)-x6$%+YO*(EOu^0M?ODEIn+C235aXS{snWy-R=>jh8<L33%5wA&?D0|c5_c%v9$
zH1)h>pi6K08-;<l4UPQ{sdj?W2=)ud_OL<(2Di0KFsrz5;Q~q0fTV>;JJia}n`tmh
z+T=mP?AC1vccGDzF0UY^AWkB})H}Ut=iYJUH^xx&dP&RM@bdcS&XE@dqw$z1jLC|E
zs;ZdIrW9p)lz7}55|C1A?gf1p;XEfLY@}!>m=%uN0R|Qu0;jzwp;pT4eHnvr4jKC!
zS;jF)fi<9Pz|_@@l))A`5ztsZM!c|$sViJQCCnyR8L~4g=(U&0q7M0u+iZ_=!lfnR
zr4`!kHfg(oOk->yeDnUDUiw_q?0bcEj<PHnPe$zbcFBvJ@o-F*<)}FU?dbZW&@;pX
zdDl>A2ZX{A29ClwhPl^(OKr$<Vwubea#fL-lwvZ$)HSZIi5n^D$`T@s5JC_I0jgG%
zdBu1-W%%;5R7)$Yt)Hax_(vEG_i@jEg;p3MqL^#jL#{^4ES)~V^WXY5H*RgyJadNj
z+8PUs3v`zkAx*q=etmBYe(RvBYl<>wJRM@KVVcdTs>+`_=kr$xh~tRSctlZD5Z*l+
zFcywbSaKui&m7szuo@&R#fCe(LpqBK#42LeJHV7BwpOI=2JzY&R!YJ+q!D^f)p%Ag
zKG@-<uYH!@bj-%aX;wb+DXtCo;rTDqj$<kl@$_pwZmnNnb-Bgo|It6>JKy*g?F;wO
zU0r2qd719&5(oVOW|Cp`y{(1TL0Nmc#jKcOZIdj|scQA6wJ-=IaU5{<+N%u5GujIa
zZ+e`!ClDz~*H_RJ!>lkcw5)}&(X`~bWHcGDwz`N^23wR!1fBIYqLoEJ&<G+*t+=u~
z<=fW>JbUvF1=o0}2>9qv{~{`za{T03Zm*ms>u<A=#0;i2RkO>-e&JX6whj56U;lNq
zRy_90zrx}|m&L^nG77!U(R=7e7zC4P&W#&4*}SuZQrg>z7H=kcB}u|?IONKeR|w<e
z?sw2a&|F<2?JhF7HAEQ8TGI=UNUf==0#!(K<h{DXW&<e%K@uYaNl`hbg=0Inyx6ah
zfx<=G#ETh=mmi|jT7uS$xV=PCYG&Fat;XF&W>L!Q>Qz*hv2gk{r#F1_D^4TgM(Pci
zj_@vT4lj=Q&ZMKmlT()GXrn31isT?gYt3jpKH5F|gO2~*LlxS4L1P7ROyk5lW;kI|
zDq>;jM2=SAsLO&Z&uKMdEP}v6lqSTDCNfPBK}2ON2c_o5pyuVR37g|F;nswQO~ieb
zVtH|aXkiW0=wd8rE9e&%d3uAj<EPmC?sJT<U!`;E6e}lA&}^n8%@i4hC~Hy9aW6$B
z=b)+;{lSPzX~G~PFA8SUDOIg}0>LAVtg&qG?$T<u5aC^-h@2x#6T)tXGE=CDp@HLg
z>S#(wHZ7UV#x&CustQ|HB#j1*<t0QT^}q@t7-TiC?iD<DbHLTTU4mOfemrP%-~A79
z`uus~GnW~?@NJqXj1z26D%wACfyY1j6AZ3xvia4oaQc&<<i1B9rrl}MS@6|_N*mN0
zHYj%+$~Xq25$ft|1Ttbg9y1t^QR;QnoA-s0+_-t2gTa_f_uqe4_p)=2xYI&pihQcr
zA6lB$alT^+oac9p@)3cQ-Xp2fG&(I>8ygUZM8XqZcSbXwyEEf=UfbpN@PLKQf5uPC
zHpOVn+50ZhyzfzVpZW@&hCsW3Tf>6;e)fYr_ShqA{?7C4e)H>`TUzAo6OYqsHCR~e
zl2;X#HkkLY9#+b*x8Gwto*`w(csyn_8hJmLc?cnqQu3W=o~5W1Pkiu$Z+eWkCy*fE
z<VSy;=F8XEf8`pDN^-JW(rrMV#|Yz?jRv&Bg!V$0#^MsPmHIM^6YTb<eC<}wEwRG3
zh<Sc?18Xde8~ZH1@LiH?DL?b`YfQCdK?#&pYz;DIrRDjjzQdr*xp;P!_2ra>NK%%$
z_ggP2tkQEGse2nV=N!Y~gyHb}ON=`UX(J^`6Q;A`t~H%RAZeXBNq1$H*>E4Kj5wNj
zJtF5_0jEWT4g4;z6GjLFqk|GB1P7I1tB~C4DPGyH2?9x17@CD))SF;5M57551-7mT
z<~Fz4u*cYixVpqTO&}bt*t1b%0n&Me-@_`VmuA1aF+7B&8g0po5@#)SU3+Qpxh#j2
zzWnB#MeDm4d2`MYBr%JZF5vELQ*Le&QqxWhX#`k9U6zzZL6D^0C9c^(#4*~!Fe}-f
z!Oa=mm{?vpsM$Wq=}ZIU$_-r9;N<D!tTdWPZ3&e{*NU<%$@Y5)E0|<6+PY#Xbac}Y
zVKmnGx;@t2^|-z9+Pr7VjJzlSaparwX73N5ioKTFy{H*m(&c5=Kd`{gx1J%(6$>rP
zv5q1PBv^qe3$pQ$c9POvT_s-Z5;PKo6cnZ5+J4Rp1K44a=chxi4{AEal#|ytnapx5
zj!v91)d9z}WtLYQj0$9L$c@{Z1g(@y$6Bl<kOTo*c?)Y*DsQOv9`0qOG<&-T?7H6f
zd@lqZw2&mpk>S`~jiE8LPo3iUh5Nbs`LB^>g4K>;W1+^_n7~O?R*>ld-DaEC>KakE
z?d9DANmVGmyE)|PgMjJ6S)MQV*~ttm+cPd*yUER)HxOY+Dgq`oxX3b^=Ijp(TGy}h
zj1(-aE^sj_SZW3c=TKGd1sjXG_su<xo{y@k+1~oyd2T{TmX?+n-`e!|z#HNnp&jj0
zr&wQI<e+yzl1*tgB~hx0N(CxJ<rULrf&hvvr-P*`E1uaJ@nW??);iDgH(%uW>#s1I
zT8_o1>1~JHzP8E7KJ{^?{XVnpJET%z>Y7wq9%>eR<=_1)w#G9~O-p{{a+BqaRb&uS
zj>n{NOhX7N=UricSP_K*LKw!AF{>-f%(Bh9eheWco%3f{+}L37>MOX+k}yU(ixdW3
z)np=~7Lo6BG#WT#xtWz{C)lbauNDzoJ;k+NO&|kKTf<VO84aesu{W8KmK9DZyg{iT
zA004OjwXOLgu>BHVg!&#K{!X{ys-^-w@E}PjZ&JDHzQ*X`yI|XUw#PzjF(WkYmaY-
z2t(RuP9ma^YC0t<YJ$KJhL!+<t`*~H!K@aviD(BJp)>X}#lC5=SvPofSa7X3WcQ$^
zE<1$LDx)4uhZTBFGs}ClvKfi9I4nysOon^xPsbElPB(Ncv>GfnQ%YxvtRcoH<3RW_
zz&({<tixJ``W~gOFbs*~m@p7Tk@Ut|cY6wUjvz@`d*l+;Y{KBVS4aXwqv<yix{^>^
z%Dh6x4XksdRZXlFJHr{z_B6Zk8SZ3?E1R!yb!!t_HVN$p*IwKsNfR1rlitnikk1H0
zhpB5$G$g&f*Z9n*Z=j|bKjPrN^##JkF5_7aN)ZQ=M9kft-@^)Mzt0~F0cjdhRRv)f
z&}p~sKIuD0n5G>2z~c<}cNstT8c7(EG&Q=?SQSC7$ZSN|l2lbiGnd4rV6#8r>)SQs
zrN_9X`aJv1FL85ckCt1bDK~iKx!XMQfeUoo3*7$hcSux8NJ3p!yl<ny-nD1>jc1=C
zn$Gx%j^x7eW#sY_!|@nuMG}QHtYXH!Z)V<2v|B0T@sRUp&(dloyftOLErF=I=IV_r
zL}$;?e&{^IZ$85_JDTI`C9#2QXt_GmygKZ&+DbXI++e8{@U5E%Je@D`W54hzuDtRz
zPd@Vv1PL3#{VcUFV^xFcI6%snum0w5;$C~1-h!mA3Yx8es3;lsc9~kZGMjOwKjPAG
zjHJix)=d&=aCk`F4>N%*Eick(#q4a|WNmE&ZS-C5Txz8l^m;UfV1GKr<ufcL(ixIk
z(@-Iy5`@Y_5_2hVGGMO=(N55>B$F~^mQN|mj6ep=iZL><?Dh63R8BHIAU_ylW@GGV
z=1(p&%RoDF0Amn?5z9M!2%t9@GMY@t@`BQL8Q;_H<*nhIJ1Tz_ML`rv1d=jRRJFdl
zQ;37IDCvbY&1Ogu2V@gVt}7O;CPETa20c||W=f|qV!0JVAb6=)k}aMmYsB2#+vm!S
zYm`OEVtfLXEHD}fMx&bXXv*&9HI@!;vyhF@s`Oe{vyAQBGN}!+5z!t`k%KY088aCU
z&}HtM!uOaw-rZPQz);l%<1uJsX?K?Ix&b_yjJVO;COLNkduNkddsD`;X0fS}bxq!P
zY!?L+)n~oi;M{7HSy}Siw`OqW1Dw3P#y7w78D75q8YxSxCueA-XBqVcGF)Qo_9oZA
z^aajN_Sj4X)2d{lE7+Ni7|jf$QV|)=eWMAB;|YW99qh0N(jk7Bs~!M?m6ZjIF1WLK
z3u{5CyE_d#pntH>wcXpq_n*UVZgX{S$}p^1Obo85DfR^~_9w`&$MJ51bE^&d)0}^Q
zZA$0ipXA))5>I~i)9g)$bU49c>k^&DaeBKIQM$})FTc#Izx^Z+n<>}KjG3ufSqQk=
zKcJ`u`$fs~3Qi72bcPcSZrvi9^l_meD1Wf`@LMAYBqvX-GaT%5=Z<~7@A!rgZy8Yc
z&;Hpz<DdTHf8+(+tV7K*jIl(K=M#WHJFjDvgpy_)VytFT83-G6Ivqxn0ZMBELc$;=
z3?hVZ7!<8`f~{*JqliO!)Of9prE&t}99kP%aY&=nMaTeEWaMQ<rHn_Z{19*6BuV^h
z)fyosQilBXzy8;G-}~P8t{>y?{@uUJ7oL2QW}2e%0-Pn_kP86vNM_xi1w6Cc^E7av
z0c8co8kDvO;XRh6lr);L|9xwaM*Di(+>*9(V2=9Zjv$PPQxD=Ss|uwwl~S1bG4uyL
zhKKcKU&8WAv(`CUtrmalZ~ZOKo;~}n9!Dv~-~1bYgBM?TfjA7XWsNf$iA72`N2>{(
z@KO^}2m<N3G?jL^Ai~K2r8TvxaZV73kU)k6p(IH{qBy{*8fg{s2=?iy=N}e`gz!=y
zNit`MmsC}S(jG7WE`V+K6jFM-XgTMNXk*yeIL_bx+kcytm6dn(7?a6_zxLPuDtkLS
zUO2QU&{d6)j!?S!2udpBeN@sYAPEGe)?`MIHakcun9L?v>j*_eC=<dk!Z|~fLL3E{
zvLF#&F6R(pUs(@&(AFWXrPFE>Hd<J%Df1b5tv&nwhdhQ!5@U=<J7}%B|G@|Oum9$M
zeIyO@yT#__CV%y>{#VpxNgPJ#qCl55GVtI8te<na_6t%g2?-?>wPt1n?am_BDkjqj
zLIi{&CJYjSFu)i^oC@M7z?KD#z?bu!a~S6+jYHu#4}o)ZI&H$ViB*a+o06+J=f&|u
z8^a(7NRx!RE)i0ql;QpFe?NcoZ~o1r<M7)Oh_zl8T)xXA^Wk6L15oo#@s4m%n6q%-
zn|H<Ubr@rg<UaoYac^O*eUntm4>Rs{4*#z?4pDG22!80-^~QVodp*}X!#4ic9fxS7
zw{jo8@8585*ZI4xhlgt+N3i|>sB!cX-d4iO^PJ!Mt=}TgbMM)W!`atMu<z~&cn{67
zHw=t_iXUuFJ<@7A)NbN+xak{T!_lYnf3?<}KYyN&fBfU`dNCV~Mtt_OpPes){(Emu
z)9>{9-u(3sd`Iz)4=BFh58vC<y<uFn);#pkLwx8%A9~k+zqPf+7r*$$cTB`@T=3sy
zr1e9(wcb`JIC=qZ7O;DxVBZlB=nYz7-(#vB{r=kp|HRSnX{~wUi6?mEkw@OuYZgNA
z>Z`BvJHPWggyFkovA^}c6>sQ~xwriM_0JDq_q)A6UVk{};e+36{t4dj^SA%rJFP|U
z_`Ui6tTp_^Py7Vu&!2x+uTu!Y(@#Im3okrRG$$b*Lfggnc&_hp5%22x-uO3fS~q?u
z>!F<A{^ICb-l}`|`dayVXs!9xU;WjiuirMa27>|r!+-zZlTT~D^PTUqvb@L>AH1Jd
zC#IP$(Q0+EmbnbLfB3@^jC2O8)qFB}y_F#FO<XAh&lrBggtS&;9Z2Ez{OhVBj$_hh
z%PXx4i8cmn^jzX$?uRY#rHn&44(s32P)0;?>|u3gUhc96qYeAJSGj)Ww<xoM(FDTg
zaUTEBCuuaBU=4PDq}E!^b@6Mo(d_N-bN|DS@$rv;{9V5}+uPgx+OPc@lgal-d;Q+U
z;Y2u`%=1NV{?jOu$hk;jnx=?3fe|=z8*H`O#Bns=1WAn1%%(FOP}d4oE0&g)y~jlq
zaq9FL0_piJ+uK|G!Y}^(T_ungUwn~Y{ncN64}y$8ypTe?New~H?}NkdojP@r%a<;4
zZ2cIc;gG%keex_PNqqS+2m-8yVLzjPFre9LaPhwLOeSN}X2R;q5>mqM-VVLq9z~ur
z%L@MX|LyO*>jVPu<u8AkU;DLRBNT4{ONhgDV?Gz0Uns40N1oZ=Z(DKixT$xJLoh<S
z*<fLz%RzrYRlmK<De&XoXf#ODgtKSPa_-y(s?7K@r8S&8dxnP|eE_Uyx=R7hnz=hB
z{MM6SpxJ7@>jVPuumAPG<_n+uJhyJ{G8~O~<dKV9yu1#CEHAAiM2K-7d5B|P_7(_d
z!5;Qc9YGing$Z)5*A;{wVtgc0fWc^cRQ9c^5}c#iY!W8E4dAT982ttpP>%`}(jB>;
zTF>G|h>$pp38N6^>!7t3W4v6-)t5g<fA3Yw90oJV<%d7Yx%=+-GC+2|Zd&8Z-gSji
zUVH1g7he4S2_y(4&ZM-GHJa%r>#HaE;QQat!g5F}S>w!^`{p%bhm?R5ND&|<2%|7M
z2Wtt!m`2(>+WmxqU)%wl(>UQJ$tR-$c{Zlb#|T%^?Y4-L2DMFCSv^g&xr{M$$2l~@
z`s5S@ewr&?;)gCIiej&PDFd7{^BcobmNid*{h!b*EX7o?x9^x(&Et<gz<n1V@XmJD
z3jr#vsH(!N(N;BAuiwDM>0Oml4sBfpAjDzk==J~0BUt9VTQSf6Z(7#h{LdTWZf{?C
z^i2LJ{`yYikjWs0JYv0vp(F?clKFdu@T<<s$}-KSm#_)L7^O6W!C>Azw`grRbNVbR
z%d3P@#DxnN{(tQK>62{9dFF}z-2JSv=AL<%yi4sHs<0JE0we*V!DcV)nO3VeY1(FF
z(*GmrtIT9F(~QYx+Io64>glG2WH$&9BmodRYQJ^&n|o$t>}Pj(eeiQ4^Db4Wd$~w5
z8!wBB8<~+2Cp^ya*Wc%T-seFeNzxQ4)Klz&-SBF0!6yyz(7&6C&);N0Q7+z7QrE1E
z(`AzaJkJFLt2Nzjhc`d>I=}Nf-(Y)to0~Up^5KUca(a6D5cZyD6{a|*NDb@jtGx2k
ztL*LX((kvqetnz3SKPhV<z^rl42}@>BJ?!F^E^D9i5c~^a9%~Z*&N8)eUj3EI)e?3
zF=)46TU>9=aJz43TYSEG^fS+>Xzm5RCs9(+X+^Z+3q07_2dtBvmtvu8pp>H1ZWD$9
zQRws13op?2R~e0l=&IoI_67dVZ+?SImp1C_-?#=-={7yb$ETkens0_u3TapAwIx-d
zc=`EP`Sq_pPns0G{K8uVQHw&m+8oGQty&?i!Bz>na<<=L5IfZkT<tRm!g?d2vBGV3
z=aU)J$$&JOV9PmCAnEpd6qO=uud%wajgZRqYnc+PL0FK=qNJqKC0bYYdh8KLEutt|
zAR&!)#NLCOd^h_xCn3znaPPoyajnho{+(~&Nm)C5x*AJS<|JuCnxxE=jDw?-htIJ`
zCrn7e>0r+BPD+t$Zr$AF&%W~!-G0Ce*Hf-vy+YtYr7glpw**<|l84m@tW!btJ%n@_
zf)E0OrYJK+?TS2^4EXqipE4coAw|yWQjhB|JWmvP93Sp*@7@F6FC;4)SGaijReCGy
zNY8`1;AxEn=?b4p$vXPhGUCcz*iuQXlo(qw9zH;+7GbA@9E_L^N8G)8n`_U#;w}ON
zQc0vyNH4&T1zISS^3czKj3}w_l=qMq#>3L*XZPnXz~&-sdIBE3D9=A^mdsCwZq88I
zMkDCa7P#|fM_O0Tb0c=o@qNc>x5m}$LE?EnLP%Pz4%S*c&qoTyrAwFD*w_TX^L)y(
zVt6`w=1gocPv?L3$WNZhXzA(soG4i@!n&f;dO?P;HR;1zXtiQqeDQf+d+k*&Uc9)#
z&f_?y-EJ>-j%8Ugnar6@XT))fUav=%6;zd@NauNhQX%cO^JgE8PM?~^+YBjci0NG5
zXl_!<MTX~oR-Y~B&QB@MV5-MwMLcv4f3_iN6`y8@WC?ty^n7x1!nJGH*xI!0?(MqG
ztg(w@!&;Ig!SfVHN5_2d!3X^Mm%dIEh0NzUJ3Bl4=*K^%-T5ur?GRIyPOaH$d{3dx
zGe{B{xbfi;*os*?;|G6!gR-&|S;pm!Yjisyps}_Iq+m@&jhWRBo*t^!y=zePvdXEn
zTem8maqsqr?A`wWT}_Dt$@Z1YY;9d4O=cY4zst$#lrUcA;`Ym|Z#_rQit9Dk5f`j<
zaz0?|z9CClmB>=n%0><)5=t>0?~!K-t<D;0ks<7i2lwtUnM_z)UPI$*XU1BD^6<il
z5J6GubDLi`p13U(Qs8+ZQhJm+;r;jas7gyV)BN@~zCf?rsin3>Exf0}R<2<!@bUe?
zEh^G2z)E>ee9mZ$)+EW4?|l1z=G~uum$%-y%I5kyVc;TbR-{zAKuJR~J>vNACJ*lZ
zl2_mQZMLs|9;Mt#Z*U5%N=5ZI?ZotoGQ-D5DG#i~V3;Q(lo#{bm;X6Z!s+P0#~che
zK0I_%<qce81s-68GuJ6AgHRsNQWk3zYK1+w*fkJA!@{y>&$aMS0C5DZ24km=98He$
zXqNJ67w@B>gIxriMj?E0F+Wrna?er9!}tBgO>4t+I-{y8=JNz&EdAaxQM*MH#Y9nz
zF&169W<z7_LZtGqGM*G9|Mo}<0&B&&nY0U5Tv<c5LEy8swZYqOe~A}fcplGFj7CGs
zvRq`Od7dMLq^vYal9FX5YpYB2x_#0#C(jFt(x)g3+9IOc?XtAgV{q~eY6$0sv3B7Z
z)6^=O;P!A05I9%6ako$V?(8Bvc-mpXEr8NxLj8V^$#}xa@qkO)SD4Lbj7Q_<UEDD%
zNtR^<flm+w+`D&=m!BKZ>-9*|DMeOr<HilHUft%^S6)KkdT6xPC@CM_eSPNRw2p{i
zJq#7Y(S*PF?k$uw{N~pVxpH;GDcKr<rq++M8ZgejA&8tLzVc8~p=ybks)pEgRdMU)
zhy0KK=Ks!G-}1^!*Jy=(1hzKFN)b|HtAzar57>L~E|;JC9Iw3b+w^)HwfLqhDR_Rc
zz$vW-t(|m}=lOMwSx{C5v*`(!p8IvSwyyF1JKyKty+36(88Mqq*<9a58|gNNI5h$z
z6xs-cmz^yd^>`w^z*?NNwF;?hNs?rgh2j3a0fz?z*Uc=9gQ84<sfxN`%tb(l-B`fk
zbk4R_QREr>`!~6B`&~YK|CdZA8P>R6ZWy?3(UYSC?%%z~>EM_kbUUE&aG!U7_Q!ni
z%kLtD#q)in5>9nhdPt!beF%Axk>zuWGN-CE#z2}Sr1Kf+EXQi$7DEq;Y=#$vAU(9j
z$zu2+eh}1Hrb5VPvtzZ?67i;5+;n|2RzLJupL<Sob5vLsH{n!I&Z5v$9Ld=*4?UKm
z7HbsEu8_qps#!psM|xBT2j>Vl8jU$U9Wx$H7>_1cD`>a7wBin}xI+}iRF$bso-F!r
zpp5b8g8DQC^rv5pMNJ>CAwQ|=?AkpJ0-p;PHhJTX*LnHnmx<$;!C=7N-X7!em^^oG
zePI~U>2w{fR%-%3px;{}jyt4jMwYoCQkE4;DcYSjOG``i`rXHNexG%GQtW=x96Up>
z;zOi?ryXa$6B?xy>+5R_hNq-S!b>l{K&R6t2>iw3F9799l4MR^WF+$`H*bFA^qPF1
zFbtVZ=e+;k2Mh+M^*QW}*=jubsgKbes3Pzp5Q4l+Nt1#ksknV>A6-^RVG%;R+KYDU
zp)nO!)ES4Y*TQ;DunwnTJUrz7z4y5B;m>(+e}}4ai+mJCC?%Os#_T@WVfVofRh1A&
zK3O_t_ueo0@n8HqM#Ft4z^EGN%tJ~KfrBLTGAAqM<at6_R%l~uG0zF<G^NZmN_u#{
z!0HSO$RKc*+6IL33Bm|Zc@CCSkHgB3%eh-itL3AXdQ9gbN(q8MVJgXJlyL9f0WUm%
z$t{k$!Zglvbx}@M5-Cb&awH{EXr$6)>5Q8n{)D}~+w{AAu5GV#?P7=3wKXc^guTpU
z#C$d*4t>IKm9lgevuT=B6gkHyf5gV+=ectAWy-?ASI$*bVXQ`*5-BaFl9bj3r?|D`
zcFm)GPIf-TNJ&woL?Og+NW0r(G&;pd1=i_hRn!_YQlk8TvV3}9^#6YzPg)CH9jdB>
zvI|5~s`|bTw&zZ+x|a9XG9C>Mj|GpKOz~NR$g^0Ry1%PM4wdJMp1MK^!P?piuf6sv
zU--gX^!q)g(<w(sha4ZDFq_RN%M#!BSz20lSxZ^sdjU&JZF;>vN~uMUNuCGjO5^z+
zo(H{N&$(xMcc1E*7suW}ZRZ(=kIdhKs9<?)k!=`OX99)if8XqOpJqPY-!#EM;5y3Z
z^Eod(|2$<?v9r6wTW`I|)vMd=?(RC_#5Ah|O<iQJdo%7oxXY`rzCyR#VLZ}!p3nXJ
zJKVZ;hhDc$;H$H@eTE@~V0EQUl{)uEKahAhD}y_?4;fD;bh};WkyDlEg_fLxN>ho-
zp%B8=CWJItG7k4|^YI5i##%!hSibc67udeMi66Aci;6UvJ9XqDWo3B@E8yt(klB2S
zF`A>ppQ553zw@nsQJcNF;<}Ow;aO~@v8DtqsZ4>j5Q;9+3(2#D<GuG79^Ym>9pn3g
zSW0@mC6phanSt`1Ts)Qv<B~-$Kt3Ys`FI9mq22cAbRr}XtyaVrf9(oyfAuQ;<rX(@
z?2=}+jD^t{T{@`+TchiBrC)1}E=ww1GMkS0@ZBHq{yX1e|G`IOvmw_mZqRLo3{Fm%
zO=iqzQ%qG6gdt%R6GSmtRWVI+Zr(m%Fqm_6xX&+t{sX`g_#sLvnk>f^%A6UBsa#;w
zx<vUAo)_bJE!v$mJU;|$Xtg|EeCZY1-9AN8pga#h3?0sG1wuLN)J9D4Sr#34t(}L3
zEKL#H6|$R~+o1?y7nwlwyXLt~CikSnX#U1woe$rXl=aQ+ZUmxcZP>-iXw2EIEG=cJ
z$@7vtFDZ(OveewW`+#@e`H+u3{FuGn14hFstE(GqY;195`x-C4@Djb=3ZSqkl6gj&
z6nI{U?}bk_cb_`1pIBJ`n}V}MUAGfRAyLW)1V+1su+@rq;f3dT>#aB0-oC<UJmlbT
zkJI5P)9G|k|5;yO=i<eS1VM-~hA4{J+}t8=wJ0lB_b7@2UFo{6<RPV?)o!u6wo2d!
z&$!EfQXTCq7&K6pu1$o_7tqfKjz?>BXBNiKFzVMZjY_kCEX$En^1=(xlO%I?c6YgS
z`4US@eSFVDNhhIijKTLk(pvsL&F0*?{V~0M7o{9oBg=9=_~0Xkrz3Z)e}!{HO6V_z
z_$ox=bM5(6e&e^U^Xlg=aBw(eXXmJf<IG|$T+Ed*&Q7@wEM-+tRVDimKIWaDeuq0B
z|CH0C9lD|7xvQ6$PsfZ$BWAOi(|Pk0QQRVGwXs5y76k`KV?MsQPns3{^hbZr;AGz+
z1IjrwT58c&>E_u~8e5g<(wSQYVH+t!TCF9bRu2RyWw~&1lMCBdF%~Lg2*MEM`4}s(
zD1`D*et=XDi+Vm5KrD`l6@?z2jA_LozxK83{N6u&6W>$ZxqZN3Fj_Eqq^hOJr9esL
zSPs@=stR3I7+o<qxX;0ZkFdHRP?j>AQI&H-Pm)e2j0Q(cM<a@&#8V!vR*Tiu4ab}~
z88M!v%(6LYnsD#t`(*PeK^PK6A%5UFwuNwh(-Kj$M+_i{;)uB2K`2R{&8hMXAtarC
zkJsP)0&%-TRuoj)xX+;!PJz<H^FxB**}RlLeVo-Z&K`f(=l<$(KA@Zhq+fAF_Tih;
zYdX8-c|kHyov5XD*+;kAuWvoN-G1Gr><W*L#bFkTsDD%Ad_eiE3&qpR^Gh0^XstDs
zF0oj8y)Li6{u-~m@+$qMz6%})2Mh)Srqc;omLh~;d3l+&wN=tI<<6Zu92^`GMKMcD
zOGqI}lZ>J$YbUt6s9kS70^g_C>k`G!7=-QPB#lS22^UBlWZHC-yGBI2I3LbG|KSbM
zvkynwa2qJC9jk9;WtrubB@Pbu$+DDd*RRp-b?Nnc&c0R(l;;E}lXQ-*a&~tgFdmQS
zcDo4SAe_O;fZMn4QWT}Dy*|Oe<+F}3^zi)<U&UOy+~Xhp(=YP;i<_JrPPlz*Z{a5-
zm8xAF#e%Lasg2HR_s%?>a{t!5B$E@AG$>J$%}4msg02`2ju@Ywkj@jZk}wQeT3(^s
zUuHa-F&NI6r5S0Ka&olC!Ok5#&m)RL!pO(-1PE6Im$f!sRWk+yFCdCz!Z@ZZa*A}0
z)dr<Ju0HoXmo8tWGKRb;T+Kwv8UplCUO*5ADDPJoN(ms%DrDs_+G%2WaQ^^t*{9Qq
zINm>HfB%Gy%@w2)XwwvI%p!nTZ5EJ%HHPu%h{^DP&5bpREM;)C&r-ZitQ5<gC0B?n
zEc0Q)lqtejq}3sV<3n^+va;O6Qv#_3^W>D_@Pz*I8c<t+8s}u@>K3wQJ2+UyN<m)c
z%(INDNXd%?&;))&tK9`-9kwUEdcuSYibkV^!1MiQDqx;v1CM7f&4T(wX3K(w)f7E>
z;{C&~clMs=SvJk<Si8u5+-uZ`#TmWt@ndM-Usr~zGz38lr6PhLU~^-O<>h6*@|9mh
zO2zW>3cc<!dwctg$777qWSL8w<ETyG$2{w9ReZucJ@zxv-2LkJd8UPAv1@j9zE&$@
zYipC|pMRdUwN;Y&oYDA{@pwX(WfZkXe7n^qiej=X=l1P8+`4%SYhiV1ogfIAPgBw)
zand>UMQd^OJkN7X!YB%fTc1sS<4Hq&$}Zv|MxYS2LgeG~^EW|Qu-3Z%BjJ>r2ZIx|
zHoW-ai~QgRKjO}vyL|DBU*N)pEym+9VHj|Hd_q}j;#Ndi7UV_BWHRK|?OS~DtuHf~
z%+Q`snr7U%ag%G;E^*<)It>!=)W>LGJ}*#S2Sv!i{*b-hBc!k>={D3^p3#aUjB%d!
zj<8_xl;<|n#()KdBFW|qjvwHAmMfR9aIm+}ObE8tR_MoVGU*{T6iLOfIK@hXR;P?-
zQ_^Hg6!`Rd9jq;gW5v<YP7RwiLEqVEV;!|u*R(Hz0;SMaGtY9;JVRj7RgRPpx7zrg
zS2MLdSNpZrsjMrBFw(Kf9)YkP3m{SoM#Gfkbc_&+A~XEi_uj#`yG&0VEVZ{g;H|e@
z(Z*U=0}&2Ga*Hr$&U*rZA1I1^g09-k=W}LB%B4%!P^Mx&n<1o6WeinSF&>^UoDDe{
zPPlddfGpSefluifqE^7*<cQt9`&`<7p0X&MU$k=WW>r~I<T)0BAGlRD&ochxJ3r*x
ze{q*<s}KblWmyrZhF|VjSV}qmw&z*2F?d>GC3XM8-xk~FZ+|>Y3vPa97qZP-8>d@j
z%;Gx@f`BNB>2w@y&}y}b!e~JYby<sX%q8*HFX{40HITolVQVP{aSrcqwc4z%uF-0>
zNYj)wO&FY>kR(Z6G&cl(fbaXvW;6Eo_qp-W$4sVE)>hXDV@EVdvy41*(3O;4z34!m
z=XidItf{iC=-Gns0;YPxoVm~Ta1mV8m{QXW>|m`&=cAeHzuNa|ja?LgL6T)T_wL=}
zb8o)E#>Op9PL4S|IN<8lZ9e#5PP-jr?FysO&^6@win1!ursD41JG}bx>%?&jqn(!B
z(a|Ay?%ZK>a}8BPI!`-{wcNWmVx`kY$$)#e5BRtL`uj`=f;=<q?+%&Ia=M+)g3jI4
zW*2m6BQV09p`NEel;ra<?I`5r_>>D97kTd5OAL+=vDO8lvTpc08SFF7Qf}Wr;AA+%
zAV|}ccH2W&CHL;#W;&Y?g%M>@EH<Y`Yl=LlDk}#)dG52{y>p-c>3{wSVm4-T39xys
zYAl_5DFUzFh=b?aB&5cAO<&uuw6t{>Oj%|qA(438xwA(oN<z^=Nf?|=DRPbHD~xep
ze$&Y;qyh=biiNvc6t(evkCT%_4)zDEEN!y2wav!b7S=+$)kpe1MV>OB4B6Yg%ex=G
z$Npi;XgFY=xUO(*D~uM5MkkKBq^sJpRl0girP1|5Cn{Mt5oT;$xXAX4Z}aibzRjiO
z1kV##;VfpIk9L7-gYS7bil)L-%_0cTnyP;qeDtK>oxfNg&0NlB3Xf+Dk9~s={q9#4
zv^~#TbjbRFkMH?(IvqebC%!1A+wBuZ5uQ@4uB_2&b=<sL!F-;QXC-NxQB`h-9z`vL
z5Kfl-Y0Z%Se{5L5mmWwctI}c2LXhV<2L}g?MxpC_5`y9IlsqpSb1jUhs*<~R@ABZm
z1E$kC+n29$`SNALARw7E;*3>&p(@uefP)O%oe+fE(YpHF(;xdM^=ULPRMVTXcs}Bh
z_pysC@M)U`fAzsRwxXj*JLq|DcbAu5eu1l3uX1`i;`Z&^Y+cx7V`H7Yy*)NIR<Ri7
zvzhbxs0xZAXF3_PzrW9=i&vRX=P0Ed<bC%(ufF;s%Pak7-eMT-SXG`17)%px+_+C9
zR#5m%CmGXm#>R#~Yk@ImJvL%-!}+OOi>CtGtq#-SeeU1C!+0`bb!CIA&%MgE=U#We
z?<|v78C9Nge6Yja`?pDxA8~kelX+TD6c%Gkln~6OL+0s>(D$8iBnpAS)*HXd8rKh_
zq*IaBnm4}iRlffxL-PB#SX~wrg`q55fYW6Ors<n;-$~<!*}#qG$0xK0yQUOZd=1g1
z!O9XLG*_;!^TG?4(OP3MNZ}zouZC;XS%ebGl^%tpEG!3yL#m>nsw~UPYY03}#|dST
zv((dUUc5r9)y8Vg((*;#e5;2SwfVtc{+Pe~<qoAOkx~;x9Xjm}rpVA`=}JaM00t!;
zX6S;5v%QR^$M1ao?-0cy|Hohcgz;ofuWQjsPLY?36WR9^N(!_Pj%8!rV)4vFd@`8w
zX+fpAJ{@gSEH3U(gRma{*~2?ZV?H@}G(Qgfz;!Kq9{qlwIF4CeT}KE(yWJrSW8Ql6
zO}4f!)DVus8cUj%2q74cCd_6tio(gjNU7*``$#GA{g5bbJu`#y40=1m?X9qj;_LYU
z=~!V6B;%gYc*fg$POK{m)E&vqT(NkUCm(~wSdBCSB^|3to08dV!f@!-5+zWS1=Cu2
zH4H<_vf}9InB(IUk|bkob)AbBFR{G5LY`~VbV6RFc%GosZBbSUL7<4^kRbG3uo6`*
zd-1FkRPdTzDQAJB(c^X1ma{on(0=QTYw^9EFL|7wsAo7VTldv~a8gE6)I1-{WIp5L
zkMHpE%P({5)?LQqA@}b+;N_QIVmg|lb;ZT?3*1lcL*<x3S&`B0R~($&<I?4e1W|xC
z7DU1R;fVdiQ<j#OpY31)Yh4XR2_H~YCdVpGs}=L+Ti0o~Bl0}Q8tVun!gbBdx`yDO
zLFJ@Xgl05OcyRw1RR@z^x6gEzlNA-7go}Pme`Sq4nUHkmY+rqq-;R4kt%&dc<OX|3
zV?0mc`;I>EOQ`Y`TN&rUF3##_%2Q}#(3K`HaxP!K#y|fT{~PxAAMoyXZ_|wpd1XkG
zY=JsRPvWV%7eq;{=evxQ$4NX-3?QBoC}}`=c&bw~lS(dMUg97Bli%Q#*RE4l8ee*g
zE@Xt0#IVM7Jl5hYDCN=WtWZ_~2m41XE%mvuwT0Fd?|=9X!{I4!y!<?I%dwq4`rv);
z?;J8rQrc~w%i9-u=f)vmk)ER6?X$kJMVcg3Wr>mjR!O7+Atk;LcnYdo-)isR2+s@Y
zb~*&@F7qLDkrYMAd_Hx>b^)I6WIU{uDB=1~1TOkM3k2i)KAmpoi43mKQUhsVkn=l8
zt*xUeH>}G~$V8q5J2a!IlK=}$v)}I%1OY)1vbniQtJP*}>jFXuI-M?26w_*T(AuTC
zp68=6SX-Ril?r@6K-FxJMg`W@0{&_aE^g7sQ9f#>+WEDh0ZY$4SbW&#KNcsyd#y*q
zyXDiD7gI~)xVUa9ik#VOic*R&be0B1Ss;azi#OUbo=ljGr{sCbQh%B2*Pmx&eFLc!
zv+1O!ha3Dr;rR-Y7X*Pv5cqhWq|yaam@^W>vy^V&d=-3A95>e1yJVMr>bkM*iSXbe
zg%y9Ly7_s!gSEBXzp+$kJYOIcl%?Td|B%;TdxIBWc#&WH{AcX%?X!LH66>q0+`fH>
z%NH+GZIv7x9w3zm!jhLM!_f&@K1F(pN*kmKm`oFPcJ_JxxobRY0I7iql#1|VOi|4V
zBf;0d{wlxs557XwiYQBsUpEXcYLM1xT?w~1)wb1b!g!UzV2+dpuf2Maey`8o?hYT_
zyu<o(4__MmKyvra$K1QS!@d0@f|k2BH#Sb#*&m_&06fj+<~9n&JeeVc>jOc#O^#F&
zg+g0HWek(ajAW3}?FOu@u8?U(VGPE=Jeg6JITnlONj$%H7ZnXLN>h3MV}zK;`#=Dw
zB?LTyuR@d!*;woF5C75IeC~5EAU#D<<xVuL_ARi~DkH+wa8~VWAOss1uJXm-_<Q`}
zzx&tdvSM{*g-e&WS>L)sGMnL3Aux<jhe+k|(#x-KbaKLMddkJkO;(pfCi4QU$E6D|
zv$nR5wN2f{L2g>xMN^!V5?=|Vj0ycNJ9~#Dvy^MspX1i(E>&5etzj}5QIt6iZPNF>
zddII(Mxm_5dxl0rYb{|I(r$%|$LH@qEr2x78xDsL`H45j<uZ!%JLgC07>``wk3wZa
zoCzhiS}{Qou(7dD96Q~d=bwL(PN&P2D_0Rh(ChYy;})}7>LfO#6NM90b%FkjtrG^W
z(a@NtZ$y)(33*;T3;cZkQ*eIeas-8X_iWF0Mb2`z;U<F+279gs<C?f^v#S2ecYOHq
z`PXnZzSHT<)j9$nQJ?S2OUnddL|K%K#}kryO25Cvl`B`--oAq8`HaS6X0sVWx&pZ|
zCB~M7VMq`JI98HI2#b+!N2!E-rsc1JshZ$%cB@Gg=U8&Wh==~J`P;u%*vh>+)O2<s
z&ZIX0d|xq}Crl?Zu3fvv?OQj=vV_6ln9G+h<GQ#jUV7;zX7f2&<t(48vLu<$na!qj
zTC46Hw;oktxpnInpMUFhx6ysdAp~Fq)@s*WEnA>+-hTT!-}=2Tv9hv6R^|v{QMHwG
zllf}xL=L5?@1;O_9&di>>)iP0L+;&pmv*bg#>NKyrFE`7_aY|Gh+>aqK4mtYvUTwi
zt=<a5@hPU1T-;pd!<GvWfw#o9Yp=L<xY^t^^-Bp#I5{a#dMHnz!VcBEVs~es_0@IS
z%j+cBF211Fx|-MLop9{aTAoGLwP4TFk7<cLo`G1LSXtmvYIlN#2wvTcdF_QQe9yy3
zyEteeYGW8%pM<JuLJ&3F;A$Yh{muUz&uI2;{g~x$pI*C9+*zS2a^|B$@;t?q6_N6|
zc6p2IS1xlhIOfBT-=iG`j79~^%iH|sZ~c9G{S}HLTU-c>CNd!~STGt8L{Uh0ZJlnn
z%gq}l&s~3>oA3N3d3l1Amf3trk!R=ps4YUcx`?$Jf#K;pzt2sjeOTi9leDm&G~{_k
z*{^u+Uwu5faF|8*b6)xLvtZzVtq^%bFID2)Je&`RXUr3DG;;9>t;{Ws4vH~G)Sm;_
zRc`IWJM{CP<IzWT2I$T<C-mofo+?Kt_XDCR0$igZz9foTL{XcI7cQ~AeT67$F`Z1A
zj^~u6Aqd<CqbwoI3fk=!aU9o+j}rrQdP?R@iQp-RHMKNJJ$a7l1+vEEt*$VOxt0$F
znKStGj1cqJ!zZcnFya1f4XD7Uztko0eNK<(q*=k$UXRVq4esB+%js~yrArst*j#5a
z9`o9(ud%(o&8@q47RO!`Ig`oQ1rRAIb%~IY{rv-uj*lNJ0DqPt5VTv)a#~7(M6nqO
z-hO3^)uk>@L$8MVq{UiY*L2+Gw&{d#YK#KG)oZWtkN^4qfgk+Qzvkk`I_<E{#@Yqe
z*A&HU#B6#(k!MJ2S?RWU@%l9;^Etcw_xb4d9RdZWGQ9m;-{8fU-oW?O;+k;qq+Ae7
zoH{RfDxlNtv9j5v6)FZtTdb^KVt8+tegJuqF`tj9>b20+8yI(u>4#wC^W*UVA^<Bv
zE6@UI;i9PM?(Z}G=q5|+Ye?lUgyfnNqM4kAb<;E&H49<Cx6C*H!9QiZ`wBfD@^psh
zMU0Pjsfr9G0|MbQ9S)fEJFIP8B=kJed4erV);F&35B~9g$!l-E=^!TQG_Z_uRNkhE
zaZ%<m*3o*EN4wJ@j$<^Iq_V^Y!cdT98CjM-v}jsu7g<%a@I1@#YG~tpwsBq_#MK8}
zae1NMTmRl#%i7u+RaG&WOcrE-`Fvgr?m0-Psv5@ACm+W-^a7=vZ+RTI0EnU(-wWvX
zm*{qTD5WSXO<7b_rK6QPI<90qo|5M|N-4B<?<wo-%*o!=YQ|+%zpo2EN+~Br@l3-7
z^v@DJ-S=i|gv!w-g>Wny7gQFN5Ldf%fV2w-brIl`gWq|r-$p0O)?|~i_Aqw}t4-}=
zHcQC!l9lB(R#(^Q^_DRjvaCdFOW;TNGH_Xjae7*D?BrsEkdBrvG}e~D83XRAXRygM
zgutGI!fZ_~ZI%TKh;u>XeCDyBe`68$ugm1i^=U~diQ<r6zsu(4I^9keo$c0*nvyG5
zu5f&OKr){(n@!l*SYtY#psR`(UU-qiqZ5iU!IKJW3~8F+DHkNIHTa%KnHmPi16RL$
z%42t*+(Q7vK|H=_fko1#;!=^2y!$@+o39aXZlJ70+ghkmoY$JG`zS;cOx*hDc^<ES
z?#o<U?_wu=XkAj~DS47HKRrZ9*X?c#&GG&Y%l$6xFrdmy@_9nwhkWVRf0w`ipZzm>
zy&hUSrGxVsW241qINMMu&ncGox=tFWu$Ea4Z4al<HlLkC6`D<?%bXS!vtOBkfa^jo
ze3UM^phs-@0y`<W^_?HkzkGq^7hkCHOS7nRU|j*TNr_c`FQs&NxaZO8Z_tXqM0)%&
zgTWyudplq>anvC<hSSlA;b@3_04fNWj7Ma7#+SbMMPC24Z}GV=ea&6y_0B>GiS*71
zIwI-<9hAC6`W|5rp-lk-DqWHmhD?IzTgsv$OB1Xybx;zEj#nYwg6qj=gASTqQDca7
zc7GN?g1XDy^VCAVyx9dxDHp*)YuBBe=XqTVY1DI{?s{lWhK9@CH2Ha+zc?mi+)i6-
zx9C)rM#|Fp0TcyFd6YrboKEg_>y3lf<+)vJv%#=@x**@D4`bc`&d3E1e;x<P)jMtU
zSwXxBoc7^OK$EpL9ua&>R`u}TK@&V&_I*~DatM*b^wTtFHk&h_B}7q+et(&MZx!DQ
znI;KZgXe_=zC;*Tpl!6s!q5dES!+MJ#>EP2Dy)9S0D@ac;VhfA=K=>c$zV|f;jHHN
ztEkauaNw^V4=+AvQ112Tu5sncHfcI1iX!^GWe!_Yw6-Mkl=bxuR#ujoOh(M+Q<j%k
z=(O9+<};pq?nO4%H#s`mCGfpk`rY6M9!<h#jYV3);P`~9%AfvPwvfyVTGG&}COj`w
zy4vUPojdIP@FzU?d%uGU!nz66x*7=eS}0__9-gg*N=c;WA^nK;%db<8I?Ts=?BBgX
zHcwo(LqJv}PDias=q>eVb$X1)16Eg7_~-xZU-Fy3_fJ?_UPo6YqE?wuN`bMp>afd1
ztaUnW8jJ6T1fGvjZi7-(6-jO>Bd4yOCMgy{;EQ@LYY|ZaO1awjqkF+8v#_i{RwbL|
z171}lw17kihPU^4@Sna%dvz6Wxlds<vW}<%RoB_nB4{**LedZMqZm&qOc;{RPO#|#
zek&rGB!tTK6-2WcQP5&Go--MbI5|C}yS&C%zV$CzyKucWmNFQr7ExA6>FBq@?e;CT
zJh?E7MOg^<b6r&w+LBbziX2zG%$*qxIO%JpTp_(#B9(faL;qRE`2cc$ILVNv$3oW^
zm0bjlM^#dr$F4@>*RRp+676Cqe0~i)A6Oa|mbFk-?zu%#EWYPTRm^5npi$lpDgBZ>
zFA%~}{G@ao{i>``$`A%cef(~lpBK3kY^(G_ild=GTWiSk{Hf>9eUN4c{pj4X;Mq6b
z(^ld{P{q06;XdT!FWW#T4FuZMh#z^55N9FM#W7p|>rDDZ2q;S@!=9u$$HznRqM#jj
zxNzYz*RMU#^70zG(xhn)RuTrz7ose4Oy!!+{J<lO0(@T~&v|b&0sUDskTZ6qz%(a?
zs3EMnnBKg$6!MuQ&)s~T1)eAV9(8Ws-Tl_q2EYIN|B%hi4Sw>IpHLKet$783A2FNF
zxp--fwe?lfY)YPIC?#lj+jV1M$d&D@jK>24U%JnVB@6-r-y;RI6@cXU=!k4Se(LPW
z0%&OKQ(m<DY|5Oml@wXQ{qOvc{?<C{U;bjf7qNJrLI|(Uc*P=sD5YG_i}X-YOc=P0
zf;L;&`2gWtMB$T`mVUp7VNSc<Wj>!V8BHk4f>}D_^I!cIFTC+pydc8VUL#oPVE2X!
zt|US^7)=}FWZi|Tb*$?8Os%SrmQYq;wV}v!gj9=$LMLJgS~eo0kN0$aVh!ZY>uomn
zQsUGyt9M9+fy1BQpm+7BT>koRg4aa80M~V{T@6AA5H9PG_2OP+1+y$8NfWAR!bT@x
zWpxu>>|ybVdz&n4#lhY_DiriqSJ^q*V`cLin^#_TyYNb%$pu(dXF!gfC2I)H3JW-+
z7S)(3)hDht8m%q!5|*@~(j`TaQ|S^VT@kE-W|UI(PW5lH_Hb5Qe^h3o8Rwy|N4v~F
zZ9M5IXNzUC2pWVq(=I#D^*6O~mn+=6!kpH&hFr|zxvtP=YWK%881THG$+Nl&`kun~
z<zlzrG`BUGhkJd!2$HhiHMz#9^EH$6z7+sxh5sjwrtePHZ+0HeF&4!6zpZO9x-#g>
z;&~zc{wi0lT;<I-KhO5%s}x1W(cuwAUZJG3IncVS#qu(Qu(aDPqA092uq@77N7OK(
zc_#cG+-LASXQzO1rezKvKRXXTYk$XINpGw^UmW4a(CM`K+Sh)AS6_XZ!C>GZW^I^G
zM?_%@C4FY|Ii9!0*5)QB$NP>YWHs$}i-Wymv^HG5a+RH(J5)M#^M<7rhXkI&))}19
zmf_J6!{N>|Z3ZuGc6duoXdh$<E0I{VfZ6ecd*A*(t*r}$7q(osBYmeA)zl4zZSZ5H
z=OdJdsVwt+?o8{F5s@}@H&+RxiqYW!75C^x%XEihhJyjin_J{Xg7jl9J^wl4?h-|s
zB7|}*Q7f^s)&x{F>f=&AgcKN~5vYYNnh?0P+8DA5vJ$kZD9fT|wMjfFDC+`xqo*h3
z<LAI*0Yq!V)#uju&wl$G?EHs<gFpKbo~gO-Ksnbu`1TKpH&^Mu{u%}kyuitZ=mNAw
z$a=Rd1BQbsH$VCXAAImWJ9|5vj!xO$=<;9wuYZH({t~NO>*x&E*OAQW_PQ)zSS3j_
z5^dSoyyA3Zs%k;kRQ03@p)uCs#|U{=2We^<oTbiE${OZ?q<}(0r7MbBO;|370RZl?
zQle&vJpE{D9F1+|*?B7-isa3L{^%mx9CK3}X)=&zadr!)!^fXAJ`K-5p8;9xENL6b
zaa#wAvUEKjt_$6wl#lOwi*9ppf9L+*QFWbCudQJr$9Qs9#I?--Q;skUiQ>TZ3DhV2
zSsz2)1Zt|qF``)nk7hd0f^PGeIxTm>=u_)QfH;?_oS$1I)k5mN-RaWpbXi(jW@CMm
z<>eJtSJ&Ct*uok?GEXUsiXd>JcL>3JJ|oXkiXtTpefs@AQ54k~r<*S+CBE<xs73bq
zv_nXV@;s<29?mW7LwCRObry@y+5IP78IK1ev6wT2Aq;$e<2QbdFTedIj*btxed`VY
z7cX97GD?Ux+vG(_nauE%;=;u(9_;i<rc=tYAPhsM(<!srjO*84Vr_kc!C((hxebnA
z2al%Cqm)OP49NzEPaQy{6#VW#`Vy~q1K#^@{yXxMAxby`f<`j>c%KK~{R^)C{<jGF
zt61qF{P3K;w{kT?KO)aHJNNJL!F%s>`}R$aPL3$bjNkvptE?^m8ol-s{>D1WTFPXG
zG=_^;w~5<r_Kx=GtX(FIdz4vTuUVp&f|2#LP@aPw#bRA)dY|eO+Ob#NnN6u7Eupf8
zs;XTMaeYlLO9deeQc<&9AD*Ab0*JAOrGAIY&tHdce3kUUeX`s8PQn<2mXhRX$j+bq
zIsW<@QFj@s{Ig<_Z0txq?%vwtfBxe?<WK+nPuSVLPg-OEyz$&N|M+*`U^qHurMpJd
z4k@w}f|Axshv|IE;n5+j{w6CM+qG<VZ3t5@2DVARjjVT77OXjA5iAOTO#xNZf{&66
zL9U@{gz>D~;YlSB+9IVvdDJl6)7L;6T5v0FEp}eb&mVrCJ3_aEHubLGt!RxJYop!@
zVj)i0oE%PGy~3&XK6xd0^vCD31Fc=bFG=P&>Ty&<CYm&LLV)vmN}lIzZfp^^<Hg@=
zt(nbcSZhhsj5N*b#d+~g4)&=%?Vom7YYBpYe!o`-ExnLQagAMP9?$OHP!+m1u{!In
zUPyVItBW+B!DUj7VC31cyDap4ruOi$DJ6+(F~qgCHP-8(v9hwp^=sE@x4Re(X_|3*
zIwVO_r1WWbS_tcETS5rZETt$CqIij=r9N>Sg3vfZpsfq863<r)#@4gUqiuFD^%rF7
ztO;xDn&Gd`Kt7F4_~`pR@$Zp!C${8spL>HZed!B~MniUXc1Y$4+uK(e42Bp3tIMn0
ze{c)wIjFST>$1MS!h>{1sY_a63|Pja5&hm07dEy?=ciaIeC5?k<C!p}wH8rU`00$N
zrIAS?xOC+L7p`As@bUZX{?YePIcQ<2guqmmgCG8a_`(`nzx{QD^ii^|g9r~H6w3FR
zO$vVa{U7lMfB5ft?}J}5nvBra;0wtYURq)@8RH3`Zo7xohD^)}`vImj+`oU9;dI7x
zZ@rD@1&+)h9K7U$!I@IGvCUfO5SL0jZ}a9FaO5|)W~#bVKGTLuTPj_-Dzb3;R}zCX
z2B|E8Q1$xs8H(r)PS&}6k*(kQHAcsOOg5RLagkRdnS6MUgYW;4?)ny@zYJ0#@bCiX
zKzO)&!hiUKKj7c~`+vi9KBnL6;`xH2$O!^ycUt5{!C*M#()u=WE1;@Mjz&Xn-nq#%
zO?d0MFF965EeNP>GXX3X=E@i(F1@naEP5v#tkQtng2qWpd6Lf35~WTlD;GdqG?%Da
z5?<S?K6#!g<2{?hI@9>10P-Y;R`Uflzw>LmPvD_@Yb{xp)v!+a$l`tW@5c+l=O4S1
zZlQOj&n~zo1IhDj;l<F5W`S;s`&CueO>s|x;N-6?#(vUJo<|S_^<30h2BJ3TIj<P&
zj=d_*>u@!Y*O_Gb`Qx9@=T1I-;d5d*_Yt3%naJ9Erqyb(vb@U0i<h`^<tmpiUtxWH
zlXk0vF^)261ZbUNjK%jOROr}iXLIRlAXVMR(CM`BeMzOyKAW2ZpC}9n{NS19&|0Sv
z*ZkXBTW8{RHRma7E9Zm9(|oTD?%wP5*xue|aC*xA{vPw$jOCRT?%ut_jT^VQy8S$P
zo|7aQOUurSN2w0|UXMNLEMQx$7D{;}NlK^P;lhOr91k9lXEVaUIXX779o9mbm$ZyV
z70*_S6GGrc0h_<}MP_$*8NYjr%DS2u(qhIb`+xqIw6AQlw7TvVcOem;#Pb5GqT=U2
z{w4p9|L(uxosWJ=5O{REEvI)TA&fl2FlKOiO0XK!Z7*RuhIyWG`@uc-4)^G-Z?nE}
znZR>av-%tq-7q4Iv1r>g+FGi5J#^PWowZw0Z^B&$8MkAEG~~IV(v<`KZA}!Z(F0>N
zLZ~N(-X{l;X1x`HrPp6$^X@LQ@BJ8CSu_?Th_c}D$3JK7^KY@becg4mdLEwg8J$k}
z;Jpv|^B?{xvt;ZPF*nz_@zE`^G-H-z6qV(|_BOlsAMlG0e@0mpOhzO2j`m5i6srPm
z-9KXEQp)mDx3&%y3(VQg+d0))BhXsb3yQ6GJZ1r<2q6f=h?ie_iPhCr?tbudvaECg
z$GS;2m8L37s@$2=D=&Oz)K|#H?eA>ydQyF(xqmc(JWA{JeSgt8-ux`fvU7jm1gVGD
zga*%l^yiD?v35bf^*mPxG3M;GXR{eek}h`5O<JtArYs$N=!}rkMV4^>Gr6qh*$T71
z=MhF>&5W~8xEZXfB^Chfxt@aaa9|_R(%c%BTaqLsNy2<SV>X+UWjR7rX#M0h)xp@R
zrrO5zdVQ9cR|vzHzz+%h03n=Pp69t*2RK^*qpOGEz4|qSz^C7DvAVkK42d%5YvcPq
z*0>%kDV!Y4v(`Xborr$7ht`^+EYR99`|@17KCgzkh=W3)DXO2-#}d@n5Kz198Cxs7
z!lIGb8jHjw{ExD@EEZ1+XT)^pHpj<%%#%4vd8BF1@N~lU=bmSMZG(Ha4v8Y4b|*qA
zK~)t*aYU=#LP|jtIc4IatdO2!eRGr6-7eZ_g1AGPYtpO&<zcE4lpxcJf%TsLdKPD&
zKUiMn;@7{*;BY{Gcv?GefK-BPXP>>F{DiI7K1bB)gYxh_pQ_aC+&|#YzyCdM+<Ko*
zC*svtUt%;k<-vnJN^O};6P8xim`5{i?cAcuHEA+)revc52*^%nJlH+qrRT4r1h__C
zhg>uui>-r#wSv+JD$N4QcO|gef5^Jp+ScYK{iPL5Hls|OorWKRK|8WiRiO*Vz7)cd
zi+&}9RnHk%>lT%Eo6EoT6$bZqnBCfSMRF9rlFW~XoP2PDYrpZ^gmu?>md?0y;|6zc
z-=-*LeDMpf^Z75l$&dc>$Bd`WL;Lu6!hid3|Cl7LXm?|#(;4IOi1}oM^cB8pbK~|N
z|K^YWuYC0J9{==z^`CR`;yRvk9D3LIB%QUb4HgSpr8OAWk!}%Kp=fMJC6Y^97x?_Y
zdXue7SNQk;{(oRP9#R%L;39RFWn}Y|swmM`V~n8Ee>EA7|7AS;P)gx@!9zrbGmg5h
z_1N5^EC3W(VHY)~rspOIeEeD&w_)Z*QB-$*bIxTEprSg5xSh}J910|BNK^unvaHAp
zXZxIG&L`gYL&Cr=HZw(8Qk8lE4>n|gUa!m2Qir9b9u|$RO8mfcFogkUZR}{)kIHC&
zmSL@>ztrRMrEQ9`KuCzffTAckI5^_yXh4xywP$@3NFEo=YXXPGZH$D#(TW$@J!b@#
zdN=&oLTxoxSmyH?_wL=J(}`JHUdH!*hNly*T)EB*&%Mmv-Vwg<(e6ZadtJa%mN}m9
z(d~8#e0M%L#RG#jnohSzr`spXQleIyywK!DxiCkPQjnsVKEoU~Lu-vTmgQGp<igu;
zF#f|oqpaMVC?%k(IC<{|vy&6Lo0m{^U&`U`4!1t|n4{x;R#tob&NqLHG)wsS#x3$v
z<4ei+zW**)xBrk=Uw)3f%$)?z@DyE@1f4e1Y07uM{XM?@!`u8<|J#4XSKfXLKk%@&
zsu3Z9P@e0B!6`D7Rppe*n}%+wT#%`Iq~b8*o8R~b*PnZd!{bB#;D7obNYa$9=PV1#
zyd+B;d|g!~LMXIQkJV~F0m5>_m<nMHt&5vn`sx=rIXHpTfD}k1p0UjC?vhVuM6Ev7
z!r|UM4t8(hdy3D$^%`IQ)^BqA)($&6hbScoJ)cr*{@_pkk{|!#1N!|AaU2pw5n&kN
zNx|XCh&wxn91g}j*gxWAaLm8@m;Zv7UcQFw@N{BjNQDuKs;n?s;wH~H&a)SEXm?kU
zN>P=iJL!C%IF8v^zra8E{eQ;yzV{y)jmNZGp@VUxrm70E!U>~^%8(_=!VCLpMp>3j
zr^#a1`*0!GJyz;`2GaD5G&Ekf07&Vb`x}QtRy9@D_k+5?t?KXV$yoLiGY_ceQl49s
z-L0xhCKDGpvn-=5OOhlZ3}zVX;Jya>aS&EPS>_HpD9gp_;VDHN`$+k0EUfd}orW3b
zKBqzJvOW&8$fVrbzMw}fvbhG<YUH+?fYcBjf*_zQt3_ZEH5FLYbm1l|(c0EcoURBy
zpU-Hu+7v~}Y&yf*TE4<61UW@fk!9{8EQ`t+8#RJ;!m_-wM6cH(i~^jKonyO*n*GrP
zK_%U0=~>3o(hAGV%Vc?m?<?XsBFoac2BaCBjL7q9QCqT)kTS#>`ny0T>Ldgy-FC7$
zV1aAqY_7v54hh`C<t`(`Y&LVytcJ;W#?sO%FTU`KQ<>F*Zm+|1R?_Qs7a6Y6+9^8+
zzUxssr_#`Bwdi&_j81)gFCfoS*Oy_ea|*PYTvzD8ThOqdb~J>IFpSuG`z?-d-{bK8
z`|jE@3QsE1;}O#byIgqfErgUzCPVJu`+)gu#)Yj7UVZZtuf6dm|Mh?U-;$;|Vc_En
zxb@(G|I`2Fzh!NuN4p&p#UWu35l11(Jmud03HNpnDXR}TJw4@r_h0-gzWL3s(QZ5D
znqyumtWaocDf1jrIU`!5D}?mu^j8R@cAWuRl;;se5#3Ig%U53{$qJ4ScTh^8D@|qG
z_1y#zgrqX=T0cLYEa#4`eLW1;RtOQW@y4rceQ=xMPjA#6j*>un%pdGCxp$A{OINT$
zF&d6IJlexk%e?*NFZ0~B=XmEAAF|ZzV5%Hb8Y(L&O3mS5#PR7EaE4HY0c~(fn)Mir
zPWd1H{r`xR@W21B{!6Z3y8=QZq{K)cX&pW;aLq0j4=X*E)-DpayO<i-7Sc6(6;+O|
zDqem0b69Kn!4LivJ(;;+Df|WNsw_(=3bJ(mOgOPIcF{5ZsAss84IR~<`<sRx)GT_g
zaPE0ckdY|WH2<l37b`j6`7dg^@<ZRH{ggV6sHxBP96nwbn9I`5TT@fgT2odP_FMol
z#x2Os@y$7|Rf{G^C;TX+eXRNDvkqMuin4NzmdN@w>qS+0i#f6Ng4TS@rZ#f^`*10|
z=6t`6=lO(TL>xCZq^6dmuN$%K!?h3>yd5pn!C!Nhmn!lsLzkMe(k%5?KsYhX`Fzf7
z=De*Jq>nlS@dL%mQjbn20-Q9;;zOK+8=K9=vr=d)?Fb^i?^9JJ<MD_zO(_be-(+e%
zsz$>}H*+bRL`OqqZkj@!{+)H+-r8cdOUPuSXZ1)iPSn4%hFbE-*(MjcW_UWK-&^L5
z*S|=wx5~l(3GG&wWIiT}0!KQK!qulLOXDw-W*KEsqH35p2pn9bq{3=Tl4texu8~r>
zg|I5fUGenkCeYS%Aq3b8Yc#F(HLiU1%gi2}VCDstRD?nyGtJ<G8|ZI*ld#jLC@PK*
z4;haWu54fC%`ZRC(eVgRN_yRhJk7vBWner@Nm2(JTWcwdr84#AXF;qI{N$aVqW|lX
zAoTdwH@{Kquvx59SiHKg$n{4lPf}^0xVy^I$|gd1XkFHDY~4(lXY@NueB-x&mp}d9
zf8@daTWnu)^~^@UtJzdmW$EPJ&y6Pp5O)H&3spL~Wg&?A%Upi@bIk8PKo4__m5v^F
zI^y{K54iptzl9m@aQBDbWoP$@Rx4xc;%lU-=H-`P<KS?gAAEm+Htqt}r9~SF>P%|-
ztg~G%I-$cbAj?z!<9GfOTbo<_i+}b{*xX#L;R}tn^+h0+Gkuo<o&E|z)S}EXr-X@H
zT(#B|dBJoxVP)+iFTU_P)9H+NfB6HlEF<(o*Vj@n&{b7APvK|SwCGBcXC+2A4LH?f
z42C9HHC`$YH6GTBd*d_F6uh&vAWNMJZk`q7wey_|IKP^Sb$(H-*9;mN7iSZiV6<gA
zodHhnBnSfhz#|Afw61E$y^`@{ST6{UR@iEVDCLZgg3!S(y<Ugbu=Vs4NF6MZSS?)b
z%^LK%AbCc0t~0lD8Q7!Dq9({TS)Nj=mJK<pJ!x$yii)B*`_3N?z}C9LVk2E|tyAVr
zs+6+O^m|J<*rg=PGSW1mDDp*LjTElQ*7tqFFhD6$Luh*OJycp_O;s<-vYzAWlfK?h
z9chv=olf2PT>Dz+nt5m~^^UX2N=?1Eh+3UFsNp}q&crm<y4`76l$5!X35TWumvg#u
z=g0ZFPr&-x2HV?LxN_w=wl80$-&<iko)bqMunCjteJ)&BCypbq1%V$rcThC}E1A#c
zq-jbNN44RbL01M}1z0Oc=P9OiIKC91b%Ezef+&7!LPw(krD}Kx9Ho8j#TQt8`7)!Q
ze2lRQi$bCqesq)c^pxe*HBNv2Js#Y6pW~_K<(EE3nG1ToRlfY?w>cRcu{WLLD<6bo
zLAgz)%c7h5Z4-R0wRlS5d5ZUMe8B(nfBJW<uB`Hvw?B{IOypA=UDst?H+~dF^q1G1
z<0ZD%8g(10q9{o7lq8+8w6w;Tzx<obrc-9~gsQAuwq9RbS{Du$dt@>EL_;Yl6iS9v
zws8OU6f4)SvGVdYMnArRwiXYIr!|usH<1Uo=sx%v!QFS6jx6)2;&^Y3PCw?2&wZY5
zcbOzDxOw9Sr^9j0qZTN8Ceu~d@Xjc>))nPemX{b$Mttwj|BUCZJ;%3x=Qr?zz!hzA
zimTQOz<8iFei$(x3^7?jXLX6np6h)uhQZ(fKkTr)vcapbzD1Ecal5K?yGy%ZWyz=B
zu$;}|!)84XH?!3XV6A>?gsA&aY~2j?=z@2~<T~Hy()g8s)<&?igRAc!`E1rU4TENp
zJ%zQ?U?bsSOe3&(24OWeuWncG0$vcf0OEOSq5k`{!v$1#idy5i>gOrTPO|#!-@3+{
z9wj8WEbUCXrIB-K{A-$wG0$^+KSQa)!8zw@A7|5Raqewqp+%NudEKdRoMN%^Da%^^
zCC^B*Gt1g23hBmOTJ6v+-da1k66<!3P3CH?#;ZN8%c>$TiiObd(+tmZFlSX(<TdLn
z&z%Nbnx-yzJwmV$LJ|bQLfkQqV?0kT*p1pa#epKLXqrlW-+kX%mNA)3n9b&kK-|QW
zF!Z@_;R0Xy{Fiv+jkj1?-atv8!C;Ivg20bzN_j=E*RGp+1z{MX;+R%zf|80PnNU>~
z?RE!MOX!q&Ne~2B3wc&hR<$4?nzGED&&o4Wp|KXvj}Ry<5?}eq{t_E+zDjcU05eK3
zSbQNUkB*t&zC}Au=->Jw+3sB?rp58W0c-0WuD$Xizx#K8pEN7@^S}5mhX==0dFe7t
zM~hn&!$pH-HlfRCx^V}K<(>E5;X8l&=Ulw7!Idi)FqOf|%C#+cJ_v=S#E;rkrDl9|
zM7Q6E(0A1Tdd<kvIeUBeskC5qb(1fC`Kui5ee4vpjj1)sYEsV=poGUg(FN!vfjlSh
zL$Yj!u@+wmR1mWM*6UPz1InF4gs}L+(n${K-1rV%Gvn8`Vusc*=y$pG(Fgc}M4&Jt
z;%i_17Q^8wM<+w-@*7Ikv@~rN1-SDkb5&KeJ2BmEi^JnRe*W{H^V0KIx%S)@G)~|Q
zA+gdqxs8Woj*bVI(y+3;%y@4XE2zy?Jch$lj!%x6%qmvb#waBz3PY9_l##pW%c?1O
z+J(d3(?^oWg(aKUXyAaeMgI&&YM>X#wqmh!Jr6H6HIS;(G~VTpyz^q)-1L>4A7{IC
z9(wI&@v17lU}1GS-NoYYsMJbxYXVFZg@j=MLJ)@jq7hQoT0KuY-0`W!Tp4}V2hz-;
z8#Q~VUa1<k+L;!a^M}Se3(i$l_x?zQ=LLAmM=2jEy}Ax{meJKEmKt(w(l)UOqVw6D
zEH9AKEud0L7d*;>vMMNwjH)VK|A_R7T0Wgl3n&*FREYXhG@_oSLcoIURu;umL#giA
zB}Lw(Q%<XEG#b^lt%|a=i)&6wXYfP~6)EM_5MNwp!e_^0tsx2>CD`*k;y5M@LyXoO
zAD^&)aKLmrr6|lIgHy^Q3Y|i+5FWGX93ebMM9A|xqcd!8Um|Wf_MRX3c%C8(1G=3q
zSynI_O#nf=)xugskrx<aXt%p$i3{Wsi7^$WE=bcUUC+bj#V1YEr%fQ$bV5-39@$#O
zSxHG}`!btfe3g?w{TW7=I1s6zyYmxTt#zJX_IbN2S?XV8e{{gD_nqTsYk8eF-uMcW
z$&}-xKXMk@vi2UYq0>br@7#B8j0NE&V5XxnAAR&L@4WL0>ubxk9+xwF1BF$JA}=_+
zcOQ%+5gAW$eE*2ejZMNRq$~>VKe)&K(U8fspxx;(olVGdO`bd5FsH@mt~*_6>?7;V
z;{ik}#`o@W^rOEdc<H*+<*^#9^9GMM*SYl7FR=Un{w{u%@Z!tcy!D$eup|?#hAXRm
z-Wm<Ky|K#A$M?B^_a0T16ZV(sw0kV|m)-kNLnu;UD`#tIt;9Z3u#V!GDDJS*Yf}{!
zCxaty-M-JI?aK%y-C`mD=^F3a?KXk$Q{)Aeg}Zleva-Cw%JM2zRdM&;UCOecETr4j
zR)!>*&{8dmqNr)T&e$f;ON!h)Q`yZeO#Wimdv-FMWepFf`l7CjxT~${N*1D_);fqO
z%L=kAuZxRCjjuPXpys6<4Cnq?(MZxaDA!q5bQU;_G3csdK6jY?Xf(oF%i7vH?RI-1
z^!KpbiOWEOz)@tEmzU^tI{^5;SAW*3_JMx-IaIRge7800%G7gKOR_H-30-ko)bm~`
zgXbw4wP5#sh{boY@QtbU+O*a+bXB=xw>A%5xaZexV;bnsF6ho>=>l4>w}kTCjyOr@
z%;yuzDklg$TJ4Z9@M*PU;y6Hgj`w7>t!p}}4jvAAX+~8!*wyIA&7MBXc;SVYSn4ft
z|NcEDlNr-VN}g3<B~cg?hJqq5Da*=1rhy}xv|25c^l<KzP9nzDotk+SjuC1b%-u=<
zD9@wY>yc(@orNWfIdmQ=W2Vz7hX+Tr;ts7=htX)nWHLo-jqfWiUfjSb5$igbsang)
zBS|J4A00b}p6^qY6~{*>U<Gj$vwM65#u<7UZ77SJJWc3CAv}X7o6!|}fB6&2S6;^R
zJ-|VYLMVjqvGIjB(EG=vH*T`I+2!?LdyZEwhsZLe+iLOol@Kir<!Yb({awPAM{o6%
zrOnH%t*zq)0WhagMgVG7T&-_)4s=tVN2}eT*Y2{??UI)%ckkZg`1n<}E?sa<y9lfl
zAQeFn(ChW6iUMmDhern-9PY5Xyo#qh2B)VCPLF7}m+-xqC~A|YDUy;RFOX8y*PJEG
zO7g6Fs6Gu(3?PzxaLPyj_P^)qKl^>;g-tLP89MB?(wddezfPRbxL6$U%9~ec_c~au
zF=a{AZu4S)!qNT%daZzi;~k<fV9WC<6>My5u-xk~8ci0vU$;Q&1%9_!M8l9b-+Y69
zca?Tz2vtZs?s9VL2HAFx*0q;Gg>DBYT}>g1Bcjkpv?F$Qc6o6B7FjkYjw7a%2{%8!
z#ia`uxqNY*Rw%f+bJxj4_g^~O;Ti|w=o+GUrUay!Paz%&Aa$|dKJFEAUgE^}{l&s^
zzOepji^%!c@Kg=`EC9m#k)y~w@$Z|g#_deY#WDK6@9HB-ve3-(JVm?RUIY-O9L(7S
z$)d;sS1-`!**ei(4P_Z?jDuZr#|Sf~qUjcIPH<hBg&&C9jIh`_HC`rX>^HesfTVO;
zM4Dw}ncD@QQ572~tZA@xY#dbWq*s;9=QGkYCkR5;*F&Nxth2<DvdobZ+MN#FZrkmq
zm8{{Q%9a1bS<(9JZ3Rkt3kb`}?L15NVfp;$zr^qS_BZ+Y&wt9#fALd(@r$2Qm6~3!
z!wWAw&(hKgM@J`|o(?%WI>uN_yVD`?ok^DOg-Ge4E6`QxtbOYs>@a%eRIwMz&K})v
z4`VDzk};d5i!9n02k~WT&hhavOMMMU#^WJ_!2mx{T)wo0?@PzLs<W{m@R-jtio9ep
zo-iJdSzTGD-D#5-1^4dWWilF5WCgm^SR<(lS4YaTguKYG#;ZkBpYB2y2=>4KLvCN(
zAb$HTL=YjAJO8aVs9uMwzxy@Bs^D|4tg^Yij;B0yQ6hxo;>I$^_m0^MHrbac`}_Mu
zen=J`6Gt9fn;T3{NACN=)})qN?ZGzE@b0_6dgU@Nz4$Vb@6!swml5f7%HW+JvDyB0
zM0d5Wot()ZbviAC3CXjZk3RUA!Dzt6t&3Du#hp8Mn9t^X<?X9n*jQp`cc1<JJ=T`{
zU{=xS#dTCwno3(9C8R#N1|nd1`zHL+e`Njd|6RJ9TMiEJB)Teyq~?vUJjZep<4Mr-
z6zO}I%B9_FOKo1CkJ-<CZVgYko0SMG@vP0Yi`&e;@D_LP-o;pGb=qWE#=ZNyR8|v3
zAreu0YfFNlMX$R;ud8YKZNi|(>CqmifA;^vJNz8t^{*n9FV@+Jn`=`QWV0DJKKdp1
zAKc}mk8kn4Kl?MPs^aIrc#rqq{eU;$c#R^PaeVxMvMQ<cHP`o1xk9SZXAN`DGMa4S
zJnZvmm-(ahk@L?Df?yFqfO8$C=gpPQVg1$>X#LuD(iInLHC`UBpXloRS>R64CKJi?
zf}$wM^Bk?+ViCvjIcr+S@CX8*m6a8D68W`EwFK8OZpib*9j}D4ES|1L1At{^b%ht6
zf6*BQO(%>;L+0~2d6A#{ZW`h45>*$+oBD<uF2F1xFAGM1=bbq;Rka67S-6Hjtb@cF
zWm805>(OT-j7_jjl9cImMz6PA8%nW&^@1S8_a(hvr?xkCqIDL7v6XWftT$1dGwf1Q
z*0#2ecVCs&GpgiT5Q?p>%lytaR(bKomwE2`bKJXsmmu(1UtcATTWoK;W=d<FV_u%S
z1%5Ody6)t<0kA0YIygFNr^FT8n@x^WbN29jAFUhp=lV!2b%P;HrW2~l(CKuU&nL{2
z2|*<3_u6!Ok+Tn$0_u#jC`xpxna>k;clSx=DbGK*M7P~xI&$ugPKa@irvi$~ea}Ua
zGoFmmRf+Nhm3@Xe6oO(t<-xbV&2k`E`N9`nT~H}37M-WOcx9Q3{nv=ep(rr2u2GgH
zQ55rHZ_dg1gxf{R-oY`-S~l{Opwr>YZ+#IDm`tXGtr*YuIXE~VJsJ^)?i|pSsT+NL
zTJaJ~{eZ}`EOj@S4QCv_^8>DDJ9w}DI`aA(7(YZvfi=RxT4l!0{X2a0@q5gYoOj>*
zfL6QB-FrK{^Ugag^?JPY{56h`?sGakrX5R{X&Be6Sro2U$UM|9<;Ew_cm){A_`O@`
z{!fVi{x_+jwu`t%voYCWGs>}{2NtR%!R93@@F<LPTI*QD=e>->exEysV}AMLAFzJ$
zGOgAU{jke7zVUbX{FlB0GT=|X{YOOcN6sHZS<+<AXgFj#O<7&O$j0UxRdI^7k}N4G
z=M&0ViT&{Vc;yWF`QOEET*uG|C<+FHLvG&wh@vbRk28Mq(;pFp0YWL#JSWd{;wYlk
zYLTQVd0tXhHH=Xg(;Hane;IB(n*BV^csl?4vz^nk3}XzhzxFy`{pwdaK0e}PFyQd;
zkki3{+1yEcJCBlbL2Wi#FAC&M;1#vasHt~z?)Tne;j>tjD$fqNzWLI_4}zNR=yuy$
z7;IU&%>j-U><Z`^-LA&>J+d?b0Z|kXhD&&!#P=KZU00kB!T>3Kr~wMIsC(2K3h7)8
zm8Y;}<rpyK(`RwuHWgKAn9WksBxiMXgKoErkdUScqtS>wFIeiYuzh8li<d6bYPBee
zoWbCfqoV_c!x4kQG57A@V>B888D*hs=HwZaC}l|)Mg&GMpQR-86l;Z(VG)|Dbf0r@
za!S>A>Iw);x7(uE?>fr5Xf_3IF2XQG3Q3ZuoE#50*gqifeJ)&Zo7$7(W7oG(Ragy`
zv6xDswP7}!G9C`;_I-NGJ;cHOvz?4Wp-<-=eD_EAYwN@>y@V)THq*%_T$-QI@eKwa
zlb4j^8Gbu<Ov<`8c;)nzxBC&_9R=L~@B@wx_gGuMgz$V`d-*k9c=a`&fAKZmd+&Yz
z?9aZ<g$pUIc1)ViDYA^6ojv;9H8$3_XtgxSWQ5Uz>2yXKCg{D7@zN2B0`dG;K(?_s
z_2V?pxOMv@j*j=~bhh~AFaMI+eCE&u=`&AKbfpQykXEZjr44yeAcbp`)io1UNby(#
z_rw6=P#cL~Sq^^sE}iEt6MUhAF%>Ij&eG(7d~VQ1L9o<E1(JL)ChT_H60R-1Zi|=Z
zQ{*!Ixm@R`JG<Pv@g8BThwwu#jwfvPSGe^2i+t|2H+lWFx43rwISvjEm`$e~9_{n<
zpZ+C+If4%RdxylTPd=G(RU8slJ<vYh(OvXU{($n$-^H$PV~wH8Gd{j`gJeGE*MI%%
zsGv*SinxFGCKtE1_~kF&AxSdl?Ov43=Lv_0M=Y<bVjH-~n#ZE;XC2mB!YE{UWqFZ~
z7EO^5C~9M<g~Uglt%#<pSJZkxQekTi)h2r=bqdn$h>;aSX?)KGecyA9VomWHAW!t7
zxNL<&<AmS<0<SiTks42Gj42rorXZlxk{BycGN&q?_kypynkZ6dH1$PpwLoiXiY<Ue
z)Y($tdHCwdIy#>|htTaUaqaqxY;0U2FLGwH8KcpVlamvUkB>PyIdKXG^LbtLD;+1k
z&JOY{b4`3@BbdpeJ`|f6sU*fa`5HqBR7hozO4I=$L<&KX8H@&v#a7@c@Rh~tjOk?V
z=F-<#lOSr?OV%b=0&uM{c2Rtl)_Kur-@~GuPlqyfGUagZ%#?_t#2QUuO0qJ0hIlMk
zk{;G1jK)VC9_=xmOi-TZ6ncq3wNYVA)LTP_eWF&EZm);$`CQ(<juf(HR*gA6IpNNo
z+uXW!i{s-XuuecTNfKuB8CjZAnF5^`WW^k$rqhXNwPVUOCnH1Hg6a5(em4L+$M-_K
zjt?folwkarLU>r^J6fhEDRjZX$vz+7{g_EIVRK`PwF_I!b<WZ038QpIp*6w^tTtF%
zlBIJ_PWC9Wl$GtPYzBSS%Ic}7&;peOPkKyukGcOBe@SO^gSfXuBq}!3Bcv%viUJY(
z1f4d<6cpnU)s87jw=i$DecDxyxY*%`M;<r#cR4vZ#)~_|ahr}8vf5qd^41n#|E+Iw
zY5OY5^Vxau0If@Y_Ol<6Cv#xVWSnrkKO~7Q+2L(^+Y)pKmCsS{eGi)^=vTf5egIl?
zbacqg+aGiF@>Ra@rQc$CWtH!L|9iArAy`KdAJ1k^+%qpZPLG*x&*}DinA#YwG0b}O
zI{t(JBB;AaNQVgzzW*~;x3`JAT`nk1y98T;NlLPENW8v;kdpLdh!@A0z-RTs3hBh>
z<>Ny(TOO-wgFm^axHB9v%~QsQ`y}^o)9Lk5t7|M>x=J>k;s*h%Yiq2o_6QY>hC_Dm
z?U9X2()k1Chj)3tSE1Lo5wVZb70KQA7!L+?zV%-qyDMb#3C9O}y!GlEynO8{?|*!Y
z`RItZ-gu3oF#O_YzaTFvR)Q7}9^B{jbi}pGSKR_H&M3H5RX4^x1AkK1FoY9zbFY&Z
zE^uU7R`dRwlv*qrxqwBDK`B+&56*-EP2DYQv2M2}wA1j%H4gbfT`2aH%MLT)FdOQZ
zw}M6)MAdCLN?4S1U8`w2#q$)MPS-JWM1|Ek0;f!D>{(e)G-{{jtkBnxAe=MXnU}!R
z*N|KxF^=1WVZ0y`q-n}%G~)R9n4_a3hQlG#smn+_RpEI~RoUW-wyqZ=B}wKa$&4gP
z7G_ymSCmyvoG==xgp(3c9#I_O2eq$5rLaPvrL3u;2BQnIR8tfMolb`+R4A!2CP&zY
zZD(k{VtKA;FR0-F<vF)b>4A0UgA@W!in_kzm|T<TnDKb@O!?e#cFOU|E{BKrnT(HJ
zpOa9SvY^OwJfOd_#Mb5}tE(%tT5ZD6L2-^8;1=Cpe~Bn=(`vO@US4G~nb7HktSm1P
z#UW{ua&WlM&VvUW9v(2Bj@jP6LQxb9PY0ZqM|ChQ2-*?7UJG5NB=c;s;fbOMfkg=q
z;Y%cvz*9W9x66C)yu<$PE&`9`<u!a2aJYZM@$rbV00`WCQ&Qy#r-NPQ(<6GZ&-TU&
zU18}z3!U78u@V#*4F|ux&DP5waq-JvWJ72!Dub*9rnHo^iJRXjB0oOGmKDnPk-pF7
z<xRq)W4@&_F2ybWK(F(Q;~{&)Jv?iuj`n!)lfT64EfH;9Vm_YG>Go)~+bnOb6L@cP
z@6H{@!x7`t2_h-Ed+VpHB}a&O8PoMpwgs~xqaXYSlu$%3eFco><ZzF65b_&ed>f$*
zli>leFZlH@eVIGA?zzIb3`moL4?g&SYgewLjjiQiT;Qm5MOBrb4Z;#YP)q(8D;RwE
zfWyD|F>ieF1vc`8K)Ns5ZwKh2K+g;^h|o!j&Qgd&jJ1T_4*qn;W~KQTFLt=pf`9!n
z+?s?WMZ&F<0o~J6;=O(LZr^2ReS`kWDpor46{8Jhl455Q^2vz9!4Yd#({20MDn;iR
zRpyh;rxb%D)c#kX{beMuv9dy_B=5iX3*P(i1FAAXmpQ}XgtDr*`SC5<aR*=d7_CXO
zgqyc+ap&G$X0th>#>-Dm2E6*_TTiwAS!-#vT5N7^E`m{(<%_~}mSqb}*<~ozLOryq
zG(v!1YyP0m=(0_B`dKZ-Tlkze81zC8rH1?;?uQT$74H{;Qfny1LZ_*rblO^UO?h5@
z5}lbu0Z+`(tbH8ndbS}J!hdJZjqbJMIA(Qqg&+u6US1|i=2T^g@4L*wF`eStKgE!y
z8RPMolapgkP8_yB9#2S;Ia!*L7ddg1(Q36R%L=Jn(9@=($a18rsAWWJ3UEa(#6l@J
z41<Lk)LFf!fj?`JLgkEz7PTM&(t;<YGcwbzaa9RH=)1<vWIiRGB^>P^aC$O$dN@_U
zZhFN2!9B9+5M@huuIQ91FR*1n6!<K+J=XgHEB%l-@({wfuH#zczN(zO|KQ|=-Tghv
zvS4+kfm33_C?wnpdH%VVNk5-3o6j8+ElK$BqmTIRpZq7vbWXR^VP$2R&CLxuoeqsP
zb7Ng#od6iVE{VMmr2-z@zsLJOdzYh~Bg(mEWo3nnYnPdv%$Uy$DidHen5xA01VxrJ
zJUU=>^Z@cHTNf^}(bI%FL3z*CkK_xS#Ve3bGIqcB6IPcxJnyG;ghb#uf`kY$DG*C!
z7*NgUh;|Dj1u6)LmX@dv5BZH%kL}JX|K^6}J9{3ZJY{!2XEdMFIXpr=*kR}68?0?!
zz>i}Zm)9aI(DNB~HsNS+M3Kz69ti@Iql*-sWf&z$k9TQ*{3GP{%andhI}W+9w#L!H
zE_;Uu9PZsE4i%?^V~V_BfB%qsckVk?^Lk@-bbQ3o`^W4Z9MoBzgIIU>ABF*+7(k@3
zc*0UxXCh_uit*1rpt}O9RZ&(J&-d`65Mw3PEW=-k!S^s(?kJAd-~}<QUXNlj!=xFn
zcm@CH0{pSI+!HNwEQzp;iX2W4$+9W=^b}7UmOY<|u^fyi%#%4;S#UWH_*`FdWf=qp
zVGLDKP^JYD1~ojOEK>x6>F|{A{qzS2ufu4X@?ie~)6s-VORUih21new`5~sV^tye1
z`m<j$7>>Ag`!<v5Yyl~0Yq<XMt50>p8DqF|<to4Tdw-uejtFY|%4PviQ%7+;JUql0
z!*Dp@`1qK~WWsndMgX(f+~K-eP2n|_W5H+()_IhKL0q#`0)!NJN)d!!-NRB#(kpca
zt4QZX;rkBrKd)!isMI>;O@Ze*?WHJkeE?1q=aX^xN1OMazG(jVu(ioeRk_8iDofHd
zV?LiVpF5S{g;ozu)1wDgfmL|QE$k~R>s-Ekg|aNjGY5t3@9)>L9tnA#kr!v=fue@X
zhNnYL2FH|HMp<X1W##&n!Z4uM>l4Qjolb{1jxok%BxiCdj)E!_oMo5pCc$-ABb;wL
zHT7Fbd?7JK$#gW~X#bFdojsCS@=S9$xP6<W^^17QBMfBCNYfaTB80}1kS9Y%#}bbc
zV=Y1G<W;P3JI~2v%HZULlamvws$zX?Fdc#b0RR9=L_t(zol6(D==C-htTU}Eq*O#<
z2c<%+HTeFF_0>(TUU`nCu18#Vj=O-N9twzOav2Kchr~*;x3|YHfBJK7e|(1|%~|TM
za(VMAvFb1zW?16{EwL79B*lEr@c59?(Jp?K(vAhIp&>iEi!UnjG<~Ky^ki*&U?o;q
zrgsm~Kl?fDmzJrtBNNCtLMacOq*z}f!;os8AhN9fvVFoZq-=H2m8O^Hd}{@4Zus%Q
zW2`zD2?JA6&L_m_6!JOB59k?7;CbxMQzp|1vn(T4f;T$>Z>&K(a1?f1Rpdp5tTfSV
z2(tn7FCw(%mp}ah^CUwPaer@zyhurB8OeNxP?CFhKgRQXmU=4;PKW&QfBF;d?d-6-
zvxl~3v4(u6PV`1v!!@*6gz#93B;9I8H7c={LHi!`x`@ETiy~y;xgBa{@P#uov9iQh
z3L6I`X+aQ1{QA|9)$oKL?aX+0YB(w(j$7cluHaEt5*dTYQ+Csw<9SLX1>d~b;@5hD
z)v#o_>(g#`@I4P@q0Ca^&_@mKk)GZ|25Urt&%5v6q_hDWn^zFR<Hn5}BpD=G!Oz}(
zpI?0NVckvKz*T1~jfWf6o_@FqS3o13;i#@yw@|qL1prGN$F$oWX0wJ3V>vw?P?jZW
zlGa&5MOik@U#=m_WisZRQmccP8W|0&rJ``jh&?Zucvk#|v&K81me>dvjG`v6G~MR_
zk4lT5We3gN!C#5{ycib8>eA@4Bu!H$<8j?fFj+A5{2(BXVpf)yiQ>2}Ksy?&z(?Se
zO1&r`3|ho-n|^<ptgew5;|!|GqF|mR42MH@cXzQyFqzIAqs<ye|MfhAz-M`RnYGnb
zTCLW?nXu6fYv#6bK3sg7a~Ge$qOgcMy9%Vl3j(Bqvd9?@PdV7x<6vi>;b8m-bMCVZ
zfnet+@AGG~ivHRfE2}FkEiV&95h@Jur9uhNrsQ;ZLXsx<CnKV`<r;65ssqS`)4_nr
zbml&<=e%=d9SAF<Df7IxjIL_faL#x#A<t5_wl-+DDzYR2A!>^7nM8jh!y~06@O`iq
zw{G3yM?d)qw{PDAE7;t+%+<@!vAnc~*22}7ti_td5ge+F(~~`hM>~}B5&g(=Wo3!D
zDjD9r4*|?3*)wIwUhP9t8$%F=KI>kAm`^FP!clLl62IL-wL?@pLMJoIsv@u+LQ1qL
z@qI}chm=K0B_x+tm-r_=k1Kah_}OX2t<;h!1^qrsNah-BRiJFepelJVODIasi#?ye
zx8ZX&tXOUdx~(>06yXVhE(%Op5#)2q{SVM#hj!?5dOGH3?|i_<=5?eWF&ZCoc=!{B
z^AaQA55M!LPU+lODpNUCj`O`xk9K%BbMxpAETDuz){RmESAAby@{mPRC-Pv+ifn37
zt%x9Q*9)fW!1eq_KA}KKA7KpLewW@>j5U@{tIhS*A#YDLKi!=(m}aC!#=r>FrOTvM
zg;$hZ&W^a&33+`z;>&$O$E%3q7E8;kblM#nrBbC_`fG;+s>3_TOG}*2GVTsD(rnCw
z;RJ-CEDE%C{Ek`jY`c-qI<&Twg_BaSrXq@J-*zGB_m)vgap~gr0^48gz-lP#?%lg&
zS;lNOTl}sl^F`JGO#nIT8c)-7(Wo^T3|vN$B+gosS_`by<VBsG)Qf=g<Bk^aknj7B
zsUw^gS7R8}yl+D_UND^&{U^<17Wa!L{AU=}TB<Uq$TNzfU_P5Mn@yO_rle_7A8(s(
zw?nVjr`KH~2;!QB<&01SfCa5<8FDL7(x=;rX|+0Nt?P5hWoOgblsIaWX9cIHBigM~
zbl!20ZEfirMG@U@mtL>8=+|kCvzl6#tl3{~F3uWdoo9-~2)F1Uo!m?4IZu(YESOAZ
z9PS;kzq7~S_=IGdJX-(}f@GR;>*KrVJNIdI+iYxIU}bHMPOn41*QeL*(26^RK>!xA
zWKNdM(Z;yi71s5}=n{!xHW_mNZbhCZEU!#m<7QpwbG=-~HJBQwZ0-4)xaE;0KE!bi
zx4PHWwc!ys+Ik$fkV<m<-aUT#%Xiq_KP2jO*}8C<wUrIxsEw3<EjI*`3PDPq%{e_e
zVl+IWN~fSpA{nzBb+OYi<us>?S!Cr;H;3R!<DCBLB67Fm(+dqcOX~WlrA!Q!F7f*<
z6wcyXR}}^aKj|WKbwO+Km1OhsCiosJS;kTv^M!fNPY*M09VaA3$s{W%RyV2IF|w%W
z4NrO9lWeR+e04owJFEyok8Y>Oa(~4IUTQmEG@`7i_HJW0U!o{A_fHe<4^s~2H(j%9
zRZ^6CA=Fru&t_Zk<P0S6(Mm$C<QXW#%1W2k<t>VIj`F~!IYLNeH*mF*N)vV3*fc{|
zZr7`e3cnMhl#fYLbd@8+h`{q{M=`Gi1y{O0^AfzM#gX=DeeLh@%e@2C?j52JQ&t0y
z3muQxhcHr9mB91Ex-rT5K}2ClGMOT(5<S=>>0aZVcRyrSt6L`Z0{gc;Y+Z2vRRV6J
z`R`mO`gy64C*2->PV@LP9oORcSY#(p5+VHSfT8{>qbdug(+Tsr6VIE?X0?-Gj^{aZ
zr=|W9y`?^#b_XS09U=&VMIVg&JWcJ0#m`)@u5E3-Iz2Z6tVAiFIBL`DF0s0@N*o$O
zUlPYHSNCYFj>Vb5PV+NkY9`&esA+M>;Dij%RFjnfKa_;NPvkk{so8kK(Sd`lP7Y3(
zPG)3TPW~*7gg_Da#3;(7V2}+NPg1=5y97~4uis~FZJqViH8wXk=yp0pb&wSHOa}?;
zJSU<cKud+GDo&1$7>!46zQcep@Cke$ukK&579t&xW|}OOq)`i{{D;mht)0J*5DMQ9
z8IPwtc<_Lood=X<#ii@lSzcKuj=Ol$!&-^PAQ#OgCCPl3!O0Pm@dRDulx0o?EO&Yo
zvy%LHMr<Vl3HBMJOoYIO0orPiu8!4fdn{krzz-~Ck>N>!Npn;uL_{8@?%4=C9c<cU
zX0E=d`~W>qsb*tD7$Bub9JW}snr~j~@WqW78Ma6&&@X?9(fT%bZ{48t;m>FdJuU!X
zL7%?#LY5;4L#Gobm5-<!PPM6s!Vqb3wY&M0baKSqdq>=Tu!{k*qJU?ZhEE!gXCPYB
zxwuWg)gyoZL!=d~t@gNn{RR9$a`<46rS%T2<t`>Kk-k7hAt3NtEmV6>emX`)AyylN
z=ObG!NK>kOMxLjHojySrQqBr|3@uU8Z*^Ew9>@3JB@!K~EaCaJn9ECD{3yhfIa!jC
z7m~mlVr#KwLEt;P&taN6n#6R-bTZ)J__#R(_^Uus&oZpFL{ZG@$|lOIJ9KMpCWNDP
zI?lg#M^`V}+SESfLUQ@?Hd;H@(ZZs)7Ex?WVVoFV#c(*p7{honW^Z>FV_-C#0N{JR
zn-?Jv!Y2r9y<34G4Dmh1#>NJ5953j@=Nl2FlnY${d?EYM`zJvIi(vE&fk+5Jp5+Wr
zPZ^KLi@HZymd<a!)u!EU(P~8mo<a#HWf6w46LeI9sw~eMAcZStyN1U`)}(HpEJ{ob
zXU*qx@~l89k6yP=xv@o6gh-)lxsfwCusO~~cMEV5GWAq6(2O-;7TxRyYb98Nz~C!I
zpcGnH3<jq>c<_Mx_aAU@aKLCdBFS<}t)KY}Zg;zEb(WaU=gf<otkPt&gxM@%csk)=
z|Cny8OS{#k-|4Wn(q(0Jm2R&?6vlX-BWU>2L-`VXN3al5xScM50i!Kt!wM8$y%{nn
zY4LqU7{*B9EtrkED#@~vvZyfDa5@+<8jYCFX0*F2bh=&qFs8AGg&Jkh71A22JSUkA
zIXS(_Y(4`kN%Im@RBWs)5eFgZG@(o?A{AhyJ?9nk*@sk$)i>V4zjp^UObI-n>({Qb
zy0*>1?oEnp#>S;Jh#X-<hCb5s!9uj$XO<iwv{SaV#^Sfy7;TYdMmibcwK@b*NR}r^
ztqI6zdjaeHE|bZ9rlvv^IX$I$<LVOab{iorRhBXymxL`rWhy)^F_m`Z0OPEvqcWo$
z4>&wNX8w#dv|lwI&p?bJzHpJNU;8@K!(-&$kc;arI$_Recb{@PMhB1;IeD6*_i|dR
ztHdj-NKYaBh{^nzo*y{&i4b^kL{Sz*;W8qbQcWiK-7ej3pVQ$nc|IqfI;g9>cL&qE
z0`n=C)>m0u?xU$VdT`3o?uh=zs-v=6D60}`YUzAFXLdXw?kuBAL;fiR-@lFF`vKi<
zk0%8%W9lA(a}pU~7k&vsNLE(X5JH^!o-cYtSadHp#rQCYsI=Slj7Jk@vt-eUEtPV&
zddF-ri{Q{|xgGV=(lV`99YES`gtZI&v+@3T+HB-esg`FMS)MYPjF?UxWR=!MP|s6t
zQ3{+SdKd<H{+X5()bbjhw}2ho@w)=Dt|5@BEXa%8-RHTpxlI#_nlKRf9^FnGjHM`3
zr>|C18JlLs^ESE0n1w_?oUz`{KR%P7S9n5VO~rIJV>X-d;K2j#-n+;C{sFUD;+hAA
zV|k&gr-uTqC7LH}F1hO{7nWJ2ndBMsvLdfEl$vs$a5_nt`y&o^G@kF8DBG<T{eFKz
z5@@wrwA*b5+erbQMe1tdAEP{p=M@yW!dO$+;!-A)8D%CYOHH1awFR=1j!CnuX0920
zKW1&E;{+|0!dC$Zw}H_ncQS#xLZ@?PCnt=@CzR6F`X<wis<3ndNw42QTSHYAcv4Z6
z1q42R=s!a%jbQ6*UuSi7jl=)+Jvy<+g*8Ps*=KgNM-cjCrz0$ubTY<|B08HF@ZuQh
z`D7Y&lGASc7%VdIP*H$Tf_C7OPAAC7((Nsi%qNtS0htuYszT51QHuLyvd2;&+1y&k
z3p{3%5qk$G7!wk8JuJd023U)bg1jg&r>BISE`brGWy!N|K|Xyvu?B(^ti1U-F8t=#
zIQ|cR$i<5>GDygiG3`!_5QbtrbF$~!k{lhnh}-WYrN<xxnQ8j5hcyN-2=Jl^v?l5<
zQO;(R(>Z>pL)`9>W(n4mRN4|mG0WC6*|@-6-_kP`q4G%Q3H^Rd)D~Fbc9)Km?-pZS
zR3zgGz3~_`8@OG-*(ds2A1>?gQJ#EA)NdhO!IS8&XK+iFmQWYOT}P>`X}SWbWF1UA
zqyVojN(MnhtJR^~P3iZS0PsadRplt<6jPO_TvkF;Ja^DcqyFot@?z1;;MEzFG2*e~
zZk+wjaDMl<(8z7s1vdUHUqhZ}q{*DBDzL_BFB{_o@{(kR?<?2$Q8&n1<rq2ER_Mx6
zk6o8?i8ht<yRXF;T|FqTHP0%FvLH({^4w9Zu@#;t2m*!G9_2Zja}zi`&#Ofv&+1rB
zFnRR*ZL&JgbBlAJT-KDN3FGO6!-HcUJb1v~!2zT31Y;~w*dmTwbUHm|Y4Y@QC<N_v
zLa}ojS}i(DeO8wG*l2}ZNM>2dEXm1|f^?pt(~Kf;{v=gtm`~>%>>oLzLVes$aXECK
zA&y+I@l|aXTq`fjS_!u(DJv(;XDZJbLz#-k9TMCoL*hFvJ0TEALPD?-O6PVTT8T0S
zo6VSy22|4_Hks2lB^iQbUXkZcHmASbL;8xr=_ztHr!OG#1npix@C?qP)>{0v6}JEG
z?{VkGErM!Bf6XIJ2ZX*Ph(oF*bHNHlHJLHD`?S{A@uQgB`V6OYTI~=gx-9Ubh{_m5
z;L{R{Vm1XUh+AF4XpAsUE=C0*E$y-1?Qjw<aZ(jr3?)^mky^95)u9w_y{L*3-*dfJ
zMr+b}N;#gw-~eGO=Zo|N<?g@gc(SFfwTLL<;y1p*;BXJ!i@9@<(hb{qflm+xgli#K
zOIZ|T!&8dEiCa2?7;T}f%wplR!Xl#(Q<hLvcu|b-Bvqah#~s@JRRSRh!jQr7l-cPi
z%a^aQA4JHVcX_?z;rS7*Zp6IGX!!xAtdOEHs1(?$Vm_JDpHD~*9*`FQ->G{vLkMTa
zqLjDT@t-ZO=Al5+JRi<}R+c%|=!J4`U7vP!r`jV!Ngu1g^L&Jm3;aI_Lc%a|?@_hT
zmAk-uz1~84+r8G=j<|mJ1w-r1_VujEanZu?2<z%>H2Ajrx2;Wa9#vTWZ1X1^>7go+
z7(6AYB$RnhGMh3UoiZGp(C_!_+cGPwn*>1x>ACLzvZ{L;T%p$qEV^b)$Aq&eRr^L$
zA(gGOHn4b9+9R1c=&)g1HJ&Hu1I2mR?|i>aGgmI?IQ!hd5Al?s$a4mxA-lW#?Ck7v
zG8izM%`t!<#`Kn!SY2Ibb#0x4!^5YZLx5P6tOhYjkx-oG<mm(*#fYdyr?bT7g-znf
z)fk53DYuRf8BHdX#aRY#tR>eaWvNM%n#m-b&-TJ=MX|_|ozSBD{ypDwnqtxsYY^bo
z{Yv1@FXc&;bh=-r_F8w=xV3b@t|*cz)$A0RPKk7ZG$oTPXPA|gPPU1qcAIvm$8bDB
z&n9$?1z*tV#(0sXnmvOSmcyRtz4Q`SzWG}ew?1HR=Fyp#wE9a(CFyn?iO86WVm2ci
zpOT%NU{}^K){vEsktl2XXyGYj9Jw`2DZ*}tBF#|7(p%Y}OlHJkh_Zsy+b2Y=W&F!8
z^V54Dp{gn51>Ln(w5t2Es%oM9V69`(&Zl$AWKMOw!*pb+o>5ifSB+0FlmcM|@%jeO
zfAbr(PmVCByR_EVu<0p&?4yFfWfwxw?yrzePRWM@=9Nb=n_*%f<@*TF!&+CIM0nuU
zEEOeD)`|1=yUP?=MyfT%e8Ae~bM!YhdG#OsXWaUa|ABYD`|s#K{}o1Io1mD{iX?fF
z5$F^VMcDC}JU0kkp`^#?WXf<bq#Z>pFZF46yRIv|EJ(AI!EnrUI{zD)&}InXE{cZ@
zp&Hua<0rC|;<3jDdDcbFS>xVWkZ~?yx^ha!4Mo^f9)TY?qCps9l*>}i({-D1zG!}S
z>`$&~{Pi>le)eH&iFsXVFcx2Vh`Nw%jk8^@ijwhYOq!&O#uHSXU7Z)ObY0+QeJV=1
zzx90w=`|puG78Ug%FuaHf~b(v;CmiKaqYcwMo(`1LEOAJN4aKq?U+f8*^);Pgmv(7
zVwY)}aqrG;4iEM@IzC}AI3+Jjf;eWm+h=KMnO<*+R=eXM%7cSvn#0>h(9)oU#1s~a
z6ec-&dBUK&4Pi*s?XrGhi>38-e*KHD6Gk!Pk*i6hN$McBWKL0OjBu{$)`DMEfWSzP
zN=uAXAO)(nW)7r6SpnKm$r7t-%LXCvl-u-JE6{}jQ{f4Lrz9m){8AGeji043n?jx-
zY(ZINO!JJBX~t1$I8<)YTx)lE?&396k}@4+tO}o1-_lc>C0h^_#aZ$FX@>#ng=~G{
z3$WV99PSb<`vgIOs^+LTbZ3NC1PCO8bUI{yc*OW*j?e~yM1=uH)Y-jIpad8fh`pAg
zDm0y_MP^<9*>H4BFE*H^WnTI6m+`;x4L<t6{hwL#``GJmP-PQV0zpZIED}OxK}o7g
zGn=_Al1j3_cYw5()#YWn-43lz2dwk_n9ODjPKQ*_!19An%Rq44=hfwP))aJBBS;U?
zIe5P37BhRsa`A!?Q{|+y3D#5ufrkuytkFz|lH*C2(<(xkl728_Bc2iZfHp{_$fV`u
z_z*u3y!e$bQLJC#!o`co-~1Ne{@x#R=S1_%vB!mrFY~2WuCaS-mux!b^J^X}#Q`TH
z$=XUp;ej`w@Sp$P-{ra2Ugy=<Ut{~qRod+~RZ%dWj<|XACO`b)kND1azsLLUfACQM
z$6wE|_?}O@-L0V}yYL!mI?>NB+GoZ~a^c?Rx~Ly!UNv3W4U{sUyJo{A$?Dx3#8I1m
z-&tfjzlj)30pd(5zIpHFb(-QlG)7k9p#}K7D)9NS2xxWJ{*zRK&x-^;?NG9o2^Yq7
z^P(0%KR<L;kr_i#l?z4VX2<U!BM)Vjg8-E)bbG$H2-?nN&tOgMG^icjxAFy*FRAKI
zd?CyNs&b#RVPQR7=Q76mYglUuf{-|F)s2Rh(P+eIG-fauu)n`YI-gT%P1Nqt4MW=P
zF5O<APN$3KhsavL4Eb!q`>4>|7)^=lUXxHNLgg<!QD(CV`=<kL?cOEowGdI4cB{j+
zD_6O&d4Z+v%f#&#MhNn%Vw$B4$0MebDSDbxlqE@J$PH-g0*EmNT^3MhcvwUA{~_+r
zf-TGP^FHji*4opV@0j=IoLQMwnN`)@)!o(IXrM7Q0Fp4t27)Fj5g<*GChaHti6*VE
zL@UAyg~H*WY+8O0Ee90}2@;8c1VCaU(Ex!Ss;BCyy2i}P%JI(k+%xWR4SraApL?@1
z8(o=1gjU3@%Dj2bKKrb_*7{H1|NCsyo8!5_L{$+8L8KIMAc=%TXpPfzgg(Nib8MD@
zv4qA_Y0W6l*-2A&OUu+jDI}G%bi#lao_&VC4cWPKk7boG5Qdn7z<Jr-R;x=lcv=U@
zW9r)N4uiFI25m*cjAHj)qBsHpwid*5bsa2>W764_BFhQmHYy4cfuzbRj>aAK@)a@-
zZI!W>Oc^8@DgbGGYI!s{WHuXd_R4j5`DMC;K36~SW8C?}-{tNbZ}Qp=!`yB1$xmLT
zlL+p-afeeq$tTqVCKJQaJjBKcS$4$pXHW6$|MCC8OCS9R&s@FA+S)o&K$fQL>~8bz
z?|g?ZeBsM{>C0bXcy#ncGxMk(2rMWYFqRMqtRsv9Z0;$FbqccWJu+QZIBl@Hrv8o~
zl+cO-bY<DuTjK5fc@E=E#zvu}W#DpF58vbY?z^mYG7P?ReKsG{>n^c*?lQ@XpFk?X
z`o&9}x_F*)zhG%~nd=|DMr*A{G}b(rN?zTX^Ws>uTT1R!hVO2t90q+_XCju@`U_f%
zB#E(c#HA}&`P8R>jKA_%{u(#ld7Iz**Z+!N`k((f-+TQ{uPgGWG@NUsaXoCak;rwh
z(0O<`SKdH9PrTo`TJE$FRPx?<Ly1!4x$lmy%KG4|5W@E&gh5SUs3LH(hOTNFaz|ic
zINqOXd7OsH@qK=9c<=kW)1PSm{1yD1NeLDVbuHyYDs3SUB_%>NzY{10N_mbiDe4Bu
z6SW;F8jdajPIx*IEAfI+5*r4XD8%THGAl8{(%3Cgr%oGCk59Ew3m8>(e&3Te20;i;
zkmm){>5Sdo9k#Z%7!HrH&Jsou-EPnGX|+2<ae`7ILa3VU)}T~~3Zf^TLjhD44xJ&y
zQ*wj|LRk`r0Wwh7B!XCB1r$P}@{IlIl<j+WkwUV%w$7O|XE=T44C|*))9<wCbXu5=
z4Z_lRQbrY_9h8;DIG-A<ikzy<v6ZLZw?tm+-B^UP2xCAOP-fU{j?L$|GKVTh787s|
zYXzxs%u2&yUa?<R92&t)C0G?At)*|~Twd+b6Nb_5F0DWi*WiQ_-@DQ8EwOg?Eb-3n
z(?ip(Ll{SBePdyu6xtbFR$!F^6;S0lR%=`>tsDlDIPef~B?Xg_;O5a4Zp+Ke!ak{~
z=tt0-?{RtdHdj^-2&6_zNuH-zqZtfV+5FfiiI!K8f#l*#A7%dbYqV9ssY{pHeC7gq
znqii5-Wtvr%rw)v<2@_cJxZCmE?EaIulMK;x+F<Lr_)Aj&C1#-KK!vy@MnJZbNomD
z(KcUw<;(oiFZ~j~^KX8i$#nXGrseSw2>$TXwj@%HxYt8Sg~~%r7+|I|bY9?sfO4LB
zt|IG@DB>u_VaUspqn%}5nO$dc`T{ESg`UhP)}FmeZm#h9TU|cnzRvQJPf_Sf`^KuM
zg)o}g(LT$Y&#-j*G^6PQ(pk>C@7^bj&T{VZCQe1%eC5l0<?aDrEpxWEOUm6jAo;8R
z!4-b|$9{&5(`SgHkZ-*D3jfnT`=?yKbeYfm@MrnxM?b}jANeTHKmR<R{p?TjkN&HF
z%&-0Wuh*OZ-+vfuD9em63<+gO6a-LvvSDpNnkFjWfaQzLRUJxt*;-4UXN#h5RaG@K
zwR$K5sdRB1Nmz{(mLzWKBrT-ydIIC|*z-8m%4|(~Eso=b=N`uo>qeJl>6`6N2wOGd
z`=z-aLK+1DG6<d+K_u`RD5k8uSBFC(JkqaD8;J%Utt3JUDpS@yH@>jwoT=rKYvb%Z
z-Mi2NW6A{wlOK(+SwmdukkWG?1wmM=^4YqEVm)Hc+J$lVs4lNrA*Ck<3{*f-mK^LK
zG8&DT&1RnON``E1Y+~_ZT~S?>mNLLPC`yBrt}fDhnEKQD&IXv**%D<zSx0CMp|%Js
zaZ+FexIhv*OV?V2)wH7kAr%3Z`DDzNsVJ(FsGV?lbj0oVZlm(jM@)kONvpf~E^!>A
zr0|A9`$z;`A&muRFlA0YpOR0HF!_wY<b<vw5(WhnWx%|um=-1D$}l!C0cJoZ8U|1w
z{hjT!IM?e?jgAP)3}q`yt-%_ka;&a(ICtp+?LixJf9t8{&^d>32J1?ac7lrgD65fi
zgv;g>2SY@sO_i1C(mQj?NmMPBoNJg(HQ(Mn&CS*4(e0Q(K<*%3Sz`U{6>i-+gBbn+
z&#dksm8z9S9a2TaNsIaTfbOX?tem?<5<3#<xc%M(me*QrY^-qo!_V-YufEQA-+aLR
z$(&cylI*oD3cbaZXP)8b|LWi3Q=j>9l2$@dWc-u=_8)V2xW^|y@o7HuBcJ8m`Dggt
zU;4{@`qQ7{pZ&9c&OiEp{ny;Nd-n&<&Esnz)?u+YT@n;2fhiF#^<ilgplD&!!WYpn
z*wWKjJ0S?;HbG*^O3Cf&0!L>~K}&OLE#dsdF5Tq~RM2I#JL9c8SBZ9R@p5k;=PbJN
z2dGma^X+>`C0ITCEUo2rWcppMtT|@+lpAl}A}?X<)@_d7`4aDJ-RHj6o^r+(tZ!^^
zYV#c5{noeG`~2@PJUZl!H@?T$zV>C}MDh90{~kZ`nV;Y<{Q1Aahd=aDUViyy{`3Fr
zAF{N(#DD)k{L-Sc`%h;)>Vbd1K8p7fA2y-0N2k=!1{TeUAEZg~apR;H_P=zq;GZ-E
z)F6b1=gHUTk7+Qhu~-eLru+NhW=y*{=<VX)o4y)vL}4+r&UZ+cWmTsgjjt;OvaVhE
zA7^ZBoJ`FQ;Ejt^0aaNspJ$Y1;k9LICA?0jdz>y5f+%j)GO+?`V=nMD8xN-u=u+b<
z&!X@&!vUo{W*txhG*CIfSg$arB*KEz{=0NrZPwS;eI#Z*fFTML`FzUm?gNIC3F&M?
zR9UJlWqhy;wF+by1a#Uhdfg6j?CWs;Ix$*%XP_!`=CcX5%88^QKvSBMNvX-TVyI(M
zqsVMP1{A_lNrwicHAtNi$&%$x!nwf;`FM`aE7}5Fm7}aBR<OR@;_|sQmV1gSA1@rI
zC!RxRan>NKf}l)MF30AnM@M$si1{2_lqe-AtK2)^_#UjJ-K8ijd$T3(bgxjZ1jNd6
z?YS<!rFFW?XUHdp_wv(x@AhSuq63zfDo?Ol8A>OxRmJ$N*H}G!g`~HPb&id2LYXYH
zd-nmm_jlRbn{asJ+uXnTZEnmn=GGCV8X*OnXD=|HW&GB^{+H|@?69}D&%gVl-)A`7
z<BMPX1Agz{{uV#`=YF1_`s~lLerl7y{Wt$Ntgfu^fBP^0@4Wr?JAdNbe4hx!52Y*O
zY=|o}oN|~vBZwjobBqHBL!^n(^BLF*i9{Pq=rYi!*$&s}4En6bAzNEhw)c)$+bDS9
z#b+pr9MS7>_s#~-jPGHs##YvASEv9xpW~_=TUE&32IbxZHmASGj2>Ec*j58(yLWhF
z`(5r8o~k2|zR7a?=!pOKKl%S)s-lLGdX>S-I1rAzcZa<F)=gf0^=tf%U;MlLxj*|C
zxNzwi{^P&@e<IHd{>NYXpF9)kQxAbP6?u`87rueV)D$HSM^$=#iY5YSI#o~R*R3-+
z>rSw{dLABUy(jI<4jLgXLFn0ct*vz^T!^v0nW@a@80+Dyo+L4#)9Lt!9|Vgi1Rm~N
zIHoIkxEAuzw`|_$t1S;1!KP0l3`2rOL;Fc)PzXjxBktV3>#6Mn4-D{1eW8y!n)`Jg
z>DMqQFAicU>tg${p;#Cpr|L+fN!9shV64xXm!%hqDvJ_lBAn%fvCxY@SK8M*j4>GP
zS(S^j^zSVTob{1J6eeCE%DM&mE07Tl86eK1>jD)lc%roSP+09MpB8zFA)t<=6w)se
zEP|@8gi9g7StJ4tXsgLw%GUiYmY4c0EiDnnkuPo+8M>-SqKI&HnWVBP2d>iI^R7Jn
zIV>`e7E=aL8|=bBn&o79ffWK>r(AIsXDh6((3K@Of_<SVU5F82r2ysWfI{bJNoWhl
za$s4D6;)a>J(|;2V4NYemUaMVS0XN*?QyCHGTX;iH9Y&N)2WuYm?GwTI6zk^(pgk1
zrpz-$5Fs5P6}rff0*XqY%Z!$2v1`_tuMRlZ?~%?*Ui<D4mF)B3m)f+#02#nIKFz&h
zKr5M|ON*_vXZZlk!2xDILB(xyo6x%bWzI!!p+%c%(5FZ(TMxd)>yrto^UplTVGaD@
zm%qd--*|<h%<5XIchrrwJlGm>>)ma>`<-v`+H2qBum9rT;h9TU`MJ;i94d_X2mkT^
zg}e9eec!oxJe|7`2oym!CK$bovn4j^fYa!_z-mj>YU7lmNHc^CsdP;#Tp1E65kwrN
zmf@`}f{i6sW5M>+aN~^wLf0h;Lw5F#XqAT9EJu=J%bJ%cj1c1yBArl`IWp;y&r8mT
zUAkt6-85n=OL%){$WA@UIIl(qP!uKGJBN!6bi5F}SMNon#!RN)Ae)zr$7BA&pZ`l-
zIDdh^|M&hL_wL^15B}(npLn6296se#EDoDvVIx;}cH0Gutg)unWA{yN{$5WOxNGg{
zx>w^k?PBXxOMS>&|6@W$r~`10JhKbq1jqddz(SR;sg1y+5s8NlfB$q<^PcxdQH=pn
zbtB`G4j}~h?>|6irB97SG3|DXcDqBX<r#vbI3f%@>7S4it85Kg5R00VRC1A?YidzK
zP@|y)Li&ii>5%q45EW%v)T)OXZ0zMhYvsL0Kg1Zjcu(UXcrAE=QX%bDw1{XtGRxFP
zHC`wMdXfN%bkggfi;#vZs#yoZLGV5T%MUtYr4SBG<(nrJ0)@9(LG3(f<6+GbfpMVG
zR9btaPM-UHr4+4p8xe&R#?%`0k|>Nw;t&-D1Wpk8U<v^1S22|}Xj@U3j7gr6TPP4z
z)}ft;m|E+TlvY4Sij@wOclZR3P)OPemZV@=1q_rR$_&$dP9zjcmjuGH(ve(Pi@C7c
zVx<eQ8H35a(`KK%7ZX4V1;!9g?ja`kA!_6BB1>jgB7%sp)uzZYobYJ0GN0qD!D&q=
z=`k-ucK64)tmN$GfL5j1nN@uItM_>J+0z{CPT8BMOpPU-%_;K|v>}XJpfxH>(enwe
zcZ$lz1hXSfw<pYU$#$A>YoxhxG$a$$^`mAVJBCN&<9)UGtv?H^$~Pp>ruUdmf0gMp
z<rjb9Z}HNLFY&p*@E5pq`!4_afAL>1OCR6K{&+#Ds={PBZnTdmCWuhtiW<+~fWE`X
z`<>Q%##yDA&!(Uq?bZM(4I8}ydeB0xhrIOsDnGvY0;vlri;{1B^ELL4?s3J;F=fGc
zK0+W_THQdY5Hp_<>w>Dta7r;rEtgk26fALLu9=F0c~(&UfaVzYeg0;x<LEHs)mL95
z&;NU-lPSOOAN(ROTz{Vb{onb!eDmAiVKg4U|8gH_2q_qiN4#_6&4&-(I(5<TUN!xN
zLI_j<0Vf&`WmMm>Ss+JG!hyl72|B=9?G31LUgrT1K?MlIz-t-Q4REfpw*FIUsOj;6
z<qg2e^x83NYV+d_&T$0gjf{t@Ha!xaTHQm6ABW)jAfrK;Nm(Lg%6L3^sJ7yfL^bLv
zP_(-pf-v;G2vYj~f*`03NzGqXgBk=Nq}M+<rpwnD>pT&lF^k4SY=~==*{of_d5?h(
zP2^b1w9+U@n%cNFhRYw=#&}wdSS1MMffd$SoOYB|;h#0$xV6H__eja7Ga#o=t<i49
zj3+bFd5W%`7vUPXe*L2xMMOL~sqj%{UKCW?)F`Y9BL!(uP!^hrkaU!ys}yZ1h=m}L
zl0el|CsI%WC5|zM@jPQzX{H(qC#Zx%BXG#YW{}R3;DI&93IgSb0!1fGh>WA94Y3Uf
zU5K!TfwU-R=|+OnOKr}t#%%OL2B9T1DNYv%t^IazPa`Phuz8M|PLbmSgs5<4j@2NP
z#8wU&#f#189BAj5<QY28QAE&@bX~yNjaAB^VB^_7KXm3CVtJW#t~of^<EyWJk18(-
zWWjVgA)RNe^!xNyPf^XMC|zLl4AU``MxjeMbM_2xZaKC>M_TTZl|Ps<!ae3s09kIh
z{oW40{oB9GbUfp4|E<5nr$70_{Ec7yMgI5;f6Q<F?(aSJdXJ*8lw`jBfTMR_XW3?+
za#|>4=shD51(>|VYV8?2TQP}n%(5v(uF2;G!@1(p#W5G6DOq%hY(As4xkeRAzVU_E
zxclA^Z7t_kR#_gsPM(h#9ZlF+TV-W^gQ6;M=@c12k*4UX^qs&q;eHjdryM&w`&7lh
zA1jyxMn@Un{mvVdWscDmf9tRRU4HVjKgo~$@Td9h?=xvXXb1#ZHfMKtdvVYmi#_?m
zsZ@fHaB;Xa|JEGbi{`^cilq*uPe5)P+vE%G&WF$*yW@X^?+Xz$jcxKF6jpNpK3aQu
zv{}%<$L|nlU7i1Lq8eyu>t)^h64a4_RKAhyspe4#@BQI9!nF1goTzWCVK;6H>PO@0
zVjg7^eJGEBY3@r%pK$Yf(6!hTuBLSN#dC3D&aH7s8P)w{Zh_Qm3i%$zb)pvK#6zgw
zhQ?n?&(>?S@!q|ydvoeY3F-COeJadgt=C?+0<;AE6l>i%$Hu93u3k9LWHKeoGUjQ@
z;qZv*Y)V;HRAsd=sM>)FWvvC{=hr$GtPVw~F~S?(C=L;YrYtLRN&*R?lvn{)NrX~Z
zC4EFAVJ2Xr4JA-IUx#t^{5k79%1QV-O=K-fK@|CGN((M83ALdk1A0o*QIb$VOva$C
zICpA^jg=N%RS{9(s)Dl0!T4)C*HHO9*|5fN@aA_Y&#jTz3K2*TD2fEa`RIC)!jpoe
z*TG`ZhX>4N1zAy`!#Sr0yPS`@+{_eO1a!}z;*Hnu@a=EhVl>L=b=L^P_ekuJgM%TV
zP@Fz}hM?6ZJ31tE8l7bnMTWItD#gsSxGxQRvmuk==!w_I`$t(??%dv?D*up57f2`g
z)F*$KU-)Z(jX(b47n#l;?VWqnfJjBUz0K|4`z_i(^c=Hkf$G+rNO_K|H1pAf`EX1z
z&p<#B$E0~ho?EKiabPP>n}W6Cdra+R-b*`t<IA_$+nMq0ukF(ttnkA1Q}jmPV!izy
zli7$)r^VURn`q~0%4IsGPy&hqYjQfVz>1K0lCXVuhip8fKGL3gGzHn&EaUCB-sNBZ
z#=l^Fb%Ve1m;OWk?9coe{@@S)9r;rh-h3LTA@6Gl{2GK3_2E~0!Y6-3^ZvSo-lU0|
z3(Mn<a1eF#VBLW#>!!Iy_xSO5{1fi<1AiyycqkG%`CVn*v(ccnoT~+^D*e=;1W|v#
z=A#yeHYaN`{_oa0%Dku}tP{Y58mYU8tPmi)@oPvG8whn%#|i4Rn-Ko<gFw_p;+pU3
z1k!9#TfrM0uPAwpLi3J;Pdz$o{dH>lZcBre0;gOZ8F`qotY71-L1R6lFMm2`X=`hX
zZWyq#vckrxHH>piXEWy6oY8o~a5y9{3d+)>r8;3mASKTD9Jz2_d9kvV*>p;24I&6p
zN}!!1KoCettR$f`1P(&aOM<D>Q<ZicS(uidO~W{kZd1+@)DDpp5J}-_QG&$FEIR|G
zK~#>Gh4o0V(U$a8MdS)Pq2$b<#p-g4ZcCv=igX3eRG3QRoWvo~7E}-*)Kk<oonz-q
zU*O>E8i};b(v<F?M`aDNsL)x)^k76bo_d5`5a5D<EHmg5CUb|Km%O;L&HDH?rh`v%
z<BbtVhp+LKKfcf8NOJkvvz%*L+Ttcs98i=67tfxd)oPKJ72#}*$WrnwN0$Y0pdk#I
z5prZpb{_2bH0i%5AVvs*t_<5-`+Vt3U#8RQu)cPRAO4|FbM4wSzVYfeA9>H?i|E#|
z{mt*uS2?<s^Y$dA*N)g|E8jgXC9NbTR?<fYWkFEobmD;9bH(=D^0BI5(AwhE{a+_-
z{SZ@k28uS%Upz}!8J6#Um5ux>blNk<mE-)G3#cGuoThZ=8RPfv5cIk%XE{}#(Q8XG
z3Ga=E9PS=bB1w6=!lG1yC=3wDMXDi9D{j4en}6{e|D5a3UgJkT{ljc-ZgTt1ou~Q?
zB_Rq#-MHo>T<@tn0_l4S0(tDgH+6=lo{>b}TV8OOIPc@0qz193f&BV|T`LLJ5Gas%
zInpXG7O!g#+hai}_ef3galIbiU^H+>-_+IAW}Jgot3@1pjf^Bo78)gKnsMvaE$-ca
z!1EvZ*b~p8wU(+XkJC*LrJ+t3R_DK2RC~_#bOFtr)o83H{U@kXTRw8Z`8t%SmDhY>
zThtlHCd!dmXOWx$6-e)CODUU~_35F;xCn`{j;bnZ(1Y`JgD55t5|P(YRppM~r@Y8h
z*r39U!Bqv6PwvPFA=up?k`6~ial~MGiS^BOmR6P+tS+&3dIRu=ema?8%90@0=(3{7
zbE=~9-%BdyS;oAqsLIlVq>V#JPeE_3^OC(ndLV<fI1L(0X)L*M<RZd2Md`E`vQiR7
zZ8U@<R**=rQnDj@lv#l&Q`%0jjNoiba%Cmqnbi*KZAmw<v_e5!N{lYBW`=cjKa{gL
zBt{05HlPv#?KmL}=TGc$3MrZI4|(_Z|Cr?ZfUWTmgW$}dMO#$$GY#nVe510~6?r<R
zB|#@KclR|{vYbd{Y@}ZzOXqC$o@bPHIlH{c`JQF1+M|E-%bacBWim_IIJL_9`X;k<
zgvd+s`98B-cUgZq#iS{*GPp#sZ7S|=ZBrDM8G`C5-RUs$sb*&#))=gD?C*~F(kox&
z*(=ZSH-6!7^YX_&{saaDNuFwU-hIHeA395VG{Vg@S}U6*odGHey@yd%n6f~Z6=FOh
zEJDmYXFi;xi<~HQoN3)7nI3X5IZZ0tRCz_L_t|Le)9q{0O0u%NN~Hxi-rIpw3UfGQ
z`|cK-*PdfOn~{x<P@>|#F}$~R#N0{dLQ=Z-G42ZZ(DN^_ytK^Cn>X3n-+Lq{sFY-N
zWr^jbWt31%r!z*A5k*-tnH0SK#vA<h@BS<P=3oE2T)KGisSOA;Q3RX>-ufDsd&m&D
zg_kWB#V>!L9Y+$2;q0+XzNlOH!=~w7SVSPsr)ZiQM*~-FE_s}IIXOpl{>2@Cf^#gu
z08Prvd)2C*Uw_|OmQj@z#v10cIa!{QrYS}KbWvO&9Eob^L%iGWJ$>06uYm9%3IwVu
z>olJ5&4)r_@zfdq|Nbfna86;YK)7QCP9=OpVa-)p_jy<<TP|t>G!dPUUdq%-lxry6
z)$zSqM`e5$zBBdbNFQMcC#Zx)yE#&LI-p2OVj+l~BQT!A+h!@Y$Pv0=P>!Bz4hI4{
z!jhLc+dDhlJJ><BB6|HE{Z5yqR+sg5pQWHhJLxiO*XyEAQ<jycEGm}5n2Rc+DmB_T
zilRh@V03}0G7uG&bd*NpDvfjkoWfa2OT*bLV_M{-wnWAuDz0mMMTy8gR5{NJN?>Fq
z!dUuPR;1&FzTiWf0hg8|PWRgM;ucN|%CaOc4l%aEIxt#NRu&=EvBO~uQ6L$_idcp`
zP0qqv#o@gJ^3#fqL6`e?9}o@#I)gq@l7RBjsRtNUgtIwvrZ8E>oH4qrz-b2UoGX(r
z)7yQAiCU-75|@uzk-Mz*XAl~U?a)s8Z0!!&KN_+5V#Z{5A2FWOFDlZ}5#%${0Nx%>
zIGAMQLBzaxR1j)qX_=Qk@?rM(_PP1qt;c3lI-M4$)>mmI{%k%PkC>-(iqf#Xb-?fb
z+u!0xKJ^)1xc)p*6!}Q?#CUuW9WX+%J1Q8@3a(vVC!5YaC2$yF5D29RS_u(-(#Zrp
zIwC7_*2J7FfWzUGjg4jEPLCCp(ktGg)DG2yAoRtxgK@$Buw*i?DAO6Sa$Hzf6q6%n
zM-vJW5>ypBpE1#no1=pLQA(jCvs^zCf2=OA@DKmtf5wmg*pKpSzxpfuC;#w2WjdWb
z^qN7h&$CyqbMD+Z0vT}s!F}Gl^Dc+OA;wq^4<~&2D}ThF`Rt$N;>8P3w9p<ulfoBZ
zZNs{R7wWR&IG<h@q3d(7eo3`JGl5tjTTX6(#tT2@)@lm2K7Hn>i5BbPd{aMpe+{HT
zXq|jbbIs;8vuR46=P0G<^?E3!h~pR`M2+H@b8v95P*5D#0E9p((lk9`^giIsIme}o
z=ee?ZmdSKNR%E1kO^)d7LSEE4kM?oSBCSO<pW`G}C=e1S>ZnC}23+dYS98!iJUB$v
zQXam*Y->tk=j%`s<jLqn_#f-MEUk}LTy3FzJsJnXS(K~Ip7qZiEJ9j3u{Sr;7^F4`
zU1G`#stQ#UQ0AzzB(R#uJY5P)DF~Mmu!hJugaK0+?DmL5bHu)L+)x2B2noUfp`_P&
z@AnxjEfGes2Y%GR1>qD)6c8m2BR%OMQlpgh++t3nON}+)3>fW@#<7}CiPJfgY|gyM
z$)^*qh*x<^d1-v3C<H_@Cln>i((<9xEk1F+!<9kIawv(cMHd-a=E%zmZ3`-09>cbg
zSOe055{@`jbUQ8jorEZOI*A7ZWLj{4Z^rfWXX#vA^J^yvFv>IF$~Yitcd)u*mgf|@
z#AI`xO*DEgm`=x>I&+?`tO)aM*2@RDz!7$UP%_Old)qmO6HAsJ;Iv_7DMlL2?BI}~
zFyv04^OT|}*eV>i_DAFf=1x;QW`F(6&-@Jk=pX%8Y;Qf_@BRJ1#}~f%rHB4K3<Az>
zp5@y0>n!yLOeZtmy>*lO+xN(_oGhz&=f>N7;fue|`Ag5!@Ao-4IC$t9-`9X-7HnbJ
z+TZ8g${D(YC6L1FPjuQutppVW811QoJKa7whp-M;RD?1@7)zLhgq>xqHq7T4qfyG<
zPRiavMxN#<=@@iE)-LwwCKi>H<XK8@smscxOV}vE8pVN5xIarNv;}AB_6FE2_343n
zeEg-C_|YHz3}?=s;q>VZn#4jgq!6qwuky^r%Urp7b<y2@aInW{JVEQ6Jgc~Q^A>O4
zc#GB5Rp3!*>!1FpJ?T0^Ir*6lil+u}{Vz6-8}W|&Dc)cBZ?xro>dAXV9_94p^XHsn
z^V9|({_uwwPsXHm0XWMGGF_pKrmSgP%F<J^%iMr9o*KtO<s8Nd53M{&T_T#gM{RuT
zNW(b53U33Pqy7_C)u}Eak<vBooyXxB)^}Ui88Rt+T_BWRZaEM&z^SflIBSs3(ncd}
zg)T}=o>S#HRav66A+VOfd0HAH11g>lg;i)lyA(*Ph%H0}C<G#ul*%!4h8dcXEpSyy
zr;yBMQ?~c_k#)a~3IjT=4oQ>{1|hAuMW8|gl^_K;r?IvKVd{v@P-shGEM;Y=j3&#*
zlxYr?Ca9S!Lj1fqA+Ui5=vZSI6cd&r$0skZ@ySbTT<nGPsgOlRUSt$SMV9;LFKo4l
zih)HW=;6`<K@id(3~0wOh}oYwSKn`xPB1f${n3o!WXO7dsRmg1I#H|JCWw4QRTx8*
zbm;a=q!3&z1xFKKaK{<igC23OLzx$hrzv}*jGcoSqfv@A742AYcBRegQjDS?tmXt5
z)-RnSUS7f!CDSV4ekwT}PjLuBrD%71%w{th2ea4d@aO)_pXJ<zbHs5#JBc6pUhSmC
zxijaucJ&&A!GN8e9fn7T92_2y<vtxZ9FF<=H@?p0D=*ONb$_q{5qOlu)Ec@nV3tkU
z9UZW-e2Q+bhl(SN6U=8R&KQJq%*InD!xXCxah!1aY)sPXAyr5@nNmv0?Sq^<_a~6e
zXvLP5zTnjP0li)aB^`O0la6M{JZ1ItX-@z3|1DXoL!~RGF5<Rpu{)nrR0bnp<Lp`7
z!4|`jHy}y~KJoI)Y@R-iF@|q_`&*>h!`S5@2<Y{CEU&Dvyu3_Rm2^6t1xm;{hc<%4
zgE{x^-ldf!PqpBnDcqhsV2=yFO+Ngjs$F9!j^A(IS1@k=<x%$Bra{s7A$ZcpCe)NQ
z^+S_tYU&wqteAHUNd;r<qS?sLN!|1ogtS`zCMJ^^wr+9+PBhk8uZRXDNqceKyn6Bm
zE&&`I?Bk5#+?g}PtrkhEg;0vpRLtfn<H>~aY{Gn=Qe-)LGRIU^UAQ%r#!za5HjdId
zjP-`a;!vvA3o#n2%bJeIrw2u2JnBeCXWn`*h#KpLKc#$+K=XWpx&R*tK_CT@^#51*
zYhr8q6Js$|LAkq)(S`2_HyUXjiuyW1lM?e#>+CTM&Oc~4yunzXI5d_D&`Og_hY1u`
zc_Y{tA${vr1T$M9;{+=O7DweQMpuldM+|G8CnW_+1W4QRH7RSsXoRTe3*R+f34!*R
zR048T1#Jr*AjTKlEl>zYCSj&ajJ8~B!Kbfp^5M&;S&sr*x`3>tNHel@Mqw%nV<;<2
zXgz$q0d6!jGD(u4Q~>p}M#!gj1n15%a9(F-`)H3=9MSI$X!SZcDJe_Me4ZikHOKMc
zgmPXWg=A^9Lw6~lI~Y)9Ia(VIGtJxg=Zp`=v_wU#ZMk|rVP&vN5{DRDdG*aSr)@RY
z|KiURgdrm*FivpKCfut8ldPh$g1Fab&>!&b{KjH$pFe$uPki!|0Ni@-7Vq7D@8SI=
z1npLf<)vkomzU}H`^@Ka;wIw6@0V=uxO;btvM6XJtw-kO(FjC%9J9sIQUWC$!`X;9
zirE{@8O<tsk>b>;B~&OVXLDSsIdyIm6-k`-jeSZ5RI?fP?(XvX{W-%)$(c3DwdYRJ
z>4XFU#Bmqr;BapXB|}!$)@bzyBugvwfBJvTAN}sHqvrQW!<YjHdEo_W3hB6Y_a>D&
z&ig5)SXx@4(iNl8kneo=J5=xEwF-lPAe0yk+Ui9Qfvbz8K9>(grFrA6x7gS?PaHit
ziCr_1C%e3X<GRB8n)Xh5P$#|l#g$HY`Mz%Q@cS1X?RAq|^W)#_A;Zod8<3MtM*cd-
z2&=|`G-<A55v2g70*o<CCR57NH(;J5gzR)Wi?mi!8{e`l8Bd?w^Al3Cv%SMBU;ZM!
zL7!f)Pj@gNZY791*CbRxx7}fRX_-JNV(k#tVuiq1=OLuE5h*I;*$A{jTLZ#_5VY19
zUBl9~#?_o;Si?d~JxFju`nr{^RU@4{?&qoh?jvvwR(ly!UEs<RTb0=QGo1Ea<<3-)
z9TB)1wI#fmQWG_orlQoA$~X#b8Rd`WxIf5POA?;3F{!PXP%#l2Ei}%S2rGyTw45NZ
z9zr{m0;?+@y(vi)2436AX)F#A_&SkPf>IkK7O5@58l?5}-~^ubU4T&%AslT4fpxTk
zfPN5=1R=^nTIS4kg{TZpz=idMpS#rOL(iNgGGNCu%DkYO&&aYFh0d|kdTki%2_glM
zB*~%&Ck#V6oi;%b*38$QDK`wGC$4}v2PKwBL81hCnR7TjqSBWAJNr0o*;wt;8+6b`
zMm10Ab=$PgY+$V~vI{AZ#xdO6<*m0y+}bUOf{JSw2Anx{8WkGiC?<-!jE?rm)0Cvu
zW^J%c7%0|0{ZnjDQ*Qn7@340IJi|C)oEK<qD4k<IIYLjz58e0D@-kr*VvXU=x87ju
zeR2R&2*M};3EJrz44_UpdLF`UE$kf}^7dPA(QUOBAfIMDK858V7Lu;;VmlJWw{Gq7
zy*qPW*ck9*AHPhJgp_$ktJ9`)stZv>HuVQe6er#?2a36rOsk4BSC%+)c9nLgOGinf
zD8N`vk>><K%<}RYN=YgephATZ0rzg*Bu)ZIRwzmhS>-vpEac{4XlP{&|MNfpXWV~q
z524^&-~9GtzLz0QQ^w;l)9Dn6taZh!nkwDa*K)k_wQq9n^d9|g_hBRO0f#n*EX!F4
z?wq6ysZm5t<kJKwV(}MnF7Sd`CszD|`>8fArbfVcW8uB&qG%$JI<;pK8aCeJ8pyH1
zvGxSxudAB`t@Coh92=0bEWIuV9*U`C$e_P+qW{5r=|%&>k<Rl)u^lHkX7l=BLjYMe
z=l;DrB>OEoy&m1c60J^)C{73>pR-3wpZIf*#0Wwu2*QXkjtS$KAdE=DfFz7CEni!)
z4@G01<=9r0M-<f(t=9NXZ07{X#20C+V@70ad^4rB+NZCq#hDUacm`Tm6__%|78$xM
zaJul?^mY15;E09wDV)0B#5zM^9hG%tr6H{hnK9(XF@8$#`OH$nsdk5>$&|3vEX#mW
zgErbjmIa=6y)NGb1Q_SYtRppEZ-z+GPGaJ27cdk>#yqbmF?EDxX(u7QutjVI=n7{v
z)>L2(p%g3!EqdLUcDsX<z5renx&KfNZ6!Ffxx}YlyvV251FCsSHqObX1?9XT&oc5N
z$0|$E5{N|5iUT@fj1+=atIc4rL|J-Baume`D)3AK&NsHUlH`eNan0yUNhm6!AmnhI
z^7+>evAN}^KYD?s<t|27C?V;tuak7UUMhK(E!JXQ<dk8=FxSL=$;D^Rv%I=Q5(KoP
z5GQO+<SFU*mg)5dRJOt(2vopqI^*rvze9Iz6B|V6S+S_!Rod`=_u}oFH~7E)r~e7p
zu0O|Dzy395^X#F2cX)Y-$#l$QJSJ2j^K?#Cl?w;dTEDw@4@UgPzxr*2RF62+-<Qgy
zEd|m_tWgZ7hF5ORITgc4F7=5+k8&8St)rq4Yc1K)5iU!KyM06ukmWf_D$Z@LU^dQi
z_OJZ!*u452<%9P~zx5^L)=dH_kW$m>_8<y*b9<MgxWjc5@_eAMNYc`>y3$$DM)>5y
zqj+xT9N&KJyS(<j?;(^}9Cjy1RaP7w4cXp)K&R6sREp_rLY}7>Q{gyn(hI_vkA3W8
zoIQ7rBz`o1_=AS63HeSowlxN1Au_e-L_gX8;r|e*z-vi~CQ7h1(nK$!gQ}_)b%(|<
z6h;1U3d;&Wz0ZIgM;8q}$HQIbZc)Q%{w<wnlvTN4>=nWnTbsFW-ec-oKxsVlZ5(^a
z*C!ngA2GI~h$sw*0!0{j_FW|f&Qy$M6Ap(%_V)J~4M(V=B9e-@mC)|===Qp_yFFU%
zHgVJ<sHt0|^c-Dk@k|_!#t0ls-a6-5b}OA>wex&!*7LPF=hIF`SD4C28M@4=vN=vy
zJ~yvRgeehLBZVQrd(5(?pHU+AsX(JCN=>0Pxi<dC3Nqs;9Av=bi6nl&96HD5a>Ui=
z&vJM;B0D%j<|U!F1T`0y#ZfrJv?!Ptn#u|a;}}`X&^QPc=byR4g{#jHwqhpJF}L4)
zmv`^nXRaJcD<%<w_1+Q}RyWw_En&t}#@qKXvk9G0(Gr5L4G~eqR974oIrAbXD=UPt
zoN9%<aAA{|pS{ddBru2fNk=oXSw)^{vUy2WX@VqXu-qkVON=N9W5rUdxA?A21UwuL
zDNBu(o@Jqtb3RrEeX?~JAxQ*SqcGO-+P#dMyE#Ai!V+iKx;SfTcYB0M;x)<BjAA_U
z!fr`IsWsYIdc6+UuWeAg^jV(&+<!=*3+AuCO8(~8X{RF`hA2*GwOj0EDeq0EL`xT0
z4<eKl%u9=qAz>U-l@+$GjXtis<?!f;U;Xu8Ck&M5^?RS)wAL}7&)D7DVQXiDs?<!z
zBhvZI)9{~=ws1~x_Ut*XU%N*BJ~;WKn&=K99Er5#Mlvk~vr&aELN2VqJoBEqstiS%
z;-&^&mPicU<rQ2Mp{=3Y?}M`>QAi$i7@S%q-rQtx?jqAO&v5TI|0yebw}|4H$`-u-
z;DE1Z2^h=mU;P#C?hi565kxK4)>esoZCoqH=B00${XW5{an`rFKJMSFBTeUw#zPK=
zd$i*=X*&0WdG5HjAeCTwc|ejx^ap*yr$}Qx&Vb0qfYd~Q$2E{+!y!(DkV5#^`sd;n
zhks)LbX8$pV?e-~CVgcVwDi*YcFTtiNL?JZ?>8VPQCQ7u>TdPAG45nEc2XhnWNig%
z$m&xVj*}yaBjU(Yu7^sZ1YixiD$qugrzv@sdd{k{KstZFc{Zmk3Z~;Dgj9<z^tu%h
zsg&o^iXwub);F)|!9BU3_0EL#DJ)#V0)r)Lp(`P9SZrBRmN~k}Df5&fn<1<w5ROn-
z0_g~q@BavJ1j3<&Co?>pRTPy&8%<#}g*8;xQaUJbl+Kfbmhgy&`9X(EhlFX5ORHVt
zwJtU<ae0X_mcUWNykRz*b2vO=IL?`CLsWR*rMJGqr#}8MuD$dUMoC7K5$!mpGKLGQ
zF;|~`jv_0N6?9b0b7wCwaE?3QhMl)wr&VRN5JXkQd^V%d1%<N|2CNXAU+?kq)pLC4
z^ajhoXlILqqdm_=TUg4<lRvh)ZC2J-SYBU3Q!ty42t&ov(h|<WY&u0(UY|s%fHcie
z^>c1_y2Nq(P+Itd4gs`l%-_5U*c<0;9|+F0A_C`_W+f|8i>fHlMS-a*Y*o-|B}Bae
zm9vC_qSfvAf0v4MeL!o&Yo-q_UE_AA$LN)R#lT3CR*OlN@%i^A6dUWzzWH5d|Mb5>
zEiDsumWbnomGw1j5+a<#=EXvn=5gcwY5bET&vRz8F^7k{gn=TR&&l!(W6dI^4Un{A
zR#%r$O49H5AG*e4O>!732n2KM*c(|W;8Ht4S<TM=5vv;mqA*~7I7X)_p_KH_Y~tE6
zAV?zLOm=7Ih;QE9;=Q9W!Gr&jkN)gm=9#O{aqi+}S|9mQ4u1Px&V&jbM7(pb$Hve6
z9M7CT!*BdQ{-3<@)mM4p=YEdsAHPJm-(_`UjXQVmk*5~(lwqg?N^MB<l%2gD;-JOW
z?k?lmtfm9;uWJgUU;6TwC@aekedb4==yRl$439><{pL-S3=lP}R;pv9i0^nl*3fQ*
zr;ZP3w|Ffy57VPQ77>JQgrVyHJBMjdP)!ZQMjUr*KWrQxN1%8N?QO2@|K1cPo5oAu
z_mS6k>ib4c)@qbeb<a!v^F%Y~?|43+G14VT((+8UQb0PV)D?MAke3Bn?irh-ND&!V
zM_dl9UZnOO0x;mg4JJ>CqJSt2y@pM6tkNd{AsW!NL}CEDn2Vaw9cw*XZjpNcJ3++0
zLuRXjK!OUCFAoc!HZ6^Y($rCm!#Y7_1ryz(vJ&f{v=Z$Ml{1uBN_<+P0*t7Arl%O)
zP_dcjlx2yHBce`%TI&-;F`*SiMvzpFFt=2zr?`3l7DstOyRd|rVe{e{UVQ#J&RxDt
zB_yk54naiN=~4w5m##d^V0n%GgAs4P{sv_faQ5N_+O9zF-^c9kkT^$G*B!$^5NlXX
z0?wX0#fP3f&ogUFxbY$P?%gMwj7ZDM6ERxPa?$H`Sy@@8)k$zvm?B4FiBv=qw-G{+
z&NI?9BZ?xVP^4K#9LFR}3EfT)DTCv>;FGLF0bwhkaE@6H_RugO;7k`rqnv}$0ZYq$
z%xq3Eo8rof&e8zc>tU223=(7@8BJ2YcYB}L?rt-F{Q;Nu$GrIRXSj6b8J0iv3EqAE
zRg%4TN!l&m-92I_?(?}{_*;DAU;Imc>mUEutbObyp8u)O((89PwXsIrYg42-g?)Sk
z@<E2SkXIGM(GX)C^Lfhta7bR14;kAyih1|FTfFu5+w316J#>vnBM@upoH;{3>9hU%
zO^&9P%}}%1Hv}TUYQt<cCNP%R3A%%2S}Oy><sKjikQ`2PzJ71YyJeqOw|Cec%^BqH
zap4mqit>91rHGc+8R~%3vSOZBY!5T8tC-g62InrEA=rPwQaPjF>!RwAe>%-D#y(~L
zxgS|xa_iPzZr;2D2<!=uz8OLYqBtU*PT4;=dTcuE2Mt|mX0thg^z__H3dH*jNYK<k
zn#R22W3CAy-)}&g0~}aH5;d&Lx7*c+`-13j@n5kpAdSTAW3@yU2(jbZMuQl0wlOa6
zqcagKaQG+FboJYxrgy9EVF~DUyTq-8B#scyk!Gpa=+08F2XAd1qWdm#DSRzPIDC)7
zf)2%_pnNpu1-!HlkYPw@4c0}-hTdKZr10$@2*7HLaTr}u8I9GRQr}olxLA}0Wszg8
zp%sS&PGD#X>jqaD|K8G}E5C6o6jP<pNQ@JnvP1|fydJg|0wb*Ny%Ko-k0&02R;$IS
zZkwpzXKD>w^D*;uN~_;zFjyvOw+Tf^WF)P!;#!olH<{5h5;e(L*<5Ag^ad-Z)-iF5
zPE}%}nA8{!?!Cu@n-3VRYf@vFAxWbMxjbOy(s{b4ma*lS&{WI{Fr`9j$7(C#>g9_(
zd*uR20Hf{u>^-<kHl2Buh$tbBBjO|^=_GV|ZQ@8#mQ$?JB&~pU;_FgXRnl&E2*Z#x
zO&4j{JkN<^FDD#DQC;WMJZU;rD8<T0K8(7(!*D!7IoNFHth5|u5mFWz<Iy1ridK{m
zt*z5qUq(bB3Q1L3UcWcywNc31!zI46bqDPx@a`7t=XRM)N4)g$Pcey>7?hSs2)2fE
zauVWJkCls;c;WPb#N{lvV!|*ajsmj0q{vG2zekjBf&Kj<hX<p@zJ8pMZ3t2i4u?#q
z5A$I?)_}CvPI2kupWx`=kooojot|K@WGS)=X(L=#P`QB4>I%tVfM~^-qC#uUtM`xi
z#zBi+xx`L3=GDC&imBjq74prmzDjo~;lkxBnCXNnD-eNXIGzzZ!<}z@kypO?HRNc_
z$668BPc4Cf>10e*mR^<<;WyROPMWn8KY&+A3Yd&1oIP`z@p$^gQnx~QnDbV)dqN`i
z*gKc1k-0rKAnzwHY^dLx_n$;zIqXBPZA6Kh&pC;-3S>e<U2>e3I))De4<Rdj=lcmu
z<^TkFR+3X5zeWg@^vfp<!#cuJH929!Mr$dnvZhjb!e{1Sb!CYw7cU}|!U0)PP*z19
z(J7p$oAZpu8b>4*VGW&@qAnlR&{?Cs)`yoHHMr7ZD}{+dDqDe$Ff~;Q!n4Wxtd~R^
zNu|B0Q&ATgt+T$7u}(!=XE0c*JO@!X2-e;6&U)a71&pjCVJ8?{Md2hGhrwbHzS&gN
z>QB;vun5dk8W13k7>kOYFf0#x#8J#}G^U(X2$PZzB(@)6+X2cr&NINql0d}F+ZoBA
zLub&Vv((46dyo~Zg)z$qhs^FiU_8sXpYKpcA*$79j%12pb8`dP>!8~S?W9L+Eu<9#
z7jX6LMJ}vwQq5-EfAAiA`#Z?Muzq%l)wMotv4jc~NP`m<j*6_9P?j0(Bw=}}hpjY6
zM-#@Qkyrf;Lxd1iRY{g*WO>2zvaBN#L0;30Jn4BWgy77}FVk7Q%l^N6g}^FSmP+Ep
zp$m=B5!Gx)(pw^4=@E2#7;BMP!OmpHD|ZTRlxvIw$#-{dar1CaPj-3d?JczUW3GSr
zLp*bN#Q5-l3r0}pC9|}oD~9~afA|06;9$sk=lIBamu|mHI-ilvX9(w^&>2tVpZVb&
zRrx1JWkNt%7KBmA>CFv}Mh^>2J-!d5%+P}t=YI5M_CNm$X?Db<bhIOb&7mwb-#eP4
zMwV_e=FCb5t2KXc^N{<!>s<W!XZY5?{g-_6t?v@D!s*UOSZQBix-a?hpZ+AhxXo*S
z@Nc;&Dx8zF6#VFwKHr>daWFH?rv;b$A-#S=vbN6Ng9mhVPSlEN>y+~Q6psEhhEj_2
z=Qc>^Bc_vtI0>H$KW>ci@HqFdRtEUOwl^Th2l@9iAScQD;Qa=q`5h+=h<qqIkk0x>
zL>lUSS3`PFzQ1XXgSxfr5#(9(p5`?|fLj#!-EpK*zim+$F&_2wG;_GKy+bgY`gBVa
zFOXMDgFeeE%XM>K-BIo=GOc{g#}{)jl}1-3rYy0QsmTl-9=BCPXXS}5|GLS~)x~99
z$7rR+gEySRxe{$EuaO}nGKxScoRb)BeS~67{nH@2ycWmF8K|&~(P&itJWqsbP2R8O
zGjoeyKiM2Q=SYM=+KTbPF0nK8;}&tp7j~0wgp3u_^oZG{LL=CZR>)SDuu>9+V@g#r
z&4;Yoj8JrGw|Z#XVz7LUo1Je_%nhNCg#K=E^EtEWlvykoR)*3F2HgRhi9}k<dZ)*7
zoG{tH$#7?vBFj0uzRc#C({u(Mrt>-T*@%_pWr83i%e~N4EL*-gP=q8=pR`Ibc81Y&
zrnAs%tpuI~bGAREGM4V3kBk&06=^xAsH!JMVOeWQZO)mGUE|#T9_3p%80Ctfl_Qm<
z%1gFJ0mFkM1_v3-otTY5i#Hw|@`u}kvp@WE%*F@&=I{L~M`_7=c$HJ#Yq)urcDTYP
zKJjVpzxf8i?mgNaM^%)(aJtQ^*8JY>x7n*4dRFkGp<;b)1)H>y!vnglkRv5Xj3fKO
zU*oCA+Ug1`OD%SH9<a8$zDUD2<FPf6oxNTD<?sC-Nf0ydhRoA|WYnUU7|@b%OLKQp
zVqMDVL5t<x8N1^d?@Zv_vlX|#`A59<;9a6lk6zG6CIz!-hqtz7%zkr+{^uj6w{LMa
z_C?03@(7fBnPF%lHG<zcD!Kg!-z0f!NHH06e}9B%cZh{#U>w@{MyEgR5yvr6)FRJw
zcJ~h`OG6L@i!SUBGOV@8P%zJC3!=S~2eLT9l<WWh_-iz;dH66tDPvpb>l;C-#Wjy#
zA;mEY%DEE<6pLR!RO~)k#D93WM_$u>ZUYc#3e`=~-Z?N<GoFk;@Z24L2+W+|)_8^r
za)KZvjJ&6*!k~`K7F4_b{z~nO;B~snT7%Wbqo(XJ4LptUXr{tA|2~>}O@05kW0Xt%
zoedJBLN9nE8u?itIapQei@N|@Yp^$*QjefukR*8ea}UR>HCbw2GoOQlM!3UP`iXT_
zq!fH%zu;bKC@W3)V24hp?V;P&5+xZ?lAxVso~EQ(L6G!OafFH^=4HVo-Qf%OevOTD
zUnh(^SQ(+MWICO4|K8hVX^PgKR8tB`lEl1w^ET2{+<I#ZRT|!*0%7R3+r*QcY_`XA
zI7T=}x3k1REg=sK)7#rjXEWkN&^_|Z0cBZXts#ygjFA{q(%BhOX-!f1x}Vq?A%z#B
zt18WSBoVUBS8k^yJ6l*RMV9g1IsEMVx#fP)G8BS8e&rQzZEjLY!z2lCvxvTsgw_!p
zNhYRXcb>7@PT5>);q)$V9b}}fQ#^P725-FcT~aJdy%pjpplltnT^w=vjT_vb-DA45
zLwB~%*ILS_WsTu5uee|OI`o!N{C96`v-vCk8p0L_@4iXlB&y%1qcx!wb(#N9fAso&
zjJ2fc3~d{#{}c1`KCi|ZV>mcCpsYL>lTiLpIJpP>Vrl;D{U<2|N>JnHkM%6-47%Uc
zPGF5+Y(lAp-_;yoN;HR{|Gm~8T8R*-K%s<uD7*4RLo}cAaFP0vNw3pC!gDm>oTJ@t
zv$VAIfj_FMDh`K-_30!Q^KkqbPxA$yZoGe4J#Owk=<PfXRn-*G|AaSzgKoF`2!i8J
z8bzKn9F15!@DowY2lK?AcA-4(O7)jGwz$XR#I1agcA;Zd0=HNhj~(vB2MO;`^!xpV
z9PtMk^ZA_Nc<i~r5DUY2{26ZXwZw^eZvL$S6dd;+9Y;w*NR(8Mac)?oPgUSM>FtT<
z=6KjCa`o@)Fq(!!|73k_@xq7xe1e_99e?&oXU#4mIe#G~U~PGcr1fY=_4gZ-$%M&d
zvT#&{s^z2I@rN(YEXSh9$P;pUP>WJt)XdMnJ+Y3#Xj_YK9TSuK6sSLAo9AMz!`7e>
zSzk}qMboEV<|m)s_t{gnMk4y}=U;RB^y$SLABjL_vl+kf8^6J9I<2S1)uH|m_8^;W
zcJl9!z1fqaUO4_G$0?aVabt;x{_rP{6W3}qAH3*~sG*pD;v1>Tifh-e^SRG`?gKXz
z9v&X@%fI}~buUGISUeI*Jl$Aap}E$HO}Y5}`{&QS@6GOgzkJ*#<<Zx=#pl&#{{w!X
zvMl-JCqKzg{nSr=pw|Q3yMLeG{LSB7+=GLM>PSyB#P|Kf!xQ8WTi*9?@0%Rw9(Tb<
zU)TJu{<S~Tj#-!&HK#|$@tg|>WmWOn&wiFq*7Ot~WW4p(TYUcWpGO73;%mtIw0r+v
z{UBp8_mA7;@Bi8-*5~&f?_ZZcXw+v#V{o2i9XjX5ygv7HpX2)V>mTp~;ni1P<?CPj
z8bKHyqcr7{-1PBSM8GEu#rtmZ;dAnFd--wW10IDZ=iAR=lf(Jo`>HA{{=xtGA1wa<
zhyfXo$NcIq|1xC;Uw-9lY@S-<Ge7oWI{k=FvO=%F?0Mtr%z>BQRGvoM`o=2fEK=a}
z)+(gI-ur5PW67M?0=GW&EsLByn-eE7?QWkiY7xi)Z5oOR>kF4P4Ta~n6E)?wHwr-z
z5l0D9NlzQ$eBo4Anyq`U^Y&~1f^1qc9D(X=^20y=7wL9;HGPG}8cSs=%AzFC3o28w
zwY|ezZ@<mwKKHo~G{FEnJ3IW7|K^`Cofq7@dza<qWv*R1OQ0l4uTPw`7EomW+{7__
zwB|~3)>E?wL5Qs3zOtsMUMwnC%ckNB$5mOPtIBJLw-QtsdaZEdznk~s8#;KmI4+xo
zB!ol;p65wOSrh*`8vXCX_uggiyWge+W<avK&Xo^;#7CP8gqDXHS9N#4wU+JO9Y%li
zZ#<C!dHbEW`L$pE6|zymn{U6v)n_mAp-)`}DLA!shA3)b%rSphqov_Fh%0Pef09!8
zH6cA~s*-_+1mitKi@#ri1n0=|)FXRZEn2NULEt-`wXPO@GYuWTsFhfS@XX479V$W<
z)|^7%tf|+z=P7&p&KJ3H{|o3GCI@hO?FBye(H}=Cxp<z&SWnqg<P=4Tv4%I^yg`yA
zPh>#8{`If(|NSTbEl1Og;lVx^&Yk6%)9W}P=&!6IgAh||p*KrGh+6SP`+N2@^gcb*
zXxlfOVAM9$IzKO>uD_K<fz^g2P6(6OtC=EdtqHqePsSa4?@gowc-G_~By6%6#_xSw
zlb>#V`<o2kdzTWJ0;isNjtkE{>y2WaGIVw0aamPp?a4N8Ke)%`XD&aH0r@xo=HKwk
zzx-?L9v(5D&$)X39Oq6gVTEF_vg(--PB;^dc8u`(TMya*lvD&^RHv_N-4`j2oigY5
zsF2b(mlg%qT2IlEM5xfqDqC$lsK>d*|7+l3{VbX}47}oHpnUDq)+`pzQluFVzVTJ2
zJ6jaM1i`uIKg5}f7wXbm12;Dqttn4*r{BDN>rXZyDo_+fh%38vTWc)!H@SM{Iabyp
zy2%=sFF!|VE!O!14%g6R1Gd6e9!eJkAxY9+^g0A8Tr}iar$Jcqtl((4$9y`(6oM#B
z==VAZ6_A#a)wPTC`x{t<@AySir=F?%>nN+5GD8FeK}5UVsp&I(C$zPWJTLj$7q+nn
zf~+Ul3miwf;+aclxpw_UOy%?BjShBE=44q)mY3Y@-@-=86M4R27*G*H922yAbO)=P
zIdhgkItI(@EUm6lS-nUTfH;PZRu)^8IK1dl7{zr1qEF)mLG20IhV`~&Hl9$XbDWZd
zt%Oz_W2HnZMR)xaQQLRxTB~u^(6sn@suYW<O26n;fC@sQII2?#zSwM?CC_qh?rx!!
zq%@8(g37_U&C@KetoQ|IJ>p0kO;wguRY_GV^hII#{U_|Pa~S$uv+OfS_Bp+Lmglcs
zr=sA>*=ww>oT4l}Z?|_OWGy!3z!g~Q(NJNSki;#I4igZn2o-o;2#3Zx!+e%98t;+j
zV_%y{1ifCDssy83Y_41+N%~mrFjixQK_Y4cB5)YAF1^5%3y8y*R^mw|k^UK5V;LPC
zl80YKw?L;bY(XS7AO7$)R+d*W)iEih?{!JZbI(4Trp1Em=}G1=CnoB|L|MsTd7bkY
zE}*N5Gv_anbUI#Hua1x%HCoVtt2C~v5Dhg<6wxST24Pt5G4Ie>Yss^W`DjF$rk?Ub
zDOzzt;ViTg2B%IFBuTwae66ZkhXUaZt*-oal_Cryq9_LZ0okPS#s`PE*S<p=Nsi<Y
zK}$%kUVer+P8R95hM}RV)fUUDWR|5*+>IUt0fmT(yGs}(ORJ~2aDJ1r(wx6^g&>O1
z&L9?ZSgVg(hpRNU(g-OD<EWNa0Hh=cLsU)h=!{48rSp_@Jf_MrzYmopNm}H_5Oh2A
zPi-K>sCIUJI!oYNT!rvwq&3DnQZgWn{d04+>1h%eYZyM*!oK=-`h=uN0-zg&JbUG8
zUGH@ZMM|S9%A%mC;p&H@qlf3`{SzhxOhy^|_X^6g;=NlB_`)B&MY|Jl{kaQVxqOjW
z2`b~j)*8b#8?10<0Z$FXaG?Y3|8FR&42l2U@Nk<qzxxVBdO#Q$)|Qtz`|MScBw}xO
zhx@nh@z%G$On><_XD_|T+QxZQ7}r!84l6|?7rwCi*7*Zfsh|$kB~}Ww$(fF}h~gz=
zsF1@w=EE^}Zr|eSvoF>cX@r<0A_xhRHlg#Lg8SgMqrQ}2nih;Qukf|CcgXizZG^Kt
zcS*6fw1jX5Q*WP!?9H>#`G>ve{H`C0r$skZrSWdBwM-5V*uHrao23NK(d+i;Z+N7{
zWO&5EXvEfy8?={KSUGo&?$UA{f7f2xiQK+dKlFVhM%Q}bO)^%1E_3qnn6TIPn`NAm
z&E^ab_E}n6gZkM!DG*K~!@$?XM1`nXWuIsc1q^p4gr#4&H(t9#JLyoAIo<!&fQ{2D
zSYfb6_<6(Dft5qxd7fkt5Y`xWS3h6lyxgh4YggRgdY2owzKM}@;z+Q$xz5VQB?NFd
zy3f{w9o~HV%dD+j;QZ!w21^?VB*q!A&I>0ApG=WLBAnl(WmS4jc4}mt6wJ#bX6X^#
z)+V_%2ywvCaF?CkZB~}n{a%vZdzC_>R74mGjLA4wk$kdoFqwe}smg+z_jf#-aZ%D8
ztkP?@Jo2o$w)1df;Xu@Nz=j&VUfA_Pr7^;jky>4{clRFqckUp{f>=scmRA_8uVJhv
z+dp79nQ%XNo598gE9cIW^g4)|oD%#2=&Y^F41Q1eNVck-!}|Ut6q6BEmJzS4V9E+H
z%NZXInNG$e8*RUTYIL2D0vQGb(i6mfkTm@JhjXwynjwXw%5&biyG>;*x~f=PJI%^+
zSl4M~o#L_$c+Rg6DLf^<Jl@L!ycl1d);49q_ItM(ZEYcSMJxl>)>as-u22;v^Zf%R
z^BF`jOQ$zkK68e+)54z}J|)_qFHfe*jWGzV{h2HpkOpK&Ui&CsS)s}cM4B_)KOoC8
z`h$VzY-_~Nq(nvmfq<&!Pdhmt&xIugVHBgXh|<is@y-sq1Y5yR{q&5br9RSm@QSw1
zf1vXZI|u{9C?X7^I@I;l0uJhQlwmp@@%#Vgw|MhAU*eM=zrxbS21y)IRTbJ!DT)-R
zsLBy{-o4BA)|*`W@Q-ro$|s4V*fYjreL6zb+71@1F%)G^7=|7#U?mpIY(7Fo9bWv@
zFAyj=IQSo7`+&p!JuHr(k)RWzR)<qq6iOR}dKxY#hmfGc(1*CT<lgQP+E_ZR7S}JG
z_qcVx-27o)_bUi|gP~G^*X6F$R#Hk&1ZX{@>wG%n+u!&)<NFVIX7dzruTQ9yr!~=)
zmx48#YCdQGV2|<61I}D~o|W?#y-}=l$Vy6(wK|+@IIT)QuSy|NzISFeCF=KCz4Qz^
z&lr9A6>O=QPNp<99&hLXl<$ua_)z`{kuUWWf)vDI3n2t)m2&gj_tBN+;?;Y6<de^l
z#1dT!jHxMXoB{m77DnFt3hIKRkO4|6&#7iL#+2;sJ>WP0^{<eXJACBjt8}_6s4yVQ
z3yjfqbXzgY57>YAHe35Q_|UZ<<J_t12<e%d5xyry)x>5Fzi|r+LP7#O!b)4iWPZr%
z;1VBx?xz`!9<X=+*C?xu;c$ph66s{^K=}7qCoxtbr3TTYE}v`?m6C`cq7+b71vl>R
z<E-P^y#vl~tP`Sqgx`P^tOet<3n~cel&10rvht0VO`6h}lKri1zVd}H5|;&+HaAJS
zJ){tnW#OrIbcLui*?7csXPfcP1J1qpVFnwUzMf$%)Sm_3(eUJwn#x(9Zd8Hi1F(i{
zGNym=5-VrUaCqkq```E$RZ%ccQ#O>qT7U52eDvTXSPkMSBn6tGLWvZL%2~1^=bZ=p
zNC5|jN1R$-0;jOTbM)DU!nQUZDhLSzPsbyqcWQ)?3%)h24cm9`^0hC1k!7Vgx3NJ|
z8;7DOsPe)yI9C<f(GioKEvCCWoO$sDx+`nIF=8+X!bM%EnM3U?G6<mVJ=0o~&!((i
zyT<b7CimZblilyVNs(scdCv0EGEUSfZ6y&Rz$&kdBP#XCUVf|raSoJ<P%6gg5qVxB
zUC6z=Lw0r!Sy}2KgwNAsG}d^k2b8QeqkaC&cm7I*upkv#I%Rj~Hj~jIcW>W7YY-@(
z9}EJjvS2nl<iWiy=Cdhb=+h|k$suok_YavIjrhb5|14n?W33ObmGHS(O-)^4Eyh)Z
zHYN-bgbFB%jM?~zDy`^r+Q>i=hESC$VVvN!M(YxXM1>)NY%F)J$?&B3djJ6{2$3?t
zRK9E4A=n*Gn58MJO9P+tT`ah|(_2ZLlZ%CcHP{ARAO%I5GC4S4KAAGy+gnJ>DyfiC
zpv#gpO&J}HeDkFg6xNYXCfxttcPa9W^UuFX5GbE>t@GXvpKCTogSP%Y!q}%K@%5Er
zHb+RW;vkhkIfJPxR2bI0T;QBQhF(fkTeI*6pQ7dn2*NN%iGb1+lz9af?%mj9KAkfd
zbP>|j#@E-pgms{TdLz}jWn0(k@D$aP=^^`v_u1XsVr%O@D{Bc7K~U?Ym&Kf;;gH>(
z9V%TC#fsUiWHj00)$jf;#d8_gu71*srwLEk*OaCGI;;>5zYbMG7<f-HE9OimLq=hT
zyqNpGg`mKgk|1nRmfE9og!Ji3D-fmj^_w~c_vBM3kU`{g>$;+<DSNi|kEqIuFbI$L
zYC|UJVUkGabxdsS*a%TKqB_TXbj0*%$ieO|<M9z2{Q*LVnhplaJZC(fFdmI@#t=$L
zR#nXQ_INk_I+s57G1f0!^lRHSy-u=j;n8(n%3@7G5D5tUvn%q9Y%*ounow!&86LFu
z2X+|v+GkylQA%R8gF2#qDo3Vvj#LnV^l2ieeC}ZTaLBdG7yPrgW})@cAV(b%<S}c5
zvlfisJJwpJhlk83WA^UfV>X+!vbyT)#Zn<~6!X;2;dp{@mOx4JdCquyi!4pK@`+Ed
zbZWzk`YnJuHS@Ny80&wpOb8Lkz{AVSEMq>N5+xmsHUtQC<)aG`D$hz?M{H6_r1o`#
zW0?N&cnq1MquYy!%z)`ECI}?0c8Dnz(^<~#+uJ<%>^YyyG+s-=)gc@>q?TT9!1zfN
z!XlL>osD?=^{=wCbBE=nRi3?k8h2rdmDN?OlN4o%os5~!W^`H+t=1}4Rk6LZ!+hp-
zwfFY_kTaLA@yykaP!;vUmJlfEiS(Qo<#F28krA}ID;Nvubf2TyO$?5z$Vg)7cG`5h
zJ&uM4pgdB=IMA-<+foV{#!x+>rniH36yaP#CcOt7g$iw9mK7Wvjalk<k45ZtodOi1
zS$lOFsc!IdKD5uKW47PD$!t8P+wE~~b&UW=tKG&3JcfL3DD#{oiizWxsw^mmp6jN_
z3udo>kEPRRSzceqIo%|%kkT6vXEaI(Z%kYXID%FiV=eh)Om;X#J6|&jh46hdNlcMv
zKDCC!;4sekgGmZx@MQf9^%T-yY16MdghvrButZUasVu|Y2|HT{^adSYz;zYQRdqNo
z5yD`MKx=^&ras&)LRJiqws`ZMuQ5%>tPGa;$VV@GfJnQKt{g>CP?R}&KBqtE;9QT%
zY|1R1qmYcpV`j6@b9(bxmIv!py2Kj1$dZy6TVl}!!XFaeZ4g9Zm#WlE(=F0$n^{)i
zge7iCdfh&8++sW#p=5}&ppB!lejN%C_}@Qy2tu#b!UQp?kc3Jg!oUL}##6@ADXYs%
z{<+s;QQrCSH5*eK4PpHOBD{ydXn&8b8#gezB1zg@IemtH5)&o~+PRt^EXPz8omLBk
zU^bo9Owj?;=^PVCmNqwt<CbT`2NYiM(Nx}$d6c9<mpM)($S6c>O}4*JaWo>!QzVW^
z3R>+ps_xekf$}|K&I5HE)_c=Wk-`!}&~LY>T!avcD3VY+UwcPmX7f3nPTS9&)*h@<
zJ7oaY3J*vS^?B$Vt}Hm(+GhLKy9gm^ser3zPP5eQqT(3iEM-x8UOTPnb=o*!JRXtf
z8OB=1lY7Wk!b@wbsQP}Ib_xk-TN_g03-6kOKuSaqP*o-A?jG6UA!#~C3W$`X)oCM?
z#MJZUvv3Vi#|H@SpTkFI)IpL+T1kW?q1{gSp&z}>kN@<uEUk37@y-L%tUge!@dcn0
zn%oWDbt7(MY{hgu<jvP!;hnd>!ru0~l=CCbtS=J>f}QOxM#Cef(=pZ<TCFxoyF=XW
zFfR(mvy>a}?r<=ibFjb9cV2x3tQTMjRDi&G&r^aB&4H}3y24mZ5GI6Cn=tIq?VTb@
zx`3nI3Az6Ki*$NR6jg~-$`=ZyPaOyqdY$OV3nV>B7qA23z|%-01KO>a=P#Y&g^L@6
zfn@t&SgQz9M;`XL)7u>*NgCcOXEf<#%!7B{Vei&0X8Ze?ykMo*M;Oa^JR+UXDKbxN
z5XCW391%nzg|0};g6;8`RBMh7581l?9zuvkovNYRXuzwBf_P2$uA&eXMFep|-0ypI
zmvgkjfHUXL5=0@Du24#%f}lCBLHZQj(`;OvBWVXjDkg9-gY^y{`ROY>|Cx&zW4L|e
z!J>hUrt^CNkM-Zj)<_&}G{$NUhxhs3TVLYFoo{hCybmtt{P}fsm9f9S%XB<qKA$2G
zwA*dkNf)VNrn8LU;hY<9Z<Eapx8A$Kt@qyaM2uln)1O!$jnt_EXH1R2Dsj#dM{Pot
z(29C=+G~VC3|I!sJ+556P7s8YW$BSKVF-9&fD{2L@CDeXuphJniBuuTfYm{Vk6t;=
zb7xnX7bQD~N43#C)|olke{m93XX;GDXm6i;Z@j_5?e~};4w2fhHdvxab0*^n>3mL}
z=SZc9<CwVBLIje$Dj20HTSrHf&T{YeZH7C01ffS5Nk3^G-6s4-Xgq0GbJ3|VLWL1w
zyF=XXp~3*+EQ4N`_0yX^LNMC*j7e3O6+Kk0NdrIWBnlu20-wrOiqp$IKK9HRp4nXC
zXgcR$G;WrwzgOE7K$>NId?q%|;_$&X_uhJw!@KuLkB*3ttoD{jr!%ILDQTKg76mF$
z#BoC0YT+alMb2<OXZz@g%35yUxWQyNB#a_d==FnG@WT0<YEt9Yr+WyZ7#T)Hoi1^=
z=gSPvva&Q_WqlosKSP5sK-NaYNwJti@xCbTk)7>;Fi^DHUh}x!iuvfv=lIE=ex6V%
z?!33l!NJjS0j$o`H-v!xOSt+F_o6a~J9pXJdKV?26)DPmhAHQ?lNgg1Ovgj!lPOhE
zA&|sz!eFpMpaKpLCuF52EmMjj<-y$>Wb+A86p<tmQ5+zILkL&r8l~qxsulQ>B&OAF
zgM=&{d%A5Q==J+td+{T*+HH!mLe~&Fr1Vfm8PrISCquq8BT$k!P>3KV4g#J#cZ%mO
zZeXor=U~J<&AicQ3VwABM%15Ay&$c%7_Axa?J++*L?VcUq)Jm<rHR6jswm0lK1yq#
zfMFET>U40xEX&c>P?iN%Suot*_NgZoEc7*ms~a{OdJyaBJcN{<YbXwJ#!#jyx+wf1
z+v%`*?i?xzJcWm=VYN-Zx7O2sS{;TkQA9~dqFQwN2`~M`vt0Z5MXIdg-i>YMvzhmd
zL=*Nm&(3plH5<ohO{H_Twr?_-?h=KHC=#Uk2&FXbRz#Lfnaz$otFEyq6_B*r4EjrG
zZ5Rz_l%-*w&B?PF54LaMoF$AR;#NQuD-ccITY1Q(@j_8b1;j~A+-_5uk}Mlj=?o<m
zYik=^dFC2IDvH9V0UM($l|reohGPa#mB9)_MHmFgAf}tdeDvx$&YxN(%?h@54=IWQ
zoLevn2#<n63R%x#eTW-NRTLaPxKBBq5lDq{P|oKB&TF^ia}TVj$`b4R*$^i&txlU#
z8|Fp54vU;(K4-YI<56Bp5h|a8ZIFrT1SOAgeh#C^uR~Lol=IYsA*5n?Wrg*P4IGYA
zS2Y-+hQ=$Oh6%#Z^QS$95sqv0qf!yeorI4)dx7=kK9gC-&i<k2*m7<`#Mz{6{r9Q&
zQVoYM^PI!`5724mzq7TJ>D<d03q>)Xlh0DB(#NhU2#Avwt#+HND9MVtK38O9(<!69
zeWa9x%1aW++W5-)9BG;x8)R%01%zSX<p<IfYYa*$RyR(uys}b@$>~KJUesrjPx}W8
zd5LB`7KJsNRgjm2LCj=ox$(|DvNMi$J7RD9i0$nI&YW38Dv7Z^ziW*3dH8zcx;k`s
z)-jnJFge=h^yv-CJZFD*i<Qo0VjRnzzCU~l%k*%HwnIdyNGr|W?gNajSRQl<6a;}{
zHXm|$xX)nuRE-$&I@hMAst^e8b<_umkuaIh8BH@N3yLZOEI|;EByF6vzE+`vx}U`&
z1OH$V5Juq>L3rzVCoMHq;Pna8tYUlrkSJ~ug(0J9%5<L6X|+AN3Klt1e-Jhjv^9Rw
zInwc%ax!6gX^AY$n2yHu)?-2x-FC0;H4vm(iPb4c$W6uMXav^Mi4&x?L`soP$K>gp
zFphD~V4cs`I$NVG#0kWT@Tj@0$VqjHstmgFsVx;K?=9O}cUlUXry!-kI)UQpQdk0*
z9Tjw?B~~%h{V})Rx{E9WTCIfLdqc+KDeZ0x=e%f+*Kaqz=qp(C2np#K5)MZ@<mH%i
zXHRo<c)-zck5jAX>BNd)$x}86t(lKTRHj0s7*BJC!#$*QEU$EN7~*!s{&1T#o6+g?
z>U8dL)Y!mXYp^s3g>f(*&6&*#;?Pl5UT`A}Bf>Dk6!kh3fj30TuSKB@$|+Q+o(9Xo
ztSAv}hfpC$lR5Wy4@h%OpaOP>WAdV;+iBGhRE`VjHPPoHGOl5wv*97GELdAz<LGEa
zI+@X1UL{g7vh5EnZD2M@v8KW)Nb`*OWI`aM7iTk?SSiMb2VP4^)pEg3EFz)?l;J<x
zAIR1jW?4qAOO#QVszN$P7=`{|b3V^6q^SFJ1kwqtlql(*oKDA2A=9En#0kid{i6w6
zdk0k7!*h2IM^t6yB>_a!x$h9dE)+)>ymUgMi;UUu5M?c^t7{w{9ug4H?hZ(zR^9d>
zsI+4;PI0!vN|>i}@-+1#YjKP-6_J#T4-e`ZNG-j9FK0O#czcbt6dqA&jA2&f6uR<h
zf8*;pVHnadkV;ic+q&Ahb$%WR)FX5Fhyjs8Fd65}hGT^Ebi`l!;%fxv0qGPl6<hZY
z`Pj$2*XLZhXd*iHpv8hlKp;ty2%}TVV$NtZCCf`LUc5>uU_PCpf{4oaMyt_qm&5&i
zw)cnJxV=MBS>iY(PytChVtBB}gRQ%qzw{isC^5zo1m089RYg%0I0Rwj7hjR){JVem
z6+ZvUZ7!`?x~+n$tca51#i~S;!&W$FQN|ER<rkiMItc14E67!jl#;x(yngczB?)=y
zP&kezDW^{fa5cpNZt*FeMx{o1*ES?n3Z3WZvS2n(5i;P^>9d&9H|z<e7HPv-dc=G_
zVK|wvKb>N5gjBV6ESZc(jE;`z^q0{3nAz7m0=`*G$m4WWp5?su&Q0DMO;~q=P#2y`
z{=}ivkP|A^003ki8Tz?>(jkOkv^S@nqy$08Y@G6?-+dJ|YvEEym`jd!C!9Td8Yk*7
z-GOS*U{Wsf;NVFz!#E%-CRA0I!^0t~%d1>{=6UAR5oPJU*{af1Wx?*@kl|#^-8;M7
zyEm-UNjahPu;8t&2kam05y$OXKG@fhto5wDRayD8Xc!>`932k%{ona2x9%Kr?xH|M
zIocR-{@`=YdadyqIt!qrA&@~0F@JI+3~)5fXoWsSJDR3^{d+f<>42iLjAj|LET`Y=
zo@kIf(cD)bI!*0GNl73CX_}(rgvoTu#^!0xT)4<=GWC02N{rD|Mag(FWuDF0I~p=f
z=l*;sD`Mq)iw^hqn9ZhidwnXcYejf}=rxUNQYnIZ{S1dAzH##=**s+pfvif89;-cV
z;p!Sd!)EM2SYKCJ?5ign&N+s&l#c2kWx)Pu$SbeE!>p7P+A^HX$qG#?X?rKmJxr-0
z>-GZbLtO<5g`-SUDit!F&RAR9sONCv4|*ZIs%TL#9!*HI8QVugrbPzA%K=A$L~G5#
z-X2+&5=Sw*1`7ySoQKAEBwOGgBs+Wiy!zfP#C*=O1EEWQ&q6F(2^JBc@b@gO#GPoF
zI5{4V!os3;fd|ip+`GR=OFClFB~XI#D5WY4K_q>Vh?-3AM3F@~x!}YJqZUd@4i0wN
z**au>?F_4Hr`cFLjX=`r3<#2hs?13zBX+j$^2Xb5FqwXp_wH_UG%suIhYARoPLHt0
zcLw_$Bfg=@c=(yI22tnli?ZO^>o4%qr+$WS|M72fb}%Ol!D7+6ruwc*R2@qDjjXWB
zAr`zfPa607nu*c6MmS0KkB*R0A4Cii(!8iqOLzc6y<ugISn|Cy$CMgj9D#uGXvAnT
zW%IP}>u7hoAS6k<iwGpPs>r8vMu!L7yMKrD)-8@^6AEo<2}@)oDiY-LdEII5=S<d-
zP6G+_UC6TT0M=Z%e3|)l#?JTNU`1C5=LqWPliHI}wenx1U?^RKgnGj1^HYd%+OI<y
z;3~oWJ3B0^brNDksF;p2jB&meV&wwOBbyYE5I9pqMTMgkwJ_GOwf%to{e7-nyv*9l
zS<aunh7giodj%)J=#uedpRJuce7pT!=11RVcX!NaUeI2u2op&l3#RF~?jSbD5l7t}
z>~LsfkkUs*+893dsUIWFCft4Zi<~;|A)mUe(RbEhOhpig8Uf|kp;Hdy>KfUTHzd^s
za)Xcol`-rNhsdapBcRfTtSD<JISU|6;{+~9I_nys0Kzy%R}}{b`^@J)0_(4@v2y0D
zS9<JuI&)i9%#Owk_xA{I-C%y_HsdU#tUNn$AV8M|c{-=l>3Y`Y6B_ZyWSj<JWnIXH
z7hZaa?^T+;8@E{&9xS7+cfXptU7(bA$!ZWvlh%H^g$PdjdX`X%qAb}vJVHc0KvGni
ztnva_7HZxx7LBVyf<*ZUAB$AJ&0$s)r1>6MUeN9LS=u<o%Gt9h6_WJ27*k?P!*qDW
z;odI1`!?zQd(6v%veLAT?-eTZoHEaeg2<N@YMonGQ<vGgFGX5a*M}r8e(d9X?`z-S
z@cs@x0miw7T7a__;am*}!6Wgc$JW}0>H5U@KFMjnpiu!z#`VLjkkaw&bDO;I;$`2F
zTS}w|kzwd@<03$aT8U27{)80h3bwX~R7C+OR##7fQ*7-H$@A1V5uU$HlC+tPk3b|`
zeC7pyTn5Bn|2kiK^|mLz4lHpT)9>{$WsWW@zi^B|U{OjEtGE^!a^yup(hvCIANmYZ
zI==DcKVmYSG3bNI3d*9W4{$*cNR$#7D?EjQgVIzqy|Pcf>O6u$2!&JuRyz;j3mg|u
zuXB2R)q9|TR3Sn&UW-K3;!@P`P$2}B6CBOvfI!HAZoiKKhSMpgs^}LbOY5ryVTi>c
z<Cu*L7qC+B?(O$@_wdL={j4Tvw@H$gr#aUaCnZjJgDIu25BQ?0Lsy!l)8+c}FQBWE
zyEkr9XibEptUN`8N08N#hG;w(?=in$tm{c9(Kv%5Ae0fg$uUk7#4*=izQpGFb;`nP
zlqebbZ7MxxAE$gj0CEAa$TP$4{e8k9BnlIjmsZK<6$g8JSXXlTRLREXd8Bk$ZHZK$
z>(@Vy6N;~V<uzXW<~9VD7O>n)=(byEJ@?ROl=sr4_xNH}Q^;6Do)-+dEBwq){R~Ir
z1KxS->nx=iB5>qsj;{0eQc6!f;lK((7%ITxYB|EEL18u0yi)o?ur(FVT9Qa{{nA+m
z{cf#ZDUhKr)O#^4Ro4TXRFXF+d1W}5ObLRBC~nc|^~g$1KF^TWvTPmwl_j521fpJ#
zOV2)wkb<}F-eGGzMmb+!TS;2PL4+;}oT(w~{-#kfK!yrKqbsB7_WFG6<1ce`c))jV
z-KAU8q7-H2&np_Q-$Q<-@OxPsQ&R^$6_%Wa08y&Y_v2XKF3?G0u3tV+JBb%C{UD5J
zY7o-%+|&?ki9kUB^RnV#JS7N2TJ1JTtIMn?$)^(nB&)*G9SkUSh7}Mm4Y+dcIiwW4
z_251Svnio-C}BxjZ33y#MOjCHVj*#>f&gi~6KIU4sx(V0t9<<BkF&G&fZGptY2lkJ
zi?XaiD6S^P_06|V`1*==8uOS`{-ZEnAW#A&eflX3x`<NZT*Z|uYy73Z{24A^IfvGk
zFj95lx)D8+Si|DaLi&TG+h0eh9=m({EH5o_dUF$;<=uC0GCUgc;<J|-40>1`@4WLS
z_qO&qoX+TULe8Aor1QoWPB?-vpufDz`s!v)BdrN!R7XDkp#?(t33Ucz9rqt>gA7RG
zgjToD{O|#NC#g!!d_Kk0+*(1X2m^()236-WsSEQ@+JEGznT0KN0V;uf0@iSGW0g<8
z^gK(09>#d8_yrIFj&sqq;qbK_Nvl7=1`)H#lwP;Xpg%xY72Es!RC$5iw6t0=Wl^xZ
zz02WrLRJ+-QnEZ45FU-ZaTk`(V2Mt*N0Ar)UTcbGhxckEs;+PO&3iC9A_x>ooS@=}
zayBOvg34%$q9AgfAy(CqlW;zwk-{%#^)%@h2P*IrFJ(ZYI!yEkAr03*agm?-`A;)g
z>3i<4C_>7h*7p{^fz&qZ&>wz*QVhBq=t40b`qbC@`Z{Sk<E^*eLx_qvgkHbP@Mw?s
zZoS8yy9X4;khEHyIkV22uWb{k2qz8e>u2b8x}@nGX9Y@xps3S!f<Q`~RA}P`%-%b=
zPph3U7z}VCA<a_S9YLNKq<M-^F+xEQDQb@QnrPAQ!{7<2+?x^9i&q8;8N^<87srR6
zInSp){5)|S*6XmY1BzonM+5Oj_~NWk5y|p0g@t*Rar)FLdc7X=*^CE!J0vn7l!`bC
zIojW6|KO0l@u-F?CoK2-?2N}AJ!LgZYwHAIM3Lv-(ITkbwdT9ax`9wr6eYuAk9Hi>
z?R2rBq%xjTrYtKj+3T!7{F*Lmd`(tRr&*t>4+I(s3=sy%Ai|UxiIjZ&xo7z3^Uo3}
z;pZ@lPMn2aY|PaV{3gQ?#1YBT5~(&|EEm?+=yuyo$78nlcj+Z<k}xDxioNX}4v$9c
zjfbc}(Cv0v>U24rO=~2iVP)+UQYeZ%_s_3pukdWX(!+9{=fFzyoHS2qM<LyAmr8jy
zV27Y6%DR`!AqBp_PI_T9D<oKlk&5>-B7C0|mM7pMa0D_&x|lPYOZ>&peTwIvy9iQH
z>cSH!HV7hWGBn{Idy_*ILU8Kz1wQ@bKhNR8|9~?&E6dBAKYxKU=dX~@Q_vYgIHtoP
zQIznJk9>rK;gIp@fKVFNR$@le9A_2h&%VIg>M8H>HpNm`H}DAWeKv+94%+lutK56A
zO`aB9x%wP$4sTOcIR?vQGNP&qtaSu6%D`IR9jcThaJ53+Qw)da_mP38MnMK_ED!k9
zhn{79W!b|{SvYqdAASNZTq|uX4#1VoO)h`*C0_s9HxNL#*JJ(EDONT%JbF!QFqUkd
zlC%;otSmE|&B!KW7*uouiPZ)XMy#E^K+<aaG*$iZ77V8~*_dx;6UZngYR4QN9^nwI
zZ=7N{o?x{B16ew!V+J(}#nWDR)Q4Cg^PZMLS2%%nnm`7`G9e`3()CmP_+R`a>!*DL
zXBSaN&93bHI$Ruth=pN!=F*4x$cKKKuYTos2m{I5$|_ecKg;=ZS1Iy2Qfjm*m`%ql
zEw8b%a+bs4A!&9*dgc`Ul>ix7V$tU6`4?#=Ey}X;3X`rb+MCAk`E+C`6;aZq+wF4y
z-aX<l;_TT=jHV-6iKQwlrqeN<-YU-3@}W+i@OmZ6@l*w1@IVG%qg03>LWqE?XE%BI
zL(kIfwrld!y5`XUU+_6}zx3Q<FH0#n|J-xz?rd}WwKoZakbZB#%Bc-D&Yi^+71CNv
zSyJQ$gOz1^t1C<=6UubP%4|v^{aO#&T~3|9fKn1?>t<8j0zBessv^(K5E8drgi3ID
zcu1?=X0WzOet(+;>3PGl)JObyu#L6evv$IIS{T_hdp_M9HUJlb2vuobxOA3}T)RRm
ziR;Eu_t4dw{j}h^tLq_B2`;|)0=s*A?B2Xh5+(He1A0r#oH>6EQ)-m6RCz&Jl&r0<
z(OX$!Hl3iel%>%T@vgtGu)D<i*|Wz8x)XC~h4Wln!Z%w>AxSzN0t~~0A^rY<_R@f2
ze@K9@s}x1PsAmU30F8R5a30NO9$kZv8W0DqBqEPP%E}?6q6J)AQ9N^I2_*$~VK}^-
z?F;60zEsHtjI#-UqbT8rf9z+8gysG_UuC&BpdEK;_1Dn-jQP<nMVa9$-<)=KZJG0@
zH#ry{@V&RbOA-d?vf|Xnv;4$Q{&{-+CCV~?C|63IE_1^94rm}rBD#YC&$U|Iv<2tS
zU*Pp`wkfI<sT{NUh@!|bHLZZFLpWC>dRQoyJn3-6QHZSM;@)BxaIrmLt=n5HsAj%}
zHH*FpXpD+D@dIJ2#q%$Jk_bh*wM{#26DJ+Qb_-pU<l`|`YlHzM6zlx~tNkS=oep<O
zU;NEX&f4Y$p8e2=h~fxS`SeqBfcQLGqu&Ca;W!{}wF%Z0t|(YpU1KkJ52G~_N0H6F
z#A_X`NRJz=f$>i}fBs$VPJ){Ic@jvvk>T32Z8kQ0b#7bKxqAoFGcRLn(_14+d{NPF
z;(m99Kl{^vm8E`wvAc9yZMv-iOWh7dF=d|ap-lnS(h394p5NsBxea#qw>cOcl0=fC
zs(9h0kMZ(He}phR?h*2azu925*qQMu%&66(+g%|HJ#_f==6QB@U&C|_RiK%sBcfK{
zs~1{Z*ME=eQPc<d(~MRWp^dBEXhC1ZTpg^@Pg?a_tG#5XSr42~x72HJQB1ERlHSTP
zpZJNNq#w3Oj3tT^+MOPqFr-Q|ipdyL8H9ychAgkHP%bTVbhyvK!9D>#I(+8(3tV{S
zYK@pX(Pt%_l$mz~9NtlkqKHnXjY5!(C-hcU*uB3^Oidk=&XIBK8wbsaexQ0!*W2ic
zhY-*TqnforAdoD}HdmKU(TO5x%u?+HVl`IOP<(NGjRxK55UiZq;D>(lv)oL+&PuD}
z=dd@RB?Z+yqnJ+7l|dqDsfbgn>*Px-9PaOO|G|SroOb;qFR^+4JOW2Z=xbh08mpN@
zXX<D^B1&T4-)CWVbVz&9XSRPx2Z`1dMV2lIQEh!5Hb0_%KJTByM-7PZ)K;bvU?j>p
z&Pz@E!4AXM-lBhM%~PBR@7ee$pb2*C)YXa9ft23zw%RRz<fr~3^B1p@xPmO5B1FJ=
z?*Ue8l!}Q&$YglHu$!=P_B;WQPG<;fxOn+GKmS+%E?2K#M>t0uCkSyYcGGNP>%8~k
ztVJkA7(|3YK$65>d98A^48*aZ$aC^6tBc2WQIz-0xW}Hu(+uNk(HRTM3RY|vb>Nuq
zOo`g@aqmYx>5D&~NFg+fzRByhx*e{6;$`&CE|HT|xi_No;mDt%VTe>A^TPu=af{BN
zk3dis6%xV5#ml_>6F<T7>Z<2l3WM6?tP8n#FWNST*8iWl{|dG|$?rR{PiFq@@#ng_
z<$ZNkyJ<Jxz><Wa=aNfO+#okrt58U>(u+bHQs_l`P=ta{j96`OS0Pc$rL<HmK>`Fo
zEC^r*<7dz_)6=%RyWSmd{x&o9kpFY5W~Kp5*9^d%sED5Ks$2K|PMV+jnkw^^o=1q&
z=qhbVh9yT3V2>r_0jOZCb}=>*?yEm*3~eD0+8tV76r73^%B|ascOSEI`32V|)5m{V
zb&#t$%wiDidI!g%cFRla{QS@VZL;(}zVDG{Q-nzv4|kkwg9uPoa&)vqr`w>>?1FI@
zR1hd`T=_VE<yZd!%d2ZxQ=rt3?n7hkLd6Gg5NF@>sn_aG-L=%rW{~IL`;a7a^4vkA
ztzD=REW|yXqV%)2Uo_P|w1TiytQ9Sa&6M=Qocem}DNmfL^5n^9cl-<YFJ*NxE}dB8
z)t~u1`k;rvI&|PXAstLyJJ|~W#i-Y(5yu2kh%OCzRuK3BmtTCIS3mbz;(85RmdD%z
zYoAmgX7PAZ?SD_9JVg|SNGT~yNhYAIW{p{v;a6=wC#}!%{BW_JWgl9g6atJuC_!BM
ztmGld<2m`6lyIqbJaajIFx+4E16`QKe{r@0lC`sEXh#vI=c@U#syOoT451WC23Tzw
z9QCM0F}|lL3Qd_8gi*}(H(uqXPrObL25!hNZPkrg*ebY0P+D~?h$@xBQ$A4`A`q06
zq(F#ssLWE=jah&|sA4D)=tcZ~D25-}!pifKqA+-6!YOmaswc2B&4Vw0o6gyjth{i;
z@ja^4MM)1)eJRJKSp0H&G}7}Bo<|rh5e6@j?BAi^-{)|96D0(3+#o9wjs|^3;}L2H
zyf9)g9FUhepZ)CTdGV9K$d%_`buXh{=wP`anb8XjM8#*YmO|%Fjnxkv8yG0cg0wKC
z0({?66q+oZAJ09Mlo%;lcs(d3oczN_+;jpcv_Tt*mld2y1G+M#oWtR*Lu#EG@k*nD
zADSnF^NKxtOeCuaT?<5vkdmO)B=CHSy#pqL9;R@J2``8!@{HMRN|MYO^m`8NH68<&
z7higb^Dn>1(#iFw<QA&Z7Y`#HgB9gDNFUazp(=rr-m$=%(VD_qO2D&Da-+;EB9(aZ
z@GcG?h-#+v5u+#|&onh{u=a>kwSA(%GTk}k@gIGi#_}3*cMX9<hj^YsC=X%WX;}~o
zRn^jzN>+0cMJ;mTomq09!=nxIBq53-JR3P3tl^N+XvEI$Hf!2293BybA;0p=e}y+*
z{}oo(&N}j#kVxSKDTQ)s$qS!&qupm)ni9`*4@f3md#@-<($tV=5Ji?O%_*|PeJ$l;
zrBo$rqusw#&-e&ur9ond63xlHMN3MiM>&T#4+t9(Uc+V4)>x!h4SuC-eH`z>1yNWm
zkd}&gc?s<RIYGw0`Q!&j4EFaaa!nM5Sm9G7b7u1?$#hD;*Q3#DG98bp)nY#VxzBO+
z^;fC2o0!rZKU%VCF_{HgN&upo!&!%bl9jT7wFYf8nKhJ1w6zp@iS)|~%3Cq0g*o=h
zP(nO&B9oUI4?}HcIbC$A3B_nPp?Bwqldqma2Fjt`taS;s^qxv&M0F=VX|oEcke(tw
zu}bL&q(?*a?k;6s5=AjqK%S+{=W~iYV>sv&L?QFZn0BYj=byXA<yT)Kti`UCy!dS7
zLO#Wys#dfWRHR@>7+kcvoWg)1$DnaStwmN~1^5dkZb8Lu5!GCT>U=+0ACAJZ60giS
zU2O228dC_E3BhRVfQ{e(3XQc@g4GprT_T07Sh5w%^Qi-?q=!%f>4yYy4Noad7?Ms7
z5lKe9TVvcC5(uAqqfTR<5d}3SqdD{0l>YFD&hk2+`{XavUO9~}GAA{Hstk;(9;xqD
zpO*_@sY|R_7+AT>#9e+Rd1;wjsK*X#p63;hrFvYIl31%Kg?59eir?~4m$d?!YSt4t
zOM`-v9L@A_PVd$s>u-dJP+_e;zQ}oMkiM8hI4N)tzN&l}BvvVkQHV^O+(DXTc)myA
z2SjPcbTT2$a`G%?IvvwmTj9)0FA+8xn4)AMvrh%f6VkQ41a5O#TwA~j_c<Py;D9!U
z!n%zY;A9ULEi9;#ZXw*@O)NzCo~?f%3`jkqQ@_W_&YWrDYy=K(-ehO}1Q$R1OGwW@
zKDbg;Mp)K;CkrHusxC$8MTAk6!D>a8@8Jb0Qh3ZJbGq#%;z-i0cNh%%Ovf|o%_)QF
zfOF?w;`!%(jv#EHbyiuFD(q9nRi5uxh%Z+)lx2z4PFZ=uld;yKwe#~%GiVxTc;*bN
zo`?!cYmt?gh?G@p_!*Wp5X?$0q#<iGn28i?EQ5z58r>t7UtDJ)z$YuYisQpjRpM?R
zy8kbHXVoC8>QWC!ct(Jfq-9B?*+dviTyHX;%}KKiZ!SraDW|Vr<<yNEDBp7s_ll`q
zG2h&fxvC-@365hfnUYWT(PHMATSLhrrp|_d#Vn$_RE=0PFvX8tem`~$HPlkg<+M&$
z)R>W?OGE$eh+3yk=jswv-^XIFJzc@cssa}7prHbV@A>Y(w@dQT6kmEyEK--$n+-(n
z6E_;9M=5Eg4xQyG7hZmz6IU<0N1)X1-l!@!TOrRp<sEl1tTmLjtYn46aYbaT^ZRfQ
zh^zRtE~`hU0s=raN2xkYi|)u%>qEp8p;imH@nV<m!w2{im{xx<M{xB1BeuTwN1XeM
zKL>&5K3vb0L{^to`$M>g#P=M8)d(iTDam}wbT);g$LX%3v$}@PL8y?Zb%KtT#5+59
zHAQQAg^m4fx@(tMT0c)&6z)=%3Mt$xZ;Ue(lFCV_2;2-r;nZL&1^()XcEW1fSY|nN
zv_Y3Sd6uDdffSC&WsG|?l$2PxK!be54F}W$pR;QnT00qj;T|Lbl*TaHm=G@y=v-Z?
zoE;UiGMlj#-dzYBEMK|DMQh7^mXK#TS(c#YGnT!8dZU4{0uclRtv0c-_=6#}dPq=<
z&{$e)Ys8(lJ7CAzg#f9JTLx0PV%3fRUFKnB1wDl0{;JDnAOmW~8O|&mBV|P>bdnzO
z7()0=a}W!i<vK5(3b02xo^~xA45YK1-M79=^Tb(}ue|IYSQJWD68Y7oj&<L^@O%Uc
zooPn%A+y<xc{;@pr*yhaYOOkDzK;|&YEg$)BO?xXSzcQwiW?XsSX;S5t-eB*C0Hw+
z=e2OjhO1cZF{s03o(5~>F{))z@hw`VWhqGJE?X>hPF@r)7pxu)YaQuRm9k<sKT8NL
z%k>)9ymjK80%=_qtT7as;pp~&Xr)1Xvf+|3<sm%v1Q1{ZRtX0ZkP<5;MPbO2DOsA5
zrwL(}(rUJdypZv5hKwSDFr=2IOhzMGOC9GE<au;Xo^mt0BCogw?w)WLpgC^aEiNn9
zf)Ul{C>PtXY7=To0a^8Dby+x`gLV7RRCm^~B#U^qeQ1FbttQv>GVwu0kd;;2Y^FK7
z)gxSr374Y<gw;8SF7)jd?ZB#qYK0)nOOjbenx+)VoLZUDY1i<hnAtQz)ndGcM>L-i
zPNp=M+GO)NG7RXPT1SOGWnMgm+^p;w%#&oaq7!0~$9^iuT>!C|+ZhThInEeN7o|&L
z<>DrA0e1#j;)!$RQ|m__5TgwjF0Jx6e(^8x__qqSzxXv&$+6@{ks7wY@?B~tS6F)W
z6^!t}3y?x#WnSsvdw4;B#bbA;&j;_m$Gh*n%l7UTgVB(Sr@H(v{{GL>ZMW&Hb}<Q*
z^D+5&M7>$3b8?CCWK5#r<oX55Jb#K$P)SDH1>?P_T4^q;Ug$?1GirI#s#(Y~%yY=K
zQ>=8htNMo!Wd<oN%6m2;Kw~Ut*Oz$t^7HI}>k*R&17vjwEEb&!dN+IcO^@hQ8{<io
zUp;raf*>ka=9U(Qy&)T0TkLLaF&Ym^(uA{}E<gLi3&e3mqglgB4?R!ONk&wUXf*50
zW^;@uiI=+7C0_~5h^n$OV9*uXg5xUT>6y|Z=8JQ%U=C(jN)~pvPl@1FvlFb7<CgNn
zH{PEbLv6VJiA((a&%DYzztv}S?*qrw#?2S9g8>`g{4zneOMT@6#&`%nL<);A&KE*e
zE|o=U+1c9U{deEt)}34I9~>~9&G^)3F7o&O;TMQc#H_5YU^7Xc4Jq>(-IZn9oeukl
z2MF1w(O7f9jRoI8DwJ{$zDuC1`B?>dG{@w>rxXUNjMQ3-HkM=tWeJ6LY64c8DtWEE
zM$kz0^i1v}E(<SRJI9NsuJZVccgPMBq;M@CBP`j_a`^56>orNRRCj8tzUR!Ov`ZkQ
zbTIQQFWEml;PIo!?Cowdo{Y%zoLA4Dr}@GQwA&%ARs(=?nqu;jR=Y_M1RNb55VV^F
z&AJ0_REfr79$+h8hy$;<Y<(d@=zs*r+wjSGv|zC0IHIL4ooMEwRVAfxlmcs<&GECe
zQE?UZYcE{lQtc!gUw%M7EKn6G&|paqHAn9su=e^1{8sEPVBbNpMQL5;uM|Q_W?8}B
z{yrOzA9Jv`$9Ou%Xw9cDU7`Kl4Qh=x%|-)COPME78rn;3q>_wh6XKOFUL2wGl0}xe
zXf3)&^(i$0`2>-;kd>lpOI4#esLpT>RZFXMU?{7O&-=nV-F0JrG=@L=fLKGLQDgPg
zY058rlG)>jB)2#4QIr_8l%)L$TVMJH!P**8v+GDoQX)j;aQ~hMR&n>{Lw@tOf0M6$
z`z!1p>{3>TuT0J`pMR3Uu*Y(1ji_0pNHbIo>NFURhU^{e(Of#s(&{P4Minm8TMVb1
zbcJ<VT!0X+hhLQX7|r5BRF%nyO1MpuN=aTqS$UuvW2<?MKx&JW2G7T75Xq0e?X0!b
z;+V!#i~PAIioukuSE7Wh7^{|Slry+>$Vw|Dtk;iATp?X6P%1@fERQyKc>C?|^7!F>
zrnAZMr5=?fd0sG^PH8l|_`amfa+J`7fya0<W;E>6IC+A&(X6r|iKxT}-EHqA7EJX6
z%Hsjt4=J-kwfV&HAvVI95NYF_;}(!bhahsu4F$(-kdL~owY0h|y6fjS_3_V=kM@{u
zIx)!wKw<p&F+1P;24_FJOs%$5LAV7LrzfU7k0Ld^{hfFC&ENekzW@IB8IFd>7t-nT
z0cnyj8Vy-)uAu5ZNirp9`9x`r-Mt-l4-Z*C@hWlLb{Srk1XLy2G1$ylQSm@rYoIKh
zg6&hIV4e^|okUKd4S8WHOHEPcc!78_P!<L&Ey@=NtDdP6252@LG*7lzzTRbjoFP*;
zx6!!Ice*`eaQlem*8*tB;~u$HZofz=nWY7{Zr|mtx87uDdxIpMAAkLetu}d<k<4b)
zTOEW1ty6p-v=$s49x$GbS-o)$F9@GJven*jTd<lNRh>?^_e`blW}a#xAD42+NnvSl
z_mZh>Ui|=xYte}6m#X&cv&=w*5VTrNn(J-4SKAzYzlY4-1*?UV&lx|O5_WvLFRi-=
zL%77#`W35RNygKZ_ix_fowvWs{@ylOo>f1Q<i*t!BuRo&0kwJygr(FeVIWB|%fbF0
z^E~0?nTyB1^oxYqsTjio1Ys5(gz5}0xLk`CsH-Zfb6kDCrE!9lW#z$wj4NP3rR;01
zBNkRVeNV4HCcSo-bnV0%>tFaZai@i{?(#R*B9YAQZ?pT2Z<3D&&O}Oh2rs~k>ZqW`
z-tK_k{H@>QH-6{eu)ViMr`w|4tP@pAsP$$YZQ-!r!+q`%VN_!>opJll9d-|UbXL#P
zs5hLN=t7d}xGH>-$!TLTR#dVUi;Vb*>3mg(0nYQiwX{TGDpZ3mD#o~UzeQBO*X9ZQ
z@0nJ0-^R-|+E-TaLl3zSi^C#O%=c2p_Ycu&?pj4|a*qss@D&HW0pI)1n|$#8J4~lz
zn)N!7A5_Yn5XX)IJDp9j2xQ=+0-sruu)DX*I8A7*t`Szg>QX$(<`zGerScMDA;e~@
z!KJD8<x^#N^`|`PFy*C?l*S!&yBMMaSSO`^9Q$V(fZ#r3p(O1ymsx$`4g4rT$wkXt
zptPoc>wS6;zK<!>s#I498RCa^l#F<I=Mlg58~=(w{Kl6V&&DjTbg9=v!a(9hf@ZVD
zG?~*M4#6sdAV#Tx!^0l8@7-fE&RJbLPaKCRPa%|eY7bU1G}d`qX!n?C?fS{?BBZKB
zswh;KXAt_t^)^}N_H3!2AOMlJVh`)5kVwxILKZ@Vjk9f97dv=DRr}C_uuw`y55`P4
z2iOXBY@NvtD)i7&@aXX--}>fPxqtr_MVZlTGzdMPKq!0xQ5-W%X3Xb_Lq3K9(o+nE
zLw5Fd$g#ARR|qP@E@2l1Z`ESP?K{I0;JiE6=1C0gla|V2JNbd{JV=a_JTZDv9adkt
zf}=iN1wK;@5lBzayx5_Bx{i99oCt_aF}yn>*&VaM#9Jj1frpHI3TwH0{{i3n*4NnB
zc!0HrMy*buJVGh)l_ZQJrn3oYmO1cG5MrdD*YC5pzlZUC8r>y)<vrnhz;Wlm!m*sn
z;+H-~mM&uGezUDA5iEpZhzvvLJ!76E-lD2h+(T4#e?FuG_~DkeRsY>sXkWX=`p37K
ze({^6c{SsZAa%*%Tko>^+N-Rbzk*QCUREieqy0YbzxO_Ge(zfpI^((Pmssg`xcUBV
ziqbHe%$dwGUViBe+nbxbed~K9^Esp8fW!U~g)Y%}y#4-t+DqqIKe0@y%j02@gLyl|
z)dE7SjiA((W>0n5nW|R`j@(#_YrOXAE3B-nar3QjlBCYG#ah?i(S;L%E3=$ZYdkM}
z=8Y%dm>gI?+o5?hBfEd(hW|LyqbM!IhkaV-S7>zGSh-NHl?;YsHa0fd+<J(>a_aO+
ze5Kgk-A7xPXE}v2EU$Fv_xe0~&?nDxlIe``bmUrPO7UoChp^eAUXLAu4R;!!vLC4g
z^T0@#-7hj`cd#pnZxz>C6m$8~W$N`h+js9(gJSz68`lP17$=cp%`vI;nPPaHc>1hf
ze}n$k7U{e1Q5IER?<qtvXZQW@aN_bSG}~(k;h{VqVFdfzd%X45_j&jBdxVkCwHsH_
zWzK`U4=GB+us5UBinaAswjOWrcx#h1oiQBrIU4q{LQrJz?%Q|y#3wp5nl*~DK;!!J
z3&7B#Eodzixf2t#VzCcp^%%QDtfXXhWr<IJ>Qgl8F?a8Mk37v?(qcWgKTPFklV@0K
z@%;E%-{7k42qN(4Jh#lWHznUIs)WJKzRbL&fA^5?#S?_}nghOg3au@NM+5HPyUX6*
zHgOno=JY9&`HcN}g0Y5inxF!oRxP4;c);%7HhG#c9gmpICkP>!CxS;iJ6yhWiO};X
zaSn)1xwz_hR{!Aiq#SG%yQmx-(#|e`8bZ?Tbg9+rNCW+?oeIdYxGbF%jV>K2SRj!8
zvounl#L!xcihY)DtS}wU$@_Wrxmtu0l;eW_gF`wO*6`!V4a=1x&o%o8eeU18L%(-G
zy%uxk<VgmDAxV<B3MNSiYcXEnb9lHzzOz9(PnnFz%+o1SN{*%@>U(>fKYbQYEhJtR
zT%jNOyoG~rYZi%4Ree+(4{+j~YpbiQuC0>KQj)>Ynb`5<EbG$wm*`T1aFRzq{EGbX
zMh(_dmIW$~IQ`i-=sntCeDfg{wFh5Gvi_K(TX(ti`JX4MO2qke#KXIH*xuYkh=R|4
z=5?Nb;Rav+@>dy669A5S6aLx%=eIdH7}8m8lVuhBbT&mrF%%`Y?`-g`-*}S`Zf@}R
z|Kq>SrAuc}(!tTJt#pI@041E1yz7BF_cWzMMa*+brjAuBIeB88*Is#*lcz86JFCka
z?r&F0st!q%rYYGxsr(EK)<99_&oTmbA;h7Xyx3#u+A`DLgseY3?nx>sNyi1#g9*!5
z9di^Qoy|Dd*=IBu5J<tr=dQ7|)aCsTJ|Ha$q!9F{Q@-@w@A3TQON4<(Q50lpMp0xg
z>nTb$4)@u7aG%{?pHII22FslmQrOC})NL#`S5R&cX<aERs%pYLE>5J-Rb!Hsl@*@5
zeuH|u$#;F9yfPb-c40hZD2np93X~KTI_6p8=N{WqmqeXqPQU&sdfVG1`+XMkN#9dU
z_I4QUZnA#<Wv3iqE&cuh_wU~3;Bc2lJK~i$K24+1;7ecpGD+&JQ19N_=KuNM{u@61
znU4{KKJzqn@-{_|3L4C3Dc^bX7XSL;_xT%t<*)Orzxo$hUh86&lW(^Or1J2D=N@3K
zDN0RQ8k9%H;ZRtsj;mo$d0fB#9G5O!!4GQu>tFYnC0R8D7A|&WL6I%A-T>53q$y@`
zP}q1mX6f1zli`dsbDj^bYLq002}yrS^W^gJyl~VTu)DR(bTX#hsPX)b>u95S=iNOD
zqftWgU~iw_{l+)AeD*9-;v5~51Y<Q`6p&;YHy_;N+m9ab*-w9-*Ixb@^;%e=*EmK6
zA)TC#vtl*qYIY%U!+R>P5SKT2T)udT_0y**jON>4{VG|W6Y0?HW39>a!U?dJ+C4yU
zoLzs^<=SelDy*HTv2?l3Y&>^jtkt=YRx;b2lZ+GMm6lWCEe*q?BldUpna`#yw_CjU
z;`5BhV;*mAp{>Caikll7G{5~V&YU;_7_uxUOH!=RsK{rQq`ZIgK5uP2=JTKZ0x#XT
zLFjuRo|?M}j$JDkCSZ=(OVyH;3j_iHffsP&x$B%bb(-;H%=f<iE%H1=ilFi!(iFMV
z&sww*ja-nepR7O8>BpGTInS*wbLO+JakO&)lN_W&UwKwAd$dD38WS}-SYUs5gZ-Tc
zgn`c+Z@kRU{KDsX_q{vp?;hd%3QssF=vUr+mv?X7qh61Q!hj$QiJ}N4B?r9`+xvZn
zlNozQ|C0XTh=2GG{%3sbW7iSFM_^E%hm{J8azbWyjwq=HY8F3k(P*u>9-m%FnR_Uy
z=(LykrC<45eC5mkp8jx1vmUvCD@CabCoE<RqR>vY);`<D19u>#gs2m;bh*oXl47;X
zvLtTfrM-kQD^Y>_@J2_6%!YlUAmq7Am$?4?bKJahm+8!z&?p4CE_vs{UG8sf5>}>7
zp65CL5(LB9jNxoTS`<va`BmoADWCn+r#QL3R#}u*h!-ibLOMx{qC75pjS%>8OjxT|
zDOiQZP)^>Z+3NDKPke%Vw{J2_=G3b1T?m2EC1p|B<yP|M&msf}72?K}SR}187dZXu
zt4zQ6W%rqst}Q3ig7MA<x+n;On7qi@-n~zMv`eSc<Y#~G(|qF7pW(Ov?e8)jPl+NA
zPr`Vb@^63RYkc)9-y^Pt1VKO;goJT~(T3gKK70E^k|gJ!{fqyOLBGd;^mqOSXV0Ia
zL?cyYX@El-mU)IYDMH|cGPR@@H)+({IM8dQ*QOMnQk*?~fnWWl|A_B>|62^l6IO$E
zg%mX8ML|*ISnFC`d6qu&9y~UMl9uM_Cf)Nb_U{a<fomlnpO(yz611^MUs7Z#!~H$7
z>6CV}#rlgca^cElzWR-Ckmm)y@=!>|^C@3@`z_wOeUrfR@T$ES1Od7<^rmBm(+P#q
zB=b4**^Ezr{0-Wzru(}pG@DYDi)B$_@*J^frAi2Eb^JI!zWk(G&;%4KCr|O}r$5E^
zqeslsj9L^sAkAX1UI|5_7ytasLBwk9TB_&UjP@rCHl|e@$;pM}Gfmo?V=q{QCmbJZ
zZ=WJfSZ;T@cJ>6TCr|RFFMk!SHG%X{LeiTI`QmrJO+Ae8E03K;4D+I(HyJaUP0`jc
zpU+6<GhYAL$Eeq8?vaoS6%Ge<q9`jSxjmjANA((B6dnVK952dI9nN35LZM5J_IDA5
zON16~sAXCC(+lD5R{di<AkK_Lx-D-?Pyx%=uW|B)`y77zeRs(q38cqtYnS1}hb*1D
zh($8!_1Qnz!S}kn`mt9yeeyKldGj4smOC8m@1qOnwOp1Z!^xcS)D7gcF{GuF6@Gel
zJDtz?oj>?pJYP|({ij^Neib3K8{|q4YpVvJwMYfV`rrk0S5Fhg4a&-FWuZk=l+LT+
z@`bBdZTRZfzsP7bsRGS0X{szstS!j0`44_WA7M!$og{oUq;H++(%PMn?Bo@OL*faa
z<S-!}&WV?rm{KsE&X|mcsG!dI3m2%>ni#EV#u3x$?6_<#jA51~%rjS<YilW;fb$Qm
zc{1azx4(<$DL((%&#}6^jIiY~R2RDd8#u_R@f5=MsdYOj-#_kwE=Y}KSzxr`)ai3r
z3-@o|B1yAjbj@M_q{|XpmgpjXruczt`3PaKXp|qZe&ZDm@7^Qd+Hh?S;|Ptz$M;Dm
zJsRy*OjgkE_c-kBQ?GY<?#3l*VGSvv+ifx$jTZO;(yXN4pCE=)usE1%QK_(0O9;XK
z;UT~Fo4-LAh5Vhr^S4-CUjgZ6XT~aouGr7Q5%NqK5CnCWyQlC~NU7aER25GpD>4-D
z>dUWUjo~|QeT!NxI7YKcB|tcTaVRK?lxOK<iRyE!T-Wd-pXDnnO!pG(*!kB7NTg9r
zw#VosCu-HubHjKtW;z+ss5iK9W*Gx`LQxL`(lm2EEW&XY^n70B8SdgQ_~F)y;HY=V
z*S_%;f<W=9Pkf3-Ep~$=iGz$+19kU33#_N`YcaJ>+nsm)1OaC)6h)3w9#^hj$7;*Q
z!v~JhznD!_{hYFNNKI2b+XLd}0WOBdSp0gx%9SOOqa2$Uw6gfZLzJ4y_L%jT465N%
z3`-{CF|%1hx7+37`E`bqDWUHZ2LXBJqNIVeERNN=El^rR`9mjJ2*K|DHedPLmkE5I
z*I#*+Fz{W6KnPH>l5weGC<R(5!e*UXyM?V<ui)-EV+=)+6Zj$5Zaj~%mi?_Qigxkj
zv98*QMNzmU`-dLk4;i9Y7l{;ZpjLg6sM+T9$6q2^app6|x@@!@jyd|^7SdSc;X@vM
z{rA|~c*OqxF2aUPXE`sw{3_4C@VvA7#6j$|fzp8E$+}+%Q5icu@gWGq0An;?{l-`M
z?ce!bc6axkV4pFS2Z%Fq5z0fV5EazuET2FrpL8}o_PY?m_4KklV=^656b0wcU*(mT
zKLJ9Krj8s$)gE)1Us?6?pJm%EEG`#6izH~qbS`vIVFg*Npp4}tXSzE?SqVD>4sSkU
z(CaaqO|f`rYq@ag3Tta8+<>n7a|T>LUJ8s9$EGh%PUy*H<NH2Ene)M|_j%{N_nFRS
zEXZS2cyOnoBN3j52tsPDmV2bK{1}bp+BA;%md@u`Ygs>en$zdbW2`+^^E_4~T|hC-
zvsWKdIeS=Z9h<n(U1$B}ml1L3_F)O0QY42v47MH;c#sd@;?b>dv9+_obTYwY0)fZJ
z-}n@#&zy1{IfsWrYtUL^j6@rWse-7~4>FhIIASyz@$bL%2YmU<Ut>0%Ii!klnW?T&
zYf>nr@bSF{?e-eFES%cGQ!y;dk}OLZPsb?f@#6K5aqY_UWNAU3<<1$=b?zNcq<qRQ
z_@l(H`ky;bRRyFJwUs9Aa}BK2$1@x$CF$Xuq(8$qlJxd2hxfPW9UV}VC4~k*in(_E
zI_*xUYTdZKsD;Id%8EjWO38qVg!tr}4g#OiXu$X0`W`oL-*(1j6=c~ipwHkCL%xSs
zi>Wmm=%RGcRrj$~?e&r(&q?Mpl;?B)@)cIs*U5|G7?7}da2G_wWAxlJETJR|gE#4U
zG}c=*PeoX<XwSNzb-p(vo2K{$%-`GMaATL=;l6{Omxe~G#ntQ2QLn|vF%+U2ehVk%
zR(XsaKctHo3LywQpQFPAzWdg9x%1$G6GW}%z3xtN38YX8;d=y)8gac&k>!r(QGMmb
z{Z?cdX_^p5H7;Fwj`q?rS)Nla?kD*~X2z)m{9s%@yaKTXDLgzGQrZkD6rQK(oIT6(
z^Oxv-^8>WCcvyU+ncTaJ-28y%&fByP9*_+zX=2&mTBhC#xpL(?^+ty@EBWBPcQ`y6
zRH%yz3c2{P7H!VOVY3!tVR^a3Y&PSYfAmeRUb@Uu`}2g68wwlYENra`Kq#j;D?En1
zJ~k_<FSRKaKfG{#(4*cSUQlOgd7Ud)Z;&Ko%F<TlS|titkqMu5@@6Y3ccFZ2nIn|K
z6O#IBliJ#tWNRTmZBeCQzB5E^4XNMj5gjFzGG)$~@u*2$3psP{99|GmTEpJ%F0*;c
zLTJxE4acbd@#*|eBB(bS%x6>X-o4F<)m3g>y@FhLRtW1ht&mvbReDx}*=UA95Y)m7
zV_wOb7{g>VK!yR0W|LEA&!UT*!dT=}91=Hk6Cc?J>B&CS_%cA@gS9B3@B*Ku^H&+3
zy+!)qVbyXG$YRcD`zBFwh9&zZq244-LJp61SlMk;Z#8)Bm5(EZk2R7<4<9fb&W?2-
z#Ul3=Pa~e5>_g9&bUI7y9qjRqZ+?>t=g;u^CttxDrx<LJu8(gd*h*teTMm1BM7~eZ
zXgRc+`*0}Bg585HlnCf{mU#JvS5dbOd0rCul{}BNl_RG6UOw_O?4UVL4Z32OOXcB9
zkJjlH^W6k9Fc=*CTF(v1=8({@(|RyQ_EL%<Wi}czZq|wGHLl%w9*sr3^%jQ*hh(~_
zhT!5@a1WkB`7QpPwfNGbR*yJ5+~<RvAF#Hx%=xpY-I=fUA<8kKtpJpw)RyUJ<l5Vz
zI%b;#ltsa?cZ5(LwML!um#=WJ@z}W=iiJR9HKSNSzn*1Tr!z+-6f31i5KCHTT4aX>
zX1Z|0l<09uwml}Cd$cwZ#2}>%b0)(vqk4_?(`(##@n!O|<ktJ|)9dxorF}B}whIVL
zRn0$jmaQ>FQH%&AySv-mymgb+ZkP47Rp(=dt>We&w^oDj6nRlH8_kI8HSh(-l*d0<
zp63jDM;M^isB!Vi6$X2|RR_jCQJ9b)4u<&<Pjo>sonYrP{2(Mr-Rq`=KzcqaFF#MQ
zeMI(f8zBsV6fBzot%qNuVG}-avd++I4!d0*+`fhH3#<_ognays&$GX`%kF-!s%$ux
z|8}rXaf|?1tg<YL<B(Rf&R}rBJMX^B^=p^7a_tftu+~9{t@N-+M&luSRjX`yX@&mQ
z7^EQd13b^;@bHk$?Jau!gq77EQc6anDV|Z(!hoVEsv)p*x+|V~wSVj-z$7`Mv?$MW
zx=B@4Cj*c6`3`0>z>G^25>E-5<DBT;J}N1>94t}z0prNBx4nm_1f?+;A-Qz*I^*Gp
z$t*ekypDCrtY&c)KE1Togh4>yd5lMW9&c>0zPiTJa=U7Wxm(k910}P`gwbe>F@{#F
z&FFAM7zX&Bldu?$hLrstv!bBhXabfrFNlOhRjn&G#N(JN@{yMil(QiU=T%{B1~`~&
z5Z76M=~Z?I19URN4<u1tvE14w(%&ZN7JTaY2;q0xp9t=}{{dd0Nas0?sLjV;`!sv|
zd+6Te80jOWIELFU4($)_LruHWB#M1Db{_HWJ0EcI(m6WIOBEUh2gQ~?LcspPKD#?R
z7;QPRc7n}=dn|RAh@udSWovtjvfN}k%4jz`l%-}qO-YgjK}b;)4w7CvPDxolo1>nY
z<tQX-!K12%u6{FO=}MQuJ3~xv+&&bB*4~uB9<d|?uBl~$PD#I*v%R^CQj%Gc((JBq
z@$wZ0g8?Q>kI8JZ;vqQT<5LQ|RBxHlG#eeVH05w_mxqrxSY2Ksh@&UJR|t?!S8O!s
zV+svI(pE4&8q)1{2?E!ZI6ORnevf%x5XLb@QIO^Zc~Rg=NvW%Z&_02te`Hl-Ythpj
zPkP6)Ayid%)Q#v|Y%}^`iZK?2#K+RyoDhs8Ykr+8VwJ{{<zPBuXKNo%N&M!FwbN&r
zPNob716LVR>1J8s+S^Y(fIkpJYguYHNt1;A-5nlpZqn_v-AvfnN;uKYHp{YLaCqeA
zutH%y7#|E-T3W*Q-Cei8w}%`ISV=R2FeFVA^4xLEq~LfMZ;t5_f65F*D9PyIBX+*_
zRq8L_Kq&#X1mVoygQYHKKKBM&qY+Ut<LXQ2c<IxZX{j7r!0DwnFZK_)+iBAqPTAbt
zMC*e3@-nr0n`W!!{vH;%cu|o8oqU7&p)97=ZqaCT=(K8>qGUMg^Web-=PsN_dJa8g
zq=xDPZnav}Yc<z9mV!t3@3FkR!pib0c~P*j@fZt)ev^8wL7o?+NlF>T<j0J6L0KBo
ztfa_4dgdpDV7fD8^00?ot}xq1S414}I}zPW9fmi1M8L_l4yRWe1ckv$#cJGSoEp5k
zVlv;SchpChC1JBh6ov%VCFG#OREdkNdYnJnd#=@Df-s^MgcxI)PR1M_9ntN!T*k6c
z&{ggtt<`G|4O2P<RPXSB<>eJT->1lPjt&o<#CBXqDW9S&D2o!~1#ZJt?Ss--O6@p0
zA9V>KIk@*WgU7GpMGcCQYCa&q3L5L@Sbyaej^6ne?Gpjdef$d7KY5x!rIgC&^0@~6
zQHH9W;AZbJ8;>57WC_jo8qG$-iPqUFsa%kojm5e?%nw|CQc7CwHnq6H>dF$P&>S4@
zv+-zycBfmxBP%&i>7hK2ZnujmOLS?N&F5@xJtU4oJgJyYXYA~45r#3H#tI5Wk|s_Q
zD=#U$sx@XRVbiR{6y}+TaDzMhWEZCnnW$Zx!|7dF>g#pPuw?ddOgjoWd#c09Rt=RK
zJkRG$)FsVl*tW;+QIEs@BU))n)M-+y)$n|WoRU=q=N=87_-nhU0&ezGYcvR>m{z@s
zv6f!1Pp>!N)R|MqD$Wb<2~R1SjV7jazB#k$l)=#vjd}wiC9~O_@n}dC*NB^KrzV@_
zpsgG9E+DQ#Qks&|Jb|u%lx4J~f9rtq^d;xGX<W-!N`;kz=Gg{jSdwkeS!%|dJJn$=
ziV>wD2tv+<U5X+lZD@Mq5eEl8jig4s+ooP`99Ni6zialXcFfb?vgdi!DmT$)t%0?M
z!=oOf;e_SYrQ^|pttvU^*V1swhY*a1Lq>x>?N%Eh1=God>1;}^(WGAa(&uT4G_K`W
zx#MCG6k1c5#jW$y_w>U76GBo9N8J7Of6aOCFXLS}2gW(nNl&56lFkd)XqHn>XS-a!
zev+tGbF+fHB&^lA+6ma-eN4j-*g4oF3IjqvAjNRz%o$df@6bEyJA|lkWqMW=$4vFb
zL5ZS>7hkwRtGP@)GDKcPJ#Nvve~08`MC0m95Y!#ESvt5{5C-5$;z08F(FU8F_n9RV
zg1~1y8gcXHZBDJ9Wc{@+(k489c!&9X%1ZMEtf{$|RvR}oDl3nPkJ>twf?}Mp`{q45
zudJck0iqK3LkKqzztE;$SXML3a@!q7V=QP(5CyDfF{433q!i;ppTH0Blm~%eWqF1E
zQI8}|j|YZMAh7(A>LUz8&YU@e=SRfS5jK54WW2dg5w-D7b}*hgevlC0d!CbrmX_VU
zJqEo)1kmX?ld7G)Jx;8ipj)q#rwK><2N+Y*J$wGReP!VZA@Lb9XCVZWy~o`8%CE8V
z$zMd(ma3oAwPLh|6R&)nwbha<mo4iTSMdE1T{wHwR=dTOlLzd5a7ffp>>qAZiz0lb
zuwKCV^XE9&+apU-sxky2CE65~{ppj{?R0qI`RA#5F=5kE^K1A@a`^Zyipw#wegSKp
zsH3Qikb*FPKw`>*hc|Drv$sWDtI=w<*x%ph-ktlr@ca$d*4vB+Jsv!Mi0?^GpKh_x
z9CQ0nSA@D}i{W&8%+9y(QGdRJtW~0p6&cb<%kqt7R)ohHKcLx+QOZMWH-l-`YMhv8
z`d*2LVc72x1|dq$@q}P?WrgW@;x24!tA2Tf`1+x%rrmD4eW-jQDe!|BjA3|ZhjqUJ
z%eAVFy7=BDf$xD-ltszT-X6pL5%q?XOF23^qSxzl<<cd(wHp25ki(-xYGKT3vvpiS
zE>vf=dA3=m5Q6c8KKq?VMAz!5&_7nbwZ<X?!O9CRqd!xx(WplVsgC_a+Vuvj6HUM6
zA#}m8-zV?_l$;R;KFiBXBuV00jn-DWcTXfpo?a_UORTS-!dD)l6!?BbnHEfL@3JH#
z=!Dp+dRYK5e0&0F$+DcIy**~LDXn&!vMks;JY+tb@mzPAW*BjBbjWbfr_*dhv+a;)
zRB97uVfpZb-_wWBKqQPeH+lSff57Um{1TnBXO9(j%d#LgDc4>;MSJd4Ve~XX2A-1=
zu$HBEofoG=4(2`&#v>ly{s2>!G}|rKJ1e~Y>MLw+Z2^L~(O^6t@$lhebZ)8DBZPGp
zi&9GbpiaBBOtW4P2Mt2MNpF9X{<r@f{{DUB%U?jQT?D17HVznFFrSRMefu3AZ$9R|
zn|Ju#fB$><zQ=nX+~L-(yL|lhSE<!vrqiCwpiPN3P!yUxcP%WtXsvzJO@|}k&Gsjl
zn}*hlCr};lSkuU8O<cm-a+_#gAS{^NI7#qo<F&$wvx#OXB}daKhvg1NpCE`rR=R6s
z=Tb(a(ea}>n@u_B^-)3)_`XA_lqEt4R1i|HHHd{Fj2alBnDmFtzjvS7nKi`K6Nqk%
zPz#YYKpW<>8N1sX^!q)wc6KUgq~`I~7Q@k)D;F=47diQS;x2XL`sq#<;R*Zoneg|*
zRinK}ACQF+s~`J0{LWIP4lF3kl;y6+`O6om2NS@clR47!oWG2xSYNO4!f3?K5Oy|?
zcu<y{DJ~G!>s&o|kvyH#@AdJ60Of}q93C*=+{F((_W&46nr0{^sRu1O?IwPspc5~V
z&q@vs9x>TGq_*@jqWcMiT0NE<2M0gy^$)mp?|t?T`n>z@O_Zk?^ap(BJ8zRD2_OIX
zD~x7`Op_tCD0YXnggn<|X?YC6{jdS%kFybt){O2Sk$X9<=hvz>q?_%Tykse?vAo_T
zNL|Haa%W{=N{#Y7R{WT=vx1%x>`n$)14}E*1W~}5ljpFdW;Ra<LZ=kn9}MV^Mg+d+
z)VGZ#FAAhm_)(2|7!q1T-0U(-GRE6`q!WpEaSg7nAnL&}LdY5m#xUyl*xq`~EJ?Zl
z_z^)6(i;xA^WYx7r?_<fJjrB2F`uEOvv5T!*ZMP0?ZamrqU*AxfBTTSpV7Lqj2HOF
zY;con)`AvGCqum4IB$1T;hjp2?|Yo^>P%(@hlQXwIsy%y<t{-KaQ^H?6b{4_#35ig
z>h;N|Qv%;dDT&q&NFkI*P^(k-J$wtbW|!$?&gk(D*$C8?6~x6ZM6Qi$Y*kg4n*PB7
zyW5+jWx?GC_lavUhy5O#n_IM-EzYc;B%Msi=LuyLyHUJ#a!<M%s2}UQJ+=NA7_R`U
znBLi>Y`;nEuXHJ+8de*8TXJH&#d4HmJqbcm&S#h`$BRR9WAOZd&=@|Z<{U1yxwALn
z{qKF7rS(%Z8q0*C&u2dMIX?N>&m+AWfAIVN6Wz`U!Z=3joauDJpx<XO7}IT^V0CSY
zY<@^t7>fBEGaF+P4G-VOGYRIEzW}G6d&0a#aCo@O{fBp)&jyP3-g_To3}F<JCJAYp
z5=9YF6l3O)BpF3fI(d^)R}aUJ$`}9GOE?XqWG99CfbfL?;|FdaTUgq&oVu=tg+icH
zjnq%HX_V&?Xvsy>WEvOr<})_$->2E_;zu!2Q0LSOXSsCkDtTG*-aGG5Z}$nqfTGCB
z(~R+CLa8<NMvGRxMn0XOEe?X2rPwlu2ZyMlgy&DeO0yzex`xEC-{a_DA1{h1la#HE
zO~4X{AzEuUAF|*Ffh&KDoFE9EY`O)k_Zfp|Er?t*_~1Q6yh8lir}2sc1Vk#~WPM00
z&M`uPkrde!&-hNDFS7(eNGlld$tymiUW>i`L*5-8vbu7LK*g-Iy1en)Cwc9)PtrRY
z@dsc0T~4nb5yv&=$&4(Wv$MNLEv&K9U8T7cGMXMz=7z~|OuRJ5<QZc6Z9G52G{1lp
zryODm=aap;{g}<2P1=n$cBTj1dvF(kdcAp!x{^W?hcU)39xRU{uOxY@&*Pb6C<L}p
zj2}#q?U3+tfT-HEVQT44HDL)>34{TimX$q%>tFamKy#+J7Oap&Fz!v5j)t_l%Shj6
zsk_RHA3M#t%a=GfI^w<Wzs*XP5C#EhGAGXqdPhCNs7ABZrsfIK$rNoY^La{HNbGQq
zxVeYY8hw2onxQKSg`g-)_IGy~4F=S^OHB6n>2GWR5QY(XS-PRQl=wmDAjFly*+L41
z3YGXw5fs&Zm`O%=CwQ%Z;B0`Cq!Gk3OSiNHB^XbDu3{K?$L&_n3s@QlTnpBCFA$9O
z`%H!d8r>zNCs|ut=j_FEoH}=o&8=;2zW*-kr6%w_(z%P_(a{lctxmJmBF2zR6H0BE
z&*$huVtaE$x{uNpb9Dt9dkEw9O_t~E?QS!jjHxdzGu+%_YikQ*Eq>sW7iFcq;o*lq
zrLJOn?8j0K_J#P7_xD2|5GQ9L@r~f%?OQZ2ohABY3u8)F%#7A#k7Q~vSw_%pqXJ3R
z8zL%h1KP6OZBm<#@I^sKF7d6+f{h3FP|-GcA)})nH7huE<ti7>p5yA}Yh1c~g@c0w
zk|gEuXpir{`Aw9Zql*dKTl>_!HtBH4WpzkcbRh6igI(-*e}^K=;Ph3ju@p(dox8Wu
zWy$A0_p=DEhUa_iZa+duk2nA5yCi9jH=mKF8I#GBgZ)ERR#q#z{!frOS8EY;sTG6A
z1EQ4(b=GCItrVhZL2ex3$FKQF31#9Q;RO_^8Ae2zKw81wdc=d_4hOq?c%cWLM={R`
ztWRsD!}{7uE?l_8(#kS};Q&)=`n@AIA3Z{rIXYMLhZFn|it&_g=_BiaCy_@p?Ax2@
z7f!)a9jqnKGIn>i@jZ|0*I!^><akn&&SvCg$===`Sy5n(W2+`r9BcL3@nvT%KQPn!
zs7sg53xZ_M*7v_d>(p7i)0dG#bD}vUuv6y!0&6v4tAq3*n~ezTO^gOlc&x59Xm#cY
zBly<4OML(C4i6sQM)@_Q2q}{pp|o^YPVxMW8>}p^vAS}C?cHs#hPw~$a_|04N;PGg
zWbAGpkjI+I@G)0UX_W0CjKRz9P?QN}^B1te3RY{TlOZ?n-ekGE#OFWrvs}G&gD-vg
z_c1o-V86#;Fkm`QnavZjEM=ag^m+s0xZz;!a6G_#Rs^axZh&g$1xI%dsV&v0Eh~Hj
z?Fq;tCo2p>So~Umu!dru<3&E&I*4>PjHwk)RrYp0<ndsW-Pr~*@K8YrIY=Xknl&z7
zxJY+(g<379-|u5BY(0L=;P4PxWF(oPH=Gb<g7RoYT`fahA`PfV19VzqUOox255_VZ
zPuShvVr6NSD=)rCcWIfM@4tty6g$B_##quUBhNE8L{BqjNkTjDpU9&a`^->sENGm<
zcri{nx^qO_sSz{<p@Gg=fX+y(4uW6v!Dx!PMTO2UCh$F$!Wy+gA_K)O;jul~VKSW}
z1CJn#oZomM2;zV%m#?vUVh!ba42MIEH9Wq5pJX~k<{6VLV>F%<qy~;A)Pp9}1hNEm
zw~tN=><eoUIK}PpaLB>_9;Z&8=G@iisMQ<Xy>*j%9CQECBgetZQnEaA4F0?%NmF;0
zD{(Sw&4-@lKL*Ats9c6f#~GX7{64Ga&r$DoIHL@8=^>55lsQ>%LbTdJDv$JNfLE(w
z!hp_NKoon3gI(5YC2Rf)zxzmXyFXx*^q3y)lW#w!(QF~R%QQ}&;b424D2`cLUFGD7
zRbG1WB?i4dTaR}sCz}5FA+!Aru5}E>nn1(>s?^Lj?=U<(r1eYx2-RLAn~xdwj(F{b
zk8yr|m3MC5VtBa8^{ZFNGR?QY^(ONqV`+Jn$B!S=>-D*O;i5BcsTika;Uu>6Ja+~w
zA9>#`BpNK)v}EvLK(pH<R30lf##gw>SqWuo@M8~zp_pXIP+=9+8!?$DI62mIyog52
z<Adog2T6~t%*iKn^1}&193dJ3akt6fsE;4`G@1>TIvw=s(<JkRquvl*2-4vpj1rVu
z!pI!yhltXWZyu2*Q{vBDhbSPQC1mrI%jd7qiee7;_ejTM&YnHb(O|&#&NhYC1VO<5
z{ywGAoH>2wSo*+LgE(z8`WYF(#<+p16)@i1W%u3h(q3Mp9U9Iy!N*CD7kN%To)R=0
zU}t2b0m_dNQnI|h%w(i^@j{<dD|I^cRlfH2fX7E;vLdHQ4w)R>BZ`{H#wtrIXXtj9
ziEA~MR#&-j{tUvF%%)TNyFFx5u(kglp4msuYm|kL6cLz=Y41B^qm<^^zX2j(I_+bN
zj8DG)2^w|DQSUKHI^Yv;e1h-fZ?nI*N0AzYRouRPhbW4XFDm9q>cAXTm8X9q7PXtn
zSPhedl;Oi6QKLpf2Q1-uSJt!Wtf0sYp06-jK`|@vDvN<;yMda_xRSt1SR-ypZclgU
z&HC8M2s28VZujt_5ZMVB^aj-G4Z<j-*{ZXC;si<vX7iLZ$*`HAf3QPiW)Za(#>)|Y
zfXy_?%}p#h(W@6g2xg-(zVNtu=_;N!OpcB)X~xAH&odZ~m?sGuM|RuX+N9lCLR49|
z<LEdfr!LEiU-Qw{0#qx6YlV$==5#j4EL{z0>zIyk=t}DubY7w}jSLjUq;zu)rI1R{
zY}b(^xKT)!qdMPfDelg87)*!A@d!K07(eQvYCf_PGU|_rYc--cq~54=`s68&j*b{l
zXQcBClR|%Yo8=_HHVcebB7y*ul+54RK>C{CxplCX$#6^~ZgTnjWr{3ka(IYm;r!Y2
zY;Nz6<|$S>xbN2HChI3pVJ*&VPrI{RmQG#bhm-0LeLy6Za3QdVU~q4Tov;28uYUX*
z%jp!qf+~8kht3Pkq(t~3h=ejt5OIh$7Qa!)pET*@1%LHogOfGO|M#}$)+i*+6CRAm
z)W#!f2M5TF4fa>p=q#<E5h&&Kh00linT;6_2ON(2oEDaj2tcRkETu?&viS@>I6&>*
zgXR)aNKULQQKWO;`rez|ymOl}PsoxPNBuEnS#tmGeVVlft$LIBY{q0d<>sy1+_`s`
z@p#OlTI=_Ry!OdYeZ*}C3Wq##o*6LNO&HzpbAGu?t?-;lj}j>5V@iY37U5Y;UO4EY
za^OBc^vMIC7K%^SPtp+{-?WA!JtHr3_UC=V#KRv2h+dC+t4UO^Io0H<2d(o0o0KH;
z8S`|;DlKB^VT%HtJIs8MWZ3Nl*_&Zb`FK*%ZZt_}a~_TFv$MSkx+G1U97R!-?C<Zh
zwt9lNRwGGLX7d^2>6pF!{mSFJis4|u&zfKO$T9S!!1p9`YtV(};GK6_J9UDWU+mKI
z3Z(HJN6GgwMTwbfq#sh|V}#5Q%1Jm!^(N_T#+g;Y-~H7k)>h!xziip<7mU)3`ObiP
zI3)BA*xP!{(#kruMiX5+<cldZWj@2?6OIN0Ci5xR*PtG2j7iaDg2*jdIzpKDQN^dx
z!5N|;;KbSzz1}{LHtzG_(JegBFdX)nP9{vIGw$EJ$J*K&?N$q|4gG$fM~^po@bDp7
zmR08wHn(^9#lQ5cAAKJR>0IS-72V*$m{zmKn%5$-0n!sr_)i9v54lG82AgPS!X!Zo
zMNkhZ^OBRY#?QAd(z1&02+Kq#OpAnMk`he=yphk~phvUQ!4Cs8&hl4li%tqmnlqU?
zy|Fd1OsK)+ZXa4tP)_D}k9vs9r?4%bFbHVXoAeJ4nam~}?e5|Un2bi`S<ZAeWq)s<
zC<>{?bp$XP4H<1sI6OFfihE-@IO;u<x4efqw#hAWOYhEzW*pNln}i62=ODOBEmW)x
z(l?k?BLXUPo{OO#Y0i6fmRd1&D|tu2Os5Qs8MARhGz*bK#c;nzv)e)W3WK{#iqc?{
z0-Y3$XG25@b<xDvV2T2hyU(l`&xsxl5$9Ggp`;c@G@=>@yL${q1BOROc)+kfa9o;U
z&cVSEN-5&FhP9UdpwD19;HcND;NPw*b9i+0L$UdgB?Ks8$&6z=+tM=s;6AOB9x^uM
zrNIk4{5ZzqE}x)Vqwp0r&9JEx<@1AxTC+(uo{~;xoGDZO>M2RAOYTm5(jdf0I4W}N
zWJpmYWYZzOHmrCaeQnqpjhQD2Sk2{H$ZH+J`L1K-N(;Iu$mbc65O~8q@?ws#mchXR
zfAp=dAp9ot%&@h)!SHB6X%)tl93AcP!3UB&FAylc`K>qU4F_y&ZZS`OP<QCXS6~0g
zn+_^7dSkG{VhT&Pl@N)R%ZjW(c?#uKt%_2ke2JFUsoNT7l%oQlFp3<!Gc7n##=POJ
z@P1)Al8Q36XiBnjj-6R-S(48Z0_#zi5mP;5m`=#^9ED)r?{I;LPF0Ocs6-I8%jky*
zcGV$WMn`?NH#d-Rjd`9i8urN2f}}F18Vv^MqM#@p1O4`$JB(*j`h)(lEW=_w3ld8z
zM{p|(2L&$1Q__d;(0t)J$}~e5@WT-0`wo4T7pPi5E+nGJ!8l7(PleR#P0Bo@m`{2B
zj3NK5$5+0Wvwaw#iOB?v(+rvPu_9+Wtx*%2R?TC7ka0Mel4dz!V7YNB<dx?wD=U>1
zvbE0iEH9{sIYRGa%?YrUjmNinyzvM{oBm+T;qZWDme4;MVob^Q&IY4NpEB1ZY0j6v
z^i>WHdu(s-{zwd;Wi7l|jCCZlVp3urC&VjtbZIe2S7|5-xyEXco}{2elmG@J-AvAp
zLbSG+L~})Su`=O<vd75FG0KqHl39_W#u8C#l<yOp5Ks9Gvmw)LMqU)e(&wDF%tad!
zTQ{>57K*&UYC(_*#AuGKH^IR6gGY?UV?-P?nv9uEN94I?KAT}m&Edg5zVBlZOlLE`
z|IRz~$3upLVKq2^vYx@+2U)?BwY6Ab$%iF8&I#IvvaCqgzVrRCdF8|IIX;IhsT37l
zGU7)8WoZz(<$1Y=5E*xEpGn{h#S)WK76n1!q)~)YG;EBOl72Q|p3N!BlBSHf;H_}Z
zMuf(Fj~Gj#3xt*UbBh?K4&9~=4{zNj$#V!p2BQIanvkSUfK^Dr!R}6_oZ-<Q4A|b?
zb&t(>{3k}|Lmv<&M5Su#2vtHVmKq+7a!xVKvD%_N1==D*AHNwR0*88_(D=fshgw<U
zNr{O9w3O7^9<Q%8SdIj5ZBO|Auw*kaM1FvcYX~VB7lzORWx`I9GMFbce8pcn8S{F>
z(h3SX&4^~BiRU|E!7@*YLm%0DOmXx8RXa(eUgy^REpqL#cH%4upSuqpFi$P>yx^_(
zZt~Xqw~u?w#^@jH@m2lPXIU1U%I%_efFr9ll~0grOzB!&RvL&rgsO<b(p~1#%Bp&B
zLsY3G%2Q}ziM@y`;TlbR9_J%A^n|g6!1Ez++E$6q(Ks==K`|jIG8)R~75@aMO`XUX
z>VXrlltQ^aZ&4B|AF)5gEc<v-Ob`Za9UdUVn0l*?L9ny8OKvThDY?J#$f0=_&|UMx
z$Mq*$%9F>2ITokFy0+@!GohSj&QUSX2pV-~pIi$tLQ@n49s-5uShtan=X-!b70_J@
z`NC?<iRDAyzM1mg!-D?U64xVyP?T1Z&q|DENQ#`@{+#(d=j3X@FMrbK^17zkhIXq)
ztyaVLB)K+}MNXhBHor|FPXI7M&%O{v5>JJ^^Y(q}(K4-Omm)X3_uftBGfi4pzVWTM
za0;r`FRg#NF%&<zl|9d+jgL&UV?blDrNxUB!UO3GbOEI<@s*E|UL_(ZTs4M93Pn4p
z^D+N4OXG-#<(Q3fK!SkK4VVjQNp%he=0(CV8<XpjlYW!eypyczkXQ)nQH?MRP(nJ+
zkk+WuVz&-44N2trXd!s8vrWCzC5S@uvgG08N2EqmTFv`+Zn+p*`xDPDe|Rlavljy>
zi-WnhtU}DSdr(Dz)e;qW$SQ_q=~}pg3K69ZUL{^=gquq(MNMA!&a*nH^H7i2EC%FK
zLKM0-oN!W2(n^XvXPk}E+HfxD@`^f5r}U{wMJ=x32fibVYKzqdFVmD;JxnBs{gAXQ
zd9b}nb9ou5Jmy)-@X;d@Q(%N;a{GfP=a?TqhCk|w4oGPb!d<KtNQkADLg=K@S#hk=
zbT&bmkRYzRKq*O4mUuykv6iC9@Kk^_ntDB^v04KQwJ_#Pcfjj2!}oV*9F0<vJm*jg
z)P+k-@{C5B(cIf5=)}CZ8u4Ns;y@F}O*)+}^|<aJs#4&273^;~Kp#9poLOe)XvpJ1
zLYn1l4aW#!t3Igv8PmM@lOyvJo?-_mcwz`71vHhw=B4v%25c!PQjPQlURXI23TKQZ
zD@d)AW^qDuQ4}G4fw3`xAFz@%xnMIk(+Tq;Co4;)UWr@|NsEjq^N0&c5Gc<3E!J&>
zCoMq`P;b-;g0TAD<Z+f|Cpo4+#+(S~&8GB=6q6^Y*$8l`mc`a8^iQB}`ll@mDyTGm
zgxiNf&7-^OVNB{kFw&w;N;)@qapXh+D?eykYJ~DDLS$K0tcF&%g$N={SrSJPm(NXk
zWvk@ottsP4&OFH(m32gGo%uYWVW+gcgjTc8Yd1p9tr^0aPrY8F-CiaNqv~-}NaY?;
zGv7yhdz8g0k2VHu@8w9jhvyw(aP0EM`AoBC7n=NWmJ|#4s{0jsg1P{+P)i0(t|_cU
z`3qLQJbu)q6liUbS|1}E8}-;_*kw!@M4Zl+*(q1qPN$?gCn*ZjCfJ6g%u9l~q*3^^
z{D8~e5=+L%_bgEqQ*SgJi`rV0@~AX9F#QBOnNuji!E{K!oHOjDl^Ep$TIyC_{wZhq
zCt1R}yAX3MFB1Eb7;!8tXf5c>sZgsBq^FMC$5y&Io+%7Qx#~m;NuyCiD5%9Tu^(}^
zTxGKuv6oKCb;+#AD7%s(G*Cig63{Lpx?#*^wM^T1c#$BAVrumUN_tSW3$4+v`aYV$
z=_YBBvo{~nFXxO7vr5BFS1n7oHS%YCbU%Cs5{78!Nr@1O(6g+pEzvl8f@C^E1p<;B
zA%yb*_9Uj%#Ek|v%_xhkf)6@9d*cPzB&8@4WEkTqMd*i|ZDy>+O_H)nP;YT)d}^Ql
zCBDD24;%NX-%eN!6svWgC;&f@<T(g0s5ZF9TAwHknGHt>U7~xtB+WkWe{h?mI-Gg&
zXSl&WY*`S2!bs;&A`k>h(rh<~8X?&{K?$&hQ*L!u2<`&+<B(#SRd%RuknD#Fqrqk+
zWu_59CEf1%tZC@hLP`lL2$;m+zi^62N85ONbHZ{?%d63p9$vM<N+A(mWzA|d!jn#L
zF)xU;rR+~hY7twzyAD&2^N;`Y4x$WVw9K8sk$`$DVCCdG(Q=JsJ|c`gOq${a0WyxT
z6<jlJHp#{#3Y}M69*rM_*t|ezQ%pHWsfa-Nguci1OM-JJI%w;Ij2P>6<nm{D`-3~w
zvOCn%lC^HgQdi-JpnOG{3zwZ+i#7#*P{a3p(qvAvo>3$_93EP3-+oY$D%~vV&wYgo
zgGPwO<>%2_Zs7?_k-7`n<Ob!t%y~fw4C4TuXiO!KZLC3h9%zA?=U8PCzQXqdbWw2L
zZ*jt_BLWYj1lh?N<%Jd-TN^~1DY4bG{g_6z51pE}s(e1c8jTltDCtq;1zx5o2UB{g
zV6W=Cn}yTmpX(9`UtrM=90}B-fW}e-Wi(}$;Yo=tKm`iv39NC%-XQidnVWA~?H+LH
zdzf6KXF0;N;CYVrpq+%#DX)!+JaS>N7u!r%6#Kh7L=RJ9Dd+?>YA8HmK`JMsAY7+N
zmm1&qkwQ`yB{Gp@!zl;jBl^Pu;G7$O^2X?s^&@J$R%=>k&a?c&izsQ4)^cLC%Z0NS
zk%aX3`;=Kh&~6|@4`WND?;$*es?|`nh&;>P4`mD}4_T`rf)KPOnG6xa62>uDiLe^a
zBs5jYg-BDr`*j%i(BmN&x*^Y<U1IIz3bjU^JhNn(BClAQWuAjI_+CIVpJU4cF&#0T
z3^?fbk2k=d`?8kWi4O5<v&vea+iuWZUBd{E=`_Vyix(@vxr9m2q26Suz!S%)8*44X
zlgQ910a$G)=7|F__#Q$i&<2$k)U0GFlazOMVV+~>1?yph^+tnc)0Me}fl^BM^^HXr
z1>lsGi!29i5c8a5Hf1`W{rSi4IE0qv(>JKEo<=D*4>@z;G$&ThlFtmo!yy1+v*A>T
zg+X~f2!$Wloe*DT$YiuZcp)-~QC@&9GSYP7=6FGfumWUG;3q7#1Sgt3lEZIe%PF!P
zbL~o#i<g$@t}YXX5%aO8NEAh7hg=kSCBCVc&t?cp$cGF@eTJhyxrh8`y;>Jm37P>;
z-9oaqwoG$rg{(BprWrzlA1NobgWGE;ffxH2u~4yfS+)!m1eGF1T2dsAWT`yo4<zye
zKi4#6K;tMQzq^l33;fh@rq*Ju*`nU4p*)YoShNt7x^N~?Wr?a(2J$RNmKHP1m`=td
zKY<s@Pr8JZv@WjU#~u>!rQ*bi73!TO=4nBa<R~TaLW!*^a^VT^z>EE34~+#gUIr53
zJ7teDDbQJt=PRUK#4yLtEgcmRZO<qk_t5j4s1Tg3cUWmO9obqb676EBwW9-QT~^hc
zpvZGXVKMWZ$z(+K6Id$zN$W@D+>IezJHh#1{j1d1){$6Fu7uRW8H1hs*kpztPnjPc
zF+Mn8aQ`9G%^k{liYZH^A2FWgl%<o0KnVOeLim!X(IyUK^mvRXB+XU_VJ&%{kWVKR
z{R8a1?<2SGBeDtWCstWnS*6~LnNKJ5cKi52)fzHTlm%3*)%kqR^sq;qJ3Gtb&plIl
zvaGcP%QaSBIgeipQ9{rOebjuy{Gg90G-XnfO;VE4ocZB|beK5wmFtI)<mGXRZUN<c
zAi=N2c*@5lCBE;whu9c&mQy4dW;n+_?jw40R8i7ywWu{3_`XLv&zR0LlqrvUo!YoT
zur=1X$xSl?LbvhlPcv8fs4+A&PhaEgE5C>r#sq=n)P)%9_ZSZzA<4-mLuSJs<D)%>
z+Yg!dkI;FJYasfJhjSJQgpd*yh6t&s*E{%1P$m;<QH?lmV0A&BCFJP{J=;c%zK@-6
z5(b(RYpZmYmIx!CVSm81Kf#OjadqRoWF%#&nN6o;;~^1}qWDn*o<H9pLj8P~<rmH%
zJxQPxT~AUD`Xomqq_Gs!f^3>F@6VX;Pspbkrf@%xmXKscrOSjv4f+nX6*cQfBQU8Z
zierK(z?7OY%_(OY=3ojBhnS-op^&sY9rt)h!DKS0$TY&Z<i#uqf7J{ynJ4IZi7!G5
zqklxe@6UA!Xk1&SePs=VAo4vrQjqm~<ii=N6y(#KbdtIl9!x3bIkv2<ZZV`q;Rfus
zGIH_+(pQA_8Wa|rS!ej<dzhl6Of&Q}L+?$IkH+Z!oSNs;YPYD>V+1fB&(Wnp=o~8d
z!-cGo)&=uvg3Z7aK84ZG1Y-Gd){j;oRnXekpX2n;{S3`^#L3eQtd~-xGipuenXHox
zWhB1vNqR>ly(4s*B7~xs38tCbgbVRDFN!NDTZ^z($Ij+>2x`qXO8HovkF;NF(Y6AA
z{WP~#i)o%CQDpO!rR6$)?AWzMkw5WNw$9IdHku$t1N3Mg<#|v4M1Q`^0z%L@zeM}$
z3UMfD#}bpzFhx!f_(%(;Fi7R9h-^HknC4D2Pf8}HBscn)K`W%&bV3QdT7>5Z*wo-D
zkFe%qi!mStDvD?jVV7&{2B|YM5>Cgg6^96;5mGqNLv=W<b=JUnS|Fwcc9J;Jz5h_l
zT7>7bcKwsATz{4BT8-s%4RSj{N=@9TIrLMO;R_$3VK&?&pAOMQjxmaZgMz%MRGBO&
zFL1mU=@Z3m0_9_}DS;Od$L-1^#Ysl^ew~%3Pa2%#{()arY+#C<)l)6-G+0Gh6vt=A
zTGtjzrxQeefYoC}W!3s0YROR2ePNx(nHH_cqaH#ro8ln|0uK!6+~WBGXvhXL%6Z{{
zGEy)oGN-UxwQr=Ske;iOgL=&kZc~FFgamQr-L6ZN?^-jhpvi2-<50|;tcC?)EbUqd
zrUW6;<rA5EYoW*rinN450$FkFN%0?KS&LWmSbq5=wG%Pz*eCJ~#dL<R6rS$}*QT_n
zAi@-eY>-eUCBi@<VU%abQz%5`&#olW_Xry`q!oz55`+<c6uG^mHD2hUynt4%#c&1s
zW>yg+K_O^HJ`1<hh4KTqBb{eCI?>pp8I=0ZEQTL$X=^P6AtyigS$g|BWDUjroeABz
zg$fj5<Pbd8I=j+rFd*;u5LgIetTB{@K{Xr>-wF^>2!+JvC0-D@fqRw{MRi);6+G(#
zxW6}GI_j}<;WE=koA>U0mlvZ1KW-2=Bj&}Nnv_s#R3*6q0!qznJfWM+nD6f~P3M0m
z0sFC+wTM8{xqgCdHYW)K4wI64JwgcZ{h%s04W`f()0AS80G3jT%Aee2&eeP5<`%|)
z1{rt`wN+YzN<+z)9zsca(*gNR(>i&Ai743K9CE_O`1KgiljM2s^pK1L8v_naRa)|_
zAWjYGC?WqdDtmw2WsSkBcR2g%&tS>v4eFHXfL6PUkb<~b!>eMLrzz>|h;)49G<<SR
zsf|0>v2v0io<s(rYc&)WKdM2Qq05p+qk|BVFz~QKG1)kv%*HIAzrn_AOt5{IlO0Xm
zTt?MA=RFTtZIB8C&dssNbCT(dc6-El=#l0Bp<`(AIw31Bt~2o^0}tXvQ)@Z^MHohI
zj~PRmmz0w^`Pf+nEUaOvco4=uNo1^c$%+^Gjtgd>)@qO?DS_uxVwsH&5K%z$+$uZ$
z5vr%D<%+0P$I9{t6ljI0jMsp&&=gsTH_4gj6UskJu*Q$EthEGdF)J^uQEp8+6q=ef
zL=7Kd1yLL!s!FO%J<4h7V%Ur@)=*mQ9uB#fF}X*?YU`v<qJXj}07<RYq|9;xF92IG
z8;|fCHNtaUHn#_O6N}WAxKqby2f?-0dH#rss{(=|E8OQY63p%N&n$)?&O|pLtRUzt
zarq0Mqj7jZIM}A%mYDewUgRTu1y-W0C2qA581mtOaXuoOPO<d><pl^&IbcNz+A8<D
zD(7^itb>rY>TS|AVVY-@X`l7;H)yP`@v*=5x48SoCEot+|AE%EPqD2U)J#H9S4^`x
zp-!OYV@4yA#916F<uN=QFd2?n>2z3ITcO=uA_@cY%#q!84-Ppz>^+mh>5sb>GRR@O
z#`=vjgkyt0&IttA$<$>bmEn@GlAuwen9eDuDO0PR{bnp&wkIpJgpk+;5lTA2I9XT_
z5ZCHX%ArURNkZ$~5{*ua)307*^Xu=j{q8pY<<rb<N!_+kUPO`R1f>Db#|{iR;2V%i
zF`Xt9MM<mPq}gmztH&tmGVU}>8H`8FW`D+yw6zEg&9$>!dE*Pz$^)VxB5qiW=_CEX
z5yPxR`2n^15=k;3nH<qSEU>x&D^Pv_N;*_bA<)`GSZ72gmBgBYP<T*lk{21{*@PfY
zS!%Dde)=>QU;2Bz{RjV)yPNM2wLVMQFKIU|VIau$lvw2$;Zv4|`N)#R&iv|V_Xwj)
zPM%m{d3l*uyNy(mJWCi&rtI$SGagO<ECX0cgG-vHIyAhHV4w+u62H`tWR6H!X>BRr
zLpEyU(>d8VVV0*(zFbJ8?<C-<;%1bnTAw(RE^Y8t2(h3nN@lYOvC#yb25TqR>3r&C
z9{k2PnGZ7b<yEpQrQ<~?;Zr0Tk(OAg(8iEhK@0GO^XgGT(CIGGXf%jxvCHg>oLMqw
zFdULpgv&p(C^{w}?Xyd$S_`jV5=P)>IU*|^hXx1W@IoJnAfG0r<CJ-tI9`aTW|$7=
zTAARvvumYBNe`{9d(d%SCez87hLo6Aozv&e625$$NB`#AL|#aKs)bH{I(`jdB_b)1
z*27lZv(yT-braa}$O$vH8*LiRI&tih{5&t1&ZZ1T!)FE>eRu^@!PNwUPNTz$DQI>*
z>}-!Rv5xt>zzkP?U`OW4DAFldO%V7Fk)buifye%2nO<1~o743toQwz5B8$=vDmWLG
zquv2Qpg8x&>*Oacv$V3p<xl+#kG}X@Z0t_?{wU`3*^9h>{UQ%<ZZJ6-@W$nU^=y-)
zp=4zxVkW`2DS!WO{B<t9@FLIMxXzig=V>(SSZ$cjCOm%ph&SK-F8}`D{~m9C=ex}2
z$+PXOKV>-&aY>`rq+uWo3UqG>LOJ<^s$>VMLAt^e1v<}>*5Xxcb)#WiNRCj9(0E4E
zMX`b)^bEpUl<-K)In&XQ(Dzuqd<MN#qt$7#a_u~KZa(H{(&Le^oLXPu{CbP6%|ptx
z<b_&9yeF985&DwUq{K?{GoSt(C(oW@{p5OO<l}5)(`3%U!2$R0Kj7W>-(z!Yi{hE3
zHvWWVD^e?x<yM<+Xo=%GMX`a8lZY^tYO*sTbASq6PAKvO&lmWCbK}eB1*6A`!ET*-
z5+ePS*7BT{vk5`$VYS6qipe}<Fgl{M+@*Qp6UcatQ>QL)aOso0^M~)Sdw0aaZinYy
zzRc-MT^_uBA5WIN_C`)Vc!%DOWO+4Yl4qP)tMR}5&;K*7zwiQ=uUujM#5!RVQWgcH
z@sPXs?(s+8{v-b2_rJugTem5H%JY(A84d(W1;jO<IuOPg#a_R1+<eOW#AQ7w&qpT-
zWnSPZiSi^uSn>h}R&gLKGpmV=VHL%S5GW-o{*z*sPnb`~v{%~%&z&V~M68}U$<d{g
z?7n%2`@=o@zTn!KCQG%L2aiV7gyI!E5-XWXk7W;%Y|47G!(aa8zs%WlXIVdaf_A&@
zB<D-bbUI;sdyAX5Zu8E&@3ObQ_h;6y76|+xq*05h0e+HG>}PnMcicu_gb?dymSvKl
zixNNZ9Y{u4(!z3N1&7j-8cof@8iFNJ!N4sjRWh5+$mTPa&aEP^oFuG;oVa*~y|Z`e
zKRDp#a7G&Y+&JALlpY(KL)w18$I=>!6-<?f_AOa5<^1YNF1~VuvuDq+c4CcItL0?l
z%aZYE#OBr(AKbjjd+)!;(J}bw$6G(@2Li|VaH|bIpe(4s$0lypP)w$%M$P#ZWI0xA
zY-yaTXdnnAM4lqg3|qZbZkreB2P=#$c*@WTG>^v{ypZ4GM00`_RD<vd%~qF{lV^!v
zcm*L~`OI0?&Yz{&*Q}jf;pI1;qq$ON?V=@-G4Jp7(8HXaQgAEF`TE9~zHHF)N2mZ0
z?qW%kIrAjp%$ajseeMN*;TL|1fAA0f6aL`$ewTmt-~9`|`RzYCR`L9)F6}Nrfg=Q3
zT&qZ1jR;*<Y|6^t-QoWTgdi0HUwPyjHVAmwh?vHbq%82Ipl)Gll5?#n=}0GZW~`ye
zQW~``t<@#Glg{j@y}m@FS!XUhmX>OqJG)NgD>`c(4#yK7&k{}+8GVs(P!2Ge$J|<)
zSr;W7w*er_oJP~q>MCc?o#(}uKgQ=i|2f|N{@Z-@8((K*bMvP=tXqhGT8%FSO1KBX
zYJ*NQv;g796nTNkb8P9{)0C%(VkmNn?<<mi!RGguIXZlv!Yq@_4Sozq{V@kycewoW
zKFt+_z)<RfJWp5(&vN?ARaD~?r7k#i{yL94b>=e<FF4Jm7cL<^%gUMa>~9UYcXyL0
zE!Z3h9=xAWJe)I1Eho=R5k8f~g=I7z0g_7>uko>$UgMYk;;-@#{=t93Z~f+P@Gt*=
z|BBnU@BQ@R2Y|DM7RKVCP`H_-vI0|B5GBHQR<Tx_N;J=)q{IibbW$kOQgGYz*=+ly
zAtXgX>?t;K!>Lihbt8yyRs^~zFhxli*I7Ap64eT^#;|<u1jFw=py7u^C)=Dlxr))6
z?)nn_@r2Fc6yN0RGvpvUATJbYA~;b5NMVl2ZL`@F&-XcV?i`n{Ug7nRe}bLuEx!4U
zZ}RnTeuKka@24C?Try(>N_a?V+%Vh<Y-Z3{SK*er8fI5^#8nJE<$PF%h2Uo3vDcQQ
zfgmd^b<blX(VQD;t_q1?nWhzahP8&MUZ->6G^*x<fR%IWB=?VKMggm*mswrwl4k|&
z^$x@7jGfVh!sP5S;vm~42?S-LI91jVEXcl=>2yLA*SK)`GMArwj*oxplWabE%-6pD
zb-w-PA2Iq<80d@j$Id{U5q(Z54N<e{$ZT`p`F_mi=q$tf0mW=isWrkHlu(30guyV+
za}IWw`Fe7V`N<XHND&!Lt^*oZ){*)m@4Xu%M&Dp{+4&rog&V3VKSHN7CPz8TXRp&a
zvCg3XnCWE3J8$3N`txUKEwyQnQy$)Vhl2+Pd^=6q-^nO<CV0x@AO5GW^Gkp6uhCgv
zLP+@5x4+6i|L6advuDomna}(@pZeq%ID7sg|L8ya&-wJHKg~b+C;t`y?|<<xNRpot
zWU$JLjVTa9;l%+WGcIdZ0+W>vQ!j-RVKds1jFiOld~}(U6$Sey<efH{c0s5->IRIG
zbXGg$Ny+W|1D>BwY54B%)W(98NZ&`#Q!)i%yG2}&sTBp&`Han-Jx;E4X*6SIc}9QZ
zfWc&!`{fR^S%IDl!XV%mpTEJ2ue?SWg$Ur?@4v-&-~1Nq>+4**{sI@STw(q6X;xNN
zxpw6$zx}(v!*{;>-Dk6~{jd-kThTE!UgY72O@u!|mOi#nl*t709Ic_qQb%|bprl6-
z1Qdl~I!W36{t6FwZy?&M#IXla34V>$%U4Nz7kKpUm$>-)eQF)w`Eh6i!ov$9vZP0v
zG-$ThsI?nJW6iyX2gDzQoW5{^&RUm2Z^FGh8ypRXy!jwwa<fNS_E=t8;xGLC-{dd+
z{4W!P?h^eUfBk>p%U}N@H?BX=7e4cIyz<iPy!hfPT)A?IPk-i9{NsQ8Px#^&|L~{w
z5W8(?Fs8us15~V#wsiVU0`#mz1R7&3W#N(-DcrE!^E^@kX;JdNinza`k#z_?4^aw&
zn$OZwo1?=yx9<<RQ52Q#m2(k9NtExA4JQQ68ug_%fs({Za_^`|9D1yF+jKe&M$;La
z2fNJX8{9AU$w!8AWLa5W<}<(W^F05`OU@gk)O_=+Ut#m{1I}N($c>ji#+maMxboa{
zoIZ7i>(4#Mul>fabNAj)?;&<DW`i{az8@p4L6!x=6WGjRW*QMlXJb$}grke0bEnKP
z%nHNXo?vICK!uW6DP#a~GoagUu)RG)JREb)+KNo9ov5u+D5=Q$W1>!z#!3gzS|Tag
z-a8;vf^NIzVmO-c@Nk!Yw#no2kfLwUJ<HiMXZZOSUgE;lOGqinl7w%3<qsJR`dqwx
znHw*CjP=uJxbfnPoIQJvYge!G>%aML+1lRzQ)BbTsPS57D57)xe2ywJgq=}lDI)NZ
z#F%+n$(&eBI(O_}Ay7e#A8WEgv1u=oUp#>q2s$0b*$Z_#t0(ci7KhtIZr#7g?X7#X
zTf1(ESC-_3#!8?0-nO$FJ^wsmXANfG;l|n(rYdFg-ecw&>^!{B(ao=ObAN|@t?>j%
zTd=xz0u@C3)^GkB4tjeG20iZHy~}&=zfBNIzWmic<l}FAiof!g{|=w{#AmsB?JEDr
z|MGuFXQ{*g{lETi7*8fYon42mI2s^J0hz`W+Bw1n(&-KXrPkQOnLimog#n6^1j$ar
z!(Vr1RLhaaN;9Co)TOi@NBt>}8v*r^;D+xctBtXgRV9-ijuE9HUh1M6G2Z5w^PMH~
zLUXv^BPmNpdq>PRw%M8OF}EqoI^X9;yGdy+Z+`1rOvXcIvnku#+wAS`^6<fZZr=QW
zD_5`a>KmWr-1!TfJ$shF`B(osaa`l;-}w4ZrKM$Y8L=S8BSDG`Tg=hAK!zbg8f>mX
zdE}HBle>T#Nf^|Te!}!naWK4uIJu4#&^YJOI8kS9<vgT{L#5e!aE0j31{dF`V7Y}R
zFDwGZ?C2J8)FA9!K=~~q|9}fu*U`m@M|UPnh6Q`uAMj}NO>XUs85fp7xwhiD^Gl4!
zGydhj{O>t9I$$sy@Xot$v$6G<?|%0?eCaD+<a3|?8UETY|4nXOf1Xc%;uHK|{?Gp>
znym)E^_#!*6DaWhU=%F|Q5saHkp-Bv#2R<$DPys=nrG+|TRP6QkOD6Z5jvxXVt-jt
zTTx&Hr&|H7Mo436iFxK!Xz$00<-~F-6i-|LD-Qg8dq8ce>#Pa_MQfIGcCA4<P1&BM
zjOPiHtpkRekJ(EP$wYxyDzKKg*`PNV@Q1(m`;5l}CX)#d?>}HP9B}X6J>GlgZC-f!
z6<+(qr#P{8f|p)=iCV43|M;(ejd$OF|ECeY<Mya5EHcx`!eX-0wVgv*Ss^>!v9i*M
zbAqr6KMW{r$pM1?nxNKINDXJ(5%qfLJWl2XNjK;2j$mb4()EP%A2$YFbpYmD1DdB#
z;MGHnuq-78yIdz5&)J&H8BS-+HxB4+KW0BaLZeYeAq2#&Iy*Z%Jh*?C@p!~|Jm&uW
zdnB_d_wL^1-M8Q7rB`0#m5+a%rS3AHc>Uwl;yVB8*ME%%4<G)C;rU}`Aa2PbBbXmJ
z^C4eg@(j=SDa!%^9}xt|vP2t;)frMqieipuQYcTdEn6rRa3YM@+L-Y0@gbe%2_O5|
z6_R<1t;KBTD-1>tu(rVD&O02>!_MdEJV6&Zvbjcnu*>rBw+M8ERt+ZJ62qNMJ~-T9
zr*M1)E1dG>POry*_22y0=rTK=)0CyDhDw&L?GX>|ZSu}L-{<fB-G9Vi`>TJO)%DZ-
zZ~j;RYi6^AfBN74KaSO8f6}FLgEvmGrNUsa+Mqm_Nx5{{p<0S8t6<5J!f1S5lGg$T
zVM!ojVghysQ+E18I-`tJYs>DEu6gVU%lR~`pnRgT_!Zc+#7rF<TXdS3@7<?6PRM<a
zGVn=bLwdN&&UBZl&8R5B7%T^a9{=ut`fW^ER7_;2d1U}*(4WjXI5=eM@nb&yg`ef6
zkG;zB>MDQvmw$yU%lOtGefy`BMGJf}$5VT*btOtLB`ONY^8^tFU_FHLDYF?8Cl@hK
z3!>1F4{Hp{HY>g&kcPeYCL|vmaOS+=%!M^4Q5`I^zZtV~t%oi&rqm7|DI_MHpwj_{
zix|6tO};_f-yln4vaG>qRAb}l9yfOfj8Z^AUO_ECxP6cR>p%X#kryturg9rJ285Wf
zzct{&-3Pq){@eVczyCky3!nWNo_p?j{ty4<|DNf1#_#{Z7k{$Q9gj8yexi|s3C4Pe
zGIy<2B`GNpo|DiKxx{1{!dQ%O`b(jb94f^mELf%v7Vhnjk%tARQ$wp>Lt8^uS8N>=
zEIUm<Fx6Z@2<)uH&ND=<g|5e_TidMcjVObFG*XPzoXOrMJLw@Ag@b2T&(FO__j$bW
z0A0Ap)U|3W|M^);Z!}?hXP1q~kNDYN_={Y=dX3AMukauL?Z3l3{pZ}jfBz?oA+Ugp
z`E!jJPq9K_^8#rF(i4<&5ko*q$~<u}{Zdk9DM1ji>q8P2tlB1pDY?BjL`u#10>a8$
ze%4Uz_9rxqbigA`HFOu)X^Ba4L>N#8KH>cy%UcQ)_{?I(z>XO1ZLn7yIU!8fki}Yf
z@8$>Gy!AmPz2cHgU2TuyEa7N0VrOTY?afDg;TQe_XU?4G#TQ>ddXj(sFaCRWc6a~S
z(0sTWuN%AJB_sUdL#!z<Q5|eSo}?IU32Swbl4LqZNS~s}(D@8qXkz6d2$-gV-qt?;
zT9-3j#lbwFf0(lUXoM#PJ9|eo^MctdLomnW8c+EsKR~2CWHO=5Qe@O1nU|dNGulkq
z91Yo<A8>cK$6=|D2R92Z>!K(bj4D>>Q$N>Ys92VoG?_7*eL$A|zZj1v{LR1qcR73N
z6#v`*;=kbGgNOX)zx#JT!H%o`J}Xeu40$*NtFVFduP9566#_r-Fu2QEc#1MDF-49s
zmeBW^Aeon%aev5KtHW9lGtrj*Bq54Yid-|EPf#W&EeglH*5G+gmyE(-(-K{3MCenL
znq?y>(}F{pF&vNCpByr`Pf#oig(0mq^FKiiHY*MJG$nobn8JLCd6MwjYj3c;yuvU4
zrC;Ij@Q`~CezL<WaDtv5M@aD)q{I>ehHRDt9zj^6C^SZc5I$Kx#pr^vEUCwJW``l^
z)Uvm6z^QW`mbwMKxyR9?A)UCvY%pi9-=`gwr1KeBnqW&q;5%z&`Twu&&4TpI&ilUK
zdH47I_U^A6-RQ;+%uaDgjyMZPlq!*;*o-VYRZ&@$vf{*%FOo`9%58j`id|l=vRtIf
zN+es0Eh?6*$g*jPlt_`BjWYwxU<S)HmIl!Kx4!$?a`E<ng8@itdX#g40?^%GfA901
zbDr~@=fCI@Bj;Fc5u$}GN><k`MV9ixtudS1f)DQQGp`yO(~p{HRaH&(5v{9_>vQT<
zv0KdePW~FQ)J&%{{-gi!XL;iB$N822;lJnh?OR;E_9<d04*xI~cakG^OK^QSl(CVu
zM))3X;9-@<BCxKDs*9%Ipe<q0q7s6%EHMXTR{A}LUd$BO+MgrvNwbo9l~U-MQa5m5
zYrqRbFczma&E|*2_&&DM42(lr)XdY8k=<t}-KVDMK{kEpCQ`MO)E{{q`@r8UxztpX
zl<LyE)Kc@MJm=}BKf}4R=lJ=r{55{{*M61pWODQv0^oqMc%vLKQrN%;L}Om3G>-2z
zvg@^K{Ky>_S(aF15C%NgrxcLpB|?u`?)LE9m?;JK_UA13d(5(&v`#6Mq*4~Cni%?F
z&_Km&FolF(h;|*E%CM+h@<KAorX0u}c8d|}P^jdi-sL}xg>Gmd@72MS>dm*Pq~c3o
z{wXe;f1IbEdYZrTvwwx({O#Xt6wx0s9+^N)U7^wp#(Pj_;5*nl18W;7ngzT@ysLqs
z2@0u5(>X{(9QTPF#qnMT-D?pY3wY);%k)==<i;VLWxVm)+l*#gEb0_33+D4Nj^nbh
zw2I?+=sd;OHFZ^Bg~Kd2c*BH|>2t*>rW6!K^&^oh_6PjSTFZ1=ap~<V)YWe?pQrrN
zfBs)^=F}N}`S1K!eD~E?AJV@63l9NS8FZx(^8#wwKyj;P(l<s^+_w?^@SRnvrY;J=
zB@BE#0fV?h(W+>79WD$PN&4&LN>P*rSFYWlnx**CLDw}|Uf=~j?Lohh=cp9I)2O<}
z2tlDOek)<(J8YW?WhJQ(rPz-$ERa=_8@D&9q@u1WUU=b4oH()0mtXz~Ht*iK?`iSL
zV<<3nfv$2)wTF|`4fN6IM&Q5*tnjeHK?u`)zZGOfMOmf@<<M@m@k7J$r5+>SC0c89
z>ijXHg=NazGMi0#>(&*f`IKQ&P!u_n@th=z=?zxUx&l++m;$LwjB%MyHRsOtxw+wS
zbKj7aQ_52Qg&^UN`VONllf9JJUwxA@uPBO~zwvW_oflttfxr1Te}Vty@BhQbI`}BV
zf~_>Vl(@434t?s;4TRht&O@9Jb4{a8P{;~Z)d<JO6%gYx?DR2S!Qx`b$-w}>8<JHu
zv+0b@8+Xu|A<$6Cg1jn;lNQOKi>@T1HqHCc65}|O7M8m`Hf_l+Db-w3YxTgU_rv4&
z;P+{3nWP!-y?=$Ws*qL9=RW@&&p!JsFTD3f{^;L-?Ne35rqbwK;?2v(CRR2$Ra~0&
z>mJ6iQJX|#$m^o<+VMj?0f`f^*bNYF#j#<;@%{=<+a)V&#-lOU_ctMT@ii26K~+n7
z?KV-bjVkKHYcq+~4(d=TebDc+p=XRlMOD-d<NSy?G-WLZ(>WL4d6&GbkhSFLr=H>Y
zFFwz^@4mx#UwieDKld;S%XOLG-Q?i%TMYfGF-8(DP7ol3!of#~3SHNYZ?5MOwc4av
zN?mHotYkm+IM<nS!X1<OXPJ*?(CIMFD!%cxi`=_C!;gK=43Duey+~0^m`zjGjxEt&
z98%RKHct^ARC$h66-Eh+iP%yBdyZlI-VW+7L7Z+jueki~b?W-77_IobfBV1TrI%jf
z%P)V4ul}pwJJM?kAt?5yl)DoWZE-}ydlBB@9h>VmUJTNpjl%H+z6cPxi!zp4)nrA%
zFezA41MavXX(<tj%V<95o$Gg)W+g%3vk>*@*9Z6gr-y?9K`WuEiUy+|nuaZHG)A~+
zKVae)WOc!8mL1*d^CKfGYc}uP!|3l|4ZQplKgBc8Jj1E)oaE}YYe$PAf`hy7la79j
zR@n518rHrCrqN&0RgElbl&bN<fC$TU>QZHjd@h+yLryM_S(?o;#ToLk#_JEcw|T&&
zZ(b!ERV1A*#|JsB!I;Txg02;->nDhl9z{OK6BSlT%Cbajg{}m$jM<ty>}PWh_73>8
zH+9y+<RInkH!n5hO>6jtzy1sSCx7)T{EL72yS)GYl_S6Qd~+YUo;Um+(J1ZWIyk18
zmqs6wE|tXfU3?p0L?cgM)+KdSupny|OrLGxFs~$T+vWE54)1Mjk=KeO>M*o5-Fm`2
zOL45j!eD?K`czqttHDU!n59{Z!KDs-Cbs$hH2w52dShXdW?Z}e0Y)o?aQMO(zQ_wN
zyu>T7zs_j<$>6CU4r>hQt$p;u8jcm<G%Rv~K)B7}*&fDFYmFDU7!zO+)XGrG0v+cJ
zs}^Gx*t2sorEppfA8c-M{mw3x)WqFIqWzqfY|QNfH=sWlAbgj4p5vJ&hDtWqQr7V(
z1COaLn9fp);?r|sw1v@h&Xp@yuu9?L^5iqm@Y0Jfaq-=E$n%GJUk@h`$D!Qa<;EZV
z0m(~GGnr=Cc8FsvwkQ!&l1*l0;~6q90gDp^q;*AA8Du5cQ;HL+VzIu&T77~IvB>Lx
zeuM3MDeqsLF<4mO^wI+D&F`?B++v<i>Gj&It{kJ%23wXm*_6U@uvLv!CGA)=LWE(!
z)(3Z}(wyc-@90Aun(Y)t#l{D>`G5cL4_F=^<8S@q-{xn2`e*nz|MqL-#SzGILQw5z
z?7w}R<HI&tDTxD`CQaQi$*ZEKDr&TBuHYQkp;8)!Kns^t)(nl}*my>kEHVlWmpAUQ
zHOkqZR&={<dVY)6c$Z~eQkFUGRzj!M+_Bn9AxcH*mjp_Klthk$b_Jy?8BJ$srTNQp
zpQxl{YkQYBUVn{Vug~W{{~S*}`4rc$Uq4cXBOHg>-D}*w_;t=c)nP2>bO$XA5=U1U
zS(A>Z6!Q!vYlI^Zu1_|Xs0zxoIZ2;7RgrYxV$m+Mx0~?pH?DE#;+Xjz&C=Qm$3mz&
zUng0eGRt$0tq$q;I^?A!tW%Kl1~%HQ@jOMV?NdsRnS!l5yA7=UFN2V3A)V)3dG9K}
z^RIq~wPUNi^!&@b^uh~AP9Q=E(wk$(gDrd@&1>Q$#3+N1jg(GNRMe$x{!h>-L}-n+
z4i$!~Dp=Gt$9CqFC%f#<bG~=^HdnW%%xXBcI3O`T{pmg0x+Je_hJyv7Af(a~zmQN!
zs;WXLg{Kvv?@_syOcspCv%{wIr$_EtYsvD08@Fx~eDfPD3>H~mTj%WAGmPFiS^@!>
zUfXBZKfqTSr;tQ(fYAo$FmaSwg{)-bD(W;zrPdZL915#J7OX4D@^(gfqRriL#w@+b
zm3tE^T#hX)5*LEud`h4SO08I4T*h_X!x*}lQc~3tBOQEW!1E{^#k|g$PE(BKFC~V?
zT4rg^^^Fa}Z+?URV1ct|&#}I?_Q=EF4}kFscHX&0&sWgXynlH{Hwjtq2Dn1ucpm*;
z7o9|?x<=L#Wi>VqxSDy~%QVl)ie77%V_UyR;eUatJVRTz_{`~mSQqp^_+yUSw+P}4
zp&U-Eo!~=9u~wQhzJ8N%&}Wbr)J0A^cE}vq80|CKpHkzIakSFj^PGn0h;v`XvMdz0
zZr<hh{@|-T{rJ;7|Knd|b#;XgHa0%_FAz=tV|;T$Qc9FI+^HsX{g9p;;2i3Mwc;4<
zYm8DTtqH6Pp2r@N+$bDn7}PoIHx*{kX5R4#M8dH+Bq=PtyF0AsDV~uC&t+k8h*6rV
zuCQfIwl^j6dQevlF5eO4RP4+SD9Rcs6tz8C3?0Y8JG6|o#-NR+QktE;UEX-@yR5CQ
zbK%?tk|ZI`j-cu*1g5OG{n{(Eqcdbl&gHl6u(}en&~tDcgYO4~i#=3Tp|zy0YLdVr
zY`}Z(SsY*Dxr*M}Hf8y*iRYh1Z?3Tvw^=w*6Zp5#`S)14a1ZHA7UCAG%f~6}f-0}j
z(<$SPTMW*f!0+W)N|KJtL1DPDxkon5$u(3*ykYWum%#U${+-rnW60+vSKq(N@BIE(
zIe+E?FFgN6{@HK*=FvbM0(5EEeQ%d|n2_cb`N5c_C?T<p&YADIgh_x>O%jn(5(*#d
z8}?;QM+rQmSxx7t>os>4V<s(!ZrCLehTfiHVQ-(4vZ7WMy;h4}e?Xb1SSeBSl5B56
zuk0bp3ZXQ<15-U?Z*oAbErqBX<ls?a=rm?)*bkwvhCHvhdFv*xeCJzy<*)sxJa+yu
zUVr0_BfUNdWTx4DcbBjqGtEn+F$}{tk!k$qL*FO#f(DSGHMLa4j!*7ecI$#Q<>MR8
z+IWttB=;75X0c!p_c4{Bcduf3Z_b)B6m`X*+oRp-kf$>orKl%4<^GIbHN=$QX@hlO
zq$li;$4Fx-sW3-OAfDsmdhQ{Lv*7{Fvy$r{T;sJ@zsrB{(?83Z(`WzU3B)3(OUus9
zZJv7RBr=`hsgx*LA#M+Fe6R8QPzqC5jd0fl+{(pC3)EOrS0$cjSxNSY(m&?l?jo6O
ztXX45EG1K-*pNw&Zg+uMD%soJ!#(b?bTDG)?iQ<^K53RxPDVILZYjmhdn0n;lA@3w
zJJ|{$c=3zRvvO>Oi<d64aqH&8Ch(ri@^F#Gg+*M)V>+EOnv5vQk}R#bdi5HA_(#9b
zFaMYSEoV-jK2ib^AI4A`rsIOu<vy~gaizkKV}c-V5Lm)C=q&9ulMjt>LqyStDyq6B
zi9=R&MY3BnK5!|7#VSKfRxCIcUua6#C2n^}%Zk}7C+PHv=Na=^hJU<AS(K=(1gqFn
zQ^r|J?HWoWADVQo>+;xz$BDv4Vz7+>03tO>L_t)Lo40Q<8IK;SQC!ER+iBBkwGoI$
z2{}(uN>Nmj+nbwQy7(radG-qo1`9_^Ab`<Oj8jH;b~*dyHOhQS;JL)@MZ!2nxQ#S~
zmNiBxs<Obi7ANqro<>WBk`=vv!piwwqW0IBjyqJYi;YT>!HoW@#7!L1Jj4}k?%kQN
zzk5LM%qfCuo7usb)(NQ6DXN^JgyDm&f}QOdxe1t;B_A(9;yB_bzVtHI7~Z;g@sYi>
zC<<6w8q(`_vDz}8OqkB5)KW4V=e+jDYy8=_zrlqwPq4JS#N9hT8k7IX7y?Bq$)k!B
z-41H+0MD}o?XV$b9wLl9)hN1(l1A9n1*5Zu7o#+>?{lgs=-sUtyMoGEs3ilfSa1&8
z-a&}#hwRPf%+i#8-A5lxkwr;jEhfvcg+$t#J7vMNEGa#8Xzuo45R)X~=_jA2C<<=e
zyg~K|Yk@Eb=ykiqQA27@^OQVGF~%^<OWwWw4o^JsG^;DCByn=I+XR}moXu)v<TFg#
zj1I<lfhB6ic!7&O6w>sR)2Q$kB^IAhgOwU34OVH|q06~>$sjG6xR%-)+*+}q4Sg4k
zu$U;p31fD~Q_>=5p_CM}98uRKT2p5^wo(+f;BG!8lNFVxC?7$B?sa=S@%R&rClhYn
zzWK<0WE6(<dp*J+LMg>OO(}{TZ7s8T%B72M^28HQv37hF-}fHs$Nhi=0$Mn1PbH&i
z&bbpy6j|1kNsf!L0@wBMIx&vuVy08fbWB;5EQy@bgM))PE5`-|?H)a^pjB<6w8d#R
zjO<(+w)PA54oc>^!pIq&$l=6kA2}P7O;VH}p_E2d1rrGy<BEe>LG3u?Mn05pk1a3r
zKmEPG$ICCj%vXQ+clecG`5zx*RQ9_)9y|X8Cr+Nk6&{<Lx3M>_p$`sFS~DKa`R;4q
z;y?bGzskvzCqCIs`^YdfP{VOmQ7Xl1w?{2i6GL$*zUMf&$ssE^E3m1du4@KL05Mrn
z(rqVrVMw>GXsN0}YyrX-)LJnrYUZ_}(gsXTEL^(6N2VoJt<XWE(yOH;6Ox^}Bv%rJ
zrnV1Hy5ZuGU-*UpjFpuYzWyg)=Xd{~uRh?-Er3?L&Fb-W`uzdcS`PO2F?)tAZ+a^8
zH0A1*%bdUPI138{Ht%d6DTdlK_=7vQXFRJ`c<iwi>Z-<Afwm4pgX;$b?F6hr>I1a3
zlx5DT70JOp7>_3O7yE>r7K1RSv$}`Y;JOwZmoznOeUNgnT~Op2v|_O6kwk*o{D3^G
zk%Um!RCUSTRIqU~Bg+((g+kkh&{a=8@g)D?AN)i5gC76npZp{K*Z=K*f1pp|2*J|A
zkS8B|ishvxWG%UI^9EWgX7d!IEIT{9{OPy<gmY)kv$AsRQza0IWiKx{-Rp9^KWO0I
z0<^G=Q={wPg)SINttC2ZjFMJ#NFiMEG$T%8!q8>0mh|ezcEG`6eL-GJ##K#PS!$aj
z40Qc~z&cFkc|)-Cz^bO#G7&YqMM`ZI(n=JHk7LvxzwkJJ=a>I$jMn_dKmX@^>)U_+
zz;ii{L%%;@ef<QTc89X8*xA`eYfV*_C~Y`6IN;rjZ}IGpKToIIK5_!lSSqX8&okEB
z9aee^O)p1SlxTWMj_2ZrjU-N4ml&<kWzDK-p>ZhloL&+V3CprnEY=O`%@qRU2~t@z
zscO<%Bl9_~5%j_k&p?@0co?*6n?8tEj77<Qo>HT!ZH<0J3_ttqv;5L8{Vk@GDgX3W
z|0(Z&j2GiNE(;4o*4EZYl7u`j*xBB~7)4pt)Y`DSv(5YOTx9k1Y2r9R$_Mw;M<x($
zAu|@FVQYVn_2F@nPPcIpGlnpU2*MEII_O$q5yVNS@$W`pWla*sIJPl8@{<8ZDW<cW
z{ezU<(VYEJfsr*)AUNKSSR8Z-9f@@cq%2ujS!8hL3_1)k#$#8<+?nN6$^x3fV2N=y
zr>ef+fP3!QXL;@m&(ZDo2t&W&lzep33Bh82k<%wnbME|kw9=Gi&d$yjqsauTYRXb^
zW8)TAuU%m<7yujz!C_3}8}4Y$WICcV7!t$*LI^Ad&-a?r8-dY{T4xZ)&HKaw?KBWR
z3!W?RS`k`XvZ^92YG!#wTGpT~f$Ojk1hj(yS4)JH80FLH#l#COwCkcAmx%}CBERo4
zn6x|OMT-7Ng1B(*JZo!fL{W(Ah=-otaa_9X4#VLR!{Ly+s;SD7>2yL~6lg1{m0@dp
zhta`4tycRJ$L_<SlqEMBGq2d%+u_2Q6(ZNiaa@cwgmHrJ2OtEpsu02@iaP*^HG!)Y
zffs?%gh`vQ6;W4`@qWtQ)||baoXK8_BMeE$WBE+NU@1m81&&|h>xkv`<Fpr6sJwtO
zciGwx*x8+;jl~s$#r}|k=@@-D*IY;N<(I$2nR90vNsvb<;`_eG^3oEgPn~9Ub(Pt4
z#$-BScW;MTnxU0KrJDCHUtu&pV9@U$J%+|m8iiMC#?ui?y+vFa@kA?Z;{tg2CY4dR
zp4-?12*IL|%#4E<1Yk6N9N+~mN@?ar$-IzE>yo07I9MXrWg&`b1s=Al5v9Vj&|2ye
zbRv{1C>@6bS2E4BM(#s6B<&7qHow331+K?4&pg9$agi)b5sz4Zfg8~84LG)ZjBdBb
zY&IiHQ)aU%Wm#girqGI;w{G&(XFo?PX&pI+5>06oX;CwsPw2P%c%o4m2k-(H$7!lz
zsU&XT6C^PR!J?zdtibhrtg!@<j~6y4`FU9}FRP{!D;1stvFoywge0DatsAk%z!7wo
zdU&k>BPE#w2UN_<qS;GapEz*I^Y06DCQ;01p8gE|e!nr7dPJWt3_=zL3miMPLXsqm
z$7ANR8ME08>Z+;g3dP2a8yr7<nkb5zuYF`ZQVUB0sWEh2m%L2b8Sk;WutKNX!tonE
zK~c=W7#u9=WX61yVYMa*BUaZVqNt5<Jj%(0+`{#pf;(F&ktygVaB{_AsW+t6N-$PY
z6glNQCz2idYbQAIg)cGD23b{1ZNN<%vYXGTq(%u?>JO=nyq`c^A^5`Qzrga!F_cog
z_2!!oVV8Z+qupuK>-XvQy2x6RB&`OSBE+F5hG0A@*xcMC3gaVBI(2w5NJR6&&U~L<
zcR&z@2&ZA1Rz=ltl#Hb)YpPlzY@_!y5P~oY0YRNrlvc4_=1dDoXdptf<Ow>x4uS7u
zbknF*wIosz{k0|hCr>lC1xQKe8uq=4d0ry5#yZe$wW;dzegX+xpR;Gr61Nhvv=N>B
zxGah5x{VKW91}+ouHz8+!F@UI!^a^lN+#n8VGtZCegY_!p>QlMUof6eIG9B&b{C1-
z3D$9_OG!R25Tda+o{i^}X@PJot#(5An4sJ4p|nQMCQN6FEAP*l?9OSo4XpuO_-xF=
zU_clLC{>de8M3VCI5j7qc%C5o8TK+jC=O;JcSauLNljI296w-jFl3U=&~+2TZo9>g
zKldC-oN(jT4c@+V=^>O>5Cn9(9lE_9olXa13{e!pq4$f$Vrm%ePB<7He5xc|AyG(#
zfZ1$J;s><aU4qCv-0(DISvLfJrOAuB;fLT5g+77n5G4^>8B|&_(-n8~oT8MptRO_Q
z82hwa9b8wSrJ^d!Mn_LuRz80YH}sf_0*qkf)f|YNOx8%Na6*q(t3{s8@82kNJ3Y>y
zKacCVj1CUCd-u*m&*cX`Nt}=*E#f$$mNlN==tYS`u%omvp3W(XoG^?&MGOy5_!g$K
z5uq2*YPShOuc=OIh1L}g&GY9)jg|)CG?i@ZdW3O^t`$-#CbHySo}wyEj6;HAxfRoD
zwQvNeT2hq-Xb6nU+DnfkT!)!1FxGJ3SB%IhRNct@!~tOvKafBc2Sd)BIgJp4-JKnF
zcXuCpt{?~wW0(*}F=bTX9r`FWM8t*^JDQA9QWE$NujPj;5Yb?QEs!|ixiHD+1b)b^
zohj2)veXG#Sy{kwE!8~3F_z<JPT=?sMmASVp6gMiGj49&<I=5+tf)D);`7+472+ho
zbuE67V2x#eXP>~2SXw?t*zOXwyL4atY2N<5-^Lq{D1(r(wbWIO(vnhGKDc%D{z)2|
zjm-~2X0s`i$(Xk;zWoqK#Pd8{S0If<DtU;)Y9v^MZ5rRkLZvhp-+71Cm6OELQ6`?a
z|K393;s`-rq&Tk2P9`X|p&vwaTQOE^RMpsdE)M%xY$Iw%Q>M$TV0$v-PElaAW;yg&
z?l0gu;0TKsG>f~O=lEevD{kQh5ka?$_v9+q|Lohe+imK=lA8jpo6S~P&EDSjeM1^J
zGz)SZioD=pf1kVe9$vVv(=cw0)fn@A(vtnig`x#&rMP+f7Q+9q>iUFLoB(Mot}lqg
zMh$j6pJ0u{l`DJHwPAIo#lmtAqibZ56NNs*V`s1!j8+Z0O*o{p1FpWi#g)r*Tu*WC
zsV*l^uHgk1FK`Lm7G;@|6*J<X%|fq7&<bfUpGLPH=lZ|<HC9$mk`Xa26-p_T(bR=z
zeCzuC>H`{o4Fk;QGd4D^bM@MTnoNxsjMr36S{}L?9?nDCEL>}vBs|Y@-hA^-bX_;f
zxknfpg)3YF0mjzMvnkrbc+yDSFD41CD5S0>N*badq&Mi?7gqd`R-8@e+?meUE=zo)
zIo@uu*ze;QjpI1@aZHh>RAoUJw+KBS*NthNSf)r|>szm|uyTwtR1~T@)H%~sWldV8
z4<Pk|Fl>mddBNQ~ci204u!jLo6tU|hgup%QA!?&A)*KRs?O_$uY<?C<%bbmkjfRSF
zgc#aGEjHh2VyJaRnoX&VVytRB>$2Ex6NMgCsW8&eN)m#FMv2@S(;x&5%%*d0PiO2^
zC5bg0?{(?7yEsPScs^bjl1-;gpSaa-V(5f)&L8KXn6mxGH3r9)Dg263RfnWiO`6X!
z#e;iY7)3}W$@84e+qap_CJ%hw5dznBu)?CXI$XcvBP&_l|4gRyl$$qh5eC71<m!jx
z(L_QEkr1>*Bem}0^4hgs-rLN0?$iR$oL?pkJyc!M8T5#g7Dfp2*#w~tVUnN)R6-!b
z0PUxoIKR&F+A?0?6Jm%052YnlQ4od+gTW9Bl&(V@cnIro_tp)9$iq#Rs4ES{Vbj-w
zG!Hg&q}Kdjzy5!7>D^0sp5V<l-+t)1w9$?J{dmlDafXASC^Dq1ky75@2$Z(G{=G|_
zT%XYEbdNmo0w*BAY4G+4u8j9MsB0eUEz#}w5C0$Hc1#ci7;C8W5+NMCz(ZSuaxB`j
z2v;*4CUiP&ToDoA;yR5FvpTH&l1>L>G__+0Jr@k@Z*OCdErI8tGzje+YD`(o$1H2B
zy5ej9{%c&has?%8ws-d)o*TwcmIYa!k;I3J%u-^sX;3hS6^k;4s~a2i`-5iUAs%|I
z9~nanZQmjA!Sx)f((=cz-C?v_@l!9av2?6=sMza}EG`m82~sJlbdKwK_)&yZ3TZuR
zJp6vd`h|4{%LBYfkc8j`%@v|Cf>zX}+g(5!NvRyVu1{S{Ha6ZTN_to)LgqD&ZWfub
z_93I5gYkra^pF2B|K^YWkc067yW9H@J(E(3`FzH7GNqHWD2jrzDp5){u{?}no-4lb
z^>4GZxJZ)3M*>`11hEM4ti*F&3Z=QcdmCxtv%@vohnC2$>o(T3VL&aLO4fBfJl}2l
zBA!JGO%NJ}!yZYijbj5M;p5O)BO@H*R)@INLa7=lG>+>c>x!+-J0yc1))Q!Xm?&-2
zDg5|6+#l`pTfg;NtQ}uvZ*TWuy9H|x6HHE?W%#~NRaHnO8?KI4_pe@yy5_A*mq_A-
zFbIzj!DhrF!nGx?=P|1*KA7#|i-2dBPtb}&jMDhNhu;eD0-w4n(UrpUy~YE^NVIFI
zg(OJ?!{LA^ZsC}K&}kv8!DxeUT#|M~7{^FeV}-$ST&l8Sd-E>s;Q-@-uD`EY_@RRG
zN5|c*JN){u{W>elOKjfRr2bg*+geakk!Klsp5Zz!b={b&8Lb|;)}9p^Z@zPpAP63+
zLLQkw93g1A0<??Kmc5DMwVMT}S}xC==;9s{<2s8&9M8iTO*NU|NJ-S`V%$dCDewbM
ztSlhbPP6`t{~aqA9;du_li91^p#0!%Lf7KBK1r*MbQ~_<-DVUfJXu(NOjvMzDrH$}
zx6#%fGA+d;;cC`eu5WB`ePg2$hI_<nlTtDskJ#F}OOmt*JdgQ&Mp@=)(;z`UoOnsA
z&5JL*$hq@-&Go`38cky@5-t`OD;&mJakDBo>9tsi8%-hCbqM1Y!f7@!nb!ydei9s>
zq79z!Gi<l<*83d$+0W7Hcc`}~6qoP8R<ltGM-xOLwXL~3-=i=NXY5%PMS~aDKoCV3
zqiL8Q4^|8xjp5y`d)(c+_XD3yDM>b;GZ~NZ-A2Qwt}2wQvF0PJ5+P`{T0Hsq6SUeb
z9PtSghy_mwlGwo%1}z0QZkJrXl=G$M6Hc9OANJ#7f~bF}!&Xxlb9@hixYalkYD=8N
zoPBJG&e_L#{KdaZtJ5Lhe~;|$pCZ&v9O2@*0bb}aEhO*X+M{xPUOXpg_bo;T810ZG
zEu>O>$eDiR98-$7E?(lTi<f@jJS?k{gYiCF+xPHX7o}>7BBhp9<3oS=g92DyUgjrX
z{t4n%`_V7<BZh}VAR5UWEhJlY!Imufd~3*Jlwgg)4}Cnp5$u$;MCS_6aq&VQWg6mI
z6h<5$bSW<MS^MeF;X5wn%{{7jwunYG;!t_lcU)$5%Drq4leBq8s7Cd-kx3>9LzI^2
zLwS4ih_qPCnyc5Za_#!nM<z`RR8_@%I$<<wP=a+;AI?MdKvfMe94>L;{CUK?&O@lo
zzgP@i0^y=@(HL&lDPtx1@%E5z*tmfPaoE_~3W2O@OsViZ7ccOUT4E6-am>0Hpg%ig
z?I)hbk(Od(kNWZ+;k>!F^?V;c@EDa-cFF_XdO%w=+E!9n#GzHNGU{+2w2!Rri?ZPI
z`<Hp|{YP`<ES9n?nN7xwx*fFElzG`8YSn`=6vAP3WtH{gYbgC7@aCZeVlDW>5(-UW
z0;VOTxyDri$6JjKQ*Xgwq@tM3uu3&0whQgS0@m{|*3yZ3SflY=mo6UIl2RBt7oNc%
zta9^T7YsJvBM3sIDS3NykGE_O&$`_Djeo|qjhjfN36mDX;UZD5g=j_SyuvEW52FzN
zpmk=9r6_Y|>4eE_L@R8deX^<#Wv`nbbX<pCx82y5I_~`@<0lzDF0O#eSTbXoX-g<#
zI*w0T*F;HkdN+9m+M<<e5=uLw5yTZlLDXn40kJR`RilI=IkSRj$Be#tiOzw-_kE-;
z*qD#m?P#njnSA3F7B_F>h5^EL>2}+=feQiH`iF+^hYoEmwX7+!j6BJ3U9UO0=!XFd
zQ4~_w6`t#UVozoZgrS4u3X0S+-ZyN0P%#J{+KFH^Dp-p<=(?t;s^-)rE5bM+=`5gZ
zbFJA)611wZ#!^UwqNWxe3#VSfdVTiZ|7Q$?1A-``s#Cu6#+W_3N>xnQ|Ly;S<<n0S
z#XTIyV`+IACyH<)g)MU)@ML||GyH%rE2+uyjLB?Fx7z^=c~LY<u7@XcXaFdkZi~Fg
z8qmroO~ODRT)<FT!%S-?(xK-hxW*x^EBax8mKt5@#s^(V!oVj;LNo@~aS6kqA)Q)-
zHWsBy+;&3u$4)VhT;|`rN~iSjgMg|mcqiXR4jqd9ZL&Z36Iv^Mf=&z9^XT+Ch@iPL
zHuYary5tct)S9X)$+9_77&U#as>T?@M{#kk=MjYgWmyvVNAsr-X>J1wEK@C+ONRxw
zjWw3MF6sC!R4q}Z#A<_4nlK9SVjpdEbL-+bD5;QEQ!0a!6@DwC_xaQ8dzSpwO<FR*
z4}H=y=bd63vE-3&-X{CKKPDI?`0WJWYvu%g^L$2riWn+MRhH!QIf2(yakZ=u_rdSS
zFz`JB-=nIk2DI{#@yH4!2z_keQA)$kSmV}kCUJ0#X76Ci(n_De?@>-uRGv4@n6(vb
z93pUt1D~pvT-%!Q>eV~k98U;u?(_W5{~~A3U*N>)bF`m%iM<cr<3!*fosbVkJDm8+
zS2%zA1po4X{`-9J*1J6Q^FPm-7thn`v{+nPWaERIl*Zf_P5h*zxtlknY0B-*+XP<7
z&dv_w`5dYA{RQs`2jMt;`@7$vt~D?J<WGL$-{trM<5+5=m}&#YXUP>9BbcRgI!Q<*
zVq_sPwZ;#8{C0?S8y2?jI@C&XFQ2k;u+6xfljL_e_xxuW4wvW+hj?d~Nq2A1_Yx|{
zVOM3W{@7E*_`LB)-(~yG7AK#(K)2V$k3*7H#ArH2a&*lc0A(~qS+IYwk8m6&lL>iM
zKJd9Xj)T^U_piK9d4E!VvJtdnBo%p~*x8hL5>BnVRCCF}XrJLq53Hds=4e@wByIdw
z5AC?PPJ?or%}Xv{9`WWoo9xZ5uyk?G7he21*4NirT3Unlk1-wnTNb+!<H?*`dxn?(
z#?O=O-QjD$`G0cdolAV?%Rk4$aDgnVX!p8o?@TE_g^G@`P*s}obb>zg#lE+-Lz)$s
zkM!+4&u2ED^QV9IXXHhFD7N&;h8H?0tEr4aV2DIaPZ&yRna?J4S`k)zsG`PdOPGYX
ziH~;dVRb7=>yn$(eKz**k!=+W%Wcj*`!tJ7Lz2}2)73WRwJC`oF-j$4x8e&gzr^;H
zyIlU(C8A->splS}mBe)0ZT!$f77B$q@;o$HYTfAXDc#73jAs+7kIgmT^O#P?Z13z)
z$@&u~VIjct1<I-h_HLmqS`36CuNCvzm?Vi16{u2zhNu(cCSHS(3VnotX;E-vvct`T
zdsMrU)oPbhPhDWRG^BNE$kw$jyserr2$;=F(w64=pM0JVUb({i-+GtM$sX&UJxd%%
zbi17f4Oc1Dr_hpUVknu-r&O}0uH}6$(nmiR&-dBe-Q~{RyJR1OLO)!A7)|@cIt#ra
zo8NnvX=*thDptD+$MHd1@@$5$4UzEZ4wr}qUA%6q@%=S$cR%CRy9Jvp@y2M6tx-i!
zZE<$@fO$Fx$0cg_m?)QZSCSSLTcd(bDXHCn^^>ba2lwdPk~j(xhivIAZ5G(4A?z1r
z#f=-A7;~#R?HTjX<|c$_?p2j!G#)*4y8cna8cXX~mpJT@-yLC$!$3gKD6D8=SQiCC
zH*mzL7vU#9A~@98!C^O_vmsLsBEt<cC$C5JMw+F0!ECz0Ip4!10lBdBj3tvb)4JxE
zw&++e>~#rCLtA<Ho`+)$*4X=#<I^ONs+R2T?_=#=bCvTbDGwp=JddiX$+GN|ClG5`
zIQ1kgeZc7Y2e7L-ejJuh7}Q)~lp;-M7~d!M1H$frsM~`uY_RpxaP`WVw=Zk%?Dcv5
z;!Wl%=Zw6|sk8g+?r!nIi*2OnQD%aLj%I(qV6<PM${G=NICp-Tb`1Sk<9R-gQ=p`x
z%p3CHrx7qo=NTK<zE5<0)YXb38id<aO_@%oN9ZC!_w)k6C}ch=aPe5c&@~z<Jd9El
zX-43N1VM-&hXjcSp1=_fN?UHv#@uoWcH)v7`T$kus8NT->5S=k!sDNLiaK#|tsyps
zab1vGjZ_A=6|mkJ5Q%`66W}@yE})b~Dt%<;^oJwM8<$aIY=feFL=1&+@O%$BPbtdc
zlP3_zp?hWlZ_i^sZ<^@~7+S`lg^!k+VxBfJ3?uy5$4@+P!4WRC(%eW7xZ{^>g&CWA
z4_nuGlOdgh1~z-{@y975XjzM|4WlZj@(faG{9ee(_5z_!XnWD2PP(9$>d?6B=(ErQ
zvovFtet&&FYMnZc)5xmKQ_Au|!@Wl&5KA;z;N%z1aj>^XdUK1UE9i7BRVhLHn4+RG
zKJl=HXeVgb!R94-Rr9^g3Gbu}?9%7<_zvH@w~5I;j_WpWzVa4}D_x#=`m;>;cTq)&
z@Lk5Ulz|WTzW3*R>$UIT?jG>P$m6kNeegWy(<xfk1dc<1rT8>KGxnhd{KtpqIgBS`
zPOh&p9#0=?Q2pp%0%1R4adnCLe1@J0LPrx?MQsJR9!6@ks}YfaK%g-QX(*&*BO9|B
zz}!{rk9XOeZ(&S8UBbro4Z59}Q|Hbx-5=p;3oc|;MbG-&y>yK|oe~y?3*I8jtsXco
zMV>coSzKK7ec9(D#n9U3<orX1BN~Q!uh(Tdn|<O0VvVJ{e1f(0r`Xxs$Ejwt7d+y=
zCaW!0JE$^8+BQii!iF)@T0~K>J1uzq%{d!4dhDkzSGR6)<@S`WO*o+DjaOe`<>V@5
znX|XO$)eU6C78_$Ru>en{ORwrxmVCPC13h%kG11{j4`C?46G$^!E4m=k35VqKk`0a
z2uqe_tgfywI+z?G2}ApMkHPYQ=^Hbg+97cizHL5#D_m5iabt@J1lG3=X;j0c%z3{Y
zb1#NW7u?(5<X(P&>m+!ijNrYitgf$67A5o1gcdGJTWYE38<)5K>=nkvjK1=DDjd*D
zx+twF^JWwGaj9`WRT60wkPl<>h}a041F@8f!JtoeFF%s-W#{C8PA_D3DaA7`iKFq2
zM&ljEP~j#4A_VKx04pTymt)?qCv3-tTxM+U-(;_t;CpSNoiUbcoIZDkJk7|bDS`wg
z6-pVp8NBlKZ!ybKh9>6ec!fA=BV}W$;Rpv0pIYcor!@Q{=b%8z1_s{ib(wzzDhxk>
z2U1lzMI=~z;aPUGf;1a3t|hI&U`kC{8?KHr+_9n+PB^icpq1il*ACcS_zdTtdy%*P
z@V9yU>bu0w635$5v(!0BITd{VD_<mZeXhR#DrX&qw1&ucdH!sd*JpcVnISDJ#=cLO
z1Vp_7dw1{9RyAP~(bgHtfclgtz#lYx-{;hc6{<32HXReiN4qzbRf!rX24@GPS7#Ju
zO<)xs7Gnja(QM{Z0+kVEJ_}(&QI}kt?@*i=(m%e$#_Qi^XS{>s^@+O2v8@3!?ee)N
zpF~T=?A8uz)<tSb%X2vwcey&*r(#Z~45in@_g%awCL7I&wTtV!gt9TW`ZTAPA2tHt
zXR+6#$}(i&6L_DzhXj_q&asQ9S$p<#Oy7Bp@oq`DXmJzJ$}pWOF5aEe?$&f#Bi7ei
z?ChodyKhbDE&e3YaL6~m^V@9gjXCbEbD{GD$HXGh)a8XA|1#6PT_*P~bNYm!){<i@
z5pmz)8yD}96^-~{DFtym#&5ORDkgLi@B@zo#XRRPLB8(Mqtk73e0|7xwnv-<_~8%r
ze*1x=EHm7dMHWx@DR1P=mBOV)Sb???GhK0zrX+Ph;)g7RF}vxMH_9DapE-|kH1EFh
zU1nvD8!Z#{*C>-V(vLa!*b_{)$EfW&Asw`qEQK+#vb=q8kCKdi?V&Nmafk>am`f7J
zXX-eF#-;p})$k7+tyYUctHpdWp+D&3x(-L#HOe|;xY}lM+@ZcxGuIMfB_PlyVx$UY
z>72L<X$KK~Kj!vqkGJ(ci=VrIoR4|u&9}&<#A~nM_l}bWF{0h$<k_<vZ0_QYYdjmH
zjb<%w(^ByE{#^=kMoM9YC5dB<>myQ4<azkQA+R1b^8cp@La)=N6@<(tV>;dLLvZr{
Y59il;7N#SHn*aa+07*qoM6N<$g2K(wMF0Q*

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_tv_tensors_thumb.png b/0.16/_images/sphx_glr_plot_tv_tensors_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..8a5fed589d17fe6ee62d6a9788519dba471fbd65
GIT binary patch
literal 26794
zcmd3N^;cV6&~<SuCAeFl(4s|(J3)&T2rk8;P~2Tha45yKc!RqHEp7#hLvXj?ZZFTf
z*8BYp-w#<U>z<r@Gv~~mJ+tTR2sITsTx<&LSFc{-D#(M>U%f(#fBBGMqQ6{mHpumQ
z^~%~w0VJvEnR)nlooKF`yb|H}xA|C<^cUE&K&!B{#ar*%RrUI{fOnuSCL$03!--#|
zXbLuXBIIR=G00EgrNBb5!{Wg0L?-(}Cu?Haf+0F7%3dlH!I7q$mL$Yecj9<`jSzfT
z^D5Zgt0)_Z#Ecth&8aB^q#^tt+jp*g;+l3Kuta%Af{32Sylxr^V1hg&jmnR}mkXfv
zAbI2;{|#XPvE=`E0P<S?^9AVTE1Z82S>eA)C<&;HFH;EcKm_ss4Tt~~BwnThPbC<U
z{y+Qy-yssCdl=>g5(+AS2SjXlQXBFG6FYU#Fiz_d=HGDi{5xv>!;RVHaotlES;<Vs
z54)3=h(~4As6dmO)fSx8D!+i$VC>Y{+4GrXjSnu(wiwX$)Rm{msq2XKT(PDE-xEgi
z_(VivIE9l($F=w;5%X502cyd9S;o)?@*@IUt%-#RKi)is3GxWf#e9B1a0p}-K^GkH
zE#RcLH-hEomo}wtS-`xST`geg^-vL1i@Xk;hL(*ihF<qw7%c*?Q+E}k)1pi-WN4&J
zrc9QIvTOJ=TWorN19X#w9YvE^0xfTy!~GL#58!=LRoJfOBeNadJQGEW^>VL8OfGro
z7;=ZwLXX%~LDV&-I)%f^Zc@iBUPr)DOdL&SY8Fe4jRr0M^?OY#0}B$nzJ&L1In9Al
zQOK9T4G@xxBY?1&=&}M<$uJ+xfTKBW>PRAN(VR;B>we1G_F=pv`)g>`Z^uYRx_jEF
z<Ml1t4a&-D$31j+QCSBStgUGDG#OX))Ly@#cs>LgunT}hWZL=J)o}-tSc-H43?kvu
z2VGc#wT9ESOGB&rE@vy9Zz3|&#<T%q7(yR6b>)3Ae*}ua?Sht~cbom|GLpPlTSm&X
z8g90~9$!qLFeLh!xc`P0$(R$Ja&h&6EezoO__ACMkpyr<B+{;|AT;h98iu@OkGbHl
zGYQkX4NM%Rey;N@d@DM~2kNR!mU_AYWtf#)e}ui_>*-p^zc6Sxuga|EOJ~>i$53|i
zk3uPR?4WZH-H$~ya%J$=ipQkS-owRa4tw(P{%YPDc|taynH>AT=I|^o0}49wU~C{J
zauK2=H85m@3Fzliq*hcd5LX*{uUE#$CpBIVU20AE#`cp-cs73fa2V=^W|F4HJO04v
zh?+`+`1w}s-C)?|2Q&Bi)%tp}=My{T_L}AL?X!5%*(mp1+HX=ICa0VpGHL53#n?|w
zrsVN!1h2aM0663#mGa+wr{K=E&1QL;2u+|ml4!Vcv|H~i8eectf>jNDUkq^angTdW
z)#m!O%sOuC%7f^!Qht!qCw#tfe81zkyyL#c-mQq;u_=hU8Y&X0P3ZX%SuC=j%cBp_
zr)IOhDL^juo$ys}!om$%Q`tjo6J*I-hLpYq;KY(EwP$7&><+R6Cs-uf#l&-#?&}Uc
z2}fsB6#K32O!O58@ksbMFteP}G2(2lnU(Xx0EQ{|6Fyl`z1ErflfT{pGkLUjTO->;
z)k=c&e_nqG{-=f;xxac&YcIobydWM*>-WCPDRj}<`M?J$>U-1W=e@wPjOqn`WB6h3
zK~jXSxO+anw%MV<(bYq5#QL@?&$9}E9mOt!yYPhfLyOPk<($)-DALZ(Z<${Ys2e!I
zEf@cqbht`&4<v}sP#}Lx4n!!n0Yy~ga^ziw<gSkSH*`h*QC-LW26MvjQwQ^Vr$sI=
zO3>b(^c=S@?tyU^-%7nsg>47yyUusdaMN|<O`K4>9?WoX_ry|GMNc@rE*&1ZWvhRi
z_S3m-n`bk(@O+$m8Kp&CbTqx#!3Ro^Y+#WD0wD>iyApzndvKsR&ejCNHW0D$L?Y{&
z-+!(ex(HR2t+z8ezJ13p%e#_pEzeK4BMasjYdnwJ?s{s>RYi;(GFI&e<aS()?XjWO
zoYY<N7r&duY31K@BufxfDguxb7oZWLurjz{vg84UbI9BdjF6-Y{<Q7%utyW)+c$+0
z<MkF<9?8fp7m*z+oS$&oEZ>z1i>3C&Ca}f(#daZAeX;DyOc))i?10TL3cyU9coSu%
z)362xbDc4*fSbB%;E}Qhj(t3Y;h0y44|>PJ{Wnh8tcJHxpgfb*cWZ`3JyR%I1eX@D
zhJN6ES7k-g^nBRA6+Lz{hr;f=v9U%{mDsRr?D`_~Bl#>!q2hGpQrdVLN8GXyT@?Oz
z+7-##Pu6db%j-<<)!r^-I*w+`UXXo=8fYr1;~kdxDdbuG7#=k3gn<y*eN*f1KF$Bh
zuM!f73PNVVWu(3KcN;%3OXZZG_KI`Pw|e|IE$cJPXsj;kh%dz$%0ABb0gso)Q9gp2
zNtf=PDrDsl^fanC@8EI&q(M1hW$#)lTUm1o_)})e6B!%S2>wVoy*Pse*8?CP{EM4h
zidh=Lh5n)*Bd=X5LAxl_{94L7yp5$zMl^#3{EGwW`CdAsJ<I9#=;R3}l<zuf@=^R;
zn<3z5TIkm7d*S{Q&!Lv#!b=h$$&Th_>ZjClo-A{sUr5=r94uSOqPi<ojUL#vQoR0D
z9@tomD=TNeq@pL#T4Ehm8}_x9rr7iysgJ_9?J!1H=qLr{xNdprpE-Y{97Xs^)JR-0
z+mCbAPM4RgJda+_w1ZhDF<f#6KMK?RRs9%DGo70zPi6-^cvsZh<Pp*;9q$hp1HB?1
z1Thj8DEHa%3MWR`2XbC2FZ(&PJ%6R;$UL|(@;7M#``OoA{*nKc(^Xao#r-E)ODh-A
zOQX3n*Lvs2@mM-*O*5azC5U{B1MKB77dxh5mKt#I>$Bx2XqwJ0Xj>oqW`wvg3e07J
z)B$YP_^e2>9eu-`H$Ilmv=_G3H|}OW)AT=Y3o&4{OAN>c@maT9IRH=16i5c13XIbq
z?_;?AkC)0lxBghxk!|c14fTiYk`u>n>GBPA8g4GyLuofED1?CZ_ogTPNTn=*7Rz>`
ztYLiEcuec^&_!PYFq(!2r=O|z#Ft@!biey30#@bucj&Ux#$iwD(4dukXbV1&1KhNt
zqjE7f0?G2qOI$YC4rS^$^olhr+rM>-n6e_o=%}lSws$wIa?!a=?|&NpC0eU;zsWc{
zKWScWoG(_FP6kwvxFXl0Buq*6*?NA&t1N%CNhYQxAHE>%%=x@2fFKMkfr^ZzsGxgw
zHOYm3bg4i(w(q;Sv}7U`6Pvh6Lm7Ptpx?d8)x>^1RQ1|vuG<vm`}&L2^y60pzYXJ>
zD<jpjZ7a;%a|D&NxnAb?nXhfne1x>|M?Wx8nh#9<4K<gGSUSEv7K#rQi7l9~Uj!&y
zKe5Cei&k^WbCf;y%2w7n*r#WSHtKp+%roO8#6_LwL1~<?Pvf`t?S*N^=01T?r|(|z
z$d^E$%(J*;d8ErD{ZEFps9f+WKOzsubk@s$VbOjhU+t{w@7U3KK%cp%uNDtZ>3^p^
z+tzr;24w)=wRU5Hn4~bhvixzHRwBbhqo*XVUm->z)SC1la7x*{$-z(_ak+M#+ZP-g
ze<IxXQ4aEqNb_=<ma`Qe!{XjvZ9eq5Ib;szIj%cDn`nu6F$klktHN)8IeDy8S5qM2
zC%E^-H3a_+ex7p9xz-wfTD14=;J)uV*XXUj)pBRk@Pnh8CqmCZ&iJ1QUo^Z)B@vIg
z(FlGVdndb1<*-47yzX?oP5~QapP+;@NEQSQ)AQC1_Ty>JQP$k`X`41*d?oW^%4JZH
z`2Ld2K6XjjT<v??>~MILbr{_xL>OfMd9;zSl*--#DwbWDM4Mp%J~N_%)nv|~hmH9@
z)4zT59opehPLJ^KJ6=VVO6Ho)jzT3gR98(NX*61GGVm56{>rj2M18)d^HhPG&iNrA
zw3~jo&)!0v{GJE>iXI;ylLdla?z?dM+T(hzjWMg#Y7nngVIY8XR3%u9|L|1IzJP2%
ztaEj&!l5_dWLvR#((N&`TQ%Z_aD4=0dIY1oPwr0rTMp@~krDj&qrOo>N7s>&+3e2#
zj!pO%Q}P>}{K5VvO(@IX)|C%9o-~d!1M>GPa<ndYmNv;aFk~!<is==b3_pJ<30VRo
z#OB^{wUn=&24gx1tf_VnB;eG=J!0z~bgcJJX~-Ya3?h@$yfOIL)G*(evb?a%VX!~f
zqp6{~#WEH-Q1?9^ryjP0Z$$eQAyAnI&7&uR5P|1uwXt}Tyv%;tYAc#KvBk|6x%}OM
z9w~hzs3Y5-^N5AaGZ~c=G;%D?B(WRA1c1N1778yvhoaD%3rgBv|7EatdJt$>$u9}~
znp>y9i{nXjS1MjmdxDO!=)mKBA}-}mBp?edtY>D#G4~`p;ncc6ZCuu^9162{V`}LD
z1V6Z{i6Rj#_U_ZbKJm)mqBY>vgDo5E5h<Emio&5$E@AF?G3Le0&bO!{+=wz|)7zrU
zih#Jkm+X}@+r2~Gu{6>b5`g#UiRW{W{;Yv3BoGE+jQunVS36Np2enN*G1C~E`p6Rz
z{j*O=Hr16UEyfQtZjl0gr^mQuLw~P`!;^`G!)B$eL<AKUJuf^<NAv;d9AiDUw_1$1
zJPo3x5S=Jc8|9~ObacN3AY*5Q!3T7T+fx>?HKwu<G$dMGGeKe#-WS@H^V?x)LO8fs
zqmwi>JHg*Av%bEfp+&rI#S6h3Du-ZF@PZN*czynUs7CxrIzi<4CKW$DVUD<li_Y|i
zG17HQvoz)>Zp4t@iHd(8;)$gN|7tcwXZ*}bVfvjelr52^Z_wbOI56$NP(NgU%iVHE
zF4%~Q^N)t*gCUrPDQc6o=jm9zod2D$s=kAl9(4KKJVtQbESUEoxkjYn&#4)gw9Hf$
zxkomp*C}$oszKqQn1bcxZ?n$jcvSPa-%+!QHfnAo#hP0@4Eu7#;_sn^YAFm}a5$kj
zR>ve*17@Gze8vcL$w+E@`m5(=Qy*Q~>OI;1*l|C1v1PMWOw3eL9{)(8=9b%IM}6!U
zMcHwGU&b-gd%E#WWQ1W&us{G=3v*Gczbd>nU~XA+RYmPvAK8xKIjItpke-|6DHLM?
zV^m(Trt~H3n8Djvl-rxNw5%%KO97+CBR|c6-1xk!_MSCj{y-=Tl)@CO+~o2vmxWhz
zY)tX$O0oEU#QKS|IP#<h`b6RAJ-6|QMRwxTv+S{FC=#_NXFVnwajvlc^v;Fe8pX9A
zhf<6M-uxR-uogNQ8hEK)IoW(Fw~$RK7de(kPizF2`4M=?^(HIkpV8g2>9wssiTL8{
zJD#*5d^it313T$)w|h}CIq*lseFB}(#~)6PjENhpg?^8Xm3F>+oFgAQ0$j`746Z8!
z4T=vRkY=AQ@V*?NvGxTdbw)Lbpeb#Hwn$jEkRl=kfxy_mFdnIZZ|v?Bk47e(MA6iu
z=n4|<iHvkUa;Ob(KQEpZ&J@0)KduZPmaKWtq0k?3_hJD2Gm*qyzI97E{Lc|!PSUoE
zK1%~C&UNiRN)zQZhN-r>Yw5OvgD)#Z`yIX8TL)kRf#j_X&X!Xre9^FbrGeT&ua~Ez
z;L~rWd6P4b4BJHLCtjp5IQbK@Yn5nt(<W&i`^gZ*x8>u(A%E*th>PWT9zmg#$i405
z1u)h5>|%j;2S6YL!d+3OTS>AcGo{_X_PRbtC^cZ~OZKy980j48DF6mKe0<jN+^Q!d
z6_+wYh)0B~28opBG018{`&UQC-Z&pbLmX%ur<=#gp5NWH^gbnTA4PSVe6dX3@YPg6
zDr(C&J6Z51?||W#t8Zxn-kdrXW+H?^p#W}niWn)GH&MG$2$2AjQmXqq)5=*Mm#Z+(
zS55!)cU_L+xQ?>`ShOy)%)$Gpx&xYS;yMa=?0Luop+4>{HHg<+3oxDAZ>#%N{Z}PF
z*y2V1MAwfF%&zQdI+R~N;zar?o(NK19$nQ&-1jXxMS(~iDLANz^Z7LflHO0Vw(S<{
z-&<r5eM|PgPITIz&u6(UmdqwH`L~qt)9ZR|&VG<JT1tSay74fpHXwRH=;&6EH^5|0
z+LuTHxkiqxO8upJPHvd7O7}8jG<V61p27Jt)N=R14)#js<10eRO&x|CW~Yj>6fR1v
zwYJ=7+&AQJkPZmGr9o>Y;3Dk51T1!Wny3eOhpSrVHP)r8j)n->0clR9O!IoX`MucX
z^Zwiy9^o(E6WZ-Mwy{?YbRia8Hzm{^kdB^~ULyM{?&zV&+;rnvab{}4=iyZAU08O5
z@MV3_6vpe|{fU%r_^adQ-fn4Eqe6TqJ&VNr&47I)qv~~M_?mkg8<^CFz(S?YiNc&J
zF8sSlt6VZ89ZVI=m8J2(?|e_Ih#V2_^euv=$5qy@pDZ&CU`?d!{!Xjfsm*EywtUo{
z7M`bz>Lm*<Ce}MtHYm{7r(J5)5q}n1Vl~^$H^%&tC52`G-pRCoZdtm7PEwjEjW?3{
z11<BF;^@L~F7Cc<j;XTn3-~YX91YzpO}BG$c$6|oZbk!Z-y5|EV+OKNAcMg%n<sVj
zl-e-={H)#pfrJC{pT?h4SGInUS*~;n1&|2O@-f=eNKvRZ4R^(dzj~9cl6p)6{Kv$q
zqIam>;&#qzeUCS&pM&I|_VXXi>@ekIJ%6OP-srY6;NEm+rQ*Vm2SP=dpg`9v=8of0
zhOkW8VtL8ieYY31ECO7=oqO;@Z;1Kzm!dv@X`5}nJ?l8#I_`)#C(=D2%+*Mp=89va
z5suRsE0+)2#jpPQZ^r2U4e8Tn0Artnx(_PB{~E_^f3atL)FN~J89#@h6>YEJ)SkOU
zkVI-6t3OWeJfD)pik~F|nR+pV9<?<+S1*Gk>ikPTcq<UXdw_)K!lcpOk1iX@!U|Th
zoEhV5Sl$g0E!rygh}-<LV39X5LUDVTxY1S(aK&!Kr*rX$uZujy4v|hnhJ2+@m<J{n
zzX?5@^gCzR)(UN#6V?@^t({?Bx@QIn1n)vheiuiVE^S;DDalwPZ+w=+Bbf|;LoX&9
z8F2XSC1R0>m+4>0E|mxI>`U=jt6gulFW~I2NBxSomP5IND~h{YMxG>1blL@4eF^{6
zk#UK;tX9dPQL#jH$V)FC!y~DdwEKH`o<?o9|BR|2<^vN!e^m`uTJGR!jYdHl+#*lk
zs5uVJf+{6hxY=%{b1<y_plLd2Uui24dyK=?>^W&`SzLFoKY%L61$Od-y6mNtoA!97
zb$UD;2B^}_laB+luIKy^Uxf-;n)u=5i3a`beLsWhyI-)zg6gb&zCf*CjB)ffyJji%
zQ#_YU?1e#Jdx!n7Hh(Ev_|i!PMx4H_`%&raZn?l8`iT!XKN5a0UHxpS!xGFhTNf+u
zxpL3g?fDyTpaY9;=4WvgLgnhu2&s^lz|K2HyQC6^rpuT~EyYofcy)zpZBfmg;sx&O
zRGn#3W%*3V<n*|z)Qkf>l+k#uA^Ks&Nc5APO_6`soi1jM35BHy;tl(LIhwN&a^a`%
zJuR6vsJ;uQlY9qsuOj!Xa9BQ;(SH{l%NmW8PkWZ6_$5iF;!BonGJlTSxyBnademO^
z=BG6}B=bz`T7Vp#E}gK#fA=$pv=-a?p8t<XnSlE>oJvm3oLf8KYg@E~;oi^ZJM)2>
z@1e%vtsSw=PX_(a0OjVZhHr*B{g*Z;W~=k%vO2VPhKypEAzx81pO<(tom_IjrLqcK
z@zif`H$2E)oZ<bN=C9xVdlaU6(E3^#$Kr3w>V2<j@7jJovp;i8k@<L1`K44d!+7h^
z;j!IeS*PpL6VpURLOzQr98v;{UR7Gn4&D`3s(EoL{Hb3^v2>FE-3#8Buc{2P?839y
z#~{t0<hQk=<*T18>*VWy@2eH+)q@uti{*C6u-6YLDb<8RHk){z#wL@!tX{(Ia1#tU
zO(0i6^t)Ueq*UJ0(u~F1x&}Xc`0a@7^&Z~y#d7ER`@Xo6>+bhoC<@-^DTsWJz$8O@
zE6q6A|0WUD9M91pzApN8$^s8ZUZ_$??^mF<Akg}=Aj6dC0d1XBWe6+yIu{JYbdiY4
zRV#EJzDK4PtaJ~0363$VH|~2P9v-U1O5Y`@`zw}ofRT_oR8hMLibAhfS|zpBh(>i=
z)P@l=?Y7Lm!mt}6Km%zR(K8vzVwM*Cph$R+Xh-GcA2iO3omIER78s5{b=pAhf9(?R
z=jTLCaD?TmqKr8O<U~<+4)>I!4VC|q!DCP=dfL9nazg#oyW^g{NeK5d*yo|ECkDoA
zntQ_k_T^y$F%++7ZZFRcq6{D9EFXT)`{k$vq&&@tPha9Z|9HA1$islU8DzR5%*4N_
z8DoCrp?%1ld;gLWA)zb^F16gFV+?4T=W$cQrr1dXp6_NT3tcXsDig~m+Mf7k+r5{4
zaX`hIACv4>RGTPbo0)OUixoJ^mShV!I=(iQq{0#r8*6Kg8efTj75)X$kJM+F@I}UK
zUT6J1kRe_tI*P!@&+7A20@t?a1+@I_mNWPu$Azzu(L~LJHNe5}vc>A0`ZhB2cvGWA
zAP_fmED`+(u^vqY)Gl~k>@lkPE7s{N6Gk9+#3TqD@!RmIUBVFrL?mpY%{jeTz>Ul#
z^|pRz2~Co5>*5Vwzu~?e8!^P_RU&lWU@N*MZCSemvk~1wkWh+dwBQK1rWMVqOV=%v
zyWG(np@yt5?UDMjtFamAdo2)+Vqc~!LSR~pYD@3!kZ<Na$LxL#pj0rAjQqD|-&1$j
zy#(nmP~x*E^M5Xg`E%7amTX+r3i*l@v57pn?=H|ni%Eevpn^1ymfX-wy$foPXGw0Z
z1mw|CuClWHN>umq_zz#BBQJsnet2!E?A{p`ONN@O<Arp+Ey-6;LrrroraY~_4`3M{
zSw}L+i>d?Dfmf(G#Pe1^hkYDJ<6WdPOs0PO-=ay0Yv^)oC(`N1#}Xo#K?6jVUH7&Y
zt>ZCkQuHIG`z{jqel&L+J&lkI(N??MwcMz<!|5hMT0kTAH8^+~^b&)pnX*vkErQ2~
zPu@;et;%C0=sU%49+J^q3X1pBn*(_u(Z-0MEB<C}&nqd24lK5gXP&vny^1Gezm2Q_
z+4}k|da748EB;qKA4<mdW-dcMU?lj5m76+sff54`Km3OuVccWpy=)w?d#3TqxE`h}
z6l9ih_x!f;a;+y-2LrFARpjx26^wNDnukYzyUV0&I13UvG2=?cXJ$kgj4+>EWwuRE
z#tsWN%OvjLGsAyYLoMu0+RDt)d<{d{D_#C6oX)J(P(ujhKG$ds$ObrB+81&6HEUZK
z6)Ny?G2BiOc@i>AX>0zY@kpJT=631_c&}0lHb_-=K+_#y@r3`Ov!QI~A~_f2I%EKS
zJ7emxmhe7MD5eWcXU=kDb2?}pJ=Z2ixgDeUC)Cr0Dgk+Xip@@Z?fW>sIrSfsd3iKl
zA^E2N1so#Rx-^RoLChMgcI*of<7Fc`lfSl9eUto$^I#>M(!@{8#<(NM6gWSTnNcQu
z#F(n@U$X%u^XVuNW%;9#eG{+J3*S0eDbzOK*JU^!PFZiL;WG-ghu)JglFs-`?gx{0
zg^83973w0~@wfomO4tTCaGoAw5ySDvK+M&*B-%3F6upN6405V@Led1c)Ncz<5n^(<
zPiF~QP0i%xt-8+Ans2bE6p^{j(LYwh4VMN%1BUMdM~<t<Q@y^cy+VT`F&9XO-UmwV
zNQSFdQ9@(h9NntM?`(8xNGQ6_Nxr;hVsb>=lxfRktQVj7Tbz4GRfJ5l**g7#AP{Ll
zg26O4>0syPBWE01LOh{Vto8R|+hzwxjMgaa+4w4znI|@APC<L_p3o7U4Yvm^QVm_5
zOVTvNz*$GCz`?SqIMSzE7St(9XFgH5M$3pE98RZ&THMK=&w-ai#*K*W-HmlvcPGZx
zE|2{<GyASOzW{5A0ac5FB?W6uG!rdIrEI7rRrbvHHSzD#WAq@ZMp<5G1U+3Ot38^0
zkWUd%2mn+s+69nZC|8j!jec@6cB^Hb=$2uG2_CMALSGUF22lMX_H9x|#-j&=|NN3R
zm3MsbUbp3kXQaeGa7qqy#C07rvSnB^KY)88=v<VG3e-?(hSrV*cU%;r4bEmZXYHnj
z=YJxI7X?am{GN_kF@k$GTAc#_;Ns8cBGR(=(_bTbRFcwwQMtQurq!C}McDx^cmimi
zLAKxbb3<jnqHw?7Ob(Pc29cn_Of$5{HnwD@ef{P(d^BGojR$WAu=r_gjLP$CQG%I$
z%4-rTB?K(yt0^=2p#hk#Ngr`76lRn^!7j&$f_FU~%2Iu9H{5n#WSYWDG%R%8>}4r-
z_j-3kt^W!$v8L)O)(h**BQ`G7xTp5~*cIxL$Cbd!4BDqq1iG#K=~<?S695vr6nLB_
ze5lVfgCMMs$o<^jHKS+&C+0>9AS9TqsXP_zV8f-mKeyL`)N8UjxE6ah@%CHBRjiYY
zIAN$UE$g3x=1lf{!xt1ZRgVcz_({~_;ON4pN2KG(x>7`@%V6tAaX}iC*iEyNuh{Z<
zP4>B7WT23ha(Y<Qv)`n5lm-6SK+9jP-Q=@WZuFraml@kW?^>zwOadZ+-><%ECjXix
zzBw)Eg|m*luIPw7+s=KKGbMdlsp~TSU^^P?aC>~6uI9b%gV=r|dD=H2?R4lgxOouS
z4y$=lE(0JSD=3f=cism*huq^ph>fl#eruAl&Qgel9&OAcjR<Alj3|8+Aa?D7u|RnI
z4cuE_e%8kAT`@G7_haq-o0?hACI10@peGt_iAj_kTckd!D#<{VT33EZe@1Y-4A*a^
z_LEZcNew!=aM4z|te477Rk)Kg#_|Dfz~k%pOW!YGvYaJ4-%?DS6ZBjA&d8C~ri<ER
zgh~TqKZmiqZvv5nGo+gIxf6*Rwk_>k)uVWD_8t)s7{ATQEaE!nr^~>g=%(4pS8+Bu
z5=G_(yS2XmS6Y<<>th8PL}|H=vh_+`rF++VgzfHwF?)POOi$iLaTu-X)KFZf9*u?T
z?)}Bqv}%JriAp(<V<mQC+Jv+oFd~ryN0m6QSZ>#fdiklAyZx}<fs<VM%#R&iP`U8S
zOCWg^;(=)Eo+T0tz=;EEkv+9t1OqTPasza(9LhQ|<YBtAV9Rl<t4SIz+SeaeFy7sb
zJmT+NaXYej@T-2JltU&JP>iL7=i?Yhzm53*70oyBuUiOR%j>5$SfU;Qg4%7qnxL^X
zxI<aYkVGyg07x>~FR3wjB?`nEIEh!feQ?EB5fxl_Xbb9SubMyRuRRP*Q$EOTP&drs
zY?gCRJmNK?Yq1V9fe#~p=K6_U2~;O55mu0P&seC38CLy+tv~Wp-x`q#x&>ZM7-Q3m
zW4eD3)V})uA3+zGj}#kgL19Tb|B00v_HN*#OGe|6oD8pH=^tCuMly2~^7SO4!t)?g
zy6@oxDL?L3C~3=371z7zURp$Z`m{P~37?4aXWtd&fX+CBi)^dH8Pk?Ah@fkZ()!`)
z1X0mZu{ka}gkauYDfc*`#Qvobr##rtiB{U4PY809iCvoe)JrGm!V8g_?{9}sjEN0=
zJ7<CK)qfT-JtwkE7tN=pto<cUp>$GABFPts7amvF?Wxg3FuLN{L~fQO2Ji_C{H?Zr
zfx1Mn>%xiT^c%1j?7^kj&;Ai(5oLtxTKUAKq*cME;>S!-zWFU`$Q5Qp+Y6LFt-d-*
zI|Yi&1D5sozyVt6Io*3Y(QbMUH6Zt&!WxBbM#)2%rL4sK&J$q7R-gKw`9Ij&^fvO1
zmLCl&t)><H@>E;}-txS})rj3hexI>&3B9Lw&HJr;em;dp7)6hFS<O(ccpsv>x)BpE
zDm*>V;wiw)K?<mVEhhW;@iA7j3ILNZMLoPF&x{hn&$8>Qc**$GE$ELU6tzKjmkL)s
zu`UM#;VDh!tL`#Wf4#+=lYGrdZfsUx$JbO|r4oTwXfXkrg4`=3V%Izv8Kj(cTDGhf
z58eQ=^xq<-^r*F|u>!{;_Em2^J)jAy{&s94JOl#v*YHux@HFrJNuq_KMj{cd4B!GI
zyQy%L+U#GslYhy*L&uRMJLVUA4D-9ylSREQjKgNu|8-2>jqtiW+McHya%Ket_zkuj
zm3$1|$Ks(pmrP4EkMi}jCyvG+`NH5bpZ;|q0Y&){coHggK%;uf=ygA9Gjg@-z<06?
z?Ix4=xt4ntSo@+9oy6cjU5elzD4ub<xZ(T}<I6I7?f4SEItGZCqH^<VV-82;is}N|
zCJ#Iem)VB}DuIbW^(ifRZW(vT?%x(jWuuHGjsQB<=Ca$*?Z803l=!O9dzPD^C0fKr
z2m~N1WJ$ToKe1jEO_ejA(9L6h#ABVRAa=9%1Jyi$K$hq6Ap+kBGrPmDUB7Lrbp7}D
zT~(ys+@NI%xgYT&n2Q#8OCR|4nQ!zGoaEPRz8-8C*(dO+(WTLQqg~S|jv4N|PBNb0
zjg))lyh7kNehT(miL}8oD{hw%^MCcSz(ND4@xi<I#8^W?mgCIKl)sq&d`dbMmE%a8
zZ%Hd$rWX!{XmXV0v02hdx+S*&R$ekJ@^R2dY1h%ElrES8zk8r$p`pt`@N(yoa$K8(
zjhK6o-347bbur0FPyBz2^m{=@pz-rv&bObZwz1EzCUD*?!EiX3xSDA}=rb9C8xwGC
zNrz1eL{|I#PEL6#J$&j5zf+699C`UqL5~2qV!2i!hst1Z=Yiy{?8%n+gz3Y*e1piL
z6O(<SfzEwKP5Di6F}9!EG|9*jG3lry?fcTF&yqX}`=^~vl)#HXi_dK&J=(Y|Z8<*}
z30@Mwy@}{(<;t;;0%r<52#kSC7Bu51jNyy-t#x>WSy9ggSr6AII%4s~kygw*$EfUf
zCY$#?+H2MPU7Dk3nKJ$X=vT#P&+<xyH8g+ku0O7RnV>45%*v>a)aJH8>pmXiNa08~
z#<?i9?rpIxfwN-`_yM8u7UP^;zQna2AKg3DlMak0P;y?5eH=PSdr}C999F+r^0L#3
zC^_JpLv#E;X%i<g;;`yFp%_PWU!s&&mSJg0X@WSs0r!qKjk}nBa(sb>uSsyGl@a`R
z?7#3vi8E>bKryMKl1od9W<sxngnJ;(4%kmLe@X$yzeNlN(!>2}*q@jjedoHB;K~H?
z;h{>mbHAOQ?yR})5R|r$vm|$Ig$ZnnL#t*8KK5}E+@@Kv8^(9yG?EfT5PU1;lin{4
zLSUjS%c8>N;~$k2tlzG=zm*BM6jW_WNrM*T)#Gz|3{UBi_BFO`gWIeuu7`n7qJ7!+
zoB49j8dB$kFYl%CkKkQAyfMq60G`QsWATY0(n^Qq0MqRiwat%l{@qJH_O!Wp|2ASV
zp~0AbnKEINW-+Zod_NkA)8QUU3OqKkv6{WUv7umiWmN6w4y`-SB|;BV;-wz#^2nX)
z8^1f-0CkHspK8Awr2)AoVa;dY`-(cwmj-#a^qWnY@txS1{hg<?>b=G>2m7{0c%!)H
zbs=y!>P<LW&QMLJHU2(FX@TL%w+0?R_@3OuR}@{Lg&|edk7qocJn&~JH@eW7osFA8
z${K&8_u3o`BS;wP(94`+cY0*KMQ^Pn!zqW=({HA?)YFG$Dggy=&6wNiOg1+<AFBE@
zZw48Wylw=jke|06tUZ<MZt;wTJkCAv8q;y;Hw%GG<FW<SZJt=!e<_wD0%!MN`hJPm
zeHrurlz8sVi+qX5ww`QtsWJQ+HeWESuU#cztrb5u?LN_a$E%6jk1*~Pa%#{$x4pRQ
zkb}<7AKLw!Ny9$X$MW3rj_>=MRpDu#0WZhGe0hmuS$N;@)mS|*4E~<ZFs$_6CKXO<
z1gc-%wS@=}LjMsBXkO0zzU&v~<KDHjs^+$;w(=)gvt-zC>72mu{YNq`RZI2l38VG;
z&9cBZbIPwX$TSBPKN?k&W?cZIRHxaR)9Ch}tUXnw4yeb9w9Y0HH=?rzg6qU0_kXX&
zikwQyBcGjclOqS5tHEx2(=0Eyz4ua<7wa614t|PXbnX@1^gG5j4}b>F?#wG?KKQUL
z>D3%@JC#ak1x48sS1KpB=kC_yJ!7{-np@4coMInIt48ux_LJd7=zo%8N)@-)@Mqn?
zCB{TDMLie03HEs~sahW1-369Q7k;$b8PLlKc|e60hl^hIV7!27_os@JOqQ{&^m@It
zOU+BY&Kd1I<4B8XNEOGQ1fxpfLKZXxc`gJZVp(4YLUrlS#O3!<O<<Li8^|ViP~L7q
z$w48`7Ze*B>F;mrX!*VfOSO6<2k<HrvugZWKTT3;LK9K#TC2QDfk??*7a5I^N#_yO
zlk5BFkcI9a+tRGJPA{oW`Bx4JhUDGs&@j*6IQ?(=Va@q^QQIdWfsS)iGG_#=im~(I
z#@c8z_?^~$u5IHC!|pN^Wg8orei*{_?;yp&F8Qjk*|OneM3YNAYj1+?PD$;rZwV=n
zzRaTq6k(wsa2;IUu}FF`nav--XUmtN>839IE#ij3B6`Z33p3V1pDGW)z||*)hzf4|
z#5U$te~Ik~F3J|qKe`<wNN1Otc$I~u5`XtwL{mgdIA4xb^cB@R&G$qUX|DV&YaYo%
z&L(UQ+fbU3rKLh<U@cfzC3O-2z?eQ_-(cwL{I(P80|tK|D1sOdPghb|TlSh~KPvqa
z>0sb25<>Dt`sQv!b)&q;?!Q=CJ{$~>URp+ihe!Y><Q%ep+4I|OL&zo!K@3$9_XK}G
z;>>%EQdotYO8c58TuoTLmkppGqebiz(tbA&68SL)Ym2PBh`wyUU1P7^o-V&xxQ%&I
z1`AGU3bH%$YNuf^IXQ&0!Tdb$I7(3)vaau0-GLq!gUc*E{8a*|5H8aRMZy`Da}U;Q
zk0R&2Y^ra(MQ4s*7$;THElm5vNno7%&0ev8lfW&rjt8_-k;O9IROA2H67|DhXL*}G
z{w0f8xhpks8E$F1j1p#xIWiXZjBfJ5^IaMxafGqW`53v4sVk*MQwPMFnyS<{K}hw@
zd!wvxfN`sRTTVb-x;RGfyoA$6%2mq}{0gGHfQRm0M9P^IUVT{HFSh1`bO^0`(macn
znI;iM@Bp<><Rc~X?fKpyA~tHnvI!PdWl@Ff{c(Mz{#_*v)u754f48v>s_2kCwF=JY
z&PSrL)N>3YjsOL_gI3lipNJ2XrOul1Wkw+f?&ac$$3qPNi@V_?<6r^H*bU;p$q=>>
z1HvZjJge#_N_vj=kpT^tx;Xz>wFlov3AYhs1rZwWA82QuRk~@988N#nN1&YOzQHa3
zdf&p06k#ycU~rClDyg|%s8er0F|}lCP@)Z{v-VzuK5T>5<=`+i$=M^z<d&;ru`drv
zD}9p7>g~GidW{KUn8}&u^k7nksXyJs5{jRl`tHwI4aOs5*?ya1z}BK`rSbW4Osuwk
zzP!7-44zPQ)BpsgLeCxhiNT!Za_>?f*CXJ;7qQ8#<^K=Qzj13VkbE1||LKI?N#Dgz
zIi7fmv}!*%-MEm8-?F^f%-T%3{*>l+mq_<~HYo0Jb>M|QdT}F~I?YSYm@dv}bQ}pW
zFz@r0I@tf#7ANC?Q{`j!@F@?F3h&0m;LY#5vfPc>xE`I4U`@X&vyP-}ERVw+47+d*
z8XbB2sGh5Hd+{W*%UR)T5f#vG^F8`sWtR@~8)B#C?QQLfjP@T}EoZ+c4rjM1ZjGme
zhKp}sp}+!o1JpBW<=)Fa2TtF_-3=QXVIC2VH8rTfxH&^ZfK)EC-4LZ7vc)gt$j;&B
zE7W-upbGnrXAHR~IRB-K{$r25*hy70%I93yIb>rw9p~$V7vkmchY;zmv47Gi0=ZcH
zG-@I9F(-ru^j;GAl76&@L6#uC9|fE$kv7K>zz5BI7}0z>OgtK##ulGp-w47^V_Q0#
zHuCjpJ4->8o$~7oo7i0@Y29LCCw36}NAQbYir!Nzr3tMo>~3U^4m{^j8Rl_h?YAiE
zyhGp#8%ummkQ_Q>=-qTpeH?NTNV6{LaTjc#b_*38J~BuhbgN0&^W0t-1FuH~#hhC1
zMn{OGvG&h*`xt>KB~YjZM>i)`Nsmwxb)F$$r<YPHwV!0V&ufqKc|0#JGL-x$z7jU6
zO_Psw{GVRRH@I>XEa__@Kz+APMRe*11R#?(&0!_hWIB<((OfR`TY$wQ>BK1O&x$6u
zr8b-zG`lss-&MNJ<t*)>83H{XmrzrxLCXP3zx616s_fK!t~=CSlew@PbyMR61D!a|
zsLmv7%{S${Gp_IQYMXWX0v;SV?H^Pci3At^G0=Y~46@<M7=4?8k}(Y00h#Mf_oLZt
zEaKFb;z6mASb~Eguu=$Q19<su2WT2XS;fbTS~>dK00jSvyO&zMzs`y%Yr2nYu6#9Y
z*<ebMtaZPY1pYJH@gdD`b5my{U|!8nF|1wUcOG>YTghKFZCC2=D}K1OtgE7My;a;3
zs<??3)@Os1_A7_?q*<m<poqUE`vs{gNE#3&pb~Jtpqr7P??e(>O4r0%Bicd7q4#W5
zm+Xvgm|hj6LwVVX7t5zeMTO<OPj`cE&#GepIHe@n2rhx=mL&(@5j6PcH#(beQ9WN4
zyIpJsyi|#V-!H)R;a@QlA^Sv^gHLsmemx4LH>VcpsaPSw)$>PxMkvdb`kyB?G;aGW
z#kZjiTUtq)U|p}b)@9@V<bL(e^H4Y^m}z1uIIpXRTP|1tiM}ZzqJ;t#i5fFPFcqfn
z2z=owWH=(tAaZh0Ynt=ie?x=L2_;a!FftKYNz>#PZn`{gD1Td0g|}}0wF!xU=e0hy
zHyG}uiLnRu=sn7Bh;pjoG`}Fdv!-#0vnYegTi{ni^fepVyOqj1R+r<N-dlesR3mD9
zm6yuO68TADe<cb6Yg!{id<L<`O<<}=d~#(&Cbq{P4PwJd2Q!&O0rFb3C?c_?WF{aG
zZ?hU~C0biaE*z>>Tla9W>4-K)UAhR`x5(nWp0izSdW$kHF1DHoWrOr2;ljn~@&VXv
z!;to}?n^_JyK<p97NU<|6OUgi6*^+*3?Ag-1|ge9mf(w0fQT!Dt#yUCTC~%ww`fXn
zYe%j1#Dvpwgl2}sXR*)Bi@5Pam3~zMMZ4~y|1|3I$4Tu}z&RUh&RMNk?j%v)?=jcW
z>~TjZwrA8#KSX8dc!BRrSBlPOK7p;OZB1NX&jQUu>3+zhyr%DF>>dQ*Vj_VXb;zX(
zIo_JSRUrlqS!=a;rg_c(E}I>bRiUcdPKf);b;qbq#k^^jYshOCbOAWZUqIUnQ9<E+
zRok82*>fUKBEK$&)hXv1&qdgv{1eugEQ%@G{a2OG^*wXuG3b~$Zv(2H3pToh28)cr
zYev{e<VQ3a$$qKuIJ2AsLNvmpKCza{4-AWhV(06z`y6=Ho!wS{5X#=ko8yJb{f;7i
zLZulxI-bl`%@^msa=c$EaYC^O>l;D$^5k5A9V}#~b=H}vTS^YGz0)O@*21Z`O0#~2
zm-{BE`O_B3P}xZhi&+@oFrY>(h+qjmQno+gN&jLJZ+iW2Lp5>J!M6k+iEQybA<igG
z$gTS1+Cf9P^nkLzr`p<6KqN`O)ppZ(S;pYC?O=*vBqr!*TnwJi-ek!Z-fC;Q`)J=X
z!!_TEG3s|BpO9!}p&gL_)2w|}bh)RXA>pKr0+#up4UogAeoM;JMHEB#{VVG0Br(~t
z;o<{@RWdi)*lYX)(Ellg;`^QWQ~uKqQPcgOc8HPgMHiIdN%#N6MK-MwOPxW=U$AsA
z6hD+|C?>Sb{SdlDXTqSEA{YPm#r(u&eVOc1!~;E`A@Ii?TO5jMV6eX7U#3^=(KbJw
z)X$50!WpsTP5Ivi;dG`1(Q1Mn-iO$?#VW{dUm|NOB$ja@QE>~Ph#-mNJ6c1`0#bL8
zb^;qmJmr$E@2Yme^?T-@OSlKhpe$oGY@`wSL5|#*;yAw8FZzoULOkIY8Pjn%67i)}
z{l}6W>Jt9?%jS@lTQ|d>jTNm#PJF*w6sHB{j<*k-*5Fdj_2a$#24~eH=jC%yHg2!H
z2%3x{t~SIdKK>vjhfqvt_1%^QjO=(dj%>Ar?0bGs(A!D$zPa)z8Xsv49exN|7Z+O6
zQRUPUKl+X}fXYb+dctaPfkRo?R%^7-e7>~5?UUu~$KcRYt<o{R+o1Mwy?Wcy$&MqM
zX6)=qi6@i8KsE9mLTjqme|l?11n=)oV8n8bC54`TgdpCv2(x><>l?;Xf7hVW&f)4q
zKa7KyF0KN+?^GbTpU)om1euxNw5TZ^rg5oVeTlQIZ`^tJb~Uv$0s|b;q%9c&;>51e
z`c$)xs>`*eJ`x%qvrd4^9+drc4!I@NZrA_@HNnBsw{F1nVkF8z@ec`yiSEx14LDqy
z(CcM#uu}3QSi0JRtqHAV&yzQlJ+KM6c}f*i3M;DrmHQUjI^~nLX$wx^x;4UaN$FVt
z&Pmm5(XnW;jlm;3k}hAJAQFW@KVom}CA#H78GhaYQ}k-~GuPz3;-Eyv;kR5brwy=B
zfR<wKis`>_?_yPB`{N!wjgu_j5w7@JOz2OZGu;OL@xz{Q;*pimlR!y=pCB{#+t5AL
z32+inpvA+yi-KRXMGWN<G-5CAPp+`+`51>P^XPZAp}vroYP!sQG4E;+EVla%vHfn|
z#|EaX2j^1(GZJzEND26^KBVP~l=+=c^IXE0_g9ivk#n7S?_xastD0^LpRc#-hdV_0
z{hMS2L%wspFLfHM`AppRfmS04pWmTbkzSqy*eo0l65!CmA7^`)_Abq90Nw2}UpR@b
zjs1LC*2a%m*M;*QVbo&1a^QSx7#|CBpliq@NG4%{O(l;%KbSXgE&km>;pUK}O#LY7
z=hKdHY$ZPGW{US;_!1p?0+d#mw!Ifo*>HlFQT4NYKK3Xx=%8w}&z^)%lHyL5sH{iV
zQlF5F9w7Upp0rr!ed-P?%I&>cfx#jb4Y6>zdKeJ+1*oh}oS%=)jNbSgBl(?h{7G(h
z3mkpT=d=<%WB8h)flTzOV9O;(hB8mL9kMwruOktKJSwJobi9HGFIkOasl_Se(vrzI
z%21W_{aWwfkgZ?Y-lndle*~}utqX+WJJEhlYJk+i+X|(+ms~(05=ByP)FJUvJ{*@t
zA0UuB_vEcydLk(s6&v~4(`<o4;R*T{&oB-8e<%$goyxpQ?J7FYYc+<99Q)lgt?@xA
zX-rc?;u#A1GRWL$gJsFtWG893Ku3h=i@qaH*m7rH?^fI3dBIlxoT$9-J?M?bJblm3
zNSf4ot>h))Cc`&Un_zxPiYeT<0))7e6|JO%<o+Dy>vqVKb!nHdzRPQar^O`$jL}9z
zcyBLNT;7G_UWPng87D7{$!@~X$jtVVDcw-^!O--QX8KHDsde0RJpzB=Dl8CGTWVX;
z@z}4;-=5e^MKO4ZMW5ZX8QNt)v*|Dr=8Jk=B(Ie^;aW$=do+aqxqxHb|3JavGWra5
zNAiX(4t5fh7us?>RPI#)^Zbo>R|As>Q{=BSGd7fK<Qw7V=POlq(v^hD5W`{}-=-TG
z?Rppmv{;`!1mS<Pydzg9Su&G;Nr$7{sv4_rXpAg+zV<=?5*r72EPpGHSIds~KTiFc
zid5uOgrDV_@}54LVWCG9hyESCUBCM=!p6trMfDSKx}yhwN`2b)sPH_?H#Gy4Z&@}?
zi-J1yh5g_uJ$K5DMwa$r8nKfr6gUS~qXm$K^ji+|s8C<D(KjLf!J^i-ww6}Zeuc+O
zkGmNBW(F-)pGO_sA2+=IcFk4BTCtor(T-?tqd2Se$e|)bBICmVjHoi24i@iUlXxRC
z7h&F0_niYUl;%?K$<?)UzeUsIwS96wnf^rnhCSqZ)Lc~7SjHyJYjSj<h})i~cqo;B
z<!FNZ;9|poC*VEd-lpWNvmJMdm~s|4<!`2F@Jj=GBLwAG-OWF0xtFuLL4ct4!Ryp{
z8M<8OnMs@=SZPLy79X_Zri#Zc*~nA5pZxm|HO>Nt-!1Ld6HSL-vHePI$O@TM*k}5G
zq^A`yH&o-#vlckF*uR?APTJx>E?Zt)SPk|+!z72N4GVeicTyS;X^jnzEuc?(yP3P>
zlN08Op$ef~naLYyYId-keedgm&K9%#VT>(izue;kL4<Y4r#AKvbVBM_M%d~z9I`8}
zO<4Xscx?5Wq<N?A*?YIla({jg0-q0{zTX#?(mZ?QNe1#qNwf@mHiT$g$_#1|qw2FO
zRE3Qn>3eqt%)>!v%o88*#7LwDY|8_^$tgu9%gq<`i_mHPkDUVII;Qtw_RB+^7)5mC
zCO)~}!GYsE*TkmRB3s_G>oXw8BYs`AY5sg;!geCmo1pHIex)aefr2!DvA#Du%ja7g
zDT(m(jrA81Ja0NuFR4?npk#SW=Tgxj>f>|w4$m(>3*rxA$Nt%CR|a@Y`UTlmzk=`{
zlJa_LK&$aMpOzv$X4@i{)S2$*P^*VjVyi`d><LTua(@O35dB1<y&Gk`LK{T(DyY8R
zt$d1V@;WlCe!`me<h|mj>`~!zR2>ac!W2?GYJ3BA%brp-`1T-|#o!YMKf7{>+eEb0
z!+5L{^jYlE?$bB4;fJkXuVVT8R&evJ-b|1^3j?y}_D<b~_Hzv8%bqq0jXutN%#N{S
zp56+LREhdWa*YoBWmMqM5huJdnuW}*xjc4<-!<yQkD?eJD{Q;XM(<yX?`p=F2*o!?
zn`kWJt`8|cu`SQ-s0u%(tB!h6I41m=3A7kNZA{j^E^;t3FxiVyKC@h@$qNeCkas6a
z!XmbjC)uAKdbkR<=q8Q0-r#r^_XywwUeXdv7D!0{LJ)HGp#a@e=Xhq8Jq07}zaP#Y
z<Ld?88FYt#3t2=VVcl3z6mP#O-&*edewH>NB@Qmhb~hIpqim9aAH5+>c)Y#7rpKts
zZ(aG`MU5}{u>`^kuxaA4s219<A|G{NOKDv^1`iL59{>GVYtPiXZkO2mN)&g2(3*85
zekL#c#|vwqsPkYpbh%Qbr(NbNcExY?J_YDh8Z^w7#{*G;43yA?l`OC2K^dy!{v6SH
zps#MZGnK;JmJV?nmva<nJXxc&4RYOlv9SV(r;Q)tIq$Q{pIz%CD_fh5NcdZ}s#nkU
zL<X#+F*O7vLH-~Sny2P=>Y@bw2t4~(UEcWEs7l%MB<!}dh)QKkNlS`fL!tZ&X$I*W
ze{f==JFtWVr>_4?CHKX#Q!hWaXl6K6-TZ5uZ=I^?gOjBz|5}r&69B4}z_Tzld}wGy
z?b%1(T+-Xdvq{HW@~S}iyt|<G&wq{s>ILpxD5FIltHoy%AD(m<*-9OrjE#=`&No~h
z&a5kP)IFE8-cJKe!KPQ3n0;~FRcZ(VX?#sEJ~^5!nlJ&J+T5a$c0~u#e_WY-)+gKH
z{YZ+(zrN=FbSpBh_D&uHpO)o?oyPXOrEcIeidy<KdwN!Dq%<0}?*eotcp(%c^=+zA
zKA8yv?l3uQ#A&X(-ptDd5i*?>H1w<d%ovWkm7jj9q7B2GFIt>R_`PpE?ueekOtlSq
zD?*Um=U0_Q!WP0HmHr;0IDH($;EznIm)1y;*Uw?!R{{+X&=}A!?<5WRid`iRdZFm&
zs+(j3E{U~_ZlOKp%p!IEGLh?yA26aZ^Jr+5KW(&=>FBfk$ha?qWhR^MGpKL!pMKr$
z7H_L>>D%j8#_jS#U#seThtMgb!S$HNFps~D>K6`Rg~$(tc6ZusJL_?`wc1}j9<yCO
z9scI(xa-mJZ;=@q_fCBw#;Z&(Mp`&azZuk&vx=6taMInqJ}D>}iJbfaut)2sEk--+
z=k|Oi_(_-E>o54eBAm;1UF`l>4h<9QdG{*Q0M!SAvrL_28w3V~MH0;eJIaZKp0v29
z&8L=>zYp8bg30W+;VutS>>Hof2%kjM&0lXT{-?6DY-;oS+IEp5#jUsnEAH+Nh2lYq
z6{p2DxVr_~QrtbbyA>}E#e;i+;BI$*|C#3lJTLPmbIoM!wXeO`ah$Pjy|bz4_Md;~
z1pC@Ph0t7v#4(dWv8uN^bN0A5tHRbFqD*Fu_d=(_{>f6i@-qE|xqU<YN8oj_yB@#2
z2jz*7*GY`t=-?u|()ju^Uq1RQDLlhs;{PqgoZc2bT?(S7!J(Y5g@B@6x8G$f^tX*F
z%9EGK)onwTK7{&cZ+-XdI#tJ*rt9yTF%A4-|HMO<fT-?D9c;wwDAKXSuPee}g_}?|
z`V-|jdWZ=>7m}YZ4OB1(@)vk7l|OabbzUsXrP!g8cW9xZUgV;}&7FRAIGETMG4Qrc
z@Rcqr_DE;dhCVJEZw@J!9ZD-9!<^q9?2uEH5}_nyZc=qL*#3$=-TZa@m4Q!GdR6EU
zY4dEtdO*_8qNu$V?|#l`$KVS~E7m$SZ(prQt+X8QNkep)GiGJi&2G26{L^daEZycT
z0URxTOx4*qHtXF&(amaJTQ(5U6R32nkJcNKyGF-aYJi~Z%rhUGfjt8)G*Wh_n4ye|
zZ5=c1H%zCvqvVahp$pm$Mk{QSf!7M&wdCw*V-Q<*DX?w&RYHw_PTRtKRD@@Cq(zii
zt?u$wxSHhJF;K4=a%UZOB&L8KibDYJ`ufhIDyMtYH6BaRStSfkIWkwNG4l&t(SxYJ
zQLQF`xN_vv{wyCaZ3~tpf^SRmTuiH&-0V_x6!%xw6rX|$3ILYPtUHWr;FSzYX&~<H
zx%C-1bj$?Gwtn3H8oNz*VQ+ZFS^gczqb6EX5kJu%>@n6+zWdPH@-*=_8Us~~T#}hK
z@N*_gzQ<t_Jq<NQ<OvytK1Fhc%ZHBvS{3PC+x9Q0{t(?bPj~v+kMhM7WcVRoKnBg%
zAgDwd4&-yx61pBK92Zv+Y|mm1x4pD&7t-tL`2FFqX0wYHeK=s~xP965x!8l_CVVZ{
z_K0x8tkMprNd6zfnKV`j*+}r$kg@GK<UJqBgw+E+_F;;cO$h*q;#NIBAn<O(*>2b^
zx|WPHq$4pL3fGtHr0?`l)0fV5ivu;yzAajO`U^J&U;4RYQ!5P{U%oY7%`9|Wu8HJR
z(-Gzn&XqZwD{+HyJ~^i-OJi@j#631hjddE()`yHmt57~FEo>vS_;W7Y*7Lb0Gi9*k
zn6ht=38plebV1|SSqhBbd_+mA=y(3l&F!asQT$W6(CKy$(Z6pXFhQ1N`5w<2w)15j
z9odK{jC;(c*JjE&C%`l3cN!Le0fv>-!1OqU$tH*A(E5nr5y;}!?O8NyN;W%q`bp56
zU+>u<+@g=_-?kcDzZ^l_9i(>K7B!WJfC<RD0TbF$YULk$o|4$ZTK3Tp7Bn-Juf5i#
zmk(N6<AO6V26B^qw`9MpSlP(*ur(oGT-m-STm930KOTO)9_R>p&eEp4?|?8Okn+GS
zT>@*2p4(bjyL&ZTzw2()&PmDGE;^QvR2&Le553=b5P>jB87lho$Vtn?l+lg~B_KxR
z%i|BP(Tzs6D+sv*OVrpQ)s;aY6pdCX&P7H}O~-L`b=Z(tWH<g$;qF0h&_~1n^QHl8
z_BwTu6izC#KH)!)2mW5RbF#3bP8sYsmHDkGCsde^HRK$tEa~}XD5;G*ImVKpg>AO;
z0SjPieNi@v_gnd&TD;`@&{MWwPoa7#qokn6N#O3+-VD&~dsROYw)rA29lkd|5CR?R
z(!8zwUPV>1syn!rUoQL%UwMJqj8@iI;X5^}PCrMF*a9p@%>$|C_N5b0vDn?GY5<j=
zMB_yoaMQ1J1@AT`y;2t2BfLGpLzPcoLNBY?9E_@=a(V1uWt&(_Az^HzglWJX(l{ng
z$AM$i^3LRJ17&=0UXtUTfB3l6H6k6;KD2&cDj{lWFj(iizhctLD{<n+R(xrEne$F;
zWTMc3`1>L>x`NI|S=4kN8UhC-(u8?-#qWPk>GvY!hTsQ3>$elCyZ1wiF<g4WLy_WL
zLuv5dgHX?Lhd<wLJ-5e=HBN`ChC~3Nb*tamSY<%&m6m>oM=6^zt53mj_xLU<jsFs$
z+7!bTtfWnIX*qrRZsl2C2PpNKBcSH+B|k&mL3diwt&@f;dm{2!9RBko{UfgEzOVK;
z(u!RB<EG$+pVCLuT%|5@R9oMot)jSED)vTrioPIz0Oj^W@lj%)0jowMC7G1x&|!bS
z8jo#;Vag_P&tS-zAD@%yT2%EL(=3kPHW|{d7tW{;cS)YAeMP7hX1nE0-Y?`Xo2WM8
zo{ZDZD03qs41DNAjs+};*>qaDdBI*2;&ndxKcbao6C<K_tbZ{)FZ!d*E6D;Agfm<I
zGB3Qk*WQj_vvgTHBP=3o(PfIm$7y|iLDe!(AEBRH7_-Rhzc=T&CRvmwFn?dQF<b0M
zqWP<U7P0kRz5>nTY5RGuXUdiv!~|-l`MzV}<lMlqzS2Ole~~-c_V|0Z(#wT@^wsI4
zr@qG&5gW_55RCLcq4&nPY3=THxl!`IDoRpT;zpQWv;xxXSwG^@XoKd}L*FO%<)zDK
zqovItLa9eyc^}s^NBSOwpo5eECJF{8SYO;@WB_kXji669J*RJCWvLGTJAQ8B)wqTg
zfj)0_8h|8EoDnVQE9Lv7`ZI%mq#5+An@W{tNTARN6$1|QPnbPSYg9La+9fs9C|ese
z<Jqn)R>lS+#5;z<cYc#G0-dtWAGWtLrFQu4*f_jDF0ESpgBr@q^&$cOh*M~p<b1(t
zd3&C#1)A@3@(@!!eh+7LW7_s(zzB-u`IC}@y}RQk*MGf!3C>J-IVr*U%&TC*u|K0^
zf}G~(aVqCOX;9?~8S=Gm%SmpysOc;p#n2FFL+_-qi<KKvX}`DTGb7dw{e4R;uXAZD
zz`RwG@dtyo=+V7FQdcM+GWiBILj12{8vS42NyalL3#a>6T)j0v^vhY97Ye|3{<X+X
zA^wT)?;DEPpj^N&TI5R6kd8pieoY)&ny`o373Sh>p8cC3X>-o}cd&TX9WRcHtwxh2
ze~ZRP9M;=crn5a~#rp#|M{`;hwOYO^_%dziUVV)v8hoIco9tODo1YCDXl76J4dkHs
zJH|uIgFy6&y2vc=8okc|G~xbf@a2A>Qf0xX(+!2Zk80(AxF-L5UPvX<_+d%5TpcgA
zqP!(O(--EWZ=7YKE7&jl`82e-2$#Mk=OIx_xb!V}a)k5)RX6E|4p?EoHZNl`{meSf
zElFi(<2+}sSK$)C+7E*{(ZZzzZS(WOzBm*IT!+MBAvn<76h#v<gae?u?iK`gXeH4w
z+w#++lyySVTbEkiMGm?8*PW8(ujv2EWNZHjS3(u~Qne`+c9FSIXXJ?32E-%c{sy8P
z>fH7~=|ALqjKjTES@TEU0!YgbU~Rgb@O-&BENqDu=QN24{s@~={eReNU6E4=aQBoJ
zzI1Eo9rrzlHmvAQ_pT8W)21an2KAa{)?y|A9Zk|x!u5+hSi>x1juF3}FZ0^t#hMmC
zTk?1Rl`0FN@ZeBs%}*q{pq``Yw4`h#`F;oqqW#>abcnmkW{@bmgRNqTdHaKWnv?Tp
zi<)ShZ8BM%r9)_kR^mH7aE)~=<p2}_W(~ZNto7dgXzZZ3@BFjj4l}$Bm-7pX8Zs;1
zVW<vNMM@6fb*Ofzkv8Ct6&aPW;4k}yzGqr(cM0_C!BnDq=CuT+8q3!Y%mVLPn31l@
z5D*ZL+{5qvoB$1cx$<~~>mmKIomZ5(N1(F&ceUC9UU;u$?qz?Ai4JImr~!Cdmrn(%
zu2>#`Z36A`a6A`pZ(gfeo5zzY_B*8_>1=Lqr=v;|31BZsE;nqdPa^0o;xzjW-b08E
z47SBlhPGEj5|*c<lUoLJBZ%KRwC2qy9n<FWPk}ESn>#eE35?#=v_aeN<$j4l4(6aT
zV&CYNpDouTGUyjCi1~8B)6iSIvL3A4GX(q={PrkFOw>mrUyHOm+EBK(61DJlxa}Y=
zvpU3;A<G}b&EnI(YrI^Jjk7L51TBb8@jBSlTD^pj6zBWX-?)_H=C|6h#PAzI{WZvc
zM8v3O7;-6BI)Y;1_KG^HJ^EIRfvGe0h6tmQR)T*>D4pDx3XhMir#gllFyHPVBc%u<
zIh0{{Q4)<XW<r7YkbD6v<Tr`O`W4~hD(!zRffL^7dSQ2s`ExQjHWD0VyIM_yK0H-=
zbo%E82StvEbIzRlX^k6~3FTU1%ikrx)V^effszye!*-1NzCUcV?sxLX`kq=r*3*K1
z--hVr-4w-*5H-<v{8JH6toHLrqbHDQ{%2EJce3OpokkDSO`2=cMEx5L_SWVOMGqgj
zQ*U-5=F~=YXAts;T&lq3;1O%5UclqG{dM6E)qg2n1#ni+!>H6;%620zIBt{BW8Mru
zG0#P>RK_T=)G^~KMfa>G_|GE;A7Ur_*1uo1drGF38p_<^QW3E@9}!zoGx1zVLFb-K
zHy13$*N*>Fr3O2qj}Vp>9|2}$wy{d$r|=3x*aCzS9Hh&~kS6BdE*riDGG1ni$r0zd
ze<-U=Y|AifYm)6-Eo)bqks@)j6Y`?jS8~tbvd~zfuO)y#p99oYEMYrOrVxuTKXkq%
zdQ(gA)Yawm!!m77y7@&vM%W&6u2*-YH5aIZ)MSx_SwZiSXxb0S_J@S_V~BL@<_2(%
zkVMalhm5@ruCOY%ErBRG&ASCDhgov5@S83%31k&=CA#qzFb$Y8gbK&gCJP?ur+n6j
zScb?LO@*dr>ImjWXMO>#pm{~#CnB(8MdF)1=^*bI3CS#>z+%>Byr~@v=zrpf04!XI
zpK=_lh{0<a@q{Vtodsz!L(DvKdJZx+V!W@|g;cR$L)M5^Rh=0?@U7$<iDdlR8w^B_
z))@5<WAaIwGXqPSGv-Z4{V3cM9}`LO9OfH2v|3}IgWr5^TfyfL<ZJkhGKn<Q5{ZG4
zijwVMEA&0qp3HMKORcTAZ-2sWLndMvc`%rT2aYLFG8n)GK3{BsgxKJ9JI#;B!Y0CO
z9|!qmJY|3-k{d`EQ-CObw`(gm7UEO-^|=?BZ#i>lECh@VP*o-AD@r)(1tWldjgD#k
zK}+*BMMMV%5!m_}!(R44G`0JHo9DMNDUT~4yOG|iC?^WL^W<<c=8!Nx%F~?e4B{K}
zr&n+)6g$d-ckU_+S~R%75W#gkO%LlC>wRcpCpbZ&tw1EYJ#W=mWT|tW?TCwVq%jHJ
zZ5KS~_XlicZ_m(Ff<)Eoc0DP3&ouUna3KCA;!{(JC%^V#j@Ed2+jk$ox}cAWLT!99
zV{S6`Sf5E!pbkhSHnkJeY92(9pB&KAeQ|F$V7FrHKUySb>|r7t@3da`;o2QjYJ2-E
zl0eP3xq|9^GmNV)WVzfw{1vw<x<UN8+a4;={(xMLiS|whm7MPX8G9BjL99`5gKU$x
zM#@>%PvvYEccJ>Gs@sPY-40sv7BDqlLtb5#lERvjv2ldvj|V@y1W4oI5q?6foBgkH
zDYz9~77WRA3H2zcWY}o{$s6S{99wPto*QrJpn(T6L<$P;6w-mrj^)^~3HEc<$@Ar#
zu1BGaCe<E?A;(<Ir41`B7_b<oOAFJQlVN9{7URam>cg=ii9d_l>Il^PFOmd`>1Qb~
z3pRU>UF}qc89<MvaT{YR_+a8OG#G;wZj!;{F;X<~G?Jngx)%R(o8Pq{&G$ks2@6;o
z_5%^nIT7igk8d`?7dk<&){Hm{^Y6MEmpLb6NAb?RV#L(-e3a}1$K-nHSQ56BE?SS-
zzKRfliD8b@0NCiaPLNSckiejZIHIcS>%|)49<R_GW>*uDH*dNz!cq)(u{KS#VmiE!
z#z$qe#Kt&K+kHG1Um>#0Piq}wKO=`<t%W;Yi=-RhpG{6`<y}m~$1DI|ve8&mlVPxt
z<i^|K)%Eu6gpIP3g$X##eakK9BP71_oI%LonY2<1uF7D~HA|i*C{WE74%)kA$@Eg!
znPYM-mJl@_i?{w9eb8vzh22s(Ln_;cIm9g8_BT?FKvXFv_*g_Mq-vpyfK$8>$y!#L
z5HC`{Dt5pvJN{Q(&OJYw_BM@;#1qD%k!V==&6QG%_nQBu7E_mv2C|uqtcIY9)u)?u
z5sMY0Mfw`Z@GsxhOV(Q6|9WU)zdO;8+(8lH8ZB*5YiQ-W8*FdZMDP=0dd^~WiJm*=
zt~Pt=;N_Vj$9W{M!4vHQ`Or;(Pp*_=$=ODDBkM23%0^kMV@2tJ)=W7|KG%?Vh5QM~
zBj=eVy1zJvbxCgG*l|2zr?j|uzMwrE+R?2008;tk7UB*)p8qn#u<j3QIHQbisSbu>
z=9A~A=hKKPEt8%I<crIv&?w|Bx@rMVvw05G9j&YqoDfGX2N3UVk*4Re0)z!uK7{Q~
z%e*iDlLGaxgh05VyjW)Kcw=ha_FGJwCasT<KAGY=HIc#g|7{h8^phzC-ap6Ma!1hH
zQ%Y&DVJ<|?tXgs931W}lWc75o*s!b7=x^NVm(!D~{OQC3<%dNk9-VxnJV~JtTESTO
zW)RiOr?*#OQCIlepLkTYLlW*!QnDx6k6QpY*W{yGaQ4Kn71sTcL4D(nF^h&f+YT{l
z-7@pp$mycb-`-UkUD+^UJRdqXH0!Fg)&M_~hN#9Uv0QeI+P7CjDaYO%`l8wO+81e@
zBsMvTeoX&kJa{IN%;DYWI$O)%Vhj@%uJcS1eUI;UN_&pK=Kq$xDt~~U3*~q)gagvR
zW0jxubv<$6<7IOYFm~Em-N37TJlCi@Lq1ZC$Y41xK-Mg@!D2-d-s*W+^1uh5P1xDZ
zx=_}nJIER;$ZyW544R0)mYXA+dc>Z3FM0FhuPIo5<s3|8CT<7%HQIFS*5I<=jE`v-
zt*y%c(_sab#6Bua`t5F+J29Xc!iz}rJBD4eW>z8pGu~ccK>W+co^5~{Mp86<2`P0(
zkvNV<88MPGlZyLr@fBNZEM3R{2<F#O$f4xKHvI=2HyQXOm?Pz9B9|^DLR@F~7txrk
ztMG2{+*9c3D!3Mmx+#>{76PN$I^<Y09(%7jq2*Yy_^^a+C}@a18J3tsGkV5DnXsyi
zEcNyR^XQtj`Ja{I`p=t91rkwir<v60+##uzb}>S*wI1vKvJB^%mra3@5vK!Vs5)Qz
zKxPI(x%&x-3=2WCh;6hn0N5$<-9|5mWU`)a+#XO$g{YP!XE_8PCpi{~AdTh5f5M_O
z<*Dk791XVh6f68NJVB?GLtFtlxQp?<{#8(I>vpkiIj+uXRt@t;heRS#P;ovb%XpWa
zdJ30Pu{=JH(2$iHQWjNI824mVzT<d%o;=Ow$5ARFddw{z1HYikruCQc99Uf^(w8cP
znh4FD*JVlT(2hF04f&DoQ*w%$mIOYl?3^rcGe>e)$?T-J1fKX*KY)`TS+`t;+iqol
z;mIsxo+5kX)byt1*akU;m|zZjd(20OcPS>rCAaW{cUnf6CRwPR2=+ewrIo2y7W-VF
zFS7;ROoKQKJNuVuOz-*04Uk0oNVFL3Gs&2L8Gh=7oE5ir@4{k%K!#b1l0~(2u0FP#
zRfF1t6rL-;Vpogd0$)qGX)Z3?W*d+_err<Xe<3W>bl^3cWsSyyFJ2TWMw9<}NIV~*
ziQ$ZD{ys+0Bst9nhuif+JFy#Wo^0?g<R5{X!!vmxcbu`8f1~fA57SGp634IDCaAp;
zza;?If@S_G7n7&jxf$PbT%M!)+WYxSU-f=|?El<Ec)H)a<ftaW={W{aH{+IGk6!PI
zZrOQ%q!d?k_J9}CdGlP8$vd+hds~W^51*e-gH5G9J|H5Bm7Lv{T%3_J4PLlb)}87m
z>5Y3&j?IsouNsM;Z`kKNO1HtFxceUu1Kd~|^xTRhcc#Vb;5k~SsfFAZZxddhLo0G}
z3XQ~ZBoc-(SrF)~O4ph4UCZZ*ti_wXrY-J2jmOlhiis*74U^XGDF#o|;_$zr)y@`6
zw8?qBzn>RP5rBm6&UEAVOJ<k({k5PV(j+1pV|zP)`lHP(b>scQxOX$9`!*+qW8l9X
zl5jG0%>Q@h5l6c*)llUKpMT03e3@C%=U3Jh3r`HSr_QCkk)ozTX`djmh^}FCtR8ZG
zR+$m09T<-8{%9MBmt0E=zEEQ%PkR|m(%sfdJkBxhXuIFF-ULoL!_6H$65z-;kZK!_
z1a9-*N96h_x>)h%1)TQ=r(d)K=^~jl86(6e6q79lPbq~DD9ExIoj{%4*L6b|9@16X
zgV|}qEQj`+s8Tw1M}1<0ERG>reUIxh_04_#vknTAnY0br@_k|l!!<*`sA<3Lsk}U!
z@UkCvjE;sa=?oLjDp?*_9iegDH>;F$iE8D6Z{!H^dd(}?mLaV<>&K5c>C$U<rDF(+
zBwcSGhT&W4<auN1z|zabZ5Q3#{B2P4xZm_@d~R9Ch&V3ZFk%$`cpJb|x#HEPw)vvE
zGgX3h+a`dSulU7{PgMB0whKp|ZaYOSA)@=XU&+F~d<!&h{^TFpJZ8{4F7u0)dEhqN
zo|=NMK0%0h%=<cW^!b!8Oz$#x;lD-nKT|YZfbqiS<%O17baB8Nt9%Vbm%X^F<TNts
zizBYZSMI9Kn*{T95B4?m%I#GyPd}~tn^@Z$K9Ab|=a>%^qoD!jRaumQ)@)(XA$Dkz
zT(^4T&z~#)WDnRDwi%qp4jcf1$$LhSF%#wxlY8}q2Bw6@OQdw8t9G^Y4RJ+9nT-Y6
z%O1u2;bw>3cGDkqMzqvv&h@`@i&zkP*uSDfu@mtLPZfv!NzXCNJ2|zh{CQRV-Lr3M
z;Wdz@&=+h5z76>Bb5e|ZO-N`7ZT+cNYk$-(xkE(rpgMI)(I^#%)(-tkDB*fTcPXhp
zWvU_bZ*%x?;7BvzwN;Zn+nzX3P)`);;cv~*ed&MKE?<KJ5O)c_tKazVES4`8si5tz
z+7p^lO;7(^@D+St&<@1AwW_~+536oHW3f=Io{~7{%e%j{O{KgMIwcP0jnB2&BU@Vj
zp?$AsV#utAEP2q<t_=jfmeOEJ^BDj@8mN=3AX7vR<P;3va5gGRbWT2`KxH5p+cye~
zRF=A>5$&W8?2<cu;5QUsDT0bA%-=1sIb}*UN|5Zo=)*+hQO{{8Gr2fMt3>sBO-h#U
z`lPO986`baZ#KaN@%x(30G!*1gOlu~1wG&M#uUBg056Bpl^@=B7zSsPAADrwOX|h?
z%==Mq`?o=+f0Y>lA=aeN0Z!%Q&y!LnBGVHP0*irT+o@UU`+OPh<7o+Ou`%iv#h|kW
z>rLj=$8^D8fw7SK=;7J-+X&I~J{6!$SfdwNi{UOzQ2vT2(#g=~yxi>SjQ?|_q{{q5
z`(FbAoJULXX$^7Sl_^O-myl|I3$tfb>NGU(2xK?Z=KSN7f(@{Lx>MY=MO;G7F>K!z
z-Dcl#iD@YGCE^3Y!`V6p4%P_b5{`a{<?iyr!J2-7uSaIUey-h3p^~S5z3M+%s~4pq
zIY<6Qd;}M^w(_a?_bFVylVt%PfVXye{5K|Z0p@&m)Z(9p1M%+WXmR{YgmIqK?Vj(J
zM@J18CP9C6S;N!?LIboiXQk8h%yi^JOSc^DLFzs_t*b%hjtIu`hNp_6<pARSbhFJx
zC75=&1U$fPwvXy)TLIptk;wKMxzCBGj}g1B8IG#62jnTUY9M5DhKHSC7{CSIiVuV`
z7}*eQvhMuL9yqT#{g28k!A~b_<f1q0GB>7g`dB{IWl$iz2o|JuJUm%mzTSZ+2dV>L
zQEWu8EL9JzW@1$jlXLDdjKH{?Wk)<WPD;mL)282uUGMLdZbn*lZ@hyCYld!x;PGS6
zVL;Zal-I?kDwC3Cu}!P}#4u;*NJ2y|QchLSk}M}fGkZd>I-;(;o8Q~5Pi!XcEv5Ii
zLotVmMB92J%aVoMyogWLL7dnwcfP$mi?8L{hk_~P7fPHD^@5_@hHP|J4WZt}On=i9
zQ%CstnpM(7#dutJlU61^2=elpvTXWQxmp2+<3lTr_E$E4ROWJGg9^c%@OMmNBHc<{
zU%|z;JF8#oEw%<P$|>{2c-Eoc7O|@&k5f)`zriTO=`BbF2evcx?_>0iIKAimc}4GL
zC#NJXniI{}<U}FT!QNgwquM$>M4=wMb|iedDvu{l?4W@C2L*l}13n4ks6cvT6Er}i
ztbj9P%Qkq}1%O40*57v0tMK*wI{r_$dkv)k{N=2RD><Av^gv(ZM!PY1C<Y#8Wol_F
zfFUAtw1H`LOE&ih2)RJ*nI?Hw3TkTK&-W?cSnI3NA8sn*?l}sd=wTfa?E%!34h(<)
zYUtu*8K0#2S#f`OnJMUg==iT43Wc5&YO)WL8+M?5zOc7AI4#z)Q^|&IS=1_SRQDGz
zKs8D!fgpfy|JJNUv8leI3B&Sp+}>8_LDw5s?hpI>D|x;Z$>2{`J`HFhlE-+GF!<ga
zXs+*pceI)d+AeoFk92AQs>3m0x2SB_RP1;euJ+`AzN7xNlP#=50gBf~U_7@K{1mI2
z%l-*CsG7lZe?0Vc%nxxy7ubgq_LWae`ohJ+c_|(vIpgQvEO`-UXFKqDq-cE2M|)>m
zzL*-%68p9`aaS7UeRc*GKG#jPo!trmsL6+%57dvN1a;4YFHWv`UX~7PJMQ~!&v`9c
zGe<(GfF#s6E!3dzZ|fvA`9y;)nQSTiLg$wuS5BF4t$Uh#)CWugKj!wJ13Z+|pW#7e
z^IqR_I?SjjS)ji8b#ONdmf}3o-ZU98-y1XE%b4fkvuLpn{y-rujITk4K-N1!D&I^F
z2X}VEH>}(xtMXATDh5|O(Fs|QK{`!DR^1N#C|^e$_Z0Zdfx`AJE6KSJY7yNoHOZ#p
z+*+C0Y`^QJwY_hiW#DU7{6@s;)ZvIdz*H~r&gZx~PI>-;B+u*Jh?*!*bw2wDO8D&<
zis|h{r$<V&<)b0dlm@sjj>JM8TD5)7v*=M{=h%6<w(iGEE=*>>m8Skx;#@my@P{oN
zedyIcBCmG;^A9>O{?BgRP`{gm>^`q)nAc!Rc55tlI8Sw_CQ9AfAR%hYen|OCA`+(a
z<Nm7L4oA68QB{RKw>gl=SidR`qiAW#<cd1C<8n;j0;VtexTBPv#qkqd+kKQNt_!b)
z^@7Ro&DnF#&atldqj}vo9C@~6-X|m=POO|_HbrSycrLE`kqMMR3(2mJj%KeV<!|S(
z9@-|^dk(t|M0n=K*qXKFy@$di#ll|izQ*%zC)p+4Z?8PRE?Y|`jAW<I#`I3z-o{!Z
z7M%r((@i-yZqSh~kfL|Acmb>GpGdr~@r(DyH_FVyW_>VS+bq!R?fz%R7suiHW@xz{
z8CIvW51-`xL?0(B#qw+|BTTtYii*ml&&Ra*-%ZV$i@9Kl?{#I1i$fnj1<|;AOjut6
zg}a$irou3fbij@gEUR6u`zQr;Mlk=ATDe`mlpPbqRV{iLog(BP2|V{2AFmSLeT!x%
z;<J<be%5Bnk8=E1;ua$<RO&Nt=?lx*4)(;tw%qzyKr`|7tqOW2%W~d~Uz>Z{TE*7G
zKtgA&5zq=-5h;Mal<JS1b<fI{rIsLd*<*!0_410^+b%l(JGFqB{8|TN=tCRj>$0cS
zR=;iPz&p#?@PR*1kXBO`<B0^zI**Pro8pY6UZY(Z)`O3{v&4h}q{ASmcLX@_`l6!B
zj(Eb9hjA76dtr~f(m=f)4T0s}S~549VNA{f1Hxz$MCFSSsfW41f5x!W+f~7*ux@hq
z03_71w;$3;VMG8TSkd44!M2SBXW>ZKU*9etEeQ<aalAmX?Mbs4IjgCDy?7tXGSBJI
z9|aF~>~p5yf^s1rwyc+V@2!>n4x#3^on?vQ1R;{;!(WHM2(v8>E>0~5u=l*uyjU@q
z0%VeEVryW-eGNHWQ07XoAFmroP^W$4C<;n=9t*ihNb9PdN<O6?znqWe_OD7loRnQX
zFtCedt^__N!ZEQ9HkoVA!-|)wMUfyqrbBklp-S@D-baf>Y7cj^j(A(`FcgoePUj9>
zdPOl%N)P2ev>9A5M;(aUkGf6<o6aLFJlzt{%toU$5{Uw_4JO2Jn$H35OJ_1A3-6-<
zORcGmDkF;BrRlGKxR_43cd|a!2R3UB+F9)?s$8~ah_`pO04wE4Ccmfw6;}O`lLTN6
zcn%?g{hheUP+C+?(1J8&m0{CC(8(z!UV+-r538MWMPO}!dvH}dXw^zwM9=tmS+#!H
z#!Z;-dw9G8<NhFj!)M`cY(LLQ8y1&!I63827DQCT;T2(^(0H$||2+6QJi|&fFsq(I
zws#p9&B9aG8fRvPS?{8wAU%jn?XiH!M5ENX>8?n!gQC(&)6yz+2s6er7&7QHkT=k@
zAv#aC?b@<x-D+}YSQqn6)((6o{jC1f9QIUC<1cH(AD5`bd7GZy3n5RIrCJsE^>|MG
z^jL1#?NMpCTavL<2H;9ASBLzgA1s8Zeo+NTZ~!b4z#CdL$as|leV<&2rR+s9rnq%f
zpiRyOAiutW1s4O*G_ZdG@}d0w_6%fHH1??TDDZjgPM!K2(Q9K{mezFk-yxGhT1sp`
z5+&=+J{6R)wQnqC2r(nus{Sv0+nn~AgGglM%nf9-B#N2}e><&ezl8`obK*`0VG>(y
zY~Yt=VPZb%07yc<uCB8v7Ut!Da$`o%`yE=47M4q!P6%MJ>|V|lcEi7*>)|V$PmwLm
zbN=#(XnC+0j572(e}1m50%Ul<`{n9eN1T9T826Ck_}2{I*~|#(O)p0^#{8It5^pFF
zb&?q-ai=CnK|F1gD3hD_k3kJHU7j{SUn^007#_v0QO<9CFrq5_?qTZMRT<7-YHpJ)
zT8+_pVKgI(^@+=(wal$=_oA4hN}w+#fJt(Z3=3aOpeXlQEDx_v`w1?9h+wX`m+ivb
z+)ZTY>t}c9n-fNE`Jm+73>4iCCV59~xd%SAO!1LyGDEF>;QIW|_rDAl-l`WA1Vg57
z^xneX7r@l_5@Qrcs`J8^7UDb$-WeK4m+XtJ3&KZ7S6+^$NZ>!&X`PU!KTz#P_GAY}
zUxpQ}?a#t%EUg?CV!M87FVE7`zVtM28Ws)S<jPCEvBkrE^BEaA=sg0$|NlpWAAg<J
xk>A`7AtOq~2ECaAz`qv_?F~C3qSUCUM{7shv@F07{vSkdl;zdrs${-|{2xkbS>yl!

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_video_api_001.png b/0.16/_images/sphx_glr_plot_video_api_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..9688396e0e95f422f666ce037ff23817f9e1c6d0
GIT binary patch
literal 894322
zcmeFZ^;aB0*EKo}0~y>k$N<6JCAd2TAKYC6!F7<}?!gJ}?jGC|JP;%ycyO12%X8Pa
zzW2TV!TsTx)xEl>yH<5~^*MERoxQ80)K%p$P)Sh%004%9ytD=Y0IL7b3PyfCazxcM
z`??Wym(g?Abg_2#GIO&6sF=BbbaZidv@?fzTDg6)b8+To<7eY$f!MmcfBYoG&hGSo
z3$VGk*|5)uOYOf_g7Q%w`UwEQGW*YpP%c(x2S5M-6r?4zymS7+p$V1>jeAjW8Le4P
z?pwVkE{7h6&672vbz)N(f;IReZPtG)gSH(w)cuhJH<ToiU1WZ`x9wEJqp8s?3e5jr
zPWv+!btf73<=9wtEX$?+y|#~U`%kDlp7w{=A1+Stwm|&<EWa3@B2;1j^Z)f+Hx|FC
z;QuNjz#sRvo@~_h|1LS%BN%%ByTFXfHZsfqDnPbZ>WY~DzbpMT-;OZ(KQ)kzcG&%b
z?fidkgvYtx?f+ilb;p_Xzqj!}E%x8m_^)mJABXz&Ur+g8ozH(;<G;P={|P<+jVb>V
zFa8^0|9^@%{!j`b%uuq6`Hf@aqyixbIbeu5I8jhMQhf<T4dJ4ZN-Q=V7ia6o6bEo1
zKu9TA04U-B)Pvb100Qi5pUjm;h`t?CKWYIO0tsV?EEGGUSkJ1OeHTWIV=gWcumub*
zfSvt01LpVK;g?;&N^;J++UQvv5a^MmA@UGQY}%)%MYQhj_gmS#zuW2j&u2|SqfKY~
zpHE<eaPFtwY5u!I@1dwG?FgjEV2)!%Fep9iEkH{g^bvpxav0ZB>-Pm3i;9`{<=6MU
z2T03fVSOh<JHdMM>l7<<kAw;YAu7%}pI+nFbNklS4EQVo@r&4ygMg%eot~aQ3^TJ#
zq8xQ@UCzJHagz)q?EO2(#6l}E9wLBKN_;-A7KjYh^}n3$L&gdI0*G89P{F3QOx7_$
z2hy0L;&6h76zarsbt1viAQ1AW{^o2P^C2EN4g`>dzOJya@YecyWZNiBGCn@Q>k{)5
z*t;eFj9u&z@2T!(w{PU-4C{sBWn1j|MZJ_9m6JmB1$WeASd3kcY#DDOZv&Po5qzHF
z6A=(8UZ7I+j`}Yz)n4qNa{@7N*GHBO=`<c|wUsI}5+`a+<YoQk>WET)Rg@yX$6Ow#
z0FbH@Z+2nrmh%ZYn1ho>bAX9N30VW#)6{`*Cn94TFjPRq347ncHp(fR*dttq1S}`p
zT;FcWU_WOQ*~(PCd&sP(L#P1(wZgN>#6~B(5PKC-upGYay*%-%gWdyAPB7bpZqWOD
z9+6GR6BwYA%ybaYw`Lm-AaV7Vq7ZD*623Y}53Rly>4?cYaK(&Yt-Y`B4UU<hamF?=
z5F)OMV$yeyT$XcI&nDjRO~Wh%0KxW*Y!`DEG)~9s)$1Fms9u?;VCZ*KX84e3gFuVS
zRX7&iA8Qi$Kq3!1THb`D#ilCN&0<58u^S&4K)~QgY(WTzMbJtx*xTFtPDfC7Tu<ZM
z+BVLqDt-5R*1M}`ynkin8{PMQ-QAnpL6;^k><!F#XUKPtoBr`ZcgIBD-V%EBTX#Dl
zdtHdnL_wD>XJRkwn}1&Jf?oQ5A4?0&#C8o*%;VRGHs44t_3c{8jFUXVDL6;s$dr20
zc}*;DMUVuMVfnq7iPQ`V;w0dEPAD=E{23JGuB?ibshsY`KrG{vY85=OfeJy?JIXY%
zg2HTHD6GCfv56r|wbu|p3?`$bIGtn;-O$S7k{BH{&{4S*q)w#{>JN$ikbZjhFs#Hf
z8Da(PZDJ|H7xuq@UxipBMqN<fij=Xl_E&1;Zz?WW{9Uh)ARD8Ua#vuWSA?2L?+9c5
z#|gCFr`<>^Rv3r_i~#rq{1Khs>mN)Mf!M33>aC~ae?H6)1K%L5P@a7BdLc9E4TyO)
zSXBpE#*hvG8QAPrRL~m#1Ki4pdrtrC=^3-jXXLjt7Bz&GUL3^IzaAJdXciQptV)7)
zgvwN!4PxyEI=OI%Oq@^Zs>JvCddJY#BwA{WBjAE@TQtf6ra<m(`q>j$PD8g(k&tGc
zBA7F*y0O5mdX{FfO3<_|Kby>pSQ*Ea2%Fj>OV6V%xxN<VG@jny^OjVs>jl*J@K^KS
zUgXQ*%iSh5$5|h|?pYTc<@)3{EEaId8+du+(4C8`2ztKSDQi#g^T*BG-h{(nU*py0
zi<s-G97|Uuytwn3tiV_JgC_8kVxr+kFRut<noN=oeBqq<CE`M%hXT{*zUTGh$%f0v
z-^eU`+uWok4{zXYSF`+hzr@~I=c~7<_~P<!X3s%FG(aWz@`U8SD`3!%dE}KZ6=MCR
z^8}84emt&3R=0yRb14!Q`YK6Ok4e;gqDyveynNX0>82l$ZmV}UFE6Y&>}Pvm^(6Ic
z9&Y#U)W9{(Ot2<6a>~8{50Y95LkrW25}e{z9QV%Ft}6Q7z)DGN$_y^N`Q5g2GB~j{
zZpKTu=k_EIhZ8#<$HE5-ud>MyRpOJbE5&7w^4N~50%c!MlQ*s6x17#X@B8|l&XVQ@
z&>Hnj<3{hZv}P=O4}P~BPkTNS(dx)ZuWbr596MJhyOELvOM3VA4PL^@<6TI!Kr#gf
zGf3FSE+zP>z_9*EDp(2Wkb+f<=o2;L%|Kdbpv!abU`ty9LAz}b9g)QarA~U^b@J^P
zR}l<KWbLexP8qO&CQKr>b^*7%@1h;Hx<4o!e#ZZ+$M#&=>A{AirB$}zU*4cF(lDFF
zqp-N%)_Fau^4L}wm0ak8Z8(EP;b$`~hXeeHk9gy^FfS&+OIc84Y)oaei2dsA(Kt!a
zYk%`>ys)KnaP#w30EiCeRM<&6#7_c;xtkQjb#`NC65`+1=Oo1B+_GiRp=<NS)ZQ57
zetaIw{8pYPS*>_RQyGU@kHlD`mmc<1U`#hZ6e-cK`>vWa5G+z>E?a0HWEUisC$32+
zmOMcVkPnyeBy$7<VP9~<KImLWqdI-32AOjp;&B5x6=xVa3JS1{rF3e$jpcTQ#KqF3
zS=qVcd5e~J0OTz2VB#diyyF`IWxRH9DN80E<b!&9K_X-bmQQZ{2hH67ewaw;Z1tkH
zoIrW4hViOliM&GBkg=pJxJAQGi|6^f)6*L%-U|C)lc8k!CYME@=tdJv5Vet&Bot95
zlhxxpxkO|n(RFT#nv8ksV%SETbGy&m#v2q6d-`JcqYDq=6CKS@VmAw-WeYsj4s%gv
z5T{L~2O5|nHX>9+Ic7iXQVM6R-Zx0_LkAm{h(`*xJn_q6S>MpC*gd`p{6-AsCXpV(
zOK%E=(9&R2)iG-@w)b_7!krU30H3J-32f;-Hn9K*)QhLr$~`COHjBO)9+_6pif!?n
z(U{&h|NMY+;!|>-CAPq0BEIW!wh=XD(MSfS0+*r$sY9P@9k0UVYVS814JS<RB;O8h
zL`+`H7sQVvxjnim*SPX2GL8va80l6kg;*jZFjTuKyMGDPP`W;gIIs+*(%f)ZsCM%|
zwvh00dd6h^EbU=eAzSVIOm3wnt>r~VQDf14SPKm3(Zem)6Egx`86j^gGAAAhebZCz
zsnh<PgYd5mg*|_LTU0d?GQ<&z82qPNhCs&>r<$=)nW>5@a>)pp*l|OPZ;*IbX8V0r
z#NQ6|HT&k4w-4KRrh5jj38hO{yPh8_11t9G<oH9s66CfCe#?=YIJnhTBmC^AJ)ZpF
zQiWW7XtwaHZW@gi**R2wQ94R%+=?PrQair@>tCqVe*0!hnF9r>^dQXJHAQ?#@vG0f
zIMNf_cN5s-;N3#W(kIUzNnJ(6gHxbHI*^XfV%d*r1^LSUYn5Ql&p?xz5|xdJr*AsW
zNMPzrb8$#WDrka)>Zo9fFX_u*m>B$EmV@*6a^JIDYO8PQgt|%qCl-qxhHsAY6&Wu0
z*TiW+^t02~V=Hy@??<7^@g+>!!sWqYZWMKs=Nt$;9jQ%FeLFYD@KNiR9}jWC63Xj2
z4qPDIb!u*9bkXXy#%#yZ5=H2u^7Z2?f3uuP(ib_af(iKde7>S{A5<9#T?Pp9v0ij4
zZxEF9Rn{*1x6U}r0>-ap+)o{IYwtT8+zc_6_u8cC&hTCNRoWA{t@G3w%*ILHg)wrV
zf*5Tg%a;vf!-&C#j+hb)Be|>ewotn+T54<;o_!~wk@1n?u(Jv`m2}plP%<rOt;das
z$^6JD7WoEywm)SG%3OGxt1_Dgn`Otwq;e)T+`I8fue5hC)9^^f6pdFGeVMgn^Fw+}
zBAXBZ=2Yt2tafjaY*oPft&C}Mjq^UIG?LFHcP(w(R|1~X8QgFPt8{D;GA}JPSf;3M
zMG0Z6jYq4su%}$^krk}X$1wPtePZ0x@9s6im#?i?r`Sl+5WXxTx#$~%kG-(vknL^2
z$e9HHc*#WWF@t5cP(#_Z)qwPu7<hA8{aZzYPl_v80H&u;zMG8j@yLSetAVW2QZ6a=
zn4r|VRvzuh?}zPhtwEi>O87(Vy;mOJYgST}bgKBgj&!x~2S=W}6yO1(H0XU}aW>X<
znD^8=XEDOpETK)`bNxo9GGK;oB9;*un=@+@r-DBMv(bQP^#e+fi@DQKQccVLd38W`
z!~P#+38P&caoxX8%4|}fqb=nXB7;}e**xfqEl7Z0evdVFq{l{qzkTm{+Dxu}o-%`#
z=4D@2zExo{`T_uQ0>M;hlAv8;9AIHFQYx1@E(q`ig^V}?ju`}}4+<v<3wk+yvB^nn
zSdBimwz9vgm}pit8PXA*Mg(V_Rl*k3L2ALHR}DJd9!nMzTqxo)Yr9@{zGBMebwMoG
zeL9rxrL9plH#&k8^Ha*1WdJ(9LEVh)hY4rMpw}4tT9?CWm$^>9+~yof@2z&}w2QOL
z<*$<-Mh%#TI5r28^iovKOb%~ml)Qu~Fy^^y#om>!S>ZH-$L4sO=Xtb)OxI$t)Osk7
z5(s=FoNX<fz`ff($P{M51jhbnE{<cWz>JC-iSlWIk}ytFqfHDCF?gZgn7$4vfcdN)
zZq~Hw5TKCDx8&D1ZLyR7T_#+q@YjS}zAztjgrl`3TXZYn`yId4-J?NxwMrJ-+xGbJ
zYcr?eXGdkxj2-;|PHfLh(C%5$*u|p1NM47Wx|D0$3Xoj$$V<g}%KLiVl0=>safnb8
z&#C(PM(elT1Xrc%kL%eox=F*H!<ilLJGrB>yb!WBH9ucaD|gX`fpR{$Gr);o<#713
z^S<d4tLId;lW7W~A6t*wMP;@~7N%jH_18GFYSIRZmNUcghV;JH$*N(tMe!n1OU9~Q
zC2#?tbY-ao+}TAH{{12R^1_dW7U1L=5WA_>`ODY^q8Z!1d5oQaWYyqpRE^An`IQq6
z`0FT=eqcn6^osRz7K1DCEg#8Dl`_(qi@c^fy?fFuD&wkDr#B4RW+>nO$8af@1nnKE
zR{fWBW*mnb@yQ`T@RUqf&C?sDj@pev)ho7MM}uK|oA;z&BMSyfXe{D!hQh*x1ZlCW
zfTWpSjS<H%Dj5px(#Qqi2+P<Wk#=s4J>zCl+^vT8S!96(n@pSe7pfTPerym`n0O|Z
zpYnX{dSKnbY)BE3H%5JDOS5jt_F8iJ;~@$3V%tDg{4M!cCwekEh}h#_dhb@Ttt0k6
zNmKDm^>;Kia;eQBZVlo^mBq@Of38n3`?7D_vABy_<Y^J~ah=Hj8JZ<F?aXw}LDh9<
zyMUKk=z6Cus+qWDpZp{lg?zwMTVGwb9+uL?@b|cXI0rhesTJm7WIHaQ;8rDrH3SJK
zKdtK!kD<3yv+^l50J>K7W)>tX@uE!O5=-M(wmjT*?DMv7T&fFAgw~=V<rxM+Zgi%;
zWlTGJZe)<iA+T~J4uI1dWlJDfYG}>NC0GL++GSP2+`E^uz?`b#=!TQn#!M=eq@Hr!
z`OPR_N!`jP8s{Als;fMYsR70sMdAd^+F}KdC?Yhd@(`Oc6=-0CGfPo|xdFV+_&~j8
zu|uwPoJ19xS})C_e_|wuLprs};&l^l{D<n-)lU?ShA{_^zt-|xouMS})A(w&w*Qo{
zEPE0<)@>zFO@h|!Dee7<?S$0cTF=s1?WHcgS-C7%ZPHFZ)Tz$lF8Ak$I-eLVe!OQz
zyqHh3a)<U(H>s(nt6-NvH|^I^<XQd8H(-?pn~h|*K}%Nit^JafYhNtdHq_QL-vJs^
z2O_QZ(lcY1JLDiS<l6g++<mDQiSaQ`ENPwClh(Esv#m9;Hqc5m;cU*18pUf9?ey%j
zB$6$&Q^ce9UYY>curQRPtVE50AyUjDk^`srM$@EtzoWx2eJdyI#w*ucfLCqz=onnA
z*qMe>Sf4V>P|n*SDA<f4I3GZfWc8b;oNC;3n0k)A?n1f!?=qJobDDgj37O|Rw=K%w
zYskSqY}Xo{`{4bu^-)tu&pmx*s&W#>u#3N=qlRiu-6(37S7kfC#V=mpxC?kF6;kHi
zMy%a0&UW{(7!uILQ;!L1?ii@u0cb*sn6&3~v%Tw03>khpg$Nz)!H1pHy4Tg6l#X$Z
ze-6R~yY~WoBQhgCcwCYbs5NN3MO?;vXzgMI{n}hN>)YIZex?w#`l|&wicZA^{u~J4
zy`+rYZA&#9uI<gAL0+x8Jq$7t^^4CJ4vM<E38PlPNYqH2M`eUwGG|emOK!osIDWHD
zt+);~>_>^ttrj^NE9sJYIKA~~X$vrBetw@eNQqfF!S$VMF|K$$mx7Oz;`F_X7o|-6
zPBnvEjAHJBBFSnrx>|$w#={;omt)Nn3G=O>v)f4Ld1+64=bshya+P56IT!wUhW@#f
zen*LDSLfJO)v2RTJO8+Ao%4vs&e-jcS%55}WEOddW2qnVtUrPUa=|i+?@<~zUYxF?
zI!Nzh9JObC!IEnupF49M7?}4Jk}rjiim7Rbu#@ELSy1ev3lvU&T$qX^jg{6MCIREx
zs7grCo7H}%E1MXV>C~>@T1{!5668lOhr99BM*C3a`VolpS*Yha#sT4ljlze3<yvy9
z`v5Gvy57db52tL~-T`!z+I*oZ{z3Rq*;6UoPqQ8GOvZ^SU8^!Q#aLbBv6a(zx+b~W
zuI7K<yeQ>0vfR$iy(E#yy8n5?x9ULUFxp=t4s;L{#SnCUn7do|5z4N<X;0?@({qbk
zW%D^iO&@-EabEZCsfFIMV70*ImjkZ1U~v`2;~pxv#1CJ0R1RBpT4koP_<$~XW)@ts
zE;1v|(U5rYNO2r_9O)&znuA~Bd@SBQ2~1_fr1EJAI@SwqmNyIb2aAF3f54@dWm3r4
zNDeWvc6mHn5v&VDKfLvD!crS=bgQqw8XfZ0eo3e2^C7X)3wNtayU(Q9dvg8rIu0)S
z5Y3bEU{0fM3$ON+3Fc+^xSmc;>r%ZRFxQ=%5lGZJLLFdBvR)CqM8dTUODN?);8WM=
zdXxUVn=2Eh<^J{#1d;s~C}y#t0-i5+eqfYP9w&Da09`7FjMzwi_{_Uv+Hr0iimBzq
zoAD?<`(6BxCVFORh0z*tULQFIU(>y+ij9X&wslc^3BgdeUfSqJ0lB4jl$h!V<T$_A
zJ8HSkWZUKWxDQEDLWuS5*hOq~*(Bg|cYb0rM&poo>!v|BmBDXh^{iH2CM394%EL^h
zV=I9JN!@Yi3Gt?Ubggob+Y2!0VqC2u@E+ME;q+|ywD()n)K<oNL1t3DJVIh{qhzlS
zVz*y^s{2|<deJ_kLiNdy&Yh5~N#Au;<~a-+i<#bliJ&5|a{BC%qS4b&I&tjF_K6ba
zs>Q<}d!AyMWXp}U(`Km_3pY-RW?(wRMbqIQ$+QLik(rQYNmGi{;D$}Pe>Zrg>*&D~
zt_`x|rI`)o-SyLimMb^?${7ZqYx*ph7+O8bI@D40$9g<jVVq-5Dwi?_je%7Y5vsqy
zi^Me<=si*H!+D?2eje$1!ygEqulwL{t4HxR{O@^t11||(C@HWf5bUO5EHF!DDj2GS
zn=@QD4w1jK8rgQSBOJ1~Ft>EcuNdJ>^L?bKst4Vah+cSYbDK?;EYr0>#0|&)dXaxc
z#D>h+Ke}s>{N(fm_9YQmpA#=o8{*|1VFeZc8i8J20yCHRdqS=Oar)f1oWvs~=VY+W
zhm3Pri=hUpxDZsfKl&>%r1M<$3!N|d;Y<Cx&BhP0au$QNq_3O0P)t6$BfgYs3fgK5
zshCYmiqAs3oPXSn(r+J8F3qqapBNgeJLX;%qwKPaf7X3SpG@QY;>7YVIO1Q_V7fcC
zk$doKSRqu%G+#*Ba8sd_Xb(ra+fLkhFj!MELQSD(o-8u^)O>L@-p!lFX}gax*>LD3
zf^*0?3P6HzFyKpPl2Ou;-Qe9skOKyiHMwQ^*NG49NFu2^AT{Ievd5IZWK|8q+VK)a
zX4M9RY^`;<Y5oSqtqw7%)0x6-1$H{4AiYgmsrDG`vrCgFvIsauU0z>*7FYZzk_$z=
zw*c#y5}y>I8G$LU&%aid>(p+ZzYdh8R<mPPKmS&x0#QRfC8UPc`5BO?udlS`@jP86
z%|^QBameVVB3I?o4rS!Vz+*#W#VYJ;hF|~2f=#I<OE*q=dUmR7?HTYi1+QJ2rbNkU
znlB&!<F8oKr{89EsW)we>x>TCZax_JMKT7g1$VL*=LNXl|5F;4EoPQ2Pa#!$k4LqX
z_J&(hf%AyGb3i6T_=k?R0dfg5vIQ<5TA#<~rZlQ7ISn*NJAQQ)Vn+@cO^*?>7Bz2%
zlyuC>P#N8aSM1MiEF2WC%Wk~27_%pT6kP=9BIam7wxq3K+X@p@l^3Le0!DR~_dYu-
z%In5I60N2E`;G9HJ)}mmC*S(l=}mjKr(On9*?ctbj!rsh*%W$BVrx&ZhHNv6kmGgS
zs=c<FulANb32#Y5-T}J(aZ$eTCC?Nz+o?Id#rVy7zc1T#iwd6wy6orEh!hKz?Vln<
zv&#w-)-BD?df_8A7}C2w1iZ#*4S6ft(EtP|E_LSSiH+R>TkXw02bVgZ#JB-%dUB~w
z)}kT{g`jWh!p*Fumod`S>li81ik5Nnw}0+Gb^D(K>x_H*xkO;|(W!*R9s{`7U_Yiq
zACl18X0GUKlLUqY-qRM7<WE+;>GLhdZNA&9(RC^?C<q6E1>qHFe3Atta3T+3OAxX%
zl*FdtEfp^VR<SY&X>8{cBizvg;#I)Su`SJhFZWkGI|C<u@a|{b$~V&jqRR~>lRqsQ
z{n={!TjP4w()pbX!u>`zrR+!hdR5Y++?B*43eYqJgWRDMDT+Pbet~P5PsD?kKMBEn
zHO|EIN0;X6pDrRMxKbwEBqbDliejEz$K7hmPIxz~ZV$#UzTkbor(B}TizmxZuvs6>
z?R>U7W*_;!(~h4yXA$SVu77HlvgkWDM<;fgw5@VjUCnMVw!-On$vFb+BFjJ_S`!Ty
z#&geh0*5WLoBTOBf!TQ7>Z#J#YYn;%BqM!}KG^r=`pCaz965O76k2O|{g-K;mxgu!
zfo|`*cObOk8c}Fd!A>Telu|pta0ASOX|hhJN96B;kjMGCUXVlKv@>;EXK66AoAIoS
zc4p>w2t#jX^mFH*w&TXN$S_}Al$6>zjxUmouvF$k`omA7s`1scf6Rj1MHVK7njE63
z7rlDe(QZCk+n?tiF&Rr7Klwdam@3bMFk=npYbBa^HqhVn>*SOp^S?U`FDl$R47`kv
zk0I%is^f>AL}#{)p{r~w%Jf)g4o716?>L$g6QIH?kg8`{I<~GaGrR+X`h%`bo-$rB
zv&qw?OVTpZfjbb=?&ld{p`X+Mj2WqIUpq}&3j2&}pE#AwF4eKDX}v(DoTgRb!HNH&
zM(XS3Qi%x{rzF_B_5IfMZxZ88k<h=feF1U;v;&Mb5acPZH{a1A_AD$DNN$GMXlMUE
zM$|2#D=0A+bgmaL?y4i^%ERy>nhi58E=A>KL>e`rw@`bR$laO`HfF>roAf-|1D14Q
z0DM9i$Vw(-kWxQ>l<#a`S>3>0Eq^#@Raj*wa~u;Y|FBO?!@$u9C=^#4XynmhmkE%}
z*j}x%>pZmCb!m^%1;+07skqs8!An!C@%p~&YRO6I1}I){w9Be|T0eY8TUZ>HA8<jE
zx4!wi?~Xqxj+nUbSuF@F6*xnZ1d90LicZQS@KGEc!Hk+LgFryzNbSOBU9WS2cx<a*
zoVqH8Ipvp`+oCV1id!MX-!L%&4VP{Tms4SA@Bzka$FU-gfHymKfKb*5bda<hS!>_2
zRSZ%jDdP>s!ofp0Y8_i6WGGPmj?l0o1X_RofoH({&u;8I+?Cz7h~IGiBt#+|Ev(ot
zv)3zp?ZvUd%QQl3;Dpkg)<j{rE!{S)3L7YIpVRpKXqfKtfiVPX^#xU&FE)?`>TIzm
zkSS>ewQ%*gr!e~SkQsX%cdelk>+EGIZq*^KG4h%hMZfi;WSb--o`rtY$BS=1^kl@b
zG>jit(@l3c#~dLejXhS=O3yy@GPA4~Lb1qQZzbc7<17MMohfOpSNahS6Aw_86FOpE
znwznE7abg?i&U<8RZPx2{%s$x%giY9kxd%5sLF`;(K9=9BfVsSk<Ls#wA9dE>$QT7
z%EO7**AvgH^5JCc#=?UpFV`lpsI%Rwx?rmT0|iqBa54vyUN^(KZ<&>zM^mNP0@AUu
zlr#tx5Mj)=PN@67t08?#d{Luz)^I-Bg<``rXo{h%*w=+nn2eI+6!3+<e<J?+b7m<C
zUd7ZXHiRUzzdZq`4)eDC{lD|<dBkW-E%teRBEXS<qV%Yx?ek}(ufc-6!OJ0Af<dXT
zin@E-=Cr&*wQf<HWn-<rjM67VL!jfEC5Ks|$7&U|qo=b~Q=|81%(HJr{434MYIf1t
z2H(oGjr4|`++(t?4xLR=od>{bn60<F9v`!RR2&#I-}@KI@ED{$7peM>QgyEDh&ZzG
zVAhEL)XL$Zy-CTadu7Mard*Mw>&es@>=yHr7wbo}`VKXW4t0J^!@L!A1xhk9n%}kf
zJ8hGG2@#^GF<1OlgFtwQEVHi@h0Q|u<|sXdRIjPHc))o&vYq$b=()+KP&|D0@gVl|
zYIt;vUp~6~{cPO8JRRaRpNfUj`>u-h%jetJ43m^?6|qYM{Yy`cqKUfL5VP68bff)Q
zxblZ~;<~5^+VQLi*F$qLGu1C^OJl)J=`9CEpdOJH1*n?E`YyM#i+7*RD8r~dz#D!+
zVBF&!3~vt2!%n3VJo>S8E$@3|vy?h5tlrAJ>VBVoEVmvhQ2T6BUB2q!aa1I#^q7!y
z+^qX!K8m=yY&uHC_sgO|ZE1!cfuZDO2*Hm^l<#*`71OeZ*|y3ji^86>sZGzfDCY=+
zRe`NZJCnbEKb1N(DsOvn{qj^vkC>%wI-Mu*Iiw(zGW^f+FkV_^(zd(95b)XHxZ6Cl
zerZ=>w3IkGrM}5eVTS4gS&OTJqUMU8o-n<hI^#a?xD58Zpn8_&<0qpfwn_NGWMkll
z#WJ*3mSzbrb!SPKk|bsAqky4Tm7Af+%TTL>iSSgO)L>?ghU}Op6D`)Y(-%7~kMskI
zqGMs3hf|7C2by;HxJtOnW;et4^IB_?!&_rK4gf?V7U7*Ad5VeX8<fJ-*k%?6!8Y-W
zVB()s{^bUJUxGC<+A82)49V$qq?Q*swO%gtGJBfEhFnX$X7nIxkx~?m`r<JBm^|4r
zM^zO4T1Pi>^My_PdbC(<B=*z1->>0ola1bTe09Y3`<|1q^KU8rYjv^aqMHLX`S{$^
zgx5hl<>xro-PPME2%j3u<u@jq;9va9LEn`&Uls*wjYvXg>U7qdz>48Md})<|&TKp4
zE61J!xvJi`Cy^Y-*cJ&$h36q-7d;|bUPZLcu2)=E$2F~72qbgnStU%-hsh?Jg$bJc
z&z6H#d8I3A9zT2KU{p)t1+Xt1YB(<sb<YHW&j)1!J$|9v=aCV0n+VDMsYy+p56jwB
zTJDn@QlfU~Us;??yL60XP&II4SWDPO*X~G(S7o#xj@<k0W3Shzstxir5K7P|fcP=v
zI5fNKn6oD*XM(-KsRZ2Ibbuo_Nd*)<JJO*kBrSf@*5GU9o;pNFlRqt8^(40WxR%qg
z6KA*A%!LL*EC8-+p^m`oi0-GgK6|&Pv7L2dN9H;wTz5{Z<sc-8^hjXeXd{dF?HM7y
zbr^QGggeg+uV8D5J@w%zTCl!wVtfk}$O9SYTl5wC$@$0kfzV)cz)Ab3j;<03{`-a%
zpJk=0T3b|9N$|?_`=`ZxINH+{eZIM1&p>1K7CMX4$l!r3f-#~xm=@8bF90Sl0S+;&
zr1Euuua$Q%%CDHO*lpN%Hy1JhxZ|Ai7y&pmX(8K{;jXw6=y<>y`J>8n?o5u)FhSMK
zKVkAA;t0VOF=m<ESH?h3vW9IGdUY}3;u&7$TBVkVW`E8F{sPqmg`MA_r02&RGY7RF
z%B+v|mk<@iN&9wx?Dd}JZeGB>KM>?TJiB!1ffxjj%*c)l5{??XYtc4Dgty{pJfD)R
zI;M@TeHKk=N}5^KkIBC~9WKT+_&RZ1^3iRajo3%xGDcECD5#%z@?X&WAI8;0_0oA0
z-A?FP^Ew)ye~h_tI_RRU6>tutUd*TOm?`%@;j)Ugv5ii{cyltviLa#Q?sho2JiAC3
zYhG;)Ap)B-yRewNnB#D|m6Y+inJS6y)AG(84r_>g3|nGst4HnO9B;1W^c4P3sfpvr
z-`_vDZY%&NCnp>cBtiFkLBYE0dE5Mb_kKC?r)M|Wb&F^@G`n`lVITG5<)K&*ChE-X
z*wjVS>GFYnyGxBGFa93MzeyuaKWb{kOT?r=ddsm95qNpq6!Sc0sCdcMcMrcK16G2|
zNL=GH>2zE)5q7cb(Qw|2nzp?+6+8aeW-}DfZ}R*v<3EhzW)PfqDYbS<=JxHrMWZK+
z&dU<LD6gPrFXev{$3iXJrLMIvdT#~;wyKyN)2PBLM2UL*&@@8pZf%xb!&q_Fn;Tjv
z7mj!!{GD=~;fs-;b*dJ6pfGeo2&%)GZYIC0Q4fK{I8fOW73at^q7qEF_5j14%JpV=
zdvIVM{HsFNR-*;BGsOd`-#(r$l2$}Z6|EI&QE}n>&3pZ*Tv8DYV<av&AVGQ+e?*8_
z`~m!ZeN?Yk7x&bUK`xQQN!5ZV#A%~r)RTC>nLt9R$ZeJ#j#X5VT8dtoI;c4rmFQp^
z({6#`^K$*9aqqIpcb8AMmamlUB1YWx841kiWS3K>EsR9Y5Ka%%RF!A6P(UhUX=jEa
z2CjKKo9!ShSacpOgz%hv;Vs`irm-EFER!^_%V08@>uz2n-7I&&u22#)()I33Amf)6
zq<E!R#*l332gL}dH0dK&aVt$q^1cqXS^>bp*yg-Bl89t-Xt<MyM$qXmP7bIMrN0Oq
z!YmY`)cq(w=JmSn5KV*i+o~5EOJjfokWuTz^oBkm=0_m}>4E3^QQ>3kFNw=YE*t)#
z9!Ba*==hwoAGG76+;`mj-b0im*yyo;Dirf6jYIWqOpcm-6LLo<sVeW+wPM$u8^Vf5
zvnB<9r*aqXdxi<fgpnocuFq+N4UpAm)6k(tf+ICv#XKV@oT`}u=GG&su>5ZCDBW8x
z@4H>4)STuQLJ@fjZabH#!*hwww&`oGUaLwLe!oFTn@add)DlDv1RM1`vT3K@-oFsZ
zgRAET06$VqdSI7+r_|fI$F0J<Y&XZ>#~q|;By;C7aQyQb9)^t=Msp~ew<LxrM7JH#
zD6E|Ky~lyq$^g3LDuJaM9Eoh(isSJXQqZfu+RnaH&9!P`DN8I#_@8zN+@SgHm9`Vh
zo6VAv`$rZ)+u+(9U(gBDihItW_?~WjHS)u%!=?@(FDMb{4n67AB5#n%t7YLAti_BN
z47{iz!&=)LdnSw%;nP}tGw%8}MJ1j}*Zt^}b<~&UrzFdOq?;BZw{lP{$lRWO^*EG-
z_orl3%z;sj=62~R0q*=yt$2r3zpoCD*qUpVwR&~Lgx!`5`7yx3-K_J~hW_D0l{WXo
znXuPUnK%3otvYJ_>Bet&VJcRVVBjIuTVHQPELGE11~~*WFy8GQTFdk;p>vTNd5Xid
zNM*`RAWql`hSjMhCM2t$7NQ|8EnzKTU4R`b02zst@67!~WHaR|s~7tD7vRLxPdLc6
zze&IX3j=j_X@zaolG)_IPU^U<m|tg44DEGNg+EVmUXEr2{bhBh-gS9CjylU-8fH0P
zRH4}%^LkaF*70E>iCyNCb3~~qwh;sG{!we%<-40bY#<07;hn^yp)7~hmk{_$04)JF
zm<$+MG@Mmg<`#*}g8mCbPvx`)p#>ive`H#uSWY^1sBWU7aU(*2CQsw!5jjHX7f0b(
zTBP|YruQte@{l<j7ws4M$;a2u$6Qi@d7~%b)q-)<w_%)#NIugHRRHKiLzh`2jXNU4
zepGCC9>n)=a*51awBg@F5(@*jjc)cGV8!gDMW-avs%U;>mf#UO#8;tU!{+fvhA<=F
zeijrgBzXq!3>`8}Nqz40R99t^&P0Rio`dKPQiDZr?Ug?qV2t<5Dt;_5j>3f^s(i51
z#at}Yaesy}!J_HnU*d92L5f(MnqzpL8~Onk9a6zcU5$;2krYN`i4QXN)&4y|L+$+!
zg$H~R&a8+qfFObC=oa<>RclnEYJ2sld9EZ8qtjfi!7*6opVp7`Z3&-$C0~OydlXsn
z=l-_t`1va)5gyet|6DsBdoaA-#|ADXpC9gsyn?vT08<gst)seL$%zWa8~cK#MpS(4
z(5+Wkq=mbY7mXYoZp(s!JU0jw+0El4SV-NuR5<-o3soxe8-@yUyb86<UHO1?%JX6B
z)N}7831tb_)9W{7rUcEIknrf0sUWJ{r2-b1+@5Yjd9eJxXONiBJ!;U)4aO30+i{^5
z#}rb{A_n)F5aWXXS>i-lu&Vz<rXgRWqXi0r2XjzQDW<|AOuMaNIsc<E_$DB3)1ZB_
zR6S!B4O%-(wmQP_NJA~O6g3=FG3TL@J~)J}V=>!Fx781x2YnRBiSx^JQYL)hgSL90
zTxL|x>6aUq_k`+MRC3ld*GatZh#78eX*=(-UK+7XIqH?MW9DF*jC~g}2p19cIY$ld
z&1R{irJ{@>oi3l`DnZif99cl8``&+0QtD7W+TdJ`vyc*svY>jd%xGp5=m>4S(Jmeh
z^83~(96!F;u5MA@dOn1#-`J`-HJK>AEvWdT_quD$wPx_)lu|K%QD6T`9oqnG#R06C
z;(%eJP~l=Pya|NjS-w%%Iu*vXi@5frX}m#&qPiK`R&ZJ#zdOn6eC~<}{XQ%({mCfo
zpp$da6ZtSD`%*jP{M^PO?}n+9fP-itW^eYaD*pxD3VODC^oDFR7eZ_CCz6^Z84u#A
z*uL^L6A6PZ9*hkNaUy3Mqs6j-2Kn$BM>+1?34ae*`^~vf#fQ2?j!6|V2uhBOqFG8#
z0ww|%+LB1-@i~#=xZAV6Rz8z(B26N`rX~8-O!`J*+NDxq>h|v~=gCz}q}OXZFy?<?
zrq5Z5kdSx=2EFzGPu^NqI*v$$+JYyj9!GmliR2Zb4%Z;c4#Jeyg>k3-pai0*f{r+%
z!-J}X(sRl3IRI0uh~k?@$iuF<xTR?peHeCP27!-CqE_N32{xo3|9nWX6c;xoL1war
z$F8gJzx<@QB15;fc1OW~*8+AA)ZfEh)7nNye|)p8y=j3tR!QG3RfaPg99WM`+RUD%
z=+?RV-Vz?I)lvCKSelBdu?A}WsXj25Oh(^|)CMUbm#^rQR#%x1eBDf`;k)FsBz}dh
zl^;cS^*_ldjdhqB;WD5}Qc3UkKKkkd(qj9=?D7_iSP58;q8ZE0aHz`nau|cS)mC`b
z-rA)I88OZw)0Y}9!SX9oox0ZO&J|oK<a>PHkDIC^d|!I1_J`O?J_PJ!B0^uxkWXB>
zSZyYFzvDl~2(J)0HX63!C)bk%wulT7*QbZ^(pNh;%hhZC5Ffqn56cSQjRqa5nWlsh
zDt&Fnw?!zYV=*t~`5;);tA!-f28{iZSm+<#9q%#6UVL0bE$<`2ffbSQnyMudgEKC<
z^o|@HhuB#wTLEdg-ytS?NGOIr{R1+fO(3dA-`8T(a0J_Zh9%Oj{Vnkox=OuHiSHzH
zi7ooNp=(rj0?%?D(0IublvNg_#w%0@%u`%8|2c6UxVni<)Rivhq&W*CrY`M&<;%lB
z`5S%xWw-1cc-xkzTly!(Vp?m_Dxfa@SOU@30oQ!=1ZAeW@r#zXwux30frrI6Pa;;f
zWgZdO+Xj|CiR+`D^4GfZXackN3o%k3=_YX)X&(<bGu_c@_y5RNylKiraF7?{m~c~B
zEMRneAW5-$yXOFPGjdBdj(3`Td+1HQZP7}g9fxw$ovNE+CE{_m_xkYZb$#Dl;m(au
zaU?rD-T`sYA_U~a-+|M_RQm=Q2Ko+8`Wk+ShXLviwo8L>1WV|nzsj~Jno<5AF$ud9
z3%5lxIYGXji(iA`RR~pi-|i9ISoXv;Fhu(L$h>Ch8*45Gt{AWBk|kf0C{(i5cEgQy
z)eaI9LeCmYd?jw^kdN~cSq;b{<N?JeR8(0%qlZ+<br6%{XU*N-@zI&Yw$@-)26`=$
z>?_m%5Uq!OY1pK!&$sz!-k<#{n=dP3OHA=~L}n~tDR`Gy98;b|Oor%&bIsE7VAJ$I
z%Ez+>UvcpVh-jqp@`AKy`c>1lz)8v=lc*_)*_n15*cO3r%824Hj!iW+k2#BV>)q8W
zH9_(!>t#E^j$aj3L^$z$^YnL^Svg95iNLUPraU5QrK=E`pU}?AfaoOWMc7Bb7BGWn
zu|j7adUQvgYa|$G>OlDxaq^x<_Btzg+H>$j#Mh=Hxyl(m6{&0NzL?$~#;gTGXr=k+
z*6soR*UodrnZ(o&7ufmM!~&o>pr1?X;{y%6letogy-8@mVVn#(Z`YM~D4ZBR5YgD9
zBf#_2_E3oj8+0C-eDq@(ebarv>mberMDhG_(-wikt(M7PI*R8hOoRI+Bda#Kp~E+$
zIwi8dCq1)Ea#M%6O7=_?L$kFn%DsP55b<V49YfFBvEVCbkK_VPP!bX&QLzSO3Q+a7
zIh~f{Jj^H)$1z5-ADb+&8gT`p1fSn!FlqA<4RqXybEWfHyb*thXjW<_NLc*-_wJv<
zrd<3HFV}`v9f>>ZR%OL4ixQXeRAsC?nx07I`;3rOG`HG|NENvcBBzz{^hF2-6RnJK
zdlzeN(%HWZ?(<6WDhG;RL_`msZhubK<hYGjg19XD)_xB+*~lNEV4ZG2K&Dy5ct<x|
zUw}Ym*{N-AWL}*`E~2#=ajgKJW0_M%xV@oRnTr4Hs=l`IBCLe@l5;=7S-^}#M%nC}
z@NDG*k>*2GKbihhz0rR}J7@l0{Z01;aNJHZ2a@;}TB75YS=Dfo?G%Z5|D-20D6)Ok
zVvrGIj8OAr+HZ2HCRxhOLpAteK;tjb$8_gD9l3QKFYWX%AKaFPi>retmB?dCnq6gQ
zHoT0|{i0%rTNb%zTF3GT^!rBt><LqI5lQT>DS<ln?ffkppPrtd{l?v<O*3}BdyqRm
zZr?ib`kwL?Giiqn5W}}8xJ=Rf_E3MX+~lKE3eX{z6V9t>Cdbg-8w*lU^I!D_$q%2h
zkJ9lvm5JXfrE&|ZdNN3mr+5TUj%;(q&hH-xFVpDz2~$?>3%|hG1t0GkTjqcEQ6@#0
zJp(d7g!4K($1;@(vdSNsAAiA)JO)rt(-W*4lT6vp5+YXKyW{>gM>L*3^7Gg)97N`m
ztrbG%F;ZF>tdbAr;84TcB~E%rlLiFRT!g{9bqB2(J99Z?-x^;*KjoY`29+5HE6qKP
zr+L~O{BX4`hsBkxZ3k3bjz;*I3W~8=L|EeS=6*lg8?xnutSZvN<UES7Bq+!g50{T`
zTaIRK6LpEK`dblp=XCbU1gJPHpgJr{vWntM9Qg9BhuMw3pB$HhK}4aw1tZnvYv?0^
zKU=E@bYf@w(p<6k4c<^r`X?lLcoNwCQG|X_gpGx``}SX)ynVV^eu=!k!wfp#3kop;
zXG-gqQ3Gv-OoRBhQ_+@^Ip2(M)-jCI#9Iw_vSgzYf`JDOTbvvk<D`iiUw1cwOqL2+
z|48enc=(M}nuQoqhJg>rz|#IsUTUtB8?SaH5>D<A4qMbt6V{`CW8L8|;4Zrn;pxcN
z*)7)p8n<ReA8AE<g5ra2vxA20jCg0lKk}PnAXq$zy%`+VjK%l+EF@U7AyZ@*F5mWd
zV3j_QL6CwNL*}^Y_pN|WZ<Z$~R!E#9ejK7y7%@EuclvyEvvn%)?^ot<B$5*RYC-77
zt&<u~rYVa~#dX3TxH?KzH;4v!x_Ac~WrIA9M&3Dm2i161e>OG=B^2qu2ac@snE6~H
z=fBRe*{a%nCbSbY7sE4;T!|<j)wV$}A7T+_iKVp>u11rb5m4{1cqg1gXc-0hz|KdP
zOMKK7F`XpM>cLe^C@H0HAnD|%h%r?qyFc$120}QcKAER*xeTGLd4&R_@Bg+#R6c3O
zV?BzTl#-WXLuj|Bv+G>%u8@ph{%Vpjgs4@V+P$U!myK}fAD18wHbsG;sGtN*_`jx1
zqhjED-K>F|=69TyUoz`im@g{J3>F)7L>k<y8B1)jCm&AatU5^Z*xcrDeMNuPU{%vo
z#&a_AHxz|uvs?N+!IP|@Q%1QCblLA!X7RHf{EiD{p(po$bf`U)KFy#^Ca_K>NS33$
z;oV)<+~E{NcK?MT{@$Zez0Jm-Qn1&W<}To`cXNCT=itcbCqZK;<?y&spah$cY;g`1
zXs9Y$xOHT|%7<LoSKn`Q!sJ>R)M_Tk#$~eUQK)hI69h_3c3nAs#hGw)dFDWPLQq7q
zjyNi&nJh#W9j5{CSw}f3sr{8MNyZ}__vMiO@p(BFZMPpx1=(5nYkGx=7|!X_#9@f{
z6iF82oJ~jlHia_r7G;a%*6VgG>g@jf?cW~LzW1iNl=UOz;IQPFTuIvq)t@($V+dV#
zzzZPM$>7FQ$imga;VsGNU#7l83AIB{&*84AdyD@tg$asnl_ps+Sq;OfJD)yH5ThO0
z-%78vcbOQ{7e(K*9{lQF$&2f?vQ%~;cmKk&L;%Z4&b|6U(0%wp>+rxP!47j!fMJpH
z@#&lEl;C4;jXF9>hv&lSzCPUBOU&QJm|2ijZ=(lMQ5IrqJu-zrHQ1dD#zT_-4(U>k
zwaQY5WM<X_3#eN!H&s6)I4({j-%W$*MBvlZT&lja7k%fy8a+)DV$DYI=|47fN!9}{
z*2=JXtaZURQeoh&omVzc%<r`cTucQUf_n(&tZ?Q$ft&PW34{;FmlPI?mscs%J8`Ep
zqss!iZdqmht<-&KANVpC1QVbtBOR)aNAD3d6Qfl@()2#tQ2*<eBE!WR{Iwc>_-zYe
zyN})C^Q%n-1l981dR>nu_KfQIm7(|;ou6da1?ONZw0X*Y*y&SUurXJUl>b1FQ&ByJ
z4VqirN=8*=A-z-%4Q{z@B@}a)q23F>lZc{FEXqw6NKMN$Lks3W8j=2WNT(rf39qnj
zWRy+2wq{_+G#^?&DLC|=u_;S~+7CT%MB3#KH3unR?81lc0nk!8%Z|WjQTk?sM61M+
zA{Gsc;p;a`#~h?<hi<|oRvjeGwG$U3mK;|WJ&T4{UZ<~A0!Ao{S4#cF_-OLZdHI7A
zkvZ-1Ra-5dw%S1r{gTf1R|&!(Luk5=HLFOmB7y>cwHQg+zC*WeJm<n>>n)O~a{QV|
zF|O<J9Hnf04F=S=Zv4F6LwPM%@BQhq+vQg(qqpl>_UkUp&)>=&XWI{P6yUSJQQ9xU
z&;)Cf$q=`l+XgSL5mjHob0EJNDng16REFAezKJA6K2|dpf*a>SPr3@|uzjj3eBy{5
zV7+xR1}q#0J4JNI{m5|Zo{kWY3+H89;Ze!tkkSc|MgP(Is9IhWX!`0iB25Zi0Li}v
z4r4o6C_jZ7E=QmkPBpW;qI6<bf}Wzpub-X}WAflQT(ciJ1t;lk<KYXQaw8a`D-mrv
zJ)4pL{Fxc;U~#xBNXbP*g*=Rs!`<I6%-s&?@{t@zI!*VIXEgam2Ux0}aC_B8R-=Lu
z#!9_gui~Saj=Q+3iGM(Q4UE}QWyi5C8wAKTe9^0GIttT#&=DR7DO~!VQOjog`upw{
zXU{I4F=ubzV1zu@ueSCRG?h>GiT8}2l#v5*Hfqqgh=jr0W-5A$bAX&F@dbkQoC$R{
z8MMK8IqHMjvvmc1J`c8%<|vtQBp(&|ck)T~V#1nO)$by|fxF1tEyHLJE{os(-W4$8
zmzlydBZ48KJYUV{|5gjU;C{jix}`k(Xxy>Q$NpBnMYW2VKU5Kr<l~&P`4VR97OTWr
zGdiNlmNAm%i_x1+w#*0D`_iLk=Hu^xR$vn&_i;vl4=*$nG6JIx*=dH9HM?59nZi?2
zt4AIdwR1?hCM9djnLHQA>49}FN2HV2{}3C!D0d2CxZ@OmxW@Z-C3_vu)GkJL)}$5`
zQWtdj_NgrBj{NyLAI@mJu}PKK0{=yDCkRUApmOVHB5}p-7UA6J5j0mo($lphlt-x)
zkm}bYE<LiSMHw(p`I{V}Hi}oz(ksh*6SMCve)xq^D>1174WWye=ZuH@q!f{7|AXvb
z0NuAso7s<)UXf5WF<;LYAp1%F_RczYSZY#6m^$iLUQISa7CmrbBDwT;KXKBKpUI$W
z=Po7$wHU0V9mCc};iaFC;gbfh`>3juAvvE?>Y~Ak_HT(nE;N(ruTe$N;gxHJWvcPu
zAq4-^`V(k}mJ{@uD|Xz9cy#7cg$|2SDSden*sJDfv9k5qnP6zw4mzPu9dbNU(z^4e
zJUpuZYJ0Mgq^k2oSCbeJ1p}!ws_7?-Nym>4XX*nfsYmzt4+eNCmEuu`aFh?+l-ZCb
zY248F+pBLdDfEppEC5WSK>WtR%=%oKg)>Y^LcUOz>S3>slyzKBP8;IV{bYzYoPV6N
z&W4Fhww7Pt{khBYgUO#AIQcHFf!^o-Si3$sHp|O5-#A$rOahAD@Z=CKdTD-R0;1s5
zAGa=2s?nS}YZQb_B%bIOthR9jJB_)VTi$X1O{3P+kW50S-r9kqeHWt2p$^Fv^<E)|
zNzJ9anGBK2X+Xo<h?cFgu%0seW|VWzd+?iU`^jnrzoO-*d!93Rq4`kly5INEqVdw~
z|K+;J?)>y>1o#;ctiLCVW;Ua2?@~<=?A1JzUq8-{z76;_B6e`GAtd3*+#%K9Z>HS{
zcS4i`{}MKkSt#+D!aD3K@#DcsK;1I!PR@D*eACPIF`04Rs!QXY5tN`~s>C-g_2&7c
z;+@nz9>*_($>b=`S?)YBzj%*(R3D%Ip)k1Rpdp4MD{!TPMa#li!kt}=lMXdAlQROQ
zC5h$6EW_)kF|2pGt&?Yj;{3THv@s_@ARWi?0^6NBxF51)G4#{jnJTWWpNvX|$5<y<
zcrPTGySTbi{xb>=kg%teiQ(8C-?2?jp-OIPe)aI(%C7-YL-L0~H<kF=ohl<5gTX5q
ziXebsE!L`|Zo^A6!4^|*kxjx%zKv69>RfC))yMgnBRd1DOle%rl*e5oJFFV`piH0U
z+SUC#{2sLs`L$N=-Gc_cMnMHkzTZ8iENFa9=+?5aIFCLNE`VYyiTZI?rMUMiadI93
zLW$1W@l%JcBs9C3oS(rsk`%2U_-Ey^fAhRF`f8M1i)*nHnlfX|J+B4KYmG#M8}XGj
zhn%xsXFubH0;y7Ac3IQhvX3m*$jPF?IM|XyJSx8JBUxQ4vLj%%C9h`8j@!{9TqCi*
zlbDa${x(pG5e`}Y4a3yXE1Nt)lZ`MIrHTs{Pxjz1iIW_GnA^U*6h>(~%qh^z&0z7K
z?p5mOG_#6r=^&L!N3>|+v5)ZE*35?`(vVll;;(SMs)RU}{M*a9cQ#`icR8d?e@4^)
z9{^!Np1$R=%&Gkv)2Z$Ek9bVp^jX`D+}}-(YumqOAoxAxc=H&Ozf!J@$@RNFCNoou
zhc#AUk0WdE+!g|e`|(EYo7P<I-Zm!7_~k|)?d4qcf-@N&-YXEm4vPRZVEsHzAQ3F@
z-XRj<`Q06MF=8S{Bv?E@^^}|}%tk4?^Pz-@nFA6BOdL>(V1>q)&B*;UXzHWsIIcFK
z9rD(_EYIZFZmxPy7C|bx%~K{s#2j?@NsDc(hAcf`Lp^y}s)wB}!0Xtedevw`mKuk4
z=r?5;lWR-fNVz<{G1>Y4xB<)gs7Td(N5j8)(sUJqFd9*a&{s=<O~A>fmdN&+Hfd`A
zv4L@XP8)a2(CO)CG<ZE!1(obYbs|$&8rfFfH8^;NNX9(t5w^wl;_DW|{$Ao?b+zb+
z%&e=wDM6$LFYg&)X6$x5J<d9(o<I9t+9o{BgB0Bgu?b{JT_HyyP_cUmeS&U3^%NwH
zv7zs2!RbT5vuDrot6%*oK7Uy8)#qPgf?#F>))goME(zJ9^hzN>2wVz+Den=4LkK9-
zgtSXm1_2N2Vg^4TL`IAOD1w~FwcKojl7436vQ=MIoa*qWadVg%VG4-*32_$@b^$AE
zW8$I{><oZH#5^(bnz0;@cA8DWaynr-K44u`vACb^fn`N{_oi(TZSB(r<T~gxX<RO1
z(&L6nPevj;Rp+VE<hb++DT6ivnE_L7?4%ktZK7TC`^vGLJwOCJxEj{O9^39O>M<(n
zMcu%rTUVtNtm}%LR+N%3&jAG%yxmSMEA@0tvbYi{so}^HDbUGh&gy~G6ns<KnF-Sr
z)o41LWLH~vk%BB47!>oWFgDLBvKSVFk*X@9xfBaFUt{ZSYR<}3a~(p!JWaUYAMiU6
zWL?1}BcdRZ9Z7}U74M>Q#}EVob|C^0c)s5Q!pMp6W<6n9*H)*c^yy=1eXx}|natRh
z!%Z;ZG0p7`t8UglC`rbeSKQyfz%wei3lY1(pdgrN2BI2?OU`YjSSKO?O3>Bb!w&h2
z6V``Q?+YR3e5#&Tg<*#6-Z(?;_f3l*V<SE#K2(;AjVGm4k0La$i>tB>Yx2~MUF~@{
zBFj0c=Oi&BP62sIy*#AmJ7Zl})YE)1jHS$kIN9;;gQ$_oSg}Q-lCsLO2ujW<xwfO8
zk8$y-*vW}H7U#|QV!1WP0+SyblRuU$*UBkY*E+c@C~V%^5C~<<kChG1Wx1xg8&Qgl
z2|unZlj(JJGHi8#;h>;%W8o9v`^zx_G#W$u_1{U3Z9MvJa(t1Ci=r=8UUNb|ljEFI
zGenot7i4l?REHg&FLg{l>!%4eCda5DNypP=i?njpwl*B_U~*8eW<8F4n|3oC=c0k;
z%UK8l2#`u{<9LV@o*iC*a={nB{tce*_xSl={yARW-C<^1Xsz#q>dTZ1l4R_HZn1r4
zCupB$-9?bHAZ7KA5<9`Ko|-<o+o>FvM?A}9^Y3mPw)1NsSaZU0T|ms3ry0Z%?^3|&
zc*MG%zyyRTV%qHyIB2}sWHJ{65D1wPAVs|-iyE|RW$8f=8ppOy&WPCNU%kDnnkB8D
zeu`tV^ZjrGmh;i5o&^B<z^Ai4>-g(B#@5T}OrLBc>e-^-`(C{5ex(n{?O8bM=-NyP
zgW`IAk;24HnF7$<vjs2(I;-KLoDCY)(zJ;SEZt~lhP`jk;?PlxZj)n?W(cTiaaFo&
z;f@o4QkT9gn;XxK;h)WrCxo0b9^SsgVYkQOv(GSx2uT@`VBRzYYRXGN-6cd2P%fZi
zM~PcBNj3Hr5cCw4wG<!$%PB(`V2&UJh}5BK9I}QqT5Z<0Q%_)_;S6xGItY^*$RaZs
zj3R`<QCGpBY1+>_9QJ!%6^>C)doh>lV1}7dp{`|ECnvq&t-XNkC~GnG@@WmRqZ~I~
zLceGVU31spD7MuDZzz)AM~-31sLHx67q{ov^lk2-F3%hEBrY^L)(3G2fT$d&ymhxH
zCOcu&dM@sU_c}q)GN=annilHmA9GZ6n}V<^lO<QTK<El;$p{ugyBg821%L+X1h-Ed
zS{>BFvUEj`kW}#Q&09R%AF%)VFYs)?M<fEvV3^TNL5g}n1Sa)X%vyhuSUoT81f__m
z%02`+seFXyk&0$Sr@BSC&T1?V?a+y_p|nPz-{+zyu5k<i1rP!zkSe&W{ouOli5UTe
zmv_%`Kkaopkp;db0-~MLB&1u5G69)2y<`(!u}N>8xOC5h0vZ=Up4hziLuFY8ff~Kw
z#taj5o7z>*#qAF*%gH<iSSP4tnXfWQ_20upwaME?me0!akqnOBl-<x~-;F~=m1Q$f
zYk3bNTpj4!fN#ba(zr1nOO~_6qRA<nEVqswB4oycF+*pvd{IhctgJtjEcaml3{r=0
z>7{)c3F&tE-eWSlkJR}4`^a%^mfuZ|-%pc;_sH>>CliK2-#zO!>U^TgaT``HTd4lK
zfIt>ByzXY`eZr{oU&OuK7}#5JHMM^nhwJZop783$E6gA~ynTaLub$)i;ea^=1Tav&
z{fiw`Q~;Es2d0Rg&N*{L4uUyEtX%Z)6FXToTYOh=u|@hSr)YW$vsyqm-gBZliGJQo
zY1Cn|6s$=XqC*HMg%JcOYeHFzZ97-}7iO3z3k}n^XqIkF3~c3&c0zkqr)w8e&oga@
zH4xc&TzdMSiV!=$H*UakJ{o;Or+-uL0c~LD!yM6Y9mc`<yZ1*T7DCJ3)fO?Rk9h$4
zgSUFQJ!4dqd&k}<G8#`ny-d_q77)BU9+6W96GONl1XGSl2OohdR!z}p$}$Gy$5~p2
ze^bq1?W=8-2&farVIoVdpKs5vhBm2M2B9fJmk`tgO&^e!gm-V=;3xaNtq|!+C{$NO
zsIGJp(rRwRY{QB!+o<smIvDGb<^^QITBR&VMJMuc4v4#H$V7J~zEp?m;A`c`WL{C~
zdM@TJAWR4`AjGK065B*Agb5++wN7%u;eL<%!vXtwhdEA&Oo(6viMkb1bo<7ru+)_!
zBJPHRq0Wc|+*Z5Jj=8>ZMmZ*etqOK!VN?K{BHcc~AoJgiYjQl68x8*+Uq_o~T$}LL
zv}(+Rde{P>f%^%9<LQW$v(<6;3azJdsT@=9qinTTpG)9t;lBp)R<SF9RL`m&n3kw}
zM>2J#uJw;s6y=&%eJ2kJb8m|pV=$nkf;V4%jh`Iu@aok|Jew@obE*ozZG9+G^=B{#
z-Of-5EYxDt=n5$#2G%VL=7Cx#*#a_;%)YfklXK~>(d%yLplt#!wmM6J5vB=35db4X
z<JG~ogRM)3WI+Jq)w7qlpZ9oJ9zZc5+oX4fZ3ShW#AXJPz>M5tlV0w&18cjRAjpCq
zH2P}mx=p^TEZ5I#<(v<d1TBYheduc3f-JM0euZGh=cZ(!P1MH8YFkvQWiX~?_-GuH
zn>SAz)Aj1w?%P70ibWffiK$nL74j&{?L^x)#xf$sk0#3k?SIy!6U@Lh8V$@4L_^nd
zUQ?rrA{XUVw+{YcWLf(Y=DMxFrmVEK{1o`T$K*3UT#oB@RG@C}tX_5P(cdD+k3((u
z!((!sZA=D<9vI4+!RmQe4KZVQ3?kdUW@U4pxa~*H%-T*zuiuZ)tJifSVIPx6gS75y
zuo930F@3uJ%+n69U%p0uc);7wKSy}=49^dH^VS~11C$<j6GecmvLeEmgBh^mgi|VD
zteQTl9&p;3uyp+eX3VbjrzsqbSn6rn<F{Us<5I9n(uJ`YA*=Q$dCACYMzL+3Ok@jm
zxsT_p<x40U+bd2$V2Jh0LfvyxPg|GX2iCrk^d#8Q`tfb(6X%Z~H()s*jpo2m>Xt23
zUBN8ux);1T4%7*lO%BF6SnZISlJTA%s_KyMy@1aCelO<HBR$9=iMpw?s4F_#3U1Eo
zI^1GlP#wRhcEdwP_~Mw|^ZZ^N{p)^qqRk+Co@W_;iYd$@(%#v&CDa$S#0)>8E2aDW
z0nCJlx9^bmdxYH%6RZO%^<*dYE~-j-v~80l(gsJecEBL)ryYoAoJv6|0zp8@1xq?%
zPkT%|^~}=MGn<4%w@$V=14Ref8V_vH;l`S?SXc~-8c15*#t9r^+t$adaW!T;@hEW6
z(~!!HiGf{A$ZNuKI&DtD>GeXOK(>vun1PmTawV<p5<RV5grgjXK|ZOrzbeba_P1t~
z9Vhuev>XrDec0qfji+RW^E&Zl`%7%{q`HtAl!^tft`pBv^M@{57}EKuvVK1E2T%_m
zC?BA0RlmLFy#Wkzys<HCu{Ov-PYHVU@)h<W;`GfoC`<^u8L}o6vRR+?4=^uNf=+ZL
z1Cj+uIypuG0tL)**1)T!fH^=w$V&pqWS#*EWRqSEtgAA~n@N_YRe!4Is#uR%B%@G4
zAz+#(>~=dynNTDkEi0DO5fAU)K{9Mxg-!6XdT_)kBFq6a6Ap(xUfw^)@~&WA)=h!6
z4P68^DCC|lodmIM#kdxi@M-X&vOIe8RYOQay1cP`!0?Gbyet>nW?L6-v|VW&mt42R
z0o!44NXsTUS7f;wLRKrsO}@wcnUSDP^s8~H>Izm!On?yScGY&X*1{ASyGcDa_WW^V
znJ9u#kk(@}MiT?X0a6Ght0&Z>vV2DCGo`k%A6b^$HnAEKn(gx3^Y9@J^hWr;V{-L!
zWitW@zMmX_TkrSba$Iky+sV0YPbQf6&LeV6O|Q0~>1MzZQ_|ZZb@f@LseSrNljCh(
zy5@NKsZCxixcR7C*T?D=@*a8UTUcWx%u~SKZU-X5{U@K`Zk|zcMiAc2ffz9FD}b^d
zPLQBw7K;AOq`L|Tda|EzKmj3_f^}UG0%JeVnz~YSpS*O#G|V{uo=WO@Hl2Nu0xp6H
zge(AMK{;lWwKOALAz-qD8juPKWu%l*a)P8}D;tC?x;65=+*==3jro?6x=2mEwY@L#
z&L1Iez;ZqsZH_-0uc}^b18D8_`$X{iNP4wvx-tQl@gE;Jcz%1gH*}7pCf+8kb?ev|
zD0Vxet((V2gOLos+pu|MPN&K)Z8JA0D(6O<&D~Nbg6CPHD>s}Ouk0B_z&y|35Rldt
zGEcgCNxBMD1F`G1)K<I-5y&P7O_>?u-b4-%B8^Rg4A_>wRD(cD-3n-iAk#KBuFlvL
zs{OONeh*Qz>fmgv-PNGPWJm8pSEraY&QS~+|0WP4AR<skV0D8|Y2CV$qi-byF+}oE
zwl(c^wRE2Xqa5>@T!~cWxGTh0C%wiGEys0|^60bQ`|djNwK}1$3Q6fflm#X()F%SX
z(hh7VKJ{5+U$@C(@9;)t@!O6sZ{>WEsXAf45|61S9NP9nU>xr5Fy(^0EI=++pV~w8
zZPR2c`*xUu&JeU8%V_R11hUu`_I|sswwFSs8IMQfIN7XXlmD^1lv+3XBVtnm((NQM
zPDraRzn7emQ$|`=1d192+8!ub>xlx{4=_y;hut0}zR`5ma%$;N6B|=CiVsDN+K==Y
z98YJ;t9%z(2I}9`%Ct@R&L@87^S)tDaSO6M#+#~61%l!5gR^X$XJ<Usw~5s#%l5Xq
zpwv>eY2z{&oaJhfo&Rg|5$OZ<G|_;&!$*{5=;?#S;wzP`+y4XWq*pz#xOtnL%koA2
z)Y-U|^^Y&hxp&#t(7J~mwK8}re5f2Z-CAnBY}?jF48q6c7+v+1N1nBL#CzrV%BSZ-
zWA(vT#^ho;Y4xhobjj8@H9pqLVXJq^&_&tiQ9%>*&1gDy0%JS3pB^KOw(lI@WiKOv
z>Y${jaR$~yjp95b+}&Z0QBO2&&pSJgK%hqZ;gAI?w&Os$ag4zTB>EZYZU<9<0rUJu
z*N%tO52?N?$7l_p6hX3+rUP`_GKoOe0%Y@4vvEF{F0G!RW8B>yV=Nazk{ZYMOvi6&
zmDWB$wUm0zKE*NF`Tn>8%lUW=t3<W8uYG5Mty~b=-tMY2m#lw}O7gj4*|2e^(BNe>
zd{@UDp&oi_d`8nl$48z`E;X)G4-lCMnCvM`>SUYN1xrpyOGeHbB(=GBb3O5>V9lw8
z8rMPBf~uDS<f0x*Xmf=#Y(TRwGWR?a_1L(oC~HSwkAkm?VBO--6nb;<<%{<y90>Dn
z!l$2oiXZ@|Y(t*dpoCUf1VV<c@G|Mae^dY}0!aeN2_>z_>w>f{D5RSPrg;L(gj5y~
z1j{={$%2%UjjzE5X4dTtr8LDE#7(j4)!wHEkr_0p!A(lBSQ$VF5g|lddDT;F7_2Kr
z^A5Y+9{c?cQ=CCL0ysgCkaI>#J?MQ8EL$LAfj5wwcbKkR*Yt-r@!6L5jWfz|`=^$0
z$)Z~~U~R7ebqi(5vNc}g_m^XWu76%#*`=*0cR4mSvnr<qbDc-?aA31goz@j=NzI6w
zQJa}QyZ~5p!J4x@+{yvwO<D?&N?+lw%CmG8tR>fCOw@QAH0WkPo%5+;u3<oWfZpmq
z@R-5B707UX3=#MD&#*6o!<w+4CXfY5)vY_W!dWzLf&jLR6u{zcs7%Y6P*Ot42~&um
zX~t<;A(9b7M9v8h4=3#QI~?X2Y_eWs<uNmaYX#YM2HVV_ep=EzEnCoD5#1*}>pC+B
z(k;DApgDpv0~y%I33t<sd|I%8kyk=WD{@ZixpMamYy3)OGmFOND*$CNLx9S0s{N#3
zbHZbr^x_FB%-=zl6>Dcn>zxO|JexPS58D`D$NBNPKaecvT-uz6Ef$QPgl8TdWYJYb
z4W?RFD$AjavaA^-F<*x^nW~E>)W+Fyo~2GvVqWJey_<3|V@fO7n(6kn_8w$g<2W$x
zKB6o?oDu-Set(CYS3Eqt#csbtoDPT~BE;Czt2Qm($a0HkCpv4lZ!|CR2ax3<aJG8t
z5eoG5;V0=$`GIoW^lrOGJ?{FrEd7!#XL7ua+tzPVZ~E^d$91cn^`F)A(Bv;kLAHZF
zbnDsWm|Pz`ZpF(|kV`gCWof!~hQ+2#WfKrx)7}?JH##xY$kI;Y95VwO@hy<vFtyPu
z&Wvhf+1E(+%t5!XvZnR)@PH{!cyV`!ISBNy1>MxzZDv)12-$i%0)d6qQ#=*qHGu&f
zc6)FsNFqomfMhj9=Yq6mosWcwz`DzT>^)3^8c6qCcd6c=8f!}BIi(eEPYI=D1Olds
zF$cyj70e}L;(}DmUo>u4Nehw)mgR_)7C<smNjT;eZ%-$@Ii766gi)&T(gN{o-(qdt
z3bh{A@TA9N=X>J@Ea&6#NG~#aYmA-0i0#>~uCmexzY7&(92n{VTYsc0j>9FJ0eduD
zoj-eHyh0DwuCJArwZ7Vcw%thN@n^LYOl);@6Li^><E;|VONYjfsHS7+s#}@X%-3}G
zK(*&(0^mv&4a$Wn)xl1BiKeW?zzAHo74%h7s@po6aj8!fY5*|Z7eAzzYWy%gnW)C-
zD$<WKuk`>MXlp%e6tA;_C)n1-X0)s2tJ_Q>2aWv^0#b%5MyA*v<{kFC9pV&0A?S%@
zAcW?9p*gKyX6=L>P;+d%NM9|`@%K}~GC^k_wP3O5O;Mm*7X~@rJkG9_;n|xPyho0K
zp<igSpaxSy?*sbd`rQIK%R`qff+@^)0;R6@+IQ4fg4LK&P^2r;7oF57$CgRLWj<;Q
zo-V2VmTvSj1^-NrH_t&Yt9C^_wSelRUFCQR0aGCCSPwG@mha|KVhOF@RCTUF!;7SP
znVM${gAgbnux^_T9NPg6>fK@{DVqW5t9psleAR~0-eqMuN)iz?xGi+TTThV^GV@o}
zZi}Fl49OWKlbr^oF*w*7ix6Gb%SBtKp}(qIy|$4yo|HI$M_GoA$z0|6tnrnxZ9MkM
z50T{|Elrk($6m;e$aZW!lVu%px-84^9ovm+i<044beWIM_<_wkHkS=)sMk~-VNA2m
zVD-CFJ*dxexT%jpmTP|5yf;3oEC=l^V~CJ?qFNC%G*C+@)xtiPWk{PC_ZpHeL-=^I
zJaipH$&UT^pnE+f**)W$xBmG-9#`c!U*^5F!$;&;(B*jZ;gFVZljA;2UP_Oca$Xlz
zo3Zn3OdjQ!TAlXx+ZFpkQ?DI(z0dwUPxD4oXr19U&Gi*s=d(9$I)$yy0jVBXT{$6C
zJ(^Kc28Dnr1O&>%BA3)PvL?UnHPW9y#`^+l{}3PmLe3~$7g~Xw5->$wcpfXu`ia^F
zwnbp;v&j0$QpaPe+ptv^(tz)3432=!6|5dfcRJ{)>jhGD`{4O}DV3}XgZ7TvCY#>C
zfaIZ{8FT|ZY-7j^oc|nf1D5mA=m$BQ4hD4Nl?8RtlVH?a2L(tAFeM=5?yYbwM$9-r
z6!eu|41a5A(Cr}F`paQJXwM-Vbem>un^;P*XPRqo1~_Xc9ho(z(gH{p)PT>BL1`3?
z8BsGIn48Ce4%X@hUiY)t^lR|qgnE)2gdh@->pt<U3Q4pwT><O`c6~9o`p}-&CIe<v
z5;MM(QrdQ@3X?97PJs{@NNK@-p7F{3Jzh=`_mL5JuUiw?w#?-^5$Ho;2mxS0NDGh!
zX=O-`2)xHq7;7f%!;C2~)~u%=<xr5J!JtD4&C`j9P*&Y$P`4a1GDw0RYc2ptEXM+p
zjCU#Fw5+-UUw}grteNm4KEwU)8K&p=NM9bYEDw;B5QOljuYZcy&tG8o{0{tP!s(eH
zO2l&<u)n*<es{oK+~br5zl$1FG%|w)m;_i*Y-Qh82Xod~D<%`8UilOV_Pcrdk$S9^
z0%Wqb#io?mh0940OOLHl4=$+BQIE*MU;%CaV-VSMIYzs$_bU~Ry9Qu5WlZBE#BTa2
zsWwwp3X6bC(PQUr`{=t9P?8{-YNYR0mwaTxx)!9AHNblasO}$aHF(I&m=S7Ank~h;
z(o8mS69L$i>gED7YT4BCoKcXaG=p*raw^rMv!Jevhf=^SI7|VDIbatEchiLD(;m-3
zxQ89hP2IR$ERsehfT%zZAkf1L^133Ygj_^D>=MkN!`SZ*$N-Wsa01@G%SfeQfgW-Y
zL@?K~Bn{Y_N!nKENIHQY<HYG?)?%Ij*cZlKVdM<FT>*Y0kQWJSc?XmUk^<hodxN)c
zzJ#O&5(COA$fpz1!x5*`32&DLM=DqYG%j9dtfHQ%wf$<p+`V#2vDJKQu{~Vu)8f0!
zvS3}hERP0|>YY(z8lF!QgY=2<`((L5n^Cj5*f=caVO+=L?!gL3QBUUTWx`@(Axb+k
zp0fa}%5niP10V_-E<AIS<y<m21ogNm)^E^d{Sq{v^<+@gi7Nz|RY%}DsbbAP+ev={
zwDGa#D*;%c{m;(wm?6#EPp-&vfM6FV?Bn!NW%+J*u(S|z$rhhRu!7X%A{9IfRhH*V
zS%!e6AQbJxG8LqfaVjgaq>m=ck{J&yYTyIfA{NZ>mW04XK<R`ctmPwvG(Qoct`j$b
zdaZ>f$2N{oAA<q_WWky<CX1OR+%qb$s>f)CRGCse)UF%KrtNINl`*-=F`@1rFC=<0
zB-I#X!-Ix4uVY~bVYR~zB5-9)*0DW6vS3{^QclPv>1mlJlWc9hE-Kcnskaf({oqw^
zSGlb_EiMgSs5?bWw+<GcwO%bl88!`Abo|OFfJ^`dAT3}9UcCMU&v!dWNq|I5K@TOE
zgqsnX*y7}ttrw%fJd^7V4$}d$2$BFCC(I$B-~?vDX;H7}cMor{#~#ykk5UrWlo2Ut
zeyltS1DW*DA-3Pqc7|lh$V52Kr2S>iIAO)HtN^Kp=zWOT6G76dY79s04+q?11wcqS
zs|VJ)KtvEi#BRSwl8obeLdr?wUZTe|q`-Q5sw}#=Xt7DNbR$?T|0ktfJHHohz;Zqs
zRXJ@d2gA&rO<5inEjFGx_0qm*(G<baP&EcIHnRmTg24}a2irg+{he)f<C^kfiLmhm
zZ8fDnOg~b?iy68$RxzNh6l}NqoO@fHO?unbztJP1Px!|)j6r6%igh=}we{R)LF_9w
zn>5sSwmjG>?Ye?u#4ZLzt}&fd*$rl?l%g{Uwx1AmJ7{y&mcCzIJ=-+hOl*^e0Hy$D
z?Ur&*5Q(i&NMNfyb$zJ+uRCs!`526KHn}GR1ZM1}86gB+krL6ZhRgu#i7Ffda#2G?
zE(shN`?$vxBbY;Lr*+iS>LIjWY)XcxSIXEv>#^)lMOm)OY4ynKEBZE3M4!mwLLR$Z
z@$mhq=ruX1_sVrSZgEZ^%@7XMX9~nDSJj9rXYZ)DQ>#0{Mg#;cHUozHYCV;4rPr>w
zR)tMo$Bd3bI<(cEe>%J7{F$|d*SrrG%FwqAvy9s3Hlwl>q`X1`BXZP9Kg0|J8t5G&
zU}B^qI^ku~iA)*t+sa;LRt@XG6ea{B1YT7}2(~?v>>llm3DCZ>t+d)tH8s|bkpL31
z?G_9U0fkIywb-*02U`$O(~_P>rpkOO^*{}Rlnlu#>uOXkSW>dBiQ93c?f)8#FWOv&
zGz!eT^|biGWZ9l~F3W9L`VO+Z`L(HsVN9-dF?5}y?8&G+*fV;-=OL|9WoVFPOXy?W
z@i;8EmtU*+_PG|Idi#vexAEaIS+4p;8_yn*WixJlTv?81vP|DmmZiyZuR>~`7(ciy
zuL(JiUO2UF%215sI9x}F+R)|kdc8py*cz0Gnhe*jYt*ePo3>2aW`ON>wNYNi+Y6|*
z*}6saT#lQJ)iK$uS(n>1uN3yhm~781c0%viRg4}%gH8+dxKG<$T#;j(T{C8->I$`&
zZ${<r6}5dtPX|eFT{KQ$P{(&mC+c=eHK3_ctr~@^E?Ey88tx+tOJa3Uy~W4VL2Jm3
z5NziJn?74IvbmPmdaU`ZvyCERTl3CK(5|DQzps;#Qow|bL9dv{90=I&_Q*)OP>>7O
zRZlN`c7Km)iV!=IJX>5Wh}ag(O?9C4t@ZsP!+88yR5s_26*pixAC0ycf*we_x@1((
zC;%JRWGQG{cQ$bbVV#t@5^cJ4svcOR9l3om4&vv#ADZz2`aX?Q^+-NgpOlMw8tH&4
z?OqXJO$kd{H}};Jwul+G>~uN}WG;htKBTp#i)~WT%zVSO+aA5g*QxiM=?dq6i>->v
zFj?<PrflN$WJ$=GupZxn=7_r(F~RDur~)%j(Uf)d2H~RXzKM%%JwQ9MTTe<cB^VkL
zL<6Ts1O`Hwu#0;LRxHa2yIllHY(^i+Sru5+`!{CL^P1Bm!(&7i!I~4mGh&RmySu{_
z0*()F5lX~9&fqBkQvmG(5P{?Ih~w#qWj*1re}?_@d+g>NLI_Q%;zCV1Yqkm|i?LDE
z=vK&d7;bLxn?k>Fc$ieyi7&Hk6q$@cUe6}}u#cjBSXXsf)Ptf7&o2TLx|HMAf1s*{
z_6`A1Y~`0l<4X`A?KB(dlN0J{n%*nz1qC{BOF3grYg4>bH&YoYrPzvJZR|rwgmE(5
z^=UQUTOw&I#~s=#=bIe%^1F%^8Wrs{YTi_ttbKwi`Y9!x9^N98U^>hQro9sFKpYSw
z(A;XUVp)zF*qMN`WD6F{AYqiEE8vkMGBn=VZr+1Jz}ts}WnFY4Wu9$7vE$cOq1P>j
zS&!4M|D&P@KaArB=>%P(D(EW}JjenL6Usz5?04A32oc>*L^GqXp3t<U6?2F{3F-xz
zRxp4$%veq<-n~0wO&Pfq<UkPahlZ2IJTzZcO$+LQ<Dn_eL!~_lZc>)3%-Hkm4ku<d
zqw^TN#?oMmD0KiH$7E~OSO;0w2^ApFxP;ZK=zX%>bc~@3!FQ15QYV+CKj4~YSHorb
z`^)mf5$md+)*<$Zc{QYMCf&7-*axj{pOhTeQKfsQ>vqb)EH^f3Ypzmdw)JFsaSgSg
z5j=-o57OIf+vOm`BdhK+7@n%VCY#WcwYzI_+>%%C9mZr$Z?U+esTnGCejosur@Mux
zpn3t&EcN5z4Wk@WOV7D-tlw2bQ0<udf>~b}JNL3~eaItnT*nnMg}oZli3`|%g0LPs
zVEqf5UP6>nkMO2o0pLY6OkWb@BoJX+C@x?<G8EIi1BHm`VMfjwOG=nZ(Kvji7F9iy
z)3zCkdU{r4Xp6H8K;zM>M{W{rD^xN#XT*sRSq**~@3)|c=H=zH7dRYth%w@HJmKNE
zAcTmY|Kg`O>~~1Zf|M-ISy7+asHxc0P#u$-_wS|=ZR^qb<HrqH&PO9KH=~9sGSq@R
z3$WQoJ{oUY=S0JI^*TcLsA~67bD16$k*Z>jW6Y*s8#MAhUR6a<JL1i|&lCb-YAa23
z7tr|pKBjGZ)23+Cc13W^?02zkY@cq+nU?!&ido<8M1vA?K4Yosx~e*Q8lMexwoO0;
zEnZoj%m@*fLVIZ~*OE)y)*z<X*RnnxA3!01X^+SeQ`qU`BQHp0hg>qqR&F!Zfp(lu
z4W%{c9rbT&ik+4;GvV%VK;(dT-@L^^4%n4F!cq{IjJp!Cmx#1x3jmze=(RtncNW=9
zLI5fSAobj8@t3H#JfQ+mXlcB<dUxaWEw5c!27s=~owZl#a#yzijden>EkB~$md;yy
zUELk8-@Na=$VO2W^nT>})z(I<$3&Iovuldf5V->_CI?;C>0|TTt0@?J!TOf|076~a
z?0;ViJltZ>opmPK2MFqPjtCmO)e5;90c*Wgl{Qf5UKc2X60fVoWZs2E1VS3uLbpBZ
z6%)^p0xau-FzJ?L(G4J|!469?g&w`VG>=^?4?P?lx*aJOHDV(JrC<gj3PDl<2c0C}
z9S(T5K1YZXz>L+rV{6O}f_jaloUI?K?NPn>wq6_C32*N+aare2fSZx!?JKtbIFs|{
zpfnC<a($96LtkGd_I!+OhO3eH$nr+i-Ye6Wvb^mFG!<1Bs#m!z|3I>AWmP>>x-Qr2
zzsFVrs>d#<x3Q;!%5l$2L(t-gssyx;2hS+e5UXqt&9pC$_J<D99zjBVp=SHq_jZTc
zXY}@eu3KL`i*zi8R5_OIvjF5~G#fH797VXj>pgnF*q2{iLrp)Ze>d+{i}_mbs|USo
z)X_nXo5pSB7l;u<z!Vsv$+|taZetdL1Ynv106l0ynu20AV~Qu6v8)=F34$q1U}6N;
zoddPLOOY1VL5<O+#fBsE2q;+)f}YxaDIdKbE~1{46M%V&`0TUK@Z!Y_<eY5+#5l}5
z>=3boAg>82XUy{szk2m^+}}UbZRBRa600m}{akI5)IM*+RWHtMReq4}<vexVfaQEN
zf~~+6L3597Zu54{xIzb@ZTYJ!K-Foh4z6dqZ5wcF+?3|pd_JNMgVgym>SuLz$<E;z
z#_^`wwnVk~QTALtjvGDZ2leq(Y+&e$`nh7=6Ixe2E_si7PzTL60EYgBj01PwGGZ%s
zR99YVKDV<r)d2|rBa#lXRpAy>07aljToVm-T2C_R%52+WIh04$wvWdL5C_b6_lRM_
z6nDtwp`A9A>?9?PdoVT_9qh-_IiFWr@=>QWKsR793*LVH7Sj}P7k3D2MmS|0Bw|-0
zzFAL5Yep#rF$5eAX22@iIs{My7OP^{%fPlslEp7MQ!qcEMsJEh)FQELf)Wqc(C`@n
zTivRDF5BySc&AJ4@m!8C=7En;y`H4@3m=fb>gNdR&KPwb+XBv486Gsb-oA%mvNiof
z%&+seUH^CgL`)XPvnt2DZwwD-S5*0ITf?xe8}%O12@tdOajW$cbwWexDOo4IE&4`H
zoOx0LWFCd;NtxTES3B_-CKMqYPYZU;8l$EqrU1j%BlJgaVvs=8co%A)k!|}B>8VB&
z7(obFbqm838Tb1;obofoX#$9`=7iL?i|J2Y)vsHUgXOJFeTF`!4JB<0l{Si}BEKWI
zBFop_`xwo$L!^5zSc0yN$zwX}T(56w)cS8@5c;z<JzbXDqj`APHf8V6iH$vV4qFSO
zjkzw%Kd3AVq{UvVuN)1{GW=Hu2kHjvnw2Lq@nx&aex_w_vwa&^-{d^37pft%$|JfV
z=~~uR7W8*LS@&8~Ki?&H@hs`L^)0fGQ}YIBU60$mt8n+=<u(1(I;VDeEUt{nqa0H&
zmy5iN(l!WRHDHY&5^21uQ4hY5W9faah{|nbJ-IIoG08ycg;f_yMF>3U@3g88ogpnO
zN7ef^6i3r$0MWyWn7AE~ROd4}>!L73%UQ&n3yQ^lEU*(FgS9moios+?v@Wyg50PMs
z0iS*L8J^wU>pQHS9d>(g04S|kc)^+r=EDx3{QOhg$9rfvEtBKW%Lt;oReIam{?Oow
z7;1e!)~K96I^2Ned_+uTkYd}oNfkKW4jM$d8X*?LpqRg^g;A+`TAR+1piMaI-)sXB
zlPNIJ!0>DXA3+0EZaTmE4APEUG<nuBUdc_J;M$&ZkOo~96s#gxMF&ZODHvxXV;yi~
zjArmsji8?XQOlyn4y*S-Ebi52bKiCPseh)-@G!{Q_cbW-#;68ZkRa7MYnjD3A%=*$
zH$U384uUDp6u`k0Vv@GaAcTk_gd!*1b8pIe2pXfKl+prE3kw(l(=-Fbh+zjp)P-Gw
zMH3avpS_a<>WP5Wql;kKE;a9ftO@IK!n&TYt_zav008&%j73g(^X3iiU)+Q5C&b+j
z({6`cR;0Y31;s82Q;c}FKY*Y?DC@+CL|w(}sfTC-s>PS6merb0E4xSZz}tl^a3-k6
z`b7wRHLOh-32m=IH7JK6&ls*Nyj|Ue^!BfThV{ED$LC&=RZeTyqVie3wP36Fpf*ba
zh&9Zrs2nHj9L7GNuX=Vl?k8s2MOA*twha)PXNTQr52>@-S}xsi(u%uHY%epT+eP?n
z;yd<g+-8r}OLU4erZ8!Ga#7F1iBSSId5$4M04!^<ZPkoiU~xvY;OZ)G)f6@6jMn1m
zcm#38H18lpm`-;%$q7q(z$7~mLW_Z;(=XPo#jtG?P(wFazfWrhgE2892SN^l6IMKw
z1$#~q7VKU=$MozDd1B<q;K=y1KlwBKqWlamn2^?tZ@&HtLWFsaAVTx1%=LI~5%n59
z>+_4%bwFubbh7pTEjaL#5@O>PWEs>pWCCZc+-~Z6!{gMxR>x$Oxq8UKQUq&tfF17{
zC(|D)%k^zCJmdpqxlQ{BptCXDW%-AdWg+uG;yNB^JycmP0zEl7=yv{;Ga^*BpO!3}
zv9X?pgsOXXgB+VqD@5vDTF~0&;5*c1zP90^EmrNZ%CR&#*1++#$g1AFY4x@Fma<up
z-a@u$D;Hz3Ko8ESx;1zV0U+&P09s&g-Qr5kYpWTW&g6KM|0V;Lz@}wfnFEaFuba!e
zCsy4aOv5#!9E-^^a{x0i#R<D<#$mt5E=KDI0tiu06CF}krFDp6h?48pLG2$143Y$q
zjJ#%)1cbN)h!MgLAtVq*6ai96V9_{tZGiza^heM!qSW{deJd^?V6n4>Nq_)IRNX5w
z<7clv!92}~b+=7>hMv4QO%t9ye}?_D171CQ0XYiZy*uKVGL|9`8!Q50sIgjK8Y4?k
zcRtubI=Kg1asKnc4Oq^{<18p~b$M4q%5WL%n0mr+@&x)IYoq2x=hQ~eZYOB*pfDJc
zsJbHq<5+e6sJ@bOol$qy&YVzB3aUHlHtH$a#23SjPz4Jjw!$&Kk=g)JU#J6Yi`sKd
z1L&`#I`Qo(9#xU+tGyTuZRX@Wco@{Jd%axyAWe0ZXv{);4}(~maj8DN4vyn$ZT-a@
z5ZES@2&lo&samF_Z+qDouWDLrB5Hf7?#3mfSTOb?8Utkt5h<N;I-QWEfI>i+BElSX
z8=3UUt$JWF_VWx%q?4tpAW2v9&PugsjJolv{qEo1K7$;$vgR6}!6pc$PbO<waGk_T
z|7=v2VdXKFU29usa@^yL)y*hq#;e+-n+#tSk^Y68{Sc9DbuPA=Sle8`b)59}_R402
zu3I&R**%@D?2moeAS<Xrj?WbQtsIX9M4R#$lT_14QiH%!D{Kx?jmK8#YQV0FeN|rV
z6ry6@48%I26*O6FeMFr`NcC0{q?AE{5a)<o3YaHIJ|UMCBJ<Ecja!)e{~)EsGi1Pm
zoroGev@=JRg0*DKq(-IP{Q+^Wr=S%E3KL#Fe*t1aI6mMgM{SeLn5U^3&owrm#ZEyt
zKC~{x_Jj+1L#Qj(<H00Db9frukSw?Md7i}j{Ktm7K@c|{l0fTgiy6PFH)V4Sd*t_3
zS+3s`fv6|+LuGl)P#=3o#m8m&hm_@3L$&Xu!P~!i*jcN$4qo$2ZF2r}WO-<hsCp$e
zPal|`(H7dQt(?o#I1U21>VY=Oan)~`RA)CIsrLOv2Y{->J|fG*XP0BL#ZwY&4cSfR
zFUz3Hm~1ZYio&@ZkK)^Jp+Szi9(gINwPdNu_}GXp<hYk4m<fRcLI{}Rq$in^jxjXI
zu}If!xBC0wMOqv0FnE<Ly0~5d3IwiO2+f+GONRL@ZS>dvZ+$Kc5L!B<)wM3*=z?{#
zZbElAJUiSWP|yY4>N;oJaB5)r-F}b#eviX$hlkS%>#}0a8A;A!@)DZIV3DDP(3sd^
z+<sbCXy^CD4Oq@crJb%}eat)&Vj#@Z3;=ldcG1pu?H)lfGD57|+XRUUN)ar%D!`YD
z2EkT&NiFw<foSHIWQG(z_c;1C>Ogchc!mIkp!>?}5_KV6&93nmqHMRz)nM**wHH!@
zJW}hEYd+cteOZ{O%Vf1<C&ETq!*IVq-GIsV*az;e#00bx*Txf~1PmcYN$T1UQ${Ig
zASWQNO`)!<oGBNWUo%S9ziZ&@l#&M0t)~Zyw9}u6RSDALx`E^Agi;Cu2gEQT@&Pf<
z2vY#4K#+96l<MVFpvz3%s|!>IS55kEiimUvDuVU2K%yGKW+t3Oa9k3k2qtDcd+`h>
z&On$@DD>d{IpMJ1<0r2_!TQBpJd_1FYam;W0SI+VWq}kRmsR7elv$MyTP3RLYs2u9
z8EY-pN)2r$z%&6<B*ZWS0zAB3k&0kRDt|FtPG}SB?WyQBfzUN<qn)h7mvStuuJcG;
zLN;^4;W;fxwDedp8~xejhAkL0hoD(w(m>QrjxXy;SL-zfNex~*25<*t2qJu$ww^{!
zh@i`{oOd==F-rv;)GMJGjqAGyZ_2twNVgHI+<<{;ir7sv?svM{3lxw9P$7gcfrYUw
zi+YA8l`)C|p@`0V%Yu?jd1WmxB0|Zwy|;Dpw%w=*SOP?VB(^PJ3Z|Sh0t*V;wgv%8
z&WKQjM6yn#1G9S2t}6lo`+3%rEK|lrggEW+uYUQb2r&T1gtD$Ed_tO65EtAtBT~RT
z&v^O#1^)b3e~SCVJx-^Dx2FYfPK$a5)R=ZvGcN?N9)Dg6kagCKf~jpuX&$oll;jw=
z6<IC}^<GUiRq8QY4Pm8V6YH|4emgpf$(tS^tL>hHo`73rITe#-oZXD~%X0nq#c2G2
zvP?E6t0ohg9^|t8ab#JnB1)Uwl8UyC0a-*3k`Y*NX#s*x$wjX05WQ(sua5PQW&sX|
z3G*%jNxfc{<BBy4-mTJPvmOSJEgoq-xJZC@SU_84!_Y07{1gHtl6tLyAf#iRq@J?2
zpj6C@vu&LilizjU>H~E?ChI{BeOq2Ox$Ngs*Y~MsKdOF-;Q~R|H^=*Gr+p^N<3ib1
zVdw^=YO_I652RqmJbkQK-Y@02p0LfFKm<%7U<$!@GH`nbGgqIkcIb~30#U#;O(-QH
zrL1x-guoNHB#0!H`BL=sP6AR+U=ViG4k@h&5+FrC7n(I7d{B?Dyk->9pJ3^Ixxd5x
zVUI7q`3k8dz_uz!01^U-CQQ4CxQmE|feh?J1VoUR70eNlY_QNExrkxHr=R=;^KQZ+
z>~NG5*1Y1~dcxb~ggmnzs6uMc(-Xn9u5`9lw}{FU9Dki35;tHuAB_;y$u>|ATsqAG
z^TgOq69DTpv!F1W2M&Xyo(Py$-PR+Vfn43CrT%|F+UgGrP|#iJ&&Hs~SAvG`fyX;V
z)l>V5NOhMdHdpAXw2!H~GDz0v05u-P*^A_Su%(uO=7Z8!ie#ezj1&6qB{Z&<_UB$*
zv1rS1s-!S!5chg|;8;*Pxb`3^Z30jqRZn38S;6#eDno$)0CGyU4MepbLO@?>>!+gV
zdqp)QsG?o>!nZOqWo(K8lev$VoYf)$Ah5*@Kz9P)@AkMat8H`Ccmj2$mWeUNhy@xy
zpj9f>*elcq)v8f81Lj%U1REXk=21CT1F<THf_aYE%_^@bKwh%O24hvsb}=9qGgjL6
z4m~zq`fRw?rL-o{0?l5N<EkuIbnETe%=%2?uLin3yRs{$ypiLJenz&}pFdb?#n$u=
z_f<LE`b1*1QC+01y`nw5e@V@X-zVMMv{JuDEcQ%b%poFLFxNUUsT1$qV$0aXxh7D$
z!dN?)(LmU1Xvm2G)#pa-KLoZ_X1!xGn*ksu@ucxIV5|I4k6exAp{FlF)RTia;Q8I1
zzN!Fi^KLeyC948F#fZaxkJm3=VmHkwrC`k&OU~vEStsnZd^ephs;5$&AlF9t)Z`er
z9a%mf@UDo{@Lg|fO=d2})U6zgoz{ghT*0|e;{CGR`||VF`dww&-a+-fF3TTBmQ{|o
z5td~5@9+uLa@7x>k_-d3<#+g-1ekXbhr<r}xT0hLlNxh{x|S1I^{C@1X&`U1zOCsq
zXl=2!flZuQ^vimC?jMe#B<*q3J_Bm|t}oga{5U2v+vZdz3nYGSOz!_;VhD3vIIWAa
zeb~IxQa2q#|JS2U;aq>N*SGT3)HS{DT#l^{^tZHk)m;r@$FDc7zCNdPeWY%Y91ZO?
z4fO^YRvT&^GX*f!up813vam#qo?X4H1p9f$^Zfx|z4;o+yeNru`zS^2f7Bba=O>VP
zwAX~QrPT~rD#w%FCv`N?G$qMBs@rBv-c<Ks96TRMod29~1D5mAxVzhF$22I5&!6vc
zILvw^vH*vP5GZ0l-{Zye7kK^Z6&{WYPNx<B^w0l{<FeqP=;;*e@raZJX%!8gE!T#!
z-q#HwOU>Q&Yyf_QI&kRfR98H!t3A*?I$~r5X6$B0DuiVjJ)y>DN?To-rU|Lo2|Q;B
zsyac{A*y4?CcSmg8LlUsUxc3b^R!s7)|@jW=d*9sADB6UNDxEBF3z|=++jcOP><YI
zmvjv*xk3+`pJ-Okg<RTDtzJn*k5U$bl+@}?EXEZ@28eMwox1KM3}C{t7RYhI-E?P0
zs)Qm5DX)ljx?hbqRAUC!0X;^>jwkHHjEM+&J)!I}hy(U<!aCPg>`C|T|Ku~g$S?5B
zj@rsWpHmg5K#X-+aXOuJC6ZZxpU=966-Oaz$I_EMw{6vUA|vnNFzeCnQfy&f@X4#^
zcz%CCNeLxo9A*N9K|JHd^B4H!^((x6cfx5|@#R-vVObX(mjx*ooR$TY1(Y?mMV(8X
ztF-5GtosOb<@wCnqjNS{RYlg6q*39vUkp{oLJW|=*v$dCFpf*PkmIW8^Ty+ku%tnT
z&YD7%->Y&w-qI$_ZC?3!-rBRxvomLtF|?zxYpIC<R0M%_!uxE$$AJT)0Fw|Vgyv<G
zEwCVSY<*8jE7mom<kI@EesY~=N{T?Zm+QJNSW~j=0)U7(P1x;cNX{rFo6=uUax%jP
zV~2nlgvH7#5)ebcU5MZq!PA7@G@0i?K%{^W?vPT!@whfmpvXZ3>Vm<6@a*|L=KX{h
z&+c?9R4I6OJmH(;1InHO8YUW6S<Ps*F+NjmG|gD~l;jw=Az7}A5~G*Z)j_#`SW3}8
z6xBNebypnfn9M}rsBsxpmIIbkzLXt&yDVSGVVju#p0a#p*5$JN`^$1Zs1XC`+xu(Y
zTRFCXKV7dopBz60_WLTwS+}DF;MFJ3@%;H6zW(i($P3}wVFGi&G~HpEXUw}jaxO@@
z;5WbhEtX|LS`%^+ENfDoI_s9b(GblTw&L3RYY6!okIHP{XNHhVH2$gu5Dzt2$n`Oq
zVg#{z9i=Q-*7b5sHjg6?45Fx8N=wnL0G0Dau0f99x0t=L&AsnB^v73kugByVu*F~!
zAQw<B2vE5$StIHSQ#+_2=d5~i?nX!{qOsqS#eUO*7#2{i%@gTv2+0U>LI@K=*dyVD
zlujBZ1JVp|7>&)OdH(F>Yy9GqPw@F~KgW^_5cK;U2kf35aCm-)-8AWr4M}RCYBQy3
zAhShm0f4!SUJ|6E?tnlbBq~@?z>L<vcO7n9+8ZOhUHDj^?ELZK1}x{JQR7NTNzD*Y
z2b|M-(hie!HEFkFynK0wpa0@h#Hc|+Kl!Iucvu!JtXt6@-n_-Tw<o-P^MEft{{rvc
zJ*bfY=Q?5a)G*D2WV?Fw*yNDTz8?ncjgq4?o3>J-c?gIhV3l-LIjxVV5AVjOVa849
zEBf{Kkic<n-k);aV7AHE*#+H5RKKcEvow$?5lo074nMj|Pj#hdD0M+~?HvqBw%S5W
zZ4hrdqv#0PJYA?KRh0mBkIpHBR-L>^Jzl5<p0Xg4L&k_;tGtZEZihY3*oO&`X>*#%
z6e8vjF~_L|+>acvqX|1ZqWU4NeZ@@x$$FAvE`kE}E-Zv<&g&DYwt}Q-to2EoMueep
z-8Oa0YxCm<`pWmi0?8Rngg6Bp_Itc~b&pRzd4Zq*;tm`p{PuTW;IuBtIbmHB-n@Ij
z@$CbS-@L;Yzk7p+cMAY$AXjWk=t7QxZJfE*yjvOT^#rt_v^4&KLD?KLNh-(D<e22r
zfOY;si<@)y91ME-D6{pg{SMkBzI%LJLHcvr31jDF(4N{mFg<}tCxVd!CXR@t0e))`
zXx&P@$%H<&l<}`k|Jp5cU1n>~w5E)lOZ!`6ZHyA4pZ5P1Knm{m&+u&C;mhMUIx!={
z!~utSMwn*A-3~EmROmvY+t7;UmeIGswA0jJr&mD`L%{vv9&0LC*NinMtT~~8w+3kG
zrP{q_hK6JP8*K~x$#5&OT>I@!(Vr)FkXgNU^mM?*q_mF7VvALw$?~NE>-_KQvWzCn
zm-3~#b6LJ2S*EZtD0fGcF3VMC1JDefT_Jipumw4<PA&q%ni7u31;^uxvSv(ov!1Yg
zm@w^T%)1GneDWFY?(XqVpZ^BO(*nr_Yf5-HE~;OD^A=xz^%lqDdbMBM#&ra0x#>7d
zP0zn}XZvrj0?78yZGS-oW<dyy7#Or%d3J5P!#&-ja(MN*9B;Z78FJB^gUVu;mh<8u
z>vNhsH{((DY8!Lk=L;J5%G!1j-L41S!fZ2&b~^6)N?qMnSUt_E3t_vY=FyT>Y|#&;
z*omTQb}Y6vk%-QV5Y-e!M3_UwA<np)c39Jbg%aivaKAfX-t93@vo0u>V%rP3d3lM+
zYe7F;r<N_Zx(di5SaQ}-B0T7KL>B+5dhB82Y1=Ee8?P=Flbt_m+<@hLG(upY6qIDj
z6A_kG_fbE5{SGJvub$r_1YoxZKKt1V{HwqD3;g+?`~_aUc#XgPhyR3!WdYsov7A<X
z^}Da|n_vGH|MU;P!T<Ze`|t7Y-Gh3-*!wO$CT!rTrzlY$EXR4tIrUnu%cnF95ogLL
zpz)99I3Z63ryBhGjJm=*4)k?ME2Y?JdxU!4@o2m{8z(QnXz8wlZP#F%Vlbx3jEJJg
z61CG^*iIjm>Lwf*yJ?5rw8uP6nB$}qf~-N&fdZmF!X{*M3AUI33<TB_154K?QN6dw
z3cId`S<;X*EQm58uLXrNaw)c*v_M!-H&X4t=&I^auMA)pCcL_Tj(wQ04-@8KK_x}w
zi0tBoBs(1Tdjt;1Ib#Y8q6yF9j9d!7Ii+E%LqSd%DHklIU?Ik8M<1(KnpTJo(B>3Q
zMP2l(p>D8fJf&eagrEx0DiDilcUtt6#;<?-22u)s`ja~x_7i^g*$aH~=_`Epvrq9i
zfAeqg*-w9gfBO74Sk?tGsF&SuzxWdW^!NXSfB4&f!2kFE^Y3vyez9%eXN5D$9#KCD
z4Un$iNz>)@JL$oKFAgrKFS1j;s`3LOI~hpuPP(Vyn2r|s+n&i(PY|Q-lF!s8UR?mK
zAF19x?Y(q9quhF>oTs~(GAunkO|}_kH17RYj1hAr%yGglPS}TtSp<|<$eNJ~A(w!@
zO~*>sCb9&LUsL<QT3$L)mUdks)E&ZD))hJHaU5bt6cvHWFxWik3Iy7R=tDC;efbIi
zA3@;0KYjiR%Rel5SdN&8u#XWh4|kXjcbNBgnCBTFU<m@Df=G3X1Z?Hj`qNU7p@E`H
z&X}eN_xtDg`pp}hPTycjD~{`ej6F*2m&ZQ7PNvPUiRy_(wcoe8dZemum79`fSZpk`
zL#i$@;p_p!`*X)hTYXg@lL_-Qp=5w%)rlOBkNtL8Zd=2L1Z!V2%CdZ`ELVNvB3~}c
zKddav4h5|7*s76jBg>a{F)k`SC0QoYgLO*Qw6hT2ym`R!;e<C|Jb*;-{C<ylj(G86
zhd9recQgK*|K`8Mzxb=a!aw}_*ElXG1SXtLE57>j8~pv>{yqNVfB0McyZ`BLaaz8F
zwvx+IsqMA>4fF%7am^HH&yX(1HD*+oDZ6wSlR*G6VVWXJ7SO|TWwBN?ghlNG$PNnP
z+V-UxLPj|rty(QFn|;ZBJFh<1(({-c+hc1Cs%vswjnTVl#(u{dCo2_=0Vis}sv8XJ
zbH;qtzM>R8c~q=l6qsJSF+|s`#Oi@XQe)M!>ezNF2XRobwaRF^I1@NB4spV>xW^}V
z&k#iLMS6p~X@^f=z5?wccppIka!HWM(wYNMq`eo=gBAd^vWcRniH9&D6~VjZgp~zl
z0&2Yb(oB9Z#|HtRp*{LRPOY{q2Tw_go!=KXU^yQRDTI=Nb*T!I;8YgS13?}#f&hn@
zaF`Ex{gY3zl!AZ%_kV}K{MUbtU;NdtknhxO`0(N-auGbdJK=Y~{R&_F=1csi|M;Kq
zkAL@%`W^t)nYpxqN7F!1MTlV&UqOq_QCBLO5+i-`I4X{S*h+<+Y*c_X7c6-NXFa8F
z4grW09&psyAm+tW6<i7g%GWwqTeX6&MAojjxj2h%aVgM*59TW97QSn%y)^vI8tAbA
zTr3`mrK^ywezp~k+WniSoapIA92oNy5P}vO5hFwfmjDt*ZUH~pR$(l5MRlPLdb}?&
zbUOeu)B>f3wqowmJWYTI)~o?&)sR-8L$roZCiBeF9|je+IukqEnQ<4@c=abg{Usz7
z$WjpWm^oX`s|Gv9v`m2LN;76v@N1CBDG*-V-Q!Px_A5N!KSN1^cgGXn93LQ*!L2^j
zBTK8OD=KxXCD%G_Rr9n4*>%-b_+?RJp%Ad970emSv48=rSy$%|`#Yp8`25SSu>0f{
ze)+He1@^D@$m@#T-5x0yeDm&vKmGL=`1~*b8UOtG8~o-sUktI$Xb9A~m1EczU@|#o
z8I4b-f7<G2Tj8rGGZC4&t51ScDacS|js>8i^&dmP#5I0Pi!i5kNrWT<k)p<cW=ym6
z%b*a8*@wlHWjmP*){e&;>Ea(%f;ES1G3bI&ljCGkQGdFpCtJF#R}?+Wq2!EW^~V&z
zF(QNsvNB4sZ4IWx5&=A^VK?O!Ij?$hQJuV)zN}(pv2jFfrI)(GSqNMhSOu~aAh4#Y
zsFF=BE(hVo{d0Ww`X~6}xBrZ{1<0}B`Tae9{>z`EWYY5*C_sp{3n9`2Cs2?|odEQ{
zkSuUCTM*fli+XF6j71Vw$sj3UQLj>ARh|jfr%*L;ph>jWom^Jd6X9lLnfescM4}s`
zn=JS12^QShZm!1Sx_vwY$Yc{Udsv_@%LPLSfcoxlk!9GV(pGg^f6!&w+L^Tf^!#o9
z(H;m0ES@AcStilLDYDD*?<vcIAVp=F^w0qH2mz|DL9JS<@uHm9+LPdPO2{Hu6EskM
zN{A;tVL2@Yfe6P`FcpA81WysXoACOVpW&DP>dz6M?~v9Nhr0uE7QB5}@T))jGyLga
z{3X8n`hSn#e*UE`4$DT4E6vt3%CU5L=stTUqtzJE3{2H1qc<?y0XQ~(7pe_XJ?_dR
zX7n1yWP3(nLdhyKJqWwf^kXybR8_g9NQTUa8l}7THt5te%n2-a0ewuadf#Xm7GUVZ
zhEC0nhYWV`LDhfByrlF;ZBm=uv<FHvsA>0b=9yYFAEgMU5C975t(ephU4&7x%4rLe
z(ldz~r%M1mF>spkVm^RM0aL(>`+NN9FMf&R>4b;l2|)rtY@uSbuP+5i$qeb{7O8b+
zvdvl<Ic20ZV@=s~el1I`3*6f6=x4N@*^Fh=;|MzcIpGE@=c7@IAZK7*ZIzV@)+Hks
zU{1h9z-i8q7mOFLUgMZo{KtR#2k_mDpZ@tTkoFlVCA|LRGXRW~fj3{h!#BTsga6C_
z?*FWN^V{318LF=h)j?Dzvv!IKZ1N|n@YYq`Zp=cP;8u6l+LNm5uAXYNmVzZGOj!*w
zGY3%U0+rG4tH;a|!;WRCDINw|>M5@VEFB3gmrbLW?cL+Vz?A#C3QhJQLAUVLiLaje
zSC!>%Pw6Q$gOP3XV`jvOG;pZsc8jWnqfmQ)-G)W13bu$I8>vA!p~i&Jm0SS<Qjo3e
zBSjl{7j;BqRoa=v6p_*mSHfViG+3GgF(wYUpLTe8_Z+|Y^yheZ^MHqM-a(cVP{dAM
z(sXhmjoT%JH78X`nhU&OV#4#o0l)n07q~YA*28JRyVDVZ4B}!&6hbCjJu5(4C4zZp
z4GNv=AhH>JZ=H3?dMeo}ZLR&`sPQN$15+TR3<v^ty8}+S;Onoy#_shC{PeH>9MARy
zNh=PoU!o9W$%Jn{{{~<G<}3WS|JVOVZN^uBfUO*3_&1H$SsRkgWRkk#Njk}@%Abf$
zdLhVlB3g7Z7`Y&t(W}^rmtqG|sI7FIU{X`WMNFRRq!--~ZOf(let~*RR*x-PzU>DQ
zoM~e6|7Y*to+L?@G(YU8YUb_{ky*FyK9}8@9n1m(EI<k=LJ{6x_y>^xGX-xHK9CR!
zRsezv3>I@ax4u^8B_iDIqI^&@ch9QoK7IB~&z_o-){<RSnUN9h=B8$<pMLtOJMRHq
zZMg4s)wZwoH`je^+Z<|fZ4h@N#n*Zq&^RE)gTtQD?14ZCf=OKiUT?kKL337s&Tmhe
z2(9lMoz+351-(!ggURQi2g@pP&8EIM-SYXXH@qHCgjBf1%<bvK&wu%IE_YWhZ!ahT
z*op0<RCCelU|S&*^lq=Nrvi{lAyqFqTC9+$XbT+r+!?I<)00)$Q5OTV`s5LP8dmJ%
zPsx@=Tve&=UQpY#%kEGX+rTSn6dFGxH&CjGJOQIA4t7M<21ZL`C4fo&QMT*=mIqHA
z?0IU-9<+Dn=b9bc+4B1(?!X2{_hTIg2g<W8|IxPW4q<1o_cNe@2Ty=!SvTjl%KiW=
z`j0>6xll2(3E3*CR<u!Rw7?KxPL)G35Osia8hG`~&-wYk{!7L~q~^@4&)#67qylgM
z)m#4RufFB~`TzKT1OJY`)c3%0KWA%S%e-&LPtPuC^DOP}DlQqq%Bb`=RM{nPrSAg<
z#lhtmTS=*~b6BnEH8tSq34qo2YJ0NA*Nv$CeYIWsp=#^$*!}Pci`z}TH!5gAbekY+
zbJH;IC&0bGQH9o=NX5GkKIxKJT|?GZ&8RI(Sov@}ylLGZ`dznY(2polFX_v3B4>y3
z4aHiOKxeA8n`A8klnuZd;>hiIL=A><;PvZQ{L5edlJCC$4$j{ZcMTq}&g66%vRc5K
zaQlAJ_e1yTD>-FSb^xohaC@gvdQypQ!n@dMozqXrlYSom*y9<%dLBR7cz1uHRwHF+
zI%F|Y6Bw-(vcPwbnf%oq`rrP48BYhs+aq%+)G=^+^@dUn8E+94Myb4rf$1=E`*P%P
zfXC7f&f4^}y8aBTS&G)(nOgn)5Z2_*=CoRI$9i98&^F@E@ac~C5Ly=)g}D}N$q1f3
zl;|H?N*BpxC+M(0P#ddjbrAL!jn~BcH`pIh`>}yf_TG<$9xC+z0=jk+EM`qu+&_O6
z!j<MIi#YQsK&gppBf}KE6cF_eR9YeDmgZ~@NeqY-XS`Ma)ZrkMP?^t0A~ToDQYtxD
zs<x8A;+Ypzsim-_OfG)dR&(&J)IzSxK<m2HI@1U+#mE#Rig6r9{`J55YhIjg&<dd#
zAy;ZP)Lg+Mh&VW=t;|&_rDme}0qz}~Dj^7`!;xRS`8nZmz$jdn3lG;vYRi}Rl~K`6
znoC;Hg#&G?CsW_|W<HH^xn>IHpszXjyLQH@kb&<W7v}Vq>-m4+_2-}Q*)M;FNu<WW
z%P)R`RHIB06CsKa!^mkm^5VsTDZ;hAw`ba#xbu5<+`m`r_W9bTRaIN3t7gX?7}{KC
z9Io2fg;r<|t1~@IEd-4WF%pA2mZ>-sumjLe0Py$7IX7B6Ye3zfZx0|viU-T~UONS2
z6{My?@z$=Tk8c0%oe&y(quH_c;JJDBmiI70@egUG0x2hQ9uRF#e)Te-T(Wm7a51;W
zT2qWdt{K(JI0lZlM>JI)&sQ%;iqSXAYAY+%$R31N&nl;sT|cM7N^jabq1}OQwLQJ8
z`Q`DJLmU`>c49hA{%qz1*Mu>8$x=!!AU9`Yq=LqRhJuKf3D?HjRP&%em&~#h=3Ge_
zS&gCA72xdU&$e4H;LlEat-5W&#;52u|H;^Lo4WN(wk_CGRZlcfJltx5q%}=N6)z{F
zLmXx+1H)}OQ)(vYKy0>5HRjTQjUQ{vZFF#>9dj%5a~(gzmbdrnfxLLvWzlC_{>QZC
z9?`ELM}H<Cvt{0Uke|*-#X-YtZp(RhKVqI_efyZWrU%aFD}V8qzvQo9ze0x*8AonE
z|H9c#$0LXXgP-ZZX*%)h<%wy4YxP<461XPY(C4c~jz4I}^1dB+`|qW(tO=hbx&76)
z1~qCED$l4yLQoDdk|?Cq{at_7?q0bam-3;6M`vr(IDJ*>YSSH%TVE?5+VKZ^ebzxY
z2cR@;9n@1T+ck-wCZ1>l!C8W(7*zvnW2xwSz(^oSM9>yZ(7Fz?f`Ezx@QF;qn+OpE
zD@v^!u;{@?E3=b8&Xt_YS}N@J8<5(0j^5?n+H>+}TB)^AtuPHPzJ9nlahwh;*O|Ix
zG}jh4xn6BGqZA>RLQ0tsBjD@DsFb`kys-KAwUSceIxpl}+D_N^`o6>avzF!r$_^xR
z=G)T^`O`a<pT~dbcm}YZ$4@rqC1XG_ai`fU%c$LPHn?U(&NE+r`7O7vU+^*p?;;|?
zFdYa65+-*ZLt(%V+vyM)22Xcb%b*Aiz-mJp=Tw@<x#m;W1Fbow?_c=)-{=S83Q({*
zo7p^@88xmMQN+s)>HZwP{o49=zaPlxpW%7~{R6k(+5h_YvuC%=mq_m(`1C$E(ZlZj
zwEJ%~533ck+PXD){d^T^jIj;C5!swqk^TD!rb;p+nXy!7095PqutiUudR?j?gk5mb
zOxD+1cPh59f*1l*j7T*O!^mf^UNa3NmK`mXT9LXr0Pf#9D52`}TCMnj-pexa5|bfD
zPKOgUIPjNCBIV@IcbhdfPpCa7*=Xh8=PZ9J2W={~l@PFI$Zh8=nHiU*Qu0i?T+r#j
zn<j^X2;=bvG2_}nM-{XPLsbrkfguXlY`f(rn}M>~aWe-0Oq$c&m+j6emh~0#fti`@
z9>s1CD7ARG@)W(qL7qBLt6#f&XkE`ip4jHoz<1qWY_rq%pse<5A9tv?_hNPYpK5yT
z&;WLOi`4CH!~@)7`*In-n6>=awio;D7?lEQXb{2>2@SR_;AvGFTd{p)t##)NTLXfv
z<38*K$NF>W0DA`wtXA&|8kED$3v^I-ymNtE*8PZ5uvR*<20l&1TGd)0)yj_B{Z*@t
z72Jen8|!teAd&Z`-=C{*XHPfT?|Dpr^5aj{mg(%^Zp%*~!uM^tefI-fwzk$i338l<
z71;g{wPjg9QF^@mo-Kc59sGbTODm1(Al6gC>1SL1qior1hZ+8OTYmC<KK|`Lxlx**
z(#7p-qe*Q(rZccgE-cqGuRnY3_96lyFdk37j^p4gJ{1l(Fq(2W?(Ddiv55<+ucIGq
zf1n-9dlTQ2Nu@O(Sl5!e4ph#Saq#zwL8#TcURz(=yk{@)q!~kB)4eLe|1VoGuzTR&
zYsWk>%iU;xbpzF(`1$Ysg4qVHT6=bD93g<VQrfDUUjk3gdZCvNHs4&@zTKXM`^Y<g
zvzHOm%8u(E4036azHjbkY<*T{O+%{`7oNB1`@IQyNmwpx($?p_A{A<twJdVoZ~Nz1
zUGLqQVn%6=1ZwT&)ATK^t?8%x@Tw!aUGhhs%=7qj7|#IK^Z3a|4#VmUIW>N(qOUEC
zyXNX5q01Fx<ak=h^Fm395F#o8^oClEd07YuVXVA*eZv=DyyWi72kKH><0`;7c}7H3
zd*0Hso0v0wDwHfd<cg)%X-u}mq@NdlK)SwFDGA#hu-cpjQDL4JF6qi?JP{-irva0~
zH7#B~CIOXzNB~mHYSf)bqnDoP_iVzw5`Xp2cRTY+vj08Nl;78Ob8BEB{XlJX>z2Jd
zPn6J{L2Dm6Oh=>&^Xy<9Q^ay{oeT|_3Sq7Bhg=Iw&b+-W+|P;M+}$xRh4Up*lc7<(
zRzDj?!4wKJnL?_I14;-YBcg^%1XHfdOsN$qekUgI@_55%x3^4j;P!Cj7q8xs=7n#5
z_Z2lI>XMP#$`Gg^1*~O;)U}aHO;96Jlv0|bE)i8>93fdD*Fs8(IV~Uef2F%?PC2OE
zr5D-sA=*pzW553=I+mv5RZ3lL%6i;no?ksL##M|vD%0bY&+hKYDIo%neI{fC3)B*c
zF)-a2uiu>b{PP<gzd4hV>zBmuv16%qXU9S<!Z~^QM0b|JudA~wo{YwTtIpsjLhAq|
z%sKIJJ)^Brfeyj7Y39X)zSfMuZF%oBvjM8om=*nJtfYxGv+MrnKEU;Bdg4NUZ*V0m
zxc1%;JP2vmQY;g+u_`QM1=q|e6|tuC6$ggVnTvBut+jbT6k>4iSgTM(scKwn;amz|
zUoPY-5N^nLcUV)OMYPcM`l=a)O6m3{4K`^JW-7iHO7(N07UO!oaJiltr;(g9-+uKi
zVNedo19qOVykIqf1xiU!D<wfm#n~IVA~_Rd3+{7g=PW=#85&!~_0Mvx#5J2vbBy<a
zZ;viKz?TL}>kd}=l<XLuZTSze<yAhQee^%7Ez7elKil$u_(=FQmuh4)2aMKx@ig6K
zHdHF2#@BBj`0YRZn#<!OP?6z)QxOLTvWvJK-zu-)+<a)q3eyw_A)p#Paf|)Ic8u&w
zcBPAqdoobP+3_(1Gz3z~JYFw^7zr8)aUfKvHM^#4h)A<#5k+Jtp8k>1t+vaK^k=`0
zJ@E4pJ6_*fx_ws6blWF(pA*D|<Cn$^t8|-pATGwHBOAmShrxBrqGil<ttipxv^T5S
z|1>zBju1!QUNdvb%-4Hns}Ke(=LS9-n2@ux8&fXiTA%Equ5IVwui!RbJrG#I(8tF|
z-rn62Lm-#T|MEZoEjPy_fAtstf^<zSo^pE~#MTl>Fot0uG;n;ah%{&BjbUV|kg~C)
zh4b~ooD;bgCy>e(`&3*vYfHa(m9D22`wUwB`Hg1)>v{ZSqgV%7JS(z4Tl!y}q6|wm
zZ;8D42D~O$nzQH_{NS`(DNNJE=`^AG$o5iU2#Usl1lNmk+Ic^jtku>FT53c2eWPFf
zo9*4!YhC4PT5uW^q6k6#2($)72~BV^?Cx`iF7*v~?NZXSL;KHpGK2f<)qdhW+^YLj
zb&B=}2+LFC>x0otDPW^a_iGl!h3j)6U9P}@Q8<`ujP&!nQ;)40OU_)E#KYyx{dM8)
za$!k@c`m4_JLR=yjHF$Gxo%dKz4NmHZw4y`de3YU!y5t&BZpz6EQvfX<Yj3fTiM;V
zAFliLW&Js|uQgNE%0g7d1=Culu#}30r<3@;j@U;yHgp>JHn-pJZ2c*XUTIfs^YQwX
z_xmo@T*E?LpJbV5tW*@Ch}&<V=`RpLBN_*$!^G)yK;K>=H9GqD*s+%)bvqUcg0v2&
zZpQY+G(Iqc_I%g8uyh?9x5?7MW$}^#{Qy;d`b;|zz1j`i&hh?}{`mjj4{+c6ypMhS
zN0ya*<eS#P=iQZ<87UW{xz5*6E4hnNx4>MRlWpd@LDh^@3YWC-I42&LnPN(0BG<l$
zZa3oH+qQjwV`_GA+n%f+>GvtZJkOjjXQmclUe6a|_L5F1b$#}|+(ArMk?9o(Fh6V5
z>YC|h-PgL)?m<(<Hmwr+IrIJ|+^w`dw+3wN*&d&g<NR#P>pMRA*T-ymd)0oY{^M-<
z2@v;Ddw#a%XIuXLqclnNvaZ?g8g9N)7evQkNtw&#Ldj+GeVPsXWClbFAw~{AV8<bZ
z*2xsq>K|;!+l4+I+CbZ29h~reDCOcZB4SM|O`pEsN@22fJ=l6lPtWB2Ci1?0?=S3!
z?D(VR`hAP$1KL)1Sbgn<5FFI6l_~*?-P@M6K}7$Iok^^cB*m<0@p^ac-f6Mhi3wf;
zD0m0Y#$s$%N7t2IKVx@;9mLN$x304Tf-qlZq9{3KO75&KH#~E;LSGIa0DHEdEOD;d
zQwCmL%eZC#<leW<_cHctvSDY(&*RTyJOfzI<0l)JlJ>sNcCY~b2VJ;t0UnpcyN3tw
zo-ss9$yCdfY9P@|3K>v~ynOSTFMjqJ)35KCEh9iw<#?FT5YXUyN>*(xA*4Vq#)1(_
zh2&1_la0F{?0o~qg`HJ}F}PEB(7<$@IK`2fNUDL5ji9c<bR}~y7tdz2n*4o7v~8o_
zHfOa)z3!WS{h^(>J7`aj*Y_#a>nu0NZio4-Z{r#o)=DvLeOJ$hDN!O~%K0*Lc{ua!
z@4n_VMUGS8ufO;gyf~bQ=2{+82&7uM0`D#t-o1O^Z-4U@-<_{~GcV+9l-amB-tg-7
zb(@>kP01=bFJ3;VttV5dPFa4U1E8Hlu<BVXj~C7re*W?elLUVKzx*Ao*LOR#n~sPn
z)m*!(XY4A{X!6p!zj({v{z+I}#ePdM2jP}XJ*-kA>zwX=nbkdCgMjy*-Y0a>&Q|EL
zz!URre{n{Kp#dbqoD=WvAFy1QL`k77F~9}x4=#Wz@cPXge*W{%`P;7^SeD#0<^spV
z_<<d-NOJMa`;uJWY!$ZaL%Ofpx+nCq(3V{c)g7JCn=a&vWus;zr;M7nx^{u#E3mO8
zSM83b-QdB4(*3|SP!+sfqww^?J{F+zF+1MRudI4?w)>s^yDGbT?7!EumQDDl6!`A$
zk#kr$=)lb(avUQ^Mq&&MF}m<|P;ymDg~#Q}-Q$H{eRIdV%gmSGy#t9HkC5{MF|7Ma
zF{4U#u7((<tz@7UOq_dDYI5zQ5GV+hLQoHizWw$)^fL3G{_-yn;O*DnF-{}LI1#e7
z1`D(-uLokKb#@(w;CgmIEty)&+N_~TAh`}>&CK~qs@}ousLl*pC$+Y~y&qb55WRIC
z+~yoUE!GA++wvb`%j;#h@C++;WButLZp$F86Yq9@?*Smc%V%4Dw&folOLf1ptR;T?
zpEf(L{@heyP78PM9;i7nh`3NVkfnkQXgCr~2@0=Yzy8pUV{qSEy((cS0<}{053}P9
zl<+?T4QP#CIw!@5;@d-73g>I$bVx85F+@r<?`zZmOrZJHwzFeU4R}ZF9e5?}&nI^L
z@$oS`UTwJp=emiVci?DEyGnzis%%YL=-`t7d^#K%$B}tS#7PL#k!T~MMV4jpIH75J
z26g6`Rbwfc%aV9^KJz#)ygNTqs!-+VjLBM8UENd?#WJ}ja?KP9m5kxBRjmbruBDJ>
z9+Rm$Sh=Ky$IAo9o7Vu8rBGs_EQOphOU~Z-CQg)Vf&#TR1z+{D!cvM)c?&hfpm$sC
zYt;nrNVpW_rd6z%xszSDd4Iy{`=0TAc;fT;=Qo}KtmpBQjrE*(KiXo;xO8nI_YeKh
zy<E=R-@QW*Bk0)9Is<W(o?01Nrl=nvDo&>#2JeKkilO$*O>^3#c-AT8;<}~n0^i>h
zyWW$T5miwM91bJLY2=r0-T<D}yqq7oAf8pMfCkj8X|^}2dOP%2ig!DBSK;Y~Fj>VJ
z?y!D#zd8rh*btN5Fg}v(+5K#FgI5I6wnO;IO>cj0T01U`dH?Xp<^Ih5{UgGW7!Ty8
z#W6@P#ly3`RhU!d;kq!FN;X4H{MN+5qH5N%hwa(~tm%p6nj34XA{q%Xv;at`nz2qF
zXxE#UL=s_)k#QI~-#sD*QG7CsnIFzwa8rbmU6j^}t&>+9fVDBvw9Qvt*Mu*>))h|r
ze6s71A{#ojyT(s>e9Yfl*IO@}Y|J<<V7XG}#B#YXUml(MFJUeBaXT)Qni&SSV^P^M
zcPfm7m(6%)%`@z(I|C45Knba=LD$o%mPSMG66i35h(_V|_UI0Ih|M!q2K6tkb?bc8
zo^kKi1)FZo3aa9N59*!8FkIHHWv07}M*q{;S)IS<26nb!-8TE{_E9?+x@OgF&9tuX
z_W>fYZ4@G|4fOWzo-srYVI<{&bQl<pkr?87ADxx3l*+^P%G>(~9xi9j*TiL>5h=vD
zxbxc>m=5yR7B^O^mCa$^cZFnQJ-ok>Eh%x)nQBH52awgYi7s!yzv_Ho>vNH{SAxB_
zf3+alA#v}(R*H9VtKIh1S+pvvZrvL6{D8aszIg1XFrIDs$J??spsttm{bSg&gF|)G
zTI_6zzLw?Lmj9`2`GbFbLblvzy!(VZ6JGxz#a4EDXyuYp4N7ilKv~OkE0tn}Qa*0S
zD&oF%vqj9GQTRjc7}@+{+>V37i<^^!m~&!0OrFpc2!i`+;1V6GUY>{hVU9^j>r%Qg
z_&U(GmH1k_CR_j6=aJRy`WpGrj@Px(SX@S92lgMVw#?>R??A7%dUrPOzH%;=hwDtu
z85MXrI#}4({aWE*g<OnlPCQ&MynDRz_VJO&>%!gR14v{%OuWaDbatYN+i`1!;B3Wh
z&slw9dy;K~_WNCvSY@-}?t`oB5(|}<05%Xs!hng(i4a*kBGUKM_7FnQwcId>u4}_{
z{MKYmK3xAFw~J>M)}PaO2C$yTPd47)M+DYRHK98eozjdum$?dWzx$T2e)}sv|BHV?
zJVbvV1(S?frsl*n4&2_Hpy_Hvfe@ABG&XQ#6BrMhe@d;2=)jOOVOcf}@%OIOvz>l`
z4a3M3Bd=e-;`3K8`9J)x|4Z)g9{JU8zU2IP&!Y*XjAtDbsTEHsDJGGqk*I+zdxuXN
z9lMtydIpQofuN@#*6Nge3=xr!Z|LrT+N!s;2EdJNr7cU;9Z=mLxYElH+!_7)>u<R{
zoO%1*TS8E#!_iw==gjeNY=EC@F%%0tTnld>X6E8rP@_bu8g+2sE?Irj4G}!sT1sZg
zGv3#)B5@>+kzt6Qh0?TMzh|XkzRp}y<}gki4+rj&XAP$@QX0_cnjn!FoeglkB+}A4
zAgVt%%MQp@niC||9d@g2=A#R_S0hLOF|uW-5NfA>x2LpYUVY5L|CG+y)pzUN#ClM+
zUoc%3Wd<7xd0e>M-SP0<x4e1rnlMb{p*zckTodG(X&ShFapT~id7q%rn3Ua)yZy(o
zE>5&68V2H$5ncQecBbKud<(Lg$9bG44&%VT`j@}t=6K-qS1-7`d*C-;eZw&hq!7HT
zPA!aagdhZ!wef`G;$AvjThOa23`5h}p_l7Gr@2=efy16MJy+CzjRx9Wx6fHSh&4Tl
zYSqhtR)BSL0zK$#!5~e~Pq%JD-N{e6@T=c_>8zMx;^obOm!|`_;|+)L;94Wic`Rm}
zQ|6m@_x$FouleqpIH%0x`ApD(Azpkh3Jrh<-<z#c#I-DzGE*cw*dfwbkGj70eZ7+F
zeC4sY!yMwk7$yc8kkU4AHOvB-(}0ZHrq1ox1BBjTN}Fz;VWkCVlD#`pN#yKWJH53<
zZOBe{p#AfNH5y<SRKI5%AK8)rWXH2D{{gm)P^#Yt4b1wZZP|?60>z*Ov)EqS&Sai#
z`Jd93*Cc>_TDEN9fOD6tsLr(Npow5;foL!lm`oWZqNN~dK|{n;-LJ;A)}$_wu8dRs
zz>Za$Z{171+BN)#*l}-=A_9kD;5d%_^<Vx4L6r0TJ=u(<B}T^}lw~26!r}M=3--j0
z!S$v>2rfO+*5)t_UZU3~YXzisPx!~$F-9wE+^-I|W2v&1Ew(pZ^{bv}<q0j;s4Li2
zs&QQkkKeuHVF+BVSM-f?d@=d&sW1(`Ka^tRl(@fM`ReW+U%$QQ*KZ$qoEPrz9~p-O
zKc8+|pJkzn%cay>TK$1**sU;@Y(8`h?pvp2v*T3|zgBWBD?=*~k*cm^i!!JnzCyrU
zek14X`hzjF(#6cEuID9f!R{qxVjMXfPK;qBN@!hY9SlRX_Z@F6u8oY)ed~4a>A)bL
zjvYUbKW;n&SkL31WVjo$|Cet;K_My}hJcDOUl-={h539&EEA<tYH)!_D~M$vq0s@W
zL4`rtWjtKOQ_PY6YJEFvKi4hu+QFm`0W8+ns~E>|;B+|hi!Z+5mp}W0|JA?wFZl9z
zUvYiBa9-X}1@7OyBc}}NZq&^oGH19fr8%aub|kE&9TK4hqrF?A&<>!@*AkIkCP?~~
zeH0w@zZfo*xjDO{LWp4tQrh$jG!RQM*}OX!hjrGINz<69FpeXifBuHs+gqkMQF3Kj
z5@|?OQF5s))mVC|T`t@lP8^RyAX18PzAO|`a=VT$?q4gEw0Jj?R*<lk;57hFJFv*p
zK+>uU68C!&M=HGyGd=U>KvpT0(wTgHe$5)q+1G}Z;*{pnoLm1uFGuX4L2X$fYnk4L
zR;`EX?wN>4bNU?=bwSTh=aB7M22!Dg7;6CN|F%^Z?j{DJ8RzqrdA^eJ3=~v+&Hzd&
zNJ&&%5H(^@7dBP@3=tY*(V2T@xqUW#Xvc0o#Xj!2C=SG`8mH;Ni<=XF{g;2qo0l*7
z%U}GA-+uX}KYR18Sl9EFoD+2%5scGu;JQ@Kb3shJ<$Vhbt!t%4Tj!}e=KBopmhElp
zX`d#iy>{<JKN;5V+JnXrK%UzF{&_S>U3!T_kS$9tsvJ%?L{)cM<A_OUC1D;gR($=>
zOXhq{<SLjBjDuhj`Rql&aCV1R>9)PBUmta4MpUb<JlOXb?bHfZw|)cF&77&)bXHVb
zNr?%8up;_#7#OC>_x9>`i$-J^D8>Ih4WLoix$jqK;#uIJ13@AI^=G*Eh~Jy~9w>g0
z0_M|nl%H++54Yu>$oj{&Wo@>+x<l(TdbZ_%3R`~v@qFyp{!~X_mm2`Bn_o*oF@`9N
zD%_3(YR3J;nPr)&B@u$7t8T~QjCadGMuU969gDM@tTe&z?gRZuJNErX8G@I$zP^3I
zi`yIixBu0D#SjADef14@=SRMJ`<C<hk@<R|B=3aBp%PR$9o&wWxgf~OnCy0pY^8&1
zM`5})x%J#q^9%bN!uQ#+?EZ9ryy>g8JCZ%R{C%TZYaY?7F+4F`s&TzsSn_~L;N|hk
zX}mIuC%yeztu?XA+&x}-xXwH-3v)`Ol<CN2DIHwgt#9iZQ_Y6y-M;$_`S01visXqM
zm&UHOmc%`d&Ume<F<XN-bIR8R?M<?}j#bds@1dgV&#7;AA$Y=DinT6|fnIi5Yi%I*
z({;1riF@Dqy}fgNc0ceap2t6*@eE)+kAITkU<s?kCT*LI3PTW%lTxdZmc(+svRp2x
z6+)ws`bJ)`T7ZNogcultv#($Sc(j2&J;}U764t{^zE|)-*27F%cfx5LxH%m7?9FR_
z`SYLgpZ+KR2?NS+{{GiIE-xt<%jM#k7!f2W$0@Fq><)r-Lc7%#-qyN0R$Jg}y}{BB
zSlh}E`Z@cwbYZ={egr*r8k&P6s!ulejbPs$(7Fm}Y7^^RDaU3E!^n%{k>hY+h!dq$
zmXt^(Q$-*ZmMZ?03n^DlFK&qA$T1vAx$sucTubJhy^l_BhizhBr@sPNF5Fq0*rd1E
zefpgHsMhDw)^%15*Fy-3M5VSB=iRqzV{x%_#tLS^vsdPyUu$*xvZrgk6J>AbUHcjm
zR(+J_AbosQz5ld6okMn=ZlRTB*cRa7-LRs<AqX$0fl><B%fgZta#{Q}VjIX;GFCHI
z7qr=N&^7VCu@>5AhGrc0Lw4LVhqC>%|J<1t!f~3oJ>KvyfALFx{`njJv%maHra17`
z*I#p)XO<-~KU^uTlNXBHabpI;4*Xv2n5V|NI>sH$+}m*jUe_|5AD-{`Sq(i%cpt#(
zAL_a?vThV!g5YHcwBMb^BT<wniiU^~95C-f$YzwBSyJXYXL1E3GDH``9`!(}Fc+w$
zdM7Gm{XD_-RZ1<*cK!NR#)dWLe|L{73u4`;I@sIp%M=%Jh7clg7_p^wE-Mugh#IKH
zwQxp_NbJAk?`<L;*r?xkkg#fgtXeAXJ%=Z@y!+vvwE>@w9mBIN{~@;A-IYcT`p2>5
z{!+!u-1I##TApqBpVF4spZ^F9;!kSq914LI#OX;=0j3}vLg2+TVieAo3n{g7(FS3#
z2e6uN{eyNa9pq4owZvgFlpkfs;!8h<z%)c&+}!ZltJnPJ|LMPG8V0^R-tfC`zF|&@
zcXtOK<G}fPUH!YDaF_;iNtBX2*}4MJveUfnWqj?LTk>!PLD=ov)_~a`c=&su`pJI3
zo0soB2igxpP-2W|aKI^LXAxfJnNkgn1CPtXWtoX%SU;_ix>$1I;XHGm7cOZbrQB>r
zL3GvD_3)yxBU`mX-HW)dK|}+vYHk1CFD$p`V$J@|D>e?}fuQ0|Jv1_C#7ZSEiNX5s
zG@ZFN@1nAkQE79bLe;(vig%K1(a2N%zxSq6wmsN*?>Rob{O9p!FrERd=ke1G@xx38
zC73Z(Lqb3_FsTsLIL5#jgwqf>&xMCeCf#4Se0vA;BjIo$RId*wB+lnM9`3)PEDKQ-
zi2@O7a!+p<2G`^$72DVF<09|1IZ*2*N_UJJaH9ff12=>iV2Y6`2I@8O@a;SP-GBRU
zdANIEH2D1GOAa?jLajVJKJeZ7Os$1sj9eFU;qp|lRw}6~gt(z+wK0WiKkV^My5MqH
zU5xR2ruV(T>gj9O4PFF-?n>I82FKbcBNYd9VvO9JZkWb_yqqbu5RW6b#{-|gc!3nJ
zgHOvNw#*#I(Se{;xGcuqr7~YVNchF;FF2iUSc-AI&OF{<DTuc{*FcCa{#vY(EmJI0
zdys_dmxa(;3ETTllxLT2U)=KMa6p2(NUJJF;d-4Jk4L6qA~(=5r%Gs9WlV{!n~Mv#
z*VVZ;YYePfu(Gv#&NY$h;-Z=FotB+u-UU0m{=&ytexFj%+t-vtK`X@CT3u^kjBdxL
zFmMc!&kjf4JzV(ax~zfcQqNF~5N;3x5|#P>p8LDsk<t|n5Mv+)CFQ2Kpz8GZTAOaM
ztN`Wr*|E=llb7xtix7q(FpcgoM-}eAeTyuK@qFR_;enUKfxr0TXS}?9f#$;f;{zsc
z7dMBIIeEKjE7cndwD%8TXUAsV$;&{^OD7Ds<E{by!RWz;v^66hFsMFt;`*LoDvh<#
z?bw6duFlh2!(+lUOdO6!P9ZReU{a6(uWoLcqL*#BHi{<=9&1Ib$K%U&=Kg%;{_a9r
zDk{n^KmUw43|t-)bIQCs&s<W%uJc;X(RJ-o$=(jTl`6p28IG}KTI<t*p_-vl2xIVj
zFBNj7*39+c!Z-yE;{i*R#KIsxF)lc6SO}ga7C~Yl#6V<(QYbNOz`Uqu!o``gorxLx
z1Az_Ls2>VGe`p(@)_At%A8*TM&idN-XXv*0!T6E3+}{V0#&B+De79wLw&j0nTkhZe
zq+;Jzb)Z3rC?laYrxOkl28Ek(;4lolIi9$CJoD|lD^wxIKrL6SW|veDMb%gyA9%R`
zik#*T?HHOJ*UgR*s9xszW9`_<N|kAh+@4NM!$3gz=67E+hQRsTw`j_Ie*1#_`4>#X
zz^{M(TP~LiDa~jM+)N|O;+nIyG<H^hf8Ew%*Kg{DgM!;c-X$MC7$33Yy&i1WW$UGe
z(#~XSBI^}gbE>*jM`w)=I&k~qmYbVfST1Pw>y8L7UcF|B1E=YL89ZJdIfQ|r!jcjO
zEV=S{UHI<ZJyzg$JaPE^1=WN(C2G5#ByGRJ_l{DtYt`0FvEpTq0aOPEe?vqzC9idu
zd2@vhin&mG<aE4YQ03+E*!<?EvnF80t4m7BSXOkI(WvO)Y|O4HD*`b_?>;I8k$@0;
zR6s4YKZ?~R*Jnb*-E8|F@_iA~^Z4gJo&l`q@ska4ENFF>m6j<|<q*{m*+XPh;TV*X
zjSE6eg>+pg%S_EKH0p;&A}tpl@9!x&yK~Yswah%TpsFj~wOs|>Md?|^Cz~I(V=h70
zb)aQ-hTvIaH7HBw`f%o}U;oajyQMOPhz!w1D6R51&*X-fxyA(_Ff>GE3r=m*r`qqr
zTXy01ZO!j>BYXl*$rES$!}r@~E1Sco$`B*NFft4SX@0=OL8M_AI35mI0?S68ja-~%
zW>u-BvZTuOnyJ+YA#$7!+)O7dB~mH`XdM@wIxI?47Vc#a)(#~9b8C?ct)0u0yEpSv
zgz0bqkw$Cx`OP^q0#OI5xg$_pi}+P9p=)Dw7@)tsvFf8)Yq8$5ShaNk+D-HJG>q*U
zrLVE={Xf~!z-R?Rb6BG)$LKb4J4B9g;KgBNo|EgiRHKwi$}^%u95O1dosgC*kN5Y~
zlG{4iQSsU4ZMBn@d0QDHPwbfW-`n@LrU^j<V|0g5%($E{gi`o!nVEATqMW7!8V7Dq
zM@mjyS^zbK(5UZw;I~=K6WiKKOnXTM_F<M#V9jXzeeL$CzHHvR_uUS*-tbQQ_{j%|
zcxhCO1Jg7yikBXxk`aM%7&r_gH;03Z?iKG6wA4x-GZ-vsVVM*2oT=5B61S%t#&JT*
z71hLL@z&kiZLZ&IWd(cP7_4QuYtys{JDY!cucEqw`-Y2li#ZEKsxvLDK&5zu2HR)1
zz(Rrtb)xDeYpVY5>VWdf3~9mD1}=Q0Eb}8F?UUVeGxJYtJlpb*vgIds#Xp`c?{4%1
zTefFg{*PkIKlHsn$<b~{)kFWmWCe&0Q5l19d+6(9<Z`(>n=V&!Du`6jg2arPgD<OZ
z{XKTvNB4d9knE4PWAU>wXzP@TV6}4p?v7EF<$T7hatM*z<B3u$-;Li=(&B!-=@UD4
zKW$UC+Q?QL^Qa-l>$E-bmp`1+5AC@5ar?yYO33OTC~OI0w$B$pH4x*#aXNB3omfi3
zEHxk3^<>8o8OI5z&a7%BfBvVCN+l&v%4!@q3=<(9FqHFkCYQ=J74*Gz-m4R=RWkI-
zl%5oD#%15Y`X9Faa|P5LFrJ2qL6tF%ERD^&0!?lIs?}V=qZEcH-qE((XnU<{7iKT4
zliZ-`K7a35T5NXwaVveYFaHpo?s@#P8P5RL^Z3bzM&GzJ3PXj1!lc5W!ke1|he3I9
zdnBqbS>X^fK@yS!m%GZBfBzdU4`*Kf#m`(n>ESKE`_-@bhkyU~obS)5M51_BvbMI|
zIp<B^s48kk5vMwb(AicA5%U(lY^7<wJ?cu`KTRiI-k#6`*N2&}F2CbWmB+vRdxmjj
zIvyBrPQ>HH&u(7ua-7IZ@{L_9b9&@3UKj_j`Cr5j)SMj|(YR%_q!^*@T(9+1*oK#U
zte3hSxVzU_%LCOeldsm!h1R*Hzut7!%MkG{L4%CEynV^5S1%btbO*Q;Z$&=3&|_X^
zrg&l;Coc2Zv$QF#9ZHrt@$T(~<IN2(-n?cEBPCbPIWxC**=i6<!=n(?6)dPt9Hr|h
zATCTAddd}=1O4<X#6TPdaxG}_fS{OhO$#Rr9ETG%pQ$y2xyWnDGok}w7+b15AVWlk
zfO;Cu=do7iQdq3T2*tlvt$|)f*d9a=L>gm#yUyKn`*gfyAJU{+RAE%*gfO;FU!OHQ
zes#Lxpn;cRKy#&BE6ZKtcmMDm`SO<2&BPafd7@P1?lJMZU;Q1w`S*Xr{BUKAqi5*C
zE|{Y`yrE6HVNxj~AJ}p47`D4&q$}>3aTrEko^D|=uJeVz%YV-(!q>R2+jKgjH%Cs>
zz`y$WFZle$OaAt^zad*8sOyM~gU4CHJa9IH+#Ki7o#k3xz|~RDEnC>`>4}rg`*z!W
z#q~u!pGymJg(^_Q2`t(jPm#6E7g|tzJo578OKxs&DXqhia7EM){3!%(#*w<XgFK(#
zu~^~7%@L|`y<Rz=XCCgZoL;`*&FRGJn^%MoNsC}8$9N^riIBJaw$<h^w>`+!LMm9M
zriHW!b@J|>YY@~OeR1~e;c#37_s>523`IG`6G5EC;jQr$Vrzp}y-eylUkGDhng;K}
zhEj3{tG<^B<SM8}M22;J^mVg3>*i$>RrHA?zm-S*5r*_9FrIDs$J?@5ZC%`C6@HgL
z(w5zhg(8sKOH|YkMY{gwvn~I}u;ow5v<G$HdWgb=a0J(b4a%#-#9;`$esN+7kr!bk
zov+*vf#3c8@A&WjKmQ|d-b@^Cj=cWMNUm`IxbVAQ{Vl)!_kZIox%ch(Jzvy{GmL5z
z$ojE%93(QtiBTibgzLkF^mygpK0Gjmz{?OAritlv<kfiK=5S)Zf8=f)`SyH9P+|y-
zAutWn?05&I?F?^y*53zs2rCoIS$t2sw01k*-hOAtyML9PQ*NTPwxuqtk8m6(K6~>y
zH!p5E-k$jC@t#x*qQ*D`K0CZ*3<J|JVwXqqa%N5o0@OHAt8xG6V(pI)SAO}+H+=rt
zXAI*JGvUjxzvDVDNWGBNlM0n>vobtsYrzPs{{YlNt)=O536|BR-o?9Af)Ikad0Ixp
zz~`TTff_`r?*j&_y0K~NMCi;otJ&c)4bI>arDP`~#264YvOz9TvS0*|xV>M0o@Q&D
z0=C1_ZPzuN)#?6}?D%>7apM`ldLBR72yG8laWTl){v3jL7a4<c7^8O^smdIM!vK#D
zXTJRM8(#hla;eD8*A%PVpWpJ$mtXSbZ+^=>FY5v8SvWsfI{m%Y=De7Z&Dl1te&iiF
z9J~qi@|%rhy#0UB$P`9MnVO9yS4=7o**H|t5YaKAgSYk_Vqly`$iiGptMd;nv+0hh
zi?sU|TXAq^uWO^`2kVO2?7ZWif#|}M)_&;aMQrDY8C{!Wf6%Uuh&ZriWpg-V2!t3p
zOcT>MVcDo9uUbzk&Dk%7N~V-R0jjv(RTMacfs$SHHy$U(VRX%rTDUG(mb^H)W7`pB
zEk7#+#0pSbC#2e(`aWy!LT||$e<KlNP+^SD>Jl?loQ_;;z)ZokYum2~(|RyAKh{*&
z?%qA?xFzpAOIT~qJc>WZ_JH~&?b%yvigNG8tuNYTxITdq=z$i^0rJw?L17Ha6at4K
zFonn@k#UTS2j%YVTYmkwzai8Ec|J1Td_%7A_T80lzx<M~fBPl#oDdCb3QyXfr8eC&
zvld7>8^E2It*EETPP2u5tn==r5v~yvOt4(Z)tDbMRpBhY4}>8y<N+N5V~o^sWEv)<
zWK6u1+7z{Q_3EVp*8ba?ELex1?Y*30yB=wo+^+Rdd$zILF@S7(8pakd_pc5>n{2kw
z14tENh>_EDU>rtbh*Xv}IFxm-VaqJz?bLml`sb!pSeC@)x^Q{{Ln{$RAh%9(QyiH?
zA|T{8C)>KOjRk9nRgj8M0C#9jyc>}SnDl4r35FQ8Bil^VL?Chq2WnX;W%dnTs|T&p
z)<!K@O32_PX-~wYu?-NXmuY$6z86B2_42#EQ`b<jo#=SK$sI7mr(?&kJCjuiIJ@}S
zmOpOGy*2#G6sz0yz{V<Qcl)Dlxqqhfjl&bRfLVRE<sW6s?J@jFTV8+m#8~a=FML9y
z6&|>-KmKP7&9{!q7?o*=Od5zH3{e@!$h+^}@vFc4HPlxu%ZYGvA{FE9-IZ^D_dCA&
z?U&5gd96e+`qawiTVoV+@T2x;zS@K~U$)tW?an--*`JpUQjD5iw&9YE8ieUEfOx5}
zhQJU4r)eUm19#V@flkgg8$(#_t9D=Ry@XLekNcUt!CT$8V`D9SkZszn54!DI^Ic{A
zccnVu(-)@Dz$=L3$YGi|OykNp>+_=GT2SK<oq-hxl4@%<R7jG&dwDVDc_F8QnGiKH
z4HFs%O5Hwk)AZBLZ)m=O|F0ra*HTj`RLk`CNKZzy+wQQJ26hcY#L0UZRrS(q*#U<;
zhox)%c2}iSZ@a0jjJBPtyp^E#Uy`<lw$E@Jt4D56$98UfVg~&@{)5Idfb~3nvT?vS
zo*0Btlu3jsD5Js{oQ-um1O^F=IZ_UZBn$uUfBYMM^_$=FF#nf)@$=XG?Ar%YDSUVL
zz~BD+f8c-qpZ<=Yy?jX=Cm17)0bLA4$pvz)<g$>ew}Q@P_DwHjszENowM4GvfN=(K
zN3xo<gGW&!2bs8$6W4SlFO_709Q<$vVJH=uufT$wCM*cYmnURNT;HA<!^mehCxS%g
zWX!datTey`N`N5-hH-?T6oJ8%!TfLu?QraL>>l`*=5l);#PxPU03D#J!#xrF;4yb=
z3zaGXG<eBng|awJS!3jQJo4t%D^7<ahiT&ed`G^_gyLO}R0OFNyDnVj#A6cjQ8A4y
z4;Kz$;P&+!=IcGb`Sq8KrvoqFyyEWh9ZO#LYPrWU9O_5_Mie8nm=TBs0$M7e8lf7d
zE+|^-#oJ_$0VJT|2q<cG?UwaoGCm)T8!4zoQg}($5i6Nm1{b=lo%L0r3dIzvAf>uq
zQBbNHRV$3RzQV;1@el%%i;K+~m?(4bXE;<tTh|m*X=3(u4SIVvzqu83`0`UaUXQph
z5yS<aRfTD)90%hV2aZ8G9s-j_4jee#CZ-_#cmMXk<?sLQSNvc8&;JFV|Li4S|MqK^
zocZqUJ-_;2e$C(i&ENCl_7xh&2Fh5|jDhRy-NdSub&W6Q8C5t8_~cupSYTlU<_3VO
zi!P_iFhyRTjvNTcTuF1HW<$#eQMtYXY9%gLxENFq;>Ka*)r-%#<b`+31LIKm>@+eC
z5fdTTN-C8S+>I?B<)Z3<sTFYnT!C7p>FU7>X*DW`Nmw=Rswh<jYq6La6l)DCq!aBO
zjE~Z6xH);23sNeFIC6V9@$=W8xt$lNsgjqBWmnacAy8Gg7B3wU5zvT;62eHXnad?{
zcX#3IZ@=Tko7co~a29JVe0RP>P5I*G8>q_t-G!?eOV>uJ70VeY1oc0mpkYB{rABYz
zKG+)sV=lRMSX0JPIgEkALKS#%`;w87Te;=k<vSjh3xg#FjFFcPgm6#J&T6%xqDLop
zPy?|C8(rTh8?``5LS>-J*q)0yC8;~fW~h0&mM09HF<Wc>z;Hh$I|ew>0IUFm2tnX5
zRSxmlmOrrNDo&Fce8&#vQUYcjoZAX^OqAX=@W<FPW-PT3nhWkFK9!m>QtjE6f5?_w
zaBM}WMX(>R<ql%$9;NEG44));1^9f31J6<YY#v9Lf^i&_;}AGa&U(>eOsB~0XGi|0
zfA=?h_wJGZ-T(eS<Mrpa-2U!Yr1YL0pIqOrbmn2xcuK`cwYUywo~b1hN+D<iF?B%n
zfRRWBRvUvc1g~~5IB4Xv(+#9beMo5W1Y-@6JP30rAXii}141<6&CLs@IPmUaA=OH)
zfk6jeo+flqW-(HQRG>6~Gh1DCR;?=JUJ=rK;-)vZ$DDk$bltx6&@;HLQ|E@27C_8M
zwW6UXHX7htDuV{T`0Qs))5PuRL=_<|X$AG7C}U8LV`K~iV;CU}m<Hx0QF0+lWLYxr
z-rjS)CWd%mj1yxRxf;wRbGcqgDYpuZfP{#N>(<rkx>S^ckTDhq$3-x6aBL|v5hch?
z(-C9Ui)(l}*X4qma6G+1sz2KpoeAejih}5XRrP(d`u8web%q0&Ue1X7^u24M37E%U
z#n{=ate{{AvAjELPZ2z2QhuNC`}6qcJDvfo=kb$`*!G`@(|iLEu%?~ihY4M;N{lGZ
z%37A0@=*BoZ+^>Ve&n1MmYn(K?HynJ_FL}X-t+QsK&J!X!2ZU(>YHNSfCB4DN@=+U
z0r8fyz0}9FGiqGWc97^9+ES>cvz71?fa;D@Ddr-Q6+-n)on=E@O|Y$l`viBl;O_43
zGH7sjcY<qx0Kwhe-Q9u*cV`F;PVfMC&N=Ug`x|!e>grnSSvBl}sDKvYN{p{*g-?a)
zKd`P4sP9(DS)^Yy7}YxCyPfU`R~Mcq{*R6G=;JcGx0O2i+6cPI`IRxTsET+DChBzF
z?m8j|ndTX)m=K9jm`I=T$LVCrm&n5req%ZAI}MyH^4k4l|2jPFj_xh@Jl3A>yreUd
zkDeN?f==?<?xai%eP3ytC@seY>naf#2BTRh<-Im!+?8xMQ?Hu=S?vq}R@E8wMQ=@u
zrOQ^QC&JgFReHbf{VtAHsik9PDw~B!!4}*DRjZvZzHkJu!+xxrn?&#dcH<5XA~`UT
z@5`M1y+@vmsa*N5rY3s0msgk1%k|%;4i8Cqe|S&RL*{S(eQy`uTY=tlTzUAPv2smy
zTu(St<JA8VRJNWp((<0-I-32hOPyb7<74CupFl^zds2?b$_{ona8~hdr|md{%5JCP
z9i!JM!A$0)fLOg<5i^>a=-=*ybiGB}PuPjFi>>j!i#J`OIFxxX9z6S#y~u!-E1nE5
zf(}>hR^iK2%_`$WfG8@84s(443g#U<gXKqVYi?|obk6i{#;M`yi<yw&KkJ6oK5EPx
z#FGm2nGT6|``d>kNfY3Gcs2<w>U1yZ*ga)(wu&^t7#Jmtn3*<xopn%67fJhhBO*7R
zU5(P>gb7SmxA;ZgSAXN`1qIuj(xNI@Bv@!{gqK|Yb%o!k-e=hpmk(MmC6`^d>&V6z
z8ANRy01<A%$O|Rl4SC079ST}i7-T+D!My0`Yrx3J<ai-qfv0<8?(lqi{S2u1AMP`e
z8RqNHiWG(C3}#+Yc`4vZn)jDFA=~edF_gOMU6Eg{O?;7Z>Z)<tc|RF)p0OpA(6|P9
z&((kIuWt1Uy1kxF<Sb|kk^<t)%{A&!zW~KQ7J>sF4g5LA?BV^Ehr0xMM1j{D3-h`u
zqlP~etC|mc7v(p-U+$O)80%W`!rb_jY-X8&^W+e(8ha$OICXa*%;Pd|ZRrf{qsW>9
zmwwQ;*T)*NDz8^rK23G~+s^Crl>{+7LTOnHei0afr$LB$eRM86TD~&V0(uncKDOyd
zB54x5i80vG?{87bMhYm9sLC8~aZ%1p=_R7D0skpiCL9F@dCu`^>qANYY};hO5+XmP
z$lTwvj84*E?1c2c7~yhKbba}rG09tN8i)*ePXOg&m5px)rAG$*CSO+<+T4_RYHQx-
zSDeaxkoFfjpMC#zCj6gE-P}e+vLaPHJS1&29NtX3Jg(irN{w%1;LXjn&*~lI%u&9p
z)bqZbZkQgy+W(y3QCU)b^IWYHcPNJ7k9?4zXS>Z~_qF@=uOvf(R!eJ99(|J!aIU~^
zP)l4!i|UY#&z+um$&hZ{TM+>mUL>KS245~7b-;B%Psbqz8kVYxuq9|jXIFP<T=6w`
zkF~vG@X;Hn5-_cWM7rO*IG&7FSuV@`{JsT8R*Tpu=(Zvn1Aoa)E)Ub*&YDpcV=JJR
zHWVOVQrZ#5jyhs$u^={ya>e5P@BAK()wl-|A@-gC@##<LI3o_vL)+mQ+yiJb>q+|{
zXIl7*ltUcaQjWNcig(<=LvMcNVRgDkm{kH*qAC<ebWKareLf|cvaDpG$*js=EK74t
zs}dJUy|cHM0?_5<yM&u4&xrKb9@~sKMiQpM4W?+CghgHTPvl-gaO~Zk$ckPAR-?9I
zeMLIP1ILU?&Bu}<XbnW()>I5dwLv;irAwbdug##IVxH5%5(9P{|9JYy<|l<IpHqQU
ziu-NhQ}&-^$dI8pZ35M{?<DATlzS%HFQ%d`ZT{e*l+e*g44RgccIqtRfg=>X_&eDT
ztmz02YlR2$m^~9uW8P99lt(q>iu$Bs(}5Pux*;yKaCU+2ssrUu4$i$bmU;^_b(&j_
z2;^5{i6r3>yv%=U6#9AXXW+ZscXA;EgN`hWrB*YeUcVVW)M!{cecd|Fbg<H7Otz-f
z;Q)Gg4>c(%+4IcqTbgswD>Eb@=7o1;PSoA851JhnmV_<-NJDo)iG)#cCgAPCUWVX#
z^ptx!76DzDJ6^EukgbR|C52dsg&fq|v?Nwt7ZS@OD6>7Nt?s6P3ae5N2*)h^yR8hA
zF?WPw)kb9Jh?9-{qaRj`Dpe%7)vBS(cDENpOZ`>h29@3PvFBlLh^5)2R%uhDW%p{3
z^4&tV{Wv&(?$B=P!C~qjR2!;2dOxkyJGwy`oQ1Gn*seWl5`HA4wgk2j|8vj%)!1xF
zD7V}s_x8b{rcI29W#HH;I<D%73L*5wgeomS9LO^XiuBPPs<@Iz9()kVezX1R$@}9?
zc(EeKD1Mq=xbZL@0uZVo01<V1H1Bdi-16rORbTrg#i8L4fU&8m1dHuzux5x-j{0(N
zXBz75d`tptM+Gg`1a|UrxKa!h$gEC34AUaOhk_DXDu;vXF^Ele6VGMp>|X>Cw$C$W
zA|JhaXj6d&Q!99zb;H}h*^Q@wTaVMF`?qjg0gln6!~O6~@JL=g=c=!1C+&b5ff%Kt
z(L|a&jk@9>HQQXX4g;!eTJ_Bg?{48ZQ9^zolz7J7^k$Uc?A=ffrk1sbub%O>)0P!1
zv~(OD>8n<XNkN$n&K3E_PeC7F;bp8{ZbVc3=tU`&%u(EeU^+th7Tl#|?@+wKWKmJ{
zP?<75jN!dtyjaYhVjuGpXaksP67sdVoyS|aOUZW5{nJb)@Lay0ZjjM`BbI0gx-K1K
zHrvM`Y3IAJAHP?d3=^S?N@1SX;m_v|xiQ@eOQw%6lkA|HJKQmH%Z*1jMdtYTr9IxE
zpmP9Jl3FU`Fp5=M`PShI(cMhp^MBqY5R(@7#n7!$JuiwcGgfPiyl{J^DzenvZf?1l
z!TwQu2We8anMcg-RFrZy8yIba>bdN#w9?;90-mzs@Apfix;SQ65=akBbhuq3HLqu-
zC`0~-wCgacGpw_An;fZ1!RL>Iu@$+y+=d9jm_c8RU4@i;AiDU=y>%R6#)(o}EFfgj
z^<eoTrvK@f)i*G3wEc?1cXnx9Hj=`nOBmw>E(uJSBI_xjO_!Y}%SNN?Vx}FT5pNi$
zRg>#J+Wc)9?jKT!Z4X6-pcq45uh;w%=cM>jSgzfXej+_Dyu|e_O1`|7&2|bjyOvgU
zJJ|{pMQhKNlhe(r|EKGnpN(if4nOvP5(sY!^6-9Hiq*2!yX!jL`~)lhoy{!gZYX*G
zA`s*XG|T!`>cXF2DP0MR@S+d4XASwCntC)?h_VQebKtRfP%OsclF6x-G~Yw0Sudk8
z%0*5XWBQi^w^|nhTwe<mtlX}z4s9fK!_^f_AOXs<_`3)6a950|9N&DSRPZD$QTnpo
z4S0Sc@MOPbICfu5T7AUR{26pw1KE)Dk)>(~O%YOB4fRmZt}>=irFOp}xT)j-oDrqQ
z(B<wYx8Fz<Rg4HZSf@`=9&S0=D2^3~;dcTavRq@e44y91cir?r@N|qP;7op+&nxz>
zT_xHhaqOrz+?arWS+Jj(t0|Vh{L-UgIXHVfZYx>Q)fk=N@5DB`PtfKa$1<-N0nRph
z`C?ik!Z54^pT_Q4cFVa>&J3bCSG3uFt9fCH$7~VFe^duOlId@j3l`j0qPAPIo<81n
zylgm4mqIFqa9dCIl;~vIDe)I;@hQ70oYWVGDN&25r@Jv&irb^gmXi)Uw!pc&Q_-ad
zLs4#n1T;os^deO^Fd}_718l$ea)_K&%OnkJNlr)mfO&9xbE5k<v+;;Nt3c&TsQdya
zs;E7hl|l+F_`WE7a%KVp!C`eawDA4l$ij7}JF-M8<2H^nHzcjawlDnEMx!0V@#Lwu
zEEMdR1_Ul&^Y#7D7S44&yp1*<^d@D7h66b+Oi}_J(9Tcxh5Ma^f95_fRaiKxjI4h9
zmBXq-%8JL5&PdOkSA(<Cd*au7`}D9dd9sI1BILz1N4E}DHEepsA$sfboY4PqM#SQN
zP#{~jKQMC<W8bSWb)Z~p<5yY~$DTa%2BJ%%A4xr4E2m3WsN1=&+H^$>p+J>$o3GWU
zfh_+ahe?U82-B*}rHF0oosN!gJSB_-m?N!Z=&g(2;G%j3&kx#A!0Il$WxI+r{$B17
z{<_;)e0{ohbN6)k-&_X6<%=V_wNr&JKx|sI-LA<>l$);b1+*5?J(teNq>zbq6lz#!
zBUDgahaf*^mc;-ENzy1F$Kl3%6C$B4qi_=<R>LJdmq<L@<`$@1v~2+WT^;7a(71_{
zQ~mKU+~sqcW&M4mdAXFVc$Aaus}lo0rp<X>U+qhNSK2?v4F+K!zulXc3akg})T92E
zh1=48>op$tby!O@*3P9f@$i;mFZr0Mfa9j=l$1;ew_;9Anc*su&o3&%VZ7h}Z|;NB
z$tVx<+=gcF@DBjTCE{n6g=9nIq#^4bxfpbg^e9f}xgRV}K^gkljt0&waw8>}CFuo-
zh4jhtp5+BVpf$l=SheQ`jkWXVKw<=`I3fI2ACPWqi26*U9!Z!+wDzU~TAm&@o^u)J
zk5Ejm0?e=EMpSb8Cn?+iL<04%N)!7Y#oujzh`EPf`>?m1?`%{~_E@x=W%#8rA?wPQ
zXmyHOdjK6eC774MaU)1Eq<emN(?))sD1O<Ci_JIX!+*}(kXLEkGx4Mt2@60xa|v3Z
zbVX3ha;aNQ|6<qMKwTnI%-ItyywHiYS{%{ohbHlPzx%xkES*2^d|OfsW306|@!tl(
zd^A)q{`92iek=I-zWMkV)D}N<Z-<&1oVj`Jsn;>h`6K;1O!snkAb<iQWcC>n_r!-J
zbNU*syiw__j{4u+!W|x2Dmd?c8-F{FoeYlVD9_Z=S46fCab_F`5JFiihZg-_{h^(2
zU_0x`<A(t)-sV)H)u3UZF~Hf*E&~w2pr$$?PB;3t(rsYcnVRI=oj^|1;n1+yuu+Kh
zYI~7sEMn17_gVhv?dqEb&qdpLLjHn)kM?kt&?W~YLgLZnFPZ7ortFo&#t~#{5zf4b
z`9<U=&^bWh_`<Z!H|UDk2#QnBc*ppr61LB&ziHAPaXATy5fr)d+#4DxR)DEoq3AZ#
zen8Xmd}itAFL?QDA7s{ge;)LE@4Wr+&x<lUgrI4nM(E8>iR?PZp8v$8KD4)t)|%Qb
zpjkEJ>rY$iW9w3R5Z9zn?gM?xCY@H4qK~)|jCbMW+ikW!sgqCb_=+c@{Nlm?&#1sX
zpZL;bt<7&ok?A7k$evq+CFdmbZdshk&NSMz7dWk(7IC%HN25S)v-fn1BW`}D`KBn9
zSvJxA>J1SyZtmNSUHh#1d~W!~VGg+p?v!fRAGiC+s{PzLuoWKM7TE#H-AbnfM_W~x
zew`(;>0IOJ7;Z3`J%Jr#EQi07#j3zvKS_d@GgC3!?J?L@{~r8mhm1=Ux$JA1dF;~G
zaic}yef0%w+GgL%Z6ILa52F7x-sWsnx;zZsX6Ob#-GvdA(Um%4c0_g9co-bhZ$kbC
zD><F16O{sd+SARoz^zk8Es#Bge;JsWiT5@ofDU@Y>)(So?U!fJAQaqFchD2sc(-6=
zH^OXdqLn041)LP6I6|ap0Z54jiIbIoUPiObGK%<<YoWTWuc!fuctCTxAA!D0L6z2S
z;;U61cyP4|aiL-<d$$`(WpldiP%}SZbq%nP=Il0LDNOfDM6lw){bKqvZQ8ffrPr|2
z2vSTlvrx?OlZxVK18#0{ZbUzLmjph+I6E1+;LEIX%0a-Ys-c;S5mQuDWY(dWUksKx
zoum;@XPk&3uj2@5$Fa*>lL{rT#%eRItOrum<l0E=e$~4L*y?8o>uvh_Z92n~$%4eP
z3mTAiq{B&ORG;}9A|OT`wo+j!W_}X7QY2g4VDkcm_$kFuSHlkM4USu|{}<9Ksfx2s
zT*Zyo1BFjEcl&k<fI<kH;pWj@ZLe=eo>vG{ldSe#{jC;LXaeh2JDt7`wV%Ql<0hz#
zm}kk_>ZI?kK~L{Wth%*B?Ca});F0kJ?tkK!L}f}xdHjL*?RD8;s9Vsk<3t0+*S3_3
z;SZlrx|%KwMKdCZ%a1c86y?f#GzKA33WhNyc@Vk19Okp_z7VU2z6?v|5=ZjviVbS4
zU9~oD+mjt5_n(hld}y;4hg1Jc_wkAieb-i0;9F~L&frwI7R+6bn!RPF_3R1fkCX4|
zP}yz*c{%swA=$16&FS^e=MU~nl>HsPqceOI@ccLnIls|;T?KDI_ahfGV$oP9;GyoQ
z%&gciIH(8cNc4rH>&e@{Muc4MnF6U}(|<(6OD&T;nN5qtFRchsRi(2d4JBJkg&(?*
zx7Jk*<Hv~Wv7}}RyBoNHZf?$xxx}M^81!+Af1up7HRwxH3?w7!P%u|~Q_@UyInQb@
z1JHK*#?6kgnk`XF{9G0m3G560r$rb_^N$gkUng_){ikbiS!09qNCk|S;gpD;Yy0Q#
zz`uId^vQD(V`CVp3hGh7z_+r&tN;8n_-E7lkGQ(xj#SjHhl-33y{ah+t;!wra)IW#
zGl^dW)iBz7u6=4?O+f$}MjLID<LDR2-D^vnF_zT-ZxpYmOZev2qiAJl)i`MQFz4yg
z)QdJq!BBz1J4h$Ol~^5R$uMj4EY~H@s+lqo1{b=OjwZ~;JUli87yFJ@VK77hWA%}t
zHSxQo6_HvKOq93kKlvSR$#m~}OSq4RUH9Zmt1A}z-<djr#&+avb-gc7KD4dERuC$O
zRfjH>kSZ%);7i~8CG*KR<L47uBNAu+IEZ<NcZ*o{$WXr1q$uui0*uF-qcVy2Od|~t
z5;CfEH$_d#-ZUUh-?Cr{;Qc{JJi@CAjSf#sLXZ)CK&V}{oUe*n;y|r#NVSIbS=<oc
zNIk8kfMQ^6?pGherI@6ndt!VKHWzlRm6;l>vjUE`aHIhyQ0Rs@OG(;zuF7<IxF`wp
z8@f%vZA;#G{_uYs74W2whP1mFl_YW9UeUuvF?!Im3e~fmHnlaL(F%28fU;hEE3^|R
zQc$EcJYjmf(4Nj}bbSDtux2&FlstNWPg0yQybTp*l4Dr{!BccueP;wkb$fEIx2Z_#
zm!&_Hd-L4nuV+<;tjGZM=k>r@z89V+fq6&0lAbUPS)V%Q&ipS|X?eQ8$v$*Of06b6
zp|ugV)+E)cm1ap%D)lRKp@pXA1q_288Nyp{R4hQj;g(oSTn`MYjb?*D1cd~RsXi)M
z6o8|{q)x3_`scfmb)kvHFj)*OVh2pDxj0_5>_E4%pH8s~VA8M_6^2lFam9RaUgE{?
z@D`1tndh$&CZ)qkn-x~14Zn9MfCDabw+&yJhhAXYD!*}vq=vPrJ|ch@{p>H=a!@~q
zjAmU|v7w&9^}^FzJI5w+%HEZs4Uxt*0(pWF+^gb-C%9CI%VJhpjXC0BPTAN;1_d2T
zBR07-$?6k{4CzAc9KaFVBd@s`+@C6?d0W#t&xH)^?cJj>9A6kQh5oewAvbuTHM{p+
zt0Yr9y?+;r!w1UlNzyM(A|vKqhhAWac2o&=`}$J&S51^#trQ9XDuthQ8+s={X^a4d
z<-w@WLBVgR{WC%dOcP?NMK%|FS%}YsK(egFKiLALm5=xZ#eiUpkhsOc!A3|yPZP9L
zS>q!$&*u5rBdp`PEes_BKi1mWpvp)qsq99b(0#coNn@;uQqQ2lyxOO^mAuDd)#{Xd
znQ-qM)Sr%d34C6R`ip_PQf@ZhfO1U8wtCYkGXh~E$ZV%pR?<Ff+Sfnu4cfNd|G2M&
z@gPC|b1H|<w5FyC3(?H(@rGw5aG3v<sYvT_x%F25gU_z@Ib((Hopu)5%PwOyj4p#w
zvYuk0$xdarNUOQyXE()>mplTZLpwysD}Ul*hB?32@3b`UP9l!860zL&h)rgpv`!fV
z-4(ryYm9AXMacjsEFH&7u#JK7bwYq{EAThK!Q)y*H?tW0uV*gBv2WMPmP_yw{~}H6
zWr9`Yx%_4+^7UKS>0A>_V+v>I(K4<*OP)~KX*2;qjeA8MHH1Dv>2QKC&Iox!pDYqT
zqby{wU}K<qlTqO)hb%_?H}L)Z8E^MP;Af@x1Ed^|Au5<$6mR5p*Y!EWl>Qmevew3l
z^-11w^x2Hkk<rUOCjF7RqCVv18e?z5oxcHoSq(gqKW}Qjct8I(FV&6mAzANB`r+_j
z|NonDXuHu9S)^30W{*E~o+}>tN3%UgGz0HqjQZz*yX{O%`IL1!5J~)B^bUd_Qb^IA
zb@H<jLPVHx_@+w(ruaxt;y#%uhbBsgAv&z75GK#X?<SPjdFR^uezhm?bzt+mX0BvC
zQaxz8ZXx=nw<3Q-EEOoRJYwD+(B=%H!DBOx#){Nnn0B`^LY8$kb^8r0{~Vw2S>pkL
za;Ec=klB$(6xXMg=$m(qKf3y=!|hLeR7=9o(R8a|CHe?Uqm9*bJ&&*)22c8M{b+Lw
zp}fSJy5qHuJ-8cG<~n0`<u7h5LR-67S$hLojOJQK0S|D-qCzf@+nZwnpB(Q6Mv-;J
zt1zTuDE}6x{m*(ozyLW+-MV__7*+Wqigk7&(-e-Hld~$5yG7qDEJKxqf^Dx#^_Y;g
z2efi<JpPa?F$THzYG+jnD44GTcPJ{D)<bNCTRTN8iG;Ze&NaVi!h%&ttxJfh<c#2w
z#@AMkKBB}$S!bHV1d=)nS3fi$`tQLY4Lrn+WA|%`|Hcio%+w99YKayAu0cZ@s(^4P
zzOFrm)Q5(6AbU&X$+_Kt!5N>hl;Aj+Vb_DzJLk@J&XcokXTae|fhi1g9&FDgmwmii
zkH3m-mW^`6(%{-kh;lv3oe-xILW^g03oZ8fhGwCqsosgKL2DLvLz=Xc3pWv6X?A1K
zo|_|7ZF-Lr)blsQ1Bn^T05YY_XO5rgTM8WizQGe@c#xa*2#|IP=T*p85UAirBPyD=
zf;4)m&bh>ynbJ8M=@77-q)f6hi~`42uP&dW>idd-yHp7G_l>$TcH*(=SU=n-wWO2D
zc{;pQ>TK9U;S}hPR&G4-0a7;{=XU`i>XitH94^tatJo#VI-8&ExveLSU~rCIh?Dl_
zq+zRQYJj&rYHqf<muF66j@Ic-+QswGAl&drD^xOnK>p;~rkiw5sm9llH^>8`pZU;l
z`AV<)Jh*<>c}jJ@f!nqGeJ3p62T~{A+6=3k^%R`mGba7o{C=6bQatxMNG?Xbh#S0z
zPYY0XbFR;&Yvs`ACiB+Z)J;Z!IjEHX+=E=m0b{GXA`qFNnv2}edR(2yq&N&&GOmYj
z`x6U<LhzBMxo*~R=8K~e{rHFCRL41dE#~SGYdp7mt7f&@9-qXXQ1RRc#Vl`V0oTA5
zQKYNRL=Tb*l7gmXS#!n#947HY%!v|O&a>)G%}+q`4$&Z8g84dR3bd_tv5Jcz1!vg3
zo~cKOISeF^Vz=`62Z`o%TEX_+U5`K)bU*biSA!owRj-Pm!b<DLV91)O$>)ZM#p>)F
z&F&UxIKj_J3Lpj`IF<ccx0QZ=zF>43j*%r`Q-{cA9fosa;2|SM!rV@qo%>LASu;%)
zgYL*I9Wons)u$|p#xlIqtiUSdWxhEbJMFR#wPpM<45lYiAt{DTx0!chbydo0lX56^
zI)1F{Nd%mR<c6%qEFuTcolWicPOh_R?%2UDx>u@3=A?i0FDw7wdnChQmbUh&?roY=
zvSkKYVM;7|03AUByW&TR)-P@10qBAIXn|C3G{Kr>ug@WC9tjFwNdr-@vP*ma&K4#?
zIYR<yNsF%J%rM^WLvbsm3AKXoT+?#FU8GtGz+cE*o)mF1$Ohf674Q(X;FDkI#~}f<
z_-mxYRj-W5)07{x)zzyJGnmp~pr<L$f!V02eT^G@oVe3jmUbY~(Yjl1tw7$9Xm7io
z3twhy@4tmih3*Qqp&wUZMCgZ~svqHCfG7G8C7MTn6`T$48S8E>cOiwEr6Hp@v5w4u
z^lct=o~1(p2tUXovWp<EVd`@PWV_gKXpyL)oLw5Eo834$zpeT)nyinKuUMyOJDaCZ
zUd-N>`wd*K7I}6%e)oZM*%X*w7PPBEJY*YU6ZgcmWRYQ>O&%!WpFEG#<=J%Do=&4l
za+i_r@b^ZbaaV(g25FpjV+Afb(8uVVd2h9TwoZ<v{Jr3r7uQJdz2ALxT_mla5DZpx
zTEU+i+cZwBc*$aX{36v%E-1S;UJ>BmUT?z%g@1e%``ug(XlT%QR%DcAD6Ns`dovTT
zAIgtUhqGt|a%?tV2F}?llj?T@nQM!aeA0C#4a#sGW)?1K+pL4ODTUV?vI|I0{DEr(
z__`?gkcrD5vxB-2^RspH3cvt@bQh{hIV$(qoL#!Z!G3592lHpYw^yOgy8G;wz&j-J
z-wzd}Pv43Te{~*cw!vFKepZiw>9ipHqo+d8o29h;I^oU!Gsy4#gL!yt!0ouUH|lO@
z(Z=0Mb@u40{aAIh=~0QjdDagmXeV2B<ASfEKAu8`8WBh~(V3hIv0`(syj>s`J}+$q
zFX#CB1cd8%_V<5@hv8;UIXX588?uch9uqTa;pSM+HD`2b$#>MdQWU_dhLUOXcVHOe
zN+{9)<tw9T|8M?h4_N3N6tyfIG}&O!yM7UnEuPm`VO;LePklajG}Hd=wlh*E0oz(V
z)7Q2(rkOxH=eJi5G>sNKG8L6%5WI}JjKa@(#8-sN0S~n^E_GB;o6ghk<)169nQn_P
zTJ=D5gZzfgy(;&;oz~Vdq}&jMCagYQRh~t_=bU{@GUBR>=TuMezDJX+&*t$oYbjq|
zjhK4lf^;r-{<x9BHSYM=K;nQ*{uf~WLR34*yf^DwCu2Lb7DcX$KKChFz12Qx;NoV)
z1s|d+1uFMhv6NY)NRVUA^Z7jd5^81`d^TJ}_j$AP2+`f!e#`VV;uFp36egYbHH{9T
zpzCC*&=T<vJmNz(77392m1>F_jc6+mC~0|UR(3_Ikh=l@CA#xZ9OJ4Q&7M!C6r+~0
z%)`1%MjV9V`SmnrxSa{plO2oLh(EZeADEYy()HKTxAe5`okmp=*V1!3tOBLrfVB<2
zq;hg&X9d-+(x@)6seL*ZPjoHeoAFeDG=eEILXFi~GC?Gkc&ZvlB@hG+q*{ZjgfR;-
z4W>XI-ycb4Fb))a4QyBXk>S@iA>{5%Q@oDvv4`*6!|kG7(sC<)r=&EG!DPB@m`)WJ
zPe8*oHO+>qv>2VytXX2|qcM5O#EYSeLo*Z8Io;Dqfb`SbK|C%6t~goI^~KvWo@*JT
zB<Fhuo85V7aK%IaZm4ik8|CkHwMaN<bMc+z&-4tGHu-mqrLcSr)7eyUijjef{j`0b
zFy@be0HY2YFbg%*;w+`j_WZ~x7~)llf^jwq8WWoPcR5v`kv@9*XBCB272z`Ifxi{I
zZk2J>O?-0tgjL3Ad|-1*7})e!RwXluG};ZKR%JR+u#5#FY9rdyE@W|z6MkAPJ4|Fj
zVStGXsYo(qbIZdp^wBdwQW@6^Yd-i*drT$BA?D3m3RwrNXb=L6u@O)sEam+o<d`Ao
z9^W66nWRa{oY`XG<~6QY#V=WyiYCWB!XZb4Jx_@$NnXKrLP`y<5CVxHZ<r*H%<x8i
z{kNP%L$KQ;_%mIJ+;5f36%?j7svu)$IX_VWqlquZDdD<e<u0L}QekcnuU1)x9Yzt3
z6F!05OtUJp6p~`Cz&zp#UAtDATC5ddOc(!?JXN2QZsgDUw<=x)apWq{tC{<bw4u!!
z*&Ob`^JBXS^<RWegiw(}j15l%#~KhkrWORklwEJCyGH1p-a6bMu|6JObF(&*Gyu23
zC;soFUvEIeE52+D;w=@A8<;#^`_zVO_~G7o43P5n@_t|t*NWpnER7WnJNQ;TGJp>~
zeuyF^`@XWn@#PTWEnA&U2y^&ktA<}L2~(IlbwBj>2IGBoi65#lYwZM};x#8fnSAa-
zFS&i8bqnNYix~1gDciMqKQWqWg&`*){%vJny?uGEBSlbj`*NJ5)pzI@_<eMOJft8A
zpOKrUeG2Gj6TicXxB>NghP4KDNfyesn(Q`+)mLjs=IclM=(+NkyHnVCq`(kksW%XC
zm0X1Cm}ic-rGi1J8{E|4l~StY*xFW<<SI}|_)D>@Br^+7LS9J=V#e>VI`sa0r;og~
zep~93?!?^Ztz;@OvT1++sfcs01S+0UZDnA7i%_eQ!4zLC!ot?_(*h%Pi&+;^3zfmp
z01^}txYL5w0D!Dk;R(gjWRF6&M1+>|O(GM@b$OKIypkm7`Y(jDe*9m~I^V+w3CPO0
z#)FUGyM^TV5^#C0y~>buv<JIPSF0IM0C<wJ{gizFxe$eToUf^5T1fQTr!P4<T}+K}
z_uDIpIXJWhR<~vO;WH%Nn)26R8hIoRUSGhFi0y}QE@d-(mCs6%Gic=!$FP5xNj=uu
zxMgu=Rmpo!<q51)Q8hEQNaBl?cCMHDHfJ;2Gv21^p8VQ=O3)4P>zviWbI_{lFgHe7
zI?v7Rq9Ua(7&Jlmro|Xp{}ml=Z~GMsC>si>B`QWQLO&ZrSoHc=2|G8bKOUf8rh_o%
zBjz!cLZ!CsOrD*tT`=UC=j-FM=g}q3-Ex@-V2ar5JWOUt>4C?oyI8KLE#Pn%&=Pzv
zg=?WyrYeQX;PPv>j$hJaGoAY7*jdJH*&3>se*o2O)oY-4dInK`vzgm5b*tu#m>Z@u
z%}{*lQykE1`BPOG$(BOkSnsb-FpD<j5VRm8wd6NFQKPzO?VSV|tJIchCX2Xf(=1{p
zv>z!FSaUXW9hvg(sXE<%Vv+5-*14)jn!t_NcsV48pWtb5cAW(li-FU~R)wC6r+UMA
zswIIdJbp62S!;V6E}R2Nd$6BRS-r$81Ie1(WZP0#*HTslGJQ=>v+!)?(NcZ@p1!MP
zW9>ScEe#aqdC|e$8t&m6zFPcwAR~db^UvcI`-A$W{Jr``1b>>_y{Ef^`xo)*Q{x|p
z#1zg}{<c*xM0E7G^KjK8kVU03_z+U{iaw{+d9(Ez*dd@^05u?ds_XonmWcpz?5~MT
z*H!%c$s&S4f6pus^mg@ir2FGlv-yO((Bdmu6n9Yo<hgG;wGehn0@LnzP5aNsAg#Pj
z$(rJS5U&PH@i4!JO(9AIOGesQ&!S1vXruBV>#2aQwjSgA<t?YtMuW5ihg&T{P6(2Q
z3o8+l_-m91nSiX({7(f2vQjbce5aum2z)TZ#%6~Ok?ZVcQl#lhIm$z&Lm=Vfuth@C
z%{sU(R&c5*zeXN-Tiz|MyzBRhaZqh(ifaY)=v&viBH6PZjBZL?9KBlZa`V5X3FeWY
zV?0=GtVa2e9E+g)u>NX%F9^~7x;UfwTqEyfbH@ZukftY6&@s>sQvkoL*UO+IEkUo^
zQ$}wd->|h<5h&DI-rMr#ceMqMuj_;i-~+0I&P>!1PDdhcDlHgmVG23sP(J_o5eI4_
zm338<2P-*zR6cojm*s*tJc;U;gPS(=@~br-y8KrjHv7spcLiRZYKi|vM_isioi_WE
z1_A!h8*r+thyeEg1iXnetRoSsHzKTnWK>IF3Dw8|N?#*($ZfR{xkBba9mp(f{G<?<
z)ZTWtV45M3TH3Fvp{p7;SW{H!aU=lNxH!Ss#0?h)CMspP-BQkPv%ez!l$%hixz>L1
z9rR|sHNPWHa`z7Pbs$^a(cQ9obsv;Bkgb$`+I+{~#0*-8Hv(#9ExLm*W_Mv4{Xm_9
zMS^}83_MPrj;m;O6u|1%IN86m>Kn654oh0uhif)-!*b|efH3QxfvdY0KJ(*9eA+7}
z%y4vZLGW5hh%ZK=*z4H<kecIc!jw<lW00D6hAkMbc}N4-+amPVN8~P-UzbmcRp)f5
zhXS5jrq1_<MYy(RP=B3@Yu`{?0>1~lu@x5=srrsOnO$h0GA)--Wpw(p<5kp=yT<d|
zp{84!PE<WF^QUVC)h5488(XE3*z`>)ES3|v%y^HPJU!GFrCgNrAC302K7|!k=<tsE
zmJ!cceEIc9ZLV98eAyg);og{KshCjd<NK-fa5lD4XWi)ni=L?@F28_g+VY83VQUmy
zpciZ*LAM`X1C4mZ;hz}<joIpL1NN&9O%86@*wkgxYDJkEs0QwEFB<TL`6}u4LnHX6
zmxEC!a%?uip7E~sJ+f$U)pN7G1X&%PGU>U=g#`P2ENleG;D|ON8%__eaZO&BGHJ-M
zHo--lAwv-kKd&HT=|yx&)5m2}p$`)uuN&4Z;BoD!-rJ+ARdL4jqm&8JsLlW5dVqJd
zqMT}x&n-u$N~~&KJQ!Xqh9NCL_`$C!kL2n+q#R0B3dZN;w$As+xHMJfy9!ug4G0Uq
zyQS%eq7`g@YLV&wQ>UpNPwV#%nb%LIt@o|i;jV-3A2)9yK^|8oU=pNe7-z`M-P=YS
zif%5p+%3c#-QY>h^Jw}Kt4nZxC63jGz#=mE`)y#%(73uKnE!c)C;e=60d2k#jspo>
z<Ho7bzI{2DL{YGQBuCkoFF~$WXiL<sceaE-snR0E`^S%IdY4j->MtcxlGKxsL>a2_
z;UtLOd1$SzV$?eT$T7RCK0PV9_;i-5?^nA6#|^tWmOU2MJgtGyPU2ImwYrWOQUxZk
zdzxBVo}W9~ZmF|$tKN0sx?yUoHuH|Z5b^Rlw{}~<%^fYM;H1mWulJx$uMechaTAZR
zi4Jdbb|7~uz&`xva|}W}tWe^G$zaPzsm##11X_C$^%8H<a7si}BuNS-G5=()byc2H
zG3Pt9tBhfXAH(N2W(hFZFBpQ@coB+#45^Kz8ETPh!FL4}(p+Q9UmER>dVV$jb^6np
zD(Kf74b<-u*Lkne=N*6hCz5-AL2x;9;QQ}6?*HtEl3^z^3MgJE%IM2`P1trgEQ*Cv
zhXrFA$PtVhlohwPcXRwblRa<SI@HzZseFKN+XBiGXdr_b4hd{Ya@r6F9%vm1oU=%!
znq)^s6(`V|Sf5yBr18hS`JUch;$V1%sj)0=$<Boj#Ij;>`OdJQFl{%pgqp^XhG}Iq
zgkFu_OcB=psXrkkCDr{Fcf2)bEY?R}5olXG&M2RoNfBu5G}NfQ5uzep9cHQi>Aq>2
z+`qvdFTB+)p@Gp@bJ<JGP?&{9VkY&tNz99V1xXqk;A)01Q-v%|Y(*IL2XM>buV`_{
zbKFX+kSl*I(H0>w<T0XEi3}GMuXm$?b!UB|a2X1F1KeJei$l=rv1n0C+S5m9^_$lU
z`~HB%xSMlDznb|SdHi(1I?6x-kvWVQ|F0Y}-jxiE!svuXg$70hJ|#sQWD^8}=L#Mh
z>R?#+*jRNMeo+^`ixcI9_$rGFS5)_s!f9|=a0JAA;oTV~mOo*W!Bi%}j`tm|V5r!F
z<2n%!9!myU9_9xo-6{c$(@qb0AK%UAtG;^`eZoZa26mrpXT=I_O7zcAEpCrKbIDul
z3Gtgg=&ZtdV;CcJ%Y%CwC}A>@5IkaL<XE4s8DnJA!Jn6V%S)&{4x<9IvIrl8F(xax
zF~W5;(XdjH)^2ZfDt+P^d1@#XuGohkcweAzPqT7xEQ=-vbkFu|@G98<JI_T|)4Kc6
zq{|vs3GZHOgUIrwx^Ksy{-#g31HpSN`6t5TZ|O=?n+?~#_hNX65<5kn;zkXC5lE2I
zJ|Jh-+|*xRoy4X}zZ>cvmwg{Sp<<V@i9x(?;SkTQtm0EFNjacl66Qmj<Rd1CigCr4
zpFR>W@7Qyw9fXM|M-`72@8}lNsM7%j9C+|uv;L5kzyM(<nVR8$scBt+)cXL*Sf49|
z-A{`207C@v68;tD&Dug9+O+Re3WkSyxE*rlUte}k<A!GvkVyBcg_KbP6oy;eQv*~d
z;|rJ|ox*Dyaa95Cd|4EKcKuaBn5q>D?gyK}gYiXac%TgoO!_T`K^;`(vZf#=U%vJ1
z&D?q#g^klmOTq8NLbC%g>xQ;IgP%F%5_;=&%p;D_S6^Y@G@C8$DPL)w#yz)xgm+pn
z4YyLqB3X4pK`C@HVTVi5j07)_LTMlsgmB2@{QU!or64hz<aY1O=|z%Z2f)OMXOwUh
z#SQOC+%78$fSaygnssdOAwhswdTA1ng}mW~JZf~1%o<sA;ZRs-Arlr}33z!4HS0Hd
zHv?smWoY@ST-B&KjI3|WMa|#j(O7*aX!1UT^K8$IQ<ZXWMoG@=UZpm`N$ZN$;kg^}
zX69DZ$x^RxC{E%zDo_J!jMo%D+%|5fkX6Uxc0^d6U}R8~v26^N8_$g)<$#$l6h0Ck
zUv@WY;l2q-1dBtdP}Lp$36a7tRQw;ec7a`4GQrW&=<bXYH639Jdg@dUxgw3-KAV)2
zlaqdGT<{7|ePOWv?6uh|su8Bfa}nEVx4yI^-!R_L9=!DDIF4)3tM<2f)XF~z=^kn?
zXGq>A{)y&tMj+|O{(rW`#69^1+~)3Jf5jXYW<j<upO2$-^=dd)CA#S3j_N<o+?B%j
z+l`+oFhjk$2)11(JsAfV$pN4v35BmIn~w`n=Ggd>7HXF0_i@gX)mZOm50NQxZtXq~
zm6KIp)o(6+*lO3+D^fN%MBwb)_BAkU{fK1-HV+cid$};KFf~_6)(ocVJa+AlfO$Ix
z%JtX~)md~t+;N%nSbe2A9kkv5NN$1VCyaX+l#MetD_?S?q=HfiD-l22fN2>`#g2+L
z;5IKKX)+&@Q>VLFU^1s-x2H{EE>K&LB&ZKJJq2wE*HH4c7)DQ^W0s)k!bB_9N@}z4
zkG!W1`O5II!Ym30+Sbv?_x%B_@7`J(bA|{=wae`S!*Q|ljHUA-?5KImXt{xwBRi#4
zNK@IVwBdw_rC*-1D*P3#kw`oEJiAUy{@!inSrLk+!gw#vrKgRy#z(3#Xmg||>+`!5
z$x)vU+@C39y&v~KDFP_}K}L}!zt^R4;62eBL9|Q`s6RCn_X9FlIn;1g7m09aaUxBx
zEuCUZk@>Nt2_%j{eB|00WQIDbls*ad^#RvSbMw$ftG7v%J2XWtC%UMWmqL-YdmaJ}
zkdJ}1I#<D{xbVm&si=2VUu|IfszuRahh#6F(PmOLPXL=2&*g^0#3yE%oil>PgFe4l
zesoZ(rcTw0Ea5P7S#2LCN6wsSsd7mjOFleM5F&{>eP>zv_q+Bknd1JxCpr4ef)FEk
z6J-jYpSczG-(746a$_i%6-C06DBzp4DrSZ50}E_Xc=iMUllM0Y3OF$l5@jH%N`<D(
z5Gb>el9s*kFH(|*daU1Y`R+7`Hmj7izVp6#e*YkdV^C%-pJO+G?@rf$e7wf|;=53K
ztx2>}s)}x{<Y54@Ge5K&JFahTF&4IAdV95P87tb<gnpCw@G;l{8$9#}Ev3zDmD_mr
zPG>&P?WBPxXOm3iTFMAUkWIuHrLlo&`*sLu!GFXK%dMtbLugqNuBOZTBs41vGD~Y|
z5F)?gfwMR|flBb!<*Wcgx?l#}FckHcGKE6szZ#70`Y*21*a!{qapG}>Z+V=XjA<ym
zQ80%dW`vOZ0yj6WLdGQ^>sRbSrWO+lwVEy-<4PMa!{pwaTI{9{MKOGoYiN&}JM>n*
z-N<|!@A6nf2@;U+u;pr3{_c(B6V#_q@>1ejR&O3ok?Xc0j+Ua#mYWwpDGhwjW+Qvm
zfA;&b;2_(ij>RD}Kks+xNXFS8Qb+fx!nMDuBcwI?oO)Da({=jbNs4h}9zy(8l1E*k
zd<+KrfAw?e$q5n94z6*-(F+s?<!tW4nti*Jm!Oy&m)TDA(Q(}fDro$tlS@*=Ub;BV
z57k=Nq!?!Mh$@VTB$sTIivqsXqm7(d+&Pi~8-gX>YFj)%^Ewlqz6e5woc8|>IdS3f
zzVWQe(be`S*Ivzxb>F2P7W^p@Vu|Utq@7_|cnf;%e|P31{`~Lr$DOEJ-3e&RcT8=7
zK(%P(4<If2=IL7Q7YfC^K>YwdjC!x|Q$dKW0NaFEk^)ay!B$_uOb>S7+r4l$VzmR2
zhiW&X?%!D!nofW;{m}C}#_!4f(GF;>-gs987CUx@2eal@j-kaEbZnCFje!Jev|u7~
zAYv)ySpk5qRyNvtrRpgpp0oA#%<QqL4I~Dm3+B=NuIaB>%Hj@Ifh}yDDC5JV@%hq^
z-&ppU`+z)0Z$HcM`ysYjb-povRESQ!6Mg6Pw)a&Be?qvro*NCG<np>FTt!-?r}$>z
zJ=;(fAGmVb!r!8f^XKm_EmCnbk}V3PWoenuQ6cU|CZ^1iZx3K+J`!Z!pK5jzBU7k$
zaAwy`?szUcspH38J`LXm*Jgxofc<AfKPdz<@bqch5qkc@s>@b&4R7}^J04;|H>k1h
zx|-Ap=x%aRq}ipIZROSz7jvLqvl;t(^Ea1-D(pcvy1g(0qGI}ALG}<qtE%n?r&5wW
zN-MP%3dI_sT8e(|MX)t)G2%~HT6OrFxa0YEM^~vPHP8@vIH}O8FCZO5-R$pNgFeRh
z4F2l6*re=)SY6j^dV;A#I5hM~g;MTipmZ_JZ$)e-YrDxyI-Zl(lt~AID!*(yMFZb_
ztr{jlgys-kRQwHqDaAg(hYb-;j$r!4#L~!9FcR0nZ}UYT)quq&9#zT2bon)AP*0aa
z%7XB>sSCOQDyr1%9aSYY7hm&VH(^Z&Q=t%>`vLE3@Gtr#_mE@*tNy3-<uc6^C1o0)
z%A37o|LsbaGv>=wxdP0rVq)|@pL+scf#k{c+^x~DhPy-Idryk+Khu@)WdaF>j8?|>
zwCU$Gp^m!dZYeynrPvBH%nc|ubk#h#ZR@4c=N{0TpW6{ZU%1-xeJbn&weRA5XCSYk
zz&3~Fkcz6y`oGFMXHB&2?~M+fOA5cpAjauRtjN7%%`iXNPEmV<y4q^9)upNsba!w;
zI~2f58PwZovvGJcIuCGIH^s`IMhw~T*mkp@*eflWG^^8mbjjqGq5(`kXLa}+_AH_9
z*T3Jh{$}U+xa2+3m;iYZkBi~w+;B*J+y(Thxq+cBrI@YvKlkkv50XP^K~ACu2M{HU
z4&(TNjl4CPg&;KqfBc@?3dSl#V`hr;lgqWOo)|JJjbLRtI4(^Ad!H@dD8LHN?gn1$
zA`T9+9}OQ6q+6WL<HBQSZnHAmvkid0I0-)RvQzaA+7^!8G1d3ufbB8t#4naEVHkuC
zgN2W&U?`Td11jKF(H){^n3*)l)g<|LYS?-e9dOX`tGhuH0n36A0<h5;b!SozLT^Im
z@!J6s-{DvH?LGl{at85Yd6y=C|8yobG%bbXmJk^we5qR%EE-70wVU}9V3`43?&q*q
zT*D-WhNin1t5vFS!=1)?$Y!Di^0m^_u)oxnZ&jPN<QJ%*o6wJ3SJVKm!+L!_>^->_
zdJp)|Mqz4$6@9+m_W<J>rKVp-Lw^LV$YOXRk;TYPM<EsIYG^Q_@0nZDpn*|i8KZA&
ze9z+E+qOZVJ7cP)*Q{NnQjE3{PBXfRlD@!_ke9UezAaRn7I@<z*(OmJt3`n$>Ug|H
zvIs(5&HH6L=x~)7ZcV!M=XBs9Gw+^jV@g{{aYI1!suJH`ZkZF0`OZdg{OL`1D$vEi
zz`Yw6=#}*A>o;GU%%Z&Y%6PQnb;ea(#mYpD5K{;13~&L2En|t8D(%1x83no2*`9m-
zVvgDz7<!ZT2bK9`mihRW=~*|nBs0KNvB$AI8BHz^N@~Yyr%lC8hEOn(ZZ))Kq%G@R
zy4w<c-Rj91A|SZPs<l1QoojNwUIGO-i#!Tj(16R&$wtcE=aGr1N>S)vHSUuos;r5d
z#o{GqQ+?%+q^6GOA<(_434KIQ3dwz+<)4?J3j*ER`A~yY+AYSWI(0{|q2%9I&9Bvt
zgzy@$&2TOkJe3O-ryIe1tB(~2s3{IRw%^T4Mg^CYLsqf^@@*|=$S+5OW_nKl+C>S@
zy9KIyPuCz!u7Flfv^M;}xmhoQ6&E>{>s{QACR4c6$7WJs1)ld0aVm9#FB-PTb%89K
z0T0|{p$N6Dezq5U($fr6O)mIjAo9yW9s43syzx4*o`GXaqo#g1-GXOHo0yw2W(Ipa
z7VmJ(cr^^{&gid<6Gp`>5+A!4KBDj1E27g~m$JNWAHTci?yW7CSHms_j=u@@e%9~3
zV!U!?I@jbd--Cp(qZ=5_+1}JY;sxuz$+jN92)Ato=2YyM&cmf`4?erMTJE$T_wKF)
zP_$dxf^Q`aa6>L`;~$M~zX*GQx@+ZJttxp%*fc7opomfo8Qmw34*GtKi4jw>cNyva
z+D@>9(~-2FW-E_NEmLP$2>^q1h|vDp6gb!@ec4K=?YKNg4{v$#V`<-ePceHUMCU|y
z(is%$RN%%$henO^s>iWDtA5Ix|JMwVt3&Oxf~>77Q(goqJnF3PoEQhW2hM=#{h+`7
zZ1bOMwx25j_J6~A)Pq?N2(`B#E2w7OxF824sp(f1b_k!HUGna>IONv1zdCrY>V98V
z=<Z+nW74Y^=sb7Tzm96fyyywjqTieSI2dSei^QoQ{oNjA^}xTTh2<73u%+D4I7-wa
zd)(FS87R`al8e0JXL@lzlaSB!@MsnE*lUdrp0ob1G4K^uY9XPFpnx|+Pg@V6G8@sF
zC~aaI3n*<ZVE(uJ0shE3<tr-YU}qPlE}|=z(IVHe)<UDupixr*z3>&EAsEH&c*-7r
zf^ci|8l6CMLijOuj(jr69enL$s*<OkVnSZ(c%Vj8GoUtlx>QL$d#bQ9B(~@Jtwi@g
z-1-1woy!yIFvNA6Ro2>SCR#1^`SST5)lF3E!6`?x5GM&qPR$)0;w1Msj6C(6hGleJ
zpFOxR#jE$-il|`u=R~K4hC+%)Mf^rB9Sn@Rup*ZYc^gXrGPQfdZ%~AEIO^7}aAMlD
zb-zzUm74&=iJy2~W_Of#Pp`=B_c;}*nI*m?d)J#guS9KB`<9?WI0Dd|3WQLpvZ(TX
z>yP=q1xl`L=Dt~&yvT;+Z5(i?gN4HwOyf#?_q(I6Dh`I;qXtQu0AB#ZdviS8<>mV3
z%|2fg^wmH#ysi9ud2h_46{WqJ9?Ez>(YnqE1{Qi?xor<}RrhOEt+YJZ7aZg;;3+Bm
z;@piKJckNw<M+<L+rs`H3cB0<*N<P8WJ}0~k@g${*r<9TDy5vROr9KtNtelVxu9gd
zs3D|7%VCNi<hQkm!&<d3OXQ1_RY1xJ+~f!?Z~fO8bbJOqQ!2xd89`8E(&=VIT8V9B
z#hT6!_`;Bw7Lk&SBHA9zf;Xq|osb=oP)3QR)?)=`$s|%{u#6Tvck4mA#uIRdsV5nY
zuikUiDY)3kC%(ASw!0~kWKXfW&@@llBDh(MK8p1z3V%~X7m2}$kamay-htjQO^QW^
zn<g^J6#an9H5rK{b;@&}?ERhh-<OXUyc{P0Q-W)Q7vj*p1Ovg6bndE252RT*_|#;B
zGLNzr$34SVoeF2}!|Ce&iiw0%!%;5vl_UK)$y^}mHUj;J&Eg$c-0+7450nv`{l&mc
zhapQI#(47zJ?pG~#YUGc58IybUSm3;zGu0Szjt45=(f@M>rZ@0e*lT9Nz}D-^J<I*
z^F)5&w-MuWaiE1YTbR*SAq>p5Z8@$xHUoRGFMF|8xW{y}XB__Qm)lL{DPi6Cf4AEy
zDWVBu{8n!S*AQb9RYvglp4S*_Y=R4|f`R^P8>~H!wI<1zNN}oipRZVm`4Z4+rR9pc
z175$M+SP>rEd0|CZ|f>UM0j&V!?3I3oqzDtp|2~rIlgqu_lY$cEs(Eah$WAVx!3GR
z4f~N>C#pZ=SXMbKFt7l%iymz{Qq#es1Ebv{uj<fOUQmnHg`pGNkK#>7O6R>|$t7M*
z-U!1*w#V?dsZ#?5%95B0CP|zU({QymD~#t`Zc+slGIn*HICqCwAo@O=a`e#QeF*RI
zJT>wbPj2ekNg>3#?D-75Qq!;E{eH&uqJdda%cD{Y@Y<h&FqcShS1Jah_5zTz45Xia
zp7iZI!A`;V3q{qpg$ajOYK?>{bts*hT3E~Ht9H|K1IPY?KfeIAT&YtUt#pCzW<ti^
z&i%ivtBxk*kCSo!ms?o~M*6M4>-U(AAKrP&_V@n*-asM0&C2k@b%=qS{F`@o?-<iS
z7>*2`Rn^{E@B^i9kOXbaqE(-ht^3UGOB55O$foI3Od5r?BC8sDV~wo`R4)_jK!<;k
zA(OeL<f7$^GKp}aI}e^MS;bS$Hr)~PG6QcJ&FYZ!grf_}s@H;d^a;PVp4r%Ef17MG
zhHjY~*$!FQ9B2O}cT_yFDr;s{*G%db%sB(f=g><XTHD?=&~ZR2mBQsZla>X`nPo|&
zC8J?jlVoeww*fpNo2I~ef7?y@A=`5^?YZ_j6Z2NRIeRxn4SpTP3Uit%*0i%4<EpmI
zoHU`ytutfG5*62q5SlP=`~I2<?iW_~^V$#54W5$yJ+wR8gI9fiKZPOM=2~j&*2pgF
z)|RF<K+?SZ@-&XkVX0i_3x&ercp$#V!s{b6G5lrO08RhQ-8{ENU8-M)8+sF;P00c6
z(f68Rj4o2E9zgFCc(rL`c+l!!&h#r>(n6&$W!Gante|zlXy9ZADBC`v=2{-ojZU`D
z>%&3hrokcB)}*hg&Q`&8YopmFb+g*;Sg%`I+uynlhpn5$7TEZQ+nR*+4Ti>0NtA0|
z9n0DRS{hq5=fWk;q?$?Dv#tFaoXxto$$smdf#OVAp?Yaq*!I#bLa-GwGGNJ>%QACW
z5&@&e(Vt)NZ1kqGVcWyprPEh7Wn-a$to!wr?X7EBw_)qP_kk(4J+aR!{knG_yLE{C
zeHQemGoEev53uFNaOzmcdKUgiu;rB<(t^0ol6kh}Zp*BuXw5fTpF^8Gq4QvO&yr8m
zmRGh^HB^mN_Ja1@S6>S-uT&VKm+EzC3#rO54jVIm=O;@8Z#G}8ug&I3S_62RPgsrC
zw8WinQ6=T#5*^|IWQeVlub+>rZ)B^;z7#KGt&&;FLaK=*6{(Qw=-m=Etet&VAh<o_
zTs!-*S&&H6XK#DYj?LF@5v*dXj_GaztoO9}ot<U4`KtRhw^cs$<fjyWZIB)7RQi_C
z^iy|lmU1Rv+y2zbY&8@#Mk<MUPM&<u#XESiJ?D*)Rawg+*Mr~xuF@21YiV}4U9+?b
zgKjMb%sFvhu3YDZP%E4c?QArvxsG6MBKo}&p--!?pUvOzRT$N5TbtMlvbKf2F|OL4
z*x7V@ADZ<uw!g1nRu3qj$3Oe=3}8KvpKOHQ7Fh!^zz|@JhKg}dCKn^+LKHy;<!~6#
zF>?2IX1-?b@4sUz6L0?F4HA{%L09T*P?stG|FidJPqrl6c_;Q;%iSYl@0`w5!<~Q&
zfa(Tm-7S-~=|R@k(u4X*f41~Qi8QeZfdCt10~f%(sH$6a&dJO@M1&7Z56fd`)&<#3
z04$ORNH5D8voiON9V>kKx4vOrGxNNVMlL3Y;F~*a4Ul_&m_{r_NQt82v}zp9FSSsb
zp+ms5`E-@d7ir5C@)W^&U0CzVk_+cD)1)!CGmXli!XDx77HX^H>NMOq4)E;<O7o;A
zfKV%g7%@5EL<hTE&kmspLy1UnL&(jLvOTrHfVt0yDnSAeFjJy<iVVF0Z>?Q6vQ6?4
zVnR3OPytq#CPY=Ie^UvPP_%<;-aMR-Gt2RW;3mB1yig6!t?=>q$S?ou8wM>5!8n{|
zh8Wlnqi<dw=*Q*@rlB8Ryt{C~$ogg|0ZlGhL3|U8Nk9Wb8NIm(#^JQmmXGB1#BPY>
zcbTOw9Og&f@R~F^d%}90y1TbKwOjka&03?Bf~f<gE%x-Ai~Q))ngA{vMLd<!?Sp+;
z|BmMq;D-CuUJf<({f+p)^c9mgxs~egI|zh=X`wVv8TG%q8RsxgjJrJ#_YW-dOs$z<
z!qizCniM)FOY^pGw@Xat#QTp6QQ&%fx<)SlF(mi_x76lz{E)Das2W-7nKf5r*B}aa
zudeZqjXtGpH1UL`+PtEfy;(*IOFNTo;Zzq8BV0vl2wbaB%vf?pL>Pt<4QEU|VNB>A
zby#O_DiZnucRMX1Gz<NF)B_BH5v3yL@(pVSMZKv{_<IeJ+GQ8+58?+%?FjvrJQSmU
zUtQ``$QE5{-P*H5oB4sZ8D-8K9uAxjXJVIVueFey(?pi@%y-8FT2`V}etJAHM8%p1
z8<gq^LeQTlK`*{J+PzPP6fg=o&s42lhPAzthF~uL?#V;nK0b0d&a9^cV+_2#-urtT
z6-@?G@t}y%)80Tf2kK{yx&zS_vxph2YoQcRHmbFf^Wtn5a~6`Jv~7sMb5Po^WBK1G
z)=#O9;iZ=UCbb+=Ahzf%uZ;olpP`ne`(avhj~MCCnU`Arel1fSc&v0_t~pceNvg`F
zmj7LQ@TXMEF~(;A0Nv}zuW79g0MuuoMhb!JDR7(@)>`Q86+?2yk*1!^dKnHXrMPdA
zpgqa@NupZKJVXh_J*)&1C72>1q7pbCjx0HYy06x_b2IL4o$ugdPom7VA!bl2Rx-A(
z%xz({OjgjS+_sSsWsHFhTzigGG+?!!x7hEg2FS){a*vRH-2@APRx8rkSrYsiV}>(S
zalkF~SG`Fe{o;B3e{UEoPhENH>8es|4+?VDl0r|$45Z}j=qJN@<M43c@#B%#?_M*E
zBUY3&Dq$K}mX+hffy2!e*VDu<Owa+TYIJYg#wNytUeD4QOCrP=DAve%#Z>(Lm_NW<
zPumXCvpEGm93D8Fj;s%7#t`}Z_L?XVq)=PoG|%j=RuYL;@j<uJ!F*@am2OsFYr9AX
zt%qk+b5%xdPf~RAr<~uQeO)l_->BoiwqAg%e^BcMz<OCf*b?)jQ<Oku^DZz%o~Dqc
z)($3ts4~RJn3^{u;weF{WHVaVT-RD?%_U{MH*}d^-5%Po8TxsH=BY?dcNUgXD8(pd
z%;$yrtdR^cQd?%Nxi{9z+T1+RsqDUql~$;&QD~mBBZ5g_&4n10ysi+mH>C*)9+4Xe
zv{n!iQVOUkmVJ5|G|;xm(zY>n)7tHVnHfrWn&kSoT+H7wKXhG2aI$U0Pt&KTDQ^c?
zkWLMEkgi#eu@!Z*{XWiZeFK{3h2v>ve|<}ck@pYxEV*)jHa?yX9P=5AifW)#$dx`h
z-QHV|>P85tH(ET+SOnh)TeVAikk#(D(E}m3NxZ6ZUKfh3L@Gm2Pr<@z>t>C$%&6`N
z61)lF*^ovYymFc|NHao6?jZtLGjb{2%ZwiRTQC2=+0!J^?>*T%q-};t6YEJhTsGKk
z!uctz%}6(f;5NUtlA@q7qEh|9Rf?xP4)lgii7J$>nQ4t`erU7aD2W-`6W^rAlZCjs
z^!Nd!ltPRV5t!G48I0oySU=DyxhcoBu;k4AZ~&Azj@?TpQ>ip*thG^#`sdtPe;fIu
z)W%XPxi?qG^WulqYuox&Qi#2|^N+YRf3IUVgWc$){aoP(qF$3y#pM~cwP1nTO=Y(<
zt*xf(mI{SX`v^F0FJk+7&+6FOEB$%UI-8;=Y5F+cC$if?&J0RvoK7<j_YYj%T$9p3
zT^Eka%E#j)_veMf@ysB`ppfg9nB)(d+7Fhs*Mgu=2~Se{v!#M|ro}cyC>8YSu)FpC
z2COAx(nukaVsh4MF|b0;3#a)=hyy`Vmo9B=BK$Mz=L<i)I)m2lA)i>6E&+ZHNHlAR
zT`0T#^Kb^i#=7hnq?0bqWug6FT@tN6mGx4~e}h`C_T=5U-P`RyPc0*U-Tj9L#Pp?>
zzgNpX_)?^Uqh8O_y^^dw$s`dvgNN_!)t^=^cYB8*wBn!LFa%TsdR|?IS1VN=U=oC=
z{?$x6(4}siZ53<1slr*8s^Rh>TsS(`@5%0y7o=U-d!^Ox&*sBkkw7saH}CNq;_2NT
zplqSiw6N4l%|g`Z&)Me;tX5fDr343+&r1fBY1|mb_`tI{JJigmjf*|Ub0xtxfVAy{
zH`r)kt>O2ubkE+6A=Ge*xo&%a|5ESyI^%FFWp7<CwH!k5zS4@nE~^CXo>XiK5;w^&
zwW4NdG0vx%<MGJ;?iz`aim<lI<LN}sGbI;J^UCqO5O;|f_2L!RtX*FFww~SM{_M7%
zdMte~Z+^|ygg^rP-axjqQ-RhPhk<=eOeqmsqt-Lkp`_m17YDD*U(fbSt?#QFJIh0q
z`aNWgr8&^<GT@$qs?C{tDjhF;I=8T$d-})mn{~Xs=}!raep!F}^#WkMtRHUm30U+@
zX-ca|>rI}|FUZyX97;8k4vZ;MMo<f&ovYM$p;n56SgpEjP#`8DMmQb|*0#RIo59*s
z3rn@`JrO7drzLYPm2)W`58evFjFb|$;}x|Q)|wfkM-4A!VO@NB6Eq^-Z>w2zO1@Pj
z1`t^D;yzc)!Y~euLt;#BL_cUE5^2>p`=kkNN?R&o?pb5q+n_ZU16s`vVi`W+ZeUy7
zZ*Sacg{r<OT$(_$9u4o;TAE!pU%x)jOeFhUBLr{EUDn|Tm)tja>r<3<%^c4&4~HYy
zuiufzkw5#(UvQjP9!ucYk4FyqgvyQ>BWs1VxP+k=o>7VYAf^2?F?&Y97H=L(ad6H8
zP7T<i%KIiL!R4UGd4^oLO^IEXY6cO;lz_^4J(9+esG&;-9f(2<Q=yffPF8&A6JsPs
z7i2bTtR=fgg9*77mg4Ctrb6XO`nV}Nep3wT4h?eoH``vjNgMu4U)Cy!l41la6sx2d
z8OA`+l~No)uIgMg6(y;XqEI7#KiB5$!{#+izrQ!&m?#h-#wWk9&0}S|PN!v|w#LoP
zl^>)^?s6XCammcf!g?&kIPm%F&nabLDeg(tsI0k>7pNpZ(5f>YhLBjRab626omzjK
z5fkinuLTs1ash~~8Kw0AQRxxS)xFX#@zPtT)$oyF6Ys7qf+>X=aFgZd@}jM=RG@0Z
z`sMllhvN6yR?E&%q^G@BEVvJaQQgzT^l~6E6LPK`9*?~L?gJ7NqbRi$9?mnrzW>N!
z&fK3@u7-pSiCh}FG?MnW)au380^CQ)E&&y_-}i+r?%oWa(&vh3L*v92CS?8KKbM8I
z3Ga4yOyk556Rj0!g=IbScna+H*Nigt*li=??95J#L}CYx>t>2wDJ5SHytx)~?LI?T
zf7Uh7{&RTRKl}+Gm)BVOFu?5ct)E=8{8Gz*vs!j=1!qZg0@Z)&T5j4dX<}yL(<)wS
z`FSlH&KOixGz6NsRLj8_b@tnKsP6Ado?ZW^&*2kU&825Sl|rM`#y%vbF`<vyy_8C!
zN}yE-JEI6Gc-pgr&yZW9G&gCt<~oC*ipGfaUf01r>wBAj9;Gyto?d8>tF!&0&;x&-
z0IYK^Ai^*X9du~4P^lp^mzBI2<FG?bp!lcLw6c`S(kgBV*EpV61m)F@OY*D-99g%s
z*y=c{vl&&y0jOqJeHtt|ElZ%fFO--%$4b>b=%gpzZNP|iCQt`soJlE9M(%dG`Mr)+
zh#|Ngg*i*KNFk!`nQCr+A2c8+Id?W<%N!3U9v%){zk7uYk!FFVR6ZUaD0PLFc|0%d
z4rlha6QUl@uvw{?wQUabOlJ86&bC@1w9XQDHH6dxro=Gy{muH%F|uVE%B$TSH{-;9
z7_q!?%9+xXW`SC>%eKv53+9IWm&uQG#%kk|AX=J{TVt`z(kg3t0<f$#RGn$Mc?>?c
z`uN|i<CpahX1xGdFY5<efhWDV3dA71dUeBoKXR<!!n}4PPGeb|B^3u{OcAxfxt%H5
zI)G4Is$~$G_c%?2n4BI=@OeP!6HVv<fw*xjs?0gNsd}r-4VDV0)7j&i*Gvq`JRTin
zYi=khQpt5;sTlxmN~L<Uyics>OkEqhxOX{R^GS=9ibNxrKrXb>knZC(?ws8e^hq{w
zJ}<1fAW<)50285x;J#r!Ifmc{maQ6OoEr>SZ<Ip^t!f|bhCcDyJf13j!X(>tlP8u_
z_kIeUCT#|hK$L*SF3W6?bK$%!tUZOP#y}RhKb@F!CdL7YBc?lwB}xm_+AiRk_SY39
z5(AVX7sn6(?{2zp8fau{%g$a?CFuxMR5PN6XeCW9(fsVy9e2}2(g+knQ(9>(>q?20
zsGV(b+32L-C+a>Dn{443E!?kL*<PJD3w>(YdIw<iF0TKGCg0ynH-85Cgg4mj81IKP
z<MpdM?p|N>@${EG9?zJ0Q#;QPCL^ZEZWsw-EVg1M9M_q=7Nk_9Q}b*9f-=aCK~_r5
z&T#2HqvNYy5TyyhJQ1^$!trz_TOeEDa5%8$%+eY`Gwbp-LIqi!VQ8?n%-W31$|eQc
z4||5)9xX~+{hkzs{p7%N%>@XAAt1#Yct6OBgXgLQji^LaV`unnJ`R^RKnN~-z^Z#I
zH8&b|!)Na}TlbXn8VH0gM)m6;(4`!H587qwyhX&<+S8l1uZ?*nLJWzZ5zM;u#$_AX
z3{u2m;OlR{<Noo;cgF{o0!tGniAda0nvyGMXVPv=Lmv`AkZ=i$ke;-Z;)KaU?M-kO
zoOqKT27*<hmCgVtTu)c*!pIl5cU%nvlO_t8Wd6B3sGV8Y1(*PNI+!*iw?eI%WzC4d
zG=%=l8E7ml?aWduOEG^HqMk5^QEPU#MwiLR#-#h}Mew~t?xz^t{Zh;SVYTcGCB1-A
z4$k~%sAZ>pp+r=*2jS>TEkCPePnz}vZ>a@Sr3LLN&)Zr0X=A<^x<4!3PpOtS_8my~
z>Jx5nZn(R<=Kj2LI?uh1EAtFv9M}zR%s)^$;y}WZS8A?Eu3i(Rx@1$7l2=N}-Mb35
z4NZhDV+~Z4g5TpyX)K2mm4#}7j}HgtW$nGQ5aU8n2evUc4cAbS*f{4$1Z9%s;AMa)
ziPi$eR@96X5;4Tf^EV(28mJ)>B%oSJ$vn78f`4uH6!0`q+TP0L^WUgEo56fTgLDR_
z?+=C6TRg?&nS^%x(a&nvno&!;s6|4T2O|#RhZvvISi1~(&4qa`7|OK2;=}z%J{%qp
zZu#(d=DWi^iZKS|yf#itV_AwnZQ3@-u9xSI-|Gri=oR9=Ri*XttZ+FqL~BlR(&+fH
z(4TmPtKBWTVdCxnhU+vighb7y4<>vMt*x?_jF15B35<pn4_&IQu;$#if$jYob8ej0
z!djp<^}#%w{~$e}Lug9>ecN!!RoEu~{yKhH|1j1Ifc3I|uoX5wxOp6GRAIlHxV^h(
z{O%)5$>idxZUwNx%_fHs(PW&?LaTIx^?FgyJ22KCWI+gVi;K03fjc}s$TYZgY%PUS
zE9Z4(F=cJa>FjAlrDjjkSXYJ+*oWj(KdV&BRLfYGNzjy_%CL)s5+Dm9Bw`HI8r|=x
zS(mbjHwHLTq++D#hstdO3mV*)q*P}?UETmR4a5*g$)#;7%37_b`AKKA^~>CwXOV6&
zi@Cqpc4RT@sVz^#CHg?a|NhBD*8hj6=wIoQI-1J?bIIh^+$=M=Pt19B>UxL~(FBsH
zrj#Ccz5zTNyFpda7QFTEpxQP877bX4-Y_)jfL|bJqM7v1+23nQ17k?s-QI9BjsyuV
z-7v-KlLXMEk<T{x-h{Zv$TrD&HhJr5Q5!WI{C#+OOzSd>C%NZ^b@T6B{r7Fx;nSJ*
zUZ~sjshle8_j_L5y<#_g&9ctq)tP{KZCG+9!w@4XiM&<{uypfL?NZH6J8C9GNkkdg
zk|_Ic++IL8Bld>N5OQrCPG^>;EKNC|XL2bl25N<J&cq1A0Ao~=1}cSWjUt_KNX5cL
zni6`BZfY%x1{l*wiq4kQpajGx(4}1tL?O^+6Anl}HOoD8;Mw6x`g;)bdta>63~(cH
zuR&*6Sg%>>#=NlAw#%HVJ^8X3TqfgYKE0&-OW4Kz%BQ_SwxA26Y!gp#FA!B8PDk#K
zM~=%(F{Ri*(*WA*xIvPqTJSRQL##_jp4C)Pmt$MOTBoHr)PSN65LgG;%{`f>l(-pp
z++OXtN&|zqYe+D^7T~PfQZgZo{W<3kuhu<Pa+j)=pl%Fn01+cQTedZFgAH6Tph;)E
ziayo!^C+b6SHlI^-v0Zi)G)u)@_%?O%d@=`wnN>2`dVhI<qjOg;KQ$%TK;}5OaB;Z
zt558Yt-*1~=W;JTfvKNnEqkva7dEJhFijITH&^VYL@q0}C%fep1~q4(N4NAiuM5>0
zYbic#HJqu`+40hvr#-joeQF2wFSw1m0r?h!WQa5L%gWk>+?0pYnR)f$vjIU@LJ)=|
z3^9;`Ai+VbIlI@^5S0+zGpnfsoz&`mUWmjHE<LJ<4|dZ<E~v!Vh^pPk>sc?`&mouo
zj;+P(`?Y!cqDu%ns4Gul7TdE5q+?xnY3Zc@>`NV6@6S7{jj$z=noDt^KV#$qK-E%N
z<^@9-#)*&TkF2$zQMf<N9F`MF0%4eZ;Gf;jqXxPB&ep?}<l;rw-)cV#Z2OuUr0b;)
zob=fH?jE{@dG9+81K0aK*ZYa9kVq2xwWMEPb-%>w17ge>XVx4nb5O0i>{mR@sx~86
zBbUZn+#{=P5u~DyP6^#BOD@;_d%YCTKFY0^?{zHCfa;g^_j|nnSTE}bUIuLi=xK67
zt-@|M(F7ikCxQv8BPs6K4STXQN~q}B_2I|U32nyB?us!D{-Dx^L}h<{!(8Cu{sE*B
zE}KOMN<_KZ-5?^oe|(^{%36)J!cv8$xKu4S$ZDigkX<BQ@1RQwuWxqj_dD{M+3oiH
z{O7+%hye-8`LwW{Gmi&na^0mnh5)xiWE=*jG*H$|&V|t2{~`rv6HStkKIx4qk;X>b
zHY7kbalL!RXRlxJ<(oIWxxQgPPGmJs^N~OJSO1*zJabxBOrWSIJc{BH%LeA|jh)hL
zm!`Kh^C?CHnj89t<`n)a8B-s-g%HqgG+9gLw9L?qF-?5<_y~uYwFnYNZtfDgW^z1Z
zWngWI<5G!n<)(V$T6=TvpHFaBmpH@8x;=gEA-P6FRTzS@Pa`QNTI`-Fb!ChFz1v@N
zJ6&_LAKAr(ReG9WpcGiw%6VNd3EktQa|1WYhMH5+`mU$CNl=T^@N4c;5;rx?CA&;Z
z1nJH41;l#R|88oC3!54~$&EG96QJ6HHX&&sFAK-VM^Xx2%XRjEfh<tn>upMr7AlX6
zQr61+cqB>W)y*B_FnJ@BUCQ9*yoDht!5Ts60SX@VEJ0CoSwl8s$(7@2;oO9|32SkW
zwWWBYR?Qd&g%B80A`FrJUYG`myK9$53EbS=^6uSRuJ5in&nu6|ds-{Jdwb2G%4nVd
zq%nGAd#z9k)(T+=L=6ln5Hu3DNic3`+q5^V4f1O4K@uXj`)kH=;-QGM5RMD2^jokr
zSd6NDa=T5u`{6-VyVsv}5Vg4vQ7!I6gaiC$Mrp;LH6f6aGdb5~bra-MW=I2X-o9fg
zjnnB2<B0Amnv#b%q#aEX=M3junbyiUZFTCBu^|nYWUJQj3AVJ$)E5y~VCcpK>j#u!
z#Og|&-IF6I++FQ?dws|C7#Y={M`B9FJyOOrQkti!-NY+YV5!+9k_ZL&HEiaf_xUt;
ziB)2)3yW3e(m1b>OZV2%?)!5&#6B-z|N9M?eqz+YCt1k6)bih~mR(M0;>O?^SF(ry
z)U}L}DbAo&LzAZ?zSQzlE!PHBDo-h7UZ2lRwne+7jNh!@Pp_7p1L#b>D2Akn)xvU~
zNx>PDr?T~c0W}AIQcv^zu*?pcl$E53S2r&8tj$=ebB?ybZct}EO5N^T5Az{Jngst+
zWyve=@9)W`l%_n+nKjogk@M$hipn?!u6BZk$h04sMq$5q&?Sm;_v(%>fAl%$b7ff@
z%X$VYT#tdI!m!ETFf2xbxe2;iA&$L{qXVXbUWTxyem@i{YLL3+h>Ezksu`ZoAcQL2
z=Hf}t>qBV|jc~lg_luy~CFxJgYQ}0K@C535?=1{CU(;%HK-i4377h;wZeP7&e{;*@
z`;~G&@$vox568vX{vjf9ph#pzSV|*=9$)~(5Q%2Q)~H2#@9zj_AbQ`ck)|sJ2fAXH
zhsPKg#y}~Jb$$Yku6Gl!Z?4z{cgQkx(MJ})?-Eef?AIguS!P~~+j1<<#<W%`L!t%9
z%Sx%ne9oNb%6T=cMFt%i<2BYIt9|JG_HXRRu+hVwh9b}R<jZjQW&J~0F96od`oWez
zWG*O*Pi(poxIWjZqd56K^bN<oiBtqJ$V(=PGNjR)zUHYp){G#LsBS)p$OWOdj>LPa
zj^F0FR7z`98l}4LR%<>9-tdddG_#o}`>Bx9fS8ik!fv<c)vH@Xg(fh>fjs$yY0a4w
zgs6Q|+?(KPc)ABZ{R*-rk@U$=pM1)eoQQ8m(>U?^_BF4rU-Rnf9dEC1xt=E4C>-Vk
z)3~G5imaI?J_K(YKxF1r;mg!QFWUI6!Bg8<)cwKV#C>xv%p_qG`fq7`rMizvjFEX+
zDQlscAQDMQhz21QFxT&|wX$TRR!^YWoUVM*`(zw%{dEY=5)<)=_ZSsPk!c)>F|kNM
znzN=n_O~)bWj7>Z0F_Sr?gQJ_e|TvQKBcsIu*uVq_W4&|+eG>z$J(SG{oZ(*$J^za
zZqlQFR~^W2{PU-@wBJ*$)t<)_XeF~Qi{D4vMAQI6mz{L`W<LRX6PPQ+$T*IuMyeU1
z`J^vbXGe;<tVjE#(?6r956DEwwUVnF`Q|2M6QANW^Lw#DHbbkMx{3x;9Eke}O&G3j
zJcSeEYQJL|2ZBa+(?m=zC%fKF1R&%U(QuhIZ+7Sx2H`gC-3Gk;#^>Fq)hP{(QQ0Ns
z_3dl+yB#qrOJ2BN@4*VIti2Am`)iweKl|a&uE%yx^Y7WdE_b%yp~8jjDbkzrTCh@3
z4Gd%CSbY+T1~epM8W4%Jl%Ofqr2FO+Y70IY#eZ&4uWD`15;5bFFx1Swdr(|n9fUC{
zAw}XCVfD$sf6hWmk!egsm5boYOdQtk!wC0aal>JkKs28yw~ZaS8HP8C8fHpu$<y_E
zB2uaS#uV+KlwCfzZQ1h=v-i<|t#0k<T7N3*rI!B&wcOb)+h=e5=c(n(Vcjo(XNkSk
z^7C5Wev9$siSzti`;7kf?_Qr+Ez8AgsCMIeq18&x4kla}NC;Ty!@_`QW3x~d+|Nz~
z#$iM?xN&!LmVWEMLpy-gKRUlVd-Js2lZMyrvSzZWf0asZM(zgsVqAujb@cmrH%%mo
zv=(vu3gPzdn%moJVl1S&u-`@R^QfqpOJ6%9F{)zW$*}!tuiIYOg}-y@e=lSoadpF>
zkdnXulQn89{n~Etb9?r0Ec@@@M9;<?eRa^(+U0(0mu}Q$!`pqjG5Bk(lw2?oLQ0qh
zH*l|hW|5vyVd54dr5V+ld(mwQfeyy0I+JRXK0kk@)?ccr^#t(1Fmze+5W(67v^R66
z+&Cnnb+tv?jQgLSP}aIRySN0hRQD(@wYiCHD;KcOy=EJ=``FdypHo2HQbZHEU*o^F
z9}Bo8!gjsC|2@C?DZH$||LX<7dRaf%l2&jd?9yYImHGHcSr?CStuEV8OGw?(varr)
zm&KWzs@DpS-#rp`du~2|3vu9nSx5)>g9@`Uta<@zbW4xgUW521RIBlDIQzyIot3k$
z0qN-*MY2EGE6m50U*`|}_HTX3Pk!_n`{9bH@yQ#i%l48ASFd)25cuq~TS_jx|K@9~
zR;JJ&rnw@up$)Fm9+AYt;(J+xMxjt>RzQus`nCGe+t2v&i=Xm;{_p=s@<U~LUr6VK
zl*pS`cTB^PtJ_=7HFNlI?*|JBh!koO#<5SkRh;gvaxs71rmn$2oBL&1YuH+`(I;y`
z2UMd+Cb!BmpMfsr8z&|m`1-pCPPuS%cLS=VAWTExN+KmB+8TH~JyO=n?NwojiNM4r
zw2#^kMlI8-A<_M&)|}amdxqPAtD7sP-8JuSZ&3|=eOlohnNI`NW@=fPl5#sG$VZn;
z<XfjFTVn{3VcJn@;oMFnoiK&4fmqTo$sn1i13|+jWzZ9tj0jk*<h;6>dp86XI`w)3
z057-Q6Ps&eZ+yxdYiO(2R4dT^PnO3=>T+g1E!5J`>fr1+?uluF5GgrR%Bib{6lx*F
zz}5a54UzL&(9t8T?~eyQ9v^Aeh=?1y_E`083U4%8%{-nO^Lb@0P&z{+XP?5i;2s**
z3RDQ^LV7jv=Jgxi-MnHNCKxAYLexghm33Zu{pyNoH<5<V!7AT={T(f5m-k{6Ts|B%
zaW(8HC`Bp(B{V-w7F(!mp=E=lj{Qnyzu)saKmQ%x-Mr!RyLbHfi_f^bzT&>T=kfH&
z|L+g}koO-y^5?(&6{<!QBS!c33%Or$ZJ4@;T@&2wuCp4WWA{|qj&fe_)+6N`?E=PI
z^Teu}GjZB84ikU%>u)(O3vs_^7$fiAD(Bibw1%t$twbJ<XJ|8{j6E)U^8*@CDoVp_
zq1HknW3BO+ABk6m;i@rC1DXate|5_+4Ag1hcu-Cc;o=*T#>99vVCNIH!b)P5%BqbK
zkg5(uG^4C1P?ri@HEMR>5o>UGoS9{T+k_G*OC)DwU7=*5Wg&$flL{pt0cVXkaN*BF
z3E_J+@SAn~sruQy)bih;mUH!cU9~43SFW%3{LXLx4)3o2)73Jf)Jy1!Flu1X@KVda
zQOk3=2Xm_mlkTf&?nwzevqyMl$j_%)%ffO2utErgI8c`p$1<~=otYQR{Yzt<2x%my
z-ZS;!q@<D98be52?`{Yov8;t<K2uddj2MN`n;k<pk_)Ao-+vvLU0Y?&mE&B=*2pGg
z^Sh}k-ZQ(Q6!T*2cY+Z4`Oklc*SB||Zp*P7CQ8oEB#V_cRNma(kz!;$o~R{L=Go=L
zR!DK6YUF0v)2ecyA*}~+^nq*5m6i>QUdN*DVS9ab&E5W**H<_E=!?&}y4v$czxYG$
zA3ySb{^0ix5H^@wx(sz2ICtq<p$~mkklK)D-B;GX-x-t+45>&@-K)gT49lI>zpg8D
zDa2NJ%!Na#4AT`m8QEPMSYVOBT#a$9thtba-vgd1>GH&s5^JqUXMGusV(urUHPfa-
z+Baf~q&V`Uw^xiqWR=9}D4ZTc4^9xKG;%dg#PuGku~cQLjcg-ou}kh1ZOXcyX&7r!
zTJdyLsRe0n;x9p2&ok><Sk}_dU<y@<C`l)3Qx;QHj1Z&)X@1Yw=Mx3E7}H<s`1us^
zW&ML#F96od`oY%bI<P%b#Bg0Sg;rc2zIkM9vB!wg$e1}xD-ua&9!pD)$SwtG)%(7d
zePSZ5N1O}AJmn&&QU#X0P<wJ+1*Z!)L#=Ke=peq;RC9B934#57&+AurL>c;=*8jiK
zQ&mKSken7hj0vlWmhY$~!@5#Sp_Kx-VP#B8B%q}m2TP-hP*wc(nGvEg4I{t(TR-Kq
zH$UQPxM7x+nipy|7AgEHe8r(Y^6+@%bY3rj3K+TgMis|?Nuh6`mD-tYo9U;zofo#Y
zo*3!hTda|3h&5uXNW%?r@u^QkX&Na4D~-7oq*KYwKx()|t>;#VRF>5}4pI_A^6OPQ
z^-b6|BWFC7ni1WR#)u}62ogO(Fr-E-#*iYp1X>M#IFEwnhPDl!Ge8MMqK#k`p*g53
zP>F0VkXv$Gu!fq)Qj7Y*KbOYZ%q0sgQ}C3X?WVru5N=~VQ^ZZsKx)75?dcxt4fG}x
z{S?;5K6nnpwc4n;T)cv`|MyabT3w!1n^G-e!C7wu12M$j0DB#$R^89c6+&oz`YlbM
ziV!;!qxA-Op^}&4!@Uau?uo?!y}p38LQDhKSJzx!?HGrV7Nbu+v$G6hR6>LhjnKRG
zX*bZKCt8(Ku##!2Jte_E?jZ8a@KiQ}N;mO0M3vjySG;-kmY;q8TU^I0Zo)M~+o5&f
zc6d#ak<UN-f@&aNedTNfdrIoGe$Q;S4o`QD0|VB;#M3zGY(_41Y7~m6uJza2kIjA`
zH$!S=E`@U`Ox1neLW)E+QZ;IJbLBZ#Vh~nK{u|5_VpRiD`-Q*MfImx1n@Ji-<A@Fe
znj&c!h$&&X1YsHzB^xETP2@np13TP<P-sjv;*&2AE|0_-wFdW4%$~y3su8;x&c-gO
zb@jt|(ay{?qC^G;)J1i)Ui)Is1Te!7B=mjEtRHkY1Ncv`j$va<Jm;x+GF8bvj=pXG
z<)xOlT5hyUEx)~c%g;XhyVNqE=Dus&Q&l_#&CF;TXxc5(#ThHkA~uRtBDEi+UuyYz
zEhiFz5wS+;wW=*_lp!DBZRpE(Q(CWs4orM12JX<`!ghkkz<SDa^C5+`b}680f|;9S
z8^L>J4QPOYffN#87c*{K9dr_Lww(Z0x(rwZL)l~t>)MUO6-t->w$|Bm=3=|n?;)TA
z4#qHy+}zx9_v)6r+Z#^v>@wCV0gZKW;1>xDLnOsO%}NM~X$Y)2b6#fmnc6<nOC3Ku
zbM%^S;_N;Vh$(P$b;Zwr{4@5$j_Yyav$yZKzTUIg@A7bXAlRRAIGy<Lcz@|nH{#C)
z@a*#1P9kzU3K}%OMn?BzqI=yLs2QbYO6`f`{rh~LEzCJ%XAdYS1~CLe6vlC+R<~a%
zUFJELOe~FwPO0mI^$^?>*qOA>q;mh)iiJvu$}sE*(bLgm8vU=W5mgw6L|NU}D)i8W
z;9WnhpV5ra-=kiD;gVO18My_w@o4^t4Sk=eX4K|GirN(oK16nZ+h~cTK+(&y(?Qzq
zEh_$jUOo%9TYS@Qbi%?j=G)8q`@LQOte5q}E$tJVp70@fDjYR;Um^ElQpE-7X@D?M
z+d^R`q{K)fT<?e>vX%^sO911T&=5J51IKy-tprn26LN*QHU<r}3hy89yRXA07c`_c
z(zf{kCJvCzwGu;OyuIbK&p+q4fBG{HhZA{qW>7Ov!0VkTni6ruC2k?|?%gfxyzuK^
ze$9M7aXze=6hv3zq@<V#M+a(FpbknI18XtXxp95H<K5dg{7?VW|BXp*IsJM^TNc)n
zGCw{rpAY==KlrD7$lvkZ@z*YE+YZgr$V)}Cq1S^Sz*?hP;b==w9}%PlB0^2x0M}zC
znh*nI3b|%Pl&qDz+=zAHI^AG3GuIO<6RSp+C@js$<%mjT(1}t7&VgLc%yl7@1=Kj4
zGFagDN||KjIz-BGW|>#i+}yPkqqWSMj}Wfe-As@&Su0tLpp9W%!5Zy);(T<SeTb1b
zC}}aI8aY&|LL!lD$JlPLd_Zg^O=m<SCq{4RN~C268YMb2pv;XXp6oQ|%43EPb>)K{
zSyX8{B2^G;1X&Og#(-IJHevKByR1%~mg;5;P4pteanSJ-*=2P@PxNUTR-Y1gJQCU&
z^T%^zh>h#1lG{ijz&H~yZe<{bKxvsdH<&YZHxeYV+eK;_pw4Jd)hWP&5~~oJ1JTwT
zP%PEWX)EGQRo6vWt00D{u^QwGyZtpk`s@pKJ3rhm)g?6N`9uhT-E~A$quS#2BXIZj
z$^#2dh2?n0?jNuq6qHpf3x!cVO{ii-%oE5|$SO1|3~}Ule)s>t&))qO|HFU%KX873
z;_%G_%Qr@jGe7#Rcic{Ik^ed}?DqW8zx_AVn#s+OT&M^-Bw}z`goz(AJW)tcM+Ima
zs9}4CX+P9vkEd5K=`s)zimg<PRfJqJB^N#{GpBW>4uc!H4uPQsq(&asGpBXt!*O9L
z@aEk_z_4;gED_>>W<yHrb*zLMF$yaW?9&^ruis+HAc77%k1Jn35Xii`yJ9&g=Uf@8
zl8UmI83~PZxu#ge><z^RvM50f2}Z3m&4l$lP^z(%(i_*tZZ{E=P@8*jEbEDRZOrHB
zk|V85bs$;fXxCISYh6h;G0Nya2lL<xF=WB-|4SWvO(@dYnx9A=;{%Ik>e3Cov`ei>
zD;{xfMi8Yn<G8-m@>4BGe)s2p_gW4CG#X8vLE_#tUN<yY!$4_)CN9mdviMM~ZNVj6
z4(rA8OD%t|mTzg6IoC|Gz$gQ@h5NYq*u|5wHb0CvdjI)}?a7-W(qDU{)k3MBN>@w8
zn!~WojJ3d8gjxd@M$m}HM%2bQda_?_g*i8DUa_{)RFDX9bS7#Qqs$e=g8--%tC`2u
zfzPQFmL-GZS}yZ!P)x|;o{(r~S+>UA-D`gLw||?T{_ICwUrqe!pZ<bUjj9tZ7zEhu
z6UxXix^a7)gdvW+dG(sd`vc#7^Bv2wIH)N`)ke}Tua$<1kVGK3NjlAe%&V(Czx9(J
z^Z)q2{@<x<C7)MbeRj?MX2*Z`-~6|nmYM(OKmC94%fI?1fAEKY;CsME^-=X^TZ|Se
zRV&5n6GN+~TsIs%4Z(rx(kfPq4@w&-m1em!1&t;`RprC+NXaV?>q>5L6%%0)Ufl)G
zbK&v0Ft>)SnGYwI0q=I#9-vSg8kHE`+sfgk2zZh@q0&ZSKkd2QzoCSN1<-iq!{POw
zQG}b@YnI0o^IVAq2gzDSFmiO@voX*}AOY=3>F38YYbj(+P!-9MVI0^Gi3ZH2IK$Q;
zH=~$?elbJ_Q}#NdHS>OnJvgiOK1aKiiF758x6#dTOAqlqS=7NTE4(nV{(-C)0PAJ_
zaH~I1ux?(uO%&}yC-4W9K``qwt=y;IO2Ca6>6_kCE0Q4?tZN2o<Wi~4b@hr%A9@%o
zr3tkdHkV7f3f3ny_9O%I-o_b7gTmeQ6~Fz{pYYk+x7=P`aXQYPQq`wX8U{B2Zjkdz
zE(OEAvi8G38Y5qR`If`u$VY+u`<ZoF8Re?8bGlSTK$?+?2XCkdKl|}d`10Ln#2m2-
zHoJW9e*Vb&!`D0>Kk#w+z_A{v{XLjH8CE)@$w)PkLPV4MiqWS}B1W~^wd_9C3?YIk
zxix0Xtfin5Ff(#6igkAgRYFXZxp7)&h9PntCx$)+y&Co$t?+)HsX?G7uu8Vd<FcYL
zx>+Q~SUV#xclo0Efl&xpP@43~Y$;GHr{l>da!CoA7)QmbC&#EN)ha{iEHWvWHqc0`
zLY|v9%i3x12xfh9Xs~%)mC_Isa&Y-qEuKHnvQdj4w8SDVd<;*F7y;?&IKKMNmF*tq
z8DmR21MJh8_-<^0hFuJNFZE;Fc}fTZaa2-h)V5;cNoOJS6Ib<VY;KLPHi8M~@r0J-
zb*T{zZk()rtZC*94+EC!=EB<9l+C<mqUpsWX7fBlgsc6IH+Q#u@#W{defJJ*rdeUG
z3y6}Yfk0r8^lXyq-BB2nAc2(jlxfG!wB!E%BZtF*6aqsC)NQ(Jl`24R<932^bG_sC
z=9Yi_d%wr)@g3{K%KBKz2c<<I7{B<7U$M%WU)=v&zW(kTmZgUSm{FHPsKyuv2zp5@
zHQA^s;!*~4Q`L%3ysI@bD{HH)R<Y*O*W+>`#KdhgYBko{IG4=17^J`uN1}>*{*5E^
z+$eKKX@ub36?uiVG!o9T@nl&=W`9YWyevrY>!L!VvX;ex_6i3*5`!XLoH6aLkSbU<
zn!;)egL;A}^fQAwOK4dj^=q*#tIN#H4@J4UF|Qcz_Yz{C6uZwxDUD?<wCc&QttnPu
z$(5#!v>PyuJsy2KAlmmPx$^AZ`hGgyr?Wg|%kTEh<IwuE=*cwPrRV9Tmj9#Ia)?Uo
z9#+=bU21T7!dj*C#A()Cj)mXjFSY!?UCZnG@2_QhUd!8_%*IIcJ-Ka3h@Zsr*Xf^?
zT$m@9BuV^z1y#lvQT1LGF&}6I^`TQ0S~E|SUR~x^TXP8^ifZaTPQN#s->bFP=voU%
zrHPVrVJ*H7sd;i?z!`CyhtUuNH$&v^_J+ILI|s{aBLruL)z-Xb$LO9&Y#y@JpH;!l
z`gd0oKmPHTJUl#dI2_px1LwT5NW1uV`EyJ=Km^km*zflI{_p>He182U^&wJcrJUht
zLOvA2+eR_rYIn<S`iKPW{W1i<`P?~hODo-2l^a8MV-59+uC-EWtgRcfHe<ESqJi29
zr70$rj8P<TDw%U#slk)Kb6#o98%dGoEb3NSta6-%K)|m2dqr^nuNYA6X7d|CkAlX;
z)!jrKm0UBqRa!L;rz0r@k~I)v+V4GAX>CZrElxD_tRt^O!BZ#aC9|IrA^0#Sm&_qA
z1R2=r-k&WoP+IdLF@mX6bVGu{gcd*})TL2PY1QS?t-CaeT{h(J<5IFcBl<sdme#X&
z`(^z-UoQaG%lg5VXg6=fffG@~ieYto$a_Ph&Akw;xX+LTLqZ@(1O%(a%_Xg%)riGd
zmj&yaW-0!l4Jw4zA9^%$Z7x|Xo@ilpvmJDRq;*zm=qD2eCKcY^-tc=r|2aSU>@)6e
zZu#cdAF0)dgBvL#>avTqGB2}(X{wB^G7N#;l=$q;=X`uWVzu&k{J@-7u7^8s_-r!=
z2d$l++uWDq_kQ+se)9Q`&^gf7hMh9!^MUWq-|`m^f6jM@U-O>#WR@+E7dlyAEr5A~
z(a;8lmi!<gt`*eK5UCWZH|pI?GNcGqS*V;#W^RjLdo@<66zkJVQDPd%$A#lLGhPqu
z;>gvK88vaYyXL(eIiDY~=C5bXg;L;tUP(hDrih6#RcDa2wO!H?MI)|PRG%!@)f1Vd
zaV!U73{0=~gb<k~L2=VZRi#$<5EI2GTR|WsY8CQakiDUyv!G1<y?6Or>j#aJ8&p`M
zHzs5Uuxy2rJ^B|*Zc0q-2PO02kaoEO8-2OYa<@$ue`|e!^hvF)v2W64slmNUthy=b
zR>x6=>rqLu(b@vR+y^Ff$xD}gEVcOko{bU|8v<tH>0dE;<JApJ%#3WFNOdlL*bTJ<
zgboLi5SPc=Q{1&6++JVtd%yEje)^Li^Vt`lb2@xrS<bMmpu(`5hy(^AYp&jmt2g#x
z42*H)CfuR5A+IW>7LKPALx_wqaPBemwYfC6Y4by`8n0eo@x{Bh{GWdRpP=)=cYl0l
z{#aQay<ux&{PNpB<9_*$Kl+P*!8y;&HGA`|Aaf>Y=Jslv<mx3jLg4~liTQE2fU``j
zvD(7io7`57S__Zo12GI(9s8lT@_1f2EQN5lV@v}v!w{64>m6EWj_Zj)B}J&`#%jiz
z{U8y7L*@}b{WkMELDh!{O`KX=t#V$^R5fRVRZm3SPeP1|X?Kk_^@Fb(OIw(N0~y=l
zGXSj=7ORZofCS-mo~Z_!gt|$_tN(whfe@2}AkF33xzx*{S1lmb554n}vDmnhYk$4c
z`XjzaoImfFyH}8OPzC&6`;?mHP4r^YTviq630w4onZpq;wfw`>vLGbsM)HD7WoS=-
zbUz#LdvCSc!DxG_<-c_;_k-`VT2_0K%F-PHJVWu*S=Jj25eJ;DC#!luj$iN;lxa+;
z8o4aaDpeswA8M`zYBSbSy^d>hHeYZiq-t~^nc^(9E`_W8nyqW00ZVgM_F6ii=A+09
zn6rWZV~o7HyW;h$8(zPDg^E&h=>V29?GOl4VvvDI<a9bwYb7LsAQUOw++6eK^H==(
zn{P=maW#&dmYJn3UdP6z*{b5CscDyZ^X8WS;lKaCFqUhU|5eE6#(GrNgCRp9m5ipw
z^>oL6ca8M)?$X?IssY0gpv&H53%Kaup>61B2enG|etV|S$$-o?6Gc7jqA7I)utMOl
zE<BW(-OUXl1oC<DfGLM+-8Q8bvQ-Wzke0B|da|qHRQO9cmS1z4bFwu>ZeHyuR+!67
zw%P}&LJEPqNzs^?rajsOtp*XYWvT|ul-Bxy1ErKou}o<r38mD+l2;xMC&svAyqcWE
z5R}|pQ#fewGgJ(yDpN?r5}1mz)|Fg^T9l^lpT@;|Y%}KHCc&}=EV}f+|2qe-KCM0Z
zW&Q2f3xM^qez@h4M;DE{Nk^+pF3c<2=()9_1B8^^*}?SE_^SC7rnE*)0VJ?4m1+f(
zghogb(b`i4%_l@_^NCI|L(OG0hB%`OE);V2QGjVk{Pf)$-re00ty1#J>3n9l+kq(e
z$B$5q6l34P8&OAO7!U%r7>|!f&LS)y4=m@IT#Ex$5*en^EmAw9%r2&bMv4<r6Su=_
zuJn%OQCUt4-+X(||N84c<Bz`iUpUQ=tjrW~pQ}r%k!*Ia9#f4O=hF<ea&vX#%rRB4
zdP$!`g|)ZNMs+}~6p9rZtw&lLvsSW2VjPG;83u@linW=2+Oyk_{QReHP?W=*xlVx}
zN#y$N7mPQr_*Z}U$2@%eNGt~iO{5SIGg<<gk=J@LyezF?;s&tI8pqR-R*QRa?I)<l
z!?y=i0)tLq!jL93#D3`T)S{(SN(tl^$<|og!g!;EF|c-~Uonq-w=5oeUZIiQGcB2W
zDP$q%LS7rW2y2T38iRPdZ+$A}Oa=!f8=j>J^wDoWf|)m(YFjqYC$!1>Ft)SGylZfu
z7JI5?3|eKKj|3f&Y3H8@eprA$?JZDpB?d?VEN@0R0nx-TOc;f>Ebez#+vYMx?xVHK
zJilLG!@YNcic6I7&+lqXynAy8E%WjHw@kZ%-PIL5JkUxdTfuZ7#pGURYegfpDjbdr
zvlJGo45CbKZA?g!t6`wD$}u0Qtr4upxOX;7)WGll=pXT;w?CqOo2X~O7H5l|TjuNc
zUvq!H=PUaYPW7JCdhRBxc+-so(bgKRDy?dlCTtB_PcxE66T_2({NQc1vDVxVW1i+!
zYUAU_k7(PWKi&Jak4lV%Pzs;Ddd0Mx;NgTAF1wgU!dvCTdEsGRSk?ioay-to7`P1)
zh-}eLxwcDVOQ?tnni36juM+WT!5{--#`y@Ujrr`}Q*oI5HL3d<)e7hH$}%?|*OP-Q
zz}>5fVH%P1iJCKq$9v2IttED0M570P<XjO|YBP!mL}i*Hx5L2DgbyEAT3gUG(1QC4
z^>Fjvbbp_v0rY8buZPRz_Y+%!GZ|WEC`o6bh_h(K{C&RE@()wXRuyTU3b*-7ZIqym
z3Dp3^s5Gn=N{0YmYWZ(l%K|+K?@|Li?ZeVp`5Tw)lUO2-&)7E60NeKD_N>QfmAo7X
zQ$ohUhf&>7T?_~Rz*F!7#9$Z3Z|SUagiFur=IjlL9tx7h8C;Uh2ZHc@kU=3-9}cSu
zZ*H&oz2E&g*ZUo((*ar{rNG;_?<l46{=-LFbA^C*u*1x~u$=W{9FAwoT*=4GX+F`Y
zj4?39$j5bdzbD($JDrg<>B#T?>>u&Po6i~69Wp1{>;Tq0_kbSFe0Bdv%=LkP`~Hu)
zKYiqQT3q&sQ7jX4VTwC~DQZ2!$F_%A|JvAm;!@#e5G2y5-OJ5;cK0h2_ixoe2qVK#
z7(yedkW?6V6V_lU&CfSdS^Kdeua#8=VuWg)A?FY3THL(eaZE)tVCn|)Yss9JnbHT?
zDJ0a4(-Dw@omSL@&}~toNLL_K4yT3lIrH&!aKO`y)@}(Qa&>);#K`@U2{93eM77Gi
z&ZOu+E2WVJq?WPLu-rUJUKF7-#@YPAHcNz#MQ%N){RYdb@5N7Ste5o<WxW7cFYAX}
z^6anuG_aYNO6$xGym7G3TbH0%^JXRV_!y-dLuPaI-l9`Xs{s*}Zd_<S!E`xI>q#Yz
z&Prxf?M<>b@3t8tAVh^RMPA?DaJ`=h-B%#zLP|aX{P5Ultr8`8m)U&Mtm@64HCX1A
z)r_T7tjEb3E~yA9b~DYs+3PkD_e~_kJ!9CDWMVzR@v!jz{ypD3e9f;8f6mkKwKHF2
zYr{4It!!s2@|ua85oiAfmpz&TUZTEP!9^Zctx@~*w^#*(++3EVArNAZI#xe+Oet}-
z+wuA5?=TbQ_aB)EysDAUzxa~7cVF<UzxawgujsrmXe3C0V5ELPtfgIi(wa3l)Qth7
zu`CO%_~fRh0X3M<*%Kkh)YEsOAIPMql$eoAp%j;pmhK0HjA)D=(Qb`O2hIxAs<bMA
zAHYiW-O_50{p}I%wYelFYzGrsZxH;dZ@>Z;wjP;_iR?D-wI^ffPiA#*((OjG4VYH^
zl5O{)8MUsMMKpMrLffWf^k%zx18vsL$vXfB-ekp?yje1+wULWq>eAgyf4trE7si{p
z-;2165rIL4DMhaL6IkPPJg~cZOH9dYq%{v&h{0tE%@2p(NWrocinu(vgn)!dH8*n}
zy6i-~V{{zc`~BV6w(T^w)!5E7HX7SzW2>=kqp_V#Y&Ev^pL_kj&wAd^yqY;@U1wi=
zfA%O4m}S;04J>_Sh1}Zo8@Av!m?ci(dTV6B$@TGb_n6^E*}g_HsCOz#cnfSbfI)cP
zWXwT7f_6_kecYbsyRJj>>2}-_!v$PDS4qw19%q=jtct_#8Jza;wQ70S)Yj6ilTk_}
z!X_!2hBV_-QC9I8s-STo`?dWWzFEqIAW5q)%VV7l;5?RtS+g!Cn>H^Z1K=%);7h7I
z!ql#1tJGs36+28L^lf}2%)THKtw#hg6VOy#h2wIv0TsPD)@`|R%JZlzt0GN-CISxA
zSabj6i;yM*V@rR1U1#al3w*S@&*es8;ySd*?BDtcY8MEO{b_~KVy^)0J^jxCc2mAi
zEf8D;>}4o$E!|xYm1@;S=%DN83NW$m=w4~FZyd%v)BA#f#~|}mjJZ%cqSoul#Q@=N
z5Cg!TgoHD(czKx}04<)xOP`EjMYD+(CUrFM3IDR`>ZcPR_VvAou`d^~)8*Ef^(wUR
zL93$HxMAdW`wp;B-Zs>|R!?1AFtnmrUFmVEFmQ5k{Hie}y00H@sNwL!RewOMDy8S^
zua_y<Y#~(tGxh{ZCB5t|;e=$@Q5ou#;Q^Z+OX^p~mrD(9>U<`0eeUeXY>QPa+qexH
z#cYvRJoyh?yZ+D}tXh22?bk(ZFDm2)Ty?jq&wa|cN@|g!bLtmOY4N#11I3Ue;rF&Y
zrUcim9o>pTwu)A<@r{}QWCFg(q@OP-;~hu85f{&JfyB@nkR=?b;pN^I^Zse9x$LEP
z{Z>L43exM1sws}R<wvQMWMQ9^EKXyCJkQ;Af5-~ZtQG<9#K`%<fmvWZ{&OGI_uNkh
z+qyH*a$h-D9y<Og|2HoEE&izO=08(<{JtaUU?o2*flDnND0awU;l!kSMGc~u9xV{Z
z^!sf%ySfgg34?G&7;VB|1lEGevX6cY)TbRRX<}A?WTU^BsCb%z0twRMML<N{6MDzy
z+_ugR^IYg6JW&WMT<wWH<~FW101r_Es4#IBHz7}g$Y()nY+^hdB`zed2s3hE@Uvo0
z)#dy15nMoHm~{xVeb2$vJ#UpaaZly@DYO5W`_uMq@B1OiT|9}mIPldR?^_)A);ltI
z&4j^QiSKkuiLBMuj+_oEsdkYI#AR4p>yLl6Qmg96#Cgk7M9555Pd`~eCo{KK!|(NH
z@6bghzuoqx$`6mXqtwIwjJxHyBGj?t4DL6Z`7Ft%gi2Pn5D7KWAygz^_ydBrh&<8R
zJ`{y!O&IA+C%#F&t-V_Aqi9Jbb)9ffDEdrHU-R>}GBg#VL%%UhIx@l06Gkm>ap-Ni
zj?>b%an<-7relq7nV;Y;k(YSppjqtZ41=<uVSsY>*GZnb#n8RN^Gn321O6Lu=a<K9
zDcgkx63$nLTy53#u(D4St)3fRVxk*=KTBd9&5^6wnrC%%#{ecY{u^CgF1hf_OI&+L
z_mThSfbh|cLtcwt8B#(fgIu}#&Gye$=N!+q*#Ni5DmmJMejd<@b8y|=z03P4la;pF
zRMGR$yj@fxeHd+0`#1xjIekn~+-`M^DrJl~jcFTWHZ|#bBxEsNiXo@Jq09xDrU7bM
zEyvG0FXc~+)xC>iHE0y#IP%<GA3=HFTQ|6+G587!e-C7rGDqehlt}fv;?1(<-yt|w
z3xjrd@4R&GKkTsvqM&@JjhQ~t%?Cu4x+${7Q?isKkX=Z#qG=eRPfzq0^k2OnSona@
z$5l^pmaymg-T6#0hsw`M!fyy+wSH6F?JBqZXX}C{sA9G_%KGZBXx-dnZ3e1{39|VI
zViASmc7*0Mcbwe<24@X*<Kv1LyRz;5qs`Y^y|0;fbj||VD@er0f1Hao+SU{xv5@1~
zp=jshRq)({y<t)$;rmu07o=qqDsl0w$(Kh5SXho_%OO&NR{uWzr2~tre6Z=}9Q<iz
zf7R-?@~Woe#{(6a6*9iyr9e;(4_tiWP!UT6x#HvWCx$OJJrWY_-F5xPd>O$J@HC_y
z|Ej@*7O7uM%^RXw6VT9U`6q|9KJ?g^9`sL8VD)m!;bqElPtY-A8A^RVJKV9^ic+av
zY{-#$33@9JqSnrFWpuYm+b^tI?pp8upS}Kn2lgq7z2X)mXhmv<yEKDfLuaR&x^X~Z
zWn6SR>vY&<HB(&l+U>3DqX#_{pYr5+%O%mvg<a|CjE<^gUD|fB5bPYEdbdtNa!I>>
zFa0)uA5-Z$m<n@2-`TTtYYq)SfT%kq(TnjDh7duI;fErf^R7BP$z8$~P^#zk84Kk&
zu8Y}!cmp|m9h}4EZkMX9T~KmNxzl72I!Mg^1u#;W<Q}Z^Zb1ny3Y8>nO!-gbyE6Ux
zuyi~)W+FU4+&eI?=fKN<N>LKXdshkiX;}OH)RB)v?>bZ!NjB5<!Rb>lLsKO6v8|Ot
zOp|7y^8tsu4n>Rd(|B-jbE7XT>=5-d_|j3y#cYiZ1ooglJ?p||P<gB`$j<;`lb(l1
zo>@svrDy`QEG0@d2T`Sl^&|j4tSMrrdotaUa%B`VGzrYUN6@nh{F##dO%xK%;oM>_
z%o5au@O-}P^R_QDH|qgb_nH5H8h^l9BuUeUrw!UW*h5vHhxvvcps;D0167opnGwra
zI)tSz>rwX#WBH9i*6Luu>BVSDprQF=cvqaLUyKTs3XM}mMIL{Fqal}C6*go?^4g1a
z#HH{i2Prz`^c-Wcyz|~a&UL2xXez_=F2ItTJVnU0ej)_tU6})kzs9}wpjc&Xvh@kH
z2OV(t;pwrUYP8alZ#g@Kt`toc-9Vr}`AZ&S9H^%6!#C~r&lU%7-Yo4RcMgMI^vT_c
zg>g0mz(Y}!q1I=jT#{S{0=6NPdiDjWvNUL<Cen>and)bxu{WF7xIygiz)(Cb|69mK
zKCpcXq*Z+J=b~$dHnUB;XSTYuuO~XPN&5yoUH<gRsh`JPlSUI})NrN~w{4!FHpz|P
zw#E?t^|JEpWDhuNdLd@6J%WkMv$I<p449;r+pNjgL;j|eTYK&D?0Tm!l;Nhs=l*8Z
z@b-&-e8KmCE-kPG863D)wv;Vs+PCV?<B=g|EFD>SlDYnYwESjcZPnEAd^}50J|%bN
z^ubhHt`ya3RbNgKthF#|hu_}aW6X{8+5RifvDWA*j;-jW*QUBNS63QFFrk8`2&B&1
zxcL*JahGa#%X@KK2&!lF$Le)auYA+VNK4s&nG9c$+REL&^ZQH<HF>J5oVf(^?67D#
zPW2SiE%$nfr4cE<V^zpu<Tgc^aDa}!Kf{0pg#rs2f2v1{;;I>QBSsJWjhF0qr5GH&
ze9ZKzQO23)K~ExZ3a>aXC6bAD6Cg8^d-8hZ38ay>NRml!bO8jb%6-X$D~K-bl1k80
zc|=*==e9E<@TN`dQieEq5|;_FDrM4rA01kVoL?0?MZt`j2Dd~4q)~-@X5+|w0G>{+
z7N~kQc6&Wbg*U^qzMU_WnRr;gyLwCcVr>0QmeV_v`~Fqe{NAlJ1f%9_M@|F^7N&;R
z={~-JE~cN$w~u4*NBK`qAd1;)X&t5~P%1z-)5?qW$tV*R@(VNf)T8l6SqmeAX15qs
z#&O5med=Nfsdc$^(V8&sh1|pyJ7MA%+L@oLE>lwl%5wp=M0RmE3C*z8&e3VqCeOWh
zw=3$p_qk?e<3(k#l4d8j?vu4^vCqJuj&E@rO-h7eUk4$8*?C~%n6@94`tc^plwXX<
z65*h^elf#@*P+*M$Iox(<Aoyp@$dL*;Ln!&PM6^#f<uFvUAOYq*ik0-?dteHf~NP!
z#X-1z<g(W}9bN{-d-62?Ih-|5T3tPzCWsh!Rz^=Zfw;aRmSpIugGPQ54r5i;pym0c
z;Kh&W_5`SGOR4E048h!!YuVZf;9z$+XC0L9BA|SVkWQK6I0X<T?j;(@`f4$Y_ZpOo
zBIYtPErfgpAL{`3xU!;I%^CE{ZrJI^MD#0-CE!Y<$k)HNWAR4w?Dqc30J#12f7pRv
zKbDXof06q#$q-e0705FvD+h5{>(k!hC7?x)ft689NXm!cCKpktNG8gk?+psSNPVS;
zgK+<cy|2fN&Ltjka*h;{J#e0YsTK6dQBsIPMMR^57^ukMsdb|L13FKc6jn;`C{7dJ
znQSGRA}>fgO#2?%2eAPV)WNGjv04zXR7K}KH!z++wF9{aLVG?sJ?5<G*sfz=%fNSF
z|K=~r$58J<HcGbVJ@pSrN^;1uP*Ml}&4xIwu0KNiUffR21iUw+x=Qb$26b+Z@kUL+
zA7$mKn0s3cX19~9r*r8qB7Ox*XY&l5y|yZC1YIgJctn+F|E~rAanOmn`h2%Qe2a8z
zKEB>fxbl0knXari-QT@QDC_@v-u^V<KsxY?phJ>%9?)F+rz$0dG`ACzW<-3E<I7ef
za)#txCmA8cIet7I#Twxq&GQrC(c5SRYIrk3CJ%GeHh*!_$gQ<vRKzC7)FlO-Zpe=K
zyqBS}F6u2no~*!;GJ-`F&Ae`-8^#F+iIDU{Z}CvMsUEo%sl{z!OUcM~%oN8kj^<I)
zMo&KjN&(Gyr>=%@##xB~tB5)|Q2N%&2-D4K#BYVQ&cyH__kH^6&co}#;A^aR$B+hF
zlWW=FOs~ULMw!JVeo*K_CNTotYD~w_z~?2(F|DnG2v5~ia3%!zqcSkL|J}nh!@_jR
z@^_hMM;y2Yn*xHIpbPcpN>Umm5zVk8B~bP+*isk?2PY&4lF!R)X;<efI*<s(X7_rR
zp{M(c^XnGpi#_h}Z?NF3HY>bttkS&F{q7r$x8WEbQa4#EGC9|)k*e>l-4XRGT|Paf
z433CldguydJ(!P>qO!%Z&*u@Rz30>k$K6uvwxf5qxa9`PC9g0O;Wo+hqsvVGbd{}W
z^X-sDu8HR%bI*%Nx?!|nLdr!C3s+r%=7U9AISi$8o$JYtIAutfwlQoVZZd$Pm}-7;
z)yG&@0U~$1+o*NDe)~cubM#m&B|2ps6Q%%qoTZ>o`z(Vr0TtW|Uu4nuQDW}s&Sd~s
z)s#>wY#09*Sz4$$3nFl;fJ}76Nw=ffkrAz3Kueg4n#yCq;wbO)ZXlHb{6o1tHHsV&
zB)tbOvVHKU2+daHP;1)+1JG=WSka>rfqfs2|2+=XC%`%pDWSN3OZ_`Z`t-!~n!lu}
zgX|){VTrTAP4w5FDr-kKHpGWkEOzkv>J!XY4x%cc0{^pH+Rvn*i^LC|h5Ku#7j*1G
z!UrHgLg^PFL(JX^`jjDOZ9U`Xy&R7;`^g|K^)SnP5X4~Z_9n{(26wGn5$34p1HW7d
zE_o*hd2>OrXPJ02F^I>nGy2`cSRm71Ois^7c_YNLniL8l3OT{+KEH4$&mua3`O0r{
zeKW0TWOP?ILX_t)v=(S^lZMi}r_}obB`1`)K5zCn)K<Nfm*&@`w2gz%cfV1O#@+y3
zg5ZV+KDXtn(_bv;5E%GAOsLi(k=e(S@gbxTbdV*Oyo3m6UPl*M?(YXQ!e-fS*)!}q
z;K%I_LK<7rd7RJ7Uwag0LWu<+!%Hj03<#C-*#!)+CTXb*_P*;P&r;H|O2?|InZ;Ly
zLP}U6mmvxm|1pZyuZWw;7RyqUaB@M^c@p*EGQ=a#B-0{qn}OOaeEbe5#l;Jkz)vjn
zoSH>Nb97*Xc6~JGZ#a^L<(B|&=6O?N9_X;ZwYBJJ*6(tKIPEy7^TjuNZ}^P;|Kb6D
z0m~CEHYRHx`@q*QY4%iA%j$a=z>^N!kcp2Ec%)lyjCFD`sL+j!@^OC<B=EvywG03G
zsB_FPe>7LwsHK-Bmp$Fm0E@|k3o{}`O2DL`dy!}d_Mdnd^XAgsMX6KZ4hmVb2yhNn
zDA8y3`RH%j6q<5G27Bx!A%zmteM)F5ktGhJ*~HvA9eoVA3*;bAUPzcfK<3%<bEd5+
z7*Sf8=DMBl58|Q+OG3lRYt@$RhShHS52#uQXOl`lVRSxI-N#D3&OCe_Jyy`?L-?-I
zErPhq_}_FDTe@-RhIjWNKrk}@wqqZjyXWgsUIIDJI$6Y%TkmDS$!f=?%B;>uiwu3y
z!v0AD&{s`W#eT7h*HU$=v;nf$K8kFt7;%9ov43#tA!m~(ez?z<lMmMLsGJ`NNQ>&o
zO?8>6TdOc|R7m&V2`QoE{&x!$lFG%#p-3;~k9xe%z4-ic*Brulzbj`vdFTwBdE{uh
zJgN~Z_7<;N1xvApl*w5%L9_Z0DXoi%trpRN_nbx~t5br|T9D8|aM&B;FVI6?4?_LX
zZPzbCjt-YjkDtQ&V;HxSCVHEj6;d_&z_a9gQ1y`2g$oIuZl*P?CZV?OvKfW$p1ws2
zZXu0+CimDcM%kOOyLfc*bZItL<_t~Ep3i5jE0k@=QItupg(AZiQDWqC?{_iEv7KMw
zNa|TurfFSV-kIt1wYg)VtfWVzsadal0Uco4D(_z9JQQbaf4H8W;P9U-LH*aHT^*a^
zu#;qU*_w(^G(CR)GvOl?)){(;A!Uo?AosTzDe$#>*&G+xg)NaB;Em5~WNt$JY8ZS(
z%&>k}k7WS>#oGRzdP#hD6uiU>OCUqaDYI6>(EJiyR~aDEp+qG52aEiUbF5p_9Kbv3
zBa>t+*j{f6oj+0RSS;p$^nR4_h^KxN39A2ix>KLfYn4nZ;wpTE@bfJ{Bq>T|#>w*+
zl$p+5YH+{{gN%fn0}WgJi?`TS4IeK~FU*Y9N>z!>Pjdbr>+G_xIlshCzLg*Nh*q08
zU<+Ws;(UmfC<~6U%!EE<%M_f?{{{W+J*y);hzh5%?@5cQ!w1jQC}b#PR7oW17vgc%
zV17)}Qn42Wk|Al^9&nmWi0QE3a|}{mhK6{os5MLKYvQLgPcD)#g0n)CcsxEaXI(%s
z#Sp6+M|qbx{LtCWW)G3NOoQq1d!{0p-gcr#S!_`r^LWSjMYy!x#SCeI)Z9^yo3q@Y
zoBp=mn_OcGTL9_U-o&ozEuySBxiP_ffAG_K|Jq|`|3i=)cA`VolF{sN*ToIB9&aE$
zeNQ7zgj@aNxaINOeC1QoC*pJlQ+#R6PJiTdX2P7q@h>;I{uz|@r5}K!riOYD+$0ls
zUOEcRb+!MHIvE-BZQJIm`YmIOPhw*=CkqiHWM1H8eKPl3yQxIyjOLy9_xE9b_8#{a
z=9z1y>TJ-!W%oa6yqZjN4rp1;-O)#4*DPjd2q_zU=pxoske|<rYIZGC_z*&PvB2(5
zaNyuy1WeJg*LP-Dd_TWqe&*T&Et4WO^D%QTPl4u<Y|$7p=nxQO)Rgp&5rZnYrIC*C
z@^6tJTfQwvFO&NNf)hshhe4SzoydYrAcLf5KGScS?WN(}f9RvlSi^e8B$CF;#)yL&
zbnBS;Btq^3pj46l=vn`AY3*}k?Pv0Uk0H?vN44Pks=`$;T_3r<TdTa)&eCTpB}5^V
zDrZ07`NaN<>#dXF66I+lS9tb*2g!K-z+O3Tg1M_*?W{;2P6Az&oF5POpP_=9qO&<{
z*d=ChKG37Pms;oG`OF)npVSfy@vM}AOmCb$9(Q4#m4bBiQ~2So)WwmGvK+~(tFZ)r
z+URNGO#t13JP}nfynML!KUP!PPOeD)xB>uvZkOw@!^YNHyl@84_cI_2Xak=q*EzI-
z{fPv3J_nDdsac@Zy>?7<wrB{C#FdI98QMxh1Q+R#vhYDw@KrWCEH}U`DQP~5HB-yQ
z%x9PHV=ivUp;wIJ{uNE1Vg+M!LZxBFA~2m~h7u{7%rPaI6e3^I@;A;-4QE-UO10Cw
zN5Aah<r1eB>&;N5mUM7!&*#5zVBV(48WZ${Np||)Ak#&p1}qW>)gI~!QjRs|V8rP{
zuNC1c3dG2jkZLLvLnv(YTJCpR;{-!clw;KF(7~sO!{j?V8J#2Ix!X7@xg67UN!iCd
zZBcSt#$&Y84_d7*VIe00;aiyA=i}_#gd`$MJ=M7r-(P6hvc|~?Sq{fD`HlrmQnLJa
z7yHY<=IPE73od*tri!rv{Do>a_oOkD<Zx#5JEWqw!FlVE90yM6aPN)&k4i5~wY`TA
z56J4g<;qTGdJJ*xEUm-BFSn%|dhPx)^}{Cu?68)2R@H)?xiOlW>Y&D>whVpqDI{xg
zmu~`HfzW2v>4+g@D%c<z5n<e*q0dX<G9+|~L+Fx7C6njLc)@256id~iP(Idty7a!p
zMteB44OQ!Z6F8<jA}blbB_vRnr!{3flDeN~c;b!q&+Bzj@cE4Fo)!8eRrPg}cC~lJ
zeQ<1Z%UmGs(x`?ddRa58FfnMjf+;tXLM56^l4CZUnmuJUy=zi;-L4GzFMeH~Dn$}>
zd;aUKE27o^Q{B29A|G4{0vXCd?L#+)9;`BEe^)wL1U|C6jNlmONf^8XCUw~{-|5+*
z=x}oBBIggZ-{Z?MY=ZE|X;1k1>QF&pxq$|&qkR_->-b<91Q|UU-SYk@3uFOqvj#A`
zef^@nyS=z7!zXhrE>62#8J`u>I2yOlTOz=^>P+4)bit{BgKGKjlAkfY+EX2SQAc(g
zToGpbrk*2_{ptkKaiL?C1@TL&0ZH#3%I9freR4{(c@e}ML3%mP*OskLum1*5_b~D!
zSD{R;Neft|{j9OCh=M68>RUh~r|?AfGy)xdbu8>KDq&Ho`TGg;XS9L;^Is*uw^r8R
z@R!lnCz6-nz3-U|=S~}6r*|_R-XEz((pA7&T+t}G5{2B&Wj!M@u|;V_)S!Cv@G}V*
zR7cW!dNbRiWM!f<hN|O!K1PX9<{pnw)JQfDqO_gm3da{;d@+~>AKX92d+08vN=9HV
z4hW7{JH;H+?Wz^ZLp)O!@{2R@zKVJH%zqww?Ynohz8wCCjXC6dr0EEQphNCVgSIIy
z1+7CPB9yFTMZOpCk%l0MF;A6N3~5e>%U1tZS|-;;;h!GZ!*p}q7d}afm33w=y3E`W
zG&$?I$nh#_%aw@c*b$f^M2X8*f>A|)#2CgTg6wDd2E@gb6VfRWtt_6PU?`^L?F`Hx
z9_`S?zLWkZqBQ$^9Dm2bXXf29d%V!wI=Md;5rYwx(NVC+I7;h0n4RG?6RjMRZgR?G
zYUb|ep3v=O<8t)x+o=IKmRz2!yX+rRy@35Or~;)rq=E-GU6F#UW39$Z))YM;UH@67
z;2TF>(Ws}MA?#w5y@;T~!Nl3}#`%*x89DA<moYZ%9BuW}ZGn))4_I0gSVahX)L=Q-
zNrB<cU*Vxgb?T&O#R(FY57DywM_#5;5Na+W<!EF$x#UAaCrc4BMz+dBwNmV*`fpp?
z15k^weZ@%BQ9mM#L*i$-l(ncNPKLXokG7ERUGRIaeD5?(&!~Kit#XIC&JwlSnxM?v
zd_>2J>(&3Y$5x#ty_kZ@iewd}D#r*p4p8x_E4At(BP<3TVj&GPDW1cmJ%zN0U0Qj=
zuO-Q~-gnCd0})v*rg-H<-%)ADd18}wJp^Ww!y&3NthhbdR6@*&UE&w71SgP7CCQty
zfM$|es+4TeaFn3jT|^}9W$v=NG?^3or6s5H=#$UGrxxKfO7?{WFh!x{_czz1Jg6X0
z**S<D72A0qwb^FDt;eeC8#Xxk9HYFpK{IjmO9rioJ0i^6+)@eIIW2qT!zsIZHxT2r
z7UzpF_&%G2ee3*bvGeEUqc_Yp3d*J6!VUoAjkF7dl|jM)#H4QeOH@jQ7)a)&K^Tg&
zf&fEZoT$YW209XDreDhZEF|C2ipb|jHPP6^3?-}(tJI0gQ!T68W!%*z3bNi)$3f^e
zj`-o7fgn_-sfK$@dN?uZgk)&hsj7BkdBYS{Vk^$PL_&e<-)A!qazysc>%M{#e9vDe
zRo0)soOBu|$hjlz=wg}yO1HM4w<5%HgEdZ{YsE7)%wxCl-T6I)ar}Q1*z#>7YGMpa
zgKzZO*=i1J*_WKY{s@sUR@$xow@m=O+>OMEWMO2xjxd<D#P>W(3$@2Ou!L_|)^>~|
zP*zY3cu2!}33N#EpaEIy2uj{siP~rSb`HT|SKsPV3g=`pLr6K672Sk<XmI}{&^0HL
z7X#IewLLDiqOl{846^CT&-ygSrZ{0j);3a-a&R!7UIsHsc39eR=kYXlRJu}@j(wuz
z`Z&&i?BkY5E`v^!^(0&2F7UNDhl*ufBsv(PEt3`g&gV*5(<a@{Rlc{J52;LeTMV|_
zKTXc~fC{Zn6Hx8)MuWNUl^K7X`Lyjbv25T!;lABH5-_qbxyy66`zgYd0*t<1{Yg^d
zK7$YgNBI<AGSPDAbUh2=#`8h^mh9KHtFG_WFU(N>N!k<v5Mwk%0-lE9R4|Pws8>M4
z#C4M7|LC;~!oUV@L>SkUS%oc1*4XJPUHsyG&oKU>Q&Gpk;&{#0Iy}qQw)wS^qu{rV
zpC$kEy4zztNjQcc#Nk~2pCGn`M5@Bg%+dA29jqJLjLTp*8KEHwXDKJ*Ap9Ey_|K<#
zjj+cr)|`eh(aW1g*vEZ2zay9)hy|JPH3~e=pu|!nFTQDFJ$0hmSlY~Hv7DfXqy8y(
z=F9c)X*+iCpE)~8O_p>samuB(wjT_|#yTOWuo?h3ZkMdl>gYI|;lpr9PN!y6*NxE`
zM{6r%bM^>!9fMp?v*7czhYfFe*fLzV);<GMmsrW{KzQ3fbjDR)CB~-gipMZvt_Pbq
z_A9jzy`XtHC_(kFHFwO$H%3SC#F$$G39ms{nibf=gKsfXNm|xigWxjc_I5RO@Sf@E
za@2YXjB+MI%ROY!elB;wQd8D=8>1YHXldySa;=+^<easGJVUxI)GJj=C?x`VT$_vP
zZn(Z|R_wQstu?sCwoYO^F=L$k0#kA`9ZOd#O7S34viU*nYLvJFBXw>Xh0n|&L`W>v
z(KLkl0yHRN!r4qZs(97*BH8wg<0sD#>LUX0JL;i61|htBKijNmS9V8EF^h{saJic?
zTeZS6H08`6w0}}a;hH5do+1~H=0mAtURcy{(n%1kl(&36r<UqW@TH9HaL+=?$fvS!
zlcp$QelsrebCE^*FCC!=ITB_FZ#jE-hzJqG7CW_eIt_~jY#erNsnIKZSiq3a<4-t=
zWKDRSf|O9>C`7u1?BCK@7U3Jw-l+rQFYgHkes2~6(<qat^0G;O%0HOGki<w&|DIJY
z>kbLlE)YB}aB3|e+4pAS*5r*P47?|72tDuiOnBzVZy?zs8`}0Eh!%Wd37?mjhpUyT
zeQ@{k)>7^_X4ybv*_ALTPV|q&k<7sx2l`$Q^krF%R60tPAx(sGT8KFoYIJT8{Do2u
zW|ahR0#qQ=!Vusb{ubspwJhDFifD~i7tF6nQK-t;N>cJo*z1)~b?Zrll!O{*nB~+v
z8YBAcAc<~fO?SNW&}M?OQiT~6EkaDJ1)$L$ko)~J`*$nXm)<uXnmqS|Oxk${*B=^g
zO^67!Qpn2ob$1yCET}GZn3o$Lv<CFw0dTPkpL4_Da`i%N23X(spTHAD|HlhH1BA6h
z(EI9*(#>}ZbCW(TlpS`+Xbn+iFLlyGn$e{=b19i7Bbll||CUCwj47jDB%dWj<VZv1
za{g9XEh!|5AJXZL(aF=WeW8s=$^WkrsWewa_(`GQX0WL~CqAalOdK`sm>Q;TdIY01
zT*{8GtT{`PD@9MV4BCX`x#mZ?%eP9++YZ%rX^I5Zf|lVtS=)Z=VN@x*r`m`+>4|>c
z8NyU4znhC(r;0d6Mx~j;4p$z6EVg*u)e(Y{zCg?Vp!!=|UNVrZV|=7~p!n<1IA<Fy
z4VR)Gy2;Dl8%(!|V7c`)9}?U;CTlYX6l>0X-$QwR{qR<GDZie_Sd!yb(}1(AM8s#R
zT1(H4(H>U|NC_5<=VPF^mKY2lAeU4Z7r$kPi?QyX$86rQ4i*~JsgQGjGKRF0fH#gT
zh&q(F!V1!@(W_)R7DI#>OI{SILtc}yNa!?OTd;?)!AhmZO^`ITjj3mF|A+4M?@gi?
zNYruadcMq)-ngOv{|op!-22!M;i(Hgn=Ia3>HM}^gWpjEQ;X{DnPlMSvhruMV;MF*
zK&@Fe;<AN@lU`f+d4s~QwSKwp7QH)JBf?jbtNn|D1(G3s(JJs{h?de4Ckn<6lIpO5
zn(!kF@_O`h;O_rLPH#s~zk3$jjGMrsQrL#%jX$JoH<ACQ+51Br7fpsv)o{S>ETm%7
z$AP+sRT@6bz3n%0awE~`lOg-;q73)_=H-&+{rU9RuD^%F=_eYU{J>60ybuCa6sy0_
zP7kL3*ChcXq$E*Uwa9~MA}~Z(V5nHS_{A`#OjiBJZ;?Q1St``uB4{u*VS~7mjI@+g
ztwEe7@)TMf`*uDPPNUgdx#deA)km&JJ<}WdoAXZQyo4%v^=ARrl%Ww3!BTR13FwjS
zG?-9k+=}O6gDma&pKs0-I7T@XeF3VgNEljgM9n#tRhAiot}K$=vaMSOoMw81X4#2w
zRT4i@SA#`BW@9>XJ>v@WIoBeYVr|!sTv!b@jcD>na=*IoQ=5>+_F=(%xPVS0zm(Ib
zOn5-6%9H;KgZ(@LE2B%K%3X;qm$pYDE2be5Rt!T@hpD>;ge@W`4?;Buqq^pUv~@X)
zC6yUj*Aqwi_E+_G@%xLx?L)ZKpg`ZqBA%=0CI&SwWhkZ*A<Z6-a(F;=5F#=clg&|=
z`82O{D?tWoNpi%8GA1gwOlhE*3nC(!qrl6(d+BKo-;xf++Iya+LYEc0NiCf9-c9oJ
z>#~8P|JUfpZNe{xl*v@SHKh~X6NX`rq0P%wOmY(p(=oz0gN0$Kwfnzue(&C0?S4Zb
zkAu4hPxtYRpX=bYm3ie)dUHuSMzF~yVW>Qrl3gQ6PCiiOR6ZsQ9mikNFjY9}F3)BQ
zf2?*qdn;)B?VdtYikV>msNK|!%%~YX$@@4N1SsggS+V3q?f)hS#xxiiPfng8EiSQo
zLG2Q8ZrYD8fEbrden1rnJ#8+JFGH6|HDAPHhe)OZ5T-l?mKFa!7pc$)O+|IXTy;(r
z%nwWnYMNRfSQW2_79+@0AY2|CXds>Ql{v-NS)Lpih|J0o9I$gL^dTShQt>lVUP3t~
zkJ9!&3cfe^r2k3V{(Ogp76nxWU)hc@>9760^C=RukE1b6OdoTs9~0gROUUo+L>n}>
zE!(V$RvgV2uf+orG<m`UR)+~xE6lSO2)>Ay*aAx5fT~80*iB2<pwA)NK#fg9(rM&<
zUf8q)-0Y_=C@>u!+fk#)FF-fs{4Vy%jugh|Y<KzKM>!u4pf8*dkfAPfD){@G_!(Uq
z`1h(R)w16OI;%rsIw6>$GSHuKy${Zyo<{$%{SS(Ex}Bbyz71|giTaWpXuxJ6owY-#
zm};$=W6N@fCUKlp>Y;okL+kP~NSqT2L$2WtdJ)6`$3$bKD(G9=02*`t_-(9HzG^Fj
zdS|*wZ3)2;6Bllf6nFLVFjC9I1cQ9R4-eIv<E{aY6RSjOxi|HR^wxAd^&GbO&;0O>
zpDH>=_)S*lHjq@#V&CrhZaDFhE<hpTG<af-Eo5;)oO^VARi`4VZ26XlT1@`^GEQiJ
z>Oq3tOox5j&qIWhHH63tA1^F$-H3dTAc?%lNX&&=3kEw_@@X+?^uo2js6I7RnT#`Z
zGXgPCZzZr#bi<3v{0WZ|79A5VaEW{h-^AKdiR?jrLW~Zt5^E7@v6zlZi&%-F9f=XB
z?H)>&d4;3Eq6#Af5$<^yk3uMGj#Y?SzR5?;O>rEGN`oT97Or5UR2Vw_xPQy~Y9JiL
z1{(WD{x|k<Vo79An|uoNLPle<!=QUns1RVXmNvM5|1yuPsOB9gW@xP7EL%c*U$bm>
z5~zvR^4{`$%&fHUadjx1iQrP(w?w{`@HyQD*e-`vc7)imyXmE2RSGR-ikXu5LRj{j
zSx0xQ`lvlEpkfRkz=Y_A0NFfsXxlrOwZ12363y0qL)STCC)v(rVe83wZN7UV#{0EL
z-F@`?BhFJ`!&yHrZ>s>&(M9BMG9@b3xBz==Pz%4Ry{3<ZJ~N7&QmjqWjy~9Q_~pTg
zUCoMKxj=Jy$K^IYB%XlmkG$rfj09sb6Y|Z?y1|r9{IQ|YA)YP&<ND4Ec&n7Is-m&d
z^7%tlJM^tGO-lv1bxdknaYEdMzY3_9k}Z>-p<UK#2b=}aVVJz=b<~TQ<>ZSkhS;nO
zmrurG<;aGjKZ+XAmIOm5d^=U6rK6?QpVVO!rWz%Rf{Te<Y|$+_tP%bVskmk((jj$*
z;JG-ZeYGmV^xl@!95qiu+s(%qdsgg!KK^cl(N)Q>n8(CF!nAjegQ+dvgGr>0z?=-Z
zD`&ZY5%TsOSz7Luv>phC=+LpZPljrzRD*U!Nymg{gyj?=6wqH6?AXX<0tw9$s^^GV
zr^EWwTh1ry@Shuzd#Xibv_D=27i{{b0(}9t=@Zx$4%q}G$WJ~`&);MEsdh0s!qb;e
zai4UDw7T#O&@{>>z!*XWW*`|PCZsD7$!=edQr^JPtyJ}n#KZx@CDIQ^`q%MbV1uGq
z7%1_U<iw0B48<Iou!${@mBW6rMXV0omntzE-NfZ`6eI}YFO!~nv_nZz9JJ^@nfkIO
zIs{Y6vuA%%Ig0%9tcVPslJnaL+UAHjOl;Xbp&0Dwb-&GtDCfI`>0CCev8+*n_oxB}
zqONGn1<wb_V<f0mCTBI@1j{!q%Giq;pPO1?8<UKDS&Rb#T1PLVOG4zCuQM0X$$opB
zC%2ZX#M!Sho`%iaWoAjUW)yz^sz0y4P|D0;po;CB3PLq8)Tx@lk&gDqJveVx*FmpP
zn`U_w_<E+pt1Rb}?jH?1zmOfD$Lo3xnm5kcn~o3>|GB;rG$WR26h(!XsMkHV)|!3&
zyHH)8;v$vS0|JDRj)se4>$j4e;OH$M2QBYe2$g)Mt5t(%IqHhxE1`<YpMZdR{Sj4&
zdaNS2Gu?=VziG90&z1JQ4}OFw=>M{5EjDatVvz$7E`OVC+6|~Z3k23SyUCmnN7GTc
z(Xp2Gx*d=6o{d6Ego<LRnW1u*DIHA`R1#sHlt&|1d*-S*9r6e}2w?RH`V7$JY9DU=
ztcCQrD)@6li!Q_Gz^U(;Qq#V=*+5d;hF|OX;AOmm5I4b{ef0O_ee>|EH$bbcVwZgT
zEEvZEb@&2$Q)uw_OLyY|DHKYrhr9Pwa<Xd5LRUk!W@7C5Jm<dO2un;f8IMV$j)gZ8
zX`pIWV}grGQXfY&2Ym5V$kDRmLV#jqB?sA5*+hnGsHV3;k6^BbWk`>3+j|{Xag!`l
z3gAx+b+b2Cma5!g6DB2=bPP+IK-THEv{Js;(&mHleNS_%W|~0yn7`Pw$#Uj(H)FVd
zXQgju8)0(5nV9ZT1E6rg|86dgGt)H-l&oUxBl6@^8EPCo_WR*RSB+e548;pz&sZC<
z7k`n}5H1WeA#V}2tK8d%@*t%WceMG%rjBXV3z+rGXpWWMO8IFD9<^Y;;cUTu|8Elk
z9*iiJT9lG_CEqg^hg)VZO$2I6XJvwJv=#bgT-+o|oOHQ1YY{B~!n`guWnsi#Csp#!
zbM>&9F7oW&q~<=8c^+q@Sg?<+{(DhNysAWSLt0D)^Wcs_>B}Hpu7^8czR5u&w>|F^
z%_WJD7l^w^sf<;>XDGHljQz7gSt25Wju)p2OUTFBd|7{PtLrjsKBj?5yOP)usc9b(
zs-MSk97b^>LxQZ_==-BNxrv(5s1O7NeFYYu75~40Q;7!2<h{gM)YU8ZvcJ6aPF$%&
z5oGz%2jHc}%BaG1TvR{JcyngWL)N_D3t7&yE&azYJ<)>X)a7gkjf|rbHtm#P6O^Tb
z>{xpxH?bHcx4V8fV3T23dX@+YUB51U=ho61)y@5fziwv@w+4&f&1(6Ql<^3TgP~|J
zJdf|(a<$C*@zdJgu$Tse^XZog_-j>cRv@BKpgb!&R<C;TTq=9y@B^;rWK1(21jfKu
zZnv;aETzn?Kp#nCL`5bZlY1NmMPmhjr&TG)|Mcb`fmh&*x29*Ec~ZhxLO%m5{4G7B
zA`13SsICl*7)>WhEu)8s7z$1d^&1RgNku<7<K)+XH(3Yo7&$RcxH`i;d^3+Y`aCYw
zCLN8eEQgFPiQ5Ju1ic<<Kazcnb+d&q{tp>$kk@!29j_$Ngw7`&=V(8nfhj$X(rvE&
z((&!{6J_J;5{1zWtegmLPgLe&0YnX&!dY0>jWb{^Vek>0^MPWZj5f-ZUMq7uc<OzD
znJ#rF?Ab~RsDh9zRduhJNk!=VDIC(O9jA`A@V#(gGh;a=oBnttV-vJHp5pAOX;Q^?
zB4x3!btld63?;w$T-xhF*1-)YB%h}o?iv_d<opL)&i#$h%<+c+>l&_}t%f99Awz9R
z-eIC0Pi~BRWw>UH1vLRKzK3kOmQc1uQ4mvD3Wg@}YD`8p`o3L1ly&*;=Nf#eE7cSN
zomr+mQCvu>ws)q+QA5L8|Gv{+TQ=RA-jjr~yC%DArHV^h{n;Uu#x!bwyA3<G3e)FF
z?%cB|W9zW)N5P+hueAE?j#x#^KUK``R`SyLxEu*7vz4BahyLP>3!8mxGR7oZWwye<
zwQDH!Y25a61WcWXDN=F1cv90slQE7|4D)U3bv_t|*w*ng_Fw!ZXJbm$fqd2^fS5hM
zGmft|p%uz?(%VIOOj>lfmjoUlS;G9^c(H4K%4z(bY18ZhJ5#-AAiOpy)DJc8Nw~s1
z#Ar`bX^dtr%h(E1UiJf`CP_NEQJbPcKVcSXFa{jw#NVHJQgh@8@ffw2HF&9585jzA
zObDLYxIb7UL69%Dk^RWt+^@)|Cz|QI@ilq{fV~jUa};3kD`u-+&U?60emb&S{-1UB
zQEOe++~y?r_B+hwBlo-T=h3o2w%yD2*RIeO!JyDA5-vGelz+V9n-rEppOdBLjk$mT
zA>Klu)y6(Ad^>&Yat8Bdjfb!M5YMI|)B9%aS1GMTZuH3Epe{}wEwLw*Ovb=cU@{4B
z+PIMmG*UGmW*0QR+X!Zn=~%SU>fo6bVw##w_K}0KQYHkMxom}@9?a9lNklagzeZQg
zY>C4wF`F83AzeK*{n@}ighomt`#HM}@n3x7VQqqHJces!1pAqoMl5*XayMKUgRJ+?
zN4S(KG002)qZ9J%5fLNTnXosy_g5EYo&InsK{tep)$wad@j=$*PGjzcu+vM){S?`n
z962#TeF!%2hhLXFJSktKtsb66d-s~Nz*JSSQ0>nl?+*3r@Aqkmx;yW(JT!YT*{ZOu
zl`s{7w&F1baX%?h(CP10US-PfPE$OmMENPt$8HoKlU4>(o0S2*&wgVre-eCq_W3q%
ztltwwjT^T&A|+ILqUx;N;pk){B1b-_Iv7nsrCk!XoLeTKBsz9WT0hy$rvSm@_a5Zm
zH{(9~v=8_)AXVdH9xf7Pzy<-fJ&Z}xw)o8Uhx;_4tZgzt$EMh*{z@9>zI!tAIsi>(
z(~k$9uws0cUYs>?@%NuL9hc&t{v~GsuepXqq=vrqVMQ&^OJj6!Fko(+v<qSqwfa02
zuSz7joNNMNaM)$IR4c4A5U@W>HdnocN}$C33|EhqW%uTx{~0m@MMvsK=O{_I9yfeL
zWwmull2st*BHhY^eIgy^XPUY-<mwEtjplW9Z?^E$r9DgXlNp`47e9l#2Zm0y#uh>G
z#bLK>=oEDl6pt?QYb<vj#Xi|IeRgMyE&gH_8^V)?Vn6L5A6P-lg1b6fZNOH%>-g?W
z6f6bX+L1`*oMPWbV8H$_R-!<Apt@i$;ohdT4ckJ9L?v`H3k)kgEWz4o&Jm4?G|2Dk
zqX4PiYR&eEBpXOy%;#F&bW~<8Pf&1i>%1;yps>%fnRJkSM{O1nk$5MaBvFZU<t6z`
z`Wv_cS&g1DXBqlx3M*at&&r=MJX=kjEp09|y4Q*bsziE7%&}Tht|T)c{^$ahSp9H2
z_TzVs3RF~XC#gSt*p4=e&9rV36$G<LNz(J(XU-DKg<nu)9mhnQsQ&jadT#>5xzT3c
z)O+1A!gv#ZxWom1o!o(QLE4{Yp66#gp-+6d8@Ee2_m<cyyz2}(aawM|QWYZ<yuCCr
z0D02-dw$qq&^9X~-1TZQj%<77Jozxy`y^!kvn08Rt+ID<d^i^F%vXKg%KCiZ!RAt_
z6GF)zt@b9|CPXG&k8D0pSts=MgMk1`?wHtE=)#6=JvtIKS&|?YgK&N0Ev1Oq%=y<b
zcvu0uA79oj&8Jlepqj`jN<bkG%c@J2)rrjpVmRyVPhN9?DRVa<Mmrg%)vTP6!5_)~
z8_xJq@1hYU@>9z)W|K#CjR}5ki@f&nV$0R#EY(0*iL>>0M5$zX=o)?jUVcJ&bPGOI
z4jDdW=O(grG`vm4q^(&wyUoQJgi#G8xF;dn6pr;DU1J_X{0C#7WR=^)MwOK8<V{Q`
z3OXGQce90yDBh(iN1<tL)?YQ2&V#U+zZA)EoJF&TR*kH%=on0FwOb-gl4kdt4R<k}
zWlJGD0zAYx3O!-`kS{i|>wCz;nYmpV&hI>jZzi56hULJ{%zPTifUgzAzf(E#tEN34
z?0&y~{3x*xLXT7-t?j6k_`R3#NDlEwtReE`h;nDRy~gL|G<AJ>nepr5Dr4LC-<ZWa
z6h&3x4fv%llGcF=>Q9i8pb^hTaX9HG?BLFsuUW2dg}=tGK4Pg2*Li%PoV9BnxFHi+
zxrywq4NRdPcl;h9ufjM1v6X$tYpIr(MoH>ySjzo=`gqkgQTx1p&uJg`HA>R^VXun>
z^Jo=UPrijK@T9OH4R;IcK7e&}{kiV`FWK{fq~~Oay?rBiw%@DeK$zc0qLo&eZ7jw)
zi}h#53@9&g%Nf2-0DKwkXi`+dwAM2*xu>O^f<5MPuCg$VuM?YbjNoBHa1%--7OGo`
zirM*|scBPPl+22d{v5(G_O}e8(@JVQg8}ba6L%hODUecJn6j^2J(cZl^QyPFIjnUY
z1hzPgVm9y}GV%{)*=vy5k_4;dG|^)aa((tx(zoU{QIOWbgC6?sToY;Cud0~}a3o<>
z!mLdOlvlH%T817~Z?6wh6`tInqU;yW&nUTbh!3J!3BE`#F7Yrw2J>JSA#h;Cs?&;J
z-JbyA{D?<N1p8tkLOTqHc!Y9ZPA{=q?Ju7VKF_xAn76yHE-hnqq9CI9(D);*{0;q^
zr5>W&$cF0p*mJk@ZA_SrM42miu-0iS30~&_2T#~`{B#!R5W%pS7>Vd{r5JY&Z07@0
zPK1xysVS{jTB&PgK>lDcDSr4ul9e+o`!Iv+-ve6jr<Vjh-fu)B3o!98PS)qO1<L#`
z?$gOeLR;)4B%TjboK|Tsua&jDf8u<euUkm&iCQHa6<hwn;07?_V_G4~%th(~F@l@~
zuG;y|9YQ#w5qTR4aZUjlBo^v0lrbFav6>#b(*ilf9-LcwbisH!?C!NaM*~*7pErp3
z#$#r%3TN}i3coaDgjm)_w?h6fbXFPBCe+GkwuAiGGz&k2K`+u{v%}o4lka<VGZ~9+
zlt1_XX0Zhe@H><jEc`XmYwz`!8#kE4WvQmEvsYY)*Z!bwmNv;ZR&X3+Ik#$I3)dwL
zg5(Ucfco@9#{81s$Mq9P{mohn47LB7)P7X_<8lSFL};g~=w1q*l$X$&rv1r1=rm|6
z8)=$%VSzc(Lh**IU}6~v-#zv)pbfl{v&O&HRN;Pfjn+M@MV%0RtN0**E4p1J<+iR8
zEXX_BQDm4_6}TLDB8ugw{Si(pix}wsHOKR)udoT|ZSN#ZT^Pr4ch~Lh+4cXZ{rcPG
zXK3RFo_16tn~oR_GHfUun3B~h4}kGQG;^(~wh>FZRiyEfRJwTJb2lQ8&v|p|sJMy4
zb1d*&Z-#pV%|zXoyXV?VxpsMK84o8teTzE6)hJLk8qmmqjan~qJ?Va9LPVpP<&Oca
zLCbhhVU&x95GbYYT^|Dxs}u`q8C8*In^-pA)hjmxBA7<ze$z5(pM{~wPHHo>;g+$c
z^{X<(%kVaC)A;a;>)@)2B_;}HemmWrKHcj7zAoGLMrFpEJEww+0;7uW4aP?7mB15B
zl*Jm+U_xFM2kk~Judq@)$U^PZLvFMC0V6KBr!B!6?MautjPdoZ)JqZR#C|brZpl8p
zDIr<6RT2#m(x2|3i?h{Rn|(K0!xxn(znT`tF_Kjm&lk^(Pa-qW{PFtA0Bm&3^jA8G
z3AK{O0gjm=d}i-Io4x<oFAjKM6IxyAkRU_4kEYm0pk5z2lxxZ9Z#=Z%=Nv?K;BBJ3
z+oDRCowfRTrZT1@9n2>EiYB+D5c1H#Aep#^fnUkcG){Vt9HP(HA>h4^bDv;KR5$%{
zMI@pTv+W*ZSI+5XWlY)8{i}98JW)lZW<-lHOiLHVo$$WG#+FC=cX0S)Fw`19@lOi|
zRKu-=?^6Iv2;C^PZk)3;f~f_lhl$f(f<=Ne+UBtKeCg%I$SE1lj%;9lmy$3604s|?
zU3M<??vsQ(d5!0fV$<5nEO%Q-B|;Vh)yJCWyw>TRAqEjvr0@Js7nq;6i_MNgI1)y^
z1Y_tVjt(j<=9JwFwCFBDr;2xqO3nW)o3X_X1J{0&udCwU^*1tUmHtd5ANy(32KL~!
zWLEJ*M4<ELQ(faDGA?efjiys+X5X{-t8hDt=F%^tmi1kS@Cl+}tXPim<L|Bl9yD1T
zI!DSyp8Lt+yaE3(3$lD?0y+EWk@Nyet{%!BImbGG{kjd@hbN#*|DXI=v33F_Eq9dR
zRFE~9L@SPw?QU7S-cy*%B%1?&&o2Gu?b4<FYH45tV}w@28k8U&*FJg}I=Bp>99nHP
zc2OUPV$~TwEgV;AGclx+`(t4z^9<j%BB2-%#C#)n;1iVRAFpqY{p&FFtI_QAtU@+!
z6IFDJq=px}a7Db0SoV_RYSta0+8(7;2;qYL;er>PNb{kFA01}pDBn@N?)ThaI{Wgn
zmSt9ahsoL%uDd9&)0MiqGh)8BiB-%#K!jaL1pd&X3a|JIt0cePSa!CQii|G^C18G%
zZfKTp?Do!r0O@^(J$Bb1@~|4UyZJ)+vzjFGka&hDkTJ(AsiLzB-WMZIe4OQ<9OIWx
zJ>kux&+w4&+YEi`5;Z71$AZk$T@L(|U@Qe2?68u94f*0UetCKSTzFeiauxXL>{aFZ
zZBI>TbFP43o}T=RE|xoY!lV7zQ9feHx|f8Ag-*r`t&Kw0yW|q5%P+SL-`Ox*^6`yG
zVJ5;OeK}$um^FtATU9jEq?Id_<MeQm*jhxt$5)ysY#iGxDK3qU*0v~Vr6*oli?+^J
zU4z)LuW8`O-MzbXSyKq!SyRNF2rmQMI+cUNd>1+5ZRYV1RW$DBFOr(!^@$7h+Jr8+
zo@Mst-f78|zs8C6#(wIhW+{cjg>X^z89ljAXhiIJ`*=H7Ws7s4@1~_hN-36URJ?nN
zgSmglvt{QvS|Cx=djbY?J6=XyqP`c{*3rX9Y8%8=zkY!`f=0Pj=!#i&Bwrn~uZo+R
z0@OQ93f4ON!HYz(%t+d9penG;s7z%NB5BlSRBB<-e>HM;j61@-&y2?mRh4!#7It5x
zezGZl!G}Sx1RrA@F3~<tL6aZp?xn*x`ZJvNaYFlF_*(l{`|rQ#Ts1iuI@x5r5Rt#=
z+Md?Ej<R4cMr3@(>)TUbLex!4H$_8G+lSg$d@G?jHU*WnszWba*J@g9z@%HL?V9WN
zHo85hU@0Y&@!^7Oa5%GZP4}|?tHkl`(^_E0%bY)gvIIL*nCn}crB{H&p)vZnCdtZ`
z`TGd*NVIpxi$WCklR0v71jVM%-$Qw)YX^1;AvScW=4iG$(ZQBuldG4uP6MUy>Ig1v
zRv`U$G#;jEqTd={d%6Gp@w?nagvBDh9g@wQ4moF$?AJH^rmZn}7!1)x6(niL>QHmA
z5hl6)8>2%95JG-pIhqjHXQpjEDI86!Z_?{Vb}pMW!9mv1!j+;qC_Ybt1Hwj;K~nF}
z$jJX=>Mgk9V54QtFSxtAyVJP4TjL(wg1ZEQ2X`m9yESgX-Dw<xI|PSd)Ayb;Yi98e
zdeLw1y{qb}aP(%_PDPky(g8H9A{_o-cq3bNhCvW_^+j{YX(|3&2T8zXaVo`3Y0ua*
zOZ;y>uCMrMs3)Bs$d;HdCL}_Yp4m!gGUG?@xoHFl2ZmH6%EPs~tb9emo=uhR{1Bk<
zm9I@gyP&VUrJ4jKZCIW=QDSPy(Ym%ITk)uy>X$7(fI3NSl8ivC(3S6yVlj&>t_;mS
z%~6xpGchSpV>Y(wSa;<!tLv2c<LIw9NkZT5;A@b0sT)BqO{Xy$A_1R${eaVitzwI(
zEYQN+{i_Kc7*sSqkMktzA?FumRva%aL-0>z{aep;F63m%hzqa)?SYbcNDj{oeneD+
z7i&U|D^5fo6u{*lWkJBTrUB#c2Kr~$tK$R+bFGD*P{aH<Hb3T!8ujw+gFfoaAboUd
zQ2GqLKqa%*iKZ~8KOYVIVjt7&c2)(o(0nT^v4dkBd0mN~KmBz7Xdx$nxJcA;na0d$
zkXa_9Ti~<@HzmTgVU^ZX0wor|@LRAp;vMtXZyJ$9-B|sib7Kq)s&hpVBbo$zUUZ(+
zKfw+45lxl^t`%}e5F1bhYRbvP@X-v<JY!7;S+h*aPQxcESzI3DOzOZHj2n^G5#IJI
z#~m#u47rKUkW?poN1@axMQXp}n)@yy47tx%@|N7&aJ2dz-?1)-uKiix$if0_=5flf
zLSS#aE&DR`)uXNyWP`N&G7y}$EA6pzSuk%4zL}(7sL91%6MP543g61+rt#g34TjAa
z<AIm~qJ0~J=7h39_&sb7RNMUAj~L~X!x(<{pJfH7o#NbJ&(b1GLd^QynQ%JFe}PHF
z3c-w%Jg&d4o;Th<Q*>XUozdX8dm)48!xu~0<<)C${k<Q#((*+l<f4#=-tBc7b&&Gz
zY}tW~jGUiL4WAJ8d<o28AvV9!8WHnkW65TQlgd?EPZ)|+=uzL(juM4SvujsiKGh0m
z{0K~q03ubfB1thOm4CU!(YJ$}uDJ8FtFYk;g~3u3MzY=0Z~=laMz60zr_){h+7y{b
z*le88;XfX8>g_xWz`;3#JQidEYeWJ*h;3lkT%RuCzmLy9Onc1fp}CZ4B7JKuHRv!E
z$;_|-ppV}D?<!pC1eI*5N2xll5l#ei?4Rxl`x!U*aT!qH$!AL`9>KO+CmWp(TSbDN
zyAB?`)~gsL?{VZYXFw=6anX-yRJ4EWa+sx^a^+*cT!9fa0iLhfI`Cww=-(tWHag1~
zlOludvNInoShf+Hh!jFZ`HrG;;ii;m)@te7GE@`21@mqKexnGjIF-hDrRrE1BeO|@
zmBC-E%$hHIa(egt0;Y4`>)Xjp8ZFQP*<yPDw<7${Gg`Z-zH=I3{=;EL{Q=IZomm4z
z4^lprP?I;el$p4`P0LLb^c+TmJ0EdI51|N}65+2f_}liwumHQh+Gu;(>cs2@C1|_c
zklsQfSe~zK8}0_g<yI1&DE?kL$|wqw#Wy~^a#pixCdTS2cij|~K%GKXT$*TK#z~n)
z3pPr{*N1b5M_p=fUH_(Q`Cc344)4!mXgx5IpV4%MW7#yb_deM)v;!jLc1xI2bRkaU
z-azEt)T>30I){{to8=h%M5p>n_<3%q$(osB;a!pND9PB<9JknEA>&@VQ_JG^Z*;Nl
za(+mLa}!;kf10|`VUOVj-nF|>tYr9?-8lOcizlta(d5QTZuITP2PeP<cou1lxE=>G
zv2{e*pR1J77y^Nk7MJ4C{qESlLbaFZCG36i+#;l$`|+!DPD-fdaG=-FimH5S*+~r~
z3yv2|Rc7d?fy{(Ps4LxT->1{IDXYzgD;&)xW~DuN&??uvx~s4_mqBK#$8yQ27j@><
zV3L1!(>3_>hsonn9$~Q&qNW~_*lf{9m>LzoRBd|3aIZ2s**(WHKtJmEI7^0{NQ&vt
z!p0-|TXvF0NdO$3z9A#W36qE^Ux=Oi!!M!~hYc)@La?+;f$2mtj+g)lWi~fcar}hi
zmd~_%&${nekWpLKem#Z8Sn$~+4$E5(f(B>n{_aCj?tny{jDu@o3~Lurrvbx<c(cmX
zSl02wejq?DWy*jV{dfR?syvHE=48nCCYg;GmLU55U6VYBDgCtZjRg0he$<Vf|7mNp
zM`UsR%TQxtaqt;?<D1Q}+vY=ib)W?I`<uCM!6B*LqJVo092oLA5jp8uiSwYnoEs!s
z%fC!o=yy_vE)qu_h?)iclj>{~Z?_pCY7w9oOIs2->Jm#^Yuj1tW_2&madiE*EyyIt
zcGJXDbGA-UKEz?!+vz`O#J#iChYmltU4yrr=wXXFDjUZ`Z#O5<#%PDi9`4a@P6C(Q
z%D}BKLXyq*mW3GQFHA&jUZ_y4Ac-*|%;8Nxxz%pL=x9j75%XXG)50iV)!Y4n+jh#_
zOAsV;%Y8iRKZIBIt%b}TTBDsvx|6#~y3n*R4cFKr&%V#(1fhzo&$_2!#iGG?LzUA<
z>zqwx7g*eor)gTUIVdrQn987TaB{@!ymf!1Aa-@*Azl6FPO_2*h4QDQDt83{{1RMk
zb};=wIxd;pgV_ljWU}OQ!<{hL*GD=@B)I)i!2ud7fL?(_Y%=N46c{aa2!9J9O<f{F
zofy}Z>ZDsCj=*M}3ho!x3~>{=q3LI>7kQm7=o%vyW1yQo0-27R-FYHhr8EB%0Aq&k
zF&@<*&&EWu?;nepAnCF1FUH|?nbvMocB}R5u&z9mJ4=}UF^sIu_R#1{NIvvhi?w<s
zjXa~cs(@wjl!rFm>Xlr(d78}9Ia_?iqC6bpk5}&c(d3{5_?<W~;g199d&uQl26|}c
zQ5a^ZEb|FB++SfxC&Bpt%R<{{b@(l<z?BH1Kv2NSO6uCu?Oywg=Ib(TT5%|ADu0=h
z{KDC!NJ8E1NSo-R1{PZ7LBp2`+f&!{kambcHSO%V^rBef%~VR7jzUWg&Uu7Nmm&+D
z5v2j31E5AW4Wgb()=9}AiVIptKby^n-2l)dHQN!khSXiCIh}jI;6Rx3r)H(u*<sq~
zOYm0a+$kD&aq5Kd>`ywd4D<v`aKEBLQFD0#9#ofUPltIQd56e_`UZ6?a@R>9hI_&a
zip0xErPk59kNayjfKZAjnXgQ{NNKIh5_b2a<?mJdS0bpcv#b(qscm<`U-}u;myg+c
z2KiV*-a!#Rf&_B|(4+(x>as__s5~gt-073ZF?b6G%STT5<F7oCC+yqqN0f?)q;?lK
z84&>P4yPTe5~hxEnbT?*r%;S+P3&qPsNAgx+&x$-=PQkL&>}f)r4%U$ga8AcQc1#t
z&OYo7d_hp_9>y`KRM`Xyf^MsqdJdF`@6Hju(sb@!uOS;ZxXxat+^b|_$U3OaAspf7
z31K5JowX!M?MJD9Nrpe}UVgSr*|t#>9xoach)%ytzDFN|ie*b;Z2ZrQz*}5~!s<$o
z9&-;)vi+q5kZmd_b|jIDOL=RF0&oTR?Ie~ox|!eiCr=Pq?#4nthRy2*ze<<7{U1TP
zVqJU^gA%f<J|=Kxw;;WsWhzsyZs&p@g<7{N6`&qrY?SbV{SO3guw5DIX6t_an|Jz`
zmgkys_+O_ErM5h*d_;Al4~1TewRCt1HP~2k#jh+it?L|GKn9vB`U98d326+E_}>!A
z3(82s@U2%>i%I{V**GiS%B#TD`=9zXk%|e2m#>hw=y=p^)9dABgsV&Mu8VgoKl<>d
zbF>Kwu{<3k0o*6c;>^7o8vRe}$MY%rHK-3vbuH~y5M;?L?Otj0B3qnBPh&-2C~=G`
zoUlIzTl@ppY(cDTPIes?8}gMXd>MMbNdnb=qeHvjldMg%#=XNz-J{tu<~{;tLBOK4
zRwVuwlOlRV)33r8vm<#?5P;W*OU}1>qiJaOxR`BJp`$`+7C)RyzSQ<kjxthH<4@GK
zQwNiCc7gnl@X^Sf0$;1CK!F65JsQ6+%!rKeJ<Bq-Rx(W~WgJhwM2&cel##AUQYYb7
z7R4`~$0M~rmWiaOJPKON%EOdeLAK($Ve7QKT4}BbTP)H7dGYEn#&}?zSNDXFZ|)4+
zna6U{PbKdeB}8#YR1%ZKI)zya(dy|8U*{`}d<o=<uHpy!tKU&A6fbxZL2Me5wzabi
z?s+*v%R2fiX^W@IcgpLgQe0X@Oo*DL>RAq#Zlrz^-QkM*0Dk+*+*<W|bd+?zJ8qPq
zfl^QAPOs=!8|m{P!q3MWV80(>{x>5Gtsc$wkdrozGmE~v1P}*Q?z|9YoDz>%a_1_A
z%N*QUM3lsmyg4@smXAdqvFZ1h=q|pfot}5-V{4Sk*ZrRjecP@OBx}oLX2{8p4tGIF
z3Zik+?3&pgWq<5X(l-N#RQI`m2DOD<e~5^M)m9elOH=W#h`@p#dWj`Q<d0e?jqqrT
zsHPWvyZDiiaM`_**F*NyPhLpA?di^&puFxou7%Q1)EG8<hp)Ky`|rZ;O#_v=gx=V?
z^*f2cqpQdG*M0b-QFk18$Njk{v~K5FmR5a-Xd=CM7>d#_5|uo>BIkD8()4_{A6Gva
zC>Q~SEVL+)Pznnw+N<GF)j^BbR;*+#DV|6?4pW~PY1VzVsY&S^)L#sJUCt8P64*3p
zTIwQ^7yJm}M9uvMcc4y<*57C!<S&{<)OEszZ_$M1zPleDM_paiD)4Bh;#bDxlvRvY
zjy2WpPBWXGcwY#YlXA{i-flD6-X{CruHSAF(DCgT;evt$Pec&wqm7+j)B<aa(Wg)S
zs1bgx5f+c$Jk_5-RbNNekjGe@+Gf+CQb#9N#-ya{2#_(G_I2$3yGjpwd3P3IuRR{u
zsFB|PYE?O3;-RZCmTc!)xFi6$^ZTN=xfVx#m6)OdP6P2b;zq<^_iAYSL#^kU^X%;u
zru+QjjEnW9I1Fp~@+4CKSl^yq_p*Fh-$CZo(L~H)mX?)jK+?;X*)6^21nxvjkb<sM
zEuBx^WxU-oFc&=(4v|ic@R44LznQCao&IS**U%F({DszTM5dECzd&lF%FDo8Sa6!u
zn5KuiT9ru7h&vs3(D&&hNtlSw6h?;scq)_5%Wkh*_^VFnHtoRv3kqY3huoqINp1Cm
zq@h$5wVk*pg<+9?6c4L7&kv1&!@+SOku1GIc0Btq0w3z3fv&ih{hQ#ufz<89<UdX{
zsTzQm+jT;_i4$rEMvCY%k*=EvKJUJR=V#XA#Yp6d2RrNYQs8SW$!Ap2)naU)D$Alp
zhi+$gG08tNJ$ngM^-F{`$g%m(yaSqmrQ&DKMacV<c}GViu_aPfN{PwRbCRBDo~jmH
zgvSb_MXf_DWRW8@KZ@}PomV<NrWYFsyHd#9+-JN`^>uJ{+D{I1en>JwNhXi8sKF<<
zj|<V9#LqZS*nf@)dVUuJ>U7O@@}e^6j)KhWOK1`Y#^~*?;%5lmsOc$9J&dtK<16Z0
zpg`f`eq&~HNzoc7Jbcmm>OZRIqxXwQ9^um)IZNueH@HIMElGXd-+6QgS^HMsZDfVM
zPAVJD>WQSdQ^zzj-W;0r&d!m(q}t&KBR-LD7rC1UxMSZ^Vimni=Vya+J~E$5<UEBI
zJeG2W?C0$FZ+v6}zVa!H1PWCFbA@AnB9{KHh#KS{;CS?;n9sg($!>(X+%i-xw;1Y^
z;vx{9H_Dov(gIxN0F;Sj_TB)swL)*!H3(p1{beA|O4Ol(3tm2>b|&(?OT$Q}Tr!<-
zIGvMYngNrQ@qX+<oTq@<(ZN<crJB5Gfq=%BbJEFN&D{4Xb-s0@B&1hi^InXrW3kxq
zNvB8Qc(LNLg~{i(g#TIJ_KTF~&#B??MgPgr>x2~V5J&d>kYP7^f&C2${!H`>#x*=7
zYka<omlBr(&{n9sgt-oes?_{A|MQ)?sHb0Oc;B_POS2C?A#|6^@)6$>@*A=R)V7>$
zc`^nbg^W<4r7T}0(-^9K>}RJ^+C=1Pj9=UacfC6}jp&jruu73?D2Wi(u{?y?;`p=0
z{+A?P?u)VcSN491X_V`3)tc;>x=W{WnV@{isc)ZOEP^=0@u|jQl@|~vlJ=))@CJe}
zZBeq1L~;)AWVKsE!k!m6d(xyycJV#}R+hyM2IsD<Qk2dnv*BE4S2(d)S=+G+2Hjr;
z#no&ZdYo&wX!3NzMZ<fF({LVO47Ly*0*C)mTKj*@V!#@sJaw~>WTbWDl|*buwlI`*
z@*2cRdC>CS(aYMK6M1*{62Cld7*Dx3%lAh`u?a7J_3!+Zuaa_d+w%H*E0O*o|Je2I
z&F`%kBQPSq{Y|pur{R<yPK`sPht;JljDyAJSvn`Gu{xyee8cFRck6yghDQ#QhgSLg
zP{6KSc&~7H-qn?}eopB&u<+Ic0MkUNc44mO#EMptjjh0YoEx?}w(K)``GM>G%HF=<
zFydP@&!I{os~9ysMKuSMu!(xwQNhoiZvmDLo6UDMv<z~<l)iMRj$spdoOdggr|8*c
z`?RMHaI0XZ#q6J0Wr0z>i-X<%lg3U?OD?Jo5M;MhGQtkP&_<;+8JDndLS4h4KcC>h
znV5@WC9=(=kka3@P{EwcQB4vdrBaUFi>Qrn{3=SdPoSW5wsI}eXv>NL^{2+__Yro}
zwpYv}4m=hnq^Q>r0~;%<rDF#)l<M8x>91l883iNRFr0r`uNW)&a*_C86~6F=FMlT9
zK>ppYP5<IeHjX;BWK8G&HZ^~ljkgM5h7xup7@1{FhZny1)+yo4j^tf3ai_;-HJm}n
zzM`jyGS3S&-^ZWcwrSO6C$ja*GD3?7x#0y}znfamYtp~iiKQRQ)IFcF#`j#0)0t)B
zcpAJg#7CwQCc_cA<qQ%pbWesCNPwh@#i@woW#Z!9Pq-dr?b{L1^xT0=06Tn|O#O^;
zm-0}|;-q!qSOP(VZ*##C?z8A=G*KG#W!=Y4PgZlmiw;ZJwP`M}_Y>9+4Im;u*lxY}
z&MQ+a;onVpztfh`l3wkhkf*ES`cE-lhwdQ8DX6->HC5-lj5Ir(Si8Y)QCv#R3&VvX
z7qK$_jBj*!%W<F<7br1u#&fK`<5{9|RRSixbF_<Kznbstw$tj^%KK#gL6>psq${{f
zcYC$SgF=M@xCpG6r?xe!YrX0r)bWO_KY;1Bk^oY>?Fwm$jYd6h?S9nZ#&th;%Enf<
zZ++hb0&HGi<{_dmZ}1z}N74UIFfE-{nv%^YHMrhYSmSJDn^r$(vY5xAf2X(r-BU4e
z`(k?qAc0+m-ml|&Lxdwgb`2b!7krF@q*8_e@naHyOJQlJi3AyRz5S=4-GmM>o#Wgc
zl^Ijq;h1XKO3Wx6%n7{oycj~32^|iTQ9O7}|1P0|&0EYXXlv>yC-g!Vjhj=vsdu_R
zBAn4agFii9y-QwDJ5|Hf3KRcxV16ohCL9glVnfox%`|k`J@=zBQy%wkTWrr)69#h=
zYyb5SOj^M8u6OeF>AqIj(~;jqLYkMTlHGA3;z2D`vWoBeTc}e<F1y_uG%Lzeme(Sk
z(k>(!zQ@A!#gMJGDZ4TI=J7o$^(o+GnVIbP$F-bLkB0l3GNvA>wXUl5W=59r42o|-
zYlv1Tlel%V@<OmM3d{C@42R5~e^58Nz1liw*;FDu-165JKjBXU;&TIm`tXjLDaXUE
z#Q}XAG63gUze?OypgX~WbROJAS)tvYJ&!$hI<72fM~B~Z=F^X+DP;RN;YTzP#5))d
zZ<?JUOR)qM9U92CG#I7lG={#{{s|xk%CMkCE_6s^j^5r|ESwHFO@W5K?0<_bcH^;J
zSeGw#y%%e!a{0~^zCA@0DT}>oy4BDp$Z?5oJ9AMkRd69}km$L}jVk0dh}H({O^qI6
zOLZ8ldpw0w=W|{mQ7*JS1k`lcHPW%K?8!uGpChQWDyVVcj)Sxf96CFHQagL~nKu~%
zNimd`I{&u6B~5U~(i!0N)>fMXP9hf-Yhmre(I=y*!$&(guIN6lrr8NbqxhV+|6cXD
zXREz~#h4N7Nv0Hy8B6EvX%)~oxBt*<Y*pYX?6;=ci-|jF#kAs5*U)e9%a&!{ZHV<+
zY_gJ=%r4m%Dui)ma!*c#)@>rIBPU$%+)h>xeO%${hs8b)`o~G@V_A?9LNZ27+E}s!
z1_R2^tz;U(HPEl!He!~_xo50`xY@SL6|CxPsV+>_lTl!%6LB)S_1y?#GQvpcyFuvD
z(U9l(ocDOs&*Q#Qp54#4a{u0!f9apMrh((1=Ry8|(0AS=_D|)Rv)&(;OI4SbqrM8+
zzx6j>my3LKHf{#VMjeGiA88zL76TdBo=g5mDEnqi)lYkJwX8^?5vbZId?9cU`|z#@
z@NlI{ztm?zb`)K(hsAQtoO{Q$AR8aNVvj#$&@BnX_{2myQ}g+Dh`WK*ig#a`*vg$Z
zZq%H{1=GQ=*h$ESTH>ypKYXJjy_KD2Rd=A#pb>~M3MsEqD9G(-f!9=xm7+_KI+)Q!
zvcD{~J3SCC>S7pa_dR%Hx+uexfYJtybhWM98eg;CgQ&_JVyL3DL(M6oYeVWobgdGT
z-hGC5y8o$liv|pPenccezj`S-*y!}iL~!p@7TV7u=>;;bOW_pZl$vYR;JZR8oFa6+
zljX-PDE4JcMA$jXs-I0&T(vVs&B3&krGjp+21OL=pZAv0G~|-a@obRSl&gm1u?$+a
zi5Zr2A%BCOH*c41EFew{neKzAW*stpfl?Q|_l$k&m9tHIZ=wg6BOIdh&N+;cT*8Pc
z4metT4REq8-R5gFhAgR2xbXxfwnc;o>A=DYa}{T16FtnyuQUTy@e?1|A~yF=m%YKk
zDHXrzRJkVBn?<zj@bj%;^wEKUysQk`yhV(4&nx0^j^Rl9Jf;@N{Ydox-j7n^`17A;
zV(0k^Y?PGL2*OrrO1x1@H@ckI#K}QX|FR0bc4xCBn`IYAeo95hsbep_D0`lIwWtwc
zZVPM|nqsl%bQ0{R!yGMrPP4SqAU5i_%G-&98muH|g%*DY#$J`S(U%V@(YL#|L;_se
z`i`tS4-@NMv-;iJ361?Y=FOB%gf7~@NZv?3$kr--dn?KegVO<>{6*VRY)ef8=LrgN
z^1dFhHfD%B-p642h00SCj{crAxo5m6o5M(<criuNyWtw3L;gxOTe&pStp#^Zgp8+!
zHmf07z3u4d=r7tRfbXzNv&L7I$!_?h#b(TVnv>CC&PF*GUCh`SoC98p(1}7zG8IMV
z2<OU(o<`=1TY||VxgHMx4wbwoF!<Hhk!EAF_2>ENuV*l(UI%-9{;LjC6X=m|>q{UY
zUQ9LpA*81L9x8uu&X0tTJ6l_|22jZ#HxHwd#R%K<RUltqkVP~9^yJF{fFu6ZWhdD~
zpL+;3NgU?{PqXK|Lm;rDO#IR@^1thl8e-G})7&oa!gZfs>)>n`bj(kUr;AYsC$Bc6
zQerf>(A6-#Gh-c_IyobyF;j$|C&K!9n&izfFJ?C8X4c&AhemkW2p}D}IP(C-S!9$(
zop11Z^e3|THoQ?G(4@;Z#g)cEYzsoWv#4;n;@@<lB<F!dWrYg5idIr(NE$zDoZi18
z9qTYylY2Jvk%MHPXB2)gCgL2eQPz;a?$Eg=&Ta{?9u~BHbD#Ei9bf^YxzZ~kWk+b&
zut?}TW2@k%$U24t@ua;Da`|Jk)(6<W_vJkAJp^w7Fl{#@5H3oojRd__SMlyHqeb%(
zKF^1m0!b$CMIe-;$4zhnOb%!JpV021+kC=FQbVtf6?|R3=e40Gy3@w)>7Mp<>r4Me
zl2rO2yh%?I$szrSKH>CywSTkKXw#aMJ!0WDu5a&3zIekKpKN!Ra(nIi#Si?vc0LIf
zx9I!)D1lhDJL=w2RuqQ?O^X`0!Bc7?WQHdr4|^y_TeW~W*(xGilG9;|1fD-3jjlG>
zJNIZp$U6OH2@EvjvA(l`org|-)73Y~bbXXB(vPu(OI`y3s2R7YsB9bGU{<AXi~rL$
zDD|@fg?zm*7#ie9tD*hgLMLihDPqBMUr(Wo%iC$h=o%445cHwkjb9v8F@*;^^3!4(
z`fS7id3*k+YdsXFcK5tXv_QG16-ilLOj|7<CBlL*;88mCHwEJiUSfZzg)k*HRjAic
znYm|7xr!<0ks1MQn1(p@Dno7BS2>@qH{<3+>m;1;DNdzw)(C;$i-O3dI>|_jiTTqI
z0qffH{~C_ziH|h?<8W6DH(&xoIk92IJcF0xP4xO#;0Xz6QV0;l5>@K^?aTo@s=UqM
zEcA!Gw6wdIA=O_i#dk;laGmQ+T}8<R>X8$kxJw0vVr2LZafroHaWR%LNbzxnX&<G`
z#jGxj{;m;vLtBs)``2>BQ<tXJZBS6A`Vlr}wE{y2@MxS#hO>*!V;v6!FJEq}O`Gpl
zF5b}QMW~>2XdFsq;B%;(B<xq_$^@6Wr<s@4VigEIXRG40iBw>ch5#xk0(Z|oYN|7S
zZKDR>9156^AWe}X2Ap4aCv&tlG5r&Qa<?ho?E;f_GG)-Oc5-)w2t8jYFQRFgxoE5v
z0PZuTZU)TGJF}3f!M(@DPcJKhS!gRI#l;BuSg?6rnfA|eDGymjS}I^Ow;TJv(XGS0
za;2MZ@UblAPc5w+bd)xf*I}+x7bBR4C4<uz`+W%8P=g$9i9=nZ+~-oeN}3SM^cv#K
zj}9PxJ5e|nt^MRP=jLrhYK6n@rT2A_CL^^cD~rw=`J?Un{X7vTp<kq$C^?J{&h0?Q
zRQ1>8n=>)2QPNt<sFJLj4&?KA3hG35_ygAX>sk7l$tdRycpX#DtTo1NvatF()4Szq
zW-_wvyH&iI1m;>s>PV$zJ{0Om$gk>*<p3!Q(x{~U1e;Y8GtM<2WlmKyuE=IVc?P>x
z+(F|F&^`D`T4Dq)pX#TWsO%k9Cb+{_e(A5j`motH3i{1%`iZ)7SD!gk4~2C0XIsx<
ze`9o~dE@tk$&D8Djr_=TmPpIGt&P@f1w>3G^R;EJCKif}j6^ruA@OnlPJC~HYM)C)
zb9%ec<K7EQs;P)Cfg5LmDJWfp!N}7R{j(>X(RvhY686vV&gDO*ICsuy)Fr~e*Df)a
zAOk6A4uPVyVdQ3Qpkkv=O+*37E!+LMOEw!IVJ}=kJ;uOm9j1#n+2E-0N{Yj$&4vh@
zPuKtPLO{N3isvXF5ump|1!VH9MLF6;^Pd@Kgwi4Vtz!?2)=kHpu{wgN^IP5K*;rnm
zXby>`uHN5A>p%A)>8#HVYD$uffz8-190*VpX%Dw>cY#9zP?@hnGHA-24o{U%4VS}f
z;6`?Y4Z$OB{OdnsX14J<3%F7JKdahLt40h{e0UJU93h&%#*4u4ZLmm|AYYAsyB5BU
z9kLQ5`DhcCfym*Pwk&PZ8V-~X{829Bz+Fk$kJv}cBx&dI`Rrz}IK!~dL`HEnChYE#
zg@CB+eG`2U$td{|_OaE`hTrU2_y~!sa9}q*xntq-7zK9l#Q@XkoJMLTevJWphTTZ*
zf}#Z<%EjFeulcrI*(MYkO)f?Z10V-`MVMYkbV#^$?tl7%g3W|iuy-6#82C<QhTEF~
zgM7|`)4CrHbuY=CbemK_Mb=HRkBHo_1bbvMU)mb9d4Jg=Ee?dg1q!$|rA7KtB~b5Q
zI+WS{xjIU>W-dScVi%%^vs?6Oyl;08#irPxEL@SYHv`mBbSOZa53@hmJ-QLXPppS`
z=`#a(lO2yN<euxACa+BVd;aMKDK@jIC7H~U%VE?S`gMJEy=JV2vXA|J5hVQuxDoEC
z0tH8@FB?z*BAq54{8FaT-a$lo(Qk|WfYaeoE3qF4sm8@(aIs7KV)GC|Jaq>Tx8Vxv
zLv(C3vm6W8U?!wQ?(WQU!R5+YCgaZ_Lc#vC>tn8uZE9)^PwR<oc`^N-X`8S1|AjE<
ziPelH3o$X)C&{|fmR|RAg*|6FO-LAwxNQ-dM0>0~I29eHKuuq32x4?gem5aZ9<WUF
zrw#ZT|8|(W_4p6-thuHB5h|ynm<=Dnvl+>0OTcN-3`s5#I4B_B3hf(%h9y~U&A*MF
zMwesOsL`{Rr;gub;}tmg=QRel#`o!avOLv%v=Yk}xWs|km)&H!ZuE;<$XT!oaNdQz
zQnD7gG@j)gYlTJ)R!Y#%_Ix(GRsM&H!a|<UeNoAhXQ`m+mTe2$XU_Lns9%gaza0=Z
zSJwGwk?$5HVNB9Gav8kf7&G~%a))*Cbh_zHwu!8a`N^(36rxI~<s(t}27ju053AR+
z&`C(#!(g=(x~e~Rkrp3aHAA;+Dizj+chuFCE~H0ib9Yj-er;7xXU4Ia*_XjDhLBI8
zm!ezou7DV+lSip3yVLSW!~5bVzm8gXe17?1?l=QP@iA-I*4*aTDte<vMXv!Py6c}Y
z1gaUIE#9zZhWJ!W(b_uI0K}E~Ij0a{+VWRdn>x&MZ4zC1yt|bj06H&SYPPs_CIhKN
zJq@i~KuH4$di^2U4x#~$V-F_$unMuIVTK^emPoiongEzd2Z=RfuCQ4)J!&P=@G64O
z2(*I2>tXgU(6l5NC!0MJIZnQaIK<QP5Qq;$9rIAs#A#^ppjU}v!J4+t(wV=IGY2U~
zlk97hUg3u^Z)`_b2>n-zpfJLJVK8i4_{j8FV@lHLIj@eeQfELsTI#|!0w8k<EA@^`
zv5pQ?KOayuD>aZC_99Z$$D>oS#X2i!_nC^EzNx}b&#?bVIKA(SCdff!r5vTKJ_Xw!
zXP^u2baM(X)lsb7qMejN=_KDKL6xv04wfdPD>LQ`vW+{u!E~1Z-2Huy|J?u7gB-rU
z;~)s55ftLL7jE)y#x!C!y>UlbJYzR@|Jw1F6RM+&CP$=Il6HG^IaDHU()zEZFTSyL
zVzYC5vDG<Bz3T0-F*YUQWTb071ss1qOP9W1R)-%J<*%MD3{(2AlgL|G^fR@wTV0Q#
z6^?5n4t!?Hp01J<IENDa!J0n#WG;Yq&PiTqK>6BlN}KLFw;tn679sx6cgXD|RP=Ba
zy#HJ17L7^`FE29~e`^kPjp1hMS%QS`NatVP=DYwE`Sm#MUc0fnI&p|%mKU{d8~lhJ
z1!_l(y%M{%XWMvgvCp2WyRDOLhL`PfULnn)TO9F(oM5!7kzi>9FlUiQ!$4-NyZ2wV
z?*7W=4NoyChUOA)WgJHoLoFqh<-(y+qfCM?MadAfik9J_mQ8%COC7RB5Ly+49M<zw
zHNh~>D1rXVeYEL+sg2COD=u+G<l9hIbY#}(G)q0r@B>p&o<t7=dOo@M>3b!+(HF!a
z*vSahS~|6Wt)l=V(`ueMRjijECMXrHS;@9!!(;2iccFNBu5f<|CB`e=scU9=B0+3Z
z=M#3uSxRQLJevfRGe9T3xR*8-ZUuB2ZODW9&rhz-HP0`XTs(|-t0{fLRaDaur((T~
za<#lxSZc~~Sac=fPa6Bajqv|!`P-fd+O&l}t8WoWnu+V1&=ENzGWIM{IEpAfWQ_oE
zX5q`0Mx$wvixJ+8Q1Q=Z5BmK5TtYeOYozP+th)fX*j%m@<(^es$w;A8i@Sn)N1Y8C
z1AHvCZKyg`8mn{PQowNw(fL_F25AOf!*c^K+A*HypvU==?k2Jg(s<f@Zd6kBEnOlb
z6#zSXkb%N{tbxiZZiladsbvq}5mF~oEi;$kAdWCsMR?x1P}9I3<*N}7DM+c4uI|Av
zGbmTtD~Kq^GEjkAjqZe^W)cwO)%U+C1!C&pkx;*~`V`guaR`SvThcuBeIO{{4#49g
zYBJW-r1#W=2$jvIbHy~j>(LOQB~-nhp#UlQ^OeuhWQ&dRO3VAtsD9s42s3hoVTn!{
z$EvQ-B6%^-)qB`1!^hJ-G!+x<0=B=(Rpp*+`pZEM$p{PZXmR^b)A&n&E6t_r)(Kcm
zD&+(+&^D%ru`mxVB;HgB$^I%Noa~_naakX0#p3R&DDJlIQ*2|f6U2pxG7Em(%bm+$
zN>%d1=5;7N6j2-Qs}icqVnJt3^%GQ~kQ>*|OTJDZmAC%Uq;a6UNoF`7TMPFmdIH_*
zF~z_QQy3OVA*pWL)b*{p3-;I;>%b9G!2t@-P+r3<IckHsal+G!8s?E|!}NoF*{z>o
zbp8c5ul>*uKWgzTWU;BjQO2sQg{TMhkYP;sQpqX9bWX7pw1D14(MiR%Kb2HrSfn>+
zJIexaNPU?~95Azm1iy$90M<mBiWD<;`fJNX=docX#uP@RPXZ^vu&m&R&=3ISVKB=N
ze_`Sd%9kH$+()o#aT3Hx?>t3xKklWRAuXIfuaD4o_M4k;(ML0{x53sD;i9im%lPqb
z2Q1u5h?WswCYxM(gDeT2ln49$G8$-598Yi@8}^n;U8`IhOjnDmE*7Y}V7h8h^5(#Z
zI!-L-Q@;Ao4Kt}&eqJEQ$jvluBBBN68qX-%VATunNL!<+s7{>tYjO%kKmSSRz*obM
z7t@a<?{f6a5pJkX6J;E9Hd#`wzXeV~s{5nn9mYs{#1Ayt_|RGN*eUd~ld)V4cR!C*
zu!#nST{!obn95ttQ$mpxU_KNZS<FimlIAJB4YH71-kPJW-e9&RXA#Q(Pzucr{C32U
ztDupE>-g|!oPxHoK58=QM+utNQ~mz28|R7Yn1g=)64`ixTn!Tc+O=zCP2wH&OYbW0
zyNUte&yr}4yK_pdc$Q!|x7+z+jl`*Ru(~VO?UGjVl9;Y?K0~;gYqy&{vSX7f4>FT&
zCAAIL69KxzVGACZWZiHN5!Y-kI46g>_idQ@afR9!q4B$4ar?5YzHY!&1sZ@m+P4O}
zU%0;I%;yq|gyEYE*SUZ!(!5tAMjE@7m18(lbcTal@2*(y3lHC0i0)(C^O>F}4X!j1
zEh~-yH@=isRzdy#KpE+VaCC(UGsixJ<^6BW!cecK;o6z5bPAwyFmv{|uE({I?98jq
zrAfjoG43=mHI?Kc!~nKB>6%`SrtsHHB!ll*4bOx*^T=6T+mqbU^2@&r=dqq2;PFH)
z)?EFHjxIC9xqKyk4rbSpzhS(kj{>xU!I5lKv>5usNYlTT2RXGb$Ycl>!v4wb$Ly`T
z2L5r)^7R&N1Wy1Zi`kLV(!fm9Az8Nt=rkf6B}13f@zwABZ=#wrBiB3cc+vvh+c)jG
zFIT&Hjyinij-w<R^bT`Hr&<oh_=TpK;h${(n~u|<#q+#V`Lmhl89>n#enqfUYuG#_
z`3*)!$seCIDz3n{*oS^yp{-q;)*rz-21SBdT<Skw=1NZyIOch6&c8a0qV5K>THd|N
zQ7lON@sxwsp$=iive<}9JC4+z7%@^@guLg{*RY-Pol=XGOr)p|xMP&}d8mY3dbY_w
zeZ?Cz5o}~vnhPt>O+ezc?F5h<sRKnoGFq&KWVv!jkQyVd_x1G958JNr#fe1hQoYpf
zqP++^k6AY2cBk6zI($l1edfDv7vbq#=Tv5mBhZ||;05MGK|r0-?mhpViwQq#_~N%z
zy0_bN_6q);nQK?Bt{b{e>C3#mIRcky0J3y{^xdB{_<AFK%KBioYHPq(p1I|DBC7OS
z@3ujR(Zz*nUIJSaa%SgLjk)3HBZFx8I?;ro2-E@qDrsBC;HnbA&<MkWb)rg`1XJzy
zEz-YWpP-K)yf!B0@$e#)d8bWmKD=~MdvW*Be~+f9^2h1A9wT&Uyj1OsQ{q@t+dkpn
zJ{cv_F@hd#)`>}XZA}I+BMW9}R0>poQra}B(u{v`*H!oQN*1Cm)ASBbO|_fKRg2Is
zTyjBLKg86Kd+o%Ge4J}`m*gk`fzAPQAU>ciy#)RaJ;y7*EzPU8KV(KUF#DRwoUEqK
zvzePSpDKPNA9uiCE^c~1w5g&OaB4e&D{O(H<y=z>DS)x830G3D<USYVnB@j}3@w90
z`zTbqKUHD~Y{|t)y)B&Rbv8Wd7un92ph&AhUg`x+Y-xAhFg@P2xyKU*yZBwu)w<w2
z6wzeQKaJXCj=z6CtbFS!@F6>&5QB8!R8vY2HJGax7tRZJzBKUzncUy7?Od8*1fBo7
zR5wk?+*vq|ypCcgOioGZSI~AoN}erc|8-<OjFqP}EQ-aRM%r(&5Hq<c%++5B*A${;
za1zR02f+}xig*VDbzG=lBB%n+Y=hKBPU*!0S<<)ujW?LClqI!y_=J;9r;sgz<<}n(
z&JmCj)gTh#>}a;;^d1F+*q|QocP7yrkA`%U8gZ;`uIccOU;2)ph~~Z{(ikf5+c4?J
zwCO>D7L~0kSp_j5ykvx(n0-|EU$mU^RIo6@kCquq^Z$HFs%hMa2wFraw-gQlNuCr+
zNcRPTQbaC@(gFMrv|g{Ie0+RbV$Cef$DT4W0vBlVKpIBd0xL-IdZ9zIpOKt{#59_0
z%{RJT*KtNKSM8V;BCH%?snB*4k;HG3hG@lf!uPzT8XJ-zSm1DH%1(r0wMbi~<7IfC
zY@=dhTC>cCPw^=vl3tGW&#)D@%Di@M&P>e)YLInK)uTk+y_mO%W4S0+%fBwqN?gOB
z<T{mUiHy&kamr!W0|iFMIMmBukkyhjGh)gJFaCiuyEpr$^=m)UW<7GBzt8CTAgk^^
z+l&>zeGli>z7Jzuk@1QeD(*E=mn?S?gPz4zE$9}Kg{jeGU0P?zK^8I$<yIx!#R}RJ
z-f{o@ODaV)&KOGzL5wO>NIJ&ZWj`P85|fc7<f<OWANWN>@1R+Hk87=+cp)q-G9Bsr
z{=dgH#|?s0sKAY0MKuDpNP`}|HnfFdpl`^gNlj+d3IpCVZiyxepnY9Il9}pX+|Bpt
z=q@j%Ya*L2yv=)iWcRE}xFSVGh!K0tc>l}KxKU#3`7}h--K7=fII84tVGL~7%%^m`
z8{QO8a|eGFFHcf@e*Jiik%`n0-ZabRiH)E`&JGcYx;D7(h(cOG4q9xtXv=R85_Wrh
zfy#m~@(fY&3%c#hJ^Ed*wVb)~4KW><0bxh=nhsX_5c=K}kUM+|hz!zNccYo?XCvol
zN4`$5fsh(Hw2p>3X8+IW1T^7HwzFqhc9Y%wEAw&*S&uLywsIG#dKl7}Gf|9-=0Ktx
zEre@khgTQow3KUC_0x;0RNAJ?%vH^>xOK$f*Wv;{Yg}X#(BXV(KDYpP2yf%ta8B|h
z)LJ4=$gP3Jb`%5OBARo!z9LOxwXCjsF}5AQ1ESy<;Ak#=4MvVncESeuo9QT{V<Fb%
zvx8=`!C0NdZ;{q)o5;*f!8mWuz*?G%VHT}t4@d2%B`{llJtq2nPUz9n2!Ld9b<-sk
z521R8fMAN=1qgC@ry_h}pv+g@4Vl|LV_sC1{sv~cCY{{+5<tl-wlk^l;TK@=40DH)
z$cxGv1~6mzcnohz^#w1X_moONJ2y1VF#zfdeI@uWaM>loDa}mmrYo$<YAi_n4E=T9
zE@;D*cwCvy1<0rCOoo_wHEZT_n%aeEg~yXh@1ex4^&KdTG6uSJ(o2CV0vC_+IgFmC
z_gs2Cu1&PbR){N&aRZ+A;XAGVb<R#}j3o;XZ#3~F%n{4cJXW%mMMhRK992A3$TY)9
zddLPkc5$!!QLl$n`<)90>Z&6&DC@>6-7>D?Oiob-Xe~PY7hI7y*^y7kM<d_d)tL<`
z^o*z(d7W+5LL6A<UshUJV=~^651JXJ2m7!l@3UwOe|Mw=cNK8C2qX?Cgcn>##3SHW
zsS!>N90l0+MybkbfPdvHu(;bXy);{|rpTtqVoqWo1q|q!aB4`qop?V+-9Qp8Vp)x;
zWX|HQT{y^p{9Rd2B@fAIzAtI5KOT?|aVmn9Q+`_9eIVJPyt62He8rGbl(4}$cP)=y
z9C~=by(Uv-IUjfdqyUc^`l17P&PzGZ%-^Z=!hHD-SF;1Yo!XX1M&0uiQn#i^V@|e?
z5d28?jPzSgR&t5i2X%2KzaXu<a8@atd^5{sO-!wwp~3VLXw*Awsbg7ZurJ9ott(wL
z?vXK>_dIVE{`czNG%0%)Yj$OvOv4bZY7^=sg*d%st>K-pr;XE)Uh0;uJ98Wkl4K4C
zBHj6#Wrx{?fLKW;qj=FM2lpsH8==sxcF{*6)9_HLoP7+pWT^%$QoR1)*@PfNdYGLX
z7pIs#vL)_xA(4^;OYPs?x3T>u7*LB$A@UBOyH%Ovw%%Zd<&GL1-MK?_GudlJ%OMrq
zFr!!rUcHoa>U)NTIQ;|i@}n~|vU;;P6xz#*Gd?CJpI`o8FVk*|wAoOx#S+Lozf*b$
zKW8Rc(bh@+_zC`7x9&=h(1=XyWFn8Whv(%M^cX{zFy~m<s<bL~Xc$IK7PsN(*MrT%
z70QztFM^DdnJSDi)b6f57+q02_PWZ5RXZJ>m@wwA;6!)SxfEi?<vFss_IP_(IfiY5
zWxMTPKI>q@tAUz@ip$tFMa2;8^rgCaQYM}`eIatG3lI@-pNuB*i+EaU*dTucG52C-
zVyePgmYQngfOU=&K3`wuPAEfb*nG?=sN6+ge`TawXU3Bj6`E<H%X!BFwn;U07=qea
z?B>WAaO$KiC3QeG8(nOKz5Y&+Ir|i#myp?ubRc^z%Hxm`clK0{J~jQU)@ItI1by79
zuA*n$*tvkjA$Y(xbRB1f8Yw?6$)?DA7%1SLtsCh5nYZ8Bu$uC-%drz0kKCGjI~GV5
z97Rd$f{Zx<?gx*w^}Vv{$Lb0l#=2Y+vSa;hsoQcD&a<;POq9zrDmCw<Rmq<wdlimX
zLjH16&R?dR)wRW)6K^~;aVO;*Tl*7dc$6iIF^<Mj%heuHdXw#%U~#NKsvZA}rwK=W
zO|lCX88_({MPvQ5iREP#9QX351Ki4oweMHq*3LKEodphfHp+;EB$~xUzlte#Z%;A#
z1WPb%wmQFyP-^G=tx5%j<p7DObX>uL5U;JR?U<IKMod_ff*y~am}NCx1fXGdKA45S
zRCFPmlb!xZfOD39lWDM^rS7HK?<7Nd_yzIQy0ehF2t5J8SH7L$xFZ=}9-^kWluaRX
z;s1^0Xq;kB7|nuAGUay@{Km~{*yS%{cE9vzRTrX<vm!lrjTvnPhYXC>d`6{bw1|*{
z^seo=12|Dos4$PKQkS+=@GTS+wdV*G*?bHcYUfa7zy1wg<l<)M4$37Ls6(#Csw)%6
zlH)nq&S0x0)S1auyMDRa*csRo?_68__PeBLkqxi{PGb*rovi=7!hh@g?E55s8hw5C
z^o~z}(2d5Hx5&z4XuX$mssedkQ9gAWySG}(^>nVa(i7^M7H4|!4{L)#;%Y%ndTjEn
z8Ev+P3hU>y4C%j0@o|-lzn0q;RJTx9|4F7kln~QksV@P9p!wu<O+nWHL?Mc!nmNg=
z!wTqZ?~OEzY~SCt1~IK*Pn0oa^m(2i48uiP33NU4ll1(W|D-I)<)M#==tN-NnD4)m
zsm#b3Z!u=vf_o=^U6LzRAw({(&_!m-e&Jye5{DFlkR2wBFw8p+6$LAa>)2S1y5Ap`
z)GZx6N(>@QjcuE=L?v^)@Q-Ltl>>W);E1Vl{*8vF7n-l|VMfUwvzxU~TXQJDqJ8o$
znmVT5r<gkC1iU5>y>dwfTB>;=u7A6vC~S}X5uyMHmzrM@$W!||z-DZw!k_o28!q~G
z7qVZl8=3BWiSmWCS%%manUh>jjT6~Wd)N&Ni3k85$j-8?)2atO5B`!l1rJZs#w*pP
zmTHi&42vd}G@L{&)(Gbjk9*T>Xc!iP2i}`pn8KM{W!)}uyYud3V~79kRMZ@fMMm!I
zoA^^x6)Tu+E>8LIi*b?U=2SEHk``&*l&rnpXKx&vFT7sgt_}VF&?b#KT{wf)^SJP$
zWW-(&GL;y~tczhUWxe@rmH6$CAm-f${Jjvw%dH~yj}zX;#`96@qj^rnS>4szHQ$%&
zjo_}kg)T3-H#5m+V^U!mZ(UyIKTa=OeA_O#Q3IOo;y2Q^>w=^B1m9NsyV_y~(Z@<-
z77ty8PV^<a!696kux?t@V+w>obNQRz<}e23g#oW{HnsnDPO94Se%>3E?`$lwPE5a1
zm1Ju6&q6*DP_VWE<^&O2$aL;|yTzElqM`nzA|smqjYQ=YORbrJT$A!CRCnPbRevK(
zJAHb!BJeU~rw|HQJ+CO$f-EzEhL5b|f@!G{A($BB$v4D90<=ebk8{*<9G<K-#Ixh5
zWu?K*P2xX7od3hCT28I$cJ(Mp$xvkyMdgssopb0sXVmhP-#&FS3mXhdmmf~G?KIw7
zwdgg`0<dZQz8u@ApFO>4#%=Z+aNvwgDM5QkLEE4=njNb^hxsKf<+GNX9;)+tf#{{z
zRmhm+hV-Pku^lSbgWtNRf1TVkp9rfI{pjG5fY$gEE_#;N-|kXCA72i3$$cdgA!Bx0
zbQ6&q;SQ;1@f)cI#@G_ru3jxl7d(^+Z*LmU1m&c6PJ{&t@sR_<4VxRvNL^ZT>(X}~
zzq{YK$S&Y$CHdA~B&X2Wc9d$bcmL7BAesbbWrSO?vXr{4_F0d0%@H#)gF6EYb###V
zN*Gb!dl)i!G9oBRHU`vZjH=-kC(|MplYvv_s1lt&g4z`;TMGDE!a3uVA-?arDXT*x
z<Yv(J`;)4iuOy%BR?7~4eM=KZKDoXZJ$Q?-oN7-G?cNGB!V%y7a>nfS3ccazS&%Au
z1?Y5K5OKu4)o;Z80O@T@T6WddetR%Fw%kDG3)^Xxi_p9(k3F=JY3XJ6m6LnlG+*=o
zLImVzP&{uW5sa@qX3c(<Zba!yeRT{p;Dg3Cfk=S#9ANvT3<V0t{AS|3$Z^fkXSZ-Q
z_PsYB5~YHh?sJLVq)Ifdd;Xau<QOkms9)AZ^;q&#c^i*NQaB5OR4k|`>F!T`#150q
z&N7k-!k;HE3lWMq^Lo@kq={~Lx5=qOMnJ~_m^gYqcpWc0qd*`>!?24&KvySQcgVn&
zZ+_otL%)=|MtCluv}Vakkb;Vm>4T--zn5OlY^KXjtDLv`{()w1>~+_I5tcbVo*JBL
z1^T%IA#}ADUX_zpK~EB;ip+<7QDUxf(n7y%tHs>HEn+a?=h@f7S2a#Fd5z=A%?>+j
zj86rxv0YW*RN-ia`3AGdri*LwRI^y)@Cjm|u83|3I(p{7Z~Wq9W;su?+TuLGd?GC(
z1k%|K(i_NOjtG8)nZ{TJA^(|#>COFp;2*x8adSuVKY@SQaJ1_nuuP~8p}Xz|hS|xu
z)Nca8?{z+LMK~g6cqe}X!W5H7*if}JSa7}41zJQ>t_dfG_}SVA8oSFMOOUjMM?c^O
zGPw~Td3uiyxSG}GaFg_V$%I$A9J-+|B*KFHB9w}b8~yN*grG9C6rx;4YfF4UwykKQ
zN9;7XxS*93U%b;0NZp8AgH(!JmC>6BPsI`yAx^JB(<O3|Da+s}k>ZjmTSiG@>NOa4
zFx1t?Jb=^$XElFXpUCbnaW0b~0sMOgDa4YOpOIfLJne5m_O9q!nR#(Nf;jmGtN*C1
zT|XJ2V>_F?qgGLWzHI~@>~;fLc|Uyhvm6H=P}$(8ut6+^gpW0oZcg0Q+JUe9u45Mb
zuTddrEuN(_%nG4Y4K0$G$1S4fFsSbL(A4lp270wO|FIMS)C!YQ3u(<%$g;_NJX$%y
zi|_>aO}uD}%n_WsWQgcelFR)1rM4l$>S`l`v-saD41|rovJL0>$VNL`PS2WuV`6>Z
zY(J#p6dX|WpJzHM4E4?LPz4d)Y2~-(Tm{Q6{H)~17E+*4GM|k%f`7|}AojLaeP$zF
zCaY-GlZgbmhnceBBV6Sa=P4bxhM64nxasqf^N(;T3#HVB!sP?11(J*Soo-WYeO+s`
zum&=*6F^OSdcP`r;eHhk{@voiP|H@(BiR>5;I);4`-{hBUq}lkSj6$DVlui!>)s}P
z9vs9p)d8rKTwII>Ge$l@$gN}gw&&5yAx1@BLSyAFD-(E8wF)Qswsb049lReeZ9BLB
zRQ8V8+`m5|^>yy{y&vW49~GSH-JO^ySqFIzXP5@as0u->20^>sco8?*%ShxC{|^Ae
zKs>*h$_^cNxE!KbtIFm&)rsGHSRPo~LOW-sl-VC%f#*2*mP)Gy8-eW1WkE{iyv}%o
zwk!xo;E&`S6S3qBsz@_DwQd2gv?2bs3TW2=OJbNWoagtvzyA%Ttc-fi?l5p%hFEiI
zW3|R<S#dK4qJtjIhHzxEZM`SZ!GiY!!{-voKOW1Mt=YPcRo4TImPLGjb51R-22xsK
zqss_}P|4jf@KeOHxHF_USSRs|`_<1bmZc&D8X*p00S&fnlv2Q)Ne;31Ps1mR<t77Q
zpzqmed|Qq4C6@okkJgv`EyspG5yzcoC>98uaN+Zy$w?#k%HBtDEQ%ql)^KlR6$X^a
zGN}x>Bqlx3CgZrQm^CU5X*HUZt<t=ahRirlm={t?v=;k))`WJ1?u8!cJ<oN+k<b!N
z4Ns9fiS}IF9e-Y}vn|&M9k$bYM$$qahA<&&hNXUen!to*J+ihFCXG#8+&wgAi>#PO
zLp1lsZIUpUf)sjvZJ~H!o{x;9P|Gzc#x!JV>H4%kvM0M3mKxTgSQW)LMTdmz$kLP{
zXR?mmpFS`@E<EZ?tHSZ<EUhxv!cqd?e0z6C&J+LiPrs#9XMc0U!+PS~>5;_;kb!Dp
zBDK}TGk`~->GzM{@i3pb){#Lo!+sz1XwM;{Z(Yhn4Xd&CnU&D;s}+2~{gEkaG-^}m
z5Vx}S;$!M_(}Ol_enB~^M_I776Cdsy_xB&TJABEI4xl%&=H!)!<B9+N-#_r-@%RjF
z<;?TKFl5HznM-}Ajr98F%lLONUJ6(*<CBeU4)O7U_xW%@`k{VX=(cWX*9yA09r@EQ
zh1qRrWVu=VmWAU=@<eh)EwTNVQwj$TTDJv`e#%7z0lGuRMr44^op1Y*OUFdEj;uV{
zwaT*&)b~#XT~8|982biog7pMkx(3Gum9upL9WE`Y-r+Kj-+zvCQNkJ-k7RmCilj_#
zAW0ZfB+9~u4C^{XW<7!EA3VY5UHbL%qz$1OC4kmxbftxOXcpmcxF*TSJQu7$TZOzk
zkf$A*6f<L4E2TCxB`Q#Ap)nvb0z*9GZNl^i9^Z<B<*5dhA*V6A=OPzGn|SM;*NRAV
ziJLb{HCA8ATZb6$^iL2E(NEV9v2V}k+SZS>N^2{n%+$KlS{!gR35e1J<`%GQOOW%x
zZrXEN=b%A;{0J}Ixy?%XbLsYO+8WP0)*rw1&6fP>_HBaCo3_Qp?UHAqfnV-}wT@sg
zL$sp|8h8~nWs(lvw4fs&FqEM8#&h51Xzy@t6wxi~k!SDe(yo)uL*nv2w?6ID8q{B}
z#wsSNy*A_2R%q0&DH68SF56XZ=nwB5*xrSe(2Q~0G&!DW0Xig6#-t&5?I5BgamJiG
z;t_I6XjW9aJFE1yP<VPR;Ul!><w-fzY8>Z<L!CH~k(@(EZ?^|8tjnfeC=BBc%@cXZ
zh{IYUj?qNxXpKg6<eiLPJdK{IeTdV%vD(VX8_gTu)+l0X;IvG(q(9bYyiu!R70N0>
z6{9j*m~(`&M}JcN+8d!Ea?#w74G}<}gN1cHqw+wSGD#hm6sQ@DwJzL0Jg}}&RI$sI
zY;n9j*{3JWv##*xFn%nSpMBYNTb^1e_{Vxo&j2}>PrA)YtZ|<xmM<}jfAqI6pZ-i@
zS)RSVNY4*sYX(HFbJ?FhVJt_<e~aa3<0Y1V65|JZ@V^$vJ*Dz3r@rjNTcWw;TRo2Z
zy@@@vh3LBm+dvvv71EH&G6E%1-d=!ey*Svc$CzyBFKDern@fAX`Ydm_tTS6)PIvz%
z-g8f*qtIm5wX(XSszdWYC$6=yqeWxfpAHUA<nMtR>r9BtX``=i1UlQW(6H@!bJon5
z2M`Szl&wgztt;M86~-|$&pof+?&`BWdCNVF4Ve}{)up6ZilK?f*ENAA=4GW-Sl1AE
zZ)T)CkR*rp7f?GD+`FtwF~P{731sF{z1j{^{rRpfyu@=-8sXk6k~*U#3id|3AlvXp
zZBZYst=}tiAdypC^)?qSP1`=jHde>wsc#(=8np#Aywr2t4;v%z_k~hsj(((+azRSE
zhjyc?cD<DCy0*Fh`NZ+d_{+vi0qbRavQgW!1e5O}PCT~EK%_vn*bti9W%2W@curAl
z@K{tmtbs@dT7~<0X7-h1dnBte<e<tW$>f&tJaBvUinSKX{FY{gxh*t#6Lgvy-Kf*?
zgbyP(`#s5PBxN1-)tcdIq$z`?K-Pq0YdGCOPzc<yT4~K`6-sSUtg&zrY=owOS7=4>
zV$eco<cWr*h%0n^@#Yo+v-AYSje!ug9W^sZVo(jC@_AvphQl<XdK))(I+OC|F&M>@
zS_<pBfEpr><`9q9w%IdBG7v7zaVl^RC%?5Ue0bcEK2+*z?8d}@_~m~hiSi%*=5INl
zD-ZVzY1|<?@(=&?JMQmKJUp&6l&jlo?$0y#kA)()pnF6sA_Zv`krpU7B89O|LVCO0
zGwQ%~z99{XAs3J?`bD_FoIHwLtF3*L9gyl!7serzk}`}V>W<Gb&)kfr!is?kYC?!|
z0<P3@=2#b!-rznm3|9=p$m)q&D!;!!B10wV%H69k_{F&6AO7XP^U!rS++)Mnt_PrS
z>56_XqqS#6IOx#&K6)ayeLoQigqvL)hFm0Qv|AG0tvppa<+4`J^~~xED=VgjVGPuT
zRU_w-X}n_EFRZm-r_%poqtu1dd;)=-6Scc<Rf#Td^A;3bp&tU{5FZSk6ot-yU^4G^
z?FL>e&7CUFvKGp`;s&ZYI*aC<*AqKEBicxoNJ*&ukROF&jRL!ci_=z0Di-wG)>Cxt
zaY-}O5*a09WC_9HAt&w*2gcn9IJ;?L+>N|@cf!2KjC!XOhiI=$p-RwgQ(vLN{ik76
zD#q0pJMQwH0t`v`pZ@3n!g?<J@ppghhXY}Eb;~&IdG~JRcs_DEmoQj33@kNsUj}O4
zBWaA}{9;<Mb@ls`lAumMc;W2IaaB@EWEse!<T0UA5H}>pyg1E{RiRpfe$Xh@QEQN+
z)nGN8VH=LM!fKVOj;mZK2CCtPi!IE{JC@}Phm~<WAo&YKgt{)gefP-ezy1pk@4mZC
zZ6^*a))RBt4$@mJe7dmvB;#L?<&XE|?O<tQc%o}N$)$_F=~4=iRq<l9);>uri!_Ey
z;|VUNDegbBSdQkLO8lKfqTp1oc%R4Kg`Qi{PZ-M<Lf4lf$3^b@C6<2z<0Hx-Tnfm*
z58^mjYP|?5Pp+D>)e-nOj@ufJ+2?wu_!<StWDLomMu94)9lNZ2Xme0vJ+vlUTPZa@
zLd(=H>%_b)Fl44Y2D-;<JXev9h?5KbtPw@2YF!{5A)_>mS}Uy@Yc)y@0sFNWOKF^z
zGn36^OTkRgjVA4+ltahX3T7+jHP(^dL3f;18s^572Xe|}GFQ`{-9GVQUOCqan#c;f
zK^b!t?5A;H+>OlhJ!@%*+J#oV?a56VjYhR{c^=J$d4-2_<L&*zE2(IP>sN0Wr^;{t
z{+}sJrL6;|0!g{Md&S}Emig2;El194rMS~ZSQ%*C;F_b++u178l+9Otn<T3=niPgK
zQPr?|L@Ww+ija~ZL>xO3`nwh=G}jz)gtK_WayS%vHxy`8N^j;1mK{lSC-&#j7DX%&
z9?$Pkc?6wtcbYv?IWsT!_{B9euHw9`m^u4tgB&~?{=;v*jGyOtDPX;fPd56V1frL2
z#g;9IKXrLGj{XIG)Hks$S=$gvTQ^A@v?Z|2s6(t*a^N6)hBRDIMrxo9c6pCTq8_6d
za@njrtZbWG?<QE#9kJel=t+*y7jU|hSEwY}mi&9~!diFRY0g@Wb16Ko3&(Y3DJv!!
zRRImh#B3JgnT<<#TYAS)yDQ@+Zm+JP-N1nR1`Y25yS>9@W_;`v_0%caiZyOsSo?yI
zA@&mMhXRT32iQ#U>b)3bY&sn0wQ_%4NGanu-@SjwpovN2=Uj~Q`9x`tJf0spuPZrE
zIKpW)iVdK*ID6cO{x>T<ZK2C2u<36IMB>#8A|j4yP(;X^9*JL2YK?G}GV*1Hg<x?D
z5%zv>%#Gr$+nXm^>q-Z`Tcw>oiZs@8;_>{RLo$+1kV5djbx}&Qf`~8<BO*_!u;a9j
zJk#8tOmEG7R$Zc}X6DoB_xo||Pu*5?-=t}{=;l0oDW!LG%~1<G?$GoQCfT8~A+`cl
zIrzxk{)VM2+@Ie874Ze53xLwasB3h2Q%Z;#)~x4B+q0CX!;D!!*v8B=q{pN86GAs}
zXfZn%tDIX<QHvSPDlRS1qi!8Q7}|QFd=V6kt)N3M@H_#-x_S3zqTmFPR}<5as2)Po
z!<ZOy#DCI2Tnf_Txo%h99YeV^BQIJM61r{CxmK`P!{%Z<E)7#>6y;76!>GJ^bw{%h
zz&2M(DLkAWIUd*O43jvmu(knY1PQWP^QEIJwr7|6=UeyRu<ekd7RX1eFY1PRLAynF
zP)a29?lf+g_@=A{-de239cgF6<vv$Jl-s&KTPLe<>3{pmT2E1e@F8ZSR7yQlKHRe`
zbJt6ZcOT0^*Hd}6Mn^1sW^oLERxHDlZqBx&Y<seqcZWAFYni$tW}hsU)g7x1SGl}#
zho50AcNEHH5A2vPQB1;^p;jlE_3p^0isgR2&vJ?F|1YupQy9<RsXqbNe~??p*B4?y
z-;<+-^n}6sD2}_R_m~Ef2&9Rf?k@Msanuvp2d>5|L=unlDVl*^s4eo{T59M+^&qj<
z@SHl6foF*OZLc5-kmfA4Fqg_wYS45$O)wn0gO0Y=#@vjvwXO|SsoovBB?=z)48?ar
zVs0o#k3ybznTLxp1YHuLYv>J{7>0@CT0<|F(3u@_3JC-})~F%!@{6<x{-b;H_R)I$
zS~cdS@^CJU2bfGq<HRKJ#g|`Emqs~<#$3yZVcerhnb$L?bAhZ_gVW+HHR0^zL8-@m
z+gKZ}eYuab&8BVD`hLARNrnUDHkTS=4YPP$O{B;DhL-e8)4R5=LAG6|KLHpbKYw-v
z>;?drsfR^D(GC0@(d{$giP+~j9iQrvQ@2i0e4z{e`yDR@te5f0#+KMUi5mI^3LO9P
z^g*XqsS~ao+Bcux8QKa8>Vh^Q*Mw9!?01YhaN|1;a$u*szIi!PgT8QeeM71<Yb{hY
zWDuG-rG*8td5hvtGg@u9L<$_(X+RWL4gIF(oLZz&D%v5<v7uTsmRc#*IIfNN^O^5X
zN6zz1sfBUBLz9rwz^Iwg8m&226-#j_*ysaEgKD9gaTAg#n#Sm?5F&-yx=u-CmA#18
zdN(a9eJW4MgJ{q0H;8!ONVRV!=B0}-(nW3)*5dHH)F}KR&i!d&#!15sr8)2a@DEHW
z@y+X3m<hEhZ{L03{{E5g9tvw!ZVunz#d)}YWZ?km7Fxe=)LtMgK)ocDTg^iSJUONf
zOHsIL2~-@Fu33grEfSE%_=5u0QbMa?d!{<XM!49DG(k)-LZDYu++!8kY@X-kp5t0*
z`<W^48FzQMWNIEj4B7%J40-H@ub%9=Q`U+MLe7~dO@z-!F!quxU&8oL$MKWk^Rq)Q
z9Ww9n@7j@WYJxWK=3HIvur%QcdmqWRC6&%#1IaVr+#Wd9Gauf)4aXcaiiKuHbZnuP
z6-5~HfK-d5840!9{@$$@bgQ+9X$gVl26&AFuT%#MlYrWcRfNY{xL;13ty0{;!)PHp
zk-S@@b)15E>!{3v_XC0VwZ^+xs}lNOA#$y#hY<51iK{~<?FMol$ywPAUC>&hXpK9~
z3azcZ@Eoix@wlInL`q*rCD1Go!nO+D;C`;?!^)!0E+ww7glQW2=9?YGjJY)4es|CP
z`$zux_B)Q}!fv>MDttH=DtnL{U<EH&J7T_c+j59QNi+lESrW7?1(M^*s8aiT4JRqB
z)=@&?WYiQ&Hl%@CP(8%KO2yo%t<%D_JAgK$c>q|q;+b>=t4cI1oR!mhkCqcs3Q055
zIIu1c9N+(*OPAW8Q^u8MfChqcBfrRJb?AlvJF$HErhi-rHCopql&BXp4bzlX(N7%9
z5NOU`=ArdN*v}=Ft;e!x&nvw317nVOKv^v)QmM617t1z>V12*dt-!Dq)m~!xCorDH
z=by~2w@rAHy&y!rn}@6AApV0m7GDu}u6BD!kfxEH52WZ#;NpRBYRa3#Yo>bU+lTKk
zjN)??R!&Fe`OI2ZAR!Vuvo_?Q2-8V*7DZAa43AKumCEV7a-3)8bwyeft?(!eXH%?J
zPHW|{79LvVxU9^jgq(voDj7$q&&1WY5*3}+y3t3i-w(4!>y2yme%3G;)4;eB9?z#h
z>WeTWWg0U>PT<CnGvjWD2$6#mdE@gvISN=x(<qKr^B&(7@`~Kg3|9lw^}x+^#eP5X
z)!{W(jN^whEh%-(oK6ep^UD3>k%!}oj3ZXy{k_qcfIVmdS)(Cf4P`6nDN02<eLL3M
z5UeLf;k?ptfrNn<utu$&em>-&8;bP;S4<xRmWwe*r*v*bC+Q8$ZZ;*NVKl`yeZ4m>
z+BTK@|4vbC{3s}64zIO_i!7VW$;<fnI9>`^FXPjW4Sn&9m+VC1usKCUf3t>Vw}qom
zrFJci7Kuk+0JCTC2^YmvV;UiA!);|`ATeSRiLAOJEjqq~+eL2*8870dZk_T%FE*4y
z)`sMc6zLu3)*6q;cbt~WZ{D9m5HZ2toXMS>M$F+{R_0Rqurxj_GY{u8r53C;vKlFS
zBw<^UW=~xO_a{vdGr2UU)I-WX2=b2C=uQrjP>i)Ua_{OImD-G2=jeiKK~jU}3^tO-
zF<w(brdRP!Ta02@q5COym!q2cCbup<xvX66Zh7_PmlR)Ub3r?D>2aMY^UC?*%(6uC
zJ5CdlGxJiJYohrbF4u^@0($_?QFxQ>plcv43WI&y)&ATZp=mfbMNS)?E@?G{hI=O$
z33&5TJFP7Ryg3qzNS&%h*{mrS-IJsO5>!bL^a5K3_ksay8Ndru7DPrY?QwIcjrZ@)
z%=a_%vR?LJb7yWVl7$@|Z6+Sc*#AG`@2@Z}x)h>3ArNB7_u@GE@<CtlEh%ihx4GFW
zBdBKvCM<9IVIB!o5=P5t8>pr<2dOa+#x>)ubVNbSxv&r4%oSi=R;m_itz@&%li2hh
zI+xHJEM?(0-+f@Nl~XIE;%J4DFh$s<0c}v`GmlpIwynH7A9<XQEK=j8lCsh$DYRFb
z_h)<Qa6WZ(L(j%h?*%o1(ZlI0lY;iF8mJ7EA%P{BhM_yLdGGdW7b@Q9%9WU=8u~BW
z#1p|um?WAgvm4?SDJ|lyHLmRrhAZw*_dJw^`RzY4pj>68wZ>WskB<wdbD&4_G|`%b
zz_Z>VvP0#7$b0C%W8xRs?PZQw&vK*5pYTmm-vw`-`Y0jB>S9<G%$?PZVvVz{3^D}T
zRxZ7C%$+7dcdo5pmwQlwH4e^#UJkvHaL1ZqUNJYcK4PkvDBk8S61`3FxDNBs`#Ucy
z<B-^mozmUm!k<eV!=H}j@9)Wh>Zo|9kGnOTf~86$gnr`;68lKfkvarr>{G;Yveru|
z+Z&E~=FM4uMzKulg~9&12sqJ<X_8SAXKy4ohGL&Amhl3;kP&YrMBPIW`X!eChmWoM
zC7V3NhEw@|91}>Ko?p5}!Ne@I%B;IhK96Hffr8H(=;d7=7(EKl%>o6bEujdzv`1WN
zz7=DP)hos7MK8+x3~nA2<GNO|8+n^r(9!l$5jih2zyJ0v_op+bWg)wUo?PBBYT|C1
zSgLb>UU|Em_)yNAmqImzAqBE<h`n$_83l7Aw6B^MLIz#Bj;>#HwYDZ)Q2#(MhLku=
z1N(iVijwn=VF+a9kb(i>9Z41Ls<mI&$9uACFP`@JSz74*xfs%nb2Hwb3jU$;-Wmgm
zFNcxrQS7S*by@lF{sZqnJhC<*4J;-U2GVego6;=dejk->t5KZCdXri%s82Ks(4oy6
zuFV&;lgGM)_d=j|lw^yztY)YS$)m9k#F<P3Eg5>b@-e~@KGH;c`eyfi=_S^SuJHva
z*Vn@S6VMqjm-pGFE1q8NGinaMjK5^O6tG^#rytKCsqr;E35vx%=zJ{}YEQk86Pv7y
zJqdNVlT1+`L+c_bLx!B9TbDhfHqgYVG?IskMD|STu0BBs7=j+KL6bYw$K9#51Yf&v
zGPvU{ti3;cc;Nmt^M8K(F6aP5=1tDrro>^F5tOAA&ZTf)Dv#^Rab5A&ms<_ZChZGy
zXauBf+3sD4P6OOp-p(Z{)Xvp!U9>!F0*BHHl46r@-l6zq1#xm}kR)K(QO!XJf!WaV
z=mo6p@pQPpOKfmWg9bnWlSb8*G>u%pdd=heKeDU~gE=CNb6Hr<Cmuh%LnI@T8OCee
zMjlU%wI<AOqktv1zzh;ymQ7#e3EbY@tmq>3z4Ou}=F%q6DFkwk76ngta2+rc$_jbt
zh^>BIL(hL(>)+Ezn&O@W>T0cVFv*g*XixTjQmIG83qEGhib;qK)0~I%BM;?)<>K`f
ziTT<JqxGe>Z|=wwy!xj;p3@U9Z8zd`c^1bI$&hyBf^XDZw+ozH`u`&`2Q=3}(j@~m
zG%15-q;E}TPHG8tVOJBiCcGmd8<lE71*XSBvqo(~ZE?s;J;C3gh3-KgS=N=`|M6QM
z%fg5C%$|YKGdJUbDGluRyP)SRg_EtkvxWO*<}{zNOsq+oDAK5DE3Vo$OXI%r1fv((
zX0UZil|%tqghbFPRC{tMLJ}Ah#&IAeC24)4bak|6vlghrlrD!}zhrNMle@67vU=C#
zs7Bq8qAp0@b6P)8P7Cip+%vdylM>b{wXV#oQHtZ^*XS@(tqVQtYeYswC$MAG%)0RF
z6M@Cq0?7NC;@!;&*FczvfyCO0=1xtD4(@7|)hn|{F@DgzZ3CCaNR!^}-4472GLwEE
zWjh#c=e><Wx{3I<763}(fYWg%{RztDIW}{aQiI-%cG`1aw?88e!)49+ek?oRi{&_w
z##2<V7RS(j@b&Fr;dGjNMst3O<@DE$WpN|Rv#Zz+h5j>(Wz8WSuq8_x_t`^9i%MuT
zsspusvRHPHgKsdPwzVYC(LcRdwtsiAyzR-~s}KAjw~lXl|NmCRvZpxSno7^(ICa-e
zEr`OD2L{U>oz?$@DM=ECbYK%yuO9YD#XBw6VOv4yd1-5nRw_w5GU^h_Pi44yU3mBY
zJ@1Z39_NKg2C^m&JL51;+)O))Imfwhf1Y`_oFb1GMFvGuLIw>-Y@QnE+VhdN6+Amt
zmP?Lem*Uk^y3@#MWEwNOQ7NvB<H#^1#uOAn;8_7{i{qz<^Ry=e1QoaftM$E7K%29e
z^RQIx=y0CN2I>01E@cjRq*<d_<Kf{0-@SWJ;|;DOYcnbXn%|(!X<mVcPI-=xO=h5P
z#@iwmrWZtkUOjif;#0zBM9$P4%q>LR-8-8hoTts7HAecQjPJ$Qmm=2IJNaIj__7sA
z=o-=~m><38%VeiFx1TI6)V(8MpU1%Gl3l-yzht}=uwKTe8^Pq;M2BMZg{^gaYi+5g
zf)<N<@Az61?#~aDG;_VbC5w;~RE1^&1frl>L)Cb-&tQ#nc}J^4tHK}ygN*ohM-mI)
z+<eK2!Uqc36WJz8S)wbECPW2U&ykD{J8Bhr?TJ>6(|P4Imq?n+%4s>XYUZ5`{PN8W
zV@@2;#%XCB&nt6tnobOAB%E^zY}tRj-SbKhWZPpc6NsVK5C75GZW6S?WGVErG9W{2
zCd{$CrxYA|3wXA-%CbJtbOsqonz)t;PfpDXdEAk8V85S&7I-{kwUKLxG%gM;3z`)(
zVRd1(f>bA0N0NlApbD0h-Q8<ix#sltcep@Fw_Lrt<(oHOaXhb_=E@&`|Nmp}&zmGk
zlI%|K$0DL?X6_zG9(6Xldj>NA0khx#JM1pe4(%?@%pMr@qR~KIdBoxFW~M44%=-tk
zsF_D(R#i_|7Amq_l03r?JJdvlMVKEyehl|#sB(b)3tqkadzNk-TSH_;*As<`Nt8q2
zj`28udXT>9?r==4ILMh?6w^{eb4Dqc)^Lg)oRXchej3JV-mRvDc;-oKu-)7N?n%+%
zj7OwY2hbwf9UwJ&TUmPL8#RzbwXZk*Gpu6NxMFwx&0wvaHW`pSYZMH!Pf0Wubt@e?
zx)jX@S{g!Eo+a$8n9Uf4)iSDyH+p6-%E6{6uL&P28IxtQ?T|c?Cu6@;L^HO8jjNJn
zB2ic#jb_5j!wVkTnRm-0=dL_1N>bNfTKWQ_C=~=R=@Bnm6nt%*`<YWab2jDF8jnk*
zYNMD?YiD0}+}+-?it_cb^0u!W`w5jyKHQO4XEe6qjP(>{4$aq(Bo2DO5FV0|l54wV
znc2yV<`t7pu{&SWsgUG~x}P$_9#c3ztcVEvy+B^*^O<RP@D86cxwc%ZKxYyntD4B!
z*e4~K>%pCL#nk-3FJMz*w?FXm_9f*!v8ILl^;eW+eD?Vrvlu(+eEa@^_m5|O^M2yA
z!g-xhZn(Mm35zz)x?l=jQ)mK>go8nJ`ovO{fI!a;!Pwc1h*4y8JAz&{T6BlrV6aps
zn-Fzv4iD~=YZjTe<<_LAdO+)bvIi`#R_tY1BHsOKl#@tu7R^?L(=9}4(+9w^2Y+g=
zQ&Q3m9BxRaq+p1Aay^gCNJB#jw`KU)mi<KbE^#XbwZa>#>{Yn2Lo{sgqrC~tmONQB
z7xD!ANumE~Y*_^6G*hA(#2{s#+w$YG<=&uorsj+dQlb>&kc?slvBXgqOr1mWL)o&J
zfhm@91i|M>1pINf+*YA$arSNKQeuU@-H<HNLf~2esxP}TXS`D+jIy>(x9F;+brP31
z7`B|A9HjO~+48bBrhEf@nzrmQy5A$$YK2giq)V`TdUD@MJ4ymFHF!n$y;4oM%~y78
zLL+fV6Do;(TgU|tlRy(%^*y#E**X%|M`u3W&IgWd;h~+;GEojU$l9Sp-+Uqmu~PMB
z2D_5j=i<P0OQ^x)>BzEHP~Uet!qOY9b*vw;W}I(d@>m;hmm_OegpA?x6S;VaoGnUH
zMRKC4CpX0!#)5?q*d}lfOes+eCiV_GXJ<=JyUaXicBQb}Wg40NVRns81MAM{9*J<(
z?$s8t@Dd%w%l4BvF-)?6CXS0Q{M}TzJxsj*=}QhfVLwCd4P6_*d3Vo*a+d{djk+$J
z7LfVC>0DTQ=CmH@+#uY5tf;I}?!0(uAR}RFF4=(FhL}Y6cj7bWSP0#9({ODq)uE*Y
zUbhgSdPt{8X!&)iUKY$HnRZdGiiwWoeL!0m5Y51;<lrr9{W%}PP?ywcXmT~Lsvuba
zuxxC~=keno&j8l*_~Ax~g98^(e?#^oOC+*^4$i%C%@z@%wn~odoQXuX6W3jc!<*=)
zvMUKyIO<Ao$w98P_rq;@M5J+ev2(g>T{cBa>#`kaA*bpIS@HfIJ<<uiTeOVzPB>_%
zwuS*!pc-TnwYh-%xpt}n$+OZ~TyLqXahNjuRA6<wvuC!voWt2h=_(Tx+Fv0=ZIR*i
z8v9Eo#YLKprLWYqIt4sW^tN+AC+Ws$nx?|Mn@Er19rDb}d|KR_kufo#Zo;+3lqRO#
z>}|MnCZ`>eGhKWeS$p9yC-!^c+!EA<rcisrrah_Lu)BGItqaSkBFR}dLa;-UI74R3
z9$GA&5X@B=g5ahG*+SgCc-epqnsk&cW!*AxV{HwGxQ8x}=wO`LB~vRy$er!@#YWe>
zPB$+>Bi5}8P=;53#ZL0I?d(0z-Ve2W+p@l&)bIs#aX!Y}Z2B4_8y*z$Sifg;=TReP
zFT}DHBXws;<&??E*q4s>L`}odxnu)dRXQ-0i4uBkUJ?RXoi49Y3SfbnduA?dTRa<F
zTcfE9>q`iMHggapw+3Wd*z~b>&b?84!zLk@306F78x(g_LDWJU$AhE+3WzxHN9b}S
zDc;phjL9YvA^JRulmyqG(+rlW)q}(-6)bnERU{YEv_n#+l;WjFssWEWpd(`B6x4hX
zsyW~zsZgfE?q<(!w`VRhZAviBNV)OO7t?`KjN76vT)TFROw)qO9+Lypd_(Uiq@Low
z``liFDcSCUgCc|D;6<kXh~c8D+kN*gd8)BJM~8`Q3Di4PrQ`y@F6!$}>sG=lTM{^8
zRo_p1*<PGl&-~jkQGa-}eYfjhUS7!e&_00b_)KGe`V^DBHk5zhawo2B`JyxBw(P*+
zVoKWNx3?W2NYVi~=1eZIFAbH%Qvc<)92}?ZObj@h+Ls@xE#m=va+X&~UR~(=MLo{+
zAKI3E@9?>H#>I#>f4?m!x+Yp5jc@!*5OT0Q)GpdJ5(YGy*2D`2fCU4O_J7Ce5)YG$
z9d7oh!*56*#Ia$^{5Wjct?BxLebR<yd-<ANJ>Gq*j!;#}(zj=)*scB8O=K~qX(ktA
zmpfgB#@I7u0~K3!8=Z0?TctNoJdmzf!FJ`txv*Q5p9w{TEE45lN^P#Alye~^Z+ze#
z3ysz~t+?OYOgXoe9-X2b%w+q7qj7^s0I}x4^WftRKiBuJF<+PcCC7WGETLQLCBZoh
zC3}gmsp~Uoj|NaW*72wi@Jeb+?pumD%iohsekKdzUB#!&?TZ6*PRtp4S+Q0*w|7+Q
z6a{T7eOc%_L7HfmSbIj<W9&(Jc0VgVGf@(mB(&-VyGndVO<cLFu+8Ci2AG$C4(xc8
z>$#2i+ENX5ya1f9Uu)QvjdjBYW%*@n@!&XPhu<G@9m5Cz?zO))6uFk<81BY*KI?h>
z*v2z}^*sJ8V>rSaATUB%8E$VOn~+cl;q1{mRTELG&887G5cl0-qD(ODJG}~}78DmU
zZ!~v0i&CRDd+n9?4-ed*9(h=fv|15W+UdkRWe(GxIRR}!x*`haWubL=`RWzB+M(=u
zIGuQF@1uRJkqeaBIQ5nH?aWvG1Szpgg?*Wsa-rL|NP<-oXA_R9q%O`l3A+Ae_-3B2
zw{h_UX_qH=VAVE_hgCbRSB||=rIT}KGU2eUq*<cY-n$0vZx6^c@tga%o(Z)m2S|m4
zdg+xp7$H(!hobEH{L3%-*Z=lk^YZgAxV?MH<LNzXSGvutOXHv3J@UJL=2!iZQ!^G7
zDF2$5uV1nI>`TadOb_UN<MGs)N@A9RHm`dxNy7Q{Hsw7^6FrD<s&M4Z!ikDtcgqer
zprzehz*;<$EDA`1EYuBzOrZx6z!P}pht9wS^QMGQHb5&SFKu)Rb<FFPZp?Qse4GW-
zAlCR0!TlGT)qjI=ZNo&y&D-8*lK`ZIQ3?tUJSq;dCK+owknQ63KFt$3WZp3Ex*1Z;
zw7z0=TI#G)v237;oF}HKu$wdW0eUqTqD?~Y3nt#aIt94cRp@FQk4Mae|NfW%insR<
z?9;^Kx^Qll-gnfr1DR;n`1W|uJ67JO%8ktIawe7P%?ETv`ii!NLSarj*4`bk?9ojv
zm@cl5nn`JoCc&gZ^@vo81L&QONtECCPROz@h{0}3H0?aD3zlcner12yg(KW)t<j=s
z*w`XPT&Kq^V?}b|)yr4>^q0TjfA~NBJ8pJ+a+#r=u}QJ{1;^#g|MtIs!>hURi<`<<
z?;d#ju%P+hayWd(t1tc)miP2@!@I|o$F~c)R3>St4iLkEguXS5Xq)rf`f&0uK)Koo
zJDiQ%#YLU9dx(~pw<LmvC~xl_1WI=Bn&!)C=<!H|2L5l~0@<f_JOJnM<=69F!x-U@
zGm|u|KsQJHQyhO<TbAteStg-OLMr*$mj5bkIjl(}k~bH&AMptWl;pAwp6W<xawgQ4
z`Mf+wm;f&+95#HdrSsI5fB!J^?V{O_#g=9J8Z#h>vyUy2bVx*8372;Z-KwGWCP)hC
zE2v4+l>AI4p&{U|+G%}7qT{3X#==V4We0VuOLMHtLaUYDD!Mvoq$L_AP2B+pO^8bD
z=atqw%lXVS6@K=!pON0~C=d5MJU-BBM>Laj;pTQvdJvB5i5ewtH}fqi!&x7pH^eHU
z0zLHmw1H;CN~{x4oX#>ar5!3>1<<;XNbJ(Y96EM6WoIXupxw3JR6P-9xngZ$#|<-y
zY2I^lryLjW%4q!xVpNO7l3`iVRD(qdyTgvZ|L^_|TNSMxTkcux3<7h4mp40TC+z&5
z_07WlePgLJRd2a{{kKTkq3n1#EgT<KeAf}WC5%G1$&<Z006axaBA<GZQIT+dRiRg?
zP2GmIhmBoU-NsFb4Z$`4E=i36RQX;I%`P-wmh1PeV`;w!NgiOB1L)^^CfNk@eoK$2
z>i0bNdHmSMGl2Cxez@_$9%R?Sex%C>CO<mBj1BNQBNlnR3&Ku(Bg1)eC))<<9}ZPl
z?-tnUH0v~?&8tQx2!-zLt6N`bd4jaV+v5!yD_e)b6t(#KGO^|w*X^5sDZ*Sbxg;bd
z8bjo80CXd_Lu45xdTxXud1RnWT%*c1y^X5?E1a|xv(}=FW^{OJc#H6Eo<TxfTyp?n
zzU3RaOw5uYCB{)cvSs65upL6`j(AS&ZVr6$lb<l}50qResnAmaEp)YL@0~$Ca7-fH
z-Mk|0UekAbjEUYfvSR*OL=3YYU|C#)MC+F5bTV$cAltlbf4G{O(OqtHkgrlkmsyPO
zE=wg2?W^C9*S|VMH%68X8Ock)d*Ld4-|*|+I^2vz)2Ec~y&hLImu*dRVGMp`$D0ku
zyj^9;wnSaHAA>PB2yIAoK2v2yS(A${n}qXi-GRuiXfuD4Y*ekZ(IO|VL(Ky!%mb%H
zVv@{0Pb{e+(l%D(G-XOnArRc@;-I+p08_CLa*j-=jK&Dg(bsK#U)j}|BgusCI2Ito
z?Q?wzWv!K^ue9#G$gbX8p=+=QG0HSCM?(mIy{9%oWHXIXYGaTS^M1$c*RQ=fgb69H
zND|kL>%zGTH^F8}=|Gtalv}LaQ4Y75OrAqXgJ<VoNe1IFlyk%|`sN%DP#z;n0qv5_
zjlT`3WUP<z_xQ;<UFK(rW7wBq_Odhh3LN`vsN!5L-)ryr+H$|wv%XC0b<ihxvgP+Y
z=P!Qz32oVfV1KO(>*Cp#{~B$1Bpdy{hPF-oZ>*nTZD~Bswq<g8{I)&^nEJ=q^7Xm;
z{ebk3$(Fb4-V(y0ox0g_{CRjmaqot&bP<JT^NVcQaeS{DEGk@wRZ)637O6;zwTt`R
zwKW$s?;ULcI*0wc_UpAjcdgIm`^6|}o2MyQx9$t3B+PT6l;~P$zG9LjZ`fkCl~!Ul
z+)SU7kUba)4qtbkpiC*&s9aS&jI#4IVv6voyP-s*0ldLe>Au!;$xPD>>v32smx<-#
z_a>0!pS5YQ({sj3CRrin6Os$aOgH7U7%nZCLA-=<-rXVGGVgCdN|bmSczYs1w=Hwq
z&951Yu}2R-bg-->n0v!ME*hq~S=MG@B4Nv;Sdh&d9)5A0(7qCfCh7I8Y!@lpb^LxH
zdRV#7ea~g{A;-D%JpMYyGl2Cxe%P`3JC_4?b9{7cZIF@p@+XN6c*p6&Vxzv<A3ZY1
z$plfT>e^UFMiQjSSWah7t#f9f73FMRjW5YXRd)$`TWM|O@eC$=>g^p9&}(y1NJ*fI
zwocOuCfv?@mSu&G*V(J_Vn6ZncE{`68&*?zcpp@M*wH=vlU>JYmnSBhLNwC9RJ}ur
zCGUcy-NhGE215K&vUfE~>JF*a)it6-y@OAJZWA40-rZ1gVwFz8n9Vh0R_)v#?%0dd
zqgA}^__dQ0Ub}ajk&b21opAS)*ZiA*_iuT)KXH0IVe=hDXV6Eq7c?7({R@7v-*Gmh
zn(+GP{{!q^@(=Icu~en&8Px?dZ%eMW(!2jIrLhtm<dW36T#`E;O<hZcC@(?;(@}mD
zB`0<Ph%7#bmn8{3c`0fV3s#@ELH32HS23(JKtzN*su}_;HEaviwvhV|oSEH}rJj+J
zDS3#)eZpg8ZB3#HfJ7j9>EM%5lJpIW^iESk-^wSQNtZ;z#-K_=TVfrV%OupgQk!zL
zPR)%zd-i<kmGg3DIX|H1$n=WQrSpDyAm<&L6P7(lS=9q|Xc_<k002ouK~!6Z&r)a|
z9v>dDwQ^wMku^StuZ@>4_eg3?(n;o>g_bqZn{lJg6`E8i>I{OEys5&{U0baOK%R3Z
zs~10Q;;@n+5hb&}r+LX7^xmlJiMQ|G@P2s@K*^b1^M(sO0&11c?r=+)6TJHhZO%G`
zP+v1J6`DDfW1c+-U8Ny&;>F85tR-l|dDm#Qb2>HF1<0={bLMcE`SM11CBj4hDb;Q`
z&wF}@`mlJG+d5(uSP`iR4KoMEEk%&1u7!2I!=#iEjNXY7=}g5Uz-bK46vk(PO(Fw>
z97gGnAHY?-wq-;zGFr*@%Rcg92PCs0z_C4o@3&3APYor<_Twy)-(x4A!uV6!a*D)J
zbfsO<u;uh@%YS9I+_h1Arw`4X4Wx(@Q$`t~vCIBiwRIQOz6yF7pqO}4YycK;v1MmL
z+lA>R{o)_|p5aNqACWB&24%1wClbYka&c-%C}}ijun^4`Z+0?7793zPdUUMKGgMWM
zt2btloH+M|wRcXf({qPA7foMOIoFlb`4L?gY^}7%<{<SUvml)EiCfu`O2R@~V3gvD
zLR%~M$1|95*zI|r9>Xq$dCI)JyW#F;&tY=*e(%oy$pWbv5+!|vf9>{K$lbfACi8Bd
z<BSmINalL=!DY>f>s6%$l1XJIPkYYi1<aVHOvhN)M%gtcl->RSdE(p0H`Lw(NW6eH
zUZ!nGhG~M@mD5^T>X~(2DL9id=OZb1y3L$UXMXq1*BpxSa_*4sc=6(f-Isq)PcLX`
zPhA`LZ{K0IxKFJojH8zkM}!5qKkpav1h{e?-R>olQDxvZ)m`*{lwMkjJ<g>9Bwc~d
zp>O5;pT&CewU;P(m&+XUTQ5)a#T&Ym?#q9$u=%bt*|mrpQ_j)pc$8j$%0~P={yF0r
zz<M4(+_>6DTpPt55ECz@%96OrGnvF>nWRE735AW3;~m{5u6j}=iI%T2O1&APg0;I7
z)RVGOX<9iZKS1-Gox;j+ryMxyt+CdXQW8y+E`W|~!F8rlSbOK|ckj4A9yzZonGk?X
zFlFJz&4FX@?D7;{m5i?L$VqZp(G;&M11J&K%ed$<xFhVDFJ*{B=BQ;)mNC6NR&ht9
zS@+B?#@gKZOmoC=+Am#{(uK7~Fsxj{T(&HW#eq=fjGC*S9G4Scef>4b+*zPDY;AEs
zC)T#o#A`jZDNT*@@c|mH7hq0Z_mT}y_xQT-K;x2+4bED0__M13bLe07-jQUHos06w
z;ly5Lf`%hJ*Z@7|AP?nN87an(vr}GI|6NY-bq3gf7W<xKGwyNkjdfj_(~M<#@|ixZ
z5fN;xUkN+jr<utTJ53DjA=g-lHHK#`{ks%PJ=lIKX!qWQ6z>vPyR#+Klv7_=d*f&>
z%71&ir%K1(k8JM>twD3G3q@$F(x-%NnN{1mC23RWiBbv=rxV}YzvZD<mfpCT3R4lL
z?Bc2~Z*DkOp;Qeg%{D51mJ&?I*dVTF$*xzRY`-&e{SVLP4xKp%UaNMFbtRqPw5}|z
z`8pNcp2aY6QP@-pCGSx4*HIf<9^n^m*<;@|r?BItt{jg?-o1Uxt{j-t4zV7kSt+`A
zHn`bwbMuPF_wRXkI?{I)%`Moh<FkaNhN$lrUlRA0jE(ZWYb_QtYONoLFN-_NdTF0M
zGPw?51+yc5k>j%teU35zzKvY09=otTKE6L=YnShM#E0W@4_oU@lZy;r-t-e362a=0
zTl$yUve)`si#aX&Y|DS8wmkG{KAcap@p}I2E!iZ(D3Bxt5Tzi&Bp>JS^?Ce}Ul3hp
zu`7PphuHL!`^t~RmTd#C!1b4kWNyliiNw@hAGUAn)h^6^yQB&%d>JgchLp;N9jLmn
z_<*!eePL;hBNY<7cyUV=#ojuAy{;?LD$*OG&E2e~SQb<pW)n7==ny_~u(ia5X_|R|
z|DJcp2acx`y?1Wso!fH}Zgw-Dy?lW_oN4DuYpV;Thu`RWYJUBr;lnsfl4O!)`sROv
zC(~S`$^g_GhD`H}B=3Z})P?i;gsu(%O5W7&HoSD2C&+~)8N}y{U~r|lrW7boDuS3Z
z6>F=Uk4J9i!aVKVpX|zdt}N%qvUG0eJ)eDcN6&Y3x#8Qs^ZTlu9=7Rcp5n#sQ8F4f
zD|qt*x4W<r8(LSMB!@_B>!lm$FtaU9WFE`O@YT0{Vw~A70kfIgSVt)?EzafffUy=g
zKsp}$hkz&t-bTFQFF988A3DvS$B$z?16a@F&oUzX*bc-ZFiYaq?v7&F69vt%-dUsU
zZ*!JyfDe8c4IKovjJ2S!YH|_tR^7Rl&e6`Swr~U_C0_4uD7J>P<?VD^_rtF*&URX*
zvloFJc`M%)p@Yncby@gNzy2qduB_d7G0#jnGiTSf`0V8iRugW^8&0iq>XqI*wHq=i
zIhh}zIHSzFEQP?p$ofPnf(6^BTXovC__m&PzKY5QE1<Jl^)g8@&dZsW6M5dbf72q9
z9EYhp+=|g7M>Ql_;{cZq*_fsYHQ`(vZy(<A5C8BF{OpS_`TXT;?7YyA=S?f>v{n`q
zdT!LUvb09azk~S^+kXa@=^@lSMCUQ@B4iNt`XTQx$r%(uS-(DzRF!pInI>^98=|B#
zL_#+npSh^DtF-v%Yl^ouFPY&!eUJ*;z$>hPeC#j}&}v)T;@2_g^&bvHX}xhi8~Y+q
z{_x@a7dM`G1d)A%5Ms@|m=8>rnVVDWTd*+9yo?uhdvnK;jm?HS;Au6vvos04tAnA|
zY1X)JkDS}e31cdSpS=E(Dy~^lTV+{K<f=$js2%OFo)gtFy|~tcadFaY@c?hP-?O~^
zmjC?Kud!Sp7hcQ<_LB#_hq>_O>sKs~nXeuK09M_~9_BPNrGi9TuA#jkBGK0SLaT>p
z;oUoF$}UKqvIqFO&{bKrqNm1bS-7tYOK;4Rcjh|Q)%U()ND}h2<8ZiPWLDLTBm)#@
zbK=Pzl9JLy>+tY+<n8-+{N`7`=F88%q`Y`d(iU0yjB4lce8S`vFJ64ffBx-Te)Fqu
zx&8TT(oJF2m2PtNS^aZl!k)Hf3!}(tl<~#NiM{;1E|i>@N`Up|flt%s0IAq!nVaJr
zpVQ14v;aNWKq=v%zQy)-owdEfr?w^#HeP?AH=&mYSA`NEkKT=Stqw$&4+7**aEM%H
z_fKQX8EY|z6RwBT`m-(nmD#c<sXAMmO0K|SbJp!>q(TZGW&;#2$d)Iv%2k5N8QR!j
zbOjJ?VC0SY!M3W$J3m?Vz>mw86Ho1UlFaMFt#|od6SXzY(R9cWiH?zI*2tF8;nPUk
zvI5qEZEmi8b?z=SXa@K7$f>P7u&|qF{_cyPQj@{2!;a5qN?Xajk(wtj`{G+f*M!*w
zOk4R`Rqr&|&K=1!H-{Vk@pr%BpMLXe@184}-R%qZb7r1|m$!R<@x|w?{SBw3vDPz)
zGVcmmTmW7I5DrJ1Q6raSrjxw{Su8sBIai8c8uq>-cy~_{8(94C_<;1}iPK3oEXDu|
z?8=@=g_H}fADg!GCznFb;nS)*fTiAaVqI4r-@j-7>Oh$j(mS>`P7e#`)57W0nZLN@
zZ~o@Lr>8p_dmg_&b6(t6>KbgPC?WK(dP%TGIeiw_w^F$<%@i-YY?1U9f!2)L^m}~T
zNMPfnfs2(5qQNL%lsJ<w#p$D2ich-VkIViu_6koVNYtKxd?t?#MWWG%#ricy3xntJ
z*D;;}tmpBAkCAN~!9_|!=`fRcQ4UN}D3Y+M^xApo3yU?9`i3;Jc`AffMPyn?G)F$r
zwbNSX)K*-WvvIbSmB!g%lEUsVb8eHjX~sxN`Nfy7Nvizr@ja75(e8qc6a<%FO-Udb
z_mB6ydpz=fJyOcd-HSWEINb2r{=k$3dgZX2Im{EKc`Mzil#r|M?5Bx+*?Y&9p0VCR
zJ;Ob^EjdMBw`^jII(rq*NP9M`F<QnpQ_k8PyumX-Qv$SmsYPEg5lT*IQ+jQjmowm*
z)*J_$gD?)n2;vSMMwaZE&~Dt_-SGE+_piv!sK-ZoUC`b@8o4MpcQY^cx4eGw8E<~|
zE53UB7O4We8~AL-#DykPF6`z5%bK7msueSL>M`*Sl9sk^iwUAUz*mz66T=ALNSzho
z8Ju*{Z(tNUcwpvCF&Al9Tp!HaoyS3G&ImGBF(s$j18M75j`GLZ`OELV#{(7z(`sYO
z_<wTV^1%4)`WNmj&ysmD@0lsgCA%J)!nvM3Yik1mW&@;$Rz(gfznLN3(3a?3IqOQJ
z(XDf~%9+M0&bZ4Zk+Tbi)^+9K;gP?8@dYoZJ^fs1R(Zd?!*Zu&=M`GFO$)=43o8HS
z*Z;)V5AV?8r9Gw0eEIq_K7Vn?l)#Gc`DZV<M>#wmS?I1~W&>oMfZ}UW@HSz6Op-fi
ztt+j&5VT;C4UR5pmc4;Ok6;ilX;Eu*Nm!c6C6lwDDN~9t?Ht8#RcVuhJsF`$+E|6I
zCzr^ncn;R0p?p_bTbbvHpa1M<B!lzup5^pNTUK^Emwq+L+}s?PlW|iDxAV;VZ@#6y
zJ#zTj9cB}?LGKA{3Ks0(embBa;u-uAfNq+8eu2)!#E{7Y|88w7|CNpL0=_qS$*mZE
zPRRJHZMKVD$`!D<4)gAY*ADpi+3~itN5cBrK4rlAzCSO^%3&^7k;^CKGG8rEfQxdq
z{flk6ga|-4MY_^kqW1P|%YTiw9HP%+A>bXX^BfwVm<3A(MrXJI*qL4hp=Ox&?Ia(7
z;{d6iu&}oBsC;F|8^7o(Pv!UYlegvZ(Z%Os?0<#CESa0UW2TVFw9=V-=e$;0S1t+V
zNbK1*!!G<zf>8>*G-~s5UJZX|rO`w$Q?`}s!TEUN+c)3x#XNB+g}?cmpJ8hJ?*3a!
zG8c9qbz%mQy>PL0F;2^gckds#KR(h;$thFH#OI&A=Jm@vl64Nd!cRW`jBnmQ@-`b9
zdagq<!nNXnB=p+py^_k#L67F_&21&Axd3@&<;;|3P+Q~u!@Va|=)8Tr=l$u(PG;X@
zbRs24*}>JW!d!|2+Cxh#rK^NWa)1aJ^>-qPBEo*2cy;%JWXkE`J=RvTbf&q`s&`Jk
ze|Y3~zxf&~XEYr+9ietX<OU4ts`Tcd@0NLwy?y8qIuM;bLA@np)28<|j8Ii+t$CHh
z@TJ4&773Q-`1$&*25fYIGz~pnhoCTAIDP}E;o6Q>uh_s>%lop{uM$%6OEXV~*RD*-
zfq(n(xzFQAJDvfo=kddh4|fsw*GS^I#)5E@XXZRHr$SGiCd#sQYV*S|4Z1U)r5#M6
zOmR3uBd}-=Vyu0oTScj?+Ng9Yp2eK!nNo7>+ujcR^OrB#=gj-lF+|rpV#)QdVgq(f
zL*sZl@pwLR>I=8?#LdltSGRY3ez;{4KpVSxW-gi3U0b1~L=ty+r(D?OnMq1yLOk#q
zy{(MMS_aZ*CZ3hVAVqncb`nurpZu_{Ay97ur4*zjT3sA8=?-?OB~;bRCD&yGcT(gm
zBnhdEJZnaaALbcnwy`!2`#nGX$xnFv=39<$-=VEK;3l1vm0d0zZg$+<-BDF}`*82h
z=yGO%^M$u0W=~3qsZ6ArLTjs`*2p;)%vC9+moC`l(0j79!VOaw6D9GH#?5eKvm`tW
z_LR}+mJ|eVMAX}*o5Pk!M$&}QFl!$LyyFUNQz*W^rjLihmhpemrR5VEW}APr*>)mP
zByltC*vZ7)GhLxY)>eDa!vkP79E?oEaizG!6}7<Do%OL=qfzPBsQw*YJad?mgo6lm
zt(+cD-2HUs%NH-`$Hr+r@%H_<SXYwxYkTS0(5->mE5CX36_0hrvXDz=E`<OBLH)j0
zFYfs4<x6rhN`{v&Zn!-y%u=GNVmMxcP=Mlz-C#@-KHzxR&}dRqXO4!}ibhGTV=tGG
zOmXeIj-aVTJ8!`xLxMc{8t=_q6gpae=dA_3M`=lt+p!F1-fvjzjZ!k7fA*TkckekK
zAE?WTwl=0|!X#rdvD-~#Q>L8Q<-+5)?>Os``DI7)MDK!XLacaaI#pDBAA8!FGdVyh
zQ5F;nN4t8vb-diD^|qB28z1P9#OD%>EF;I-7?x)vI3>1-$SynP9q%!g+clWQ)1OBG
z_o1b`e)z?)ueDw}9erw^vR(ZX0lfo;@-Mb!vEoUG_+#%vQ-T@Xo^AQB$(F}{KF)3g
zQX)$M8>LM_qx_I%ZPzw8c+xG-X4wMor?zZcH$4w)4B4Lh@#EnWx8?7&WB)!&qDW?+
zCgwCzQlY6*RJ}J*uWSu;;=gXX@#|cTR)Qg~G;y|?i3W>ZX{MXcIlye(1!G;#+~2?F
zvsW*f=82zv{u!9^aDI=8gGrYXHU}ZBJG0X;mUZFHy9dtenT}CXqD+NXFK&5pb0E>#
zO>W<ZU13Vj+c!(Bkpvl^1bpx6y<vIwWO3|cBM~xEoueo&3XJK!@pz1i4r@G|P8`=W
zQap*$OexoFBqVG+2ZQqRm5Ai1vgvJ!NW??Jlnb|q1MS>d&yQGd9y658TD)uN>2&7J
zw{Kx;SYA2nM70T-Zvmn80IJ4fkPb4c5&ZWf$%|b;tW8`!u4gMP)qW}12$_xTKa&dv
z74sFnpSH>h+5RdFo!M>M_!ymHG+i2d&cK?#n-P6sXsar%yTr=2iF_CbJdYpccm}YZ
z#}79KTlDHziA<8rp&Yo)dtT2s6!FZR<&5oxW2;n*!#q=@1l1QTc`=yK5%o+{nLLZP
zRAX)G2URksr3-X~yepLb#8fgldj`aog?|5@$5v@&;`PgyNEXh^k*Vxpp1fwiy9lWy
z!IEo2{o-d|vM+mH%{yN1_uTDwOmikrLa)yLN>)SLt1xFMMJOWf?6>9_C$nIr5t#%-
zNB`_;!YBdI1&!8RoYc=G<FqcESy^aIC9yAi$itZ<x|giL#LKexThcl`S1(6YqZo7%
zzt&X1JWDiAg5Eo=HcpR^tjpPNYVSyG)aKeAOFwb0jrZq;Uq2qnZ;yQS@Sa7K`{R4s
z=@A<2_n-6gFaABPH@cqb-JQp-E6Q?(PYvhY5b**<6Q7q_8>JL>hn;7>aOX#xlBPJo
zA{&__AP^nSa5R!O?_%YLZe*g2s9ch*6}mgNf5@xVtMC8tq3Lkt#ufRbhFo8CboAMk
znf<iqF7KJhq%LSll$1E17pnRF$~k)%FcLk2Tqn<TGLW)2FF3Df*4{B|?t~>llQTVb
z`-$D&Tdil5NnklGeBB$rnf?p%@-6@R@Bc^kHS_=d?TNMQP*u;C<_xuB-PL=TGA}>>
zjJ%%VRJh4E9P*wQ^NvIDYc)~IK9O=q*AsIo%w;B9_cGLJ$5du=cD6&aaI3^M+LTU`
z8NmYxkxrKCOi~f7hEo_y%?+jLO0#pk7CfF$lq3{xK!lw_Z=GJvg>vz(ZPuZy2d+0a
zH`Hp>s^P#Iby=uub8+Ria$ZhoUCFJnE{~j-g;o#LYCJqF{MYka<lC?K?VC4zdw<W{
zwNulMmvVH5NWLY@gwy>MSaBwWvF-RWoa5}Q*8o1$pjt<YAd{d=qt>>S1LkeT`5Mbs
z7@8g<L^a8{a*c-{&IICVHaHT|-&YbS!|35E==r@C5XqEa3yiXjPib6F#h=)gMctN1
ziS)9nx0<*9Y|H$0*|O?TNE_utzQ&5CO+stqMx`9fQ8#RA*m*#QTQKa&rSG;mF}?ob
z3f;LZ_{%em(H-%VwdFD5vMty(Q<>S79f#t{(j<wTG8tIT&W0>`Cg+5y(%m#&UlkZo
zmJZ-{ZPc!$l)Q_#nd_&~+3m|mc5J<~E(?cJxS1w?^^gC+G)?^CXJ7Joe&m}sUwH{#
zO62Sep2PsBm1)1@^Dp=GDU;`kInNxX9lI%0vL_uS8Pfz)Hl`G7BY>o-2&n67X;&22
zOC8#)szNvDsbORXle9ZSzlT0)Fp!p$Cs(^Qs)D46TqaT}^Z-dq^7F@<u|#!)cJs3<
z%|o#RRR;w{9YE^M==E~;WJ#!1L@TW=)LJo7QZ{yniEhd_Z{M@%BdaC8)<jD)U;N?#
zOw7}sr8=9?r9qmDmyHtd3%WxDRDiP7N>k<b?tlY#HMC^efGDwyuZ{vpBt3l9ZH0cp
zW=)%ZRc0`keW<gk2jKb9xI(ssZI8M9?dscsZkZ?jdtEUylpN)W&*QIQJOfzI;|CpM
z+q`T)!SP9jDNW2dQ^-{93|ehTnIf5zGC8QhTYwU@e?x<j6$N@~p{wKqf_6!+B3YtK
z(8wI(8G+uY%R=hP{<Aws(E0;i;t(*~Kqe6=Q(-5@tsL0N#4cs#oS8}q8Zn-o6!UH#
zcs8e`L`mjdQV>MVnMXLIEG5qdUs_S?mTer#%j@){`(;Me?wReROv%AHpBH1~mRj$q
z<&8mL!EhLaCP6OQA|l&rZ11n?B@(KkFBJ!i_8!+Z*m0e8U3diEoQ31MP%)O?Sgm2F
z2Vm!!eC;9RsoV8zhnx}F=j)*IvWkZAr{oODw%1@QG4yLk{2)y>p6&nl1l9EccjYk~
zabS;M{2>6#p2oc2_t23cmF+F!7r*DF|Kf)SC280y$vKgOF_TiT-aG)#fzn2EcApaS
z5XZIGG6uA+4p_(BZ?{Fv?O4FNJ7!si9-47_eB_j%Bw;FP%SQVe^}uV0s45^SakxF8
zSy;8QPc!>8Gv`D}vR!2=Ay%)QBomSnB`1=MvV&ktTsdWqp~j~a54e=ryKr3@Z$!}@
zFis9QH;us8dX%vWUAr%{-a<QMDAIa{xD%lEmL~8$)XW((lERKVS7j6K+?#6bL(wL~
z+&0&G=3#}eIPvcI$m4qEte`2;Y$Z|s-8RR#V>HIyy<Q?CJP}c6wTu$J@yM8K)gFy0
z;;CUHNrpr0&T$lPA>;M!+U%Z&y_?T|jWK^TuI|~Bwf$WW^EVvl$hJ3s%7*M{Ab$#5
zR!eb*+?IRSjUoMP%YS9IJj#;>fE8tFDcBV@N}GTIk_lI3z%e#vk65tczW+QPAQzjy
z&v%5$#uJwIkH(h6j@_S2(Us4?ywp}G!uGq8+}=mIt8&>$y~A7p6J|8+aV8J{<zoAi
zKDJ}`50xbD`>y9l#FU%;p4M0Q(T9DBcMFxx_JmRjH}lM4J+bu4lxB8iq9oZ^k0M4+
zxb|937e5r&v@%mnFBKXa0D85TxS#BM(~zie7C!MiXA}0V6QA!j7G-MP)XRo(J2o|X
zi^mbA@57g6G-JAS7FE4m$Jk^1y1n6rgGkEgj@`6#JR8SuEZw*_LCcQUTFK-nR}8z1
z5<>qd!qC?n_ArX#{W|(qZYe--{=O1133jpL(d<CXHFPiLBHL?%Z1?rzLy!CYgloy@
zAj-HpxqSQk9qP+>7L_G7+J!%lznbw3U_FmN%SeIE`=KIC`yIP!&zxO5;*#-H=8`$s
z%#_5jd~@cE>mfBVYsj8wvSbuxsVi%5U|vdCX<&X}dvc(0|Mm^1$45jU&zbplqAx3J
zYy9%_TTI}$zj?#TM8!xcHsUVKQuy!w=9ipWMZWqise9Y)!+zrCW{=64W}Tce``yG(
zzW9uXrSX1g+!SFq309Oj7p7na)sDA@mn)q~83&!9>Ka%nxg#KwL9=TJn7K&3R4U4Z
zpmbWTXm6-EKy;14q?{<bOv)KGqewziLiWX})e5aOdRJ2MjBH<wuEKtIVA}1FIkTv8
zzf`8(4S5#co|JPn&TGf?jA`TDa^|tF^iq(N$-6|JjM^W9adpP5VycwP=FnaPb(cdg
zz(Jx%Ve-yRMx|N<I!U6GLo9$R@p-oBXJUSix5)NppKR}DIUMn!*U_Up)Bv+S@@ajY
za^>{>p|JsIO{q&I&(qdv>MwrChd+40OE}9(lB;P;x;QS{nN!Bf&Vx`HigHj<pIfnT
zxD87grL*Wt(+XIW61u~aQle>NsmlG^w=Bmavp_1DR0{RDaB9lqH{ZgP`RQk$^J0HX
z4;>g%X1}{($}`J)PbrCCe)&2yE?}<{JG^UIPD;~?qRdlbHz(#foKZ{M9ttHH)m$(+
z5oN~Nb-bjT2au*%5mbbpAekaGXZfknA_J;;$>dQNa)N2zF-<c$3+vgz-t+Q^q#2VJ
z0rabD$RJG99%F*Ju-3{~-@JA3UlP`{gVE|v?6ljFrh;Xmb)l}p{_v7KcNSY&TIYN|
z(z{T5;r+7kSXWM2APdL#4b4{Omg#naw9@TJ?eCFXaps16uy8IxshXEglqr#>1Ub=x
zwKq>~h?g*53Y=Y3dIak#;`p@lGCgxoLMe={dIJ<sWd`H|a)0j_d(Bwa-){&O=ah~F
z!|w&h{sPC70Kq@6{E2MYnX<lzE1h#YyD0y&E&uh|vbZH>%Ta!m5emWt#A|Qo-xRj&
zc9@vrYydN7z^BRQ$LdJ(efP3<>uiMX!;`;!_>529mbVdhEW?hCWQO)=h13Pf336sv
zCekFxG?;8Y>y_r6!SP=!2^CLF9@jJ7oB`W=XB~c4w(qv%0A#f*V(`<?zhpnp97^UK
zAd4DQHKx!yEz{2Z`KdF{H@v=m?R?uL<Sy(|X1AZ1<`kv5Mlt2pVb4!L`wZ|VMfdBO
z1Bq6Tm}a^rq*p{QV2<PhnyA)E*2uDePA-?xLywbfdgx}X#-@$B)|J+r)uv>2R<<sR
zZ8_1+nCAmXXC1#^D>-L!&cRT$D9<()8jDtzb)_wpc5YsoQ3MT<_WQ@TNKd@i%wo!l
z@lXVjJ>8Cgp_<WNy+l)7+GDJVamEczxUmOou$;vFcc#kP91M5qUoRCb*&7;dMG0Y6
zevE{UF5(@^lc(-6L5Vmd^0lMslB#~cX=2-1etDA**86sTc{7dka%5K$F!B4(@aOSk
z8qWaM^Z3DraP98IZ_J*wST#HY(R=p{K+v71;+fJ-Ehqw$v73#p*qv$bK&PHw^UR4|
zI%u@6P!sFw0F7UXGrD%u%r0d_y)8KE8IaLlLsBA_f|#+}%}7ez&O1;jnwd%_<tSV+
z_0mQWZf_1W79MIRlgQ?uA<M{EUjH|;rk9-d#+V|UtH`7wID^YvfV|_*gqlVP&ZTps
znYWuyjIz04x0qrGITwn$z^a;O%q@bEEeq8>Q>2llYT7CLJ$J8Olk@Dtw!iy^wKmqZ
zfkLLU($OrNx>rsUqyjQ@F$lbB%gbLDL>#hmm8Bi^^<z1>7NmLRSw;XA+1d!kBmycm
zMzcFCMbS|lfD+}3BdaIlG{-Z-+X6P81i>=8q}d0juw8?2^5h@zIc}|Elb5P|n&XK@
zZ+SNdUKLHm`pV1`Nb(!EW!=qC8RcJ5iV`{9(UzJ3ti4lJ2Rgk&ZNhoY)E0Y#x;==v
zF2j^Fd2&J2eVK!uXW=|$C@@Wv3&Tzmb(*=EXLdR<TMqX5B@2ofQ%c+(_N<!d>Ms8X
z?2U%z#t@WNZK!o5#ELJn48lgii@CCx3ujWgfNjdcJY`C8wn80_;^>$b4Ik7RSm;5C
z2hW4OcFD4)Ab8luNVVj_Oe&ef?Jc)*!?~YW>ca27`5o(8J)7ALZ7WOn&Zb6TbZYWX
z??I?1n4{#=h6w8ij^`6S5TAx2ayvAGWTPj>oLvQ=vaPafTN~ed+04*K8gKleZ+y)_
z{jRnA-6ai|apki+U(~!DW#~73szbiZmbbuOxAKsuw#;|hvdWnMgyLs#53QSLTmG@z
zGGWW({%u?gM1pO$P2f1hZ3}E8l@Uzv!QObvw2;f}$M3$|mj6&<$!xrqeQe90s4;K*
z=k~$lEXEmoBA&p^IYSZGQ`_#vwJ$19PUMbmzI4ZE=8QkY8Hin#+R{@yj$4-zO3K_F
zZkeY7%>k`BZ1#GpbINEoN}0%|Fz4)`*KS8{!W<1BQWE#i)sUvlrEs&~v#bl&ZIkJ!
zx;TK<A$3L6$Z3eqkF$W#O>^ecMi#|wOMzzYV=IPaPp9<WeP7Th3$!RLH>EYB4;zi<
zLM}xy)_lY}TL)A><`ko;a&vdb{%+!{Z{HxN6W`n~wBDUbrb<^~rBhM571BujZ@$!M
zzVKZJ5LP5%Z!+|`*s3rr=EYaW)g20B-&}AT>u3BuK1kTFKQJmuHnWfO{Q5Bcjw{^0
zH@@GFHwLKBeeYKZs1JSe=kcQ*&j8l*_%jXPyqv1sAu<GuDs<ztETmE>^F*0<B!tuP
z=#E9Y4l{g{YJ{_$y<A}xqlqCY(K{^KFk3O}9FLt`gq!)mTAROKS7fdH{L7#5Vt?Sx
z!+X|N$+bGzq8*tsG8Oj4yV}h2o{~C0xqAV4N1dB#CMTn7!?eNDnSq~v`8m9O&%5)9
z8JN3&{*s*GW7@+#@FS;dWUL&OBPBnqM_Hu-q$t01ZZ{!~i{P3W=jwAsf!6#GJCw|n
zgxZz!y22^3Cy{btf4C!+#$#)=-Z-zFDLdOsBh#zVM*ZAQ9M2~nmNTz^_H+L3Z(pE&
z;ql>-|MUO)|KZ#B@40{fNEYLmfi%Isa5}H_W=#1dlJ7{-DFh~lNUL`ri*An$FCsHE
zKO{2uCXmH2@tJdB;R=$-8PBenAVU~D*m*s5mSS}Aphyg;&bCT;YjX>SOg6_IzmD(l
zXgydN`QnRy$9Cdwe)J{C9=KRt8|SVZ<{hE_`Du;t6QY=fgm=X<36bNqK}j%W=G~5Y
zmpHErIy6SiUE?&E8rHFB``$5D^$^Qarfa8a=%v}p(v{jXZ%?q+PA9ulTpRiJ1-o*?
z=PzEfyF0K;FKFk=dcU|WlOqFKc=akfcqlVzcYv2KDViykDcMl1^t!r!Mr+)Z%)k1@
zPx$)Xk^AEr=^dK4A=h-`Z1<!*BQ{Zk!FZGA5Mx#k3{%+vnGL`V$M~TgSyP1xcKgE1
zL&2JtR4sz*?D)g!O;}rn`aqGFh)k%tfa|;Ci6o)tBbiL1wo5re#~fUoGx>Jr7hnFI
zFLz(??fc*H_WfJ_@BiyR^5*S(mUX4Kg;pPd36p|RNbH!(OQy8LbcNKstVtTmLf&fi
z!eMu0DuP}wV@}!ks}aDo+AuXzD$vyhU~@R4l94Nr_u9E0uj^gWBv>-uaAANA1N0u6
z93KX}PXK61(J|?}mpt2@^H*uwsC}hO^OKUKzxeST0l^4cH=eV&sB>SVEMs!EkK6M3
zV_W88%g(HEtsFmt4W4cJUtr5~*m7nKTeknkwv1%Ay(hLTv|2HRlnT<iC)#oWvg=%>
z<bc7GC^bAoM_9m&WI0%qqomRq=|Ge&eY}VryBT&US6|<~{AIJ|4{Z4p)X%vF=hY$!
zB+6<lT@$QEVMi(x^SpyeI4@^Ua;70W0WkEA{JUmKvqrT_!>9<0br!8?S`le1OQZJm
z#Eug}QrM-Ln{wdiU;Gro_|4njA*#H7@sj<#^Q41cIrB7=QlZ@?Qu6cb<^JFj8<7;=
z@3Q-Ub-}8^0(a9MUKW1++pp-0i>ib2(BGrY=uIgtlQogk3ul#yCp5c8X`$ene%;w#
z0x${PjK{XdbBz5ok;}~e`p!YArj#ksHHXZGH=nBASeMFL8>i(=N}0LLkz^5CU%k>n
zUG5{5LKoptEC2rA{7?M*zyCM<&;O7AC%^jjZ}^Y@<##-u&vbLvnqpK@nsgM`<dfky
zOU0xD%^6i>(2-trJ5DnE*058&vv0|s1ad$xF-kXvz`Kf<PP%Nz(<Ve4(7iQf7*=*^
z3ou0Dhv(>WDc^@q*%f#-5?D{q?`_X8Pf$1Yvpj90^}jzM^*nwY;~Btu9zWQ)Y+obG
z07KAXX$vP3r?w!?1DQ#(Q&59FGfP3mXe74oZDx3z<ZwWuWv<}|&dC;5TRnRkf%VWh
zICW*|%A#;wD)*0P+H2RbINm?dw6UKi=2A%3LoYyRgf^5qz_AO5O(avYgC`bR;6*Z%
zz&?4$pLt3Y3Pmz$D&#VAmQHeo3|3WIZAcSRlqzlqiFm8&gpn+mV*-{!gP?ohk*Nf6
z-a8zo#9`i(FxTEHX`?@jBMq2#)>>KD#!_o^SW*{~6!CW9I_mIYN)r$FNB&>`@BhFT
zH?MfPf5G|wf#dPS`}c1-pO2WXp1C&hELBe=xuHyV{N(e$Wy(8x>n@<4GOo38j^GJ|
z0V?@@vnvNjLS?i>|H)**3~P8c4oFdt8I95Xn2ao=ScJlx!<-NbNoLH6_867X2?g=a
zmA+b^))u%H&v1rq`w~W%q(P6qWJ8DF-a<<vdOZCJtnAM)MiO40h6N^2wX*iis+HV@
zzE&o7Ol68hxG|9;JPshiMrLkP!f-tZ3>Q%DEZSJK(X66ZtUXg@sg2syJ9LJllBbD0
z72Y53*-JxjvX|1<vu_tkka-75NX;F|EKcJ;keJF$$rEKVl60mz_N$3q5pL$pZtk2~
zXRmi;Gw)hT=4dCRub_f9zx~s+L&A<)*I0K(5*N)KnQk@Z%!-OKB`=8Usu6H2twT50
zFfkYX@RI5VrVD0`RvW!Xz?dAl(YR;Bu1DZ2COp1-<UjrI|H$LvJzvi+`0AUlc=!IE
zckkYCKA$1Fi#EY+)D%fyQl{6;``2Vitm`A)I?XDe-mPe?l@#R!i75F?8@o1{mlsWv
z5lgfctVw`JIg+SeuCOs%@OJJ5Lt?B$!(8+_*3-zqdYnDFAPwD)kF(Mj&5kF_JpTBh
zZN^HC{iD{#QO?Y{kjkgYq`Z7Y-?yqZ*1l48MZ3_KMwxd^Ib`+Cn2fOH;`c;>9Cu0l
zZz-^RBhzr;>e-h67+bFNkO0_3k|RKai!Hl`(4XIyy9gR3K+dxA;@%R8wN@iHX@ZcW
z1S?~b7_5+u9b%r9)a^h~S`Weh%@Ab^dL?KZ-W~c<3U&2JHon>u=Einf`PYytu*+Z9
z+IZybTBh<T+Oj?Qye;gwtDgzhIyn_`zkmRfxDQhYTSf)afH`76Rxh(v3f)}Gv{~N(
zG7TAp5AE0&tkAn_r5;ZY7|Q+m5t4D33k7w1mOW%I=}nTzmPp+|8@YC-Zp?Y2l+bv>
z_0<ZAea`HXC$A-CkU~xqqM4JvcUBdhCNo)PialZ&)s9qMg6*vv%hK0IB!o?<0p?Hh
z#55HoD@b>CmO<^ENfnbsS94#wby{<leeX}#Xz%XvjRptcAWRZ3D}3|jE&uqR|1)~3
z{Kx;_Kk(Jp-*EqM&*^-|)Y-6Z`3<A9pI<WXUon*f$uc#cAt{z>R1<h3D=<t(QGVDQ
zo){O&mnw~>(G8N#KMTY@x{SuTI!s7>?;B3xV3s-)I+&O#A(lUYlPy*mPe7^brECFS
z{hk$1ygLTI4?7(+;(<@Zw|*Y~qVWu1J&zx3JOQu(O<O@6r}_xJ@vwhIX$f7uGfJXF
zmQBbq(J>w^T{b!!UQcZ2>e&$j5_-@hy;m1rOaz$W2lstf)^3~(?oX9(<&pKvhR(|S
zH*Y}T7r*!g`!XTc$P^}lZeC(%>H_jNd1PR-3&j>eSKqg1nIH)_2@Yk)Zl0KIV#2{J
znI?{Ogvw-%qn^00_w3q^xliQz#@iAmkfh`^d7XY}i9q&ljiRnqLL;%@iX6_i$~kj0
z?Z`<$8kV37c(1_(&4a$>ymDGA$F)X)=uWBtYimx+4G?0N4!ry3p8w1L_5a2fw=a0L
zKk()^Z#W+p9`M85?Cju`CDx`iHOl@c9QHrqU;X0WK{tN;tAC((KRi`dlm}N+5s(d}
zOLXy1=6~?4Zvti}vSsh)*afYgg%O1mr4jBFkHRB&l#{t*x{11{<oZ`hfClEJAIb<&
zgp?)>4^sR<jwQ%AI)QZQVmPuzPs<7~-O<2xAG{4UluwDdF~$H7!U<a2N;X*fN@<|y
zg-J5IU8V<Bel`b1S4%O!9^o2p#R>O$z3Wk;Dm=96UA$P)p#Dn(%sAGS-py(BIQ*FB
ziD{ns=IyuammRnHHGN&t_2hoI2+XrH2NaURjI%$@64Sh+><(x?V;K4D0BTM%H)Y4&
zUU|1v=H9vEC6fuqenz!$-yfkl!$Zt<UgqhBWQjF8I1QLRiP6Y(6LL7%N6ussUgS9f
zs!m@U+B&tz9+oqjGgdN6qOY*Flh4C>3Fp`Yak{J5bW>VW7kRek-KKit-B<7UU;gL+
zg%`UWH&fwP|MeRlAJ3fD1|3Rs=d~s1p&xW8ubJncbGZ8wv~zm@7=h9WlTP+ZT)^+4
zh!d$S^pcPinmCON%&DLzoEPu9CK^F+@=iitBJ3<ni5eJ736jOG;0vZT2cJcTLvCB^
zZ#QHcs6S7$xZed(4LDV{1j*BfT_mQLWo2DgZVv~te1dM2eE9vaWn*b)aufP_A&W3i
zhY0jwXX5vbX)A-nM&Yjm40HG2o6J;Qo^AP`W6R6gZ8=yD+TG9TCLR=P2Mt@+Kd&vT
zB)SwNNB6Z}y@O-Q7{XE&YhlY$ymT;Suj&bgvRC{k#0r{|C%i(hv#u+QWLEOBrmgHK
z)~f`onmDJg0V><d===k0?H+_HTef(tYc+|%d0APOmD`&e%Jhky$-}vhaw<fqZDlp1
zTSIz6*UFS82e3kbf0p1}hHsmqp@1dm78-<I!7|27GvnM(G;MV0Azsm9ul&G{yU}~-
zskX}7`)^U}ynT4f&0)tEx39?E|7JNsnvvuszf%{N@vxP=cIIl_&NEXf<Q%?(DpM-#
z=RM~IR-(IIo+xQgXX4nu!K`CclH8Hf0kH#S#<|~9^+-n%>m=`0PLf2I>T6MjDjiKQ
z?{~bqyW?CRS?dDH6K-qol<sAkwJU8|Ij@b{I%{?I=2)Jpo^01GnKX)K6Ilw&Qu*!g
zzT*7mk$))vf&cPfe$D;Ek#~<?!Jtu2+{2E$HD2z&;BfO3_R}4v$|)ZqHzcjG_R}Sa
zFT?-!#Hr!7l#+x0Sp;p~EW#b=3t&ttg!}vi|K0rHQC4mat`mu9z;%Q}AJAXYc#>Sc
z#$NqZ!g6g}+nad6Fz<1We}Grp#t{7z+1B&;iyxn$=k+{3m9cG1DYhI{EE$p<Ajy_o
zgxfONoRZ7Z$+;j=#x{fm$HwK~<jBUmzLmR<(P+KbC`sxV`HBHxm{kYSro^##zI`}z
zT#crQDeoXU1!A4rAnQai&nPAp>eA5OY0K(e6mzB+6mvm#RLHv>QjO)^1M9j_*M($?
zO2?KRNkK}&y0Y}QocjaE{)XQ7RGUcs1%;W(Zdt9e*qK7b^JO7qDn@^zVIbU;J%@SE
z%bPp)^NxL<=+^n&>07FG*0oVeVwy7N)!SU3mX+ERA&o=}2Eo2^pGZS@B=ydsorkva
zlb0`e{rMg5zj;d+*F!kWg<pR83m#7QeEt4wx;IrOr#o_b!MdCxGm`vHS!ZZ|v2|N1
zab|e{DKfMJTyWkQ%mpE%-LMX7g+%xJs~#vvRo^yJtfMh^CXl+tx@oQi@*QB3{vI0g
z$Q}*W*!cOPzVlIpNSwwV)aS?+eG22rWjgF+;$T#@N-^-_flQ_#>SFb9-BMY$uK`dy
z?Ce7|%9c#&)~Vf<9=ZjrD0pLm1X3bc>CvIf4B72S`vdE_lT&B6+eIfZXD+O*Qd)C>
zM?ElHn!2EBRjeqMHT)le>>1xfy5+tvEcMYd*<D#$qpEP!6<8^9#EV*&V3`TB8X}n!
z4=i$I(Ub2*WIEYPH)QDEb@$5~er*C0+1nJIbu@&DQ_gf%TJIhlH)U;|ZeBhu!FJNF
zXb+x(q-Y9ZXltBLXWqQM=hfdF`0UG9{O;?w*wM(dmwz0XIkts|dfs%6cJnRsbjRbp
zi&GyCx74=K`Z)-PQ!I}G0y(25nFUuT!018mFi-behtWN2z`U{rt{NF?x0xiDBEDo`
zX5?f;N(V-nLv&Fa@oI#um)77<#>Exif<eCD?#DSacBau4@sk->+mm6TGZ}*Hrj*dM
z%AsXq9eQ<!LJEDRp_`(S++bsH_&rvQQYLc|=x1C0V{N&!)aH6OsrWpnm{0Ao_S`R0
z`}5nf0A(a<#26hLlO=*QGw2-~6r3U%hmG@q(776g^&V848D#)xk+6~!fxGz`4iZ4A
zUUY@7&utqlOAUtDV0e6JpV!xTfwWEJJ0AO&Ied>Cz=3&ZZa0S~c8qu_nW&!$l5z-X
z8`tOZ^_hGD@P{T+BtUfaGVM#Mk5r1Y^sek!wX>R6JDh8$&z{JtwW7P3d7kO#)l2Dm
z2kA`7fsG<=6Lqa<Z5|UeN2&YTeYch3Uu0^poR$;E^O@tiGR=*yg|j$I$+G*-q6^wx
zM{eavl{2eWR0ya?#x97WoVp?f_WOw>#=0(?j*qPC8EY=}QO!Sd%Gvj)Rym#*YITi2
zGf$Kb?Wi7m;s9G**2>VPtmlPy@85I#`9wL~(C~e()97MAM`f@e0VeM5zTj~932)xq
zW7bLL;E~ag(Um(o3x))*vy~Y-bKrV~&R|Z>wcuif#mX4zqU{A3-^vxRe6jce7Q5~D
zO(cFhTq6l85nt%`ewPuzpN(3I>l69vY<~KZwC>d;&ZPV_?D%>7%f>T+^*nyCu_ed^
zMp5%<LEGrmV7+)TmwD!;b%f&GFJe=ZZKH|<SMjbYcnQgHiY**wtPc$TiY|*EXyO^r
zwRYaO#@dZ;Ldi3dG|J43s*NdX)IDP;QL5AEwUM%rGh_?A!~`rMr9ki0R;krX0&0Bb
zTxYVul#&ZR>LY7^&)VLBo>3->!j$%i&7nWABD8Q9d@=-Fgfre&B(X0uujd;+yLrLQ
zZqJ+xOJDiz@`m1&)*7vMnkq|ViCp7@pkIq?VDuhk3PUs&P{XuSyYjGfEE$KJ8Oh?I
zTx!ho#Lqte40&7m?YlE--c2dVo+NwPy7*Gq4)jq<;NhT`Oyo;2C_@O=zdWGI$N=$r
z=wfL06g<X2;bdf7@q;!6MH+D?8e{Gl3S@Cuhd;c%>von8M{)e@YjFt8UXIG)5Pn#O
za1p=U0`xT4U!TDE=rY~;RdG9RCUkQ(oCz=Q)vn=a%I5I+Q!jRX4HJw;0Tz`0ZZ(|c
z5waL5d&W?5+t!ko5O>lvljfOrKGCZ(O%tpY5vT0e-d)U9G&)QhwRRU-Z5D0lUF_BU
z_e@HeDbFCTXJM@)U8&V+_N#S}3T44_&kGf2k<fjuw!E-%qOpR7{#qggQ;kalhX7<l
zBxlGhm*kzj`eyzqm{p$Tt-aHFgd#Zyh~Rr+*U?362+evq)4EpfACHjW=5FszD3XJ?
zSrWTEu`HFocEAJGY1%X81LxDqToNz$dvZ?nzDDU+)KPEw-C@^4Rtr|-2w8ov+?b0N
zV96E8>HC~q{pd{qd{5BLA-*mZ^gYZr@cpvDp8OE+e|0`xt?SL(|2?*_UADz8@F_XV
zaj7OmBUOu1UJb!!-<wSwe8wCg$|)mh2y%}ACUWI|O9-;712}d84WDiKkF@31T?pH7
z%2OOjW=7vi>zp~G|1E4;{j*qT`CMuZ<pN?}?el{XA=i(Q5%>lOW^4*|u5WpH++bfl
zIa9^J1`BR5p1BA`k2QWlwfR0k+OBMQT%&v{C4>pK&lbM%Sc6aPI8&mltE!U4^;|b_
zciVrT0*_-@_-76-N8(W&XjXxg%SU$H19YrvRQ2XUrg(Q}Am>7t8qJ&Bjwkbyw<5{+
zj9MKi>V||pYjYNSN*VBdsP@iM7nZuR)D=q|rLc<YxfwNlkB0W*0G2E?t2FcTQ6!OR
zp0E`EoQrk}iNoX_GHa`>=Y@4SbD8-#lTxC$#?l(+wNh($&@g~l2c2!(4~bXUSmzog
z>Bl4O^()dOsC8Oz^yq*p;xUfrdlE=tzkkVo_k!Pk{SKjXdx%n93GKnFb4;$*h#zwr
zaSa>r>Du@KVcNwehJ}tZ_F~85F&97hi5+fh=o%EeI@dR|l<itKNlN07VSJ9OR~_G8
zJJoo}>e?!$lyPCt<3~Q80j%fogO0IXKXD<Vq^*kbXcdvnXS+L6%2?9vYlv#69>J}R
zz$Kh$cV9>-IU<U#ose}!@!Dt%a!%ac&YY`pu1ZBYC0ImgDf9B>E4;p*=prIcg>Fi=
zLY7OQvZ^A<d->2D@F``7CFM8Y{Eqw6%5UF2@aA;Su`S%p6M4!kwUc{e%7*pISs!Wg
zfRsnpeq!y3+Ez+G&<&359q0alC3i;0gy!N-1kx!vbGv)R=Ql6-`Q1xy4+o|)1rUYn
zAc$}{97w59oAJ0b*48|WoP{fu{H;r$b$fEtG#o`Re*K$YvmW2m7Lby-KR!66Jh=lw
z0@Xsd*$=DUdGqb>xY^(E^Pl{j<LNDL@6X(?hAkTGVi#ogH;W98;F=gCql&o5vk5eb
zvLlF4D;$ukVD=gW+ni!Y<&qkhYmp_QlUYtk&QARf$L%|v-|<>cz^mUkMi3C-U6NqL
z{83N-Z$E}pVHe+FI3kMM@zEAU5_dN*yrWQpB$tj<f*~D}22){V>gDoEf_E-T7-(at
z7fJIs*QAhzUD*mv*c`$AjNRRE0QU17c(Oot7G^g?TPKm2C1c6KH;b|s5F|lcnhUyS
z(AIeK_A7pO|HxNw-|={Q<h-6aq*q9^&^C;dy#*R3D~s4BMvo=5FI0VCu`|sw)h4p|
zenRSW>7i+3STpv!12_8{F9jATyNOEYw-4V?a;CRVt&O#HPE8%W3Z`NjOjo<;!KmsL
zBL*+jvI>3SpML#MJU)KS+s8B2DsPVu%(<|C`2|%@tolHf8;Z>AW??sX9v<ExH+w$&
z<==2Tf5o!C#k-gVxGV(IGMri)OsD|Sf&r-!Jc{~SN|I=?o@{8^jTk=GiGxQD6G-g$
zMFK`K59&=VvfmbVjpN$pExw1g#>J@~;MU+&U6%ZZmVYTgyfDQ+De!8HYfL~dzJ51o
zD965Xustt#cjPj<migoIm^ZBveO*A(Q5FmdL2;7lvn~H~Y}vJDoOxy9zDOv*IU6ZT
zkxc*mwwxT5!z2d-x1%D_@CDcJ7a@`9VPiVXE6G)LPtd(X2~y2{y_^Qn?kpY$=?D$5
z@BxC^;ntY0@S71_UxU@(XPYCC_qldZ#FJc~iZyQ~R>6i$0oD$mci1sxZf@?p`(TEU
z3EdoHQTgah-h3z>K!zs@++Tu3qK6W?szI@*b{q-YvlvH#k_yu_@ylQSl4&aRZo#+?
z=?NNnUCAU$H0~KCb|TTMSUdG`A!lKKGxK;_cz6Gn-@SXsw+}~pGGxm1)M;j{Dv%l^
zm&Y)v&X`i=*x%BuQmG-&QJM3OEE8hRz*?nYmT0Ten}g%l>gH$WocR3JD|YiPI;z6j
zp&|4nTP{S#y)xS>4m3XhSg9sTDv9~9<L&)hPRF-=^ZpHw%b9a`;Ce!+rmUvycAqiN
zugS?s(wNIclEi**oL56q4cog0ege3%OYdu!s}eK`$-H5cnUFN$rGC2I%OUcA^`ZFC
z(z3)ZZ)ISrPB7@5A>y9%SRxW7!cks%1$;lX<L|e~?ft`tTv|nz(GcJh0ivJBzi2!I
zSkL1J8{3|AjfKeNZ%%jabTM+7+>uVB9d~p$3XniHdbnph#pH>o^oH!m_laQMNCL^^
z1k;phI=cA8VIwX~I!`lR+O}S<&z!=ZLt}~S_ma>H>`HI0F;Tm6e?0PVUU^&>&b3n2
zI}NIcXNE)V-BjszMp;3sJAxvZO>{GAJ5l?YW|d-zE>mPtgfxV3;~*$`VlETA;_ZnO
zUX$N@wD3((va_su2+?k=jf>B5b)d$-ajI-W_;DA{>%#q`GTRL#7Y*&&ycRx0riS*7
zhnMPl2d3<IJ6fM`=DBU}ds#1W4{d~ru~IhAE3!4ixkAF<x*T+oL)v~P+iz_<7{zo0
zR@kyh!=Vkzv|ZlPuW4M7T^L1A7W;ryKm5~`sHFJ+H1vHwr7@V08%9WE?trSDE=n%Z
z6(?Qps%LOv6ZhmIYERg4Jah|khMgf9-E<S2wJ{a(tnSq<YCr`_nV8C8etG$@i320L
zX*3K6Y25d`1xLf-I=9u!^UjsW<;-baIJbqRSLWowvws@jGraYDr6=`LC(ihNUZ{3<
z@mm8m58ROMLbzCbW)CLjT)3SN%vmT&D7(U{jc=!klrySISEH%7>+W{BH`nWKNTOVX
z(Am~5IHTjVoO%Dys2w^=?W?m|vO@B}A`15W_Yz${ZDzNd$OTjvE}LWAwU5A|t1r7;
z#~?|_c<8mSZN^}mj&-&<tTt?FnD}MISUmUI|CP;HW&3yRX_pE6@FG4QmtrG%^6!WC
z{%MTMWM95>aB$Vm9A|RQf@Fy~*KI91%X#e4PxtQ2UifUw|3_?jJLb6fP8S=#thPV9
zEsryKXp+XC22<PpG<jk|7bb{YUiRwe%g*Hfnpr~05IRAZughgg_&u9_+hwjkT;0Ea
ze22Y%V#YkSzv~_8+KzY0&*YTE0j$mQjuNx)IFoIAVgF|Qm%{HANpuQ=O9^~v$K&bM
zjMa=Ji799Hdx`SJE1yr@-`5^h6Dr%iN7zL5&7xt4!m>1u=Y{u=k362;&+fq_b*(<<
zp%GrbN_Kx#7F3kZ>P-uRVW=o#IlMHlcIbFXUhm$JsJBiAZ|swdR1*7nCYKV;6}*YW
z@Q23Pl(@{tll44iAm&TWB@~wCWzcI|Y3eL7F~~$mZ5k<Org<NrVq?n1l{0chk`h^=
zrzNph)eM)(A5Ow$0c~e7T>M}WK0f;{`_i!F5rJuF(2a&jri4Fud9H*rDbd~aNq5q#
zZ-1{Hk1fFb-Fs+aUhk35<HtRo0j%fogAIj`eS~mmRG^9S(2h(wBR9nlFgjINOjp-N
z&}$dRV#9P$YE(xL2RTWm$c|ZrNfI(o^z7hOo{aq*K#Hz3{JL5$XqnligxSi&az^lj
zYL54>%L3h1wJ_Mvd-kO;Pnmb`?zw+F^7ee<)GA*sXO2zz_N+XvE|NZ#j$|dZl|mz_
zGsVvRktSy>t<)q`5mLJ++lhIBlfGxwGuVWvP!isWC3Q@UGEGc5GtWCrgmbT~XFt5|
zAMQD|h1wcrF3j`HvR0N_S-Mdbtan6U%3i+XHR*nfwvIkRlsi^0zbitSGV@%J<|2`r
zz=|;6B+4;EsNN`zQdV|T<>lQSH;0?(+NembNbX2dY;={haTtaehIyx_i(4iBH&=c|
zG*P0}wqbq{f@@*9*hZ|m%RdsWlaq$yCeF-|p%Y{&f*?P5Mex!}O_5>LZ5?jG-*#SJ
zISfz6ShwSE%ISQh)%+cg{mUFN^Bb#M%*%0Ayx``(9hs)Wd{gjHx1-J5s+%@vMwn+;
z{RQF`M?_-I+7P^J=bo8d5~bu|^%-ftB~3zkS$TI{`S$VAoj4J`diS2B!ry)IQ!k|{
z8S)5BeO>qN&fwD4aI6wJ7Y?_s%TZG1+>G;S;eY$D|3Ek6_U4YSk1Ow%Gmp!ewJ%KR
zKuL{bUqB{u*)hs2RytJ-WOW<RGgTK(y^`o;QA{Q#5mNT&lq5_P4$ORc`<h>V{!?yt
z6D5o5d-R33yr7zw7_F_d)b5O>DEq~X+||n=BEa)<X$N?dxiC5@dkK;%DNpS72Tse0
zX5Ljw3heJDPGq?E($1!^ws+)`xxG1Xdz+bOD5W9!yaoMkq24L#VkZjX*=I;0@GNNR
z2*Q;7Jxn1Ygbl<@^0ct!+AZ}?XOKe-zH7$%;-Gy52qNZz+ZK4m9;cUA8rgBT<3BF!
z^$YNI>kPL~g-RL2o&v>~vs=12C+4A_n99WNc87?9bXu+S9-{dMX6fp#Y`cTciKJk3
zxX}2sE&uVh?4S~$&RR&4!#MmLa-BDPBkq4@TMj0Uncr(MiB4x8vX||?yUyv81d@1a
z=NdNM?&n|;SwT`pWTH3M_q&2jw;|$d$(bAD%)f?3o-Fm>KQ^FnD**jtEa)#|Tmg6h
z$-MOUi5;hj`F6%Vx=yRB3w5`VR4ROUCd)uyB*gG7QLJ#_K+2Gd@S+H0!ODabKC<JS
zgonEFW;v7R9n+jS*`tFTlH;`pdlr^;B`LI)Y2M3+oHMC-;~`P729L{$U;pEO;<zrn
zKQEj~tR@`WO0`ZZ?oS$%mo3}k>r9kfyvtsfN=wRW7eB5d6l)X<5qAtaEays*!oC#t
zQ=#O9>}JRbc8R9O<7wgCm2+2W5806b>aojNHtfNI5ex$A%~l?c4=n%n%-{a}r+og!
z=X~|;d#W`m3rU1|nlbJ4wz9k3^Xk>-bUm_Gc=_TL$V#<uQBEK&-fN1I#qgu;3JSN-
zl{05d`v227u`lU|C=(u7*mv=oE`VH|0Z$S`ai5jY3B2gdilk6H@axi1(*Dn|<BJzA
zuASCH9RHbt^;a>T0j%fo!;Z1}4OD%k%EEDI=FAN#c;-dBcMw!>m5Vs7Jg};G-UZMY
zE|OP|2}KGeB?qwbjOIj2=K5A!3s?(>g^PO@*#b$S`w{G`-aFDg@JgZIwCP@$XHJ_M
z=Vj&GDw16T>s*&j&tfD$R68b4lh$PP$oANHKft7c8O_dgTioq1rHO{iH#$CzD8vtp
zZpxw!>n_m#{(NNBN}jR{;zyRKwoy`a2|fJ41qX*L%Jc|r6kzDgNX+<l7F%S9CGLwd
zx{Ua&Hd1P&-2Ahfk+QRoYF#kvq=f61nBOA@GOmLwfB%aPmA_XC_S7g9_3QKtla#{Q
z50;i)vO`<IDkH0y84{kDg!3KGji6U#lnn7f>CrVld!54i*am$4Umo3?mt%Yq<AP-l
z7bR@@V#n3oZc~a<PV??}+6P;0tO>5>=o*;BC%DI5NfRX{O3FyiA#dh7JUR8q*4y?P
z=UO?gi+7jIX)CuuM%h?&0K&_Z#A2O{au&m>{oUx@TVi*Cy$B7Xw&vX<HL|-SSVKus
zdhEfqMR~7-s%jObZGo7>y3N;|4dG|k^*EFrhuw~wdB<G5v4Y#bZQEw=>g~xxS1Kx|
z5WJKp%f<S<biyAn-sciPhMtIAScSyzN=c+F9;_tqU21kl`y(RRgIwCCG2$$g;TYTH
z`|E`hw_7A#>~M2E{Xr>$no-K*aeCP9MhktqG~JFvyq=4F4k!7$eja<*HI4EDFUz*3
zKOWaJ=%1nYP8$)|Cp5OX9cOCTvI0%L$CwnN;@aq1X<f;Tv-q;#c^q_g&n5P0d$#3&
zt}Ty}R5dTe+!}pcExiwR%|Dwh%eJ2o_BhVQD>mdXoGU*`WJo=TmmrRZJvl37V`p7f
z!c*HFuQk@)g;^QE8Zoi9{dSwLC*#9k#2RkBQR}A__zwRj*(l5Y(2jfL11-v{E4^tH
zR}b%rCuj1-7GvHcsaJCGHIzWI`&cU8W%8*V`)ydGDZvmf=V}jjzc}n_V}nUNcC49r
zyt;3!>Y7>0vZ4s{{=n(Hvb5&D#;Dr3I4F20LT^-IB-f6W?OnHZ-XnRIWUM?-22Ck*
zv)gmKzh#~ZDJ9N$)krnFFetAQ8P`y)+La00m5BRx2giP(^zP?~YkaZoG9*YDhAz~+
zE8O1g;M_4?quh4oRPSl+jEub`eE7?b;d|VbckNd?G|<A&znG<L=fie4BH6RK!Eu&8
zB{_Rcq9djvh0MWa9>!;v&Cj-{`@yvx%lFvv2fy2rQ*zlep2v@SJOfzI<A)m;ya!-&
z+%C*jW7SHdlJ-WgjkYYD>zP(NH@h1LgeVt>IbO(7e;1d<pf>|c?DK&+W#*KSX-0CU
zQL)@GtBj6!&h>&yXR$`Dl_Z^M+R=KauAQ|t@1oTwk`;E->>Hyx7pFC)nRf;`E-R;2
zx&7JCXs4Bj|MCspTzEEFLNudVF@gQ$Ex;SqMdy)?LX31YHJ0w(Qd+=lPHnCfl6Q_0
z!^GQgx86Cm3fVZGPOP=^=KdX$;qza9fdNZ(fyrg9R1z8~coS&a$f<bu$KIgpmXUN&
zN=eDKPJR-HzG`$@7T4UVy3jQ;$X0<ynNE~)=6rsn<cU-gt*^X$_bt6wN=ek!+k=-R
zIJ3^nASJRfW`-p{yxp+}i9Uy8b`+<wOOgnZToXvt#b;9@${PIu&bfphtVZzb8UQ_h
z7F-wO;yhc>oZS%!In7v*?hIb&`QPCLkL4RQre$5eGn@67Iwa;j$G^JSFsK+yZ>*(5
zR_Gno#<Cu%Yh%B=A*B#KHn})IxK4wZGYc`nx*!yGyF&!MLMg6kQ;9d$2vX6|STf!(
zCvr|m5f>yrFZ9}+Sr<-YS7$0E3Rx0Uo}F2zpjD&Ufzdj2O%8tV4)kN?xKvi{&caL)
z)S5fSU0bw@9t8=_n@oflZMWtfq0E5+I8mvThC9z0rZVxx>(_jCcgKtUKE&}m_tiUy
zEv-|#i#@N+GrbKjh07@zlw3LeQKG7&>?%O}l!gQEniW~(a!&*#3DYh!Er~S2&79en
z4Dyz>EoiUId!^}-t|yd}?+q9=bX77fz#)wcZrTD~&+rR93Jo$5n5G0Nky5$<SgM=u
zniyUgp-)*^<Ml?ViJ~Dk{(TPY2<`_fY+K<U{QG@lfN};REXx^f^$$GfFKhVR8rj?d
z?B<%379h)68>=<a4AvB_jkO+GmzA6S4Jnm4%by~Bfm?f16A{otmigJ1{~5L{>7odD
zkwplmQF2@E-BVb9ep~kYDxvx;lmN?hq<}&gJz>f9HM5KcD%r)jI}VCP#Cxqem0}+o
z2}Z*<TTUrmoy9-|^A1xfAHxFQUC#gU2*5?)bUxGTr(?_4;8o_zufDcp^)p#pr}fG@
z&g8?5gCgx}%>aHTrz-#m#0o_~c96xz_7P{fH^UQgS9YAD1n{^nyjvzty)j$b1ms)j
z#wu>d?qm8nVIh9*dT0Z*^5kdjom#y^`2OV!e)5x_aQ~nFk^6UF(<dN6T4R*HdJ<JN
zMWV+w7*%u=p&uY=!v)r-9H4zSa>~5Ex#4GD{FI-6`7@*xRE+=l+yBbrdgjztS~nM~
zZ%S_s@>r)AZAJ`#&j7h}Bmp*{>~eBTNS>Hn0n=HS_L;USRO98V9Y6cU9dCZsSxzU8
zr+e1=o;Q#GMAt`Bijw3g&qTam>S%x<5tyaqU4m0suMJs^0YY7=!i{o7>3svMM}^Js
z4F^D(QlfREwT6k2Qp`V|c-TPg?Til43BJRQulI=Wvu2BRs}~K#X8`N3Vmt#_&*O(1
z^5h?v?Yzn&0?|ggvDSsYR@z$Kx!$6Ci|k-*phM&@ezsu7iexf{eVLggF@=z#4M$a>
z$V4G(?c)NDr!x=7M}GF=mMq3Ye~3(#cl}T?pm@Ne{Su7f*;7f0JkQ*{ddbuk*&Udh
zFqau4f-)8Q+DK$_np`Z>8|3<-!6n<EezjCdG-(?+)U+Y8f_hgkpSuwNW}0?ROXqQM
z0BgQIkaD8c9?y>h?J6MDm?Ir2JGrV`AKBb3NQrBdQSUPj<V*HE<x&3A$&#>?*dK)b
z9?qwgZfmgX8m+CgUTM8uvUYJ`%5^=wxQ>=eu<?TAZ^L<u&ob7Am>_y#cF1V;A}Ovd
zvcX#yvsfqBho(I}C`ZSnjS2R_-{V1i{{cH50~hLe|3i=d3muVPyf#BzV7T?p(kj}T
zx51855_cL$&T~AG%ae&Eg)`!c7@8*!cx6i<)4PL)vt^Q{@gz*abIonbIHqn$w*ZD+
zTzL}jW4Sd12%r|rbZw|XP80jx4PSow6T09XyULl8ClC#m)5S^saBjlu_Q29ppbVWa
z5AF=K2ba>HKT6(m0<%u-tCs_7<DsoQHfO)I)>+oNb#;<FvZCWsBJ(K&z*2XpujYC*
zCP)HU!)&mJg8h_){eEUX7qY-qGP~Wx-Mm9&=Iwi9T`J3RWUWWKE?)MN(zbt$;vh>F
zubW-ynW1?xf^-j-H*scyN&CbyWV^azLLcTFl6V8)9t_X1Lixy^$Ir%P(!S3E|A6x=
z@c`~X5x==asIB?A@TrWLn!%*pVwL!Ea7d}zS=xftj@Gbc2^}f~ld*q=o>-J0jdE_x
z$dVB$SbDbQKgyQ3n*oD~<j0qnCk}z*u#`W)Eyq1Jh;qWTU9Xx2D>@<BAEPQ;BZiIg
zjh8Nsf^kEF^%9cq*6Ce80GF?9SW?`JQ35ygdF%tLXIr0t%(*NGwbiXt-X}36WJh5G
z6+W`#IFqe*`r15UlvDV3^3j>R@o#MVABBPmq-^A1Ic<z6b715{I|f)<<+Pruy`q|#
zvWvdArbBvbV=bjrxNOt0*BBt7ePvLlnS%&leEtROW&oyn$L?_8q-VDiMvX$hS|bdH
z(U*`6Te~n&U4rD|+e{+tcM~se@3=V}*v%8i>PhOgRa$GozEalO(bl0Gve^ljWfxn?
z(7%%C;;F&*6f}}Bfc20Bh$rIA%Kbb?aurhQ<Xq{x&?>0j(zR3T3AM%9Xjg!wY+s&?
ze!{bgh2y~S=<a!0OxM`;wT~>e@m(XSWE+}b3}`NH^1$n^_gJ|~POELKyln-_ce~k7
z#x)IWyk7VOf9ME*9zTxp3}8KvA8uT2K!Nf50W%yxTnno9#%VoaOGD3%oP{ZclV<9)
zU+}DfCR|I$TfnONmM4-aX`-;FBs>VVj&)@fbx~EBnZ(=ujt%qS@santnZr-MWXkaH
z?jgE_=$65dIJB+%VJ+f^t(rkD<bC4RJkx-aOq{E*FMC89d79{p(3eJ;GbPOwW|E9H
zzCFqh9K7+2iLxf6(lB#zLbC?CpvlP1qqlCZA0Z+R0<M+&(<4i-{Ot2zl1t`vy!X-t
zBuXYyWK2ev5<IvJa7aQ+!tal4fQtrjG=QHWs;T`#8&65d$(@`m36gMgo47r|>AcY5
zaBaO(+d^$?I5)T>YB@4Jx;a=WUWdr6nfRQjg=7u^N|`v--NXZ);kf9~<dE%g<KQJ6
z={~64v0kqMtSwU{##V4TUVMAlmFM_m@ypY1hf`w7yH)+c6Z$W4T*^fxU0{*kHSBmr
zYoje|I9mx#M2UxvXJZ@xCY-yN&mvQ_<yMLqIRZsBqqP+T_BJ`I5#8!XJ4%xt<p@e{
zkW`~pw}Ym;l(6htd18Y02DmfdYNvN&+V3d49lyN!8`j=Ao-4a@W;gAay1CYl<Y4JV
zKpR<0Ns;;0?rYUt%o!xI;*qRYFOg1`86c~nkvVNF^~7U1%kS2eQ?In>Sax2$vy^wu
zB%DutQw1y$qOu+Yx_0ViZJP85`W2(1R%6|J#vtd!&CQ<OaiTDS-kFy#_lWHI`dj0?
zR8Hr64|@BFl=cYGc_<uZ7iJA-R?Akd;cF%wb8}t6JmxM|M&KVVpAp70g5C{x8aOZx
zUiI$4#t<65Hbo28&M1Sv1T%loL;m1rH)1h~YlGUVvk*R&5!ZUXW62hY_yd|MYj0l0
zR~1{^24I0iVyA9rvJtvhOdgctS*0QqDv)x2w&g#@mcbK7V+dPml6T|vbKjvm`?K5f
z0HX(h#m1QO?vxgR_^{<<LJ9=IHrnoB@(O)4@QB3FV2w0&HpK?L%f(+y;=rd_$248R
zR#&A@SD4jP)BJ8jbx07_wfzx$!e7o1LMPB-Z~Hzwu4~w_vkg-&n51AJ4b`s?&g84P
znq&fra~wtJ;Y%7-{hYtH<KZ8i>q1^!r>^V7A<y1PuQvy6A(|N(nSA93T(^d>SpRUC
zZ<+Q7rn^0-*0^ttX}9BWcgOi*#d_cTRBu|KqplLy`Ma94B;o7&wWa{E29`HV;$}DV
z*^5`)9d6jo6K{?W+#erV)|It28b<5N8cBC&2@mk(`ZJ7s;_EmC2;HNhlY|{70aI_T
z;2j~|0&+^+>~}P6B#Cm(ywckvwF2_Fm%p|nYAdP@wJWVWB8G|PPbTj3r_Hw=6$qDm
z=jTFLpc!6CZQI#9uC4@#?B&GqId%AY1~TUTvab!m*%ME}n%bu6%5UE6c!SlxXN*0@
zJZ_MP&7Q}PemnzM&*KLlgC%oa=A0rB5$%w?ge*y-l-XMvC&5yV041DdoL-yb`vI`%
z)t(1zT~C@;FExmk)I<0?#lfWlZ|~o+x6b1)f5v_;ti3r0s)%2tXn+lhi8HR|GSjuP
z){0D-QYLaPC_+0m4*Na7{MpY?T~IyoV4ZXCwAxusozkp=n)f(LVwh^QxvzlIOzGCW
z^w7L3idiLTC$mFHtSHHhZZmVvOj99Gg>GJY(`t+EC`vV>HADO5XdOeMb#YfCsb{+>
zeyC^3sK8<Y--M1>&WWDOraxi=Yh6K<+e2n%hsuuKWaM0F?TG4uY04m-Wxc0%&p?X|
zP-q-n9=!(f8DcIBkevN~>tJo*^(e8vl0g})s*y2=EA0+{?3NIY@{tyT_M;SO4DS^9
zU0{>vq(-*rspEUSAU`;G<4^kWU;d(pyFqMDMferNSH`S+t843|lqkD{3uvbSl6t@e
zfTSyj#6sM2yh~*FR46$!XJJ_y)|DywI+Ye{RN2ZWx;7TAR8guZ)K2%7TqF|6gpeKl
zRCVyu(pH|bXk}e0`@386T-a&HSdn(-#my~$|F?h3^6-v&esqSWIu(CpE}c7ev8}`w
zk&@_}AMK-HlHGv?iu>1y8B1SzQ{nyb#Bp7@?+dGTs&)rTa`E*$z#7;ZU5H6g=sl~6
zx)bWx5-xoY;4wjxLP-Y_i<fT=a42jk1KL*B<%pycB^#~pSl5;J_is3z9(`S?+pm`(
zXk?+iszkbk^BXpJ1?C#g0(Cp~(pw=)sUTPwUWQ=-Oj)?jqtlYlxKh=-tT`*>WA?`8
zr$@Nz9pj=*IsT0|xLWa-Ykl%N0$r2jd+qlG#-GY~^6`c(o7-|pnOt_{DbCCk4LWo%
zq!K16_G~Z1(<oPQMh?%m{72ie_<3W3&@a~&wwzN!C?qo%YX39aGU_GX0U;qN%Iya;
zEO}6zW%M78acx6Uf|##;!C=H__O+w6dElF#7@xmum}yF}?ksIW)z`Y7eE8IDLOvc2
zTqI8@ej=tkG>68ye1!pgY{%M|Ql{*7zRJX#uXq_<u=e6i_G$A)<2^WyGWKqSEZG@Z
z>SE`!d~C;JE;Z15=e(}`=Ih_V`hpjCzX+Gw!h6~>V2q|58yiP`FV7;ZcE$|md4}v@
zd8#Xha=5wSbxIuT5v_{Z%212tDNN7x^k9(m^8Em=wq5|^;U#&g;KMt9_x>&KPsgph
z_@eG#lbr=<UE^5t`Mo|*M8p|dO0Ygtlxn89Qi;;ZOwLo-Rf;*t?B<!du1rM$rL~6S
zMlOOz2gP+gd;B0BtOOt#uRa*EBL1F{gz3Uo?b0lHN#Q1wWen&^WU!=03F7E5eyv&O
zesWU>uX2{{Y#1ed10-{QIua0x?X|ZPcH1xby*8Nw-17VAo!nb=T7Mot{_zZ8J&!-r
zxa>kB-uL8yK`;iAgbq_84%wj5Gs+bPwfM4a4$A)Zi>H(*xp;OoqU=GU84MhY#~x27
z-iv36awZOS^Id&p`><_u9vKk_pcKgyWtxzAha%*9VxA^mzkEe&Cu$Ye<C#?)Kx&u=
zWERgK+=S2}@J+lsSY+xzj#&UX-rBd}9UD8HZZ2$_Q)WLGicECiSQjs)(2>>D$QTFo
z=jWh1lrbL~|I0e4)$XwPJj<qA;Rj-ha)Z#Xfw+fo*t?U{W%7=LQc+EqHF8N9o!U+`
zZFGa2ZF7|U<s(O9L0ldTsnA|9pr-(@hHqko9z^^e33eWv2xPqUMO!OUO^OB32}7L!
zvT7CAPTP1}S8o{QD;5XxM^5rsY1gN)5vj&c#aH$r7tI6Y@~JL?k_&`3Q|2^<9jA>Q
zXx1aJ9ZyOz2N4(#mveJy6DgcGF@<ifi-Uz05}WoyB1#woKx@`X)}1w?Ue=Q2a^X5U
z;&CN@tVZh&zD?6a?wQ;(n6cmO_~PybM^P4xhqK$(Wx@I+Y*nNd4yj43;V8S=oZ~2~
zqx)Lis7kPEXXy*~>&bQXqSeFDy&+{xM=3z8qbQXy`jyU@c_17@+9{MSEFvBgXL6)8
zA=Kff#_fq*ym)C=tCf^iQp%W4)Y>>6?^)~lYTp=TKViosjT^Un{OWobY#qo<qMYy&
zLb0*2V*Nd73m*Ua$8okxihKE?>pJ#rkpXDSRf&p(?#HzbjK98!ZCCQ0)@q*~5dGo!
zFvj<F<oB|hdS~A<k>dt7b7r%0IUC2TjJ7BvC{E&Qi4SdAjglh)@Cn#*8tZ1$#<~uC
z|9~yG-)GCg#1${5#n`&(83bxXusg&{b<!xmOa5MeMqAbmOum+a9yUAJ;BL#Z{nnnE
zPYOVLfCjLg;}RW6l^)W0p&@j&N=Fy7(Y(R+j=0>D>)W_}^MQq4Jw~=)T;11CV2qD2
zIL!mVGCr_l)GHCBkfst$L7`i{*zwhwEZf5NGCZG^uvL?=<K+6IW|YyXa3oAD^1gV9
zVQr1K_wTq#VD^hGk$stH+13Fzya9{9NF*fu=TZuk6!xfSOU%2O8xf{?X4MtUJBD7s
zql+!TI2XCP_gCqTt97(BzZPNXjnk=ee?D=4I(g$Ji=>GSkmhoRvF)Lkvr6Jj_B}4n
zWE-8WFLNO&k&`zT@@w+7K9$0hGbQ1ywbl^nq>`}ik8QnTG)Nny6K-mJzn!14pNVX4
zZJ2#LW$0kp(=$81AB_SoXXJR^wXT_wa@5>xal@7AZFZcto8eo|<@KC-dG-F=M)IMR
z`WZ8-MtXo~&*R5Go&l`q@q>-ASyiG6%z=L&u@4pT&K{P%1R!MrGFS>KxtE4?F}e)l
zUqP%Q*1W5afh9i*lDKee&J(FC*@PSG?liYfjx5N~Tsbx4{i>{L+@{3mFJ5v!uROdz
zLMrqGO$sovi&VwBR=T=&(~V^2vg01*j8X3rtq8oYk!idkPmMfPTJBhG)XBS;OjE(y
zMCuul6;eZV26d)H$EdlwldAwTLh%;A4rwu2;*P?L{R>{addY9^zvXy-<g}h?+F5Wd
zuPzdmmzRvi<#2MzSpTyiJw=IthrVVR$!t!klBrfM;ie`MGk5#XILOSwly4s%xj(<*
zwCq{?iKX2z_nF=wY5kF<-#gtosNQ0*D=vE9QF1pX6}q^h0ioCBhhrl8MbS)<1bGU~
zQH+|Lnmw{SBhhzVHKem^pwx!zt)zMhxQ8Pi8fBLQGk`m;i{WGgVmuHwhF~RENBhMO
z2%UlSskq7)JMQ4gJEG~bCQzrpr6eR#gp=!~r0if~Pt-mfzLk!$N&}gILJsW=QFrSK
z8A}!=Nd?Hx{>Yh<Gh`)EDJB%}e?U*rm(HoL{D*(~756V+@<09E-!bXJP8SD{gCW9|
zLzUzr7@o3GCS%Ttk`m|AIp)q|n?Mu$4x0Bc8DF0lj@HoP`f?^pla-S!XosXbEUl<0
zJ$d<0v7u)JS(PMhEBni_tDUT{ERECg%&AtYb!x42GuBZWG1h2^-1;hu$RUKDI}@o#
ziA3*FN^n(Q5+IiLj`kJb2`6AjX2+|X(dmYV-Ip}>bel<2akgoj>AmvpSHI!7yd%qm
znRiqZygZ`JcfLdQN;26>1T}dlC>XNK7CM51x#7(lYIio33pe9non%ymif8AudPbs)
z3u|{ODP?lWNQe+y2qYSvtwed#+ZaI-3=n@f%(hMRrBTB*bmrf<_^>>&?Z5nSeI|GD
zvhQnKwy<T94*CG;7V=C|e@ty>x;1L=n=P9ZkiyiJWUfWImE<MpDbsK*$TUyPb77JK
zxszn2807zhwmbl|w3PuuN`E3-Hb~59s+cHMDr)Uw%YcWVjcgK3r;9DG=f=68IrYYh
z((3AGvJAFRbllB;<~j9Hw_x6p0B2&P4%QX4k(eBgjv!q-^yVyQKaqYITQ254f=Wh{
z+tUU3F;-QkNCHjbka_K6O=Slv1Zsg52RB@dKY^NmhTc%?q@2k)ZJ@u~GMg={y55q5
zZUUB^$qhh`a;HwRyd?*(jX<u9!K|ZQT?GFF2KHb2cp9c>Uw;~<kSRej5~~9Rskniu
zpizq3&CBk*Sx`k-{X91OIp$4r#GJ`H)?Jw<64|v*bEn(l%&L9?(?p#4_vXB^o6kI+
zR;HY}nF?p?ETd!%_umtll(K{7QqYcfF&-%A4tjyJb`~{S5ws-ena^K*$?g1tH}Af2
zkUY8CZ8b2<?%|rVN4sS&N4|n#T|cX_E5-TBOQTig{rSvssXU%t%dV<XYxB<FT+Uw+
zkdn3%Hrp~qes(n}-pPENzZ=MwqAJ1xatW3v0fvuq+{{YSimb-tx9|Ce-(ixN4-@V5
zNJ&SMR8KVPgeS9_13L<>WUr_&FUNP5oZnMXp9hI^v7z4JLwrq*q|)X_AoMF#`jS|3
z;CpQ=DGAeDB3T@}#8{K#d+d_E81AxNf$ssFehhyN;QRtWkJ4!)aa_-|O%L#S{Fuix
zfb~3nuyKu-e;*z$+x9rh$o#hhWEmo+9i?JXGAATeB=n(>P5v8S(sAMiOQ<A+<a#h0
z8zA1hsZqOn7ZEjfbD^j1+0HGClb&S%uc3*Y5>rkg7Vp>Dt$Vv-wP@=qNOssmQx8U&
zM*X>fWiE_`$mUq+KyCUHaUEnj71=iSp#m#NDs=O1Ev<V7y@#{BfhrOQ<JCR5&Z2Lu
zjd9aNHa6C^7dV7$w+X@Nz~qT(+9Bx?vz2o_lQg01-BD@_y)USRMgh*QN>?dw#c)!z
zMrlt#Wx{q1Ugr+Yj3O7YSnn1XpJ`mLj9PxS%&(<rv#!DZ*=WuKAs9}`WlcJO70%i9
zM95Vz{AAe0S>|F7A7_yMvd87R#~q5#JJ<$XgQ4*V(alM52A$hQqqExTe@~ranH*ov
zt{Cq)UdgkPN!VH-4u5tOiy0ozHdEd{JaSW9vm=RTYm*F+A@1*!=^sI_kn+f+LKmfp
zQN_ywa%--QXU;-tDCBV4>;kZlj;Xf&=T~^W@3@=wlsRfZH(^!d)H;jyU`<BHBy3|6
zZcf|w5iFLPZ9bqe%C!A`h7QGMvXK$?l7Qhdddb3+gk3Hi<^$(m7}-sYb*{a0K0naf
z!eAh(d3P#jHJ13yJ?@~FgQQ^9vRy~a(J)oHUPo7pZ#-_p-n@ezL+5bhu}2qD^_ke-
zC0=|0ejmIzSCbN@5`O@ax_bPlHlCcxF}H#zE*NWc>{(mKUd5A2M0t-!0M}WqMzX^q
zDDrtvVWB?fvVpe*D47OTgNtm>7Q_5=Z23F(Qu&kFa=v;GWiVPc|H1I21(Cdn_rJE~
z^RQ)8dhhZ2SLZ<3)Du8;`+d5cJ{}cgn<sT)PCGd3K)G1?!EAYSyd!xcN96)WZ(c+s
z6p19zWn9^#7@7=i7QmFBq27pL=<v9Vgnnep7jWe1{uz%L{~P$g?=jKwP#u5*PyM35
z=y7q^HoF;iu`#gH({(of8$iQgXLXcnUbYEJvE+$|@TEujHsO_lS2=)?lXt{cyPR>u
zbqsixr8myaebSOLIXzt)+le4tY(3bC;P#9-AyLJv7`i!=v9(5P)pwlaNij@Gk>uQz
zQf&ZBFj%<Pp4|XhMyZ?YGr2E)u9e4S4d32Nj=Np_%m@*j#k+l|i~qU#OL4<Cw=_Pv
zjk77vYo46R?%PgTn3AjCwWV@;JhHpnlS&3#y-FZvJ)BoIM)vj|<5gwbqleBMv40J{
z$ZO9R*<6fmm5pz|%e;mMKjvfXm43ea^M{oD^*w;Abv%6WCkf=M2VGsmIJbseNb8|h
z`#gU1;~Btu9)Ff$;lPf4X)C=j&#3p-&{QBf2xJH*UXF0qM^>vm>KP)tSb8R_A9Oh-
z4rPbNa;|FR>X~7gGSi$Ob#ziB%LXbWgS9D-QhE3Oo*l~k*#W`n$}Sk%86xhgLduzW
z+R-$$KBhvSCyvL3$94Vxv-fAuktEBSC-ze__Xq%)wVm}YeP_CdEkX((_`?7Hkrbf_
z`9K*FawtRLO!r;RR+W{R03yQO%vAZHYUUnD)Ov1BSJr{0AOQp-d@(mQRsB@;Q<jG_
z=hKCEZ{ARvu{NlH9w*j^m86YJd1P%1NM*~6uu<aZ-Y0-Zw1!h^7T}j}&<*9(l&>yu
zc~H)xynlS)v|OC*){K&ZUr3!-L!^<I04L?_n!^hhm4i9YNYWZ$L07^V?}QW%xfo`v
zD{|b?zMRiAZCtqJQ8V|=c=z=?NQJxOe~;x`zJB{BPUm+(*}SE2ic}@|%EHl&d&UT|
zT_LB4ds~jSxdHTbz9s3`9O|*_=b?9H%yHAkWi8~yb(`*-eXiTOeP(l8*i5vpWW|5)
zb*j&3^{0mMN9g+kn*;Hbv%m@TGuRxj6mK0ZLX|?*%AyyF6&~9eO&Lp>BL^nQOw2gD
z-66^xW}4yc+_Mucr|i5d+MU?lq^dz}#-HAN#iR}(CxPj3bdbZ8GzDRE?PVu2j*;gh
zhiRr{V`cGj-%F)_{m61DTuz1a+E{AiqJ<`KJl>#cXjvh42}v!uTikn4Tj<@vHW15=
zw<<Mc8ZI1Ild)9gqLuTqQffgo$k01c#SZAX0-ga0LtQaPCxYZyQj{PO<-TFmV(v6f
zO*ph(ew)bC#NqD9-50O0bH&!iMK4%wXr8HZ;&gfAq|uaN7^}GtRD+e(xssI9G%_%}
z3oC%Mb?j^4&VpBHB`-bb#(zYGE&EMN(kOv;Crqnn?jGhF@|2vLuU^pBoHGSuEi1i@
za9hZ-D}@*@(p};1OI(d-yxWgdjDhcx<per~9x8Yr;6#%~lg6qS*U39sR_-{+0Tp*3
z4>l8;HbAWY4248NL_MR{G@P9PRepdhry9*qi0lS|d^~cP{~cu6TIh-d<|X_3wCG17
z>-HI0Ud322&SjyL)pZp_sxSc<z|d9f`PBTiMsU(}P(r}V4WpsVZb#7uQ?p;ZEXV4N
z455Uu@v*^PO!o!JJoy_}uXTv>qXDpDglRWyY0)T86(D3~L$WF=4HKbBrCMcaE5$00
zcELN0X718~gG^*=WD<uoVJVt7;NUDtkc^kXi3V@2>!lGeHKD2SeqFgMD>c9y2_Qr8
z0B{!Bvb#C}Bc0pJNQW6q!daX=emN~H4=0wd-}C<dp7)nCrvRhZron000j%hniqJD^
zd--V!`V9!(EJQ_s%X+cKg0Yx!QKi;(Gz&5C6s0{nGoDDd4`I~j(Hw!Z<X|7_S?Xfj
zJwpU!O7PDn8kspC_{}GGq^7*PJ5eqdzB+%!Z|{E3d^mFd@Rrm0ee4!PT7ecZ0?+R9
z09f7@PfC#~@8Pthn-1vYdV_w$wpkpYXgnbiFB$CRp$;Y{a?U~0u5P~rEd|kjbne|s
zR1-)9tiMqTH)w4SeDyCi!xSFR=Z`Is@@4$U@d9AIjQ`EXP+r%s0*qLIM}?x6+A7Ul
z(N^<P5#L7LfKw9Lyu?i{JO70Qpn~io_TEWnBs}fT-U%?PHmZ6HTSLPD+cw9p7Azt!
zJ(|z}Q{_?$=XGU$Tqx&-wN;kVIIWd59U$ibp_H|pDY`PEsS?rMtB8$fuw-+>VU=F?
zqT~IZnW0o`tgUb=7w?Xu>c+VBLY=1>SG#+%?PrhDxxsGZX+*h0C^@spR@!q`brV4%
zBW=n|QzmC&Su19RyGdxRa{us#^W}jn_Ly*_RO6K?Uf*=50}q`qXLy(Kb%kd1&u$r2
zyF?+M36U|sem}{La<*sQpRawOP_IFQ&HR0$biYdvf5c-y|27_B$gu=5ag_dIP+FmA
zrCAB?jp}RH%+v5Sc-Ca{5(t5uCN`HMfI;z=_RhN&7bA@9NCVc^Shbg?j9xhLTsIif
z?;|3|e&Gn2BF+`JTng*C(9VVRvhZ-ea4uf<P?N^{Qy8Fo`Pv&<Ayd25pUuqmb=#*>
zaZZi3d2@)W8u8SpG0IqP#&!<k-fV-L`+IGik0B~Hj^oGdnWH9#iJQSq44K5|+$!hG
z#lZ_+W72}ew|)Tf?6b}+jwvwsiMA;9b|w8;Y_};_&uF^?4L%Pqb5j%VHYA<kT4D{R
zAhz!O8foM70Q%y&z5-&uFrM7fFM52lEK6`IdAoOCJ3`e$(L%NQM3$jY{8-70nbguw
z;rC3MYv8pvZ1K0qAdCLKvW$fM*j5JP8P9*MEbnx2pfmltuM%^+Og$sZODimO^>WKA
zLD>T<JtJesIR!4u(w_}<y5ey;@FXij!U6x=kmZrE9qY?Wm7W-PAAkziuRgiiZ~Dv6
zay;F;Hd|@;8~{XIClh6I=I<%C(zJzr9eSDUrLu~b5Mv<ONz>zbx!u`=fT7!@QwT(m
zWMmaOB)h1xG>y*QPOz=ohKbbex=ku}wyU`9XIWxDXq+z#m-E7MUbtKqE|&|7L|L0{
z66}G9>-E>OD6i6SZPSL<3A)6K#4ojYW2fjgy3ZEw#$wNI-gNTrxEA~0$gte)F1A_c
zp8a?&fPFu)o|PpYSl42xRZ>n!PAuiZS}(hO_30OJ&rS`pb=H2yA=*#I({$+Fjd_NB
zUlV;_bRWhuWn)i5BKtZ54XTlxw(b8zj>BHMx}g5@nAdH`>A%UCL!);#9I=6y@rxcW
z0M^U+Z#M)ifTbxPf^SJ_cx&=i7fM@M+J!=m2q?pl%_gXMHcn}fvqrSYjW+LkZD0$@
zEchAx)Puc5oC{<%cOF}vymfVQkY*C`?keVOo&}4Jc+SzXG^MO7Zy(P*oX(uSeD7eE
zR_>QGU*A9Q>eZ(lk9TzR<^J?Fm4c??B^61c#7aX+`Z?5$0}~Z>4jj=i)VH)+#i&+!
zEDPsz;!+lBt<HTj4E~PBIr!JfPoK{~h^Uoh45JVP>YCZg0~pm6uDm@8Jx__8gxfiB
zS|-+-xqn|NWhLJzwLS9B@BW4J<tqk&wabC8&ZF)zqIh;2Fp|8Ktod9KPEA+1mLdx)
z7M5q6^x_8D9(>QzGK~_aN|FOP_txRDIt(Y-H6ruO*x=5=cFDCW-7e|r=(fo($~g{W
z!~b_jFr@=fy-ZZS3({%5P_$AxyW>ltnFFxaR+2&GNM<5Wf+ZuTJjzr@`KwhD$~<Lq
z@*<@s;(RQ4H>`!ir724(WP@sLl*@?e=vD@}92;QSGn}Txh00PJZ{Iv{IWMg5&)lC+
z{P}dxsVa*qx36!RB}c}YdQ%6pAgp$mPkXk3G@e>hZs-ZtOClATDfi{f!{X;7N;IS!
z%EL*Om~Air@YS8^lhq^JJ&UNDcc#&;)pnVKQRcH)<7^d8?$DpCvPwg<QDvp*h5N4_
zS#74;=UA#>m$;c8J@)1d4tRTZh^FS5pHb2!PEt2))Gia+0hGP*ACG*vHRj!51@DZO
zl9y_>X5`5Y``*23O5O-$$`ghnEi*QmFxWf!?1wIEM@FAoz<VB@zOQ~f4)1r4XJmQG
zw+Ic@f|_Tsoy*GF3Ja$XWcl10t0h`4&TWY}FIj?4WMhN;-Lr4z*~y7Bu7`i9EE7Qn
zRp+Tf1e+$9@?R;-F!ZcVk(ty3O+ICbooGj;(v>V<&fK3)^iHhPkj>CZyVbWj+5TI?
z%eZuu5~d*?6^pXB{#nngl$=<oe|xgbb749^-_tDBt4r$FKAylYM4orIA(F`Xhy=7*
zwV>KKl_mJbmIx3pG)ZX6h+e2B+~hkZnaSCC40D<=O*9qg-K+!Xnu4R!<wrvg`kwh%
zl*iI2t*}T#YooOQFGXT?cSx=VkCV{MGb?3ENEYswmGipt)z|Mh-=BGW`@p;NJ@@OG
zg@mQVtHbS9a=4cf>)@Mb|5)$T8DmIT@Qhf3VwHF2N0zm*w6-nsp*~$#Y}x&`*|YJ+
z)HAVV1Jp)XD%QBfN7c52k5WY%tyPw~q9}*M<W(58G93@(!-2=k*Idr;H^^!i+#Oj{
z=0Wf#MwZ~vi;{cW$uDe$h<np^l%>*l!PYH{uxFG~+&|0Nxy&3)^d7ELn#kgb2fds+
zB_Zd0t&{hG={}38gLoT&-t}~UjtL9j2(13X?Ch8EjpGHtdKv$1#=siFkqW+yW|dWy
z)s$sDN5r>!^1cvB{B3Nos;-(>Tbb0Hq>?GVJPIi-=o*_xYCVFOXekF!k%Q=5sjU&5
zhKNiN8FDgdL*x3^P~OI+tek4)biQypU3hc<o^xIKi!PMXuq0lqUsg(W>cdsfG-=`7
z#)b8WWOHN8NeDHz|E@iH##R7Nibs#9?A!p&v#)ATQMSlxima6ZVyVyf-gNWMMtGEF
zm(mMeK|4q#5b42*I$G8{|JCtuOR^PfXO5!W<;)aiWXEGdRXN{(%~BtU;s`c}am$bC
zPL!Iz>e0Q#*0Dr(5$|UT1pE)X#nv@+ckmM;87F;}F!C}@d#6b^fRZGS3~K;;7CZvh
z0Wpc^#)vw8Fg|#BZxuZMk&k8=;hc5gGYrRww3fEgw6fYlrBb!f`<I0|Ju+j8x`9?$
z-Kc9onvpbnXD&0$s;`dX22j6<ZVUn&0=g}gG0l@?W|BY8SPv%2jn=?`-akI_SXNG_
z3zzf4{o{%Ia^X~0isH||YFJf%5-)2KddE;NF^Ie-8+jwcEbD-r_7}zHV`&ZsNN*93
z^XAF+whc6~6o9{%3&lbF5hRB3Xx-woZ7GgrC+_$<X}2H>Gnvnhcidzz4@#|)TjlNP
zk)>1;D<<CE)ON?&*yXp^JIJ516TuGfTX2Wl3~Lsca$I;%t=kUr1ZejI^zN$NQSN(#
zs=Bd^ASyI$+c((C1ARa93401IUq`ZGbcDJxto@#UF^2L7vb>d}nkl6fT5DX|g{pM~
zl=iaRTOzCqjfpAEB+cZe&XJXHg{0xA=0;l!rPk}Y|DLiuk<jLPZdBrD{{3WGU6x7J
z%U*)luWP+sah-pED$5ut=K+P<(BA=~_kfF8oL_gGG~F)pi7>M3<9_lOn;9i8Yk!ln
zyxSK0IeyUScE>M4A@jlf`W|PfR@SyqTcPOUfN(TciNY?=*b4x6Z3`-j_mnmyPb76<
z)tZA1DK|vCyJe|0>@gds4oh?do>hY*R(q*wW)PTmz)^4;H#u!Y3j6l{;gQE>;obR>
z({kbcdgfhQIax)RNfG=N>pT3l4X73X*6uZlc{1yb75sO}e%+^K4W85;k7US)b)L5Y
z<Qm&u*arL%?;AQ<Hv4^dr2`+^t#QmVrVjIgo1<`hr--cN)O`K7$aWXq*_t~Y)%~kv
zBjL(VAA#)HcgziY_c5%$w+Dh{WM2;Z*m^0ZAIkt$-^cSj4XOeOosg{eUd(%`h4}mT
zl4JSCU$%Sx#VF*L@qNY%fb}x|HydMf0<L@lyhdBCVa@CE&std4N7UR{)`+O6SQyKj
zJJD!Nmb|M<ZNL$c1A_oiBSkp?tdzi#ASnk3l+fhFtyLRp5|N=yCI^6(P*%*8wXLM&
zR2G-Da^H-{(}l;=g|F`4@lY22#LAo|Zqk9O4VINv-C-~7Ldy+J9x)d2ekQ}UGNV!v
zU{28`q#Fi4U48F&<v(Q_P->w>G(5%TuewuO;$>1!t{Z?+w<kr0BRrb@c7FyKMCIZU
z*fdRBc8Dc}iJQYMQY%%@+@dVe1*)aQ?M-G~3#a7`wJjJOfY^*Iv8@cM6Msv}Zj`40
zuq4_ON96rlUk!64eovc0HB6H}4_>0?0GZe9>prDh0C7Ul*4L>5D&zb<G>rIOS|adG
zU=|(x`i4J$BqPkXF1ifVq?cfsQEGF3jkZ#_1gAvpQ|N6Bto25Gn60QZRy(+%SsF|l
zKDC#vHgk>yTPw9zddb06+S_nq#oNzsI4rYdX31P?!OZ*XkPMPXTs6Vl`v=~gPdq*@
zoG&YH9v^wol}B9>%Lo~(v8IkG?pLws_G3}H>n$fbh_$&!-fXSQIUVz(nQ|$sJ83xu
z7>2L&xW{2I+veCNGkW8NF3U+RjO^aU&p)#Z6}zL@vngWzCld~2zPR~>LSZ^2Qdvo5
z<$wL^Yc6NyW-gd?N2SY~Y}birVMu`NlEnAWo(0-hP)<8A!vShP#&(^ypC=>k$?vPx
zC^PA$8Il4Z5GI+KB+*J8*Vb?PnreH%f=`TP+ho|r9Rl6?k=@tkczlfG+hn<SZ)@6E
z%fi}LF6|u67>euUF3aj`T?|%pp2dj+nTb>snY`@G%Zk0@7g|9sl;AZQ<$B*!mM6(v
z;<IT2$u<BimA_t=)%?B)j+d^Z_W;)T{~1|UAWeB>6K&<_9yE~@`e~FucG_zh%QH!=
zdr7+`cdbBtn{Yk$w<*itA%o94zV_s`2pD{%vTQ?chC%0RtY-<C(pqJ$3roGQwuKhN
z|1E$k#e+|^%b{IRDw&CfVRd41iKe!OH4U0~r7bI^^tHqmIN7dhlp3o#0lTVsXY64+
zMA>WyYQ-D7FgdXD{{21omkV!C58PiCzAh)uy0TKaHNURWVRk&o?Yje5js3G9P*>P@
z*0Y$iXF)B?1!VFDMeM{+x>}8SCbBk%9u5abyUoea$(tP8W-m%S#C|;4u1+LzbGX6C
zyn5w(P|j|;sCt%nZwj@oUm0^MA-jIeZu5GlcT~pnk>v840-V}<H1F@#!K<XQ3*7p8
z@oN$0Y1-yN5;^7IRm775t~kTwTjlslFJJvlp1$xh{szYjfb}wd($Tk>5k2j?C2SzC
zF7T98Z568vi>|Cz(J2Tice-dO!+6PwC5=T_vP`r{gIA0sp*@nH=%SU!Wnpd5Gnx!1
zrCjPtsTtuGRME0hIj|xel8_}Kv+QjhNbvscJ%4!j$X9RP@_4>*Usf*W#P6u0qO7e@
zsaWz>%}rLCXvD7#F%1rgjf-owl__U#4o40qB!eZmajqsefu=-jZWLev_LygGB^XAn
z6V5U?gGAHzKzb)B5oDSm``md~piZ<JhHh|1HC-a==1xf~E7lqp6N=1ex?!5{&~!sO
ze2SQ$R)FTpb`M_lTM|z}mdIQQC=F>5(Ulaf-y`;|PIfuT6o!OEc_Gk2GIImp4b(7X
z)`qGvPX~lVYfh#)rOY8^5Q}ts&ukmi2t7^tM@4yiMA|k0_ahyhoyxk=H9)^c|5s6}
zG$ebu&<2L~V}u+_@SB>l>Pof791e>1rU8Tln-tN7b6F`-uE%Z}mj<QUSeI5;ED8jQ
zjjN7~4F|9+N(L;$!^0zQ-@WI{_wQJ1<1Da7x@S`7+{-c{GC84Z5T~jd+S)MS9k_DR
zSq0~y@!G--Hi?X>G`YigS(h<K?&$S0LYL{iBj#mRIyOYZ$m~dI<1-syUpJ!rM=UZA
zXUs}uK)BI<UM>_}IWG^Cb@9k*GpriL7QE}x_AL7yk82GFxM}mU(3Ht}iY}><aZ&3?
z`R|_*_j25UMTfWB^@d-28eD=7HE&>_^B76AVowg3nvv$@0E_B~KL3z0fYPycH-O=n
zFuqZiQ=?TcpINk0tRS<JP_E(+Y(?#w+(AuNDJ(Q=BdfSs4FJo?;^e>B%2Fz=0}{!8
zfGiWb*rFkcm!$<)@?R~>8a#BS4o2JloN<kV(%sXAuVtAYjPZcc@OndLwk~T4scGmI
zcOrwBz5Diw&~K5rbtTJ65@DKV%<3;+md!JKqeaC4A%5@}SeDy@AH7oEV>ZD{<;l0r
zo2b;nqJ<(An_Q>XHCBD<cC~rgeuAQIp=K7c8m%PeG%=?ffnG(_?b<dioJ$!%u3txw
zp89K+AkB%}bjxK~g6q-Gnu6rJ<2$y~(VC1}EAQUC<=yF#x99h)2qa-nvv)DB4tCYv
zt&|NA9rPSl+@6&onPa+Y%w1{#!eN?_Ce(;tDjIydo>Me@kjS|9a_Iz78rk_=6Fsvt
z=0sed2p){zx)t{7fB^j@IVDE(wW(`EnsLa&Dn_+NSsOJqyh)6g`0j&Ldypt~?l|+x
zf=N<<iN01gg=SdW@sC050sE{lh71eAStkP3&zHo@0!@u1IZ6j@n7#g3{k=^!_Ox!h
ze7hX?t4z<lwo#?DD_iMh{Bp+&fb}wdvJsUl1ExqKXtS$5sz(|;a^m&g{hy^?9^_vZ
ze#BOby_Mq>ZKwz)&B-KNr#mt%`(|oEklYNCdDjik?2;(I^R!$~hp#A^MvnAaA!P|g
z?}k1}xGW3r@85H}e_&ZxR&%94v9+jmE+9xjlCDyNUn1^83=QSjv{9`_Qny6PzLA)y
zt$CT7>?SqlSj~B}U<w~=7kwFN7-QD`n@K34aH=)l>meJ4!3k^`2|Yq2m?>(WJ!Uws
zMN_DjFquf{z?3JHTO=JoCT#Hb>;Np;UDI&Z-DsF3lwHpb6EAj@FpPS}R3f5{##ksl
zt~|P<NarT<?pG?1bF_-qD3OrFlzdxMaWAb$(D$=Ag6ZRz?*_H@JzPF&4LfdQ17muu
z&Z0E7mnF0y{P&Vu|M;HpcyBs2ao&jxpe+o#o|zae+|@8Gtktu*hAhPOs6tcE8cC+8
zR$#)!yA*9DpZNaLIWOXCSys-cGmobSib0!i!GtMIq?*XOMYcWSxB;3PTzM7-ucB_x
ziXb`6Uj1;5q*%A5)-b+1*tzk#1nf4lIK(aBvGz86WqUtE_WRFVufWLKaG)?8ax+SE
zKESfBXshm6J4PL#dq*z_$Gn&RTyILfTqsAUSV;-3{?Ju5vMg-3N3j#EMCVsY5>9HA
zOm8QTa^T(lZ>7eOp=sNmrm6wt_s@UuC_Ws!_gcKLeQdu8e5B)BWEt@_>}BbVY7Hif
zwdWtdA}^2i<(aHrZr=b}l#Q5<a^Rda6i%hI=<PQ$5&U|-Tb2#7?=RLtq5kjRL6$p@
zUGMzc_bZan>y4f+p}j1NNUU99PcGeO+pL$w#VXpxriO_z%K0|OcmFjq!;*tNu|@u}
zWjW3{$f&<dkoSL<vAnjQuip5P$#cK`XQrTM$ha)315&0`6v+-MYA+M*pe%vKcDuZX
zqD*MbX*mR=YIUwyEw1lv4V}Cz0|X?>(Od)!Nr`!$y~A$xHCk1f%*b6Y<LB;GsJ;g^
zmh+kAe5Nc5I!)yH-~{!tw_`FoYe(_3A1O))HMnb^ZZ+e}b`=ZLRw~#RtVI^lFj>3w
znQh<+Ts;p@SPDBPlyWb@l})^cHQb-6Croe0)9Uh*lSEdg?@>;B;f36rG1M2vwwAli
zE*MGNH!+KSOuAjvs|akfE5@Dnu@5_ky|J?`+&1ZUy^LlLeRm!+ApV4!dQg8~CGp(&
zE>%4~J<3tPk8*z*ANzO#uwKSbI+6%P22d7pW!S>8qq_r&N<k@9snM!iHzgESvPCw7
z64VvODDP4;vKZdb%s2I{zR{n{f{KxnacS+z^*OM#6bH996VZa%!wD`;D-+#d1E!4R
z#QAdK;q4o~xcdZo{Ry-M18!xa8Z2tupBK*S>a>UE{8nm4lTE>tRw>O`D0ou08_-Kz
znNlLBBUAFSjS7Q?NljUn#lf&}94wf#WGKR}pxud({?9z^IeYInG*YmeRG&LZh?*<l
zL3$YmWrTAExR8XKn**kaoQ%8U4XP^-_ZRL@#^bwtF6(0`Gmpq@(bG_cJpvUM9j-dr
zHloR*6%(aNb%&-!bUrzki3qglJKlK-9HVH%k!?6oNJ_|*$1~4N-cq>wqH;x_7?)@<
z@xV8i34S=P)oG~O05(3B;g2n#74s5Ru?;{DbA)0JK5L2&W!gxxMf7^6Mt`1>kP&W_
zp%fCvEK_95Rd?bH*0Ok+(xhCPN9}D8S9m1v=8!qg6Phxuh4CKEAE1&Ya~FoO?V1E!
zmJ<)^WlhJF`0d>%v?MIIx6FCwI3IZT^*xV|XDrW<GC4UFHHMe@7IzNR`Z<+q&KY++
z+#<D*q_}geUPLP*9F8;AjI#7K>Z=jh2x*_;VcI)l-8d$`Y5c9X(e4?GZfnaR+(o=>
z!^^=MO&ZJWymeDPa(i2u#F*w=kVZ=G{p!v@-mb=F+x^x=N(Yk62pMYztvX(>hIGT*
zBI^YS8Fu+}U)lziF1~ebK~_y9bHeA8r|uT`j-rjuJCa1mk^@Za9!fatKd78H?sf0L
z?499{XnY@8CbXZXhUEyFSn|wV+Ox9UqKd4S$)0s_;37d)VFi-<q_tX|15zs&b*4A{
z5wd(6-Ea*C_`59Ue-~M9;qa4v|0)V(VLHs-BqjJV`~5$a<t!2zCB1chU!{HY@<G?T
z`=ywoWYA?Jq|4$6u?2YcsmrpYG)g-ECS+NITz&{)`ku1fL5=Z9cCsv+e*XxD2p$+|
zn@)(S_9q(Mq^nd~s#varE6F4{-F7r(o(;DFpk0nlFL=N0#8C=`L~Wi0x0>^0sWw*a
zJuCYvcaYHu;Ei60E3ptvgc4cA$yPFnnGTZ0Jz_kbPdqLcUcWwacQf;!<TsR*SaQb5
zm?i%C&6k|kg+f$qNbjE7P_0f_&xY|>4g&FJXp78B!GlmJYptM(L>_yb3%U%QydR!n
zs*y!$V=kSyE&P{m=bCps^^SZxp4$kn?e#KHO?j+NUs7Aew1D~bc}bqep1tRj>e<^J
z1oq(UoD<1%$gYwUF4gU-Jof<E{;qmXrvH@sGtJ#zho1S`Iq%dgWOXY+bUo>`5v|+e
zW<N}hW&f>5Q2d!TaeK~{OV=YW<Ci{O0IZkslZ_4-cxKDC;U&6qjQ3$fp&HE<B2jj<
zGpWX&)c2=hbZ_QYGNqYNgOD3DR;{SGkzOO!zeiv&Cw)}4NT=;x%KFts=OK};ONe;{
z(7kTw>AAZ(GEWB{iYty+>A<;GbY02OMz)3Zro98(ZmwEH$(%o@|HZSaazt-4i75$f
zPp&#LueLd`Fm$&UpJokoXsACY#_t2@k`1Hj^R$1@?jet7owB!K&PkZWI~iu2x2P-^
zYF&2<t)JaX_j*)jukeN*6T*rQE)g+g^1JE=!~QP2!Q3tV@ifB==xux@oDTmN*MM9#
zYU>xXkDq^|ewC*Nz;4QckqS2M?&Hw1E)JeW)UDet$74>qQ)Qh4YOrcV=E4?Z>&6|h
zTj~8PG80VQu~s$!OWi!BK{eYVl-Q8xeG6w|&*?H!d#5!SE5$#HQEFi+3)7Uiy_q>i
zlY;YM<}e?*JKVB9EG)88!?+ytv-9H3TI61BV7s@G&X5#RrK9(KnWf&jX{_UCR@iuH
zzjk(eNW5gSt$2&c9#U(YlhXs0`y-8LdBQsci7;nDre4ZpB)qF)H@i1^HF|4O-zo$N
z4ig8ZZHQ#TM$7d!(pq*`vU@%f$GY15cVp8n1mSFCz}dE<`p3rlVt>wW7$0!i?SAea
zVn4>m<MaLiS?*bc4V4zUIh-?g^7>SkZF{YvV2y|fc==7pax*VGv(_lK<4cksBFj4D
zaId%i{bX53M>UJP7t+Xd%JIIAyFDw*wkIflexZEG<+zt?*+jQTq;~JVmgP+rf9bN^
z=j{r(|HiS++0_|$dbyhnAB!yeQiu${1hhTq)Bg`wzw6`RN-{RSRoLrZA4>2m#qY(U
zV<Z(ZFY~NXmbZ2}7Ngqr{6<k|hf>BLhpi_P<*p8*2BxtEUSlP$>&m(oZjJ|JhELOx
zf^Y_!WvU6YY!wgN9@W>6dO2{~J!5eoN=y;$8q(LbH*bi&(KbG~vB6!<uWgx2yBWP+
z>U$8Yzh&FF_I{8*Te*5Bo2m+KUUAyF-uPe#ar^9!^9qb|zSv$R1z%=VE(j5v_D+x{
zy;%ZRvM{P&01?~G^?nt_Pca#(t5Dl)m92k{j5V>BW7!|go*cSoR>_mszu`Jw#xH-o
z09Y^MCml&V(qY;=hiz%*0^~4QELGaHQtiT08keRVCW$Pg%}5-3+LPaX3s=kvtyS+b
z6shfN7-?Dw(&#w1dgMq(I60Bi49QE&5`ZabHgfh(If5|DL|F@urwg|?NB;32e#hT^
z@@o$Hz$cH3^D?EGKfiy^ySEQygF~Ju`3A&TYo#TlBn#ql&n#MPr8(pi_o%d1XeP`|
z9Mi->3}vO7MG({|Qpl3X5=JMR!LZrPwmD=G_0foiHq*$c3i4FTUE49DPUecMBNeKK
zFuNj?6)#=PU<+a=Y%S<osh9VZvr^iZR9!u~q&@QN*VUbsDSOLoYvK9_&!ii{UWTd7
z34!OEiF9emYC+%&BsZkAw`fMU^0jD9C8;^lfeJ*&rYxC69wn#Eddp<D;yRB(zt-<<
z<8~XxLVX7r`FI{5(MTzg=NYZlBZq9xlSlsiVXe(G3@a<zR;+r-uEO|^FSRQ?Z*%!F
z4WsCpRig=6OqKTnbHgo|6X=~WC-bk=1XNJ<CJe{vKq;$}Js%RJi5BcBrio5ae0qH3
z>&JKe`q!T_|Nd7@C#Y3<zpRjFWIFPY?|Ddx65Rz0k&P1BXC{?q)yu8$(qzS>xFR#N
z7)Kflg^E#QZYL3%NZiXcoS^$1?h3@Bdyu0=ck8k$opUFgg)qXOMpr$np|5QvnKEO1
zGS3*wB^8=fN^0CBIAq9pV+}7B$p|ynpNx1!yai`p3gFJm6%12litCyWH?&q*O7%t!
zd2-Gcd~fJRtsAm^BNGsp=@xuMhxsrFYP;jBK`5=}B{JGHoM5i*{~O0{?gn%*eBj;Z
zV!yEOza#uC#`lqBLs(j&+Ct$(X^S^bnB5V1PL}ax3TJkdUS-#4dLko1tJs}IpSc?4
zgWhG&SN`{vWlIyyVtpyjDT8r(eBkTH_kX=CQ>jFj6+*+g7SnV?YK0kLjn1KK%<Z)-
z`vg7fZ^gW6G<~qNRn!SyHtEi2##A~I4kAH#u3KrmIc32?Hg=xB1zAo}&Ztf4UF3GI
zdrW#f1=#@+yJxVC4n%w$vfP{L$ULK^YsUV^ZeM!_-O^T?u2fcPTWPI8@N7r%*|jYl
zvK@)&@p{GN%&LiVaG$xPcwSX&9o`N?b6`f85N7eNp}x1cHI>Cd)x}|MT5=B4tv<k?
zg*Ldqf8=qw@axY%WtPls71p(JbE&*r8%u4-+DO%y#mflW4z#OO=TOt^cCI?l>R~>T
zq>^L}dJY_KSFoh?%=`}P0wXJGfKmf!Q#bG8*$hrG=MW%W9cz1rJ%}_a8N^xs%v$UL
zjhx}tt6P$J=IY!kXN))N%2GkBAkxP1g{Y<wAx#qrp_Bz}#+;@I3L3|GVoHhBGG@Yj
zyrrq~s0&j{%v0D&0YaHN(A7b*a53<X;cfU4UaBvnJ87VT{!Vk-t!K7DyOQH;Bv1x&
zwq=&>SL!#7^gZ75GJXl;1;Bb4Ki!Dkc|;UcpV?O=4B|Gjk2+;ww=g>Yfo&hzz<AlV
zG~Z`^6ZH;Cu|<YL9@$zE8Q9Z^mr^OE;oT4us+-!{fA#CBt#P?5WJ%l|Z@nZ(grm9Q
zFJ{3J&`AorW2#rz-*FP#UEh9Q)Os!dNI(vDEgf6Bf7_rHK{oQ{F^$)EV&Qnn{yl;_
zzzWU_6bgDwykDio!{|rrGu2u}v~90S2AQ$cP%BQNsk+M$v+a58qtohUkOVoT--{wi
zIHZJRVa}P7D=I?Hi8Li7+3q_1e{T}d6=524IY!SmlWkoPGsKuBVl8#!o82+?Cdqdx
zP5ahgf3_ngC#B@%qV{AKWiv2pYhg@>9Lw_=-F?Z9^)vpvmUSZ>JF{qcV4K0+<w&+Q
zq-s%a<mIGUlCS5sW$)RvF>5Ifkhj*TwUH9cIWdbzj}IoSRax3XjZy>ayOeoWN{fBV
zBEsvA`<~UpyP0)IFiS#32Qa5mdzFwU*`I6^)ScUL$o-6_c<WYt<1Yl5EymW3GY%T1
z8-t%>mqyc?BGyRKD5i1WKFLC*XVk<MZTnM?YU8fTlss!x`d)1CYM#j{Q4_3Utc8du
zN6CQ*si||bXfILdJD-dQxmw>xqt$vSf_-K|U$;FIw=Xe&>*mm2ZEqj0(((T{xwapN
z%NyfcWO>6De4d*|29k8@bk9=OCx;yB=YyYR$ZOwGJxCzqyj?Hv_kkZI%PGp)L*|+}
zDE99n%NkwNtnU~8rD~)gbIwEOhU2)GWjb6ZyBT=8%p&7#WMS>xES-a2-EsTI+iGEP
zwQAY6ZQJ%bCoJ1owsqoKcAd0bC)>7L#<G6r_dL(Pa9-#0y>VUl`vNW-t|~cP#7jn8
zQ+Mfz-l9W`_{HlButi~==V;R*VgGeI*C)%D62kabKhDe{y59#-ujO-~m83*-5e%WI
zyEuk1yK)o42K5g&RraH~L+R@a2j{v3QxwKtQ}tP`Ynf7Enyn+HDMNksRryLFN&!(t
zbL-S_x9wMTFWS`Ktv+cVXqUB{cq{o}Sz=KY_{%BomW%vP7QqC_&V)Ag#?2xer~1`^
z#WN>7iw6$%c0V}s4|8q}Tij($wf9wOHb6aO+zv-;CQqvx_vsv?uh-N$%s_?{1@gU=
zKT66+%9RM=)vz%-1)=AhWWRxa#@}l`zmF|`KkPWh0`zJxyR)u75-mF0PAXS-O_`h{
zO_2=!%Y{9qNZd`oT8zboIRU=(Ni-o^$Nt7!PW&dFQaYDfoN9q)JieZZ%HTn(7a4<+
zxKW5~icyV^>LAl`TlOc8HB^n7;7_8Yr?22m0-9!yRrwr5)<w7+tN(7aY(Y0I^!^^5
z(B9<g@kB?FUe<=N0&Cu1$z`KrMV5}EWK910-{~HFU}^IS_gPqXakpRYFOVG-(!qee
ze54%QZ-HE09;=|d&o_!3%U`ZB*Qgf?Nb%*hp32rrxiF}*uT+nPAZP{aZaQF!cTs{-
zZ}QO{gdwcoczl|=LKG!O{iw$o_qm(3iFF?>C>vuLxj)OvO#t0|h7?^ago$~%ij+bh
z6NX*j%NcfRhkc-Mz@XthW?i0)%OP-?fK}3y_ocIPJi7z07(#6>`qVjBwbAQ>-DfuL
zr^>@m`-cL80Qp?%h98xSXfSqh9p%n+xa!ri*PI@c%)SG85oS9!Riid08mZ$Er$)`T
z&FH$i!8mBmik%aS>UuU3z>g`B!yJ53H-%+*-22=A%8{f})l<RTN%&X;g&&nNDxBzX
zJLyAbo%1vAK2ubBr8Lj9SMrpD+Y7Alse&v9^wYMEh84@m3VZO>7o)b~NYq@nsW?%3
z!2+o|E;8A_ENvreyv#Ye5CSnk7UcFtKcXS|Uqe^w$apv=(nBM$3&9PLUG*|*G*8Em
z*0zq`T{sk$SI$%^av`hM<jpZw9<rfb0k4@^);k-Y?{WW#Jo}Lp580Ec*?r(zp%q?U
ziPE`^i7lqOiwzbaqx)%3xlK*VN4QT)FK?UZh-~SZmwMN8iWXpM(CxE~Bg`-7ubt)-
zXeomAqqg$%GAUlND2}JC&x%7eq1K5Ir?_d=r!4JbQXIvOQQbun@CQzu9PZ8<Mgn-3
z-#!viH~n(~EPBe|zx%AhbEZZF2#nY9SW)U*5_(O>qVwb5kol&z7wt6R{f=R2Pny(D
zW8QL`X4!=*hJRBS{@L<VdPg0)hUxL7Zae-V$aK{jMAbU_?lkOzNl8;;Cu=vfm#a=8
zrC(?{P<FBhgIjJ4tGg?4wa%Ev@@st2Ua3i-2{y%!P@<Y!b#WFS{1<4e)tB{>5>~ca
z%sDAuStJ%_or_?FRTiIHw45Rl_^Skbec*hsd4A#LV(dkmQ^hc0`}x=oQ114+2n%^j
z`|2~t7@6%GKv&#*K18>z8fW<>){T1g)ed70hx_=cNF_H~A-%?^?E9ELMhcuxtz;Nq
z27iFgz|VGPy{SDfC-f4Tjdq`{*p1pHMH>bSzfW)gYLo<xz7vY#l#Gn;ZJc0AG`cJA
zOjf)08sgfFeqKbvFKHVcxb8o<&TUvfy8mER!7A+I+}~=6zCM9DW5oabO_}3+>tA!+
z+o|fwrD$*#hshzoZDEB}vq>MzF0RP`=H}d)&DoecM1K4Mn>(hev`#t;i|pQfNdlXX
zr%Bbi%f^ZlY|FA2>sYcu80+OoYopjw-_E?QUGKwPwyA3b4KMiv4%Ezzvu+F%silh}
zy<B-m`;<EKt7#S*FXtZ<HSAvM=t@RIwFu~qp=o8*vK<?1LurE7fN=#h@95<9W;Kyq
z6g32Zvj-0(fBX-Hyu4?NiS4~pjrifok*FpDW(RCu2!tyO(v9-frjP;;bN!uLGQ7ti
z?YF{QilDrGnwr*+Dn&Y{42zMOUoru0SIbHy(-mWwY;H0llUdXnRLKwPX5{-Z+z5rt
zk+N#xGhaFRDfAEFuY4Q1?WufZdUBna+87(yUK+(7S~N-)^%$LR@!s#U>CO-ra@7yb
zBmarRt&bt`FsT-*!K|od+U8boeg?^RlB@2@hYgOl@f~f&gc4q$_KgoO$!IurPTKz>
zqCTL^sHPpz*Vb2-%>@;Myy%&6nNa%nit>M`&Wj!N`T`ar4KWYFj`})pmzTNEE1vb*
z6Gf^GYQG`>63hdv%>omzP)0ilwk>|l$|uJRhuoY3)zNNv_G{Y=hMcI&NLy3i^gOYN
z9?p8lfC1kIRTeTX7#=!F*<oP+qEqgh>EF|y#Uud5Z+`ZQl{E+T39oa?d;-V141Byq
z3y`Y?4;b#5Eo6E2-ShC&JnT7c{b7#456BMT>$STWGTXu9w$O!v!Gw{nt?y=Lm=9W`
zkUOv8#rU4cW}0);U-wNzBH8%rYivvVlXbeB<1HsrrDL<0ujFnPK+lne(k!8);G!r;
zB8|tCo{PX}<jzBwP~h+KfWT|lDr|O;v(=@e>ra4)VH!h2(=a>P==BBlXpU{hPj)Z4
z!eD)40Z=004YdkDuN}Y#f}I(sj0kJdH;{Ah3Kyh~V7#767pf3@%H8gU44Er@<O|Pz
z%Lli6_<vAOfq8WXJZ6QVNKr1L?^_D<Pg=&Zn<63`c&pXdAUBRk;ab>?k=2u97MZ@@
z(+tMe7^CoZhue48|EyCkaV;7mJMFX19O5%;RMTon5&t4~Er~|lLP;D_)fh-;ms`Lt
z3_jE39m7k8O)f-C!d7m)oT9Jbom%7hmGW^Z)_JM}#m>auvO!x8etBjn#-5Fi`}Z!s
zzcfCnHh!Sjo|W5I{S;yVF(^!+=!?P762llbYHOAH#pSXM1{J_g8EX~}s`%Pp@G{G{
zB5HfHwvYf3Bn8;fMR6{!Gi108m-K(U+}&X%(I1BU@RKZ`_-@tjGITOtbhON8d-UhO
z%}@SyQFMJcQ}mL);1{sv|GV8_cV;Gpz|9ZJG^&z;;}&=>v&A=X?Qod69(#t4^=|(i
z>j{2*7%6pYO=DWI>AYIqgLh)`RC-SVsj|y!wXLN6hH)qaea1=r%3q}+`SUppN8U>3
z#*~%YvA+PGkLp~7#zOB;*SEoxA;N4D@|pAcZ|r^j?HwK*)E4$Ps29gV>H829<@<J)
z)Bpnl{Y;reu+8sY&O~slniZrFRDQm8Hv`FOj$;V#NM!l7F|Otai`5(HBvrc?!YiZ>
z@briy<Pk>14kuf))N<UYh-6}1BRA+(%1?nqb+i=?`CiJ|hrgVXedQl!_q%~(E|iZX
zgGOuu_nw>oJpp<Jl6*GZb2c-_SC`QoQ=<(^x_<5r8S0axi2@Q+qeE0Yhb?TFR;os|
zlT7ha9ThpvHp-pTjfwPx*D(-LTT@-mWL21An(H#T?F71mOFn13!FWjTx@YFS)($P*
zw78k!!)WsiC0TN$G=<IP_pbdd5a<BsHD#osBgSdm({PiD-A&Tkw&#lRk<c7u)253f
zmo$%j09jrOx>#}O3+~pHC+@Zw@ZOyN+i;wMVv|dpx6hmpOwPLw>gX?QQmP_UA)b?j
zFi?y1oryFN)z~XB5nN3JWy79nyyc4&y<aY?Rs%PSg8Z_8r92JmT}K1OLr`pKZOG}x
z4T3My{ZX;K%f(w@mrS=hL_L8c%hn|lONhrlBi@k$Nv=QVo_l&V=oxlu$a8(Kv=N&s
zn8-1_*^>ElSBGDJO<Xf2-)sZ>IE1Wa)&RR9ilDQY7hUBeC}HF4uaZ+Yx)^4A%$qgr
zpLMB49`~da%EFZdLxPB5%b<>JLjUI@pTj|}++^V9=<db0v=oa_g$lw1Vq5f_Hntd7
zy|nerpx9A4V2P)>6?$RJiCLk<yU!39mG!A69f(lM-@bTAbHF^Fp@^n?g$mBb<Wt!j
zp#2gfXeA1d^ps`<BL}s}#YjrGSi$J2jqu&6vJ?R_{X-ZfHtE(r;KCt5-iiUs!rXdj
z=MC!Qtg@9lJuPcbKKSYG@RXS!`cY)`mUxZ<Wb!8wVMwc+O_<Fz>2n06q35IFk#vVP
zFbpb%5|iGaYkOn#+q3@l$io%wf=XFQsMVFn<aD!~=om`$J?EsqT*Vx33FTZ{<YoT+
zkzOnG@Nr|hCSllg`9o!jUq_VbbRSF#B`8~vHd-uBy+v4yj=FZ0p)ZFlQHS`3zAAmO
zUzs$2#{B6k`f>6#HcXl%Zz0GF^Sb9xivb^wgmya7S>erFBpS&l(K!skZh<;zOfmb+
zwEz9FhDC6I*K=-|a|UKj32aw~*ZA}E{gHmcu-sGSYG0FIy`!pkLG_*jI`Z<O*a_xI
z7*PNTQT81k4B_SGqv-15In@)Dt5>rbNoPz~&VO!9^6X0;bh#qb2}GSYB8KP(WH~u-
zj6g{a4R9HWy~cDMg{a)5)OXW~sZ>OJW~UxTz_abJa~J~|@Qx8Oq^0L|H)sN@<`A}g
zN<Y64{)Zw0f+S@NSvok7dj<1rwO*+n>*(5uBGkwhnr5j#fynUPuJFkxs=?Cq)6Kwa
zMeM3tIX5d=GRURSg2utJ{cHSJAGp5y+DvVcTuPpl@Q&&6qln`jrB0?>XV&PgQcunH
zboNhbKiJL4*%YX=EKQK4x%S_{<l0lWWBYHp`_S0laDoY&F#Tbi>6amTm;SuV>YlJh
z?raC0b7jqA75SDd^q0e~cN%@Jx7!mRV%COgUnielnj4$9(cU~=12bXi2G2n04h3@-
zIR%Ov`w_lY+U<d8dzV7b=*?`3L#6POtAc+3lcvSm#mN8wBtCi~H;TZCu0>;?w_J4`
z!{`g9hCFe<qO>-GC4rgdb>D<6${W}|0<|GCqoU94O#PdZQJVfu<-iZf%+{;<wrc3@
zfC4CKtg^~%Mf!0~`V+baPTb8`a53Y8mE0e{SNw7_E$jZ+e(TRaqe}DP-YQqJlpLf!
zzCKZ9b7CWPBY-8FA^Kvmimu!8L2&WP7x4Rf+UvA_;HU3mp(eve*LIygC=>jvDmItS
zlF~YmY~78R!D2#42#;IKn0`j1bZ{`SPh1D*$j59NqzL=(62|J}W$#QBYkQFaPp8P5
z^e;~P^Xp7qUC-E-SlS%KBNW&O<2=H#C1JK1J2;|&E%g=)O<>mBO@bbgH1%)eS;6<?
zNe|gvzT;U`Iw}vl-zG<{Q{?o*1T(xJmsh?HN799ZN)z{NJcqNT@YE@1Q7jp=TPB3w
z6cYE(=-%DH+0KA(w{w@mEuxZlof>I%<8x504O|nD!iaF&od9$QOp;{*zi<$h{fx{9
zPdgPV0xs{kcBy<;Qc~(~!UufncdK1DP7ht3(|14U`;!ia84cH=2KJJ;Xfl*K8si5%
zj2aZHq`o`)?z0E8$`wbc%j5MFR;Pn0DQON)Kjh$G<ApyC(Rg5WjW#nLzxDL)4iEn=
z>z&w~!sb_&pETtR@=$;_<DY(UgWO-LkabcNyx$U+aZ*-ijcsYxJy%|Y%$wk8xju@{
zoT)9J-j^0$K4f1^GM&@o$!f&iB&wKLqFt~UmZl{3AKmZ&tSYsQ2y~f1o8gL`5x^?f
zj>7MH{O4kP7#ZMy&xH=$@5J`!3V?|CUkkkbyJK?jDg;@|>H{9WfYyZLi8&`8Yk`%S
z(s+dks}0uei?2y3ZBh*g`Qpl5YH)cYPP~C_mQTBkH|bY(asL_m0TN$jrw6!W_BjJ1
ziIF~^M=T`J+0j+Eb(EdtClbQJc9Z+Rh5e{br1$Hn8otZtwWq4XH%x;lir<`06JY6U
zU`Cb7Xlc+gIB~+;FrgsQynP8V&$OOtZ*&~7y1)PkfQPHUR$`5SU3`{PYWQ>{o$=?l
z*c{Q|^3rkOrO2)EJstr3PhxplkxJ$saY=9Ak7DdyyuBp8;i{c4j%KGFFY7jHGoOqA
zjr5{a<)!B^Y}(YtxL+wgn7S2_6lHGN6b0+MOuQW5mZUUZ9jt@&;ARIZ1aBUF2fMg`
z(LcJ2CtCT=un3<ipTgG?%+rxeH~G}WX5BG%?d{#Poy0}Ac`Gks%-7HQ-CkXwpu8(e
z(-1WK6|x}FjYiC6-MOt;4N2G9+d|vaSvUg3^?>t$jbXPXd9>dv_WT}$H}cpNWiuG%
zkiY7G5=u1&uIiEMx1x|Pmm+X3>u?TEDc?KVGbjn;WH<@92tws6<TLTpP`sPvv7}pl
zd7kc7yzhUd%I-EN-ecsb`s}HFS8j5nJdN3BI9Oi?3^c!Svta#AJUszxN1hx(Xjs`L
zzv^mUX)hn27?1b-XktXf6Oi$M>tGyGNTB#!xeY7{j5B{%My!eBY0pJhgm3=aT@tQS
zs6ZuGaC6sk*0Fx-mSlX2{tweC`gM-L$yV6`;8^U4(LiuZVu4_URN?!e9)Q2DO005d
z{soYmzOQ_+I*J?#G*SX3#5VKf_C(Y|;{`hPWoGClI-VXr4T0;Hu`YLkkHK@RN-g#`
zJ<X%n+%;`uNhDOG`Q|bpt#`m)W&6<eq_b?zu9885c2Yup0KzZHusW#-GO^DXmWTJx
zp3*evssINQk;GjFVL@^Ku=YQ`+|c9d3limQG?`3qK7*iI67``h*g?NozQ3NAYDF7B
z%mYEE-aZ;VUQXS4-9<=L{gNFn(TUYdn=xW})IOgmkHg0lrt`LF^K1?(gvlB=6OlDy
z-8i3SCRst-b@aTQsS_bnf%1<_XUPYquIdzAQj~0m)|cjPdg~(cUf22v!<w)e)K)r*
zUT*FCvyVfTJk5w3a!<^gm&@B+gPh5}_YobVJFRUo;*;RDKA-z1UECa<rv3sXk)qXm
zmBs1Ws%MhCrTNKNGju=RS*ZHV_rEC`E-@6_ZOOAmXCfh~<futYEH;{Jj6G73W*$eH
z)QZE?J*s<$QYfgwGWTkz$Zk>0(FZxW8?XM$yftMSyh%quAtAnd3s4ww)>pD|%F%H^
zQ4Jdq)VA0>rbxM!yvtyZ-u#j?Vfa-W#8(p<J__koZ%L6KNp~Xq5&Jt-iUAiIrM@^X
z6l`}=V&#@J`gdViQK<mX^8qdqi84NmgzKpLYfkYGfUNC461#c?(a)n58fx7c)je22
zO7LNj?U+kSwrE47DF4*Sx=Fw_U*3k_L>@_(^hqcZSu|VG6$V6%)!9>P9AB3FaF38=
z;admJ=m`BzQE}SC>0Macez%zA+u!8PYn8lLq4(=_5!K@nKglR>6a}`ib}h$v7$7J}
zlVA!}_FoJy&v!)|3}gowag+F4gkxEs@81F_-NEsnsiT@cE(pxgCW~eohBgEx*ZoEu
zQ6%1HBP#sI)teeySL7G2*20mN>^U$<>J$&3t}`c5XPd5&Z`zRNsY~#F8z0bP900j^
zL`@2z1KJl0Wa3rfWCj+-mYC;SCIP6&r(zawP+hoZ#Qu=Bt8csN>!m^TNJ_P*j!8w?
zdo|1SbF3N#Y=ra)t~I0*HTiaY30%-9x|+7R?^=$iME$&zTo2t0c*QFzZf6YT#JHIZ
z<jk5YZtPe5PFC9ia$%HG73%uC4PUh@YI8wJKdCSWVmv-_C5kEAqy<$VH&jO;4xakz
z@*SA1<v`jzE?SIYS9?+O*4ua9|K+`*`{oL7FaC9TyM9RPi{vBHMfp)d$~QvT_Pi~K
z?{c_`I6+z;))dtg1r<Q_pp%5p6x|(j8$5RBdqN$J{mBIdM@C-aeoWGMwlA(2!xu+(
z>h}l^Qlt>?u!Wym(g*XVvCEInB=wBVg~9bdj&}bHbV;W@tF+n0J2}SMZeLK0X5sDf
z)4$5<DSI}>P-V7sT#}E9+)RHoogY`uoy+J;ShnkF)vv^Jh*A&yYFP2gF%-`4g?wwr
zZAGH8&+FNmhY~<s#aT($4_7ge?r`z=UD%m!L<vg8paZWuoT5qTF^38QuTO!&bSi`n
z_M)*ZfU%eqOMAJsBt9*3Yk8*7S#m*Uo6A$SKq?~Aw6ITvAWaKSwja(iyt{ZN#P@6X
zg8%&D9t<6lriwgGl=*&@Fh5fyH8*M;k1``Lg`1RHhCJqLXle;KHfrvl402$_Rw7@h
zFvO>iw0deEZf%YtjotjxAiw*tSvu=9z(A!xCcZ-rw7(g1;=erh9DKjLe;-;zw|wXo
zFkYqE1AYg?Y6$<r1X*uK7o|)sG;R6GQ)rBT{B|QN+hv)ji0bI)YJ`m^^8Ssk)7af8
z-&fN4h5fDke=YTCU<q-ZfFjn7;B+S1{$iw&k4)T`{@9ulgK4dQ6V;s`s!dBMjd9;Y
z^pF3-mnBNgUCPb_jrdsAz=<VaL(b2mY*I4h60j%3VOyw@TGes_b!33-yWf@BJ2iKw
zf($5PW{C0ZwS@#??b|N0XZ=lZCM+;U?qv_OKuNsc{=~Dz?#6hn&@JbVtO<n~A&I^F
zTwpXBX(6;cnMa~&5IIJCwVI>YN>)_NkC2L?^Lp%V@TtwAvL+d4yRu3rI5cqCJ(Bun
z6h;0y9eu5sg6HoL+xQ--ZF-K=bnK_f$+FFx4F`M%t3X}zo!#Ko_6WZ}#Dj*>hIkxV
z9oPB#ZyB1W?h4%J@2pCMXjod~eBY!Sk{z9X8x|&`U6d4=7aC6XXuS6<sv(%iHa)NU
zOZyCF8UJ3=<TJvVxItA=J6L{X^1hQTV~9wi=7Es|O=Qt0Dm!OnD>liIgMJ-vS=*3G
zg#m45I22Q>Fg!innuZf<Ec<_N$h+HI@sD*Fh_(s28g#0lSe!(+-I%x{trIB&?xX7R
z#IV*PW7%sddp-rJ8l-&uRR{nKGsj~}1t(g67jpR|Mu~tv)1*Lk)0nJ5eK?jvW6yiJ
ze8ts#7FMPN7**UULhHP^t3;k8$7_YVxi37+<!_djRIG+N_5#?@$p$w8j3S@1+IDJb
zE)vg<NX9*t(+;!w2&v@(@*Xr*li%O*D=caLr~n_M6vEGfNeUS6`8-?I-+t(A;F0@H
z3@B{sq}8Z{nq4xM(9?i&I3VS17#Y$-skDGjbISDks(ddf0waC5DS218+A^tPb$yi0
zmv->gMP>ibNVv<zVttOuRnlIOx6ZVBYVaPd$~a}+R$kD~S=X16Vu++HUofq@Uo2s9
zi2jbKJ29urvp~rym1mWd-YQx<*q%s+InB=5mMnuPMuipJ;ToVJXcP@Wuu1M<y!WsZ
zD9l?NILxCqKNp6l;7HmUaOuiV83F3C3h?UK-uaOq(vosdQe*vIYFR~j_Q7c0HXfI;
zcd+Lpjt5{_jII?6*zXN7av&WJt)183APhLH(<H2Kd+7Hb8V?AL>RS2K#Gv+h+l#v&
zJ{C{M&*j&~%Osf$m~i~Bo_3}K@feX`o@P}ZPiGfjR!O?{8kpy}ooyq2RZ_er>dnXQ
zcO=@*6iABCqQ)E<tc{!P%#V-lH+YY3j7y`k((2=qaaJh9+tga^kMF|HVCGHciDnHe
z7B-N}{Yl}7VA~(tcHBC@OY2PnE%p)iMR&dKbxXxI+lZkmCtvXs{Sl(S-|t&VkuTES
zU5iRLD5B{`SZ9kx<#V@GhyosFLt+}a<}mptiM1B7_Xc7Y{w!AQS!R&B7HwZ9kSf|v
z>F|)(*gV!3l4@S;CxO_3%r3u`L~-nAVhW>@hLNFSndP4jsB%rsLUv}|4~e*20RI}#
z38RFXbXApf;D^Y7akD?$7Hj0jd2?|6{a&))J&|wNUHqH*tt$b=oYRVB!M}P6BuEQ$
zZ|7HGY{I8e>W|k8<_=XLpd+=1uWNAMpTWTn`VY^a+5)OcHow)=63b+x_&+Zn`ra)=
zXWp*YM4;X}ULPtG%b(lS)ZX+H;mGb(zLc@eMacJt^)(`Njv5EFxuzNYf~t)Vl?L)S
zhKjI#R~o9CULL|X)FcnrSCGRRlba{UXwXch)4Qr@%BoJij86%{p?&@5V-QlF-!&3F
zgb4KfLzhkdRyn<uTfZU!YaIX3-HeNNXh_(s4K-r^_YSy*PXe<+Dmh%27ALHDidCuf
zX<(6qDLB<mpy{7cqd-rGMg~ePcj(JuPV_?2X1!djmg)E#pm_JEiQhHeFutA8^?Z83
z!TPF)F8x?DgpCANSJxc<)>M*1cM+3kQz6Umy=*3<%87NG<n8?8p=8p(ze2~Oa290J
zAv8_Awf*`@{d+EYA$PJBB16W^$}y;92S`Of5iQ>JzLf5bOM#kX<#3reG#%NPoF)KO
zQ2=OycqBtt6F>3dh6aI*A`GK>0X-#o9S86CH33_`ou07zP)~aV0RBUqZMG!bgfrFX
z6;LI!<{L>F<4rc>vker|aFPserUYH`cI7dy!ux*Nzh@r3h6OH{dXs&NEd8v61vKi7
z4oKY3+ogRC{}T%Lay&cv@c&PM<^2v}A{{uSQR9wrdwRR~sjzqBO_b(-QzG51_LMKj
zP`{0QY6gZ#rk4~s$I9^ovCZzAv)M$=pK%0db!seq_=<sja6*=I$`~iq<hN__3HvZf
zU{BE|fnWo>oQ@CaCM9E4a(?8mb8)6dFsGV>CuwWI9t@{p++oVP`(GOe-7)7znKJJ7
zX3G+|WEvmIcZC<~928+%$uOpKs!2Lt3yN5z{?c*2XIE$#OWe)RuJ?HdG2|2Po!8xw
z){W#65U0}EARdM~TCFUWk`BVI53Z`)#u{f-&h~}KbI|+wtjU$i`?iQ<L*M)U``f#z
z5qnq8`%e?1uC4l26E-)$_}~P}YK00a@YP*g0bTC*)e3_kiOHB=P6VDV5;1{`-+F)?
z|8!ClbOO?5lgUAN$*!0#`~NMvePFpu`qbE6GG!&HliYPiC_WX3X(K0WgLd5xS<#Lq
zL+!_IU5}{0P|4?(1>2eil)ziz7!oH*%iHM?M{8@_X}U?NFOaq?ird(89b`|D4mB+G
z^lQW>9CdQ@N=t)so$jBo>n8+_H~22nzvx*`s-nkfF1-opeBojvxV=i^pTb_aVAlVg
zNYB*7F=>knvn*1kT09XyHCf-$OOXBt(-n<TY)<TR4{ep1zAlvS`MuVAgnhXUn_L-2
z<hOC6dUo>Wg&?pvg*<gd0sl$<j}lfMVVv*Pa!ErGwB7F7re43!S{)mIiAsr6bS7Q(
z-4~4&h*>*CL%F+-CZ<bTpUhd)^%;hgCNB$LRhi#!SiZj)<-NDWn#J?oFOx~R3bn#5
zPki{sz*4#Jb95#7u<m`Ce{qCsW3eE*>eD6T^Wg!)*;$QGwA*((Y`e-8-EwtKvx9!s
zAn8JNRU2i%ihN+IH2}KE`C${qU8H?DB*(|5gf-IlyW#@tn|-xVJ-55_T240;{$H1h
z@%k?vtKX&_%?=HhE&iQ<bKo|A`OJyjm#|JyyG?4(dW#0;_OO5PqEo2SHR2W7xN+oj
zp+}|<Q?xBwPh-WN?1WV%8_X`MJG`q@+fCr`@Y<Q*9y;6J@|Y22h9pwVo}ms1{Yb!J
zk&?|xm}GIb)p(0}dX+t;usaRFIhi1eQ1X5DX)djcVU8ye<PLyUu}C{<A|=`O)J_xP
zDD%k_INP55gEZW9i<o5`NE1)Xz0oNx<F(Bqmgcc5GkOh=<f+Hn1T?tOwDWCayl;*%
z((%t`oIHTLtmX-3km;U@W_-6DRQ`F8(lQL910KrPn+?Mq=B#S0B&83Xqg@#}qkNe2
zjV5$h3MKmzGB&(Tbg_1bje?3$s4@yUDR42njmtyZ<K0XJ)m1Wh87E{9<Lp~cM-DmG
zHenadOIIDQd3ZVhWiLHgGk8MngxgKZY1W_^47jY+L$|R{kZrD%N%J3}m>nz0UtJU^
z@X9WI2Y*=f3pf?=phic?elmdOx6-fRr>kcd`(pwyWmyx+OHU_@zF;GN=DRq3;?-ag
zXjcL)J_csMLK?pfvqM2!181($JU@=a2F5E>>tua}mPq+h?czBIiYJA8tJ00#TTX8H
z%W*r;^!!-Mo)=99Hh2DesKMP@wKkSf0%Ae)AOz(H&MQY98DwA<MoCQ(5*_C*gWv@@
zQYsnZ*CwST(+Y|lnr|t$1<@pFNE44n)U3-Buzn1fGpsd2R_G6-O;mDY>8R3n61#hy
zimkuvr6QS6tk1wMK>urV@)Z*wrV7>Tud_B^aTEqjI{gc)k%79HlXg7<y^SaE>+I#M
z5YrPZGnC(bwLYkasq*ejM{0F4Aw{(j;QL3yIA<C!t_;jl>2l|hF}`H3#9$P9v+PmG
z{`JtRheNH4l+M0lv66UE0;Nl~y;{3K%Kb+N*|=(VC!)rAr*~yl2Vb^;k&6rAZIQQG
z5$E^2pYM+WPc^>mej_y^0pr^TXsLZl-s3;T?c(hV+n^U=hJv|_>?k{D#sZ|Kt`J`i
z8aD342o$55!3)hy_J;dG5%E7L8P|Okk707`P30~m&2RTGm2*U!Xp~__9BE~>z#!R4
z-lOgzWoefnJ}Mj+|A@ETH?t4L*%>oq`V{<7r5U#(+h*5H%Q#)pM#jVQi@m)?CbqUw
zejsC{1izy!$0J6O?84NF?nJDDouRzcc!u-N!2x_3<n@_Dp`vp8pYR2So%89<v{l)(
zWJba3p?=ThxrBfGG@1Qk(U>>Q_$yuO(qQ(=(DhQ5*IV&)W;W3u7)mVXV8i^GQn&r}
z$DgyiJ{vFl@hP|BWeij*x)a~zsh!3^+Ock0lA=;mr;~!i2QltFUViP{JdbH6>+9El
zufl2+@|mMLf>necV>VyjNHUn;qbkLPRVbsC(^!AM<F^Z8KBq;%)nuBrdbjKF2ViCO
zFiOj#Ej1h~Yq4l83Am8N7+(Ni`6Ow|g?29MVl4rD9(vm5H)OxbByM7;3f$bL=cf_C
zW6GWPrm*cp&gnpoLw8rBKJ0FhS|;eZ78Hj(_G%_+q4U%OW+c3pw2@|lA2ww2+24jf
zXG58_xClo<!iPOinY3X(pvO<{{;Wf*Tg7jGKl5x{8abaz9{X1hggd`%16vH0yowj7
zTD{~6MVy#ZYlunyWW~4VI#?^!rhQ=mIIViqr4G&xbFdoh&yQOwdWl<LxaGF!BcFHy
znMPJ(IDy5kzl1*b$IJvL@G?R3x9)Ct?o|z>s0zLa-g5C=e3SrRg+?N90}+impYTe}
z;Hu3(Rv|3s`GqlS5YV0ss%ur!X$v(}6+`bg8Vx-?Jx$r9>@5!4<so7-Fvm95%5%iP
z*1=^34vwrCK(Jn_3h05HXsXe%Ox^~f)VY5TUEhQUC;p(SU;~*D=u&?AMvd>=jx>q0
z4pX_ISuly~*_&6_4t3m3F41L^uyI2RMd%ywW#)g^lVl&0Fb8+X*017pn)6hDY&&=H
zT2UslZx}JaT6e5U`EG<&6Pr+c#<+-%k2BGbOm|TRVE=AJ^tuMz0U#lk39jU-;?q^y
zblO(>q~nO76%?kCR?W?Jt#*r-`5p!I$cTKr!y8BoH?xhI9zmm~t@F;<&AzyV-iW+)
z>a^^EoGZp^iK|0^%}T-#-Rq-jZV$NYy|@C-Pt#}!U>C0JDvH|bH?+w>g^U<tiq(4L
z(ETC>`)+TD1B;yN)X&rDMQsLjhYcrv=mFCoQtTvrCNFDAE-E?zQv>PnnMaA)UOao3
z=L0!XriZ}Gc*Ww&!5vr0Gr~l?B`H~^9CxyfGr8x@gI6r=E@xml>eNYMBPxgM?1*ao
z0uqH3e@RU7mW#N14)cES<ND;<9ia$!#lxf7apLu^po@pgpk<A4(LCgN5W!;rY+8`x
z$15ro=7NE0nHzt1?oUjaFNqiIIXF%UN=Xr<8*crw57(!0J>+5y#>J)|nfZ)<h<vyU
z;m0lB$RAEwpf4E}^2*g+ft;*PCQ7iNvg%i)6f@f7H^^6|&cysCMzyn;R@dha_t5<<
zfFPF32x|^_a6{Hpzh3O8u^ufwg4xne3b6~UkO`vVwQ%fZle<JLHgh#6=-qNuRL`^V
zK-5s4iUC)??3=ugfph-^2=alz+!-Cd;XBT&V^&Q3cx43{$%ue$UWB>;ml;c&)Jfy@
z>*v+UKchT~0QnBpti-eKlPr~o^&0=)-YlMMeQ|vW1u;uxp#*Viznb5{v$C!TsP;K~
zJjDm3eT?Uvi`p5C97<p2JO}bUM+zve-W%aO`Y;}3rJnips@?WqA3sRRb)s;q&WUE(
zo$v>iFzEU3NS@OD@>c_oB}1-A$VS;e(k2RVvai}AGxppk?W;kod0A55boHuB<sFmw
z-?fyPY5t=x9qcFb@F1MqpL_b<BQ;h+5mTJLZ~UEm3sE(0k)fR1;Hzxq4TeL8woLSW
z&rEZoiw6<3n=vzf=Na+nVz`dBqZaKA!R&{fJC21BwA$NA1;EEN^}3tTB6_~A9cgL{
zL4}euD8ovMi?=pq<(ED|*K^UST@w^^YsbD<i`p%vz7%Fuf@vgJv>C8;l5~VK$twjX
z%H)4%fe=PjB$ErE+4EOkT^3yJ_r8~M@=gZss{}k?DIp&^TTCAl`l<glqW)whCqW~F
zl-JmUlr;!XzE(KKkn`|bVseSz^+p7etV+$MMuRF+4kqT##f`}I-Qy`<h2#s!d465e
zeRf@)@r_F&;-%gN>GmQtzZ!AG?yAJ3C;Ny}@gDdew#w`b*m})gOnqY@FYNxO|GURj
zS6*9Ncl2=?U0-MDd^%=0zaW`f$Z2B|{vKv9DGJAiAW8bq)wjBQPuwzY`a?eD4on(V
zC8y|6_*h66wbD05Ff1I0HoX)>6*i-#q#*yYBAw06`Ok_&ul04Lc@5<kc}+eOX2T!5
zhM#TJek8y~BBtMj&f_c48QbIf_qa$@A5*;^H?-89RlT$$M<6Of!L&h>hsBalhPOC>
z(M-Kcgyv@3nLKl|aTw>U_E>4ysg2+5&G=>Ve=<Cu5*jyZ5qJ6znhV{$sb?fv_I#Sp
zfQD!X<G^TF>}B`<y5Eo%!s2Twx@gCxro{Glxku{i^y~lo*7vymF%4rBkz>}%;S~Fx
zDLdE}MXC)O=?*MOFwwr5?m^>PWaIS4Il=+O?x4puK7OrDs_=LvmzLZ?zv){)L(TJj
zqI2kcLB5Yi|00J@!xMH}9Y?n?=hCe;_L@3N&?SIK^!G&E^$%!BqhApO7O|sXllz{e
zD|6*V(RvT9dKcp(-n5E?Z|z^_A&)OGHst9bx=rT{<JfbN3U2u-Mqq5xj(izDq5Cmu
zS%f9=PXd?1gpTz;*XM29D?lzn!{Mx6!R*iGT`WdQJK7YO(;E?%KV+@6V`AL&iUm6S
zV9!9<ArxP~sE4X+cfK@>XwVtDesGD4;`H*066lbQwa+OJ(v?sgrb7rfiaW81z%L#t
z2`BKsf8wyeNsNqCCuqbUMr-HqJZR2)&+N*!_w*T_qSt2N{*<iG_!S|;{3nK}7;lGA
z?4&@2mo=CHqp0x!CQzx<0qQ$70!ajD>bT#`zm47vy!i`5u3{Ji9uHX7@`zOh2^Nk$
zD<;*D`TUb`{M_FC<lsFU(}PSx7|}KIVaY^{)2OIz0JrCzl*TX2dK+;af0fiI-Vyce
zH93dQ1WLwI+D|Y3f3rqW(swqV*w#j1+~MD!#e<A~m^|tPIJ_0w7`s26;A_#Ndse^B
zX<QZyUGEd58Js;03@Nl{r$(FYyK>xWr)=ifZGncBGPL^%UOiZwm9a)6FnS>Q8xt?3
z+=5MQw$IOzi-)Z~ycx@zSLOC^&b3@-iqA{WN^=lU#TVG@C!6j_OE|1<NIjq<;`297
zYBWO;28I;u^XZ9tDOm6%S=ND0$5Tr7zP#@H*$2Yn0orseuM9s@-1b=lUXbbK<xoNW
zt1Yt_S1lhALS6q*j9-Yi*ISMlX@>}8J*>sOoF`8{KC}2oL*dQxBcNX2t=lt(G=1Kx
zPzg-?VnRF%rw*^xJSwDO(1hM1$?wpMrqDJ&&oBc(np@8IRbLcXAP<XA%NtDY$_KqJ
z(a~q1@8Ex5I?q9*ub|`26Lw=y9+YSRg=n08iMN^5W^=eDFrnbD?*K|{hX*B1sSs0v
zuoQd!|Bbt4wujm&b0o1hKX!XFSK6$jDSjK}Olxus@SZFju{BLjvfw4WC~dradlRw%
zS$A%qdxm%oL^b;vSKL_cOWZ~D>z7tcGY{l#zJW9hJlnH7Le8_oB(*WmQUAp*D}4E;
zC#e)#`0A!2G=bR=x$`ByU1P#1KcZ<%+>wGcngCa;-K@0)?w;>2Aq9W;K(TaL?CRE|
zw+%7d^X#HYft=j(sK$ZdRq63(?W}YL-ROq1tq;Rrh>oSfqG`hy(_0zT_6L#BwDOn9
z7mvL6w8a~ubimiSmM^EG44<q+-!C5>cvC<k#&4<R*jf}S9DCN6p`gCur)v&Y`VskL
zMXW2t;_ehoCV942^zTmD+f%1bp%wE_$oP60@d%v{FD{CqA{}16kIoiaMb-(qVL~Ms
za8GJYcmuoj_1_JMS36goF?nT&oFKDv!FivCCsNZa(ZVTCr0CM`LPUCa0;N)UPCzOw
zG{)o22^MIGvsYa8iiOr@Obqw&xN$Dqj<#dr6o^p#aU=Ia3hnnf%CB6X2Yh+Sxt&@9
zdS{MVTe#I1#~JW5iqRHf9I()Xc@ZYLQ|M&Aa(_rQ8@t8W=e84zS~*mtSe<XBIS~{*
zpsN2E!P&P$Ydu8AmtjP}^I0jBN+9wqYj(Nqn^MB{-WvWiaa;8#)g*5OZi?jo7H;Ph
z0+1Jy<A@`M80S^q`#H&@$b0X5oXO_P$Z!ZrC%j<W!S=tQbd%Rmlh<Po`+%Fghc3Uz
z_x(B{a1(y_@Io0QMp!-!r$xJE)4vgbhU3Sh*nh1AlQ_tBH9Fpodn2lEx}(vZc7hb6
z@ld0jzAJO|Js-DZzpnTf6;W?#yIV!lovS4@P1zT5fS1VWrb-&pC2~~{!BQDIwBvua
zSkBSlcy<EY_GFqzri|Zu9Ct&Y%|4+_CN=a|&hvZX(@yR3edU5$m#Nsz5oSXkk58~N
zKN>2*PWiBt5m4{`L)uBO9soYb$Bln7!5mwDJUQT`;I}K(TertVh}GumG)BHy95Ju*
zW%ruueb41BhIPY4!xGHanXuWtF0=GrUn8{A=HoI~E82o_<=8dUp0L~3)K1g>PV#hl
z$B!t8BkQ|3wD$W`ZDSTfV#s0TC$~;Y{&bCS<rj|U@#*^NLLD953tjv*U*5&6AEzEk
zf~G${H#+zINezK7ffR`g0}is5M}z0je8<{2g8C1#yVYTqiT9Lt3<DBWm3;m0>E0jD
z?_bVuHm{!ME!*1yWVp=v#|n2tHi9@EK3@HOLh2Is?=jiP{WFK-N+FcR$y*C4{^naZ
z?Pa4=@kP`eZbEMFbe$&a-$cz@_g!`yIDNNd^<dN!?>`&25<)sMzQmG9ZG!aC9U|_d
zvfry|b9{i+iQ@+trDjwh$&*K*%Ju{G&`(8O*2H2-j6MtX%#30X{f5Q;UpW*OhqjbZ
zH-2?z3N0S$oAH=TwTim50r1y@zumr<sQMM^^BQMzK~x2i>iXtPvCZ-K*(>bT_ra$X
zMwdPJH^lx++%El7su)6BW=%yK#-CULyERJ!)R@!cEGiMr;dO#hxI7i+R*7ayaI~r8
z8N04O26n*-mKpUSiJ5f=gyuQu*5oLY>s%$QDU9q@8imR>9MF9Zh}!>kx#r>Qu&dZ8
z(<3_ZoL1^{J$>Fr=l2|fSt)K|ge$j)290S8f8CUywDcQ~azK|M!5-y~Y_&TyrO6KO
zb25Lcul~K}Z?s_{Y_7b7dJ_yKHz4(wmSbVNTvSTC3>W)aiogSJ`Y;08Hoa&no@hAB
z^?~tO1!?D8=pH4jH?fJ)P=7rr;I&LPh0$skN*&h!V`r=_I4lWevXazk>ERYgHUajK
zVlWG+trPew;KY9LnrIPl{qe50KOKhk_ntK1fL}I_jghEpK!WfloGw(H9<+`AwI7#j
z?@WFq+r5InfH@%Zwvp!sVp2q7xtr)y!PbYGXWuMYX14C0?u&>foD?S*EYB;-W3I)j
z_`V%u`QJa*@U5U$X{OT<o*EuQdUu7I^FB8>^NeYH_9vwwIj@j-5@r((%n7GlkBs1h
zz?anvkos*vU{{@x&MhpO<L()r9<S<#*#uceZJ2UH=YXZph7YYZ9|#|2Ryc<G*u2O`
zjdI5qH&~?8YSKqdSVjjH8eIXrr$D~kz#<y3?e0zWhBth%i?e?PW68gK#u8qYNqhy&
z$>&3=(YKNdmQladm5gVTIL49`BliQ?bmwi?&f0}zye9(Gg#<HnMz7vqh)!|q=XyDW
zUH`n~eE7Iy;iVGKXgXdzb`sDV$Q!`*eCA9oIf%Gg%;Q<??ef+XoiQ_gjM8a%v(gOm
zRnlcf{#X_G6?FYgn2<*71$-(<IY}CtvCvlS?oJ~!`o^V|$2%)Ht=gl-U&Zfj@0jDc
zsOj}VQ)blu;<BLVh&$iN0&9bIFgrmw5^KJIm&6{fu23gQlC)xVHYtcu<c<Q_=Z%--
zac90{jo2PkLeb$aRvZE}H_MkdON|<fe8gbwspSw(`5Jk6NHyU%(d?yhGd^+T)3x|J
z-oG*!ZTqjpYF?lN*7nt5yj&bqVB`6(z=letVYBlxWu=gx{aIIfFYG&wgI*RlQ16qr
z*<2QhyT#DP%hHeHtFJDC>K-Y5xZ$u-b(NN`*oB`5q54U-Z#<=j{LV~FGIo&i{231i
z8t=nZ5&m+UO?{^jn^G*x+d1~&V>fQzd)N%VNvBZq_0Qhz?zZGQr~wj`lv0lukEG1u
z`{+3N9FB#{;91+maO!_1usiy+#%6r^J?M<-q5&p|no40ZeFRZPGP}7#!&yETgNp?h
zk^0#-O4r=_&kP4eR<n{$aQ~cu1{q-5N<xrqAw#RoZU5gxx0H^ybwMK+f=-aNGO^w?
zwhgM<ZQ9>&S`2hOf{$y|SEs0nAuD^57dMhj;(7#m${bz%gnvL6va_gM9l>iE4(t}G
z$nk41$%tKV|8!(?TD$}vgN_DjIWZ_AzgGo{jSM<oS4^h<ER2yw+c7Cv`1lQh8WSst
zor}+kql${X)z)gv^x`*a@>TEEy8YFp!`hw!J68N5@(FOeJ-0p7g7Ska4pvF5<3!R&
z&MTA{0%w>Y<AH_Nv!z$wRnO(G2>B*NA%M{RHQmWr{jIz6FsN(dkU&&-GIex@H(i?O
ztlUVQop@ddZ>{3GTI)Q@^25ef*lfvc;$s5H-U4a@Vg8f|4XQ|92kR3p!|Lmd6xCLy
zigZ14L{~GWN0;=5yPO-^vsp{TxqWjMhbiDSU76ESMDUkHQ42km&Y?DR|8gxTrm><v
zO*u!G+?X^62{;Lv|HvI|Az6<A_*71*DHu&b!-86VGP3)8U->s?gK>>hOyjrNTP_`(
zW~iG{j-htDA*HiYwoLzXz4<g9SxJGVD)+W>Tzitiftv+iZ!gLZh!{;@h==pQNIGLY
za==5t8?r-s&Ex2E4XrkUN7!T%G6?H4v}e6!8?21F+1KP!cU_o0aL$=kO7ZVjq)yh;
zQ>{>B%6`b;ci(Kr<;H9-$w?BX#d`(CJ0$HFN~_}z^{l0E^A<(G&hX^KN9Hhkh!pvZ
z=B&cixI=|ChcuvqxRQ7-`|l`ibmmxB?MU?1Jsabc%wIHqN_<Y)mMh@vw{?Lu*C)}m
zWJDKtG&;N5$_jxKGOR2{WW5;4=!x&|XFxrG^T&sS<!hJL8=Hx+`}&E>Y&IK!ib_C`
z$oZ7+9dc^Ce%i%>fwR1zurQc0J;B_>JogJ;0}`yNyBxCxXDp^@YkNSOvO5}ky7@HZ
zagG+(5NZ|8W)J^6VE6NfA#><#k$R@-*cXJxBMYl+$}9gWU$20j+cTzczQkaghNTbs
z7SvZyjsWEWZ~BHO;$cwv$YPPB9*dx{@>d1TP>)gA^w!!+%O5Et5Hy0=3t?op)drO2
zUfAvtGS@N>?8y&|XJKwVV;l2{3s!Vyie?x{d_+Vs3d@gO-`At1@a&rLu=e(@y?B)7
zpc|IwdrpQSlp8?-YUQxCMHh0my6R7;21XD|Ot#4llw&mTPfs8eGzyXnf#oQoM<3nM
zvqef<_B7eJ42NoYBi3Gnf3L%M_J7J4c~zb>6batukc*b!mnDTA?X?!by3hfP`xH{_
zblO+M6i$5nHgR1mhtY&Q%2L+fbrREPBIXAv2U&J&2>t_PAC``~lx9fP#;<Bhgq#Mp
zyD?yxI$GK~@xL?+F<5SPRc~N9ARj7@8=PcXY}yvYx<t2cg{3Pr7=5+C<^Bb7&R6@L
zpkh$!e2cinInIg2sqcWRmG%aVEtdasd$;+ZlmFYVeqBI*GZf%Te!8yEC_}N!GST&=
z2$?c(_Y31yGOtt7WQRx^qEL@2e2!0{OqbF09XaGIraV?%=a4kQ2h*vu{;T|G4`gU!
z*~s_3ZWi%h48w(ZgOmQliDNeP`2lQ13`z&iWa$w^ZoATWY%Xo!`RRyCONE?|f|=8~
zm&#rS;l5W_B`5k9tbyd5)gLB5;iYlh-NuJ~C}(viF{q$0MZT&h{a2Q*`vmLKf$B4h
zqf#-y@pD)*X!172$k@OR_SbNdP;`=@C%a~|Vpie|ttjA|cFgzKMlJc#21;EH*@Jb?
zWXGXEe$&>mRKUOwNuC1fX?eTU8Tuk@<`c+_w(BC$X>GQC@|mCc=*2~)mBI}ahxE_E
zU~dKJl75RzzOfqOAhSu>PeVX3#X+&YVz)vIbi=fC5=Q+ZD-g7oxV_x2=;hJ2`$}$4
z8PD|F;OL_sX87JCq0;!MvcbNSo~JHLy93RLK5riL%k~>fyTJzE6wAWN^wRsu#}=UD
z9hfPSxoRSm0+O+rdnVQ{f*M#T<C{m}*=hW1cbDYxL0Zp$;9&Da_mvGCV+md8U*9M1
zU-@}t+U*M*j?X7Bwf$}!Xl3MlRZa5=?WCVbo^6ULj^O)OP780aHz}FcmbsC=)Mv#S
zyz17s_TVcbvG%&BXZ-hXyelH6pk=?nB-d?R=gK>AbH~j+k6tJIX+VT>HQC_G+JY<6
z1SMwFG-oID+Z4Hau^x3+x1vxbXv|zWE18T3+AUY=Cd|lIA7F3LF0F9Z(>d&&9greO
z*Dj^j#7=7k&#{A_Ye=qKg4&0#Eb4hZ*U~%SWM-G5T1~&1;hKniSp||%ES@+9>iCHb
zW@l`J)YX!f^npe?<$Stl&U6%`>J-o89C}r|z(pN)*E$q})q66U#EW}_ale$oeaP#6
zpM-%YW9qE;Hh9COwS5;5I)A){bPDqid-!RHIQAw4ZG)!Romal~^-Ddwh0LY7Jj8bO
z@J?l-DY!poC3&|!rmOM4MmF2Eh(P-Ne}USe#TkX9pDtcco2G-~tiQQgYv2oDoUBkq
zGv-DB5LKg`_31VE+4o_;BiVFE1sm-6LUU$#^_?xNmgFVMMCF=UHxo5wGp(Y@I0}7D
zGua0vzS~{iwL3RKGC$v+&_z%Zlr%Y;;Bsme6XaLJ)%N;dytxk^@sR?)(S0EKQ?vzd
z<cy%iIaBEN{3{&;kt>iuN+^`)tCBn&$|KDW2~%Zp8*wgZy0MJue7_g0<dDWrM)6N$
zs~eof4nV=?U~Qy_$S7;y2dy+9+3*-A(c??93Y1lb`_)bm3SX+q{;}<E^l*G-^LBkX
zq(7l)#QPVvf|Hnrr<(d&TSKIsZS|`ePe?MsK)JFje-X+s>MNS5UET`0#IqN;;{F93
zMr);}d|%g;6&oYfkEC(GkGQUTJ}JZxfu$Sy?)L#b&+8^}2q$i4yZf&?uhC~mKqhsu
zU^3Ngl+2+5G%NI*N+An#e5p1*a*6kje-7_3LovHk7Ct-DtZyH+Y3shBHuLQcBd*5x
z9$J0A=1>JU`&a~{7TImFj%cf6V3E(oy8+>5qTDiPQ^VdqCavCYit<TF!fLL)gHb|Q
zyChl+op|rgJ7<j1^8W$zKn%Zm;Som{bpq$$WnRmQRAp%k)v9OACE<3Sp&8nY)||6W
zw;3MK86f5ON^IcuxX?t-Fk6D;WYLeM@pw7&KmOZ4^QVV*{J&3sqC6G~D=9Z5D|MPY
zVtbJw_I5L6E7git<Na}q&zYK0n>k^wC`ya{C2yba0Ra=N1LjY%HpU(H>3iDN>?0m8
zvi$Rw<*rG*ykE#^0y5V6a|x6pIe%n#2A6fAu7#o(C$-+mvQ?!OFOgKU0oL`z!k@D&
zrzx6*B&BC)n3-o+O5PmgUHQ?~Bl{&PWI$^Gk+o1;^~^o>HR4qX``tQn1&u))(KF(R
z_kSNbo~I-0W#x1(q;zx{&PS3YwBkhS6<T%MZ|tFAd}$yY3VeNfAg`6*iZJOdH<M=?
zSyPfGCXOIV7UN;PP-rag-ciLk>B_lYxL+?U+E}ge?(#rs6^U%k4VL=0jeE0oo7d%`
zt&vC^W*auC1Rra+uYDqw&Xlhj+EM7OfWRurvU5%1{9sSDj5C}FQ_dcAZm=vXmvv=r
zXByQr5Od~moY7J#m*R4)>NYlDS+4@W{d)?2G7(}A^5<Nv@t~DQtsL9R|L0%-3xE3h
zOaA9K|HeZr+)G7i4ur?GdK;|7Oxn7)N5C8FomICkqTR>oUFMbQfZTKl0Kd!smU-ns
zn_+P_B>M^-PSeYH8Q0?lz<L=!*--V~y4Ee>__^o9j6G?GC#ZsY8#?E*+U(eCuAKd_
zec+>WH*7os(Mk?gFYanj_>i6)D$GQsT4mLhTX8s~R-$Ca@&6nt&2=rD%Z0TT&SeR5
z#==qyYhA&NDQ6@{ui?LDGq&xAey<Lk$WD7*Pivo-9yAFxGv-s%77zCKGA48cU?n;`
zCRh74c(9sjlz|4kV_ZelYn+F!NQd&;4?oFOmRMYAM<-8ZLF}G4U;U9U-@oI_hqruv
zx$t#4Q`(uTg@sFGXdNQ^&+Dhe#5=_Fa~W68f}gL~vPSyz2Uf7`u5<hI({4x4KHBdB
zDSzg}TH6r7rokun%p%(j2co<48E=bu*qfGPR{QG%A@R2Et-+KtIZdQAA(ChnkI*+A
zkx0iOY#eBV7?o8k=ekg*SG1J9%&eE)2}~KX3aLRY6;*g>t7mGI3zt$juPbG=1-Dp6
z&jBCLjA2k;F9QvOGL3R2Cs%&fUW+xk&;08hK|8r~cU;mgio7AXTRMMV#cJxU$6Iqb
z&fy4(cr?9f$Z=38@5IE9KcJ26N$&1d#AvFl+IaK$o<F?(l84iYhx3^~-oNF|^2nJB
zs}xoW$4SUpf~#eVFw1`7j5Tt-U-{bUvb#O|;iS=z(+z!h%DFz&GsCXm^AXE2yvXv;
zU6zMV@8>ZGCFJVaL?*;8>$PW|)D|3*ds#LpRY#dr+;x=jv1L*I3}v}Hz#HOcyKCzI
zZCO~?&*_&tTC4HQ@a~P;YZu_jl)`Qg?8!+z1Mhpwu`xh@RlLJ@cFG5l<Y3wpIexll
zZY7%@D7>^vHBd`YUKUp?62WMf>4Q0tv5EuPr{%<XSy<MERV->7cHPV&_))*b)_x+#
z+Ed1NHlUxs`wrXnUsWUYbt;|aW4DtCqf&8u9NRgfN_YIRcj{Jg`&!k@^Hsc4chlxs
zvq6a7;$iLT3;f)J%a&enu^IBfxh(wU{w>pAX5Ku!<DcLBi8l}LIO)n_g^QJueU`K@
z5#K1bMC*QYJ-@U2cDviO?rRK><H;)9d?_Jq3+-w=^8kl`@{ud!m+@DR7Xa&J{A8n)
zC4Sl$v^WX*o;#?I?t4^)l)Uo^vZdNTYh>!CWX-s$SI>E3nhvCtMoPM8NA(ENHa7Ly
zRiUzI;nXgCDt9EAs27mgm?Gja=ZVKwIG4(4xu>+1hq|IxL03+taw&zw;lRy2ac<6m
zMg1!#)%t&eBrlJQ8wjWDxeslikeMs|hSH}>laW$H@T@BoSI1rNC1el)BOcK%J<>Ld
z^QD-VW|rpdu3Izjza)@4j_8H@uz!};O9wro>WxlZnOhPxXWm>M_`2Tn|N5W)N4|c1
z<X<1&@nDVj)+j8fROpc?Qzmct?8olq3%i`=))Wzs`q<u?#nZH9XRHWmMB#ct#@|wv
zeQRs10qCuvOp+T`*R`6XBG?Q!u*}QcMvLg(-#3eu?nlPoyrsJ~p%pmJ6Z7oZlo*s+
z99Ri(#k#d5jM~M`>1kZr!ac$=|C*X-N_CdFG<jQanS^<Ts>0*4vPk8t^E;X~gvNO-
z++P+{dRbbJVt32@+S&Wg(GAvc$}<%qqZ<2EpacVKG9%h((z7C(+2{<KlN*vMUP`N`
zLEakIB;5h_PJ#UrRE<?DYw?Ux2`-KCx6<d>Xu+kGJ7>;to;v`tD_7_aCSJDn&=&s7
zn?Le!KJmZ)%YWtJeCBO2&w!aul+>tMk;BX(C(1>)4rD3jH;!j95975CKG?=%f^Hy<
zb8<uzhTT5soPoQ=_ct<LWclYT%Z(|zhgh=R(lpZU**GPFktK~TV9gvjTdQ|A+skrO
zV_DnCc-IhxLjXV{K=?D0Wha`iyXPTzCaH|9ZAMvBX6yF&+;g&ymi;0|tBcFA5glo-
z<#@MRd>P5E`G+~7`&eIlF!9^uxG}!+F*8jQxyx}`BjC}3a$|4b^}RSw9zgU-p;)HL
z>?Gv?;w7=O1VEyO<(ZXCn;LDcyuE+RvaVc~6`j0OpJOR9$|FBO&%SFuHNaAhZsuqN
zY@Iu|X_P;z?=gtP`t{Fi&tQ%J3Hcojf%-&?j(L#2c6-1`8!h&OQd?wK`?{^QP)c(-
zR?oa@f!j_43F}sAzvu3CdFE#q^tYD>mUsWcKY#T{CQ1D1{!7kn<vhch8jCLE<XN)5
z4@b$fMHX2Q_#-L&gi&}rWX$lnQfpLes6{fvR@jg4*s(@M9C+*7L0^;m1v<ulZ+sbl
z%i{&WdKo|2(4dI~Cv@I`i5{`s{E6PO7qP*G)zpU5<ZtTg?>cf3x(hKV#n9%FqSh3l
z0<*&qdGh_ZmhRULWn%r{qjj~iu<)pfi#Z%Jr9@7i4lI+9ro>uS7OTwjjEQkMy{BnK
zaDb;Zb){#*JWr5?qKyygCx6fg{)&`?u-Nabb69or!FKo6A)}!V1ZOS~@6u>^^sGZR
zgWNT0@H0Ydp1r0x=vA9XnDgY|POVC97DlxHt#?xv4aF20SgJxwHh@Qs^mlL`Ep0|A
zg|FVdqn?$=`-&+1@${C*vhW~<GmUDMN}-u@oM!J(sP1iB^^Qsspp!+5tbKIQgE{U1
z)aH)blyb~f=O#$8K7&IDqx?0G?;{(nQ;<fxYd6g78icm(n;XpA=W7^uGQL8NEwTW1
z6PbZzg0@VOnWKAIo(0spIUXRQUY~dSFJ6rOWMM&C<6_0T7qx_BI7}W<Oz!!TGUx<m
z%37kdC^eSi9gYq+2j+Q34xV{=IxRGN0;_+YMG=!E=CU;<Cwonjyp*ps=b=J2pUZ53
zsvUs{x)Ayb?`wG<jZZ1Ka8ft~DQ#f0wlFdyJ0d{Z(p<JnYa{C1%ZvvI6mo28cx$RC
zv*=|{CbZ^lvKMLm`SPB}a^Y<%EQh|2gd=aUKPWT-k&It%6uJbcAf(}_%U*y(0gMua
zC}HS;wOYeTgavJt-tOF;{GNIFG@$Xz951r`^Oofl^Ct1Wrs`QBIMKE2^z}UD;mD~5
z5T+l>a#QPA;MLfJKtD%W_Pxy4rmvJJ@r=w5ZPByZgkC26w=!CAeo5FfZra$ev48+c
zHxQs_-23Oj)cenUFF8g)C#Edi+)RjM2O5WBNW-q{V8#{hxVsL3)e4VP?#qQ&%aQr!
z7(Ir<DUgg<*jlAw>WxK2plI>HOKaTR9++>B<WecEaXPOwReq>}J?J=McT;Q8#oVHz
zpvzXz65Te-KL46*Yeq0`z5`=2K-A9pnz94YkTErC^)i0N0Y)=O%>&M*Hfqx#+uwru
zvb%$x#iVM(XR<H_a_*p#QKfNC6-~w@0ymRqtx`z}xn(9YbIO~%8UflVmSmTaYq2NA
zoSB224$SwgD|T}gCBT+qb_DBsxZ4)iPdIMvde}F-{xbgN#tVS;GJe95M>dmhPCC>?
zx8HWy)G*@u@6Fhy?GlmFt)yY>UlLfpL9Jm7lhCHF=w*%!BJZ3rN+zz-P_GSTON~`l
zPAXjVmYIaKiPRj#$XS?=iOC>oW0s6Evy_F}3R#Sz9<Az#(KNX+SQm9eY@f0JgN8dB
zDZsQW69bXk-;ueJ=@FnU{t=(YGcqKibE?hfX!B9o652f0qUwAB)s*PM?S`e{<&@3&
zO!{?n3a%b(?M`rSr@qzC`}<Q|m#Wfg<?H+RJj%kC@_-mDm-noyoRUZJnpEcmKuP(Q
zoF=UJB2kS@lu)Xhg?w#HIrmkROc`TNU^B-0q@<Ltu6Mr%f2oL8=grtxcWM~RJ3DTS
zp2V;j%Uj8Al1=H@{%iMlLC}^+Y2t9m&Y==1_3jFW9P9SEKrD7gj2Dwp+KSpboZhMv
z*@T?kUz-h#jAUcYN@<M}<-d~D37>1_n2sEd2PW}Son<LlYqTHp+^w7a?)0YU`iT)4
zfPD?x0Aq|ErQYVbuR-4*ZOqAtbV_$r0dXQ;?eeP{B}aWf4moa0QKhI|?azc$+vTDI
zX5&Qh9COM6N-HW*lJe$q&pTjEg3h2N9OigzDfS0dqAW!)sq}L2Zb)|nc>`OnJJfEV
zeK+uZU-M>)ey?8kJZ%6Ldk5wl9@n>IZ*0!Y&wsqg^3PkAdoi5I<}91(;N}`!kqG3p
zl?B=+%UvG3EH||)on_XabqBA0zOt;n)Zd6KBDenB32s-i{C$=1x5@IyGBhj`*>Rk1
za=hD*b~f%6fHg9e?0d;^GQl*HbLJ*HiGP=4|6m$&yf;84KpWeYege^%vhh$B9`nlQ
zdUW6t54Pl-x+%_76IFs+4v9t#jbh5HdE)l=hBP<UTDdGG=$`EREQQ@LI_Tve_v~h(
zEV9-huKR|qzL)QBV84w$tm{sXWJ8B+r|1j{nN}L5Ra)|F&XJs`J_)6@(c~h`W<s3U
z0K<@5)p5=~1(Ia^-o$9xIjJ%NH#12yskpwM`aj8|X@_(fj6JE}Vz(PSc-}Lg`?YH{
znc1yDQ!niAdxSM42Qhr#lfLJ3+spCCq=R3^zu$NPuwKSbJHoi~9`C?zli%4bgY%(p
z&PEv7p;ts;icPx*H$uN)0@IwB^2~95jZDUL^r(rM^FNp>wXSFotm{)FW_V~etguT{
ztgigs;n!HYal^A2m&*e;dFFPy;qI6r9eMwFVXZCN7#mZD5=MDSo^Gvmp=o29CoeCc
zbA)XM=JRv-?Ngj~%$vTC?~PrL%Gl1(84(pMlym{>gp4)-Esv-vqrHteAXkB!w&T>I
z(N!?@j=EdP{SN;z@v_HGDCrqm8(&Y7o(k?!=6Uu^x?)OUbw~Vg2rhsEmdRoqC(kC3
zmV$G{Gp4Si>8>$Tp256x<i4L#u)IWO%9f2WBH}TZJqnjn4kfHZ=RPg1dBz|7Xh&0H
zsckc$I*1V@FMIQ059*mjNib{~k2@P>B)x}iH&0IVJEfz$w-c<<U1(js{k2x3YIB|l
zdveZUmT#}B_z7B4T52>)lxjSlR_^AR+v5#~>@_ZTIir@L7OXiJL^WeExWfUD%lVO7
zD>->yPpeP3SH8ys-;0N-Z*Xa3MEaK`q4ziM23|@+zz8GNK1@PV=f$x&jwE`Azz$+1
zNtD*8rBc;fs%z|%ZEZ*r=Iq&&W(?Bq;rM89cSst>Y-<Lhm?e@JD}~OtrJ_!zElM_T
zCyog7L6A0aY<E-|RV%3oB+r~hCCV|<la8VrfMyuY8knM5hun@*>L9CbvCpM2p4suO
zT$3F?T}$1O`z1*5HeO`;=PJuW3j*VuGFZ?ii<3Ne&LaXCoO1Haj9MuXgz4wl%W?~{
zXCTV-Tpb4gJd)*SE6Yhb*wwFZ&s%mC#tbL@`G>aIn!g-bHiC`9*vWAUIkv%0bTx(?
z>kpFSoF;O1IrePAm1fFX8fHRU-EOLp3HaQ&f*8a>nw7>vWIL&N69MszA%b@*M;GzS
z<0F^ljFvJg0oK~cmzA}y)Yh0LFQ;pzeQ){ooArtCGs$k9CifXyL(+9O>m+G_M;q|l
zfv~7%NDAuj^A3D$!C?uHC(0MSj4%MK=6mxr#lF-$(ArBs4F(^fs-YsD$r*B@kujN*
zuZ&jJEd)spV42cdK~#~Pw|i0{=>*l~9VTB_*5V#r<~nedNTFYY1I#u+Z67Qj`c}hM
zvCTg38PR^w@0SVvRF1E+;$Ftz%6I{=UdB&1hGKojtMB{1#m}}GtNp9dVFj*~@c^)F
z+r*P3Pu))kCJZvJWP5w=W~`++DANt1al}2)v0X*A&JmK4e1KLkZM0IbCL}SYBpgyE
zSvOGa3b)zz*~mIl^)~KFAZg$#_RY%n>1$8sPr%C&ItsM^h-{$B$g<e&@jWMN<QZQb
zd{Zu+4-4Cs{8>B`!ND>2o3Gr#r_Rq_k@dMz-7Z~vWZ5>r$kf~Y8qJON6j3XwbpDI%
z=o!g&1_cBcgfS3PBeE|1#GA5}m<=C$a5G(9am?CKmT7!$f3Lf@{vyU6yx5*^I3L}o
zib;6Rgy_arVw2ZlFmLuC0@0Q`Ax!SzNkYUO8!yN8^3r`5Q+aY$+b`p|0*i|@>C`{b
z4m>&lIp2ks%4R@1aZ8j&G<-i&H3xrM9jWu1uz$b#+)G-&wjL;V;@E2!{wbi~^J!N{
z=oGSD=6#*J)7y6i!J<4Qx(k{UJdbE^a{~A#a((=rbL0Ew?)t*XQ0-hR&I2OiO)mV(
zlopwm{dq}BBok7bNCw$#RDD?6?TK;25(cw~mmzhhS9QC8+hrT<Cvej#UB~niuIDqZ
zzOUc&F&fM-viuRqa*$DW66qwJlS<<Cq3VGy%Wi)3M3=oR_iUS9owL_-dq41JDa%*J
z)CSh~ckUmQU-R?N=)1<pCd;<_;aZLv&izw+ZyPb+KSGX2lO!q-O+HLq1|uM{zn1Nc
z`z!ApqTb}Em;=rdfrSq6Y|qy(tgMB))+mwpi)*T^wNh*GjQd!N-+L3^UDd|_>Q^gW
zSj^|N+u3`tWcVWebRXyy|9v0O9T~8GV9%Jc;Bcf<fb;{#GmR~55ASpt+&8|$LXMSP
z`K<*Zym=A;E=Q&nRGoM4Dr>JFrwi*<9yWWSl-)#k5UcBweWlIS{%w6v=#$%5gYv;M
zvk#Z>w>{R6O^#p2UpHO=te5eV4d*<FDA^vQ+3<wzhrN^JA09^S0I#}%E&cfYv%);h
zWSOy+sjaeTrD^j{Th&vZ%h+D`mk|m_b|(%i$VAYTxVif+ujb!UzWOs-7MLn!F_z1M
zHJF-kXu?4gsf9rh$eRL}b)jk<ZE1U?J7=L)^S0XKhQ9xpQjQK}1;`jVT2gnz&yClS
znjgBNN0+KNH%u0f@?g{8#Yjn<C^q}-SM|ub!Ir*k+p!c_J8A8h)opa(%XSK>Qx#YW
z$2v1gCwlST+A6ieWKPvO=OY=ok(m`^)#efEl$d40462oli(nWbo6>z6AXIBKE#vv`
z4HPw{2LA%OD<6Df>|Kpe%zbWtfF1P_jWDG`ZWRB1We4?l?&e|8ias%}{qYMod1g*C
z!i+X3YP2Ra>*c{-Dzm}Q*DmDM=nFLX_fkHR@*Q=_BvbNh|LCe0X9AJLdRn<GXQVWe
znYR$f%XZGoncCKkLvKzbGv+)EBI<ET-*Xtf9HjFA?c$fV8DzaHQopZWCg=O4J9Q?C
zSlW@s$^eFwG|Vb%IZ*=C-#F>IBh3hc_B~|m$zg%FrX&oRg<h7c>fbj9vUig6@~I<<
zRx7m`b9M)zGEr41YtOpOk-cKj6oDX{-W$u1B_k;XsjkHy@5I%8f7<r({x~`W@9>2I
z$ueU3%1L<6QvQK{&__C6Wcg<y%WSf&qT74AwURn9d&OF1xtvG0GVR&RyXW*xvfK~b
znvI4GKSx>K=4a=u52Mcl{_rl8qnXX$@)#v*KAx51$Y>i`R(xMMJ|N7rW~|LYOf{i~
z%~z_+vG238JHqq3X9rW?FAE1<n11();~|k`#W3&Q9mhJCg~xT_{{EhGy)c=;l*zXf
zmr{6KE*vHYpe1Kg@>1U3fMi_Cw~S2-W;$>T4BI?qM)p*frCwIozf@c|_ORyYEZW}?
z<+>?5HHVm@+E~{mN-{mT?p+7{BTd70Pdo6cZym{>Tx-sUW@#FkQLS|lSWpp?HY5OP
zvkXjh!n!U}t@G$@w?wpG+y%n=^AiXjt7Kl<>m|(2E24H4R*N#~xDoR#Gr_woLhMzs
zVf8H0Z+=#P3p)8_e4p_GV7-i=a7b+4vO}A`eSDZDv~Ap--x0RM37%%2#knVXM#1QO
zV_S*d_u3S%-eh4wWssS9y2JZZJ3!KkM&}hnYr~qCrKJFh`e>?1!*zyi*25rC{ZXSw
ziuw#iH<L~@DVq}A(w?_Z!akZ~+cBljvBhKSK+aa<KYYG8eSc)%I>%2DtpJMr^X3M-
zbbc1weJl)EbyLeG8P9#Au8#M9aQ-mHpYUHv9JkNGXoYP1V_AF;!c2>D%565E+}_S9
z`Gu+1tEl$(ytU_|zhkBLEy1>9Cqj->_6$X2N1jS#3vE%#emr|}{O2#k|JWL|Cd~O2
zb5@dg6xjl#EX`PJp;j+%5{s-!ylhTYDb=E+)DJ5-*lQ|JlHb2$e&}nvgQXjAno}Bl
zCPTi)=CnD!&pG^#Newm(=!PP`DLHS+Ryg-`&TN-maee6Sd~Z|Q&+ovvMTfd>m^y(x
z=6NrbUXmPgk~|YeM_|LF#UAPIbM#c!Hs`uej6T`lyT)NRUxVXk+f4#o&=2na`fm64
z`pXzEvivia<)_ca_Ae<y4qAoKN2a?nPzGPwHeH)6e<08D^OxmKUvB`X4exgE9&b0Y
z&j|TNk8hIWr$5W{f7+Xb-%pMM(6InoENrLU&-=CZ#MIj?OpE6k=UAJvWaW7CikoA`
z>_lyqQrGDEr(9~`QVOM3YOR<#|Lc4_aH(f%Qx3tIW!4)UrLY-)z*^YJXeiQQ2lnj8
zVL1+9wd<I!Q+WQ_3{1~oNCJJoayj15-QNCnq_1a7ewdNU6Weu@V+~L&iAIwS?^KC-
z*YTb$*Lm}{?95GXUTqRjkR=(fS?FC0o8GSOVOwn5HTLs}L}r^^Y3FfYSLYh9^D<rr
z@B(1HjGuJ$-cQMm4f$~AxiXIJ`R_wHx~(*7_XU`fE7)T`lJMwhQ%|SwQH`cXGth{Q
z_*VPGn%jR56s)JeFQu`F@b{np6WE3KYlFJ7JQkL6W~~dCbEVXR)2TawyqOcrsw%6g
zcRfnNT0Jd)%1)3<YFCQk`wpkuFdD0;@%OADwzk@dhM0`bMhQI1YQ3{wL{OrJzNa=@
z0*e}YRr3ym%07{a-o4lLj|A`3{sBIDw%y8(7fgHVdG$4--c?b8tTTlXk+NYm`sL*8
zkEd0TM9LGIltzuNlQ>s_WGqLhYZL-XH15?AAxcI#R?gAUjh20v8F#?EjL|1)caI{w
zj#9t;VITbP<o9oJkh|a381|t%iXz-eVosSk-5@g4no_EAX>Ay9Ra=yq$S}Nz$urLT
zg!>EyHPnRnr<Fq~{O|w%pE%CSLP2CjSvai=kC(#x%ZXE2NKA+sWm(Y#lEt${3cS9V
znUWB_3dq^{pcoAszIzS@+(}H3awg#<xYLwLhXbosnl)PWG)`$@e5RnjHiFM7d8t%5
zmq`d`!wpu%iB(fLyQQ|+3%3l{&5`t<Tgd6Yg~2dw?Ws-cz(6%cns+49?B&1SBrPJZ
zk?d4T!A#I}h|-6cGz&H$*%1m$4g*_BwiWZ5G0MVv%l5dK4F@p8dPk6P6Pt_BA&Sl2
zeGYv7fez3|HeO`;XC}*0^e7}3+bE;=ziLo(3P}~avTOF2`pvS;?sfOKBOvuNmgRx8
zjTNwgcwIh|MTf&F;hdh|{oFqO@!Wp33+RpXJFewe?dta*F2{|vwaDZlvTkBWm7by5
zpCQ{bB)}M#jscWC*2-}*e*N43p3h$$S;~QT_iuUo=FgPkfYnzI4?JEjJYE(qS{$&t
zo4Nh;Q_gEc&xM>bQ%<baGi;|R1KybD`;0z|y5o%W*EIm#MBFu|I<gkcwjHYbBwI$C
z(djvQIX`%<M8ZHDXmfrmHuKo%MSIfU`9#S3GwxTo7ZJZM9c^8myK73piJ9H5PKn_s
zX{%tMlD&ymat=J2_?yf;0}XEglT>03uC4~r4wzTJs>H^z_SDB3Kla_f|7_0yU@zlk
zJU?Clte5eVjh<9IB4tkv;~%(3J&o`PFy%Qq<s{T33|zfZ2#qbBw}1L-gA~JbpMt(c
z`PATXedL~d{_*D5%xUIT{sK!EES)eb)Vi|N%2F%f?Qtb1CNpL;PxFT@<7#hE;6RXF
zTcO`?E^PD~8wyXc;MCX@Mh9;&k9KqzFNur}f*!RBYOW;Y<ltK`-`vmVZkF~QV0Uq!
zEj{T!@4HG`+=u@hTmZ;0vP@DCX96j@Dp_-LUU<@p<^Rv#zcopcV|kv~?*LFW^Gn<^
zvu<6reH)qCnduqIJS_K#!UqcZ!gqd4ev9P;g%sH(m%FpFJ;P<sPFHtVWmVoH+}+Gn
zfy0LbK%r*l9v%^yRqk2%S9!XJT~q-m4$l3YlD<Lpa7Rt`e#j!AmvN|S?)Z{-LuIIo
z=dFxiDc!eB@>+G9!2}h}QE*3(_u1pDeXi}qaP$E^e&|R4Y-QXMlP@kH5=iRm^V3yp
zvO@pmG-*0&mj6}<VIwT%Hyp5wJ8s;Kwcp;1eD>L&f!?9%ExP}PLz=joXYQt{(yMb;
zEO`?=hcq*n>K(EOuVq{xLdUktr2ehVs}>@Hd62dJhTWHBwy9#UKHhJoa64k{w4G#{
zSE{#sQ7)+UMx+>^wFuwM=SqIbLJ;fcQXgunw53Xo=9H}9mueS9+81whZ>i5YUbhzO
zDd|=9X;uwEWAWJL&&YC3RGR_H>O5NhOv{=VzEaMy_WWwjw-X%mvuA*(Ps`<bK4)4k
zl7`W;)vwXY>&oYn10uMzD9r2b%GtapwzS5RwX_>qblUv>q9m_rxkQUp`Z=cMb}k=C
z%M0y+ID5{Oma|1f3N5dS(DMPbyr8aMxkfYA^4az)?@LE|t~gJ}?H>=!`jK?(S~>=5
zaiuJ0Dby&yl%3RVRNPTA+H%e`5%DvTeLC>F@4m&k7rgoW4QUwIi}2mscTBgp9MZS!
zqB5%)bb|M}e1==oUBfP`oLT7j5h#1A(hiy(>{bJ=mglf5Q_aG%IM$skHMH_Tn*FRx
zmO(adsp=wR7aGpjq>e!ro+hx)YB_$M>oeA(B(wJ8N)%V7=|(Rr-9UvB9j2LP%hl|U
zl-J1`F4=f2T{+KBmrRX9dna)VLc^2G*|gqQE5)LI>isAYcq^vg@uNjIV0HXpNR^?b
zg^#u8N~;VV3(DoQx_-H>!672}u)(`PjK*Z6&KAy2t_z{Uti`Ca2xWyCAom=N)tjJv
zv-=u}0sq;*=3*S!2IZjd*v((_P7fTW9XI=#UED(u^TKWik|w4&ktS*v5GjRh3xluT
zo)7mqXk`|KnPzXcE)ix<vv4h#r&D7-;4NEU3I}qE6Sa2a9tNjYRZEw3Pf+VQN?M+!
zMG<QjaF*9jWn_zVa;5H0d6xONbr^z_JkI-UR5NZDt=l>Wol%73EnQUOTJeC~N?n!6
zY{FOe{(>>@q@`%4<?JhkQMVg%5sZ?w1I-g}<cBKtmqpR_Hh)M5e%5H+_vNKyIzIQc
z)#t6?xOW?z1Y-1=9_CD^ix#;khFW}~_FgV}+^A{!;-(IoU<vQyTflL9*z@B0ieLT3
zzvOUu%Xi=X7H%Uuz2V({;`VSL0yt-$+#w*tK$>%{50R?$TJ09wPbKi_@d0^8BqcF5
zNh$igq+GRYrg_P2$|=|6SVZ`;N=QPyARdCVE}AaW@2plw#u|2I0X)CC-p>y_IW2->
zQ5e@su5%@oz{Z=<9Nn3pUXBQHVzou0CX&urZ4L*eo=rZdQ+|F_yH-|p^+-XIx=LBG
zXeIn&?31$E6oHx5Ulf_XS=0NT+>Z!<U|LQrPNnKxD&1}?r9)liO3SuyZsI%SiPEzD
zQ*)OwtG~I>@}l(4Vl#EA&_`}3pI4y7ra2GJ3qh0sNJ-n$A_p0typNV!!`ZnfR|`~g
zOv^d_2r)y`8Cos}l8DI-^K7(SssV}keBGQbB_Be|bqXtmcUkRhY*33bUFg;|*IFp`
z#~0OcJfE@vYk|71-viTM(=nI<YlD|SN}f`<-Ol|)kua0axk*4yAw#k7L6gFlleAoX
zF;$2<CJx-jTmJ5!{(-1J<In%qf5~Rr6JER`{^LJ!6Mx4peaY>lOzKGD+4w*R*?^TD
z!s|4~VlYpXrq}PI<GNFFCPOjrB(%OZ#|TlW-7j+ngCsN31|iG%EQb3cFOtlWl=-no
zEebR(kxEV*o?P{kLT4vW>EZe0Imeig<nY0+<30JTi}gd5_vLz9Av=pxY3|fGWs64V
zL`cqL<CGF{Ww=ypL~(AMjghV`xuKHIOevo&A2johKP7YnR>u#9q%-FZu(KNd1De9|
z*_|?jwR~=es>>~*6-h-QO2iZ`;zo;zTv{;a!?x8z+A&!rQqY8;-055V?D@@iU-EKu
z#p@RX<K{J2FJ2LbH(Xu3CAoj(-QgXH-$PI~R~tgy5f5*fqP3sS9k!bZ+ZdeB#nzh+
z-uI%&cS5#KIh-F6JWdN2cX*TT%=4_^k}0mnsB@Xg%p}ctlaLfC-Z4p5lZq0lbYz2c
z&3Z}sPg)*Qeawv!t9XQ}P?H7aGBK4Lj;^r^QI&z^IwF#a05O=*92{wmq-br8t#N<}
z0kc}{T^XMOQ8WwD#-yW0miFXpp}bhB%KDIhX?;;~*#OoGYA0L#3<<QqpFECfvn;6G
z^Zvak-_d0=WkE}VrO%{F@F_>oCA1Xi)}rt(sV3TZH&<h+-d<M)>jOLP)bZur@8F&C
z>h~j?jpOq5Pxvf>jMp64;BLNS_x4-xk&A0(a*;b1+20-515+%m!W-qQ7;HXBx>}Sm
z#Nh}iOC3K`JaYPhGetY*GsBCk9fmBmr|eZzddVrlC0S5oLSw8Mu;OyersRUt0@ce~
zWKTIm$ASshDL;cu!z~@BoXT6R-{hevM76YOmv8E_0l|x%=bAf@0yR)1W#zdT%4=O-
z$(1)ma~CG68dWDiF>5YbuuauIXYIVD@j#cNwk3ju2ljXK{GJP*2rVZq?ycp^i$OZ?
z;Z?*2nU+&S%bqXqe*dIs*_K5R0*e*Q=1^1coE26sapGR=-vHHw<!nsFylr2MW(bTC
zP0?frBjy?%rIj@;w0!)CW5a+H>tbkRtw%hW+V*}ia>*$zYbnWW;|U^6*6N|E?=5Rp
z9!tw@jk-mJE|W1{kI4-roO<d4l;}0h`^Ke*X}LyjJnCWp(V@qi^F4HY|6V>r$A}qX
za8T$N0nHsl8-84PafzH;%vcQ_FG5s4i;J1oVhN+}`SmydndI;I#Xr4Z7y}^$E?&Im
zjgEYA8wuO*n7;lUi35JK#|4i|5Y*VrG)Lx?n*l+U`9Ew;sSGbs_)nXbS-Gvzgj>EN
zPF7tM*Yb#3CV?~Oj<`%~F=oRpIfAlu*rv%8$k)hO3GE)0V&p2fs`PslI;vWo!;9m|
z=-P=jQBYN!cdI&f4kd~=Ju7MEiMr5=miybhy$I^e`L;N8Gw<ani8+U7IdydpRl{FR
zF)loNaa`6#IP+)=O*nU)GTrfGMK@q|{D3%$-F-+n(EH!FL%njQMX|$-SC&3#ti@^e
z2*mlQ$Z27wjgWkNzsmfj9-=~Gr?*HFet-KFDJEWAe8%=-!^`U{ynD%ZyyLr_kmMWo
zae^Und9g(&VV`cf&F#69TDm|rc4(>DWK8{oJ%ojEI&a+`(;^-?mhfuHqND)HNo`|Y
zqK>pEZVI9f$%Z@0UA;=(lB(A_6Fp)(j&t%<`;Yv7YK&aR(@kn)NGV$6RsBHYr%$;W
zz6-SmS9!FOyCR8NEsT_-?Mwl;YbD7AYAcH@`=K3)uKubKAkALV;(1>DiRGy-%v08v
zZsa3-`<V#K$0!kfWxgvsK*!5R3N7c4XQ2ggbzKu_=}46p<CX1%xQru5F*bv1wvDXr
zIdfpI!q@w6aCeU1eRIv5&tCB2^EbRq4i~QY>g~)FFYq^t^O5Td$DSRecUDINi!u-=
z1ixJUqs1jHf~f$(bWYWrsUbOH(==6u?CGRNF_v!@Ba?Y<BT`x>Wlwa9xwDtp+36aW
z{K`=CJ~}RsE){|JbUK<+jcTcSl}gshUA~Xi5plR21!KR9g-uxsK-YBPeEqbd*JjYw
z;J&=qMz@-BAz6{us;o7~3{BCFJf0u7ApFOsWnFn~3N5eCPrH6g%Vb)1nU>{g&~hEO
zL$YXHiaalByYby5=4cZZT5kD6YXA`O`9(8E7XyN1ZMmi8_y}5F&I<dt8L3WrxY2TU
zbZN>FcBix~xq_02MXHDhQA+)u^#-cx%X}y;Q)#)ZR51t^K5x#<e1mMns<eFmfz|bB
z*XB9m`=sMyzA8;}N`df{qt;`Jc|R;1xjKb`#i4Zns*4mkG!fH*-@g3~go%Ip{m*#$
z@+EIxzhS(%=2|X!^?E|x6<@uZiMnSnv>BEF0RR9=L_t&>_uywznriCaG@HY9Gx}u>
z^mz9@5|a16EC#D(5sO?cIWk}CR;!UktDN<cHfZOJ^Ce9<JAu+-VRleu@?^HXCCsus
zo4i>I{T5m-#`j{xDkzSbRO@z?sIXFJ)|sO>DbAY#Fl9^9>gg2&q&AK#+9#Ylm!}-v
zT)bM!@~NXtlQsft*@5>xv1376PKA%=vmf6atK&JL8?ZY5IA{-}M|`+FC`~l-I6TSP
z-(&^P4)8fft2?EOH~zh<NNvAz%6MmHcDm!M_)pCHcgUA-xw^gJ<?aRIM-Y#&Vf5F$
z+Weg8CgQY1W|u?$tzP3{KOrIntMtxXQ`sy!>x0lpn@(N!juu~;36o2;O#57!S!ygO
zh2fgW?HcE|nT)0wnNzfY4$a1#%t(({>b)!@4`m&(vA{VGA|@nN;X{i((d=_`LL}mr
z3LLniJYm9f!sSP41>B27q9`SDnjEq6(F$sHDX+BOkGs}$$0z9d-Z`A}s77KE$hB2w
z5|buoO(b%fC+GZX_pJFn<4Hu`rr+WBJHGrna<|`dcYB4hAu%xc7hG=#zIYvodWTM5
zWjjZz!ldRJFU}7p=!rE-NJZg$<XTv$z1HrkYvZQNk%3D%|3bT2M9(zn7m2XBxK`&k
zz#(NiE?<|L4ZNfxwU@YOS&KjVzBv|}O(`+Q8Sfn)zbLJlj-?j2&CMsA3C=~76{JR)
zuuR#6ZY}NGKitcHKQt?0iDhN}1n~!=<yL)J-@JLXGK<qsCX|_(x9Lm#{_bheG9@Ci
zB<E|L(UQ8Fl|Ws-m1Hr}W*hUOfE8l<0crWZ=Ur)8BQZ*rbJn!1YKk~mk!V~I$87dK
zrvt}i5p`?(WT)jH44**9F8{A=yO^UTCsihmS#HfKp>^LM-}F2kueM_5UkT+V-SF+f
z^Sf_;&%2v<{Qk?|gWG}(9HxZG$m<tBVU9cI^cB;5%bac@Msubn%`%@iV^KD!l&*=7
zkEXN@>Yd9mXl5jJV#c0YT)h$f3Lmew>hi1nN7@sh%v+cX9s7k|TRJY6xQ~vD%2HlV
zaUrhL`I_-@jx*kQf^#R2NYo6`&Skf3ETeWS+O#>ide&*0NwL(vOB6Fzbzt4Q()YdJ
z7(&0dGoj<hh;G2@_~YWyI)621R6~UmmyP@-barViZG|O^0qZEfNmz4+`bK+LE{r&l
zm?g5~9U&!d_K;*kyb?B-_^_!8rgtL<Oz{p<5fY74r)XXf$(@c;uHTeC<cvOHVzSt>
zYARe|nq|E*BM|c{tz`_!|IeOW6_fpP+D_C|J;m)$wT(4nKj-vVIisK&3D4==l2aj4
z4qvWCUsoQbtR(1i1(CLJrsUchi{Owa&SF7pAAM1oKLmR8`%e>1P~->GaalW8A_$Zk
z{%%UI64ic+A^KS5Jt5x}*1}KytSEcE;V!}5;R0cY(;a?zfeV)wAt;`4*f7V)A?AJP
zmc73|NXqH*F*v3I`}f+IW&5^-_E(`<*1ND$dt=y^@*S3rtVNzK>3NQYE*7EtXjm-I
zdMPWlrek}r{jw>g;shsZg`D#>&3RgC0<x~*lRvOpPc1@up<`Q!Qp@m@72o<$+n*0R
z5uQ8#IJEq}8<oMFDl!(ur0n(f3DL4Hf95Ne=;|_Npvo*oCaK1&<4cz9j;`vif1k8`
zjwi1^m_KXL(I#`Xv}~lPRrU}#s&P}E-+GxkE&nk1KIpg@Y$%drmZj(HSgh&5tLG)P
z6vO$dMQ8Rwkx>N>I<b!j?sj*~hlw3K{4gP5!(ocVm<ZmgN5xpqnB1B3<m%G=_d`+k
zu~dFm_H^0s3*Xhcca_GNbWRQ~*Q_rr=b<3Y`xXTyP=d2h=veFZ@=NVAhcgtke86hZ
z7K7B9CzlJ<@kl-owOpe$R4&)ldfp#Lf33&YLiJRhYu-Kj-O+I#-GJ5cEa4>(WWZr{
z58_;cnN#%`EnQHG-^6gg!rOjg5lGbwNfO_=Z&118Rf_2Jj`_Q<*^6=zWk0_orkS)$
zL?V;hBVo@t1O_=UP(pLfuT#3(eaaE-MRGgd3FiXddnB57+4-ERkZdi4QVnM%moBr-
z9Lr_lGpQAP-_Mqc=`lsLq|-Tsna0EHsPzX#ml{4k%c<5it=v!0=DT15!Y~9>A;zg)
zeVZ%h2`hzfa~>&@*FhR%W~#L_f~8-mtJm{6B5kTJw2yrPk11Hz@m{D{5mXltaaNN?
zl*yT(JU1e8@lLJ;!Q<ywiRxI0CfA;UuhTcAv}cknce5vb=Q-%_nbh(1+dB@^j08|U
zFwI}1LnMqaiv3KvE<|agJ;#rWqSXAKCEfRch_|1jh`O8roI?lo$SbMG7d)~Hq-5z^
zQ;I~5RXA=%{0Y;7y(`_O>^~xULdPyGk(f)wY7tViW5N5pE;D%W_?%261Ru<+Kkw}x
zBukxe)<w1o?SPWg(6l^Op@gfXrncdy3%cbR9~uVRvBC$@@}exkS@X&9#n=CImcsK@
z`kIt>PlA@sj#KKl9E{hOI#@PY`>VFNSJWv{ny6@KcC<B8DHQP`w7k5Rdk#N9%Ro){
zgtfwHw2T$7_F@eoR<x|ddB##5Ep^9x&m6@J=s7$(sk2%-6#n07`KiL9Vja!}9F7##
z=vZvu%`4eDf9H|cK8d=?_26zbF{X*z-8(jZ!$!7z`|eBbripKF_8g`N8c~gi-l2ZN
z4T;e!mjh_@L?qX9%nt0N#Q0%5%GDpuXH;6!`11@b4K^EVvJ6>@o0n@W^R>lU3{+C<
zhs{ZeN%1RluNlZDj1k_)lM^iExgOxj&gE3WT=Q~{36Vtb<^*<{=~$FFAILYg?@2EC
zp(T<rU%S?g%%@aZj8)|se>@c{jYpnOKMy*7%;*NJj;DiTtI7&lp=&%N#fVYzp|vs+
zJ$r}>`FwxK@_&!=s{#`TCKXg>k}8K7If!zQ#FTc{k)?Qfv!TQx1n~^cb08a7a)HJ=
z4H1YRa8Os7c1cW-rmhpk$V2vY)KtnvJ6zqZl~^0zp->C%mhV<ZEN0MZb!JzE^`(Bz
zB3$R^m-Fz;ecBM{k~S6F;#yfqXzOcO<VI0^vgcXa6r^IZTRJ|~o0UKHILDv+w*^Y+
zoOXiFV^8r2IxfZzd&ZKbvD{oBdkRzInJ<@GM7`Apd+$J6lh2fuz!YcH3F;hkoY~LH
ze!5{l&snIPX^T30Cyd?`lOw9sLaR&d3sxS;_d<)jX-a2lvF+TlccuL-nXa|Iw4Pt7
zTOqCYRI3NQm<-A{7CO#!B#S|zd{lD!#Ug;~iuP0&!*o@dovY8V2;R$CwHUkVf6G~>
zt0!Bpx7<sX%bn~ET`$OeORnoJQ)*&$BEAQI09saT-=E18JiTA@w>~jiW<@`xXtM`t
zO?Rx;;dn!qby2cx<Cg0_mX=cyi?aEuFz%0|W&a*pE(|<hRf<?9H9EZgMm1P|A}yO7
zvDS4>(eDiTEFPre<wN#3OD9+>;;vXzos!vbA6-0l)r%2Jmluet%qg<l-{D8c<@VB2
z-sXvE+A$R)DGA36Ir=I%&lm!CNh%#9SZ!FbsIU*B<7ST<2^OAQjIUZ{<GN>6S2g9<
zvUUFDT;(+ht9@l1#9J=6aK%E0Cp@`Gw%UGrEyAyFF1d?hj=Iu?#$5YdS$(C->g`+W
z4-`6H&hsLnoLPm_ksa3{DKdai$5^j=%(UMJUG$%hA2+%ItK(V0JI~+;n3Xxr4C4iY
zV@?)zV{GnN2`Yar0^h?}zi$SY$X&c8NCG!8GZ7VL@kDXVav(Ah=Y;c)Vb~h;OA{A<
zL^qzV54&8y!rJ?W;Bsd%i?GlSGGdv;Z$fg)f~d2<(cG3ERAZ_AT2m*EQ54F;PpaxH
z7O50f@%eM5@Up{{0Mv|CE_+_dUSgpXY%P&s`DRYgWS$+DIjCLRzHbb%CSKM;m%)$V
zrM@p>pVep~(N65yLt<jaR>4#HT;VKdzX-S~d%+#0TIT&MPge@=hj7x4`(P=n79E$=
zfdoqFSk;L`noqt{%&_j!biDp{%56aR8rjd0`EFvo81PqH^mbxC%<R)G`}vMEDc%Ve
zqeUPP6I(wb7oPp?fhi_Z0_S1a<j6sVd5)y`c<+BHgli;c=YsbhaRP}NO<0Z4=eD1g
z)yaygGlAQQkfJFb(JqH?zUIW?on!C`m$XJ-wt}TXPK935wx!%pl{vxSr5cx=bA&KZ
zYL5t3=-JdEsWy(>K5bE;brGn`lXp(CF%QbARD|5pX?E?Zc>jK=^@p<L?n>vU=9w*5
z>_YHk@Nu+ErDf};)oL^6@3Q)SvV!jio(?TrjUVGhB9Y(+hG9geL+0&9=T=bpoYJW^
ziWu?WESk}a#b`c;mP6+M){(a{@bEA#r)0D|hN3)MRP!*5@1<quJ*&7Hg^(Ri#Zgom
zN4Of_Qtt8LwA?x{Q?dcs^?j%1Pk`j~P{AVvL<UMB_n4GJ%3ZwAK=1kcVJyzvsBm{U
z@b%ZfC%kc7eExH84il;eQkscrj}mkAzt(0E7sCZZ7<o6{GR4_w%L`!`tmB{8$J4Pu
z+1)PX&Yq$;=S;SBT8p7q1Kd56UCcxR&f$W${i=yMMw=5H9Rf=_VaWBn);w9uF92NA
z*;MLaGG&QHQI((IgVm^7bqcrczlZWIw-~0%HGQ_$Ehck|t}Q9g<$6MK0(=;qS!0<T
zxtXKI;=FVpT^q<AMwJRXssG8K8?ZWlFdTP+I6mB}9P;QZTBvzUkwF5!9Lg+t=#R56
zoVjCl|DI#TR_HZqB1uGDvXng+%roO0>Nd>ka55pz6WoYMLS$g@!k8S6eZ5cdXsyKd
zhrDEGhNnfrm6@bwtV!8gb)24PO&9gtWt#hEsai@WFlx6hYb<3R+0yx?w9}SEO-oc3
zp}9$(SbR3kjgJL*D+p^p<R8r~ev4796l^cvuO;S-=D$s`Ty$Z!$NNz;<2|$7XYnV>
z7~j{?<zwi$s21J@ybNf9Nb&Ht7eD&i`|kPBA&@tq1xO||PM{Gl!p3`aR4&IYPJ~Hk
z91y%mT|}KQI3ekbW2UwXS2M<Ds<Q!nC>=L~X{{$gjwh|{$946+Quuul85>t_Ax0<G
zw4gnCoqe{Hh^zHZd^Ss~(M|OnH`7@u#JUy~C9iFbuFSNIGb6Kg8dSWqco=mtkGb(8
z=R>EYPnNY1DIFNgz9}nmq@Byh2(RnN${#z4fB)T?KAo2Dzx44~3x`=~!S>|Spk<4$
zPMLisVoFGi=A0s0r{H|eyj_h<mGd>Q>V1q(9!1O93S4?9({w$4h?dv4lgeLLJoy|g
z3(VRkcblA`x&MNyF3Po*xtHFPpH9o|_l|eVkC~RgUpg)uILlbx<6X$zLp3)#XbeE-
zco?Vazr1<+PTdkmuckDcaVJiA58F-PYThtRj!7m^#m@=xia14eMs>!CM@sI6n#0uo
zUa|*{r+YFV%6FEiE$e8W5LY_<B}!alyAG1&%=-Gd&6jzz&SuEMmHJYJMzb->2|lOk
zt$1?2usgnddn9TxZs&Se&Q(HMvL7@zUa;sgFAi}qrC8@s4cw7>J~k^<Gcy*B{IM}l
z-!GMZ-`uU^X`>skI-V6`ZpFPFwrt#n-C-uKtn??x`w^7YiC-ZLS3UY9S(u9&i6qo1
z5<Gs`5}d~>)+8pKn8OZmc(=`-*qKf6BoeZ_1tP_N6RjN=rEkZFNMMUsG!dB*JgS}~
zktAw)=?|{tg0}yUg<j{@_C*(UfX|`G&f#)eVkoMgkcE_W>>FZE9hEA(R%Tb^uj1_m
za<o-puu6*27_F)q$xyszFEkOCeTB0wbI4b3T7TR6mO{Dpx$3z+(D4U>*fT^E!!R=V
zE%V+Y?V6^UA0dsjR;}gB$KsW!LQJH2kEAU@gv-&hagHgzWSkD%=^j)WRlr4(n@D~_
z@i^Hd#hjwb6v=da#>;!3Fx#N@>ZaUsyBc40Ej-%zkCsNM+GtH>a9Qb<OqGSsih-*x
zk>b5%Z?4NxU(I>IQn<B5uv+tm`zKbE-XbV#vh-f(gu#2%Ibxct0%Jq7$d!Z{#1|#P
zWdl};nq)buJO23i!DzYt(sEWGpW087mKDTV4Y;TWVwzEn)d_P<fA3wIqWC_FmMt=)
zD85Z&vWIE;$XaBym&{<L574qYQktr)AUNw{mL&7-#kr`GYLkf?3DRl#X`taV!3Tyg
zGVfz{CX$OXtXd&`0M>Ne{@31?%nK5#X1qE~JB9!+u11_(bC@GL_M{kzcM)-la|zW6
zO&-q<ht;1gwd|0D>X}-l574ouL=tP<v~&WEBwX&KS&U+W_%-D^7Q^CMn$GW~Y}_un
z`_%;{MgFFC?_BfbTqYz70I5}3GX0hMSfJI2B_;i_=t(J|Nvg~+IB(51)Ga(Y8?c;|
zf*d(2O3bk~r@0!k^w?3+(eVdDH(+%<11zVb_kqEMrMMx+EJ~_PYW?Xa9@0-iHWj6r
zySQT`j!oF2%%~RzpIc||OB6!3AgO1RfDACY4HL?oYYNz9bKBaes{vM<&;wA7l_`lh
zybMSXoFw-9**vGh(lw}+{*SI*(|ROZotAPT^{05IMJ#>6iJ*D>D5NB)m;g`&^Y)gD
z;mF{e<W#9Vhb)wP@9U3>m@<TUm7Vu?p}DCHVBu^_TE<CT8?B7F^Vz2XYvmlQDDk{E
zV{U$C5p7f~qv=*cI^I)joFkg@4clL{%<NIV*}p@FfpNFbTVL631AY{kCeoZ>ujb7a
z&mbP>;X-DVJ$LM=US7!_&SUG6-ZP|=VVl567F=ES!p4*Z@RI7r(K^WaTO8T&mVamZ
zUd~S?<dgux=ZMExJB*fA<c;~y{?6lCgl?&i$cfb=G8%S-U}f9|XP#v>0X<Q=smVgJ
zWcj?6%=4eg4w-5k>v$UYR9b#7s=|51z@8#4oAzg}6)_&>qNVy*n^<AT#fg*m`+fZW
z!)V#(r3pE5%F;LAN6U*ived8>j6aLYgS1@vd@|!ZxLO>3s`A+4+%l&UotB>xve6{C
zz&uZ+sHEAve&(Ldov3rFc>0KRu~Wm5E(TbyBuU&%cMNwoTzvPIFc|%aSAqy5jwp&U
zL4*_qF9Rc<t!zo+xYgTgSSjN@siE`$9hZXb#aQIM&nbi!jbuhyn@DLevZCE)N+)yA
zhUXGloa*d-Wmj8Wv~^pcXmKlA+}(l~cQ0<mU4s^P55=WOaVt>VT~aI*hvJstmSV-;
zJon>$|H9p$Gcv~6<D9kEo@>gn%D<kyw^*DLI~*E}N=b@0*^xcgMA>tSc|^gtjb2e4
z(~|)j%8OGT7gY5O>T?h~Sf2P&W#z=QzSMl}bP`$qA|OLV<QeV1%MT2xVMsXgbyg3`
z)a80?+m}qKo~(7M_uZLGY6goG_f5$JkSQ#dl<>nmnrtm5HJ_gfx{BeIOvrBD4fMEZ
zP=;{U$DDruqxCO^L%t5CY{Ibhh_*l459|1EYr**L0ZGMx*2Ry)S6{Ujmk3iWVbzI3
z<{#lG4CFr{Nj=;nmW{j?{+Fpl^2a?a{YpG^IhlWvV^pp!eMU}R8zIMxgz8CD+~0d+
zF3x?dn)uGcQ-Rb{$<;Aub6PN>#J_FjBF*$GKol=Wt|R9LD@MHQX%fIp({D?0c?9?{
z+237a;Cnfs9j&m!U2?L_<@God!79VX_~&H6yA{SU+b7Iypx`X#CqXgQgI1(Y0STZd
zuP!A`UBKwcv_<<I{mP@(bDiwNN!!Gr@1($^QP|1>z~ECXM(VYJJ{Qn-JKlz~P|Af8
zOJ`h%X_qN7xqLq$vEELr(@pTErb2PT$ZoHO*Mtx5aN@2#p<Q?MX+31!hvTdQk*_nu
zD>3I8NWpQO>yxF3Pu^bh_E5*DHGI-GRvRV5dYu%Q50PVZ+KbE0U3R!n26a>akd35+
zqzKDLYTl=4|JnJqD|iynfFDew@b3ir>(m0=wQ9`CDVX?;299shPczn;Q;wWVv6D~a
zX*YxnIO9#?5b2MZzjnI~zHbvYA|b0ol^wfx>zmFAP_p`xC6`%e{wv!45_^o(udBlN
zaxeYdxPs0_2bnzK^X@eyQGZXwaD@4^hY6r@F@WXau>Q3u{msvpZ(D1`i4{kL$_{52
z7IIgjBE6FT&4!z#XJr#pWW=_#m8g+a&iDpucO)X_OWcP)%YL;EuJol!)uBk?KvEYQ
zz>FU%bAd#}F>+VO?7g;NBQK6%nM!BHO{l)xj%a?9v~pL*N#elXU{4)jSOs5**2uzo
z-0!ulqI;E-G9qjf*4nMEmgERmBCOHc%6osS4ZUFC@A3=j0cPS?#YK!RwDXtmcT|dB
zcjU`Zxv=C4gV(MZNdBGN?b4S0N-sYdVSj2|)SGA33<`bq@59x#bUgCq0TgA%)Dy6W
zEk)b^Dw`y=R4<RrQ-LFy5quzTTZlHpS5rouHX&bChm#McluBv<kh`%*aodJz>lFrK
z{YdgN!OS%fsU;XdN!A}Y*6u1%(<fxXBR5nhQFr<!1v<Hi#MPW~2pdt!duIR<_KKfW
z(z;F;Q)mLR?HQ5Lx5*|fE;vyKm+*V#+>dSHC{J8BYNTai_QV4acvf~sxAj{N)6DOr
zP0Gl6y1~hbcGNNxm5W|II7K$+;zPKOO9QzZYqD|8%|HIkZ;ju!q|c_lVQ4w%ihIQ%
z+%$Kk6CrG7g&qsGG5pe^W6k3j{0G>C+3x%v-{X6nug4%Y{*rH;2VE>?2YL$ru4-=W
z*DgS^)z5Qutmpo0LZOH57fZ-T%{xsI#0DhE<}JKVNJ(8ZcN8f2F%_T*2|M~a1}iH5
znot*j$EJO5VpRzN(dsAa`oKD7s4zZdkP<}TKwlN;i1T4+_;k=&vQ|B7?;K|eD+L&6
zPTJzu@)}VYSHC?-t_@w7xn%jDx)c_DY|r{8zlJ3u0i%+7RQD`O4+(t)EiG!oaI%D4
z;GrzOLZFTE?Z`r47=><8I9LRO1DfGklPvm3qA@eO(^fYycAZdh97Iwp!vC(Y&SLt0
zbSn2pPc?WDV)%n|PS)Asfl-c}TdLD<Qy=tQ?~i2f%HT;_xzSfbQ`-NEqkSTS?W={(
zl||O`@ufTQHQ$v-&VaPb*F+kL(#jCp_bNsEKMeLIBz@4mi4U_wj$$VW$wcR4r1V1;
zY{tSJYn&u*t{7%7@5qfH;6O9O@#zk_N`JB|_xysRvP>KF$jsJ+nScDJr88~%Z{Ras
zQ+Dd*^OxX3>a@)L!&%R?1)Y+JET{d9UwE{)9`)Bml8Kp^uj_PYA0M=uPcrA(l)d2o
zMrQJ#3fx8Y+P*xir2Dfl2Jx1GB*{(=%Sv@k*oVJQG4%?GWYYH4Ri+x&##tvSkI$qE
zrlz{)GV{o!zB4ad?nr|g`NtMEk5H9b@p%UQ8aYoes6PIdbX8{^m{&GGXa{FK_(xU)
zgsO7x@=%gjaf-|-%NdL<PZaqXWu&QtDMJWjeO&5F{@2<?AL@Cbqx}C?h!~<kDEzMl
zPLoG8KVJW9i8=h=PycTR{%;2&C2wAd1b5$!ImIwWA>$8By$OAjNJV6k%ZO~@4x%4N
z#NI>-4}=aOk>k&r8AuV*o3o*-ene(`O{Ce9954MVMB}5jCNhlh{3Atde^j{6MQHr@
zjcu8JM0?0$&RtYZPg`Gac9AgMyLG(k+r7T-$;V*%iC%<)`#XRdPY6D@RPX2EJ?6H7
z$Uw-xJ3A&vwh=}9cFM3i;neGY3;MPWa!|I{ev5KWx$7f49x02b1SrIPjcbWtjs{PU
z1@0|OuyMAMM_YUh_3x8;b-s1A6P|F2GQ_==7m=O4JGtt64jBHm&QGT_Reky~AOCgC
zPa1b`j}g0ib70$M!|J(zSD#HhpNQ48YiukgK$7<yyVSvUIT4HIVq`=>WO^;|mV-7N
zzCl&CJoV{n=#2^f$A0vIfN;2TD;=%7ukVsGa!lICWi$yuG|h)MWK7s|lT)tXkn5$8
zp^&HPXO0*6%i|_|mFp$!W&c#tvuEWh<bLz<{O;_u;N|lCIpqn3qSYOs>PPYN5OT?d
z=l%H5_kj95cOEkF@*w#VGBYqlx{O7)A^LXa5$lTa69a^gj6h#sA8Y8<N7a<B!_NU)
z$S?|8%K(f!x0hWv$=u#&O+S9h1n#?!^gI>6(mz!F91vr8{X{Ra@v8FY5P}{CBZNuj
zgZcs<K)!8lfmyUV+Op#6H1@+?v6zSHpuTZ0F)~Uw&bm7&G<ualgm`e24v;6}Ohd52
zw$pz=T8`|YJS=>9G(xm`o&=(?@aoo`-XCG8H|yDfo@K6fA?8)G^S{h!W3B#@7c-+O
zsm_HF(na)HVvz^m?=Y46Z5_J>ClOpgR)KFy?}n&~cc)&x=~}N_L6Zl&vb9?jS<OzK
zxo@qu2|bQ*1&^9PU6DM7Jg>flZNA*xK2B{uVg_IB2)x`Jkvz_=p5KK${*x@yf1@8B
zf@%EjvG5tXnNobU`FQ(MV2&Uhg*-fj&^o_;@W6&ee!WeKW%m-BDkbgc?PWwJNYV4c
z+PccgxP1ACqfrR5D6&bCpOARI^SfGg>)VB{zcWt*#Gb0u56Ta81o&Oye4Id_#j4%y
zot9nqq(%iD-zvm3jJc-*H@BCoVLLnFB=u$<z+2F2m$B=OC^Vb2H7oXZ3n}i`55BqW
z^h_(Ui-&V!BlUZAseJKGRD1t5QkRq42V1)C!pv*;7%Cs+&t74D_s1a+5-u}1SMMF>
z^W|~<ecubXFG@v<6Ir-l{_S(}%Ms#z)p#_2`SwD-dE$Ei&pm<qO)x1b;!_*2d)`^w
z$tvXI=xgt^1;|WL1W^@`Xw-R$VQx922%>wBdvU)X5vF;qmBAwYsr@1xHSiyQK_Cbe
z!EE;7ZU4Z;%3YsGKl-m+5D|k`u#ned(#)kkJ>wQDO5PhX4~cD=9j1EotDUP<#J+O0
zbs_n>-2(9H1gV4)+bxE=)<IoE+jw3-{J2JOLQCUhR8^aJr^aiJRX7FdzD<p86k;zY
zW=x!-WRc*^dC=oE2uPinz*$Y?rxzB8+$0_nSG#=I&Rp-Ir|%v}Edp#}EaCyLVcXoN
z!o%pJY>(VmX(WY3b76>lGJ=7CQHvj5CmuO$Fx0Z}53y(i{@p-lE<vWomRN<0yLWd|
zq|vowwVgsDy&VjrlpY&2I?}10Jcy*fBfjV9m+O}>uD*ZKFR+&j$(_14XXAYPz<)=A
z4V%-Cr&*h++Q{=a>1ZUo{aZoeFPNs)N7CRd{YYFzmKC@X-W0jEbxT?y3Z-u#8Dh9n
zvP%`K_{0*k$3mzpr$fsiSKNbue1Ce6Uka5yQc$)Z*GK>w`3n8+$de<KxNNGu$7j3y
zI(6}WEWs-`bt?aGFi{7**q@og7h*Nc5Jqx_=w#h?SV$k3!&j`DH<%a{S+<7@Dk>Kd
zs_vFuL$$-nZb*{pA4xK)3xyaa!FurThU(N$YTU~4^d&qCk@4tNHv=)}K+kp5K5cf%
z*}$uZirs}Lkhw4!+%jz;GnM&f|LR}N*DId&yoa+6L!mFCyWCehJMlGBnX4YMU!6ZK
zEHtuMMhCY%R68=Zm2@c!M@hQ$6fWz%%<O74=glW7q$+Ad+*euM3z!PKrpLVRt$up~
z;Z`a#z2}ZUR_-)89`9RxYK@G_i)^&WpJhtSYVXB|A0>?BDR#tJ7>ikt#Qj1#t^HZ#
z)bMV1&^FS~Kngo?RGP`qu_#a&X^9#WgKT^p^lng&G3`rX+Ht{)Z7^Sm+4%bu$=XxB
zzT|`+oN()iq|?)C^u4ZZEN#C`9mNqq$;ga0G<%7|A=4h7E8%HUjTzJD`!pTwb}w4>
zUM{~{m~7ade-$>Keq2P0##LXxg2T6~FJw!q20t>R?rNkA$^Gvd077*Ky;j~&wf&n7
z%W=rfDJj;m6;nqa=pOrnk@j_6>-e~a;;DapC`2qA<&VLzgiX2OVXbr5<AO=M7RG7J
zxJMoum0R@p?B6?%+LKAQ4OTUIr`@wNf54d}E<I{!n$pVMYbLzrJ%d3H8nTf=%Bt0s
z1m%P)YAO8GxJ>qWulv4`A|k#g&9WGH+)@mzb@9+XdAIofIjneuTSeCKXH-Po&UmP%
zIOJyUOr{g&!^j5cKj&R#Gn0$Ba(+<;!xUt@*br3MLe+^8zcPD|vQ6Ls>xrJ+Oc;jJ
zWW;80S?o$R-$7y}=jKv1ml)v36<FB+o}E&t=62<p?RkakIlR8_j%hGho_2UB(Uo90
zvJ!s*YYtHu@Er=2W@x9^8j94BVZ=y!0$y?R9i{11glFekMIynU5yW;Fph*(p*UI}k
z*YkJ8c!{4Ua7LqK{~TR6uC{(JL`eSC{Ed^A?btZS4Aw<OiGb0f>I7$fVZYI9@fb&v
zxu&^5YK{rquLBJszF<Ncm$n0+1bQGv9M7kRI6Qvd&QRp-eQ4^s*udG|+lhR#4Yr8{
zoT&qFjPqT<_+seo>}7ZP?h~O0Gv_&-WIL;quD{5OLhf;4M(sIe#%b_^eMw4YXF|gr
z!A)xO(s+nyNtrMMlr#tSKI^llg)%y>$&i&C#g15RXg1oI{9c~S8bUpF$<5%fa}fLU
zeE|{qxRGcvYfQR3ULgMHPYZHEEP%8j8{T}_q%!Wb<V!*y^k?C{d0!y7>xt40hmWDX
zNvseA0vG=`d(zC-4%rOrTV78<mDY=aD)7b+PLc&(@aP*|GLDfjr+)OWKMEWPNJp6*
zB%0%8f>8{J<`Jf7GB+B2Q*SkyHXH^=L}WyFrQ>i`&&z0;CQJEapPqhXi(q>bF}0$~
zqu!{P5}(X6t;oJGu-{Rh{5Ma~r}05j2aBoJtl0Ye>kj)|MEZO*2R8LvPEJn1Pb&r4
zrjIH7xzgMU+KI#2noPRCE8l1~k(}U+(cG+r?kCdeYjvjFPGjC{kR%uHT5b^%Z=c-p
zQHRrs1hY#-gC>@b`Kq;ieFnCAW;|{@F(xRe<-iN8e|i_o*Y?A7ExL8pukE!@P4-X5
z9eI=&aKdLicJJj$<jjESIf}ZWx>l6Q1S)bKVY*k@2fI-a@JVPUXtk&d*+Q@G@z=y<
zI4OZg{I$D%M)p-1=~Sq9;1eO;k$AN>by!e^#L5CJAD97ZIlJs;PR?7>8BES+T7Aj|
zcKTuNQ58%EHLLDlQp-(bFg5+Vm0>P4LgCC}&hS(zDv+`!TNERxNKI<EC%Avqgv@#F
zn~iK~mEXpC^xpE_{=Twv+qk%g)!)Uu{4MMV@QqNW=dnZ5D{v!D?*<|h#quE>*7sL3
zYBKD6J1a&wR27f?7AuVvK?Rq5{VbFuv~*nYg7t!4k1X&<QW1&g-*}X5k3DC*bRw+*
zWHi-C9-ys6CX!ArNx5Fvw#$;k_Q~pB$Ws_3-;M$@%b$a!!Oai+@SPGK5tuu~@J|S@
zgq1vt)JUQli-z?tC+;r_+HDeKHXX1rz^c9KSjyz>q?2KjAnjBziiiD$y(%Eg11&Kk
z!_1xZw3^Q-v(it?%-n?cVc2__Ta92-!0DPa=68^nYNkv99pb<e`;QHP8UYMv1aPt2
zZFBY!=`@yJgcoxJzKW!Hntyy)ZgM~O)epb&%ySghE5jRZOf*8CAS97f=H=6LezqdL
zfOwuJeAlH+O#0a!i&QZup|;|2oeu;*SCBTuUydl@xg>C}p}qHpKrdUGMWwXy-l7yS
zKR}*g-3~>`)aawl8v9ta7(oaV%47EM6r*<^D&m)7W#w#tSC)Q{g=3ZMjwB#n#_=f&
z^JDiBLM5dfSdPCSl3wpJ>)%e09Y@ssCw|1}txy1iXG-Vc4*VB&WcP3x>Y}b{*z-bj
zXZ4#Ww7gl<NckOQfHbN5hJHP*_JILNtPpnCzS4Nzb=}n3o_>JNM%2i>D@Q7+j^oFb
zUr+mrkFsm7@||uC)2AsO1ZH0!z!mx7uBZok_3#XCD&HO5a90LT8MSbsZCTNt5O6(Z
zIsm7#Vnl88$8p(A!#Tb)+CZ-F|6FV9TJkb{m#-nT=*!35^O{YcKKv0#h*F%h^e{AC
zfSpJ1<8h;*vqfwK21#>ili5xAS@ZxGySFfcWWAsnN#sn>K?a*AUDc3QJ~RlXuXYZT
zl8B(x<QS%3G?L-Wcot}M<ahaID~NRj!I13lNhBXwUZ8Hz!e4wR+;(1P<Ye2@myv~)
z-Gc)x6tSW2p~G7U;c5*~e4@0}f&*5`HR5Hkm-_&jOh>&n@!CXm^xR8>gCQ9ArOuCH
z-mKzOg|cAH6Db>6in&XNSB;waMK(=k`xp@6Xzs0H9jmudCX!Vv1l1J}4fhY(AQb(5
zvzmJS%{n_-Z>2>g0i<bw3OI`W`i_mod=o-_uI8GTUu^RQlETkh@rjNX6}H^o?{hJ^
za>!cEhO`(^nDlD}?FW?Lqaj$@qP2WLv>RvAK{~Z@9l3QVG+9}bH1IPq%^Q3Is_ldc
zlHu28NhBxbPnLUp)G2aS;-O?7zWmY#K8wKOU66*%e|kLkUnc4kEV}P#+0;>%2duB-
zGfAnMT7{PWg{}_nzCY<WcdB``LE*ItyQlXuQ*D?jJx+)%t4d$H;bn$-8mW||=E4+}
zImQ;ld~5?H>aDT4n2T<WA}`q2Y_Wql<z;xSSfUAykLCN>CR<}=qryb>NtQ)pDtAV5
z@ZS^VeTNVbX~+9D_!DQEb<^|TipTI${fx~dRHINNbOVa&J*a+rmZ9>^wS%r>7k>6a
z+CY^i?da~bEJ&Z|8vc6rH6lW6R#^S3iMmu{C#E_fo({4X!#!yppqNljgwJZKB%?by
zL8yUMY&vo(n440iYGFL6jFl?Dcq<I;hn|=4%bC@o0gAhHE&+?90A4~HH7s$|DxTT4
z{W^zmm?J$f6J(gE#du!*bcc`PS-Nkr+0gt+Py1GQpt-cP)yrgbmWo!puWx?k0uts+
z|2m6sIHw5PhPKfBjYB~!^IW$KY_5~&b~_s+;Gkz}Bw`j)fd;qHm2Hq7No!<qKa`&h
zl<`8B#B%0r?f!BZ-o?{M2Zdp$LFjeM0d^%`&0aDT4ixUf@5_F(FW(zR9YK3Uz|qcy
z=+|{Lt<Y*4r@fvly1g-_jM9~Lx=%y@a+!{MnoaKF%Lp1eg%Fn_hI$Wy#Egqq!ba1(
zV}dh<Wh)9v_;OA7YJy_I(rAf8p`-6O<%<&(e=>uV=tlg4K$$vu$>+gbf0|I(r6apR
z2q~ShyxAe9#BF`f6X-AY&75p^f#O`^sG@Q~Bpp_PmiWU{frC2W{IXZ}o4t<+`!3tV
z-m+%dO4Xk9cW)MS?RW2{;|tD<5Rj#aN|3KVjt!ap?xKWGZ=o{yW}nZ?<R*hPHTRf=
z$>Y6ynUyDoUyAY6;n9e7%bOC>TYC{=yFi|JJuAh1sTwz@45Ub#kd`I!ax0GJ-+2w!
zj@7bi6y-==>)7W}Nw+8Tx+|sJ_q1tUq<D<wtHwKGMJ#VNtTS~|EWPV@JJaLwb=i76
z|6bF+q$L$$Z|^TSnVGxYt)B;j*VjR$QT}>wr2BaQh2MUmNg&H#p7aK&<XW(WNlEvF
z3?}O5h8P<@HCAK|7+80S)`&7se+7w3p#Mj%MJk@bZW(u70WLqL-{#I$pxMLHOJnw+
z{|cy{y8O*Iox}`!%kDMJcfQ7N86Hh$_kYR`<yURBc^gN|PEvNWwmi<FmK4f*`05yk
zSc|?cN457xd$r83ka!3r7+t6<Vv=9;L;BnTj~5g*h+j<^DEQeyI-kfspP#2D=Z}ft
zg>H0j=k3_ZB0|Jli4I7(t@^R3-L8EEsB&W0GYQGXZ6Py-_;=qXe#3*MPXX|KBrur;
z&$IDm**^R!cWsZBYb}v@A7U{qr}-;hMX5@&)F<r=l|BE>?F`SSsHzii=Db8EXgU~5
zftTMm9&p$m*h_RKTS;XKJ2_lBv+KYt+x2=Wn7#bhql>jEp2l2-%a-z^fa!QX1?aU#
zmb*j(d0WcvEINGT02q)9_$g6A4<O=GWRlaNpTgax!5qIgOI{nqnYp~$ylF~skmKIX
zog5FfOaF}W-D(hJ`s-UxF3wojX&UltqHKZGd0N(l?HT?%%k0^OTo!h_LiT`d17_Fb
z&Ds|~Y|Y%$my*4)>-(~jFHWB0X;TC4k{pH9F89Yz%_e@IRu9i8G+be={fRPX0IzZ_
zb%c!a;&t@vxG++bw+gbcz&djJ^NRgj@vw#N_&JdLOdvf64q3b9M2m?hUDj~daGvds
zAlbB9nd)o6`30cd-?E@ZzI?GK+e_e}lRpVhh0KUEjlEG<pl1xI6O4G@4en$udi^c6
zu`0ADxvrIoB$+9x$HZU<?TY1yw0eSsT(>hrEU6LFPdDZ{jeAa_yzBY<4LH9&a>Cf{
zp6<PQ?l<%OZvicQ0x6#>)GNz;9F<4E#cZLe1HR6o9ZRi|*6X~nowd+NY%*kZxiZ}A
z4!NjJY|4ZAnCT(;iw;IBEayP|JUm9G&HF5FZ$Qn-!)QFv@G}Dp3qs)AFK#Fs6<w4L
z#No){evYxW!kA@%P@5TrTokr_m1-L9dgTrXyaaJJNd0#<e6Eb@4O?I_!j?9iO|u+d
z3@OA6Pw<YGsmggZ9+bFe@#CGmc1N1!_1991Pa8J%oJOnNU);R&pgZW}E^#nx7uXYb
z|ArXDh7200rY2>I6m9qO_`F!!JmX$Oy~RIu<a4oM**vJ+fos&F+{Cz*dZnTZ!z$I1
zF3hF9?5AIiw#6$s;kk6TBDK!M<sNITQI*T_Sr09%VNAdCdjz&+TkBHdnN85e;sNxX
zaP4C%{SgcE%1bS6@taJLI#s-wPOS0!U94VNK5|#yNmx;P$iVo+kHqsSF=r+~FoJ38
z>z?A6m|9d=KnexJ+tfZ1;9i2CRQQ&>aWel(r|?WC(^A1NTW7h;%4Ed9RD*->@6r^C
zti&@hAio<<%?**M0QYC$Sr3knT4fXag^04oicRv?L>YijWrKL&iFe<4wou_~;PyNR
z2|xmO0{Z(zV#NwClVc#iorZ2F-@9^8Jyu*7N}Wb~weMuj)mKkcB5!q`;~mW-hl>@#
zq%-O6{V>UU1QZ*6N*c>@d2<G8d!YRC078tBZyv<;N%WoH;)s-xP0Q014DnD_sDe=h
zM(AED)KacCc7tf|LaVve;{26W)2`N^`}HWkC1BQt!}gzZ55Km5JNvz29|%fmPJgDc
zsNJny>E`ee4q#UqS_mP5$>KXE_%=4`M!ncN6k?t{N|cRJ6KLpBKv450Go?Alw<wqN
z$puTm$wgOt<EW6|jf?_Dpw=*U&s>GM#_EL+^<=vHW1YN?WuJENCL|UZs`2-i$4W+C
z6&Za?K%jrN@P7jJ^$Oi2y^m)cZ4csoFJ~NSjErh)Ya5M)jIH0oc;-BnFH32Lf^eDv
zetB^DiQ(kp{Up}a!+y+k(TYbX)cCAi{BnT{7q*f}UjF%5Y$V56_Xa;laqYzSq+*P4
z1wbIcllc15$&Ae<Qb|y{f<1=NNWMv?c#%{@^1K2=O?i^MELVZN;5a|RIz^l|wF^+*
z?Za+a64>YUqXXO6R%Suoq6|3!X`g};7F1Z0LI|t_;B~X&6VL!S&CR%xV=(oG#i6mE
zOIXszv=i^X`@v=^ClS6Bd(WtO^%I{=$j}=tmh%b0A}U~(Z`IXeTkdEPa1&RqyPc-#
zd${BSALnS;dG9^t!xLrS`uQ?7_F8uUn0=7{?M)kBj02R`FC1;du;|vK=`{z-NXJdO
zXhv<8oxY!`(Tv9e(l}Y%C#0$w9%QA*D|q2W-xJ9TMaRk|PRy@b*>e7dzEy%-ROv|J
zH+ZQLD=et_zv)~20^?Z}7BM0pPskz)fuOA`AZW%fRj2!w@nO<%@yn>!+U4bc7!Ea-
zE{(C7;fpCYINC!~Ov&?K(v_P-FywGv;gPwo;m&cDu4k<&xZ-L3&y|gZ$vzSbrgUmY
zJLTQzwDhl{vk~`n?VdOS2P|rpfpIIhoy$o*X?b0fWa5e^RXi1PMApH;j>lM5*UKVI
zo^H^`Y5eIpiwk{&kCM@64$Jd%fjEo*+KFDlsl4r%yMHy8obp8(LK$*J=E+kurK85*
zqENgQlKI&d*|khsA=<aL9r6x%6nRc!pIS>IM}O-l8cg|lT^na{{H4KsRv0!~5tm6B
zQCZacYv#-ds~?M*^DsT-Gr#LSXrFD&b{C2K{1!UujsT%+fRHT6*<{S&o&CG?7KZIL
zkz>!%JV?TV@Xz910!w3c2|4tEUF)G560Y{nKfGTOys~$OX;1_$K6@4jp9h$0rx~UX
z<1c7ZK*2bHa$wIdko5EVxsWWkCG2p+6;;R09Bs>f<dyGX131VY?W2GaWpO4w$HjOa
z4!b>1>>pZ^J-jDPCzp28;EP`1v%+B7*fg)GX}p#zY53*dpp15vzIQ=pl<kXsz*G~P
zuk=JG)zKFz(Dit<y}*kv(MVCAi7duC06gL|kuZU2*0+ohth*xgJBny;#xRH}@~nsl
zM5~%jp!V1drjjMQRm6r&_0yL;w+2Q&)$PJO%!z;<e!huBKuv~lWY`{lima+*N_sYu
zy?R{s+}qRlqhA>71-}xGo85Vs>Z8ayHiI2Ks4db%Fe)9UVX+<g*c0bh-6}(BQTyyp
zy*E`}s!d$!a2c`k-5VX{s#ONvOk-=Ol3fI)+cP;KRYhVC!xU3f=s4iOPe7!{?$K`f
z8hx{JQyf9*KZ**Dz;PaZr=OM8-QBr&WSKzaie79|y%4nuYw1TRw=^mCue&$O6^dGn
z@t~S@zTM5$E~ANW-%qUCWP2fKCzUrdMF>lPW4=R~-|j~pz4n=vj~_AU1r)x1j#esT
zkuRL=)V(QU$1e84$k>>!Wtz@Hv~lnLYvUq{dAoHhY6BHB(IN&vT@4g`lfdJ^``0b^
z6dOm@WQ2Wfu=l=95KRQEn7pr;8^VR!7F6x`fhgfwY0Bb{4>MlmK#UPND?@E`daGOu
zG{C^aKC3>zej6igYK?_!ut8dHv3CC5*T;L>ZsW=&+Tj);CzhaNb@<EG;Ui^L83=A*
zL4h<F9r7Ub5<V>%FtxBB{p1SXe0d5HobEww4dX$(4E0Cr;j?J+JPV*wY`W14K-*Z#
z?B;e)D<mmWlvPv`p2Uk`x(n0h;6?7bvu(1VcpDC<@WgYYmN#+5q*x2p*-p@}_lIAh
z2#N4GSW1HB{AW!pq$#VV^O|_SV%AjBr?TM*rXG*;lW%u?{6kYX>*1KBU!4J(P=PXu
z^+2r-?t<jQ&l5l_T56$tPp`)<;*sn9HceOp01SpbtoruXkst@c)oW&_l^Z&m**eom
z5xKG-8`?8<CnXOD?WS>LQRyFtc@pW|Bwq5emX;omvLvm#-rrS(+-Z~wON;%y1WnhO
zdkDJ%R8eOw<U@Io%aMV~wv}y*dER+oxbX~6Zck6<&T{#+B3x(Gh)^@;14|0@y~fQO
z)B%(M6pi1*qwf;&NiT<;or&Ls8;L{!j_u_KYmAgFXDK>pKFOD0Yw6qxpeyw+VFOuw
z!&wh)uYM%zvk`K%ZYhkWVV1Rb3+MYfw;3y@3~hfTMK0XcJ1f~&&6xl=As_78cg**T
zhyci~FbMZUM({CB?H`h$d%EN*q(aV|T<u@{rwX4QNgj9LTz6+RA)kF8GW~rOa<Q@A
z&>#;dV*%<$YsIZ1d62QuvEk0w>~huJj1UhoNjz2#IC}FBf=Qt`8q=TNT+7rOwCe5c
z-Fy4EcVZitrk&ee_nAY%v}F|Ub|cewX>{O$#xhinDU|yw!g$&O2CVo;P&0%nPqg(H
z6OG|LIW?9ES(_IJgbK9n0DZN}{Kf3R&mx8?OQPH=E2muAn_E$$jbXw&ipku$Z5ZGI
zI3c~X|JB%H@Q3Vvl2c7T53OeF6!l<P1WDwRwz`!QF`WiuhCK}5Bb!UhCUA-|1A)iY
z!+J=G&<h4o%}o2Y><WV*h`FNv+Isi?&8hK?ZySY=FD{)F#S)1=J-Te)0K-MKUm|H5
zaF}rA{8o4z!qqwSo`&xsLfu*HcD7-HSM<#haG0?)$okoJr{dc*TK)F<%cZ1y2_JS0
zi5sQN9F`6}h+Ei#2~T(V)OT|3^-OvKyi2u7x@+&W*WsN*)qHQ-+Qs&Yjwcwrwfpan
zua_8pEw_%QE;|dYy#z&B=@@&AF2|xTlsywz-*PVS!Wl(yeb+Qmd)}cwv^Jj2sxaGR
z<q5&4q2-<ey4o}9{$iy{^nBSpz;Jn%0Efr>1+!Z>7gk;yH*){f26)}~@V*<Hc~D@p
zlt<e~3@WELy0`xw#h=>B<l63t*}SGDNm0<7)nYrrNcbE;O$Gj{Y;nzd;H=GxA#2Kn
zDR*nZy^UNIWtT7Hnq#RRqd5{c1T!*Z{D{s~S9d=`BDudXLX>6kLyM-2=*r_no+oa;
zXTlQAM8b>JM9hX_Zk2@+Ha&n&$wjiwKGmHJTDld3uV|E65nBY{t+8LDd@e|+=#$;&
zpp<or|L%0tk4NbQ@3Y<CNGSGOV2(qj{fIywtowJ@9p1vHY=;)EiR#UIWkdaQ9SPzG
zP}$rij5z>G^EVp;1r2_!OWXc9yBnt)#38Qp4_AKUE(mpxN3i|DS{{<kB+K+bIx#tr
zM-dP?F_kIAWyTv#!={!KoG`#xuOGL;DsM*)Ig}5ovxKu~jJ-VJUG8OOnikoF33JBx
zh$&7?Sxcm3P6V4iv8UPlUCHEe%{3bAU2ub(xj6qbA)xVt`qQctz$q>v1bN^%g`XnQ
z*230m6FGVy>fr$QkkFwAhn1_QOe5jk{WNYYtlfS_Eab~}>qnvtvZ>ZT->zG1T*&GQ
zZEz>iv_LE_@{ChvoJ+xV$WI@}b9yfJiJG2YAKY!vq#yHpO*WJ$u;H{Ov+N{-Oa<#k
zjY>YTt}k|ov~}xYXO#FCmRNPoS^k{)r?bHp!X8*Jf4|fe>odUxOdk|FZ5jw@-Tg)4
zCb2uR&%mot{uK7K&OtxW^*Tl<M%fb}W)=vwv;7XFA>?uk;$+A$ydG}sm~@j9JN?H|
z=vc2Xfdl8wJgAn6k}|E^ECXC3Q!iuF4fr2Vnoak34+KvweXx;PJbHT^G%Ih*QQ~Z$
z!&^qpM(ImYqJX%LZB?%STCAVt&7igZlMEe<x;>V^9oP7AKmQ;>KkH@JajTNBW$=zq
z!7ScWocaS6aYw`Ec}Gu=O*d9M_f}7v2-X6Mw9U`5jd<c|zYKmO3e3r=2W`%8U2F_i
zmU_Db`3O?kYxBD(MFCDqu4ooE5V9)ERf73rqWR-qg^8<JDt_O~sap<IMi)Pt=I82Z
zY)l+stq<-PMl)!DPeBC<V3T`XRsJdaC;IAb+zkUqznEo0B&+2NA-w{rP@)p9P`Jnq
zkKwqMdnuvt$t`_9s)c)4s2yJ-mc$3vM-s^=l$V9Rf8@_kFIR1W_fz`aq7;hsdCGG(
ztZdSa-1Y(TD^0pl=dhSYu!cfkxfOR&Q`s(%s^w<&b_0I!L_pKn{Sq|8gDGyzA!<ST
z)-v(LCxCP4a{xv1v__gG^Cg7#)tXiojlMReHvA+Rr2+bi1x96v#(@GZSt#q@<m)3q
z0Cb910e%*1u71~00m=Iv%;10j-t|3A)pv9SfW<UI-Tghrf_rz*){}@d)KgA<nmY-W
zgEz5cT(mWp;FELn>|-QF9i?vhL)uQYMP3XcOgz(g?*-d2YwGc>=pBZI`BXc|!Tbzw
z+1GiHHKeD=6NlIyPo`jxM@*92=fSt+y{qdBml<d3rxe-}C7J1TG2X$t?&ut*KOqq4
z;ZB|oOW#m(7!x(M(gqADj=Uq*u&H74&NvT_99TEuC&{sN2>o7)V3}fS&i{EVBCCg}
zI$w?uj$~I~w8DK1{><zrWCh-gkUTD!KMMP91`lQRJt>s0`j!G3R!l;v{d{97uv4DS
zJ~va)t)52X@^b7hM_ZH$_1-Ei!2WtPvsWwTq35EtifpA%Eo9zrwNtdM4I2k)r=Q3w
zf4~dRJixBeV&qvsxVWdFw#C1NP!h;ZW$QW!D_p1dcnQkh9qvSej+d9FWxvbQ<TNzi
z)Jue{-*(>~Fo<L(pWkZ=JX>3`tq1f>&&ZAye8rIm(;$~0_CIr{32qJJ4W|qfu!#M>
ziJkEDI&?=BQv~aAO>XBI$oiW^W-We|0a4%^lJT`vvUIR}otVa*sDAjlp4jdG6d&?n
z{_Lck*M5KGv=MwuwAuRrIzAN`YTrNI<^YP|;-Fe{7eG^GN-i%gD8Az23V?Ou<NR}L
zgKL1XY%>Nm102{qR1$M4RUv6ZB*FTQS;GC)v=%~!;;gWhyRhakL=EIQ@7ss|toQz@
z?KA3MR@XBc3F#LM)2CN7bwk9M1o^9_(ekIH4<cqEpk#&r!Bm)0t-O)bFKmd*ZBgy$
zToA18P&NyZx<3Jccg}VrAC*P#uY~SJ6YO-znx1ky>e%nutlwr@)n!pUL~lcz`2CV`
z4DUuT(4F+w@}Q+r-BjULmuGfm0`N=4ai6bqAI4<yu*9mB@vB@86S5kh4Y0!2?|wFa
zl1LoTV_YS&a$-mn@TJ1-)|@-K+xPxm<qn;3z?+%WQV27IS#yzLZ}YJHa&lNuDqJEV
zDB<)fn-&%#3Ad$)I1~1H<a+r<15Aj^S%{$q+ff40Px6D5h32q7F6sb=iGnBImbI#p
z>0|-#G;ZR(25^|p<X^uYr4^yHnFyloJ6k4BHVt0lSd1K9r3#{|b}QTEU6TeUxf>CE
zzcP{m9|ja#4cPxAjPXX){D%X-p26EW52x=yw4p8Z0dfw<#$wnj2+a|PK-KRCrvm`F
z)BH+t>CHVI6J29~Eti{)dpD@{`tR&-*N5BpBCWf6StwJZJKV^eoLqO0*W_={SI*C3
z;i~bDlIZ=BFEm)hYPswT3Pd|tEQZF;O11{Z{vla5b-Od4#0Fg(4x!VvM_HkACi5}v
zVyaA;t~{=-&hupEXk|4;?+rC+<W$-SNzfrB5)~98QmI$>3Hpx<g-`Krt{Wc9)XT8B
zk=iWB{hqe8z#04rAaigmEYb?$d$R~^ew_EaDsfmr_mi^qtj=p=q+<qZC-md@e~TE}
zW_E#?fZFHRJ}|`TRcNQ#cazCHsr2v0(mRe#+|2DiP6zI?Yp5GJ6bE4&_15d%x=^47
zU&9OG2u9J<lqmG6fUuUi2Za|c_s6RK6^PqtSbp5$k%iHcm#-V%i``1Ld7SH{77-Ym
znb1&eVkcbFn8N&Ll(hvsOQChS6)h*eU~T7irU||9t5^(WFrQ0ccq@v>*+yc5RX`}6
z5hTlz(L?s^h>X2YBRKs-2@Lg@I$Ca-ja??^?=yH}D|D(0($l?vArx`V)Xh5o#Oq2m
zz@s+L$B$XPz}&`d#g`h(jhsyVM(MNOKCX_-mS(ZD3Ww+QO;F9T%jj+D`ojlGm>lj1
z?JMB6Zxll=8$r2Fnv~<e5<%>)PY{OjEVdpj8g8sG^XC(i?%r-e*o@Z}^oBAw{j_}`
zPKLnG1B;Uzn-2etgu#5KD;L^OEWEd2eN`m;a$NvaHW%~d?AmeG=yH!CS82>C)l+<&
zG%jXVcwR6jR%Z0$g}aCMz<;x(Ca8ikXS-{siqafO_xm@xcxng`7QoPw`da-0!;B$o
z(s38MXE^5puK3S)D409D-VU7-5hPZFYQOENLW$%klRTpMo7d*`BRUED_$loLqvz6l
zVgk+Zr~xXKVTCuqAb+A>M>3}=ZOsP4#z`sUAN|%pKSh-~9D30cccff>{8SYsr<^V)
zt>phxizvm~rLIP{5ABrIFiuBq&Lm1Lp?`eD!CR|f6&FVPtC?zB<X+D*jI1#a*#ynR
z#Kbemd#dKxDfsF1yraWE24dTvYs*(2=a+{GAkoZ;W8rZ^MxlOV#xw0SyQWk@8f=u3
zCFfu-UDOiPCqez-;9|7zKtlzSM-0w)!u`wNS`}#3ZqaKgk|m|YZ}nJVy@pZ5uxt~U
zh)gr)ymmt12AgW|4fpHKy&ml#1wmRocbu{~F>y2%mq_-6jA*XrCbo0k0lqm88p><_
zhfBHxgK^d)%L8U1x4Icn1?#8tqFvgSOxbA<G%UkIH5Nk!(Jox|$<VTnGn1lS!lnk|
z+AP)5+_yl%I#c9fyOEaP6^s7ObB+!jbxT2BChV{8`Q8Ye9{jtCFARPbhM$2}SILie
z($fKUR&{_p@@4FAfAH)jPhBW&$?edPSd8Szx{Nz+bJYbTYsz$Xp1e$^tFAnQ?!rl?
zw-j1QkeY7&s3%dsCB50mkz*M7;a{o^;|!=Bp;5X&teL*c3VB|7tnYgkIPVRbn6$>=
z`t*mPFhg7PC)eoBO#{C`U*S6pj;sZZ^P+vr#OD&YwVxOVCUF(1+SpoRgxquSWNtRX
ziwE=kat5#Ai&Sccq?TfblRK?0lHNE*RejfYo?3mI)s*4|>3N^uHjK$zV^IY)0$q6w
z9T8^z4ex~>uM1y<FWtA+h0N2SNDc+;<O-BLEY;ZOZ)ASi)wjg$uXm9{g|`)Gs)kVc
zKJR~qjZbAZXdXV&7?~A|VAS%@r?OMNM@fp~p9+=fF1yP0{Dzx8gX4#e2HFyn;rWV*
z?6DAgjo8U)Gn31O=gf66Sgzbnpd+ijfNT~_GuCH&ZSnn?|D`pqQ5e&hbUhePBgYF^
zdqUBbx=-@1a5PP2lX#Ut`&<{2W|N^`Qk61@62CxVf{$}WjU6b6k@PLaf+Dt9mH8o?
zJyq57sm27jSZBA%b*+;T`Gr^}4Smbz`e>Pda}LWOUPHK~=t3){^LC=w<(-&rh^DFb
z&Q&Jc)toJsiJ8eoR^KC#bFjb5t#6x~vVC<Of3SZaT6|@hY=^syZA=eX+v@n^e2RrZ
z45C~(8W!d@q!)E)QR;IBIsa)sTr^miZ7@}?M(U=L>HJ}lhd&owm!Bh-KlWVG_9fsj
z-r@zjgJ%l7)P#iW7LzIx0KGS(XRsD#W@z}AF-FU!)B3Tz+jqj2O@0if1g0zu|IfH7
z;=9NqA8;aN&F4;00uVLWqpHb1=o#%VmOD7Y<}w&uHu%}$Z~LRI%al>C`3ZW`an+QU
z*td;L$@4~9<IXuHLEDV8h#3DJZn<)FXm-Y*$^&Of9~yJ<92X|a1skf7TP_<3xnf+W
z_j$#1f*WSyB0UM8H^=!CirsSu+t4HpEywb-q_+?^QOTDk^Bw-neYfW@-}4O*0L8NM
z2LQt>pzG?78LCkPx0rRXYn_9IF?#_yD5i{*k)T?)_J)vOwYg3O?~FvRo7K}e!BFu!
zI3xXw2eu@#EFD>i|0ZlWoi+k1ahI6swe5(1%xy1*-kgO-9vuWF7JdB`9is#}mB4Z(
zfwr=V`<nTN?4a0wWX=fL^J<(xp-C-%rweUDxJRd19<|l$q?b!XVq<j=zANt4?e){d
zX@i5Z-pQU>M$qi6B4>yq#|k@yL?_XPI6Fxs%{SBad%{flEYC6N_)SOK!0Y%!-~IF5
z`8+<sTGT^7mQhE?NDSw2ScQ2Y<az`pW|jQRb)%uLOU`OI&dME$hv5dED2oZ-Or+!V
zYu7DbNu2p`XMr?X>fc97N<63VoMk&A8T5+Mm=Bq|lZBEpK`Ri~;!^{fK2tkQHmgTA
z9tA$-c%tLv89}TAZ56{5GEu7%DIFHNO3PQdF=nX-^lF86dC`M-(fOXM9|xoe@qL-h
zs?P3Ma<PrwND@j(&w!1>uo%zOjAP4QgX5=NCE<}tC{DFX!v*g;{q5XUF2E+ExTSob
z;uMT<bo*9Ui!+sD-E7%?H}!3)JNQ1lxy$O$x`!{ELrP2v*@G(PXCY%YEic=KA)!%j
z2H_M-6oF-?)L=V}I-f^wP0FP9PL;qu21lE+o61B*YAOO8QnV|Z<7H6`qiW4cJo;MC
z9&eq?L<qK}*rKGsCP0e!o94y_R3y^i4$}bl^d+C(zx=Cz&V4zPeAGZ(x|Y`Vfo(tB
z@M|TZ*_BnVCge-+VedW8t4fk`+2za0o<1r2+UuV445^fFpGSaHvQBAxi5m$$n+sN0
zu1t^Abf<W%A5?3wKbJbKGfx0nU=Z9lJCt1xJKWRqt^1F<wchFc)V324Q^)~tTPCfs
z$f8)7GECUYB<xx#q!V`BrIcl)Kc!}wIGpZBV?2u6PAcf!Z3m-0qJ<czuaJs$0q_L2
zf$x((REhY%;eMqfNUGv+$yDZ~?gP%|Hc?pego^wa4OJH@&{3M^oDA!HtvsyXABIUU
zxd&sUcXM*PW<mmm!ajbi4Y0-ybiez2&BBFXS6d9@u#o?TGbfmwMrMZ?k3;eCT!99Q
zcOA1HU~mzM$NAms2K<@o?P?>4!v^P{?2i^CQAMp;4)1uuzN}ot8-aWMCS&X<BQu6C
zL0;PgNE*Slgy@3Fi}(UtJqmeNkkM3OVui$9Ev8huy4(m6$|M=$=ng;Gzb63IXy>Iy
z6Gi<(@q%QEKCR8Papt5sWAe?}l+&iAdOWr6ryFwY2TGn%=U;H-yfSZwDO0?tsdkH|
zm93}Gc!mbG%;#bsT0XW+6>9}fOQ{!vJ@n-X{(f5;(WbtHro?x=^BvFCMk|SO0<XFD
zNa(Y-urkPs$udwf9>1c4R<U_TtUz)@<{%13W@2V{Fdh5G6U<DTIxem^Urz^gMCwMf
z&f`GB@E)Mo<L|425-VOxOam{gGAo_OcqzjK+@;xC&Jg`?lkp(o_rGaX9rK+WVfDA<
z3L6^Wghq1&Z*6zs8nbn`zqxDoXO~r57hjmDf+(^_PB2Gc$VCBykzmrFppfQm>nSlW
z-BA-K1T3B%JlsK`FFHPD5(#oT<I|)5H7+|q8tySl6W|oBlxp^Lf@!52i@j478VxH=
zn8w5`(THskjBcR<Tc4a!>t+&|noX@Rar1C^v4>N%-#x&Lk?p&<?;D?zT|C3ldEl$X
zmuBWy<?U{K&^TLRUw-?b@Jaopm3V^(M-rSi4N+V4V1Hv`Zi?_yWuUy7e4vB}qk4?S
z;pQ#GrcOqfo&iWr&O(30lju++VC@Wj<WTskT_;}U-Oe-uTkX&{a)p$1UX5U^{BGpr
zjiy$JJ|#}cgA^usqPjVbiItIM>;vb^e6hI%lM#6LWmexE5tmRFmbZp}6)W7TR_vbM
z)L&o&_I7m_cCdV)UK41b(Fnz=Ns-wkpcE<}*aznNTf?9D@<houo6i_Lw7Wh#W>A-k
zkW>Kp7!uHb;OXa3FNqubNUb+1h#C5Yf?9fhOSkIQIp3sEsHtUz5F)wVf1VwtuuF)9
zL>n-LD<f;w)F1vh%bJqd=8>R~^;p3@K7t!V^_ZA!*7agE@)e9@dq(izb&65f>dlO*
z{$5l0N$8TN5CrlhZXoQVuz-0|8aXY02XOhhy~o~ozqtV-H#0SBct<B=)h>I2r3P~R
zHM`4Gm?ugZqI`g7KpCP*Xlo!w$svl{@O37ANVW#1JKfNOdX|!-Y@>a8o4cg9G--E@
zR7{^CLxf{3A)#{1IS?<g`?*h&TLY-)#epRZ{ZvXy=V|hxl@j&-H?cv}#4gcSt7oGS
zdcK-JLO;F$;dZ*;&TQFwzl9LBtt8%prp<bwj!g__z9()Q9uGI<tt2GKlAbIPa@tdI
zVTy`h-75j~^4j>;@!Sft6e(M{g`yF)VvrG?&cvoub}Qgyv69-1J(0J<VFq(qQyz&4
z?vAT12(_k!+bFvw)~<gy?E{Ztuw=MS%WoSY?mGmVv=q?P*J1JJH0Ksb+?h!xCP<$(
zf@$G`E}(~lt>j~VhDTq3Nv|-I!PJP)UVuSahzn#ptv_sk`Z?qAQEr(MJJV9XE`Lu8
z!7OW))h@WlVqpDv)wg(_ff<BFsUkSYY8#3@0OY}b{M#f{Iuusl_n7MOx4V5AcOGk+
z6@+eK+ZaPGmP^~YxotUI775|m1g8gcF}lmE{Ynv4H7v%ZRsh?qU%?%;Cq_gsKz^~4
zbE@M78_bK>_5|{Ly>d3oeu4w6u;KYq3VMs)RlcAH;&ly0)q{^N1vD!jlp}%N#KP_N
zu0}5|rbV)A;uyc-gULym3A3D*GmJ8I-<mehvl&bF(?2j~>sYH3XSB$Mui-3|{|1Uy
zrZ+b3H#*O$?;gazVd!+UiG^Tz^L*nJTfqWap(U2U^d(>TIAg-NUM`og1}>rh;y(&X
zkd%dJGUmM3&5k`ulluTB1zOXn`ieUJ;s#aT4zHC;*;JKOl$nY%@Q!*wwiq#U_Pkji
zr?64Z@sHRCg|X;@ySF@FwGv=aT@Si92pI1?htdx?eS{$?u}e3XHYx`P8<Y4EwVLq!
zR0}(QlqIZjVli{4Vm!}IeP5RlhFmI-D#i10g3ptA9Z6wdL2h)>Int-uYw}hej8mI)
zWH}dSZ3T|dulMD7kFKe<KtT{bbLraFB)#wdF}_Aj*g<{e78Gf8Ivyx!f1kXj0d~4?
z%E0uJ1Ls7JeXGX*n6Z*j)FxcqeTxOT9I-3BT$#vv+^v7%iy+`^C1&V&kF-j|9sOF8
z(e^TNj1T<F6yFCvIn_m8-h~9Wu=D}ZgFLWeF|jPH3(^*xugeO|`HmQmTo}&h(Toup
z55oZHm7P^V9bMTl|LN-(IC;2`LYAZKFkf`&=XKhA`aHB%?r=a|bjlR>R`_n=%f9)Q
z+w)c7CH3=}+x?TTIeh;)S~v$07%YoAQzdt$`&23`cWkNo(D2&>Tr%?x@gW<CeWvFv
z|H+-`d8Oy0O_+~5dE)8y1f&+Xs}vE2=Q6()UYHCWwMx0qBA6i8pZwMyxk<8LarxOX
zF4qxxneN$))ItqO#Ho7RoX}8-;E!2T&WZB9jaK0C17TFKmQX$kryzMZ+QX`Gc?417
zyEo+Lr0(<#OV#d~b$jA7_jw!HU+o3BwN>@pJN!vo3_P`{#)0+Y^(>$y_I_~<pX>VS
z5A8M_rgcSHgt7Q8432NUxMva{HGai32+A)dP7y%x_t^&OtD!Zk)R<Be$<}30!hC<^
zLL(Z_I=`G({l0v<wLff6T-uKlwlfv7N+gb2GV3nde>gB$q^>}|s6PPA-Dn%oh8(YD
zONN71c~IQc(7X8t;D|HsFA0{_hf?j;p;6x0Y<&Ub8uE0)3gQeo{wO`l4!nqYxPyT6
z%&tC_8hE5#_f|y<7Zi2;7=)*AQB?lHSYNZM#<Y+x9FpSS*Veq@8Ci$=9Q*&6I?KPP
z!gdRbAfU7;p@5`x=fHr{jWi71LrFIXO6Sl+4IteO(k0EnAT6CUbW4MtdEfKloIhZ`
z%>LbbKhL_?y4Kp$7Ney@SGEPMLG-M_@d&k0*$r)`yn}wl(or*?2})p7zc9C2bD7Bv
zoP6Cbj`*>^zn1R?e@YeokmYLTlGE%hd~*aHpJg(kHo>bQ6<XbTrKzA0oK8<;Plwb-
z2%4&Hecg^s;48fl^DleWSeXd;j0o&Nw9XZdu1o?^98)ktj@A8{)KdP6GA&EX3(5|b
z*(6%Vfs!IGEy1twb6jXhfkpsp4L9(%)slBL=ZpIQ|LE;KI^y6ilVV!&nNgl^udhLW
zF!TC5{61W2D3fd_{ME(O=Bt!|zBj7B$8IG>b<5p5AwEo{)Y-(>r~88O`H0>qm%0VF
zx)ryJEN2SX>G`?rG5Qp9fp0;mdlI(&nmZw1SXvzdnM)~PlC7inPM<EPb^!c#D~Qqk
zXL}Go>zfjxWmg8Onzcz};?o{ix6yzd91mS}<pY|g`5IM!)GH|6AI8Ke2VO0JAGSW2
zoFY1ntJO!YFfvO2o^9oAL5BWAnELyjhhaiIwvYwFT4fp`MDb`s)vd=*OQX)G8@)Ff
zhL~F_-As{o`%G}s?-QAs##G$D!IPNeD<v~W*I8Cv-7+iA)U7>OVNZ8jt#SEk%o8R~
znt9P1^Kw&-O!AmKJqsdV!c)A_2^Mp@S;!JD-0N9;9O_20iEst)Xi~l1p_sLvNM`Vr
z|5Hn{;kSHxc|_nn9hyb2!au!2vw&ZE-BcJGDdtC9Le%eX81^b{pPow|M}VkOgT;~i
zb$p>W)P$G5#fpS%fP3%Er><HqZ9WZcy`ongJ#rc{&pb6D)|sNKTYLBW<Ql64#hH^_
zWJ(~Vq&w`N*ct6++7S>u{nDVh^mO9o&rYmctKAmLiCbq5569qIJ}kZK&f<OnN@{XL
z&n3t`ohWp%bxi$F`FRD5C11Coh~hI<7$ko1A<u(@DU5j&%f!Uk4FFjVxD>oNhVET)
z^5ORFF0^4VTTo-mo^FLx$Q6%6vmPNU9_w6e=q$3gf5kznT#SUuh-g<-Nq>ltt0=R>
zUt;R{1?zEr*knu*L?Lh-hIJv3x3B&j>%;lx6ryY$$VHC0_LU8TZ60DyANRD^bW0b!
z4CpE8%m1?t)_c7z|Eo^ITQvOLJI)=JJaeA3FO>g)6SmQ*DNyYt6?meJ?o~yAZXJiT
zUaC!*H<ui<q~csi?d&C)yq?(Nprme0E?gQ2VwEjKwNvAe9Ipn#$&gcD<P?f%3Z$v=
zvl2yE^rY8q77gQCMl_*a?VXmw<HNSkW*k;ID5&vlt2cFGv3J&CuY=;5(%OI_LiDm(
zEJ|t9Jpo_o^u^Ufx2Yy2lR`x%h=p|1_<q(>Qw$jiePBY_(IZF(NW5`kikFYfqwxU*
zoB)@RK+RVIs0OCDd`%MZZVG9clBbn+pwT?fH!H@wzps{lPOOverC&x1UiIDm-Ako0
zv1A=osq=H7`}Y}<HoW0dJ1?y!%bplwFSrydCvMP&GcVo}tAUo{WCR!@Q(i<{KhAj=
zRZsFfS7Ioh)1Ue2JG2`rjLxhsROmfC{`=GH*txUzFw2eZ`5>hKg}dY`{X+K_+C;dX
zLN~v}^nO#owfzH61!borp`7UwUfPavKBv3!;ywP<?R^YW5P%3;vwaI95yQd_G!rv9
zKOcyUHN0ukZw=E7mL|P0-*jmQ`%@;aJ2d|^=doOB%xXPl_|WM$(uC(oia#=IySA?~
zAs9a4Jf#BvKqh+ZBk<omPG8jd9BwFcF2JS*muo&tgv_?$bO|@@QXJG3)D^5kO8LuR
zlB`|63&A)1l0{>vY^JYcCK`9O24&I2*a#Vk)pw{Ft22vHAj~oz)U<D}j3|BO3gYkM
zP$_!B#+xmEn0BA_3zqq+P_(-k5;|Mq<{18M#e8UGT+E~+pm)}$ZynauI?i^m&giX%
z;YriM=l4k^M}q@2T9&VoQIa9TY{9YRgr8MDewU^Ht%X58-G-^bSIyQ{sOp`wNzh4X
z9!1kD3#DOT0n4hdP~WguEzd59qlxkT#T5pCO}?Q=jrx;DZXqy+!PVu=S$PHio=2RK
zPl9?`c?e`{G<iXqZ)wNff4`^y<Xu1<7t5uY<`e_dzd+II)va*I3stiCA|cFr?bxvJ
zgO@R~IPW**Ctbw|nSWj)nU-9)QAgfbIwnR&wbYd?ZL(6o7a2)r#84fr@+XQT{Gt<w
z`OkmpGZw{(m3$<_B(0q8*R`AoXBA)MK{s-kco>^*MFzD5aDlQ3D#oo_Go#kglrn!)
z0`6eek5XKNHSipcCK(pvw0-Pb*Gb+7J9NX38K}-Jrc0^9Hdad5U~i}Nqkq(c-8(c5
z8m>HI`!=E2sG?XSMQ>7)nceK!5j&^wi&_+lL@$eAN!85T9}jVMz8_t21f>n0U`!t=
z%S;d-ybzM$xgS2H+TtN5dCxH_g5aHQuyUKT{_A-L*c=(%&*iu;s#ew5Z~6t}v{iNO
z%jdU<zI~`|T;|2c+DTL3crlLbuIB})qF<?ms%3L@o~2QENfV7!b@e`1qt5bbuYRyQ
zu&Vt7+M<)5T8M&z37Sc5aW%Na1FmRS^HSFD_m16$h%_Vvc()~SL33loH$Kell0{2N
zuwkP`!q|RS1TV4Ti@(Z{%RB1%L{2wPq!=eP_nIqWc7PYNY;KT=le*n8H0uh7&|9qf
zhas1g*{kG&IAM+Uq>#xPdi>3FcblHwkh60@q6Ix-GN+wp{QK~$aZwISC5|S#%k7L5
zcEvm`V~?lL3O72rzQ#{dO-9Os$!{5^Ys<w8Gz-{~N~dby?A{+=?*bWkZA7^ezfT=5
z6hD8WB-q9<0sTMso6vwlSmNu+-e2Xo6(OX<O+dY1j|zDk-%_K$+kUb&Gbmzi=1wt-
zZrRi1yS%2JVuSSPFWffq)kVpyXnWKOE9maKuYa1Wh_f(jZ1d$`jy>4}Ur~~Sj-=tw
zCg=X3eFb|dU)GO!GxS65b{p-PTR+v4(<2y*hpBd&%xXZi<ivFc#?HbjXk@^Vp6OXd
z8aG;WpAVbf)T;jz%KKeIv0+mjiv;D`#{vmi+k%}F%4afo3%AR*s?e%Y;N(%@jNWBo
z&>3KblPdX_6tOQ&@{dj_jGC0Bzx^u4jCIn2i_yU&zwteB4?*uZ$m`PCxjuk=Elm-1
zkz)+3&AlGGpdn$x$5b@4-~`U<2<@@+Yw5{@m4b}$LV(;<)2?Nvx>f_T;9T@=jpO{$
zqY~t@7|XC}RJ<_^oT!kD63DxOO}Xm02<mJYF$#OzRv?}#8EkK>;8m#~c2vB$^@GHJ
z-;%Q>K%6y9bkVXrT>ZPge)uzv&J3IWB^PO$`IsAvJ&VNUcs@3XNIfDxB|H|wQE#Cp
zqO#bwfyB7v^w}$azRV<=D;98&nq|w|?%{cBsyyV&&omysJDgvdKKp`%6edhZ91O-~
zPy<e{YWi!zS}C}F&`t%O*g`ApkFGpEdJIeDbqsx)S?+=<awPxi(DQg>2N;M<`0Xg)
z?^|q#;u#-n@U$SV|H(d*kELmWkbTq(>vGqtvVe^Wkrh}Eh*HdLq|?m@7tRQ1c<e!T
zsdyJ&FsZdO7jbatha~E2rD14t0so<5xP4q(kq2u5BWTBZ(Di(`g~vBZSe3B}-gt8w
zmZUiY4YknRqsUM6Bc|BNSPQy_2*_XA<-QJ*SK(K(pfZZfOez@Qeo^agL(5nPY(3Y%
zSRhWOOtqLnX2i(a6~BtQ7~626EtNjQ58SrL%Q9ya0XMmy|5c&av7_k#uO9hcUp5nK
zaRq!PVdg4--L1wWD~rOC2A$2l(q+?vmd5SKjx&33y!S1H#*&GtTuA-c*^~d}%B#?b
zJ$Ci<@|K%hj5Q$dcDK$k8>e7I6HB?Vkes49OKIqS^JWa(2YXpoWZC2~RhU_j!I`}b
zsEh0#OI{K8FpA|;j{0h?s<!vrN6V1BS|ll-%{_0cBY2?hd_~W+Zv|rdxVd9Qg!%T_
z2!t;x0X0UW^W0uc*Vydmexr%g5Bw|wZ$<MUv&9p2Dnxw74}lV;+~QA4fyAt`nr{8*
zMAB%KwyspI2U6;u$zpNm(q&atR&=MTz-($6gG}IewC_pPZ44e7)*mc_=AXS{){J#b
z;#pfgJ3gq%aCVw}w2G7Zsu@3-KkOC);UFs1wvRAaF-&gBW&XcRs}R*x`r6NBG9o3z
zWHV|>brmt>{jowzwTNhqfn<kIaJ0rM#pKWQ#e=n=Lu!AQ^WFL$mi#9EIpC);_pw->
z_z&zpa~&$|@*ixH)U0^rQjPMxIF~4^t-x>Pf7%;2hOz54u+T8FMoSh1=!u(Y3{ua9
zIl4eR(pxC718>nA?7IQM0`Go$S;k~DW3F@#4F~zcNgPd0q283uaUQ;+VH1XggM`Ad
zS|v~>$A;dgG%i+M_%vH^#@vp^n4&6wCR>%y2_<?FvibB9_1vA;AEu+oi00W;b8}5r
z;>yP9dq8<Yzm3;okl+m=`dB!39dP)Y88tbn)5bpPIbiqh-->&}3M_}^u&t<NwY$z3
z_ik<i38;5=b~|($bBdz?hzt`0P7?)?kGTAh=bQH*f+KV>fO%V2)SvfmL@Wxa(R$-=
z+xH|ePjfsj9u`MwGt#Fa?oHLHR`fHWVf)N6Cb*ltFVyA8(?j@fK5Dxpoof{JD^#&E
zJ>5*>uLRau@VH3`9-I+VbJxtf&*33<2#Z<eol7?b1WAOV4&dyo9ex8M5EV2o_I+Ig
zN2uA;_TRrh6NGJNitGLx;zJm%dqt@LE^u=({xaGEO!Ep8POcj!-MVbW<7Mq<wEP>8
z3kMlH9I98>=ZejD=~bpq!|l;5tW6tA#1?{I$v0St>WlnO#JNOFOuE<qLlg1tW{%|X
zt4JdqH7K9MB|LcQIIa+^&v4)SZ18&vV0M>Ld6u`}jH+o$Mk&2cVq&+&wV#(0&&slq
zi|9vXG21|{E`$8VUHh`<xR~=K6=Eyfn0X*QH85uQXOt(<i{|9~0@BLTin5rnPPqNX
zh%=%nB*EP=|62MPk~npC)qfuyhn?LtM9f~{-v{od`^9jqk^Ruy^VtZ}w8{<0(7th`
zS!#Pf@AuBmGEA)463Qy1k#1eHf1VD1@T8nZJC4o2zhZFA*olU~J1-^xqEM9~Bs6Or
znwFM^7R|oI8g=>%%!*`g09jua6txecG}r{caIG(de|yDD=swm|R2eG!6vc{if-JZk
zLo4Gw8XF9m&+0`8sNac=4-1qXgy-+~Phk<^6#M@0k6wp?DU^g1NTYJ7)gzBmpD08Z
z`toh@*UwzBoqxQbO0$%OMttHA1IwXzhT;F&8TP6(I7Yjre}7G{%O=(5aroy$L5n$e
zv!^X0&quOTK`^#3yiW0PVzTL^HHYd;ULk&w_D^TkGaZtv_9)iDQHp-MODi(TI-b@y
zvWYP=Vu=QZ>pm_OY0<RmjbGmsnN6D3m2nr;lR=fw2I+V7Gba;AzpFnuNt4k`n{T>N
z`94;dp4syou$kBhl$W^)!ngD~$Xy_lzB5exW=z)n?>#a_RP0CsHOXA03w{@#bG-Ic
zIx{4H#@7wE_|4|%5CthGns_eZ6V7wGx%uLfxbXSA8th8W`Ly}}*bS85rG#FSJ#VHI
z{_2@GU0c~|;Y!J*RH)#aNxl$uv~cbsn?~{h2Uf6Eblfs0=xDXWtT@EHB~omOBazNz
z>FsScid`u+WQ~!8-XPc>uj}AX9$|e#85YhUFY7Nkjp@0{|KLp36@eJ#oa3HjlT(mz
zg(s>~@IA5tI;LMmhld-##D0D7KDyZnMiWjPS$+cuW=ulOK!9s6UFo07lo9#oZ7s*|
zqj!+$#QONL2m=?flgfi7$3*rhba2deJWX#t;OdTG%cgIocb8%;QUd=QPxzvwHVA)$
zmAz-^eu~C3Z!0V)_3Q(=X!wWM1}3&LFw30l;v3YFKK(P!svwO@asx!oUcHcxatY%p
zcZk;P8^!d)Fr`IBsP?wkr{=9{5~_NN{EG8RVr>47u7I;7w1Ld!=I*T=bJKc*vVGzp
z;EqA|twI@QnhQRbT&@BJ5GPBo@j__;rg1?fIiq>#zUaYwJd2yGEt;WNguL=i^_6y=
z`@PwOT?srh7z5mVy}qi1-58C*=_jy-Ybh&jJ4H_|RCR-|p<LQ{27QNS2$J_|rPO0d
z6AoDfFX)r-6oe64pHZjyF6gMI{qf!7lxuPP%auaj6O!DM)QT9Pq%Tc-R)*?v!Dwu#
z)hdP!OatBIu&+)G-z`O})KP%o9?e<k`AeG?V4h_xn9FXv4AES98x1jJ)}#byeBw<I
z0^VA<_hymT{0_P_ys|^tQ?;%-p7AE-C!s>ep7NF~v(8adRI-=&{OIf9l1-M(r>q-O
z2=&w-&t?{WTXApTp9*sgl}u##vpahQ;`XRyQ6$gQs6*jR1KmXq0V@2xSr6L*W(^~I
zkGwjVP;Z)%ReEL%K|4!6OkI>L6p+V}{R@QXt50(T3j+1w(njOx)}^@r9mFFRR_~}0
zl|mUo{h85C%5t14viQ@dlWg-H<nL2*!uN9Tjhfh(@1HZd@34)v$*va3e0UXq3^UY8
zOQCfb@8TH?eI9AG(fV1JanEDyvt_+SrYoSw%f3JwVMg+9!YKLWG0gFys#(npH^=p~
zBk&@#f(Nb82@em1>v7jmKC$Sx4ik@x;&9bfa%`4$#pPl(F;D4{tQ4&NNJd*ld~PoB
zL<#p+q@M8(NRvwmGCmS?H}Lg`Yr+4qCt}&}VPQh%=eeVjey=U_^11NFLUVW~04q`o
z_u_VoTU4P&9~iOXeb{w#-}Fhh=2m6tlNcM#nJew2CTaR)Le8jMAP_}a1%Zt^f<>Q9
zk7dT=T|Nk@DB|a5l}Y?mvgofq1!rzW{(BVd)$(I8`g2>6kfy1#+a~=gQ0w3scvXI=
zhbK06Yxt=B?K00fJojJh!T8r2W=-8TCLGQ3q<qRgBR=O^M*bV@6I{?9Bi;`aGeq<T
zP`>|p7LUMO{7c!Z4G$xqzD-W-c*}nGqaTrJZKo!oy<&sC?r7bwsZ++qRuk<9#~K3X
zcB1z0qH8G?pEl6k7vWJA>B<xAlGHw(wdR<JhirTbg;wx^?5X7}9JT!7oV1E$?3oU9
zu*hdTpdNe>!O_k%g!lEAQEY=dA8qT};g86_3mHXbvT)1o+9iJH9}f>P+%2azEm9mW
z$}1hBqLNrHJ6tn#jUl6K!qnjy<eb!S#D@uiPyGU0#y$RJHY_uOjV$nB<fVy5!@iry
zY}rA6sHjWt1_~CPa(?0NCDG3c8d|g49EazgCkxm7HZgj+@OC`<r3^)QPc3CbfDWpO
zqDXd54*PSCw%y{SM8-}i+UwZVoC$vHoUD@?OWgMCuK7@B)O_oE;(6c^eb#}ly51A9
z79ITN<JMUQg)qtg<q9w1>%PtwzwE0lcZ1z098nuXzYKV$I_j&o8bx<{*R7nRW*w3c
zp}rP9w)*|})h1N?REp2u@pm#B%W5Z81|Fi1o45?b>BW&g8=DsEJk-BGiJ@52Xoi)P
ztNK#$#@S{f<$Je}q=>F%1Jf?+%sDO_2I<g}tV`X>x?Du)*9(l?ry+PgR9w#`dUq=g
z&4<HQ!A+tjev#0*(kM_0i|j%D#rvqqM)5x6;n}7_(*Bzd!lK7F0%%Sox|7zxf)S6U
zg5RWpz69x9|23ItA4;-6*{i*pgj&ncvTwaAlV}VUTI>F23ED*7ShOj~XoluhE*D!x
zb63@>yoPxC7VveRn`(%^K+f6NYi2Zt?SUEA7c|nYuCDnvEY?{QiKGk<dg!Ucd)+;K
zN92GP?S>w;9#S~6OH~tcB$l?B8J}}(&QASF0uUPHZr|+3pHaIvDxBTT(NE~qnz)_8
zgWv}HXR9pl?1iOhO7vQ&gjLe{%o26T0ra#;4?KRGfdx%97-{&rLmooYta-Uq=c9_R
z-TCRk57tc+UJG$G{Qx!*`5~S4v1j3ilD8w1XNj*mAAFUDK!z2U$o&Mj-|cZn7@~95
zIr7Yd=u=7zvsc2oq;bk(e39i>;-_@@D)*R*R5fU*n`**hcBx!!>Ve>>dfv*^a(N>L
zIY*JCdWEt^=vWd&zp>P&=|d#sQ;sj0T7&m$MXhU|ZRtWEcay^LE~X^~Ei^JQ@)~k<
z$+YFwAC^GC_5zpr{{3{F{T@wwhXy_)(sCjSJwYMRr4$1e)i;^1^qNlxD_18C#`e!l
zc5kTSdv9h8Tm{*H@MJhPH}(l?=f=J2xRk<KbBs6(l|injXXI1XXxO4GZ6}mHCG&oI
zKk44G2B(`MRwk}X3(~K|>ixIy)T_C2!~RSBkV(E!=JQ7PS)*5CB_gdHpF|F0XJ6{P
zCe5TzTf#%9ifCgooj#yTNt9U1B6<<ugpQgXn%@Y@K);xXMRRQhRVE@?Hy-N+*%$w=
z?zVLPoz&C)2MzOHCKA7Mc-%lU2j)1900|zgC%kM#4yz{F`(coxj8k!-w@!Nvgrgmr
z^(9DU2p4}9U;OWRG|wljSR!)l1DI#MNl_U&4a`nYi+()OR}HH44#>)uPw2|vu2NVk
zQxeH&1d(vmol*E+xE^KLIbr`$^65v>Ry{+oJ3gI$=c-v81;^{(Q^8-jBu+?z@pLEH
zxIG!=8^9r^Qz_Z~+nJF_{QTnk8XE?@_&^9cXlwW1pe<osTI3!0Dq+CrD+L>v14kqC
zTd82-9|`jbLJ#}LyvD9n>S&Ht@(FQ#ojDxS=UG)lC7E)1KPxzijNp3Xa6WmwFL5ZQ
z=of$(E^<Psfj%&=H7l8VwGMZImy^_G=JxFc=f=naCQ~%Z%QRoWM9wfQB~Dg?+{R=!
zEZE+IJzG(etP36yOyn4zIL~eBwlSXU1|US0O{d2oY|cP#674Ct<v=ft5A1<zg?;$E
zb(GjMzWE}Wp|2!$6rNw*k6<nGq`R;Odo!#)^RHJzZdaw1zEEOX7LBOOdXf<7Du@2G
zM-k)*$Om3$tzF#hGv)p}6g0<0TX`Cuqqn_R2hE4aP9MR8>ps4oaYUFAg@?r%?R_$D
zfFr$JxO6-Hwxih5mF$HaAVoo=>L1G&8i5nPTq*q_e<sTKfxMy;Lc|McG{svPrK(^q
zM`!=BU4vaDov0cCwKCZ@wQOQa7=#IVY0H2+N^f^NBR^SZv$a@}R^#LO<yl0Az&KeP
ziC~FsO5L{8CxmF3ntR~;a^lz7;|F5nT9<65jz7i}Gc5Z3@j8Gi(Dsx~+ku**Sm&OR
zugU5{!yr+U6yp&Gn`L^8Ua@d<+EuEAzHfTwc)f`hjg8%ad{;Tm{;kvfmd@6V<14gO
z`BuK#w9BkvA^2j9m3M%$gn`UAMc6GC#M&5{%n1F<$n0O{)Vd=*bXp<_w-Aii9BYL6
zx@{B?x+{pJ<+=)KI}};7?^pBoR?gCGWxlfuj6K+|r)*TCi#LYPuydx#e;-fROXeJ;
z4Mlr``0=T(NKc%cZU$yw7kJx>ZhwD^Gh(lc4aEiss_Irb>;h$reQh8ls-puw=4IW7
z87M1O6<?|gD>bGaW#Jx+0H_i?hCXA*TgZ(;K8s7G%}bepu9lM1qJaWF`a0YFX1!(6
zsj4noSb8>aaVumC{NWZx&TY!$q-?BYJr;ciJIrXsWz`eqX01?6mJN@Q??<Nm1D9T*
zeM6jovJP0tL?3DP^s}<cxDw+B*K|#5%0j@dGJgz2Tu5?IDb%_09<bPEQD&cjzGU)<
z<ne{6QCHa?KMr1?(p``7snhsBK|-5Ek>L<tB&b(&zoMZ`^EvI9{a&Ynh3H|6zc)PS
zY?FyQ7_1W9-_JN{g_^smx{BR5${(ql1}57KFrDgq54yCPtGP7m39v=<88J7Uy@lbr
z=Ojhekd!KX6pW}}7(*bMB~H#&8qUh+=0+g}suVp@+5C2LHvGSy?fA5IZr-`|4frU+
z7$W*0eH~jp7>qBm-p3tkS>e*CquTgYy3Pdn)aHC)B6!g1JB6X#m;c>R4Ai-ltEV>D
z(;2ite4+JEEXZxUmi)pXH4A~o*0Axcq1EDi7|lp_%A&#mfmZ%o2#70F<$J2pZ?Pga
znDsQF@_u?os`j+Ki;^9<9a3yBE+=R&xpX3!D{c}(m1u67d>5DOQ}qz|h05TCgHmdE
zKe9BOj<~~NuJ$=Y3CqS*Wq}Qo$iKp*DR6k%dKFYnGiFo0?&ltYJ<Gh2U3hgrNeJW_
zkcSvv6&^Hb??+CctJrd{@HOmReKCq3b9^na(TTTk^Tuic1W35TmTr8tbveVuOjqz*
z;hBU^pLrb0MK&AsT<b@6=v`PuXghPHyAmpK#hmwJsb?;mRJ!r^+f|0$mH>H3R5DS)
z);@g3`ehq^x*+w$0%;4x-oA!W`6fVlsEAXr?TrjO7559%Tj#`dI1KvC_-Y*e!&s@b
zHQC*_;on>-T(qjC6A4zm24Du-CmbL;gg4J;Tfn@MD5lgYrUE_PO`%V>jH#_sX$#d2
z?WIbY2gkVCQbe{!O8!ij&MuR`gFJbqla7i^VX+>?N$nbMiF*OrXE#q*$zj0`d-0zr
zjorybJFdSb6SXe~Xysk4CZZC|o5&B8<r|HHzy?swPI8GzC>-#hpMwwggKD$1BQYD7
z2^G_hEqRNn6kBCT8-Y6Cr~Fz!zLlL}bIFGr7itLNeQrLW6`)?+@3NcLDVp;3%>|;`
zfa9Igxah=j#X*XNrOfn$Df_mOJ;v;og>R%y?1YEop!o=^|0pqb%5Wubyx2T4mT(6?
z$g*SH<*yq&uHY(nHjA~-#;G;0l{SnPr2A=C+suTrig$na;z&7Rk!r~ZZgg6@yY})B
zM#62JRN)}Qd2x5EI>uiCu<S%+4lClcH^tgkJhKow&rgka;C?V!F9^@+j)&lNl7|@N
z)kyIQrq*-@eJOisK~pf(<{<biiQbNLQUn=LEiaRZ5kENMzk6fLQ*o00>nuaMF?F8q
zEr~2iNudD$CsGqG%Aa4w#!@<>@3J4vpgUR_@n;t7T3|b0!0ZaaTYe=q97cJ;{DvY`
zhEn|qg!PxVpw3G9#CtpI*h|1}XsXH6%$TYryaK#+I89@wNx80}hZw}j>4(ow6vtlh
zc0-R?8tDFfxk=q&;pM9wqvdIEXdmPQ>;IlC>sw#3v+UyW+jHy|Ri`|$)4g)FiJY5S
zjCST4ni;!01xoxtl*wF1CKo{mU~3#IKcBaT1xOUiepT=DPc2$++4l9-G0J>SuRxK0
z<=yXVgV4ZaKq1{#kJ4T-TZqmK8#CFkcJd;&n{Zi6R-z2aSnqrC*-Z||Y~MJ)R84Q%
z9!Yhso?eUA8>f*Wc%0~P%zW>ak4Ae$tQnp9QBh>m%CPe%h`o&>!@!KR#~0(TD~+NY
z4}W(NJ7_m}-!t1HY1AlnI;N6*y3)M15z%vX_fket24#EDNRS3f&rr-CGjTC_&IiVr
zty3I?o+ykxjfgwF`~&I!_vn1{T+fJS69<>YE+{6erUmy~+B`b;bis|g1!a*s&tds_
zd<=n;bKk_6J|^kuaRZ1mjGs^Bxb%oPs4AXut%rc4{a@-vx_yTI(Kg<rJX`*#%q}-H
zK&ne1821#&)D)b<JI{NnxrZIbqDdfhon?9Mc`Jb)YA!9Q;&15tOAfeV{Oe}jE_;j{
z&!<2bGpa>cTCUhqhm=b&q9;v|b-=TTSlGeZ5Z)p~Zb@CNKSWXS^_qtoTwt%IJp=%I
z9=eIHw1g?2RCNLY7E=})>fk$+MCRF0oyR!`i8X65O2||VNS>t}<3~$(FScwxu~H={
zB2pMhe$f$~uHifPJPzGq%_qI#;c$?Qr?0FVPCY6+7jdGkNGO1-kpfUFE)_MW!pUc8
zK5_aTN?=GiOcZfj3tm0>MR5d#5g9qIo>rLtjUMN6YgpoEYVwoh14|oD-hDKp^_du3
z1>OTX7t+MLbyIU++s}O4#x;Mzne|vxWtWUaXy3SeZbd>-ujctP*_s!~1TrO}DYo9x
z=d0y4sLtRT0=DvUMgOFQNYWBly)U>pJochOEx?p5z#4H3=;NC)f{Iy}#cw0KdcO_X
zamDIgV?2XOtv5twbab&O1gFHtgW4(@*v?-;x{9r<vQ>-N(zs2H&ZoIe+*3oVM5Hp$
zMo6v7Fi`^=1LW0~*WSp#qXMeLsR^OVLOvQ!q^R$$T{gnhX5E%SEy$0IYSHiMVuj`?
z7(z*&zeRQYU2-vxMg@QmwxMj&tRTHQJYm`%)#%x}oe*%t0?*n};u3Z49$7!>PAJ3X
z%)SjFJ-(A!LLxd~;C69L7S97=(5o@6Z>?n+QzX3?NjB`_+reDJ&-g|Q0TFuUW);H<
zd=-$%MC#+yAz<F~TdM3Ni>Rj(MeP7K_sq26ie4LWxg~%8n90I)+eJSoqLDePMK+H^
zY7yrx#d0c5)OdnjN&j<uOMHHV^MRfsUinKg!ia1^3iad<aw98(q{A>Vxgi;)naIkN
z*iT<)L~R8(IY@^<qC(*vWAm?D?@}P1f7YqFwpj-F0!h5|t>#kb5zIoEjM*&<YxPUc
zEyUIbj@2ghp?O?y>F5C84tY5L9L;Q?dUSELQ$Wx9-F?q8K@9$n(RjjNySI-~oDq+q
zOX#jHN16Hs<_r$5Q5d~UFI))zi^N0Dvz#;+@bTrKc~1=SjOaf$rE?cynPn{7P39#$
z58{6Rsmievufp4};TivNQ=nymk14DODuF~8;5Sz|f!@Yba`C+<_+F2!vi-i;J5eI}
z?_o%&Hot9jM1b0ULm`^R^<Pq>nM-A-P?3wxB*vRBM_5M(pcF>&O_%aLkz4wFE4ewN
zriY}#K<<GpR{`wL`C}J&29!?E$L1=xk@dB=4NSrLxN5U+wdV`*x3%M(#SOdw#z7^i
zG<@J{4ZgVgjxpJTbSU`+S~&79UzR>yCGDuG^6R6;0?TH_NNv;;s~cSFGIdbAZltIe
z@OFap4=Wx^@R#@!_S+VBYibM)+8GbXjPcNJmY?NSl~_*H^A>M=Q%S`pKAIEtF|8UF
zh7a}&P}?h`VveBbKY^dHV0_qDW^7l4kiprc+lIkXkg9Q#f=!?xp?tW}-!=*ff<6Lz
zdq>r^r)CYdG<B5XyBtW^uL2F%RP{AY<IO*HEyU+43=J{Iq3+RNt|k9p5*PHRqnYp1
zp8h&wqng)mJK0E+4jC}{f<oJ=^&QY;8d7UtE;K|P^X(32wwlt5up%u`dF1R^X?Ty#
zU)|)vK(R_iNhRF|%Tu(F=JQU~w+A%;L8HPq!h&_^Hhn7PEwvS0%Nt-LrN2poW;?UI
z^!P=nKE=1$BGPB7xKZokmK47U3udv)a{R^oJKWK<Cxr+@LRTT#PlvZPkpA<2E6}i;
z75BKD)uT}^iMZaI>VNho;DH?7wvX_(Sm?>dh~Ke){MN`oi)q}$$<|24eA5(P*v|WL
z<KavIR3ThRx~)V0BG&M3YFd6dhXr|Pdn}cQOvX)^1<XKNmzDLv4)kvezSJT@HczX6
zomw1`o{4Z`Z-u)5QuzYtJ_ppnmIzP8B|wGA7Z@Z3=3+EovXe&mq6*Qlt*GC{JV&At
z^%DUr0<AISwyLu-oSlJxFwjw-1pu1GX%oQpfwjZ~Pw6xV9)tIjxLoo>YCtoJH-+Nt
z4F;RCHsaox`O{l4(+26&F4I9LU(BPwh_w$xBQL1a#J_M;3(wn6%vrdIMfeR)9mYKv
zM8C(W9_-5Bk^(O-f5fw_^G&JBJ7^Ls+L{h4zcBu?@1X;4F$7ISZEJ%n7EtxlL;Lbd
ze|bT3Wu8AhZ-;i_yR8%rRb9qW5)HTG?u?2S6mk~PRBl;1?n*LLq~*K^CL{QMVdM=m
zSv5Jv{POM`NU47r@5m%2JNlvobz0suqa357(nIOQieVyL&4$%n+={%QH&&iMSrNWJ
zdA3Chf@S^fJss^D>W)BAeu<Ox9pypl+{ph*#_N3iil9B?Kr-oPwLX6JX4&U!T<?rq
zrJWI-CRpvV6s+CW)-2_dRe}QD0Y0tXULQfK_}{jGF74(TrH8eR7D*0t9OGm(M(Kf?
zvNJ!qEuN>1dvkJrf%oYBbjzq%?o*$~bRbJ?YD}1e8NeTu{a@3OD;dk+wCD4}j%>Cq
z_-T11qhmwBn6Vtz_Sp@%5D{jXmqjD#W{J$v#L0X^RwP6t9gLZwBGSL{@k78lS2Ds{
zZDvJ3jizI_!<6i>sYI7<l)nvcdTO|B$3<=XeEY7H`Tp2T_~uNIy!YZtck*#;e1*Eb
zFDLYTf0muAhWm8YtLU163qO*`$k_Cv;=?^tgtIdsz*`IN>5$P5hhK~tud{QGq^hN(
zvYLdch}|>33zs&kD%$Nq%=+Kzkn8Gcs*#j1o?lAva)9PmIs|?h9XxpbafpCMc}Vt5
zS>MWXG%VJfw{o_wG1DpVvm$LEE&+8~Z?^60BkTA*{mD|B^(swq&~x>U9(2@8Z<XKA
znZ`|{0}m_McT@w`ut=%=B6X#q@n){n=V|#*AoKEjo>)xF;xJ;ln4Zk{!G7U%+a6LC
z?av_THm~IAU&<<)C?N8AjCIRKz1Xvjvp>f2->*P;FnC>CsiX2mY&v4mf=ERSJ6epg
zpb%b{u9{4>U=*3isvr-zaGF@!OK8u)XmV0=Xw;+yVX4oBxOEUQ>D;wWc`DX~0F62S
znY<djLrpb*tz;+R>kK0PgmMK^3a&gG4}Z(*zhR!^`0y?Br9rF~FOvnAN{v0RtRvR~
z@&<toQG%yT!0ct2SFRA7kx;|9Tca*U_;O-?Wew>H5Hk6UD`e*<0Gk_M4Y&GKqPJp*
zDZ*WdJC^~e867i9O^5g~n3-zUkFOu{Gd+b^4)$*0(+TZ+IcinVmF_$4BEWRLY-G8;
zIHum79Aa?&f<~QLkdIrm$^9cK{U}l-pQgAEgONCYmR@BVBb!;kKr593tZvN@3AjQL
z6g>P}xNl%qOk(ULeW;@<w?Q4dZfEPP7|+^z5Pf}Cj4^gBG1lG5t=veYEFjRxkN4||
zvT_5(_lY+4iD6uCVSaI$(Ccrq+Ox|H_2r_8>G&!NsYU)R?PX=rxOn_QV-K#Mq!bJk
zmLPS8UIj=VkZ1i=%hNnihI4_yhc2*jPk6+SOX+nM?d)Z$p8CYC;^ZaS#oL?v#km6$
z+qI=I0zKag6?SM@_iD^z&9+6h$N{M{Qoka*^-!EwY)l}ta_1UlIWG6G!d^Y2QYuwc
zQu|%CLc8U`(*KvHQmUB0?YpP#T&I8RC-P4A(3ti^WVYuPne%nwOf>fCs^dP4T1l=%
z8{@5iRP_qwrxH>wzv}AhCluk&Ns{L)8S?qV8i#EL3$WfCNwO9_^=zBO@vPqGkFbX`
zf{kCt;L6D3e%l?t`geNA+Mor0cnUkgl_2iiSx7DkIDG&NrGdjQ%#e>&j3PdFD>USB
zWZzwX`<?x&S6T99VgJJs1I()TGeq93+(g8CKbdPk>x`32Vp9$eAz%P~CbW1Kw*GT&
z3A!Or<hNn^JB;QFYA!Ee#b=vT7uojE0w_7Z@jlqG&<S!9#OB83vKwr-epjh6o62wR
zx}F9`=a8Det~P%NbL^@H7&xwW^F$24e{;LuMxbv5^m=7nPhR;p;q_Y9rZ>W{s#BjS
zFczbXF;xwV9C(p(xIk{Xpo0CWD1$p})nDrHRKh^f#QG_xy<9VS_m8_OCOMR&c1Y1&
zXXy~G11QZzqw!U--KGzakH*FgC!!0dB;aNv0@&(nF=`ms*h(q4>Klv){dr8?-o-qU
zk#fG7B4~ZG55FprSWIpM&3F#1Rd3ka4MR&lyJ{!I95&-EdkpCF6jc1J4Ft7n&I26O
zAU*YnnymT{G7xUd{Gb*pat?;?srQZq*E$dfl(n*?N1N~J_%}u(iS^CFtVG>z=Zp2g
z4*(MzH-AXEtZ|4p*4ij;<??X5I-K0`q)pd!vzISN>DjAFtNd@z%M%0{7KmvuGk~i&
zu!}+<$HXrmxx+`N@n4mMjSj~$g4*Mk3CIV|R-W-F0cLk~5L{B-I+MBW*H-Y(LIe1L
zB&dJGC5$v?U-TV*KO5zwCj8@hCQ11-9dplqE~LKj!H@HZTFs}O)_wlnxbjQwAID3L
zk%hHxiB00xehy5If3rn>3;Ke%vme&K#8^ENW_Q3CbaD(8KW`-Y+<DgHXMv4AoN4GX
z!<`L;WZVl@J#yelIxU~;&AJzc-ISFVXfiL&m)zg}c4BHZ9tT!d1Lw{vp6YK_Vi9q`
zH8@TS;GgGWho^4ts50@UMe=bcaV(IqH~XZ~jW5?aMx{EhQ~%|OLSGlM*?Nv0JU9^O
z+m8>cMfv8?GASVr?-9H7Y~No?6`qJ*rSi+`WS}y9b0(7_d#Nk{s>|;t<Yc(37uk#o
zcNX{?8$67Wv%2;jIiDN4%mG1Wea&v;&l+cCKN~&KuT6k(OY3K{R0c#|9$;oKA5Oo&
z@6P?)`TF5<Q+E>#Q&?)Sye~OA+W%fy?_%0?j`BNXs7c&H7SJrOUD?|Eg?#u3S%BfB
zb%Q+XM)C}mHnXT2-;)5qQm0g2o!{W_Z~M3sy8b2mvSCSD&wg`WjP-~bBZ6Ui)PgAf
zZ^I^c_T2Ks;A{01Sd1(Yd2Y{fmxU8Y%%X>hI1fH?N?6xN-bpt5m<)FtC4LF^E*7zO
zE}9e*m~<cU^?bk+_D`}a#?9~}30K`ks9bvak^L!r?uc!RPC3^DKL<~&w-ojZOK(kN
z#<+c!w$Y>Pv_<{((kZ~XPw^;R=Z$HYQ{X%m*;eKPblixzc@FbRYT*oZu}9BtVcFP|
zr8DEjfm3Kd{<rnPsQm*d7Rj`zBK|w8zWUm&SFhoTY{lf{u><etm96%NFUS6_6j^mf
zokXCZR+}Dizk&=P0V5xSUK-gS*{b&Rb$gk?4XK_o>yHBzJhEa2tRXTstgkyr8*%e>
z)wlZ@QWIvy<mW+Yk~_f7quU0uU|bWaxSusZK$VLNT{Nc?D5T!_L`gzCj$w#VWunS)
z_t$kO;v(@rK|079TT|meMSGs1<?D5?5^YDxi6}%<t4Y|Q#GV9&*JIWI)-|lPiF^$Y
zzYUqY#NmNCDnz^!N%=tr7mg`#yLE}X*!PaV`hoo-klp9<^Zdf-nS&SbfzHiQxPi%V
z)#U8bt)o*5Gm&Pk*C8Bu(`mSKyM*+)EycOExfc^*$ADUVFG!2@#<mc=m=To!6w1M|
zUR{-4b74CO=6VLU<JO-m#?W8InS#v=7?`mNRfD@y<L5MSU6tP#DKY7#5hQ5xO^YP$
zw0dQf!@j;KO?|o}vDLA&FbzJe_UA9ect?-$*Q~SUHW)6V_^wmXT*aPq<JtO3pP}>W
zgFHLfHt5s~%B7FpTcl<}^`Ki1iU^8(5R^aIknqn|o15?0xt<%j1cK0=TV=^h6(oPs
z!W=FPtuy>Sc;<3%Po&T0%`UA4pFmrm5RJf1|54TNfB6B9;OI^xSg*I@7ktWY%ulcP
zq~dSh8)v;!!Tad<pT6HL3KZWS2uTKWv>D5eB(X~W9Dw}3P8AS;syWf?b&?fMMlP8-
zZet75WvRlmUk&!Me`1Enu*-1)6j)=LrN*|{8h>lg>2r#|keiYyd0~chq@))9BtppD
z^6t|hDWg*Z*HN{>nZ^=U)|0UU>KWMeT4&*h(2dTDgSO+u*;Lcq8ry-3ecjJuKc20=
zS$7BErCBC!edzz;l=3PI4;sAJfP-;_bWcrv(v{N<z3fOG!L~OVJ#(WrcA;pG%M7+2
zmtixOdlnz)3<&JocANn2VC_7<Fq3?i^`eL(RrWOJF%X=Q9CwB3V9Pjrs_v|pU7~q1
z|F<fuekY0jt4`dAooIpPDMBaP1cQSXe85qjhhB#DdKp_!y8aaPjuF@X(sBiauCb_B
zYF&s-{8_je_L2yG59#ZvCh9llL44#)T~evU+XqhDK0$xVT~_&bv`rUo#u5zH7LRMV
zxY)^eN5f_V(mfMb$pC@ET}wVVW}R=>)=H_xl#b4e1hFiX`cuLkSl<dtm6av6PDQ5Y
zJ!AiI^+(#fcEQ7GAc9Ov_P$Za&II<E6H^+#r7+t!R5m}SIWk&QR5PJ6FqP}NBw~8s
z5vTjzrct1%w)5=*%fNu5ec`m=w&^rcrs9HxLq^uTJ}q1@E3vhe+;;DeAp9rrnev(H
z6|0*4Vw1a$q9IDh|Gj%;RiUJ(4y~-DlDqa7)haayN)x3*J2~PGJdZqPEsDW!tMXE{
z!2=PjcE$Y-tN>H~rA&GgnBhRG)F|l?YqCIIYUko>d6r;hCCX&i=o?+X;y-zoYnRJL
zADG&P_hzYgrh9mp-=~1)lIWR_r!0P6`P<n@B3D>u_vE+hx26nv$MhZDj1oqXJ@i?E
zc%5FO79;8;Y#<N!j~f;>5-{v<MU1YAmUWMpD@c8NKG@%0w~lAi(9LA~%C0MbL3v_e
zj4Q|Zg8ny4+oJS}t(NSpd3asrj^enwBZ+z0$C{yf`T&=vj_*$?+Yk*GO<z3ctO=U$
zwaci?FEiY7D6yYCmwC7M0HGRv2}^h93t7aA{Bq0((y>wEe4HG04r#C<y4r2Cl0^8T
zjy)zwgH(4gBUT{vqnG0Z_kj_wm}viS>DPkZ4y8#e(M|C8os9w2uw~`})fEm_SM=lu
zEzjPk^+S+ET=S?S${${`@Jw0_>xI31FWtXNohe9U3;dwN$i$cWNZhGmD3CjQIk5On
z4BYapy%Haq8D%&;gH100YW!q3asBJe*yG&6YT<KWH6H!k`<lPtie%yXjwS<LwKPF?
zW1(yBvyMx5=$bX&z6Tn5$gc^0_?bJs{o6NQZwR*dIEp&tljjB}TLx3tYbkoH)@O&D
zM44lg|1zT33Du+1o^!#h?MQtk6qxsdkE8IhUNqaz7ZxiWg`hJSw&Pn>t-Iwca_I;P
zqe|J)oACMfcs<PJ`rDL&>(@!mZi&@S7H&4F8hxod^cJZLa=Pc_zTk#G>dQfcZw8cu
zw7zdg05dQ3-F2X}^5R0AKBIEX?6%eE5sup}K&WNs4J?l8f+%?l>7C`izIpxjHp$4k
z=Q*WdCqB8TFNVO61N2GD57U5klMI%qsylJ8O_AWdhO0cceV+IkouJiKzkD-zVnz9w
zBOz1WMm?!xY{U=cezFnWpG7tQ%szQOC-12N$o{l@Rg0DIRFIZRlHR=t$BcaZigk1f
zS_~x;ep@ZBLBbX$o%~v%sBzm-y~tha4Wc99OfZ|~R0hpVs2&S&?E*syT_IgD%T$Ew
zFD^tTcbTbNnwF+J|G_FM9HK>{D`rSu8<o5&Z2Ueix~44gyME&{VpZ<{P5#y|9GvJP
zBC*7M{VMEIE}TTWRo%M|PtwKP#+M(O*M{>u45Bb=6Bw)K0|O!qk#|obd{(Blk@JU=
z{IPCvSE!?R!du<;i!lN_Z?cW3O?@8*uK=^4l<}ISt>J>^eJ?p0BTp^A{vwR-F0c;S
z_@eE~GBmA%QX`zqan<3)<gz+w$FPjhGw!dwI|D~Mdr54q97HK2V=7?Me2;`pZ%4NL
z+zEz_qG-J5mDN)c3dC=8+yC8pr^}ZfN42_tbhNn&$d9M;^3Nw4+v*3-?<UO6^=1J~
zoHX>d?2${q)xuKcn}&Q1J&wYIeO+-1-C(wJ&rCmK^Z)J^UV=TF(4ZN64i8F;g`X%x
zn!V`%5d*q9e#t)nHGoqYe$`jEf`~a;T5Wp5Qe>`K;~TohG$g2{!C^SHU85~O_|;%C
zl|fF}I<p!|-IQt+`!F6?2^3ECW??eIe_d)H-thQ^m0&Ch*|qp(^vW-;yVwg)pE`DQ
z0~xBxw*}N(#yEp;d6jClRWsau#GJw;?6zWTY<R(<l=+esB&3@M4VY`UTRGKTm_h!9
zg40s_ecqM@W(|_%I~<r5_SSphRCnuB%~f$YBdXs^AFZXWjci(;;*6g6v42_m6|pw&
zKn@bKlXz7i=;eO9G!J)kkDm88{(3{ebEOV_2Z`vg9L5CAc}~=2klFpvWEE7mVhT~-
z<8+`@We@vYbaQ~|tmI%e-ZKS7PVvP&+X+tnBNk6aXU*`%No~g-ud&{ZEPJ)BvuQwo
zV*c_;Q~!(<BI;wQ|0v?oH-`nV)wyLKwC?LGH}4rKVlhu*VdNO$Ri5%gAZy}t{W8iA
z&xBJ*(|EdgVwI0KFeM6?<404m;10g|-V^=IxUuHpv-)cxXm+)A6#9V#7N~b@q*bU@
zyH(bJ8<%rB>pRM`TdQyG5|(gw47YEV<isD+l%fX`^e<)MDfJ(kx-T~)SEs|imZTdW
zEz9nxm9KPk)X1uoXRla=AsK`}L=-Kr&6*`D&>qA>2jdqQ<;eV3-bf0tdx-}v1~h`k
z%JIBwA~A3POo$RvR=F8}som4ogyM`TYZqc^tL`Pb+OPgX-eFu`%fK&kYbV2PiO6Eh
ztg1SiaJ==lu1r-r=OU`^w`-p}Fl$%iPCiHOE;bR~-)*CM0@rw{CQ<;2Nqk_Wh?7$C
zKFH7>;?7?<{r+Wu>zw`gT=_p;=2vwr5q$==F>IQB-(JiP7Zl%G7>N*tJ5_h{!sGsp
zfSN_|RQ>8n`8aTni|Ge&yWIx?4C0PWNUukoORkBA){rp9OE-&(5RSJc7KV$n^<|}(
z>d`lG$8$eu;IXs0^K*?X?SVtHT|0i>19>NBy;}i217F#jtn%J$06+laT|YNn%_(aW
z&LpfP$8$m9o#n{yF=PD{B+OWQ!>?X2Y=-^vW?4w76=jj);+G}O&v-nJzRs19pAJR!
z%hX3&tq`q>`R(k^OrMC}8An+0{~+vQxKlW-{vlVAl=31ojLYNUPkf-%ytj{M>~ueT
zTD-*9j3Y{)AVE2$xZ12ijAzT#NJdC;qPC85HQKztaCP3-^K^UU+e-D0hySNc(gAkX
zK3@yE96P=}X`~v<RVKe+MZCP*p{HO<%vWqWZ9UFMOr(#i_G^QO<9CchOvw$YUEyZs
z4S>#=Z!LBeExf7RT^!vt_eeg6t@C)J)Jx~&t(Z=5LMREp_v#&+m9es%(UGq199NF=
z538f3SEpm{WsHv;%QV&JiMAT-?Ht%6;fR()i``wfr}+4ana{b1^8w{Wj4U?k4&qC0
zh&d8sT!djpL4}KlpijtGXun8~Ys8L6`*57ek{`@Q^AF2&S6!P~KEhz-KXm6C4MZ&f
zqXJm~5hnLZJgoQ|-Gqm+aXXZn;Q+;sch)e%i+HkRRBrT-u=^+i-~p(QEy1d;COW*!
zx!3b9WYyp+#P92`X9^L%bsm3h6q;nNlx3`UZSC&!XtVTBLiF~!{eDQ4QdeCTW{m?-
zrYr~b+?9rc484_3qIl=vxFXHL)op>h3`19m95oQ_>7_JXOy%45-S?OdTO2<k_-pg9
zb#WE2vtMeyil+$5_5teFzv@U(#<a%|CF~UqEB26SN`G*2e6C}i1LjkL&U)dlj$Q|!
zps$h&{1~5b2bp)I8~I5;x{oiDYE$R?ZseWGp5@;ltW1YGzFJR<xCH@UZW0*3+q;Lb
zC<VEjH817K0{-AgH5QZ7OU)mh((&PAZuy$gFo&COd(hY`x_lm0Zv~LRQ9J`$^M!vu
zm6KN!;`LOr-E+R&^$fawV*ItDFQ@B3WQ8Wq!Bz;Q`V=v*()=u?7GKcQ>#Af~AR_Ju
zr!45kKFD;b&GX5MN$CPnnrPiOiK=NT0SVc}uRifuU@%1U^#hVaAezv14zM&&7@+M^
zs`Hw|s7<>(Dvq!-UFY-9W8H@8DL>p|y^cyZl{%%To=pL@flBa}iPn#2GOivkcfZgk
zotX(xZMt%CeEe<1XNSP+c2Fw2BOIiA4aC*m`#xZ`61BH9ZTEz#{crSd<BgJjVAt0q
zbuaj~*iKJalLv^8*0w6Wx4noiS*s$+?tiE{%b>QpFWk3Kpv5UJMGF*n30{g7D6YZX
zEx{d%YjF#qxVuAfx8UyXP@uTo{Acc+dGFUVlMiRlK09ka>$jdq*Hop9Y{Co6iY^+_
z=@z25&b8Oe9TPR+rPA2rwYgYZSoCUAfjRHK&%h|p0QT}4Kv9JL9e)`TrCNzuSzF^8
z#RU(Ur+Ad!4D#SA{?Y3Bl9Vpl@V2_7-B0j;;v4J#yRoZF=P%;GKKM{9o_k&L$E%gu
zw~E=ET4O2*{vO43{Xx~h4b;ObE(jx_jMO-zDpkDQ;Kpr!^1N@|o2f54GUlAOjP_}C
z<m3r0)jzJcD*G|iF!<xqsr4x5m~hAB+Cx*#x^3VgOS+Nmm~ZN(m1y{s{!~hP59|Xz
ztqR1cmp=cnHg7L|s%Y9~5j>aPT14_&HF*n=&QcX^TbXHW`n^?OwYVI!OiiC;pG=a;
zqpYEVlZDZQ=U-?Kxm5P36t*_+IB74==xC)v6H$4X63s$E9gDag*YJ$_kl?tHU#Cay
z-@Qbly|}zSvSp+{p6tPus*vu!CAR68^yk}eWEu9S?iRbB_26&sxp!3z42r>q)boz_
zPftvZ>y;i_>#<Q3*!_`;NjcZPeH2xTneXm9ryF27TJ*0B-f_N1)Ns(B*W@}nrq>sG
z(1>Zwd}X!pcSzZur!-M%&evT&T$Z6*wjCVtTD9LogH_T!^4K8Sg>0u!*Z!TI)xbzj
znjEerIc^@(vDw5xI!Ew{dbchsIbZCo@E#VnC-;382jtB0tL9f>@f^h3rG167hZ^2^
z+#iR6eVTZ&xd}s@Oj*<9-+qYQb>})6c6qhO4n2KngZn^~dON^=Vu^=MoKv^z4LDzF
ztkDKz!=toe?%3e2`M-~4Lx%BEm#Np{W_AF}WLsHZzs#qW3qhC5y>noAL=*mDn^HDb
z40+@1(8v#4=Sv#Bm&XxK9qF_Y8`1dPl2{kPlnh1fH<qg~Cqb0aBM&{^4M{qgT6X8L
zc548dB-0|7+MDrg_Zpb1tJ-1%C+mo-))dvkD7dWy*|6cG%{qc>y_LkI%2yQWlN_85
z!H&Pj{5UzWXz<*STNlwiKtg<*EJG4VQA68zHDpR1%J$K6sirL9<tGK*fH{_j18Bu1
zpdD+p+I$fxD2Umb`Kbd*MFF|YQoB}%Q48Uitf<1Q_XqL|_Oj>wZQnVsfQNF`O$&Qw
zdRyJW4&Mle?$TU2UK4YdwntswNMXAaJeONOdEIoP>toT<`%K%l>Zc}iN#LS`S^P}H
z54#R~ZGmMkovWn&w^B*X*WQa-Z|P=sAcV`k-^Fxw-}aSIlv<Ly*URr%7k!zD;!em7
zzAanTtM?16-9t%F7;AO=bwz!goLSgQ1h)q7znX<%mvObilj}>FrWm;DIZf3CO_Hxm
z9~$EV=Jxd@u31vH-L_k!|F1IcOZcT^lwmt}!UcL<hyq>F<sduH?!2BBo^*3wttaM#
zanK&)xsb)=C8mZ92^eE=rAOw=FlFGa&K9)sx3Vr394y7*uQ_&>qc+UQ_yd|VO6O*X
zyQTuay~ne_QOr6)_uILfvk^G|P{o)Zt=R00KYi-8#MSQWoAF#FkWTu|;fk+2{Zfx&
z0$R;D)W^BFz{T<IVPW!74BNu442Qc{(rA=nRwm9z<#9%ZMp16q(X0KTb<dEAsH^U-
z#F>VjtB~Z>>G;ZWDEYdIe?7QoHb7DInaGdqOHZ$X&keW$fD@Q|1udPJra5tRSN;+%
zLz^Hy2_fpoeO!eq2o)!5++mA-Lb15bDCj-(-r8Md`uO1kOyy(AMb}S{Hph70kp7bR
z^@_>sw@5XN9G(v_Fc{o-Jqtbbj!h9K6u^A65qUWnke}{#B$sLaRSWWJH<JywN*`+8
z=ueAJ<TT)1Z2KOg11Ng>wyOn^#3!s&OYXb>w$oHz=|(QT+ibb~iorK?BgR!ifRlo9
zzCd&xNpZ8`pie&_8I6*!WPNYo@~dwc<VCU2>~D(H7;Y$N;jKZNEoqIM;c_vh3kZ&o
zG+mv$xBD~-(aC`V2vZZr%y%b;@=zr*8tvauB^vuXgM~8jp!yVdJI`2Yp<I9;WLGlE
zTup6Q&aE<p!VF{X^#3Mt^XSLY?u9nnwv-WB|CG*a1jY)r`40Qtro1k3a(Np<EZSRq
z`kbJtYKusbGdd`;&6E$m2cOFco_}8ZL`o3~8y*wn7Os0#RNj)^cwM@ilT(eS6)Hk6
z2>o8hQ~SfG4@aLbM%jP-5m=9(l27LZGvd*KRAfC*26@6C+g&a_6=qa2v4>Q`y5hJ-
z4=>bp{`PuEPgI>;qnzDcu1B`Me5&z2tn2trKO&WE(|?J8>2NOb*WGjf0!f#VhefTk
z^;)eg?E@_#jR&hNEL;U7jAbfpA)CDuC1H@&wR0haJ~xT=l}E+dd@e9-Ah@ESwZ`>b
zX?|1JQgS-nL$J_)B9eW6x*7{^)7y@9;`XUa(D<Ko^>&&NRZ@2<eJYn31gG+y!KS79
zC4>P?7a^^(>Fs9BkA*(nTNcc$R?&tZRip^2E@3jI{}P<H$SM1)k&&q-vVF<gEN=T*
zeW2o)Dmw4ORBE%wWuoZwz-#ZxfzuW|kL>Dx)QO*CUm_jt3c^rWj0`P$G(#NpW@~VH
zY_3ThW%IKF;fPaWJGCFi45T<1#@>?WAj!?uuj@NK(&<+wXrKP&5vyZaul+r76As@n
zb1YRL!+Tlj$<u?r?S!ov&ZQ6jz#BD0A42~UJlWx!VY4{BIyGXpChQA^sP?aDSBXyg
z);CtYt-@QGLydI39s=i#O#?Y2b)(m+&z@WAnQ$f_b@4+H4Z(rWj=A4&H_Xg3ucmV$
zs5pJxEmUOt!I3gW9zB5m(MG`HjW?lJ3oKpIxyw=jwkZ#SmT7$pF=TjTAe6nvCFtKs
z$G6Mx#DAl<2K<jeu#D~Gx?wHHMcD7OcYxf_p&HK+g)Ka<DfsOobN_I9|K!FiW;+}}
z+kZ94O?!4ePUaB1vc3j(b2Crh8!NQKIC!YHTEXbNx~p8q(!gl3Q(lLOPb;!q-ia*=
z6q6~md$y;X+YL?-3|jI0i|n#lW02D<i>~x>${<|l&bLB&@O(y*S=bZi{UkNmMgl(m
zHJrJfSxPIsc2~q0z8&8@W1h2W;f2W)czt7Anc-M|Ywy}JV;^eP98r>6M4US2sP%Iv
zntwSb7GUAZN+fn91O~5s+d14st(5(;ux1^WrKa%4N=UTDxy_32i@yUzEFwvDl<Rw~
z*%#(|-OGdf^tW~08aMe)AKQGEW_2g`<_UwHAWmDUh4Qe7^USw1n!>u*2eqWGt%y`o
z`s5Dfk!r)E;4OC&z#4h8m&?xC*&x3HY3aXevTw^=uVR2T<~Zh6KUUH0+l1|T?+xld
zvAa#r&m=}kC=MyJ5e1U*^N;sFJ5Bk|qJLO4cPZUR@|+pfedm2Vs5*E8S7jfQxxikN
z<qj-2<oFhLm8!ru!pE`<86E~U9i%eNke?<0d<JR@75UNREr_lpRLYKlQMwcr3*R*b
z-!Abm=hhJ4RGF((fS0K+GtURDcm9E8qpWzVI;)+vmpnNmbLo~{7EV$_(-IvfwKA29
z^S=$V!+yh|1CX*Y;ljBWRJ2We2N}D6Xenzw`ZjV~Fq+Hl*23Ok_39`?u>VR%*2>BC
zRqk--rgm5TR{D6DC&8drBbxq{lj#A+`h+|Ik?rncTaMMBF`zJ1z+=}cZ@6YACctx(
zy^-F!#y0cMoP1eB>Z!NYiW@%~nG&lN(o(e5NP97s$=hsHMyC3w5fQ;sCv0u0j75z0
zmL=*IS9>+YhBmi<KGJLx*`rCz6<{bfV(T-M^_?2m)JYWfFGw1l<cjrxE#NBf>g3^*
z=);{VRM;=}0Fl8$VY@Tcd@tU5?z`P*)&292&K^x6BM5qL5*&K8Y8B*Tz#-YlAyW_~
z^#{FyP3wW${dwo=+;f4m^Vq1KK`Vg-kTJZYnAp@>3VOPXxl-D7f3DkYeYFKKAAhJe
zSF>ZLGh#t=#ys!soDC(MFW~z`y=y$qJg?_m^ZaNOzXtlg=*z*R>C#5=3xjPIu6H0)
z)pGUA`ehpLj-DnT@uZX6^Hv2omMQ=%J7A7Bc{NLEaAp%}tgnP6a&d|=rmQ|shv3VB
z@sE6crdp7(Y@3AvV<sA1s1LXCjh<7G@f&SrG<h$Fj$2rqs+3&yi70)N;{2@qsptA)
z&FG%3p}S2*;NTv5Q{r~AFvAoqocF6k)_+6`CnC$F_)omFsFnJkn%jh(+B2{>?zyP?
zJH5uS>i2~I`aO%j+W1~hXMto9!3t3W0aqRy$Cdh*k?@3oj>n9wi>s{H5C?q{|7(Gx
zcFAbrgAk34_&g86bQ2Byq-D#PQ+nnT4j`jb0ZqT*F2nR3ziv`dGLK@9?6!u53m}?w
zd9R<pSxx6Oi*)Rfq~~FBv0QAifP(Mi{J@RB8x&iJ<DK~56Yt+(^1=icn(nyLNz<Tj
zv+jmIFbG$)w=WOM0#R#?K#P{m>eTzFncR0=!`!%kumhQ;LgBxdTczFa_8{^Yr)WCz
ztndzPc{k})EBg1)i{h7RPF~tVUsBk2RM>s6*9_ji0B(6*cVuH$sR;zvZ96Sne)Myo
zbFz93?O^$F^~akE!>)(*#yCrr31cyn;XqS&7q6YOn_g!Ur_O15s4t<%L*D-0-UmKD
z#k(QJm&DsOudI{-%~3#{I`KfbLwoafis&nk=;LzMRiob{T6=ndwAAe7(8I#QL(i@|
zeiZzE(K2{3cc_<FH{|?7xEekGW{&{mOIdeXFudm}hj+&_evJ&UzpY+4hT(Xd5(>Cx
z(@?r2RocA$)nGei{kcKvJ&<=C#@HGisjV|q&flB%v=@&ts4{q@M7D)a1S5LQasZVf
z2Qkg<{e?Sh!wa%R9T!N=Ha|p+s};Lx>Gb~K{_&qlSFFdTHfXISd<hp%MS-B$;s+AD
ziS`Es0PzNh#Qbca%bf?}B9b$FF^8tQT2XDQf4x6;>^>r8B@a`OAs$v6y!yaqF)Xx&
zWDh;}xZLR+pb*)-zq)ll|B3Y>`y0gK3t}_go*Gd|TxG{Y$p3cF)%@$B!~K{;`_9cx
z`%&>LNWjBo{lpEX&0vil07(p(_n%ZJtmgkN^taok?XM&sY<-jXkhbccE|WRdG;KDZ
zlylLS5R!Y<W!jYb?Xc);${s}10_G(+Tfn)feLaz3Lrj*h<+cIDOhk>Ew)KA>xFsf3
zEcVX(@9#y~2g&3J=o}oVZTsj$=>D)1^31=jx}}Az7Oi?!wSMvM9*jxy#Y5knG4ojF
zYSfA`><htwb6CpKHC_w0$IA}G>!rbyoH#s~M^HuKPZPWv6C%}2(UOWN(iOUrEM7nm
znqd~vj-cYp*bWsX858fiD@{$Z;twDYV1E32kD};)vxY13aK|9}G8ub6MDZGAMf4qP
z%$RVv4d6njVblM!<?y;^@ihB-!}qjT3!nG$!g9<;S-Kmi87L+HGZ8YvVM|<3Zm`?L
z4!hbD_>n*&vm+5@Y6G4{G65?!FcZnH(sY<uA}TUsL?n<Tu0!Z66HMwA+FGo3Z8q>_
zSeu41PKFC)l<VWmb@tNsqL`9eVF9+low}3B25x;aNg@8)j1k+#>VC=NY86B6@F@%s
z=d)fw9_n`;lyOgc{vf4{qyp>ZlWzkFtI$|zM<}WJiiZ>rR|@ey;#Z+iUq#1&3n8E@
ziK}lA38gcP|0z!_4fl$hOFkO<P2&xduB+vWYqdV%@3u9C|5hMG$j3{`sCnANAWj4c
zoZ+%}Y@b~|*Yn=+Y4QnqoCC66az*!IUoNsh<cb>e<#1A^^)W#4<skNMGWIp>>h<|`
zXe6bvUs_Gn0nq7OFG{*LZ98fHYTk69J1s_|Y#quYjmA|TTEtM%f11Rtz0Zlt@viRY
zwpGax;TWLo$)E-#vS`7&Mkz;Mp=(W#^N6e3pc3t(ckTk1W}?h7uhY4Iy~!R`c4m}A
z4!PfCLO~|1+*{TPrIC()p_ak8I)qGY4Z@ABjiVR1--KEh$&Abbo27umZdQ5&c%b2(
z6}%eCzMnw6;qO)U4C*>553^q5LgRltuOlc#JGQEpuY|}C`f#8(x~I1*x_TT~xZsPh
zD?Ah~r4qynB+z7fIx8N}uRcvN1-at>aB0cGK;m0tsT!t0rU=h3_HqOd?1FY9tF!)P
zjTY!8yxVGHuG1X2c>)QwAKQL?KnKHjQAb|K;?-L0+IDW1?`SB6Z*T2pbJB6IjBM~W
z`_&<Q2`PLYOzh8LfA02ZzZ=jD*qVN1-67I?vjA$|)qc~H9x%PrU7_6E%)E#T0mNvd
zg$JSqEEw7T8zY(0n4^bF63r<<n5N)S+G@%1kA{=6(>D8EE#AxhodB|tZnJ9vfW9Ki
z;~y20!*;-h9{S#PiTJRKZ{~CB*f#H&L<$A#zG}i@a$6ecSuH@9k}&dDp)Sdo`mSP4
z4ILq(zEB0pKk622dqN)ZL%ov^p*HQ_*Z7#@(iGmEvzwh>^NPyIsjy@Yb)Wa@mdq~=
zqWD&ER$X^@c$~lyfuJDHWZQkwnng<>e$e_!{qXASjcU?KK4Q`zJrFuO%;m}*@g;%)
za*CXX*RtWhE@>Z&_<H8|jP*3)_nh+7*YUJ};)Y^8ul#xb^dxn^_t|OZXROF(&#mZ|
z-yOvZ{aI|`(zVN~RKBmC3t}WQpNDTIGa5#!6&s>t;|5HnQ0t^^kpT>oPoLb#R&`E~
zDr?nmbeXdga@~Dv*y{X|DgP`Zn&01}*I2<2QL>?CrTtm2hFM3u!qXeTD%<6u7!^Q-
zsP|qjP34h>MMLBdH>Nta#e^g2+G9GBGV`~-S!R0W8a&zE7UP5;X*@I~?)ZOzJp}1o
z`l#e!HJ!J{$+Bo~;5+JKTT#T!_ae|)XbJ<7^_|H5Uz3OS++)El%UypPLjDZq`G5AA
zw#vm6R-Yu>uE<<bru)7V7xX%jjWOD+3N?_1xR}`elqE&$6Hz6Vuhtu&xFMp;0K4#P
zn=>}4V5{f+Z0}4>N^x_0dVSe`?VrAdUgc)JT!^DWBb*k)-BpHHvgp<x|9*oPX8o^s
zqA#RpF#{Mf=C!pvdOAOo&03lhx96&_s}=11*8QdQqU>8khQf?Rx+J6onZlWBl{M<9
zMfQQ;+j4(@=FwhoWzghl_1rw-x>iF}sf;tx3~pY{;LUL7x?h6cVt~*Ip75sS$+LTA
zL)dM;g>#Z(LEIO~5^t<&pZG%sj=jsr0L8ZnJh{QLo(|(G5oRjZtm<xpoiwb0jS1~D
zwHRgpRBPjMBh6dsfO4J|H`%qF`nx5MJKKs)M{e+i*}I2oqs*Xv6xKhLaFle?c4}5L
z_CaCXvE|)vU%!Zb{Wsq1`x3=H2EUS8L5V=dbG8<km<Df!d$kcMJNf@5<0b>SwN5h{
z6=T+;&w1D0YMuAr@`g-Ct-h;YHmn}}ITKAUu0mD%OE!D?{BC0#>;8^jbT2{l5x0L^
z7>L@cr>MczF8cgW`e_pWeEpT})!^l!^7W?vllT0;BMe7csZYuo9>i{P&yOs>Cr-5X
zNjVbJAO(Mcq*`KSBN`0(mv-;KDw#7kl{X4Zr#he3&A(K_Ja==3)hG2|KrY^mMH-Ja
zfIOs;mR~3l4Lr-Po*o6qd-YG6OoRboIlb3Q#)zFjSa?phxx<WNg|_;s2tm6^I)jR8
zQEA-or6`rwpNTqi$-<Q$Mdk8cI`nkuT*8QfaHI-oe;#Mi&3fjs^l8CE@%CT3j|1hL
z7MXpG4A`ljwl*E55LgCl_b(yh(;k9{*f0zdGweC~m)iNVluqDle+_8KGCj^@0n=d7
zKbDjBm_CXKN)Hbw%EyeA1x*1WJo^C`eUWVUlZt-5wXs<jT@*f@p;_J%X`EqI%M?mv
zwE`mdcd8=y2dXcp9dLlaFVI}nKC=tn7eY=p*WT4^TG>Y4nz0v`Gmi1sI$WGO#iObJ
zk?z#xn!&G4Cc7NU6z1~5kf(N*Y#1rbejmN`%HHo;a7U|`Gn^Hjy*y4HG1q{8U|CSP
zULJWrm{sf8ek4PUdzUdT-&~mpWnj)-fBdJXeo{S-^vHwkOoO`Q7bFmY3dh-}@~$n2
z&Q-LzbqlRrzra^wHpAmESA6Qw%v{w_>HELWw{R74<)5ZJO%!JUZe+o}`2>(I)hU1&
zySZ-8{<pEPgBnY`knt|6>#N-lR7<WHJGy*x*bs=d{=1(X5^<_Ifg;y_cOq&I9oE~y
zPeqX{nJB)`eFhiwuVYykFNGA2K39pW_700x3XW7j>0<PA;a7hw=$~=FJ1oDkao-!c
zUp2P5EWhgCC7CAe?)_t}(Zz}G1K8KFu!z-8I>{xch^rH)b;}zu5=UcFDd{FG&R*EZ
zoNFYXLN&=VR1EOJt4dVd&GiXZ;HCK}xz&}cZ<%Fve<w7ONhfOnt7^m{6YAJZy=Xqu
zWNX2yGNlEC**rX!yaCg6mB0^>;+V(e<+1lmU}Y5!=LH73T1J_XFU1zhcpWR!6@gTp
z*PEU6@c*$9r!D^R-db$Eja)2|&p~1eANVjVMW)3J%ZrSeSb)O;HX0F&sL5^B=eMKz
z?DA<&vXK62SBBaiZ-)6~WMKIlp(#*KBc`LE?|Mbm6u<f!gTww=Dw@9Cp$U72*RX6$
z>;4F#!DY(pgTcj6)>Eo3;60djMZ{ScX&~{FB2;+yVF>?qQ}l&L^!1)ivlF#8mSE9P
zyG63nez!^H^iA<H+!1`|M{L7JArduu^2g$uA_DhRo^R;y#a(}{B9;nIoT#E1CPsH{
zXbJ4?J(LJk{51#I^B(6Qm!4U_(O~+ZYj->VDw{dsDBMA<mMCvgJTl4={~pn#LW)15
zSh>H002M@Jh^)PmTV4T($?2h(_#ynpQ<255YZ!;;p_L<KEA5spt#9G|XornSR7^qY
z*^`2%w8iFo?B_2o3MnRAWwOY2oy{krJOd2j&_Y>PXwMsF8-(3c51X4?-}$R)(R<ar
zt2`%T2)%-pF=Mzkt~j@nEQVGh_2>wYoAZ`;yF%8-gXw==rvJx<(ip)e%Vags0uxKP
z<F!^_Mq1OR9Y-RUb-LjREqhrfi>H-~pP*XKRd`Pol`B}mShG&kpWH8ZNY1V4@Kb51
zc_J6!5;lXpR>=Ui-gM@-U2kjWA{L$DD2m85$~j+t3vif9fKuWXB!GNz_#n{j<fnxO
zk*YQgpM6_6Qs<JnsFJ8uF)P$DE~?GxXe_P41URxs={Hu)V*IX3ZrP5*H>-JvWUTmB
zJ^KBzM52t^)Q#FJ%!+oZtp327#iz|-2H|ECrfvFH2qUVPJ*9>}Lbckjah$<=l{Q%R
zK|g$5&JO>~ltWnyY|nU~ugy1VvegbrO{jl+<}Do+C;E~gio`NnI2~Ud563f!S?18z
zYQ6-;qXN{Y?(^}w?Fwc-4w7~%Gx~)k2R4LVT&B$O+n49pOA;H@@>V_{Vu4LZi6CGO
z&<5i`I6mZM&;Bz6p8^>+-D^Kz+qzck=Nau3B;dW0JzM-pi5A+l{jfLO5@YYJoIPA^
z-|#}Z(<F{H0(DYZ){WHI`it>Ly_QFgwSGqZP?qWga!&G6OFM$7cJ^qbVa4T?eLYF(
zY<iuha-RRR%Q$tV&N-YyX1d$XZ)4qUBECQiRoIRFHuI(#+p&2gxioEP{|HhqUxFPh
zu_crV_<>u<y-<7*q@7g!MZ4U)i5Zky4f-k#l~zeFF|r%Ol6~+0Ei%`%)va|PZZBeG
zJ?m}^o;EuU_32IM!5?>>m%)C4Qv$Nz!&3}=F&L)%VE)zdlcw54frWTRB)F)~K}UuR
z$puJ&#l=kZp+wrpl1#JY-6b+Zo&8`g%e3OXuBx-ydA4~h&%ERu=3q7Xl=}T4f4bei
z(zk<<P~zUJorm<&LcV{GZM!&OO5*I~P8-vrFF5l?1J?<qN0bgvGR4oh@NN-c={S^;
zdDRGw$Ap*((HS#g^TCHSgnn41Bmrzxg7XB>aTB<h3z+i)LuDIKT=07r%|U73*!r{U
z_^VoY$Mb{}-zwQn5fm2d%hetxUAx7G&cHz-vtrX@7~m=Xn%*5t#M77%dp-iaw6>)5
zq|uLXYzLAy)thP#v&zFO-)+MnOW-l2L4H`o*(vo=ovZ0A8JW;M{~~si=y{$KF&G4w
z?>?2VV<5J!#+)s3^?MjME?-!PK1;)0tJr)mz-Z|8KiA5Sttj#GR1l^0kDAZ1fj#Kp
z9XXr=2w*o&&7!X+XZa;rGA6oP^$0A3q;W*TYWQ!Nb_hwcZ=vZ~WlY8;)t9;U@_K}8
zjp4{Vp1{>h4713+2Ku(Ewa{|x^TVW~Z}+BSU+wGi>(7Rclx-M$M4$e35<;CC8JKAx
zEFD@|djd%!V#P@D@1LLJ^xd=8+>H-QPaTw}O_2Q>m<n+xXlGMQ*hE#+sAd}4S*Y=K
zPNQi=+TOqb$&bl4MAqBSV-#7RyN!{#|0HW<F{5qs-ORyb!^83z&O_KjRY0OZ9l`Gt
zp-f`z0O%OLAL(XPyP7eo(MN55@80F6(axWrre@ALWQ$`c5-ao1*(Wyao3Oiu_9S(I
zW`f+i00n^xOpUI;0qah?2tnPmWB9KJ^rEzxbnSPG4qq7HGV?06-D%tV$A!1zLtH>I
zOs2Q$>Nk3dg>mszk@0X5iOd~9Ml53*svKlmzqA09Jt?x7Ygksx<!mat@NNAs4xc-#
z%NOd#VO^Z2BuQK_%1vmmNOQYCnHYH6M08Qz2ufYh>`(^iqCIXcp5`!2+6`qo8tqgp
zfn0tOR+Y9DGG-%x8#FdzBP>4MvPt=EjPtzW&r;8sw!st{gBy8Q;ov~x)tqyL*pT%j
zq95oy?>o?!2KT~06(#lG35;^R_tMT8)A^}qh1`ca&Ha)=UGL*ijH@2tAMzeGd2-gj
z{ia$)o1vBfjMvA>!knYZHFk8BPPIyQ##_k(1isRnCpLT`vh}B}KeON}{wM=nYPjl7
zD78^Z_wex3jUp<SY9<&+XYGrJe<mcr+|lJ4s^qNw<hY!S=ei3UJ|^W?wvorh7vDuM
zhXKoII`HaaDVgTq+)fYqJKbO0ocXpfHh+Uyl_1iyRM`eVlk-Ycy%1kQ48c2;;p$R_
z{26(a#*+b-FLlpLbIs;jMHhp0m!kU^6PP@gX(`|UX|#sj!4{d?DPF#mZ{*&uC?g!}
z)B__O$38QUPB5UZ97||o`qu?_!(Z1|K>V1bDf2SN8SNoV8-@j-iv_6F6qKY}t6ic?
z__)+*Y`tgFI&z~s&F$poChc)a#X(|`5(_QlQzTNLZzrP{*AU<3X8?ZbYoP?@tqKs_
zr^Dj}8`lO<(;F2WTC@x44pZnHg=rcrM|&=khd34hx@|rnqeGsFe%-{YUt*jlHIUy#
zh}Dt3dKTnNn2i3k2%3EJ|30|-G&0VOtuWV=mAO34y6M`!)!<kv87)819}Z5<)phxf
zNROtml`i+yo4taYo{4`>Bn%}j3SUp>Ys~d;uex?Z(d9&L)~_XFdLqAed0zLcnHzO<
zs-`_LSu^|8n#E?87U_?%$^dfNMPpJ0LJN<P`L%EU=HtzjYZs0*yDJasfiv;|x5%Yz
z>0)Fs4_St&?OqZ0CtpR1a;~-MW70Dm*<qP60&BGL7LlfiUvgbGU)dt4Qi6ey`5YcJ
z{n?T(-qIHHmIlAd)ge}r1G~LvJCexJx{vcUfRs}2tk9s_2#ExY^Wsly)Hza9tS427
zWEELL`QVsU-^StMrC~wmIFngaL%V>(G2wde{)Ty-lfvi%*s>@yx%SYw`gh=4(sErp
zsu`P9dp6*BVklw1^UoRojG3|7J5q}1`N4~f{yBvPk4IO)8bXoOITa7VGAD%{_3|xC
zPb&`wCwaN7t^VnOU3}f#Z_ncPB(?bK9E~w0>&6X@W+hO~s99T+a1e>w$Fis@-lzlP
z0i(wK;q&H;x88f!jr&8h7=_-7>3g#3N~h(!r8sTNK4!oN>tbRv!TpP)Fc_k1C3qOJ
zUkoy=cPS<Y+Y7F%xEbG9O8xf+x98ckx2O0v(uzt$z}*jutz5(8eY1=9a#Z;e;rFoq
zYH+vB>5sMN-P~s8jhBBe^vpFdn*c3k8h0EnLehJ$f8a)5Of{#bhhNx_VQAGWK9uai
zwO14Py@DClLRn+!XkzTnX{~)vEqZPv4HmxXy>*yi5p+-<>=qEuwZ^x*XU))s)LZRO
zj`7Q5)iL&uxyJ78n>%3TpL?B%02C~I<_#v4!5ckTMs&heFN6$^iv;G@0=*~<tz$Cn
z$u`6pX+9=CPb1p1Ht9BX2yzvOyT&lv%xGr*k_!=vt8t-91DRKlk(p5;>a1XTC8q=S
zfW2_lP$v7F8fiD9(ttyerqQ5{lK#twexnB4R$>WZ1y8hVuc$<KXflkoq^68Goq?8P
zx0heBUgCQOhFwTCFS|Vw?phV(Sg2$3R<!p5E)WUYXkXquJ9Q5PuJ`cWO8Z{jTRgcx
zf!DxDE;v*htI(WoAp(I)GW(#?(odS9&_i1x9={?oiiD)E^(v_vVYRELgipL^tVSD&
z5gW#o81%57&g*N8Y!K|15%Z6jXm4kxfw}rhnDpbH*H1g{Ix*FM)X~zpw(tp&q^mf$
z(i=*Op}j1%l14C^;O?`kpA|oQpjRHIspd0Yk?Fj@24|<_@WOGDhJ_PVSn7jk!R}8b
z(#YwF(-#ut^aWVHL9Gp#Szi%81#om7-VCxzcpu0BIL5gIyttQJOj+coZISQ0lXbEx
zaV|<+DCGfMDY^hjlB&I<e<ReXm0)x7kITy~?RvS&hS{Y(%R*RvVh_uYT<&I=j(qg?
zp<Frhcv%eA9{8CjLrDFA$xvZMC>o#Q?aIo(KLU!0Wm5d)L91)ioNKbP`R!`vnYP3{
z45UVU&B@Pd+#>~lT$6R)SJnj<V1o2(a@2*WToW%wxqqCrl3zge(B+X}0QILI$z>J1
z5S=N&+=d3?^l*vaoYC#{Ud^F@q^Yb2^GIYK8Yy+Q1BD2`F@<CO=P!+K((;{{DYL80
zTTU=>vJjiki3Z?ofjnajgN~G4mv(>c0vbp6jxMT6`ghZsFN{fbK2&k`OLjhXLnY_-
z=XbM}zI|DKmki?-S&qwXpEsHxAMgjJ;|}>hg8OGVcRr^97-$(35P)T*=HRLkl_Rq!
zDG$~}V{qk`_4ij-6`g~POCf^&WQiqLa@~s2<=v-(^v&4DD5hS{O}W)y+MB7UU#SF)
zhKH5lN412ate}Lj*P)Wj4s~zpqVNy+s_7cIt*mKleybG!+uc5o|AgKkrc5z%0U+EP
z-rONk0`4{qjd3nlh<{DD=J}aK`2b7U@8*v=G5(A^&g!{`@NMWl_6q99`b}%xpY@ad
z=C-vtDNYOnBMr@B*_!qu0Z69s2Jb@4O5no-YT&v{@79EE*P9W6@<coZ<-#Dn-|sf1
z1Pc;9T-+2Hg9EyQXg1U8e%1ZL({;KtR(}UhA54#`R7!8NZm6@QIvj=2ub7ULbD%IL
z^@qdVRXLf?*LANpHzNQTLD<cH7xCMt!rRX{vL#JmjF0@~qZIvVn*Mo6nkp>HcsW=s
zzQ(k}TWN)(7u@XoHSdP--mV9RG09jjrd}FE|EC*h$2@t4_C}6RIbk@CS{6CPXzOhD
z(-pUIZv<0iSG@5t=bBdw>%=t<-;AgI+x;L*7TZz7F0$q1{I{|=Huy~W%vR$zBAtkV
z3!c-bYkI0ur{^VIe6(CSZ8NqgZijPqO!gxvrKGp+vJD~gAB9(*AhWYR*xq+@@2AWb
zY@Q#Z+uFii8_|<ST$fpfbJUk@A{~dUIR)Us2Wgx#4WOXr0(G6di8o)BnLeo%n^9)G
z>IG>g-&i>KawD>k;8<+FoXLFWqQ&RdVcJY)kZ$MzE6lq%G{o@gD$F0$&0R*?9xh>c
zOU{i(Un68NRJ32^UQhCLoY^VYRi<n=tlMUS806qX<YOxov7-!&<Z&&?bvF}hQ~TfO
zTmvew2Km3~e|YUskCB{Y{`_vJl%r+L#{2!1<w)6Y$q-Vuh-VQSql$O?VF9hMHgc#X
zrx9^P26`|47?4n3dswD}4X`Fst>Gw>Ve&n}#%q1)_sAQ_2=B~|;68d2fgp7>eTDhA
zx;DdChjmTuuU2-wY#9&2@2Xh$sZzRyeh^INH^eH*^hl)J8Wfe%x$|nz=(a`3cGs9I
z>HEgaA?cX{=e-O9La$>8#EiLWEA_G(m1431D9oQ|Q>qth0m~$G9R5enBH%A+`n~Du
zGpq@+c9xrH`UusvDd=23TJIT0q>h;7_}J%Vz!D5{_i;0Ay_eS<x}}lKm5bOJo504I
znj)oh2$D^c?tqJe^H`_#^aabq-lkTeLULQ}(bUMJU#7wHfp4+}FiTPQDyNjpL3Eaf
zoG4lT^S&O-`E+`@Q`}C=L37S;g_ICX7O(Q(^&>`tkNDe|c;r8)jwSa!Z)}uvA|(hN
z-&;XswxbHtg{hK=bkP7Ni|Mm0KZ-ZnRtLK4n@w+cn)O_qTRTn<*I)KNj!vBJusY70
zm#mX$>u*T-G#ShxbLJE6069m&=0xV+DNZ?fAKqtAiKZ&puWlM!u*Ego@j-rKlT3za
z>HT$OlJWRFFyrMp?c1Tn!9!HcK@^ed`y+`|9g?(dR#>#ROV--726}PeK0S86B<*)S
zO&KWB4zO*=zFcVVI-MnHnyabITf$5VYs<t{`Dsjtn+z~B3p5sqsQD{rj=rH9S8x4R
ziKQO5sx8924yQ~gem-zmygg9Y6u6X|teaYx_9f=qNmB;0GfaLzjvH2=XPy}#r*7;e
z%U82hWO5Rvm(l&6?y*0hwVK7$YoJBH@o4Q8`2XECCj7A<Z;fYWO@=VpmbAz3-M6mB
z+YJSv?*VbMF)g-$8vc$-(+Dojc`vAN$5yFHAS8-tP4Gr=qzngVF#3HpdBdWB`H$L<
zAS2orn_IWE07)0;49N+PP+23GO&)l3cr~>NhD3<B?)DD97lXN<7k@`tut-(be?#1T
zN7xQylaIW67EKQvKptzJeDKqTaCTt0psuLajFU>kSE)`bR4l7GEL|OiXC4q)&y`3+
z6q;mO_jwNhCX)(rDwtUQm}p#E0aXF9I2ElBsPSo)1?`F-+Ap)Rg+xRr;;N=IbPAb?
zye?KA2JYX_#+*cnkY8KksHEGie(Qh*<Lds)b1K;B=BCA#c7-y~hgg|H%%@U+BU;({
z=kJX-KH0Z6&sJ5uSzQ)^(^xi|g>(aZVeDP2r)lG&j}ca>f47E=Ln>W<p$DD$kCUe)
zJ<mcj=)LqIQlMQc)Xn8xtq7vDr?NX;LsqOaEla%88L#Odnejc;AzQ`oO4wn+k4Ae+
zsI0)m8JdG<rQASuPCk&(IM)^#m4NA?)-WpGGb_`rXaX6V8R~4)2$8$`Ws70Ryw1+b
zaB(^p>ei|LpQ~$P^pni~-U}g+mw;+r<)o9wlp7u-z^`Ocpr#}V9)`;`fgmvtGPT+_
zLK=qrH2x#`I}A4v&63soVwo+qF#G@^M_hIeBuF_}2A^eUwYxUUSU`GXH6WqG#X^eG
z1ZZzbj^)ow7TiaxI;N(6C5T0DuqAvMIO0Z4LgQdC2}xm7e0t!^0$*RCtTg!#V$vF_
zaX~9Ox-GC6Y6VZ4+7~Y_`4`St8Andht1I-h=CC#**e!w?wy!eUVjD!H%AkbWmT~Uu
z-lS1}m;A7QfN2lKlh&msla?lE*e|fY*`j7eIx;_>2Y8t|i!Unq@D_V!Q3SD+naxXs
z&CG*ES@Vl^v|b(vfe!8629VhMt1f$tprFzBMFCM7a#}9MaF1d6#Trl@4$4~wGn$o#
zP>Df#0LNH@+1F{yT{MRA{<P^S9*RK_4Zd&|@)Uc#-xF@mJItv7#;emb52A@?%C*gJ
zzl%7N(zpTKq~lpU1Tx2YvoYaok9~AgK)R&MQ3w;;ZUF?ZM6dbbVB`<q*;wc?f`cY+
zzV<I}WJ)C_nClXW>6=>MzFF|HCT`aA9+cj9p#7MV3_Ch+5utBOP=X%f>Dn4|d<@mB
zq*EED{2y<v*Z7WId@QJ#N*K}&K^qM>I*s{fDTE4BYvn58%FM?;A{hg70$BUf<Njru
z>8C9IQ)0=g6RR0=O5dl|0Wvmws`gZl4SRZry6RxPI)BirXneK<t%QJK?p1+KTrdWX
z(ydu)2^hfmWGrTC*eyR;8_i^hEVP%M_6%kbA;1|&9kcZZ$x7~PCXI~|(RkGKV4~E=
z;$$=6MMro^*TNRwJ{}<%%qvf;v&-viL@wU05Zw7Z>~SZy`Oww-rF5Mrc4<uYl6yN-
z+D7)}<Bpv-6-%!62q#4|ZzpGK`ymOM2C<kX0jUhBCaZn%p8eW@O3H;n`2SZ!6s*ui
zT(xw7{52PX2d_6HHpo;syIk;Ax*5z?yE$p;;hfd=H9l-L?JMg>Cbo%|(j$ply#X(-
zR8UyYSTlJ{i_|^uZCh|v3(<RyzfVUIkdFpF_kjmw_CHJIxBX}A#aKQq<X|($rP2<o
ze4^`MkhZ?c!;k^`EjbHexx60J{gDKmP?4uj&*s|HRtwpTe6b|vNm8oCGb5uH2m(0Y
z{jTGU-}5KgPrNWAVfdR$cQ#Jrwjcgzv~}Y=&Tb}7SYp)XNy4-)RetY4nS9Xw_Lp2m
z?}IO?X6<?J`z{m0gU7)G#38fGL<YvDBek&G>u0DqCw&L^8M-#L=~iBH<U&hQPVz-u
z$@U0sr84-&)=!Y2toFzD4c76M>}o~_P4Ni|5Bx+*(UxwgAcqSBP6JmN`_8P?03G75
zk){YpHuDg0&N-r{W-Ds@f=UqIFh|`CRH+tq1;SKRJA}Rb#+DAZ!N4TyZe#KG6Rm~4
zgG2aQ?P&}^@rv;gq+P-&gxri&qDmiIYRhW=*DKj%ww_pgw{qH8p_6|sjAXO+T>_2E
z48eSRg^`hh)NgN5NHc}Gex{UuX3R<^EJ<vhh2>n`WfOCnv9yE<J=$`2j0>gd?dQtn
z>*EuxA(!_75Uzn`FAz)$#*BI1jnK3&`|h-@+^a&tmS)pl;u4KOsSIvgWxa9+oquCc
zsT08`ceqS9&{<=Y-1L1}qNc%(Fp;u539Q=(EnVeNrmoP^EJyZEr%`SO%H|vkh4xgK
zQ(DkC+|Z5mcg@H&GV3ggthYD^Q#LI|s9<a?t=Lb;p@p3s7<bM-XMp!!&Me1BkBIGd
z<%u<aN>Z2GFh>&Bxqk4b9MQ3;Ym?@!bRZ@K0sX3yc6m&So`IU$8muq^Lo2tAu^bS`
zmfDxvOvU!(mRy^;VJjkj#qidvP7g@4Xe=HAl0kDs*^{fw43ni?v8!@%=Pzv<SR=Y3
zV$P4~<2`RZyA0i8;=I7EbhK>`${$<`8qeuq>hTXO)e#*!^V)nc+vh*{HP!pfW4@Lu
zCC32!It3d784bPp96edh<$V>S*T2%oFO5O=4H%e)i>1`qdGS!h3Orpjf*GILviKa?
z@4-Zft$MMy(vVu)!oCKLjPvU<EiAm)9zut=zmU_bpd8Um)*i?~L}kWZp)y0k*(*40
zl-6WES7lMtb*38OTwc>%Agmi=n}W=+3-$mKmaiGH=A^cJ(#LI;iSJYBU&Mb*>?PF8
z*E%(*tkYftZ3AiK@Ov607Ckx>ECi#L9%UKw64Ne6PY-*oAOL8R6Ko}Gyy%^yN*9#(
zXN}Oe<B!%q&^WdC`v|PNKCtw0vg)<kls_~J*XKxnUzWGt{~+=kz!A+wOADk4O8VOL
ztJZKfj90ta#dqZZjpnq{6)en$-8|8E+QOvni$*zv?yy?*u45PUWbfU0f!XQF*WNf8
z1+aRTK?u!AdiNm$#kF&KH6V8VHO%MzFLPtRU;#x?J=$;B4rkod;Pq_4?Ydu#n}K-s
zdo%PHpY=G|1{u@mr-><M3CF%o?_R|ZsmVm!yJ>}T{UKqFZOdSn`u+d0L}u_hWH}em
z-k0+RlYecM%VRttqcApVVour0iRk#;X9Tk(E>FtWs@x$X%TxR4AKiIlnrnuwh2fYj
zW5)Xd5~0=9<sV%cjiKhpD=vP1;(Mbn{bz0NSY6cidg$)wj>fAdFz}Wy|5aaU${(TV
z6)dhSpDPU2VIeOE2k<8{mhdJ#o}Am{Q^~Qx64Bb2FiMiJNpyd^8^uWb(_6ucirXkQ
z=_%bZFd0%x7gA`5nB!ZvF8_p_9Zphre27`TJNf~TzCD@m^CkDZOq1i+zi17LdBqpA
z7aTo~a(}cjKIvtyt@rr7e2tnPsy^l7nj4+@;*Nva`KSrU6zN4b>CPOdqmD7s(yO0S
zhog3I(oirTp%&v=TPvDHupoyxlXXvRi*0Z0{ST_V1I2NRB7Z5>ZLBK6@DcPAH$jo-
z14OtT()mC*Ow+#T;dcKj4K61<3RB;towiKvER;iBsg3*ftR&zQ#7te1hn=^y11&@y
zwJy_Y`FZj~X;<uy=;Y^K;{Wbg)Bh5H)x_Vsy5RTcMFB2mSLNX6QF}XuyOZC}i;&~O
zPcV+R3s9EFPm&?2<8vc3GlL9LLJ^F;UMI&k(@oDzyo-~nE%)3ls;4)yvb`EKwQ<4v
z(Il2e3}-L#%>XNrdlFBE$!2*HpBBRkLY4YHWnp?6!A|uXMl_{o!GZY9>zhH2vSo#&
z{B8T|-vqvn`;-DqpYAvQ2E{^sM!QZOYt3eh-%QGO><evP$$Y1I^ck2goOi$FIq=R-
z05i~y(`wUc;3C&wwy%A#A1;D1&97|*&DYi^qAs{sxOzRT1!_%m$AttOlyc>3j0yd~
zEeVdXe9(wO(9$vcAxzj%L*`Q@EMFX>?jpeSWs97pL3jA_r$eLA^r;q5){c+(-57DB
zS8*nGq>G%Lo+xhp$dPZTrZfp11T2-V+gbzza1?}Bp6ff2(6x)iKtkYjQqHm-&UTD0
zSxGyzF!(b@P}$<|r1c6ly$^IhfknMzWtmvV(T&Y{I5P8B{4&3P@vo7m$@L2gI@A#C
zInl9i%+egCh+r*)I(5;B95Bmm#VX=$_E5CuESmr6eQJ<aWi4e1vvuYyW&dMVvtmp-
z`@~M{WbTldY<Kb_*$X8THwWCC69mZ+&us$E8KB0aaT2<=!i#GZWA`Do@z9MQF_Ck0
z!s;<*UjyFZ=i`QE%yoW6oKAk8Jfj$CUvv?!jui5{fpCv|CyoD`lr>gW3BidO-~vMy
z2K169S?a2n0;2~MB*LhaROrgF*ExZr)RBC`@4k_rBjm2@V3(=K%-%RL99ziDQk4Ug
zpe%Bz=kk1j*r2d53k9$q^0YF1gu!V?NYVS^8cjbF?pnLeCWwg(yY5wzdqCqptB}I?
z2WpZM26MtjJ*uP#V8jRX;;`C7V++#*NnDIYy9SBlrcCYxNk@^Y<C{St4u~|j1rVik
z4_Hi4@L@2}4JSU{Jik{Uz?g!4zZp7Ce7tp<q1?PkNX}OJCk0hvUnN=U(`)_UMeD_C
zzu=qW)w{Kj;Ui0)9@Uaih7xy_(2avX+a;0i0gv0^iMZ5e!Z{>-Qr~Z;qOioHs;Td)
zifU<VKYUKZG0zoPt)G#KqVO!{FraoDSx41-FF^q>3czb#FHx^&Y_B>6PShB<J={(T
zT%oKWOju_bLOTiaBP)AhH{^u>8pg6nMyrE*9zPbKBkM0~#-WR8Uc3>izcd6i5sCeY
zaXj9D@~26Cu`w`TD)kGsTuXiBB*6s~!Tb4=wh()8dvwewHs>G70e2aW?O0+CI(#IR
z{O)O1&VC?mJCWZUEE6K6Zm6IHk=!mTvO086m8*y;8Z$bkq@7i*r0W$xnRTKl<SN{h
zVIV#iIv0KQKcMuaD$X5E${J!wXSUAxen<3Pua2`wS2aVg-WD73MPcRRZbMDi3imUr
z7q6Hj*E(?4649GH`J+guw(rDtJxCXcE^N^?l<456<1FAyaBHg<BC&4d$~T+u#k9+j
z5@FDa$Ve(?YP7CP8UVqlBPSNB70xP&&vMhx5kDUubCdVpPcJA!elJme*K9Au+w5Yz
zp={az)+Pa8RlcI#VM>raig1nvJJra?&<!XA#j{1Pn@b2<fXZGaceDtlsqg$NSBN>J
zAmEjC3a6`4%1@81TRgaZ(^j9|9o}%eVn#85MPt10kMP~LNBMesSKarx$FQZui?a5Y
zJgTMAbQbkS^8Qf=O+ZlfqRbMg76PB_rz2Wjh;oSD(|6&zW<$AbOf`6n@_!I8HMlNg
z5|VwEj5G}9%}}<s)m+IzOTM>ZUie|35Bg_l=Fku_qTp*2I*eMzyeP<n*6G!AU13bg
zFMS@&9sTENk5*CC#~Qp`1GHJJD=__3AxkYGJufY0J%@b#)X;DP`;aHknwqPv=<I}S
zT5g44!<_$iKzyvByygjx5N_Vr7Cs$jy?of8PV>{THSi3gydkra{e{-*L+LKP?g3)l
zS?pxkCwT=W?LM^Gc#dfkIIxzUu#Ryb-@C=JTuaR#Vu*s7URMczOC`aiTb8=m@_S2{
zy`G#XcLHUzZ{utP>X1Fl@M_qja0@p__t*by_!8Zlr>4=F5upi^=~c%#A03fIeB=1_
zW#81$8H09rX_YV?6Opk>otv2Gno^5$mUuQjm}+YRp>TWgZrod#rqmuEe~SAp6n1XQ
zDvNyb1Uj<5FZNNZSe$0amwgzgZT;GYRhJoGU#4Tg)~NIGpi!2-QC2~|{}0%`T-V&i
zK=icCD~suW2BT*k=C57Q%+?0OAPGVGdZpo@WMY<ZqED@Q(rBCy)}hUYRROA@bldM-
z-u*`SW4ya<vyroQEcKxQbj$j#1^a%Ip5o%)A@0k`_-mjM|Hb+1sy_mZVe%VAfNiv&
zIDOMb*rcioRUt6>1brB%m*kJ5hq=t`Q1hC!Zx40Hhkup<Ij$nQOWCDljWPWy9(QWN
zB4>X^*x|Dz&v6&-JMzDA$FI+jb863EeRsy8++b1km8?@2K75rBCyhVg9U*O=fK-ff
zgq&!Pnjv#EB#SGSUY7mJx3)xlWtwOjk+#rM+2~sIF@(>FtT%Z0@Y6AjaTWP)1WX-`
ztDk$JNn&he2o^?>E!I=Ekt!q1cFNgHBp$E=JQgjReW<a?e3Iq+3-oZ8`ws}@mkV_y
z><1P1b8194#KwrcgPxQ}9k`1mfRDvZ3RdH94Bj$xLCL5djwY(f+6k=cFGb3G3*|hI
z5Qr`)Nlz~91Otc=LfiN6;a-#Q&-1Ca;|^_#`|8@l;Bt1v;SL_NTa6Xq^KXZPrjP$U
z_Cc;Bb10;06r+)GF^B0(XTj(1z~=bX0m)D8>EgQHMWL23Qs{RbXC&=c{pD+o!gc<#
zGUka;!+d^Jw(~Ro@NX8!TeG(~-7y=b(+m|Ak(>piPGt8F?&Hsi{Nrg<7I*_{d%;g`
zdgY!Jk}2ph8V&ybxAxyf4a(|&Sj%o#5*Nl92`NV;NecU1lI!DAEsm?uLhzUcG(c1~
z;(<NV;36<gBXFx|XM*Qj*KB=)^&7P+HSMIm^ITRoe?E^oXVH@jIzUB?G2Z5c$HNBy
zKclx*Dhh*0F6ml>O?VSQ*4-%C$AJLh>)N%?19H4A3#(y5GU4u3=%*wc?_Bl9AIGz<
z4hJYMQeIE8Kq6W9r<Hy^rH@|i&8_)YF)%sWm&&vxg#(kkoix=7T8^pCc@&WDZ9aF8
zor$DR2k4KlLQ&o=tGhA##J)56m!3f}!*xm5ul&8xjQ7glqSw+JISNLt((m_P2A_q%
z8Jq(i#t>W)MTzPt)a1uT(&|fgJ<jYzAob}pf>>kx=anT|YoA$}iZ6_(zP2phV>R3s
zZmXoxM7-+VRu`BTDtyH{GQ}BTvP>S>#h4g!D7C>0`s}sdXgRix>W}yRTE1fqk4CW%
z=;ffL`9#LI0)^_yh}q`MyB<5cBPmZ`Uf&b;B8Vhk8p9OR7|iHWh|hl{XRJ?(<R<+e
zrrv_7&9-aXE?%S*D=x*0I|Qc`FIL>$J-AaGiWRp2#odd$dnoQ2+=9Ejxn`bkzI!J5
z1xe1V<5=6i?b|^Ka;Dk7f8KN~yWMkEMl^NnH%axQul0w9V6b>`j&Fb0<Nh0{txXWt
z*lOrZjsmu(i@+j<30vy=PA4d%roM4hk@sYGB;#v%r=kCFRY*C0(KAa{rl^}E=!9$Y
z&d;hns2dE_46M*`p7nA$f2yJ`9AA4u?GWBpkrLAT&+Tq$o8p920}E1eNBg=x{c1Db
znM)eh9v+F;uVH4bz#CEit&%gri5ul$ptYcxi?nj$MWAh9Q>2-0^;vLun=qdp13@CL
zZ8_cL1TmO|iLDU-yv_TAJ-X3u-_cR~!uUzv{BW*?cY01cuZ`6mkYlnMsb)T>!(PBk
z_Pmmlw7qXXdZ7X1jm~>aV?r*Q?ktYYD_itymRlk|JgsgiPsX3|Bq3aJq&!;`8TwB&
zmzB@;pIcz@panAo)F?cJvQ(1Yc+w|gBiJlP(eVaH5%5g;cG`8nze?fwu*>#Jy>ASA
zlJL3{u)jm^5nO@2_b^;W$4J#~jF*y6H(10}cAJ;xE78|OUv61m-|3yj+-e_-j=PWW
z*x~}w&)0`Aq3|xF<5$Ce834Ki8Z6Nzh~M8$r62j5X}HC($t)HW9zT;Li^fj^wLu{4
zxN2VRaf-sz(a_>lZp9~K+H7y)@{A0VZIDD`oTt6*#lsE{NxtIjkNuszb%qHo`6~Ui
z3c|6@xFBBO*8zj&hH^tym=~vl5RG{C!Sj;M=9rEDoX~%3JE^2xUDh&pb^PzClX6`w
z{yj|YvHg3~>C2N9Cge#|)0NU`&TV|3tLy$a2iPjAn-XsJo5)ZW85nxuJ9f&K!%-ZM
zB{YTFGtz6m1IkZYp30;A@u~eim?&=S+hyXT()W{d^zgaP<K`ub#n7K)p&QT7saD_+
zP|Kmu{U*Y+&<~aET~F!JbQR8R((rH6thl>?eExifxzD|F&8u}&jUYmMyw1?*8vch0
zZ48As)`EVPglSSls3!RbYgzl0J^3H4Sc_21elTtjwOZdyC^?qJr!ci<3aLl^olRyO
zMtFX6@XUDm3T~fX#7lPYWQd}d^Cx9DGq>*=aviEVs4A@KNw^G?`204AU9mdiKYsqC
zqERSFNE?&yretstTE8qq7rLO>)Y}xIbIwQ$!^IKwU6>h?*^opQ&23>X4#5k*$A3Wn
zmE5pQ?gXHa5z*Y%jgz6-k8ClkD`y_4v4C%AVcm_9bG5x|XMwmmp9Gtq>A3}h1=qV;
zc+VI$oBb|m-^RLL_?~wI-U7r|4V3}GbV2TW?Q+J%XU%>$p}_rXvZ>6OC9xRCL{n({
z2(Yxx_gmQ1Pow=ZQiL|bmUZY-xQvA7!`#C7%y{ApCYv|`)xNP2sq`5X*;(PuV;#6C
zeQZ^vHLq2d^n|V*78<u}%yAij&Cj}Lbz7Uiaq4+_O#~CZa=T<(Sg6zCs4AYE!Fy>g
zZd?~6$+Bt}_oy_tQ2??n`lSU>^`y-yZg&vOP;Y6{B@LoW)_-s({+<0|JJ>4GGO+%4
zp94W7tFRNlR^?kINV+6pdYoJ79_!E}wiDVDm~!;ls?^7b<tzC<DuIwOL67dpZ+jGK
ze|Pc_85uiHTjqpvyz|_uM7XGO`GuPYUe}|E=7Epy%~rmNKSDppf?JmTFskyVQHGut
zl+NMh_HO40p1a;A-l9Ifod}CO97^Mz{kZV#5dTRPdEq*CIyR|1adIb|<q@J&x1A#2
z7Dl7-F<+xYi?_H!zf9|`0<Kpf7O5~UF!Lq5yvlN)`oUL9Wqao1Is||LW5f;CHh=;9
zfVku#T7q`mDoo)aP9WomT(3(Nt$#|)TumAJZ|4E7shh@@P2)hv)Lk8&RT}NWWyW<j
zpOS~<PBFa6t-Vps5K}V{uHnXdf&=t7f1;4k+GP9wNA$TK5E{cL1G33gvPr;3{fgwO
zf$|5u7%e^nY1!EX5E|WlbbLW1|3t%hv3p1*M^FQGZT!JLbBW9Mu)0=tl)PVvap?Xd
zDRp%}D9?QIG9X)^z%K_f3{vaa-#NfD2fq)1THnG|1SgYi>dj0G=mCkq8*ezHA8(3I
zgBx@Iq&1xK%_X~H1eWGFTYp}BbvM}r;vEPa1VQ?aZMm6X0E58Vl}Er0SN7}ukkHGN
zbT^rJcT``#XDWYA*WUF_Y6%@?I2nh4I08wz-&(|Flo<ZkL>_|+^}$R~?=Lc5eoUTx
zVl+PtJ&Ie>*4T9vSHcU1s|z2fi%?FpXikf0HW8S5e?GU_LYM3|UR4%@md$xp?7WiH
z@h*5cG`)r>F3;eU1S|L_{4D11X}hEQkW{R#zDbq7(A`t^rqE<{&Oao;1lQGryg*LN
zg|HyoDme`P&}+-~$_8Y|wYPWe&t1qkk@1BOUww4^!ie$B0)$pvza+tL`DIfM9gJrS
zug_UH&g`qDKw*O*n02^0zbS<)_cz(4Rm~)%)yX=W|3k2OBdvL};FL$;YfaPD*AOiD
zb+5gdkP>P{+xCV?9*7RA6K{<&nzHN@kL8z!kPaDID>e<C=)RU=8g^>rprbl*$BzG9
z3CJj%-Yf<ZO4>-FlL{OJLo7CC9=b&k4PvQ1svXk7Fik{6W=waY8SD2Ec}Qein`RPd
z{_o-fNQ*#Qel9m&L84~e_yw9m^A|fmqc?_<u+&bj{rSk2D_Vu>GwCDj9{InQk4N1J
zLc^H>hn83DTGTvNwclW<=}@6@LnLYkZ`}s@P0+Oy<XXhrvYv{VVDi=^bG%>^A>2P4
zJqX>z_kH1J2N6G@Trx*hTSkXvnYsX!uCgrz-l^#SN2wTz78_CKBw9d>t+OC~j9^Qh
zWr6fAg=cP#z2S@PQ(K+urNY%XcT96fAFdcnl+=FoRfFtLkC4q&biQCcvkw28xOnAa
zyDjQK4Hp~FoWM_D^}HkLsB5P@!>*JrrlY%WMAeCq&2c9Y*P3$Y6C?@CqOEbuoGgCw
zxTMj@AUu%HwZdQa7|S!L!&Cw)3BLhGpFpE?`+fKK2Xq3d=|?vt+$=77dX2iv1kNTl
z!ynFc5&y!)mQEM%grBM)%!$cURy|YETutlIyyBNA`-(2!4OvvFJ^Ruw%j>r}Ut?Do
z@J$M!K2?~SASLNRug0@rw56{q=JM8utLlsa;bjTV#%VY%qY>y5mQP)tW6PrSQCPZ5
zX1aK7^pHJ=_-Ay!fcn_NlpFn!9TMBcX*t1NJ7<FErhF@7))CG%SC%#hO(lNZ_IP<}
zHOPYhp^bHTA2lo66=GfX2;J&;E@aroqSd#F>~Zrc{OKlK1cVC`SEkf30By=|=-<?!
zXJ`7w;T9eYbCW(yRQr!yysp0+=FIn$@L_s4m2NHM;_>uLg_Cl&LeO(Z){Z~NPQT;>
zc^RqbIXgvAQ<S1Z42tuXZ-800<*lFQSE9us*b!vVxC(TE)Poyrxj5`RHocNS+SJxI
ztQZ#Br0n(1;84ne4Xb&Ai|sMsiSpJDB3F^}6An$Kt;U<0W0aO7MIr;YV4R))@wW^^
zdgZpT3+p%oS2PzS>6~?%KlhkQ?oZM%z;dqzc1m9<Q#790KWz~5KsX=OsjA>{F-?y@
zh12V?%KWI!^;!SaQ)59<4x*H>dos?GnLwME@S&_Z1+xJ%)vXY&cG=iKs+H2h7oBTv
zf7v#qOEuQ1A%r4NA&u?r_^kY7a1gX_v+xa!;YFi_U>I=msqZ1ghAI_K`m<Gq(Z)D8
zsJY80G({exhjY5=sR47XqJ_6(e<AwucA@+^?I$0<3w)1{cta(R`5+`dN$R%to9$=M
z4c(<D?!ChHiVIyY&B$>Wy&Lle3j;*lyf5&c2swJZazA5EZ&4<JH+~(KdST)0$oiPX
zDrFLJ@<bjl&xoK}=TIz6X285s<`M~t$pL~a!hj}(bnQ_m-!_l-7$ie|kVDdeq{qN%
z%mQBMu@19K!}!OsX8;fn2oPe9(LCZ;KBh6Jp4TZA*13_%<~&+0f=gy5IxKudS|YmJ
z39_vB@1(6f;?yimqP=Q>3oZ;Q`0)9P1?#dB{obgZ_J8NNDRuKXA8Gm;kHSki{%p5+
zKMdddlhILUi(P!e>GA4-3;uwYu2*3SUth_91Jp?o6UES#C)kpAd;b}hr2Dnyx>s||
zT)D(E0ObEGMI5D-)pR!jOb1pmBZjENeCZj@ZzXINxOSx!8}+@K<6G_W99Y}W!Ok+>
zr(W|PdL8*dA?gFZ*%Bj?GFa2bPl<4PFKj8gFh3~tS-~(%_4vpHtr$3^C-iNOsT(KK
zU*>xYD(mzw?2KuvbmF6aDhn}dLl-GESnf;jwA%Cagot6XfOCa35^{L+p(9Q&l<e#H
z6lkwMvc943Hv7tY^UQNEli)7m&e!(_)VKAkYUAQ^H{KX@IGP@q?x;sws9xH?prh+<
zc^n%A%>R)upTR6p53OkL_AzWhE?ZVdsK+YYxRF>=l`HU+9-$r)bgt$7r@xTzEzMx@
z8&F}U?84>#p#qoVcLCgQ7tYV!*r(Ziqf4Cgwgu%0Hmx7fS|s2CD8!CY1YlduHp0Vi
zo7&K&p>7nVkMqdhGQMse+hdy2ozDaNz8CeA?@?%7<ByBulae*>5OUrg_n-G1-!is*
zO}dV41`A;|7g4pz5ZdiM`Pfue5#@nVEOItE3TmaQ+iOaymh<f>HW=*o!SL-8Cfi=T
z5r)mw%P^DDgU_{ofqd8It0zt2QeD%duq8C2!ns|_ozHnZ<E)eC-HLsKNv1{8%YK5X
zL<B2V(Ej0#YU^5TpZ}x~2I#wS*G7F@ApKhQ0ol{$1!aPY$t~(W>5Q2o6gD&OTko~5
z)UW&(P5MtrNs7Z3OFwA6mxf|77b9qG9$Y$(Y?AwZKCJwYD){#-ofCl|(<IoUBwL?u
zPh2zU>S6fsH8%5=sOoC`N<JHXhJEW%<^W+zT7Uu=t&GEOj!@msKgU4Z4qVdFb&B56
zQQosQQmi+RYOrIqR+`$;#1~1m-_t{2U1#j^n>Y~i3^5SuAh;!eG<pa1dFmEHUJvl@
zwiPc9TygB&bQFDyuJ*sMJ?rw7y+i=}jFM05HMa>~tQfuo8NSU0Z2JmK-DM?X@z{xL
zc>O3$I9O2mv9GL&fWm`PI^i`D+}AxQPT(OF;Qb8`vhi@)l;iTYjnws;*;N)Jged{_
zYFxP*x!MPyblu+&fWia4vRRIO0wxF(jHy3KSX;Wn<a_n~j#{>fvOS||g#yR!Ls1|T
z@7{QZ`^TpS?xH_t*7OnPcmmiC4h{$3b{_Y_7|hQU5jc=rMW#(-Oj}>z2&jg(<fnq5
zRIpXj4D4HdHtv<-<^K}x-?aa7Cu)?Iv@$c!SAEtzK`&(}0}IrTeQT_UEs6@pRgNn%
z+jmSM=4MQ$j-q|vV5Vv#B^Q{GzSqxA?;CO70dZ<J!P%YT4t!58ZSCCz_%g~K(|i#5
z6yZr(4qPndishV>18(*_5$j!Ho6P{#{^CGabYv$+r7s5_I}}qEf`%zx7m*M!1qc@T
zw>CrkuCLc!IzC=cU2RQ#(xFql7=GdF+S+(gs=hCNyQF{47xj}%_lP(?Uzi*zWnayC
zI7!X97<;?E!zhmTUm)7l@=8H-Zw#vPv>VfM3yK4CLS}v@;KRs~53_Zq9=VXB!yUEz
zqk1+N4@<l8yvHX2h0Ugn?=>}1rhTB;Fi72h>`>_UUOFqH!>!<G$tstY$3RgcEIw`S
zSFaLzx9D8jsTf<%qxtn%>SDdSTxV2fW<;YTwZteI9*muJesY95B`rQ74sy<m_*{!X
zF}&SEfjk3{!her81R0J>y?frFe@n~EEIz^X31i^M?##QYfrfa=6=VcTwkf%dn^I0F
zTPX7&F3wf_-Rv;$X25?|)#m(D@3AQ)$tj0CA;X*SLP*{ALZ0)%%Gn0-2xWuA)|e4#
z0)C0{BlDu06lf^aN_)E@8XH^AfVLuOTqs>(VTe4P@i$oKoi(yY7~#3dXD=ORxL4cX
z|0Blxz^3CBQnJ=oqCfZtJ)2caZI*f%lqz!$6FxleE#yG@xNafYm_yu_1|33|u;;|X
z<yCdsAW>;4=Sr+&AtMg;7rr_xW>Q0*=+7pMGHgE!i$;?^$F9Tlv`J>abg>l^aXP3^
zYVESUsD9Nzy)oO!V9^jhg~73(a~0v6jM_CY{L2PQD5S1L5lZ{1g6x@D|EzIqz0i35
z(&o)RCf@u(*7{v~#RBR&1n;q`I0_p|%AiX?Qkpn3FwE*sx()VZbk4&wm*esq`P?fs
zDG!?$!LA*a0Ui7QpodRn#q9BkIDVfW>ZD={aNybE^La9jIf&Gh^<z*9lC3JHwr*p7
z{pNwtrX%bM!P1N>j+03)Rqw(PSib))LrUzpL%cW$yWhA~kCFHKICAF0VMn}~m-;nl
z2u+RR?f4E2^$jZ8{>6a#h@^~ET4}e$X>5!c1>!w}(ZmLQ*uJU;2)Ven#pa>x6|}bs
zU)&HFo)1_Wa>mprC;Y_;wP0U}Xm?vJEr3>y))K${@Hc9cL6_9H9u}0QpfW&GF($WA
z<7}44$o?rAYMOfBx7oVkG6x-YF&Q3tZ#|F~Gqrp_+lZiNmiX6wIp&~6QZ%`;R<A2{
z$a!@TAJeCUv<ZV+e@9E=pf&a&2+p}DBAMjRUORoi(WV2-Cj{dR%T~f4p9{x@=m(9d
zQhTd5To%(YY&rAmuj03RIOf|*FQ!!xnp&F~lQSA}fwzf+#d%}0FGlZ^b66pH{LTEj
zUN|=(gKkl1s<sV^8AN|-H@f<6p))A?yCc=*cjVDNq~Eu$JP5gVO4e)7O)m$hzfDW(
z<ZlGY$`x<_JrpIz_<Wg+O|k$N2WqkzUQ8^o{r2xik|sj90(spW4M(vl&`+B7wipp{
zugiF&OYM|bx@dFM<SNhOAEz1EJXTajoVw|LJ60pCFr#CbvHodoF`dqeEU<sUfdg`i
zAXB>2)bk*tfmqIBDC*m*P^qWr+l$oLfQ$0KCsXJY^+~0yyo^X`F6C6T$}ULj)~hjE
zKfs^NS+Wv#{qCC@iaiKE^r<95yTMpKy+aJqKZ!KqBo!~YvnSPnaC5{)_RVDLcbQW7
z$SDHSBiC!u13^VL0gIhHX$#oro;=%+I@so74S%iK{o3i+=H)8|M2QB}^}i~zPlga9
zUV<rUb6v_v<%{T7i^`2OW+r10S5L4q#Y(-`JvkaKzZ^N^S8F6PPc{^k*IDG+VSXT&
zxp1a%?6ZOknj7R@X>of!2`Y|?)e~a%8RQ1O(gB|?X@?vQ66C^VXlQ9C(w|cAd~XH_
zd@q&=#`~fPl<;Lt6isG?%X^Sj7$~EyOV#{nqZDnqmUd*>MgQEz|8u?c3IXo<i_9R!
zbuHu0V&|DsQ1Lw>XzEL+<I;WuZuZN~`_@$X>+lyhYA5a?sPR|bI+9$T+txE_-}ID~
zmrUjnJlWa0R68SPeY}@bfoAF6Q#O2=Leg{9yV1NBnU382_NR7)AvP)5c$_y8ykHk_
zyp9$M7Mxwl<@i+ESiV$W7?Hd}P_}-R84Iz0$?z?yRIR0xzvbiAEj?ox8|Qi_TaME>
zilc<ER=FfAi2E1WGYWfXp>gh)hKOL(5#*P+^#?ft$K9YZmW2&%&8<d!8~1Pd`b(g?
z_KE);U2FTI|7^ynaRQVGLPfe+k=;44f;Y=W>!Y&i7kpk{q5gvf=WGrEy@|y>y<o17
zT-BYe5CYM+528<PhL_im#Pq^P!nB^8>^_#e91SFL(~$|%W|$G4dK5nhdZ4-TgZaw&
zH;SK>dSr9&Vl6Mah2#r>QnSIHu1&%38NlP;j7_LcV0GUJ6v93^>mIAGM@>7-!@1P0
zy*E7oeVo9SFRBa$#LXE@2bAJ0UQB;6ISp*R2I#&Uxix5@R5d*C1~<E1>+0fLqLt$9
zfg&4%GxWcE)(QT5=$^ab3Hrki<2qrJ={G9mt-TtEx5wtoi|VBi7YJq4a~qbexD~_#
zWGRtCl~y;%y<XR|>G>uPP5xF{)%Ka2FVNfeWPJjzH(R<^A)d`*6zg2(|6+n42nP7D
z@+Hzzw(62%AJZNF{`d8uCDU+fB+|eE!|o^>+axb4(l}d&Ki>xYm4VBeHF`gC6TS!O
zwnGwGX@8}#y-ayn<@W)qU)+w-px5zWgR2f4acwjAeusc$VY}i`Cy%M&*MrT^0=6S-
zNC&35aqIU3dD8XDdP)SGPx{-?fLo^3x97JlY0(zrfQbHkrs?0!KEO$~*H_KwmyKr-
zJ)e?Ol_6FwRd%!~7^WgkD?()$iDoJ_w|Cv4l1EipmoVT6ua20#-O4WZH{1nqaH)c*
z$w9y>P@GJ9gn$d#M^em5!ko3MX^-c%eM2pqF7NiZRZ?$mM8$1pgi_+Ove6_)7iL1k
zF+DQD+zgXcCkvN?f>9tBD9ZK&{rt<!`i~k`xzY*{_241i{pJW4wQp>e+cWHjys<7y
zSx|0GsD<iRc&YMWAVIakmZ@0)fPFG@OUUW&2$Mk2{~98i8%3snl2*rAR+o7ffp`W5
z<7a)U#Yn-(7Q^c?B{Na)vHtp9>fpSpNt*Ew=iGR0)`hthOC}oA@J<H~S%lOV8?m;Z
zR+OGf7#b<^91b6#LL5@AtVspUShkVh{PAJ})D-jnv=T269RD?2I4I(VOs#A%PMy8O
zmMnvOR(x-87&MK<@mlttmOpHkqoCg!W)5r1ae0Bh3Wwj>DLwcO2w9vNkUtD;JO_0>
z#RW{x6i*&J!?_36-O<FD5hy!KtX;X5t+gcmzy*7`FK;R;^;~<FbdG>nW*Uz}SOE_2
z3zB41O@u5Hy0=E-RZI&0LA?d|&;2*93RV&w<9s2)Ib?Bzz(u4y#Hwn62(n=GXdObH
zL{Ju`DJ|!u+3{$qTwh-~^9nnkU7-kmrI^A%JF_khx<O>1U~Uj_U?J7kr`V}}V=BBm
z846^KhS-6W`6092Z4_N^A`5HToEFr(uaDpReX?tiw+&z_NeVju>lxAgBd2b2$_ZSL
zV%@PipE~*&)l+~~Y^kbJTz=wk61urY0pFjE@=f^YWh2(FQG8+<*plxHsTb@<4H(&G
zJc|U^r#)ixvaN|_W-x}R9lC}n;HP{{FhGPdh3pecJL^zbN7(m5PGy8DkhIC9Q1%E9
zwIyBR_aYhO4wykd{YnvKz$lvKyqTrmTQ|I+?}r3SL<#vh(uMebs0g4H%#|x&2bn9U
zHVnA>)%BKXc(LvS<?!H&sDYqDy!HlC*&c2-UNa1z=O(>gUsWHQA<Ry$>9+7hl;3$~
zZSJHQVrA(eMHwvYq`N<zXpY;toziXU6Tkh;x)Vx_6`~&%ig)Gh@Xwn10vdzO%h+Z^
ziOr^W*JVdL!;Vnoy_4UI*nWFXbCmi@m%^p1@#UvEQ;b7$K)JH+zj71XkO}9CW}Rh?
zL$Ly7QZwLSPukb_$Tcd#vNrTu@3IZ`>yHJnGK(O;|Csp(6~BDqTg!&6xHvT1$gU2_
z;ldKq)fvDp*`X_{Tem`AJ*0{R!>o7a9!T<Fy(_m8hH1XW0vZn5Hzfjp&E)X814FC=
zdxVy>GjM1D+*+ojhO;NOF`gcV-&K4#Lse5N`}NQdeQ7q<Ihn@IvV~d(AG(3O>DLXq
z51@%gHd)y+M1pP~wTe;kJo7~q=I)dt)cByQbSyH;GiC2FojI}~_7<-X8vBUj%F#3B
zY9tK%?Za-iJLeA16qf&2IT3P%i-=UMlYFLQgEW_>tV*`-<L);kG_l&*y8EAU5VlQf
z!Yu;s?M4Z9K0cGc1fD=xXUv`b6EU14HGBhZ^#IYlm2&C3k&5&R+bo{Hi&^#am)kbW
zwRsIsi_;HfH9sZrfx|Mgu345%ajyEG%Nt_cFk3}*7&0nqGv*ocYjDsPeja;Q+r%PQ
zETNqG*i@8<YBWM^<Lfkna>r~Tt&!%;C)=?HAKFI>*(a^NPTR5cK1(qvae~IFidv+?
zbeT4QGSPkG+~(BnQnHmy)7%`z5(I)OD(zgoW9@juV>p$-QzFi!3JnjMauXXO+{Sri
zc;0-8s03B2C^vl9_r!RpztB@^r$@={(8hgaT<W?$Hngp7_gDE+REXvR`I+COND{Qj
zzUj`W)MGXh@hX}{-gYTa&pYVT4e?Y}>Ykz6poWj+dcfKQ2*K1FzAwM>Jo>>?1KHLY
zEq=gB&76+Svt6>MwZJMUL5ngE{6A|&vvQ8owzaCt&_Y|yT=U+!ANer`e?G<^JX7@*
z-heeyz7(dIkm~bOia_+XYBu5-QDpFIIbjl{p`OZpIdtv?473&S$FErWqdJ>kZ;@F{
zn7{L%a2{#@kxa#j-L<??m*Y62-+xcJ-NApS3I@?ZmcJxh4&F0xN``>F60X!ewYnh*
z#vqFzfnU8Kt`>~>C3|8lj<ej&j%0-SS&SZ;pHW)YI$v%6uGLAA#?&;3d|RWo+wmNc
z%9Xt4lB}5JNRo*rqtrpa(L<XNN_4T7B~-uVs&Xf`vn&g=#vj+x5F@V6nrN%y`qdKJ
z7>VE4QW?$q-L$aEF_$#h4a2^>434uEOV81*E~|k&UW`B>_Rg~N{u$|Y7Zv!+a)_JG
z!H^twbzlhe@R{6lQrPc>zh8Q-J8NLj^s2gXJ7UECYcV#f(;PeRWys0fDrQv&Ld*EI
z)&clF@&(^_N+^HEOo3jaS>Qxg=l)V?c7L>)aSY67RVU*3f#%axp>@bGEy{AfO>)Zn
zfBsAEW0zONdNxa_Fz89Mq-#bJ+|gud<HD=J1p&khB1Fr93%KKpLW<;@R<KP9%sp&r
zzot;R<BxN}{zZhi<Pn}~`dYNelu#HnonU~Y5A|FV9g!5Gy|y<`K9IktF<t3jokNNr
zwFz8=*S1r0FNAQf)S{sh*LumxIK5xpec_?utcygqB)gV305jgmJJ3xyl|(%{&AZ;F
ztko>4bAoz)j}orFI?G1bkK{7~I#BoGJvn{0V`nEsv+F*3>BHxmD<wb<edqHw672pq
zNOOo<Q>prZkJqqg2T+JXn@_7<QzkY|1ngb+u$m)lt)S;!Fi}u{G)nJP`B^Y;uEE4+
zjp<Hj_T-h#c^~8wdi;{DvKa-jE7!|JantX9au;9VDn8m(kJa{{Z|ou2^{i(}884(8
zI`xYp%^U~ppQwBN3HpdgLA2lFX<Uq29ZJ@?(MHIhqf`Jahz@pB?P`=~+V5~%`y{Q%
z1@YW-_Z$ICu70$vqS6cXI~8cLz;Ck=l!iu#Z2L)=4>rHy0kvkbHit21)<VnniJMpG
zA_w+9VLOkEFUbXF-Q7A5B8_e5s5w-Me~{6gUo@$=qi-K|KVKicRn1am$&|ls{E~UX
zyR?@pN)GKVe>08!KOeqweWdu0AiQ>+*};tQC1X`N1FS4U1U~1_3e;>}4`Gl~O;40I
zg#RjLQl$q~O$6xI1EBmb{z~=XUk0mz<l)PmYwDTPMB(t_J)r7Z_k<ZK@-H`7gECq9
z@5fQI&;^)co|Wq*U7f4qV=Gc(5~bY*m4JfoPq7T<D%n^lPkz!7nEM$seqKURy0)7Q
z7h@*(ON|H`13SF0Ko`!xmm-V7%mu>}RF*6gT2-_u{H_4KDKA^U%zC0R2s=kCt`e-U
z3R)q=fK0sbn*df2EQw9v#Of@fKnj%(16FkhyFD-!I*EeP_s3yn9^YzDQ2$&8Xq|R?
zz<r89QVHY_rjty}#oq4kM&^hrD7mnV(6Vb#b3RP6q_*9Vb^_l6f<D(ub-KBM5ySD*
z!X@?v+wZ<VIhLA#r_Dr&c47LXyzgj!Qz=7yTJEMR4sw$yr@5#axL+5STxnTm)=e&v
z>u&AmHWIIO7PXO8ejndw0H8qWygy;`e=67gJav-|gwZ~`@ecql3f#%;4(StjJVrW|
z&xO5S8H|f-M*Fwg$mC}8)5b<{`I0OMo2}48eX;#$up8UgweR+mW=!AH$krz!01j2R
z%?AgfO?HUXpZNIQrvhhaN6Q<~co%p)8c2?|Db#;j9ka4Hw?7x0=G%bM)#a(xQ)~$N
z{*A9|-8s2=w^j9RHH&Tlv*{e-@{>``qcjAJ$fZGS{b#7#ocvmvb)Y1b5S5TE?kyh<
z@dJ>I*Y@0DKakLiI0wt*PgHybd;?kIXGGNZHenw66kLNWM=S5(-o2BR_^MWtA3soa
z&6Q%!@Do-#HHWLpT+U+MyR|fcz!<A9g4K=_+@PprO;sqaxM!TC28-M&QhhaX>m_q3
z+hx(w27sDQl=f?z+tq&lrL?O|%NH<GXNx#0x~|XagIShghxb55%94va`{WQC2*Wkv
z2Jm3>b~oPeb-RfaDpI_YjztCGzS^FPcNI-Mv&a_qQ9Nk<bgxN$OICzEQIF*2{E67n
znKM-)UI$rXw<ynT94GHi??za7eKdk@42#pB0EW)F*o<LVzdJ>|=$*+V5#P8Nb2Nyr
zSK;15IMJ7WFD}@vVbm-jZ?fM{8>Ag9_p!{7uV!xBPTxkCi5kc6T<RpI&`rn<Ogd>7
z*lY{}e8eq=q|>*(e-GiNV&GVxL<F{byo9h0W}w<|p3Fh!IAUQ|gm_pc1SZPg^2lC9
z@^uP1hzFJt>h8Y75p45Q+;K*1U_SZx)5;i5P<QXIo2q-tVG^ud3>5|;lY&=;d}6QQ
zgcUS!)NFB2ea~HcvK2_QXfc5we)f1)IpJlcZpP0>yMTsTFc}bRwelwd68F($M78S8
z;nw8{iM=`aF@L$asf{7)h(+B#m&ssy#<+Gut~UH*wq>q_|6GULD(=T`_}+e+Vc^K(
zf+|cNv|7<dCI>L@jsq0M#J>mG-OiM_AW2jYGd9$l<B~Nkf_tPfU$Pb9&Fag?%$%9Q
z-)>8|vOxRd0+Ut*J%ZWmdbn-bhV9MN`>+lyFjJeKWqR8*8dA&e-2o$qy5~w~-xlee
z%T(o4`6a^LO5M7pSj>JiK-low;irAz&Y9;VIGmXI_T+JxkFC>~oX$5;e_`rwiZe2p
zzEzAqn>!Aq7?}g>#28Ep#B9OIhUU!vNpAA_0SQG{JQ#={b~XPAk};Rk8}yPsp}Jol
z;WJJ*Q%OM<rrYv|`{OkKjmo^-pn3iDoLro{8?XQ#2QW=w;t-CBfd}epgQ~sVklso<
ze-VTqQ<<xMF)hTbQOr;Iq%$9x+8Re70n{G~CrWb67y192L06pSywFUrspNR6rEM}g
z8B6hk0Ats{e}d-3D1$2Jgzvibc?~k2>iKnT;ysW5s8dsGGk`sLgPMr17okpHtjox1
zXxOmZqw0iuy0Q0{+ysL{GW6ALM~^9OV#%efbQswrGHO`?k(=m~Ir{%)6};Tyj;YqF
znyeyifB2P3<pY&jSelI8)c8gTHB`q@2}B)Dg`3PFLd>EjaOULE3FxvjgGS^qeKas*
z*h}e0Zo#;g)EU%}Gm9SFWC;UQ!u8{tm8pFKHw7QpK5dg0U&zeM&-ac&m|%h9IcUXf
zS<<IrRhMT$0w5zs>84zfM?=zpPu(oeyIQ-zmZGpGyhy6Ktx+jv-Tpjhuur=;U4zQs
z#hblzLTZ3P`$R$6TG#P>NyhA~CAzT6FSnj+1AIu<F1dyx?wyf8Tgtl=D4*P|eI&wD
zMiw@^r2&&;ngzlimkW)vz|Fj3m)>=sFi0eWr8CiwkqEXOuRUK6si^W{$z2PkU-PWI
zmG!aZAo|LeTXrxrP7S3;$u0grPGVud1>D)pTo6b15}Zy9t);fK)TZORs*w^L->pZR
zA?pl4oHQix*k=~K&}*^gFq<xlqwVwd2Urn`*s*vlTT{@B0Y&$h<>iELQB9-Hr#x3q
z85X24B|6{(?hpj^FJVs3$7_y+>3(hu-<KtHf&Hbsc)c2zS}M~Xj|fi~gOEgj7Utui
zgs^X%7qC3_;2@lC^O(pFn#Z6lGSXHW5UYfh#ljdY6{>-5rjyl4b2oP4B@c#B>x}sA
z#Z%i-69hbvIWO_M<~`yaJ=maAz%{L1TDg%`&%C5p2Zyw{98vyfh&QUNYQ@}n5CO#l
z+e){+Nz%WG`pf(I(jSN`t^<H$=aGVKmM9G*3ZagbuMw+V0-U?!+!PEMSj8~b>+meZ
zEN+OTr}1is=+aJsK#uo?DMRy}VMx+3k&a;J5%Qe<hBx!fPU{M8y>gLy%7yK7_x*?D
z6Wvlvamffccfg~5NemkJp76kftnQsw=I0bbE!AIv(>l<(!3?jF%S679_1_sOVoayT
zTKU=j2)CK5?SW+W<+Hi^0Q0eC)Ip{j&@cQP^0*#%^bq7sT&1e&=5<&(eJ=>sZq{Cp
z8Q%v><U4j6f!WQ64y56cV|BeM=3VAVP*Sf#uu%DbB$zWup*Hlxs_#^S7|90&XU1GO
zsE$Jc-%qS2;UmRi2*=&vVPbMsREQC+wra*{yQZU1_*_+DM(38Gv=eLrXWOWw&1s)f
z|Ba$T5{X>1+i-sTfF(omX=!EunU~+-JV~QTM+01{Cur!-S{LOrO%5LFPc>ZSn50KQ
z+m5sbdEQL0d_~BT%F`;cmzB#nIk|-~s&f5NI+_A04&A*;^srXy8Z%P9=iY9sp{fnt
ztJofJDBY0QL<Dc=z&2^wpd{nCl+x$eH86)lI#2+TSG*zFj)HEfZx@NqAGV<xyWyJ0
z$2!l$WH^6a5ygRl;H9{lqZB>U?FGPY6$S=;gAw(H9XoU7x<wop&qqYt@Lx&I!LT(M
zt0yePk`yQ6opbgtUeFw50scm}O*9l(bsVW57r>LI=!SVgPJpoV2}kpQU8U|HVWHdv
zSQRdB=;6J&_y`RU4|~sKLVKJK^_oLjc@WfGm2O-M#U|B1sO;$jMSy2Q#n1xvYE;BO
z+m>xMmS+{kbX~tqu6GH^CeT=CFUIZW+E|>CSC3beRb-!KEr5;R>5SvRp`Uzs<6s5A
zA;4oc8?ga?T3P)>va;~-v1}`6<f48yE-ItyjYfznNtaf?VOCBI{v{>zLru!lGcB`}
zF^WCl*^eT#nf?r4T+<GbhOUBd^RDsdN$Q(7z0~QgcKey-m{HtIyx@b|%ZrX7Tt)S<
zr7otC+o2I#BdLAr+N!ep6ya|?Srhqb&t;8yw^Q7`Xkv67CikWH)>J!QO0GhsM3EJ@
ztg>{HTdsiL`89*$wpCFC-h9ICB8?RUysFu<#5EI!3;NV#t}a>p+2`f|@Pag~KFG>p
z3}!x2M_goXu<X!~UQR_F38%LTTRGC$@&yF2_*Wk{wz;+qia3x-a!489W(X*~r8m7P
zs;_Al^EOu2RA3s|>*gN*HYYB0c)p)eVxA<(_V?a?GJHvZ;niVFCZBP>nPkUd!5_Ql
z{wskwaI=LEzlfYl9pu9jVQiZsBJlbf4m+O<FpVC}xK2Z#Z(8D=kYfLeYFhk4x<L)X
zkx~6)r>uJ%wuW85t<Qw&GbH>wf>N*<RpgRn<wa+l7{m0tlj#_6(t#px5;}1j^K=5;
zz3aNRY<pf_IV0mRQ7oh%D_U!wDNJKP#GC>c;x}G}(t(|=jR)ikvj|i&Wc}(2p33fT
zF!hLRe@LaS7az_p6Q;rL>Z+zFTn+<2iJ#{pHpID~ja+<A2RzxF5M|L8e#)TY#|j1t
zC~_w$?bEB^sR&Hx5x|O4ntuF_t(`t`l&u=^H4Na6HGDB<XK?gKZVDzPTr3|0*dXy(
zDpxV3j)l%j>knZKGXS;075kwZuM=@XptJkkri~U1?=;R#EK_*YS$l{ItFO1$_TkC+
z=tqx25>^kbGbxMDJy0dIc#FW29u(u9r8`2>O%%%O;|1?{TET+_XbeMP$~tx@T?OZ_
zxsEi#)SkPNM!Y?C%9(c^$oujBiWt0MTmlcK_y7d?2j3XQ5rw|3?|QWXf5jXb)1j7h
z2Ipngsb^9&v~(fd3VPXWLLy!Nf}le&pIDMFosGfLm@~kf`=Spza0yC<Hggk2Sym@^
z|7bf9BEl~8))dd9x7X9dg|KJ;={$-A_#wNdRz!3C6gh{>5h8)V!n4+4U}K_0oxz*>
zT(bJ=8!#Al;RHFd-G`pJF_TnJ&b&l*-R)1Nd*}Gg7~!S+h$OoeeoB3_F9f7TDIHFa
z$>|dqfZ5L8G-4RcLHt|ebA7Hfw4&vsjTf0Y{^9+RASgpaO(Y3BY$7(6X;+c+t23nH
zog`0?$fFUD@W=ep?=+()QDWi82~FEvnKTCuN+T9|!TSd3fFslLxEgd#WqvKu${dYL
zBm2-Rqg6X+9#cB4V4UA^5eJeT-|hQxw?7ImYEMmPat6>Cyy`7Pm<DEj;=5EKX=(7n
zPZKu*|C85B9n>)V;&yRZt5O2%LkG2td4Tqo3O8_S(R5<)AIZz)R2_wi7r*}7Y_qd(
zfLLA-lgMfE0y3SmEvA?MwO*qpEdbVccNA*Abkv%>PX<vj_6kX7;5Viw^H0dn<mQl=
zIzktA>H&>!KY&75lcJ3xQIMA*eUd?{nktgy-a6(Ab8}q+9ndhx%;=!G@YVD4xJJ^Q
z2gQ4wr|yJaZ*3Wl^S!NX59qqHFThAvYd%JaVlv6x)K5F@BTA7MUH=DV$MG<k;dCj}
z8HD4MFDEW;cNQ_mGH8ghS{q8bju`47eR6?%!QvBl_~p;{<Q%)G@!gb4YQe2M6dIqX
zaHw(Ix-=?RN}>ViS=yL;P@<(wEy|(=lA~t&9KK|OQVL&Dtev^ku7cIa<HwnK&3DJg
zk4q`^BL>Q8?rb<0Q___@HYMCy@ALj@)&6^reClkhj}Djf?*a$EMv6IERgScHJc&~^
zdX+PyD9IBrA+)-2v3J&Q==a9r2#rgc!|mDd^WAAz7pV%}Exb%(i?;rY58vPK+DvU1
z<KA-bsr5nYfF0zOS2AaH;l255b5`@m9%zK4-d--P$^uMfV3nfsj73E9Xen5Equ^7Z
zlk-y8MgC32^-0(uBDUZ0Wamc^i3oHM=Q@5?t~5Kk;&`;KDydTW*P9%rvkb0_lg>q5
z3jx?%pFfAKU8qBOy*XgEt|e|H+_8(2!YCtpNNLj7+g~a(LHv)kTd8`iOr=$&AQu4f
zt`6X;ov3vuRnsj*_8Wu4{^!lby12(Tu@5FPtR5JVAd8$uD)9v08^)udW%f1{w~w(x
z@g`^C?e%TXDv@J5<ARVh@A&n$YrzJh#g4cj0ttuLSC8U;!)Z95Q25qsOln+##5`<U
zCpoWfxc)CYU;gf+A8Y8n|1zkJZxy43&aVV2iX>gFmT!Zm(@++68-v>eTqd8I?u0-V
z&e6W!zRsc%-;I|Tb680xgX3pP<#0OM0J}tL*7i4MTGajj68HW=wASRcFAgmxl+pgc
zK9!mGc3lsF3@z3nT-Uy&%^9NYl<yzt9ugMzhyv(!+-~|>1~1+v4c=ieoWC13L5a6}
z^EcH;bB7pfmgjBP+5f=e5qY_%PEAEunq-qeU85eO{nfy2V=dI~^=&w^$?*);)~N9J
zbYLa-U!M6!9GA=O!}gO&sR)1j9#L=1bmkUn5ZHb6N<#i~?Po#`&kGzPJFTurs`z<g
zg^Yl8LVkJN315}Gf=EwlFX!hTzsdxfp#F~E<WE*hXfWhT7(12MfDzM1ws?4TJSd@u
z>IZvg`}QqVhq@*1zFsvmKF<ip%4W9cuSEnmH@CP<K^h>3U{u4(!r^)>s3B}NrjOeN
z#5wX&yvt!vRlk^dm5j>q?4!^s8J1fl==ENZf;kV2$$pXh?t`Uri5M7HH9OrT=-(HA
z<vXc~^SwX~q?n!neRWeSx52%pjdsYz;ke=3wfJ~e;7ruwJaJ{?%W{Yaj0$8WJo&|K
z_)1d$mHWcc#xEl9ZKal8Zz&NWiKsJ<)iL>%KgphDR;5%w=;NAmE~Dtn8A{5&G;2A>
zeq>*tPu~5{O#AYrJW}$r_xxA4ZG!7d2#18)e7o(0YLd3J5^<D$>F&lSEFA7nNQcu)
zVzDSDuD_!(aRy08IL>$i&dDjJ`im8!P7L^uTZ4LEo?((Ny!W#2Wxtdq&A*&vAh1^w
zrly8Me9bcGP@0MA{PW$|8s@e6D4PvSV@cxnc@a-lZ4YaG&3iAG@cn$syOP{YaU1zK
z{|5UgkYQBLUH$pVt|XpenVph6=3L}})G$bkU!iWM8}I9CT}m`htkaWGziD#&mFR9J
zkO5`hzHolDa1ry&XT_Yx4Z%*$S-|rf1Im|)$>Kej!<=7GZtnLy89)5X<f}ks#XH}p
z*AH<F7Ym{y*e;j0SV9C2-Z7^4%&{mkFVILxEZ^1B3v6dkMot!O`~1u?=O<cL*;-d1
zRtaaHE@lDdN<!bg>|MJU?px~*_kgx!S{8B7^jCX~a)tBr-|H~l;ir$t2&n9h6LaEF
z7ux;Zle#KP#)`({49cs-`mhHZU9cs8FQ-R+wPwV-U^uLOGC>s%8xWRHl&~}$l!%Vk
z`88GtrRYBAqLS=QTAUvc#K*@TDZwvbSY~wHQ%4ig8W@%9v&!WsTy76#ru3;m*<zeI
zv8>agh~PNe9q#D6X?*n)DCnwsn>uFtdE@qCXB%xw-f`(i#j{Oz&l{c7@noCkBWW#Z
zd+|-%2>nxSSOXd%|DYe*f~OlQs~13?>uP6R!SM|0^|rTe?xdTmLCZptsSnuML8-Oo
zqrs!;2aBK67!BKwn+0riC!bY_D-$9YeN({|P}sPYo2@VGA#R&U3SkW<VIh06W8-x%
zon|-ddTD!6A3S{0e(QPri~hSx1=$UivU$t)xafHKS$Gh0RsvOUVEwkS=qog-@pViB
zkR4l48t-y5sZLTlhB4Gr?-P|~u#tYvnibNKkMlkvR3FV%fy!Zt)pMJs=Og4y^50)7
zN-WwNbat_?OKfw^>pxWZHW38`?R=jw*%lRmZEw}|BEkGMTrPHR8|63JcV~~>FAvp5
z3$~7~_=UCghH8RLXGMbKG)6ks=2yTQ!c0%n83@1)L9JQDkONsoekTb)4*Lm<p1*Dl
zuMVlO9fvKUQ;~ePE_gjTxno9T%ce2GuozT#uoxz<Zj<m-uCK_)Pg9zbA1TBPj$8-;
z8znp*P3Ap|tQ^ffA^=zWUXXW)mHPLU1}8-L{5`H%8hCR27Y*uF)DLjrzuLOhdrM)2
zcGQr+u`1zufyb{Ta)DZw{lpw!ask~;oMw}JxQoc;rPW`imUO&3s)DW&XGMMfJnf&<
zx<VXHSM?m(Dqe&!ABK2Rh8l@l`Jce>*tf#sy5Aa3CC>nSRjGF_D9H<Vn>YT?yMrA6
zNJtkZYe=^07jj>;-Llo+vqE@S{<i*@E0(vvKGfl>6;dSHfbSQcDH=A8UClZ3yt^g;
z(M2K3JZYy(X?>ncpU=7l3t}e_y4MV}_k3M6m6_;xa4^-r^>kl4GS1eXmn-+Lf<Af1
z?XGa~4_mnoQ*_np_qt!n`P<)hb>>k&<u|^0NK8x!AF^45`8?0gr{@c^dba~g{@osP
z?v{Qpyrv7|^2$#Ch?_nfejAn=rF35Q{cGj)&Ys^56Dyw~uDA7<GbN(_nE44$p5HyD
z>FaIr9oh^bWO%j>KMHX$vkeva4M~Jw$UjJS?#oD3hMBvYo&cFbdYaot->)e__j)=+
zzE=CMy`Gjo-7h_=nDA?fMm@QdK;%1(4ZVVUuNnw(t<i+CnauV5<SH3-naPJt9Wkhf
z$o{CsK|p|U6vd(O!{7Wl#%rtPbJ1&W=ZzOf{PC_{Hj5+(%Mec?>qh9q34(^&+RS&5
z$;V#uA<UmKzR|4I9fKD<8)TLP6xw{icFWP6{f+8vErVQL$n#}toqpBXJEbNTcF{o9
z==Jv>dy_sE`0fCFDscu}^i@G{+0K*oNp|E);lCr6Ol?#P8@(q+>^&7)s)MbM=UKoZ
zw_h1s`5Y9Y>Exnc+B=9iA!&{jMctoP9V)R4IVhYRaqu2PP}E2Yn-hBC$AE0o2pcZ<
zUwTlwosvy3Q#M5=e7f`z3R<PkD8{R*bWay<-opRCXUT9UVlJgxaDhl~1KuePa%yPh
zN*Lc|D044OFQo2=+R|QFNuF)6={XqG>{Wd;wRdH__v*USw{rJ>uu&t$#dsvN>7&Q^
z)p(=BPW!)EPk%riEB)(jIQn<>OYX!rqPpdG9#pJ?!9_t0uaWR=juC|ihlTh**I%6R
z<7`azHMe~~5B~e;@L#m2SN+NleK4k@?ae?cdH&<`l-GGdXFyJSn3DVN;<uSE`OOF6
z$<40Hvswf>c#D=F5aJok_5j#V#-0H}z;WsI^)3j(N5==JJ2B<Ol;mX$)ObEeXu2hU
zJQ{@Q>v<nmoGM;k<z+Y#Sa@MzDoUeHGT|L)?<6CwlE#5!zEYzxf}&()ih?-+Kd<-p
zDWd;i@bm+ucwc(Yck`6P$fN+&e!y`t0>Aua1#JSir^D&Ud4Q7d*Hn}+Q>X@7Wrf2@
z)w=exaQuQuhW~20|M~iG<?ZoXap-)&D$iv;rifZ_nVi)Gy~^#cnepPEg9mi^^SsNg
z&Q(%8TF%sIxstVsL%@^G+_W(yQD4tVYek_|+Sn^H{)zTBpRpI<-kGbBHeb>8@R&Je
z*!Y9W{k3mVNpbzm*1<G<8(r__cs2*Zr_-P2eJdRwTxpi@QB4ZcA``5TMvE$r<<O*Q
zB;|mH4Y{Ql#=}1zI!KsamS21zo?ZRbhDJ2M5y+b>!%fdkIx7agKDy<ruRSG%u@8;E
z_--zAJ=2<(>xmFL#(F$ozBG}lN@cckc&qNs1U>++oC2u!WjFzB$!_I;?f#ZMUprq+
z&HZq1S?ErE!9^-!Pn7*zSx-(=lN|I{3XPf{^Cd%~;Cr}!P{#AikeD?I15J^&c+`&y
z6X96XXd@~6WC;_TRO^Byc93kkOO8+ZgkO9`bNl$<PT{N5$Ll?*UsMqMwStLx+K^ra
z`L_TP)weDkzI#vx!JooMOGjMHcacWiW!b2_lwTQ+(Nl}j%aMp{5GB<{CN_80O&J7M
zJmj1NX#8J}VN2}s_70`z?LI;Oi!CzQcB1q12@>#rXwK^%b!x-;VSB>Ufa@C`W&eo*
z%bu7zk{Uc6>W8xxljSI7GRcdHO0`sGuovCHsCg)WUzH;)I%E;uFj$PO5%LMEt-~N%
z>$N@lD3mQr!Tu5DD`R!?Q~k_jrEd3Fak}KFMiz$_QzfHLG6NlXtVjEk^c$g0uA1Yb
z@9>=uLN|RA@0~xW&c`Lg*{X;BBN+E$b7u6lZR2IfYa}A~NQDMFKPgm2v*@5(-3%Gs
zifpqOfLqP$j0VoAIhZlzX)P5)8IBAb+ZzOX+tJ7hkt%`4La*GnoSLb3`^DE#sDCym
zt@Pi8)xHfJh+XSgnXhcP$j8^nTGD(7H0ra=>tyvmg{m?`EuB%Sf6gXhRVZ0Ctan={
z)v+r`V}{E!SZ0$`Fib!^6<AktW=lTYTCdEf;jEHTcRilzu8XW*ys(L$pS=#ZJyEm8
ziPo}OzOF$<>g<zsc<1wk9Af_LwWy!Fj=jVce6(|R<o^KI5FvM+*4U^Uh|$ng`dPvE
z$QldNi*86<J6B$AjnUfu<KUDGNpu?_^{3N$xJ#7nd!7g%b#YfZV2;F73yw)Y@Ibzh
za%QMCR2JxDuNyNBzibhb{%y4V>6V~<r0b^>;st6O>en(Oq$3+FDo%yjN=l89h6Idt
zGS1|2oAV~o*Z!HA`rx}CRy~;&W6N48=*O*Cb7>Lsq-%M(`h=%2;&s{{kLEMKdnWE=
zOyS>kqoW=8=&{zk2~Usc{+*nzjLfI+PQj67dwp5VtK8Vn-jati3Vk;Q&Ex0GgcPMy
z!{G<td=1ESf)G;t*gEZ9{o;YoZZER`_uM^pU!GTVneVh7antyP>;0Dm{^`piLtt&Q
zwCZtq?4x|$=X_62%QX0_Bw2=5CB-`-FJEuEjCd0<_M^rr-n*HMASaJud0=70Sf3e9
ze^Yo?^c~<`-bG+lmsFpdQrN^qqGX5)BMU<<{jf;yOGEs@tBUhWpGcifl|I#*n|)$z
zkcVqymHw<PJtU>pHM7Di+{`oKaQ)&YZu|cM*FY%0hjWgR99$}JcRcnH$HH+wxU2)N
z-B6=TBOr<=-G@9;qIc|5NVIUhw!3-icu&3C7DSvCWiCB(*{Qo=N$>8wVtx#kMQ8tL
zPuilo)W8q7-my1v04>~j`SdAJ+8Qbix-qRF=(vodk{Q?d_wU@z4ZD4K#<TCq^5+Na
zmvYb!R-FI7%XWG!4<8G|3u0B>ySU!L3qJncjM|l#qU_QYe^L5q=)g3)!LK$qCqG{n
z)^%g<vSZb2@t0#Hr^uWGZF44N>N1$&hXZ(WwC^`J$!t%utlz#}zBk?!?7j`PdchtC
z(sTn*j-RnR&)vlF{!)J+j$gx{(v!IHT=yR|is3!JZ=ODXthxcT?(sZWB%Ww{C-&}e
zqNYX;aGHD~l5oSwG#osNKHMMxIRxh5<A@qi#el`%y>upzb->qYn%%?B!N1*itWQZ5
z9Aq#FuReDm5IZp7wU1A^tgRsgS0{?}xb2rSjO#RBCvY2=4hme^oqnIkmA4y!p<l#T
zaeRGuxBnRz{SQGHU(B*m8uY!=9Q=sQ_A4{?WNYdjr_PpcMX4fGgkrYG^7Tyi2R!G<
zG$m5@Slp{hg_vDhcARId8P6%vnlUH!LiZ-DVjtUX&32#)<kkC>Zh$1+!1_vh_OD-A
z!{hvOmXoOMjfS0BivQbd=t_^-&trXnA_F4%15be5{>E;<@XF_F<S|!fbf4ze3;7Yr
zQG~-Z`CyXj(*|=lR^M+e<NDBjq}1~t5$5DlZAQc9=CL=cXa_LVDnni$JJ7s)fA$zs
zp<nBMCinX?3i}kCRnz(pYUWdFL+sLFWG_+2X@2`bd@xMx;GCc8AqYXeN#a99+TLjT
z{0fXavh&r)G&<z9hv983+h@1!&wT9zDD=wqZrxvYoMTspn;Skli%)6I<+u&ct#^-F
zx~&C%Ip#TYs55iUOgS;7%wd{2%}2s%V~)|~{AuIz(x-eD_dL6`mrJn^cWB-)-8I4^
zj&I^s)b|l*`*n7-;_oP2LDUh)!Hj6vbJOTPcs=&_`}+suxL?b0P35yMyjq{f@t$Ch
z8{)Zt_u+MudjI}_`$JASaF`spGxG_LagRsym~!;K$EmYlLllA}q`A-CU~G;nsM@Dx
zi=pa_Ix&d2tX?G0SaEigYw<h}>UpsU7Dj`tt08dqmeHRRZNxs!H%Mn(?+i?5$(H^x
zuB|q&cY@pi(Yuf4$R&qtpS?zTy_4-b{T|1k-NkDx+}yh_C62$YKW%*lu)eOJZ}p7O
z`lj9*lwg2z2tG#LRGh_8jWh`{CGJlX(*(bJ`5mQgoU0)UnSc_ABXKi7kORWI_xA)-
ze*3$hy5VG_JU??cXMXklH{=-DUd}9~@^smlB7Fbh#N($6S}VDa7q7K?p&EjBiAIB-
zYjP|*pvE=f9hBHw5fkq2?pa#l<z*#Fc4kM2<d`U~S@e#LbdUzPrmOwu>}j=pwjqx{
zxp&L%vC;odZhjX+xRJ>9579qIwFcF8&yCg%!9q7+KFtRn4l_}NIn6%lslibrlZAbX
zq;{jJHe`vVlA3Waq4a|EIM0a5K4B<DpLV4_$_}iNyDonU2ZBYPWMzU9Nd$Blw}YEj
zijcZ@%UWxfJNT4f(aNf;Cm10yJPIU6LegHSy9dg)6+heON{VneID?1kPFtc7q7Y{x
zoIT-tRkrM)!eH243vF;}dk~?UYkhf#>v*<>E*XVdl%Sxqd&6l@GD~sLAtxcnnL`fz
zc)76F#@YFHR1H+?eomeGKTnb4F>`mCvDSEeeB}Ic=Eol*VOx3kaKfZfYvt+Fv$Hva
zaCb<&dvl=vxUx25o@S<;sat{Og_vq*YcN8#Cy0%msWDVU+7v};UAk~MBt(qMWkXXY
zhveXHPLx&|t=nteJR<Qn8UDj@yr*Uo80<adbsS%rSHkDJfOpv&RvKC=t$2q!bWf-#
zp?kIo-@SW>8LW%@eZ^)Flqx%>LCQX<b6pm!R;GSYVhqT9pf>pJ$LF3{M`(>o^{#UT
zYXvdmm`+IdBkBrk-y>oq^|5otxJ14e98y3{SZkw-QccP0g(){A1T47sj@DkB$LLC=
zTK&59<X^`WvuGf3FQ_9hocnKu*sqi1YXMBZmp0xjMr*Bi-oqa$a=w72dUH=Y0}I1}
z$bd=Qhpe=%q9J2hXq!-0VTpkv!bKcJuDAqn#4@V9xj%GfqI=}Myc9mgL{xbH=Eyun
zQWSCy%-Q{wLYn#Qc_YUcpJ*q-d^j@YiLH1nhY&mXd5vY!;lqYExVnj!1UHIrYBcfb
zb~y={vRpPM>5~&BkV2&N>9noOO|K<U2eF1zxb^+xhA{xVdZIoDK3_$!$=)E@lM3@Q
z*u>nIuPN4Z9=CV~OqpU}5+NIi1!B`3*&fqZ)r8iZB^A4e;~Wx)l-SnBR(vv;5PRXQ
zG~N6;XhLf%NOeY|L_`v5ax)>e$3=*k+!!x+@IA-O-87+5xS$S{2e>r%LJT>fV!MBt
z6J({QGH&iX#fcmza>&pb(INI?U$sjq-MeY*Iai>{nIl)`OwX)#X0+eGSF!y1u6+^9
zOs|bo{L|J7g6=s}(+)T`-C2LFDy0~gXjB5V_8}@5SY+;v$%gUa!&|1qfzy<kBtTJ;
z1rA~2boUlJ+yRl7%Y`3Qk+88;V+zjRmFAMP({y6anRVGn1*RC?|0=jWh4*}&RUaBi
z3(y3M4!BtJp+Y+1ID7MCTMLqW>TC@055%$l@i=C*oBH{Eb8g>|hx}i@isRc1`?{Jc
zS}NjpslB1*&t(jO!!$8P_a1E9hN&^l4{n&>x?B?Rv(y^lej?9-<3l2az!WB4&V}u9
z?Nk3UF)Gp;+xCLeu-bMuZAO^L%<+h%z(stTsrw3NjJ!9HTW|m?_RkiHOd)YhGZrIj
zFsfFXMhZ&AC4q`jO*!UFPMLLCeV9*h@HtBMIL;FxxxGhB?xiPysZy&?7mXpfmtOGw
zF?`k#9u2)9f{r*g_dbh5e#n;>$6wc<w!Q*bU)RsJ26MOJ-Co0|Pl*bpDKRQ#Ode}W
z*&0~ELf(CIl&hDN;>p@Bhr`GzkaHkLL5uk&+9#%5f*<IQFVDmv)boYfjIF|y67w86
z&4FzdrWDCR(AEe$z29E%hOcW<2G!zP{CE)%V&MM%4YID3<qSkmtoy$)#itr->kNf|
zKRf4osM-}T8DHtN?AM0yzrU5s^<VhA)ls;3XDgUX07f!83b!CIbyLi$Mvj4uk2wxF
z8$%Jctr1abD414+#@#Wa4c6k*ZbAn!_V;9dI0Yb@PdrM&r=fX*7I%HTvs3~mO4Z7g
z{6M^{D<7YpXvc}15>vhdoYf~15#sCa+hB8uu#YdVFioE54i=quL*j9%ZTwP~hl==N
zKR`^ubMfm@`q@3a(tEzFHIihrYtmy-6^bfRmC`4Cw&wC+qf#wkF=Mhh3&-44dBkzp
z*+s%}p4gfLRBhWhpD)Dd;KQfK8EXX-2U#!Y6$=s4%sge@-rw=~<RI+ue5h3k%?9~m
z57#SPA>8^kNJov$nona>e}4~m_cZmQR&Zw64klSI5KQ|NbKB9NKbAz@zK>U+^1jb;
zEsp8W%y_1Hm-5P*a=#5R>xp@AcA7bJEQZ*Bpmy&XL7EbY<S<x?R~o)|js5$A2qF12
zqM5*XF<Pm62w7)dnKdW+_hLLC<TN2s-1DsO;ogDuSQNZqRrNSFGBHM8mJ3mpoI)Sc
zaStK56r{g~Q*;7^fCRfTFjc#cUnNFo6oe3nk?y1CS6{@CFof1dv(_hNjUC?|x<YUG
z-U3zoj%xNt@9&@c+D|Imv0GaI!lp)Q-VDIpXnETVlZJ&%J^*$2miR06$7SF5M1-6Z
zIR=(>@xn|i(kg9TXywQ}N1n==ZQFRcynrNPJ~~4o!`fdbMh75Utv-xEu8ck(sc=)`
z?A}+h-!s)lLuul&)HiP)ZeplULfoI{eod@1dp?ij*U0nhugUc_#&7oXJPc6!dLF-;
ztl3v<t?!WzP7U9|99>!~K&zE%iZvlfBt@U>XZ`gJu&P*T#6u*fK%5hSNC*>JFRVSc
z*=5G&%$f1#O_UH|3Jzd3k<Nm4Fw8KbIAb<+W}V)~vN_}9<MT7icHuM~Ii{KW`3-(T
z1d3^A`nWVyx)-8Z@y|YB`%V!v(>^D~3+K=iOA&7@UH5?Ynj!u9J^Ta5ui0x}&X*F`
zdj{1XVW0Tupiwi=v#TRmG!?dDgmXbHASv(W=T`mfsu%=sCiu~2&Y(<zvTj^j<C8Wf
zZRFa>$@8|$y0C65rTTE9`7kr3m25-hC?pb*=zycaev`rgq7AcgR`nqbdWDIaHkv4n
zN)ULszwb&v|GrxF56AJKWR7EX|3CWIcK1c!J4P<fU~gZ&(;VNrwuZ;A@jL6r`9Z}g
zOLom|hvQ`m&KB3=HYO?d(@?GNpXM=M7lItz#vs}?PXY&jYyjZa&!YbPw}v3(6v<7<
z=FMv)AT9yH2BWeMTu9<<t8r~q%!hM`kYeKT{LFbfb4oL(sd1@=DZ|{Kq0x*o-IJ(&
zzw=LbRZfgyHCIc7;b-UFr?D@Y`LyY{KkWv6w|On-?9YFf5`Kws{B`|l>nniub^Uzn
zw(vEVku(Eq#@bwmv{ogGA+^z}>$u}Vu#~9go%mtk+WMiFV%Q7LIR_3?Bu1gG%K2sK
z1*MUrVN}AS5ya5*f=a}~%;7F`cbvFCMrsM3bccx5inZ>MB3#ua^4eEwd^rCe1zKm{
zxMAk|x8D#iXCGxNLhDm@u+H`nLtAlU$Gv;~`x(~nZOE^RaQ1_io`75r)>|U)XRm4O
zW~hT=CIq5|YoSO7+WMrjxqEm;lsR~ZMvGF-&32m!m!%P`atOscqbP6g-(VK_@BjNB
z*_N^=r#F>H_jDDKMUo|wOrDV82JHkHXXU8TDOSlck#pwh{6ae~)Hf%N^UODIPq;~J
zBbfV`K~I##L9h`0aBZz2)r+*_aYDqsHZZ!E#PHR)1+aV?5(p_}Y=0I80sHSbKJ5!U
zTt^&tFv5hrySp|;nzB?S;JSp`D!M97CM-@676)v*|67BeAh~}<2*Tankzz*KHkQl6
z<I~IV2<4PuIZqe`Go@9Cd1ksx9OjAXF!9?TUf9-35+R!VyG8Y4YoCfV>f1q~-;1~|
zauVD9lC*k)_x}BR)^+1@Sx^G4`2`WAPgeD5J1Q<YwJxptyjr@(@iiiM^IC4>_?icP
zUK!tlTH0ByrjWYi!deG?ZGcqScL{Bt_$o+oHbb7Ax!C%iW9HuauG(?tW}Mt8I)u#m
z<r8Jy_5^y!c*b7a%({=2aF~u*Fsii<27`pyr4X@Kp3dkiT2T@)20lJN@vM)W4w>V8
z;55B~5IAbv3s;E&LWIy|2c1DtYo)e8Z9-~-4FE67?!h7h!0P%IGmm3Ih_S~pqrM6^
zd(__<8ei1%IQFAxsO4*aMAqiMcDZ!#krr9ixNHUD%zSsmo-SxBKC}Vd#qcp)Cqo{l
z%q2%&UYy;m+rs53@@Y!Eyc{?j5-;^kZH@9&nU8lo+`sXL(gYrpLXOUms@q1;@$Bq1
z?mi;`Zoo#}ql>1^nFAe={N}?4)>2r?idm24{_lpI#l6yQPS{bG-83JrIP=XOvX@Vv
zA)dDrCiXAh{?}J+%>g16#GH2(1VS=WbO}yf{aj5%BxI($<Z)})GrO&gV2%8CVx9ws
zlQYedCD!xGlUAQLwx7X<+Qu+cmINsYGc#E-i)^0XNW+3tra}x|?&<fu6K1Gd(as3}
z&Bq@ob>qXsTON-0ynFbDV#ZqC2QNowMK^<xeBg$y3s`hBc%(N#uS^*kY)X+`6&U(<
zBIZDc2&9|au)`oW;D|rGzQi~lYXH%;<2AKQ{7dzAQ&mCNMyV^6K$s6e#Z1xS&v^kX
zIQTrIcqvCx46Ki5HdW3)Rocs$`h4bee;`hQr^^#-Sy{@0#mwRE#GKDe_`V_vNivD#
z9$3S>t51{G-8J2T6|p}2qji7=qiLmDqjjL={ksort+3X$PZ77>`r}m`@9}86IlJsv
zuFDi(<!s^Rs$T!~{(kq(pud7#p)9j*EZ@Xc41G{cz~dUj2h+?Gb=?qc<T#kC+GX-a
zT{f1wA@YGd&7|N>;z3p0{aIm;*&ux)x;8?{%qepqFsYxF8bD(5J$>K&g_y*db;Cg}
zXx~pN04ZdC_xJ;s?aa4t-t*>E`9)p1i$ac*PZ*40uP3}_x;7=H-%n@w>2+U~y^XaG
zSTSoOO{4Y3T>NY+l~?!2?~(G_jNwa*<FD&aTVDaJuj^-9EjaU~0o9zP(eQE0=dw|(
zaa<Eq2pkl))>zwyE*lMad|B99LvrX2!RoMllr9DH0`{_PJUzYmXyE8F8ZjdUB?sXy
zdS^-RkJP3twQx+4`}v*^5ATRFv$Y8<_$c101XVAXL#3$z(PfVkD5_46Rz+GMYb4l_
zXpPVoqWCnzLx|j^><6k{yZ_TD#yXv6bYFgdb>F5vQ62~FNR}03OlR_g92rxZdh*u~
zjnJnRnHjA%T5BxD+jUh1T0lx5CL>1U{QSa4QOfBJDMiF0*1W^#{};$Hpvl=%+MsL|
z%fggr;^E+3T1&KQ#FR);i88vsTtPdheGx1X$mBRv(~9QBPcadq&IF>f%z_`BA-XKU
zihFE){P@T^PrQG4Ppxn+MxG$gX*a_*aiA}x3A9T#)DXRPeaA$ZQu>H*4XE}cVJ~F-
zItNWog%8dYyG}>!{}HEW!{{9UU1{{6=V~E(m)5K^uaqI=d0JOCZRFW0=t-5WI-n(4
zu@EU;l9MEXHK1amE+sL)cd1Vq{Pgrp**2%5_eUerNPQxj5!uX>rscfw@a_%w5BL0=
z`~%On5SvFythltKb#PfKYMDmFLc&BS-Mc9$geGo6tTUzwDH}-wZ%;>-ln5cPwc5QP
zKt-7PgfubTpJ}-Tn{L78TNa^>q|rGfDp&S}nbNEgdiT<xPcQ6cd4K&70^3&H=y?^a
zIS4_Ah=Z5FyT8VeU0WQi)kLUnYmW)7!CE#1$mztCj45hoe@OR0sXjSU03Cp2ms^Q3
zb?>W$8>vSOixs6ybVX1(IuL>TW5f<(HiStdp}}(Apn-f-KsUB^A*YPwsY^{dh!OoY
zJYFlRu_wc^gHzobtOI_v*8cbE;D%K<5GE2dc3?9Sv;$uK92sl026oBK7r0uN?NlAH
z$KI@YURR1%BxE<^F2-6nE@k0A`N#j1x9>mj@&ER(S)NyNNFa?63n3X2l_rg1jjbr_
z+IW6GyCk<3AAn+kvM9;&FgrsvVCsd=aXRw;%{#UqU)WYdUkXmlm@xHN7V8skZ3d0l
z6ww~b);x|=>)tWVWs7VZk-)?0#8eA$i&SmY?h7;+#xVqfs(S))6UT66W*|t)&hS%j
z-VE0m!_cr|C{)>t>Dc||=x0y}q5BcF-CL}b>axKCQUla3sYwtl5-f0kx@Q){YV#?p
zUCJ8NkdPo`mZh>42=U}HU+tb#<}&sv20}1I6zj~Z%#51SOgn%OIi`sui7Ewx(2_Hg
zO$rJW69@_E`^p$ep}=Ow#EAn3>P6XX#nv-2DQP-7*csyPktAV8LL!F68EXcuy0ky1
z3Q-AR+8I}fQgj8YL=+QshK?wKK+p*xB0)$Rh$d8R`xH~;`hLIESZ<*M2WI{L`uQvv
z&ui(Rrss&IR;nsntNhRZ%m0#J|J7gezxm((OaA6JzavmQpOcM5BejBbFRk<Q!t>J$
z%gY&2C76;#5UZRQH}ihh&E<0&<YVS^eBj6F2_kF_z}V_W3^0d`SfiCnp;DzWaUw|S
zLkye&RC{halZXUVXQ~!h3W>no{fW73%&V}P4<T`%DSr}U2*fTQjAq0RUUr+j{cgv8
zw)wc;&SbQRn4z{h@H9Z~;rwpwtRq$RY3jm0q^e`RpowTOg~Y50-72kSni;3#0p`gi
zblnfQiW}Qk3Z)HyHxP0Mo>nktS4Rupm}SMJ`CuUfk&K_{MhuA=kIyE?qIhg22wI6+
zF@(|rfjf|higq@wOTVAjg;*meJ+eG)g!0Vu`yWX9hWGOcV#MZ&*5kfqKU=o764j73
zP*b1<>1?v@i)TiNM(~leV9i;+_(bzc@g|zim_#f_qDqfr_xjY}thp%Ri;m;3>rY!>
z0j#g<XIs__-$wssijC=DwbD#Emx6{!CbHJX^H$herKoaQE6rfay*%tSpM*Y|H%9X0
z-TU)pMO%T`i&Aql@g#vc2ss8wk*yZ07N!_ErUUn<JIo@VUVJZYs{2$d7$aw$!Q%x^
zpqU>$u_>uJ^Fhd+`oH>z7a@nhAtpAecUb#mlhk$I*J8KrU5ndJuXmIj#0nO!6ePL%
zhQ4VVc0HMQz>C+f$JHaIHAPG79ixD#gE6sRyydd+B1$a>R8nVh44H-(uTcVNP`S+k
zEInp&nwaK9&Jo>|&3rPQC+^pd=;*MGVqYv0NJPnK*pOVXZmJ$^A{bP0UmExJ5=L`e
zn{i$*&W6!SGofvY%Tv@PN3gSTLd;z*Qc%SwV%~s*?%CF8&Gu*3UNPjnXc3Lj&1_|O
zjosV|e{TA1VRQk%tfk^fQB~=NyyEOQU?~N)#%0ScOH;UPjpwD14j?JIq}PlnVW)e0
zH<MmC**<3bvRr7T?o{zn8b&eFFaVB;qR!TuLgek;19MCcstsEh6@ubPX#ZKw+)&yi
zy2)^JAFtNTnGY%i%R~#%7Ge~pc}BX=nyuy%<Nm((ICe_qtC-%GU0E-lF5l4VaX{5@
ziO!X8;NN{^ZVd2>>u7ZReb?HsD&2d-zaBy$nEx6@Vhqu@kJu*xh5&J}#T>LPTP3GR
ziiw<rY3i}5kW=Vpyu%cI^!*jOF>(rtB!2zWoc*Svh@**qJ!JOJ#<iFS=?pGURF}38
zr6AToeZo}`XW7!P5`7<#;Ql;nx`R`WmfZkYeL~gP`+TyJ(inbq;sADVX+;p}X2PEZ
z^528i`UIzH-CywLGcMbPT40)-)t4%~ES1Mg;h#S|@cmzW&-8!)ubnY~Ph^l_gy06<
zic-{lSV~crW!Z;=B!L`+Rw5P*G<M^4*+I#9nz_Hb=ZB9gn{Kc+NC}G>tRd=bhJDXR
zd?KMZAcsM05JxO)M4CZa2qAI2J0LNl7)#mEa-Bw*WXMK`l;!uqO9zu(@;3Z^#C+d3
zLYFq`fVuklJ)Zk3U1JQdu;zyCtr=zU_cOKG*=Ehb?jX`7K0Z9<m;!ZMu|X%r^M)W1
z36bs6{Jw}!HS~LHySy$AqkL?aN+u>+&(fl~%y*L9z2iL2Qt;war31i#h)V(U*wH$p
zw3u?w%p`$Jd8SCk%Ib^_nfo9Wu6iKJes)Gb15CZCQEMfJ)`6ID?Eo>FRgBilb?d%b
z&hQP|&tmgtX|&+aj#`&nvNO1@=e7Obx&7z79C-KXz9au?y^(p@3Yak^Z;+JQIB$)Y
z75?!*{b&3a|I7ac|HuEs|CQg>M?&<+$U;A7&8IADg>~7uTvnE4=>q}a93h8BEfAw{
z7Ju5aPDCQ7`RJ_lKxapFW}Ze;n*HA5_ph3eB>>r<2Qie|8?Q>X<c(8JF5O=@ViXQ<
zW@;3Y8oDjsQ0sA(!)I}91CZ{`%N^$EhT=AM0O@IhpB*wx6?^=A&fh{hpli1Kcs0XK
z>id-JXn`457HD;*kzyR@i3q&BtlfGbbswspXIb52^XgS+-Rtf8KmR;eU{)k~P6gfD
zFZaPp=H7d|6t@w@n?y|m$jK%B9iIy#G;LhAje|xeBInD6rNMG}q@+YpZ_<aHQ0@6~
z>NX>ym_jUtwk29Dh$RAU3xik((d|_pF{Xew=LsSjIwf2P<^UEw4_Dk4U^JD!)HwdS
z{<QTK!1}s=z9r5mC|m0k-=5e7p){kJ@^V?Ql)0PkxGaqyA1`je8|f6<KuUszMhlIY
zT>c_`W6L>_Q=(nAT?!O}Cw+4iW^qg_C`b-WY2yAENEmf_=HZw~>BOhUkCe3%nh_-v
zrUTO4C{!(hCHiqe)zazO!g^ki%ffVUvr<!mC_Rb94~%~PS!-CWI~xetbsW+e3DPHu
z44wPv-g;pk`U?X1q%Jetzcgw$@T7KlOWaN(4Bbdkfy<{CsBj1;KWsxrq974OomFK9
z5+d{ANHk}gX?4#sKZ}$skyR^CpPqO)+%rx0hzYH9yhvP*6?!$Sy(qO_TsHwDFikT#
zX8PbWV)U+xnxU~mFk17G%NS_&Zq4y<&qF#;w}}*m<#Hji5jOYZ2`6^>p9U|sa}4g!
z(mfyC4n=krm4!wv-IJ+RPpH+CWEEvF-S)0rp9W-M_zL}zkdUk2*_X5YJ=fyY@^F*=
z7j?to^SUyHh=jytDSUcdxI>tZ5Mp!&WG^u)5ON^qfCe8X@N&NJ@^Z$S^$9vKML13F
z0~5tK9S#uO-=S0`Y?UcSBEo5k+|Lm$g<2umgoH@Ufm+-<OU(mP&C}$j8&w6<$aYzg
zMVaQ<iBbK&HfLi-Gcuit(VW4E4R%AnOg#ba?k)}vKre`;OJks;@cTa+MVR!;$~riB
zWuU!Y+F5)}8>Ke17@-9sp0LJ{m_j1Qgov|XLyTkz9Pe`nWPP7e@hPn-M4q3YF<O5{
zp={d0sy<-}g+9^By0kWmv#ksz#Y~c^d;B^3N`evNh%ZBfQ{)7X<6zKgVbRJrr|)>U
zzhka5DZ;ugE?F``Vj}45zE^+*g=mDB06}&0*YA?&5F*>QQEQ{NK513^?6YBTCEzJg
zLh#oy#-{gq9Lr6``=!RQ--BjGsqVWHV^@(IEVWTlrbS2yxxr<vyj&VD%f@A0K%Cv=
zL&Urw3Q34l;IggMt+Ff|+g1lz8Uh?+;4TGDIdjSriAYr;gys_qBfPzT;COn&k3XK#
zwK9i^EEC!p8Zkmy3$0cMP=YvcCC<Q0>I}`YZiM1PVd&<>0Z8Jq^kk5&*RI_!Q$ip8
zjyGtoUNe5xMhz=AfGfYA*OrIJhm3EMQEdA6hYqF$d0hmL@@ri;%B3(RkKrjMLJFju
z2?T0UH?qb^IXMWqHO2(kW)2j#+7R(V(m_6#_t1Q*=kT1deh!-&ff)<Z3-N$eu-F0O
z*dP>$8)D6;$|?>JhlYt?U&CX$Y#Yz5^6ue#9_D*en@G{PY%3uJj`6;qofAlMXdx&P
znoE8Wmwy*$_QiM;Ykjh;4nVpC)Kwcz6_^ns!DeTYb-vBDF*snzvAtgb4c~LWoOlj_
zeQ3_`RC7rgtXsiaBZ>n-r50WmcwXS?`Gxc4g?Tz~JlyfLEQl&OLl*aG`sw2*o||%c
zS!i4B2IKy?#sE3OAq7%$#;SH8dXAa5r#lW)X83B0fvg5y))&N|-%VFqD^$*iG`4B-
zyJdyGwIj<$-6|z(w|?-3MV^w&ZFB5QBxlxYYyB374%m1A1U&y#+^%9@1D9ev8&HGw
z*f4DW4F2p%=&v;Z6LHl;!`MpUa#={t8N8Eb2gWU6ML9peQ0}q=s4?s{fCxxV5Pc|8
zKFvIwCY~?P=#Pc@n{P?dbKMS9xny5yT4+{Cu>)VdE^tq?Kn=keirrCIMA2ZR5Xs?$
ztShvJQNWaH6+yedEv#Dj?(m+s@yPS?0ztW47M{}!&*hoj$x5;}h(kg`h4?f&4Aj~v
zTcK<T34Ma7j2+V(Nr>(4Cw_Y-JnNYqS?z#gkg;iBcU-%<kzZmQe_emt`U+ruT|eI%
zSzf2ZcCrpIbn8%?(t;-@ttq862Rk~;%Dq$Awa&xH6WEIjgFXeVcOysVQe<FBop=)<
zxYPwRYS~C}V$QMa)Z6~PB?QphakE__K=)cOp{R5pn1;sWdT<6<GRk$X4W5poSod(a
z{S0Gc9_*Ye@#5;5bLC?6@4-thx&7iunqPhIUUltBKnUVRc;k97OZT<telbp7@qZJt
z`)Ev-D7I1B@CfnF(3mKq&6C_=ggN|x#(VW*z|Eh>yR`n^LWmuhwCgpnanxz=xQ|21
zU)R(+P?0gq1o0#uQ>+T6eJl)PqL|%W%xhPCfC77=$~93o)7^7Od-6Y6B<%Nbe712v
zU*D(CzxE{+zgJ&m|NnvS_*c5DuxansTW1ETQ`<EXLf*f3FA{?T&1!vO=vH=sY!rSm
zxLNQZB86f0+n<|B{dZOyIR*|n`n|HNF6wsNv)OnGyC<JEM1#k1tC%%3W@5TVj;Zdy
z{3WlZrNe!}MiRUSqU`@azVvwvkGn5_;BmZNuVeo8Gc!P0Q}?d$J<rEq$KX>4$L=BG
zvfCVjGd`N_Dc|@RqX<w{v@@-op>r#|u>F7IPC1MhaOAs7`fM=EUI)T(tr)K*jE<`B
zxhZEFnGl?X;f%`G0XKGVX)q~$pVrSVkAI@p9JsXJDc|4ItGm?Mlbz{$vuX8b=KB6W
zk67t{|K&vTXZwzLA-se8T~b(^@0U^8$H6a`h3A)Nz=+Y~jIn+80NrYJ_H?NZq|oCQ
z2ai&4CQl5(eVoK0pV;r=lxFfYGo{q!K>clwbE^l4nghw$tqF7k3j`IaRj|e(rj8R2
z@5&B>j94CpP-mg-wX9vw=YE;|X&>>q=N%&oUaxE1zc%~7jpZ@na(uIcG0j0)5#r5c
zH3xgIuMzj@ZG#~)4$9s%6F=K)b0Ew^_$CI%QPn*OUt1LHm#JT8-G4!kIol(bVa&XL
zR&(z$2|iIek_h2?e%gL<LN8<kBF%y6AfQ*4_*D+<dmJ6Sy3PBBO@e<Z0?yjG{f=t_
z_WL)&?Dm!RFZ|iB+p|Jompq$6Q)j`HT7khtg<9a#r%(L$cfZ9<-LSo1mtfu6L4;*n
zd0sb<<940e8v1?_IvYBNKqk>TYc2*qds2*~90|=Oc)^`6JoW;eb&Kc19boA~f0rsM
zoe^DW7&QnQZk`>n&UUn}#_#)pd46HT@XBs8>g%bz104PEHgg8Pi0l)=ch>y%sCt!G
z3mqQqxucrLZ)i95M&J7|_H)(t7wruaywObXcnBRhUbhWswC?|?c8y;h=l<Tn?ttB^
zGuro7k>OlAqSZX%j{$~E-g!|dO2q9D`u0i+i8*FUNT@Wl_F*Y~fK#BOe7M(a=y5Ea
z&Sg4)^X5S3&HJ;yXZhYSZkPcP>^gDQ&*YJVbk^L>!#|?@ON`^M>rY!>0j#g<XIw5Z
z$Wz#vqAr=aEe88EtlqthI6Ee#h)xM&bm^UWq1;PbRNWZ;a$dNc7qmIUCq!p0r4aYT
z%=*ZsfP1dYIk8<9etLZ7{=)~RJaNc)*Dgdti<GLJH4xF>rLSs;HP*VZiBW>ER-eF<
zH4;<g-SL1CS(giIEo}V|)&M2}#XB+*#1D)Ri7|sNE-C0B4Z*J+Qu^UEjLfyW%k)F6
zCz4&>?k3g}NZO|~?Gv}I`J0cqb>m|y#2E+shEymgA{(aK6^6k*tRyotIY7kL4Y8H7
zt+Y})+t;OcF-KwyY%hhfR+Q#Rup15m6(n;_nuqt4qE=BWv{pPB5l_}55R)HKp}1@!
z#Kb1bi#bawnXxT}gfs7ITd8d$#OwujKPuaJFR6o4?IuBJ=J&rJykJyqthKl)ur*YZ
zr|wt0&5MN2E{W33${8TV?HvHK_631h>fUWrWXM`-^J#qpBx>eNq#%$2<N&6XS|A)U
z61u+}gOT7uymj*~;`zL=F78z$(IrPyj5~`+5sDf~d~icfk!i|UUHR#U-*K7~-@bpt
z`IGw8v*d|vG54tOhckgBOajenk7L+EVBId1wsFvzlq2`k3_#gd&g;hW<>KtR;4!Co
z!S1pJ2VbK^==W7dbsv0yL>%mlz33g{$SZ&VzQ2Y3KQ}@h0N%AY9c!>EtvO{s<wz1A
zu#kcq@P-(jz0`l7LdMoYZ5Kp*k8E0LS~#92@)UV|eB}N#b9{58ZpCGa*8TRx_i(@W
zksv-{3e2AYjFk<e(qstr2G}7k&jxR?o3jT}jLa#~P|nq-S|tJI$rIGh6k!zNo%ti+
zvLg%Z6OsB<GYLMRp?2wp7xbGOhqvLA2&NGwdO|<o5L1tB>jmDcHLCyqAN`Wz7{H&S
zIlG)J<;1eqKC#sS&)U2oPAQV2k!k#!|JVPMzxmt$7R^WQ?oT{EzwCvtgVK>tk54?E
z7oMM<SeK1CxokFxb;DxWfwG)kK6D6~s>bE{k^8q3r{j@hPCNzSa(<@8!s+dM5M#6I
z9x6%Pt4184QjE2(UDoTdT<glQ&Ey<8<m>?3x^UiBo|p3uZ2JUF!HUbALU6uTBoC<D
z0ZswL<!qrF<9MMv210bnkzZnfUE)4GT3&<E{SU$S&rtvm)^u!<Btq`MUyROp$vHAf
zpjPEsi%%oWV?YAjJ)9r}e*Ez#?oJcm|N1-Bq)SGxa6#+#2o4JL;&1?&3NO7tU8PZj
zvPJ*4(7a%F53jzb#o$Wfm`Euxr-Zi3a(U*wUdT~M$G4;|<ujBjFr?HHi(jvty+B?o
zt+_9s&?ixLzpzq@gQGS;u-FS`u$D0MoN;(9Vq=nZM!Wrj?U0P$gPkF-y5GA$|K&2Q
zDSMpeoN4FH{gWj0fl1yBK1>s>HMZ3Fm;d|!9sigA>i<kz@5nhY<%pT@eVvQL^ZCN3
z=NB%^g;FYq?9=L|U`)}^V4wD!kT_ukn1mdKm&cDxQ{=l}en<H66WdamV<JoJEOjU%
zSV)8ktOXhY6W{;WqApLv(6XRyrIm=}z?*r3XsqkPvaY<WXKH|&gAY+LXGg0-j4lr!
zOyAbQ4?jK~Wm0MV&%H4&)@8CCL=*Pr8T2^rHW9lI<G33mckBRM3V|%q{~H5o@+My0
z3L5<VzN{-q;{N_Xo*MaZV5uADw(<DW3pv0q-oNKKWfHz`x+1^_UJVIjjBfE!TfI4+
z`gCMmZn~sM=`}}DpeU0-ibCi)Zs^Tu6oN$ZG*g>$t}EqyK}5;Ndor1HjOYe!bKo*a
zBBY35%xQAUwR(+V);P5Zv*_dm1g(A>qS1j)_e3jVK8>90%v&0<HzX9rG-De3`QKS=
z{T__w!j~1tU)P_qz5-ZZ*Uz>@Tuu|h=nYsOZ%hBvP4U(XH1Aq;RKA<djLx}39UYWn
z92o2yx%Wim_6>Nqp-<`RFpw9I)~B9y@1oFINfzUt$opp4yD{cI5_AJ#t%e0<sT-vg
zO3CD&1V}$TyjT$Eos!WlXuWs}b^|PpmfB61a91>p?u+ym*zLOMp6x>jZ<6KDCX0KK
zH<IuEK1SnXr+oK<%n$CaLGK6m-U*_2{l=K^5J+9}ceyMa!%RxqyH8wi^!U8PnA#<K
z_Ub)j!>6lZy;C$^V!Y1ikSoSgH-BHt>QhzbLqFKOi(A`93T|pWo{#ZPbQB#o>>2HH
zjPWQbx83V&7kk_{8~3QQb8d?i8;@-N{vFKPU*OMwc^o?!`l<*U$&y{oWMiNBcI{Ml
z_MABgImR9PFQJyo*1T)4s_v`6NF0aM)*oV3m%NSF#XLc3>fTL<<4iek?&+4~D#`1h
z$MC>t-8i)NI97p6DQtB^OJEY?G|xWW2cub|)Pkgl#IVQDkO<yh4!>aR_Zvq1*?u)-
z$OLb?yV!^{**oFC51#BdWN!crxgX=-%FVU6@ig9ev+c2^q!c^L)%QS`W(zMbXL17L
zz|OZC&-`F;$m{p>>LLu*mUgMqklc>zatrA8>!b!t*|27mb!GCgOsKgDa8fTu?bb|}
zA?1)^j`Qk0j#zDdh=%Qde|_77jWXVx^}^&<cKvxb_xkb2x%1_$5Uy;Vv8Ru7ZY(Fc
z2`MF?f+bM5mF3b%`2aysbL-ym%nHJ~t-PFly5R6gjN%}cKbK*@2hJ99wox;6NxAxK
z#^4~hn-ePC+Ql)VXgebk@g@`IU4Fr+t3O9u-CzqeG3Ln)z_p*<r4~|7xObJuX6ql^
z<-j9W`GaTjxaT_yRj!=iK}hPy;?5o$c=qk`GwQRij`Qu70=CDD?+vXsYVE^Y!j+jC
zV(6d2r9L58H+WXA1@`p}+jE3hKS-|mz<4ie-P^}{BPn)p{q`SUSry|<rYV<gMbl_N
zG+r(*OerxX2*Go<vDHc6BV><7&z~d&_oTe>u%pL!+XMQhQ8!X48#&Y;+ebh9^A{A#
z<6I1#VW^{MxBU_7Ia=T_O>CtQM0tLGVXF)G2RFLkpAEr5peU@Ra9P%U0&@`mIjJ-C
zuWVLl3U+=Lrj6xtCIAno6VFdCXlcD!=V}VV0E>F-QvHdkRuNNgrl7DYrLHtxv66`5
zhT$-z^Od!%^ap9rXItM}#^=AirPo{=u20PCza4Sc)Ab#M_Mo@{Om=X~bKqe;G4@mG
zuQi_SkvC(uH}aYqh*FzUTJ!6^tvuJ2ux@A+wyku}O50_@rq^>v#v2*u?#-tjd!A3@
zb`yT-8<h5%Y5=&VorPvShRr3qPi<q<hU!L+!n@#7`?9Xk5~vSI@;GyuVhF*N5#p>a
zRD15HdnkH5dwgpDd3%j8wxQens9*D&|NOD+=!=Tuuj@}+UjeMI>t|azPks<|8G_QV
zW}uC(Qt0Df9rTG}WPt(rNHi?E)DN^v0!Ps{3X{HXYjJZ}eE(_~781mSw(jRZ0^Gwv
z$mAH1FoR@bOoSw~kdPF~bE2NTb8kL+kFB}fX)UFX2j5uKSyl;{n(^*VT@rzht54$0
zlYQho#@3}zaX&_JNmUfAbyLW3E``|5#$D?-SWag6)FL{wjKR<fSCH#!;;{p0E?4Lb
zihkB<b4s=dUUU|a=*HTJce-P6Bb#ko+>}%B0!NIPGnv5MX+mlI&ENc%Uw!|9!{MRl
z!_sNib|cku&?`uwbzmfQgIR5rRtiN55+_1GG|lM758CBT-c9=5%O`)Xo(uPR;var^
z?@Tn?Xl>(VdFDQ){yP0L<Gvxa6*sMpxdWTfsu%c(Gj1BTgIMFDHM+TOh`9M^+mmH4
zG;mJ>8K#=tAR7PSV(dT0%4zbFuDMx`Y4?BXj7l*<8%Pr7c_N<+!93wj5iYIJ5K428
zg(%|+1uhq7IMrG(tN8RhyMY?rSX))7rFRH3SX0VY*|y478i%)UIo+SIa-mglLsQ*N
z(pzgZEr=Ow5zxxIZBC0{E9bhhw!$&OtP^kV?mD7mlva3IUYJjAP&-KZ)ttEq{TcVS
z+r9AEiz$~K_kt~OWmAP-h_x$I!@c6<8pnYym*_wbLyGN8(-@TsVAii_|78ZvDgL2P
zD?`0YTS}#rMoKd|&rEq{n&3DkO5OOIzx`Xx-t+B;yM51UYJDQ-NJMYSgPze3kFV4V
z<xnd{Dn$z^PlOn+AEFn;f)CDEq_DnhUMR21U7l#a_=ZAbIX@H9f#dNFJ}^NrlFLkb
zqB|HhSl{1*(we_#2RZdsp)+uWT?8f48M6Hfccc60m+|k-8fS9r6T%$O#HUzxP{Wy&
zfxFX@vKmoVur4h%HzF==B}(H!H{uVGr>AFr_~A!NaYoG)J(fk?D=o%^S)`c|N=0HI
zrWsmgJD;g{7icR}G!9cB2S^$6v7hP93!>KQJ|%2xE)&0$4K0nfHqQ0J);6{z9L#ug
z=P{|KlvY@_3v={XuBfv#dmfQZ8NPP@Tm@(6GNz5WVf?T%uwd@T*1>7(5~3g-RO(G0
z384d{H^t{@TKS&<_cT(|D^S}dk6N31!CBzt`K2>p-AgB=Oq^!oRM^fNTdiE4AE{+#
z`~LlaulJ!3tvZ`&XO?jVd&dzsDWrybAjJkKg~mqn`!=QsHW;2-0HrmmHA-0t-J|I=
zPi)~Al(O=hzx{!4-@tco-vEse`d$)oKRauL&~~P2h=~*vmvW(%Myp;}qm9}cTWx#s
zYxL{ZnPq;J<E!iY_hpToNDx#jOWCN!_F*&Dk&~1%@7}$k*$a=)8{1M?x5{CvF4L|n
zD%BbJQK(bo@v`va$48$~Dndv?5@8O$_qM9;p%;{#)h#``?Cs-EpP267@#}9t@bdVG
zZ5vxz5E_TW8~5E)Lu+qZ)#CgA<lwc{GgT{_ZmhPl*+xnSCd+(te?m}7-B{K$&o56L
z;ypQEr)bs@Huy8Gy*UwrdrUbC(VvE*?ujKB*1=QfqS>wwjAOF2cq2Zfursp;XdGg6
zxopf(d@8S)c3Gl>zJm!Ysx0Tl8#Sf51ai6X>E)T<e)@?YFBd+o3lBlK%ZZos3u2XU
z$UY25hqZ;rZ;*IxzVy$GS#JUen{A+lW8Fwm(Rj_j4QFThz9LM`&-nsaDVQ2RlnX_T
zW2iiYiSG{w)Zp>Aza{6l<muhMj|EmlT9=qxXNJ2e{XSqtJQpdYQB2ulV8hs2?Qwl=
zdRg#2*Q=S{NbivU?m=k$eGBcc>;HrG6~Owsezs*c*a0q6+15tNK2>mOdNt$h1rO$?
zuOSJN0u{S%e5MW(;Bq-PWj408P`1jD8EW-Wvl5|fg;kZi>453zczE(@lCY52RIsWv
zynvD9GJ+VR8zZZ*mX%sJPZ!KzqiJK)&EKPm`wfV()kYR^qeg_BB262ugG#=kyKQic
zvD5LdJFv;<fcD8D^sZ`Wrw~I2?$klR!T#SnDt@}ETsubl%inZM0jC~o@8Xz$-Ptm>
zPwv`ZuU5)hSf4J0F6-LcOTkt2%d*m_G+TLlx#M9w?j6(Hdu`^0!RTZSN`N0swzt1R
z45+zOR9kaHdOE91H$3&|o&FrTdvoL=PaLP2*eb0S2g6!VD#Z4wSg1c+t<%*T>fJ7Z
z(v$pR#-_^B8l{7!&T?_nOzlPo(tQ~EL*K9CjU$2AA^YbXx8Tb;cl$F*pfv|S3?&38
zhJz0MnQ?iCk%!H~@LTK87uf#wm?F!%^7QnAw(4C=u3b3yYVvzDyibPR5xq0t%wuk{
zZUh>oC$7~2<S7wafoim}pw%HVc3`H*u{P9TD-{uOnOX2XA%?(Vo*bASg`RqH9B@gl
z48MkJzjlO_KQqCTy4qYm<paXpyxm`q?i0e5;Uv;!K6|_O%IN4n<@Om;b)Pt*r|8nz
zmOfF_fXcdT)M`86R{M2OjF-!qQa0*ZSl7a~mc5&=uh<V=xZJ^`Ye7y2Ky;TcIH1>M
zyXtZQA8g^NhuVgy`*j>cB8S9jp1995r#OKYE?YxOWveSG`@}Nq<IbIN(x~S0VH<m)
z8D_qJ6jh2TMSJlw7;HEoqS8&gLq9mi^L`VfI~x{W#qpPB<{rd7g|wfoRU4?sj`i!a
z*Zk^isN)o|=z&*TK_%?F+kWpu_i-xSJQ|#7ATDb+>%9ka)-=)MB1V@#9S#QqMp@4s
z=ETFB6GuHF5-GYN5#k|{%>7r&%Zk>*)@8d*)qxY+d>WewYpMS1^@>x>{iAY<y~ktw
z^WA(aCq|!~tgXWXorS5_-o@|7C+(&~Av;(+CX}&1JN~|6=!~=-tiJvIj>Go7q04)M
zbdS)=<-GW3oDS61SUU6V@%fpj%SNe<WQA!;)Y6<av->x4)70`e7NSf3Oe?-`;MN09
zRKNNSjfcp@Wm?TTL*6YS8qH%yVTzeUOdRfK?xvZ$klpfOYtXupyj4HnJIX)u0g2MV
z=CPMwt5o;HvmM~NGB@bL$k4&;;alnq{3}V*ZazP}>L&dGw&kDe3L$lC!t0(Ff){jG
zpCkthtokIrIm0I29BtOvts<dI4wbD|_oZ!a<so=e)%I=CrFz=gZ>jIKqS}X>R1U#7
zPLU}IISKdoM^c>7?9bAW;06)4QfOMy+9hn&fu&U&8%jZ_TSF1*RtecLkSxL}Wj9nG
zIcCS5e2*VKTjMichjQ$kD)W5X=YV+b>mIrt$kprqHn>ZkZeQj6&R+73VE}m}EjBTl
zR!m)DFZe#Wye!nMq9IaS<MF4D{P^jS$MXv>TVbt@)11-h{-GktKU)W`fG#zT-ZV<7
zH=bAIW`FSz@i?gBS6$QyqDXJbTc6DA-|Tr&Gp}XD;3SzC0&gDfxSMCDX{Moke7W#&
zcSoKDWW=*SNFf9#69n;^&72{qtwD4KaM=o08j}gqn`7R^69=ybV_pQhMS+($V~Xl6
zWqAKTUtdrhe_emt`U+ruT|e8h=4r0e-i`B8c$*{h9N^-tEZYq`d*?vhhbhjHCu@D<
zb`VQ?br)iww#rtG%d)a<8#(*f<W_t-AHcd6D2<k8nn7t-gF+>tF|kxPk8cPT9Kedv
zr<WxSL=(#6g|$>LFMeZ4wAOqSw?x;pZw1xHRvLuBgpv9<J@=To()+9WK^CJM9}WOX
z$2hZoFgjosI&-O0J3^$j)xnp+!0N?XKR>W2ok?)5MRwUkPxb&(2e1aGL1~Ef>!`Tt
z;wV11!sUGC(@!6{dw9d??g2j>4eJ2b%jH6B8<)EB&4&lJ<wUDq{MLP>y{7;nTxFX3
z8#17GtbDpvqTyZ`+FI{y`(bA03=1(x^O2k;zB|3=O`f@%59o5?Tzq1cX~QRGDQ)Z5
zS+%p`oXIDqY-{OLApM!xT4QaEi&i%EDNxdl?zMOMoF(Ia6a&P$y~g0O(s2EFU*76m
z?f`7fSl3FP+-hVk{aZT7QMU@(m=m4)F5?n8=+$LIF}knX%lX1jA3xGMfE5NPq0V$D
ztv@@l?-yYhkwQ$Kl&!UL>>g>BdU9PWIVRF!B9;{?N_lbcCXjjx>p;~~H@YPQwtDBW
z)Yh33MoNLZ<IyLO_JrL*8c$1ONFB(y3WDqg1PSJhs@CHmPY%-8E}Psv(oHc9KFJj6
zz>i)B?+gHoYdj<DKU=l>IC+y@<}-Xb+(V}E^7w>=L`VmhJDdC5HI&DvN48R+Zd}%l
zWmz5M?o-Xiq_Y8(52((1!arnFO`WxBJs}M-qP<8>IzH^|d!BW~&WxNV?()np-@NBx
znz>ICWmPU~qi!2pT_Hgv?JPjWSwTgWtXDw7)G&4CUNL3U#%B9i4aNj0Z;H4yRD$g|
zgbe`W`uOuIbYJDmUk(%<ETOnBTh&I{oGKUs`0NWqrhaFel7nz}jI7Zp;uFo<4bY+;
zp^@m!x4Jb_-_~1oFqko3T^*FxXb}DRPBCJWaChe(G~4pb@stSDJxA`S&G_B(6FFt(
z!;#z@S<la_vk1xVkExBKg{5uoHyH!#T0Fl1XHo?B{hFt#10+}0QKS2Hr9^E8%~?Rf
z9DKh5SYEhCryO|02lM9#2l<WOaI?Xt7N6?pSM^42Bs)kp<biTCzRfzwh&V`IYh%^Q
z^UDPV=6s;F%6h4MdVb=^Pak=@6lzlr6HIevTbqM*J<swrq_N=a6rbuw^On$H9%~<g
zI#|di&Mfoz!u{+_P#pAY)SizBr+nmXI`Hj>Z@8N#9>T;@8gVtGI67c7_sP+Wj0vp*
zAq`Kvu!A!W#=G~TT9+x&r!S8CLPbS70Cq(J&HAK7G5k)yUROC5H_d@R?-eD}#6dvo
zb^h+QFhCW>8U0%$nsS)^z7_L5+U5AeMHQ-gPFspnE8HI(EU49iK}-FdC(wE<J9yKX
zL9J42A&PM~N9F`mhPU555Mtu7t~dj`BGHIh*v>0!QRdC3&Bn^!G~5EzAlTxQWVfx7
z4P+9g5P6snG!bgk9V~9J4}sD?iE!|}_eN2t=QT58^jt*t{6zXtBDoImF>f+^9%p@j
zyG|q2*7_dm&z<!Ae)z$v`h^Rf33fSOSS~C1&0DtC_#b}wfuBA-^V9PqXBAcx>S4xG
zCZtS=sn-x4z!D#P5kn$j#5`fOI&c@fX&$soe0v2@B>KMHoH>_*;*-^f&!g8GW=f^|
zcq+VS=1n;8?Yj>gjx*EYz_u>@^mHcXH@xkQ06I{G1qWAyx`Kv)s9Y^UY{Jx7w}O_&
z!2+RqzHJ@6HfiW?aw#B*(QJr2xbJ(}Z65R`#_`wnr>(C5*4Op3tw`uhon==XU6e)%
zZoxgc6WldGaCg_>?j9_-yF=sd4#C}NAUFh<#y!wD)3fHoynmspR;_dHeV)BH6*K$}
z={HEwH@209SZ8Z;$a-tPPFcWkD@6?Z0iq&?E$lQjZI?(kn&7o!W9k6%XP{;_8WCp$
zqpg$-j7EaS)W%F3+8I6D(J2U4L-!BEeDKYqcj4lwZ9bnu6t9;XuR@K^24bblRw@pq
zQ#?2T5zi#g%2c=jN6dMhNioax92M=PZ0RS7Oa9WK>&xz+Z8zm*jaoWbDMn!+gH*oM
zLHuIzFZ|h>P15jn?6dKE-%%+Rcf65f>jz4MpF!2RwI=IxpyULcqA%3E#n$_`O}97u
z*h`^RY>EiBoB`gi<NH2`>T8+5iV5V|+ibrDxx;Io?br37;QcJ_oWjRc`xhdoOF|z_
zsD}W3yx>gTIn(P^!M^|Qe6Z_1_5u-f^CeNwhky65v!4=!oF7>sK<->60i(5C+4eO#
z6zShbDi@b@s$I5-6k*sEeqeIuU&3r@&hOet&<P~e{aksx!gg>p-IxFR^ILqb(PbPr
zMY-sV*q@(i5j@3s`BL5htS3Hfw5-xh-^Jq}kN2`XwM~vFM;@ORHt~n9{|aNNU0{h_
zy}~0#>3xBuOSEDwlM)iF&V6=od`fwWxyJcjnK)bRxoShtzdc#-V@vQ&ODpSeI|m9S
z9l=a_N)2~SSRb(YAYdWYScu!5%u*bjV`mIncW{U%^s~8nC4wf|K1+}ZlVi5d59lQ{
zehRQ<ot*wL{k9dA>Y0WdTUXY`!b(&FTT2jyu`2zmeck%+Xu0z2!AbuKFeqb6tjpjL
zm;gy-pA07C=c#bVs~2!p;7%`sTkE2T^C##EC9Rn5CAR(JVNXy|WYIDuJ_*Q+|IE-A
zb4?UJKjHN9lHMY{^7$?38lBgz9<5w7KdQ%bo0@c>K(0>lpi|qbM%@6d&cP+!23(D9
zY#Y@}v$&tI7n71<F+>iJdhSQ;cD8~_YLC>kVu9TUUDfsf%V>gEF<t)oiFPw)JO?kt
z&`go-Ug;eGos7jmL?nbK+2uD;papi;jc&rpG#!I%S`}qDnqp4S)-buD?3pIjad(~1
z7|@inM^-a-yVFuO*F#SVdf4R-X0iUVYuk3U)mL5o82aQ`9fBkicM3}$;GlVo%-c{!
z!d4idP<x~yJXchr*{SI(i%D1E=7rQYsNff@3whe9WSv&R)C>-R`)7x6n!jc~kDN&7
z=ad&foCVM#D>M<p6!TN>98oD(!OiPyFhyt=q|f6mAaa+JAab+6fNh684b3pCfnpDo
zS-wDaTuq@ArcpT$|4#3J+YDDyb+w|FG?&TIf@8x#P9)$v^+i#YHc(`DFw@enThQ6-
z_r7z+#KhWVt(4YAY~NyB>IrrNwpn7rA2w$5sc#84#$_v?pn$1BB2CoKMvG-oSYK!{
zmyq3&ovnWGM~trYQ+9Nqmo4`o1lEnepmz<2T9b-XNGROSg!goF@m}OdG<JjtL~LAe
zqHbRa@Y#iP=Ffa#O-QIIb)VDn-8xp-)ck5P)AP2UMf-TD$bg;f=~<(8CKa2Q?95=L
zhxFRejlqsaFH4yO*`uON0kgBPoTQ*E@n6L8*2p?IV(FKChjC+zpQ;imY39E3o$!~0
zEY5KC+yr!WoWK`2ie4J+2S*-dtPeD>nHt@A7bqWJ?lj3G57jL@tEqe$aw%+=)96H?
zMr&Hp`)z>7a+C(X>Z6HNh1F6|*K2epFGt#om9w-<%;At0QU63%w}^`*?9ufz#LLR|
zOYTgl43*68T=baYKVo{!N<7uGX(o23;r-D&S~1LhZ@z2KPBUR(zyoDowZT_Czeo9F
zCI&l~S+EP!lpV*}&Z3eUxJOfx9#hZ#s+Gh#b(tntoOwaFg|7ndJlsdv)$WC3|MCCf
zuv*9Jl_kEyrSuBwAhubBCxwMij`Wg@41)mQKYSKqx;-gtKg-8!OZ{rRAQ{Zh#p>l<
z3YEe8+Aq`F$D&$@XhE6Hy^ven@}szPD#abwtySAyNR<-6sq#`-=1BczlNAYd0t1-R
zW0Rc)qB3d|oTFV{(>9#MzpnQ=;!aFXpy)-`T_x0Z9JLXQDbPgdGChwN+LGYro#XOE
zn#)VeGN3_xlaAf<3iS8ZOch9xlQvxLl58YHWkF!DJ9O2Jyi*v@LSD2L$~E@xw1?={
zJVbu27jP*9>Z3X@u*HI5t2#riS}?>Td)zK~RUHq6wmL2?vxd5g1GmIyHpt>PfSl!t
zh>tx|uRoXda&t{_n_I}jZ9$t|XjO~ZHdvzKtkH!U2z!gr-<$CFNRl{?o2u+gu12nK
z6@F{Ab2M25N=YHSqUa2CtEA=eg_8Ss#xE>I#A@PZTvI}rCyH5YilHoh7qz!D@h%~6
zwOVcK<tiRf@`du+b)K>TKBZG9l(vwQRL!?7ri?UG$Vdc&N`Kbq1y+TORrt0O)O2Es
zQ&WT4<(XZD^7!&DlbSnIM}^pD{**Rd`m6gXw8h4^D|2t|dk4b*9@PxInyKr4*6#>%
z&oKTMD13nAOVRy-^bmU<E^Cu}Xi`>SBCX}e7G#vd=O!yt$s(dgDIHRZliqoAbC{Jd
z%PN$QpV*&@t}&EF-{T5S@bnj9@qehS3MlPl#hCw`j;ir6yba#^B(d<>zZJ&qa+`M+
zavdEs1jrH-V|w_u=-*^pi@NWZffb~)P7_nBiL4Th|4|DMemha<t<4T${SB)Juspn}
zKm8n`ec8BTfl9cv+T6O}dW`g%(J02s_JAc#+>Z<``BMCM09rthcLYin@z3wra(|)y
zIhv}tc}T4v5K3=^V>07T(AQ$jQ@X9q8QtkVao9B3!~)1jE4Xxd>}DFyq}Sv>7*7oz
ztLrHAfl^b_iKLz&MwgB+$1asDxF@S9!I;Cz;%?ML5HD$_6uX}FITULpX}DdgUwzC*
z3X<Ur+<tM+&iUq#Q21he-nj5#)J6Cfd1#U*8kjfMQ*^{s`&GxnP)h+?nmgSA(zC`^
zW@Qy(wMs=wlAxMnn$rY??&fs#eX{zsV3E6j_HuN!f=UE?@2?c$R(?d$75la961;!=
zOdn*E$G1T;-mEOn{n;9$ijQDkkAm8&#cunoL_R5-r)cD?mF%U@)6)S#GNMLxxCbU8
zKvh%FZ0c&EHOAflJ2qrwTOa@4n>fk)j}IA&f}H&#Ai>IR>)6A}llK_UL{7}+AJSH>
zG6dMFbBE-UaaYP&2i1~qxTW${GpnhQOs5A4DLAx_HgOKTH+UVVHz~!w@4}w~R0$QI
zoMum%4~0^mWsFP$Dab@6a+-r1fX&UE`-gA?GL+WNzg#6WkNLH648(YqvU3c68uN;_
zz>#mFX3e)9x$M74+1WP4d%@pcu)=!}{xTW$ddP6&Z)k&HUfODZ-qHO`sWZ8p!1XQG
z!20^<@JcaLqTb4Rw*0O80=462I^l8>%3lJLs1kkLdLWWk;_l5v+YxI0%^<|5cLDl9
ze9s*)p-x#A`ERBOcf{6hr?;8nPfI>h?}$4CKMyisGsp2<vX(bI{HR_O{P9IS2Yj*M
zX4h;8XFsIu{3`x-yZ7H8pShcVCtewpi;_&WeMT7lAP%<PH!DWP@$&Iqf{v~n*$4%1
zh#E|X+cm!1myhjmdYLHU2%i+6H)0x7Z5yh34!8xwZZ0;iu>n(5V^{v~OTyG-xyNDC
zZ>9<+$A9{4Rd}|VXw7WOgS*buxs*;<6yd`|C!fP_a(nlky@$BA*?|G(Mgh;v6#eWz
zCYsa!&_xW(UoF%41iKi@=gR5P>I<CVB!`0l%bhO%P>kVe)tF`kT)0Anf6;9nV&(K}
z)SJ7$-_RTVVmWe6!0PjyftUR*!=LO{(l*#e*fyZmNx&jCp0)k@4_o;ttk3^FYHD32
ztNRX@E?Tn9a+*XY#$Ro|HmDXh!@9y+xe4J1ZUtpcRiwFV?DU^f1DCu=$qG9u;2Sm?
z*6)}kag`=6+nAcNZWSaG%F<yi#wI46oIl}Tyz#;IHX22m8p;HF@VBilMm#L?7PS}?
z8#KE#qzv#?_;f?}Q+@S$i|BY`_tzxbb|l14u38PPFQ2*l&>QiSC=$(-*B84s*D2NG
zr}<s@NF&Fl(e1uRn<_@BLU6F51iCqpj#4MLCE+y2=AE<{<uDRw2GejNVzN&3JX0bF
zTWj<9-o;SY?$B?m!nOv8z2>);S;Q{s4Ze<l;&KTD$gtl$yc>d(F{l+nGoP%SK9Y&U
z%E{jGR9qvi9aMONzlOMz-9(i#NR<(&aw8kkVr>33rM7ltVuAE!GiebtSK4xl#(t{T
zu=~My;nW0P5b+jQ3*8R>FoUkVXwCK|=C4@LE1sj*H__!C^b2LUd;Ax*?f)9p_jDqi
zmxuqgS-x{H$TF$*)M)WiM0vxPNyk;UAkOO>$DAlI2``m-hg@GB8k?WG43QnhWRgW+
zpCdEpZ=Tqm;ta91Ij5~`sx1y~1+G3Ggy%$zST4X~GXL2xt`tX(VLC<LSg&#`R|`zq
zWgl%r9RXvCkD#hAxeqKQ?KjTwhYwOQ=68H-`Xn46n}j;RVw9SNMR}0eiPVup&I{)P
z7(Q&<Gva%~m}9thy|H5eeXd+!z{Fx|RYs4Q!Sif$NARFTa~&oWgGQQ39hi-*C9>1Z
zV?}9o6mt9`uRi(H#O``Q4@rZTfki%gFlYbU9UwCyi@{>6*GF0*N=Pwvukk*{_>HQJ
z*yk1avE%Y9e+ZMRXxK#l7nT|&yp={(mDlr#1KjA{{tsE|h+$&dc$Ymj<n@B82TuOW
zeo!wwGF0p?6wHstnMBy`W1O)fyXzyihe;+x`&#na$My)(m>|W_W}cb#-#QkPbf$m?
zu7$zj(nmKUH!IJozn>#vU1wxC(}53fl1mmN`kp&JR7IC(JEDdT=9Z=k2Rewlk#4j&
z^fcJOnCKUPkBFi<awAWRf;Gf2Q_971mX&yA0AH5_V_M4#&`4UxsAigqZtZW&&RMWK
zG(euUGhvSAq0oahc&=Q|AqJH;M`9uN(QR%sL!IB>yp=NjX-1bT<a{x6Z9|uf?`4Gt
z7EFA9*>UZ4JEMIp^e8ChHN#46x_N9=ZKWi1f&!PpplY@FevOi#maw|anQNPbAk;&T
zEOu{YpH7|z@1}zDma<Z(q_|8tC>oP7w)$Qm^Xk)5{gzv{)%!qwfMOoqtx$8im`&Ue
zanykIYyy|&OqC<a{AAiDsR17I#HR&fY#omTrPqzBr5wWOMDZOHfn+uEF&PI&U@C5)
zdy}%%+NYm(aat$?(3SXK;sep@C)f>Jr;rh>nNOJpQrY~-6>#Y(U$0#YLq&7qO9b~J
zB{TCu^CIhg9|p+bZPW!u6nfP9VK!xt4J#7oYIJWlC5n3kUqjY!Q%cG9XN<p<7g+v%
zG4tK;e85FKA<$2ajide-$TB0Yfe56R#Gq|!z4`YzJ@oeN?f^(V4dx#C611h@hLWX6
z^aYXzu2q#Utz|rGb6MM%CfnFC7S}G62(zbd+}a?9@a@(XJ~=;8{&$*9pNq??#iDs`
zq10?yzO*{Xi9cVr;dPJm?-qn2aY>+@UN{)tvQ5>~Y5yE5e7f94UR->IN~*jlMQ2&f
z0#@7LwW>{%+^p%+%y_kSZU%eU4HL+)JNvoq@R1-QGf-cmdHJqz9Z@kT1L9L)#LiGM
zlZO?LL(uZ82H28jc#^f7$dz%VFjwp4l@-bW09?xe#ThM|b%NrV2k{L6O>?<9Zy$*w
zXIDqr49@zOrY@Wra~!AnnC|66i2w|2%FjBsyWMp)ttyKKiqw};f2O}_y8fmen-@tb
zc$wW+rmXAFhU@$BwJjt`edd4|F{QaB7<KPAjI(Zue%qxOY?T|ci~;v!OEeFCYr*D*
z#Dh^?pS)(iu&%A!2hDw*9v1K7MqHlL3bQD;XbqZ)zb?(?73)b^wb<RIC4#?SMII-^
z@bDr1<Mt7;3_qsYhDhHInB(Yt!s&|oxp^yp#IpyzxLAbL;KcHz)duYA%=3%WKDPF&
z{(_;{YB0=-iz|9|*Fsd+0+JlX$aqc{0(dVvNUyN+8uUJ#O8I;%v#|Gq0)t4J@=k33
zH!j1gqDkb4cbAtMKH_P7b9a2`W>^nj?(eGm&5@FesQFk(-T3iEv)P<rb-O|QuGNBW
zM$g$v>{wefBbfc)+n6RNk3&sulO{N^NSTED1at%(S8a5LDcC%(_GP>Wy?qYkz&<%8
z`FO$jNvs@|O2t7_;u{}}PI{vg&rufZ2`URk19`F5v;SEFM3t^@Eb7&xqAl%UZ?-&f
ziCu^bF=0{fs_n4gH}_s`1th=Woxg#%`|ty=(F3>6h^97Y>+_q#^OZv?IM^~UND|X<
zkQgwYvo2k|$z;g)wj}rheBka!5g7_a^h0Wo#6%%*EmsGNO~3JeOJE!zrjdK~)Y;ai
zM0otbUBJbQk4-w$br1N~(;N^lZIs8#bT&D(8Ra8qFQSNp$zvCg=iHMIL03Oan^@(<
z-)=f*Edn^5E=l<y;_5pIp&<*D#kO<x!2mLx4(LDoHHV<S<qt>VQL6YB&Jctymj-6V
zS3SIYP>uizj%+|aAP(EGNH>i-d^BJ*4yl52v!PeJS@74{OJ=n!+~?pY8O~`O)eF>(
z@l*;Wu}my?@{w|-2RA{)Msd<hw^OKs>?5c54j4JGOViwbr!so!AKQZHKM*3S=KWjm
z!%ElX=;I8n@AR|gIeN>mHt)#u{fCfyUuqC4RlGFz?PT$d-4!)CSp6I(?QMI4plWY+
z{s%dT;RWYBN4oK&;4Bc~K64kq4Hd>{y)aGV&_-leX|HLcwVKWOJNv;vL|ROa_%dN5
z7i}YjpTu&jFHk(s>edD1s;jdS$L~9%b9OH3l2JObK6bUy@bP=XFz8$Rma-LQd<4vx
z&<$|&8v_zM3hNZ7oo>hCNQCM?Vu;}GZRDA?ql0#(uaku5hcBu}LCcua!tz>q&|RBq
zNPauJC1@zf5~2@NdsXPQXJnwZ8e6~TK_ye=HhQUhF#alqjc}fwDd=L5pog#?tegob
zmyj9YJhWxfcUBg1o1G^x@T5Prd^4KzUU2+t&}-c@a00cwg#7&_jA#-cyqEeWMA}B_
z@PXRq`^$)I{bu*#=O2gHBlHk$W4ZraMr|k8(ltmaT<TNkRid=)^@~#S|HidKi=cAa
z&Ye4-i1GY6Pa&nWF#Yxh!$4Ps$1*ke<o9sFIJ$-F${RM|ohVKTI{1vBX|z1TyGRZp
zk10<Lnkwyn-fx&xYGgTeKdZS`em@DE1aH3uYiXGxZp?Zau;2^(zm7jOHPk7?US{nN
zdwuNGS2*mKEB6gWt7T(m+np*uK@}OOTv?J^dx$E>6lpfDPBcR=aNS%)Ts>aCAq2bu
zc@I)CcRp@bKC;U*qyswr{!LKtzv*9<eHINeQjh^-5g|BoTFWJdB69s<k-rLyBVvFD
z32&BYnswnM_IEk;mP&ZDd|^~ED~S5>eS8c?#?r=lE%i<-N*3uY5!xghO{aQsFGcCK
zz^;NDZ!^)FP1D@=ASD^I$-MVtL?6W8ygvN&G<ayXL4$^!6e3Zk62--X%Tq9Hmhtxe
zFHxyOg=Y4k&t@X+@af_)K?)2dw7#LPyjc%dn`Y(bB1K{va@YQnBnvZSh}+0t5>6$1
zwR*^1#sPkKmp1>+w>OoO$6k;x+mzb;+3l=DlFJGi7Md-oJnMy*lNeU<9O{xCIPC`7
z-OUqH)ZkW&o_3?mc88PUCDWwA%Ib{OT#{j+V83?d7mp0(YQ{PAS>waEsGf^3D8%!x
zGXAxHGyK9!_dXK%&iZ)hwS9a0kE()wj8L;Qb+9~N_~!NB;O%V^9`<&(e}wJ6kB^v{
z8A=o_OP7p9GSmPVKpBVU2d|}{rsYTUBQh#tA&D34^f=b{F75ZR+5UQ1A28t*@TxB?
z?0;bVkrOxqIqFlM#g9;-+O#T8Fd{H3a$J@fBvem7E@=1{6@rM1aX%3=bnfmD7R|m1
ziOn3eC*<p3&Y$F+IGVI6VCFx!>F%h-WmudA815op%Sj9qroj;0am~}$KC5Hk6eTBM
z%16^!ASV&i2Ha0!F90g0A;8r6EmTC%c|s;paDzh_7A*fpA%)74P8(cGxOJI_58-Dq
z++fXM9yvzyu-1QkMvQ0c&P!63O4yzOWvyaLS=3uX6nrvg`%H@n?(VQso^MN$Q9D2S
zuT-1gH?(g$$;IS^5epqJz0#@Oo#+sX<uTQ=GtoW?!g8Sq2~#L`$aC7DL7?x$<60=}
z0(QDCK33JyB|Ef39N9$c{ukX83q(x%`zIDA`n;3td}cTy7H$u}Mn$=(vTrtdx38<O
zgs(+j@?VvouHTuDj>sC9g=pO@v#m%){0|Nfvo2nTA4-Vth5=Y3L>b(_$%u_)NSm&X
z91qEef2)B5)mnxB%*nXzbvXqLu=c*eeLRVL%zE=FLHc^{z`&R7>m0IxslNC2kLdCY
ze51TgYN4_(#HHw8kJpei{THT-Z!G5>(9{s>{N{_Yn?K`(kZZVE#sQQxLbip3Ra*vb
zhEELl{QaQhGvRD8S@ic8p29eGCA$R#>7*!Zf8ZfF&!XDcCz#vVpTfW1WHy)|;{g0^
z*-HCUv9{k%nHMNj9Y7FL>g~Tsd)%s=w|WN}GMq+HU1TD1lI^1Y!){kEjxwf-y3Fs1
z#Z%9ABQ>gPd?i$-QB9Jz8F%&iX1ZUjE#<sYwHj%+el)NXNC!oGOi=hcVLSABore=N
zkgMEJ@ux84sYedQSpq-##WKKg!PfB4x$*N-YmcWrT{636Kkf1+2(xq1#d{(2zKsZ4
zrozOPZ@>4?vltTnH5vH6aQaK0GsD8Fi8pT{)-i4@BdiV17Fj4`wCKkOol^w)0XZ?|
z`<lp;{QIu)TQ-C#0tQ^79gmS=a)P*$&^YhjNM=O>_HIj<Sn;jS7Ui;O-}iE^9yb^0
z@Kn=}=w>p)0JP>a8l!6mK-7_eq~+HmGEi3FO~MCf@5}Y)j{~wdTy_+OFCX`PZ|$!a
zSCBXLd=?12n?gAZ+v2-Qk$0n|3;Yn|Tc*R$hG^y-PJ07sYqVaO(8X{qs>4dTjF!dX
zlF~!TMEr}wM1>#}vx8mtvG4{8p+LQg*4wTOE1tNGTqR-|;_|3FJRQ4VfGA{<e^Pdc
zGvjL5JhfaiOc$|o6gkR!b*7=InWj=@qM6ViI}qz_HV11$9As;#1SF}+<}|I3a-uky
zQD&R8Vg)CGU~Ge?np^_a%4#HL`~#=9*e*pDpnJ<1RIsZWDjG9g2IvtLDVL8)*^nqW
zf!KINX}D=}-1mSf%G;M_g5J6RZOf<rUpQ=(yVH1xx0vd@&f~6+WYNO~*lFub2!Jh;
zaQ#=reKaBH4%Cr9>6HSW{f25KqXrtJhX-+w6nS33nlSozcZo8GpU^{pQ^gh43(Wrt
zh?nm52^{+y;d^-*Un5^nm*PcI-c`(-Rm?`lq3&+B6Y1P3m?%fhQWufHem&HImF+b?
zSRi~a71HN5GA(FYzCZ$D)Lmx;-k^JbT=(-jZMz0~QJ0RVYY7VnlsJqwv2-(K_(QL1
zu8XY<`>HM2vr!CQX0&=`%?Z;?D$XbU-o`((ZwRLth)K%myy<q-z($Pn5f->!R8$8-
zmoiMXxs-evcfEM-G!D=C`Le&eQn3z_UI9evW~x!LrBGwu1*LN6#S#a<FlMRR(!gc7
z$||1@f7XGYg=%&TU~Q<&n3-b_4&Ne`A11J|6n2=a+hVenN{yUNiq9k0H$Q$6SAH+;
z(ztv_Y^O#$o6daqkRbMZ5kp%uO#1ZeqJY?Q2;&sBX!jc}$7Lrb;xI-c@#cA2q?Q{i
zzPgo9<}ae98siT}m~uBtK+s%j##DHy3?LLz&~hT_UxI{IG6fZaARZ9E7oZ?o^Oqdu
zyMu)+9417U$!RZXggEuvD2vlA(K{1L>MIrF^e_wkr_FMdjQLeXM|iuiY6iE+cc<PD
zNL8#SaIAjoBiU)YcmLu|UkqnUNa(s-=w|<hZt~7p-L_QjZ)xjR6c?D6kmUZyCU_UL
zwXqSietJ2^u9H5~sdnXBBr~HCbl2I$ElHDD9(6)qL4$z1xT?3`U3&m9)LI3n?#5*d
zdp@L0(T0=Bir>s!0J^70o5?A>wi_pa#nt#Qcd3TMOP#ODEIzd4>~q0bhWPPFi+W!5
zEoGNyVPdu{;Fv~bx~`MUWy_T)-9q0d_Yd^hM<<uTt?zh83~^|$irYQSsXwi@djWjm
zMv-Xn#(|EWvQ_RpBQun@-<Zon!(EtP&*k_#c&RWQM%y0$&AbsO`%lqSY}YgKjrS-+
zX>^KH?~oc~z)J=z!*R0~gmmzN{+fY?x_o<|uG3TdYdB^J9fF#;^4-6R2~Pup|13vq
zmgauj9^VVmq=uiTNn931{$d)ERjZLoD;GC&&5o^9jvX;HE%JJAe7k#raCt`pj(|b;
zLI`-e08c*+cXVGbj3vD0lm|+4N*2jGlum&(ajDRrhi6C$65?N8v(Qz)uT85!5BNbz
zoh`D6xE5c9oUP}m1tfrE>DTv7tB1<~yxYG1$=UhZ{P1O4I2a~Y@llcpbK$QutX~E~
z8DN`W%M!_HsF5k?#HJON4&~f#KhYOmhS1PT#7Q89;#4Ziwu=42O07z(D#hPi*8MZn
zM6_eFfV3o2Ev%|jJ$9${nJQ5nK1P^*EeQ`{Bc%BR3~&?rc<b19D;;S2@&XPFC3r<|
z^h~I~`QSY&i<+Xx9YRtLBB0wdAGUEl5I|s~q2axF`2Iz%Q4Y~jgd2W1{#Uwi(t5ro
zSiHE|MLbfpV0kqBNYw(+cpBnH8Mk1v*A?Nr1wp{<?n>Z{oJvh1seYhVP?S<oBp#h(
z{UfM!xXZU1We57XeFSxCRN=Dt3oz`}<Oi!lw-ETr{rClT+P-xL_k*|nXa7dDPwx)z
zvbl8TKfeE!OQ#a)eH|WqOd`t@eio%A<Zp1d#DkU1rPzoluMp@*dvv!OsPUc-i`6Vx
zy&S86RwOo8!!E&t4icX+q$%*fimYjEw0B;s+OB30=<0!=R8h$EfLhh9HC55~XruIx
zcJZj&0B}Yei0b4(D_rV+F{6y)W#`zEKpDsxLvk{O!;BPGV9eCdO<jK$&66til(_Fq
zSWRR;v-Y&HQpc0K=*W4#qkf#Ln6u+66d5b#`G@d<K3ehD-{0pyyio{~oc5CiFZtDI
zDW!jNqnx-OmUM{4N$wmy@j?<kHoI~OK_McF<9Lq)<N8<jE~{aSwFc01HFaXhTxVV9
z2$kmUVTVeKT1`W3Rzb6VWOn0T>cgC@8Peg4@oRG6d;UA!>uDD{ARa+(<$%?KqNIO!
z=nERUbd471c)!~?;8iAikoA2J#=*g!HSNxIx2f(GxG*TspfF%FLuBBI4}7bP=DB(p
z9_EESe@OEI|F!-j<|B=>iVZ@xM<Cq-JW4Bn#F7O*kAFl3PFw}N=)XVry*LqbAu0{j
z*RmhxIfu$ospy|Xs8R0xSd%owx|V{*JskO(Lrpt@wgrXv235I7H9F*XaN`Qv{wC~_
zjV2>}{b704mrgy^#HDzypF><!WeEaZeowZ-yC9tY0S`_10~RB42O8g7R|tyq6Es&D
zJhS0(T^&nu*7saO)g^Hr-QPMd6}sjYt%zmR>?fbMlhR3dr!^8Bn;!ZCTzy{%-B^SU
zLersWoQ<@Fg)GCX<F|4SrhS#qV=4Ar+qhDS9tpmuwgM7aTz%cd&nXR8k__5n68oe^
z77h_Kh<SOnr_rrI9c;t#y;n(oVyb>eE~PxzldtSdikPIr-Oce1eM6Pf@)4f2Fu8V?
zx^W{-<bY}#NcV-xj$B)L5zScp4v6L8*2N#+<@w#U7x<!ezO)`K$MS<EXZJvwou{CE
zf+t$6mO1kmEg9RIR-J7P(lDN-YB$$aaU&I9!c?AzUWYPgsk!_CPP3!nifOMipBm?O
zS&)(4L`~`KTNf3dG%wQz8TL_S1GEJvtwW;SHos9%<m>5?@pZNFD>URa-^3G8Nv3w)
zC1LoEZpNRv?37{~+;#o_xFn$*1jgWl3~>==#;6v2_S&7uyfg7^*@u2g%llT*y2+Mu
zppmR-Yrld&E<V!nyYRcuolYV|Mc@N&_w(oOgS7MgC%l|`5B)v@54{D5Fi2qWi8~wd
z{GAHq)iJ(kA51t6^_~y%)*h<Oq16>Xj%RaL;w*O)@d?K>oehE$61<(%W_F$X!MJ!S
zpy@G)FzfnFT+BkWbjU2vN7&ocB^FlBXq+L4vkxG_gHb^H^QQY)YSi(^cAP?8Vy@U~
zp?9Z;*piEz2YEietINNc9OB=wMc#XJm>ddL<2{B_%QmG<j5Bc2gDqdx!ZJxV>#PV6
z7F#>i8SR)A>?kbE?ra|_yj#|_C*qjfaVPZhp+$+<dN8%W@Hl9XR;k#Ur3VO^$p<dM
zW6TLnwg^vL@%|{?a}mZ8mv)g$kM`Oc)OsEWCXOx!Adfvs7(zsU+&oS<pZoP7oacmG
z2NrliFm8VciZ@wS>$Na$*N>Qn)7WJ>?VX38JF@}z+4&y_^1ZLG@-qqG>2h?x_YuZn
z{;l4?lC!VA+w7!h2I#B}|A5orzIS9f+J|_?4pZl(2y`+&QnDw|Ta<dGdH%)ikx40=
ze5v{0p@7?wBfs}&?vJC7%a2H$*$5|P_o03awAaH~k+F{c*?=3jUjMt#+x~BXCPw&?
zpU8m7h3R1cMs`p5c`1Y$Khw6rXf0RfR4#SZnQT<z)OZF(rnp^NG{5aa`Mu3I1hLq2
zFTqX%XR{p&Al|YHHp9T*{a~4c!+@4fsn9nJ=FWg@-4>2oMkV4u9yNM8t$QBqtTk{_
z9^J&njXM!R2pKv6H9ut<HYP3K%~;=SxTL`y-y+z~0|(}WZ-vyLF>HARH#)!y)i7>3
zOe=E29U(LDZ*z?b7H>!SU3?2b*Nvh+%%YYU+xnOwxcHRH&c!|31v%iKc{vj)H0Zlr
zb7i8va|`FU;>ML~%2LCJ4=eqwkL|eH(ZuIP3BBs3HrHVt@Q69s(0mo07_EKoy9aK^
z>D*FdG3V%VC)-k#0d4mzG5nbOx_N#ZNv9V4Vv(q5$&jGkJDd}ehYb+W<m@S?7h9@`
zrczj)HfwBrxBbfck(!eO6Hy%$;kB*m23ne0iS$k{9qgnmpLD$br>z|u!6?Tvy=~c|
zv3!?xb{S3fc1#ybC+(%}V~lO5xML+Dq)W9XsJ5-{F}gDNu{Zk;U^NPyFn-_b3)1BC
z4?O7X0^Kj+8+`0Z*BCR!8Xo*@H@*w}xUSDyuuYb<qTrcoN^a>lNdSvk4A{^ti`J)M
znw4c}xUlR3%9rgHwfS&e=@demkj=}dYZj3Mv`h&)i0Wm_CoU10F+H+L{}_003+{Q8
z1g^2$(#nE_EL`hq`I-5m@K4@cggI*I2H7aLT!iU7#Sda~U|xh>2?R(HGK*CV$h9Nt
z^Xh)qRa41GuSgm_+Mzi%_;*bxtHbA0sOmIOpAe*`u>P`A_WD%Al>@d>YNN%4=}NyJ
z?p7e4&GHD!hAGNjL%>OimajgPLvd`XYw)SFu^C{`Ea{{qI}~{5dbile{&jEjIuQQ!
zmp%hF+G-QJYv=D$zo0Z3bDX(u`e+%ZUjvF_C37hiFO0ZNo@OhSUrDsI0r#dC90P8W
z!i`wvU^mk!qFj~%if-+%z{{+bFfEkHrGNwJdZ(xJCFXs;>}A2)nlsy2zz%Dw-<42r
zv&@gG_0l<U%n^^Eo~tN->At%2=PN!S3LUV;9?ci1MByDgxI5>zYJ6k-aTo}^cntoB
z=i5UMrk>x2J|&zVR}T26j`0X=?{lmn>ypr3>Fgn>2;1J<<sH@BZ{dsJYgLNFOKzig
z4!ZNRL`+mq@E$|1DP^liNjRJ2r|FJ?{j?1ya73`Es+r}s*J6pWyCjT3yttkW@?-m`
zzi_QEKr5b>a~C5hQmtRYI*ry+I^wjuK9~Crn7fiHMy@AxF=bF;;bNoVy27}V&(0Xe
z2r;D9kS%4)O+YmnbL1oMePTaE1!hQh`|wE;n|Vyf6q^s&3}Z4ZWzi)pPx-D1p>Jrh
z{SDcgF=R>Ya`2-rKbN{K{N&s(`LIM>gj(PpkD-9Y!#a`a@tq}OOOL1-$Z!yCZ38NP
zII7YZu8qSTR1Yq>(>T(jR5mf<I1Meat(lzfR$i4+<#_>jC6RqR%A;v(N9x=rFT480
zwQ5k)5BS(lFjZoW=<wII7Ra}vSw`lqO1d%(FSi=$*DE(xoX@Xfe#?XW9j&_kZlZI%
z<ul-Cys}#B70d?Gf?TfXW3vB^=&RXExs>jyNdZdhPDjfIg$B=a9YMPf6RC>cEscCQ
z9AQj3vQB5@W}|o2UlXQ==NSCYS&{4$*WGlkDFOaJ)_tD$)}n?G3eTW%XpZo*&_?^Q
z=(*KPF%XourJn=(`JH^_<!W$96X-fIet#u|4S$Od9l{?Oo9lc{cl?cK!rV~dvzNLg
z%y<>*lf}%S@lA=uNHLE_X}rULdSUq+Z<PBLm8^J*(+h23I;{l{ZCcBWlilHavz^JE
zLY2r5y7!xeDR~Ff%_Fmv(ab<qGb7E+4D+%T(&xIdmvma4w0V_h`N_IFSTa-$H<TYn
z;T@LgCD!44D0O$%no0@MGYf<JJ`^-5IjT9BUac#xt8k8W;O>C8^=u*=nxjP_%~KjU
zO!DylPD_vnkCayBZ;T7#)Onp`{ovdS>>!Zl2{}DZtS;V*=s9KJ?Sk>m+JzOMuiiS?
zkBm=Sa%ysQZREWga&Wf!;v_pl0B3$@eb>stm(p(UW_gFjVN+maevIg;Rh@`UpGJ*0
ziX~QXD#x5wN-xh)v-E0iIwP@->6JA05|s_XYttjTX_aa6u|}QEVVhDjePMj&k2$iz
zVw9Ui`7O60*}1iVj@O;knyhA7=q!{cBfJyS^KjE3eH1zwo$wBZxZ72$_p#)3IcWr%
zSW=PY<z#Th*wfqksQJbJh*jpWT9^SU=rB}1YusySwKR&vO~`gV-nx_Cg<${AwF7>f
zQ&FlLZN&>UN+|xE9l3HB%m~3PDQRuQ$>sPR39E0HW>r%)N`$O>@>5X%Opm3+1(PP4
zvQ-qw2d#1V4jECoIFBY{*}fO6hSY}<K)16R-99fdM-1iFok11iRUh(t&IrxAZ}tbQ
zCM`?~u(svX(@H*sNl3bnlyd4~Q{o;WD^0>hCe$sZLr3J;c^Zx-7=L8`eX}F&(>1~M
z{-<DgDDDOF?miMCd3ru`^6>?Sb6dGK;%MY^;2xLG52EV_7S~z;ayfx_sw2>bbI3jx
zwLz!RpYB;yxRW$guj~V3V8oBtM*2HYm}qvo6Jwgpl}oSr6gSS{0lfWe@XE}L{7%9d
z2CJ$68yaIf8%!h9yO?kMBJdo*me^|!>foR`tfS9`l_H5^clGs+u?#DDtYc?!`{aW2
zGdjE792j@6Ew7#cMK5pkrzq7tuUYSF2dwQ44l)fs=OQZsRjq(QYc9cO4J_*c*&eng
z$&D+KltUl?=bPGn<$Yrw;ZHjGqh$8>1N%8i)p8E*&*f~cbYkv#y;m5>Q%#BDTx+Va
zJ86DOWr6ATmXoDw`tJ5*KdZjZl!L056l5hQmUfX#Y%H16Y=D+{ksh0X^alY2ka=Xl
z?!$9`;LTCr&S6Xr{0~je3`0B|6Z|68Hn+l<Tv>$bo^1?4XRU&Z=W)%AcG&o769OM0
zW#W~k63nIUN`;Ke$eB#Z{GR974`AT^HbwK%g>he(jL^Ss=M9E-h(kl<dMf`dS)uG*
z(1upq^Ove&j1Et+^o^Lu4hwBFFm|0^qlIKKOtbmVLnU!Jb}nP;J#thKVRqxw@KBSp
zo_5jMaW_WiFYi@Niz>z=f&FZ+MLqaujJeL>=udtZuC9$+te#opO)4RDXi0h~47E%)
zG)rRNgy+KWDv6<o8oJyKhGVkUBdp=vZZx=!TQt7%GaFouSw1^74ZYZe_mp=L*638o
zF%x2R5)Am7;;WR?CnYN`Q%HpNl8lbtUeD})`6A_kLOVytP&iNj=Nh&Jk!*_X)YDhw
z!otb5!{PG_eDJXK2x##&rDXjO<pywMUN%sdxuVLTbwCNbtt#nqX@24<ne7#ZE1f3C
zlE#f+MnjmGg3Zt0qJzo6rHhK!>fa-B-JJgxyU7M6<|qq0#76K^tCuN?F_sE3K=QO{
zGp5^7WGqD={csIeWBh{<>{gKTTFTQAzE@lxlO*zTn1B6rf!`h3en7S1?-7j1!Dm1?
z!Ag9l!656x@{4y3r!+G|o@;rp?V(-gxQbcxOs$Q(k5&$*;k?|9`M|Ro^C-JSq5>;Z
z(?^FEL)jVg2o;cQ%f5a06aF4H>U>2eAEoPuT8W8sB4aZi<JTPbblWB<kFbE_ld*V{
zjY66fH4c#~Tl>E(-r*E`X673gNPNpxp>^z9&gQM5%N_k{wFE3xv2T%)X4$g`MF(sc
z*NUB0JE7e8EyrPpdLyW1V|;9Mi8XxCGhw4`9PqoekQgfj`odk6VeEi)xE;s$LPeT0
z@!qU|c?iFg{i&;RAXm1H+bs`!hg@JZpoCsXSOG^BMz2t}Dd$QbwZ~9q)0T4)7bcH$
zaVcAgM0>Hw?{}9~Ha7ZE7`LEAkL=4x|M1Sv^W5F-Lo{peBQ&$;grU0Fzmr>iIjLVP
zQkSH&#Ip(B*rjt3r1T{}9o3%feWUfT8KIOI>jZ*{ayVklj;Xr%uImHtGr9%*M%zDz
zk_1rdB<6Mbg_X{q36D4eKbAzU--f_U9f{RNcHjL5a0|aMHC`~>Eav$-tZTmTXDh67
zE9Zh77&<=bBw36Qxe$U3bZpDCj&CQ~P|sh2+P%_}vF!AmDHV`rs-3e*B65-(SFmkI
zS6mwr9(O%vA5v+6JB(ectu13DT!9@Hiqj(xy)3Y!b#(3|FGILAEape=DR%soG@FMs
zNg<7WZkp-byNfyPY1^mRj1F=yoM(Cy9lFdF>_va9kNi1HGVq{xW6YYd57tgzl(o*9
z@2?xpv$$JJWHmp{X<0c_D10Hcg-Tx@6tMDRPYiU&B!gz45&Tlsl3fBP=zYv?yj;Rr
zrKK;RDp3AAyHJl@j3Er&D`A6#Zbp_L-$Jd+K!CiW@gVSaf7a*i8BSYU8=ps;;lU@{
zpq(Rk7M5)7GYpw90K%y%IrHSZa%q(2Zcd;0EmEC)imG;<e3Lveg43C6Hw)=^9WvJ&
zZ`?M|2Fx7SJ&pAVP*m@WI#q?dj(2rA`SA7bUj^RTT`FB={WYY}&hb@3@$U=MTQ#bT
zOD_Y$v>~VG(WNP%;4*N^-{9n9Y-0n;kLBiVqoQWqI4_U=%u$p0>0IO7<ACobpBD{$
z_{g&AYuAXx$iqb$P#7#&C@?B~#tzduSQI*)aZ^`)!o_D-!KnN3+4vxLOFd32W~KzI
zBJ|fs%Yrk3I%9)I1(YeD!TUeK6d+hlSqcL67Q*@*!Y&_V<h&}j@{wd%bxAiVTZ)WA
zS$bv)9=bb21>6>7wQohat^22cpe0IZMzwN|zV3<zZ7G4sM6bwp|1-(&jxLwhDQ6gO
zqOI5pSZR--tsG(sC`S%-I+UtZuMTL5MjI+M^YV-RBf)~UQ*&iGu*csp0P<3;$Q^OO
za%w9q#hU1+KrPAHBra+F5<L379=Bn1<>TUO_RiGWRF~fE_PcYiRifJ_o9`PbL$BGx
zZn7m#*Qe437CJ7Vx%SQc{2Z)+6<zf*D5F@+re7&Gsqb~9Z}0B!>@KD|o~J9Y+Z@Dy
z7a*r0as$2vPalqp2)%#4N+8QDTkm#G)fuod*_M_dHu|P!$f>rdMPL1ib0(9j8ZXw@
zA^^T9-h+9%*XK0SOxFs;?FPVJV$|056(G)~-w&X!O>_TA9XSuioSe)?R-c4L_r9As
z@%y1SJ$ED9?u4ZE#)k1co~%8etpf^c)e-VN7Kr|OWKc<2bCgwp@tn8%s$<KtsMa9}
z1e8UA!RWBWxqRMF!K=Dnm4D-=!RY^neqF`66yx|Ey9l$nQ<c)Q6AqP?Ok%)z{NnS1
zK>Wp0U>${o9ZsTL**)I}cGqoVw|9T|a}7}^u5Y<jlLJ1vO@d4`Cv3!q_lwJMd@(#?
z_82_BRBAF9O(>m+CBzuY_e>qw8+K>_??s~<U=y$p{lfWME9`Kzhmk8m_-5Bd^BzfL
zfdNxT`90<YQr&lPf0hn{yb^{9o^J108TdKL5)|$Ro2Xr7ypN8>B)Y{VvQ~z@oVK`^
zwcv5CU?)LM5C3NG-?PQNcBFYl4n4uyhjD5%7^l`-%yx&oeLmt(7lYfqZ)6>Uefv4Q
zSGsq$viU6?^qF3<cAsG44yDa>k;1=3dEq&+jJh{Zb+hkJyN&G+@8{B$4{i>9>!5b5
zt0y#UxQB!b1(XsviHKJs{=5odfNTzfCabmEhgmo(zFRWX<1eGE`|+%Y*#@>2pHay(
z9MB|zvMQ;ua5s=g+*sgV<+7QSzDemTYofqOPo{Z!St%u8J8kvS{mWaG?)CE+0xrcb
z_)*p~x)~7+_5c2Y$y#H>jhBYF6_NfUyYMG9f@Lkh?R%ssHExO&5kd7u2Y0$=&Wk5^
zB3aE}DbEJ3#e_<&Ehua5w*Izw-USSpk_)<JF3eyoSX&nBJ6d&*#J|aPO9F>z;d~eZ
zW|C#ex_|xBRc=6M%1Vqg7N-*_DBb=-jb(#PjchGG>d|cOL2r&b_<gT7cRi=$7Nv{9
zDZgwFH@nb7F}&YyY5aM2RBxL&NUS<XwXU&TwlKw%4Yg;Eq87MZvaQr6XagM;Yi7WT
z@z8;{DaJP`+Qw`pcDE6~W_@j|6x1#AtjZ!d%b%DRe$H--yu*y$7e3j$%DSZ%rPf^}
znYx<U3No&DyC_}%*)1~LMBUDfN&!yyt1lHgNaZPE+tqdlJVDosl{cW|Z6WdQi((X>
zfgSp$G&CXBaKDO3-ujxN_|*6k4Lyf8;4?JNBEt;Qi%Y?kpjK_F2Mv(ZO_rK@u`}5p
z1o)n=V5CJEU-xAH?q;_Jcdc5nBDewNrW<RhuP9v7fi(tMCDRS)J5_RJ6U&-Zctkjb
z^n?aHV7GH>l*ls{p&aaVa6f$vouJ2kvjw))%KkWeH6xsDCs|u=`eolF(w`=$izkR(
zB-;$|g0({R$A+Pg!K;8{NNeDn+jGg2=}2tind*}(N*p;SlT&)+Z%ChP2cr`o4V&?d
zK-tD45AfHG=2gFq1DAs8&UDg4FGxWhr~U`tLi;^O4<pQGB0CzNh2-JJY-KByFt3io
zlbXzx?y?lZAZvui*UF$KHO?YZZD^67H)a7(#rZ=nKa9e*3<>3JC%T^=e76Gbg1pA+
zy#WyC&uboyGH9VB{a&WA0qr6uwTVV#Iow?H(`Qk-R80m=7Z%E|2b9!b#}21k<3ll!
zSpN9zjs*Nz0sHyf5%ydizxH!CeB~q4RR;w=wtu9b_?1j^_WE10d&Y(oJX@pQ7u811
z<9ywE$dBbXgu9v4B)Fkq)3$p*qG7kQ{3CZntcJX>A2q@sKh_2hn{3Gue=I1c+3N#N
z-`}6kxe~!BE<O{wzaQxS9CNX2*C?pZEGqx5g*gl-mo1~;FsawTL?`VYL;x@UjaeEH
zq1Tk#r8r}JSeXRo+E|#_u;&+U^LVv+RVK@as!`rY=iz}_h-;&0STkL-ICss}cFl@f
zrxGS2X|?8b&q%*+GrT-+TJ|7e@9&&h^kmR#tt(wi&g7cG?fO?1&5p$ul_A_!7Kx}t
z@pG5$;oJ*mpKr&8qfE+b`^u{f>=o~kJ$@(V5$(1_S)5z`_x_hklc8sfRuMQOc#wCp
zJ(~Sbm7cl3a}W!fpUVqPe%~W>ZV*z@1C#D}v+J}~*-y~hGw#^GypL;;lg{LNKV3FM
z9TpJ^;u`A3XVJgpX>p)#|Fnk<$Zf%a7A>}I$8l;AlO4!C4^bOte-(R00cunjYk03S
zBm_ffy+5oKXW0~$<U8WP{_rZdP!|p|#!Dpn)-;K_Z5~YA=DhiL`3am<*;K>xlg>{^
z3VDT^47&Wvj7N8aHra##X+PM0!CAgbm8)Y=*s3I_pVOad&k3Y->wTOnJTp9g3hzP`
z%N-jVHMBI(%GCq%7NwtNhHuT{?PS8qIZm7bz2gYHS~LVHO8LSQYFI{%B}M7c=8Mkk
ziulUswWncq3GM?ffMr_wJqA-<VC+)Sf+Dg2?-LSoZ&3fNF6{Roh<O?9D+VMF=?1H!
z`M?YlLwwk=``!L0chJ9U89lA#5f<eEUzqV+cDZ@e&GDW5<F7N^won&uya{RkeMckj
za4T=jzJViMcM7=*KlS7-mj>U4zLg)w_q{xm`4|X0Po~3ILdDmPsHL99NgkLp&hzxX
zT6V@YMzo23x*>0kN{J(+v8U}qYhhEwrpaXY&`#A!qsSi9Awz}wd-h{<-$164&5YDh
z*t_#9lP>E#v)%jpQK2dRRb4{p@kDDAXu647ZOVUYZSc#osbyQUK0#OSr;m%b7=IA@
z81~qH79^Ba=y@&?HN)u+L*Y%0LSrExbq4)o^unn7!eVP5<4zeBO(4?FQLPTD*37=f
zz+W#T$9Qx3B&{4r?B!~T%lp|tIKA<cV~%?hu!{`pYg6t-HJsOVOFB>=gXh9rlXFT7
z;L|NRmOcAmAP?hORhme{?PekiJta2j49Jj&3dyoD_EdYuSdII}YP{vjcYdzR2>GPG
z%smPH0e{QyS;sUGU7U$-p*CpwRxXi~v4s#_$HcIvtz=nOx86p}9%9KJzkil9`?!WL
zLUoYWDPV9?K%MRZ(xJONgW2Bp9~ydMcDfZ@?o?svQvJ)>_^tL7-i22s;d5jg*n~7G
zKDpzy-TE$Iz!ui#JlF%0bFF(=d8~RgATTXJ4;otSo*3=VMk(J%8n;Yz%n8l<j&Igy
zAZopXr9Hc_H9sPAJx~%-+Q3xJzgekf2Y;kPQ*8PG-Bm~Tuut=AA6$Fbd-rcCN7n<_
zRjFk3c<waGRC&TTaP1{Tr^(3iUetZGY33_FPbplx&hOA@m?Kt&Bl%yG==eh;|DK)*
zA5>#wUbTOy3$)aZSNf+a*()yDE+d{+`40zwl1A<|x+QX1dN*)S!k|j9ZS`=hg41Mj
zXkQPmWx<}G1bjtb7`CZp3YX_5-#G1cIbOZ`IXdugk%~2n*CZ8#J=;{{Za8*vFN1os
zm!>aYDubfnLfe5NoIK)360=x#(yaGj`cR8o9wfT+(|h2odRDg&<p8PM9cO~+Z20Bw
z+lTiv^OcgZivpPp;LoxOx93G0bzAdg)UjEUxz$lU;k2p{&McgL5d%?Ki<(9)#i7vF
zcVc+7{qY$575C5m0--YZS|N1|H@tyhuf^_UxnbXCmYXeXvlk(`9=3*aAEL<C?|&+(
z)o_CacZUB8Fopq8D)BgQuLd+ta`9k(8IFuu$GSP(isDb;TqYz43C;NN<*lNB`eU6p
zN`U#jS{A02y;|jNZSyp#b|77kKzx&ldboX7)5X^Hd&>Ra+>(J@eT%ymR2C7Q>R$;C
zB!NU9MK?(O&s^_6hIrpSH{CTf1U24LKwDwf(hcU3@pE9NsVEC2_rkFlD}3pGsBGb%
zqg!{)B5z0$_z3U$AN75p%l9B+yfOgud)9>xFY*fmjT$?AI!w{Jo!edx9o)IV*pkUu
z;7vy0;L<&*%5rK212({;=An3x7#3FySFF)=y}KQma?OuWVC>R)#4oQSkPj<FeAxX6
zq-FL-*L94-H+z4beD<C$?jsx6YwDwMnmrzKyZhSj1@`judX0~GiubxF)43lvP>?%B
z#Ph1*QWVlc+{^KNjrj6B5;C205)t#BEVP6fi4`fl=NDnu`9|SRZTvq}on=>C(YCaM
z6Wk@i-3jgzT!Op1yVH0h!QFy;aCc}VxVsZv8rR_V_8ITFW8C`#`pX`@*RC~Z)l-PQ
z<CTntpn&h)zCm2`k>3Vn9tN!hA=rZ9jFQ>fQX5IzLrD<~e+kp@2fJjSmR%4Y;@ILH
zG9*D}Uq4DuJD}sjpM!_Dz`vk4GFx=|gP--UC(yt>)H>w=B8pveEvHHmUfaxwijFVT
zq+%--uMPvatEGn^$ZCuinHXX=c$w6>eJZj;6Ra3|us;><76FP&2pp<)LMh4YA)8r#
zs7L&dKFSX&g^PFqL5PiSm#t*vI+FR1De^yp1S5x%c+h;4`zc?7PFII#StfLk1>_N5
zSM%Ou`lA`&AwE;DzZ)CaUBE1Pz{xkTt@`cDpdhzur(ilwsoO5o&J<wCVE^RX&7f5+
zffo$@0iKdI3bZ<NQpe?ak6Qt@v}$sqEf+U<AY%p2Q%i`KBb&Qjl|1!FdC9pQ{L&&r
z3WZ<5Eq1`u>hvfn5jubJq3xzo_tp8E*Xwh?mLSVz#%*G_tEmHrIBT~}i}=32wq#aQ
z|6!`o+x<%mt1jz$1`0MpqFjQEI}pQ1?C3a)zWEU0XTbK!eVSo`8luTSke^by*fr&Y
z7rNia`2*}IFS9p3qA>fvI6nH*%#rq{C*uvLYczaOJ8J5KY|WF*iP}1c*YOmYUlIy%
zuFfeXNq@&C$6*k*uzk!-H$=M4_-nV3Hb1LCQ{&EMABE?BA(9_)uOEhEAzoRLER26#
zA#dOHhc|L5U>_Kn0f?PAa{(_gjsAF$WpNJA?QyG){nUDM=A_%D)ahTp+GR~=x5;|9
zWL0h4aBLN-Y6!l(_NO|?$Zn{w9;lD^g$Zbyk<=R(ak6`=eLpw4-MNU9=4;UWrD^xq
zoubOEglWtra|D+JTVcu3P+jH6*hv>SdYQGodv*RHZR+m*O{gYkNA<gNOlOE}w>+Dh
z&E#XnKVdPJ6#Wd~3c&PiMKHtfD8hgmVfZ9921-KSg|~Ij7F0OR^Egaw<GG+N%OtoY
z{l4?g^OyM&z|*r+;AdPz+bP(N3d>!d$8GS@k;(Mv^Our&*6u5s?<TpA>D{C0&8RRe
zbS@B$v#u(&pSR(Z(rW{j{Av(i#B}QcL;x>F@VR^M+B&ACQHmta>iux*gIPcpbQLD-
zxls7ez$xi0W%L+8(#5C3@OgRF4vQ;tMbPRbN%sg?9i9zgX_nf*OrL~xsNe9pxcQ29
zy>2!WiJ9AXi-cTiPhawthV7*g_K#F<Fj7T~!n;Xf_*tDDIUb2Z*P#7n6*KJpdKpcP
zTPC+bSof)^@&3Wcx;1ZuYSjl|75HkFCH{QYck$yXoMmYnadzswIXiaNlYri$WM4v;
zy^PXZ54)S+lxOBHPUIun$kokv@Z;R*1DEvTQs&fFftpLHFJr93+*b2dR|nZo=C|v&
z81*X#6ms_aM99w8is;@~z)k+};v(1N=(uBwz*yr50>li}9Qbl`9)%S}rmK)dLpT}t
zhmQ)1SM2Oqs)Sb>y5Vu?B>Z%P(Xlo_HqX%IeJXj&_XaxtDK~sr;C-hxu>a*Im*;?4
zUMNxLs@kw-QSgg_*>yQXiKs3dF?1ntJ7x#@S6{vxl@(JnqsXqg{BR8sDc+4$b2&zg
zf{exx!P6>hpox`}s-oR4$np?W_j@23pXcl+8a;wzI8YS@MGIW*V)T!qS%Yglgjykg
zP}_4eXq<o_c6jeNv<A6~h8P!95lo04V<4eo*?j9UB6kdb6~<meWzg>XXly@KSNLbA
z5FmG=%f~F8rmCGprm(rrCoO_P73Ri*Pd!nUx_~x41?MT;02{wF!bZs0ps!t^w7q!A
z9i6)dO$z>BBceY;OAB4REk}*#r8XI?DhE>?#FC69oP;q>gE=dl)pWC2?=RmBUEJCZ
zZM`f;Ip-;(3H(W0PQ!w*qa3A<(tpz_3eHMqH%G0YC-yt;pVXNh37o1H|1iYHqF~`1
z2WW8OH?e)@mR-ba&r0^77PuSQWcaF)DW7h?uG^~E7-1G6Q1krN??f3qQkk?omV`%=
zE`4$M@OSzW@aqz=AdQ91+w}@~LC$iLYUc0#!IRe<!;P^UY44nf7=@h-XbNN*vOpz6
z6vhA%(vkiIuAhF5mNmrQaE3dP$qCNmtBLw;KP{8HSw#4%D|mVTzRI6p7>c1%o&q8e
zRgSHQU30@-=-}Jzs2BT=abx-W0>>p)6{{I9nCD&+#qy%Y<5aWxVFOJ7^J;S^WcIXV
zv1iDTjR90ylX3nfU%Ar1Ice5WU4&Qhj!>n@V`Tr>E%2wv>uh{9(EFt~w|1v`VCGz_
z<-(|XQn?8$=Z)AArl}nE!H;;$xmloYit!Jh#L3}@9B#d12k(-;j$27Doc<LByi_Y^
zG2~I!wR)dzcRd+$eHvGVu_0DY&bA%JXJv{9#+HpzVo+PjS1>AUs-Ie}*NAvuQsiz!
zoBeH!UtW;Zm1uhU_;EV%<~RH@uQzoJY|hs4?;iQpK9Q|ir@^RLjpw_h_E%f%`RrkA
z{?=oLe-WaSTjD}mo56!@?!?P4Fd>LJyaR*qVp3836{Pg-SHe<-*oAJWXI_sj0X*<a
zOAHC{p=732Z58`T<%50PKc=h=Ju1f5=%5GRBLGoN52q3?u~OSY|HP%YJS<-;@vOSp
zH2vz9-4gp}ms~kT=M4^Xw?W$>fM!_^;x)r`!fVH~UzYGMuRsyKP9@tM2lOZJpG^eM
z4y``VVOBq2>W8B2@T_zgo7d901%edP7#A_SXosxq06kV2%!s`~mN6yNVl`q*X?P|2
z#rvkw5tH8hKEsYgE2tr(9u@UaMU9R^i%YJ{(!R`#zj&HgcJLF1syLMBFy;Fgm~+xR
zNw}uHdxh5v#lSw#7hkk|PIhMo?U8=F9;T4C8*Saz`E9ptvbkAWG+P6F?-b&O4_LXD
zz6+e^aq;FV5rOs1u-4)DKYfP=BH}KoC(IJ63Bc+v*ON-s+zU9?)?h4?W2v}Q(eKHT
z;g>|?)is<hpcGe*)(xzlR_I0jwmx+uyrRj{!iPr!tIo`UqSowcP`?>2+U8cF86lU;
zy$}U_=#o6=7F@=%(PkONazWwVVoL1OlfjvF@}{Oi#(RZWSPHz3laLYBvL}m>zrpJ!
zOZ=-2=8|nutodubO#D8=SJf9jMz7Jejtu)|BL^mVHnK8J+vbYDJoJ{h(kQ_0c>VL*
z_O6Z{+dyL-Jjj0AtTjE>o1RHh$8NA>K$i*XnmrXG*)e=%=Hbkfa&Ki%LZ-c;rc&0o
z>q$ViI{h~kQnPkaHKcef2~|U#Lqo`dOH`iqa@O*@gKtDs$jQ;`$lVo^md_GLill7$
zO`D)I56>rE(qFu)GTNVdTPj6FlB*3lKMV^1k}E!1?FsXfrK9%~(GS01^<hlqXNNSQ
zS?j7IK{dsuDD5e!!^abQfR=$<E`#7Aw0=>PS?%d*h93@4ko!q>8hkgQ{9l3=pX<sW
z)g%p_(bcGS`l5LAu2c-d8gk#zIVTU9!#M)^NJCvB{`pJCqzT(V$n5ThYG#sA(rbiW
zVYC`3$~EV{zL<r`DcguhBx?TFm6Ul?)a{(fxwXNh@8IFqIFZ=E)kj!k|1W`cCP|+<
zH=Aq`I$2kS^cD!j$-VFQdLhG^7xZ&6I&TUfnv!)oEywPL7tD8GvB=j-qc=G8c(2#R
zo~PbShSNp`Me%__QI~Srr7u)Pb@2i<9d?>p37<sXR_#1ra#`S;EPvCp<yp;vZf*{#
z{g*qKOKu)_5-l>F%PCm&ZuDPr0U5IGu5!rDyeYkzj1_Qxy~7cT=j=1Dh#eSDuZPmF
z`zKme=yARV2?V^gR^9C%Ts==$U*9h_w$$89l5~F2X>*z6IE>>|&{@~_n^wavo0Fb@
z(MycRS7u&!!vaabN?!`Ggp6A@@e72@kcGxo&a$AY;XfUx8{Jqzt_Uf|G3p0byBxNd
zBwKlp_b(4u@kX~7=K>^mkV81v|I&F>^_U7i`NKXP&Ea<?Y6P9aZUa;-A7mK21bdEw
zLn<oN#bI`w%>YqN;`MZQT~<WWVr`boq$^$s9Wz{_z3;9t6QKsAgq>67Kf0)KZ71pE
zB8aCL4PcXsYS0eq0o6JUsTgfD%K@eHod&@g)mM!ioukCJ!)|wX#$-`9=Bitw*wr*%
zjLld@rKAW;x_lcV`bf5G2WJSUZC_XJV2ELIq~6c0dUo&bkV!=R<QeJ??W?|CUb&F{
zBR<{8;L~X6K*px*h?|9g6(S9`@kQQ}-b{=M_~pv86=`qR`pGAqx7v+`oxmL9oNOY(
z6pOQz_{ZaZhGm6B)94PunFX9*>kd(jRk?nfET6Ys7}dM6em=!LT}3&X&TfN*XwZA!
z(f=}&dW&?~F@=xmvNKGkbjj+oq_?Y%8nd5RL4BG-@~DPyDc1xcTY6UmgD8hAgA1no
zTIDRZyk?MDYiqrqR;^xlmEKe8qyMora33G?Yq)AETK4_f2uDC8KI7MIAPOa5(y0UB
zeMWP^l~;4y{@NCnv}FIUJqk*P0dGi!Q5R`fDCXWO7=L7&-FwpA@agOhS6vDG0DBxc
zXQWM)5@kII=+#ioz3QeMh~t&$z?fm!w-=6z!0?^U2dW5O5S3S&O?E?R7A%g?h`6?m
zmjopDWdQ?VL_qxl-`&hS0;tSxxszLlm$<jGgKSGQP5uywhM(pFbi6^Gt@%ipb_{&&
zRTlWfol`--H_wlbALsMlqt^RTFT3A<s%wp*ED9{V(0<Y$$dtn;`yR={-jUn=@XvpH
zG5V|L3yFh$!Li_cc&>1-pKyT&;e&6sw=KirEN2cmJxrY~b$VaJ#6<EpCGb6fp~OJ7
z65kd_A2UI8X++^8iL~bt+kC_2gB#n-qB^pt^X2O%@z(W2e4tbJk1CM_6wR<Clu&@;
zB;zVo%5Kae({!yCo`KXqC{t>07?12GDv5LL<%x|qnB{&M9`Bk5UojD9|F+%dliu@j
zJq{Vx)DD&uRo)!hLysZf;eh4ar*R>RT%9yr){;>KI}UkA5^Yr#EDd}!s^>!?H4}x=
zFRHQb?K-0B8gaM5-tj0fo5(6~4X=%Md~(vX#Qw`ggY~DrSI&wwm0#{D_Xl*bIvtCV
ziTjO7f_Jj>@%(DCL$E@2m(f_m99OXYzJFGV4O@zFL+$VAeV6N`S|_KbaHxI$HMK9#
z=cY5PSF~{Q2mb!tH%aTaMQ}=@{^xq0M)w|=YjrRog5L0pvoZet>mbBk3DHrTJ%H&H
zEU@HiJn<Z=;6&K4egT01p!qLVcINcyHbvgXrmKIW*^0*_S+R$O4*u_3^SeOJIUA{p
zxyyS`_!K!gMzBpV)0sGxg?zq#61AN$ODa`lYHAbJO_HqasHF4k5^A~CFk?5-d#E7p
z#YOdwV71lzb5_r;31IqbccJpkAspF_dNQAW@)53GDCcxnh3%;}bOAg^6zhF=2!a)W
z3uX713abw-9V=z50Vl3>RJ2ML@N(w`3cMYfk_D@PpPpY@L6Fdt;B<<<jN9ShPVdLF
z(OY|o>&-%f3YUnO9jAJNX>%D4%*<7pb}zE6S7^2Z!^z^RlSV!OHE)`cAs6R|6fx}B
zD(~)qjKtH;Fzd|j@mK`^?FtG@0zR^V?(LY1oW$}TKYDl3SN<i)`g!8D%llRNDBJmy
z>G6@=zKalt)m6c8vI~p;O7ff|TMbl$9+F#^pW**i;6IOsglx*%DsY8kCleVpWhxr>
z?SrI#Zs!0cn?3q?KEm(abjc<&4nSBKKWDCQo-<1jZ#+BvKE{%J)YME`FMfNt6u~A>
zOBP&Guz9;Gl3BX{QY}o0dibLnbO<L$&Wy}uAplzY`nThu{MUqaHDj}nZ8>ZY5dZv5
zJ(^g{%IBWN(|vBQxs=L`e&QH7{FHl1(sSVW-m5x2(;<;sikvL8Y~`SG#&w{XZ-``L
zQPh;Lkrh93pD9Rk%YR*c8vroVV>@RM#|OB0NFoALklCaJ)Xnl$b?hu_BKc~oEU1?U
zcsJ|5LYF>}&gfpXH${s~NBQ~Re<%@yCav#A8H?+U7Y;=Sfsd91HpJe__b(*XhgOR(
zQ~X`_!p74X2-TD<dg-1ocw>WfU!j&K?kg#^vRc26Xv(#~vN*BzX>6!1d;5Cb3@ofh
z9i6WRpoo8QQ_8%T(Z~vzS#nuj!A6pwN~g0jTX(<-*Yp$88^+NF5BTXK?h<t7rt|Pr
z`2>f*R9{D-12N1_d4)ZeV?GWeOzD5W&CCwVI8RL|*x3Z2AC`M@UHoWKX2x|_Ol_)4
zOK}@zvQs#PZ?aE6s}Nm5DvK?cH;m*hyyW}0?9DHSmBv+n7{)XH^+J*Ku?_^zjx$d~
zma!n7zfbT8DXroNfI{3^{Ze=N*8h1>vg4$wdT}YSWxRA0vzg<vz5a4WxBbHzVOeWP
z0$T|a0xlTOd~()r6WQcowW=8=?w|2{lN&_ns){e2Rzp2=sV?dS$Tb^z&GDbE%^6CH
zGRs!swU3n12V0h;McG=MQCgFApw(hZTL%XE8rvo*4QLApYl|h@g`Ev`V_3ytv!dk*
zPWi$RRn!M_E8ow_Z5_Tfk{;YdI=pM&obmV(of70$AB^kHA%C|>F3@r@4+9wJxY4BA
zrXx5-=Havc8rs|_++DJ<xDcELc)s<H<1a4^b2$BOWGff%EfLexZHe!<AO^S`x-5C2
zm`UTqYznCAe*L;kkjiDDEFfTWX=Z?Dt4mL>1M3H0uGbU-EVJR#48C0_8axTP_d>9O
zyay=%8p2KKTS^ad>GJ??uemSBagqv!Zft*GSfcA4ZenX4HeIInpR@d)kMolqz8Gx=
zcK%+aMi)qL&ddRx823!`$tn8Xt=G2otZ^t=9TqZ4Lc+*#Ho9#B%oEAgd;YxJ8@TNi
zvhba2y_d9}qsft-TxR-ut?Qdbw&@bI7tl>v>heLR>BkJ}E@eJ+|2dki2e8+^EJX8^
zJ#5CoGs~DIJQ@SZHU;GC2A5jg#1O?$h72jQlzwr4K5LP5BngVG?UR~ap?XtdIvTQl
zl1aO{)EhRwEacWy9#p|t(v`6_k*W`t&K1pIh;?A0`NJ~iYWS;zV^-k4zc)IIF(ESC
zhTXUyPX5H+TQ+Rmu0DB_D%5~8EzXTQoEsT|6uMb^msg|u!oK%ha5@2-i<*wvq%D&t
zx3U>%5xL90s7K6X8q3MNEl~eYzuhuHi0z@6y?$xn*nPiA(76W$P9c$51GTk0tF;Rb
z4%OO@7SssU#*EBgZl&1PIOZn-|I|Fe^Jhaz{?e+*;gG|JHWc)aDb{^D=h@1%D&yV!
zD|EGUEu-+68P|xIuhcVKRh7Ye>mF|r`4k1<Nm5@T9`ewl$(@hg;<LFIy^$D<+>)36
zX}9d8T#`<<K$pGBE{=3>hH`K|J+64DLp|o)R+G^o4||Qz?uUJHO{=*DEWTj3dNB4a
zl%o<+>Tjaele}sZ+hzzHV!>!V=Vn(2paVrZ{l6v}Cen6Z8IkNQPsYSq&D!-ps+meT
z(Bx6p+GSE8;SB1QnN@(wlH6}+zrf-u#{QZj{NcQ}S8&_o&z|=am}bI1hq}y5y4AxJ
zqO`eFlKhbp(0dq&MOH^Jn}6K-O$zE(Pv#j=-g<ds{JdX?xv2Yeoi#q=f;f6hP4S;B
za<uJsswx2wI6HM;37&^mT!Jgp)K#->wvEnekIWJ5rW^2<nr_yIYXKr%2bY_#dRWLh
zcoMDu(t78ck>n<;6aGF&ua$w6o5d>(zVT2wC8LFUV!$3>ajuFXrfCn&u>|6MYyhMK
zJ+>Fe^+4#Lrz_yJ;pX<sh`7R!)4tXRr>Dbr;w@oFCa9aX8he?iXZOHlxU+UE!H(h?
z=%KJssDzTUBu!ru%mJy<z`&6>zs1Hs$I9ktM&i>6SMdwSEBHN@iM9C$Tx%o>lVlrZ
zZHs@z{edFk!#=XHpVX{jO0(4$)U8>$PcczhXRaO9a&$bFL=&6|=+ca5B&h{jEe@dn
zw2!zLcJFN&Q?&kleD^tK5;sYMv1N10iktjEvC!XfRXn-;{enNh`$NXi@_rfSh2W_X
zl9XERIjHUFYIQ<&H-KzXnXVBSr;8_-6`{K%42jTU!hym*uonEJ>vGQ7@m5PVJBzi3
zuKsPNIkZdJ8EG6T!TVwugVCn-D`SO|65LsUYhCvK8SD^hbUr6BJ(cfQXB~`&q}>Cj
zBZBnkVC>;1VBpJn&yI`#_S+3fGH9uk{X-SubO>A2Mcca6py~X&b(GOk^Cjfu2YFkB
znsgaj2*8i=q?r1>%6e-@vTZY#Xc{6`wqErDo*VOS2GfnmIuom|9*jPIhVMl>ZTW4#
zI`3aYy4brO;g**XPFyS%QVE#F^zuyhJU@Ta$>X;Q%9ug2AcrTn;t0WY>Si*(O&|Y^
zM4~T|46vz+rXX=d6DT7o$E+@3(^bf2SSg{YMXty1AUNujA?=he^?@l9$9ML=9b_;)
zyksa|qpDa#aSrrS;Fj&<ZfOCAM8_|1P=klrSnM)9)e+~uFli{*lIok6huv-nwdm+N
zM_LEje`?}PeY$P-`?$~b;CH*ToxbC}Ne_IM?{M0>=GpoX5{A9s8&T&-5wjUSRYUqF
z;S#gCbaP?+oqX4U8GwJ>Nj&@liLGJh_4o`r{w3#iWhkC60$43VO0hE`mV8)#l?H~>
z?8!<RZ#jA#4b`z?^&f>Nl)gP?mo+Qb&^H53EU$Z=f%}dD(_6yt&1_@VpQaj@dkg~3
zsp=0`q-2jN2dv$y7GjoZrnNHv)dOdySU`$lD+Jm=QH8K$!S5T`*%-v2K=1vJYpst=
z6C&@oXKKbY!~01uCq4xZ@FJ)Pp7iL6?P9ogl%8#pLtJESQ0;tF;DL(NKJQ4<r|0P-
z;){3(azV)(!)k^J<ci`dC-$d6_?32<ovX0q$uljg6iZnQB_f1x^4}(A5#_%>Uo8IU
z6cuOe2%sC_6I?yIeex(af2nsH>aEaCM1@bG{q1FAFQ>~Zwef3?bYAz}|5O{t=2Kv`
z7@l;As-}Fkev>QlL^Nj`$t49qV?Za)bu`RF7W8sKJiZ6hxem;7*`E%);|O@!K<M^)
zumgGjoDsj?hZVMeDV($s*d;dPxxL3^&F5mvv7+L=e-SF(%NQQUUVa0v5cVrJxjx(r
zK+Dl8t2KKN1p~87l%rf(2!koQxaR!o{AD>{Lvu)<*x$49FStxD=mDk`$R9hMfzumL
z#7;L(qgCo-jr=pc>lI)9EH-Nm2w2mw{;=|<>&0p(R;5$jNeR<%>on8?^GRlnv3#t!
zFm~J5&aZcW-HffjAiUohefTb^$|5EAp%4wsHKH3h#V%K*WM)Lq-QDN5Hd*?$ajLZ3
z%IT__;tLv}&=^TvVIi8z78K&v;vP#Y7w8EjajShWwdB@NsvwAQO;?>>2UW%iBm71I
z-_gTbENA@q$LuB_eO9i^H*MwFAB3JW`p`9lI{T^J!s2!d2Uia0(L8|*{@ZwPlr?R;
z+Xn&AS~c_yxBq8MGyC5$&Fq<;Zow8bSg#pmxgJxQIa?lqmE-ZV7?udiCeo>IsP9I~
zOJRw+&~oH67?+gD;E3SCS5<eEzQA_ww`M81h5SBJbpVs*HsT@rUOkJcMwI3Ax0x^!
z4H>&#Z$Yss{TC9}IN>k-5)yQCQCCYI&a2;<Q?gaK;Kd_|*(?hiFTPjE;rrCo8qOXE
zRXAsp@-fqP10FNO4`8Bm>#Gu6k5BADDB&F2qaHIXJyTZMYJTez%$uViu8Lo@gOFCU
zUtgD0aeK44<<>CYV@AoFIo%W=&uatwv*r`ESZj_u;#L7^T-Wa}4}XPnpYHpGv<ko{
zwpNzMx;n{FKB`Rq#W>WkXLFrT7EK|M7_qDKNP3A;w`t~Dz$&qF5-pe-&xPGl!b5vv
z;!|P5JgIEu91o{vx<QjOrsJRy)#8#%L^fQ;{)*C+B^C9O{A}>R#@?8!-OG!hB(Z#E
z^*2PRQO$=l(?HDq)?KlZE*5#sC+(>fuw>so^*K_yt;HyAcC{HNb`D{LJZc-33&BBv
z{Hbqu-12->N=#ZP^4K;-cn#Q6WLJ*bx_Mx>o2T&~Ja3f!h2qfRaiP^w6&*uK!dM3%
zhYo`V<D-U&4xX*t2s`#SH=A42I>{<T@vCBr;=;rEWQH$4o|{UZpWt^T?<8XXnbI_a
zU3b@CYTHG&TzWRy(jysYbX}{Fp9d~53nlos>M!q`xLF=?!BIfA3e=v82*)2SEuK03
zzF$IaTw|>R+kc^uTgcfM>{RRn4lmC2<^*rgwTv3wfyGylyK@?~Tc)O;3jAdxbii#b
zU8-?P404EMCa}ubw=8t10|rKTAV_L<sOc&(_Io;Rl#hO_%z1+Ci_%mx(D$$scOj#x
z4LGzOlfq#CS4BU_azCCQf?v#E-N%Y3;^cd1&~(K`eo~g!q&~0@GVz9IXhM6jxHwH7
z9$hN`_ys!Kc3pmhvJ+pg8nuP`oXJr}nD2`p@n>BZPj8%J$nTlf4j`8gH#oKLPt-;t
zZAj$f!<qtWAf?<m0=TuRDrkj&GUTCWM)9BKMXS0Em7`8Kr<!WuRvJAQZEY-H>JwD*
z*5Fq$`{n&T*x*cX3gmUNMo_f)s|n7PJ2LO_2L|syO3FlS;)h$x(pzGu>8<Ow%|vRo
zY=E__!;hwWl<;pD`rEdUIRup_r2&3Kj#Z{e!g%PzB%}%~4D8AZQ(EoP$d;p@rA=*h
zqdH2j(l}f`=L<2%DV3+bv|eUZHk7#v=S;xJ7hL)TygqXn8X93smX*}iFdDE-Pmutm
zHmze>RGwL8Jg)otqEsUhiN)Zfw8?@iwg-Rxlgz0;!1_qGYmWkK;d1f(aWKMRqN0UQ
zj+1usgqcGA8o%jqLj$h*Ouxj`ke|}BWoNTljGv^p{*ia&I6ck29U1nP7lZ!3N7&OD
z*BYg~nCmcf!k~rRe&{gF>HM2t#&fD|*JTYPf^1z=1r94WS6_k`HoSF)tn-h8F(i-R
z_;u&#^uPS1n0bmbIQv}9=bx)Mq~51`lx9O+BF1c-OHsha$4Jjghmj5Sr5mux2aX8g
z$hR^aRh8jVbA_yPo=LuB;zj(8$zy9v1Wkbi0>&S1AO4nH@*Tqrf@!%BTYyf!J1ite
z;?tF<%kBkMM_x3@F=KCgPC82n0^#&+4EQY><@gim80b+Y<{fL=5+tTE?i-@jgc>X8
zl;fF4>xnHpENE>>BN!A$Aiv4;Mc|R&Iy%EHber`=34>-s>so{1iVXZnZ2!o&f3a}P
z?o-%{OV>+F5B{)31&@}JWT6y~&~wG%YMU#H8Btca@qhX~9m_OL8+y}Oq{OVJ3+N<~
zS&$0cLTeUB8=pLET3(`7NmeemK&Nt(W$PgCyv)(`=2wAJDB<q^I1I;!Xg{6KKZeB4
zm9000v~*bRGsTrq{cbKae-Tn7N8Iqdrr!1&C|6K%F?Ih-p$-x8VI4s=xhtPLzd8{w
zyye<7^v?v9-JqyDw;b@p%uY1b?+&OQgYe~ARBO`22%tM%!Zj^2u;bCuA1m~YJVLh#
zUp70~)Uz8yN|C3bs(y@)?)&T_78IQ#%?d<WsxtpG*&f{C*yYTRzTp*d_ImEs^6D|E
zk=9f^x5=da)fFDfxg?b!dMVvYi4|cm^aQO?QT%&R)bI9>4}ariuiBTJrKqgxlsnsW
zALpgiwNF*5Pm?0Al!14oA8F}Y2&OUFDypMI@yR8M^slS&0Yi(E^0!<wwm(yG{mk+f
zEAkH!EL<s)nPI_^Sj=1vU+Ay~cKl$QS*qx&R;Jt6^cLKH($kaEb!y5yaXUFHWEKy~
zc$`GkAV9mp<FADkYi8TNc=<Bnn*GcHPrQWNTb699G#dOpOu+yXE{aq}c)V;KBvG%F
ztbr81RKHU4nXv{hayv~-kHh-e(KEr(9a2NW7!tv-#Y6C-2#g(718;x$4X2Si7yv4!
zv><P;hIpaY7h-@xIsL3hV$CTtKwAGaQ;ty-ic9x>MyIuLX(Dr3YL(}B9e%B%Pmndm
z;}0{3I3nKVdFaZIXDCj15Wz|!19u=h4gx+pRsAA?Z9e6k*{7w38o#%l*1tKMZo>$-
ztImr^Geg}g!am_MveN`!e@P56P3tZgXe+8xo+Ff51Jr6<geN)&$v5{xCOv9tqT56H
zjY&dmxIW1b>{KTFj_ico9YAs!zA3mC%y7;U%h#rVZx43DfWD|W6YbwPHFs%x`kk=!
zVJ@j3@yHz~P>jem8D+Z=zjWO55#Y3IhR;V-6063}B{x7B3Oo8{193IlWRiX~K-iPF
z@;Q@iE+q7jHMVK_m94Rc=Cp1KK*QAo2PKrHxh_o%;|jMalS#*=XMWTnXmm-;+Y|F$
zTWZ7HZknabGRPrUy<BW;bV)_^d8XKaYnO^b0wPgKYYqxi@AOedGO_F3{O+H)c>3eg
zwP0K(Q<@qA*~Gwrmjnf*Eh-zm+rG$Ab;xQ7;FF&)vN_eMkD35?cIzbSP@QU%iDHkM
zdjym7@JIj|<qU*Pv{NOmTu$4%T$V%_6uZ&zAB@GuxDQZ3b`l`K>rWnVEgzVY&wPQ<
zuNJX}*S6B==M0A@CB#~r-TjqtnDIV5u|HtD^74Zcy|gtV50mY_aL0*euW~I&(2A^C
zO-~?J38KH787vf1h_2LqE&(1MlW+YnTRnh&bq2TU>`Fou_j<D}$$;lUFi4mvf|oPP
z-Y%UZeKOngxHAT|*Y<|BJ|Rg_5VN3GfruUs{{{J$Rzq9e`3sAkPQ{9@)b)-Hv(-2J
zUfr+ALj<B<?Wo`jkb%Mkw_^x4vAl!B5x1#j-B8-ILC+P{e^kmdnL|atP+bWiyTc%z
zoqElH<0a7aSrhh>y_K`F8Bdq?$0P9cNAa$9f`J;lGyBT4kET``mGH*j1(NY$Johl#
z-bSqib1i=w9zjb04UIo3rXnXh*-f~UH++>UbdzDuo~DgS^jDw{v(b@ReZ6m5tftVk
zs<CNa6P#HQ%RTymJL3T(KlP*hn+nr)jro#2*Bu6>e7QFTtEbVn8WpqE=3oE8TG9_-
z_)S_d2Q^ivt9HXcathvxmzF;2^6O}N;Mh$3*^p4cr)918z1nzCXO{ESO01(@2RGSp
zbdlTU#NWlD!707${O=JtP}&FX67q+(e-qVAt*0kjYZbTGBt1>`t!lfsZAIqVjbk-k
zwxR!X?!>-X+s*AiEoOj}i@(JQb(#!l2qgRBBbLmy`a@%yitJMLqNv6W;zjwZoV)2H
zae0c#(8-ZE;dRtmQ`?d6$0(!rMiuBd{raedC3nwV5t?k&Wo$OL>y_+eTrxE<cDPNy
z58xq1@NY2<Y*P(<H2ToLIV!!*TmB!pU4@CMB2TXB=2F<=n8AB1o|Kkb8nxU<m%Dkl
z;7JyKg4ujwf5#XZ!y={&m!w)q*_MbOu>Elb4(!`{zlUJrwBjcg1zG$wLBaiWjhUHG
zH4{XJUB)86FqsrT3n=LbQpG<c)?CWoE+FYEk%pd<V-S0fGHX<F(|kf9kV)S##^-Z+
zBpPY_mV*rju*D;2hyiQ(rOw-ZU*qQcwft@`T8Yx7E&Y`&-~M##XL2?`|A8|+7uowp
z^mumSW=as@V%(-mg#*EH_LxmtB@9?jbv9FLESJjXhsZTpDGfw(JOyDtA8Dth(xv8z
zVl+y;y*@{Ye9Wew`3m@ayuVzI(sO7+tT&G0!&WBSn$S7H`+b6wIH3mg_J$mKn9<0S
zb?YqS?DQ<i_5ALC3vx>1Dd~t#Vc4{&zF%6$Fy1zetg#G3*xO{%av2^_W47CJ;j5mu
z-xre?)uOE?<GxqD6zj|s7B{KdTF{4eS{$5%T$3hFm}2E6Wi<#eeVpdtmV>@%khLn3
zHZRU{Z&j+fmg6hch}@fmDZV^#$2M?}Vy5{tQE&YcN;IJ8u$3$`{rzGX%Rw*z95-*>
zt!NeF2TGZEM1JMD>3NCn*<RlA8?5bli%P*i<RV){k%?}is`s3{y%>#W@WdPhxyIP2
z9EpWr#&~4d!5e&L;Xng!O2a$-AeU}dCkI9Z@eZvNXvl2N)!5dXZfV6=Zi1M^B$Suf
zA=Fc+c3;rm<STA+u8L;vT4gROEM{n>qY|l)qgU%$2QljqzP<rzTZeE$Uu3b>`}OGu
z&*_@q8#mt_TrvAM)7t}G+(N`e(Zm?lGt*0~yL;;0>SX7}ty{3j2b{bd6aDRCrxzr=
zRZc)4fFfXch7RiP+O6Z!IpojNNl$$Xu)<ou7v}yp^~uGo=j^ZbjM$_Mv&^WwKl<$a
zrPL=y>SR|*RdTsMDj7%Kz4N0{UUIBZ)l<dHc9sUQv61Rc95mrtGrPRgi^3x1DR6!*
z`BADkNrh7@>0YIgwOcv7U@_eIL$OpbO|z&_cupff$y|rtugKA+stM5-ig|*cKURh<
zn0cg{S}eshwB;?PY-41~&4CP-ERB?N_2Neu9^R<-^wo>CbcC(E?zldA1#VRveW;Wl
zFCs6jPluC`zww*aKjRI5mF&vLG8qjs)szg6(S3HcCbJ`H+w)(gVfqa6I%m6diQQZC
z=#=xvh+OqO7uSPxjXtD=;#S{bbP=b3J~j8gQ+5>0$cjvfx0VD~y4vYsSVS9(xVIfQ
z>*yc7PSYQWJmbqmr^Fel+RQlAk%lYyKoZtmt<GtJy%A-pp@zMZ&I8M~-TV8W-9~_H
z>44kGt@;{57kIM8=(Dlp#gm8yZO+b=@|q-QRJw2ilo6vtF?=BrpQm80aAJ{;_8k*L
zJ{CswQjT_<A(Lrdlo-3g>OpA7hNNgSS!Kylf`}~ndh<SOw)-0O4z_P+(7wb4bauJb
z0Su`I_z5$!q*53=`dD@>30SjmL{UHG&cBB|>C)peqv=4)D>B*qmJur{6jv&sERGCn
zS};rg#_wT`H&;;{^TrbgH>`U8*D#6_w}4lPek=*~o10-iO+`vzF3oB$eu(%R9NM!P
zsSeLasU@urs-|)%lWFOcAeWS`ljVgX??_MwaUtd0)VFZ5QN2D$D%!qTVIrTS>OC<X
z<u&hp{Udkp%{R{LvW<j4*t?rQCl+&0PrM?t2!PLAztq{gjkZuhlpfYZ^V??VDQ!*Q
zKBhDGpe-L@aRrluNSsSfklcBQJBeJ5RDv{n$BnOAFdvjGhZ>fxb+W4BXCc-u(zElz
zvvs}r>?L?5ByXTRM(rkDY52<|{QW068tC70%v#X{2fNQtZ)33689L>Yv<Ud{HyHAh
zBsH2Rx$otYvWtlx($0w(3@lI;-dj{rwy~?*2p!~<Tsl#tzVBLr*U0)2|2koYZzlkT
z*B0Tlb=c%-Kg<i3WON0P=gPv`E%A_q|M8aMjN;%cla6L;!#%f|Et4K`iPbF7?^SL$
zenS%ycZUFYvUzWO3dd>wpv_|I7vG1Ks4eKdsT&NX%k;!d@NtxeMwNhA`eh2Xr{ZDn
zP!y|2nS{j8BBl)7PfQ{QR9(7<0r3%acot)+)ER7VZ>|(Wxl>-;%k(1JIq;%cecw5o
zG*Ix}8kF`#UjA!|7iaCNS*pf!7z$B>FVsUXF9Ry**qu7-X=!UXpXf@XFW{=<yP_>=
z)WT^ITU-#_3-y3DZr|VZ3hCTbmXHkmmjt#P_2^^@yok*R>dn6mHcL?aF%9rXOt~dA
z&@@8;3}fg=uX9mqEk-$^)FN7+4yXgEaLeP8QaqTe%9}#wt#50B&3P3cKmx*77q3EK
zcbd^Btk3>Cpw?3Q-D{SHQ*`}!*8hBDewqx=ToO9LEn8f~8M$Rf0X3A|^OrVO%Kpv0
zpFpDqx#0s+o_tGEWV`EA>O@?*(ERRqGiM(bRF7@`nUh|V&kX_@NeQ#?p(`xAqkY*`
zz0gXO%ObovmT|9;8iMUy2EPcHN1Sy)RV4X1)t~4@nr^~~o|^L*q1N{W7Zz7l$rjLr
z2CF$SQE**NOk6Pkfv1JVgz2eBs(gBbpm7cc=xdAfnLRghZB$-1zZZ8QJiyo)ao?W9
zKdQG965OglY3qrP6KJ<s+Hqi3&+t3IWBuB(CHL(twn={D^YFZ`gy++2aaRt6Lby>s
zdLZuc28=^qi0;C|$(Riyc^<^&dxTpR-=zH~%Nb%SXb38(0C`}AFuGK;lRpY8q||O`
zcG42}n+;IywjfFv!=GzUSDz9ZQ4y|u8`N~Eqcb-Vek;%&_!?){PJ0|B=h(Tewxyi;
z*BBATjHsQY7GhBD7I5wi{8PD}UTl3fez|qG(9kkfBv?<~Qd4=3NY<ng<G~D-I+lr-
z*3L;$6+~yYeEoc%#$|B}oke4+Vd^qOa|U1YoclI2KRolRoO!O3t&Tghr!0R;RK%qP
z-shMF3V(fbPptPY@y&S9nh?fOA;o^vY>Euy$AXP-eeGsPGks%Km_@m_(`bm85^i`=
zp@w#2hITko5+m2@yGnH7V9=tfi7xg7IC4Mp$?leeUiT_iV?l<lXMLednWyE!5rui#
zx|($Ok=Bk-Aenh(R!nIoI4UWpx}j<j6w5islplDRq3_BWS;^BYtdtQz^!PNcHwA&1
z?M-;q_Y=BPZm&tGuzh6cW|f*T|MOR3hRRcr`ZaGWb;Y(FlEXD`^PQ6;jzl(Co<<~!
z<yT78ud^jecLQ@~B#-)bhB#qZRg{T&F-d1Oc4o-oOi7yNC79w>)satEf9}yzs)uD}
z&)#CN!uAu>2)Y3Fd9D6IpeCHr*6(}IxZiD;mBVzWgLm&xknIb7Y@#NT23cn{2{gEh
z2Gc6(t};M0oP_w{YzjI4wvQ|~g?-Ok5+zZYA*s^8WpPQ=anng7FJ?MxUa&1_K@P$t
z0(AT>&^=G8*lxnep&jTAI6@I66m+aEHhrmpF*EX>VLmH@`&XxJJe_S>F@4^<r3aJT
zPDn){dYKgygG8NW77EkAp9HUpyJVj)rD8^=f$OSK{DpFHsWo3>-wUsa&7EiIY3S%h
zdul({i@N0}T#GX*hf{b`GE@r6z#)!GWSM(@qNSxVSm%1lx~PHr3j>V+^Kh6+&z@4?
zKg#!ls=pil!x%8%p|^5qvvR5>m=EE;N9e^ec`vJd)A`8cu#x@OMF}g74>1-pLrNYx
zM6GL?BG~Ac_1ANiRiHP|uW<D5y_X#^6zN%ts8t)-7YD)*a<_#22D^NxnTA_D+p`rk
zn%PPP%~sqZ5boFJ%cV;pU>^A}lIa40tD9%MEm~pFhv?K~oY>@hvGq4SWGJw0mK2@@
z+2n}9La=Eex;MJN^JDiOZC+RLrF2aVL~-%D`en<U$}q=O(fU=x(o@}~FtwR>TSzj7
z+1j+#Eb2DsSQ41=g~qY`gT3=i@B3xe9aVzwm1wZFC#mf#Z*%U}0^Sw%r{-&X{#5@=
z8&V}Uno3POU^T$V7lRX1zO4*u{AYSuq|?;a(v9P1vt5pU7#IeaFF|TwBjHWQ^|NO5
zas#bVExWFYsIcAN54CY3zI}vOUe7<~+4(`&3CD`8cr)NL%yDNWA-6B3p;>259w@F2
z9U9>bU=Qda!7I!yUevQ_Bhb2IUKSk6<zk=O{Ps#r=V^r)4jQN$8j<zOep&rEJ09R@
z3#zF5{X;k}QJXO62%3k7CuT1K{atLlOPGiNWYex9**N-%NRD}w#d9&D+2BZ-;v{qr
z<SLvuGFUK|HH{eOzXfSW-QRkh+=BFRprX%H+KuASGiQwN_=RC5%(0Ec9r>Vj3e8Fn
z@#kCZ^uOk6v4&-7`ATu!jvBq~gNaHR18x_$KE=)Ko;Pz-+SuSY;udn`k7~3RS54TE
zTOB+)52VJeYf(N%bxcf7iV_*sdr`*qTIB`WxW<fv%zrm!lJv+XWg~6^Y#Dj&P+_-<
zl7j40tT2n46r-f_3v0I5iQ3M6lkrm+lW!OA`aBCGRDcUu%iV##g-jWDCQ@z)C1yRR
zM;*N^|F~Jl$NH~<TYir}@oPoEgQwGUX8CTKI0d$%zZj^I%NZ)uJRPnV>WV79fwA3c
ze$am%pVsm_%58d_;+h%QSu}%n;Fx*+Q~W_mqW7G~R#hc4Rbt2jx>MzBn*Cxt#U)UB
zh`F<Xl&Ubya6$&p>8)B;hFz>VmQ;?8ofIbC3ibh6uAHN$&)P<c29;4h%aHmuiRmH>
zTV#g^93c+wCD9JVVf2?o+|2Dze|wCZ5(xI<)yUR$Ev?9C^da++r`w>+P8;Z8(S-F;
za#<_GO!O*6e?2@GxR|=+Wu{f<n9^`f2E(Jm5=Nltqays|o#`ZrH+yoH^6x_rt7L)D
zo#&dxhT}y(g7*N!4^G|E*|K&JG7gq*V?AFz+HZ7)+#{x53Y)|0UuBj3?Rl6TGUXiI
z44o+9E95btfNLXF9L0D>YON)%tO0<*EF=%dBU=kcP$fBv`ni6gAgE*PZjo7B<Qr3$
ze;ls5P-?_M9G_!HE=K64flIEYZAQ!@CCx`az&6Zc6y)nyTorl8tHNQ8SyStYB0kKE
z#`zdkIF?Ob+gJ6?lDeX$P%FcrB<jyUwNawmi7yO;8VRC{6M$S7aWWJGm`^;Pwd5;a
zfL2K9{pWkY8vEDeX1{hkf(Y^on!MLP8CB^LmD@4jpI)ZSQRBWquOfS!jD>j^W{6~u
zlujg2F88Cp)rB}+j_Qu=I-A}!Y&Z^&yGT!k4bLZ<@XW!*pTX%^9yT<rpRb&IekxA6
zu903)OV5bw<&Bdd#c-N%0G~a~Y6}_Q43U2Wi%mr~vSgbB|FCrqhy*-lzJyb5_4q2y
zq2u5cbCKgA1XHHPRXX=8_zgM|9QhTqFpj%ZSd=_sv*!>#r>3_ivUq#$^p2lu{VM=_
z`+eg$D=;%)I89<j*Av3QwJd)6sr$uzEV7v{Y8V|;3)z}{K(zI;y`!C6WB7ShOfk$t
zp*Y)HFsDo$-FUQezAvA|BhjNe1%>UN6ODt7ZtCG|$j<XEB9EUFU;o%e*BO=~7j4?(
z>`H~;4Yd?N-h)ssR7fmGjhifTsQP%dgJvK~N@-`HecE?7Il>Ex;AiM2!;JcpBn3IZ
zq1druIC5nTIyP4xGlhkl9d@E;v<E%MCE?8WfoM9bd^?*eu@{604L*#8%D9MdrEOZU
zNr$h<rde*$iq|AXY%Cn`5SNyIvAn-3RshBS2r9n8n|_zq9eS0@p?pD7<h48kOF)kB
z91d0Knp$6p&>nwp`-a*{3&y&wOYgfAI6EDg;V^t2*3VP(^}!*3HPFHtYwi(1{9~EQ
zPD!nnS&IalxEj)14`uL*U`lJ6M~FcZMoq2?08!25h<n1{Ml(LH*FS*6^Dv|*0u?OU
zA)cq^U+oEjrx$Rn&#Z*mK*9EGdfk!Z0Z!lF^Ix~MKy4(e!<b+r_Mf$Mb|#ZcqR5-Y
z=Q`M785Fp*loWAPdahR6g3Nz=I`{58s62@QI_0`;!umV@5)4EgLT`(47I>-^Ad$@?
zjVtzsl!;@%;hLU;uhPk2uxIU`Ls#YmE$V+z%Y5=y7ZfhX%*0(d`VRIM6o$}R0c7w>
z?F!ky!_;)Y!zdz@rFJmz;kk|qf_&w5@`ZgY6@FxTdvuxc+`I36+Py~MJ$7--N3NFz
zQtk@jn6V6hmnoFdNx)Fq3V|=?&;CaSS6#j8D9B^2Knb+m?dH1wDhsX5t^97LGn_Zn
zWB%U{&9|8f4>w({PR+p)9Fc*M62d6!#Yt(L_Sg9p7|V>=fXU`)sZ$({gu;Du`^d_T
z+?U&^1QygA@J%E$34Y6z2A6U{26<E<yERX6hyh6v&u_Fv0*m#P4x_9ibH+zXDD5(8
zN-RvUUJOV!C9TO-aLYxXB;%%Bma!GzmQseS4fyALf_3=v^Ark22}4*a=ttFjqyZlW
z75hC*NgQGprtgcOd(pD+@u*}j3h$pyE&38RUaWR)tOCxhIs&31iTlkY`@h@tgblS5
zr@<2hW>TmVZm1I~WEN)U5#ADG2VxrQgQF|xlT_ZP13M8@nAvrw0=BpyR;AjnFK*3G
zq(UUqDO_r4xb#st1U}P4C*oD74x5_T8|>KWg!s}8hz49b<7tO#3sh<v6Ja9-WSH0k
znIA)$ja^b(+TR2(58i_q=L4j?-w8id(EaaO_2w%z<y=0@iH!of>R&^*bOtBybSr?q
z3^jms&aBNL*ikzgneA^EpUDliB_p_ZjkD(UN^@fejoEu=CPYGz$h|>4toRPbg&uOT
zPqfU5(hIZ7=~gfC2AkqnBS|wpm{L49u?=Yx0E2cBr5`pFb`28ZRpU;sB{E<i_*J&&
z#LYYSDXlR0fb^z>v`7p7S=3yNu=*_(TSuCzbAYXXOI+R#{rdAp{^zh14%wC~B&EA*
z<Wr{e`Oe-M^fGULEC^`U+*CH<7P^@`L#JdxPQ|8%ODWH8uNqT~>Ssirff<{lu+w2G
z&cd(qKb<Xc<d3)u+Co`O4eBK+W5@QV+G=8uLS7d369vu+Bt9(~yb1|t;=;qKqk0F=
zgUBn>p=+i5SZZRv+k<&c0uj557WC^I;ef4lzaN7TGvN5zNbJ|a7SvU3cU$l+`t*Ja
zbY9M=o_>0i>R@hZT2xCj=<*?SMF;D^Vw005Qa72>bGEk`(t^=~=qvw*LayI^#$?N?
z>);sS6^e2GiuYaL4AId#yQTSzIEh1dvXfd(g%t-0eO@SjJW>rHAv43v@2S-=Z+#^I
z?-WxBL9~ElOEfriqg>>*w850)Pna88eu!s3g>)Ig^W{(eeB4XOdev0g9ebJ-WXw;^
z+}pD|{@?EPr@UF+4SMbJDKQok#eb)XF*|s04=pF_tFvtCRcv3w8yURD#H8*HzV0_6
z*6^FfIQ?9FM5f$8Czedz)>}JE!>br_R-||1S2{Y0(2<$^m>0&k2^b&x#k5#oQKxA_
zj}b+ob6&NY1gt-tOJWCGaDDD{E{Abrfh~$}8lC)ks7>r@;@IT>Jr{FVxIzMcHOj=E
z=GdZ3^UVZ?u#RijI)?0J*r96v9G%X-+!>yDW^N}1L?xJ8u!Sw42mVUAa;RfA1yG6s
za!sb3gX~2}E3k^F<%bB?ND9@hVi$Xd*ju^=`s7;p?C&bs${dK3ZL?0BQfx-JT)YB?
zK~q$k0;~t+)Z~*x#xaSrmjvcy6~l~FNRaIHQk}MkMUGOjM<0N^BaY?Xoh^9@mJ_~h
zCF5^fce-<YZ1n{&g&C9JoJn0awj3A4l@o$|$c~Y6hGbFcv1Yfd3jqjmr;Z(V9ki^4
zmkQua#~kEv911KFp33F`wI`Rzqe@~<XA<q2w@Yl#Ob1W4$N7a77I3K^m@>6(41b>A
zw#x3Sc;IUm%OEqW>g9cfN*$WvNb3epu6H52;Ft7yV!U0UyyukU`#`+{w_bdOV?SR1
z|2nom)_y2!x0h>GUv?~MfnqkWRl7`SumL+d<vvTlK!X7Pi@*CeksUM6L~<dOg^;D=
z5X|0UZs+{OhD{+S3=1YUQYKZO^)jCE!BIFGKI5hXO&BT$z2_$ZFoCHwIP9=+A?nAv
zvR&s994RIdVF|L>5p?Ni4nyLJ`N8%$eGV`*_-SPCcq$IcW+)2L+G<f85A6-v@S@us
zj8GW803jfAvbHRTPi(x@0=yse-d#V8>ehQKory;uc&{z7)(gGq*H%P4>2VzpvcJ3x
zE(DFKVl7N?vYtJQ#)J$mfkUmS`7L@#BMM4Mu-mom0fnl!YpTRB{)n1Yl|?CIF1k9=
z+y0-}E8?;nJl?^<7!j4tMvI^Qfi;LB*d0d%FG6fI7bM7vl!lbTpT1}Omi~P=Q2TMs
zBLQv~(l43>bdI69n1q}~)b|fd+{JYs5A+KhBpA(m`Fc+J&DX+zix9&?C&w_ba<?Nc
zui!nG@z{Ago`^Pv#EV$MoBVUvxwFm@uE~@P9z>p1G9v~Zq;}FC+p6v!s`gM5IG;{w
zDx1`+;W;fP?6Zs;&_hEr!u}rssX$i0ng^Mpr_AmC1ut%IxGR<8lB{5EY`yla%(OX1
zd;^?o%?3{UfcK8BA1q<XnIBr#WY<t<pq8bTHDl3IE<p5lqtRM$;^=}WwDckS?hXl&
zwN$=-_YHH(oacnJNc!r{cg5wNART*u#qIEt*ZbFKTF{)SWd*G`Ndzf~RvzB3ET=d8
z^e=wFb$`wE{>IiPUsJVEbFpBC?WXel$%_H;xti~tU;zY46mb#ndk6(2Ayi`Ygb*q1
z*+*x_zFLa4avR`ROJ!XX%aYAA4mBCG2oXzaOju%K^d3nSlH!uHG%}!9N2iY71-j_i
z?XL)a;ltgEme_(jY%WbpW?s&G_sw?<G4jQ$SELLnDcvv-8{npD9#@akkzd~*dG*<s
zynGcnN00N4zRk;7OdJ%et-#`7^G|c3N3it0s$v&6eF9dt8^p%Kxd6}lJ3oFb<M*uP
zTq-HsoEO=^$u<<UfP`mfvRWe6@1DtEV06+-t<b513}wl@x_ZUupS|Y8a^_)PNUdq%
zO14G}g7?n6SE9#z$2^~L;t3_-gX8Ars;zY~28nkpsUl6r?!C#^y&5~owFFPJY(H2!
zU+|o#N?li$TA0;BAf~kN{da$74X+L~DN-Hk8iTf6v5yCSvi~u?1bPB>o~dOb_A3%A
zA{j?QS$MctzWVi%7eD=i;bzYluYb&WKJsw?fNpE~au%ChJ%|Bt$(y>aQ^+Dz=fPXB
zP3NKJm1#MdcTV4vi=&iw{?^R-d?J-h=$aHL<#zU)+_uz8ZlG9{VA8I<GMAY|rlyMP
z6jt-|>9WTaho{ojK&%#;)Aydx4@j68?<X{Ms0It#DTR5RnC3IzfB%+UJn*wuU$EAh
zwa)DKdxpNlxhDO`(tXzw&^55sKu(puDr8HC&Q)2oFj2XiX1+U~Ir_xv3?``RI4{`%
zfzjZ{+CZ}BhUYf5pPF9=a6E51{c(>iwehKQA*|YDq7BU11d#?38K5W45Z!AuM)<Qr
zMO1LXe$A<;h}kVTb>ytf$CbnGn%(VdJ{<3Om`<pe51!MEwfCa2G@E}})k>`e?_f9X
z$bBK^NZ-YVl^X|iEvwCU7YQ!VxsE;t`rufmBROTX*t;<qE7agQmP}bDm=#$Zi9{)x
z(|SV1)(mPseg+6TV!h(Jzvh>>Kc!5C^_cOJ2~N==p)S*@;;SM>nNDwc^L@wo`Win*
zK708E<9K5J_FEj4&Dj#_#Y#b)Qlx?`P3B&i)`k1WGe3HHz`ICoof4Hm9|o@W*BqzB
zNfW6|tYx9=JuzByxoBq16Kl>$Hl;`(A|Y4@p;XT*&g?U4Z8BQ(>(m)!6+-O^1!6KE
zET4rKlx|SE;ef=z;cg<Aj#MnDr&Q&9o|vYY@4x@Rt~>DK*FR>c3v-z`?DvdA517iL
z)<7zr<6<>PaMpyD;+Rt<XC<dh%>}`+R?mVb6;G+2>Z}f!my8b{q&9+q1vQwP@l*AD
zIj=WcluzUTZ^kD8>(lrHj%Ve7Z~>_5CJ`5`d&*{`xfK?T%_N&y@x|z<b%MRor3=l$
z*nk;?7On2+nryH(Hidn=AM}n<ln~78AvfSg#0+|!c#}%y%;sa{TC$D#H?`Lz&h9bV
zx&ZbIRL$U;<~;_6zC&Xm*NnJ8r7|ytC0FzPax`|3y;!G?C>@7z#dUYXAs%qnUPP*a
zAHmmV@M>Kt(Mp;p=F<rV^8y*Vo?J@nE^a+<y)ZOn+w--FNwf)XS~q^pg_JFBSDeTD
z_TGwJjvd9Gs2KYtmqJP_;yR*h1rkEv4<rhwQYrPxZqXPuTdTh@OD>6P4qrDvECUOD
z@b<nNW5YXVmx1PEqovS!&k!Sr-GG)3%^kxy(DhpansuUcDJ*HlcOAYKrd-HABh}-p
z5bBl)<tWnpLGUdwv&RLvzIjbf_blgA^K-H%yw#PP?C2S<;dj3N_=mKKB2T1^e|8;j
zYt^P(lU`lc@$(D6b=;*&YxB0hw_!8lEHr@NNXx=`nJ@B=He0YcX;Kj%JU%$YYfGNQ
zfQR<21gN!zCy2=#gBQFvkWFgq`nG9Ed;Lc36a%_N2u`eHT5I!Ht1QX>{R#AF?*Yf)
zt{B6fSG(7YuE!M%2ryu*f~)2gA>gHgCQ7QLb!J@#BpMqc`arJw<RQn#9I$|jZI82c
zZN*-w=4mI+{B6WpLZH@qVH{}N6Sk;(^ENXEmvc|gK`U*|m~>Rc5{O*e{Ipii(*#x7
zm(INXF8g)!F>Kxx3N01OBGN4?yX^_pUWG&04?RQIG4>r@bd*jQ`kol$h2fzJxhiu`
zoR)<+&jdQW4`emN-c${!QN?yE@7twLEa8YitwOGlO1*$q+jC}n))!X8@2=sGV;#fq
zU(2<sc?_wv%YaSb%^e7yoyjzYg#Obr*<vEOOvXlhuz(Qf@tr3x%Bn4Sm~D>OT5-;`
z6i!>uI9qekM`HzxHa4Zrqf!eSBhz6_%n$>T70va+sL{s4QnB-Gb31ZEYupGaS90bl
zUDLI5T`rsrZo=mr!hx&q1yVAS7MxmAuJ`6k<J#PfUa4i~d{Xwy3t}2^!Q;dlI>t`3
z55c*%C$zQflm=1@zP8m6o`r*Kj7o9Tsw~Swt&R0in}<;&GC0{-<n}#Fsia&ezTmwX
zt_SZdX|+^x$s`iB8c=RYZRXYKDo>1^tx-|!8#!sxdT~^5dui)g;+$~U@7RTrzKaBu
z4x#Hhe2gtr#hAA#Wv0`aItUOrw`QYj&8IAxrTyMj3rj6juNT(3s$2p_w!5tRo%Ly3
zUE}ONW6@vmppRf|p(OycbJq6g?bn}vG}f+x5!;?Gmo#5}Vvt_^X6?YVIcdP}7T6M0
zEYVUNSvKDxVe@Eo?aWn_5Q8<Y!c%H+a31G7JBP3^ol?zt%g$&(Ji&WwRy7DtT606&
zS-t^pa$z!>4{IuwQh%88tBs}Roe-ttb@vnYVMnNYU?%!Zz`_#533wq^^T6yJ)U<HC
z|3J9vab2YE26D+bY5VE+{I-2@J6E+*rQILISo+gElXGcILfY@Gl#*GOg_JU-6q-G}
zZA12dlv<gWnIT(1j*2z$^$E&BTk2ma6;{0{TGXh=R@bT-S!_FDZT+gPo{la+^OLIx
zSq0SUl*4{+VHMG|B^^TF_k<V?Xm4||l)}2K1Q+Ro;MHXOnyLjEwf8ULSaac67uI6w
zu$#wZuF?qcn+qVDZST`FZ*%;gFr+^wR`jRwKY4rtus)4H+>p!p_SEQ6qok^9NAN`&
z^Hdz$CRkgMW+0I5U)8!uYF$BBmCMPKY@_x*;A+CxOf8vQtPm8MY*0MSLYAURKLmFB
zj(a&$Q8N{E5qw9jm3dz1RU4B;e;_ZcV3H#$S{G_Hi9*#%$t#2o@quyZ@gacolqC4j
zQ<lQJ<Jq2pvqVNj%n%U>*>e?d`Q^=j#o!#h2*=}FmgR^G1?QAu@aEl9t9gb6$NBLN
zmBfo5-_ZHU_5Ox=SvXD0<r%7IE8Ohs$pfKS5hM*ka4^k@X-HHNO9t`*EoQ#F&MU<$
z4)c6kv@qv|^K`-oL0q&;h7)eCZ>i^%52w@)Rr~udY;jSef@`MMOwdSFN2r2t=`&8D
z3xVJZflXTCP)($i$htzUygD4%4?RD6eM_!Ax%3>y9bJf=^Gwc_loHFb@NhoUcO5-@
zQc1WpTcK7ReZf*0#gkpY1%+${ULOvOy90mmCtvX9-9Pdl{^M7q7ESEz7H&srV<9z-
z188VP&Qss+kAJlH(>kUVn!NJ&tz)3g>54u@#LGr@-!=q~P>@0sxf)XwUlHdBjXk5T
za-5F5e>gTEcI&j+;}qu!T_pB_*atLP5h@$7-8yqB#!Qf^bj}*HU2h47UO-D@%~ffa
zfL3JX+Kh<R3R>?O`jI4#c`8h6W-Z%wf0H!ZGYSN5!;g8fd&Mt5|0VPBj`PD?q^vky
z@ngYxOAd4b0b_Xa|6%XXo+Qb#Gf(Vy?xJdD?jBoaRuu{bG<pHamKpMa6yg6bd>m3J
z(nFyLDc#f44WJ8!s?3P+#Y|Q2at<HPRWr{7Fg-*gMG%;{SzehD;o)wsc9(O%^DSCg
zE@$K^GQXXOxXiCqq1KI?+}FR8`*-a!BSBG8SArT65-DdwOoWsWo9L$0ZI|2*(AZ>z
zA9QDMNhhTJZPn3P6_%=OO?^KkT&0b@H@^S)19Qs!N%$q%0#UtS4mKE`SA%79X@f*2
ziCFN9ugfI7(~+X^>+im0N{NT#5eXB7$>n1q@X{)609aj>ODk|WQ(`9T11ZlO)g>|v
zqxd<vQGAFAY28^2aRzfxy)@%e3pu(>Z3nV!pGvQ;Q(r28^>?l1Hi9mCh<=S6=;*#)
ze|#oKLP&neeRU>dinPI$-j~8K-Ogh(v?OfOxM*SjG#gSK!DYJnkV$hS<Uq`k6ceZU
z<OyfJUfGttSDHD<>cT$AIWZq+>P2a#fjYory35A;xb}x3K@zeiLXAAXlp(Ql#-Qzz
zs_1({k^_G>|6A_nJ06ZR%jFN8%M+POq>-``q7kF-Cn*WFU3hx@%>DTr^5Fz0+~dW<
z@SpTcdkEngERs|RYMz#=M(&Z26Pg1ik*4m;Cj??h^wxQNd?d(32$|YFQ8VP=M#<;^
z_5<CFQaf4-n%r9mWyrk*D~+wKENrw^(bnBZtQ(;-Kxv;vCeXSOx{<~ko{97jhT%W-
z?)zcRkw5wMA2Bg=CkIFw;(?G87KEj{Z&PcXbF2LR>CF4Lpfb}_Cr9CsT$a44(hX{d
zt#=md^efQl@L|=xCo+J%At}Ae2kkZZ!<V&u$MD)>G-hlMq=8x)a&fMjy5T4bmfYuR
z09YZ6kPO#*6Nayy^Z`U$_fkcK9)%)8YA_X-k{8)!elq0O5TgUXigGy4NEWgrWH%KL
zffOfa0CrE(Q<R7|E?f6AeV#l_pf$IO2r-WBW}uVi;7r7QFhvqdWv!m@8Nx2lEoksF
zJjcXczUTk&{{M{h$oc7s1f;`CQa?j;cHg|vjU+zr)^g$Zzk6bScTYZ@xI5f=JrLC;
z!$S%fBqEj^eD2P!G&+dyr0crzv{lMhhaat<za3>=J9RA-g`&``4q4?$S&os?8=qeu
zIllRh)5*{f=_#UkV@}jPZjHuT3cVWI)on&<CsfEiFewhqiL=>T1*vcxb_a(zv0IOH
z6H4>7bjX3<{P8cCWage3$rGd#G9@eqO7}$SBA$49t{0S<5CyeHHm5*e)u6bI!+9zE
z;k@xuDw`S>0!pB0psAo;{UjQ)@G#c#-st+`S<zp|KX`lvu)dC;ZAj!*M_+tn5)b>m
zu_#87VO$vnyD_gzx%~h}cEY!!=JFg<7lJ5UeXc_Hf+rnjK0KUw*)|?8n|qvKLt;JL
zrGp<7RSR)SgT3KNToUBk&2pwd;B-2;>2m<>-7jRc(PhkFfl~ZH6O7&(T^A5%K5Uo5
zS_)H~$f>>3f)9E(Lg;}M6W`wdnzz$;e0%yeZF{C(K9USV%1Bf~FfuONOCfpj5QJqt
zQ?zjUW+r6ecsydjWhvJ}Zts~fI@Ao%t*NI^8J-Rja6btzBD{cTtzFr1tvA0Qh|v1z
zW(Ek+fhCE7+A3X?>3AYFVg2+%3=>nFT(d6j)1%U<THV}sa&|`)Xt)`4tVSE%y;59q
zC<2n=h!IYxI@|3uy<<*^DLPQ8>W1<a_i|7jG;e_rqx%G%U&u0(dZ13u@QA^=Z4E<5
z5s}O<wpE4Gbl}hb)t?Y!;otvHLMfHYa^ZeHkwSdE5`69s9{^MNvH+}|QNhjcXg9_F
zTkH7tmtd^TYcFg!$F_0Vuz>Wz1Poz--QJmPTAAa-bnuRS(CS62?XScO#+Y1k6q5MI
zlpVm*0kDL01-snS&V4p=3Lf<x+<T`tr3V>6n^9}0Yp3@{qq7YOLy$->8+CIK*HiiS
zK6mtXHNqj@ag6u;(eY0>raLHtG#u0!@<|88_cYBw&dFP$*4dUbDn>ZG#Y8zAjv!EL
z84vSS2Xasie20^<KYG%T#)uJU>P)4yUjXnfaqo)uGJG!FV@gdS2veTOlN-(w+|yz$
z8>J3FA8}u;=82tksg#td_v(H|!TmO(23l~xp;%xV>8f?>;{v&}qTF}G{Rtzd`{*pZ
zU??4J)ghPQDO+krYn8QZysQgyxFed<TPNae<zPbWMi8ZoI@5&VV}~$BHw14zq1t%4
zTs$OW%HuKbK7srDD-LeN+bp*)WgWxcy_Okk*%<r3TrbD|IeuO%11R`A&SbMbB3U6$
zJL@N)LmqTUk@=XhA1Y;A{PV~6WbDH+C-Rh;4w*Ow4u=C#oGTCrOfj=3_J!b!+JccI
z+)p#dLnNgJIS_PWJ1@hFaeVGboZXsYT?s_)4kxMvE-#ISvHLt<!Ac3dJAB7|de1lc
zI|6Wi`UJglOqrk__~48#a|3yCHdf9q?<w2H)6*kUIMB5*&qqQPN?l&9;(hc1n)<p`
z^)ssnW!*NGbz?gE9_6pkS)}esSg-q|r=vm2*<VNYb)4qe&%3_TwSfw~Hd?FHUakqg
zzyJ6H>v`c<`YR@JxwS-Rh}EDJRi-V_L?IYut<JcUK-^h$(cHLMl*tlFGc+4g=n8Z!
z22>^tPdr{r<!Rjr`2cxlJ7uy24kLxOH}}faCUgl%8VyYXeI(B!nQpMPer=A0==12-
zwtKy2?}%&oVwf8q-mzoN;>PRTe8sqedf+R!VdB;ohJoy8<dF66XGk|R4BHQ)laGUd
zD79~Hqi{bVA1hWDx;FX%aCR14Oo{n0V^fFd9#3My6Q}pN89by(yEmRQp33(=%^ar8
zl$EG%JFvCpMCd(1yLVb`EL)}34qcd2WS%lxX;)uP5pO65@sN;T-v5R-`2%<1Exi?9
z`k5FuV(>J=V9^0>8~_X>NmItUuBEWNykG}&|10&!2zITlWANtoUULA00UAAJ&$>6h
zhp!*vEHu<X9tqK#rqLO5Av)t!M_^6?&TAz+FEj&oHnDZI`<e<dj0U?>nuA%Q{vJdL
z5Y_)IfMTRpVN=Rks9U33M<g-DBPj|Y8bO{B0$CzS6QR0-p=m?R*KrH8Gb7iPb5s6M
zl@Jf4=^l+R@6B8TZ<Rrbu=LJL>AwHQeO5=1><z_Q327ok_u_?NBQfg1*G$`MNBryf
z-^BO|V0|4w+ptj&^gdiP#iesy>~^soT890w-6_r<DSf@FzUp&#JT962&g>=h(7UKM
zQb?Q*nGg3TmaVds%2xX|%VL~r5QJuxro}J8QGO(M+6~-ebi>lqab}86yKc3CGLq`_
z24HO$|1Jr%H;OjWd?cmh0M=H>F_UB7AHeutBiYG>DIWOl{*QS#|AKFhf5g+)S(o3E
zvJrEF;9yk5rAJdtgy2Q<^Kzjr#<%NxrjR)vPxNjC+#p_F$*Qnd9aS4$-%-ruG!a9f
z)=F*aooUl6fE6Urd!wUHCpSTZx8e4VFCkD`C5Vy_C*r1Tbz@F}Fbz{jgxcMksAzQ$
zvAp+Hj5d-0r6_51>g(oA8W1FgYq6~6iG{y7zHwgyb>>KAn9{4Ec+!v#bTXvK)++Vf
zImA2E5+%<N0&N_&n9+k5iP_vhcrC_Zy5nE}*}vg@`2h%&TDe?|yL>=WzRuX&|L^9x
z{%iR{4q6)*)%_xC;_utvwT^%MvwHEywWc2bNEii2!1{3NbPe4AN;}mW$Kdj@EJo1A
z&6~<EXF=S1Crycv1c{KQamg)8kU&cDN^1|npLvo%&Vgy#3(g2du;Skz-0Q7tN43+`
zllO?35TalBwmQJj+t2GV2bW|V!y6v*dwzNM8xo1SElBeNH;J<+2;Bj-k&rg$JXl~#
zSuU)#aD0Cx<iKG*psHLr50LC9V=!vQ40i_2V9_{p!`!EC_}+-@Qrb?fz9u@}8CUDX
z5SgaTG)?4ra`q^&ZpAyYa_fAmLDSB%uH;l{_YU}mkcROy6zNn|LV-RQS+z93^#otX
z`z2C>5M$(Y%!nB}WEHI`od_YIG5hyzl~M~Y>q4r9N&WM}@Ii_e$m)Q7kxn&0;wX`g
zDNX+TTE?{Ua#=XchfyexM=LkE#ZJfCJHW8B;l7OV_pRmEAqx22z)xM-yXCjf-`R#g
zJ(F!PBgEp65(U&c)e6UW<aqFB+{$t_CAR@!x_mGW0Ba&mk>kljNZ1@0N#i1CgPjdv
z(M^Z%?`H?3V@Glz5!uccZ&vsZj;Ya0W7`TUp1*cKAL$bKUY_W$-qXFyF9`3Bzv6xV
z1>eMP*tQElJpCt*xpOyX{5Iwnd=c};g&2f3ynrTZ+1Q>Qc{n)Wb~qeBl&xG|H30T`
z)cpC18yoM^y55a-D_pjf`@@mAGl|64O0DW=+|B|W9)#26AaqQTG)?4r=F%FQb((hf
z4(&KlZ*F0&;sDk;z3}1Pw+^^RXRS#v8jYfzTmqd=Fr!p&$PofDC1P~1vM?niC<i8_
zMK|p3ovH=H&7v*Os0pn)82x-X6Q(0fN6IYZlv#U2x`U9Rxv9L0K#Es+V<&^SMMJmF
zT01d15Ru3Jf1A&Yrqa&Lk^y9WApjQr8hj)%j&LBuhS!rb`@@h2`s#L%Y;}M~9DNaX
zFDftAO@}$Vxs8K}HxiXj(-qPk=`IsKHmEC2JN=ae(af2Y({ZB3>HzbU-SR`bGlW#H
zN`{Wh-7}FxO1!(jCr8N9MsPr)AVbza=5}ubOj>9|_CLkMoVzoxUs)TtjCLZ06Ti6o
z*Sw#;<F37Bwa?Jc1gYK#HfKgAO`g;kM&l$jLiE(lRyHm#F9af`l@SoJf3F*WWnm~B
zUV))z9vW12cyUF!g1cz-`Pgm9_+<E6K}x|T#^&!^LpW~*l?&YjiOHXdb-Gqeln{p$
zRE*LpYbng?jH(cP*X}qFT?MA*79jN;X|2=ohII}HJQ&yrF%qP6h?&sh2v%sUwz<z`
zjP%1a0IZF3sr;}O^8AMBRK4*R!)WXpT^b5h6&6*_rP9Ql(Qn?!LMH#YY9lB2%G)hl
zZtM6K=Hk!U-}wD}_&WZ1kFNmM*YUHBHZK1T75GNe$L9HCJI(Ru!$(Jl--`d@mI&sc
z%ed%_IE!8#g)dVg1qh-n+lBJj`OUAs;lp<y`2GL=g|#%lgbcd1B=7i!9FZJ|F?h;I
z-2GL?>l3HdrWmg*rzmbNpCn?Nn*zJWoRO5}m(V`slEpkGxGfzjynA>+ERb`gX+$+M
z$xIFtAMSs_U4Fwa5C4*EGcQjcY2`u=LXu&0tQip{%1HMjBhn#s4#x=#LMazi5_b6U
zQ`YOnL57LsxRi)9FTC^Qhpjb#_I=R5!A1g#zYS?f!@3EfgY`xV!rkG>6q6_UjUxO0
z;epHWDf90a+;CJClCyC>Ue2^R@aD~s6PZy$r_o6@BI-T>y%;JZVP$wDboBzQqdb)}
zsM2et_nF=g9Ht{NW>m(-w^n-V$hZ(SYiuuPdU5GjPo0!c!`Dlx-JjQ+(}`I!mmeMp
zU5UdJVu}fE%Ay<Fu_6-a+Q)0N(Ip;pDum(3@?{ju9riZYU{Qs^+W4t0?ijWM(ZJ4j
z`>Awebjrlsyj-ix-%3&>8X+4?S>bcxSKqwj{^5?_e!5WV0<XSrOp$b$2+7xPj3X^`
zm`;ZvV=wWJa}eQly5}xO9`2m(-&=Pdrf7&%U`XbgC)~BJ&Qyvnv!g-Oz$wp&bk5B=
zJrUuM-ZG~HfBf*TxX<r7*ojs*US576$;zA}Zu1|!(`h!O2Y#6fAtDL3*8F>FC6GBB
zj&$wAe`uHg`QGPNs500yUd&dB17z+<?G8luNW!=p01Y-x&TgVysu6`Dc@BQ@2%a=K
zACDX>JTEJ?duQ7(qs^IWD0S@dRoytgil>8h5lR?fbaS7JqQ!x;0kVY<9rz&7Ez-5}
zT+X1KUMgt9)RP<|E*S|nq>Rz!&82R9e)`0$LYhv5=zDaGGo>_ME}v-@INjZ&Tk*?q
zH&Q<G-LHPl%lUhL|NRqfmsVc8!MlH3c-3_HGRNP$mUr;xwm|ti&t$b%Nx(lolSB7Q
zajTVX&Z@H5=@YDVWBT;SuO8lV$}@lY@jtS(b+Dsc`gJ^=`0&kp3L7dqz&0DQY#X4?
zFojN|ab?c-d-vuIAvj;Qb)k15#@U-Nwz{1_Ra)(|(#E_Dq@0;DV?BVXgUc}_-cD~h
zrF;JL@UOTJZ_&2W?S&Z2)gLa2OIo`c(iIDV5QaD9@Yid7rE1~S4-U+pP9Vl*eR02=
zy}4q}ND$n6(*&j|Q(C1hjhFM8!z?@;1JhtSJADoJ4hsRAGnMWS$Fa621Yy|<X`Xp_
z_nwbGJhE-oy+n7$mM9j~nU2y~nz|fny#Lh<4G;q~@sHk>R*Y3Q+VC8cz2P`~N>vrL
z%1bX$BeG_chEhmjCLp99u_n~kSj)!KawZ(_h;^kbiPVLYCO7-94qhb^Jx;_7aVwtC
zCP#t<j`N0a;c0#3&BFtS)4}%_2L2p|C)6%ex9fEOQt+!+>lmafqCzusFIWFVxK`f7
zBZ}a1S{No_`{(S7e)mLbXxFRnUyXvMPMwr(ePa1T;qAi_IsKCV_iw-FTq<ESmRhgw
zMY=aq#9T^f!lf?5@6Z{8rp?bI|E@%V(_tb9BZiLFfL3pW#Asj<?Ty+yy||ZB>)w=j
zn-9bgd9oLZf=VEe$ddVR_#@urZ}<g&!lB*q^71=cd*+xRN+R`il@O{08ZNIx)a%?3
z$uY9kh4rP7kc%n>00961Nkl<Z<bVo1Jba*)jk-KybY~XYt}k)VI!541Qs~Wt4KxJE
z32VD?iu-nk6s~t3_hom_FoJtQsksm9QVT_&uqq?q!k=~<vOfiyI!p2opC4JL%CGnZ
zLSVxg=usvb)up?=5P+p^?jhEM7#f5Nq9;T$Jr2*w?E*C*WpecjjVu!ZVe(JHtW#Ry
z@p55W&%|wEBQPgnGXLBlhIXSiC^3;w8CplXOTUFgh$p%(yqurtCLGi_9#4c2ug#R#
za=|ZT>3$vmh~q1O^>zGg<9dmOv9S(G(NA_7`O$yL?eE#^g8KG3JJZ2%9lWUnPGZi8
zs=d<I!rhyDjyaQ~Yx@V=cNFR}<bvTA)g6Yf#7HgKUr`9-b4PiWqI*^Z0x>p%G`a=X
zybo#P?*HSau-*S%6=`rt6Kg6X{m*kua+o=$BX1AiazDM}F209mEbAk^6=HBx$-u}x
zF)zaQfBZtC<QT9hwC39XFpSGRUFu3G8qeYO0Sp0>OY>|PE&9(fKqq^3NQkp0L<H&H
z!NJ_yW=_*cP6Hj6nrWU!MEXwibzhZkL-TGj{AVf}2FR&K^^%&dfw-4N-DQir5$QfJ
z#@Wx-i+6cz@y?$`x=zmS8JAKW9s~l{q_tiu*hVXbU_wvo#<%L84<H<J<QNiB9F($g
zVMv2*&|1G;O0E=VyMj7i>G6{%|4(~(89a(*^XG2#|Alotes%p3cks%1T`V)e7&9C5
z#8ZFVN7bmU5?ABy%{}=rlVaHa%B^@G9e)YFro-?_uyFftdG&2YV45c8DRY<-y){bT
z=)u_q!d_szskL_Vwa^B58-NZ$9kgr)^p2Oykz=~!l;7}f{*L?fmeev$SITxKWF;lP
zfcag?*<M#~FdKk*aD+h3i+pUPI)!+X3%;VyU#}`M>_Jev1iEl#8@ArvLu{XayGAVd
zJ|+fIJ3)jgB|>s3Gx+>ZDU-8O&u3}__f+xq?Vz?xkHoxq?$XiV*OL(T+}8c`-DtUw
zR4KWdQpaoPff|%vF>Ul#+0C)dronOx*X1MlGixJ}N~VHIrb-*Bl~A?wvRzQQb4jED
z&7VmQ@czv^RF|=5+x7YW^!)MPvY-6km$8oF?_SHk!`_zaZM?c5ygJMO_P({V)13Oy
zdPf9mcP4@c+#GyaIZYEG<*P}vjBji}Oo_wsK;1klUItr$AvwMIzU>>!Fpmzx@o+?F
zbRAuM84SQM*x!SJrG0#e`sd_u1vmBu5hRk+#QpS^chd*nr4JnC4y`L{g%})2_VZUp
z((eF748YkB4w#7hb)lFwAh@KjW@k0;4JG{fb^(_aDN&3bpjT%QuUlbVH)`)*3tYXO
z7<)!ShF?rh`Njw_*lI;WVm=%(`!vF7#wB|^&)pZ(CEP}p(R|P!114R>{d3f<x0~4r
zFp*dP#8IcYA5!fVH2V834X0E2`77p3q20z{YxVlkYa>b`V%I>IV1soLA`%l}u)qQn
zF$$-{9j#TCwQ`U7`LLfyLwa<T6Pg1#yY%i$VN>esUkoft$M1W+vA?flju4RHzq!|c
zJ6^r-XS=LYUoq9KLu=k}Q!!dCRBcS(yytY7z480%Og4L!#5dAVQg#oleLaf^y?U}-
z6xiuBB9K!g2M<~)&F7g&^aky?8}z2v2KQe7&ndF=5B)6+M46c5i8uK@@8@s0lQ*QE
zc`1L1>V*{cl)gZ;PB5h-=m3IzEqKE^gg|S>tujhPQe>V^1QXi&g3G9N09gj&XQzyF
zWN)0?XtWaKs!;Kg*v~NQ0|MT+J{gVw?rTkAa=Curpw9%2L}N4oUeyEU%*~f&LAuZn
z-U#T7?`3=R6P|!-M#b@OuPeM65XQXmZCbHbp@*^6cbBD_H)ErDLml&!)2$SiS~<6Z
zq`)SgL>v+FuxN1wXdtBIhV)}TNJvcag!IN%Dk*1DcFPix+ljk3^semuFJ%3G9sfAv
zD}eQN{A^>tnCb8~LEZnyYaQikzW0-lV8ftY6_Y*=uUoef8O5k7l?JA4OCcs>R^xy%
zi}3yTe_*xFAx}J<?s!_(s~p5VHzuZOW}ZEvPU7&+5l`3Kw#_4JlP4U-?yR3UMMOLj
zz9geWsQAXO8c^NAtQ`naF90_;t!5P{fzTt_5|d>9^xdEG-TfbNpMJ%}iDQ4mR-U*l
ze?yW&&Tvir!fC^=F%Xwj1tbd^;e2_45}5OY`%sPI+$B=pSr3E3<g6^E35fVHx5IV&
z<y5jD8lZaZi9#69B}Atb&oMI*vEb&c8idD}3u=M4zj%j)z#%14Oe8Xwrd-O#8kNnA
zbGe{oXT0pd23`zR=^#ql8r{TWvSVUOnVeig`B)=W@r$b%VGhhBCT34f)J_=EzS0_|
z#wpKqE3{TX7EBtdk*)`}9?9vz>2y!4jkY!>fjJm&PKg{EfBB#Pg75$E{g{uh7)e*8
z<+d8!5jc*2e>niw4mNqw@ABPlfB!m`0jY-lzf^`1`m4+EAhfLNZU~zdtSNC5vI?i@
zpB@ojK7D4<PQU@$s(SOBh+!CR7x#0RM#7OWDlsMJL?W?WE;KWa^F0s8Bk%8zl(KTx
zCp0vgG|aC`+%COdtTDX0Ow3sTJ#c?IK`A_LFC6oc5A%Ee=;6<KfA|$|(l;dha@ATP
zgk@YLMi>M((vpHJkX`|Kx+nHTWv#2naZj^9?;+)N4>(VA>#B&!uw~lUi=fdBp+h%b
z&I>UZ$GhPBTiB`ihNOU4V2VOSy&w+$zL%}BZp!g=M-MZ<{q#tyMv#2H(CT1hl<J`w
zPmfPnJ~EwfnM4fq<SU~!MFPZb4&4QpPv)FS;fN4fFV5<eqnl#Sr%be&$b@urpRnMP
zwQbvIwGc3BEwtVto#=`)WTO%#b15s&=VwmGBgeymw}%4>;}4%-INb&Q{9pZs>93Uk
z{1^Wpv?<$K39|zOU7I&1M)KbHsR0?V-IubC;qP0^-EdauD+uEH`8Wt1b|$0gI48z>
z9%u5R&Sb4LmkF&#%5aE*MJu6H`jjZZAx2IqaoL=Yt*TcB&hzDk+6!;rza@uEsUB?N
z?BJaZ*2e`FqGaBkzConX`k70wtgX7lM;T1ah^U92B-B|N7=7!gR^LZdh?aSG_knNU
z{eoZTKj(e?C5Lu|-g)MCXgYH^da4#W(s^-1b2XzUA&C%(wARqtSXii`)2$In16Ye!
zV8O<vUV_kT^(19=irFfa5R55Co}Ql2zVfSIJ}~7>h<@=`>qPOdnNmU+%wl{`xvYh!
zb0y9998P!6<krIOf8-8xfu<Rg=Cw~(qAKNOBjrFm9-L9$6dl1it+`j8WM}aagpeZB
zoH>PivSfbf?)wu$a8~Crj+EGh(}nB`M$A)Ws)3Y+wVe6>lW{W02md}vBFAxM*~};i
z1h?={1V|Ha-%k8r|1bX?fA!mc&;Rq^{g$mX%J3b%8bOPD673AE8_4q|4H^F%bI1a%
zb!u_I(8POXJ22}O`$qzqkv%xRjNQr*F)YAlP}IC8QsYcyQssUybvJ?Y^TtIA#|(FK
z;Pax_HLBJry|DJxqYL9`_NF1##5^bnWtyBNxvn0t&|BazXYNlEE8UYOZ5_Y`4v?3Q
zHAv*^*XezuYIV;l0#lgz<(og@Uw-qy<3T@guzQX=(RJgp{xwiJ>^_HAG|wo*r#Irj
zYS2J6rK|FyXDW?%Zw};`InH+o#^w1|`}Dz76?CMWRzFh-G|X9YbeePrxdlX=@tM=?
z^|%BI9%$5BrKxe=3cVR`@80tGva&2MZXrS@X{MX7bfMr0k$Y<J%VkA3s8-1-Ayaao
zkxCLlLZoiaNR=F%VIC?W-Vx$^j`<BaX1;Gqv&x)YlAD4N#Zy(=U|P}zy4QuOovwwY
zyzqGWM9*)S;!G;;OF3QrBw@8c6@N<Zzni(cyW_w6H~)f<fB295$AAAjjxmzv$^AJy
zHw%L=$xHU@_(vaK0j#g%XB%=oC<nJkZykc4ZaFuX5n+$Uy|%-CQe2oowN4uvZ1*!#
za_8DH$r35OuxTZVQ*mE^l^CPPI7g5A^&)+L4Z%GGuRa*B015Ly-<cZ2Sipb2p^3j1
z1g|LShK<S_|LDH?<}fj*BX6d6yqmsZwg;lkBuf;9+7={rGCE%WO@_6<BsFzUjPc*y
zly$I=Qaho#U&LNtO7_R<1@LZaj<}DG;Bu?$d$wP8MRwZ=_GF){G5p>U4gnp^hh;4^
zg@<7r9YYw;Z<sKfOD#KY@?5o}t-mTLhabt9MQ+9)t_6PxF1rc>rrrs6R-KV@AXy?=
z^oaP~x72W&ySW!bz>B?ZeRQrWLH8800bYuc#OSR$fI_1;<M+S+1CNhSBdFl^Z0$$>
z2VeRTyU!Q7Ms@$!*YW;7%k8sXfqFK=95(CJUb(2thN8f1ft(^uJke_Fl~vmSJAx<K
zT`zrs7~RJz1ovhOZhYY&^2k%)!6OU@UB!XKKqQ7t5dF!z-9KM`#6sEWPGY#1T{5OP
zampv&%x`&<-}4aPGjYIkkT#%=V%1))UdEnbL(Q=>f1Gtysnp=H*&QRt>1vX^uZ?}(
z*(-X?-tDz<NpNO^`{x|yfZfgP7dG2Fnd2fPuUH*>H@CD(X@-P5a?V#NXc&L4=HI8A
zuTRV<wNNFx7fG<|26FB68&}w8e-Fc8VIcuE|BxX8ol+oLB53k0bbQ}#>T(j*aal9y
zigiyiRc&;Mw9)mB(PfpvgoKgBNHH+y%$xgrXYq(L5Zp5_?a1PFZ}*vSyYTFP{^g9n
ze=YmCE`lGO$@{vx0$V>>_$E-Sv-Zj+zLtY32O_yTQ?VKpj140NVM>v;>h)`?Dz!Jj
z<#8z`_g@Nw6LDoK$c@b~*auOD_g%<@5p{KC(ZOvF+Wr&m3=TJR7raJD`N;ir&%60O
z_u(B!-V%o8aO(?QH)m@0-6!bwv~=tZh!MNjCrv9W8!fo!R5JBs%-0WbXC2^PL&1bd
z9MIA*q7EJsvZkBVQ~WOs7Khvtx+Gi~Q+)t~mv!TmPfSC?b!8-Bbn%8!H(!59WcS?Z
zt)Vjf2i36PEX%OJb{j0ueZ~YCZ1zMTc|#A#A(BGk5VQYO;_PPa(9Dw)Rrjyr=WJ&s
zl>%|1W;eI*;_sP1TYrr?`8w6EOljiV58v|m^gA$L%dItX$PTt$pY*G9>5CjYG_}{*
za%KNt3Hi3yX*Va?aI3O;{S5ZxTou@~v-QS`16b7pCxiRVZ9!Iqg~B>m)nSzCVbtE3
zQM=n_gt50@6QdOvb(fC6Xu{QhPkX>D5oIRGhU|t`d!rUl+hxpoUmM}N-lGU1W{&9%
zZ}MB-#t+Q)hHOWIFxGJaseaBN#8+qGm8CRnA4a4UN@LNLCXL;9F2*oCrN;Xq-l+FG
zgR+0cz3sf$$S&C)?6X}-Fzh(|XSv;dnkuDrmaWiLI7~B_Q2qTJ=Rf93l%|~~dR4g?
zP`gs;LlV00c}8yvDp%_Q<0d1Q{r@GnRA16aaup=`K6!|F%p~0Y0hbsy>(@I_z|dYO
zWnrsFG(@Ih%rE0C3F2(*+WW`=amHPa6Zdy-I6s}P?BM<1dlj{x2m1_W-xt{#m0!m{
z_wg0L`Z|8L;cOq@Sb8@~+o|SoONtn$tR<qj33nPE6fxvsj-IYyP!F9RjZe$MdAsm^
zf1+9iY24?;uTDpFJ`(05B^9bvW{euVU%r#C@9s`~`|StvG<jEP*$5$UNYh9(^QGY?
zu=^#<i{M}r+C7?9tILbJ+I0i={1(?~d#2i6P*!J^wKvccq6gmJec(6W{+i#Uf5V6H
z$FveL^@CyQoy+z;SY%FafWN0_IX%UtMjcNdid{IbE1OiF)-#eL`QaW!IGs+MFG{H^
zdx{|92T#@JkV6QFDOe@PL`2wX@FKH2+k+tNk&pWUV&lCD;}YwRIL@N@{l`a^t@G|z
z6*XZFvs2ld%RY;k%Z*GaLuydZXL6p1(@YAPm;^};!!J0>sI!wG69@^2)IQN{=KhqK
zC*l4s5bBJS$uHP#$eViO(ib-C1dM1#8khXmgusd&67lVc6b}RvQ}9H$Pai)*EtCwG
zOX2_ayZ;xPex~$Fj53Udy<<0KH@D-LTg@o_%enXtU{^+GuT?LYe7s)AfVl)GO1gf8
zn2?Y#y~(}`jm=hm*q(VQ3!nNET^cC~Z&D^Mci3^pjuTH|<zkh?JmIGQ%K$utoHK{R
zfiPu4GNMHyk(3j?cdEJ<PY8i3&RldKIe30v31Up?h*(3LIk2qPYr9`~o%8R-pLr~G
zRN<Ho1evkBdwzNPBmVT^Uvg@H%v9f_R#6JIKhyO>iXcJ9MZwum%AoNN78>G7V5M(7
z_A?umi*3xu6W@G$pf@9hKv#d}F8B3rsuz$DO-Kw+JG=<*)%f`NGuVkgdjAVwJG$qX
zjpS7VG3Y2%2RNk0_n)45dRh7ISO1bc&qRrIItg?C3>sBke&A)h^0GaRgw${RRod{&
zi-8bbs#N;kns<g4$xD5+K<9vpkyGP12j1Ns&|$K_(P<49EiAq8R0>;F=Hcff0@5qi
zR(hKdj2e!locZp%@6c^VOQmcts5NKf-JkgOn{WC3_YcVM4coTDah`}*Abpp__%nBt
zBN*{mKl>8=M;reOYxyR%)o~efUnLJ2NW_#0k`T#w#PI7gS@~gk;q&&w=l+D!nRDV@
znmBC_Y{!V6BA>UF#TxgAKn$5h4a_+-tyj*M7w+Dim<|)C(+NS)bu@oWJq=HU%jH6A
zjo1R(1GR>c#yg|+LRmMo^#N#BmvlG5x&t@C#G5`<p(#lN$9UwI55M8NyI=F``#&RH
zenBh`RC%IV<@xDvP^siYCcN%Sh|ndu$6MJ5(%8zzS{IhO@~qGFpuGRpA3;dulxfwa
zj5d&l5sFMTPZ^!YN#96>05QR`F7)m)x0GG_L>Ok*!}Bi3M6k}PF2^g~cv@Dze?0ST
zz9UZN7i9@puXKzxbQbHZCUlubBCPvzEz3ge!eN>qm?sfNaSzZC2x8P)ed2@w3`21e
z-4eBBRA!F%iBp<+pWm=m4_;WLQ@Zl9Ev&6@Szg$-6;Mi9sq2d5d%8yQ-HeSyyV8`G
zrEp#YwFuulykicT|NOh}$t3>lH-E;<^Y2M9(pzI)R}N|Rb7%*lhxGDGIfDPz80*;C
zl>WT2J{X}scCbCfPNWi9iSmvl^DxBrCeB`?R8clMpSBAx+rlS#p-U%A<hvO7aLGK+
z6YVhZVxL)Uqjw#ErIICZ7bmp2U)#escO37I<S7$~%*P|v8dfV|auq~T2UluW5>QH_
z=L7Ht>yPwOe1B5&;2u@bI(~j{UJ(OU6?*3(PMqR9{^i^MBX9VQ_s`$bIih52enDfU
zbw#BS?3rmO2+#v+nI_&4(ONKTm@QoDg>yZ#$Oc*X<9{(T&v!_M5Eb+MJP;4~33SXB
z9IYF*F+xo4XQkSyt$0H?%1GDC5T`59C&5{bFWbtpRsQhlh2#9dHxKXWC15R~He)K(
zTG`slxt!^t(PE<w@P5-yu!dEyJbLf6t)WZxfQ1y8v+pTetNvOoqERVvqlzGwX>CTO
zkdKkOH1TbIORa@kDi$kSQ@-CWY$I?*Oqm1dcH#WksV2n3J#P>1$srLXlY=WL%FDvX
z^M$o`x&~4_ki#8^cu$Tq1h*4d*M)p=*|1=v8R2U=zHYvo`<Z<m|NO^S0PE}c*@o_&
ze0G1L*J6wPChwizV2;KI2~JHN9t!Ta!?n8ZC{-JqRyM8PJsh-}W=d(wmX!_Vyj2!m
zsp?&iXn}0PY%t}(d_OTAJ+ixL^<p4+H_w_O9b$uwc;_;NabMSw-c?ccv@>Ti`{j6N
zA^-%jVOXgk4ML~sLJ;8>_rK!9=^NgsUohLrDfQh!wiX&1inCZ`Pk!1vMP9fB#Mx8L
zpf?DOXI)sVaA_OaCd?8c1g6|slNY>KAD!X-&@h5@h&JRI!=)F!cGr4q?!y&4W*NIh
z2YdHO$EAF9f=p>9-T~^$<+8ADjoy?XjTkoSD70%Z8s3qusyuHC_Y!zB9l_l1Ahi8<
zl`&lM(!0@`(L866Y7r%&x&n!5I$@TeC0gjH7?*NpsT)gcY^_t9x<qJra?qjIl@=4N
z6%q-F(0gUwR+4#AOw;OsVw?~X+HEjy+`p^u<_3cq;!XCxqka%Z63`bgB-~`D&hYz*
z2gjAEIE1HiV{C+x__p_p_g|<QwRehEs#Wg}4!;XCVe5^oiDb&lvT^C#z)!<q!VKUs
zc)C_+@!$akf)JyF-WCT;qvlMDVT`U8eXZMwpI5)Iz3yzvF0&P%_d63YW{<9qV$5;k
z!|^>2`2#aYvZOtj95Z8U3#G4alDtDk;(KEsAQRRJ+MzYpR#@xGSvNK+ED0qHlVl<1
znWjFE_Ptd1TiKX$eMl4oo~9HKdaJZnhX<QtQMvMKc5jjpkYIxu=>XNbRW8d$sqQml
z#DSkXNLJ1FVB?Bwv(mU&;khi#dEyXf-w&JnX?eHar4H&5hR_scaepjn*Uq~Pc8vke
zb{`LAsSD>)SbKGek^tsX=-vt{LYs<%sfUDE^nGLKP}){dEmRUO&z=A9AO1bR``z#S
z&swLzHL3I!>O6jZ3l?$n{D0Kt_OlLO%R76HAFt&fF`ZxWujJM-=Xz1_d0?UxZLEFu
z01JxCg^7sx++DN|uNvXJEL>WlX!E_dyS!$S$RQoMJG>>y@KSLymdgNffU3JY4>2}O
zgI^ZaSj)<koYm#b%H-@M-I-T_bz*=iXHb{wMvWvf-X7mD$patauX(U<IIbtGyW!oa
zL@3%YX~ZG>^}56Nw`$c*j`!N27cRAMX)CL_FV4Caa;h%>lZhbHU^}#70Noq}mJwz#
zk~GJHA;d&XF2xJOhx2tm3jq!O_lN44+Lcl&Ww}sFp|<XRUF4An*6C6_6-+8kn=_FG
zG{BQyFcXe)@}O8eWs>gc!zz1|PiU5WEtf)*Cxl2fgDb-%jMo-un#Nh#S!!V^8%yuB
zu5Ng3>ZxSf=&jIurHjzSnN)}4?2IvsoN{Ikk<*-+r+5tp5d`f*uXw{T?gl@5YT30i
z9d(Qi(BTUhVF#YvOQ@^jExsEoWj~XH-Wp0WYRG*L-|Z-rMs1D77FN5kh6R%$3mgiH
zmEnihgr!s#Ex=&F2M3quIdM9ixIesQ4vA<UZewPYtq=@yyuJ2@Z$g0x7AV%Kct{UJ
z`g@~b4f_R&#0WW!n%H0_qG96Oy9bW)p11NHM|(rGX@mwj<6gDXX+R%L?9pBnZ@sqO
zuF0CJO6iSDZ!Eo0Lq~eJvXCP)fixQ2JEL*<LXA7go<-$gsUp^VF4u;bho+>oH+SMy
z&fQhK4(|bNux*VT)xV!4nk4suWTcc1B||f#4$|5#ximAL+8F}eg%g$(%WlgORIujz
zHRdwTN<x|tLg&mgTUaU995@$9L}H&PO`W0L14Use3u~z?wNv{D;AqX$-`9mm<QVTD
z3R9d3c%X|UAs+*Wo=8nbJ#ScxZs8E1HDlY9)8xsb*NtyKrN;U-F+_BvZ+;#B+{ae{
z>+AU0hBrF}lXxvW6&n|wSC@j}-?AtBB_dhUxcs^%-8*Iujx23uX@05IS;xIXXuzd5
z@~lKret21US<lqtfd*tEvjy%{IUN$m4@ZtCk9)pchX2PjW7@E-5H?5!P6Y1a=+p!O
zEA&>`)`F=6Zk|ePB#C}`>Uf7G0JWjDK-#d-*y?9;e9!;>7ymtP(mUR#53n3PQIQpu
z#yVZ7@jSZ9erVfZyXgMd@Q#!=rRhS`7k<A!^U?|=;n?pnoryV;r^Z?$5*g;XUOXD+
z7v7+lRHRMLFhl)4NkWLwY9VDv(Y=Vg8#}<O5U#k4p;WD`U0It^x>39F`SWMC(y67A
zVkb|PrEfenr3Ag2Jg>d+`}1d_B)&Vnqx3@40yYyQVtGO_sw!3$4V{gJicxezM5*n}
zfr)v3OIM>6KiDdjkIQGy+sflw*h*vDn#)n8W9kWXZN;O(muJ#Jkok^vJM;AXnTPbw
zqtWY%M0oS&1FRdLD_5VfYX0~0=n%yrPZRKBD!9K9U&<L9v)BF2hSbERmb&w}Ukza}
zutId7GD%m#0z2JaJi%%$g-r_^8zK8Tnq@-7c<B`^Vix)MbmnEfkRRSSvr!F1NnM#z
zVm=*tIbUeCa;!6mkfxa+{*JWXsjZ>4Bdt^GN>31v_h@a@y3kt1dcVpbf{k)nA}Mf{
z@QOh`3aUb+F^41Xj{lY;69=2fxWRuHkC}eno~e3v8IA}@q>r`IaWF=E18W2-=WXGo
zJoBh4+bHE~H$H5QlmdtO<QM6>4C#gWMQ;!?$9<hyl}U}5B2faZ7Fvx+=$NRp-NHnY
z$)$TXSf}IiSi7=x<>j*R^mOLUWh2iCCHsXmh)b1aqe`QT7wd}gM5XtI!{rm-+`ZxM
zcyJTi>Q@I5VV*J_VRNn~jmE+YO)4Ul<guoRWfC*hTuM<@s0N=dA9>kUp3CZ{*1?b=
z9c<efYlW13@0_P2sL-^tyFZm>rI!n}HlANL{^BqHg3`~P8mo0Am_7_C!Vb)Gl^~1x
z9vTUJQ65;)O}%xbLG}S;slXmRzRNEhyh;SgF6SJ8m5LF<1Q-@JF5AYZ^^s?LVGS!W
zPX63EK<Iqdl{=g1UHJ6$%=6^}(@ZtkO5+s#e0_KLz_;)Ily&=zmICW&vNQ-GamWX{
z7%eCz3qpl*#(L-R^2pucNQ@_7CWeOT6WaT&Q$KuQQl6nTv~`jg2h03Xe9O_k<yZFK
zbF34mZK5RTal>J<=ofaWOayws>k*taxbzjZP8MjrvTQ5oe&MWV3I;Lo{BkD9#M_5E
zQVgVcPw9qgp=qV+29oJPI3F6h8HY^}NkdjMGar)s#pTG@2HhBagfnOrT8K0PRXdm3
zS(cUM=`-uY2c~27<YQ*;;(<b;^F+6eE*DC7i99({1%6jPbFs=PKX5=e;Tu;l_fHeS
zy?QLN64|to+L@N0u%MugNjzQDOsHy<9$A}E)fqUCmq*U)!lgB~t^n^q5WgI{puy;&
zV|_(yqO}>J6Uj`|%)8S)Q_8F{@orAc0#6S!564X3maCVT7V$mf=)nT|nn`K9SCQ9{
zVI*IE5hLZ{xvHM}S{p^xl?LMCeHBHBq<$hvCdb_<d-q=w#16Qazp<8;&+8|i(kGU@
z5V<1~IN`oWPr+GIp~CZX<)v)M2lp(C$(W|h-NVd>ci-^K_rIaq1#6wQR#c74(}j6T
z9PUp;9<J0O^B0>CODxtn&~Z#9ky0KkV+GP+EhTRt<)Ii*<soN|cHox}|C+gf;KSw5
zT)9Cc1P5n(Z?x8EtVHPpL>avmX(*jiH~Nq`7VSLOjkRxVRzW>Yu#J?)%pHlv!IloF
zpyrKd8*>VE(8upQ9dx5@8)e<xt11N!heV7CiSGZpX{U6dv;aM#RXATNIT~1^k=bM>
z%Q(yWsIyUu8rAR+o8S~fwa)L>kCagPc6tY)W1$m*5Gz>SvmhHyB8wae8cD5kIs7%6
zI-QkClq`{~5vfP&)_Exli$=CCe7=0ovTa;e2eE=e)s0dW4Kq@eIiBd*$Z>WrJqd(^
za5rTp0(EWtP+wfMy(N0{;H9m&U*Y|skdilH#)7^!TD+?i*mLf`j(`5+D}eQN{A^=u
zH~o53qx6x`0ld1=6A&^kAB^I%9XhZ4D&p{C_KVflIwFB7P4wX1_ijog5@jZaNQ#i6
z_-2<9q7doM?uv=i-5u}Wzo+ZwDM8H>F0`NBLu*gi8-=QORy_%%ic@j-q!sV5C=$#)
zmUnRJ>J1Qx$rHAud5TYtiTmRl4t7u0(>2AgqjalW)+g3>MzTRvPyQLg_A=N%(mXRo
zbTBvC%7)TMe7ARPy~A@j&cw^jL&o;_ayQM{-w}2%k3H3~t0#47Z)gr9xX2GyHvx7g
zh2Ic@CyUh1Wi4DvV@fk2z*1LA^)xe;#@SX18=XoYhRzVE7$0?E?icRLg*<gqWOxM`
zwKY1>P+BRh)hH>@qZe5*Fj#t>%U00hk?Lz}Y_;-yxo|EUTlWjx>w-0iUq-|_+BTjZ
ze?VJjo{org=EFpY2}lHU2Ey81LUnbR5TTR;+R2C0z<2CACu~pr3i`#!7m90RN5Y%~
zD!U(nkDX~2#u^?UIxhaZ+~Jy(qwZy9MvjSNIIu|>jDlfH8v;Ru7>t~RDMwG%>M+Yh
z#@ACyk;8GO_L=A*Uv%p%>x!ArTN`xqIG7BruhcHIesuXl9jQhGfHgWHIh#(p1HLW^
zb^z``LQGl9LY9eRe#4<B<`z9s3q!U0MYg=qZ6Rh)J!_*|t7e?r>S=4{P!?8Mdt+0N
z?{>D?FgX?@<;d}P;8M?5c91hveI8$vLRAJ(>g(J@9Q^6t8MfdTxi~!DZqeJGLe#pk
zmCkuBm_~9sxk0-v?rma*#z5_rbGuMe_hNhy7z!QH!uREcJWsqc_W&c<HN<1<z794E
zh!)nOEODiU%9J2k=eZO#1WsL9TW9H=+A5_so-P;8+s3MWnBMR0k-=I~p)6-2k=8bH
zoH(8)o|ldDrTF9x8Iy>_%wf7iOQ&C+a8`jczM9_7t1uRUm}f333>M`VvXB{T836I*
zsjgSImi_nt>uY&<y|Axi!6b2rGa}wd!ot;yB@CcIf-)sxo&sbylRlVVh^!k^p16Ow
zM`WXR_n84t4b4M(SGSc?3T;SLrW6TLsCA=+KuMY2o3D=@U=M=CK%O%iB0U9?B&HZR
ziU&@RjvV@lsU1np4d1bjs$cf{U|*RunzrE=WK`{RH7<3*P$W+@ZLGcc`9dN^MbeP^
zY%8r5M2x%pBlTs)$^acgq)WWM6CUs&o`y<5ty216+YCwN$Xvlph(T!Gs7={QV=a}J
z^FnWtyW@AD6YEl_b)##gC#6|oX$#dl7V?z=zB4ygI?^|OcmBvX(}~mJ4*WAH>KDas
zoo29EgVtc%sE3XwWeVQ9{JaznSy4H#HDhhY+BUXUd0bw&Y@7R9Io-~gQD*dBpf1!h
zBM9l35sB2&(CT&K=g&`sR;gWidR+M*{vZDXfBpNv7(RKOHk4c;1ULU5`?!huS&xjy
z>Z_F2f8t}Ux0LKUYU^8^#H}$bjNl=;HD*w5XEMG<`vB@g$Q;v5jL>z|;_7>X(&$nk
z1th!KKQa@<6%D83kq__R^Zxxi-hX(<^Vz6vBgupX<9N7Zo-(Jyk@LkFfi45gX!M|l
z)*7`G)^&keUA?gX(7`>{QudG_i$YXL;nmL*z1iytpCcJ`Jdx5iN?VE1y|bF`6i)|$
zx+<%UFc^tcEo^=Bbls3KE50Vy3v)hjNFZbm^BX#qZLQburQ*yX3GOY`l@Zz!Ij<{E
zmly6%M<$EZE_AI(UD+y>W|Z34O5xM_iM3Rw<i=*_wXhavQzr@Dm<^SxojM?1m57<M
z4MlvPx>zCW#;0=TV8T&$=2pL|PSiThjFsX+p=xY>qos;anUnFf8C?oug_mAgdgELg
zYim3%XO>c^9jYlIMF)xZz>8kNyrw7=OpMTlW!ZRH3v)X%5lMAJL|C65Y3nj<5j^D?
z8*KWW9AQJ{6K<9ro;>?ytmCiazioU4u)dC;ZERJCCfql)+VvHs_Ety>G74ocnq}9b
zTo2hjt;SQVJeA0e{%R<r*m2*f7$!GZ3?rc|1|%hy14<cG)4=g~<n7zHJUxA+wc^_F
z)+xo9^Td>ogIOe_7VwzouF&c*MI5Gw2F^(G=rtdEo^SBNDfUDtJGI+8%BT5(L%k!`
z8LSURl&2Lf+Y6;%2-zU$ka;)&u=b5^4b$dHBmtIQd9E8JDVF3)TbAGe);woIjOZwK
zZ4@~Bdth>V)`pBa*1f%&b~i<~gvI^yBHDu|6Efrs;$7L+I_p|EUkX!tLzcu+R%%mX
zN;GRMx>2msrH=GIKqAfHk(Glk+}8{5VkCxeHNv;nJnSUk2`#Ns=Sa_i)8rhrE-o`&
zN=0ixTVQLAtu|g(mmZY?pja3lFg9dO<2v74;pOQgV#591TM%QO^YD!b1Whgx>xHH|
zIzTR;tF=P!a5xRd-t9ouT{gHA5Wj#?n(jR=ztndf5=maw3A;V)@Su`NF!2l49yv(3
znwGol!2zrwxA1{|;<}HP7n(T<Q}?twFB!9^PMMHmVm@R_n+eri0;kGSR(45Q48ue=
z_}^)|QCpy^US+5{*7D$pA_M%DIG{;!TV)Ypju9~>X(cmph;Nupm;fdE*HXXqF69N2
z!f6_^!2uj}Q<k#Otuqafvukq$T^x)G`@A28Xi5%X#TWPY!WdFMxq?vsmiQXprN)N%
z<*>UyijJbzhT*e~M9J=!L9}L+(pfG8z=}s|^{#Vk?zg2OQmHI`p@rbGG4qaZ!RLq8
zGxz<(W{o7nK5y({y>~iLZ6urZ%92-VY@8&LP^1=!k(bsft+Ta8sfAK2&&$H46t>-k
zeC*LK=TJmDbz6v%=)Ewd#Nm*6v<sK@Ofra)KwKg`pYAAZ*kzas?*)Dz@*)QPY9v9}
z*sZUoI$y|eX@~FN4$$>~z*-*X#5j{}T$HCc5rc3U;N-5x+GitT<OFjLSjq^2gBhs;
z6>`elo$hGuj8r8=KTLCWCTEP!zAUvk3qq9`oefxPp@v8)<<?vRpLD@rU!F4!PsdGz
z(L4{f3hBVy4@}#Eq)wd}t}!lSv*rK?Zp`is7g$=Oc25fhV@{E(olUC)d6J2-5u?#+
zaqz7bViNA|4!o?sU&vVRJq)AkaX_ft&g-t7Faid8AF$&9u=eAmjs2|D#<o??=N0<Q
z>F_ORVqL1&J6h<XG&a`0QpamWhM(1_(>6NV3xBvgGK=tw;~o2iuYJg}P#XAo-V|-!
z8)=jBkRyijv@R4iNR2ueBbTyy>g0N1Ddp;OM%ar)XQx>cYF&tfk%I+l>GT#b6`meH
z(YF<jCrT^)yMOmzxYQ?yv&|D30j@zFFebaFT<@0KUHU)nV*44#TDmuQMyN^YZfO5^
zujAdCWA}9AmAB&nv0$7zkwcuxN!Y56fDx{1K_o(KNa{p$8L3GiCpaFCynXYAw{IVK
z`}Toa7ce7<2jU!NXWkyB1Lw<1t(}lFA;<{(5VSRKOs)&5^&fi~2}l%TO4pJi4(4x&
z<L5~cv51Vb-t1=bUHeAqo1g0gz}g$5ORd!2siBW8E>U`=Xc@p+a%F}Y+qSUgg?<pQ
z%wfJ~tryJNb$a=8fgU@(GinS6qR-pLyaduAA!gL>79`f8_CW1{G8(Rrmlta9<m_P8
zW%Ki|^@^s5h|#1`rF(550Z)Bj{Y{O&lP%UrT6ioM2$@sJ9)f^f*_(ZYDXbQtb|{Ur
zF4U<IgHXHhv_X%BAcgZ4z}i?!;c2<B)`s=uHB|J#6ikp|k$_fAqkH`cL=CLV!sU74
zkPl>uq_QDutS`^>btMt@xE%*l&3Jw2j+IQnrGqM7zkMD5{Kr=S>+AU0#(62D0V?$F
zvaef-Fb*c$HOz6uvhMNHH@QX-8!G%;no5rD;b0-t%sAJD5EF-VWC|0BOpr$;tfX-X
zv{6xq=xmO8o@uSJEDN<d6G#v&8zDvJ!%PkkLLBTYm%0@zS0G4HkGYTDF$@ZOO2%%?
zYEi%<sm{a_C^cb{33kMKqE+|q8GUM+Dlh9Nx-Ae~*WRnrM&j1FKcm*jL5K+55=A=|
z=;PoF!KHoO3LQ@t+9L(EyQ%&z6|u3~2M-UX2c-`NT$*NLo>{6Qlah(_*6F1Zs~^BY
z0<CtQUM@}{-a60E8;>t5fAYtFM4k@($L}Ah%?SBKYsS;IQPELo?1sSA<$=IsITQ2`
zXt?KHJdETfpBswD1-IU)rL&1MCCXL_6z2Sy7!u!qEY1|ugmh0ge5sAKcQ$jdLPv6n
zYQK84xnvWXtz6d6{Nek+93RN}9>Fh1BbSPeLZgvJ#jhP|uQ0&RAS0dXC)55*itRox
z1;)?Pn@bk1Ov(X7i@E1j8Xxg09NfKhoQdHD<DdbmgEAqC%LcTw)izRu?l`20Lz;-;
z2?-m45pRvJ(-b0y!_3{?J+)mxI=vTKt-L&~<P<p`W+X&sb4(McP-|n8!n&+j>kbA1
zA-Tt$%Fd8;e+KtQ;p&Csz?c!jFwoWP^LHcw^}*v(lx7=7p=rLd8aFLeZCu8sH%^nY
z(yX}*Rnh>%_4>U-=!B@GVYq5~UCyt@t3)bxR;vGggYgj%9a4(65><&IVJ1ACJEc^f
z*Nx}%%IC*3Z|=Y0-P;E~o?lqX>U)$JXWh7HWyPo_bPd<n5`yn1n^rz;&-~S6;uJGe
zjHGmQpR?e1?`B4;on9M@Ie_)DRKyx{I5Wk}A3m<M=6hfq#vYq>ni$pkPO`b;@p0VC
zy1H5WU;p+$@o@hZzWF8HDrt@c^9zooz}>@<fAepD!(aXN|HSV<|K(r^!n&;(%H82O
z+Ui4^uw!-DPZhNP<cF_iKSNtLs@iRT`d_=2O%Qe73TA|H2g;I&(J#VHjoLafB~H_~
z9H+z~C!#zftj=n*fLJ5N#PK*2W2E;^&Iys3ZF`~jPT4wiLz)p%VxDFsMO2m4s|PgH
z%0?u~im4k*=P9G+!5>|WMuU_{IZ<kJZ=N<7Il<3CC8I&npu@`$YFpXbf<)i8w(5Z<
zBF54-sycHhj0TSL^D0I}NbY%K&K|67qxK7NnmL^gOwZXHC*4M}qC*a0ydR1}a4@l%
z`$laIrsfV(+T4`61R)E2dR$r8%G1ljvTl5Oyzu69<lFDQ<I~F{mzQVXe}k~lShTTp
z9E6r(27Z;`C>o$rdD@iUt0LRV>G*~zA8=a(!=$;d6@Ia<+eXPQqh8LHpp8%AnHVxT
zJ=2?m<$KCv1vCa)sn=$Y`b`y@ORw8{rV-KJNg<G@L_JrwWx*^!m`BpE``jrz=%?67
zH_m!9)t12_mhs(QVTNDCxRjBG7oazz4OTR8TgSNXn5QU@ke<P&f&p5g1Bkw@<vAob
z%r_6CY28yC?+))AaGEmD^@%DgNhVYxL^w!<);y`XZJh279FEzWA8Le<sipD5`f-3V
zGa>sGw1<Z2U@o^t*<89C#e-VZl%hgxLNu4bTQEZRM9|t`QzNB?z($wg%pxLY6*J|$
zJhRj%hO&;OHu`8Z?Ouv%etylf8-;7Fh-GJFjv0GCKNCoNcvDF^^7icqmfErBXC&Iy
zdusI6uc`+~L<zi{3n3WqKfK2fK0kd1i-^r^HBhSX^mt}jHa@=;V$A&Nn_sgv<MDE4
zT{l{(XmoV(rChjZN5YYvxwv=I_kM#(=d26Aefk5>l6jHLe7qy(87sctn!Wh_gM?H=
zHe<^hH8%R%xnQ)*CnlRY>V-`g*1piyn{umlS}-g-YrJ*$WpgaLA_{H$Ot3rd@7|NM
z@8zaXtd}$O^gxVo5<^<yhrjt93(w3cVO6m1U|{d=1NhU$y7LbM^2My<uj9XMd<C$+
zj-PFm=7zmd7!=y;zJC4xy`0$1pLe6!n-2$0g}s7T{@GCk3*t0d1k_RY6f%)auzd4p
z;<}x~xctgsWOUUbAJpsj1P2&mOvLP$NUx{7EJ8b`4yf#`jVl8yj_zR?oi5C~*JhqJ
z5*4lE3T_!Rpk2oW*d;)UQrk*rBjv#m>Moya+9-YVZWs}QUoeIUY43Gk4c1lL=p5;_
zSlK^bk6lN8HJjeSoRB~QwW+7K5UE{hwGnD1C`9ohWi6GhR<_zHO=-<YDKkyeV1O7Y
z3LRrpH+OXX=j*Ph7h;<hp38-|H8E%JzKGaJXAzn?(ARpSc8?G4Z6i==Z9{16Off8y
zz`I?;guLxw3J$1@lKJ}H`y6bvzH(Wf2#3rxA8{BUx);n2e(m#O|C+iB1;-4w{RwmX
zIj@YjfBK`eeg)&4c5biMu@8S_um6GT&wl;nef{0mGZKgc95MkhB_O1DAda;xR~TkM
zG^0M**J#e!*J=nZM^&Yi4VXB}VMu}9BvpnKxDPp+yaIA%7^B<pltOU7H~~f!IpzlR
zK4*5Oz~1iLnNWk7X?}T9I%<CVHPzACtyHbU#$mdGBG~YiV^12~_r`H0yS|Q}KHFFC
z#{gH<ZgbkNqkgf&+!w6XhEyXbFL>9jT%|`_ZEUp@Q|5TOL!Y0hwGBp>QM6G^X$ali
z6U-mjt>dlI*?Qr5J7a04$IKKmGNgnigN>$*){7^Q;xa3=QCUf$v98Vt)o#o=5eGox
zVyEk>y~<jHzYMx|v{%mOM?w(Z+&`e^iEd)fq(mU)%)4*C;nVZ)xe*wQ+8eP79FAjd
z=x9#JEAaozS;w%mDaW?^x6WiA|JAh&z%a7ElF)V|-7tK?T%sKaOzGrd3*$oV-(zoF
zK^%4f&a^dRh>#+e)q_%Mb)d$!&f)HkLz)L!M~UV>cfG48aJ5EIu8gJ-hS4~Nx-T>m
z1#4q{s5<aLAF@Md;i}x0n4ewMmwETKv#+Sy8`Y|>E0OEI;Q*Jghf?eSm1;+|0Ev{G
zT`c>v-H(ic#O}3#^9_Vu?%Rg1WE7+I&bDo6NOS_ry0WZ|Wh<<uQJaB84u>PWJW}fB
z-y;NyHHsO{h84hWlD+?+1FY2na9k=pcE#f4Qpd2%G+mmk(y3Z0y-?MgN`2dq-YB*p
z6yF^UOvTNsgABkQnmL0)<W_%(Do96LMXP&{#_Z1xI=yZFY(ir9YzWZ-hn<;W1BmHn
zyPMrGdA?PmZg~FXTzqTy=j5R~w=f3V4=Low>-g0^_2bZq+jH2bZrG*e!KL>dUz1Zj
zx|$()(@COxL`D*mn6GbRG<Y0{#v)}aXg47H++9~BM)!RRUPJGl05d-$duv$Szt=ds
zN05??Car5{eVsUa*wD6-v^aANUw`{Tu63id4WggVtt(9%tv8BQ8tRA>Zov`4eyaG>
z*`K5DNuE*|Lm=hDV2AH)VEf6_wLud0bG=rV_#y!lVOdvDuT9qC0G6-cN~y*i-21X>
zVXLd}KO=;~@DL4O=94p>>=h+*eE5Fcd8rFf31sf3i76$oXHcq@!9G@k8lgkiM%PAd
z4g&XOA!=jlLbHv^3gUN!Hb8$1W6v1tdavum8ojM(NQijAOisz?n~pcql*C}sFK22j
z$mFbTmjru#y+;GxZVtwE@yd^@?XTmX@AwK}eH}mB=;Lf`+?WE_t^F!57};gMLva)F
z;K@81NaKDeKoVgNBkkTi9j<FZ%t)3va!+iatC8fulsgRp>(_!*)m&>|t0(i+I*MH(
zNOH!N?Hj6L-9|AP5E+}N$hAP~nEL_<SeNV1I%+#io#E5X=mMKIrmlqH{ZVTn<{60@
znqa1s_L(kTm|Ang@}h-ijcOYvBf%s_PKP7gx_KI*dI2iIpex(@Op2AmYV=x$9-YzY
zZQ}wh*EgqYr(5M9lLKJpn*2{s3q8zC_le80vOS-%wIDqa!%FMQ%lX1(-FQ5oiRsAu
z_imzoUKTFrMpc+jol94qE3}l{dD{B$Wf-me;NaB7Hvang2R>guaXfw?#XQU|N3M8x
ziC8o5aChsVjiu@^IIVQ8G&PPfki;q67NC?78Jweo6K(It3dH2lmnm4G)F;jgb=#Py
z2V%%bP@q!U8SAS*KL4h}vq-L(&5!2(IA<Mz`Nv#sKi{YdJMgsMvvz+QM%t5%grn<a
zj8VAVeo5eVM@vCC=14&Y5LxNk$pmJZxRZNgb(2>t_@!hN<#|ek9zBMB&_{MDfXJSz
z2Qfybc_sw+#7Mz+n;>A_xSTIcF>y$lRz1VXMMAz#EqWo3`S)q1(wcIvm8Ex{)(gk5
z@a@s)%_y~^OxOUn4aH(dty5Y>dq-QNhDxIk+PffPq?EY7n|N6l`l^%xlBZ)N#{gDQ
z-Dq{fv=1g2!vOaN!0q5sA7E<2y|i+ikdRnApdO~MJ+IVrC&i3e;PF$THf1RV)yVO5
zPs$mBQEF!^oixuhZG2oVEIoic48U((LToqO_1Coa!pH4}qK!k$OaYf;j0=89@%q%8
zZK%3mmiu-EN;4K)ui)FkoQ1WNN>inn?_NEKr^1a3x1fk9N~P3icyXirV}8f+_y$$K
zw4W~vOZAv;wMM+cRK~Uy=*rtSG5{<a>-l!!3gOGS^a7M!!EF4)*Rt7_6^-dw%ZY{J
z3AD9fYRr;2#+mzgkF`d<R99d|4-AMzCWK=+@OFB`{oNhMqsvum8;s{62TU<B-5p4?
zN6qKptPINLYrl26bb9JzjgNTx7~L<2fTYQpVcW{GI!L>&m8}`)zH%35-X8{tg5ie1
ziEUk3>&1a+>6nGFOEenZh>;<3nbSlHM%M#XGet9ID@r9tAqAuL7iSc9Pv>;+&WK7F
zGEGG+VA?5tW0H)9iI^r*jy$gwBHW*ZW!w1p@q6gc&}Ygf)Mh-NFBpOMAHL;yyrWVm
zZKal#a6A!$@WXRqRl(%VD^R0=bGe<EHa2ZlqtZE3L07gOXvGUWw;d8fDkzN}tIM2J
z3F?60)+46S+DeFt5HlG^PoLLy$nbBCjh(3<hEa7d3u{;Y=68R=!^1m%@!hZJw$kiE
zp6>}vbT#I~#CN}b!}p*5GoSwAul#HmD76rRaGd6`T>I@{8pN#q-1rhk9oS)eg*-bD
z$5_YvoD^i#)m1AR^tz4(e}4^}4IKkV$!t=n+F7>6nNwE>q9fKC+Fioi=*$ucWF~E#
zLgxMPj{C!b>EKD9y?4&%3$-dK9?@24r7|78@uXpRyNZAqVP}Zh?FaH8s9{5*ovn9j
zZ)|O28`8g49gzC?_&;+mh4<6{X{0U=aNS^Q3u`@-f`c+^YpAuseD*Z+9E1C6&NCr2
zPD{W-f^Yz7NT?plLt$$l3Dc3g<B2>4VpK}qXniAaj|8C%zq+k8raTdI!cs$1z~)4=
z&g1h7tte$Py3TaV)U}{pd2@QhG|yP_kfTx~B8@4-rCq!sD+iE3NJpftpcigM=J+r_
zqnp4M;6)4#m5J!hB<ob!K-`c&G`CPFZK2Z8=5wTXZ^WJZg*3TBqZDV&J#Q;rAkPN}
z|8lxE?SgfrE8EK>IyIL3M3Tty?FWAG%{Lt86D%`Ym8ET1H*#>V%;#kznyYiF3eq8u
zfUhvBc5lj$re^r6AN=3S_zGZs9Y5PJyVVHS-+T3E&cGPrgmGY9eWhe)dHnb*IHSuC
zJu^jztWJg*d>k8`3Is_c$u!%Gm2s&wy%xvb6&;4iqr(F3GZ99|4(rZrk2hp$XMpTo
zS7&+1&0J6&VUy8$?cEtnwJW80*Ou`)S0nh_X@gp!58ADbLtxM9)BSW8kPt}GsmVJt
zWA{103~iO(BE2;*BZWW^cgeY$&5K=4Fn8vL?rbeQrg<rqxh4lCB@$9*ZJk~^QH0)=
ztrTi&7?AVCG@ZQ5)fLP;qD7%<@Qc2LS18Vp-)9r|2`Z+D3PCG9IH)A*2{xlzqgzF_
zjRu&yS*D3on@0f=Wq1$`vre_~^V{x!B-bMTCb9P`o@%2}z*Yx0Lw2c{HMDP-w*44=
zb>SOKoqb|+<uQy4(vKQ-|D=ch<aOHL@w%^_UIX?H`t|<1vz)Gr{JM_!d9A}wC0Hby
zxM_5EfLJUJd8_~RJ@VX4uU~hMwuT$~VrL=xYrZk47<-KVEheH|f4`d!R;J6zRPni7
z8?|}?U1}l43aVGHDxX6)&$fMI>%)(0w;H#dx-#~h6at57Vykt`4S$|dMj!)<n@YQU
zDaMHOfFlCqxxDJIo9s=10U8dklBR=V9|BzmV{&Vl3Z)ExGwp<sIphNb@!*8+z^n}r
zxb}dWGap1M-ll+^p|x8>8A>%z`ivGqyUVIEBnV-2>rJ{RIN7Uk4BJiVJ0>pSXk$(7
zOu*f`Rj(cCS6fQhAAh6w!nR#V%p{p{`BOx#QJe7e^vGqo?9;&Ky*hK_Ht}E0lq)*@
zMd4SswY=|B|B$u3?{lx#k6ZmYqd|v1o3CRbX!OR202UC5?(gQVdcGHgS8H}p(Av*c
zaS%3!c>SDxy<eBTy#g0s7arngKpl;l+Kn<y&P#1Bhi}GK8#xwix;-FIUX(Ep8W!|=
z_TEA{{9C!tkb4bDj2x!}RH5nMQc56(K#T&`Fl!z#p#wA=t-Kqh;0l5Jo?+%}5(%tZ
zB_+7KOINR*S}Sc+dP}G}3pB>S;dms*$X2~U;AZ89im<6C0s30l&;0!ml-oiYAdta;
z?fgR8cQ)0SE5X$NRD2z4*J~YO7GC)r`crdrb>DB*Z0rkw=}*{1<M*_&Z7<|0v2GVs
zD>3d;Zo!K0PjQ+&QT5eukJAQuc1FsT;s5&3|8maceTwg|SKtOrjXw+f`*#K1$J3O(
zTG^{>KZ^ps?sTl<rVUewB*X%=FnkleUUXkA-<S7|<+b0W4P>kj=ArBV-DfLE7y$9D
zmKJ3Ak?w1IZ&3Lw?dpujmt}F*RBzPb?bwH><+5H#alw+I!C7i?bPLTCweDy8_}O=M
z=Ki*a;CoSu5!1+&B5JO1*z1ZYLO|(yLAzjfxB{=Y8pM9oxV(Ljh@x~0s6|3>5UBT!
z+BUjIdY8cl6%O-6%Gtw9TB9{}`E?4Es%*W3xh&tW^S2O)+hqCie=0@~zK%Us=Ej(1
zC=?R=J*#8AVY=5eBME)@p~Y}hf9RMSeVoI)80ESZGnx{$u&oyk({a=@>ON@EYd({R
ziqg=*gbmK{HfNlM?q6+Z%<psMHG}d?vgr6a{^8>*fc16!Y@`1K!1^gJZlg1D^*-Mj
zSmM4??iG7O*mW4_#u(YQmD)B|U8&Xxl88N!S|8D~k$gHpf^~Zt6mpMQ@4Zm#%DOJs
z4)6YoK<K@*EDMJqWO4SxO?D?&;A$_ni2;cLQN=cK)5b1HRbuf&xAe+otGv|0=jUhQ
z^p0@CSw!?JYiIWa>wTlyMwjMM*&^Qk6?LYSxasW_Gg&fO5_uT-s0tn|VzjzYi&3|U
z6omV`6ID0VD&01~2!UbJ>jy$prL+nr9P^Q;@bT$_Op*6JaJYZV!~I+S=kNcLm&Y@(
z`Xyjlk(8JZGjHC$<?i&Bzy2?O;Cxxh`9xRvxEiY-iD|}SMMDF5W<V9AmeNsN2rZ3H
zP^$D{&>aXaEt7F^>MLla_c)5|jPu*mD6%`%TsF8at25!IQK)T&{mN!Td(_+L{|ea1
zJgn_PvrcV`7tFy6a^22A{mDG{g1!yV(7)^sVhsuF6~X<dKRUOI?hRrcV_aZ@$dA@>
zw~mJKF&H_9UVSdaTsqWCWo;L#767tjVh@BOs06w|NE1_-X*$?7GUi9Ov7TKD<YwpM
za))Uqr8L0l@ppPB^p4WdA%lS1Wd;$Km#VQ8FW_5i4xUJ2T|Jy&Y0A<Y&&z`36+71b
zwlyxH@p`F_!LIB}6NU8StHXXdpVP#<y9dfv_}rak6hmZ=kvR&H23jcPLKL_=oQ5Y}
z|B=LT_utraFkF_SsARe&mVz7U-rsY6tZZB1>FG?ZojJXAa5@A|cSqj7`<9oNjgOB{
zY_-wND5h*CRJ})cpn)eGz#2yVcoJj44OOdjswg42Urrbx-UudSPk!vprP00bAo7rN
zSsTDWU>mZpGC=4a2f$Fb>+E#dXCO$%%m}8au3VPt47hC}&+m~ixi`q=h5zw?{A-rW
zr%`YZ=uM&bvFE>HBkdD^hy1^oh0Kr)zFy1n->{a)`z9#Cpp=cJpZ#kPNah?6y+j(3
zVgZR0DI7>?27}(ts6{MS2P$os-UU)hWB=UwO_3N}O4PQE-WqvNRSev&<Em~~Hr}0=
zbzvzh=k@uTv{|asnqo514PKTDAziTJU;_g&CWzt^*IMYQ_{A;cG1bFhfg5uaatuT>
z@-*@8;TwK<T3NRXSppIw$2oJD5>U~;QPu@*eViA8-Q)4q+d&uiudCLX#O2g~^YMvG
zoq7N5FNpJzyEkw6{`|t@;|nn!5OH?PbU5+u!#mcc^YPOQYi$q$n>E%hT)O*ekxm$e
z-Y+)*3$I1SsqE)$^zf98hO#*mdelNlXZIR1y005?KpX@@94yK@tN@H<ux>5*^RZ!2
z-)bGrxg8wz&o>(9<s(g3dfzzYH{3tmjSIg~)`m4$oA^bS{lX6Gx?x>;INn{Gl-z#O
z>zVweTzp^kYi{q)k3i{t9s3@21?GbI`>5{Wxi?hy^CL-QTUVaz6B>*l4mO1nuo}<+
zEp!ftiI7grVWO+>(h9A6Dmc+)oUNDfTHIbEPXV!+Fpm_$QA>+BTbgdPR@|H%U&r0_
z8d|Wm%IB9e|KUIVXK&n2c^r5`HKFv*)AMISeuvG@rzU#hCs9B~E!dhj7DvE<83Tw4
z<RHu`a5qh~py=_Ar3g!D<P=CbahMW0!B)N}gpGO4^yZDHG*S>5!9*q=#Nd8<0kuSF
zMy!FQPTU<czy25hg3mvE;^pPP&}yd?NMRzy#J3;rX$pV%{KWH1=loK+dy_dFCjQIi
zg>x}7Z;_Bj6R`exq6~|_+g`r|*hZmxT@v!xi$S$h^}^ge;G*^FjeheJUBf(VMykuP
z5sDZs7%^vG;k6>_{$yQ1gV)4Spf+cfe*eR7(O!9Ty5sr0@q8|v@-GN^BFt~-weqr6
zY!>FbH|V;dTVYu%tqE@*-i$`ZFazBe!WVQFe;xnr<12vmb^MIu^=7hDJoihM+`2FO
z1p%(i81@e5mDS*5h=8se@)R?XNKO-naB#d!yuePHKxgLIQ1z~gn4#S@?CZLDx0t;{
zW!LU*t(7@vB*ncgc>76S0a;z4Hsf-+P)cQ47cXihlA4hklwuwOzg3pKlM?!?uJO%7
z>GsHm!^=(w68TZrcW+KQOCvDF#3>(G)oJAkH*l75fyI=R630W}`P>*`=U4Wazg^?d
z4Iba!&5fX2hoy9uqFgp3&VeaM-oF2aIo-3Kt9y?~Am)kbbR?xnZI!k9hoy*`P<x=`
z-8)#g`UaQ*(Tj{-)adFhXOY+*c0D}DOg*7Sa0Um?yi&aAA3ZytcXDHDG#d$N68#{z
zy}P_ufabnm>~Q9gL`lc25BAo0aHa%y{}172oaEJa8$J~~Q1WUijQ8$K8Lux@d+XOf
z<VWk6>qTEp(QtDXv(b^XS1oQau1QLBbjj*{y5krQguzfUQNOs!!MpV?Nakv!uRix$
z@57s>U(?AzXv3sgfIQMBd+$H$Q|d7Hbx%B_w8r!K;$6>LKr#|1>QZQ>QOyG>N^KNr
zSNgiCjxO^K%BUC5qvL;N{EgSYUoafR7m_4$NX#J)EL+Jjk%9*>?8RCNk$IX4%jLR{
zzTy(x_KMLB4LA#OYn{jQ#=Qx74jk_uh~FKE>4{n@qDKdfVj$(&r4O}mSy!LhF|anH
z^Z;_)pWbyX+_uQq%nca^+1z}E#=5q_qTBa6U*Q4o$r>I7uD&OT@`@)y8(nl2-xbuj
zUc&a5Ffc^_`eR>jy0NuLmxP3jZ~u+<@<OTSA&X;xXTC3}D*E@W)-P%ukMZiV`wv>n
z+?YrE*AR2_<0KIwk|c4L?#7wy*^@EAoF-!Ils2AgYxJM6?^^4gb~y@$7(E3M9p`|0
zlcBXn>>ZO^=Y3B`6vdlML0HSid0qJYd~x5eYZ7lDdaeUo+bB|QtWIr?*bH`pahu<|
zgTz<%t~ua5fWIt}<1{nHN;D<sNXWvRl7n21430~$ky3;j_GvoCK+y(B)=UZAEs-yq
zlIBQJK|&%Qj=cMjIp(*t>g;a~0g0JX3tQ<dwa`ql09#X5i-dSjw1|?qvdp<j#rM;1
zmu?F(p0!--Gru#A<j!#po2l0j!<*BY+|IB?T-qD(y{Ta|^K?|%0c&10NyepA!eC?4
z=zU||&I8i)5E#`AVswR<^9$>?zMi4(OSvbCj(zyb*w{f9voC5b%f61~^|SiPI^NGF
z>_<MYI|O|#Wf<?gL5j&OB%)htoZ>yR9^7VOynl&+k**MaH+4{Gc$9k5urqE#7-@jw
z=Y6nvZLpNuhS9#yyWTt2-3Qb5T5G)j-G&^mE7sM6B*X(Jh{3H0#+np^pynGv+8LAk
zIkTIV@4pkqUBhwdh%v`V4ikEJkFl|p!juEk6g`Y22&M&94~^MhGZryRqZwd>Noq7I
z*%GFaRx4!-EEo5Lle{81@!hX~$?~GCiz4b^wIbBsxvU#ajd}KWuCxI1#%)iyALr#V
z&ZE(^17}IY@bh}Lugmb;wb4ZqUw;6dPBj~eeu5I|^x^qBo>=f0qpCK-VeA@SbDd9W
zXpq|(GlG~r^|-KZXFh-afo7SU6K2iX?vgPHBSAeg9geK6vz6j=N_~!FfjyY()%0gf
z@UP>41LG@z^>zHb;}wm0-&L-rlCKV0FBG{JgfNOx8-NAwHxjxK&COp!5KifYMB#Y6
z=b#gzj)O`dSte@cJ_Vv=^7NU|pw)$Sy>LFCy&H3t5PA{bM+%CDNX&77MYk)AnqFUf
z>&Ciue*3%cIiJt;-jSHdcW;?hICi*PI?r3<scby=jippdXhR;TFp}-A4TJz9JAkF$
z(HV4E9UzNG(|YP*B63OxINbAGD_iemsUy0+Bi2DXhiT?;ocQ#(AryAQ_x)=d>YcHW
z=)sHet{v+}Upv#%_<Zi<G;umi{Q6J+45`B3{MBExZH47nm<|U{_YXWhExf#3cv%)M
zTO~|)s7BVB(U1vo8eTW`24KmAdvCNWXsLK&Vz)lfHV^{Z8^LxP{XSS3hMLe*8U;xi
zpA&W{i!uNN<O&gv1KqBjoD{t<+u7^^9TDo)i$F#JCBA{`xM+;J!3zNil)>c6A#rVk
z4UkJreth8mQy)KB$MOpJ6ojC|JbQf7UL5Vk;@<hYg4j-`B*ws$GKVmcV&?7X15*#g
zGE&Z@a|kn2bmoFxFPYs7sJ5|f8?_W_sTj_Nipl*iTC3=|n5D69dUrO_4lMQFK_Ve0
zT342>^8Ker)@@^17DAjz(-9d?xoZ?Q>zvERB*lNaA@h}FgE`=mw*9_m17uN#*{m7K
z0%@$X2xJn6m^tN%Xd7LOL(0r4JD8&lQDw>#cc%lNpP$D(a<0M71~P77Q8UG?5u3nT
z`I}E?s!g2k54`*47u@~g#OIHn*p`jUbM@6QO>9LuUkWdmg{Sl4^!^YyUzDPW5Z-!+
zo8lMdR|$Qvn^_N2<q)tSBi+$I7{GnEdi5MKaalzXbx#sq5sWY{Ojx^KP&R90?e06k
z*gC~oUr2D98-v{x8iA?{O&iiSWEg3yHP-VdY(Qs17!mV;IS4v}N&oh;_~HPp8@M<y
zzyJT&@(x78Ze0T%a!P2JnV6Wu#D|A($tuKcJl6neKA|!3{kfyHQA?popjmM*n%*hf
zMycENT+TTUmukIU!b|mwV`?1@xB0)zj=JuCc!B5T!sq8_e*f|FFn<?vnn@{hoR4UL
zb2~H1;y_MUtQ9Pe%jk~K?LjN}g}pbAo$n|yKn_AOVUom~!voLhLZWe)GSeY&oF=9m
z$<Z&Ft!{vlb0Vz4FZuiVzQ0>B(mQ@`)s4`E8Z+d?r3-VMIXu8OZyq>s<nzZf+gd4j
zTH^N~pSi5cxeRx|1m{J$^hnCzAQmXCVp`ntV4RBB5JioPxCCKDqjKHU7!3_^Lv>|<
z`50X?Dv3=UnCv>hy6AhlXrtjCht=FgEQIvRI$}RA9fTC|=A-mV@5-`vH+a`h3JT0v
zIMB57@zZCv_2u>yN7JJ3b*F^eADBZhd(P$;H(ucbJILt2Emx0y!<%;24b2WT?XuM2
zV&PKM(JXZE)dP59$Q<$qQW3(7VWAUaC5M1WzSdQBZ!Lu#bn*4kdmj?d;7otS6JYVQ
z-euV+tukj%-fX3yY4$YlA^jB@b-B4W<kCH0qiH8fB1&XFrXh=SASLK7rc*2_J@~u>
zPn+&|f~^g|MIC!wK*?g9QbJ<lc>D&{kG!1EoaQ6P)5IZXpYpcSTXvKB)?LBS^Qbr6
zb2mmogHb6=CRkU>T2Kvqetc%#BIo%tAKtv<&;Rt#dHS^Q^0?BLK&$ZheBrV+9$zk`
z7`T7PTx#dMfXFY&=}6QWwSwsi_9`G}ALbQ;yqX<IbZ}6Okh;&8;7#U6MJ>=Yx#z3|
zin?TY(?%3!irz}9O(+%zQ*^{g?e&g&Z5@VvfysWkfzHNx`OIJcw)6JGA94To7p#}+
zQfQk|F=FhTCgsijTRv`ujk5#M&A*OtOFov@>wquqO#V9lq2nun^>zGw<0{bKweABf
zFfWF#<<cu~%ZGR0RXZxd&6M}QhkIy9NCflVKosGWC$c8?TasBIsTVBA!-1D}p|nnG
zO0v#cH<lqG1g^5cK4dwn?#t8L=p62#8snlmVu`n{@cew{e7<l#FD@I43F#gkFJX4c
zQSB_Huxg=;x#=*Tkksg;!7$i+VFTR)xF+bRHNP~P|6F1G6Gb>q6M251T^^w|PRAoT
z3xRsQ*oF};><wmfuLAZM^Z^_SNE9kAxs=|BLk@Cj@cE)xKXZOLbLtl|i8l{-@Nnne
zBDJxs3!k4K`Sf^pnbDMayc9N@3F$4ET)T5>^;P?Dp9B70@5xRQ1i?G|*H5{!*Mx47
zu0kDA(!K8;C><sTH{pvUk4J7eGs}i?t9LUU)#}#AeAsWT<7;_U<(<NDk?Nys$!&^Y
zc-MSgH!LSkQ`-OF^}oKLlKF41WB+aQXY}glxUS++#2DUb@2Y;->>X7J1aYQ@q6C2?
zfgHV~sx4p^5hh}T&>Qa_PF(uRZ$JORrE@gax^cOjY1*%G^QH{HrX5Hc#rt?oy|$~J
zodZ~kgoK3P!4B(2ZEeuqJ3$+*Z)8aX;*bLjmXH{PK71O?S%J3m2zHjGxwLo0x$jJc
zC?0Pf@t=K8nbSP+=I+2I0S(H-{gIU2!>&7Z!<nRE^0YSF_tiaIVwVs4q-?{(OPZiH
z@ObIOS$SKbTweHGUtBV)#_8dNnXwg50{s5dnbILlGs~5E^@b28I!CY-)9nfz+4Z|}
zMK^|>`ZdXH2fBs>i39kfiCQ&KO{hB76#*HZC4v&|2A)VvgH7wQ)cqo-_<R4F5hep=
zF`pQywK0bS*1*5p{<?PY+?9KI{dE`^#_J*6j=3*lTz%KVZEo)8@Biwx+)4Z2v;BDm
zG7%(`h@{~DE?A%wFiCFNUR*-<?(jg$8-IEJJr$!xVJREW&rh_zI<vA4zqfGnQYytw
zn=uG6qH95M29&%8Rm?oSOv9Dc9rNtR>W~YyYNRj`(vbOxVlw7I-x*e&W*be59|y*b
zacSO6*#YPjjN>73cbs`RoOpA0<jup(;XaZNkwcz{(dz~Y?yq$^%;f37vYlzoC3WCD
z5%;EaXvii1foMY0MDEJd22*G(ti04tYA<N<HMOjTbzFRvL^@8C3au(3d`skx&bL5E
z_+jkp1H6*9?<QOsp5muqXEYggtQ(Qx6F0y^bGFUaz4GXNzfKDbU%_2k9mJCeoBA`=
z?v1G3Yr^0f5V6~m`&+DNFKp$7-Ww@4g3Q!<=d9fNk=9?X)0%O%o5`s6_BsC=i1Upl
z{B(i;6CJPDru^tP|MoiGlNVty>b0TFKEJd%J95-gsy@f83t}Bf!OglQ5iMaj8+NKj
zZjJo*#A1#A^0ZR>Or^0kS2y(4QzOmP-$NVEUU#;XGvZ<l^rl!h4rwOFboG+N7{He~
z@Y9sL<H_sHQVF~FEfFv@IOyr1nPJ`i>w0J?1(&c2m^1%8*f6q7-jg*l%3+R#2vrgf
zcLz$l=k1%517)oPN}jSOiHeZ(%<*{7+7#)AgznE5p@~sKqeya2l|_0_e13tG!MAT`
zK5l1z_z(YIL^GjB`Wf10e0*A|4IWPK3852GL(9ZkGZMcc#S>8cE~I<P>npRC;Y}tQ
zu0MB=#wN-fJh|Bu6ZSnXP$d&g9L(xI!McUfz;I*9MklEFM~EYMY3u<_Zz&A>IlqTg
zq%=VKO1H|^7apHKv6Rkn{y@$LPRBRY+ITL%XW1^q$7kxg!jPC7&YBmTRlS2((flLe
zq_VH$pa1v@V0|4w+t}&N;QzjLt4C3DW&Z3^l5ssY_q*!aQHj{@ZMp+k#E}-LAcZ7>
zV@?E(erN0fG!itBB=C@rAWp><?GyrAD{Sj_HR;`{<-4i-ZbGVU97I1l6K-s&QX7}c
z!m@0v>qf2B8BCgCg5x}~HJ1>swRuN!^18o|lCg>BmW}F|8yYfbz-m-=1NZ&5cQx7_
zhjq#mha5Qf7rMadcpxRgqpNS1mKegdL+uGa5cecE8Im`_4IT}W_Q9M<tlfColvXy*
zQaSdOW0?5mpZ<m^Wg77L@rh4Qk365B-Rmp8K~m!7veKA|`2n*AeH#?<HfqV&0G5NB
z?n&XrzQ-ZoI;Sp+^rEwCq^Zz`No{BUIlNGnE;JcDz8H}qMKc&4YyLGIr*x{i4C%@Z
zZSqU8Tus^ieYi@F>~%5hZ{t>+UGlm!w66uOpZQOJ{H=8iubzus-<N$YUweH+%B!l^
z=mG!4VOuP_vrQt3l0zUx@2qJM(C7y4!(ZZ*?!8MmjK=x@vG=D>k|ar%C-!3zQB^ZP
zA~G_wva0%+o}O9IV6h9r68?OA0Ej>YfB*tv2}BEgz|8C%-CbE(nHh(>o0+N{%=o|}
zY8IJUJu?e*Q%7w`m1jh_o7<r(2lL~{@1f+HUCU0jzqi=Iiq_i|d+S7NXz<nrZOvOg
z8HtnYGZAEWCv}T$|H3N40+J$A6xUF@Y0Yr!uz?5Muo!OEF$^*2VH)r41*7dKiZI2<
zVV<}@&df2OQMo(1qtB(DXenam8h}yeu0M4Juyzq-*>_bJO;(WByh^+&OEykLI6a@?
zys|6@<}~r@ZpKvEww0VWKD<1WrhCG4pkyVth=eZ)=8llG$`H{XodO<Qj(0^5GOyQ0
zZ3tY!Xdm|VB2<N<;*fivBkI}b&W0rR1SLk_KT6w^)G<`Ad*`lG*k|Zq^LjExZy_J=
z9IoKi_>Dg4$H`?kYhXO!6GE`AYq_(Z$9wO9<40?GfL-=WYq_)9Zfg5c63B4X>j{$t
zss-XWugXNE@#^6bG<E+;rA1*ag_oBTA!fW)tab3g_S;j6gVu4Lh#~me7T1v*3C$oe
z4bf%8Sr#G9W2sV$x1z)_5n{MD7X@bnS!aNl;U*!LN7QoN;Em3Jj07=~z%oY;OX6-h
zaDQBQc$iu46KQd#sf*Be2Tp;*GO>h(d7g>uigazaTU(db*__tFz|fSn8K>fjXwr;a
zGO`(WX(Ae|mlw`!<#frU<;daw$fiOqAn^;LyheGA*meV8dCRt6q|TO5zm5Yq_EshK
zb#ExzBK;)nKvAHmu-%|6gKKYU#NN6eCh?+!;4w*d!1B70u>;csP#zKwW`$ghQVS_6
zK^AXWvdY%p1KZ7{AHVO#z?+WKzI1z`%mzUA>5Y%xwhT?RaTj~6ZPMGo8rMFDu1ARj
z77-z|;Qoc7dAp!Ml!&kl2$NJRLgaBi5Os51SuGSAxmLC<({jdYWfyShEy8YyQXPmE
zXIIry5jCbb5kq3MxVfV?h$iaR(5f5`2TTgNdKZEiUbsSCWIptwg^e{lT9n;_U__Da
zrUh{{3P+SgW1a#r2~i?<$C;vcJUkp(mWk)*6NWKO$u;>9rZjOl+!0<@(CRIFy$!G7
z46sTgTO*k3Q9hqRlkn9t^ZfLIZ~yjNUL9X^zdT}>M%$G4=ZrA(#qk?VGRg_HiBb+o
zdQA-Xn7%{Ey*0T<1DzrpI;goDJK?!iLAxWcK=Zv^nz!&b5;tB8^k?V-{#~ZS)ZGhP
zAVfD4r};YW)xD2~|89UVc}EC_wMx~Ma{d74!f$?c$1*Q0%RR7Bbi?w6@bW^<c?1^z
zeA@S_{uaZS>(<Q(_88;y_){LA0j$sCA8h!3ITE@Kl6V;52`Z3!veBpr6GCD&cb$ke
zb1>r&4jvGj;JOrv`?PQ;k$9H~(YO=W%&0aKgCRlDqHepI4}|7o<<;Ot;cUuTyUqr}
z)P)R7aaTELR4vpNNpo_M*3dPybfRM9cBZT=wifQwLOQ~GPGG{ZHkMFe+Q2e~iz|vm
z4ryjd6CqxROicDG2Wfy-ORri9fy5LRMBJ^&v@-!D?n1kBf>*C6;#3KfvW0~y8;=h&
zBCxe1LFRtl0BK05h-BgrelD(!VUr)aQLVQULT>IvQ=V7k{((Px{VSrr(DcH~3C_j%
z@Bj8+6HS?G<@x!A@87<q+9T89o;6R@Hgk9U=Tz?4!fRT4M!AqWrMQNGYIjsKa|cHN
zakrL%>&f0)DVRl6yVGFv!!$MzM8vx04WS1&K<(9<<k4|#*Oo3D;9*B8KW+=^7sec5
zGaWp_Yn1)dqte~bJm7BP6~2U#TDuBN?@dqt^v77oFtpnI`vb&9thb+xFTZA66-<=7
z$%wd)s^6iwr@fC+AU4v$1F(Z+)B;;gdji@>sD&PADJ3dZAXuPSV=F5f>?Y_HNaDMS
zW?FYft09pja({TuAthp%z&b;#HjE9e4)`3>1i-3|V2O#@gKLQdeI{t;J|<=nUg{Hf
zdQXxE2VBLqLJ&_pF_<KwL1|RX5@Ie`KgCE|67w<fa2K!`sUeZYTMwe(pSe^*oY2-=
zF`{eRg|2ZWL73C*!Lk)Z8bOrJ`E5kGBgV*dcZbajXJ&K?*x|r=tH`DD_Wj>Mt&~e)
zE72>VmM9j;^^RN@WP0N*lI<A_6|C&4Rjo5scNT;6C5>XyI>sAzS4B^D56&>H-j*A+
zvB+pTjhF>ccfnF~P$UKMmP%Z!&k(eO@u3In*TDNKw(goFy^Ym+@noQDAI<mczZJOT
zOpB32v<^7R4*ZyCZ;kjg#_d|(K|XF0`e4o=5I--`fLY>FjV6`HgM(5*{X7c<g3N?q
z9H$9MjX1A_083bDHL}$RaYAB3b-{?dnq8&wcmUHtEWa;Ii($t?l`~s@CN2?)0@D!z
zth1;z7^`H?cIMF@-N+*ykm&2if^>j#LM)QkOew}UuYScLJ)`wIa)zzIlDzfkK?7C;
zVm{ZJR%$xJ<2@FNYuw0%64owutR%$+pciFcBUER8N=T%;nRFDw;Y^q|LTIGmI(C7G
z7?gSjCIU0Z<HB)%Bs4|y1?vpOXzIYW!P<dz<A5+xPu|Y+{v~oe&OF@RLpZ`b^L%;6
z>c*vlTH^Kn1C2o5V7)wYUSG4!f6kO1IjK9jo$Cb)*$X}-OqPK-`;6(J1E!r6Gl6t~
z!|xEPkyh_Iv)UPYf`k8Hm@Ll32@xz(G%=aCbx$oIrY-<(w}%URQEIS=+}*CTSj~aP
zT2H_k_|VsG#wj4zs$<2^n=~IuL8yA_u|eltnuGA+6FHNAX#^aDc5!V(y2cvPu>MK6
zy3v~P#o-7N5v>fzTO=YfvE02v(mm55dyxg9TH~Z^XERS24}{PtY?N@JriKQ?T438M
z&!?5xH4H5#Zv#vLYmKTa*^FYw*<1tbp%f++(iBO_&&FmBSXqjg2vfRe4w0!rZU%aG
z^P(78;yo(DD{b8IfX)dd(Kw=ffVN^Z5GBT$+2El~j_^d|7`#0{O@Xi^(lIcn#C?ct
zIr9OS!pxjz2!*DVmrG?%aCcCS^O3vw$kUI1U@gykm#y>`NO_pq*5)pn@fA~8c)dKp
zJacLlJuJNW^CPsGwN13Uj4jIRLm}44bdJ2wGrwO0G5nTSuf8HrGmXMUHngqK&bY2@
zMq1x1<OS=xq2Ti(suF7>%}NSJZJWo&;Ye(OCC33@??9Lly-*Nwasc;eHCk0?=f%LG
zx=vQ6+~{g`X@g#9FuedUG5gsWD`FRn%G1k#;CxvSdF$X}cu%u4r|nE@1!b(y$dndh
z5UiZKOo(*B`T=r%y6gD!_|G4o0j$sCAAWTGkz025DERKl#m+twnh6!5m=ZfPYiI^J
zKf?nVVq}_y7?p`e6QkOwD!TxrD74~jfYRTCOPF6-6wup4%*d@G)?D`pXJQ4?-s&@|
z8(zgIm^FeKbBvgIh1C?2xjs`EtfpXI=`nAciV{F7L1uTu8jKAam2KG*_@G{;BzW6L
z2u@q}Zwb>Jh?9^eVV(lTB2)6pRn~|LC)!Q2=1ii!*6t^RSMKNujnHY|ejV3VXa@7~
z4yqyz%DGaku%2E>TDaH7>2l(Hc_z$vgelQ1c9F)f5DpY3#Liey2g`7`IWK+cPG}hI
zj6*wUA0sz57Yl8MgGxhxijkc5G;~y+oKmnSo1!;7i}9TDQ7U)yXTS48XN25<H~ld`
z1+fO#!>3p1bL!{9v7f{ktc~7^10Q|#k3t2%f}K5ME|ltwCBMGof7aVyQ%Fb*Op^<P
zCc2h}D$-jv2fxS#Jk?upJAmLV5$(#Pv+Mg+HR{f&)csz}abZgCt~9_3*EX|#k4Z85
z-V-BYfnb4{5-B7?cCC!mSro-eXZ<Axvt^*5Lp0Yd$GdA^ll^-^A|XX$6jFlJgfxZD
zBy`<9?QM>&4bbOS=i+7#jZ0tykhSjUYP(KbObJPe7Lpq)=uBHrSZ(O}0<9qh)FLr1
zl<LlPY6pxXAs##t=z2L48e+bejMl+@KA7*fvhQhqnRfulEOa2+b-}E+{)G-Gx4!SY
z-k%H7MhiD{%-2(w#_QKzZszMxE|1(`RdTz}+<fzAQ{O1n*RF-I6D^0Z^$qdz(-=2v
zncM3IAAS7&4Cs$&2E_)K0$lpoKfH;Hx2F)C9XMqqIFrFE?fuc>_uKE6Ynz2`+GWu-
zlv>L;(<6Gt-stVWzUC{9(%a612-aN*WvA8HzRp`!v?+&aq0E^%#XS~?Au-3uAtqE{
zsKls(20>#2>xCn#w-$D0i!UW5qyV8qE8-s{q~Oe=RFR<Ffp(lhE==88_16;!5}D%S
zObKlrn6-T!IZL|l20cb9wUMg}InM@6GjIgCBVSfj3t1h>n5P3;m0B827fcUK=>g(B
zYx&^L$+)|*^p+=x&XVZh`FKADH?FQD&KlE3P)+W*s$S@bT*2Y)wYazBs(Hc6nIYO)
zetkCDbxoOm%y=IUk^bB!hU#!$>#W=Ta2~yg9t~(OQc2&FZugOW-~L37<d62av5w{D
z>!X;-c0jV3V&b~57VM+zV7MVriZgLCmf68DD_!Bt`Uz>?6=4#ndZv=Qu2z4I+IlBK
zXV-Ohy_uiGRSQjxs*uffMd>YJAx38}?Q6!{y1gwh5kqpOC$76F$=l$T>3~V(E?!u|
z+_e@1DhXOZbPoav;>Fxa1nmYP0S_TU^z&Gf5R))T^ftTwy8EVYCTJB}@cFsK#GH;@
z)3WXR1ZwV5J_TBCRI3E>LI^QIm}nM}7&zPuWsB4;(53*Pu?&%V6RJ*J+Qj4hibMJm
zlNGUomF8UyS`oVDtwgW{oc8JBe;UyYQSG3u`T8G^F^H{S|8RSRhTRZAL_jBJi)usE
z-(%81vs)+6Yk_1iHis;V40(>wFl`8s%QLlJ{G3356|+p$$`EVcpRqG;rMD2eG|gBP
z*FQdwKke}u!1_G?k;cB20bV8d1U8T?`X&en6~tQyn|Z6-Ak+)yIz$*DK}f<pCt`xB
zZ?k4bshQGhuM9`ty?o$BPgL{6R3gMFG0!tKXBUn=Ur<$QDM&C<awd(6AM&nkQ~iL}
zN@<0yZOqHet1oBThxfEqh&iDZT8pIR!0R`!C<*KwxX%if5o<)PSaTO9^2Sh*36_X5
z5eCpAR~zwRKP|-0^zgPR4?f*7rk2Kb$=n}~U_vV%faj83?<qt=j7-zyi2%Ja7T|hZ
znMW-EQy2HFjp-2h;?04{?nnwhl((#QVy!Ps^F6=)>T6D?C(`j-o=yi|E|Dy+Y3YIc
z`)?>FoY!}#<ldH19RS3&xf~qwcCS4d?@lx#kq{DU-Wm|Zo!L}ft4n(j?o?+t1Ze%h
zG{qAg1Y-0+ifasbouTgh60Lrm%yoU00UTVD-k(0hPlrEd$o%vi#(yLHJ-_x(aol{=
z8-h!3yYO`!JJ91A2Nkd8E2T9i+D-z*8U(L~hU9BGr9g<TWl^h9YsHD~AaO!M;OU%s
z(TnTwn0iv(JXj9mjIB}zThXXm*y^?ew@nQT#uR6!7+v?RC*`V*s>Zy`+}+=CX@$B8
zdO0B)P=&`guXy$95k6F&*B3s#yrW6;wKy(L>+Lc0_CRqxzKFLYBvNmQ5F#^C67#eW
zQ=+m#G^Tmte9mkobC?dEzUD$p6RH)niQUC8crvJ})b-*~c*HZ01g>YJkOK6;d|ncd
zk4IuXAZ_M-`GKq#XOu|btGm~nPL;RsF05PQv>D68Z#cwPs2thah4aM&<<wb{YLRHO
zGb6gNJ$nLcH>`;s4u|APfO;!O(c<q@21x1+)8n!#IOB{CR&)mBcs}}LT@2bmN6}kw
zDge`UI^Qx({>VBuJUKb7Ct6C}2X|?70Pgzr$&FuH%QCcb0x<?>8g{)C!`p=nBiF)&
z@DQgxF_EZ{=w~uW3`p?vU4gY^tQN>ZoD&j-AKv}QNl$EBBPLIJnkl(ff-OW5%9ddO
zw=rOIBG=~T7zYS3P<t!i9MUd4FCjzBR6Ub(=6qf$)i~ZilIw{zzh^lv9Hs+bfAtzm
z!t`A~<8}5KVHbkA&PyUPNu*kEfMn_ok=X&Q5CbVp#5fUBK%z4hVoJ=@fm*#~YLX2T
z^|)b993E!od1jiXE|wgr{R~4OhDe+eaV<>Z40W?e-ZqjfEX%}l5Dte(NDoN5=fm@v
zVnR&@O=#`FrEGlourVDUdHnh-mitF6%$(mE(Hb#%BC6UP6ePLpA4=1oYg9UmGSSqT
z9aBsWE^2Y<1!*wcV)Xz(3m32mcEABJXNF?*c7ESauIt(3#lX$pY6B|k?COqYd7^FC
zB>5N5WW!WBU7ndmNIAQPT8a$#Qcd=_>ysEi`KUUy`qbHeK~gW0T=ix>=?|{)NcH*J
zHA#nqZU~eARSF_XNX?y;MTjvGQbZ$|7;is3ancjDdXc5nOfD-aC&KYah>>k=SZn0E
z`dMa>OD5Ms?rpj?OiVE_9Tq|yt;Lz98800-%r+O1e>hV0#D|v)5@x1p<}d&J&p0d#
z;X7s0I|9WEF9WF2NV18Ei8wS#mgwx^Vh9r<OdRIKah{1{n6xfTu1pp<zg$=^8zmPA
zaDQ}7Otdi{XCCfeaeV!XrJX6|?7)mU_iYZa3YbJBIneF<>dYbC6GBErIm8)TCSKkb
zN^P88R&+D!Gt}jgFTeQ}5Azosrz7DUSo4ZrHk1{Kg<xKBQc0M&uBYwb4uBWmUCyAQ
z8upmBwBp*Xqi|idgNdMabP$*qu!PtG(g?}V=BA^u+}Tjq7FE~e^5a%JuqQj{I<^*5
zqguwhRdRo?!`E;;kFWD)%4PKq4#@>gfwBaWIdPzcpLH;P9)F_aGl2DZ{3DH<lF<k>
zd%)m_>aLQ&?n>2j3^h+Yxd9JGrO6%1fZzZ_YwFZvhTe+ON@KN5h=Qu@0)u<<VMZ&(
zHLqGF#9oE%Zf9PR?A3$*F$}LXsrmsiDmO!LUCc-^a(6s%%o}U1*WhT_!|zs3cI`u#
zz)%c<QAczUO{d3p0E?>=qZcG@6FZ?bBWM>}RYSF+I-#n>5J^)am(pDg`B@210$0&u
zbubTc%kQMoH4!9Y5;0j&6Sfj4HDES@OoT9#mItD|Vp?W*?%TRZtn_Pu_5<zOmS?*z
z%P%mSdIf0IP1UsPa_mYA10vmLoes7^7_N)gx{|XnMh)^-1ak!I{ds=U+HeEm{3m9;
z(X=x-#tM)>kpuQ-9rvqu(|$Qz@%p{#Rq(;SpfeLib|_+B$IvzK=*%IjKtRjJrn$FB
z2-a**T%2{}BE6<A2-{kt)!sZJ4pNzR6wSQV)cX5W(_KW^>GYn2S8ZKS#5f#g9v|*G
zl{-`#;r#yQTKF8k>CW=4t_N%bgpSseUvUuDfoOlN@kQ76skPBUXMzsXgPl?4?=i%P
zRWI~fYY2gq62z{v6MY|WLF&HV#D(A^lQTeN!ld{*7NM#a4~#^bCKidr!yTl9>yBU`
z&G#{ehSXTYeLwmclfgH0RZm_+uWFBbH{8n{Fu5sq*vBhvhn%{;-N&yrlrY9@)cu#O
ze|OJ?Z=U!1>^~ny56-oJzFP{S4DQA!J_d-@vx!@#q>n4(PdgAilHFaRd*2K8r}Vv4
zI_sl<SNoZ#>zT7%dTW~OEQhA<-i1iluB%v4Z)FU)vsr1>*1YBHX1Dj(?}aR&wMuQ7
z+A^hPwyltBMLV-Ggvj090ht4jcL(MnPO0YWyMG=B10n%u@cQTK1`6V0?B3Sj^?HV5
zUk5w2c|79hrP8FK$u;nd&T0||UTwH*3J`@jO{C^5NK=<q7$;l|e%)e-#E=Mag0Ntr
zL9CcTs|g|;j)8P{PdwZ)ErTi10bWt0e~tUJ(+h8atFabGTsDrxfg9=VIvz7;h{o+-
zW6sIP=fd@scUJtqH-40WkNGuN8hrG6*V*^;VVD;?LSXvY{v_9R@ClB6*4y<u@9kTi
zl{#X{h;)yjYfQ2e7^Uy;#@^04_DvVmS5(9uL#=SGD;LfCern<|Ssms`VHox_tX357
zfXJ;<YM~9SL-9t@$cP7egd{>OUNA1TP+P@4M_k9tbzKt=#{>8GN8YSQ$g_8ms&-Zq
z-1KhjCq#^{mFtBLAp|21O|0N7;cBh#L+YjyJ#3Y0D&ujz;QE^41nbtXdrFa*0`nBn
zRM7xlR2>e8022~CE{7RpAq58{wL5gC<%k>}2zN*3<obUyoE68$Yh5F^7u|OaLu6k*
zw=2WhpjSP#k-Loi=I3W5+-gMnypjEG?M18pN~3-M@2l2IXO|()n76(7Ir6>zfjAq_
zzne>eAWS#kyGEWr`JkW2pX~SyV0|9{h+_yXL+FB$X|NL*!LYaI1XHMn#cqq+gU&rW
z>>xzWnb082fn1IARy*53Sj$SS4XuSO6sCs>N@Y7c^T(`VTCrNlrxWTW8*~alqQp#z
zU_I${VQ4`ncdu$|<Gf~GE{dc?V&Q2!qt$pF5{JXYZ@>A<nI};Q?|z`=g5->*=B$yB
zoE20nR3pViYO}kB;Mx##NGu`Is2x;u5#^K`F-5|pgftP;L`(}3GwY@Lp06%WoAZW+
z1ZieD9C$!@`EaH-59gz~oVFJ=l{g6`KSWz?l=FsabGH{uG!5<=rU{WZY`O4n|KmF}
zuawmY^Bs40uerPbYnWcUc;mVv=YoWc$OXcNSm`Z-f&D1F0kE#;&`5Nlqfzea)~=%1
zx(A>mfE&+f9UO2gW~H~MsUNX}`{&g1cmr_#!tovGEu{RyN%`}~t2Qp%8H=&Et@%7J
z=qC$e4G?aCSKU@&ua=Jx$Du$$TNg4NOhLO|+o6S}eXXb9GO=5$E}9jEOIxX}vX(PB
zY@}lZHEQ-CuytTr%$Z7B8`hLk3wc|e#gGy;W_Q;ty_IRg-$#J96}B~VUN2m7h9ESI
z^LFMKjvS{0Z(cv}7r*_AX(_yWIq)AoeD6u9sB6Z>32V)hZUdThXQ1fLff@*!y0WS;
z#fcQV2x=0hLt@(+)JQ4Hwid(^hS9Rok~2gJvnSY7Vm=%=ot%|cYh?-(DWxtJE!36?
zQ7LspO2pJ1_+m&TW}+MrS@1R@VXKvIpI$JHs2q9y<(GVYcx06?D4KZxbOPN_+fZ9k
zy0!>?QkWpolhfM?K8!g^sc7^4RaN?zgN8dO)n{4Y3Q`L|=xU1Q&BiKTA@(k(&`*&-
zVCq2hH4q-P1L+{6w}y?`{FCJRb{(6u0c!^sM5HG?J)yIoulY}Hj6`L?ezBj)$)vMY
zbx$n!wLEJF^sZ}NRFE3F7KsqkgoH+wjnSiXY3|f}xqQ#Y3I`>{z+qkxAlJejo)os$
zXt}`J*~2xmzN}m>nagxST4S;3V1~B+Hkt-Zg>}1dzMR<d2VTw>{`UK~)TY<uGDLp$
ztJfSK7yhfazedkr63>3lS5^YEJMe~?DJ(AkEJ6y41MDM{O~Mq77$HUoQM|B&yU}jB
zQfs5S*b=!_LUIkn(kkce0wIv*8H&pi7$7NexW7YEAx@R;0=Zm)-V|YqfjQ13IS|4f
zaejr=8POR<*)EOyhcEc;pS>c72a4Reo>nP{xR`>lffRyNPhi1e4)OMW+fiiyS<^;o
zbq4}kv)<m^&tmhK$&bg<=hd;V#M+B(!I@Xua78ceG7j|D4+zO)B*a^!_AwIT+v3s{
zC>v43p8h&Ag8`Sz1&fjU<<R#C>)M*$sQD@9y*@J<QKt?rthf%;!YC%q1ZEObcJrQ^
z)b(Y#u7A^vS*7F?5(9I301V<rk%k(a^NFk*AMyt_GHo&!PfxCwwQd`)9v_J@QEQ>*
zLMX*&TM%;2<h&A->)V#HQMHlgqqCwy_Bj|Lrw>1JTF*ScJab+P&*w_cCnn$^iMz1y
z%@?nD^X8uL_$w|?4?O+5UeHl5NFx)%1WcGUVi1mTMyw$gAvPp>%Wn`kOzw{Sk}GS?
zl)TZ3VIs`S0Wl$$${Z)|rbxBQ%X;FutUOmYD2g$#JU*~YiFsK#l@sU7g-hAky5?I;
zLgkJS0&!l6^??u`$ngWjjZ1)dyywl=|B|u@+o3^DSTUMB)9g$NiKY?L4IBDzQo*#F
zSR|S?K3Wdoj_DXNdgiqaQ<flI>(#JPsldAuf*m`+RePIf#Cl%R7_s<G^en!PrOT6;
z)+?~p=N&sfeJz-L3`$?o!V%HGoS#`@<jp)Y3Q2nrF?_l}^yl$kHa-JbpT|Gy7^y<S
zbu;eL!C49Qs?#)}jKQu@Yp8i2Jfm$x`;nU>!>tY)#j1;dYhbM#xiv(U8XD6wIW<|`
zio4cwwQg?e`Y&ckHMD@ueu$0j)RQTrw&qDquC8G*C*FQ|$LVsRsvjEjl33=%VTnB6
zAE?$?=1APG3<I^`!9(Zi8rG{25J(`cyE%=z8K`vEAE8%G1%k7Da`r>1wg#;SW3I%A
zHuH|6AWZW_1JX-E)x7PiSHLBNo_O>&tm=o-G`miZSLKp}UA+ZhTN^=*yKv~Oh>hdj
zD-I7|@+=bt2%$R5V*rOk^J@TBb`7DBFvw<p-NkxE@7l66m?&4~#+YA&g)#na4A(mo
z>5hv1HN|$#$6Geuo|ATV-_`lZurB7l_rrZ;yL|(CU7yaArV!9TuX+#W*rz*2wKlmz
zTwlkDSf{Ra(e^uG>R^L^X4#)1NDq|EoY8?`(JOJq>L{RkJD)a+7HUwI!$PP5+UCqE
zl!$@_v^KO9N-nfkv8pJ(u3HMUAf2@-H_SbsU!~NYmGE+U;k=$Z(@97vF{QvV1s?D3
zXfzJfOkmnsqyWzB^3O6VzT<!3>uazchf`)s&cMpe1OL<c%w=0yb0&njQ~3R>YS*to
zNoiu47vwxNfV`z}o|g{TxISPFjhZV_8!>r{Sdc&pgN5Z)&@ES@!Q;a#QaqBTM~WpX
zv0t;nxYGV-*WT!biI3JW?7-GFIPI*9zQ^tRi@9db6$|pmh9G-8rtH^P`jS?L)|Uv-
zWH(3<e{n_?c%QF7nZN?ra5nuY;{6!B-r;P!wl`jX{wc5GKE&@@A~5x}oJdSMd%HZj
z<4@g5Qyg#*6M}UOvQ7X{!woYq3XM`L)vALWYHYOb##PZ^?!q@uOv#g+;R=knBWNuJ
z+p>#?H=(YXwl=nP>)Mld(1AqT?<uqfWvk>|Jvoe#6lb7fT3MEbyW_nJMn~gUufHIN
zd*rfptvz<vE!S$3ex01znA|&+CiDVH_V)hZV9v`a^Yn7&!|8?Eyfsdygy7HJn(sMo
zsM3fbGEXzqGa`}W@yPLb<ni$pW+(D?c4m2JP{-iKrWgWKO2jr3EV=$;GxAz674Git
zn5TJ=a5^*H&p=hz7#y0R*0nMBGj3nY*3avnK6d6t>TP8npLvU(?Zuy)EK(p?!cbJ|
z{>^cAJ3wapo`ura@s&}zC$ZzVZ)~9lDBXX1j~je^d>)tDn8Y1rMWlbO?p%AbUwo3o
zaNt;!iNNFXNFtHi(pO#Gi*>~eQE%%OM%&j|$6c4S`CgjY$ZAvyxw-z@rW?gF#grC|
z<&c=CcoRR_9;>T%HgqXeHL@Bdue7ouQ-GMLQk{!$HyoeN2sEu&>+CDj&W^sekYe7h
zpF(5~3pGrH&|Cwn(Q^feOhii#&`IZHg-A$(MeEWbLX1MIm14^I(s;RSJgq0!qm@9K
zCJ2Gr{2JC?aF4i|N{kT|m+5%*<_nM8n~~;;bg0zJMy=3-K@7}uVoD1s+>_!1kq4}u
zsl`pC!j!msb>t$-O2*9#3~N1pQspD+cQDUqMy_O?jXdTEzGj*!G>B<Nklhh{KX1ce
zKl<A96FByGyY6SK_WNx&Y$!i3Wsgg)rRbaPhW=k0OwgNgg|dDbQm(a;M2IPQ@v#5W
z9=d-*&cL6?|Mc-0!1_G?(Z+2kGD%>O#251e!6J%uYu(y6A?sn2Ya9iOT?eW&Wz+-a
zI8DUZ-MXZ)YUw}|JlB=n8VPUNS?&&0gK8IYEo|!r>8vYSMXFbt&7fJM9b)e#_u#nU
zz(cN?k~4KH#A)VmJn;R$|2Mq-@SY;ZVLEU;F5Dd_j!WXp*AL9g#NAzF-k=l@&W1CL
zh=X%`lB7zK2^H;P?Gfq1?}ky;Ma82u*MEpcEtQ-LTh5e{Db<6#`I!6wuU$th3iDw>
z(!>;JKj8cI0dIwjNnG4jjdNKsjN@TqN`clglPtRsy_j%56%xXmSC5F!ln{CK`d7U9
z<~N+*&#c9X<0czsr6;}a^ysaGbV_!wh?4pd*sD2a<dRWU4vTMDRjap~3@{3Vz4bC&
z85f;ipuOc_2-fcBXa}ggO>Bth_X(>%X@?krnR&=}{n5t>?{&8OAzo@Dr9cf{?G4g}
zb3f@b)`OBs;>|ivAi=dQ+QtZ5MzYkjcA&IF5f-u9#mZxFO{xS{mDMt}D#eslE5#aS
z83FF@?kFW=rS`T^+d&Ymg_<|k=hOZ^wJBDGIt6OP>H8*KlgamrQYv}P4uVAC?Ykd2
z<;?@S5Si1&G6j}N`118DraAIzIifmuklA;vD|q7kvxsY81$CAh5?oXrC4@$Z#x%R@
z+9emBPiNk~d(ZRBi7gk`@r&KCpg}H5m|P%xnP<#g+!p*=q-o-CIPQRJoQ&8Cxvj)j
znP+DLMiF;G3`^JXfb%JHm}kEF&9Aw;|ANO?-|+h%-|^#z7pzrWUf)?8?ri0@<U867
zAb~(a<;uozW>(+xx(=Bsu9;HG4nhrQz$-6Lh%#mid%|LUPC>i2J=sXNhw|*G1&?N9
zAA`j?_B$c&>)L*C9rxeE=jpYLO!uFB>SMHU&m@*G@p^e6T7nYEwqew6H7{tZRKwTu
zXty>8JG9~FNyJ;?Yi7%pY?-Y!Hm#K4*HNt^!Faeoa(_HH$Pxl|Zy|(|GiBY-%j$DU
zlzb`FwQ`zXXm#S@)$DAiUR<!y%^NDLYhk@)ZyB3rj`JN`+puEXAMg3%@ikL~DJg&X
zt8X|>2R^(#gGQ|N#7Kmu-6c1Ui4&MYU?Q-Z?@<ye5(NoHu8o(M!Vhns`2OvWE^HqY
zQD#CCYJ*aZEst|BGffkR<%nFKi2=TN{f7Jdd)~ZxLn-j|ywbE#RY@^1b<M9BBg=dw
zlm)Q`QNv2*^n77TGhcu6x{Im1Ls&(-A%tsQ8I2JA-|S~`XDRN5!O>3SU`K08l}hTM
zW2w!V+(TwzXHv>dOKCzzyA$kfhj#P7F#58PeJx)>>uVw_*T43E_aewg$*%qE?*`;r
znL;GaQ`bry0DSzk!7G3s`y`Yl!dG{{B58oLlG)gB`3Z<zIuQM+4Fdi9^jInSNtWEm
zK<f)zHHyHwW?u4!rbd&-)>()T4>N~D^22nr#(AGSO}&-Btt+K8wyjd@hPDDT#F)sl
zdZCl6o1(`^>VT7GZ#h&Wr6Zwj5QerAL<vk1nK&}ZOiC5holZ>+Qz6+*VrFgb5*SQe
zzb<*9nxM>!FilWPVJ*ssm&Uu73qPJdu*8WuPRz#xAtbg^(J3)oVd^?=Av(aN2Fvll
zuYUV$>Q;Du`$SqMriaFMIkVPE31Cy;a5sBAP2wPLn8G7k-;=kMToThf@%Ysd+bZv$
zTpqzC5+bd|E-BEE;Kd~mKh)bq8LhZ+2a0{Wm)#iVaCA1JTJG!6bCM{xF>aUoAR)LB
zz=)kddx6U{Hj|HwR=0(x5##TZZU58uqS8K>ZT$3O@Y>ARaSD;NxB-?~L&Unf{inW;
zKac<1@fpDSJpSRvO$DyUC0$ffh|L*TkIOxghz6(Ams;4WS8Ic*R}Omo(zv>WxmdJ#
zyO1dtt880yoeXhV&?-s`>Y-2^7QKP!v=EwBcMBBZ>({Tn0xgkCS$TJULTaOK8#>*g
z(W@0Y*jn8hZ(Uc;nt8Z;Ktkl*<;0RE?v|MtpyUfdgk_m{{qVrK7Eb3g^OP_%YAaZ6
zUNXjZR)lmOtTnF;g&sIte+<}Ou|8cc<Vz)=D<4izobts%rkI#}V(pG>jiQYZ7m~Ox
zL2H$iX0K8Q7u|I+(|`o?z&i<d%N?)p?m5Ir$OSD%D1~Io>xT!l6h6Fr2O0=A^YY<D
zX}_mUe@+ZXv^oG~I+D@OgxcLj#(##!h<?P#sMQb2!@+~=Qfd#v)hpjN+(-huU!?f!
z5V}T?j#{!IP51VfU2AIu!8g_2+abRgAGe$E_~#&&nvzRlo&v<h!O-#OUoI~F(;m0$
zI7B8(EHd%9+>>x;yIeABiPWyyA{}fPz%tmb#o=*_xBN7W^JVkIPHd-wx9Ylf0YQ>5
zrHGKIhnX0IyIytnWGe+}#^dp>YrlEBOUad*167Q8?+gYV`fyEYRjExmUpCfUc=h@V
zrm~SQ&m5-2G6hm_XVDatIXRG3@;yNXR9$ekQ>EN8x-*XO0MmusgSXfT7~6n3+49DB
z`H6QQp7`<UiPoJJtEzWWE)x=i>-uOVsu9z~+!GH$I2?||m{`}c2b8rGZ(q_%ZH2OB
zTCPZgS~FVDSgQo!{&+yl`2Fv{<?imx<y?7QGr3fpzZqO$zZs|pjcNs<T`Q<v)L(jF
zOJ{ZLk!kFpKu@S$INI0fF7V2581i7($)VSptb_bl=7t{;0_MPo3ysMBJ3}Y&Rv>!U
z@7e!*3rPL68te?UT)aI!hPXev+3h~{iH(8B^!CV!#39T)E_Wm*WJ+9eB&>#7961Ky
zES`q5aa;qcGYm0rA8W=rZ*0XiRz?e9(@LX}!bFI!6PZFJ^|*i_yqwRh+txKH8!yi<
z%v0d*xNwY-FbkoZeWVz?J$FCnyiGM{XMJb`vxQ)Z<I<2~m==~0SfcpcHDOk$H0Jq<
zYWL8N4ydY;n21cYy7>i2F1RivLJCYVlIBDV%7>R{v<cfPysVi`8^>uz;?&#jgfJ%u
zQ_4z+6EPi_=Obpq-Q^h*V@?yTG~WI2mdkm?aA3c-N~vyCv#l#-OKxYMD{LE~W+p4V
zzFW{{y!+vI&hAPR+}$zF3uP<RY}Bf>+JR_u(+q|t%{YgNTt)r-?pJMCg@a7KN41K!
z9;*!yt0%m|0)(Din~UVjb!}qoT1+1U-qwp-wgX-M_*+)=SkuG#%ywk#=S$M|A8U86
z6>OIwZQb4WlU&Mu9S`TY*cK*RxQi20nusZJUQg)BS;}INf)IoB0<k~IUI6GIh&2a;
zx0)#m*}MoxiZm%SY0S%!Fu`$|I7|~Mdci<V*|Gy7*4Od2ZkQTV2tIGjU*F6PRGPUA
zL<>-Z$DyWyyur2_HG3oTtA~3CjZ)7%E=L{?Gr}2DVF>}93Wxi@VUq`1e%;+Mhd_y7
zRA>#`E~p3^0*50^^90SM1it<D4_p-9W_T%uJbAIsLg11&7fVmci*4H(v6;jjvqkR0
z%>A1;Ov}uA*;r2(-oJlh&6{5ntwa;TrJzq2p1ys{-68SnZbIxCbS270N^rj{1XF(Z
zAOA--iEJSfbtXu{te`E^+KQ?hM3~%kHOSz*kHWC^Vo`9N&F1cYRdwG#uUOImpGJ&6
z<Avc4R9V8e_1-`(*E4y<@@^onZrEdkS9iOP2eZ>haLdY`7mSm6tn2=uzDCssXLPW=
z{7e#4+Q)eJ=kX^zJ_A^v$3OTO74K17i{fmmiNO8r?FtRZt&&5;tdM(v;X#IPe9es5
zEo_GvqjuM?)*7Wcv!>zVi>+J8yQZp(HpUowLS$V}%GpP+?;n_B;t(c2oB}VG7k5A^
zrBie}0OAOb2Ur+67cN;jE(cPYIM>1w6Nh;w2BFmrnHHuw@o+ryDsR-9nL_A^n71-U
z?F~o4D=`jWIeTPOqy`n&``Z0#sw$Ul<$TVZpEh3dnXPSJJsS=rJ?U%ef!jO<QuN=|
zrpP3F5@J1o92icBp0K@oc;NnUWD>zzLvuq~C7N=7JdlC6FE31C;cj~5@^a?9t;pBE
zA<UkLYX|*2pB=|f<71IW51jq#%Knq8UV)b&z4}zUrMDv51!=hk-UHCuIq8gm*YCZl
zG4iVGfq4U9?FsoWOz6GHV>e$vJuz*c`?VCBraf48oNg~ueLCajW4d^8WU|N{68F<W
zk_lpB6>oQ4OQZIn!WoC`>v)LpN>7$dgC`c*+k2fz;|Ycay(Y>rkfMty&-2t>_<a3~
z8m&~K!u{dS^;K%&)T+B0mOzWXenJRWgM%)P-kOrPN~w*9`v+p3`Or4zl$fIfSV6o#
zJw<r+aL0Bz!g&B#1vGTma_L_qv2L!RgtI0*@yiVGB-hN-`I-06&wMzYSeA*@O(76r
zitZ@n&Oy$0h-vDettT(basY(9xuc(17lQ8&W2y};)m=WDp{1Z%pj86KVR2X1cR&8f
zdd_GKtON=Hi+2pyKnDe^2cfkyH}t0UXlSa8)!W#4P$C^r*3N+QWGUEK?L#@u=Z5v9
z%$G{={lc8VwI{Cwh(Y^aJ#;&gzn`BlDaS;<F2BET9S?BK+mB*!_;viQ+HDB_360ye
z+*@5INz7s5ZaR=;az-WQ?KIseU|<hwJt4Be>KLq70}H4L>srWd6rlWqxpq`z35giG
z(_U~FQwyG0UAEP?*w)ad<aH%7Gv6QG?T|oa!>kc{BF+wAYH#sxwAS0<Y$k|un4JAv
zOJt6LIrP>VZA>Oi(n!+@YYVNUF4%0nSumkBV`7MS0$BxOSeT|jOc4pn%lS;J!nQ`v
zxll|wbg;juYwRq+^`T1Hn5H`nVVY)N!}H=T(GqF3ak{*)<?0EWDotAlvI?zK7qzd7
z<%VtrS|L%nUnaIK^TQ86FikUu!vo8#Oo!9~`{sIP>O#;)b5<M^euj<&HuU-JEre!W
zXn(ZR2504|-q@xXAg%xV9<hi}`+h(9o+yLS!CqK0Mw`-jy}riv%Aqv9>e=n)9XH_i
z#{l%rTHcTI9xs~T3%d3d7}c(xXP?aY=uGxD-)J+@7M74$(!`XGs445ZIXVQy=<CLC
zmuJIB#E*)1&7i5rxCNmQU4kKw=uJt}LRy5w6uX;dKO@X2wcxRq@5?o3V(`a}w%pp|
z@YVtGQA+UET<P(XbwJu3wC@iGR5O>$f%|#pex3<zm<p2|Xt6RKex#L($O3kUSz@${
zt9h}@@`_48Lu9&NIL?u(?vD6!erC166Ehbr)F|EEI<S?3pv+8Opv)_Cs1S_F0>?1%
z=G7a*6nR<CoG&XcFK4o5ciZi{Te4LwW-jkuNDl|X6hSYTod^_CG?v+c>>qyrJM?&<
zy?S8b!Mic61J5m^R<O`n&vFHIjbS!1inKt_tkeeZ<3N|0yTgxeVV4$Y{p3y^gev`c
zhJKaw>#En)H^8esroIKHVZ`5d!`$8_0>qi-*U1gH>$?-kdTDF`Si||*=TBqA2A{{D
z`uGfBeIEZPW1O>Q-m!O>7Vgp=cX1++$YMx}#57ULiDpVli5Q|giH$(DS7mx*9kn%g
z`7!U0k5Qmm7aA=TYb=L*rX_Kh6D4j0g(*gsW#aPgi8WWgsDU^w%=Z&t+}$x9f6M9o
zfy;J6Oo)<@<VkkZO6hE+hlhKP@xb%T3v1rEKO8uuiPv`r?vD%8VWJplt9*I?fSGbS
z{YWAcTJiRx0EnR~G@IO+MLY=`?2Vy)5rXUdh?rXxZ)bk^kKdzJXf05qVKGsA&to-b
zZ>7m~1I&7*gFuY4Yk<s0OqE=6KcpktoISE_h2tUdn_vGKk9QBO&nKh`G;#X98D^PQ
zSI(CcYA4>y2Vz<X>A?N{Gm%KiN~=PQ3(^Fw1*M=;5m)ExXKQy^?7AG<QKt~R^@tR`
zV%G&Hhht?hpo_zXUNs)cu|Kct7dS9Dl4RQtdl=N|4)|e%WcTqb9!Z)EFiv}N{!0gD
zPul2uBh_w*A)k1ONd5jCriJ@-<QQfmiArGU?Le)aX(f}=ObpRMBoP-??X6tSynxag
z%__MR3huC{t&&ToXr`Jn9cSV+^lSplePWuD10l6ywej@s9d~Kww}1YZOfe9n@%F=e
zo?f0=%bAupJh%%?F;LBggomcu<HI9&2$xo{95_r1kH>qyetgZ{adCZ&>}&VU;~nR)
z@bsK%HA0I-7NR7oMF&z$QK~0-X0#wo>JEgaMytx@TqssJmyNAeD#j8Mk|tUcQ0wf|
z&MNFW8!1g#6z=XGk!^EgXm?9;pgwr}hMKcpmWjXmtAD}WJfT_9ZKLHvYZ<J9Ra#wn
zIzO>h*ByH)!rFxUFBd|3AclnyFP(kWTL=gHQRpBR4BD(OUnSQ<RON1YAjCi|n}cxS
zx}-v9e2kS82Rp*=*a&rDPjkcmeUU3jB{CRTH|)-z4eRe$_h<j<Iv&f^&n;!k8!^o2
zat)hy%<fYjeUH7H7ar1)gSp0`65ZDQaL?(yky~}fQwWq+X%a}1y70GlCYQ0*Leb2c
zvp1cEL~EI?uI}oE67DAA45*Sq<o@oCtbt9XyM-F%a^VmX|J(n~f5#yO?xw)=<-}>d
z@b>&f=>{w<B*M`(8C9J%Rf}Tg4rOn?{@N|#-)3j~nepcJBVT|0ND4q}90^D=%EL1!
zO{}LQLGKZpP)n%A-uhdq^@S2<Vq7?wky@l~-a=GZscNiEy<M?`0w$riE;}$Fs<h_q
zvvCRp7KFq+FI=_@fA@F4LqwcWsgSihvPM5omg9l{<-huGNI}u^LcVO2bD@<p)>dc{
zYnfbLIIn>pKfLfoc;J3IP}_o;FvUPKxU^x)P`aMqbTc`wbcWG>8=|<INd&~G-9)3O
zw0(s-5H^&hyf`NNJZZazrn7!}xo1pBcFi8RqNe+FwT}|ivGV2C5tWY@=9Qu4k1x%I
zyyK_<09QW8r#VdBoaAv?cuWT#<blZ|v4VvOIX?2?vT>;uSr#Nl-<Jt41l})Eh+dRv
z;%wjZ)~FJwb7X6UOUY=;SWr@!2@+Z2#39Wb=9y9>wFTDA8At27B2D?~i#I$h2fluI
z&3Qd>+0MMIjjc8=<%ASvLP#<UJbHmMDQOXY``ZU31nv%k=)^YN@y!>n`R3IZOmV?9
zvPi^YVV+h>=m45DM)6EBDk}};#KywHfdh*OK@uq)2%0E%rl?X_DcUFq8WI*Jww4Li
zIG7^Zkm~x9vnaoM^#%EO<ipFEwPxOZ_&{q4F;XqES|g7+7>sa?{N+D?Lkb`iwo9X}
zMy{FG6s;TEdf^Y>{X5c^UvYSSN30W4?@6*T#RU3-X+=Zn>&6w$OnUL=W=g9mrK-1L
z&WUN7ItXR%f~fd<@9p7%gzPKKdaSOf^un`+zJ3EDE|Bli0{bhn7mKV134ioN9>wj^
z&Re*yXCEJ}V^h{roS+~e=6S%?_4f1lQy-rJtk2^gZ^(7)iou<^q6F_n7kV(^!sdZt
z5gmG!z_kfSHP9sX<IVP}Yunq=uD0=tc53xt8THghoRQ^DM;mK6ld59#fiy>6-QAJ%
ziSuPp;l0fQl`GI8!aUE&d}P};YHQ4CVono>WoDidAqEF7)L5p(ah{#Yq1r1bL-4pW
z4y+e0!~;iXDqI6LX9IY`5JMnWxU3sy5!Kj(%I@xEbhgW=aKPT6F1<Z4st4G3+h*@n
z>_DRn(yBp9F7~XoK?Q4r!Dc;~bx^6P(WDSNb2#M-DJ`^?Iuox~!L*`b2txNmYCorX
z04l<6_b1nH+eke3YtpaFey99;?(GL6H;)>B?`?|E!4%uuTrk@SSid}sUH#n#^!CR9
zslD{;tgmqU2|rB`Yfm;h05yf1b&Ts#q?Fv5(427@#KGTj540yKMi<u~48hi1`g^Q<
z9V{K)K@#b&5hiDEsJV`vI{-Ea?r4>QYgXMK7PJ*gJ$Ke#f1X_^7{+{C<^_{TUArS_
zNG#LD-SOb9ivpM;rYv)!Zn3x1Rhnjuso&LIind$Dcb6`!bTvP(e3~09G*gP|o`UIK
zCA|h~wzmR|ckJ&u#6*adQd$?F?&c1iDbvY@NC+I3BU22Z-hx|e?TNKpdpC8^&<uBg
z%5Z7I{HiiV+bg9XnJ|p&*RSh!?3tl|j{d%OQv~|+jXkP=9=loYA9JUEvW|Q8S9Ui*
z|5Jvr`|a=kk@Y;>xvqbgpZt9Mx{Gc%)m<#yS-3y@l}~Ey>Apdd$Q&YRm|ozZ;uI56
zN<XLl>{IQ`s@Pky&6&=I0~%GG`PU>ki_)Auv@-;T-IqDDa!M1mc@ompTaId}*c5rZ
ze_#$UTcIrzB8^i??o!;@(9&soUik2}7bFm3<Zze?O_)=nrAAFN%QA5|B!ajT*Cc@!
z8YyKYZm3;QTpvz5GsmRI6jo3Zszo=TF(c@;=ux$?)lBWK#cE>@@N;FH=i2q6BuK}I
ze9bpC&gV;i@8bS4-4%>}jYAB)eteDAjpyy;4D?b_b%$jaimx;+9wX$!lFy`iLN#L6
zV<Dl(A;X~I3L<@ccJk4m!}R33dTeLY`hR_I)qYmW-qvsX?+>#JKO4qOwEbn|mKA;T
z^0x{3zCXzL!Oiphk+sZV>=KD7>>!)ne40;fjN~7LIYyRvJ%0rwN?@KQ!j}CPy2EeF
zy?so^I`)$XVUMqCGg{F7j2;X;-*5eH$F7+aW29*S_101YYH_KK<Kf6*nwdo)B@zeF
z$|RLs&IB9sAUikDS44o2g!{=I=cZ`PNti<7FeTUC!Zr9}f+P^70hOj3O=I`RbMV~#
z@+ukWT2Eehs|a&kyaPbuwRqgyR!3uOyE&7)RzR0YP^Or?*zxj#OI}$oD@Y*33F`p+
z-hS?zE)S1K(CYer>LU5I)&AO%+BPohGif_OzL2yL6-3;n(+g$(KCXQWzsw8vc-Olj
zBsinpZ>4p8N7$KCVXWi+o1#buwB#p_!Hlk$VaOx{EM{L{z}|`T$7Zqr>?TOIeRl7U
z+RtQz+Nv{nV{c*a1#S7v!1|xX_zYlu9{)(=27$oAs)z?!qdjCG=}Slut=qx-j;9y}
zBifMac4ko(Z%H_B7e8!+P?~E!NpJU2X*3fDl3MeFGIYCZYvjD~Z~x|Ba~CJ%&;J>(
zzWRc%zxWL=De)`=2y%H$2}TSvDJj#EIo{nfA6~JZ&a_&&KOA{DEPVa-8xF^SM6m24
zmWLQS<#{4jSnCHGLIn<X$7Bnpf~t!)E4`Ya6(#Rfdeg1DmmN4vN8Zf8=H;^Ta&f_R
z%!umR?Q1PeL5Lc<YaeXu1q1Hy9tg4U;q;8DS1)4h%(l|Os~~JO^Y8!tf8;k`e$AKv
z?4HYZVLNYJHX+wuceWLq0>L6l7CzL>xxDc7?%y$8Uh(GsuQ7v21ItuBBa#uRh+Hea
zqYY_Kbh}nojG+T|U2j8rue>TElMDOHfDXLZQ@LsHu?`YS>Zho;JzZVZ?3$SVk@3-y
zXycIlxH`<w{xu%&Mggt|;vBK~2`9cH-T6m+zF6zL^PZ}R2utkiICkbB+7lvY$hjJY
zQ@>01Njm2<jX+JV8B@(g!sBpn#AqcGQ)EtaZvi!GF3^;}__IHQt@1zoKmU8~4>MoC
zdF1iU7yS0?FFCExT-Fs+b^R-nyK+u2&6Rn9yT?}$B2^1&jmP61uOA=y@~c<O2Nx$7
ztt4wa#Dv8_l$EN5S})Ae19=Z;U9i1cqgp}Kyj89NlE~%a-KyagVhNRmghZNVXQGwD
zB(C!ni=x4Z8oTyeWxHHx&B$A3+Y09`6GLE5i?c*arm6?v5+_uJw?Dk+_5G2rUfr|R
z6YKfnBFx%2Z5J-<g>zZayz=Aanav_q5|6A*@!oYda91d!fNUTWxI}$V1ARS(;NsVY
zAdZvo>stHr+gk^m8j1uL^T&GYNe8Uma@*&Z5r|>NXao(Ouv4xee|L*=tn!xvS-XjX
z;Xem~GBaxRR^WXN0Dqi)`RNQ7^bgK{FfSBg>aDyYENP;|f<$9VlLIbRU9V<nqV#M<
zFtQekx{*VZNCAqv34qUc9N-b*pq{iY)8yJWVr=J3&Ku{`iDDQ2gZ(>fyHK9q@#{bT
z6<_}POJ3a_S<6i94!*bhaEih*CyvJhsTIPMnL^{?F!_FEiNhR-Q%`<Gc(o*=UHA}R
zFsT$)oOkiaswcK~LCG{S8xCSwE0`A4CLYpE3JdczquZJ1(+3BIB3$xH&Y6_l@l1*r
zG8QIO8?6<#wUW2OyQdc}<$}h9#E6DSDV4314pv7@R?HNw;$H<yE^N6_tWsz^uZ2s_
ztYWMJXB0XA$mLz(82*xAOLt9F)K=6+J2UK=C!i9Xu&kYlVGx6mLO`3h0l6O70HKF#
z@&pnIVf6&RF1Y2{1@Qf1{9XvSp@-F;gm<Sxp@VGVx_iGcyg29YdAx%^DuC0U&rCTz
zzi>DnnCAOE;T_F{pXv}Q!IWc)98;vM@2N4O#|51XjW8EyxgKz1ky{~$(Z*>yg9(K*
zxiu=4W<n`mP>j8;)C_N7jY2eG65$|$)+Il3skBzv&S#D(@%o$J@ZtME@c;h5|6low
zKl?4e{`If;)#E*iu;nw0B$5h2fnv_lOTpd4qj6XcD2+>dLXxsf6DpNe&%B%)wI;s)
zDu7iY!kXXm-G}!`{1q1Nd9in_JW<&wR8|2C5K>^(N~no9ufHaQ$jjx#ww)=b3$3|k
z)=SB(!I-0<B3!l=lf-fe6f3N{@qT^cyk7Xb@4lmG<#0SAF(GLtZ7^*{w2VfmWHc&w
zuivnqFFe0Lv*yB9Hct7;wdzQ$VIxcB{k!jZetP1se{<%yU;YKn99%bf!omx;pq1@n
z>4P1e`nlY^)wIi~n7JVm2BFKAxTaM<w`FpMRk--$>kL65mOa`VF`chxH-K`Ifdgs{
zV~P%iwR&5Q`f0%30b~&`O6d)T_0zd7hHB;Xbmn-RS?-TS@f{$Vek4lrdHm^)&j8lv
z@ee+(p=HPNN4ZG6qT~7t=9{kc_L#{&VmkTn^@L^zCZzYzdrif)uYx-WYJ=dFg4!xn
zrDWHY5Zrz`#<T}NQ%oGDnOFA@9Ov1AF7Z~w5k!ruQw-jVA5tR5K$5^b%^YfFj)^Hv
zF6tW112U~pTetcR%y9w@l(y2$$S6@GK@$>2S9h=Y3_;ZXIi*)xoPglqUTwMyU>eSJ
zs9G^Gl18Gs6PjxUY1hQ*+6G<Zc?6%@Tf@B4rVJb8=a&~=-QCmFczAf=i^tbgJyA~^
z=k-i(wZA@Ifj6TX6wT09R10QTTDVae0{tJ&6&V~8cTS;GmE%ZCn<utoW(l`0E2Bi)
z+k*To^zZJ}kfFOY((R8ANE3Ib8Q;6!_iM6xtp|;S`pOsiW5e6Us_R7E+`CV5+*ZP6
zRAQB8UB94f*~Ab@$$^YL$rNz)3~|>%aj{kxlhqz#IfJA2>SvFqXj-8KT8%UW4cC^B
z*4OyF+z~90)5Ox*kSPR$!4wm1N>}#4xUV9FUX6?~A|W!x{uNwgI(By(QwLMA$b^zG
zu#6FC2w5*6fozqm8`Wyxn_LG<2_4LG!SP^&1p&1hCV`leGecFuDk>4RtK(9$F6O*7
zYHqus!D#Urb8<Z18BM;9Urs0H7-$c7+~41Ge|X@#KfI@&F63ItwbIP?^_y`8D`PmO
z8Ll!ofZyu^9^kp{R5?S;wbBN;aR+_&g(v<0a$T%Ht)9wtwSD}fu9vq9sSogb{LWAP
zi*6^d{p{~oZrd*}OuPBI-L^heCMEC@khg12+F&&+F-AgXR-|+-EDVsqSFCGEjb>_R
z{rOy0Gm2K4b%mR8MrgxYFV=J|#x6CvGXqP`XfvKZJds)<sX9>J8i#2nCOVnxy8o&U
zwx;?vWCQYa8-iJagF~jYno*sw5C~%4UWn3kCAWgu9jH+3Lb1(x8`7vKO6lueh}w`a
zArXo-swoX`|4YfR?CfUKMpWtV#7Aq5ZOd+6v1YcCDXkG=gb=9O_r=b58*4#Txm-3-
zZ#BBRJ92+`<cA+F6m4wn!dk04KWd<nx^`AWCFAY4J+=_*jL@4XZ5&V1rM*M+q7R)J
zFI__rcIMAMp>DM|84j)8{TJEU;A1WO+V^%OW3VXu^U2sTuKbL@Z!%w3wd|{UA0|7y
zymkW^c7}&D$ijc}HU5c?zKsd&hCJ5UxLeIg2uMHQQtA<lQ{w#$rVG6H^$RXP(F}$|
zW7FE>QQMUl+AtN!kyaNZ$Ub+P`a0gW4NB$t>527x25pEcp&6;ahr>Lx#PpFq?U?Sa
za|f}T@kCQnfF(toc;kgJ6QPvSi$BU%3m46Z956F7D}_u!sp71%8X6^5HfzM%dTX3`
z+h8wrMTsnFrW$P8_Kprh!X(g4$*n@!c*!fLyi!fO(6}G&R=qm`bpZN61;4#OIulH_
z(Tch(h&7gZVhV}V^O;8BvYokXE5$O!luB?0s1)DaVnM=&Xy%$<jWbO;<B`6ON30A%
znzPu&HOo-z;O@{EW?w7TK_y@2qhlg;pox7S7+oWKp=rd4w`<D7f}8J*_~&-D_Q`d#
zKYY9@uXC)n<}I4s#-o3KEdF`?>5R_+*5~n$GOn#74$4?hf^y4<MoJMvnn?4^v@A^1
z!d7}!J@gH~XFUTpFoR;n*$?7&))#GTTF9CSNr_Q8Z!d%<#HWdED<IwycXzz&BC?Uc
z{Oxb~)#DfZcmM7Gno~aW`w!m{&3HH*n7UoGwt|Q-PYYFzv}C7EE4)75bBGgb$sFgI
z>5vFgu(H9LVXbcGogxp1M+!5S@{Ual7hSQUSc5~n!&P&Pic-uSON@XtPu7xFMn6?9
zl^@@Ipo#12crw?>^-R=&lmikcnkA?W&3Ol|zW@F`xi&2NH3%tCtM;m}YxYQt)TVs<
z;}0y$kyB~>*Z=MRivRjw{!9LM|NH+3|Nd`&&+p!U$9lO?B#>KWEgOXiWpX!3E#&$F
z7O}8lvH_bX8aK6)T`?m}UbVGAh=$OZhmKON*xriNHLv0XA??|^QW?f)0v`oc!=l86
z;+;L<0s9qX5lQ|1C_fd0+%<kAy2e2BN8Sd^w?Q;r_oLOydcF|vTyNs<Z&CcGJ4RmN
zl=|KpCyk;Nw6SF)P7_mHn3kCs6Gex_r(cWF#@FN;EM--y8XBXsfpn!>q0$J!AR6a<
z@^ySZ(k8cwUXHGNx8;?#73Sk>?(gpS^MC$daX3ueKOWru=aP9i95^nMi>QkOV$#Kj
zV+u@jBzlg+{jwltyvWQV0yJm^E5>nNV469vCstFi4_=`*cxn}~Bh`$eCoXoO+UkJM
z2>x0jsSvGs3qUezR$eZJGzFIXdoGtVS~9t8L<`*6<gQG(ZN085AKqQKTrw}~#YKtT
z+zIgO+HCim05eL?{O-H&Xsz+Z>qq{Z|HZ%Jum9q&_&@v~|1adf`5iBpmGin%2sr1{
zS+7luCL1cw+%(xhGGa<#v_N}HuK^aO9q5T7{hAs>RReVZaBl;a==yxUy?6+I-$G3L
zV%8B<j|cQtP3ue!iQZ1u6Cbmmhrhe<v(W?dagly0@x2yZbo_3A@lW09Xvg%;fg6IN
z!Oxtgz3pr{5W_;87pA^1ngO9FwR_u&;NtmEt5Ov#E>ta?>IoHfHf{g9l&(ctX11KU
ztQWEciaE1kT`#=6ydYbp{rJS!j}QFI-+aw1k?rNoo3DROToyh&Ju!(hJX@{QQo5jV
zCB#BPi53u5mR3nb4spS@jhCm!JU_xbF`I!FOdDJIz=z8d78=TzJlPLqo>|$bR>-1M
z9LNe~MK$pCt2-pZhvz3QrD9=WzDwNSPn@1!$ZICojY!~*I}8V<&zFr$e$V?KpShej
zp4thE#>2y-gAZk+6h)g6($HvxTnfMc?tA8B1j+oXfAKH)t3Ute{Ga}B|7X_U{f?*K
zKXJOOyk9RE3&=txQZ1mYh*TOGlR`>D(+F*1B&GvB2m@FSfuM%e${{9FOkLRD7?mqO
zyF1%fT!~rK+0_<Te{NfR%|*p(Uw2_F@3@0Xqs4pZ++888KW3RQ3P~{5uH633t`puc
zN^zlo&)%&Uu5Ll>PiI`)ww+z&&d~|lOxD8Nrw<(Oj@;khlMXWyz5UkP$b8KX-5@{d
z3Z}-U#fzr#4%LlI`*3X&RX~M{o+u68zh7An5hj76U5l}h^TvnwA0QXjZ@=gDGVvGR
ze8a2xNZgF58^pj@uU~tyc=G*Nt-9n#fD|26o3zr*8TcUt(*49C3BelEybx02?b8R=
zHS^(l<!OE4i4X3A$RlTaBFmW#cPB5gp;JZY;#yP9_^Q1SW#aAWg{@SyCGHMK?(QbO
zefhw7+py*!^qb?IGb~%<vR<$veE<BO^R{|%Tl@^JrLyJ=wk6aO!?jPXP|M2GchA7e
zBpY9T{SA-zUvfGr&+Cu;{>R_3t|!j+0i9kk9lk(jL8TE6g%nm~c|z)$px0umx98s!
zm|{dUkfOI&Msc0Z6lc!ciYm+`RC=r?&W3iENC1}rA?CiOT;~*_cc?hv=nXSNPW^Ut
zK#l(##`1X0!6Msd`n7u00kzyJv$`YVtsgj_;5z<1{>#T_0PFMkM;R`-?ZT>xP@0mZ
zy2!XHx;3H+DNZ{`U=liCr}t^Axa*C0pSJ*}u0f+DnpH|abX|+X0~GDL7_AAdNhhdB
zf=v)-EPB-=C_KIYK*s5@OG<>WU|OgJLT6r)5`n-(Bs9fZL2@OO%Du}tBigHd0%?kf
z35Q~o25oJWZKbsz*{rc@A=#0NQ7q6bv)Y+rD{2kIeV3|%X>iRQ5!O=O<tYWy6q%<;
zsnA++(P&)jp|wbD(5ezb;j(SyTzhrR*(ZY$)VeF1YhS6eV>+X&s8@7<`0#=M_`Bcp
zhabM<-MjBu*Oi>B+j3jQLdFUCk;CydhvOs5a-h}9d0jCwA*cgTc1=*VV_9(=MIm$t
zBiCmM1WJ!NorLDWbqj(-I+%1VR7dSqpkCD*v&lInu0P^VEB=Jne6A<rW<LY4t~F|d
zTc%e3&(Du$8+kXyhA{jmN^riZ@(!!wODitYKdCd<in=yK!qpN&?4Y7~8?OatB!<46
zJIH4?0wL+(?UlLY+A+3Y->p;(BX-gL<9y&*l8cI0|Ng6^CzPcWRJ}psI87J{(dr$f
zHY$nU_BTm%kk=GSBR1n9Bo+-&yj7x@F?ISt%axilTYX_|!lkcGEIaeWR<aeEbIN)P
zWy2&eiLgi_puDUXHf_wy#54t_DUi!XZK1c&bw;MS9?58zyyT6w^!?7vh2arucU8o2
zx1Y|;YHF;tu-43_X5KwNbNt~4-oAUs`=|G;c_r5}YQr>9QhH#S9$DrGrs>EQb9adJ
zYZ}bkC8aZJeV*K`0&`Gl7gSev{R}a=GI~PnRdg82=qjTv=;Jvk{=wLp6P>`}+rfa&
zuF34r*6>Xt`_m_vULqd9a_e$7$|u9zA@pm0lR<u3qkpu)zdhGNAo4JK;u95uy1vsS
zff&7&-q)&whN`Wf$$j57(9ql8OK+cL^D8~{Q3e1AwANiHr-T)E$8*;*V-ArJ;Eu@a
z`v<=M<{L`O)K*CqrU<im%Xr-`Uf8haAZeI6Ob0YHQgydJWOBxIPufdql+q9hY^uDh
znYT}c=S|5$XtvSpM3xI#GT9;pWevUm)vB{sL|98Ewn)|uNs+IPkA!G0M{KyhUtfo#
zkXKrxwGEZZxm?&ZceX0NKQ+(-){$tf*72ShEOE*gEJW__0++n<<GZ)K{qW5D=M&qu
zQp$?j>Hu+!9G2HS+<nc28%R_;Q9EdNbsOsw!)(8oaJ_AT=q58pH+t#kFn+>HcXiWV
z-iSjB=Zcp0HGHE!x6?6yX)TY(4uJgUYxb8`^7SXzsqTA0pPTp`c4-I)uYx2#$;9_I
zskgwTY)lr2_&i@oOd5&RHQP)K3*OC88ql?#5Ra|3-@S%Xt+APrsWfR_*-Sd)e{>7D
zjD(7dl`D$13Z+0USglaptmp7>;8$P$2Ca?LmPs)%Md7aNS>{#{8YT(r<_RezZ(D0d
zXb|e`U2LHrHEP73M|^mG;dEa4;pxoty3%MM8zzAoE>vA9)*TiNCTmPqkj@ZY*Od?|
zEo{Uk@aFy=a8oGjfoC(QCbXh{3{|CuObNzXGNo1yk`NPGv+l46bR{l#9}id+%4WQO
zd10B2`7krh3-huNf)UJFTh#<(BFZC6{0%9-LgEbKady?xi*aME4u@F3w+L>4)HSUP
z>(@764TAc{uAvuY_O=E38O-kL{c&06`uBkBg(Z`5l7^cHjtZ3V@yZ;3y>>eMG~&Kp
z;kU1EW_v5E-Yn!`sRON_$DjK63}Ag8|6pTxi~zv8b*d+fYp;e@Gh(aQx;ZG6j(egS
zLqrWO=JfVp*D8PO{o1tF6XdxAKo?IKgHo-ct@U6i(Q2T!umf0$8R7!dfypMQ%6H#=
zM_dB(5V=b;5+=wSTV0u_BQRldBuHSf2`P%T6<b$AYdl(HwuDqcl2<O5IUotv51CSh
zaw=?jqwy^({`o3_#D%+-Q7R|<K(p0>7Nk>s3l+gM5{Tq=qevqhCYHm@a$Go_ge^x(
zId`3mMlFq8!R_uhUM?$JE*OG@=<QQo>r2%gdfeI3Jc#b1yxAJ7R{r+8@A&Zkg@6Ay
z|BfGic*|+4v}$N+Xed}*z0xEz_m98k;r>gOU~Jo&{NaSMk&^99j6I#4JTOkJc@;<e
za|VQ|Gb_6B;kDu`5sQLFh<2TB!)K5nXkZVQ{q8rA)=5`g8%%El>yh;Tr24b}_bU5(
zJ&Wz;>*qs(jgc&VS^(C^>)5Zqi8H94OGXvGJ~omD*eXFL($p(~Jt1l1p7nMw52A2}
zjMBB=T&Awt0I({`NEUq!o4{quG<Cg&Ac=>=1MlKHD#f)kqVVSKk+rV8y}a+in{dA@
z-p1B4rUgrjyEBDEl*BBGX0Py)3#lrv!^~m<)<!jh8b|_Qlr@vL6X*PUE;Vwh3A2oH
zbTHU58=2}&3wt2j2+R^l;Q<J|J3k{aay%|fQ)Ehkb-Pe%=?Su5uV%(pye*?*oVV=l
zgN)X$xNBOd4)NJBm%BK0**KStQ(5`;$L~3xH~#k9-|>g<zvr}`sI_?ub_cNz%OiKk
z-*UWr!*V!sIz2%i>SgKxb5CSM90aZmvHtPjo|KDuGL9kI17q<$h>^IDc_vp6!E1}2
z?d$joU;)8F%fY6&nYsTZv-~IgjnOK6tr#2rp8O+tdwP@z_UWwWtIO(j+p2vnKa~wt
z_+n{90+T{)5SlSdfE3*g&!9E&q_<y-U2+95_d4lf?v)~|uT=|(b;sG@iB)Z0tcc43
zb)%N-8i#5uAu&&pFIf2c)f@i&FaH_e|L_Cv-oGPlu$XZWKpW-q<f82oKo%q<?q-8l
zvF!v+5oSUXQWA_t&W&|zR1>yVd4F2@_Q#dy(pcj}sT;L@;KGR-HrAHNqMQX1$~-jM
zg7A`8f&|vKay;Jin{U2kU02Q*H;yQ+lFi(kueYVIxv=TRs>)U}O&h5@xN3G5ixmdi
z;_PS>A_uUEr}acw4!k-Zcz=H4d@B6@`ycrJ?RzfgGr68IJ0oExq{!Xf*SvZ4SJds9
zk}uA1)eX}O)_Q8IeLi<VcMGTtfKtE`We8FCY;`amAz;D3+{#{EElFHU%O_xCIGkN4
zjNu}AfOIXMzQ<mRN<UfIH?j`O6>z?eo+!)sdtbjDn6v%~d-stSyd?3doz2icsd;o`
zA@H(Ypeo11LI}b`m`FM?<*Dn(Is;BIuwKZtio44R)q3g3D%BdRHHsBlsD8gqyj4~Q
zd<81iTl%A)fvsjNXV@~93sNhA$o=xbpZ(ci^ZW0<<@+DL=YB~nbL7?RjLvOa2@*-M
z^%y@f$C=5xc58!>4@7HBQV}z(Dw7&wjUV2==fmm3@86!esJ9A+Z~>{*_(GADt#)sq
z1|cd-%ShkLPL~&gL|VLXSdRSq&1;^YPrQHHP;H1dia=1I=IWgp80WBZ=EAmRv^EYY
zA-G6>6ZKG6il3VXw20=wwkm&k|A8;QSormu1L=5VS`H*>h-EiL(tyf=IlSR9{uOh)
zgK5DohT2NeOw;1)26x}So_yZqt<ie1I&=^>>_$(qMuHkO^fNEYF4fZd9gcf($|s!k
zZijJ(2ML~D07Q@Lg2mo8JRE*ykNx&1^C#HB>A!y+U%xXR*SjBSP4@yyvP9_M?dS2Q
zJw5|ipT|Gk7_^j84KR9ph_baz!u4)~1VZrOK~<gUBwk)|O{bm;*-+{5=N&!W{(HTj
zyo<$}RreebcPw)&6TyNDFm9#raz3$b8>JRBE^M{%`)|MH@ykbk_2qB6{ccaVRjJj1
zm8B`sV2+Vev{!PAJKJ=T(7awqlQK=#E3J+9r!!Bdl}jmX&69L*rh{f@q9~|4^li4f
z$Rox^BWWQ8*W=p6t*1i;NyhOolOzyxqpcfxTdC?$$<PKcpjPke-%4|qhTOP$jsFQ<
zeAy2dHP-<U5_v29`1VK2%Y~=uz$IU(G*0<Mt4boEwo!E@Okb1cuX%j@iu=1SIK6wv
zx>j1dpymg^4T{=m8QM!Y?oKy!y#%4?s46*&XE=sATgp+AkwE)Xc6TF&(0i}>_BrXV
zt@~l9+O<^VcJ6*WY#g@0&<?pWH2z4k+-=)iuhgmCf?l7<_@#Ah2G+>x>$qBTH?f#J
znCVqDufypcWCMHR;|YaaE0PvG>BYh4*7c(z176VmIZHKOZ0&5qOsR!Z3(uD`3obnT
zbb97r|BwHgZ+`O)U;pZBBy@36Z65THT+2?3B!Li<JMguJRM+VkY>8TxG&vLW{qqZ@
z7Jhv9!iV$9#gxq)(y4Mr%^~teA@A+LWA8JD{@<x>1gR`Kks{2J(307*QgWu1>N;4R
z&Trx%R6)t6l;*6Cz5P|jTy~csXUN#@dS(hW!^`u9Z-4(C>+_lS%N@_B_iSxtZ5MZd
z!>A;REzE~EJUsrIm(#}ioN3hoU<nP!7G($7LI-m8HazPsjEW0J;|wxihuz(1XlwMl
z>-&!|mbx$5u}3(sQ!wI9HCV5();>iCOk+QFOZwT+n?FBhiQGQdc#_tXtrV6h#4s2r
zpV0W}T2>%<s<O3>XbzM{+~q4uK-BMvAUli5#`?ZFlhtrwGPon|Y8N&tZ3hvQ(AP%o
zTWvF6%j;U%)=b$pYA)1Tnec>bo@OrVg@5~R|83VU>I_}!uV;q#qKCIcscRYaqD;=E
z7iFB<qrn-DtrcFqc_f#{|L|}A4X3SgE;Hxmg__0%jg_Vss%2-P^!XCSfw0hlxKrK;
zA#pe!2toMp?niPiY+ELmLc<6#5|e|qiqNXB<E*|%m`K<2VrLhK+$57tI|FXCZYG{z
zE}UMz<G_)Lne+Mrp>okPSyw8JIUHH$KjUz`=Wr*S3;i_;4yljxbU2-kBFk0XaooXX
z^wutc=Iec|`f-+ywdk=&2ci0!zUINYvpt4%y2voh>1t|r<0vZd(Nex0yXMdB0dxE3
zPuGAyQETnfUCTdtqt5Je`G|8_L6xr;?Pc0TAQ5f261tk9x0k!nk(d_@!NCmF5E3Rz
zwo5N4v9p1jT|FxWqpE{}s?Gq;wRU~g>LxBG%uLjr`Th@o;N81-oL)}w)hniD-amtN
z?Wk%X6vW&th7h}sXk*)){iQ*%s;CONHWcG;fB1oOspxV?DTT|Hi4;Vv7iG-Z<71DP
zF4^f)n|xR=NQ}(WOfYzQ`<BaAC?)r^5jSI)oz<vjTuNre$l|QZaqcqMR$M_F2@Vb$
zMrnSAn+Q)Y&pe<0j=u@NXATGc=5POw4^Pir>WR%7O(v$}k*~h~5(^v8&nMb;rk0fu
z3=6W$S&T1Vk%7^O>w2SISGP98<bdapBN=C)a{E1hI@6Ec%tv?PHE$RWqhk?Smx|~;
z|GR;UuRZPO&#rX`H|rSw_<FwXMuS;io3fS4B(WDKKCSTj^Z1`WJ_A^v$3NV-J!JRb
zz^i;k3mNHpSl+@z2QBv25^lsicX7axsB|G<>CPE8Z0fK2TGj!O*b_YKO2Q^WZr;w8
zYw;>@@@nwYhY!T##57H=A)~!n#@t37iQcLJq!7FkRV%etR9*BIjGEo9I0SEb%B67H
zR$jK1+$ve2(J1s_f|ZCODyUZ1_M%d{z-qM0j3*QkBUm6d7n*)Z$y+^YLrZZT5PBlm
ztsTw0BEB`FXrpOoHpqT^`n|ZldDrAuBvNbR<>duUnKevo#f5FxdO@2ZOf)N0tzI>o
zUvW4*uq<~x)%Vn*G~F<(z>OGiKe)Wj^x7^pz#b8($_`+O`LV3G?{yrway`DgAP|0t
z`rn7(?+~jTJ^pImMsOP`-yW{sktH|Z9|`sKirICN-^|nTr@2V?4ubusl>q+9j6b%H
z&2_7aIPf$<2ZXvN5x4i__V;_(JJxUqv@r3NYu>&(S_m=eRrw(v8_+JCx|QtfuQh6G
zTy$lP6Dn|CFMRv$w>-XnMM?>ybTGPkn{ss_@)(UMLXb#RH<m!vg-t`)lO64L)z7Ch
zmu=(ebmDm{WNWS)CFWw!GFYLNIsmK@3z~Kos<K)^Oh2+aZ?)1&p_ST`@;*y~Czjrd
z<cV+XPMiCj?ibkIMr{YNMxdwWRQq+!ynp`z-7?$!!g)JW+UBltYCVyRC=+SA=XiY0
z`SOm|3Wu2Rwghj%HC&hPmjt1PPNDT+SH#7vN5TsuK_4K9U4aSl)wQo>x;XcEZn?G=
z{G?;?<B59nvjIlhSePHB<{$s|^)1@XkG!4Q-60e7_SjFVTj6$p+AonOTe%eyz9%Bi
zfF2rW!SNuMAKow5$7k|hfY8vz&)r#cFbLIbcZzEL?y7m>n{%b)O3j5<ugvKXgfvfV
zCG+jK-*Pw}IUEm+*&!X!QSC6M;e{Z%f-inPdt!_Q%#2bBrTTnVYy8c(zh|w+^6(8?
zQ?fN@Oo-AX({`WK=J<1Px5F5G-LADD0glrGYMh>*y<k+;4Hm%NL}SDqzC50&X4IY-
z-_9dNKbnosG*B0UW`I6{b<Mmy{lLt`L}IHIl}5IWY@6TPFf&bWn5LOD=^#7eY>>u|
z5LuUHxLt(%-5vM1uho9uIu<WuI_#6ohPIpxaK*pxN2_o80SCGElye8~U<kR7BGFHk
z?5=0CFo3q}dmRkv{_mfz&;CTJ_UW$WUs%T@;Mg=1IuM9Jl0>jb(D2cY=ZqKWF;(d2
zYVhKpNpJlQu8BwMzz>-7qIZU@LT3u~woKK2Ci}S@LLkLJBGGE&!~6G~PiMBgxg+y5
zIrg%%a1?7skm`k6p#z}4&TDNwmZzIPc#*iO^7i?e&EWpcBUKw4jmheMJw}oCy1J~J
zJqTjPsccA#eEE=|D(9CMvKp=RVx>Sxfsm57OExz?DFUTm2OF`g^_w$Zs{^*aCj2>7
z!GdsJSDv4~!*W22a6WBpC6n95yDsEFnr9wf-ScvG(}z%<dF5`R-qPzMFxEimdRt?H
znBXnitx;pb4IDy$Tf@TO3IctEDUFYSE7w|IFW1Lh0cQuwZ2!o=IC1&=*75B;-6!|;
zUpL2lQzO?#oUZz{pCPROd5q5h*5~mLHm+5Uf!bgDsNMLyc9)RA-QCfv6H$AAzF^Hc
zsiJ|PPK`Ca-SFuBHNuY7pd|X?+uLTOSC3U)J13=?r{@>`!{7c5CdxD=(*2RSD))DH
z91jO#nD#xeTDN@<iq*=xx?o~Z@x#EbiQ;hWJD<)hEfK=RcR#-4$EWw4HsMl~RlHi$
zT0v7H&WTnFwG|K#Kr4kpBlm+Rh*5)jH3}C`zK`CHabAVc+||Wu|NICkCLUkCf)KpY
ztwu4UGzST$E;gY<gQ#iO9~n-C9!6_$R#rMp93EzZ25M`(zr5$^`GrEpgs%6aRD($-
zVp#b7w|~bnzvY`(zh$13@7}&p_0p4GS5(+Fs;qT8Ya1k$-aY25fG`f^9XuLAm?!H|
z=-{W%4>S460Vqr%f<>AJXZX~%C$(y>lNCZg(0fZ3z^&8SH6VxULHWy(ucwBCKn*=k
z#C{qX{1YC(ypBECY&iSsVuctI4~NOyGQ|@k5pf2bS=TUk0NXlkM=TH!QkcBL*}4dG
z2P-1LaGp$H5{M*fGcLKYu7!17dqP_IusxI1_;7lm)XLr6J$J`D4u>OK?rk_quk<zy
zH)e><ll7*p>&E&0mf&DqtBqjp!1Ztc<KOZA<%!o1Z&+1%DH$bp?H6Y?nD`oMLAxWH
zUrBHvCj{X_0hHIv9cISM^E06beW(%;9u9Y;DH5mTY?9K*xJzK?hk`kW!kx}~OU;;D
zp(m%kI@$=tdFJtpM-oc0%xQb(`Ep`yTQ8Odh)3r5nt7T@$<SKST38+u7IC-K1`vBf
zvm>>>w)>DC1xsg6(gx_{>oiEjWV9VQza@@jcdTc-wjLP1*C%n)h8^rw-Ph^>TlNBS
z-{<6}Rbk9*KVWW>jGv-j<B|Q_wKdL{3v)bTbMgfIQ%QRH$y%1amirZ{{Y+lXb?xrr
z4AJ$6Oh%I961~^6?|LQ&^TcMdNF`9%C{`R$?dN{&iCs+)qClG1YUBI2?<qM`x9n_(
z)<~vUNZj4ulh-rv-u=keDf006KnQVX@tSF8JkSA|LKLIs>MX|QYf!rDmK(A(@|Jo3
z{s{}h;c%pt#wl-P5j2Kg5NkAYFEx2P%oS`hVlYyG_glua@mM2P8+lz3iKO7oED`wX
z@il3h+*z<S)>0@4O<aq~&pDrt&Rp!tJiZV4`!}px)*}lsRgU)$gi@gvzJLFFp3i5t
zw)P}_Lc+u}CEmUJj_2pU<+tDbmcw!3-~Zj2oX>d9cPjwx|K90T{fw0kKF3~A@WfX@
zeXoznneL4aK#L&SUB)^<G!lC2R`+4w<!wI`{CpkoV~0rnA>RGJw-3TN*L^fTc1QW+
zXR-jLRn8avG}rP?7(8O7V6M5AQJ`|sLJX06IpU@@SKz93kVUi?4ZA+AXxBtCckL}!
zXk-eN)>-``w)@d~&~0qZ$knLX$ho<jW-e$gOljhfB5w|_m?W^C&m5M8<NZC$@y<o%
zhoOhkB^@A2#el30TB9lc9=(EzgzhHU8sGil4-{4I4)+8z&ZV%mN(IE6#XS%i-Q~XV
zc`L%4CW66_mrMm_1gcRl>hc0%;(nT$!@}bikHi#Nt2-X9rBY*r5O)?*(E-pJEjUOe
zA-MKy2o4zTXI~^O2kyW4g6$=f&za})#A!WKY{O(>US1JXaNXSPh3$Oii`Sk{{rK#~
zd$rbq<^I`>9z_YEjZ~?x!RGsm3dH0LIo0axF-4lV6C@(O_jKkSA$F?m?YTK7;I%X*
z;4PQp-}c<d?^zFu|1;LH-8|bLUkJ4+=WSzJ0!#Q*7Vzis|Bvw*!1_G?!N*PEfgMf|
zPqeDKUR6w~8%h}_+N(QyQ^2?a8v~G#Ubz<rW6K35gSgGJN^d2Iad)Eg#3r!j%!ij}
z?(dJx%M8hxBl9#frKz{kbvpe3LaggS)k3rY8g}4n|EwLMYo$^nlo~l-Hl9vrHk--K
z#k@!IX~Debyj4`R)3Vhm()a-b)`jb}rw^_Fw|cd#LTpBC{yw6FUiFH^<c?suw!NK5
zdwW91V?1!RaaC_v8~fez$Ehi$7DN&fBBi;rQbT*yrT5!+3bF{~Z3PwPuV+e|P|KKA
zhWp=sY~ALKHz3yjO&g$9KcKAZEA3Um{v>jJ!~1)`tssd9Kq~$I=}kg=dxnmLcLyEB
z*)u;mO#jmm)Xn>l+sxX2_FVte$1ksAXA2Z-G*Q=mqt7E5K*sG%>ND>n%wPma*LfKF
zy0+`0-S4-A08MG?;>b;hkO|~kShKh77;rcoNGUleIi9n>w)G&bX}tkT?8+p8eGZtq
zFuFG5e7W#)x^RF0Mc>05aCiH7U)j3DI@a4RKA!I)4&wXga!{?6g7ttph&u+QY4WzC
z4vbg5?i~({-K^7|1dp*_FS+VFjmoH+n_C2os+HO@%|>gM5(QG4`&2DmPc9IX5n68z
z=$6&CgT?mI+;FVKKlNbC$iUKI-?zv8m5;8_&7>R8sMl|fJ!80H?N;85y?QV>#y$M`
z_-O6!OWm&jzImMMkBbZG_f}x*bNAE9dOsdJJkZaWk=WM0mWiPQ5gl~CDIWanOtvf7
z7-Xy^?ep>4zP%F@gnr%!*JfK+7y|Ac=ok##G)<H&)av>%ap*RN-hys7^WW(2wK>yL
z$HchK1urr=h?H}N7?|dniN;GdR7T-L`&qf4);H_R_VpFSL40p#sYvtB5JiX~GN%c#
zz%ot5nAkLXt8`PE1T>7-8_68~{QU__x4(b?jGaMj&JGonAj+0k&fBTaPXVT$L@Kpz
z)Oul>gegVL3Tm}uL-J7^GhW$tz^Q{*$n`$o?x*7SayMnTon*H`#1M72oAX3vpC#N}
zgZ@T5b{o&|)8`-TZT6La^O`?jm*cG(eLuDK>8|CUTxMU!em*$!w|P6}XcNP33IW$^
zHu!2{<L`rG(+j_HEz%+*K9v2MsdOD8HEMMxv@@1nTU5jgKg+zhVN=b_%glT@xGB=u
zS9bc_kcRMe+*);-rd>bjm^-F!RAkN1#^d1#s%&`CJ53YPr0)aQbA4R9q2Cn6OQX#y
zRh3+Xp#H!Z0!vI-5~d`~^MVApY{kK7RgWj-`qUkZ`#7H5@75LgQG-&w5Kvpg#0W9C
z+i7z%m=FRf%^*st&wd^Um?m-cfF5nw{@i^%-I}0`MI|G?Vvo6`2X)dq&?W&Hahlv*
z_M3QZKg2sQwI^3&CjP?u^L;}0>;9j&j&I_DpDxWlU)<bAKYIR8hDrH({QopQ16ZHO
zKkVrHj|@6#k6g4LvS-cAabh|~hO1F4E3Ngm6cez-zUhlczS3LcM4i$i7DzI2AMS|}
zVuV+7gea6GXfSl%7zWrMUQWEcyztNe`Y(8OIM9-_0A9U%<nHdClxAA#tw!`{yH~Ni
zA3hM9pz5~KF?vNfrG#o@+gy|9oLAm`c;Um#1zi#ac*z?^;IJGZg3?fN=7vi1fFX=S
z>3UcU{jL^3Xq?-bSfDgxmdGrTlp<4@INaS4QlyG+?rY9$3V9?3bp0{JtJzVBGPHr%
zTL6UKa#0IEzWsp@@BhGG|K&gD&%gOCD(XBjbJxZwf^h(Q0BKlw<}e>Q%rnyjF)PT1
z4VH?fenwpn^buZ(DQPe{B#`Wyv^tRJ35-ku?d?@jdMNY@sM0mb_bO2&?(7xD^Kxa;
z(Vu5jg0GH9Hz~|*V9xcaetGn;IYz=X(Mp9IrpiD0v9IGDXYR=@G*M3V!ki*;aSa;n
zOeX7e1-~`}HXTgN0V#z&h@3)1tc%S?MM6VjLP8+R=9&$MBNjGfTfLa5>hnEBF7LkM
z93oq*q-EjPzy1w}<B?J-Aw*J~8FIyl3sL9X(Ao$!I4(mVgs!|5BTXwMXI@?$yb59B
zaJYxG_<kax>sxut<1j;L+AyhCKzOt@MIl5Wdi&DL<(Z_yRKvc%J>1_B<^(A^yHS-=
z8|zl60=0*j$hDO=^&otJT4TTT1lISiTq{4md&j#Ef5$)j?Kk}Tt2Yq5UDISJ{222B
zkFOTsh4Wc?eE6CeUt!^U)GjnFef~#oq}Bbu-4FphNRGtLiVB?_>)J<q!r0ea=!sU?
zL5H7@J;~Ma83(1YN7y6iy#cPS>)C$cY3-Lcj9-}k#z;zr5*J*{%Ae4<X7Sz<$Q4x5
zuFaNGBFux`R9#cib{;9WE|6o+?#ZP^B9SP<L?Eybq#>f@0S8IfQi?OVDnGn?&%-=(
zKTXuGf2ryKQko~`cHnqCa(8#+cz0x37N#^I*<lqzKdTkj!L;fH8;lqyFr(;ou9Z?a
zUoIThOj;Jc`s!CKxpIEKptWJ8xubEbs1;OPmcWPX7cxj71>dWxR$g9CERuMb4lHSA
zNef>d-(V?n$s0u*n>I>qY`LN_VqI@o#C7gGK)SAh{fO#1a@HEEm0Xmkr}zBr-~500
z?bmPk`t>7~Le<jWVss!PWMux4gG5re=hdqNaSBM>K(?DR6~OP4^>!jNw`9PKNMw<S
zXh4b(`}=A88V+3~d}u8B4j-gDKg&<?tk9X6V;)^wkq22T+$^nORN+7W81UGq^-s>^
zKQ{e=t^vxYyOwXxWbYCi8b%&lpLAnRGjWcz+9)+MEFP`zlcp|UaeY1qFd(>7cyBQk
zGeT$0NspD>TcG*##Yj=(`Q^gpvXXOlre+LSX%K?9{?<aZOiGDqn*5xV;JRD5tE&Vr
z!nW4PtrBWO+O^#=?B}hwfUetxH;-?)n<u8PUa^|-wl*%RoNDFm4{td=KQrvBC0yq~
z43T-BnflHag0q;<oAEF${MpxEaFB&#II!I(@)US?eqqanbFF<nUt8zmFahvj)6ZTY
zcJClRZd}K&l*0LZ;dj6P1DDH>ynFv|`0cNL!`H9Bq=ibh4dkAfj@-XGp!ULdfB1j#
z)ti68ufF;XX?aVjD;BEXjlZW*)2>Q$B=O>Q3KKyRd&fiQIZpo*hYvjq=?vW6ZB|^T
ztYOS)B8Groyd|<(>mk5kG+yHnpo8$&b^g<J{PFnXpUHoE)n9D=5)a1f=kcdMJ_A^v
z$3OTGvrciQ??>0(>7tclUU?WgR<%`HYhBpgLkxeS@r;A@bY-_lw+)`s<UodWi}N6R
zVBTUR1g|DF_0|&!0ZEb5b|F~ci+LtR*L!Nczg~N5j>$C*7c+MG(w`+tP)cWY1R+fe
zcXyAREOtGND@(vSeVg8{U^Z-SM_$#BosQnfI1$Z{7YTux$RW-g(~Qi?{e@HPjDcPm
z8=u4M`d)2p{-b5@WA@cY2co%%tZBVz3AqKGI@Z{Kx9Li4Pat0rgD31E_8?Vs4_b#%
zs~>U>T#cr?0ddgp^cC!KkOBD#^RC|)RqKIqPpJ4PXZs0g#?Q{k0b*V2SN*++-MphQ
z6Qu{{ABE;W8`j^mgBt^K__WGIx9fOJ4a0+)+8Q;Cb?m^5s$tD~(lM@&Uq{EaZf20t
z;nL`>pq*|Xq7tSA!4(=>fZXqU5ZAX^*Oh6WNz=q7et&JD6Crj{dG^=1CP#4nhS^}4
zj6~_Wx0oueHdGT5;o<&)Fwd^{Ww0C$lv>EG^tC+}?@f|FCSC_tyiHXZ>Q7TlJRXn4
z5J(cS!vryqca~VU<Gy+C68^vJ{cDpX%aNvuJ!)ni5t&(+a}LlAO!xHe?2V*wMacjE
zGx&iNa)p$=%ygq0=Teu<j0kr(Q{@L$Gxx}<0?<9g5``%P@>Er3WQ4o9shR4lufDpK
z`xwp+a?R8ZC5=sFTt^pxZn}Ee9RvfR=j46wDU<N<dLnD3=^53EAS`-eS+8_mtsAzc
zX*lwmZHeTN;Y-2T*jD{!o!h_DXPof8sgCdU`yQPKWlI3F&4FLTAI7K#@HVdQGcT4N
zlSl;m8++_a46ncF5ecQ#vBuag6=Q9cDitQhnmV6QH3)a%G7j3+c*9RVqKLy8B|%O~
zO3EQSLqtl2>`hx%GptoklCWGj9$eIUd2(>2vk*?F6Vr4cXD{tm)p3RmCYEg1(oI|>
zMdEB*Z5DG-rowbQaF>i$8h0z`y0Xejn^({5dz(}~StAyEMcPh@l7+kJz_0GVXOhB{
z3fBy4bm~;I&6fMS84g@o-=hMk3VZkWPT?bXINHx#*MC`~*#nH)D)I82oC<0Sm&@--
z$L~0nH!%-}<P}MkUam9Fy*>L&!dr<8l_*Kxvxf%*V|QBiGg$_uOvW|89D}vK^G|M<
zszvuS{0!crSKmK&zcT*%S(Dse*kk5iV)eD!a!)7?-+G*>#-feIsxys6Y1EOBHR*as
zgNyH@+mNJqZDLC9Q=5d~c@idE$X!eLqY`DY*DH1P?tq8sKrT?1XoezQu68<|I2<NQ
ziH@WiXNelxY+RRj(N(r}(^-_RA!*o1l%*t0rC@hQ1mmtYY+YFeBsnL#^FjM^?(st}
z@fUnDCBva)9!^Joc>5g(D@-=Arn-|D@P<fL-IvKQlJomvu}|EzL<k3@)=EkXm&?rM
zdgXeZ(N-Nq8`o>34EL`lY-zN%()7x@p3(Hg+MY05aLCv@Iri#?aSa>78AO8QO*uO2
zxp#S{SF|wn>2&8;Gv;iUAhYi84Ibav$NVMM9q!@#Y>kh`TfVU4EvD#u_Lq+Dey1AO
zyD_=HE<64<{%zwMfc0(sS;tnN@0)xwTntj6ig8{SPRYwxR2!|X%*&O!t{m^2RUs)H
z)!hcC(LE(XVv<dyoH!f`Vj5z}UE9fns(d(-t9K*8T&Ousg)|lZ@t=OjcQWzc|GWR5
zTqd5MFP?c-7l5^DSZZT)^zAW4wyJmTb6}#i#`Qdtrc99oKYaf;e1FLNPyhTcJU^eK
z=HF1QUars*T364kT7VqMyn9z9RwVa)s+UiA0Bamm;#c{B*LSaYbN2@B4`>#C_wgM|
zU0F@z%e!@GQ&Q<n51*e}9Y7OZ!-*VF10qo_gqKR>$X>P9T?)pZ;SQiEOZyp?4}5p`
zcP_*`8L7B7gNO^sY6lS`2byFZj=MX+7C~<!S?7lM>`UrmxDuKik##W%_MQjKUU=#3
zuuh{_^HPsy-T_m4OBMD3c|SzsSRY-`KJSD>cK}}gbc+y`Os&eYu1u0iU(q4Ex8vxA
zK$8Hk#JH{tv#ieK(T3_8ojH}`;S|}uc(XA@;e-NBT=;k5z?2M`gegPGh8!|vSSTzs
zenBA_IVUccGe7<O6aVf1;eX)a;eq%6%RgZxGz;1}{E=1edw~YXBf8P<x4@_CpKZ^k
zHD#U`4tGbUY2qLL;lE<m#y|b|jv~qrzdG{a<2#;zendhGqnGLRV%-rs3*?-cN+zcY
zDoi=?`gG5K{pRmDOcPTnyqhmP*M-aV$pew*okLqVz#9Xu>j1eH1fgNqh0&!4kzvOg
zAXAdfd2jJ0qbv~aUnTzCKb-jZ)0xZ1N6y!Ku!-~fdzSS>FmpgvyiGV>#)Ro;xordx
zX#`-G#%vF@4BddI+yOb;mPiNtG4WgI1c`8jTa=t7NgFkgPXR3dUcai(1j&7=^y~V<
zYsRjZGAWm@F5vAW8D#wwz^Y<A%~uXq(3AJ=Qmw9!XhzAWu%|xH{T+K!Dh<24E2jv0
zGgE;>fpkca;TQeL+!~M9ifMzoa5Q+83cvc*ub7TUetf(l3fH=FD48FA_<@Iq2M*KZ
zjY`&NNlywOs{_{>Yp9o|<v3q%156R&;ms@VUcaV@l2_%N?>H<A`Ff?jpP5%@CR*wM
zQRuhzeK03KPpy#<-aH)m+qd8IfByIXCs-9-8vpBupSWC}Sj<>dxwh)!@hNqVl%KgG
z-mMltOVyOMHHs7pg~b+Bd#`$DwK!PiEXamnsX&?>z{-i!=@skp#P9z3ANl*={Exi8
zf1p)olI7`{RL}nGB4F<4Z4aMIi13qkP;-DF+Odn?<#}f^B}%iBwE>n6^JAR0{`cf1
zz-?`G@dz2EOxv^C1JPz%>6455?7%5LXTAHM_P-_HPrp)I-h)?NBCJnEGvjf+a*&BC
zm6|GRT8186krNyL@^)A>IuZ(n1F!Ncuu9S^Wy+K(Lo$~%JWQ<Ccz1r!ysW%?_Yvws
zHhBB$HOF$`m-+XAuwopK2fly%3UYx|9Js8_wXd47&aP9d$;LSsM>rzcLD}3p33}Px
zx~`;>nC>Q~?;bdapo_x!nddsAG*(@<b1vS~lAz64)=DWc9g~p{!mqzS^4nM6^MCy9
ze@B}e>!tGJ&;P{x^GD|Djax42EKk?vjCL0zH`$Za(S#;8Snux&X3F(?rd9(xu&x*1
z(}EGFk%&cB%0c;?-`ugBE6-+}FYkDMeqy@+8RMCzkIwGS_h|DovHNJiHlML0rsNuS
zLPC1|Cx(|BJD`{(k+KBq%1Bzb#NX~uiRahE$p8icNLu*zp`Dldw8&s!?RA6qcKp%@
z_|n+>2!4j8kc!BzG4^fz)sJrg*0=HJ9WucG@$c>|8?J%1*2>&gY*jDY6Y(w|!JdkS
z(jyWXN-Nd_6+dc0kfaWFs78k&gXI7A_{to-3eSCBXU^9P(i*LJ_dzKI#RK5lbm*(7
zD!pcY?}&MDwrOj4rZ6!bjy$}2#ae_l!|8O-oeB4K@lGmB4c2_yK$2}&zRe?y&kGS<
z5gtwxugi(=k8gN5z2eo~D;{;>x~|M^W!B1SN~@kJ>RtY%3*PQnCOvP_N5TXCMlW^q
ztZzy{h$j01w&goY7Eq{d<>A!>uirkvTA63>IJ2%Zwau8V(LB-}=kZi~CRp&QApA_D
zpRs*jy6(I9=MVpN=OGIY_?fx5@Zc18$H%+_ll9D0EU2Erf=ztVM&^6>Aa0lVi@$t%
z)LL1Vh3RmLj+I}@cwxsq;P3yc+E}9_kF|!@>W)RqqXeY~C*EE?N*Vflg&Q)-gd}re
zQ5vZ!TEm$OCv3$TnwT-qGmnpOoo5cmBa(ahLt`qLH?QAN^2EA&TW@W3V<KuB*+1K@
zo6e95LeseaoDwBZq~fIl53gS#t#Mf^=V@|wLh`6?2rzf1Qp|zRp&%)F7qbj+UfuKh
zc*p(SX#nI)UAfLPYYk0~=p5Qi_f3UHxn95SP0PkhceGn}zlP+2d`=M%#?47D!85n<
zQsC}RxJ;FcEG)}2+6t>)sBPI;S3`f}_DK@DUzXx0Nk9@UCZl(#Ol+Xl_7-l?DK0P>
z(S|KFgA=1Otg7pbNtY`e@3&)=bS*{pAN1KJ{^BoR8otk~HgY};VD4)g^70=q&*WBF
zn>V?r>*@hb3Sh&cQ{nDR?g87z=MavZ=;f;gBR6MsnJIV4gVB#vXG2#{QkCdp+ItU>
zgp?CG3y0*_;GH<Dhu@*I)0xa_Ey~z-07qx)#`&VE<dm4E3C+-oQI1FCvT&ZyC~!I+
zNV#Asp~1QvvEH`Id<v3Kire3t*Z17r9Z+?Bt@*NWzFxR4Gt0WLSo6}GU{?u!TZ?G0
zN1dVhlT+GtDGeVZ4m%gu*D+6>t<b?D0dg@8hlw|DUX%06^Ybq}K7B_n!m_$8i+EDU
z#&7M`NX%hBi`A(axQv4*0J0MPOR&?~*nRXJt?zq7vn({~qGYRcWO@fo|LDG##eUu2
z$ojb);J0n7?lY_Vk6OalY|DEeQC-&4lb+_?rWdWO&2^+&TM&dI0@gd_DzXOy*)ekC
zyt?NEG*T|)<Qk0{%^8w0SmpD)@czRG?y_)7QL2^_BaM^YSF6^@xGp7;FxAkQbRhQ@
zz369K|5w*B?cGOJmFaY3y1V0}HZHaD^V6B9>y_(#rCId`Fue=(cw-riO^dLg6v!p<
z`r)4A@xZ*UoUb#_=PMttXU@yj8zF_~s&US#X*3{k8`8YZWmwDjUms@9<f_XGcq56D
zGo#iJ1FHy?d}!ol0eNDr3+pnW`2w<d!=>0m#_jDAIY#Id#{RAXEpJsIz4=l9o&I-+
zvb%2`Wu)=*JeYh8|HC`jy2dZvzWWg|VzqJiG4J|su;VRi*}uA}SMMD12}$3ejsFtH
zx2ODV{MpAie7X~|D-&O9Wl5FGx<XZKRnKZpdFU|prrRC9`J}?x;6Mpvqq&BPv=KmJ
zs|QpyPFpqr>pWk{&%#MP(>NU*z-qXTO<mjY%3Eu6LCU`QZ_e=^z%oP%r{f)O-+sqa
zUC_32cmF^u6R)f?U$0!}$H)j-up)L2$<CkM0=&?M0!k8IKb-jf@QS~E_@2}0fz$DU
z5AS~DGGDl~=)%%MAUc4n6rV?AK-f7hd-28;9QzifjjoLm*JC=eX5Fz)$tWpdC`o7D
zzWIT_{p}mx{U>LU%=49bxw6&^h*DD8ZgMP?&HzcCkxdM0e+Q_#&Q*_{+^4l&sY1L}
ztdfy6ieK;`#HhMp(o3G8*W^P0m8ok%*ybRN(BQ?|?!SBE|9>t>b`kvR)p$4@$>}R|
zmT##nH+NWTWs#NZx<YkLsVNt7&O7<z7EY85t(P7U_HxZBO{6RwN`_j|<{E2F8dBVL
zSG?n(nQ^^du;t3-a^`e8A;mL`ORJm?cYOcm6;&NHT-OC{-JT6?)fqyPhlACfu9VW&
z3R}EvBIiPxCeoxNfj8g3p{$Lk=L^%bmlRq`v2TUj9_w!aWX1r&OZLzC{_QK?9A5Eo
zykn`2d2#USd^xjFS<RTE1X%G-Q!<ufs6a^HSlzKt<6nK>><hEeyC4eU45$!g9#{m<
zmbt%&$3sKXm3euju9+rhs?NT*BAy(;8Jzu>M}h$dqcszng((E5`(C$i>mFsl+EGIp
zd(dcjq}~K+T%-hm)!8w9PIe^<LwJ}TynF$?`l2283+%t=HmlJpe#(4(cD(yb90i}8
z$!n_2wbE*(t}Ca*0ZA#&gYC8ZGr1Rjdr7)*FZW?fc!yc7tm{lmjr)T`mx{V*d~@hy
z%Ed+CAxc8LWqm1$V-Zpkv?;5$U~_K?hu)AOiL0PUfLBODk_A}m-F0&*9F9j?HrA{h
zUY`(aJe?nzjwkLOUXx4lQY%SlQY!8<^s^@38nW9<&WZ28d&R?aWT`99*9(u2Pdr|p
zd0wtu+scZu4zMp6V=uZ50$#EyE&`t|0IZJE#@?b*9l+|sM8QyzSnDE2IVi6l4t)RJ
z10Q}|`1trQy#MfuToUtgX00<R<*^P1<9Mu*jX7%~4q)N-()!akFB4DUxF`;67!eG!
zfmz8iy`!%L>%}{#D$Tt6*k0iI#%5!ib3i)4NZXkv+jGIbWa*!duhN#c7=GNVO888%
z9nM;rTV-wyt1H?Trku$VQWo(XI#YM>&MlgdRJX~8baK}^DJi+Hy6TyxtJ|QuyhL5E
zESI18_5D5f_xBVMCD_On&?Gc5*5+q<5gBX0nS+SiiM4xvyY=A7rHpzz+vu3B93JjC
zyt?Q4`jHQpGynX{Pb{@^T~;a;Ie^algj5GV!V=ZNnCSrNn0Wi<HOF-1X_<L{e&UzM
zN8Vqad0MXCF;~P{-Xfl)QHyf+IQPO-t7)8aLJqCoes^XKZI$J^ayY=<{VPf-?7R4)
z6=9_u8ns*?y=SfmF4qEOLDE81MfX|U@M#^aZ#T!8iZkcM90*Tl(b;@kR2~C(EYR(k
z0I|GOQplF^4vUezM<{W0cYVqn`)==8t*X1EZ3nMB-uRLo+ZXNl77R1-_-tKQC_>8L
zppE|$#<!>ZZT#8BXjBmyzQ(77Wu?)q(R5`>6XkRW2UKWO7m1V}93?tk)R{Ainm6=K
zQKIB{M~Fh>!mBwIkj7e9tSVdT6P9`5ngrH`IZxDD5nxTA!RJpoA(A%%L!f8)!Glm|
zY((&wgolSC)2YzJNcR&5$+X8u>QcEs-IGbI`6<dSJfqdDAyPeHZk=@Chn*0CK=wmq
z%8<=7!7A`HU->XEynFh{$IF@4R?LhVAYhXyp}<S0#G>0$cst45k?z4_4*in3CC@Hd
zh*1h8Nk}Sw7>5{hlISvHh85*7WlB0SFISec@yokkxGWz>hPHLRuW_J*g_^CHVB+0r
zG*G?HayGRPX|w^TjVw=3T!|MS_E=c8JJ9}7-T_kwpqj)3BWqrY)kQ7$dSM+%v?zDH
zNhW<Z;_csOWWNw^0B~s)d-I)_n|%f2b9UVC*fZ-c_Swvt5r<MZPHx9Sa<I^pCLW|&
z@*qs;+Had9Q{)iqtb^B{AtWWLRe%kg2^Pbu@bU2}N@_y)s(5pSBo0)y)k}0y4d*Te
zU>o~M=<oUW{XK}l1_8<5Fd$7CJ7g5)dcE>|nR);2XO?y0ba&4j+8Bzrlh+hL@eX4F
zcXFpBp3(LcG_zG6uU9@iKlAZ=;j%0&sf*=CIdoTT8IGj|h!lN@w0Udv)B}MiZ3s#~
zR<WX9W#k(Q`?Z_1a5{Py)_JZ7#&JqKOQq_{d3jIMIa;dwzSzqFrGr=l5s7U1VAA$p
zNk&pgqOm8j1J5>4bChQfcylWw*9fx6;jH(0?!$clj=27YVXy(Ff2tjS`kGoR^K9Io
z-h5R|%FUKLTV})U`sinJwMNdFyVFF8&b)P*L%3N7g0tz6c(*=hK|^j1MokH%CZq{S
zCDE`dBr`9YX}+dgtIYGv`}ZHbEDsy!DlnR*<dUi9+3or0%tz6*u{zDca>4fi4NynK
z%ksSxGIKm0u;av%gm=IHg?AsH_;~)vR2NqDWP_!IdH?yrkdD~U*#_0i;SU84Q?QUz
z;wr+&w(|4!nWuGSX>IsF8g!2y*tH123`iD?WbFoV3!Hkem@$R(0OCL$zxkPYnRh?$
z?2Q%FNF?q~2k!3*5BG%+vT&KN{PN)+Nm+*P)TIwRqz}NVQ=a3Dk2KYYD$O>4K83xO
zn3rC>yw@Xh!FC2?$3Yk0;j2~Dl$7H)(|$n)L&w@+WCdJ_?=M?|;E!4Em&Vs>%f^nG
zj<c^UK`{_Dsx($@oOR_;gfg9?+g_qxdna1AW4E({n7BVqV@d*+Ia&7PCq|uJgk7?i
zd?$6r&ANEPH(BNI-H{)@`<`W<xn3?*ZKOzKfdjo$DNKh0?R;@9sO-tv$ZI*fQhk#u
zsUVU@vw$KQvqUwalq1KZ(o*JX%FplK^ULE${^@uBOfH#nJkZ)Yz%gJmV1mWfAmxck
z5?LFmC34GDgJ0UhyLIO0<;rubEb8k{qlRCAlpQ!WwUOvd19{1a3R!!V26Uce65}vs
ze*N9AY5I|6`QZME8HtIUP8decN2V#$4vFaiYAe^dQtbn<&{p50#HxQ}*J>3{ycvdR
zf@IOSAV6g?QZiUu0N7~&gq?mqlGbedP%<b{&Tl58bbqdurqvm5BDnVHz9ewlYZ!6`
z|2cMSH_x=T%CZ=CC&9%3HvZbjHvsF~__K|H`tRe4-R{&y!@F3cWW4*4O<>8}$^zbD
z13Ej$AsuG`Sg>~daBv6KH|9h#KXkK1*Rc>|AD+APk}jdN>g|8EAx%Ooxobap$<NkF
zG492-Owm@`?cY3{Ci0ZL_I%2ulAYpOE7NI0OB+<^QLApci%f54vD5pL2qlTLH@4Z~
zS{Kes<>_+aGGD!Sh=ArCoT{zSZVFnGx9=9PHZ*xVP}L=ud*_$_*akqjlVo-+2JgU#
zHCk%qoXAqBb!A?ar{|B%^-On@Rd*lQgVWB)uwdw<w27_m*G<fbGda3W*AAfyB(r~a
z`;=XwQ!q33>^a8azBxC&oUE5^U^bq`epr7-eNpspj;ji1y(u(*Hebsj`#C#?ogC2!
zMnYpL3)3`#Wq+3DVCoQU--!m(^~bucg@}6sG}c*;W(-v~$4?@VQFHmy^W{pO0su{k
zL1#pV2zv+0dih${SdpC`L%-vcHn60Z{7Fub6PAo-!n{^4^OdLPCz>hK;lQ#ik%3Gb
z!!Z~^duD^-`$`rW`{1ID=XK`8^}==Wpw|^KHZ3=#2QC&)A@2B-Jjm;UtQ`ajpl)jx
zQQ3%>y+}aDa~5$iToGYeJiskQKw5NcJfpT^BIL;Q3Oj(f+eWau$UE&T>r6cB#iaD|
zaCu=**e$5#Pja^#c0jI|0PnNNeScOB?Vhml#HN2X!sqP+_sp@++i|yhZ{5Duo>=%g
z>=?H5syB-KL;+`<$zHBna$(8`Z>S=I$u-Knf^U^9?A*`fUh?fAbn#?M1bP@`6Hi>I
z<F2iFvxkHOj_bOzo}VEn7X?TXT5>5^v2L4H8`>Jt*K+%x^s*$~$)%+@U4k`e9&{F&
zAZMz;`wx$Nczp7bBXvKcYGc%<?$+mir8YHMb&$OzKc5lr_Is|C$7OaVw5peQ;gfIJ
zj<K{~i8@%reKm4p2f>z6SZ3D>-F_PTs5v_&2}~t3O_}4NfLvMY!t>?53tL;PdF|!<
z&_%!p%XXaqG5~)APzR8^yJBH?J^6#!*n|3<4IhHdq<k&*hB|Q%9mPOKY1DS^?&svi
zf%K*Eg)_N-!~HV8I$MU=6a5j!*}Bbjx8YSAB`3;Za_u7psM}5~4#)`R=QW<p$>DEK
z>bgxyJsD)x;NZzr@MJ`*ZVO_KTnhL1_k8^Dfq9<8e)I5+oz0vmIWadyYmG;4gYf6f
zK2Inn@&^8hCw*FT-P4i_WjdfqSWJ0*e&*xj6Yt-D;B<fDA#YU?FP4_ip1Cch<UVgw
zA*q0R4Dei69_z}}x-x5Xu-SI|NI!eTP}u}A(m3<O$51vhV{a}bFqOpjZ{BdZE?h4j
zp9n;EzyisTl--|>Qfed0ysu}l8RXzTOz)CvxL9rXbHYd0Uiyn)Z=Yu~mse5g>win?
z4VHA+p})g+=4W3rn80cdV0n_QKS&4uN7Joti~a%sdOQB~o8xn|+i|S1Z{x3jd;_q)
zjX&$yLlgkTf>~3mJ0*P=45yeB<!Q}>1b1$t6r}@Mn74B7lqmDyu9$JGxTBqfW1g7E
z97VjVlFdx}GLZ`B#|!l`^EziFH-1^3c(^<A`u@b_+L-4#I?5U}7je%Nk|m%ZDUrpU
z_0|&amS-#p)BQd3Q8<eqOp7W%KK{bT=V#<xSX<+nOlwA6q1h3&g4!Kw#%gB=(Sm;f
zMzxA)<sgNmiFEar$<KEy@1AFVdAhKy3$^tceY_=j7myLKEOb~<oyX+P#R%VS>(FI)
z;=5RDAt7Y*_PV_vlD7uFk`mlyIF%!Z@<1-XL%H(LKmDGjGw6hohj@FE1xd$|0d#R;
z8vsOC8%dG53H}q+8s2%*U++P@|4uPKJX-{;#5JlJmvxP7@4-9QOc~V{8W?`K*ShTH
z!$%5%=~nYN$i1AwwVrx+pKZl&3t#ef_BLOxOeu5vifGO+?Kt5g`mK2>pCzzlh^&|-
zniObY&4FkQCn+A0Yge>}^ezPkeT@ko=q4#>$=FmODMgP|1z9Rqy{t2r!f`5me177*
zUb!>FSo!_?AE|doPK0Q&RqE>CQaaoPc$7)z+0nQGzd2E)r4Yw8S`yD7Pnl&Bet!Rf
z$IFG^|Nb41moq=V`;l@ya=5>zwTjl2i324SmUi7%h5N-UwNVzJWw6ZrY_w_Rd41%=
z`NF%)%-WRZTTVE}&OYn&AEXeUGZfStTlh7*exJZ#XZdIdPY4OqM6#6<8h1)C9jbDD
zy7Khyk#|4-#FR6qhXcnurM8MZcYr8GU~(Sso~Imu^0r{|4fwNZ2X&TJGbNEI5`hK|
zX7moKn++P!&;YgP^@^HOatgGrb92)OQk+p|n1wLE*FVSmbua*=GkN9KjyvGkf!mZi
zP%B)kvm5?CXC9(!-q$wzRCM8PYaRe;Zdk5>+p^_Ct6pZ))Iklg5`luzmP%a|J^DV7
zv>-O2_-iaHO%tsdOFmIcAx#seGt>Nmim&qzmkaB<a`(e4rsIJ_QmPpLJip^t>Bw7o
z0IjeVrlVp9P`#k0EVYtFIOassiKH2A4pvT6=3sEDnTr}vE97#A9E5ADTrV>}{^Rd>
ze!jxvg54+DG$XB&RVk{xSd0U>JxJy#nd<`AmUw(`s9boizvJoo!eedJsx-wxESx!w
z^uW(SgqEDiroG!^PO+LTN;2E<15!4GmHPYvTd(ANbRR?)yg%F#WTDl{<>z0xlefqZ
zZ%E7t3)jm67%Z>+eWitX^cqR7h6QW!6oFuBMoyc~l?q6-L~Y_WSa&vF+WsVzU~3Pk
zE4F)|hgvIUN-2J(c)+{ar!#*;%}>NxcsWyRXd!myh)sY6v&=tt+jqFJ<)Ed4$YAlF
z%EYVb>#*fL9yhkMaHH3Ffds)ei&7{|c0el7jKUgI@Ze29xH<?`fyq1HrRrte%`!O|
zDJfYClO3Jzl^RlKCY4S70bd(y5v~FkGnS?#R<8AtyL_bNdnP?I%`59Pqq&W4o4L65
zBns11$X1{>tmzQiH!4}Z6#Y;V<cT#kYMc4_-Os##dZGb!snn+nq(tV9oSuEH$FfL4
z!e2?usbRUnvtvTfA6Kd=@8>g@%S@XaNj+emZN&^Go!l43STigcrj|(7n;;05CKN^U
zidaI-Q$~eM&W+a-{9j(*^N)JuUlt=bxI>~*x1njpQlq3o;y`P6+~2!@`|-m&)^!Fn
zbi=j^hP5srO3D+7%+xYPg~`PCimc2jvn1i5im0KQDLFBf93Z%2i~DUU%Jfr|6Eq7p
zxK)ILA<>;t$$5(P*^5)BAz9E%8-0t*lv1d*2GhOEk$97<76gXdl)(!SYaoNd(mbju
zV%&W_M9OdDKVy775!AQw^^IMuun&NH7auc1<hKVlTYlamGoaRJkrl3K%T;FW!lwC#
zw3MtTIKtTp76-w07{rYDowhcr8KjH>t2G)*PLS|!hMsxebZ0g}#xyqS)Y(i#7HrD2
z1XpcbmdbT$T$h>aI>WlqG=yOz1LVJ(2Viv*_elqH;{HSBTvX5o*Vee!mCLdO=;uza
zGd9e#s-vCD_F%gH!8j!L3C_K<PG?AL8JrHQ^$fmellrwf`$C&xR#C3Z%f(AiB3+{<
z<Y*VXU4aLf5@1F@w>m58Hjx^$By7&pc)gF#2H5|w1EM7*a`F-=2`Y2yU5`GOyb*?;
z$L<?`Wwu?_=Ca4%_UCN~WgA|GnZFjn*w5MV%ZZA0w%KtwaM;MO_X+V1cCP7^QV&!*
zqtD;hA6atEjpUs-3x#ZT(Ch(7PI+Ie+il5-RV(u{vlKzw5l~8kwU;Ue{XhEfkVvCM
zZ+m{bz(+uGq6w@Rm+Q>)`NGrX%=vO*T^7ihsZ%G+pRev7{%ycvkrhq>3J6-2#=>P?
zSz3+mdIWQgEjk!j-zI>cI$L6V-HMyuZYD0?>%XF;!Ox86Y9OYxx^RAe=GDVJhtm;}
zipa`-jT>G2O2&SB)2)mS6YbXw3?`(#yiUVguSU6)y)f$Cmv()_4w#M4*gn5kaop|=
z<7#feEC~>GKa*eB27DIH%h6H@6@N8jkKc{4%}Spz+pVT?CI?ediBd2OtkEn*_uBsE
zg;7Qtwr1kYMURX@G9)U4(<899#>JGhR;+l*ix^z$%&M^$poq4SQ1H%OzP}{rZw{*-
z?cuYZ5y?VV%^Z-HDYLefxw$=GE;Aa85Un*z>Mee6o42^70b==C=iukMdUEAynYk=0
zYjl0<=kWk&yPv{lQN7I8`g+8|Mz*;f&k4htgU=}#W}U^#I9Cj<E85z2MoDL*8rB)I
z7+4zDG3=xNii|3u8^2(8CAZ~2QHIYb_uDVLZ6Diyx9>+A%cdK;M8MoNul8C<pL7iz
ziI7jnOIsHD{BM3t7WryydHB39V|&{b7zo^GtD0;T2z{c=bxV5}@Y))Aiq6nuLHN81
z(OFWmS3VG|r9DF*SFBx4y$m-rs#Y=P^-7zBobM59Nb?f5l%OqJLzQul;-9DH$!_9)
z#+QYB2br}sX02T3E6ciW3~UV$B?XIXcNX{(b-%qyl=7<I>$)1NR<3hpUF*i;f^ly6
z3hf<r*;P1rqSB2eZR|#N>pcFQlksNC{F2jFkYI$aYj{%HSg#9hRYVHYlsT#J6l|&e
zk{@h%^YpS<p|c^q6JjEZM;Xo<1B-3CGpYK~IIbre=yWC*M5$!#R}wzTsNfh>v#3Fg
z{cLy2_&s)w+na?=KM_7NKJ(qaRrsgT%u8FRucW%++xX8L-vF#{<Ig(!men;Ldi#0}
zr?n&fUN4Qe?53h(tXy4Jr#4n9XPYr`kmp{G99$;S4^b(oB%0*m2zn_Ih%Qd2p9*Dl
z&6W<lm>SlMb~vDi1C$AqG_nG<*5NGqVd)tb2>^+He{mq>=|q|m>HfsJ&b)uV@bUdK
z&(9am&u5nT%Jn*vD4eF(E>^C_B`K>YM#W@zD2{`I?w0!;8CbbvR^jq=;rTjqt*f^r
zkIsW80tcVhQ4Y1o9J*seIOq+896yM1Ok9N1I?!fD7G=s_E>h$Er$gc2{q6T?Gv-;T
z^UTjb{>0^Srmiz<J9}A&MEMg$hnC1nM{sQzt9l8jcS>`{UNU#qyG!hAYV)wY+_B$T
zDD8+ra*dKyJh-;i9g)NFG=c<)vbB<&sj<}f{+>$Gm6xQ^OTzps#Dd3ZzhMny><u8I
z0q?B!^`wbj+Oe92XlZ|zf7p)I6s<y|F)tU837D8HQ}A*g!Ds<2Nw8+DuY;r<$`Pqb
zK09a=@rRQms|VOkjY?)&JsT*%-SLh?DQI;@C0aczE5(CT*h<pkJtdq-HA6}vPX`u*
z%Vp-54<C5{_{6*CN0xOB`!Q(E#bC+))k`J`H?yw`wZSlzB(#)i3Ll>@T&^>hb#Wal
z^Ck)kP{Jt;{=N|`EC34zf|Tems@vxVNYw4TJG|auO36|7)fGZY5q|Ziurl+z-~Eoi
z|M&ln)5D4D^^x=WjIsn6?tzbUFdc?)<*z*u*>{kPN*ePp0)br#@0fV-bp+U8#N9f%
ze*ONvoHM;s=z!USHPWKP)jr6RBxGE~c>AU&LVPZFVf<(IOzUedoc){<Q~G-B7(TV-
z7Gll=TL|Ru%eHJ-%@~!sUce4WK6v>}6_Renm}0(V7AfSZKuMI+C|NM8E^j&N7%jA&
z&Rpx2`zfO(b5X%gM;=ah<Z8|a64%su_k2c8Ih>BswQ{#-CkL<=t%wRq3l*ga98O2_
z-I2?Of8>|*NB;5Yk#%)I?NAM#D`rQM6{M920GGiIN#@S8nwS2r&9yTZGv=mzxL#S7
z)%S-eSyiuCBIQiif3aTk)n_08*#tz(HfJLaz;*j7$-!@Tqg=1ANky}3U`d92FHn?a
zS*Y!d+9T^a2iEo4KeOLdJcI4e4mEojuxlD-3SiPZtVX%?_`P`Y0Wha9o{Gjigit%K
zwdDF|zSrmMx@Y~E^xyYAz(EWmk+^|aDWxb2?jJ2<-R>)nUfOa$y*rz%iI;Sj&%)M!
znd4L6ni~SW?3>h}th&Z|qE!4`sd}NYvgjjCSCoa5!DL`o0^NuOc_8$m3N{Ir8(Atz
zYWUf{tmkFn!}Bwb*9+I>%Di5fs`2B;GY|RK+#i37T^8o$%H5k+NXh(o{>8fq;-FZu
zz83s?O^k}yc@|54&Y5dhWtn*Y^GBZQ%;WQwYip#_fkVmMm4dB8t00xI<2@82K*4l%
z1wTjnIIoSRDj%P()C#o<t;1ERK&+XbejQw)UPux;T}UBoiiScq5-+a78gWoDWP)@Q
z)});4>RPW+);B+$`G5c4{~z4l-Sg`HJCfvJkRQQT#y--)=)MRHw>R9&RUBLzShg8S
zm5CJWV!wCQo^0Hpq!dnv;Y{2CoQZdp^oGu^1zbvsv)vPwrv6S_N8o<W`?Y#WX^y0w
z>vVZys%si@^TgQwbJM2XY~gr1aesgRZ!@ZI<Et9q0IYB0&o*{=`}PgkHZaCM8IFoD
zI+9qU+DcUyhi>t2rvPVG^N^_sK=J@75E-S@?zc<^$;-e}>e+yvHS~D$x>l~yvc9Q$
zd)r!B*Oh5<0b9>N^%MoZ*>`83$b4mSaqb98=j+1xa^dOund`E!RPPK}l8r#YgYkYS
zY#ny``}#rN^|vHSkvwDM?Yd{?y3$NTPiNcEH}CjmM&kBBu)h2E49Nb+;TU$-2m3>^
zzq<%`Q)Z@dvBrA6umV**@NqpcW#8>@Xw8Hpye|bBnr1tZ;1PHOJ<Hs?*^OrsN4V|4
zr=9`pj2753Kd!M589Kz*OTXXoIve1Vz|e5IoyivmET6>P42v+Fsq&R%iC)^Vjj6Wn
z1@Y&X?6|3-mS|R4ZK0QsBr^~PKD>0y1~BQi8EaoQ{3eNPoyydKpp?CnmYUG4>$`-*
z78+R2M6j{O$2v3FyzdR|2@;9hb3RO5YU8@joX;07*9&v2UINzv)KQ<}V9pR6wrz?A
zMuV?Z#XLi8a9vl{U{H49^?hL0f3w?x@16Ta7H@vjZDRYj>|Vb?Yh8`1uXi{WXHRNf
zgRus+Ev$7OL3F>b9_DWF7%{&L8}4PHv1N`6>aXcp+RpH~9lL)^7sT&DtTXqB{Ut@l
ze*E5N9A9BKH)C>dRyUuHZ6Z56bYKa=I{RA2=WN;ka{Ib&JD;~@-?N*othPiGjO=Pu
z))~aM-;({a190g3L<&a4VD{`lhQU!VFE_SbgAp-9TXTSNtR)%eqeQ^n8NC+3k(b?y
zm*?e-l+5$_%G2W+!8`tyoX80*noC61hTcMcEYH0ii;a0)wXrtuv{uz+?Gs(IZ+BJ!
zpPtEmTa0sfmniAJjceKXJ55ujSsJ=`eo0&K>KjjEZ41xmGfP`gX#onbX@b27=lfrF
z+q(Nd0XA=psr_pC)SvLJ*2Z=9bs+>}Bc7k6&<TpJzx~?#b2Ni!Iv4)ec7}TgMcsnj
z;lF&&j`w#M{*^>?jQ)H+J+%KK#~-xg7eDMEqFSR-X*91iFlP%k78biY3ptW6&YFy~
zAsYQioWm(6EGy&`TC+0PS_ahGn6FpnWno#?U>_>l@UD2#<vEq&XL4)Kq{Id2xBIYB
z`7rJ;sv#_icMvYFiM6a3=G7ZPnPjHZ#Nmedv^m8BJA3JVx7%3PzMbpJ(u`Fb&4fm0
zD7qKozqNJI_0QNWVVs8J>~%|!oi8x7e`VYA%o5EK8;jaC+Ro=^axTbg-<v#P*B|`;
z%QioTc;8M56AK-|IJ4c681_6^Qh&g7biV<1cC9+gw1Z=&7WwaubtTDXtDhY`mXNUX
zt~1+pXUA*4bS>;mbzG8-^P=rLIer^|wc{Ir^=<tBXYBn15mKH&{GeOa#YWF{rm5Fp
zuU1+0N|TJ{jOa=wbINyQnaE@^fH-4lZK3;=ltUcEC1py^ln_*RF;$_e@$UJVNrZV>
zIEXS!=6pW${QS(@w{MxIiQEGwBq+1Zb-7|IR)lm&*py+KSf7-~kI(%4?j7$xKJva@
zsHPMma5txECnJLB?C-g2Ro>Yl{@N(GNT5lXwNmTK$J(MKF_=__3$~2F*FE1+bkrUH
z9yp>99RUR|;prMR{&?I`>{yZFEsuNQmjRQ&>qA28!ex2l<9X(J&a`}oCBbC%K!EAS
zzj6?(XMDUINm7n1Q})sm7#?vKgA9+?XUN*8YQz!fc6|WT=-;-lc{HEfk_wX~FGbOY
zq|7I8*vk`o?du@w_j$T?;{9;#;?;hCBLFfWXDDA6bK|pP|1kfE9ZStrDp$^k9FS~a
zE5(v$;iMp`L|L&=O+)LfJDm+dnQ~$wkyL5gnCHf!93dB0Q>qyice+bT)MaH}&n)vo
z)y5HD(>W{}hz?j$+NUn@;cSRPK2GGj6F>d(BR@ZW;HP)L@OZiKga@^E#e+N5EwEGS
z0I~K{)u&4VONyq(wXMv}c)ZU39sncI>uY@|z(NNI!U65WP-lgzY$X&uXx!5RcOaH_
zLE4y!B5;2IT?{mYezR~{&z$E+QrKuOhyunwu!ZLCgqB<em_N#(%yu@obDD2>vHiz3
zu5%P%HD6mfbT3l!w&RA9Q;xFS{`yaS@xi85XfJ^1Pwn_cnNrYUy}`idWPeQn)*l=l
zAb~$(%ZAhg%~ocf$hlEOIFS)cC`TAUuW#>tt1dZ1NhDbv@Y*VR$66?*kPpS1H3(y0
zw%S4})0nR_wKYF~LOi{L;F848zpf)wgOQlQQX-Xt91E1p?|*#9@8AEz!{LM+GWnR%
zHdEIN)+*X?#-=s0>})sAdlm6!C?)hLmxZ|**C;#cTuABX?fy&-yA*7Qm5cQt(zd9*
zgI66uitd`P?oSxY0NLm|Q`QBj6lCUnQO?)O(`BZnI!c|47wccPD;sVDVc#hjSd!=^
z;2XfTi^}(J7KsE1z#%@h&!RT(cA1lFiEf>o83_u+fv;L`?bHG5Ov(jg1#JVn=^uYX
zEBu@-$C@En@b=eW%eFl~3jxraQta53#jdQnGON<m*@Mj%7Cv}MZ!#?1fgG?@okc`*
zyHaP%ts~F=00961Nkl<ZSW2WrhT^_jrhv*ws$8!#Pahw7etKqESIXo~hfJZ>#=I=t
zPX|t?J3f5)z;&KEpU<RRIGyg}STmQT3|CCRMcwgp(<x1v>Fz)&iSth%c)niw?*0v$
z1bH}6*9+_X<nPsbx`-Qar{gsi;EMxRjStrgtHQNaOnawZp(jog5@1_L?m`j-PcE0J
zlh$s}1jwhGwqk87s?`(g2%>^nK{>|WVyGx<t9<ylP!0zuu+%e`%hmNqfuY?Av6=QA
zDk`?a@pMhJfZdcx?(5K+l5&Wo#da2tO7NJcM%Tjf&k_-rVbMk@cVIBr3(0ypw|~Aj
zxPZF4J$naki+v=MQwf;~-IB|@|GqrQw*+)g<oVy5DwlQUI8Xm7v-vi@&hZVv`ZoS|
z9wQ~e_LmfByV9&O*O{fwthRW7r9?36L19xJBHpVcRFop_@HR^{b8%p84OGd+MOc@0
zWsTCJ8+S<Xpx2BR+0^|8+1Xerm0L$Ndqzcd1A3-uq9)<_{KWIq6PFK9EY~ZoE$&=I
z&?PBT<6JZ>nA(C15APLF<GthWJ#18Sc2hMOG-iu?>2dB&EqP}w1k<T!ga`8?o`b)>
z(~h?o*+vFA-dp_DnJ6L>?_O4lXR)z{)&_FKB;WeCJ+Qf50dei#CRrjtb=Dg$-i$#w
z@9vcK>-C^@<a9I`JBe_>EV7F|%i2FY-k<E*qX-&OR|bkK&Q15FzGn8GP_TMFiI#VF
z-J!zn5L9g}t-kz)uV<KT9gjxD^M~x%z%mst2NuLR_&_r+O!2vCqq9wGUYgVGOiG6(
zVlB%SB-I+Mf@IX3DAPm=k^Gbrl;TXMTBEekL9T@B<k`NqAo*vw1DLSlVB`6E<=y-D
zB!&Cwz;rx%FjudUz1^{*I>;GD)L8)*`$dbE+or}`7nbOJ3IsSV+w$qbTz_x2OLa`O
zJIOIO);kU1`$=a^be5xa<&M7oHw#JvQ*trv`$I;|Smu>3h%S4tKR&Ln?ZK*XD4!(V
z@n`pX=>kf_2K#G#|J%QQD@^J)+f=<&G4-|PKwgnT4qZ95hU;#>*ysD+$57$A&G7z7
zcD&bl^K)q(t-=2~hTF9KL0dL6XH!zCy0A9ivsDssbZ9v1ZOL)d0AIY6tg~tR9?~&Z
zHA9-B$yu8!&Nh(b097*Ma=kJyGn^(;@<1_CR^xc?_TeRaf@*eqC)gG!Wjb&=-tqSM
z8p%Q{%9+aCj3ycdOxtqpEDJTGs%x%v!h>O4)`hiqF@`Pj>wmvF2{+JpBY*b)ZQL@3
zm2SVAlW=!Bu&gD5#L#p!-&e(qjBv`Cxf;*c3(Fd;vH|t%gTrmndooR&tr1{SPGww+
ziGzmS_L0tN@4JXNU>@3B+g4#CG~P2Mrpdvp4%i89$2leMY8C?aM&K4vCc2d$cMxHB
z7hkmH{VevUX2INT)V~7ehMx&>>!IUjQN^&fnWZhEO=T?BP`rE&X&{xkEto2WgO`LS
zPnsIyY@{%P)U~vF$B=Hnt+6gErDRf0NRBo^x;UUL?qP0FYV7@(Yj8<drreMw$>IQ(
zfHgm)lejKr#aLRUw#vit4omLOx29aznX1LtEMtlJb>ipAQX5mTkecw4)a$a+27t_V
zc4G9dy4^B2AA38yRs6hE=|1=NdS7#%DE7*Q1IJukBwjKl9mZ!2AS&VZm=jA~c|JdT
z5`Ij(Y**CpF{(AZY&k&wuxX;~*ThE0!ja5N!G`YdC2ZGYAsP}vMhA_WDhJ6(5)N`4
z=dxG5bp6849v>(E*dw<HhLP~yqz1UPW7*B4_Ql?x&lU-M8-M-d8-Vp~{Mp7}!0-RP
zMRfRPn+8~+s;qToX;-fG!fJJt=sBw*x)Q2VI{?v=cUegjDH*HD0HbR4UVKv)Q1#aI
zYYmoz-G~-eVErQiR>|2xk;q7h5M@I)kfX!h0d0-Pr;j{-eB}J$iDkJ^+v252lF4O4
zQeq8dR25f`@zF3UL!{bw^6-QGf6V|(tTZYK?ZJPrDMp!0A0i`eOmTPtyX=t9rIJDk
zCdGkoPV={Zuozpc=J3&o`NctlsjN&TBP<Re7C)#zq5F<CupR6@kjXi@ewj&6^+d6I
zFh3-ZVA*dk(5lXO>5j1aRcgPF@nPF1rX-%pB{};_L2_g)gK^=`)qX<zjmYgSy^IO)
zs7&Ip?~V*fTxh(md=2BJ9ebeQ|4aWGJ7%I<qf+BN4YNiCikv(HuHH>d#i*t%ZH3i9
zjVW(IWM>6vWZr9SUP_aUyQxqP6S=rPQi@Us)yBH64q!zW%$yREb@TuB0vQt85TV6$
zzMOgY%P$-bM-GP*<@Fm@t9*QJ)Bso5+i@@~x8cB6Yep64Zi*P!Wn~Q@I$q`Mui>Wk
zUk`_JEM!K>QExCH!@m3P?)etJml)k4^{>uw6Ffkil5i}USI5I(yQsEcaqR->?%bNW
zK3_Pi9b^{SI;#!HCy6t}I@@Xo)D7Sy&WTa3u-l8ZdC*%<&e|}w2(X=fl|;NOMSB_P
z3s=@U^t2D!|F|6wmPr7UogMeJ0nwkc<u3*LU$$ija9V)TE`02|g_;tXd)KZLb%v6N
z2RYV|8g?fNs$)CeGx$usYhN>qPO`x$NE_gKzMOf!TzGr;gEM!6QCCHyTVAYuApq*=
zyeWyC3{4)09}Y(zP7nOe-FG0yg$vbY)&{kDvEAt7y8}o(fn-#*(G6M6xXdfnTra0@
zC%uH$=k(<R*}gGOrLA6Sl*sQ-YVb9Ge>yVHh1`S0oSZtUzFsrJT`sJ1<@xbSOPjgf
zuJ}E86+2kU4j48GkiT~<_5gG5BW*0oZO2L^XUZV4zx32YoYTR2YHA#hd9bfMSa*HM
zlA?XD_42IkIxRr9?fScq^5TyE5QyanPn2XzfYI&K7ykv0FWPbZeUQ(LAmk`5%E0G#
zWmTh@F<T9vDqbj+r9rcmQ^^$ZXQSfhQ!}MT*E&fCp}D$;3|f=~PRApXT`MnJ#pY;e
zAa1LxbvBQT>#nUip-9BRN<z!lAX8jABiSG%yG)$2)Rk2m-@UrWFs}8&xi!x75@1Z(
zX+~}<rmC{E%CWhCe=gn_>2h5ijPISI({2&<B+`z}e{*_^V4?M<-S%z&cr3~@-7mR2
z969DfF;B#la@Ya0;S(pw+5a-H3s28a5qsP?`Zr&Gl=}6g?4WRp^VjF1j-U6^K|!MH
zx;LgsLX;iu93$E8qqRoU=G{|MB4^hXZQ5vSB}t|fPYL&ZV@M<rAlX>J-Q53F_$kh1
z+Zfk9d!aA)*>B^oa(n}@zKuWI7~4>iq4zX?5CLpTf+VLmueLC^D~qjE@$x@0qp4>L
z#uggv55-t*_JD>1)*3Q5r}QUr{UvL>JI`2Z%*V!MUB}_SPmbo2C{rR$$pH)z?(Xh}
zW2H@5FtRB}S>3r47jqoR5^;*{b-nWYAAe6<D!+RBidTAI)yi_XXGM7b_~=?0k>PDp
z@zBw;!p&?1a>HR(s4c2kiU+2bx}s8O(ROyPpK(Ji=nmkZZ++om4Sf;^j0_jdM3rj6
zsz~1D$)8~y<ah_4DNo!Tj=a9R=VLK4jl=04$p@b6Bigo3R{d1pLUyl@&*08-bj-;C
zPOYsuWjeC|s?kABB7-Vy42#}LD7-isbKqr0k<Oy$`EqjY9M3qrpd<!MUHtGj^Kuq~
zrNv9f;nS(?p?)9Z)mfkjmu1;y+5WP}-j1=J^<u0~iOi&l-q`VKl_rW5Nam%w*1=fw
z(xeuPa@ED3p%z4xRV5gg7A@Y5^L1rqWmTi9H+S&07xMqs%Rmy-X(Atr>)nV4VW;~$
z$T>R5RbS)LkikXrf*pw<7Fk;{=2p3!pGZk~_3EDQ@4w?P9kAma&+CPcPw#J7RUoL<
z;lvI=2G2Y%YFLGQa4_gRFElb562g0(3<=Q6!ZuwqBY+x<3~JjpGQ-O{hA2sYA3DoX
z$SF~B9(pqUX9`Ty#IJw#H=I|nXXUc4oZo$LvE|Sc8}Hgb(`0)WiO!y$yJz70^ioFO
zFRk@A=%8QcHYLP==;dKMGfOFTAtP_qorG2!LZYO}&lvUKIZGzFM124t+Y<H(-NN`U
zuY3E%j>nB>L8ioczTQ9!f9d0Mwmj^aJ)UKMz?Qo}zQvmKnObylGKO3zY8;}3RCA$f
zWvvyY1!H4jt)0yUnrr}L>%2l0G#UAL<cGigp8JP~2)F#1+`WBGzB@(`+Gqj=3(&2O
zzy^deO<)PD&g95Ncs!qZTCW^VN8aSjNhjv@%7>5dx#StwuQqI4*T7Qz_*AbfOpBrl
zd74NF%XOuZP;$rw>_W=ei@*VTi}1~pF){a5Zb9#!e2Pcr)se9$tTWbDB*1c2!&@#+
zh%HRH@WcHZ+Ij(<-IqxTL`TBwb_n2Q`ze5nsW=PAnnpL+j9BROC7+N?ibO^)N8Z4u
z4s3KLmp5_ofY_POkOM5Mb#*4U6jInZy(Bu;ST7m!&Xa2}r~f!>s{ig_TKCm%5%Y@h
ze7QsujKBObc6W(2xh?nG*O}66#dM{%nbsE6ssliMO%u~>O4Ay9WM-`ul|;(k=~7ds
zsj*f=rIEBzrpjtH{DAmORqC>Et<`golkhNQ{^oE0hPSWZkfws-;88w!GIv@R8VtSE
zW<XorUua%hOPPR3l9=1&To*3QJCXJ^w$#cm??3RkURWEU-&Zl$7>T5;-*gqYs!~K~
z&9K9SWN22$n!&t5gD2@EG(UtE<)%(z${dq$Xt6(bP$j8fyUEQn1N^uwFdY+l0;E#g
z!o19wR+2R)GbRz<95Scu<-?|xF1NAs6t=%xygAUTEJ`Y*Nb+c$fkFmrtj$6C2@BA<
zms{HCiQ!2oiz00UYYsF|N9HI6?mEFWMBqy*p0Kw>ZL^mQN4cVEfLVkG(Am^%0M-{z
z)p1C@u+_M)_m3Z$=j)e#``h>{8s7k{Z{yEDr2l7!$n+n@zN@q$)ZVPoP|xrfA)3ur
zW-y#)FQfA+{fd}!t%j5YDZ4<WHmY7xX{-sB*0w-O##K3eI6FmGLg>AVw_h%oC|mTy
zbuz;*tqYSzHd0(4YHcgePmdhZ#H;&z?rlObmg5^1gY&a@a++y$S<T==c)-=YUWih^
za5ADIDaRaYQ||`U=cNO4Jp|dEXT5yMMjEQWE*^{EKw~)61OsbhI{7bcKiiXjBS4l!
zPMJfV$l|%yX_`iD{hkjo3KGM4gb-|$$o3dce~(O?scVA~fyw4v?D!Pnk?pV7pS2C(
zOM>0yWe!3f+2g(*dpU#z`!7Hd5mE<gzU1S6?)x?#%UHvjgr%-K!RfDL2#iiQwmGcZ
zuv<H(q2knh*zCBAfvG7f?s!(K4qgNUYc+35plFm4Hx`>ckdE>PQyV}uY~Yb6PQ84=
zI$$A*G95SowfSea(CJCq%Tb~X2MIB1XBnziuIqv%;dDIm`u?6{IZ{)hwKaeM3sGqq
zGhvaHwcVGOEqGv;1E5)JbEmR*v+MT}>#he_Uh2~KuF$L77C&R9MUc9GZ1Ii+Htl5^
zvJ+B`E6*wOaCZ+`d0JPVFO}=OxK@RHhMK<vf-IUYfQAjl9y7$YEu8=TuWvuQh1`0%
zV!tif{Z44|CEqVaQh?sQEG^)(v9IhL<9N=Si~Pp|R@?n|H<wAUEX?b=W6XR-V{gl!
z9e>1@dmIyiL@zbOjMWwgL9BW?FgchT)ssv#*1EE49X7EChg#DvQG^|Y`}+qDr(+1B
z`#7AAq^WSdUY(I0C{H#|lJu@w1F%F!fRj>USr(qo&y-Rqxsb03)VQ9{tS7~eQ8twv
z=g~1Mew__JkJsw9b2y}6<gH#}6^PV!rpXXO|J)fOaTs<cME_n-s&tpYL995jS_KW}
zDP3M5Luk)r5$>iFhrEvF0}_n5em9#P^f?l|yvX;=TneIsR*gMB0%Pj{9OJwkYe{-+
zx$TDi6B5DJjRQn1&f`&*<ywe+G4vy2KV7|SDTxdp>`xt?<u!b!z5@!(>vBW1`AZ%z
zFVjBrkL@ICs8#PWXce;-a}MlEZ+n)riI>(_ZJ0EgY5*Tz8r#&|j;&$Y_lQ139KdSL
zSn5gx3Yoi-c>Ve{clY<6I8q0zOUZ5<(M@(J7g{50BGE|IuQTP&&~xNNO`QQ-*YR8h
z2S=W-S7xoj8VfdWPX_Ipt(dw5#!~(C$p=QcVfX*yUx*Y$@*p8ww^fU43f5RMqjf@1
ze<aq=$lYp<yjd=B^o#qrwRy+wekK>-loKW9wl_oys@jW%&}~?DahIQ$DWp6`<CHkN
zdlI3H#ov>GQPCj`nOZOG+@v*9>P=d7_!p|q$j-@GUt-C%E1Qme!z}qeYTb_6+3}}w
z;>+>*W-spVov#<#S5n>ZZT#nrZvfV}@xRM3+ni7TcWkY(k*&7UWI>Z3uxLYj5Z1*#
zA!^Kf#>p_6zfZm3a*`nks@WM>+L-l1Ls_oE8t=0Ob)F${@#E0nNDf}*?CsX)*#%=~
zD%pfX2sS3oBkU@%p`Nd2uGb5ujk~*he)GdySQM)9&JH{+6}c>+3+ZU|-X@BdW~|ZC
zgA%N@b=#=bwz$-Qsv8J5zPj@goqgKwB=BQc$D3qxhp`tP24{PKN4>k3p+R{TQna`y
zdG>0Fpki*y$Hy~|PtW}L(=S{v&3*U)8b*e2cO?V|O0w&iq+m}-N|7Bj+U&yC&c5=(
z3`w>*VA$?|F$vag03%Bl8ET~*r_fT_$jhn?GuKd1Q}4hQhq%wXd(uh|B+?%{U*j%&
zHY=tRT)$y%Jbw7dxdRqo)!2Z?5fd5%j9YCPcC0i^n~V5Euo-MTed2u#V__j-w3cbQ
zVr{|uWd+bhqqS$MRj$-geqaQ!n>N2~1Y)rW)>c7{Tr!g!SkKo1J)0?AP0iT>aam)(
zF|PBKcaI;)Ir05>Z@EuLq`^P^{-61{T)14%)RZAdM=Xq0y=+p7m%+5E49O5<fNdat
zzak?mP6r2r`tL#K-hlRy7$aD>{(HT<rK!7{QOdY;+eucPRhSbsc?q@PWssNa%z17+
zU$4~atf$`L?~?)>|3FMK7R-}UBy_nlV(L1)>iROOD-d=rUeeVItVTj0I<)$GY%r5X
z36yCtFeQ;varWf6GeZB$nVi7rNYvSr3ELj@pD?VK4AR40JQKbcA0HpNe3f<#dt2@S
zeaCxD><`*<%TW$w+i}s`nJg?c1g%$83&k4Cy0EV6INPu5nb}sV^)>t=6pKfcQs(>b
zzvq6+G`gNoFrG?gS(ebsTWFfl)HtRSla&58!4^|m^Ruoc<8r<5{Pe`#>jzR!Jg*m?
z)(cPStdo)t%sEk;u%<Z2Ml_lR;?~qy*G8(%b+vl4kUlADP|`-964~t7eDB-J6D>-1
z7#rm1lG@Ko!X6E#<jqs|J<H;ku;gs;^X0;_tlM)<+|0Bywj)VHf>aPG)F{=hV$LK}
zm#wE&+PYHf;`>#ws&*=zzMfJq=hpzG@(#ob=KVN-4UzD}B$IP-(QQ+b6bhT4QdK_*
zUM2K=#-#qyRlofy1fWqHk5A9c@^vsb#!`&UF`gcXcykKdI5@K`sI0W){xGzV&*+*~
zQQ#Rs!mNvz^_w?FvbuPcLz<XdWo{K=1*yQ9i~3pJ%Tra2+6)WXv~_9NDe=SiKk%xQ
zQKp&+O+!;kDTpe5I{1#ehVNLy&rKRt08##EW~{ZctSk421IJPvsH4GY;v|_<KJww*
zxGaif!9>a7kG9>0Qw?l|y!u((B})1nLL^fYqr86~&uBc0&IHd^{G_q6E0^njrEdXF
zsy41%S<?)qVP!${;z=g5&vi29WrfQOPfreT6^RekVMksH3P=h6SUiT3gyjiCXsx1_
zFzdjTjo55T&ZKO0-F)dTNGcLOxop0X8m;=z6KMeb5)95`nH;pL8v12j6VoD&>poZq
z!S^xCzv6uE&ruatxY(8T)9mWE@mDy$0a)M0pKbKOqdRQ3z)}BiI5E~jbhm$P07BnP
zt-qBxS$%YeMO_~&oWT^WrJH%1`_;TuZ`GYp^7!-!lB_$5vb#!lSyKmpoSBn`4)Jy$
zer>f@7otpwk`sqgu<9xrx;E-kosJ)wo&M)=WcxkDXG*dy!{4(v{T|{bdgnn2Cm7k4
z-^hN}ZTx&Z1=~HR@v4}Y2>AN*lkuVcODZ-1#Q{nLt0@yi8}$RsGdTOX-JXDj=GM^V
z2`67L*GW;Mh@g^&UPX6+KRKs{SL*k$l^tl55Og8S+c`1at+{@Wboj{++ny!=ECp*r
z&&Je7gtGm3H+{9Xjgj;hK3>{!2a0U>iq0HR(_zOwh2HJ93p(GxFfm1qp!wsP9J!fy
z;pso4QQI1fMZ3{B*jD&<Q4-o|7VM~AMwa4x4yY<!uwMpv)ysD`UQnxzWnRdW>%2*I
zosaYN%5|Mx7j6iM_Q2kTJ%Di21<*DpHr%-}&*SG26pm!7+&+W;TYKyNL*vueZtK)=
zTOJ8m5hQhXq7MhKB+>=JYV7tb2~rxJ)W)8v(OpSg>^iAo`F7{u1Mq%<+rG~ZRC&2S
znJK27&Cy@5y=t(n#@@Wy^LYO++3}$IZr|L$beuzKv|4YU?O*u#oGtGj`#2N*_C9aR
zTP6Bu_J3Tf(ySs@Nd`;n4M#$dyETlJ#XL~DspcFP)7OW0B}_@c>X`hEo#=mRWOtf4
z<HAFwF3#<TwFhgS3>%GEOowJmHHGYA^d;Uw=^$Na;O;Kgw(`Q}wUhMpSO=o*W^(0b
zo;%HK>#TdraJ*HA@oU&OQW=G}W-r#V4cl@JM!U74m~?qD=?<cI{Yl*(>@b(;(%PDL
zxJ!Mt#9v<?c4LpZJ^A+=mc{$}^=_=m8|sy9v$411n>QPu`Es|o{k;C#otqH0>>cvf
zuf&#L-euoa`#f9s3$YMO*YMXY&bFSQi8U7>mV+>z<x^41WLrsQtO~2HthRzWfTgv%
zj+X4sGrKKm)L2TT6fYN5_i1~Hx!J~mMz?cUO=;%)dm1c6fAb+F(HgB*O0aVK?j$*r
zbMj`-vLL3hoMrgdU8LRUC4WE*pt_@Rwqp;EiT2;?uxU?54gW@<_vP))So;3BD}eA<
zm?y29d186L!|r`05pxaCWyNI#z+{UJ_UuP`1EeI*iY9^N@H@QHLHiu~q`FTn0m{XL
z?EbDv=={;ohMm`9op~k0j)?c_Ov~62x7WSc*Z$SxWeD-&d)kyOh5c>(HIHuq*0=Fz
z8(oKc$erwmFi>|hT;oW+q|TkBT?eBmtLq0FY_-!apKCPM5E6E!3{@fWAp;az)IPb4
zDUO=bl2ew8*XJ+G^2z<EdTC6{ekf(hBf!+=8|UTuN?jVS?_YCQj-<JAJ<nV(3;)~C
zk9?RH*8)n3gllyqb3w~$utpZcJ8D6hT&Q?s3>d9`5?_t&y3f<l&XGP_n?gmGcvk60
zOT10n92y6cw$nj+)|txC(CGJWASHA6<~0uw5B%!Y8~&&N_<Jt%d*<_nCKV0(gp5MT
zaNbjL@%&~?*>!7#<O-Pzr|f{!tK%ya<MDhZrNSXJM>67!siCkd<K7b?wA-rdq!1bI
zV3YK=#ugq{cj!9h+;xV0&Gjx%PJ`aBirp#thn!y-mj^%!LE*1q?CrP*M*Hx=V>1`|
zU0WS?42@)sd2!v7u5qE(*QpKWkJ$<#eVi?!Qi!;d+8Rw1@0O&@Y&vQ$LP}@2DNNIm
zoU(Vq?3A45rHJ0X-EGjskX&5YT{M6iBdb4~CdRrhT+V0SzJA5sVdDIJ=5kqhzh0QN
zQOXg`0$Hf7x!w=K0*V&dM@cMdEG>cw-1!D>OVGmBO*DY=4xBRf8tBd^V~^Yex&3Ft
zmKh$*j}ftXzz^QIqT&uYC3E-YEu|{?Fr&X*xGr;uMB_<!+ZaiT@0qDgUe2vbzPqEh
z>Ml7mB|+Ls(#CXmge1(DCG>7&BnVpHxBL3)j&#q1+varFTFKcsO-C>1wm3H;$nK)D
zK3!iDX5KrwdvEv;+wskxO>W=jFMNE?miy>%MAta{k}WqADsDR-l)3hu1yE6KK~RcL
zES50M@jjJWT}!LlHS{0T09p3aF2FQF$>uDpk`XbI#?cUL*$%AI;=asSlVZ|hUlOdU
zFZJ1ICDBN<wW8}vx?W;m_Y&c|SKmQ#J;{&j%rduVvL;N&;viDnC{k)6g}l}PeLE8|
z?5$V%-mH-hKQx>=>Hx4vXxzZC32#24;m0Ur_Yfs#b20p0S`Pv8RkFa8Gp}Erc%EIN
zpsD&9Z=KWB!63%+PdSsPX{Q_4Hy!WlZ0mxxm28!4g(OY}k=<SC{_nD}ZMG5j6brDm
z203#mllvuF$C=z(4Sk&sN!}(lw`r9ZX8Xs(<E@O)hlaew*D!K+z@)+sK(~=7>UvNH
zT-t(!-=@KgYu2JuW+V)A0TWD>RV!kRrJhNwRLv}wXezAH0bJXHQMq#Q9)@XL!VRDu
z$FBRwwRw5J3I}}e@66340T!>W0b5OJ0$B`lCI<@=Gi31{34!Hu<#K-Jbax=-L~G4U
z?vD4ABpkBvC?61Am>1WiJRT-g&C6oh8pdEonLFTHwzU<Ay_JMeuvpDG&Ka$WCTL{W
z(NyuSdqIzDgiK{7ViSMANxa<fJR8@=(2xjB79BCEq+(2yaDShvxp6)}Qr9KGv%LlP
zFHapy28s*cd+A_Up!joG*M%mYST8wKQsOACa3vyCuL+?^PsF*PLLts)=~WG}e3DnI
zh**?32jCTZFhLL~soX?!{hI!@<Hkz+?<e0!liF8P3jA&S=Z$Xw*0=Fz8<KVbL3lK`
zPDcc1W~?pB3xdv?#1Fm>TJ?ju2XOtrBY+Fmf=WY?DA_VzW{v&A;?o11KzhE!4;PCL
zMk8=EhLEWWwCu0xK#quzB7?QA?(p6nj!d~=W?a{W$N9=-U6@;?iENp#xR4jbpAgn{
zL2;##bz_+ANKhl_@Y~vUZQ1=7HYaBbQX*qA&@YSYaCXa3xf<zW|N97dh9bQ6QhwLH
z5*cE(9xJbdnbD8j@v)<8ojVqi^EmAMa*R_hoFpN~BNgL(nGq2WicO-^U@$@=`y-(N
zvboYdGihTH_NFS<OA<y_GP<Q$uk(kI_5Q+1+nx0ni+lX=vte=f^1gy`YsXtAd^pR*
z+?uhrXw`08aIpP${BOh>kpZ;rXT1=~O9|WouF~0=8%wb-{(cQ9fRKa(LRt+ik&PYK
z5+dR8i!B`LCW=WH>W{rcH|N$Ofb~F328ePv9Kn>5yupJrPuR)?Vn1=A;zBl;ckY?Z
z?)(iliS1X)XVzNZM#o0sldxaq8}PFU$lNY9|J}|k(cKTn-q0CnwbJU!TGyc)(x1|1
z5i#!)nsL%KsQeWP3Fp|YvUes-!PJ~m8jLlp3;e&_+aj2cOue_|PD3;nc7Wd4(_!dv
z*>>l9Ebo(b&L6hp&yM}3{S}OqyesQ|KbM^?L!X~7+p-@A_JTp$S4PY~ick>}i?z_R
zYYstMuqd(6O^e^^&m}&j+jLL{1~D$<j4Kal;~a=v(yp7AgBgIAkcQmd2)cUb%3@H=
zm~!Hn60fHTrJ>TutWnC!NO~CJY+&2^ju}w~$F^SgK2ASzF``tD?bwYW(FoyYImNuh
zcCoGP_@U7CD92TRIJlg}uX9R?l6D?~?Vn<I$~OPZKSar4!M%((;W|Y*C!n&lHWEm=
z`%fmpgYS!FT=eF15obs2gS~TUiY~Td{ki(<e4h%2j%|8&eFo%(i2LnX`{Im!d5quq
zlzdfMb~)_0_i@YV#ooOAho)UOaZ7rPryq7~p>1SrJ~PrV6RNBT!fH_I6%C8qvnf>=
z4%Hr9ygj9MU&GmT(5Oog^yE>q<_XE<2c#0^YiKZXw=&(F5+!8{X$XuTQsUt_Q3NIt
zN_ql7)Iw^&&!36T{N9$y*#F~d)BYr~K>s80sB{)Gu}8|V7x#1eKI?C>?T|j)Zxxah
zgyMQ|mJ#jAV+Udqa7Y<dV_C1>q++l^_gB98x5S=T@`0SZi3Milu%V)g<*uja&maez
z*{u4riLAi3%613AIB#v!&K-ZcEWq%s<8`WxPO^J1Vq<XaW@qe^yE%M0WH$qGun*nL
zujowvHvaA78-Vp~{Mm*(%MhL=JGWvCZ+G3etXGPtw~O_7jvB;?&bI2KAHoh;0Z1^h
z{FR-s)Puu)mbn8mJ#bbGv4%{=*;UoW+FepQ?!$ykMi&>dokdCNObhR9E0i?ROt{Py
zMELIY8<IA(Hr`*){P^?$^*{>0k;kfB4anXN$ouFuKPt@cx9{P)8cdlo9f1XEbvdlv
zVOx6e+udf_JI?Vpl{hqX%tOWx;;uhpi6~`k9&{sHiKFzh#B1q^4lvdfQU97SCWG7*
z39)sva13<D-JTZD&Tm5SI(f;7wk&{<4-cMAHxCF@jqi>pes#RV9`2Z}@$TajlHfjh
zyW^%r5txhj`e8T`N-ZA9lM!jddFn8RnR@9*2ky*^)%x{^Og3HW+7p0#;p6SkUIzNv
z;N#}i?&NqW<bGBE<&GOWCZ;ry)I9@8?bePvhLT+*g{5wk^?gJ<-VVbE?Yq>xt6u~#
zJy;$T{O-I6sDxl>N)&TPbGa_mLxCx~6KGzhn#9ONH>G94a`7%&xeHWBx6|g%n+DLC
zNEA&>9YIdQo44Ol&GoLH>dbY1j$n2RHd|9tE==W!=}coC7vBM_lrm274||R-V}owr
zea7t(gtO-W&frQ&uq%5x!Vt~(J;LzyuV%saG}dM#e$CxsZ$?`duIFdY=V#96D~-Dd
zax9)y-*YU2P$6T8`doE4zN9ojXhXPYq9)G=PbqO<GRs=I);5By9!Peez2CM1IlGg6
zm`X4hC1UO^r|aD#2~DwhhuHiEvt^q~{<t0UsT~hgLBA?rIfXu&gqC4K(B=!)b0$W{
z%Xcozm6ASZ%fS>Hr6#T$X>negbjI&8%5++sbTxQH4qnLsbb5f)S%zNfo&6cBMP}b9
zk|-k7rqCLl-jFTP%-I`B8WNq;1Xx-nsRJ#|gw~Xj5_iWVCxl|c!=Z3Yh5OS3YhAc5
z7k*4L*)HSpCOpAXvP)Ag9Xb|^E@)8(6-)-gN9p=;9z=E!M@Z!GN3#D+x1>^Ei7A3y
z!F`oBHopWZ4*;+yU!20xm?UwO%t;bc-z<SH7HFf90Rv?6vfqYPIOJ)tlm($A;o&gx
z<}mT`S-H&0)%>~-;^0A&#HtOol_4T9Ov%qsGiQQ1tIFr=kc%^%<Lq}9n(s?qimd~5
z*Z{s>qO|QvjL+$Re%Ua0_zu>!$=VLh`U@YWWYmnNjf_wR`**V;2pD2Dr1)2FzK}G<
zp4!4s>&YRZqrK*ITMV~>1YyJG`gJr0XRT)hcM3-R%c}`eoWsG6X`R{V?>k9BlCdUe
z&Ct4H#hG2J8Q#@%4gbytkaR2)563(Ag-s{|rA(x|2a+1O8ULI=gx}`?7%7v0Qxax1
zuG-@8?9f<%S35}yv5ZrwivL}r`a^)+d~d9fH9GN0hv)+^i>|Mu<!gZMSrSuu;E?Yq
z91)$6=G_(rXPi!&c#w&8t(>19X&Qd`iv)tZoB=Y05d7WrmXc3c^Us@1$Z8aA97;w@
z=Gs=*)XN1-nWMzptYQ?@GNu9K=kS$bFrriT&fZ0GG;;DiTWcGyY3hv^=&b0t-t;0q
z+k;qt_*mV3+H>B$nMAN6{)h#A8-ESs8-Vp~{Mp9Xv`4))n>Nj02sqGjRpFS-!K)rf
z^<bBsTOqqYW3O_bZwsbx0=@Cq-7ks<9wtW0(Pb|gO)DQiK5{r6ko(i9Q8y+)DYCi~
zskq*$n2XwqM3*I_MQwjl2fj=MDTQ=6@E{%lSA!baE3>-KS#B4Ng?5jTOgKuxtWmdc
z70JOw$XgbrGb%Rbg3n`bbB(m|`zdI}PF#wNOB~wW>FfZaJKbKQ(byc|UIy08u<DNd
zQfriG5oJ6HY4v7p2(C4fBxADTa9(HM$zWss@cJ!(_wa`Ib;Xt|S)m|2SYd6AWogV=
zkrZVH!^7RszTP3y@7bNGuH#|afwkRg*-<t+)}kZdt~ak9_Y+I~_8H#XaGb0DOq^%*
zwE$S%8`hXoiAD)v;llP3m$|K2>Y6CA`Xa{fIo`f1bpR~=mC%_TRoF_Dj1esF+?EJ_
z`@Z6KpcU3ydH?>AyN46)y9d`~%R+U3oRW(Ji0c~bb@dExth<1)@a!)j!l8I)O;O`+
zI*{{(NurvtsG-*<&UI#~DY|T1h@PfcR!SD*D6m)rqA4>NUop>O#z2ibFcZks*vG;c
zV>J&(+P1^B5j^)r*>@PpzLuA^vNol*9ux*|#=Kilt&Mq&5IvQlBi0)e&{+}mJ-&ql
z+q2LiTHBOWSANI`?oSi{Yh9`4V!V@NUQI`ys&c6dOgVzmaPW8FRR^bg2ewf<7G))U
zeRyyl49cDX-wh35dLQ)eZU2jQypL;#f9ztyd?j`a-IjN3RY@MN$_+;uHDy-UsQT2F
zL)cq(yP<R)8&K=g$E&lF1%|q1XVr{ltYdF(0mQg=jSCIeRe5~>$mw)I?<XvYV>jZ0
z!-tYU8Zxh7UdCT@MH@)<9sf{HZcjz1m<yAa%yF7H1|YVkiJA&2AD!uwynVg%pMABg
zO`SC;kZBZgoxWyJhE`_Y7>T$BPA{qHC!mb~3R|ng&1Oa!KRapkU?z2~MpfmqDogDJ
z$rT1~)or|)qKk|s7vu@alLLZupw`A-B9YiKGqMzKf)K}{m~t<fUmqTLg|Hm&`A^Sh
zRy7Ja8Y*OJ1ip;Dfnmixkkx3YBB6<tqwBNoq!P?-t!gAuT7V6GPUtL;oh^SUcD4Qd
zvn%WD&&NK?%=}9n!=;IHYB2x84~95D&66mS)NMA-UuVb2C@CHQ2<#4$c)zYy$HuT~
zJG(F#YYL!sr#!jv`J|bQFwZkfDX8S=WN4&b;x;8`S+3QSAvx2`Sk@JCAtgAZ=6lQm
zx0eQo!-S;l!sETMM<h!${5(a`<e-<~no}9?x|hs>@y%oOvp45x4kChdoEitOw4cou
zO<OGXQxjTeX^jJW)5aRCrttV3fOKA}ibmID7m)AsoBUA5uAU5UYLE&!9jGa^=x!K@
z2Ik-Qvrfs<sOF6kw9ecMyt+HlPAA$__}zz}XsuGDV4{@HbhL_4{G64tV{0j>OR6;0
z!MMt4JD>aBGS-I#xZnLay9GLLVB=4W&_4UuExHMV=-8lk=<wV4YaibLtZ(DbHqzL*
zxBqsb&XLJC&*)s^@Hqr%BvHt3K$pk3Ibg&0i?i2Dec1N5{{2rv80n6pzgsHqEYn;o
z>Ep+bJlHGb;htn&eI<h!MXu3F%MWW!u8m~kS)*!^zcC@V2t;!xm5F>f@@kq`O}Ne;
z96Yy`&T8o5ir5ym_0D+{LXi^n=o{=IDU(vBgfOzHky_8BOMp>Z2>>HVv3^&6(9`uO
zS{&5<fL8WljhZ<~hoF*U&YfG;3@_VKWm#8uo}=@UXU)3986C(-L+>iZEF@zzYeF(P
zPyF!uJO2Lduc$x&#G(tOxo+8mWv*4RrIC&el7{#!oteHn6gMtMC~Tpv>uj_L7&a;W
z&h87n)3|qc6uL&hZdw2FVY`=_U(NnG{-uwcHhOm~QJ>GWFBvhKDYGsJiIQUNSueTk
zU6I(zgy01z1s-}unc!!%gl}KF70pU^Kv!#Hsf~BPeB_lG?d^l>Qgk@PAm>7kH?>%=
zCSGcjk~ph#JYo-+kB39FgogXWi4zXuRZA?ISWS6cp1G0+?ceXFucJo7I7kH1W=IFH
zT8nk%Yb)*UcPD)pla)hI#KwxY)J9o|``<0ZTlXU((G;E7)<$hcRewI+=`{1sLbZB(
zay5`#uto^KZhAXr(IQ*jW+urFVukZAh(C|kczgGXUmXwp$M-+d+DbYU4yo{ZI<Z`@
zSY4nTLGl)ijXAJ!|2e?D!SS*l!0HG-I+Jfaa2RWlzP8w{!Iuv6@QQDz?~`E+`1WfV
zZp-nHK9}NWSZB$CgA8-~BeuLdmqqO6%rT&`dU;aX)^aS}&7S*aAbt*Nf++wjtzMG<
z@%<wYukL8a*LaCs8YO@^A3!RJGgucHcY$fEzKTjYKuTD0j&U_3mBisNanQ^}q8*N`
zDMLECU~%%sA(_yRGFX~2B{(L?qF7RvU?X@p%#4}qV&)X=Hrs^B`+es~;yH?!!1C~M
zEE555lo1J_ibazf?b=XgoEPQVx;VG9COV|3=E*NzU02RBkPDKJq_jkMtOF_X#7-I}
zQR06UMR2C|y%hfD_<$ZyG_Cw{{TH5eWeNbx_cPz?=s>h?Ah*RC(W29G>CD+!Gi%!p
z+`eA=w&%&vZu30|esPBG0Q)~>3?h#?&{>TDr2q2AfHub-ws|>yKk9mE-c(4D?9GFK
z=<;ZTwL8kt<(5_Lxn56X^n{bddw2ax|FDrF39U=75XcFp@bs?p!mLJ{!dF$}U;z}}
zAAEDb^%hIXtn13!T+=oiq+CfQ!2we9a?|7Sz;Q}kF3+^ufP{)0(W=H$tNXD@p#^{y
z;K?Zo*^C0C8iCluQZDocA<nGyvldGmAQkqj9)JLe*2|1-r}`FaUr=15&OfV}QcbB;
zRFs^v16Wujiv8NG16Zas4Tf1hLB2Y6)wWKAz7Fb0Ho4-TH#|X=wl3VCj{M#I9eV$Y
zI!*lY;YY68C{iFKSX%SVS|myfQwcj+0wiN3wK82MA@;h)rZ6(j$9^{Gh;=r$ZW{Ky
z`67Lx@BV!QUhUFbCZM}gbklDD)_)=68-Vp~{P{-J;i>xJ7>->c*awc8m>F|3rX*;}
zkr9qf5vXnWS-Xub4qhW__%~;#zo_SdZ@52ZH#!vE6}T>n)s^4<_>q$o^f$#@wKr#3
zfgc>0GpDpFD+`o@<q%0saCe+MfDKA~O2XT_SEO{{l8!72AI~#sz4CazpejN5@lAfu
zD2qHQ%`1XPQaDIv4N>BK*Pae2XbVjNw(swQI7fE?OC!USLTjKyU7o4hfnCzib&2-8
zUMA7Yb4r4#WKPFBmf6r&{re)!+HM5aw*$?<8lpBdH2To)h?;{^d7Ak3{ex?j2)rss
z{;Ti*p7-xR@bl#zXKP%|1&93rD`UO61HuqY7bNC~2(pxcScN!sXJAQ6UM9H<@L%ML
z{<N_JM83*Vb!fpvHK3P`(L4VbG!L4tYUG3qf(i~E1pv|GlZ~~(SX+IxPDkqB3miu5
zHlP{9qhQZ2VyqD*|NQZZx5tSUI1X>zulsm7ASrPuSJu{8*VVI~x^l<8XJ30CGl6_4
z+}nvsA!C$aUoGbQ*R^_C;@$C%_m^iD)nJqCm?tC9@O|NsJZS2sFUh;Dq0P&4yMq=b
zLfRed0p@fdI8()T^B9vHOs*SR{SbVyF?So>5W!5&9FKRH2;XbtV?D!V_JC)AX8t=P
zh*A%RQ}*D!1zS;2nl`Lf-rm3F-@p2f|8-r+*E5fG;hqEk{_WqQOXX+%$w6W&COZLL
zY?Na<>&ve=_2=GM8zFq}QxhPYb@1^2ch2lfnH&$IzarxfTTTX&ENsHh5)aI}Eh|?u
zaw?y-<=VG!+qO?XONijf=o$D`hWT})*p~5)EeFtH&Kg>4!y5d{Pan7!f#3eH0ke`^
z5PB*etQaJ79mb_rYE{;?xvosPV=BtA9Q?Wx9PbVs4w<}o*=tGOp?F;u=DKjLtBaoR
zuF*lHga&gt;Ud#+ccbralm$Cs6*k7H3{bhB!AftABJ>vX)u$-tEcvZ^7dAGKpf=_G
z`}dreM_NzD^!de*BIHBi^ze$ix^jQYENWij*P_<byB`ccY12Csyp-D=HX~s=alC)v
zS8snH)rD49?&N0{T{$$NSRzYdZHZ@9e{WGp`DRNph(Nqjbc)I|hfmTDcr)qez}^q(
z450q!7RLVPjILeuH9^r;_sLBEuVfd=ZYj7?>PgkJR*nXDNPxeFWoK6n0r-C7vi;fj
zcOy~7xTH-NwkHU;3hLXLGeX1GVA00&b>??JzC&N#^Y(|goKj*Up;6jz$-$sZqFIZo
z4QCO}>l#TJj}1&4sU=EH+#NI1VPbB@%l)f=dR;5?x^P}*2f=jHx>8kdDx(R=6mdw-
z%X6ItoF#{^W1|#bc3;Eqh&k#hS@3Nvw$_eo#qTFX@^E;(`*sF7p_W<d!jTH8Ii{1G
z@qIiT_}kz9hW7MK`RQkGIF*vik?h!qgZpH?Sy4i@qPmjO<j*?^Z|`66-~9SF{CvLh
z@o6SizuuF${KsYX#Dj!?0VURnjcS_Sp;mXL=-tR$(<tfRaq}bjo<ahkt&GF~L(l7T
zV>nJXuFIz{>1$h@-L+$-eH(wh;~RkWZT#7WwQb|~KSKyv1nI1y$OL)D!GUy7w%Lv*
z-QRG$(FXKzyZ`mSj)O)R+0F5ajnf<co>0}GH6Cr|GS>jP<mSzC@)q2+d5MayJ?qS1
zVofD9cccpyCr;A=BP01pQMj(gYRxki+Bg2loJkrt;WT579H?hBv-n(f7R=4;O8-t5
ze(k~LE{NL&z!E6WzQitK0s0XuF@K#M(gmjn2*~#R-YG~!{8t3mv*_h2FC9d)&|=x@
z;3K{X;7R)ja6e77>A)c+PUXPc(;Mov^2B2to(u=v`gJD|?ZJ7hz8;YbWp4+tY$yot
zCZGcVL8bq6?7#QPLL-*e7k=;S33{!CgB;i0!Q}*D$KL77%(IG(wCNDBfrVoV_s+6x
zf9;Q^yJS5BEHHxJ?V@geD|{y}d99|L=Y{K9qf}GIb$dp!kdu*frbZ_rYu*KMnhrP{
zB<*ZFbC@7kXCmaBFiGSVOw8uMPRRwL1mLAmz4m=4>@?qqnZ_JQhz*Tfl+j6QIL%{r
zLIEpuU|u?S+v|JwcsFmmePMeHHuhD?g;FMtha)A;qfC!a0IdF=ZbN;K*t(sDewiw#
z!-=;KZ+LUOV{VP}x^j@h+v6+l%3%Yfb`E*JpWz4xklgP_tbcYa5Z%32wrl;<$Ie-P
zF}|eltbOjjlCigC58VB^C-6?1fK{JsFJ-aM+VZ&N7iY5n-PenV2XGD7wM@G=__8f;
z=V4Q0F?D^pwPM+UBkSi8&bpMgrE8K1)*Pr=R_BNutDom3HD?72<dVDuUtLGb*+Smk
zvbM&m-l=qFw!!x6D?h@!zJK*ZQg=rB!O8aXE%4phpIHD2{yRMg?TLc%y8UE{cnRU<
z%u*M>uJKC0W|znkXMY#(1lkN*bjGo;w`~iy-B*<1j!chB2IMkvJop)Gk~yZ#B#DD%
z%J>Qf{q4^7VzGps8v|h8_7mIgM|Rh>+uLmX_X|e(^4ovpOcvV`hhGbXrM9IQeGk&;
zitR~Lqcdh<p-~7qi|(exa~-E+5MlbX?x)Ctm8HbF-;W#XAT66w?tf3+{;L{GZG3#X
z@TwF)lSRB_P6zxJ>))Tp*kr{0Z7@=;zQ=1NTY?r{gge#C($MD9)~KzqL<j28<O2ay
z%u=wM6FEhhEe_TRd1x~wQD-pPrU$zvoX0!*-e_seeTS;kU<CK|y+4)Wl$3&zsMatV
zl5{)(5y&}nfB%Z(QXLaE2eieu&+o6@KZ+cCmRciqKsdqaaO8(qZ@69?*Lo$JgKx*=
zoC9fI`Y8ckci}XVPi<pB>OzR~GsN3_;;H`+Y)GX^{O+H2Lhp=!|Jkzt`Q|J3z2)2Z
z>mT0$tZ(DbH#)629OMjSgRBAg0IRM;Ag}>fASbU&&#}?1yYp-x@CppN@{L|Xd{$@b
z_iXX)@J&&h3%|B-K+m;tZAxo!!1XaGil`wgzR@3!T$h>i^Am@=d&==hDMpgS{lk%@
zLTR3Y)z&zcLM{iE=2T^`7tfpyN0v4-uTY1R(BT{3h-<B2UienT4@9zqWjW`O)%CMU
z0sz=&hC<|YhzZ4?{h`p0)a^4nv%%wt9=x&x5jr^MjFhz*=cV$y4<A@APkem;$n$w+
zS^V$>XQ&85%5jQHLZl(4qjXEcd9Bh&<^r70&(D0wANVd$yeTLC!w-MYU3uUYZ#c`9
zV>*$!fURiIv)%a~?l4Ak9i+}4gV1*9WttMsj0Lz5Yy{i%HvUax{O)e?^}2;W<beOB
zjh!8L7FOyl?6V&sOLPu0TmxejFCj?7S*FKUw{{#;(ODGxBG*Zr-3LBp&31Oab6&j6
zLF4BRqAny_%A|6j3+^N_Z!ay0VwovtRF#@C&&$H4Rgj6p1cthe9uG)K9CAWZLz>gm
zn<{lxmbLLTFI;U2fogx&8{^91t5ub3-hjJAnOrZW=_QAIP`=-y1X!%vcsbnJ_k@7T
z4v6aG#XR?a3ZMZ>^x#2o8(3=N-G@ikb>`{$k+v)xk_%QFMlLy+At{1^gxLyO$@xUi
z4j^_waT4K>3wOsmUOl|#zxkWr@b>Z={imP!uDs!Odf-k@OiZi+UU-?JkkaJYa;@~z
z#mKWp5FV_NJ(|+9!)R!s1UO*-n$Yxx!@RfS7r@O|SE?{<IigS9Ztrk(TfSZwx-`|F
zWXmr<Lbm;)2iQe~QUvLCx{on#0+B6)uK~_p5mvnMiRr%d$wIhV<4lbXmxQj3G+Hx0
zUM@__iXI!3gq2LmnJfuCP2NHH;=<zdx^mH#CtaDXQDHF5{QNU2-hHnCBAF75Wbj1D
zo)@&8#5-f`bfi8k0!G=l^@!JYvQ+$jPo`*Ve0=<Xl*--RU9iK`_&u}4yf%LP`4`?j
zKJxTbDP`hR#P=yv2e`ybHZkw4n35r){*{!J3Y}H)bbjJr{^eiz58wZq?;gJ6H*bF6
zc(~`+(`(F(rLCOT1!?C9UeyJ!=}F0cXQAFV7^1!yT5H3<K)ckx+jrkj`u+b=5bI0h
zvuARD3}25eTh%cKo~#pZMuaj=69~-fI*hamT(fY%C;|r!7Aw1%csZSYk4v%FDojy9
zagvN8oLdcd^3z}J%e;}88~phGiPxtS>2xB6)7~mWm(b;${CVe;$Yc`Ax~#0txGe6E
zr73Zgf~3TCQB)eM7&Uv>;A+O)l&h|stum{Z`i?Llz^ZD#uN4Hz#TBZ=8>%=w?V44w
zqr`cPCsyNJmzxtH1VQ$BDDnSvOhv{wH_g07?^rC;n9(#tnggZs3V_y>kI!c=m&MDX
zvyzjGw1bg%cI;&2G-4<Ukd1=J7u7QNuO9gAuYbeDk?#%<{C}R_bI_H4_~CC^>Wa0-
zN4>DN3*OQrfrP%EWOxB&^&M{musUEU;WY$XI{>wp?5KYei1iseR`net+n)-IeH&l<
z_y%Bo8-KR3D}x3+xGFhwC?y2UZ7Yo#`M2Hp%%0T)u$v7kPKN$jDR~R&p8F7e@xJ>H
z7`~*3e$`y(sWwBSrhCr>>AGG$b68u$YIDagJ4-7?>bVQnkD#&nVbpIqdF{rg2_kU=
zo<?8@QI7d@r!N!Pw3`X|@{pGk*?(m(oU>0$4hP#l^VTnq27hnY&QTZV+&!V3c5Qeg
zi0N82{ekstdXJ1c(;@?h>|(pcyhQWW;elhiBWWgC9__css_0Suc)mfYj%R7x!vpKb
zt21}fo>c}Q`_{fYbm7Gme97@04$E$R%I#0SiV@xFcMBg%DU_TYKpyi5-Hrn9+_5rt
z)Au<$e!0}*c}yv6ptl0I+ojq;p8m_-acX8GP~N(d`YIc~rF9mC|GuiSR_{M5B4~!m
zyu?*f^2{iL1>jj*r8Z?%XKbuiqx@xeAJW&Ch6|hqbE?ztKOHX}#GAKcuM#2q*W3if
z{v|Kws_!p3Z(#2B0=JKKt>X)`=(6g_MJZ|M)cH0cqw8Y-P%$0Xm{a0ZGRNbQ!|}wa
zOq|lhL%!!O-%+FtaJyf)VK05;{UwfjwODBe<1Fbq8lACqvsT7NDlgviwuQHV1K{=|
zOy2Fo8&vr91ij$QjS<M<vt3elR^d2r;?Crc+w#rW&DZB_x%AFiVl>(9D>q|17tCC^
zyR);M8QITIN8U6TlqtE^(*}&>!Dw8ocUmhVB+FxOCkqy0@r_0`*JxU`F>5p;;l<Gy
zbKx+#ci-#tG^Sv%|8|T*v)h<!|HY^NjBxw&!8()vc1CS0DQR?V{){~<&V~-4GD!sN
zL4`}jo8}zIm_kzt@CyHPJo&J%R#}&Ygg3Q#b-JgKIoZ+A05g(orIaxlW4_1v8*_Ji
z0$2#t<GOnB`y2c6W`%s}1NNW2u;uX!eID7(t&C4bc=0=b;iKE}Hhs=0E-8CM%iaCD
zepTODcS~<jcE0>MV~6ejOOjw92WZh3<Y!#WKF8gLm+D}%h5zbI{;es2>ed)U=n@6~
zq|MCFW)Y-$lZBKZ7caSUwrDWE45~H%it@EvJKnLex?fLc;iYk&cvsXN0JGaQiSPBb
z)+_8@MlpRK7;9E`D>v9#^RiM~i%q3rvNdBcyldy>dUZj2fBFy!L_eE7i8U1mu$n<a
zJz*E`B60tQpVwNAY{Ee@57QlD2Oi2j*{-2u+3mR7$GD+S=D`0c+a%m{=5?U&_QdJF
zi~M1``wS?$|JUd2_+OW(`Zm7y@eRQGHvVj5;ZtY!`!{cS^ZGUa%Rm1U=j#mkMwX)M
zTy9X*UBg9mbSv^hW)}pzAG`@3?>^*sn0S3UP}@Rljm6rHfL8A)=M>=$I#Kg$zEtDm
zwJ{xq`zeD|OuPe;siC#HfMl7ttc~aS%=bT>NU6}Q`W~JgTqy~bYh_th9-l5euZ^Wi
z=rwI)u%>s);m*J^ax%ysIA$?S+_7JKd+}Sn$l=%sMy6*$`UtSfWQb52AYIEtZDija
z3-Yol|JALe&J0Li3OZ?GB6B1&O?Q+xKk!)J^SoA4nIIJ!+4YBx-hrkzjV^$SuAY_g
z{GhLxNfNJ8;y?W6xBQQP{~!3}cRz7{p83`BS4hjO&xTcD$_0t6*)+v1YG<+afJea!
z-mbi}u0m)YXN)vcN-7ZuxufJkeh*jwgiv+=g1w{i!T}%8;VT&n^s1hq{2Q+yUh(GD
z1ONEb@3}7X$iljoj(id*_wgk=)?u65>yE|kcwJ@(i#jNUUyxdh!rE;vP0g7U2G@1v
z`C2LBdK<@sXDa7a(dO((u>j*tnFP+~E6PkcJTOg#oD-!K*UxcfW|}2xRi3UF&Q+NW
zo|Y@`t{=G8rN`gn5vv(%Yh(%~PtmOh*Yi;gr+L$Eu-zQ=k+yAtP0h_fys<#XoQHd8
zz1(gq=WDq3N6DGv-I3|wS?u0<G?Nh%bZxY?QrAjOaCb~dI?zm*4iWoj7kX_~snx}0
z#ay#U`qfF|)$z!0r+a?=!w-D_s~_M${%`P9`TK{z<JH{*2RVATGq$tB?<*D?4j`Jy
zPHW0pS{BZ-s$WB&H#P4(l{#c7$=Mu#fMIb{X31xQ;F}i*7#eh>uZg*F-NhR%*i7HO
ze#5K#dw&1nCzf@IlDR)(%hCg~Eza=sVi3E=dMyb~sqm^CnCq2QE1DS{Zo4gaJNEK>
zKmVE<S5s%_l_S<xw5?$=%Ch=d;^n)I)+#w=rj)1?kd%*?6|;qz3*~rVx;t^y1Jg7y
z9gZY~=UTbc%5072y6}E}V%F8!C;i?Nn0WbrAyHC6qYJU6-tn!0iFY$K2KY|eu@G)g
zV?ajdP6-Xb)Pbr&$c~bx6ig+cwG|UZgK1OpA?z8~^+I?!9(Z0S?oLpOQL@45h$57G
z2mX#yXiZsH4KTV<ScCl$XMIyBO2&MCrz3y={lDYQWnuo8GnGo4lq?fwg{=#y?>;%@
zU{ShVmei9^5{wN+TgZZ^SVU&1dOtP^R<t(&N-q{;%;il2YU9QX!1LvZ%QJ1Rg~@33
z@fSa8pV}N#RpD1}U-A0YJ->hV3+uZ2_eE$q)S0~34GCk;(K%W--=<Gr!oShieZ$vX
z&ffL!?s#BrD@_|ITQpiRM$;hK>izvaygkIM#$1hgRqv`T8N~PH>t)8&IFtzkme$CU
zILPGXucACHXVzMo=b1?|Cn>!B&bWVN*l~j7lI07P$8~lFW#FVqqB|mN8WqLuE-9pv
zAtm<<3|ezxs51Foc(a;@nXlK8;O*d!`kGDIOK3IZQ~aC{*05y8q6@!eU043)_y38<
z^E*zbOrBOqk3b#!-(^|(-S2<mj3P&{+)x&d$HalmDSMaY!?g3QYYP#wxHfO4O{<Cb
z@1OXe|L6ZmS}WF+zkmC8)I9MfeGmE(Te|E=#-s%xAMKCg-yOW_UatEf?rWQ_IK!;C
zf3G$6PKKOFN|`Mg@#zqG32pUH`m!D0y!6}nFJgQHu)d8y+ZchEK*^a?_O{7buaDlp
z*gIutliMxIzOXm#G9LY_zh4reWTCa4v$)Gt3BBEJ_RsB!1H6OEwN>7opSe%>yq*e`
zUjLuS39@;~p+srVT9wD=C#GfQ;dFwOvE)J1s-DqZv<{%8A=D73tnJg%pikye2ps|$
zS(~22*n=dy)*suS(lN2E27i<Z+U^6!9QD9AzM;EM6Gn}HXE>;NsV8_Atu^mDW&(Li
zNJ%clP9`NN1!&=P#k}UAwOK!Cvr!E0NZjX%`{}?t+>tIik>$XvyI0KhiN{h%DkDf3
zkrjma=6;22Q=<_84_#}KanGH>A)zVJ9pn7D?CyTZ%ahYSIa{}R=Bsg*_jjC&+woMq
zd(Z})_z*mku5D`{-2t7?*>MN3K6y&!PEnDe%j#t|@p=PfWSt9YHAKL>9V_Q`;m7wM
znC|a+INiC!Zq9_%=z4164t+_PlD#w#HLlA-)kaCe@o@AmV*rv_6&8aPWktP#y?NKZ
zruvx&?hGtTM1a{@L&N>uCJ75LHw{Pe6F|!tx;dVwZl6A={Xh2RJ$4TN{gl#pz3U=H
zS#{`#C<)2ozFMM`D8=j~izq4a_uRziJ0t5<3P0RG@NhhFm=4sXQRl|P>4ArG&#4@k
zQrVt;03t%zui;EqOW`ofcyD7XA+!KIJP`BFdgH#6Upw$w`|b9<=>!gVaC;_ffX$bF
z|7#dGwwx2kR5+AE$qowieHs3+Esx{nW^L`hxBH-uWn?yS<6cs>x8;;l=qR?E&XW!0
zdA{<`KmE+#PDhSq@+5=0CKccS2~gC$d$2V><8hnrYq}WJl#yJJ40Cl0ytK+iE3;M>
zt*otu{@3nSq{oAb4d6EtBGFmvmMPSACb!d?{+j(Ut=kdYCwPDg+`gYvvDo(0cyEuN
zZM?VO=jrvjQfrMatwcGIfFx4sT84&dmt63@N&-0((_pH?>(hzfeE%!%@9sR|SOr;y
zTr=Vg94Sk7CQZ~tOH5j~Adz^+9pF$e_3Wi%{yiN&%HEIhwbdEz>2|Vj*5AIf>dSF^
zCV%#P{u)NtOf}a!nu;?VJ;}NOD1BiWMvLT=2B@+3n};2LX43k)6~(1ARG}%>T&v9z
zLAr4;D;MvKU8~CqbTC^HmZqF9GpB=aUrIdUEgsNd?WL5FC?yv2&c53GzP#I@VPf31
z#zB?3dN=8&%A$>{R%UI?k*Iv>DtBgduLf|l+xGuq%uycU>?|*@-3uLj4rbY&(dT@>
zjbR4f`TUm;PdqIbmZnto1lfM+cimmhkS0&8=H%LQCf*Px2f%1J#xyR*{CPF45R9w)
z!yVtfc}qQCsmscP9jIjN3bsI&i7X9Roq;9_O!0=*?>iD438Fh~Kd(o0u<ci!)oYA~
zSia7JgJu03|MV=ixc|*z?DKMK$Ag)_f8Dq7S3kZ1Sl`B<ZG`_x%5XXz`0nmCIc1h*
z9bI$!!C}_CeY=5(I_ronVC{#&es6yH)kjV+rNkV;*h^1F@UobkvS;{Iv1n!8FdpYC
z|JyG=ai4$9cXvlt3muP~N$U0REv~<|#&ucwc>YMr!aw}of8cb;SjntaX{+l(T-N}w
zMveZ;$Llk5FgE@WE@EOd@fPDsluZCO4i(?dRkuT6Xf~z*-}+?tGsT59{P#^5-G0J}
z9UWEr;gdE1iwFu@%w38@sJ6;=UHEvpP?M4GCnRSiWezfhkfb{wS`C#5nzN_LS`GR}
z%A73oofh7viC6gu=73~b4?G+n_`C0a19s*4{3B%+tTib8`dy2}lRXmUgYoPmB2lZc
zKE1tqPj3rCTbr}X4m<EF7=pFdFN0Vw1KCmQKMG2|{E+{GN4H}VoQ?;+yL-i<93u0&
z*R%1q-A+*#G`WKd5qEvgjxFRK;&pqG=tLY)>Yb^&rrmB@IZP9$>BQ6J$up19A*_n=
z?)i!T*MIy^9RB|A_~GG>wRvE+tX|4nN`a<mg((XMDKskD8jp_`1Xz}thx-R4Pu{*d
zWiHy7)wqhWh&QjFwQ{Ws9`=524!bC_L~yRO(483*9N*s+ZTatg+WOAtN4%4$7fRUH
zLtu||edb|`nrqD6*oK9+hf>#-d0Dxv3py1n7pCHqe~3LpQ7oHx5lvlGwgt#^pqFvJ
zKAiZ6w?FX1!)xwNC$5);`Dx*|ufON@>7F-tC+?<c6cGwQWU&p!QV)Vtl!C{8Wu1kk
zVdtB(o3oc2TLd`?XUU|LXswNX@^j;+yIFrU`vuo0@#{2YA8{MU>A-h)58NG&6iJ>y
z(m&0X<0UTse)HB_0>p^GHUi&bO0v*=3)b*{YNwJZ`M|PXUf8no(}(x`-~OL7|JQ%`
z2fqJzzolNTXl)qI<jJLwS^~?gx>B1_EP46i>Vod`bs--b2L(w7QWl<O=ia<qT`2ys
zHqLcnX_eKKPt!}zCow6Wq-v{o)w4|JBm|^17*j9JanC@#0kCd}MGgkL>FYL;-Nt&8
zn_&8?8f$e>0d;$9&Gqcg%t!$#hGgY9Wl!E6gGtgfI#POoo=^lYA5NwuOH9CT-hR*j
z_}~5yt{GO1b~g0U$k)Q*zHoST$KCuMYi%U12=}X*>r$4Iogf|8Y1U|wA`}V6nx(Ci
zC5)Y+->*baK7!jXjs1_h4>(@GU(7E($gg3<K3J8*lz4SW9I^w!ZB^RZ_c5I1t1gYP
zhJ9wWk?(_@>$0=$ZpR&imDNSfldR}msm(|Q2p>~}(RrK>+~1veK3`bYb+D$ZWG<`n
z^M@x+uQIRikMss1&h#W~8m2t4RK=DSauI@7(A6NCEX>xpua!gdWK{($yB6EWI<wT3
z>$<Yk`U!8x%)904VD|>=$Q%WdaBMULbe)YF3bodD>nO@zrAvmqm`B-lwsypMFb@%9
zSr-1|?|x#ok33O1G~vj>-@eC42CW%3CDJKzm<oq{z|{TbqxgQHxTfY-?r7em!u)e4
zDZHNU_|12}=D+)||2y9O{1Z=4&wT%Ua2?){Gun(}zNd(ibjDI6i9+jLQlsi4C9jl`
z4koiP$VVcv(Ua;nI@9Jc5B`6NZiW^*pnswrzO-X@Z~L{G^0)DC9^U}0Z{yE8N>1EA
zoVY*Ub9aBwa$Tqwmjrf)WyHo9Ar>r<5su+-?DbIgd(!}~WcPOn!EVpuq@*seju9=%
zJQFb;I8H~NF40yx+@*%{v|Rb8_wV@as|V)xHUiGhm+{S5%=NBvE|emi=QGy)@JuaJ
zk_#ECWM+j+Q>s7{ShR9&E0=X<)k^yeBYyX3k;HJ2&4rgc7r`BTwMMr7-CF^}EtA>r
z-LoBDvc%?{cw`$Hk?1-lFdCz2e@>g@>_WCKaaq@upMHMFYL#=nQVXOkewZ~c$@6e?
z^~2aeXo&Qn-9>nN(D?T5f&cdYZ+JM|N0YMzDU+hq=~xP{?@s*c?Q8Dd7m5n2hUmIz
zT)1~o5eqLXILt#tQUr2wU@cG>7k`eGq+5BAst#InPYAuhMWb6|qUZK@PvQ1OU)vC8
zYurDacsSm1ce<x8m3g^(wqCXpKr!#Qx0((yO&LGCdCT!<cRco?L`qPy2Rtcfk{Ts4
zjKun}W=y4Un2yMN_BY%)HgjwIa(?Eyt<)q~)96YmLo>#8Boe7Dq+lC}3T<(5T2aWk
zu~dyx7H4SAt#Vl_*IJpYdPxOs$jaTYEZ8os1#l#iy#wo*m@XST!ZrV$U?p}?f&OPl
z4+G4Smp|JBLSeJRj}~!|#|*T3V~r@cYqj#>!y~gU%(c2xp1^CV605fhbf<HWCVYeN
zj))RWMMxE1O-KIb_4gFX)MZ6TlyczB{TogtGe19J%gUV;o~dq|J(|zS0q6cirL$aa
zCszaqoMVsPcX4vC%?1!}W}|}S6F=9&&PiW>!T4Js4;}3KT1MFN-Tls%*UEBPytMvL
zuw`Rs%l2u&aC069LvNarm#i)e)}RzW%U;^@@o;3V3$_aq4Y;<(kLL?dtx^+cPLXsf
zuHV#}GfaBJ2JrO|@9l3_P%G4yARo!o9dlh+TV-yIYpXoY7tYIt=lM*v&oCPWP;Z)`
zLhK=+^pZg@Guk@)GFU@yn>*m7clShhcD#KII6&-2myGxFKa-a$dG)Wa!eB~eeBY>R
z<I$cuOJhz3NlCe3u~#Xpe|9xY<J^yc)7OHtJtj-s*TjiTAv3QF%Tm$BnC!rR{a^i-
zHC_1q`j1F$lm`25jlG<H3#<M5x4YHBp6u*d0=Q06K%$H5?o^d1ZT*rhn?*wJr7qa^
zirsqcZ2CU}i2Vf(!@K7lj|W~I3*X&Otm;DhZ2<JHKV<k)YwCJX4KF41vnS?mXUE+g
z?Cx?ukJFTpLxN)oz0*btO~lr0*nKIP=}@q<;7w0Gg=)EQURM6`$DjCaI+5Odhb}9k
z=2~=upNr;A1JFcf!dgXJB%=+E<$zkEX~rn5%`nLr##~pfb>VrvFt01u+3kCK+%onR
zRl<(@;TS&$XDs#Cv#B-pxx#=H{$XcnSr>kPX~*$l>GmhCA(oMvC^>UiP8>Lr^bkPL
zwyKw4VYMnWsa{TR1wy~i2#J9IFMEI1BuSFwd15~nQ8jb-h{&s|x~san=fdpH?9O5r
z2VsF107Q5J5#B(w-x>r0X!ab;>`qU2U73}cad^0!nW~5|9+*W`%-rM3%It~El7EtC
zxZ9yBA}TC=GiTq!ZZJ~{BFg^C(hYBj+)Oi1uC93=0&6LR9N9DT#TU=mpJx8>?U&|I
zDr|>&owtCc2ZpW7cd2soc?pD8i?V*Lo&$nKc2#<-KRqL<zc>A*8^;ADZ1HS$$G_pD
z13x(o1J=L~2CJ14d2(~Z&HkFJ>uU~&6ZglH`OQ7_7`*HO>IN(;!hT)qIot^wo$;Wa
zX?+?)WNX<SBoM+xjM)q?^?oI=-|e{GUvqyvP;_nAFDUoxiQm6G^77|D=a{n<hIK<t
z0W)A>5m{5TnldRVsU((^2r(1I>NzNpQxOiy3|J*#DTO(4zn-j!<)w^OPkh~8*aSQ}
zGgSmU9jZzoH9y@KX)9{%>sSTmcB5*%>ekr*r*x?v{$AB*Hv^U-r<HHMeL)Gz5)0vG
zM&=1!y}pGr(^OXHN#e$^Y9ta%MH0xOe0u$afAaHR@^p7?f@QESRHA_>)(GI~)t)at
zdCJZG%zUz%U`3P=YrPRcLiFIWY7p?wMy_3{L?n3kn4F=g8L(=)U{#`AAU=!5xR7<<
zSCQRB`5zCPjWP1{$rG-3SM2v!oQ|)&UQ|cNzFY<MbXhFo0@_+N@$Fn*taj!0nJJj@
zJH}v9LLo9wrZ^IcrzTkKn|a=~bjBV_&U~8|4k?prT1hE@chU<cEQf3<N1};POmNF7
zld_VNGVc;=E=|Ed=FBN2?$gRCB~B@^meuR1)EaS@ijcjQwYUcsYW<IzzPc%5-NB|8
zaeHe2tBPh@$88l`YE~7y&oYGZvrMABmlHu@aszKpnU^nbNud}WPuZen@(OB3z<M6p
z!4)iWtM*|Rk2-Ned9vH{vuB?%OJrRVBGC+3*Pjq#;&lHNtSeU`5}E<a+!@ti+|ro|
zo$<=fteT<(-r-d6H7gs9N=LtsDX4mW+}6~bx07d6rQi0Uj}kFPo;@{MzS>=LdwR*?
z-i*=jOUrMYwLK~4AHmXq=Q%LVQwvF&rf89`9WC$XnXBENyZgHbw0z8&uhYuww6cm?
zbWa2+lSTg)C1<60i59chm@!(jdSuz`MyzNNSo0O9<ALM4aLR>4O5C3AIjlzx>v1E<
z{!#nZt0g2uaAUYfa2J-2M{E(i+o^xK|2lqLon%#?Zu?T)4rI(&oIi!&Q6E8?3>0v;
zR#8&Pyj~K=sGN2%&E~i&PRnaK*~892E!x}+SQ!!6O;+RbNsjEr`og7@I4&phqD<`g
zXMgigxLv;GtKa=CA$gjn^Z2@m8Z70{4Dw!I%~Ov=d<`TtV~Yhv2d8O8NoQhFpYGe9
z&0~10^1FSv79;=hL`j>2_Q`(FvztAiKHc-jmoK<G+-)gy)EOE5=3>ngBAb!=?26j`
zTeh8PqLLtcfqjHIQi3I}TX$Zl5p)7fb6}n)qmfz`LKMQp{hIjQ*RT1fKl=q?is)i{
zx)x8DFrg`1X^j#HDo~O|>}NA-@1{LPBUxuuBTH7wWOQ;$$!cG&M^4MaVOcsc`g{|V
zY^+A3Y;XH#YMtC#7fsv0O~j^3<myiLjTh@C?{xo~jas(%yvwQ-VG0vh(-pIr0ZTmP
zp?2I{6;e@4p$zHlj7i0^0#a?BQt3_r&4>i{`^dgTu2(Z)UGMft2&}4vJTc3jzxe!f
z!ad9{0x<*GbtaYWJht;E1TSJ3XhysuR^twLG?ivPts|u=YbK$Wrk}hK&A7z_dHazM
zZ+5SDeRSX_fnmTJ_`xtu6T97>Cr_S{Q{w9v&p95CtR+*DZs(F(q3GbLErmb{!by#N
zJP;~$GriOztI>aUn80lHa)RaTd#2qS15pBVm<jA?72>Ow6l6PmOJ>c=l3*V!Iw7Tn
znxaxrBx8k{5q9AjqAPcYM9L@R$=<v~D%_@(+x1A2iM1&A%ZcN<n1U9BtrES@PkdsP
zK^C<z`#4#Bt-QJb;)<gbuaC9Wv#DKTdK;DMCZf3>?7!qrQLHfQVX6F3O~?zA*DzZb
zb_iE8k<=p6QpQ8FgM~8ZY<h124PJ9fJxU7m9tR2JWEw%7CIS;X_WbJezu;f}v;UlF
z-jPEfPBYV#Sn`qkCGg_@Iq@d)mp}g{*SG%`ZmqC>Grl!zfNwH4Vzsv1E_jse2EJ0L
zg=2$EuKZ6}lvoYr-uB^nAXTkP_91NZ$D{b}c6+X_oQ~JTw=Z9CI$HE#snK-Z-p#&e
zt@t+V-lXvAR!w3v;#lI)gr^lopw>GPA&9WsS<R%NCIFt+#XY^D(Ln8zw<sJ{IF`V^
z1Y!X&rFTt<l1;EjVbai|u&g^_V3I2|PaIDL3092sKCc|p%H6tH%IxXDl56pH-Dv9!
zrmz(kVym%J0Ev2@egH<l5=?QxQiRmk)v}3{9K79qXUtL&iqzUzLT$m?1nhvFfkg6Y
z;h?t&$}~liM@}p%NSxSR?I<CmIa3yk(ukgFWYe=~C6Pjic=2sJV|P07H17F}PkzQP
zfAK5QY2p5Kf?(-hG6`V{#NEWFKl_xwo`21YSHI)c@fB70w(eKb^3=>Au<adJ&9JDA
z=qk9h;^Zv~rYWej6r_r6)dO2%Ydsj>6uq(G$3w-2>+2h?u68_q`U&f@@XgET78RXy
zi;Qh}3oZBa?xV5#uZQ|0_8|>wX?GzKJ^i((<sj4f7%HC-s4d!8RSvnZS0QR(O=h@S
z3#}-;s?|IbwN&`sG~2$P_Ao_cH*t4&WH}{Xot|?_3%93*Lt1#f+)+wqk}U$IL48pO
zSck;mbfvE3)!HysixQ+n@UCs!VOOuWabq(G7jWl`tMdfB0fDCs1~Fa}a9d!o5VxXA
zN|_iT&cc$c;l%`mdE(Qj6O=?cu7qT3nJ3%t(sduS%LR%@bx|@`amU|${tJHo>~r>4
zPuS(mQb5821(tLq&cd(%`Y-vKJFZ>^4!RYauV=D)WTgnGN5tY0yjD&um?DhLQ-V-5
zgBO3dO(o43efez80yxedns%=|<k2Bm#D2-=?QTA1>~5a2-%Z?n{)Ajk{PFG;ck79@
zlor5h-X7e5-$xul?$$#tmZ;KSU3I}yJ4^9>I0xb!nTthp;ygJ>V3L^_BYClg9#&*h
zJ#)ayXB-PG2ss2o+z~W0WsABhOR};axuAJMb3oHVP~nRwPtC!R3KD0+bj?u{r~3on
zq$9_Yxz9(|vT~meW;|m%#ryIjm6nat?5vQ+xcx8@G5Yd#GCr~yTx(?SdAh5wSFPXZ
z*VJnM3Z3Cf6qS`FynuK_V`h~DQZf<>F$6T{CU;I)cFB~i^Lf@}(Vi~9n+$I`3bk~H
zl!ZVh%F4}d$6r7Fl&8~<<K03AWEYvkj>t?v+0T)mfBreIUo7zYTM82;287kuV?s4s
zU94aR#^y|`azT)`-z?JI<r}eR-hQnBufusxciTF8_(BT{els0!G3S>D?Z8hJ!+<sL
zgCT~=iY!kPX<a!S52Tc=9)h2^^*Cy^CKR{I8<V|Eo`G{?d~f#&0y9DoPm@zqx~eC1
z3RduW7k6k5&4^~#^t8M#+&vXoQeih4=PI5v)0D$vZMlO+X4#_>Se6WuxlUF;W7WXP
z3!CR4tVLPULduD=u!~DMtD9{#ww$xb%__AS07CT<(L6Alg<!K~xs_)V__hM?xBxDl
zg{Hmh3aV-$r$iKM$z4kDTWw>h0(~iJbvar(NO2ENRnYMw&C-<hnZV8dn$JJ^+|#wP
z3+lpBRzebPPp`SYj!gUL5p0_UMcec2SFmS%CY&g2V|49Dr~*b^=g#xyzip4HdXs!^
zk374_80n*r`6I?OO-$3wG)=6_%Heb*r$jCf=@^;_NLFoWajwEVpktsl#cVrQJRZOW
zqAg<2YDY*g<#(3AnuIlFkK_#7h5dRNi_AM^WiCQ=>0a^3j{@pay`>zP5i16($}|T{
zuFFX&VWo(2%qypyIHlEpPGrr^+5ga?SoaBC1khGz$>!vrYa#nJ`e;+)tI@`Fd3vMi
znL)vg*V0!g^4kVC0$NRx%*qK+kD3x{>{66SjFbJXRs_2__GJ^v5s;vsDz+ILf(Tb(
z$MtT{)zuBJmz6bJZ8>6~NCC+tDTH0*>gJmGIyOUPH)KDP>REkBIsG2l)?ZBvZfn!l
z(m&LqtokcHB;fA(IgM*oV(R(uW1?i3=eg1HayoH19x17`0^pZ>_i<W&L^-%HUct6H
z)izriud3<0!D(5TBvNu9=dwjAH2ScsDMri3RM@$P!f3d7?Id?HxU-_tz;2q%NTahE
zu;!VR63eo1x7>3|3wLYclvm>~svg_?7%ltEIwMupi2Kv6^5QP>s9Vu<nWq2CXiB#;
zRkk&#kyouufE#|SD7tk{6S36&HCZxp7Ga83H)lDSk<z={pk8OR(SZ~s1S`ny1}x2)
z5QLknYini^C;y2?6<S!!itOOYlPB!2cElRZRFtjI{kA8XbB$_fu9K;}KwBrwzNXIE
zn)Pyzwnxyd^<#ATxbj<h&KubEM~Qi!ndXVO+p(TboYs@cnBDj<r&5DkD@_Y+t3CRi
z-V9JrWP4Of2)ztv|4t#Yn|5Aky*TUHWL(ukR;wAgder3JWx$ZKIm1fve=A=$J{$?$
z?61&L$hr{b9n)^d>6BR4mDk66?(@o_EM(0j^4a-b*{WWw&pA)A8gN@Fj!xoh&QarW
zZkKwXhnz)XwJq3v;Y?$(C!unLupj{nnXF4&W3w{Zrz`Ikm#4F59uqbm9|Bf4QliP$
z&W|AzMR{_yX9|IJO~%6~lc_>6-Z)Q@PoF(wfB)QNV)61FH6qQ+aZn>c&!eJS;|eTV
z-keHdOM~{BTxZeMmEX7PY{KB1D{l0}+U-(3-?0NfZ43j}zz>G&s~chn+}_@@oKBoh
zCyOo5Rn?ScVOKo;OT4{rwpQ&yf``$XN3k7x7s7T;*S_8fB`03JP8<(H&cfBsTD8y9
z9<VMghttaOm{<!Sv+FKa$XHcbfEV{i{_y;kzj`+FWR66c(4w5~Pvj)%x+km?a}InO
zpRvlyxA(7DPnnn|Qc#YPIi3=wL|(4<9Mi(f<t0l_WLA#?^U(c{x2c=-TJ;PMwY8|V
z$pzhLhN83*c6;XOhMW$RyjoavP$U$qWul-lfH=vkSeHN%LXK7gqr}!FD!O8tb+#HG
zVPyvP1lIc_FHVJ(1Xnnd4d}eHdMA^mNv5T6S{F)Du6EW!xa2~NE|Gf*#Jo3gVLct0
z!p#2aio5m5ZveMINWZ@MhTZOp5F-o9Q4?RKKayAFH(!6r>)*fP<>>`qrF$0Hdupde
zPJwuB0!yz{uU8OU8z^D2`Z|TyJ+UcV#Ranh#)7XAO`w_Jg`+PGF}*_EzuNzOuErF8
zEV$m?Ai#?kFG$P6dRol*+tmo$U?N|{j7|w_r~sJf?&8z+Z%^zb1G5NEb`f&n{_dV-
zg_4D<vLggHthln8QsQ(<F1T+|i#WYx;Q6b2e*61x`KO;;v7aWUHIZ|%g8W$^Pn0y1
z)67I6gu*XB-=SneS+ShLd%n7V&CB(`eYS2&-`?F?I&kf1*~wp*v$SHTzIU{sGMQi}
z0g2I!0-Aw|Ah8vC51w*Wj|mFqU5(;h6|KV<SRq=PcQ<%VsLseGR{N@U9!%>>Ntw(6
zoffMFCIOjuh}T-x5-DEWuc&erXDbw35+MpP6!Y}DQ5uxva*s}#<;e}NIPvAHmy~rS
z3FvMjmzCchf5$1`^X03*<(uP6{&;xKi+n_Mvhz($ZLwoQ$?BaP-N+=a7#FoDp;Rmq
ztN8h@B`|pxQEfZSOfc#m_H*Zm{Xc}@ij8gtt*Yc$QIOuo20tP?22*@hm2bZJhP);;
zW$((lDQRcMP&}AJA&JqZtwEn0!nuLk?Z{cJj@`3;Ae7Ag{gJd<%GzFcM2Xfq3qX}6
zWsav4$)hBx*Hnw#NWfRGj{NQKU-2(K`-H3A%#-Cv&YAnPl8dmWNJ_$z1es^9CiwN<
zin-sV%)Fb~U*GU|FTduGuV3=z;WhW`%BtB7Y}$%WKdw6|uo7Zq-eC$^u?T2YWm08|
zGa)`h@`}Zl+lqR#UDg`u<IPJH>mnC}kY%<p0HWfx=_Ujvgk%A!Vv(RZo|vW`PxenC
zXHLggl(Jgt-+s0T(%EWsh7u8GRD=_zyv{Ot^jJRA6oqKd23X6HJ0ic|9r!FBdCAIA
zGg%Ak@jzPPE4?M4oI>GNG6xc;Y~5L7Rgi-f60fPff<^RM9cnL-UD8>%OQp2ES<gW`
zUjre|X4rD8x>l=9zjqt$cboDNo}9B%ipyI+6213&mw=TQ-`rZLO9ovcRv^wg#kF%^
zH=15uOxe3ZrxlZsEnutKoM2;Ta}0dCp9vK7ctDqoCWu#iVu(ZpLLdh?o-(IHV$E*A
zEcUEdRy(Vs@b&8>zy0z#fAj2`n`vS?nWJ;jY&>VZvNZmbxW1aWiIFE)*Jvpm?(Wh3
zjNa_|`rFsMO!wUCfrG9j$q!_)tr0_MbF#DiWX_jpY4Ek!x}aRm=7b4xqU4O0Ovp})
z-VRySyCKeIlqq^7<vo%nLbOHKNi1608_0;(?tUvUAt5u*D{(GN`w8+4cQdoTLSrIK
zlXV=7(K8lckwntW$)gbWlksFy@{^;OpGcxdOey!vfk|dw>@qKQ_q>Q-lctqv4!mAp
zkn+qS|BCC+KjnY_U;UET{~zVQ{oUW9#n#ixvRD*3D&7pKI)Te3+j&+=Wtkc*`gG0K
zB|26ITBzr2IWvISKj&tNb2@&A%HO*`2L4Pi3|Ip{7{n!uv{)<Wk~31%_P7uhn%8|(
z?VJCyvjx1-Nv=@a!mp*Ura~#^EeKxvJ79HkvNM!5B~ygu+|*w^z{ixhJse3-CR$Vx
zY*8B;sflJ9BO--~NbNLsN-KByNJ^H!ep(lf`9vzoLgTx!zx{-6i3VqivbaPnu0n|S
z^4d%#Tf|VWrqu|n7Hx%V$6eauS;~zUyRMNNIcKz-tPresRwD>9MGqq`9ztx<O-1#r
zj#j<CxVuU`MN?XOQenv}cc&u>Axsl55BEGjyynGv%c>cb!0qXlSBDoIXG=MCH;*as
zZ3od5Qq^r8svB*4?IOahu8~SxUtF~im9u^LhPa~k1UP&5M<O9#K#Q_2D|t0Kek@u8
z>WbTMN=@r=e|xnDRQa+RqU)Fa?`mi%UWizWb-)P{Ej-@c$m-pdavjfw_FcWJ$7SW!
z-5u%ao)Cje@qHu{tpIrRy?(vh1G#m+D-zLD;gl8*32%9wE%M6u``gfxOtn&^sRn|3
zRl$WrZ2!7Z?0i4zM!R@~qZXpG@r}mnp1m|j^ifhKJQlU$^oSXvLJ(v!0llcLZz-Na
z(xNE0IaNBATGtC24JS<;(}~xodrC^AHE~~7UZ*2+|24;S;H&$Wyjos!yBtZf)ex&u
zu1$1xgC)LB(zmp%_3b>GvXOXq1*xe^sz$&5>YE?ix$(;C2lj{9kCTp3GuWp!S!$@$
z&{=x)_R+RJM$2mHsnWgv7Fu>&15*s-0?8vKrAAW{sP)c@^9$=fh@H>UDV%=c`_szv
z*Y~7PKk*uNSRrBE()g;-5P~2P=4mE-^v5&>cC)3uCRUb`ST%12+q?O`Yav@)uK=}X
zj?^<gLW=^Cvq_XLY_%&B<xwJ)w(DJLq_njE+6XHR6eyZ0u56ku!R8+(H-P8Li)+Dv
zErPP0+Fg21Jx7p2*34Zxa<?9NeLRqJvQEn7#4%^Sy#E$*;k4fJ>UgkvgJq+!dfAqK
z)zR{{er&RhH$o{j!h{;_(La}svi0m-IzOY%7*F@j2T!w!1wRhD7B$*U>C~crtiG!}
zx_K)tTQsHDZ}WZE+v@vcujd{jHQF9>$rhdCh78MaFxg9s?X{H3gJHO22nct}%C~n1
ze);su!-gzkOG|B56hJZ(lz9@uWNC_;GfNb5P)f+;kjWvD$lmF)@`P^iXA5_^Dg`KJ
z96~HLPHUs0`h`xO*5H>fO&_RyM8%wXQV%}aP6=T%R$o4#5G(>ZV9%mtqrV_ZjAqyt
zFWq6$rv40@UeM84y>%^Q=Fdj4232?Y#O-?E<#I<-dvE1LDVdk2JC4Vd<&Zh1)t(EA
zbv}*tWU?{iacX@^pPy%2V)P~(H~G6)=K3dbAwbm=sL9q3=vdF_SkM0canSI<pN?U`
z8u-DGmz69^TC6BzNx2yWHfB~2E-&C&aBq-O<7rA2);yHFIodi~LsYNrGVNFnE5}1&
zns!WaW{NvR0ws$J$&ga<8V&V;^y60R;Td>-e;}l<`I}!nW4gMb+?&Fev4Y~n$Sg2T
z%9IN`KyIFq1WqMz_v)74eDf7?o{$)MakynkYmMM{M%b^xW7s~CRY?@qY>Tv;JWRNl
ziOvKVywg|>@y{)Kt9Z-m;>DJmk)?(7N7O5S)QC3od<G^0Q?TNWNf%C1IO#$nG0i)s
zDKPCLd0jaj7gF)HYMs`6nFS<R!S*U7)bxn`eg+Ms<;3gbYreYsmayz82><VY|8Mx}
z)pLGx_ce<qPBAmb$h?oFP(1R&=CPCLeA2tM;Eca0H3GxfyBVak))Mp06k~0h9DQr3
zNd-{}sgG%?PpCenpC;o>r^N-fwMDB@&XxaOENKR4kLbuMUf<hyyjDTy`~qe`vhJSq
zwB!DG;_i?LVX_~Y2@-8jdCF&AJ(8+*mh9+Q1T^sN-D~u;^3VU`3&QT2d|1c|32PW%
z5aI-Tuhl^ircbYow(=DvM$$C#`*cS>yyDez;Fz-uc)^Q3bQ8aT_^PNv^82X-HUquY
z_$s-0WKoSuDNSfI1Ikvx4WNieGwAeS=T$*kgVnLCMqMkpr;su!7IILMB(fBC)5H`5
z(>{>b#Od^qj<wROUed9h(IR5q<_f%ieakmj&-u;GR~$V8_uswvJumO>Sf2kqNi#=|
zl%QyUwFpH6F-%^=$mc>$BPys`%0jNiraez0#z4u5oUIyG=u%sJ>42KbT7~Qjp}890
z&y<#*7Si!#sXTe5b!s~+(!0^JotAa)*&1fodgYAH-a^X~n5T*Ll(;{xP$r~StklWR
z5lbD)N!`fRo6YJq1vK*Y?JdjU#DD&)U%+lpl9NSeX&@;iKqh4egjtYD*gvtZX}UlT
zfh8)d6f1gM$Nsop^(X?MC9~!ev&=-93A)&NI4^B!>EKF!FkYiw&79!n2qai-BEL2&
z#-saX5N#>W_4SCLDRWGB=%i>gr5;t{WHb@xNJ*LHcw#N4<fp97awM*J8n5S^o`Go>
zDVq7_+pqaux#IKkgzNpDD1qPUw|skg&HwN}|DL6sxLsb51@aszi&zxB7%jFj2dsmg
zI~{C(MwK}yth?TXmdsKzK|O&iL^lXnq`H>kX`wccvtUVj`u=Is@p?E>w2;!;<foED
z>*{S1tf5@caRGKNcqwY<o_1^JOE%*s9<{v|QP!j!Rz;KXfRN1?%{+OSrc%~oslZ!1
z(>88ZAuIgu)oa#M;nzR=l&4Q-llL`~mqM0|PMHu2u@vU(NSuUeh9pYi#HlFvD=gt2
z-LI6_*_u#%@Z&B(B8Hh5B3d%FP)IBON=L`FG=|oq>DJt(@fO)kR$u{|f~DjaG5H%Y
zXE(0b#M$Vrq~uP5<jtMLYD)%zDNxqLdRkc3DnDjVFWge&?X%MqngPr70F$i?@cVDS
zg404duKeoeFL*XR<u=^$x-9(mkFU7@@{TY6^_SeG74|z=kB~EAo}G>@QY~WX$<P`4
z1VoB=I2Eh2rrAyzwP#<?ol>$#nMbTeZ1+gIv-qKG-5JaV{_HUfSOY&8a$a4jQ42-w
zJY|=WRD!hPy6XQcT0If=_h!}Km~_qIuMr`on3u?O%0LLw4skJosuZkPq<fv464Si<
z_3D8oXYSU8C1+B$LglsdwMsx@VD`a<5J|uqthUt}0~&>_nOaAwnoeqZ!UcbP6#ZL;
z?f&&WlS69vo`+)FX1v;dn(@rNC8e4Q8{^o-z*bMk!-8`uB+aZOQZC*hFjI<HL`p8C
zlv~Qr7ERYrN}EDS?>(x64)A)p=eN(lA|a##U*Emrb~$j8%qp2hGFI2D*6Go0?7q1I
z?#V^YBe103Gp-op-Qms{q^(p<cX;re9#{?DWsLlBB4t<3)H_NR_v~@eV4(^!x<y}^
zUB{HKO2=*g(=DO@p&fBlT8>j>inA%yrK?Tj^xUE@sBcs{Hsx?F${{7MMOR!wtk!Rf
z%MV(Owbm6>{8B}PV@}+sg~M`WP1bF2YwNDt-gtxXtlO}tQ3giGy@tBc#ABV``);CD
z>5gY>C)=-d=V%qa{h~D-KbJxxlS=OD9~sT&Qb=#0W6p$D+BN&hSHpa!@alNSZ@>K`
zr{%=5uG}sM?#seK6GaLsBx+r=dWP2{Se#H>RG-r5cx#*B=UHC}=WIB2A7ILVXJwRn
zR6KZY{hqGM!B2y(Q>uom?EA2~i7sdvXirl6V$PiL<oBXwBBf+<hB?l}Fe747m7eA8
zQO#=UD3#gQ=h>KLut?^dIpxBV3x$A&2qn6r>>aX1txi)Aa<+(tJPC)ga(j2={&XNG
z>)!Pt#;b~=g_N_^%5y^iq6E*qEyYuXJ7#U?K(}wTSoCETWLllKZS5QVVrwASIaSjU
zODWc!jE&D%x>r?l$>f}!mMsdcE$#NdO%XnO_R?r>E!Ls&t5?sNqIF#SW_iWSe9z%j
z$eLL5Y6a^<M4jh~$G<l>T&s7p-CJ>IlGCF|J>xgp-8Mw6^I2*60TX`9-#;B=R!ScE
zTXZwtX*E-Ad#3-b(s93XO25neIshVDZ=ouOhlpw+C745W<K%>h)tD<-c0;A(vkSJO
z=C-F!De+Q{+~>rpC@WqMG1k~jZ8f%nhkP!{`PEhAupas5<vp*LBTK6FCp$&qJ<yJq
zMs(7h_Z~5&_S1c%qv6tI`dG=O*hL?R1NCf@(6K3V(v+MKu{6c3N>*w0Mb$fBr<5ou
zo8ujGyViSj?Em2$t8>~@oR_rn`gq_Qc}++&xADXu?!V#nap4qJPUXa_^^S#2-)Pqn
z!3@YAp+<LeCv6q+C^}ru0%~_ee9qEhSM5<%!{FudbB3QL9S{6z7zV6?9}H_c0v;h;
zl@HMe-ZHA$pn8P-jpUz?9f#GPGx+J<Ob!Kds!FX1dRFsg2C%}@*}Q+}9cOft%-eW`
zV@Z@XbHA({*UX%wM>>0&ONi`(SjVDlUf>fe8q9MUCTX!g0WmL=Tts<)fbg7^Ru4=R
zFRD>fI+Ua?khgGM6L2e&&1#G<Koh!EOwqIi#2SfIm2;|v%)G=XOU^81Vbzs2Tk+gL
z0#!~SkeAGQ%63oj`U64SGrc{-mO{{L>y_C~fv@jha(el1xjmjZu8BiRtWr4b3ki!v
zo2@O2mm#rd7+kB8TF8>BrLxV*9!V4Qi)LzfL}>-#T@ZV}N7F$s{{8V|U5~WYasz|3
zLc9GZJ37`a71#{Xy>iCc@30wpOqtz`S2@FVoY~ELH&7Rn;uauJQ*HHl&gocONL)(c
zR#)zpmHQ>JQ$Y#drN{R7GzDT1kcf)1ltdP*VDtQR<jcb=p5MLVcs!6rEZa6ejN*9{
znyRH<5W8%~gH7nZh-%ZiPFw0gRrX+0_WIV`H1+s=b!!b7iRKMmOJ>b0E2#;?*=wgP
zYb33SrKe*tn(FCzJB-C01=8ISl4QQRea_vhg=JkyIdOYBv6Rdm3kaSf5eXt&$J2Ug
z_t;8PY0l3kRaclqvEqHjBmLCV8f`BGkxpTgwzhSgkE5gG?_l8gSh23BcJ|h;ebGDI
zozZg7SD~Y2K8TjJ1Sk>BLQZhK-?QIcBUs__;?b0v@p?yH>d^&HhCo6%qTH`5ck9Y5
zlckTx)d#Isn=J})o<QJkJ(9F=+DD$>-SOL3-*Pw{NGI!}_n~RsMk^k<p%a8)H5Rex
zQ9;PXYQEI<5nTS!(i}}ix>Tj$dq3AZP`}h3fF<-;dbI@II>W6Ny;)*$TCPf@vZ#`m
zOgd##eEnV$Wn0^B`n3#0Yi8pH?4&E-+`U4NzhgO_Sl7gthgTfR%3WTFF>o{Qo$t~N
zlOnYq-_|J&m!F!9g3Tps6*ufTZp~(-orA$U$Xc;&SAyU9_<o8!d2PSfsNBA|f5QbG
z*PpzF`<<;$Cu&pes->7SaEddknRS8d>plB<hfI5{&Xw)$q+YMEr{l8&!lHK$X<^AL
zua=e9IdPH%u@GgpsLUF@7rb_=DtD)a9F+WQ&&#iG`44~iibaGKqpc7&gUkEE^bIvb
zRW;hyy^pv-O6p?ybgJ>HP#stnL>Rg#UNeA&X0WYBpXiZXcp9~k)1gY}EJa!C`l)sD
zv{cbTS(UU}bEqQJ3hA8{t9y1%$02yCHdc%4m=|8$-%;)h-^4p&nE1PIpL4&SxObjB
zg<zqQadHEcHQO*ctdZ_)dIr_NWjs2U`W&R`A7`72no?P#)r^*P`@kEY*7?5!f9@Ct
ztbrd4so0_I&HK&-nBJP<{-O46J&3J^x4jqG>KPT?JiryT+l8=Wj*;uDy?Kl<!MJ);
zP^pbD8qUodRTZ{6Rtb1{cjQ^T=V#MX)04Ciq%Z|#4v{%TtJS21eT20veDmTtukLPH
z(!%wAV%|+$>lHC4?(Y|}7Y+T8v)H^br7<fjI4Q(Xy+xbRsp^s@|87ccfi1#Aswa5!
zC~2(;;N4;BLPbw$SX1)C)yBM=u$mQC>r7YMk2jAhn|C#M{!xvbsltPeHG33ZU?p>i
ziF_q6PsEdbZgRmjh!9-~+EQRT?|oe(pmv?BePycAs<?3OOV(1NYP4W>!KD_Au5Hra
zhzr5-r-I~)WG@I;#Y_7aFGXN~FT17^yU*6W7|y=$@|Z%&LX;i*eISI%u9{uksmgJm
zroclwwjmXgtKVP0y5)+%U+q3;l0Xh#amIzVDNKI$nBrGEY~3FYJb(V2q{@7A#m+;(
zmgD+i^y~kct+i|xs7igMctQ4Rz^O@w8@%)gDuvii)6$)x+X|pn7O!W-rbwrpSW00n
zg)m!`LoPMiDw#qovPFuWJ$$>8f)`k>bW9tEx?wGeav-5Ilo*H_nKFcIML(u#;yjPb
zjFeVF#Lk@JXG93mi!FE0+zs(ri`M}v5Vv|)wq~6NxEkUf4XHf0m$WT?_QsDJsZ{=U
zeoTf-TCUdCiwLFcQ6BG0%czo)0+D(0RQ{sWqQ4oe&YE|*1aT)#H-_#ZWGyU3_`|o)
znbyKTy80PYj1=(#{cezw5}0&C1yZuEXE`L6v~qtuz!aG0Yo^0u(T1Bn;GL*4h~`2T
zZLK3jJgpCtIW{MaE0k)f!y$NNLQhm`3T$i1T36MqK#i{ls~OPtcPUEF$&Cg|m~8K_
z*<(zDJF(n|Yms7XX+Pa#F2$NgY|+Xzey6~qi5K}`wJK)_Wo9bEE|{`BPdi`BY%-r&
zr`$kiyi`jyh1N9GZZDS2AW*!vpc%ko8*eqBTO`L8Ij9$V?+kBy??;NXRXnI7&{>)K
zHq{F{uJ+KZW;m9vnnab^>V-BwU(;gkMm6fcMp5F;aMUB4)t^m`;IeCNr%kn;slwOK
zUvQO`U(8ReIm@JIuBoyXZXKrRj+(-nS8iY4bANBCbZLs@D1;OU8A?j;{hW=Cl{H(s
zBEFBLMO`&Q*_QGFIxBFs__+*?H)$~gy+v55q$Wq!dX%}?UN4$iv$7VCK64!ecXnyf
zcI8^)<E$I5^Xi<-Mz{VDE8j0#NRl~8Mp+4^AQ7VaSgFP{u`@~QS_w$7=i+)pY4UqL
zS*lUpotJKeR;^8JOtz+u-4FS`y5CzL-A|B?2mUk+1J=L~29Ku2Wu{Hx^@GS{5~<o+
z_oDq<!`&-m+4>T7KUG*3GqQ+I%+pMmY8b8;_JtM>Tn%a#lCNi+sTZjV7IOdkaO7p!
z^TpFAUhL62+eIlvS9~oTxUicAj+qy)Uhw+<jwP>L3rtgFFFPbfj>nbqpm4~$(L%G;
zBf!ANW8DawF{b_hOq8)m1%JyT;j~NJtc<8o+t$>3m^@>`XUfTn<QFAO_MSzC6*Oh@
zr29>@Tg@gRbsnh}l~)Z|F2D+bMCKSXC7O^FwV>JlwXkyQe50ir?&MtYI$H;EC7=qt
zv^pfU1B%F2|HF(6YH?}O&7~?N?`E~znl~lp$EyK1I(9VU=%)W%e)^22JPzTk-FX$d
z+jrHsoeMT42_a1EcUJeRMQ8c5F2$lZ$~)*-v6>93%JaL|Tx;N0KmXjKGd(RfsBQ)<
zJi^HhoNLM)jz?a;d`Y~yV!GP1%LS5k;r#FzJrhcE+03b1-7ec>;SEQ_cJ6Izfyh>=
zs1m>yQBqua49$Rr3&Oc%)>K#u>?b_c%ILp%k={N%Y#T#0RN1**qg`tnXHW5nkSUzV
zRy?0T(21A?DXSZdqY3C`qt|+7w-c;cGhCtQ_o9Ew4ANVIb`{uB%v0{Qvb<eHC-^^H
zX=(TD=KU5r{xQ*Z%j)Z3H7EV|^LP4f<t7zBFW-}vQ8OS3fq9;(^;%j%@mf<R=~?td
zZ@As|RB4L$*Dqf%Erq}N_0NehTGRlbC0W#$1|oKD<Rm1NEEaWmJe}A-d%|?JC#FKw
zwHZPi4|xY!Z-#(AHp4n~nbFrAO?`T;h!+!ZW7{SOZH}ak;HuZp>b|u|EfJ#U?^raK
z)$&T&cK$9}x_|Nr9TBD0D{LkNzs^}1WaYxCR2wuvHCilcb-(suvKm`uV%9(kHt!fF
zQeH_Z`-TeLeB2=62D-N1-9i_Cws;}!x{)OKLus<1_}Ye!=f6LG3Ou<6>a{88CR;z;
zx)13%S6Q*j<{cfk>k95Tus^ErAc$o*Nbow2Mw%GaZUz;(I7ln+>PhVRi(9T1<(I#F
z%6^Wlu~1f!EEe`#3Q>Z0wi8lPZeJfb92blBijZeZL5pfl&d(O$9gjGx$-q<b2)34+
zTMbgM@r!q(8D(otNT0c?HAAr8oK@aDH(+JWq^y>rxw9IMn!VnZbzf-fxUFaXy&9=3
z8>BXOZ9<wK*6yBSk!;Ua%~=#DGhm@xc@-2d%~2%|kFxUAXhq#1<vAYesoIrXTL`%}
z)?LfOM2`xq=I9TlVfbm#@xY&kVZa*r;c%f)y68(4X3SS}ejvlfh?@bcDn=c1u712)
zw{e~VQ6^&81LmnKdZMV+uTmBA;<aB&**t_7?7n}cD&EiU56mU<kFURAlE5@soA*Gm
z^o!_qDHM(;i`aPe@&$*qQh?*KP!N(>i2rV$OkrHkJ0HCZCa}UzcGkMsJ+f9H(#EPT
zoK_)qj+1p~p_pf>DB2ne+v0^N1hrZfRq0nXFY0cNlmhpsW7k#CS0<Z`5rYXai$}z0
z^}5Q&zt)tABJ8~agL#TfK$lrs2(l2Ym}bfxmIEPL+u{)IunGrfoI%lGsa!P;rA3xi
zAsIFXNv&04wMnuU>ue95YShAvdRjkg0(BL(!=+}+Cg(g*413}2UC~<~GcJZ*y?;Op
z(C9yMNyn)a^TzsTV@Jo?_lLS*yq{<GVb2_Ih#^oyXvT@8g<O<1TPVAhs`S2{j%x(N
zi^G9kiRc$tYH^$>Sc8cuN~n>6rEpkJ+%5<H?!_N?b-bfwGv1a2r)1-ZF_`fs6+Q$x
zXAroWt~|xr3zJj4#k5%3tt)zn)gOtfNZI1cYA2+sU^L~y70zJgz0Anv9i&pq?jbb+
zI|Qo<sA|<}jE<8(6K#vBZj@MaAqvcXhFkPZO<$=gQ63#3k(>&rbY${0L3~^jo{qiJ
z4vQ`MVw;EEzAD;wc65oC9x=AnB(ptP$RtE-q5R1eiqH%TkM5Ih1E1~Zp9&pU^aqx1
zy4qv34DC1tOhNo0T8=yB=?PN^?4zaRsKRnu$wgVRD@&!cE|-rDx4jdDBEq+adqTGT
zcr{JzLLh{KO7Vij6KOfImO@U6`*h^*@4w{r=|m}56Na=}T&`F|uSIrld;VQ(D2O+-
zfc=$t`e3w#*Wff-w)D*4%^G5iMx)LgRkIas79q&g9j<laX6zELeHC{WiFi6J7Tp@X
zkwmc~_@Nl>sVd9j2Hk4B5aZjbN}6(W{NV;PGbB5&6D0&2PY_GzUMYws%w=a0XR-)I
z3P)Ki-7HKL55JLWlyj8IJVNxn;=H(&#-B^66$%CXi~~>&<dUq;4z<SC8z0Z3_({=n
zrAwbu@5WHh4XeOeplfzI3@$$h*^JvY`ZpD$@9Sx13Xzfn`~3~m)f0BIBS_$|T7!>u
zDJ`PCWRoweA+*_P+n5z8ChOhaAK51%{~~}+%ppNk%2bSR73H)ntfg>!y5rU9o<F?$
zBd^oSq7cK5-Mr_t9;q?&kDnrMSVA$nTGGO*iR)`C<{swB_Mccc<kCcSPsc~CBe!#{
zQZiX!3bPxjZ4GlyO$ZJ_kT5}+K?1wo4zzGO9!;)Mr4-{?wNZ^oHQZUD^H85-rQ@QU
zmX#>554CuEjm#{lWOs6zJTXs^QWAIhHM8yrL}U^sg;)eFkuXi(RHB;SU~60HjsiDW
zwTH^X8defLvbUIyP@_<#m(3q9LfK#D?3uP_`%|Lhf$xlAz#90$(4Q%r98+J^jagn8
zy8PS58vM|1<uED}*6q4xo(Mz<2otU+wGUUpG!=_{*xEB+jJNw;-OE>i<GOGk7NSI^
zIH7XFj^EOZ8|rQ6mvv=X7S_eOBN4<5XzPM{#C3>5e0Sl7bF@jDGP<>ycjHg<Mr;C%
z2_|MHwA8F~GeiBadRi-!Gv2G)_l4HQujrOuu>ET{&G}q8>>iUvoP=$F3LfR*5gq<;
zt1^+WG^Yq&DA$VPiU){ot5JXHl<l*pS)<i9CBJ$N?YV7hQlqi9%~Ab@m;2?RvE=ft
zw+T@{HfDQC$8bJBGy3sPItHWVpqntQb-7)#o`QMuCCtssTe4D(p2cZ+ixk-IQxAg&
zbnL>_F|FK(1(C_p74UQ%-BR_`yDTp2%Hee4<=tzJX#va-m$DZ<_xdj310Q9rVOO+l
zMq6LtS}1i}cNgEp>o}FoOtn$(rtG!foSvi^urMQ@n6a&<ARx3lTb2rF!f9W>WIL_x
zfA+a<ZT%V4R6^L!k|-wNTMAQENR5tz7kDjh7~3K;y5~tJ=u!n)UjcPBTlMUFcD(ed
z4WXyvrkrih?aeCBhbluq3drNX_X<L#<z5-CpI7CBX*tM@#0XO$=j;YkT*<P#ZuJK1
zUAJf6##ptP0EfJCw=Rf8LXChH5K%Km2ds`=N~D}PrInY5J1h8%*Op1rjAoSwen>>6
zu+@mF-M%)5N!{4}>elqDUV&5Rha3NDwVlq$!OhbmR;$mIRCDHtr*KuCYPGmfGgLVF
zpFeEX77bC)Xk&f;dI1eIqq79D+BVcQW1DNRqZtKH)%FL}YH6PH>unvR)h<+*E6xX4
zf3q_V^)di8y-UKQ&uU9~l=G!O|MiXE`?26G5Vbp~&gk{xZLNQ>Ie@9M*99F{xmoaN
zD+#CuVw{N6WGQDVR(L!YmYiGRvp%}MA444<&<t<KHE}Nq!p>{eo(N9Ecvs-OW|o{e
zEJqH@k^AMqsqS-+#{ZD0_y?;aUgn@=z|x#U2qqu!7+<av*eW+QeiOB`9Bxt7m3GA=
z$&g^3b6uXv#cIG+J*oKO#~7R@AM&I6*_5;C1fd$DX|hsnqGi%SAV?rfCaFbWMOAje
zjqb%9{~=sNqjr0xIh!u^m&eFQpf;yy=bp;>e$~gn`q-uFkly;a9~m7F{Am~ltbu=6
zT(r8a=}Bkj&Srr!VUw*^;iaN%Dv@v%XXY>i(HLq`iV8)Al&!Gj;&mw=OQSf4`4$Pr
z=5N;nyBPW9um4xv%z@v3`9FG5Wo0cVkd^zpBZp<-x36Dvd%7nS*zae;&7S-F6ZeM$
z`~Bp_lqVNN!&$rScb(*At72hmQx^t{L|BV?prkUO?t+_eZAwS&&u1#$76(^(i$!lJ
zS`sBE*44X6O6~tu2jPX3IX5yk1<1#jvqgY0)ec1#>TGp4tn-@}$pTSNC&bzstJNYa
zH6lZ-EpbL!wLr;8Q3~;bfeDsKBCNVt1RJ%8?v_@uB}#h4hEKv;9m8g24eh;0chFBx
z+kFcz=<scHTrR!_54U=)Yc~;?+z_3!we7CmGjlS9Y|XXo_!bRl2(&o+_P2K@3x(Ts
z&m4uH{qk>kHqU(Z^}n;OnLyEmsB$`;I4+5=ZlCjVx#v|nK@g_vD{j-u{c_@Fzh{n-
zL?QK>{s%3ddOetelnFtZ<7CkVRjR9nBQ92KGsA9|t41Ta__T-_Yo<Apa^Zf-b`J>@
zPv5GVO{PXcZP(Ch{B@dEHRBx~b*T#LTHJUQtdMTD2$UELQH0sMmQ`bu$<?8av0!wr
zY>LhH{Axn73(lT8TD#R1EgoI9t!^!LZ_$#)mFaq|;`e-W^hS@Lq!H^Kc$Ajwdyg0d
z-bTxI`cCA8)xvU{jit&cYu5;&oIQ6qr{#J^8xOoK3llS+|LPy{$!_M4U;P^r!5mI`
zp`?X1t(;Qg%i9;cOb1@A3rIkAI}WFn<Ke_E8ZX}O_vD;8=F@vUaJl@eNK<qzHFEQk
z)!o+=+DGd4A&LfzN*7hHHA8cHDaMZj1uFz?X~1Tnuc=&4TOI-3qSU$A$1Y42EiKxy
zrs5}6s|PfBq-Krs4v{FZ6G3t&mqIL)^-U<&;3Wr7n{0)it1)k@Th;YWZAIu)^7?gN
zzMzSJT8n{ORHn_*_lC5@^bayR{*-ZPQ)pu4C4tIWl_r;=ySJ-g<lhepQ6_e~YfdS1
zykD4QCJ<PY)A8y&S^T~VkJGWgBIdk&t|yS0r+@X&`DBjVe)X>jDHF@eD#9v(*LmUY
zwD9*YzU7ckq-^PR^Ddy8S=R%lB%lyyv3;A~Z_RQkl%=o_SIpDIB$L(B6^jsFYcyc2
zD^=&EjlOH&L=Z2{kv9XER0nf45LSb1c4w{0oqZVri$;VZVv(5_0p2Ylwi}W<5AW$%
zLLr7st4t<=X`1bOvo!)pB@y)+h(x04NhLMfY^kBenofjTc6OWoHm+)bQ<H7-x>}c2
zLUUB9GiOn&aa{EY=#Sk;7*E}cJZ_{N`18bNl<C004-b9A?{D~=t07wymu`oc>3Ox2
zl^t(ZW2Us?zqP<=l^r+!+}&LI*ExGVd3TjCUoq{UaP{O9ru{V<A_-+Fnd7>0TvtxX
zYRjlv+Rtvc0}14Ma6@x(Sv&0a^B8O%tUB^70!5^YM$j6WaWTKgCRe@BXX~I%>9wJi
z&8v2K39Zj!Q>V`^-&wOBFg&kCR?`x-bY8ovzzekSY&T-mj*5M)j>-_bDRp8qXH32p
zo;+LB`q?HKY^(kLoAv!XNk2Kf&!js#zT87Dcll*%o9P+}gcTEaI<5u9A6}z>E^_u>
zGkA+Bl}VMvgvLGBPd{b8euBiAj74mma^jR04$H!ET}hS`!+dpxgxG3Vg({d<I{Bc7
z?*dzm9`Z%l3a@r^><mnGK9Ai_sq|V`a#IBR3OMb*vx`V4d|o^o&K0B9D5ATuIj^0D
zYjj-c=}g*rRcUxTrFi)H?q2qFTV(0RS&f>P6ndV0)y<V|f=yt4vw5m5&relSeoPDa
zfIa-av>cqV=Axur8XvCqoy+Sqx@9xEob%&q6gQ{Fs*w_QTs{4S`RWNwJCeYu(elcD
zS~#X;k(?5k_B-&JZ(F1>oAcwn*P)+zjRuvociTSf2upCA`XwztU^QF%q4XUKJKndg
zPu=*k$vNF(o{x0<V<QLqke2FLGMa4s*~6^*u4=`N$oRPGwc5Gn=g~I4PNBDpw2qAD
zXHLI{+UUJ@f3tE3+u8DGiN4^~#lG7<^B$M+pG&81$2Q$OBOd+6w#Q1#Ty3&wxwzJw
z=@>K!t0)VF93sp+uAV+)zP>?YB#E$S;iQFA$sBWH$=T{%HQCy_i}&5}I=kt6Er4pb
zSZyusr|Jv!+^O^EV=XtwQ9sj*om`1tx;5!C8XoR@wl%(ZE|+VmXTNN8TymvfI{sQe
z9a<D@rS%s1NtHV;>GA?K8Vh_+*gZCqJeSkA432&*^$?WzdtL(r9YzG!z&|i9-3Ddj
zmK80O$3N!(%WP+zIR=Cq3Em0t3XzGNEd?m$t!;D4Ry(Tr_evEGp<{IQ@a|czVMRG9
z9N6=8|0%!z=l_ZqFTUmXzyB?V(><s4z}+cvTnon%IjI?}rfFjT>=Rz!FO>TuF<7zL
z<C-XH9k>#S{DC!el<K|An<DMw&(4u2two|Y4~O|ef=6H(Hig)qh3Du+Pd}n$6An5d
zxJj$(UeOxPUzbhV>b5U6K>+-Fp>=ACA)`^PSSMojY@!hMf%UkM*Fu;q<!cf54kf{I
zqynLsGEnRKNUaE6)3Z9oA?JjWTV&Wdr2F?V4evc9!arB^5^Kc;9hdIyy6<Z7=!h|+
zS!VMD=E9mXr<@595bHpewVXxiY<sd3hLA_-SV#z`0$<)P#N8+S<A3>I@#6U(`TM{B
zUpU^q;&8a*{%Dbu&yPp$(!!dR{nZtp|K%@u`G+sj<H^Gl1DedU9AmVI%=~`)!_+Wq
zYcCzG)?AK)cC=#V;hOTc3Cw*2lXYqBQI*|5K{S)|>h)vT>bi;-igntw4B&biT$p$K
zGp&S_M@%Zgx<RTEf^~C>A`mSVF?-Zv)`I5J91%&9%_c|l)@pHseqA9wS4!&M(VD_!
ziflE8;p<`_w)B=(a9pyzr+vjm@%x;u+Iu~}pBygv?_;#wzs~#9a$vIBNx5(;iD{06
zAg$PMY3FnIEY4_o+mlAxf`(|_1y%XY^LsR0^Pl{S|B@Hazvef8_qUvGpY!^3<Zw7}
zI4yj8JaNbivJ|dvuK45^Kj;3dKT=MM6N&8wkwS>loGI^Kva0$+DW23HYE7+R^HYls
zhGMCIYS9=0n9dGIrA8HPckun|D>Fyy_P1n9i>z@iRe25`>7aNzKWrfjos3gyxt3?B
zgszAgq5;98DuYLXh(s(Ltd1zKO0v(!xiw$YQjk*65JAMxrusOyatGoO9y#YnBjWoe
zH#Yw|+t6g4OIrT(LM>q5nHVnV*k67d9rp&2x-YA3EF_2`T!k6Y!0EW6F);0R=n}};
zx*;aT<T+701KMMGgpSppXABS{X~`T_d47M#_0^7l{4f6tzJ2};uYU7)+}+>udb#IS
z%G_t=;Pq>FMY-DT*nROSU%q(3tNSC9JCg!_4t>CuFu5>8wz_?1#yTz>Vr^4eJ=D$i
z;@n5o(rV2Lv>An@(zl-j1k$=tijp#9#JeI}gDA8{CU=9pe3tUa>lbv4nnPOsY_a;A
zRyS9(=^k0Kr=(7Rv_OD-R%6uV0s#n!ryUl6qOJ5op|nn^xeck1veyl(`>bx)R_K0<
zu6=yd4h%fRFklV*V0g4ud9M+@jl(?hHSJ&bONTHKWTFtfNO_^Qb=~sQyR@;c)1IBp
z=j`eHzV+Y^q@w)&m%k^i3;*oX&$;>JGk&H@eEA$+zv9KKmBYG{lAy&y?nAV?TwPs;
zf^1O?HNqh9o&vLvDHv0#46#KTRAwWbU9cL31i=EN-MCeSiOMW$fhj4jBC8j?&zDit
z7WMUr(o-ENT2<gnS+jj!+dstx{1DVauxnIRC#u#R5T@Xo&nCE?edavO%R;a_F>Q;=
za6!F|qbeiYo;@>u)lN}0RJ^-SANBE2DSia+de|Qs9;IX0EFo{3&ztDDdrt+TM509S
zj+!+^tUfb~I;h=lO53%zU;JWCn7>qhY?p`#S%u$x{RdPEfAu%di2EBp``KT>G*jZt
z?IE*hrf3A2+3l{F<{1s4(M&F;?AGyZLfD5ve9T(wQ^Y{=g2r2P<Tj-ifzeS$RiYo9
zjji^DC)wy`h~7S`XJ<ukvZ{i4dtS0phVA|B*|(^u;t9zvomwi1C#uw_48MK|)tg@{
zjcl}7>YBFWSI}JPuRrtp^g;zs)i__|%R|b46mRkE#)}J0;9V42BX??7y&n@EJ(Ax=
z%a7<3m$Y2>Nv%s`UVf)#54ln;l-4=6Gg4MsaGs<+Up_l9Tv7CHf4}?Y55!RTr{y!I
z-4j0h;xEWCQl^>1K`5($?1%xbt}^pHL#WE;MpLy-{d;S@^^<4$R-H_75gF7v<qB4e
zz0!sgZY}7(8L@<|evCT(ozrrw_g3#8+Nd~_WjdbRKg1@7bPUW%vUP42xBF^4Uj3{P
zuZyX)XB+BwphZbnYHfBj5zwszu&VRoO2TaS+KlPjv#%eLJ}%`lD)ddy>QDT{$A(77
z?K!`P)#ed8?q7DlBw%*kfYWky3KVNLVl~%X4l31=U{~6<U8Un~e{GVm|0`C9?zi9k
z4hbv2{`Eg$zW#*IfA&}C?gcWx=D56O$p_}JCr*LueL!}z%ln{N33K#1ZRwm3zU#4e
zA}WoZirKV9x0;V8EV>-wJWMQgwmFy7d0080<91HqNzV3q&dEBZ_YVPWe7AOntW$Yh
z$f#@E#aUm_aS&}O|5bJnwfdqG%o#4KUNS}7IT)oTA8sdH8;4XUp`G!4B$+mr*hOc-
z8J(V8ug$$J!Gq}y3_ONmz#8}mhIXgac2cRnoB#H2X)(z&PJ5zE<l==Kt;LQBIDYt*
zn$r!PU|egXlt->F?O&@!5m~eHumA1;$ctb8ihuSm|Id8#<Qbp*;xCzg_gnOjU-9>U
zc*X1Efi(-cSf|1%OvIY{g4tc9h0SAQ#oE}i>8y{VG6gYN!a7LTC<JdIUzOg@BIdO=
z(NZe{nNTQTp5dzT1vB<kmgVXPIVDQTNVMhB+zIDrnviP(7Y3xWC2cxVA4ZIKT!A>#
z(mze<bwiq8S44?ZBm~dT@9t6=WEJ|WOW4{5XH+wk;s&ivh!ELQG3<G`VL*gZq%#;4
zFTh=A)o<2qay^576xyxdj~QoqW!<kIO2@u;W|^5~0*w>|%4+urRnaPBO=PwGZr2f%
zOy}~vxc(+Mhk`|D7U949yMM#0<17BvfA&w<Up?iQfAf@Z{S`d_mam>4Ih90VPl%DH
zH<4+#Z%3Bs7SV|t)OxS=`{+SUj0>1Bti_aSZ{b~)dTC0mG(|Inj<(y?@^(15q0Wra
zwbQFwExD4tEl?eMr#1E~_h5T&_cH~@_j*Uis<h5YajvNewT4`7*Qf^b;2r5GRSZ6F
zu28gCVOsI~NhWJ<XP%NxxvxfneqHNmw_-XzFYScrO`xw3wi5XlD{{UXx)r}}_xUm7
zU9?>BNTBc~Eh|<pXOao65j!Qdlo2f!L0L3=eIxg}i}5ppr-(T{Z)dHc9_>LQMd5$?
z&A;Wayyk!Nul_l^t7rV>ub&_{&j>GH@Zwg<Yb1t+U6`1!B3$j;z@^ZbY01k)J-zpU
zuo+-{Ull17-HJblrv1{C`7i||`NUR4zRE1*wjPaU#2O^j4r5xfwXbhocWcD4Zv6Lx
z^*(%JW{gk|!Ky~+YE#R(P)Z<7n4zN7>*vPq618)rMFU0E){tRUs1cH-dy>^iP)hc?
zS9yEN-Db0ux}>X~E=@K#PigKyZ>LIsD!ln<K2xQCM9}j|_0#x3I_|$x={QJW3c)&2
zdiAxE3+tK*I)REs)fR1$|I!NR*?yO9zcr~^tm<kgR+p)0;eYtM|B2V9SNs?M`G3m(
z=?#DV<QME;e9P6Vm)sxzz?xohGkt?f;nQa~ERs0rV&Z;=u#dzjl#*<Fn)vvBaM9Jo
z%uG=W5To)T=M!4x#cJS-!Q^@BzIXoCXxRk&)}+N&utiHBmXmc__soNy7r&<{VW0Kq
z-e_1i|86NJ*GGSbRi_Cm<S<d@LOEfJIH~I)R`)AN@cN)eMGwuy0>vWna@pdiidwy_
z+HB=qKK6)J@g32J^75uhf5>VK{5fG5um*lGT-G#c=KDvOtbP7XSLzfI*Hgud18NaU
z@D5zAVI|G(+3-TYofy>w;}(I{51=1YYvxV{Rua#bui4+izyA+^%gxOVH_x7+M4mqT
zoL~IK-!NajX8r1GLYh%onBvT=-lA4vSyt8%tV7V5;`yGaYaqqlV_u3~Q#ZHDL@U$r
z84IocM4M(aD7hCX25+Kb9@AKbeBFK;D+;x3x+x{%56~FjmQJE)&ssM{rWxBR!&Ogv
z5YJTDgjxx~1a_1dqRDBsf4jQKuL89;MmJvDE1}gZ!aE<fRp_Oh&H$%%2WU4I>f4o}
zN~=lYv-5S`3@1GrJy@8*k49vEoQ~@+52eKS-qOC7f)K+@2ouT#bwAX|NiXnQ>nEL+
zC-%s=v(TGMI`)033M-lC>(|^IgfGAR1N;5VJO@%%uC71jvoC%{m_OkzU9sqiWx6GZ
z5VE$^+nh2bC{tDb{D$u%W;Go{KqaF=D7BZlVrK$dM38u1go;Hn6gNs$S~ED2s+|c;
z(TzLmUUbf-TVJ2xQg^M-r`2<;+NERo-oDt<kJGW8wWU<UaQ#MU&qs@QKMGSbTA1Q&
z^+t+9uE_`4&bbmcHB_a5n9{4Vty4`!e@q!R5AdeOiqcFZXXy4&jaGS6@Zs5N{D>m3
z9t`z+rDa>f7-wRf5t*q`8#Qt=c_dxdLXnJAgAMA>R{P8Owd){*8F!1KC6HBl$v1p+
zAB~oGGt(rjsj!=$@X05CiG-)zr5%YAa(zdcm4x86(Mf9}=g{mwQg8Pj^EP9yOBga?
z?kMNF63?9@_@0t#wAzX*+v~;<tSv;aLg+z^&xsqcFhdtPfj4YHYt$M@s0wHopbTfJ
zRa@yKZJ)cO<?Xp|{Mt}})Q(rH;pP0<uV)vvx}%oHTgX_qGhD4N9@SkmqhwF7(lcRU
z<Eu8{%gECjm3zMbdSSyX656f%_PPEUcF{mTBBtCs;Wo${H}6~NxY91${udEq+!4Y=
zacU{G(IAlW>K*iw8F{*du_`@*H`s7Pwy{e;t1b7;4LHh+<4bN{?fLD0_zl<lJy-jg
zWi3py=h@BAk@A$=`x8k|geem^5M*tlR_)r?=J^3fa#7HrXfVTzU@87uqaXeH604D^
zQLP)L3$<K@6qmo95Q}027^BJ50>RT`V=x13s761ROFc-@wy<iOQ!iU@J_l`v=iD4L
z+Zk$fLPy6{BZ4Zk6!EeQ(fNK6g4%pS%1w)^DWJ8`yk@qVm(9B_t~4}rC1<>)i;{2d
zU_X?!e*VlG*uSTpY<>QKwZ5jh$@&97c?<*Az&{wOfbfWj+G%!gMqgJc)Vfbv^VRJ@
zEM6xkdzgJ+__J^AYy-OBcHRAJUVFg8gjdbHTA#DKSAO^V-}A{QpYVCf#A(OX^;16k
z{1=4jDPMnkM2nE(i77zz`b+^>*9@U>y`Rs5mAXu|cS7s9rE~@?tM}ChMs1GRF2%s#
zJ(Kl%c2|{Ns;@@3+7@iruHOw(JBsPHgmNwjZ42sm>msjODJ}ncReVc1PstMTP<727
z&MvK1Ny#*!w%YevA$WZ(HE0votD)~~b>GPO>Ss5`Tn!1^q9_*wf3xKH9{c@cMXlHL
z=*-?ZuNs3>#TEAm%0vh=9xr5dj<i^(El)wJi3&CKuI=5o&~be>(maHP+x&tTr^q+o
ze#6y%$8HzdUq55Fzv0;@KSSaZK6z0%raSIqMkPVcR`Wuvj)W9t+MO3td#`M={`?b<
z2ow`3<xGim<5ktH*|uM!pZb}WAnfml;0kX?%l#36)2s-Hp3`xAf1_pHw!Du+-$=)s
zfa^3SunDi*XF}_+Xz77P)Qr&T36Z6yoPmXr))lBxl-=`hzjvFj?upLwru95~_(<B~
zNWH3bw|vWkeGuL7Bg6Mf%f3NFn2<1oOukPN)qrJbXJu0m>7%f=sOg6T@_Q+eL5q-4
zUh7L<o+iHeW~1e4_k?M>;>nZGK(6`Z7FIn{t|BQcoH%+UGNfcivz@nLbCUQVeq2-F
zEV9gA-vT8o|JnAc6+|zc;e<wTmX25HIYO9>mc=>-8b8j>fORQLZKV2;mK$UC4=Ni!
zK1+%3Cr7{aZSgCuS94ORb;Q&<rWG+`f9pC*q0|U2H$vv(P6=wHC})J!H*h#Vew-Gi
zMv$H>%G>r)lMT<(IrWXSygiE_3o&2J;d0n|@%uaIxc^f9D}<R4XNm-~IhE36wUo0r
z*KwZQ&g6o1{*~^Yp%bnAoNhjsLej$Ra>t9imwfeyKl1eEil<Na$aF&pSKRDBg>u8E
z*LR%qfn#_{AT?u;;6~w6um2q&K(HEKBDk!9Wk<Ln${X!&_t*%~0y)K<KMK{rXj{#q
zJVlg9O{c9jyTrfe4kdpK&48<yJG}p8Lp3V3zkO`leP63=?DA9nwTSb^V1}m{BBGvV
zDA{Vt`g5}g&F##v#D?lv>;~EfAwHifv(+h9$-Lj3mFac`5}IL5y(6-J@*-NZTkP|#
zH}I3jFklV*$3w>ezU$YCo&r!T7=6{^T?5U_SEB;73W{&L>6^yl0{(qjc%2TsI=o^@
zCtlyaBuq0BC+?4MT9ljXC&VeD%Sy@z4!5siP0Uz<^(NGI%*AWFzPGo#s%o){kba$-
zy0yiE)FO<E6#LfdJY`eVJVmdDIm@;}s*p+{7f*9+0<CV!sE_}0{VsWI{oU&O)a%sc
zwNG2YTPQj;=DuCX=d86fb1$e^H^-ut^|?W_Y=SUVq3+jFzYC+QckMk&f5ImIDd4^7
zxEDHvD#nPX@hIYjLzT4@R?S|6En)GBHM-#)x9P`H$JR-HiNJkI{N|hA@nk>qY(Jsl
z3T4MB32Rc~G&9Q;Q`xhwnNvPcQYI*E95wN+H$;DENu;WF)zZXvjjcR%4MyBraGxs~
zP-{kkNNltsbQwm~V_LMG5%u*4bllBl8{ft9>YM1;%-+^0Au_pmaV~Tly@Z~Qn}M!Y
z`6)htts~{b{iWTR-qT?HeH*o$MPSi=P8f82fj@zkY3cHol9saZ-%86X9vQ7p%a2(T
zFY$(t)QFtNXgTSQ%*3(^YchHYMA8yjDV#V^)?|@*MUA&f1XZu0`9AZ1WIz(yuBvB&
zc(g`qwBWR}6@z|cj+J197U9t%F`B{YoHq6ByRK}Xt<}Zl>-Q=xKbTHW%j~^nQ(WN^
zEt~)W0wlP*yIXJ=+}+(B0u1gJ+}+)M7~Fz81b0Yqch@`T)cthcKk$CI{dHGO?b`Ls
z)7`6AuN7#m+QDpvTV~F4k)U+3?Pcq0aCe({&MezseXx*qq||Vj!rilWvo9p7;{aFV
z2LZh$&?^QzLz5@0GNKZ}@wA(P&QU+Ar>1Q%d6K|76ggYUU*_o>ymn`4slT7}oxIdZ
z+y4Ejgtd_boVISGs_CO;WZ-d{ok&O_m)y$3lEwuXwPe+>JZB32yh^5uo6n&|5&h|P
zlYu2pUo<b+4qCOf`<5k=MW;%CE2J<y*dhA``GeoqjRGGz%>55*R`r_yd17qQ;`gz0
zF{rp<m4Tv((CuJkdlq%#a(z+d%%p~R{wYu@>q>ElOZR$Q7d97B=i7+8pqMuQmLs%~
z{(sZakEuf8t(R=i2^q4<N98Wx1&@W<xG5iK-Ooc0JP6Y`A;Pm0l@W98`EBuvhy&rj
zMiL<#LMLG)ShQ-%dK!QF%oJe2n|ee&(_ys|^4^C_FkB^|&{e)hMhG0vg|tzvq28H1
zwrtgR9ej6aV5?Y4*EQ8y;@#+5nmX8)4&gvlsEa0nGfB52Ev9cXg*r;aSLJo7+0KT_
zH+aEc`MF}bh3%YY>D$gk-f<DeVlTvo?ft#V-F>?^PQ+`J>@<9(hP&e5r@V9C=8n`B
zlmFd&!0|RF5!xC^o0A>MF_~}@&hy8H0;aywhSJ0@6Zx;fZhhu#&wqm_BwP-?v+Ho}
zGb-A0_|#>xf^U0*`j_}v{Ws(+3k^#iw)$Q$0;}a~c+!UG36~Bm{ndFl&sP}Z-&7Ij
zkH5^wsHbe+e>$%2xT)~%@elCt1|n@iGrpuvtvZyug4lHK^R~T3yT>DKi_ptE4oMk_
zm(XRuo8^pBYNatZ1&wCNrNZdw7++oNTrJ(c4k_@pSAbh^qZw6-#HGP)YS4{9j+V2n
z7$Jx0BbA5^LEYhnx6QX(D>K(-Sgt&k7|R@|ZBoW8f|=Zty*Awsm5==u-{-*#t2-5U
z-ygpvE#6u|uLl7`(%sa&o9Q)WpE|Dk{A70KCV#5-+@^XjlG<~Y1o7!lcxK4_YhCXk
z&}>ztj`g&VN92)II7v{`SWFxNsZem{{=xe^I?sFVi#>uH8eVcbv&RCIsEXyc_dBya
z5Gy+i?n<cpILDG*J}1wf8GV_(%7B)qD{%u>XYJ&V;@=H1-t->3ACb#Foif-uXGo?P
z5ESpP-$coB!|m~#wtg&%(>b|QemuOj-2GN0{*N#dd1#C4;#Z1H_iz9jdUBG^#L!dV
z^9BiP07Tx*l(w_8XJ_U8SJer<0pWR@5Rqq>*<!MG9JLxdo+gL^Me|<ypNv+jFdaa_
zFUY%KG<oXrFQ2isNM!PMZt@=$Y@1?<J79|!Pl-}tVqpm<mrFvBgr%N1zG{^@D9MXL
zYFIUro;bxkl#b7Jy`)6k4bW*7b-lT$K{6D9GQi)P%T%rcFIMQ?eC0r^;Sq(nq(+#z
z1ZX^7&K|TnEM?HFdkhnXOqe#?cwZhkN|8E06Zxw9um1#_3_hUy9_Rmq+|LfY9~$tV
z6NIS}!})u9D1TA=T%ATjY`xbyywBqhZDrPpv67?Z5b>~qybV#IWtxE?m_$56U2y!2
zli-Ez$9abLt?3_=+{5=MlMYUA;D9N>v!qw!X3AD;9xzsa_O#l{{_AW!CANyy_yl$Y
z|ERT8l$$d1%!O5gFG^+z#t2;x`Z^$%^Th|yW8{X8dJ;cRZ>%%F3QXI-@prV_2RR@&
zd4A)4?<NuRcSFYmunX7C=LzCHizYRBMajE*z`(yiBqsT**Eeg{H;J_u2i~q=%u4kE
z2QVEe`Xl@s{l2kg-x&9rHjY_s{}j30cQJeHB+0jNJl}Hg)PU_~2*g{ad=|o+Y4alI
z$@J}HT(8t@*lY7{HbNGHs%1=mnh1^;r!eW@)eQ%mpX5%CTDNhrRY-7meqYM6YBPUU
zA*UTA{G=qta1E=ok<Hs8?>6Ip9wEEn7A*LD{2LBX?cq+c?6+WI&Ve!QFjB1_5$hf%
zs##u^3gNd<z*qKiDJ(Zy{Y9?Jy}41$yIjG}yL!Isdvk`U!eA39MyxF{D>DvA8yh5S
zjL8`j8&5V%s=BplPCd<e%Ws4jRsRf+X#fs&dG&5b3%XgRo4j(!q`2friu3S^9y6&Q
z3^#oMK2ZsA((z>4UG($>$0{cYt-8N9v#!#}u4D@zuh6PMi7<ML1ZKbbpoz!=i7rsB
z$5s@_pRq0-7-~R^8DGxiZO>rOCR>NyW&#5r{sXl8@E=ZhaOavfodYO-MWD!8)vo7z
zAj}5$&0Vyg?j5X)(!F<d!D^?t3MlUrk^+Nl6n2wG%~`S3(8QwfGEK$ICAD%yFvNqr
zj@69mJzo9Fl`U1isW|`<pxUgh&PkmJ4JwnPSz}h~M^v!8;LWLjTVkgYn~Hx<<5|Zi
zCp$<m(*I{^4;7pHW_nXkpY_0NN2_(-rW}cx1bmP`fV;k9gBBLl?O;L}FTqKO!uOZt
zRyd)Wt1;JAUAO>F=22`W#TK0F+Ac6|>yOhgrqoMl%+5osT`xr0EL}4zAImKm^i#>(
zPHVE1;H5^A6Fa-^Mjs40gAi4-mkS0K$Ykdchh=IGZUt$TH%>#_%WSSRn0?mHYPzXX
zakGkZq!+w(T5?^PZ^06$uWX`jD<RtspPK4jFc*Vu^BJ2Yxy*}Zor?C)e^-jDiXK{P
zm(zAQ7Vc0fLC<f>O+Ge%eP_(A;%3pCU0pNVw|I$v2U&Vs!%sQ>hWq3ErXM7!;c>hV
zVOh>blci~Tu=DENPZq}galVYr6=!dwaMl(^D8QhPtz`*~c+>k~aRu0y2;HLm)*9PG
z2>k|of?$)wd6mERqGH*S!gFZLv4~Mqy^7R&cD>QHfL(Uwy_CEVru^hasHjdBbu;qk
z%Cu;&ex^R|z)9!+(4mHUh5Td()kX!8O9*fA+w)<|R{h`l61Z`$rE}+tFsk9~JX;}J
z9~449R1O|Gl`23+Mc=)Mrsq{JBax(Kc|7Fjy!}cr42Y3^uFV|fNdQ4cq1&nkG)=bL
zHo$0C>Ge=4rbXOkvWDA#j9YVT#~n$K-GJC}7lU=f<t<TKI!4oig9ZG&hZXNI2#@KV
zU1A&?wT5x^h!LnaJTK5L<5Do&Fz>nC!}zWbh1~_%sMzxw+I#YAhCN8oe0$o&c{VLi
zI-0tklzDLD^~;ldPAb_6WdBd85eGd>SVA{(le+;N#HgDPRb#W&NS&9|&%syRG;BTk
zuA(f{jK{x!Bu8*z5Wg>#EoAA>nI!&$kP(?*l9?i|X*$Mfc+z1XVejb*GfHPEH2_-{
zLp58KF8P%xtTsUEC-VxID1D@zMBXf7J%?b!+lkN+5OY$4-#2n`_J^hUXw?FWON1V;
z@sE(d4#f0e0tyZT4MaO^r|GK2dP8gZ8r>`ThVXQh2}BIQ`r<~)d4v><3OrUIW;9QV
zO1PK_WQ8k~HyuqJ1s!OEA^WiP(*FfpbC&qUhW)4CZ}pGUf3~=gziEc&CVGS!8zrU!
zZI%+{ftqas5Gb?Np)6gHY(^yhZC9^#4!c^>{j`?KK}9%2TLDoA&B-~=<kQy#h<Mu%
z7x8}!S6K72*nj{OYWe`Wh-;!4aeYNQ4n4_M*XfErd(6zLV%;w*nKDh33s}wJa$gmO
zX>8eenYxq=mMCE;w06VkHt`6-^MF|yd(-qN4pHreQBigsb>3JemL-?f+B+`ZDW<`-
z^iCF+vqKC<1o}O2Np23Vh!4?5Hj)4P#)`1Mso=T{>*bHtHky0@5?|>LQ4q$uYID#=
zySDneb&wFWkkHn2<d^U>QuRNt0jYI~@xW~$FBxT$d+n?IO!Oo?cL(+9KEwpWy_&M+
z8(~=h%ua+?_v9wOJ3pOHr-i;MBtu!kHZa0jdMh`#Rry1<uT96yOY2K_VXNgEzB}0U
z6{VGqS}8TEfD(Nj5adWnlH$?##cuGil4$_Zy`J?PI<>HFYB{NQ(W#Ael=EQ6jeC)-
z+DTTqq=hGXlD1TzA1_vXOPf1aYn)Z&bsf`gH4l;tc};i>?|RBdS|{xiVpOG`)0ybB
zbrd!R!6pkN)8?l)TgQm2UCPUqQWlP%Fi<d4DX%&lFFc$Z$rYC5TugE`$z;^_o{wQ^
z<h%eT!r{9(;T>Y5h*4Nkk?)@*XDO?%q(x}<!ODtBG&$>saH17(Av7!a3B@KG9gK6r
z>#UkR7Pqa%3$qylw=Ch}oUkE0S_0KrpEJ%FWMsXl39HhWS$XfNacW}C98v1-TB$;4
z5-=cww)V(k+&h@ovEk2*N&0f`w3MFUJB{70Z_Y2bMaz0N0z?SQae|}nmuEuWyzlIN
z{~gW7U)*3=t4JJVd{WU72~A|L*gd;JemY)@n-9}#=_qNI#bV5!O`s$k^(2Kc?#XKH
z&i$(V)l-ov!xW7IfJEgNR5BoqO{Lo&OJr&4XUE@_ui3TYr?TV)Q_{vhMnyAzuV2vI
zy@{kXIj1U63}d?4T2>>_Kl{zpw6nD@5<v&^FjJG+qlVz>UQYe<1<c(lEhw(OokK$U
zQh`Q(p=?1h@h@zpe8WYI+E!gGZ|inmtbDrq1{tktGdMF*MmFn2$j>wMq13d+K?_{4
zppKaP{&OPZWiBwSrKS9Q<;<eCwmLX-h~AmDno=8V2FCNS5-c(e#p?ZR$F@v<py=I$
zpQ&tQ&7my%I&G<yGV*+ok=Hn*R#mYu*O1=+o02e3+tfVjw~~(RHC;7YaRM`A`$Bn5
z`Aofd!Vqt##MT{(Z*(j@fKzMlJX-0(Y<Y1vq3M+PCTnYTaaI3#vP$5Ry0FOHC>EV{
zN&>x23FMCI`$4b&cUI;sp>0;d{RFi-d&YT@u;XpuZb1wlx0GJik^!NBq(OM}Tq?5Z
z!g`0qqV<at*Rvql?xL#dAs>yKw_?k?A7J&;<Iz<i@3wlVoq23%-5BxBB;w5A>+H$B
zhnQOz?7}jpo@aGfA(yTWQ>4v3eDx~-Ka)RR3ghQPyt?RL9m-W)+@#^>>*G->XS!6~
zghFb#iaz^w1#}+};d`Jnc{n2tu&__^QL&auT)oW<!#Un+KlMJ1J9g?qn7t@zG4vaZ
z6z2+v(}rDCDSZyPt!<fKe^+3D%_u0-_@G@-+(^+V_?RagiTX`}>$qvR(W8e$>@2x@
zSygt<m2VA%imrqm`06|a(GMYiokujpy8M3c?lX4j<V)|Q`0BRwg#q=x^;POhN+jS>
z)-oo7V&-zB4l=&kQ*7`QZrEb2v@i64T+p9kwcnt6a#~-cX5qY#MK9r*iBI3Aq5s<!
z1h%*Ic7>S0{k;N$MBBBPkt(1L;r^*8Ik<~4^M9$#=)ca~aUKK_gm>7MzI|UH)U-9q
zBbOc9T}nVnMoF%wuc#KYV8HPnyuQOMZtZ&>iQi<2*+!Lua`?pdnTeh%+;+2g>@VS@
zPh<l^m!P4zE{vKiq41yGJD1lNpmX4S`B!cYak9un=$NNP1?rOHNOf9!Tid0tk9tfh
z$(7M`cBq6wB~o;?w<=#u^ONtva=i_2vlxGe)13bj<2n4P_YcJ(cgoaGTQHk0reH#v
zc|IhL`)dzekJ83c_RLGcerWM|x?A$OKVtT9cA1<j<GZInU5cPIceuc2H~Jzz7HQ)p
zJj^318;UkXF1^XNizNntZ2)fon5Cl11ZN9>tnNKm+{11m@-lsuNlR3U!&~_taKEBd
zw|UCP*-q~99Ns*VHoLJjb_IxT6Ia&wvKBpmQD|Dni*`+QxpY`vs<!}+LN42#bm`8~
zW}6BTBE$ysU$~u?*>Z@`Wq@m|r?R-v&be;+@_vh&ODH>QN-1j*|G6&9d(Y0I)=V}7
z^Or?%VxC{HqfG{sv?of7Q9<bAh^}Rc^ojP3_x+w$)7G8Qg&2vpxig0lJi5B&$f7$H
zS=Gf>rc`MbQ0lRZiVOx<0ucZh&8qWd=)tsa;+G1}tF;N^`=1d0vWb-Jo&EKUPIVe2
zI>UE)^Z<|yg|~PC`7~sm-;K0uGM*QhXty{IT#Wjmb0KRJe6{yWi(NPABW>}KZbN04
zrHEa3f)>pnAu-IEkXBsTz|e`R9f6*gYy4cl=a3U>O`m7ro)>njAtDdu;)~jNHVB!W
z(t=7`i=I$3SaCXvIYtA!IR$UQ5z5&T<+vDbt*3cOCqB*Lp<H6k;uR_gKfaP7Dtu`W
zjf$a?$NQ?Xrea)UzAMq-QkZ4WiPvH9R9$yir9-`ewaL=m&rRx~vE=e!!u;_!TX2pj
z$G0lpOQsWzuIaX^QcZQ=f$8;&;Alyh#XRI+pKbq2LVwM5BM(g0P?>_E5_*kX3R#8e
zQ4!P=_QHqSkMPQ2f0^sM(S&nb8Us&nmNv79)2IKCE=VuJ9#Gk+ZKWC?5MdOkwBmWt
z)@uiB0^yg?S2<Tvsz15Kk7R<HVBNoOhNyWCyNbiJl9?QZ(A8ECo$-l;;6QChJobi4
zDw_^a+L|_}_F<1;fVF_>=EAh32y+Bk&bBfP1C_6lXed!wE{Z>qQX$@3cKYFmsYUl3
zRRqdaC<V_s$(w=SwrnZP-6u(URo2-C*ydK9m8Xs#6*;)lyVt)KFvX{m5U}(2jE8uB
z<^k=80;nX6XNBRL<HpF~(b)s(+e)SEfYk=YZHvBMhP<@^7x=7V#_hgJO%2CV`?yG6
zbj&6%>1nKcW!~=w?4ngOJVRV=sv6K(;?}{nCzzBd+hz$I3f?KPI}`&v4zTu9M?YYQ
zcgTTuu+T#NI%y(Klx5ktT@Xv(j6fkmAqvK_<Uh_NyEy`?XCB#GnlE~7Zh^>cSr@jx
z?2O~*Aw$q@)ifEj<qLKge~>5<10%Gdz1s>yMtn!l;b*z1_uj1AFepO+XO&c>t%6z|
z2V<U@-Vb$$U%ow@;BBDp&jceFd<=Em!bYX{VlG+TTR1nTr{e9Bnt-u)#1dWks#pep
zNEl}pH?DaKurRFBNgNDAI*rQArWTTUnSR$5po>a{A2!_1Ugfvfc%a_A%0}!s;E%A5
zhAc$xFZ7{{?UP7zx%rib80P0V5~QCp5sro%%B2xx37m}OwWm}r4mAb&;v+n0eG`6Y
zR6II5^iO)JeiI(A@Pv-8_}0IY@glUIJvF*_m(&7P5Nq6pTEh+MP^sCIB@xV<Om1p^
zdvj2$WRsg!&2ft-`7Y{l1*fV(FYW;K5ag88OW}2TEcLXDcq&Yzu{_lIy?g7*5Gd(v
z;39E%lsH^rMf7r+N;uUDa37e8hJ-!^SHpFa|L0&2e|{he_a;_Rz)*MF7VO9ST(B^A
z<%`2O^+ijyE+rW2JZ<3YErI^)X^9(}Dp?;!HA)fI$SB!RkeJ2jDT;!}W8HO5XxlMq
zWj|ss=N34onN6i8l0MrNhcCDcVd1oAEOm;~e!OBhWk+c(r<`?q%#40673J%f$oaNa
z@nCp}EK@x}`B6ab*lwTypPB%(vt$!F%`CXl#RAZygN;s%)+!(uP7q2hbPSW6T|vzp
zNH5&&$<PyauvjdUUEC$B9MD8lha7rxGT_h>U=bn*&u$<w+3rWFx?Tul;WqI*L6qHq
zb9f&P$#yNvSs^J!3Ln+MGRq)a64~JqmbZ2zO6e3SreH<^$$%Q#33|z`4dwKW#QbDr
zesRXuuJv=3g#WFw5f#HteC(u5nAb*6<@%aO6`T*d(Kqs2$#6s+g+VgT!ljJh%=W^w
z8~<a|PBVS^h7+r&V+->ZAy1$rsZ0!j!&IJ@gp^I%>^q-E1%j(J!f9CHtaAyUHQ2;D
z5<e%`P?xCF`N{$%j-zsGkr6O~Lq@98m6_l4*j|QbfKv~Eml!AYXgE3Z<jAdAd!%O!
zGrDNP)4&F=(~YDpz8AbL5y-TQm(1T5ArFREJ_i1oae_}(%5t$VZ}Lj8rZmJjfhGDU
z;MgIuBep3fCCnI$+dLu!vzYDS>Tl-mJIQ}BV^{H_sunei=@OpG5W>Pf7$NN@7RSS1
zRMCQ_B?PbrncpEt{is6%&A1Gg9rvWXh1o~S<JEAI#`w|Y-7z3oC77U1hF3eFMw#5v
zs<J;ujBi}zo7y~IU_I%5Ain3UUbL3&1SkOnR0g7V4O;3YWFw|~=0U{s^Cz<sPX<;Y
z{2tH$cj?yY_jJ3;Jc2;g*DUS1#oRWKNCljr|2FOs_D@<C3)T8Kv-8Ei@wL#S<6l`n
z*bDMQcaEeKGH8OND2Ccx#aG2CX$Lr!)H>(#+H1Z1%6sm@-N}_t4(DvF_o!$z=Cqh)
zB<`S+oNM8i_%M@`8!;W34>-SZ?If3Bd%W{|B;#^Gt6}~UMUrBQQP#lVM83kPHci%f
zA+xb;LaDAQ#xlR_TN@aCmCk{aQhGo@G-eN0wlrjV2}btew?fF65hiOYTN)_2h>3km
z-L?!qXrjVgMAM%8zRWsENJ?k54Fm;^_I_DIAw!c-S|1df1!zHw3?fjDoBPH;;Kq9u
zs}*|{P4k{zZIT&>#l)(&BzNGgYsAb-3|nX+&meFgEGR^8t2E(O6k#XVEC9LefD8)5
zZXcB++`Nm3yl!<)C8qdV^@mJQ00)V<KF@=uz&I2K>4n5D)pUX3<}F^zW5Abpg$K4~
z)1Gk0Ja(N$Le|r86=C+_eu!o<r6JW?l|hsR0-_7U00Rc543pa@SN-ly-t4icHAVTf
zwPYFgZC`Ae<~%(+qBb}5volSs!&44|2qyr{Y-YStX06rj@<MZvL*i;-u}RFjTV_-P
zmaBhuYE(F!$|8ngQZJX+S?%;<h5cXh5d3smX$tvW0(d-~n;blOnvo*Z;0vdx*kZCi
zIg%&OystD3T4LQY!s!tpYB~DkQ4bvhWgc9i+*(>=*a)!*Ike}}JYxPI?0=T^Y%MRJ
zS6?P-9+E^3qD2a5wAsazM!zO7ifJ=mGl<!`sUXNDjna{7MW9;e*xu#<vt6iQrlm))
z)OCc2O~o3oHuQ2V=ce9K=VWy}15AGP4_orFToV5m0x#;#cu+XGP+#S+%MHB{-6_Gf
z_Cn={hM;_mRx36mT4%SjFpFalEsgEW_hlS?Y^y6m*nRATMqt)A<hs@LQ*v&wtX<G)
zNnbgk`(``3y}Px+TGaFirKUg9s<=Kdl@l?WZq0VXl?Y?R{D$UeaEF+)UeH1fl0>hX
z^LD+9Ov%xYKA>!_TnjK{Ez1p57*X$-f>WMj92f=47oyinja(j$+8na7a^SkSo1XNE
zpKjOcmgk&0EV`uKdslm_9os+EV|t_^?!DT(Z}QJy6z?=H>B|dcWp8+obG;-@5<!1_
zS}wF>T3EXjDC2L#tXe}uw$kP=>^mvtBu`9tlBX<GIcj(Dy>&~wV5{{iiSm8RA+O=v
zQJEfwm?RPCC`?PV;42F9S?<2d9bJUr;{?ZjG@u$nZng=D@t9I(iy5*KIOg}(f|%Gz
zoSJ!qHcN$ti))mo4)`qS=CK8NE>|5aPCo()fD%w$ZR1dN4v7Ee;3STGahDtKlG{{5
z*K@E$?4yQn|057b(P7g_H;ifu5$h_{qeZRx&Qp=sVZD2ak?9~`kbwJ#ri)8duEowy
z+!Q+RlfTl<Z-4&q@9rbN|2HI~hjpZp{y*Wv@gM!3|BX02NXx_WKXIvS3ZaJmZwT7_
z2Uq*wkc@$B{o?#jJeL1QBk+HNm=ltXjs7S87X4=`{=aYQGynfy_WylZr2ju0gVK*>
zXq?BRZS1@+V46%3wgC!ZVbo=84w{p~aAA1BXAqt3uj-$Kq3{ERQvSxF`6M>}5ExhM
zORPg8W~kBx0RNP}T3A^rL^_!!p8DlC^{5bb00~Lp5BUKc+LC;l&}YZ}M(XLu1;`T0
z{UH6h!PXUHD5i%Bl^CWBmBJVYeb3`9dhVv)%~~3Bn-J4_4=<{?hco8ax$hb7oq~is
z)$}Jie+uOgSo<810VGzR5W>k=$;e<rRS;yEs->lI1%18x@BD`SdL#otOv}+J5YG;W
z+PaK`3>2tch*QkkQk}N*cDx2#Xf$)C(MesOd@R4`b;xwjW6%r4TsZEbv$bm{nO#rd
zGb&962?`m*Nd#qGy*D7+{dIc<`>_4z!yR$Di5d3=&AT31FXPd>usRA1sxd+W4wUAS
zx7gqJccR}z%%3S)E>f(6te+u|_HyUiFOT1a6(q;JgcpM__Pknl4)XkZKX!WFd#;8Z
z4KF{QFW$ZTk|HiCKq^Y=x$D>pWdk1<am=)o(2)q`@Jk&TJQ5UtUlt{6WpvOfEps>c
zjWelaXfYTWC`FaUM1qx|Ft)}%p11t&d=u7?e<7FbP}G-`Rr&hmhLZU665=<a#RgQq
z$$c&@FNUWhmKd(gn9dd4Ni%%SPGdH_dOnZE#?T1Gw*V+hMc|}Z|LS=>Rn6U4zg$b3
zZsW`S!9!@MFTM>XSKu=sKwhsy<bf7nshH`%%>`yhB_jGf-_k-jgL4j07y>l`XaQNa
zMLX7+T)YYqnstuaGW+qG2sXEvc9XSg;qcyMOG`^W&qrnJga$7Dtvg(G43v!~wp1Nb
zYNT`wCxM)GECS>T-hNl7>_Gjtm*;T^AK~Q<2`a3=cYky4%GEg2c-pcY_PfIHLzws2
zQFyMx2*z1k>IB3oKKo<;<J$jV;e_u!%>UzpXBU0gXS<xf&8&0{y^Ql_ZAl`nI0X&H
z<nMpim`R=kVjJhKlBgXEc4(|nK&yKGNI)cFWJaN(qyv%(W$mEC-({Xf=9mX-H|x2h
zSyoSLnspwo?!)v5Ut67ZV#`u+NE9x|nFSA$$`l`8(zbs3^Mg8npo%STK=d5g!w9zs
zkUML^)jxx;c8_FGF!QQ{eiKBeCJE2>sUT$9<p_nO$~x(EckQX6g>^3OmB0vx2n0%p
zb|EZjF8N%KQWL~J5GXwObgVvLK`aXN#Hpaz`#{-Ra|m;)+u#G{Jm1LLhz=pYx3{@#
zd5cxrms|Ga5yl6|QQg^K+ym#2Lm5M%h72T~(7%s=+!|i4%qc!D90=_|CJ>HKhJGJ>
zW_0Bk=jRtR2<VD!9YR;b2X!cXn;H8VDC94nlu2ZUlT^uiB2v}uy?P<dH^!sm#|Cka
zX66g&B@-`GVRmdJOn75_fLe5{=&!Ch)@SIE#UvJalt?tC+TZ5|`DlM=VyLU=A`?gN
zy(NdI9UVO(A1*3Unq~>)!aFsf6|Tz4>vtWp<@>_;FpNIDL=E5PgSiq$xCGj%+RN3v
zkR1kziqX(y^y#qi&{ERp33`t$_)YUG!pj(J_9X?n>ObADIcjq@7ABj3jl---B}ma)
z#}cx#sC}NFED7S^Vuzu_`U8^@dGvx|#9$Jj%$&L2HQczVbp5Y+c2hbI;;!=lPP-TF
z1iTI4M6iO^&OYt^Y~_B=bt{LDtnc|Qr)p#pMpo@)uFd}7PQUlFm>}-{=a*Vb5(I0s
z_eR54^7p)t?T-QC+@8PV2h5Lad$I<*OY3=94t=p7ca4TgUprynU$)-8-#5~F_MVT>
zOE)HVUj|2(hjLT_F8}uTe_69hv!qQJ>)Q!Ig#^G}4!;)t8-?B?{7+eO35zpVPwXnb
zhgK;w8cH^Z^pgi21B^kj?4BHT@f>LAk9}n55$B}ucRy2i0;hlTdTdcIV)P=eWXB|q
z>Do(8gJFoT%ziw71mEp2h-4L1cjO&gg}_~!&)@xSU-=&O{D9Y^L0N^S_SL4(K3VC^
z#dVC!s)^PA`j)ngr{U#37ITC$XJa42<nr%;Oqj33-%YoJkF-c_C8EgMP}bq+1=rpx
zHsb8x&)F~VmP~=ChVSnW;W&&-S4R@6619Q(Hj$O|Ri8z|!!6Xl$du?m8S&QAUu&#h
z#jPY1S*>)#xlN^icYHv`CNJ-SLsbj$%nTSwI>|8C{!wz*dp4CS$((zj`-gCg_@w!x
z5%4jgwt$5BL-E{t<w4;s)np1L<s5GPu}lBOXL@X~XM>uF`=JP2?f7)^E$M{$Z0N=B
zM~b8b8M@MZG_J@bhB-eoxy~Q74LmH=G53SfLq51IK}pW!<@VLgtbJOW>;D*Yk)l;9
z7Z8HE+V!)a2FrwY)_#t2Gdr9`w`%n~ryv*X2_e6d$nEj{J)4~~1y&Cpw#H9BBq5U0
zkaAd9eB3kC&bF%gEo`*ehjY&~q_eyjnF<itV-lX#JuF*q+}g+4yga;0NbK(aa$YG(
zW0d>js{KHCef99Es45ImC1#cMw_h%aFjZbm4W9G~E;``v{&8f<?>A8MlDF3zldkKZ
zr##~SRFHoemqRwrIl2CEl?+&GYXyU#$~Zg7tQNC9vAN8zR8Y)KKM^~dL}}c}N<dr6
z0sHfQX1DwsDNTYi1yx|EOcf`?f3d(^G?C-7V&;aS5CeZ|W5INH`xOxKkZ0Fs^r!Oq
zki8+UEkd5X@(a9KXbA@3msZFI_`qhB9X)+@HLAu`g(~8e`r|RLAsfyAVR#9WhFkT~
z2KO>6cVEBCuD!pFvaBCBywSky4qq+j9(_|tiN#h%BkArD)rkMy85I*7C5$!q8S_Oy
zsCCwR|H9`bov3JMc5d^bHQyoJVa)eMISAIG2G){YP5i4EnFwNGP$@iKIB!8Ra_Npl
z-ZK`fp|6k0L4;R!p%k#lq#$##lthj-bA});pYEQ&2*v~+l7tWzMqXDinGL&l-5=c7
zY7VX1@EEgcwxWhka)}>*{1K?jMjN*_#@rdr)BybA`HZ#_)`rUi68?T)1O;Kjgdm1C
z^eDR=FAI?c%z+X`@*Re~zN|2?9gH;qg)TDnT*u5_njQr(MoQvcEDUSY=gIiM%ONI6
za%>44C_u6@`agsPdci{%6G<+#j)cvXGxZJUP?aGlqoncB#Vb<+lc81G9&R@w)Gzd<
zPz|jva(325eU?j#k3H`f?{TjkJueXJq_8?|&IB4uAMK4Fi_u#CWIgXQ!ap-^w(o3C
zeS&@Oh%qkK5BT$UT(z}1k}v?HA%${PWTX(Uh+jG57n;q-5>`7!pHpU6!idUtCS8jr
z>s6R>vujHb9A{A*^m5VTgfSx_*6Hun-<M{if2Jtv6dO?`p)@=>$eQ}-$cnf~xCE+j
zh0Vi_eCr}VaY9w7FCVj?Xw7D<^*BAq*uZ#tqE@7(DO+<S#U9I~v6WDT9mvx*YFWFg
z>UpGe4?_>D>#=BOG5Xw9cedeDF2shk=<o1f{Zx!0BPPa&1IanJ!+Ft!3m}62vud$H
zr~j3`VfUpC7k9puYVtAV7gCO;aV}A7IDW=;!4Xf0(4|T{hn{xG$i5ON)XAaxol>Zx
z1R~Y&oN3p2^oPGWJWa7s=y45PF)b&6x8oB{3KJA|l3G|^j`%w$QU%OG>i#{DR?Ll6
ziQ`tb9k+wXlb}m-bg_g(q?#@ZvL9a1dS6!}(1-)oL0i~{0^@cw;1CPb^!FGC3Xj1=
z6oe-Jbo%@Y15vR|;H3S!QnxD8<cjNX@No<LE=0wzFl@J|dZda<gjy`LYY_LU`!LS8
zbU|n-No=d+8!kZDa)E6xsSY3LnF&w2`miv^@UuXp*mn|X_OnlqP1lU~tdk2BcVK($
z-Q?!xCXVXGl9)>xOjf-fGaf%Bd`&y$-KF}3mRb^D$TlpwacmaSyC8%ziH)i1xAn|Z
zLthat8J(^yJ%Sq4iYjYYT%Z-{BBmoUEldLY&r(rHJUu?Cej==SvBa>-&&R9wwA-~#
zntPrYS1^QHb8xRnxFq5)S3b8m;qqs@ygKjs9%5!Io*#^om4)qRMf-Y{RSPgFE+&m>
zm1DN3(Pys#j8OB65dZ=1X0?`gRGggSCq1=SPVuOjU=(`+lUs6e`HC9LXv4Z>xx)NR
z6gE|*(oZg6Z-32Py}jIf8%0$G0y?5VbsLko^d^=se)mW0Zkp`Vj)v@^gxZL0W*oK)
z2`7{AWODEw#ix<=Nm8?NOeH0tyO!=`+f9kmuGv>;IFL`Bhw!lLG{*|4+xDfZ#dd!0
z4UXErVuzh~vCz@*wq2;#dTbw~61v~~<j)(9cap`hjgV#50^AKeVbR-o4*%8^G;yY3
zODR$=vM&Dg#yeI|^3?_jidr~na~`e4K2C;-a8rD6rrtP}=He-wvQSTPOzd;@nvPI=
zsh<W1t^XQ#m;AJ#uDPutH0qY*{aN7{q<9}IQ_Q^_yu*~$dAwd|ky55v-92r0I6lZ2
z^zT!EMm>T_y&1kSc)j!8?F-DK_zxxzCkhXs{aJbnrTc|MsA#^+3tMK*z1^}a2hr?e
ztt6i=EaqQF1)ypaz$;Sc=coFad$J6n7S#j^qArS*u#Z*gQVoFR_@&lx<#Gd7)2P<T
zkR%LLCs86U^o%e=tnmXTuH~6$ZG<)jz`2CZ4iNzApT86q=Ue3%z++4ntLW(bBY=8g
zapKJSZ7o}2*>|m*7lrsyvdm8No_+j1(WPPq&b|>1$|YRxMvRVLCAsQt0Al{yxPY6r
zBZfu_ors9~?|^%Hgx#mTW4|?*%Xqo#c8_r9yi4b8^#O1f7x7O}Vff3Dn<P8<2VvMS
zOd@idWtV<kKsOv@Dzz=3ND6#9QRc-eGp*qCE5;MaiKPSrt}T8e7=hz(>oa>7B=1Op
z(bJ?b)~3E$ElY>3?1jNwb>6&_YbT~iy+caCq#4Rb$(m-RMY%;7mGq+esK6>|DWdx;
zL7*NvRh6?6xRcU(^ngC4JtuY{dcqt{Et39wDl7pe)nFm*+s_!N4wkF&ytix`_Ps3f
zhnpPS6@9554zZT<n8u|o!0;Km&=;Lw+*f9AL{?9qH%pAV&Z1Ou^t!L@@n-LhRi|=b
zP1GN=sq^MQXQYGoPSr9$D^PK;Wk~V;Yf+;B0^h#O{N-dhvRd)5C^C<ms_i_p`^6c#
z{P7*AJTjOoNt=s+J5i5r5|hss3c{e78G@J_A7qcV2T<%QrBj;=VS~1g_^x*G+Ww7$
zugpGLmOmI4>I1MK%z%Kofx6uH>}f~=+@tNU7#-L|q0;c|qQe#35sCIy;$(QjLA|zr
zfR$O}<huYnzvg@Ozca9pbbpG_2hL?cY{S-xr@vdXHjCGeWqpcoxc?Z5U_k8jeTr2T
zDP7-+3r~@04X9!Hy&B~DvrHvBoRnlLiJYbFlM*4G3h%+qpz#|Hp{C$ndOj6dO}9=*
zK*;_d0o_bGXc^||2gSi}4kGE{a-?jbznLV`sZwbRF_P*Ac{<%}DK&~_v$?ZaYw)(F
z9Ne#bjxE>xmTuzG#*uB+;iFF?>ERbX-KPn?@btV7GJ9Y_L6qxn9Q|%uq?!-RrvkmE
zGI|$sE7(43is`CWIL9={=-ee<pdG!MZvWpR`LbW+b!7G(#hFTX&YnNJTMc^RenCi@
z$BNCJZ$;=9YWD3IU}H{HiFI1EW~rW{0UC@UY-O#Wi``bHBNa}r6wB3(B|4Vn<m^pZ
zAA(4f1Zy>jS|)d?bmMO7{ki?K+o%gH7%#y<Mo9?E#rg1C>;o|o#7UZ9+_O4kyyRkP
zHeadxD~;e9n|Jk4mtivEhD+J;U&YXrl%i-W%QHHT-=#H{r~D<ln9lVXvQ+fzT{85a
z{P&Z>l9bXMP~B=Do=SvUU4N)RHmuo_45P`$8bz28DPL1+hvBqWs0owxS6%5FauNUI
z6>aBla-iM;#Olnf$=lj%?cqnSP)00t7KmO3K7SCV;xy9wj!r|=T5(t<rtwcmo!2k<
z-g|A}=hbM3gQk*Z&rV{(C6_%ARm}5`B5XhZPaA?-HW;wYbyaHgTE)fix>p)q099jk
zbLA156;(#D8j0_fQe)v%v;Tw*grZ}sP-VsRSUH2UPVnqQGka%0l1mT8+}PMhyD<%g
zYT#7W1Z?E1%-0Y*i!!vK&;xygcpML0^`Eze^xKM>v7hbVhdjI9&f?ICN0OMSp9&6^
zDDj_2bkpsefA2UVelA+tPWnokeZkpuhB9dTOH)Tfd?TICB(g4?MS8jLsnYyGCE>1b
z1b@^XRb#<5XwG?U8aPDT5N=2mwf>V_?WGGNbFgVjIq1#f-<_pmE*qSObJLEWwwCk0
zYctgRaA*4}52+bDnYp;s?h83jO>*+rE51v!mR3l6KPU<)!TVQ|`#GUH)Rgx+N-_5;
z?tPot<0?!jaZyY#n}!F?J);$)6myZD{rj)O>0L0&(gbIMW;ka~h1rE0oORXG3aQ?c
zKk(zZv^wua`L8_x3Qp!#WWBv%zPNxdcGW^2VEBS9G7`R^Gb2`(Y2jTOCB0NQe3pP*
z925DjLS%r0i7;qxra$<Um3Q%2tyIIl2{{=`Cz?n+JFMSxw^6Z~cZ?|c`xP9C-yKSb
zG+Cn@#LYXz@Z(sMVLZW=$H~8Q51jpYcH^iG630t5nXEZx`i?;ss-z(!?Pql3mvDS9
zZB%6K+W$;WT#&k24ww0(Zh>T7PycH4?%Ui6`9Cw*ev&?i%;jmmTK6S69<(tHCdw2s
z8Tum+F;%v<!Ed@>tQC?P4^;g(6E`l4lI!5Q+5elrLUMD2OGoC}LFk*sFRdO7A+4WU
zy6*n&#1)7uKa43o_~K9L4wzVf@Zr12BU|+ewrfWR<!jSo2PZ3tR{e^l6T3~Y&_#V1
z&#<p+k*Yg)tr>Gv=UwfX6<kz-nyS&TcX+e)^_mtfFl8UWwA{ThYqG-PkEtKDa!eps
zL5QarkI2p$ZQr)uY_YGlBtQN3StTaS>;C<|{1|U0J=ds0Y`Ql0a(r5F&%--En~g_w
zD$A1QmoOHxb*#pzn*af=bh?>I;-Sjf{ckW<Tn(xKb}q1Hq7@H{XUl%9e@W0mZ`P^B
zzlw$YT>oTqcuNSe<f8Uj|L{@1-`gYbU7k<6c+F<7_n=C+v!w18EQ6U^WyPQiqe0%w
zRCuk2{>;ML-D!|DtXs%66#}vPl;tZL_^3_LxA>j;^;DV_U4|9fMOL|93bday6=_w<
zfJjLx`{sen>u@5n0chq*dog~=<e)6B4UBQ8%I9A*WS39i%xiDEmjMV%m+Nrsg|L{L
z)2#_`J)gmB#>QrSF2?tMz-V?ouf5vob^kZIX1f(Zy@cEWdfT5qUc+w8E$Ir>>Nt6U
zd;ay!W+9K!ra!2*9X*iU3lWquZJp57-Lh+k@`h9?_xC!`!PEoDh8!C@Y_0t=t?^#u
zr=yk$DP42hy1Ok9#uf-7P6UR*Ac8n6x}Ot%<XN4r`t;#gV<45;$W)UdF!JBgN4(s(
z4G&NbAHM&xUr=6yG<x2;-c4GbPhT0@4~&v(dT}(ejQDB?^P-dEVqHxD(<whx<KSpi
z&ZGrVx#4lFNVBUz6z|b=faLz`)%G`jPxQ#cIrz0)x3W*bwW^Ne2hm~drQ{*H<_G$`
zt2civ%7vke9(qPzZL#nqN&Y6wvNK!fEB+P`!P*KahY0z|LpLe*>l3~2vw<Uy56hAT
zOoyJ3weoL3%>vGfJiB6T-P7yMw3edeqU*nX4kx`>AKwQAeg5al$!QB5u?UP}Gxt?1
z#MDIIvKIbg-o<X#;Ox<jZd;-evK#`mYS_g|KSe)JKq8~u@3FE8ClhQ-h~qJ7H!jE`
z3PE4<1m$$P9JcoUYL|@3c3jj?Z>H><Q1d;sko<opNyrFy&6mjv{I#8=5RSy?WTWv6
zk}5bTqy*iF4q^rz?0f2Z)O9+AV%uw!HdGaj15p2&ybT@BiXtZEkWP~>GI%@?j!BKt
zpcbl^UN$mmsW~y@paWKNE-l%L_?mFi&K3qSmY(TP08`1Be@dv?(O`}!uqR}kw2Y!S
z;ybD6NmUo9VVO7cv|;jBAE}CV<m8Z(au@@_eVZnAc5$7AK;XC^hjs6fCmR!0*oH2=
zgpbF-T`<FXm!CYE@rHc%Ir?~JYc5$5Jgb)R{)Qj24ehtq;i|=Huh*4@2N`q^NHZbu
zjc!0VdKTv=QUXCu=uMsfy~{iD$B~fadF-ph=(?;0BYjzagpW=-TQQOuePmV;=r9s1
z>KSCIxz0|_k`Z8Zu@K(=8v}r^$rK6GWQD8#z_M^)lIEGGqeG0oDK(69Hh>aua)KG`
zDWzNGJ*b<1Lp^g<w4-tpo9gF#bmT#==P(!<j45D_UTAOlg+f7$wj2xaAGmQ8)1|t?
zRpT)m1%Af1tbl=sbvVL0_YZV(vMp|NX3E1RP_HOrDp|ayD}3r3)i$`OTajItu3$7m
zgD)0Igv<qg{nWzme&yh`(f?YJ<|HB!>)qoeyJIJC%lGZuFlWCPpieB@t=g7qFD2^-
zD|8@CL4B5`@TT-|3KiEyZJABGaJEacnlx)ag#GfxJhqW`N$2sv|Bd@Ks^^{ldz~W>
z(fQNI`wO3xj9~WPRuDalcME3>YD6!YjB_Aoku6o499Gdo;TOh0xds{rp)^tV?&L(;
z_v)E|{ad4?T-qKXG#_}yT3VZ9aCQn~|BZlETaV96%Zp0Wzj@z0b=|hs_i;uL0I_6u
zO3K`up~5xm3p^>?V1_iC4#cy~O?7vrk-ik=jdwXa=r_h6X(<|8=~EMmbE+RQ>=sML
zJ)^VALK@kM%F$`5HG(cLdbC-n|9oQCw41!<#SN)lVo`%wjs_MShU?z9v3;%vv0tCl
zdUh|LDTn)QYiB2n1+KQaTk3010bn&HtVkvb{nmttoWGRhh&TMCA;{aK8ql&+LmM}{
zct_fLsmt>0Dr~aaZJE#=Wt#kmimfwcrw+MJnPifI=ziZL%PjaVAP87UAB14eDA#+E
zo<diqMfyU<;c17EKsMSwu2JN9>8o+w{FG@aIz_xF84CwwI1pkLR<?GQ*&~iKh`NZ2
zjiNK#bB=O!hgyhkP+39B!Ui&+MW<4Wn3CpBSU>L`eRD}>-!#zUic9J;kO>?mi#ZYC
z65TXqKmNdqwAew-@;t334fw6ywRUfRw{#L?!xI-9D48}~hVyj0!v22paq2(r*JJuU
zAsO)r7{tkL{<wWay`C4#?|(IR;(+jVN8TkQNBEnf?MwnANR{#S_8a^6-pri&iC=#=
zd6QBKg~|X^YuHhb*@&rRT5Otg14$dX!LX$!M_EZQr7Wlo$;a93w3ht$6SrlF+KnYh
zN%G-Ig*7}1!u&ML<^;@YhI*+Uf2At;pG09s0{=UNhP0niNFF5UcGMFYuSy@K8lfxI
z$63bv*%1ar29sB`8*s0EZ+UtiN}EPYTbtGd-ri+I#aI<etB;~6Q4T1IsTD-|&)v&k
zPHbz^PcpDJ87J_Sx`gizL~2_#)c-I)gomqi5qh;j`OO^rW^51mwH%C2e#_%;`!{-r
z*}l#m;av1Zd2iQNBwed(U-Bj4Oy@WY@B&fd?Ub%n1Anb^t+1<o=ND_sbocRkOIEz+
z?#!%fm37xBqQ?dwrW$L2OEApq48fR1%oZD!FG|f2<V>W(`~9DnP0v5CiVMjA?m1TM
z#5!pY-(5M{wat&DqsJL!M^x3gx`WjFhoPsZG<hSYT8um3DaSVObl01y-W1BcM{afK
z@iVY<rMmq};Jrg%p9}B4OB1e@P}^c4*JOA9Ig@&6iNCF#R51_)TboAMMzK#*%rIdx
z<S?Dg@<qxP0qD+hYwEj$(Spj&doHLeu*11vb;I&<^Lz2-=#I1Jb=$I&=4Vz-cMevl
zowbo_I1J{vAG&Jaf+<_IsA>V$=vH$uSNmz#iL2h(`gx%>SNjFk6?uW_<9B}T=G$t#
z>+4>A?GFFDSMN3KsUqU(RL$oX?PEhL{We~rgQcn&rQUCeDX9mZ_wYGqPh5fh!v&Cq
z0%Fsm@y5Yw?)5@FCQs)VD{TAgN8_f+A+lka9Y@#m-k6G))}8(1OiRT)Cl}v-CWT@g
zY0~;EDv*V221Yu(S#q&a3}t6;#EoRz2<|IM?yHdqo#Lj4pL|>_8UrV8x){8wS)vpP
zT3Lh=Sw;F)-@3)YR|AVOr?LWM=Oktuv#Pw-1=1NaL3>l%6Fq7#>7k4GHbJb;gxC(o
z#cf~TyygmZ?IYBLi<f-U<CU+v-!2<%?+8FLC61-0Up=0+OLi)fYC2r-rpQ8@!^^z#
z?Gtn?y*rC`Zp3{E98;*DTbQu3<zc^zw05^|uS}wMKc2FyrI_=msF0Pxtf*YM&B@8E
zuzeO&=9KS;$`_bX=GvS;_UX1KmX#DB1AE*yGq2}kdXZgx7v9n>pV3JcPL`rTQESFH
z=oZxSmhdzD=)fz<^jXgvx1VgksklCkqOR|{{y7Qf*ZyB9lK?d6#b;9-Sg73s-N42j
z7r4>>O~83l@*RscOjI4Fiu}Pov@y$gZ(GOh*L_$4qMF}SBs;6qKRIyZ-_mIPv-@Ij
zeYS$$;EL_ca#K)?Kl0%0r-2Jf1B%Y3!=JrvG4fxSBf>Wr!8@I*=qpGJ2Qs)*{L?TU
zct^^i=#~zZ_jYC=mcNR5f;-IrN!7PYkjC;n^<_<0A<R2<=d-g#;_tZ=YK)8#r;Cl=
zWGV3WIcdD?K9JWlS-)n_g;pN8Q+}jt=S|{KoFsnw_ObHPNHv<s(1ZR$sNS_Su`aMU
zQQc%~ZG<;4VPp~FLmrZsxrUC7i+gEJgAq9q^!X3TZ`@B^YMSQeVyzu3U6o#sKVy&8
z+1=x!tIuKRLHerv{TruD?Suj^ACC_;&qjG1s;Y$nOrft1dy*Y3#Y>kL@7sn>RV#EU
z(qOBCOk>7m+(i<Cl6-{Wxr9z8jjuIytx|;XlJxOSu68v$F%h03hzr7gaQ|LhY><y|
z?79c544)+XAM6Qds1Gj^dh!Z)TmUJ<#6Lz@@Coy(lDGc~&N_c;aFqpDnw6HoSpz#k
z*|Et#bbFzqB}~Ho+`bI!g1qBD9~%COy%iLcA5KG`RoNYg!pWXYdDem@_2Wp6`V;6_
zhU4k!+As9pC}i5RMtU;l>^=VW1~=wRmfSb#FzuJ=>4dKBymt-;9YSgUfu{qclTr4!
z*Zz9Ds{t}#_2S_Lbj5&YoS@sA-npQ)y;FPVU}-1tc;+)YC=+7?K^iiVa>f#6BT-fO
z=NE_s7c`AucZU;czkqbi6c?cI;9(T8`LGkcd852XvihPOX7bci)*F)gsIk93ncC)!
zC1p9&_6pg8UkN>~i<Yv*we7+DcsLy~B2!GbL>3h<N?#-VUZZ{>-gQw$gR`|`shwAq
z3A#4;-z1h(ev26eIq=(D+W0*>axA7aX!O?`ULMY<2@udn|C)_D@<2M60hJIGIpsdW
z3B6I))jB>V)pZx{rfhvI7`o`3Yb}yfR)=JI`ua{9Gp;jdl+X=B(IaMR+63W+fU`)k
zW!kt&_IgBf8C2wnc+ssUS~=m2v7x1mXHqsOP9uN1`-`nyZQD#^vUlKV+fZrt57Hvo
zHk5yhT4zzprwF8lsU&Wn{Dg%X`C(vt2j5_JOPFO`hgOLX=LuZ)pFfL^U2{{zF=n$9
z<fm&Z`O`jG_ysmc;3lD`hn{e5v({r_!gz}xXJ7{yJ?C@{hf5g7zQS8ZNNF`(81k22
zXu11$k&QgVIxGW_x2xFng&}?YPZ1I~v;t$(%`Ti9Szh&2+>u4&yN<GXWW>j{(A%_!
z%T1zo$OEz9rKbO#=6q&_4!@TW>d64o`b6DGQ~mtJ8N{uiU=dK!oEEI1nTL@X%{*E2
zuGZLC+Rj>^cQ42!yJ}?}I!SAEU>^0JYJ>u<!+2~H8=;?`u`QO4_R=4iMprxsGrq3|
z4^s}%j^asDpHZo3Lvy#DQ*95{;r(E2LMt=lRf9LHsn+?5=QY27ATd0BhHN_xTYYd(
z+!ZsF+Nfta<wVn-AWYQdi3D{?oqaRg=xzY7zqQ!@?s+tWx^bPNN0LNxo&uWxBpceG
zoLssb<(-rnP<9}kqMz^k^O8$(t@P^yd}-_ftHB16D&&SIv3g5ujAAYm9rz-c7lJ|l
z;<-Q0?da*`8oJx(rSoHBO9az|#STxkRK7sG(9^KKM>Lu#)bWSdDp^*$hpXV_B*OuM
z1Af^Di1s)*)KMwOiy|IMJwq*><A+}w8bzXsiO?iNq15DCo2i;SL{bpFUWfKT7WV_`
z<1m5KI>;P9>X!4ZfD>O|((9VjgJu3g>i+;7LF2y1GXBr^z>`At*=LUvyqqsy_l$Nm
zG#Fh!pkv+Wxb-K<M#rrq%Gl>ki2W2g?%P=#CXMg)C(bVQvu9L%+c{(9xdUImeaoxs
z7hLW3mJWS7k=K=yvz6l}ngB>(3JDFBT@p^wqA6ocgqR>k3-urO`$y<_=;afi+jnO7
z(``IR2eS{Y+8RPqA-SO1&Z;P0kxHbW%^P^Lr5to4eN#M!W8-YyWjo8y4H3iI>e(l~
za^>gTp5n<c(3$XXlb+xFC@qKK>^nOT28~vPhc{mQ`PejXXNB*__KsmqTjWNo6jt2p
z&v1;3cRqyB3xJ%xp!e-5B6J#mq>o#g@^}txH=f7qJ~UvpX4#H`H!olF;&6qQLRnVU
zli^Eh^@-I;sF8@UpC&?82$dz=lcn+G<QnRy)AFbDWW1{37G*8f72s&^k2yLhcCm9u
z1KIJorZlmecO3Qy?!p~VO}P!O=+-Uu8)e8Lce+;F#eD?YqEojfzELktjh{U&pNn_>
zqvn`&3(BL;lD{)Le*BT=ow?}nV|3j6t=sP*aL$u!qhsj?XBg<XIlUTB?ynx>@y7mo
z^-+;CUVS#A#z)TjJ<zRfx?Vi#?2T0kNk}PhwcE3sCXV-atfwQ?;)X`!!&ThLP2{lK
zlWQTC!n!Qvb;Y6#gTY(4VQ6Drx~=zR^L@WXk~O;Bz~||6r5g>60!<rre}v71xO{{J
zGzBC@Qc7NhH6U6z7ulspObfIchrJ2gSw2EOoL%sk9`iAA<GKA{+Oq-cxqYu~^T@Ri
z`ZxWkYK+9J^ZqYNdHA^D!8|HUK4gvG*1@5FPu|d!R_U5it6mhNM-x>WO-vAzfh2>l
zuBQ05aN1adc}1k91VqbO@66ZV*m$UFXC$pJ+DbOIYA9-<YFnfUgMjx?AZ|jQOP#|z
zO&O0O2o}|}-_5*v{fhtMfAybH%gXKVzUKYA-|@|dciil6nC3n2m-pnlP?iOm3_~T<
zfEHoi@0jKZN#-$`VxpExUNaKKDlvvfDyubP(jYMRuQu*?_KOX-t&tF;bc1MI3hB6_
zIin%6ixYJRZ(h72pT6OCIdvSY>H>0f!3aQ0MKiFk5EV(A(BBm4eyt1H(h_0C<ZH(g
zZrkqD_U9i-dzy~V_f}v7H}8fSeQ??E;K#rFu5^6Xzi*{u140=qi5gCvAP$WZ6Qm0|
zu5|-94-%+edfSw;(1m67iD(9@Cft4s9k-MKqu*vUYr^owL<zl+Ku=?2@LY<pn`f@}
zd;Y6`@?UVX-y^4m)A647@87eQg;KKNV`)IjWQr#f`zY+DKsl}Gva;JBnCBTy393bO
zeF_~L1#i%-Pp!9g-VUZ#{VurMJWj!Ydl7a-4sK+dLLz!J+ab;*!cd8+ovGp#oLmqu
zY84bqMS@!2?V#NRXvI>BRgqe|A|OF5o0!j+j^PjB$<XoM_>Q#P6<;`yCVsMWTfZK=
zb~6-jX@a&IO<LFrEHvdeJ;IY4Ek~zi@D!yO9LH&(<)J^ONR37wYrRMdEo{7<{Ubka
z5m@7$oBpSAZf@^v&iP;j_A&COKmP^0nD`I>_FuCsC)VWzjvq&v3ALi9i73ML^$S)l
zoXSeQU(m3!+Z~vunVKe4Kb@ApJ5Scl-?Q4>OuLDsf)smaIjzK!2qkjL?<xDrVZZ0q
zt5<vpd|3X9yPPRYrf4=kTsOR&po-?w&s+rXN^v3f>8GV-nhq`y0$K~r2KLl``(tiT
z)3JZnjI`tDzVYNiB=z{<hu?iCI+l%An$6>Zv#Ikao}}Z;=X{KgB?LkcR)3y3Tw=dj
z(eo5&9;f5We1<mm@!yBvusyVmtC5Iq{Ws@nW}YUl_B#%{i9qGA|N1v(xQ$r9Lol4S
zqQnxIC2;lfhElDvC{@wo?+Pu+-Xg+FGxAiWO1;=1+Zb(ow5sHKJ`VrxK^ug*7=%@Y
zwW>uvR>z4XAT>gWlzl>KC6&q)CsK$=Sxoj)Kms);YKRa6!P5}?Ipg!Wfst4gt29fh
zG-IaC$LC7N&+QLy&jzgL_Pw<&4nsF1J#=q-`)*9AYN5tK*%V(H1Ztj88f3e!y{C?h
zxO?Dalf9flqI^p*O!HVdenb}>q<1O+QG_W);*`)J6qKb@W6420S3Iy*AB{(D7+z}A
zr0E`oGiE9thmCZQ+kl_rMf-t1ggCk4+l&f6zq)yRb-0H9ChOaj64R9U`Hz3f>#G~?
z?>~^&h1<IiXjS&}jv&TpqnJWDPcxDdF<I1>?hmACCdAmf<HCx8Y11Om9%DI=3Q-2E
z-WXi}bl4|0m4>}kOQGaKSxPU(Y>`<3nG-3(;rhT-?~%p8CWB%g-U=dCP+hAaM8w0X
zMJvH`$!aylA_gHUM0F)M4D&qh*uJEn51W0AjvM_9|FEf*?lJStCz(Hpj@#djj-#^C
zNuy)gFy(9mYzWg&$>%C2w6#>8R!xT}glu#?r%0M28lX^Fb8$nod3QxE#Y|L4Ug#i-
zbj7(Zw_KbpqJ90<&k&QV<-vaM1~9EYUz>U1ETW(pxY|)62HsvP%!kN+nn@y1D);xd
ztm_I?SJWpcc74Q*?EC8%tfg?wE4#F#?yrf_Al%Vb$;=H14w&337LJ4~?Y!!r*y^`P
zl;$ZO7RKn)^@&hD60)VWRePqT3b;dHKNA$L>Ym;GOp-`x0<xE0$Gy?PQ~rX`!FFSh
zsFRMNB{buiZvB*IhWmW!7`_8f?nZfE(MDc>08f_T@6o;?qgP+IXZ15cHhEyg+!~v@
zu{YuoH_lTDflzF%=P5EZT2^KCq5>kg;u}~o!&1WTk1gu2bxONEU^|2KEFY;4_9@td
zHea1S`igA(u^H8)SZdC&(fIQ18{WQr#p(8z3{=g80J}L^(E<sq8A7S-(nOWOZn6qs
z2wWeoOrcD(r_v<^lPiR<(Xm*x#5vH}qbt0P=7J}Kr4>rBv&nQ$?J`SIQ_yQsb}}>R
z#PPJSmP|;I7=;-6vFxdwO3s;5tay%!DerAhCZA0+O13@L)}x|2RLXsD*giOZ|ESwY
z$Ns^lz`Ib}M8VLa(HXA!u5`@#PVY@GKJ)oL6AId^F64!=`)y4kCQk*YLx~o(6By}O
zsZ_G_<d*U+A$F%yQjgB{)X)uvV3;zICnD~25Tmuz{W%|NvfwE9U_8UEwt~wGYHJmO
z8Jym}e$A^FSKNN{4U1U%Ukt)D&;7(z?5w)k@5oxo#{4Lb0e0B!DHzt9rA_(>^{L4}
zA751^oaa++|DDnC_`Chx$18Sj)U1@XP_z{dFegt5>eb1sAB_TPsl_Q7_MOZ#G9@!$
z1##!Oy{|ik+vaVjr+P262VVLe+uVI#T=}{Eq3zj#_1wP4)(%u_RX9`dM7F4e;KE;Z
z?iXY;Bxu@>Pr(&gIIqa%|8_>xz7tES2Zt)u;Vj$&WHzwL3rYDrHLRkd+Zd3Z^BBd1
z@p(5h?Ivn~73Ekf#kqHKEGX?dc1t~K>|478zkOnrJ7i$P1}k9(6RoD`nUb{$33kN$
zow_L_eQpGLdAQBG<jg_Y&55hSfnWdpmmJc>hi`t*vM$^o??`FlYImSy^RPxDlfZu3
zBPkH3$n|bdL?A^Fq3CK=XTamYt*p0tg_tH%Zmf4p`53_|w<~@68@8Cz!y-a8p{LeL
zSqtkaQ;G>9q5+Xem?IN`>w3*}`heU<BGn8~F0=>{1OG9BxTlI%CAr{Jz3NJ=kTg&|
z^rrgW9U1SoQ=1y|xv<%Xbo^ArQ|JPS(T*94Y@h1p5Z#nj9;f5>IW7z#CZ?%gL@Bg&
zA@r06Z8!VAYJCxD``@C1f;Nisfw%xgfZc9l+9gUboRrOgWzklFKpE~21Q)!0e1YEA
z7KNl;%=LZU4ACy+4ifS(39XFT@cqR#%~;co)*=?Epp{)pT<vCJ1<je;+xMvBWf%^f
zRr;%1h#CoT;`Pld?oKDx5AQh42O`nVFhJ@Cq+(Tztm34EfJ)d7-Lo~?1Z3%F3}?b}
za7)1^e`}A1Y7rL{Jl%0iPfpi6f(kb%yALxlCL-DF{r!Avifytgak8ybYAFwXMiS`f
zr8YY5=gvq5o3Q-(&@p@}Pi{(toy}ggQ#QwvhEL<k0%z$Q?Rzgu+l?R2(%hM29ndhk
za4y{(vdz0#v~27Ok6Av%dH;SlG3}z!GL<!FH>SAa8w0K2@#JgY`qF3_cT5=XY<Mx!
ziMVZBxi}Ed{~VG9od}GpZ#l%G__@>~59~fc`RR{-#81BXlH2#+pgB`@B}QT1&E{;8
zKonS|vQIM_ECMsv0xRg%HFJ!txv|TM>LG<f@RaFjSXgY`-1dTO1!pcSbPsuQbb2=h
zwqY8IM>e(ylbV$sGcyx+_Y2FKNmC?78%y#vR1pfW=9N+_h@ZBOUDjz?wshmRw~YT3
z3liy$zP@$3;B<NP{xLVv&<ktyZRjmHL3L;f+4fPdQriftMP%_UbSy4s`rl`4sQsb?
z@8@#wKV%JK#Ckg<qP8?LkCgVRmLnb4P{~@HC$|W!h=l6k=B+Q{*jUnGwS0oMIcbAA
z>!tM1A7jbgLhTGeYI9&)1h5WaUJZv$)uNVCh<x?M7rcFS!@Ixv9Z})!^)<U>QCZcE
zx{B#S^{{6xnfJ@RbzOwrG*kDLwfb60gchyc;+3SyUu*+8TP~0Ac0(aFe9A`47x-R1
z>b1?cua87(uH;iD7ooTTt6&DKDjSD1-fsr1X-Z7<MBML?IYEj<F&t8xVxc?i{pKc5
z_SE4f5~&cXMO=#$!skTG&+QLy&jzgL_Pw{p$eNKw85H0ETE)-U!2_(Du-sbE?Wf5d
zb~R^0N!RE6zIG_Ka>66XUo$dj+63i|!{f#iyTSUdgh==JwD*Y`^PN23`?xMFIh$9a
zjjSsEfeTDc@Eu}w&ZP7%JtD3ELqFF|_*IbHb=M2euMSH3Ha9KuVT_h(N__G1-7jR%
zoDzp=$1X;Wp`f7<qD7g_DRL|;MXgM^Y9&eSLQ?`JVD%^lX(j&7km}p-TS44yX?!)L
zrHuF&^-^7Q>8;}6pV~?mxl$S|wJ0QNoV+TEP-F9o1%f03iO51iNm0-w#N??<5;*K;
zcGFCpOj_C2LeL=8AQb;sT9qd*72b^&9s^JPV{cE=G3QENUq7+WLfr%o*l0L@0v$J6
zwyy>{c0sr6k+n8#^ns8zVmQO0k<FOWFg3*Ku|*_KQzFE`DW6zMW>p6QM?27U_XHx|
zpG{C3o6+lLP;U2(Q7xe;an%7|?_%zolHZ<*pO-By35(jO7@&SrdVliOmwff+E%zVZ
zBU(uj3)Pnfa~zPEi3-=dL={+$@2P7hQV3cM)IPYyqPTR5^u4V*MnSfD7skc>6gt*M
z$38>da8(u2Kq3*uB1iC;n=;u35$_%uOYv3A`yL}9B?z$#>rs^bv|~3X>X1fMUFfet
z(AdV|oMh9_eLQql@j13{<;jYtle%*00}Xwh>BfPN@#OYdo8x9E+Ic^k!r7H%!7G>A
zebs6CqF7#g12wh?Ed0l5SwdgBIYmNfw7jyaRl6H`a-(IzqB5E>x<%{t@rd?)@8<b!
zK6`qWLay7^(M_@e`q>p|hJ=PQ19W=6FxUx9DUxF3FaP>~A!y~tKYGm+9J8*P93@f-
zQF;5~z*;N!rw>-SuU0fKP%=|6{Hx$g+frk@UbtDUpc~$|IUkE17~Nh%MGLAC>ULkP
z)`wauk}IrU?Ui8kEXhIIE63$Tt(6o*k9gkC6Z>Qa>d}Z+YvqfVuedqvdH?PmrIv1J
ztb(P!`q|ahb2UYJ8_(t5`aD|aHBMih?{s3a1(j-i(q!#Gl`UeoHvaQ%vUU4M&;6b)
zDtcpE8+w}J#tg^Y3Kx^C!~EA^e@9Gqj2;Py#LiphWO$CtdScC)T<P+4ZM=DO_5j^B
ze0fkmwceZc-fk+jKkGcEz8U&A%FrhBh{-6mSG#T)Z3RKt$HY~fxxHUV5_t2YD?2*Y
z)r?^kv>1P#z2iHCQVYwvvMwh|UXft55Q$VTJY+)%!=7#fA0|K=KW?eh+r}ATNVbWx
zaS66Rsw%Z)8(%-a)Cx6;r7@OLD0xLSqf%%Gcys0jaV)TD0tu8_`RPx8#OteT?v8gX
zdF7k?J5)WI%YXVF^8MGcDq0ljXGzG<?GJ9@+b?-;&+QLwCK30;bYQUknCT4TJf#6P
zK+f^rF{j+F;=!4v7eDQK?Kce9wL{RK^D%{_E1bgSf$&4Q0S-sUv|4gzt#u5Hu=fo5
zke-cV(~w5LvtEAR=V<%v0nZuk>{H*qALi<k$61w+N-miY0#iyP30}p;m{157mYrgB
zc2RrScM`R#C0Jdx;7U->XK#zu=WSTYAx^t_f-W&QTG;Y92z5<IHJ;;eG*C&L>4k{K
z*a5d|P*fV&r=^4ib+tGeepYOFD8VY<yt=w(pJt{Ix4j}>sVdM7S$f%L<a`Z$?Db;b
z{_(bNp<~g_+uL@wZ!8aJdH|}~^u3SKar?ZH&G^~bc5awagEp*OQjOiIYGLhd8@7_x
z+f8m9ixCOJS~Iy84~usJ*<<8;N|A58hd5O(@NOYe4|^r_RD@6;&eP?bjRPMXk@k^{
zO0OB-U%t5E&8t@|$9s<Vcjg6-fjK4S5Sb(}$G|QI_9-$4p)5ygS*)^=|2VpZ0id@O
z=wYJua$^j?eOEd@TPq#UF*mwN?DSoOZbje}H+K1X92}c0IL;tO>nTKblPSc~lwm)0
zYp@D!osLJZKszTd==kI5=YMSN(|Gc^!D8dfBLMFw@#H}fHWh9|xoJywrd+lhGpFUz
z_;udSY1vj+1^{;HMz|QG;p1G=a@WSsY1z)|W^f+f=Zd)5<1dsvmt&e?;F2eIW!>e3
z@q2l2)x*P5#0=B{<^J|Pr@LGBPP6+Em<db@31t%HdN*;HBa{Wr3tCsSX0&8a%d=`-
zJ@u*eK*#lK{b&)SdS<*DD}+6y<<MzaIxW|2ZHx9aWC>RIr6`nYX@ZJw6`HG(Vq}W3
z^J80IMQi27^)+u_zGMoKu~KvuwP4Y(?Mxld=(bM1mxh`5IkfG}KWCe^6fy{Y^!*c@
z?jH>C6X^KzbRR_7F&N3eZxO5yW$}@!&LUgGB^@^dmiRhtbeu~j7pLPMH8dR6P39aY
zs`IJwXER8R6QO4jR>Rkv)4lPr%ek}(ZiPd%6^x*rDU~F`Bmv2lP?f8BW)~xpGoch>
zsl=)Tho?wtPWVtOC9iHgD*il+gceWP1<zbK+Zg9E3D9wOdn%xu^JH_5Z6n?Ok<Mqg
zId(z{mQm4Yz1Ct8o&&wLhqnC~Q-k7w*Dqc==xOKx0RR9=L_t*W<1fD8_4PG3yS-!3
z4!l{~`pPgQp&73%DzQa~KbtoGKD6futmpP0wRu!9b}XH97gJAyD~g+-%P2GtcRcwH
zKf9aodv@RO*_q%*QwEz@efGrR>MnL~N4<x|3XqaTF&vLa14Z=?DduG`##=2!HhQUc
z$K)7WOLs(Dq{-F?3mqT(bHSy@ZyM7XB$q8)sVUgouwsl%UbXD)%a{DczyDvDf^gVd
z2=s1F?5D(>CgNPl5}9ixm%@D&jyZGt;g<D&AulUA1gZq)IUyl2rJ0;9B`ZcV6r70#
z?zw7}X0|nLW8Z{^ttv?ycJqc8A8w%5%DSF_K!`h1oJk>}S`dlMDIo}@RDx71x1MH@
z!1Xk<n<hdE{MoO6%|H3|KjdHi+keS#-+j$*PPgP*9mho}!E^qHSaplG;Y=&|?5iBL
z7BJ6!vtxJ)tr4pUpbcYfv~^tQwU0J=;*ZC6{8`g2fa&Sy554fdnK0)H7Dgs1nL%{9
zyQM6-pC=_)6|<5{2Ncy9C}P+;O`sm%)&eh0*c^LzE;wIUV9$DO=}azUUZg)5NO|qm
zx!N;S6-t41<^KIUqQH+|enAq8O6*>A3Cs!P=7yzY-kt8KB~y+I%iSHv)53JPBE|$$
zLL7gp36AP`Z_?vy)b-S6*!;ujI0Us+!5S&KaJ;`I=R&PQEl^5jSyz;3<+XDmFBwgR
zFa>r~Vq)TOH8bBPv{qhSU-RPnil2V<CDT0d=YR1(b6Sf<9*9!JQy+pU`Ar?O=)6KV
zjQ>#Rm?wC0OZ||RlBF&1u5Zsz;mK$8KGO1d9<*l;WO=ST1s+(0`WqjiW#~K^5mHL1
zD9hdb-<p;;>KZKsws$u47CbGl9s9az{cUQWXsSx7CKFUYk5Y=HB>e6-zk$1Ze){E?
z>{H}gEOkAqvX)BDm8rVpD45*vdfxHP@yPqzJMQl8IUY~!u7qjYvAa5;QIQY{9%g0{
zX2s`o)A~1_9MAe|I6{}0K@Vv;1_g<RQ<Wf2!kiMjG%;UaGwEsu2?6HBTz4E)i615?
zk&sU$Rrd4DeoE}48SA80{_IbF#m~R`iof~oZ&}Jp5GVq<t+6{Q`gx_8VZM#w2SUq1
z209)?sMJzkW3Uu269<)M!22$;HPCMu5faXV_KZ^*Pg1pT%dkCHfCdZkpeY~$aRByC
z$2_ECFo#14n6bo-S|UP-2?f@Yty<yb_!_VG{~CtWccSCVLygNb;@k-{&ofuM9mn_Y
z`A{pr`pHk(r^r+*r4-9#sKJT})rQF@Xcc0HiOBW5r!1MZF66o*QrXRW8$-kYvng~4
zhKH-Ui4UG~>t|-I=X6|aMa3hRTda$R$F}rUD}-WE&eJ|JEhkHfMl5|W2AC#l6-rh&
zGFf32RGENOAgs!dUcTm+Uwy^2to**fUoIJv5T>-@VYS`hPH?;*CLfK4yyy1!p*<V0
zp4)%a){QDn***KxJpnDE;o|#_o%#`Z<AF!&k!wb*sQ!JEh0o>EN1OW)HU{hmW)Q=I
zL7P#kx*<heVAnIxhPYPW8GYxJj|mZDhq^zvcihs}&U@JJdu%K&bX<D3gDWhD-JZjK
z&+T9Ro+$_~uU?qPB+w%_2nM!VqY?#biDn4$C=Dx<U5$}V$!#0K!<1WBn~}r*9ZTgl
z#(Fl6E-|&+2a(~-Lef0Mc5R4}m?k!`>2VEot8CN+pkUEg(bxFJ;ff!>eam0`#eomu
z{q}?|Cq5zxUM!njZm!Rm&pzsX+$iHL8pv&TgTnAB&%fO9<Coj~oe9M_Da!d)c&c5}
z@ut3;0j|=NTkW20MHrPdd!M+W%IR1y-f<>9KkSuWjAEPDjmo*0?=B=uPmDQxcL<Rb
z0%grC#}fgyXsZw`s<Lig^yJT<-IzRyu(aJeSm2sxdNVvU;lFj?l#Qy*h@~`8=&gM^
z9rr*2u{ke;QK8L!Q3L*^eO6lJt~mW}WxQ3D{WSCX`i2+#E7Ft(%r?DPi=n^U&b9UV
z0nsshgeSK(kd41QMav)M$^O4fy`t}DePjy#v$k`-D=niNKMZ1~)xQ&3K6|IlNIy4*
zJV&DoedLTB$L~9DxJU=uX>H!kTpf1Qb>(<_$5*ev;A%=1m7&_J@dfqtBlVPGg(PD1
z?ihhMjnoEcH16Y0@NKl0Jh|apZT+|N&7bixp6r{)BG(10niz;75~qnUP4o>H4VNq?
zOL0cMzyeljEryneZR5V#?Rj-|O_I>NFT?ze+#cue&5O60Dt-X8EN$Pxxf!L=@Rq>-
z;MPx|W3c6UXkcz{y5QToefXHM=drf+?p(GOvAtLIoQ|7AN0knq-ki0H$<_~qTxD~%
zJvzuwrDMP5F*<&PlT6dh)!_;)m3%sJb^VHcN=Pk6!<>u5s2QQ7<MJg5OkxhFfLP?H
z7*1o=t<QaspR}26#kQ_r((%Ue9=Xe~27@3mn8T@MIkbIf3%9M0-`#!-+M>^eeVTeS
z+9cBD@g5a>){FCf(!ba7yXW>pZ_fs-=k_1BU23M|SEDz8N*F*~!^dy!k3Rd`Zx;Q?
zQU0jip^NHIzDXn!($tl!6kGM8K&^#kS?u%Vsc6w4W>ww0A|Mh`tvv#(DQnhu>sW)H
z*}CI!&<aR6W0Res4=Py$K(r4`5Um%2>DWRH%t3hb`Xz5)z2e{f^MA=cM*hj4{*)<1
zOCKtr1@`lfnhel9mqH4rBu5EQDrH@{J1t~2qnFPdF-G=>;N$jCXjQ}Nc*1+q@WuvB
zqf1*l%uVwgH3cTciT!-V%NH{;@8RDIxhPdEJ)vkNFDuL4iTzbUQtWy4(`3fB^_V%t
znIB!h<R&Ec6{Z?YaA0GstvcJ|n{-fmwZ-{;#?rF5pvV|%?q7*jf+8q|>c9y$`UiMy
zf8gu;y-oLOwEpo<gVSK!A2RgH=-34P7z4E^$0bvX;bcXL#NbfTE|3|0MXFkf>n&=-
z(=}RTh+NX~M|ycd$L+IGsNZmPHpNSalGra_-0<?|z}LU|4X59J&7b|rFSwp(t^%Gm
zRZ$J78fKQ%b}k6lX@`(dNxp8vDHn1sX4orsB3ojSeHIL01<;NwWqad%MZOCimr}c@
zUfB#-&3CRUtYXG<f-5l@qDJPBNNHl0nX9=lg~+-V-oALjKl+oOvkMbB7Y;R2vXGl6
z7Yw_n2<OPZ=X>Y}L&xxOo{Z75RZ6dx>T^NM|Nl`JzMYn1^ysk|D7A1}GUe}rmJ>ZI
zxZz?#19k@nx)e{ZJ~QHU`lj>dhFL6*$48@Yfj{}#&-v3||AOEBtAE4#;R9E1zv3z;
zB91GyuAsq+4|pV|6p&icwQ`6%e*EUZF6{Ua?zqcV*kqYjaxRn-tp8FiqN2$h)eOW>
z@??pKSWw2N@Z@e%5up-Dab`E|Ib25$btR<<i4&R<#O&2xBT^$NO-$45_*Y7;Ct}U)
zfPGf>t8ys9slchMpouUiN;Q11;sqz9o4bRrQ{}U#W!e6Dz+?IcK5E~u)i3_r((z-?
zsBd#H+rNXc88?>bg>g;kvUL!MAyP{=o?O`spdE`=-LR!*G!e*JsZXrwhAKPo*-xco
z9;f5W7Nb>0qzYfY{gPk(^k?MX{uTM|j`^*yLm(GR84QVB&Db72>Q_;MMB~X?ShcW^
zJ4&kPlF3y$9#51~jAv`H-{r|J4LsF<Z*Bq&kI=D>#B7Bg)byDUiy5w_M4Tt;F%x58
zH_ybBSgBU+GzC-xwFsr*mc@WdMzk;?93-)qz#d^LmFSM}hNE{@a7TlkIqlR}K^k!W
zQ{wk?`$4v61J-l<-kOg84h-|mYu~Q9)MhW0(c!xhRJEld^zYi|2*_jNh;Ekp;RVv}
zQO?4MTi;w)?|tcBNxwHG!&#}dQcCUBtJ<(kR5GEomwr_pn4w2$jbN&RgJZYPtkYXO
zL;Iq6l{dkl!2xwoN6U-H`!3qO%%dSIxDjqWEiCs(uJ?NmDKSgv`dqE8pLS2g{CblF
zvU`W-nAp!Vx7mO!OD@)*3lGUznwihT)~DKkKraNIZM}PRTNi+Qro2RYkVr@nOX;D<
zRz+(|$?C>Ml!nn&9ULVnux85XgyxKt(t&ujZ8&Xvy9vhLi!RMwh>H4{U+a&vJx#|C
z?J;Kp&tpcVC-j4l(Xl+RKsQN6I<a&aOZxvlX<${i6kem_WS-ut<XRo<nJ&CGePE2*
zP(2@{oDbhte%lt?zk~EY(m~(bPRD1rT+qOXqc%>_VSnJw%h%Me-=R6Pi;+1-0%e;o
z{|Rc591$}LMFL3-gQoomw&!axB>~+kELy)>AI6<w|5h*ZP^$q<e-}Ci6Pj^@R)~=_
zO~jax;8-J~W{mJEh|(#sITgIBoWed$-0b)0sZequR6D<V61d`-BInM5?bpwjj^WdJ
zvbt`6Np|CD_FZ^#qahhU(9e4%n@#gl7Wl-0ZM1BW&(oBMLC|V@C;uJLGBjD?yzii0
zPH3#X(Xw7#dCC8(R*#_A<O<u@_Oc*>B$2D#4$WqajUpXStL4m@Dr(QR%6)+;L^ILU
zbD(bPn@3d<yL@4L;hb-^C=_$P6>5#tT0H%KkVd|hmMucPYNnJ7Xy?&lWqvy!ZgUi4
zBBmKav9vNT4XIdS(}WrAq!b^UY$;6IK&LM#^lWc^p0wPi=yFd!;K>gd^!aDsiH?0g
zKN-o@uHQELcW=J;&*(TMVz6kcQq3S$*`kKd>3I7wsd3;>1J>=uu(#KHA3lkWNA}=!
ze3>>_8!cn}bT>`x_6JgmG_zZe=oQ&yN|!-{s>`AFK51D3t#g%Dt$-LkRj)3NZ}<2Z
zH~zZ`XixFv%Y|&`^H`nT99c?(V3^ZHiph3Jm#=%-3j5nXuMWduj-!Is!kHmc7--TQ
zV(N?ICkkg9PXF+&_uM|e_H4j<Zr^L`1?Q@rWv#^VS+OnMDi7J;+7B^d_?(SNZ$fc@
z{C15CeN_w<`AjJF2MXGw3p{F_fPsiU^eX9QKvH#qxR7&(Y9*rQ{SFC{b<LdCOv#nv
zhitBuD%GNWY9*_M_g6o-8<2De0$DrLL_lIdTCR^YuWWE(;yi7rk3E=&qIsB<t<oAe
z@MX11OxZm))n_mT!-abHyRYH?$e;Z5C)`XEI|Zp`n3DoEJ4Tr*$pvW*=5+}bu6BE-
z-H!XJa9UVTN0dN_b03F=6b~SmeNCI<Tc|Zq3dEFX!g3ed5U-}@<0^)l*1#V_41^E~
zAzDSjYThgfpvfjU#z+c@7!rlTsvsevHL#YIS_-==<hy%5eEl`~_6|L*-6P~X$)m`;
z%(yE`wKkN9KrcVsE3JKAZ1z*=STE?fftZe2DI*zvDjjzKVaB!h-7vgUa2${e!TZTE
zsih+<%R(*L==iWF#K5{_PI<NcfPwoPSUsqkAf21@-q)M3rB14sxW@D{Vx!}7)`LOd
z-Go#^Q}#EbLiMqYp6k*|XE%*;n>pURe8aDP_6vS<_ve(`TlNy!OC*%+N&-}mq>NP%
zFM=QhCMv5Nw5&d7AUe&KrB$E`W?V)GywpMI9R*y?lwzxwrte0_U}slV$hoka56HA*
zH}BA>l#;0!mQHLS&Yk9h8RG&>0#k+SH1YOu%{RASbG$zit3{A%2viX}|7j_trpRme
zV0`{`44=r8*&G#ilWr;2@CVVd?XIUZjJlm4<+1O-WlU$ZTnGY(s{=6x<H<Sy-O#c)
zFRhpF-waT8?w`vo?E{4kgVO2V6!U_LMJ1O~simMXKuFZAl>3D*U%p~@b7Sd)BF=9F
zNr6QT7jAN66F?M2VxZ=gx*Vy?O3lUaWF)vUETpif)P-z9!xsIiZN)qm#G}P6jj&qk
zpC9g@#FNE%@~Va7a!=Jd0E?TFYkOdeyqtD>_O)<C3Da!2p+T?!Nnx*nuohS{DMPA(
zB8iMSA;b+@r7yv@|D^r*8Pc*zPINeo=77C4z$r$<owf(rx)JPmp<^zteX@Vu68-e|
zkKte`P|778@6YHs1HEvJkG(n_wFs=DX5g&;ee9FLC5+qI+eg{uqjanz4UcqOv5;W<
zuGC7=Le)Y**@^J_i!a!tq-rvC6pI$C?LelLLE0QITQ$Rq<bsrn7CSS8Mxq!t?V68f
z99BI`qkVF`)NV{_Qv@3whcQx4ah$O&D!SH8)=aK$l&+QQ{ek_wqXG7d1Y$@`GJ%K}
zPf7?YwPw&tP@%31`E=~oLMu-1{SudfFo5>z7@85&^IC=Z=l1ujJsYr|+xOO*!3PE%
zs~cRz0oBKB@$JC94CofzbNE=hJh^nc^U{-IO76MAga>T@Tbf^7hm@V>MLeA{x#a%t
z@!xH%o7B(+W|OIV|2GNakpVWX!iA9*QPn)iW4Qh(m2~CzLfP$o0Cu|_FR!kt%fhJ?
zu3x<3dY*|DPfaLj5u|KO*)V5Bg+yeZCTdw(a_LMr5U4<2i&u_PLNHu|ro>-{dUwNC
zcX<zcNqNAaZQ8UImfDnsAeJV$EUY=(J4FZ~GEHXi)3T*-h6YmY#*hR^E!4caw@5Z@
zFIRFH*0C?w0Jhd5i9Qz(@7O*{$Ah7ww_%7@(ta}vx3zmzwta$*yFR7ox5;pwIB~b_
zUR_ZcAY0sf{IH*6-_6h)Sn@(HWjmRgYq|raH`8U4i#@eX^#L6#L!@4R)-afXj=7-Y
z@eFM9Inr^vR=rHUiXc&_rBIfIH!oh0_Isuf3@3-SAGfDE`i4g^D8ZvO#0~txL7t<U
z{Y}-&ZG<MEw)3@xi`GsCjka6gH!b4Z=~&eikUHKUEb?SG&D7|Tu0aVBm_lNoW>(1?
zfV_83)Iiyj1d<3XcElVCm5Owaed8B>48vjiInyzG2cCTHhCfquTl-Et`H`>Ac(T9a
zoKNFXh}ORm67kz;d8CuIWY&`ZPH6c_o(w$#q0w`r-LbermpNoG6!~X$nB*4GAQbZH
z$npIhucp90O=f^nGvKsHnCK%Yc-j%J)aQ!kie@Dem}BaDeO*?}zSO|kBd~#HTi+X3
z-cpYG{C-DTRyLU_B%x{{*VSf88jS{~G+Cr)vAauErg>uWf+`^d0!j)nr$AX(mgAA7
z6ml(;D30&fq^)huHYcggeKY)gZcQ)eo^1`#M#oP>GK}q$=(rsR&U4PA5Vu`yyW(8_
zyqw=r{Jb<`H8HQ@B|S#R4Yr|2`L%3^W}Ngh%+k%JM=YHMa?2A@(W5@{Ep&V-S2u@K
zZ3ZsI4QRp?0-}YyEX4iH6eEI18;Ki~)gug}rP!*9(~~JNT9lfL<2(u}7_O{Ea9Sav
zQh?@Mx;(@C5Z&+<Hb>HUs+p7Gf{yiKqI{uL%|KVQ^kN`}Txa@#R*TwB5}9Kn5lJC3
zrHK$0ibtQSDrH?M%ZcT5w8BdwhOZSid1vU5?E*LCx2}DDl>FTOHtpGf_1wO<);t*P
z><P`mpqmic!q`V6k8y>E9JL*;RUX%WTT_oPXaca-bK$Tba+?L~{PS;VJ(8)9PuzeN
z$oXUjtfov446d>%)~%lky&^}@aoC%1doenihGjgY_Bm{QYKm`HNTk^syr2H~CWLfv
znZRLp;LVFyoPPJ$oNDHJo~@*9DJE3rie#~r1;xYkUF5Aol*oRXI68(%EoQ_rBN41?
zHYHt@-43ga(}GVP6}B2KnqGW5%%L`g(!-f`W4|hb24T&aWjV2|nG`LZGQ`AgXDNnS
zD$RfuLqw`yBf^XjN;LzP1@%$1Qcz@c=M6$s+gj3lX7kjPmTwm;|9INR==i*ygPlHF
zW#~J!M#rYiiWPxFEkX&!!XUQ2Z;><BH?3`tjD0Xgstf*)(6M+V<eFF3QZ^f8v(o4$
zFt_<=3U6+L;hC`Ph9*Y&Ycn{H5iK@myTyQ6H{W-|Nwz+QGvQMm#I@9lMoamtC9@t+
z{P^Vyu0voJH(-gMt)VG7EhSmJ-)5*1$H@wa;{~~p*$Qlkr|g9m9$I@REjmkGk&r$Q
z%_ux>oR88mgMm&!;+6*83|PgZHG(VW`*|jdS9<LREPIcYFSnnB8SAp+ig;X(tM3~g
zn2M)kD_uePT<I9T8&4idr*Wgp@qRa+ti#`%A!mG!^-Y{t4YYh_Jc5tXvJAXVgeAKH
z>+gb=&oufA?SA~X{p!lJ*3FDE^5ZSIq4DcRm!@cnX4d18`}ZG6Z(nklCd0R)8L;g6
zR(`?f08iGbm9l1PE|8TdkvUEjg;kZ~Y4Jz~qvaq8A-f^YpEWG1iyi#ow5*hD3^XL6
zQpjcP;MYN0Mbz2O`eNrtu9ZWYn5UVLA|YB_Sc<}=X23e#A6d)Fs)b6R65N<)^P(;Y
z8GJ*VJzw9?kd`+?;$U<b32LO{PiWtTj+=nfmcQ8#Y!<7(d;9F%Vd|d`pPf6u>~p=8
zts5QZyqy~@)Ve!>>~p0>@ky`d9qPbqeP?d+@7W0XE_BRbyy^xlngPoZwL)Mw3!>Ex
zSUcvJEcLbVbroW2;}IaBRge}1RzXXpt_21ImLQzGSW?cO-cK}aJj1zo&{|4=jLm3$
zzE3X5x__YKl!)_5H(q55vv?+npFchs=%oQGM&>jV!?CAWR?L8vPbZG2dzN*v^SCoA
zb1Zgwpl^Xqz5*Uo8=u<`xIG)Np4<1<N{idBo(|#%njMDj@qQp688f<sy%85CKjnNf
z!-_h%-$$-v6G<Pxu5~!{KU+Vx`<hY6l&w;Wh5M@ACK1P*@^m5aJg0L3LtGz!x-S>~
z5>xJ<nvc%pbsNQ5EQdBFLc5X&j-B)BHB~byuS^81Dk$pCp<wAYD#S7<0oD8OR;|2d
zmeazNW+1Vy1&PApFtMx)r}ao)Tke-_j*i0>LhpAFO92$KT6C{_uiW@pbK^iA@U!MA
zaoFuR?01AYASn@IBG$wd5|KctR+fAYR;@5O(0f+H@zSMOTBBC7C=KVi?W8Iu0NQTZ
z;L4*{UMy`;@+oWl$J3sqV_`Fp4tsgTp=iNIt%+`_VWX({SZj3L5>g&sfBx5Zp3@$E
z;b~NzjvX-EvCcY;$AHGr70@T?*gI-s=u<+KUi?0IkT0)l3hfhzgNnsr6a};rfxMhJ
zez=1dSEM<3nppF+2UjV#2n1IfY`t34#;RyBt!7TMRUBOvt(NA5wDQ&4fbIO=ln-2a
z{DbJ&Ppp<6l4~Y}f~s(}+f!q+e48-8pLV=BTyqL5mBJK^w%fdgU><GFg|cQ-?t@T+
zrDRv%%k5wA^ddvob~>(BnBX&{WBBeonSqusm7MR)ll{yP5wr~4NSsD4CQ}}g!^dB6
zLCdJ)<`m<}TSVdC0WDt)dPIb4QHPgc)(RawyD92pO0YVAP=qSNJWsrMz2oLEvkNVq
z2bbDC@}L+!n&R5585mAutw<I7QP8Z^)l&ZmLP}6dWi1=VWTRi{qp>K|5X3RjHd^+@
z{6wA%4JW9SN-h}@G3?M{hV_ZSE=(Nu2Z9D9<5d<buXoqH+8>zHObUTe3sEXDs>vq>
zm=ao(VNplJx$N_eU|)o@$V@&{S}vu{IrKUyPRA9Yinf{>{YU+*6y4HH{~$Vc<7AJ3
z8os}r!%vfEFJ7P)LHH;gSEJ*W+9oj={>McRDw`v-{dCf6Om)bQY!0Q^WKlj&$IZz6
zkY6|1)0|MH6!yEBH*a5Xb=Z?oq-4lK3Lfnr?GTaz(H+ydSek4tb)zfCrwmmfD<MT_
z&Z(-Dve?<7+nMrs+wcZ^1Zq#2e!{TW<P$eGHG{Oc0ji+HG%=-_!?a_Hlj~p>(Vc)f
z1ZIg$DRv`Si^x<lBiBSEuZ6W3=ADM~s+!uOyPc10_$tbI@1NV>hxTm1dT!rat1W=Q
z8EiM22wT;k_BqhJD5B@8>H|TxeI>)+!Wpx=c*KJ?GNnzyyp&4MrJCVr?=L#9rlLax
zl;0CJ55c9^`Mg6oj?&{yzfILyhvo4hKF)MoYRrIt&?mm#2U9ddPrSM&<&}hoF?+h4
z0bFnRAOeP)p|pS)^Zw*Q&ebZ!q)07_Xh6bDE|pr85)_eY47|^wu^|;qKo?UU$M4U^
z`9NSEg6nnL5h(<wl$erLMG-N4tQaCeVlPb`anF`+UZjFY4^&Ils#=Gf{7s7Kg3Z84
zP0-tzaGUnesXa!=t@S<apS9n6`RcQa&-eBNY>h!6JfmaIKf7#?M}s}~@JBxB3aY1)
zI!)B-6^#_frwf~s+wOfzTGD}DA1Jz)$CviK@!%o4ol_k*XSid0YBN3ruxP1L$fpIl
z+4rb4Xf&m$<LEWbZHokRgLF|Sj)_wtP6?4f%$XvNH4<!|LTE=)7yJEXE|?pr7p;K8
zIC^{KV|2VJ6?RSSo?|62PZQO(uAX`nCw9|Js)0-;N^}f0ny%e6FiWv4#XP*83aOl@
z9&bIAfsSSC{bxqU@ZETF`_AT_^YMNnPj0_9g$Y-Dw-;^p1vsbor`va-WlyPZitpbA
zE&G`_#s#0LS;KYP9JH}+My7VBMN`Vq5!+`mMQ*MR?C051fO|B8@4@P+LU|j2*~YX_
zcu$|Q@NXzem?OkM48E@Cz<l&nNLoA^7X2WYLDy+{D<|{qwA?L$YFIy2EBIPSHG^5g
zQ+#8hghG<&<EUIuJFe$FDMrWGa)(+Rj1tY5A-WkRWjjN+88ZHllx>MVGg|JDi36K`
zNXNc&wNK-tZ03{cc-vNOqg_N#4gcsk-``<eW547HI#wqG57BHs2H5yqgX>V_?Cd)0
zjSM&R4i8wHAE9G)I^NEX(GX3Si&U+^$ko*qyJ@mQNN$u3lRK6|+|HSzgEdimv{94W
zgNDJ`ZTomi38|ISI;7K`P39x>e5L@M&r2WnV>}t{j_)?e7O@;c7^15pMXUE$Jbho7
zqD5fI+4yX~MlV*;WG_bDsduvd)uLCm>qFzQ`k9gQbNd6^vjOY5eSdB6_O{lPgqAke
zJ-*sl2h@y($+}@=kf+a^Gv(21KJw`1O>KbN#yl>bad~)mA^74&Xb-P8j3s+e5HB-c
zGNnasI2dXOLd29P^KRCTPt}-G^-?*O$8HlMxP0BPXtW7wn<Oz}>YVjdD@vidx4HGc
zs1mztxuqWX9M#4|gk@Q{|8U1IUcBT*ip*7zQdy;R<zN#sY>oO@N~IJF^R8LF!k1NN
zOjkQ{h5M5!$IFuYyzb(TAVPLQ!BQHU@hYL}p=`E0F2<|lA0}%E-3Z#TwZy6`?vi<*
z^IBMIMzR^`QV6`<UvqP{<1k-2An=hTA6ZyH3Q1tq!l@RfWTDsDXSgZ2R>@5_90bAE
zKe#bK&W#wKMSF~nwb8GsIAv7s+D}%rY`V+Gl*EyaFXe1o!!4#QjIs>Fb4QtMeBTAp
zw_%t9s!)ra4-%-^qU=g3Xt8Lf<_WgcA{V5~IUN@l#ESRTyu_Pe-J&<n=-7$98v%#;
zu=AkM!G-<oDd?6J;2u-m@(Vc}cKrB9uX&vg%qg1E3#e3&mJ<ib*7k^QFs4{~f@VRA
zr}R_<c@JSiV`M*Dy3z4;k5Z`xgo=bZ2$Vx~R}3zECaMM`#CFP`Z-EDNya}c5pe9q<
z2O)*XtE(GU$(&9rSt`5Tj_dt_m)AFZ{o!l0tVAqo3JfHBDGQ2{LL`^Ml2=QsRH%a!
zUudA_O2Z4$(CAn{Upj_Q=E*cYs=DHv{a!u4@7s9t16p<ks8PUZ(j429YVW7evOo;M
zBJ66R<Vy93v%ec!Zlwawt7Sgg6ibIHo(hCuSaFZHs`SX}tHTw)`1vn*$;?Zcn4{xc
zO=iT28ZA;%0hckPT0~M_72J`er9i3XoQa7VBQ-^C(!{c^R*<Hs(Xuqdftw#ZWUm<v
zG=xseO|JPiT5fz;5K;`(6sdEhCMBmzA~46mK3CF`S#KBe-HCFyP>u__7Jm8Vk9l>t
zCauw&f)PuJ#0pJ>3U$e7F2pJJJ=_d<#d)!sC}3y3C{@bmNz0dXtc^bm{I`jF7){n{
zkz3WP3ysSDx6$#%SG35?wv)pX1lw7HJU+i$VIhAG?Tq*c9oNCQCAi#eKT1nAvvko?
z=|-S_bc~jcPi>cUEP@6rd{Tha&IYjv<JG;oX+N<)?0I>(;buN?mG;arl7cXcK^CYN
zJrGB1m!OE+VH=`&nzw_pXOpR?l)NZK;N{EL<XXrm^}cd{%!F77(bxGR8f!D?d4+CI
zAr`Ssr58^4iv{b`7lYwh)vDxNNFk8s9Y22aCG6mLx9@osZ+IE6*@eh;O7Qv(H-`gD
zsVr+DX9tB31`DvC_f!J8+W9OPP<srK5EZSK;@s1)+pZd%W1kfrKexYCdp2M_x9_)I
z7=zUR_ZZVOgANAZ*wX3xLA42xmp2Z&^XS-paP5OgE*bjbQne1oh~cRWN9FLmEoEUn
z3-xUk6dRZ!gHfYvu{|92G4J-~zuE$0rO(80{U{jtds?`$X(5-=$JV#jS}A%7pf!`%
zl_*yRw76s(W_`QI7pASxxz)nfFX#s5hy>!iLyB@-7snY2pqWyvR-#JmwSpRpY-x{M
z8rdd!YzdiLiVL`LqfN12S%(kikRaNaYc)(Sbq2bhXRda8^B%FK3*th33|=Lv&7p{!
zZ)8{&p<#$<@5T873vJOqqxJ|LTQ$pvG`G<)8y3~DZNGKo!bTV43%9Q~1J>YuZSQVh
zNxzz*^JD5#>y*<mkV~<OU~NCCeIACpyNA$?M%wnh`kJ4eQ|%oa6<>&x;0B~ozQ&Cw
zc(t954YCb%n}>F<2veGPaed7}W@eHJ)If7v3u1Ft==-_@GIuSWLHYCkx<PrGEHxm;
zNG?_d&}4S$3nt?n^vAVmDmnXkjE=`Ewtb@%2SbW6u%Bmw3d^!mk`-o1F|nIxq6pOO
z%n%>HVpvDfQ$^c(X%QPkVR(P>_TBkN`u6d4`0UyfJbC!1|85ajbc1|reTzSM&_!#s
zEZciNsynti0KOl4-}+5{z_Zc0eSA*K5m;-b*0MRJ&iH~G-#<pnA{R2pwlf<qKBr3O
zyRu#0<()<wBQ1B{**=mj-8BZLY2vWE;y`4lk%(0?i(=S}L4p}pV;XYT>t^#Fp5G;j
ze$EKHc_xIw-SM7M3&krcHv?Pe`Qi&B0@{kMw8eUgC)YDWWoHp4Fj)kYRdFGRR|{3K
zv{tv`ZJ^o8E=}yxL|Ha~@F`8q9$6h@a9N=u?(8y|QMwS>Qk9Ht`~YbAVn5qn8|ZZ4
zg^h2OYP|U}cJABgxJLlXN2Y0HV1Au8Pg0|4hR7L=E?Px9yH~c)`3N2R*gIIe@0U)W
z;`~L~&UHN>+QyMbzIy(_*-&-s@A!S+_p<To9{=V1YB%keVq!{x7&fd<tCZ-RN<AG{
zlxCsarqur1>U;Vb<&J6zc3#9~d3a8r2^O1tz5N^d@0X{u4g(#J;P$UbuoS|^n?*Rx
zJJygnEhox+fJq#FM7Z7`IPCTvv$HlD7mHMjAyKNDbHP`x$qgc*eZshZp4;Ew_H4j<
zZr@vrbiK#JL%mW)-J(;*mWz^W-XgCiJ67kcLzc5Wuga*1(jzWf6qW}z0UP6ODeoDJ
zG!Ib>m|T^md0A=w3K$^ekjCOnCgRV|nWUf+IUbKJIfI%(YZfm<Ux8d$*Ok1Q7pcuj
zahGTFB(<~^0g4-Wf`cs5fRY<nvTVn=CVE(L!wRr@5j0E?10h*dQ>$KNkr{!a!l_n5
ztn6o~dgAtU$MLk|Aj&?7N2eHMY)}*ZszMDWrDPZS%_|&;C9xw>Lm{i6A#i>DmRt+R
z`ZpZH1o4Xd+xIM|m9Ps$nb^4^EmYQ8SSd(uUgteQqG{fmQC)E~Z!(MF^h}e5Dr@at
z^cvxQDufl#%2)e??W-86S;-%c)V!c6^XBzSzIyWtL|IE_(Mk=bv>&D&*Xckgk-7kr
zJ%6nQ36Uw7dQr4N$U;l8sAYi)Nj;5H0xeqXb7&E0dQ5a7r53|IsOR=O6v$N#tlyn@
z2D%He8Xd30KEL$ybvh27@xo?6YUf4N;X8ql98*V?wb*ACv2`hsd1oC5#4w@U*rCb<
z7SU5Ot-4-RC01oBN_3%gU13>;kP;%1TCG$&5(u)m*HqoBJPM0C402S`eJ{lAl3KSi
z-h$Dwj(8tcI~}X2rPBrBaCObkfAl9DN@ZV^U3BcHVCT{#6BO+jRw_zN!3=eaVSZ_;
zXcef5AW90#QY(bO%a<=$b7qy5_wV2F&HE2rT_2FRL(2;4X~VShrIIxhik)#mM>;O3
z1E?cVs)0(+X|l-HYJ04pteUwmC*DM1KkfM9#cOWYJO1b2{u_2BAa8`-ZpY2d4dH!6
z7uzREg(!&-A{rs3z&=H0nMh2OpscmBB5qhuR4HZ#kla@flv<9TH66o8Xu0vZ3tC>Q
zvaG(B1tzu(m&(Ai8r_b{(?H8vjqlEWe~aqyfHvdF5kkai*@7LK0%P_Pry+%{(%Jd!
zt13Y&6O|>e<eVw~4p9iD5~UziR6b72A`6lP>EWf@Zn#i;2gCVB%k5hOb=IC<UW0hr
zs(LX2<F5okLxoAW-raD0^Mb2%&6QTJv~mC@l${2mL=p)COp~yxRnyF2HN*F!P%9y*
zRS{(ZxuvqQEE%bRpZw@c*0ORcN7i?l+i#A{^Ny5ulyHkkC9&W5y%q|YPz_uva<I8#
zHN)0G%e5|4onv$y@D}ac#%yd`6WdN2HclqCZQC{*+jf)2Zfx6bIKi9y?tLHL_gV8{
z)|#{aXP>=)+ai?bucUX2vba`B5~d8&V7;-7#sb--(?KN_GR31aMIKsegq-MPR5bz~
z)s?fsSU%6)A-25-=m=B4=Ou`Csv4cZTB3-TGs!@=WwZ0wAZ881UFLkFd0!`sbFwME
zBJ_H&M7}B|!|IaKiJGPaalw8xl@)5GHDZd5WW8quXIpz&AfB9ocr@OnTDZ=PS<{9B
zj;XU?)Ha^az|emLO<tR5tZ`XWUHW3U$i$s5$HOiJ!(`Y(vKPK0MSKO}(lUexRyMU8
zN8_tn<hofWsCvU!Qy%KFC0Nb@3Z%D>XOv7q_2{HfTt*y-3>pJn_XPOmYJ`&u%0Ebz
zgAxk=;L!{nTq)CUcl#^mpmA|AjcEmHS3%<-05pn!X~~dCZx$k}p}J}~14cF9APJ<@
zGv7Dp$%tI-6^P+x4+Z$z3dQv<Yd!OXZq~Oy)3EDm6IKa-E5D4LGjU(IRd+O|Om^1(
zL}*Y~mJ8v&x6}1#%hgQQ9P(%x<*O*To;D-Ev<m03(89;II&=noZTqiDW32@6d668g
zPmE5$E7sLrN@x^ZyxDZV!XL7!<bWkV)IlxD6mnIsyDdJaUSjR@d$UObiCxi-Eg+M8
z2zT~I(07KoK<x~#0q@{0w(<RbMclWNyig+I^o9bJv~<v}mEe&Cdd-d-9gIju)TF}b
zyk8Jd4l^~4cPlXbcSDl~tQcpXPWy{a9^ul*U0(=HAj&LV4*;zRRyIE4aPCS8*!S#)
zN+Xs=L^}vv&n_TVs3*$#_-KFxmNhgMlmSSMQf`@O38ip6^syzsR_${OTGP%zVqv*$
z4L*87cOxg7VT!LrSN;yIA@T-G(NbW}4bb%8zxC`_+<L8)HocdwHEV4;xoE_*-1)@H
z6t=)6K|aO%OXOMg%C4t^htqC$@9mF)b}6eBCp+T5_gc*(QpL=C{&<Amlc63JFDcb>
z9D%=7mFas>_%EdkW66Q0@~H7x{MoX6KSLudByUrKfSJnuWWUX#-%vKPft{@R?BW%M
z<QX=n@5HAbn!ei<zNul(jg@fSg$bf2$cgBIrdm^)QDrvXn8UCZEMlrpbCMc3UgP50
z$+kVVb|V(m#K?pLZS!}}@-TmvV(p-W)Vb1J+@>QIm)1=2b~|{UNEL}JaiM`}t<Yi^
zw>TfCF&)=gSgDR8@t6D-<4?)W{QtaTw$5aZBJ3Szp`QyT>{~yh=Yg9ckc(KqO`u5u
z>P|l$U}pbBw~=xY(GWcVeW$X5_+#V+8Y|S7TuUrSskm4x<}jv?OswB+B`k>M8j%i?
z(pre0Yk9p%(G!in>JdV0<-ovYz={v1nE7i*2h><og{K0k73*pz?<c1okZ9#|{T4YV
z+$6JMmzE|#OCF-yNnjmYD3v-MJHERLzc=h$Xsvk{(k;s4t#NmgoUqM#KM5wJ$CObq
zB-VrRv|WF}!)cufsV#-&oAm$}L7CXAe+EIB7f-DD0Ku5gVB+r_E<-*nI(%CoDrVI`
z>t6h#dAHgYZ1zl*7Q@tbL^PWzpZ-?5JZKP0q(iD%>6o7pdq?-}-Vxr_JHhZOS$=Ep
zIcJ&QH8<Bg!M~p4#ljkP^>5H~MA7-y4Jl^sjilp?-Nr(fG-YSB1(-^vWF%G<c^&G}
zCubAY!PQINKL&2`+6JWtOOL?{(*MN(ytgRT7N>?T+u3mM>pJ<B2YHT=G4t$OnYM@C
zkH*StCdzBNg&DBQ^PNBkHX%e`x_j5JtNqp12)V>Pf~)*Tz9m2^Q3ZODQ-pf$^NxIM
zz-=Jl*9<{x<%zTk(mggP`tz1=%aOp#DXg(F6504ym1%ojnc7I^;&`y?90F{Oc7~Vx
z3W}s+CvRXgP-0;SvVeD>{aTVQ>jsf_>}n~yh<bH>m59+Hs1w|O+fHIsavq(FqFZtB
z_ofO`+-QVBNeW;~OSHdofk*n{b-bfThMMt>ma<fnv|od(PBR#IfQ_;z7Zph>sn_c>
zly1M}00MynHc~Q*)K!BRtC!zHXbbwf+k;Z6wwdKQZR_B63bS#Tk52v7bKF*(sWdWY
z$q_-wHe=DjXJMV2SPs08z4%)a=uT1tvA$Fp*E4lp`NrH#ZVr+o9|lGHr4oxDho#T9
zE*pyFfpJ{*cI&-%y)4&m=l;&duc?<76{1`|P+|3-ezp+z=}pr*b&bCFV=fG>Qc%cT
zFmB<5fqxkLFatZVlIWas5Sv7kJQ)`&8N!Sta9?FAxWDvjE`;Tl12vapP{0S6j1*7z
zA@)sjQl%<P8|j80AbRzhNvcA%STE@juTlzhs9OsGSyvnQWC@ckCTelRw9&XhUvTAh
zT+Sl`P+lvqbdZ2t1vX_~200_Y8l_YmHZnwF9p%kODc+<AcVT@?rT=?m+-|akRs(sr
zBp1noJZ`E52FJ_xFUgL8(Ye?jru?Q2l6wrrtt~azfkWi4lEAH~mUX|%7TW0SyX9$n
zY{tSo{4kgU)VOUnK$~a46l`fxbx{~0?W(AhGvh&4MPc<szZ0*rim0-aAf?^W?Ms!F
z6871@7FDa-I8wS|X7pl0y|@=l2Hu0sE_ak~&Ye_XJ0GhU^BaLkbw8=*;S^Pu?p$MD
zq<MuQi*OQ=WqRRI9o2%zU@;0qkCSqS^)3n;oZ(p5M5q^}O$CFyR4Lvl8tqVV-GJdV
zwtX*2NoEo~i>3ruG+ILkEoRj?v=9T3$n+fSiL=Rss=w?jHWQQkNJxxZ?=f$O;p96v
z#gFBDH{Tgse3CqI@$vL$<K*~Qi(!nPcT5i_T+nsh-;0$5vA6iang98acLVn;taBZ!
zY-(yJD~3-?w9b$IS#(XNdlN?Brm7Z|gO*Z)5)axpTg|9MZubc$<wXVT@n=lx(TA8k
zce+2dmBV&za)5ky&m7F3eb-ny|2RMlJGa-QToP8Sdv2T|ie6eSc=<KOE>8#U-;Rbq
z5tCr<v9pwoZvNQGfRp$!8|(UYi@+{pZt&XgDR<r~(Y2vpk8DAhwjam<pUbnVr&>p?
za$8=N7-xIP&acyU{5bGGCJ0AW?L39@$=0ka1Xi~4dbZ70+pw^d$W0K)W>)s~=C$F}
zm<thGyEKMgeLUIRsA*o+oVV2M!LajZ{z*WY^Ah!^&5boL1ks1V%`o$84<p|aFR0#C
zF<BGY6ZVn(ZwAmyOjbaS)~RJ1bw41QJP@Odu{<J+v?}dFt!A9r?kSD?1C>!OH}0vJ
zmW5c)1NRE9yD>2p-|b59h7<}WqI$46?)y$m>r^D?sf7w}(6iqi3&4{SOBQN<L_A^q
z^g~*Pxy0J2z9B1+F+l^OP%0IpCI}3NC^1R1q4bk=txl|mOQZ|l+mq{+x%a~_9f$q|
z{o8sqwg%ko%HcLKwQgJPzVkAY!izt}a59<CJ2N&UtvxW$H0X2TIl8phk~itjH1NLX
ztospVI$U35uDCH)@hqilRG+~Fd)j(8e37!T__pYf`DD%q^I3bkQZEmmmD*TaJrS9O
zR=g9&%y1SlgZY3`ScV%zZ)zV!ipF18Jlpu}+Sb;FHJdMr7`fS8<iINYFQL_0Y{5bH
z*@YKN<7cNs<-cGPT?L;V2p*=0GJ?Uvg%%frE;o2ZQs$9^du6R9^cIYfo*-rVGS}S?
z7m9S7Aio&hpUZ~HZPR|RE{KZrjn6y)n+g_VwT-2I<L1fC%d)n>SXUO|1D!q_{k(Ty
zPpW_3FEolD7(h<SRnQg^qvAX(c5^rF(RN^(cFGm`nm?Or+js;(l$nYRBh{57!JKQf
zEia%Cp*#b<B1l8{QTj&Eu0R991xYXjmej%9TUHu<ei1>L`Hh<**_ulFwz*;k6j>FQ
zCpuB^u*j!On;X|$NLgf}%9;QNs}gYj{`XIXMzkL?X?}?)<K+Bc3e^=hQ&KSNG#mIG
zHA<Fdqz-24CLQ&}dy6{|e|!&4!zYZV^#|uLzqkkGBr9;W^NWsT%bKT{Lg$|6*)FD@
z{xj$mV${x?y~R}%>|YlG_fb3|vongZrwaZK3I3cIdBjol>l~9hX@AlOu$JQ<28r1(
z<Rqfhu-av>Sj8<BR;dwBVp;`1CG7sPx^gPuYq6qT$hWYl!pmP<WszrdjYXWFhY{dC
z%Id41`V;J|>3U8dJKU8PT%OsQ{xEv$5^g2gtQiHqsm@0NWk=F=Cy=9REN)UZby&mS
zbL)<zWi01Q4!VhLV>vtOW*a83eR5sEHV?P|**vg{G)Aq;iftyHR2P9m3ua%fq)cW2
zYF`;4uc#qcXEIET8?h&Et7AI^Yna|=+;=Y$%|vIRyLDP5jFFb>Sw<(y*fZ_rekfy?
zi(>Pge!HjoQ7^wWq;lw#)ZR`Y+W^O62R@?<>sP2v24Iqh59DI>*Xx1zJe3N>(^S_A
zLB+q^K&p_+<C}eGqC*dU*3-$Vo4G!`%>bwrqeP0VHSV?h5M>Nza)S{|94ImerdiHh
zXxSfXGK@J(sk$uujlMa;Y;7atAYG$1>iZ3=%lvz5P7?2C){;?hW|35KDe1ke9b5CH
zQ$*W;+hbT&dqY=Pg2wI=zNRkzs#d56s}sSUUExRB03!xRV41&yB3C1Pe!ttpK5JO7
zS&{LI^4Ha)j@jJA{(Tn$pKIIc9h<q9%NDy3m+vPl*e8k5&mL>RUdo}Ip-=n<#FpuH
zTPN}J4L1Ep)N?F$e=1iY2f7)n%aT3a_b-O40#FF+!W4wf2CWoZ%IWDn3EzFz{DL$O
zTi4gM69PiU5H;`P)(=Osu#Y-uchZ8itV_M#Hp^yiGmINclvInen0>NCbFgRR{#jXb
zXj5p#pD0(ca)+AC(gHQpLxF6x;@kMC`%Zg)JX%)wDAVhf?Zyn_-^1c|9f5Yj?|niX
z)KP9*^aESp2LyhlEgl`A375?_B7*cR;IroMg2(WYgg$b5zW0Vw^YsZxPT1DhoLxD1
z-CzLPvQ5NdhGRPO`jfeUSeDS?m)m}+kVY{f7R@%OxH2-}KJg`}SW8&}c(YC`7UD;n
zbN8B-GVN6#j{Vg8Wh!IH^%3yIDYhJG_Lql82S>Avf1B21qR^pk*1wCglZ(Ev=gP6s
zs%rDI)nitXsc!~Ih?Jl-L|j)5<nmCanTDPnYgbq>Ud(}$ooBw10zp1tw$*o-yUv8@
zJc!^WyDwLVFh3^p#7B&fW!Q|6QV#FjqDy;=jZ?!}>FaM<z=SRAk8UO~NuW<}&~1Kn
zwe@#X!vq5F)iOJ;;PK^mk2mAL#-DFWygluo)xIHdAm{T(e$?X<c#b-0_+kuB8Ysm;
z9Vb@^b<!VXGNQb?k=KWA2}X%2C4~*9J`ulU)nJn{$_QmS9bDI`IJ1pCG58Q#JuRHZ
z2OvB{Aa2Gihy4Jsp{@c>yqhAraFdWf%JJ)E8otqkJWCe@OiHXy@6%;>t^9ZDy?@9g
z^MpY${=(6=_kp+gzb)b_BP@PKa9|6_|MNkv?XCu(D82S=#K=u)V`TNK-pFus6JZL#
zyO|iJm-uGqZ(^9ORPOND980~Zd|oxMy^%vyE7Phr(sdYFGTTHwu3DZM+Yw=1$`rDN
za$5SqwryNfn><pl!}hOeVt?eK9&Fb8A#PWs4sMo2o_kJXRIe$;KnM5?*9#Ew4N9x(
zljL;|qsX!_*~ZE@G!)kZk-m_J9Vls{TRTc`y_-ZiZ{nE2rhT)BlDaIE$LyBta)TmY
zjtX4R);}US*ca&i+_cYph#DPSG*m37_xJq_R5PRaPj^>z6@7W?L$i&Z<BpfV!N(rb
z@-mky+=786YD*;?W(%3C*pGE_DE((Dw&$GHPCq@wRw(7T)Kt@Y&bD8(Y@-X19SY$M
zfl~H~Um<?*TiA|ElB=^p|2mC|m!{h!*xz){9&!lGG2H`FnBIqA$vnuc{))3EPX49e
z_)zrQ|4WIjBtMe&=*QB~b<Y|8aF*bY5mA!CL3Z4~YPwkLk`d@ER7v}y59}849ya4P
ze}`lez6U*-$y$Myl##Z7_fpEEbJf+YPQkd{BehKcWYudUNE)UyLspJY)=xA+8?+eF
z8<k}0dVJ!dHPtWTHDRe-L!%U;Jd6t*dA3nJl8=_G0o%%2>7m`#3rRJ^_j0KQsc{SM
zZynEmEJNzrlmWyjgahc!1Q#0%>LBtAC+4n((&l|#IRhICjvS|tr-3C)+^~~q1fbnh
z+!wEXtwOQwJgc)=yuP_jqM1fciND*|<~|x2LZTBKn(%T++Ml1C^xN}!Fvc~vdsy-n
z3G^J;qKs*J8BMPFd#9`kvX{-6Z#Fg7rPcK1*t)0AHt1T>B?MS0Iw{;x$l(wYY#QhN
z@1Ri7JfX9U8BN{Pv&k{kC<M@ME{>X^fTn`3f`)Y6eU<AkqrB?W!QQM^i<&*-WHRo8
z^WVuavVJ`R<ea;{25LY_{Xy!JyL<k)VcY&wrvu&4yg#O{hIC{C6|CFv2~Es9SaoNS
z_&v*#jZ}xPsI~}>|9=pg78!f!yj&xj_6k1)`Gw=Jy%%|W{4zNQx4=H8*%Ckyx>F92
z*L9Iz@h7QHL%-CEy&6C&wgKpXe+g_aXIpTF=LI2UZ53z#CBe(mB0F7MUi8Ii)Hv0D
z(~1_s<L0X*Lc*lz%u%bTJOEQL-v4B=i5g9T4QqyN=X6jz6{Ug)7-TH1E+zuY_-}Gi
zLm#!ZA?&!M(1uZoSxG#bvZtw-)>A3n%6Df<E2-xLm?}||^MB2!a{jnKJ-xkA_a0FD
z-?7#9=ely#`)L;$4n*Mm)wMh?(gv1$vthnpoIH2DU;Y<%X@5DXeMVhAMjCV!C_g%!
zpxx6A=P~dbTc+%tmB8*&a1-(<*J~j(HQx1+FJ|I%xiF*BrhmO8rsjK@0UPw#m#B)X
zRLjkWRov8ZzumUjJ`Si;LyR1j;56vj58D`e8H1=q9!j!B9A#K4{y3)F1u{#yp^_{{
z@;Mw5i<?At=o82HK9*YXq|l!l61+r1VWTti^lzEE3wRiC=H@9_96R2mUn$MOyYlFK
z$gs2wTZ5{K#qqbkaYy1uZ%EFf0IfOj&WK&7*R+1=jhT~cRqRnYU(X}a+2;zfob_Uq
z7W^ILp91jMMq2SNnwlC=;OgX*a2@P`hk)DL&Pz7V>yDNDd*Z=v^kXF5yBX=)!={j!
z()r1u^^6w@{AzX}^$_CDUc=ot#w`-_9?Zz??RpUd4GPSTHH|h<pWpz21qekY9t+kI
z6b}&-X`J<&Bab;=wHVE#$glLVcM4^pGB+nR;xcS9&Et(d%7jMRoIiW*i}s1g3;x04
z#x5OWKvu}q;8wY)1CVfMlo_ThvNa(*<EhqEYf^FmsUlSxcrjr-D@Wg2I_}`BrrKx9
zx)eW;<wtJV0Wcaq>mKX5^ivaL_yW)yfeS~HCK25{Wrw%Nh)~M)C-D<My7SSWAxqpn
zsA8$Xe@R`x6LvP5yMgyhXv^k{?Uvj3g66k|rANb}sWSr`tWdk&c+15o-TAHc5zR@9
z=wivx&45LOzZjfFal$f*v{RFo@o1x3O5Ln^HqdlsGvUoXaSrzP1t?Q@P@0{Jh(<Eb
z4m$D-PSvPDL}*oASrMv?(m_8i=avhBwS^_|s>TCMi64Dr^6Fe(__XVib}j@#KgYjE
zr%A1L-mb(cR(c&3!x>2?HAxqZn4A;oy*=Bo<Ei@)M@iuMpKrwSzi0z{4~$*-Z%j~I
zT^5}sSN+e$(rl4!CQhLUn*{!Kajtqi9CAa5qU{?i6-yZ`LMzmn;o8UTIm0wRMidJY
zS$4Z-gMjp9^J8OEh`ZK|a?Uu)t?TG`{+z(8>gNmQ?wzrLe4Ac4T4<{%0i(8SB3q25
z+(LCyejIyFlf(eJY943b_C0ESUkOil6Rot^bj0^0$7y8#%DUG7?~yywN(*G^<L`Ue
z{bQHg;q7j%DOQCG)u`Dljsor+fhU~oc)gYnu#-1X!Fk`tlOwf7)jU&^q_F>Ol?nk8
zt8em#bi9t~j5qPAOYKd~NWifC@<%=sPs^an3_GX<fuio9q%TAqHTn_mZ2z)9pOu9*
zWnx>@d`%Z;1joWAk?iV_fv2c%9>%w^+Dnt;tpX2BT9GxvMn@l#0hAqNs8t|jSpyDQ
zaPDi)p3#S@0whY|`i;h$)gLx|@7K3C!Agyv`xLuAditc-E8#*+Bk_`v@d;;MF7z(u
z9^&Tm(O5v(W^_%Isw_+li>TgM8TtGN5@<6{WsIQ=y=?vafu~|j|M_y=%rio5ogFw+
zsix_)yV>>R^KwV|xhnKAQtVE>^&9lZQCHKNMS>ezwE1orN}`a`3Nw{xt9IbKC=Ap`
zH}LYCxJ-R^$YwYBka&lFd54q#9qTfqh2)obO$+F)ITy^lVA<U^xY2Um`-C{%eEI=-
z#TiE##Ksk>aXok*G6+J&*`kK(rt~QL2lih$L&{%+&<Ybu0h9FY0a?B|h~{H7{K51U
zG<55cRE9H53#S2Pr<y-S9Rgj)Min{~BDI1rgB-+?q-~f=QejmTd4S7{;|6U{p#klC
zK-rQbIk?oldwbNhd=3)1M;o{Ut7b0OulyeD@%;J53x7is<FZKs-hm;UfnTE!;MD*U
zEYK=f?>>ma6}Uf*BD%Z%`V58GE}v5N`)QfV)ob1OAd<1T8L90QrPw3#>>JWyM=Z6^
zkA$BMXIS_pCscBzNeitDunq@wOI0bC0S_Eh&Vs2H<0f|OD;%xPL++|>=yuGJ?cdB3
zUDaI0Jf0qyBy@2%p0#4IXEoLq3Y}$Jcb;u&$tZCE2+b(0jwNllj$e%s73c221iE<!
ze-gkzTIaTh=1+*B?}27>NTh44TJ#{xBWkdsC}to<u@r!znHCz$m=$#9g3ln)B;Ihc
zVgOx=RZ~!c6V&#HJ=+oV;c+a1M^&B(gpjFkcwS@YjyP&>v~&bA3L>z$(3{SV12Kzv
zEq*`HM>%9rQ6W=ZAIOdODf=IXKx!&JP2R^y9%2i<;OhVO9RN4aA<3bN%2}XU(H_kA
zUv2`fuk8VN>#-(Sl)r3Ijl{=?ZT(=I8Hh|(F^_(%T$6QiCySc=7D{y9z3Oq3Bh3L8
z<BEX;3wx{$AyYg2tfPOE4F}#-WesgmJKd&9e));F`r$})@O)|U>=Boy^~EF1R%?>V
zvZWpXuM@-cmi0Idb~E3xT{ecdl`nwXLxk4?P@OzeJY~1DtT>`O_1jI^<^x)o*O_<#
zt=Ia5!DhT$x;_yy!zjoK@n7Jui%cnyXBRo$@bnrb>6Gw#+utF4ak5i&sYN+zY-T=S
z$cR<q;@))hUKB-<M@}?0;{x-WikxaVRK9xQpAcqwws23gvP9KNwJ!>ZI_k(8Fdqlh
zLdW0&j_z6txMJp~CbAOo2vp4f^nw%nzmR{tKBuRAZ&PJS7bZZoGDqgnKZF*TUs90Z
ztp_=SK(o0>n$wrw)}-LRvpzo~v?HE|b;_l&@lr1yJ}_~aPm2B^sm;m&@J|a9@LeUr
zfXyftRg--eov~KXne`v2?Rvw0N3R;yb}J4<)on|JXqk(?(%oc!*L|EharAp(*MLZC
z=WvBr`bEOr^!axeIpaFc-JJb)z;{=^7gW?FZ%J}7@oZTbLI&^lwp8X?qG6jPaR>)r
z?yl1O_FHPmAPNe(Y!TD}>-q(y_^UIt)LdGmwdVTPqXC^F`2?g~yky%e<N6Z}M~QKA
zd$1|v%z-(FiHjyk%{`jsX74r2M?=#mM=jQw^MXm=uybfQd^3HB`mMtB)!4oOp{8`S
zn1@*&t0G;t!Qs;iKEC*uu1-O@nnU$MQ+e-PNKBeg_q(&?$X9HcHTNl$*y)m->_6b1
z=^Bplwa3qsUA9h_0tL=BoV>Z$?Z<1+%3s^_?0&cu^!jhIeP=3<VtlrL;U071q$tTE
zn9EF*WG@-!V02Fxz&omX>g20D@v>F{<aM=#LR8*ZP57fqGa6knceOd2S*#0x?V9k2
zZlvYH!G0u}ctIIW^SUU?5xXab<muo0P4?s@6eCg5(j}Rw#B?S-X2qrwodwZasHD0l
zW?~^#-Y{y5)%g)h*cMeMb-J%cP6IH|?EuVFt0#B8YTxa)(TLKb$6p0zgVlgD>xE}%
zN=KXZ51vn*b#v^&;tvTG)e_Ib!B_vs0Ka%Qor)2Q4h;@|A*dcc!1t&k9i7D#>&sx~
zwHOwbp+SDiQaUkGL|(VJTiVNiK3EkbbD~bX{;qdd=3$2WyU68Mi@4gm&^gEM*E_-^
zU`Y*CqNoI^tsIlnGI3TUgoo;A=R~I@K;!E#cXUe`->=Q;kMnUSl1FRP_+I}gN|I!$
zj=N*KHn)#dpq0u6cELO@l-93<7#Zf#yz-=El;{u(Xe6+XTuM>OjhLWvRSGY0y|AX(
zU3_gdzO<#*k~JT_tKb3wk&*YFrQt~8t>2f2H<m8d+Igb+b^fw#_HWiJyzl8S1{c^S
zf_{FI8j3O2>_MDc+(Xh)w**h$PbGCt`TSdh`@hc`JS(Ir>Cjq*nL3}?cP2b*I?<Xz
zs>q79!si^P$mn+}NnS_WJ)76m30^@b%!XkzuksEC4F1u{@dtyE2bTTZNy01=zCwI$
zWn+SL{!)gMS;#Xg%XXz-jfrTza}Qc@^Rn~z|KCg7HkS;35SUR4t~tM`GehnoJV5sh
z^r7qac!?5vSNve%F$ZTJx}L5(dV7x_Jw|}6o;3{Cy91j+Soa@dd^(paRu!_B0j+L~
zCP;s0sTI10S8mWAE(MB6y5Ouu7~cHMz|I~WgQ@-GlW7$-qZk1++1oW=LMYegmo%y=
z2R52o{Oe{Jiv5VJWiT#F4LgVsf*!h*+mYjX^#_F+WqG`i7O}SGYX)@Tq606YbVuJ0
zIs;3@5}ckiOv;A2Kj+G@O0cxw&{+r)ETC$efF)@>cn;N_QmY|*NY`c2D77f!A;qF_
z-|#Z%F7^ha2Qa%(W=uIkZR5o0n&i5DAF{Hm=w$$Mw6}K-0|at2xn+b+QaDp!KCb_c
z;AXN|F^!2Df<CM$V=5isVKmj=ad%bVM}&qCZHaNCuY&CsF3DpS!abu?`)Wfxo3pgV
z7eH#2*zVbDSl-9pGEs}l^E=nJgSYz#=rFDY<E)|asDJa_=$q@hVo&-msd`*XAKHL~
zX&=~Z04EN;*JWrJ`@KyKK_r4UF@u2V%#qhFC|5#h*lw~h%3Eg4%;g%2Fbh?d7=ON0
z`eEyYQ!r7nXut9-4<=;XFU!N%cc`bcX!VwZffqIUPx$~F=6GjPdQ6f9W);w1l2R{X
z@?I`t3?A%Rc&4z*;2rTPB$;$9tPqMonTH~zKkCN8I27!Kr(B_Kf^UMXof8-DLX~Yn
z6E(;})Bo)`0Tw8h77MF`z%(I!5+zc(dd-rB<vP9l{k*rVw(w=J7HO=<lZv^TX{2XW
za`rv@KcbrgUIh3@mlO8zthqpj+4}fbW(Enk>DvB```$+U{O+X42=$AuLsg+{kvWy%
zULowAU|~gDkq<3zy|(_W_4C)0E?!vvY%Gru9-_icHV|l|(h@ALoRv{t>|)Q~F@`X}
zw!YO0TTN}zf%g$Jg7UH3EO#<o_-OoU=^WH9w<`e$8@SA*e+wSVJkUod8C5*#yd-{5
zwqHW}$nt`qNB*gppgEIOr>;5GN+h6oN4T@sOUI3bnpO}fW^!mmX-HKSh8XBU4-v@R
zCQ;Ml#HH-7c2uj-1pM5<@=>EthD1hN(ZA%mph|+BIL$r33EH&Vo2cs%2LEDi`QVbI
zN%5ohVE9GnXJ@6WWtqR2v@(J<IcZRlUn^!)c9`c_B?p8^fLoWB)qgu2R;If>JSChw
zK6N2gJ@4rndOP>3h(AMez%X6Ur(Ht#5pTPA$q`d~)aSB!Bj{RJddt}%C2ZfWi)Ebf
z=3fQoE}tIjJvA+8Mr?)8+F6dHq0hq4b}l*JELnAEZliuK7Tw7L%vThr?u?awrvG+P
zpr7(M&vW<G_R`2^t4UQg@bF0vw?~Glbz`H&jjdIUT8W#-*>EHP@h?=W(?pC%IZcKd
z|LxYtE=TX5R89pY0i=wi=ch}^FrrKpsv1pbwzz0lv1hMV5uMfxmVV9IsF*@|ge(+a
z8AvE2&&PSDXB<b0ArY8}mi#>X28<MG&-yNobTDs4X(FD`IcpjdIq;DMc9of3DN<jh
zp$|>xVkNen+ScZEBqDMJSt49-vj<DPjDkUSE&sKt0H0>aR5?&~=Er1)k2s1dFCN)6
zmc8|QI}CHp-0mmCzqjSYfxpSyX)I@Hys^65+kYBl4{mo^zUE=1UgFPJMwU`H?##|z
z>t+|*CUSAx|06&*KG59d`guY1aNSm3i9<$Tf<HGLi9=q`+aq&s?;L>Wl{3~dAjW94
zF;e6e^LmyjD<80QDKk$PB0scLWA9O!7PJ7!`|YWqD|ItBG&fI_b;GiPoi$V&+`vTu
ze}4gaf0uT~ED%&Fag7mrlBFPvy$(r<!eOSfLMIaW#G_swhnFjA!iCEW>Va0~fY4Gy
zNR6BAi6{NpsK%HSRi@9r-xr22-aymTlA^cN(Q(ez9dr)_$YUk9Qn_&KG#S^|GJQRI
z`P<<%PM--*32L_UUUYP>IlFQ=gFDhDva!3p&BIQfCI~w)mXhISi3JHfpFmG<ALs5l
zX`vf;iCJW8Co@89z?m9eFfQqtn{YnlFOioWd;8`B8P)>17Z@|sqsO?@!a)c!N{kqK
zs@^V25BeSF&(Aum-Mfo^x<Zcm1sOzm&yH%ub~bmqDzmbmfgjAwC@HIUo_2S@WJ%cs
zbsBI)x}s(H=UQO^=|GoWqNGgtY20pGX2k4Nnof?4p;q%^e$X(hokxe%S`v$XOt*v#
z2q)s<3AUlV8~*w2n1jl)`}<c1=smDa-)F3cLWc3cNN92T#db$Xhc|4*vg1JJC+Q_*
zWa7J_nR`Y?(%s{tcF*gcY|qF0^2XNY?bKL~%%P^{^5o`wq`lu~;5HaPTQ)Cq+7Y_;
z$z!hDy2Y6u%+h(Q#nbCDpOKCB!jd})0pt`u=-ec6;kBEc0w4DLSE_!!2R1WcDeZkp
z*uPy^Fm*6bs>Ec?Frx7KilIU`^=GY+R2;}|vT}gYV3AXvIP~&MgeLwpZErD%UIb$>
zC)m1`o$^6!nvA;=iU!^-Y+MXj`j8hUVthP-Q9asECRw&TAIBrFbM1EF9yyQ`AySc*
z_;Zrv7aDv@OM;OEnt*Qg7z#@k>bryQPzu%0!*<4vJO!*(`u%^O4wkFYd`8x4T4|F&
z+a9w(jBlX#Nkbu)sX(S4kE!!zdymgufu|AgE#m7*xCU;*Z9>nd2XDX6F82*W?6gJ1
z1mC!nSQw<86ZxYi+pT@CX}&7+5aVfs=Mxs*&V(AKQ_XNFzT>fhz#_dG`}510vfbnc
za#p`<P)WSfYe(pGoQ3!GFkfq%=UTN`I7yo4EQykpqnKNag_;%MKzXd??&B)1%XD|Z
zcQ*y0Z6q`Rr6d6pHZV+f^=OblFAxh;dyy=wFkBgK@89f+2j9GnDn5~J)kpjKr3Vjj
zr`*p8E9mOYILjzGXA}Kzd_M@cDN%k5-JE_)Tf#1^Gjz{|(*V)?y)DMCULrGCAEjI#
zR~LHELqHRuIFyAx&O2)+2XgVZ5&QhtYhSad-UrQ=P^a~B<N^cCtdOI<IhsiC_5&pm
zzdvINeS*<Zi%fn)PW*|>lgijp>R0KLx9v9%*IT!BpNqY_`fvCCL;la_+jk%DHN-pL
zCB9qcaUDfrvRG;Y<6BGXWzbQkd*^l&Fu-vM1EMHeN*VRIfqB#RI^253P}8?AI=#bD
zR*sAJbj$@t5PUXg&^Ef9rur1VzxrhItex7({j8Nt<&Wmtiy#biyCO}<k<vd^!8Q1j
zfoY=Ui=l`pBekf+Dwfa;z%`SAglganP_wC|s$4QfTIp2yw1s}=jAUm|ku_LNMH51_
zG7T<8H-V=(GaUpW<aM9spab`$-eFTiPfC>}Uh4*<ES+<uNZ3YGY9XB6$Lv@rUJNuX
zbJYS5%0~4{mkOdE$C!Bd%qOdH7eW0A$nKs~#E$eWB_WVYY8|~y8;sQK8hZNE1a53h
z`n?B7M_Jj~gF>%O`gi&t$)8D}F7GtcT|a>&>!TBww<#{-=X?L#x|h6<aQ`v=kB8;%
zPlq<#I>xQd)0G}*e)uxDhWHjck0cS1%^1ye7sN%ff4TYlt$r@N(6}=;^pdhir?wTU
z#v+0C=w&HLi(9X+r!LhgIuWEduR{pnY?^dHabTvw&W2XS6p2P#G0H$m!$~rbp5=w;
z@~_-Ac0(B;^ye&4Qz5fNbs%~<AZgS~w2F4BHH|3sbZt1<g}|oPP5couWs$7H4aN=`
zrpM<Y%7<V1Dt-_$HAWRy>3K{;h-}ydwF%i4jRWQVk}Qx`zT4+EO8;{&y}I`$?6B=x
zb7?Webg1^+lEBu^-@^qle%^`4j9p`|3PJ;viLI~0GL0^x_;y<7-!JrD_Q_{wFXX^a
z)FH6hls0+mV#HI^?e9O*`-=UU<^Qq$sra@C_B#%QW2HJ56wm#W(U8h1?fsQja&NM7
z;!t!TZ`xgah(S(ErNHoBdCRdaIS~bfac(O7hwM`<AUZK;;VFD{>XskdtKLFmDw1JL
z0##0`Vnt%j{PUli6iN~{-ON(kmwCG|nvFbG>l%}kIUc$r2D+pEsJ_bl&EwF$3XkDN
z2y0sDz;^{59y`figxrf8h_8sq{g7m}6pona#G0_AH$zjhkMn~MAAjsvH%c++L!p8s
zM5ekTU1<rfv5vS=&7(WfLQxQjt%a!V?ChS;r1YKI*TmGewl{AdKu-ri-Y`NT2jG_Q
zY3ROk`Gd5UI!$UDQ(Yv8kc+8q<<d-4=SA`30{d<L^HS&#JU-@62f3cq8|<{O_s4oa
z|Kxqj?vH)$GWAmRc3r#WOFFTRuQF=4U1!^Lk=p6tHJl7C7iAbUA#>!CwhvUG1%c06
zmtWO>D}335mf1vPUR2qi_9z8H?j<)qEL{t3xaGQvx#0dyyw$N{-!soy;lT)&2yMlg
zbbA8&_*jzF(VusMO|?J-|J$zK-RyPmr7X?pyWQr5VsL?#mWXl?fAJIfAa_xxVXQjl
zfHk1s&JJ0NkEcpqe6qfIk<K-wmr7${yeb>}?cfl<kZ=4d-CLO*B28p~h7V~H4YiJ)
zHyl^Z5IHkMO-D|JV$h^r*J`+RnQ8kb@3V;BZJE#RI=)+o(6cv(<}RXXPTPPUjqQs0
zR;k%r5S!P}&cEw_X@)NS$<9I%#9k|a8G({1e0`Mz-T3u(@3H@L-Um<JR?n_^?}r2o
z)HKI@eW+9I#ulLa{Y+M`*LO+?<p0*Tv9+Z|`Te54_C)>1sc4*KZdh7j@rQ`{m+{P^
zQPGbPI&n-0QFX+|&D^vx2341+qgySkW~<|PQttXDNV+tu;8te0KeDOWUE<lvZu5ke
z+}P@r6GUfoe=;-}A9Iy{wh?hMMf?*Erh1fOsaNR}o;LqB+iSnPv`FXv1EUkPrLP<f
z2Nx0m{T*bB7Z`3(CFPglPD7SknAoK8T?M_8l*}*)r#u81g$Z5yXS3#+*sRft3{8i+
zg6Uy`SrB9?PR3DW-rF;^&h?|tWu4~o_iC@^C^Hp`%EPnp%8<ek<;bf`^^m>cfp@=Y
zp*JwNcxZk3!1S=4VWS3NBuT3Ug#!&%oa=-%5QtwrZ?Vw0Kz}MKq|5rssdf@yLa3O%
z*MiYLg*z6d8c~rzdG9sm;4k%_-Md5le*BrC|5oG=_J*~ACFgOtpE9x4CEz4k_sgEF
z9=Mdo|9<*IyXVD0D!TfFWDGpp1Vm{%{Z{sH$7yQWJr$6;x!aF#t8s$oSvD8YMrr6V
ze_eKUGJu^>kjt_U-8%e?`I!N|63mtRiZ*EPu%YWQ+%?+;P{0R4uCJFlaS$S1FP17h
z+A`v{{mw-aUjM=>NlV(yVFsg!X+tnzDf4G<I2o*plB^31#=2V6Woaw13|r1*7!^kF
zP%>#u8!%BJGUNA-F1{IJi9${#ra6TmTB?3iu0edWgTHX&`#8Y%?`PDakQ#=xiVQ1Z
zr@X&YK&!c10(ynAiOkTUK+r0HOXAY9tE^30n%j!h6~Lj=1j3*0{Wo{T(D8I^ij8`6
z4L&{p8HbIvk`t#XV`SZ9P=&jOm-?GG(f<GPZoshSxDj$-oJGxn)#?05$<6B}Hdug2
z@_(HE(C_yCyzcsCUVPXpk88nW5El*(EtGM&;yH^cA_fLX(AZK{pH$>#R$J}t0(k-S
zsa|OA-7{y{I83!IQVW%3@L^|!byoa^R!rwxbWC=1bCLzeuq;?#XFWd4ZCuap*~!Fb
zXi}=H%Q0(VOEaS(_nIMU(n`-!C(o_uZZOsvRMU)<RNW^cDX|4Nn5reTWK1i91~bW?
zaz-(det8ZZBvywO%G9<plTyGRRV%d|knZn)t<0Hs8p?;Pl!(mwrt1lM1Sb~ScbWV_
zTp>8mhfIEkrKccoH>DK`?HJ{hPT6v~NwS~naXI{C;FS76>e0;f#xfVv3-#yDw8#A!
zpzKSV>8_ZdQ)Q;8X_|%QJYJamGc15lG0ZW~m*}!NBUnOdwEz7FU`UyiKtnx5!<y5q
z1GWo$`R#)EgSL%W5WI`}q+VA5Y3L4gI=b{d-iPq@dc21G-}OI*US6~Pm*u@%R*mtI
zz7Lo0Aj%9`!q;7myUFiohLK+-a>sUP#K5(rNAM6eR?f0T$wmIe<k4!eLO)t0tJlp{
zZI-&UZ?{D>Z|3>Izmk4=KqODt`AgC7DBn+=ql|JL!1@M*MVb*U!7qb|k5x?^(e@+-
z2BoCXqk1@c3<*x9D9dfrhIZ{of?YrF?~(hKyQ-{1^Jb~v4ozq`eou>59nc1q1lO*L
z{0prfGR>P|<l-riNOk)j<X_Vcq5HF1Giqs^R3U^!r)2CpI6z7iR0@lhh@v!t{!3YO
z@L4)RIwsPGRSGB<Ng<;w;=Jop>C@hI_`V5~PP^dkV;O~)%mMDbKF!S|RRm8^kN>?5
z373y{F&fns)FKz>21`v+JJSjeo`|2PuKfo5A9`Q(@00aEy1j*>A?)<0_~-&lAoY4Y
zhoaM$!H>e*<G1P0?a!g#On*OBIT!jHLY?<9f9YqDnTBr;uLA$2denL_@m&jtg2TmI
zuo9_i2)x;|g&R@$<zTk=Mpl&C@#)-`0{psWnv@!U_*$HJIdvGfj^APgJ1x7x<x>70
zw#jTmZN7CLYJ&!%QsNoFs37+9z@JTI1+Pun&3pRZm~5^EUC<9?=A64}x%T@=%^TiO
zH3t`R>R*w{#Us!m<M-#_Tii^gnu6QUL>&jjX2+G;#^q81#8<A({odw@8xYXD2+G}`
z2wL)$NVIqhR4G*vT9HdS!>2LMzmfF&we4&4K4Ti@j)EHzI~~%o)3US{={Pfgnh-fx
z^q*N{Q)n(kdW^Pm(hrA8-0`%`cf#>i(Tx;}bMVGh{0w%`VauEhe|Tbs6MBc~J$UCL
z=HAL%DN98OLHSM-nEEGLgQ=sZW3LxP{F(D=|FM~=RB5+FYoOw@Z!l{aRf}sd0<Xfm
z;Y{=}FHp(y3RMdkforP{FpI!yh!W}yZH$*)lJVveov|W>izwN~;>5F*Q3CgSik%3p
z`7{BiPjEVOwj4PprKx@kx&DjlGFYQi#GO<vVtK{m+_(DKMKvfa9f&7C+vYqT^A)P4
zS$Oj2;&6&JCxp7=P+p|Ykiba+e!W%bnqYj;QQE74Gckn=d73JUnj5Lp@AdnpyiW|J
z!+*QMCF&;hUzCEfwDcI%Bz0laXf-9M4<`d2UKjD*Kmv`xwfzcptOnxqrr^P<l1_8J
zq|@Bfz&I$0pvUG$nC$)dZm{<Zi<YLiBwvI(HONdQ?2)PN<uvaj>fHUhU1-ZOVaP+A
z`=`VxWUYSBj)Uuv|EJIg{>N1B`_mC*B6tgCsZp;c{lRw1>z*7Kwhn**aE&O#fWE76
ziK?kJsyE^{oM8wPlufUClXnI|5T2(QnTIJRlV=T3rvP;6&zZk?m04CRWi8ceUk?mz
zoj@~y**>sZa)xyx4E=4Ljl|RunZQOhELFoA&DXp#LiHL3@C#q7U00pU;Fl(SH7pQ4
zB$KS(LSzvvPumjFb;L8FQ7HY3Kx`Koj@KSsl%5Cxo)tX5{8aUPKcW-@ogrKa`j8Bc
zPWykbjW^zK`M(gakY4<@vQ#P=hIBN;a?jY6jIUJJq=by;X6-mUe31Q^PybAS%u(&2
zr*AutRMV)$;V;<>l~}-d1Y7l<lDjW|PZt5Kr-uqvO;5iLh;p03J2*k4)sFl4g8M<=
zpxUm59C$a`LWdrIuBZgI&~er2IrutHeS&&H&S$+J?&>Zg(07Z`#H|d0Qx(?M>{Mt^
zd6J;G<rPG6^FdtUp}(aum*0ORmXkQKmz5tTidg(TCq-_;2Gn=pkYDh_HPpH1{b6%P
zdDlKw{3CLY|G-5aDWrXB@93^{+%)A_OSbVVt{ZfO&kdl!bQ9_^gO60MoHeKbZ}Ht-
zUHcyvsI_HeW+o}pi=s7*pJz)WXG<ce^u<m<EE%;B;abn=9Q?RKy`6xqK7g8Xfd-v=
zC~!p)<|qM{R;i`_0BSadyS22$2Pi@>o`55*(unaFh#GEY0oBou5s878;7nSD$GQGa
zSkio(c`=6|KI`Fh-H}&B9=!Uv30Bws8v-NJJ^@;)A$Z$aF13j2rfz?H<l<l;(Sz@=
z5F{Gq;!#@=qKI`4siArAcN#jny0@>N&;A93DiA`H9EO<lDwLX5x!;hSRr%+B!-nkf
zto5a2G${XM<Hy9t(5Q>jv@)v<m8c3mU5*l`_b1%vT!IDQR7l~L3y+z7iFQ1<aoFEF
z^x!;Mj{TJ8o6-J<^}ry|a7*kuY}IkEUdR;|u`rkFc^>YYYt#%I4r@w#0FQ=9Oi87p
z3eGg;A9Om5(?6ADFh@!Fsb=%vyI7Z|7$nr%6;-x09?8|<)@1ICgY9*qrK8>JTH_KS
zI7uaUV~}JA^-1EHrbn-5f1=N=dt$t6#e!5j{2Zr*o)_!J+!6fW?LQ$F=c_0pY9xka
zJ*NYHPDCvU&zJ;USUr=AKt3)neDSQ=$cuaX@R&%HgVPLcE>u?(gr@LPjfOX-X~o5c
zke6?Cee=Q4vySUFnlp-LR1WQeass?9>h~JAg`Y2{wp2?83_Nh$Uq#_4)E;c-8$F8)
zfe!0<2x>h+{?EO0`MFs?esGuV@35=Bcb2k8W&5$W_lau52N~}~{G)+I6&dP&UhUIu
zQGJA$2x(DI(;=ZWN?_wfwu`wG{=-d(9UX$>y0D|9DW5*&3k{AzmcFz`{HGGvV<BSR
zv1kzmFw}xI5M9nlvuqXS28Rx;aZ<X&X#~ob$ITiuSch7$vs|0ALkb?0zG-NX$W?H7
z=Xe!I(^qu-tdtOGO*(ajilS{bV8><_Q!d&cycj+Xvf|4oqowHad5>xXK{yIIyE+eH
z+ix{)5O;iLZ<9kQOGk)nT%bp&k*Y$?3J-qKT?CM^08tZpE~Ui}RL883G+Cpu)Z^U{
z;DL>VD~^SMyzbS10^`n6yqDLGC8oYxc5n+2VoO->ZhwZ?b#%sczsyc5W3hvKNTKdO
zvdY`OsQMaa$QIn)>wjboTxEUUf1ZABRT22X1K4JqxHbeuCXklpfb30KNljj+E6S;A
zA$U{FjR$fWNtmV)M;2#&_;xFVU!$X!>~c=WJF;fGUlJ#KDOV&}>{W`+{wzHKgvGs7
zUBDkriA|$~jQBeB_%?H*ZNpS&EzvZq2@h?-KMf!p%k8SXd$<Jjf{qA#d-b_%LqW!r
z8kzyVokwPCU8J2_8w(>f3@;`!jiQY7CAE+(3G;<+K@?MFZ*aFUX{bMSQ4RWjnUaAe
z(W&lZspYx|mRmFT0pE_pHfq@>z4}mCrJs@IkP4<H`EpIWD#{FdtkU8#P%Y()k-`gu
zEKD1<2-k5>KAKkmR=NBmn{y1(awTUstDa(0wfGIk^TlJ)iL>EQ<w?zjD7eBSZgs#L
zsP|cbi^<IIA2pLEM{<Rn2F#q{cdzivBmC&Nr2c9D2m0DNy_>_&k55UN_S-+Fnv*LV
zdh-_Zh1j<pD(<c9AIq>SR7;36vD8v!9BcvNF<f;Ef8EoRwPdLj;6y1DzBV>`W!3a4
za|2*z42I(Xcq5i7m(*S4K+Ce7w3~L^qK$+pEH$3HgDHe1?0%1)iIr$GnJJU-s+DWE
z1>nn+z25-WcK7}}=!qbz9wBo6R9aM1dCvv-?BG7Vd@Xnb78JTWJ;4hdo0?f#%T4wZ
zv8F=mye?@K4?-|SAwCG4DVVR>T~*5)!ip>njKcjIj5VR2xN|4pVzu*+!uVcaiVbY7
zxb7flCEj8Vk9^Ya4JR4&y$IDTZnaR6Rha|=3MZAG)N1A`#@2_PTla&0MC3FpJ0H5Z
zPj7DOE^Ar+H7js9b|U$=hjJ-Ql%$jwtM32(zsw(dTlP1up+cV8MX~Vf_)HwkOvEZu
z$?pF35Y1Tq5CK5U+0i2X5zY$n<5ccm>o?L-6tvpxuhBCpe|Fy)2V?y*P$5h~iy8HF
zDYY~0Z5a1{OO?>v<43L)R^goww>k_~r|?In4u+co#FyiKIQp<^#IBDMhnrKXWDqnj
zLl1v*#Om{=nhurBJKq@z8CpFhD?LNAc4azGOA(l{1oZ$iWElbTcjq;%Sz4Fkix_wE
zr6+gs2E$yCQFzEAh>4zA9c#qxaWHy=kPQyMg>LR%q1+E4JhgsjnIIf)iFT^gN7Pn^
zhJ7u)l8=hBR@)aJ7VTQrY1hXd`D1myH_K&+9wdsrf0o&B=xoP(B}=S=e+#?_{_aj4
z)$h}!_93f6R*IUrv-4$pd}<<Hjugx|L;tq$b>z2%aO+d3x=q>dF1zMAfgJXtw0A{|
zig1PGkX%t^0m_jk)!(WuW#yFBGG@$*zA9c_`|)m8(og(xlxKz~_z(b}lsK50)!vJW
zz=0vw8J2;!@@DU~o!@jBc)FZQ-MfwtRysX#X=DU6)>O-9iKu`QS)Vwa!$qWjzu6t0
zmzIfi&ak5VdY*kOAoe#Ay#+{>L#o~bHJJ2Jjw1&a1K;_Jz@+0;)UJuYSX>FAM--Ou
z5m~5OB$X&|X+UY=sFKy{q58ne{Wmz_8Vd`h8lr|L3_|PAZS!TKBGdX7U4;Bp_ckfI
z$n%)8Ta7*N;RLdg9-uHm>U@8O`2!4QzLuOg8;^h11K|L;jO~jcg?xX(oE`;#@0;Ic
zco5*VuM&-4vU3t!{~X?)ta3FYA@;b#?KqoZ8UAqmz`cwPOk0P2lRIOTN|qu3gTG|8
zoF#wH*`vuq#2Mc8)47G2s>Mx=g9X|xLhQJX`W%MT^YW;(II#smMSyEZ?56S<B43%6
zWXof@vaCRB#QIrB6FH#m!HOej-9#Ktj?|Ik@&}7n$wUygc*AH`0+mp05Fhqr28Jj*
z{K8i*)y>(HaL0u|zJaCy2Fj+CjaV;27&tC6D~(8SoyxByrnp_9BeTNH5(6wDyLB$>
zue0Nzb#%1U7J|uH5}6D%l=3XyD}EHEqW$!`jQzpI#ax`z#U6Ztu)x$dmyfl-30?~H
zzZAMQT4*8Um#}UZwp#zL7nS31`__iyP}3`*1eY|L*ExDiX83}O=UH)H$$l;Q%Eg9F
zuU=&eGH=CmV-q537U37&gTeskD{(A%kDHg50QV8{r-|{wLC9)rv)I+5y6Qhux6Kaj
z{I_Qh8ajlviS5UJTAu#E8P~%T?re!v9Q-=2OseC>xer34fkr~SoesFANB`bBUuyZ%
z)n{VCUJmTLtVq&CQL;Q%v{3dk!Gb~$)%Gb-zGUwscw~{E*2|to_3Tzief8E=Qbsk_
zLW<VLXtBh6g!Tsa13=f|+YZ_F>C3L>|D8Pbtj*!I$Ky3B`csP6b$-CBprjQNoPRTf
zQK2{%C`=5VEQglCEUl>bPxZVt0>dUIyPoJd=#@e-LI-Loq2SLbU8p!5906&2Y0(bV
zGH#PNRW3yWNhFX1CTBJb?Irq65dQ}NLI1u(+Bo7IDQo3)I<cQ7rWDyvk+s6HT69JV
zO^MqMRs&9Kg4xBSHJqz6G!PToeA))(RnKm#O>uXHpb6=NHvec-@v48(f07XQGgA$Q
zJ>}l9(7fP5(V(P#q9);m&Mdj`&HE4JS~*f#2xKCQr*ve6Py-1w8m;+6)=Wd$*~AXZ
ztIxW!mW8m`*eA?$ldA~`0SH-<Rr%t@4PRYdv*sfyO?>#>@40>VfzyXusJRPFA(#<O
z1Fl6`JVn+@s2R=1DzDOxin83FK-H?!Err{=BWtPM>u6qHOP?%^_J(Fz#c@G4PyM)}
zpG$j;j@7G57B_C*R>N@#_nG_k#45McxTi8vYeb0bWJ1+&BBr^pe_gq{j_meuI+_9{
z3NZ;+yA7<fM%$m!1()h{T*?qZGNulz33Nihf*j>St}zk08Gj_gS_;S0iL2d?DMa?u
z#HzyGDT6rmi7*<0gP#l13{H$jj)pn$9qCvGg+WJw-uvqB9}RdR1PjrR5dvf_tZGq$
zbD9VeDNDn`>QCc-IdNPvAC4zZYv#A#+_P3CYv7~;X-7S-tXW~Qmz7#g*`-pnu+~he
z3snmNi-M9W5CU2wnT1t1IzC`&fYA+~LL`^M?fpI1yFEX8@rL8^NH{6?zkAQ!hg(h`
z?g?6mC?&5JB@(kmSBPP3)HPf5T(opTRbk#A&=|Oy_huMaU_BP@?iX^crYws^TWYAb
zrx}!<N9cImV{m5I{4CmIw2U3`9jpAVMCz)<+Yfv=eIVEOWOfvyVbVzC5S*4*VVd?%
z%jtWk<;{2n4c`r5P8LBNB(Ubf?dime-JU5%=G}xK+^rcA16GD}T5e66*M!mdyu~}g
z<~Q>@dOyLj*P2mqpykJGBmEp`hB9rcnnics-{12$fAcpi_ea*_iPwiK4%3dWuHKMB
zV5Q+e7QT6R%kRE^Pd)5O^A0T*xwS3}_jTc2zT>zqUhu(qxr)UZN-zW2k~2k*R8B0n
zg&HGam%PY{kW1mDcW1O50+e9UXc<q2R;4VJdSAGaJ$0%4fByS_iE1U4L|#uU>%!qG
zG9`#HkW!?~lVgCceb(B%B=7GO4WlmS6)i%pfu&m1XiT%MU$(UN%9h&JF~=Ba+1IkK
z(`RD50!y{*4j5>x<&2Jxnh06>?fb8}J-r7O7L+0$5l`eG6DX7#nPz35l&kB+{s8y)
zX2cSOl!U`Bbo!`nGz;#4Jfq{Hl+0iSYA21jE!X94KZuM*JhoAVgkchX$eFv-JulPF
zG6d$?q6m+t%@8je|5dd}<KRx4PRBk^?WfUW=t$xEfSw=Y$(ymiUAg@?Ho?vq!J^Zb
zC39a^<4K2WUf#UsD$RVjzo+EF7q4DZ6vMx|Kk?!2#ETcN`QnQ&sTg+F*LSy20_&Z!
zu9;duqE{i+)eO;fC6~-nECNfTl8>v0VP2c8fTAG~TcvFRHsc=gO+};<Nkjs3Ow7~7
zJk3m!93I4I3$17=Od*=1AVwqxbjb#ZqQS9@E4zld6{FK4-Ek(h8n&hzEJL!Kg;E{c
zt3#=bKlF7P69_*ix%#>N?c1{f>$!bzt>H6{b{gsI3KLFSrNRR1S~wn8;#>%#R7nVl
z+7GR*I@~ndkI34cFCjom2F7j11;=7)lBk&R$%K_59jN{9k<;(Y%c3+tURNHgdrvHN
z$I=2yun_K1-|5fm?{#6ScYw#G+>17Lu@B2VA9~UM0h{sZr#0hRi@p+P-v2-L{_RPU
zB}wnZergwYkBH35syf}LXL>L@zyQP&Ab^FeED3=K|7X7NkrYyPg#wC@T*$5nLCg#W
zgX#OZR#j$P+}$q92Q~AEtW&3f<?xJj^^|qzsjAG#2zN7AGu2N&{j{andQziz=`v@*
z0%0|QuT8rv*QP6i0esprY@0@bw$3oMw2o~B?J4;8y3s&IyO<ClSLkG$hgvFWO+<0*
z<G|B8qp7f77V?s=b#Bjov)_4}v$jZE=NQGZLmR86Wq+z}wfViQsL!u`>GSt(4%O|N
zY>rL`x_;~0{O*ho+wslNrax1Zl&srmDV5@t>J(LvsHgX+wDK~u;}AW=;8>T&#1kvK
zxwT{Gui>Mr>A#fP*Xt@s%FT7Q%WBNQo?*1_I<BQd8(gfF)S3ziVj#1-YCdhjEU=Xy
z-8@_G+wuE=<@@7%?YM!-a`UtAyidEa4rKNB>(@|9p;S{DdFK1|t8eSIq{L~S8-P_f
zFQ%JQ#F1-2By!!c`|m2f#dc?4YAdaNcBNe{8=l*-v{J1bYc%p!#fxX~ktJWKHM5*%
z)_I|3t37wEDMV*im~FLZqw2M{MSIg8q;EPzwVgLrpRK1|V{d@bp%&TLYTvbE=<HH%
z*5G3p-*3yZt!pV}%j;T6*_d|ORf<@H4Z&OX9zn4(kRXBhw_wZ9o-v^by31C&L<@X!
zx$49??<}yo4M1-|Dw>U7P2}3V<$e2;XWx|f{>=xTtLweocpEoofNp0^|7@C#<(#=(
zW-gZt^D+}$fGVs*0w0LpQEFvb6Kl%koY@Z}VHjDd%xmVfnuv2Lg<8}~9Nlfu-&K36
z>D)Ccjh1$%wNf2bhf=<4%V4F&?W<q=+B(j9vBJrx#}g`wk4?C}(cwA^ZQIKfV-{6K
zsuEN;k?3vWda|+GCf#nOJ1bm$%-b@f=eBG<+?DzB-qZCvJ?ae!r9Z>c0N%c5m3p<~
zoU7S&=uG+6>_P$Q30035M~IFX9rJu+$4-dRzzfA$*+04^?w;9k-(;@zo<7;%xsSe&
zOSj>+Z#%)7MdZ>%z-y}#a2s>;*<;b(ZS&*Wo9pcdeP=vBhrV-NxBtF5L%(mk-%678
zdLC~T2(=Q!z~Be=yFF7FSl87`h+UI5D796Hc+;RPRxp=qZe0T*29Gap^Z9Qt#@Nu+
zKCiQ{+M2XY&j9wWjoEJOw({c&5!w0;F)$5-CCLz+^S18ADi^lBtAWB@Am02gV@uU0
zGf=CxB;5P{Q}cVc;m&Pzxa&QBzisOGFUqI!dp$k@Sf9oZHhL1qNy}Z<mh4sCDojX9
zb&wPu->v-SS4Up>iQO<j*drP(Ym7tufar=-&om9(-Hp6^w~})Kse}-irlA2QN@xdj
z@4Ms_3sO^N>747^og1!{_1n=-fT*rfgxMUclrpC!G1kfuJi|DkRj4(!vW32x3#GPh
zBPm(t-}x5IUBQqKva-H2n)7}0(q2Zi1$1(gUA&2gJD7AkTw`ze5KtiHOj5<G2dOyc
zh;eMrcGJ6RS~*KCoaW5DW}Yq=syYhMYV4h-W-H@U=O|iO)(guzo9+e`75g&g6|~v~
z#3|0HWn;WYY66TlZJolD%e=5G3+@m8nEPSki(mZ8bky!2C~In>k<H=NUbfg#r&fA&
z)7sqhLDYgQH6TmDG_2G#h5GFEa>I(-^u~H-euLWH137LVw~vXr@f|zXCg84?bv0+v
zsO|F!E@$C%^3;Q%UI;$o)RRj^@Q5E8KhJYGOuTrp<9xnAsW=g1GTzv+Z>*(Nnm}p;
z%Pv}Q^+v=8ev6mW*FuTy`l~{y6>1@+%q1oEr5KwjMiX$ZYXhpyN$O<}rsGy{0^U7m
z*xd%mzkPhzj*SU)wc}@ZaSNLD+rII(OD!zRf_IVNg6aG@5bv8to3`c$m2aPpeD&rH
zZ%!BHl({5FQERL*3<nPT*G(Jd8mQ--NNViaTr#<=S0K9UB&(Hjdmm8oCXgPR9hbDh
zn4S450^+SK*yC6qXHv-&gp?C$&4hLz{r0bTyE75%);iGlkVG(TI<WFUj4fC<K)nSU
z&8fD)Pqv`w8Nm2Gc6^2CKA!PGTgKs4Ow&!FltdC?K04-0V9o>U5GdYsk9_syZ6N|q
zB3{zpiY@C6(j+2`z3i_PT+PtSz1PAa6$FQi0qM3}*%)4~Jy0FAQ)hrl|LE=GKfu(w
zMKtBTZ})S&H#fOGlZ6&&Ha&%F{-%{uwezZ|GA|2n=0_gal~YbQZ_R8Dmkax0V7$L0
zFAMM9J#snEl%o7sKmI8%A6_$`Gf&4eW%)U|cuECT2kHqv;GAbJi!nrUCf5dDnc+Z_
z3owvKNL5UeGCa3ss3cdc+nKY#*X7E$Z{IN$#~<JS7#BR>e(?o3i2H${g@~~qy$jdQ
zt#d}#V#&-}&|GkFK)i3)XY9XR63ensYU$PcO&?KPDU)ex-LB8JH~*g4a%z8ezy!dn
z!y^q`Q0pWq!t&^uPs({|3`i$LB2FDSwY@q{_)?IvFiiu8yOGmr!4d~T@SbTJH?YvP
zee!$lxOci!HYuhK7+%NK*1VSu_!*osc%_!gG8fiiL2DrdN6yB;qbM%3oyoMbq*en+
zK9p22knZOEgzr74KV--Ho*lpc96^7+t||Ab<{)|Rd3FDihr3rCc6+?@ETJ_W34tn(
zC0E3aOq1iyo3~u%3&q2{Cf=Uj5p>|;#ewxPvy`+E+eAF7rhS#QfRrmTc)c*R3Z)3S
zB_xKTga#<xZn@TQypn6?{{D{7Kl_Zqd!z{CII`x2%jraLO7MLRpv((7s&z?KPY8o0
zK#IjlB97X!5~U<dWUDcu363I=JB+1vw%;Zmzt5xk7=7wb<NrV76M*$;{BWZk0=;O3
zE~d_w5a{4LL9Ru6u2x*JOrM}FJ1#)W{O3xHfoU4?{)BU_HLDY13@z)SH#PZfi&Jx;
zef2F1aLtzV>u}CJJ7fhRhE}2zBDGZJ`b6}m$xte!q%7;Yu@J<G(e&YYc4O0XX!q=z
zpmNde8n`*^`}}CT_ATqea}a9N((3cs1LHPdSEgC}oSx-vL4B`H-xPL5teXg)l57S0
z@70zSTT9~ceCE6)P8VfKh2!OdN&qJf?CCQo+dXv691)>vA=i8}H|_449*y~+J=m-_
z*)nsEa?>PJ!B_x3pe3^|$pj#^^QXF2N2L_Osu69ndg)4wya6bhjc(p%?3NjG`&?Y}
z{=G*3<{oc=!Y$7va!nh4Oc|s1?Ra}Gtt8W)S+h4+X)L&vFa~@Kn_XY+xG`ahGK`V^
zVTW_(EQ@H%_TJiY3pnYvBzTk<U@5k)`pnKiT6eo`D1_*7&T%(RXsygoN5&zT{#u2c
zVO`U<DYz!)ZbZ(TQi1lY(!dw*tQ=L_1N@BLa%0PXQ#-CVcHCdj%~!3)d`l^Fs+pxI
zd72n~WaoDvj$BQ!+Dax@PUo4m6ccN249b)%&J}X*r2ufHg|!f-F*jPi_xh-`*JumU
zRq9P3T)BzZjkVF7RV{7)nm9WODJ4oRH<s2lz+7VsUDvjh&c=`}_$ZA9+rbkmTN1%a
zA?voTP8?O-7P(kA#!be(f6RA>e(z6yH^&ET*;ceO(AvvX2?ca9CLC_U2MAOH_NDi5
zv-zzR#=iwywys##h{L<UVYdU6y39;tY~A^qJ*U({AlVu-i1a}AYRk8i<DETQ+Rj4h
zXL6q_;aNgxJEwX66g;!#_eQOzzovS9zWP~gq~^u|)rw|ih!Kec(=ZZ3WX%cROSl!5
zHQ{_@+8q!awHhl@1jZQ5UQ%Nv^~Jw&F#5Wy1HBTgo?W|&f41*!w!A48DXvyy@Q9f1
zTnKm{=;hiXjyMdsl91d2a8<OLrIk|93eE-lw_e$FRoX7o*eKmk!C7fz&V`|Sd%aw+
zpT||H?z^_Utyi<*+vo0gee`Frxtur8;MGs*e%2%oQe{5Rgedsvny|UHgvr(KLxpJ^
z*zb2ZSpfq$ycdQTuIEteQ2Ajy*3DL=O{%?pKObegc5ATW*i9ob1jcbBczY20UeOz_
zcoBjV!uFy4ll##m9d*Jv)5|hB5V<m_^v214@4x*4xP?m0CS{wRTPVBRx46#zyv7^e
zSz{#Y^z9g&b<Fju9G{-hsthPTSeadF&#;Ii#KE-hE|pocF{Mvu?gq!WAB=_N#7bee
z=e^!|%GYEaxAzAi-kdZk*Kakzt*_ZtQ?TN^BZgq-u7eOfxd?e#af5?kjZ(BWA$cjr
zhTg8>`sv#Jf-2NnY+W%wyC;}Ssi^PsZRLjjklM8E8rXjeMQ-QCM|2c_8h`Wg3BdX^
zez?(gmD2w%K+`7ZdQ*7AO-2WU6QXm(s)UwBLZG^a#Mg?HiloXkjoiQ36T@59b}$C-
z*o{LIIjoeLua0Cx%v6*?9I{qQE1niw7Y%0z&-L=H^Tjlb#3AzP<tuVZd|fV#(d4>w
zR^|mxrwb{r?Baxo5dAH*VH4q;wU_T@doctPNqnYvWur-FHDB}QNNo<}hg_|#gW}DP
zOt89clzp&Gexv%<N=e3&!dpiqaUKZ~5l_mgErT6CC1;jp;gVLq{N`KUoiChL(=R*C
zE6$C?(CE+&wX{pN&+h`?-U}rq)-ogBwS(Rqno<&o8eN=ff}?GrI%}h7C97qAi->i6
z8FolonJ)`kN&~PQe)CT%>k1ST!X+y?DK!ErO&d#DOQLt1G{@by=eN5%g2U+cU;%Zd
zv0F;rx>n+Df`dN!%_;lv3jU6c57=>)E*{F(%SV7%q0Wi<>B#*qFpd%5x)2rTP!-UM
zD@sn4{cgw0SNHgzJ#Ijl_wpS(u3GV(A%g`&i07R2nH`Ix0m%O9I5p5{jFE91`26#q
zQfua$@Jr$l5t_Jng7f)84xZh(Gsn}LUAY^<Ww!|}o4rMEt)-1Kbd4tZz9Q~#ZpWn-
zytoZ)zWwTS|57q3J1)yYI?bfZ%!_fy6i0sg>c^-zhGnkCSUR0&-aVbD<3Mc=e=f?J
z6vza>SjW^>W`JHD+|vG6rMb*4V76(g4J<YRa2!o2x`EYAhfW*&vG%Tz&LJ`4TmTAl
zN}vXqcp%Wa8Tw$2GipVupan`+O0m+KDxl8n7-&G0?Rlvf_?}88mtq}Ooryp>aV_R;
z&U6RoeqZbuK4{B|b)d6umKGRzg()d|UO2sZVn4tzO^i)5EO}$B)n1mFmHqy2$Cd$t
z_rw_a@sB>oIb}V+!3B@^E$Gd#ToyEEhG8H$JCjXO{;EN_mBp(Cgl5Z-CEozeH*?or
z33^KH#+2V5-^0)-rI@Bvt%U7-))x4g@O+HXx)K-RemXEr2kwV~=p5(AlO;*SIxfz0
zA`BBRcK61DooCi$bG;v93+6qN<XM+aX?#TkF*+vEm@wO|R}@<xZRLPw%cTYMifF~v
zN-RorFnY%{4ov$UF%DD;hyhx=!ye}O%8bs|B!w#0wM?D;PRUR*DYb;c&?F7A1-_>L
zW5uH>XO=awQxKYeg>@G-u%(zkBJbJqZ9?E<3w*CGt^HFBMyyJE#?6R=2>0IO*2MA6
zJ6=qVaSC{~)2mqfxT`ON<8avX^3^@R{Q1!k&2GojNUpYCMtH4+{+V3u_)-*I%N0J6
z4pe=&x}kT!e)Zy>!(q=K{lOn{JU;T(*I!r{%G%akgkcz{-Z6xxrQ5x0-AvOOykG!o
z@oL>=<z{(*%VuqM{DBwfwqxA<e!m}^KLe8u)bj{F1cnd^Auxo%K6nP_ASJ&1;w$EP
z;g5g(IesTBDY33A>zWWZFphzg7Um^e=j%1|_VLX4YT)ii6K~JKm`}ACZd*I^$8-aP
ztQ=de7`-(AUNTPVR!S`0mc{PFY8}i<ymt&S;u|PD43TxYu%1tNHzAXC0WMXWtqQgJ
z_WO+$y)~r()j}0Z7Mcj9ogbwZQqHWopo2LQ-U&EkQ`Qb-cANbGlk&e{i9e0sv+)VQ
z`ZRvH;cvu^TUI32rfFdNJ`Y2%tYcO3y7F}VhTV^LOzsZnB1mk)sFhk*v?jD>VsuQ?
zU|Ha52dr~~3jv|x)S=lj?Is=Gf-q^Fxm<II+8%gKSO~r5e`tr$PygU^9v)us5C2#H
zn0Y?)Z~w;`1)Ni&hcRN^bp*m7xZsG<Q*)){mRWVzLr_bv({GMe6Jd7FGODOmQhaE9
z-{vfeksG%6y?w20j$75vSdv!WD2#E!`@mWXkH-_s>BNP=AfCJF9-_03W7R}s=bSjs
zGsla4J*|~oU`dL1j#AdvCb<QLVp$$*V{N)Qg+kR#)r_c#?|N@@AU(lX1odhqKq5Ey
z)c^u#Ew(F0Js#pnRpH(Ff~aM<`W%#$$;*n@XkY}=fJC=>>$wyvg_u|S6nCA?k_*eS
z8ql{nzAfXkQTfC`z*0Ww5Z`8`o(03d>*IUu*w@BD65&P5>UIsXcTaDbUcX?HfMWoB
zE8jE?G_5POB!<DW+l@#o(+pxv%J=P9D+KG8WMxjxDV5gZ?ESgy`a4ccmo7NRPd@vM
zmoFdq7yr$F%bFJc=l^LXW$U~bJ&Yc+W5ACfUWn0~c1Ug=bmVHc+V%9BlW2C_ng!Gw
z(bx|x^#4*j7Lg`;YD|a@tf1Cq&xz@1EbGeKcPnXLNsG1GRy>zwK|{q4k(906cU}^;
z!t00E47(l6rE*ykUw-?DNU(jXUW0;b=QO2g*#g;314}bh={Pvj>U*BqaZ5(jR{tQi
zm9RKRN{M;Cm{@OBzI}X)_rm>d+)9PCDq1R8ij^I;*wcijdk|uc9yWME)NI}d3-oI5
zvSz@d>mclgsKB#!dcS3SRCWv>wB=C1SBr0)z#-U)ndZvnwDRuhiSe^LJOjQC2*H3u
zRVayCR!Uxp@rScz%Y9vg;D`Hr?(Pr#%YXSVfWj~S)1Q%ZX04eqIHEu;g^V%{*6l6^
zPi_oDR9hM9`?lPd-&kMH$<}GDZ}rcv?dHGlkIjB>0(TMFM9V{CHqbsV1do$Q3;~>-
zokNT~?DpL6@6l4BR{VZvtgH|y70%~{{mVW3`v;C`=H2|nH{ZS^r%K5X+@8a<<B~md
zt36W{dlp*hs5O4brcGIu;Ccz6C4vfUrMPCx1@*q&n<aQuEI2J<KwNCVx5JVLK6nOk
zWN~C^5|vfS^MX=Xa>7Yq2m_MA7280o8dwdWwQlR9#dN>=T-#n?4Slw?Y%DFKmVI|T
z6YKuKv-dkY-1V93_ZhU6_P1p0U2sfJSm(;+vT!^d*_Q`IJW+z_QdWUdDSAQIGi5oO
zVEb;QcjdJ5ZLu=%W*|8is)854RT^|V-k8tt;exXHaIV=h?8b@RZsH&Q$=~PY%Lo4C
z@BKZ#`szz4E2p)v<P0sz>f%5U#}ET+G0jS?O<k|Ox3&aEKc{;|#m4ZzI&F4|Jpt4w
z@&iNIUcTA11Ji}w_N?yb3*hm1Lh$CdIs>%c9v_*HGjHFXs9E{^#Y=(@l$6XrO^Gm#
zjMK<re?Xk#-$dqR!4uezkx~-kln^~-Z3)5l?3a=$HB)NZ^st<CpR*c}3#N}H{e0A>
zZ>85DsO?ccTH=UOQEx2DTr;^O(wZ3SXwn3=>6!Y77f;Hi&4~H_&f8uWLSS7jIjp%c
zu$7H9{)~CcjZI6NdDoIA%@5mv==N;gZqA=TtiK!M6M*$;{BXl{N7)WF?;Jr3p~YUo
z3HQ-M6;e^kx^g_d<yH9u)Oiay2-MOvBQ!%z)RG9nF^rMkCe}#<iF~KZSERH<t7Vot
z{k=9p$Zg#2geq1p=qw954I`g__9K4slOOXh{`o)SczWdLKl}Gw=7r0e2p)z&3w*6q
z!Kov7M=G?<oUI9<X!E7y#=+uBrS93>s}81KFH*jLTiy=Un}2VCvwaYXbr0)Q32j-p
zF%EbaNV$;G!ZEL$G7J)kf3U;*NYM^pDND+nmxar+aG4j*Yb!x2c93;|(L0aBHLaEI
zFqFog%1t!ZCPT4<&UM;ZTZ>+<pwjyZ-6p{tV!F0eRYDvHMVX(T@D7H?t^p`Hlh=gT
zc%8EjV08ze7K$cPiO+rpcF3-4+L(h~Lr0q9g0=$9Ep`3e6@I^~%|~^>e#nli>6k=k
zwEp|S0FTS5ur4dd(-W^tK?n^%_SffG4T$b`JWfM<9!<Q}Ir*?1Hz84zQD`mU<vly*
z#s<W+JAx0y5curnOa9;|pYt#N`9EWx&-~_Bf5Fr7$m7$I!NK5ZL0Yp{XTM)-A>RbO
z+X`2}U7ans+Ht?2+RW{5YR8)s?xeAx*nsE`U@6L)GLOqbS_?T9cEgU~j5X<0*$t6Y
z3a9ghHCd_b%lilJUp%lZmE(Elcv>k%D4SlK?V(mm;Tj0tzdBE;BCWHZ1yHc5!QI;N
zrfXpqBzj%%&h88$QbOf)JTs1ghr`awNg5)krBY}mCXN5+<;sxL=DeHVYd0XybVvjP
z?z_5uFS`(~CFe-fF6!cT9~gS`pndGN9E)i`MQ?!9eKhUBbgs;o#PM|Ebyge$P9vgC
zyRjCt<&r3Q{b6mnb<VP+g>Zkj=k=>s{L}ySKc?2q`t&uY%ZaD+i6MBR11;9hUDR}4
zV(`>d8W_z7ZMj85ZH~Gv>-C!6w`F%_IDF_y*~*W$9m2ZqIq&h_w-TqOLn$!$f#8FI
z)oLuWyJ6z>wCCw`Ce=!uMq(Tg=cx*pdF9SW4u^Yw^X8GK^CLh1>Q~ePhvA<4De|xz
zh~Asu)7fv0aafA6BeQ0-mmgx~jA%u*v1zLI&8&ge{wA^1bb?xf&|$%aw7H5Q7;6q^
z>|JNqP^*%cM5VIiOz>h2PgQZf3CqSNxn6_wZez3AGF(OJ`yT2V=x%~<-*3yeYxmLE
zatrFYdA34@s8-1kJlsVq=S)jwnG>h;nUX6a9?yWold4#`W?iUhMwScS3*$620HN8j
zNNeQNO6r=3-qun3#*XEE-}>2gw<HLb_?pInyZxR&{-ZzQ$FE=Wr+@M%?4sw*H(&7P
zbmll;?7CAsD?|vvlN9orx1>^^lj{=SGLNoaV`rFlJMOs0Z#j~;>6Z8Y?&QX*+vi%T
zof0KjSK#*fE_fc#C%*dXE0(kA%ceC`Vy31-$(fQ1qjOBdz{`hwVsw;LxP$@JF~&$O
z$zrDvs7bb~QLwVSjp=+1Y7E?|2L2#zvx8pBFD<#QrIqPA1F(h=?fbP@mtKWh8%UP(
zwukh_TA`Ip*0N%}Y-6#yVC>5fJnPytL|YlKN=r0kYr1i}5YLuGuJ?KSSZ#l5Pgd)*
z?|&M<-{TX2^=bTI!_gdNM{A7>)8H9I7zP-E<HvV*Xi?Vlg?v8pAOHM6@$g4~#KTWN
z2c8fDUK}39yTax24d-vZLN&ABPsHd5jrAaQ($c%IST<15TyfZeNpsBlR@kr|a|A>P
zt^r+C#u&KU?@Tb&2XtQWx$u+MpK-dJIbJRtPZ#FP!evQNtTx*@*zY3enwe{{%xNiG
z0NTrnfM-<Lo+0|~S^C~~e0yR`ZzOK9Nfo6?HDS~as5onJ+v^gol%ulSjRfa;b@zfH
zz&Rz-<wCw(xc4KI7amV%YEkaoNJ=ZG%ZazgBkxWpf*ZKsM_&E-o~pt;7nZehzLcwT
zSzF0{>pW(%7nxdDDw!%3-vrzn!{2mBT-TSf*DKLh&gmw6A3Ad?CC;rZL9;SHof(I~
zxZiD#MhF4#Jn6h5>Ik8AI=TUqJc6jh_m;m6aH+LeNn#h>t(FCHTJh>FxOb6NAZeTP
z8;|%l1OL%E%6!<4cg>Ec(K7_kkM=v(HS<lj0O)+aa9NJb<q01>o?xAXsNge~$8R}(
z{Uxd^yL}`y_MPkE!SC5|6+yB=m`*GRi&D2l^k&EHU(#823ez}n*zXxbL{vF``!zWy
zUhVD~C9n&T%aTdC@_3r5SxvXgIS#u4)x^0L)2gebnNhRjo3-6Oy9dTC;JuFjay#w{
zQ@v1edws7h7KalP=(@Y#8I!{+z6#{T{;(sufxBtX-~*R+0aac;yrSsB@pwU8;Bfaq
zaKYG2rQ%fC$4GWcE|5H=wl7(95>Tu>DQhC@Y6*r~2~G0|tvD?pW>b~X?p`yhL(@;P
zH&hBz1&Ug!NUnGVs#2DW_`rUg&|;#_(8{87M%1CjfcOps7`sYAp`>K%)SiRhy346-
zU#3T7O+tcmL>DbMQd-b^vqZP+EOO&Ce>~&Ew(J-XrpXb5@GwoJHS?w%5noAVX31yf
zd?ffgV_lStbA{j+&X3>n&DUR1b@{<;*-}t2jU(eYaG3V&Lge)3Td0Mx2tMrCKMV%I
z=E^tkPUKRF&YN$rgQAIgv0789WdomEGq5J?UM;O6a&tbs9}<1omOEphV$QXRU0-b(
z6=O%-mh_9s2}BWI?qA`YBi9x09Ag}L_w<gpUw^~poKY42;E#TUQ<yIo=F6F7o-MF-
zl@||t!USJ`^M++k%<IhERM_tV@hpgFOO%Bc-X>I6swPS`9ZykXQhV<ZDdsyN4e+|!
za%BJ<MLiiMIgGXHL2E`+!i`W}v)NL~^K8LujQ9|VJ~rX{LM?^TN&;&&Slov|jDcmH
zQMJZAG!UYsLS8dc?Fq3mRa+P597AhjQCoTF`?joo&DxJXCjFc)m@iHUB1GreD~!N?
zKN6$kVH}|r)^$dtkk&J~o~e4FiU1ywir0cNv##%$AHPLhVYfHN+%r3FU{Bu~zt@i2
zS<x9aH~3Kl7fnwq#=v2B;4tl2PZyqe!!K*$bbR9FxaX&@f5Z?*{^Kux!Rd0LW?Q$z
zZlLH2HMcVIs~t-_lf4fcn$@pC*znLBX3qv~*?*;_Gj8k{K6ECVmZZbz(oUNE$uQ!5
zurhdQBJKz=3{2C+{%*&QUOq5|J)eK}6UGprW?XUj8X4TcI7|%A6UFi47cW@GWQ;i%
z@IK-f1B$d}YO#IG3pg)Gu2!Ye7Agi4$T#3oTl1RMu%xQ7CcQ(&O1!Ju{f%Mdes_SX
zEb~HY%v$G!X&8C+@QN^u)Y3G!oRtuIXFy{um9;GRA>yOy;@+|bZDkBF%O$g>M9PJd
zD_UA`bd$+*EgmREuGJ~sj&I5eukZb7{9PEI0IW~r2OB-s3L->-QH0S8gBNzAFa*zG
z8le>CvQXBUZ@&DR`Eo`zQ|gQ(5Hz+dTw<Ni9G~8ywGczVi#&67`%tb(kTxI!-$KFn
zO1bRscCp=Qbz(XjAp`?5a$z~0$a!HG2ecGwE$qgD-4I!qvXz&0ht#{4(P&QLcF4-@
zl{OITI}Y&&CiEsC?m=xAr`!T&y5RZk{q+~z%m*e88$)C_j))X;omq1srNqd97h_uu
zE}Qe23(K0Am(^NiyT~pC4)+gK6^_gZ)*(ohEs)&0nDu*Awa#UI9rU^2eiu}Bc<3cc
z&=^jZrRgki0n_oSWh)0MYeAcMX4f%sy(49@a;M_zMn&%N?KVXluw@=wtyXKatskuv
zvwgl3(=F+=czGUyegF6`dCl*@_&z%xov<GRV+<Uoksywz=%^)gI-Xe9Gr3%-Zij4r
zuPdTv$4`&%P%Xshn;rw#`D>%wu|n&*W97e{$MgOSUcV>XS_K~%W3+Wkh2y(-poLKa
z8X`;+ckSzZUP;=s`VeElw<ZzYj<?Tw|5~5f@o#aX{tN9Gsow)=w=xE2;LJFTIB}$s
z5La<NFb*TTVaE`x>>|%ISu5f^s<18@T(GXG7z<-fD=Fm->=!Q|*k4=8ZDV0nEp1-<
zkCeU#xea{yfE{lc;`Yv+)lrQKBN&4QwSBlMWY;saEnBQDnOx-=fVE9?)oo7E_I{no
z(Z$-)Z8;of%U4e5d!SXn)*Gk!KeN)mbK^s{JUPc8X3H^p?xvAN9LMNT&CHiGDb19+
zP}~4pNqz+q>vHDl-CH}8e;`{1*%*~$94yhKaCv$(4W_IN-dGCv(~cMW9q*oIN~ws-
zCMNIw)s{E9#<pkQthd~*>+@&z*?4BlH#gA(uYM-q*m4H}w)+$zh_SM!aYV$C=M`Gw
z<@j{w^mruC)`hQ@*2E+wN-kzIsyI=`A)+!ejUyCTN^bY=S|HnJ_4*~PW?waL7iQ)E
zy*x{CCc%$9vt_fI_Nt8$>U#H2wT2Cr&#|)IlG7#}?wmIezLx?wrd0>8RC-yqx5h42
zjot~yl&zK8_6U$||83Gbh{!W_(GS`3cOJM8I$pnfqp2Hgl?;P-Oac?Ip8`Yl+>Il(
z7HG`6%lSlJXS6O9qA~5XqGZsObvbi9K0!N^-FtR?d$RRr2EWaYuQPdND)qm4=LjJX
zy(fq#T^25?JgVuQMdvw;6X{NwVq{rX^UIp3c<`mY-}Ox1&QF0YLwqxCed1e--mXn<
z{&jN`-*+b0o{;V9Ag!7q1k-L(C6&3Ymz4wWb`vkAd-lVD{kS7I$C|S3QR)#fHno-B
z!W0J_!eJaKnAU3dGwu9U42U%LtO%B<+0L(4uDPAOmcXxVCX{L=g<Q8oum)3O9IP}n
z8|%t9{lnlr(>Q^1W|OYxhblsCrTx~GQZyL2*}<soq0kLA-Wo}59fF^~?7C%r;85u=
z`v{XspT^&Od;+jOjUQ|rhNdGCJUi#uJI5|~M(?=W1;*(4@%;g<E9ZFP_~uvqkN@T!
z|Kwl(34i(z9;j(T+<?D(Mah-<c;Poc{|!I?v%g|KFO1V*?VOv=Ql&Oc1ic0~=2aOj
zHac?6q&%bKXKtK%td<>}!odAuXN-+p`T93s@$}7G-u&uUMDG}a;}C?yVNZVjn&3zB
z`fJXYnVb?s2;A=`>Y^N(jm^;=G_Ifta6LtfVhXlmP9N99wpZPEEs%C>w}Vx-jFENw
zxjl^R^NP7B^eP;jBRb*b;RVr0LhzJYSWC97S{%R!ig=FmY=LO63sN|p7QTIZ<WK(O
zA8~*Gz_i~nubJiiHG_xUAfz=naaOxMY29B+X<fjyfqk^(jB9{l3rwW8^$=re6z{-!
zs<g60El9EOIL64d+u@4=O-_}7u=fK=O|V*vKoO)?YJp3>AkO1Lv}_`col%9dST`7P
z4lR|mR!jtZL{@W@YnxZ|-;7~LYy#71y#^>9JbiC$ji5e)@gX}tc+=N-aldDbk%#HP
zGAxu7Iezmi{$Kz7f6t4D!t49Y*)0S=5%+sa&RmWse*Lpw^NT<KYu3xkem4=kIc_x2
z_kBB-Dvfz(8cRhLYGR|qzc(CKGB%9^gZDUvcVEBZ{OwzgUwvUVI!+9u=dj<AKl>5m
zbReyn$J2?lWC(#5yNSG1p5(IGajmT>LieRwSGI~}9&0Xa9U9vr|1Y-V_H1>>xwfzE
z%@Fo+A_UJ}6K{Tee9IUHUcWeSKkm63_LL=|r65Da4+9b+xx(o@vw!)TyB80<JwCF&
zUHQ$I-%yG&$jCJ9xVyV!PKh}SA|`Z>t5(`1fusq^&9U`vV=|<6r4DcUh4sdcgYh|f
z5U(D@n^P>MpzBH)4%9I)1ly;4^{re{@O~hMk#b(CIkOhawhwV?r7(ryL+jE6w7npg
z!o1F;w6=02o6Fi7FqAgyl#F`2AF*v_oAVC|$9_!U6@1v1?;K1{xPM{VO83JZ%eYYU
zh4VLG^FROJ{vSL%cz*OsI4&ogM0ST4l$@APkNoE6zvkzE_H)vlelT0^fJF3xDMl_&
zXV&AHe>cBl@QxjUX+Ls^2Y&kUCHra5x+LB{otQ6~7y>U26KSms?t+(oCYvqWabu#^
z#aXF+?^xM^H+g2uZ9Ln=TX^2flxV<E14(T=Xc1Y{6mqSB$=&|I7zf5USObAtP;VW{
zPBrnx{Ej<$!9EVW7!T~iKwUFZVH|f1Lu81KKxF5H|N0O9gk?@VKAkx1@0q5dy~ZQe
zZL{iK;4mK070&BSZ1M@twf)VRJOi4kHBfJCxklr~xgvNq5P3gNe74&&#c0w~V&`Pa
zi9%&{g&Hhz;C!%-YqROujT!Jt4wdSmdQXiGf{BBRC{BC>!wi%uJ!z=WIB%FLRz+&I
z0N(Y*W?R6mgX>2+roqw!!FfjK*g4OOpd36r-0v8Mf&1Y=)yic*a(eTY|L6bkZ+P+G
zcy$jYn;pmdiCQwt`H5fr)i3z#fA^Q1j~Dj)X)8_dcHCK%0M2N2@Buq^RDy%FP45+e
z?j!ohVVKZdnU80F`4_)njE)x%fe@T^84r;k9}c{G{hIwS^5$^`6%H@<tgAA}*}$IG
zT;P_0leKID6*T{n|GcE(Ho6_#`?9_7^-P9F_d)?59J|qTxZ4xr$a%StN~X-28jK<K
z`K#Cb;j33f-4ir2N@$52kFO3Yl(mvBnH0cfz!idr*DntoFNrr#Gr2PqwRPlF#fxw^
z+@V!D>v`LM4EM35x@rDZih!X`wULs$tv6s?D<K9R4hK9!Sre<D?OF4|{D{MWwPsSu
zNU-z7%&byUwr~bx2fE<t;+}r+=8NZKKDPG;W;#Kgql$Guw+^@Myv?mP*0*!}zoh2#
z)A&0vJ^@&t#t$|+Yt;8Y9oClKLkx}~c!uaG)#Hk1$t#yPna8(pIlXyHt^*>GoRypk
z$G1nm{_;z{{LPo7IpM`q1hUwJthBB(eUr02vjGm!8!1ZztKjO>^zI#ku@6Qc2`Z2)
z>9W9*IHfZ~@c6^dlzpOS7bEw(9fxtC=0Yt(bq*&H?-M>YAf&Z^SG{sH?3%8@Y5kx>
zb!7r<?4O=b(UwDzjZk2}apS95F6PP%5nyNm-Vg%bxve#_RwcTCi;ly1VDyneOn|a(
z@8!rvnHLMvO^3tE>~hUOW$+I50u|OO4M=p2c_q}=ISe&+#%-4~n}c3rWDXM5S*H|b
z3kb4kYqn#Z6xjj|wKl`vZ<%3{D>z`C7aRRsOjpIS?d?$X<`Buv!=YjTmh)CFsA@qt
zE%>!L?`vCc=3BPEeXwK7N8&)o7Q9&}IEyE|Q+zP3jTj(CLGcXJL|PMHfAKZ%zWIiy
zx1W)N$NPzxE~GW{^zM<bzx<M~njHrT=1(<-jUtWVb}a{!<|rB_En5QbeF#BsoaNv=
zQyB0#BwJTWP|kU!M9)sF8>JL6P<<TP4+CjkN!eD?iItvuFPj~6V_$vGsXhD8HH!$h
z0QviNEL-NYVW(SGyTdEbnHsV^N4?ZJcu#CX%BY%-gcpJjP?fYS<T+7GWeOIJ)>b~8
zbHRyq{91}bb7jd3CEFSd<7oSX?-^X$&S|WwD3Ks5u9tT;0KmDe{}4BKf;Kxwb6>@6
zf=r942|-hh$>pvByw-iF%|Y9zI)kQGV~8LPpa%m7dbd{^P2kp`w^gwqUd37eNbT*p
zwRKl)b}Lsqrho4v9pANO-Ol6?x-EmMXBu~;HS^VPzT)jSUvqr(fXh1^k(^diT6uc+
zj<3J?lCOXLCHXtH>^57bcaW;xz5%(OPuFMVcGx;qRf2fN7>Ul|RM3<uwX(Q^21gu6
zN^;bkh$8HV$lWxOQsQ(rmVuXm7jF#)o@J42RJ7?J)-8DIwtV~Ke9)FRjzim(Zta-&
zY`ME1Vl2w&#Q-%R=R|JysZJW3+cS;>`>^K__l#jAI#11u*<u{=E`ZwmI2?yzB;p}g
zVvN+9EivX?TZ@8F@gfX95+xFe8$-`*Mk!tJS#7`CPBs-(lbv_%?2`&nJiBqix3Z)v
z6?HHU5z%V8K;kH!ZKZYV65N5MQme7$y`w9HJMXq@vY_12HA7pwO6yj(X>M9DZ7+5o
zwB-+5@yB7?p)tOr69#8h4MTtslp$JUp&?pFxKsz`^UBx1`HFYne8cJOYkXdCE)vR)
zoD-+VN51~zOTPH^7thY*rdDoe$JV(TY-0W9TR)pMdH(O7tm)<4UQ8p_dvAJ%T3J(t
zAk2fY#~dau9=ve3o2aTBPu7jl(VG4^Hs7v~ZiAZ_)}C3rpzErCXQz>VrCky6`hLju
zeY?-yQ&3JE!3Rs4cw?qk46x0)b@y&QO!Nje`r>i5oj2&Uk%TH*O4FAz(6(0*s5-pl
zYn4a4R{DA9%qQ?-0`irDHhn7*+2rf4L#b+ha8WB0-auW2=sdeP+FlD=g+uVBo9LXg
zJ)qcL;BW*Na0pdbE6s07Yv*M1i~6xE($9E*wPWpTd9%xE_vx^EZMN$^w20qb&`;y{
ze0&11K8+u4w7|htz*VaAE%Rsx?id`y;8{^tnRxf^#5Zr=@mK%h7rcJ`nlvO3Vd4eL
zCGqW7Z~5h4{*s^l`Cs$uvj@gFaS=GV!X1HvP{ca*P%0FV>R5F(ZK`BUf+%~06GanU
zr+ck-*zE&M-t$rhO3Ku<lGSMM{ytNq5cJ5HM}ph)B94qNcg*ijy!9i$eoUkikZ~e}
z%5*;?<4Ez5QjLLRI#rPD5-H9cMaveo(zNdNUPE$vnWHu*t=2-Jl6@t6PZ3Y43e^}y
z{V*-v!9)qbtHT}BFo4)Jo#qU+;4?go2kwU*KmF_{9HxnMJ|e(e3rUsgJeMRqo-2=+
z6^0A4J~6Kg?=EkME^@!$5q)L#mG$H(>Zs~bb>!kGYl6k$!f2XSWd+5wFM>mZb+q$U
zp?c7WcSmzr7%5cFLm{XTwURF<%HtzRAyIho;tuBnr(|t_y|+$<sZ^v`%lE8GZ~?qW
z1qzCc!2$*XD3o<Yq~LrZL`R4d%1BW~m;hmwa%I%iWL>%RS~_YDdASjYrUd|htCsjT
z8Uq&ChYT@PD+%*L4T^+}3ql+fVq`Br;ykZ>{U3hDU;MkjAcV||htCK;QnPTmtbFzL
zTYmi?e#tNY?B~4x(F0@Lkq2d+D$!4<_a^k85_wGw!+>|5xw*=NYZUnM+<7%T)4cBJ
zJo`BC+5R3)8NFm!G3W1~)Ka*_x6~RD9T<l_ul75B^5O-9I)3)$BkC2ooA53(>>?5)
z)kUmbxw*(iD|1>+KvgOkvtzxnV|mYxt#n4H8>7Qss1B+V>a87P-N-DMGkXrfg2Z7k
z2RGHix+ae23qeQx-GRrqZ<wE+sM$Jo{=J`m4kE0}Y+vh&_m0DELdFAOb@;CXsjM6?
zCw99%`{}?C7QD>ZUV`bzNnjrjR8<zY(xG_Q^frtQYu74L5vPcHC=p3V%k*lrKtP1x
z0%I5~==B)O%UM$f#4|6e>G(Nkt;SV|xuQj>$spvS<5qU0-WiL^c>|=1;IYn4!3X<3
z)*wWgC<tp~<a(+#tWZa3YuwqhrD?6e6~y}3+~g5^9z!Y&vEp$MphO_7NL0eG;(X+l
z{*=ox@$dfMf5A`x!99*A?uQ4w^DL!uzAXIa>o<J)7r*8=fA&j0`|%5Q!_;h931Oo8
zKrV`=L`^F-OjH?6OGYivj?`y*Cp2)Y17Imv4nyGc`&ag|RH)~K<cg#b9hBK!;F9nx
zxM|1TF!Bd4U$BdjzkK=%YQ@!wD2d(R@IIif=_=ClBwCr%g<3OJ3t20h-apNqaV-dL
zEC%mAUWBYY05wpx+j0>Mkg3qSuPIHR(;3+CFbY{S>smQ4M{+LA%gj`F_`@APx_imP
zIPq%woP9j-FdY!BoO2?&k^9|if=tvEqPL<x9|P3|?&8EHB_2<2NjX^;&^n`42;y;~
zct3)N;vI`y$&!({Ladxz!s8hwP^C1zIFGA=S{@OB1!Il5rLcP#d42zy;5?bi%1ZEo
zKltpkHf_RXy>MPHtZ7EI@G?v|CoHKY!g4ZD%{ya)xlA05%_$|L;=%85OEQ*_AHa_!
zH&~O4l5k?cSoTnw2A65`xo2mxtz!H0BRZ3J?F<>cV+@{M@C-vCL?w=eF&3twkg`M1
z18=`R@#}Bj@Q?mKpEHdM6FZ!Es`p$j3txTthF|=LU+~xe{;&DbPhK*P6B?mJW$+Ua
zO-Q{NgGxXJ);c37(RoW!sHdt}F+nqvUJmE6lEHn9j4?6<PdP2*wUARpf^Zotb#OQe
zBblgyKYabbs~0ag&Y3k=R3lyj`)E3C#knis)Ijm7_N?OD-krNnmf=ZGdoswr*}E3k
zO?S0{*G{NnU)M9)*jYh@7$Oh*7wm>W90E(Ol=Z^fH*ct=Focmh*e2qnH7Ljlso4^v
zW=%MALMih~k~8k*NQjZ$&Vxc;3Z4YTO7lZ!eTy|1LY)C*5}Dk@%1V)BjUv3TYQ<Fp
z+$<qf%^$CVFEdV{Bt=Wbt8f<wetiD|Kd;nWNY)mdv1tU0INQXsRV@S@cheqJcw0{7
zY?Tk=-fX*6OVd|T@Zi+gpE2z4ve>@afXC}X(U7jj*e+!Q6#L8GXGQl!;HU9-V|)Uz
zK8+u21ZTpl0j%?2b6BrAOW3I0)!SAso;Tk-@r$4RhU8`tnEVS1e82pbZ@&DNcc&AF
zhXZ5qZ8L4<JJnbS)M6C#BGd|56XJxOA5e8IA=2|gmO(%)upGlcbioeltW8wdTGvy&
z9f(EIT&ZhDU1(y@J4zM4JU#*jE{Q0^5OBU{cG_m!K#bZ^4R>W<^pp7cmMq&J-x=Do
zLmk{|S)!InwSd{Rw&0C+?gPO`mNc{G%w=8iTA2XvJVT6JE@$$xP?y;#zIVp7a*guq
z#Zhu%SuQ-zZ&^~}IA1tKID{SES+?djQ{{l@9nnX;kEmGLh>4LJ4ci4P;b=8ImT{G?
z=_T9gbQQiAN-d<4@gZ6#!`9uY-tfIeo3>V^>1>#@yk*yJQ{>H~4c=3!NVebUngds=
z`i3;x3T&v<hdp80UU@yEKZfDN?AYne-?2bxlQw8Hb!pDZGOv8~#kbu3)x_cO)Pzk(
z3wpo!mN#E~!`tH%cMp4}U?l*8iE5kPTA~!oIOc4&luB;dNZ&rw7;#%XNoP_TqbG<b
zAjYOz3ohH$YISH8v<fvVHK_q6qTKIxXocT=bGEE=v5uwFWB^OgdZJgGt<o~RZBgI1
z<7>{hr=~j-pe1LUv%Il>xU#Lb2c@ZROS`ui0!}MhGE2^^DWf%mMnqv<S5B8RUI$}%
zIAfBT;5ybh$Qz4v(gAVG>xWk?!@?YTS*K;nLvWN<_66p=MsEVlK?0RzSy3-I>$%><
zmYS2>?9koVv1_V2m15=H-rL?)tOd98j>CSB_nx9H34&5frARih`ew&cn|QO+_hd5c
z-_99;F0NdSu(LK=`B4Ycv@t;CX5~IO#2$l>$&SI6rh}@|mP%;0P1gZ~>XaIkWnDR*
zPyFT=U$Wmh_VH_D%Y}jCW#N}!e#7IJ-}3JGj=NVoronG>q7`bfq>VJ@U@4hQL2cO?
z#RA{0qU&}GZ6;$HgN`U}(|vMTkz#CK`?9=M&5F;8bILwM%2=6#0aaQn-U++ih!2kU
zxg)nsW#DSdx7X9{x1VgAS$B`dUUb#}zh(clmy5L0Hotw9WM|lAzEFx9JJNdw9~k!s
zkR6i_1RPG4t)WOeCp8z!ni+O#O!r_wau6Y9sKvDEoY?nq*E7^v`w~=b62;Zl6sM^}
zBcSd&4>Uhcdwl6YtyECkznxka!LlU8DPxS3Tv_Lpk_t6dG%Dh)qnfLZSr^+AS|?cX
zH>~M)t-*U#JjHuJOhmf-a~lQI7Ui}%&dnk}8zS;-%|8}f4q~O9o{C4>xvX0=9AE=y
zs#lhE;kcam^)J8Ra2FVwuPwvG<+AX_mv8v`i?4Y+9eJ7WiQ{ks);Bv&W!o1EkX6aG
zSaYQJ=J9>@FltO%=L1t12#A%2r3zV_yv2<zXGOE^>E0<(gond{%bIzb3&A;d(`2RV
z@A=kOpUgh{7WA--<Jog?`$$~bSM3>U=lboLtnGe9jEyw3a@krksa8@c%<}?O8N!4Y
z$LNEV!;7@CVOuY1yKeyBm`XL{f+r4xt)VKS?d+7Mec8^gTB~iOy#he59466rCVSsH
zgqp|J8EWQx8#uRBDYSi{S2a*ud2Z?2l$}8r7`#G&;EdTS%^MWwt@K#?j%a{!ZP#M}
zU;_<lrIWessp3q7vIA5Nq_cZ}Zk-=0>Ax=BZ#la^jo+K`3BdX^ey}k%%5YFCVGgS1
ze09*mLO9!?OTcSn@B?4``WqBSPDel)_jjBx3%~l!*R1EvdMUhKUos34?*#A6k;$c)
z*dr~ND-Nx&XlhKH14XSpeAgiA-O7v#9NCXM1_@9qnv}XKU_zo)F!?2O!186KMk|+l
zxIYkwiJyIQwr)CkMMN2=NC-ktU2SJ)r{iy4pgi-QZZ(pw=k9M2Dr~2;mDJQ$^xtUP
z&N9W)+8=sv;_Xl8cU+c*r_+(i2c|d?y=NK#3PJV0BX7QaLz-87g=rWV$7piQ!4q9z
z5Km4E=X2&i{pPP(asdY9AiUbWF#2~e-3rtK=HP@;9GSvUXDFFm6XFWaJK_*66RZ`o
z&9yN&?C7*&CBFesw35_ZGILtFn<guTi2+riG@)mlWqg7JmTJPjsZ=ER7Q}Q=tj)40
zUL0<S)J35#_L-%&HpS)U_qxY;)h~O#Xx*FSxdZhP4euLhRl8_xuk6qk^fn!*h9u>j
znJ)`J|EphfIi0t1f&Kjh$IHymfAtmlypqo=uYdfK*jjcAf;LBaSqeR>O$kt8$qDBL
z)d+}n`q}d5fUvO*LkLExudwDynlqkgWhNS1Mz1+YbHzy^Q25dPJ-cDYU;pYIxmHTC
z&)grZ<0*125xvH=H%P`7y?@6~zW%+JzabldwUw<kFSYB9Xal!9x*%8(H^hOO&eXDU
zKF_SlKznL|QdyRn$I}!0@B%kPf*%NOpeCzV3?UL@WEeezo7mMr{mCCP&nr)FpV&_a
zV(`Qm2r*FCej!+yLh!6*B}k+i4-_Aql}!7tOJdn$N;UAxSvNC}!%0A-80*vtF?wk8
zEn;k`SFau@RSk63iZ4n@nXC|sZ=%o@TqRcvYMk%gA?<k-5khe01Q&Z=Y^$^q4I1+n
z*VbTrlx`y&z1s0dK3vlV6RHXFIyP;$tve%tQ__$*pJ(2^d&gh?`G4Z+?Y9g9K6v&o
zKjZ1L@K?Y33O&#0Y2kHxNz|de7f>t#lu{xJ1aYLavNpy*@E+$vTVFn~FE-h5XB{E~
zj<gn*^M$EKM7#m3RR}6*anx0;ERw=~9P#40n<mz5ptujh-Tj^zEK9NlK{xBWbryAE
zxJ~U>b8Ex5*Ln43jLXYaA5l6uXRNHAaC6S#e89Va6Hm@F^SbhMd_)jl-oIvyk=+p4
z??<L_V0p~cRguyd=D9ZQOzYU0=LNTSgy`@p77%CQvRH6U>o8Wm`9-xIS-y!zYsug;
zLcuw)E`+`(38c}*^*grgeV`~LSE*ic9$W%nD*1c?A9=X{jHRp`FQ!9Ra)AL{5T-aG
zRd}pRn>UjK5O3w9rGQp@USo_@F|lh=nrwsJi>q}5uI)q?HfXWylHIQLM|>s+X-qq<
z#x8aCZ!w~oY0U|USISVC&ol4dzT+?d?5{aJzQa>2VS9MsJg@xvi#M$2ne{UBhxsSm
z5w;|i7NxpET9!@NeXW(G)imOKd$9X+-p;NKIZ~66NVOnQI1Hnme{+Jh7=YFBM-@sD
za#AZpMscnZo#XSD4?N8)$IG`2!E=AN#}C2ERNEOUy=j2<?=?p72a`%0{(SS7oBrfd
zo}q!;eYHAU*M#V&VMNP9NegpYSyJYFIwJ^o`xgY~*$)Fz6^}CXQoCY0U8x#K<D#cZ
zrPP%;3=Go_@``FTlB!se*C*RM%_nWu7HZnAy}Z7|>E3%>^l0=3Ug>o|D$RARqSmFg
zvFa3C*Rmu+QPb`#Ciw5voWC&+cs$WX#t;b3H;`MbQ=(ejCn&0g)<Bpk40gWsWE>7H
zkk<;>d63&4D!p4`Uk=!+9X2rWCXxN@jQ+^%`P2CSZF~Z-K8+u2^bKqS!>eu?`K~!t
zTASIyMW(nT@8ELEeD&oMDV@y?jAzzm<=vY{f(8bSMDfOu$x2@BNb*Hm{dl3O)okO;
z!9|3%tW-Qj!ZZDu_A<P4OyiDNpw88@dqvE{2Xl~}QnFB-pw%N)aXv8(5u9T;3Cp4^
z%NZ9v(=OngQq`nIdlsd49x9?6ZMnao;V93>vusxXo<7eVstR4my(gnSgm}&KOi7vJ
z<;=3qT;?-j*b#U4<aOov?k(@$J+dyDSBIB~3j}f0RG7~TL-4$Q^}y$!f5xyMD1)-B
ziOXDwc$Q^m7)EO;{BB01TFt!oxELs|k~yIBLM=@f0n;dx=5P=_&PS*#N+FA*ijb;q
zEzYgGT%e*XC3Dx9VR5vUze5<&oY9nt&NGBS4uMj!QbM3;My%eWwSTv4vWgOYV0DgM
zDvCpVKsODIo-w*vmiBk)VB~F3q{bwuA6nes@zFJKdvH){u~I$l4p9pfq=D7W5rbzw
z7v6reSXVM9xW}1gS$TZ(mZ5q^4W?=1omF2hmFiks=k}f|)jmToht{>Ms0vxH33oxB
zX))}_iNOWS^k;33kQiW9#e$yF03&bCq?QF20z*_@JWR}sby$<ZF;3p}olLyVK2JKJ
z=Phe|QwQ~oxp6ZTTKjhS&JWD7Y);#4V147LIp<JqT?5yJw9Xt)CsHm*u@a^IH1OH$
zmkfjF5MD5a9S?_l2Ir~kio-L+oiP$~!AZfZvKLSE1AceF`9Mm^z*?aRS~s=@#?FwI
zb8XGW2x18Zi&@M&cG2O(xY==PcAOP5T>#fJjw9k-%j}yrPlyrU=A@=XSrdXII8O{i
z>&BLi;geHiV1_MIYc0$Tbaz!ZEfQ<i(e(D@%G%n})?!)qs~z7g)5m4H-x+OrT1Ps2
zj?`VX%><8l!BvPQ6Ex$fTuzC%Z&uT4a)LX3$z@r1eDjuF@$5CWaxO<*p)87qOcA&&
zkVS}1iwJK`K2$0ygPqaR1FUN<C$(;{yD^x^f6o(WV`%_Y9aSA&ty#p%g{lK}6;I{m
zs{@xM@%VIs2*c=bhOj+jjvAnMYs)r%n=QXTK6EBu1A5&8-=0*;rE-}Qmu2R%t}M%9
zrCn8<Ek~jVk}Iy2y}8yL!b^a<ur4!)IuKlAS4Ynf16q{zm|EFmX`Rv>bIn%Psir$;
z42Mc|_6$^2BF&a*0{no4h>ydwGg(E*r86LD`g4MHHRP%X7_DsA?<f4l&R-`ECm!!j
z&cMvmY}7TT3EQ;U`WW+t?fY80Ew@xgw^v~Ur8^+;zAVnieI_@^R7^)K=R{B;c((z1
z9hvbp8DQn%a!$Pcb}>ee2s~!ib>->J6J8xrESXeQ$qDkRxR}V*v8p$*ejKIoXRDT`
zTUwjlwk9RQmTv97l5alm-~)M023oX3Tz%8VD+cCNfgBaDik3{pm`-DL#9#?!(O}G?
z>k9Xq-b!_!m6v^w9baF-29(@_Bg%VcGS@`1swFm4%A|GWyj+;q#60I#t{!ckYcl4v
z3#LJfux4{0#xdZ-r6plZBW_(2%e=4)9>u0fxB5u2(mlE9zDrB|VuJ3f4yWGYB<I;6
zb<xJU^&4AH8kKzGZaG@>g;sG;g`O2kwcx!c?svG9@k=HK>-c?HO#5ohg<PsJ=}RG}
z+?b|j?@mBltA%ZTk+xsZQ<<CoU0>VhOiFDaSX;IRV2N%=!Kd+eV0;3wK8+u2_?twR
z^kA%&?rdc<ElUzyV2n^ySmwf-Poz9kw2&o{Q(`@xIm83|aWo-8@n|->adEU(tR_-i
zR0;m-oC?O&G5pN8O%KziXjy0PtR;O|3N_D`(bxiXuJq<2XtDb+T?HkCh&o{$93{g#
zFN|Se2!jPgdLs(Z=ERBH4)^Ph`#q=ZhDN==Ge3Xb^}N14XnQbh_q;4CnZ!KLEXf?;
z;s>ISl(ceL&K!?N)>L_Qcx_H5j#3iKWhQvXi~BoXzI<SJKcPXGmz8&K&!l-}In9jI
z#4wIHZyGtyd2k*VBZb0RCzdsk=o~oNoPR};U@Q&I39U+LS_BzmsyML#Acja)%kI%c
zA%{31;+W!Kf%01L?I0KL5pUjI*TQg(;ng$a-AQrI$`nddz;r(F9WR(3u;1kFqzQD}
z+i)#QdT;eU5}*0nb;s6b6V<*c{n3WN1hvjH!F$%Z@}z5H6)D-T<aOnIJmSNFA&$m$
z6EU`9QfjR2J$vUG12(H}2|g1CR;3p-m>1JHdSYd;(=ZaNP}g#`<L=Z}S|b5tMOenV
zP>bTD3DX}ABZ%_$d?bW{7<MfvtJ_=vTN#06J#Sa{YRAu(S>FH0haZA#Q-GWD2-VF|
zE2XloiFI8$U(S@OjBdgSOhdrExMQ~)+4~m^e&R6gaaFlo7MvQmD{26ica~k9@K$a(
zPNb?VO@y|ud2ik)tws3_-h)(W0@<z1Qq0CA1Y_CcWaXh^8h)8pHX)WJ9)<z&25^*i
zuRfS4uEVqBO0Ee<3wlFDnuAR>086!X4{VNYtyUUpj`KA`YYubQ+t6n=y|ol?cepXJ
znjQC|=x3|;k=QX@Z&w;BZMJM+G+T2D5K(+B#_$kjSqe{2t3j5c<l~WbT{s<&4F0f{
z%7_<ghPo>13Z*J*s;KkEehkjajMW&HRkW4x+g{sGLj+^rIVDDGCeXW{QB4M*6`R`i
z_kcY^2?Z?{U^{`k7ZaR>H*enJd?5OvJ$<)OQ!7<$w*3BC`@tpt-XKYzz3A4Kfepa2
zl1gPsg=I}FX(6qt0p;c&j4@dGY%aLA?{>u>*BFE)St))H>&O{=AjXL1N<QTV&Ke-*
zL`*xg7F%z@bVJoNeNfj#^Fz}FYC`P-&PQA@Th^-7hWeJ;1e(2E*=zJY%=Qo3dG4nH
zmkVyrCdQ3v;dFd&n|Tofu$mfve_yk`D@*1oo=+>`mzE%~d+s2HUDI=0{+Q2X*D{oL
zf32IGcw_Na1NOQdyJ{e`_pr`|<D+$ARg~;c<dm3CCwB3`H14*t%9@pul^U$EfpxML
zTubmN+PcHSHgR%I;@I5E6&R9MekP7mp(g86EmaJ_YW_qm))lt&((qbPr^GN2i?fP`
zYT_hC^eV;C#`d{4&qn{r?e5-(ZeOw~nmm(~=&w_K_WdnUQmy;BC2?2gw360>L}OQ4
znOw3e1n-H#^vkl^XNDmn&YO;0qgk&hu`Y>y2y{JLw5@9cv3jTMc5z#&trPI-a9HyT
zFOGug2`g+sl2mQ_oY%Qm*Z?eRWpM)#Vci`?0%13r3;`Z5mg8U6L}?}CrB;*aDAn}G
zqTPdm*4KuR8^9=TV_S8be4p9AY0ciO%WbrD8$o>}X8dXV4aX+{>(ltbM$ch(=cLx2
zk<%t<ZW}+X)NmYxFpfOU3zzv!UK3Rd7oX6Qk##}hju5*zZ=q(;0)BA#=!r3MH%tuC
zv8)R@XD+En*{0plGZ9?}LaD?sF~o?E4pJfKYCu)>Xz0uhD4Evc*#abOERM36(D9F6
z-}4v)zxehoKAP^61yv^IS{j$OJ7~4ahbQmm<n5brOLT1+>f5%)+p*oUU(x}YekfP#
zv1q<<rt}P3cGMDaM8-Hmx{xj>4!a%qhk@5GUvU^mhA>jZ<~ziZX$+QS%b7d%j6@{t
zNMmGi5%Jzi1xjfSR>nCcxE6SpgjX1yl~MSXWmZ*il~RQyzLhJPE|y@WDPFD`M^SnQ
z7*9?K@xsI1Jt&l-NbtBJur4!{1nLoq)S8e|t+b*suY!ws=g?%O48_?&T#BMN#6@!$
zoV3h+t0@IG)<(}7ZFB+m!8qqVpX+yebp0U&3tZ?LVoI?}U2?Y8$+;GW5nc^<q@*mD
z*_ezJE;3s;th6FwPZ)v)u_?E=7s40_gHiLnw`}2gU5qMU&B;nRVd@{TF@0zbdBwXx
zXkEKxF_uy-hJJ0C)YeI<W}%2s0%$gHs?<agxVsxsDqsI*!4KBv*;pb?XgFOnh$hG`
zT&X-a<3n~_8|%R^gl8HrHr})24dk`g=3H9Fx3VtEx>!X~h{h~)T6oy+8R7sb;;P4|
zre{$EsSZb^rpkI*2|JtTF$#F=ta_YRPRBD^4JI@esqNWRXN|6F!6^g=#03Tk4LG{k
z&3Y&$qt%%s9h@;^u+krG>|5`Hu{FJQzM9vC=p*<0muQ8YvgwSJrdL%8eh5fQP|zHC
z=Uhu{Sh2CT^7NcDxm0phiV9Z9Vb4Q9oV1mAi}R+X)ok2|fLA+yKj8hC>=@X@i*uu_
zIA;l<ic%655SBtIl_jqXqwwP&y`)xQIW3?<rE=CIRWn*vT-XzbfK;U}$x54oW84pf
zC=5fTETyrod<)vF6K=}cKodOIqp$<hg7cC6ZbwwdI<M4H5oZlStOPextEU!g;*vxx
zFe?gLkXG)p1gLq%4^~oYAcpxLDWxm8)NR?xjTmtIOFM6GgW}pzU2Z{}+uwU`%V4ax
zlI?44dQfZDb`}qKH!zMnVz2~GZ2{{LJT4f3P*o`v)O+GS;)5WX@h;;1h-7=;S}KBL
zh!bI6kOl&Z3f>t5e$6X2+q#WJoD5bfR4ZDO*}qn5^^_bpTeikCJ@9RyMeiO;wG#8j
z%FC$jE%yhU1foO}z7C#s%~Y!;Zd<5<h_>fjg44I4e*<h9V?<nNnma9tpo&tg+**uj
zr!62iTkcTBcgM$M%NEEtfN(o=96~_UIy;xfPOWuenmj)q9&qAV=R_%$s+Ds+p|xDk
zWFM^zF=s+hrpfW*VUKg*eKA0yB=EkKnVKJ4v^1N%c4lr61f?2dEViz;DCRe}3J23C
zH66B+!Fed1DXmRQsZ`23Ln%x{AXi~kNSdwGKNoV&n;rK_x)t>90El(}dp=Yf2JWtQ
z`fRRm?bsOz?;8s_=fb)cmbtLZl{G1~1k?qnc6JZmTi5me9INtpIgyi6^T@ZSM}GbI
zmT4Rq$AR$e8|HQ8emCN=5=={)+n=>u0Vk&*TFJVaKU<Yi27));K-8FWs3ml)QwnCs
z%~m(8NTlgwnqM8%I@bnonrWq0f<PDtBLZlk<ds}2Qy9Q0&aW-;(E#ykvIEhFwl|n<
zDzxX**2A?9he@;f2A#d>S~8Os_}r9rf3IzP8h<CoCjjfy_`!y?VAZXqw6#{{8T+c%
zy0Hqpx8U{hoUH|XDpakkK0z(`YC*<EE#K<Bg%ATn3{2z1eizwKk@Fdr)np<pL1^vL
zZ+AZt`+euHSy7|`ZJqLgQRO#Xn#RPkY>WX!)lO-4<G?Z}1Hfu`mU=7PCc4-YgndJ9
zpZVLe2Dcf<@8%fV5i9S1%QCM0K<+K*E&JLu8jI151<MYpbEp$@c;k-UI5G}H)5$RH
z42cdOtc#e|N>mH6Cl~P14*IoLa>+O46Fo&<K?_=~b#G`HIfo#%HU@<KS){g77ulE*
z9mH~U;nH@GB9v0_QW&RbPQNCL(7dD8LQ0CG=_Ykdv(`3u=7@Q5<d#wF8Mn?76xqae
zU~`qORQ^8a-?Mn%86V3*>$ae+XSA=`NXzy%K#Pi!@jkGN11Zf-bh&lwlNGe!)ih45
zT+@Ev7?6$_1930_DtHJ13I$TLwYFDS_V@ZsY!>3WzKrQr<zm6kH6z=CW96Fhtu(CB
zKyeiq!qCbwjMdrA6k7RnXJTy-(+9w-4~#y4*CE@yZr_pT|F}Bd*Lg3ckc*PD1p&3)
zkBO|p7y|ooBv%7!oSLmlW1LC@@@g@=bWZTWAyp9dT%0s^j^4~~ZCH*PgRPp7v+qE2
zhmHE&Xp?EzO7Oh|NN?u9-{Or%pL3L2@lpvspw6z*TLgvcTi*s2%I0`)U{>EFZ<rvp
zH(<bC(_x3rinX>(x7l*nbL+3TW`331(uM2u_i>D^a(-jGvT5ejO+TlX23cc_eHb_%
z6KPF=b>N)q3KR=|#F8}Km|Sb%F-`+R6sDbHm4&rJ@nYl&Y|TpBf3>lqZ>|(D_Vf?}
zJ_{w6Mvk(zHTG4zi}pNOz$+G{s%ZlG&~=U*NGa_cRcXnUdi#m?-ts<F`TWZ0a|at?
z>$ks}?RnjpNwv1}qe5xHys8KFO&iN0wYKI7Rvs3_83<MF{lqz3G)aU~GX?^A)52@o
zVuCcyxop5x_ZQk6m6BVRzRuVPC~dB_*+P3j3RhcJsWyjgGPa#iJkFIHHkR!r;yfWn
z>QG^@v)zj`FeSIS>U&(<o9ygy*AhbQzGU5eqHQmzy(HXxI+f?%(hUjoqd${7QwaK*
za09P5hui#Z+k<29cpsS;wbJlne&ABl2hU`x5?Yt=!+zH)$gC0~rA+Nz@Tw*4I;hz3
z+-+}%4dC<*^xA;n4fu51wQF0${$8qTY$Kr8Mh4e3Y|C|D)_S$$zK9(J|IZCOf4i6b
zJyv{c#<XW$Iv`ys+0NpUp=1MYuC)cynBn%k*J>J5mw6^7(|A0s3&(Y~uB9TJF0)Cb
zREu{upR4cDdi5DPm|3-KXM6~zPvx2xm}v7+TQbJ%!C?&?uBTkz|2CaIF-DgYv%#u*
zZPZ{}6*>2EeAniwzehWZwS&oRK3dr}t&!5Lj@8ywVe_$S_t^MeqlL{+5pxqi8aw_p
z{$}G7fc0toaHC@u(hgG~l!|kf$*+yZY~aGoFzlG7NJ?j(9v?}?G*Omd`SPJExo`!8
zv@irX>^y();~(*Gf8Y;)@|u@-J6_)HI6b~$na}*{>$j|HCa0ApRUX%3nkdEajo@hd
zM@Fx=jJ!}=_o^zMDh}1sg7xOlX>Hjqk5kKHOmQNHL`sm>N?udb*LW^IzBy7NTm63@
zdNX$$Z5A3Uz-_ue&Uq6q)$;6sy*jOy(db|1)YL6AyDmy9j#@^9ktsx8{^)aF#LvO)
zKq8?=0)k{g6XZp~d1M&KYe6&<b0#RfoFbXXqVUDHZ+Ln-QdMwXD9#R~c`3~Eg=LxX
zRT=OM>@4&xmkpv*w6Os|#UatR5(sA{L#Z^$jTX?OToz;7yxe(wkVexsaer~Gbg3Hq
zK-9I`e>UgaN}xQ|UC~4<Gt@0`>gRK<#eg%{C~?~g_AHS!%&xI(+Ar#FIfQ>#hpTpX
zMXYq7X!9lI3h<RyUMD1`F*4mvynFk^)A7leFTgAo@CKB;(^fTP3$WXH?n~tJpS<S&
zZqHAC{F*5^#vpwC?Ux+SN8Y?M0e**)N@HTx#%OGTX^h6eGKaR1Y6XoT0d=uy;?>rS
z!5py4E!ExvyNKWwa%yE1rP7g%&Y0@iT@f)Y4u4a=(w+kox|eHU0IhuMO8qxS&jQuk
zPCKeMz|2iRP)cFdm37Ueq@bgTX4lBzM_%5)z)LH+N(*H0afh}N4yv*)E1E}q6^0NQ
zC(q80P?XD)X{2Z|9YOCrrOxDBsO`B<WhUo@C`@i-^pUI?tp%^1irV}(Ah_mc#|NWQ
zdbE{`7zh?!APxi32UrSVdPIWY2O;CgIEG+*TF%;%1cKKV*l#5r28c%ASfbdzx3;d9
za0RbwP5}7G<|tcW-`3ST(9(3*t|gow>u~j|8>ec?o!ZKG`uB3Sb9rzu>>|_M9Z!#6
z@b2jy8tnYJ6w{CLPD#y{JLC7T_Y_9{=#PHH{oRhwU%jNv3wd7o;_(en^M!97pRBcd
zH6~tZj2mrxO+WOUIHqxrlR`OPC~AF-an@8ssvt#htz;@!7^;Z_Q=6#sl91IHMp_GL
zO<SeYrn>}N;D!ivVQTxq7Ag0@?)obR7<G`>J5SMK!S%B<S#JQH4!GGftV?0d%31(5
z29y=&I3f-oqa_7=z~i{AD<Z;lw`W8N^Ao4_!ugv=?tb)&Y1%bdKpI;uP*#ttfx#Ig
zt13v&lv*swP%<?a#@)nj*mLH|&dt*9tgz`2YQ}=qy_K7~)(KLpbvqr0k+C>NS6XMe
z%K3CA4g<Tx9mD}Q38AGEQf>f1Y4Q(Cd!8bAHzH0N14KZLxwfo@oD;RBBn2orAz&qD
zw}M!z?fjQ#w)`=l$$%vxFfGTVl}shdHSt!80fF9C4*MOuyBGY&pMSx(Zy%eKhLA9@
zHz<_7fxX_Dq{r*m_e^&afBL6?pCLFVQBKDbr{kG7mot}?IG)T8PboDJi!HIVfl^}P
z+`DlkxIkKsbz=Z*FfA)WODLAD+MyQHv-8{eXRN;Hh0B_FdU`^JaMfj$)(5XAeR>IM
zPh@b-L2UjV&#$$E%$;%k{^)12CA;lhS=NPRURf@gv?#d<4y!B}{74Xw<_f`l#BaWR
z$MJaPXJ36qDhk7%FCO3VZoRMwX48*Ha;~IW8pExUt7%G=Qd&bHTSK>rombQ3%@&7D
zVaK9JO3Ap^mGt_kJE+zS!3CR;+?eS$KWzsQZ4G-nxYR-Wwtn7kKK2ytol66ky$=Sa
zYxk`KwSj5AHIhinO35pw=Iygo3o7-h^TzdTeg<OoBr4zi(WmiuV|)UzK8+u2Y@18H
zatuoCnZ~PVU(X)&E;(8(Ly;`bBp1sWxR$YT?wUa#$H<gN#uyo5U<i?E7}<|I+<s;B
zp5rAkhn368#MFJNTj!-toC)5*i_&Q8))%)Oz(~tRO4EC>t~{31xMsY%lBiWDTA*V>
z!yuvMZRF}OT{Uj{c%C=tU`fw%YCmB0y%*hz6kYR;*Y9;tS{s12)u-DREmstrXNZnz
zFyKzrIz0Mz-<@ZzrWnaqir`!KF9}G2oa;67(q?eWw1`rxG3rFYW8DF*eeQyxdk$ks
zSSG8BZ(qfRn`RjnxGL>AX__q+ZQzpKyJ1frCyvAw;Ok5cX=1Y7@o}Ar(Qmo-%>5Nm
zls50r4_Kj#%slfIZf)UqZ9l35SGVh}u(hq<?q2&-XLBFCw*ajalD73}b4dH04yvju
z;~3f3k!gtR#*xE*&wdE(qbH{e(FtjFq@2-gEh?96Af{#bI&NqsWVK};?W4>Yvq$pQ
zUYj0Y<(k3nUS6%u(X!w)=Yo&dXGD<g{W}ZfnH_%!=V;&it{wNx=gkYb1x4AwnJyYz
zwQd?EXlFjOXBla1Ez<z%dxW$H*MOsztyL@Y@h;$<WrsV6+}WGmmKEJH|68_K)yf0h
zKyAXWrlq7bRp;;6vABNE+BB`)RvuJZCbeP=0WTi!J=Ilud4qFepkmejWIktY;O%*E
zy|MQd#Hv>hoxVZ!DS~VFCfR-_+%97WBAzYnM`g#?<Dwr~{i~1Lh5^>`GN34_lucl{
zbdp^=7rpb1VQigu_oD3gBZvKt7k39Vj?^)+Qt*T4ysV^@njG}@Mw<;>_u?x9QSAq9
zPrqd&H#-*W;<|37b0~EygB@a|Vgku=2v*wP{?~!l54L5O>!m%@1H856KB3Zlf}8v3
zzi_?Ut1W9aptlxnr8L(Pn{FGvt%G2}T~>o_swk-z&g+Vl716}HCQ?oWI$O50H)MO{
z_4WqZoSHPZ!xIS7z(E{VZlukA{TA5TvdhM}5V~AJYcioqrlM8=1tlZZ8m@>F)3+;m
zo5S~~T^oR*H+I;8d~MzDEZO+np5@v5boQ7&12Jx&^N;pSZu{}3=ckPsd}EiEKrsI-
zgl5O8q?{=&*;3SB^=UET{L<L-;}{`2?hiYr5ZHNP5YOlWMFh9Ztm{H4f=P2UW?`FG
z>EMxoI7bW--x9fI%MNWwa?f>dU`)5+n`gJ*Z7)MiYuY*-U;Vj0Ir`pOKKHF}=kfYu
zdH%X*uJ*0v%Jm}}i_)~8j0v~dutH7+ExMH(YE_nb<#JibSxMCd_}2#9ZC@7gP}M-)
zF1>QS=*_vl-h$ZYnmU-3>8{)Tw5+-I_wJxq2l@J0r&n_lX<)Byt4=!my!YU~GL<)d
zQM-`-fL`}J>9f-8w2AL)^Fy8s<ZoPm`}ftk{LZpzf7rioT|bTA>+uP|`ZRv9p{)~5
z-()LBH#PBHv5tWC8kA{iH1%9bNx68n%zLI4rzogr^!C^ml-I8x7^aE+5E;ZlsswdR
zZeV|SK&J!XIbAM%Q!C<D=By0PTbIOQB~trg&ln@iyb=-&-kbiFbFJGETj`A}3Z)0X
z*1=A5GDeC^h!#fgjKQ%M;waEc^n&xP#=8q6U!63d)pzT&AcKaB^m)`Ch<{*gu4KF3
z+fizzri@reZOU4=`g89b(=af26J#u_34IRZz3IvnH7x?f7+hL#cLQ;B?Ct}>IR-ay
zIw#i0rFH#_cwdoHSl5%a$IgY6?P-Vz1Ci11kl?t;(j4hv`bNY}YPUE8uzb5mk7sa!
zT^LdCS)7t<rc_U&;%ORVC|N7J7>OaW%nL>9))mD|B>I6E2VAf+CLc^3S1<=9=Y*;`
zAy(xi?N2}z*~-amk*o!Kw*l+m1`hA8rC=Y^0b98`PhCej6~$3E-=P?31}D}fZg7FB
zFqg_ALP4zMJ;vlh1Ctv}<ERuq`|N?;y61kk<9@&AVSmrd!^FdW;t(8{%ZcdUa9LL7
z^NG{EvZRY?r_r1?!E`p9Bd3hi+Sn5IWx7+kId8_-cW4%4mIT+FJEg1{smk4cq&nez
zp4lDt#37o9xOme5df$%Q=+H_BeocYy%HZpE3@w}6JLI*o!L)tP*~%44>#7QADdeTG
zg*r<Cs43G+A_ni9E>tU}be;;VC7B*s%?4ii(Gba=wHL=2B0&nS-Z-Ky(_0$biUPwl
zu^a9v-UM}PZQ5jbkN_^29joc3SyxKiXQD=eGi{8um54;O{pmDc7~{ZUx2Ke9fv>l2
zck^o5XjQQ`+^OQ$g7eu*A)O^8Osm|S;JmH|+;ztoBieXpPx<>rBu=j7ZrzT3FC*^j
z^by%{A1W3onu4D)6nj2<9GkPPQ4EnW4(x^jL|Ch`R;VJT$3nH_lty-bqU6e)R$l+;
z1^1=$*~2{#hdVxd@sd!$7v=D9;J94SDxA(|E~h6_vC=f>lqNxD47y<8rE*OmIA>if
zO&`a^p1lgSmF+mO%)aq5jU`#~iWA4n7cY5SE`0s=$o<QEcKg`Q0B_H;8nd=`oy@kM
zZy<_AOW(ES{ue`DZ$^KfuAwzzYy@Z>AhR)=N;{Lm7^cygW=!u|7J_FzTS@k@6iTt@
z{_XjRpMCWuqlYm#-aZ|1-VqKXXcNM2th-W-ZKZ8qta**chY8Vw&zXUd$cRV;DJYZ0
zqBWWpO=HTSv?d6c@H8+;wEewQp3=f4CC1SjOGsHM#WB=k-Pp!Rx>QQZ1Q)>7MwYQ|
zz^IVdN*R>kuIx3%N}+RFsX1Hl+x89Lo)LSHTce5gES2B0Gug!PTQUhd%ErpIvO%q+
zO`I^rk=-z|8zKrUrI38Iv##jIz^W=l=b7h)Tr+p~6ZhWn*~2~iIB-7<<hw*(64SR|
z^X7B{tt^W*0?LKP#MUdOpayDs@$7edVo~&Bg5q=^dxNnBUmK{^k@5<)T2{EI9YoVO
z;Eu}WGP92ZVKBeCv@@Az$6ZGWxB|-CncM-fcA3q7H(TcBfsm^m_j^)>k~7%}NfRj*
z@``z;qIj*4D#yo1)=Ah6pW$$%R5(7(Jifc&!w#pFV_sPRzZ(g|NQ@K1bRbN7QmH^9
zX`)ga8wy11JQNjb0u@K_BhFVGh0*QUxjl;<sglUr=uuUyqQp6hitV)xh!*Fzvg?LA
zQc&}!MG-1%QCu}qf2Y>`1ur;ttzRi<^iV2Vb@SCj1L75{7V+XxGPNXX$+qtn`@+=*
zh;m)OmOQ@&94nTHyJf8dTN&Y}@%ukM0a%~L4>mSmuEl0mjJPAk{w!*=aa1W)Sd-$;
ziK-3>R_>#KW~0ky6k4EWL7)<%35up=<x&doYGJ5_m<usjz<OC`mUU_3&&oKB3}GQ^
zuhbI)0gp#pR`7<>(HsTE^eOCqq~DO*u~uqnZD)<fez?0syk(C{QAmYN8b&d;fHqdr
zZMNuHM(Fl0Jp<G;iMmnYJ4o2r&s=l8{on0KZo>JlN^jxR2tz%eZ7i%YI%=9JRS913
zh*6Wp9DtHDInOwm2tJzFw%mMb1K^CgVfWXsyOx3=#NdfVh%M+qoKe9AdS@xs-g=p)
zuVWuj#WFV{);9d{eB`{I*@uz+P`Kp85MgZ3P-kIjqpe@(#g13mvdy%><(e@v_E_sa
zr*#8Msy3~Gty6(z&-+R}^Mw_jmlD3W#J}U?Dvhd}!yHrzwc9b|QV}tldrFmMDWq&&
zMXNLL$jXD-!|E(ET(orwq@r3_=a~z_oA3=T-4kmiEy*&Ot!qy39SpYB^W7d1q6-A)
zsktEKs&}JuZI@nq@k(=&JEOAy)B<xWP1)`CBr2;|r>C6DGh4NC!m>T*eQX*^zs-)D
z6W!ngddl+Fj;}01dwtm)HWPBCn&wrH&rB1pgI2;Q-oT35GpT|EPuvYulw~zxZs)w|
z?Z6E{bHl*88TJ}0FEHoS+Jn2!WOM;g1NU+58in9sE4FIEf;3H(S}X773#7!}4~%|b
z^m|l<sRzozx|Nl!6Qpd~942n7)Lh#{7#jew=gi7IET|FfGQ--+z^~WU*S1O9J|D`p
z{OJAp_((=-fq}(_R@$4DEs>;D1EbZzt(+@)wS-d2PzH}WX%sAH9l#oxzm|+HnUq#?
zsjQa^Gs1b87$p!Tkd}qC&Uj#O4jlrL3aR9#wbH&|>rh&Xfyu5PX55&dcmuMk*|<|n
zb-69xX{-$g<1i9SA^3=@kaOLflnykj+Re-D8o#%p+cw_zvF$Zi7OHX+z>EFtHU#eW
zTS{p@gvaEs*{<D6YX*EPqZNhY@yO|X;ay4;b$Gqw?d8b3`OG0kR0HREVY<vb?4oIN
zsx}sf1&OBlD9{>GC>YLM?AoFX3~nU&NcIKi2Bd=1V4(kvE%$ArZ2OOC1}@9W<@5ye
z%!~U6hA=S3fQT`#e6jOZot^s?jLqd6eO((fPMZ#oyRz1uvtV0mMw^zVw(|9gvAP4k
zEy<M`b{qR{570+^CYQ!?s?EisLKY>9b?z(GN;9QVlvAptTRR@jj-?^+8)hMu%W7rk
z=kp11j;D9;&~ake9jy64%9L!V>lg!~O3emeWNmBIz!fX$v;SxZAe+M~Ew@A~8@5it
z#t!NwsnqU;Nn2xcB)mAr!Mc%axt4sJhU1k9c4L&huhM@b`fxL@kk+&BZf;|<<@VOK
zDp}39H*citLI%^jYRs(C{C%m`k=4b~)((g<FjZl{PXvN-A$`PoJEQB(UgDginV}_8
ztFz`x0ywKOaNZHc6U8;{GOLcz?fzN<uQ9eoY6FA}*mmM5rEpqi=6S)FLfK7?J?B}L
zh2zr`gLnp7(l=fiSmL}fD6z8ATMf}}$6XK~*ph&%rG52o&z())+2ykNN4Kv0`)k<O
z=hOK8AD;lMPvZw0w|htrd~-{R<Z3=}%H~?+D%7Qr@<PVp#tE=)6g4%i1VEkX-1KZ*
zh@Rjb%j21~R?crS<#Z+=&+P9e{NQ-HJh7yOInAh#Oou&VJTqX<yHp4g2?V31Z+vEp
zvTCcF>FvhN*)<Tzm`+NqnL?%%JG5UuykadYOIb*%Hhmbq9(=0Ql3U!|K%REic7M53
z>#wY;=h=UivWc4mSKWt>^gIiJEo)axHIK7)y0+!Nq-l7?5ZH|ad0CNKh)rk}t;SRk
zCC@8!UJ-ds97cjGR#ecaYF#YX(+(%q*1j7PZ+DR~L?#?Vb>KYJf%ssqW@XF3`5?xu
zYv53IwUsDX7r5XeUq61sWj*tw7q57+&wQR24qk|UunvRXZO%pyAhti%=CyY%+i1+v
zfIAFyv*5REz+7qCOQ8V^*O0#>T30KmXoufE%8_p9fL00!K_L`#q>F0H6q;bSEEy!R
zBqdcxiM}~&HfIGvos~7@YTdF7$jV&K3+XbmoX+UwHGWOxvXD!nq-+cw9~eSuqQ>Cc
zo+1uL7a1f_Qo$9|^Dy)3WfoxFbBw`2F&&*AJSt^l7kPoh;hs5XN?xehfNBHMdTGwp
zj&Dmb;Kq*S19p59iEp4;AJ6Rg#-&zc8G^A-tb#`Hp+Czl<LLrJuy);}vuFFftah!K
z==1J=N0q{9KBLeA<<vf}U7yl|kXzKiJV$W>CqfiQbdkXicpq@C;AAkZ3+(z(Ap~KR
z1{5k}XM4J9<w!XfzJ2$G%jwAL!wdFtV)PT$!Cnu>Dv1Ftm0T3WTZv#p?+n1gI_iba
zCNd{Dgr;v;O4Cg1Ws*(!S}cKa3o|?`M`?d{=FPSyAEA@HZXtx(aRyS=R+eTZmev^1
zg06MVEa%L;C`+b`<ls7!$AkzGugqy>&4p#TkaOYu=wM9@#c>$-?Bc|EK66=*c&!ZH
z5vCC*iDk{!j=cl1f|aHVjZJARVOrU8tL~@bsfY;)w*b~FjYnyby~Bi`r)grW%CKg%
z3hQbKB<b0Gsc1<Ja57uo&STjgT??>on*}sD%l@i2roAES_MUwYaCS!IT<|3#)!SVN
zI4WLYzFf$M5${Z!Pbz%-<{jU>eaoA9M#VFZulV}$E#IC`<Z*|p@OZiqgXhOD_c(2W
z(X~+27?)a&ooU%MyJqzcaRn!i!HrCQBKQIC+nG-TbX(J$u${>^Ct2zSCYMq;UoITq
zydj;Qz<Ku5o+0jV&M|~w^JNh@>fBZiNduax7FauV24XtjXVXftMD0G>nT+<cC*xE=
zueRJ9y~r~Z^5Z{~ds4~HGdtT>QKG165~-qVwN9-R{F(`u%)C~!W9K(JHnj|A`etN~
zmxZ-t-o1MS;G4K(-5m%oZ2fB~q-Cb0mHjXfoujM?!L;m5tlcz5d+DdpcHP?}Qluq_
zrSV@|H&PltULCIZ_C8i(t70n`oa1i4x9e^)7~TA+pe55w*`9NwuS0K2=h(jbwRJcK
zHUz9EwVPS1wl1F4x@ejo38*qiAZ!3uV+$9tYYnCWS8_p}u$v;)54@;Cy%cn<I6op`
zY%X=BO8~gm1>4+ahgW5lWP6{Olt=W!=seMxc)U0ofMuI&Y1eM&XKVT(1^|Qk0&C8^
zJ3gW3Go(bR4-6p^7!8n1g}dDzx8LJJfL30t1)TS-$%27b6gzM2InxG0U7rgBunJl;
z)YO>mw1yLXJ=z~#6X63%r0;-OpT_U~_yk~m8b8=5&TU|gVho6^7C0Q2WO^s<#7WuO
z;jSg~-~H=<%}@WUKj#1NfBZl3n=ik{kuX)%l|UhsWG%aka6Zl)pH9rDvy}{KC5RxJ
zInO5MeXOgM&#e%5k^SzTZ-*!Fm~bk}npexxx`?P{nG%^Sg@L{4J9*a#1(l+(R%KKZ
zTP<}YRgbP2ub#UXds2#wi({pbN^O}-17o}hp@BxFgFS4l=?xSKVh1=)<3lT^@q#fz
zd{uk~2@Qz@x^SgsvYJ4y7;w-5J1Ia7c!j|Q#u}(=rsPObWxtzX9E`4C6%`?iv1u}i
zR2oGa5Ml$K7U+xybvX413o1qpRs|8Gu2<9qU$HjWMU=U=11f;48Q+}kR2+^2h^I<z
zSt22q!f~1L*|V=Z=BE{xjvQZoL#Qu!HSUe3J`TNK7A0E7cU>~RDzZ3oa1<wvP1Z}?
z6z3Hu#>O%6*Mef*ZnDXr;&cS{C;?v`0pU`OK3!WD(l?gK6=Z13^tWe<f1@GZS|R6D
zw{nHK{kfl4QmvF~sw=q`>iLY4_^1Dyf6l-D(U1AR{+s_h-@JK)dXJJRnh^@BiSt}o
zRC#(jbH2>zWhFSrE(T&PkQUaml54iiun4=|9o|QVVIYKs;9)N2$md$9IEKOW*r1Y1
zC6|OFG<_u3p63RlR7JAKB~O-tjH9Fz4#$h#fuSVa>R3u9m2A3etu)ag_|`ee3J|ZE
z#-14$5dy<AAgW>wAdD<jTdQyo-=XUcymiJ$>-ywiT`QVaqy&UOP-)!-jYXJ?HnDTr
z%AUlxvf{y0y*;Cp717M&@r-w#L+BW^1w_3ZrPgb<-*o`N3C??b^hD?IQgM<HQj1G0
zsCQc0zTm9v$?d>9O2FEI*UGvkPRE%#S6p(OPYXI!^7IZy$LIr<Vx>@G!Eb5pxl0Q)
zToH247PCwh5#4mqf}a|YS%4PwVr2?NvVqk)U^+h$Up-!h+!!mjcHDbQ=?&!gvDh(K
z092~7tnEw?IEl^UI4{h#SgDazlCC^mGTvSI`+xsW`G^1H1^@Oh|2>cABOyfVR9F)!
zFP!EhPv^{~79O8wa>|tB${3Ga4rlImI}W=YPw9!Jtjy;XLAbkrK)CR9T=BKwYTIMQ
zF%6L^1}F=)dXn=jrIK<bXuyjngq=NWu8^IwcnpwU1sz7z1yWteDyR$Whl!u;9haO~
zawg!do1Un#`H-#b%$kPS`Aj=E?R#+pMoQ~q=$d^?XZTgjrmMDEbDA+%#9Nu&a$1S0
zGUU;cV>rAwwsFpdB_+;jWhhpHQ!>;nlvPM5M?G@V6~7Dk7?C*OhaGO%QEDrRBvUCC
zP<nwO5Y<ZV(wfLv5+VYl7s8Nm!v*mtoGgsl8ZxBOBpzI;rgK$=>IaGwvNMn{R%M=N
zzJB`_cTTtkPN$Wsju-bQN)4n7P!gIGctLzXSMzTq7<0CyWb97EHwX?=S_r|h-yc|0
zw!~3mjo5ypWU*!~(NH<i0be}<Pbw`L`g?LFn;>?j<YIfZ!d$Ji;W#hkQW!#@rph|6
zxZ{eK6My{2|A2q+U;j7!+rRoVo=!(&FpChQF`wpC`1bh3X})lLJfjM~NWbB3+OfVo
z8}Kj6ah*-Ldw++=bCi?8xV5zNtkQKQt)nLBx>8EhIBaJOts=mh?f~ug)I%zTITdT{
zQDF6^tCrNdq3%a3u}iHRuH~G=R&I}8mGXLDo?iY6O;3@Y@3k}8*U)~oG45-Jrz@>?
zuK|aq!*<oNv#J4+UX37IGRcXf>jT!JgmGeU1D7{%NLfv!J}ZNZtj<dI&hvtZV_BnV
zBhZs&#tiq~<FYrQ^QPIFOSOsy=Paw|GF1vyDqtmqb*rCj%x}$&_2#Hj@J<*cl5^qR
z(-Gz~QWE13*$oq~4in2|=8Io`!5{wg4>;T%wtcgx`9bO(QkxWqSkf=&ginGDZGOa(
zN-ngR>IR6;not!ctR%VpzkM0(T+O*_>)ic~=gGrQ<M)4j0<b=fA8b^yJZNqQhB0R>
zu)D0uG8Mr#-;@jIwQySCPyf+B=70CE{<r*J{?GrGuk)j|YZXvu-K%`>2AJ6d5$AbE
z)Uuan578BJ0`G-$)4hq+y8G<M9U(;2nRbsc!!oscLKv;o4y9-&tFa6qw$4SZ%%ag=
zV-28zs?oAW^NRD1>BU4?9Z@Tq?6B{xbX^O4U9EJPV~4iZ)V2v*+$l|8qGuf}Ft*@U
zOp38uP$@lHZj6SO=GgQQ8ac{JIIEh&s1C1=5h18KkEQ{l?8X5PoKA~n`6M(QDg&Rg
zX1d0J;I|Tt&I0NGv->Gq5G)d!<Jt9+Vk<q;rZLv){H5S!g#x8IkbSR4vcOo31$9|h
zrfO8?^X0-^V7@$(LcmoEN?Z($l8$K|sgk9l6?{sRHBfv))acu#rP!Qn0IPQXpi!I6
z)yc+YsSV7);Z^X(zyz&I#k8_)^{&i~E&=_qILp>9uo{41j}9vq*qzr*)h^YmtfjD|
z%$zR#@Bgbm;h+Aqf6l-CpZ*>1zIlXT_HXPQqs+`p;gk#K%gp66Q%(y(pa#bv;IUTU
zMGI96K1TNY7dRJ)VITyLGlC!muAoIQNVs?mgrMRGwvja3u#SV3wNzs9gyJbKkQHjN
z4CQVbk!;U=UFxRC=S3RpzyctrJa?A$hOpGMW{kCS6)1JhrcIYan;p{-D`69D0sFfc
zh@ESpEQO&4ygGdK+diTylvZ<K-J7&^!)(!@7m9<W7D`TNUO8W8rZ_Mih7C09z*TLX
ziQX^4QLIcP1k;gm&CZeB?A{(5SFP8TwlXhdLY-JQ!=`DHa^`Yb&;-6XQq8!klyru&
zCpbr{mFjd8xh=)sqd42ov=m%p;m{Hzs>Sw-R!%AOQll<DtLEV2a8*#}0V}!mD#j-L
zfE`O;*N@GP!Pq})EH*3Mg2fq-bzCi}WDPVbOHI_gQqq}!^-uph{^(Es5r6)RpK*LT
z;-lXVlw1;*^}^$tIHk<_vXa+KPK6<O(!8)7b}Y-pq6S8%MF~E#8~2n9o=02*Vk;Ia
z8jH|D&8EHQJXx!O$VdZ`S_!FS6E$~AR3T<TgQw!i)in5Uj=>L%V}N?0B*o*XEimn%
z@itd*Q|8k0L(U-*`sqg1OjXmZ>U;Y$Tc$Q(wAD(XRyvg{%e)}P;fr-)GXTrpxs+-F
zylpr@YkpHbwKx)$V_I0W;75=19_IrRM#K-c4_0GU^h8r{x$M=%zGr8CjCPRS3C?FE
zEQnihEcmh`J7pd+A{Ae&F)h?l+(7YAoNi!2N{Oe_nY*PBvSXPu;#VpQwXWo)pfy3A
zk#WKW56NsA@2!DPErjZzbrW?Wo4%!Q<>MxbZ9s^K+8N-gby`(y&eh}8n-8VGUuUv&
z-X5)5T9O!&D!9UBZKYd4*1|C@)I3u&{HuTR-||O){~z(^zxo-+r>9n`+nTN*EV=S@
zKJ$1wv7QoIm9(tP(}hq<)BZI77$-#cU?p#I73r2{53SEn`x$6Tn{HiIa{{H<{w<*;
z2K1^>nA_P*G6rHo$=H4)>Y0WS)<mso+mkz(a%JdTK`iLZsrL6?VybM($*vpQOAtGw
zO3nDU4MDk<q}NiB8f<^?ThYHYVCW4>8h}*__&DIjn6FF8jBPEW_arRAI<E;Qjx`s8
zxGMnHN`QMuZfWux+JJPFR%)lHGw{lugm&GsMF*gz39|buy9{ftQ;KrB%=mS|)y%xE
zrYY%#oD*Mt^))XZURd{K!S;j>Avr4tc1>*GI$~x@u@vG4Mu{=N5VPYRKNR!KHv8;)
zR#%pleP-8Yea@t8cKopz@~81P8lM2HPvZw0&KVWBmb%qE_fno^O=u~2v68ct6Q>!D
zGdvwnoG&NFVPZEOc$#OVR$_!GrqlH1?K_U8ayiYEH8+8A+lRe_=wS+0KB9PPD@hxD
z<Y9kc8e&hVSoUU7tUF)Ei4tO9tqY|jGUxv<dw=#MNs_I1V!xw{nYo8YL}q1HRabAf
zfr|?eG9*A!DCCGxDEz+(zK}8)4u%5)gCQ1Tu-psuURAk7gu9!Ws-DA#qh=A+4L*S_
za;uQqlAT$Rk>T!Urm9E3^DT%p)?x4x)Cxgy@13PkOQmFU^GOMwxH3e<B9>!kB2`*z
zv|7EeeMyD2=R>jLhv9wf9iwN<$rieEPTx~Eqhi#H<G{204cWX-jI|W5*O}BJK?8$k
z1V|RJqFkQNl$+cSpW!(HN<gwpN17S=IC6U&c)C8J?+W9aw+=eg&ep;=q-ILfLNiZ`
zQH<Iv4fiw))DVdM9R?aAeo##zlEV>M7HDoIiMbS0P%;rSShVo@^qTv4;AuWXP_EaR
zr*!72Jh3}jN%q4>NJyv<ANxt&r5t4~lr<s23x9h-KW}UudNv#pVZbfrM@LpWC=+%E
z$?f0oTk8Jb+t2;pbcvjRwkI6LehnIMS=^jANB2Lu%rkR=Yjv<;I*i<mCyx1s>-vaU
zb8nr}Xlv!e;}h?%3-fiMtc_Ouat^@(u0Ul-LbJqL)I%4la!3QiXE(h6xKi2!T}41_
zk=P;>yaU_|ETiiT&7Pmps8nmLtx~R)yb5LV`K3M0&^Gm|xTGnk=rXIOPI0%F9(i8y
zMnd;;c$QAPB#B;N`gQE;I+yd}vdm!HmF*xZBT-^6RJX#}*%Oi64-e5h=^98qEW(@F
z#)l6NtZShSnP!3O9Qpl+cl`Tb|B^8!#u)kb!y~G29Palu$bEB~djd9>E!rjl6n}z)
zbRRRRtU6<|AfW&Q^qCVgq?nOXJfwrtD^c%O69hG-tP5Wp@3}b*e7HOy;vQ<(W#z-u
zdoP|tL}WlhBE|&Ph(kh)gK@ETP^Np<wOZLc?7;fLS33W*g9#)D)ft^x@|}4MKd5<(
zlS#D+t@XV!I`~@!E_3nN`WR@coUbcYowYliPJHp@=iKHS9&A9Dm132rW?a_BI<I{E
zc;@50Fqf^tWDtxSX<(}}B6|ZXs*!Wx?)Hvmf)Hs0R&{Aih>62+!xRIuT>G=;3>qYn
zED<bnEh`}id8nugtteH6IS7k3mz-w@xm2q&oTJa>LyUYhPrK}HD*$Zi!1S6Tb<fVV
z10O-3h2mjHxx87of!R?V9A}sL2ZRa1uWc!X>-9<?k{P=9m-G&kVO9Em9$0E&uF80K
z&#Upo)5njfR&r`gOW`h0470!*C@rw;Ah5b60zFMN_)SkSJ2)gOZ0nRd0CH)?eb4$D
zD1CP}l$4Aq8B$&H?dw$NtY~FM8Dr!$O#JlC=iD*yyUXvfxv*Sey*>cOKsmqi_4+Tl
zyE$=pbL)*WBIM4<s0pH?$-S^zYgDaN3rO@sd+C8HLmmlS(CVQ%!J#Q{6xaY3dmq5-
z3O~{QaOU#nUAJzGSYQE)!sEQMwups9ZOY?QanOjscslX+^EaH*kp~+v`rHb^!N;dF
zPv?cprC_xKG2Q1;D4j73M#R*^N0J2YZcd!z!n|r{IC~8#5=g=QW$Rqs`%(fy5=bIM
z_fTq`d6lBGgjV(Xwglth{D`7(>kJ>N5%R#{aD#F6=7_0>fP~&$a>%U9{n<V7w(07E
zMRwW0IL(=t%;{$|Q*k?!-UOANuv}}u_(E##z2=fvm&&FPdt=9+v5JFmvleDE+F)og
zYFYNCUbD(tE7$VCQ;EDg3-vJYYM9)s3L{2H1eks84k7e=<jujR#$pv!qlgC>1l!(f
z=(TK@&o?8Q*Tx|POqFXbygE*Nc5}}Zj4aA>ePUe~G#E{kr4{B@xYmU!3n^~F9`r^(
z9W?f)WvR{Os=24X2Cp$zD=2OEYI9oN=3EzhjU9Sj)j{Z9h;9Jni$fB72kz7OFJgQG
zus)4H+1NpSdf^u~gGEorsp8T>aGiK-Pz!wg_>r%_`5H5I!}fl13mxb|gn3<fT2@M_
zo>0Xz`@v=6MhZd>fsDugntGC45|5=$ITD&vH-h#KtGcg8ctQKO4zMWBIuTWu(x@O>
z{Xib1GX%B|ZkTj?=B_L94$9Uc*nC>#J4MshVc0r?zCK&mgu6GBN-x5;&ct?|tzT!J
zyA@lPeK&J$?hWMLUF?Rr*yi=r(?6v5CA<7ndhDKw`_7mM%j!Iwrp-NO%-G^^yIscK
z$&%d@;d!I+IzbTWdinlUPd5sVM23yx-}?8vOG2Onq5>%-#+WG~IisTb=WNzGyKU>7
zMMR=AtAhKb_$KQf1**+~&KK{`_6@PKHaf82n*X@RKtH$W{jjnf^tXHXheh-D416EP
z=H0g677_Q@FqfHCp_`N%tvb7-R5+h6JUl#rK?-pPkhY&$TVtLVE?4)=@iag7=UBwL
zlq4WQsMSGL%-_Q>j-;4&un~IS&b*k~bp_9OA{&mhv2fHEBbTx@^CZI9zg@QHQFr6;
zE&kcfzx(sEAA(-K+{<U37YC^4?`zv@^e=BdTra2oX862)WorU7?P+>jN3|yvMsa4;
z24F4A%6y&uFld468aXdBPxBQWMr0V7*UG%O%$l$VqO*UpUfl17_|`v92e{1ze%+ts
z`G9EsMoJW<2|JUqd&xbsN&R}}=xKJBaOozk&X!!tOilws=mNgpB#>ZUbO)bzX67JS
z_u$*)DBGGAI%{K_&$iFi@LY9&XCCip^?pOX&+{1A4;5GkusUX<=F-zBT{ha3vKj@W
zAk6E^yv)6E6ZhzIH3w8{ZLDiyS&D<n{bw3<Zzt;@LqD2``v<0!NHHP7)5jF)p&p*r
z7YKe(?tU&&#LcJGB|Ek6i^;@|=vxDGX5OqT8kDgoZCIc0RW>gc*}t~W<?UYf&k&5x
z)buaQ{^!o@*nC`s7juKo$n4v|w%H;$8|!&5Za7OK#%&+AeT&)HAhjpqhBT7HfLv$L
zK#YxKfuTTHT((}Cu{QI$-1p*rZQRy|-jv$rFSc_MT{hb`+X3nIhA&dm4b{<3#og9G
zwnmR%(-b0ybl^rN;yKWoLW>k_d^|spa^&XbcDsVRnYX-<7H#{o_FpmjHQ&N0!sfwc
z&+}dOWA2N&{9=v#A<bpAt$w1f&Hiy4&g=>O9yV2KRkx22^Lad92___o6ou`xx7Jve
z)f4rac_4y3*O$9fRo@48U$-EP!@yd#%l9@<IDdXZ=m2u-&SQyfV)$BavVWc=p*Mqc
zKPN2fiqc3m5`qwWnsL~g8bNmHyO^+*aJCi6_Tr}VyhraFt=aSRe6fGM9QNW#q0e*G
zXKJ!-^LB$#`&@BGW`DwZQ+KJ#(!6UGVsNz|pxgWsx5u#RhWcXyhLP<#^(KA4;hV#|
z2)m{o`uyKmO8(lmb<TK(a-F68-2C5cAO!c2owAUFv0l$!$8St-p*Qj>mR{_a+t1nV
zLBGD*!E{qULs$pg_Ikh98JodBxL4@Q)6woR+Ut=&xIz2%*<SP0_~DOF0M@7RCmb%r
z8nTd5B<D=KtnQ!GyEM%jQG{t2XsxlP#{cyn{9pJ-|NVbMTW-iXFyx4PL|Bh?g{RAv
zk56Z=^OaJncL4;3U<}bnV&s$<azf&O5nynG!}G%jh9UC#m!A{fePo?0V@zamS%d*a
z+%qE#U@g!Hm?Rp?Qq<*X7+PkuEwmDm9Jm_?h{m$a%*(=gxljYt?D5|XL90`-CG{hm
z#GRP}-N>{%#H+Oa-@O<Rqlux?LA0&F*bl*-S`1sD&%PP&;8F+RQt-5r6r4em2G_~g
zwe;jT;k+y$iPLE!4~;xc%ys3`RvzA;$pOB2{hGs&Nq7-a>&7;Fj&zPOAsDsQ7eIm+
z@``tc<`gMGP}B>nq6`8l3ZZwB!`3ZR2olM|NNvicE|kj^5hWi^WHRXx(G}WiJ(VpH
zl8fEOH1v}j)ar>w$2MRViJcCw)o(*II?(B!Xhn=@kw``*5okoKU_F9eGp5lQemt91
z`dwr%&i|-N`s+HhI>lWerN}rY&U0ZcUXXU$pR<UPcLPEB2mjz7@ZbHT{|9x6&h!um
zmPbHq;pu$k@#)Ot`AR8W(kYOma2&-wS%8q-CoKmDvqhV8-%?~428L;(UKcP=lM5Cw
z*Oo!(tf(!m4nboFH2Wd4w3VhajTLK=kOOZHBZzV-SFWY-w9XD=i^~`q4!{Kq9tAEP
zT(Rbow%|!WNj%=&r1is|bV-0vq8GH!2VK|%2+k#lFA6v%gh))8g~d}AlON<<er${}
zlA|B=Roxrp@!`zl)0wjxi@`%PzIiya2;?Bt5LlWr*T!0l8}e3{{DL1&$POk%5%)IQ
z&V>E>tzt-MR1>NNmJa?$x;c0?jtD~tG;L_bfu`msE{PEtGPP8ebw=kx2yi#vky7Mo
z{>WiCAtX}j2bb;@7GvxZl!UYEn)@B9?oy$3b(SpPhxIMNtj|@Ij1mcEFXnOU^Emu)
z<}pC2M1w<sJOqX@@i@D=vnKy2A^Ec#)Uk?1jsM$!^WXCS`XB!%t{)e22rTL>wfg~*
zD$8}Hl%8OS??X96h7cH&Fy+Ky7zi3X$YpAzA=7FBjKkrE;VgvgyS#283nsM1_tr2Z
z(qM!L<(jb;kuY{(1!{93HbiFvp6v+>jW6RpAw*Ua7HM3mE=7vQB1RFT)`s*sw;iCe
z3RE~pp+zHCNVp^x0COfpl*lze!w=mSoY^M9+27FusZ%)>m&xk}<ah8kMxT!%_~CR)
znG^zQjioGH=b3yspeb`nk);|V5ffnzjYSHJHokp0v$V$BQ)23L8N->-F$Tskuq>;8
z-i={xZjL?-iKSjB=Y=wY#y|_Pv&x*&v;sH4fO~Y+*7h{#4J1xOW-SYsRtPncV&-*v
z!^h=4-#q??>3HBx_)!N2eQhL|2QozQG-y+bHfpWZDi92jgtkrE%=4kR`{y)`APEi8
zS#VbImX~6hyy5r*nafQJYLJq%31f;ZYok_0LPTQhDc9<LerBvR{_p?rAM!u^(|=04
zDmg^v+OP(>2Ii)$RaqA|B_CoY1>umDDMgYQLrRQ8rjbA*L%@y056+}C`}o|P-PnIY
zh*9=ueoGTIoJD5Z=ZwKXqH1NWiq;jC%H{e*v&tz=3@LJ#4><o%SwvZ18rRoG3`Xjy
zdm95Eo`r;-_}S~oe!oL!`P9(C!H%N@=<iit{Isx@4{kS7=&OdO0!jz|0^Qfr+^4G5
z%6eJ(?Yj>=t}ElKTW*gB?(>OSE6cJFTVk-44{JeD-aTBf+W2C+?VyZ@*9_66;M!WZ
zP0(F#n^K~*h51^jdBK8qKAM~4)4{@slM1X`vPj>Pw?;I8t{&_$gpr&hhj^gm%FWG0
z2*$EpnPMah1V;Di6m@25jJu^ntrZE%pq^a2uHuQ#c!R05)%^qk(~R3zY~C^TxiQ#T
zv3pj=3m|I;3iSn&_i6l~$0q>m)A*B(jXe<hF0g5|c>$zvMb9)0tfde{d3rjt)|u1f
zGOdlp<6h>1Mqw$1>%8oA(jc-~JUnaZJ9{XGou7qiW4>Mp!0qwK)8iQ}TYPIkLUh)T
z$TK@hy$GyUkse9iP*{{w7n&|uaZ_0v2f&RBORX$rfgw42iQa|jT;^?)e6f!{cYpQ{
zp}nZ7whqiz{Pt3NFUD-|dTool$ji=ac>X)<TY677@s3L0xXfs^C&sCPxMZp{FIunb
z!c$!c%Zf%}U5oo?1lu?Nt;6DPV`~xe+!ft=S-q2^_H6#S1=wrv&>($t(#}HL!7pbD
zJhqip8>%Zg3a^4E9hGH)mOxYY7X!p)icXid-njCU+Exhc5-;_8wdw8Eo*!JZKWOf4
zH%Y$p+aEQGzH=+TCnJa-;)%WU$Mf^|JeOTkHVy-8^F*Ta(-Wnv+)S_VveEm2AwV-_
zu7%RnJM{DdMZ~+Teg66L?axVyU9x8l)e71OVqU-pPx#U9l|{(@+ywJqoe^oKRP_QY
zTB9}xZi*HvE0jc%h}nqYoy^iIOI?uYUJ+a4(|z5nn+E%}+MB*FuHVMQ>?;s2y)^c9
zpnvx5M@jb;*~WI*+`8E>$8(q5*t)@PPvmlf*31*<)R=3fm=RKTIjuxYoVB3pJX71(
z8QDFjLI^}$l3@F)w0Us!*W3<^+uY`Pop+F$UCyzsdHRAOa%n5AR%RM`OIuQ-u7w;F
z&Ce^xmLeE>l9}|R)U9B4iK8DRHkR)8dd|qQ=lg~2b9oPSmFI5#%isTKe2?ZaI3OAN
z!8e5GIric^Fr1;9bD~r?B|o2^XswcMbpJsN6D3Dy6wSp$SkyYGDlQe;nF99wOauXm
zLKI`2uV|}`<3LLDeqRMV+1KzrEf^uRE@KwVs<RCCXIH3oWiD5w6$phZ1&PiMi%|fj
zQJHN44bILpbB{N55O|mG8~*GE>FWmjg|dE9vX|HF<@Uz&^me97KOS!<ukAu?fVK4(
z<axi|pG?&~#jLI?Rg~JFb5SESch6e!_q$g2t67Qzk>kd;Vlz8;_Ji(gt~}q9UN@<2
zOiIODZsuVO=D<<wK)%oA?gJ>>+Tm;aQ(ahBVF|d;>2}SUuQNp}gKbjY?m@P_zYyJv
zjcvYm#<bAAbo{+PzgNEd>*ZV)evorndgCI+_??>0wi5dVYj#M9+8UzD^>U`x$}t`H
zpbew1iM7#+ga2Dy={}`h-iqyyZ`>9RYhaymv^7qV7vY<R%TxuC<hMF#m-cDjueUG+
zamjM49^6t_nib|+FqE}ce;*m#^Qn3*-o2cBU$(CD=)9sG<#^s(`=w%gJZ#%hx*fnv
z^8E9>c;W2YpDpRNz#e8KJt#%?S7c*hyZ@Ef8>M(d<)%vw593IUk$|yIP&IHhD6LY;
zLM>^p-4WfTbF#mRf5vUi4TSDBtQ*S4)^U<;zVzT3`m^$q-DGBzR=F-$RE=atid&=I
zuSbjygw;|VM2pJizPtIDZER0p-mEjY{JpB`4n{xk#m|+6udn+X+&<#-&D%frivp_u
zo$%c+eHuU9@d?2CH2!2GbRa=+|B1P*)M9(*!a8!2Qs&jGJDQz&cv@L!kJX-r>V}bZ
zL8W?^H41fzJX~kqeR!aC1JaO$B*GYaB1Tn8H9}Bw)}5NWtScYhe`Gkl;-w2dNB{u<
z07*naRL5^W<NWY|tt)Gp5gOBU=ZD+w;o*#dLQ`cPJj>4N#k)khve?3ED=AG3mU(kJ
zA}FP<%*%zR^CMF{k@M)X0qg9duz@$7?u!F?o5}U2*Dv~P5N)z`cKW8<^?`BU+-y(w
z*ccrlgzhO4i763dhSF)nTl!f~?5f?wT~wJbGuDh!noA((D<98KeEsn~@2*!qE;F}5
zxXFq0<&0P*Oqr&-9|*h<a*rP5xm4IdvKSx*VYL;saHuOODjMxOg@#ISgI^Pd=7;hE
zSaFllZ_AaU#vxR0!@yV5gc>}2{WUq?lZRJ3I1pG6X~gJ02G+cos#bSXPS-R*DUD*v
z8UriF+Nu}+Jt9~|vEXT5&N*@E!rt@S_T85o?)PxyJOa>~0s5l&*gWofpg~UD9dA%B
zJYAnymz7cqhv5#1IItBOV-gk`@5;=zR;oRB1B1Aq!VtS04m_Lz?<}gpyj&p~rHlj`
zha|M<#p|+OiDH~?56=g;=KA2)!^cax5)3kxwalECGYbnv3Mpm=5^5(zT~;wKS02tE
znT|*DFz#tWr8+xs(=3>|>_$`Y=c*Uc?)>DSSu&RZ%XYBZ(Jt8|vP~SAiNOng!ZwfN
zvxmfsLs<y2%d=5WP!+llo#I{Blu%W!Yo*1^c)Vqh<S9vUf;drIgARNt&vF43$SHva
zu3>S&Sq%x%<zMm)@K;;GQh$GKFI-m)6=PBljtDp_z(Egxy<c7*C`;vV95@XFi#0+q
z*2k52e`Fdx@o_U_2O}mo&o!y;xo6KHT1*iPE|-~kDS#*Nm0Ej&D>Q5FeYG)F4Z7z0
z2RDxaQuaMLDInV+rymupCk1OIg~0vkh|DX?Wu`7O=URAk8X$?A3<u+O43xF;^-{Q2
zH*U@`k%Fg3rl4qBX_|YufHSU#3`XVsw_jsv;O_p0@xvLBLJ}nfnAbBFGvhEJxe?+8
zDG4i)S_`!r#b#0v1`Axu%=`HP?F#OeD=`U|xiBVYc2=EvsE-^N8OSb|Ylc-n^u!n8
zRwW881hh4GxyM1GCL4g#Q-VX^ieI285;lWPX9|Z9HxGmzScoAa`ohP>ln}*3Yt+5x
zv^HvKENkQ8@{U=B*$h)7&lS<epvodZ3S7!UrSjq2Xk%i^haKPuUH)xN(c)>o5+Kv(
z<Sqwn7&8J4t;=qu(p0EK&M0$GPWW}#9_%8~pPy?jynAxP_Pe_qZpX-M7n&-AzM|F0
z(|TqYqGzMY0YnI)L5xVuzHW3TWm6^Ri6DXNbw-;{$Y$PD(+)r@ib(SS!X04@ec$H?
zHJ5Y9`ySA=Icv5UtwOCne-IBdIUNUVU0CKTT1#(W99-HhbQV!X*G4OCuRUXk<QT{y
zkhUg;Dy=AawB9sYT>{HZ_tJ8vZtM+}AW-W<M3KWt2nMn2X6nuS{Vc7zly1ostz65(
zye_P@(7NHQEPz3n9OPQ&GfV#-w|`LT3ufbUH^}#~t@hl2DwPhfsdYEEX8>#e^A6fM
zD?0TMB5-R1lV>1!3zP_AM6vxk&^_;#Qn;3dS{hc3!!U4jn7BXOxO{WS#8ep$jbb0U
z&NJnDW=WNhGZI&%d%X$WmLo>2bZ@&9NMtgRS{)>k-WcATIz_hH95lS}dQ@|-*?tx|
zUoX^UB~}mU`SF|27$j1w*T6%_(2RASDdR+rxwG2DlP3EbC;>w3z?7+{CrY5sN-d3A
zI+z(;*4@N?bc2A)`}>2k*E^fonf~7$@VD=N?bG-#V|)UzK8-)w2qATQ0()j@1b^tB
zeUkj-VAT=)F+)#Prdgv0Re6agUbL*WQoCnXj@_iY8MfPH3EI1>&AfwP=H6A-8m(2P
zU>t_Xkc6Cs)9FBpgLnDG-1rk`m8_-Ew4$|3*lJG@(Z-5W5bD}cgt}HIe*`z2Kc?(L
zpuREe9YT2Sq->o>+n8g$GfQ8;MTiNu1Vj-}AT-~Tp6R~*u-qMCp4nGM_ujG{$lF*~
z*3BGEJ+)2n&dGe9scS_;q_)Px`wzVP_`t*E%y}&=wQ(FX8Y3ZOr}x`4a0PT}ho^dl
zlpOdeOE2EH&Mi1EB~rz&x~LJvi&5{4bzch+|J~f|lHLVAN+yQD-R%uG<H#_KG?WkL
zE4Mc{<k34<o38>Uow;G%^T#`c4$L%&0W&CT!AfH=LDYTPws!SqTx(m)%`GPV@X}Lb
zzGF(TpN{+Z!_LuP;}KHw=SS5&8xo}IX<}PfesA`f$DdDwP|SL2<o4`spsIhZ>hm+~
z=GxNP8{qfdJ+@*r?>t6(UJJBQmIZ5t6dKdusbjbIN5)~GC7%reB@O~1axK+Of!n+{
zW=mVBZN;Jkm$fOZm1UX9IT6BvDQE7EH(o4mo)g<)6`#A`@$Tvc?{-EK90ZCI2+`S)
z^x`>~VJbVQYn?{#A~x+FAHAcjZcN^~R~`L2`kLYXZA~47E`7ZV!${WX;H)U)bf8J)
z&?2#csZg4+)JD-4YYrP5ag!wY$zFPwJp^Ylx>umHVK(S!Ph_-iEZAj|5W5LCSOINH
zEh{<<5QJk)SPHzpz2!I##5|z2QEc74UY;Mct$3v$ej^%OrfK~;N^nh_na8ZV8UD6L
z`t`LSJ-w{|!0*gseo*rm`dl{OM^g->9EfElw7OURF6r@8e~d6C9Qw81wRby2|Al}2
z+S7b@&=^0IIGZ`{bJ@c|{O2}RE|*8*bmVw*LyFFzK8}$gM{e#Wa!6PaBuB#G=*%lC
ztn*6g;V%uCgRrVg#<mpU)s)cuu$x05rHNx6`D|1|iAa40Hnu%C$ul!+V+zvk3w(&1
zA_oZB_flv3^aY@MWBR`A`&6^wa<33QnNz~P_ibK`z8-C!T7IZ@@TU5lZpwA3T$(|4
zgXq)g$XW_EUl>EAig}V|2wav*7D2}7;S&*vonekVr;Or0JT86etE+(arjuIO{kSX`
zM)h<xgchTlLKj?m)T%RTFm7Wc=D^+Q#Bm(Z7$Id2hZ8X+XVB8uhko#vP9+VUQTfaq
zx1DYEG97iBqrQchzrW73?GQ|Vp#5jKq4p1BF2|VIR0b&~h9OhhOluWSeN$T19YBo&
zL-6Z_J+F7(WK<_osW+E5!+e+W_S${33fL1n-Io*tYp^aewN^s(22@VMkTNMoPSZ$^
ziE&KiD5PX=?NF6<Sy*eM($D~_RDX}b_L|Ls5P9AoRS}Ei7)j#c1KU0u`aYXt>@}V5
z!wq8#*KvlP?4D2(;hE7yua~9oG0$x2jp@8?J`%zX0C}C=jp6C<b%z4|K6!)17^jgq
z32$G&=5%x7?ddJGI^&7Q8A}UyOC_hk<>LpRm-4Kva7%>ey?S3W*p~2KUj$!Yo)>od
zeCtDcFw2Xve|JBF4LLG|$nkh!I!tc1FOX9v#Y7CgHmWvMDs=hewmEO?bAt2)T<hnU
z&dS#9`RcU{bfDk%XVmxfotN9+<8$D`&LDg7DZb-B`f2=N$0q>m)A*B(D48aTS=~YO
z?#ta9j=kV(hOUih$~1ZbUCcX6D$kMv!%ar#;(i;o!f6V<%j0qA=2j>ufu_zHvF^!Z
z3fd~Q7NQt8V`NM)WVnBGONfbwx*#Y~DiV#Dh4r$q6lGk^Bi1Xds?4pi2B<->*^MvP
zwUP~F5QY%B9VeQ2EcSK)@bm4qA@xF*9yQ!04V$TKGb2V%`;vaJ4B~7HQ|TLkqMg;y
zaBr^EfxOK{M?1SBc-#@9eRJz+KdOG=LN6e%mn-viA>ZAzw#IM1{g(G1pLqZDz(s|{
zggT8_%7l~&G4*1*16bn8HZdd;M$7|Nt8eK+$SD$Zqndl6StQa#kmk&}6cj9ByDy3K
zq%ft@Iw%RR8MzA+Z(n`JbQl??iFKKI|9BzhJMP^b7dlXd1qWAygCa&i)B&v48fJqV
zV6O!&jmZL``2oz+bp(^v6J6+)O@a<8M%h5hXQ>1DM|hT1vKK~w-A9Z*k7<>d0@ES0
z76*cD_apIN*1Y?B90Hf76l3p($K619GYr=KAQA}9xax-%Y?E+0a3lSY=$G5D)>!5%
zw3Q^4;|L){Ucb2|rp(j25Ui0*2}2|fk@NY=s)g&idJz~Sr4_0!XpBe<tQ{b_T+U46
z$Z?!F<V+s#xhxAyDc>oU9MAM;)4?0r3$h)C7Vn$Vy-GUd-V+D6dEDi6x`EM+M%blv
zX4XR%T;97&qzykbswd02OrbzitQm2fJjL<ygsE~&8HtT*D{U^=+L&7-u8pO7cOkIJ
ziJiHcQbKeCp}as%eSU`!dO$??e-qnfzS1Q~4qzpO=xmK3u(;uWDJxCgkLNZGi~?W1
z{)}lFNCZ|^3SLaR6kS2RUygMEAh@JK)um+nAt!{r(=P;aNS-VyT?U0Z^D0Do%b1Nz
zY3i$viTz;aF?fV~YaVnVIWgUgoYzJub%$I{We2KK2e9U5R222o6CnyQ_%&)Olq&n-
zL;8VYyEjO3nNU-PF89sRC4x`q4@_d*rnlq-V}je8%;7L_|Je;f<ovWCLna&tj7KV0
zmdBOFDz#J+M#lGHLPTq$HDxwu1+Aqpgy@XJF)|%T=BL7Xt?tLO-Ai$hX<Ns_j+J(^
z@jkCB!jK&N!+h@2)c52n9W<r01RH9hGt)YN&9=@#&@RX6d(>77`21}S&{bz*J<V4h
zYvT})3^{UtcgK9a((2-VdI3sGNDMs9Gf4zqMPiB!IS^7L2KTva>g>mDpANfeaz9Ll
z>?ZE&9&<qh7Ar_4l1Y+jDfvZP)R}Cxc+*jY*Ew+)BDeRiNYj8?K?dP;dq>WKcCR_<
z5<=m5?rq@=o>1r0(t#Fnwr{WL{CRUwW|OpTfez1e*^}d9Uz>g?b2*08S@V@VWRAy?
zWoeYP_BsZ)fe3y*V}LOVt14BUQ0RD_r!FhDhdTrjLUjMNrn*1>8+a9B2Xx(}4pn8o
zT&SgVuiMC&BR9v9F=tMPqlYG?<czKqD7nzqmF45i+?-k!6c%$<QOFrquiInHNDBUX
zU5XcTWK4-gJSTI*Ok#hYQ%pWD?KzpxpK-b@*ux22Ch7Zw^gty)mpCBxGM7^NCD;~%
ztqHpQT*v(V=XnFj3qkAF6lzK_<#2N#=8+$L{yBHIcf5W5C2J`>ozIXTu=GT2Fb<jJ
z;kO+x+d#Budy{8b-m`?%5ueTL&AqpfZ4ILrY5sji*a0jtLwYS1`+c?;NNHe7nbX~g
z<1p-Mkui@uhy{jP^`<fbk+@$om*<<Kgu9JO2N3CY410^Se}8xeqmj_RFFY#=r8CN`
zgID{D**=H=&}QM&_+gGu0M@7RCmUiyv`%^r5tHl(do!-*Mm0l6*$rEx2vZJdmu$CI
z@QB=g2&qsmE9Zv?mQs3pl5bwBP)%8+62?sG9XRRWhzX>W38}~Yno{fPk+ey;mJ6)}
zzMaoRj0q)4Bodg$ktoLH@<cG>b{v@k<Z$SzWkSf({kIHMNg`d&7b$%+>3`*kL%|R2
z+T{SIwzD|cep5Ffcdv2T7jErfSny_sH!{tO!0<v&NS7jo5Z&w;Ks^D>x2^6ABQA#n
zmrT}LxL#+jmxb4_-x9>VZx&PLTCp&Z(*SWqC9^ijO&Q}R?V)>jG=F_DFTz{fyR6R~
z{(wj$NMUNl8FtZ#7>P)==m!8CxNy0G8_$~vMU2BVaX;KJ9!K(!`vJhPYeh??Y3U$C
z#9Hh{g?D!>I8ajUSyHisQszZYiO`~ZukApiv@Yp<;T6;+7_xbF?Gs^}4;w&fyOHVl
zVTd{4skJek4!nKy+Pm5a%Vl*F&e9w#9i$t-1}Z%YyRM!vrp=jk%~Q1&Q#T{W9kPJ#
zW#f#BRE^w>LA{{aQshisj(90EBF3xx8|Gzo31OYFw(#M4L0TiV#$lWo^GL`>Y9M7L
zhRE&x4TJ;A9dm0etudDwP{tvXb0T%%Ed^(Wt{W4@JkC0HsiXDSP4{|o5EeCOG8kBC
zRP~s1lBW%6r|fU@(WA7x{;9)UvK8mSpFHcuYY^LiZ*2ZHX@g&=n*nmj)EKy`Ggn0g
zsmdV2BEpJsUKW}*o|XaC$W6$+^q7jWd3e~)=yZeN-bpp<2UlNnl6JP1si#&Fs6+~$
zD(ADr*#R+nfn8`E4hL?-KpF)}iq+sU?zvIs!aUDRS-fLz+2`BFh}(?Pu-!Mu4-O8T
zRCQBmKj_lsE<zD!fo%1zu@%aj#F`(>JoeEKRSMI1;O*PbXe}W2NLyBFQP#OYiVR~U
zVNA&x$F(U-S)HBGlYeOqEuag8!4Jy83{0tY@iihCLkvCPwX^mD5FE@(F}OMN?a1eE
zZ^+T5xyw4E22b;qxWIDR4qK5SHDoBv>c-Nay}n~O92swKxy~!gQuxg`-?G-?#{UsQ
zgqR}ZAk<iy#U-w8Vovwn^8+VlH1TymEPm)KDIvi~tx{3Moaqs||C*@JUx!Kh0ZYL`
z=xwDA7{$I%_eWPd2+#sFiQeezGbOkdtFJeaK>~Nj8xF(7A&-QnlxZTzf!mwRhq6+%
zaz0-eXvmO}S`cYyE8ghqd!3gX9!O&w-L4<vleoG5*qk*{Wd-r%x*8%SL?*hJX-gC{
zB@daKyBm%nbEt_pHYBgu63_~5uEZd;6ugd6Me7yO=kN~`rOC$U?BHRiT5U~_TGhRe
z(9Wdp!8U(1m;1rr><6_c_lzHCaL6N{-M?a;E3H&atNV<V3JG%XUy_AZd>vov+LJx~
z`mWXdXR3#9`12w(pjM#<C^dO)WkQ^?Gm_20;-||4#oVXzbTeToF%APUxSYPIQnj&~
zF{8as4UB^jEE2PS{}-=cBPk*&@y*xoczFLv?P(lif-y&?DHEscVF62^uFtn^V+H&B
zbVj{&P;N{0>|Ad`^E%vXz7A^jrktmoIwwecIiQ*Et2u6r!SAwuVDrSX4x)7Vp>AGJ
z-O588<h!}MVY)eR(Z;;4e7s(HxSn~uoH^bc8HbUtf9U{w=pMFB@#mc_*{{26N^fL|
zHyJir{_Y32rxnsOceY|TkV7T_t9Uqti7_1pXNQe9+~$$fkVpY14s5|UZA6>WN~Hqi
z!T0vfIJSFnscNt8I@`-OscU1cm8BHsb@2ce^GstwI{qv#YNB2z@m#<A&+|W8-}m<f
z`A_2qKRy9ipT?hT1oiVp00;qzsdvGQi+TEqA40sK4k8RmD2>i2Xf%6~UeY`v?|Qj*
z#ub3e!nAjlN>zqzJo8vLYeDgJ7KwN~yV6==^NpIzN-^VMzL3rBOQ)0>V<shEm>-`O
zKcJ17kdk+aa)Ox1W7oc4&j>~mK|NB4m&SzsQtTV5^f$j1QTy1sxXoO_kTz1R({;W2
zeD3ycTQjmbYHXY6i(hPo8XF9=eP{o1U5eBy>snaX>YbBFPkk}><cfh92Jfs(q^MGw
zFi1~fVx#<9mpgWboN5Pq`Ul)VA91EeQt!yNU_=w51)6zW`t}~Rvn7Igan%CkDRDfE
z<RKALM4C9WuedR(YVC&z*rK*J_-vyydjz;`=ey@l5&m4MY@P{yvlJI1J>Ql6B6bGC
z_I<V&G>)`B7dYX2QZ0A?PfV2@6Sp@f&QBMXWo5k-OugHyt_^gLn<&I4MC(3sUD{*8
zoJ3QVQeC>R-3Q^>x2d#1?d+}%C<$SMxjez<dcDFhaySe$^8}{SR;n83`2tNzRS|Ij
zH$+H-5N0DKINpxL&LDicE~IsZtAaIhPR<hS%ZN)8Hp!H~pX~tP|LgO3>+yGXscvk^
z=Bt2;(K;&P=|8%&K(^~Fw#gZMCVT(%#o<bxG1ed*fZo3!9Ef&*m=^pXzV=kY+z%;P
z0xN!aUP{4Cxvp-keKRDso^6m#-fVj^p})klq|CMzZo5{t$rRkXELxySKrL>c#6N~~
zMyR*}{4i#2(nL&j*`#;FwJ3G1tn11U2P`z4YP)?}*xktcVSf+)P-mO8?xoYrx(Atb
zP}>j8JIi;!<j?E#4`m($&)s)}A!lxHZ+U#UFkct0OF<jd;`dEPzpQy92&qweQ>jz`
z9H42S1yc6=DXmLpRFUTFtd@MQ_r0hCQ4;!r(?Q^AOx)fLSn*V>QWt77uIrVcN>O7R
zCx&4pXIGlECX8d^{`DQl+gtA5yy1NI=F54$a=BheDG;N-b_zm{k=Q*~tZ$&5nEZSv
z`?Ib6@Fyy?>}=_%>I`q|vb1PB=(T(2Y;#~^Vf0D6u_88xz_tS`4zg(fC&CL%*53}X
zr;OfA6Nh19${EqdI*yFP5w(f+>50p7<+84<Lqt+QV(|IYDy_-Bep&y)s=dYtVc+}$
zm0-r8#Tiag36h8eX7Twv+KV@5o5xLHJd6x=WGIQ4D@4U?bf$f2XmK#hdflUS1r3PB
z-UQgz(XF0%o?q@6>Og`6ah+Ku&y~z~=5oKF^1QwOAm*}k>6IwtkhnP<ISd2WoLLs1
zKdmWIjF4sDt77qcRkcC~22Hwek<vp9{Q9+x4Xi>_5Aq4(vixT;zo1Q-uU9N4(l9ZO
z*_rtrNLq_Gm#j5!ptiEl<8gFgWH?M5r-`>;yhdU|Qlc(}wiK4u_lFEaPUP%<t|@vW
zbMxB*zEkgFU3q{vi}!2S6N7_=-URk)&}3WJIsmgj*S;rfS2{dDYyA<D=l$Ai-)-9W
z&7{vU4`GQ|_oYpPgMjCC<+3b1TrWIcuADCy4zF*?DQ`WQea+Nu?%KwZ?$7ZC#75uW
zR5k$X#axHJ25t5oem-(Zu!CCLzs*DDbUbi>IB}mxrsyxac7?^3aB0o009t5-y!lh^
z7of8vb)T7A?d|etpU0(D)>ggV1Bd@b_c~|S2^-I~7VMvS`>OT}<E|dxk9quQ{O65N
z0M@7RCmW&{Pdw@!ml#t&Y^h5KObD7m6WSw=hYXk2Q*uJ<K#Q+oF+nLS4<8y;E0Wm^
zb=Z>Y9J_E&C6S1Ph<2c)t(CPkng~zx%=M%4{^`ooc}6cvY4Gj55=|IPNGv3T*RMWf
z8b^Ngqt_hAk)QtP$K-J!k0VPdl-8KWNDM}MdT?(KRj%{G+RC2x;ay&k5J)kQQ-)F~
z+GAoxsMY#HF60on9S)=v$T3peLNjH-H?Uk`+e%fvKxnlPERd4)t`a+twNZArF01xp
z#d_B!ri9^#lqyg}`1X9Igg_e(q#NVaYr|x~WT0r{X)e6`_{cg%;%%b#E~-fPI}`U@
zS&K_JUk(jImB4C3HKC@61)+%_{%Ii5yWCNlGun<LwFTyR*^7u%4!j-%u{A`6(2P<9
zVPIVvj}K4Wo=%Jdh^2m@po1zEHTNoL)u^Shvid=;HEOA}Wno!Yu50NOYB%`T;0HL3
z3fV|vy+gF6@!*@RY+SeRr!rr3-kBKGh0BKzPzys8auk}$p7t2Glso}hUG71v*bel#
zm&&Tn+K|CzN?1Rf&;b+i$oXc_a3)-n%?lw@TjA+xA>T#PVen4gTsU7nB;nn4At+=u
z)>XO87ruJ)mg98CS3i2iG>-iEM_*ugajwma_=krN)LJ-nf2p-WwL)u4H<N!pSccGL
zIWePJ5t>VkEpSnnU!aW~BlqKpUQ%Mk!Mn95dr6BO$j492&EJ!R?oHQ0RR^z)PRaDI
zcVtZkN?R#fQ1QZWsg;N8N3IP|Lp2y=aFz_pwfVE%dc^sMC&hH+&GD8H9qEc8kR^2R
z+l$VP5vtKa08QO(EHspgk=UI_G@<5eT@IOKjR557Kv`Fo<>Jqnz%d&4Ns*<qWNW5W
z?<}ut;c{M>@`0QTboGbRwg!QFT2=Lwrx=WN$<U?br5Q_EnTyZks@<=rgDu+3eb6#5
z<}p90c?=L+MN}EZuyx_`{sWZ45CW@s#JOoCO6GRDp)^?90t++~tX2m*`Wdg_N}tud
z*CDAN)(qCx4|TcEW%Wll`b<GZ7{?Q6<m0y=n9Ga?<9eMbi_1}JGd`XRF$hx%SZajO
z__JU9oYOQSA{-7AKmFNPT&@f6-al}2h#ZF#hw0B~S_!&h)o2&(p)4zEmDum1Zil((
z((({IsWzz~Dm<BcX@<2j<jCvO9j#VMU6Jb2yQU5TXY0P7YET72panysA}w~=R)6M{
zy`kr_8KHJ|MQe@LvL}fGEd*|F4%`kSzxe4-xj7tpHQjNUFTDG3B8>Nh;fA|!KXQJ0
z;;;Vw2c|KRV%p_fqP>aGLClnr!$j%<2!%oss3D+Rb1gXotKsr}G%|$&$S%*uWvQ4_
z%0h~P``Z(rY2|h)Ec4=P#WJv#%Bqh<8xcLws)N#DW0#f=e40|UQfj5NO03YTfCX9;
zN>%2vGMCD-wmk@<E6n`QQ6V*w^n<zC4{I*lN-(#IpcU53g(%{Qc5Cr<x*8G#x6=*P
zgrY05I(SH}NEGsr(Av;NH-(`4R+={l0Zam(R2v<{j2V;Q`*91jrmX9VBp3&8S_LyM
zPiIQ0l=JGWIgAtnLyny82R?gq&#RyRnBy?=_Vc%xDevBYgLN6&S6{v1^{ZQ!)vYN~
zs-9F`n)4jh4a&3s3kQ-?B&0-6d0*Rm(d$W!61W|2xIG^E>5ty>^z_KX!v~h05Xjbu
zF=y|YDWaiYl+eMf=FJazmZ(}sz!iVZ+M6>&iliJN2F77T;>gFca4j?c!~ghCSc>u=
zfAf*cGV^$OWL*oTG=A}uUvM)VSZm>C$_(Q`NXeVXWu-K_p?V{xj3N{h4;s=lB^FvV
zius(6>YqJ^?hS?ziSCbPelAHVc{o*D8B*lc?TG_{WYvx0&Di_^oAkgGHI{jytRTbD
z8Kc|gi7OjwXPBurrtD|IQiWOzYg?I{a%pt6MM8CPfVMSk^Umtvm#9Bm0bX<)c6<wE
zcfR{H{tFnN0IW~rPc~#XLhl{UeVOru4yQ8&Or@s@_J7yqcUu(f79XF7Oo$zXdC{GG
z$xL9AsO^2Y4N!OnZJN2H%@$)Rm9?rH!!~DdwZ2(}?3>|qI&gEGIFUI{Bd=fIk;jpI
z7`a|&mSttk0YRyCr0SMzr^pi6%rHZ6eg3v6hrQ)%JKM)=OW`n09P+@Mn-eJsF$?GU
ziKQ%D>+D^RjkWQzbG-i)d2vmi|4Qxo@3zkGc6LU(@x8hsy-MJ+uGA1Q3B;Hgr-^0>
zwZvQ+MV0HikV2w#s`YjoLYGTuXHwY94{7w&5eMv=I8$V!JUh15JBLXaheSvLAtBY6
z*Wf16=1D5<$6~!(?$Vzi5momk)7JXyw+>)!EU}H=+&j}<`lziRxYVyl>xcK<Y+L2K
z58=i$-=&eV-v;~tQ7G>|9DmJ+bf6`OVOlB6jI~PG`>+BQ9c<n{QtwO#p+}wjwUXfA
zYK@@52+@#eQ2S%C&*5Gi!n0q(HmAGMa#QQ~8LAqkOB21ga~MT%->IB4r_<3*{bOL7
zM($tVxzwUH2PCVZZK0OP(C32H2E{!^wprsQ{avcCp;<5H*mmDTj2wp}$8q58{Vi&Z
zqJ?>xDYdX@VZ);~f1S+${>5s*{>odI)n0sUJA|_(`swZo)6|&Pg=_Uga*#j{k=UgV
zX_t`rx-hSmDaU;c-Img@8A)xdtF2>gw##@_oN>9Cwr<Og-_LGZ9lfBBQAlLi*LIil
z#bCssyU%RvGED5*$3(l_+K<#c&*dG)+N2dOZGq=QZPnJpKWsbG_`!r1o*COb#&|Y?
z<_9s4!RPWOr?OHg*DG2bRNTT7oQD!g5@{hkceF*=4D{&+!66DEK}Z78wg0^@0=$^d
zY>$xi`{9Sd6bICV>t$x8qGqh6QI?9Sr$)AB<QSQzk?An{+`Yf!{^m#sFdatj-`o=(
z&YUk-h!JZZHlfzAl4zx|)Itj@bx)h!)~J5}Ux4r5>H0@!aNVR6$7$g6H@B?I!aQFo
z<^akwYgb-84;%8_f6%_by_n0rfn@u9c37=#pbXpLIFM2z=FIKshMU6;hv|q0BhMGo
zG!ll1o121|c7K|_PRSOw@Pe6WzQkHLNLJKA-Du8MHupl)ohjyWBX`ww2FnnFCmUv{
z%~O3t5OPYSR7kZ_RQ=G~JPo#51x8W`-K(i0$rJTFOl7-{n`yhEZSzW0v_BJ7b^o=k
zhg*Ye^AG)wVfD5^|4`<#bp{onTB&96T4?i`?9ECckV3>PP(^l6vF$>qUZ1tp=q?f<
zFP>+i`y@T@zt3!=ZvSCh>C^WI^#*h)g>@~|wO|GzCvr~QoQ|Asj@;ZFczt`v&Ed%B
zZ$G29%DkL=D2%|t&Fa(QO~xTRs8OCS-Lk>=fGDB+?S&V=VrMOF6^by9BPk|cpFZQ&
z?TN3x_>zZlB8R|-^O@^1qqRPlsN3i2KjGz$y#2*ir@pKO_kS;Rxv>9z2!TvM>Pl;s
zufO?*r|ZIB{^~t*S-310TE8!+kMCHfBLkiW8T#7fYlDMv8-p?I?@mx{EY>`j<V6j-
zy<h3|Q!o$AiQ*FOC_+ooi382w_c#os*1fT`?~Arw2ivcOc3&;s(hm3QVm*1?sMYtK
zTD|_*0%zQ_Qz^|l@gO|!AKO~Fz1APHs=iwu{lQu1doqtdjsL9i3BdX^{)D6KhZr|C
z?29hOkUW0bngc!8OG9#nTp4ff8PmwOzki^WM$C!B&BWU;UNNFfG#<bHmbqR1z+}B!
zzR|Qg;L^+wwV6x^k(@?!Z9JT3+8U@S@KhSJf)%5hcTI8-4hVny&ws`*e*R;A{PV9k
z-c0C+cZeBd8c5T~c)H=~-AAsE%6z^$Tc=jiFz}N%pYzS*M`lsxr$;W=nVbe>%oHjy
z1X^g+*jQsji<>ZqAWY-P|N3wKJwE^J4S)VOeoo9n%Etfvmw(CczWJJe@ymb3)A>v}
z)6=zZrbN!<<nKT12eN*wGoz@obaPrvSgg1gK@h4aYc0&Bab5~FMp_cyef&rZfq1+j
z#mMdPh81B&c>3tcGvAz_ST%BUJTj3u3<H-nxG8kbM1^mwZ&V?qPJi~)sK$l4T~H}x
zYZyiqBc#L-Gcp9i5cug&-Y^WA^F!llo_Q=6q*T(XEVlB*%zaEmGptrZFw!`o7P!tU
zr=v1tC33-ls@V(Z{+bp*X*30@8-UJg%w}9M3PO|UzNHl)A$VuZLZh;vQmLj8B4QC~
ze>@{XbO4j8FRG!x-XrMhhNx93SLNKQn`(Eqkr2pnAP{L)s6Y{Am_}qwJYH9{HNxor
z6<>aKqRx$Stz7DvwXJBeo`B?1<CI5_QzxNBDuKA2N5sD;4kH>Ob2H{D%ta`3!w?Qb
z<}?kwxjpi;pMJ$xUwzKoSN9x-Oe=*LBR97v?=&~#+uwi7!{sB!N=VA-^qLSNcM>>1
zedNRY%6TnRRmNdJB)JbxaOP1BN{KGBXw^L=j>jAR4}a@_$LDW8<8S_rpRlx<>w4i|
z{;Pk@uYUVm{=<h~F)s^i_Y+D1H^)waoFGNpC5jdl?NTlGT9u7+gcRozEDCc~X5BnT
zl&9;3r?rx9kBoWX_027<HkQSupE(Ykt9#3Qc>D<3Sif*ahWo!ZHW`HreQgJPOKFsH
zhSVT7O4<N%_u?9G-@zdbSQbK3e*EQUj5$+g<>{PwIzKTD6UT9aq^x0KjAKt8nlUY?
zC6J7yOjBj8-t8CM;c$~vRCSYFFN|Ggigih<R%Tm0Ikp+<x@-q1Q}THnmBNbR9!G@3
z4{9C*q_()+zAELZ@qWFcVE~CF{4))4B!%RR7Mw^VIgu>(@DM|8BI6MG;y6;q#B$2a
z`mraW0V<5g>`jN508b*Uoms0EF^i1jiSzlw*WZ54ddw`x#QEBopOjEsl6HT0;Hxh`
z<G=blKW7{=L(aTD-Y~?(bzO-$(JXUw`-Z=9^97fON3Ks7zWwk!T3MKei8v)bdv)aD
z@-5#y{E9Ug%OK>#h)L|vd!(VxtWDm`s@6E&-*9)j=l}G-_&@UY%{@Q=8=v#|_{fJ3
zANXhg@?Y^EfAbrbOJTK@rFyEH`&_!=bxKN#Mq(st+#7SK(WFqqOw@!D2+bMeOHo!6
zY#5LM!q{1vwesrzp5x7lMx>N0XI<g)gj^~jiD8-;QesGvL(bHg@X!MX?wsA#x?IOV
zh1CjIy%3H<N&|D9y9CdV2n{30OiGy+4U3hpK6}eFWG>fAeoB0N`oItea!iysa|sth
zJP}(&%7T`a>ngm86ZfZAw07luS-8C&F$q*%-B(QVMnq-p;R3a|$Z2iLs+HAN=2p2>
z!^lJ#T%v}!%s3i_g3_oMgz<+pmu=n8=*wb!cweX*u+|&JBBU@7#rI{pXK0)<S){X<
z;4mJM0Zw^xgM3xydZp-#j*>-8-J7<l+el=K1WV4i6=O;sR<taI>)N=imE~HfrJ==`
z#))72;-`G^<y-#z&wmE$;T}^QJZV%Rr;)$;H~)fpUU+;w^Wnn-tDZ^X;Tv!6Z?UGF
z<3}#*mB;y!6a$CDfpJ_}sgO3HcXW@dti(1W2!H4A{B7R6{+$2S&;PgFo{qeEdt|;|
zxSY@Y(|_?l@gIKuOaAgVf5~;8sr4Dm^pwzUh8}DOto&T#*XIwq$UAT!lFK1&lS_-s
zrB#)`{N*qCaJldwKF;2F0YY-w?_YiUYmPDV_uqWQ&GG0!x`+qrtTRig-joVL?2U&Y
z!nGDI^CM#%7zS6N1hl6-L$+?ElG7g8bpQIE90TXa3u{?0tK1w8+}_?WoG--H{qNlF
zqhZ!4YM2Ru<i2yksIAfjwzXcXhYZwKsY@Z?-l!=CS~J#lVO`FguQLyq%Gv@-h8960
zISiQK;UUruJ38qB{o{$efqdv6{ne-O10A0LtWV=lHnxtzRuy>1-Wdm5jICmwog0X8
zAPz<x21wm~(v*~*weFZwtP+CTPubQm-@3<PV-$c{0HPmw#m&qixOpJvzA&v*`PDnZ
zk>*|2+Y_%}ow&U{5=F7Pl5#@2ga$RFG$aJ_n9-($7^!Xb*E8dgM&6u0qt2CTh1op3
zeT#Z77*zy~(WO>qyuQEZ_U4waKL3jQ+k3`vWSRz!r-85DR<ay;cYe=@_<?U9Ke%C{
zJhN1sne}Ywtmd*NQ6-wEMxp9SQ&v2^Fl2A&b%W2U)(c1nnbZWSg+O4+nbw+{`1VfM
zQXBIM*R}e+8X`FdQVo<%BGC`Qp-aQW@QqECig)~|&XpT;W=al(wIRUZo(C}|l0cH)
z$&%=Rqhu0sk0&RZ`xTjYhe~aG(O#*0QKs0QwzeDTcL}5?x)n-whF(>Fo`i_(-U;6A
z-Pme9bFFui|9DvMgl!$H@68aLanTKh)taY-nzN`Bp`b1tQ$b=R#-X!pA|bkYe-a^!
z%MiAr$JtD=r&9SnbFP4QcAL`Egr3`MAtncox<8W6jj}4G7}~psH-`iFrz4-geZ}qV
ziE$WR-bnYMK)B9jH<Jx9k%p1MD$mi~E?*sok>hm3(^42Hh*frHiEg5<eGU?wjq>^D
zpL2hD#oPNYxI5gEn8@SA?ae(Ozqk@?;BkH8K_7UMxwG>(7D(*Ir!MUk*_c^C^uvka
z2jQl5_i3WD(zG*{{QTja>BBg2Gaa}+ooKC*^NNN<$OXF=mqHgm!1!8Xv|6x`UKHk=
z2-HVYPdwcF&Lz6EtOI*v9vJh$VHl_xED15EE>&(MiO$xLNNVCqQ`>q?aNn}fT)KpL
z1H@WS$=gCHHukD@Pa{tYt@W9KRjV7?m#y2|i*IL=XgBxm=G!7JmlFDC`Q!1uoyYX|
zD-cuY3>;_0RO{C-7%PoI15)YpAiMFfL_eU2F<2v$h+(1yb8j4TL*~}}aGcFUT+CdO
z=8b$`V^i{bpw*~ml&UOCWl=*{KQ!hbjKjd^pS|ME>w8W&xBG^l;y})sxi-|`dRcsb
zh}lU8G4=yvX9>OVW=dgVl1h`tgD!jem^a(9zh89*<q(CRe)%I_-F?RGc+Vl8NHTFc
zM22wSr(c~3F?0UQM{KRE!5I`Lp}WRsu<BBC?Q)pV8zwJ2xST;0D61a=rMjfa=-xRo
z(2Dy}ef#k}kMouDtbBa<$Zvo5dzvN>U^w3T`l;4^ZG8AOu9UT)oq-m6bVd=E>@96!
z)RAbh>&HA@G=#um7@3C5amcK}&>+Mi5OX9K$SE_7BVtf44q~b}2LL@m!oF5Su-HF`
zZf4pZ+~e&vpNUO7`?9k(RE@P%P^Gq^695&}LaT+fHs)$5(E)Dp+RG&LU>Z-1MBKyd
zhcuVv#hUEQG=E-dlRl518AP;(L_s8X?wsuMNmWmj%_*aygZ)-rGTY_e5ohZ-BlQK-
zdV7K1P-zOQHrCbmV{b|cclUSP91q;z-*K2GP{=X+eylDTotG;iMkYBp3&?s?HuYMe
z?_;Ppd%8&o9;DM#ar+esFW6#caJo4$PMObNf5Ge9x8ybvN=6qUF>yL1Uf;dud^z*0
z-~W=CZJqRMBs~bp_gK7b($j1$^rnsZTrsVHQfqK|=SF!8J>}4(dw0byp<gN5UML_u
z&2+9xRAU@RZ<v{L-$Lw-fTk~=R%dxCO0&wUm5j?gvF)7YU{y@X8{I?p&lrShNaPUF
zlnB~bV)Dp|=4{SqmrtKZ9_kYaL}!OdM1ntq)XvCj_!^+@eY;x%cE+B6z4>{i8Krq<
zdJG+G|BhbA*NC>&Sbs22?ZxkIUAD2rKaC&y_yk~m8h^4O!4tSVoyW(j?tf#Q-P>s1
z5g0Pl&BW>S8V1my$nAiwfg^)6w$_EKUa9R$jFr<AxSo}s2JPnPy<*d5Ktg07hD3~!
z6g{D?td(;WE<q@Zc29r=3WF+NyuIV^|Ghuw{{F=2cH-gN-=Veg=CikmM3zNiU6HeL
zoFe1xEr*jc8|L3utSRf|!rgGoS6}^v!#OZ~P~NG_hKQ@PN_3=#L>&^UGch#&tH1RZ
zeEIfA{GC7h`>+b<-#>9T-0}AIOYZ*cE7mU;4nyX*-~5{Y=lg%mTvn&i2R|Sz%Sy2Z
zryF7vl*N4lG%=cTz&j(YC?OO=h^XGVXT-Tt2jMyyt3VCLjI(dzY~*y{_HaizUpbe;
z5EFsO_4<g_#=9|)?=s_O0|%=Vsa&qLCkA;j5G1oDu0739T-P&cIv_D{xfG65=F9sN
zH={FX9w(Ty&_beWAZldL$ejj~3=Bi&RlMUACUV@mN<yg+tEZbSRF-unC*vF&5V>L4
z0V}H>171s{2A3zb)>zgv^K#|sQdt+!nh+a^qR>_%f>uE#Qb{1zjow{4Vy$ob=<eyH
zzI*RHmhZ!`=m(CJx+D<eQY&*)Pp*oAl?AhbH|0o92gd2pSwP0a<i^KGGX@e`W0%Ox
zE=sY1;wL;tJjtw?`xXULrYdN3!J@=K8WUU>=Cx7lLcLVZ50zzAGH^(VU%Y+Go6kSv
zr+@YnwixUAGE=uMrYet@j|{`a>2yoT6Q@@LhkAo*WxdYcS<aE0<B3C@XbW&LzR{Uh
zE2M^*OJ^H`g@|M$1>wK`d;c4L^x2QOX+J`n@&4CWKL61hzWVY<+~ptf^pi)*eC0Rx
zEB@`fUpnBP`hhnLv~s0s?F=K?jm^WP9t>fFGmNx@$p}#*YweyectIEAz!>0*H*dJR
zIq{R%KcZFTx~$wSE9a#!eEY!D<-*sOul@N=k*1C1dO`Dqr2#P^5M26WF4bG=%)|UZ
zXo-*_LmRs}_LVdae0G1&amt*Ina9*vqG2i0B#@dgkeLn#CKBV-y>ujG+)T;SUuzMP
zT|LRFE@+KCVmo>FyLzBSz~`4C9zM}3?PhYO=`u5~OJ@fZu1g2yiA0P@r1hlJN<^dk
z4#n*N@B^9002TMe8OGcXpXyzET}Z)G=+3qv)rj*ADG#LamV?6toQH@C+|-33j)>fY
zK4Rq~Q3YuewOkQ3ZU%=uTL`oes2a%4Q{%!gfEm~8>ItAk&JUH#vJh%uByxL`xPNua
z-~YRRo8#$*o6{X1KYYv6<0EhGzaR}K^s-Qv!sGcJDFucxV3x6vNO6E@2$h%;$J0d7
zI}T4Pxvkul!n<<ex~zm;2_b>pqPkMoE8|f(9EJb$|Ng(`?cJ9={JTJX41D|Bnb)sg
z@v|>}!RuE);_>o<_uu@MUq8)Qoc;MAA#oz4iPp|osVovmK^O;Mb-Z`b4l=b2mbffW
z818d<S*JbtB!$FMpP1*BfBY~0DW$^ieskgJeCET)_gu@u(h5KO$<Meup7>maJZ3_-
z1h~}3`7-nLc%fE+z+{Ok${gI2=i_qbZcH57h_ws`D_0KV$mh48b37z&j)^BJtfJV#
z)9J=JQ-=dLcdxLfeEiMtIUJ6R`9yC;z!(T?U~CboLahp=2X~13p{>PPX@dLvwALuI
z@3S|ji4+ypnN}B;<-+5#@@`QIexCy?5)x~OctUZY23UbCLihpA<*=oTbbwG*S&D;t
zEjVaVi_%JiWhD$k9u8z{WK$LisKByRBqweU$1dv<iW;f`^hz`sEt1I~2OPF=;E7hT
zE+KCOmezP!3(Ksmi&3f(Q{->`oxj1|-HBIs&Y=4JZ@=b?x1V$O>Yk^|2i7t(KVAqD
zxEXIeQG9PiO_UDg)>hG4Xk~UEzmRtM>i`T`_j~gFX_FuS^d~>%CqMp@pMU-rOx*DJ
zmz7JZoIks8x;t?D>d24Y{)BNH`8R*{Z#X|)kkEQVDj-85<bi0FP>nDg2vKO#u-Kfr
zZ>`%8%x<%w12&GdR{@I=$&qM+C`{wXszzH1Wmz}7S8w!outzJTEZp7Qa(g-<)on3i
zPAEoME7;bU4>YS(omq9^%EBNAvLt7^YRAx0WV$(Wa~QZgjx1JaqMWA8K;)1Hu4=Sl
zB5Gqfe_+&sV9}Et4WvRSkyHjUnIUCj0Hkp(l}2+ftLh2r(iCZ)>PYa$lL57?)KaO%
zD8*Qd5oKhE0z7(CRG!a`-JNWI?!~EIyHBOj8v?x{y{9gI8b9dq3BdX^{$%5sG$A{~
z)?;PaY^pjV!sD4~LRr0gYr&X8_O84-qh)7Rb=FYI#xQn|gtfT=rGki&#Sm?vl|TYD
zK3<-9JkKt(LAb(hh#a+%Q{wfj&v^UhHK)@J!!Q7XHuIFV&d!<VnPHf?nU2)5vd#-B
zx->540gH`-BDNCBg=z&6C4@#0PlyqL9Gr<{0;l^sj>Eu@-~O1_H*b0T%?hi)W8^w0
z?}rOdPanBlpZN8!f6F%?zGkhpGYTM<iUOl3J^awYyxkMbDdZOT4g(BRs4J@`G&Bf_
z7!qoQVup2n{3%aNY2=s_mm#sVM2JFd#!@3n<MDds{x}h$1BN+8w5`<I*pu7D{XJT1
zEV?k;!fl*LIg*Eg<1{di6B1!<m3daK#rV*SK@tZ$kaOns>(^K{C`KF-q=eK45@?Lw
zVA8#fnmJpf+2)twG1At|3%ur_V{Hj-kw&A{=7#!>o`x3)8i53sP-d`3?CkMiew|Ii
z9@o4N**Zz=sXzM{z6Znpz--R2@1#2UA*IJH82!3-nG7%{mv+Wh{qyTaZSh1hi9#B(
z`=p2;umQf?^X(oGW}JIs+S(eWxjC$x>w9r{b3AZ=cgtZKQ8ikts5VB)SP(9k88bty
zJGz*$tc7(cgxF;R!MDcR8Cgxego;5BF_7L>6o{ew$b`TbpMS;c+dJ-t*Q7GBe5+_R
z)(YoT`7mDi`rS8tczDOV4<C3qyYGdX(W)Y`I`d1+{WCU?uue_ergJBpcF!r@JL1|6
zW<!jm0b;Bq6MppN7u?_8aesVGDFr!SDI%<Dyn1!ZFlN5}jeyxs#jk3+%s|akt#-B(
zjb@cXV+a{Z$q6R`#%bU<jy{iru@=8qYcL*oB1__)BPnO@?(P^=2u)mW76MWoek%b7
z2epSQTvzHiphQ}$-py_Je$pA8&0IPu)(^?Z?xj<!qAD1PAR{4|vtv;DBe!?Sk-$#u
z-NyHP9s|2w%N}n9&S)@t8W*G<LQ!j_1!t!U&T1WcBw=pB*FLL=G$a_PC&s43=pI{x
zKeL{=>y4~PBuOCPUPjkz;W`(VQYo#on{m&~`&X}c_4?MO$RV(<E2T7ARnAXmtSXoD
z1q2R<iPj3&r!yi4pD+DzCDoI#9xo3p>r4>Eb>Ib)f=b!Qlmuy{s0@Sfi=X_AFFt=w
zm`ARwB99|27!P%U<qS3Q>u>*xZytWn`-exaYh_j6=alA6jxOnnQ6N_LrZI1%C5Z@1
zYb#YF%@Rwm6>{|Oh9JRx)T1%vOi|@Gzx$fYTKVSP%34+~>r7FXuYdFKk!4-^+2=pz
zI2{PhnN~Ru&RR7(>$Ts1^W(Dv5#~&(+<k8Zrg7jfjf{B!jAdO}rBTFqdb*H9;=~C_
zk>kzL<zSnkS_s&(c%oz*SbOSpEtTteAt#qw)>bfavu8KJ*L}?n`2N#6u%*qt3R?lI
z&ICyeWU=mN=XF(mLDcZnxjz`+_qi+^c=fz5N@qiDQ+!*;TkCsO|GOwK4B1(-(cB|T
zI}?55E=eGcqc>0nfBvke^GRo~y2q(Nt6b)lYi(T1GaLGNI&wH3IGt`d91g^os9Ky^
zx2#+)7oMKZl)5qwk$_N^;_|X$`*Recr{Wsj<FNzp(ixpi8ymu68_Tm+rkpvve$5xJ
zzToq_A7SUnyei>3VbQqe#&v$+>kr@ZSHJ%^eDm>le0V&w)Gm?30g=$EKU-qmn$y8g
z9Pg)TrI}Dw_H<0CjTj1-d1ftx-=nHfl_6&iHh6H6`XPK%zNvbs#Bm%rjfd__=yO0~
zWEcj_gi;%3o6nc8I3^X7*5!!)|NB{ENDMhK<$*|`mBy--YK;dyGe{%@q{!iTB;}#2
z6FLOl8m$^@fwcu%$SA>O(AF?~?P~}cvWFm5BdCK|LmY@9I#_2$ZOU32%jyiQV6Lo4
z;-MH`Tl!?|)?3f#N!xzg{?}_%lkd#yPveI_J^@&t#-D7sFO^G=_JeLei*4L?-%!1n
zGi|JMB{q;C96~0hL@C8-u-bLlcnmEOMii||vAIi*+|+vrMo`~)BG6jn{f9?BJYBgI
z<I;?)`2+|+vc{ALe)8ilc>DGhx3{-Do699B+I<X_>*Yd<nd3C^{{4l=$47?YNQjY;
zN6ac~yE^x0HHyw4jVP5UmD;?%8^RV_44;4YnzyfB^Rv%?&LQ0KSO4^ktbr3FXRECB
zk#9fzmWQWr`S<_vKk(uFj=86^RE1`ZV2yEfADs|;Kx3z+Ie6u1Wg6T#wr(AtHnX%y
ziv@`zDMXsB6jLNQ-TD4-WQdtV%FMM;H8W_Ss_}kJXjL9APt12Gq^BO`lxS_Cl*(rO
z3@!z#b!FCtt1a9nBM*`3G;$b6ro$1f3v;<}T`L!BERPp*O#JwMVi*SQUfr@>7naMF
zFlHo0WOjp~k`ncW@w%I_H)s11QG>ZGZrWWGs}<2kZBv&}y8NgXsMQbArh;fhY{2Ld
z!ZcF!lTU20$N2Q8{pEi9azmB<d%hpr>`OduyNf;x0ki!hM{xi|o6^>{f8#U^xMM+c
z*^+M6ba1JU2x%O>`)=-IV&;W(7KmHdxBJ>Xov)m)?j5x>W7X#Amm(Yw6Q`RauU_3T
zj3cE~TC22Hu#_-?^VyT{4%3l{5Dn_Qa(TMCZ%>R!$o)_??|>^fyD9`lMEA7q)H}6i
z@Mk~%8Nc}H&$;6bVHr99#tXk@@MtR{PyEN<{+8dr{|&$W=372q9$da<U{#S&2*JqN
z0}w(oBq<iVj6wn;+ON4kdmDSraB99Fq?nK)65|X6e){9DczyqhDIU1a3l<}DQ?8}*
z`gp@}UHP{X9L$gY9M@(HUACajlhK~7(hH4Bp^{=iazdg5x;LkZ!#FWb2U?rCu2*In
z3(9hx8Di%4bR>@huf_?M73LL5fsg_$E2hGl@DxZjY;~Zi-Z~InYr)KY1N{tco~G5F
zpM_GD(Af!Y7~KIZON2Q2;jBF2k~E*CLHUmPFZ^)kF+9&@+dxoLpUbw1HXLvYK&gcm
z46RBq56YN^;65|O!BtO#Y)EuK!3gBzL=?lqm9DDwzz0(jfh^A4HLaX4E9dLVTpPv7
zg#CWN>$jir>eU^2nlLl2*DIwIT2me$A6c$5m+O^b$h>*|n)A8x@!`Ci%Ev=?&&6w@
zwZ_BK2ijT@sX)bS#bm{T`>cr>fwBWDfANdI$zS}9Kj+<Ft~@;!4iEQKtvtM6SkJ~X
zul)0W_y6ISU;icFeE7(9DbyDLZ)ggw_N3AXam8W;GJ!x2-gMEnIxF2G3oAk3FnE0<
z2q^?4!6A>VRr!y<{w*J`Gar|J=;ayM{@us-JYCNmfBxs(9uCB6Bm&cPAjC%u*+J<Y
zz}i51)FsLpsD7wCo+ggtfoYtmZRRr1ECGtZS|7<VGu<2sIdV$3WOWvmBZmQ73tEjU
zc?Yo8rExu92`85zmbzjDmr^yCfm?GHWhNnb(jD3@5~`s^uo|E`6FUbywb+c@Q*BqZ
zzGnUa=dy(UTc5};BA$b?1EhftU^RAstQd^cQ|*p8P@qjwt0*pk3la#^(dY3{eBEyq
zdM&?&0}vu2#`(H(Esbka%IXaNFHc|a>h&vbPq$3d(B=Bx2%hJKfbjVEL~Vte&tDS=
zESH6~D9hZ4hk-aIw+s-db#`E;ha`y^+N~i>6oUDF(2QCNx5qnP-@WFmH$Ub_uYSz8
z|L#h8YD|{_jmFHCkLw5CEx+Mk{?$L@x9@*RW%GvXztq#dtpRBZduZFDhKU0>%|ZvH
zyv_tcDx5D@=4AqWe`l#o#{-%Il@4IV;7m!NIvV31RL5g~E)BzN4u)aG5SH3{{>aWh
z!lb%I3A+Ce^+fOuygK9oXq2_E*21C{y*7prc|F}CDRDdu<UzXErjJ@1tr*J^SZkoh
z!INZL^XJBjH`HCv>BR8Pz6u6o7GieW4pR^6Sc@{R&XN<lHHbuKYne9*!PgA8H3&Q7
z%KyTZ91i};K7DO3PCtzw`uGH3eHwq#@!cZfxjiJC><<V{J<h!;<mMgP=;<jS?$6YW
z#auEYO-V-wEgt79r8Y|4x&`V-E2+@j$0g2*wJE>-_`uU#c&yt-;Wia11a1!_w>Jmw
zZ%+)vz_JwRTbW2C#>~^xg%~5RU%loqP6YGfzLZ8_?FZj3(U=ajrE;DxZU{VtE`3o-
z_XZQKh($t6eD&(b{Nk&h!_&mP3S+o+mef-D`nSL5o8P?W+oxagczws$mv=1n>OM{8
zzDqRo0Iji0DSPsxb|gy$ECGYiR;O2oke#W}+{5T;xo~F(Qp%)I7^3j%X5u)G{OJC3
zH-?N8cX(`ih?%ui{>eZ8=UgvOw0S1>dzSOSVG_(9d)%+HzQ8xsu2!L9T<d~F<2D_L
zDKV>}wNaPCY?Z~7Aq1o~u8(Jy5O{j(+nBg#L+wW5O^x4t`wdTzGhhGyJvj%4A);?j
z1OhZKpw?c{X~V_(t#-elL@QTHY3|9gHcACCKXBwU(So4*KnQ|q+0(0hgMAU#y^rwx
z6~lp=jY9K17+Z0_vDfGfKIECP62zHI+8Rw7l}1dF5Cd7<XtcC|Da_i9!P}-l6lOfJ
zW-XN<!Vui+0fTkDqP8N3z|xd&K0fexT{*8_z%+v#-B0PuFWz#0d*h%BZs>2FDs{Oo
zM5i?nLu44;S7n*6l*K9SG7B*&(=-B5%usEVRv-##Ojv{#jb@(M*R(ODiQD51pG}|h
zDt<wIpQ%^(*nq}d3;*HQzv08L-}BptzvAQd8{VCtSeqBWh8rb|0;Mty(T$+R_OwUu
zct>JXkGB`^nAcvsM?6M+91m{z-i*Vrva0go;|G>?MO);0x$@!T1Hb<5@A&%Lcih~)
z>IV`B8&AW5Ii5YvzE)&esntP%ko;LL);f^Y*=S-^E$&AW2P6pBwz^bcDa_Vb%t+bi
z@%em)5Lgoo5<K-0UvXE1QYv4+|CXnxnXkY3z;R66Oo`84-C##()e{`6h8>93+OXOf
zhmn+$cjrx6*UGvoYr)g$ERtj(1tW+%r?j?uI&5dF?$p@tjPLh62HeX<wsz7BygdNT
zy-q^kmx%$TQfebgWJr<LrMz0CgZl+*MzF+?4Na9=oH6rckTDL@t^X}B&sS;-9LB^_
z8t>0nE=%om*}MI3zxa&1(~;BNi9BY`*UK|7UK<cti!<=1<H6Y^SYK0m5~Vd-^DvL&
zG@(%`W$s>e#+V{`7>T2=NmUJkAL38P6R&Pxaj<*p`vE&g><VR8u5IOAdE!63f6wE;
z`G(&;{YM_=_gqTB%amu|&(@%n#t;n&jSwb6Pf(@%T3J+T+n7+@n@Gf^U<xs1B0?6p
z54XhZK!9qO)pW+e7K%IOK+eh2ea(YuL;_>V&j(*(->+UAb^AS9m&Mm)YtE<~B1E{f
zg<4#$Ig1<n$01-*n8S>TuwW1)Ipw~v!_!h&=EB#PE7xn`;r*HWbwbO5aXc}Mk@<Q?
zO=vav!M7N+3av3s2XY!{>Qb>~Ei9|Dt^&=yFk=<>xKiozvvv1)dQQ*$qnW_>dM<Y<
z)%L83IHnZ?X!AXJV-P705|mXvTq8zL?E}y)rM3T9gd7zKN-HkOzE-1JBk014l4z(z
zLUoqb+?0p4Fc)QR#t<{dF>(L;j#sbmTy9<pR-3QUW?ZfpmU$-S$YIRnls%=+JVjbt
zL+6!RE60O3hf&|twgf(abzQk!ufBGKz_M0S2pnYM_33l|{KtQbsoit^+F%`!2Iq3&
z@#&Gj`t@J&P`~Cw{T=Vl?^)0t7}P!8)*Dr=DLM6|L-Beira+6jKVBx%eJK@8DXnq7
zUYVC8s@h43LNetb#^3$v7d$PmvBSwzluM(U+Y#KSkz>ppLL{iVe9{bwk!moP2H9_V
zYtArj>Hro%D}{BLnKI-YIUGEJabAnpx@Be2>dmiYXf&?1xy5ETtS9*S!6fo<zHq&+
zeEqGbJTJ}311Etz8WN0B3#%%Z^9&XUmKbAltC8Tol1nLIN|Qhfkrw=HHO0ofD4}_M
zX1(s(6Q7aqdVzjt^mI&fCB!o?>(lt5k52&Br}3v7-;o||haLaV&;j-!-Zj&1erP>L
z-D|u^G>Bb7Sho(jLJpafGmkbqUA0s9&7g%Y`Du!VN-8UJQ68>W&a1M#pi>9&2=~J@
zaC00vOcN<ZN+~#7NC7vRUY3<%7<hem@FGpU8>CHX;)im=7EPO}(pc+C6YZKl>yn>t
z@@-16L?UrFz2fcd7p%Wa)GI_8Xi{0$EAK8J`K$9U`S$!Pp2~X)7mU{FvONu@fB22)
zv^8fWcqeo-_9O`{deFxPHetR2Ud!s{ox#l+V}LwOoDK(Pv~&RLlrkC<ISs6(@UQ>f
zzeDRvY1uo?hD!*$&5BEp{EhkbbP&rA?NC~wsF1>mAd%7;FSRy`HYz>UNrm;gfItmO
z%$Ypq-T_lAWGao1Pmer2&b)s*b2BBxGN$emWa<*m+L;@s;w%L1y&VZ%cBY;})lL2(
zxY=~fq+pclV5PYeEn@v(@}GdOo_o9EciwouN#BDZ5_VHe$CT{5t!%{SGwgD$rb;j`
z2D|q=)ubmGX>VY<NxH<qy0)EK9>k@6ic2zUT~PwL99diA@j7!}D(hy90M6hWGWT~U
zZf}nJd)D^+O>3#1+Bw9|5{%H4x~{b9MR6-WkHgsU3!{IY>IZ}rcNwnA7Gv%SUAKo@
zj_H<(6A#x)y$Fejp)9rX&EtFi>fzt>&GIYG^#jV%LEW%-eVZD-l<2cuaG%ezM}}{X
zwY_2M5SO?|Sd58HSTN*_wZ`M)BiChNUX06S=EKJae)YTG^1H9U=JmQVWCxvwlw9)J
z8D*_0NbN?f>QYW{5Oe#VI54m3hT0)|k-uo8)k-a1DA#^C4?$UM!GbUc!*cdSN&0$W
z#@dvRk54>4%{-hhm}U~gE@NyRh_21&Ypv>uf=zu5PRS2u&Ht`8r8WT#L>IETmz_9W
z!>qa_V?Wsb@#Orz&tsr7Z`d}B=a0}2?+!>m16wX@QF2J`2`1u3_Zs@1tj=O25p%;r
zmzp+X?M;h88ru0?YiU#yQf;iYaaju2RarVKIf`(5d&BG3cZ}mmh>>Moo#|;l*IeGJ
zNNY@aBIRDjcEH=z51y2Y1tZ=Z5b@T6nR}%SL0qO61xt>Xgy8>f940=ydqcJZ?HsX1
zu!5V0uPcx9h2K5>o?kxvTP}QLm1}=_U$ZuzYO~G=!KLkf@O^&1WMf7P`+V9SCRKN2
zasz2W0vtz|?R2rWuLCb?Dg$y#3^}s#B%1*e$ld(8H22Fyd@nZXQlk!rw^pdDAjuE&
zA-e6r(h99sN~u)wfEQ0GY^=dk$_kLio-W;Gc}s1amz57sEAzGTbe+io4pC_BfB|J$
zJlLmYv^n!rnzQ~S1VW$7sGI$lW>j^lej>QBbo1xedM&9tGP}i2{J{3)5O!Mz@R=;q
zfvoLeVCUv^7O{tGq|VwBJV=3#pNhD_drpBU#@u{ATABl4gBnGkiqS$tG>}tfZOU9d
z;kq;!hsbo8I2;cgj+0mZt<u&lLe~QY)~4KzCsOv5aA~?rjjj2bj0Tq?i#NPXL0jAF
zky<P3>ON9PpjzMKW#DGK<IUZda0!$PSaqr7wOsjl`N*%n{uPh<4IlMu3JZ#PlatK^
zegl<#4MgI1_u>%Th`yh5wBLvAuBbV1SGxU3z~^j0$pT-!zGG=dJbgxO%Cc6f3eAM6
z2?OHZZN0f?cmtvttWAl@ecg1ER?_y`OtsN!L5A#2#32z0l(k`{VCzclYeWSwp|nOb
z_kNTX2u;w?LCkgG={ocA>FV#1%yrJpvoqFX_C{W<E+-8nl8j^tghZcLM%C)g2z(#W
z5IrG2B%&5a#*<*trT<anMGf?wqSNQWi?2vO%e-V<ei}dQ@d?2CH2!2G^lpaP^Wqbn
z8ILW#DW)7TGuCz?A4c-&Kx-?l7RusAg((ggfvABzOblb-FaGu~xPN`iKl;c2l=mM$
zP}bF@VkWE&W|VcQXf@)htPS3mo<Kzhu1uIx<}hY{@n=8dbaTTrj<i<!@ZlrVG`i-z
z0xFCLkEiz5e*+UjGf&Vm_qtK3SPe`eAO#?h^2l&7S_l-NO6+NTaCdye>)T&2mN#5}
z7pd<eb%DqFk%z|*{N*qIHSgLxzHZ;J=$Xp$e2~!UV4%dk+twPCRj?@`mPp&d10V`v
z2$bL~wcB_gNOTYZ<>_Ieey)TN8KN7SqsAd;e)fw$LnP9Qab63TwKCU*wG>{xy5~3z
zNIUU*nmm2P+|+twf`rghTdYyr%2KXi%~^KQfy<l*guvtZjI9f-E*%vS&X<`89C~`n
zY=v<enW~#k5*zKXQf%eJ!+S25g_>3@1wu~Vp<R{p^2E{_OA*8(*#?FX7^jJ1aOwUB
z#b7q2^g?L}iDS$xDPoqHTf}Ho+tOgRqUVpt%K};~5U>|S^}qfj5FDiFMjB>41xPw`
zL6wl8iUSj2%;cLRwC1tf*A+uZ3?Ai;FiZ!=G4j)&e#z<Pz(4ut|B{c7kF?nVF0INm
z2sIk#vbyih$H3f-ch|yNm13@VNeFM>yyE`umixOC<B+`rS{h3&q?AZ8QH%TQ4b$Y2
z@C5gMsg<Ng)`+f!R^aToHN-|;DrH&FZkBEc&AgBg5*XrvS2thsv(Nv6FyHa<H;wf%
z(2DWC{GJa_ANZet{jd4ZKJdQ1quAL~vcAJ}58h?AE_u{|YQ&m@s!>w+t-%w_q8F_Z
zWgJF?K(>HLW+|0qF1!5vv(H#s;~)Na|2<Fh%-Ue7mFv9lv@BfL!fzfwGUd$w@{6Bx
zcRG>8As&&iC&wY4TIa+hrF98mqfvBq(2d$Xdm@<mJT5D&g(iNeJzp=(2*)X40j|<$
zdBDa(9RijF9ac(R`FMWMvNYo0#d=H$ElRnrTy3GKv#MwTT6CW^ftWL47%AO<Z)v{%
zHx*LI+@!!<TzXA^28vc3dC&g|=9P~>oX64+0N<B+3_h3XhXL#0>?Z4PO=%iP5sDfY
ztJoCDhm$vsOwd(Pgk&CZ9wjgik;9aD`{irK!@&RhU;TR?p3W>~rK<0TqySl{rJ(_?
z!^B*ThuJ924}Yg|<n}P~`I~!Q-`^3`Xv@lS^#tH)oVuTp?I6zO{DiIpw|O8cOd+w<
zMr|`@1rcNU=wQ-%Ewr+r%ehO?%w;Ua*^fyg({RhH!&m(H?(dS8iTNp@YoN51-(3EZ
z_t$s)yN~~tb9>;42P{<FFU{Xo&yfdw9h*zzB%Yufi3D>9ID|kPBBfR;#*h+85(6Va
z0#lk8vKv}!Ye;Jh!;za5IW)MJ$QPgAQJXTa#Zw*;etiFi`@@lAh=?h3t5oZ3>E?{6
zI2{p*2;G~qU#A#Dm*X@)jA~y4n(}ac#A=07Xd$8^oUd26LO50k!YYE+3G_&bhBC0k
z3v-+~&)#Sp55bvRLC|8fYvWQE)>gS(SHvQbktt=S<cDvG!di<b3I{i4Zvr_F+-l(Q
zvY?B>;D+pTS&$eZCZ4sj-yPqb%e@itr=NcRe;c6#gx;8gmurwW=AZ-!YFtZ2hKQtb
z2Lf8@fJn<b=rRl=(-?UBqgRZ_fq(w5f5pSoh3mSqcHgl&M2ZG#F<KDPTA7RTv~G<)
zxVt~`<1gQGoJPcq>*a!oFh(~Z*XA<!5F@n|A_k+U53c7ctq1VbplGW+eR%MOR1#0(
zH0uDoGqA4LD=iNwLaWMQ$b5PF2`4^>k2mPj(4t(=Gf&qG|NPhgjED6jzpuYxW%d9X
z=RbFV-<g2YnYi0LK3LvW3&A{5xk!cB&BL{Mtzi*@GbQ~oi%5vXD5T!J4`%%A^B?)S
zCLU-CwJLK_mQ{KD_C0Ma$g2TUp|pZYAde#!BC84qA|lBf>4qmcixQ)_Ct=XujCKj8
zYGavKs1+%VD$d+otIMiwEu<6~$BC&WAh8k{L*h~%xzsDODuKi?k3><{RybJlGl<by
z+Y3RIm<IBgxjCKCwJ<L$llU4=iqH&8h2!x^8jd_YHKeXs2)j(wb|d%yL_<mB8LSod
zXX5*_8-E)Ax#JUn^=bUc#*UajHx^wsC?4zVZMChXp*@z?qqW_)#EaD^Zsn}vvAW}!
zxxKmL_3b^kHz$^5;r#f7srC3~Fq7_&ql8jen|mT`o2;jH1%{NF#*yRkz~L|wx_8RD
zu3fI<{!XIm#d8SG%4q768Fi0|js9g;QH!+I0WI4<W0zi-Uj&l?>WAOE=?$TctQTmD
zQHt`oJn&(D$G7v>JlF$Qo@f*dXQ*rk>E{k3yP2n|&>b+HdfTIGH$Sq?5OMb&lSHuA
zlYeY4ilV@>7V5fEnlt1|H6E`Q9xpSc6xY(PD{97RnmmqK4cp5q_buBzdmLzM9egvF
z;dK^=^wh6Xt0$|exjCZ<TD?nJrFz#EB^IS62Rp2~87-k9;s@Ttlo>~lTNk7owO6M)
z7a`yquf#x15jCT<)|o%vAwq;GLJo<R0?8n5u`cNIzAE;ed-A-E{z2!rca$*pYxLKB
z^yV3KfNtx4%J%0M%(E?p(L%tQv)#0_JEElhdxkhL4FmV5d+zUVINjW^tcAzNtFsT(
zXl9tWY;d(oT^5##gICR$kJMQW(>QWGjHDC^q5B>+f2Qe&_0aU0S^{49HN{+u>k>$F
z?m}IPd(h+_5}}@CNcXGKemz72F$|3H$T8hR&8*FnbxNsxnBVhm{+4g&ueoq$<^tl+
z{!W|N*T)Wms$k}Z*gK0rw)pw}rHx{R>^?D}&-vD#JC|ejIPG`uKk(t{iGs5*=HjNS
z#oXlHT4NY8<1lu*eowbz1F+aT&3nJO^ZB>=Y{5I$(&v2L=5e#<)Y<B6%n}P0gd$3k
zMhy*XeVq*;(G7EB77h$d<3Juf2Hu)cTBWF)z=}np4us%|OmudL%S)}x#{B)K5E)Y-
zwMtZ%qiryhkL|u}>*{}c9zXAH_Pv?Mu+3$EHf&?;?(9i?59`li-MZN!In&t8nHu<Y
zH1TNjA!WuqaDRKp&FzuX&F@)CMK5P)<`Vw^%>vh|R54^(nX9=;f8U#i9J!fB4#U8h
z6S_9{BrHArVX&PYOZMl}TBC)E7Jr6AXK>Ke*|A-Q?+0h2Cwgo}0<AM_O@u&Xh)2e7
z!@wO>s4J|sFxQ0->$iMZ-|?`1VBt!k;5l}EO|k8X>1*N6kco)IjVjq0ZbFFSG97W&
zq65e74;I7op-?)fBZ?c8pN0X!xVycj)W&sQX&u0NJ>76OOg(kaCA5myEs6t^p+DRE
z6WkFGmqK;fnHgKqg#op7c4M>sL8`^M<V#$T7^opqW20pcMIjj>3NZ;OJAj>1W*9Ot
zX9ozh8*w*dt<KyaWZ!pFXOd|Hb-><w?PS47F}b%6(Hm53Yh>S_<e9Va2fOh8r)?+u
zV{XS^@A14g`R^}9Q@imt4$8GI@%Q?we`nu20&#ChC5a3vbAP(y=I+SN;T_9b`FOFO
z0&A=hQZug2-Le9{C)du598zSOMvgZJZ;YCUYlP0WQEik`8IpTjVZGMx^6yq_H|LLT
zN*|O~3q}BoXwzP2zhL3?dRRJ3FOs;)cMMFlMQE$<JD2&&<Mn}e^S7Mq0}B_N-`}>2
z>b2_jA8bIs`>5LPB)-k#O`^IjqR!wF5kg2W{HwbCLg)-ktB4w-M+-B=?9$g@L~Vpn
z|G(`0X|pBElAeb>?h%ohYwe+?Gu+#Ky8*C45EP9jQu-kM0R4KI(MXv@P~ZRpjlO-)
zSf_@)hRld?*9VVWyUy(fVIu9Cp&FIP_Ni$PYh`9+xWE1moNWMoqi!}I%;FRYy|?<~
zkJm9itXRj@-|0)l^S)W^u{vsN7om->TdiXjAx<p|)|K7;KP3Wyn-W<vOU~q+*nK$}
z9;J*Pgwbn+AFmE>*lx)}gdHVCay<8tP<<%D{Pvp^@RMuda(fo;z0Q7o`kBtZa|ZS0
z_y=};0kFOtf3hP%fIVYivQMjTg%V&Sa#E(dnVOu-dV$)IX5=!5S$aBxBuo>BW#PCi
z+~s?Y`Hp|~kN=8Sk1u%p<6ADZa$cRg!-Px=AFiE`EsRuycFS$rJgKs1;p<ngczkhW
zIZmW3)NNzkRxamDm;^Vcp02Us9Ph|Upc}P@_MRr(<w9*SwOs2;NBFoJRVo`Bb*;hT
zs7~vGmXNVfXO6t#H_LxQyL8^Y?}#~<?aaUU!+*zz^({a0dpaE-o9ymg0o)$;y?0a-
zDj6RsEusaLOtu_G@qGj{CD06#V3EWmg`AIQt<a5i?L3_>0b-f))8Bs2+aG_7*j`Pz
zCjRz^pZM|ZJFeF&y>;rf@$&wj|LZ^hPf0_f5?(6LjkszsH;@kF*Caz?B|(42N5pOm
zsc3SIPj$k;)!=;Ia7omhgY6;E&%^Vg6bjC7{_G{*6H`Cn8k=L*Oj|p(ZftGi=~5{=
zqf>%3F)b5YZ?qm}{z(#P=}aoz1Eo()z3_26vo?SX;^Y+5Xp^0Bm`FFTuh>p^Uo3`P
zRS|Ie)du_@^wI4Z#2RZk61y194$*X^1VyP?qadvf=|-9svKH?1T{OKohiT?+Iq=oP
z8{WKn$$$P=7ry`LXMXzezoBhmpkJqi=ZSZh3m-Q_PbZqexkX)-fKm$g)6C=Xz{Bwn
za71s^ZKaRS{BpU_YUh4FFy#{9vvfAUVzpAw7i<%##3a&<>$OG^Gzs1-4F8VaI#wO;
zf@GznnfvsLqrc{`y`gQ9pme^Rc)EV%zx&(&C#U*>5Aq8(BFAB@j{)AtM417?3L?Ss
zbB`&gcp?osrnh(oq$wCj*`37?2#MYm@xr<`auHGqK}2$AyR2L<D}7R$k9o5~c410|
zWtw^Q@)eJdkAvkhCeaokg-aTv<7jlJq~yLAjA2_FP>_I)b=;j~gav;O37L1H8=OvC
zcwo(m>_Rhz)zRF;|7%(Jvp;_gU77kqPKBhIWtxyKZ0p9Qt$a9LNK!bId!}QgT{UZ5
zu6=+q4%386r<ib`GgHr;Txp`Pg~Yn#jH^<GTR|QPQ2(xV9DXtV^CDoyUa+%9$9*Sp
zS>ul)q%wgfHi?bG4YUReD9J)fnfv)37iXh#EDQJZ!f#)H!|OLMc={M?`Nto_G=1yP
zGBgrz*Uq^cbN@)QVYCfQz~kMK-+cXsyJccZf~OoH>vhG=NGWqYU0AOh$K^no<{R+*
z@%=ljR&1+?WwM7UxO?aH6yhM)gmk4Kj5$<zE^Q%wThftlm*4SV-@v<jx@~l;eD~pd
zemVclzj*uKap`BSoX0x8-4TPnvjYmsURuY#JtH!Oyv0DfkP+rNQ#Yk2m`diL3zN^_
z8uCk1%mXY;InoGohX3*({RRWxeDyn;IoGvP2Z*3oLmd`zlGp&16)1EOsymY;L<*4|
zEb%^f_JX$|({wDTp5yJwjd?<kR9I_~n)cK-G#B(ZQ@6Np^dKw+-rPTMzZ{U#k(QY2
zM9PI+CJrT$MJU<nvVlB-Ldwc=oVmZdr(RdK^MyGFs71~JwzzUPC8j3)@b*lm2H-D|
zg6g8wF<^*VBoO~UUdu7a<Qa(gOxCiF=T6kt1K=2dN(@Q`6Gb)Arh+655U6g@JJV9g
zDYKLVcqHj9IYN4V^ZHx9`sNj<r_Ogj{>+a*{Y=}8twG8m<^B0;T$;h<6jE7-yLsZ9
zS1)+|>XFCCJ0=SjU>&v(>O}hB$0uIDe9gn%0ihypoS!au>$G*lO~cbw18AwW579#~
zlwrvFdyhb$$45?=M(v&469<lbv;2-?2TpG{R==?JmH+u~|0Tbi-||EMJ&kn?gk$iy
zd!_YZC812fKA?-W(0iDZhn(4O18UK5Kh>m6R!C%WT7Y1Tc^9_YS*FZ1O|-3IwX$AU
zl4kDj7u+LkYU>@XEgno_<dV>=0RpDNyd1-))#PTuvJcOD#4-(VCjt=|`&on|_zfJg
z9pw=Te>pSyR+-9RwX0w`QD)_sm2Y0X;qjqy_cGGB54JGpNH)|Q&xkV3%qo<8A(hO0
zTp$TsTTykAD$^Wt(QZl>VRq<MNSiQ~iBe|VQ%E-V!N91N*n!Hy%JNV8$i0^O1k?R;
z{6}zn0kFOtf3icygm;%Ke2S9WU<DaUN7$@WJ#_TNh<N)Ql(F!D5CjaoJL`GnGMy<&
zxmy-qzIXt+@ZoY9GM0?F&<0+z`>X8p3MC|jQwf%i>X=Z}7R7STSnsUsI;J||Eh8da
zuUBZX3DGrVJ{BoUYBs>K8k6BP8di~@c6UgVkfm@Zk1TRWX*0E7*sO9oJ#kt;@P2*E
zseYjP@Zw@GFrTM}gI_g#GIX4;HvFZ=6MQJu_GB6rZYe}D@6Z5>hyrJyd`vl$Br%WZ
z?!WoNcep5Blpo)H;Fph2Sgm+Bk}0iKc1BLEJ>r$e^lE3F-3$ccoy2vJnCz&^4d`SP
zoJkX@N1Ko}$7bB~m_Q}0OBlYIh8g~-)ep;z^u*jJsw>Td{npLNs$|tTqXL>@%G-J)
zNy0@3D@-Cis#|R4uIvs#KD?Yh^+SlM+-Us8Zbgxg0&{1DjPcRUeD`xW_Up<g*Jo#+
zgmfhuoQ_iouZ`|P?oRTMxT>V^8xt8G*%6n#T`Sk~m8Hx)9*?|yeB|lr%)5Gt`S2*#
zwiZ*xF@U`tB_)=5VwwsiCwdQ$g}R05x|B?Boo(A-K9X|K>D?@3DXn6?;TCHFkNYwW
z=2&mf(_R_>8$(foiITZ5k1TRTdZN}V*IN1c^fMnX@Az2XbL}VESjYP@p!`^yL#`KQ
z5#dP_jR-eZ$FFaeSTE9}DT{R+lUY1u6^jU-g2Adpn&XlLJuJb%k^uGH;4m%R&xbH@
zwLO3vkVJQ<n#J?_ljUpsWLASABJMHe*XIefj%QR7ZcLx7vEIZ+lVwWpib;fZ$N<_C
z^1P011`if4<GjQ=PH{a{Zh-+o>Ik-G(EzzML-H`^FOoqrC1p}e1nee5(me|0TjRR}
zihs{K-WR{i@G<*5j$3X233#=?{NU5Xb5d}&G&auaWOK5|c~KNG9YGFp?_~;oHR`2u
zKOb4|HeTM}aXnvIpH@T?IZqgkt%sz??780%hz6%*gkasYYQer5`$%gsVCcOEV<~60
z7T#>lp>N^0c6&`8081m--D*+pn)qODLJ_4%#I|P=N4=-`gl)#Dow&3!Kb?Ny-Q{P_
z{Y15u;pVoV$s21U7y>!ud1;Ehk3Ai9ApV4Dd<0(%KRe~-Ka_}jGGXEQR9i3;G_e;D
z561;aJkB%SV3EvPo!XrBat@CmiNR3o9ZNan$>Scsu}26-$v#*Nwl;`4TB4rVrJ51)
zu!kPRnn{vD!voR{S{GUqx(Z$$S_DgEO&phu_QaefJYphmEaxnV>}fDplssi53FhHn
zC>mB1A^|!pj^s$Ebpm`YIbIh$1JIwY<$cfk<F$M<d>aenb75}Wu(koP?9bQThpiKT
zZa9@#2;O2H@6Tl=_F%>yog|afOs`J8H13xpFCR8u-ruvHFVxdIo{Jgt9_h@@ZfSKX
zCyvJh%QTU5!q;HO+mJnks$#5P6-p`LKiNA~Vo2mYfVh1Os_{!C^bHoY?EA3@IcMA<
ztC4Yv6tWbOPq@Qo8}BdgdD=emZhgzg`YymL3CZ}te#XxdiWZV2vM%9Qm4xOi-L9kd
z^V`?$>?cifV>O0EIFTgSXRbq4P&-yU*1I#58Nm|mtwtZ@A@j4594KRNL!i}2yiPZO
zd<5sn0HgMbU$_B0k*Yetxjqu$HQ4c@>Gs?TiwF=}tmEeNMkqNfzOHD(vv8afYKhqj
zNkbaheGH1mp@S8_D=y3<y*Y_Sh>IvT%Asx@_xQV%2E(lB?LD&p-}hR3UjTAj1HtHu
z=kV>CekSYq%kh8g_yS;kIsSCVUL*tiQzahE0RWf@O%%iE+l9@H)s4qIJZf|%X`yZV
zl*6MLT`NC-|1;~y4;)`Dyf_^APyfkZ@HgN8!ta0nYmy|EJVQ^k8deCgh3*f225_^R
zw;!U(vdZOrMY3{tcgN}J%*T%>9_M??G_foP#Eo}9eNV=j1gu73ka8F&ryjD0b8nCx
znWJz`IYWykEh8M$f!~zB<fVQ^zRvV^;bJTQ)A#?9cjuq^vHcB=ZOlmbVk|bjQ1mED
z(0!+T2!Kx^rxn#`;@u2sAsJ8$)<DP*!iYLq==H@zA`)yR1xY*{??|Ft+Qv`sKk(oD
zi+{=1!*}T7a0Tu~nMApt=Gc(U>DD5NW{&sVEE-@^zU@ByCL|*%amtCQ<d83p!Z?dY
z@}L{Z4Ve$%!n94)0h+WX)C%%gD87&_JhL7T_oyjz%Ul|qw=g4hoZ}QS(Vw`Ew8$BA
z<9t1HTozKE$isJIT1o(N>qbzd6XHVeux%0bzRRkjdWdVHRQ9Broj84Ca?2=S_AmKN
z4(HP%WlXa@-djgTTAv3?$&GH#LoSq*$y!KKXj|M~WWl@ORruw{_tevc<I9Eo;syWo
zFMr4PKmEdokAF?lL`gHv5|_%3i;Pw)eit&td+v`1mU-fOKG9oayKdxVW?5!xYn-0W
zeEsrkmiZ81wKYC`e2-WsnUQQ7#@>6;q;c==2)GcGcvmuz$Mrtwfp3q0&N010Hs!ot
zc(=aifBgP`;Qi%i-uaJ&*kUhvhx@t(3rbODW>PBTX~w+Ky^*vLWqTP=?{jyx1JI#m
zgiI!RfHmHnr>8R$uw-TO&LqxbfpjQ5eDzgGHJ?tj7LoNY?q6}Y9FVrlg?m`}<jl5K
zPHUrjFgU{U*r{G|F>($joSIW|q)E9&`;j%1-6?J)58t;oEx1FvMzhtL(z;S>!mA=(
zkQPAm<M9sl#M}#8jWxGBz8vR;l$DR`m7FK$V<eb@)2yT8bC5`FvecOzQY++^ne#-c
z4YP#O$6D0^LZ|ToBKG&LWB+8?e@^Qdp4alT=qT7VjbM~rYa=C#@R15qg@+U%%I=TW
zSHcWDC)5gsnWvx5^mF6Yi$|6;^H;z7iJyOd$G`db9VN{imj~==<JxL8)C}Nebdf0~
zrm0}wgT?sq5`hmIQnPJ~q)@XVcU>l)KCI!*<b>q#(kOE#7bq=;4Ic@guPj0%ymc0o
z$%SLi+~rIK77DNOS4@4u>Xjc(-|@rx1OMuWf5FrCjtfuhOz8c*?QGP2N1j+nsZgew
zYKhGPEU#^2$_t3%Eg-uj;ftertmy#-3W{jR^`1VS$#-`g4>Mk`&>OEG9!ZjUnCI}v
zNyb@>t9O2A@2PdA%Yi1srEVlC=v)Fc8ge_;-C!WR=)}XOEW4AtQSwZgCbAbCD5;W6
zNh3+t9HuhU-N?1kNo*Ce3AGFT>R9SX4v_HixR5M!m?o}GxHd)z03oHKlT;`rv)l(O
zXsuT!g`Ab+-Hd4FYT&h#%-Pd_m$|Ui1!)sb4|dpXEe}Z-V|MbZwd})7>Umv!R%?0G
zT>ER@?HXEOha-FhvNyaswi*wqP*SYn<SG0EiT$ML!kmwse!0SCyn1+K$umFx`EPmm
z%X|K?eMe52q7&|wW}UVVlmzB!;`Pf{+}+(#N@hEsscXer19_J9_ufcJIUE+=T|cm`
zD`p$g9c@Mu4W|0oi+hge(pp!1FxN#Qkr!eJIbR!1;aFx4dPG~HdSmqyfBp7f^WFRJ
zc)R_9&CfJhM^o}ipoIgjyqF)E^TIOUkxOR2D?FXw@!|9fNrRp31N9eABh5IIx}sfC
z3dsw_<F!S|t#G-lJj`>1MvU5AB>>vXdSP(5!V_<|*GOR$FD`6r1??0YP}`8)<)k4C
z90S?-Y2%q>5!#Vdp^brC9)lnvz{V|7BfA?e9Wh~B8`D&1J?vaaouV;>)Kn19lpe{T
z-Gr^zNPU%%`X)&O?CTgE6GL;P5zcb~G`tY4v#wkCdupPT6k#$}K!PQicHsFo>fi?9
zncU{{dno7g3HYoQ@|WX(@9_n|`f~iqjvWc#*;jk>=^hiJM)(q`b(*zbiEGpps)TO5
z5A)NcVHVts?Ywg74VMi`@Z$cF52uZJf+<IO$1EXDh$J9JK_f!~GjS|4Srp7TUrvKH
z5UEz{Ws9jvhnW25;UN$vg~0*{KqG1BxK)+1XaGIejV56Z+?&yw1zTjGX_>-{!`Cc&
zplw%ve)^6du0Qau{=msU&{Mm4P%tK?d#Q3P_sn_b&Er>0d1jf9Aqy<d`;ULax_v-2
z=5T6qW1A(EG!ltO!QC)V=<tv56q2lITEf%D;chvwZk0J`Hn+V&fd>hr)o)*a#jA(=
z*yJLb2+{DeFb^4$={|LisjJ^mP2;c}SeAMC0|=8Qibo1dGz>x-8(;2m&a(@-3#o$)
z>EWftWYs0Iu{KHAXhoaH^sdx~dS@ynHuD7Do$ehiLY|a7g>=!c;n(FY;@C@xghlT*
znq8q=jfvJsGu!vsoqGPq>j8Gq>+{)V?FJk@zusU^3LT-j4CbLz`?jy(!PZDn36_C<
zV!GLbvVQ$|hHay@E3VG#$CqqdW0@0EPGr$waakDm27@3Ga6@OFXCChEFzZ~eSL#-Y
zaEJ--&eQuR+S-Pv%pS=+$P5;eUG<RkBw{Tipe6kCF1<#|kBr1o>!?ICY$m8Blo>PQ
zvOe*{hu`!4=?6a4pSb$dXqEQ)e>4m3AzOHyqCt3l{}t0TbGW<1WTUh3@$`GP`jPO<
z0=(RiUzL&G*rNI0xs)Jy>ztm>JTCKKP&f(aH?Lo@wa(+q*VOJ@?l-ozQ#a$V9GP;+
zZ`C_t=$t6?0pE;jaRF5l;t{OThrf@e;j7Y(Wu95)WxN)cG?C4jP^tuf@i2w2&s2L%
zgi+Ko>KO=$a`WL;s3D2yq?pp8k)np7se<(=&{Iryop`NNilV(tAs^Q2;FVfegptxH
z!CKu}8fGnIlE0$6`M-Z1e?H3?_;fA9E~45cC7l3xvJVDSYh;u%JSY`Gx<AuDEZ90$
z8}+*KBu2d?w)KpO^X<zwNQ3XD@0p7-r4+EB5qm(GUXoN;a-!sdsBm337=A|#wrp#S
z^%Bia9$_DJsOyHejY8~Wb1LJR5e!;$VeXyQopn9a)0I{#UD8Mq?Wh=&X3FrKTKk#v
z_QX%8zv27K_ni9&Dp!7G^F5HvB*9dEIlpAi3lH;4rg`G{aL2VjaqaK9oPVVCQ}|wz
zQJX;k;Xoe06iM6?fA{WdoMp_7-WuoAnTKg1-3>|NkRk5PTBMFeqF}eqo+y}eeBYgE
zbz|FDobN2bN>qXgIr{hEk10&^%(5&IOYh<HnGwnWybL&$*jG!xAS6}~W}ddVUp$4s
zrtRJI;G_A@zHJ*Wo#GYoMoID7cIKlK3=~x;gSl~CV~Ba)t|XtrCLw1^DX1k7qkI1q
zID2DB{`=Ole?E)(*@ENQP{RG$_b;s8!oTsfwNNx5pgvf3!O)@0jpbv_Y1hu_{fTzH
zaJ{UUIN!W{Lo)c~haZ`9uvc%MQ{!~SQdM#a#`E=Z8HVc?%u6>;PiL&RVU#YB-d!8L
zZkQL6Lm?RQCF_k%v<=x_5v6%}^Tp5HP1gygkip~efx~>{%Erg_7k)l{&(D`X@S(kB
z<3cAqG53Mvj(I%H4;=D=Z;xMbC<pG2kCdfwcro+w^1#RQk@rvEbKRcU>25m!n$lpT
zj<tJZw+VE@cHnZlvK}V3mSYtNpgYTGf38RjOw^g?3AZS?yBX)pO4pgRh4WY`b65?G
zYZ`urig_f#x+rL{6Ju}JF);8bLsSEd!yy&SjK(!c20aGtR!4HH4ls_87fOGY&)d$>
zwhpO>g~_lBsI{|{VCI!PQBnX^zRMpiGZ9{Bs)>1?n5&`NM0cb04V4%aJE3s74UES%
z{FTZ3&9{<$s_^Cb2Y!43u)Z9Bvg0`)<yk!FL(UabB8hl<?`Us<@M{jRN))e=R+D-d
z{;BDROAY|b*O2&>CG+afzvkoV$}$U6h9WQvEJUJM&`Y1G<Wnx;qoIme=X^dP2uFFO
z_s(Tqsm(Bp7ElI&We!`7*l5oTw$O-y)x<O{k@8hL7kVV#^~TmZB`HfOtldZyULL<C
z)k0m*{P^zo{Po9w&4>OI)vxqi15mT5D&3!_4|mf856dI}$#4FOWj=5`+>sX0+4x`n
z`iY<4T~H0_h1yQ09}T)Flcq3p_k?#Pg@F^Eg)#+z)lW}&Z#*21v|3rldE4hVlfa`E
zj`_e}y!n=gy91NvkSY#2ln(%_54lh<8p6!gbTsnPIUHt=%OMI-VAf3T#%#keNQ5#U
zP|0wO;wuvXR_+28*JZ8X=Hy&Q<C_P^MOfWv-pQ>YSO8chgN}$|???`l$}&CQXE8Nw
zVPaW=dF0(;+{b2HfSbD+!83++yI#-7Ck?Os-GBWL_1KwLH`$}_#-xMMKmb^Hr`tx6
z9PBv>(3BR1hK}fSGSCcT<I)=SBwWt1j^F&*x2#*`FlXkRC<IuU##A{@-JV3|EYr;6
z!yVQ)E`7slMMXI*3)gk!{fCdyB#z`yk-hK^@LW8+){;bXRz_@n0;D&jhL21e8Q#`A
zNrJ7=27r~A>CLF^1K+*-8~*zJzhUJAUFs(QmX3Qyi6^7-Vt&En^2lF&`<EOKM;>0h
zz|u<B3;*N)`i`G|c^?4Q_zru)tr|Z+o@GMvG{9-F)_8hasV^P_z;Yvr@#k;eU=EKj
zzQSB-4wrLfy|$2rh(zX)cN8NfWm=B#bftEK6wg5I!31lAeV0<gYQF(kcZUPX0@Ru`
zk;R!DRE#E|(@3wotl_OR0IXyMArAn{qR96O?ldd0j$OH0qtRGs;4#V1SxH%;cWO7L
zGLech&A~WHEhgt%U89&!^O*2QQyiW#ZA{idJU`15J~{i(#~+P*KaXV${Lxx=a8b-7
z_@gsGEO+Qm>y=sZ2n2|PM`?!0^Ukj3c&ltzu=T=4g!xG*OW`-a{RXd{!=wYi+CxRa
zO-6kXGT~*J1B6iJG+2~}>5if8r`FEt<B7YwhhR|>C`w&dtZqz_nRQ|*CFD?mmXMXz
zRiITqTu#`$Vx<mKVGRHai>6rv1X$aN^YtA+oc=9;c=`hyXM9MG1h|DOh>)}}rHOCu
zzu{qi;LXc#I2;eW`053h{tKu26aUM<e$U%?zr<IIhe7qYg_1<wu-h!8(ExXGdaFEr
zJoD!M1u6l?F|JL})<VbbKzFD00NO>JoI+wNL$0_#fA9z;$tq-%0Z;|2(grYPA}q_o
z;c!6PGk{fqX_vo?(B@3XkV}o>plb{pZFr}p0Iytjfw>LvAxyeg>ois#Xb1zqN}m8&
z67r`s04z}C+-a?IS+6Wfn3IxAilCO7aS3KebjahqyWGe&@Vs2dD|!AK`Fs}g&Y-*j
z(F4SEi;-V<#-8O=Ys{KhGz^Y3Tn^%G_o$6UA{BDGG@feZqQ*2oQRc$8zj=f9&fg3c
zP0?o{`nf9I2P+xtjmv3;7E<v%1q=G=>3x9JQvpWm=~n4m#U<82kwlTioD)cd``BPz
z6%x>?gMlTZh6+$5q=hf;UNDy<7rFBD^(X%N=l_|X*PnUnPr-y1AN50k=lj6?#r%?&
zhZp?W{Wlz^Babg$F&!uF-W+&ZANX)N@IU?Q_gvNwx1mPt#X=skH@A4+h{V7cAE&r=
z&QE79_l2!x)QA5>8b*A+vpQvfV#Yj&2bYM@TIY0H>AI4K8el35Xc_>M4_0Kzx{Wr%
zk5Yst<cTt{+a%B6nT3{_l0${C4glY}Km)7c{pFel*yf~ew`U3-ff}tF-HoJu)XM`*
zT^oLYRPq3@;@NBfT}YN#WI~ccQRaDKsg5$?-e|U>?i37ly<PA9T*a|JPxMw&|7uNt
zIsQQ(UjVEx$Dizobc=Xao9r^Kh^uWPY=B)aqk)*i%w2_hS#Uifen2dT#}SIu1Xgj|
z=ygc6n&GRl=|^6(@c;O){_i}!f6LRmpLtv+vLt?gI>)AyVzRke7_CetG}lQpi{4}A
zJgqBrt6VNuG==&1;V^SJ&a6)>-ZqxBFo`gcSaP9g#@dFrl{#p(MQ3UY^V`WYDu(4w
zYa4b}dNM9J@6SK-!;jzd`|V%xfge~cB7J3WV|^G5D9Zi(f~73HPLCXX#@}C&Lt-iS
z9Mh4z`vdK_f60p%Ge7+J9j!KMvM3$9NBprL$y&(S&}xWxiVkhQCW=nvv@k6Vu^K6U
z-0@%kvwwu<!hF2v>3rpU-AMgNB6Ad>^hRODtWdWp{D-V_wXn-fc}CUg(&(D;RA^q=
z1m<$&INvjS31H*WNG7BfJ~&HJ4)+hJC~Z42wNBqUxg@d{`k`@NJc=O6ByHqgSk{G8
zYdrNxno)JGha=_<efmiD#G;O-g38QNCdw|&YEE7Yr5l=NG=)dU)}3=Rt}Ub1_+`CN
ziZG|+xYygSnMwqJtM)uyhmeoRXEN>G4Su772)~VPM(<8HrI}#cHX5mkt^w3En=v^c
z{Roo53b>$~;<nLzC2u7dMAnejso(#?V~79SKmF&d>xFf_a6buk$^5i+HkatbMiV@j
ziHE}-CJW}B6Cdck^K`zjURSP{EB6m~+}%weP@Ca(MXWLBnbeHjdVt<JfEGNA&kN3d
zIdHDRdRZe8&7hCsGi^ZcJk<;Dp5F2OyFc*b_Is+nqgxxvL5#+6x=m)^9N%y^-Sc3D
zxfcAF73mYl$AXrEE_eLqtKU#2=W@9Y25b2AB!dEYju>(^L0dGlRXg;?da76xlFo=V
z#GO~iJ4BTG!!ZCX6E5CZwefU5(^@4h(cCn%Vdfn{5Ycc4^F};8P&Cb`xA^`8*j=V0
zhxtfu6PTl$pw^MzD9gk&CFaKmM1<Nua&3*iHA;tM#yYJ$m5SG3o-Q&WEg=_Wwa#jd
zY7G<Pe29iVtp*)h+bF65K$j9JSK{HBVhL|bt(}|<Pew|KoC@1!G&=8^Q8JWNLQ3J$
zr0<cf!E`aycD~!-)a}fU&*b>^S|%puV^{VTOsp!=?K_1g%Bn)|GsXgX2v&_v;XUSx
z$;Rp{YEwMpyg~0=zkkbdYy8*0`wO~26F4TgPw;L4Hh4pG;qLAZapij2NO>Za!d5%h
z8@5JLW)fpAg~Q<p4!w4c`H0BOF%>3Yb`a@+F+wrPQZN<nrbLs%w!+on+MVkL)|C6h
zE0WIq;pqq7oqpn{_kY7r?Psc<acj5wL0nOvD4r=^`1<e__wzl=S@849`SOnWE9Ld?
zUUGHe*x~)FuZJWs#n(?l7GVNJ?kF<z&?hEKXpaVT)fL^~m!DUD^Lj(m7#<i%iuZTD
zoH2p{TYJT*0p{uyo^EEW=SJ?CIbBGq9G3eL_5w{pPSs?gOAA?;C)_hjUbrg<N?!2Z
zk!yf`(8*IqOXl(Z5k$E9#I-fNZp`keH?Df-qbS~Dr7S67n{fTmsWh6vW=2Q(xP+Xv
zKh0PZ+^-xD3)5j{Ni$2z0UVm6Ewk9d3w@x>g*0Uj2V-fDzgxJv@nd?=G)>$c=D-Ge
z$FmO5Hkc=(eGDH`Fwu65>oZx)x1q4!I49#?aH=`YpNGckg>H#v2~QI+VKM=QWHQq9
zlwLclS8THgqG{l@vc8{~JN&cX{yAH#Y`yZL2(7|TRp~Cw8pEnAw|EkoZS>aZ*G6rP
z%jL@PaOAK&kaVVZA<4v~oheNeAosx@EE?WWt&LP|3GY%*&{NoRq!`$=-r20O5d+xj
z8~@)Q{|7#te&%icnX8=#s1q)Si4>lx%sedjyg0n%P8a4ras6<{Qpg$a?Sa={KOpmw
zl3(+~-3zw$Le{8R_E3X7#=0VzF><5Eqz?#V_P6Vm*Zzp)66m7GUK*Fk*@s%1f^ig9
zS%ONTcjxW<3v<uZa^mTHW!*NWbcB>?eW8mZ=?J!kqQF3X1SM%OQr!alPNw9Rh?G`1
z9u^ScsjjRQp0<QAL7UhN&faMi#Dv8XtvXNV6Q+u((!52zgT_F!`Uv!yFeZ#bmqbI^
zEF{R?4UtSq6Ot2N5@}MTC~gOO@0_n6aTn%L6p|PTq;Q+2#*jWdy~fGpSy=ex_=kRc
z0kFOtf3m}FTWC!9qJS7N&1jTEt<p&}7RvBqFbXbH$Yyr(v+t(GM7eEYNa||wKujyk
z{gJ==?Qi+ledFD`?^#MFOXmCYCBTBgAn8P7nTv+;FwyLFI<;?Xtx?y?JjZl@Dv6R4
zZLQEdMe~pxDOoa8N^H&IA#R3~NqQvF(U!)b#c_4IHoO;lb1qi-Wqr@zeE2=@^(|L<
zVh=NVPMPt+oJt(>9rxv)JD-_6;g^cbhSttZ;(og4_2VnjT>0hgFJZ>0(WHnt-80=%
z6l)rc60n%aNkl5A!PP1$lM>8XN%O>?fBl+yKJf7PigzELc=~vuZU?L@%}*pVvUI8^
zdK0vVZ&`PvV<Z(^1!HGm=}5|foF}F{lRb<jm0*q~6Ov|RDjesTB*HeI5rMh2A^8H!
zM$I5z5QOd_Ez~tqWv;!Fb4<+oG|>kFqJ#-`PB}b?QW`0W0UD+rhLxIzF)uPqt)UGz
z(WN&$O=L!D*+3p+&@9gdv5$>b_7$=FIQ@rz+)Cly09VHSysu^RAt_6A&o^Hi^Njlh
zJ`q~Y6qB;9xKz4y%mxiUM^o0GuFP}cFW!935AS~D`|XLP2=^uPOADER%g%)mk|yRH
z1(IQ0woEr?+cvh^Xsw0M+LVU(p3&Nd5;4}PcajZO)u@3YRy>NoDHYaj!#8N|H1it+
zLC4hlRIj|fzUB8H{)R2Rr_1`Q5-Q{=(JW1pd7NJGa(O|%tY{bfTJfopnlnkn%s<{e
z(m2yLXe~^m149Z9SCZW<^%Fz=A)l$O<2ItatsxG}6v>*0GGQJdyR=9YI{O*7KD@Pd
zgLWaMGGx<?Vlgg$OuvU@HY!U*pyXhqBo8kt_3%j(53ntj!u`CER5>43WP?j@Bnc)@
zPmNWH7Gp3^se)adO5@^fR|?QmNYR5fB3SQ`6wM*8OesJ&cMwYfLfHV}B3S?kKf6lA
z!h3SEFs_{4SeAP)vd?vYy^j5}S;z24YdL=RtF`PV;e9n2JEqZ;m?c~$+$LNyZlej5
zj7zZM%qBpfg*2r;T`21O*^3vvuPg8B%AAEI4VlU~&#J;Q&j><Y8}l@wMcKA1y&Jwo
zyRHf;X@HDF<~V6V$xNwGFp5=nZi>R#PZCL;DJxi_wUE73>(m}zoF#={+Ar&Se!l$7
zcjxa|b;Y#((bXN6^)ho{;W0n*VtQoz_yJmHs~7rxW!8+yf~I>Oj*rxS<@}*XLU0PO
zWHQB=Knja1#dH|kJ6$_lb1pRuoJH(86h~q(SlbqLlqlvEr;$?{UY2pcuh&YKif$D_
zZ}Llb$E0G?2UFQWJJw@Zm{VcSVTO%&@HVzGi@{#WJS<0&B%Zd0ZY$?rLt^A3<rToD
zNKkdfHe)?kYBFl>bkDdb+Zagnp5i^$jvqjW^eAhI`_zNImb|dY9Kj8fl5>jp-3rYT
zPiyCxjDsj?FxEUVSW4XP`djn+8Ls7LG3%gokJ{FJml5v&8ccihM0dq|Mx@}H!|2~7
zr2VF0{?7>G5oi+LO6|Im9KL#d!PE7`)1`CB$}z|Dd7R1d-!tPcUWc)+5w6f$czKi@
z@;9?F$kQB2n?&8KNzo+qzNae&Q!1&@ZFrlxfmldwt<&gq!n?HAjUPU~<FdWy+|G2b
zqb`d>(U6`pbC>UVwY(xVK|8dz;!-IoF&~17TT(&uf#Vzl8n<WOSqw3Df7HS9M#uH>
z@i+__XLY}YDflR4b|&TyLaGofted=0BenOs(eg&RtZcQ?d#Dpa%9k*5_{@6b0gn<k
zB#t^5N@BPR+2UCfKi9c06Pg5H7DSD6Q$#X&_(yJTQPZ2_U1?kAd|i=}A!oYvV962n
zy$wZ&l5ld|e?(AAw-`L`tidE{s5Mgba?)`xnB`E{WNwLxI%<0t*}1U61QWSgtwgC3
z&$lnfKm6khfc5406CQ&G4AJ4-d~lrV#%4RSy91ZqCIJ--a3~MYCJ94b+=K2e3D2<+
zG!3?Z0KHbq`(LnYL+rrg<BW*#KY#d$#gh{NA#-<k<jtFxY<;CZt-O5knyiV54HVjB
z+}}U&aCc8~4VrTb$=k5a3ZqOn+?9EranGF2PvDK5jHw6*&s=I{Ya5qlT)J?$%UtBl
z|LeQ|mXD_o{9-?2z7iar!ONB?zzgs=z2xQm8x{?ar}Z=4JFT9$ToaemkHA0S<>P_7
zazN8-4*B==t+Tb*By$4mM)OMNN*5!`j7y?3AzrZtr>86Le}2bTkA*T#v~|P0^LYOd
z4XwfbbjSQ~WW5Tt8gHN8<0d3Iu=d93{Ep-Ng6VJw#z=!a>W82xL))A@XLJhc@H`dD
zI%C!&O4y8mC`%aDrX^D5UR6$)l~dioGwKV}4L>#7)abqO(~sZNSEoNEp1kpq&Tn7d
z^Ke*5hu63yR2EcbR2QZ^iuGKOG$wo^ltrlGNMDg$@HB(x@cED#B!JY|EJGp@o9K{w
z?HlcFFWocYgk~1WRG-E1={ja0De<-%P7{WiY#*`gjtyG7NFjNcKNlHH9{P~2WLz^Q
zi7GO{+hEO1-Uq&#dHwjv-CgHze*Y7#H|U0rRI_fK^Ys)-V>oj;Qq<_{6BNac0g#I6
zNVkKoBX3O-S~A%)Wy+MI^j<Lw?=xIR0$3DeZR=cH$R1aW`-g>-ZTz=C{<mD$C(gV>
z%?L*=#v1NqV3x$I<rQBY-jFOImz7IB0}gGYJ)Nk3{ef@){LlE!{h#yIi?3Pk3h#e@
zVr$_=l7=MHcEDlaH(St!@cjc24R&dfj7rEn_v3kDAE0~fE4^>zau?~V200fV9wt=Y
zvRxaxbTkP?au|KD7rKIJ3bWZPOl4*+GublUjNUqKksdlprby;aXG%`oJuEz3;OTru
zQbNlNeT8eMyR%i}$3J{WU!k27C#jsI@|zbgc{m=){vsr-+ZLe~UE*g<3nd?!G$z;H
zg=CpUW*!fBEQdSFa>w=Z6Cd6`(QU!RSlcNE4bwPBGSqA1_j}9YLx%o&u4DN7*0S#m
zX^p2?AI8w;DTW_-{N5(igVh`<evZ(_TAt{dXrgSAQJij_sbr@619~>JUitd*j>9zZ
z%g-OE)quo2ExdX23STc=w+k~fIZaF{4Ea~7sANhyAg=VSv(_t)#?;<|aZn^tavm4U
zsoOPxO^;+=nI`&bT-M6Du3TH?B8{!UWpQeK;t#)kPutE^F2T05Lf0M==<dv-Jmv=;
zrh8t@4|wf7t#2_pk{wURhs#g=FaPU*&*RGnUc3s4@S8U;`S1Tf{{!dCHGIe<Q*;jC
zPa9Pi%muCCk&}EuFq#`{?_BGJgdr-3=Gb#AKt0urS-=P-&C$d;T(`zgZ$DCw&Vd6@
zm&Vpd@}h!UV(ZYi3!cKOMs={m%EVM=$`<>ZG^Kl@OMvXN6b@P<1^TY=K^jj_0U}GD
z5nu7zXx6!I8{hxU5A;=Nmjmata!xD1dHEIhcXv$rk#5e@`W7vj<#<nLqIbb{24?DJ
zcvnOgkSop;ZW*^6VJ;0!CQJ%jGfHc;zA{OoWX1DHQI*{_YELcvJlFCLwS^7CGw=~>
zHP*v`wlIT9JQ(nnu|AXX0U?ohU)Vh%m|(Lc5!5SmSkG6=lsP=!vE@!qE3fPY`EcO7
z??2)l4#y*<WZLyYS%MiiPX}@`w#yY!VVQ&d@3rBr;k}_rkg|Kg4O#a{la-_-OQc~>
zkTge9ou(kkuf}yVuC3EVsRDod_9xD@^V9h)W*f%FE;Ea>mbl}QL%HMC{WoM2{HYRl
z>W=vZ6XWS}LEf+Y{KLd_oR|-VZ{Pfe`}-sR=HL8lwr#~zA_*L(0yXZXh+yE8s0X9A
z%J$(*-M$XVed5;ZpbGOG0H7tfo?-3uoG=x();ZP2W|ew5vHqM{TPMvE43F2%u55m0
zYbRWstck-sG3UZE6{eJs9x_X+Ar?^6G8N`2^ZM>c%85C_$8+U!y~H3(BSp1$IG-!^
zy76zn`-!a@m!_mCk><q1-5t|3G3g<^_}J*(xmIUx%G@W47sM4e$IWRH9$Kw+dKJ0|
z-@kpw_wUYZE|_F!DH7g&+}lDt|85nG5QEWA5I~+u5?_vg_{SFj>&x+{JzQ=LuMQ>_
zLv>;P>2qhyEKDK@&v!$_o#IS1x3}-9W0Dapo(Rh@+FD5@midk{g(iGA{(EK+jOMOn
zwtA(9Q~T}jvpJ=NF=#2H88yeW6K(g1rX^lOc*u-5722klMF`}iP{TXNBBWzXWh(E^
zKXcu#RJJiuiiz3amXJj`l)@n$S!AJjB01E^^NP507dGG6t|#hti6&E{_?$|kZWRbH
zMI=DHJv%NY8H&j>E>T+~gWyAk;328XDH0E}=9p$?r)dxBvg}Qs;MQq<!`ucbTAmF=
z87X4UkVZZm%d0AC(UGeS(|$A56Uig$d`gNh6U~gAG}6w-bk{BP__cPPp3YbWZ!=dK
z8==jQqJO+t4`b-=;malx9v<$Ikm+NBh3v^|iW3vdPwzh)iA=krsXupye|IU|*h~BO
z|GS^_AMCNOV}AYaZi!CpL>(K94sQM<dx~a!oqV#6HDabkhtcow?y&H4L6lOa_z6Ae
z?t6^22r<2L=)KdtVd5;gpd#cnOpi5~L8@u2*?7eP0>!gH^plUidt~iQm1m|C1O~n@
z^DwQid|bnXJ`7^_jK=ulOkvWQd%fqF4lFWZ*74pk4?i~ZU|C&Gjk>;q-pQJo^Ad*R
zs@&4gc81UZun=Mmj;jIOJWWLR$>g}@3D%1FFmCof&cui(AD%N2F`q`NpG0%KpZE9y
z@qGdYj3#7X!xC#Z&XEq87~#|9N)cg70&}6ZHcayc(Xozu_~g{B@pQUiRS;X)XjHZU
z7O_zbhm_U(SjVEEyI*0(ZI_?KeV0XVK8vjx+t%VWlH%FyJ9BcM<&Ix(hOpx-f6nU|
z{_a}-WI>P1H759u`B-zmTF1A@fZI?APr<O2aFyCkce_Uxi>V?xvCJjznNQBKsz&qM
zbcCYG`1@clrxcSeEevMe0AFGFbB*sL8Vm<^2HkGLAFnak4*{^p$MBF+SX<-Tj8lCA
z+XzX|bF(A}v&<}d;F#|?q=i{Bo2|6gA~Z!nq+$Ka>El}thm80OrZVH{czf+(P;Uds
zij&<3Yb=R<!d4%)3&ZbdM^(dvD`fvN(2Jyd0L9K4G^{(9b)%(*Y>hgkMZ0N#BtiS|
zP9pZ+B;*l-5TOlmD%JG%+)J)xSF%d@R!y0;rXleWTw@>gF4P*S(WmnTZyE80);rZV
zoWU#`PfWA$aM~>V1-ol_@OacA`?Gi7oZW*xSDUd_i(mjW7#;@hgQGIe@$Ny!@Ni@|
zum3F9^1kE7@rfJObL}!dqkE_!?DpUaDXYjQ`||BX?7<d_ORx{SIce<Kk`&FEd5$z_
zktal=Ue=_MW_?Qv9Db6ww7}tWH(0&C*Sxw9`GVOiw`YLd-s%3F+5>QIHU{QlW#GY(
ze!8Byv`)1ZyaB`b+l53XOW~LwIOHR<&PZ?2S{R|L4-uS!8LeG7J-tWto_Ws9(~Rpe
zUjHZOJE?xc3VOaGx8Ir3tp6HRk>|H4X<UyXOB|0by1{0BBq0w|{&l6B#ePX#Y;(iB
zM-a*I##B|NQepr%!jZ55Ds}W$StHPC%9&i0W`(tSBp>Qt3&b;G-74#K<8--3`k<Ha
zPwUQ}3K>@^p8eg8tyb1<Y$j3Dh56x-5Qdszub<7F>sC3hTl_uI3%Kp|VtDE8_rvp9
zj%W0~Z{H@XU+}B{D2^`x)|cZ?cBt<S0PNe)Hl{;;fI0hC5FnAp<~j_~lLvc4_8nP<
zjBuyyl7c>u<B=Oo?i=^VM~<();+Lm)e7HpVj|kz}LPXtWP2?Pg@#c}Dkra-HBW>=K
zR+y*qN%3QBeIq$cTF9E2bYd=<IV<bsiB?+}d?ToLih||DQ*ZS30#`>?rBdm=aoSFi
ze3kO(jQbcSD8(LloFDneuYO0rbhf8Pl15g?mx^m+awnN1-MO58;>RB*4qrWz4u!8@
zf6L);;`-fpD8_Tnk=;x&yF~J!$wtD+nmE_W`;TXS`}&@w<z^h~5)Jjk-6Kzz%E>FW
zU)cJ}G$%?4kg<DXYZtcGkOoCFb1vi@`p|BPO_WYtFR@{wZ3(}EY(ln~+@aL)!Aeai
z*_jq+TJDjOxqo+|^@-XQWq}UYb>(u|`00m_9Oip|^UZHq{KB!Hxw~JO<_Wd%&Z(BE
zJ@K@Lm2g%i6Y{oeHCpqCY;PSw*(|Y=s6EjwVWUXDu8ou<{m6aDJR}OheaHrTQ8}t^
zk!O(OGdO;|j^o&S`e=ea^~m*6fCYdO=~Zd?PK@`j3^{O)>1$G&hsP7$Xq(fXuIPN=
z^}`)cA3xGA=bJ<UI!O|<PGNHDjkT_rNBp~ra(8zZ#bMTvi7IFg$??`Z$#&|!Mq;My
z6YU|B8~|DxQPw$y963pk`0>_x`}Bd<;o8Ot$WgeN?=$|)1K;piUh=zF|CoL`Q$KzL
z=^!!57BP}^XvXzixqi6k`t*ucIxd;}LwLWPw-bVoDf<pC3`VLpl1upim`8G|2&p8j
zjfX9UJDkoR0<?BzYtcAIk?)bRI6z>ROv)3knKn%B0o-a7%H0AmW<~XYloIjhqOb{4
zm23rRpjDWgP`a`dXIdba2P`KJ?@lzPkQ2Gmd*{Q44_q%-e)@3XFyHg-*S}@zPaNt;
z4tEQADM<H7qpjPJhGrHoG)gf0EQhIe_ra_%=n=}%Ou1Y-=X1q|{9@Z+N(MCcsEX0D
z-RfZ<h4SaRj^XcL%X`0g!+Q4jNQ-BG2T_moI^@P(xpn=0aD_|4ld|@Z_#BdQZWaTD
zqR7if4(W{MD@A3jWg$uU8K<Qn<w!~b7&*tlq@<aYoiam~iKGP*m^h+2TJSE2CL~W-
zTk*CtI34j$5+#=j)eV4@GDYApP4wFM!*@T?!d;l%0DSOCafc!^e>VR!j`@zq!z+9{
z<EM`#bu<e(wMa8e3EIZT?|$KKO7vH+#yVAw%K^Nz*-FwB4hfOf!a5q~lq-m{=}K=M
zTjS4%!vSN(YQt*Cf~Q;p^mQE|`wmn?MQI)0eY`T+L_cg?tI;jies@?KtmZ%uf4>Tc
z$HLulVoo`{eJ(cE@=WRu-I-@)CUG~xya*!qYzq08i6#Y;Om7o)Ret{Y9oNgsd4pwM
zc=OF~DD6E<OU#EEnKHd?^k#^3)@r={#V~!vbhu-4CwnJb!+Q+atW~l@uTUM{zPs?l
zyAvBi*92R}$}XeoxEQy@&Edfcd93HpcP;NnNPkCjug~DLvj-*EkQNSqNA&igBF}rb
zEAAO2<EnIZTmuBOYV;2$T84bM=g0(kL-l9WoV&xqJmpw3rqG0)k952{h6*4DU~<R>
z%|f$Si~AErG%-y_R1JnFoA(%IE8#aKquyv{Y`w)Z-r%xzKCX?Y^N*2Ro%YdM?CGF9
z<Tu>s2mb2yKgXK!;r-7f@9~i${jW$TiLn&0^&@}#zrN$QfBBa@zPM*j1@yo&O<1qb
zz&ImWo9BJ`hWzpjir#?L(NAcK!OfHl^Z;v=(bqJnZoGXub6qRbv~X?C>2#slNFDZl
zSZHHF+i2CP-FbC)&;8-RlqNETwmH&-Yzu`5teHtHD91_3bA~)o&6#q7R3e?QIos9v
z>F4*X*A-2fX_<L-d`Sp5EAui(YNBX_p?EOWf4+Q#!x6oEiDGODWuS9bCAJshdiE|n
zt&I=Y$|jwLLsMFJdhb+X*btJ_#HWz$&WaxB-7uK{lP$L|$A3V_7Xa(a@h3ZeHIwks
zaE%Qp{{8vah~-rvjY*#mX49B95=Ef6BEw4ukpURRL(3hSLG}^%EJ_o-jXHeb^}$Ap
zM_?`^QaukUwF_Ajnx-f+ak_W9#ne~WSybV>Ai8^b#Dv;io*AtJcoUJEHXavhH@ZE$
zEhutJz)C6<nfPk@bMEqyY<Iwwk{X%tz7Z8jLa~@YPYPC7p5FgLewfjzFi#7uhyTwF
zIDk>%?!{f$_ZoY`l7-pk^-60!=)Q29mi6A))|FZty?2H^e0XS$>2EjVx?WjYV=``W
zzWxm5j0s(*cY`e`#o`1A*_~uU_DC9(abB&5^iU+G!$mXr@X;a|c;ZG@xW7Mgm<}9D
zCd<r36whTUbd4gy2<FbVZe%S?Sx9b?npC6lwekK%l&y81Y-Jk^5g&JVv!UxBQk(r0
z->&~B_qhMjuKM}RdY{+vv+vdOI{x&X=>Gqk`87Az#`8KJX5c=G4U=a&<|saThwK_G
ztCZ+D4Q9-krHc&U5W06#iKM}l2XkO2hRDA1p;^Dxl>t&Yl60h>*?4WwO?))(7E|TO
zj3&rn9zHsKmz?<X)K35vW<4;=k+1Wgv7{rkn6jp{vo|HC*$VSuFnLybz4GDR&*YbP
zq$P1Y96_9Gy}&SC9fQtKfR)jd`B>|WXW5OJvVWcrWAjm9OZ>Uybo1SChi$9$RFJIC
z47GjI`N??GZ_o62z69eAyjcKOpU5^NR$L<0)nu@15=Fy1B+>wlYH_`CQXU?TEakwV
zgcr^{9a-jyJmp_C46QfT>xF5VSWMXHQPNr;)82-8_^(*^h>N#K!S$U<IR+{_+r>O&
zs=uy;pZ5{|=x;xl<9}-{@AnEjYj3}y{rMg4dl>@=^V@azXn=?R$-e%vSZ6s0yXP`x
z%*xFmf4@)DVBiJSirZd@M50#&Va0hjx9~|a5S>T_c$G9*fJpd4?XN)z#*x>39m_Ca
zPSozKZBNqTlXO3Bw2wP|<dE)pGk+6?@ZlYR;IW=nnNp<tl>}~%b)#=*PEWtUQLz+W
zIweokS_dJ*Zu?6gtj6TPNDdV@T>QqAb{KU+JP4za)9`qc!MY-zvYkj8?G9UOoGur7
z>jQ`mw6a;O-;`o+9d|{%e<xVy&5`EeV>SLwM@{7x>v<|E06~rBJMGdYcT|N^5_b;=
zmU7@&GE<&e?7+Na(j+%v*l@Pq_^@6%Zb!}?F0CVp@HXqj0Mr9;>nXe!w;mt9PgOi>
z$lf|#gf7D~>60<b^Sk{_*Yb{$_*kC%koS|V==O=D?)~(wibT@)&d}YZ0V1F7-6Ns3
zM}5}Eo>f%1q#i@O_-2$`$T{8WSotL7b!RW99BgL8X&%Y7g`p1c7%d0;2av=Lt;L=^
zm`1&0eFtpD^GSAZQXA}@@g7ITpLHh5%**8)?(+lL3*EON$sgkiiRVkv!7R)vytUl%
zUa9MaUIH*msgQG_%F1m|km2#_QAfik{f&S~czt&J<l^~AG92f5XYac!&794Q)A_<y
zErJ<nnAJ}pn`h>Aq!*6o@cycvNm`+4v<ThURS_Y`*jcJhNxS!L0!<(pqegmn7ML>g
zG%=-_yD5Snn&wcc=wMWhb=<wPS?5|ePTR^+CnSgOaVTn1JZn?1t-AS0rU%z?R?>uf
zrPhs{68UD-{>eRl+syW&d*4^S9RD#KUjVEx$DiyN^<RKK9u)D`q1Lg1D$g4y2_b{$
zFd5z{$daOP9aAQ;gj#6XW3tk~7;+cb1agCEGr9|#?2N4_VpYNj770LVIm}22Zx`83
z#Z}2^qMOlrB<UGOBKWSYNAWmM88M@^9<Mc}c&OTFc1LrU2;*598^X0VAZWg^*=?8_
zu0lzfyK>JzfAxPsdgAoq0~yH6m26>*TuKV>i|%9*maJU2Ge3Oyf%)Y<<#^<7xdU-h
zvLRQBqBLXyc~HG0UcQTut=>3a&d9p4tzi`DgT2r(uGdbzZfr)-{JjoY<uHm&iLE!@
zJ-z4f_%+9)V;bOt(ZkT$47ZNcxNaN0MzJsE^xDx1r7M#Wz??*J&?-oUg9|E&B@Hv$
zxC6KvWQ<aT<6nHs#KePVe9HKAM3)KAiKix5jfu|Mj8i>h%xDtK8^vuW9^7g|)x_zt
z@#AUbX{)T}xGF}Zsx%X<TfhcGx}`&k6=uSp_lElEI_~bYX4D3)!m~t!(bOtB%0e5T
zuM$Ai^Ew`)OzB7)9tL9^r#T8j1Cl8RQ@mt0E5rX_ye3u36iH~^I?Ka>QVLlk?JspO
zGd8QZ2Vh~g;^GtnG_%^#G_foTSOsfgNN+wSrC_yUYs0oa*i6oPH7=X6J|lQ>7*GVV
z@-ThFi~O4Z?Ee2utDWzE`GJz0c}YW2EzBj8)tEF#F**tD@`2y~+Y7%v{8R4lU-SCK
zYnFN9!_&Kv%h7T50bq@Zu7ttzE#7)O^B{^r*Jljk6XgL(9aH15OqdmthX2*m>548B
zS~9AIl%^;+*^|h|+cl^4Sif|O_hrzxj#eReCA*V`jN2tqn<7;>9F#1HIhFAB^JwB%
z>*Nem%DlS!mg0qjCA1uovS3TblX9+ckGMFOy7J2}S11$Z?gg2RickW1ABAxRl&u+U
zZLD2rp7E4%60OS(z%q|Pg{pBgu^;W!=e&;L-?NsVCn~q*co%_($C?L7ENP^w?`t`t
z-=&WyosK{Y8)KcSER(XRa5;9i40b!OgwJKlnDiJ@DIiWRDL_z>fr*D?$b7@E8%f~x
z-7zG>BuWxe0xS<kirl`xnNzQozIN!4M*Z-#cHXb`dDgIhnGmUin)&+fAM>(&#Xmm$
zQ`Ys$U;p%PD6%o70nBCLPQw>0s|<ECw99*b{JwMd^*7|>fq6RM?p)i3OQc^$7(}L7
zrMJ#(og_|ja5K6JCQ2s_+4GQHiIVaJyJU`FrPvs=trRwo#ME`&czXMex&_D-k71|v
zj;S+E!LVVNW_N3Y9qriK(I#Xs2!?3XWrnhCaW5U_!juY2E|^caPjs%FG3F(4Sd?XY
zWb&E2Rv_I$&a^wB7h$_7Sd7xn^~`r4-jVL^m|lEM@+((`X@pVW;gi{0qo+bo&eepA
z2_}0mO~HDj){QFRznD{m%Ivyp{RGJU9FINZ=LY_bzzX|20M>0cWglDugDsn*PB9-I
zY8vdP+lEnaj8-eIpe{^PVbaX|)L2shu`0@8S(v7%t0f6K*j7oxGZgnR@N}v-2Hlaz
z-I~G%TU!H+CMi8t8a38IcbZjdZBd)ju|XZxl-bN#t%ZnR;<Yk{E=zu7$uIfK7yq2Q
zbVqx7k9B9s88Jt#2SZ!oSTZ@mG%4K_O+c%hKD>it!Lu;WGd8@f!j>ZTSxW+aFg`!I
zy9ABB)s?M}fsV$75(x!Du=V%BsB*29r)}e>x4!_1HxJ*iwUuToNTR5cc1fLtT}jaZ
zjv10eVbn41P(dv$4GNGQ+&aB?dTrz^EX%}FV9FD&5v0|9W213*Ka*_YMOjEbQS1&Y
zPA^WAxU$Wi=J5n-7^?*s?N3h>c|jwy$8yawBgv7{V}CYhL)bLHesD@Uf;@4#TqyI*
zJbyX{j^FQ$^xIVL_WLi#e+<VL0PD-~Cp*mhtrqmL(S(8XJ~@fMAD<1hNTC=TU36sj
z$;LY-LtSsCi=&qt6Q)k>E96YMpOCi|uNS&C`she~p!6wc=4HmG*imxIVGzsC#0bOG
zEa^6ZDnu&Zt2~D}Vlk2_>WQ^A%osjO4!zTA<FZy-?a+leC+4ZJZtW&BipdG6Oo(K@
zz5gAL<u!Nug5EZs`iZ0~Nh6I=Jy9%&oJwOOi~+1uQl@Sjm!~Iu@$kPIr;x!~iD`G3
zwceOn9!;r_Nl5KIW=J|bMI<IFSb#iX@~PqPMT&&6q_;G>4LDsZ`td?@AnhjnJ*e08
zNSaZ_Ft!$8mRMYaBnl}u!cAji&#l60Y^O-7>fRB_OzDUwhh(Uq!~9<-vL$qLwr;Fz
zxUxu6n$A=+*6W2+Gk&awriDD+V+m%FFzt3QIvv;&Tzcne8`S1>>u^9yOrmQ=^Ms}}
z{0U^R3<j%b_)F~aLPiqe=Wu+wj*(b9Qu~b!5Z5+7K?befZyoUXP@sMCo;?h<`ygsW
z#^b2I2i-1dq8ujt{RjHC24g_NbE4!FfK@5v65cW{&f&1cM00I)iz%<U-<rR;YT>8$
za6E9x8KIGqqIsrm5knnGlbvct+Zw%E$do+5@J~+2%`YO1(+^}>_~!6Wc#&T+xr1#K
zU755oD>sIh`rsy|g6zzSfcqmpyw-6ABpPyydpJOdNYL~|cu_$3#TZC;*4ntNE6IY5
z<iq?lr99ZGI>wz)at;72DXkgrKVB$(VwqZ;a{~BIlA>{d+p`1-ro*~ctPh`>!BpGH
zJeC3|bDU|5M5yc8+13~gxFk|qAUPzMmx<(^NiwM`Y3%{DSc|<ODHWl)(X6qy%7^vJ
z;rfF4vQkRHB#}KpH3zy1Eh!b)dRY5zLL|14P?>_wRpa><35L@^do*MkWAeTG=Y4ML
z82-1{vfun=b|1!PxvKBTS47xCld*S5c#RRDjUyo^M{qZ^J5x8x-AtJ$`qL@ym3{vs
z5-R6uV)xuCb65xzE5XiO<G%0=(tCJTdL{{2YmkKFJW)nyLfoHmZG7xmy*qX5v>^}g
z?t@)1)?>WU@f;J#iDQ1jlJ5Cx`bXUBBkSoGYF{}_Msg?~@<%14yd5Q%5@6n}R58}`
z6GwkT&WVTndva0UodS$D>nIr+0A4C720Q{?qQVJ!Z^&z%04m}VjO*aTf5(F<(+Dr1
z_ty*OYh|7$=y0wZ&%QTtzm}AOsntA~i5{%v+FHmL)dn+0Ze^GyFzKip>$ZU#muzuO
zF?@i2nkRBLj(1LWr-+i9Qry{kXQk1uAth=y5B5y>7)nlDt?~A9qWF=P3rh(6NjgF4
z_(5o@G<Pl-S1~*#go#uRpl7a^CzknuOd%J<6$Yy^25Lg`@>#BB_x|ah#*ld2E@Lm=
z8~GTRjJAI~w`Hgh7@m*fgND~);9&wfzL};`L-vkJFs<wy`=Wbz*xenE9EvdK7WZ+2
zd6_9%Xlt<f_tajq2v7-Ik9ci&hdU<K0CvXx3>pCizR`R5oYmH_9vwpK202s^y@!=m
zylX)`k;o(|yq<r<<MJ(c`bg=C%X-Gyn9>e=#4tGx)ZG~lcquVup?l}U#}C|1FDPZ^
z?s$h};lsQ4gr8mXF~OAVgIT464P~H?YGT_OA3t8WTY7-)8dv4IT`^Ov32W_KTIXrI
z^0aNVv5!`Bs)rmtrQyrzQPb2O4w0sVDBw;VHFet>SyQkyZ44M>klrPd0k<`#7u}(4
z9hFh1rW9bLD))Cw*r!+u`v(sX!CE_A{U+x|7=Bzm>ZrPLxt@4eCRTIiq-0f&i3obR
zu0v@ftZr1CXN#DSNLw>x8H{^IeKC|dH-EtWMj!Y5XR?gH9RCN8F96n;<4<<%MFLJ3
z6>3n7eRw_jnAYwOdS~AZhL2283fP^pqEQZ3Qy&fMr_^ZoPWLP1JMvvY-!`aMnnkiz
zd{RuTa!yQ#iIys^PMOMx6YjwfG7Cy>N&#xQQ%Ed1^YZ?llAt6XGMUUa*bw_`_1*@U
zbfFEs&y+HA=^+99M6w|o09Mu`-`xEbucoiLYcII^JLo4=Dutl&t1H=ZBsFSO0HzI1
ziHStpR?bgPD2c6YH)&=3tY-i#ir+S*TRoaILIPNOTn9<<{=;~nMs44H$oORVT7|cP
z>L^a!X$Yrv!{khNK~j!$;+<|4m(c&Khxz^1D%WjewwR0R9&ry;HLAeW1QDm6QY4qq
z5lIx?8{a;dKFX36t%}x=Jzm;IvW}!gJ!G17w)M(sZTx)QDDxxJv7(abNe5Ng>C!Nu
zns71Wv{jmTuv&(BJTk@ms*RE-RHGr;yF}loq@ytmYW_|F|2%ftKVHWH(mCN5GghV$
z3_BU<jD)1aVN|Ie9jNbXHhdtOfr>Ly&bev$Y(wQhITU!;>DwCqQb5T{&Y3br$U>f;
zlgE~2CZTNIqv+OA=<ZZ`0)L!R=3zNd2;<hA=8y)YjA;)rwi~qusL`8;^yn5zAc*cw
zcmTbc6EhFY`oK5KKjuYwBws$zJpintohg=ne4PgyOVb28nm_AEH>@}MR=N3^^!Ta}
z05I4PgjY{z0HrZa_Z|kS=XK?vJ=~l|!=gFZG}=cK1LR?DZ5r$U!_%2r7UaItEc|0c
z5~|_N;SjRI;YreZgEgd;;X5>L0b+@Cg(`9F=DL9=CY-Idc($sdc|vl7>`YoHBqkTM
z3VQ9Vy>p=sv{R^4p?QEATV45hePX$snXgw4lJJzsgB7M|ypF7F4%Y!-)d2#CWP}N*
zv~3Mb!p!a3?u+?0SlPko&uksTzh^D~5ijND{x!g>XL6t1)^ZG_hTOcz?l;Mh3hj*H
z-YLwKxzOJm>$=_K*+?`<^HgpiTA53Pth53>(hOY*BRi=`ckqI`2Y`ied`M{TNHY|$
zUCwU-%IpCjtwyR=NLE5B&L`@!SZ)d_bI7lF%wO}m{3S<bF7N)|7y)2OcT$((E$QJE
zDFS&ylA}4ZUN2lOC)_(ZDfjpHXbyIZnbS1@f+Rv7>4|$5OrwP)9>YnP53kP~3kuL+
zk==}tyU+<~^M|J^+t!(<nYLAIyP{H19j8&CB&E89d~(zyt#@jx^k%pYi{v2<Hk_7;
zdck4cD&CDGjkRtO#E}FknN$iT8FyJI7_!;01~9H>TzbRI0ualc=AC98!;zf1^ahs`
zi!K4OB_)eeq#!vMfVC&K23HvXmS#ku<O8&o^|I0PGzLmh-yjLxFt%U{f2M0WgjWNg
z8YOUe2|eSmu`SEr+s%gxRBq<&;%R)|?x6_Qe6UajB-p=7)Xvt7y46q^<N&bd0+X(T
zxhRn?ooCMN5(6z&%mx!*hi6-hA@tpHpr|1IGKNh9#L$e}8s3asrM4daxnn4$DSV@>
zL61iPAgB-f2tDw6`IfJbf61{wfHm5B0&N_o9Fpi!bB;U_#wkIQFinyAcm4Rp;l({k
zmAm6zc<}uas*t3<k2Z!>-7toyCl=+83;^rtOvy65HKR7aZdcqAZaKmvYU8P{Jk>4I
zBqijy)k6ufrxA;XnxVCgD)shAHejob>$*{=PDw`gJOXtBY_p+m=m^^uYjHwo)y94q
zEI3s~lX9F3#X&DI#_7$%f}_<z?h}~ewHia>IRGrbo{=(hHs?4^lvHQ});jBDW6qhJ
z5}U<PK_rjCXa-Ng4jcpsgJFTN4_oXR#1i>C_xYFOANKJD!1{9h$&L=U%~QG|CP6b*
z2I*%^D1c|5oM%dWfGc|f7VQ2_V&V4C%t2!clsB-{X_K*DKXCbR<Hf@PIeyE3|F=JJ
z+UlTP8?9Hmhkuh1AlY)55f{$&63J+T_2AaXw^qhT0>@>dsFSo~Rk13lW>WHa;PyuC
zoxTkd;ohlE&~G*I!JpVHCLl_tNa6MJTOP|BzTq!e+8s|%-_hDf4jGbUa?dvjq4~&4
z*$WGWtcj8mYrSxN+Q_nC0uK+bY1>M@yvOM=9r0+yMAL1;o<;&-t5F4+DwK@309=Go
zbY$J>9#IGn&oDIf{pJZYbEzAfeZZ@bdnV;^&30b}h7qS3KfZg<b*lV^ZxED~NP0{%
z(Wq26THjFM($?@0YeLcndPeMk$i$YE=7wKZq$yG-Hm`JQY=EvDdL|jT8C%<Ue?Bu^
zPo(w2iZW+m@+s~{iOF<ru%%2n7HA#oZA`N=nvQh4@O1h>cVRK-a5$oxMj<pPMxRcG
z<ayx#IUIlAI_^UzbHjxhbZ&<|q0@(K*axcdt3U^jM4UbHyQOI6x(cXasZqBl>MxD^
zx$xDiSCsTUGF*=}g(>oKn3)d+lVD7`*pQ-)^v#k(%;-j_j16fO1@4xaBF-e8)|KA<
zme6M%dW$r?)@;Zjqd{2G#9e{j)@WQUlXYg&g+H7Bf>-4mUfBcXs=Ryu1LhmE7P2YY
zqdRa6&%timf-*8xe~BW>dOg$CsnXc`MxFvF&1zJ8PJTp2QXEQ8iJek%I$!wZmk*@J
z!eLS*55Ou#N?;ZxsnMj5w%DmAthMp}{fWgV<nTnRb%@9!MR2oc?;HTl;VV^p=fnAo
zdSdE@VoKi!6q@N0tjShaw8N!MY`sGd2awVm-h<ItC48&av5v8sX3^j(1vO<(iu6c4
z?HHOSuD$a1`~lmHMQ84&Lea#OC(J9A6;i@=VoR{9<0%IVcBG@7P8T>XAv5QBhl*Y5
z6D{VAKlJ&mWB7Nh<#9hnVB`L7e6T6w9h$W;@Am+@0=fxi9v=5y_y!ub8Iqkm1?acw
zMyY4`=p03PF&~*l*l?OxoE9M3cvR*od{}C4hzNO3w>={USS=X9sw5BD@RIW|u(lzM
z*;z?~)kbZ`d5csmL#Y^>d8Bfgj1*KwyfEh@hvlB%O@GN(`8OPELfXo@egrY5e7N~+
z2HU_PJDx@Z<_6wyRq|kUUe6!t$ym3QwQZnqC`YUX&>u4oAyq~iU-<QDPF&Z<FYiw*
zxd-?@7+8|RyDfN$2`6!_SFT&-=eIxOmU(sfis~0wFF*@JY>)HUeAGdbFbV(1wK;w<
zF10afMOG(wB~h@Xm=;VbPN!}cM3ieguo}=aMH17HjAMzbTv)x*FDud~vKDGHS`W}f
zx{@Wljk<4i>sS-Eb;a~OKl#F`z}+0dIg3JXmG%0Oq6<kUF4kGy@o`UDFVIt1OO(QN
zSm-^C=j85C^y%^Wt>ymdyBIaa^J^cq2EmjW+`V9wyU|AywQDC@zyNT3K)-H+H83?u
zHU?1Pm1=FnHf5P69**~6gCUTMFwKdj2z~sVAhkJrf{><!v_NRp2ZJ{voSE~)loUx8
zzI4^buz9Z~gtm2B?+i4X2nCYi#JazACCSV~dd<V~f|qx1_;&g&kMa`RuC#VxnsU5m
zAHyffu{Q-;gc5tbJ9rEBrVS4+cO#cXw)`vLM8lheUU!Abr*~3#K3=XXZ{NPBl!@+2
z?g3zJ)#**BRXCqIm$mWnvc?)*9$-icw>G?L1xRx91>1b1(lE)9gN1RvZdf;7PmNR@
zZ}DU=BA5!E1r-DDwAG-+KCrdnI-3;9A^0;BE^X!NM!#&aw)!Eur(yLXNDQ}ke=;+u
zo$KjLTQ_80@RGxa3#WUfZ=G$60Y^?oL(o)6!vOC<5Wg?1^{6?7NEvQsh?Zc7S^w2J
z_Mvq6Wd8Ey_>bWD0$_bP{$$6m3WY(!xDDus{NPqRjHXvS%(<WU0Kd_n$M^t<Nsfub
zU<=i@M^ihkZd7YbZ(ea+=8y;N8~mW%?n!(;Bu}}F3B_RlhzPyLhLvtjcWfRdCyF{*
zU~7>;5|rrCoI3P244$2sP8-ZQ(Ya-207{aHIURW{uXr`T;Z7dOz3{aCTg=X6-BbFM
z<U!r<NKBY-@RN{6jD2ext%Gj~$%%P7qAs-S6EGr<>6kkzdB_iTIz7V!$i``w8>>Ey
zC}UdgJ&L_?s>0s}p?4e9>tH$6zCxc-OT(OY7}5F-1EZVsbh#j1=t~qL5uz4=lZ}GY
z28*r|6916p8A1zIif1;4=TfVK=Bz^m81dXLPIkfVnF76T8<$!+wGGM2D$3-M1yeMp
z+o1rB_-IsbGE~u!FiLN%TOD$?HYP^Tt)Pst-d@=3657vYVD$H{V}D-9@pkrK$2w-j
zu76?@Dm(0QTgNhD+()E+camu|q0;HC(yz{RSU4<U(!AgCdNYpAq@1G(-myX=4N!F$
zkPoKj{`K9XT`8$BsWS~@|J|H_U&Arc-i_J^lhGN;Xoj|Ju<8Q9I;4AEl-InOzUJUF
z-Wr$n38b?u3u=T6aL6j1V5D~E2GXm9#Ie?udJTp`vrbBROv>vZoB5}g1aL;K1$Y>-
zpUxLv-RTV+6A{ofQl0j5BjZ^J-HqBh*L8#I#(Z6|9v*kQjr(rgZsvoL6G0m0&UM>3
z>qZ(WNj4Y&&REAzFv|>T*CFecVe}rrJf0)=-3;oATOty}f`1Y>qmV{#WxI2TBwFvB
z*9+Pv)})}Bk_cv*3#<8%ZX}vO7Y9j*7(&6jux*uPo^C9LFajQc&nM>~o?)NYI)?v^
zwfxEZ*_(h+h~Dn5v3KlXv@y)8*yUFdj^U9ghvPYbA(w`=ieHrp<&Y<`>=w=OJ!LR9
z5pEtjE-k>Fly694$c%lwHW>}5Mxcr?m>c8m`M!<=SgSo4pGe$~y9i6iI}(&EGfO&h
zKfUB-dBv;r3aLWxEruuxvWF8$8g)vKeL)AX8@ccVn>f>Zqi!obLu+AY5K{L{lTMR9
z_R_I2y5kl=p^8vj<8)q0cOA|C={#iYqr*JemD-H8RW6q+Dii6xAi53t>!)z6V;(?*
zs}A;?ht$%8#Z|i@gS~2_sdNc%!*~XCI!YbZ0idW0`4+aLR5!3rts9EchTO<K&ZD@Z
zosdrYXnZ^L+UV)RIc;>G@f4N?-Jw?FyjG5-Fee{o^*f5AH+Hl@0O%>eEca*DUogw~
zY}fLrBml;aCAYtEVE{jjfA>$mVgzs8c)&b2{evC8s{^c$uSpRmU=ob>-YO*}mT3+k
zRraJ_C6^JBW6qdI4|ZehLA%%Xe!K54VIRPvWyqZO{(xIA5Wq;>XGj1iod8~LZ$%PW
zCyx1%hxrw+k6&}oBa0t6^%K^%k#ZUq8tP#--@VfvZWYFOZa5>rY4i&2;fojnAN$!2
z_N9yh<QYgAFMOEU*V;K>F0553Ez@f3qigN72HPrJ)&P;W7JypG6Yj!RYwU^fCW+9K
zU@Z0_>~~4=oEBkgL&DcPEjf9w*M{uhHTFO8V491?Iq9Y#!Awgj>H~ME)`Pv)R`7x_
zv8Th|uA}(~blLxG0p8b*t!|{=A{E*M@i61x#_*uI;Uh36LN^4n2!(LQpwq(}5P1&3
z+F6MH$&QUY|NWQaANKJD!1{9h36D=7CQ(?l)~VZQJY?6v$LA`A#382ad?LG&PmF_A
za1xikvRTF2h8WD=naz2~8FN?#P9Im!vT-PIH!JThyE$)AeYd`G?blm!x}v1KPknq$
zWsPZ?Nvd40k@nS_vdo41<HVKD=AG5A1GoS@wB+6(Q*Pg<_myfjJgZP9o%#0hFZd^K
z{uvMUn#J!q*hIIL^YsrvWtqx&=lf>pVRX35u3S-zLb03iWG5<(myZi26%O+qf^+_O
zD;o&U<(se#^*%hl!YH=6(}==aWHbvBW}79?qd8SJI_8bmD$ShJy3w2S;_e0SpRQak
zPel7!$X4hsT)MCk8vDJ_c{*RQ6{=TC&d3y8Xs41TOg-z`q9HDdiGfx$-J$6f$8twW
zi66F&?ic21(A4wBEQ!K_b9b6Iq$)B@+RZxMHZI!}@6W%`%Og{o$y*P@(%rn+;Oa`1
zgpEnJro!Fb9sk8Y{YSk0@n7+;{-^Iaq(q*lFwE}^iIxOI;kKdfVD0C0{M~h2duOY}
zwcysq?rk8-0re7-)*YaccKcqQ1gCJV7izEgR#9~hB&OUs;?%Wr@=DJOswqGehgu`r
z{;*7Fg4$~^j8Z~#6hO;ij>fUBF-`7HPQhGgGN#QwlSG&kbX{puZ&P2@XdZ2_vLMWs
zQ;K9Qv0xA?Gmq0tzJ2jK{?XTe#l!ZBrQIXl*{EFG$0(~b*j_4;=42iXZ@8XE1fE#K
zQZQ7nPiOiW?q0k>bLRg31$Db}dHQ7lYI|YtsBPd-HH_I)f{&*Y@_pm0mlO9g#d#l6
zJ$DHZIu%60wv}p)r`mbC)+m10N;+La;#ms26lJqIKAfHemx4*7<Gg)(qOQ(C7V-^D
zG{ie%A?K^xXr1Qan=#uk?L9CRLsRF}D{d=GsUg@DA$um9a_trgR%^x^Buyj?sg3P)
zibDC-9X=;|+ep1LSEaYgwl)rPW1b?NOj9NgU=o|@mPgtwAVTlaG{|6@hP$jjuZquZ
z9mC(dCwnId52Hcy-DeGijQE5OTjF3B*(YmxZ;xlz?MmyF>v~07tmUGWC2gEfmEO+G
zUwy-?B_x6(P$c{m-G`LCUATKZGR+0w9IF-~3I?^seP2qZZbq}t?43o0rM!T)(He*c
z@H()rSi_wFfCng~!$U3=1|vW?<Ria3{1aY3e8ab|{+zqZE9QFSs!v$jNOOSl%>vMM
zAtZQ~2wQWb*ACK<S}_R&_S5A9SHDmW547%-X-3@fYYV$Sk?|ZgIvXaPZSG86Ve7p6
z<t_NeU;gEhhlfmWaUD+A_u-vmiKolT`MUDc`6H+GN+;u<dD<FJ>&D0HmAhr;FcriV
zv&N}zT<Q3{;My?T!WOS9=i10xxg+Be`@OqEtC$2EuQmxFerdGSk(tnECn2SRs&mzj
zu|<lq8q1u+v#$wTtHG>X<2_-Dnvt7xJ)K!EA1UP(%nQ~U+B2)CVIgtH#WK~w$&@sa
zBvKLUsw7QJQ?NWgTu#pc18Mx~!7jP2iqCB=?;#x##fhk55<tf8FBfc`F_IaqrPzxt
zrZdtBPq2L*3+6e%(`Ia5krbb!sgqLUw05rVFU&8$;`MxFDv>BU<ve7a8<Lc%%;b_u
zNl7ISpTre)C*?GT$erd1CDB?0dfd$u%9V9JMMy}s+pI?S!Br|@RiJ6e$XobYN@2+_
zcv0T)pFRE;+|Ms~oL{4Dfu1<mcWm_ok{gOW*Xsyf5?utmqXyCpvr27eHjFBr$NGZO
z7OZ7j6PD#jZxQ|yz+voZ9e9pG_F)~E+18B@^}^TRE_4&#zkAQtDrdW3UFkJZ&xLl)
zxE7Mm&|^T<){rTd5;hf%vh~W=SE@i0rJ2$_2CS<GD6|gmta%?8qLHjzYj_uGXCu5t
z#T}EvR0^a4>&sw8_mAL}X(&Rb{D51-t&VyI)~!=(NS>jgVt8w`-ud|cEo)nO@$xk}
z&EOd&Gu_>V)k@v)oXB$-&L+xc2app;;<`29!rfgar;vNP$Isns>m9uPa{R}0d;ze&
z9Dl-Nho*KrSh@L9>~;gURA|{FW%p_Sb9UQppaFxSW31M>TBTA!oy~(H{*c3)_=<30
z<LU#X8SDfl*h;uZ<D7gntIvAxFj)={EE`|wGG><!))n_mk{OjXrqQ2FlOp{i{w`w2
z!{{>DU*UP7sc^`TJeC){NUxdwk>ZEZ;D$5?sl)psBD!}^p&LudLayLr4r4SfcBM%h
z6O=Gl4^OFaJ!Cg4jWZiBIgEYPLt`E~_F%E3(aFe|B83;u?*A0ux0$iE&Sl-`CM?s;
zS%WE}vYULzbf#IS4L_WniB-E%iHVGlDYRQm^7qK|m=4AiX3uJgiSY2SksQ=_1X4mp
znR20cBD=Cmr(4jFoAucEJc=Ug8{2katp`jK(=g`WsnIGzb3zvCuzLcPG;x3T$mzrJ
zHqYFDy>WTuxj262WPNVO-@lGO8GPP8ZJ!3K|M5DOUwyYeJ~r>HR;hJG)ajBqm~+rh
z9gtXCE1TNz*oyl`C46&K8of74DWt41HDpL4Y^U~*X}bjDAlM2L>>l5lT@Clj-?>Dx
zziS!nC`cL;=OPo6F1(yx@hZRKg}z~u1DQhC71w@a2$}SlU=Q(!%QFK+Yyer_=)Ggo
zxk#9@9v;66;JASI_6ai;ZXNpmK_6gQ`2JkiEzF{KQjLuDKbr3y!tNf-i|bmcwWAuc
zl3M#GMgEvdkN+YvTgKEkL>%i@DZXNp-PknljyHqiH@_7Z58s|)6^9C*B3(2gL(3NC
z-dQtKnn+zxBcu{N7`L@Q^NGZq-Wm;K>lNu0Q;p}>ka=}7)d(5zcos)98%#_5`JC8|
zlMyFe#xrm{-T(J2qW?gSfA?Dc^fQEKmXqFg<(-GOf15a4od{Lgtg-fuURTs2fp$)v
zLT9US-coHeN#v9$In&vJVvFbh2v1Sf0J%&ESH;2p-4hEWzF+H+3XDaPXVM9wfp6S4
z<&g-iqQUgT#(NU72t^ZF4&3Jl-b`<JncwiJugS8YBsv?-xBbM7pc!N5qxIYR@BsI`
z<6dbJpmM!7)H4s|9!JS#g5L1nxzVWx7;VE4Uu6dnVcRxNrxUGBxF@!VR1I$+Q_M4K
zjdbH{t*os@awsJj7-oOZ{=`iN9#1v9ff$Z*-8LpNcNB!(94P$tNc+@rteyRm!1G$x
zM3Tau(vfnaq|7o+SQEO$piD8Ys|AC8ys#v}Kx@6SwGC}6Xd(%0*5lfx83U??G%e>s
z8}Cj$6Rb@cgO*B>f=aM)f#Lb2pY2-q{eQdn+IUi#|DDfzUdO(#-@n`C@3mDM>_4~8
zY8%pHKPYa@*{RKG0@^y=+bv8W_E4wy9yMpu2shEFt6go})3NgU9<DMzy2YNLDTj~f
zSQEQ%<<4+Q3Mpk=LV+P8Juz!xQstq%;Bk7%V}8j&ADI0vyq?5pwzBmrQ*L|$@!b}h
z%Xm~6G$MigvaU4gRPC`QY7Kh`6Q(jzaoQTvzF;wi5@fIc_EQ}-b+DWv>S%{h0n|FA
z&KmxFE(jAzf{ok<!x9O(<IY{J*4cU+0Qg`n-Ja|5Y^uGpwa(6rkH^elN)2Y5xnVA0
z&7uSdt%J24T+#S<mk7?0iKR^RF0^LDDq{D~v*#qt!B}Hh2e78FByhzTVJTUt7V>Tr
zNI6ohdv|Juq=k7}aA~Zya<Gu6%k2#B``I6_i_ddie>wjDJH7x|UyeWB;bXIuFh6#3
zG!vuIg6CpcQzubLTuJhhEHkPZ)dV)C<b=1)$v56^A2`(uAGu;;Orm_95^qk0kMl%Z
zCZ7BqS6}Hp<V31Q5#=sTSZh3;PCUH1<8XJNOcBLB9}akHc&+FZP1I%quGDU1VB0di
zEWjh)-_y5kByYur8r@mFO?*=dLqs3`QV(h3m|pTvUi_Cl@-?qMensbiQSkW*OO@6Q
zlZN_7CJnP)Y{9(Hq(@^~H{2WU7tZy}sh+sV3Pt$ye>5}CcSr$EhDUxLJ3UzG;1%&s
zVIvKb$E1R5#;jA@2I3*<ND0vbX}SS@qG2riv|hQal^=h3;xIq(=HV56Q@mx&XWWEZ
zD{H%Q+D>$BwA7+la#*dS-tZdK_TD>fZP=xvQ%15fm!S8rZFn1b!jiG2l`1cQ8E=P}
zDjtMGnfZ2lq^=v=x-qvfoc-l|VQm2vh#Rwlt!GZ}JJp4>-1B02MbSvGDVmvbVSBpp
z_H^c2JKdC=7K+}nq<c!5L1F8a>-9oe5|W~!+Kn?m8*Rehcl<r;I06&;bAdKKG&zY*
zrjs?@*0JdBl@o2$=A9?&{IY)JTu*$kGcLx%TzHj~yG!CaP1szxJiX&e!)cTP+XmB=
zdHw2vt-j}axpH_oa5x@Vjt9Iu*Y{_VCYD1qeeQ{Or=M82D^w-BvTZYmJX6Yo(6M@r
zKF%2)G)_q&Kg-sUky>>0DUvwbBQNr6{!g#|D-OKj_~BbBl_rfoUFmwklY%sC{y^)F
zwT|dLq(b!#?~OhtnXO$o)h9l#AF)}{sqode|0#31ryLxyD7^Ou^d^Cpk$luc8kmz(
zd!t_)t2L?^Sw?Qd4#dL%Wn*%B-6}u6{e{+?7Z0!4JaIm)^xnb?ZWuQ=8r6->Ks2FD
z!&{~E;ps$6o%SmHI9!Cjb?UW7>S!V+@&auelCdmwbvzpfIZ!;2)&q_+znnRi17FQw
zv#uN0b;G10&G`Ou;aaz0ljtNG*R^rJbZmM{X^j_m4`eNL%hBK*U$Sn-$J2$&8iv4G
z!sBDThKwtvn8Y`r4DeYQ1^$hF^$&C{v-fg#Yrc{Q86j&lE-{f;&}Y-<%{ou!yjwqT
z-k$hyIe{9l4-=0`Ns}N`L5i~d;TJX)j)yz6%v|b9$%*S}!-|p02xuq;or7&#F9tcY
zOb0IO2i9#vas&@to;ul#^CfdAh2yetIyHKG8gdo~3r6iBFKpiN-pCkciF*yk!K?W{
z<8}Fp|NQWuqw)fOdd1qGsD9yN|CzPF#{|%hO&{p4G|w~=O*AAn-u{2={n?WwOLpdo
z{mxNUGk1@O-0o6^>P7?8n0^=;5eoT63i+2)$PvDAMEJ-Dvf1oq8eORE-kX^b;qGRp
zs^{?GsCmRK;GsD{K{|odLgMBU8R_n(%hB(A3#kVyd|RHl*p;(i(WjX-9r)MNFDZuu
zZ{B^NEoatGKjJ=tXT)1b?Y7ucPbG5qk_gLnWnC_8e|eAj%+q<rn&H)HUD%p(y)>3<
z=kxW-W}SBrzr;*=K3~|ZljfO*vuqVfB_zY51Ax{Mm;;guSmO6Te&n3aeE0As&O$BJ
zUfF7_ldMY4;iqTkgeO8en3XmsPPH&sXIm#+jMKx3yXl^94)3{a3+L?$(%81n@3v=_
zZDqe_&&9Z`5n}Q5^pQ(zynFwSqCl(6(}}y&TdvEMwJjX(?xA!(eg2HOFir32wy>-Q
zvQlm1o70<+sO<*Zx2w;Wxt8tL{gA<~N|7Mhxz)<IjuMbWDHO665@lFk>=LnnW3O%F
z+*bNl>AmsUK7(}LEQR|y^PNlsArL<Q@B^zyW4O!{Wtwo`xLy`cB^X#*BKV_dA}667
z(|F0k<zj4ggC32k>$Z_o=3P3VDe83?QgVs<vrOZBE3uBTO6x*%;i0_aD4D-}|3C6Z
z-}9aP3b7ek4%B?Xb>mz=bKO4jkS*rY@Mnx7Ie@1+I;%_DSn7pO*N@@fm^$U-TQ(Ek
zJbXvW%7+i{xtyQ*;c~szH6{()*{Qd7-gs0wAsKQ$Q1TteyDOfA?dd|V4cnG@o+UHs
z9We2+eWqdb6in4w(?|<79=8kEex;X$OQM@#y957F4}5w&(+hlf^L{jW8I~p;$wJm(
zGB<a614~ZNqv4;0Y%|F+*b_+$?;h@%(t$&I;Idx1Y))?-@13X1$})Tgx7Jx}9p_4?
zUN2$VKFu(ln9BjO;wAI>{Ky}E_{`z%o@t(Wd-sOAb(ZB5_op}f@)v)>Z~yRXK7RTG
z5BG0|%4G*<Q#3YjRncGwe+97qS&Xj$*4Oc;8^g#NqQl+IXrnW*Ct&SeaIqBZp3w5)
zI&PAHcmaIGVK=t6u-TPO7LXwe44wlE;WyQraNR0b+kof@sv0RJ<~egZoVY){W!6mc
z90j2}+q$715!3r?Z#Z<1*XGLRor+@-$rK>lZ3B&1b|jHfB&J1|8)g!w6W`uFaFqAF
zmG3zE8<I~WhPMZ`+B%&Ebdsi<f0vCUnARJe9)|8_Y`t;mjcadI?MT;~?@(e!$-Jaz
z0rb&DZpgTt4VNH<4&Ld#(rb&$p+wT4q)1!qF3}k>!7O6Ntiigr@CG8|l4+7harKh=
zR7aPune28pH!q6Eb_RtzJ>l8#GJHTx@U|byA>*uMqzS5>3s0o9unr%jEJF5#wi)Xo
zO{qOmJ6!6OWvg6kr}jAbtc3*1mMe+GG2KBDrZl5i81rVK9F;@Q<R&y~fC(+(7Aa?~
zIqPbircO%m(w81?Eg~S1z^UwJpa1O--@R_+WgYih5A(&0HDaU_E(Ih-9JhvFj*g3S
zlo-y|E7!Jg<;qo8j5@rb1e3%z%w^4l^SW~B;h~Z>lcbFfTqC87%85h{AGO-?VI;bH
zU_~d_$SLD_!doyLs;yKLn(rp?FWu@$KpF*l77i&<2RCOD!Bb}PZ+J-GaL{*TJ3#Na
z1}~#-8(ZHZf}HS|OT<M&#BlGdV;$p>lz6Qh=T^CTLt2<rhv!Bl1*$6D+Ko0J{H|b6
ziU@glPX$w@MkjsH9K%B*jFQ8zC;;_ddn6x?Bv3Qum@}9TW`&2&dwk;Eg1v-JAuZZH
zKgM%yZd}?zQsrI_v5%)A=js{>ND@p;7ceVmb97x<vd}frV0e58cqFYg6SnSb2Aet0
z+rqMKY+GlkJ-mgiMJnO8U}>Ts4-pAYFh*1o>$dTHUPzia%?HYq>8;Y136X-PMC;D7
zG^S%lZ|gYrQxQV4I9Rz~(mDpl<vrx`;X!COL;j!3s0O=E3P{EinELRI3D7VYANyKf
zxyZsbZ5Ta3&*4DWnPISa9xn@vb!Zz!ARlM$4l`4p=sFle&1qYsE*0;NmfZ)XA_^%d
zWQ#jWDh(|gt6fn~Bs9ETWM@_gk`u|C$-)oOir`6+o_O&0bjy60f61HlHYER^>7HWW
z6WHp)TCboT)!55)5Ev36XI{GvMwn4s<I+~H-dNlywKGY~&w_I|ow&Bj*;<^XBIvku
z`Of;a33sFVhR|ZQn_yj8)`hw?dJD7q)|G7)>K5xoB^pcSPVWMyOcBF?ig2}+=8dKy
zopKqnG28iR9!A<OTx>-JW=W7bA{~{4nhrKHSYGMk)UL?d5IF-v!Kq16DFK#RBnWS<
zN3vV*Y}R;Q0-SE%*sQU&HYA`H-eHod+X2Z?Y{PoO`jvHAcs?&Qi<9MTnP}=fZq8vo
za6Ej&x?R||6*~;(_PG4IH&mkH`6Acy@Fv2C<k?<`xI1%iR~)EJVb38ML~qPj+1Y=C
zLD;>swi@RCv<UbRWA#okp>{(}5fd)gE9bgFZ{bp8Q9)*%Nfa^~q~ceAg2)SFMm3RB
ziIhVZ8lCEmwN>iyUEKet5*}?yqlsx69=3@HHPYm~J$!@r6L0bd?&M*BE>U|bgkd>7
zo;y3NH2iSsn03M%HbAqjR+e!VJCRDh){S*t=*JfRr=F=K6d%U;AqBSNU^#Z$VWHqw
z!{^g@IWs+2$M?qvYHn;&ai4K7F&C-9MvM(R>Trd%kJL^wox@NH815U;xasIbSXw2s
zu=FcvA+!5rh38~9i4Ym$OzsGET|@Toot1{t5jBcKbA=9Ojccncy|FZ7ZJnp<!m@5G
z7Ad9^N*m1(vyS&lZ57o@PagHswlS57cW>Xt+2q3P!6VBeoDN66|MtY^ryu#je4Ozu
z-d+?H`-0#V5A^H!XEMG5SYOAVZsan&hKy=Pt&Po$sxMSx69Z4=HltEVso(-#8zMnF
zRzKjQ6mE56Sw6GmN7lTMxFeyQ2!_F<21`mSJUvyOw-xyi=8`Ep(^R;7nECMT8-DTr
zUsC;ow@zCsX3pjD!aQXT_osM`YE%#NNbw2r%+(tQI)SNVaxR2vx3Q<ANXikFTlD5R
z@K6dzKk$o(zhdqm_;C44@C+nW8z#~DYOT>&NYV$05`{^V7A_fe4U=jSHtRgqm8Gw&
zUcsDf5p8eXDBO{`2H4U86U-Z?8zvhfP(23#AQNx_TWRacwyfb*m6gLGlTt=fq4kh(
zZln84PgoVsmr6+vo@o?TnJGhpwRLCh&J_#L;8hu+geTc=m!H_Q^6m5vwBy=QjbtXT
zdlqC!P_A-BEs<O0a`+vVJ6#qg?HojzEI~c8t(~V$xLRiI%E#;PxUMUgB|t2-05Y~}
zn9fKw=5(SLr=%In1W0)h?xw<oQkTY$^*JtRYo@n^Ct=;3zIN^p8#xbu4GGGNB4klS
zJ?QlOJVoWd<zX5=83HtOw%Vv>n1;7Qz&f#xn;%FT9u7T2*NQT|s7?q_WN8;Zt)Kav
zo><Dt!D8K(1d@d7@Sp<W<I{zw<w81T4kc2#gif<7bD5d)9oZb;V7;!m1^vIMhPkpw
zn&@GgnNp$K$^@+anWc9YQ%qVgtZuT02~{DdjZ}mwCl)tWgJVt<%e*=LD~|RJvwa6<
zbh}0YEXHNKu(k^+$FrZh#)VxOg>vI*t5l38!qOW*u9uN06v;eOYZwj0DCrPV*0$2g
znCIxs?s8$*;T<@->TH;A<GqIwyqVG3l}L~)*kz-iE9biMyj^*|E<oXMm;$`&7Cv3r
z4g7AsMNmRMh74$z<1%DP7$3JM9OJ|34$r|%aR;i<33K<Yt(az-oJfn4w`baEMM}ce
zNt%W%A+c?6-JGY@Sh}-T=cn@{*L9<98*3j%u^vf>eL0g5>fIrPVj6&D65(=P_`^@1
zxVwAH{b^<@g>BooK2^{a&4slZ>$-ACi8T92s{lrfe3Ah5T=k1u$1ud=5@Nt+POXh*
zHUPc=4a4j@VH|PIpoN~5rWWttMj(U#9cy`xwVYstfF!!iVH&K?508&LU9a376Vsva
z@a`MlKRj@ExTpF`W2Ik>wJkh9UQrR=zj;Hq&bnRWB2@}~ZMb!&yN#TMv;Rc*8RJNk
z!QfB{&!f4JtaGSqNK_`J^h7l!>o1_+^WEX!Ff-#-sVIxfkov=Qy>Qu{IoxlkD8(`b
ziF9IbB=4*}>V&0ruD$Y8yT<wi?wTRehXL};`{R40Reo!01SkwyZpe@V?2&$=_batu
zNvToP=$5Fh@c8t^vMnsF4a85e>k(Uzq%tEKCeFPZwP(bQBPHZVlKGqViHiqdY?5(3
z;N78ZxURS~wjP!TANvy~jeAwlhDs&Ng!U=C=%mq)Pc%<l+X?E5JS}{>>-g;4&4q&=
z!NTL`sjr;3m8ZILX^o{detNucT{k>M_|DQQn^&N-T`u_AIUNg>&LK^hG`6ks^tkck
zPnG$uFqOogAMaS3@wh$n_WquK{pWwhzyIsM=GVXbJ?&(8?jc(e2R4L}miQ&E<pHkR
zu&Qu_K0qwh7v`0cNZd#8h-N$|y6*7@JF7J2UTu}@c4d=FAF?^bS-g>5Sej9CBK5?l
zpC0*qd4@?T^8s2%HE~D}%$g}A=wYQ{yT)|X5e}p@F=s`~N_N<&Tupgi&v<PE+zF|E
z7`8hlhxaSFf*I92IW@9$e);AvDgF)L>i1+Tq@Hls2rI%!(t(Ol1IQh6x{%K`x@~lC
zA%}M3yk5Eb!ZUT}GjUxPQhw(CZbL;l=1esuNqEL?-l-{!)rEq(j_{&JZ<oQ^Of-w&
z7VAzX^NaVtVA(pK&l`P%ULCUyvo(A(HG*e$0Bh;tLs{iO(-VmUPR0Bu%;K|?fnc8N
zmEIeV%O`rCnag3MwT{&A`VzjeNu}4ycCC>(S~8L<l8pP4agfAA6U2?)FKq3~TCZFy
ztkwAar$?4`LnMN#PEzP@Y(TSyb!XYGILcj#KmcDGZ%*&{{+kDW`@4^<*OjAgWH$~8
zzIpq=U;X8G{O*U}U=FQUS`$)<24{ey%4<Knuj8N3_zGZs9e=X1)FDd%sySQlKVxpN
zcPJ1tDWcsa7&p7Q;vQq-9-Dq7RTK{SfTl>z2!mZ2-Ka*F4YDF7x(CRN3QR>g9glo?
z|DN~n-|_y#JD$!?Z7W$OT%F_Lj(IAa4oA+H3)?mdom8;Y;u7B)b=z2$E7Xdcy_7vF
zhzhx6eE2~mi-SjY`9=zcmS~DDk-zA;x3;pig_Oc`r;Tm{9w1ORWAR88%95zIvG$dY
zN5rgexI4>oVJ-&_IeMXo`3;@Qy42gh<(h_+U{Hv~N5Wy^yevFkp1C_6nLJUu(5)hM
zVXd&W05aBX<Ky{}Wvfg%jLpu=#<EpftK`u=s~U!_&Tf2@!8b-D3KQqz8^u;WZf6cI
z93_}r-TR<=`!N1=7B-}Tn6vhkmZLbCv-7w*T{grwo_pom8|N*G<<HkM*KMPAsBUN;
z6y%YXM6dC>EU~fS;%FDH*OljGV{S(#61lF32+QMV+Hxhi5Vn+Q6qNqrGvtHdtl`<?
zfU5nmVa0#5V`-7ml|{!o-kqJ`HOov<W>3sEqSSl%CKzF6te%1=xu?$=+@-=S(0w7U
z&jDO)Z41M646-35Bx?!hFeMICVk(*AeBd;laB5s`B}X8`T`-nP$t>$iYcG;s8JF~}
zUs)QQw~TlHnXE!mcrj!z;USeIVxo5wF5plm4mOe3G#HeCgv{@{URc_Bqyt44J3L=J
zKp^W}J(zzY!q&Hlw1;3T3^QJ{Fige@;t@k{7P7|O{cdk;ZyG*4*(1$KQ}}6Yz0=!@
zbf>MAkk1sJmda(VJT6zxZAJ5pNJM<M-r2VB7t9(SMa_-Pol2OuJ3YLkMi<{O8ja7}
z1qa?-ADNOcO}=mSgQ;Sy9tP>PrMP*uPH*7FNg8Ic$N4Y<5jLKeE05P_mTIiE^Smv?
zJ7fTA8q5Fzdarn0Sl0^}l=2YgLk}?hFwdw9-a5Uncx_CYn2skNK77lkkB^+6&o`ee
zMyQAEESaGs{L<Djz|#6HwXa&F&AqJU{Y)MnbN>KqxtnvT3liHmrVd#-+}&|FOq`Af
z9v<%a=KF8BF3()o3#a)&&Pt!+44m^62Jcq!!L`15mGxjSc~{yN76aCYujOE1YD$A)
z6Gqc2%AgU1Aux%FnQU{+Ax)^iB(|-#vf4_qFn`}H7!a$&)`QJ-RiP2UTeZsW&bBjQ
z!hSAHsHIPc6?P97VR+K1Q!}VJb8m<_#WN{S<U_>TizZHYCsI!IYmHExbx4-f$vWez
zEY@i*xGKpD6PY$Z#9n85kF+tD$gnXRm%u^aiDa<O$Lk|oT1i`_%!%okXuS>KqlGuk
zdZl$`S&Z%$EX~$nHl9m(6E#s1nZy&VZ?wL$R-rcM+!t1_sI_3$Xav*X9g&7hq_QT>
zFlR^w_rmwTc+2;{Jn)A<d}Lczl5P}?W0tu1{_8jV^!UVYfB!>(_q~&Ku!;ACyD#)i
z-YmdlH};kA3VwU3t(8drEFQH<vcWjx#RHGwfhPz_5;^zqquiy49$`DR2RkySkdRkk
znhW!xpc<*bQ<=$WFrzHi;Bh`iok@5tZd(V}sDD+W;nW(ND9c(&Hr9!RkJhej2<$x;
zlj9+M?h#(}hI?k-QtU{&C6NwUpEtT!!nEG#Hkw*Kf_!?X66v6tGv*d7H17j+i?cYu
zlMVBR$i$qEY`P2|-)MCB?o~TL#hs_=($QDInKPQw-B_-lSvKRmScDC93D%%FD)+&f
z6A#R?bk?o$Tq`c`NW&1_eZl%3KovDN8h%vO;iu~(ODcFdkjq4#CUOsqzk9`KtTkl3
z+p2g{tSO#0(9Y70BEq>nqhusa^tRIaO6!T%g=?>DbdozoltaoS8rq|Ai*}avin~%z
z!(xI7v(B=d!5npEy*|?{>W<zvo}ZuSt&?@adZZ?5jt@|FA4{B(P(V=szKnJJb^QIt
zR{-nl_>+xg>(NC?8qAGh+<0Sq0=sQ{lIVt|J#DPN0(@M-?Y23ii9?zwSy-zLACx%I
zjUGN7sUf+OMm(#FQZgJ5N8Y}9!`rtHynXvXt)XAfl0k&SEX-5kFdaBwmM}FgaX`Dq
zWyV^Ic-ZAiZXxBnjd9scIgg?=I`-zEX&92{gm@bArZ{v)IYsR&TVF$ZGytqUy8c>&
zUR^chH!eAnNNgLxS&n27ch>dFQm*uafENz)Jxjgd-frDFAErKCLK4<&2e6d$x-wst
ze8`9odB9-o)~?j9Y%PoxKVP1yy;DjYoR@WDtCiL(mZRg@WKhz_)h0S7L*A9519bI9
zwvEr*1+;L|G7M;U854BxG&dI4F!0$LXS-6Tjif^D!s7}(ZK!OVZvfWHb=!EnURY|w
zd&aQ<DuzkCSclfYDsG87jC+$Q%k|3T>B^xTD3ZwAikP!JKhc*Z{7Oc@lOfr9O=pf$
zDgc>Cd$B!8=)d`~*0}U~lDPDmMREIztu2GqF_|)r%WoclLR@;oa6byqSVzf)lgAPC
z?8^XeBI&F4ikp*&81GO&gj9DfiMb?7P8{c%yTcu=T`_MIEr<x^FyAhY5b2l^D4^&F
zG1%J5T9xZo$Oeft%(`~~OUdLL=)MOM>{5+67bYvDHRg0jK|+eWtXHn>LYkwP+Gsb*
z`&QRTYkRpzu~{9V*sufkW={7O9c+u}>XgbYwmQ5^B2F9?wm|Tix{@>_IlP=U>$JAv
z%~+N-Sf4WUyjCu2<7rvAwu<HxDih{mvb)udGKFb<aKVfUG!c*zdzXcz_XWU$r*;LL
zPs@RaV;NE43<g(7HdcuXLbXI+D{ToTRi7HulsJ~m`##aFQ?2s6TzR^lSyp3f#${XC
zS{r6Cev97@#Y)AhvaJhZI;1dQEndW7p6~|VI@Zs4H;N|i4@bWJ@D11V%IE$RAoPCI
z0K~&zBy4!Uymbs?Eo(^C+E~lBujN-~vbr+)-+wKe!?msCE+ms1fHfbF9FH@{!-4zz
z6YoE~BR|EzahzsSQfdisc1i{7J{ke?3cwn6lLlrHMzFSe1DbaOX(d3%q_LKT-K(;b
zb@&5$A$u8;9}RgvAywKe7(r%JNagHB>c7@2)f!e3M(ioo>p0_HgIAzBjPb27sJ-lF
z`mTYUv<%vvzBZCOnL^5ibeKkJXySN0qA7808#bg#O5C-RPPk@TKSx5JPAHj~f@!5n
zrTav);r$~<_jM$#?ox5^0bqSvpIKSSQn};E!{HP~fHy=N<}Hk_E3B8!W}VF|ZWUiU
zsT7JT>!|bG64@KcD!mn23z=f2k*y7e<5<f`Tt=l*+ZM+3IfG`*3*UY7f%hN2;d0)n
zmqud2-8m$9_i)F*`j`KLfBWzMV@N8y(OVdR$6Ji&^QE52+p8p60mMhcL%?N!mazsi
z(OfHpXIYRZ18fl)VJy2}t16xnk|H2sm!P|cMFKgD{Huon#r<5E=R!(K$(d3PAQka8
z8VbTZ&GDLhy&1$q$mar#3TCe&Y^hR2m?av)yS&o|cy)8Z@L-==?^t(cx#PqeiqCOA
zN?1IE=h+Tmg)K=pql<*4Nw1-3kVr+JHP(HzI_5(jKC_)}o!&Rx8emN6NY=}(47vrE
zNC1BEh;-Dh(Hzs!{V?Ke>AmxOerB!CWfO{LGRn4HFmuZECYs&S<NKrl&@Z(^IiYfj
zBhX@hyoEwY$OL=%bMR?>VwH^4jrn-ukR|4W2!jXC)fRe#x`1U3V|iV1tz2s~i~sP{
zB2;ER;=SU1118*+#VXBvFwI3tIgxI*2AwX(x?OP+8x56-oC3gVy@JKPXj?83Af+d)
zZ#+Fe(Q7A52{w5*@&K;(PxA)aGA75K0{M0PlOA6ItgquwHZEJcC5QAD2jXp8+5N4E
z1i&oe`I1G4Z_-P5$gyEq>^Go&(hNz+ZsQ<|%j@0YT>wr~;i*1SWg*LiNrD6iNzfX-
zRcc!~-5od_3$?cBcxa(+jUShv1}HP5C6-CohMUKR+Zx-tqJ+_EDP<J9LTW<t@DFfx
zv^&zB+F&&&=PSxemti7F!oA{doUcz@>tpy{Ib3V7HM%vHUa20E9aUwX!zkQpMZ6%k
zQ?!lq`3WWS;Y}r%#M`$YxYmw8Jt65OKG+2$y!At)PnbnMpEp#U_aENl2%jE5fhWXg
z)~ak(c>H|kdR_VSypdAjm*4z~wK<<JXO?B9Z8iJ=#ChH>T+EPkH<%2=Y;v5r@uPFL
zD}VF&2cD$xEQR@aM=Enfs^c_&#t<4AsgA79T2^Xm^rdsbX_t>oK6A7St6f?8l|C*w
zi+7qjUIHL!J$%`sD5{7-TR)+G$Nk-VO3BRAjN2p2<xD+2kP@84k+$*U?|#dbC+3{-
zDtPaaB-53Y_j&v>cCLet_x>e}>uQJ?DUCF@TX6>5u95-Z9V9bJrVKXIKF3VB$=R$&
zfpgF>PemLp2lj-!oF?8(_sFa;B@XREPnE0%P>xx6d%WY#-I0g8BlpJxTO<7Owk@PH
z)o5*FU9Zf?19x{1^wz^j)W);NY0a?aH{VYg&!xGI#KPFKOu{(TTE|+<#nm>{HVy<Z
z8xiDdFpMmGC+>+9q3gB*AthS+62NrIs3?c2;9W^q*t1wwjHCiSBI~z?n=nuJbSl+q
z6#v5QO5bQ?6v&a{mc!q!_Rg}d)MexOrz>s=x0zarJ*hg~GbPVJ=DO5iH?<xTGL+}#
z%F;VkreM5KVn2RWJIavSHE*2z!f!4=a<-YpkCb#r(!(J57}s%W?gp(@THm;?7h0>N
zE%T`G{P`!MM|-5Vjn+0=6MBcUZFI3n`yLYQ?)&-D5#OS#>xw)mLV}*iSvZKXS&oFo
zdL<d0@&kvQD5?}qOk{j`{f(B&3)zq_a~%V55fNi+b+82d2AjTkPX5EK<&p~T$`PLe
zh&~)PJj_(^&6(Vh0R%~c<LMBpYcq15Ff*Q>&#c?V^|~+}X38{?X2nydomYB~umDvd
zr2!nN3<+Z-WpzR(X1&vDqt-@kjkTRo&%B#@cm~xH9TkT@E>`XV)byrAvX{|pqYf`A
z!zmJTyOii!)jm<Z(tL{*El>*Naqi$sNvClx=`eqev(g2cM#D~YCTk{prq)=;*Xu$v
z<8Zu3GIV_dmj-L6wI{km@tNd8Q=?!g#`F5b$NG$=UouaJNIUbDrT;!slZPyHpXV;l
zY7zeOQI*Yw)&;sFR!GS)8kcqji)6KyIva*gj+6lA*Uh<HH<D%!^F6(7Sgo{mp*1*r
z$Hkf4$Oviy92$-f8Qb;zNUxpGhbK-GO!EW3`OV+(*Z<`|alK6BRQTcXZ{k@T-tqh2
z|C;~H|MYM9&%gdHZk5^#sSA^&;n_GAgnXH6`MiW00Eyt8B!4V`j59e{RhhJ>NAG(z
z1e<gRV|#cNWf7*)%$1abP1gszIu(*;k|yr&PUs==_U&8Vym{c^%^mY}WGY9@R?N3s
zitXic4WQ<FA*YG?a2H;ctz(`dak>Ya)4D~Xo(`WV4as@OX`(bUmb$Xk#`C(pFbyXg
zjHAITvz_TA)K*z!0fK?Gv`7N34EqRQ<KGftN*NWQOyRpXYo^;u@9V7r)qM*eCRf}f
z%+L2VGoC<rKgHc3cGI}COgZ!X9Ma*_e8*`%GM!GWEf`o@6|Jz$h1QJKXBvql&a$rj
z@bS!I59Dd4`3d)oc^`n-@Edz=h(?@=t29tJXQOJP(U7)8Ltd5u`Di17QRu!)%L8=W
z)&||N%}B{O<PxdBy<x4>4c1*PBAKG$bB6ATzK%#77_DCU@u$Dx?cF!LegB3x@7{BF
zf6u@DfBy~V=SSLh1rhXk;^XI^`1t?)9gojXG^;dIvJF5G@IA{}`k3`Te*Tq#_0M5^
z1+c!3KiQ}~wCj79w2%9a-{`&aVw(K3{}F%oikY-ina7`?qNq5ok@VIH-;^mGksPUK
zngmG%vy03m9_EtHV03j#$>51?-LURJp>?M&3zAa!K4}^o<9Ib5+Wy|+5*q)Oj-<Zu
z-4W68KKreM)q6<dtgU3*ZnSNY#NJw}b!BTSq;V<jMzc65Hm?yCJpevciL{Vc)mVJL
zNd2oLr7&qEx8yY9vV9cFJ`VZ4%jM1<FJ7y$ZVe=im#@nb9TJRX3jpi7hE#B?&aBE|
zK82*QF0>vJjoss8vqp14(u`OehpW6aPR7NydFQ!afr?VNn<l0lmrWx~ZQ~MUs5#oX
z89Ub2sn+S&E6EyD7n-kB76yffrh{s)3~<74b5h(8Z}hfcngg^^q2wrP_BeLj$b^CZ
z^?IhJ4ViXNg@~gz^Krz!q|o^58?e0olmE6ytYe0xD%SE&OcB}3#3+{&GJ5U1H0Cu6
z?IpB4qOU`Ghd`0I9Byu?M2Zn93R%s9x|20HZmuwwiK%3!k||S!28cMz5)|`RLjrfb
zUYX{ZQc7$-Izl9DtmE#3X%p)&8ki&Unh`aRkI(|6_qez_u=S1Xb;zi8W?>-D-7DSe
zm=i|rBfdiVSPP9LX_!(+i7OI@ue;%YQo+4r7O9!;1fbp>kryUq{FA#O?Y^#J59&@Z
z=FDjp%%$EmfH+H-r|?ceL55*^kJMEk&)Z$udS_`BW0yX~*WQSRa@!tbozghBmDFHr
z$|N(HHulBv3o#e0Z?V65i<C28QI7)jx~?JPWF0IC!<u3yRBbo5f%qQc5g*3>^jfRb
zRLAEY(y)0>%m}tQwRY$Y@c<rF92>V0CElAk|76xN1Zi~#h#=mD-;UOwpUM9)Ygssy
ziIzHDjhrKea#wonhVae8RJ)W+@8L%&5}*9K2J;vgzLY5x=pOdw?u1_(2J)(DAn}^}
z*yp>9Fp*|qfZjS=TTuxHMw|n?#EkIo*)N1dqfFc*k=PoWZ|o~B<Kda7iuYh1L_kBF
z$Ev#*Tbv<+lG8+_?m9333qP_h@iiv#1k`mHpx0m}8%T~CM{|J`7j(BsL{4%*NVwQ-
zJ=9*gZX0r%(R3HGY4?}3##c{t2Qv+)iLr{2=oAlzvRH7aoyOW5;)yIq$LNFUB03rb
z-DvAd&W@BG=X7n<8qI^60139O!f0BKS0j!~BraDn%XXpY0hxv6dgXk5MEM3tT($)S
z^aPj7ndj?Y^Z0xX`w26e?Y^5jK9kqdS-z;XTwAbxbg$#mC{JG1)-hf-XMgqUz3lS6
zI32%M5%ORvXCj@_JwhLnX0jwy61ikb>Kx~p<8)w}3e%LCr|?K^t<&0&`PEKsm33KR
zP8@X)czHQId@$|Zc8T^2n7B7ML-O3C?yQ}4>$Cx3xbK4cxDP~q694Vq=@t!QaGE!|
z!v<7uY*coz4-F5oly~M@G?83heP-kCFp6h(iTTTpIQAP}Y$Svf1{0duwn#KS9S@|e
zOmikTL$V<YkVeQ%H`-PL$QGx$a$cdQ1DYP77u-TT8sP)+{`Q8}%W`(Q7)_iS8!j@!
zW4sHl;WrmV0V0Lhjb=T-An90dw^}KQk}Uqb_8y`I3nTleUHrzFdzpTFbHu^`etqIN
z9RXu1nY+6aB2mjXv)D^T*lOjnJhIiWO)<A2ciz9TvEL70Sg&8%SN~kbR{-nl_>+wq
zwHWOc7`|V88A%f(2PO(+i^Ij#Z%o)Yq<4HLI_`&(XtGhQb6u~&oVqy>9r4!UV$>Ur
z&MXO~Fj?cIh4;rh?hgm1Ly69P@0`yUYBh2?Vr`>smFXC9-lj67k5Sm^@Y{O%F(eHV
zzDm_PYwy(FSlh}vq<^gjAobJd|H!>;yr2FzB+0juLLJt2WvOSf1}L+%(f#!{ibJQQ
zh_F7)Gpdc#RfpN`0i+>XNn*gp+I~W(BX`FWWl~bIAz@m<Pf;v-6ddceG35hkDtK;K
zR(#Gh?|goGrftTyI^AZv7wWQM-FS0)!!*x$j$&e~36aKB;L<KsH@X}^5}NK2u3+_N
z;MIPUa&el#s_-n1X=Nhe7MJ-dYjhp6qb+0;Tf0(e3}!dIM>Tiu7xE<JX=dAkHTSeG
zbb&G-0{kmE=+|bbccdHZ^Ji>oT+1U_634e6`2L%3ILs%w&R8|BZN<A&Mk>tHbtQQe
zchy1yTO_W4x|fOr5Rl!R^a~kvupaxU3xxejM@Vt7kl;eF**rROQ-LWB8P6`w8wIwe
zkTfhf)f-n|aXdVwQX$C^rvYsgHH}#Z+at}~NhVc8_abW);_I@~S}+MUCFXeogj(AT
zXckQ%(K&Q`DLQwSr@7-vY2A6c1lyqw7S+}otw*}rPwn??B0T6{f$ynneMsq6Dp!^*
z*j#51zZgckli{2Mrkwcj=AP@KJfD9eNya2D1JmSe%Y~~bb2*bTeE9G!{nL+JwktZ!
zZAC(Y)O)ayoY-@{J0H&rO*`Mdd&hKYtW`KauRQ;BLFIr-qT~aaMD5Xy?_n53DdCZ~
z)y}yaS64`HqbcFbxc9uI0L4ccay~@ylxdT)lrZTP8}%lJ<rZKsEy6|T0Q%fJ?Y6H<
z$d|oRQX-eaW)@xe7XEZOXOarDCZziSpQ2M<*DG%D>FM{pyZeUk-hIROzxW03KYYXg
z<A45t@w?0K*|sGlr?c_t@e_aj)BlHBgR!&qhBTq%X~;@n=E#@0jv?lS_>knlo-y&W
zGdY4tf<^H6U&|9oPe`t4*|4?K`o^3KN1d2V$mXaRdCDZ2`1I+@x^CojKjvR$J|6h)
z&wk0eRkqsjtQZ=}WaiXuLCcJ$%ng_rEUe8sTWdV68?|?~9wy+mHJ<8cRGsy>Ak4Ux
zIP2X&Tb!q<A-%G+MqK?QB=PK*CpapUoOpMC<hpLGA3J!2B^;-T<Dt;oLai50dE+oE
z@87-W^Yaty^Ao*C!g90NQ>*m=14-fCw!-I67uK!t_WljsD!>2ncWh0#*2HOpxjAZU
zz!&)r(L`sZZY$U86Dt|Yw^+Z&>_}}J&4iEyMnYqZF!mM!&9PthlBp9csl#r{Z7mvs
zDT6`l3!6BbI8s-nHF^ucUsK4%Y7H;U-Qd0hSj?AIgOQ%gMClMA#DSElEAEvafBetr
zyT+fr|DInwd_&H^<lp_f-|_kJQ+#R8oN`7a(M9PR&CTn!Vb*!`=1rvi3;_2_UCZ?q
z=BR2U3k!(Yg5yk%unwUTph3eqlp~U2-K3qRNO;Jll$bSRw$N*(OT)Y)R>)*d`JR+U
z5@2oAe&v{E?#clrz}b`oB*uq~uE)9L4%0M|Qi}PY2`w2<7reLl_tf#jVZ82uVr+oW
z)t%q{^pVT5@%gd@(42zdqnhbCLQ71J0T_A52z!p^!Itc{(tV{7EU%`S+QW~rXd$Oe
zE}1DS<!}n7*0tUm;CJ)P-DxIi1z+iX12;;ZLxyPern+cEk|Gl9xI#GnnkUw-cvH4i
zASv0MxnxQb(iGdyvK^=mWSuy##>Y=r)_RZe0ptxTCqx!R8qF7)cf#-Y#btPBl?9tF
z;Yq1kFzIw(#{Q&HLlL1#4s?w=uDfAdG{Iv&MZK)5H2ikgF(R?mXyD<(kz84|4Yf!@
z)KE2igzD7#iSrqL_ruIm&wP4(<oWWM<{LWQVIp)FK798*|HEJW8UNwm|2uyD>;KGk
zUFp5@@bD1+XcD#cZXv^H0KDq8d>#LU$5#OB>-dw6y|XJf=G2X7G&Y<UUl|#fA9id=
zUb;gs-C=Ibgt*AW#nz2T(Q}2SHzP&yU}`0agM8OLrQI7tbzH)tyK2^>ttb))ti(m^
zhOx*C(6&=_eeYn7u1B|ETRdN{!J6ug+Q+50Dcy|gav`THo;{+;#Yy8f;cl4sI6U?(
zS-AZiqW(%M<P_1=Q%(_yJ6@yePF*&-U52^if!((ARm=TVzj0jL;}gQLkT!UVZpbv#
z`$}yq-4eY^bn!(v%oDj3YO8U<Z-$I6)>e(Rckm-f#{I?n>gSW?=RY-^u0E*PjkuN?
zZCoz8$L5tpFz<M8xaoLnyD~of(30w{Shg`YBK8=Pzu=JZc|?c3*Nt_#aF~u^qD^##
zB7PlQ5+*i^q3GOOXlrP_qdmTx?48dsS9XaS<5_(PV>D=QYBAV5y?fv5g*h1j(B9o8
zE|K!`3GV>55M+to+eqGP0{{|1?Y_hv2EILvPhEUSR)W!%L`lle{4s!A>dyYJ+%6=g
z#DASzhi(VEVjbIfc{>Awu~kI3_xbt)U~R3_Y)A_cnpV1o6xKx|&1#oAV*?EGSKcH9
zqhNfVdotc(n&`dWBHcwa0GkvN59^iQS2X2u3D5yH?C*QLrSbXN{{OmhT`HR4{ktQY
z3e)ir(b=oxR=|SyV6N0HUb{*HC47jw58tXtn~bDV17BZ0OW@TLzja=HFiD!&U|q@x
z=m@Dp@Ivh6b?;(>{c_3AL6#dEj-9LGgOMtO-TU)R%{`Kimi39P958>C^4e=8BfaKh
z$hgpr{kHW^YJ;^QI|$7eyN>s9Ya;#8ym>j3Uw;4Zy_Q|^{_00_yg4#O(?(vM#XBfw
zqp4I9w|!+l*KSfqm$A<FJ-g#6fb82E9BX51jqA4YbXh_cXoG=eW3P@~oK8mvy;3}2
zdZ*ujVjGPrBMkuGZ{B}8E~>|Q4)&Oj<XMEA!Zd%EAepU*7{~d@`Fa_Q4!gY{C-@L)
zFs0-3fMscz!o$N|0L(cBL$E_h$~<W#D!Wi)lLS#Z!nGO3iKHhQGk6)G)$WxP|MFgj
z#c#y(k=JX&!v6|DcA#Wj9^*3%2C}wbepw&;BRM`Z_ZROXms_3ZqyE_0a{E4vyfE^?
zX?>x#3(NLIE^nFBVSua^G^0cEJReRxy!oEB{*Lqc8lR!v=pgZK#{~Wo*YfM+*L|Vh
zRAYOsJl>bPWAd_Z`s=s2pPS=-D^RV6TuVY8n@1Z?cbeCpP}};jCD#Zta)0ril$W&&
zGFa8q0GP)Y?GM?*Ox;C;5%_BTs%or4Gp=jpvTj_qI)WCW#*D4`)>0dv%lJNoPvc(0
z?TMc@!i7wR>|A0E@5zVLG~ralHq=8}nNyt0(VvM%od}dNp}mdu8=t-H_4H`UI>7|R
zJeoKrY#sqrlgHjMl|&Lq>KJhbvqh5PR)y{}xxAz60h0pKqDG1hD}pCvfA`!z``i2;
zXR=8IBgLQX77+1hcjlOlU;ALH4`Z>Yjd)QS^jCG0jky&j`1a}!^HSXJnU{E@^_A=T
zM7|ynDWMk0VNbyDJDleu-+lL<-~Q%r!pqLuusVrwuiF_WjQ8=fmcNdF`r|8r^>zHo
z#>;^kd&=(XwEr2S@YB6W-T$Hh8i!h^k%u865=P90%%^4HxjtbU7cLb=H^r-B3QaqQ
z!-VD&vrcpqp4&$2Q(QWepsCRMc02S%p-hVSjLsuP@OG($lp-3qwJpraA*6&J1$Xbp
z+A5!(&-{P?)Bl2~lBaSz(d$SAdwlwYmUs9ZtPN5`&xW=ly56-7PmB_YWbO<gN+~ML
zS-G1gni_Vz<GKmgtx>X)OXe_TN``g&0o9dxDfAYRz<Hz~)^>yDaFGF&CCoEhb5d2V
zb>i+=_|;$h1)qNW$n*2Bhdc|iPUMvN_QO5R;18c3d3x@gpDTB73WvkQuP@J>H>dCx
z)f~mEjTESTGQ8Mfjk;zO7UFEwI3%@z6Mh5Msdi!R$}CFjmA)6mD5i6C!l)x{TG+&C
z>ZDQ-j9M#V;ZN2DOos$cptjD|SAO{MZ?Il@bGqZ{yz+G3IF;|wGNJPudaXRK6`zIq
z?hUrA*t&6DBI@_;!<ztEZr%j@B>(bGSo@=Od>ejSuD6;rE?^_^N`{|_L^yysx1vWC
zF3W|-`UsOyq(lUEE%=jSS!fAT4<qU0oXIK2jV&vEDz_v})kI1j;10x|kP<mh0e-sB
z+VHXRfHPZbST6*4+eZ?ifGBILJYFvRr{DfI%&uLS@<d9R9<6@z>_`%jGTyu4y<=1~
z0=c6@DT1avis$h;x{w9tth~E>AmL1+m@2Nqamq|HU~Morej=9#PN(n59~J8(-dlDV
zF!*4K*4|O;Xzl#)L*?-~@#f7Pr&HnGFFx?L3*Y~l^8D$-`DvkVN^7v3YxoNuGTjq3
z3+MB~dE0QlC8YzM3D|IZX5X@2jCIE)NOp7JkOVIt$#B?4(J73Y`wQE$d8hRjj1EG^
zw*%=aislrJfm&%j<Y|-;Fs;^U-RW&e#FUVUO8{kN3!gsyHME6OIq>oECxn^b{^619
zx-#Vhu3>CE-{14@!@uVDzx@LrfBK2*veH}Q?fsh}$J|SMzSwmPvF&!^PNMs><B!ke
zf0(tbHPbIb--MP0e-3cNv}4JUc|zxzG9R$kskIOJd*yfxxk^r9-g`cuF^`K&Z5w?X
z3?L`cBvZ@@>#??7gf8&$>50elm0y2+gz=suEp4K?k)*SGz~Gfk4oM`PP9-N}%8~rG
zUkKfxcjusqk}{Jjb56Xwe+cl-Dw&W_9!ughXNn3+V_kkC%@g0ef6w#sG^EC3UTK_v
zoAe=9BqYZTmd8ff3UA&&a5x<K)vtcZay33bHr|rCC-HgdEY11D7O&nkb0zcRl1PVN
za+rR}+A__;QwY|VS$qTXUz&AkFz#@pwoYdq9+rhUWfX<s3E0)y%vifoS|{}m8k*IX
z6n7;N{%hO1!T`m)@&aaCC9g&w#lVZ1dkCY1h4b|f{HI?#-@f~E-oN=J%@(v6<Ts$v
zB>S^}@#p;S{!jmV{%`-+|A`-e_<`2#wzt24RIdT=FKaFLS1<i0qS|ZgeTy1|GQ)2m
zbO1YcGg1%c%r3u*I(%&lmwM)!7ethkJdnN6Hwge*JN+8oaC^|wRvWntA2|&+m!@I!
zkQ9}~@&1UJ(pnnnf&B*FT5Ay`-~o;`kLQ)ALO*xby7K9A<+5&^t3?yn69)-4S%5gN
z4H?5K!`x|YrS%1;fyP}TnjfJUFd(N`zgd+jXWqSg;OY6y4?ldOs3R$HOod~fI22*Z
zhWkR!@c!+4R`2+FCYoo^Bp{tW%*j2a9j2w!Oin^xm7~QPJC(xectl%8dqbv-R=EDK
z@qBhZKV|aaJ#XIs71wQHtt;M_V40<;FO#GJlm)o#?7~R7^&O0H3G+T04~aYqh}PpS
zQ$3;$obV$SO94;`A?Kw-H@0qU7XEsOut^^c<Or=<y;FDl1KYuVC93pJs~0}D-_gw2
zwua^-r~8Rz$yf)g#{K<!{`<fBANciu{xAG5{0F*KT8BL``lX4%4VeB?K<KaI??1i*
zSYOAVbjS``y=u4sH-9T(-)u*smKmOQB;S8#qU`(kawsWtkO@hGO6HXAneBkIFbfp1
z(D0K-pMmHIh=(yGA$bX!^I+F_097i2dcwwQ-D;rz(RugSkh{fJyr(!3sVUt<;<9$b
zyCG)b+aU@mOeouUUy*b>4&J=$<)f3dEBf31{@GKXZo$0FDN%I7?(T8ckN_5Cni5k=
zWEI>t%r;uBx7Um(-12xPe(St<AujUc>$g?8UN!_n%7T>4cfa}t*Joq78e#<@VT4ii
zWm#$F%yT@=ttrTx@ap0|7!tOOuBYEXE88!=FCL@p^y_F^`7ldibl{1rh5JYkimrI4
z-?~4+n2N4?wOx$1cbEKj>2`B$zr_jT5+43O<{Qg;=F_JiX<jHL5^1`DNr{f3Le7Qh
zaAX-WOtbKPlMzGddqw#&{bM*^%=qJV{PLW3_?h9~6a_N2uZ{T3)2r|02)mD5Ntu((
zlsQwhaO93+bEsM@{7=GEcEY>UrIU()rbyZw^TCEO``!UgX+$^k(wXk9)7n5w;nuYl
z7}jupE55_up|Q1*xa3h3XE$`(J??D0*WF`jd_J;=l<#YI{}CO_NJx_?5#7T$HziKT
z1I?kiMAG7<0SsEF_XV^0C3US2rpUO$?8T^er`v`}uu0mgm<ZK`wK+>&pbK7=wT2f+
zYa6}#@Z?c6<?u1-9?UCMLT-GB=ge*(98WWzz1%vjudFX#*6$9mF&KNkJM#qFw3|n<
z%wVeERD-00f$GtP50HKLHL_QQ^}e3th3(83cW5-W?ZW5BADE}Nl=6V6k&5Ddip!fj
z@9*F6-~F3kuq};`KmBwIEMPoM8`7AU?)Gg$e^KMV?o9rNTFZsreF$rfSqqZq2wLbd
z|9s2`dF2}*BH>5oAqxxnSp&&8K$`JizrBf<@3CK`)e>6|Nq-v!gxcOrGdAyJqKUO%
z@&*wv0B5(`o?;R;L%T7BxFxwIqL`jiVal1QM2+DR7vho<m<2ei5&lv}h=V&(A80U5
zU7*rY;xgIBB{X|@-ds%(Qw}6@5#HPjcV0MnX44K8X27cR`Mj}h6Dfa>@POMqnn>d!
zDs}@aZz+Txlm;0c3Axcci4Vqh`Dpw_AIy!mvu|b87&k8~rww*&@<=-F@&b|Gm_cJc
z?G6pMc{uht8^1PdZ1uwP`7>An=>fX88F!&>VX%EX+=u-4_73IM#rO86zv#7mTf^h$
zuiw)ST-?+RSTLn@u-&j%{JWR#Q%;GQ5+RkM%pA!StVV5a>KyWnYA5A_q(Ujt+<En}
z5@3##QW)o(?MdQshV@}SDr2Aa@thsTSss3QtsAwD1na>LmZ6Zyn#%Ba-T(mr07*na
zR4FM^5LIMYBE9U>!F(U;im{G%7FNHp#e?}JZo!I9%5j=$bK^8mA%9EGlru^7CY^Sx
z0Z`9(O!<-5EOZycR7N8@n2Bf_O@o5!fvzX2z@_!EZYsqa**cGxjmzphUWLU9_ix{m
zruR&z6OB&wI(*9=tReKVbYowPg2rAZ_iY~Rk9(Zg!@o1u)nH*cx3#FIU<s4KCQAw)
zK`K6KblX8lUVzjw+js%P$MYTMa`FL+d1u)!xFrq;C#3^uj<;ofur=e()qSws{AJSL
z>bF;C*$eaU>-eWWz5-ZZ$De+@E=Yvi_7a`Sy^9o#oSlIal=X4Q$mlXDc#p&FI896`
zz=NcPL-|171bvPUQA(AfibxshEIz_w!F2~+hLoU3f{~!;VB+j?w;kttUD;Y?E^(o3
z+lJ*CHE;_Z_+Bt~H#X~ByN7hp`Yi=!KIS2di=u7VI{=1f3ucD|J7m8s)REBRqj(OZ
zb)$%f*F(x2kKbVS6VK-}r}@b7G;t^;F4VrzTM3i;))5oDmpH6$W~*`OZWv}8?wxJf
zFjGE#eqvb@=lK&K-n`@A{HuS%<Hsw{pBLIyX%#*{UAU}`&(9ZfQtlrLm)bdBLF5;d
zd?eY7`G)&4=D`bf82}e{i`9L0An7KD(j1-wstI$UQ88DV<uKY8WwUS~TCI`9m{Jrf
zwF#T2FsB@|;l}i`5yvZv55xMMn%r4AUqA7izv;aF@Mqk={hs9#?11JoCQeG7(`3B4
zf6Gtn#>zP^zAe6wu*+1rF_h%hk1uMxdi^jY3VV<5RwVaZT7zYyr+yQFmR<5Ef_KMy
z<!+ui=ETD^2cs{4gRX*o9$jLo9P&ib#JWbpqud{moGEFJ`9PQjw;Dj9s*rOOaXU!a
z&Dpj_t2=-t!3yfWH*3KlmlD1?t+KSrVn%c52wo$W`Zy($<S0_TVOtAnL>Wb)#@tqX
z{9UNBQ|^tTi6UW4pL!>`Fz3Siw+~#`jq6sqJIox8nc5cUonD{lb)q-Nn$w%zK7AV(
z1yRMjQ|-!bQf!{7E-YQ?oAdG0PoW%YiENo<nR-3r)tJk#Kr=mqTVtu6#vy_#+B<|9
zd;uCsUIGwa?oF?O%>7rvq!bM!U>g^ViP+4$_+Yp>#fM?_fTlL<BniM|mnMPTF8DWb
zxe%#Uw+cE~I@x7_hK=j>k>6Z9@8A6k-abr}Nf0fpOBrc;&bQxw&;RlN@W18Lk01Gu
z|Nh^{e%qZe=^FFpws(BN>lpqk&g6fXwd{*gKW;<DQ@PJGu9<6VEZg=1dW60kJWSeE
zh1ymGrId+gopq~&%p8FnXu?baKStRj5>*t}U~VPvEWNSxPS*@QF7aD))QoGtl1)R}
zvtRfHWH6zP9#&c^G#hJqNP{B`M8d~LoJqp4ek&8lIR}ufaGEC$Wx{&+SWc6m6p9uS
z39*n32N5v5;T9_95d6@9$7WZ$XC7O^nsO}4VOHMW3x_^Ydd6p`ROjMGYn4wwuAtvA
z&woy{1FcSUyU@Ic#}PePl!RRXHy}DtVf8?e<8l~GQqhoxN)NEhEm$8bQKzUTkZ8E=
zqOFbL$kcnJzoi5W#^o#2!zF(AHIS-1xa}U2|JGRinYR9p$B$>;efXZcyF1RykzS4E
z=^1NqnC~g+h*HL!4<mBDA-Ln;<jY*kHxO}O%dgfk=xrYky5K;FG<=e}+pWgEftMck
z7;~8^v%(w}46{9us?e?-l4IF<bG$?5iBb-vGIKiJkxSH-FFl<&mr9wyJy?+rkOP#c
zP8v-MvpT%c2f&37ze=TU#<Dihj3f{bMpu@^lx7b3NXaQYz1seUUQQ*mPH$sYiPC8_
zCxTTx)}ojrR!Kvedv`hk@b2M=D&(w`DGf#<ggka7<pXby@A&cQbAZtMxs)Aiovke(
z3ACeH!nR9iQo5(-cWh#OT2~5<%k{!tat`YJ<`18FymWs5_!cR5eEX~a4oUa8+*8@u
zd?6XA7}7VKI-Yg3w{M5x=v2r+2+cggt#s_$))7w;gk=)-%4jOYewq;z`VLP?cn>xs
zWSA+@dmFYVX-5rjQpDZLp?id|C=qUzrO;jIbk5h$ob8GC;>_~{$O#dl_Xwi8URTz-
zvh@gaNxnDp<7{%_1}g3=;7eM^U&sIY@fE=OI{swib!SL^_M0A?`m1e5N7mCz(nNqb
zz9awp#bjJgq$%3NtWmOtS+``8hg8SKn5t7+qr5$F^~SFs7iyoWG}gAUEGvE4=q(aT
zTztrj9BgMx1(i7^4MWl%-KUiDIBUaSCx9;x;HMdP#}jL9T(^q)=sJwt9#cl47yTgb
z7D;;AFgA=jx|km3j5>@E$K@e=qhK87geItxdAK{UwR_&aIR#MGf*f6@Le7aKLMb!H
z<2_4@g2ty65huIQlCe@p63~ozqW8?FXE-^0`*!B1^~{g|-+zl(L3^T~p{>qOj|;WI
z!|6S$9nB5fCYD-|^bI+kfDQE?z8*xuEWUHx$Mx6r_P%>^RxxIct{0=*<el*Ak%GDe
znAKx~b=OfaD@Yh=s)g^)9_ia`%%}Y}a+{Pke62K6EZu=fl3V+g&rct@Zk^-&fl>~f
zj&G>7@wEM(>v|!5exfc5SRdXC;pGzD#g|KO^7x?ysq9M_f7d#WdSP$t#WPc&5NUkD
z+=%p=0C<JNSi>?dM#+@a6A0vLXfmXPDeIg<3T&!`Z*Yx3g`%uttbOC@@{Gv|R7dL3
z@IvuXLXCZ-HlsCvNw2Y>#j9jg3W+GDY<R-Bcj{WP+BuY&Y9SvgS}2k@<|8Q=mLVV3
zKK!`)=qkSO{Nmc?PU{sFXOcoPNaD;H<}3k>r_BA`#B|7<nsN6qQ;PEWX&c?+hFPPo
zjo$b0ivX}(1|uqUOhQ&C!&As8n9p3=iPMz$=3DRveFa+u--Nq230mWMRhkGrDVsUZ
z*Njf@nGO$Fy+=&xb&CQ~JL&=M_}G(gGN{44FgLn^7)(jRuWpz+n+_p?g*4hsXg0VK
zyNpT0FRwSpO%ch1<>sWGh7U&@pw(dA>@(IK)jTe}1xIPtxLhB>g)+S*<pa}viX^ml
zW?NT2ef&gQ8ztoc^Y_=|(X803<1cX?!+-smEH3{bYdNLNVNzTZXrY_XZA0gb%$c=s
zEOn*TO7BkVv7e}FT&^(GGrO<cx~|l=#HGueVugY-PbYGkFjL$!A`^8xQ@2i*U}>m2
zi8Q2qBV$p+6R}wUQrYrI65eleLl$E~vhg~4C5uvYrnr&~j)%<MAv2f4={WKJ?Om|L
zlX5r&Lu6T(@XQg&`9R;mDp-H@PzXc&<PmycO|T@mI*AF@6F&Wlp5eUy7u<xB63Gla
z!eZ9y{O%{?yk(AW{*sj6Q4YVPZBDO7U(Ug9<BIq)lEEN&-D6hlta!5kBK9?=@wtb0
zU;-ojFvprl16)8HS0~KQEC|#cyoJnrp62*?t)VS0q_+SJ3x<7386PfrIvPuvneIaC
zV1A{0<@u_Ve#Q9(!Gv|5pP%>-|KUIJ>GMZ|^A+#GdcT2TAv=pl{iUttpQ)q<Ddl!H
zzhcaZLngg2RC?@ZeGZ2DZd)&d<DXr~5?^obfRa)ndt#mh(~j$~fv5>Qni6fR!G4@I
z<bk!btsA{J<|(e3wQlh|C0LA2cA*+*iW)nOIycxldH5A1@I=W+s%=PnB1`6&jvUey
z#`Arp`85p5eQ=+L%E9{!0}B-Idoy&@hGYtr>~NSe$2l|QL@$ZE<3!FW*!Man39kTM
zNo1W+(u<Z8fUp3$Vz?*VW{NvyfYR|9AJC%YH{kWkb@|MCUC?c1u>&$?e);Er1^J$Q
zJYtrE6)j<{(tX2x#Uef<fHfifFb3sm_vH$qhYc^VLe62m5faWEii5tpNXiTRC(%e&
zDX|}nec#QI3^}K9Co^v0;)Fl(rLN&eJbcW|M|$!|Zru5qB1R;ge`2c(b4ftLQ{(qP
z{5Ajg|M^FL^P69@b2Q_AHvT(UW;^KqQlEZ>u>LuWuK?E9@sBoswxx)?02r@$H^88{
zzYxoP=NJ16M@Mbwv$cm@P>5K1W@ISab885Rhv`7FNDZl0sYEK#x^DDs!&|#uzT5+V
zFoV`x#CH#NRjVzE=2S=;vWD2OLrNx=X=}%taX1`ssccQ6yW;`U1-KTeVLL@X3^04J
zIlSStTVc^(Nf32hixWnsGZiIeAxYwNoTzr<{{F~37oMKZI8G^f6a_-bg~Q>5J}<*>
zhPVuh($uN4v5`ogQ7?4QJYB%D@Xb8&^!S<I{FmSG=J=Mo`2oLn`f7Z-YzPzYj^E?5
zVVp59)OtYjTaw=4_Awy5VQ6W-gH<7=ieh7*1p#7KjA<Bl(_<b;k5a3XhruMbGcC;f
zho2WA8Fh<<9ZexeVo1|=X2I(be^)^ZGJs6)jcN<^@)<6bUw(IDo@VCx4p^zS;@cH{
zex_|3d%6*DlfW@v;($+3_rdb}LaODxMLove?0RXKhLkf(%>iD(4M6InJLmf)i0I%a
z7qX{deDsjB$mrmgGzIfg1>Gv<!gMUO0#}K3e7RmC&U>1|zu{J>#%pp9pN%%Wb<C%b
z2@e1(Ngfy3^m0K1KxtdUnltAKF~{0qPHAS&2Q+2QZqd<ni2wnU0l3?3KMBu=PPZ23
zu$mF;B!e=+oP<&o@x<|1m<K339}1d;$EPRU+h96$+Sc(75zFmEz5~IeG)#<2jjl2o
zJmWrb?FS}izI%JXo71kHH3cj6K7m-{Y#Pma6{;teZQ?k;WuCvGOM-4JeE{~@NU{rW
zo7c|3J04<$!L#k)QVO|2bWIg=8Q|6E+WYu9!+WsAY_mq37m^W;3w4V4vSEz7&&wg>
z9Os4*a*qHg9rMPrKGS>ST|Q9e8DXYjZ2ru;u6+LZnYLB3Mw;jjh6RY>!&6UxwvOdX
zS;z2qujSh>WSp`5gRSM15{FU(9Ft6IHCV6r0Y-1OvGq!;k-!(=>^N^*i06znKuu?>
zm1SvUok%FNhhNoP4#A4FxNwPNT65~wNt(&pQE8;4OeEBDoIRp2<V_(_%Qx_5zme|E
zv*bDU7q>_jlSIMe;#(4jDKnQu_rk-&fs_-P^N_L{PuB_<eNuXB^co&lzMG2!l6-)6
z4Ke6_@DT+XnIs3id{57v>*d#|z{!*vx?)<dY}NSjNvJ+^diN{x^oHs9fxaxXb>Xo)
zN+Z~mTi6`<V6nYo=IsEMN-!DS_r&B^?CG&)ZAdv)WXOqSG&gcI-pnE;bTiUHDJ8M>
zN*Bdjx&c`3(99@V$T=?T?!)>(giMM0%nh@Z?iXZDWUCzJ_XrtL=X^Qy*MI$AczSwz
z1z-)khyiT8ey=+=@ujTgpS{d|5#P?_oe>>-VtU1f^O4kQ!(_XRzo&8&fTNNSm8eBU
zQO%^B$)>n8Oe$sx?|rbXoW3<w6e%S_a9XFf2+f#FeD5~obSBUcx(L($Z>pjxKqYo5
z@sN9lP9c%gOim7Tk_dB}nYA!!#=X+|LUBW0Gjzs><$aeg3PIx*QVvE{=}PfNa%akk
zsbor0dQpytNFtx6iQ0p?c5A3hNKzw7TyKfSVUH1K@y3REj`d-&Uxvm)5uK1!NXJ6I
zCYI$&U#|3XMbZf^cYOPce~G6fmkzBa+7dNofV~xC!+irH)mw(7Sze#V?qQW7680zL
zXr2eU7d&-moV^UtaAzqBXq>$}V6>lwx;Gj$Y5;L=fLZs>wQr<UB6*S^Cy2lDDFY%S
zG-~{HdqQMEIiV5+fuBD8$ZvoAOl{#W_Oto#f288z<x5<?jCK5V{QbsP0PE}cM;*J_
zb6f-*X*CKQDZ?{Cxwi09x<4GktHiEx$hwhGObU8@gJz=~HdKR;(7bcDC74^i;M}9y
zsjO7J(sIYt@vf|E<LSJR!fYpGG6I^TL(^<w^N_Y&JUp=OYo(ZwOCo0|Qs^FlmFI+u
zQ1Ts<CQ65`JJ^LH;ej!yh=+Y+ofG%iltD6`BgSXw3qE`)Qks|??t8hl2#?wjGjdUM
z&g5hGYussK-8Mc0MJJ{_fmXUTp0AB5!|7lgrz5BIz~hg<XQ@x@K}G5Tn5FvzbusGt
zOi6DjI`ekEhiT&68+Mp^|1XZvCzf95rwyNtw}(n<iE>GN+9rOxC`tc}H*daSD-)f{
zRaUGoc)udPjvS{0PNK^*H|jMpFR~e_b@F6nby|<8ZA}Mq7Y=K>9fNK=Xpn}5RYzUP
zJd8xa=)$L@99sBfZPZu3T0Xjdkwztiv@%&5)vW;wNayMNpSdnaAQ%JYE45#vt8Edi
z9=s7@%7rB1L#E33d34ZjZPPDh{PD$ibbKX^pEn!kzVW3Ur&U9{qbDJ&kPPsSNF^u5
zxFgL6bTZ1^!^6oFy*p2J!$!AO<sMDWeCSANs9mtuSZn3!vaq?nINePXup1b6Oe)(}
zx%Nu!74O1(YfK4FZ|CUtSqqRj6&gnK%whgEf&yBiUeB!SCl2L3^K{}gpOBRKe7OKZ
z%Y?D5^Z=WzlU*q!L<$K-v>_VItAk`FgyWr1?j}-hB!Tzbb6z)|mK9+NBYI6J0hlM9
zD7kZgJn-$?4}3a*WUW_Hi%zz9r+TMNJt+R>xG&_cEcKpLPyBXk&=;^L;Ebe&-(4$K
zp_UJ9P5AU_LDD;Z^{;+Op57qkMEmq3ZC!9(aoYkg^Ow00UKG2Rp|U3eWfvwg(uYBF
ziodjNo#sXxk|_W+MQx8x55P_#zy=xLUq(}-1+cUyZJ7r^rlVMng1NoSS3uP<zp~bq
zpFVd|I&jQ?fxA-unZN$^CqDh3|1F<B{y?<~ZYYU7jDpcHIM2xNW*okVU(z~;zqOY4
zwvgrcL+y>^O14DBsA3$NpgqBTT+U>4)3a`<c8X4<X(G-1d@VQOarJR=HA;FznjV-Q
zo{$+d9iS^ep3m%A7a>$M@*%(<wLF+b8&x|kbw1o3Ii|u^FNiqv6dXYHP8Fs3#M%mL
zop8TLq(neV->7xr{q#Mj{Fb}^0Gi3VVSeVU6SF1Kn$Xn9T2X6g@02cp(ey$QWg>Iz
z!fc7dDKp;}4hKh)^8W6QwRg5|q?`~<l<5$1=W?RzM6E`4;UUH6Y{bQ1tdphyA}&?P
zS;!N?ys(bgglK_x|2<FJGoL>G#GzNFw(|JY*jnQIfAv42<(}#Ej@}b(y>hWD%~w1x
zcwdm^890aJSvp!O)`k&|_}EcjNuMoIDD-Z4ufw;C;rlrxiQN!WN;{xlX!#O*z7wf&
zse~lcA$NtYYy&(b#ky(&S7}sZz%J0)0KrpiHU%4dgwyIX);6ThK%)1J>ko}ToGX{-
zPwWV{<Mh@7B<U>-yHgtN&hX*;QW#h_084gWQ8Y*urE7q)sRySnc>sl`%F+$-&e3%=
zR9{AbM@rHu8h$0DpyrV@h-7*XhMeUIJYh?uX`inLa_^+jT>}z<IXT!jh%+4u%T{^1
zJkmX3D06QVoKzy9We0)cJz97@BB&PTqf*7FPaCc$W+pz|zv1n}U04IOGwx^H=I}<-
zLI2Hx&PWa0j#5fA9wd`Zm>dr2z#+k0gsB)s8%>NXdSh|z=1|MJB1y@IyGZ4%oi_Ys
zbvKY2cbrz_)-<hd(40Fjh|HtzG`t$NIkpkNpiCGEGq|n`QUp?oZ)%x@Fi9d4byk2t
z4Ol@NN*{aa-i#Qk1cz=7>j~>6LReIh=A(v=(|ecs3%jMpXfiWiF9G;V2Wf<)lZ<K~
zjjLE|m5L?7ROq(%jD!7#juct%p^g|JKWto<Uk5or!L8#KfdQ}FRzsmP+Jk%AWXL29
z)0eOOI{s;ouK?E9@uwRvt-RYoyBng)-k=*_67GxPMG<K<?Pd<B>;{bSwI(TKaRufn
z4hpZ4V$;TEhNsv7MQL`Sv5m;wxcKxIX-BRShOIJsjvmIX&1%GK8*E;&;hrqSB}oT^
zN=7HZ$7ZjZ<FY9c7i`v%Vg5Lm11^bEx-#oD6p@NahOR-)9|wg>xMLL&OhOW=L~}$_
zLUTg0kg`xDI!P~|dvvkf1#2Va$DKK4rhFu8^nt|m_WC6$S`vM$G;gRlNd!_3>4zsI
zDTh0uu8ForhYIb?yGgwZ%?j6Ec$nUB$RBW75U+UMFZW?Msd#v9jPlEGpboe0nOJm7
zqdU0A1;>m@LH`(+=g}1-{E)&vH=#a77tug0USl@B8ENm6^wtmZUG_CD)QnD^8Vp9(
zR-b8YL*z4N0&JK^r|~tVOJ4uaEu=Q1)bQ~hzqs*77vI+ldN=Ca0W61GNW#sC*I&Am
z!L%CYtI3&4M?|q21xd9%f?3G9jETRD#$d-SQa`uaKr56?N-_+`qr1FY_IGc=OfV;8
zPYqRNnw6{)x8||G?9tFkCsOL9Jb_;6-VhP87UpS2Q=(i}6d97>oh@oNU?oLKix3He
zGB&&-iAj`Pgj5omVh&C@b6q2B!S_OVRKh|eeyKQ9$(#-+o|eyewZY^ImV>$zYKCA>
zbdTb%_Qci;S5w3*ij#&{;DXUyk(B5tt!L)>$lb#?NSbji@fszhjPd+-meT;QL~h+z
z#vUB9mhm~b<K9CGq`|alZX+!;zR7OZyqhkIzxb8O=y+o`yi(%dk+E+F&^e@>w>4(3
zKEnX{!brHXZjQ=Enx7F8-4>qCXFi`lhIG(g<TdeH_Ddo|=E$pm=L;M^Tg$_+d;D7D
zhV3&}Y+Q)s)fxr_L#4&MmRNr&$<NlZIaCYpI1@*5fuyAL7?)%T-;`?|5+o;Bt2&BP
zVR&H1It~Vyk#jU}HVFwy8WMg6V;OXW9RQw*+t#>-c_V4&kdGV&xRk}f8+~|BVG`!e
zJ6PudZbZzz1#i=h?3p65j?<*16f&%7XAkvAJQj)I2f^tQmv`%N#?B?<R;lKLtsOo!
zy$4GrX*3EV0c<cyB;)fvwKpyeT65-RJa0}n<?X{;@^nv{C+g~WjYL4xhHFQpBUXVv
z5}qyAx9rdM1&{{D<udj)bH~k)6m0VD`|Ap*X}{=)F+Z=?Yk2QO?YEx`alQ82I*q-n
z5q1tcYkBbFZiVCEvd3lF>hQ>W#!;$NY+10&Gg!aPS3D$y-Sy>}@(SK^$G*tLeA^`%
z4(GRD2|m_mEci~aZeO1A`1_--igP&t`;x;*y*Az3LlI)o8FY5A)<|_@Yc0}tHN_g)
zzo}>hAu#H6j9OPr@{L)E2jf!r6d&JNLlL@ULkcIO)(8+J_KxFp;LYj8^A)-@e5<io
zn8(rSQ9JAj4e$yuNrT%JiNHC<T1y(<QYwD4w6NaCwDw5M^+dG@X=**@uc(Hf-M(fZ
z*h|QMEYi$|#l_xe*}<dTNh9`&KK@R(j#`6};_29{b8_s*62Og1bO1!h*X%X!UXzXU
zIZ`NX%z*vg1AA5Kz6*)}xZt~eqy0Z)t&fldd6|b%u}3rbRlLjXJ;!w0L9ubz3=pmL
zb9~(Zt8aV7tMAdf-Ogfe@9iaA;>N!FI{rzIuK?E9@uwYsv<KQKNz_twj3knHRKjm#
zze<WLwKXb1!xtYZCWOC&N+KbMR74ET9Zhk-rIgW}uoM@|pFTfwwll4Hq#w17tuEv(
zqsJqf66@0O9;S<7{0m#%BJ$bdfNEMONhyaJO?mV^w(#`Wt%iLK6Z!ii&CYy&z9KqN
z%EX`li(he=XY_YQu@g!S511XOu|)Plp^$tW=5q?Ve|3a{Y6+eGG=&^b`Y<LpiYu4r
zE6a7IZWR>nj(O<wJJWIE{`7|9+c(VpLai6v`Y7a_NnsIiNk~q?Mx!B29P%Bi8zROb
zP54}R{#2>0bADd1)oD-A<_A7}{~h<!JC5auUX*29u<MF(MRg6PQ{N4!!-wQ{OWFY~
zkEEn-4KuwJ)7AP&leCeH70EF{d7t8NlS_QP9f5@AD0I7Z_GFawv+_Kg7|eDPg8(YS
zM`*wF1|Yb#EBFHP@5YSYcmb|PQAe0}UzbW2u-d3axy~?Iuu_^v>f@I>oD?!pi!QB5
z39v?A0IMLmM|aZ1$(j%kLv=~eMWSd-MwSRd@SNz$KpLvBSHmWq&(BZPv~oSHq_(m3
z4cnC7o&0czrp&t3@Hwd=Emm#R>q1=@n#O!s`$i&@vyvpJ^wuuaUa=0nH@5AOx;2~w
z#xom+keT$rufF@h;ZX2o$TlHs10w*}+YUEMOo`%r()nf<E@^_45D9rxnZhS)>*1BP
zE(=M9gk}vd4IdZB)*8p-k;CbpAI^WEo!bCxJ>-!plxar0(vz?(8(W8!g-K>|&SXiX
zIZ@gjsm_pvZGoSDsHo5U`M>;2JU?J%4#Vnx!7Z3q!LsWE5Ew<UD(-2RaR1TONTOiv
zEk2`^3+}M>mF|w`7sPX?{fGCJ-OAk;aEbutcWK|aU_>1qJ?LT2^GAhp6!}RAIjDKO
z2j93XzaiFb!>M%E@Ktmt(hlMIdZnn4d!uGJXJw9_Gfnx~`SvA_Sj#CkBnjvhNtiUE
z53doCE*mF1qCHUtIB8@|6PBSQ3jpWUT5id3HqefYbzHr1J)hAI?@kZYwsCoWVk#3Q
zPhsrdop!kbj>}4GD_cEtJ+EAr#`ED0=FV0(lIBRLBAO_w&ZM2>&xl=73A5r)=SQ-4
zj(~YbwDT9ge9K%8{P-y0q1PLeNw9A<>J{%}%4a1>;iPZEXT}mK&*YNHxsa10$)N}M
z-fLq!FUaPoCVGpt+;zbY6FxQO`xEb`U+}w+zoTBaxcU2Ft0sZDpi{=Dgyja)LQNNv
z7-m-_ckbR#lnS|J9!jCxfi%sSDr>#sHQ`q1%|e#xVLMQ?V7}4BKswHbYMhP?IbV2S
z2|2~3-)tn=nb9RK^_vBMY&Q`Xfu0vkR+1IOGo1^j#wH%%pDB%CTSBBp?+JBr*_Xi@
z8R0IsgoXiPG16RR2a+pW`w>Lwa)Y+UE*uiY=J32+nUZo!uzKf$aqvu$Sj%7Rwbyy6
zorfnQBNRhaNBaXJ#wGy52F$lr+(-Ijp$sr~nuE!ewa`}Qx~^=#viU}`hMJLk$Y9r6
z!+hT*BzQToStQ49bs-MO@SSbz70Ll}NQkvNhtnM;CrXa6jIA1L4SDg?^G6=f7d~#z
zwXL|udn%%Qc=N!ozW<gVA1^G6^R%3YY5R<3Me~G;1_(^loU<M<S8~q*U<u4cIZVo2
zGE>@{T24}B-8R;BMazV!(xrhKn{F^o<kTQ5`R>4asniBK8ZBB9Vj0z7cW<`QlhG42
z8zeDxdIpXPp1?EfH5zbtIe_A&t&xNdbYJOx#jPRU$vQG#-CoXQ+2?(KnIn6H$DFop
z;dd)gjqm`~0lmGJG<$%Trm>p(NYs{b+{4~W+PfQ09>2E-Q3?LH0Hxg*R<*qRT4mIY
zJK+2xg?%;6F>@}<l{qKw@=#r1qe&nG?EX6bIgGCW*4OcmHg0_hf3-pGX>`L6YCmKW
z66s()_H@{bqrvz9c3+siP(c`peHMCeZ*;F*+QQYgTesXL42D~e&0RDi(ye#Ag$%4A
zY`uk!e?<M71ivRFVxz+!Z-Q1EwO4AZwBClvXt0K4q~F{h58T}ydA}TCn+En_gO$X1
z=lhR%zbQqbERnMtMzp&p*3hvxkGNUyMyqX5mTz@le5RwrZsSLnaZrc^n=LC-i6YsQ
z=;+v<fCx}R&Z!BQKxXnN@Z)_7c|IbCd-QY+`J5`UCoA4Ac>%VKykmBD12+c8&&j`T
zHgx+jd-{!CGYZ!~Mo1;DTzi^QOc%L5E#Gg;V@CNtci5{}-~Z_4TZ-Je$h(=a`#R>`
z?YQNprHZd$Z5t*+9-b;)UjF4llKUdZ-ktOP*}?D**(F}`Ge5H+P&%G=aAT};AtvgO
zyUFMv`u=@J+sy|P^Riy)X`>YzpyEp``2JLTABjUP44~0YZJpK(Pj>4F4^sUo>PImp
zVQkuJrS(d+##WuyM+bSBww~shyW`B=VaCX?j(xyifXu!Bt)umvQVvEMjFhz$?2hr-
z1?w|>yn5^O+Hb&4kPb!{7mIbDLsn<R`cx6hJfS9Ghq1FTCD;ot4VR91MpI%gGf4|c
z47t>}z`BEE=EEIbXEqgRuh@_KdX-oM`_FHyR{oAjJ^q@FeD1yXAFK_(ef@2=$m{jF
zTe83W8!zj}e}3EC!L8j)daT2nN85N=JNtKMXt^&@D2uV1@5}3MG~v5M(s2zQ#rzF)
z`*PPY{Cq9Pg_qdph*aG0Hi^!Er-yV?!c<&#fM8r4Amp<@U(2!A$KT&O|HiVdWR>A%
zWc1eXT#zJ@vT265n3GM6-fJX{?*2O4#wPnY`Lh2oUOza!H+pY3=2Ej3XSXIw&den<
z&l7XWq%_!o<Js?iOuNrlto=O<qy!s65-Ej0&}|<Xv!+{sMQy8O$&e(NmEyr7wD|WV
zD~SnI1E@`r2v(AYH)S6hZHqcZlCUQgi+56TrkqL1(M@Qz;4M>Iu*)q4J3N{~uUL|S
z62)_ueJwMd<zT$<0!Hr)%Se)oIk2Z9zqEnktHVnx_9c^nChx@v9d=JMACGFR<^3yu
zc1W@}xAq16^O<`iQpXAV@}7P+!+ks*sI8$ygU(d~wAnA>^0#a;zqqkK?bl36A0I)~
ztu{Ox2avD!a{i$MP<%n<_x&LPIU0TZ&w7in0ZD@x+o3Be9nUdmyE(0OXl)dvEkNRC
z(dg3fBnyyq?9H#<i}g;e+ie|3v$x-XO98S9B`c*Ut#>qUSWmH*<>hR-F_s|e=UekU
z{J%7X2Un8#+{|MQSL<xGvTYkh6wNt+B6Yfjq}SYnkv+|@oH3x}akUo1doV8P;UQ@B
z@%qM87ef=|DPc8!AcN89!I+Bot}k`ZtMfSI{QmOc`3sQj`@`Ssv;koC_R{e0<NM#%
z@sJJfbLh4%_wni>?Q8f&RWMF!*;!q${?6Y*hvjem>}OlW{=5G3Rp~gjMc@=Uz5K;L
zdTn3FKmG9)!1_A=QOCH-M`sQYSqLAf`BwZ*gug`DASGu@VVar<W|m6A`-Xd^Zf7JZ
zQ@$ro>6JP<gO}~hW-Fh!&#Y|p;$fbi!!K)DSKd54kW!e-wymPI(hyb^wrykE7E(Tt
zM5%S9Stn1&U`1(!LyIQPpMT`MTzGnZ;<8kpE{$zFQ-FhHPCE1byEnXle}_JN$Mx}^
z$NxwqqF5tX%>|(V1@}%;;h3hu2uP4Rl58Bha3~qWc-|UI+o;=0ua1i_%?HGVtv04q
zIF&^6#`ALKsV+RV6*r4SpojbDJk2xbdggMya;+<?8(lj+3yo7y%cq6Z?oqvGOP?XF
zTouysj`!dG6?GQYLx(Nn)#>s?_X}AwT{GT;>40pwL}3$`?K~*ZF;Dh$sI|yNPAQJk
zb(jDn;3|<yhzyU01_Ekx6faW8pCp`+r1o)1jk&m=lOfTH@^uskw|ZdkX-4()f>HI`
zZ2tKxj^8!!Jf0t!Q{r7Iw@Q02TK?8q{soP_=nCX1E^o*cmrT|C%2eA)?&xBWobA+x
zOUE#OmJ|t`-f?M>2u6*h$Ta+%%-GsWvyG424|uA~^MN{|n;{ae_Cjylwn#gw^#zDt
zjk-CS!-LAKhoNe7BpH&Nk`HKBo<IJ9%XQ)0R-V?CpFUlwehDI&2w9c)_Yb^%bK-Hm
zus)wyo(K5Tdq{DIfx1efq=Y6Rr||f*E|e6O7DXXrSY4@ZoLgn78*STIQ1b0!the+i
zlMV?aD_schoCMQU`1afHu%)tmJhPBkP&9>fnzVo}SU*ta1MePYSio1|Q~N#Zbpe@3
z<&OF84Xb6W5t6OmZa}c;uFI;(@G82=PY55t*OS|a0e){CHzVg9>(~O+&IA)TWG}k}
zGtl=PGJ;B!D!Uo%U{A_k*zQ-8Vfecj{*1ytz5W`axE`_#{1WTY?d2kS`^vb;JYAld
zbLMb&fTU=O_VX^Q`|`&Qh#UrZw_uv68o>m#;a%d6FOD~(AKMOJA(9bE@%w*kEgNK`
zdS&TbG)3rq-X581;{7iSu}<$DRLIjrU$3m&N^X^I##66^ltkDzVOxdg#|u^}?~+n-
z1R8`GZTDMIE=%QkS%dAGI%XBsnM1nc{rwxhd;5-Z2uWj^H{23UQuyHMHP{1|NhGpn
zC?(h$64Jhga+nkIq^LJ&j#$Tf=ekuc%azOdDK7bwU^P-g8J>-uiYG<W0Z%#P&xi0_
zyDt;AIh!}G%~`LV>6n?z9P%bwu*nYa=3&Aw8DE@QoodR*&p$w#xO@K)$$=@>n45qP
zwxKmlT0H;3X50f`x^2K<MU1nn){0qWE;BimNN%*9Wf4;0UBb3YqMBY}9omd~M_IwA
z001q*TwXxn*Bgfn@K6Q-K794k4tDQyA!n$@2VlzU1^jBnse4_QIJ55FKo+#Xo}_0#
ze+6IQ=wqGe*a;(aqG4o&<b>Z89k$qyeGZ?%*upa+8mWEN=o<CZRyVfZSy`zxDm~cy
zt#K$vWJ<KUk<6n`ox+l0+bZk2U^Uh+VytUpS)J@gvL4MMk6J=T{k7}GYOO4n>&>1)
z7+{kORvsPoMzY2<MQxW4&(UaZ@8ex{LJn-{CPOPJ6K4NMcIj-HAx(jYBnbxCb*+3p
zKeJsIw(G(PrMS?nf(mt7kTlV#Ooy3w`905<N0!yOe|Uq6vR+qO6V^3Fifh~Gte7r{
zI+~1>H&Qn8JhS3#E|hJf^$4;sS87+RYV08ctcsd6_MAPa0LJv!VP4Q*)u;ne>b(Vs
zc$kr#Sz9C<qbY#;m$ty|PPPx%A?X`ueym|1tf$vs8|QJzx%XfH*R8IdjIbkFf2;;H
zhs$-rr_94Vjk)#Z*VSLg|NrAFfc16!<BkEbCou}j!~8%dllvUeu(TW5xqA4ZoV<?$
zCp?L;T~ZS!paN{HZ00m7TkmYmSZ$^H7BU!h=0j#GX<T;4iy5?9GY|GgjSKW<PTdxI
zU6G=YGEG{D%lw78!SHZ#Yj_{*E4PrJ>E4-+Vk|DeiAiUg7F2t9V2wi-A~7p@#%&8`
z1=01@EO?5pRnFlx(p#gtak+M$uPcwsIlKf!$)$iQt;Ye^uwjC4h(<b;Nurd(oA>V!
zLf(|8LOwLwb)_}v>X4La%9K2l^^TnGN!;W8GT15!Et%7sBUdpNw&?bn<2?>J-=oHO
zDVXD-Dtr3TO9wsvBcnS+hm@!BoJMDSoQI+>thJkehvb{@gN!e>@w|Q4@9+I0AO20_
zIYpwH@wZ~B-Ctc+U(6(54$RjV_xSqOIz@zBO87t6=mcvN0$=nP-SH$&0q*AqGKp*@
zI(pau6%(5qx_2taI)+z|+-0P3by~ktTkzUet1P{a$mlk{R6|b2c!oY;NB3L$QLl}@
ztkFgFpv(7Kp@c`BsonhbUW`i{)`hmMv~`WIcVnJ2(i2v7bZ&*ni*C$0V?|kWjrY;U
zoR(c4ow#)*H6%%z1<f8mQd|&Q?X1>$x~`npm8Gu1$iPG2pc0>_b!u-E3lAO@C2Jxd
z4$Sk+-TghAE-dGTz8JLy8^#qx0|dUCj~ojmdVmh9Bs~t_4-gW~f-xh*S1&N7SL;`z
zVBWp3xREU6R`TAaiTjY|##$XlDDjUuH?Iy6eF2?z)~BR#sp+r2A3tZ2yqZHVb?>O|
zZ;7C<R;lcJj*KNcI?OjJ$IsWL1Fd(8Dk5=3_r3$@<MRC_m(8QoEeLt|9^Os&WRzl=
zfGUkhp;9($q^_9d!Gd_bmgR4)<*h>(XYDKJwxEN(*K8z74qwXFD*!2_8(`9EjWcdt
zc7y`2onnS>3+4?uOsGn*2@uqUx>o8|v8~~?VOvLf<2X$mv`|u}<jhp^@Z^FtEoe))
z`e0~=r_U{=g5iHAf@ns?$pV@|vXgRn_$*7~`MUD)`I%)|SgkVkigyhuu!ejq4Mv-)
zGNt*}pnLn~ElHJL8`U~Cb*fd?TA351tha>b)7^=Z*6`V44xox?Otsb#Z&>eW4^YF*
zN5XT%dIhP3bT-zUXh>%F_2}4*$*mf0QLS~R67u@qgQX&>k>cn3tvV7$b<}#1CL3O7
zc^sl<V=dbYJJK0tfe$!*7-;Y78W>(&+~&w1Pv*E+-<)w?{%9P}W;Q}2;*8dK!?=z!
z{(r?1`fp-5?6p*a070kYEfrF`L<*!UPGa>AUb*u$+DSWLG3cenkgPddU8&tMb++EP
zjzn+KPP0b0O!v`N7?SRN-*3HRbsL|1XIm?MSx|Gdhu5tu?CFnAFvI!)wR<yKHF|fB
zhXdM$rA10$*4RtT8Y-Qfj67G&GOJXablcx}#iHN&So?a`q$Ksx7!DBjx>TMoE1%A1
zyftK3Z1gq&+nxO!>!y%6>Ve~YjO4}n0KjrxgJnu6GJ-)C-g=QZ9{W%Nv>Qz)R3YbR
z(){q_4@f>RJ$$%5<Gm1##Pk;bWtUv+XQ_<4%{V7-Mf0eYY$yjL>Wuw8?fzPb3~PWk
z7OIXjc?Y6={NFH9-x*-9kJ0ht?_8?v=lIKrwfm|OJp4reuE}m5>sVC7u<^B2^6U6#
zGrj^?U&lZ4hyo{L%n80ZeMdHhi?Ok?;>fU8ybhm*hhf?n;TUm|GE-bK<P6@h_RQLx
zDsX8V&)bUSNCsO+6?T6=aX4i5l%3m#tP%%iYxHHI*3P;%T3e&gn;@mkHW`%{Bi|q!
z^dV*1aN9z1Xim;Yw8uqsr`WzkZyzZ#k#hqZW|ro-3E3wy6H9-Jm*#Q6<{WG*)R<<W
zlqm3*YJ7g~e0;w0<M}glDom*`9S^8x*4nU=nY>bHI6b~!DVW22Jn-G0{ereuo_=~H
z&xLZ|S+5tC)~E_zl*4I?BtJ3-c~8*?tbJlz7q*%yQ{my8Bfd61Jz8{ry3>22_cRKa
zV6BBAI)vVzAtM#+cIq)|4hczBHx9=Ghz?Wg{qrA@&#ptV)^QPD*i->}1bFpxz`|ue
z8(*Kg0~i@5uO1c<Lqhnv2zcp0338LB#+$j}Xw=r3G%?R}Fu!^YKd$t5UzWd=5sHrh
z0g4!NQr@4wC7Ys4W$l$K+qnEC7Vnr>-de|WFA^Draxm+x|Nq$g^Cn4>B+C!{v52Uu
znY)KaL}u32(}Ov2(9#m&`+tC?B@kg3yTAgV1?K9muByBu!Vf!CRYaKg4;E1~kBqFT
zo<;RkG(w6ZD-Jg^HC16@=Esj8`{?D<e4ru}Nu2A<({je=At)xUmllJ!A6Bi@s6_F^
zq~cmjWz8&yW24(w%Ib-*j1^6Zx*w<(THxpU5w%oGR@U>0^?af%D`IeeH*s1r)drtN
z2WVZ<64{Lj%Yox~Ml52v4JEKdlz~V>C=FCe{ym}+M<XU9jv&!k*UDkd{P=ugKAxDD
z6NWIVG2QfiuUIXx&d9JQ5nRV64&1+f&HiT3+ppeo7@m1NJW=Mtyew3;HmE`wQ{c_r
zJ!3SItpph%45V-(#YmI_46L;{u+^AL+5l+Nidn{c6kP}F!e#<#0IEP$zW`so7I?>6
zi%1)Z-k!aLK<}xPAWZ`i{EbSB#r=K?ZMsnd(rl~Pw}jIiIP?m=Y~Xg6qP+lP-PXHU
zu+a%)7vQaX8n)LKD-`uMP&36!ffQSN*5|*(B0^vgWekD0`>%;Agn4Cl6}xs2MrL)O
zsn!t*KFp*YdvtxWpW1TdRL?vvM}{N>F;-Itu$r**T2_!YAuKp>k(ZUcEG*BDbhI;S
zWU_?JN3;aO{wqQf;t;?bIH}Jo%kz;sFR)~EITK8|zqw^1G1<UwbS+64Mu<k-y(d=C
z=mY92xdF0X2ttI2Iw&1eYoDrwL6Jd7QzQ(5!*Sum^O4`Ze-9-iwGi_{70{#*gf$pJ
zl&YCD+%S$ih7d?1{Q4KaB*efE-@l{SO4%2db>&zN+(#n~$r&I_ym|8mD@XEqCfFSz
z+=9KwbfIR~oU5fc;HshN`&DYmXk9U8kf!0-7+681jm){|IT~K@Dn(gyWw!%Dq^Rri
zgb>=mH=&{B1dYyeZKCaKOeAwAjHz6}G}F!fUJ4pGFVja=Z!L?M{v_$O7od|{vY#ba
zMarvx?-Laj0d4G+LC_el24{SpOB#GN5|||M_Vz1pPtRwn)y8y+m?Tc-Qy%Yi@Ipq9
zgL)gdiZf@ea9*;H-HEOr`Me&%X77$DnOG9JiRBNmwXL@n+MHL)JfpR+luTV#>TxDa
zg*X;;5G;71LK-Nty>>0+wX!bmoA2M=veuD#d1e<AL&QN^U5KGFBxTy4y(k$LkFB-&
z0SO?!&x3>(GeQz!j6w|J6S<q#+tXp@{nL?eA3m^;k()HI*6gh0<edMCw*-s8G>i-q
zxxaZu2*S<H9a<}o_T*DKsTB1BH&O|)qICrmZgx|PAE2a22?NtKV7Btxzx^$7-1E==
z^&5|I#m{Od5NAwRv@xOGg5p-30a`0x2e%r4+rUA^{jd)}P@-LYWN({p&Eywfy7k~+
zWD~l?LvICcf6dls|3Vhv<+`?K3C#;P-`xN0CI01q=dGB7ODRwZRzLc=FYEsd>kEMO
zW&Oh}yMO@_Al8Y&CT`-ukOpE(oaQ5XROrqnxMNW()TTWoEkL2e9o6+!ma>u+RtuD1
zkRnwwRVw4OCk(Kk26n?hihfY4Da*Qguu~PSm1SNqHHP4ngkmlVXjYvvZ!Ria1C#)2
zf@xq~V401wy3)Y?-EGs~Iq_=RbGI82PM8W)2<VX6-TszE?x^c)RFfx)^j`&P#g;QF
zf`-6u55qV>%`EfEAO7$?XNC7G9CBtId?KNRz<FJ~tsy0fD)VweY$S2RXpx&Ra{Klz
z!!$CV7v|%chleNTb@6+mg=j)JXY}dJ(;wb*vrF9H42V60&O}*A32vu}V9M|R_TRBc
zWC@X|BS8{o9xT^7qq;Z<C;m&s=yR)`3Dp>OW;k8NY1h{DfgH4Ox<|ZXQUvS1bMo|4
zF6=6{anVheLodo%SaY<}CbwL+YP+U3T}I^MOatEX*xyAza+|jOi@SdD30xIxm0x$r
z6}@6~eMYNcc%iW`l8y{Eao6lP4Yxd>kE}-}btX~_-U2F4H7}MML&?Ei*T8B*+$EC{
zmS~iqFho=`CQ2H143lvGYR~QME#sI_G0t@%EFO%CcPoM#qa^mjE6#Ri(V1+4wK}k<
z38vki6cR?Dn35~x8ps++<2~lb2wF3%4D3>5mjY<UYGgnuab?&&Lfui=W38FV6QzpP
zg=o>iy%2~YFp4n7NRous!r%VkTaKmhe$C!4ydTMCXvv_FU;(L-A+}bpsz_0$n7AEx
z+}_+0hs3*gA2`n^mb$W%$)QkdAZUO!P)~)2AHL^yKk(+&4r*=PV+E^~k{2k2=cjjs
z@s6;2>ytJuVdjaaSw?B?zn8FfV>bB)xl9jK^(1GSTq`Ko(lnBs^%$GbaG&^zr$G*e
zge$hJ$)@W?OG=C@_J+B=d8?bWqO)6}nR2bHx7l(7uRax9Z+|5XrVas;U!$4=fgo(^
zV4vmcwF2>zG}**x19xd+kVG2qcsd`MPfv_8kYs=qIu!{MqiFFz_Oa-nu;n>bYA_cg
zmx7{9lAOVa5rQG@_oWnPP?mz`0#(_kk**`TE`?GGvj|ni2Vn#u41>o4t5Hl?O<2@{
z@Na(kUoh89e*6PN6^04}g+Vem!wwk)TZ}X+dbhG<VO|48odva)m9@^)5QsG*d1M%b
zVQkYx#aWakXCB_Y<HLOBVP3cyB6oLtpFFm%t-CHT%7yhA)Yxg_tJ~KM5e~;AwJL|_
zBe@o8_Gzot!14g=NX&a=O)O7Gb_2Y*{RQP1DLHt8y((qSgaMKT7i}&pB7u~4&1Nbg
zm}^vZHib1mOu$9eLt}XO)KC{%h7bc1oe5UeHLW}@YpuX7DJB@g9npk+b*-Axf#xfU
z#>ZMYPIxYjK-p}$`)zZ!Sh$+Hyd6E-r%ZOGy-VWUt<-hId@zuIhHfX1b=K?WxcWJT
z5OpBQ#H(RP3K1J_IffI@dTj0OKEXQlD(rTqRR{9bhZsrnm}SilVAYsdt$0gf49+Hw
z(E;ZGmC8BKTYGlN1ud04&seDp7KxguVTWm8&Xq}op#ev6qXA2+75eY*Ufna+N*WU}
z8@FlVW*oWQB~tS9eV-DeRSvQ`xMVZI1_Wnw<yt7UP}Oy6#gtHt-8gZxzoAOTr10V4
z38V7vu<|e;IaOu9-*a<!N6nQvuUPWS)Hcv0P+OP8K#Ja~`0()HcI)LAinjLQ0GbV5
zjBw+uxI8cH#(}%LH`wV+4rjD3jYa8(skIO>Z%1zZA8G})RuC4CndOB&_0mjeKkL&<
zLlDMs1R&SW#A=PX=2Ix8Gq5za;*K+?r4<)~X*(-=*|+@+9WeC3&Tx?Wnlbrv69QGQ
z{d5Df?4>1Nvo-s5o4GbxY&M`j^ZsAfKfd(^!1}WO{HqgJ=!7RB+DNpCDI}&eFr+<d
z$~-TQ4uOy54sEK0xjbbOXJ_?Q4XPUJ>cMg*I)D+SAD^UQB2B_>@Rqq4Wec)%$$)K>
zb;`ODga4cda82pO)RWQ{>?kgTu8o0HvazntI0`XxyW63<az5|49Y=1*k-&nfFvyM?
z3&ZXmH5WuCup7*hx3J-KV=?nq2?=P347U^eF;dj0BOXrA%occJ<gA$z1rwjBx8#hV
zj0~tL>pC-r0>Kz8unz-o@81%J$YDNlI?o&qCstj_nz5$4B}>8L%K5{Abhjf65%i4N
zkw7LzV;To)uKf7z@6r8^dVj~n9Z`A!T%5IH+3#NHqn(?ed#@T@*HHRY%9e;~tN5rk
zk>wB@v*k)`(p+ChSbIzkG+@lGKu=n*eqAM^k6*uXj$LbO1D=FF_3R?{{IX^05-Rt7
z;(MmH-}`l18`$=lIocO??A7|%0IYo&7}La%c4So!^VtJd0v5%`z~dG4PWFB==2@F?
zYN-?v)&~5MLK6yBrg3DP2JUWd+3k146sc+~HIuTm_FmgS%`ptz4R=TZi$=Dl*;4&F
z#%Uy4gxYBI=74pzNE+@CT>QMmfiWb;7#ULlE2#Q}$ss6dJW>`gM!NK|zn`L-UD(>Z
z%{B}J1A#%jMW!hH@Zmkjsyri1Q)IUvD6@;^Mk@r1h<ecIqW6kaWekbkFfmO#6nOaX
zo;A;8t<D;1jhtcvMbM>keELArTeuk;fMqml2DN6?GI}TsdtprXZQ@_VTTeW(G`wxW
z*OlO}Tmx9X)C{YhsDv2FrBG|#{6L>_<n0b^UzlrYg%%Xs^|@ddscdK<b=yQ{T|=uQ
zQw>a3>s_laZSDnt{8L-plHQKkhCUs&T~W8=&oSxkw!E=l+$RrsQE5ygW8AT7B`gO<
zA|r`)@B$R=d#hD?8-AyvUfXgGKI1@xxp1~QTVU7P_`sh{7@`AWZGUL3XsMvelm=o*
zjB&?vK9Q}kS`!f$oM9NFCoWpujtOqdlKAS)FWHruhr{;-6H<kU5~Z+f?SDB6VKB;m
z1rm9UWHoA0vSzYoN?k~n7*rT!U>u`Q^h8`kv#fmk?j27#bI{B<CDMLJIToyBM4iDo
zya2FZ)X3fL28op)4+oZY<$Rv0t=&zl`gbi6%LAbfND1WQf&JZ{yVtKd&l%I=dS%7I
zZz1>~9QAWZB#`3NT5PQqZ5oKQIm`=7QJTh_-&XgXA#_$%7vQhW?i;&O8f`f=VR*eo
zERf8ZZL4$*L@uA*$RnLR>3ho=YGF&(`=)JvRbM8>9l$X<<K%j)+eV%aZLfwHC@P!w
zRlph7AM-svuhm=Qe3JET+A&0DsZ@pMJi8VY$)929t@usj&dYyI>c9wBKy(o=3>R}b
ziXph@xr0{%!3T4!W%alK&{ACEa?MCpk|fL$F$NZlT$Q3wRlE(jX)s-wvVQK~+}to4
zZDMU@mqzwOV!w}gjYApZ;QLbwu%fzv?Gcn>9l&xC^-wdxlu$ur$8MTfRLNR+KAtIc
z<>9#Slou8?EG4GfJ%`5wYtGKBmLAWj4^xmt3Tb2iJs*zVUK^9=3g)fS7Qm{Z*_^Sc
zP}YnL1Gl?7PJ>TQ*6L@Em?5TU%|vlo25Brg_ru;*;k~|HvViR@Zpdz9DjHNH#>CKc
zSsj4(TkY-q{nmHK&@^CMq4v^j8rp8*%4kE!o9TAy0G6K-pDH4K{O8Wx?r;04*t-Ay
zMNhWbJY5X$GmNdjtpB9-1;F~U{y~?ik6eE>O}t7w?&QuB-2xT{WdDlC^TN3lWSWo|
zeN;AqAq~_Xhy>SQsZuCfI4#wk>=?a;DX(Z+v7n?d5+pLkfn6HejU%~4N)615GmhqY
zMym4lo44FeJHEMl&1pVzUQQh5%2F!ld_=OkMDvzlIYdEH(j@%mH+M(~-0TF=fn~Vi
zmv3J4%lkJBal$k*NyK7i9A<K80Gib8Fm?rwnTm2`VPay(&XYh%q_iVwBHM|qO2(7c
zjG!T5VPL7NSD4I2a-|faRz^{NasP&OzvuCAVlFEm9v`VyAx4UM%hzgtEjburANi;M
z{4FVfkXg=^IvZ<Qsa4Uku*_$^|L)(DzWSQo>l<PjkaA0sg&`);gST0xr8zPm7mNx>
zrjOM%^U78!iYlb>;%qTvh-quLbd9k#A<BdfjYtClkwOo~o#7D>8CvUDxpIyTT53-J
zx(fTLaMmZeG_mOGcB9J=w9W2}y*X$9IdknIA5deTdgWjCx=sqzr*OTRChpUYSMi=f
zQnTX;xqZdyR9UhTUnPi<yksg4O7|8{pQM?{2DvJWR<a1$T(JL~Gm?DBL5M!Cei#N)
zN)AL<ZOqGz8qCiJrWpBe{>8s0WW|n|*XfRWH#tjDE9dnHRY`i`3<WAeCM9LJhu_@X
zkgM{1JYjic2@|hx_k8oZP1uyg-4qC{#PA-qg*jI=-Jvp2$YfULdPKxHapqQTxQnk?
zLPRC78*hkA+`at@ooC)34rmZ+h&}?m!kT?t`H-YB9tx%vsm2%wQXKg9w|}7cGy}3#
zW)rHlDWzS`Hw$Rs_AW9^m7I^%#Vf(39wDr}dX+ezXMX?P@7UijydLh+GNLsi;!~TH
z#Ti&!DzBGK;lH|`m4nUNCXwQ-jiv+Qqh+PFXLUs60zAnUe22#5=+$A@sjHeCL~bA2
zCnNT@VCmpD)*{}Y325c=&%GsagP=P*0i)E)X+C>WFGf-&^|q$}15E25d|hxqjDQlk
zA4l$o9k0`CMoGjXXxwAB_pB$OWC%kbq(EKsr5(Gm<h*kK(w0{>W{h*KoJu97LQ3%Z
z^;>RV?YO(YZ*1v6*2?jGW?2@NdFI0pKX8)<{;&V~Uwb=QHS&HU=Ucx0@QyXl&c0P#
zf37+QAO#~u<?d!+szS17Ofqj??Fd<jIWp}aq)gBwQ4?>cuQ+Jso&A<o@2T}`=CW`u
zXI9HBES&3^WRc9oSLqG+x34Ht$y6qZ#9`v~H(yiFC&J@92dt-&xJqMi7pxfTx)LRV
z6-xH^efaQ!r_%=>PG>4GOgn0>oX!i1GeL8iIj;wT9DTB3eJ0N<$A`-LTv<*PWM&u(
zZ|)=K^~}HhH~$m2cfaJDul^M@P|&1+6u*Wx9Kq2WKMCo2l064DZLNaN-q?W4#wd%{
z1~EjR>}LX@2Cq-C0)ZTogVF(lHkMb2godAE-G(nd)vRd@Z8rQz_H(0Wc$-!C?biU8
zzk!*NOX2D8OpFi%3}z%XM705BpXX{PA+L6B$FI_!BmuHwX(Zg<^LQwnS7qFfNQ~6d
z?lnG9IRGYYUSaDv=q>E$JhRk7HNhgt5WW371mc(|3QNvTsxL0dAqKy=f6EjSZ})G|
zyt2;EoJ(OTg<O|y+NfW>O9YHVWE>M?O~lD3bXGD$NX+BN&2C_~_dUC;*`+~5sm;C~
zk3TRZ*pD|XWoF4Ui_H`}bIK>8C2qK-Fkswz!N&@_n>|S;ZpIyDp4ppAf_!*Bl5=)2
zI9Do=(n3lLb6Hr<Czf3K;r%-f=L3kdtt+uH8*O7V4q;*pd$049is+24a6Sy=`AE$t
zs%DfzOqEhk{O<Svfmi$A@YSo=?W~Q6RWNS>HnA&zEc78e;sSPv1_Wmt#^4en4s5$7
zZm?ENk{(w#z)Azb43oG`v&3pE?)koxkhbEQo6qI{RsHud$O{1LkFKV{(dx$+z^b8R
zt_|$Wy7Mv<MLni8+kXA5zyGrSQLQfk)|d4UwhT7b$`~V4^chsayml5PFb)G@S^Wh~
zz_^du?ck1>ZlHf>xRh$tpxgM{su!ndo8YwT#>5z@8bF0oJ=iI^kPP;_J-cCK6oHgT
z>_7{H6xMulu~OQB-@cX5sP##>9Yzpgh{l+NAtZJ~atDhG+{Oe+AW8)aRTrwpHXqM5
zd0gVPu;L7&D9NkJ2xFX_RV2}W)@l=_Zi3c$3lqi_Bf*nNPzQlbSq_h!*O~b|g9Jhx
zJeg71z!hh~2zRgcpoR6EDe5BmrA>lUi<G)>o}Wp}4%RbCD^Wot5hQuSYQ1LTdL^F3
zI-})631}g#BU$cTV{0`tYoOs(jYb-X1udbK#^5loGGp7GyNpC{Uhq0d@7ebsTUSZn
zg}Pu@FZ-#3HZw{sPT-(Ta8bKzq4>;}^e5%tyW4Rhc+068DFk+DVD?CMtI}#KwM(m%
z374R&wux5S_izJ1sw%-=j&k-v1)UAo>>USimSrIh12@xzT4t#;LlkylKtxz;VOcMM
z(e*DdLUjF=af+lW%(IWQAA%2nn5NJ)H2U8I;`Y0i0m6ty*~X1$vbVrWWne%<C2yaV
z2qBV&iCC)NBcPZOtpP-;m?#6i(pZ~@Me%!N#_4#XSY;eW3azcV?RBJUjQMxGxfw}8
zD0QJ0rDkV37=;)UB9&!65~dStSrP4muz01Nifh=}i#vDyeEM^tU2_j6r6WaVmsfS2
z8|ko9V*%h5YR95&dta<q=$n>SA2;4V&0KE_xm>{oPc~k#Hn^}hI(YuFeth|6YDUhP
z7#&P-(Q2Qv`PsHCbiA+~VZ%>s83u{$(}+r?T04(}12Sr~NsziRAAuLPT)EhB)#}L*
zkcNR_oEXNTwQf19CTreUSf<umfqei90^{J*`W|C|+MCF?$_HKSJp^GGqWfhjM5#>4
z+i#*cbAzCmt)Rvb?ub&!`b5!0(R;FHR$H4Vt}N!QR0Bo{1){gKD@thM^Kl%Abs+j!
zVV|gKmz!;!QPsrMSIsChr}LTP>BPFa{@kz=thGrRO`FPGM+eDBfMF1VRPwrzbEW2r
zk&#eI$&j3RdVWtxdsNm9o|V?V-(BElfEb(bV;6zmHcN+u<cbm9EY_s4aAYgqgx2zF
zQX0sa9ItNP`=uUTJ0t8;(%EcUu59)b`R~Q{+dhy>bYPu!b+zF=wosOw87+_|XL1KK
z!snRy_Op%489c_wE+!H}GJjs91omm*Ag%39r1|77eBax*T5%D79Mz1@crCSTfJATa
z+lCcI7yYlQ6f0!)AvRW({j_5j2gar!H;yAoyp{f0*A~;eeQdiz{G5qx>Z=J9&D0<a
zF_K~+B|j5-3mN#3fi=TYP7r1WGQ}E$(wv!`t&pgUGEyl%wGg+5I1XM@DTc&IiixVm
zdC5L3p`A%I8?LpmmdtTpIL-^FdG@_Ej*a;bUCQF(D}4fUf3GMCR(xnwarq8wBNHVU
zR+M=@GRAq!YuqOM?CpS$ZTIi0E1&~W8@SuqZa%P~4;H$(MR|$)!gdw_tvGB=R)hB1
zD~_~XwJ>&b(sqsV2?PFUT{E#>Z19g_WfL}^OdD#oy>9EO__F@7uP*@Bm-P?2ghDW7
zpCbDd$@2pxMsz=+gZW7K?5u%Z6Dw2%6={QFJvlRVos`UpwN?sV8OmAQDd0h|89tdT
z3eki?gq;LxOS0BnsI{=1PV7_S^)G+T;}75Szx?<A9e@2-zu{NE`USsub&Db_>xoGc
zNj<nHn@<Bvfnge`(b!Epl*+k2AxW8r0hPk6Cl04dDT!~scF-mw%<Fr;dwf9RFR*aS
z!9FnaL}??7=jQuD3d~vvCGqz4H-r#5oR2KaiF|CMhHK@JSLR@h(eK%LnK4OByWpeD
zb>U%t;xwQ6-FM%SwXoaok(iJ)k`@>iBif2aD6D8yZeG7-KAm}fI5Dr8B`+M;C-Rah
zB<8TNO5x$d4?I6T@h^UT;x}LYHPyVe!{iAIM_gdNJM-prT9R!cdLX6ME49)ANGMvs
zAhbZhTFZi`O@o69@dy~?2|^R%ZA}Qyrf?>SAc(s2w-&rq%r%r`%dvlE^&}u|*Y?vr
ze))}R<?wuBKMuUwkHmKMQT3`3_A^*Ni9IqxWs1TSVLg2yhlmanDuzr0yQLBoQqxgT
zt=n`*vle8jbA;x#kZVDL-`84QxI0Qf@K&Pv+^3kmcHk!kSPC`!v_f*_!@GAJCGa<r
zn2$3bzW=~)e(?)_`PEzQ(jH63VLg#zcYQ6`2|`GTIF4@L!3Rz3#ueQ`y}e}`1%d;(
zwMK>*2*F2+AJ>_8%Ok-SB;9h-12bo47HF02Q34ACB}37`ZvQK9U;6P!Dv{l=V_s$s
zhiCS2#}EegyVwGCKd*|eJe{6-Qsvu+cRZh;Ip!5ah*Q8aa01a0A-0wPW3h#>d(D6S
zfBDxuKYrkE|K>lSi=wNtmW4G}&dbrq%!fiQXPyu5*p06lha2LUNtOw7A!rwM^>ca)
zn8md?A`XyOUFmr(xK?2oP!D>g<<Y?j@q$JPvp<uq6R3?&?`;SsE~t(+`UWjY?s=8}
zyovoJjQvS#K}*Cwre4Sesr7$W<@t2vb{e_8-EDT9nkL_8wLZ4xkP-2rDx(<7@g13n
z$v{moO%pa}hFYPH2tp}~pWw;cvn#^*30sy{mJx&`32*N<mRgxqNkeiO!BScC%DQGw
zrxP`={8#_-U%|Ta|M?&PCx$d~yT9j~SNFWWz2)uxEw@u)H%tu40ZTO(JXe4;u)Dp*
zP=-lZOJx*q7aZb<h_aTMoHJqxhRDp3ck_2FOJ-h$gPxh|!fY$fhp4C&&M|XlWL`%W
z1%mME%|Af`{O>=!<D3hxcduAZ3;*HaTW;fyF(mf;K(x^8ICEGI98V{Xr@)7&?^yDQ
z-Rsv#2s|E-Sc!~v=dH-WSfdeAVbv3Hdd<K3Z~i6c<C!16eaBKfkzVagmX)P0EM;b`
zM^4Kl4^Ka0+z`S@OiI*BDKlmTh}Z78?Ge}Z3?UH&2gJNIh8DjUo`{QU4%$mO$izq@
zavtBKV#IayqJ<W`9oeVh8R860JUQ;VliSu5KZ{kbM4dlMgttQF_20j+$%`%bg2(Z6
zMD`PR(@oRRG=}<_+44trY^^PHHxBGlBA*W!kw_wnFv-BHSV+kjED>rzOQETm1J2Qj
zjg50I-rb-$D4lce9Cx}f`ZP^U`-wONOe;%Xy?{XE>G2WEng8lv{FmsO`5*ot|2=Q+
zUhyyg$v@#$8hJfATQ6H>2$3X#Ymtr^VH`sPf{e2&Lon{9iCszv4!V!Sz;QXV<eBe&
zc;xv6XDyUEQOcf!K5?=WGYdsBE0q+C5YL?B%*sNTBD;9Q&6^!r3m?9F$GonLjO1nE
zSWm7VCc@1yAS6tc^RlqK|B;9D!sB^iEi-0y(`C)sC{=&~Sek^R%LA<InK4CP-A_=J
zd44391+~mD2~`F{bOzLrCQRVGJX16`@pwf<iS$<XZ6|n3taO_Y22%U`+NL8nv3#uv
z-uf8ZhH^Xhsz&$Hr8e8!bkjnsG_x_bTvkA@e5q_xME^c()2RI;D~N-yhx3X3ZeSR9
zO;6YGzUjZ;XMS0q=lTL*eOdo-tCffZHbG-+(`c5mZk3!EV+)vCaiOyapjGcLRqtZf
z+6rR}PD-l|+BW>6)<P~uUL&<kUM*|~e$`4Tg=JYF7oMM<m`^9ril`E*ky>W78%L&?
zT>KKRLV7jPTcKibO1c_EQ&NB_MPdL4OT4O@b7sw%oYmW;bVX!`nc;JvTEHyMDl4Iq
zV_~sMEEN?ONzPiy>aBKBB2yYE28-541r`ViYZUOc7RU>Sb>_Iv6l>ZVt&C-*P}M-4
z86}Mgrdp|4(NYnsOyj^1636EgmCSiLab9M!tz=UQ!L@WGyHkl72@9f`t)^?e8g*S=
zb!QtwBuFKQ0~kSEs{(C_smb=UgkIV1`-Cg@Maw>0Yy4I@w9fR9jRt+`uJ=v4T@IfJ
zm_NC#Klal8w`MMsIceMq+uEvqpN*6KBLPSZffyoId^Gc#7bLk3Tu6~r&8zV$TY}Lg
ziM$%#TC`N#fCOoO&aFDEbUg>v+BA1Auk9?2QYw~R;8_fwpPtaA63r)MQH?l&1YtJ}
zOhcrpfzyS7<;s4OIa@A@2g$pk3AgHH+Rc(zmUUrXGi!ltg4oGh`CLYwLdC=g0kTxG
ztauI9wfaP0Ri!wcrFEJbE0t<6s1P(XW~ouCQp!pxfm2>tYJoPLv8uI6ecop056<sb
zNJ1I{DS1$ARIkLw1d>ra`B=(IHrLds3RMeLSCouO@q5;6WvkBGrTXu7<4@P8MH)e@
zYr6W++Erb(nVjJXSicP|CFO-C#>U!UV`FW})QjzU^>df2^3VI&=-P$<@sQcyrPjJJ
z5VQsOpWXVzG1VCsA^4yHQ`fr5ssu@dVI)Sy627k*5ur&b^;3J(c_~%hw}^{*XU&(K
zrrB&Rm0atFZW@Nx>McNdMRR5xMt?Td2vL}Z$r-p+nZ|)6i7~l<s3rUJF`tY%MA^s^
z!Hi@u#ze44vOv|!x*Bi;A{@^L4(B7MbtRWG#g42xvzQl>3fiDJSj1Q<tTv<0bHSPp
zP%&pfM)Rq5rB-Gu)BvLnNE7a=)mTdAl!e0rt5u)gY3{2^Ro9=}7+gLzRcmgblM>Se
zvbwg9IzvudyWg5Kxl|+!-r|!2wbI1Uy^T<tKIO$2{Xcz*u86f3{WhTxAD$B0GieTW
z+e+5}l8K_BF+YIb(sYeZN%O<o%k*-NTz;tA@uxN~^Z_fI-?z>)`}v=THUUxl&pQC{
z_sp~Z^R7>Y8~FF`>|}2NRFB0Pn`sOYi>@at)>+XPJGKrwnETXhwdq1Cmq~Z-Gpl(_
z+lwoh@2^Uwn78ujx^g@`!+K#<qDD*}JM4ym)PPVIAaAj|c-vWWwRDdKtPPt8(nQpq
zwH#CA;rYaIK5;xR%&SnD5L*a(BAdt51@&JIpefRL(oz`GNJy!57Zh@>s2WM4YwI>$
zFm17@Hlcefm7HhJ*{3gi7#iE4b8A|=PD1jywN1#B+U(e+5mYOsuI_KG_?awiTBbzC
z@s=Tt?Ll?GRyVLlTDxs~IBci-<~&0Kpu6NoXheXYZmkRa)DQQzCpLe)&8%#Wy0`!O
z>PoXgTKoT>Xjp&T1Hb$kYkSbyx}N)anr1Iw@MZm@US9yLFY6y<T@Ep?+Q;Zqsi-5X
zX5K$NvcK`^Vre%bF+uTQ*I5~!pt*xEs<{ToqM2%n#2Zu>&h@cPiIWyH31>Z$D?B{R
zOuGn!KvuLWC1=)k;ql=S*3A5eA9y_t{Piz?$^Ez|E=JS^V&Lo7uNh(_3|$AJv<e}l
z=uUQM4oItymrl0>yCejw4C6#diT6*B%=60Q^UTwH;E6|{g2*dQ_QWbD7L*#494k5$
zbj+?bRgJIffhYs-j|Y}q(3-f}?YX%b_`~6m)3RVyFcNR~HxS`mE9d!)W#NbC2TsfE
zj1zGxamj^wJ!4Bk%_p;&7-pH0XP&-$24)6X`0ATq^6K_0jz{Hre#f`(e$PA~IoTsR
z+%xRnAfupC3A;=RGcr99@`<Q|T4(@H*c>V%XdiDJywVlLyrm_MoR%3?7)h9TYf7JZ
zq+8&V{JF%8SN%tjv;kPvycN|kuReR|dXqMPZI>3PkHM?UDlcAr3B3CR9pCF}#*z!;
z5IukIdG`gn`#jc1U)C&QB+@uC3?rfgYsoyE9@*X8ayz_Ym<D1B=qgwR6Mt5UYt=*(
zvbiAmQVWHGrNmM*$MS^Gv{hQ-rIJw%-p)j2QExS(RF-+BtSeP3x)#3s-5+?fzvrL6
z{WbexBore`B_-jTSGT0-g3`U+E<|GxV~k2rWv9ibcZNt96C*+t$a&ckgm;f0cs!hV
z|KZH@tSsWw!fKs3*)t1g=B5=PDUf1j2nY6zhz!iSASH5GR_3)(vw%jD4IIl#!5EXb
zCKdGrzbwXaamLqiIg?A_c6WmWVJ%9nLP1cCpaGbXAl%-M>~9B3R%%s3RCc>treVj^
zLD8l1d_HoR&%Cb-cc$za$)%urV5ujHoiSFNQ5NV^hkD{0qy^E9<><xQh)`VjCL0v-
z=@)(4lw7t}2}%${gKg)lvo@mZYIP%12+_~@+7cUS%q~Amn(!=F;?ynL?)H4$cIbQB
z`q%i$cJBaNDN3mgXr)1CpIhP!FTR{70s$C^$!(cAQA+0N{K%erZpS;)1TqwCiKrTR
zL;f1LVX^t-!mOEM1sfAfT{&d8Wg#{2OL#h;kX^V)FI8pDg>|kRjt49k!Y^P<18?tM
zbANl!U;XN@S<lbp`N;h@uZa5r!&|@R(=)XRa!<+F4Mse=a5J;Z6ug>ly^v9i<vg=!
z;n&~1CC0>m_}hQWyW^Q3pU(splEGOQ4t3?kxhdNDrOpC7R&MQyDZFA#iM28DRGfV%
zs{>UQh3A^FS{PT@kr>NFHsho-PcuAcapXJ>lx%>v=vy@xb2cSd#Fp7rZ0ds5LS7T=
z3d^!ms)JX#!cvUK=anTJY5a=a?OSeN-*bFc^082Kq3D@XPM&PW>9Q@f4T@;=;E*n&
z>q8XA2(^G_-RwnLVXgy5hOrJa(UhPAVnUGymbE=&c$*<e6XLd>3;U;CFlguRSqD0Q
znsHyi?(#>$E3Uw@TA@~-Zj2e)&yT;KCH?1J{mX9mO8m?ZZ4zV-)LJRa%rNd4rad=l
zVvLESwt`1%*ES37Ec4f(#(7=Iy0C=c9S#&s0wR%O*)_wOKUeiB-jqtNO0CBEcqE@^
z{<r`2|I0p(eDnHi?xr1<6&ZvSg}3*6EWi*zO#KYO1>&a^NE%p6rc@>Q*l%_zA~?`3
zgOK)#zxg-c^W%pHo}O2#4U{mU9MB499y!U$fn<y-U;$PtSu&NCn{m%Lj=Xz%$8nxn
z){I!-CQU4wS+lj)e&e>fOu$+LOD#N{Ph=A^!fqH{`>Oc9DrUadir}|3wCU&zh|7%-
z1<gv%Gi!G6Y%NMIO0EII4R7!N8WEQc5ygijNWo9y?z`&M-e_ZkT9*Q?1A(Sx8HtES
ztp2pEo<0OY1Zx0P*rYQ2eI;B9I1Y?Cn5hkT6A6SA9Z>4+shi*VboI8M0qvvzD5B+f
zqOzzm3<1-r0p>J$`7i7L4C@Pk^=19TtgDmBEO5>Xg9T!XlqM2`Mq-Ixy%fWOCx4lK
zxK)s_u{6|NF{)T)F|UeNsWe!mF*=%q#;9Z;8y#Ic%4&g}Va-@6kPAX$cempgU;i2{
zmE*FKVql2E%@BzpuvTxMvNlykEs{cVr&5hjA(WAz9+x`%z@1-E<?;Ez@ig<}(~0MK
zcKse%FbR}!rszzzHeu0Vu*zT=F{fA0^GpbZ8W!Rdczb&bxTvAE0JNHqtgczT%2JdP
zR&p@rypl^{CkZjH2K$6Xpx2qrKm%HYyeJQc1Jh`XyOCj>7^lfaUCmir#ROv@$}6V$
zYf`*N;s~;02&E@ATObP3rl*KHOT}W-<w&0X27D@`Rc|?>DjJ&LddnOIsG{xt+LP|g
z0P7k+zCr_+46ZbH!d2ATM*(*|m#b>=R_XP>dl5#rD)Wo(B)fpN&DUbpwm^T-)Y>6G
zgY|QPguywl3`7P=t~(RN?OG$zz=+f!s2I66riJ+c25XviidQzP32Ut^R;j@S?_FBf
zFCjL2kK(!;gr<mAD_V;SoQsRVP9bo&+w=7|Zzv_R)|FeE5CUUzJ1*HrZugcJL73t|
zQjn^yKUITkmWj7Dt_CKaRQ}=Jd!9}+r(9U7QiD-h{7jOy1^otgsbB^wAv&OSI6b><
z)CCdab~kZeJ&2cDoRL{GQjMujV1ky)DGTS~RdKo&g$mXHf46}|uTqLR3$AA6;r$cq
zx-g`HukXI*&F&li@a{y(&&<n0&dQK(nZ{S#+`I-dSi#n&nbEVRmTj7J{UcW>6ax*;
zOf~6;UuczRZD5mY`dm*F)c*P_L$b!)3Qb(r?b=$WK(obGaScsN=gMM#x>-Mt)A4&_
z^r~??Q?5a*?d#fhKdFWVKWD}|Y{~U!v3`o#BRETGt}7>r3?U&=h{<&$g9(F05<+PO
z1GNE9f1M4~wT}9DjBLhQGK+d!sQ^(NC@{n|+-e~Np;TpEGo`rDd8rkv1zig%4*cd9
zzaoanvYs)NI7CwPmZNF~%qMpFMBvcAk6*iqxA3SMrF04)<fbF{?r<c;z#l$5@ib?$
zLbM7xqb!(s%%DEZi>|Tg%poH-#`y8@Otng0PlyV4yBl(;9Lj>3`{jzWHZoAunAgfu
zjf^wI0;%n_&X6!LpGaorJ}@yL>@X&(x|s9L&BXrpo`D<A|Mq|5u$;&ls!rVO{>rC`
z+8$vgM85}S=D|s0V|A2PEQ2XQ(}h`spIj;di!P>5XUA^42xp6lw;*+fyn02b0nNf-
zJCxAuttG?Y1uUEm7MgFj>#=B?J8oR8Ym@s)vitFCE?~8_=^<SvCsdC+RF$Md1Jyp0
zb@jcieNRJKYi5;%1!pkU1|Rznn@BVVXvGYRk?nNbw2rLVvX|*rb0%>K9)F6sHdavo
zS-H|Z5;d$UIhTvCC`yXLFW$aEfOA>f1oFrjLmQNrsZ<}1Lj+0Qz80bbX9k+xjzoQL
ztX4TMEAc$@d^+)bK2t-Y7|gm58?;hw#X>>5#%9A(B>MH&S}8RnRXCkb%*(=*ym&T*
zL{w*!TXC5ZV!}dXDaulvJ>94AS&Q|pv-?!HXCuu&`sC8cd^(e)FfS9)B6l~hC~Nk!
zaPi{wQeYSoH#e^-#W*i3W(C1@Z?$XOwsWm(KnYz9w@(ecxWCQ_L!B8fo51=7guW>J
zHXYn%$7a2#dsR(pcDxy7^VQnuq#{a`E+f(Q>6N{{#Lw)dJzv@V1*m%&TVpu8Pu!0O
zOP`ld{<8j)*B1cm%lZdd22JleU?Ff=&QO&7ZXyKXE(|0c7}lYwst6XfwQ3ZqdHV){
zicqZ4f2~lgGFv5EriRiIMsfOcPv`)Z;s8iYn5{0DxUR6Q*qV`22t;nDJO1jg{srHD
z_XmFX@ds|F#56|kN2jVT%S@0+ij`^sEHTEB!C;xs4!Ep4qE!Yd?%<0us3BJQ@%;mj
z$1~r)KXF!X6%64FQYi62mYJorDno@3m8q_6>IfXq2ZBUuJhPkj{Oav%o}Z6AJT2bh
zq?If_?We58>EsxvFmvL}vaD#SO-Deet5RE#ZsLN@{@D>-VVNu6KRoj0&BU+X?nwJR
z!?YtwMQm-Ny#bXSV|dG}_|F*Q4Ga@@Hq>Uat`zmE)&>Dt6XaJ9Ts1`prUHmX)S95Y
zK`fD=f!s%XZ*8DP6;GvTdklXIlKKP~m8*8tUO0+4^TuKmBJL1a7lb#qDEG=2Zi(#n
zH{I2@%=Z(bSOy-{WUq)pV5lSUa{;hEUL64h4$BM*_tTyL>|=63t3-lD#Ecp*VzAX}
z(=Ad|3KfS;YY0>oPC1jMP<;YGvttjDl2s<Fgcyi1Kx~sB^NQB&I&dPy7`bEO?cFPW
z^H;y*`yapO_~9AVL?EymTn}Pd7J>wCSq&pXVxK0=BJ*;_YQ;(bs{|^g7IN`3DO=&+
z{r+1Hb0)K6QKJN<%1mWtmBqhD4_;f;n5e`g$PoDO_<^cQu18Xwc)h<tPRddi2VYe=
zmj$J=&o?MAYvojxvzm_r?=7_oTD+~oLTxP$&IUmPQiZZqzWc*FQYuWtj$i-kpYUJ&
z>R<9d{;&U@Wyze*Giw!g!`IyIzUKb!8;TyO_003>(Vt;&CDCqwK1RHY`WphlHL@&t
zDIu6qOAB_jD_>po<_Zo%(^yiWo1Ca?XlZKUe-r3#Z6=C&b+>`hKE>NM2JXuZ3v8d+
zm}2IMn)T~$Khyua!C~zgw%8jf>N6-{E4X~Nt6_iv<G3z}3HQ?tB7vzUQV2v90tqo#
zgEJVa(HL?CaJ#D_-Z~`4f^t|3N^Y&WMgrERmsLxl*6f1t#aQM{UJJPta$_vjb7r{R
z@lXHhKjU~j@a+%Zvl}DRZeR?NMD*#STB*gE8zu?C!Sk+zsRc?g)^kA;#1U3?&8Xi$
zK2X&7_us!~sbITTL@TJBsIp=qQ-inDN@KYM9h@5)&D(Ng<l*5xYgt*&Pwe6?Z|+_*
zFBO@6;;9$g9Nb%gRgFVd=32<8w>r0JfC65`vqG>;)qvGNvVp-SMDBnaszOSUH}^OE
z`t?8K?d@Oj+rRlOKb(JHSp>ojuV4Qw#$lw^0TCm`?C^vc4DI~1Os57rqoWWtfJAEO
z%!-IgL_=(r*C&9|up%SWiUzmCY7Lqa?VZsK8W`#gMQQ`9T#}$+XRZCyHLsAKD7UGI
zeY0NKb+_e9s{J3?=(I!^Hd_wW0hcPqREE#SCvOaup2&NUfl*oV!c0JSH+Y%Q=L;&G
zO{a`NMX70UAVjYi<y%|@S>SA$D&A`6Ev4>{{lRtyo1drAU@g@r_E{VHuo}CN_^V%j
z!{L19_dkB`k_usDOy0q;E(@wqYDB_7jKU!Jz@=6^)U|9=^9n}R%Jb<=G2#95fyd*K
z>Fy21;FQlKEfmdU&K~=>GhEf<$Cv=|p$cosR56|p&n!7}JM9T10*TQEQjJZnLy;jt
zN}O_In`vu7ZCZ*dY#`QFm<fth2j3C{iGkzuBUU2E(;fG_9dGVmv#vAed0~Bg<}|OI
z7YK3S)!kPd&nF&_2e6EYPb2o6s8?NjysE7uq57FZ*!GJx@G2LB>aETm;&wR=H&$y}
zSKfu8o92T__uoyM)b$zK&$ME26T{Qmo34$J8q;f&r@DY{Kexi>s|ERpF@5pyZOBY3
zDxn~?E8l)u|Jc_T0PD;8^R167-UdE3s&dLRsPfIE#3uO}NFK<w34cCO3vY2SbMT`H
zM26N&FC<Kq)y|Faz$V^XTjJZHiK2L$fwrn|Ddg-^k5Y^T6GjHgy7I&K-}B+a2abm$
ze0|R_joWvyrX5ubgp635k`!F`qL#|CSgYJBRuokil2weq`|(FkxuDYxIcKjTk-ast
zRk3TYrnHmV$4|Q-d7RHkjEuubFnD_Zp7WB)d2LgXJg^O8M52qRpYzI$v5IRlnOPH6
z)+^nJt|bAzx>6KMbNEk(XP%G0<G0~k#<1hJfA>2cpPo6FqqAmYVA${Z`kSw?u<-nR
zq%J2)p4$j+oX`<238OQhTtqgux{=YQI~64i4tRF4<sP?nCRfjbueVyHPYGjd^)b)V
z5N-YX?K`Sov^gZQ=~VgffQz%dY~ri((2L6N)!%($6;tNoEo?Cfx=sE04A#eXOrIuY
zJg-M;&HO6eGgx4f(dp?;-bM(NZAzwig7%Re_ZBckSk0BAvb5wAB}3D>s}!7CAF45!
z|9!#tbS?!{5ff5~oKGkI=0E%mYjxph2|m%NO>GBr)?%$jih&RWF}H^`7uHf)vy0ru
z;JPxU7T!NSA`&>PnRBg_1S>v4O{2HMI;(6e9&SmNm{CgsTbRvRNug<P1)R+k@be@E
zVcPE*QgYB<l{r@m*J>m})y90<7=o^QQq5ClM5qdf!<ql(hu?8`7x~54JN}2q{}=x=
z|2zNwxBs5U!z0Bi#2v%*mT|gew-b(M5AI7Y)KXvUfz3LkN4L#>KH4jlUcmK<s+SV#
zRWka~cJtKLwN65mrVn%tD0|yN|2%DCBZi+dBF>QEkJ@p+etUV*e`N1hHI}MArB6b`
z@&04U^Iy>VlWkep#gRzxmLHKOS9QhKFdxtE!#4s!14*6LYCxu-;-H4h&{cYKH>G&H
zOxCTfb;;SaC|lxM)|E9b)OF?Q@rl3r+u!on|KwM^y8oJC^htttu4r|gM(ZHC;QA}=
zvE))*BhNhT3L&sAnZxnOskj_@-0esPvk0aY(@M3BN@E9iKP<v9IDqa6mQUk-JUx@k
z${1mYhSrq<<1S6kenuE?_QVvKOD3yPn`pQoh>u0|>B00Cc_4`E<=|R$MOQ3TmU`xQ
zfA?GVqj57Szx%`A@weyi_}zEk@&56^oGasW$Np|l)xyKW2ZoS|lF_<SOTO&;&6nZ;
z1?~uKgQ@)Y!4%crk`$Zn4ruV3qjq6%^rdY&7)?8_N|Wg}JSpuOOkA{bR4;ZOI-1@0
z?I%sG|A2k;Lsww*iz1Z&T~%$i+(e<}C+6{&y*|R|JO1#zoY1PgPOnHBTr;X87qmF1
zu?ejs@tu7op@9X4MelZ4iWiI3j3`DB-KIpB>RMunfGOm4b#|p{^ZCXY-Oqn`|H1cu
zcI~Ga{j*FP+gZJ})3wnewSi-$RB~3<wV(n+8k+W0<?->E<8fxU+w(d^N*w(?YV{b>
zyr85~w!FYPSkrt*HRW(Vp)<^l35I1au2$HG$Tzq5oQH*?6^nt4u@;vZ(4t6eAW;L9
z)GnPLZbuD16Lt2u1RkFrS=UHj-?Dpi&p-d?|CC(6=ke)1r{%=+>6tjbB@B`C`pjAu
z2Yo{UGDruAK5nbMp48jCld;A8f-|BzSPtDjLeuiQY%GsoWV4}u#`<C5dV|tHU!NxF
zI(gRH?E4YZr}qSZzAm$IZL^=UU)uG5<iC52(f*@mlwzdlx@j-t@-OQj^ZEi{eOZ6L
zbsaQZFay9?bR`(A4F;09z-rL&B3AMKd`A_;yxm951t^0FKJg@wtu#T#{yOc%*7m19
z-rF@esw$-vO7)h75Q2~9B~eS|@!^5f@x-z&NC?DXXjP}?lFT(Kf)oNC%(jTur+bv@
zRg6Fzn{DJ;DXP4Ger7Saefx@{-gGor+3rVggVU=6J%m1*+BoI~sgbYl5)|chIIyZw
zYfJK^X;-A=?UJ=PSSQ}TP#e8jw|=_ou+v(5UaK8-VQ3Lf^UU+pci0-xES!!DOI}&)
znSx-lBaI_>_qQBQE2r~9C@WPn5{w|da>r#FU2~jB8zKT*vZ}RGVsas9x<-R+?(Dh+
z>Dp;^LUibXa=V{E6M6JtSzFTI1HaH;E&L3{S+<Lp>-*I`>a{xcWvhFTq{dn+!4j!c
ztmP95ntzGw6Lu`(3COuDpi0(4Vq^@-K}qcbz%A&ppW3mRvvWL$@x&-wp@{m#QS&z8
zVx9m8lxj*TN~+#=qvq`?rB-UKMDynpV_;oYzW?rf#{I~+8=Wd2WV@bCWK|2&=j}AR
zC{-IMri9=-t<=JjSDueYkib%u)tmvCQF7;#H7zK+x_8$>m?zu9skI4?p@Dr?>CKEm
zyag&vBPk^mA6SvAQsw%6du!fH4H};USo%-+DDZjZ+u#2kU%lD0-+j#=-oNMgu<~#`
zGv`c2K>{(|k%ozs48(=8YgKB|?R&h~6m3e>r3!pW7#;9uON2Lp^{apH-^2gqV#po9
z+U}q!7ZF?ky@+h>8rQJePp*+q*>V5V#|^{nimsBJ&5q6Ocrw=yv|t|(F#OA2f2u7@
zVu}e&uu>tlX_Kv8=!Gq}T^lrbOA20aD0sofT0&FJDBUh<r4-*&S{=YD#is~@3(D8x
z6Gm$(oKGiy{P7*X_{Ce&FgSn|+SyXOAh)7c9Ym%-w+2#ora}`KD7kVz&1k7;41{qc
znqVp|S@(FtOt)mWW3{mjY>@_FRVzHtXL4P6HAG@C&`c8HCXVFjieYIQ9Xy-Ql<u#a
zPwRlLwzFw#&J!GMprTs*><CJ!D<9rJQSSzBrakXI{J{Hf&%A$l;CMchD<r8*yOHBL
zb2vQn>Nb!jYXDujINS~Bk?qXxf8D3WdfO<}Y6z}V>3x;9O&`2I@3uYGTm4(7Q*Gdf
zfA5REI)EkHdC@nhg`d3Tc3*7!!0VR(kIpB#`tuf8UJ9IU%fZ`C+W?QxVSS9hU*Ds1
zSwX!`HW*Ce%qvrEMLKCV547Si=Ag%=sODgX0$CKLfmUrdXm1x+MXPORL#d?=Q_$7|
z3n}<~yW`=&+E`5BHtqtKG$-NySWKyuZ8~JF-XdJ8+v3<Be#w<uXG9jpaYDw4Lp4HC
zq_no;TD=9a`{jPq8i>{Z?!5Z&lB^Y@BId(jf>w4zBuoP#3TrJ`8<<km&n;`LFmG*d
z5*Ph{w*7ANk9`2i@pM8pkXM%}`Rc1T?0)xz2rTQ&c{vl)F9<2J=95nh6c<&H;0#8#
z!p`M;c|3c~K)STQJKN5%+HJmdAZ(7{GWKNiiS0lTXCPdxOuD9Aca&{5asQcSXBV?;
zvGogkeHm+CAyb!P*tXd}^2zPL+KK_ONe%e3{nV-PW&LAYUjVEx>(947cCzML8xg~Y
zT^G)pi6L?;dql#EB&@b%QndHkD`qZY7}Tk+LrmT}$m%n{+LSO6qo;2zJ7lTGT8y$9
z>*`at@|w|-8PdQmMc(f286+^DPVA<M{p~H&{>DY+t!qQM<2HpTPQ6~WLakKAudy?J
z0+i}i;_tryo~+8v?v`N2Dd#3$?a6DTe|c#<tbYHTkuj!$VDRp|Qh*VGVwAHIB{7D9
z+hJr36R+O9+N!;4&dj+`VuTPk5GQL>d=jGuu;7*8(C(M@ao6Hi_B8Fdee;Ipu(F<3
zp7RsO`9!t_lZkP<C#FEH?^u=t%jv|M*A5oFd-e)cPl2^Lsty`mJ3N#21Dkj^#1y@n
z(}JZCBUM`R-jjgl7=yUPv}m7V$EL9rwI?wbFhPWt40Y$<{J&m;J=?O_j|1gTB=lFS
z|7%g>v=-9M$Qb1^hVOG&pRi-sJ@WR~MFO*CQi$Bf9Y*wowLO<&K0!~E7j}$7Y6)-(
zp)v5hUD&Iq)&M2{a2fo0AC70AE*1yY)5^SN=5?j4E4Lz~7@3CAHS7*Y;y93okr0zJ
z9kg&M!Fuo$EIU1O^>$y!V**i{o|AbST*=De;0(z7``3(S9ElW}J(#xDgG0LreD`2W
zq*Vw5m?($iOx4P(F`#RvK4;P}@yl2D?AWo2GVLb}IGj)9>Os6l57c7tmPl2-3Y7g`
zX=^9(AiXh0YvJAD9s9el+28CrJ!do)Wbh`F`FJME$gmq())T+`!@uSB?vAhC-tz9p
z6S+KWT1OrGrhmVl#L7h($7mH<6+-lJ=>0_&q#vcuRBH?$!NvUrzyA&p)HbovT(czv
zZ#}w##F~&w`_xtd>p&F0Bknf#vv&OPHgB)|1S6~eF<0ZbRN_7{#7^1#`<c^!*7Yaa
z@(?3q+WSxdb!{3At~sTqi2773L%l7_@b)XSHrS+3(W)*+-fg)h7PW!I_2ew*<-Bq}
z&z#O@_QObuu2UBPD}{hCjeAt#v@E1(q}YZ-Xk-7VPZiFpUZkqtI+#nPs*vIY(%4>W
zVOa}bef>3YKXJgPh$f57=L6^CgVRB1?Xj>WY(X4=)+!w5XG%Ttcy_I;S9>7_WwOe+
zzv1ov4c|V0;IPcpTFJFA*XrWz#UQ$7Ulg~GKAgt+x$b8gZw4d*iK2mOUNkFt;qmdz
zx8J|x_rHH2Gc(*8auOISj7MxaQp=g!o7cSkYR|XtBj>d#mi58{ZX=Fx#EXaPG}54r
zT@u{CN^C7c?U`;smv62hLIi3<pi?Wp6_jmSS{e}>Fqy%$0cakBm{rW`=BN9AwnFIj
z`lQ~?b?W@-mDM<{>b4vbelY*hm;6O8!CB|NZz^*GNooR-$S#ejI^)bFw7Ap(7irTZ
zU2EI58iPa!uS#R}nr;jtFOoVN3o&x>f>K=_jIQd8-Mp@-Dp7orG-f<MJ|n}xxE~22
za)C)M`^T5JG!}E_@LH-5Nl2;5VFYJh*UHVfW1M!}-`}7y@NStoE}8svWImrc9%s~i
zhN#M=4cvWzv!Bx#n4f2!=b4p-6k&IB!%f)pCVa&~D$nA;mDJ36IlC_wY25)W?lkB_
ziP}zYZTA<T#(-t5oQJ}yl{zo{&ENc6o*#bTyWjsUfA`xbzWwf*9}hpUROSB7#E^s!
zkAGmlzvG*4zT$NHf%&{5!5Li}G5H0f>B7Fct}O+%ttjL=qHHnwbu;b;W40j1HNM0d
z?)@<Z*Hl9t_?Px2Yw=d#7>5SBmlu}!ae;V?30~N-UH#0@`(y!fRgULEoF*n8HT^SZ
z!k6`rZG8c-zN|m*l9#{T4!L&37phN(>uqA~;JT_?Z1oJ+H|f9M=3F6oOIBxA^r}R!
z9@6OHYR-fyrEZK3H4nf&Aq-68L{LS`$~cXTyB#4q%j;$0(?z{KQLMFi0&JIW+KU>d
zE);3iC~E9?dr)N|5L8LSfK+J$>{kbMcQoRlOcW33t2uRf&EhTSv}=zEOTv)ED^dZ@
zOZG~U`dIVrJ=w6ht7jxHVhY-&s6j5;+9kY_6=I;6i({*qGuYAyqU7=nmI(odLGX|5
z6zZSGh(>?z-IMwM1>iDj`${NXXsr$xAj#YQT4hnLzDL_{y;5|k{K(a_+Ad@ZmhI{@
zFaLfWAoLg5Pv7Sg-_HSr7C^Pp(4SST{1ZEF)i10KQgFs&2njX0s@7h_@7q(pvg6AW
z>60k8P^t%+FJKZ+VAds*(txhDRc7n<8JGv>X&78wRompr7zuI(qHG)AU!$fT%qwjl
z%0-Wb-WK4Kt5l5`h5gMQQDe@E#MYKz){?P{5%<x~m+$Eet*X_*0RxvI_jZf@v?C)3
z)kSPg$y;LXK!8|i_fA@ywQcv#deqA0dvsaXs!A?}d0i;lgKVFiT6<ewREkzkr<L2g
zJ-aC(k}+GF=EW!SJTupzXy4Rtp+Dzq8{W?3e%4$zl>b(Qk4%@&N0Ej~nHl1!*Qd6~
zc5ZEz<DaqPOQ7ENrd+-F6XyKVj-Ac7QAt0i^(Wb~xh-F7NW#Xzl73!W=Nr(TMQ>k`
ztM~V1(u8=go>R^B@~ipd?FZVv*2Ywfp$Uuk{chISchI8*#2On5pTyhW4qAcCHXAm>
zef)k;1)}*(m%O$Kyj=Y8i<7@=Dk79xShE9Xh!Im@2!V-+4Ffh#OvA`h3t6*kcecWu
znhRm~`z70V?q|@IZ|Sz9fn1!GGQ^11%+u2|&&Owu=Yunx1mZ9f2dEq|W}+zL6xrX5
z#OU9ljq+;Dh{n9wm>U;%u!G5s@9ZSakJyaD3m{io3;)$d>vm@SW=dz`TGO~`uiy5L
z^iAXU#RJEWlJt)f;h*gBpC-NA>)JjlwfXuG=s?QnxGoacK8&Y`QfPaj3AI=40RL4G
z2)Ms}`Mob}EZKAxtw-EFPBQneRn=LLeM0M1`yhzRSr90t5>sLfsrgmep7*xhw}Zq%
zXlb#8nlrll5F9nP^AKE0W7_Yr6gUQ>h%n4EYjzO5+tSBdL0ViGqTBCUGBy|z14D{T
zY2YSJthy3vMS{0XYl|y3*VB!`wiJfTsd?Fh{pUNA(ad9`bzOLRene{F$M+{5pU$jx
zLB-`JLWEM!z@BLum=~o_1HLx89#368VJ@HSKXI`xxw>w-y7<eEdXW#{XVSI#^@+cY
znC0>dT#oN+`?vo2&F;3p@4oow?0DM`z$Z+z-_KIpw+;=w`oh5aKgIe2V0~GC-qisI
z>yED_gE)Y7)P*sP#4%D!C6|>x2E8YQratND@_t=QC6KGKT5SPQ6E>#DQ|`>wT!qDA
zB&+d!ICDPFtn2E{4>4f5LI~{kJ1CiAD=8(0Veo2^1lQfd2ok-OKz!uzS_`36q+Z&j
zLYsgDM$VaeIrH|_TW-dI;p=;5Gv1fVS(Rfcy#Mh%$LDAI80#yt0x?9!ab#!?C<G7O
zPm6IkO#Ib1Z`jGiKI~X-6YCK8a5}K8nNulbtt?IZNyNo%w|hx*;z0DN^(`^+34tY7
z&gYrmfBQY>^E*B~{2PAri(m82>#wMxu-XE0OH6xi?{}yj`0o4vnXljeOMdb7uSwH;
za+$HvXxy}i6S#m$?Tfljf&V^9AjOds27)9yShck|#MmQiYGc8L<S(u*$R@B&(n&*d
z0OM#_1r07#-^HXa!LhFm2&1>SU1!_<=KY-4{iz9f6G*=vKX^YY>F1B^xJR6|C9@|o
zlPGE2k#k1NiZEW{AmbBu?CrdM9yrr4NF>sx^#yMeTC=ynNr6h?{f7r?QC{y~k>|{E
zT1YW5#6W#KVWPZ#eaCuUsim@;_DsW$Aq}LESTxhe5eY&{0d3>IgQoUaBN75rj9!5@
zZ<Scr!tLEHw|95k-;O*l3#adXWGM?N30ggv59WflwI`>dm;EIMR3sWP2>aWS*W(?3
zee)Y)90_5ho*$_B%yB+DYojF!wfc0%C_ui_*)ZQ_)=}xEOA*3A2s>aR$;3CezvTRU
z;Q#jD{oi?g_cgcsd)^=3b2!ha+(CpYKT@lpixI-t?DvT{o*^B!jvVbl>vcY9*2K{p
zxNJ(W;JSvbgUfnQ9GyW+ZRZl!^nQ8*-%?)jGdu>L_9oKw>ej*A#maic__|!ncB@<R
z_49W8iB0}TOu)6Aq)lXV9g&|%ko={uKf#uFvL_9%P+OqPXlO~kD->hemDn!4Kc%<j
zB(Ox3L?SRsZHbsLudDCjVPs5+<utSAimpbqz%&e)nYR<NGAkU8&*Z!^4g)vS4R`xH
zA_YoyZMMeBjZsJe%Hm@1p(acNX&5)@fKrvU`t;FpoVdTg=jP2lZ+3g0&NGj{`>w?c
zEjcf>F?pLFCx%ERrIFn<IUB<=(<lrn^7hpYug80Sv->p<dF9=@kgc)XmeqZ6lN#t8
zln7WgA7s!Tkk+yR1zkHZuVAYD@clD<W&F4Q{1^Q0_y3On=l}b^r{;lD63@#K72!Ak
z^uHkvBi}##PmH|co3Fm%tFLZ(^Y(^uR|#>UsGxdbHN|FovTYJS!vK;PQT>{!8CUgk
zwjFV3xP5i7LmUiDEmrGHl%Uq)t${EM)apZLf+dH}H3N-H*(Sza^eWq(>6z}8&0X{~
zKe5UG$Wrskw-zHZ4ul-qWWCR8x!$L3c)pslYT?8A2?phxn_Hk@R;U)RD(%_(P$BFi
zaHpLINfJ|-c$HpZ6oRZ<p=8y9;FE3hTv$?OJ!jU%!R8R0C50LH_xJ21aX!qXG_ac{
zhBzS7bmU-bBR2EFJ0U2hezy9+h%Og`Yhsnks+BYi+`hVHyxpTIunUze%5W57G{7hJ
zNiPhyGtR?9@!=996kA!!!q;!_x!sSvzP)1~_l$8QAJ2R^9XQoOuC9xjRms64*cAOs
zR8<H@ADXmDbl}$>L=n?AkvFg#ZkWa$zy8G?P<;qQnK`c~GAou6#DzFSb}2GU3R1}B
znbYzPwZ;8m1Cg%d`3@?9+feTkOd)PXye>D=^Qj;SYXgxkQSSC8&VVi#!wWGWaYPp{
zZrAElB7^v}!L_aq_V*Z}#njtG?{>W}!K=T&{TTr)Q7w3^AmP$E{AK;4US9yLFYC{~
zS~s2dh&Rxo*F;vWtmXoXD%wWowyKO=O{}|ivDS(P657uOhA?;l)<F0Wgn{P#QbTCO
z57~L1DcPq3j>AYyP*zVcO}t%gx7#s}11TlHmR{9W^Rd?Ly1dfV19^bX0t#Goq`bNu
zf-t0n?I#3dUn-U}SwMo(iP}98>tR{1IBmKSP4|3Ek=xzGSFhhNT4Jz)97+?W_Fq{q
zV}8BxYCrK2ow*X|t$pqR&5T;T#p8TlIL|ZlytMJ>UZL(Obck?sH(+a})=brzoKI*t
zkm~`n6>HB{o72)3NObSpTMK%nCWI?phO1I~FA914qwf>FifFB0|J75ue4e0N6~Dh=
z%M>;z`%+h+eQv8VzYO-Se(YyfPs+DlEDbpOPjJ@%IoGuvZ&xlJ2rV_U2B+$qR8)GK
zUS~&jvC)t1xC0d+CX8u9iAczV=t=L;bW=hLe)`}A7nO~SX=L1aYwY<jd$1J)X-w>H
zc8t@=Fbs~DXlp5TquKzVzTbVvT$r6PxNaVHVR(iZi9<rBfmjQr6fh$-fLvQpc>Ptc
z!BOz(OfmVioc(TMKknG?chojuB5Pr-#kHr@UJRt@&!XSM?OogT$<_a8Rnia=DF$BM
z-g7#yd^pV9?c0P>Q)iGQ#mFq{1CfywM{iR=Kr$jLJ>cA!IL4+`vgylRK}tds)CNB%
zJ3F%7;m{dr!P+=)HpA@i_AyqxodR9=%WTsNyuQsBJC>KP{G)AuKcD!%uw&Z@ET5en
z!=GTw(LoB6f`t_&Hhb~H#*2qy{hYOp*#g2CC!#1*1F&Kg63$?-Au`0sx@KqhR`Wtl
z9H2M@vv%gNxyH=2n~2Gi*o}SFAI8;GT5CXVXLU9cpigfwGhz%r1Z7A_ii8xAAmqHb
z2(`T6P+a^)164#W+KGrU4v}$4?1zyt4OkFXt(?|{T#K_*?7}X8v8}IsvTgAM4P@Ld
z-1Yvna-L_F#r3qzjCq+`QD>mG0W)zBLNe-dq7o@~B$pd<iB!#~mCYBay6nEHgIGO=
z*&gnO`MT)$c9u(URtmkfGCI?}iQZo{Je$9j%bC=8xc+b2LAriDS6X}RtZ~K;KQYZ7
z3;cw^@h2p(MgydqhK^~ECqJLn@jcmM>x&;)N@fi!)iRYrwH4Z=%P!t<b;@qwl<!Gx
zXM125CWLs=Ho_QU^?<S}TB^&mDQGofjKncA&kLv#QzV7NG)=^DWQeVp*lcQxZ?CRL
zRoZ)AD!JP!-oBr9?b<O>&B&#2I-fYt3$+x?st*qO==Iy5gWIxC-S%^490sOgBnEGz
zowYF6;^OmFU9^3>W8IzgGp~1k=*w9@^5WAG@nXz=f6JIMIcErcUSXHo(02WbghJIq
zt!GL(v*r_PK6=dC-&?M}+QyXf!kKT4UZ4$ZO!v*!+c`<|Wv#p6>ML9U>FrGH=lo`8
zvOUXwFM20L2OeLp>)>d6(a)^wEBWYM+G~4s*)?`QN1yy<{r%S$0PD;8v##qv*@?_e
zR>52c<2lcaGEk&YLLmpQ(5e|pc3iYgUM>sL0-{7><SyQU6@t#BA(Dm&!F;6iIFPOK
z;q<_=W<Gp)g0d0~Uf;c9N+a)=?*SoWOw-8Q*LM&T1ZQN_Qa1;km*UPQxCidRgS5k1
z1Y+!y1ii&6=S)a}VLvduxn&eVSB2A&V_8rtSy!rgwX(mb1##v_E~F;Yz+l8t_~z}z
zukYUSU;p}FQI|?S7k+s6dmc_t9=urPti=-s3B*2J_DUhv9Zm)F2~E<fvhIwPWjRv{
zXiemNrsRw^CQYYtBbhWRzxdUj^;9^TaXNqCcswxNyu&zB_1T#$?3`v^4QXt-K-(a4
z0XRe=A-LeK^#0H06#Ka15PEBpPs_1dw?sqi;=%+xpuxqxO~9&G6ScK_i5j6*i?+|;
zDp9|bvo4v;pAaPX^)jUwg2WdT>%a8%6FZj6cd8hV%aKH2<xG`K4GXm_s0N4w{uDlC
z#~uu+1tcb3$JdBdL{?9FlOWLpp{hRd>;3VOHD{_ByJ<&~fxEXiged&&;{jBt0n%>b
zt7+oq?vCBf4O$nLS~|%QtJMQVF}DE}BW;0Q5ExT(O^?!B5g-ncI1PkpfMw=%I#5xj
zX=-|3mms(ALTOosp-l%%*zFP_MZSLfn)~67yXlS(hZ7ISBM-*|kLw9(lgZYWSV-`L
zrbBEVn)<a>v`JKlOD9eXP}4ASGwyixo4@9V_sYBHzjY982y}fYVy4=G4-aQv-Mr!M
z{#S(H0`4XUh)+<`CFxL4wg+4*Lwe;_F72FA0wE3_)N6I-WYd~)fH3$u8mwvW0eYpe
zWldY8*_t+nsEgKzZC|OD3nQ%W*{#|u8!+%wJO2BvzJrUp9vRkoG@sA<Q*3!4Fz=sG
zS+IC^z_<<@%fnX}M)zgG!P}a>+33ywErTdGQVBj!gmB}&<w!=Na5x-YTB({BjP`q$
z^9;GN=FB24lt1o9UVr@+X&TAZ*}P6=XzgbW;OS>NC@};GiKvEc;?eHccDspTO30X~
zQBZN<?)msg(bZE(@mOVp>ss;_+9Wnmh>9e*z1ed!PJFeyCx(fGW<D$@zB@c{)=bvY
z7;3?9yjq*a>(>>c55MqsXl*)gKBy-6Tw(_z>dNE86E}lTee)PWNhy-jgq&SBE~G+C
zAnTcR%`D3^H;*fXx18q#xvmgKFDTUx(5bV_JKzUss17zsbnPJtTbpMZoTVzwKMup-
zQ(~=9#V9?AZud<^keJZWrq7v!XC=4kXzjh#u5H<_z{pkv?fQjX*rz?9mpq~k+WJ{L
z@Gbz?Selni()d{=L=BF32|zbZ_KNbnoES_v?~l}y$z>+UfJg$S4YcUX0X@5bs09*{
zyWI^zl*rP6V!=irfv!rCOt=d!gOMxKZqLo#j&hz+U6B-=J-OQ>AraEVAhBuVIapH~
z@B?mVss=R?`y{5O&t<5CPPLGig^&W%?Vk7RnZq*kyWjtwIcJt?l(IUItKmiUq3@Fv
z1*B4|PZul-H`AWG-JU3sRgHJ~z>o8hr*)>**8YdvaM#D|dr{3}3JmvYLIatv!9n#l
z%5lf9e(@``9QfhKf5W&-+`ZnTp%wCL#zaYjOL3eIM`~Rm2#3@698W*Eoy72ox}a?B
zk<i)A9up(66@6hF?j3wUhqQM1&_uEf+w63+<K{Ef`Xpr8%&=aa!L$N&2-41yrgq!t
zSY2awV~Ppe`}f%4PhH;Cpk!6nYV33ie!i@K<m+=fufMF%bX^}O8vK_J&@i%AR&{D_
zt%X`w1R;q)qc{7+6WM|o=oOn5+!)-j11cdVpAsa3TK3@3TE~zo%d+zD_{csAyCB}G
z7TVZ*z@qDA6{|#AufFzROIy-LYnA!vJ8s!iYjbN_T>W*!ZerN)dD6<c6y6<99OjvM
znW<JFft8OawawXhAm@p9N)S`v)$I+_G_vH(X<j&<W}fB~r*$TqZ6nKjHCfFA{?1C!
zc)6Dz49Vr%+sD^hz%oTLAnbN~QjA>qAO=<t3URE&bp<()%Sv7cG@e0Ls@B$maao}U
zQ07iwOxkolL&HTr*UX5*=3x8%?nTG022@*ME!vVZ6BpWS)5-#na0XVkYe4Lh_}Yt`
z-|LOaCh-1owD7a*r$1x3)<P)-i#s0-@ENUZJ2r7U_99?oQda+^HB)p321H!Y9dn`O
zu7z}|7`6(bdbM|wkr2GiM4E<@)dHd(42odYLE)uT?)SIcy}4z!GfSC0015$Oa3SG#
zeT}#848%>_=32J&Dwr~9x@`^AZ9)CEPsWH2!sGG4!{LeJaw4Qaii4j?qOFzc`tCJk
zRofs6U>Fi9B_aV;c+M+N>&(M^;#@LW9gxxnh4qZZlit?8)7t?AOs|D-{je}x47t>d
zG4bl|p2O+JpQO=PHHt|krb<eY-E_+sg`A(5^AqRwhEi8&ueH{m1~dB|YT%r|X9uD^
zbk;T)qPL!fCZN;LO0oVfmweQ;iND)5Q?gZjRS=Ed!ltfY=%7MGLKAwo|5Q8vRN(t5
zJNBuUdCf4~5W?`2pZQB)|1n!$YH?e(OtrZ&Q))}X1Z8ylX%rB*DhF<(`AHk=z!+f{
zz_fx@L<)$pn+9qy9u5bBB*t{ZI85wzBX!L_L5vHVCk>HTw*$L05LKvaMr+wX$kvrd
z-8Sciz3c2CXBl;{a~KE4I8g%Rvhv~Kfv3ZfHLoZ>q0dF1FW>LNy6EkN+3!XarZF)N
z$=e}WIac9uotaB!(L&Km=<)*nplJ-jt=)(Y+;&jTAJWEh7Tc!zqDEd@`(#R9=u!o#
ztt~7&z@<&*V9gxP?}+1>QkM-JzD6XiZ;bwhHzt*fu{(R&&*?4FvCi6puKgFJeMfW6
zI7JA-Tf;|ft@L7`3DJPr+8B*)lh?KkY>XSkV=TkjcU%0?ulwqcpI-LS|2JDMmArzC
z12K$$><#`R*NZzL9h^3bRiCV9UBjxE3;isHu$`~{gxvgBJF5jqHF>LD2+r=ef&@Pc
z!r(DXu9XxAzIpo%!!9!H0&z%KDGsVdFTSJ@7?V$MmB#2)EBK|m=A=*U_OspNeAx`#
zw|*-R1-#`w7!Rid562@Po}Z{{q+!|s=MD4x#NKdufJ}(SFoX@fITvLq%7>+J%BxRA
z?fQ-Feu?-Vzc9C3;d#?}?2N3-=Lxu`pH|+ydPObp?T=m%$<@z^Vzmt?5_bDQ46vLQ
z)FMlMAeRHB`XC!^=fwr9wX08h$)Ykp2vKo6?mj^J8QT3U&{>k5dA^1Et+-8#lYE+V
z_3Y{D9NEr?24rcqEpO_qwXMC7jrZrT`S?8g6T)r9Ct&6kEKLOYykhGw>pyLMdCFhb
zpKrZ9OKJHB)w}BuV9~--3rnq7$!J{}VkAoRo@%=a>X7ZIb2g8m#p?jKVTT0ooewSe
z$$Fw{CfJ20Uh>R(e#bXAH{9IZkO-s@QS)k?2BC_PYekYdkk&?YS5s8`^sA4~*6XB9
zJYdr{BG_6h#CW@Byu0BzKk;}z@$LH`SxaGFGlhbTpvx6cMJ|rebw?cF8Ab?G;Pw40
zrZ90>7amRr-akL{a6WQaXDkG&h>sL+RcvdOXn#kq03%RLH}J}Xmdm$OT`>(v&5R@L
zZ|+DbUD_M~v>;^CSSjfY;en-$oaY2-MZ)Sr@>hv(2af`50YI?Os?<vDj?c{Xgroxr
z+lRL6^yzgju<X&c6FPjNlnUfvB!p-}lIHYV;u|Gg5?|nDv~g99#J1L<pWEht^k2l^
z!9{-shDb=$&%W|6YJJL%L$l)+*z}}#t_A9f>P!+P$UxPMx1Nljv}40Xxsyc_fhZp6
znl&)2CC;*IOgmLdshro$Kl%EefAPz|;@^Gu?>U{G=`u$WBR-hHg~_d!3xMT<TpowD
zR-n!$=`AiHT$0FZz(WE^1F(L0|G>M4M-JyB<8I_;ikHc2T(!Z_E0fyL0)t>oleZ`*
z0!5V%%b9oci674gPOW7j&;aZhwiZ28KZ_8roZ9wWE&{36fPV~B3puaYq`ba=%j4k|
z339dHoqQi-Wt@b!Z||w=%5r+(yu4>v21;3|wSvYCDBGBd;bRN%5Y`ZC2v~A$8R@!Q
z+nsJni9z&C-S1hel!x|x^eWMk8l4HN>Ka(BEir_Egf5=%0E3Gc`}O>s9sji2`>BZ8
ze)6YM3d?FR4a8yp^S1Msw*DTre6AI2#dP+xH<XsF1gtU7u9}WeDitFUi6)G7WR$=@
zMT)NEx`1Q|h5g+vixqx+|Hy8<;hWd5n8t}$`y1xt8I?k9*QG%j;=q2tV~hjI0;hAP
ztP6|@i4Ly#dBUcvClXxerWG-w7(%6LVK?0{?)Mx;IIjzT`@7$9oM+~BAw(l=E&cUk
z$Cr!`RR=u`BP3y(2G{gjGP4%GFRK@m%0g~!z+(%yQ*V8Z;*)kwUWi^_o{?~wRiH>v
zLUav0OjypDQk5``P=s0)p&(Mwy83%*pp=!<eB|l4QbVBBxqZf^WzI`oU0C~a2@c{3
zRfH<e+DJH?ySHS<*fn)32t*NLdm^R1PY98u#!ijJj3NegXzc1UXho^sUKL_%UH-^6
zjjb(Mn?|P}A%Xw+`YCwje-cWmtg9mXku=`?3Gek6u&(Wx=AGJei};aTa0!WZJA0k|
z>QHV!6R&YU_XR2y6(t65H%xUPA&ddmI#X)~DTpbuPY956VH$V5dG!{J1(nP&B<dQl
zD3}kxCdI%m4g_%qWz|A$^TI{MCl=!8|K>9ra4De|as7&8d*2$3ch67!`0&8HhX+E4
zyuBY#HILbQtkJ(`v*W&xYF!BthAFg(iNa~E9CPN!^O<E?e3Cdcj&X3Hu=%Tt?&Q^*
z<7ZK|9;aT|Zdx-|Ghe^?B{@fa|D($WEV;0jCLLf(Lb$ngCivlDK~0(0%37Yt<phlF
z91feu{dj{&yC#o|yrAR&R`9~L^&-;c^}SGRt-$P|egosS?hV@KMxXMywUxiP1(z7X
zTJ8}pF@P5bI;-oaQS|1Cf8vrqvSB|)$+;j)f%JwDKa<${%liLoeF3n(tUv2ATK895
z8o+`Je2N-HDp@Nh&5TJ%!>&!+3zWH0jOCJ~m?gx7YL!Z52m%(_S@bqAjIy{0yF?#b
z9~5UKtvGYR3gerJuim_6T^8o^*(;<m(sdhR3=AnTj3f2rEfgVo+gqo+y4a{EUMVC*
z95h0NBoVVfF(IXiX;Ny4%%;5i@WA`$Cw~8jZ(Sc?8mTpJ;26*xKucK7&<Y_A3=)W1
z2{jOFq!_%fD<AU0yLIMRi;o<}nOB`A9-_AfSYuE?uaW{pM1`oGQ`|2=Kmx-s^2;~h
zQ1yv*eFRk(%x55m9flF(#4tqa7#K!S%gjrm*dwq~bMWd-_kh+_t1j9(UfzWbS);XU
z1cU5aN)?b`3_YpU*7kr<+7!oViD1T{4KT%n0K_=w)%C%wG|<r-TYEBZ^vQKwS(y$g
zUTRvG3X|~(|N5!b;+|3q>tg7QctrP^tdH%udTvB2ZpW2MMp@Jt;=nN666y#R5p6BE
zQhwf!+u(*B2~3dUEID<Y25IfOW=5%SUNi3xPdq$6@-*CWd_HhKogEa7-kKK#f-2(}
znWlj{SG3fYRvF$c=uv;<5(pYT-Kc9mcsGDCjFC|$vKY0j{P6BQ56=fmuB5?Qi-Yx!
zhXyP)po|vG#n7nxhUJP`aNR4dJkHNN%`>wqxn@c&#9%;h_68kbYi&I0q3K2ctp(2N
zZ3A5+PJ*$Vj^s2kj&GR8kztq!L8w|<q99<27zXa{29~o@W}(z0r}all)6D5wfC3vc
zq8}HRG_rAB+SF4OBL*XSZwp{VYkR+aw%X;$E7hNMz}q8IaRx{gAw&f+);6;J(#Q}I
z0ur2TM`OPF(hF$G#g4n}{=HT^nyM)!D;C6i<36MHr`fWxYG8;v#$;IDLmg0B>$Ov;
z8{;iCP`aX4+mndw8HqJuYe2PPH6VlU{kkr!S~$*yVT!zd^9IwE$M^5adG!lxjDeVv
zGXsa@{`&MxEej^KF(iHXgEiZ**x*sGR5*y$5_V}o<}gO~w-eu<9(i2O9M5N#C3{=3
z_<b<DFpw^gY`=;A?<qOkvn-46$=y9mt~|`%x)%zFdg@#Ij&mo*HqTzQv5mWqOk<7F
z_oq7#K}n+g)th_1etl1hiIfJ$X-6ecT{YWZqdwGRzl)TtlzQfHIIxt;GFP+)zoMog
z+HUw29NGWA$z6-Hyo&TT<qKfq!)+QU_fgSpsIA%AgoKf(WD(c9GuPDf79(MsA{=N8
zc3OMmM?gmVf*WfZXX*R|M}M*7um6Ahy7&>JlnTU%q|YZM>VWNb9&`Ooje-|{6k$vw
z!!RPnfyZ347iEJroUoU1+Wl??LLrdS&;}<fSRjz7wxXt_)*60ZW`<$l)!hxxr$-K_
z6QwHFin%dFVlajX(<H2=dQl}@beybxhQX${Rn3|VR+o-|S`5S)&q$yuJfDs{9-mpN
z5*3ztp-Phn=m1tX-u8JyYyeMX8X{E^>sqmz$;CJqr4$FJk_oxBCikhh8_?TP7P$cD
zwC`noC_rO|2H^PoNV19XuYOGoKCr^VMMLT`X;Z27gi0XS1rgWuQe9DNt;W{B*L7Sf
zTv>%%rW5YGaMNRaZAI<PrY^{}G&XMQx^5!)7bau~UQAS6C)XL?o^K4Tv%tD;t+q*)
zLD=Y3TL68DT{>f~oi6mJ^r@opwJlo<u&NnaKw{jOfnU}??)Bv<e_4O_rTCfI+5No}
z%+FBU#>A%>NaNrXbp@ZC(o#ZkKt|e$?d>2fSQcT>=<$fQvD?L9q)nzXf<i6M#1ivy
zzBe~FJUu?LEK75o@#=tfy`&TaOI5U#tyQMwWu7FOS1|+esvS*;L8#S(suUAx7||fC
zraT{yJUt(Hcz9%Yv*R{OYfZFkKgbq|nhTwV00Cz+geK5Y4UQ%ATr!88S=6iG&TMgq
z6xuZ*wpCQYY39%yc-(2xf`!dLh>%j^&HX*|yfUAyCCIetfC+*jArfO~zdR5wTFLVX
zYylZB&b0MHx{LPYmgd_;g^kT7o6M^_Lp8LwC$yBw0fxrG^76Z1TWi&b<`a?nJzdoW
zF|}8m9a-?UkRB1TL0WBXHv@R-O<Qs4>PLSjif*4@wUSGF_nyFfCac@7wxH>S9ivo~
zY(@%^ly*RZ>IyEsoegZGue+FR#;k5kiJ<{lT756Hyrq4H-`lFLTXHy`IG)a&jwhCR
zA+Kx0T>Bn!n~5nf3;|u*984|#hK)Jk-EY1ZIwK(j*O=*gL4pr=7-T@jgU+YJf#d1a
z1e&!4c%R-oT_>hrm!MD`eyS?0MR}fA&N;I-eI?b}TJAtBZkFl*S#QDUp(^x@uCt2T
z)<%H5&Xl|`#Ne#2G&DAYZh<2f2ojif5iL&1DLS(j@qn2wxP1A3YL`T^gXNnxSNm`k
zYc>CNKd!*4evhQzqb)0A`@3NMWp1Yg-LwLGd+?<#we$O~u2b53{t-LAdQ>m}@bQoJ
zq}N)~+d#?BZT-h>*~lggl1OQf)*V)d26SFPVCy<C8d|cbL~%bgkPr*rAzqsgYhZM3
z0PT`}U+r#gI3H%tr=#OLaM{2i5Mm%jVXm2ZIW@h)OVUjf3YO3(US9k||6ZZ{2wYft
zoCeNMXP!?7mSv@s>b^~{ThuPhuI=7+7iKOlo?>u~A%%J)WZ;xDluBrTtw$`iIgD;F
zo>BIexEH5}Ttt$)O$kawd3&?x&CQ+|gG(l)gsvZ@bnUg{7zn|rx-u_k*1A$kLRA_=
zv-$qMCp@xZ<BOOyq4wDJOb-&;^WJRLw)fd~n0cYHd+XTP)UMGdh*Ye#lzO{o*xIx&
ze3gD`T$n7t<;>Zx(|*pXUtG_N)!t66X>tZ3d}d%YTx|K{rQ82NLS%@^Comged5f(#
z0Hi(3ZpVT%@<WL22sK&++9nLTA2Y42>q6CArfKB)WUR~L?Yy-86T;RVks_=`De99f
z<?7l^`+gESTx1g1NOsM?iU=_#3@=>H%fe}%z0F;4t=8!5&CS1EeUAH<&~BbG#KcH!
zez&kzzs6b(Q)%tLwsjtCZH-*^_ZGZ2Al9*A+2ZB?xyyXw9G)3a?~JUiyTY!o&(DQg
z7X&Xfxo*_DF`i6YZqOzi_a<p@vzCjH?v^+8jpF{d9c0qlG_N-lUXTmgf20=xj4rJ}
zdp@d)*6M{uktTqDDIhsXI&AHNyx6PljJ-Zz<)UkNy|Md{r|Tu{?@g>}@-`okK0iDD
zvi|<-3xM@y{n^%a2(lee&GF@a-UknAqbP~QU?Z_e)5t1RgxPLg?}XVwr~(W=`7RW9
zF4erk6M_;ol5}WeYy+tjhBnn?69X2=NtgxBW~@~SWajeBK2D@~!=NXIC6k8*jn$nt
zo)GuWn_);qOHe9Sn;5G&c|NMwRmbE));Uy4U3mB59S?_tiw~}aayUcqnsp4vHpR{F
zQ+FVlz!EAJD;ymYIzDBJDG$qu^Le2zm7s3ai)GAU&|#}U=Ll<rp@!D3S1}Q47%@?@
zWP&=|BQ|<)h?&<z<-dJ(!|!zByVZyl_W1oYGc;r@R8mL;Mrz%2bK{Bk)8hwnUbrZE
zT-^OdrE4S*Nb2`)M2G=mRF)7~gD|QhYN!saQW_d)pp+$p`X2JOM0c_o9~B*If$(yg
zWDLXRP(3*qNG%`<?e7hU!F7sCDU?#TBt3)#X5Qxdlb!!_VYF8ZOMy|88_#Y&kM%L=
z(CoNu*<stPtOOg0j6`k8qy?%8y6&h^$#Es*KwKj-i6^z$+kG^~?Sp~^qD2O~1kKqX
z1M9M~{P=-~n>ReZdEK<Ppe&w9p2~?ZMD9jGbVkn=q##MDbDL7uzz$JGWJKbC6h#zD
z6#4)#Ardq(#XT`9bv2f}ay*|oEsMi))nF-ZJ7FM7!Ni**Hef(qyGx?$J1r--g{P^o
z7UOtcS+k<qb!H;b0ivozy@0tYf(D2Z{p@Q@OG0ez${qv;BB=;bjrsf|WoLY4f6WIJ
zArZsK5GQ9>5Gci2pyN1k&;#!u-*IOvyR!E-isJU>0jFEA8ECDpo05D|)mR4xiQL4&
zN7Yy8-oRV`AeTZ!7`MuXV%6Id8~*6wc>i8P1KWI5xBHI37^eny^Z*?;V7et6p|SF4
zc3i7NB%WXRepA)$8ZS`SOFM4g7%*z}cNna&WBotM%Kl;2pJ2-t3nd&;8Hh|I^Q!n_
z0vbu{L=BZZt%NcV*MT?{Br2j4!30YI3yQ_!OpEGlj6eiE5HusWp4f3dv*^k!6`}*R
zYhF2<@{rEl)tPvFCdI(r{fL+kS2>+q(i+>@EJBrtiU5suoD4J&%0P${F=j#-!4k*A
znWyIir@SHt`ypXgsY~`mNO~&W5^aS#Lm8BU5tA^4$YC8g1Lt~REg-W%79@itpn<s1
zP<1L+Y6BmHC<!b!Uuu39sbN`BOBm57l1gSw#%_RJG)yZWPKC4vf(Eic!FV`5aAUW8
z^XgZGsZir1ri!fI%CAikvi0M(O;NqoA7dDNQd}wt(L@RZ6{T7wmO!X3yPyK+5!bh>
z4*rQ5rqRDo13Dw#vNTtvh$2yZ5J#yk0XN1lI(QX9t5J)#)?m3vH^fK`ffz?hX#*wz
zFK{*x)AXt}@*^wmn6`9VuBbvmNSc{g{tq{|Kga5LeP~A{c4-~#lSO0W6|2oC6TSl-
z2~xyae{#7;4$kBTj6l%V2&(>kVyw-XMPeAeKv64mS@^rBN7j=0<u8A+vE2LMggLKF
zLD)}`Dw$<{rih{g*j&)6l;CG*l7Xs5QNu`B2u&ACne*I$Wr)MbloH2z<@m!po{lHx
zH6t-KS%?nC!ZtkNib=0(EK6Yt!jb?Ig9O$l_wjsMDcM+;UYdg{p$(krG9ryNNx<i;
zrlx;cH$ZhOJcU3N=e91(fl(5Jtt6Cv7$5{B4FvJ`0WWI*_``cdBJ)xi1|i1FG*Z>l
ze4Llq%=+}eN*V)1LDWIWU3AfYZ~61$Pn19qYiDJbQ81+TE(x$!ps(h3>;SL1A8!7_
znq)!>$%}_{z&&n&V{8DHPf5KLmSP;-hOL*?Y1*FtjC#IsS)T&2e0s83q*^4!${0VR
z*!s)*PhMXDtS{@&wl<yfkE=s=VP*L*qto-HarV1LmIvBCy180KwNh#&4z6t@TZGb4
z{HA>&5fe{*0xqC>T?O$9vrn$b&1q#ZmON7jA;ufTDpI|LEd;2mJ8-%BLL%-U)S94d
z|33YfIh!L8$+fa*VP0m|oav*#oi*fQ$0BkWiQMj$3vha(TCvuO2RR#A3-eOQxm;LW
zupx#$>Fouq&SD8ho08ZQKHJ3N`~B=y-xQSlA@V+k3z4B!Wwy2{t>k&7W<?}143V7*
zhtQns)w7qYYjnX-+xClBb^@`5Th1D56(yswI(nt6U5gr44M|sfhQ_G!(jn4<?d@}7
zOL7~#&-aaN(Asvbt)1az`L07BpIV>zTWSAk1ur!r1!4FMnHz8gNPohPySV>01@byc
zZtS|Mt$L`929FPEzyynzuu!DwskD}2YoVgiH0{>Hl2^`kWvzV>M5TnlS__&W#)N3a
zDiHm7Zfu7uTKvXR?9+-qI<v$ANt|t1*1}@NC$p&%VyiBy2VvXzXkgn;{eCr;mYPuu
z7|N0>Yc+DNt>WuojM#-%ALQz$ce}L)$ai{Se-Pbv`@L6fT6QY~mDpJ;Jp#pbKe|R#
zYAl2`udI0?tPWE7u!qa};AY&}p6V@s?fry6lIX30deQRnv%rJ!OWVR`7;cX7rFC8O
ztXq|JJKLOn+rA{*-)>B_OM61sHj}GFrvKZ^|NrC5AA@b%BnBAbpZ40n#PuJwWlNU_
z&~+!ldk&?s;Cw<#OWMqRt!;1(BI-JW))*ZUoBoPFF;!*B3ropltIjSK$Z8Y?6?I*!
zs)RHWQgHTSEAI6D)PXtwI$h&Unq*D$bP^hK*b}q77S<)VGa!&sbOCv5uWfs&g9Juv
zYm08fu?<Ek)wLvN&2HB|ccL-YF7I<W4_%JH2u*9I`IKf8eP7#Vqa8X`#uS*6Fo-jn
zwDy?5OE=YwWt~aG;H)C=rFN!AJAYnW%cc^f+piumcw+8gx}PUXY-f`;5W6#KeJ`0O
z)!JG*FHRaW+jKAmA4E}Ak!TxpeIru#=P9lI=<2?){cN~;e`_EGn|7gH|K}4X+L_%w
zzKQ|TXX28j$3L52pxv+Q6RopR>jpA(yZHE5=M|aND^{Jqhc#ytu*P=rSQRmpytNRX
zmzgDJaw(MNi>p@Fk|`;AtF^TWdr=5cD2ngP7Ka&~#oXOYz@#>|pKEr7wkJ}EKB=^3
z2SZhfA+>3?-8o+Qk$#Wb|7$gmBeeSp?;0o-mRzXCeY6i1vep78I0N-^CTu>GaABmv
z7SqVqblZJj{r|D|=RKArOO_}06IC_4*O$oUE%|C|^b7_K0wh3!<oEvw1OX1oVGpZW
z)m>e0%_SChzt_xEMfit^s+oI4=F6(;oXL0-n(z#d@ay-Qsi~@n9653X)6F@JVw8Rc
zbcS~KU`=^?y1JgOCC~GiMyJlOLpDx2x}IpFYn#PlHcuQNUGpy3RDHo5Z)4c>;{q9+
zX7|tgbLuY`Ea;y6?ZWo{h5}=Mk82+1YS;GcYxcvV`#ax1;w+8z@evX!KaGFI_yk~m
z8b9e6E=<Q8dhF28zR12g;Gon1g3R!Yg&2!oXl$(dj;0&R%GqEtFe?OK)+x3#q?Wx!
z?gW+_Nz_FnyLi08vaY;&eB|Br!gal}Y!{|#{ORozcllS`p8g8EtSsx5o6qi%lKIoq
zx86+<CsOyUv^TK#di#o(ClyPf!Qt9fStee8^OlFY@b2--wKdXtW-ghVQm{>E6@pQJ
zr~>b^STNn(4G}<$r>(J8<?Z8@TA|j+s*#*=B%yE8gZ>`k1wiQ9Q5mAt!-bZ0loff2
zfd?%VLncTkVN1$W-7qz>`|m6dPyGG=>;KKo%`Nx0Uy#Jv7+FqW8}7i$!5Bu?*Fant
z2g7sSNzxv)nvqnQaP1&Bv_Ve}tJ@8D>L3vs+5<8Sb2Vq^2+7MnN(#g4j<lDmNwDW$
z0)ZMOr8&yQB5O6A=slQ!HsY`8fVK6#a5|s4z5O@@V~<Y7Grim{4l{1s+$mG_&R3zg
zwx(^A4{y40-QGdl5LRYWvH^G}w`929Ro*BG5`~gyFX>B-q&ilQ01wNRKfQg;>vykN
zt{2wp6X?v-^2Ggo%jtYZpEkC2A>ZDSN<lPAhC}pwFE!HP1e!H>KBMdfpv)dT|I^ps
z@aFnNT{em&ZeQI|tYNkda2egFrJsA+{Hr!4nXxtx=)QY=U~R_3vVmsrvPS~Rt}i35
z<A(Tt7)Pk#`Vv+kiFe*nXFr&3C>yI*NR8X<#1E~jf3=mZZY-PMXCi}CQc`ZFvokIf
zu1^=*wg&quMF~;=@Nk?(K^hW!-tT>+BwVCD1rCyi5C^cINxq)U&!y1B@#oNgt|V*(
zM-ubV_npW=fh&25Qs1jxB-S(GkaMQS&yI~Zo&~%w_VU;Rap`>@pKorsySw9K89#O}
zi@~Hyip0@oN)@=IiUI|r{+PXtlG(f~a4F<+Mp8lZhS(Kj;ksV=`kUXg*+ymK2B83N
zS|g<!KDWYc&YW%p%N{5`UKh5ld0>Xombz;)+3X-oYtGCy@A9eY<!jS4qa@baczSo?
z`f%m$e9tsx?r#g%i||<2uqB5X^J9WaNV#S2zGkXCt-{iTr7jML)J!&SFrqkvN_)_F
zC>ON;RLt#gQaAoZ3#K7?p#F&(E&qOTf5WeCZ@6!8ufiP#<BaL-z)@A+zI}_C+q0=m
z+?;M%FITqB8)^{jh~s)U_V2Yhn>ph$0n+@8@pixkNeMGw3!U*3zv%&fbzRNgpwpsz
z^A4_fNoBRlMDclTVP`uCE-vGcl0eJjEa>m!KaY;2gNeccZ9FXK{m6M%vNkqFl8<UH
zcUTh+cBb2%n4gpyqy}lDH@PvJFiWEZyQ-HEwj>A@32)JXuxa>2?s8Ty4=$NBoxG|b
zL$b!IaM>zvpB{Pl@Q!!y-tzeHfY!>}o3~6>`KzyPVDnDk^ZhO7n=@}8yupu5p_hoc
z?i~{BpF~axGjmaSF{DAB3e!CC&9{Hz>vwOsJUxLbw|8eUg6KBpN6r0_akj-|HN3On
zrioP-s#dNvw$RwLCWvJR&~aH2X9M~hW$A!%_cV8f)ExV3=X~`YqU5BMY}`x^I#1@T
zX>=o*+c^=VS>x+(zh#<EeEH?CL5%Bi84P0Ux}678)vv>>dby~c{gbnUXksaJe7nZk
zZmTu*ZotWtNr5(xWF*ghSt3?QDPaYqljq5UEnQNkWM{-7U~LDmjzoat7`O9cAHT$}
zUjDY4`xAL8%=5?BP4v_Fmyb^X)~E54jREM#u%pF&=D|6`2?^2m$n$R2Xl86ysjRGa
zbt6pn!&_toRK0aQxJFe%ywoqdPK9TM>6i(%HkRv^WnEd<%`^L`2bt<lsU#tl;w3Vz
zd0joO>oBfex(>_X{zN=ulqBl0T@!2FE-c%|lqO6v(|KZ+{mqW;YI~MwIJiC37B`q?
z%2GFHZfK*K(C8x6;Q#`AwR-=IqXCj(Bf8;rgS8t64I_2brf>%+amfA9wDI)x=!}<F
zco|03usy<V#QO)1oAdWg7UH(9Md6>-FV;rpco@Ey*6gt821R+?jMLyXvg{_fE<xgu
z@d^zLBwDe@@Aa9NXz_f^AobUL-?eaXmJeR0nX?Ysv6JJ^c*yvlkC*%S5F3R4R5*>+
zwRBp{TcfhF={4RrQ6xo|u+TV(O`1ZR!$F5a%DptSvwQ*uHSc6pTV-n@NZzk!fET?z
zck`@kYeVOyulqq>3WCo535-C>vI{nETV>le?&cdT31yzKx?)=$d#h)X$DGmK*Z2@I
zBzt3nrB-UQXRH(t#`iF#YeI~vf3~ms&*NSK-0^MVpFB^6c`oFfNjZ_S3?Rnu(Lp9K
zu1gPeEgXNvnlKQ8aRL3#j(53>dyjyz`=H^Bn)y0;&UTb|hg}rjRQGomhLEsl+7n#^
zX=q@@Voc(JMmub;zK=mhS=fgh;ty@*=Z^C6Oma=TrXNqa!oO@Ur*Q^0Yt*pGe!f<!
z{rSBt>}5`#Jjh<_B!%vNnt2QPMs;>VbOQ4dvFath$@iXj7OPyZSGKJ&-`qY6DhYN*
zmLse90C=QCNd+^b)<!0AlFa>dA{Qa`WLn5W*?8X1rpb9qVUOJ2ciXJ46W7aJ$$me{
zu<EByQi-IUy1q{^fFzh&5<#Q&GwRtrz^N4GQpia|SE$mY`sakb)8;JK9PChQM%Pr{
zYkzjw>6y5)Fqlb_{AbRP>8m5OlL9G87Al%z5L!OR(+FU1H1Tqeg^z?oS`ZbDKu#$b
zb5Nf#H+FV)O0Jh`$9J~Fy}Ve|&%mqkatk(CqkbfN`DO4vNRsw}pT*pH`F6!fG^?J}
zQ<tk~Mt~ccBC)R1&a!c4C6Txzm@}cfWd6*XDQn$Wwv}b=U284C%do{^OIvNU090#j
zq?z$>jx52>z)KQ9mKrGw)<wkC_wgxBlvFt932^PG=M@Eq`G_wy*VAl~Xj7oJMr|_u
zh+YLS1Y+Zw_6H%(Jnpvf*=G)~`oL`U#!@Bs1WhDRjj&0<MhjN*v8j}{M%`Al>EZnk
zrIB4Dn4Qs>!l!+1-1ft)`$zKZ=l#bvuD^Ff9{VhOi{9`kr4$7vey;TYeE%l);A>Ax
z9c|6aVGO6?aP9jAzWhCZduQr?OuqG}@oyWS0IW~rfBA7ZC`Z?k94?GCR$Hl5mT)Mp
zx>9Xr<qe5SN`|EykQqxgGKwt@;Q*epkmu}r8mW*e;avH#e7!C_ynV;x!z1gqQ6`}{
z!<kka%erzq&799Sym|A+MIxV`NV#x6-$o2$jPAq{^rk4Cu%!O)mYHs5N=ZC@{f5Wu
zl`n2TLz5tPGj+SLEf4-)EgZwc9QEoT_d6YYaloqa=6YdMxVGw=6W#j^HbO!IZ0p8D
zhM^us0uhY)#e2=X=+2+0Z9}yVLntbk6_gW-3+t<R0P^PTN||RUu+}Fomn*H+xSvD1
ze>emY-~>Aa3-B@8qPx%Rr50UlO3Lg8@nE>EV*=7cXdMA>r#H9F*HS4rV6fE7&>0f^
z^GkC5ow_*-N21<e)etf{mFUE?cPV`4IKOnwj}95}9L>unF5ALsnSSJk`5BHE&Q1>?
zQ*@wFQ&wBqv~p?9oi0{3TY0K)5bj7xImsDh!cu@-$y;m}ZC<936Ee+^y#Ya{K6752
zQeWA&m33L%iAr7$BDmQ4lrm*5yt^!{TXn6Bl$cKrf(tI*n^GG0G5pX3QoQ?Mn=>R>
z*NwV1{`QMsBS~0G<7qM8Z5MauRYtJbr>S?tY=GqqlaS_wRCroe8s6SogaEO6M%u5X
zXb(>P>OB07qP_mlD*mB|t!C=wk@xpEyt=<*KHo5(X68AQvPAKaXWgfgK^xzE{Wb62
zy#rudYjj^a<3icpn~aXBV~>HWlO<A`=$)8^gs9N)fZCehQ7%(lbB6<Zps?p<+i=Vo
z<q2wysvGls>-)NG@tQQwEYJ7~TYbJ@`#zSGN#?*@9};B|&sZD#Pse)<Vx}rh;Ib?{
zd_<kgJ~%$KmqnaqK}Ho;TUpx1wKXrdkEGI{+slG7siUKoGdcx}%4QG;3pA2cYR%7A
zaaL_zSGFwJwqlc@xv;D&*Ja_G*RR>u%2(&Jmm~(*<0YMiA~Q`J){JTns-~QnByrA#
zO_l5Q!mZ3O-SMlN&(Ng2;jN!tsx)&@s>>O4lA3mcf(khmWXe#zWKIJN@kS8^a`ydR
zgt|3Ep-|{u$~3el%`#X*G!5`qI)D^=#8qVy_w$*%^Gu$vq`8vjCz@`wT9D-WQr3Jg
z%(FLZ*p`J>TbRu5v_N-UJor_-Aq!}As69`vsiUk8z9C>O$~S#4Pc{wP>lC0kBMw03
z&Ko5XI7aV)IF-p+R_cKHwr&0lb1>-zOcyWN%z5};NfII{8g6`O-^P#+{iT<YgEdz<
zdb(bjKH}P=7cu@08mV}71C^>3HMr{P+HuW2&Mhg4OcGCWNXmHKJvh78b<7|Dq(EE0
zlmJrka(oNzHXQu9T%UNnJn`=FkyaZe)=V;^wn~x0>3rtx<0B7G7uLmr()m0)P!(sa
zX_w)EoLomrbAjZ5>6U~FgFXMto6kr&lTU@Ut$epMPj;uS#uR%qrl=V#4d!YrRnY7V
zr0Z4LnipJ(fTbgV73S3`!Xv>ZQZU_tybd6}9M1j~PE+PIWe4q>H*G<^v$j%cR%y)>
zuDK*{I5AILpDLG!3w3Son8e(JYYE0xVe~HQ$qRrb4PU?8d!mFFBTDwAAIIUUhaVKB
zm?`{i?eR&%waZ#Xa;D@=j$lPYH?%~VXHI@!UGTl{U*UjJMG|T>?fSFhka-DUR8_GC
zYpX1^{@`za8h;t%Blo;MjgM{|V0U8&=G`$enr&>lu(p*&m8!~O3yPP|xw?C0X)C5m
z2?20PuCvlJqbAXP&LHXso@F7lHMVu56lWwzc3zihZxikzT0LaydyrFGb5X)j=7}F7
zSscKMj3;3{nz*p)+MFqJce=rnXBAshuG>P@LT^ERXuN>}O_jAars7PsRJ<hOx^A>%
zirH*TXfLU8lX6g$iNvdSInwxh8;qweNl7<!;zG7J)66Ls7m1flN%Ns>#|8)_Nsyda
z>&D~L<6tM);c7<42fx`a6OwXd+fvu>aor>9HAw87nUrMDTzA~AUrz{5b`+w6MopDj
zGLnQ@PJ{85Qe=>l-~SQQ>tI0-ss?1$TNuBWCEDk?Gfdpix5M)RJ3f=Ygb`CCc*eD=
zvNhkwTZ=3s3)ME7C`l?J8>?oh?i^=Ht}|iYn86*{9aHGf<v3bvtgG+YlwGv8l)`D6
zDa9RYXHp2YI&-Q&fu^Bllw5>3#rN*l#peqSqKWU<r8P+2ne%o&)2#8dUTL)jm>cZN
z7wbU{sv65?l#`G&(V?5#!m$qR4igLJ!}Iq5JWEc`>iPwn)R5UpvF^NEP{PK3nkUz#
z^NvC4!F`-y66dN&rnbu0-+aT<<$C;7#=&uj1|<UDXa6iR%7;fVJ18Xnz5B$S0q4K$
zFN>e2uCW1K%S3|d*!NpX+1Zq_4>6)+R_s%=!Me=ZfwxFDKikJ+`C8wfeE&WkIJI`&
z+;MjOMU4m`Rh<F7ny|H%t*xPJqo{2(e#~BWc7mz{lGYB{1|&*78`tZ_&*3a?FAZ|@
zz|NBRzDz=%3+E{1@e-)!<@&h<c<o6gOVE;cI99JqcOM}cn&4U&u65&73TeLKB$=x=
zp01%!DPU+YzWa$0_Sj%+aNUgA^_SV2Qk$1$i9jhb-lHd;QtF_z@4F<$8EXR=3PyRb
zw@nF#x$AxU%55MH^gTA?V#+G6iz7n=Q<wsvet3At^>U##i}_BG@EaSJ4$VnFLz0k-
zdMQ@ym9C#eB<M&WQDT&1U$h89yH3u|9!vXPCPb*CBTfnc%5_AWmw2W^NySUXtVNQ-
ze=qm-35TcB?}L}y`n}N^TsDk}9w>r`v+8F)`Z;(Y@b5p1Gi*OY(*ReiZfxpF;%4ev
zsNtKYRLD|DI^)VrIc%b9(3-`2G$VzJ)l%;*+!<IaHJH#6&-FA-%+utlLJ?ZRL60)U
zd18Lvh=v{<@htovdSb#$2^nkUS~n~Sd78MJPUMtX>I$vJinej)_xE!U$N#Jbo5E6~
z@<U*&!lodSWBu+G7Vyma`5_~_Un;hj&A;rJC$z9hMk&HPCvu6>J&5w$0AxwQa7*Ev
z3Z-}%-c~yURt{hBeBzAg#Q}=%=On{|4WGt=BDZ<Ht9Kf{bj*4pWJQj;Y5QrYAr&!Y
z7oYEvbZY7vWo33o>%gOIe~+HjjcM}aVSt1Y3x4nTpiQ>1cem!cYoEqn{`dr7eHuU6
z80YNx2lfD=8`&OwSOROFk*HU$?aI>JC@ywoP}oI5yv=R1jdRHq$)S4^RCAn~D`sRf
zdbheGWYD4{a5|lkoG~idDz@n8uG7SY-Vt`OVDn6qMfq9u<nR&)gG@oiXpzYjJY!XZ
zUcG7Ki~Cy)<618~wZ_x31{hQ55;{jGt*WxN%Bcw@x#80Wm+OW~f?&OPzxfz(y-D}K
z3teE+|F6M<vgi=vj)U#SE@;y!C2r;ur(7roMGB?Nd)ZEWT#gJAIk9Xj504L#89gon
zdG_U*TX$?y_Q1!pu_63S9Y61-gF@(?nV?Mu>t+{q-9OT>AttI(ijs5j?rqv=TO~=R
zloaBZUb?2X%h=k0%{q8>0IvS*I0V$)lE-AL{P1u7T!y_|Rv+5e{ZHF9C)RdlsaLi}
zbiTRLXf#nsjVzT-7w>$RXAe@v9kQg}v9WcJGX-1d#k{vxSr+dcl~eMpUn!iYnUaf_
z8N}Dt9Kdn~AMx%t$sNdC9Kh-w{=$2XY@~y-wJt0@<2X&+&S#dovc2oT-0t8F?PeUj
zQiH7-S5Zz4N=cs2wBB)vQC37RZSBr{1S4YNfjCBYn~}Zr#-tJPB1A{96lGJVd15Ms
z){NRh6GU;{tPt!L$<(^?r+0rGWjv!?!kj7M<G@_P9Y-4nh-*xFX-VggrSY?I{&X;D
zP)y@~)820TDBkYQb}T+|P%24iZ3MF`mrO1hPdn5@7o!e7NaqlY^?Fp+_~9Mh7k>YW
zU^m6Re<b6_?_~$Dsx?}JwO-k_#oNrAqO3n^FB=KWwF6gEQVlRM?WNbPH69-?oX;~S
znP{<;BpMoY@Jch3JTuR3V;T;`Ce4(2rVG#e>y0gYDT&2;OT`1dD$am?xL%m^#OX9~
zlFVA+@$up|KRLTY1JoSMJM&JcOM~kMG$`4#{%ZqOA&XF^6lGQBofst{*~{Hr)-K_i
zaV|;F8z6WP?+jl{;m3$;k@+Pk4h1$IoAT6*O`N4HLr5Ry(@e_3+jnnxdb;4ay9dLn
z9mI<F8`+xJe_5Dv;5xnQXn@fkC-z+&z(VZM1`KiaILAfy^U&8^Xr0#fOw`{2>V}Y*
zgALn(3dNI8Bk&c%U-AXJFX_IJkBgiSZ0zGd9{h#m>+)k5&#?CHM|Za>z8^{m$@Q3;
zDyuH6ZDG+Znh-cUYhxo(P?=(l8k3wnKuj(hq3WRVCPoVIvP}-?HlS)_*%oTB?o&$S
zoH<Vu=Q2ep8Elxdn&*ku1Z%;(Yy#O?hB^7B>)q+PaJ!$m8Lf@WwxS8jY2tpmp&)Dz
zSI|f%(SaJCB^koXvKg!aL|gpMTA^8Dp5|eXB>=E)i+j?=ZDJbW6?AXU1=qp$Lw022
zV9JSkN|Y&~=FOQzQZ#Y$_e&``!#W8$3uSU&xi$yn{ao#p2!1`<ao8y9REy;7NbK&5
zrN>D9?}F{GLk@bCJs@!#-2o_X4C6xmvq;$d@PES3p33CURx3u7{l;b3VF3?~wP0X<
zcpuyM?&HpS3uKF`mE{wF^}mVn3BdX^ezMVB{q6vEW|T*D><5w>rWI`)tuEAdrRm}?
zl*p*!0e{!L@$#_DRx2url-09~IXf^>T_jXeqfC{}Y6OXKZK~9D<yxJ+kS5`7%KZAP
zU-S8^&qz~oc7!T<p8P=GRvMtu5v3W>Tq{Oe1ej%lFhH0e+Dl!zw9Nq&b7Q>L%D1oI
z@NT=XwccK~IwKfy$31Sk3S3nwBHpEJo{$X9>Tv3?8b%<rVIr&8dRtp_D)uP}v+LPY
zwM!_6<x7mr&9ET}q`-7a<Oz^UZLV=(T1nQJ%$P*@Y|fmscvl(iQnzw=v%~imZ;k_4
zhEP%=rA(F-LiJAm#u{6T_TG3v;HEO1+AxHwy$eeKS=QW%nocau%Pv!L2cd=td`X2Y
zF1TCUf=2foBi>JX$tBSZse@QH?&y0LzwkU_98(~?e)pDTxqjdme|}@9c^?8fc<<)B
z;J((4y1711WyP{H=~^|az?wrKy4C0y<k~T!>Ro}%zc!q;WGk*8GjBBW-^qdWu1^bZ
zYh#P<NvG`ZmFI~(dDc?ZsHWW9-ZD)m9xsbCqD;g4HbgR*j4~S!c87-rN!ZrLb@MXC
zl7wZwa#^pudwfERk)}km#(L0&*t!lJuD7vN<LT-`<d+Jyf!fZNB+B_x%8;Ci=0I=n
zzi2KHE8@Vt30r`@RCZh?u6{*PQF8V`J7*{bN)Ejr(<ng{obBKx)YkQQ;ylsTo!h-W
z(a#F;Kzx+mB)niF#APKzlz6Er+BOFha`8DT&dzZ^+83BM-`zPn21=ZV%@najrg)iV
z>VaIM<Z%8+w<dPG6J-kx8yWtW`}q48|NHCQZ&M&8-abB1w~xnJ{_%SmW;3)_YTeju
zVNDw?7>e3{++MCITNI!zfNT|-Vy7!=YXo+F5zDo*)y9$&Q-ZH<3iI9UrM4y54#9Hx
z^2;x|xpBemtyZcyIFdwB%j7I<(V48|+4pucR+PmW@2(f#K0a_uA(MDo7oOIIrK;Qa
z6illR00UT*ASuYIRD*SO@#ijJ-*p(1xS!C}m{Vs628%M`Wr8LH#8LGUWDOunLRdND
z^KD6)CS@+hCSDSm$z)ClC&V7cny$6-`t>(lF4ut>I|$WF2#wB$v%WS`aVD$aK!Tnz
zb?~CE78AUyq~cu=)mf=BB`*=~3rcrwvXorTK1C4Tno*_`IhRo~rUpe4bDo?9-$K{a
zy2f3UKw0c%3vHesolF|L;Q3!eU+vxH(eLFW7)QeBdq0bx_jI1AGZm>TYkOj|g%-l(
ziqWF^P^Mw(i4122HB2&FNLwTvgkDVDa3w|SmD(1jWazf>`nxx5>q1rKG!;+uyt?D&
z<}^z0)V!Q`KAl*qMZU!;T^5-WLA+|h8bf=Eaa6Er<GT6zF=f}@Lyf2F6Bjm^_XBPu
zm3R(YHXOjZEI5rYDcU7KBC{@|#5Nk8AR{r;Mbt;8I^oTb;sN<KNd6wy2Ag~;iPMxR
zl2Gw7KS_ycDz5EF2j2WYRb|^&*8}YC^Rp!!2h^U04iGJ266)%jVjnq(?+L}`jhK?h
z4Q|7g%JH*ZHe`3(d%6Ga%XcPqYa6EBr%nOtc@I;=2AB;Xx%>Wyy9l4DGqJvRe4kB@
zd%s+-yO7?e@fSTl0a%~LPc~k<74|)UF^dd{R_`*X)xn%NBs)vOslOg{HBl1cAdGlf
zLrQ{4M$OyND~hCs<w~>NF%fY9t2JY-8x1IAZc5_St5@9I-g?$4GEt?tbD6St!s$d;
ztdTX5XhEqKQtst+j(n&oHChSxx^KaQs>kb<MJtMDwlrw4etqMp_uvg!YXq<}qufw!
z&!Bj=sCY2mOI*6=YK#CQ<Bn(RSxhHdjo`rKAaEQ>o?Kk3%F7oTwN*N+P4I){oD(JH
z){VTvvvP(0z-2dpVRRhy%h4H;QBKqmPZt4)l6Duwv$3ImBJ>1T6n2Iq6eS80jzlkO
z@Qzf@PRSAYS_`LnWT<);-6_#NjRzO#&&9s)Io5u4Ch&B<&_4Do(c$(6M5&5tqv?w3
zMx$aW?*OzIEklkHC6!DsiE)jD-t^nq3)S4oMitA>qzg@CFsNJQQa!LH2?gJyDW?#j
z*0E2@JdtwY>ADRkj_8)tTB9Y`o<PMlru$9GF3`Ny8lNv7*l!!lwjq+pbMdq1KxglP
z2I35|s<3R$8CmK&I%*-bY`y!Fm+p%6?Brp;g#KC-8bs+)LSlcA++X#N+#S&@y~Nkw
zGv_Ycs)vr_25M-F$S4JPX8Eue4=l&{u9ppa%|g!#@7Y{rKTBJH_&%@k<0PZP+h5y*
z|NXvQWZ44;FC7fKVR1jnA6O@`jjbm=7~r6<^n4$`w@rJb^n=5%?XoOb{WyZ&9~>|C
zvf%7xx1}r1wjmU5L{<Ke*~`*FEWyoJ1Fgukp;q_LHfUAZ%**>V3+48f=`<l(##-#L
z%gxOVInO*kK6xP4I?il}c=3PJ>Tyi$Wf!ZqVUMrt!p-CWtv5ue-XJAxu!F7#>vdQc
zUJ~rAN<|%%Q5hV?-ULQc>Y-q;jXEG5Z)u5Mb|<ly{b2KNQ|L0L0i;7tMowO?oJeF#
zBrD#H^l*`_HJ+Xx*tQ1AqBHOQgkOHIIQxaL!+yq0ZZ*|eSZ3;FngJN@U~mV#yOv+D
zq`OTHhV<ZlsgZM{ltQi5_q@5i&MA+saUI_r;6?afI@p~ZX!xEj`{Ey;+2k<=e(lS;
zGH4<{^YNkE=-(fI=101LHmq$lTWPiqxeZ^L4%!O14w+O;GAVqxrp^M~0*tPzZlSHA
zZKG;+g{0Padb(iR=p|i}GxKR;K1~NQQZ!Zyz=EETGzNH~^jdx$dnOQA47UYab*Lq$
zZfo72uTs64j4}YRj;X$!!=?^W3+Dd%2PRs2Nneyh`?Ypk*a2WaC=xPcQNpWrFw1tE
zlT%M1MagY^<+KB}?%Ue<zSa%Z7OXJ7XR+@-wV%`7_fB!WyHBm7ELX=ryF`i{-al-k
zF?^8DoZ5X7#{SXhG+NT+-pR1R{rT+=pm+W4F;V}w=eHv-k-86$!vSr;wr#t#%ct=d
zJw5?gpT<u%`nfVHKMp^8{%^)66&3HMqvBQdDLa)_8#Ier@i_PvU8z7Z*S)d2GLu0j
zwz{%zD@ivbHQ>TU*T}$k0Lg3lyU(|+4Lc`(@#QbLFU6Tsc%ITVKa^4mqRK`${K6LX
z`$faQ)KItxCW)E1eqXnZ+j-_x3Tl-`gY(2$GUt5a%~RvDh5)Om2L<M=>OQK0ZIHJO
zlL!dtWijE%jYM^bQy=DBW8PbUKiP_dJ?88a@20YiET>rny%<-nY-xegu(G0g^|F1k
zmsKTWSvOB-Jv=zTRl1iHVy@u8l0a|8?GCR8?3O3Iw5xjd$U1Q4UkR9^+exe2?v!L3
zpqNWdi3jvJ5ZJ1X`!hXi17uERa?q-3=$Gw6)km#>9w_(hdFxTvj~v5xtHK5syRv;G
zrANohJo+ph^Ki!v5ar61CL5cqG;tPQPf)h(EP^Ib<9F1&b+p+ULaPh2I0&<1ET%}U
zm{k^g;L0+}NrqEv0gkD$YGu_%L*4&vidl-_FoL9#kme4zMOLt>j;@&|F8(g&!ls7r
zr*&DlTrSKxK}ocAL&Z4V-&3Z{d@4NZJ7`xlxs#hxhn5FE#M*o>FI`jl=nVFG=6GHj
zVk-eF_<D@y3E-{n-O=p&ST4I<b9F5~Gfqi3A=qZn%=H1qFlhk-hc_SlOtO$q8LfiV
z9wNSUFyj%3cZJP!j@IIVptT;lJDAY44F*ciQ8v`PM947Lh_}ADdPv$GdR%%cb#0Ok
z1bC@Ymh6FoT6A1{-ofxg``Bmp&+lU|UwrY8+OSQzJUp?b^ugczIgWqXUcOLe<(d|n
zR>YbISthJMaxWKwT8zaS3r1=S*hZBHuAw(cw^X87vT;8D=5POoU*Dg}Y9wpU7_8n&
zR7A)nfN^z!=q$e9s4g?mHC=37frL^}Ro~k=Go4SNWeIC-SP{Ose?`^GrW+S)IE-~D
zx5MYi+1bEmQR>)pMxpVBjphouGZ{?+Fpley5@$oAN%J693lM_t5mcL2YP`=Rg0>Ca
zHge6(t8ruCie&Ww%=%eY+lHd#$*`KR+VADKpyT^2R5f%p%H&Nm3}?1e1g#DlB+&rz
zeQxrc!%u7e`S}PqbjG9aX9qE4_!qr99%jsC4%VeJ(nx~VEkKB)Zl<~&6@kzqg<~@B
z`-gpKuWQ4q^7!z?nm!H@_cZ8L{`Z0P@Bc&wG77e$y3%x|QEAc~MKK(BCR%c+I-`yi
zF+o`!kkw2UV@i$HDyuc98&KKgk;PW3(=yR^-7hO`<m_uFg+#-+xSnu{3$ImOpHV~O
zN|emx&6je1W_ko@wQ{*!nC6*NnOT>Wx>d9|gYfn|@w6;lmM2=z+;nZ#o-GkimTmqm
zzB=4*ww@GLKZ_)l9S|DpdK!Q4)c_sV5_3P&n>ty@3;0(uD2i8!RM(kGwkK9x>(@c8
zX)2^-JUqT(T`H~t+-KsY?NTIXQpuy+aO-{#yZ!5O8atTlr%<!XuAv)0>+5VJv-Sj*
zCx^?}!xr5Kr&NZ&)yplD2o|<_NNL#K#ZdlR*ic76xn369@{uQpKaC$XJ^@&t#!ohS
z2y*;;A9nmVF>3HGns`v@kw#=70<<DPEC*pa%cpwDfNe+sNt;$Stt6>n6_t&}Y5)>@
zmg-n)5<V2AxOk|jmjosxUgu7;h-ZenNl;T-1db`{#oh4jz>Bs<tCb?oGU~Z6$(fv!
zm-WbsSOm~<-~&UX-Q&S<zgjPwlW{Z80x{WVJJHz({%^gc-G9{s`);&%2G;(3dPZsM
zt!aC(JMwpjXDObE-`34Dwc=oajC(j-U+82=a>KS$M?CO%TVx~f>{%c0RKs|XBK^M1
zsm?I&X?&t}=2hsfM2%$XfyIDBWPcs|y#K4laJCQMAO207(tgm@{`rg#CdPQazy065
z6rnrGCY=h?HJYHewl|4VTnTw#t0)Q4%{#~et9i+SXa(C?+Q!le?tLF0*Pt-E33leu
zPJ!C5xOcqiIu%Zp_e^vTL#-bU7UjGt+K{)#6y<LfkR(j=iFqoV=ZW%kafdd<pHqs@
zvA*w&=6)6-BN8AnJILlUI-I|K4nwR}^1hD`E7xm)8FOupw8l@|{kHxC3uk@=J-b+L
zf1X3b?*|u0|DSkwzftBzfP<ak)E(nJ<Gk;%e%2o@==i?<3XK7J5AY@a>w@mRF#`cq
z9X{b-HhySpj(zM+IOHR$R`{3g<&ACZWe0OQBlgGcWlchgeY{!o{cRgd-B_$0o{NmD
zu(*ix`NV0S@s57p*uViwMWPu;A3YE<VjV~TFJavGxY_r$GakHiR`PA`T94(-);3ZV
za$DFzt4NN-L>W;a-<NK;$9-4_vwG4a;4K3$i!#4fgiagmjPn8LVGv!KjVKH5Agu<Z
ztU@+FFZDQ^p$F+8JwA8Gi`!>p4^YL=2ls_!_+9-RJ^m+HCxi`Ih><jcF)2IKjA2`!
zT?XBhcRj2$jzH;Pfw4Ds===L$GmbX-y??W4Zt<^t@Bf9z0IO^Vue$dG$8>|CR#8{1
z^JGCkm*wCO5@rGO<SG?0*B@Oe-e}tzn}r;Qu0UmLTa@p|w1#1J>4jZGuCI^d2U=%o
z1=Ge3>|aS<;(pvxV;JN0$)MouXvIjSAVnzC%u?$}fNI}U!)M&zTRDQB{RIa{Ic$vk
z#<DN?xE6oees)>BPq*JAeu-z)*ZE*Z9Us8>eH+i&%ve`%py8~?-S6jc?MAoTDFpN#
zMALD)cQwiG6T#4*I$V7xK1p;19z?D&56A0^Yqq^9i^N&^!m5lJdFfmK=<WR8T6p=J
z{xG!Z$3mq1H2#g_6M*$;{A42sgQ~(n^p9td2o!M#?n$qBuRdpDnKWv&Dz~8Nl1l`X
z$`%>nwLX#9sG6BbR7uq88Z)X_jLNcJ*hCq+r9Cz!^o`8P2iicZGe)usGm-H$cUF^l
z8IW!o<xLv8lLk3^QY!*qS=|jlV7*+qJUw#0nLR7gnwQ+2ZYfEav+z#dAiA=w-tzl2
zPaafP&pxX7=V3-!qQx*}pU<u%QxZa<3od7k%vM!2K_j$v)E$Iw3{}kMKbe<o#`{g;
z<%Um-aa}!QB+bFPW{pZE6=Rx&+uKadji;w~)NKv$>}c@%mum+|28s*cd+A_gKgFNJ
zwyhpKq25wEB~DR>-~fU*=WQX_=>qi0XGJ1_jxJOpjM4<xO6lgbJAgf?cP*~Y3K>T7
zhXUH~8P9BJ|NWF8IYTdfBt|yp^o-sQhv^7ks<;3I4=j)>S~jSzccV@BvRjdOKV$1$
zMuVk2kZfX>##Iv=nXLm@!P<DziwD#(Z^tBb%C3|1#jAThyT5a1Nj*f&(QQf90iIrx
z-~g$Y5Uwg{QckBE2f%7L-x|6rO+byLFg1&gX$8u}=cill&NqC1cgLR|F6j2onUjJu
z{M4}3k)$enKQSFnWE#M1Uz09|=U?6gU<Iq7$-8H#<a;(nJS_3M;^iH+Ygw2Fy{UmH
zST0!d^T8}3z5O>{R_ON|UC(60YblR29YDrm5jd+cG3A+(Gi)h>lf6?BsNYSB%<>2%
z{En+R8%&AKYF^GH@u0Ojb7nf9JxB?`*__g7@7{-F6fP|5Liur+JzKaxeteL%5dgfm
zi2s+r=l@#cU$U3q%7UhiRu)n<Z?k@^mP~(sFJ~1lL{iBbq6=n~ORLm?j0I4d@2?H4
z_{W?~I4NjdAww>wp;1WgOxf(+CWDdN5VR;$YD~Acq?BnF9qUn|_HQC_N`*NE^xI%g
znY)`Cnr>9R^7zd+yjvfTU<;NUWw6!D_clq8A$4%bI>_QWQ(~^IsOp+<Vi~iB+3L0i
zgEWE6SgH=9gchHd&nm{|i2#6Fk#$9DMys=xwG|R4455mSK>d6^quO|U_v9UxCva9)
z4=RseDcmb@#$`*$<ZIhWV@lf=jN}XtF|rt^$$`1eHUwcVnW`1l7W#BHJ|am%L?Qdu
z0SV@YH)QC^!cieL5`Y3EO3J9n&b}Q1^1opm7k=#HkA+J)mEzh~<{hL*LLe|Aj3_Tx
z7JbC1w7j9CG_!+PI{?$?bxRvcV{HksiIo;?du7v&Y8$2t(5T|w!TV`qc)4gUDUwOO
z(S~=EOeSOvCcNR#wFqhP<f2P|G-nr=f|T$U&A;4VH?k#eWMbCDRE-58mCWs{&&X5Z
z{_|VZgm2!wrB-D#<2281dqQWaS_^IP!RHR}ClCALj$~mEl^uA-@Pso$4G=D+$*&!6
z2m?D_7w<#{=IP9Qz9W|tvbjt`iS=wDJ2IY6+i2>-<SulX0#izHEqd4mVO#Sq#{pAZ
zL<7?(di3OSNxpw`mbhL&BL{YFKnV~&1Sb=&^I7ZWe-Eu+Q>6qjC$=Z8RU^q$B$^3$
zUGeDp`>z~&Wb^gO{qSEtjlbaW3BdX^ezK8MitL61B{?xlb|PUh0=8-%G)ZExqPjKq
zjA~?8JFw6N0mTJ}tGE-jnRw?U8r3$`Wuya4ibLPthg9U{*EnDuG!pBX69;~pH5VvN
z9&l+mUq%;2c4rcc?r14fLRA7>(Dm*#Q3NItN_qf5qU%y;Don7KL&dz>_+L0)hbtd-
z`eA@~Vsp&5^{ns?@K`t2N$DBiK0MPZ-X@pgnjMxA?P(bsj^CWJmvL=ZRBM2i$1KcY
z#tkrMFO)o!(?nNg&gRZ-QN?n1&ionlk`J@$&xS^4j`h!y(Xr&f07=2r7=OBkkR7Ph
z?oc!6()-G=ky#qF>-UVWeb{cwVFtt({+@;YI3nMi($FC5bJ#n)9b<sbngeHJj6`M6
z%v%p?Tu()Nst@qaaggczIJ0W=j$o`9mCbtjF7#63z?l3W6<Tw^Huc0|Fey7zBE>$=
z;_EhKHlifYg~B@vp;x=Q(wk9IW-b#`fS;CNlEk^p%w=*>b^49sYfjub%;Kxa%ti;(
zz8**i!T7$PEn(W(fH8-C5goI(5RBFMQ{qe-6BCc8)AakVzBmItK+i~WA#4YAtsjPm
zSH?v5`xFg_)xqOL#;mJ@1PjDmr$mxBco6o0tA);ra1hXCKZl7jCjY-ai(7sA(y$>r
zIYXZPx4-)Ref+^4_&tuXpP4*+ZsU1;B%|+T2lZWdWpr!o_uYwnojNGqmDfeb5A*z?
zy)0nVc>b%;xI(o`wTjsmd)Wfyl`sc;$zAskl7+T%w3oMNyx}$|O8k=yA!JqWN?1ep
zFo_3}RozFBY$I@+QbChZqD1~QWh^C6USzL+&(WA5!bXZ@4=$;(y=<&Eg6$=e9RyZ$
z5G%?7ZD>KtU_+;1U>r$BUH8U;Tra5{*ANVDi3Sgfmk~Sco1XgxT`@YI>rU+KXKwGx
z?9YC%BBKLviqbq$Kab{VrmC@47w9({U8k>ifHrd|!$ciuxBc4p@%-P$9_|UHZnrH;
z3j^f#f9`V?=(qexoBYgi;(eRkI~aGy-bXM>@-Cdv_i=L2@5SMz_}(+3T>eEjw}TnM
zM9}1?tpoqn_n^3@kT)*cXyI2Y%>muRryk#L*DWLrX?%A4nXI$!*nOhIGc=2ir{?Tq
z?G*^-$;k6GQzVfQTI4=YIZ>v<&2(Z)Wpu$Fx>WI563OU+n8rGj&`a&>pjRpMyX!19
z|5?A=Lup}OiZJZvU^d0Oq@2mQFxnG`ue2wxY#70*c==>o>R7Kw8@6AA3(kuPsCr)l
zN%X9JG6!fgHsXcO%<5X2gW=e-;fHlJo=^C5-Wa9(H@%Gi`CQTcy*>jG*!7v+f5LnA
zxcdOd_c~n8vv0>f77p+IaqQzy<KH$u0a%~LPc}-)A>7-@2qnAddL7Jz0Q=2YS|w-m
zEax+Dy^D4R#Ta+J5RKR&x|eYAY_2-sQ&ARHYM6OH<T{g}9gcnf%5eaDrYG1jt&MW|
z1WZzpWNaxiMRmi9XB#&&luU|jr1W|NNu0{W-RXuR4j>hQQYO;P9Z8McjNj)s5d?<F
zaq?jJoP<S<s}7Y~dY%<}i`*#3vkaEON(e7I`waoIXPa_`tggpVWs9;8Z%dexx8o*^
zelIi1l9<XJbH1T)LUcl!*Z&K_O3}ofOl(`_>G2&+Q`ANsEl-%0$c7U_@Hf-vlzhgT
zf8Jz5R-<TRE*UMEYuiW$Q!Zf2(p<g3tm15vjHw4UlEVcXSzLD-NXbD4N#x|EpjsQf
z)U!nZMCkMwQ6|wdOR-juLa-kUfIsll&OGnV)5m0AODThjpCnNvv1xVP3wY+cY1N22
zL65{8i<nBmYVUx=m<N>w!F6j~G&luoqf)78h{$@C=$<0%3=U}yq+Fk_tfep^6!7k#
zQErpXgW_2Vh>&e{apDFhN|VN_E0Ae5_UdrBGUw?8XF+1!Ch_$+onCn-Bnz?SE-<XS
z`<qg70B(4p;^d|Svs9Z>5V9nidK>NDiLV-?90^TG^|KD!%eRaQ$w*0|sk8jzBT{IC
zc*&OspnKljPq_2>#@V5A(2p37q4krP-Ang-m8N8Ep+{J}L4q0Qlvq_URi@(Q($}^k
zz?3J{D(kv>z#A-<rfC3&E=KRcR4;!~bTFO-aeX1fSvc_}5r}jk%N_7wfgIi9e?<m(
zRFFib-a*NK@!R~bHl~tCR@@US$!)T}IMWPdvo6O(kQXV3eXJn_H8^=OwM`bhsLlC<
zC~W5Ch&99l>R5gI$}0FC?Ky<flwvTQPo$b?t0L-2AjfFU^;R<Es$6yDx~`A}D}F`@
z%C)Wt!ZgjKk~rB6nwXiG#2LAtpIofGslZ&Y^UU;3p@UbM#5oBXS6{=g(VU}sOu>?i
zo?}MU>RokZ3Sd7#7lcYL5iD-E+CF(E9kekAqLTwz1@{@sJd>vaB~e*e%f^y6nq@?L
zD|_z<rWy%BptXR^QUcS%+Sl$@P=pYTPh<%p<_%6I^Tqv{hs%Z6>m#O{pC3~~a^gIn
zP_5MU(OIG5(h3%xVh<4W=m?sk6c~Zh8@K3KLeXtHAetneR5-#jX$PmfO@5A39ifyD
zk71LuYxzKD_5J*Zv)#O#bxw&X7YB)TFTpjSiSncyDJePGU{Ln?dJbSgOI`=j%xH9*
z>}_#O>j0v+*a_MgNofx6=fdNan)b3i>pr|1J)Jv=U|B{YLV7kT8LE{{6Oc*brD5Pf
z!>LSsetrebnO!2;wfLr)GEaPVzUP#`!@B?8%i7GV3N*4=%|IieD@h=E*Ier*_T2#X
zZ#zH>kwo1a6CCMQ`;0`y()XJ{E;FT^uvO8`4Y_pf)kxSl+X?P&&ltwGwdl}{#|D<e
zCu`3>$Neivd3JW2Wl&*OB^#U*Y{@&9uIiN}p45wXF{QQu%R(9j>E)MxIBPsxawg@8
zoC_rvXYS&ke}tk8i|X?zPsD9}pV}SVdhTEU`5ioPM+dyb*O7^!9}SoC)A+ZHPXN}Z
z@so``PPaG&BbbOxE#gp;nxWOflQsn|FNCsW&zkuA4EI9z5AhCMTO@=%u$2+`(60b=
zFKBC#`6apV`J|bQuq+E}DX3&x2m+^G;x;8vMK*64o^z&|v2GjWLP{{F<_GDF13(St
zc|uZl;c?Xgh!`PK9zw<%VjPBRPG!97UNUn%0QOvqmtvKm=7-@IaF$e#j5zKfcS^GQ
zHMO1%w!M}1&K@&*nUKY^X{t0fRFst6KvZ*}un}ON8g^}Hbwf)k<TO)Li_H14WX15d
z!k%4AC|MfSigyC63%3IIH)q=UOq&Y7d-FA|Rf-f$loBj0ta^AMuAwz@OQzA9+oD6I
zGp>3GlRNspR>r)#fB+sr+vsZOOf`F^-!Zbw5_&MlEYVA=d+>?PY}ZHo^IPa=F{JM?
z80oIgY8pBcV&T}t)|NwHA5Dn{^zniDkq9=Gg=F0UX$Wluyn`vofkEhR&y*69gw`6h
zHOPLAx$DnpKSzXfafkSt{d2Mw)>gg4ptumK=<G0AGMZItbn~9(iM9lTr5Eg|vp#$2
zG9%0F>)RdZ$kUpQl0ommwr6|01gw~ASd6lc*rPUrUT<sPnUv#PjDyloIe+&~Bz15$
zz;QKNt=QBHJL5k!MBTw}cu93inKaLArhXRMC<Yy)h%?MYqV%x&)n=5FDn-qogJeI_
z%?9Au^ZoADM$7Ni9cQ8mgNBw@0D{8E8hh!lMyB5eW398*oJ|+^7s327MEx*J<oJST
z%QV=*5+J{#WVa<B$#Br%Ig<ePJVV#BY6e@2Bt}VL7wp+y_Fve8i1;hLZGo~N*7?_s
zL<?=4!?YfJf$_dE(O<uNOR|kGZa;UhvMHv{Dsa|@QIdq_T}CBwNm1I~R#-2jT#(|n
zc4716R}#`xm~y7HE0{4$qM9e7nzm>p(#HP6nn(8*Y$c*Yp&VdRV<+)QpGyfXIT`mT
zQ5tNPF>#<eV)ormiqQEIt`OSDEqQQg&N9yUxnh`W1|?@-CxrXk6RM5N(<9CCY%jMh
zg}NBlI%jKV-&uh2<v>PC4N4O3Q$cdZl5n-`o!f%VeU^zP#!3r+ZTIy$V_G^}XLaU!
z*RktN^W7rvwsBu04lYQ5p#$K1xtj+_Z~z?a@aN(E_Jt49JnDB&_s0?V_CVh?L<#0_
zw~T&^EqQs`vZ<Gq6`}K9#(sbHb(nYAK_6uWGqx0o1lt7`9HR^kQuasS!}tEylvR~N
zBm1UskQ2ZfqT!{Fv&rJ;4bzH9C6VbAO%sEhm`^ix-DuIkK?EdwSElRrO5N&l!om+u
z<ILK%nk)b`_aQ{wrjKL&us-)$^Yh1(oZbHJ+^oGSBIe<+pmTBtmNzniO;^y}ZgqPs
z;8G&k5JBI+gwCSupW>yXN^2V6TBSFzDF)}9`PI$Drpi?tuh%P2bz`zbwu~fCcyD&a
zQxChJ=anu2KzAl=%t|`muYtz%^ia$qj_8`j-Cjt4Ec?1UFe#tAh%X$*4?4~rJoN7J
zoyq!X{AG_%0M@7RlZ`=d9VhJW^bD%JI2WL8@!H(Rp)Ao9BlR}ebi>mf@EzPTT?g0z
zJ;w0T!QtZQZeSX2M~iqvmI<dKUze3djW&7kl+>6lfTG*kqdFE+N|ch>wvDa1*rk^{
zR8mPWLu&5$pH4HUDRH?x(jpS>S{IudYpqU~P6{mmtPr?8Cn1|rU@+GRpe5xJhBxLl
zoL(&L$siRlFXikt^ew=%!d@bQ8Kc99)!~UWQ>y7;!Q`ABz@lgA{MxJoSmr{grA(xJ
zhJ1COyzQL`eM!{F#wv+MA~LqBv~A_~eBy6zZ_wL&>NN4~n?G^YMv(&2JX_ujw1uWo
z09XQ^{SU2&U}bb?KljXUi&!BX7<11XOC2qhy-dUuOTBZ_hZtD>x$vHG9lmyt@{};p
z)(fJ3_M<x-CJqSnDD&vlS*(Xzduh+%4T7SfA;&N()A}0i8Bp_}SMjnSL`mX-Vm&xX
zGUiS8_TK?a)>>&z9l+APq|GcqB#A~10#6Cl5}Fdry0Wwdk^@*Xg`UN(W?Wk%r-_n^
zgFgY1iwUC6EM4lwyKmxWpW%%cGQn){%L>RUvXut_YyW>RDo6Ps^rAc?qt+P-d%)6v
zbHFi{_6j0{+FVnRo&|4h2tkh!Ca||D-m+Gh=0eUhbz4CFZ1x0!2QHY9CA9Kzpjv_n
znYA%R$3sY1Dy%rPA<aRKz6pB?VwUUxA|P=NX|?F`=FWH`m{)Q43PZ&E2&D8%4AZa|
zY9rM=SR2pI)G^r~-pAuRiPBU>Q+n^UKj+cIs!@fi7W>clRSzgUF(c$6p-))jJdLkt
zH2L0M2E^0tG}ibm=T&Ag4DXCJik<rWhjf72&f5CZo7aeL{QIwd;moX72QWE_2X9jZ
zb|RUU5+x_DG`7}|HBoY5&Swl^u?-htD&(n<Nu=uvYRvJ371!5XwkxeSA_=V<3-P@q
z-cc>3LzgZ5{u92AG(^!gCMU_^o+X&lp1MJM0%@<91LAD3jB<7UVYBEe>i3>SMzwn{
z2knCK_qS(ME3b<{HClBRaw&znZC+*+d)Z|>mDb=?jFQFKv$%HNl;A#lXEP}os#`9J
zApb@Rt9TP0NF;TAJ#gs+-&PfC%C;&lnBPl}b^xm<ijIw2x~=TGg@Q{;_<?W)2@a!!
zqJ1v^-2PN&k=7se;-AN8A$=gpOEs%D_NF}UJ2vqSd)L*FBFvKM6s%rtu7ryiZ06oA
z4+{WGz-*(5VaY)V7iaH<-vOKu<?+&!8DX1DSX5b>G81KI)>{j?|0e33#8HxgMJZ!*
zc9d3f`04I`gpJPAC^0MomV}fu4QFC+o3lXsDsfHAkZc#XC%)zaauV;ND;;#U!)iDz
zv)yLszWXH-aIsF4?RLFuO!oIm$<L)6?9EMAlI7uVcE3<WnCCKpO0h(i()0T9e|@&p
zG1nY5^#)cJ@jwA?bK<MpTQ-HIs>dGcdVnR8W`tguNN8wjBq{@dYApb0B+mZcSgAjp
zu&D!tj_L*reaiv3$?!!w1GmrZd*Ia%Y|H)mNW?JoRs|pag-_!zYJ38)K8>GjRNeFI
zJ!dP~MY3Cp%6WBh@+YmF3~nMsNn%bdj%*`v^pMA&dnI+Eig8IcGAG@S=}elJr=T%9
z$FpkV@w)K4KYfSZ-}3n{KIfbg6Unt%I=Epj8r-bKVeJNV*)~KPCfUW3wUJt))Wppx
zGtCoAEAAKu#kOvhWm|b#7YD&~*Sb<w)Wl0_$S!t|hJdw;OlHZ^JZR52$l<F;aov=-
zYa4;Hc*a*4xK{kW8}gAMAz_k1&ZuS9x^kj=YBu49Rb3$StH1gU?ctH~^*5B}WqQ{h
zNsiFkt&x|bfk;BNqPmgN<j*?^pWoi|-~Z}2eDieW?ZZOy>L||AP-|S4EwX+oAhgL#
zr!cHVCn|#7*I6)OG%+us>Be)1ll=E{PM%$F-FEit|A4dnavY594lZ<0sJoa6vl`kC
zKmSp+JGwal5%n%!B{RBGn(IeB=*GG4<7fqMNf5CJnEX9EbE~hi!B7CIxV8(@vv<2A
zc>LYtd-bB=NfOQAVc95>xPy_sHxE09_El9zHo0Z>GR3xHEwSP3fjT8B!eVechts*`
zA(V^3S}SW?d8!Mw?d8dSi2=Y>U+Ys&L-R_46;`qX-KJe_Yexlapsc?B+=+FE*cn~;
z9v8gS<`95*C4y`2ynA}&cDiL<8`_K#d%VehKID?gWbSTHUZPtoS)JM83>+Q6osR&l
z5~LzTEhZ_?ph|5kznEsey1nC%>xFj@4~RA9T==W|J1+Ig!<%nNQz7LEG^3VA7}kNZ
zlp?S&BPWOPB!1=@VZWZ<@k(P3Q!u$S<zV9mn)f}7sbM2V#M9U4VZn`WW6k$6r;lXR
z7TqtC?-T2KaeeKC9ciXK*y?*(I@`}iNg6Nq;PGfoVNcXNAYS5(7}&S%xYYm@iRWkd
zxx6WV{Psf9f~FFI=C&hd>Oru!N^1?XN=*iHqG@B*M%or?t*nA4U2<nrhptu<wB`VA
z3#Q6-TY1_R9_@i;+s2wui}Mc)Fg!D7Jna=F`7Y!fcO%_zQG_V%3t_z!z;QC)!z#Ou
z<Z%*JqSPrYQ<X*)g?UEGgsxXmqojf%&V0uGpLw3q<^WL%+n<AJfn6~gM?rUO(PcF$
zk{xUW!IHAE@^E>}XY+}_`_13*KYjgMo*tf<rZZ=O&vNEM;+qZ-2QU@8l5X$!TT7%Q
zl;q$y(N(dv`piW$T$j*dUw6jo5w!o{kQWi}_XWTQKhgazJuJ+h_o!`WGWwp1o|Z|%
zJ^<nWtg6hNiR{P?=>vzB9lm^ST(&~ZLTZ9o*T>&DJ_{M2S<NsSOWpX#-~FD4yBq#@
zUwskJKG?*P$d(W(!K{Z?f+Z<RZ6RJxrHOjUwn5X4l24qbnN+jeoo)lFvRUPEdt#{@
zqv$<=Xis|Oh%u7;B?)?Wfx#W@=S<wQ-<|Osii(xk;}GkNNJ7&Jj>*vO-!xTfYqX{`
z^@cDKb-c2F9A?IHUBZnIo2N#LGE5sQAfB}<DThBH2G+>Q&!LSKO5qn@e!=DO!rRyH
zn6)vR@mb1DlDTYN7@E5*32DSvhn3;;j5pxQDUHNZuXs>9D$+jRdR}=FOs)<@{}=}I
z4`R#v?>i9N{`?v7Y5YZuPXN}Z@skZTduC)k3!B5yi$Vi+hj67)aF$0fiCup%oGbw^
zlk2HYd2TwpXmr=lNbx!`7!8*`5WJW=fa3zCjkgaM?n`7jM7(592mA)ojm#lZ6r;8%
zgQ+#*x=OYL&9gkxsa}?_=7FF_xzpNe2+j6TQXuHgZkJPt?b1UtylgJYX+%*CQAc6S
z-_h`NgOV(bx$h41o*mw^Bl|3QG-Bdza5c}gO4<c0L|pLp_V%9BS_w%TlF2AN8NR>~
zT$B*EHbt$Gy3>&0JfHZ*{byXSjcdJ<&B3=*a^OLl2UDIk$`~2ahOoB{h)-hh?gYP{
z2;90p))B_?*F_g7`W*hq@gXBC$8UD`yMOk_F#0|wW|sgNlL#q?i&H(wTP^l6QRWj1
zZ3MvZ3<r2NJrEhp4q)|IB;3k&nBEt2*IQwGSuR}Gjoa*{z}Xm%<PLfFz{Nky#IwG&
zMcJvX4w|Gw#moJAsg5sN3Bgp?uBxrEX`_uj+%MANb9}zMu|?dlIdGT9z4zZbOweIn
z#)}+~ODO>8*tgc_AwIX$&f-3=%gS|GoSmkO(&Hl-(b*8wRGc|xeINJtJVrf6XPHUr
z=WPI5LkFjE%9&rB?|8djdDu4Z@LGh=&L`5+T%4zPiFR*<@jO6s;4ckO;Q2b*yOmj=
zQpPD7(S-~E=KF^o{{MaZ_`U6FpdS@{g$F-CUpZ=B*xuz|41q{It=giaYudGr{jd9J
z_5qtb@PhzgsxUz35%TUaAboO2(5-+Nk6Y#GT2a!_o!M=gziDUdRPz#i$%1Xpd~_`r
zMMLwhHxnqy!SR#=_>J|U2AjGb=TqCk=nfDctXSCI2=dGUEW#f9xfOTQg?GjtvH`$`
zVn8=}hrqZ`bYIIpI-q6AOAmV~&rsp%i6Vg#OcWvZHNDGU2Z#HEWBjf&Cwy-DDXHQ~
zwPm|9pKti}-7Eg_yFyl_m^0vRvnTodx$Zt39qhLK@{I1E?d+?;x(Y{4BFG&l>7g7*
zj#R&w<HrQT9~c7yI3(JB?xO{(6m~n-_g)Hv(xTjV=*{)IP}}#fzd`)j_ic*5bVl4L
z)epY#3vJ8&o&1e-d}~xyzIppV)`tH23&tnMvy$j4oh8vi27tI$2Q|HbHg!N%JW-bN
z<aRJ8x5u&0G)m>S-iaInnD1wd&qbCFrV#8=XG9J*crOPW%k9OwIP5i#2c8{hbht~#
zOz(a{2e7m^5Q@FN3)$}vrfyx2Dn4#7hxf$R{_gHwrU1gaPNx?fD3XP{+Z$LbSG=sS
z7~EzDuxeYQV|<)OMwCa!-1A~|_GHX~_C4HLRB{w*_b=b?z2x=3#a@%=fgrm-|7rZ?
zk52&Br}2{wt2z#UcYFo0hM-K-1Om&p?Z!@AmuAK&0<-Qd%AW<U#{;VeZYe+kb;^56
zaE_qpsnyUWfBzSIoYzP!4gU1{fmi1<>3k-sfV$SWYc&Ok6nZN;#1Iq8wr*_AxU8PF
zN>kz_1xbnPs;D$JF>3a*r)tL1l&fw$S!GdWJqEu4R#o#d$ATca1kKxpCfy}fTr0@4
zTHT2e>xM)->@e;-VczMm#{>~XbO%$$Kc-QZxatk7Wf}{b7D!7(w+T_NyMO!m#O1PL
z70=?PN-q)Wt$L?mH>o9Za&51!-dk{|tXk&w{*J%;)o++M@x^?{e|mV$tQ-IC7hkc~
z4Qq|JdSPpq$ewx@7wOdZ=T@y3$?O1D2MoJ7Y|MOe2SJYp_7IT$+eQ~G?#{-?F|=y<
zYmo))fetXg*vE!aQ(`GFMakT(GMI%Uj)P}@Z)a#UIqc&CrNAjPu~u!)j6=d9wecQJ
zZRp*-TNeK3|NLA2yU$+nS6}_w9Z4eszf4XNi$9M{rR4-xuqLzy%d#PoSex(HCPtEl
zrov|GOv>inQ<rvSqtSB0dW~RcTC~<k7TIrc_NJ?>xh|BA#bq3dq=9g(N1Mc6ZrA9H
zj=`Yp^NQ@C()jW~|J}Ps?(;23BBjifr_kBjpYys^iiij9&5#O7D#;dlXN{69*mskI
z1J)>WCUrKRRZ^b4gj<EXn>+s9mw$yx;lAARpPs%Yllc!{{ViWVyy5SE|6jaxxDnY?
zB@OMNXJJ-9k^H{_T(!XIda%U^csmQz^dSHIZ{2i#a=+Ig{k(_PD6a^uNMCZ!P@U!(
zlCVB*0b(_(WUd8DgRC$|qZ5uc<;4{Az1?Dq()V)6*fa|}Y6SB6>^FTU7)U8-DLghP
zCfud$T5fHj<uf9gsZ0p4SjCc&=i<Wh3Tthk%1YPlxl*PR(=5({O4+-6RYU59#b5y*
zw~g<f9$B<`MSdi(swt8ZIZp$0f?f&;X5OKAfL7A^ADwk;1eJ>(8K($f<#6vEobzvY
zP^HL;A~!5s<+^UzYS;#%0yH%3EMl#A$zkZ)6%txkv~A(P2q`sZDMmvDRYfL&36K>`
zXabwT(+XDM-e$f^_ta@aNt`+HH?Mxdcgq9+_)tQ%u<t$D0fL@<3&HK2vina-TpGaw
zBI%%H_zfcz?WIP|0k#j?<YBLW+<xq=m>okko*jrk<I$QYcr1nAqYgG?gwy#12_D~F
z@Sd4aMW~s*FMxEwQ4ePsV?KvJrJhuBom!YqiPM~@)meMjAV0Okx#%pZ0F5yI`1;Wq
zU(spW*GACWL*k+{ur?`vUs<ra(X7!{XF_VIjpqWQO4iCIO3OkfgBshqvTCCWQ3lJt
z9y*$nTrj<tzlpe}tIHa?Ttwukg4x|2<*qr^VVUr+gOTou0ROWQz@NzY{eIlOeR$+X
z-mv_#d2^ODA-S;gzk;o<#>3l3KTDgEEK_j*gQzMH@puO$y07BhS2IX4Qu2cBGzs(R
z#226alKXV#b2;(!2e9oefA`rhIhDfi-@N8cyYQwh{S?`KiBTrV*k{8Q6P~TM;ji?U
zb=@oP-m0WjFq}p0jg$TxY=+1J^!ZSGc`$q$f2rdWfc0toWMj0w9f7KmbEc&1osACQ
zl$R)Vv#^JB_Fh)v{o%iFBU>U#$RRe~jS3u`o|nJuIF;*uRM(2J9wdukQ3Byzu51{o
zp*Rs*3}(hwtAB12EIB<n7caSU;qzd887};6jf#3^m81W3U}MQ4c#~n?Mmf(AliVSK
zF@wTR-yeSUiwW&Ua|FGEtwLl`jAh+I92<~^$u{1@@T}71dUZj2fBMi%&5}?OWD%y4
zhQpI^jiY`&68CR-fTPvO=GqN+(+y%XcjcCBSMMg+vu1t;yQA{Lc8%Oo7hs};0>`-y
z7~lh0*dJ{QU#zV`M4g3FT!9v4ArJ_`B@o<QgA**cyE_DTcXx;24#9)FySuwK?$B7{
zG}CY9y_!$xs;+zQIcM*+$btHAM#BOnNC{zX*~r)9G;1I;lvwu%Us=G!h_8>wT=Z_m
z%rNkA92+pA6lDRs8UpVleqm4I!r(*PQN6hlPMs@H9Y-=cEguE-{3qqJ`<dS9#%j5N
z9463brRYbR^z>QKtt0&s1N$DHTntLez4v!7rZSj+-82u=y1oS{4V2-)JX6{#@6h|(
z_PSDoNgy=b08+E9=~k!pyb`dCHyuq|0^?8=hQOuVzkiLM$MwgFUc&%y6bQAYPxyy!
z?Srb>)iHkaGPfSlD66B6UF(=ey~o;)(yJMZ(Bnt!hx=A+UDD69ZNaTL=P%x=j|M$}
zT;M)x=aR$$eZ@MO)YVMI!G$Tc<xIW51M`S}-Sp<GP*&%8W&m@WGQkoCCi6M(8y{%)
zy0`fUdS>53yVF(72((RW4jQi5X*k#YNdw2P6fqHxb%+4b;_KY;UJZvM_Mz1{B-9dH
z+>ArH?I$?>EAsB7mCIPz>-2mM92c;HatvO*<MlYSws~~_nbgi)Lqx%RKAcFeVu2{V
zGKk+iwYnTz(u7&Q$G;@h?C{CS2|$?=qWyqTr2lEm&-Z9SGIh}N=S_5nhAJCTT5RH1
z>I~hi`FX#&nFm`yDYFj>;vWjAif(sC_Cz=F<hb?9X}otOohx{oIKrNUQJ(L_h<+T?
zP2|x-?+oUv#MGmM6G|OK7gxa-%y{RO8{5I}%XS7g9NZ}(Wh*RK)+?yK%s(Dhyw`{-
zAymEMwLn}z9*;hFohOL4*YG7%T@hrrqehn)LMA*uZiDxf)_|^d4TrQW9>3+MuH`d+
z6+g-**bOx25Er3P_yYoLuM`k~sb;YEr#}gTho2lLPur_?Qz>;bm_~^d|NO#9<KlzH
z))@-?hoWetM7twpbf=7znBFk)qK<_O7^|tQUczUV#e7_dJolNXcn+h*k$+D3on=l0
zl}O51ZOd}BTGRA_8|g%5DUc4hpa%8H5^HS1uRQcv4T?U7Y?a>h#QryGqx5>3Lmp_b
zo0bQ-I|HW*`aLpkG(Y!5&pM&)>X<~bq+a}0#^D0N0014YT3eFrZ?n^D9pJ5Ov*zZx
z#)SHzH7ggqrigBCi!S5cf??%YQx$n-;g7MLigVMx#q^1Tuz#LiPYjqU7@0Jv=a@QG
z3={nn84XwhAN{ReTo>#@PGh$z;~lm}b>f%vu>IR}<@j@Zao)<=hkr2EvzGyxh$1Ot
zz=F2Sc=I1kT@FQ&7d46X_>zPLjbJ1UrjYI|JubzK=9_Rqu*NpO@5poZ*aF-s;r8L6
z^)2MT1Oz{ED9|{4nm{%^bdjDknO^IQ@Vvc>Yob-xDLUI*7glRr8a!V?eH##b!?bk1
zS?PIyv;XJ0*KkjHqGPi}0&Pm6_D4K%gQZfVljDY{LZ&i<N!+r17pVb580e~~-QJMe
zfN0i|hMS}Rg69A3$71#PcSI+K<b5BP<9oPX($}3kjha?}*NMFD*VJ>o*YALl(CgGJ
z82y#S&>+arab+BM0(EE}28nm<KA4$p;5Ff%FB2;HD*->j@gkNT4KBq9cKeOgC~z5e
zu@%=f9eGoH52iE+P%j0~QwXO9{nlNfwqiS5vS_bA{g(cRPi23N>+UKTr^D&T;hc73
z2~{tU<`RumF3Ea3-TEi#o?(mYA1$E@7`;Sa;&8OAdsdh`5Pl0WivLQt<4E(q%90ng
zgkVQ5QC@2{ljgXEw2kr)x-m|i{+Y;*SB<|ZknZgKpy9)vhHQp8)GpP`LH4M#c$Fw_
zh#=Hh_pwu@1M&embWdl;{uP8X9_@a`yN1lBA%B?4K~F!-)aB*!i@Y;(is8M_y|+T`
zjSu*#ar@}a6o-t*hyV3)KmZJO+#CVseI{P`*L6>^D}3L3oVVrd?>@j~uis#!Q%f0|
zIrvylTochxHiKJ4Tt7CI3p}LQM%uY@`@OaJ=yd10Q9y4$HkK8Hl>EgiXhi!t8~~2A
z6}Ve^W|A$a9X3B6w!kp@*7`|DJX5hnykdL<ByF3bPMjf+4^eEkxKAc>13HCOJ-nY&
zOKtZ%K>Zz{@nZ4k9Vp|W2?q+&Hjm?<$b_<Fbt!`z8!+pc?!!}cJl9=FQB(*{sa`cT
z`nQF@W<GW(!;}Y$ugPh|Vg!Bf!Y)g0&Q*r#C_o|V-&TI8RQbGvxVnimhuT(WkB}D1
zJS^wRCX13sOM{dYvwr)#Zk)cjYhfmt3UqWG*ZALA``u(2)LQHdg5UiQowO2Ugf$f9
zG#_4NX@gT7o6DMJ3y?@KX4hFP8?BVSUU=i8tLg-xZ;u_b=aGyf*YWeL$NgP(WR9oM
zeGC*}Y0ps*o>tUZKcw<IDg?#Z1!5pLc;~k9PwBc8FJpIyaq(Du(NPSLUVH)-*^gMg
zIFKy$)M53ewR*FsO-?<BEYq7U38i>=+S{n=I7=Zza6TbRAtPTY5g&^cTf&y5Q-lH6
zzlH1fF4m7?;k8wm&c)RZmAQ0R`5fJ|Xcq`MF0`uX<x;FSG%Fob#>048mwc<atyK%K
zl$`?p#n`le_O@US4$&_snF~_;$qtwZQ`x4TjktGO6?5#89UJTO4&g$LFXBuX7e$YT
zm2Y&DLKqYN%HgX5{~1%9HuA2lRk97QnA`CA%uQVRwv8RcM(cSfRxdp-S~vONa47ik
z?!SAdvO94%3GfBW_&I-AcXxguaqOQ1i88KX`PTmEN@b(F`f4q4+T74ci5EnBAu2Y&
zfcVd@*=mCQwp$~2ACnDt{Ooyl<5_C9sN$zf<{VBJo+gO@FU+hIAA2O#lXeaZ-VXc5
zxp~cl(8J*jgAv<h`}lW2uwg@K!tjZ>20MYXn#J4;Khn+LI{Je|TCIOe9abI)0mMz#
zD!ELE$6Bq7rpXE*2I10wCCLeBFgS>gCr6KmTX6gvz6*E_bgtGY%VqUe`{gNO_|bM(
ze+bAcC4oo+{i9yMFeek<0)Yu*U1tZokb8-_u>svW|7mZ|y+0YBRxq9)reILdLz`&T
z@(J6spG|paigD);kXo}b8a!9}dD>t#ueK_@x->i#3Zxgl_O-U>*gx$x%#<l;oV7ce
zv!%r7^{J1tCKQv}yyx#1G~U|aZOtByD(?r{&fGt=MS3nKKwlj-!y8@|lo1`(nS<R;
z&JG!F3D)ZeWG<c?z-{p`(Uj3F)cGcW5wF#e9F>DtZfSOyw1elxq4P(7;MzKhh3jQ;
z(Sr?f5`ki4z8ZR^=tje!Rc(11U>zFi>@z3hLmNzbMzYgM(@+L_4t*A0mRcq!XOre5
zQM&u^k=byut*-Xre6t?c@fvb`<iWEDm%zPp0*>G@)M`3e&Nx8GY(opdxBGMC{7_Cm
zKo)|uQ45|2pjpD4*q#N{4+Z~8Ji(~aq7pTnW}+|6f`W#Ek`fhC&D~8S5NeHV@KPPf
zc(k7WqLSH~W8d2I5%Y3|+!*_M`*mJ)8KKTtsF<)>2|!>As|n6l@#)i7es+rcG&q8G
zt1r#q?RnTR9UM-5>jO_*bNqRS;BS#6>`&qe0Ra@(fFhYuic>e45w<^u0CiieyOo~g
zx1U#sfw;q+pdz8T8k0CXt>~+Qu@A;$NaI55JgU^lhmb4O)w<K~HO^og@e~|QqjGa)
z!YHtSrp!ddExs~AR&B19Rd088tb|BXjn`mOB)e@P9!obYZd?{^Z4QE+H*(qN0sh84
zw=@qJ)07;69iRTQ4Vd_{QX2i0Fl&imRm;~7Qp(b4W1_dNmVKml8Y|kA2{&OOZ5bjJ
zfQBDxNIREGokb1e;q^#$a&nL|eZts=5=vLz%l?{|rejx<)R5Z}W-DIJ)?liiV{jA4
zXj9oD<A5!N*Q^#7gR|xvo!_x^$+8!1!@~2%VfBi8(t`ECCtc>}!2F2t_GjjM%@wk6
z7~H}L%jAkawgIrT3?k$5@X`F|`WXK6x)hoY<R4dfxyHS2!v9u?=FR<<i`8o#H39*S
z#Pjg}TLn!*DgS2MkGJ0@D`o5HY$Mz@1)O_ZF&LL`%eB%qSwU?b>3#m3Jjw1&eOATM
zYzQo)jaIMLZaW!Yr;|X0j%jkCp##ixbcg3`^ZT6HC3zE_KIwPNr*y42jL*!Zg-7IZ
zRnh7!bv1Cmu3sJI3;;9)$tIK6IVo^W<+CBIB4OIsJN9q!__E>)UjoQ5OIaQL1)Oug
zQsT(#kZJX+Go$mTWN4K__4?#)=B+ldsz3#4vF5Zb%ap(?uF8CIkH)8rcr7?T1kVed
z;#kjeL(MazFYmhyzX50Jdd4Ig<01TM(EsE5?4ohOs%4Z8zGtViMc~@1tS{%VcRuLG
zmJef(_@PjD<S8&un82aT#>NENFj?iAh@?r|ibqkzQ}e%H^U1NbaG4Y3F;#9fo5xZ*
zuO-`W^Eq_Sbhh(f;`%!>631R#BvUV!udYi%E*t8zQzW7}IM$oX<D1E)Q(_p5_|_X9
zIylP^Bs1r-w!tvm)!=@Wr`#a!*RNgSZLl82P51~n9csTS2~m6r;dV2EIX|Jy{2}l>
z&^y5WvFv)7=l94F-E|l^z8Qk)%)<Qe0$xVS!Z^)NFYV7ZmnBT(7So^-6LD~9y)H70
zt~wv52-R3X?s?wRctiaF`|otvcOUKUCNLA5ry`$ie6}5?1t^Q6>!DfNHHs+YkMz(|
zMyLb$Pu!=nPXa!Dx8<HKpTlJ<-s(Ue*eDo~_G(7sNzNxkFU!sMHgf40p+O=&h&OL_
z=EB)sO4f51N@=DA)y97Y80)A2{Y?Mx_EK*Bc;3-@Wg7^rsQgor*b>F~t!kW|5{I?g
zyEbWP!jL_a#v{gt;cFIUCYGokgKpAavdZe-0LhzbezPLiU|W%vSap0gd>xh4vw}M0
zRlG`M?8!B<Bf>9$%KzXY!a?H&=h0cW`S<e`D9KlG-Ycf~*IQ6J7k^E0fNR0L1sB30
zq}6sIFt$pGhc7;O+G<s5)jFnSDOS??iSyWpaQ*cRv)WLa2t1}oXdd6pL@}J#_xw78
z^UD(^XRbtV;ZVlXDV%u~e4fu_{O*<v?yen-e7z=W<P8!^@u~O|bZ%~Eck73Ll8#i`
zdZhguoh{v#LadE)=fCg&9{=1WV#$%lm`mjjAr_?+j3Pn>5$>GD+q^Y}Yxu<9`Pg9I
zI9{3{{qvj5!7y38Tq?J*hBYaJw;hI{>vXc&Bg&s)yq^R5pbC4vKSkjUg_h0ANqv@H
zJ@6dg+}cj44%1%=)Kd+W*?4*<1`pQI%KE!UGvQ=<@qqE<ut-$o&DWRzm==;`H(akK
zyH@F@4t?H4ul#C5;#rby=jo%F^CKrl3{PQ$+*-!~_5qw`rcYGS%q4U4?qe9736x>3
zh@H>!6B&6%P+6j0PrL*z46wg&;Znlr=>3Wv8KJ(qMdso%pS~2$VDprM7Hu!tchYi$
zVnpzqPEPK8i^5N&II4#VO0@ji)X=;{Gu3BGi<u{94!*iGxEZ?SXWxDCI@}rvy=QSI
z5A6XD88irdd2ZP&UbMNn$eG%Lwn)Pqg@IiQewel)4fwV44a2L=f&$i<Ul`S=kp2ul
zXbJreg=$%bX8-wI7wg?(Krk`Cn$exDuVMr9v8Cnh283U|LW4sLmD=!$(TDGxAthH-
zKyr={Q@(KQSAa>^kLV(-Eu*Ettgk%ON33E6CI;t;)YVH*dEnC)8-uqX|Bsf(2wdDk
zIJB+Vr%Fg%Vq-|$9bh$HG7j<rv2IhcO||^{l2+xo)_l#I><$`-T)?~Z+H5waXYmQS
zjI80tETkl8KTy7Pl<YMec@Pr#!Zm*{xm1*w4bc2LH`^Pd^m1ms9_Fu_<*3JL<)ZG2
zUFyCon`pUJBGyol+-c*yMn|gvsWiSW3?^el9{ttDdBEu`lZPoLqp-Ur%*(NV?lAB_
z#ItBJ!Vyx}tjWD?%d&4X>dQ@D#-o!~juj{L1DyWVq9OF)>hY@qe|cZ6`^WbqPX#W}
zj3c0~n_Iu~*pc(T=iwNmd*%_T=MrjI8N9vXi|4QP-@O%NLj*WwclM|AapbKlyS={u
zwV?wu9}O`d?s_RAyQg{^Zm@}v@8gRvd&(P`U>4DmItQubtIc$l>=DjsH$c0kBxW21
z@k@O{@VX2->|bcVnel(gWG-PlZD5{tb=G#a07sZiGW_(SWW`)Aw;mCd+AN#=N116B
z)4$BHN;0g>RqcA{*imbPzhY|Yy)3LPK3gR1LL4Q+2Ss%qbK&tm3Ulck`a^UKUgo@a
z{Lg@WJgUr${~qo_%%bR6MwXM=k?8zdU}NxID;#2lTE3KPYV<dj{Ll0S`4(R|-Fqjm
zr%g*6((tzv+ZZw3kMX-@BIQyRS*fY~ZlX1^QzHTXukg%0iX=yBaO4BQc!^#<vqO?f
ziH!-QpB^feHZR#k7Y$hId-y*;Vn`G-6Z7~G`aMmGyTb>^n}1>I_jiVT=CIZrMV6<8
zmtH;#ph#5iZiTm_3`zR=8xk8EmjRs<Awj*P5P4y}&OwK9Qhlaxv>AzJs`PRdFofI-
zU+y@(xA6RZ?oydc`;QkyDaFj@b0jGUnB+8NBquENEOnw$<b}tkckY(w*Cz5NU2B$?
zTkQ}b-=`4D@tJs%K9GRR0;O_&_9H^DYDNOkLHjdgY;_N89^$n=!{^;B8jr_g3~+yJ
zL3BxnUs%EebbC(>fX;-F_4}7*q7K`=cD`?AG^W8(K(F}_XQ@zE6wkumI~67)86oa;
zDl6|sFYxC<kc!^U!5MW-5OT?0wq2`WPFtH7ZE6t;^w?dSF;Efi2^YI&2`ZnL#*1O%
zk|5B+94Bz38&S>`9_J|+Qd9Art*J9VpX&JQCn{tQk9m~Xnt}P~JY0F};dyV4A3b`k
znq#^eg9#J71{!)V2Cf}|GrQ%-5Ez-rx*~yxeBaC+s9J{jdHs`f>T7{3R=P`?wP+u?
zV_m*cPv#uo7%+IXahi{iP*jQ|(v+r7Md{c)f7Vr|yj&%ePK>K_5v957a>4(7^z``Q
z-nR!bU4=GTDMoiC;fVa32*=U^GsF4a7Nd9x1@0#lgA!ilKmc0L*6(;~R^jHsfe!a<
z)4iJ)8K(38`P_e*ikQqpDtXR4d{}CzlWG{Wp3gqk-K+Sn7l(Wgw|kg6m<!p@hZP1l
zhwo`VZr&sI{;v|<&ky)l6ZYIwT%*louH?~wC=;A-_xmIGQ-^iOxRV|@7HvfU=T@x0
z{NOvk6MlXD{!%W;u1fgp`^?TY!|qA@V5N6qY`jrKYx4=8sh!#GRkDTc{?>qw`Met}
zWsE4BvF*67O?LRmg2F}H#zNl2b=45BPRXF%zGFrxL!`<WbnD4A$7@>`2aI)RcQs!#
z+(Jxky8*Vgh8=Zm$@bhlJL-EyRU1>)hQQ;cYM1hF+i~rImPLrzqIEg;yglq*$iSNS
zn!PggUf!FK+ZAncN)hkMO^_{*k#yfM>!*y9f#R#r`qN|~Jk)r7CHuy|Dpt#=`NuA=
z0zqLl_<Ci|uUo+p!*YG&xBTS1)5a;Hp~+)=dq4;zmD&FWe;qV_d)=E!K=8VEgpFCO
zpGR}V;2m2-lqc0w@N0<jA(E54bhfLJ{PvWyo!TlV<Fj7FtZ#p??MLrRm|f&>4^i%}
z<7$Io(zECPKPY0NFgJAY=6JYz@7=v(`g`5|%r+%FP8cFxV`z_zG`9!Am~V`p4GL||
zSVONQawBB3%Hqo9Ze=)s?Gbf{vX+)rL70j_DfwR1zo7nn30ZbM_PlYFnb^28w?&uA
z&}rH-Sye}xeA1q?vbbl;oj_=kZ9r(4Tt{NGgH6?Lv%`<yfPnbhD$o5WdeBG~!_oxB
ztxb4gJ@CScWrXoN7CPZEjcTV*GRR&~0a+@j$|xu$A$MiMQlZW~Y>_aU7(u)k(+6fA
zQsQjd>$s*tic26rD;H|KGBW5&(nRhYR_duu8x+H>NotCd<Fap5JU2^e28p?bUOB=a
zgWt6-G!epd<fL$gC3}KB==7dBgV5dN()_A^@a~=v5~cugVK(=?6Uv~yUC-a&ncvrt
z(P?-b{n?y(ajuRd?TLv!I`#Pncph$0L^bY8=USG%fu2zMxgSu>o*9^vk(>w|p0m8+
znjThrhG#x|^)xySyC&9^&L~jJQ#g4y(RavN?r<fq({tJR2#q0&A$HOg$gvnJwA7H^
z*bE7Zq3xL|qQV9w+2gUvRM}K#wP8*2N^}O!j;dMs^3P}ecTJd|iWk%6bR5dXMaXtz
zzW*fA)kB)M!)9Cz47#K~+0#~sJD)x?3GFz^<86+qRya39Ny41%ZmSo~LO*&H-g|~B
zt)$5j*Do|%SzH0ZEl2<M-6Iz`m1WbzB|gWrnaw;iyYYkK?LG%k!?TBk;}yzss6w{+
z=mv`6kN{Clw0oVV54onAp9gU~{M~!NB4WR~?&S!qVH?`VX>uFyH-U~Ek9UvVn|ATi
z?QW}0Itk|B1e%u}@9w8Qqcsr>8w~h@26w#bPh7>V>QYXB3WALS1z|c+rhYr9?OTn<
zQk3wiG`9AT?AiR1Nw7d`WLj)$S1-a7gDr9x=PI(KlA}(os#}uCO--q??d<3_T}n|~
zdG8yJAv8zb;4LQ-Qe<}L5SC(Hl^DUuY-I-3<Ii=8{&fGE#vc#8rK-ut0W{2HynDKi
zOOoZ&r%!V=UO=2;E$7jcV!nvw2`Cn}81vk#2)-Z3MnVCZ(IX80CR(4Q$S~Och+@9d
zcl7rjaO%3gbO+pK0p7gzw>HULgWI1$kuvbyW@u0~UkZrse{w$%AgzilZJEUc)#u~1
zzC5Q<P*A;?5Nd2WI-{5$Nuc8Tjm`K+w8{=#dhR`BxAlza65%kdAoD}b32d@z@Q*Td
z_8K1?KKlOTjlj%XIqu0+`ca}gO3QZ4)}&rNfsu$8)8Hi{f>(62h(({!inbw?XNbqe
zn5!5yOe9C&#Mvy_$PKl(UStho?QHT|vsxObOdMFT6_ff?5POmy>8o8iGjMa9LyIf?
z)R8=hl8fXp<F$V9RKDd`DgeJaYm)pVJE@tLVm1AwzR_3Fm3Qk$q@7Kw{0rhVL?0`%
z)%k%zm&<@1Dk^3Yd=Q}egF6kg=WUOk3Ap_?Z3#2*lfkBM7{y`7IdJql<#QXI$kxRU
zsf^&G+ryHdg)1*3O!k?S5i{=QIaZwWQ{N5JUxB9|%*1;WE58EbiAMy$>|opf9SXa<
zj@fbSdO8bF`}2qZx2=peXD&vivKMjcGo0M}qn@w=Hc+hOnsIk@J6~YKFwK6CCXcF^
zDVK7R_w(W4#H5T_gEUr#dIvM4`&p(nLciq~i;lC6;T7|oORt=v7^zf#tE8-~roi-b
zbO*xK(IrwVx9wV`9G>=bfgmAyg-Zx~Mghd?lH%39LxX&1?gT%bsPboIx2n)i2(D^n
zHHiY8Nwj*JpW?$sg149s&Mj1s`3a@*JKd)8l;eOt6+cTC%~n#`UEO;P-S4jny0opP
zE4up3LzB^-XyS2_v$(Rj3+3Esy(90;+t&N>W?IJi{@G7zXESNGjK~njOxB*4TNxt#
zn(J#uGq%i*bY9KOFu8A$SK8?Ryhp_OYNJ3$?LWo8hUJ9y`bT4Y4yG}9$t7jRK7=2c
z%K$?cottxCbw6cWN^fq?XzF;X5VvWA%s2){*33s9ck{_gRIXnM2tFM$U2XH`1lEWy
z!O<oh$!1WkOd<5%vs14)q*FJyliGM#((F%Bnr<^{fWePF$9{A^ptEmdfB&(RrOEO)
z(W_&g9Ap?UVzHL8Ltf{MR^qnwsfe68Xk#{V&sR5rGFy8LI&ER!U+)}UdJ0^y1Nh3d
zDO_huden#YQ?iBag&*Otl;B9QKh2OvZ3<<wryBYEO3Z(y=22OC8dw21SjJO~86v1a
z8cCSd@1gFv3{vSu3EF@Wp^3C<tkWr$Yp3MRHHv<U;kGY%dDLd;%QD4UL8pV}Ee1~h
z;FyUGcbhas$Yh|{ajl7?J-%A!vw{&~(TZkRzl@G`uf-W@BH^@ELqTe0``WhXnL2-7
z-fho~*BtzZIrmWDcbp)r#Pi!4;a&3c+Z|g0ef*{HlpS;NxdEY0*s#k5iL3du0AQjP
zcsNUpl7$T=AH;6z<i_HYNu@kZ^o7G><7*$(85BAc*#X~OBPVz_-r$bQ{n5wOdBEpe
zQD31~Z?|2ndZ;!bL&oUl)~3@>{NpLq@@%v9?A4ON{lh@6iIh?_j~cfd!gbuy#KOMO
z2shk5U7XRgN{bDxFH$I>OiFv{zXz;W_y4%zzZizHyO_AZwhx*kQVpkOZb+k7ueWL!
zo*wk3@C5Lg7v2*w_-=6(f0>jDB5Y8y1{j0!F1N$ex$M!%;s5FvDHjD*+f1a4aAlaP
zu*<z2O#rT+r|7{R?;rJg#I=V}^A^QMQym$Hx_p0Aa#hcMp+hsc@Cytd@Oal)*AA8Q
zvXmu;5bbGF@Ux)sl6O5h<b@NjPx!for2-nJlLP^dcW0?>Ci=c^a9bXg8k~3>BY227
zjuaD<xs(wcHdFbIhXNvJ240#IL1}2I$?4Mhva_PWQW;V_!JJmkoOvIKzy!PZ7CJs%
zdVi1CxEYlrwecc2+7L=LJ(w0wL|bj>jDo+qtr@IELDO6mV|=<e5UtjKOlZsMnd(F<
z)aA>&-Z9bs&~9;jp`om1x1iL0Q+>Cqb)7K`B_JXNyrS6Vnp_Z-*dnNieHtM@CViN+
z<pQs|QAz*p=yF48@A#_STKTA;4+}^r@|m_@cZXd}SR(5@82TI!gVgi9Wh6T}JJf0^
zi|RC`^jrsgPW&-ONvM3=q!|0U!COLmvvwG#;~ieLq$b_YVyRmCL=I|@4okftRpMk}
zY>B4*#T9ZwaVmUICQLW<sBJHfc9_BJUIrTCAv!=sPLrG)poC1JYyW#FW>Sr-j^0f+
zGcauQudmZ|gg?&Uf-r84`~x5Fx#9uDGv&EsNl(RCwgtQqq=_`NY-yffs&q;{K^}tz
zHlJ1}Iojm+kg8Sj*C%1R%#2Pxvxg^0Z{|*KA5N;zv#T6sn3maW3=e=JOPd`dSwp2a
zk`P%|jUImcO+IuuqYZOZxIs(d!gm2WE`G(v;;!w-LRIAsS<J{ve4P1f$X5SRr|_hz
zTKjqAac@3M*hdk>-jj<CLqTYtmI&WPwl?%Pbd;T|{=54zCM9KamDlWAETmVFBYgz9
zXma!Wwxr2EWz{}%T&=OJpl)!W{fwqujt{v~u5JEeTWW57$*n}baUZDMFT(P;YcykU
zm+*b<l#lIDG_&4%h8_6TWpF>v0U2A!V&&TRLu^Fjl?cVh%p6+_mTwM3MIehVq2n}t
z-~dbbguA`h`mc$zdy5DXEuh}SH(^VJlPOc1#gGyebg8E(Np8!rvnXMx5YDKxG!t3V
z-mhomtEN}j5N^P;q1NYnJ?63|EuWLI2w8<E*qua~_Xrm?=RSibUkly<oA&8?`H9Bh
z`oL|{44byq)`dzoz-SVAcq4?vlH?|OHN+l$2>c2?`LW$(&R4rBL@GFtoE($5Rk&4N
zfhRqkeT3<pv9=8Sh5vYrPt>y15#^>?evg|tAooo4U}tJ8*!xqaJ};Dvbl!wt=zY6z
zt>+`xH|FLzpbSiLKn_z|vM|VQesmYvuD(2_QRA}j3=8+dK9~mw6XS=lEC?~8zf4a`
z*YQz2)B??!Ib_Np*S-?-3(8NEj;S)gEo1G8h@PDUjNIL!#qlu$lo#x*->6t47ueY%
zftKbvCd&+}fmL?8*elg{^6t5B4tuIShA8aUTYdx@<Fn=XJx+?E_{^8F@E6aWuHr$2
zN7qA%z*q;QyF1nF!_7$x(bBou3fX=%XFzP%9Ss$_`A|807no<H3n?*3fs8Y^gDd1l
zbWxqFx>6Oi<>uYJe|3b?Mpq!sDyTu|py31Kw!=GH0pmE(<7p@HJBw}ndBm3BEw9@g
z6>85TI)Q<QFBrI1z`Itzf%dvfusCBOdpD?xVLA@^LpL|g%J{lP4Q&p8R86?7K4$FK
z9gF~prh^55r?ugPF^*<+=z0V3sO-xHuXX+Pv^9B5|55*B%D50AQP=GsFXJCkBw{&P
zsO6-S;55pr2A((HHtKuoR#!sYNCg`-f?FewGeq*Q<X<qSgkHS9QwG2LO?D01S$ZLm
z?SW+?@L9g6{V*`LC%njjP50KrPL@`P&1^TwQ+`Cw&bVZXcfVnhg6KjL<=_^N9+L`n
z3WmC*Fzm;J3e>gEQdZ(?+7dDvl4lE)rI{vGb{~$y7FffXG6|%I;+<Q7UEAoM6v=UZ
z5m%hWX{bb5r4f?Je~2F4{MGB3JIqHeAMNGif9wt9)wMuiC<p&dXgIQI8193(fCN|;
zhSm?@RqfN=Z!Ev)J}*D{iRYQJhnvwmT-%4mvmQd=L1WGoD~QdUmBIpfy1X!t*V;i7
z#ZJE3Wlbw%yqCTWQ|GS@*PwwEfAE0&TZabMW~#6jlJ&6Fl=Xc}1ERpn_*9!i78_)x
zB%3m^kLi|KyMbxo|Js}NGRWKEP!~%_itWd3`lu;0!rvC$#;rnLYN#9H*w3wS6er5W
zG+Tx|7bHna$(6eVYcaY=7>T@uS%4eW@~^q`{8OZ@hv4lL?I;!nM~Nl5f;7)J3rs$F
z*$3>26xAqqf=N$RsVe2NWzA0~5D`t)q<Rh`=1g;hFjG3N!g(keX3g>N`E%sTv$fS4
zjmn;px123z+QZ7G&oj@*%EyAi&9y%lcmx0@KM8UyF2bF==xc?}T$A&v<cQwk)CwRH
zof7?Bg<r+7LV=!6ZGYL#zF+&*xa%q*tkaxD5DUZoJszx1tgyo?GjKNC0$`Zi!dyRU
zS86<tNjK)1!_(SSNxtgd96{e`1I+p2hw_*U2<AxA)Z5gmS%|msL$s30Lf>+mV~g7j
zE43cXb8~7`nw<J_N+iCi({EZ*mMK{kH;Hc-#8A<Fl#mrYhfs>#Mw-QrC?`CNc6S{T
z_dtevAD@A9P(qWM)z`3@AC@0{-LX3pWUY$>#$4kE)-)MJAzLsk%A_*eIih9-ie#|s
zm~$P_^jBy)r&}c$(u92hyo<4ku^M?e>REr~8#(6GX9B3-2uKi2GlLEdD=m#1hRX$N
zjy-Q@V&W9PY@>8~q-rA*H6xCF1<2#)t-<sdfJc?pOFH`5)ui>;SUcOFV?OQ-K8T<F
z!7Cqw#Nb`IfU7ef6=?F2+|#{FOYZ(e8z@c5(Jk0t^-U?uLFCIEp8xtr=Z~S~A6tWI
z4Q~ftZ`&+f%x!%5yr13RmQw7&PuPor;mt#wZb@%8%#ZJ#>1-A9d#2}GD8#l-GQ6iE
z*WyUjV?LaDd^Faa?J&mraLww*)+u6nx=1-epjn1YKi$mhaCe2uc*L6V$R2Od=Qa*y
zwR+A(H@zWPOEQITS;-v>uw<RyPIb4ed7dpye0==#)i^GCztsOU12?`B`JOiTB2~YU
zlVM3oYCzKw^_s_WSm~?lSQn4VeqF-jz~2&-68=_vM2**?N-vq>%S=s*1COC<Yq1eg
zDXDHo^37vJI13{W@Sf<t(VxVGH4#C>$W!=amP?DCZoRHq6{lRE=B)+m#FUJo{TnWx
zy}J(a<MdX7=z^`eK=)8sntr1*pR2!SqG$h9^(m_qa07biZSNd#eVo@642_AZA2Ool
zR_UJ{b&mSwBLXoo{`{lA5ynY*ctXuE1bqS3lN(RS_uNq2|IWqm^OTODt2Xa7-ASnF
z2Gjxw<ecOLTF+sINf@-9URTv@Tcpyikth(~PrV;TLOZ!-H@LShmRF{ItElE+FUCbT
z(!*}98YXh<^>7c?H8ZBVdDxBo*!Tbn-uXXo^c+S5HWNk1N^Xz{4CQDYoqs04$}6W|
zz1eJIulQ!pBm2)vfKP83YnT=Ibk$`!cV=nZVH9)X?vUm-`7D#%D%;s=wMWNG4)Egl
zby-M$0^52Dwizfc!qZT2lq-Q(?y1+_nV~aD`}BP+3P7|UM=l~i^6g?x`f0r*P1(ef
zU_wh5usD@OVD-zgAPPe5W!p`Kq>I10H{!HF8}b$j@Gc1W?9{!Vc1N$RBU#?5128J}
z1`qf@0T#dx18)Xk3Hy^;8dgN*+jO2fl)XU2ZLV@p!b!*S?Am2Y?k@rSmpB#gfUnrB
z;8(KynGMA6{+kw4ef0Kc_vHSNILE&cYN+#tr?D_iL~}db>Dq?j9_n|@LgJ^K$ZvmN
zoTXWK9XP;Re_=x#Eo&=e-5gDBuu6hCq;2sKmfw;2LvG#ubh)~HJFhh!k0m2L^peZG
zdKnOkNXPh|81+jms_M-0DMeHYi^XM@c<<BW<XQINwzFFI4PMZz6MC^~Qd;=WUv02s
z9DAr*16*cUqTqE0PU4UvLmJ4!M{eJsi)Rz1pBiTKc@I7TctZ1|8EUk@+io~3kE}Z2
zxLn%Mxc|l<H`eiJ(er2!P@c-$7!I=9pSm3dwb$w6t+YQo=cAab_y4?dCls=rZ27EG
zk+m}_WUN%acoSrPUT(gbo08;nlF_C4_ht|5mqU3wr=_9R$Z^)*wK~!Xze*rs?_5Jz
z^wKME)$5w*>qWSF@4IdTR6Hy$)t9#}shiFwyuWo|;k&M&(zV_<(&EW#z3xFjY=xgH
zJLm$=tG*S~lVsAU2sbF(8ZCuZFsW9tTX&;=Z0-|C<ah=6rdJZeKVJvoK?np6NJA4b
zllD&L=rK<CJe74@#rJ8e4%n{@pc(Az+gu8Ll;2VAAZNn9Y8m6iRl&%)s<@|Z4%Zmz
z{W^Wm7J-|i+g}ah(;0NJDY2#PPE9Gr9ulI0S)OkE0up-W7ArVwf<C?TH{aL@NpMd8
zsQHuDzFvrM%Put-UU@$_!q;6uLOmmBD$IgfE35E(J2Rxyud`b*T?~Mpv{&Z1+iGq%
z8I$uSo+L7O(ClLi7ly?dLZ}{b!cACIUHQ#yaxpIBxC%jSM9nWl<y$r~88__0FA+60
zmeP`J<<)b@^bgnJ!;iT)zJf!?BY~?;ws946{$j(rNQ`1eZ@KH$o1=kC7Nj*KdQe*^
zsUol(hmM{mbwj}qt6Y(_mSfOc%27JZw!vC=zTR!_<e%3b>pl9fra*xWCi;gNRcf%t
z`A$arb#}Vi|E@BAWmoL$df1~FTpYH-dAlu-?XImi09+}GUwLPU7wrlzhv12s4FMCx
zt^M)8C0BcFL3WivLP>S;26148trjM1u+jG&5dpi4!~M#qmy=k*H-sb295Z3-xzF~B
zoT$JR*Oq`Pj%|IMYPq>&H+(!|3{4Ck?<QE*AMi(W+C7T!@ztR5b<Rv%pYJ}6IQ{}3
z85z_(PN~q`uF82Ev8&Eh`$QLOZNxRmA0quVe%%ijT}w+EJIXkM^df~#Y6=k`rktFn
zOs89AU92T+zb1D}j%|RPlTRG_9s3XLicUSE49U&0F|8WNQy|YGdNy@<zqyt9NBQdk
z%*|+nme0fc9N|TR4D~Q-D)8yF`(ztFJ8vW4aRi=0jid#DqYjzWJpVVTbuj#+ZL%96
zvAARK6>M+!T9}Jfy|WW@R<va#80sRhd~~FCD`g-ziK)U%n_{8<o}evRA)l~Z8D`sV
zuBng2mXG`Pa0#<f@OH^8cnr1^9_l1~FK4&ZwA8mM9Ct|?6QjXY{gE6h7F&wEd>@j-
zdEYmL_i<QS@hK&n%4z&6roV4ZFg^CmC+`Yv_dobdo0<K6YAe~FdZ7gopqHsq#6OQI
z3G871R3L(~26491Es518{i)~WN)EA0vDI3e<Xvi{Va7`6Q<auVr$Rkaqvyd%{8G9Q
z@~P+bA-=Zq#C5;MzGqLZ?O4z2Ku81ptYGi?GC=PC<o^~p(%xt3271C)ei$Ps@4Nm=
zwSsYy)NQ@9=<q3fN}0lF|E1Hq*b4sDiJKPQQSJc#x)K!^5WiAnr$BENWf%TxKgL0K
z0ceizBdn!Oz0G~*SYB^kmL`Ktdg}Jm-JuP-6<kQ%opYprPrJNWAuEAVe}*1&@GX6J
z_6$3=|B>Y<jHwV>_z`or-H&40%>2m=t`+KT@B=cSI-YMdCH5o{WK3M<33Y1vOFXXE
zs}=?obI%#_U|xqhr8Au8ZG6FjD-eFj#0RQIMAPtE*2WfuCI4!=J=(Iy!{Mw(*(nZ@
z6d9BOOUoZDi0WvhQkxrC!ZL~~t}A4tUK*q4Cl4;;^}HXUo(zOOJdDTYrr%`%TYuXg
zDgeN7z~z>={lgZ%#&5R-BiUrsWRkYtip0MHiZp5RZ&oZC=K{wj*cs0dluwpHJRt;k
zZ)B^U&GGS9!1%QhVpNaMwR-A6$(FV07we!ils}LT#O@<R*cLY~r~I(fx@S-IeM23h
zn@qM(cNVfPQTsQBYp9--Rzb8@87?u0#?+gHfO-_G#bR8BVyDF-+WBS{M5AQ0+}a$f
z`ZW(=`uE1eb3~>4I(g+d?~a_PIE}0A()kZ{yOB8QAF-q_2!EMz6TbhbB-9Rz`SPu0
z#b^$(t`UFwLVEhAbn=PtNo2!B^Nm8!yM&e3PA0&xLaDyKB+3F}f-|MdwwqowHEq-F
z8k`g^*&Vqs-PC8=7#Dm$b!WIE|0$?Mj5GKZ(CFUT!45w5fBLu{DYbt-ZUNLs7@i(n
zRM2qTf$G^9US3OPPwsA&P^Y`Hw1-U*UmS^Sm~KMC+@h4s6(70ytFwqfzXbKODqZuN
zhTXOv`0=HR_mexb1FiffQ$;*A^1^bH`X7}29F9!K``akgIhx@pO_j%IJIeyY<jh8t
z!|iADjEjN9`-vDE?D;az!_JN=kr-Zfx!-dvJcEwR)pKIj={B5pU43_Ky&c5!PT+Gt
zPy;nYb?cRnA{Q*pw79I09bHb{?a2elMyDed!?+aBOhfM**VBV%RRDcLKSOXn2bb_M
z0157Maa`Cg<|bg?_qUGo#ziygf%~pTp6&Hz!B1IN&3cfNL1^|q_6sP<V$8#e{sT0X
zXK@^R$5W?=XRAbH@DcZb;`d7IKXMKVtmvh>tsLku=n-1ivP-z7j0Db9G5WQOjWy37
zPtcTELlY6r&jtEDq{nrG;`!164NJtpZ7{pv+kaJV|CVHiWNiudx(<FP)Alw1e_G{L
zHW+V9Je&fe&)2!lFk}86m~u7sV=<Z}d1&SQBQCG9ZeK-2#OJu-->+XB$8Wn`q>Ijs
zHi=qT6A?1AuyNgrY~n2Eyw7g4(dhQP8yxTgsrSHj*bT&Ii9}{Ki5Om~&jhoR!8}yb
znFVO#c~J=Xw@J`2vt_}fugHTg^EN(HHK~bq%GPDz&tgL<;P_WJ6*NW5c>VE>L5|DF
zH{N>vqDcOvdk)!A^>f|+D#rxg_5<^}cX!+Ci(<uaw(a68>(s?RX$P6EG!5lK<1O2N
zIVr|<1n#E0jBq&#2bUfByHWvArMD#?)!iDH{h`AeIZ)>r2&nvze7nK6udN?&Pxl;T
ztz!rk!q$H%lT;i%)(hmNhRXDeFa4g@kv%l4-%c;_nIP>b4KgOIYJM80zXLkGh>1c&
zkY-8j@K4-mPNs<{IuDUEe_F9ExE&7m4wU2G)gH5a%*7f$KLIc~Z0PU$lOLOpP)xV;
zHC$FYik$XYgnH|lq1QE#sEYOWbD27`=tT}ZuhiMoj{cQ#gq<4P5+sz=@6LC<svh;t
zB-pIKB_43CnFoo?sV^8d5H&2lS|xPHn|`=Q@@xtn>A5_<7cTj)xK7{qYzctBw78wG
z$1Z1Su4;M%XEq?7Y*gDJ^6uHds3Q*wEn$+im59$3{VEE2B=0OX2FF(Y=M&@~Z(CeG
zu3kethe|=Zi!z!+4vL)8Cvy3}Eg;FN0e2sFPZJ&bl0Q-%ep~Sb-NSfy$r;C!+FcaP
z#fF6!X>UHfL>G$o(onSDi(W1c-<c(NP57TuE)HFAmmyXzVQYL>m31>4tADM_d6pN4
z>TARLZr;y@5B1zI92tlj29TdR!K=N+!T|<neF)QGe!N{H+c(2ww+Hm39iWC>Tr<dV
z6r2~L?vVkR7;6<W+|0I&9&+>uao!4TZg<@cl%E9m=Bx%XQ3T~i@RE+bA=FK7R!;!E
zkKbwPx)Hxi6{l40;)T>_DJ0x5=pzD!@psDyar~fJSx@J#>u@f657TYZnLs=9+l3Xe
zrzM$KlqgE2rW-Tgy>j|_dLDfN`R80<{6NeyUc~u*b<>{iXuqLceXvHMr*cm-H?x?F
z`B!mQk;mY$!`gj7KR`~gDQvcldSOKkLN(IPNvV7ifH?R%)sR23<~SZGH|m15UIF~H
zh_zb#Mjf|Jou)eYwjB5Q%0o{tryOa`ibIcuS$9x(`d`pHwvapU1W#<}4U9R#6uE~e
zvX|Yhr+FJA9n5#PdE4Ht#2?L&3tH&^K0zQQmab!*j%#ipp(G4v?bZ)F`2Bl~O2L<i
z-}!;U1qTQX!_ECuQ;-E8`QZV)g-b@68O;6n@1gwu$1x5NH`9hkhaF$zIN7)QMAJ{M
zrF=3b1!R)Gy6W1j>lK*xkP?YM_?=iwY;|v2`S(ibd%{3_!-@>U&xmggJpZ~!00>>k
zGYJ-&b^k={VS8PnQiktt!nQDyEyB-2)R}j^xMun@Vt4AIrI*W3%&ufH8T_fIM4_JF
zSS8$K5)E*e=6!7GYx0QZs1LK=B;`XtZZF;WY!MYINw$&0vq{X$K~TKjI|ULYSkv4m
zdS6aOQN;=u`+&Ppnz5fVhx-w^%($rq?;CSJo^Rr{MR;kX{Wj4-tn9B`7yJ8IlpFq3
zWXTt{C+Bd-cayqi6sB-Z18?s0`A80(Jy7UWc*8C6z*hb38~>-Np8b!R?L3fVs)=Xr
zDllTR!ph8!h)BjxPfKqpxGcjG$t(<JI)ESuRtUpq2OqloI{20z4Hi0xq?a^n>m)b#
zs{E{cGA3q-i*3nU?BIHrAc4Nz?8uI2H}{l?^g|YrOQkE?Mq--G{Fx6{n*H-nNeZ)k
z3&W_n#5j0Ff8I)ZiLuSA<Q31uA*1}kUZxA4(LjdynY7>InnEUI!x6bvg;P+MXJPmQ
z6;3+uY~`kZbE}u_G(tTV(zu}K_=x-HQ0UIRVCoC<9r&sVRO(9+lWm9;M)m_6O;Rb7
zR!_KX<owd(M1JkT;z?Q`4SD?~VDk)2OwIz?Ysu`UP)RGgD0Atrt-=G6iiXRY)4!{9
zoW+@qtFf#P^Vu{Uw_1I@A47stW=l_&P&`(MkArq9C-M)>Ml7@8142}`gqCewhE3+4
zn#Acpv`c<{Nl=dI9zz%l9F0-$FY$ILJPeTD5%hUvoH$Q=Yw>^Qdb@q0WffxR*QdI>
zl4p^jfZP3}akr>?c<8nBSmQ6DpHd%$waVsb=(}k3Pf=|-*`prd=;<le-9=g@gQ!>e
z(2_JbSRQ*?@p5+8RZj#JsZLpYYlt5phmQ0mfh6JdHyNzv9dg@d;za-D4@cdtEXv*s
zQs3|a#`7P}F4~;-nIj9ge3!qbc0RoL>j~_&lu1nw0dE^0;c7w)k_dmy*}Nhb?$im}
zh{HD%2Q?tDjrmCIuVz4vDtKA7<sWcHJ|nT7)O^oM4el)ic{KAs@~uKSo#v{qCOPqR
zdcD{}IoBHI>uO`e>LdF;i5AQJ>+$T13_pl^^3te-`8|zXmxVqM!@t1?20uSeQr>Xo
zeFV+8d$ZX@AC5_D-VahhQ#f^CEoaY0^UFxRfge7%U<qR#e`J*%)G4YdH)p%u`Nb2&
zUrAY7SA+q%epUw;X5EZSBv(d8d{edfM#ACd{mX)WIT3*GEnS~}++gA)e`0j#p{bdg
z@>zzX-ub6K3jEz~y@q34wr+-{*E4xk(?Zh_ZaGN=g6MC}ZF+V*+gS6;9*$X>TMbz>
zELt3fY}IP+)tfw|*+c=>uaooe>?2Sg>Q({_2#uoAQ-<xu9-f-b0*BHtFy96OIIjB?
z?+Rb_m&#+5KB{q+jf<@2R_GY4W52F31x`JW=G6^VtYDO15iz^{wnY8rT%&%2Bz3jt
zXW;#htfA{X9^gA8rtZfhNIE>0T+NU=JA3y#ls3~CE0m+A)5mXMDmHY!m^QHxO_{`m
zhL&HOne|XZt09y=$O|_1c+~RKHfOS#zhlX4dKKctZ)yQLwZGQ{Y~{(>jO>3zr|vK|
z$tiH|O6W7S8glN#6!j`%3$Y*P%c@U`v$B?1KQpRRd1n4ru{+CXY@g4ia88b?$B-5x
z6Xlq$QQgmI=}@V4FhbOL)!9zqteNj@j><(4fJSt#ly+gG`_gPxRCNeHClf4)&OqC?
zSprOn^R|zp3lhVza8jlxzm)#d%%MJ!6OEs5D#Nhud98|HL`R_(C)TE?VVAbFjSi6j
zEAiLL{w+VOP2;$knsU}>lu!=9O2LY%N>aWn;I~8irtp3|X2el({0XDNq8y(w4DaQu
z4b_L-v$DYkhA<<A8NkH_NjNz&Mh${js;Mo}i5QaiRjW!JHbJdLixk$vproaCyRkL4
zBy3NMiw;;7_#<sZJ1+M?paJn#C#gyik%@k>Wg3Vy1w2xo#?a)r%VX|EPw^^E-6&55
zk=0jKOZTN}{-e)$p?ZTo{l3#$Z5z*#$hwqY;s#-@Kc9KCGrp#OHfEpf<11R>$QO%n
ztPe(!`*(q-&>U*G>+7HaUELuo<e8zaib*N%_R61}aj_IwXlOn6SlURnywr)Ch`n5t
zO7c#I;Y4=uI{|b&`5~uht(H#K5iQg26tbkI`pt>8Mm)jFu&xP5^ox~#@Nmb#5eHve
z{l5(_9Efu<gR#f>7h{Oxr;HTczeV_8;W5l6{}eFVTd32N{jGH4-iO&}*lSr}HA!q*
z5*dstC!7^CjTOcgGJPmyJfqsvcWVr>=;~>C`M%7Fb-XJM7Vm-dWJQw}&bE$D^{1fY
z55~;zR;p7aE?|gh<E)Xdl^=`l^RFJxlT52N%&A^y$$(z?a_=p8+G3$R+Fnj-(GKat
zqKh#a#YElN$xH_>(XP>MmI};1{@%gKDIX_y20wt+vo_(FmQ5i(WE`dHq&OAeS5S-p
z0Vnf;G7aV^g`~3Zvip7E<^yX-O&F4N#q>tzl{m?D(+W~)7FYY&Uu-ZPh9df+F;DPB
zL7NxD5{yzHK~<c^oG2~5CBOE)uyWFmL1{4yrUO$-*Rdz-VvXKy>PAj2GM@$SC~b><
zEb@?FTBJ(c7DmXCSXH&cpV&p&*muLtc{=2Gyr7moZj$ZjY;kI6@Cr1FCI$ApABtKs
zA5CXNtJvZ*+c<gSJI!}Wql|j-o(!+gs<W~cx%Ao;b`-s&MAydOz1CwdO*Gn%qiE~;
z+T0zx?bLcMuxbzKonmcdfWD6h1)1zVGh%_jZvU~Ng0~~Cm5bw1yy%=gvpVuA<w{le
zTJzGPP9N8aOC!><#KIi;sIN+vO9Ob@^{D1ew~OkWn-lx$v8|ZmEk#Mk;-A-G)5~{4
z*F}IW+BW`FYHml>Bc*Val^MxP!tCxu<q1m_43;v?&SXeAu8sdZ`G!`Ts(fw7Jqv0j
zo7>bTxG<;%+voTC<-Wz}{(2Z9^>CWqbN2k*%bHK0*VRB=J_V<87oy5_#x=9xJ4;pT
z(?01om34xU(|JR5qPYPuW2ur_R;dT!27mk43}GcM4X5PdffsS<yuOH&Ny(LU>tVLY
z>{)(EH7+d7kW?Fwm5=Usl=J!6a*pe>T>aKk{zV{}<&l(z=%HjePxG`7$2g)67gIK#
z&A1}FYH14X5?}8F1*hclww2cZw{iCEoUB`dx^G4Goo-a9u4Vc9a5G>T(81sE#0PSz
zw5?{aU*3}*#iDvlyhV#JG@A%etY=3Jn$<lyL%8Mi(g`h_g6axYvZ1SIH-a#EoF4A(
zhJIwE>FFqwHHT}2UL2gcmzw@-k7SjCyeA8HvCIu{GFaO3lynR?!s^%5<=Y~nZhun7
zwf+53zi~NdD*TmU$H+%q1hRHqL)JVBO>rWf#@zi)#jqBW?Tga&nhb;|A!P;r9{`U)
zaKHCXv(;zH3|PiH-7{28@ztvMy72`t8jIqInuYt6xZ)Z118PQg#EnXYlx``ql4Evb
z!`6-B$&A)zqvg&X{AgM(rI<H&@k()`5EPvtMWKWUlNqkdTH3>km^URgKOdmod=@8o
z+jmZBImqUDX);JR?sZ>ihj(wEtA}t}QS{S7<;k!OL2pEcGg_{44K%xPwIRqo_f@CC
zATSe&1UpcgB6EzS)nu?jJv26<NoY2ju5s$^+jf{2$6Joe4Ou33yDRc?G``~%s%v8N
zQq&DtiEm$h&#@#j#X|6>uoV%xyLr743SDs$x&tK3q+3rhYdQ__&NRQjF88cewiT3Y
zDF6`jYF3&zp}n*U=JqrVE>joFJG<&a^0W!{grIt1*TP8s+-%djc_FG6axTPZp%1Ed
zcx8oIOh3xS>KW%~#b}aRjzI}#4ye4f@^P&T!3CdMsn?G_s`T9sI3)xN6{c*5>^(1*
zt?e5v4-C9Jh5>8f!=lP%zQb<k;`VdqWbS!0?Dy=3gj#(*NDW1#oqVKXUayza%oHLe
z2lo3bri&-+WJi#|?OMoLS(maggq#~=X)dpIZxvS%y)^dAn*;kK<X;BRi8&;QN}24W
zNQ!b?7S>XDd3?=_;|;%i@ini~%Ayd%j@`WHxE_qj@lL#!T2gvR3#%qBFD;pSm_1%!
zqA|D{)g*TEW45X)I+k`y6g%xh@kH3}-C#z)3Cau-*zI<ph2!BsZlQx_6tsLQ>PoVh
zqN47)?(?V_v?#}AB?|0<nBt9jPvl%sPa<s|W5PT|N=dxVub6dbO37ro5@HdwM8Y&#
z8YDIbS#y_(_gyug4s@RwK*dhu=U_?kOYvls;_Gdzu=9fz-%gVIl!awDk-JQ>;XToD
zeV&>V&BZ)1hlwCF7yCWCi(sC21lDBX5^KsX&}BFN^g@!gu+^fGh*iZaeEs|tbDFsL
z<uBMz6PIGAolB(=g~PIPNQvjSuX%pF;l;YLMpz{v?5&VOyn&KD<j!CDh^fsvaab&W
z38gS!&aUX1H`P2(m`hcrd2i_U=$2<A6bR<kR6~ubQrHZpL97y+M)Oh=Ebo<MG%0y4
zlp+)f?)ju>43q+CDH|R8yNi1_9goXf>DZE`3sbO=iet&7l#l{3Tp|+Kdm#!!BzrYd
z^-vgXnf!yx%;k=GWiwX!yocghuoQ_d|2N~DZVGQ5S5CKvpA&k0LGc3;xjP=s*zrhO
zZayJuo|Kmo*~N+dZij|Qk$|RxW>}MWf?r!dtrFtd?ATuxSO?T-HcHr2!V}_r4J_=^
ztv@e$-lxoq+Z$dU4}5+6o`bFknM~wX0-4#XkREM)y`u}q>zmj7<L?v8@kmaI-Tu-;
z=gcc8P8ROh!g~Zl4u0s8XhtGGGzt>E605Cq<E@pU+n!)Kq5>iZr{#rQ5@k_Jw#tQd
z-zbwAgwm>Q52^Zd$sk2J9F9y*%eH^m!IF7sXX}refi4E4u9St64`@p4<r&Dt(RsY(
zBdt5{78+#T51wG(8d5`$Oc^%=+%!!nr6tH0u>&$Z3|iVAQ2Ds6j=RBQkE8BSnIAVV
zA?fx+xz7j<C+0)%a?P<6)VA3uo7a3bl!)_yIZjM5661`9h{gy@q7*xfC`B!wQp-o!
zx6|IQP6X=hTmb?+`^8`It6%>W-+cW)P;OtNH!qv9WL*>MTKM+nB`=S!Ib_S1cr1#}
zd#35qbLy=ea=vFBylo8O_Hf{vZ@zIkHrRo!A7E9pP^hvXX1oa{_$sU`JrF}J<G`uh
z=l^X!UPawV7nx@JKCOvd5-F<{#_{M71WGWSWnJAMR@e>?-TuC&ga%=bu^sZ|WNFkx
zm_US_7TY|T+JUb`+o!K}rVv=g3fd$~WM{t0A+!`JwEkAQsSUB!wS`-5ItSH|Pj`^4
zfOxR9%iPtc?dQsq2Ywia0c+rc;Y29zWQzNk?43gt>2gKscFIyudVP;<6%9l1n0&R8
zvvHbC@m0%-lyhOp*|~G^6NZOc<FEC19@fN-BnUgJaKzC(IO=JnKwdLT&fG2sZkGc$
z%PmI_r;B3wcHYt1{d9E16b-Xd?Wj2_)CHN|)!7)t7XRM(soKeCtlk@bZFOyKCnD}i
zhFB#hI|cI}?F3;T{1^i@XT^CB>$NHS>6C1tf_B2@Oiqm93YHHmOD3s#U`17SA(|jj
zWJ?T8m9**#I#p6z&M-M`RrTZ+#4AbFMW)-hby{AvZ@w>ll7jO+O@1$Q%x1t6skCCD
zzcEba;b4|ZKk0OZw-ig*U5(K9Hq{L~qGVNGACA0~BNBJSIS^SvGeMGt++bzCm$Y)b
z9yzQFYsn-D<f24^7ljDz6!P)#%85Cn)$pZNfGeIGWAhGeURY=2TAyVC<97DC$;?&p
zHv?@5Zpf}l-?60aYbsdc^O``uDi=N%+8j69erpxOHafQVs)tPVbnH2~oE(!^Zw<vN
zg9Z_z8$eDx%(U(c{=OFP6xhrpcmDA~@|Dylf<N1g8Xdo9^82Ldxzqp8Kkjj_C&wdb
zxyTmIG)Z8RP4TsyIUDb-g)}NnHwT5j9a`IC-X3Zt%1A{>C9vCF5kh1+9zE=-Sm<iD
zJa)J1!tHWoB@=d#MOaZ3GyVn%6nO{L$;V|~6R$N_S0}0>E0m7+_P0rIYiR67)id06
zmCuYG&)qxy1_~kx(L#}IiP}@!5vWo|3Eh}>`$(f@wGbamR~6ilq-}4#L(7<PDvNtO
zYi6$j(TQVrZz*2{O{|h`_wvv`lhL+njJ=1ngldrTkYnyH)h-Ilbmy~atMq->AGd4D
z8#aJXhnBbFg?{axu>qd~zpmQ-@V(uy*$N?@@??Au4sOJ-d`J<9C?=b2>!!GeYzwYC
zX^2!#WkMzeNCIKrGwq)+UtW_H!gNE<mcKP?VabWZx^TNLB*_+1=Y~xQ7VfI@rWt)R
zQcB#G%x;>AaiYrFwJ6j{VT1N>Cm(j+Na9vvJ-K?5NH>w&=r~66B(Gk{UUc{5ZN)vd
zi`~-Zs$cuLXpPi0ov7gZt9hlOP;$*r=7-keQ7TcGOGL5C#Y)eQE3@d8H}9|Xc{?|3
z&8Rz)c5OXf6YbCb5s%!4@4k89=ZImz8u*jq+*`D{!k$WdCM;t=RkVx!B`2KLHufkJ
zyWJ(nlsVii%rX-QtVu~pSyw-$ks3L0cMI(I&nzZpeWyo|nd`s!8$O#OFaPlWCZtR(
zE2{{r1YYHZ*T;pw`~F*Q(~*>o$<8~guCuPUl#+l#oc;7UJ#Nf$DU_wK4;ReS#3UAu
zBNEX9YjHoMxCz1AF->+atSWq%tF!gg{@zYON(+Ou7=BA~X>t6a4SF>^i5P>`drq+1
zseNq{l^<Ge4aLGbBot!Ew9I4@n5Nl}{;Uu{Dv79<KqL~$sw^cn!Kt3ym!gEAURdz1
z@C@dTUPG)($t+8vnhosq%QSZZ4~4M%`)n-dcu0JD2VVS<BXp9D8!k?S)E`dA4X$L(
zD$GoTuxCk`<6$Kd5Dix0(+$z5hN~(v)s|JWu0L8VSIZx6UZZ8l|Nggs!__YE$KU-6
z*7d+`PGp3v!i%)<dO7m->lfUnqY0T73qz1hltc)MhR*o;$cfG+!;-n!U9y|zRxBaL
z*@Sw}Eo3BI>98h%_2k8}<=Wdae{_ah5hN%jnBQ5P1SAk=7s{7n^B7I2HCKN>brxxr
zCfS<tW;#|&{#wfYbgTj;21>RZJ+XyoETA?fCF|ze)2)XZ#&EKpT4PK~uU=<LX<k;j
zT0@KNb`&U5%UP6jyB77N<>w&~`-FD1Y&c!Ss_#h4(yf(jkDs#{Wpff%%`7D$B|s6D
zHIquQkVe!4|L*Rx>hD4=&r6si(|+O~{`j2T@rwWL|LVU+%fi3>7ynPvapkro7Ez8U
z-=u}-%Yh{qWEZ*o?3zQ#9G6=rgg8MWAVoOl^_@2CTr6(!VmC9#iJ%s;BVEqEP1cdN
zPncIrbdKZsIVo&Zw>3gUJolrIb5JxC=8!2`2qlv#WKYFuo3I(jHB{q5r-Ih`siFaF
zN~mJtoyojB@1x}sB3UNNoKO)~*+GdW+n^Mkws6Bann#f5Cp4v6&U<KM<_5%^Ynp=M
z@5b{A#a0C_s2NE&0q}nOENHpw<U=EI9~~>S9D**{;l;u2Ic^)1V(V_l#H152k#isw
zA(v=ja~_7C%9bGhq*dCl>n*P;649XitH1qw^6S9=?7#WvsIL6GfAg>T?)mq8`}H4q
zwXED8kG$5*K_F!a`<bu4dd5QGrYsazlB_>i)-tbOMPjf(Nwq`pQj$?<+~m7@ASzlA
zEv@pn4YG$l77vY&79A04$ejW~kRnJZNG$B*M9%OsuPvmu?yus9T$QY|o_e`&#z0R6
z;M++Zhn&sg<-v@YH&$edlgk)^EGC0TiQeaxlA)}&2h<HjJvLRT#xaZ1P+v%;_zqTl
z2hY$l;rik$&Cp+|-X%6kZqR-=*?M3Ah5>8fL&EbPoE({*Z&T-T^nH)<e{0y1bf4G1
zo11S<DygzD>O#vq<6mv9kh6b#h$+8oQ5Fh0M3{G6TtDS_yk)t0MN(xcg`*aZC3DD$
zC1+DiLu9v`IZ7rc+o<bvMILzP0|Rg(*t-JRPHEImQ)>0HUTBf7qb(`9pXAcA@bqI~
z5^o!0Pu#klj&E~V4^_+{b<*5B3`x&_srTu2${SEL^mlp;f2pO%TdH@L8?3Vrp0--H
zZED^uyS#8xbxQlw_i6~KM@D*j2-S>(K3vrcrMBQy|NaS~&iNY_?2n-1?JYIugm`FC
z6`I5=jh<Bx1xw)F&ETn#g>JQA^jR{@d!~z-%j;(pnYdXjS?QV!hg`T#nY9!$N(|<8
zfT}BJxvh;8kNL--w!1gJX_u(p|J&NEr_TD@f4djg%7mu_=|5Y=*P0Z0^I$lmW0#gs
zyc>8N9hU;~ARTXhrv>vO4Nfry7x*_#`qZRS=ej1hmKxGq*FX(luKPfDW1V&F#{I+v
z=6;;7$)6S-pW^|d{UNkm@6nDQi*e;r9iJ66n-xX1hV!3fQP~~<{SxKG+g9M^?G4|B
zS1|9H!_3vQFW~mT8V(%wmb4ZY_3&V^Lz-!dge8;LYy}iNzhnJ*r@7Y6Ls1%$nOBIl
z*1zqTaid<vd6N4!HGZOZhhDH8St>{rh|)sdxBQAUxu$v6^t8+-^W5#Dbo&x5&)mJV
ztlfShcEE>wkTv<LbK}|G=iZ%_mZ`hk#y_jaOV5eh_;qqmy}#0O+g<LNyvOq7O@g>b
zKK)oB=OcC2Zeb(Nd!TIV=EMuOqXUsDS9`VQ;vPYs4{Iwlvhk2RCg>_|R<f>((x?Bs
zmoNB3{FX(Pd78O;`ix^KT;3kIO)tpDM3IO}AWRlnvD?k0v{Kf*8H(R^>b51-(Xr8F
zYvz4p=PdVJ4>`H1e&5}1IV<UY3Y1#iSvSvP*=QBr^p<lq<xbit-M&@7r_@|@-XJZC
z#=6JU!U@F-7ir_Em4@&1uS2Vp?BBPLU~j<oJKN&w7RC0=`zP(^BU=yr7z_i}zz0M)
z-N1Sww8vWn=AKE`xcthrLV_s5MX;O}$HR)oz_i<;OCV>fhL{u-3l}IIgRq<x#>Sco
zrM|_+05Ou5%t4j!ZeDYFvE#4*`G3K;-+jZ2zyDia-@N42a>I+1xyj0Hu@Ht`Q7(2n
zc3(c@_uqfdi<<+JhzkYv6p=^o#scJ=A;W&2PmOg{Z;frVJcOQjV*06R7wx(h!|y_X
zoroAqiow*7v@Ra{WSMz#Fi)-Eb7@Lw*U0m^?E1O!SZhVe8^aS(OT1qDDHg7!IgwK#
zOLl?F+`PByF4ZV0Uj3!!90`^Srvw6)d_Jp%O%w{j905+mzj{by-W-17C0hu^olvHq
zB|1sBNy*(;?@Gu1T~wpvIqreL@sQ9Mh;e6$M-$|1F86k_zbU<1rOGCCiQ9H+`%I9#
zVdWJIuYsqRPxxp5{J-X#Z@yvqyT9e|;(K1de#v(?3$JdEmK!Qg?&e>RlF4Z$OR<6m
zY6*`^u_VOrf|QcwjF~242u#zCf|AwqLZLOd&)2yHZS>&<m4>%-bf~dHDw~uoJ~lZe
zaw(@L<+edsHBKk}s&C)1_1s6t{(DVAe;*wSfr-dkGNoizSwJSThmqt?Kkbd}DxMg+
zY<rwlc(R1K$-OnI@h?{|7FVbnt5Ip|Yzxqg=sh8cKM&~H<W{1sd)wK^4?pD&E#v=2
zg&2fs-lGJTBNVTKl{20fv2??)K7Z~V*cm^m>s}Se9>xlP`|9`H+$?-|T=?qqXZ+27
z{@?Qa<#T@b_17H!@%P*wzu~Aet4>UDW|{)~S-3qa$5#n?GNDs+{{*XYOX+yo+eZ<_
zJh5^v6ftA!H1Az0U3$%y8dexon2fS1MDvdBl*J|z_`s?tAqFPd5%&`z3dfrR$qnwP
z3Gpdd&YD&@Cu@y}e~^|>lytwQWH(~nOUoWSR?Guzkt0YXO;|CT#U2t|dshPLdF)JL
zsGb&^qFsuRN<u}6o)oC$YSOF;u#0>A38jWGwWow;0R8cJ%cB42(D(TPVe$=K((|v+
z*5QnY0EZbNu&$7UCGIadle2r@YAd7TvRG|L&*4E_uBs=Vs_<|B@bCHl`S<*@|L|LW
z`NbFf@-P01>o0%BXTSIh{?)(vcf4Hw5nkVbDA(6BaTnN4k(=en&CSu)R1hRu7+p&n
z_YNayKA^Swpz)Jhoz;v=HJs3TAs&9b6%gqksJ?}{Br2sHvW2Mb`Kg3ZNXtrcnZ9zP
zB2c2YNg>vvRG2XtI^&g`&ZBLU_F>MpFNsF#`Hu!6WP6~Xl(djiCaV}7$IxU0;|S-1
zJa>$1N$pUt_`&eDYu9P6G72rFiQVfPlT?Y_%2T`_Iv)7BU>L9lJ}8=e)Rd-2T5aCI
zi|a`RcpNvDW2&BNmSO@i?zXNT5WYF;pK44s!Q}n#HDgfG!hiVYACR!}+u!~(=F89c
z;;X+vci$uPD-O#mmVC<`_QWZ0xev&0wvyQeG%I0_NDy*Kr*?q14M)!;-O!{M!&VA8
zXJXWo>uKZT3C-D(7grY+@e@?nd;O#pLYMs}XH$H8S-g%(&#c-A-YYC^AF98<9n`7`
z%)u+FHDwjCB)h>X%@nmtj09u#>WQbLSm>8M`&5wK+FU()wo@V9zH+X-Dd8vTowA*z
z_4@9)?AYIP-w>Q!&`AlO2(}Tfc%-oZE_Cdk5{g+D6{9;Xc2Q9a|EwWnO=#W{<#4(p
z%@AaHQH5LzhqChbU;hVkIk2C;B1}8J{OVVfX@|_0yu1lydCf&RQY157f~-f5hgYmc
z$V;(xDQZcUeSJLI3(K2Xt0#%5<-YQ8IO~BAN_Q+SYRSmAhzt`9RmJHwRFRt+2`kjf
zY+gOilU&%i()N``6ID69v%u@5)pp6&X_uUp`{@|O69oq(D54&k9Jc#RIjCW}XKU)*
z$b6<Sp;f3B%p*Wshqk@`T<PtG@V<Tj6Qav^M#oFuSygxoE!Stq;0Fv6u&nFV1E(g_
zr_pcozLIv9&-nzX^P9?-RE1kT@-pA>kKg`*H7{K6CYCj^-|cyF{W;Il#22p<({#lx
z7lgU6n<7y>d^`&!$>ufmgCQsUwkdC&h_Ue{*(wM_OM9j6E>#ugx(8I%&ypQWv(a1F
zyq>)4WaO&RO7AgVok1&Ii=NEyy|nCkDywlkmfYS8RYBeMFY8yW;#R3~>sPMcTRuMG
z9->>wKWtQA$IW^z?s?t#NWHY5h`yh19EAS*AF~$zbZFj+_ikKo=)<|t2AfgBzi)<`
z;AtxA-m10wZ|?Ta8p5sJb9xUOJMi2dUYo6GfkQs>D&6qy^Y0Oqu&$(BSeKP4?s#(b
zjO)WKak_vfRy@fpK7pLfxEn;Mm9rluPv1K%wj(v~;OZ&ddHmJ?^-0G?+(WH~3f5U@
za`KrONSsbY$T^dfr$ackf>C8`inQmQEYmY1R(qxNwBVt8qMi%WJss<&5Ntn^Y%-Si
zd()X^PO7Y?+l$o4t$9jI9Z7so?KZ~N!j8|^bs)6yRN8~?^snpIzyL4|SOXs%r-ISR
zF5ENG#yGjlm(Omm*DGeAnL@B?8w88`&V_Z&1f7iW=Hf;z+i9ixObeT&SJB!aen4RP
zbV|<ri@*JsygI(%zxdDpQ})+a{N<Bhvj6^DE?zw6=Jt22=>=EQH>ebzJ-uR)#H}uN
z;+!Guqj_|Ae7)4%6_0^*dLvc_H;7g3NjZTCG1jVG+IT>0jH|K}Hw3Y<AkTQ^pA|y&
zumWzEBUu%dKognolM&y-S?XzU6T5WFn^cM^FhOcg5Hnr{6>^v;bD<ouL7dbLlcwwk
zuf&$rn5GszO|qmw5%-j-Dw7)w6wgnfY#9bNBba9gqGcF=V_#1tr=KLFy`x*|$v^PJ
z=oo6k=YZu$3Iw08M7CT}2t}(gwVm~LPwFH(-98b_KtwdJott#a|M!3V7ku~pS3JA?
zueiRt;<tbK&)7eG!PSe`+$>jw`31K*p`oz<Vn=xWJvZxORgrRLx1Wu6R}-3>G3GI}
zELi2FC=gsRk75dSvAre)R{*80K+C0}mA0!}eWl~7BnC<wd!v&n1Y&@^ENI@m8v3b}
zY5+Qye(#yI>e<<L4&CPPUOKMr4*{JZEZO!O&&5%V6IJzX`;_-pT%oo7!gAQ<miNj&
zD`NZ8X|p$@d+Ci-Z>8fiPo$p%^nOBp7h2wwHm`meJm;bYf?7_NoU>Ki3u34Cl<vlA
z)_~?RQ2(wrA~&*VR&Ml`l9hk;xBr%Jo<8H(&pzXFcfpg3E57>Tirod|^^%vbU-6=R
zLzV@m8+I`vd`C%;T(#c9XWkv+MtA$e`{Bwt-e$`;q*l=j!TqF6Wow4JvMH;WH);{_
zn#Dc{iXFt|;yEbYi;<0v+O;Q+!f%?ms(7E$g%%}CaSzb#dlB<i6BXvw^Vif$f$kbF
zS`CAiE2tT;1fP3xkJI8_r#5cpc@5PVZ%r|`xVqET|60|!>PM+<?VTcYlFd&IZy-oc
zzdE5~xOaOy2_tOAs|g8K#a0nBM5~fCwRhBzvucpOZ_Jxly!$oP`9tBR9MA;+@bw?K
zdHtH#`-y4Vv7fJ);ssxR{u^?R9M%Kn=|W=R_3>LMk+dp1nTS&)OKxF$Z(?ERLsSh)
ziigwHh=r2fLtm_TUo+0Aowlj<({jWL1e2J9;z0~mLGkfC;e}>6N-j?{PF3?oRZlga
zi{7AfpNCTYwkP>I203Sn2(#2eOWE2hf<)V!NoGLTY(~nqpLL3w@#O6QsHNELBr)*A
z#C8ZQC(qd0?}ZwCuWZkGa7i0a{z=Nz13wDGfHm;JuracJ>!};^{Q~nF9#-!*2I!0@
z#vLI{6i<lS@@)VquU<hfIdjU}!&J?J)DoJ9d)}!gFshPR`Tp>ns~3Cz!@vJ~F86yb
z_A|>`m}Jk>t6w1HnwK|6k{$_DCU8rTwFwWkYM-8Pna80hXizjLC0Nz3Os0s1K&S?C
zE~Xf}#IOCJxd165)rpCE@>^WO(Gc)78w5X2i|!o~T5=6H0r?~-AT^JTzbc4_=QY0T
z3PKS|V{nzBZ31XLA*upV4W|;YB+LRtSNe+()RWJr>;^Zbo+MX~7WMM%HD8uBQaWi(
zx*un|oeO+xr3=W7F`pz}knX$9cz8a)CRA8Es*&PTzzH4K?%qkl&|UusItDsJs1WBp
zG0bLwE=o?ts^~?bWG~rWvV{;fm96u>stR6P{>7Cue`PVHt5Ep%#dpj>_~zT+^ZM0}
z7rU9;RXD6LPnTR>2$%Ecth$ho3xUYxlS>Y-ZqSt4`^0UV=Q-dW?0TnHHkDlpsU)&m
zjw-|xs@ML7UO`i8g{}~U;9)}Q70axQx}Btkf*Yc%=Y}U$Wivu;E4bzw>suy%s$A!3
z8|Rw8>Q4WB;(I!-VagWh6SlrffhbItOvjDP+d31mij=q~ZP6MAoGES3w(;3ow<${X
z8Lh*0{N6~%{!yRgT)1~Z_xXq3X(RtL<4#w4qsP1Q<o-H*r}i+qDNN*oSIJX|LJUUB
z*(;0{3PCDO_bbZhMa;kzDu0S<JZA+|NJaU>w|^up2WGit7iV6CnZx>oW7?CFfbPg?
zXCaDONZEFeQWQzWJvz?StjD9^VN})F;~_)UBcZDH&ZX_UXm4u)>1sSQVCe^p!8uKa
zfMgKO#3Dpst@b@?N`mV3u!a3JQtoKE*>PJj7H{7sGPJQMKJnF1X0&WPH>!EN7f)fZ
z3s%J}NTehS<FXt(*%jg~b7cXF({a{p1yz~{?beT~^w$RU%hXNRXiZU3*Y3SMNxZ+c
z^W0g~pHAh$4foRV8y4?7IT?D<o-j=WnJgi=mL~7nfnoLvdOnBM;3#e3b8;wHb1`{a
zswt$1k~H(p^KU8Zk*hDR;dsl7^qkve=D1EQi+N=&Hwg`yX`0LkQ^Szek8orgg12^8
zPJcH;uoC>xD0nnNPs8f6veCQNSXry$=~77uVjuVATAKw?w*IR{i4?T)*&y!uRVS+&
zg8hGD7T}F7ZPN4hvxy4Mx;8ql?J1sHQ$^5d$=p?q-zuBTR9r_;QrC7Hhur4g4%2&&
z_-z8ZL2f!oCr1d}Nz^7Lr&S%XPk3VM=*@K8oW$>)^nsrmh5>8f1LBO0obM`n_WL{R
zI`z^JW<s1P5=_x4rETIVTi%uIAtxgBl&I#7>BZJge5>yCs*vp@F#|7`SA73^&mVsG
zHP=@cTwm{z>5338xY|F1a>cXD*BtXLhwz+0YQ~->6uKek2dO-QRU#5&pyG@Ju{LUn
zXeh?ys60fRfvUL+d-seT)Q}7{VNfhw&%{w<8YUo1^M(k0LdL4+>4{2+ocyxEj7r_#
zRqkxcW&J5$=C>x~ulcGJ3C8eZunKOX!IYp<&ti3pz4zypOjlh^YJ=G%9{X2n?x5|e
zPEhF9M-#B^-{vkY9fR%1ej9*(G&TYKqsLoT>kp-4g&I#6ryWsdiiauWTtEVWK+2h%
z6I&2)PsjbsH+nXCQ7P3ezHBu3{N^S582JAAH|%0y3Ig#7Dp$lfv)=`F@hO^bIIKrR
z0{i`raC6us#BS?&P&v3qD{g|On6O_^D5)h+uEK8ZWuK?5J+_4(`c|dc1U$hh*XT92
zzf1vXp$cboTuHT@9f6*P5p06zM(wAxV$IjnA2Ck}aU)h0a|yPeOi`K<u--4eEF$I|
zQcuWpNo38<`?5A%yXKwg3L70)#kWr&a!$u=7x%sG(LS$Z*?;&+;J(Jn{a-(pmb<}H
z0@J)B%0y0hl`rE-K?12{J5Z9Ulo?;XdurbeuG6wgP4r}*0tw~Y7vH1#$mds?h!C@o
z%9rHw)ZVWXB}FoUtV+tx_0$ZbF}UJWSM)=Mfb+TbzpLVFIj{<+8P`lfS2a4%Vh2!K
zE1Rjd&>Yb%yw*k~8ZQX~*^GFxC{g^$b$#o$5A?s6o+z;QZ93&fy;UT+XVe@2Y{qCW
z$`h%H>Wzy{!7DIIupE2IyjeocL8zNzS-dEW9V+H_Otl%Nwgq>-{G@%_cAKW)R=RHO
zuc4}sGEVh-`p=vvgI^qw7!iUUCb>~T-AGUinpgt=depex@4iOc^J=;d)SYg&1Cm7M
z`&Tarn)%h!iM(WPUmsc0C8_LK)<VjeMH4a=rd?=WLR#<?1McBe-}6YGtZn&J6O4O3
zcLuEL`%{fcxk}EB32(Htjk7(+4y9o8lCGoR`8-YVWDCBvN}K)r`o1&|wyqQP-OcW$
zXYmAjYKX?$uVRN?Nh}n2#t)~0ryS6&AINsVuV4odTJwRLiN1Q{^#grE&A#_k{=8{H
zzVl0+kaHewzrW$kv|h1s^S8xL5BwxB3|IqyS~MYHQ>NYpO)gX1eb-d9lq@bk7bR)2
z(+zh$Z%UK6M}TkQCv|U{5du&AtDu%Y%1YwGl9k(a;nmF_No(NtR#|jU(mluJlC%oP
zw6NsFQ8Nct4jjlip+H&_Ap~~2*%G6#>Ao($rT1qNv<Pvk2`H-)Rt+iDo2k3IYKN-L
zODa#spMop8C!r8()l4X2O0R4lZgmcO-?ejjvH#N;t=_q=p3r7B_qnL<mMvR2RjFPT
z*%-XIU`*LUhnhfM1p}>llM01I!AgTyCNAB7w<q*%@4kC*t@a=OfCc@j;z!bPgUd&y
z1ac5ouga6E(n?_!H76Zc1W)O>H=Xv4tI}k!bzE>S&B)EN@cMYm?edZZMRVeqzvGxE
zZf<AR6>gT-Bwbj;%A!XOdSf1)C@Cd7wJ$C3?wz{7IYab!cLK9`B5Z{lr{DMx9g9*b
z0Z!5$Dfj`gZggzK@kl!MowF&}RaNv2xD-Xo7E&O!WAt?Fp+Fv96x5V`)dE?Yw}{~0
z9Q1h#yXS%#wM%YFZ|p|n{n7EeGow#R=Y0t8i<Z^a$(nr4<l>dfIHhH}r2Ba8r#8{O
z0@JmNU<rw}8op*!C%*aqd!#Jv{`wiZW)8=d=eGx59|OmOC5=DILSbdmTT)t?CM!sy
z2=g=%M7d4tyLqY9)>T?|T8gr*-x4gjdMFn8zDTX6+Z9uBx|7O_Yaz6W7D*XnJ#H&B
zC-OEjYWt8@(tiUj_Y+k|SryHt?PYb>tX@gQ4Sl&}2ufh`y+v4^UP&v6rE%B{=HkXn
z)21y4kI`^V0>1?S>h@x7E!#SI(wEHyaQcOw-TMxcS+@qc^SVz9?}?6WeFk!}*$ml2
z|AItM+O8zbkW&wbyEng~arb{)it_#IS42?=pF=dy$lKeOyt*yCzDcA~(48VvM#_Q1
z@z(CuLN2L=@}`{MiH_@Pvi?=$vl3(2bPYey*edUdSNGd!uI^e=WfCXA5=>qRA_R?y
z6r#DNQnV*$J+(r4<C-z+q^&*Q>5T9G+<JZe7OdEWjkSA-inm((TviDPMyt4kn|dTe
z$LOlzs+0_OIQC<puCF|_BDUVX+f!})KShtqfggrpz#90Ip`{V`{Y4+WPxbdyrAa!0
zMZ6R(tu%LWkAvchUmwm!|N8?7T17rVh=>SDjgcOf8&FG5d6SMD^30kRa*iC=1F5X6
zC6Q&NNb1~pO^^&SQLFp>7*wVjLg+iV8LhPAe)YSiac-`{eT+X;0rdRdQ)PW)klS#c
z7^Tj>+uGLbXYIl+Sb*&mXESuQ@CT&{+SQ9rw+@}WLwy|VWWSm)`v9(bm-;hoUGzC$
zTH?y>BfqKtLi;Cm?fB^$vgGWSA56zp=~A!+vC@oGHJ4L80nMCyBkA3SbZCkgn-`xF
zPz|iPn2|Xhh^mkiH;Zt)3OBcrH3>^fWKHC_l3iS<(Q(;?q_^2c-Y^uWt#!JU%?RBT
z%-Zk!=oqPo<8A$JA%fD3S1q*Qo#?m|%=>G!($;C4LU(hezZ$Moy}8;rd8m}M{+_GX
z8^74Xf9ubd52MoD`=jF%_5O{xyYqZ{_~Ep?-Kb~lB-bc7=R^0=^0qead&(&qEo+xw
zM<S|$+hyUfOyK)_DTyT|ZjPDb5-hh|R0@f#E2^0o1-GG66Dps)>>Y6)wz5&cw!B0)
zdZ_Zt7TM-`8U6Zd#74IEh2XSIM+(hT{O<c7qUF}IZo{&@xO&Fm25Y-{^8&Gg5$bMU
zp<bklYNi^kHkX!@m!HydQ<?q0iie)j#s>dR;AF2or{!%!+WL7HkK@Vruf9(M@0X4(
z46ZZ*(T%Pi&TfXQE|6dQ1-NnE`d9txp(ZKHVO=<;o#!(I36M3h<dtJuNZBgR3K`aw
zl9PMv6iN|dglcGdClc14Uv4dy;JMuHd0Ojpm0wPn>BfWV`mSqOf*{4{n5^yDaze+O
zT+?SqIN4LW@vLtYmyOSPy8YYB0i8b7XYUuxvss`ep0D1crJx?VSZS^5S1oP8qeuO*
z3-go<zuUI_DMdF7{1^-a*1#tQ+s%WA7v*B74go6xyJ{h6X7!vYHLuJ&zQO(tV#ygx
zL`#4$v1s6TFTUd!&z|u4m!I?M)xynk%Xcrn=WtBCeDxgFLYx%l71^!K6!sC;Y{Fv7
zw)vk&m_4d(f+th7DnTm36hbrHRmNHqqs9~Ckb8FsO?WnSunQ@YjhC815Znuz0Dk^z
zYJ^8Bn(&|-!>!76pXMn#$IaQ58#kH>5JI&0|5Dwq-AF)2N+76Jno%%!c3jaf(GRVp
z87lgpYOkt-j`?x5Sf#VszGtj`Flu!b!+F@+&lNvbK(PC31)sH;(b@7aSX!j&ny{$l
ziK5GK@E}~V)B5xh?eD1&QbdR(&D<<2|M15@@Y%&BUtC|a9&Wi=j(q=W;dsm}YXT{R
zIT0?D&#w2}91=H&6}&2Eh&6lryPZKwFr%{_&bqMciXdvCu{E4;N`7M65z`4Bcc0ZH
zw)VNWpsAh%?g!AZ4dF(|;*}!(nRJSa2!t3B35|{|j3{G4$hH&37!fTj`Cy80%w(O(
zgj40#jSDDfwsYzbLX9Q1{jIh2%|h{@<G0~G(6YMXtFUC7rb~E`mZi?cc@dQ!C$wxv
zlMq2CMLZAIYYG=7@%;6HDMVzyC$7qrS6<%Sa6Bed6A}ut&&+Y<dM~^>DtS#T$F2Kr
zk9nsd2@;4Qn%6KjXQX)-1r?^^o=94llX)_fXxG{y*=G02mG_F}#*+|DY1d$TLh-%E
zQ~y-D)NC|d!dqxrHoH>2-@?Jno*bM;1X`l!ilW60h&82v5Mn^3kn`#adL|P{MB~}T
z517>crdE-y@!a=J``gspQ(8X1legUcQ^$Ma$zu7;a<M}t`^>%Mu{FDvL@HbUvTCe(
zXdWzl-NW<l`R-*}*{ut=$@4bucFY%<i;_8JASn`(hv{Z!QTF1YY`H+3epvq-?H5F|
z8E@R1tGK*cQ9Ms&DA{Qwb~YU|=p|?}x6!l`gYiAp6)2ugOq={W`IuVePa2&fCS!<#
zmg-rx8OO}XBt$nj<(j9q_DS3jhtw2iWKBqjNSJIU6yh{dvQj*my96jMq-V{R+&#_+
z;wf5cM!&92)u{LADR^K5@FxxoJO;ynHSqDE-~8_EVdpzz6|nqABDhO?OPE@dBzE^c
zZhzI&pQinpa`x(Q%V$sa#5f}%L9m<-DKDPT(kg6-nwQWN0=p2&#oUIsWXWfJeOu$&
z_jdo?%exbi8YbCziu)kd#E`XHQUzJryg%?<6<t2DHqtK01W`W`tHz>Dkry?6MXN_P
zoSM<PsjDZ<A}1I7r8V5bzVB)d+1FA>;an){|LQDR?F6g-UL}P$D^N5;W3LgM8ZF+H
z<?5T>`;nub?moC&AEM(k>r-`051wDa=DFrqN$zP-PYrD?ypx(zFAm+jVF}2S77ok8
z#k>OvgcyiXEFVt|SH%QMtB`0Wm{=csvS#!7oA36$jlo7s?H!1FsdQ^ps*z^%j;Iek
zq2qRkhb6h<S(jHeYWIiacj&l!LY?H{ssb|Ypmw7wY5R;li$=rud9<i{xP^xjcBK+G
zTCQVS_a#;Q$iMTz>ISh>nLZ>P|J2puA56>V7fGex;JK0ni5|XN!+JVeK6$2`{PJK!
z>;J0JOE7Pcq{@;q#K0T^5`-AGaG=R^J4!4lkrH<tmX$>drM&sIZ|i&Sy>a4+wQ06g
z-``Mwul#>=b+Y8mZGY2kJ$JO!bFWJHf*Gt!JrHub?s;fGzty`-PA=&}f$KBs2@`#F
z!_I32&rumdtLj{pYw2_PNcRI>F3JaFpst<wqSA65iH^4GSU>Q{pETYZ9iKf@bbII~
z5qGrc<A1u7pUT_!e?#LTAY?79rEq&#*u|MyW<rd_DH5X{(1u_J8N6cgE=HE3t)k(h
zaL;vjJdP`I_<HI{xyzB)^-BA3sj&gl4yoI|ARX5_JGqwb-%#JfR+Paeccatlp}F--
zP6oKrajPO4>Vf&WSM^4pn~db#UVQ%Rb<RM0ue+X_-v#G}_50A%a;H5E3^WV_*1)F;
zt=n%zWMZ24tT~bLS_6K31I*|>+uxmyuxBV*?^9HRKTVJ_vn+{!`1UnVu6D%fl4(xN
z^GZaCs$5=N0fkAANHoub90PHeD0yK~tIE`ePrQ5IyK=-@EBl<f^wu&&rFd|HC^=iQ
z*;;%lXFvjTw5mcHgN0V)%#v0wS#Qq^Vx_$U!W=A*&NNMkh$U^!g|wu`%%>2LNm?S`
zCIpC8;EOT%X|X3{N4DHT-8@WKS@sm%m{tqkc(p|>ZuIc=(0I7mb4y7rR(}&QItAek
z+Dr4+Izcazc{3k9aVK5-w4tZU)em|En3sV>B9r~>F3^<Fw373d_o!E%-blxH-hfrf
zkqn+JI!R$s<;CqS``ra9GrN7^NemngnIJ21o)CfkRFK(d=z_=|A&VtHB~F$jM)Mtg
z>@ldhGh&FQ;&yc<QZ%#Jlug)e3D)nSV_T(x$y_NR1~i!$qZP6!t*ThyE**yuJqM3c
zvXcFTIY!ILQ$$*#T#B@rh$4Qv96Zb^vWdo<x<f8@t;F`cKroUGSXH)S)z+-&f`yAt
zRf*P`{JA_}VfvwTe9xoqeE6vo>Lc(TXc-{R6EQ@lH4-%v_j@!aG^KlJ+3?u6TIDOt
zo`outT7eYBD)+rOCZ6mf)6VFrD#02e*OwPaER^XLIqZ27l;!xG<Kc$ncBE#n`GLbA
za_Xtr<vgYYGe%Dyg}kOr5hw(sEfI1J)$z3)vlnBDf@V*If8t&Y6f-`@Id~z@n*YU&
z4(3_4(Xzd>dU}ah`*QCpEZ=0D0un5rRj)Rhcc>X7jd}y~6s?k?SdLGE5U{zYm?1NU
zXkM29Dz*+NKUn^;if64^+kpGtSU<b(c<(2V_e9531B+X1q!tD8;y9({5j}tXsj>92
z#8fNt9oB{4fBl@#ub=YE&z~^QnSF-+t`I|EH%%a!EJ~h~b+5cSE*x^Q>f`6Hdnc=h
z@v5p-G-<INo*CU_v>A4^6pC0rtRlwOEIb-Zbs=Kj&{A_~VIHJzblb?;8W`hdhpMhv
zfoL9_=3dnd=G1FM&}7Z8rY1z0rf7$97#(lk%^MH(axd*58m~rZ#&-8!_lw2bQ&c@&
zi{85i9tC~N%DB&?{bxwd18>JLU=4g4&?cFnm}_keGt<0dlF0=zOX7Z-R#3Y8^;G70
z=vCc2wR3y+>>Vs5ULO~}d+~}?3cGpY>S87U$5POekz^rHK>~9Sb|DfZyC9bzZo@-N
zw|#Ej492_s^v5}&x%Qrl4}GZM={R;TDt@XG{k{;)8%wq%>Wz^}JqeQTS<d{OxvJXf
zYzV=XM?bl#9wSPt$S88g2(`A`xx0s+oE$cJ@5INUPa|r%8n(PS?ytfo(w-0So0Z@G
z#QE-d(7+FeZgU+1Q=FN?gqKe+AxZ4V4%oNcP;aJVZBuG;WbTA<s~HA`qQdREa9k5>
zGDCF?k&E5zr_N@SntbhvDzog^hs1&8_ThUp-t+e8-R~?nU-b$s&G?M>EK$g+Ev%>-
ze;=e{#b`2ArL~-RXDOaU8EuBFo{ps%bnD5j?=QHqD+sw0w430K`Z##lhk8kH-4uap
zV3Mxd+bJzO9sBRS5uq<YQU87z?@q_>q~C0fIJrl*kAB9`=A^75#A#}@jEBD6OUre2
zK4{I-ZdzUWYTwkP<b_;TUc7ikxPt3=2^pkh_H)F`2-sdzycU??=JK%R@o5Bbw+Z(y
z6xsf<f)3S4EM!;kOyQ;Fcgl@ab<3|41zJ^6dyXiCXd&^rc!KrjNvRN|^UACyug%NJ
zD9Pwu1vfHRe!T_$wBh$(Gnjm#<xVnW>Bb-gGhx}~b%*iMaQUR&<d(eZ30j|dPTP;g
z$vqu0{~%gEf42VECw%ItMipZMF-8ItF$6zYNMtXN^oB<Obo1>0kvE$Z1vApVJ{;Li
z7c5I5uLaG8D8}YWE{GID6egM2PkRnaLAE-9ZJnw;^t4?MERuFML?=tfN*<-<!Yy(t
zcZl~*YZZnnE7Z2Id6yTbsetq6aw_*$zH7#=GZ|jCUqT};-E!Er|M*^}kbOh0eEht_
zS}dX@*!HvSgY|_~mGE3(l`+d>(m1>YN=N_i@+^4^l~;BlZ<7@UekvFStbtDmn*tT%
z#60b|*k2Neq=TLErlMW0lkng(-i8*erdzV<Dl9H!UL0=thu?q0^|L2jKDpr8^~&Wg
zaI+S2N~|}j38wonQ(|V4g%zvZ(wMC?=Q~Xk?lx8JEXWkhj_bN`FIxLvmG159mQ1!5
z8EgWqn&L6h;_SCXI>Lkt(w@9Z8k5$I^ZPSnbS0*k&>SRC1g{9WG3=sU(pZ5g?sV;c
zZ?>gP>1={Q=X-1}(n|7~=|dmt={()PZywycXbFRRp-igcFz}<`Yfn7i#%_N>3=^c{
ziko_A>*7^G&4{uIJZ~{(DAbVRcC6@Pno!6EUL6wG7b}NjAx!2i@Z|D>{ge@oMh_Yh
zSG+I6C6^JFbmN{RCfuC}&F82~trOIWC!STYTrkx`O7NT>QmWmD;_H2?%H}<E><8ag
z<xiEIbE{fd@Q|V48_ffBZ2N>Oz;UCuP(8Xl{JC1M>X2qXkqrR?laIA_Bvw0K->IJW
ztD|Gx!Y0%=O_j|M_hW_Thv@iy=&J9{>EV-wZbpy-AWWA#rZ8I`CC&HKa%aMl$I!As
zGhjI_TP~%{nhIb4@jJq^z*qaPysBR2av#}E(ehknGa7};*K}fLVkOfeXE#Q_*~<F4
zy@!@XEUdV=7pB{rK#eb@V6{(!s8uAZ)}m@X^7SsED#Z3(L`X$R>xvY6Ms`ED1m|#h
z<EO|LZqz7FfCw?TH=`%Rt6{p$2&S6?TZ+0(ZDM0|+MZ{JXTgjbxf;sr0ID13lon#`
z9%h=7<-`kNOAh{n&?)^?t^fVd^3NL^9RnbNX_|>LnV~mlQd%jVW7d|O5yYd0zCHJz
z?*sytAMN{B2clf?*~Lmu8BLi85MVi44y;KclL(hn<aJ00tMMeUgVQ40y1w_F_t7z!
zQMp!C>_U(W+3CZ(;`jGec!|N2w9|51ia_-Iw7F!&^GD@UD0wpj_F;xWs<F)fcG{L0
zHX|2G7NI6P5a*#aEoA}A(<u<+-Gt_<RzWyKqvd*tm0X`NZ!2~6A|Mh_H^kMPeA~44
z1N~UIG5*Iyp}gtoKYO11^T9A+4SX<sh;pViUb~iZJuC}B18G@V%RRYh9^HoR>;$SZ
z<tm=gPEQ0AAde~W?TgoxX`)DEO$(|kyGco8)=<zQ<QxfMW<Lp&9Ed=w;`_Pb>Y)>(
z{<7+UqL^Y$AjV)G668+5*QW5Qc#e#&m(A`ag2YIOkuc9p3K6BOmTbOdijE#)Fh|G*
z(ZV!Y=+T-po=e}2OvboloJg`ZuM98&e~Q91nYF76`m+<iP*45@DK*a!ehLUBswuf4
zni7wqsWS*-3|dPed-&9*EIc;h$vqeh@jaqgt{hW`P!-l}A*-?!)||*SOyMNI#yg@e
za;b`)x__621Rjg>`ndA^^@05^u$!Ue44RoIC1*$pv<SLFjC(FG1Fv7N<m_Hzl$P7#
zUfe^+Ci-v5VylR`38<TWrwX}mqT`kVMvO-H^Ny(~u@sJZB^3+dGb{65I*yYG$7$X2
ziPUfh3Fgt5#P*v~mDXZjA~BlseNwO?+P)BDAY#e4Z7S4~^auBp?cyOs6kR+gkzbp&
zdtV#iA1U;H06PBJ8?>bwJB`;F?IuX3_})j$La!`8@;wpI0G`xb5T?oYkk_g3a>=|n
zB!a+hC$3~GYfQ~MSqr2EqRi~)3l8ZP(d>PY<}Gz^o9_EuZ8wSTZ6p$*D5b28prz=E
zjU`lrGg4DWx}~`qe?2Ew2wd!Tgw;}`my*d^m@2VF-$^{YM=+YSuzuN$&Hxk-lMLis
zT44v=@E9XZQy@fjrz!g~hKX9AE=p)dvZBf)5rS~*vnoRK;ts{tVzrQ9EnSL(_eC#*
zmftTrI<~NO^KQ>YNu{usY?AAi2<q_<t~~A@ny&{X0a$ZkT?@yfGDTsYXA~tDJ77|^
zH60`n<4lwT6B9{Sqa9C6aoRjRT`Tv}v3TS_H7JL+X6wN8d7A{I<KRZKphgp+NS9tw
zn>X?lqDdcW{$hDM*J=FIs)?JHF}rMS9@R?nZ1Q$!E87ODoe8mkM2Jy{lkEldD@)U1
zbR2?(35(~LoPzmNCQa0^$+{^u<Cpe>)`vU4JI)Pi9|AvXI{q197_bIDIRrOm<eXVq
z5e?+Dx@+^<iNN_r_Tb*>>Yzej`N$-)8b+v%Kdw3P;^x5a@{)PpTcs?`Oi>|Nb(Vrq
zvJk?|6elLx5k2Aai3sox@_<r#N$<fF&`>zB-*s|g9g)g(`ymo&NpGhR30{pwF@{{A
zd5Tr$sIE!JV0o;PhenEXd2wcJ-v>`xyD^m-cMqF)n<$n;BLq`~s=(tD7Q{~OOQ?1?
zF<$TaG^&?{@pavDg4ww;TVUxUndjTwm48O)4bkt-HCt*GAA5ExR*GGc8Kw(W+0|3}
zc$3n5-*@dGwFarx0t?Ej$}weL-7dgnd70c2TMG;(QHob%ieYBIn~1ls2sv4PIH7qK
z=_8j{^&F69So5aJR?2)MbVEIny^oIjl^P{5#fV5m2+d2X?ptyX9ZxY3f>2VTX4ey`
zNprXR>?PU+=Ss)H6U5f>x6eG!o#Lg#Yf{_TA{0tBI(vEcQibO_^R?oUS?<})AMdaR
zKQJA?!;bP%;DgXI)m9Ub5Y3Cw*<dc^E-lx4-t@q8TK08fad>7W2`B{P6o<8Nn+mUw
znXA1r&q7=k4T{8UCnqYd5L({EdD<cQNK35fVNUO&pR3CZwSa-PVk29##XTPSN1hs>
zb))xs(1Q|1nBoK_dhWTc97&^PF@r<(teQL@T`DwMt}=}TQ-1AWXbb1xXxV<O&_Xp}
zh1TMLE6-|%?ij4UMXQ9QZNIgUV)6W1eoyn>?AF7(;6u~0cCYxz@geEhih5Y}-r{qX
ziytWbAhX38^STGoJ%V#q)}*X!vC!8kGL_&5B0?_0vSb94S))v5bm|Q`C+~esYdas%
zY9I>1=(2v?k8!t^o&300w0Sssh@X29TRIIvJy%!d=av6e8QiLN8V?Ua$XTKD{}P)Z
zOP%4N#hW3^=Pqottl8)-#*L2cP%1i4_Atjl&6?17FuF<1@631KJlr1?pAsDp{5T8)
z*1)HMw&B|^xg?S%at@YokjlWC3ihT*{WoHhog*+eBTL{wpb(3l)Tji6nJ9b0v_hFk
zIiq<+<!H%~r%OthK_XG2RYcMs#tJOl$`Y-Yx=B~T{zTG$;QsE%I$Fg@KPgIuqKQH>
zK_z0Y)E;vn=J>=!bWf^ahKP_+wX)&r%4P`FNL8`9WUGfcg<zwQv&~s_Z8^q>X7ix3
zSbl+NXa9(WAQW#uCDU?>-RbSX$BNBxmB=NNQbCB!I*~L}sM-48{mG~+Xmzn`1g1uG
zlb8pVU?-f%C37(cbTY$>il9q2Z=?tz%!poC`D`srVmW@6vQQ}I71$ftPU2u6My-G*
zZn3J$UQ`hGgqnP>NP#VUkh^s3-8SJfS;B0J7orf**yRE`qvIHac5Ey(C01&s(HODv
z<-G<{>DZFb2N5(|wh$AR{iN4(eXN>5EzCmMN{w5%gIW!6SEREG<-xnw=Lj44_)r}J
zjFykcvJeQAa6u~XiI&?(0a`6kWe+X%t}L0ut0N)5M)QG^W>kgU4vGfy3iG^YiWe+5
zFImei!H-HxpO;cket9FbWQ)qh^FmGLolr_4st}71ixR5G6IK|ZjhowiYV5sbU)#+S
zH`?OT770>Xytq5XrMLxmDekU83x(prg9mq~xI>{7cemmYoD{c{>%O1;&l@-o&b)!_
zXS2WA*_p5WZJ*WAX&#^JD~*EpAtg_Ocw|hWU%2rV&(GJ*9&|7yOVRA>kST|}tqV#?
za3a{u%k*#YyC@<+!l`EKn(A2iB0wwaiy6e}u^VuoMd}%C2`>WP_L4UuQxK$p-G>N_
zSW+v{th@-ueD0r;mLBrBt_4EEi7K!diJhE@CoKq-hD!A#3>cB5*m-b8y!YM>+nKuO
z?fvp?ev+9x&?ewPF?jIpBQjWIwG@)F%L|%k!X`$WzhBi?VkDe3YSLdGu9Cqdsu$3!
zBQW3D12tMT5Qq_!dpM--_@Q<CWB}6xu&1Wq&6k_je<*X~Bgxw%)8g+SB$#p$47tKR
zWbt0;>B#~7*8`Q$cN%RhS5s#4b(6U%xY5J8%A=#W&h7IOr;5;*hEn2%<X7awWGK9)
zAw<?)cZ7{HT@{AxZd+MXIXqc3<NgQNeqHwFTIr6ls18)3(>zwu$>&fH%hW>DOYahG
zx9*=VLSwd1-GhEb<SZ#>B~q)(+7HDF4AXIgxM^bKS)+9g0-9qe$<&+ccfQRe4W2Dk
zGE0?WK5lc>th|NAYbQT)VQ-7bVLCt^W|15+;QCB2f5Ft+P+BQ-D9=R9!~OcqMRw=-
zmxPSWprT#hDH5_A{??L}5A~f>-X{foQNjXQ^S<h<XR>n-`3>KHlvieQ290btF1_}P
z{@XK%W9#7{ErFDtP0W1jI)PxLo^ac4`p6(F7;}ThchYiElHRPYYlZB~#c)$8u_6E@
ztUtYX2CvRb?$Dw5$Z8_9r3JfQx0)v2i74<LY>nYH%J~OY$HD_+jnHZwYvGnpD@}x1
zHyZr<P92S>WVhUQQ}6k!3sNA>WSbtpU=1&U9caoy4Y)Wagjx#YP9ehl2CXPJUmT`!
zI#X4=*u6N^-HN5y>lQ9PT7)pAKIB!aMt3}Z^Um@6=~<$-LZ!xQe8=VyF5)@exlyA+
z5zDg5T;bW`C$*ajQST9LeDWAUZXNm?$5GvUZF@7=M4bGg-SDl3!zaVRznZ>hIPDvc
zu;5@ZLaA^8&#(Uze?R>Wuv|v_e}(9W|MKAfhHNZixcdHo;<l6%fn50Cpx{M7jsEAy
zzWy!vjVQ6-{7?AL{`Wrp--!EwIJF%A6X$B?hzjI?L)7m7U+n+CvuOWcI0iF0?XQu=
zj2<uH--x8S<$k20U@#y(NaF4KsM3gMg`pJY|G`4=#fU;)mCL<+4a=pp3nB_8)J1*g
zlt>t+Ix6|`m`FR9fuWBFa|T5n3+ec&Ems_uM`nu$Pck>Rlqp$BLLhW>e*t(G%^uhb
zCuRp--Sckjrf8+my8;<-<I>-AyDA{ep4e}X?=uhs0Y4|FI8tLWDh#-^1%;R4Fj$O*
zzeAVUt?9S%Yfh{gQOby$#%~PVeo{z`7F=Oc)a(j+n@cnFhQ`Nf`y{U~>yZ>*RkN|T
zwslxs2EML58gAYQ94SvHEK2^WGlHcak0B8j02Xfb?ISkvmr!74tP&1-78z-Cv4~Tk
z#7{4%&q-cM7Jx{mvq!?g8-dpZ%FG*27dH_YR5Jee7$`X=FH!^dEPbT(#qZPiFc_Mf
zn^7>8X5JA>;BCEe#lJSB!q;WTAF02*-(~gHo&1ILTI%o5BM}sz6!4?9$rI#dJLpnu
zA7{ekJp3#N-h5~eO<h<eB^F&Vx#+&Dcu8Cs85R=`>V4TidtrUK_YRuay#f@kg~nrI
zXvCB*-!m{46uEeM4)83^XDIG@8968;;Uj$rHN?Vtym{F>x{<V}z^C9Q8<bDqx_%;W
z@|F`fqxtQFLq_aG3m=%7xE-kp96<a7K6fzQjp7t7#e!`YIVBCN4DRmdOKed;H<>{s
zSBjPlHXkmd%l2Q)Q1-za;NTHnN1RToj-jzJ7Z4zET4L^o5{rw4kCiJ$kwXuxzIz<0
z`AqJt2SMhaF!{+AnL@@F`8qPHzfa9k`1%49-mrhT4`e**JE{qC{=yQ{FOD(QrbSP*
zo(|s6=1>mKHvEP%-&)It-OTRX(%Fizm4S?`fk9FIiVrvNBkbqxiJ-BUYxWn#=edn1
zvKQFd3piB=ThXu;beCabjM%EX?gz2+nwQhF+wRAWmzoY1>>Bxg5OqJI-rd=k7(Bn2
zx|||M70>b^WgQgv(uX{H>33oINviF5S!Q6tt3R`fZBldxjE`-Wd>LtGcapQWGOZU#
z`3eAVV>3tgId$>r2VRr3v2{HELeJgp#4YK8GUJ*qRmIX%m~6O^Har%bA5Pba6BOH0
zU&A%AP%T{rpHEV|ciQ*aeJ)^xHR^O__}fYh3c@sIqe||Z3nu=(Kj{}9p9ip=5RM;R
z-KfcTk}VnRATE=O_7xrfZqnyR4?`492PCO<+z^y&B}1{Nn3})~MdME2-lNP^QWFX}
z#w2FoyO1A9iJt|@CXz08CPag-dk=iP@zRTX`ceN1-}(IAyYnujO@c^aQ07};b^Bb$
zyv)4Z2Mu;$-5q^;d7f*2Tsv}RbPvuwcG9^v-6NiOM)+!nZ(IaDygc3PM2j!UR<pCE
zZ%h42|Ld-qpdkrINNS=7lKzLp{S@LhtSlDDvuBW;-xR4a%0U!@k2GImSLI4h_jR=6
zi8*4?GQ_0j3hl9Y<_OQaOUq65L7|XB3ZM#;UH)+SLzEEUfA59v`YFlIZ-Xvb*ZfFC
zf5n;WwR+{-a_68Md7<U2xOU)C(3xSi;VJvl37u$=n*^0Re5)naXz5<lk5>Fmusl){
z)@$3D$-RivRPUo@AZZRIS*pTdSiZvHfi9t>ONLO%HngP>E1V((e6-I7E{C%A3#EpX
zx9eh;3Z6uF|Gq)(e?gT}Q<JhQ@U-u41`6)(O_@J{J^bgfkAg8ILQ58aEz2e@Gys$W
z1+4}GZW|ugfFm7M4C+xTYNq1b>}IRU3*_4fo4bS)hW&<0k|_L!S2829d{3=nWcuq2
zJ@;CNOF6s#dfEG(pA?@Zgzq{GzC&LA;hNwC5viZ<ZVVb$H$nK9mJf9Q&}Bj-V?F}A
z;0Iu^bYbFdPB!@H>Z!||g0UQx)hg?q(^o8Nt5V`C?_1+anH%-6DT><hAY9<-E|MPx
zgAx@!GU>=N^9RSc!`-ko;h%9rUhq)@^GuvM^n+WnjD;JU_~!}a|7z)y?>V=tOMfe#
z#G6J$33rka46S5tdDBffhT*Ja+)&)(WLdvoQDl(2?v&o;aHd_TQ3p68T`lyR9B-MX
z^a`+pW>HAmzp_1b_~~Q2SB#{h#81rA{17|a5?(cwF<StKG-f^nb124?ekyh{OMK<N
z1R<kXOpy(sej_jl04qZhrL@w0h&FkOs(}WK*FcTk{5p0s_cs8E)Orbq@tX0JxFKTF
zEg4o8T2TC@yIaDKt{QaXzh_Vqa$X-j&K};93;q#6s7u+N<;&(v$}41rfLMo(1ht9d
znaSYL!}m;Y9kABEd@eEPlH*Bf;(=F(5HW;(!<UXg(i5l6qoSm0gH}3)5{a$o#I$S#
z#CBto5oblU=e?uiI`euQGkKAHysxOa^?unr6JTjpeP=UVXesS}frs@M?IstO@EX5$
z-Q*lrtFBJvG4Z@K&FL@LV3<PBxSp|vw%}(YKnvM66K?HMcQZhTvvEuS@UtKvb8c-X
zF9}fI=F%MDye71rsl=is)WoLMDY{!buNKxj5(x~5Ks4Bd$R!4+*;m9|XUaeO?zh|7
zOvjd&`xqu(?9iq=knt=TT2o1-R&{P%iJrVu<HbOc==bk^CiU<S8J^V4%tqIWLQFae
zuP|CU<i8RlV_Ir)+im_hsJf$zQMlgk5+V_Uf(LFs^j>E^`-ag!U|+-gy_?x1=X#6@
z#+G(aH+)bwqmS&X_Y}K20*4ofQ&CLb?*;EaKwm0uGXiMtuDym#d+$O2m#$jK6marg
z`D?>W!hKbf1~{sX>S;+O76s14<U4<=F~6X~^)xG?Wi$vdix|fpFe)F*9Ne+6^{Tvf
z5A@G1L00EK0DMx$v98pzso2&bIXf&WSxA{y;{HQoPxLhk@bCo17K)-y6ktQ2u8$`S
zv7pSg3z2>!iP1+>9Irrc6EPpyfx0zXY2FYkpWQINkgoYdFMPB=kzOa&|6)nw8l#L9
z^Hj^SudhFF8E@A5_ktCMsMt{QHB;yUpM5q|<y)XZ(6T2Eh6e?+sJvixgRjb9<6-?$
z0`uBzqdB=ArKr5bvF5*bWY|(Vwd=%W2HYxyHB6)OZ@keiMFoi>DNpk@`^?XqPQCp^
z+XJH@%nCytVD7<fghqI<Cf1PA`2*?)PQiCLCqmA@3Jz|wz~$FP%D+Iqpn=qdoQtcD
zaL?VVj3LH?R$SaY4ggJKYB`f7;@~fiNQwWk!wPyC&%q}9MXxZTsSoi>$q;Hdh9-Gp
zI*O&d`~b7Rc-Y)oE&pV}u57Lp_{masA%0ZhM;P7{`iR?klMSk$#a_mJqaR3dOXab8
zO!tU+sLS`D4|r4|3cU|z%p}&ZZ_XIZb!uMRw59xEv+0}dPxch1fg!as%W+L7Sp9bQ
zY0@=lEPG<L=VzvMK8h8!4q{91c8OUAk&oPVeEde*s*^Jfwuf3LW9sZYNb!We-N$<z
zJ2tK|U12?R4*%)C-%=E9@7g=}eko`yf_)iS$i~Wu4L{o_Bc4%`vn$i4Ni#jg+rH7B
zcF<PZ>_4&pOCS9o;`J+rL@jdHjN_l{a*c%zJ0ZM>VNrK0JOn0CMX7oH7M!L3*k6#X
z?^$9Htno_&@R^WjWe-8z7Ej{VyHEn+rJ?Y@x^s`>lkpo~u#XUjZ`N<z+b<=1R5>!j
zycmdIJ?<Q=Mp#xOY$`I4&ZvmibB%j)iX$s3n6KMK;5>D4!zUL%7ACsQ$1oMpwBV{(
zjW5@LD0KyPEhkeP=M0LgcSKd{F-Cr=rr2_Ib91qINP3|Y)isEHG3oKH(XHov!Nitc
z1aj-Ni)ci}uc_iaCgadn4yhzQf{MnZ2hS@wR0-4F<rkd~c52bnuLpksQ@{gnDYot=
z_!xZ+7SuZdJ;}!1g%%>|l?-`meVAll0Mr}kWX-2g#`-2>6wF>=GCx9fAOzR!6<4|4
zZ;PA4j<$}S5%#XHQNfhwmzPn$Hhb=$C<It!jk~)}VjB5mk}!T^T##=*bNUdFEeP0|
z|Mi}48ORlG5A3TEdkNV%VPgzOo>-U@=2NG`M=V?d?766x?D3h}eveB@xo7-&H3vZU
zt3+w!=5Y-&MZ#JUhNnHy%+`8*!(-lnQnHrwj<vhZw|CQpvh(p%^Wdi=Umj!##kdsa
zabG_v^-m1>Q`}X9i_g&GewqQ{Uf*A#95#2dTj7(g2DX<aMcemRO|z8}bL}R4Ol}!-
zUO6qCQt(521WM;q@2>rXmE~na&rXh-LK5$W)Zm!U74+b?5PqUZS84z*?rZ2qB#3rg
z(EY!Rpi5xjFW(MJqj#%aU7J3_ZjbkIt8B8<C)09rvmL!wk)!x&yz%Q=y=7y**ksLL
zIvxKmpGk=4p96!QfVcGBx2P|-4?*X(>oG)i$VW`!?Zk(3P7_+I0Em*(Fb#kj0;rsZ
z1(>%?ae5!7oeFh7XA5tmiR%6LS`YvJo>{93?ZOG*LB;ihdr5}FInw9G^<yUJ+>?GI
z=jG{l@i61IU_do@)x5uX#1t@43MGn9XZ!tEDAEiY{*y;>X;URcE4}Hhj<CbnTVd~I
z+Ebo-%v#T%UL;RDp{<_1B-Vt(Dw3@J$Al9rMSES$)bx4SYN1a9w8U_*iO5|Dd-rX~
z#`871+>xZdPJZC;xi<6ku*_cwq1vbM71i`sNqMU==Aob{<yyZ!hl{71iCNpgV00C`
z*;c1nWhaOei{{b<H}F~rEP#v8`i85Nwv0Gr5PeyfhIN(PS)fvOJ}ZMdkC*<-oRdFm
zXm$3JaPqQOR@t@t%c<8lF&t3|!P7ynk1tTL(R<d+1=LANG^w%gE@HBKowRX*pUB!N
zE6x$AkLn-*Tjq0~IUxF>rb<^5!^M*mzH^z_eRpDVu>*^mfQpzKy({$GbZfRw>`C2a
ztZ}x6tbA%ENVI<`Vvwt+X|UuK43rhl0BcR=A~YQo8Fw1+qqxdk7y->(&HJeRUt;*C
zv<41t=f>>82!_ICwfodHI>pZ%WJLZ{8Obb@Ff8+pfB>Bsy>i{^;*QzqwuIBSw4<ti
z<>0cKj0z$3?eAs?NFn%)^3w8Y^wMh5G<XW)a67SqJ}&$2Rjs|Z@JkH+>bmQj^Bplv
zR_`AXIW(*<uLR$j`Bj*>Z4KRu9_mJi6=G-kG|F4&+EDpaQ56}@?5)-?774&9k=g?+
z)`3bhp%I`>Du{E_Ue;XOo+Jn7KSz&a8B*--rwflPcc&jVE3#O4%yisgm=ftFkrPnS
zU9lJ7vj!VGTkG;k%NTYBQoF4rf#}OIWBbaIWIi(xZ3LArEFPPd-R4<b_-<Uu85`yn
z)H^o^K5E&nM}Z!*jsEE%GOI`4QWe)`_(_eYI75YCZdxk)C}U+F(HZ!;z7hc0TPv*`
zv;^p64reZ0^Z>E{D+^z%rTD*8SZ7^Tmb!a3RZTxvqkq_c+~$mFI<--b0JV7HaBtq%
zD89<}_F7?8{&u->By>Sj@ROwGLy~y&%<DrwBMRiX6C>|?*7}qA`o?zSb(0n6ohR=n
zUS5JPttDhDER?#_10&JcUAL=mbh~e1%*NgRy(dQtE}&jvbZD5;n?AU6H#e$DM|Zin
zS1{KnePM|X%rs^3And`q3&vTgIBz!oD)aVdsP*WSa8)8!o?K?d@(LttMg$l|1nH<R
zmKsJ#`|`<3pJ*-xE_mf{L>g<Jr5LNWC%MCuHXA6!$eJHc89E!sN<3DFOzD}8AF&I}
z)4UUGqgZQJ$pj(by~@fNzr5`vuV(t}xR3_t@w1w3T<Hk4#WQbzuYfVJr;!b=K}^ix
z%#K8zt6}-RB>zYv%fxR{^s!!T1wN<^iOP%J&gUB;&C{H`E_bIl<zebv>a%Vnc<GUj
z=Thye9Nj-LELcgwvh$ur+5HQ38XJ6pd>(Id_*hrIH0+s9oh+3!_}@Fhb#c9fDq0B5
z{x!C6p5$P$F7*G=c@Uu`-osU^2L-fn>Pu23zoBhymb&60Zc3`<XNMug!pVT5RUS9=
zd6uJ!j_o$sZypdOr_iJ;{n6rb^}-rx%P-G-H~F323XJ>Ik|c0*Xxt#xY-9Ip{S}X6
zxPlfR;PYC6l)lPL^GscSC-DYOuKCHq9?fd)GaHmo3W=)63EC3qHW2i0At)BocX8dr
z)W_r#CezZ^A~lC!#H<afA+b=DDD*L2<|b~G-QpA`;iG6=*KN!37Vd1>qgFb`GVD)h
zfler_F{@ns`xux?Jc;OFFg|cM#_GUbD*RwelAiCrS?@UHI1T|T$8z3~;5;(?mub<%
zh#C(n>S?=9gUr6=Z)wS7fgb;?$=(OG8?~lRMU{ilHql2!zBd}Bs#l&?YsS%f{!VVB
z<sw3oKyqz5+fq*}?wLzc+&iEzD`(}&EybtjCP-2L%4KC~>Sy47MuC)CKT+KhH$?;-
zXYww5MJ(ze)cb}?*KE-&a-S8jr@$Dt>$4MmRPxCGvKu5VADOxlI8l*0(b@_MgDA}_
zjn<|{PSl8g+U-kB))`?9w=`Wo!5l6AcOB(ScEk^~#10r0@E)?~r^+8!e*5qxG_eYn
z9Km4_g<?f$hH3bxUpj(ZhLclS$jg1tOvOdjZ9J3%Z(AvbyVgaRTrbtvtl=luOs~q}
zmly+;LP~;*+8_CUc3?S*kRk)?Je6zukhoy*m)s(Ln>Gljn4h-hZ(B*@hq+ZrR+u2;
zKn!=Iq!soI=9CRLgDdN(#$OmLgfS!4%!x2z{NQ%fXt#6bLWD=jF-S{A@IyRos8E$b
z3Q31JRk*WlZ{%pAr4=+r@q9s{mx^q!uNv#M)qxgr+S=v+JCO){6r2^eQ%_0tiMGMc
z6h@AV0*CmlXo~#+KWr{NH<;vOa3i(m+FQ-{D1D<QSK+FQ6JjeCfqRV5Xy=O%uYA`S
zVmVUK2hdt7`jd&^sK@8LCHpn_Wn{oMU3R8^>I*Mjw64kNq+7(Os}&Yz6h5~Sr<<pz
zbj(0v`~E0FepMVId#9Q`7z{~T3|{T-Bg6?<YV<T@6uH<Hf^)gm=HQb)AM6M4OTz)w
z!o5OM^+9FEk9j{n!C%|znZ1Sc>E+7VDmY~qD~$Nof8Jehz0<v55M*=mKX+<8DK}0%
z1D$_7e=pcIqf+%dA{@$il0SxFU2hWu0MJb+GasA%g<Xb#xV~Gx4P#+zQkk);owm!m
zHmuZIf5G)IoV7q!V3sgLqo0i?Rp^Xw!8SBJzRh_CKHG~1{l}+sax#XWG;h=ve_~{{
z!Az!(M@~pT8nH$&woE&zcEf1tLznK0Bh{zTH-{WW``^e4WcCXE{*#2rnfJ$xI;K{`
zuhoOUZYT?IVM-y@{gvU86R5+Fxo)IPw$o^Q?3GS<A6wHV%@lEAUMrA_A1^<uv7Yuy
zkC4@8sxsUf^tUAJ|J2&0+D}5Hxu>jlqZPg)X^AerW0^aV0OSahjrWc7`^`F@{IdYh
zm+FP=Wk2SlX%)&Nm8XB2Ly6+d@fPH|00}MyyR-7MBUE8i?hehIQ^4!(VDg-sePdQw
zUa!3&Xa)Ab9qpx0=hC|2?&&&|G@|I_?4;|2QyCqG+a$y#hk3jeqsm}M&s^1~mMQ`_
zi~wm=53GOoY_?zN_(vb7Ra<_uQqop5<tiaEPm_y~G42;uo;9R9S^M2+z-X;?vUP6>
z`0?<<JJIA=^>5v2<?`9%{juzqLASs!A|wqrb~n|r7H88JdNWJ|RJv|ExUdZTVi0va
zW5$NuAbn6bz2LkcbS{4KtyvJvUuBTDu(fS9x+iIb8PO@8RR41{d~y9FE9pj&fk~I<
z2ggA6%}*D($P9y0C`kVx?CoU}#dXC)WpiBN_Q}2Arcs(|fbSG!M}Q|*KBAXNdK~T$
z#(nUPbd~v63?US-8}lRRoyQvrZdUK@dl~(%?j;e{1zvOG&oJ<NnGqZ6)PAq!pzf(%
zjiO!k{=TXxalE~jPpJrmiJ6>Sx#K~HcA=jR|Mku0nvEWCuVKxkZ12mYn;7e}V!_I|
za#_i4k^y;ry}yMI4;i-g?+Uh>qmSq22e(z9lY&;@c~!AZbV5)EY19Fc{58PyaU+$7
ze1DmTVQcC&yKn8FSTZ;`NyPG@c)BitLOQMF&wft8@0k&LNW+uMUzZl?=qb+exLA7h
z;?Eiibe7zwnm~rB`q%~Zze86`jbcwNDx7Yckq6(2I}%_+v}yihMA;#}45bVN(Z6ku
zO4;^g3OrX;Zuacbj%~G8)~6<xdL~}~S7h{>680?8Xl^(+n7<3rxxs4q{0~;U?UgrU
zy$-)2+m?Hx{TK9yjw1(i7{l4(bS?tVzOu8M`;~?{1LW(4Su=gnr;!T?520HvZi=qr
zJrcDxp3U#oHLc3Cs$XVrHOliX#7-=Pr8XXunk|_p41ke^2S(|~G0a-hRkAUCy91~%
zBQFC%gEP_0&(F>;B7-Z)HO8Id49Aj`{X;50OxTj8F>EJDlMj*R&1DwMTBT?xUF$oD
zPO$<KBQ8uuqhVIUFp>(8!lr4){LP6PFS-mRp`~&pFSCH6;&>+xcROg%BcN9rqY!to
za~Y5t8Ba{pE+md8w3PA<&7i24`{U72-_rfplmZkr0~#unL?UfgDKXmeMshLQ$JWDJ
znz)bjK|CIU!(UQ!t8b-l#(#y~XY7?=$x&$vin<6_2B`}cCzLcxx4V`i)qgGObAR%G
zU`q`tP66Rx&wyWGIWO*Pd+9+pV$b&TmbC1$#kL6EMIosi$*`+L64wZLEpYE1hAEv7
z>MPZf#O9>bZ!2tNv$202i31CY%QtW9d4-zi>!<+FFj^R`w+KPYM^j*mz=8)(T+P^b
zy-C<u>ynu_a6>2$?}a5Ode<M@j`~_Yp-*4X?FH%I>e}2nLO+VdKvKSaU~RwrSWuhe
zU1`60-IBN$B7?nu$2jH2Cdo8hk!ipqVX@`bWd^mzVm~+gy7$B(J+ZWex*?a@>L=s(
z5l-<d>BuJlS0D0VU@S6i&XFP;va6zCH#f<|8bp0v@ndk3bMo>_ii>`$Dv<rBHMIM?
zcY!$&uXSje?8(W?^za>K?4K~-nuh-4*H?CHkZmTFU_{aBW~@6OmK8(4eI!^<CaI)b
zq>O3yuz+5;?fA7nil{+M=vOVm&fs`ovHDs_7eD+lO6>8OLjGvDjqTXm@46iuVlJbR
zN2>I0ZRA~KE<!q-oUURokRO!Ri@rUizxrilPr8!{COlziWQZTFs?PUezVGCO)vt96
zc8S@pBNXDNZ3QS-y&^Rg07OY@!+musu-1`?27WtZpYV6)B}<-%#!#UY5-iI+y+1nc
z0sbG}SBY$Mx6O^3h-aNxo5sZRJa}n1U6px}G1hC;Fyre?%fxnQvL73Me<B#+`DxUi
zoi1P67lIkq`mYa?^Vk~yUUHFd-!ILd(DcmybZ8&J;l%E;zWg3A16Iuil`xbbml3&f
zP5_cH)UgmB0ZoY%0rzD3H@m#>_Ro2MVgi&&2E*g)9bSXc_K)A<S?S-eiY#^4r2aO{
z4&bA++Hfr4-Y*{`IxSGX^E`7RPHXV_gV6dcPFwL0^-3h@??b@^^hfxdCc~#k4TPG&
zS)m%Up5-NIX+Y@MbJ<~w@iDstx~H?l_95Z*XV+c3g6(_DY=Z*RrGwATn^+VXVCC0k
z^s!I(fu+s9gGj11tIB`}kc>MWbL9x0f~lBn@n_BfOf`7v@<bA5=~s|?NNtX_m#wVa
zHNTcfeC+Ra@ZeOPhr3^K(47TB<g{IvQYaKRjoLJ7G%4KtXz#FYS8T~<HYR>LIRX8!
zg)jdDH_XNR)6>ul=Of6b)%g6AP5R)p+dmHKEc(=wbTL|egp%chtqcaFOB^Xl>iQ_b
z+&C7(%wGL;Mt401Ki>vjh8ONWXTpL*H@pMo(B1aA?=**TMMH%=D(TJDnwz!Bz>bdH
z?1G|>zll_=;oi-$AtMJ2675mJCt6qHLw(fd7&|N<fz_93v(V`<l80xOe~LeDR0i9I
z{6$otXW1{uhZ>?mUSlU?_CzeByssYKr+=}aR`>7OKWJT6jgRbuu5~2VN*#8FFCj_?
z>uvS)uVl`cd2p-7Xg4*UJ|18{E^OR^xmScCQ`WY7Z`nsn&`IbKQ}&ctHu~Pd!|(a5
z2T1Z?ewLUTVY%)-^wvb14KnV`gbk^`blkhg#muHvid`!kERqF!Ij{JkZSFSn#t%6b
zoa`8)L;(8bg?kHY>3~_VH*_e9(+s-)LFZD*bPN?_5-E<E+GC4zrhbYyed{6%nI1r3
z`<bCxS&1Zp${Q}&cW+SZ{?HsalXyZ87a9cTO??o_3A)yE#~vTLL<$J>97x1&2T@1I
zW5y%5yY!B^dF*|}WArMI#jI`IT6nU3ucXVMuM#@UM7na=!Lx9wLU`hvH6|xrNF$Mi
z#7)gDfLVbRPPRiBGO<KJH!_zu>;z5s5?VaLB=#%zl%;G~e3-$Nv9-@jUd6mj+erb=
z_MvO94n`x&>NAb#`-zEv8P{irw$5V)1Q@YkfxPWkN=Y0*$rvxQmdf<R7Y~mP%#i^0
zGug&3Hy$yIL%Y@sK%eV>QD<Gx$UiaKNLv@rZjXlmNO)8(idKZQ!U4k>+%@m`154RS
z;e<@_&cx0VC<@jQfj-9fP+<tze!QR<+b2vl0mHR^vJ{0tf>*VEcW`nfGM;|VvOXTI
zGRlJBBC6n<5<_qZ<G*%XJzDRhon$bpv#hl)w%Kl@evO!j;E%X)2=Jik;nB4rrdbW+
zW|9kudM}K3;u=suV!goYr^Qu}kVC*q!3{tK3Dy;lMt{%6r<Bo-0)f`pPU|c{IFhv;
zdaU<{5~#N#7Z#Wlp%7wq$33zo<;Gm;Wgq+y+S;=ij+_s?Vj1ANRvR=C&NHfiY$u&K
zg6lF*4XUcn@9pH<1M-Ncyx7ahjhn(+(aJm=!eiT}eYjxZg}og+T6O-Asi+!UaYi!b
zg+QOHG05}7;7r|hHxGFeiJ|1Tplh9=p_-rz+MpZu7xV^#uE$IFy*02=#O$v*+l)W2
zL#3F~x7+G>A{ob$8A5pgPJDv%NxwF429qY*ON#=XP0G>O*Nrgj&{r6l1yQ|-iIv*Z
z-^c!lW)dLgDh3b^;7m9?vp8p0z3uf;9(&s{RoC=gs&w6H{uG${3FlGu|2#RrPhu@B
z;PBDy<J-)<Js=48RVIXdY42oAyXhFsao1ts3szlWJi??;@rr!H7(;ClQ{y~iZs(dE
zn-+<tK?|{>?hC{^-oxnSTTV_^H!A!RsyZ5*Q`P4Y)9cQ?_Vcx(NJ6<M-zoI=XpOvi
z0;uvn)9s*yGSdTD*KG^8(y04(3mY2XHE<HGGfM6Myo32*Yes?MD#7uIAS6ty3}M(3
zP11<sn4(>7lMfRU{abXO9nL3Tl?SmGpGcOQaPFadHRstdM*XeZ!~q;4#Y23V$TH;5
zx0l`+e`$C3MZql?s5Zth5W2^Re=RioE4-_tYs+mt!7RU&7DYRh`j&eQ)D7jr{P;`D
zRx)8d`!J5^Olr^<hb%|FCN6K=m_xKzB!m}>4Y<}CsA1Q9v3}rFfki{5D>r60j8I4(
zb(b1I=}aa-G`afR05fCKqCX~MIF$Ebh4(XV`z0sNP9feZu1E59O@UcivTgGqaw2Lo
zQw@a9oPddN``+BmMEdvhkrE(6w3V#utD==QKWW*CV?_@_JUcN#J?$IEL@q8(MxFTM
zPGd#Cl|CNMq${xIX3BLXt5n2pB`IxTp9X-ZV?&ESkTk(ewDpy^$$FfMV=(gbN&KPB
zCO7o&@zRwB)X?z!@jOcttBBjp_brZ{?HM1vW9CvruFA<h%{k}A&UGBOv1=+A3SkCz
zRb(xOGogAyN&T}U_W9lzhg$My@&gmL8k97qnOKQUL{sR)DGeBAvjgXT-(2=Go}RB?
zA3im#+PqfOG(KbN$`OV_#@t&}af7273e|Ph>B_kjzk!-W6p^LBhv!Dh9^K!I$@`#)
z#n5{jtC645Y*9sXt(f9AAlEfoBtjs!`Uau}gHx9`%@gu41nZ0(q`~K$<P;=e5Z9@D
zTen^3vHDr_VQsL9=<?X+C}7GpN$<!yyRynn^!JM3(XOLb?&Xe!+M5Ru7c?d;p=j<+
zytb-rF2+)479CFM&lm3jhI>(A@xiCS8yi6<u_vD_U%D{dQH)GP$1olr&xcD{iHZYH
z_#f~O{vU}+ObAK^c|sx;eWC86eGH^o3u6qU(CP8#omXB#$v(xOBtI1$k#Ys?dEPug
zgp`m4lDd3hqCW*U3Z5Bmu>iiO<QFu|J=-7nQT5mfMp<%B5u#HbhPbLR;0nUV#;ct^
z1e-QEhx$Ko!S}o$#fucVWZrM8vm2C6kg4gq8e@UVU*}K6lXEknv-|zFQQ)I*^fANR
z>nq$(0tWfo`j|#3V?ERNuD|EYx0oZs)8=ZCcg@6pu1IULbom=84*^2La?40+k?hui
zfk+On&qmg@GIrQ;81!w9&dpOlSNO(Yn!12vBVKSv1Y;np%*LP#R-ZEhae<J0Bnf({
zc}DlbzpdzgE+O(XUNz4XSHo7Ty}j!DARg`Z$7(rBVlh0M1m&os;jZkd{7Af=lYK0u
zm8<l_u-3<A`jJ(%A{CE|o6Op*J}LRI1{$#aW_(I$BCo-2afkY)LRGR^WED53WY(vd
z0VsYbt@VFnG&uc2hiHeHJTv<YYw4Y@WpX915nkEp670!l$Vy~mxTMcQTBLFssEFV?
z=E}P$<C$}R^;9@BIBapKKReWQtN_`6@rEy5-C$0`B>npXW7t-4vmGbe7RQ`=>aqCk
zwR|x{aT9c#gvGm3QI-x`vd#F%+P|gcf~NWSNj*bdLo|3N&j5!9o0T=s5g(i`Moj#N
z*t>mxcOx7~iQ92+C!V(!yoOup6c~53mdr~^^A)7z&3NKdKUr{(4l;(IP$%TqF{;As
z_4Pr20_&T~3!IC$9og5Dxt6YBluA?~qfR~edtVU-c#YFZV`V<lbLv<>vHKAMU`pJU
zQOR>GNG1W-G3?bf^4;CLxy-V}w~lWDy37467_)5XGSNJ^o^y~yKR_!39gW1C`Z1;(
ztGb_6WML1-SOu0k`;rY=(YDg@LJbaREK4>51Qmg$!-lp3azdK04INNDF5f`&hc|-E
zl8)XrmEEzB6xM9K&ttDH!3|k;-;=+ooBeI7B!z^jVhLf^R(v8%NG!Y0wX-&zyy;D<
zcnkqmm66#g<71jyuWP}=$?ec88!fN~Duz2hveqCL8y^G7^#9GsT0vpKhYZ5=<u3+?
zK`0alt$w)(&QI^npG0yr%XWwfHqo~v-2S`Q*iPxLAS;GmibcLI#vF>2`)ek~JpoXl
zN*!%m&UiETz<`122x_Q&vA!#}^9#vW7w+EV1J7aKC}VFL3%(=w?4>)i<<|`sf73^b
zOlvu98l0Ra`S(yb^TmSeuC7#7vM@r|;u}+Q5v4!@BNZcntN(N6YChhdS=GOyVutRi
ze%8>+C)v!|xfl=0@!>YzjCfpX32I5xvB*XXx7`nBml(!!=F^e*_@Q6ljbTwSKj5}T
z0QhNhPN<#uk`PQD?WOy1skP=Fw7~lRz_4PNACN0Pj$-PB+U?@JZb*|=G{+yOG@uo@
zcf6HKt?ka9^6feuE;vbUG=gz14G0HENk1k`fOE?;e~!S%aMgf*tcNL_Y-)%*;63t@
zuZ41?W+fS_cV3k|pdDgESXr)`cV(2>GZ}qpg>=C=$=#bfvnYMlZ(Z@B+D?VT@ofFl
zql}xJak$Wh(E6I%P~jD8VR8ZsWK~!%hdx<7IW1W?nL%7-=SrMTUl0C7D}i;T$r5^M
zsC2*d&*^g#L7gV_+V3M$L)AA&zUDx4D(WHOih8=tVn+NMp*}(9(e@tzB`eC=e2BQF
zQY~iw=#+l0-raxW#Gs2goyUtf6ZqvdZ;eU!HDjjgM@V2otGMclUPa03;i+t#LUk+=
zMXtF9Mhr^>O4OpDRDc7y{NUVs4H1M>3E;2#Db@l*yu>F*qZ>DascE-ntr}3-aJi;Q
z^yD%>VmA<NvH4i=k(bt&?-geBtj}+Geona-OlqLkkYpJev#qt+3ivd%H^nB+lustt
z`7*6BcfK|iSgl_jf8496X~9V7{198>DUBcMfOhgq`&5v)xNsD8?<?jCyv=88{twTI
zA*gwO=3o*1UsvO{`6|I%{~OP3Id&Fjz!6r<+lH_!D{i@onf@921yY5OAs^{stI$8x
zmF1O1xb<cUt5#Xb@BN>Mo-cK}_nJ+fLJ+gaj6`mE5AHrQ(X6?oQ>o{(dk<?fQO-vJ
z&}IK&%Y1+u%1vN_e(DNbiQ=dOZTwmf;fx5l!{SheL|Yo^B+j_G>Q|%VNYBNL!2S{X
zW1MaeRefJ!=fy4CA0l)v+=XGf%8WEh5_f!JHAf|~D#@&+Gvi}daLZ?rO~#;d6F58Z
z-mkjoytH_5;V#zQD5NRauav6HU$q8S3)oSgTBn78T#;!}riCUi5!ldJK~{8u1Pn^e
zj5#p<!u-7G#^EJ>uy@|JdO1JjZBP5Tg%1sxH*sdT=0ekhK0D7gcFT>b%aXpGom<+L
z6h^S6NPTr=GV6xAMCKHqpG_?$_B>-%|LGx@jfbIZ_4W|#39spU^ZlBLhau>e?^V;{
zN26x;Wa}J}?dS=y?Y%qf=krV2?z>B3!*RgFalJ2^yu!F@NT);;zWXMZjKcUTQlX@t
z_9e=PS>f4LhK=Rjqo>zkhmlWN7dvf+zMmHOq`PG8lX5OIbd4$ga26oM%P8>~EU8k%
z%w4}g_FJ*3kN3#@yj)(J^8ZAu81|t}*zmug1=i%8dn07XY3UJ7#jo0SAN_86q48o`
zNaVM@|9#6iKZbpF9xaNv76B(cK!twP-46hEGJpu4pkSA4`5N!P0lj4%PwwiMi!7`U
z?MSrV7U{5TtuU>#OozVumU!HPlvZWN1UILU`abRTX68+cAlP2f$`b-F@UL?*(Efg_
zbNZK8<0N;<$ube91~B@oUm12fF=staubUHrMj~T~g}e7VwtpzRHR-uN;5<pjAwp!w
zJpB|<l+M!>Uz@jUV#H64<4mNWX;<q$dZhHL0vm}C+<Z#7RN7Wq2wp#m8AUXNxo^QD
znYHp@>58}eaSMBT>}>{Xe|h#g7EXVbPo@e_H4xnsIvzGD<Z(>AVp^JZ0qbZ-*t-N8
zxC9V|Y1@|nL$WxxpGf{4X!-B-eBW>AmxqHrdyxg9yP#mTn6$ElJ?+@;aR?MrCS&g`
zy|7SqM#sBh|Isz<xJ#-|#7M^%%j~Xwx~-Nx&Ob0<Ogcshi<X(unn99*$V*i-LgiY6
z(;4!96h0oa>{(f%^L^}rk!&Mw1~;0?A}4EZgA<m+B-zl5gCD`^R@eR4Y$=nA+Ui>^
z{@NBgs)FVXLur)izvi;bftiv=STy_6;1H9R=b2_%E3e={@fk3)1AHom=g%VCx*w!$
z`fnliKHkS`OvL^rxiAaD9DY4weqg)Q>97Pz*>!Q%_cAuL3_}N9Vh*fLBxPo~@{!_9
z*L!R)ki(rCj(LS5*T>^oJH)W$7$L8q!xWCZ2b~R%qCRd<XI7*Tvy=qJ^%~rk#j{r3
zn@|DC^M3gIw|m~XyV0LrqqAziR@nOrVpdtl(mMmL$|f({CtX+c0>+uhZiF8Krd=P7
zU;l7nCz~nm@wL9ASX7|1Rcx%IQA9P}|Dd28sfN>T=pm#xIEK4+wJj=G-JLUYv4k(^
z=(3Qb{kMqQc<#m6$mnu+-^Asi9iiom{c_0uf(i8EHWkn{)*5hAA{5@4w}9zlx!1x<
zUnkk~67!@H`wV6F8gv+$eZ1*RQjOKw7jA8q$(iY%B>OO%chNw~n)8+C%w`TT<ePR_
zp9&1Xg%?6Cs3*d3S@oU>kThZtFS875V9sUu{)YZP1$#f%Tr;%$nd#K^owjpPLL_n@
z_WyGgGKRDWF)pjTfvJUuekGJCuYel`)NE2&VycjUndtKrKWc0cnB@IHopzmq^B)}6
zY*7XOrAMOERGL9y7&&1DHR?JM!jG9+bx97GJ`Y49RW3=sYVcET_Vz}uWAwjrR%>KA
zIQg`u5~uGic)EiBPn?0sZ(XHWs->mnuW$GM6`a_=yZUj)<T+~OeI#QZd5CK1sL7~C
z9z&8$buqRE5>3(#iKdUMJLRsDK0tslx%!{K$aOoA{gBXTsgFfrO2w&^+_DbfVHi>R
z0{3U&lHL-oTaor^7iqT*5B-zB)DgB9X&9PSixu`km-98_&S*78svw+3$uRtnC^K9i
zBRc<Q>`S4S$=H?U>0yY?;rv-T!FlACv7mPCx$=m2e1(~&`%1?rGUSsVer~ckq9tN{
z5aah<XX#oW2SPi&3%4~#0cz1HW$$|5<i~;^s<xhY$BISK`CnmY@b|XrL>x_Y0uTO_
zYX8yHf{F$Vvx`1t{fgG{hr<ZF?@C@`=x0#uB^69%9fsR+$80oSVH%Ex*qJc~kF22<
z^;E2cWpUY#Bc8Vg25Tz{EH7Y#%$`o;(}Ha)l=Fw5R$By1CsxBup=&WpcvX2AYVo#i
zICzSG$hK#H>nm8PFQ!D0YPSgVWKc1J>d8S&guq;b75vgj^Kz~$+|*lw3+v4x=f1le
z)9r<lh3o)oa{P#nk3f|r7G1OTKBjm#I-OttNPt8^r}McX3aJH`G4$P$kj(v*BMS*u
z932;CQ^v5833A{_6eUx&CK_9bh&q=%QS~0S0-iA$cqdC93b-_~ZFSniX8~(oiU<pY
zItCK>5^K*%=6z;8+oa~ILf_6@G<&0#Mkma2I5^{GBNM4GLlS7#@Tiezo`66H*p6fu
zD@|RRkU8Vfl$gUv_a#+G#=LS%doP(S<Kn7h#_T*QzPN;gmGc{&DI^BzzZ^O)-D_3Z
zG{p(Gc5E#mZunIGnyD27wh)d_{a>FXQ>eY-37#mG*EqNVt!Y+>S9W5(B;Grxcv-Ag
z25OCuHG>kJ-9JBe4tIJ~+6jpI2||idED%GA^|RKFmVAOuFQd;Y%mNrQ!tSlze|r+L
zuo2ii9HGS(?^plf3IpN0sgHa2Byb*XLMn?Q8E^f#kXYr+p~>G~M=H6kL+>70T%5<d
zf`f6;Of^);!2zRz4cR|(a*X-DKOpmcgFmQr9i2`kl{m>%4rUOZ()`tSJZmf`kyHV3
zH_fI#TsXoPdkp<6E?}axjcd!ZEznFSkwX*rWntUxwl!BKgN7wlgzqncFM`w>55`V~
zj}8d)TaiiI`6>bMpXl);@K1OKUSg^AowD+cJ*DcT;_?7k>wOLu50;Aktl8>Q*=aY~
z#7!Dz36M(u7E1oX$oKi6Zuyi)K1``2;M)P!G+a7i??fiG#%v389a0tM1<l4>mHcR>
zwkQAkmQ?WCL@8M0W@Z+F+cEN%tSLF`n)u))>jat%dp>oaqd<(Toi<;4u0?({{Wpsd
z;A}m9zxq7eo_bB(l(I2(fl#eYd2$+@IkGRtjn7BOLW8IJq#8=LGmcvdEN>t&W(}m@
z2Q2N>G;TN2!5<}B`X+Hg96}C|ecj)#mEQ5N+<C3x6sn|=RQw`(_aSSqIZP8btTd!B
zrrI_dSk?9;=e}9w*&;Xg2eVd=$C<|OoJe}-iB_R+^u9wwNh)W;?HC4;_YC^;3L}k3
z6yJ0M`?jLao`m(<-r2sugsKAGu!^-zwRY5I?hnpfE&&PP7;zYCF|q||H&ToUSxCew
z_xw91+=^|>HGQI!E@;AcW_^M0O>KsGo2{cGzdbm8Id3O$cSA2dlJl7?hx|Hgrq0h&
z?Xcy8LAk@+Sf|t5DpZftlrIR_LxY%a*X&utO(=K*y@cCXs~<P*%g%qk4o_hVZ>y#-
z>&ZG4g!m43>>tvF@j6Ia<_f4sC$+qT?f&^Tcrc1sLZ@BX2c4d(jTc3`%*c&qgaIXy
z(^#Bot0Kw(-WpwkigzSmswAroPuOoSf0h`I3t5Hu**IyaA9`1OUYt3<{y8-3jK_i=
z$n~Bap_lw&(b%)5&kOvQ?T{X|`qte^QOmW6bGEqt!CK=D^s_+yUjL@eT}{YGH^~@B
z0RNuBs1Z4>XNlBFY)Z3?{%#4xpTJpsqUSZ<#KsP7{*k-y@3xe@6FSNWT1)mw1VS+D
z<+k}H`SOHc=}1xu(ho|Ud@o-`SA)-B?8#<pItGz7f7e$S%!rN{^r=lJGGK8U#*ZlI
z>ugo>1}u`z*0L+kGf*a76=c`SX|Pn_*T)*M#*r(5Fcs25InWj4LK*zMJjdn+Xg?rx
z;PHG@DwRn@t6ye>3K-K!s_n6aCSx+<MPp9%(w7szhDEK3+Ad{&7I1nYDB4vy@BA`&
zg8aa7>s!^Dz0OZ&a2<`_B?OoRcFibrIJwolTrvoHWh|(p2D*m<?OHSJ`WA_ZXn8h0
zGH=57yNoyZ14rg8JJrkcOSSby4W=U0xspe%b#xvF5Ui$w8GA2*pK%Gf!=%~Z6uQsW
zI&6*<49Sg9kR*fTXs=tQB^7H0ArqpOcQ%OQ%Me`%$@sGR@jy2uS3;il=iaJHhGNsI
z6)~|+gvF#%j8HtHb(p?K#wCptG&<T7hB7?HdNh(6&!s;4xk~q{M6#i=sEF~<+6oi)
zY~Jw~P-uSp$LG?#6HUcca)f-Vde^mq)ZBsaoF#`$W~2uOqB^^@sEdJ}<bKJ_cO;&B
zaS#S+xI87+PPuW*Rg?@*q$hG6wI%W*M_L0BslVzVO?PgUmYq(_f)`lkQ1+UK`N&m5
zi(Gj`bc;vdDe726xOJD2#}{X+)FOjU{m<!<I(9(ILIe}16dQXEWuiC$VHxhmD-~@m
zg!3WG2|Gua^+oewF;+$zAqXTpI2Fx=Qg8Jxe1}|CWdzz(FkQDKFqYX|j2uTj#$ZWv
z|B5U}ud0FM>7Ox_B?WwFp%&A{BOnIA-c%m^)XZ*t5j_EQy*ehzBT&$97B-`XlqgkB
z1~WkiwHfgBcmeY$gDL*He}*J6wh1uuLphS7b3wiKzlE{pIOMv3O=zC4$73RL@yuc@
zw47207#L)rhDQ1F0<hj8*4j*OLleWxC$6O6wFYYnZyLRhui=+>GxpE0g>DH*e?5yF
zlN?+B`%xR(PR?bwImq!Pp%X@h-_MEPJ^z>F(wn;mOgUmII0jW}uRTA?v{jYvGq3f2
z`*Zm7@oQgt^g3(IoId;$3ZYH|OU{p03m{zTjhTW4TeZG(-D5F}{m6H%l)d@6@`L#c
zo17~wgwSFpqx`|c473a<!c-sCQn)#m?~DN;$2uP<Hr79)xCd&gRT$>;-jd0FCH*D!
zpap8Rl4iqIl#vmmJ%&X5y;XqTXv5}D0=N9}YEFsRl6{o)gNjjvHx56J9z$A4O;H0)
zjXs}q27^~U*`=2oK_6^4>?BN}s&H2haVOFPv_q}_*h#j#)kB-%u0piau}q@AvsX)y
zU1Sgcv)CtL*XtYe!?dyei^tC8x8^lMuxvtilLv=W_i;`B(ZNaA0Ozx?B7NYet1QFj
zchj_y*W|1Gfmi7_jh)2JC;DsU5p%-t&5Y^Q{+g}V<p_t)p%Gw4*@%DFb#q$vN~w_4
zM-2#b@L;J{D&21{^~mDwOaPyxI5wTyDeTklOb2PXN6Tw)Fp({ESA3%=7&E}oDvdq1
zGF?B(kPXLBn!0X&6(0arlu=x5qe{^RYSU!;NtRsN8rkNck&h(h@k@H}Pbp8u{T;e#
zs)DH)tu&deD~-r}ufFX<W?gvT6XZj9PBA<OLjC!^hQN2G1Ewx^&}}+XF%@_rb12s;
zOlmBo4|NTswYj-tEf!=yt&eOOOzN~hvNDqtSWYonu9EFq9S41zf9SgEcHGv9=GW|6
zff?{Q1MmtqD8tU7HmxMuDj83Q)0YTf{t>dB1b%-shtx!@cGQS@$~&R1<(ayg@+ANC
z0G$o-!Fx)woW`dZojHHQWuuxcs7`xiE`qbTi2&CjBoW*8mxE)8KQkFH7ZT}}4=54b
ztaA-9$T|2Kz%Pl{#Rokr*Hs1$=$}lwr3W6CcTv}sxw*O9K8u9^A+=`P6XBGWQit<Y
zR7Fv_a4H*~TvsYu4O#lh`sow_a{x=rx-)b>H51l0f9b*h7s>U}rwT)xKm8#zRNV&T
z0F(FM2KlzdUi-`LIPE5X(8s_32;7j_MaMC|gaqy;>OoLh5jKM(6`eX1N~9X~c0WRB
zt>4hS(ZPv4F5wceHrF<;7ZRKi;2r-%*6cW^R5e-0SkD*Ostc_eMzK?05G}KKt2bMi
ztR-(>9G@m>ML#V42Eh)|F?==S+aPs=$L`Ha^rht6k1Ck@lCXG`)o^|O2-Xtqr9+7&
zyutmqg?UM%6a0IU%Yu8s&_|eLjbHZC=hAF5WRyHrIodi%zdAn1=+BV?gZPi+s4I5t
zb`Ey|lk<#7PzC~KO1-wHY((^Vy=!l8ZuI2D{@uvynv^X}hEe;9#*V*^d}^N}e0~@3
zd)*d%-y&br%1dhZaOkV@DYfZTD_h4C3k8D?maC1#x&LZ{v>(GtDt3bdQ<M}i=LTn7
zM|AQa@hvuRs#jUy3hHHbJ~^F^JgmUw57W%8Z2|t5CIGX>H7jV=w2EJJiD$YQr5#mr
z|E{n*V@S0uR%2h1^DtG5m3_Ilb|r(v#h4nQ(a)JWe8WS+$k7>cEnU4%5uEIfikzs0
z@dZN((dKI0x%j+2OYIv1omoO!duqQB-OTT1n;ldf6G93Kg_Kh15nQeTo%;xvsn2H^
zGg1+K15SJTU+IHYlRfniR6Q2dw69#iC5D6#gSnoY8o(7!AXxq7-sAi|)S+Y8=I%Ru
zM<w_e<mNbNZcE||&7uH~j5iB=jt}J}u8t(ddMkd6xG&r9Y7eKkr?u0mWd;`CO6W48
zu%qRVr>Yi2erCiA;9>oLXnGHJIR9vC_@9J?5WR#9f*^Vj%&3XpLiA3Q(TUN!5S<Xc
z4Wcvp=)KorbfZK!h&FmJuk)PqeuVqFul3t|uf5k=^;nnp4(vImsps!lo6IX{?Ti>m
zf;^l4BiGWR`v(`@z)znju}Jaz!IJeZ^fBAUE0(fqr=xg7Ad@sH;?p}KRc7X=N3CP<
z2JqL~x<=T;vHT;nlA?PF0O|^L?Jri3om8S?Ysxun8K!N2mOT*J(8s|wIk{}6F);ib
zC_2kqQ0PF1KT-bz?ppSNHj?IC03#9LDR|ObT__dMU8241SI58SY}xzGKnX|dZuYo^
zb#5UGC8azEbBBE(P5BEHRuNWXwbK0&Hju=AASSdFF%R9`-Lothhoql0Ws$=zgr!Ij
zWm>Z9HvXTZQ=QUPgFWYuHr8;D-&HCuJ95l9kv*p;*Vm++jD8f#iM6+EG;7!Y7Rb$H
zf-P%-N2X8&MX}8CR(CNz3K}AZX8+?LL5F{qU@!cts7w5hk#0^V*QW(f7{ffKEKobU
zk#LK8-%}Wd+)CHq7ONOsD9M%+ebxkB<X(it@BZQZn_sb6wJ9}ZZL0v@<I;6t(cu~>
z9-A0(do;4En&fQ6WzG2Z_u|fA5H;Utae92dd*1S`io$BJwEvk9rXETfsCfqw^j3{e
z*5owCv2~CLv|uX3{gSi|IlcvvaBy(vPm`#4=@?=@Pv$mYnL8lP3>vlHVyIFr@+JkA
zHRCEGzDEUkM(i>3sr3@Ya{Q(zNnzr}Mm&4xkobNYSlEbWlmo>&!y$O^a9l2;>9Odx
z#9GkZzlXvfOJ+rsPV6-5<n7MZ+q5lsG^mOB5xl1dI0o-oSR-IK?ZI*`UDl+3>rkxB
z3FF(x`!1!uLFgSrlc#nU$VaY?E|c5k2~1pd@&oKo!!OIks*=KN{%%;LXTML&w}{v)
zQyC3i0$CT9FmD#=b{P>ni3p-Va?=l9Lkl|O`K5#S+BydWxPMF-7%A3Vl>6@hZf1UI
zg$@hL3%!={F#hD%%A@eJWfHVQwkc`A*B&<M0=Q=stAVlMk;_96NSQ)NUFD2NMOEQn
zoYewWBurbYjFyAH(U&~srzYxFs&}LQXhQ+<hS=euMJ3bV-#N59!$YW*kyFw4)B0A-
zAMh5{n$hR7G+oUDx{bjL_Y1MizmJ8uT3I4aW1;pB5$DxyZ*S&*F$L&9_ff8h)qLqe
z#Q+8+6H`HQDe_bc-T<)SVE!@6um0ho`odBr?w!6D%egSj#^Qih(LvKy@bAI%v<`fE
z`moh!9aQ%yvX@zMLjyRhKCk0mBcE1|zwz>FRRn`Aq#Bx2i+5JLRW<o8!iI;MG<65I
z-{_Gf61TyM=L;<#kaJ1XQN=Hi*UqUeE@r~L@Tq0~`25?1O#xpDi9W~FY5dtZ$Z>0>
zflHQ~;h9l|?Iljt?M<-RZi-<405qr59lJrg&&u!z=|RbP`cA9oQu|*%aPG<_ed;Yy
zwDx=<NTt=XdG7G|3Xdz_Y+Zy_+6ivAN=~;kK{d6iP#2ppksw3*m0CGWVk<CFv(G$W
z7mCo;T8zdtDg;AouzXdFPv`;a7x`r+B<dXG1mLX1nSaxvpO#&%vVO_(;VGsM<U9-0
z`x$K42^Z<lA<jUygVN+JFPzGxi$nLl1<tK+a{(#@rG-a>GF(RI)B%u2+GvEe@vrol
zar&LrG<5gYRM&}=qPLO%KEo6Rb^%FiS|Fq439IU5FT756;Y{9EQJ9>*8Ar!Du=3^E
z4kaw#n%~BztK{XJH}l~?iAuF7#98XfR?l05cqc7oDsy7^+b{Bt^uN_jsbkEKtUfHU
zLQD+|^j@w;EZP|U{I3iPmh>vFAgnLvgW~&J{!D0sb>?6i5u~VKx`%GI8+y28Mv+&(
z*`hY9hoz%poV)+?6DJ>sP0mRs(Njs(_|2G{y}Sj!`x>=s!8_^-1+|-EpKs1lJ?19}
z{+H7UCc7ay(zyZ$)jlc%VidKy+B&e;2J4rMw^}Ug%_|KYFcU46@5#<!-I@5?BwK6B
zyx)`V-4G+zND6!8BT~ba#iQ198Zw}q(^ezUH$g11$LBokz5cG82wmo&m05c)DBI+!
zoLk81N(MN7iiu?CIfo2p{=+m>UL8IxPcHO5Uue+#RCqN-EoXRsXQkD`4Af(jq<>;O
z4HD*mkwhK)y^x^dw;3A^x+WH6{GGdT{-BgY=P%4)78zu5D)hOeqRlZyH_rOkUV1l-
zf?UXzlXSJ}A7Ys~esuN7e*I=(<NAp$qt9LAG8KY(c6j#6D%EEvipOj?_L3-R7i-3U
z=FD4%*_J1>B*140j9Nm(8Ls@<eVW`%LvRJ;u<WeGsAxH0<uCe5z=?V$QSjz+ONF72
zD(|VbICc*or++B#oF&3*^**x<eEd_#0hOkoq%a%~CCPZ1Jz)AJH6g-z=S9acTs-oW
zXm8IuO6UHrs|wqv@HHo&ZE#~qYVBT*3!Oo1nk{)dq2b>hneu?8VP8N67+z!B7PEd<
z>w%fGt84MPkDzD1HZQfyDL(Xi;_*&>!_hiiqe-4&ysaL~!bSP_@RP;hAEcx7%xcP<
z+q)c1w@+ki=4k!w$;t{1{dcH#f=cs0Hm}Ws9<8QB@*1(G-V<<m+aJ=nMC`Ku`j&S&
zO%eE_*^(@okDAdl^Vho-qmJ)`Tq2Ezzmp8UVpa=}Zul|*Ix6e&zcij+gQ|&Q3rRl9
z4&+P=nuym%eMtXQe)ZqkH>YcpRlQKEFJpk0O-{Z8P6K5Muw~mHtrvJ2isxE`Mr<K=
z3Nsc)HdiTOV1a28L1pYn*3t<wlt31B?;uivHYt?Di;W`pE5BKjXnXW+x&OwnUr>@}
zP^ihar(2`hg8f*^A<X3y=F4wgto+ifdX79jgHS#=BFZ&{%BNqhrs!pX#z7tBFo+MW
zzgeWjBM4sgm6-910CMWua>m=$`W1dJRnU_EgM2#&SK{qU<U2iFJEoEpcXSr1oDTSf
z`Ik|76urC1q)n+$q@Vlx*Y41Z{;@N=+aDJ6Oz(S@B6GC7-6=^3h}U@r8Gbq4ntiVx
z1?QQcAj#9ojWqzt+dD52(<W7izIV|;F66UG>*a^?W$Q2PFQQHI|C*#G?9P>ud!95m
zpsV;EN%oKucqCyoIf40|?ov}&TZ0$UZFYs{$?}uWr9rhIyC+-r{&9L(8ZC>}dSc6w
ze57xaK|=k#2k@24kd$u`S|K$lSA~>+*n+WU#+x$2CFis2^nS(*a`pLeMbwclf1vRB
za}n@4n3mo{h3qjxNYm5Pb?{`L$!|x`+7u^x>4ngd2SAn7IogVqkwiA!4|OuKsmyBc
zWQrywKaC+ooMUdl$5ct;`AU8<!*_>iN%G|NJ#l9`Ri(>@%<L<9v9$hj%x{wJL^&ci
zOfNGB%_&m9_c>neP4^0LJd?S07<FrV4AbCWaJcl~*s=>cT{8cg2$IpNU}_#LcW+)K
zcCv^HPYFs<cit3f3C-_L)#+Ecq5<=oz8fR6jCYS-5%CRC?k$%F6eX(qe&;A53pH;v
z$$Op%t6Up5xw?4SuS4`Me+uqxV<DzI;E?~O0mjY&EWZ-js`i#+j-1-$VygBEUO$=3
z0*1XL0Q81SGRKCo(&JFxb*%77BJn@tlq=T`=K>UqK#LX?eSDgJ!vP5+*_qhMU(SVz
zughhfHa=h|L+3*QAqn2uRM?7jJD_L`_DsYiXe;VOtCz@2Wm{C(ZJ3LA%W27!YIWj5
zI_9g_5mWlj)M_EDCnaS|YIqQ8+A#4WaMP~Z+kWuJOc&z}R67Q@M`hhRhULY~<fVCT
zob09#jD6Nw4<I)I<>J@#R}tEtYjm;ruRKF#BvhS1L=j#zG(-aOlr!;+nLodmbqQCS
zdT++{ZquAdg}r-nKHp1J+b}`b+u>i-^O^$#SBLYN>72?I9h9v`Zi5^npS_}6T(2sx
zIKYM8v9a|Vj$yICwgjY2=VcpTZ%56ZlOeOgu9x05TqpAH@tRP48=WI)GWcih`xuPj
zVKt2|*$(OIJeO;&IE%g}-%<@E#XQRFlgM5OC(=85QVt_*`;0?^rOQFMl1d<Q&Cui3
zp_amQ<A1vImR)L2?}B%&&~R(JH*PFLxsfEAfL-eT-tlqAVW;JpKbNB0K9qpnWA()D
z$JXHIt>zO$=in@#gR9W2*1hJn%N{=I-XpcQ&JTD)m=X@;iT=Z?ehxI!vZ8!^NnL&N
zyz|AVLv=Mrq?*l!HJQCuv?_tP>mTK)3mfcOAxUlr(9a=H8=$gFr#eEstre><j&Sh!
z4!U|J3UpA$(t#49qmGcIL}o+8vW(+g7hMjoXJ0*YaNGaujMPom*eU~jwFOzB<*PU-
z$oRZa<A%kC#V)O1V{Cj9ls3r_;x$V5EBo+I=4Xv1+_rN7O=_-q*ev;K^PqXsiFxec
z^gkE2_+&hx?qj|GD!@*@y3-IT<)kXA7)!;%6f?_(@7t1>Rvh&zT)#{BO1~(wg=W4z
z^^M81(~$|kD7$?T0&=!37{)NfbFR1UeeQ;{M`<%Dwv^j-qA&WL47IHv60ui#O#cS~
z=+-`n=Fb;sS{QAvR~fk+`_^<LL5l(Yg=34kJ{O)B3^dhG*9S?oQ3Sz#jkcb(s$WBo
z;vm8P2>MNEvGTM57@X?vYFZ28VhTM&G?z>2o6gULbl&;uEs~OV7rK4+z4!eycHX!~
zgB12T+=NEZuyPXk5L;Su{m?skx7_AjC5y%St18L0<U$I)RJPmHR(|e2X!oEMzxLQY
zQUAa0>7UH2lMS~`?!09M2*IrNMx|Y(4q=tm(0rqsJ8#0wP#aS#bX&RL!*u$sU++bI
z6rW1>TIu_zY_fm?s{#=9XYb(+{5C02uP&-B$WUA1&)UB1mnVba!C_utMXhebi2v&U
zWG|&t?z*y}zo>n*WTC~5D_3v-@~xOjQm3f`Ph`Q!$rsXDy_nSL?h#ZzOuwG?eEnU*
zF8nrENrGuT!%RlL+BfDRP&?-2e>Qijq&FVsZTxUUI$k9AVy1G1<)GIe7bae=AQz{M
zUDZV#J+ZOke6E`nc@R&3>I&CU##V1H9|7;Nm;X<?r|Et_pCP)ut^yZ7d?@HK@bDuW
zc-KOB+EyPYS9PF5;aO5PR|duX0xm-=Rpn(?VE)PSXaS$65hUx7%^TEMv8D{q@zy)R
zo~wKB)~{Xev8v0YYhSDva8ycdD2wRtmIh&du6JQShG)GFXU{VjT%-)XyvscnNg$ac
zuVLQ`9?=<a$bR^a(utagO6*7S1a{TYHO$+2igEN+P_=3vaXx=LQ_y5#k!KF}sD1M~
zfy;_%q5G=CZI+0?IyI)JS#*7W)-ej@4J_~3D+hL$HmCeP{@cVnBPpyc&|y<KuwxAY
zF4Ln~T%xn-8ny-S#eO^Gxs@H~U=5xk@z>VPH`+m#Zj}1vZf`P;#n+qOe8tG;pZ9{8
z!WiTBz9uP97}teMVSlb|r+pQ&d#RP`6M<HtkKgrttAU@i6#@8K4-B43yuoz04RzV3
zRxh^G#h~q;YhT^M;y2#qEA1;Ww|{TpX3MtJp`aID9;ib3fSyW?%s9T2)(Bd(=++c>
zTqL9mMaS91!>MkN<bD@{lH5tbdyh5n*^;F|^OtTXoQRT4_=zH!_E2M~)Kd73RjQp1
zSF{iYl5I=%F};n&q0Yc??{$cT9B0HO9SK75zWSRXO*J)umMO_U%MoCS$m95>@Q$3t
z;p#l<hzlLa6HnD!C&%=61^X9WRsWa&xVsvz=a670&Rk!de);`-7<@F65$6<1{BiYb
z6T{^GJ!y~UdZJxyS8w=+;d|u@Ihj;-a}tEj{Xfcz*XzUdi#(%zlOmM(B75V++VB!X
z2<^1eE<+1lPt}aA0Yf|wn7~@5msBEVH2hE3M?Jn^D9LA$3G4|GJ{p4UkGi_Djddws
z3YSjb5UGUp252H1l>rKDLFZ(lMy{E=J6i#gyN{rx{ePgemW_SJ*@atEvp$)uTRZ?O
zL3y^<#gjB#>_&cB@!Pd>RvXjfeOpM7s7bAHqoBB(bFAm_jf>_|F9>Gj?wL{hpTJy=
zlt@Il{!on8=fuR)<U}Uqpn)=3%=9%^-+l3aCGA?B73g`}!iw_U?ZCp5Hgh9RWI5AB
zlg~KAZmqc)5WsXSq3RO3;20gtp@6yw`as_64%z#QkW3~WI=>DlbTHG@&yzvqQ)iey
zj|8e~RST?&21SELdpw!lBfC|wxCGwR-F=Em)ya`h`{mw2UZ}@LV+uQNh_>J_Y5r0_
z{N0s*-fbl|DiN-CFVb68j!{&nRj>2DZ(eJVRvJy%Jh`eN<P_JVqny9aI(}?uVD^mo
z{ns>(rTI}^1TWcTV_$4F^&P=4;CkG?hi9%vyS5_YlQ6ge%HRnky8PpNoFzdAas3dl
z%ES}fR>$4}-3BjUr+#a`S(JIz%%++r=S+nNpBnO&BxsJ~W(<74<!M(N^mXCx8(OhE
zLs4=Y7YUehm6qNwj8iUz>RV3}3AruM7@}&U<!W)euMg^0WH(P&z52GpmAqYYmr$7a
zL3R#*r_`2?%Eh!^fx7;kU$TP^`75rX7uQRepJ+bH#2HYFEc0^r@zdORuujq`{#Hq8
zgx8aE&v_*&h{X2ORbi9H^6}%0-Kddkp<Zgbcd8ONz%9_~l}PIfZ?Dr*nM(jOM!DSc
z3Uk|?#82`*LpJ|*AvANPdjhQC)HyF}s!PF9fAZAcM+RYM&L8Q_Ax@Bb+jdjB)?nJy
zs>OP7UB7bqEcwA6SpML{5l!cDa|1E(NSyfq^U6&^_Sa6e*l7J4yI1<&5@Ti!oFPxb
zv~DReFZ5?EZIP$%2NhLL*<Dx3;S%X~D0Kifq%+ZKM%<dAo#ojk9$>+#gWx6Mr?hm}
zITgx<Oxc738`BSI<$mSHsNrKTkC%mE$lfQ9X(44wR8Fgv`q9O8#lZ-~5PeDR%MQ=(
z@A2IXEK+yupDMUEZ^eY{aSlJ;AN~LB*EXqQ&)2Ap1)C!4t>&kyA6|5p=6PKX#nBrm
z)?w4)yG7`o+n*9K=HVMc0#m3=-~FWU<W!Y6mg^_q?NT3=XIVk&ahr81>_GPzkIaR5
z)uG`$dK>vZg@}l(M~zxRs;9CN&w4Y#7N?EsujM__V$x#+a-vYRiGS&*HMU>C`}1iU
zS8qQOSG$?Y@58hcW`9V3ztE-9O#K+&I`&=AMU04~AzGr1%i^O#44I`-8ZShdh&<~P
z3!^5^+90ifqy?R@^KsY*CiLk2bp=)38^8OLtlY+RN30Ek4XZhh)Hr+16k@3;+Y$?h
zZyRM=Newv8(y1v=z`09J1jTV!Ir-mKeN~W;F^Kz7id#;v<@$;J1;@`6Frumq)x)=|
zXOyRgqYK9SjiP|Y=4Q#hfnE-e{lU|}v=@JAF04NEdUXwOu&Yir)i|+!`*WDQNZFco
z<A`p%FnxE<gWgOiqs;w8;!M`WTIXOCe6=mS>Tw=THiA2oTytXM(v(;g_Mcz4SYgX=
zqIw)g#O;{$s)2?fqt+c4Fw;!Nr^7SGx$~eGpN-%wlVgx<$BD2Zj_MoA3g*d`Ps=i$
z1cv{Ts*5tfN7>j4+h_&<CSwS63GkbAcJ=Li+nG~3y+Yp2Q_dIsF;R(kWHXF2q(fr%
zJ66q(2(GL~irE^*?~E`gd+U)}3Lyy0!Zqo8(fm-0t}l}v`7oK|lJ#bxcI$U5vdoR+
z(P=<}BAtZGLt;7hoV4h0XN_sWq~Bg7jb1Y3V>fg*hLKif_(Ol{I`Aldk?CQBMO^l_
zV-bpzor^~!&!&=@%|F8q>V%l67jIShcpq(S3D`Imn9jjhmBG9mI`<m9b9-}qGas`*
zzR<-#==FU1(K>V7huZ$ZunJ(tEYlnO<AN5`%<Nq4j!7Q0)bW#JM(s9j<#COc@ij5&
zC@XO-uQo^fZO`osoGw2>G!lK8Ch$|HQ6@8nx5W!t1cPJ`{7!6}0?}<hR~VL5uo3?<
z@rJ^GBqxy{99q=$MZM$PHGLf+1&|nvAWmOSasIcGeq^XpKSvxx3c?!=*}^wLh(#B$
zY}yGTWlb9iACtoxNa@2r#HSbvh-Diz^rjt_rE;Mk!((3}8Xu3z+;IcUkqrr$g==~8
z&7H?hfAgA$FSl6DbG&E_V&m}{D$RJe9I-C#ZzRO@_$ewqDSP-NM8*nM)!^oSPiAH&
zaA@)mcLG=1o^_m#2J@374EgXsGlKVz5gOLEmZ_fvv^h`2k_XRRHJeGTkUAKVId*wb
zSn8EWP+NCsG*Z|r2&vCwD5FQAg;U4PGm^i$WZF5~x&H-&gWD6UdLF9J+cjp)^JU1S
z3A{^V`2Gz9W>PAXc($qL%yyZkt1G}^>FmU)C>;Qc`^Y<j8^_0z^c7x8<25X&(W3rN
zzi_<#0$);=R(A8fl9_oXgLT>y$IkU0(nBoDLz_BD67Yl^8@s_mF^8bu-C)`Ls`c%C
z)c;O->%&?8$6-)$xEgfVv4n1QJ-Q%@1NltBT+-IS80Gi&M$DdY(xp4Taa)&p?Dz@%
z+OV2co`qD?em^tWN~ShEsa9vBz10&Gs?DzR@fWF?k1q|{|03so$?^hU$pCcDBnd|I
zSY(5>C5Hb^2hs)GeV$qLpTC~w&4U|jC0nsBE=|RzkQi#UV74Vj#|V2w;*PtlRT)1h
zPKAQ^Ys%I6y_^xhfid`HuEnrArO5$^$R9BV<Z8ys*JCJ0vqF?3{G_v~ctePkFX!il
zUFb;MNWRUGe(Nag+1o5?j$Q8-j_F*rgw=Z=OfP&iQCm5&cjxxn3P8z{pSizr!(3m~
zPWN?EBbhfe&brcN+(7IxA@K2{Tw49RVd>CaW9De<RfXf<a?IYbSdAE4<f^f8OJ5tr
z{5_JB;N^M(PpO7ku8|S{+bND;6LoC~y%Z`YNHVo=;HXFEU(Aapu5nk<H8^X-c>vSN
z@u?mKe!W4}@$tk+Z2qU!4YCsDoVrnP`5-S};DraoO8F}k&z9v&>t+I<vl7-;<Lus;
zd|0nutPo|V-0xmMNFh*cak_sy$U)M2=w4`fUs9)hJv44`gsAiwd?roTb$ruy=xvf3
zk#VDQFUITXxwk5eHZOL~5Pj>z&9U`zJ`jwovLje$mD#{T(5y{(SGDNvVUN~u8~fQ&
z&;@|0ldN@;mR})wvg~3qnE2Dn{`&7~ip!%teWI=Y12tbDUH_f`fy$-d@aiUydM}T~
zBo2B#pf1+Cc;_jjeB_w7+;S04-4z>yye08=`4@?%r7Y9P(2K2}tkUnk3y*?&MeP4d
z&RoiYd+I>|YDjbO4E85A&k9(dZUrH=WriH$-dMa0ws+xu7_cG?#nX@-<z26i_hM?M
zXiWo`?SLdQVz|I}_P=^~D@|D!zuO*^GFot@#&4)8uwq#vJvHqON%<G<H}kLyC)%o=
zQb2~_Z>Yu@$vmS`_04Gv+QaFa=r0vPt(V1YG93dG(Jizyo2|wfveF-3<hw3^R>_qb
zVl`G~+A(r6w4#z#6Eo0yS<=*7WD}Y*24qwY;dlwwd^)aoU+`5pZ^JxLGD1F2Pw4Jn
z<1J8?BB{G6VdUSOZ5lw&A9>wjgr&ils}+CBo)&dZ3`$?M$Um1L+TE_*f2NPBA|G85
z%}A(m*4+HKz+WQ48?d18LSJqUJ{wpqfO3xN%U~i?`>yh+xBKY?;c<3<9X^T0MW4UV
z!(Y~L{NL1oTRA}o{6*SVk%9T)mAqFq>O{P98x9ZA-zDC<b3<Ckuw#mNX=R)gfT0jb
z-Yd_#qSk~cVeGo}SKQu+hE#QCn=4iVHSL@g<}k`Oc!90SBXc_%V^``3O^<boSXUT4
zl3aPMGhjU`(<`mMqO*L5@F40h9kQ(54zcCDVT9OirT)+eo~RxD$W*#mz%sP<$O9)1
z-=MC>tWdg98tlFw))x}XMX4E=u&-CPdsY!BbMs!|+5~(XU7QSZ`G|h_ULarSo`_|Q
zbJ6>Cz|l{~DsLGmjbM6gQFxJav!IVE+;-Cz4~HxeEd6G%U4V}-ex6mIp+ot1Dc+M>
z9eEvva3WGJR>Kduf}^p2Xh>Q>h1f!XC#!$LS(y&{VpoJRjSWa++yT+mR9CJystl>`
z6<iWwhT1@*sPO7V4-qp%a#ljb#^d}%IJhB{5UPDh&%ynDGO(B=9HkNj{-pT%_9AJ}
z5~Z$qG*3GBRMOO*L|mUk`-LvLf6yY8B6rFh-o0QuBy($}Lu+YjWI(0&RLS)Pl4dN)
zHFzYlU+`U)P^Dd2KIrE3HUfR>w>M=(Io}Mr%G*Tz*j(|*Qu15AtW@Hvj{ngeP!Q8y
zMfxr=9yS!`q?x$W1#JmI*DA?aG4))W-_kR!-7>K+da-V?k0y>XEpY2i)CcNle#&It
zZtC;IE(Kas7HahIl27O4^-`53=`w2_1-#_>DMW99-u>QIuv>4?ev92n_j{VE&itEb
zrMS)Xvqr2y?J4~cE`8o9yNoRfE^dZAySeYhEz6DHL@IF&;DnFk{ju=YYoE{v#+So`
zl`0`X@MiA6zk?ve%h5sJNsaVw-$4gMTR+XjYb1WZ$~s-=|3XTDx?)V>gPdoJ7W<Uf
z?T~F1K(R&-qAN|U+gd?YwJVtTV%H4rwI=mP8+Wlw1#MD|ca%9f2=()+7I(D|EW_Wu
z*0a$l5SpCdP<S58aX_h`-g{|3_nZ^fsP{vlkx+4|cB0kMBmoN7i~G+CnP)8#@(C4O
zcK}M9+D{5DwI?h~tV&R3P2nvm;eD3`uPFVkX6o<$O}XkfLjNlMcdUe$5o(%ui-O4{
zgWWOCtvt9pX9hON5M;<x2mP5~u_lrOVgTjp%CYjQFoIZma2H0NOZ}73p2!X+BMn`T
z-QCjmIopeC!=$L!I6SIrYM~%*n-D^avFSyA>2`A;5eD6xePEK?vdqZxn|_wpFkq6V
zi%XewduU5;l7ff-{a>5abRM>4s{T6K=TRZ+wXv0HNrnr)0RKaZqVdIo4K8H4I)GaN
zDh=iHOmkfPgb_6=%X7<AQ~jO4AHeLqBFK?!Tcn=ow5V6DHMEqMFoz)=b}l#(btY3=
zZ+U;wmd0zu?Un&Cql845XzmJVAX;{9vAe6voEiGUJ!y|vC8D)#qDE8zE*V><&EP0X
zEbXXV17sDMqGf$Ry%z$ITf2v(3Un>2o?r!~?%+o6(vM73?C|{B5d@vQ@a5$|VRR7Y
zp}*&MGKU?`Lckk#mX~QaJ4=_@RkxBu8!xI_TI_$dzDyk@L6n{rzo@f;Xe6(U$AqC-
zWz;&I2->iIAIta2abDSYnaGD4d30)_&QfE(w(>Ong`64LkUuNAar32Eu*Iy;9$?a*
z^h&fCUDWzq=rN0Lk6*t}z0Ex01RAxM3K^NZuFIdLq1?3>v4!1->8n>7oN_Y;Hey<c
zVSw_JL~rqEF5op$RHZLdH?g6KP3*hG!ctg6i6a3+uv)G`356Y!m0oPUhtkqL=J!CF
z7BX{^&p)HKP7DE4S1j?(xlpXl78>5Q<(2D8d}>v=&t+qyO61(Z==SU6qwWvkEw>D3
zg(x0@r|v`4YFAGALVYS=UN&$R9bI%*-^H8%-Nc62(*GsrshsJ0<2@CQtL<!BrXkL0
zjZ`RBF`jV7Mq!N)7AZ`I|8Y=ROfJ!ss32%o2Vc=Mr6<nn65J^`HQ$>WS|7tB{zLii
zAG&LztcKvMnoYbux`tC;%8Hoe0@8|d=^aFRRHD{$xveziSc3d|(5Zf85Hq4cfhVy@
zN7Gv?i<@G9s637*Eu<kqAce$;kIQs8=^&~%Wh5NSh=k3Whe|89CO+(L9#j|CH+^XD
ztzI{6yXpn`Jvl5wQ<W4mD1_#Jp(O>d3w^bOmqDnbo5y<chym^jKP314O7iI-UyM3q
zl2N?P45mU@BcpoSHB}d?y^HcaSYL3A!dWw19Um+Na#dU@149A{Z))vb)iqSMOsVAX
zqSGrjHh!Y|&^u;Bl#kqcg{CiJOxMvT2d7?f{Dq^BSU9)_>z7k2qP?cwke;_N$>7h(
zAGyh(<xZlfV$AQ@J);#-WjSQK5j{nInWg2~R4uSF#zYFGzxx+?x;TD6bi+X7S82rw
zkpZ1oNAppUH{xzNGfID^3-o88di4XZSqiQOen0s#C8G!e8xS>dIW{g5pC}!64%bWs
zWmC7N`>lM@T9+Ftc64t<>!F%NErApY_4VMmm_aIg9V42gUV?jbo%<h3M(q}z-TmLL
z?k~x7?z7F0yLF82?R5&Mld@Qnc17En>%L1WgW-!R;<QOy!1QN|d~CF3mhy{ix(8<Y
zOBI_~F3baTiAs;WqwW(c<a_H24t{+c4oR63#)&@M=n}4HBdhAJhkT#ij34<#A2tAL
zYsUZ<Ju5Kk$HOuV+fY@Mv2LVOD&&ki>5SCN@<LEm(oWOYLP-k=IJSbsUUN!2&V)PC
ze>7GB39@MJX6kI@nGMRjX|j~#71yadD_tT1I-Tw=es7&?4u<?7%$&BRg%Ty@`p)L)
z-f+$8#vud^8FKi{*5;MG32y0UFzYm~8bGJkHrL|ON(`yxXSKVPqHY>7?!i%FNBjh$
zGi2g$v7^UalCYDw`P)wf=tsT$T?AB_*~jwcAfQ{yfUCF;jY+I3fLZBL1!%G=X!c@J
zXr62|&*bBCqWzhEjBNWotj?TnCBpAkzP{A0jH7wSD^JQw>Ez0d_O9_=%E{w;s)W_k
zOu<UdAIuQ`M|m`%=L9#ymI5I?C5u89yXF<;8&LLkV1?oK8<pDE3I*%DO3w+g=>r*)
z$$x*!qIX?M(S_7$qNc+2+n=%AKn5m~)AkcPZtTs{S)TW&?)L#3l7zgaHODxBTSsCw
zOuyqKz9t@(CJ|t2lR&e_z}mWhtsZc^&EX*hoXjy#qV4OBvuSd}?J%CCgUFx6s%*e5
z2=cY0v8X>%{D!4{<6Q8&hoCH+ls$PoYHDVWX@P|TZS`fNEkWsDtJT_;m_gMbO!BER
z#IA*0D{i$H@C3v;hxg^`h5%UpdIpS}US-ffpE5OVyF~fg0mGh^u6V%!O_;ii>49kK
z?YfuRKzG)6Pt3v6OIe7UN*{r=_-v4h7OH5ar~2PB-))>g&~l0M8h$?+*DsrGWHp#e
zioux`U1EL1K){^|{UxW#mzM(C!@PCh_{%q%dX(a*AJtI1c?`_1E8_*Y_ddz1(fBhh
z)?~N8zo0-4&@YZ$Q&oe8onT?LkP7&|1)5|s=V0Wrw@SF%Hs($&T@2mz+Bc)v*_Hpt
zmMGuD>mfWtzp1Zu+oo;v{Q&#Lw3&KREBotVMZWO)+9P$7EBe$uo)?Nk(^h&>6Qj=&
z;z3G&-!zoUt-IM{@CPh^fi>3~PMM?)*@>Idy49zknOrAXSO|$hGOoEdoK>9ayuI2w
zS_9T|yZu3Cth~C_bu#iYmY)|0H4&d+vx;-P6RRcwN{HWg8a6+=xWRJM{52gh&gg}a
zo@38~g*3iid8=y&>Ev$S<hcKoRR@2O?zadR`Fi;(DbMT})FVOMUU$?8^xcwYrlFy8
z<Kl#S08%(GehsKm;K>zjV!w#%ER>~=>qW8Dc4oG}M-mGAFj~z&hAYad(t~4o^Z+&d
z26`6h@CBVQ#NMBfRLfqc>)G=e&rkS;2fyL(AerFTa~_Pa`q66V)$279I)E>VxUffL
z6vwe@k=Q4c85qrXvkn1eNhi6IKe}(qFRB9;np7(vQXFqTwk`i(r!_;R#JrbOq%t$v
z-G{r;66Z8xny>rPhOWu$_+GUZr;H7_&zJteP;okR=`x@SPPt&9v`ra0Gw;axppQ3q
z(sb3!08C#{$Oh8e8TlTF_Kl!(mD|^UQkq9L)m(5!Jm;UV8=XFW8Bfnip|M#M+yry^
z95~Tf=_frX#_G&FA-jAD+)R1>;b&i+#zHVmi>U^7J#C=3|C9PfGPl8d&SK?#zE+)j
z${1Dx<x>0lxWlC?nz|VtX3G{}OCcD)=J-><h(-6wt|g$oQ8Ycgj^DjYS3pLySl1(#
zHv2~z&_RZy5Lf^tAy&6?HaY-zlD*%Z$!D3p5&)nKLM(p&<%?yfJw5$J_j*5(o#~Bk
zGP{p(G##`Qde=BHyBxEQO^LZFJsCt?@$Wc2xY85(=ztpMZPbNBtC|G1*xH79hkH8^
zug9Gi9oezV!S6S3Qwneqxj|&LuK5PDB09+WK(pPKo)N3)i5f8WY#Rqt34wEg$CloC
zu-vB>u(bzfAdyUqk8*)Vzb@aUK}KrNVmNI)C5*L}V#SsfoPr|%=bT{(`4*-lI}qRM
zJL=~flDk%)WUSG6TU>>2n`##mp(S76wXhX}#xTZ<uTgK@v-tMua4qXX{Yzc}8w)7N
zZepwcZjPT9F4bT<omfR-VDlL3kBd!!D&+Kn#@kd<p5$)%me*($Y+|uaeO_<l+Jx|%
z$FBqRRJ+l6OU^BNA3;F<k??8*5M+L~Pq=-4v9WmtK*f|HgIzm3ubN@7CMY&sNM$PX
z1j&Q{%onxVY8!N*M%I<!)L5>!WA%CAq_Osovo7U~TaX0$I)W#*nqh5lx|mcexz#BS
zWeZ76D5WtiB&o{$%F-Fe@MbF2Yyq&>c!Wu<Mm|%ljvD!(qYc5x$H}wOGwcbx2_$?k
z@ZcG}FJ!pTP`Ae7-LR4(8@XP1bk`UEY}eWct}0(yEwPN7-pO~Fxgv66@~*Gv{<l7e
zVoJ`0=e(7qs(kxDu9WzWbv@5#9D(F}Ap%`?{P_e)ID0L_W%)ZZgzuMk?-<(CqRJIi
z{&ld-k<}1A<??t3s<GI}%XwK!h8B>0fn126i59KZYE4uRPYXB4-(Y#(R4%r=q^G`*
z<6TF4Pf=?qI|e4cXY)8RFg|6L1DNyj#S%BuqFm3ZSOv3F`LojwTMJJDM`J<kX3$Sh
zC(hml6pbga6F+|n>soc}xOvq{XyvwH?wx9=8OZTSQ-mcCIB5p5V>dMoS;roE2nBs)
z;S0=Ift&)bP4N5>TEapKg3MLNohzc;S*_;VFVv#sxQ2;%1f^*>s-sF9@FyXr{=kHp
z6GgW60;L)9tKuZJAv-Gj#`pSq=Nk~*c^lwoI)EhRwS>wRC19Br29Mh<Tln$ME9RSF
z1sT!9J`nj8kUI^m)=tU1&Yad?ZD2dQC^yR>(LA5^ei#Qn@2+&$*-(A?WM83GzyatS
zwBp?S{&63uhb&3V(24&v^KvuI-O_$fpS3hY#SjNS!xHBS6C{N&Xk(i<teQYBV=^(?
zCTmNW&}ep((<};ri8d$RZidq)#jX>1p|TnNF47z+<796EE&Jr>>_(pKDN@9<g;m4B
zQmS7xWMaISizyK+SVDBOmmPj}B!<a}D<TUfhyqLlQdl7SXUgm;O*WllE2D}1D!Y*{
zq^amDya)hdU;oIb!Uu_dq=p<Y2Xe)Za~X@`2nj@>8JXq+bl9wiXM~JIcGVJMccT)C
zex#>5wzO=HSEw^h=59OZ;o_zE7Kz-)xRs^L+RTw<C!!o%HkDkt<%V3y(8sA|EuvE|
z!zL*MEv1-~Pqua{w<TcL6%&b08TmtAzox)$ZM!z?I0~~W7b<*qv^nf+{z*V-?#wO2
zjL+lRPZEb_eV+yr4Irp(DcRQYY+N?9+~w(4P-5zBe4qDjgohEYg(LSsiWyH??H}(2
zE$r}O#bcjL9FpdP7rhJB(B`e$PC<g8g}@7^MhDiLjD^oyTWxM#+O#kQT-EV1*sE@q
zCjKe;1cU)7Goedfo0?S)f)qlvNqz{tP*65nR4A8<bslt6k|GJ;`_w84mxR6m@W?Q@
zoyUFicYn(qU&UiV;GDPng)*eUE}Y$~{QV3(*X52w=s|w65WaLq{~~04|3?t*Kg*_-
zGDUN(>y=9u@dpq}8#O#>K>Y@6WMIhHIj;CEGtoNwG5zYW{eI9NnathZ5&ofn0FoKI
z#Ky~8i{-H@6c1O|<uhL|8`eVr@?%HDe=WxxZFG21XW7++HJkp+Z7sYASP_XcU1fG!
z!4h}L%gfWzD~%)mF7Zhb!3ePpu<XilvetZ3X|6MUI)8nQ*71J-&&1P#Sc-+a<a@=7
zyo9hqXSp*kO{U#|Z1DYbQF*E|nz5Kh)8?bRP0#A>rzCV1SfK=GeDEPzDZuM8OUh@O
znFKn$%}w~2CJWCa5^fohc3vk;Q8!l(2k9|Szb0LyS;7tP;;Wp&k5up1w2apJY#1&{
zDbH&EY6o4~8el`5eC%?V+3CJX#9_PHI}VD$BS3kFSn*%tdDgIQwPFr|9~1G3{wk32
zv7mboNH8L-eQqOLd3Buq-@&&wjrYB^N0r$pjfCtAHM@tOw{O0w){#e>bwaj<o&NQu
zRJk5!uu^j+9Hh|wG#j6GcYdJ#5#O6IDXvn@K7+?ySvfiXY<l&S2WSM)fI~TRB2E4x
zLnq~zBG1cNZs_qconp!yhhO&3PdO?ip!_u6s`BwD97KlI<{?W+2*=&di(ZSFDVzJO
zCcVFPE4eH3&@_~@RM&D|DK9jFqAy4cXHd;~&&J$o7RoK>WygZLg^FC4-aKZ-;UEQM
zp^3q#6MS1zQN^G16+_ES&rI`ivUg|qxwa)<*wCmIdO85x>aWj2Y{UYGa9#YHY2&+E
z%t?M+U?W<khsCLqMjywbrOZj=WtMntwTeaKUHALn|8H0l<ygQw$Wm?cxv&gJpWUKv
zz8bT*+L-Db_?vUswqleIGAW2eU+k>4c=&X~TSvyp0&_9KA=!*U1R0{r31qME9w|p;
zswWaS$gJOL==~DyLaMBFV2)iY@sB>$!2Il4<fRE9OOp}$XKMk{(spB`-9fz#f85zk
z*9Mi6>LHLZbSzD^rLKwSD*osaW%eKJs95dS71OU+(i#WWI7)W+V6>dCLeHAW?K1Ra
zl6|XO9}q>Ee47=U{%}jL>$qB4u3+PrwzMm~lkPN$>Zs<pH{5OxhK5FPDuIoJH#dHc
zOVGHc1gi_*(Ss|d^lVv#F8QRVcg)Va7XN{|j>*Z(R+|5%3i>K>-9Gg8xIL}}Tf4mm
zaeltg@I2zEoZG)VW)WxV>-i|VwAg0mFXHH46DF~w$G&KtIaXv9G4fWV${rkkJtJpu
zg|{m5>XLrE77*I_C^m>usP)bFVs->dY;O+)Ilg61nY>~dpRj7Z2qKDD<9;h`64jTh
znZeE9Rh<CO7(G@x0`K1x`{pF=Iy108m8VGkD#n^{YO!wbRD>s$HGHMESq%vwV}xWX
zfhf?Z@5avVfK-cqAdAuR(rWXo?lDN1f|Np#ygX|l9#H&084>{LFdxP0c|@PNEu*bv
z8ZMZ%tJ(DzWWq)*)7bdL#$8$={57MXpdb)(Dyh<b<dWVY&?WO!H;Tr5TL>=kt*&mE
zXSJCY^UFAo`YpF{%bM}s_j~fu@+`>7K}fUp#&3pd3pJh}MApxx_}yx#Uo!C)QOQNV
z^;@R71Vw-8gdoHB>$a}11K&CYt|yNQ3pCRhlL7iYHy8I`-5hm{tbVingp-#*B~H@+
z1yu_8zz*=&fq;6ePNX00!Y#{Qn%;2`OHgiaixc#hONAMZ)N}!w<;(E2(7{!L<SiNO
zQRKe0H4WC_{l-4XQ3qj)gmL>2j+yZF*s9MbgHIVO!YXR?l*-L-Y0^koWr><Y2~T2K
znH^$C`>i(@@57o1aHGwJm006j5?^d35K+$B)fKrFNemAqOwE5Y^KqG8`L58)fg}iE
zYL><t5^D20kK^X5Siw9bK>!e)4KHE4ZEqEnw?HRj8V8~))f8&Uq{2wes|HE>+~p+W
zf7P3Oi@C3Vz!tySVfo`I7~LYe_iIImqjHR4G!Sg}zUH-W_Ys@^V^GMd5L=QaU$f!p
zS)$1tqOg~@6b!F>bJ6_mYGb_CEZ=92Fd94}9;PJLf9&-O#+H*Zeg!g{lXoSWrJX4y
z=kzXl{P&*(PkpYA$+_K=#o3aayW#qS8)#bAa_T>aS?Dh8i_RO&3Bb2KZZJKwq(o8T
z&|_=cMRMs~Irler$RPKlrCgu&nxlx*EZytE@bvXekc@yNL;mnb<tdjxu(cUi$n5e3
z7!IBRD`jBoNq62=xjmH93P0&_PaY(<TN0jJOk0;7G<TYEGoIor-aPWz-V?(;ucKT-
zm9a;&Q>IXC#c~;$G+#)@Z1yBrUI&Ml0z)YTTGD*?TfuJ|=|2SlNAuv%<V|MsBeS_1
zrdo)WM&(n@T$84sniPGB+v$-eW0C2O&C3(6BXeZG;18`X>;F->`?PtHAd64TGNjzL
z^C}y2*{dxst(g)-O7C(i_GxfO5$Tfsc!bT(eYIvOZYJ3I5xKX;c;g|35;U1Eq?9*k
zz$uxW&|fd|X<~`_!u41g?f;J=DNB7WF_dZ@F^rDd!wz(|6A)eaB8w7!j5$P}Y;iX<
zwyrvgkkO|P<O_JUi#wg1htyYmZ)ki&RXAY!JdNT#latPa01N&X-ExWUzMT|xP;8W9
z5ZH6%`fCD1kFE6&XQTpn;0pwmc&qO&RJF%-qhFmrJ3;o`_v>e|Er^ba+ugN-pUz}D
z7>%!I>`IOWmcWT7@-kKUyHEI+Cb}V^ao%wJ3OV#aQBPu`2TCd!DvCKE=pjon-TW}>
zarUl2qKRq@Fg}0qnvWvt=a9{&GliN!_cD2JzLO33%Vk%Fb4-QQjxBGe=OaucWnYGi
zAm2)$++8wq2{Su!MVEZc2zo#*)dW0krOECDn5V3|$t<C{Cp5_mkGV3@C&PY##`Nf^
z0+nf1*EU1ti0tN_C6A?7n^o_TipRfo@HSqD4<2elS{Bdk>O}8WuGS*AMEuYHKITVw
zdb=Y4jMRK5&{;aMC^J^zbM?xXYV*Hrn-0?6ScuM%Cs5v5=Cg2TIS2m41e(Qa>IT&H
zI~xtCs93l<IZlahko|^I^_fy-)+#9|Z1#o|8?~2JUJ5B?;~C06@AH%?WTfvdSRgU@
zWsx3XS!c4jxM<i|!M6@=NN46#j$>h45iQI140m&1{G`+cWMcQ92`(`ol$q+qeb;6m
z#Ff-4stui3Fj=`?iz^w4HsV<4!x|lPJ<TY@h%S+@cb7ij{iFvtB^drjWI3Z%@H<i1
ztve~J<vMQB!qPULq5osnC6#TDU*iIT#xukWCV58;8<YkxCg5kh{v~%u`adL#L9((A
zsD6oWu2!eq{Q=go6>r*@;_YLc>O+VqMZPm!x@$DkT?oEn!h4~dF{Ob_ABl-Z8_c97
z1=r2TPYY>ZK@P+{m8^{7h&fLzpv;?AUTOnWEL|2T=c+fe`(8$EUVR6q{U-j8%KN2*
zSN*Kan;^(6U8AQrqFJ#1?-EUFLH<T*BF%FAU_yVfST5g%^34zIemGaBS?8GST0^F>
zZ3AaUvEpw93LjgD-N(}6y13bR-17TGw%i@_d0(Tp+?|=T>xIU5?4s?Zsz!qefhez+
zQ#aK|!{x^T)jGd{Cdz0YCPdUQQX#Na1KY**PGH%f&3W;)DX5t(+4wm{8`<t#eZ<C>
z3S|(_sc3R{k|*uw#4^lwQ?&i|;rq{gNXiJpqcC+LA*Z3dgidy4Xa-XwvZHF@;PGX)
zyMB^tX`0FLOfeb1H2)|eVZprD9z}2>%{<kvtg=>k6yy1S@MTpcAqrYg7`T#4vye+o
zMp0glbfC|oO|~I^oBl_m{)7H!#`oHQbMHRL#(p-=9Q1Vz>7zT|`sl+c_hr2QMPXk1
zO`50jmwQgxIR)NY5hHJ;qIHDpHEeo!Ke+ZUayp9on`;X<jqw1m&?9K2OmcZZBU6`b
zc1W}OK4+*i(Ya2}D*U)$?TnAUMW7X<Qb&b_o^?xEmx9{1&_xP-HP+_M`ReqK9U6b+
zf11wOe>9NpF1Xiz|9rED?qQtHxe?ME00z8%gh_w+vd>8pN>EU0U!E;$?vlJ|(B{-v
z@ONRAYnl_osY%Dvl2j-?ON!~!tbdZa{x8yaS=n1%@=vnK%&g`_cAuw~X1Jl8=Lmn3
zi(58th>f{RpiF#pkc^!Y(qdRb<aNGZnzaJ!$!;NZ@-QIXtV^^5WmsIJaI}x_MfCDg
zn;FHfM5;5~@ObH|Lq3z!FYWTFLc)X1EfF}^#hi%c!GX-?x>U4v@H(DAdO-Q;?TP<U
z=q={qtlgx&-3%bx*67<0VASPX2cB%ErxUKto?qPo9qeJngxn9PX~c+(E?7(>xau`9
zq4Z6z7IDi15P)(DZW@85Cj3fC(3CRP-0uL5t{HRfZIGa18ca|13P@Nk+CBpGt%=(b
z^l>~eR08@1*5>{VW`52;dwRJ2uJHu!LX2z%f0OX@#a}g7_;;tLzhU+=4OuUJOT4xJ
z8gdsRGFtH4bh)azVZ}lG@dCDIqJ`d?`5*iZzijstAS0<!viWrDkS+vtvUePj)p8qn
zwHdvBB&S4t6#W6~dWpv+6Df@Nuc{l_^h-zuBh(_)-Htf|7JSUsJ;W;KE^oXaR7_!6
z`v?(^X-SX%)!g4yJO_f=goKrZJ}y&kFY%WxXvqkM7$+gjqgl@pj|ZByJ68WRSXj`|
zEgd)Vlj)b`e~OZBSvuh6ZOKaI7WbkY$;F6Aagg<Cqgf+Yy9dhQ-jrd1-I=++=j>oj
zw%-YM@d?SqBxeh#_!~1lK65Vw$q{4b*s58%1j-b6=(`Z3uh7_!uiD%RI~Nz%tRdEY
z^}}A)5%&YS-CRg{_+&V}xn7AUE{->v^m}{bs>(yR@ge46(f{oxN{NT-5Vm~!L#Z5S
zB%FiFz<)f)l6K=JFu3zbEk&m|_*vUOiQP=#ASi;C-KH8)lWa>kV20+WAL@q>ygW4f
z;}yn7k{h#t6)+_(Bffm|&(q8=d9s61$WQh(&*B4<YK&t-=R4!2o-;^Dr8-$Cg)cy`
z_8qY%7bGh|1#t(O&7WWyhSS7lro5$zAhx`wncz3&NvMLj-eMba?*CAf+CEpe*x%dl
zslSAm>?f7)&f0ke7TtM#=4WW@E*@hXW~JVA{;bDkS;(d=QjyvIr+ps&MnEaFCxszb
zrf)(A43W`I<wJk8k5<ap=!RMpqe!o-RxYi5+W&7rj*6!VjF2EBFe85d@&L%kD3YEV
z`;j)gefB6){ND2K;v(-CM&scm3NDY$l!iq*&-)!F)T95%jqyW$i=T7-YiG^EvWaRO
zAwG}Dz2DMjc0+Rt*&`a6Gx?{|yFY($qZ#KulbNWIeq`&ToS?rP*PCOYFkzKflrAy}
z>eWL~QBGIHm1$%DIIk`2uNo=SYer7pmj-jBf?LyJoGi+8TL9;X{3+H|Xyu|kZiZzV
zi^F3+`9=N4dYH=4k>k6SuyV%y9suJFWbMGCC(;sSkDJaO_fl|ODdIy+Rf1gLMwGUK
zd2hqWW(t+%jo;A_l!SzoT4+P*8*9NV%9eBxoyKaHm)U(Z@T+-sT08kskJ*fzS;7_y
zAaJzVxS@?9mE}-wQNYsUTZJQjD{0aRh+0PvdkV-tNj-AhaGWJww=4d4FeSyVQDn6s
z{rUZVi=k0XqwGig{zr;NX|GTzLIH^vou$IJ^Oi2ip&q^Ef2%Tw!K4(#a`)>k5vOeR
zEZz^CBy%@<x!w5zD(ogB#=E-p$qZ^Kj`R*E91RYjrglx6n-k}qM);(vBOerzq?1a+
z3Wu3o?~&;wRv2I}WM2lwTug2@J(8{XqMS6>y%iIwn0NP{rNZE#xc9HkxDhd&neU^z
z)Zr=p+YX3`lk-XR!h(K-@S3KFCzQw)Us@Nb;~dB%(+A?|Q#w}YOIXYs0)7cg;qyB3
z)^Tx5i*mQfWEqQcCc6i4kg)HHq~@4cOsLh6Q!dnc5U+jYp&>MIdGlOEeYmJI+~`j~
z8HFk~WYZ(!PwiD;piHm(M<W{I&p$a!)A#?TnwpBM{2h^1)5dK>+{NZ+Jfx=aKl}ff
zdJDIv+c$35KmkD-snI3fvJny@ASqn~q*GuF7@;6YgQG?d6r`lPQ<}k08tEL}ozH&n
zb06>V{0X~uT<3LuKl!}Z@Zrf;l3$tR3KXz1_It6G$ch7f3F|r9y=$*Z7N~D*T|R5O
z3;0~E=2BV6Z`RN;0m;<iog93Q`j>U{q6_^6>sv)PI{aF9C#En<8z0{1PRKAZwe$P<
z7UE$0#U|B|V>TwnW?T))DunLxQPkxn04Z7OD-?fA?Oy|@4ULhch&<NWsVM#@ao^)`
zmt$y)6()<U4ZfyLf+lPI)jv`+Q>n_aXxNkOK61C6y!9cpLh6i~^N@PCpupjS++cdn
zuk)^mfWx*c^P^*#tH*gh9Nk?8hadb>KX)8vZW|0YeDd$Vziqn=I9{KlCvrgN5D!>K
zzEOUmG#-(I#Yo3U>4>(&DCcIesYyX3bX0UFw-1)5H7i7AUoPx(WxRr-z`7;qY+qwU
zziw&07rHpZ0A;Tco3vy~Z(v9F2OapNpGavSE~k4>>nFMo6ev5W-w<oX@$$sNs%tOz
z_yc;H_U^&L>}<J=#;k@&FUTE=B%#=s(izYfHUKyz&E@2>=G(mf#U`3FaeMU{9OD#P
zZp^XcXy8s-DsIwRU;lfK{+$p+$oa72?~|C(;L%xRa;-A_2hH1X%Ty?(6UTPTlZK{D
zw2yp{6<V$PAtQVGU0Gh$q@Ml&!-mA9<Eb+b2)w^CdeQs(KZ~q`8Db5QxIFv_2Nun?
z2vn;4YD6~Yg3`P@NbU~w?Aj!i!?#e<r`B2Cx{8cK(cholU*}ujI=Ms0#F$I%WDGHs
z<qF|7Mtb?sqa?rIwV6>I9+l7DF;U{`fBwENx$=Y?tli+a*-Bd>+G#*X-1i&K{7ied
zS@@S1fGQ-*CP5ctkf>qPn0Rpd`eBP<bpws9Bhn=8fNaVcR~}Dov}+DBoIYdHC^S;!
z28_m$g0?YZx+qwCu@gp+jIDf|o@#P9;+b!iT|f#=KK}E_#@|`Fk8<)J^df+8@^|Fu
z^bedIgX3&b*<)vK`-gZ{KJLnrU_jAIF1j1D0@J~8xkpA6oPkfoRZ1Q_eDL6%f}HM`
z5eJvXA_4hvmO+k*8pI#NS4dD568Ln1G=WQtKb1!08*A7~AO*RRu-UC+Sq=gv+Ji`p
ztjd{_nQFZiVHgF&Isk2uL?a`}rxeagon<-vCO-4q4gnCu+2C>bB&i(4MM-;&nl*Vc
zd56F5NpAo&XvO!wjj>l1-+=w7*r}=KaG9<2T=jFCG;Nfy^<NDvXK?KC5c&OEUcLfM
z(`~j@(`1#h@NpqEDROF;7@o%q-Euaw^}RypQZ%}meVFy(;%p{wTqH%T9JV*5)3y2>
zCke9q9s+vs1v06xED<KSg^<W-AtCmU8I4FhXl)h9dmFF1K4EAh1JfrpJP465!fF+`
zW-Rqq-cPds;GC@5ie*2ehh(O}mAvL2%u8gmBb2jQ|2qA4>fa#iG}@i#FeEq{a9B~m
z1IE7ueX;M+Otn%`6M}@ZUIXKs+{<*Am=l?EC4MdedQqXgihrzVLf0Hexhjr!wqPCD
zlnwn?{#d{7o@Hm-?f!lA4JP3DP`H5?9I^=tK;u?`C+_^L8V0PzI{Wn+vSq%gVc%rT
z$$Xh`+kuMKNt|}=@lVTC=efabP{o60f>mewE?&Q{dN6@eWc7jJGqEn+0p5!Bmz(64
zG{S25L#ZGeB8VlU5-VHTHOC7)lgfv~-{#ZQ^TsRvW$a&f3MJt==i;Ytdp-+~QP3k=
zQ67xz^k$M=Dw4T)BE;dY;hj9tN|b{H4U$t$Qtj}I-^ARne9sCv6xUzoAb867y8^*o
z{nV~u;oE^i9sgA|LQEW-x>Ds5_W(M;7gf0L*M+-!MkL7M0Xdu7a(opBDmS)G(-*gC
zfa&lRRW*z~(CKusx)lFlU^d`8DGKM&@N|;oe8H(dp@#lt!JDhVz1Am0S*#ux%MAYb
zqL=sWJ;kUSc9tCMxeN<+Qi#A-oY#Z*qkCOG8zw~;Qj-hKoZIvlXub%4NjB8lEzD{2
zBy0kLg*M!C5eD_NubO8~$*v{iqnyJsAdy$tCOQWR)uERJdAp=tCJ(f7cEiFuK^uNW
z#cutD?>W@h+bcyMnM{IFUuuT-mMSJ5eDB#G{f~>LFvup!O`2iAsTuPc?^?LMrQ_%K
zYrVLn6p2voa`nApj;*RGJ3A_VDG~Smeu)vg3NMc*-x~y-l5sdW;P8;>+ktPz$~^iU
z4Qq~mre*3GTT40=s$WJIZWyK&b1U%L$Z!opIPyp_zE`Eaus}674jy)>C5<swAD7-?
zOuUf?d59>b*Z=T$T{rqL3{S*_qr>-(<!y>#nt#-%#Dc2F?r9A(!)$tuLrFzhR_zII
zQ=tl^@y6CWdxD4S>+Q*9{=dDdY+#UX3X#r(q{%W@V6)E<Qdr2YnZUa&8p5_tlR*>>
zwHzQfvdxL@whJsqKXOQu18g<_V){2|p<BFLhEinW=K3j=ka1|^gTES_$Oepw{bSv%
z(gAArZuW+n*|4J(0cGv)|2di9hpp}5V5YpJu_iHlBV!>BWZilf$S(HZyb;MKtph5n
z9u+E&HN#Kf0r#FPb{1nuG4bNXkz=Rk-Qp8cl<{%;!!~;5O(I>0@AH?B$z`rP%`4|V
zVlj_<6fOk(F=~CbPy^sN*>^?jr}Z&w-mN%AXcaN=-lVD$i5gl*T3G{>UYROkq{Vw#
zjnw13?g)Fu{-qBELxDL6xP(}AcTFYX!JpLM%b$wQ5)!fr>$bt^rur_6QNNwdW1T90
z(tKk(8??rrr0jkO(J*ir2lIFL$%nYT6$&4i--o(}AB<j~PR9lMo-KTWc5cr+4m%7*
z6bH{BJbH`c03qJ>Hc0Y`4E{8cS<hVFzU1c_$x#JKB;zUlKc-bUwG}Pb-C(_vlZRC(
z0g~Xdg-_qSd);cf^Mmuk7d$+c@zKIucD!3sntvO|x$9#KZn$3NIG@xpe?^q~D*nkh
zEPX@I?liGiuvM$V-noj?Thv)oI3D956~4CYaO+~&iX+^)-Zy-AVCO9GsO&BV8xA0_
zX1dm}PQ84xUUshW)H6kjqE#lzfN`PH&7Ub1+~kp%kWX_dW<PFy6d=z#thHEgm&V$#
z%1)3<@+hm*Hs29QsY0bdZ0r;9W39kwTO@_`y2)g|;h?{KC(vpj2bA;_NVDH8CuV~x
zAcz8&kHP-M@DSBUF7?{ptqb4RxR_=X_x1~W+%1Xk*H4{(I!xfHq>9J$Z>sPMYbklt
zila_QH!N_sfdQMH-h98;ycaKekE&6-?G*M|YP_Welm$PCveN1z%2HE=a??bJfH+^>
zzS!@$mK{l@;OlIVkSTgx-~}lDKJ(~J(V|{P*azSXBlWvz+`5+BB%;d2sn?xI%~sQp
zA!aAfW_b*K8jI_Et3?3N4ao38W~u5ROa_J(|0P>6t@Tl`jv`Z`>$`LGuCDvVH1W@%
z@m_hx$!9utz7t(taVg|FkvURXhOuIjN*_G$bxxl<oU1iyRBwtIwcAlwF6IZLBLsOd
z57XkD8|$t4*5bTcKr|t0?Z|ZyEr+DcqA5QOxLt@WKlKfOMohR#!92~;ur`A?PW9#5
z?TEw(U0af=p=nGnM*?={{I|C}C+D`5*Vl)$AwYrl>t`-F#KU5Vk3>QiA!L6<A2LZW
zQOo>ZI~8jo-tnSqu6tAIpWKp%7oT`g{P(+O^@q4$Yax}0hlX4&q@ih#`&#9&c{wtg
zq&oGMq6`8;zu!U7NbW}JypPg`IKwTE4Yti2rOYtbAJ||%0iKo9swW+oTqy$wKoO~W
zgvSt6!Og>T_|*<I8-<;=p6-t4C=#pS=`3H{I!YU|RZU@?v9QKdYBp35EE=h4EYmKH
zwPkf0%;VMRXDwPN0X|3)=8JNx8Tt9|HdOv>F1vc1qTnefU&+*kvx;v1(L=;P5xyS%
z39vDstmSElBzB@0>#C?0K`?`O;k4UBe_Bmm1cj3SysaQoZ83L4GISjVZ-e`Hc+i0)
zB>V5U44;138`@l5Ji3aU$DcW0oGJ-__dltCuRj}nvh@4ebm7pKVin+{29NR+i*mSu
zn495VOu~Z}dXiH%`-Q*jKka~J^%H)K5DC;3JMq$m+M!gchG)sDGtxbGwQAe%Db+@#
zDn}7*or%u^fP4iw3VS?5;(fXU;_CDm+)-N5>CZ*!iZ$zz8)36r!ha=^>$%8wyb~HF
z;PCi7eO}zHt*8w!`ehyDZO73erT7UG73eyN(LE5G=NlQ2cSr9kp<<v}$dwwQFE1pU
z1{+0}%D<T{)eO1XR7m42D&cK05}_W$gIOUypWp2#&beI73#T64(9|1rx-fMfxtrgW
z8k*JKv<^B8Trk@l{ri(_vxmC&2lJ*^wOhM$XdNqN@XUTdZ8jdDbyH`BzeT;y`a*m5
znkxyq2{#<L(^K92QB@QD1Qo^e6D$xlS_%5;*`UcdTK+-9;foFg%<sZKZ3Mff!I(1R
z)G<;$B9o}p$Ro(}?>MaD3DuY~B?Pwh&Fr%!ge1Jio?U%>)a2m6>1gPanNegs<nMF&
zg2ozPXQ;`$*+&CUdc|P+N%S(Ehew9jBMD6_^M*E2a{;CND$<il(T@AUNb=s5xLg0>
zISb8J`a^EM1Z}J(GAT<ezfXw0$Hz8W1nO>0sVgmHmnT~Hq;#C+v=RCW*K?o(p++?g
zCSEv(s{FhEmpXx-tSr-@OzvR6j?+D%)62nWFB!-#{Xwyiq$kYV7&A>XkDraU2fPo*
z0^!2fuA4j0(<o+BtqN`HYj629>el5<OR<>=D~$h_yj0ADLWqi2mENU?du2kHn6xf;
zXI!UZl-O`kZe%w4VLDj}M7qk;qo)HLFPoy;Z3)7@at&sa1eMS5W{7E%zHC)cfeO!O
zB6xFwZ}sD$OA<>vb*FI~B%w@%^M4=Kjdtzc!LiK;3xCf$Sgd;O0o1Y|$%)^}sZN+G
zA<HMs=<NC&dFTB12+*GAq@M0acZCJ7MY;DAcvKdFn)NteRi&`?>Zf}&lT^c4!upuh
zefgDo4pydBA)_|iEPZq~M{g8zu`UKxX$EPiIV^8kK0a24Kg*p}>w>2t>?|V(`-RKL
z*hf-$AG;XzIccpUQz@@-xQRF!nCW^oN^jU`Iu5&0$u~FMs!q~Kk<V?+j@D9eZx%Cc
ztuLlp<>bd5$`)b%D2wVnqjV69whF5P-$-IfHA~_SeTiu{vit5^o9rHD=zGaI?#IgE
zPgp<lT=A#dz311Pc=IBW@ydpGnXL5>^^~<XBy5g%Zh`Yes;WDApSEG06}lr{YPlcR
zBz{@X*@!esFkub(bJ+g*U=kJ1Vl@@yV!iDnLlW&|MdI;}!s?m<p3Ygl-0eVkV&mD-
zm^o^JvR8F!c$uQosSg;!Bwt6vG_>RD(%^p=BvAGpI&J>v8a|JoGdScQ1A{Yr%<d_*
z`ZQs_-ySkTJw5d*3BEZ`gR;IZmzQiinZD#JluQsMud)7!XIq|4GQ0I}hoOJEm?2J<
z&!*t14LX(D4vPYMpKh^dY%6n74N4xnvVY`y=*cI>;}GeEUW}lz9IQvEkMeIPYm3@Q
z*jPs9P`~3MgJvgm|0-F2rq8Ij;SeTIx7T6d-@1a^S?L`4m;K^_!DM;UMr)lKniOID
z+sa4=&i;5gtm`O0slVPh($`U9+b-C~GZm>wz@12s<|o5w+pZ<Gh?G`D+XiYg=+<O2
z(xKvoxj3}@Pc%TIiN3G2np9Qe&8CSBkeM&~jIA1Ww_B86FC3Zy2nMHC5<CFo5E$}V
zMwlIcW<K@qH!2=L7?X2p$lTjuTTFcuD)yy%WboYk@<USL%NAfMzh=`WpUJzV;yh#d
ze;{Bq>vd;`aJjrDUM51g7jW3n7|!cvl<}P_wd&PHcU<0)KOlJ;DgBf4T&a5l1UHEd
z9UtD}bvH~1rY0|2XthYQ%LBfdt`m244Tr8eQirES(9F-#`@rfr?JPugE;do$V?~-j
z`=N#`Y@2CJdMq*m&~eeR88+oMD(`D-OsD6^oJydc`Wt^Ld#e6Dw=V8vN{hgX{28W$
z15>1X)1{XwCOMl6ZmsQRE7zJVxkkTR{?3io{3pTs@0Uewm~{w6e-)?+1BmERiz+~9
zl)E^5;FaeGa;QGudCcFPf9NOtACmBiVN73Aknc*HY`ALNW%y@YL|%@y=1S}cSh>aI
z6{0ta)dOA%1;lX3=H{fr=CaiqSiH%+G9@cU@N=gc_&$5{WobxXaQ@X2%05B+1XWoI
z4TXpb@hnsB^J(WKO8x}j90}#VXrrxcl=z~07Cs-9G)7!l#^YpYHvVrY)9J%@Ox!8q
z0Mh!`F4aSbe%))}wp+Zmyh<{;9#2!Ml_k7?YgY$iua28X-5{wJYf-fcvhT{R2YRIi
znjeYfD1>hL7PE{Mo`-0E+pBPYG36)HnTN5gtFm_$&SmaaM%ua2VAOxfj%EACKf}-C
zoqamAS)M{Q(QmV~rNtGu55{wtxz$GD?DKSNZ`*(3icwqk=Y{JG3}VZ>Y2XQ?1BUuR
zZn+7GI7pmk??$xD7ptso-jDK-C%spa%if9ae7vyCrksf*bM+#1$YRM!P119xChvbL
zV@_>hE4b${%IA0O?_YH9oKVCjY#VF#QUW9~BsItRP;3?$A^Ebgf@mQPT*xNRXWbhR
zC5Mm2C?J_2v{F2W2(@UOu4-yuJeLDfuIDFi2z8US(L{IQwiT7-i=?^xT@Pu8g}HO-
znV}a{(i8Prw6Zq)DlL(Ppr4gF>;pWdG$m;wg}DmKfdigO{VX4+lTZ>UM^nc6<Z+E8
zkIXf0hT#i~a^8YQcYhdgatS{ex8g>!Z&gBHsU@RQhBqldEfriP9Or%?z;DP;6N?20
z^z{RnbwY!jl8llFKyL?Q!fi}fJ^7{wiU_0st?WtZ9>g|fx5y9Q-3|7>0lH+rF|acg
z=2XCz!cjm>!*C}1MiRx~tY(k02s})=%6J~qVuBA$ZQSWOxlM9c%kTBTWw370w$#h1
zJ4`W9uUSM2IWf%l9s5{MemFhl1hyJT6=TxdW}k?KpOL_madUq$?7IEBw|DR%8N#Rg
z)K1KPTn_xBlU4cviw7?O|H7od5`)dnjPv_eoax_!Xn*S_izqtjg=epZz}!QQ6S@sS
z#}uGZrCr|=F2omG_W=+7FFGz~xX7>$ouCIJv*F`hzt*5mKAYJ(-*@rXH&k;Ya~mc!
zR6qt-o_?#3GmDh+Ok}?8i~o0t8g5L`Vh@@nq%1!g>%Eg8)mn!Q9L{wO&5$;1eQ*ob
zCZOkxf?v6O)x>|tnyx<Fz1BWAhi8qXKHRhw5oPVBdMFZsPpZYvvGxN$xr*!k+y2DF
zM{cDIP8I(OA3RTgK0|V#pqwgdwOMLLa|K7p9Y6dQ>AUy#x-u<Rh7|C&!bBp|{@Mlf
znMOUs96!58<Vn(l?ug?jk&m2kQagMk7&opn(FC90OF%BTj*2Gl+kwN*5W>&B#{%Eu
zwAc5%h7SdW)hEQb*1Kjpv4!t(<}LmxR21f>1NB{`MFX&+TuP)I1d443s&T7pyn1NN
z6*(rSQre;j*76-^aas-njozs7e+($&7r4as_)IZK`=GH-b~V=Cbh_H~O#c~^%~;!>
z&Om_huV>xMeMjg@0p3nF=Cpn-6{-Pwz4RrZCZLD4;UoZ7|C>NNvDdQs@fhtLZjY4~
zVo7q*@Pj{YFHFg0{#*Sv!{I!#ahlZ5DGP8#@JSm?JMFx`0`%^LnD+lWUTj*K>bW#F
zQ1Za)YuyNQL;NCR*A+;H$6dl;*oh23cLOJOBOTTpFrCQV#;f$i7x#f$Zq0i^IJ|R#
z$hUu>Rb9-w3^W*;pq0YoovY`cLvLr&<NIUX&=n0E{h)-NIG8x>)A5jr|8-FrXqNtR
zch{AAbg+r~aO803DrlyYl!lW?X$7wRG9z9%LU{xqqGI^nWEwg8Yq7z$DI7uM`qDW2
zqF;{WEz;c+(ouvF=yPS6wrM`EU)+47gbs|3%@)>-^O0K45y^I-{IpTLqs;f?OO!`P
z^cJ{Ey*rZI65UnYWhppw%QWV-B36`@s>j%VXhkPOos{q-p)re-&CS8GiUvyuCRHPT
z>YJ595AOb84dyz1)|9+f=V}}T-$|3W;CSk68Sj%FDria!^(`x1{+13{<3t``_4ZUE
zYSgUP5h)2at_5jPvzK2YIg=~9yxKay8kPn~=Q$7HLXc5kfiv^$8L1I!;v#hm%`2y_
z^KLYNbgWWeRyQ0HvO>r|>@5B`*?`W)mALkfC3tb#`6?#9pYb<jDDpLA>`@w@NPm=I
zaQnCys(OJuMoBq>H!h2yB&xylTT*<t5*MjJj0{t(rW?3_lW%T2GyV7`F3inY7cb~M
zlSDp_&gZY(v(#&9J2(0L58}?7Vb2Y2*<1GUUdnw?s{qKtn>zcpU0p6P_sidpZ@Qsg
zO<1jxEnG0Su3o%O)hZ|rzwk#q0i(?>UfQ$N=0vhctN}46afm4j-6%oaa(psh@mzng
zh$~&hSh}q3j;_(nd)$<)@yq!<U>hB$qfoUJgM(k;NLxJ;TKk~}T3a5EedAsB&T^2X
z`vgM%4h&iaa>7MlrhHxXkNu<nUnf=3gZT*X`+p7-{}|3$RzdFdGa}CX8=)QA;2M%r
zeHH@J_ZA{2J6<M8ip=8Qfq!6G9>;Nt_In|Uh{BhqFHONY!0RHW1eR4pib-x}g&)WC
zGrLLO?>hvTIFF=8!emLf;<9!7kd)@t?MV$;iLCOGIHF4>X<C?AMA8#uOH$6bjMGmH
zWPd$<UX0dp52)gYY6HJS#(`OXqCw4JTtl-xk}kx!dDxUTnFibR)a`tNz^wg^>(a4y
z9;_p2xsPiJ;L#!SR!10{!TNH59m*3oOo-pzm@}rqg@Tq=8Yk0w=W>%ME<5G2mD;GS
zA;U)%wr+pVFqe9Mju2wy#&O<nfnHbwR{>E$2;-i2{F5$!b1I1+v9NQ`TdL?WzITk-
ztWD-@y7unrT{iskLq5&+OdRGc4tK!|PFvsKnIH~fQ6mY(b=zrKs((8XWqDx=AtX>m
zX$R$mi(@YK+)+_f0vDUf=Gb$E&Mq(;I(#E8oh5H(?0pu#m{E6~jnw-0!MGYF5jK9$
z1^xW~^J{~`T#2rv?Q+5&@(*@1=IYiz=59;qlCK>04b8>*z)z<CdIs*6Dve+4Ma$Si
z7Q28#X!0-zTfB+ZSwAJJM?b4d-)Iy^ZGwHE)dw#!k*S{ZBWFKn1(SAlCNy6@e7HaD
zEvcj(ohw>jYSH0^4q@_(_-Fer$NmX%X_CLgrgq}9Frvs>=lBt3ef2C@E~QyZ*|27I
z5>4apLwx=LFavp70FW2szMqavEOvf1cb?o|(lj8^S)y=hQhUJJ?jG0QVbVh~&8pf$
z-<Gbn4w(OV_njaqp&g9P;XbeIp8n$kf6Fn6jae!U0P`AX3~s5#$qT)Q{VHG9ow>ta
z=4nKpiBxsq_f5kL1NE`75(lu}i>Y@%69J9`%CGh`V#M=Nw7t~uu+SA!(G?_w>N98R
z4FllK&==zc6AGoIKiBO=*aKgYxniw-Gn}^%MCVGGN_E(%?ED$9*xr19;>BmaTf$em
z>0k8a{r1yav+`C92e9+-D%03bKH`r`TnqF$CU4Zkxsjpc<lXTyUIWk;yKq4*kt_}x
zU0g@6T5XVaXDfj@5xaR|4vJP_mpylaUSuR|qpjTJPN7Z9HdoC-M}|H7uJcwC9JQN-
z#>FXebem7c(4(7b*Bho@7t*-*f8u?Z6*)Q}P}I0_6J+D|%lOR9Y-o6W*&u9i8ja<K
z7@+zmqZf0%|DTbiQ7gn*C!ERPa4osod_Op9(IU<J`ijoX?9ADV4qQ##Hy=p;X_Jf{
z$j|R36#(g8xI5)TjKTITnrOn=xeaEfL%j^GQf0CM@!6wf-vm3V@P6x94Kgec(Y1a$
zjl4fHzjvDVKTd1AD{5OD8hT#WAqQWzPu>z6#T|{PGF`^kD&e>u{6o0l;doKe$DxyD
zK)S$TrTL_!e3{y?1m-zV&CfmEn0sjno=`47C)y9`yz3jXSe<)K_H%gyLZeLcEkaaj
zd$5kCzJAKxr-J!Lf0HEhBkY7<P?MF&X5%rW+MQGd#|9nu>*Zw0_&*<b6~Usgwuy8U
zMJclwnv1fMJP3ATDyw77pyXTQwUMeYATsLxW5KI{x}h2tNFsrRs*PGV7K`u)qH!Bb
zSv>+T-4w(`FvgY3#iaphk#XhkgMD3T3{ywg^K!mU@)+~{K?KPB;P{-m!V2MKIpOZQ
z?GtaiO>>*BYMA;HaTufCIz_D4_K+W&JVy<$0!fsBmz!4V-(OK3O|Cu3Aa<^!yIZ7^
z!I~7?R*3^HaR<X5uiS--%0;IBf#K{^2OXYCg|959?8*z0%^_o{lUr`Tys$Z&y`>u4
ziS~;bSPm}b*=z6n_~r*eV2TLs;Kny>eiCH(f1(v28hrh{X*ZNgpiG%d=$q9_-sY9;
zB(5crrk524!}a*N)0p8&;65*5;=mHly4GhQ5&Pb;(@nnznqC7P+bc!k>zx&N$RG2U
zZ~9`1T{iZ-+e|Y*lyfIxu9Rj7<_+DCW#`Xzd7u?(LVuz6X+qx}8?-CO$s}E^G{Vp8
z|4jEA^Z&Rtfalb#7=dn^T0huCO2ZGoT7dqg?zG1pq%}0p`n#qMTFZT<ZF2jw_nF7$
zO2JUWtMuGdlPnMk5z6G9gC+J8^9PIzye8)@;^^6IM+_a5(6?+j*cDuE7JV)7RW5v3
zB=b6nAekY>V&~IKL;iMC(9`WV<6lbm5|m=rL`W?(c*zTwlikv|2VoJSwrozU=>6Qs
zi;Fs)6|p2#bnuJ-1Ax5B1^j=X#I+73XZ`wfewqh?+e2`fE6#xS`=x~&y1SF%yT<{y
zllLM4N0@-2nHgz|DU#+x6ZW`e9I-jH9z!RNH}l2&llD)`V<aIt#k4Akn2BjM#$OpU
zO62SbBg$hLc*{~a8_23-A-vY@DMC8_FuvZehwWaSFHR@qgfW9sO7NAzLm+gD5r6*p
z3M)|r9l5Cd$Sgt;OfmlX8lRnL5VV=76RDuy6(2T^PN_oub^pe9sIq;foG{Wflq6u*
zY(09;u7h5YXS#rs09pyuC-M>DP=$kViQ<31jX07hYC936LpjwFI^C^&thqdHo|b&q
zDoeRisR0e)Vct8LJ^PdY<;~0vq;v!;yjS%kdKg6RHD7TUYik2gayb_V_a+y}VsI7c
zt$r!J^Oen)c9^C6d+hZR7g=-RiKcPjQXz3QdOeO`QP?u<zvN#SshW<F*y^+V;r04p
zoXY&5>iZ2s=B9htM;a#&wjc;+ec9ZAy$6g9Q!e=EQCKFIO`aiTKD)8WlQ0UWQm&0k
zylEYEm4Cz66vcVe<#;Sw#tk1n$3fv)`yc(rmYM}7H>pE2XPQ{U4!^)2n=Tw$o(Dx`
z)t${5^+7p;Gk0t4Mo*oaGQYRn#CJkgPlp~F;CX#M@Rm@pWnzwx<IYqdom5+R+wqcs
zoGX<Kl#t;mJ(?}cDb}$Kaqh$}Y)%e+E6R*2rrEa5Q0Fju=9I^%fBW+A{%k8ekWG(|
zsJs`(xIw2z+flbSyA$is$&thu)77N;p{&U(2U2zR;$mXD4i}j<QnAnM?wq9Adf5R#
z6MMRBWv&^!boOzehg36eFBVUZ(+{N9;i5M>T=;|ocbZVt!6thxPkn?*1$Yg~+^z}u
zoF1{uTyGJyUhW#Gg-QBfpGx*X6vtajwxwzJR}9OKv`bPE4YZRyygX3}FaBN6$dBat
zOTbE=ogQV^xN|=00OeE>67EYj*ZG+%Y&2$e4rQjN!`SrXsp@A<p%v5R_>V`60hnMa
zY?Kw)419DmO?j^n&@p$|elR5I9tu>yDD6F`*MPqgH_&8RnGqx^1`}Qln8ZgW7gEQh
z<|@|H{H76IsjS?7YGx+=MyyI^c1cpn>uj2TW_sbG2i%Vwcri2tVvRR|b1-`ZzV~f4
zxrC*ba5yoMRvp%?ZUtB0y5WV8Y<*Q6B$I1xgMGif=qLvduaIUaj(D))oPQlyy~4!v
zVO8@<lfpDh6rBsI9A&B`MXF08o$jxAS^r(^$kNIAp?{ecUeY_hc&orJ3&Qo+<t>U2
zAN(s(lPdL!#vIo9$ft5AUP^8D&aeWEVsawvV)lOWJrPYH_hFl!UWG{{MrpqFGBm2p
z%y3;VOM}D-0}Ha+4;iwx50p5UwmG8h&BI=oRMh2+Ypruwpg?BZ>e{iL6jm$Q-TmYm
zMc{wc4XsTNbXB}Yw;roatGrM}HQAV^x$xP^?3?u2z}>G7mQ%j^0iPz^jyr6WNAq-j
zHPEiCY(R=&Gq)s001X)~m+R?t>s*K9RaDMlnRl0Pq)4v=FSX%>U4T^icaC`m&R28T
zw_lbDzzlR}j;br>>~!p@eCaf-jyfzn_n$r<Pv!7BMi6it!9gOy?Wf&OUVEc9o)I0&
zk3}mTpv%7-IDNeFb3}R-dci4`ey#_vRwFq*?+v7?#JmmY(JsFdb8MR?L<(N@0eg!x
zd4jNEIj8^4gF;;_6HP!P5%P^ZC=7)rN9HDJjt<_rsj87G;I%4^329m)d3@N0m8_)}
zTLhi2+<}v6SJ@PLxrG!(lL%<oCB0Nq<G&tUA}Lbl3;2-W!B)3w>(w%)%=M9~yhd|F
z>$;oZ@>rdCo!Uy*+WI4bKQMTKVVJZj!C{T7mEH~~LU}DU6P#av6-~@nSL}})O+CK$
z;<E?(uTPYr&!kv4I-+JvAK{Qj9m@BH+S}_C`H>Ot3+lzvv(`?}3{v$_IM%niDj40h
zNvB<_5s3ENR^}idV0mw{F;il+hf+ZlsPuS{B^L7LV?)d7-OWOnPrxo-vE$i)N}YNN
zq41&dj_cA-MGlJ~qjUNEJa9DDoLJncp130>tDNz8!iXLH;;+*hp32IE!!xwsQ0i8@
zckt0ED`cm;PbJ@BdCyH1-7_vvwj$aW6#wp0Ia4X;CF&f)8zS;fUD|6zg=lF7(yaes
z#!6$#<<<WhPoloxAuz+4hg+8-t-h!6E4zHXw`%ACT}0Dspk-@}l|JzGC6Uqlmx*e6
z_G?eDR97*+g-D0~%i59WIs&8*_Y29mJhEJWT?CQH4+B`mVX%6C)eT>M{?dBtr%xB+
ztQCKz{hfIZ-ob|RnzGCp0N_dr64<ZTPsi(_rHpu=?YC!d9D_ixcRXM^s_=DZDbN<5
zX}3evRV@#9XgLbI2xOX?$?3y;A2*frIBBcgSMpPr&dN<kCnH5PG%khY@jNm(dw!MR
zI}$P{)|Gj^m3K&W8=p4%alZ)M*Er=<I>Lzi3-U@LbghrVqenU8`lC9xMlF2!5;W2K
zjItaixZO4-d`w4|9~w-Nw$-aDbR+si#h!MGeD&7Mgx8uuHtK_r+8^4VZwK(-#jzTT
zJ9-N6U7v?nfXhlg^t%<~P}Z_|FnG<c^j#n#M*D__Oe~SKRB<1(Y?kJFA>+@t&jMsp
zn-`}Be4u}Sx=)N$cPrnUzxL~?Lh*S0Z|rwsqpbzZz@w-1VP&onLL;zDQBIyH(S2o8
zM3wTN)Ir~S%lXcu2DR2OtCyKATMx)lKTh)U(Qo-}YdZ6f={JEQ@VLpc1wjTo{+wP@
z?xUSY>I0%f)<b+GPu-A;XV^BrFj7nqsH|B}4fuDtxbl)y45K$*$EA@~2vyNh)F)~Z
zZd<43*2>`b=7&js^3q3x8?C2WFPFV*GsE7m+Vaa`oEK&RhSS1axg*8J0FSh7BRqs-
zST|QLXQStMlP0}>^{_nE>k*}VE39+J;XmpBixPvbh?-EZ%#7a;z+kVF4u~{ixZ?7s
zFiz4KSnQd%GzFS4rZ0Fh*xfp9-u=@#EhJ*Mv=etl&CRmwO9`LYDXb%LqQ3(q6{&2d
z+*!jATP&KwjVKx0vuv_xnwB5fS!i9+yt}@KLjshrC9(4-!2;%oS3~07vOl?Ou^VUs
z`3`x!NW9LaZiUM`4L9cM$3D3#bCWq}%@tt_by0UOK|7WHD0_wC>35eMEKc(fh|jyp
zQ#Bzn9ZRI5?zhgZqU5p*Z2KpN)K4g#7oj)Ps{O8e)tk>us+v-Y9OBsyT~4c)-R%g3
zmCEeAde@4A+RsB<pKb8ylVQ;!($`5Fs{2wke2uaw*}rLjwJ2;UmE$AWETf%6pUYex
z@yL_)ti<(u;$Uc+IZx<r#zky}N0G7UK6dD<-)b3r&tDgDnc=vuwrU)H7A%F=@8#(c
z(gzo#vs>)gKcG!;xNMknZOYKB2c86ss0%{?Mig4_r<$fnB<>nM`aEr)WR8g&2$vNF
z#JGZ5KYy$LMBGamRTGUe_6mRjCPgnB@7_#~hQQTj7ORd<7jDy<%FK_47H-90;^NM4
zFZ)q;U^C~-bn}2W9*jK|T2h9-pZQDXw*ip>c9*|;j;`*rhbji1^~!fw)Cwy{oxLEc
zSZn}Dbe%7%;1_YPc!=6nA2)j+#?O^oYzW}BVuJWGpuszM@7#<wI@`TQh4Ke20DmD}
zLn6uES5XA~UdcOFtJDdgxEkuamswRB-)S-Hc=Ao-Z=dz3>+G|<g|ZkZRa3~HQXI}h
zD`uE=M3qi#6$OtzYGobMuc{6Ft$e4=VV@J<ZreY|_3A5;fU=-cRLwKZq`(M}PIMKa
zr8ZfvjknOFK?t05xUA~c9qA&q|4Uv;fJo?C+@?3`a7a9Sw)Y6T_n+z;TU{co)<-{J
zb#;dfN0A%YYXFPye`xG<$(@kp=IgI$csBoRTXblMM0A=*wfUDl1r4@iCO4(@XvJ!}
zvSb38sL!3IBKs=?MA_KQd=P_=D(#u+Di8|-)>6X-1fIdKiPYl`_u#iPbvKu7_xFbh
z=KlSf{=TM_UH~{g>zSF`&NKMe^PX>d_V3!F)cghDC>FoAKgQk|?CMm$>9>9xXMR7l
zaOisJ+Te37cHIO6duzfPR^}i*5$;lEvHOZH#Me5Q_^C#qxyQCPx@=s|lA#Wlenj;_
zp;S5cm-tZkemJ4%5IK=<YVEvuFV{=@K1Wn=qZza3-8tjxLIQCX$C|0w(_Nx^zZ%1|
z_NCIZR<5SmhcivH<mx5i;AS6!X`y{i2~7RJkKxCEPt+Kl=b#+x%4#AVMo*pBiiJ7N
zA$H@(^<CtZGVF%=2-7qotB0GQp&Y<8;lu%#`tiT>k1||8N#Q!bFJ;Y@@B9Cpcc5cw
z5w;E^vD*CTvG{2-3`8;ay^tS#1y_|2t83~0PfN$W@Qs{DeWS&SwVbq`TH*=~1rgex
z0aE@Tfb(<l>yK9>kXPE3v4MCJLf`ycTwP_!O)&&=mqsT!c*Xktxo_WftE5aka`u^#
zxNW{+C*p*YJo<cKKM)<pYJR`P*e68R8-e!Y4-G^Oh5AGSl&+&o3A_!OsU1y?x55OB
z2(>}|M%8>tZf@7t&KAQVyYCQFYUXvmMM>vCQi=VICV1(7su(5NxV4e`@-!h`;~$}V
z)gX_0BU)I{Pp?lx{h2?p0pHO@j-#+xGDxkDu6r-I+846eov4oyYvTTD4k}M8`A`@U
zqE{j&NWGMs_&C+Fe9pxUM((2#;xWf!u`K;rCU!Zzho;ff`aQws>s8w+{v!5Lm&lO9
z98z&PzPG%#i<dn}o8{(!v4S}(YR0G)>JT4M%)-)H5^(RYhe8d{2VS79H7)V%e5)Tx
zFH2D$EiZNd!t>s3W)}7|*KbcnxvpX2gJ!^|O}#<3Ug->6)?KeT8xI}=bRau>yD*(S
zeHl-$vGa^!U5x;(8nQ3l%yv;LEpb4qk8{SURAl36>J9WypYBh4P<c<1a>U>^J8mAa
zvJ>bi_lN&@*PkXt!KsNCxap#qiCMN6(9#&(5R^Y7d*a%pH=I7sjd=3XWrO@p9$MU+
zSmI~@0AVvZk0+W5h`Rb^(;gC#>tAY%Di+SyDT09ao>%Z=(}Zn53d}E$H8x@<#K#{~
zw-`j8D7z8Luq9c<_2Xjyez{{}mW0U_rB3$@*~>Mvu!RLV4uSg=H8b-S%vT$f{)=l%
zQOuKW9xJnghv`8b@$bCI1TW&s!5L)AV&1Auj^8kqy^4q-zlxjoq0`;6OF~754-q`@
zjd?!c{PVyp0x*#@m#;@!N-t~M?P(d!xwt(*PI@uxy;L!zIBm13Faf_N6^)%K+Z7uV
z+Uw?(`#(I7j+euAR`=$fx7|rvWS-@CSVKKSwx!bNMe^JK5C{K+aMf^&Bx!dT`!86#
z-kNtn(pNz6VWv+<Q(uAnuY%5l7dL}!uI(r=C0VhE>nYk-aw>8B&YL&we?D%sfLd@y
zBU*UQ6d9)or6-dOg8x)8vK}L-WAOLCvCUJ0j6YIdSLyN={5Y++R*B#JX*sQR7x=gY
z`Fwr2RD)Bjsb_LTIsHvyE~4_&HhA;lBsQJd)QRH9^A2MPaay<BSN6shmuzB3tUoR{
zgt@SJe*d={q`xPx?kfm5ya+JCG!PAhgkdJ>LR&GFO-w%l82Wv)g@gMW*E>dV%Z9ht
zL*};j2DsaJg0}VS!hv|erTEQw+cos84%e&cTMhCBE}B9>6~aeeE=WqtXI$_wy3C>{
z4xeb9oUaB?ZV)1NBHj6C^!f|R?fR5Xk#u=`fh-8Sxck@roN>Oelewq4HW0br)4b0F
zamvfM5vuggXA7gLEDCd<R14>(=ENy8zc(FfNi^4eSe<ECK{A{e#Vr*=c3Cova4xgh
zsGmLa4n)`zUBlFbgw^9H{F-U)<OLS<@sr8tjV38#7Im3Aq!>VXahr{H<ifX!y*?Ii
z+UYsq#}Cz0fx{DMDlCk<-0*KlnF~@R-sSvB%PhxJ4Dv{NeO(!!4F52qxI;zjlhOdm
znzDP+Z{C+?a#r%H5)yc7xziq-KT`CjKZ*k>U0fB{5vjr+Zww}YCc2S0@%R~u@{KM(
z(pmjI*!1#>acY{sP1Ar1bAr$0-AAIoZbdkd<|;|NrtM9``j7?bIqt&lhXZ<t?-xO3
zW*!Qk1EG(XDmm9uA@}gMOVoXz`R%ogRX|e!Nv}HBBQQUGNcLM<8HQf@j~^FF`Cpm5
zdZU?XM`X3i*g7pjm>Ct-UVeDeVzq&LLvpb-uDaI~QRX`DE`@aq3l6foGmF-h!HA5-
z(0)n~?F`Ut;c?DVie(N~idwbn`>GRe0qggd5!MumXz-W={LSi>tf<|%m6foiR%$+Z
z@e4LD(a*KYCB$G$1cTvxZ<u;!Iw^F4ja~PKB2#);!#e3f=~+7TW$Az7Mh0J}|1|=I
z6c=X|o*N5)@48S?W*ZbNoZG3yix+W#pFHZ2bVowW>Raor)5z<@uP2PSlC@bEj2t!L
zB3l_WPhT%eGOKdfvO6ezX+2)W(a{zq@X;6jtNevB5G*WpTlpqx>DV7&$B_K{rqk+@
zFW}f9pyPihWfK59Z?Qi5_rUC$)Sr-dl9wf+zp8)r=GNMso)RUuy?Nqe;o~L^MKbY)
z7RL84_q872X*W19y}-Q9Cl!!?$sf%~fa0=QPcfmfl~)FiJ()=<W;aO>T?i}x^qPS@
z>H4=ACDMvyc*3vr(N568vSa5T7}lfYuFc<w{r>8^Kasi7yrH^Zm%)6$`%7i-zW8YA
z3dGe*4q$ZIn>wX~0`bO9im>_8jK77DnLjg#6&>Od0jzZ}SWf!m&E3G#us@wH>6=f7
zQQP#N7a&r!O9A(1PZgA-?nlou1`&K#_WU+0AWZxd{s|k;-6>E5<!Y-MKUBJ-0#Ul!
z8A@d^OdVunw0n-bP8~QhJ>%27kJq{~UX(1=)@V6u{`WGFXPOL2q8x3+WP2s(2Kl>E
z?lMoeAIpDWITeq~Y$`dwvtJKlNMr2-DUl58`=;y?7KI$}DT5#3<~|gqNvQ%ge64ed
z&yIYviLCn<mhukDVcYRMHS7G-<raZ<TFT19eE^zZRCHxcR+&C61ZWo<$pJO?*~jas
zchE{`Ba*A)^wz|@i6gKj*ELAqTz&h?krW*!RDpw!T^UgStqfOWyL$jaqNl{41uLic
zWvPPs@jwJv(b5@f+vWIVv)^snOr0wpSL$P)kKqu@`@PVzfK#5krTzQBOCagS55(tj
zDU9*cc-89Rh(iv1{F(~|*7wrm7KrYH8kuhb+=d>nZI-R|Hw$9@ViIJ-NtXxns=TV2
zKK2Sp=<;S$G?V&$7YB@v2k6~@l*=BEc4YH-=abA)kf<{~6YL#<WJ)?*?Zyd)Jp?mx
zuv4KcQfyhx446_kYR})R$f{r%_<|GhxHz>3E*9}j>k!&n+12k<r{z*zBTmA3Hm7Pu
zl)0MZW#6hVfkS#Kxqpq-aAtc-X#Wg4X94Otd78ao#;CoBdgND8sw~IyP^16=UwpG<
z&gG=<@`zNYz~nG|EZofDZ%I(zY$i&X`48`3R;XC<SyoSAGQA_aV9~7~A?w(VH0O!N
z1~xDouj8NXx!i^%vyEmlZr;kEYrw78J{G>ZZK@kIf|PdVXDwdvVb`e1o6d#%$W35U
z()<@*cKNjN>f*G@IH9+M&@20v4yRyU^db0+*@=s;ijmV+Hy39V33&-6X|5tvph7Gn
z=1BUie|L}v{UkuU72AD}N!l~@k*l7tGZ+}#(l3at(iV}Xpfyh_mQZsL5&n99($V*!
z1!}8oyAqgsWZ$ggvy=#3E*>qQ9S$ab`>tdmgSlG)X@^WAvd<1JLJLt6M!hhtR#d8T
z=nzX7iJ%BeEG=~crK-rVNy2J?VV+!6UhmiI7lCWh34?hA1wq8}X}P&Zcs<{&f~$Mz
z;&nP2%1(|%2?j3e+jG}w9TAgbJCusQNO#xvk27NvQKhXL6^qoZrW^q<-ZmlL`?<0d
zB4R6_kEZ{mZQ@``ioa}5DErQaFViN}zQlA2duxIF6jfvl!y4;|Eqz+fZKLg~t#7uJ
zur-$#L2DxBS6icNGNegP>(#UlejPP0l!nQq;>n0cY&j>iZ5<|U-s;wdiQeThQlZ#a
zl%A+3o&K$SHQ+zaTD0X+ME@h>Juxtcks#>}DB!C|!z^f}rSd!XPwD9nWH5+hZ@zhL
zE?s=K`Sz)L5K!}vLNv7F(Dh<`;XZ8>NR#T7s2-C1(vI(wVs}ES+&EH9<>mQZ-D3T`
zl1){RxQ70>5J?BLg1VY;-C(Y$QqBfL`v2NNYd<S7%J)Cu_C;=Oy*6~K?_0IexLVwq
zXgpQnFlZ^YW?`O}@DG)xlFC|C3Iy5ZQET-BSs#cxhg)s<vUw2tGu3CBecr^RrbWkP
zGh&eaT!FUD?X?67l<V&IJ@!Uv_Dj=`nsJ&HPzB&(fT8LAKSjaTsa%RFtFGf#f>-B9
zmy|lo!CCt$tm`SIpFF#(OhF}uql4j7Ga(fcgb-|*RC)Ku<|)|_Uo)!H;YD+e<?O5J
zGJ>SZ>ubY8y3GGSp13EmD%>Va_+rqDq>+K`%RUP;o~gK?`frafu$AY%*pFn!`<>y2
zR<c>^-i8Ij=&(3f|I^U6i{rLsnVWv|qXdE04;!z#G*ZP?_Qe-{mEnLyiYJT`%?5j`
zF9hET@~}1If8COa^L>!y;j7mdF%PQE3hNIK7)A86bp76q_v;%~7)eEcJe)s_G6zHb
z)NZa$74{|gufGsy?wFG1{2M8&vSwZ4?TXp`@l?B&@tqsm`!SiAqA{YnN5nEXB3@XF
zIo!_eF_lH<v4s+qFNZHYEY&Bo!3P!=>kaw*Rv-wDPbKx*yRn0^H|bZIju+d0hbLP^
z^dR_k+j@5q2GC0<%C4H8ChQY^v9me%)Pu#zR)a>VYnVJ+1=F%tV$VUzGSH$EM4ja}
zqBfF>3<+QK)O>FdZqjJ|+c!PfO-)%}H<*jPeMvNhbQD+q>|$g3Dv;R&r%i~x0~W5x
z)?2%w6B#R*ZGr%d8*4blzy;@r@Zf)*aEwlD=0!`mI{idf9u`Fp+2ZATa`w?a@mH`i
z0S-LH5|VawQl^Y?8i%G`xLMn^^{a=Yq2g^VYo`l$(pc{ImK3}1)K^ecdo2<626QBw
zxBl3Fv_*Cieg2iE1N!cis#l7Kt2_Q9XAU^yS5h94l)cX?GD3^yY98;loB3TtuQd(m
z-x7xs)iY1~$Q|pveJS%DiBSIIp^d4bK)Yn=4nJ)(fX#Z5-nxCs5_yEK#IB)lOG}Sl
zjJMAE;MG|6iK5U?8ZQ5f4@Q9O=|+HW@3;%f1q%#A0>j;a@SPaH9*5Q4C8GQT$EjuQ
zwKJhm@?`b6xu-Ee5d|_8WMb!1`|d71m1!@wzK&~Qi6qp53H>KOjv4nU{ee$yO*%O(
z&Sc07Q?yxMJnbamDH4^+i%$uav7>C}cqH%2?1$P>qsbT|CDls_SKgln8{<|MOV`zr
zkJ`5|O8s8zW-oAqijZN+|1&}p9=ck%V<fSK!!stL%D9+bqlmu}Z*!4x+5_&|5_mjC
z7g_SJ(payAzPTp!FA$@KH=FiwP;wIT?rcvgS+wGi;n8fe9%0MxH{To87@E^%cMC>t
zvQ{{ga`Ibe6J(b0znVQ6*bXa4ItYV)b^%I?Lu%waxf6<x!I3&P-`02`OYfoj1N?aL
z>QXoP?d~e&7S=xl%^u513-N^Y3uoqtD+Qistf7^K!@^}pbisx%R!Q><yoKQf<Y^Xt
zoiWjC4}K|)WO@(gaK&3mbqL#xPoRsofc?fz-}$hJt<kX+b(4G(Npt_&N4FEXLx!L#
zrolgw+FsGWFuS`Z{-3QHPpQgY{g~J-opGiiW-Ys4qPtr$zk42VmwzuZY-YL`R5f@R
zua!=!Et;<q8S(E{!2JO!fH7=^-O#C7sj1a!^mMf|nB%}mSt|vnS>&<j{T}>}PiPU}
zyw6E@5&+BS%>2ES)lhw99#K~OsMUtH+o*&WsPVE!#1dLBmI}}W<S!2P=S40!WAOr1
zhNlLFov&<k+{{W~r?&z)?|;bu+e+h}H4^F|r5ITi5A+F6Jnf6F755-SN`4;@?U9C$
zSH{m@pH9A@hK=&5SXhXQB6or(@zFFA1U-AL4$L4W(-6@huaH1=Q&N5A|0V|5qV$Zl
z%O%c?vgpktu9biGVf7a4cFD@$mdcP=-|Un6wjNAAX%A8sNu%e|EU4OxPCB*0{0Q%a
zwLGaxAqX>#>wRWP`$No_TO}_1^li<Lgpy<x7xl!66`?0#|4*trlwD#{+e_6JZX9IK
z2Z9hC=(~G~&m9s6T+@}7;aU%b1Ap714h0Xd3|ud+&2N*^J2MSHBK2e)1)2AV{tAG7
zH$+xjo7t(R;CFWHtH;(Q_2HyRUPKa@&8`G(1uGW-3CK^E9z=H^Dxq=~U%4w;>3hWx
zI#o62QrtPWX)xt$Pg%^4G=9_42q#NC@0(^@x}h8!M5ugusj+blZ{6N#QR-KNEPG1{
z!pVC(!(l&+vwgC01d=9p0Spz+uM*?C+1P5%>ghHkzcLvQDt}QVG=Ua#i^o#D?IG-O
zWYTRdqSY#kyG~=}tWHUxdB~TKs%%0Ap^US!&#x+To`PGZKeh6j<hT1Sks;C{m4Sbi
zpB%gTlGKfDBVzT}cuwKcZ5H-b&NS%l{L`Dn?R<u?l;@Vy-V9DzJ6(X+8V)>`xmGG(
zsUQu&-r3FV!d+V1%_zOF31aB9lLk)QtTKF<E25v<KhUr2!}H3?Lo5N?(&Zw}S94S_
zru9v=GvNbzM17K#E4?y(%(&4Y<fUp9FGkm!$DIWaOS=q0mtCkgfBznzXw}qL{>V;-
zlB%3=pAWNaRd+Iz*1_npd=Ig*5_g_Dk#7Cq22XWRiSF$?6r_vTL+?TIlPdQ=muvSF
z+rMPu`m0Kc$^^Wk&)PnC9XoDpWgxF-;KpkQQm&=Xb$<T>U|^Pz=BfAG2h$bw$mUt%
z_b%bUW}8vB(85Z91`=}Q`S_QnM@OZ;?yhs;vtp_`lIN4>lKMZunL72vV^^O(Z*HA>
z4e^<KwLonOV;=V&2nk<?^;FV$aoJ_la%Pi0PA24Yjq=ZAhU_Ewu8)KaHH`X6-FZu<
zp#KN$KoY;gzX!_?kD3k12W0a3uAiI=t2cj!SZ+KswQ<3HEQ@%=1#qcchX<D}Yq~mr
z%5|#z*jSdvXQ@r_y7pBHElAj~{DE<*O+5}Qhk{+Nm|QMdcTN^f#TTOeteKc*rfK5W
zzk1DI{pH_6j+PC$*>F5eeEkpK@T*_{mbZ7eTwZVKd}n7x?HHQX_)Kenrkv-g-Zzz8
zFE4X^5Xa?l=Y-&meBL9>F_-(hGl!W<yDv0hRTgu?Icr3;8v+*>Te?ur8=$ebt+ue<
zN54Ccmq+X5owrQ>%`niF%)&TF=9oAhXO81+jYERRZ!V03&KoXVakA9|QfgMAh|zOQ
zGsnY(#*9~tzfS~*XG~UF;&B8C2zeDjSu&vQ40om6xm2dO>Uh_3W=Ij_x#`vR8IIlu
z)5nXWzUu9cOxJZr+@G;#7ct>#$+&mM%P01pm}ffeR!sm*?{)}%=!}5AG;mm7$Gw$r
z7pCWe_rbS3CLBMF?>C+tSWn}V4UzNggW`x#S}MC2a9*KTro)ljx9|9iKlv*Tw>OX`
zQ;fD(91k<U`Sn{4(~;wJBx=S5Pe6?SC?#@q%+p-*>nYLHM2x8zSEjsY^rSQN5??oG
zMG$ACsjg$=Jcl_GrwFA}P)?qd964DEJEG`p4z9~#z=^WkI#e=|nb4a9%VfS%TXWY`
zopl%BdhWje_$@C|Ie9}h%CZ_ct#^X+p3Qb()AtOW=MZO((}CN&8xDsfcei`G5YTLj
zZRer)p1$kYdMhKGq8SANPY_2^a|Q+9kz(X{Jkoc@f{`^+`Bgj~rzY4$tlUo(<`l^}
zw<~d4co2fOaz!bQq^!ljFtQrQV5NLckZe3CUWJfL=OJjMqf?p@@447*P1(g-sC(zw
z?+@gx#FR+M93z@CDW#HqhZzpSX(UlBqN~Lzf!h6}#W+6GqgVRUS(LX7t=ioq7(uLv
zz-z{%1ZU~y3yu|6I{ajxnc~RZ%`JcaCx6Abzk@V^2;JuL9*$EIIvLE6b+pb@(=>y(
zGKOmOW}*?3dQ+;+KP@e~Se3A)d<n!kHeJWXupu3e7PXu~)mzjt3$a81GnQe5072=z
z<K?qUG{fE9RGgqS<4afBH`VGpC3Cm(LK@NR>WOlLEay(WW2%l>aqOIBa6Y@ZG9_3n
zQ}8%P4u_e;e&+6OV(0?li-F)fGkSv~^qqyVQ+2*%Bst@fbAiov;L=0C=?Ps&N*!fU
zO;N{N7g+VHGqO0MQ?k<Rf;}nC6TI_8JLOQoJ0q&rT6UsF;#`fGK6G@>7apCm8s%)x
z#poZp0R{-_>Adm2SfDyH$ApRydRrT<7D`ooTLwmLm{HtNt{4s!U%;wcUelSY6RS9{
za@m~6ievZ~nfwD|IcBEgL_~?2Fyi+_Op18H`C_cbQk7IIUtlq`sPK}H!g7d8L`lIi
zHggy8Ufzr4>L72wpDfripXN+ik2RLh?t{+-%aR2r9>L+%n+#(y`mNE*DVDwKa6X{k
zmkV@O>Ll3$g{q*v8OQv%Sgtp*N-2~QU(ViTOPZ77?5!<5`fQ8aemxQ_SDEZo2@0KL
z0$OsC&Dr2d*0EU)$K!G2um0oTS+koof+&742i9+Y^Nz!K;4mHe$rrB(Lsz=_Ee@<W
zPpE3+lR_aDzHa9|+0qzWQ0XBYQyzAuc|qqK0bz<6P1fa7ohM~y+((r&F9|wMxXyF2
z8@S#MJb!jc&Pqyobw1k{YQbaM`|p9{^$;<qQrCA}UF=Hckur`m`@_iH-N<2_KxXpL
z5uSJSKF}5C4l%Dv@-6u5y#?YmDaZZD?adyIi6D+mAFPBJ&mo%gxD&_L_v|$;#)G2-
zC6*>D)@a4BQDs}s*{Ty9rTjKy?YHfsoj8VqV`tt^J4WV+57s?VRSDVSa>m6>Z~^B$
zzBtvy%_zz|PlV3ld;n*G{*-1so)EUASRG`wv-hd<u~wHfIow?4=T69Wo7wvw5ssh6
z_Zv?Rtf%qGh7=9X%BX}Gl~A1kSuJzIVWkBcaW3P7aNN(leSIX)W(v_w%v0p{?V+Tk
z4yD^gWS*3HRPd3>!I(Tbn~Il+7VXtyY&+`=l5;+(5U<ip)f6cgIxaUmI_IGXR?UuV
zF%NPmM2K0)$$p<=GUrRmI6N1-4VlPc3M=7#syw->fUQ#5PXZC=kp(rCvzhApfY7EA
zDx1RJcLUA^ra5t(Cl1qr!+790PRyg=)Pn{bp3XxkN@Z74_HT{{;+#p7<K5wo*ZX(i
zJ$T1&Uccrz9r-6e`zhXgl8z<nE>^CiR^cxqCGw=il7Z#K5lX7?rW**I1D(nxn_}+8
zx2RE8XVLYdeQ0UNn{zHvX5&P0&Ppt^6a2QvYbK-wrPkk+T`du2bCl(h{@ryw()h9<
zm3^y3zLx!CN`XkS&vAt>r9_yZX8*|lpS?eSmLy5;{J@Xe;T{q1$fK%Hpn(}291t7|
zS9V2qS17Xo`BG$McV~GGxg?k5fE;qj0npubS7yHVBEsFwRQCrp^N6g<ZZ;aIMrAV%
zc%4}<-x1;NYNq<tuYSMV{ktUd0{vMWv;FVG3ew>S@zIM~i@RWbGmbmMB1C1{C6;61
z>o1OIt^Q0-%eUfKOi*pK720alP)K4li<_{vc8#e<*K!H5>(X@=CzoPnWIJPoz;)LG
zsqa6!st0f{$_y{AntC#>9cPK!Ey05^r@*t@8=7{@=f1AVJ#Qt)zbeCeYYm~&x8vCG
zE~SK?h)U?)bud+yx^P-oj-_yXI&nU&ltoD*q&SmfCdI%cfgSFSB*|FU+FdvO_f=6^
zuMfWZ>MPp1@Pm(E5(UcXh^e94Ge`@V1T6Rn04TLJOq;venJ39Y1c^u(dsIMVi05`k
zq+aA`2N`pWUD$O1+L!CH)<TStm=m!!Vr|~F4}qnWo&XQ3hM-MpwR$(WJIIdfrmayH
zXI+MH<*<b#H-q!(_j725d{e{8Kf-$&$MAPJlmCHYxr$LOy2g_>LhviCJ#Hn$ahM{V
zjji9fo<m<ThVYxka`W+^;Y=r{4&uBkmV0qDVmZ)7p|6hWUr7$Qyh29);G;6*%fBD7
z+*JGMbHIpb2v`(%9>l1vygQctYG7E8mkxIJ{a1SDru~j%xwXc+wzp$hE<9-q))n{r
zUHE1!U){FQMWx@j6=axvcr>+EoxlXnSkmj9voIyNyO}7bm9M{eq?W~VtLaE7mE)Tu
ziV$V;#szE)%4U@s9aPe;f9inf6~j0T@aoscvA-TAcN2rC(m~eF#M|;%2ezAuGj+s)
zSWzR?Mhbyvw+Bw=m1UUqfX8yU&g=)T^tX>=Iw02{ZF2Uhh9Ch|<6KsrPA8s@XHLhJ
zWvNJP%#xYn<odJVdG*Ff2I<$C&SrFAPlzdS*zdX79|%-WE1M#%A+7p#R&Q1}Rf<&?
zK2Oe?N)j9#j^3Cy3v`>&?hq@zQNb0%&d+25uIm+?v*}v6LakkccJKSFiP2I<lTS3%
zS_vt8gJ1nj4S0N|ln9}*)jvH?Gi1xBJXh=NdQ;S-KilJ(v29<u&aKDiz7f{%V@Q8k
zzs33hV0~EM*^<8T7<fS>Lj*D~G)lJ?IL*hDlCUm~r`Ic`cxSMN)kR8=D|xpk?{|Kv
zv_>mNEfA&A6jpU1(58xwWLZHSOK~u&Cw}iLG_@HayLEL`7w-0)-^J!#51<PY4**MW
z4XEH65J6zKyJf9Ds$hYwn>fZ$dYS#L&9`5*?XwO>T>vXak!2Uk^!0j+J}rS1Ga@eH
zxvqutvaqZxYgwtq!Kwk;L@{?oiQW|wfwdHtwNcOTcs}vw{Dj1Wq{!E=U(;&gXFvG~
zA;ztPV$CZ%v5RxaIH(7kMu3DT%Tvsx6j<%(?^^;QnGjO%%yc1W8wIC7Glq4++t7gw
z*Sb;jZjVffSQDvaw5+%QXSj&F_St?1E*n#FJM!^)>DRoeAHLE>IsKy{+w9bMhOh79
zxFNiC@Jf#dvFk7Tey`AD#laTWInd@^{I}v5!?Y5DG0ll}sXV+oyKa&hHC6Omaok0<
zD`<h1JdRZpgEcb*($z)*B|W>mF!_39uvMgZQj|h38VzSJRP*G1b5^<(?w6tv2i3*3
z7N!`uIqW%|7kBXL_wB8F%oPZI^{*Ee!S(VYZ6uP{aqMg#?E=Z3JjUSeYo?WDIdfiD
zPV2(?yl`GBwHTTpg-i^oTdRBFJN3~8T^dz=bW2fpPVB}BuV24L*Oj0C<VU0+tWS^L
z1TcCr5jxZGl1#o8$1+9`x+Ym?LtIQ8goxjVG5KWTMA?`Up=$sRCZ*#;##$>`ggm=0
zUyS}|==)hM11P%~MYLKW<j&^oHs3hVT}#DO`v|OTzY7L{7&;8nfe*Rd<H6E-pW_(*
z4rlT|SS$<8nrldE=jO?HzNLdDzDaHa+TT7D`-pq|4P&{x3id)&1_Mg%typf_3*W1I
zJ<er&OGxGIGx?iW^m@<v=Gpyyh-GKZ&>_Qsg@BH+CG;J!EIm2`KISyQ2ReWowoz*6
zg|Eo(HkP|oL>NrU4Pd*#qw*_adF<7fV)$x|>^&)b!?yYOYTCMvn+$m>Jf;+^k`o;E
z6J@PDygss?U7NI~<_z@nN}PA3dDq1sIv~|t2TdC^4YcO&oGDy#Eg4b~+c^ENj^mLJ
zrQoB$q0$+@f;UR8$6H5C;~H{iKFU}dDF`?FxdWZ8V;X*Dk6(1{n&|7djN_}94sJS-
zQg#4O8_T+IUKWn0Gt0S9)`lddWh1BT3?;KZN*Kxb9#ggFF?HNa2XSI@2i4uQqt>JE
zb-l}WF?Seyo$K=&LNL!$llT-w3UQor3@5_SM|{z4ghc3%k#ADYFbn<PxEQsx=32E~
z>SMeYjc&rxs%v8o856;ORtQ2$(Q|dXI7|-^N_%}Z;yd(*W(bLJ9mjgPQ~rsOtZ_9G
z*}vvJI+H)F-+p}nus*EsY+V$Ku2>7^qh(?dEc(W3jU<(ueJ134Qiv>TV=aZ&8mIaM
z+6bkQ^6XtA3j{UxJK^>=aC?7)SVP-FQ==*5!%Rql^YQHNLgso70q^L+bz&ytTtFm{
zW5#M_U0sk6B0>DnvIZ7=2LYq1dACswU8{?==0b|Fn=(ZUYirb6silqvME|woeBN*=
zx*eZ+$00jfNFy5?#o*=YZlhCUWD;>!*>YMqm%>^ab%j=(qAlw0D@g2Pd7d0SJv_0l
zg-<^Foby^)s_^-n*L;0`qJ`v^)p#d-?eEFK2JbpMYpDYdf-|0IG%BhUZOSgqOffTA
zLYk2hAm+l8J>8B>h(KmoAO{E#b0((b9}O3;?AK8dnuyzd$4JPTTKsoTJ;83;U6cqu
z({J6iFs7?X1*eCXQmD0e9x<Z0Ym`4vy|WZAg5okjt@hi+&G&6>T|(0?p6=si(bX6b
z;#LGj3b(sV%Fhu=EQ<rDO&hD75pCW<%{!t@bo7up2lo5KGzVgYqQ$$s5_?xJ(298H
zkE?>C=YS&zql43?o@DLjOccRd^Ls7H*<C~91qZFQC;!%@YjAbIOkDRXM>wH0EodK~
z+bHdF8OIr)+d}nSVccJ=j4c0Kam<!{hu{LHYpJwSI35?CmW8Kv;dCsNa~FD-=tZwY
z?^5?TwyIQB9-fXYrSh~kUcEW;*~2ToI-YoQURaJho*xeU@zaT2PA+n+oh{kpSY^a<
z0xH!itu?GDbKWtfnS@hQjSljS>uZ8^zGZZ}Bo#wqWS(c&)d<1$T!x+za9qz!VoYHV
zDU#<IZ3BebA<>IoRimwykOYY^<$xsVPIbOtYilg$Go`GwT7c&4Nptp92>lM4A{O3V
zeElA*cb>^U)bX9fa*9xN-PjlsoWWyiU7yE&U!@aaMt$bZ1GX4QA+d|oZy3woXlhLe
z4Pr5fep+A0vTC80LalYX)uEpq*BS%A?o7Ub-EZN$->X<|>Z8o6!wGaa`*p$Ucg3<r
zqyn)*7$=*5e<nJZsOkdN$nQLs*YiTFD_BE@qop<9YrKorE8eRcbA_INQ%+}NQ~q{g
z^*di3V`dl-c>d^)$$^Ae^wBy~<_TinbGSWFG;lgEtjkJMWzhxcEcKWsqD;=Rt(6=c
zG}Y!$OslO6;CBXX&_2ouY8A7=a+7z&u^S}O+3Zt@?9xPA3Rar4gc!`f=wpMuL2S+Q
zx-sUgrGdy*(-p>O<8>VCH#!TyaU8?Oir&~;{i2J=rrmaYJhIlp)A_{H<B8+LnR0e9
zasf;OQ=G{mBI=rNyO^1iGp`;_i$kq_rcYxa?I!jjtfg=~ow?o55R9dq8BD}+eI-Ri
zoYB~Dy+YIGW=COWPLscSRYG_CmH`f`n<4m_?9&8_5JDtP6A}WdO27d)XSf<`Tl+pX
z8V7+X@5m{#uGQTU)muR+xQ?Gl=#4NoW<2Ii<Fzc#^p(!M(uNPhb`8zJjot+9_?5Z!
zi0SL;)er0Udwl?~KCJI-i48QJEwS5np`I<6IaxP|GNnN1Hm#+&Q(o1^x>m#*sktjg
zkif8Sp6AHnX5#+-25p7qSf~vLL}El@+<x69v%CNe<18D=vY2h1(z<mBd_#2*22YHv
z_qV($v<Zn@>%9Z>fx{?J@Si(`kKfLm1PgB;wBNXj^vl=(cKP4&U1PJp?mAo_6wW?q
zw4yF<tIj498-lKWfky9D43_iC>9lY@t(<D*)ZkPKYpc|r94w`=l;(OkDQrc!+UR~=
zrq%@k(aSYZNF=5(c{kD8+d38f`@$F>=;?^|pBB2s^61=12k+?i=*D*R`2k92*Ytz2
zPtt51s`33YM69?1CiK#!?hkfPjyJh#F7N&u_xE~@Zc)FP?@N<t<oz&;u}in|A`0wD
zu3)Y=HYHArgMVx7MQ<yJDY<92VWWZ@yE$_>Ow79kDB9MoC>*Si%N6eIm+|daV1qNB
zWb5kXVYS^n36f16xIdE}>@o!l7v_caTO70lsbR47`iBniYFwmm7eJS%nC*>>_i*)X
zS8?3IuWiG;0PL+PwKi(0)Y2$L9b{{E0l~&*G@dh{)kf<;{JB&fPYaLd6Hn{8f2<sE
z0p_(7l6VKzwQ=5#V;g|mb~F6<B_cVbzMpLEf5a3R$!qA>-s8pkHHj&9+iwSqLkA8F
z*QpEI9U4O*M5tXuFNEuR>90|n>!E1;X6N0X+t!-vzjQ|BxaQ;W5v-F4t^)L3K-PO2
z$MEaV<nJVwML}2p<xQ?0pCNAUJLrXFeDz`n=p{nf#@c^_SnfsI(A4sYR{NE)yqStz
z-AI#Pb60;&EMJ~0uHMf-ZY+D4UhC@(_V&0L_KsM-`n+{VRO?)bK4KX98~jI!<yt$q
zzS(&&wCcvg6RZm~U2e`>_v@Pv-223)91PIytm=E#RTk%?Tv*p3jLdQ3^^@n-rMhG8
zS}Vak)=hhEjbDa&%IxNeTC4Zx#^ZSX=*WoMZ76s<8n(6VF_*kFbY^tx_j;TWm$T(!
z$}pH%{TE$gLVVi6&6zH&x?guDzkLzk9mm^?w(FeZ@-v5)?pmsUtYwR38?36Xa}_$H
z5$z1ep-lxHbhAzG%}4kIqOh!m^LgR!F!zzju?C|G2;O)Nq3Z~Cas5D-=aB9sNS|D=
zQHS{cOW#9B`)wp#K7G-dR6YMT^SrreB}wu<qW_;a%Ok4Ab!Gd7x)>mh`+xo4mos_f
z)tBe^vWE&+=WK7zzw3r>jLNIu_F?`0uMYs$hxMJU;j-7vT*S{{3dYp^<xOGf9bKbx
zb92k#?iv5fUw_J%U%jC?6htEu&B)C-n2{%^jfW8V*&qFwAFA?4Kl>S4R?e@#=JDyo
z<I|bfOW|BAr+|x(nqoEzD%lPZ2Pyn$2+<GL93vGMly2R1aC5rr-J#Z8M7l*I={OKv
z)1#F$Wj(Q+pB#EI+s0^`wCy0j6t9=`cO)i$5wDVgQPFQ8i*ARtUGLcDqM9KDnl`jn
zmSttB%36e064n9{B^WG^C+ZxCasV@0RlfTCH4l$RV%{-j<G~hMh!7%qnhAME(vB7<
zPPO#=t2A?_ThrF>tG{ly6Ca|Zo&fLJv9}$op4<Yv3(UH?N<>2>T0$feD1cFFb3K9B
z@0}_k&!pgEjV8gh8&FCqi25wPVm@XuE^P|l-EKX3HX(-SHqCwJBjreIjj}FiE2z4r
z+R%rwpl_Q2Xio&dfi$OhZO>S^y05?D9RA)e>BWmSqnM&0btO<hJ()9{MVIG^X@B5<
z{oBv@`qcxgb+&~hVo%y;RYW{-4kG-?pZ)<qDV4+Rf%Sai_2*ykbUt%DFTLyQqR$e<
z$M`mVwf_39MqI$JZ%&EZn;YWVD9cK#LQ@A{1$Qe0Pe@G#8dw!H2t+#b4>Sf^OibtG
zZl2oR3j1#wz^9DPWPd%LeEBmFwnTr71YZlj%QpHjo{3>}Fk*gP8)a=QXJcKV6anpy
zUn=OLeDUe$?CTx8_{g<@48HjMD^ABVB_&R4;TK<i&6}qqUp_u?Y)}wRYUJ8@csgR*
znEJ@GX?3v2S_g5hGc|quHy}idOq!Wmc5tZcp8!O49XZ8Efz^D^F%wED%rP=eGa&@5
zDJf@iaItDNVXcMGvg=`pATiK%p*16?PTA@EVV7;97G+9)@YLG8c<WiO)`GU`>^bw3
zWlOTJDxa|{4~M7$R=%~~+c^IAnLJ0n(^!7^;xitPCmzpd8m{{&U86{~g838m_94bw
z=x$DlyT3y$59WpiFXT<+Td|ybp(Z{4F1mR=KN&I0<xCdd9m|{UA%c&$UpQ|6*s-j*
z!?UV0hMN`#lzOR~NUm>l9n0F#;N!MpAdSv`PMCQ<G~9wqslVG;HltN%QnpgK=q~kp
zc(t{PUG8hQT;9&)@rb=w0o<3fO?oV<Qi@@5g5dcQ1ZTqpzc%xf*&S~9`psv&d3>OD
zfcM<<DFb#0Q<mPqP5k+v{V_j2o;jXQEax*153gC)!cteTan|`HLM;`Dn5Az4j+@IK
z5Mw0g?9Jr?RI1ShAh;ITa8m~93@lu$BXkK?+!Z+m*KKv3L+y2eT<&?eY$E^Lar_Mk
zL<YF!uPFr25r+xTdR$p*<-9gZ!Of7mc2g9OEY}fp4bLd<Y_6-b`A?@6H6u?GC#^ix
zg|E*i9*zs~=RfDw;lT6zzvvF)o&&Y!jAkVIxMk>IpjrQjWHM0`G=JU&clI3bOS_-6
zT6BqyzKb?@TJZZLi71{AN9kWD!sF@0oM-awj*yfz`2>oV>Unwyh&V%*)-}|s5&~?3
z^&KQNRm!?LQK1z^J>yJD5OrrwkLRAJ_xov`E%z<c4c@;q`NR6{*9QRW!}`wFwe@%(
z$=?`DqmweI#xXc(H0OkwverWF<6zoHxkBqefC#k~mQ|Q$cQSc)cMG*rFy=HdrHS@%
zg!95vj9U?AUCi_v<=_64aG|Q(rI*LmLG3TyJ07Y%Y1Pr{k?oMQL1DK(3r;U`W$<4w
z9pKCQ)qR7}NqW~;T=M>J9Kz<wO>0W4ZaZ#QuXWhrF6Wg}=Ix%TDzzE48Jhjz3Q<XU
zX1CjWq7)M$WkSqXERE~#iMhD>_-|qBNYRUiOeV`j;oM*ErK4`0xuHm(uDtH#TooJQ
znq&>_T}efo&}yTsE3sRT6Fh0w&JiDi{@Y#dU3^b!`t-es-zn2>HGah)yMAwGZ?iFE
z06tg$4%ii7`RniY_iS0epFP3UEsigYhtcf~BIFd=O&Ksst5orkp21v-qu7i>sfD!!
zZf^JN5bo~oSWi1rt;B~12d`E`8(Wd-dPJ9(cb%}<=uXmgb7Dl$OV^PL;y6BMT>jiF
zolQK3>H0aceZ^Z#uAceJt9kp{T|nx~hiqSa`CgOpTwV1l4d_lwtr@lJ{<r|_DBm4?
zSWXM;I`@6f$N$!)vMhy|W=|qRf)KGT(wamFA^GR8L+q{letoh1H~Pq#Gj)=9!dt|)
zWEk*K6zO4o6&zcy)UNhK8IEEX#)vzGwhIvi>*CBqsj3$!{kgJ!(2QbaD^kNZ`@qAk
z>AF0*fL8v9jBf8Y?$I|oz3;f=-|O|O&*bkYmd`upT8J@Wz^OEfsuw!p?eq8DrO3cG
zIy2xmh-K~UjY}-|FZf0*J7#&^NP4$!&O5K;)rUXU!e!v-?RWH#9m{t0{(DgO_a1t&
z{jONnp0{+?<Pg@rj^DQ~$nP+g#oy0$&hf2%*!uU5d)tAyw>esuJN(|UDL3ZOC6=#a
zc;s==g~4NBN<LCrYNbk~0ag3xX7o}07{m6x&+hM-_j{J*!0C7-Vmv%O(zI^u_t-U&
zgtxEluUaDFqTX_KPyHX*WIXz3wddIz1M8Ap_rF{;Ft`2s`Vwuwbo-xgU&gBs87KT(
z&%p?m9>@K4JBa0RtV1;4I{U51jVUarGc{zwP1^RE^Lb@0#myKns=;wtI4<Y@r`Iew
z1lBsn>-?o${=7WB?(*%@`$;0Xn=)Mh!LCz=zs|QJ<*hUMvZs2<GxEKDW*F332qE_C
zB%VienQj;uzF7zEhDKT93Wyr_%CJ7!;0>kzL=O>nvo7g!_1gL_`BofXZP;TJ_`Qka
z59{x>J^)xB)_1mQ7wvOKtsil<S=XcTLbw(e;@nR=_S4L+AF}6KY4eN*!HT-xLbq=Y
z!Fc`fhNmOUQ(!*q`QeLaB!h5!>&%VQnf%Ky`0~vor{xW$ER=Nx^X@~lw^cwy5<w!<
zG!e(#znSm4A@%R`PG}<_8mP^ImWXssFHrGA$*dqWm$xF&<RU0tjn3jI*QLa^>yA+t
z8O#$p^Ghxt)=#r>;C_=mFdSq`DYV}4b+FI@w_rlD00mxu{f4JQ!1#z-Yj>iGkany@
zssyIPj$1N6(7>`Nr6{}kmT9;1BwFx}{$P?wr$qOIs^IL56lZ6!Aspm}n{dM_CwGiw
z?s-W!f;LaSf;fmNqMh>W6ZtVDVxCxQrPRXQ54!c-4^0#17}>={%z=;-s-B=pb}gS)
zR-#6Nd3QGk=?-ZFBr_cRtkha*wV<`Y<Ogn_lJaXWmm_un5oSg+CH4gReO&fd977W*
zRmr%ZtB<Hc=ngjfY3Bas1_^<sHIHKnwDl6lhB4*9<KrXi`H7c5{Rz998$N#V5p7vH
zy?DXTKmC-SfA$%le*KE4^O-slAH7P!#veV2vHon)i{2a~O={PPb&GDg!kYFXrxAJq
zAg*_(m!el&?JkpDqXG5qszi*|AMbCL-<L#qB%N;=E{r5kMj_tgMlS#CIygY@ELW?P
zrLwL*s^JC*hD!6js|k+}PdptG=XGUg8EujExUxP~UOc}cPQu-DXmRFoRZ<JgO*n*!
z-8_@#nJ^_x*9{~b<Ag(@#Ov5VmKoH+fJ4|b#~n}Mks7q`mpD!@qgZt}HVZ@GeRN{G
zt7l473~Ln$4x*gT3o#_Lxw^ioLJEY~9bvEIcwGq^Hhse=mnKDukr+c~OZw{w_-j|I
zu6d=&K>@jd?2eq<)*xuDZC(5KH9>yYnOv$;zoS^TwXlBl5kLR*Q-1O3r+o6&8y-$)
z=Inc<*Z|djg>s3xBOS(X9?RCXv|5088C85&EQb(?Av#;waOTlPCF)yeGW_aTemj6K
z=fOXAEIZ|nZ6q;DAWKAILdD2p--zXzr_`uPLGl*M7KApoMxn>D^jP}c#q!0)O?_XH
zOW}*Tj-dn>sg~=!Xxu<^W0s2(>E*V)FGut>c-73$<h3dp*E7>7U5m-Z_2)RT&oidR
zX)UZtXb7z;bBqvOPjt?SQVQoY{Pd?kCQUOb2v3hseEN$|_}gFplE3}rm$cBQ<Wd*)
zmJT?*BaR1XscM8A*v~UT0<9=WaAvi{e(xNJYv4@$3M~iMQyXV;HI#<hO7AiC=Oj@n
z{daz&_3|6U@feZcK&&3e<L@qI5zyxPkY!OyGg=i;*O0wFAG9<+|Ku~~AKtV3@s9~H
z($x6;^DlXPJksVJD#E!{zI^qD&%XMSSI3pdb7fK=t9)}j(&h>23~0ccsaiX*glpvm
z0nN-ZagZG~cvH7_vm-GVTeq$eI9y8+Ka<}A|HRp+J~G&P&6P2O<5Dn$s?FJMDRqF|
z0dW)8xKwLC0de;8UJl8f4LeD|^I>~?q`!-tRc(F-hiRnNZ!z1qfbDU7tz-OQ{qJji
z0I)u+?`#dRWu*f)230z<Rq6Jt5F!wbT9vXYYcbXa+8y6SYEPI16Gg4?bgp2^8kK2T
z`Qo$B*{94+NUZ0D<$NL-OffRYh_r}wkw@*cuaWd#&vMUsB87yO<|Bi_4|+2pXlH<k
zCtXc(fQ{aLF~A~7Lt8}z=4oQB&S-*OAgOfSr=il+JBS7s1_oOJDcdPJIz|2!7fu-c
zXOGXzrnyz?(0yUZHu*u@6Zbe+XT(0nr!M-KC*mA9p3jtis5gU{CPFQQ>Ye0LT}L5y
z8uME(=87d{s+bt=c<AGwIS5(IDaRe00{RFCT}0db`H;cNYF+?`NNJ5X$0MPv1Xb=N
z^d#9>O5^l&=GomX`R)!0$q!gFcPI?eMM$?!ka{;r8$|2BW7u~O-GNK#=c1($NpzEU
zF5YIYFnO&z`OPKpEvo(Zb+w)-Y45C|P{qYcRo(GZs`z~hMk$T+xv;FtQbFg?+ix!V
z8bX9%R9$&EuZSu2u;Wlx_Gxk_PxGWWiGz?eC2AM$6h=|{P2i~i>s8_6vEFq*gO7@e
zZV(egdqXjr`G-qR1h4y=I%{vVhHZ=-3DT2l+k$Jngpp8P$1&dmFvY}U;_B<JJD}qk
z@oU^%BS*dSokGCGySgDJQV^!-?n9o~DYZ2$L^MTSJsqhbPz)qx4mY=)mxinZf)5+(
za!g%pxcJp`+&TavArQq!c~cOAC=xo|e;Z2ag`;_5>Q07Nqn~|`SW4klePAcvb)M%v
z7}QcI>&m*E`$)zBm7M)*0wKi>M8OQ}qn#lLbm%Q~04v6@>6i4I-yP!YO5EJ8_So%_
z+SWJ!?R|=4_%&zpcNEL9i;(-2qA;gK%@a*mbg*X4wzKf9HNdUT%K5v-@&Kg+f?0TX
zEO#x<(2L`t$$6nV@QyS2+p+vM{L%m1&cJ{CSdLttTj>;9llCiO*+E`<6X&h>=!YH`
z5oi0|$8v~-*hjWu%YzgjU6Ot+H|B<jKd(J_e_gn~FC}p8=S^>;c`j`}&ehDv@}zcF
zUkjX<N{9>RrBSL;!T@YQqLAka0+hD!@U*g?Bkgct-tW1;xgiB%KV^1PW|uRIDb<X`
zh!eTe4}5zKMi@Qcjxl1*Xj&PKrV$@vS0na-!<60q+stocxP)hvK#D_a>N-X}hPMpi
zH;&`0_1431`9f8{uAAGV3V}!p0n}SkIVBfs_oL5$UsIa8X4bM4((Zr>T}!sI)(QwI
zCXnR7e80zW{QgaQ^YY?+a=;ty$r4#I>MdjcZn2)5THka1{fzT^ICG1*E4Z3*E(@(x
zb}}=Om?pov>$;L+qFM8)l7P3j#<r>hM>1NgXpOWQz!KZ2J|5HKEOuiL#*=(i5p6p=
z8@qKdps(ZDE_uo~x8M)!_j`Q+us*EsZ29Kv4mjhZh<A%dv8csGRxK)VZ3HQt7iB4c
zL<oUI_0C2Dl}6DsPfusorLeM+Q(|g?`@;=Ce*O_przaks9uYNmDY1{47+w3J8SWm}
z>DCtw3!l&4?{;Jhw9-HwC>gt@YaEH|Ga=3<ivdGiOtFu?`Pkk{kic#?Q;!Q)%62SA
za{$X)>j|!Hx$?F~UeL#Ih7kD}Ymw0wWsrZzcgyyAQFHB!iZfH1zlS6~zLCO241p;}
zLP*rwI+M)YQaLDZ-n?N|CHEvGP7|Rxu(URXwQ^qFF>Vf?e1<{u7dEkV?dXHMlSJm6
znL{9p5ko~n^)63;3)XP4-eAPo!Al!mGY8+!YvIe+uaL43R0(Nf$`i4<^W>Lbe2x9|
z$K2lCkOtkkCp8k?879PjExW+8whECBviUQSqbJ0=-H=AJf|(GR`dtmsY1@~9l&gC=
zZq4=Q-ji&(j-t0;e>GSIW}HSJn_?=2Bv@#imjWytPmS|xaN|X9@{E&|5*Cc5967yt
zqArD_G!8d6%odnq;vjB4odkBh%U4CvAvg;d?HeOMm!c(vuoX_hI&)913RW&jFbK_z
zB3>AX`uDmMUv<m*Xh=6c*zRx_hql&T%&tMKu|^!*_6pzn=gq>}u1NgS8^6|Fx$ur}
ztLB~F*y$ZHV;UhwrX(EZ$dtORwKi5OXm&vN%hzwHBDDPu$rJa_o+A%u(3PkGk;qz<
z9F;u5s4u-Py^L9@swcbA3-4VH<RHWjU=h8DwvOJ7;RM65pqOOF!|ye4;AuJYcsg?%
zGrN>{7QEA4mX%c^$Kw<8G!do=3l11-zkiY<(i((bv{cu;3xN&bj4={pBz2MLabvvR
zasV+{=zFlTH4Lt9AsE~E;B`^|9>p=d`%GSOSIF-omiH-h8#0))-gY^I1(tK88Aul%
z7HeELW5&HP?IYsae&bkH+?l@h5hQx!qn2;SaxiBj3v@<6?Tv?Db0+_aSoTlaThGJy
zA(ovXGve8sTcYZ0s^F#VH)9!T-ryN;<_g3Pg1ZDtU~pW1r?H$ODM!M30SCco)({Is
z$y-Hx(I&gx!s|NseO(pXaM*@}wQGY32eIm4xtTy{wA0GE9(g)d&MUO+|CE9Qhx1_v
z(KxS<4u+~xa^mLphTEKo);P?W!#r`AXLMb=8>p|Ltv9r}#xm%Rqq2>XhY+yRD5Vk1
zbF0hwchw;3_fDm=#k#v6y=g24DMX|<V#9bf9sD!~IBtKpIR4fHVpsnpSXCN<);ily
z9X5^utZM=8=N<bT`*a4-TsO5EEawhj^-;XjQaP<I2oITQ%1F#uNDkgHrfY2j$U=X<
ztpk9VbY|tm#6*=!G3_^`pULKcb{p4PE&!H09D4rR%vjDR>ca!K>A)`Txa*?r=ku8u
zBehkMcpi?(*4WgpZRnb0My-Lk3h1ui9|yc+h+X`>H<hAX?w3ToS>46s(QSlz10T5n
zk=OU@_a(kRtlw;X0I)u+?`)-*XzJr!<Iq^7@o2{WymUv4K*Ly%l^`c#c+DUD?2q}Q
zfA-J$Z~pDS=JELG?m^<hlC{9`q4M>cBggX@tdV0R*Ux#{Jy4G)*6Ko&RU>M`{oz2F
zGOr&VU2{Ss5*rdP#puxXsRULvVkva461EnsL5!IggcMyTOM)NOs{ySM#8_!mQ?$8|
zdP+0Nl+av%OU;~ewmA=ulXQ%<&^uK)xD`|XD#L<dqC_2nyVkp@Y)NSlFTX-h@@rfC
z_v%2xdXkDT<%tvm`#G?m5_8O`1&V^?<gaH6)S^f&4j`Gbw8E*alyxP=$d{)?m5Cqj
zcb(3!#2Cmak|dA_%o5$|w<np&j4DbEN(+VtkmjAB8my0D8kNGLjY(&GE@SQb-H~d>
zu~w*sppA#+k)6)m?QU34m9HP3xL;OQ(+y(U0FSH0R!wQGgB=m3)LmG*4Rs6Z)acF_
z5LKGFyNd<8e28_2LkCA)b5#0087c(6ykPq7V(%Yv#gNd!=Icf8NgEc7>AWzv=K30|
zQjQhjjPSr8{OphT<A3+x<Ny0_{))$^Cz8P=iT&<CYw+sLk<-J8x)v6c$JbBDsWRux
z?L6~%I#Fuj*|R(DC9-__8EdWE7^$-%t52$lbgc-F=qV&bh4ZnJ#K8+{>jksB_Y8nF
z$BC*7r51KJAqMMtK|}4M5{<?}Xt0kHSpo%d>OMkKpOtgok#{+f(4P<RF`L*6b%WsJ
zLcx+3fU0jN*`2y4BF>bsMyU(hy!adsR|?{m^t+iU1@=1!%%>?M1eytHp2;Mho}O6M
zAp5f&gHT0TC_EesQ35}Cx}z~OPnkxeS~o7xwh7Zl@z|_Qp*forJU&zkN&{RskJ$IS
zeod-Dw9e{?c>JyfO7(=doH-#(yFEE(LQbr$@yVxO@bcwzrro?Hr=wTv`h0<D&WQR*
zLTi;)14i@FiFVQ1xYn^7j+&u^h+{l*yA4AV>h~m$;oWEQ83&QRi&);}iMx5`;e2E%
zE4O#I+(_i<lZB<2>!}z%RxS;df-?$0Es!K|$UBmyzf&w%trQIeHI~zghC+EqET>-h
z_5q*{v_V2xppTPxkJ(_qHkLO;p))R=k@9zl<s8}X_sl79cL*E~Gt<u9J)8dav0P~s
zD^#sCb+&X)E}Gx)V*i`5Y*f@LNCv=S2XNJaqGHCx3^+RW%~)>hnI{vb-3?R9gp?^&
z`Si1|c=__#-#eBW*MD<L1z`drrk+PwtQgV3Y1NIr0vpMC>-7zA?tQ6z$C<pAwn<x@
zv|?(^=M%e@m?BukO%K|MTHw$B;xG6=ef%;1kN^BX<JIfe>{4Lry0546!qf3gJ5^|&
zTOIa$^yFix>v~3O;o0pic6=hNE63IILX&g>4MTCzaey}#2o~7MM6$$Epfxx7ID|m5
z==oH498{}RiyN@A8V(j!2i8n0mCCAxn4>o)Yej2wX5N*4W$cWu(19?ZGyZ$bcYw=*
z=HA>ikU@Rw0MTFwsVb#*AiW7eA$4Pd5Zp9E5bowZhdJ{6R=C;q7*dx6$@@JskrrHs
zEbeZY*2dxHj;4`jk(_2EW~%u)nH-~{4dKH6dLz(A4byroi6PNQ(Ket{duPl-*m77u
zlPlR0B7VPWgQruW99Nd(N(sUoD${&JkihA<kkZu0etWy1yO!$cCQJ#d4PDEHJ=&SA
zrp=q6s@)uAaIG$9a_miS|9o?oc1JF^ysuvc2jyTBsh7*;`LO<Xus#4-AJ%uaLZ_zo
zoz)o;O*xx!Dg{N!F;TVgdO5pzryBp~zxY4!2mj<x`JX@i8;*|;gg)XCLPV>voGK4b
z3y-H08OT9+rAC=gNGrZU3LqJg$jxr1<ix|%;@yi7kvKpk=?9)~p3Q`2N@=x^kO@{R
zR3pS`(`48f3HSj+YtBltMnj$LL?ER^v_#N)0kAguJsW<tu`fm^*?PYrbp3}uf@k7~
z7H@qW-67H017a6SbyEl5oIMl~^Mp4AVa|yZBXfp1C31)?8mIymeI!z&Gf`X<%sZpa
z8moHOD2VW|?8r%BKW{H81W4V{G2$Q=z*#G?7wb(N#A-u3t9Myi==x&;@2YAe*{%Z?
z<3a@*Q2OYfXyde=L5$mcph@F+K69>xV#>7jimy7xeY^wftb|s(o`K&VA6Hd0T(12X
zFB$Uh<MS5!-x+wMy(PT=)+EMtyXZYwA%s4H+O|4i#dQvzN}-w$Z&YnOE@xUbS~dQ|
zfBqlvvp@N#{0~3>UwC};+Mk0EnDWe87f#E<>9qQINh{<O32S4XGUuDU|6DWf4+rwJ
z<C8DGbY|pWy7jSAH6J%bM(<b1L_~#kRb+Ao4m89XShtB6kI^8}-&50#8M>@1Dh(#r
z#i^|k!^|$ki*`!?Q`G^hfb?gh$F6LT^ahj%>&hI`i`O3i*OM35Bn~rk@UT^^d4@Pf
zHADg-`PlQE0`nA@aw6r3wBXLFITJ0iXyx3L*s41L#nc648xQLl)Od4xA|=ILinQyD
z=oNr9T)?y!gk1=wC(TOh8YpgEuK{lu^u1LFLruUsqj4z}X+GAVrLwBAl;VJ7nkZ}I
z^}}m+hdnR5JDiwo!m@PcN6rad-6hmVfos>S;wo9DBT`pp44d)Bbuf-jzJI-5u}|LT
zIEHVW$sM3u#W23RSU&YwuBP1WXY#Zsef~LR#PZe*@;HpcX{xiQ=9mc<{!X!6)JL3x
zDa*R}xZFEpIgVIn6uvHM*d>p*{p2@|<?W-dzf&y7z+sx1b7VimZpuBDy;%JF#B!sd
z<|Fbx?(fC0c7u{{#<C4wkNNv=!5d}4z-T^7Y_s3Dcf@jC3!xcHtprJgn5d=k@bJL?
zu=k?$w~Xb%Lb=4Ub)X?T*r?t(upz?dv!o7oamkl1>@9!a?-LNcoym56Cc8%LQ(39j
z*i9KVW2uGHx$<->{9pd?5BYch{eRBC{BQq%y#D%YvLtc`=FjWO)9I0Vs%SOVvT~Sb
zcH(EAQjw6^?{=K(N>k;u7RPxyYsN+VW52fBu8CM=A`>jScC2;+bJsZ&>p&R6pH`~`
z5)z;oR0XZh3N$UKHLAJvM2MYn)f$8@Uf$1|;X-}QjHmZ5!}!c#)^ZiES8?n$L;nI*
zTCE+-O^EtLMZ~2z2tpum$P+hn;&w00Q|htr^-Y>)GLc1u3WPk7rXBPCKx=}wK#+_`
z+SqiLBdI^bVx8FuLsEe*(5}Hjp%5-7n{9iTc&!3`UvLm$o9R$vSt{$=IG2XmN(suU
zjU?_GzbmebrM(6UqwzCtcXYtmvt{dh#fVeg8tNBAl*^yS&>9n$GZB367H7&{zLu-+
zxaQEke`oTC_1mux0M>`~ovjt`3@v)uPtFFb8hKgSY2<dFC>4&!=5y{M%r`f@eEFQ4
zu;Yzp5bqkMn5b2Ge0ZXa&QAl25XA_hL|dWT2wN8>eg6D8r8PeJ;#E&{9Ej1@6CCg^
z0zyuigb?HEhnxue-NA*sf@g3chfOcUnzE{)tXOHB+L;hwzf0a31BgPHA{C{mBGETu
z8wXGC_=J%IcYI5vGZw5Tj}<S{J7BVH+KmhH)ejMCjoM&YR$>cGZAJ)0%YN8OS8Xzb
zu76T7N-Zp_@%gJ)JS`Oo#_fK`$NLv(Yt*tbk;#^zt`u#2_SFNeRDO2*+{Yx^3Z?|<
z31$bs)ad#Jp1c=V?4;r23^LxkdJ^E+-RgQ`90(G<uu!EeE4MfIe02YU$tuwr%lQ$C
zVKoqA;^z6DGzXd$)JnhCexD^8G(ja|>Y5RyHY7l<4)(eGr6+~Wv=;!jfm4VlW^%1C
z@+(_E_V%Cl{+zJiiequz91}RJj~qp9l&Tz$N;9Z7;`jwOVaLuU7c$15$>Zt7)9JVY
zqjL;IbI?^=<9I$2dU3X!CU*NBbDpWyC|6ooY_+30-CSJvW?dHMIFY8b4TQzyBgwTh
z>ogcmJP|I<M;e}%!nqo+pI$>$=KJjKJ{E|FM4Bd4eAI9Z*o@Z!-c_E4P6t9ZNak@o
z?s@QxpA9`7KgTX%JeWRpU3?U(HbfKA=AJYL@{~xDxSjV*Ntk0LB?q-?cOG4ej}t9g
z9SoF64vAfyurxz%9A{;%m0x`Q1#7$Er$4xL=9o`3ks{lE(K%yu=?~Y2mIl(Ogx(GI
zO;&MWBAQ@Us7+Csi4qCPSuM+X<@tQz#mya7PN2@>TI<3tM2te!1q%(yUd_1ZYk!#f
zs7WC9!bi938N16`U5hi0fToS57EhotjL^piFC&CrRz)tlgxAsV{>Cx9JC<8NGk&+R
zJW-2MHcay>mhGaKW*Bu@iE$!L`8SH?+8U)cp4Qc4`S=Q=G2f@}K9=ng$G=T1@6yB+
z1fepV7i%?IE&t%LJQ_*wjAd~_{INr+L_%mxqEx9gj5+KH;oY%Z_fJ?`X|{5%jXCV7
zRymg^D&-$Ame);qe~hsWrnMrT%Sj}3NPYyg7sy<740*3&`JHDno>iWz-=7czwKiTq
zoGA*c1wx*fcQc2$V=ogyjMf^DkB=PBCmv4)Ssl#NRH<UT`sxc}3P_I3^9}QyxWBz+
zzu$8_Em&*B4xWhhnCPjskFQ2jaE@aZ*QE92xEm)hnjZt67?6vPK%UPhl!m2>S);Xu
zd5Y|J*|p>_7xRyx+MTAk#HiTNwR9jj$<TzmI)BFoa7C{jXYzPGZt@dZmNTM}-1cDI
zjS+FVgK0+&i9^opQ(~Tc%(`LTpjjK~uxH-EUw`%qYg0l#Py_64Z#bVDZB<T7rJC{P
zd<MzvcF|3dXoTSVLh7U3+6e*%K3*9>qtaqym>~_QM|-niLTEG!m09)-)^syi*2>NP
zjvwBBL|e};uR*0~!D3_8Ghe^{lAHZ4x4T<(?3rDYE~P{>Ai5KKE5)14TqG9CT4=Qb
zJyx~#_Lu*k40zE?9}?aSU$voEKy;MQhgsgwIR3Eye(M8(^<jNyORtzJ<1p)uFsd3w
zwd?A*h-?#RK{&64)A@|3khY@8lN+tAJLO$au`zNG#C0~dQ4n(gYMy4K)$PC<`<-J)
z{q^;3z&4Z3MFS>;E{tlO{>p`Zt1je0m5NH`R98|A?B*_SIhZ(|J*2oLb(g(jb4<JH
z5MA|sySOZy#)mkz^48Z3R)7sR9>&*O2l^73PNlZdcM4ZYwgXLWckROLPsbyV=R&yO
zGo{4MZilr-3Bocp%A`DIMGGwJ!m`%RB<Tk#-X$7Iz~xN2(l)@hYr3I$J+Zvj&fzMx
z8HuN=5+pFkiCGfC8s&WK*GRgfWF`cttzg!$(QoPrjsujz8FOZvn~GVy{P!3&qIZ$r
zzB_C*t8p)2D}1+mGqA1gp1)rQ?YHB&+beUWWk-+g2GyVd%erztpL>@wK)=qlRu}(Y
z+XfPk;Vdsut#?pkA02irvJ@jR1WIeTZ9CuSv|X;NHN`?7`RqSaI=$b;i~YH3IHRy>
zqcwH*PcxpD1&JUvGD#qf*BQN7Yc}*6-m+2fXU{IvD%XYV_7bn+_<D_f=^8K5=4KML
zIVceOUJ;PZfFbpdAYGG0=u8iz6n6~uV%P=orx2M#WR8hlOibBF0#56h)0&1^8HReN
zSBA>l=&8%qTXBsF>+;}(VEWcI;q5rSyg?rm3Pg5krkal**<gN+xbN&nHNCo)S0e*g
zd&<~zF)l3G*0vF9?JSrP$=|*;mu~QOuf{KW&*K=rFR>gpXUL-8h~-h4{5@j1YNOW1
zQd~pv63gM8v8>;tSdMaGPy3G$XPbOSu{@l+z8TBgGjh2hS0P|mU=Z6Yy^iH#1r63?
z*}ua_6MF{r4;;%dax2>sqH9&9asRd$zPd;6V=TWjj%CELGih8q&Zxzk3mO7Xrz3CP
zyt(K{(J)oYx>D9gDMsq7BELRHt1Cv_w8%t+InN|BAN8`fvDcgxbj4_*$6CLq4m!xy
zb4lNS`o6R6%Iey525T*}KK`oKsAU0}Jjd>_JND?|r!vM$N8bGH9PQofn|FVF?Ql63
z@J6Q^NyqqG*tG6a2;`9ZxV-~SBHpOhX5MhmJ8u}Qg{3v7Fk?W>87YQ|GoG3mr*&aU
zV7u6_jO$tr_V)nv--<+WCv6pF2%&o&;(y;_{|eL<8Q05Db%*c$yk|9~$#QiTdw##H
zXLi%B$7#<C#@J{VXjd2R>b06~H5TTesmJlC8AhIc`E$s8SU(>wO^$vq<Z930{fy%e
z>+iQd09YT^cek`6201~9fs`Wa+Gtgg5FHqD)-6dwAIDSvU;m$f!GHc={ufXnhe&H|
z$?|$$S<ef#bA9KSnUnYkO^nPbVIgB76L43$-8>=H3*crtnnN1}7d$Br7%2)-4OqQ%
z6op!qhH=uf_avKJs;{fFgG?JstTa(7$}UH4_Ud!sv9f4op-|1mbG6j&0O(9H!Dsyu
zcTgG~6QGYZ_<x7)TGZsiiXR<E=?9FOxH}qtXfEf93e3qi9jGC+6#|6rM&$3Oa9S#l
zr^4$uk321vAN=4sH~T$zArWGR-JT>fyJ^o_U-5K4V^%=L?bR`w7KGrAMlpJ4tgSv=
z)W?vL2xgVi8m4Cmr6&SH0#t>jff&1(I!f$g8e=@GwMIEFEO=rgL?(&M1T1Fm_6JNG
zZ(e`RUOEsF4$vK`OvIUY!4ncS2dr|=#1L4PLal9c=rU`lC@My<2`myU_T$q5(qOXP
zm@Qe7>yqD>jggc0VO_;>Op%-u>r!ad=?Ow0Bq6GhLv#>S8~@LL@h|w7|J8rN`c%jg
zSj!o$fuK;=h8CEnNQ!|u&&)Y6Cn4v=Zl2wyJw$?mG~=+}b%&fsMqxFQ7}tIC@uMh$
z1gEjL<xK9+p%=NvO4-JCQ)_53j?0nra^hHhbmq<aM9dM}XKu2vOr7(4++o-4+a^tu
z#<uNDW7}$MH*Rd(Zfx7e#F{j=nu%>2XP)<Y&vnipFh6~-+4s8kUhA`__}e*(H??h>
zX$vG$gkbgJagDezjn`v{q!&Mpo5YzU*U0B%*EwnSx9;<J8c*OxEP$#VKc0!c<OnV0
z!0rpcfN^PLExq542R$r1-QTllO?|mG+#Z%j-)trSK+hqTDtox!z*(bCM)6WMHw3a9
zA9kqy*d+YcUnu4{nRO?n71y-uMme~Zhk@HjhW3^dp6<53v~7m-*P>Z@625d#%m&^q
ztesblmKHh(3?`$oBkDFIcH51=y0X7oMihZf$H8N7qtYr5na_kRofFZMOkIt-kgu%?
zvKff6_y1k!aOs5!x+G2Mxu~zQ4K-v_r7y0ks%*e|BLq2$CXTTpm_VJZ<}7Yxrjb3L
zYc8&{k?nlt98Q9In7Su9Bj!s{LugDEq8Mt|n1drWhyS+}*YT)#(iOR9fDM8!in5lC
z&(V2lTm5(we2#m>{<!*ppBdk(GuI-rsK9VRb*MxUId&jemJjfml#&@?0}M%FZP|N8
z6^-297${p05dTUA;*gX>F}}5z#NkMZT049y<d8wg;fw@1l_NHxH2tZtCMBFwoLATS
zb;nzSh`v+jmZ}1W9W$p6E+b_Z3_5&vSwe(cpoh~At!x|MSq<nt;d1?1dXPDxnTlx{
zuyUVyzn_+7UoA%LHV;Fq5In6XY+YC*L*9T_rLCn?SRIu~JHc97{ImL66H7HF*YAm2
zwtXw?L9j&Igce?&@Yo4GY8ij%O^?U;kjKSlXy^vKY}C6q@OyZt&ku{LNt<sLv58Gx
zPixY=iSd;B+pc3Ybbt##FzQ#)0r2f@?CWV1fCnIqFOR9_;)Tt+*zuDGsxrNyrK2xR
zO+yaNvSY9wF<X?0*!caWACm>~LW{sp)35gKbMJFqcVn{)n#n~kf^(O(U0S^hx}%)H
zobBk7b3s!7ne!as*)OQNCKag&U(LZyJ-s717TrI67{NGtS-{2p#)UU~Jn3s1g-+Sv
z9*NMJ1S4TfRpR35?MyR73zbMLE5@e%ZwZn4eFiGlkzy|X78cyg(}50e(KBxZW3NrO
zK}q@b^Srw)(l`xwt=%(QgfEXnBG20$2QhyII5##<xa@~rAUw$_#ZVlm)-e7DPPfiZ
zj~U?Iv=kfJ-0h&UI!3M*@pg2@MDwQ_=mbg72(L<llrzjfv(xnr;an%sVKnu#*EzTc
z3A-HEnwF0s?i@DUF}L&_kNcDKMS)4`1A7xH$CgIB_ir1%meuB|6CFQw$yx8C^B)}|
zCCkb2yX^dL*<-$Mf_}M+E(Bo;=$yb$WugOOgN%8Fm|`(EPhxY)T4o|BRi0C|`Ysz1
zVcyOxpkWqgObTc##t~1~eTi6|1`eJuK%0a(2B)SJ=^sZt899E3ZhPAdb=4#BbiW$5
zr3H)TA5Em=YuldKU;gNw$XMSXxKs4^wQylbJUDfB2uPD|#O_4N-|msNIb+lca2or5
zI{VR<alP0Gq~%WUFw^tvmc8rWW#Knz{pZxk%VL=3VRI_GcXYE<)oe2=XnoY%;KCSl
z_AqiU@(naRK_+qFG<YBtZEEr(l!d1ZX4a%p<6M@plRLZ2w66)p);65fXrWdJWy5ai
zCfy^Ir6woftpB#s6;q?|=Hl;7F}a_SQujpfv7)NvR?Ui853vIRvGLOj(b&!}T*V_n
zeFkg4X2P^XJdO?kt*;M5p?RR%`E@Lb1HGeYxP|qGG{egTkl@&rQDmiOS0g}UqfKq&
z<?dvsjrDuIKtR0ZLNS3tDS9!}8gUMgNZ))-(}^nO<M&i~n58&K%=4WAT0(9tCHJLc
zEr!G-gOp`R?EE}|6sAYgoxn?kzfF7<H~sy#g2!TZ9@ncHKCTkC*oBeX1u=o7+x$0B
zy5~Za{5S*a_rI{;+~9Ee-7}D($$PXN#?-eU{Q~~qF)=4=Kt*%}j=XfJkm~xZDF&EL
z@(7jaqO>&QeGFFan)9{lQA;gu#_|g&@|)XzCs2Otflke~{jZ9gE4QDG1~~XtWz&r(
zS5?t-3c<5<aeL7zUlwsD_WHu3XQ**ib-u+DbT-wu>T)!ld9bOqin5!-X6WB1l~m{Q
z`h1`QpWnxtdw7^|V&|3~_N(zBzM1v^yRRT9J-cXDgBa+H=OY_40^2I}(U5@r>R1r&
zu<j+_*HRkh?d-Oj2IFetcz1J-QnA=>r}7Rd?;;y;U{Z9LMlILUBCaNiYQ!vK@Rbj<
z<hPtE?lKV$Z7V4YhL)cdj&#%rT&*|v-3kZjCmnf8x@l}vIKv)#>6Gz}#?5&PHeuCA
zBYPd{qKbTC7O1*%3(pqC)5eEk8W*{`0U~|<61gckVo~m7cqYU7SRa%nkzz|4kYu-1
zA8?J-bGxFNdhE-ME!9Ln@%y2q_v^LE`)jG+F;ei!==&UVTK8+lls^B@<H(kI;9bmY
zr!@mm>YPa$M$dgIDvjWWN|wdR_KwfCUR||zuVh{e>mr#|Vei$0uG-V_6{5isr-dGB
zbkA#{pizda#bt09=B$32@%pEq^+S5u5?{KT9S?2%uC8FtwwtJOUU5;fjMU}JWm<m_
z5ei%w&-C?8aDC-Kx#Dqqck#G=u+an6-y@ZAcoG2HO4HFqHd!t#zJSz)=vR~2reOO$
zzS@i`PtNn?{atTXH7A~LMg^q(r*%0pD>jZGPEn4#kmD)QmdQIS&@CT?12)P@SKhAM
zAXU$=VWCZ-s|e&USqM=_Y#be~>n(p#i)v(3DS+e4Q8<PE#R(kHuGX}McFMGni}#jx
zWDG3%#B?!CmnXdJLmn&tg$^O{hvBI>e};JYHUNpqlU}EOMUW>G0K@zm@*S(Z82*JU
zYcXU_`|%3Q=;r1I+2NqRfOCNVye8$jmxfb6F8uyxF4>Lr5l*cW#7UL=qXj0A>rX?U
zkhH}~?{4xCOp>_x6959)LuC^<up=gSeB8QXF&ZyS0hf1AgXX221mnxno5;R(=A80b
zsiijaru;SYj%L0Ie_w((Lnim>Mp1%>2}Os(|E?rhx85%5!ydWE7DqLY7_LCeL5UIL
zB9rxK{C7?ejQQp?#pZVS%Gq@6ed)Q;gb^Xmv<qJ6`^}_n1D+U9BGBV|<4G~sOJ_Mh
ze@`PYzis*yfP%KC>#kz(HUM(`xwI9LT?qKD)qEDdA=a_c+WR*AeCJR_NDW(*<$iw(
z<K?}5bVqJ<ZrjjzgR$gjW8v~mlsG6#4v@exKtha|ZS4H*#89&mrv^Sb@HYuSSK?^j
zlB{?T+(|w(XgdlklV>zD^aRHFs~h*14&({*MH#&upUUP!#fPuqi>ymx4!7BU+6Za9
z$!0;LGdmBS$P(Y3`M7B#lYu2dj{_&nvn#<vE)AHX0C%youdBZ1c`5eFqJw<*Q-*4C
zGgfTz<=s_)lXWo0iekjUi;l9CoMoU|KFtB#B!<2qU&4^4tV^qjl5rqI|K>aI&WPUX
z0E^GD1FQ33Alv+_TnRg|IZ`)vc)PkiQUVScUdR$8I=z$b!?I8FN@J!Rp0tO^t*~}_
z0ZYbkl0bVV{Rg+R1|^V{3@+2q%UWO$Ain~}(7Ef9UkG^W&_}2l7RR3IdjsirF#Mxw
z*9*B)<Lf!de=~X`T>NhqAVudF5SSUxvL{9xvwFwc#OrC8{~#1CXWl7Qct<EBB9t+`
zx;)6Fr|g+(lv&P6u#yVXCbfHY6w%ygn?uT$(h{%yYceA9Er6DiDq=USrTwmLBj~#t
z{Q?#-tG&#pA8dcd!vW7+gvp`A%9J$4V*7jvqPu3T<njrSfxy!ij2cK?Wmdyq;KPay
zAABBNe9!SB=~)15zc16}K6u20+YXC&p-KvmivE_$3O>ADCpq@{&6sA!;S!CK?ust8
zy>er|w!Wb(P9x7<6QZKUT?_y67Q2d9+v6#ei7E;m<WgQ#6kL`GmF9{)G&X@Q;`m8E
z3oF!Z#t8o!!`|QGgt7fz)ViwpvH4H$+g2~B={Yj5tBSE@Ooc@B`scN^EjgtJA7OS|
z!j=5mATf&Qrruf~6&Jz?fl-#4w?mwcl(5^SiF@s%sr`T1`S0(N8khW*=v|l~;SpNY
zK(|A^rXD33Qq47<<HQYyZ&_Ygtw;E}ahm!rQLU3@TLy)Q^Wa7dQi!WGCbo<E2e-ui
z&EVS^qe_l$k7#gGB6WyS>rfyliIpwmTPi42%=Jc0A%at}wtRudf2#mQW=lw1M14j3
zN10+?23F)de@Vv1Z@56Dh_v;Y>aKqlCW`o<D$-48T#bx>D=K{}6DsRE{!>}-aGmwM
z3~2x*CH$l=g0Kr7B!0gwUEA1bRQ%@X=;^DeVKITr9I!6`paka-^@F*|gt_Xh0wmCp
z`-QYa^fy9K$Ps{@XNc#%9me?lezD2*E7CwlYJM0v0B8xSxyUT-n?GA3L(Bs|!z|KM
z4eZSA0D;BVuFK`^ZFkyHjS>?t=RxUY3V=hx6W@y#mHh|u*ey{8zg&}*D5cyPAdAi~
zUO~UqdIDqO34FQ4X=K?WbLSHuVsv#E|5Kh6x4HF@sskt$xh&_9oQU8Hk`&Oy)xXV<
zm!A^^83sJYPsiqe?~A$(LN0*YTre)FBSfjq>cj~xBE8Gz!h<X}s@48KtwsM?I5$M3
z==kMKgz6`|>?flFMjL)6*vz-qV4^OStxcg@1q@8Yi~*c4S)fx<(@WExxOi6tkkorF
z5O*MLo--~Nt~a}<6SHN<h~j{g{!Jt@Fqj~oz{ACR$LDWMHhTQ#rxG<VUK9DRmw(;<
zTG^Hsgnz&5vW~C&ib%|de*Vdd_qT=JY8ch{?aA6ID<#Cq@>I}sGRYjt!%;c`r5#M{
z->Xdi)LF+sU(&Cw!&!CNML*PbCK+Tz^-um5SNcc=<sBi~T(=b<dSlaQPz)&E+$m~q
zpS-l~Ik4qqJ!ogGCaVz9`*+`sjvMp2zHWienQE0pI2K)*huy_Gg*8j7Hv2pCKUzWo
zY6P>3@EyCX_KT}@h!na1-@PM!sFX^vI^{ucXC^nZxscZZh-PRmqyEl3z*xo_Nj~DS
zZ;8VF2M>}cK7!^HQ+1=W_MC5-yBnB=h&WdDi`$f-gveeSM@iDk-YBwu_pVtFH68va
zu_)ESpre3F3TtJyX0>utkkm|K*&?13UwkPpy<%QS15-F#3cG#dKj8GSuOd8eiDC~~
zmc|k#?7Fbhp@~~}#1yn5-sSG7`1Z>dIn=G=1l3H`e7l<`6rgP<ssdgm3<Wj?MVfH@
z&zVoG5tj{oMWV_%C5iE=tJgN$)eMNay}j#TE!wFZqJrJIvfX)dxz>onxo?ge4GfFh
zZDaZxFn(7QW%0t!^Ol$}G{PeU?=lIaWvLU@Ff=1_5PnOP&Da<RL1L7o!(x-@%ojOV
zR}+lP=SYJj-adQ<6}wp-FY>2hPQ}unBXk0+K8LdS9dy~jMNM4*>m16Hi)(D*7`j;3
zhDwHP3_7y8AjPnz;z8}7yVCHRd)3^oUOH=b+%r0b__FEXtKF7@U9`YV*<vUctJM`2
z3rb3NAX`4KcgstLQJx8KAGU5)>(?sGi%zm}Ks<Xl<vt(fw{XAPE9AD73oW8J0d<nP
zOte;D`pz6w=}zGXor;W(<w7155C9ADgq+{^&K=EwmM;O3SzTs!+R{c;Zkt}^37UKn
zvS_?QJ0go{Q1kkmZa>Yd5mABu&1B*17db8Kv_$0M8lHH7b`zCEeVj^GDia*?&(qqZ
z2wMr0?rV|j1S=B}kn8nFu=!L@W7dnl3T9=pnZSB#Aa+gRA!4;KtB4~Z&(ba~uLN;B
zN+*2w7XW4S;zLy^oIdn7VFvf54jy>Qv^us{vZe3kaOeRqnY;oWT?Z@|Yi0QwSXKhv
zKaeHN-Jo5(n!B;Bi?vKntvWvcoYXHjtE#q#aA3w}qt<(;!LUj#BI}`gt+&B1|8c`W
z#!8RL3QKjG3yLL1`aLjiZ&7P}^;z@h(&V2s0-^b_5ha@igB>-74FX$_?w&m}H%W}G
znH-_)cFYLK<>a^Mfs9u@(;G#L|Lx33_29)OIAqO;Ft5ESr~BeW&%s+jO<mhf%~eAW
zQ(PXF&33<+OlD!Uf?m^5k3NmtUiRB<2io%|B2_{Zbn6;%y5S@5ZQbU^>R=~YDK^LP
zwMpK6TKkVz?ztpxexf%y7IMY(%$Kep92&B7to5qdDt&gNLn<B?`G_nuH9Fb5I7FW#
z`9B9!w&|ne8{YRqji2$pMb`~gXrIK$d=Rca*nRluVHjXmaZkW=G=}H<xm)547<wa>
z*dnB2+wC7@HvaKJ|9LahzGJzi`ki<}aI5a06aM~tG|cXO%G-OWGPw%hwW-IaAQzjY
z?dX{_G&FLwqQ)T@&4cDZ?I}}}H5lh=?nGsdv!uh0`a2o?)(ci^JwUbL=-F0Es@ucq
zHBzN*U=g0?)K`G{+#bu8KyJW9Li53_GfJhPuCIEeo<u-j?v5+K_UIUj5CQ+Ppr^Gy
zRsWyO@75iD{{IidV(Vf9vIrRHJa}r<(e{CI6)S79s*)xDY808k&FP51hjsDFSdC&o
z1JcGT7K%mjq{d8AMo<whaMM%$8ZlY$ov^OF*cfePg0~`PW^s9Myb&mMY!Y-Z7q~dw
z!7zxC_E0~ygb%5^-S#nD7j&xoy0pNPJGEO!ov@|0*Mv7!@Z(^l@gEeB)};E$g#~QZ
z{dVcVkyHVsgYn)4xY|=zMA@Ai4_D>y6B>(kz=ye|)FXBOE)D+3jG1e3qhe}&e5*@Y
zr$(+DV^wj5S<FHQz~43ftiICtc=juzdp(;naL~reQII6c$>GXY<OBLi$3`?Wt(_vB
zy7FH0k^V*2khm&X7;;Bl+(+nSrB6<#qZdC}d=eh~`rfs%glj#p@|D7)G?ZfP)QzvF
zeP<8MICbw;lGCYW-K0L#!cyKivuK*X(0{&}<Z_vSJ}0GC+$6(<h?-x8WV~?v<sBkC
z)0&?S4OPq|?)dVc>>Z$11~aU)-Xe=j>|z3aFP78@E=mcR&o9c<{L!0|7o!_Ie=BgK
zHT)6XE$_2BXZwrubm@_1qt#fWL9UvFMy3?{Qs_pY;`rRrvRa+G03vZTHdoMC($-WM
zK02BS32shM*<`!JPza{|=6W+|;QeaeWX8vrrKv!#Xg3*55sbk_a374JNr{qZ4@4&T
z!>g?@CzeG;EFYT?|HnE3`UobYo|8Dl%$k`ddjSv<tW~@qfryB>K5=h$Xs<F$F9H4A
zf&NzEMr!mZdRVp;5Ze0&W_<DK&u}|*LOMnEk|VRrS-0257I+%c$eN-D?WC99M9e*n
z4_v|h4OlDCDH8OO@Zh_?bw7tI{cZ9|8)(>|JM10FzNr4Z+*?df5D2Q1A-S6>Mv3_B
z7^$kFt%3w3WWGTp27@nu_mE2yTi0)W@A%OSO_!woiP$ON6P{;&p#T%dFs4-=mbp&L
ze&L=qeN{!(_lYl5`EFx-{lh5uyoz>zULp2k@9@0~7Y1H0@}iNOgQ?{Ug3Iq{z6bSu
zYXLzq2_k-7A{5fcz2anFr7ptvh}Q)VeBBI@0-N^*dsx$Tc8CR19N@kr<^vY9r=dAt
z>y?TOz<(Uf$AvQ{7%zM@AuqZ*^l7k#c%^cb3AIY(;IG}o-&jY{1PiW~XofyX<62T`
z@bc*Lz``xVk@Ny}?pzH8Ym-qpXrefH0Hx~-|Cz3Z?3T0lFFmpY5P7xw;_3*uz31pI
z&dT2SnXh7O1ol*}15fr-X0G41ZKm~|rYwXK$2mSXczw5h)sxxAOL2Zdy>EFvTwS(=
z%B{4pCN=2EN_M+^d^l&c+HT<~w?40GZW>57(^pH=o{P8|KqpAAZ7vlXdNmyynEz41
zhF8w|EE=7Ol@|O7b^lFlZQ}Kc_$?avPv^Y9hBXZxRD9V~<$aYbYZCm<my;Tz9oe`S
zij=oiE8i)sqw_E4LJeLVUiil>{+?8-bRVL6gE+Al#u5rJCv{Dp73jSAG=b>$l0NPC
zZttOrDPxHf0`|VtE^J=+P9*}ln3O*^#Q+B?I-sSL_SWcRRZ`5#xRYEbo(DSn_~l>;
zSc-ylHv}>_#{LtiuKsVJ`dl49y1W;Il`{+q2x2`bHY~RLwE#O4z7p`=cyk_aHo9U#
zID<o%p;M`~J_ulSLu7m$&9roQFK-tPXJw7@h1+~qTQ~-T{3Q(xC{kOzSrIPZN#_z*
zrE+wxcscVm<CI_PdBlNBP^X6+H+hJpf1oqm*!u=)lZ%^hire1vmZUvD2G1Q&NMH{h
zb*&>^JxE*zQNSo#Ib9|UbDNf)LWZ{P3Qwr!@?nbxWKWW<00x&eYh;)7*<aVf%c+nR
z2SXgRf5hlQnz4M}#|$nno%eg5D*eVR^|+H+t*e_B9|v;PZQZ<-D@oS%WNF*^J^Zq*
zn2_*{H>^}GXhK8~ld+z=5YH_r`dofQ`)iS)0A>GJ7g^Z*D;Gp13J4OktxTOBM2s98
z$Lo>Ie)aW|Y`{>3nsa6JmYb<+?BHpnS9PdnxB?8!5LT?xtm;w7dp`VZi95R`4sMUa
z5dCoa6uz^5MMODf2f8w9H5X*Avd+(nY&`8CmCuqSkW5+73R#zT+g$JrY7N_$SaOpn
z$-}FG7Qk$r{25+TUsCP)c^SU6N<>{W6T9Q&Nv=BCiMEKs0`7D_9h{V28sRbO{klV3
zV6{7CaC7;bu-EjbAxh@rf!B?sGA1qf4I>>mrZTC2Sc1cG*2aRS`SDSG!)fu%?MZNP
zEmmDaSg@}r5s@7caH1YPLj$Dqd_g`#RQEXH?qB|MjS~Fk3rc$9iF#}sW>YK^BZ^Z@
zBYev2dDDUL67_pLwmY-57L(H)<9JDnN8b1S8I#qM(+FZ`0Weiau`g+EtyY|FjyDUq
zH29gqe4#-DK-=w40MEwH?E6_BzHNJEoK6r{9WWB9Pv1J!>d@c4_<K0p$J2m&AYhSq
z*cZP#ZWsZ}VnIf!*x84LypH?Fovx;sgE$7y+h%p^oiFHrrjQQ<u(KC(Wf?Wh7H6b=
zMKWnWIqNN^ovRJoEbND~s|&e2*X)LT!|VqNf@nPGhl&Z8eTc!>bvrj`W;=#5F`v%0
z{VSo7IsCCe?e62c7upzP{(!@G8u9U)tOn;@VxJqktcte$lY%0hw~IW2;*pfgo%eH{
zVNyP{PYS)7tCE_35F))tN09*HOX0NMp0t*}3TrAzTG{Z&LTQuzPt2Rw7TIfQG{cpT
zKl|}Cs+L8ky!cN6%ihR@5r|KImoo=lW`tlrV!&52)^aWmzpy-<)85;0@dU~)`gz4E
zjj%TojmCzh@pUC?5m$aTbAUfKB**Q++!EoFf$_5F>O7_Vvwz10J^0tCL6Cb60h-|B
zBY6&(e}_t8FkT99v+r)qlOsiYiz;nU%x!2pp*vNHukdIc!GhEDWkUC{TT9gQ#DX{V
z>ub4#3BpHwP8W}foIfq~d4KExU;P+v%EC&s>&k<UO8jwEbNz=K%e-7Gml+G=g0St0
zzXfYCTXC7v0^z(~nDTS@4olOaNr@|C+6DHSmMvkPx53md@=2>c@uw=}6QE8Dnm=uz
zg(*LgwD9tUS&I6!+5AG<{>Vg=WWZ=((xxZ=gyoYFf|$XX^r~eEIWAiSO&{)$><XO>
z6>X=S`1sf9$b>Ak6J@QjQBR%#ytvw&wP*EYAf#b(zgLnu_v&?pQ1E3Yj_cGFa)K12
zO?J_vEv}nNA2?63Ty6AgIdB%Zy1Yx>rSX2<UQ7&!vF266!uCa;NuH+^Q=WW7!BO=T
zZKOW%du|lmI^Pc^_IVAQ5_k*ld3)NpZez$gi3}AmEX&7+@#P=RM_J(Iry|__1B=$`
z`CGTcN&?B=%gfMjnH2EHk)mkCsHZeL<m1`k1Ji(C;5uq&rnv(}-4es{@MMacPySV_
z+levO*GDwEsYI2&Yo|c&v)=@krnRY7+(->cy<0_f)U*ysSzN9m%1|e;WHsbTPgAeJ
ze%V}qoj5I#G>fy2E+dar=-vT)9@%+@v>VA}T%3hYf_>-8{l3OvEkUG6-G~w1n%hwq
z9yv;x!H86ptLKhyw$?H)j_<B#AjRDK#Cjvs1ruap#4SKtF-~h@bO+ta?X*dC(RDHr
z$>t}~>|$St9;zLblqJo*8&Z*OO)fpS))4C;uGiFhy?;Bhiud9{RwTE6#$Bz*P>R0i
z48y8`Y>thDJbbnlEBHP@e02q4c7NK=i96T-BeK2mde^h33;)3BW$X6M_c4grvGHGH
zCUsqILrQ<o^U?<^+2{=rp9+=g>cFWw#3O?%?W%*Q70KufNpwh^ZZyvK7V6yA_GvEn
zH5$xTpK0#<^q2WDv**2)kHJ=b#`*47O;IfL^5RyLy?6!m1=RzEJ4I-<TiAk-$o~0x
zC<eyuDOsn7$ZSqA?Z;og>)h*S_g%lYp#z3u2Yo_{vXaxAaSU|kQhsj!gOxKvPu*v!
zFkY{g#j;{F6Ny0=e}%KbqgSaetr^11N1HWOd*)Svc5DQIy;0>Yg`eP8{WYqRvb5?l
zHdWoJFxzQi;yvr;I08yyjV7MRN}-_+Odu4w^_5bsHod8y)>87EvIif2v*|e1{G@O9
zZQ~$3q=X8+aif+i6uJJ@C(cwTuny=zRj`20P1W$qWa1G{ZFj>pjT;e|5Y!glq&wMe
zH+&7LY-B=N!;mb>4>+g6{^&Eo%>?N1y1UFEt3YcQwXr?Mw_XlBI6?B<K`~r4lfITH
z#`|U+`3?mZ4jN`%pDu?#k!QK2XB{5zRMsqxl4#p9%}Wr|27`SSA;3C?02ay$+EP;^
zZe~LX-dbq5HA18{FyUh#XzKB?5S!1>VM^c}7P{N_4X^Lf?fq0yYiZPjc5MPZEUSIx
zj3r1Tjx*_4t`qn#0{fwHj5F=0j<{;NF_{?Y0m1@^67;xj@OqScb9#>J_d1Xp??gti
z)i@=vcOdw|o$L38i3}<g7%lrb_qFbL6^O=su)>aoxU<HBH@Vs@fW(<m{fV2>6-emy
zRXUyc_0Z(p_FR(~TiZ$F40O{!a3z*S9`?dtnVsIS^u41Bk|qk}llzG8KBy3ZEJQ<^
zv!~02<!ELD6prL*{J?--j4#q_EGAq@-@KUx<pQu(jK5>9;uabzlh9C|6B2b;S7{fv
zY^R}#sI-`M-E-L<{~>7Z5Y|vRksBS=j%kL`h8M8P`3+UVWUuz1s@l`t6KirO5Ox=K
z=ce<&^NxcD*63u_aK9fMPF&)i${Z$^>@TwvW6yN6_}MLgR49@#@vwIyB$oZB`!^JV
z@y|+2J@K}()*OSw@Wn~*Lrm~ED%)lmZ9Tk~r|+>ZK;H8UTaT+(+%S)+j$S-?T8Tvz
zp=7YFaSru)o9U{&rdGJ|_>I^ZT!yMA)3W7YKI>q=X;Pgvtdp15*u;A;vcbE*d28!y
zUz`9{S$&SeNad2ffTQiPp)$3`Q**aB!j106J#(jFp_^B2_g7TqNifzD&{<d)gROKB
zLXJR=i_5Zd$&Nsw!BN}P#2!!bzWGC|m<lENd`CPqxj=Z0WN%^vFCDWSMHnnEYsj0$
zv&S95MbB`L3U8z5MDmyxbt~HGW>;%3Z#doKBD*T_F!N75`?Fa#?R3(juA>Sj_K0nW
zGQR8?n|Va)!P&DFlFWk+faU$rr!z+vpTFFfc##uAaA-}t5}W*>;u6hA^CEbl%Xd3`
z?RqH{@lS)UoHT6HoUv<|)JZs)V)V%>VucdZ9hnaW78!g1kdpQxt7$^WbL{Tx2fBIj
z(iiX&Mo4%{G{~!%Y)ts{XIT+lMP@fkOV(5d&c3>nK?JRga8nA9kmxEY$MbKl#V-mK
z@{_)s7xUNMLgI#vGp@DMJT(R6xaAWK9BKtwBlQ0Dn!Cq=7HhzED3OlNNSv*OUyb9W
zn5D(lRTm`EA!kwT{XT^4{j?vT*zGH>RLoJh`D`o8cxt>apn$XZHK6#S{+&l6Y}I7K
zKuM)I<?Ail!KT;UUV5AJu!_C;EXlBOfMHGkHhIYQq=g3lio5>dX*DT{%_^wm%E@0p
z=}0Oqb9PYpgygMV$<Nj5#<q3^la0OO{Ot$w&#mgiSD$c`N0hM(VTPc1Wvc1A{i-3q
zXiZl{<`qrk9q*}`ua~7+35LNCq+SYX(RydqMP74^^Db}1o@|Ok_6=9*HQjKn-86fo
zsRrlSOZPe|h-14Abb#X%MQV(Nn^F-g&y$Be2x8#({^c|gB)ewi$I)s4ZnfwG7fTMf
zN8|C#l?1j0O!{&rD(}C)L(YEuQ|Go`3>&Ot2r!^IRDa+<sqz>Pn>EODy^EAy&F%_}
z(9DwOI5;Iqb{%$@syDfmuz_vBr93+zB$W{GyazDnc5Q-~{l?i$uLU}YAX}j$=lxCw
zwravnVA1@)m;LYI-OmpviT&dsN03y_;N<Zsc?IC{jRe{}wFW*;F>b!1O5EdZKY>$G
zQoS0yH$8Qo#mWxd6chUXHJAWDVuqSXdqcF<VYR;Y1+SJI`C)k1Lh~?G#^{soUsm(_
zMRtR*5)D6|XQz^Vxxdr3gDp#Gwq7xf0-&@#!y~%pgLQ!)heA)B1FFMA2Ra|Oi`aTl
zOPz}U@%QWe_g)TW)bT;b&Q108)TPbHf~jI!&At|kWTiD@_$4D(q(}0tPMyyPDWEfo
zuP3|1rSa<0xu7b9ThbT-!n~#i$Q#)R-ErS#QO|EKq#7up-oNnI=EU;x^T!WzUJE`4
z5npXT!N0%tygrOLkxlk@WR<yG1y|59=L$YR$tHt9lHIS@vLY~1^i%rlg;a7B(;^Dy
zn(B+qkccJ@x(Xi;I<0VIM7*06rqt*eY5j)z`@qM^M@(Ff&-gh+Zkd6I-)Cgsw-w)U
z4?cDxAVTMj^Yvfu2jT~~=ZV;zkBHgpHf`vF^`k=Q_Zvp{tf#OS8W;@w!UbY8Rh~rQ
z0WT+Iz9C7YR>pK-$v_pL?MRInNlRpe5Wj6HKK3=5BRsIqTWHhjMROc*5y2KfD>PqQ
z`AkeSnVY9q2}I>OxzYTfEQ&6Q;CkU2ktFt$i%h6B=#o&Ro;})idSPQ2w~(6O7s6Cl
z3MvSd`lY~HNh_bUs_X4ojaWU6HDD!bE$Uf6-7)Q4{U7_+So&Q1rMvR{57JYBj}YbU
zeaGjJYHULL%{um)QBDH@iBerbX3gM(uO8@1+haatwz*e#UO)UtV!s+eSTF|cqBq-o
zCloDeKfwt15BUhtlB5hH6CNYjwH;3Fs6UTsl7NC`igk1p+raOyt>3`=e$FD^n4>FV
z7uyxK-;+5Xu;&0I_=F0;Nf3$muZi*0hV$6YeLXd2u-;l&ies?eYpc@#ZAlqAn6z1!
ztE6I__E&nj_`Ib!Ir9y1F#YHVPBhJxmkWEb<n>C?cTkd9|B~A8cB%~wcBF|Im<S+#
zn<Kv7iSzBxeLJ0cV*Ysfi0pYEGPoIIj*L?o`y|rUVQ(mANrUBmJ8F5SYr=$p3rj8}
zwd<kyXWzjJ6=1}vs0r(RHI(Z8ypL#*4qur#AOU7>sk7LLdr0rhgat?ECk3ifme@w4
zc0xP*EDkDT$L@*Ip}-gIGAwZ)RQt=&CMcWvEd!>|na|G&NFJY^Qe3XAvk+wI0dl|K
zbV%`S>I8xb&{cA@s;8^a<{ui0G2No+YMM=|$&N!VwDE<Ll&6v!>S|=FCFYc#xyO24
zIEQHhPi!!5IIXU|F#gBQcTm5^R`6@qnIudiTxQz;@PaF)5qlNL_rzK&A(y~5v9^A`
zaz>@UzV=z->DVdd?p2*3lLDdL(e8Xll{>NY2;%Jf%G0Q~YzEivkAxP;N(a`Dg~vD}
z5ru8A6z{=4SsHfWoQXdr9<&YLSv6JC8&I2HzAzu;k;?d@g2Zrut}Xj5i$8asis_K}
zOYoZHx;Kuz&d!wsX|2elqyh-tRIknJJrF`%dcWRyVuc7<ci)FBb@j<o>KyQzQ>xv{
znJW>wS(bu%_v<zykoNLapgxY}+me)Z#p_|Ac@o6>ILh{|K`fUIR)P<%9?t3ZK`QuH
zVDY%Y`%n*v`1KJv?ZoK^de}`)X463!v*7mAE0pi;4Dsx)U~yVOoJ{`g^ONbL^EXnH
zA6ZU)Viv-1eOel10-+j5r}XR<U-vyEp}BIe)Vq3?T(lnJKKA@1)OHdTd-9SbXh#RG
z2Cq8fZ{(7MXy`eCgHhO1`E1=RDC`C8L0;8}e;a|t97}1Ho>Z7pUzaw}NU4{gi0-D@
zAoaYb%M<9F-1*fG^MSAs>VZYP)zWlijX1ujn{3j0+hvKPiV5w)iXYBFhv7t>v0hY5
zL(29}<P7`c<XFUw<9=sNm6{TK_%9I)DF}t=+iGPu8oq0&v*aMr_nq3ljg4leov^!?
zuKzVE^Ux9T5Q1kOj@gtx!|VN;>Gd>5rQ{w(M(Q?15Z>f7Wwt<=)wVQ4$d}NNjSP;@
zp82)@2DeQT*)qIyB}M*|qP`|sImts1Admhs5DWGdDMA(b6^V|g(e#zMNt=9#a5Kpn
zom9S@7kx%A{1A^WM>_lt4(Cer*Y|+6Y*Dft5oy?6wwo(SW{5V6^=YF_NpB*GXPZ+%
z*GAMX#L|?#Q;;CWCa}50v3H|^8KZMYN_}Y~b<PvdCXdZ$%tOrR2$F)WfiqEyNTg=3
z_z4Oyl(23&${jQlGd-b0bZicUXHzfKTJIQ@Af_*xw$Rg|DA?v)C_m{7@pYowgZpu=
zjswLFO@*qBvcCx?$G&i@PBF~qHQ93A{~Gwzh(00Qq^_sgz&>r3wAw-EMIMh6Ex>M}
z=a#ovQ)|}cMcH41NS4pp`+CRO2cKkjcfNivKQr^yhd3qLmd`i0&SrkZMdz2bq?J>s
z1PNK@<g%e`N1kPD0xe}pmXwN1R~D*9TTX=3v5>Wf1q(4VJElTZ!7Nu6elr`(o_*WS
zz5V9C>y)oxqA`V}(P+9le06(NksG{E5lM*@8@bgQv^RZ>#Ez>>866yHMjK|^^GyyV
zONfx_0ghoZt>pC?tMBD&*z4vVNbU}Xd$P_f9@k4|KCzBr)Waq#k`*?~=#z<<C!c5`
z&{B^@eMs$J2F?$mZQImptTK7-?8(!Hx;Qw?cUH;l-=c;sZdUTIM8-rL@`kPd>NC5@
zYbxfI6O=7TXFL;*3Mus|*zrPIxZ(8`kJ>JssZ8CzgU`(NxmimQ(%_9`i^Jmp&~TYd
zaK{Xcc-ZEoJfe+m-PYA<A=QP}GMwWU@nDB$c5wHi%c=88v;n+L+y_ytJGp)<ZG|25
zKx@bKh5W+UhZzHY>dT|RU_%Dk7h38J+<I=Ds~XKoDvAM<mFQm|`1P;-CLT4d!8=G=
z<Y*;5ZVa3bMpslYCE<AN5c;%bMOLi&NWZ+?zm+#&jOj=G<f4>MG{mxdV{4W}?Kqk@
zj6p_qRjsMe_Zh|}^80yO)gw)ck7WICCbF}m!RhdZ$;p#0D!*v{h|btO-_X{FT{$9c
zoBR)Mv1@QbJ?jczy=zRtVZn<q2B7QIk}R015>PCpFzIiFWkFQk_9~XgZf+5>se(Z3
zD&J7T_SI^~UAH%QpmXQ`J`}Rrb_J$eX6V*_7iBjo$QoOjo3CNV7WtJ!{ViMsVyR_W
zT%10>;<XB(m}T#L<CnP7{U?8}``d?;*xkWvENVdZyafBuAudTON<iG+lcp1L)fj@r
z28k527G-QAInLhqTFx%0TJ)3+-m9QbrrbWh4b`44XUg2X(f;V&?qxG=;t+VdWmPyC
zr8ej)QU6$-Yj9*rhWMqz2D{u9c;wLm!$jnW*4VBprwD&-e%G?r!sO57K`9zp(-dc=
zndcJyMYGs+i1x@ai7$f6^!ZT|aWKc$xs_Sm_|j>lH$kYXV=X|pUIjJNZGd>-itSuG
zzgYXQPsA0?C^6>~jdK5a?;US<l)()D0=vyj?_H(@f>`$Yzt}^O(y7o*j^R*c*?iUF
z!SJu7SY_2p$Ec{}tmsW`i3Hj0CM+R?=XGxJqhwLAEjhGt=1u*C3AA#9?xI+vJFM%Z
zB7Y;A*#%y%Wh3F#s33fQ2Z26h{q@V!6~qT-+w_uN>f8_Z2hOR-zOh^sJc?iqWP|+G
zwtqd)(pi<%w{u2jN`-9VQ*lhkt2t}7EpZ~_KHf9qZ4&|7P}yg4bICZYtuKTh?>#TX
zm!%)KxtD(p_%DyKLzeGHe7EPsxlj7bySMCo-zj?@ci|PrWYkR9TXB#{ZCCXpscEnt
zP`+}jTFf-dG+|rnp){ju0b*qh1-~gHEi2D7vz#GOpr-8qZlTFhT&##D+W$SQK(N(N
zBIemZd)b6VG#R4yL&~+HqdMyr(VSmdi9te&wX)T{o@68`{XDlZmyFZNxE@nH9l8OJ
zvx4G_&3b(<l?c)<l^k7ohipP}AcKf!WKpJ+Lclbdd`Q@ih$zWKc1stTJlyel0)QI|
zN^Q|)&zp%rQG%Wj8A0zIjAqRAZp~%W0e$6wk3)Ujs!@wX5B2)=UW*F)q&HXwH0Ky+
z6rO>2o$G7Np(%$VgrIEZKf}hu(K@tcr;C^&MKn{g)mELL%O}hAQY7>T;QR@S<d-gn
zS*7!S8ctKd<Z^;&%(&e5TWasupj1BI)9^14AJ@d&8zOPnYdENMsj8H<Rsp$YLM5sv
zhbs|!-%!lK^EHvFOwHGmwQzzf6D1Q(PMYbwH$U$40U8a4w(sW9S={5+jo|jw`(K00
zfmFdQ?2j`jJECYBx53N^&WXpp0rHQrO2tP5zbDL|$7943hcmMJnEY>*eNmh3*i_Zn
zuhyT?O^ZE07v#`^sH$)k#?lpipSYPXkoZ3OoamA(lIQWtEs4Kg!sRdaS`lZ$R0u}}
z#2IvM?lqcE=P@Mj>_n7_Xwq4$;hmpD4vUV)Gs#~mwMi<lxsVbvfs$bGpOT2qfAt&k
za@_Tea8^4hQYp}XR%DG+1~JgNDT+K1tdyEuf0pBF@tDYL9UaNB3R7^4=Fw@>t3&hz
z{`KjV&8p^Nh^Zd8b-weoOlBjK%l%A|jWgY03Wt|_tZi)&^1ow;wHDRqH_aZV6m|)O
z{4VDD=(w7I_>yQClRw7=r~NwhzZWzo);QLqt0aWDpnc^TK3IJpuR-3=2F!f?IJ6g<
zNa%LQ8{{ddlYduw6WeSR4~=()gJ}mEvYu|SbM=i^d>`shf1CM{TaNPe__eSnOE4&M
z1a$GHJXjhdp{xDKSL^qbsB}x|KL>{I$y8HEOHBu1w+N?qIE;}^8=IFpZCN17iCh6j
zABxK{l$Rw+LZMrgUvY0Uh$|<Jtz}71zuYuITBg()b$koRiZGTpCdL7*VdGZll<Fz2
zVu-FDj{Xh3LK2P{Biwa0qaAE?{SB&%dx%rHgpV_zhu24H!TtU75=BGRSP&~<z?;>9
zPw?vQO4iju5g`JHN~q1x6{Jn4co0H1ikGx!DV#dm#O8Rj?C1{;>TLo<2tJ?@Zqg$9
zK3~DVU&p=w0UlAyo#EOEs4W&cjMkm16Dj0g3HrPXUTwT>_Z*}e`1GHYa{kilBOrkB
z6L4lL&L6XQ<Zi7Pvub;;8DPj!gEy`ze0X%;r4m06WX)ud$N{e>$nN#LPQ9CSdwTCy
zZaY$A>QXC^@I(G0e)ltq2`F5r$h%1v#Rd8P-_;f}w`NKe=t8KDJ$ub8@DW5w6<M7o
z*EUXnI+$-}zQqpMN~B0w*e6mosWquyBQ!$0lKu#ZmI?kDvG8!|$Vd9rOJF{EETnp7
z(s!V|=3cD%t-wdMF1<b_c`?b9zmWlcOq(F<e5XqIi+mO)n1N(Jc9(}hwXnzBu7cN|
zFo<=KL-WMP3hDnPOMsNKgy#J6b(Oz1fZ$0pIDdBW*^m{q!RB<aSC+Jbe*^f4iJ{^D
zw5?q;H5;IeC+i}lmgZTU5QmK~-D^}!pGaVV+*hwp{KQ@E!;R)sdi2KJGZo5<$v2+H
zjMO5xNTDqdpn>-nI2oD9;1LY;19d()G!DeHd`{D@9o$l>#gX<7r?9l^Ye){a>k;bw
zn)Q|zWh@#qpJBKo=dr<iVyRT|@%54Ob;7)7^Q83Kq&R22CnwYFq$&%=W0+AP>Z#sx
zg<j<+^E|_*uk&;U-k5(K!+gCjp%kI3P?g>{zdfw2{fY&=kp8PRXKhOzlpj4Ehix{V
zxa+e$Klf<tF365lS&B&2Kd4(6B8;%o0f#6D>k0mHLaCx1voy+|{z=5m8YZ=RMq+Af
zOmSzcS8T0Y22@VL#6rxP6+24WE{Q2fc41TvYUqgYuaP*O(t(uI87~c9*A%3}!_hzw
z^1iNu&^rm&Su5-?+BB?33Zn5Iwe1eNG<_JTMhs{5WbUV<UU58I)hBSVCL>OwJA$Q1
zR>dOgKcWILC};peV{m=(SqX<S4F%gA=Y&)i*+xI{XmSW9=f=;#IdOvT;i=3yZvR|O
zw7#D-Twm@7+&Gpb^-mZnA|xy+#tEcn6*Y1prU#xpW&smTdo+SH%DS8^O29+qHreOD
zcyrDSt{aKp(UI0>)`hCjD~7e;L7bBBt{wx_ianclMDH23`aTm4RhiZmeI@*;_Axn1
z!N~jmrPC&S`P>!^`u<hcLch0CbA4g%xCMP2!tTC+dGnYYyCP>sq@oA;l(5{qBL7N;
zD615yDqs4n_|IZk4GkuD=&h7A#zMQf9vj!)#-91EvmWdsAgjn|q2M)3Ov|vj(c`C_
zaht0K&w{l9`C54v_9?VE^O0gn$tB5i9{Cw4+o!6j+v|!0HM%0s*t7fR9%?Zao1Re#
z+zM<-=q@MkKXME{i@lZD8)?D4tq^U+zRF}hdqH?j3<kM>$(-}|FWeXeJAL;`a4j!B
zK~K^|T%BnWnScfK4=3A3rz<_&7&ev(pLu%m>aDVhuWlbhd!9IZ9?wj)UcZaouu8OO
zwlJo|nSDIgM`M3`u{0Zv@HejhNwEFOXo3qNnAox@AX&M7JKnNb?)b6qB3+qX0tm<N
z%K~4mCw18kUHv1NcK4xrFdN&fL@>HeN@Sg5`|}4>Z%72%-pTmmO`)P3co?%Y=@N$W
zGXMlaM=lLn6<TPXD)9g{9!#~hw;T2R^Qh7i7Zt5mvdZV!a5IS<MB&bzG(=H$<uCS+
z1Cb?ZTUt8^c99gt*Bz>Ml~f-zzT3@V2%~G6K{O`c#@i>YheO&&DIK-$nmC37^8cjW
z6^Z;Nv1_|GodssK&z=pk|HiQb5&Rgw@w!-qmgCSJL-kj4CX0|n#t&+yBpbYMqYHHK
z+kad|O%erhQ`cf<C}k+wZ;l2c47E*Wpq-yzv~_hk$VmWeep@q@*fs!?hjGPwAA-yD
zet~1+kZ#0cXFdxi(e*!h_iLM0d{)|HnQDT>BTkTIiniu3s1hprBZE}6j{r31R24BL
z)bqismrQnUd3OXWJ8knKj3LdcGB!!aUaqm}^ST3#%brGJaW8rL<ZNEN5~g)i!_>cD
zueu4xi5Kk}|CfZcb|l{N>qGmr1sQnPzl5$fx5?y;7G5xZGYRU^b_wpn!NUvwQlNmH
z!<TO+6FZ)Yi8wiqJFy<x7`~o`S0a)WNnzA828qOQ3%>SA^7-9=nHxU2*(MA$-E!x1
z<m117eM01ZdK96x_qNwLXM+d&7)PdyPv)SwRfpua=XLFw3uT~1$obIdxwl+zpL2h_
zPJR6Og?<);vc#_*a#u!hlAtK?LFxB)2Wb#}UGS6L5-Yqdx{(oTONYOz;PI!-H2*g6
z-r6GZ6BL)yGFzpjW6Fry%nGb(x(EENibGQ`pz*ujNhOdq&9>#NpRAZxbuVFM4d_x6
z(VJ=cqrAJ)i0%DuUvRbSU*sG`2Ytkv=j*RwrMDtW0UGgL2ONlDeItuI>8FdK8piiA
zPNfp)_EH!$6ijT4dtPu@W-nAX)!M8I0P)p_2D2A7LHv;?9d#Ug_37AZcq^rH3J}5{
z?Hww-+P4!}&5>{h=z*`+I=y6uHMW86Fu3?Cg&SAzZ7*q2a|N8MFJ4Wd<!d-PZ7Dc7
zc;fZej9e@y4t$YY9_y&06qQi3@BEQUuzj$oRhhUWd`}}H1MnMXsl28=&OW@B$xE)g
z$RC>~zR!tqk2A=ABX5mfs~-6b^wfG>xl^6b6}&%`(sNBq*sh*D@>MB|4spakrK8hQ
zpYBQ*+&^R{oafDTNc+2zEM;E{g|{3GZH2x5jr+*jQ!Dhg8T?UbMX*F(n^OCf+x<@8
z&E+#n4eGfw9y5PsB6n-UX^~xx-iE4?Dno#GV*ZsG#{|nz`BP)(XbT=q=4nOI(?qCF
zJ6}zIjW`@QE<ijo7yGuA61EgakQ+;I&+AHyDjQV%cyTf0Y8fLVZW<bo+hu|7G_!#r
z4M-@!W&^--0-?CYzp+BNvJev18TmpoE%Z6k<aRi1e5n`|)?FZlR*7XETfT*uXr2pL
zpjl!edFLpi0EL=dlZ3-4(!oiXQVeqKuq&6blu#jN%N<(YBgBhN<o9@Ry9CT1s`Mcd
zM8+|t-CY$R2?1LY^w=%d<L<0Yo{r^WH=h$X4m2mzrv$i_D^N@Ie@mKEloD)2?lG-y
zlQ`Rel{CtYdw!%WX~;Z{Zf4cZa>;DnJz2kpO?@ob5sn4r9~NdJ;zH!8bq(~Y_Pc`#
zH`jC;i>)hf++GfIQiHD_#EY^LvYA=p3YJARcaBwAY7<BdOON-+>Y>i>#O=E8(ZQaS
z@R}QX<}$~&<@64oyZS_~$cpx*e3+?zQ>Y1$F%UU;SOQ5VJ0j%A`os6r51)Xj^@*{U
zq}Zul&hL5keQyJSK%M<8Zb81E6BhZiRdtK#060L0gmHAloQB;`_7X7BAp2_5MjT{n
z(8?{Ec#YCvHtzRI)BD;{vP>kD;nt_}WeV;;y@wIx8hSu=d~c>RNYU{2cn<7&el>8I
zU*bDj;>xl~lIN8t2uAi9*FAKk7MQ&7Zwx9_t7MI_y0OL$rlnFC8#Va2+sG4qO%r@S
z+*=MzbA*9o!;}faf{yw^o42ZqVn7no3o#11yZT+!md9;sx%BFR%CcRC8)ycFMg%<m
z3`))1)!Mr82EtZn`CJUS&ko~}wT|ziKpf#p)eOeFNpPfhv;B$@v%y0J2Wx6&JQ-fe
z>hPuNmRSC$9aj6b=^uh#e)ojZDOnzYg6&Zi8J{oN$;eZU6RVDEGHTN2ZaxGbvgVMT
z_DxGFtvfVz5~a<4nFTX{8kf6cBD*2cK)lq=VE^?Mck&wl+t#^1uVG1rW9<lau6OPc
zOVw!7nFwQyGJdg(_Jg;>_jHUO36o}V>w*nR8wiQ949|Z(?kHBz+fhw^yT?T^pX9zj
ze=+pY5?uaH%J-58h3(}wE1}E~q&BT~v|wniOG%1M*X{D{m*@P6Fkv-7nmQPvcTu<P
zA7}?JURsO-X0q9-JZ4spo=oasq7u8)^Cs|!r4Z)<`GJMtkm*E%egpkaRL4FakfXP_
zV@{bkLrM?zE93F4JEqx%HEG+V0x?#%m4r<C-@KCusr`v7{PS&{lCu<FfEAz&Y5n!?
zp=Y<-d$)Gf;HLgHG{%OYfrRuaInlo~50gE=GESkR#Q!8_aEr)TyL>+zG_i`q?mQ8*
zk_`9ia`8~*g?2M>!s(12E?G+JR@gMd{cInqoqq9S4&!JRPG#5KO*TA2J6Rfre!LU5
z3L{Bia1(#^s}|BAK^hWzpnW$6MATzdv<>%CA*yT*fFzfmBhTrR{@CC4DX(V}apiJ!
zzuiSpC3F+G%h`{9j)B>x)t(!EYCxj|gh2s-T(w$m0PQ=kGQg}@`>n(zYpoq?YIA+d
zk2O<Qpuuq2sc+splhEBfCHhV9%E#G3K1Vy~YW)B+6iLFFBr%~3!S;`}v1;56t@yP)
z`SqY|JjUUo61E}v84prm)IPdqHM#H`dHed@Lmah04m>pZK}04{C)SOVz}8N)74xm^
zP`mE@uf<6=sJU)D-BO>g{RqrX$f9k7^nU>EKoP%4h*6RQFui--{&2pKRJdJk>HVn?
zHi5EMN-bC~%;}Dwzx<cnhM!VDCdx;poH}bMJZ@+H-M9ai`}%?3Y`@~tPgEaZB@jjd
z^~8vDy%O(a%U}fraThpK*ZgVLu&$u$E=2~3hKk$UhZLFT*{?zOIGbbQ&wud?RyEdm
zB*a9@Gv~ug)^nkj4bjex2vcyDYs`t`?HwPt##$SeBO+<&tTd`_p19K-3@pO&_Q*U>
zOjAaHbz40d@9GxZBBc2SBf`m@$$Eg<gl57dJ}&v)hxeS$8-M+q_ecoLha;DpM6HK0
z0yM@rv~#+gND^6OVVN@X;)30!ZV2vfq#-&3N0epCyi9P>jv|D9?<1KW!~<Y(M&|Fb
zHarFfPVLbAwea~|)*p&x5&OL}+5g6mJC>~}!2+V4zO{iI`F`9je-g{lgu`<1amn7V
zcdgVv9?RQ@LV0xE#<QLI-+%dUxGy7?+xhPl%O=hg_F}X`H$;^t9Slluq<`#K4#IH=
zwNK0JO|SJtQ{46MM;gnqrE7y|f4U9V?|I%glF?}TgnbCt(X`KDT^*l!#v&UasdkU!
zKJ?U@HrF5Pu6s6335jFqt#^M8#0B?zt5`F_KB|XNYbAzA3z3mdA~;a_?%~Y+c|*;(
zIUf1u?bp0`^@0#1>w2MV8`6{@!dh0kHnz5M6Blxxu<jb8S{uC^=YDZ#&?&p_R1gPy
z%=q@Z@0dd5o3~$)qURr~N^Q+SXp}hy;&jKW<zI3ezGi)IY>!4+mG|fGxxd`=@9zH%
z5AB|h{X3c!oL!rIUlSeAgk1steZRjtwC(_D_XcL~b{$ENG0;s|s}kv~{mi;mQl8M(
zklvtGq%~f>n2D3{^V?UH);X;!)trGf<;0RQp;mO;h@(+xAyU*QM3fsu@@On8-72Pn
zm>V~ki0e+ylY_Xb&So_qr?#N95M4hoK)+OAj37#&jonQU-hX_@<EilNckd7pj`Pf`
z8=?2b&xqc+ti@TDVIo){sWT*Va(%+m+AtZ|CpiQqkb<xTVbO)&y!kAH5h}Z$*Jygn
z0Q0Ym>1XG*ARx~sPrj^w^6LwL^=18FD-7)?U(eEoz+eOI8G|#Gtb;Z`mec403&ScI
z8C|DAfp!+Nca>CVqo~_jBZ?7%+hB_Y)@E#NjD(HNa9W(^oN`7n)^!~MqlUIh>4oFX
z!n7=IryT-C-KrS%gUfoSNd%Rv5Mdg#M|;=a$BsOah#`(c*GM69Grwe!TVk1MmqOn<
z+KkpJ-#`AE_vJmmF2ADiNV^h?eNLSH1AqSpBOx5R2sXrW#l_Rrnp1yG=mTnqrq9Zi
zebXOKKGqR6?hZ#Fk_UiQyRj9eRO7v!X}y!;JhY<>iGg`puEML=uE_Xs6&1yEfYF5x
zuIZ3c#D=4Z)rQti4~hmKe=s_su7sEf^Fq%4ei*O)w5>dx&wO}zz+xtcNGTIrDa09w
z{`7QHN-HEWro0fN8w9kq(tAb39mynk^4m~S%G{(#8uxoo9(U4}?97a3hv=tEo~D!b
zL|Ve{{-qz~+T(Z!+ybAzXS2PK56&bQ`U?TBBP@bjp(BE-gMVV`F&^ZaDE8KADAknM
z+)Pu%$8H|ijq_GPeDq*BE=-4+JWrHXY3i&4)ZF#VTyt-lGchIy+PV@D7b+gQM~Wgf
zVo|Pk&s(Wv3*1f4%2Cy;!&qxWa9TSlBwihU$}BIar$9Xm^<u1L<y0Q{>-)drvE6ge
zJ9PXP#(#qZ<6_S2*3%67d>Dz^ct(a8Z>KvofA6>pG;{Vwj_xWNVnVffVlGAsfwwPT
z&<NCcM8pf&i!5xZ@o@ivRtl4JQgARbq`-VQKs`}=MRGuhE|jcBw}#R@f!9v!a%EtI
z7%)|8ZD@DnhmMat)_m|ise0VY9Ik@jgZIS4<;26onfu2FV$4X2)Y`lmFi=Dp_id{c
z4bH5S;2nKKY2Aonbk0L?s$huZkT?*S)6CWz(!#Z%y)F^g-1sAFA5VDx|IcOpu~`1y
zGug(o`eTpf;s4>!#`r^v&aC+~mWA7Cac!2N9rL%xGSo|_^unc|`Q?Yd<e}a3$cMjk
zEQ^f&f5ft=AB@iKbMa{r`Nxmtd_$fxF(tIQLs}mr`ajxO#v6KG>W<t<yh!%-#EKp#
zF=}Jv&i2oJ9?Mzs`=5qGs@U~8+%shzV|LEIaxgUo&mGO4M(9<Y?PB83v=1=&Ol<Gn
zK@tWlMj;Ss2B)oX-U@`oG*7&I^OD2ifCQmdkKajLC)0HqtLN&BQ9xC1P!I9+)(p!K
zV;VrEgOO_R@$r#4CSJXGHU7N=SX~`JSCo_jb6$9L{DxVMY>!6WI<<5j%RL{~cYJ&L
zYc6_X<;;E&T`C}s$KVQK4f%)txfogBxX3ny+B*Zl(vAfOt6K-Pj@IJS4<Y(kCgO}F
zQ|^uvX-eF@e9L7ke3VD3xs%~?n3z){szPt${t6P~a9Qsj3lgt;O}Fmxq<aj8VWN^o
zePGnuXy!ol?%-HMzly%Q$)p=`IBU`0`@{Lf{o{rE)1zxnVUz(nIuNe3R;da^h6}PA
zEyWnG&3qTPcM!{lSX&H%EP)&{Idr5C$507Z;919rwbwUeciGuj*SqtKmH1yP(7vo6
z_4)!}eOW))3gGCtJ#&lMHINigqlvSa*4CY7iqoDaTnEWakQN=Ru|QrHmL>AjpTFdI
zJMq8%@4w{!bm6kDgr;m;XPYaR7HAjfB5amet5VfimYFxNU-H#guX+3WH90Y4jP}e@
z#>O~$c;=&KG@)KDw5p^&p{A%*wzIOzMjvdPR0=_ioNXK|Mr*E}mD0j2Af+?U6Tf)z
z&pGOAK7QM;K)^4*`%B)R-tiafzv0|ZoOl=%e@3z@By(u5JxdmAQ`9!5Jdt8x8ZE^h
z1dULW5mg2?*WK4rm_~<WK`gSa8^?Je=E-N}yHYL}f<$f(Ge}?_bM&oL=*{&*qWKIz
zt`#=Vi5L>5u5XZIfUV)yfDL2_-9|$a%=5xLFHWCsN~xXD%@1M~RE*1|B2qb*m3f|d
zIemp$^qZ%GHR0j$J3gK-)D)Og=5<U=A~Z7+X=s561Q~!?VmT~?<Qf|tcPPx$LJ01>
zGZDBY&|9E2VG1)*eFP18dakdJ%xH)1tStDHiRjKqdwTzQti3zE18^{k3Gt)|HF}6t
zgSBpi;P*e8KWp0j8inp79+->er#$iVKlz56n<M}GfBFwRoG+AGF@@A&)556+H26LI
z?*7cBIP2kXnECoAUvYbLAc}B4pXgn=$usVCWW*R)(&8R!TjipUL=&W&I}7efW^bHY
zrJY7n+8SG}U=b_b87d*T-bPn!txy`N3rBAF$>E<9Yv%p0H)>lcb>-jv>fiGH`8(eA
zUs2gmHjpqzF+65n69^dYS>2#jrG!RGg2X_cW*;GvA!Txf7W#Fxi{Ods+ppemJl=3R
zU&t{MQ=+sps+HCnA!c6B6Q)A%xS(}W&TB<$CEi9Pii5k;#MeLhhL81I`r}42zv<Rx
zfV2A`REH1=F)+<Dd74<uMsJ;MEAGgtfjY*d9_8$#E!#?(632Y=#A2j@86Q6WmX8l-
zED13sjxi9-*h)i#fVoax@152wDJG`FOh|!FN4xK*)3gAIXhMq0A`{sXize<bjoSRR
zb?XDT_H)KQNrnGVH2m(bpGzeF;aJ|^?tUhBF?W&u3}QK|8?E$RpH3TFE%fj^u}nnT
zKtULVuNk(sKN`!ec|8A{zy5#o-RZY{-}YFpfA3fpF)(L>UArBAfRb|lM~~%sUdYqJ
zdA-nU<GgM-qVVI5Wi#>n+WP==;va{kMlSFPH1YX_@zB0vfXz=Yd<QB_eY`oCkB&-u
z8UYkDY+NtV$v1=+$vKkJ%%A@J8;-{#|M6G<nakzOx)v-1=A3EuLbFENZcvGwo6<w#
z?VC5edG(4nuU|0@uZCJH)|EUhF791c@|1Y}@+EW5L?Uh5=&cR@lQW65RyI@W+MNM^
z-q_YkQNyfr|8U}FiNr(l=Pg9Ex^B{vk1QOy(Kl#O-j@^S?ZTz3{KfnKjdz#tIq7>k
zuJvX3p0E$6#j!v1(O7?K4EwdlWHbSu=z0!W>zHcrkv=0JQ8-RF4$60Ddrv8ma%7$o
z|I@$vSF{GX-}qRb8kg4DTIaN$(B60%BFE?w9Mhb6@#;0}{e^PgklFR#Ox+n)#<YTi
zmJwAUB?tUJK0KneQ7;8W=m<SR4F>H*5+TbN>pVn`IYW%@c#QJ#@jdsC7h-}r1#aev
z6auX|Vijb<LZNpT>tE8$VR=c+fsmlI>c$=tun^tQA|;la8>9=;1Ctn=2tg7e%Ckot
z8-Ut%jW2eTV6MIS^ff-KSpKs9?&}MH^=18F>l1lBV+bP-GS29F=}07WXAX&csuN^F
zl)=T4K$Jv^nHRTrym)!!=H@-6DW~%pZ9X3^!CMI{MguBFZKKl?6UXC`d74NeVmij@
zh6wuDEuYa>?Rr4CGmcaKUC?)Y-jp8v{YmLqFk<2gVEOwnk~mR0n2^XE;~i0F+GZ$~
zZk3Pc4}7@1<5BO~>_Vq7W+Q%IYGrgU&BExwjPA0A>sBAa5QrR=i|ZW>(zK-E;4$2%
zhE=sQDDWhx4?$~nmPe3L>GXY^Q=bZ;?nzS&F4F3+S*}!cP_7*GrzZg!uR*nY*W52E
zLH*yE7`1t^Bc(W)&^o4r;WwCMLEMog9};s+9HvB0aWH2baoL}-Fa%Cx@{X?0`5PfX
zN-iGT)m;W-8bV293?xW=N?>-r@&K={dKz%m$+_OHXFRuOPxz01?JT`3ZF>}5dmQ_n
z^=@u6NTXZ0KSK<rR+NNyhNIuB055Ltc=hT9x3{;Rz^$%z(mM3o-8FE4keAXaT}df1
zO_{@CA!i?_?X9~Vd(5s;q2~80coAc&^iuuxcDu4IMGb}}dmHYY)^#wST%gpVUT>S3
zcUYqdQ=FLMLgqm4Mp+BzdSY#7-e2DFaeGf?rP0`_B5siIY0(>rz|K_M88XhExU%)&
z+4t}`l)Z=+85ZOVF=bk7#G%FEW)3#uw-a@IZ#809qN<NwYaalPjRH`ZmxU10*sDKv
z=JV~uKG~B(aQ(uj&B5qyV+O;nY?x9TU4&Q*CO%%Mx~pH0{Xl#?WIkk;n3=OPVD)-k
zp2v{Ch8X=?s_J07_~#5~Qicd28L00~DP)ple@4ST3T)T?>XYpGdIO%V=WXB5NO0p1
z#q!l&<$5M_)fW6Q$FhzWEec_r)2Q3BH-0yk1KO|iPgizSF#c#PU)qVaop^Woo)7Cg
z3M)$gd&RQZbAFOtpS6g<i0L0iEQ>pPwdQQ8y-@)_(pV1pDSg{PwDGN<8|wVtnY>?*
zaV<WVCD$j&FSU=Y`qy+^1Mr-%kLNtVepvgnMj#FsVb86FB*Kg19WP$K;QPn#5rGfq
zkN$m9WNVeR6;eoOh*UL@z;RhvmU+Z*b4F5kEm0lj3cYtS(FMZsO-7rK-i47*1RL?)
zj4saL?9CfTA^C`CZH?AIHGHx?2ZSsWS!RMJYTLM!jr;Y1$L*1i^#hOX<Is`qBVWTC
z*5<pJ1AIb|0ZG}|Gxh122@>4EXTRp-S@SVNKa-K*g6;E6Oc^y-qzz$=V2e*Tm<qWZ
zu<_H+rE;kqDTUgNDFjjo&ZG>1WmyQPf+|mAh2y38YZ$w^x!~UJ>{sfBw(bBSyC&S2
z4C(HcN(iV0q|1oUJ!VGE5rxB&nA5~EXJXi||2|Of4`WwnTE+NOvlxSn5EBy3wYm(F
z3XR~+bs}PsD?Rp;eEA7*e%e2;nBud~jQIY1Lix-3d#o=2)|d5ztuT^<eKWk~{IXMX
z0~!-OMRZC&;w1sA6?$ixCsN2Pha12X-#JI-n0a&ihOghe;$Q!Z#Baa<!2j}_{};7%
zTJ<hOMfuo`O<hPBg(z@)nEB~XzhaJwayff}YMw|7K<(t1In0ZLKc;N!<}=${ackJ(
zn|g{My`u^jI?WU<8!>>S=#C+H(iCFk5L}e>MvmObEjRWW(@5=|-+cU*59jaskMI92
z5A_3UcyLk1?mMxWAL{;xKDucW3_(rLN?Z>h(m5U%Vt^D4jbK5DA+wMnJJW3<4X`Pp
zn%GL?a5Q8Rv>4jm$wcUE<${lhMYdY0z0tK{baIFU6E3F<VG4vLBVi(jd4QAVht{Z&
zf*9E=5wy24?ig4vE86|9wQg8_(kiK~0g#%%_RC{uBGID297uEIpZ(+~thMu@XR>8p
z=?!^Kq&cJNI%O$kCK;B!QIIS&I%``wP7_l~91a->oL2|vs#b(dh|%q%<AO*=sV?4$
ziw8exQphL8z+OxVfi%w9&jq}KJ@NHX3^k>Xaa8}D4E;6Nt7zj{>mt<B(b_o76DdyI
z9!DH2q~!M9uW#S-&D*aCG4tJr_x#WQ%l|?vja35MxwE!L3qk`PRwKs5H?Lpu=G9B?
zULKi8(rw)xhIWd4xm-BRGgG>Am$Cp`xgfpM)`poE;W9)SlDKectrb(_<%`!y>pVU@
zAXx~D14S_gj(K7V6JdVEA^n7Ug3>oG^@0EJ>%ZXF_rK!1@=G>5)2Po1)t%kr&wH{=
zBrt~~h@znok|Lr6(ZPyoLuuRh5(IL}BwI+y#gwfXQ6}d3mJr6sY;tS+7$U)psuxt1
z_4IiC`MtPhwzfiV1+$3u&bDrt3WtMh>@<ZL8>87WEbo2nVhVtDS~bKXLZVi8a#a;~
zj_pQQe@;#(*Rz?m(oIO`N(qscDmJ;`_0PZln(|uN>_8@Q%fvKIXh`&4P*vASF(rgL
zoU(*Yr_{z_o=GV)&j~}gG-s149g##7NO2;?1HH@B*pochu^X3LKK-+o2YZ}<e$C!L
z6wB~*Chub%1n<!Q*kf5YADaoI$PBd;#X-2|u^d|G^6&^5WRWXl_WvxF|NU?N9lyT+
z6~Enn$;z2V_cHVE5zF51Xrtmz<dDebED4bSU48_ytiIiJVQU-O)S1xwBaY>;pekb&
z^XdHEUCP`{=&2aqeF=wmE}zuEpTXK?8J@pfWjI7L#=p$(+osK#Yef3iu&PWskz(TR
z<_^G(5Ry0VUfsOm>(_6&eR;$C`w#rj|MUNY);nv;e7H0|oG*k+$0V?o&OB$ndh?p&
z(M7;}bwi)MyS1(x0XO*Qty8y(L?gsVTQ_=b(A-8N<0Ff;_&6Q{ZLMhSyn6ixX-d7U
z)Cj#lNfx*}YGRtcCWV*O(m0hf@5=Z5$KU*We*N)V?)&#twgGU+xaosMihHfj97&SM
zX`)lv>_o6eEsdO#3$&+*Dim{PY6;1wBBKAh7!y+rEIIP#&BE>7olADaL<*T!&-7ji
z+DRev`smMY?MA5HR67;LN@rRUDM9Upm=mwxyy4gPE3H2g)Cj=`WIYZ7D#LTr?5c-_
zMb`uDjApk3`n9?C@YXgW#*29&&zUqQdhPUDAw`nF&wlcTw^dO+5=ks%mdW=MHD_|l
zXbOu6F~y;u=NfEN%+A!EGH7DDQLHvvtkjDTnt&xt6RjCyom#6KGKBaHkQ(6Da~FeW
zAlK9SOal4K`g^S}0M?iFgDu(7@aGiyvCL#|=pkk_oc7$CcM_w?(2j8XUc;rZBbgK?
zdNJzSxSbEw+s>PpFF2phto;Ny8>*|g<#PA#%<VjJILs`Eg=F5fZN0lAV07Wbx)#C`
z$T@*Hg}pYXWqU`-hHl8{rkLwqBpa`P=t{))q<^(v3&BmrZVopr`G~5p)s^#l=KITc
zyuW<MV|!rL3%ZL*%HClyqImc6Hr<dzBF1GTe9G2V)^;Jr=1zZm_cOR4v5ro93|?dk
zfuw2JC%Xc`l})A%b%Ba&@uupXXA@z!2uV5m*%G|Nxvtg4^|OoDss(g_9Nu+$5=GvN
zKhGtGu1ZsHE=zJT`QQ|0SG3sY;r;ryqG-2qMg2AAG-1*?SSD#AHK$ms?HXlcRAO|A
zO<lWyBuMhni5LT62yR@H#I9qE3{pzuoT0mhNH;~yyzg$$%iBF49`AXyyv;_t&OYq~
z|CrZP9K+KugE?tK&j62$E`pjm2(3082@`?j$rM5`?{ub++>Uu`+$=}hP2=^8mu#1n
ze%a{OsU3<LZR9OMgyZ4h!lZ#=E$wmX<GS5-?7C~lrIfJl323YpY26uE2r-NfWe7AJ
zz-rxJcV1j)XHzBY4r&9KoO1TAD7WNzpi1MhKJxDI*Sz1p=e~Ymvvq)1eg)xa8=Pb!
zhs^8gjvO<2nz5j0Tv^+RwLSWX5xW(SgRNoY-N6EJ7#FyAlxva%HyZaYtLYQ72A;)n
z8>2OX4A46SXD-H(eD<c81wwR&gGjmpIiCEFcV&*9?<28$OSHE>V)M%H@c73&$^JS7
z?a->BjVD%!2+K4Pq7U#U8;*A&_{g;SUc8Iy+D;o1zn3utVv3{`2vM#K4^yM{N+c1;
z;}!WRh}rYX^;7W5*&K3Z8jfDF|Ak$D@As&&Ki>5RV|jl&<4o4!KG;N%A9E~s*NilR
zH;o1ZVt14K-B>QxoIOdrp5HlzKN`y)R*&VvN~hsZy2s_;HI@Y-#V|l=2T&rBI9%TL
z&-~-Y^4Raxb<Fnt@<$!Z5>R!vIfLDG1+n%sc;quXtJ@G{Ka<C@&m@+wJJvHM^u8_a
zXR@CQMh~tv+N!xUgTd~cEJF)GO#MmR<G3h?Wu`4hUcY!jEfq2U-lYL_ET{t&Q)Zbb
za!MeETBp}WOzv>lHV1mrGF`8K?~T&CpB1em`<Xnh|DIa}^K=u1oF_t1th&*FYuW9&
zuDe}F7SP10o%nEh&u<>T<-_)#Q-7dQQ0Z6JkRjt7ojAmqIb?403v!w`9&hMUS?z&y
z{m8j~9JzuF7b<_1-t@q06?2W$Fh(^)h)mN=jESa79=S5(zR`8$vwfUH-l*OUF}O~8
zfOa<;$<yQ`lQ8y^r@ZqimY-s3Y**K3LqbG!jE|20Onvug-Jg%%N8@Se`ZD4;219Kk
zXe6ueBMx8<2J^Tk*PSYipAAU#=AVSo=Jh%1!On|AW@FSKI#T5EpO4Ev<;^?O%5UUz
zPS%(8-`e^DV0~FX=yD;+y;KhaB!On$(QCca14L7L?bNMt8#5_H2W%`s_arR@CCkhb
zZ#dr<LcizL+qWF^!oT{nKjq!Kcl>4jrGHQLGDF2EO_@^U&8ru@e)WR87q_rgY^$u@
z#RsQ(LVM@_<2^A5hvUSyT_~lYr4YJ8akfwrAw^-zLQBEVn%>!Z<-Hc48{gt5-TM%P
zU`m$APu_e(mN#t8cz1cnw;%p9zx?=Dyt{nQ$~{VZMrj{P^%!$rIOGF=_WDn_nT{;S
zJ3@wZFn<01*L?f^uMjyQQhkh4gq#wsy3nO$A*bnDfXq6m+!6?5OtbaIT2_L3@eo`I
z7VQmijTj#bP;PH#%pk=`>&AzVXQpG~W>U5WRRet->^l0_yPUv&<m~h7K?1!DEAto#
zWCC%FFDQgOMo=rg8|$Ugjte5m4}XO@PsC(ELNpSK7X(ci-=}#J)0-OxNT-yQIo~iZ
zlQUa_Yr5<Lv(^o*N=Qmbuq+cxRe12{rJ290<WKrIQ{|H*)t>(<49frKwQK;i{vMoy
z?&Cm1@5Q`OD$!`IP`Azv3$tXROhlVpC>>87Q=XW@!sXt@OMmjyR~%kv{^g(jIq%=S
z=dXYHH(2SE25aP6^rjSf{puyhn*%x+hpjY!K2<4Y8yS2f#K7_9!0F>X=ku9dtB=}P
zB1m+nuOK6-@?Lz?4ec<e8CBujJH#-GULPNrmYKlBVn0K}9hY?B+sj|`7k~5r;X`}J
zSs&?afQ~}cf#JRAy_;@%IlbUdzxrp~9B#OM@d8q(O(*{5H-E`r{pvrV4yn1TS`elj
z$EXx!i%goY4A)?hL>jT+Y{jlF%)E=ehcw2Fdm)7E9qV9-bW#>9My4^|vu%|SXHuF-
zX(0yJgbP8~ohEmnHZVqaMWBN=qm%|}B-}3C?XeV5r4xNbq<1PpxhT2_(c<-c9S#R3
zg<_G=Be^CTLK)+pp4fYXLz|2LB1W3CyO#|Ehm^dq4(gr%%jLq7Gjp~v<`FTAm<D&J
z4bPel&zKwj>lO3iQ<(bmTfZC2&(Gv;<4lgm9OXwJ%XA;POOrcoNpx_z@M$b}gL~V^
z7Tvi}{M?xie>|4ox9>UYqwCiJ?(u8vY6I|hh-HLj%4qinOtQ#i`3fXp>BkYvgKbqe
z*RslFkCz{7ER*Q!&##(tWmb(E_jyxE7<KiP%weOU^ck+}iw;+T>^+X_*r{rO-U@ZA
z+=c^lm<W;)OK2I58FL-GIV@aGm3uky(_h??k9Yiw|L#w@e|+S-@7|+!VJmKaKoE|H
zBX>6kZjJ|1j%aWlt5PaA(?p7q`wtHU5pHh}V?5L-rJ%JTYel*ex{tc$oJj~NgeSrH
zW<-U<d>{fLOav=5T`|#tyhcQ3vfZ)u$Zs#d;xFI-6@T&Vzhko#l?}w%rVNKnRLHh)
z;K)zruXr)v@%8JkIUbI@{rW9ERL*?lKmF=2_z!<Q#LLrep5*LGpRNdGGKuV?w*!<;
zA#*(5k#b_Kb(j-`>zP8Wo`cFdbg2TR7izB~NA2uF{H^)a#j@NWDN#+C2WVv@jsf-&
z!Vt!IX!mhN89<PvMDJj2#CZsAJYh<$AY!y4^aj=pqPdpWa&QMhP0#~Uf;*b`?jW4`
z=!6W(00)V6G}DDNMe^i2v@SK`;CyWrAu^=|=CU1qcfcHTz%nqyPX`G7-HLktwja>;
zSqAZ!^><xg0IV<T2U~mJ@L6-nY0kJ4QD=A0>rlo9Xipk11A~Dc1^_Sa(zkWa4=Y+W
zdiBll<?)WT7PrwAqbnp;;$Tdr5V$?wFi#V;S9;yhwSc&|XxR$d`&Gw<s|J*g4<niP
zMsxCxZvqFG%=M%+2n#bxrM2pZNZegQ46}wI9P=%`NB-*ZZ+QRsEx&pEB@gWbMJ}lM
zs8BHPz^XCk`ZI5)*WApvyo`4o(}9;aZ%9iZzwj}RDM!A0|5serdrT8ZBwihm_8w%w
z2L@b=OK|Y5w?=F3Bo}f<+VCLP;GLd9v#{at7z5bQst^fuZQMVcnQemQLhI@sUyEQq
zTdxExBu~agM>i#Sw^agD$aohn4lu}wGFVY*^3lS2@sY#g2ds8?D@ifXQzn=(t0yO7
z?v&F|7oOGTBQYlQD%ZqUBti&R!D3SvxRnmcGE1C^c_QY_yckOr!sCKzXR9Z`uEM3e
z7Q&8JUDM(#iZy2WW!K&K@vj|zzRJZ~@YqoEkp*++RaNC&x))mAlb;a8MLO*%84>XW
zsu$z@xQ^$nKrn8ng}NM>bLKF0swz#L>K^vayvPuu?X=pUHo%#Nb*%`-ocy_oL*%rU
zLRM$;P0@ACoVDS_TV&{3Nu=W%IL<tAy8Yl?IJ0xy=qLx)+W9};{cn7C{~PZ6JsTHP
z>NTOXou6cZV?1&bZ+H`5a?63(gwWw`IU<vxhs3tMp_UWx9)826Dk2W#PTBSM7>TSi
zDg7BT^=Gx$MvDwXhh?~Edd!*(_Q;-p3xic?%u^=ej%Q|YzHH<inGX>iBL_o`X1J>k
z3v6>i`4lrLX9uEj>udos>FQpUqmNj01r<eG@e;EKk2eFKmh@~@M93{XV^_GaxM{<*
z(KVnN7+43pYiLN381m3J^sfa=BuUJ1VxF>(Fllx7IWo|$T19(uyr;>}L)4!A?lTG9
z-&=n$mR%9%dM0-fsu|}}fAq1OqVL~nfD{sjKrx@jvertda%Di6*;TsykHvCfMMHgd
zw`~AT>F*xP5}BemK1#0+oJ$-GPyF2a5yWy3&l{%TVAMpWb*TJUW4VDv2icyU?mx1g
zgU^Ov$mi^iAOA9*=hiYl^lDDf%$tQ(l~eU`KGWUFE%=Nr!O`R~PPfPL<za(rL@r2F
z4tb%^%4t43sF++bpqpzvEz9g=o8DclyEkWAscS2BRR@_<@{btVw9rZ;kNiAG*Idid
z!8cK(o1eQmCQPA~iixlUZ!DJD2x&|hgb9_2%hq|=9{JDje#`f#@5dN_aR7N|<e4)Z
zCl1`@J6`9P+=PXxM<^9bl|!B&C#IyldGU&mr?+h7L~n&OErbwAF_I*Bjv4~do|u%o
zJL>MF*ZXkZw4G7Ol~vta!8IrnG8_tHz>N>0175W>)G{GvBxFRg%dHHysln4IJkHcQ
z9*)k^R>&bavr&aM43y${Zb}_2OEZtN=FYROi|bm54)_kOJ_IDvQJOJopzddh39WW?
zQfFNa6Q?{SLfSQjZD`Q?-jHI%Mm}SGXbE36xa`_U3PI4mLv-U!OBl@H@Dv$efUN&!
z))xTl%lg4q9~AQ^!wyfBx`BJr`Bv0a+15&yz%0>aouc^Yj)?0(1y4@38?7`>_l=U2
z)5jBe%Dnm6SG09=04tblb2KH6@va;KFK%v`=ZURvv~8uY8`I=7>zB*<3SeDDYrDC|
zQNwj|%*PFu=q#jwUBRg=6G&o_nX1OQUx+b21F*nohy-q?8y>d6Uq1Yj52wH3yVGBx
zQc>}4ryvA#%?H7l<iKsZ<L&$tUM+XLj7JV};U~9m$v27ljc`~dZjJ}e=kGb4?m<1V
zAIIsnqoK%kY%t(-<JKF!CD$Dw5c7l<caU+axgl}f)RFH5F?Saf2ZDMxK0cgTn2^JT
z)*QgXTt!|jpc0*{BsNAQ_mKoo0_HIJkC5P^q3Q=<2m*QWF~tzECQuEngB>NV?@(rc
zT~qO-R5egdsn#e%vnvMIh-lS?uOW@(ArK`KLv}_+7wCa#aER)l-jrAtVW|O;1E!6w
zJ`jSBKBn+YD3v`S_G`?Q<Ku;c5D3rn=pXS?`<;?t(A~8UOJw{*&9G942FZ0>JyCf!
zg15)9b?8MoJ*?Dq<MPnRQ{cr{FKA0)o+b{abm^Se#-@80adhL%=(VD?VYT_skw7V4
zOx-L;QjDYH=dPI&9Auee8kF^&g0Ewgz^&PX7-ikCI`r8_f#cwn^A<qJ;mBGZ_;<hh
zUpdzkkNpusfzY0=(-4If0ypsm-^^e0LMINE3Duw(x6_e$49MM#_7$<pWqaUKE(jB0
zumY7%Z3D1<c66I-I@4;UMkvKa>3s~>$8;DpeAy>%gAR6=u`Jo&QH-?KIh|IP!$e*N
zL&?H$01_AYMxKC6GmbGa<@wru&JqbxnD9=qHXZsQuEo){vlV|PtUKF7%{4mIz-E8l
zsf}mG0;-MD8miTcUIJYMeI(nk7rnzZkLX3Y`gmoQ%#vs3X(BJq=B%~j@t;vEpCrA%
z%h>odRONSm|M_IgzZ=V9L&MH&oXMa;s3?!6{-|R)NFWa#xVYEh2F(J^!Y8q;r4mX&
zk{3E=1B?z{4E@1ajw6<@9XKCX43_`yu`H34oMByRMI;bOPd{S-SU;9no?|8;%$C_y
zY1V(VvAk>hs42nr?CGg}`iEj!kSh~wZwP!2Yd1vl?Z-zthqulUZXdC%#$#!CrW4R4
zj(QGfA5)Lx5W|RL<NR=;UN*{UB~QYOZ(hS<oR$N}DKk%j%j)`C%d)V{3jvqo*vg8G
zrs6K4p<S^Mh$)j&bQYF*9zTuzG^a=mLJmX6YzNyUFvW~@sKwn7Wtr*S<7bjY3K3*N
znK+lmyVC>z={Ns@)h-mfuuER}jSpanC<_ZOd6{4H_3;g{bV5^fD_BY#awaZ<&Vk35
zuQ-*j`S!OP+v!4Hh8uYDAyx<CeT+JcEN_sbkl=}Pt&Ki+RQnY?k3q(mZuJIp6hu3s
zojCM>;{f*R)*vQAnmz%rR0Ius9<~de2c~J}csP;>#4b!Rfx7%akfBK{X@FqsRZDD_
zadDN6;A|*mJU2UFAOl2Knwf)Fx;r`sYU%XFnQ?hADN`Ezck>Gq!Z1i022PM*Oyzj|
z4udG$^E5vbbkvf2Jt?-k!ManUk01`(<(KtOetiM3zN{a7T{RcXL7!ko?VYNgV(czj
zYLQNY;5r)(?=YZBARrd8-f8WEG%qCQY)ETt4-d=+|HCi-J08yuJe(h?wi2XKv=Wjq
z&zWRG|9C-5MYVC>{E&S-JraQ1`G85FHU~i54%bBzO<mkJ&Et(3SJ$5uT}(Y4Cagg^
z7uTPPLhnXvjZ<xOD;&#(k6Y!pkAKZt&)DwfVNTbDfc8L^iQ9O~?R3kF>5i}or}Yu0
z6OTDjbmisSJAU!iFStF<q<qI;A8$B69$e*0Ak0FR$nBOnsxWmW2Ok$`p@JwOHa1=P
zo0BnbkvBbIc=8_Ucm_I;&Qe0wj+Ktaj7g+4<y;%5ZKH$}HNWTFR(du^bp$p&QRD(K
zjV??lgh-B&6pS1UndULF*L~DO+(j^^$b4`B>;2_OF}T!>_C(nNr`QS6&=xTj!r28N
zPumF%fgU4f4i0uQY6)m`OI-TIL1)w!n#=@A)FHN8Q>J$HLN^77!iUS5cc%@l6JiIV
z9f=wVmac%SNgOG29IQfj!U{Wp1tZP*JhJ6a9Gr}&5hQcRnXW<)VQ4QT#-mK{5tZN_
zzrAlZu953nDD6_P)nHO=ij>7i2!8gHZ+LhA9_~Bu*N!N}Lna&&7L2Biy7sH4M=9=n
zx0S+_63cSOlonjN&6$%Srt#qd!_Ee9w<N3+R7g1?L8#k>E}d%5;>4iYa7Ee@t-`M#
z|AP1B#C`uj(*kVcmim1U7CCUsF(3Hq?llvcSrbhgk6KwSCz{^#t8Y8EukZNg)z5kT
z@&(J$`0oB&&X?0wt1sjM>W=3~hDIO&lM*tNdf{x!`(AnZYGFz@gsvzyMm#~n<b|4;
zqh9nY=g1JY+WGc3-!a{F?$QnKTj3mFVIo<kns@WN7$nymnU@2{WeC8=MCeDC{OQIN
z5;+I%Zr%B{T`tsGc;AkwMvR4K6Pre+<a?-&h5@HaYa5sPNS98R4#VVOH%%~^X_k4c
z5L;mCfhrM#3-ot2ngkl1riowy!+5NX`{Ka1nsR9~guu+e5o`!Nn-|>-X7HX!hfxa4
z=LTR6Pk7fy8Bmtb9nr>_Z02W&YmYmd$(k1#N*KV0gS$ckHB2kzywZZA6X+Bvb0XRg
z7t6a?au_assy-@@IeYt)SPn`Jf)34`N~dGgOQN+u7Rv&fKpR5jz)Oz#$k(@T{(iBX
zoUJiQCXom><OTvT*^efcW6A^*2NSxHLgF^de^D$aamFfTG&>$Et#7Q=&1v)ph~=gc
zW#UpZ3xPQfppJ3gKHHlyyEe|`xgP6h63bV2Yj!YpXgitbn_Zz3T?PWZHZP@)qCZAI
z|LBC_<|rzlowBX;qDVKi2iAuZ1bF@WC8u)XaXoWtXKGWX!-2d^oVS&3jjgUE9CUnm
zJh^_II@47cBW>1^W)6;rKpt9j?wIF#rOqsN@OGIJDZ<0!MsJ;mRj}^XI3~~#&X>ws
zJHPz?9S`NgIoyx$GaRE)qz9r*ya~VH5NBRaM{*Ead!$=MM9~WU^2oQpi5y?t^5UDf
zyngYLy!<Ka>7IJNkOvkPaw1wF+eDH?lD%04u1^&x+Zd7DHjYz5j}Z$EtYe`eAz;bh
zv#E~)MuEd&M#<#t0PXva_sj?3=H+Y8VbU;t7|*w}XZI6H216_mLPii$2~X@N^I_=2
zd?2U9+8VW19&00nMBs?9U^1b@+{T6$UO87<D{S?Gg$}WgvoN_9p5kD%MUad`3~|87
z>f`C%#~D+M%+o~INK;{JjZ3NY9<cz;CU1z120T0g&_Q->*<r_ET${rFvp$bF{<8jl
z>kEMOW&L2wuK)8SUxHj&KWfmgg6NK*tN8~#AkpdQx{GcaF6OR{-Yen_15c^NXzNB2
z`0B+gB8IlYp)?m^9y&LB7ocrz2q+r0H(FQL%>k#o;}OI=4?0HNB<y&;cS%AV?@M7<
zn2ab$AmoG$?HdQ@d^1&5E~U`4V}0ecZJf4u?p!z+U(PEYN03HRn7E0z{*iCkwzGGo
zd!tqX>BPk_`#C8_ZsN>wUO42Ls}55VQV+~Il1m_UBZ?O&RTPs>5+CpQxOTR73)&GI
z10|z#B`G09M2F37H!OJfe(T28l+rt`6#Ds0(Z<mE#YAWn8qGUCW0WE6<AII~`uLuC
z5fTZc;Et0iMUFXR7?(IfhgDp=rZr(xC7VGDNC&xewpKaSGb9&!#k}vV<|79CSmV}(
zsu8tBC(=hJ)6D2$NaK!CQtisQ6i(ZQX#_H2lP4{@3sml6^-rR+GBj8AviB+X{JgqE
zd(rF5L7iQ9L)G;lv|lB?pGH=OkU!?lUk3yj`YEmtlYoHIw@Qe@i`!eadf{|&-2}{t
zQzT9ve5##ZA({|OsI5CqUpt1o0fmszk>v&g5?m}a1ix3|+Rbwz{$NTB(eIu3NKGH_
zH<&v6%9L7Q+myfg_*)*@nblUXiub(ZptPI2^~jOibjw$VH_#NVjUtt5mC!1NaQ|S$
zSrPjM^E{Idw;bk$DP^}PFe66sy=SM)k1-&?LGR$&W5qgWE%e?9;%;LS3>^jx5{FE)
z?e+rC3n8ejbN_f|iVM1~T(rBGHxaiOAN!d+nL^||xmazALtkgQj!XxM#FV(14=xJa
zUE2PfCTb6~o-vD50(I|RH}mn}wNhH+u@odKl8j&(u{hW#0^)J&ntOZlJR)tdi<IHY
zrNK*lgH@Hace?s`L{lGkl!3nPW4>JDScaayYmRvY$md4D`UH%I0d?5uva7k9sDW#%
za4r5VI%~r;_#VFVXC;6trk!RLvF@w{)Ww^B;8^zaV~=I?;@EsV{`X>eCytpnzZzb!
z{?S+-<L@#|6K1*LCf@zMV>v`2_CS&}^rFI*2_%C}^^YFQIT1pFVbEgMh`~j^|I1=|
z$UzKRU%iZ2ZW|#Q@xOX3E2?p5sbv?;zFrK5IQw%}^Z=s=ptUnP_OE|V%kyI!tRHvd
zR5b^zFoqt#|0xDUcU{5(>IFQnQyZXB#jMiCsC`=tan2lW7Y@t9VV;O#1&o{>=xnXg
z&Bx?|xGvqc6<Vz<(d96F#0<2lvw`+zfP^94@l55&o?wWa!jPjFAZD%3`WZP$H{-lD
zPMh+phwr)cjT$%4SBDvbnL{C&nY-|cW1M)IGhJOv3#$$ScY|8EfBY?$8h7@FWuBRi
zuQ*N%Nh15>N(@Ae#J=;lBUcKa_?@Y>A-xepaZ?GJ|D3@JGdzzr*_#W)r91gpT?m0*
z8|U){;*q0_x#93Ms<?A%pJp2Tx~4IP>&>pwaI}7%zc{4KGS8%GrAlBmk3qqwNsNe!
z3&~saX@~97DNQ+*PRi;Zak^^Js(H++dGnva%7P)mpaXoDJ^%NZS98Oy+SO}7gKlX6
zQ*wRM_qp|FpBW7&A<BN?uhvOl)<5y}1;F~Uez0{-t8CY`dIoCw*>e?JubP-f7qaVw
zwgWNVAR!S|`~Yh+n4&1c<eDp1p~H4wnU=)uivx=sSkf)GQPH#TtF;a-DPhVPmIGZY
ztykLPh1ME5FXRy1VJr=HjF`(R*hrFt|0*d$SFAVJ5>Ocez&5(`;W}EVqaZK|Y{02i
z{_6YxgNrt{aB-&(@Pp4H(qM`ce{%O{OfqvIBdWC1BN7^E=}4%UD3|kv$B)XpZ{Luo
zNScJNUw_3iXa4%zzoGVu^hmP6A<j;>CZgSwd3WS<LDa1%>t<Yj@&=VmXJrgbAR!WC
z_QF}?aQ{h^1{LMwR(ZS>j6|ufh0?n)<pndLo6@vXt+KVV3(khX&GD8O$0JKVkivvi
z!Kz|KiNUSe4^!qaWnSJL0VrGH@zPmaM}kqUQd^`Te7t{8SqtZP4|MGm8X-nPO1ypX
zio>!X(UV*Wa48%2@4mxgBF!%_Dod=SF$Ymfp%2X=*XRg*c-Z*<<C$s^l|-%19%Hyv
z3DI3?#Au;AMV|rs$=)d%px8fbIQ~~z65I|ub#b>XTyMkhWMdnnlokc;36q%+ZpXOh
z(3G--nX?gnL^L7K;y5jt!;2#??(R@geswCGdgaB7TMpAqzih}9$;-s<R#d9es*;kk
zfLc*X?5wpQL0Aq4Akrid^5V&k)~EMnooI%I1R}Jq)UFh3By-@g02Sdk_wRT-cfMO6
z*`z`$xOfCtR?5rxf<rv=vlo9xGNC<exCUNBq>yvNq*29q|FB{cy!!Zo5E~L3H}jFN
zU%%n8oKRD;1k%uGn>3MZXbc+k&^*2#3psRZs;s49BwAN@dD@HqY4V~`BQZGe?jtA0
z!?|#ODO{>@c{p?8x74jNPZJ{EAuhF2w6L)eW57~knHOHYd`*%GROrnC?$8q<K}f<e
zW$yA!eEEXX3a4IJt8l3SB7;@onjjA!?kQ{G^#0N9+goFr1M?htdGi(Xw9umwtP>^C
zI=nwua)Ko%vblJybRRj_+OgIhaP6Q?IG4()6tD>)V%_(2L7!<nIRp5~A?+!S_g?V6
z&wn;6$P>fN`!hq+(BFff*Zxeb=0d+IK<kkvnI+y5BoSB%772X@Hp~h!CPK(W6nP%Y
z$D7YGmUoTDm-!WkbmXUpKO^xamj6p)xvS8C`wt%}TjBEYLhYTkH|9C8`~%1GwFP-S
zgX8ag?ODh5vs(L^9Cg>ZayE{RBztXtJdP8a=eIrLT7Z=hBQZFOE*R9UVq(M|SW@EE
z%^m681x&d+9=SUnNOeQhNaAALwHdXE=Yh+@lxKRwMZE_jJ?A{YjX+M>>x16>mY9)-
zHRi<}ZnmJb-ng94#AKw|xb&6Ml)rgjcw9Rt?KJ5FcvVabk%>s+tLf()<1KHGUlX)b
zFYl0`q-cbw&QyzvN#px>zoALz<@U8d_t9A9g%@{s^wB*}ITKnWw%yHpjLo7L-bDTG
z-NurPyKKCDeH?o80^&I(=5izwgpdL~F&uoK&X`g=zy8fTZfc?*FFck?QDqu{W$T5a
zg|%(u5M8S^&)h7xM2*m)RX|~~?7CrDcyl~*Tn>~7TiH0j`#`ryh#5s#dmE00#=GCV
z^W6HO(z;Q)u^a-2W8}r{8>T5E>KriDNNItG&Fi)lCrWQbQJN)h8a2EbLO><aWkO@Z
zVn+0Uh+@oRjIjB0jXM?!dmIn+mGJC(@$A<x>!14i0$_bvKj8YbJ$=<9c-H9JUFpo8
zM3SAM5y9Lo&(p{!%U@JR2PtCG1`y`mV^u|BAkUdRWm0SGqT0KQj|25G*aFer4O1pX
z7ljRDzWaC7Lj3Q>QP}Y0Dkg)%3S7ml&A9%mW^`57*11&WVY^UhSn5wL1pNEXapn+@
z+@u#|BBANE!yS@BG@=i3fOUG?I6vHTxH%Bwfq7a`8Zky%Z`aoz_d@L2tKAD08*};I
z{po6>bF%+axps7}-OFo76q+iRbz>`)7!utAtv1Zkpyx9f4P2@2VtzkkOw6;}3`e31
zNvgXRdUsAp;uP~aI<0t`5>pdW^e)bRoqLDNy0Wc>(yD8{glEa9;0&qpehdfI9&6`Z
zDwl4|&Bt3Ae$jnU*msdsHM|h&HdqKdTWf!|<ZA2u^m7NB<NA!qQ^EV0<jVH^e>;_1
zb{5BYn#Oxkh6aj(TdjxD35^JCe2cJG$_zp{4A8fCL&kVw3W;S}NXg@Fo?N)s$1<IQ
z-a9@5<2RU;5;5&vS08aigcygRfOzreMch6DHahZSZ;<g!?ZU>c4-;r+6je^8@K_o}
zH&2#Luh%L{CdtfAe#tT2u*3!FO6zApuVTyKYzjm%Raz~aPWMc6WSRnVo;V(lT-pl#
zNysz~@oew9KgF_{dS^N+IqXhpBi_X+_l(ZCvr3$QAOuW1s>W6-TkRvcHnz2(UHxYc
zK8~6WJsR_CXy&6_c|y8VF3p83h~{@Tig@!N2q`5(FmehGSoP6SHXozt80X7MS+}8+
zv+vm+yI}}Cdk5RuJiRMrE3B<kyHE#HCFpPw)m=|xXd8~RqN{;;3S=XWuL$$AM!H>p
zw}VzA0QR`}tXLc49ofL*;AwpAYV_eByKKx8gL1^O#le~$3;=&LAck=^I;*X#5kBKs
z4kV&X+~k+sq#Ne25LD5=@+Fr4&ttipA*vqBTcNdnWwZXH#PYSi-Ot{qhSDbi|Cz+{
zA3BqRKNjQba{coD_WxiUk0(d2G%bVZvE_=7gPGZtRy~)Vvp0_c7cU=<fdOansSc5N
z3EqHCi4fz+lROvnJ!dyB2=31JEZc@F*@o-Akv%qw$AJl3udKcEST;_z(`>(mPfVOF
z6H_>FmtJs4H)Kgbr<t>h2D#ZE!I9>6?bKFSw-b|PvUpr9^TOITj|(xv^XmuKib3(c
zVBH)9I>vNuR*zHU%J7xo`LvAR?^=F_8yEEE6C$z}mUX3$eeep(nquZs2r8aOxv`PU
zT42s*6vd@Qa4>U<i6tfC?4WiUEZf}~!s`};5y#uQ(Yp*LEZ4d~WZ(CFjMZ?Xq|%J4
zLRa4(41AC8d+hqY*41b_Kvx;yGS0HJup4Lals8{P^f_qt^qBZH`TrxBzO2u7eF3n(
ztbdT@^Vf<9L!i(%tfz9{?P8s=B*+*<2xG3<x70QO3)jiAkX(c{C=k%n=^q|p7H)3t
zn0s@*r|^KOa#~Mpbw$NHJ~?HkJQLH51tW?0h8_aVobuaxLp6aU=6rN^i4}L-Q7^`W
zh&%GAvYiX1R<yX(vvt_E#%103_WLs*wno(npOJ6uu5?i*QkeMXul_Z+`6UOrqnDM(
z%eSPc<Om^or|1y850OGf`^w+^=YP*PKmP@HH-EzI;f86B%xQ93_egr#XW7FOXc&I)
z9)W5=H9~NwLex9;-dEopNdzUbRoC_U^>@ETRk*u-#c8XYP8Wnglt>sBzhM+h+8h|v
z(#dJ&#qCR$un;xSi;e_8kuWA;Au=N@*|Ut}aiXD2mx(C{5&_djqw)UZM`~^S<mGEp
z2*f1AJ#ZrC05L#xpscHp0{;5(M817RUMrYTB4q6YQgI6PE{58>(n6#rcQ;fM=+0<!
z@TEg{_SkcU_TPcDfNQPzSuA5$u<~kEMi((N#FKd<uqz%5VumFdRPlY_ExD%3b9^tj
zRzi=4MFS$e813T;qHr_cc&9T(Uf$htm?qM>kf({8;~m>pS+~kO2d1nXCYX~an;{6b
zHDVAqYq0G4Ig$`jBt&=B9Q{)TtahjJkK(9m^RZKl+@A}NTjjUwg;VJi3Tis+e>@St
z%Wt_$Z~5o1{(F|VaR2dpR4YlIM5{v#E?k;?>@vj#>y^LytAESaUwy+jU;m8PFJE)G
zS$MphSo?sKcQT(E)w*k{?Kp}VrEEx!Fi)r@nhFwq?1$>s+BrlQ>ordxdxy4lw%YjZ
zKblGe00961Nkl<Z{YTc?IUMg;JDj!+nvhfo);zHd9W`|;ZWNJ5iiyK~L@!<p(`=Yt
zKvE$aECd#W!(pas;4sgqz-qG>LoL!;q4&aXzWa_^Dqr8dA&+~Rr^qxXh|_Rm^Cp7!
zjdL-+Kb?u`z~NwMt4t{QnLShM>hHHf4@R><mjq!3iL|!Tn~#H=g~0;2wpacr*8F*t
zyU*9&JsP6D7$F9d1SB-*UbgPA@~B_D><K=a7sO3QynJxAZj-^RB+&wB7_nSG-&np)
zuenLD_~)<wFD&7}#}DsNYfN8a`EQNot)5ux%D3PD7H!HeUjNjE_;V!Z$TY=&*jT<s
zu!G^s?mzo1GCub+IfNmG-UyumKl%DYaU4a6WS;lj41ts|0-JV%xa@*#%32;EMPA&z
zV9LK4I(b9DyLTkOoF{67s>b1Xg8<7iG3D$a=>XhhG=p6`YB#v?*mFQOUdQ^!R&~us
z+vSX^U?KAHeBpj8JY0;myBo5Y5W65M1dOi_KjY=}4L?16LniX^;d?N5EDuqM!A2t+
zERSM?fNJGe-~I>QyiA<~cO=lZtvl$BZQHghwmP<LTOFGn+qP}nwylafd^zXdJI4D7
zHLCWiz1N)co39ZqHGcgS9VvylFH>$k#fflD$8Uv~0`m*2P4jchoHZ1OoC6M6cU9G8
zB_Hds2o9|0bcoU<;O4St&{{0hR(qXn^au#N@2^_;EkctE_6$QVyiIx_Lf_)DQ~H^f
zct!#%F?5;$y8)=J;UPJNwMMG+?h8OcGdJwULg=tQ`*dU64$%3`=?i4BBa|@KMP5jN
z^Z<c+uzcvNzJPw#qKH{HDX;e1RILfG6uc0!vtPqTC~~4onnyDb8+LDjvPJc_arFgt
zaO9f-^aQ?4yw=oq|4`RkMret;*gJ(E;$b;+i+nwyI9<SfcLDg?F~hlqS9EOI>oH~O
z(zB0h2vZofL_OUc=%7JuTgLBZZPLb@=jIo{9ApJ7zv%~FWtn$G@Ry`&8wq$@!)g<h
zRm{J#;Kk3G*lQV;V#>Oj&MKJ+BH9+G;NDB?%IP^piJxoMZl~4<f=fD^t3?kQ{qWx|
zw@6j>tZ=??RuYmJOtr}4cUd-hSR5?tmTzS8-_=ysNph(Wh8wSPU5Z*iN3_5t!78up
zs(Tt&#o5*2`MDwp$&OpBEj_`=z!)$QTgJpHvtL`+5O30^B*hwb@cIT>?q+q8n5IM#
z3{dx=uZ^ObNJw)$G|ld$p%jupA(B_O->r;&J9ItVPGjhDq^5z1M+b`(NEC>v&YKT}
z7Xv|6iojyegaLypnAgOUoc;8CqI$Q&i7!!db>0S~u+N2!QS7gxkeKq0^&re&S&!x2
z?mO@{>RNxjR>#+l+}Z9}tE=Rb8b;(^CpWrzxse}zY8PUz4SAf5(Y48XCk~hMQxFXj
z(Z3a{guvRsg{NLIMe&o^V7RH>fS2RQC!Tvc*zhh?!;piI1#voXbR9iDD8cVZ)C*wZ
zNWdE;%F6o;P~tmLkt!!v*$6XY6YPbJHCfv4Y27S`7-U?{+-yJpjN8*pEWsk3ZP#*S
zS*5fLc>?r4?`F-|CSJd~q}UX=k_`T2lJD(v0rNpPyT+%U27G7d=XVPbs_lr?p5$up
z;Kgu5F~uCSN_8Z_C^^tvs}JgBWT+gx)Fp`G70HhUB=*L*wVw?~UHF#sGo_E}l*6*P
zF#onLScdtd2jhEJSXi>MVE-~$p~HG>O}{{d>q;*9K2Y7alDHdYs3VMqYZ)S=3oiwi
zmB9)m-pDf~x<lE4bMlv(NDg9)zT_DVKQ<NMOFtT-jxc~~os;|qdlJj1X(0kJ{~0rp
zCVupZ2L>VV96*@}B+2Vi?1cBg;d>zFW6q=ejm&X+k7wX&&r@b{KZQeQ!_(o~N;PU>
zp1~n3-%6_Rq{btcS0G~M-3pE&9#MkgeJ9NekQ9@OLI~CwXj%-*r?}Qs)32|eM&sfU
zL!cciz0{Xktmf91`qvH#QuZzrXXfLXj#lH_os3GMD^uga)kEaWXd_&+yfx;w_j0F{
zwS*}ijTE)a-pKLWD4iT0_lKV;87F6X`F3gtD6J8g5BdbvjKu96Z`(QNKi?x6DTvML
z1!CwfoD|Jr$hz5#`DwFP{r)S&%kI#2nLzEDzl0BY6l3~^S16DxUDDs3HVp|zj&Kz^
z@r|(-APVr}u?GjdTw(acR1iB0q9HuHO1K<Xywx<)>6~)#m!h4%w-fh+x0=#Ncl%y-
z{2dON1Y2pg#jOK-m|lLqjM3}!uFXdnq!&G9|GQE`tM>Y5pO4fcWczWp8|*XT0Ajog
zjomh%KtaEXiC`R7egi>=H9+-@p5=|*4x_>jhYD96NGh}v|6@dAJl<lw8tSL;pJk50
zm^m%7()O!Lc!Qx+*B&R=(7|pt<0$0jcE$9K%@$L0jFxg02tc}^w9w99p1@o$B*d5V
zmz}$%tDL)rvlG(b9?xexq_x(Lv|Kgwb4lSpn2Cf^GVqmDAy56y*_3q(XNR0jCPl-4
zS%&%n>vQ&cW$}qoNNtq<nIuhA-*}tkwge?Jq5{jz78Z5T32n_2W$b5{gF1BjDijfO
zaCu5CgYmzSI*LeElyD-Fw&#aiy}RtEs0A29<$<xjJ3799nS9$YI9e8jYNBzQK_WyZ
zRilR0?DP<2-KRiU@~B4><1^Qsi_+?wq{`?SZ`e}ot7&d#2+WUqaZ%iWLNWeEnT!_Z
z-t{MZ&(lS2v}UBn`M5s{Dq{z!{uT#a_tMdyc2kIJ_u;%~i=q9o3b?42=N=|CS-g=a
z5!QZ-y`$o4mPn+?#cU9?hSIf%HIpatsMp?#rY&uM5(zWM!0<?xNE9FtgRf!}Ay#@)
zz!Tvl$C@#GjxJVp)RpisfJTu7Rajr;M@APQ&Fh+#DsaU>rYlmg8fcFZ!InDHUKVN+
zgQ?y+>8u|`wN2fp@pT0<yd2+_95(0skVy-Ra2-C7YVx!9(1xSxz!WN+HX~aEZXrPx
z&OyPI+Ug@oVp(a|ymWiDSGl>qT^`^K4z8E`@mpQ92KTEn5}T$~(O4<bS2<NDy)9so
z-JW0n%ulKwy6XG;r8Q7G&?B3$bH2S8sT0`F=B$+eY95B*A-<3K9nAQ<F9aQ*@$XN%
z1NGR%R9+e#NcZPCT3=AhS#m|2I4f&A%mI9Yn+l|m+L1%o?Vh)(C$EzRs1tI`T22=i
z#bRQ5Uij~h`ttfT!Aad-^C6itoe`@Qkg(B5STCtKx7)|FS3mj-><LcDdJc~Fs-M}W
zpgDk>riwsi14b(nS2Ukr42Q-9c=e}*3M&0!Io~TFyDu_+u6HO4o&p;ZX^IHPcjO5_
zD_U`FwX2L8J(TcHIF2w8RuNWA><0qY<bxk@u981qO5Dkv$O>O1FLS)A5g#mYV-@=6
zIB^5qS2)>a;i=gPA-N#A$nSYsMg~d!b!q6J=V_FnCxNiAc$G*5Ty0;Db5e|=XVnE9
z=_p*Dx{>e%vRJ*p)hrd7zjvHs71zbBo7=6X#7V_M7P@?wAGP}Nk`%TJG`drIcMTc*
zM@R1{*M4s)){p(H75p>k)oPN%+4x4TDYKcUs8T)xBb09j8WIgW=Pqo=m~-$d8@KB~
zZF7NnUmlS#3qy_TFpEA~qIn(Zl*1pY&iZ;a$W*yg=1sso?|o+~b(zFv4gT!~n?HNP
zqq<WdFNj;5<$8}QU{%hq>tlS{;;~&%t2tgyuG)z39{1#pa=EkRf!<Vo9Q+=_jlEWF
zUBh*v1Nrk}+By{Gpp{*YUw8hLEL=PG`jd;i8}jK`MFSuI`tA6uE2|I=NI9NMn`ZMH
z%)1o&D0WkHa?Y|MDRVX)zHRk<EC1v1N_NAp@oT@PQ@cI?8=(KC<?bf*Bsav;h5J(^
z4}u^V1#&=^Q3MiN&?ag{)bngnnh5;ng;*@=yE5LNN_OyYxzG2B0wVd9Fl8e3hQgj_
z$37~JiW@j46GAR(e{l`q7Lm1>;>nw)by!L4K7r8b)c;nPEKTKLLcIsnfbt$0e$W{T
z88EHTQi~<k6)WfgYGC?v{#jf<8)*=ppMx@j$PBV6Jb!;%C)FvKMUKQ%tZ;LDaej4B
z4a=-yKR=sYRu`aLZr!fqbis6q;xhz0Rq<AT&;PN1VTd~3EgSal2DS3<oHbp)%hVxL
z-9fiJ-L5K$C6eN4c?zcUa|KWH!2@k7e5*3tO7cZkeup82`y+x)phA(KvYTEGd3V~y
zoki<>4OFX6g$vN&AjA@cF(9Y2up=idCpNP+hr)1)8A3X525=WbL1pzutw~iSad~k>
zSYfd87P#DW$+g9>decvBm!vQAq|hT3Z|4~uevs6yvV_Sm9z3bD6QIO>5Lh^u%)!qK
z<Xeq{>k~5$NyvqBnI5O6DH@JLWl`KGj#2H)f72uFKkkgIuyo|W7bPT9g)jxyr@?XS
zD!Gx5#VpPV3q_JUVPz5BjFBZ%7wa-Sh`B<+817d(0q+4AXLK<C@xhsY#4`??xjmUf
z_VGmZbxgu$8-=9_a|?0EAv`~fr4cb_dO|{31+aIN1Bbd0tj4pa%RIzio!f7(b#{+)
z4>>n`CtZM(pk3nl3^*r>tv%YUi=adu7l|H3Yg!~n;fQ>)zppB7j2fVk>lj4>u`UPq
zMQ`Tv!M=!v353mM*Q5@d*FsQ=8dMc!BpBN<(Ls@eg*736X~I>$g(*K@SBb`6Xi^s`
zGx&~-sB*|1MGc3Ci=dH&CuoYdh;i8*k|CDJV7^LGLxXbB2^i_96??ea;1B!Kyxm-_
zN3~7+r+KqGz!ElNSWm|_5eDutuFlE3+&09Pv~I6~0RT5AX@5O4HI5Jh?I1x`QpR#D
zXnSPr)GtT>-D`RB#0R!N9H*#%Y8iRN#N#3>-TLit;5}QM`aFI=wb)B8VIA!J05-8Y
zP+z3e#yGOu6jh3=z-xZgk#7?}-v~O8CGDCSMDR9qHM6WGXX|^VMi10i!r~YKBjV$-
zBiU~^3B>EsQ^isGz8594G7U*22(&Wq>ljIPczQc~!?s~nd0cZB1yk}ssxm^W8Ng1E
zYPfTC9UVQe>CyG<VqS+qe_JPgzdw?9{GKCypVvv>qn_LIDWN|p)<u?0S1+;i>uhsV
z1#W0M3&fB&|JDu#B$s>`_;OV}nw|Zd9L4BtHVHNRpAS`@CRa-psViPy-BnA9qu^oz
zvp;q4_n1C1I?m}HMw<TXiSyh=%9Yg8oSeG^pvc}cR80;VFd}Yh@W`7M%J~@<dEIuA
z9R8xg-CS?I&r~3w*sf5~Vk;4rCD^>pFriodjn&p*Ke9;wdEHctuL=dQB9V)p>t2xd
zo>E``s*#%Q1w<UM&6?Bdc+oZg{g?CO?0duf{=fX3^=NN@BGqvU$@@o`7mTOu8tood
z__YsU$E@+~K31m^QUJVVB~0kAK+@`Pm6daN+Bt;tIM4}9+3;>@ga~hMYy;s?n<gvT
zExg#xLo8?DN-2o~JBGwn;QnUDy!NlL4kq^Fg{!gGv}#67<2mL~_mPYPwa%=bJ)2jk
zlTbaS3n}<)%pjWFr@}RTqPIP)L`<-S#8Vv)_tuj*J4pw_bPCVY(=n=6hn=OXGN3Ip
z?r)Ldy(G#a2R}dJ6Heb_`;>}c&1jpz3Cl_qt|}a@i*I-=B4rc853BEj1qG;r2ce{h
zq1^}9sO4{<Um<ch8{%S>yKG3ZB5u#RM5)RdpTSnyw%QLG&OdB3YY4aJ`;B-dxmGu;
zOK+=5>DH5V<l&3$3i4TJPiT<JS0aJBJSJ9|ax~wTz>`C47$zho*N_ldb%e$B%{7|Q
zM}5ER{x&e~mG~61S5*F26Zr_7H~mBUY##9CRzr0^{VOOs9=P{*B=^rNfkukv1fq19
zpC^c8RLn(#$Sg_^J1jGWnWJMcC?vBRf6D@TFmVn*?%M4N32n!=hC8shT|2Z&p8yfM
zu^RECM0m51aBTs_>|zu{+J_!cEuER&`(`8jM<(7=gZ*z4BuzfX?RQ<q`fd0)@l3w8
z3pZd)#>bMi)w}>s#41ThL5VRJuSKyEk0@B!rt><tbC=jESr6VCk+U!vTBpH?Xs=~H
z4cq>+`Bv-bx2t|l9AuX7R3;6ZOov;XQNZ?ZrDfgK(zpT)a#LsR#<s)|^rO*!`g=IQ
z^t|_HGSEXbAB?{~_UspMp9RwKXMSGG@lX`DDLgMDbe1M{Q{oPlCrrtejC3|=%kE_4
zDvyuNqSGEd9d1g{poj!iArXk8CL&*h4Gxj0AwK%{OgiPt_UjKRck;rwHd-%oMfw5*
zoAZ<8bEUgq@}f5=N}|<MMO{!%CPFuwY_@*vpTS6AGRS}lbzdU8)W_b?iSTtm{pQ;D
zq+y6Uic7eN0?>G?yqS}=bx-pf(_R48LnsgfPG%lX{K)nCcKB|(LDhQ*M&3bwzziI?
z|C!GR!y87#a1)YIzaMnS{s4lG9DtC{dT`$a+T+o66yF_~6(VXBhevS)gTV^pV!fJ1
zxDfbV7?tO7kboPcn|l5ZU^V1qsq7cmHdW-jb@ZIb_*df1_eDkZ$Gqe@dRZ+1S0_ez
z1hNJb`2h1r`~^Pc=#Sb&8y=K{ppcKYx(ca){b+x{4}mXyqfPf8>M2dXrAk<hB7D^f
z1H%lu^wP?{r^^|<z|!=AMNt})yj$aPooZ(ty}|or%>3CyMRK-gp=(jIoXdApIjSTP
zH&8nBr<x@MmBb}w_lW1h_6UUv;qW){o=RjTM=K*Xnz)s7$zFQ|tE`K|4!_jVGRveZ
zE*XZ+#4!xFUK@lKmX&-Ig$XBNT_22mGLgMlZBd67Qtjz*n{+s5?5>f2qzaIsu*-09
z)AT>vn&#|CiI`&Ib^^<l&sn8<nroE9UN90Q<lwzz$Da+qL{+B*ar0k8tfb(wtgqo6
z-JB-G3WvlEk&`omS;;JGXea)&Dh?fiX=>iHJ`A-XjRRXnw2B~N62$=JV3A^3r>E?>
z`8(h_b+%Uhz94I1S=r>ETVWNTjyd)~*VFojqKl_{_jFC8dkSO7Hl5;@_!`gX(T^oj
zAYgv5e)Z$WEKk53wz5iut78UKG8evW9BzEWLZL9C1k_j>fE(Q~;u4#@fW)V*^5JT*
zVnvU{;P~JJtVl)fO6Eq0#@{h!QM%AYepo506xsyH<%%Xq7C^el-cBWo!-iRi)o5tl
z+0!$k?2(~c$B42Rd{cs$<)0J)bF(~rl!s!JD_P81(jAHkNi{d9zT{Fm7}d@uLgEiq
zD)0<0I^4IGMOLB{UNzTVmz4B25{nwOG2kx}qR5CczelE%2m?*DCwm!2mB>6O#N%5U
zdJ8@d{~K-Xky<QNjW?}%<}gmhg+tm4eTMxQRm|2Q`h*Yc>FXJ3DvLtApW;_OMyLG2
zkO<x-Z;FUj7fIFK(gnxi0Nu!dkem?b<Y1!2-@q`$v$4qSLk3JuJjVS~*L22`$%Ka3
z6!j8yEJC0Xu9cc62cjE48qScR(sGkMAj^kUlvuj^$kwwjr{~2<H^j~zr<({FN$bqD
z_$+u48Jy&r+cHYd(3x?NMnj%ZOT|;~EU)$uC*RoC)^3aqcx&$pDu~z)Ih9DOt9}_3
zjp!`{6;yA-yFNaFVaG&_Z~<1uE=BWT+elqWVbB3tV#*p_$Z7;muyh(H0S>Gy!A>;H
z3YZ5cC2w>~X9EDT|G9h$TmvF5S4Ku;QCCXhx0zU`%>)A?7hER>Q>(F^;Dj5ET)<Su
z*S*1$z$7!7&hsaI!_=RVh}lSBw3ktibWVc3r@T7_CB8#A%ti(QWKMMwZPxUdS~z^%
zkQGdGTDd7wnu{(*&m|g9<Tob}>L1thW|ihj@0*8EQNZU*vk;7*5%aYdWt$@_&-I+3
zADR!?vlWJdKDlDIPPZ&kNBT(CL+VV>%g0Q`i`|3wJLnVnsOi!?&PE&UBSx8$``Yb(
zqv|opnHHw<2%|UI-8pY*-bdOxeNQJ-Mf}JkzgDNojTu+`_+!4$WLzsvjcykquBol9
znd6hfF7Fhst?g>F990%C8H8wvX&5ZK<7Lz@M?=YYQlXnH^!K%$pC@lW4n5s|b4PHn
zALa+aT1crQG}2?H5r(HMsCB08-0yrBKI1;Hg+3)4-jrpAS@il?mSP&1@3<VC4^lz)
z<=UXZM1+Kl1zV0!(j}lwI0}f&x(b2|L5FQ*IEvojF_9HPk<?bxluV2tcpP)K*>|e;
znAJ6b=>NoLpgEZsE@NF@wm2qnL2j^XfW8(-oxPr{Xu@E(Qit6BPjaj4n31FZ+-Fgp
zNINHW8QU;=A<o}iM=+DkG%pWR6txg(J%nv#1dW)Bj#MR|(tvfvOGQ&b)5^LDXip^*
z3Z~Ri1Vb9tghrsZNlgV#pqB~OV6xv%)Pxie=<D=5AXB|X{1_3E%V9~UD;)$*UG@Fn
zJxc%c{Qt*<Uguf9O}qx&XkSk08bxJXef3*Ua>gbj*cQ<t+k5RO@K!`d2h13DNc?+g
zF`ZnPjD~Wm3WD?)rzbDVLIhjjMWUx^?2B@Uh+2poIMp_6UD!QE&Y1Ije0DU6cJSFU
zaBd0+GO)OSu)oSqgeeftfbZ1x?XMV+)@868yNF^fXk^j)=CtvUc`~H;Qx@}%89^Gm
za)Xz4ajg~vY9x{~Uqc%1)6qi#KAYXj-BPNu&d8C?y9`_&J-oxMo9dI2MqI@*a@l6b
zMJlk8!uz^H?f_Yd4HzPO(2}0(hn|EIBjEHK#{%E@VJhwd3V0tf6gPQgrQb#95Hwz>
z?cAK?#6hQ9m!=_b4ps9wbIeMqy$@@B<iagOD8x!Dnu^?2F3t6ZTCQHth?`U)BI77J
zUBcFVqCq~3CgxeuK*z&?6=b&)FTS(W%b8Z~nzxhMhxB{i5>anxA>f$WdxQMfl@U`E
zy#;0+otiL&N~}v{b^MWJ<-FI4J16miKNW==tW-QFz_|73a!7o(B!_-Q?U2wY@ibUn
zreKYJMvPZHp_}Bu;hm?X)P1u>uIwi*GAUyTSjz%TH}_`{j?Yxx*1S!@T8M%|MXA`%
zqexA#KR=2q8Us5XIRsicYXoK4RM2;-g9Uka&tPWdU8jtXVE#c*ILs(fax<ozJyx8!
zNJWaS@<_-^jP7@{OjUgOH>89pP*7<|xtbh%cPflJi~bupTYw>In%NTlwh*4;Ns7Ey
zqHZm7<-D;~xCpW%D8hF4p*Zq+P=I;<!|j<9{_^|F)7tenEnwN0s@^(Shpuobeg!Q<
z(V#{r99qkX&op0Nl`41wL~**(^@gS6D3^ZP7RKy+<#8sT%)H0?@xcLK>aElPX>!^-
z6@LCZu_m!P!`sam<8Q!9qspE9h9&1Hb|2fi%~kVC!l^u*jE~Qa|GurT%qjD22JiLV
z2WhY;`ITvJ_^-a{1Ak6;c>aD(2%izSeB`mLW%o>6lzh44>M5LjC2%J9@#0+$d!)BX
zhFdriy1HTJ9EUG+P)1{pjA}k_i4JOLXJfUuNk@d^f=~W-9s60$_1<)5#)&ffhaG%z
zq%|j<rRLiOP>9hvTI1319;9p={wUG(>>M$HGRSAgL4cL`0d)((q+y?8X_ZPz=bR|i
z@a+~$-u)3vWu<kOZ6dHJ&Q%j488sHNyX$eY#a1arCLTMm{{A|AE%dMJ*G!Amfg+6-
zj!el;nHM<6-*X)WrF758IGRPsrZ!%2MvU_3^mA;N?#Ngs0Nraiq9?3Bptm;016OP4
zY18wM0{c0l*vxW!-WHpcRz+gH;`#70^q>TEz|-{))f*YuhN>S#V=g^^cOCJ7uz6~h
zSEMY+_iqZXJRme)#GfV~9CiPzLshR6LGR0vSY|DwZn!CV016Noag8RgJlsUKhBj<Y
zX~^6XSc=p|4I#owqp&#bFqUI6u64Wjjvr|N$!JZ}jmEC&4Fvj#Co<4N6=?0q5pyID
zl&nCk_bNzHVmA&3#*4~}@9aZ0z2|vuW-DSor5NcrUIwZ%=%|#M1Ui@0{zXU59r;&~
zAJZ3&sYdDw7<TSdbaZUyBUoQQ8>c8SGBO=D-ye3eT+$2Y3hf7>zfvWRz(<@I2jY-4
z%qZG1(X**54>JEI0|r1`3ZuaH<mV)iU@9>7l*dI7hxC8s5T2}W)nVYOx9?4Lzr?YD
ztcW0salLKx07dwiEXTm`ft_8~F9eI#GN{UGiDVA{VbmEgr$e1(oMoQ9ncr$p570$V
zh`dwEzo60(9BkDfXJ#3B2vB6OevAr<`em?%ouaq~TTW%V6bY}?JmRHkMT>a|CY`;d
z!iYn_ul;g+hXfbyJsCu|pi;^QhEf=A6XmsuTfp-AI{>=;tW5D8HswMk7OBTyminG*
z1LUi7Ac$m65#o7rF^wQlvdwCQa$i$ZwoDUhZe&zhpnbLlRJG0HE@8uC*m0mTuL|)m
z<IGwP8aUN)xY<n@HWgV({fb<X5-oCC>b`&NBr4VB#W6O{lz3dwsS)EZSz;p6<4%Fn
z#nyR|Rpqq97+kw=;rI*9?q18cuZPfFOmjaIzw|CXo7xm_pFQC)BD~|7GNK$t2|`Jc
z64i$HNvbKje>ex%HZ?hAvx{4CW4Oa@XWl|BMFB5&m0pM>pyTn1{<j73Gz&5XDQWSM
z07vWBZHy-NMiELYnLX(dm{j4iAh7}%aRctyz9u)pMB0d}0o-&otjHuT<LBpxm)tYO
z=lc*pPyzobXamGtp~ynG>TmHGcly2}{%|4+^NhNO2#@d!cC1mF#y(w_ZDai$+d@FC
z?RXJog-bgnmC(RQtX%B(J2t!}q&N+c!ck^&0*(OCeek(QQqEWuh2RB43>6P0gikNA
zZ?<4>0TYPDNNjldPhK`<#zX@X&)R3-OxwlmlB<{PI`;B(2>z4j4;eb5w%5FP_du9Q
zfu(SM43is?cX6HezPMCt=q4f)t96(&=*53W)ITsAB}H~p=$4V-Z-hpB!Ag?qvR|$G
za$aPJRbrt36OohtjNux9;CT4kDr95Eur_OI=<fU#&$$fZ32PDzV)(JgISmSBK)2-N
z?nfKOg#2Xj)Bj=81leI5V_j#3nT%1iX|7^^Nc}p7R_WLoA8!7*4gr?i#BDe5WiW0$
z93QI~f10X8t9vg_56IvB(F0`eJZ}BORe<nsX?NV=hZnc~T@Hq~pR0+Qqkno#*b%si
zAyvcM7>yd88kI8~+);abex#Pze$P<?VHR;51ec5Mia3!LxxmWhZsxy^9s8~2oW&rb
z5X{`vZIAVtQ&-a2Z-3K6Us|gZiw}>`Not$LF6aR_)zR^?+!d^>CJ-1!pt|BdU!NpL
zdAAEka2yW9L!de}YJ5OqSPI3uu#UJJqY<&Fcvev}vE<=IPy^`AZ8;HSX>Qo99VwW2
z@><m+MeLJ$#rBKqZ24!_#ElfYv=ovSF65z+NDi3ACSa@TbcTGXF*1{}t)1EDh~J~q
z_Q%r>kl?H5YYr~qK8O@T%B{hUTW{HK;Qs)kWpm9ej>E091V6?^F|Sbzz<r-ETq91m
z8I*D@O03Kl0v@iR$_3b0vU9OKexNYHQw(Im1qRB7Z7@T?(MM4;n+|^LZk4UdCavR<
zqG)d_h?R`*<oEDH31NSc2Z<DtczFp0k>USIQ(LgoEndNGBUuxf4@YKK-qb2ma^Q>0
z!pEtm4a)L4(z%jb8gHv{E_1ecs8)-v{xH01biXm?IWU*kdw+wO(NLFYS8OqIhPkDp
zW4$srFD0AwhTn1B<`O<s0<Q~7RZ5C<Ww*KJgm^H^AMvRjYF?+DuU=7@LS%D`#zy!P
zL;8=p19kWUo{))1rW6(2WST|8)N#_8M#67?iF%61z{Q-i>}-}up;498+PsPq3eNY7
zlu9RQJqa^ZPGFoBMKy9TadaU-(}*T`z>#ps{wf^6H}do#;K%IudZ0@mMJ0b7*n={X
z(+M-9B_4J8%Bfwonj>NMnNcmZB(}7#Gl`iBRY)b^L`qVAL2xMz;lw!H%5j%Nvipf0
zz`_<ax3vP$y{Wy7$uR}IH4p@VSyy8q#ht3e{&dhu;r-icB*);MrsssC7yrmP2_aB5
z=R8yxoA123I049KOf@OieMy^dthG3(vz~sfof6Lp8;qM{RL6U=K**8@yu0YxKo5^j
zPAf+DxokM5&#^y%iYpQQA{e7PP5PjAr&x}}E^sX<HDLB%4a4f$ZEAyCRHX@yC?t}z
z`gqOUSe5DE#yo^-!4R~qk3S_cNJhj)4l+rXVE$`@3b~KBR}DL|BFEN~{F%<1tp(()
zt|9~{-{hOVoqo`S1}(%qJAUSs>L1h)07N{YR@A$dbjf~1br|HR;0{a!u=+UL-{hlC
zPU<?pY{T4cwr);N?jL2EFSypShw6WtfgMr#HrHvd{hmXbxGsGA_8mT{+alNHIiU&Z
zr;+DR79!T+f9@G^U2)4~tBe5|TeragI9Hkc-?+ZrrIMT7*pa*HDsJ$a>KVfeddN=;
zT$sP)o*eCzbnQmDfx*+MpkYjLhNMNTt4bY!t5tDvZ8sGH4U(7EOF(kvQ)5<)A4XtB
zF_J4){2y>LmlDjhg~6?dh>SBjw=+2LYe<_XFDd|2-IB{79j`)r_SP2?jU`@#oK0|k
zlGB!ipyZTBhN;jrK;L%s(X0JO7mg7x#ZR3gc@c=-RzJ>+I;)Zup`~6?l9XkQ4#lY&
zJM<L9v#uC*-`AS-SkQ{re?!0$538ivyzg?`cp5KPR1-t_$Dj)dED?|KvcLxRrH}_<
z#1z%EgpT+Ko+(#{sm@}tGkT)=#WVvZ$aS0zSA%jKWE479eu%IESo2Jru7T38+TKLd
zZ}E~Mal%e4+qWxs=KA0!Ta73E%_D0}Z+XY(e*(VNXW42It*bQg)mcwbZA`V0y(1II
zC<@FDZWV|QW(Z)XthwcPwOJb93EWqtLZIPPC#TGzw{Z<y_tz&LTQ94bVdmME+3mLg
zEEMjK38q0pBroCYFt!X+lnr^DyqDrgo_ZgGL66TPJJ<-C4W9;1YGMdUtIGE;No0Vo
zo2;E>j=$r>3I2}F@BUuT{?rW@UzeDQ1Sqm%x8{z!t0G*9EKE3bN&E;>&Q_^n8TnwV
zy$q9~>wfv1>I^yVLASXaBT>*KruDrSi0=z3W8FH{;bN>K@O;bLpquTeO&U&w9?9gK
zPV;+u#B0L55N*?@zBu-TF3!R#d1<fq6*JCz5VMFLS>~)TJY!5K2+<Wa>HOp2=O;gr
z#ZjaC-IAn+F)yyOs1>9d9iwGtg=(GPu-o6+N$9w%&4YAwHBr-5u~|TnI)92zg$DrW
za@ygPsIFQTJ^C(%oF0L88HEP2+%%(DXrwl{p`Cb;7s^1UZ5;hHxzN6z!A(x-{457<
zb=dv88bQE~g>b~56ok5z`YZ3+bA1ksn!hX1m48)rluc}JA2<dazIpS-9tb=+fOudR
z8Jo*v4iC8Qv2quZXKhFyUNhb;xRb^qO7eQQmX!2PaEOy)7O~CqGEIduPoj8}YJ`nd
zZDvss4caa)pB2?DPj3>^LK*hNECqo`VEZppgO4r*NeT?%4YG5$xYfkHCvaT#oDkOh
zvv_qP=gUtCjHMtosWQ9XcxHtjl#BX&7YnbR)USqVNDE}O%mO}*uG0`m6U~OrpHhR5
zsiS$aNmY^T$?G--`9xW5Vy&gw2%U+M8yM`@tIkfd%5UOR9OvxVd$!(odFN{6YY@jV
zMkX@%YEvr^OOLajDce-C)iwt+_K1}=zTz)I62bjme{p|>Y`X`xaEO331(*{yJ7kHL
zs)#FO)MDyCc_j;zAMu1i!8{-Su)EeJ#;(IwKvE5whq1ES%wgcQ#X(T)JpI|urg@{z
zwNUw2$yhF_6L(W!U^<iD>{{OJ_$)+tI_R3jSZNdy!J27)iL%Hs6~hqrVVhT@xgzlq
zxcj0lfzu1@)D;sZpKnrl)nE1v!1tgXv7#*@DvkjY47tg`oQit8nk=&AeTpq*Jw|nR
zNR_yIVu?P2dA`EK!GJVz{tU+apB)i$ETz<R64kG5d|m%vQh~#~+Z+7)v<xUzkyZg|
zjHn&_KXE1j8ST%pSz}dvY`R~~oXLC)7!WM!E)lNREi6_v-~T*msCKjT<`{B$ih&M5
zM?sNQPa}2`*|Y-Ed*|KH&umTT3D&mqY&Jg~qWY$3A3k98uQ-d#0{F_-wamb}WUvPC
ze5_=2GnPo0;Ad5LmsW#_miPDTuW{M;kdzNI<19pJfuTZ2&6|JjpD`I70&<8UfAx%0
zO3l@tbjom4Xb4tft(Q5KMg`LbOSiD(LcMXx5kX+^BBSY}#hj|zDyGePT@8yMnWkib
zbX&w)U`gMb?T^r#Qgn@@UQdUD!o--tOs1()>Lyj?J{M3h%YI=R-sWiZzYVY*fp+RA
z#u%zvZ5!k=v7OoE5Y7&CV~o@cD0fKuC@K8mLfV@}Y>4Kxscaluknw5@AQ!`$&{<Gu
z2a2l8nyowRavS689{_@q!8Q!~SQ@4ob2V~u$5bYVZ<r79RPK{pqX2oG%O2y_t>4?q
z7s`XzXU>YLC@y`(zTw^FOdAdIH95mT@gqcZ55vXC|2+~#s;A(8wpwn$=4NKaBZVb|
z8PCg2A?%eIv34q;6*Uf*sZ4+Lo8lye<f5W9(_1&S{q%(8)t#<fZu5Bjl|LJaM9NHH
zx}Pdrg5yjkiOF8cqe7}}cJBi|Q(EQ0VL2(+=9PmeTDvQXUhFJjGPF1nmdxtV0<POq
z!#S!I^As@2+^&oF9c#g0=gDJ*WPhLz(5u<5>sF|muhhW{i%^YVV}DILQ)v9X9*jyt
z`OTkdY*x59&rdwyrpEgvE!NoaV%>#EJP5AzeQ_3hJ8oVp5woyg>)asjGcSVMDjyr+
zX(G-Kn&oOqinJw!FjW7uwkO^JQNfEQ<)5F_RAv2m7$dYlMn>Lv-J@!?tr4;P@9H8v
zmnM(Qv3zj$H;wVGKU|%Jn!!b|ERR*_{BC2FcJmZn=u#?~f=$ko8h5mG+bmsN?Dm|^
zH-qqa$i{FBwJKx@K|%p}*dguTc|{Aa?x)}7hgJ=#WWq`Khlujj<d*y<b3^&XJl_Tj
z7%hPWqWYg+->;INkci-|hxBG<Z3_#0d^$@iX;?@nxNaOQDrkFwrD<NqrYdUB%BE_@
z!Q^toQ#Gd4<@v;46Fl1hWSu7Zdgu_ZD<{t5JK4FTgH_|K?87Jc+>T!+jYDooYwc;s
zK#PVlwKk$7c1oDMow~+;ZI~0+qkWcOP|OM(>A(wYY%a;H5tF=9;64z-QA+u{ZzYvj
zNOg$p*mc^Xr`=Sc7T&G=^d(snD?J3O)L8j^JN{-uY%>Y<ZZm8(R|TFsek7j5dpMp0
zmOV~;7FD&Fd|I=5asmt3tTnlSgrU%hz+HsGdVZW#gsxh_y0*^}v9yFy3GW^$oL?N#
zlxt&=obi+ald8YuiOX%{Vp(qNGsPN}*#1k?*hZa&6|Ru}+*-O%%iwftWjX(EHFxcD
znw1t){~3zp?o|w+-Nk2bT{QtR7RaK6jp<12K@35e2(Da{%1Oqy3CdJX<u-nWW^lxb
z`U`D)g>9CR+#j-T*aXfe&Pqo^3j_gat#aJD>cTOI(v}V;jnh0=e}4Z*Vf~Ec)hN6Z
zoTiRqI+uwJ9-eb({F&>mqu=?*X~;Dsq39Es3~TSvuSp?LL17^_f9PyaH{I84+HVcJ
z$+5~GJln@0Dd73npkcp@&R3ApKQ;*6ogqg!!Z6MvZH;F2>j$z-pRbRUVaIV2Te3K<
zwRqfLzuqeb>-W0b<hO_Fs|sHYXe%h@zgY=YiR!>^lH@6T|1^kD$s|G3_8u0iNGoT$
zq&DT_`r2i*7ks0C`1F_EO6ENtvmfeD#w2w_p`ehYJeK&xzD92!J`R(~2_m*z)G6)X
zui}QP%C59>$O<kXcRn*A7p%i6cU`_zHu*f~@JHT$RlEq{(P-0FT`}Iz!w&c*O_+&x
z*W+F|gF<i|;=DX*lNhUY7u$|tg3vpW<!}gm&R74#5cex)i0?+>VVOsoCGkO&HY@vd
z(Rk-3?CT7XY8%xue2x1jU(w9sn6jH;HVqUNIK=^%g2V~eM=&lG>}tZYIm@C&?29C5
z;pttJ&;c-IGhNuE!8ouX^87x<)AT@Me{vi6DzpfwDW#c%qd>X*$Ib(7^uav`086pI
z1MXO{v1B2<YPs%!&LZ|U^)@%NeDAqe5LQaKu=rTI=s{i^I7}Y+PkSmd#7v!DUC{!l
zt368XacqfkQc}JL-qtxyMqI~l6r6^T9StsCN5(T=`)_XuVLziym%o5>zSo5p;gJ&1
z<&wJ(xbVOma@y-(O@eH3i#%<A5+pnIXD3Us9E&ZSLs90SRtJq{j{4P7D5|w7QF+;v
z_3x54Gf%9@CxR^ZFIwbztJFSUk8m8Z)n}%2!4MW;h8H^kMSE5z>F})s#M0)sr-+<L
zPUYeWd<-xeKc)I;bMM-KTT0NQ&V5w#>W5GjaCXrtp(G3=;W;{AbP#V)W!*o(3EF=t
zu~a|O$v?3rk!9)`3&qtYR0V6+Xj>@rilBFiBE=?Eh}2hLO<zp{?_&g+smH%xpix#H
zhg**Xp3f-M@%kiKE~1)XS{AP4vQnx?M|UZmp8tAN^Z+b+)tz5cJjcIglDVQt78q+Q
zXBN*hi}m)Qqtgl*jZM-BJk+x19=-*bpVufu3}G0zQsxZc+$Zbko76KaJNpA7fbL$C
zUZ32hRTCN-6>Rx3!=LZ3B1yC03l?N>(P}6%3AH6r1<C@AM3%Hg^{59|Konp-F&9hN
zJ+rK`Q}r^VO)9n>L#)hS;C>_sQJ*0%uDti>@o3HG1WC$8Qu{mIvVNkP_tX7g&zD=z
z;l;DwJ;K)ujU~+@HIWSL+#U%C34x+0AUPgX<uqO*c$9u}iG=H~oRMI!kTl%|Vq%ua
z&8nLegjmtNI(=ss063YJgto#-O8=*M{9p*nn{=bsF1zPYYA@yMvqyM1T%}*rr%5b%
zGL@wrn|zw_S9xvY>^5UOJ>q(RR4wgc;_dMvt0LJf8+8CFk(pq6E|-FTFdD<pYM7eP
z8BF*bAj%h)k(X$NaH2j?Mz|C*<tV=7gkh{iECmGWik(ABN8?-y?F?#F=&`?P5_Mk_
zTMkYAeOxgID+S34-Bt|V7LCo|Rz!=>yqk0|nEP8!a8acmi#lNN>z_D)@qJw4udRkT
zQY@>Mz{m1J*}HcRDRRKnn^!i^_xDYM*WTs_E*)VJ=rW!nLQve8B5mMacgBfF%3V&3
zOj_{XzN(=95}SkjCx4eDbYulbG0TlZo;MWlWF#@opxe2dn^sayBD9_c*4aU;ecXTG
znU{!cSj**<N>0yui)Kf29UUwY#x|tQTIG(;iV#MQDqbVs3S;t{mcv<Mu}IY6v+2=L
zJ{6F<Br(Z<eA&tY(ci@Y|9y&}+Qzokp%0K7K#WCP29^39cUn{s5yV?UoaoiIb#u6!
zOyUEI=FbipGl1@U^6bg{&pAx%G?hi>-G>+W6DvpIK%g4Z1a|+zS5@8RF>WzR<TNW6
zi76?@@{-0>D(_$4S@q@^^Y*-aJolaI$tMF^3W8C5W~d0Ii=_SN8b{W2K-f{A1lH&N
zkiW1L?~=Xe^g@`%faSrNX6t0q$-ha(0BBBcySh!ufM46Zv(y>xi{h;mI8@Ar4Kb@I
zQh}QL&ICg=6q|m2Zny{YIBvs5|NR{urcmlins<{qpj+|kB8$t)Ec_UYv#wkcKtZU{
zW#Y@6O=Qw4yXa4Clolst@3Xss)A&pIqMb~$YkoF`C0WWT6d53f{0rBi#ARzzOlT2F
z$HT*2_!zF{L@nw9MEy6qgu-PxrYJ)tZ^8%JF&MG%rNN_2nZYgzw@6=U6)Vk_D<$?=
z@pO7-I<7Rjr^XbPlbE@(4w2RhS%KbH94>=X*hq=>cZASH1~&%6Q7*F>35VGT+ua-B
z)ZJ+!MWO^qFz(giN{JIhN+czr!fRWf)HY2nM$N%={-NdIwf)w=_X(|6V>W#;`8#vL
zdDrE$GN*<8jjUuf>3o<(A+GYMC2uA)uRUSuW#SP{Y5dUf7x=;~&ji=qC%02~we~Vu
z6w@SPG_je`DDh&wA!z6sW<C8s#OK8zMX-2J>=<)Jd<Km5cExHrnm&7`lpo_Jx2Qae
zBPdll!_RC0v|2#)jZNk9+e$Op(DKj4J>MdgL?n}2b<Mw9+6wR24##sA&=C0`@rZBW
zH`+FRvVrZZLL}o=31~3~y^#wrAxwLshD4KfYmuZO2B6QJsBO-5FA)Zd&9(ZICI-KW
zI5yfoaJKr|F|9-?e>GrPIh^maGIVJndi#Cr`SFIBj44UL*G`g-V@+38FAExJ9m=`P
zqf!$3y=$}mIt~53r-sT&U{V6M@N_184WorU6zAx12>KPv4!`&!KCH0-pHrJED?{Pg
zprE+#J<#9E-@u_UvIu}1t$DDHLpUH}k>ufF+7xsq9R&p>+MMOfl*hdpnvyWow5qi2
z)XCAPoK=?iCoo=rQK?L&Y@yw;YNUf(zpD|?EL>5U5`1~Pi7sF+@tdZ|5MWq5sY|s_
z%8@|HW!6Z29`X2WsY+$CcXvl4cObo;2{-}H@E-tYdUmhuFs9|rlH2#u&6`fXr_|fm
zoz`}bU+nl47R6&sm-`CRoF{}6{Nm;a<FO~ts5lcfrEnkOcB#Lh_$Yn@Ng45ShkMW`
z=EvQ1vBF>Cf9EwA=1pOHh=!_)Qm{tXL&mq{uxA1ZwMnkA^G}>|qpMk$lj<u0tP)#~
zii2gk8XFPlzuj1eV$B0>_wWho7MM2G&-)cskJ}R(SS%s44YizYL;u9gHx8dCEv|G|
z1iQ)%Xl(RvqVo<}MV$+#P)dabOY^^ky4zTh{kF@`jA2My2wvYlz&5X;Y=l`;T7i!e
z;m0y&))3AybTQ(@k4uchuCAz%3_EHkY2()t$2AZs<xF{;+{6wJdV})bY<fO@CDx+Y
zK7(Z8z~7Lu$$jjNi|Qk?K;=@?qSV#Mq|~?RA3Im4OI&|>tdA2-n`c^vXpzN=pHS2&
zXKD#Zdz>$iQzA|a8Bh>!uC@^!?>A{<mbJ&G@cjqO0jl|N-96UqIQVq`W?M$F`neK9
zjT&_vJx|@{@;E3v5)7Aj(UM~qZ~o;x(qg#w^+z2YZZPTNQ++z!fi0p2uL<4Zy##ln
zJu>zAW_8Ga>ffiKhbLBJ#Uoc-hC7GXnMD1!Js$3P`T5XSi*0kT{0*WmlXu?+F+5*P
zs~;*(gzbCWsnRG$`)Sh~BjF$Z(1Gvo_{K(LDn_qC@3o*KJq--iyL7LQv{2|-{r2tB
zm^DQz=UO5m*xLFTzIp66;?RS9jW_p8!%1ANy0;iF`OhrCMJa?ePVB_lyN?mvK`tGf
zx)xJ+Z*NaTE_?!)@-x*?uy2U(;a=0AD!dQ$u6+x~dLbRAF~M`v$f=;&eos3&X~1BY
zwUcmS_MmmafzR>&IB(?8&>y+d-T|g++;yv|RV=trP(u|>QSz<4S_deHCeo7XpU6CA
zS!uGrm#~FrjS}$+*918K!)DYydAkNB3pVi{)g-0^XRy2XkZzi0+Q!jCQ&w^c-O5dB
z)K<-#f_2oJ3uUz>VMx<|Em0dL36fnkOB=u5KH)@)dsTXO9<8Cs;Ty&hU~w8+)pUE&
z!iJ0K6J~0<(kz{?eLSv+BTeH|vz-pPhixGexEawvtHyXVl=>sjqo{Gbo>s;~fYf0!
zqKf6~=4<6^a_8&yhPcH%S`m7nROc@P{bcdkQT<$PRf9PohVzs4b7T0w*UGMF7Zk1?
zr%wsmDhwa?M;fSLSmtp7p#mSQpOw=~2&=a8K~2KeZk#m-vW!Zd!A>EMpvfLa68)^k
z-?R@fOjWhcfM?sJTv(4E0xrKe5-fKhGfjh&h7~ciQ4E(A>gW0*GYOG5Ec51?l+ARq
zntRt_44nPvd~%371#r!%XihGErmWg0tG45i&^9x+Sdq4scTgUX57T1@MA-fwAMc!8
z92=G7#K)&(548FUYMHt^-bI4<&t@JCdl@}iC#=fU&E}N3y?5Vl$Mm#lTrwYuH-gjx
z=ovMn@G58>d34vVL&+ng%BGgsyPxgHXid?kcmG0m_J6(zUZTzqzl>A5HvE)Zk^+uf
zidD$b`6K_9b=R<09Y4ZKzgtx$3=*F@@DS!E=B`!0`D|lyX4RSpC#T#(8pm#;=jH1m
z>ykK3@$0)#tNHhD#&6g!_0tz4<)<LTp105Oo!uAA2@vb7QHdp17zwdEW%&{3w4LJ4
zjbI@4LY7oxNYYU74aAl=h?0>sh%^t6g5H7AB=VWwY6s0vwBQdWL~;Fs+3oCFuP}@$
zrsFBzphlb^MLlVu?X07jd+I2cla6U8R+DN1Rp4><1jJB@x}X#Z6fUxPgz?B7{pJ>m
zCrf#gMB>EwJhk{~Xx5+dFgHtM@p=)E5`|93)5g#?RM?WRy2hk6Q&!P>KJqiRT;JK#
z{l@Y<wH3`P$u37;-d#iOWH)!H!wFN3nvHbHRz9-T8$p+n)_4R6R+CgM+-qt3A6;DO
zU{q#+TzXjOCRF-~q@gGwp?65vUh5%ie;xmPjJh!O>RzV|3phtD%s+Vqxyroo6}sxQ
z)09It7~M^(x2jxv^kD-I+<VuRstU0eI&|mPO`517;S`~NEl-WYtPKZg6fx*8&}6aD
zUe@&9gBZGBxW0BKzG9}5Dt@I;YWm}{T9Z_3G~{y3ng!eL*KN3==L_5+JqGm5sN*}f
z9yKqEW_f9+ew+>0OT^f{3etx5iMzZNjMQ81bMno01~E<O_MQt}-V2Z1V~6_Ib>SoI
z`si+sI3iQ98OTWWw>2kO1}E)l)op4HX&#;32ektkvl%I#++g<1Nj3o$!KSj;(TxQ3
z$yFIP-a%fD;he5OfwqTDCmN9kjC?;oT=9a|li45Cf?G@KAhtdcRe#V#V=|eUDN6>z
zGeZTIIr8Y)NYP>87yK=-)1*b$L$={HercW!MGDzMorOrhy`EqGYmqWXn78n0QjwyX
z1df|2?Z?S={V96ea=I9oa4i58I|!y%)Kj-vi`14g{cWQ78>O+h%fHh7bieGGMVxRf
z;g?P_`5Ug~)*nKz+?y(lsu9-xCZ@N50txH@Idy8)eQ;-PnePhh3Z?M~^^2)}-{o0z
z<{ErdPps`Y6JuKzQw8g&KqbAmplYfU?N0q<%t~;onJiL#)*J4H&S0e$H*U8UQ2L_z
z@o`5$0epd+|8>71Js0am{sZ09p-Aq7-Q9-Ke3psE?~vJ-!0v90Wn^+?jlDQDEsW2B
zuh*V?_m7%8zIK?ZrVuIx@E!M{c^sfKkh+fY$p@SN^{(vCqlLHK0q@FasE7o1#wnct
zPc1vauztA&dzf|Pvek+F++ADs$bGqMR``pJNJ(D)2?5iH>T>AwE*M2>!)Oa^;xZ;E
zAM`Pq!2LAD*da-au<1-LZfbfgs-L_fXH#n^i(6}>Ngf*N%l#V{NlVcza8~d?t1tRB
zvyPU=6Q#Y45S|CBN|6GJF~0rQ)I%H?L<OFC;2wxSLxf|lui0orlmNt)y;8$@Xv?1D
zCPXD1kp1mqpcv(%b9A=&8NaVRjmpz9<@Y!#!?99_{S(1;aQ+%^A0pL6a2RyOBpSZN
zrBwi1#X`Rv95g=zn0idB6rw68yMR6Qk+khL99ZDXm*%e3DCU-=difhBa0@hY1vU@`
znM`^X(!t91P_oF+r_&W7Jo*;DskN;@>`)PirUk_+DK0Wpp*EW0b6vA*^q9$0Q&!$T
z_}|U8yZlBr-1O=fv7B?gibd*ESqgW#RbWa=>L?EL@A$fUaIB$#&lp~QlMXsw`waet
zSc)r=xfp~<pmxRai;@xGAukkXl#NqYbs7UxzeAx+5(Su8BbNvzO=Tbk6WiJ4LP<&j
zZ;v0Pi_dz!C1EAgLZX(!CS0mfsZahzfR!9b7jo4rmxsDL$_g=Kw!Z{s4Ec`?sIRMN
zXsxW#%v7p+P|MREvr3mn9kTA_v!ZoAg0G-GQ<8Jk1Ti`~H%eP6M+IY-B2j}$p&Mil
ztDn<X6@XB%k~va`u}MMcv=YU3a(016wQTg4#zcbK<k`X$=Tou3lW>Xm?S3@bFTvOW
z{H8d$#Fcbek5ncUP?`5WjwZFKZ)UjSFwm-}Jlv{MeZQ##xs68h%3wMqx?Xo!tC9N-
z_Vr8(5GG@CdHMLe!i?z<WTjyS^sHy0TP`L}u2%JEpoU%Lvlimp&k%W*ke~1$UXgZ?
z_!pm5bJ<l#;~Q>0pj!Vfk6^mr?>Rbso=4wKC{!ydG06p;bgYi<orh_U`;U83N#G%H
z9Ec+5aANijB$K~K7eE5kCCL9gT2+9J={(1lHbaayV{63G@4MG1IW#;pG{s=~zB%%+
z2wn8&qd?VX#h~Nq%hq!X;s;@}C^crsU*k%bz0Q{@^=rI|*`L9bzkkRQm(*GMn<EvC
zC9j+zU{CB4TxIh<bGn8uMiIH${67HVKpnqO?&CO|C->a?TXEbASHaLl2!Zb`2^Zno
zn+0Ma0zah#O2Uev79!{m4-X8(fR&Z366CZzp;HD|YBAQObedi!CCH=DmY{LgY80s?
zZ{Wr-B+OD2cO9zte9DY-3AE1heC4vNJY5z(KY$HGk9qo@+{;eh`VJC>%F-4DWo?N{
zVeyrIIdBQ)!|^ZyjcZ%`^%3%z8K#)SWjQk)jz|(tr&llw=lLmSR8-mGzb%dxL`Skd
zP-Uf2I2Xf9Lxuw)4XK4PS02yjjzAacy5J7u_>$y8ON~n@JYF6tuITWJ#!K)aXpip1
zBJ$y$yy*!0n6N=uwV%MOb#*Djx?W$k6@!T7(y+B+7TTnd$_iEbxx{R5LTIN8w%*_w
zMM2{(?2UF8Fy}g-aaWeIFpdWv9*(?wd}dy*w0y!OQQAzcjmPDQ)!u;$Yps}FIXw(y
z2|YKrD7N)3H@?U7r}5`9J{7P&jqh!g8ifg!K%;o-LVO<_df@)><9C1F8iLM_746-#
zF7@b<YRBJc-=qVAuf4nwkBiPg&e7HLu3xovQQSim(eW-gskmA=>$rA-kPvEhoM<yj
z2v&9!)><obDLh|iR(Iy4Na#CKRtv7Z)ew^PjU4H0%dUB)SwW@ZQiE>iL7!<J+E?4-
zb55viqPBu)?zDv(f9HW3i`R=QX|;Ey<#yH0H~aK{y`^*aWuOw>jn)%<Q6Y<>X~1(7
zu#!6|HO8tG8g*8RjZ_nam#Dl#q#$zp^pPyy>*w@eBoctCP!v=`E6Mr+sY~hixu8K)
zu&_IZ76q)1SPNgC_^uTRO%l5m1O-Uai&i`8;5G!uy;0kRwOm-%XPzDt^IVzNYZvtm
zoh>||<m2(gJTI}4TXgXEIV`f90|cPVzxkGb(xWuCTr3Ve+V9{V$F~tEf?J}zUjyGB
z+l6P@j>=IsOnqW@j3r5jxDPD^^mWP{4in=zLTx1NBC5SZZ#9bCYbl5*hiL#T;+R->
zTJ87T<K2D#?^d2DE%egPbC@K|ZKZa=z0he=BK!B({^U)Vb}NUP1s$b$*Wn=0S&fj?
zb%8ttg-gZaz$%%nPEO%cX|2%OiYDbaJdo3{UG!e~-md-a&?}NiY1}*O*J6~l@=6Cp
zg=t)Y%z+U#1rt^WPX%3~s9e2KEm7J4>4B6VC>B~N`~2nZr<-hS!!$H9su<#}H_{#P
ztyCQE;e^&uvr28@x++2-%LrlX==OTI91m;mz^4S|$nLJdZ8mmtOm|_MmuHssNUbYt
zahhLQN((2=j-U<dZ*=D)EsAb#gsOb_eI?$*T{;5m;~C#)EaxqHWfKet-Kan1SVm+6
zeX^w)zQyzR5X&j2C`#SMGQusE{fCa_`~A4*WZplAd>G4k%o#ODJ8CP$pY2<*tUYJ2
zK5t5oRO_)^Jq%D3rxGZeA3B!3AGklpJ$VCCdcP;qY!?^5?F%t}XB_YG-ecIqo$pT3
zGElB2Xo}nr7tEa0FV`?+#xdxqTdt|9<Qxus)vcq^14*c<uMb7Te#t^s!suR#JMyl_
zu{pIhN^4={(JJe@vaW@-7*->3jJ1YNpWLoXkK;}$v#z}wh||_lL|xmi(*%bhR<Wc)
zvS5Y`DvTL|3cO-oNtzkPiCmXxsM47jJGETc877hu9dH^(V>Zy{98z~^))xC-Zzzjz
z-(N_109VusUMfpzm>iHaK^u2O+(xB?Er0EWV|xC%#qn0G^*~>V-_9Tn>j<n{9PevG
zpBGz<w7&Mc(f4Gf^^?S=Ew@XouoYFI)I!pgQdV*v!mV4_xDPm+18Lz>DyF5I1%wg9
zMh)LY=Jyk0-@hEX?8v9_=P*7Mus)6NZIo?cbx<{ug@fJg5M)FjClVa9O2s)0M<i)%
zA|1-=2;M?RW75o0uUM^kiiI&*BI`ibBby%7kSC5(m{2AWhGSyLnIRoHKc6{2UrA{I
zg!AQ*l$6iD_#A6-FfmGWVcoFWsHKt;Fr+B<3>o!+<Gy=-;_cHj|L}OGy5Qr9NtI(t
z;6Zy_ZDDQ~Mm>-m)LX!)DdFHp?OoAI>E9V=*AdBC8HU8+Fp|Z2H66l&w(E$A3``ny
zhGv)U;J0>%AV37oAYE^yi2;eEDQty3u~etnid_OFR<A2B#>C4!Fy+iNOf2Us>%8J)
z*n1DF!czI_w^uBE#_841S({Q#Fh2*il7>koX-|B&4G@8TaBT`zoW%>>Qm0!gL+Trs
zX>@c-Yk|zdv9e-bFpsij%28M>y+h*=Ns&t@T{_oR7&W6gqm~d;ToedxEWv5MaGk&6
zzx>Mh$rpdcXRi}^fa#cN!|Uj>)y&U-{#X28|HuD<|L6bX|IYvRAO1hg*DI|iUOl|%
zu1VWwOW&aVhTQrHMhDN}4(cIdT00ssBqst5BmLlIgccM@(4=I#47~cz(8x)d#)0B1
ztrlD=;?5`&qmB$|+9CV-Ffu%hjALe+GLt@l8Qv<YK}oK)&|0N-7t+>(oGUpY8i<m*
zR!SB8cnXU$1d$QV2U@EvwepW&zvc1y%=^-4;xJUu2~UYJ9eH(pja>(<22I(!sBt8K
zjiCNCsxL8zQi$ymIW*!hs&Y)3(=;M(93*4C<enuns6eaK^~yRvqNy<re?d-Tr$)y$
z+-M8oce!=WGe+WCjf)v9iIjzDhKCFXgfT0JailDTx)dG`4-A81S$O+A^EdzaYfN79
z;?+;NHf3!BtRM@e&6v;8ln~$g3K7Q=3Hzo6b}iu|7`la<uyDkFq0!7)%0>kY1YT4^
z;t=5}b$7Rht0H6Q)EUmwHXe+12R7NQ;kQUjDd=0ZJ@WWu{O7-Iy!q@GynOW&N?Vv)
z!KVkjfiKE5J#c*a3m%`&T(4J(HA*wav>k$NcVP357W-Hb8{7CkW4U?>XF3gZUiNSm
z{?m=+SF~%Q)o3`V3CZIyRlV|k#d6l4qN22R)1I=f^;CNi=?9Bt7LsmFw78#96O1^z
zkL8WBlT}HwH6Nh(cg1ovjnF{;pb0mlmKM_{2V%p-SS*U%ZvKPD@_xAI<J*(lcCVa3
z+I3o{!0tdNbjtJZierQQNeC$=Mj0q}rInTD1@lI3BdQ^yZ{}DHgi!$Y8k(Hr!--Za
z%k_c~PSWS|h1MEF9vO#j$>gEg=%vt>6{}IuWZ0;Tf%eJcz`Ta@>~G(_=R7YwSu`NM
zl|GM0+(u|)M-)cUIie5TjOGQTLhAaO1j^f@I!4k&)<kwEx5jZiAQ;jb*&(HoF$v=U
z(kf-Su#V4+!-1DCUt=`<t%k68BIvMgoaBf|CTT`wq)KA-#A1oB7w0lKfBEJOS)4IH
z^6*CZ=`R!IS}1ej`5La@m-T>EXE9^76V(n#e#wwd6f?{U?hCN=+t}BqK0mf5er**L
zl7)7w8^PY_MMp<iJzT}t+6$M?UYL{A<5(h&-5k0ZfD!j)rXVhiu1q8jLne6!3mdK|
z8_tyC{N~HQ=he$MOk*@Ce)`213><M6Ru_!KDIZvwITHnM@kVV%YrQCochYEl*{xWX
z<)gM)`856+<5L0a)A-)T2MJMcoo*e@ZPC$wvUd+1h_-c7?^mG6G`jXioEglGW|gL4
zpOiLQ9~v8(byU_)gHTdZQcmGq1k_qXR9&RIxx|}^#Afaq?dmAPZ3&t6zU$p*Lny7W
z*1}p_7!TLbpbDf!M~w9(ertAah$`!-g!?}7t#cu}W|;QFQ%=gD!fBdVT4+Kgk=Tf}
zYYcS1LAT9nK6T4)@1U_b2<|8_4`f;+3R}%H;)&v!V#;+{5Lb?)5H*>G5xN}bdF8q~
zl3p;3ugS;H@ugs;fNkBWI5pk!!wvbx?f-irz<VNmQ|#Is)I=Ai_K6gqoBrJXGSUxZ
zo95d7J>#HYy?APb)%G_3`rU_~014pMSnbL@pSg@rIN^k49*M9yOw)m%eDPC`ha-6C
zE@;~)?rr|xb&1$Iq94KdW*ocSI-IwExDCJis}0z(>56Wq2S~UeS#30{SVyv?{$JdI
z==vwgftKiF975!@v`WM)`?`{c+#L_)!_OsgK(f}M?^`>%Q)iu^pf05_mzAY@7b@-e
zMC}JF<d!1uU-YIUvi-?7@`9wjOAg+3ZJdOM%M!|A7^%TpW)Nk{kywVseZ{GFl#jdb
zv;DW}mu#c}54uychT2H>#A=DLL8(U4K;Ec2DLKg!Vu^WOSXQHQK+^-m_?ogdidh^2
zWJ7M)x8nHL^puWt41{NNdG&*07i>t35WS6FwRUX|ZfHf9JCx{_pazn%^>17W$2v{A
z3xN~UiyK0#QQOL8e&(=TnARo4cQHcKEQo-|apL7?pR<<A^?Ic>qqY`@UfEvswkF;E
z@1q&tYb+O|SUB7oPL|E?L-rT@(~IS;QFObH5l`Fh`^S&v&~Mpi(FceD`=MjmgUaXZ
zXW!7a=@<Gwmiv{8eU->Zy#L!bV>v8TBCFX*wzgec*-3lgI~FWNx_`)6zMJJA+n(%q
zalEmpcG;f~0`}H8`&Jz9)T@~3((h?##sxjp-4PE~LTGBy?{Vl^NiQ_!obVPh6)6Rx
zuGU6vm1#JJu34wKZu^yQlcIM5M69n*il&75%32$jW#K%}Tz&IZY&45fIvz$5`xm@h
zCt`23JEWc8#rEx?N*F;Tcd{l#ltDX{RfHiWrWBk;Yqd8^hmIE0+ukJfT>k!=#=lo@
ztY$1`VN&Hvm4QJG$-RM`2DGEwie)4Ti;Z|Qma<Z9AnAn55tx9vFI!P;-_9=^+Ub_f
z-q;p)H*tcd8=r8?t0P~g=NsD!8DGw~vl@3wadKlfw7w=5ZwwM_maP!&JwcD(Ro3;w
zvJ7133+^zCIhx1~Otfbek%fPViv3n3?03?)MU1WZ{b~Hz$EO0;r}4dwtxFV%%x$xU
zNStAB+V>*EwjA}1$9e%tw!VEV#{#Tv;ZiRwG?HZA=)@>vB&7ycoy5qPjA0xYr;)=n
zl2SrETrHMmVHgf!@0>!ovX&O`|2)!4!zs8oB<yVh4drP+tkSzWwHc)umWI8kve6Pm
zLx*8J3{1lS*G_zt&PR9e`Y*QI3w!D=Ol<?#MoK%naTo@sao}e^`6<tr3(If6;&>Q1
zO(&*dAZJBWLu93{XT%DJ;l$cz);hD%7c1R?sJk{5H*5&n0CFP93C};_+6w>aAAZdj
zuTC7N1MZPHUu)&v)%jok^KWUq;xB&jKceF+boz|Pr>~LABlGnMYS0@+8)*}~zT1gz
z0=jX?ZRTNH-V1Fwr0yVel|nINt{ysKBnHbQcNU1mygS9%YM8cOv|4e8vK@WB(G=1&
zU~9Na<)o<e?l)VpOle;D>g}(2`|d5DfA&|5<4fERfwyTF4%2~O{Oqqd9v`|kz~-nF
z+G6foI<tu-hj!1$(kE(LtP-DpOMu}P&ylEb7oqGlw@ohxwq$RktHN3rF6B(kMv}~s
zMg|G-ey<%?ZyEK8vaF<}yn1-WDl7B#0<Qgl8`t+IUwnaRrWT=<5J}yyeQVS%V$1ej
zBr{A$>{?h$<60_nt(5**F&I*2Ovex^UY9ta>41nuQhqCUa-*-tcQ@&z!WNy`1Ps;~
zt#TYja#nuw*=H<k;px2cVwiaGaKLiw4p}Si16J4m8vQyaLI=x&CIrgR=^aa}T-wa+
z3nzcgBrll#gw)E{&nqvF&S4sn<P3-4?XJc7-~NwZaVeSM#lK+~-f%el1$I5-%glOR
z2;I~Qgs^<j=xzt3UZ&rZ<A(RfLGSVQt}f1g;I*C{n+t;n(x&z6*HQ@Vu1<8irBPe!
zhv{(2THR?X42KEn_dI2Vq$I+Uy=ldK;j%m;k0VPhsJ}oo$F=PTpx19c<Nxx1{x`Jx
zKlASEN9OB{x5mSG3^DGsxs=^6ULQvcZ{z!n<)zJB%QFirs)+}k7<5Ej@alhpvAphd
z<tRvPNapP`{PAPiJX6;zt(4Fc!uAf0i&MQJq#rPr8(ZXlSR?#)ERUj$+8c!UUi@tB
z+p#<$B;vpxDA7P|RF6ZxNiULkWH!>X#2-AC-9E}a`JT9Zf8Xu_q5D1LcgOLKbPIvL
z^d83-)>KguMj07oK&-!_K+Yq>Fp_d6<-{-y)OBT977mBZIOK4STUQ=lzKVu$GpyA>
z3zibp!H)h*Iv{SLQ7S?j5=(vN-SabFK0fg@Us+^CMK~k~1BL-Hh32;$a=#$G_}bn1
zTm|bzQEwI7<nA5RAY){09L6EmrPGO08%uK@#*r7NBP&<db*8qNlyx@-=mv|Bph}SR
zntP|n3e_5q>xJjKa%qLul;#JX+kkmvk|@uLEo6mZOrR&$nptJ!+E%`Nd?w{rJUskc
zp4NrAtT=0Yc$pCj(#@Ur+qspm`dShQ+bF2(R?M@w65;L`x|EQ1diLPaSSs6fOwqJ0
zcYAWs`-~uw!=)7lMh@e|&=#bGt{}4M%T})IJM6j8%s8ChFdkoFC5)Fmj~I<wg3h=%
zf%k@y?w|rZ44`(Z&t9_oH2xgMrvlce@%@kc2I_X&;g%q8=z^`b?-ni+=E2L>yGL*P
z?T|8)jHJOyGLQ+Fb0|{A3l4cilcISD+i8h|W}vNvl$A}!Di3+D+BY+-R%+K)Fmo{L
zpYzt%(w@wf>Ok(ou(k8pm8{C3DH65TMP|)Ntsmr+eY3s66N1WCTJ0U2Kzl`(EwCRU
zzMIg~5yyFPI&hc<PR9vNo7O?6>a{`=<>kX`o-VMKwSWGNLfCpD*}3l$UC_}L)e+|i
zGQjkjg7Wrpp)D(IUHR$>o)+Q7>z_e-&EdsMNQXd-sbLZuTrlPfP@@~;8_1ixPIO0T
z*e#{fu2)7^IB@U9$qs>zJxs{1xTU>?yopOTqq{2#R&DPtZS!w)j)c%si@pD@OSOG2
z8qHVCjAec51=g@BcP%_VzvDms&41$Y@jd(Ixp$3-@32jqeQ&)>xIUKgVSf1GIez~K
z?_K-buW_C2Hr=;^k3&w0gACZ9QNWThVD}#S7vT!2;0ZL)7}noodt>d0$&GRniEBK6
zv$aM-blvtuCz3Z2ArBcdSe6Eo(G+wbQ6Z<1F&#Lh37S#XncNy0u!LTv*8P3AINmgX
ziq+_1OGj;bjf1gU6iZX);dmgYaKbzt4xGjVS7utrqZ*A?D|oDF?!MEHHth=Wo=9S}
z)|l(U#TE`Iq9ZPcNDA|br*&od^;Zl@NUD7KSUJy$@!>N>4(RC%M2@ju&IE;zIPj@7
zZkE-z`MckL>By_?bSvKDc%uymvV<<YBpqq8Awc6^1tcUA>xl7<hP&sjz7=5{&57gb
z#4wNhC)$uQ4mT>2I;CDXUtr9T<bcW;8XGh&^OfKJcIJAyfCpkj%Z59VJw1+Bj?Z&n
z<oT$^_YliNN|+3!G2CHCCbA5F+OcfC2)z|ow^+WrAb-qQ_AW;t4%2YJebZ{%G@?TD
z@?R^Kg(z}th_nx5nLul`{yqAh?0p@)k7aZ3xl&hKYtgidJy}3wU-S<DbN+Q=+5IEm
zlYh_Ha%=o|#c>pAqa-hT%nvEy3=Bh{tVY}R<i2iOFCug!BUO$3);ENG<ZQus$h6LR
z4-ozhIZ2;8wy_e}P(ED)u{rBnp=-e@f#^|52%!y?R7iJ-y$vCDTj1_WeZJ**feF;E
zce|neg>WAqa^iRx8JaN~9LA9;W!5JBJkUa$wQUHsjlxVwRcPq=HE%tS5A%Yh73Nxb
ziGpNI#vs)dVbH`OYcy;XN-?fW<?F}B)9j27pP}*^9S&h=vAgiy&0mSw+rRGJ8nzi+
z+&wf3SiT`6x98qaV{z=+G_ZoH5Jf;Dv6ik$$PL*Wjd3Rpnb)sg@vhF;+4lLke^@b^
zt<38g9Y%B*x|4kBnr$9N7ab$E<1f1=+R%}dn`Gd=H~a1TeHuUh@u`6IX?(Bawz&31
zM(z`Sy=^#QDG{rqHjqYh>z2}Y$#UQH4&%tEf=n4DGt?v573*+v!#EsI$Rwn3LWZzz
z#X|sAMVO``#3>aHhiU)JwKT4m+HKt%)QZ>8T-iUlMpv-4NS2k6F`YP!3(Y_p9J-6p
ze6F-LoLE|Fq#B5cz4x{6GmdD&#aYS)M458eY{2he&cZkjOw-6TWLjAm$B}6oS$(Ba
zX&SUdt1E*}{N&BgXy!bh&yfs@Q6;oN#^SvgBNQ)GQ{K3e<$&=LPcXi&%*!)>|C_I=
z*DGzA`LFL+79aS<zyH6Hr#D#of>MonZkQA*D=sB|xDpsCi35OVWEW-MGgy3sk*gpn
z3VQ)jPPl7Wv^Vf}OGcZeevXN5ww2gWtMxgPptemwU)Z2=yDo%7Mt2cQ*&fH1I7day
zSvX&R%d(z19S-0L*M)!l^6&Yd{-<B_>tFpU=DXh=w9RqriCv07_!yJq`#ApK*B%qQ
zZmB!oqAS+4VcIA?2^o4i!d9?wIF2I+1(_yPGGjZDjlD`4w&;Re3c7E~z1tr#=q9*A
z9)?JUd!eLNq1Di432SA=8vW0ayftg+5V6!6T+bDei99%|T@lSZjITN5BggcB)XIE$
z;!tZiac!m!olY3kptvkXqIqL(3nGarPf?t0L4im}Xa^0M^y-098pmN`I*d%miMoVC
zlX-lnRu;sZ({#d`v9x8M@2#s~l_tBR)Kk6i+-6>jqUnJ15>uldF1-8dANlRy{bwEy
z6NhQ!AHE)0+L53B`~QJ_cm?C<tc!6zm$-jz#fUggsbXIHHT<xt5c9#Y*23AXA81s!
z+ZpSOx$fr*6~*?#bbqZBldG1EI+y_KmsxHo-X3RZ7<lvM4TsA+^ln|<3L_!X3uI#{
zkDRaXd2uv`{DLWefw?fV^7Z>?=KuILPmk{qj8d%A-nRq3_X0+r_}nNOAJh20V)-zP
zjOuUz)y!xw$R=SW_9q(4$7Wy>DF^Y|-z-M|QDfN~RtpK`;q;1T&OFb1fv9yv)DIWS
zZI5N?c}7g{{@p)_<q+yN`r1swx@#JI7|YG%R(SSk3O0voifAutbBocfVEwNX%O7bG
zdK>p2{k{;reOL29+J7UC^{x;de+q4|<1lcT6g_lJ(Q;t$9EB?rbSq$~l2XuoOKDiM
zaOTD_s|-WX>}n}AiDs@f3sylddU}teZ4O0z3nEHcjmu>vrHOGGsMb)FK_;*SrVJ@D
zOo_5YVKoy)@Y_Y%I-Oyv6xEv0gUWd%HAkx<(h!y15%txJKxsNkni6S9JQb+bXsyy}
zrL~H;Km$wPgH?efAw+JGznVe8IhVq@RzC9-MQDAXc;WFPoQA%xXqcUxFAMLV8-M$&
zg{nth{^Z}`exj8l$r_1<nufrBciG%>OTS;x*;3<%nfBsftZiH0VlSWff^*1?^ds4p
z(`aB5tH%l<l8|I~+y8d|polXZCVug=pYxmR6Xmy+AqCxXo3E0L<_nj(AxSzDr{O-L
znn7XVN|eb%WFMf^##(|3EE2k7cT@hO+>bwvfBE=S!1^@4xA6^`fz8;a|Jz;H&k6h8
ztqHV=j8BH@hQNYQpK-E>HLwdiELJqfUe#ggA$$&Q!de?^t$UYgTa!ZrAbL?d`l>Z*
zbxJX^h8=eG9D<KBAd;~biCId0BX5FRbR#9jW>)6~$%*4MF_N&!$quy?ayL8H?f|3?
z;$1IBf_64(0};lYLqxtgqY7h63|Sa549O|UNeW}m9LEE#IL$&QE9U{xfvgh``6W|+
zOV)t+y9kXCW?d}dk}w9=7CIR(y30@;W->m#L1aLiVC$LHuYB>d2_GlM(-BhKkMObf
zpf_na2O$g2ic^rC5XnxP^uGCST2<N)h=zp@nK#zfH`U^|gVuI{lAxK2MrXX4_2gM2
zG4TyOM@MwUSnJBW^CNRDpdG<Q*n`s#GeS5PZrv@X(CotH@@pW`+_AG!e${w>evkF8
zyg-QEc3R?AH|xHbAJr!NyW&{-Ki+5x+xLq?GhJuf#!We?zn(kl>~V~GKR{{kv^r`v
z9BvTAGfpO_fvmYZ7lo^4tx*`xI-;s9%hFK+<KA(O&(aZIn!BKQ=(6o@O9(0h<M2Qu
z<Lw!7B{6Z-1D_8^26Ykv*LQC;iI`8FQnR5sl|W#MIHQ0h7{&oFwQGcA(x?mxh7|Ke
ztZ}u9Fb18uAEY-&RSdEYJe*!rN@JEQt<yn@mq2lNgOreAWJS4}(mVkf@iAWW^iy(C
zPM@!ob!DC_Z+`J_(Zg%fi#MQC7;T7!-k-#QClH7WNQuI;Z@t&8<=G2)uy?|f3P>P!
zLQl|GtL|t|b2<ViuK$fHzUlOVq{K9&-8o9OK<<mU18Xb1JwLM46;UCmjn=KjrGsiT
zj6_oGJ+(b!wjveG62*+RK2aaP!q!G6W8s6i&kyN)R8YU~;^U7QY~S<vUSnA^t`1%U
zft<!TP}2TnV;Qf3Ch7un-kTfR(F%X;SPm^ACF{V8!z-3$;UZ5|jb!GBkLAinaqs2f
z_`Q5bEVCIZ7$VMGqY%-00qnzA?%7(8WAkXvO+&(DiM_t_RNc0Qk^bPZ?9x8UJ^B8_
ze*1HO7{^KC+I}mJeOtk!sG)?=de)&=V&b0ZFriJsbVpF-p3rP29h*jy0o58_N+2Rr
zVzr?1S*anik6;!iC^{%PY0Lp}@(^YowSnr$IG)h;ic(RlBvFpzi(t+bCrQKXipUF)
zv?DroFVbvtI}D@tGOY-fg~6RsBm~WqhI6PoV~zs3wR_=`TR|{hJCW1G>G%>~VD%D|
zJ~I|psyCA5Ffh=8hR__64h(6+7%@rc^aYo4<?$aMY4d`uh005cPQ3Z~Pw{->`1&W5
zRhX}Ev5J=kvlXm{J0x5F4IA?Les*{liVZY$m`907G<Qq?uDuz6Y!17<S-^%^+!|8K
zI1JI3@QQ4X78cjqJ#_od%S?3(O}nmxwUZmRJzg72d8RzAjJc5Xi1-OGn#7p_yE%}z
zQfR6$O}F!6zb!#;z1{Io<HtWf6|g>y?``<~DssQ!O5ZfMvkyY*!hJ?`Way3~&5Z%T
z3FoWZGeE(7)5*$+^wheqVBQ5Di#3Y1j;@j2#Vc~OzPUxS1hYnKPN_y7M^egEqT3^y
zdUryi=p>Pr`qqUsMMtHX(CUJ6;4lmfX$)~yp^HFv3;4dNgb_@KFpESdNDNVD(9{oh
zFp4m!G9-y)ImKq11O`>6JaVm3BuXiT1+OMj8pz{`F%6;fA=2qNfuvZi;yPk7u(YrX
z-vr;|f+u`<0iIwmnkdB+pMCxr<HL!wkC9wEQH<yZ>!=#lUc@TV`Pvd?>B;f_Hvw%=
zupd$z(HWz649&o6JX(|PG@^hKNAb`hP<SF;tk?EK)`xb(PHlz9%X1ta_Dyyxu=S+2
zH+;7Q&Ami-Z8cmJ&jZU`cskF78%*1PR%my3RFUqU#-4+JOyhUOF@65uM1R>eM4XNw
z=rHyBE?qDC#^WV-+DH70YIpXeC!rSa0XIVi>!>!fFk)EOg?tzg5!Q9ZT*#Vl6crQ1
z!rDGb?m9QOIU2=m5J?RA7<X{)0>=zYI`KLma9wd(`#cU40}xslirvpwY(mD)pdR$4
zoTBiJlXB3PQ-Yj8M7tgMMs3(MVfq~G2ZW>pr^8Dw=QEplE;z+&Kb#mM14&0#U@`E(
zhaf(LX2@|MRbhIyFu#4z`EusJ`{^$^z5WU3I#8S7*S?ObMG;zB7p^VHrkf+a-%H;P
zm)m^lcRO|3Ulg9=^$Ky|-N3fn8nbndLlnHnpoTHUFpSU|t=*^u_mi@;%H#FSQcLgh
zHbUR!cJR>NtVpV;K&=;=%_NjCqfn?%to4zRk%57-eWwlmA)DN7Z+Q6q_x(7=_ZQ2(
zD=(WViIQ~KU1a}sW4X_jP0)I09ChbI{<yKc6-#m&IHiH7=l8clju7sLKXfebMXR{m
z`^M_ujODJ;)enrtZ7*JNAIlPd=OS#%;!dZPjuX|4)yg*}Wz3<qRUWU;KS(Txx+ov)
z$sbE3-^cM*?B5E~oAccEW#o6n@veZ?bL%iA&@eqx!K6j8u;-1G{yf_X5z<jj+qI2f
zqJh}L*gzGmgfRSCQ5RIUJT1(Jk}BpcR)(zMOt=P$bI22^7R0tfs&W`kXfu*mk{ISo
zG!?o*kFXWH{cdh;GWAiELBKRIs4=RLG$bpMI$4b&gH|Z@hG*JVGIpwB<kpA71B)%N
z%6>CeFNjx?8{#9%xKrYjMgiYEW0H~S4cEEy<(JPa=L=<?dHMMR!*S&0o1Y=$1LN^E
z^R>`?MlM~F!wYT|p`S}+%bo8gNS_Axh6qj+6g7q3U(oV;oVrKx+@q1szUOUOaTKtI
zVZd9V)cfME?fXNu#<eVgj7^D~-r6=#qEMzuDCJD4XC4lTF*OEF2#(2$_cKFKmN)cU
zV%vdkD+g@UQ(#}uKaC&v_*B69G`_d7Z?dv4{YdZ0qH*}6#-fu5n6arcx6pJsc9HJ;
zrASr4m6Rq_Aq~NiZx(dKY(s3s){9kHIOc&-66G@EF05;4KIt$bsaNc46w4M{xw>(Y
z#*l^MVPH5$QAu+{vXXU-?`KXkqxwo3mE?(~oG}mj;;Z8;es=l^r-Lw#!sDY+)`jXR
z5-o9PV<e=W5scgu{9`&{WV{Rvri=`na^^TDvKZ=X@8%lS8grTPS!mV*qK}YV5I-O?
zp_(Y-T&U4qlK~kUO&agkimEXViN#lz(r`VX=>>Cni_5^P=_T*qe#x)?<A31Qix)h+
zJn{JD8UJ#{J{LS6up#J^UV(asdd2ICw@RYo(%AF|_D(HZA#pp9OTgW`vZ{Kx*DR&c
z3`!n$F<t4dIO++tv<rg98O-*i29Im1IHm!6&haOneMV(KH}U<QVsJZ{$42h3ZNe@X
zm!8Z;<<n_VEbE|rcPO<$+SAFsE9mYW@%4R0-XC{-R~%dK%!vlYRhpnKWNVlv8c?h;
zQ>e|!y)YZ{D%M{}NIIf2a2Q&2Q2PPWVrihNqpc$AN_)Cc>wx4;s}@=sBFK<B{j=g4
zbkWE2yErb&z(`||N<JxrLKa6iGlm*qp0ccz^2~UA04D5N7)g9FyyO?3ea650%fH}y
z`HFQu^Z3qKTc!B{_jt`HzVAmOG{qrc)I%KPO5@1HaT*!b$O2=Ib-{JOOJS}HEP|h*
zmKAX$%ShINECV?WjMKoo=l85C$ROk-JnA#GMb~uTz)@k9B{U>@;i%dZsSVIZ-oAUw
zfB2983xE03zvSQi^}ofRpLssdEH7T;X+ow4G#GoRSJYROGs+V#3s7Sn?yPSS?{VD^
znDjG3j~Uucd!v;lY_PY=#vX67$y`MNyvK@J2y&a%-t7-2u4_z`gff>afAjS}u(ZaI
zlxn)qS??3r%sPd<?L`h3VYaVvw2S)`#5a6w#c9zT7$B)r%F4!h_pRI9owz=h@yCi~
zPretsG;Hu<PfjCg7!Zj@4(@;Av0Pc(O0g5kmE0RAp}!gPFbxUS#H7bKHvJJ|S@OUz
zj-=@;Ruz&YMB({gC6?pV+3bF%81&|gs3BUnUf({XR50J04-vz%7r0`r?*+21$r;6y
z77f9^1jJeq>&9F!{LR<@m8ID~8O!da$6*|_XM2Aa%PV;5_pp3tEN|{~c-MitI{^Qw
zkNY^@elgYFxa)CTh0&GlU=jk^#f&S3(xNCCbg&TDP72~0MWYO9EP!Gy%m$jS;ce|s
zwnDDTx-5u0R=dtvSX6gIf<RFlk)yu7oSD+d%PAc8Qogx@Z>3U|&^Wrx3(LCDTH|<p
zAbaEd<uhPRY2x+Ek)OPpFw0oG_GYY6+t3RtM(u^q?8?DYP}5WRV>*f~Bcm#(Nf`!X
z9EEWj7^h6}BiHM~<NHUljbs~1&6pf?F@#2%;GiR)z5JZHTxeZZK+HI^V1QUAdt+jx
zcG36k-ql*roZ#Ulmv`^@ufKiIo8v29z4<AJ7hjMLCwOtdG*Rjr5+nw#)2EFEmn+B&
z#NH&*QM0lYokHVIRYO|ROV4YgDc`&5M1SGKQXRn7(DlA6-cDmJBolh`T@`%O*b?Qs
zUiq8f{yk6EGlMF%HQK&-`2PQJ5RH&6<%w1oR+))*e#mxgpGbjHR>YlJPJt$CU0Oh4
zGh~pxY4B<M*vF>=)~E4<j;&l3mHoQ~b60YgF8RG31@9lzU!e;*CJic`h*Ij-Ug(`p
zj{{M3*JAN&B~Vzk)}Rci_T@b~v}TRk3e}C(3r!9Im(Ly=867&mb>htkdY^~>Mlz)>
zL18@Rk!cut7>_)hU^)oPY`7I_@-9LiNHV?izuEe<gA6p!R8x`(8D-E+PB~me<Yw`0
z?x7D*7FsV1Y;Ip$QN{^z_9}jRCk2vIq>(kK0gXoW6_*g2Y+kx}Zl%<P_m@Wwr;!)K
ziSk^r=J09(70lPpop;<SSPf{ng_-uYscsr8?%O@R%?^l-Fwk>18b=6cI+k<qy4>Vg
z!-0mqOB65Vy`aQCC-x3a@3^{$^Ws28w|36&-T3y4$_BpQC?_Az@AxFOMOQI(So2LB
ze(P(0?4x(tH|kg~NlMoiF?Xul-tpb_?xY`BM0&S3=r<xt@koBV#w$iR=(J|EQm_r>
zuqT3osP=-U_2(K(xdwt^$Rsy1hGavt(KUn-H1wsKcTJ0x5Io-QjW9~$G>*J{bwn9Z
zSDrPD32toE9$$W=)CH;{ku*gz*23L#(nOLqumVZYBs2|*Y4flh59CRw8E-3t3#tjt
z6Ou9}F|Pz@(x_4+_9Zb;vCUdrL3(TkB|=%&nfI3`eyPI4ix*tqU#X>H?HC1_)H|5o
z3sBw&m!qH$B2k3iP+$>@@p#!PBYj5T9*OQoNoo<tfv%7nl1a8B@a?(YNyV*o6p7uT
zcKZXBp!_~Amk{^u1ttHEINo%0cEFg#b3-_I)7j|4;oEn@<D;wZGz{+M@rSgCk9qvz
zv20tLotWpo$8#cSP}@biUL}9(v0Q=L-M6+oemk}Woiz@BSu^HBtN-n>ERslh1WkJ^
zEAp=s%i2-kohHxr_$`DaP#Ie;u_G?5>q>QPn|lYpn?gk2E5(9F-w|)y26)>K;_6qs
zkLCP(W7)UHQNPE{mw0SPZb_eiVLA|~!0(FXt=jfM{_qiw0K(nk_{Mbb-d2;H)v)i$
z;&(Yw+=!%!l!PHgL#~?A!~zA+#!?GW@K!5T12q-qNV>}<gm$U-wWEb&<c7WOOO5o#
zR{sV-2#2qA4I~$Vw%09T^2iyE)5K@5UvR#_JeMeTbPBK&W<-AXy;4UM#b*oCA8)Zg
z=ak6Ff<~ws$fOi;xzrU`K~e}gCI<-)JYr7L%wb5Rj{e`ws2ZU!8-Wm-Rl8wD501DO
zN=r%wmTG7|a5}wVIG)hyKphh9($6ma+~C_xZ9pL??bg9x(Qx(7JlL8w+x0cS(RFXl
zqix9jZD2#R?!mgPiv7&v{qyWA=N;;)+6_sm+^s=(Zxe{EzQ>ASzI3FOgD55fU0W=k
zCY>R3GveA~__if|<Wc--{By>q0@kPTgAU(#>22S`dY4ZLLI4PrM)bfdFvZeEriH^D
z1H+I=qcTp>MaVj^TH&b`R7X??Ocz8OZFS15T-PYDl(k~XvE+DL!5U&gJ-e>Yoaf5(
z)5?gD#2IQ~aNCot<~I$PQm-uQGsl-F<TP;p>TjVrua6VQ!^CHA9!OI|vvD|HKxS%v
z#Oi^zP6&;p3rV_IA#q)%e8Ad3Sx=}JjDn`jFl452U>cN#M(gyrwicGP_5`}q>9TOO
zGs&Ek28uvBPR#2}t=1va%F9Cr8AviBveIa*wz677#?SC9JYHsONc_cL|B_QaA~JG%
z83nj<c)-$uRSz^-shoqRU?Dso+FqeGhWDgd;?UuDO_?YP+Fk%NT2wYvqe~vz_u)iC
z>c@@f5?Md&1Oh9PS{3$$#`^|*-&vwx?LeCQe!%_iu_cAt&v;wkc@vese{Rr9FY}c)
zWL^%rZ~WV49ki<Udy8d%;$ws6_c$I?VlV|u{h(&h%h1&t)eTY5J#iM4%rGSKH{!Ug
zl}jx^CQBeZM4Y9pfkJ3oLbaWI1ZzQujd{1SmNO|$q+z6bMXVw^l9NWK8Sv)RTAA0G
zd7ZhIh0EKoSlog_EY3R5xD}=m<|!j>fc%UrEcHlYWE!*|>I=B_qGbXVMx8KItWH6F
z5zvva%g!S?4_wO1YSC{jwSun{DdEm!Q2?6TmE44p<37`(lPpYQCad7;k!aJqqBwb+
z$cH0MjHO-q?bCZ&F8rH+_m`Yro+z#I)1SW~<$=FD4=k2gS7Ti(%Q{nQ!CC{4B6Oe`
zZjPc!2h6qtz^$<CUTJmd2+3YNcA;3Mnv;BD6U^QU%bOpT?}v|`9q&aI_Bg&x!=RuE
zVB)dP`~elEI5=_F6$aVqChXrOkd$$_k?r7Mwj-~qx9@oMk8J!iV>$Mm-Ud-e2@&D(
zM;6P;os5zQ?Wzf(xlrryXAsL#6><S<Mp;V~0*J!AAe2_97Szm5yYqI~-A}>&=k|Bj
z1^T}^2RjC*-+TVaSeAYL2sBr?mN|I3d;hy5_>Vo7zblSCL5pT?BJ}7WOb?nFGJJba
z&MA?zFpPn!RZlF{C@YXab6=X9vo0m19a=YB5W+pjTQ7(@bGa}th3kAF9i$f<8*X9p
zBa%p6f}$3ql)|!HL5w_QzI=YqrK~8*lm?DD^YSqA*^3EbL1g88S!iXXwjmIqIf}jN
zN+vOQ4smdab4H*}2Qr~eH>fkF(D4J|S}Nyd;dwrDaO;|}4U<iO1d11uD%0V}G)_$8
zNEN45&|&Vnl^I8v%R*B_4?q??B`RQb<;!<Z6s`R1uYb;8e*Rbd;?>U>9!8offA?sV
zT4^%wJ5^H1$tdj#T!6~PHV}nG3XQRzTnFMmn(gbllE@?vy)cYU&*rYu6|u1)siWAo
z#ZcsOw$^ZD+;Vo}JR#5<36<x02@?pvq0WByFr2j`#G(-Gnh3mHi-x_^_nxlLyEUCY
zpbKuf{-^QhFg_KqK8+u4^mS#^KI$cbNN;+M6Ps*UTZcA=gKi!FAcEQ6DUJpAc69RI
zd5K~{vkI-^o}#O>mObe)w}9%`inrDm-bSf4pt{v4rLwMteD`@}>ne6Ad#jC73e~U7
z*0nZz7b}F+6-}|R<gqdsq;$obfKQQ}O6#xRNp%tjV3n{N#|`ls6fEh;G9no(wFXtq
z<N4$as?^$WsUb-2HDJPQ2Z-p>Y<A^p;I7;8v0u#=o!@?_a8`159$p+6YJ%n%6`@f|
z#jJ*)dAk*YL~qiUH^JxFaO8t;d`mL>1nFHSAq3vLaPYQy_3mQad%tyzKTssTPh9WX
zGt#f|T|;P_J>NNH-M@n(<hPuB`;Fh;Px|vpEhGWYiM^YD_dP$_@|E8@=iBdhMp*xj
z;+v_u(Ycm;!BJ#8ux{AMZ^W^0y}qrt7nq3F8nsCzH4=2<i1RoIyLAit*02`3OQF|c
zz3cC6<;-}i(1NJ7gf5**#C)r_Ph&NHy7r4CB@LA{tUYcUq1+bpZwD<%%<UuzC1H|D
zy}*(*P-o^)o3X5orBrIWx%*j;&)Ce+Zr$zPqFB@CbL;C(&MA_1ah!y?_4qON|92Fi
zcwxx!@Ngg{p_W3aD~3@@oIOmoIKI2r+Z@*q3rK8<?+4x7c%pUJs3;Q0>)miIx7WU{
zt6Ln(H<D8J=WhSxa+@prTmGKJ^}{&cqO`xSfAGL>J~Qrt+!Ucc$n`(=@h^yFx7esf
zwxVC(SAJx%ynT;tekMur9si7CdAq(hbWKO8ZLz%lVSDQKI@`EI^~G)SJM@&_pPBd2
zi^H^kM=ai6<Gc4{?`y;M+&`mO{#|jr-JAU<?s?pUd&Kblp4{g^zkKFfYp|opyQ{K_
zP;2WsRp^H{99j!)rjE|7wFD}%8nw0_8(U8AI|^*OW!8*ZD=vY)s;yAUN+NXBrku%1
zQI*hv&4#9iq;{87-M(>1O?2d-2w{W}nzkE?Sw&D4YBQ|Bd7il}E5$0QRdQ{#6c=`n
zfu7Uu^J#-*tB{8rAx7$=^fe-Fmm7%x8-{y9)R~SWs$B-87EBehgq!R|-_VlUa;b1@
z-=+As3DAGt3W#xkW0JJi5DgQI%^~}i``!X!UnjQxQEvYljpl7FzCC~c_l~+VYd7<a
z@7$Z^?sIRi97VFY(B2DGJ)Y&3KWymlpe}A}#*ZYHKaD?ld@5jl8b9pV)_60Nfa0q_
z#mETBVZcc=Sc7NnQ@}`LUCxn6HN%S0w1U7|R;smd)v$`0GtU>C%I7lS!1?)^ahN!r
zUQlXfsVlZFL2FTHtukL1mKD4y%enBpK66STlAfdsOZUUsx>V-N%=!7uuOENKrLH7(
zrXeh?$5F|Hk%oXEKRhhtp>Vl;iQ8+MzuHA3Rq4rbMFa*JAPr2z*de<i+OFC{bD$aT
zpJ$$)FT8txkD#b1htmu4IB;ENv^9pQ;8fHa$(<3!(}0OGP7}CM^2%D^x~#D|ir}(f
zdWjAhFzGY-U<}6-%e$53Y2i9QQJeF48SwnT@L~$vbpz4NkjJ13HP|-WKwia$97&Jf
zC7ZqQ*tVodgK|QY)0!Akg9ae!wjsJ0E`~7n1okEn>t><L)i>kr>fNE({XyL|k>9b;
ze%J*LRETo#b|N1<&52IY^Z7y>2VUfat7BaR_$GS!@s2pyLWtx;YH@ru0<NzD$(@8D
zH7IvJG&Zdh7sXM$ffYW8<I;>O?8#5=^p%Z{qghBfF$|e?Y0S$^t1YaerBG|7m7uRC
zX{6Hd){D5UVY$WJ@<^`d>ltu{X`);nnb#FfDMY3z)Hbs&&;7j<$A=YFVV#%IJM|5P
zmkCO6m#P*eOK(~0k(??i6^2n5M_87|T8%HC&z#R^9<R?N9mqN`y`0EtU~ZMBg@c#)
z4y7X1NakGEg<_T04-cUUQ(-B_)1^_p(x}X3W?mQ4aYB>uLY$_FV}6M}8<(d?&QD)q
z2w&SzAsrb{V|QN)?H}!qRK62P=#7Bw!202#*F6&4QW`@NQV!z`_t4c5mEQ69yGYog
zN$!4b^a1O7gFsNEw}UmX8DQL`|9^iW>%)1x6}@}O#=rS^D5W5#QH+)-bL{lVKLG9Z
zr#}9<u`E<{2&AFz*FFCzV_ArXkqTrPaZQ*`e>SlU5i_+Kq;Ou|Ky%x}#G(ih2gvQY
zTRiLjs~fca@cWIp=}+ldz<x;o!(!R`0ed@kZ|~W@@%SHiEdQ=J4ku^1quUZWuQV8<
zSn$m~xdEC3t+f<oF{dV>dZCt;7jhz{#5`X~HFG+hXzrm6xh~imvJwu=%fh-e>f$V3
zcur4DtucmNR+q!*6TkvVdA(loVPF^&r4^QS;glX2(#WgRiNi49-bh1ZqOlwb?uAmg
z&Ef9&Nq0)zTvZbpB%A?<WTa%|Ay8+}mz82GfA{O(u$G0j7Mdg~%5co2l(3XSZ>-B>
zBm*yvby-51M1+S|FTsUYU|t%_yr4<YoOXI~NOBoTS9CPeDRG$}`G>~~Y08iW*74_5
zm)@v0inmS!jCEK#`YI$kZgKogiH^?ljrl($h9Ri=p~=-Ly=?k*{lN}!Gf?omLNs&+
z@$QIqi^u3!S^x;{OhNvs*}F-;+)NvKp>At#Z3S*14MVuVw~CwH$#Z-fe@^350qfKF
z!N#s%LWu8XNxT?O-R9TOCP;H|#YrsQzzYxF6UnWkvK2jyohQGRg|!(gb+_T2*DEfC
z+FpiOU@4?LkaC75wALayZZ(u9T4lb@Skn-L?~Z<S=y6Js1V-*0aNJTVtr*9{iR1^Y
z&16jhO}B9T8Fe5JncAYTk@A(6rhU`U-lg>3#frsy^DA;R38Diz#o_NdFR(g~PgkC=
zGuKwgni(W0I@W-}$vvY{AWOinPs7A&l}csUa?H-}H@PW)#POwpRU|3n5EPKng_;!A
z1<{PR%JcadGQ2v5sA~wD(_14IybI8F8rcR{7mdz-be8UVz8=Zt4J`+vA?Ds&V2E`M
zFG;<7+cgg23&qCQanIX^NWZ@OeC)}#^zNJ87q#xjePVmx`Mo9EzW;f97oclRSWrRa
zE}{O|t5?9*l2ePr;fwJ^M#wF?;MmrgOspQJ4Azm1I3WNnUDL?ygE%fNw8>gCsu{Tn
zVoo#N1&Oy|tx?vMRsz+~sMJy^b0H_i{DQTDStR4qmBM{TP#Se@)Dl1A5Qo&+A!|aD
zApucj%!=mBGOW~MpqIF%HpTVuZcIV{l5~0vimHZ9cTPb+a^v0mN18bAmW8#f_>gxb
zWND37K`aGoqie^gLlR|7nd5X|=<7uwRV54#q`RsqI1Tp@Bv)~c2c?z_ZJ;G+3iDzt
zwejxyH8MQ#Ccl7|x=93giEP)e>0+^TtL;G0&;|H!=!n)@pgcA7iGtG5kD1><QUcKx
zCk|5ae%j-geLdOBT>C+}AHsL=&$o+E+qkC*%dStd{o8j3ztib%rX19F-4^<1_}CA<
zG5(pcY>jG-EU|uvf^J{)e}u7YZaY&&3EKUiODsc#MQe?uiX;uqXSaRd6uTj<#kYGR
zx0AFOwxLsZ>8Tqk?|w%Wp(D-v-+Y5<xX%^)j(vRl#5*M8!|~&f<?o7PWkiB1*5WFK
z)})cV>k;+~x9!Q%6)>!YX3|<$N^`0h)mtyX6iO{fP2u8~QlNRN;k98UWI)WF<vLSJ
z>z|=g*Oh^-;MwnYq=(eCQrALVSC(2?`~Ar|k-J`1>dc0mGueX_nQ|fJm2lPsA@?R|
zunhLZ1K$f-XjkNN$F0#!cznLFw1ww+#xcfeLL?X*)k25SyfXr<Vy)pdn(;##s6B_?
zinP_~+JP03p=&?JmmRV&q{L|$qqtW_w8YI=)|Doe_w##9C(`hmt(LQA7;YV{v?b45
z2EE0x>}%!DWav3@m>`5B`j#7ucc;d!`sJ`G+qps@_vCI*-g4jgRJZWi)(zj+h}Pu0
zlWVl2|N85ApyB%QuK(?ag=~j#H{8&IB7yM#1i$*T8=neTpT-Y2dY4R4vL_}P`26%4
ziiH(&=T}3*C3EpgF?eMIB56-}c9^>fR*mMB>rz-9W_Q+BSW98%39M4u=VVK;&J5E?
z9x}2<QdZYOD=Vv2O07IUo*87|l!SqX8ZbV94y17)Nv4(1x0vUdWhs<my!reK7!vmQ
zkD=>f4R3{7XL1@j9R`|5r#KDI=sJRT0dfiA<R-Ey?x>`$GtwB7G95<J;HVn!pD)x_
z`R%(i*Lvk_g@c4W@!Vn)ADBAY$5H75wt3>!_`uus%3{v2^;ELrZZvJZhy@~1!1A^t
zDRVkZp$Fobb(mRGLUqA$-o1N^426d`zd&4ZlR(Z{=p2}LZo9U9bA@gev~7HTqiMG0
zA*Pvg<We4@&mle4A7#C$l@zZjI{ZCJ5prZhe(N&(H#$KZ5(*$~o#nQtz<=QAbQ<~a
z?;xgVvxX4`Q^!KgnoUstpYb+-Vk3$S@lBE^Mj83y;j<v$tSfCTv{rD<pqVR;W+kBU
zq9~i3@m7fT58}AEup%tA1ckQ%wNaX)Ss3m*=B1P<I@gM|ZbDEh%X!ANF|0;S0X2`M
zBZ?;0Q3bUGl5)8g*12#x966{W3!#CR8%g6Je>`L)C$8tODa8?ajMU0r#yplNbO&`J
z;z&Jg$#c@s78#tRaX|m|Z@y+}mG>e@7RG4|6ie#|Ef>-;qn3kq)V2BsO-#eYFdS&&
zJUu-|f;?m_S#;dhdLy-AzJjkL6<)j;SQo(ylx&cUWig(wE5G^r9cg^Q@$?yqBf^oc
zyiA<#YfqSsO<?&}{1tXxxPBsPEwln9DOMV_-E{ukth0;4jWdt~X>U`6a$eo%$8Jgi
zA0B+gQSJ9v0lXt=zCCY0EJVj+<IV00Tiav%ug#tFe5K~ZFm?)A`iO(~8{=OR%K!{P
z^S;#!FI$PG&eq!BV_AM=u}mNyEDC0cUcCA<iDig`a;-+r8A&OM5bajL(#<5L*MtMl
zllpZuHW7CBEzaq&9H@?s#ueI?aeueN=y%nkzS&UP53VAelKJhbm-FL}<=+*@Sp=$4
zmzAY1Ab~zw+`#JZ-IH4}s#mUa4W%@JN}*C&%0gNs1kP2-!+_UqFR!#^Wo?zx8t3yl
zP=G!H&MMA0WIPA`Hf`?2ZY=Y{GS95b%yVsATFk?792prHat?j2Au~-AO&2_^4C6wp
z3Dq?mtR)(r+uROxPvl>z7ir>mhBu=%pc-$VzUES9-d)a2)5y!$Z?Lt|){3<-EvjxL
zw{Q`pV5O2!reQ*;G&inw#>}Y}4zA5BNiv$W=lzXoNE}DuvxgJwMR_Wfwl<m=*JWj{
zjo(}gX*lu4Pfmzy$ArdkZ<tqZn`6I@grP<!Ywr!9C<@*5&YaLR#B-0|SR1wLUfp)Y
z?Hv`QyBOYs8}~-@&G;mWWk$-1q+X!>)}Ac;uQv`u+^e3o$LsXZwPokcLpnuB4Cz)l
zi>CO8D6voD$2~q3us)3+c=Q4mnNdbwj}Igjnw>c7g(o|A)Ivh?maN9o6&?I2g1C8D
zxaU~l)>2sAnFUr~S#8CvhCpSn-&=P#+pww%<djfLoFAWQwQ`UFTtg5&D{E3d7Xk&<
z7ts)p4P?P|Jn-V>OHL0D$dD<_V8&5K4#xrW8i)6qNP0k%@&4&c%2h}jt}eAopmk_8
zTc{Z7X9gyYGLmy<924k*nep!Y#M%mrg#>GpgpL#A;UUCY*A;0AsmeTGq6c?`q|vp~
z=CC#_CtRR3XRRT;Xq`-+e2DiSj~u5=RwI$fHi8SyU@4h-RhBZ7B;mfI97uWupRp$0
z-nev)Gu=0#TN|hUULt`p>nMw1K=VK=3#|qeSrCTab&(Bz?vC|tx$O>V!qz$3w9{~U
z7dXg&)@cT@83%>*hV1B{>TbvVpgX%YrzGC>TP{g=&1WaDY~D2iG-(`+;BM3Y_{Mf<
znaCV<<jwSggwUii%atn^nt$jx1nk}S?cRIu2!0sHa79^drT9wq3HMB?Yuxu#NN!A;
zuu}0-Df2?BEnI49L)S)4!ZVeJF*>&*5{IQ;42^DXMsw!nt2g8as2Uwp$wtOGjDif&
zg-)IsL~$zSm23G5X>dq?O%p*o<;}cnfyAWS4+O!>5>yn`jHmD>GBdn2+G32786}4W
zd2ONFhqC4wk%3$niZvYJe7RDrqB-bJORcOmy5nXM$6`wIf!8lya+(zU5(%qhN}ajP
zE43w@3~ahSkrSH*Id5|{+AO)FSdni*K&J+Z^#kY4<N~7G8&>KC@i8LM@1{jhsKrAp
z+k}1y-8#zg{|zY;&RV-iAk{+eP_uTl<A)U8?~3DF!D@T;yE)k(yZzg|%!m)9hgXDi
zVe19)Z8iIN;~4%0vD~};g6xN*t-J3(qF8nb>bi6^V5=nx1pGP0@=d!^B#<_%uT(4X
z-rW^Wq!%-YR$Nd<1-D*ome_YYxjNQ_4I$m*D3BF~l_+}aKM~8@pZCA>ZEo^s8Oz^{
z<IkoSWJ#Ddu3Q5VWEvv<Nq6_iZ|}+Rcaq{hmRczob8eJcXr<EpMeiXyu1-#g)|6JG
zFq=g=Ul-POg;Ikp(F!#i1@N56IR&dBrS5{ceT0}VFJ7ED4g<PeS!-jo%s7P8srEkh
zG!CTUz&n{kI>cWSuP|XiGpDr5QkTf}F{-U3f)b2`4plSo7PbaeK)QSIQX27(5}IPo
zS(XJiIGrk97pgyVUT2nCL#*A6;dBV}+2z`EdqX6Vbs}pfYvwo(97ko!6;FmR(R^WP
z#&xY+*225TNAmc<t5-ikd_tIdRrjWQWmw-s(>@or2HG~-=IEXJg;NQHli#pS%C??q
zG@jhq(6vF$H`*qiArZ@phDbX%S&dEStF~}^Pd)kjUAcAt$G)vdcQy=-t?$Z_-JR#l
zm1)SlIOW^Cbq{L!mUw>}Kl<^hfc0toz$21{L}HN4!*C#{5Ik%KPi00dA>=^G=nWcf
zy?6!TR0odQ)`2Y)U{MbTJr<H%2$Xa8BlH9<K27TGIX;J&`*awnEi7#t(V#Q7$49EA
z8jHNBGL93+hsyDIM02M43EUYqF;0o*Tj!=CG9px-FORh4z)R^c<3=ysG>R4QhO{fL
za7vJpG7JMv3g&RW&dhbC<T3Q2L?I94>A-clQfeXFgj!==XT}`bR!DqyX+SYN1?sCb
zr**h^Acqz8O!9##kBoUBsgXRBWCTLAy_Ced3Z)i2uehzC2UI5@N7C$Dw|7H5q`o<A
z2iz^|j0DPYyv4yc<#16dwGw_%5*<YZG!kX?*j$CQci3)+(nNHaw|S?~9c5ta?mE?s
zkWbq*LcWn0-;G@mem}&~Klk<?9~Q4%wr|-*FWCf<`FJ<kt>c9-Nai2|FY}QkO6hji
zixsSPI#(}3M+|l`e!@A*`r-G3ICfZ6D85oycgjt*iZms)I9%omUK?IpAVo@t^qZs2
zX>QEYn6hBKcPX3Lw1D>HIi<|Qc;KJ|*S1nxSl1_MjAKJ`*m_Ui4{9`;uPo(}WJl7N
zy9l&&2fZjv?}_IA6LUu;I?tX`bWyugPireOC_`47#jDg>Ng6sJQdd$x^fq9g7nZv6
z^36*`gtDw*Kw(?gRS-`k5)Y>%$3dZm@q%ckw#u?B{ZJFh?<T5?d(enJINWt~=4Qjq
zeWGYP0^%mx+M_~*&3ZfGyP)pvSh<y(JB+Fq(f2qOLAC?8Z%1VMYzp+K+vcj;1$p-v
z|2=UG``RGx-w4NBQoPnePD08#3i37RkK5t?qoyAHOJZ5s_fPljMLd`>KcZNsAK8LB
z6Z3cb^PfvB-)~a8RutZKcw$dZ5{F`wK0UT|ed7Vi)-&9YFA>K(?!vhfhj+XMY8(yf
z>K_`*-<;I9de)yoEPq!V4=0i+t<T@ZR;={7(2E$~+LPJxqOI^5NZKNe#a3#oA;{i~
z7MtjV_M)=2&}dZ$TMKnw*uYO-!|108z=Y-~QmX3S;PSrqt0t!7iFA6PoP~KUXcdN>
zB7f?rqah~-8FAjzOfVLtA8VsG*<Fw=<SkSkwc6K>py#^AHE|2q^Ckk4Fod-i<Tk$p
zF^{4gLbV#KihAX`uB^RC-P}05dJ!iTzwVU7aQe)F$nMnPF)L#diW({-%`3%LmbEc2
zh3E4lV{Np$kVpuJ*sCn)zud>D_d*dt&)w;LU2Gj;UoW?3Z5oo??aA9bipe1O9qwzp
zZlasZVc(PQ(_&;7aZlnPDN0?xE4SXBxY2pJMet_$bvMOv7v_1vt#LZ&UijMM_ahDA
zPvf6IJ{7P&jURM4hD+n&F!FjhaLT<CS`5!4`tXYP&z0+HjK>j4iB>m2sUx(&&UD(H
znbX`sG@J$R98)$uhizDb)hkceGv~{N`Eud>)mQks^56fPf6J7|5NDrf&d-lL&ll$T
zLT)cn@7udOt(Kqwh;TfP$lx4K2^8ifHo{3ktx?v(GS8?x@8+53%gocYLR%Tf3mGH1
zb7khtg=#ZS<-)=PD-Xko)eqzfX@HzF51+juwZi53LM_g`R^GKU>!n5!Nu0V?6rtq8
zdA-oyf5}%*U-8w`SG->?0psqwSY3tI63B>1MKQcNRtzsSvd<KUNTDH}0#T|{3UMGQ
zEAJn_<}eDU=|v<p+LMGZ4DNfa?!c>%XYbtge#n&$LskLN(6oxeWoQWjAOuQ-ZI2B*
za4p?qXNaOz<Y<C<2*0a)PuOp;WQWizYtXq1|MVml_T=H4WCg$bRSB)M>$;G{I@%;?
zhsMXf(K?QaaTqhFapahl<VvQIhKWNKzJ9)PUMl%`KvNE#n(euSoT@{_?p++abiEB4
zHHbu*7bK0K&RiF=DQ+3eISgZ1e&>;&e)dzu;rZ=5iZ$k*=xNufNOwunrW28dL>@De
z2Xd{fO(;!BqP#c_OgTd{)_K8f<?(6e{qxLqEe!rZ85%1qvtKD56pKJ>0B($O;FL1e
z9Ix<@N2Y_4+A~_DuYW6-^-8M?vl>HkQVuPe)<PV2t{1*K|Aw#6U-RzzE_7ftvz91~
zwvw=R)0WciHk3RwKRu$;z{}GEs*NHG;!3M<nXg>eISSEh;d(yD!I4uWy}jd|fVw5N
zcd_V^|Duv1MKMYG9Tt~<U`mK)q=nf;>aph958EE=LGZ5PXuBIV$X3n@=P2pxL|oqc
zMEEYD|3mdKWM7QFohWZPe1F;6sf@Z)mHc-esz0Xj&yD4jGH%iNZ&3w_L(q5Z$scnp
zFTS#R$N<FQs4?l0Asv4Fv8;LD;{@yI4B2xlB&dT*Ag|&a>ugkg(M>lj(8qfkwl$M_
z0Z}*cBi|0&F(cWvd3JHTe`*r<DoOvtehTg-tRH(U3v672!xUPl$E;*DG~V)v$^-9g
z!CGS&M~0L;RrkX^Ij&TqRj5u!BN?UZuWghy$3dZCoUa$wWueweeSW4sJ@fNdpL6`l
zgyWP_dHe2(wKbMjN#<z32hx-EXbdN&Y2wwZSBx)Skd8-88*434;h+QNg;teXBG#G}
zo}S+FbasSaAQHw$&7v5z={#NQ6%dL!qaM+GVmnnF#(`lH)<<V)M%Do{<MDj%E{4iU
z2PVm67g}?k<}0OMd77Q4`H|;kCM6|l<~kReLu>ImhIegVK}^7mr*~gthmkMFpOQVJ
z-e`5@x(M@iWm#qt5v#3L(8wXToKwUR4a$7{IameK=k(?_E-I+EPS*r+r?oD8E)h0D
zFf_JYd>i5D+3*%n&Lq+`DhYa~3n~@giuy!>>DJiz!^ZZU00_jKkUN^<R^y7In=zLa
z#Ar4}3`5_CK8-)8@u`6IX?*`<>(<_W%a{{m4q@=*QI#J>Ip%?93B^ej>6QMzFzxiP
z*kB=YpXOV~p;!Jpl1H>7j<#--vQryst+d)$mlaCkG>ja_1Em%QGtx9->l~U@uothO
zi;ff8D>*BpWQG(OQtH`sqHh$n?rtB?=Y_}f!de|Mqq4BMq*PxiJ<WH*X(Y4Kv|<us
zi&>SFg`^`|VbhN?hf*7tx&%@x#lg;+)2y-7m2$oCcs=v}dM1pvg+|X&+XnRy#1hiG
z?QY#wP@?d}mH<}p_5tVI-D$PaQj27KKSRbf+Yw1Usf<EW--<hZh~Ayua@_4F=FRq;
z?KQXGd2F&kP!w<0>Ko<gCKlcMQg@r{zTxd{$KNr9<wKFzz3+9``TXFM`SyLI^G@pU
z?S9~kxBAGE<t>io_Sthz3^|eeoOb|KL`Jf#C?WovRAWx|)@}gOw-@zq#BnFNNWZuJ
zM-<TLeBw$b#C*k^Lmo&755plIJ{>5nGPK5OmFBGj!ET67Aare1yF*ZS*aEUTsmFIp
zQM|6TQp~u_EAw2a%@K*1tzIa;#sR(4<p5R>q^3(yg`)W#Nv?}x9Z?u-7ZfvUE1^qb
zn~1rywnl9$=fyZLSFURbN`xq_#d;Tr#~w625fu+eek%o25I=P>>g_BUDdAp<_2{~b
z`hF<d3lwZ*pIeaZvD&Zg2G0q4LGSNwF~8f_$H5WNTQc6?-1nQ~cb~66!ZiwG654>#
zZ1uYsy#4&nDZC>Axh33ez3T5;Uv9se^}@Q`J?<MQm5*lpH^y?DS30s_-`KxvPyTpg
zx%QG<u>u$i3H9`65z7&;^kTAxSUqw_Ig)S3*=-`+UfJna>K3GO_q}!+DtkD7FhBpv
zv3;-m&+v^sc^f~<SpH^iEimN7IP^H~YvdrxArA~vS?Rf?YBVbE_hj$aO@e;+K^zCS
ztD7gpdXB~#Qfi^p8e-?Avd#;K&oVC$N6d^Ol`^f&rC{BGTzr$|@O>VJ*-2uWMuu@9
z4VkItUaWxurL{t{gq!vEU6|J^Yk^S;bc2FTBWk15Rjct%KA=8?v5*jkK^oEy=CO9E
zX51>JhD#%EOxCqliF3qKR_5Y7&6T;VR5MZzq4rv%xbGg$j<<ZU-z|5nHLPjqx`tx|
zX0diy*N}}!k-fm$w)pGlO8MYa6_1TLr|00ks6{+p`$MgLIM@5J<!&!$|IB-~>wc?k
z_4Q~&Wp3~LT_N;ON7S|s^qp94r@TGKmfMOIXP!ng?Q`;m()?Cl|7rZF$EO0;r}6!c
zZ>%Jegrten@yJ6yV5QJpkP1nJNd{h~O3u#U8EuNyE>sw{dKgOTU6E_8!F?w-o>D?E
zv9j_ooj5#9jK>3c8mS^w8t3bkHK0ShJUt+#@L&GsZ%7EkFz_%P_~PLu`II=NBM;M&
zAt!1rnAb>l)tL^7HVbA(ZWTm$7zW%L*@8Pix5CR8FVH;lw}1a@e*5;3OAGz0Tz*To
z1KB3d{>X)EbW{WikOrrmRxA(b2*(#khC}AJ@6KGBaV?GJ%8T&@rB$xAprw(jGt~ns
z#^t(F%z18&bA9Gq7RG#F(A2L1m?>^4e(>Pqrv_SRcA>UJYtcQP7o*e_a^cMvBZpP_
zo39g+Mqa)643%Su{!$?sx>0&az0lORWZ7gyZd+BdjC6M<0cx!{f~2%tLbGko{n#0r
zX1#0Z4cyQ!;wq}aDm2|G2|*d~u1OHdn5Gb#{6j2%BMI5`0ls5`-zsi^(=57^hcPo8
z$9T_vkoc%_<!{C@LKfxWaOA~!B(;WmB+MZS3lAxw-6=>-sI7Lrnz)8x1Qi+ZeH>SF
z7F$7`bWp})W;%}Om_e20T7yhn8}s`|=Eq0=Pk;5_@$z`$Z~x=pAkFy27oWue_;lp`
z^33ykVHh$wEA9<9P<3*GaZ*w?R9BiTAcaG+Kq_cs8iW_8k-z`)D}MX_kzf7xE!VkG
z^m9}ym-;nlE<Ev$i(QyIg2OQ?D^-ot_>yrL8S;orqBJ;PW}en7>k^t-X*{x&!o04C
zB*YVDP@D7B<C&#8kE_$Xgodhh4Y*RGgIN_f#e9e<sKT<W%$E!D2D}!YAD?N>DaA;`
zh11K0;o*U$9(nrvfs{|YJiWnuqEtW~leQPFRnjfd?DJqlSBUlkvJC7fN|d%V!tMK;
zXo}lGT17S{0xsQIiAdn-v}bWhnn)5I<ZT6rMukKSD}PX2-Iu8DbM6n);9k()s1M7!
zkdu&;-#{xn>f<9Cf23G8oUEI!Q6l+(c={2=@|i2qSaMo4nACYSmLGjAM=erFj6Ie^
z2dJ)wn<C^+1r9>*tvJ;Bo}A;=q``))jOqsv2jSF<n$2TB-e^wI)$CBQZ|kvtYito5
z8uj1a!+*4~{H-_+frf|az{7YTwT77B&B8HM#yz)ILU*dRhT<RW$y?*^veb?!PE?%Y
zg)t3G!^CttGCdrSG2!YwpU<?dfVEcYrLaDpnOEcdbU{+$r*A%EZIy@d%y^g>($LW}
z9t~w{kd!=RCZ8CmL>`sv>WDd-6)%<d@4w_Qeojtr_|2DJ^3}Vqc>Df|Wl6j^UdW7y
zU#ULRSSem9G-fY-PmZG1L$2X&IE<{`o&UO<-NSTZX@$iX=GIuOF}6&$)IFMv_m@YO
z<~)~%;NiGSm>5-WU2`=@bGDb7Og!X7%=n9+zGh6y)4Q*@URRdujB4R<N=)lO(-CF^
z28J}zOjy^I)}lx!bjHF~>G{CCpclAOytjJj3CEhLENj6kJDvZ-VsFHO(sO1c?hTs;
zF*xBKsKBNpwl$NzcMbxOD|_+jcV^dn0W45l){CyU1}ArYEP=jeG;gdm?%|m7o)75n
zey304M>akcus)6NZ+yqueqSy(bd@=nhxM)sDJi2SJcYiA_{O#7&WS=girJ~PQM^(6
z<L-MAnme5!br@~$WYk)zrBF)cVI0YMV1E1>b;y#CAr|Io$cP?-cQ1mq5*?LpNt+b1
z5>86iDRRxaaC3-*Ru`^IxW8PN!crS11A<U&4tTNEj^1cn|0EhR>NMG$0sH2F@7=DX
zDbNVgt;)B?CcQgu&QhJ#Dwld?Z56xQ2m)E<dr`)^dzT|<f?-e|9*!JF1+KU^ngs>K
zDVR&71EQ!791brr%dAUmXCdU;6WBgJL>|Q^>#`NKwyh_DOKgN)yF>cvRyum=+kIg9
z;Pe?UdFw>WCSt&??&NV9xJPk(D}nuvZqj|XbAPe8=pS?w>DRT?%8+$;Nxbhye-z{Q
z#&IOPW}O<@3|G(;$i$W`?w#H4n|VL1?T4oi<G8u)jyAU<W2_et-O1)qG1hfunP-;k
z%!E?sC6aIp3{cL>WSJ%-D(nZ|t=q2(L+bLQ)^<9Hy7k;E#T-#MuPbk#BPp*I)R~5}
zwr7f5DRS*DO)UhSF$#@=ifP3NDn@M;FCMf2H;&UGzPp9=dzjpZnW+;<lBHIb+V$Et
zVVORxqhY*XQ)@k@`gt)WIGqk0vf%4;Xqwc}u|q<Gr#fgFNogYIiDsb%7@h3+M9x-N
z+E!t>6`_33GB=}yNUWusfl7B(y~S)4tNfOnZprE%lOJ}M*}L_A{}*iT?;F}j_>R!>
zZ~wu+>v6YFarZv)*ivJyOIiDK`*;6X;u!u&u?%h&d(<`$w%hMNqF9zf!>IdA@W5C8
z8N~9AW7ts9+DlbC?dvX*`}N<W@_vr>M$Hz=n|6_S<TLT^n`-#|rMS@%zcq#LDNc8>
z{2defHaYHw1M9~g%fBm*<@VFek!G~kkWOtIG|^J-8fAU1$%lLLw#F7~fg%)WG+Yc(
zC5;1l7)W`*HLh<dp^dfayd4immU)4-Qr5;W45&LpQXUQm(vV3h2ytd@*Gx-ENUjVf
zByq%}z?Q^OiS@nK7Su;~p64rHJw7s*(4#7C#brTkq548)p-SMQL$D%>FqO!w8-S<Y
zbZsP+U{q+Pd3;~Y`d+o6E92rXwX!y&S)2uSO*bNr%{|<>@8T>9xiRmDqzoN}+FGHt
zh2|BP8bIcp7={C5|5=N6O+4df>(^tb2jV{6w>aD4dy8WU^z3G!uoa<i;|`5{hp5~b
z6dxwa9f!Jw+8*frX|ffsw#Waj;JVvqx9@TH*WX4)-&U(?QKU8Xt}VMI|DVR6!}wId
z`ZT`Zaew094m7Sbfhta^j?~I}xiE|ehUv&DPmC#ZHrWMG1D>jVO;lP7YwOFpQfy&S
zp-D9E+!XiBa2OdLCUnSXN@N$T!n!V8&KIudCw}qEzvSh3;Q#k`{|L?a;?)-nlJGk7
zaGZE^d>~BjoOQV(u81DGMw4?)AuKsxDy0T=+G%*I#FCo$^;cif>}#$igsemkILch#
zBc_NYW<OJ;hNg=GIj+%a3#*S*FRbPN&)%OlS(4;vy4V9%vvZCwkxOM&ueXQO914XZ
z{Qv)9D8dh<Tp{ns+~J++rMkPSbBV>>&oNU~fFA%=v*QtwnU&qylTqzj<>}#Pu|-Xx
z0A2tuytVwhckg&w7jBRD05ma{WE5MFDr9pfq%`BAaM8lM`iQ#$8SBM8=G{gbN>!tl
zK|jsun0fW;z+eCMe_%@R^*{bMN^Kr7wuPoE@17nxUn;7Z(|pfYzxoTFo)#{bHx#X`
z+NgS=>19(Y5|P$Weq!eB!h<*yGDu{Et%`HnZxqiGpoc~JGNVjTw(N^4Zde7WyW3b5
zM5zG=dL%fANS)AywCmz>Z4t}%LeISFFYVsidE(;`O}#brJUMx$m)t}t?5Ct?pUZLl
zLouo~E^G0K?pioZGe<dcJA22Sr>0&S6L=&>7|QC#p(;;naq`k66f{UmoAXx$nO#a=
zZc<8(&U4Dc!vl}6-|!FZ?|3sGvEpg`G&d)w+<0|6b1M^*D6K545fPQ-ExB)x6Ll@r
zQlU1aIniqtA<2x2QBZ#W@Q(lS^|zeTJt-ZiOqf;PmEUnr7tZ;C3y)mHJEFP++elJa
z(ggu>6W%^O0gbv|n6vO#fAQx$ENA}v*MC1stC~PfLM0<6JS-QgS<gwLKLH4*7taxC
zs+7`TO30Ks9)-L6ng8%tf5DuL-~H|1QZG+DJU+3mE2UPxeO!4eAn9}F!z<>~XHX1Q
zkBXPF(riU38%Q;31v)UiH!gG7uLhRUDkR;Jtv+)=d)Z(Ves5TGByBkFY%ki1u5jk=
zNGHtDel6NOtM7xNwF&Ec;{Cyo$F+zDM;@0mCz+X#C%=}kvOggy_VF*4Wp_G7y!yq1
zC=#UW?T<#57oI4&ur`NSDlm!uyO8A``w?b$ew`ZuD<DIj<26st3g<I`NwWaS`kEX0
z(|!)Eq|{N**grb?4N=yn(ZTEo1*|XR?@8ARSw<XGeY`xO96oJXekR8rILHo~nzF0~
znr9ZB4kvErnbUM4*~CS+08$6DI(X%QTIEs;MGLDKMGBUUe8?Q{jvQ}J9FI2?7PKjs
z%R(!Kbvg5^+grYP^(DXkum6>&H*fj%SAXVwtjiOp+Y?`W`2|fI%^Ike^0wwX{PEb!
ze4|5XsZ3TtJmaXD^Oe@7T(t3b-+ayg{&#=N&CRcw^ATO&GKq1+finw_^2kNbEK*p^
zS7fFl;*Fk?CS*-CbMD!jn*%8cvmF2<=ZW0BZ3L3gl&N@QhR0>~>k%-^5E+e_jq=OZ
zDy>zQ6;n)C&sP=UfG`79DryVs`oLONF14VOAsl%1**&MzXS{m#l~a&ByhU04$6D}B
z?v2oaOop26L46N)C-L<m>diKK`k{#P!#YssCErf#;YUr`Mv%LY&WYPADJ(jOlcM}s
zu8MoVcOm!S6S17*K<qWObu>&MM<@Ap@k*MM5=YD_T#Qf4wtg8udAtBvFXQJO&wPJH
zNm{9OrI1lM0G^oY{pk`SnQX#j0a#>b<Zy&P93b&XV#QnEUd&szYVv5&h-5a6PBu=8
zs%q4#lu}sN#iLWeS6_ZfHu&oEFUUy|3R6x@IpJOOETSJC6(R|97AHWghE_$ZB9T6u
z(!^3K=VhVdk=J7Cgs0hC^wxU8$S4_=hNK>W#iI%a1*53B5xrcXM9X-LQ+Ai`(i)<Q
zDCB7(C&5zkXsov29O!nAIGK4w*?S3ri4yIj!lq2n5?TtHjH(vda&1KIR-0Hz%%=l+
zI`Q~eSXM7BF>@nzi}rSNJ#CTEeU!*-gl;>BNNa{zCWk^Y5r_^;w!0$-v|8w_V8Jq+
zM?@Vk>XD%y7#jUrY%3eE%+8q}R^d*EZe)>HA)@EYy@#7qKUS>8AR#u1$k&f3S^UL}
z=W;AYP0CUiE|#b=Bh9%mQks~_BvH>$Y|f!#Ue6g>0Olnd5OVBgvlr|6OJ#Jf%6Vd%
z4*0Xh{hy}P)+p7}$e%73M2)X*zu-6@Sj3I18>t@Y%o$PdjH6m1vOC=|&QvN|jk-GU
zE0#fpb~YZ{LMyL1U(T5KW^;!&8BGgnZ&-Ap$Ri~pzOCb&NvH{G_S~$jjZ)84TbYuO
zl29*Cw6c-~rj(fT#PN9IFi)(}A+xl`h}}js)hV7gKgKzR6cJCQlulG^&@KyzQJc%}
zOriiP!MbO{;dmrZ6OT`iU^$#t=i}QJ!On5rAg36e^D1G{jvEUe9^ZLJ^M#FxXa8^O
zN5Fb8w?|}6ecz4CkjUQXt-CFF_X0oLBGDhQps&O_WcfKBP2ZQ|S8M7SF>1ZLIM3tP
zpTziAkY&x#p_cx1`xE?Qk>$!7??;|!O6b26S?&cquIm#bPO`7f5OS3N`Eg96E2!Bn
zsI)D&D|w0+#Gbz@Byna$-A6job2)uBx+)$s_dUx!Obk`eSU)uVv}O639Jgd3jZ3|7
zFrlUy)PrbEBg=}!lrmHEV5O!mx7%*ql7mZCpo+4Xv%yvKOu`OYXA#!0eOqmmb>->t
ziO=pP(&2{F=>}=WmtTFw@i3E8LqsE6i3qedFSk?YPRuD|NtjY{IZ>spO4UYAMk|f7
zUU+y+{QjGXb1m56h~)z<O{f*Bo@w^Lk}jOn6ALRvl-l{DlF$;=w2~|%mZ(-x3McQA
z%d#$%wvwtck(l#L&J&U|m*Uq|tkJvPY3Mi4Nb))L@;(haM@e+vHGg)ccn8+=x}d6@
z4o6mX|LC$7E(P*@BIN^bAD$>hndS-3-sRMGC4&RR5|>$#{!BIY=ZQ$`0-6VUqT)fp
z?d#q8-x@1*dp5GWB#jvEXnK%#u@JMtyJZ1Fd*|<7fzTK3HpbpD*5P1aWV6BTz3-?(
ztx8S?2OWE<pZ^n+<(Ki3#tVS;GJdu(YO}lDVnMzKl;uKEK~Fc)$;7dIC52g~9Pkb&
zZJJT@QoC&%Yp4{e7^@ghw)n=KK+VApZ4GOUlroYWz^c`M)^+7_Ia5o;PVg6h`llRI
z=Bv+t#ab>rJbg!=5{D$72CU6_Kg<EGdCEx6NHe4;Rsz7P&dDM+vo0&IU%#W^Y1k#-
zpwj`FXYWX)PZZ0nWLj!uB!D6iO<wbE%A&@l8SCQ{m4(&{S&Z8`QB+xrB27tZ<m1G2
zn9$^krnG_@OmyJSQ^m6?B|4F)7?u=r^Xwo6sR`CZJ)hCZD3K|X)RnOU7Y>|7INctY
z(jBi~pFL4qfQYUckrGY8&8Oo==1VUlumR?{o)WY{Ymh9HamT?GT^<W+U`WvdyN7|b
z17I~HT1eE99Mv7)$$G2boo-A7NEV+Z<C78P4BbFC@Pd<m0ja}q?*m&B<nO9zZOEKF
zV?h(?69KTE%dwi5w@OkTmq$`=l*0*8SZX0nN9F*cClL-Q8Vy7uzxEc;F}I3R5mw;I
zR+I*^f=uKzGtUQ(^U*)6I?9_`sjQ{2)XKy849)mI|Ji38=9xEj8AQ+XbRf@(daktU
z#LD1g&vQx~vu8%M3hPB#n<DvuG^0EyXL;gldE)Wu0ijTWkg1AcvU0X>sH~Ls#A;_&
z>L}1L5lV)2Du_-bt-L9ZEcJ<B-yS&R%<}M-8d)NTG(pbX+}v?E9(X*TDYde+4&DSX
zByR3?qU!GST2fMSkkB(F3>qwNAIZsBYmF|c6BjX5D#esa=I-VOk%_l&-!RWdZf<We
zT9jA#LJ*rq!sBK(_jTODp`1Vxl4h*6Xr~<+lijQ~oa={hDn+8?ScgH|foLy3?b%n6
zaU&h8N?wK*m003_`Ec$mI6h4F4z4^yxMFcjKyZ6}T4WgOwK>?Lt&wFTtN)aM=zl#~
zP9e+vU>;QZW0K{f7mBr!UnnB9e`m71mE=xt2emec@>_4DkVe~-zRE&p?wwHCIj{j+
zhm+j*g+vh7>-M$1A39{2-9-5t4!+5*6w*Q`j|MNkPQCE}tW7!A_Q}ff2jv*0C59ZY
zhZCs@tu?ADrMU$*OHfqc<m!rUIvMX6TBK4_1c&QF!>F1l1{EWT^X6GqT5$ktSymq2
zy`vm%NGWmm>J>M~8~*$+|B{^G&Fg<~t~yb-wG6bButPIRoO~V?NQpJd56h)enschw
zWo0?Prlt!IJn>ix>E;egCmJ(lUHhg+Neh?!#M#a))KLmNrA*64$!C&INS#>qLbJk=
zJIsuy%OkZ_ax><Xk;8#$KEO0PuWV}+Z5?Apf(fpTMd`3`N=nj3vPu#qC-KZoG1j$E
ztnzSvB3a`0=7z-@7cD$o79KCkpZ@w5Ne909_BGRd;CQ-$cgkAMel`iqb6~>(e20vU
zEJ!=zC!RX`G@4{}d2$uH<HHj7eRKe2hF3{=ph6X|7zw?bLJm{dyiLQ-9}(ldY;EZ4
zhD`;xtn__or<>V6)7Lu&>uO9{5miz&!tka|pN1U2jDOj90kB@i&pV#&Jz*>HZh1nR
z^7-_NNs~t~!_fyrI(C3Z^oOV8L>jmT%(U#Kh->vtd$Qj72-aoc@<;#=hL%cOD(j_C
z%SzM6<>}0l4*cfs6_mz*`)~h^!!h&f^P7=^tLkaMqdSO|fZH-n8!AE(s3O?p8@y-j
zE0;^*)t8@hy1V6kI8Y>Ut`o~rd3%0{P8t)H;u&PPB1yVMFjBPPetNp_xIU4l%A5q%
z3x_oGr}v+6JkHGPF-qq=dfv>!IB+H5WslC?;miWN=M(Mj-GNk=Qh2yr_#c1s_nc0N
zo6izo{n?-JfBN;G^1uF<|AV*x$KP|lJaSp#{?)G#JM!+~TW;=-eDUQy>+&__;hm>B
zhZ5Rz1Uz3P!*G^Z0}&DGSykPb?THrs(i1|9d7HwxZD~`0k_3^wac_ir?gWt{VR{(j
zJ$rU-`!jmfX}6%qZ~5SBy=#*vuNgTLttvmUDADdx-NE-vl(Vj+2J5WMk~rka)9t$)
z@0<!x0rwy2VaQm7Eh{xO*5+Az3FnYAP2S0~R%CV#mbw<yy9#&*p(!)ZGlyw%C#WgQ
z;}fTJV#>G9eW%UK&MW{l45%~{s(DGk#-iW=hXN9v@(kn4&ws_;tIt`}4aG9wylvF#
z9n7jNRJ)+mQBKp%FSE!N67M%qRJpW<W`#+ZTI238b91`q-BNh7R8mTwm7+$q09FY*
zupij{u4j?W?LiI5Mqrwc91e*ur^+$SyqfO#|NQU&mEZsEYcA)?suO?un|siiZ@+oR
z@pRzx&u{tqTV*XzkP1CNs-I);L%B?KN$(k$uzAS6MR#O&w4FmM;v9n*Do!2_?6L!V
z@)9`~pUZKgm%T(+!;kDPn{&Q*W*Ls}wy0(IjSa-=@+)S8A8J$*ThC(ogvN_3zsT|r
z$a16iHw$@dJvBTs90GJRu~F(PILOy!xyzjZtvkMPjFh+ZWH08Ik<GI23@H0+ln!F;
zWcj`E<8-pbv;3rG`A5ldW=@lrVh%ZGSYm_3+xqj{o_P^v#RJ3|W!j}yE|&}1DwsFZ
zsI_ug7nb6{&ay0Ap3dCg-80?6>#x5-5*%(0<SBzy2ScsVnuFkGZq~O5c&Iw<K&^%_
zF-Za|n0kjl$DK1br(5)JLedSYnWxtm=V-N%V?EPs4Ngj*Gwt?XtTQv`qpR8|%b7_M
zM_`^N4#$~w%9Lm-P^(7yZ>+DbH-?f3A0@P*C;xZuEYJR1mWA{A0$<#4cX!WU{f9rN
zu8s5Cm3M#h-+B1`18XUmWj_1-Gq9O&zJ9~;=ExUc-0|&eDD6VZo0L9Z7G3&!*_}lx
zxA_JXog{neVGkPhojvlUMQF*<&b~`nkM5|?pLcJJjQYs9?m_6@CH6xyHLlGiKh8gP
z*`a56{`)xwCTYxn4~S*?#N_y8{N(WhV7-i=ZEOw-LqYa&t}7@MZ6t%Ec;=O<I(Q`!
zWjDYgT+`dOB6v&n6eWX36)%&^#)$ltwXjA}v;bP06HTs*15&D#b)_r|r!RiZS}*+L
zKmH@1eR0pLFK+0YsVmgxr7_@u7Rd=+%}{|sRi4v`KGfzM7RSShFFyN%>6lqXsE>_R
z8;f30&2EU(=nb%47rHTmDI!M79ua?7&fLw6(#AuY`TXV<n=|#0S?WsB%B2)n%C;#S
zk~?9}SrO5X_)pgz7FiHg8_Qbw=YRg1yW5$+`ur<C`|O^-{_Fq9-~9G#RE>3ASXSdV
zzdE5h^X|<9H+PA<`y=@qr8SQd(Vu{IKx)58cIrtY?YcXdJ+j;J9){1vcsZP>4Pp{^
ze8WiXU9`|WkhU4WA_=TV9|;3`cX_tW(Waa~G`b_af2q&$RuVbI6}SVCK6bEQ)cBAb
zTTIHTg~^=jCX+a&9FAuj?c1M6<K=2C1fZCe7DgnU_}@~Ltu(Z{6Rc`ztxhnl&OaeJ
zlXGUuQ><^LE(>}ylf(;?EskqZ!U!qIW5vmocbO#p{yOWDg=K$#drz7t9@E678Q-1X
zfCWLjE@)fOjg86YZa50@LhM&X6-$O>V=`l&GpGB{ki4+Ug`$<Jjb`SNW0CC`JIv*J
zy=}74jYx|JmeRyDP23%CxIf<UC-=YM|Mz!)%et(rtM^$uosL+`wC4OYcXtPJGHP2$
z63-LaiORm@=-EB@GN571dLs?Fk`vh+U!p{K9J9-aA5o-t3XN9iJyWu86~U==^?ev?
z%tn^xv%mZE>F0XS@7)0F?tt{tYlqLBujNx2FS5*wEdK~u-rQifL1^)QuYrmH)5clX
zfsxS3UMIhk<ut4t=hRB$`E`h@2YUvjC0&L$pg|;gkZpj_!uEvTTkYFF$IWeaj~Sn=
zEdOXZmdvrsvAQk!><b$y|Nb=`f@z|jof!YCt+FmF8eQqMIfzwCq1K9;16s?n@Wrdo
zIZYG)^bfyB1pegiizq8s%xcK2Mm9}~DvK74ja-gf^Nxw6{@lGvAu9-(<8gv4G>+5?
zEj>nuRA>Xhs%}rmoORlaZs)04lvyfT3nqz!%#bJSbRbWOloM+yQT}TK3>)(n$CbhL
zw%a)j_Z<w()culm#S9Mfk-zv~{({Tn!neFZL^z+%(SbX2b8~~)gxYuHNw~j1ktb-n
zB6-L8X4n6?%2Wh8ZNumgm`2?!o2)aw>)T}bU}+BrNxyr~;F2`<$*pvd2neTv=RG(3
zyWMSrv*V+N_3q65DmS~j(+{9FI;YA@2G+lu@d9AIjDM3ML}jyi{?8>0nKpq)<}@9m
ztj_DDNxpG<S-(3TUUk%?MG2EkYZ@(HU1@YB4sc$b<Dyk3A1t-9tSjf`jA0y)Cr|h(
zjq~FZ>-oZRS-D(R&dbVtNTlqLLs-|oxr-}=S{tnitwl$Y+K?%sc;~++kms43(}`Di
zcbvXFaYp&=H-E#!`H5m@L?$F1y?bF8vHq*k8D1nAX)>yLNyM*iPrSN0@)w_d#ms?2
zJ@U9TXr23sAjx+y=L!@u1_)SaBMK=%&XXIJq8x6HobFDvr#C!Z3jg$vXTJHT-}1ly
zzy6>6-EY6)pML*_Z{8Wj4y<cqo(_EZt9xF3cEj-ilFl^VAjt_VRX-MsiRT7n_3KOu
zIXhRw;OI)jN&sAWX5vRufM2pX&$0nnqc+|Yw1nZRp;d!8c7$QC4d&j>al6)SJ^vx2
zR^{njnGO@DlN0Ix2%Y_l8(ogUNnndr77|NR4k>Xn-C_pK+9(m}O&CN0>kflK3pX=~
zXJ(kUhc*IuZ52^Bko(yUyc1?~L)Sz|hshlqRlfW7HFKW0JKS-aj|>z%oNvTPsgs?#
zkzCYhRj?*x$x#whXr?@AWu6b*9#7od-EcS^xj9apwej}pk%y<srCcyufdb;4+q#3)
z`PZ^IF;mszj=LqMIdi%>@aMN*k@AeC8{U>mL0Ma+w#KSaS|nSEs%I)FCfWC|KEKwT
z>y6W<Qd#W6*Kgi%Itxs1`R%*E;XnWGHGluRf8y&mU-P)!Vkz_X^*iSIhF|^aOFsYN
zj?)d0&s2TOOuYhz`oFH1uxB@93%<KrhwPd3pO~CZARQSSNd|+6uP(9C?Wuom3ut)h
zm?_?QRzOn5`PJG6sO;qUL+kkJ)t;5Hm+WUiHn@JeEQLH}Qi_tJPf(8GMV4P=`TJ!V
z^GtAmQX1sUGHevj;J^Vcyj(AJT^ssZIu|Ma+;UJGph6P2+e(enzI_?+ht_M9CXHcN
zzn+cfh~WF9{*lP?_sVe=Zl;rWjJ7t=ghAubl+nu59}H7u=QIk1)tP2v30#(iS__xU
zg|e<*o>&`~%Okavk!^TcE|k*^xiwz@{u?AGUVVA*cHKNt){0p$--?$xnkvN{^R`}|
zW^R`zjB{D3CXf@y<B>03z2fuxTkJUV+i%{n)(grST)57?XDl8Z-aXnLyvQj#hi<FB
z_DGzjiMzwhpWOYLBEolKv|4Dbu+~bcjbdIZol}ZEFnHS_O5kKv3b>>v%-l~XRipc1
z<Db9zmWPKoygB~^-+l9jzy0t3$Tx4^@!i8CDSyE{-%`p-PQsU8e!=IT-*R&UryJpP
zBVN*um)^#meL$~uIkfO88f4Fa?B%qikYnAtxJ%Ds8i5@fSeI8VWO})XW7<kB(G_;T
zFvmaqj}+(9F6=lxN0;kf_0sR>*@3P{Tz{>~vR0;PV$SK4l;M~0)5Z&c^)mj=#ugRu
zyqqfPSxf>sh}xU>B!B5&im~w|ZAa@HqXci3J3=E~WIK+dFrHg$l;)i)Br<hWTa-4m
zalEm0#wqrDqC}+1qeLk-;7wr}cy|)Sqsi06JWt4RrfQ|sg;ExvxkJ}G<L&P^fH+qU
z2FaDCnlaBa$K!$1@krJQX+C3W-uXm>u+#eSMudC^)X+;U?fS+dEm)F3a$_S)VO<s;
zmorZf@A&S`!g*OKHB&87N+G8Mhr`S?OLP}<<H4C{g~8t3@9o*2fhq-2Z2*_Uc;8J-
zJdeReWkmVKm4&i>CLGyOVbqo7Gu`-MqxmD6@PjY+6FghD5&2Vt+AJbY+jV`4WB88*
z?*>Yt(ZtBvL1mhoR(+lA9?k$$w{-PwIEc}Wz78TPy>!z;?pj1^d+K^hi6n`-7G#2(
z<B2IvV~-daHuQDXjVFt820veKLEib7N^49rOw+_+o;e<7)=9`sN!d9rOgtJr$_94#
z)BUY}ml~`xB8e$y4%5WxbikNctH0aWMT8?Op_j;Y?C1I%Hs*4y@%U%e>QVDry~!5~
z507v7hu{5{*YDo3)CJ=fEU~T&A`^$hEz^{kCL^WbFnZ=KirHB0CReb}c*fhFCgZCP
zX6;_?GSD6StNo7<mCbP;PqE9ff;USFLw$f&?~9k&vty4*+!_AgBa&T?TZ?-i<rSZ%
z9K(w&zsT|r%QF3%y4}#7&9uo}cW$p&)^?JfHR&?rvb0gx%au&-vt!25cHouU3(tj{
zXJ6<qZqD;=k{wL{Ba-Eh$}#3;durafJ4$P>H^IIh`r0Ev5r0p1XK@1j8vJ!Zg0G%+
z)wIQPQqO{F-r%8@iZtgnH4_I5yFn)6fyV&CRlW3l;Aw2~X1gnrl$fWPd^{pIM?~HQ
zFy-u%CW7ZXHl8nzVNu34D2k*+CUcl)PKN_`H#b}`tSnw(Vcovo+v?YFQ3r3tc<-Db
z!#*#qQtHaGX0QTp7QXrRTmJEP|H!#EN;Br;#5^CVtpb^8&P-DxO-9a6D<XkC_UGTv
z8YX|&hygx*K#nEmLWcGXv-H}}>Id(B6?EB;%9zeCIoyj_h!z^Z`xU$6_TaFMuWphz
zGBI~${Z2~%{xGvmCtH6eFXP|Kcmc3p#=pr(79@XRtEeffR=zvGg<1La?H%6juhJ}G
zO{nS$Q>?}2dp!i<ATyedKxWFqX-?jvw>3(Je3+aNy;x*cEnJqBBt{lvT~}7iwAM%>
zeE!)N+`qcz_U4YmoJpbZTPSY;$xi0mS|Qny%*2!?)VvdssVgW+#_ipS<L!z0_DB_@
zwZi%F4eMngi(pd2!1fMBLPRtmIvGhA)*cCKC|})w&X@N$P`#}1-Q~=;%M)+bC(2rA
zwXvFxF7y%vie^6VDKt$Bh1R=~IAG>^O;6{_&9U%*|Es@ZN|o2|{s&DjtgJ98(@DsS
z(5mw0%|El=Jn_}9e}$xkam0AS<SmknY#D7IS1c(se=;RmVv@-X_PUPslEm8p&yzb*
zC0Gbb0EN%%{)7a1CM6|j1)FiuarJiY-~_@bDR~J}cX&KXWP=zQw!?Ra<HPFo<HO4c
zGu2=PW|2>%SmAoO9Q(gTfD9_icb9jh8UFn4b9{CxW;s&VTdydA9ZB~d>l<=x$-5Pb
zdFf5?+^p+LS~FT3rB*kXYY8J6UcI{Ki|K~+dL=8I^30r^@IU1V$p);$!R;sRK(`xB
zGMp@p^5)0oiF}yI$LznS%$v90@%Hk><GXLTEX9q`ZYb{}iNmSrw<;d4%u{CO#FuxU
z^Tq8Ek<8Pw^4ss;@O7;`)#_XzO-B`8>PF$_SNJe?j`6rD7y6WVczWbGzajnkUossM
zmX*fJ<NA(DAsrI0rdLRM#btS*ww0!ZSFeP_;WG}$#^G4#k>vfO>w8AGGrhAWr1vh>
zfeTC^CC|3Xo%FVaKCtKpCc_AB=ADw}X&R*rwn`AgTgb74Sk0n&#%>$9%B1Ws4qxp<
z`%yRlKRy(wCPYuEFyuc?qxnUaUu5}*WVx3BS#Th!ssoG&ViQRcIcIt+4#Q<pBjd9B
zzotqeGtY+z{57nCX~k4V21{R0LymW4ggsN|x$4~lUH0C4-#66ck3yDzNRGMx0#MNA
z0F23y6WQv#{rkJ0IVBF3sf6swMDvg{RuwIkx>U+ixs(-D$>5w=dhs&DJaaoA9Uzh^
z?6?kQc;`utXRPX3jiQO9IRe%}BpxfQ;objKj6AzNC|PLb!n!<>WF;rB@(?Q6O6Of1
zK#X;pasy{>Z*DjoCjR8h8}5$>?r!h7oX>py`k%R|vNX@`?A->5RRa;5wf(WLVg5c8
zSXKGm*Z<7pyE9+?r~eE2I5D3J%*JX9niDB!zWnM}9FJe{`1l<#G38rIGe|R}3nB|U
zCRYZhSwC;begdpY^sLC}s@imbN-6B;G~V|hLv6iF<yLB}G*U`T)9jhENMy<dSk>=6
zkf~7y%w}Krz{M3HeKq!S{BfP!Z40&JT*Bl?2>8VKWnacm9xnja%lO&HKI^}0$F4w^
zT3FLcv*G}at#}`FH!6I@W=MBN{BOlbCQO#ON!}&tDB>x)xiM!$at`hdjUSJQrYyT|
z87(?(7%3%k%1k+fMeFoImtx~`Y{!M-Nk-mYd~XcqoXDpmGG}Unho?v0ou4RWrM2ov
zn0Zb4MqnCqHvVHcW1^B$WOU_BmP~ENQX3C-<<bhPwzys!2~W~Bnb$o!h^2;z?q;>$
z+ZIKs)!UjM=OdW$`t2Jo%M(>AYb{*XirNGMX<m^$QQL!qj@)2UFzS22quB?dtIz+C
zgo&G0Jt8E#e=~mDv#g39ZR#&Yda_2U_b3GnH5<xXx`BfYrP{B`&lxA^27yHMiyuEc
zq8LfsX3X%(bUf<gs(MHEZQqTx6_&J8t)djviUSOhK`@>Vow{>x=<Vg$%R!CClrxRy
zrAM^}hn@Nww;&CTNZ*tv4u_d^gh?_vIj4zdWr%x(n@>uA)}H>aZjSDpN_W+j(i}8q
zN^UrpmA4NMyg5Hc%Dmr6|9qdPv5I!ss)&%XFwGOQByyS%5=(8IOX0CDEdBl>V`CGJ
zZM6+z)(;tg2AwO=Ojs_3^SpR_a#eo+?eBU0_BExs9IvHvIw;d@q`YFHENf#e6HUJa
zDWn`xar~Hq4TF~PSnbM;_Hx1gjD;;jCk*2Fo|pFam9Y2f_l~uUd+U+;=dw5Uhz~|!
zW1g<|ijSGbA2FC^yQ*IP8;m~vZa%5;BFitb{NrUAG`%l(d$7ZWxtCP+7dK73zu)0t
zhfdtP;6BHlJLf|h*2XpNF8Jyb@4xc#@u|!556f|(S*2N}+8TgwZ{X8mzinf5`SoLD
z%S>c$(h&l<Q4&oYsIsuBTXkSsLXM5dgp4bh=82Rh=6NQiG-Muj+3+I+%McaNt`Wq`
z|11I);u&Rmo=EdVGozK2%W`2^&X`6YKoOiLFET1W@K~*bE?J^VVkYMaNr`hUoLgbl
zN@+&zTxqek-6VVlQ-vH4+1bg!Fy$?{c_~bJB8l+k-5Z|Hj}8H=QO%g<#Nm`#mKANz
zFSf2qtp`+Bx2?NzYJDwiTC@wnSkED-Dto!vt)uNLQO#f@xyp88Y)0(q+qZg#>(%iB
zpd4W3hs5eij`z>x>Ok8ZkGD2ZY&6)g)GHc4NjZKQKW)4KSTEye9m2DIYr^1UZz%7U
zCuS4QH;=ShDD^^;8IcKa-i!TK<ZzDR-)^KP9Bk(H=9Uc1s(!*`tW!f~peVH&t+>&w
z3Nj_8DbZ9t3fxwD3)hr{IZs$CtgX_t_v|r7S)#UREA1r~SrStQ#Apb~l<9cjc)Ep~
ziF0cF?ce>D@7}!O>G6@K#xxyJyfaac&_Wm@VZgaUu8<1zF>|sbbD9xLJgzGb>&owz
zN8Xi{r53b0NM}<x1nPNM7GRkKU<ZO_pILWKLZ1Y2=4Dw~E(^zW;Cz1M|NCG5H_|M;
z`uvvH4-b6z`Uy>U<oU$?=NG0)dAPh`I(>z5!sG&}pbb==#N4AW&sT_be*{Z{N<vfi
zv|xiNCGsu`Bw^)t{sBCl1yfwnmW4?m!6CT22X}Y(;0{3>cXtUI+@)!PySux)I|OTj
zyW8}<c{RV_R^6(*?>_6S^(`8Oh~<#NWn`B1<5Rspx!wkp;??{P`<4<f@V3ixqcrFZ
zrkxh#r%e*D{ZHat(pY8B?0WRuP~VF)(6L_%y1-MLnQ%m)QCb192(btwKC%kQGiOA|
zR;D&1(FQycG=KDKCqbJU!cC{IP{t{8{nIq03h~MBT|qUfE~Fnu&QC_FwNdHuB=6z{
zV=E<Sb|5OdLyTf!#8h%jm$9jN4GXmDQwM$796aV*hD?wS$bi@T<_;h4T?==6G>Uvp
zI?cZL?dTB(B~~(Qw7vd`7)6Ci(Tla@i>bH{1Dq~K98j4`aQ!qA<zc+MQ2>Z&=gU04
zT^YBt?$;cm<E{Jl3sC$4(J-V{5%&G&V^4S<v*;*{TNyw+MIYcFU?D`z1Nx<#e9`6g
z7ku7l-YLzSR1~F2Fzo7eqRD<64G&MVBuve5HOJonuwX1tFzZTKk@V~i>UWy4+g@gR
zv9?@ynR{ZDq?DWSQRh9&d)-9nzVB)lk|q)9@AB}Zzs5G}n@3rARR2F!JubmK2~$xG
zolj=_&4-T1SzwQi?F3r8-?e!tZ(zOyrkpG05c6iP4f?!ER89=Um3RM8A<T-4jLY@+
z_USo{gly^BYWzAbVUHri#V4;ZMVA?log6QH_qQG4<Up`IrbLQ();IgdQ=S(Wj{Z6m
z*$l>9`CNSj^DBlihTTu5H*7;&WVfDttdp>Yo100h2hRT2=L^)@Y8N*@5oW!<aVCC+
zOa5husZ;q=w$;n6|LQ%=*JppT?eekVa&x`f>*l=I>w?mF)Utx(#scX@_C%ziEDc5J
zW@5*4-%}v|MndiJ(4*R`-q3UIpMoxb=%~HqSD9r|zgz(7Nj-wrDpW-(t&M4A%1f4>
z4C2Q@zzef}YH-xey%6J=5lM9W%Rv7g!BXAzr3Klf(dA%L-2!Yf7W$N=#;QogJIRR+
zIaCp*NCVG>HcH+NrRcITQIv_<K+InOgILBgY#2rChvUNBwF92mBfR-e_1iXqZ|<Zz
zIk@X-+~>nSnosB5S+DewWj{FZsr<y8U)lVh|GUoDY@*JwQs?XDYv-p+!w8;~m75>R
ztVnPA49GauUBS}~fYt^m`$T&x@-#CBMS-dU2Q4X5Vl9H%JcMlFyec^YZ>>6}N*U8a
z-G}%0VG7q=U6s+Xex|gWa=5eY_Ntut)-h&7$r#016y?~E^xM_*P7rsTvf-a^DjJf6
zJ-cgv6EQ2>DDdH9fIr9Z2G>2h4YY`t`wDM7GzvWku11ebyRqf;n<{cT4gb*{Oor{W
z7&7Uj01^>026PF+N1r+<y_V9~#Kd?u`Dljw_xPD?!1{)>u4Gh@){chBrZ+jy0NUAL
ze*P8q)8iineHOOzmR6#{K0Op47A&PRi2%O4bFpkdXmncWcM7)TzIY<asvvL-Ppoa@
zFTEmBtpcw1H^1`jWd<d1JANvvsKA?5%=<0hh&8K`27zRX27fQR%!FSY*p3(fUH|@S
zo1tFX_vF(+IwfyO$(QmNUA_@A1vXX^YQxR#>jnHKh7vkOdM}}bQ8Q|1Y-iju6n!5l
z!_=5`%*Gf`4j(>WX35VKG%{*p_7=n>TUX#Sex}AB$huJVgOU!(AE)6k!dVhg@5!Nb
zW0__6X%-y@h;fCg(0{Hpmz@e^3X@UBGSR*Wv?sZS)H83#{>$VJr2hfZEMrQ@*mYLg
zIDEQBNxVM3!IO=d+q!D-9X|2s2I!Emx-cctb0wEgmI+^s<iXXH6`n!+b%inwzblFP
ze)Nw2#1xA)&tJN%VRq$3VcRu)pMl@n6K2I}YVQ#IE!VvTc$snGf1(iI!&BF=w$TRR
zF{%pJ{^#GmQ4M$rAsesjCXVd*dSrHJAZ>5L-o<{}_}&~Z)_kn(X~<Qw^z}k>?CE>I
z9rJkSjSjOUK39g@;e&MKg}pZLV?OPGgkpH>vX5ZvR)DeD02&HzW_&p*AfOJfc0v?2
z_Kxa?DZ6ww$^+lFFsSeBC>0pmpPVo#T3LiF+1lD}o@tSm0+rP+3h)a+wK_SlME*S_
z$Y0eDrG+{S%|@k}d_he7cn*2TC%f*=)XjJ1>&4zfGc|;!2uKt(WF!{O_g@toz3va%
z$3utfyqmPyEsYQHWVTv#kPjslnw+kJs*SeO+*f5S(mA(h&GIU-<jucL&~N%X2xJ?v
z6a5m<xNfLhY8J>&S$1F>zm7l-j<>_=pc&?$l%9^j9uaqc&Ai7#dDFY~K_BunwIZy|
z8ui!1#$u4Lu&$_PswePv`;vH3E+*7c@(`d;rs@-+JDlNt=z%&B;N5gL{@6eKAB#`d
zEO+jT1G(r>jQ)mksVfz|C*qm{EYZf}0+Ht;YFP;fehdSK2}@%JFZ?jj@+&8waR!`?
zD@BDog<_;uvxh>s1Ss_GinC3#Ofxn-lDYd8xF?XrsZv&#daR>w)ol0Ggo?sWqw@yd
zUpPie)qE;relybHn8JYxr_28()dMxI%9~O~;tEQxIP+9Q@zDGjDph4vC*4JAQakJV
z*?C7FVk2tw@<z4)mKnQJT(0|D6MS2xXz%w7=qz+|FhYtMf0|J9MiRS|P%)>T>k$ts
zXH2JSen7!BT)Ifd**kJBI5IaAOP6u|&1QMGH!cZ-l^YuyvOeG{!=5pOzh)A*`kLXo
z*9tIAX^>3H;>^@2B3#95qgwYVpp~X%(X<u_R}os#@MVzto!V=#ib`-G*bE?0jBZq)
z7U^h}7D_pL$RH%B<-*Xui%ySHRoBTBjx&oa9l!cpgE&Pv%fd^jX=kD^0T+LgkQ^<b
z$kFMk<5ao;Q!@Pkq_K~s>3TeM(bgQ>qrg^j;Db{eqqHc3uk%csQ`JlUOvASI+)4w!
zmH88D%CXi7t@F!wq+lKTVu`)`<(z=25T&^J47gQ9&~!FWg(3#qt$VKI;9l%uC&zkl
z*H*J2rUskPsgc3F|HgO5PTS-p54Gd+C7z${OQ%AM1ItN&=uogiF!#qA=}ma?Malg@
zK+98NGC^VI&K2<Sxt&5e&EwpRP0;9T)|0J)n3=1@wPiIxqR<%12!I5qH!7c&uf)=x
z#%&iP;ZCai#+xKc8Dr^?wkL!-LzIugcgu^IW*4a4?)L`U@%OyKHQb}pF$rcLgYAFi
zZ^_hvU<QKYYpH9}L}U^#7k^}L@vIFulNJ0{?zYH#+q_-<Le008(fZs?uuXpSOhbq^
z-|bcSc!*Gu$5|(5cc*Di6~l+9^J#YWL91O%2q!jb9RU4qA}fTLC0Zx0>%LxfFeIcA
zb1F^F;1wM*IeFitup0zwme@7ut;f#Z1jDURpxPj^>B0U>Bc9el{Z(^riJ7vGx={cM
z7$JN<dLrXDXP5t5Sis|^uuBypGF9nKwbhMtD+qZ}G|yNMF)-1^1Qff!?`v<Pp-x|&
zGbO+Yn^&2krF(0WtiarT+Wvc%cbg%Y{z~5#`u8L-+%|PW)MCslv_nh@L}pIpsI?{N
zA5s|FzV1_`sXgW)Advm^ydHXU%><QSrP$oK{jv7_OLadt_3VKlxGjygc++I<WM8yF
z2`1zD`XmU+O%6R~ERCZKzr$CToWfhY>aF=U$u;YlCe9NDtk`C4r!6P*EMa?+cEya(
zto&sDT9Hq22km>fwpyM1hx&ip&Ao(3IYmR}q6q3$lDk>$S(5eZenli9ApE#W>{yE~
z8nn43S!DMvbLJs@I=(S-Dpl4$Uah90+V*ANAf!$d)x$Y%YpctiV>x{8q@B(9MI<Z{
zA50%+c~3GfAjIH%^B4vRRjcCP__`L(5%S1dIbTj?4Fm>yOEuC<X;`LUmM)?mN}=za
zndUZu4K-Zy9pxe!g3Jw&Y4{<&+b%RL*oX;k<w|#0sz$M@FM5K5dXSGI#0XzsNk&YS
zditj%Z_YFX2D7&ZeK?-csBcM?pUOf{Nv4{%-Ok}*k?zZ;$NibhuoeWJ!gt9%#y|Jt
zwN%=45;>VyguRiIT<>km2S&@0jUFC~L8Q|_;)i4CMxLwH<8r?<hd15Oi?yeZxpHth
z7Ugt7|Iu*N?8j|w+MfMJ1SL>&QC3X@pOLO!f$$)30V&h+_-4Bx)TNckGJZ6(eMC)&
zG;aXmCQ<V)I;K&*=&2{?4Xiwp<Ux<|r@6{~tKoBE3qMi3x|Lmw^B1B*0KQ$nK~V~r
zOg)~dTaKKyI75p&ht=h$7lVKx$jj4P>^jnG%rfC$=E6hq?F6VnTW>b9LIaXP!95)G
z<RO%Sx5m=NYFN<@@4ONtsFDhoBKoDN>NtR_lJmQOm#{+)J=DzCsbz0`{%~W823v3R
z1_=V-w9;iwR@D^1o5|TX*O7rJa?>A>$90P9xdQ7ZOW!UXKQG$vw*enM!*B<2*3~%U
z4lLz_+S!+;S&J`4K+${vF~0eTnBpJ6o4UC#ej7&HvYFi{Kqk0Yx?7oz^mRoVuBK}g
z5h6-2+C%j{N#iv9&$q6fO2neHWA%M9M2Bpg8bn3Wv(Tx6NK7dZf^MQ%a$>PKA~g$M
zodHAIwv_V=X#ZBV456GG@V9<ON&a_p8WSUJC(!r+^ZeYYtdklLO~`G<0kN<-#+#1R
z{>U1ym7DHK^mJz}Mp(5#8X@e(w<s3|TQJDy(xxjZghK~hFKR6`U0*)Nt{}xz7SJVR
zl9wh_&gH2)1`2k;y=rx{9@<9;ZqvFU^W!*=5guYR(rtHmjjta}tbVu3lucE+^La3v
z#TQ+DGrYhg?PRx9GQ8G3kz?G&Q&}qUbo(V`<)dtOvYKQ-{aRLh$dkBU4c&3QtrV0?
zh|11ApU}jhZAc^_kG#ldJI(FTWzGl?`go0P#nYGuGShE&LYk?k8Z>)09Ax~yzm%>9
zxtXDo!rHD5jEtmo-Nna-OuVh4*+2!w{GK}Sq{Nh{XBd1q1&LbKhErcR_BU@kJd3*9
zctMYzJ{1QnSQaHyD|mrfh~9e~x}%ky{&Te7Hp3MtM|{|!E^-J?qy`HhPr*%ayCU{o
z8IPm~)TiN6qto$r|I(!@Vd*$Y&^V^zgDRDFb$~=G3}(P#Zpv|~FgKJl(M?hWe_8Hp
z?A^y?eC}0y{vD&ee#+lVDH@@bhp6!$sLBjBxg6#~ZB8w)*4_sG+l}pvmDICU+7!{r
zEeyyKyv+7+avob1&>T`HQC!i#WCCMn=SSLuaG184vdB!$$T2s)QxiSRHeBO)A`U>X
zg%u3reFx4KjEIG)e-VE#9g*+*!p&6VxYM(bWG{5#?Wd=;)T+>TXOSAyN4lC5P16!)
zG6d@)76X1$#<2a`yct+&<8)VEp$SdCo@$A>>xhH-h+mdB`arW&cVLPyQg(PD8ZVT-
zZ)3%vu8-PYPX<h%+Mo|Hr>k@1<!)B6pG=RNlOMvj8RN#St1So$`eues8~KGrsk&|&
zBnN1!d^_+Q;jm%gY@?3(XAwy4NaZLl=e|L&!&cmn968}l#~^R4qgB-ILpg1!107%M
z^iCg62c%;spsHAlh7`YYPMMQa<#Gn2hB-~s^{6+V$W+_5M&Q~3cn}Lthl(g+Q6yUV
ziJwlvsr?-DuGemWT|C84P>q_UZsVSng^p-@XZ)r**d9;|L56cn+I!Zks9rMttf^3w
z6sd@iYLv(7HSn-4^NS%kT{QmD+Lh@V(43+Pv9vBD+V(6(WR5KI3iE)vA6PT6xgWFI
z_OdA150sT-cV5g}0a!Pkb+%;Vn^$jpNe9`P&swSG`l*9%ZC_gWxZ*ET4vzZk81AcS
z#uF;Dk1~d;)O?!xPv(f&qFO;|EVt<X%m}y9Amo8qre6uy>1x!`-2d8md-`T|TrV!<
znl-Ap*R$PfUM_tMuNQx;o9R#vXTG@(^}q_1B~2diP(jK~)J^O((@<Hx`w1_UHqQ)T
z(TN@}d7TK>rZvi$9Vyrg2j1QU(+(#3``25TL{(pL3(P3h3YH(;Yka~TaceyYoD>;*
zcCnk<v?wRMOsvM%hozYce<<=QWk`VQ8sL<;Y7DJAF!q_2EsWFJNM<u#x-|*m>@8x=
zrNQy8f^#WYv6r}dI-71kx!CHSyJd4*WBNH<qM7`qsA#sCo2OdFo?iF=IE4uW9gnqr
zu2=wP*y)j+fCL|cUISzkcvq`C$$BwP>Xte~3|(m`<m((HyFQ)-A5P^1{K<x40v8Xp
zDyst@2CDVG*7?{81NJmphhT(@Vjrr@CDYVpbAS-9YYvwo=o+GcOZ`Cd65(*qM?(8_
zSvkz^jE`6?OE?0QArz&Csou0XVw35t4Sq1Xw_{0?O=Pi?VY^l*a-5a0ws&V#m30Aw
zmz%Y@{6C%_ILdzp!{AkB7IUvVJI&280F@p#z|*`rIVCotI;`y0@Zqwm-?F1TH0!=j
zwlWh}lHu9ML87?y+Z}%6g>;$dPdsEr9Q+h!C<^)!b8R!IIuObYV;nFkAyHLVjPnLO
zswZr;_647EvZ3T94HTLjrV&Zjz>#n&0uGdeE+ua5m8WT|ySJAeGJ1c~dVRtP7&`i)
zeT=cjRAL^lS4YUpqLz3Q-{}K;V$EPP=!vMxBP`gqlZzIk5og(`8E+<|>g*Z}G^3Wc
zNS)WF#RtMl07}<V%s}6O5nJltHa)v^o%7q!|EjCPaA#h_<n4+i!^zeWc(hSOo#jfm
z%oPe1I^WgZPNQ--_<TTHz{rF&#Lr1gVW`>fcf5~>o^Jp2m5H^Hkna7>i<+1r`R!Q{
zzVFk8{yyoiA{lJ;zHvnH)jkb$mhZCB$+1_UfLYxT@K*@0kuG$&*<rIZ8&F0qhjq=G
zkXTy*$ka%+4pFX}Wl5GVpvjbS|B>lL>Ov221Z7EBJNon7oIOQ)Gj(KMv@FA{Jr>_*
zn#g4l!eD(V%}Tg945X$%9QR14Vlc=4L;3hNkFcxnvlgw1&d8amjxRXhi@K<AwdpcF
zze^*l!Y2)NzO{xw5Z(Ryd;3z$=#6=U5aYn4fZFmm8^0U0JD&oiB;5@w)g)bgNz>C)
zKTN!8Qr5%g&rb~s^WV<Vq#=XZ5%2-->K-6n>kw|pQlvgE@@f}+r1{sx6n)caWVJ-s
zO4l#$!e`ynVqNU+G_I$L((}Q4Rm-B`;yfsf;as5nIDRc5uobq?1TtgON7<|~sLf$v
zUayDa(#%<=Zdp25UUad-3%v7et->scsDio>-g(ScqpJO6yQE7+4~j?F*1gg5(N-~L
z?N$d@=39KeWHm~$7q(qeC_bHhJlecJa|iTp#;Afcov}*bUBwoR_m0xHlOWeHamw_K
zUoO<#HZzi-jRq&!vy&3?89dkmGBN5;dbm(U2D0aC^GbGaDisCm#aA<DZlN**BNEjl
z3uh`wdEH{qiR9oaIh%%%-W0lwTH?Qx{PcOBv5p(v5ZB8s-dp<~Ga79HpCq+l=8v8)
z-eeHV&$4#mjy*D0axYS!FU+46B(T3i8G0rU^NcTBm*!eTkugA)@;90<QPJ{ya?>B3
z)>dJ*Y&KNo3+<)QxiW4rPj);o8%G{J2`+6AT6=g8xslV_?iQ-h`pyQfM4hPcwg{fA
zyLp&@E?IV%V9W0T`C~8V9mFx|%asE(LfjKDuv{v)!{Dk5Z2Duy#$w0t45vG)*_y8A
zrO%X7bgc1plThL3Xie<a*8ptr)<jVfG$lT;bQw}9aB?sF&TMpzWAu*aF+5R-H_w+;
zGIgE%{^ljbHhn|;8`>dkHubpI44u;x29*t3D)-4S1<v5~n;AmosfaT%CIKC`a!$s-
z?SFV0#ur6X9%(CEE(ftXxmI{0=G`ibF(-9tj}H<}XJkAg+fOYL?I_OGOi{wsOV7gv
ze~cM8Brw#V6_LktV*huZ2^m6Fq+Qj=5Gxwv==e?<&9M7-7<h<Pl1~r?{{g(x3Z|B$
zHIAY!A{JATlan70)x&k3ISg<o$Xe3s<~LU1DHSM5JR{G8@tMq3l@xA%0Wmn>30%ep
z&?A?Bp4RMnt1^ewke)O3W+zC&c2Wz*_><VIAiMl2Dpkou;kT^}wb)gvY;~WpwM2lD
z=lV97{Wn+XRT^Ig)}Au3<I^T-&!iRj{9YyzJJ8rsD}JTTHz?!!wn6o`bf`5W!{I*!
z^G^|HeY9cZ)|3m#M3G*n!)Gc4Dq1Drhdead<pC(+N`i)_Pty4U`2;h$2(YdDj?w1z
zj{{hC0c4jo2Gr8r-sUbA+trGm_owbQzJ!fI3V}S_3eQ2NuVO<kyj#SF?r-@w6C=?Z
zVAb{$jOjXPn4(@AlfOCyj}M=}vC2~xK_PC8-CtAX>*QA$>i+r6Nq)2$Xrt>pNqKm7
zv|ke;d7Gon?FG%Zo6*j^m|rGVnNn;niit8mfjO(UxEm-nc0PC3A4Mcgo%dQf!{H;X
zc9j@2hhRPRwk+$uC>Co9`if`!M@P`7rW+HLR%G`K=;sAeVt;@*W0SEWC6{(l;wO0%
zoJ|;0+}xD#IUXrFjKNAMU~>`lhvebC2h-MeyiLwZfW(3^%*C~y*pRQm89Sd(q}2AC
zpWUE!J@(Mc7Goe|_xQ5niOPeJ)+geE{ez*hmuG9Qdmo@XO}}5j`AaNLQ7Im|)U(7s
z?mLSPwMd?!*<3aPO&!gxGis(Qi0v&u=^$YQQ1$V_qI2KKH8#H&K+eVYoMGpl0N^+?
z*BNr#08*b{pw@2rUYOC0NBI3n)kx9Zq|*{v?9NY>qN|Jo7|1(5D(K^U*yrDkN4jA?
zjkWs?qoQdEol=QSKR1@s@ycG3vVx{vX0F;49FsKg0+*Fj1S4M@*ZaILkEK7Ee%wfN
zE1KBA5dPd0KPuH5nuCHZjwq7OZnf{8w^#lSl3$6D^`<$Ih@I%L=t={R65unO9OFGS
z5#=+F6XI6tI%XfN%KJ4AeR257ZhJ_Cj3MFJ76mW6HBizt1hJd8W~95^+p?X#mbCkL
zZ#7u@gkWRg%jJF={5cEG{&r+u!^+;y#<xy1bG9g0pMSO$(#wsQaQ3yCn>}YcC*IO<
zQPkA^P=1wwdJ(Lz$~AVl4i(11Mwf*FoGoYY9-{H|^#2eh7!pmZ-)#8h=fmi5zm@&O
zzs?z>Fb6ONHlAd5DxHUi)B9M+r>2t7tQnJd-7qu(sGr5ReenWd_P+FZ;|Z^?WC{5(
zt8#h_v?(TG%A<91{@JDMv>?XQ%*W!cab1cyy)!K?SDd+`oc5LDHN8O_$M7yoHtX>5
z{W*rwJWOGucEM5-v-)Q;Y)Y!~gr!GMXGT}BT~0g_eqS6ym*uLJD4;&ey9~a3+N4(<
z6A%<qWujN#8_)6qdzpY}%J4ZKnXQ#cn(X=*koKq}O~wj>sT|Dx4Zmv$nk@99aWW5L
zUfF-2>}Z&2=MYk3QmIa#n9gF99F0H;_Zo||LSNFPYh=?fRuy+-8E2`W(@5!lY467S
zt7{W;qt+X|4ei!rQC`Xdn@|x`_c@ryupYVauoR~@ilgnY;2dt4FmzTxa#gjSYf0=w
zVSY##0b=kMFmFhCbe76HTTN!?rI9n}C%axzp3=+ESd%OpBE0|oZ)-Az;L&z$uSt)+
ztS9-`{A&X0N8eCff~f>+WJFn*+Mh3o*RO`}i}0zOV?K2-o}N_03AwN!S-c@L`Y#5d
zWub*wprz28wmJrS<@r#G_(Ou}hE{=uI9Pkb%^UDKw?K#2u49@^l1qylg{I@G;#b*8
z`E=+%m1RJ^+=`Uq8w~E7$o8j!_6tK79UP{mj(NjtQ@T$pLlSq_+*p>0KV)g~oXInO
zFrAkaL*@CjKMR63`zRFNL{D+jsd7F@%{j5SzPfx^%F**0v7m@hC7B4Pi%&)O8(U-o
z{2yKd@=}Ap{5(^~&Z0jFZ*=5wGqVFXm>*GIU9-c7bcDe9jSYJvJh3L(3D2F!w=$Ie
zvmC6Xl|&5pB|X60Z16pCJY5ywiN^pKbiP^w)L-bfU_YcXhKYgiJ<s8Ai7@finN|ql
zbS^HBt`j9B%;U$g@pzS-3R>z(fKQ%fzhHgmz&vDpf%uQN=3d>zd`bmJsPX2rDrRPW
zpg}@X!gviR-5u{(%WSH99STa?yFPQjeK3o(0Ao@g-MPNh%}&p&Lh59+D$=yyw#*o#
z8R3Ehe-T?GD-H6acs20SiIv}q_?b@xQIUmg%e)Obrw)yeA5J@?C=w_bvSmI-c1~$E
za$Sg}?v}KE8pK4htb=`#hH_zW{&qeezD_hrEr-FDqJxk#7QY$yTjVgTeJU_R4dq)a
zfRD2byNnYyJTzv7A5?gfCE}ow-K<Xzt%@Ocy6iOJ5%>G#m9j#k^f0B6D~gUhg&~8^
z664xRDob@O$_%!cqsa(ux_|Ieq#8~kfj6B7e`xOs5;;4YBq|nQEx&9Z_F&MR!BFx0
zYu;ekhu@6Nr*cGef^TPUX`P%`vDhIe#<-Nu!uBy^e68W8$2IEG8u;l&*9<k&bCuFS
zD0+_ph|+hk@k66oSdAI_o@fUQ9+;d&Rz5^by9q7YVv$9kYP8`C{I~cW<uf0hUn++!
zBVRn>Oc88#$NY{zk%84$>H1y%mkuE#*iPQCFuSyYrw_iBP2N`YEG2v06`r#e^kI7R
z0Nf_GIzQYKQ#Mi_QbQ}dA`?n_mFHKs_m$ydQ?JfwWqX0`wQvmfy4q)g_acBeodh5g
zJ@m68{dYM#{F%gf^P(G}z%z3H%~S|@Ov_>_#8n5UNMH8b*$UCX@#+`&c)E6{t?#;>
z`UQ<pwzlk;ZE35Dvy^nVDPm&^E_s=?!0CuyE51u>K9+;g`ko3vkv_i%(E!^`!+N9F
z%f%89K`W9;8zH-4p^98${h^(8E?;|&o!DFb=;<a=(O{pBb$T@t&*fKNdnu*=p%BiL
zU*qJsP#KBbRn#lPlwH*?Dni%QPHy1fkFiZ+B8vbmTb~ogRE=qje-E4NOq6#@5&S|7
zC+3)uHfebz)%g{<>1u0E9s+6=dwgBen_&rC4GAtLnZn30C&MFGsXACh-&MHzBulUp
zZH0;rcaMa1$Zd8r2;~WR3Sq3ANohgZ9xYZCjX$!AUzsaw8Waj8BTK4>-3;h#a+AfD
z@EE>av4a@VXb4t;2AX}VVtsUK9@oxt>e0a-TH;L;lHL28wmRIYxm*{HTpsBEK)Jc?
zcw0%L29rLphM@7ecQt<hQl7ACujy~swyEcq_Rd@N9TGovm=<*1u|M8{(@4|hU(ndL
z*%Jmf@839eMx<a9nkwd3^4(cxTl%|lA6z7`m5ztH>~bDrD=!Xz+rAvT>!+D^ex|K@
zCo1TBTia^CI$!=Y56loo;|+s+=6ilO6Omgr1=^0186oA2P$-9lz>QCD6bT1J*bhl&
zt*{*%xm35*1c8n7!(XQp1Ngq}kQ!6wO3)0PuKp4OQaD=)eIKs{NvBHl%#*~M*H7sk
z_Gn+cQm>=8jwR-ixHn7RDa0!J#vF=SUA=Lzedm^LOQX7Ym~$ekU3L{|JK0CwnlPk6
z9rkMez-vE22lr20-CD&9zYwo|9N$M6x0xHCX$;1j0)4wz@Ut`IAtIKsDq%wnlwQTf
z>YKo>yZX7ay;OOOyxx$vzFzK<*nC)KB7{02uOs+S4%3a7;I!N-`tVzV&=$>@<2+$s
zoy_KPkqEpRlE#V}h0J?_iK&$W=k!lWS~b5e*D{~$LZ;LLWy&tc{iR`?HCVzuZaM`(
z;E|G0#g@SJwtT4EW{*PPNe|d|W%In`IGm`{*IxR)zJx1_yIP%Py#e8H-HoN9v}(je
zyUq0$d@E9XxgU+2WWMKM-xLJBcf6+4*`D;go7w}3i^iQEu$mUzvXZ)_jW!Xr@G>@G
zSK|CMZEygA{l>fCN2`&m@yE0n9H<lX*DRFnk(J#*3m)S`3Fu7D?_aR?B6rQ58)qeS
zDif$!izgJjT*4=GX;F}&sL7H0ciEFe8#N8P{lw)WZ+r<eexQzL*hX=fCl0u}6c|Ff
zPT6goW^9E#1-p~V2r4~sdVqpiqL&QBzm?=t38C65APJcJpiy*4RTQuq%wLrJsJpG2
zRd=u1Qm1B+i&GY)u_sy%;FU4KD*SB2nh75s@oK{*tYE@}Oo64<Q2>a#77dYUv=>)w
z-lr1RdrEWSQ`&&!{FPy^od>R?4}MY6yf5rzwitiL<^-``oNLxiZw1C!8hmQ_i91}H
zCaM1MYvMiQpQag_qK6uHj1__nahZgqC1z^`M%pb(s3HL!71-;(q6=J*E&Gd3QC!O*
zu4hQ#>FkkR&UnL1D+|@@dU@|S-B`E*yc!z57nHT0MItc7?B=T-+snawk)W=Lk981<
zIGKzYYj`%m#>-wza9FuI>Qu}lw^Ms?^y@%sUfOc&-j|LZ^^oScZBF;VSdc=swPBja
zV#flcV|{SOz3u4bVd14a!u^v@&Bkm6{+q-?MD;?#5v<`DH$eOCf)d251+zrCv9Ybv
zc;(7FJcfA~GQ&#6G>i~eq~w_$T8LjRR%&R?Q7H6VWh*K~G5D80`d=G&8ejRkw)+~4
zpK|DTaYYJKQ)DUE4UMTN#ad23`}z_*y}hN&`R}dJ`v(4^t%3xg>HZK5Og<3f&7Cdd
zC(f~-7wtbK>I9!qqJ4EIY<lX4(^>%(HYTj5su2?bRBBXZWz(I_LT_pN?F-$1W{r$F
z+Y^1gSlqIFpNUU!s!MTtc;N%ZZfpj$IU|vzndd2#=$O{Z<vKMjX3v3j+6%EL%~e4>
z`&o`r@1AC@`JcYdx}BjW98?8tik4sL%57z9n=OAkSjFdX@&ihCiqX-4d7XI7Q09Bb
zYZ3v9#iHRe*_}2zTOykyQCrz%@_ID$=jNgcG&1N1>PiDM2jyvh+s;q%vg0Xz(oC;T
z>5t5=SYoRc!x;F#aoZu*go38#TT)K~16?!|1^D~p5@c*Mq%WAy-)uU8rTq<WoiKDR
zAIBhO!_CfsXrsR?pd<I?PA6s9=om{5GsE+Z3a44N{S_~XPK-wC^!*-STtW<$_Wj|Z
zDD`x(9Gxh(tLg||d5m=mcbZ4K2%lN%L1Z0BC(q`e>x7KjR+m@Eeq`&0)DLZPQe4sA
zAD>@nR(~g=%>XTRA$LZ3rA{6lWQ$=Ml0^!2@S~$?svJV9F?BQ9-X&cHyjnE{BdV<x
zW2zyDF|_?Ip_n(>Qf}f(hN^UIc6?qiP;DG;#G3jvG~L81|2DBWm+HTg{E@P%%yi-*
z<RL`yeE}0N9U>OR00sYAGkSrK?6IPMit0BF&d2S8M<mwo!<mNZB%gMfLh4@N{}O7C
zdaKcH1|1M}IOBn?C74wj=K7kHvh3`pZ*7grKS>+{n=Ld*@RvO5{Kwa0o6J|?sUa4i
z_QUe&^KLDF!y%I4|BL-|4Hb2ve(QL^`(U*Ce1js~SgBr_+%)%D!jg9Iy9_m}%4uLJ
zzjkX<X?RF3_Us_G%1lTIUMGV{b+Ysg=Jtv}miOw|<R)X<#n$!ZWzS~$1f?jU4MFSA
z!wAPgv<dyu>3K>O?1Llc-oSp}pD3UAi*NdLw7A&Exbn0&eQR32$b}T*a?)u;Cv37%
zGh_a~wahqO+rNLl#SD<N>mJqkg&S>~pv|xqbHyX=I*yqVHpB}@;U+H@dnmAzoWOM^
z$Ilno7+2*M{)03+FS#yFWM_Kp0+t)a+f#nC6qnkIr<WqwHD%6DT=w40qK%G^`+!|7
zQ-I))2WK(|QW>$^$t<K_#OlxyTI-?7$7+C4`X66cIq_NfJ(~F&B0-kZlF{P~;1wg$
zxnf;vF%;reW`yp!Mq)VJ?O~I28JSUvHNVi8kS?>w&4eJt<aKm(MJJljC03V072x47
zi%V7WFqhc0ld&WTIqG=Kdh$)%*~|^F6;k^5Cy6UKuLW2EJDx86x>{*h+Id8mgnwle
z^bhZK<nn3Nzo5aIH#gEaJ%MKFT&L1oKDxg8?7PCQoK92uol&J1PZ(#Cw#D5|IavW<
z+z++#ee*8h*~XAxl(pYY(clt0%*cKx%qb*Bmc3*rTnRMWo4pCynmI(f?f)a{JjvM6
zVZx1`BRes7K#6{g-7&{l8|N%Ydws%k0GbTcU!c5bcP!`YZEF4Y&HAcQxCnvryrs{+
zArcB}(qw(Iat6#K=ONnY(SZfH@eY$1)X_0?lJnFA+J#Ycr1ZCuh^g_c|4ENsE0NT@
zz2p0s`!Y#-82^pxhpknS3h4o|yG&V@HlPoB<;FGJ0))uN{bKL=oSW@UjbuN!0xBfk
zWy5otzF?r(9?p-yb!?)b-xGZa-H&m<v=3lv_R@<(<95-x|0r#|x|GP<7g&m4Nwphl
z$x<9PG+yaDN6nvCLVB}9iPxfo5JS%}$O6tYEZ?TDa?T#$!eLZ=8r`XeL4rp=>+f5e
zetoA*4*3#N;rjV4#=-t|$-dY7tGdmUq(szd+dkzwFFdzLGvpr8DBz7Vyg$L~K1_^y
zjJ6KL2;&9)(kbrKeZ@(X9QmD+3po2tf-U@oqLFJWIh+*>olvcEv7F8tzwp;$dmf+1
zF4JFpzCVCxq;~Ir&)N3L>4@voQX?^|BZM$FqzbVpIbnc`My2|iG>$#z)|#qu>S46>
z6#5KbuEh17)&N6Q*Pg_3(>SD-73MwXq0Z;E)h~fR|1BC?WYh~E5^LATn;zap*>>gr
zxCfxlZj9E_KW$2lvtM=MyWb&Fyz-$&`LaRmgzu97)f)a;<@gxcJSt?N?hsgJiqOZ%
z&4#MtEHz0i48xDxeCq(PY8qI<>7W>^hq>fi707u&$vYFziZ)knwAd(eL(%(N+Ynw)
zOuj`rboos}aEIwi+$;wXf5Pw@Z;rgzJ-IyvE$XRr)305<c0F5mcu3Oi#Pkk!jwV8*
zCX8!zr*E$ojyr$<E1mLmrR7%t^+%7$ssp{?G#lO4f#F5jgZ53%e4*K)P?%h=OC5s;
z32Iy<MQZTh{sq5j{t0mq2XA$1gFqXX9iFy-mwib=@(~_#C%ci|ksxHh@9&~BdT^R^
zWP5(%hZ;QReUZ*X3-oRxkx1sDinG+19wU!FJ8V-~p<lhg%fGeksPRd(Ju%=QtW_2t
z6;j^rVw1dgMv|FIZ~dflJE6;5p=r6)`V%Q_3W=-s->+2{UUyS3XsFGr6c+R}b7Ke;
zLNfymg*T3gXAZ|=Dn5HyT7#}pRj#*=l_>UBayQI&uye=NFQ=>m=McoBSTFseOErxl
zo#Ey7f_HfL*`e#VSHjM1X6)%pZ_fOLc9iM_Em;V8@Z#=ulDJf1h_^&YJBn_*czgxG
z*sdoYxmZ}__%A&klqP$F_fm@S`qH<Th>YAtqVRHj%oDSww_;>dO%sqB<Rmd6LC*G_
zZ-mKH#}g@XDxA%&I8cRm^{^CL13t2@a)oBVOLD>W$FU+^62*97hKb#83DmnfpCYZH
z>tX#s+MqR|d4lMAmw1B7JW=U#5}_$($Jclz;^vqWvw3X3r#{{33XX+BoI!pga|;&J
zMm#@dnt<avpoqtd>$A97D?jkPd#Y&OHp$Kk-ie&m$gV%&1~y^0HGaU{YpOxMu*y7M
z43+4KRR{aqxVEFA{pYGvc6V>B9)<zp<f9&~i98R}69IWCo(Ez_e3YXa!9QtEnUwg3
zObF<9=R}12a;fuB2<qHWjxcu<QYiOBp^ZLMB)Yc-jvnqKxN?E!Ql+7d{`iF27f1V@
zgcpW4UgNYNIyb6eu#UgfOg#0c$xkh2AE}=2*{3Ch5q>5)w-&o`wVcbSbX2O?s@R#h
zbPxplg=ZVX>-$6Z;cw=y{Y5~ImsG%<jp?|fkARun^Q6SgPaUl}0ru$|(mdfQSTLOv
zWEUB4yFB0AgZnOj9qWr?z0xCX%?w**Xr+>6@FfC<;A$I_jh-Su-pOwk8w*m^i1PDJ
zu=JwQpv+}#Vt!xrd;h0F)KA(L`b>Z&Dn9-=k=w4aos3ir7FqlCuMzuPXYYOlmuo17
z+qVCb!w4@VyxJXZhAGG=Za5wT7QIvi=;=koV-%_%OeAtI9&9}KZPuGSR<t)r@VQR^
z2pO%pUDJ>cGw&I3{r9S%N~D^*^l@4VKnD(UF1IwGv&&rx29uIjW;4rK>CzR^vd2u2
zd9BK6C>P?)L$xE(FWP{a5dnI%jqz$y#Tgb+9C6qai4fzw)mq61(jMdm1)1TGve&cK
zs5!3Tm!Xp?*zPXg$Ro_ix%Dgjjo+c;YivOICF@{yqJmQBBK+hfkVnWEguEo?n9$<R
zdLjqcgha2U-6_b3cBLsR+5ore%!A&!!S-oU&9h&1VmHR~q;M`3?2_anvA4xd=%?G1
z`L0s@u>~8HoUjq&6T)G-d$g_edTQX@0i6$!PUuhHw7u{|>fiMU)?$!byyS30u8}uL
zxQDw0qMU!@j2lz3Dnjb-XBIdVI0xi}9&`}YRQ^R}^smBsKYwTJ-E*cE_Io|)E~t&O
z9Djds;pK8UZ_%u2t6PdUb&@WX$|w#yWGGUn>$z2kWkdqAD5yHcKm189f^2X6Oobv!
z+MK!?D9!B6;d_p)nYdn{mhSFiO^^BO-Y0zZ{ss>5Jb|I$^fu`q8YxYhYq}ypTcL3A
zr(?(82a)qTTx*6enUjf5wE44YXYk*O{%PW?EU@z=wj_Xj%KQSkO?st2TULzOTR-o=
z_`!s@f8+7^@qNr5+VWZfNm5?C9OY%XEI>FR9Tm2ScmB!c`t)nrqTw^mDZ<2S#sFt4
z+QnCPF5^@@cqPAI!1JJL`--){-2_a`Gx98LD#wK=;cyZ#65*r@Y{;-QJ1=q?GT;pH
z&$F#Ff!In;JWI9MT>(7Idp>zUbcKwYHkO?2ZPb9$JuepDaMRddK%tAJ?~T|M`m|>X
zDTVzx$q<7XQiEq{`2_6|1Z>*rTN?8=-a9X>o?3V)=2=wTUK7nm1Wzhoe<IWreQL2q
zIhg4|Y%^(zRYcl<`Zt!_<uV{ZI~9^kObhaDq@HPk9FZ{8<R~Zo&+3kp0K;iB^SP*#
zy`rBL2)xKlRYGlDvt$=}lZ*b=5vVfnbHrHC1gAY{sqxX{h>Q{O#n|B2o$6H2O-l|V
zRk_`21nCCQQ!&tJ+A-ur?D-TX<y<u0pN3S<LelKxDTM7|bH;Zi*qP?j?hO+_dL_FH
zJPDLr_#(0F1%6xW{I9r6>U2vojfr}ui34QHy`e`z`+Jkf-Or2F>0sze&f;p8lW$QM
zSO_ac4%T!FMd@%$$xRjhH}k#F36O?YCu$-K|GV<-%k!fIi2HFl<Qv6|jw(z9KTH|!
zASq|@{=eY%W%13jbGCIm#dW_SR)!DzX-#(LQz{W>rfuXgY}Ebqkajf?q=<HJ)=E{*
zRg<$3DvFfk?MBh-s~Ge`3w3jQwhASCXq77ogKg+mplnl)U<Lh^p^*YerY&g2;C!i$
zZCbjB!w%1-)<}VD*a-%9_MQ#Tg?Xu*aJSx6_@yT}z+P>j*VeM2hbPD+W>F~S@8$o?
z6SjM;xf{~k40aE^>>eGKDHfneNd%e-^aRqlR89?#g@VLM8PAMJc6Ki-O%3tV65in<
zYx+hn^#Ws+#J_S>G<eznp1H~`fJ!u~nVvVEMGMXcBAG!_K6)!id;}8l(cb68iyhi~
zCU&?Ub1Cu#IO{@`XgRe!n(T4d8iV|OB2d~{!Ds{)Q0v(P0p4TUeJM1bOm>LglfigM
zK|Ivt!Kzqxa>tC-vU*F35@X?$Z7%A~fR9g<(gUIlR}swUJ=(bWekzvnf$pERxK87@
z7!BG?C2;Fx%v^CR?^K~Zu+`nv8q^R>TcJ&NU~aGHHyckWOxf~&@{7OaqvfF@U3uw^
zjXhnH{W?kG25lQg6)upxQ4ga?=-EJU44ULF4!@t>@>M6;NwPRRm194-%8xoA7bK{c
zdVV!UuG=Br$#MpJrVrm}1tnv^juI+KqZEq8))D;ffadhBCRf9|ZcF*w;5gCA!LvhI
z#y$V32-+pHwLrjZy#B+)4k>=<BN|3Hv5g^a2(5EMCXAPmtz-O<q{v+1UE5m_vTb^1
ze@roWo+BYX)ml(w{ydeFax{zq-du9mKh$25_ADj^oIl>%;!oqRvkAVmt^6#vXK)B-
zxi23cwVR*1+FVL5OT3l<<<Yf}IN}Zyr5vr2P9yIavUy6v`A`klcI(qsbfKbcZyKch
zEfH6$fN-(2gi@~bQ(;+pq|I~?D=WoQBTHI#3ri4WMr7r%UMLdHKUPPb6{eJv4jg~m
zxtPGtkPp!!*H!jss1g}Qpp0+<wVF($PofyAlOczI>9_9e=$P%`2!Q$s)ZnxC4=%?t
zK-wL*8-D>0Tu=e(T7X5bAg^LH%4cjX67%_zSTBV=p_kG(N<Q11x0CjyRoHh$*cuQr
znoML_jH+;qtz3IsG4j;E``Pypiz&A|{aEAiVmIR4Fl(Y@l|_7BHpR%()UNQQtx6pN
z<mS;Vsj%$tIw=gZf^9ZIlnOTkn{U}6wV%A7yyJdaJH(r{oQ7tfW&bcp&{dP@?7(&s
zamz$e2&bsFzaYBBI-HuT!iiAm!OWH?<FU)0*nHl9(m={;g&r{6U3d9-?(m=k1Zs3T
z>a;UqvaD)ef_#m3FEH7%Tw+BY<5REx%u{MZGdAh1U5q<^M6EA*!3qr;&h}c^CN295
zoU9w+_Ow@2ZYrW?p9Z}MfE3dfx5!!u=3YMTlI(0;59x1h<y*!i2yklCt{yIYUo{pZ
zpNxh`KK^^YpJOlArhMgxHzYf$MgC&&Lr;Z?1L62cB?6AK+(i`4ZOKBNe#<K&^m~gq
zMb!SHm#nV1TW^80KVQ9eu6llxoL!s(wZd1(xOZ{!xt6KrN^GXfQx!qKx$0@)$#~Ue
z3Y30Uqir(Sj-^+i*wImN0i~l+b84!LQEvFXQ)$r4xBj@2X#~jj3V2DEbM|0g;O@2k
z!Kcvt?kVERZq}Yls4^+@q+}K^5u1nVNK7ap!?%}UtTn{>6j>)S#Xj%43x3Bc_P%#%
zg(X7pE5_|l2Lh%w%kS0%&TG3^d2@Rq&vU+qD@n)TLp2@kpP-8#=b&;%n42a(Lng8v
z4P%|9P}+1N9ujpPZt{?nrR%7^dIiSU0d&2R<$_-qumUbD4-TJ<zqDe7oJ%QbPYDQs
ztRR1dtj@BYdLo!OGZf&JZY<+|vvaQ8oozpqke)SkG*+nDkaZ|z0jTdu4lNp#y->T7
zsTOk1y;Ssa%-7wqjKqj!?s^0Yzg@e;VyGx#_eirB0b4se5A;baCm%kU2r66k`H;a)
z*GJcxoyT$a)+hA<0|VLuLKPw9eqZ$HU4%TX)XEzyEJJunpj@hm^<(qNzIA>Qqp@zz
zhCWOH(CBh)aGxG1z%VNUmNR?g-4_pv3+1NuouQXT*pkd4nZb;)Ug3{upUPycBncrA
zE;41GaG#eA`(nh2t&+ysISF||`mzwSS>q{u4f=jV5G5aH=sKpszPFTUKsmu-*iBi1
zfTUCbKMtK8>6^tL*mQ1SEY}!jM@EQ!HbOYV2ZwkwH@ufX&a5(}4K!$!jbwQge(*GU
z4Ihu7_Ba^2<0uuc0%EExqiTtLmoejIBL$y6={EA<Zv2Ia0$&aB+Urc%KvQS@MHy^*
z)4$RUs&v;Kb7%C_GSeYZJ@Ckeerj!$bs+{QO_jxpB>4c<nA_=dGjrw~8`JnFL(WQ3
zn?J^CJZZV<q^%yLi0>3@ZrIx>inV6Y5x5(Y{?}yVdiOp7VKeoNIU(ivb-s7A<rT@O
zJU!F174LUI8Qz=os|4NDIE6W8n*!wKI3!Z?KeV&PjoVib<F&A%egyYOy`h2rGTudz
zY|E$43|oPP8LDaQYu_80D%7_aP}MbbHO|$*)XFHB!^!3Ng&LHA^NS0wHiicCtM6e&
z6jy>ePI8g1pEH^Owo-ZsqkorkKBAI4Go+_5c+JO>6TJFT#q_LQzao=Tj*K0Ss{LM`
zlZyA!E~48ruQ&X>U6vSKla)^><XSFuAa<;y=M%^&m)7|tJ5lN;#o#XB$p#j@J1vn(
zfD9Xh9v?N_X|7H1?yB|AWQI)7%Pm~e%(yMt&C%V2Y)a2#J$-$B^;0}*vXfrpG}TtT
z(jF+N0&!UhQFXChKCUYls*?)C>G}ldbS1W~!|l6J74g`1Ln(I+T3@)4=aFx(gyaIi
z92Oy>?Eg<APII}p0SXp9MhYEr1dDPt1$S`^Vu?DNS>g?5Ba*&BqiIQ0@D-TBqlIGP
z>91MV<*Dk}<hn|lNN5HdiGrw$Lin>N7<)gv!q1qF4;-PP#j&_wmeC*Wq-AWYiKP!9
z!g-e)N$^b3)M;k6UE%o-_*bm_N(z>}NE}=QcypNG=&gO?+5K9t7vH}8>Vq;-bAwHF
z-W)(VmD33%wJP{nbB_~8HlV|4%`CTDdHHCnq-2()#{Se6LtwE7zBGNPfAq;V{z=<;
z9^*Fnn_ieEVq3M_N1E0sd|iBfxba=m^x^J7z1r4cL03l@#_{cQH~?CNArmnDs2fX=
zLY+d5mb>X^kK-5~OOp4Gop)-~g+j1SkA)F8rf+Nz!KP9ii6)vF66PPCPgv0OmR0#U
zIDxb(_?elx#!HyfXWdb))bqXisG*ox%l76L%{p=9R~je88iVx}`{82Tw|K!*?T@Ds
zTA_BE<=v#64VKHTHtR~yZK<pQS*G5d7BPPtFjmQH$@QM>xKD?1s(8P1#ov4DXQcWe
z|7-f}og3$>N{`Ik&94USaO_k0Ve6v-Nteb7R=Hg)4H?$AUsRbsvjZm9?=Rk5-sqf#
z&V`})So2e{B|$kuEnujZ7A5THrsdIgh2{5_3Aot$F2au%GhV)BY4|e3Rd!*eRPJ}4
z90NRe#bui}=WwN>aW7%-f1^C%0E01hJ@IB?{{jXht73yD<8E%`r)&AT8Z=O)OYNqU
z>k1mmir|N9n%iEh8egzud;cB7xsdQ%Ml40lVH?pw=gTvi!;_Eyt}4Bb5OTJ!xt_}#
zFx)GAdJlD!^{O3kN<O8gIa!XPS+4N-LL6_oplMZsiBRsP_62ndcOt`T%fsUywp4Ka
z`{BP+2GN{aC;T$H)kKfEn2Cv)p<dCo`6~_)gK6C{OJ~CsT?EGa<WHBqBd;zkmh}Q*
zSa{FRYk^7h>(6V6o2!d#=gt6v3s*H@07YSUQ||NH_6ctoaE1M<vD~N~tJhG(+ih@Z
zc>ZRa`?tyVl(8OTyOQ)?i^0u3g&_A;I5<`C=l#Uc-_;GFmkBO|EMjkxj=J)ISR#v$
zk%`vRRAmY-hz=L}UmY&Jc$B|NhX6%m`|X(;_w*a^?o*0f7S5b?gg0hE76$RogzM0>
zm}q5f6Kvunrlw3-;jw^#E*((id!aY7_co=5NYWo(r?&3B_j;cN{Kl%SulyYCh2U2<
z>vRO&4Z3QujvyQI_p1|QQVm}1-Fn}Cv~)}qt{}K(Zf^DR0@=s&orLgV4|TB^N*WIV
z$+8>#(weo}Y#6J|Z=;qjcgdGk0q3wSX=_sOfz2D0U<B&I(^b2^h?ZMg$La?h^x=Nw
z;(5mBs=2YKR2gJc1LZaT+;`nW=o=h3Dh$T(a9%7b#l`skARvuG$fO}o&DTn@1(ra{
z2v<RC4l3?tAFE6fZaLvT6X!(#!}c{f^vxP!3B0-V9|G5fYd}hpJK0oP!Bw>04_fjE
z)QZq@6~S{YChX4bfG1NBASYjQ(^A1I8RPT-SB3cOQ&O1X;<}#Rgo6OVCrWwO!630-
z&?saE%d#yYSU`ASnV3uhDwZI940nU0gKS=QB8z>iFYI$7zJ3NWa>VBLVJW*8h46zV
zy{kzOunp#{PSBGp+&9a-iWyy}J=JWPwuPw3H1&!TK0*<)@F2libsekyhEH(R!IVb7
zE5A=AvtJmfbXd}aD}<`xWb^g*zB=CR26Psz7V^N7NXqe!IT0+uRkgSIaq-S9rP8LH
z+z?Dpge7gnIQUjKp2yM1;AwQ{$VOctPmcFcy=-0Rluu>@|2EbB4*=vq8^6>T*`wV#
zAI>h38Q|TbT62!0gbb3{J+j}TBbMscjjj`J|L92W4~;IeRnb<cb7F0u4hPmeP0_%l
zey>h8x-3^(DxPuFkyz+rwe9Ny^^Z)JJJ{t3YJ#L72{wL$EVBc>6jb81UY0MxW#<~X
zQPIc_6g-1WBJ;VmLM^4YQDa{KniB8Ha_4y)0Pkj1`Tw)ItSE6UnUdUAIf`e3Yf&$e
zWT;sO{lvRBdx==+PSTKDUtR$Kn0Kem(L^VQH0Kmu)pjzf>biF{o#?N>2aRnx=<O$F
zQ(CR4Hr9(EHUIDzegR_&IaY1-pkv2V5Jo1YQ9mxndq6M|9cq=vidk{KH9gSF!ZHL!
zK;^;{wK}J3G38QR*G~e6!-1PucS!a!SW(|mtT_i^59UakkS0j8STCN@J7qy+LV+Yk
ztJO2Lawg9cU!HE=UaN&l*n2IZO?ms>8{S=>Xpw~kv943+XcCT74uD!CIlS+IyWbXF
z_Q;-(7C`6LgS>a*>vm1-gt7w&Wz)=lC9b#TdFF69Y=Bm55*xUyl;sRmto732zLV@h
z${iRbkaA>pH5qc)f|F}XL4p^lqY}66I@#@p(dknx5Y1r-uvK|5fIWAlY}{9;@bJ$(
zvrYOvs*PZCm*UNikNwoz_vmZeeP?_H1Zr)lD(9yQ)?PBO{ymHr0PAJ^Y$K;M(p7t{
z_ZCs_!~39AYwq|8yRr`%+m%Oqee-iZER7W4Lc*9jVxkdkqEd<a{D4v7biR=W#f)kT
zLu?9unH{NNcxb*s%N7-hlql<U7;Y81L+_@TM|@M}IL*u$S&X9)j+duK9}WjnUIq_A
zciq57R@ZZkqi^)vhQIwiqLp2T2%8l-f-F6%w5e801aayX)zIKG$eMcP;UH-QHnUP-
zp2F#kE`Jg{Jswr=%DYErJ?hmXXep%;KtAB*P1+R8LE@^O`IfWyL^4y)(A%X<k2&eV
zo4qj*z@0lHGX6K5?;hQYXV}*&Y=*}M)5W~*qL|y`vFi`X^*=Wr%$S=kOZyi;a*}be
zlDsU_ntoW0)k{A`?E3zC7DfLPKct#Q7Dw^a`2Z+P6GlZT6f0D`c|&8Hw~@Wz58OF2
z`h#r^=ah+c7Jx?@E;nPHcxhP+-i)2YDIm!+m{OWZ>Kzu{2-m^;QG+Ob@bSbQjuKbt
zba65|4|TqY0SJv0L)q;q+x(85!8UWR&*6~evBq|v=|mqfFJ?Q3Fg|~nC+0&YC!uw)
zpqCBEwkPhM+}PvX5S9^r6@*isFe=Ncqf=RbR*6KG?yT<4Wj)Js?^W$~xVX9Aq+(=+
zj*rHF`+n0iw9Gahy{qN8{r$aiygh)cIqQG4rhFQ#IX-fV&7QqZwr7C{->s`gciRu`
zQ`@71olXw-Yv^PrRo`TJ4Gx|U$?|sK>oxm7MV5#Bg}&-G09&r1leF7run@nmWV!3T
z=Q+Op`28=tl4s9O@m+Peg(1J)k<<}{k&YB`nCJIp886ebw(0ji*H@Q|&yh$TX2b1a
z#;WkNuA98YrKjZ2p-_kZbPmZKk$t~6?KXytMbo+JuClLeb9g%+=Wbzom$#oR%fkrT
zUK-A{H*{(8X>jLwR(S*<gq<Ak_V&2_kIM1Ruo=@I%^6f25ELrKqT{!M1+~Mfg(8h6
zZeQ<JikX*vZ#LMFLZY^9Sgn2ST2N+~28gB&lZ2%-_I3m$c(ht`kX>S*Pz{<9s}5X`
ztSvygoHzj7*L(_@?HsAlZNFVc|CUC^kQe$A+)bW;nVuOqfUMBRSO4pFX+P}P12?4k
zFwdNh2dB+2uyBle=gn?u?01lEPYivW@{qY>o-t`G+UVS1<_dY(AOjrUl^pjjsu0BT
z-mr7i!Lb&F&Oa+V8II$Ptj10Dc1!OE#_m&{H)QwD-ei!t?^Zu8o%}L>`gj4bUdGQh
z4pX8jl;%<XDd#}RWBV0XlojJrSJK>_A;J{yd`6L9942!APLoEJGh&%UL(E93s1=%2
z7SjP-xdFQ~cJb~;%W~n86a41n-F12dReGJeF{O-VLue7s46=3=QgUsqJ<J&nTgo13
zxz$9riqQBx&ooIC5+x<t%{}vF*~dfa)$$XNk^^Fcr(oE;DUbxeF;m$Y(cPJ~*z0?@
z5AQ4k;S7d>VuEu}rEX;S)hz4E<+5_8CpSu?-FHr&oz!rnC?&YPaRcyab>mu68fr=u
z#B{qiYjlG6oP?AnC&g8GB@=>CN+IO~ayattvao1nj(BY`kMg@wGN?=Ggz!ozS!EFU
zW=Z6f=@3?TcDf;<*Z@v}=sx9by!XV@`;LUqQ}KUj$gtZz%S7$RbU^!Mj5#^z;=I*D
z&M6ugJeOm*)Jo2tDffJg3Q=r0l~tVYta3pp<Q^SVFDo`0i&d(%0rWoiKfKd@L#t;M
znR@nI7SBGKNRYivdrR3lxBBx_rPPWfLy{t%E<`0F+jA!6nJFKzaMVguiWzHbtj*hc
zm)5Qw)Hr??m%m1LyfZ1fj`mW5l+%#&ae!|0Yk)^XKKqX5vaB7z62nA>!Q2f>nxd^a
zv8owb+t#>11W9mzf6M7K^ZNC7tR+gP@bca*!D(MKWuA?hg(<_Fy-c@NHz|L0I-zCd
z^}}1LWGb1WMpNUEXR-)I3Y~LlpdiWje(5Jw5EUkzbui=emn9G5|GEp2?|t1d_BP%g
zr0{$;hS5C$$16Ghm^5#n86J$dLMc_9Apa8?b2KWEgg?J2kH|IPdsE2e(q%cPaL6`?
zZ4WqgAj);~wN6eZB$Xk{myl)s9$BU-R$ZTqDq2I9^M_=4Yr3(QW$y|&K}u9AmV`9@
zp|b2*yM8a?JUObUHJ9a-kQO>_C(G<*Ip)ZEJuD7@Ik45notus7!vVH=>X!~#w&tBg
z53}pf7W4xmvpdg8epr?p$WE3e9Ks|XT+7KzvZ{LU?ALd9=(-|L4^&J3)7&08%rhwo
ztDeJ|>>WarH^dP<ps3JRzm^<KIj)VbWm#=s;`S_iCCI>Aix!NrR)HU>le@dSdoumR
z9+jcg2A_ydo>HRW@{&R)xBg&4CkF=lQ90fMKtB3btP&tp$jC(>BfAE=jre3M)v9xw
zb#Mtkb!(O7a;9qZ+ioq^Z~!VHGYJPv1w~^IGF6&YM7(^_lYwkduga7-oQ@7sRU?&3
ztBtxgiW+M!QO;OL$>IpwXxA~pxh6p=F-Y1Y_$uNZtwl!Wcn7aKr<QCp#@K*j{w(C(
zVnX(NiCG8hx;~0Yll~n;Qs(o|UU7GK!*36N4-2#gVxCdmbjt(lmUh5yBr#I#>nTrE
zYb4-{!wuRhZ)#z+=FKN?YvP!_`+CtejwLv<<<BRv@v60In9AV0>U&0(G)xoU>vYV!
z*`|AY8MVe9*!RsoMnkLT3*L5f-F5JO2u^#FCd$&V^##KE_b^@nte5e#jh+UcqFY@H
z4>nnHa%Vu3n{jIuW{D0K77-)uwdw&(n789KVL)0b`NrMUGdLVvQCc|3+5-v_2z^8z
zx*mUeJacGD;VW9sNZmzTq^BJ#Ni<6EHe;%`f|0{H^?0*lG%@Fpn4*<!HE2qjJfhUY
z1D*JFS2Q!`t{eB#Gd3dl5J8jfKW)LKPzEC#C4N+TwB2j}he1H6M;)%+0KJmIA<rD9
z15*ti6Sva6jkP4q5=M(_+eK{f5;N<LxAo$QP~{SiQxZfJC^%`g=Y`D}cX_5+rdDAh
zQ!*UX=`>4lq;1(6IXFrZeN~8LC$Q!^>q-Je>LjrHxwhWvDM~}68@sVyMudD9RlHWR
z&s6Sn4Q%ny@CDgAg7XQC-u^vh86`FjKE#R>FF$AoYY`46Op`wkp#!O#vD|-phJ<4m
z5~RrM2Czy+Q=^%8LK;BtGp833zI*eQ8{o6gPRv=D>gKM9IH7TpG%^f)v3f3rs)N%)
zH%t@O8kPhvwex6q6E8h(+MMsBw+^o!@$JBqJ5pZO+8s!D)N*w9vM@G%O)5Hqz21=w
zkH$vluVEhVmrH<$E`Ph-!xw4DTj#QKBQ*6&J)n0m+^^|wP!Aw(mx=hE!5%>LGzpRv
zq#=^Yrp)RH@~fO#ByrZ{W#qtZnwgUDMys32asXnUg7c&^=yZ^6fCY3pmk`ZOj?Fwf
zL(Q<X;~N=Zpnq-aMEEf|{*falyQy3+$Y1zq0UmW(?wx04>`z^mg|))L5|ao+mO}`K
zPL7|Ar}A1S_k3pUvaCBGdZ%g`eu?n<&0BsCeE#alEW*ryfIcG2ge<$wF#unW&d`FB
zOaBmAra$7Yd?k2#+(wb^QHOK(gR&e2!B{WRl-OT_AU&MBG^_!4c3bFF!8yF9DG%M>
zJCe%a@A<GS_dSxB$29IPK5NbudC1=Q!DwWXH@Yben(``T782)fyNPiVvV2kXol~N-
z@yHbOmEzr@{oXo2He|U2kh<9{;N@hd7Tc(QTCrhIbXop`bn^E}x<Tt+{U<QGo1e1_
zQ5~FZVPgak)y!C{d{B-*W_?H;rt9QJ^;Ma|7Lc7z)}7uFxAbZ=*3wv;v8r)HpZQ*y
zkeIXERi(M+?U+RWS2H7Ni)#@`*~^}j8@-kc&Enk*lg9p|ELvGxrD`vU3xH=l`p7&o
z*U72t<+$^nRook;jLfyJM<l>slgQxeYcj5Lx5ftGHS{MV2yOjX5p>!(wLEB^Q)13}
zaN>3VC`k&NzAx0xqB2z^C+3`@#92uKlL((3j<lM%;IssF*tl-%CxLZ+3^@+DbRmm5
z3wnZK<A)qeRCEL|w|#zGRoF%&*D~}(1SobBGl1w1toAFz<o$Q!5N;0?6#dxZei?sM
z;|0Ka89&>ot%gDC5y_&J+`Ht_o!D@Eo2wCT{W|idcAwjf*nR0i79i<p*=-^;lxiFI
zMX0b);62C5b<@-16KP#pNwhTK5l=!%1S^#6TVyA4mMAwUXw-gZ>uLVJd0G-AXXe91
zJ6D=D(i~BZ$W-XZ>)c*@5NjiUb%Qn8?z+XVsXG8YqoN~RdjnFVoGVGR9S*>ugESVI
zW}(vucfy=!j?;lD6J6)rXmF<?(NRI`5kJ3ZkL;)H8CpYmZ-neAdFIt&&Oy{!#wU_B
z4rY8l%`7CI)~+apqb9P!L-uqnf0PNa2}#kHG1Al`w#y_s0Icr*Zh@R`nD{+3>nkkC
zK21EJ<X%yvvhxuo+rf(u2*K5B-Qh5*%1>nM__rTjj`QTk)!Mvs;I4pYlj9;}b7!W@
zaW9$cYpfeL!^kj~V-Z8Gg!3m+{#&VR)H~f+?qBUG|KEQ17Fi2_{p!yk&CFrI7o0a~
zLXhY*l*D<ZES&CU3}ZP3IIUVygr<T@!gQj=HP_a>V^{F3)lz643LggRFe^h&W^s~P
zGk3+3bpR_IT{c6vJI))3758hA5!e^;kh}hV|4;0A5gncK<XuFQM5ZOfKs95KU3cSr
z%X*Nj<1yZg>k?)5R)|0g8B5IC$Y$heBAHN|u;A<cHXS%5cx&#qOi`}n&gtfmcF;-0
ziF!pkSP{l`__3QjZXm<}>7t}V#xvRE_M^t~$K<#VBaCSakmmb8{!1HGLzc78P0{Mz
z>^2uGWWrcWa#>E#WSN~#wq9=ZF`eu>PD7Sk-==rh)PqSKsQB)?H{^BYukZd0nP=t}
zL9P$TvZN`><o#DOBUy_Khm2ufj${qZ-c;qs%JK$?xCCkkqtk@zzFrQz5v>nd?%-88
zzC)JHWrUFUCKg>mI|C(SqtyOyy4^7`PqWLiZ&vZhy7T*evfN9~`W<buY^tO@F;B?@
zgw~iy%m!1f<fdMBX+l+DiErM7EWfeHC6gS$rVE8{M>;nAB#Ajq(NrSSvtF#(GY8h^
ztJ$_n_9el6cir~bKTs#{hjb@h`GgX%BCFJOa#QDfOPYopbvv)I)5*`}xX(>2pS{n&
zfwFi;6#lpx-c`2i<f_}}yB$|G0IX`PZ7g9N$jp&F>A`D>&bnI}c?4C>Fpa8;C{J(I
zIrT_hiBSz&GMZ@YLFQSMTD_r4%t0Uh-(8NUGzSu#>7a4yHZ_}0?i1$ga?{feom+r<
z2E!%_%Sd8x*x4(472R$5{_ZHXH@BguDGzQbvEY$R5fEp0ppX~ShD=j*FEkM6wVfpK
z`Qb<v;emt-tP#Kh?|hyGr1Wb|o_S`lWsZf6sqbskl^jP2Zf~o&yTQRnDSch`J+ALj
zjD5CeMf+rIKE=4#4}el*KE>%p6j*i3OnVuB+~WnndKo|4sBPP&9UYH9gJ{&Z7)R(=
zRp&tJ5q8&8!%ypg)3)h%h1BSr9ZiIm0#u3pIpKunu$AGt<KwVK16C8>o)?aixW%2&
zWbU|Xv;v;QNNGZ8lxRh(ImpDDd6|`D6bEE99mqK`%?GTlm^F-qDwI;WP<YcGIWJ4h
zf&~A7&r{DG>Tq8N@92uGN7H(AxGRHDyjzsjjpNZi-h4u>TNTJKW>OS(RGd(E)ka$y
z%eoHbq3UH#8v_H>;$-R56vY9OKzxqU-Ux^YiA=7X1%p;eX(j>Z%Z1Z);LpGKif^Bu
zczXL5vXfK(<rkkZi}2>#Z?UA<e84n-E87|7(o3V<QOFW~WV$`u4Sf0~MU+l(M`xsm
z!<HKdBNG38<3k4XhO}(2MO$*=2afj(88whLcTVhz6gX!YPQ>@h@v~vOAG#cij7%oa
zPH<2<=p1NmgKawgNWhb?AD_@tSoP0n;-wPZ)a=dhCdtfsrl#f{%XUQMMGdRI`VC?2
z8CX*yOJlMc$nM1D^2k~$1)~aB7Dl8o&e}6U_9kuTiE1H#-HlZXKz4W(F1JE#m?mWH
za^9hs5hETvdCn8m8(BLMw1afQdbzM%7E;c^BjaE846fja>)B2&bO;mLikJ9WcZj#L
z48{yQ7^NbpHJ;9AKA#T!r!W76e|UPwySLvlMXUNheRan%XMXq3|LCA(V%wBIv`C23
zrXbT!@!b6$dKs$?8Fr#=Mi$M+^Q7(+u^G}I9pCHt4mc51T}OQX(SA{*MfPuZ#(Jyy
zL?ZV%fGTSItLbD*7KGd&*&V-Xn=Ho#jRU=Ye8OsB<<Dp-IuFZ-Ww~@&9!$utyR^9r
zsw7q|z0R2W2gq`FR-0CAWC3<}*RwZ$RrbGND<1rSEVEm?JwuQHlWae!GLH7M%d%ON
zG&MXsr&|sVI{su?9uKw$$sBz5k|Wp4LnrrgqaToE@V!9UnpONHoqYAObQFwTapEt2
zw5oCG8+d2pa_sRu@odZQljGemV+4Hrv+E6bdfyO95)D~^B%};dZD0(34I2RU?c29Z
zD4%`#nU|RjaMDlcGSk7?jHdW*a9=euOe@+X?xUhfcq%Ka85eC1q~*+67cRB%t~_xr
zYg~tH<{=JNoAhc42>{B2%3VKnD8Fmh4e0gVM68FA+3BY})YUgOU*9CzWV|0MB_w5P
zRcdX4<L>tZpFa&wFpN>ImvTmO!)eWWwq@8HqbUn`X+~p0$Oh|j<|K*#>#zTe-#)(M
z>D_luuaGjI9}Z-JcaL5iVmWykf0qVfcXvTgkaLs{=3c14#yQ*lm3`;KzRoO;(ajyh
z<l~m`55Z6pYtA3E%W|W$F}{pH_VEH>y^No2^e!E}tCizQ5v>fbt3{$t4cCa?iq*Ew
zd>F~Q*EY&0_dqp1V5DKwc&3ujx&yw)Qv4I{ei=M1g>%ld!_hZZH8k0<t36}ivnLb{
zquJYSt3ndb;HcI>qBt)>nv!#Cr8Er9+LU6Ab6qH<4n^C6J{$KZVIun`pKQAxx*8P*
zcr_F#J<T~3&W6bBer-zF=hcSdHoQ!%G{s7z)?VVkE&{bprZx*23E6<M@%}IqH$y-O
z5>j_UMg05KDv}O-c6-OU6f#OupZ70rZaAb&{@Nk{6<i|**2^<?YEW>;Rrc3o+m8q%
z^JB|;bAFR7j(+x%!+({roBKW&A4-<~B@We1@VXp#r?r;{*ZRY93~UxjU$80u2=LZ{
zC)w6Y94fJq5|1Yxh9U@;wJ@pk*>p-c+pPzSNbV$g5Z-#3QQVWdEH`Ty!qU7uo~UBs
zwM&rM7j3LsBg@matg<l7`}v$ht*6chlofkJUz3}xfb0mL_t_FalvQr!4cC!Mi7DJ0
zg8R7?YH)>Yodx5a;cs@{sQx^H6XcmZZ*shyUd)YU$QV#cLDI}uH?MeG7A8}27~!9t
zjvVu#>W{S&``xoeG7MdD*K6bn#XMh=H}B+le|@+kzb6|0=|ismHNc2ZXbf2<pcM<;
zK^%-0k%)x<E9vAR$)UTtv(~ofu}z_`vddb?=2_HDWXN+LU6!-t=uqpjZ0634Hlwyi
z!@-U<N?Xz&D$8BJg*+yiLdJUba*Ci2&txQ&YOGU<AW6Rki+S5J`F7)Kv)sP%v&!-S
zkb9}4fcP5t$+FyUFK&1=R|uZFG#(}cyxLy+m@G3gvHmrrZWo@4^7T(=bUEJhx&k0L
zr=mcs?MLZk+kLk$<}GY$(e1XcoE`*-&+Qkl^d;|l@8P^~N{RZ#Jtou0;u+wOnZs%#
zPOV{H`Ot&18Vwan^UTU(4Qq+DRaT6m8kv;Bs+C0xms(hB-R*H*MhJi>jdI_O{vsTd
zZVStE8#io5+uLk1=3LTlHaC+uUfSetyw!uwYS*?^d}*JF)xRWRk8S~L=vlze9)GwR
zhFKeSM{9*N9r?}8E8eXO$&`uAEQ#AZMWA=L1g;iH{}q5)2gf60R=Xk@3vsi7daw=B
z$cXND**`O^J^LPuxL?@SOY;6(V|~4he-GmYz<L=!+h__Y8M#|Bj)Rba<LL+q9^aik
z9T=!0)a=orq(pfg?AV`u?2TE^%%^pbtSE;G4l`JA+%)TH!`sHz(^ZpXR1$yl55MDb
zf5)HwhyUQ^3E;i5Q)D%0W|kRCctn>3$(3xaO4Aic)TWSAw6IUKRw(9lnCp6>He*rc
zTvy&K51iM!S$kqb$@(=2Vitya_OhntUHY7xt2qJXZVMM|L_mDX4NN!Z&%G{pI@1)6
zuXSG@f_LNfy9d7h?mNC}w^%O;lMQr&wsjxYFf>zz21x~}6|~a34bE|oEoJ5@GfgQ<
zOq{SYkvSZWoQ}7Aar*^HCJxg?e)E<$YvDir>0fZ1Xa3)R|KCtuxtNntYZ&ewD1hKm
z!wuu!Ick(Bw7NyREizf+e!P>7I=2{3jB5M{%T|8i{l4_%`p-{g)Bq@R-sCvpa(#1i
zM3V6C?bDB!<35JX7m-FKvUn7&FlU%%C*fDbxwm@h7~KK)D1AdHl6Y7Or)lCe&Gd*I
zc<Oz|TXRp7JEEHEAoHo(DAl6;z<4SJrp#f^943K0Ip2mU@9N63UU*j<k7c1YA_@o{
zpc?s6DQ(vg*+4=#Y~hu6VBg2aF0^wOHve+e82Rt_y7zU*Lne>|)r8UvG=pW33{*F6
z&6v7Fs2h)use)FpglWaB;)ZyXScp6CQx3!2yt##>?9Pb6BokjAU$NdpA@jS3cRba~
zpWpwQ!(ry|4+lzHsZwI@202$OlU_Pv<D6p+si4gX`$uU9z1v`oy=oJ-B+1Ev`2kV=
zF>?IuL!Sx=-A8M#A0)>%9u=2ajN8)*Ny6Lm_^+pv0j=t?JmFc52t{ptwhuLEDije`
z$vjl$)Rfbdv1mZDm*olPZ>cG<$U5YyuCP{h!gz&CU7Y-Rp8qge4q0lw#4mfVW14y}
zD^FDQ_VmNwGdFVln>$(FIlrok<Zu`z<W&RYi7c-U4jN#Re>PbL&kakdgKmC5y~E&7
zmgS*#ZPUX-ILZU*@GM3Roh<SJS>Azfl9*(2^7?;``g^(h^+}BurCd2hX=P-^XEE+>
zk4P5EoBG3Yj9!7!ZV$r9LL^wT%+R7t38oCaX-Jg~05fx0sU(WP?_R&<d_MDk`-?x}
zMiNus=GE=1gm>yTgnabSu!*OEs(GWBwHXg<Wo^n@FU*Gn>GqBl<6?>inf=>x=IL_g
z&C>&`+8!StLp$J<)5!km;1Ft2mgzdD2h+MOJ@{33o1SM*r|b38-|n@DF!rUCGASQ|
zm#mS~#FQp%2go|O8ytU<MrX#94O+!)!D<0*-ZCKtP#wO`lmpa^ZjoLS;oVs$gkR?q
zZ|56+bGkwDM3#x)obHhX|3fm>6z}2ZyO4zJc~<RsWjsGmy3`2RHJA0oPKLLDfk+yh
zm3#V$fBD#fhcDyb)p!B0UdGQhM%L9XlHdg1X=HAU&qk!RI~hH$U>`9ro^O9ct%UJ5
zO<u0I&$^QRm%-Cyvpc~Lmxa4&MpMR<n<J%NBT5vdaXGqR4VJC|M!apXhGA~Zs#sG*
zTM))3N^6uF-BMa*Err_L5ZFgMj4cv;rAT_rp&NgFl6Ma6cC+z&7vb@jw>j<q9rgdA
zyumnM79O9@e097VrBBb#w|_Db5W<TqBCAT=f6wBb>`0z@HRT+b-IfB~Go+FTQ=U1^
zNA6EI+&`_fnz%jOaF{1%naJWXMUP5v{2}97!brC3Sp)W^3LT7dK(0F%=_>SeeSaSs
zowH$h8{aqLd3xae)6upgum<!<Wy%iFCi%nVcyL0*2RgoO5F9Mw2hPGYdB?_&&8>@q
zAHM+)mxY_0`11bX*QRnU7GWHE7o6mvF}?>>RcbSedIvTEQZurFNupY%HZQGRta`M!
zl}#K|IHUc#;=U~$!C?&cNb1!DcVMVrCs$!mk9=-&7s(q}K-j=)<5Olrsg<=>rop{4
zj_yDP7FiY%4c6`IB}n84a0Od;vFv3ZDW1143UiWg^S$Kuc*AF6{5mO@(nz*&%rkRx
z0MQILmTl@6)9-d$D<bjM&gmr<$bP27QC3qgn~Djt-Eq=x?w`xwzXv&n9p1FN?3`RD
z@3(pX?7v(myKMR6n6ivwy1ss|uPrh(#DggBE*DNo`0VEPdu7?Px_i*VLY6&4qo}dg
zije;>S(e=$TQ4cvX+GI!8}~Q1wIkBSa+BHp?TUTGiQC&MHYD+9lVvbeGC8Ocx-84!
zr}>Gp+;7MBbF#N#T=#BSt6N6cUX~>?H^xm~-SdZK`9}{Mhx%7_r~fU+?$YU3o<*V~
ztPCK^-~Dkqc`w%c`7nu6P6f#la_n-}bx|Lqfvmw&Do+Ayb2&V)$td(aIXVAQN^#!K
zEajY5Rh6YR*6Kf&yfUDQzh9bCF_w_uwKb|m0B0C3v9fG~pGM8XP8Z9E=3y5Ay`GyP
zoUsOXkHWS?m~9%(t2mr1s+rOGILGMO=)k%bE|+DOT^TYNe<!h#HI*l~c1@<x&vAFd
zN0115!`*z~i<?`1(>{0pGE3b2|J(bQE=iIj&lmfPs+oI4WM&nB!lN5K-NQTQHp#{1
z;g$=o`N_E_&*34@o$2Y@(-&P0pn%F(JlxGpRfG!>QB(7X$Y*7>N0t15tjzFmH#bu=
zQ56xDM;^^tM>glI5mQ^^Vbx(){l#@?>DFc+A9F`{T?kuUN;`JjOn!t1hQGTDkA4bI
zOPYgEB^+467ek<SMk#B$$K`&9XBRshct$DJCOO^FuOKX{31P5^Dc-CG>#<uAO3%!*
z1v>-Sk0Y+Gb~xS~v6Q+RRRmDB#j*%c1yXg9e0jTIH||k(ml&iXl{@gF1sX4)J6dC6
zfHJOf!o&r4bwL$D5y4?zP?)iZrVQ?gFbH8z87U2*{fOI}YaEsd*HgvKa>VuVsQLah
z_-Lt8AX<}W05Cv1d_Y`b&$@_lXPoteV-B@MYuC2b0k-{wr$ZnKgfbE?P_ckf@$%Ih
z?Dzuz(?9$k`|(m0)G8VQCaYra3XCsvLn4d`T0}bmd+Rv({9*@637iI7^VkdzMuHZ#
zE{7v(zC=dG<?aG03zlhzUnRz6dIlokVz|ViGKB6l#giD|Js<<TibYCZTsNOY45(V<
z3$3{GN-9{Y6~7!>be^{C77S_YAZsJ&iqB*ImCYKj3-zjt|1fC~KL)P!d6%XP;A%hO
zazA2{g0ggS43*;oU~QJOZMEGiBhAJmSdmyX{bM%_xVqY5I!vIYuI|;QQ|S(6<${z3
z%vJE;|LsqB`{EhC`9J>xdrlZDY-hMaifDwOq_)c`s3^5ys*GbPxGNP4Gj5JYP#z(>
z3(zRoQ^s+baX3!6;TcDn@cM9rg{%8Bwqqs?ffh_{W|!Ukl#;6H9=TZ23R&ry!Gg9J
z-g??AZ+Nx>39OzQT4XC(4H^c3stWI`x7T<vzQW(_e}y6KQLBC?TkC9+idTRA9+ZH&
zEC31epoL#|R!fC)!f2<s+wG8Y!nhyWT%$@VAW5vAJ7YJD_}#C6i?4@@-=qo8|M}nX
z{rA_nq>NG-Wv(c^V95m;m&kMhGhgBE0=?9Jua_WJpxau)*@MHVTu@~}MZv`|B2&_u
zc36wZ4#UtH7VbR9`{j6ZU!SfV1AzMh-M5heT<%9)>_$v=LFL*U3j!<DjE7|z`r~@&
z2i9b_YldOK)zye)UQlbru{lfhoRDCKUDXJqQt@B^^$&RY)pLCPKmQ~4AnYEH<xJ|V
z%&?k2b-`2xhgxyH%vdVpI31Dl2pP-A%d*(gK|sk2@1N&y(hiwuU2M6kj^Rbz{cwWS
zCziyX=w<`E(}sY`)^%-N)wyaK*O!oGN@Ri9mXdGxH)Teu>62x7TL9b3FAy>Y_=Xd6
zwK=Q{!#k1X-7q5OY#W5)zQg7`Pi7F`b7H;AzV)x_od3WRlVx)GQhjj%aCN!I<!;1O
z=GH0reRABc*KbvC``dT)TLr_YYw|K3QA)*R_ajl#Ii>1%E&fbb#>?9y5)$^$o&g2m
zWwDx00|wG@*GMjX4wxBpWgJVzp%h#nkC>JM2BT8O;aDI;h71F4=LxsVgtyZjrsbe@
zTV2s;*QIl|)}CV&R_9M{Sp^oQYYrN4zvislqq?DOj$pX+S|_iT$|tJGkWVCbOdrxD
zq9sHMV;puEWrW=6bDf8boCsM2|NigaVi+@ODJI*DA&JT{5^@%kW5RwvBIRWDwtUjD
zx|*ecnQ%M^FkInxzxoDWjo;uujsF>c`R*-lZ;tqC+~ZItZ34g*ECnPZw1JeBKCq6b
z>gZ0$mUdZ-70j<XQMPgj7JVwLMgMo>t}E%`a=WJBzqc;s1GtW`I>GGkPdHZaGzkY*
z@I_GrRgeit8MxRFxY+M;do!UFhIi?LDI#30u&BM<)*nBZuT3^rF*+T@=2_oibjOOP
zpp3%+0k>j@fE$Xy%z{H<ygkm41nk_JVRtI6<`kG9g-{kjSv0DuSl66sS&%q^h5^We
zEZV(DM6eK0l3<pCMM?|N-dF){lZjS#S`7{}oVxevHCluMnN@4^cpEqN-BfD3H)P6k
zuVA;cAOOO+y}iX-dW&VQS}@dA+tV~K^j~W!d5Mt$j45M3Bn*i#4C@zpm$9`IP6Z_4
zBJc2QcZpwo^)=q!yv3WFx7f=b<G90s0hL)>sJ8)Fchy=>Jz!zV(v23a4y@+9;->hz
z5w3=*?Gj;Cihb9pUG}{N(hZu2)!)slri8b`lTQJAv{Zpn<L6>O;PPUR!|l?^F-(r3
zg@N&i9IuCMukp!u`EF&CW&cCKn6srD0Uj}<9ms&|X~yM#j|GI$o_Vh|(dutlEjy_M
zDOwP>R#mj-Qm|N>ZX!l470X;OO$!dwjJxB6yLrNKnNd)+w$cN}y#J`X)LW&huR)_j
zMw@4qkg#qA+wazg551X1{lGSP(jP=BZg21KX1u{~zW&PO*lYBxN<P<WX&ur&tClvm
z=*b=Fy&s2!l(i7E8lTk)NsAsw5-#=^xY%D{*zHh`H+XY<gS?FRDu0E`;R+KL-90|p
z;q`HnHGHLeQnY4?*Wg2UuG0P7KU4ba76-ScYq_!e(Q@qf1W!|rdqm}U$iT&J#O3}1
z$Ad)#cMIx!W%)g8vYK}$%V~`uCRlWCN)YjsC~0d~72xf(;9@FRC}A|Cu$N`=dn(D~
zoKcHtH^Ex4l#01(jY=Z<I9a|wTv+`b-J59rM~kXjJ?v%}SDP36iQ;6<cNtiHK0k^q
zQzFoi%u(k~U4~f4j8BzicyyZUQ7vkN?Ec$Y+&ixxkmYtnd+6L5?a^<&<Eojk%@Osm
z9B+Qp%>~_Ucp3<FLn*l&XI$)eINnWK%Hm_>80)3`-|o2bpEE24G_8?zMC!g7hKyR(
z+1fNg()J#&Z|^V?;n$bXbc3TQsN2~eAP5s`5zO;~d0sFrGnTTTRIM#rtDr111S2>n
z%w@qe&zOp)8Or9le|XH=)YV53bxj~V;+HnhvKD&By)~;_{-yhekgYjsJ&x&}hr<y!
zHwRovLan^fGyRm7W!4VJwn^o6xd2V$gv&7z#(}gBSITWeOJzlWAWFErxWInD*N)v>
zK@%fSz%V6TKHpo@BbCdxv2!8}twB<&rDpOfo?6LvuB(pk$5IDBw`Xu2+LLJXdpy1#
z16w&>YmLb}J`wzs2nSa1#b|n!X#I<eJ@DdckLQ<H`0n)^9FGO9G_QN)tF@#hSSK=Q
zG(mT~wrKFC7OZv)%;vDNu+>Clt===PF0QbY1&8BZTW!2s$w7d(OU1XZ-{N-{JN)uu
z)QPY#itwtuG9Yt8nFaH);#d|O>jKW&n)`4(fG7iL4`g6K#!|FS)|`N401iB1s*CM{
z_g1HA6+&98Bxxbs1X&lyPJvFo%qFWjL2SD)f?{o+h^vMwvn^1v8VUZjNi4%ZLF;))
z#mkp3QEnKwzdu^duL|!h*c;!jM8S5UIade)!*0NkGoFtbqor1{sBz&+b*g|cfl)w6
z_~rg<{Py{;@DIQLd;IhN_!qqT%Qe2a_y@eW_y(g~fa?)i*7n7oKxkLdUaeNOUn8-|
zvW?UT*tY8S9m-XK(9mf)aT@f_i1jvX7s&^_81l&PMF7fAW|8fUh%o_|7r<A~F7WK~
zIsWqY8q-FObv(IKCVM#su*vvrdKu7)|0V(kbChZE?)6@zoN#%0fm#@cshE<e&%rE$
zzaD0!-5wJ%Mi4HfLTUw9joe^}HdUwuUIcX!l$kLv1#_8km}V@6wH}uwEQbZa1WtE&
zb8~~+;}I{9*O=<$qKY--<m8TuXxA^R7qZ$D43MVBsG_TNKU!!#sX?w$9H~V*Cb0?~
zZj&}S)_=HBXYmbSDfsKlSHLmjZ}TrO%Z%ndaAO<PA`Qo#-a;0OY$V1oXsYN%*62#@
zFq#0e?r?w>$7bEx1`YV?#W#5I;%kii3%vdFU+_==^<VML#qaT({crH~{+Br5h<zR~
zB57Oju2z|*C}<Z+?u~_J95B`bI37h=;T=@Tc80yyhTewNAQt=G=CylIJMO)DpQk6s
zjys5V18}(qzJ9K<{PN}+^W=44@LpN|pf%b43<NWLMK@Id3`DrR*r8O$G*v59sT(&e
zz+dkcq_N<bN9>q@6Irehtr^7t6#z>`T{MkuDie-V!BVv|8$&)pmbY^`L!VGFA60Xv
zl6U<j5UVlSr>_eMBvswW0s!i&u?Ya_dzGSne}eotvYZBW{;}4T(|>CO*Qd&Ii<~3^
zQbHmfJ)g^Tw3@Yz$7I<X>UhUBhDMg8Oj}u|ZgbB@SKe*=cDU2!B!~8#>HIXi5xBY}
zym-FD<?aIC-@e7X)b{jz$eQd^Edgy~v0!yz`Nr@_`s^$=wwY9i-tG1%g~7FG;}dAv
zj2VP~`~DT?X~EzB;v0<S*wM&Xv}ncx6{=bY^Hg!Tn{azP;C4P>nioxXOM4J8=DRz{
z<ppFn;4n?Nxx2&NJZn8VyWYMtbw6Hyrv<A6OWf;Ysdr>4y8c9w)pSc%XIGtr-ILON
zP19#=!#T;iZbP)*+pAY^F)kPQ&DC>Ev;h80TXa(bVVWi&vDOA{-%|lB81jHIXIu=L
z-H;Ml>4prc(mI8ME2>m&!tm8E@b&W-sEIJ&EjV0XqkKPu4vb&??Q_hzAPZUd!Bw-?
zWm4dJan2bU`R>n*z>6NP5p+6>vuzqDkk+~rtux!)ON940uik|}nWytf8~Wg9Ksd01
zFNaOaSNj3G9pUnlP<g@i&09@XP>Z6inpFgIW$XsAsFyyj1}!L|IbXyVq|UFCGg2aC
z(iDY-1St(jdB8Z17Ui{0puRt!^A1%A-~Q!$j9<UN|8)5q5LM*F5LSh>TDPJq2`m{>
zv+f9Wx%bObkTNhY6<8`T7tBjRDVjDoEe9OR5tH{b2Ug47Ct&;Uy+xL6lJ31(rbv?z
zl4Si9ks4M>APM@nMJo|ftI5+ZB!Fm#CW3jg>`c6%`H*b^FVM1{+~wMvT?swE1r<5@
zuB`sP)}n<uWv99V>#;}*QXY_L#1-vvx!WVB5umK0;24mxQwN*~h%3MaQY(ZK;Pr0W
z4W*{kizyO1X4WLz?FN5n1g6Mq_brGZXRQZ<x)zmvDCMZ#zI%?U2TbD5(Cc4&D&$PK
z+z+_eCtO}8uoT?hU27W1Mvk?(>0B9Of*NxV$Z;1m&_>K}i5e&gIh!0qPzjpinKO37
zh*=6mmUc`Z--?Wxf$y(xA!)!fWSfUnIU@*|y{^`v&qY<ANU55#ONq2`17!?VQ0EEP
zcfjrO7I#NYOXgk-n|8cvPHmUlQVC(*&sMi<XE?M`(C4==!{p<V>|YN)GUB_K(#&9$
zNx|`Ohda3i4G9?;tqvTMIRUthc{_bl1+>%!ISoi;#ezv7QXoQF*xI8gi2>3%AydZl
z=Pz*e>^W#0!8BlAGM;_)OMLVEZ-IrelnGSW94Tt_3Rk2-i?564Lt1C5d!0|aNK@_-
zVUl$ZZK)o%TdXy)JXQGd-P!wc+}+#lU3#)|45%z$?KAfKjH}B8M#0_TM$`A)F=N-H
zvMleE<#(;gl=RujIU%=nSd0G28T;K1G#4CMWw~F=D>7yQUSHoKr2)^x4arrO3o}><
zA_>9?ybx;9bmx(h_F<6$2_GrT8>`e_@u;t~&RAZkU6UATjks`OBKw2XZOmzUegE)?
z8ONHwq~GsuurDjiN$X{aCx8pQ=ECR5GIDP6r_a+#{UaY^9}}_Y?FVGJ*4m;}HA>T*
zS>jP9_ffl>PG32}XG>^wx;X-#1f(_0UR>_5-xDtO1eS`s<DEwJ0O$+dy-$um<onV<
zx-OgkSOJXVh}LFYI>(1Sw_w5bwBT~LV1YK+*(YfN(7Ni0t;wvyp^g|!k#%LkQf!aY
z9jX*8(CeL+qo%Pn(KUyZpLl;6q_QNfX~C-7!Q`!Xs;p{_HuO>FwbpZ$+GatRW{uX)
zfuS`a+xSvb3T|$0u`6%!%Zq1_>^d@|)&(pTYgg@6KM=7xowXn*z|;Fcgi^b51y(1J
zP(f5dBuELky1K%})fLj^9?Q)UZ(lFCOkd#_&o7X91oN!L{RJS(AaS`}vl7&a)jN6D
z`n4Vn+w04K%dtC_)~pEWMy`MVqYvMP(>Z#3&phZe_~{c4tl$fxitDqhJ@yv~7ng+N
z+X;s|ov7V$qNw^(cugxjorqM?uy_0VgWm8ZAd`0ZAu9q2!8jxgL&mrpwY4x~H8Abi
zIS;4|{PC}^@HJ(~?=nDwlmv_fVpR~T2&(MBoFSaStPZSW1q$l|lY}&iW2umN!89+J
z%YvJ7z*G)6EJxkNX;<aQ=3l3#mO703$v!nD!HsU%lA?t;1h9xGZq6n%h0QV0j4^`l
z)It~Qb--2O@tPCTqOr6)SRWB0qU&~rAdQSf6<Q!&L!t#4Vv4@rfCYpFL_kghF2@}%
z#vM`~0LsvY-UE`T150av6`)o~(b#!PxhcnfQw9A+xiZ76HsWLKz7JcVm#Psu<G6=-
zR1A6r=Mnnd`iJe_drPoB%TGjEhAOVlu6Ed8WL#clOxH6Ghr{ZK?p!S_T3nUW<8r+D
zcjruWIi^gl!;as<am>g$V>b?{+^2%A2V-+!y}Z7~Zhry3xYB)fdsf}(VIcs68gZ4t
z6{rOmC}T)a2Nn`?W!x<@ZZPBaaEHS@p`cTifsLVXD$mU37=|nL=Dcz*UiS=`zPOG@
zH>%BnBkPN6)CM)@L9!jsDykG8k4GHj4wRD>jCDg%Yuzz4wTMohTdy^gk(x6)5!7M8
z+7VLgm{3!e6A^|ac=r4`E}lIDh5;1=O3k?1{}R7?{yWUeAF&)~?c@eEdR2TZjMJkZ
z-S~y3fI@gv6x$yhk*(27L@0Lc!!X$ESshpH4<2SC$K4HhhdTVU<QQ;SzIry|;$pz%
z)rk3K!QEl9-EpAzpt4jf74MYgcdf~ZfHbVmXxQ(DAt4Xyz~UAm>j%`~ucQvFS2uUq
z?e_T1{_;UtuABhLkea|t0vB~ikEG7%#G_WU`Dj`8E7ZT31=$T>a}4=zOp`=2@PtL9
zp@jf)=aoekW^c$KO@=)Z9PU^UcM5%BSr$SyS=RbqzJdco8b4c>VPitcWY3&s>L}vP
zm_I1XtOjv*hnOuC>M!XWo~JxCCBJ7)e(bfggfwmM#fwV}BVpVD^X-iJctl}!%5EK4
z@0R0JM6d?N0{1qROzp%tj(VSpJ0}Ic&EogxdMdcu7tBJ~B|=Ww99FCD#X=BH7CA(%
z;saZ~Ds{>dz>H~{F!O=~FF4FcOv|*1xYe)c&GqyVdb$?qGpdnbU9{FoO1PLZsYTMw
zT|+LjI7uPz(hhw`H+jvhwT|QM%`NggqWtD>z?p0=U~?#!-VL^SAejXf0+MQ7M)mie
z_AL;rHJ6}H^=kS%aB*>o{pA(ta)*TzZr&99>K8BYt7pH0@GZCqsEWm%xsz&rLOlJK
zD0L>AYi2f=*+a#stAoncaR#j$z}=C>pObD;?!Vc`fBUKJxpg=B<a~plI^n<yz8KHG
zy1+POTwLwID7ZNub*C`vzIn4uI}d#ot%p$w>!jB?NZMwMw6)G6fAr5FgaO)BX&83M
z5bSm%5@po6pcajMS{EV!U@S|4P{Nc(9GP$j!k#jANsU`(L0t-J5iIkl^{<N7)!@ni
z>%wt4?f@D91`gANH(YRgJm5G_IO4{Nq%$~^Mi}@7oN8h=*;9sw-*N>dE%GQ{huq%^
zm?7*AEUO8`f|(bKoMH@|kS)c5e4CcEaI`BB0tORdx4*>2#T6=mg&9&9To|?5y~{8I
zR?nND%KbP17dz79EP`T%xtCcWsGvN<Z~cxc@?h=O6XQq|-YnPn_SGNpcD_a0?Qylg
zz>D1;^E_il!S8?bJK!+l-)^oUHERG?77NzUg1X6S0g&i6nXs%Kw#XGDsDY6HHy{YA
zh-=$)Qb(Fp)E3P_?Y_7|R4wgg-Ejx*FS!2i&N_QXXT5Xd`HKsTW5&hR9*hOocLy!r
ziBma-jjx(^mA$O~19EJ~w&?RF22LWav$PvB^0-3+<6<{}h%nCu%UmE_^}e|S6F^QG
z)57>4fBZ9sU;GmP$G`ikn=G_O$FdX@PAJO&E@U-wGIHKwF!ftV)wm%g+{ujNJYnG(
z73H2|dCU4c5o)P3fSrQ)lR)ljZDW&>*JGGGw`eV&;QaX#SYcIaQ5$56FYNfR6d|F6
zr3w}jOUGdV8Oe5M2a8(OTr9mwiGZ8|SdspPRbj4WH8wPDfiPr7N`hJ#N1pMg!%H9o
zzkB@^zCXM|#f+=V0l$2HiErN?F->>4-0v|}!FB0swVe2l{7PVsKb8XTZt8NZqPmVI
zW37d;dREvBAnjP!(!WG3!t^mY-a_pppQaoG!1EV-j6=fiVuvCHudi>wGlSVvKClU1
zXqyu6lI3@-$sj^bx`DLcjYwk!Az`=I2!ds<D2s0@**t9bE$0z)5&Uoe{wJhg`~v^@
z`xhs&%vefA<%FdSC{=6dq%>k2_880oL_7i}iytS;0Q?;S<WyBgkeV#3;bzaQX!?#i
zUpx{E@N_1&2(E0#CbI@+604IB78ylY4L!E}eA4bSUq+TCK@wT2UqL2~*7+=1Hc}!m
z6V`5o2_hAht2(v1HRGNvdjts92n(_eI6$Mav^ynl?qW;r{9svbp1vm|%P*d*Zk@)2
zg%{l29KnmGoVBR5Zb5ai{*X19-FehaZHW>FN?Jp09FWHm2?^t{g9u@&6KWMLec_uk
z&|?-#I8?!(-`=3^c6hPhYpk6x3K#AETNtH6qbQlQ{u45gN3FXF0?Pm-Add-CEx0}2
zV&<X_6MFx>vr9!*tpy&D%g{7iGRGdlpI1Uk3G4G=8?tTfkj{2NORFPy5~AC50)cWu
zPO4!r0Hgs#+?=|qyOM1fn}f>I<;CPX4**CbDkppXwMiD)UDed!Av1<7;G9s2@$Fy#
zh~x2y%Zm$KAMPME;hPuV;P1crJ-$8sckon@k&q;*GmEqlRMRJX(d{;ibr{U)Y&M3M
zS?O!ZHuI)?4QU&x_2B(-r(_E$z0YaHoIi8H&x3Ga1z(JDKVr-YH0Y!_FQs+saePr}
z|CQDG0(y8&BA|u+pKgA4ehN(=F^5K?)H-IgqsXe<FmvmkrUl(A26RyXB9O`ub7nD%
z6+QKx2|^NG;6WKh1f>+*@{F6~9gg!6OPV0WZ3nBR%XG9T16RTbwn#EyyORBG(zYs-
zUzd-o?JkKr8d!iNy+XK-K5PLN8<R9-<Uu1{YSrB&FQ5x}J(D%0pNNo>V8~iqWoaD?
z_3&1f3W>Zkm}tlCWa|?Ivn)99gjaWOaV#_PFkm+f*bf6{p1=@%_520i;0SrMru6BW
zx~83)`K~|PdAE%+x0QW&xO)zDTw9e0P)Z<kUTu-i{eY8)^daY!8+xwT**$t^*!20Z
z+m9G>LK+gv0!;G)Vcw);^>VDnhbqF=90m{3<QSbC+asUJYT21WmJ(q|x?IhLE#1>T
za^mnPEO>Qui`&a9lNonTK#fc*?Cj1neRdLZ9+1IWTgsmAlmS#hDT~(7Y7tbATu?Wg
z>5nSQ(w(HV_nv-yHLFer6(H$+CH_uYoi)w-<L|e5!m8jG7F6P9q>&rm3HqE{T9iMY
zT439hOfz~2glcnyr!8xZ8+ZB%7!^0mE#6E!e1G=_cgq1F#yBMG_X9lbX*Z4-r~>7<
z8UcXS{yMoFT9sw@8@a{h^QXR~^GZtn{DyWsmsx*r_p9#4ydSMq=8AlCaXblj`vEy?
z6joUn$9aZXw9Q87A3JwoeXuP1n(Sk4MwS=))@My4ZixuXtR1ZS&q-VNT$XQeb8-2A
zEUOc%vYwwW?TIpS9x;k`#UvW+y5z%U`2_JV^ykoDr+yV_Z-%<RWqrPEzW2x>mst@~
z2bNHCV6}(T<^3~dIX@)Jw3TI@O9Y8njcU?JiktZkZ$CkneIQN^OzLIXoJ;O!@&fry
zmf_B5@(-JH;r;jPLq)Q!;+u)!{Mn5I@?g(HtvJq$mR55)KDm~UlH>m4_I$fYCn8G;
z&PWx=G}u}%tzMJfP9Il7h3eF=4+mU<fQt(|QFENLZW1yRgtI-9gjNAFc|EEDlm+>M
zfFolm3u=G1@08;;TCa7VZqJoXT=$|cea+}6Mp+-&_0a1n_<Fp4nSHm~5-DqxnU?f0
zecPuyOYigzX0kcZuPr4YWw$%ToP01j*0*UtL@o60#}oi>?{1K3z^lUz=6QjL;9|GO
z^Q-3=j+zybQ%2$hG3T<Ln9mJp6SU7~e(ge?P1nN6{YPtpMgA<=#gI?<C(r9c_wXb1
z<bJ*le)@z1EBIoJv;)_Qm#=RCDOS5^4G#ukZKdtIJ^+qYC*?5-eE;LLASYX7kI$8f
zfsq84Lj@P$%^gFkU<U%x9z}q+H`ibmP#(IDR*KiKD7E1Gx7Yap{>vZmkKg<X&%gc}
z%W}XpEuaFO;C;)P7#I614EX{t=1EU_$ygwGyPq+a39qLEW-chScJM20hAr*e@6Wge
znns`sKkdzt2G+C$D^5<N(PldFYuU{NLM9C#-laVfjTkXvM;SZP4sKEf#hgi-ZhR!*
z3@pon1CIFq^=r(0r<W@LmWlwwKs>)4C0vc8DlHHQbk}#cACO5<J)gf;Om<)W6va~~
z31B3Ylq_P3@R#Kl^DS`k-G9f$JmQz*9+#I1<Ap%R0Z5nln-_nN!>f$lpZ^6#G+jip
z)if|oBBf-T-d@;UQzA$;Fg1LvSS7(6k|M}?O<mAiq>HgeUTF;-M;(1+bzSgiEsh>F
zd`-QRr&BOCK^@R<-`=6tf?`F^EtF*|$0oxCz=45<fnAc83iyy5vjBNW&Ec~r!gCVP
zQgIL<Rp0^%s8^^AO!F;DNdQI=7=3-SMJkFAj-2s*sd&Q`&xml5Qwy&ro<Ne3vLIEh
z`*OA4;Ue!bA7=;)uJ$ibNbouxa6F231rqj{vkyWbsIq7y0pzBX60c4VB2Ujvo6mY0
zT}t|FE~4i-3XnM=OTu8!4<{?Pn{0qd6@AJAq5{x@yTdICO-NPe4kB%dzPsFktOYhm
zwd8SBIUbWb;R{zxi$1^EJIS?OY@l$0fN;6ngDBz6;WcWVu}4C_uDAj)>?-K0;_F}h
z3iX2UTlyaL<`usE^B>G%A=qb4+XSx#VcU4n>KxITQ|Fl?)tsDKm#d0)K-I2y8m-}7
zN!Pn#(J5{~5I#zd0iQqgW<I5Cg@gf87}qyPmF0!n{)jt{nk<7ftsfsI%VM%TWU`^r
zPFYtFU|w*?6|xAj0ExeX2;q3V#ZnRkBS@@}yB&9yJ1mrNM8?bdm@Kmf8xhR8wcv6;
zVwXoO#|gr~)%X<(30~3NC(AMblv;p9ko>SL118HUBO}9CN5X(?^)#8av4a=3A1GnS
zy8fXMI4|0*3%Xgv*dZlB8a`c?ha?yiV<{T#vJ{{>0oxh~t!J18IRO{DJrZTSIlh9_
z8UGyv=(>EcEaz<H`)!w?Ik5z76V4rU4BZSV!uMr40f_{X-59PWyKCJgErKs*AghD!
ze$+;{Ky`>kz5uX!`iUTfx(MDbM{p^q1#NO9Ym%Vl{DF^^V>>+ZV66$*y8&x7J1@Wv
zBalXbcaX}MXMlyE4M}W0HjZG-GvojHhyNY_mtX%H|MM?ifJmG1aG6oc0;;5qRf+(m
zU=Sd)ATz*RAVe6RT_W!XTzoy^?dw;#d-FyU^N7I2ALO|3H91L!kk!YuF=~N(<~Y@Y
z>bkZu7>QKwD>PzML)6?W<1Rk^i>#oOC;3(Yi5QX!L?$q9aX7vOs-^bo%7<Y{U<~VS
zfTcl?L)OlrS+rwnt(YfiO+1N!oNZ%R+)hczS&Of~9_}y|M*Sn<@^*)-w8LSz!*M*~
z*{{FGubx-@;`KLJ4sY@O-5dSxL3L+>?G6Zh!h;vScZZbvw^~xQzACCERAEg6O=}9U
zH-l1_+d2<>O~rmzecv7RL2nYykG(hesSysW;L9PpTTqq)h?e9=#k~e0?Y;J5wH=8;
zJkQ*Wjc$!_C3o`@X;+RSI^QNU0(Ef`RA$xf{1cmvl(oR@v@CdedxOJQFEmX?<Cj*Y
z6}Dg`0)_#U2~vvMkmCR$2GgvCChLsaUE71R@&h|Rwn{*Rk&@O6sTQu%yWI$`5N>O`
zco_X!r>6U#Se*ju-g~ORK)?HLKA^}^JNg0I<w&$Hf!`O^l@o@nyRvMxH8hgWhnc$j
z<hNzLKkj@ulo_uNH~5;L;WF(|sNzU7Zsr4Kn(>xzaktE%G=O+}#;wS+9om%<Q%QVC
zvEB_ar>RD#Wt@g1uZ!b@H*WLNLsl%o&!8=?(r-{b5p|g@+`kquS5-3IlW<Rt?LV~2
zvEC+Hod%uIx&rOBZTxxidO4N#8P<paP4}x+%+7sa%x!12BH>sIu8&8I!wwgDfIB6L
z)|Sy^&I?Kjl9Q$Q0=v9}P{qUrlN2iw`{}VZ8ll*Q(A%>sX3+KF_1ow^n+M=B*&Wlx
z)}kv~aGy<5o{j`!EOkM}VhTCKJHaJNX4<H-oKnIt3|b3^t4fuS)f`jDP&ak-7pQXt
zC6rn)l^Ji3x40+?S0GG0;jY}_?Q+2Na>QL(FxPT#JUtq7wMM8#ZS+blTPlakF>mDh
zVVdmwAU<f*KJJZL<@kxn?L*k^eW}(F%Ic>)D9armBg^`T=vtWw7%akpt7rrV%vcre
zw(Ct*I@?8`V-YiT5{~5&SvCXKWjQ5iDqNFg!I(zyk08tbyO!?OJxgNiE}5LQ3wF7M
zJ>8_b#!50L>1WEa4XVhRBGxvQI;BGIp>B#15^9s>JM0Uv2Yw(~ULUu$GqApTPnP|m
z@Mp~*w0nAsw6}|&z<L1mZ+>`97CgA7oFfdA<D${f?yWK%=dp+TgXGwTw!LJI!DI(#
z)=koZ2n(Y{Q2H(EzuNC<<%*kW!l5iEfK?c6H?w`Y*ND{ms;kbbB|m@)#<46|tWf<%
zh5rbebD2zLyJym+7eEUXT4T^uj(e!f<Y-rKtsK`)ayJ}I#!a81%&ScSpIeeTY6Z=K
zH4KAVC&KL>b5%cAqV@UhS19R@S%@)7!OeUC3S$J}$P@1Ph}$yZwk#;Zmip~B(|YRI
z9JL?ERmXL4REMai2yWMfd(l)IqwZu3PiYgSU)kxo^hbKaYjp6zgaa%1VsM#IYsGS0
zRKWsl=i(vPz`BA|=o~fLPHsyn09Fsbtm%RPPzj)vFb;clnpMF#4oH--*rJn=v~Eu;
z6c1ei2w+>ODuB21g!$DS{{EY9P=*nl6L?SsF|pRz5+sOZ)ZKuZ6Yi!1Dlsm9^#Uvl
zrkl5zc)_x$5%?*qtasoU#*F=bk7Zd<N=3B@EC8M1OYJiv)C%&pP-KfjVi1b$(!g5q
z`f;*V2T0JC+bI+Ndh<P)X6$x5P!il758zrb0NNH;1jCq-$Bc{1J<2@ecsycZK^4Nn
zy11o;woq#>zvube&WsQiDYz>W{`lrK$_VUs7x+_ogO}5c|MAy9;>~o8KODZrgbEBh
zP-UR<>IX9eA^FDdW^J_8wys-^Kn-C}WAl!N!md6`Tb271Fw>6PbNgXSWw*38h5l(E
zi`FcemKkd<0?^q#56iJwbd_}lRb@(l-40>4bBnxyDo{hpV9H=h*yRC)gf~;cTo|RO
zBdu4c{5&%Jyad-bclf72e2ahl^>6X~m%p@n8wI;OqDaERf;<o|_B+&+z==@!4pJG<
zh8=3X!*_4~gqu=uTePE33km%Qh!Tc8B6)*}q6UX}$DsBAt1>L+VD?B!pZ_$ihr!U2
zu+52+Dv(5L50OSh5?Q1I1R@pJ$6H7)*zZQnix$Eq77QRWc?Cnx7{(D-S63+WjOp$U
zr4p(nEL<U?9qiosr$)SWr?R<(pwSa^nef+}SGbz@SkfiFz5NdLuY~{CxBm^_-M++s
zee*vsgCTiDOHuL720I^5Ga~E3sPn$1PgqO%lv8uav>uF|6Tuyp%>j1bx@r&b<G1y`
zvb;UO6ES-6<H)j`P+~P%#-wu0lyNy|-uU%l!Cdq?<US>@J6D(G3c=gkrz6X!_TH&1
zQ&I`i$Rpn%v*+GD2l{1gef%t0US`aP1xo=c6PBWD);_0o)4+`RoU?Y}wY05eVH{+}
zcQ>zbIq&d1?aot{;bW~YJ|?k`{pQK7$yGOGmuW$B%1iS{t<DR0tSmlUj^Q6<SFAM3
zj{tB*DFsPJ43u#<Pne4rWwk$DN;t4xlM4%Og|P==gg)CTk#_d8vq(uBwv0SrU`ELV
zCSa0+=`iE_?_XhJfp?c>@W*7C*}17xO`rFy(P*;Pmu*GY#nxo>YqCZ3Yko)buWDqk
zMP8*f(?Ame)DewTZPNgl>jEmQh2Gr>o!m~+ZybgJdC0i9x<s94Oosys>jvIr=ei%=
zk)+>81Wf_VVhtaX+C)<EW<G!mgR0;!$L}%R?C?+D{R{qj`wIW|`g<I$D>(exyNxwZ
z8SdO|cE_|-33qW-0s*@mQ)=V1KAYC%kZUJ(F6aHv@4vC`fPUWwR`Am!99Y2@qm~M(
z+$MO<+S2Lk#CAaM$#(~pPn9QcTtDl^9{{hz)S{$Fpo>N_&Ds<qGyL>$nwAGwh)A73
zm4HIPQ5ZL6(XM)wz{FY*l>jEv`d(hJ)x0w$F_v1fusU}V(YB_*N8k%VTb2Ap_#(S1
zdH$jXY?uNe-gywP{w21=psTfIW7wS_c_$<i9Lo`)MHexynCEIvG)55(kIuHOtLbG0
zOJVEQX!o4_Tmd3(JJ{`lO!=XlAs|d#aa|7B$s63%f|Pdn)A1#4<^yiZgaSbsbQj*o
zUfXq_riA#;u6BU|s5#yIVOpax+{xH4ny!dxSC@whN3?siv08dgKSK|{e8$vTEqYPb
z=t((=6v9@H8BLFEa%_DuH*&lgt#&I2DP=H`)ort1w@5~F=sXZf<FgAv6~vc!H@Gdo
zM6qxX4`a8DP-Fl@(jay*XKNL~>*Ecs>j8!JS#LFyKG6|fYQ>N;1}pmJwQj@)q*w6B
z=Ob6iL@e4-YfO2WXMs;^4?T0#c5^8cZl_yFDta&Neparnc=uZPSe4YuULadgMHS}t
znOz5=)tl<3kdzZh5@u#36d+Cbx9jh4Qx5p|+wbvuy1|i)rurn`LGG&;_Jilww;E}E
zz)&J;5gcnY=c;sXRa#3|ko#Y8?^$}^jQxb<x93NZ<;{UeyFh|8awDxPz0Tiuh)rl=
zf5s-u<-BFti#|8m!y~fXDKI*+Y4eo&%Czb=KY%P(#=Nk~GOH6~I{;q8vYq}$mLb3_
z*8C>P8OicPCTst6^<Vq}JRzAb#qPJ4aS+$4P0zCTN&D&J<hUKlZbc%(FzCJT1_mr*
z&z$SdqRyL>6X@6E<1*vVuV3Sr;}yOhF0|HUD_#x&RiifDky?}40)n{|OtoMt1!WEC
z`Q$OY_eoiw%l;C8RgOKf2%@sko->P-_OwA@6|Z%JN*%cB8xkCr14tNkP+8X6oP9$L
zX|hbv@pj$1HIWeeZ8u#|FM6*3YL_F_S};q(kuv^x{TCc$#vk5%hd0L?9O{Avf=ap7
zA9dZNjhAnJlONmd3%y4IXqvUS<8(U@(B`!dS!3TNl-rt?>zw)S!v;Sc!hschG0IdR
zmj7IB*Fae(u5}aCCrPmd%M>lzPJupQLTmvxC1?%tRcC{yv)Ba8ge02NuDdmBrz(Og
zQ99jf)CZyFgqvFM{o#P;0LGk}4`Uz!WgtmLN(l?B8x9c;vfxe@9BV;^V8~jCwtnmu
zkqt>x1_<!tmR>`wbvp9vuv9PZlyosgwsO%iK)>5%ileOKs~y5gcW|iUuG|1rkVatP
zisN!bEsR=$*_5K4Ft`@X$0O#sU|Om-iUJd?P`erJK-+cmGw@yKJPrUS9F_&kbd9%s
z3n^O9>aTUij0!=4fWW1IC?lr~BJF4;Ckl-#!|q5VO13Mc=hU2iuDqMCvki<tWJJ(y
zw3gk!8Nbti{nS>_W$|>QPER9jnlbcp?1~P<r4ja<6FKg_V>SD{2=kzA-q|Hn(2{GY
z`}Bk9w`vw1mkIwiAMtm;`7I_(5lpNdw^Ai!Sp6a*l*-^*Q6ZSY_{*C=<4*LM>t**-
z#F&L~SY}-29Yz|Ovw&GAucqKi(+jcgs+)nrYNb~cWM~BFLpI_%o;u&<-2h4%FK=Ff
zF=NV;Mz7Rr&gR<Esl=kJN?EWh3yyPXa_lm*I(qG&9$9ITiy-a#DomIs#zHf0Q^D)M
z{5PZ`c$Ebc7_&VcR2j(Gx;^OyH6yAOd0o>5{q<@Vn9pQ7r=y$qflY3gI~ByjGqD{z
zY5)BYIexnG{bR^-FC+wYM!PYlQAj4y{`~YsQ~IG5KOe^7+-14yV6NS2ACqO6<Hp9p
ziqfwQ8uY!Mj_Zz^&z9xMJFa!Ij07)Wy_IEX2QVTG;{Y&WS{6(wxJ}FX$nrfo>~zTf
z0qTDD(EFbv^ISCQWep!!9r4jZ<oIAY_CM7AXeZHe96-pL#?QV<Y5HbMeU*2v$=A2H
z_+S6+5BSI5{vEEqxq?)UzRx*n_h-_&XM4&R@_?Lm6TGtEcA9YH1xNSX_yv3-(mz_>
z*_121cB&uT<+x3l?fGR(&*fEi?fUOMnWWAu0zd;GRPpBSH7GN1rFFn+tyoIYXlJ`a
z85vsdRyU4Sx0bdpuQ$ogm3)cTpIJ2ZGv@?hLMaOl1PBrS*YExr^e4ez>x83J9BD!G
zVrMPZPi^yG6=vb}>DapEsg+9cZpv=E5Orfsuy<!0v-){AG5A>%4y@pdF)eGFj|J%<
z^(!WwD8JIZCYIgDoB4U8nz)yrM5CbwB#01l9sx?$^;SFm39pOs!w6vYHmYb8)t}zH
zLgI@5<yU`$tNkU42v9r^U3WLR0#yieS#Vg6c=PHO6BtVZkP~*h5j+>vrGC;p&7{gg
zDZ1Or%m8T%<CL^JTe9dQzrKWZu*f`sy69Dn3Qy|rsbX!E&A@1hh&ISVf~jf*7V4sr
zWww*c6#yk9g4M^VDB|e=t4`nxm$pdPV~c1PI9JjWWsueCLIP?MR4|IQEEjP_+iXut
z$iq$xidNf&Y&DVA-5oZMsqPxOBd=<K(xe5zwcE=58{<Z~=f)5`t>HuXX*si&qQ(B}
ziQd?*>>YC4b_&^ceS10XKIDe<Fl1y((89a~A_hbYsd~f#OVv<z??{Hzzq}AIFbU&&
znbi?(1{p7GN~X!N?|L#&iBYKHzzb$nq+Lc{7z0+PY#V?N#EsZe8Ou^Yyw>)BMOJw8
z04LcFb8U`xSEk7n=nC&j+U%F+TcUOYs#Fxrs8t{+5UQx9Vp$3n)+kDs0A@xhf@zwt
zEM=8p1|*+exGJm6Vuh=don0q`BwC*<Gk8bfkpLETq)XQMBMt4lTUXK6JvOzDORfyN
zLzBGlhkq4W^oRGe=#=>OVGtfxm>=_G{Iuj4_`<S`b=}q|iiAuViL?%rt}h8dwx}WP
zK*lc1Cqt}XZN>OSF!A}yva>p#?9fE_WVwBzS;i{M*?riX%{Be6I$b|imgl83qgO=j
z^Zt@OL1+^MmsF1+TI-Nf5=fRalI43b`uH?ietNR)X_sEp3MN_Z4Ap;q=9+97gu|FI
zq>PkDb*6y8RV<=UpUdXB>I;)TuqGdC!JlvM@Ou_u+#{8-MQ*7RRh=e@)sb14QCOev
z3I-Bzx!<93#Zv1hIicHSik9dA3vwE?nNZdBbn84dIc_U)Up~T}Rwm9H1zL}jpdOs$
z_b3U3Di)d4D4J)?i#nS1dZoz$vuzg7vz1QR%CVgs`~&V<@^eU}x<$LW38<NI7-p3*
zCdiPG1V}}T<-^j4nOR$y`pH<fN$aL+U$1iuTerw9Epimnott+)eMlzz*G}Klzb^>F
zffal~mbxxxKJD}iCU7!6xr4+0?IF+2Bs%$AuPl$q$Vd=QNW*|Mj99p6fk|4GGd<4z
z$+9VNV5^f=@t5m2m<|X0<JZ5#I9x*NWSjTuyb`Fxrm{P*CVYQ$gE<5B#SX}Xkp|35
zy??PEFb}Jap-MtoG{jh#&9FBI7Hv%1tuf-SdNF(vb95!U2KydQspwCQbns_6LZl#9
z4RbC<(-2tJ^dV*jlRA^y?y4(E?Y6FGC87%0YPg#t%a`7yrF#p^;psK0+?kXJDG!zo
zSaes2)dND%t}Cq&E~ygjX(h#qU=G$zEhP_$rtJvOYL3Ik@aM{68dJY(oBSWf=Fz70
z1tGM~ZmoL4=}=lkp++}a=O5Agn26xrekvDrKuT#!qwCH7{nAMQ1DF6@F9olTM_lfA
z*yoH=C$pE;J0n$7s+h%M{1}rgn2A9-BQY=#W9hZ!<~91TF=PR_GRj;qq(OTQnZ<fv
zS=P>vB%-n*C*6Las<UqfY*oBf$tDmO6-5?QUO=eG7M-#z1qJHP@$an6CbyFs$#P$g
z+n~dK*ExyX_qw<cG{Ph?I1l>!S&fsF-01ChcI|!=+5IPgq@~H|J=K&Ivie(U_-Nfb
zjqL3238C$V)Arf;x$-4t*`F7WRL+@fe$VY`gl$$Zi#o5kYSc&{RcR;T&KNbo_`GG=
zj^VYVT9o&RET<M3O>In9W6sL~$R-#~$1Z)AEc1yhZ<aGNx|@zf)Ew&6<QN*NE%nS~
zxxbK4^kzIk`a<`vzT9i5(7pdt*JRk5oKps+0k#ha7Eejl6xLJueQZshE93iR!BGGj
zcEGSxCk__<TuRNCt&Nwo97mCg3a#~cF^-rEp>W^G<0BE@;3Ulw8Aqf%z$-FsYOmS&
zQ0qyUtPzr4o%kpw>JM+SZ)Htv?bc0Nb6}NO=d5LE5m<$_4TFyuvlb0d2NrMS*uTTy
z?TO(1n0R@OMkXx}kSj<`2gw7nWDs}Gs&*aJoR(<EQab{6$35kdl6H0U-|0N!*%o@j
zuDi5U;m2<fK1H72eazseMmVs7FNW_4^&!PV^2pS*PKdq;dsmDk041wEg)X|s4xIpu
zd9aY<5kgrbsW_o<#cXvT%EGpA_bu!FY?7H~sjcX@%$S!MU;X+&;ctHXclh?({~s0{
zAbEysb+uMZwc_=Bhnx8pg(`{ww|5i9Jm4aa7;}P*2}}97Vv}CrbmAF27wmU??00*V
zDLr@`HFov%Gc@u{+Xd@&=A@zE8pgjyfV9m|TX57cEm)Y4ce+zL&$FdwtcHpV8I%as
zoIr(bORshYtkv9DN`WL`u$|cA^-zeBvnuQ?6+#3WGq@D6RE(w=kG7*L<6d_FSwJjp
zoS4PZA}q~v)uG$wQ>$G-f@PWD>6LEAv87G0DP*ejZBx8XhrQDaf5NOgj=sysB#&G>
z_a6hWHf6vAa_l>KB&ytvdBm7U<UE3*&toZ?CR43O#9H&=v`T}Xqbk2uot6ZAe|L-h
z%kS|w|LH&D;_?d9@>hT=D4QY51$T47O{sW2Ut_8>3Iv=9X}`mmDv~6i6iBVj2>ejQ
zhDB_$%4tAOW9#1J(F&{LNY>(<kXvM-DVL&pax$5Q=nG^~C}hTOf;Kb<F*sGEU4k&<
zI3H0g1=uqKJQca3q6#2IQ_Q$CV%VJ4b1fJWVNdD=XHjP~QAM^+4YgDN)(K7+r7Y(7
zACN>a0ze+EODVTG+!bX{k=EzVYXs@l!5w6xslj!b^|^9)za%w^q4S9jXKHdxKWjM#
zzN9SsWSc}gd+x^_#$f=_V6`I!W!9*NQq39NM^Ub3+vZ%EHG+uFSeEOek-&v35+$vV
zW!KQlGTZ9xhMYA`#F~ub&!QDf7O|K$`Yc(teH*>2>W9L@j7-4J9E7c=5tGd?WE*va
zG-TAWfX_^pf3EiG`lPIXuZe7@eA=4qHRX209z#lC&aLLvTm;Lao%Cv#3aekS$pb*Y
zCRY;B2-MvUa`7CQXVpO`frj;Kmcg`C+)M|&o^CLkBYI8+B)|?CiA2wd%sI=a^-te3
zegc%jxEOZ0-0zTR2cp48MpH9MkXxocHOHy1$!se#bH!Q<SKs47g(3_v+a{DNJJ8x+
zBSW~L6pgNOr+4yDO!4dnP1USry*HqBgTLC`nBg_#7{q!mr7#jzjcO-N=dOt%iBS_E
zOc+EXDTh3&9BP_|-j_CD<{_mn1;2OpVNEHlJfp>e(i*7^VI*^6vDfhUX!_5bdR>E`
z8{xnTz94Hgjeeg*?)_u6-F#=Tsj>E+WD~bf;2Xz8pN4~Mic+%OBC<)j+x?@HuQ#Ph
zS`>~(`7D)D1<3m=TwHyP@!}GTFzBFO942+L&9z{v3nng*tjApnYh8}4hI!6})~w}2
z5p^CBG;-)-cY&M+)V2fZV{$U7iyOVLx?h$aNa|9ovfht&6!Kjr{j5lZNJ7pUz2)_Y
zn)Av&-HJ)uajgkUTV(yzDqF;ebsu3dqu8R9wMez53|fz6FxYnI)nLyFrqoiaIpSX+
ze%|Q5yj>xofBtk_UE#9t6kBM0XGo+eEbTB4+&fwi`5^CFV*4Gw^M>sIYr0yKKVWn5
z56Q8QGU2`)`{&(7lT$`Y83avpO+*k>bD-$`apzV0(-$r?$4)Z`3AkHkygu9k!w&n)
z=g7BjtnP{`y@e~5S}-jI(=ua{MLTL`txB9U2gVAUegd7m=otB2wWdw!Tqd2;xc=Ac
z()l}yysdyoP52k+bL*+Vx+^c*<p=};+m5+M9ikb705xNt&64p>p1Naib24k1wO9Yu
zIo4XM`?ZKp1itfI!A+*J&qI`~Bdw<NwKnF`%pJ6+$HEKreiGVFu4ohw$acD$?5^{o
z0O&WMA9GZ_PmX^&d^uSbplzTe5OUHQg`VoFM(kp1F<U25bV6k{S5EFy>r!~mvh2m_
zeQj-PDK`Hl@qsm2C7t)`$&=@`qFgRMTb6A^x3a9AMAtAYFZCc0>zvWLBAWmFoMib?
zVg3xB%9^}hHf?_X!PaE^T!MB4)_bhcbS*`fO<#gwQ};@=1VB5HeoZETGf)6b3_Kfl
z$OPnJfi~`2R9~+ZGpk-*NOf&#>ECNKlRjA={}8Ovh&oSslOK1Aim1HlI!d;#c5)1R
zPTOFwvuGd8@0l9pUkWsxR?{u5tc7rsW4%B6UlF_Nx<+03n%wTW>C-+E<gc+dv_KM2
zVvQ8F3lI|PSPZ*1Xe1;Pyd(7eax1P&_qFZ1U$t|oHV$t8nJl+@rmNt70TB*<MuY<^
z__B0!;0I4sHaUBp3ahW(f@K6~zA^7Uw8Mam0aX%e$tX3WR!uY0-QIP9*#;f^L@E{*
z><ZheXuY~S;``%*zyCk|FPM&Z_~(E6-*A0%jl0(`@%`NauWs%z2?Ik$8i6E&G_!7E
z?Ccx+xZn>&M0LV($k^`&q%;CDAZJw)Ef!@HPe!7F&f1ojkdj_Edm2(_{8REY0b8n)
zex8JoG9%}TF{wwcSVV@8m;m6aVaU}KALeaAcv-(n9S}t8xUy&+1u#-ek+KvM$+pQu
zm@5NPkVw5+Q&r_jiarnttCHKot4(32oItE;rJ9u3?e@0?7v{8LRml6XBsCP6)SBPk
zI@&D0Pebkfkuh1VKyzx|lVg~oEFLwq17QFo1NKL}!<MC}f`*OJYv!2!yq#qjz{}$u
zZs!GG|IOdy|L_n01^Hk1xW0ai+kgKv=HrCBrQq#!#H-^Gw=#p1;Q2RSqk!@9@CID8
zPETz+j~@>Mpp>xN?Ep!rRcrN$kjenI(}OC})P#-5rJYuL(2Q4xdX3%Dr=!9k5hN<m
zI%1l7=`}RkcVLxkwuz*Q-;ZuIeNML5yc?AWP^<lht2x)!&a72f8#$P~^ai$=BW@ER
zfMoX@VAg0#y{mdAT3?6NApwiT6agoFKZG`Okbc(ATD%STmfo4O{an!aKv@QSy(VND
zkuU%l(TpFlRBkKG49e{bnxj@UQb(f-*LC}>WqH^k2~cX;7=F#E)nwU}7IcG0$!&Z)
zFG0duM=HtZ$#P3aJXxfwFp>}mV<%rG%fd)hEY&%Il9viN7g+}GrwjG9!+!emd(Ub0
zA$*7&n^Sdw)v&_|LWXeCg6+_hF}4U?eM)!RFL`WDP5?&WU%&ekmbZfc$N%mBiogBl
zm-zF)ev6yy8@##w6N(f}Trdk`0l=9dNw6C&MXl50m~|fc)O)R6IxCVjWEh4KBEVA1
zsu#C6pU832O`4qi*IiniMTC)vHbj6voe2V9)}r^M_KB9}&`3&70Cze%XEA5gI-DEp
zMOGOXB25bfTbieB__08!Af=2kWe5bE)-hvCnGi|q`M1B!VP(;otISy~G_0GxIy4%=
zMFiwzUC)`&qPAT9cCRU{Gw=g<@Lf9fPhGZy4<j5{!Iy(hm)MhO$Je+=W-2w5R>O(b
zxpQm$!kjkHT_Pr*bkQkI-5HQufPNCP8j8NZ;BOJZ%j-Az?$t|t{mpNXc6(es{|ZHb
z{WKxp%m6J&7&NSTm%x=l(`1L;7JB@wEeD_cJ!IE*u=P}Bb9mdofhTNk6~S(2*T>V8
zP>|NP(v50?<`Q+qGMnRvEJAKoqECE7_sd3gYoqV<lI;Y!p`t}RS4WZvTX!?P>25ST
z)E#@f2(G))dYm;yqjz43tpAYicg^rPJ^2Tq2j22iT41+HzYkp5d*s+ZC3IhoLCu}I
z)<cujlsTbQj>XTVZ7JJ99}z6SpOqQ26dX&%R0X@MXSiU&#j`ss4CLbknP%&HN}vP`
zyFraFYgg`FyFE(r(-mg>tUz}tiL`EykB9yJP^2bp{+7*a?LBsmZEQ)9NTVCXnmBBu
zSY>l({oFQ6a(4n-nV@D!dZZ{ppwU-8fB1QFi>`6S*XQ@s!R#t!*j%xuS~fZE<k`>J
zJF1<`0w=G%LyqsAM}M53247T`?Qc&9)+E-X&m5_vrghTw=jb#%#Qxr{+Gj4yKx<Pk
z1=bZAO$JY8xtBQV4+fRjK03NT&JQ8W{djf(%xELdx-piZ>oVZo9eb(&kCtWHeCFh;
z-g{4f60&^nBIN$>KT(cXQC=JDs8c!Fh6y?sQvZe2J$R2Ry8_^DI^tz{gNX&mBQBnQ
z1xCiR=XVgg!(m}<3Xn2}Awe=w6RX1nnk_*|HZ|iv7U}=kGe_PzpdS>ALhNPb{;*1W
zZr7pg=UUj<l1WXqd9dpo3bcRw7qm!9*{)C9HMzxj*)P|Mz8Vq5)(C@Jr}Fh1y9?_k
zs!nrWWwIYP*^b?6X{^UxvmvB?fHp_zF7yE&`;ENwQ^C)Pa9{;rm{T+00oCFiGm@PC
zTKWYR7?ChgMouGu3@!;wgh~L{WKI(D(h2MLH_gxv+p>8quMHZ`E(!R5|F{1S{{7o;
zF@F7j$1h*Jz~BA-f5F$ccewoOHz?_ksQ3%YEraBM-B-_WSZ2Jr`3_RGR)Dtd7No37
zUM(Ltf-H<uD~42&h_$8&0IpW2$!jmdlyafAyH>@kZuqiGnvXB6yW?&^n8K(T9|_}_
zFlNGXxI-zO11q5!A)3mlwWvs>MaluIJ<^;-=<dAzqgGYMNh%WQ!kQ8=B$$HAsC9u9
zfzTdI8O4fUtAW%lke%u-$rULb7~d7D>rK{$9D=o8me)`)6~in4698f>bi7@ZS2N>N
zO`xa3i6Zu{x4lEN*yz(fwDx080|pwAv4b!$O$-?lWMGtvmVVnScWau0Lnr?>qsScE
z|MYKv04m|X{O7;LH(!5)@#>d&_39O{`x9RCx1eLiI8<O1TwLrR^Ni`RfHIJGN$W(A
zcHsP!01mdulYs;|i55Q=!n`b&7NF<sQ<Dy-7PI#OOPz(G&LU~Wj+51x(OP#v%8W4s
zV-g(aMLU+2UQ%pRw+)+h2kLJ(ZDrYw&vpP%lVhQ^@IH}X7&1r%84Sn-2!Uk}5kY|f
zk}$wL30^Em%se9RW~v5DH7A+3>eh56oBulFNxWMqb(PV&IYQsQO(~Z4;BKGsbnlg)
z0$)&;?N$z=9f1cKfkK$Lnk;A3N+^Z8=c_k-X+7X$r}ZUd`2&Y!bNwf>oKv#R6^p{i
z-Q3j6GQ`&8HO;r(6A>*r;2#s!TT)6th%9gB&L+$K`&D0(BEV3Q%(;=0V4wjcd?vE|
zaJ22|D?cwIw_i=W#px5|*f%?K8Z?TMMte306@&tyFrn1dZLxlz*S>nlLE#S_@aAxh
zLz(cW*RS#7#n<@l@Bab6ER1K*ev9wE{~q%n{(y@L><e)5^$s)wZ|+`U;i^r9EVF=w
zkSs#!qeR#rCrS{~nsh`Aq`^c{Bc8qX-gZsSy4edfJ=M35*C|Zb<~E>!%0?#SEXcWH
zo+@gwV)!D>F<1fkCOcWgSKnyKQwTRh>hq2Xxl~WtEFel4a@L@+R5T^KbuDK=v^c*;
zQmx%b{W?h8T`BOiUlyImSSqSJ#4I|`&s9%njW$)Agw)Arla*EKeB=iG9CYj8a|j1k
z@Wogcw};e<cT7wV*b{xux6h;MnDVyOI;;B*BBQntd`(-}m@wEb_|(3%k7V*<q&Ld}
z<KZn{zIg+wj5<!3jtiE#Vn{n&UVMe;e8f(UNc#+-BX+}p3Sgc~J90CmK6GAgr$=Z7
zcVX!T09dLiyH*g9^!j|a8vvL@jow~~g|r0+3?Q3(i@MYR3mLApYO0_wXh~X44ZjxZ
z^vR9B)DLlc%le`l4z;ek;%lcwHH7tAbg>j6q@*2?5&<loL&+44%}y?;Da>0{OS;dU
zIzdjKRl(W<*soDK4Pt%YLq~WQPo-l@Gxq&_H#hhr<+%G%r1@TvtoGGvyqJ+w*wSKF
zLuS1jr}qGxpL<1`MewKZ|AG+-&tF{NVi-~91uTSdcZG|K7kGYIP{ti(U=Xe6qTij7
zl0Ep9KTWqHS3djw$}YXvqZ(my08_A=<CRP)?shvxEE)pZxr+!X!y+j49EAl^t9CF=
zX4u{vE|mu|nQFG=$?&SE5jvH2Rj`RQ?JMOb%LHH`(r7vxJGJlgI)=XMY!-FzURqDV
z6!eW0bN8;zJKL*%e(U-7(VeF^@TadZe{oqRm*oNOieXC^7g+i?I*YJzx*m65>voea
z=3o*0`Q=}+qX92oTwwnhvP^wm{Ks%2%hKxEtt$*fYg=t(xl5Nx-TT|K&pL>d%y~74
z$#ZV9%(X&%4mur1;9*(r#&aXf{--|;Uf1O0g{eiOwg)1uokIfD+GJV$25q7{wz9l_
zr=>PMD9hUm@^SP?{PkdL(Q5Xt_Dhrf2h_}W-QuTG`^xdTu#K+iY;f6n>|);|$F3Fu
zTJSrkY&mVT)uFxKnk#ND5SxSh&#QtcR=<>4P^9A3^)<eG`3eK}$dXVPAll*L;wxNT
z9Pn%|*zZSR6wES#mI=4lx7LX|BUMc~-Co2;jby74EAUJJG+nY)Z4xxemA2{Y8iqZi
zcFls-cF=&X&Jmq#<CV2*cWX+s9%*wKt-0g<1@YdRywy=*PAQF`l$4s2i$pUr+`%SN
zAgmcIAm7y0q8qt6m}LE~_GgXe((l`zz4-Ik&56>D8QnYLgSY3SNGY|>$iYvaa9{;r
zkoT-A{UR!z(gAEsDReLS#z!(^R*j_rA{t`lA+p-NMi)>PFW}xO$?dz%(51FC+m4Iw
zmXZ*zmm3r*`1ZT+aC?1=n;}CugX@4yd%SqI2jv|WnXz0Rk<v9TFLt<_7u+2RhA|-z
z89-?|#_9FHbBKt5lv|FuR*nL=79f!Y8rik`Ript{9S49Qn|{Il@>8qkqNil3nA+VX
zCsjnb3YMjyGHVe}b0$ez6~=zIxL@Snbv+w}89*r&M2sP4O)2DpY*AK{H8rD5dgeJ>
zoEP~s;X6;<DG_#T(xS5qmu{zdbzb$)Lth9~;Jd-w1A4(yX%}FhZN}LBq=ugi4;uUa
zyY1UHa{R!6Y@+k+P$>^cc|dE~3@sA7)MCv82p$td=#{?J&3OOZR(BVJ@z3A>@A&@Z
zOI%#+@b%?0yvRE&1=#JM<HfUIV=jBt@rarWW;y~g*>Mtx22f%Usw&O(<L|Ub6Eq{v
zN;eA^FcCOuij}uG7YLXVT2f4x`n>Mf0$?>JQX=hvKtLXVA!`>gu7X;Mp0m}t(?{E$
z?{+iP2zr+($Flxem1O;aQIIGhj~QMwV<2FljFL2-Lgoq*!I*d2jG-ONQXRO$J1MFW
zv5pyci-H{uUcbtpfvxLfGx$xrf4Jg?^V4@?eNkBkkcS;|8bO|NBtR*GvMd&CvvoMS
zb9z0#dtU42{M`LhrQ%=y{SSEa`VB7jBfh%4`ea!qZwB&_oXWDG*z3Nzqz_zQ_i2^o
zlLV|TBEn@EDQV=>C>Tay9QA#*FzQlJD_cEDcN3h*@&mKt2J?5gvFb)>t0P$o5(v9-
z*9z`u0x~7kq%)Ur0f}J9`({AvthCB<3!CknZE6~qHU=)s{k?8_S<@IE+i2Q4^ZgJX
z<BnQ88a^$#fB0-=9WsCABjs2hZ9-0?W;U>Oy>7l<)4i?M;ku#Z+fk=apZ{&ma*B1%
zgW!*U{T^Hh<p5j^dt71<!4A)!{Tg3QjEM(~&k`^)uw&5O4gUE38;oPd<!%qo46@??
zA3JTkoNE0hvFqMO_i41GySL0KL+dsA9I#$FS{L9pZ&{Rd9!R5-Q6W`Oi*{CqYbieM
zwNYL#h~9C59!V);5tIHJ0~NZSY6O<moP!$v0zinZyR&v0W$XN@1?;!(=ceX1Kcz}?
zbM)Cf;-x#pbquU>xz(~&_8-4-A7LnAmE)%<%fY7+4y@pd@v$4+A2;U^KnW=gNR)LU
z5V1x$TEnSuwNO{F&O6xxXUw&=@g|j~|M_Wl+3HwO@%rW(s0F*L9m*`2?ihE=fa8*J
zoF^1nFyjU@AF<!<Ftb3WrR`)@GUE)U!++rQYB76VDypf8?#R+;2!e)gC-_cgB@OrP
z1BIG?(rr(=d9S0A0Iu2<C8y>vV~BMnB7YYW)}rRFyfzZnI$f?_y5Z~ErQ76i<r8g(
z$elzMhXqO0UQ4a^hEk=%`XO!3$I~t=Z<XV1T`s@N`n$H>Me7qF?R_iMZXorCX*=zk
z;wm2c?Mja=uup22=%MHIQ{{L&WZi#69kA9mnq_rTS-DvIxGP40bV**@GTZ`4Uz!^L
z5e{WSo-=NbN011nQn4(+aRLs91;T`~5NZV~X7IA;yum=J(mnl}BKPQZKJvOiBMDQB
z+OTKS8!g!LDK?0@3+l=w%w*7ctL@CT!6Y@zAxKH1t;$p}?z0tO#_Hg<h)#mGwf003
zX<jL8luhlDW?v<$^cOQs`x(@)v6O<GHKI;ThBcxRC^kbS6MnI_+-%CV*p5CCwD>G*
zM=F-8#b*5ky8+PX)>R(+l*{cck<Cx)OE`6=tg~Wo^z{m{Itx>i<0qvo{|K@y_99oC
z$64(5SW^rL+SZ3xk4A3uG;NQH7GW%FD)-%TL_Q|m91no-XUOu2T5`HBf3PgGISTAa
zOZ`~fm*wQ&E`ao3q6-Kz27nlxDj*eQ5sU*NBemu>;$B`#S`wg>Wj0wBUY8@kK$it=
z)?g;2pv}GgJ*c%wzP^VD8o?%5>|T<snPO`o!)00fv=NDQDrMAKz^t;|Ya!bm%U&jX
zdGBw@>6+}D9Q~Wz>9ET38mYCe$r>4xuq?}yk>zc~n!fTbx828FR-)AAHvv$qrE+53
zyz?wkAgh+=^DkwHRQPk)xq9@!0<`8MBzXDu4X6UYxOjmn9r5~5FfW3`#5f%8z_nsN
zEI?sgTn(7Z2ya;J(VS!*ec!25KV*xv)9K_`tXzg~7A2ee#n*){dXThUs~;<Y)*4n~
zHb(YM;RLeiGf7%(U1Uwg6Wg#&@Org&!Y0Se>oe@SfLFj}Sqrx7y^xdZUDxEC^Sb#@
z01|19b@Eh6vHPxNH#mFuLTQ^)Eb7o^OD$Yy6dj8-+oY3YJAPXndPQS%gXrYX>CJom
z8TWPhN$J+XXAusp;0yAeRr6F)e0*+t;G?F3X59tiA{Le_kn9EBb)sDj60eD*MpGB;
z3*S3j>nhpKKIsHsgmHa$i!2pijhC2@1vfVnj%AODM=YhHO2xzn6igV00XZ+wBWXBx
zYR81z*bfiUy?NUmI$~9B?3%1ISrs<N2wEgm1+?p5OE>Bi2y{-YE}cnRC9Cl(l9~~_
zDvauWvk`2kS+QL($?i|TgVQfrI2rj>(UkUw?aMtS(pxAeGgcX^LeU#OZ7CH&<zfqN
zzZ1Q_zd5b?+AD34Xggs3_RZqFUW=dWso~tdRb;b6)FS?$1Ybsu{jDu^6k^ZbS{PXW
zsDbEtL?o74`jCQia&O=^9tg*}fR-cfrXz?|K`sk$JOWc4z^o|@l>{{v)LP&L%DITv
zmFm43?^2i_x#IS`65#G=KrA}aT{~tz${H<{sZ;BkJ=eMD+?Ys=oEUWxEm}RQ!s@%T
z?!X7E+h~(rHE7!bT?9ytR%&N&bvn9j-&XsM0EkA&0KhPK!m}CK0GQN3NF+$2MsGD)
z@Ce7Q0Xy7a(5_dOGJP+v0!@_tlb!;imrM8Fzl!#W9HW!tci?jT^!W0!>_59K_YNA5
z&S|rM^DdlzQU99#&FlPmC^LxY2ax3l=IjrW>l0ZPUY#T4a<)2EdIwhS#>pQ`cg(M+
zpmLit1J%+7)6g}wkj?NAnjc)xsW}+z_{pE^bvS)fN#$P0#?wIkbJZKg*nQH!^xjEx
z;!87pRmkoA_EfaFqZz=}WZ8^pOTSzl(I<R>EVsu<pvN!=M61X1&>cKkliPJZ2|nUH
zKSGXuERr=C0nAB{RbO>Sr#}n60SO;RIr+-g)xb{1?lD8~_U0C}0Jp!o!$KK{*VkAI
zFfR$mvOp@MOf!fXS6>g<4Y^JK)$7-FV*IhJ`lGEj!rftfL(+aSCsCWvOa|4DtZ$;W
z31rtSy6u#`$wODD&UGGtDUo*F%oYXaQB)f_wn?@f-u^Jq99ZOkch})^%&eUgb<9#X
z2CH^YM4c1bH<!A#n75Hv_ts)`_uF1>PIMA%Z_~+5^BdX*Uhtte_vx<5!KV@qtl-C`
z8Bp@ci;f;4fsnzRHPWc~j$Q(TQAKm?D{7l+@1MxSuL<4%9Sh^-^&4=R@tdoJ!VFG~
zo9Q*)9%gVg18jeQ<O=L7yrPY!@l;f{Vw3K?_(<c|mOVXFv0WV^ZA0C4ZzB7>j<J7k
zRbIWul_a%L+19f}1j+<tKt#`1G#+|&GH_dR8@<7KY7SC6b6wb_vajOp@_nDK+Rlk8
zOoFxy%Nm)YyGF2Xg#=VvK5fBk&Os5|C8}U>bGU7S?%sp$k-t0P=wnb_-*>vG?bbg@
zejqu9xj!@&IO$HkDM}u-!NM()?!*Pr{q50Med+D{s#0ObcQ5~n7yCVabGbv2g1c$P
z+e5+KT!BF#sbZd9VXhN)87N>BF35I{n*)9dtM{a(%4ynAXZ@ShBvv`5O;Kpwbrvs5
zl2wk~m`*8yasmrkf%jF-)Y@^LQq3Eo?RR+ZU88%bT$^@d`BeWj1C~XjHz+~p6ao6$
zS>*4{CjB0H$3V(%L^0T-v6`|kR^ms#xE#N0svsidoS$9<)<gWjvJA=)(lrt=L##P;
zDjrW8dF;#Y-vtJ+d<j|p*z4cOvRRxK;kX{>RF>E4ac2Shn&@kC>j<2on{B>kdfR%f
zamUiNx3>q@rp<P0F4x}MM2ke54ghqK&)a8SOIoQxTx(qm>xpluY4kUlN1Y_CvI*EZ
znpQ>xu^K>?(Z2h>kbT4geM-L(b?P($-f?d=#mjHoljU>exIHHcL>apIpiUVf4OU%i
zZ@G@vueJUF(xc<N{g>dl%*e>No$gTc08Rx9FStHTxLamSwSX#vilDrzxSfvZouPoq
z$UP^@M_jTh%m8F{z?mZ}ofH{g#a%#&9ypfkb-u~AUaQLbdi|BP2G+WhyVl_K<FT*F
zvXyDQaC?`m^5$!D>pJeY;@+C<dJ;h>fl1W}w-3n5v9Rb<klH}`9K=nxUL&x2XV#-D
zCr<0)-NT~SS`95)$M9YBk<m|nO%6VYa9{;L8g%lHk8vjW2Yn~bv+&tE?m>L_psT|G
zUh5XULowdyu8*9Lh~O|!*bhll64m2_Dhrl!)NV!qP!|x1*_7)<EUn!UK4qQVENXt9
zZTGO9e0Oj;$)T6-!~Ow6OB1Bc@$_KA>uywA6r}H-)SVkWbSv*yK>)9c^67PXs)g?^
z>*Z{fq4sHC&3h+M+x}l`jNHHXa*`(Tz6bDqzdyZ_cfHk=BzLrZs-pfM%;(E-cPw+p
z_=D-GgW?V>toQJ~^J;rm@4FKMRpM@s2MiF<vmJy4MHVb|#<VON6_5l~7O+$dN$WJV
zs6;!?r|CytmCt^X?Nrc2xcddQXSI{1&Q9xP+ULfU2%Jc3+4)<kn=?Fm>VdH$!bt>_
ze|N7)u7luq^Tu9;|4uAw6sB0a!raN_=?RFM4X5V>GoEZ_vbTPDIezRmx9JCX(!0Jt
zq%5x^qP5)m=ofclH3!Ul&uDAxp3av7jRO3^W%+~Ux|ikEsk*r`w>BS=W$D)Kc9Gbu
z$smgov&|Z^<H`2z8b<ZNp-$K2buIUK#$GoeZt4kjbCG@H8XXOjZMUq~GwQ{lZ76Nd
z46ARi$#QoN?Y-M-cBkk65cj4LzmPWg4}Y$oL!Pn&>$Bvze-H@(d2K7N;U*qzYtI6A
z|Hj&p?>(%iw4PYuilr9Jb<rAYNozwEo>1zHr4-}=AQCL|g0d{_Srsc#zFzbPUhm1Z
z-_vD*7M<s@N$p|nb!@DkUDpAXjn*5eTkAaX-o9X!V~cETYx2pxYmO_dva7OB*hDnB
z46i1@_Ib87*w^IMk)&VH>f`o;>GKy6wzv6ID-*1`wcNYUkX}dl5M5cXuMPJD&sdY)
z?jlz?exiaLd=%lp3VvihcVf}cCrZf3SgN23;Q~5A)?|MOn1!_tRv%IIj;~jx@nJKW
zMey=?01X3Z_YA`z*r$TaQm{Z!Edq<NAdkT1)ecJ$e2+VDDVS|zHf7ij<R@IYFi0xM
z(s4ixOR8(f(;71sRI4T<QL>zOy&&xdwx$8<M;V}mG-M2-DGsI1sLYtn@3u>tbI+`b
zs_*6k&@@gn4E$8s#?i}eTYYK0_RetKE;j14s8WzA18GnrM1VXDs0_?ywrffdDgH6t
zb+S4tr7jjSy)Lp|X=QuA-|g&pSK-W0)0O})w*8cYuawV`<5i8xG+;;rDqH6pE)Y@&
zjH^Oyag;0ah=Q(aT}wAPK|m-1yu7;wC1BY94kI%5bH$zs3lg*uA|sEN7-c{j2)A>=
zJQbj_rM7<Jy@#|j?2TV7teUV&0Z@sx?p(3=%6+9wb>3BhtQKjQ01PQ1jUxyQuB^Mg
zBFNsIRHW^=HXpG^VtTP%b0VrOx<)}Zhm@r^_%3noG;M(zC%#KfN$c0;A%WOx*9lv9
zQFXeBs8eFFntH`*U%4Cu06%bVe{4B!^B!p`8u2^d{sj18WO+;@hBTm5y^kWro|_CN
zO*&b?C$jt@bXJ+aLWBa~<>5SJ*%u*=iX@<m+_*2xv@t9NmfkC(&Vz(bmIXN{<h%n(
z(VB3!o(*Qr_Ok31t2|({FR*{Q4iYTwu-e>ff|LlzsoQ)L!BQ6p2BegmA)fL8VPNJ3
z03p+Ar~_`4CtH&%SPJgT@`pWpBC>T>Jp8T?I@T$P)KQq9d<xR%>(+Wl29T&6voKJ^
zngx+YFt!A;kLCU+j>2FrxS5Xt3BKO%khvnyz(5JRl;M%W14azQ7@+4>%3`$^PYe8i
z%oP)|{pm|uXg5tymGug3rntMqxaG~Sv|3cQ77A&kl1M>Hd+f#?q|T_-((!<{39Tu-
zvaYd&7Es<CWztCyd3lX4?py#=>6%0Onw(D7<hAp7${P8;lo_CeoCqz;LdbOMZ9P9f
z3$KCK=0Lm0r)#qQyvqB#pPkcLs&^Rb>wTRoY3~$WlTVzI!B3ZPU<E%KT}0@^oX&Xf
zc$3e~Y(|vVkn|~(O(7FVH=B0T>WNRQ-$&S-Mz=FqM0bm(rDC9joJWj##5jx~vBJ|d
zA|+rz#y|;!P5i>JLg??hws&2-d1+jc>l{D*t_Q=S!+O<lqr}2j!FFvPcDsI0AHmca
z%Hnk>Y|QK<$+v}=u%;}c<@WauHpqSFRo^+J5C5h1r>h%<HMvl~8%?oY6=&ZuJTZD6
zP`czJl<4hn-oHujGLtu-e_-e6De+lyEbE)vy=@U2YquXY=m<jhlNIiLS4*FLT&ec`
zPJv+Ng1J_${Q~U}It&BG0os+57?f0UD+c7mO<pv5<K5%);DgX))=x;Z`zKt{w1_2Y
zmH6P1k(<e>9jm(m?S^!Up3MiIYT?b+l5`5^>P2gYtLn_AeDj)k=W`BRe7%{1ta1!m
z<rF}ZL%=KI`TOinbW`zqI@@|QEh2L>o)0)Qeq=dr*Mg_5EdP+Q?9$AlwOstyj5RHo
z<?$|%ZpeA0()ynJ_Zsb8_g53k-v$H)1f@umUIc-FRF&SOC`}PC2}O_qLa5R^(yM?J
zA@m*~q4%OR=~W0NROu2z=%BPW_buOh|AzP6J@dn!vvX#4=gd62`<dtSJnpSc{^lTm
zJFUmn6fg6p`iHI~5SKTX9t<w(=bB^^Kp~+GN-y(~A4m?-LY2sVcLcBIB@v}xHs&#Q
z_IPw)!w@~c9GZm%NarU<ApGglsF(Qlt`-&|<KsLM*Zs$Eg9rJf%c0+K3bdzh4_|w4
zj_a~rWUs?~|ET$}&19*@85~J#rXQ#3Z#Q_K^f#F&|GGLQs?uYR=@K&^BCGwTI6~I-
z)C^7_t`tE?!=n%3q1b3Gzii&)hl=@Ps{=eXEADWZ%-8F}5zyt0W_V`Ur%=~hq5U!C
zgF=@h(-4G8F$U*-fl{wm^ZnyDTi%~?vI%u?<gR__0X)j6_(f0l0+Nsgg;i%CN_);l
z)k9-jP6m8jF%yz3o)_i|Z0k3cJP+!1(`JJIn<~lv#m5<$9mVw0&W6yW)LT%jk-u;2
z?IZhf)WlxorC%|!gtT`(;@ibK<yZd`q9xzs+s0S{Pr}{pATZOa1Mxa}5*at2@j7Aj
zd&p#Gosc|xxDbsoCznPS0SBkGfh~h+{Ch<%<p&F)!&$2%M7lZCp|sF1iKQcD3nF-*
zO((*^yy&##DH_5TsPbUY69@e#Ivf6#XjSAdDWrAgkU)9O*A*S!c>J!Ko(|YrasDGo
zJM^%f!DZOq(XEn$3cs0Yl15|6Gth?*xo*x$7XX#?ei~Gr=}8N;y*0P8&kW8ulvz9`
zKI>QmG7DEWdRwb@<_;64l7u4LpE@lqE8;}FM)_6FXIN9no*YJCT7tAfF0W|5SO&Ip
z{b`5W5OQ<ViVPEYswI=_AQt{C_TEsL$f%ndkS-<+nv7|G=gbgPTcpeDcGE5inrh68
zg{)ywtwt}A6v`x#IZjHJJ$m!;W3{&{eiJFsn^6Oyj*mHWfTx`hm>Zpew-2k_dI)Da
zCYn@gonE^^U}dt|sAK_&9ChJ>^4}@QNSMf}D;#<Xf@U^~0WcgFw+8lgOD#|DHuB!4
znJ;It%Y3+}({78HQb%Bu?8>c7DXYSD7%I_%3PFI=k<?veakZA$WNc1h?TG1iZI@cg
z^Jmy+Pd+WYt3&I-Z4o{ZHR`g~G%R1pTtFg5WL1+C+U-yW4Da1p;~ZvRs&Qe9Jd(K@
zpHk$7vWt8>`E%=J^B_k%o1~b<#Hb9?G%?$e;gU(_8MI?C0p{zFgG5bj(AcTN)-758
zwuI|fG^aCFGS;iQLF|g3$Q0|q_FOT<ou6vy4XKeb(?>IlUdPK-(#3$?yEW3DOr0c4
zZ@&?`bl_)~V2n%$%2FvWigUJ0hhmJ7j^vG$IZa1L5@@<aersj7ZInT*VsO*C1Vn~+
zy1S6H{=yz^-`e8VB$-`8w>1@EGQvozQMb}K1hnmV_hQ&gfPbPna&0wH$bzU|7^<%8
z;`$qAr|@LGzj#_Jr-0g4$~r>i#lAR%2a;)}M?`DBFQOBM`chZhTYpe0x;NJ2#N2x-
z>(4ISFz4|nez0xnj2Z5;7?gb*|4o#|2z8B(;Mi26<QC^CRaN%lL#QXSf*F8IT4M{U
zvtMlv6NR@uCW1sCdDcIASg2SK7?D8JP$yCt5<_T#7d6R)t(|jao*y5s6@-G4jp2H8
z2ZnvqfV*4xhcan&XTp5=ty}(S>~SPQt}>5v1hJYhcFV)nCZHM3`?_1IBSycLC>|PZ
z$!>8O9)ZgdrCGNpPQ4yZL2F36-hcImGbui?&Zc3=j|%KOX9a9N+<hH?WxPwd5S*&L
zx7&r@{o)?{o9R&_o}M<z0~haYVE2o$ZLm0_`sxE;jYKM;DII-kRb>tUh0D7l=0hOp
zsuM}2#Ab1km{^a5SwTz(Ir_ljCeETkPv;7oKw3#uE`N&<jT$@^Q>`C9A)Tt81G^2{
zu2s0FlKF|lS~Q+TH`}!g^LF`hg2ls;i6!iKs~2y%g{EVfW_dolPXmak+CfnQ*UT-5
zOS{gkV(h*`<Y@A0xP(=;Mh73;^WVYa!#3#8{sakh6LrosrW2~{n-G>cm=;DZ=jze;
zed)I^QVP!#D&<aTK_}h`SB(Z=<e1sNSkh4ktPKC2l_b)rPkrEX)})Fjj{Mr3k}~b1
zTzZCW+@>wIa=JbFYDejvZGaJ^0Kp&fx#gq#E!8LNXR5FG2BGiXfs^@Vnj{McM!Ig#
zDim$C(q8VU50liI;_}t~+8<uejxB}lE|sw*56Nsh+;vfw|M0ZJek$)Zp+?!Wt7Se8
zGnNf*;e$XM;pNWdQf3ZC(IyM3G8ODQ4-(<qp9T>=nYG(3r)+{~U(^j#3D@yZJF&K}
zZNH|tOekAM%pn-KdgAfhi5?~2dEc+^!I%w5Wb=`ck!xWIX<0DP&3sfDIDK)<hOquU
zhF+5lK9-S_b4sS+Y9P^K{*@i=qkgWz7SY{u&327hn`8Suw7r*@-2k|z_{;erB)-1U
zHi^!uYKe0q5TJ;ps~$qz2QPfA%gkU`sJ^_CxI|Y?HVVWvgp-lU7_X5P4pC(#$X?W=
zeO{~eL`46jxy??H_H&#IxwT&6{EYAV<hY;CenQ?r2^TfS3(pgV`k1^aPJ~X@hLJ<5
zXJhZQ*6l~Uj4l$aHQ{pzAGe^Y)BjF<xtM`UNTvN)SWo51WpnjdbQMNuFi+O!)%xvm
z$vR#>1dm3U2tBu-BeOAhNca_LfP*PR+(v!deEZ1#*niOPvwyJYbqmiv<$^|KwdaL0
zN>J5aMa!S<F7p>IJG*(g8G3V0h94PT&$6nhGET^@4j;1$hf#lNVYN-;^wsCi5F}?;
zimYszvG_uMB4I{<Gk9UZciEV&NiWgjRfSM3n9j!f5qNzj*|O4Y*e)&`aY&OO&{bED
zV_c3*wSuEDrf;lc?dCt@llFPhmsIJBFc&p|VmDv3Y12-AL3cW5Imy!NA860_ZimL7
zhk<+NllRV#Wr0)Yf4nZ*KkOWm-xk;&HU9Jw33x*ga=A++MOiP!6__+di1`VO>HW0R
zn?hR*PxIu0zg14r&17qn--`g`+m1sk-%o}$Fk49{STY0Pw<17@$4LR9<++icr$!0w
zRCK^^rZx!j-9~bgYHwynT58JKVshNA9VDUe4FoTeAauVrS`WS%WpaAFti2`!mX64;
zVdl0A4*FUkBu5GiDfR<vxAwD6LjgD1=DrsXCe^s{xSo<d>E#RbkiIw7$QrPoQ1!q~
zYzZ$tqtUsrK%}86L{xq|mG3NAwT>fr?_!1I%(ESf_*IMT!_9H{x2LW1Batv#$GTX0
z(HUg;8nn7BISIfTpT7EhX_Y2uN=RQb%auR1-O5i#sC#g&epq+mz2T!y2Zn?8u@@d4
z!o5_Qm#9bfP)QK%QD4~)(6lKi^8hO-|5{CcXCk}a9OY3&;m?NW)p~+wj57`cIk5=%
z`yc>HZ`l_5d_jIM_ts5NH_a|6e|L}Yc2S`j8&e(cnuOUrPGmA@AV^@A9v+mTwsYE^
zdynTH(5{?qPgHa6*YQQ7YAixC#&O(fzq9InS?t}8qRVooj;zl-MaJF|W2s^sFe1Iu
z9DTxAcNX1O8k6(SS4C<M;|Jh{Gsh1Kco?7OSGUrq_&7c!l%hQ#DA2M8e0UQh{z{ow
z5Xlc7qb2J@lz7-3YyN2F$p=TRuinXv1Ja+HT(i?SZ!Q!+Mfro@n?!^>yAft?U6PeY
zZif_t4RA-xd(;|^PnfKA>5Gm<WT?e-#U#-Pl+hL+Bun{;_jJ$1F~BX=ey%43XU`8i
zOjn=MK%YfE=Il#}OLwVXFfpAVNmZJcffWuvCkBe1BsO)oT4LW2-?O?0&CDctX`1Q(
zR~#CYR%n0cjDK@dJN$8+T{9yrU!ZCAm0>Ys<QA@P-tgUKW+0?wX^0}OdLIW7e7b}K
zI^wPZT3KpHBfjT(`Y(Q<*b<j4pjJVfGpfawT825PBB3D(3WFEai_iS<7S$yus|@j$
z*W^YuYd0WL>yS}-8%miMSPo{8>33izF$Ci!4#1*)7wRSas7>ND$9o7lI4Iwz>-8IU
zW>l-@ZQln*6{%61)k%#PuH;L_DWw?NnQ(MBcYN|9;D8&%zgE_P9SG%N>O7khfE|uf
zkzD4};*3UJe&RNM-HCh7tzB+EZYo*Ma)SfDRs-IZG(jwx04;=_8_(A1%~34*cVl-W
z^S49|zL55fJv1v1Dm0v59GRgExqKSr&i$c0U7LxfbKSpNpP0-e&^}A{?7`8pLi)Am
zC1^R5fxnR6uV`tLSzrv%GqTy1BD}}bjIx>L%Fw27O$s9bYBe|*@C?mmG>>|kfTu0X
z?13dQbEobfv{r%_RNFy@QHmmK6r65n(AN=!kXYX8Hjg$#`GgyS6F<0yb}RaD!SiLO
zX)DL&lvI+BjIXa-ab#IH|H`XiyxjeLw}cTsULYA~CTSq6+ZKTn|2nGTii7(X#rcMz
zm8+7X3Q27rP_%?N_bI-E&;t<1MlZ%!s#-+3H@oCWj#n$B?DMoI6AH*7n732)nvmEv
zj~(lGHg^!V->9O&14$BM&m=0o8k#ZwRN}JQ;T5`g7}6o#K4W#%EJ&g>SE*|48zHRN
zk0&>1)VXLtP2gJ31L%=1MJ9aaGEMa6d@u1lCEquB?51D^hh<&6LUs!FDE4<x{`9Uy
z_lY8pCul6!lrX8e`U@oMsLDUNKwQa{qi{?o!K<lqKM1QAF(Xg9vM!_EcZV5^6>?xN
z@>VuQz;C(Dc|mwi2#;Ytvb<%f6p)Vi<8_Xc{?4dfBC$R~2Rb*PQ%k6GdHA3V&jFS?
zap7D^F+{N}N5)@zZY5b(9hrI9786&Ul=7$TXVHi~SmW*8PMylFmh&?#Hd$(lsVX9}
z%YI@reUj@dD&=sep55L4{D<!=P`#wo194x}>!m5HoF++m9A?L{%KGvUPs#f>bnW-y
zKzQ)JXeh6>x786oRb7JI2mww`RwSr&_$sO5T6Zo0)oxBQ_{0x7l}%g^2{9jY^|eBV
z-b&&XBm2X?wCEBJ^OU&hb_Y?q+B|^I$(;PPtLFcLxRlMZ(_ZU{#_%^VTK?1*2Dr}Y
z6ByiW4<PqtT4)~L+Zd`E{#mNmGyKu=^fZ`nx_2qwPDQ^Rp6vW|M1w&_b}o)!=NySn
zzv;=lGeHt2%gL?OZW7iRo-ugCn$O8`zb#@3G9oyg_IyhFL1R9Iwf;oI2SOX@=F*c_
zoX#@m`}1wKG^>L3QbAb%2m4PYNr13fo$h>@LYXHG)W#>DQ8S4J-flX<s_HFl8_aUB
z-B<_h6H(jowCsaM>Z#2QTKTU(*Sx;$9-|_UpAk!);l-oqWcg)b@JXVdWcvDKEVgP1
zQF$mn`R-2_?~Ms9sgnwqJyYA)11+D{mWM~9>Nu<Sprm4yj3Q=gQfaLv=jSH56=ll6
zvOstcz`B<~=&hFD$K$SjD=OkZsUSr&eUl_0t1^I&Q6OGYu98M%4RTwYdp*RN<zurj
zY^c;zL^Ha(?}oX3Vu+6*y-~8meS5+N5o~PmNQPdWexRi4#}bVD1nb0&%o;4j1)0f>
zxuCCpv~MRbM;f?$uptZI4<s+OONeQEb_8Czdw^6Vxzo$UP5G2UoD?QJ2$4yze!cla
zk@04y17uAsz2=N_ctL(^bW^4k1x{VmM7cy8J6X<Nx?@HJv<SO-MI`20qWIXO1hS3E
z!w@e-ZhsV13^p~=A$~#DCGsBpsG_B(4tzfrLGDpKsCQYfX%T<CE&Kg_3HWK2B7Ifb
z0=D8+7J#4{%?yBIKR<@S`VzEH+KPUFa%m0bX0OKnbST*Cm-eybh6<9#=riHNg>v=C
zy<baPuY&NxAPV$MWBrR8n|HpfYt=F7Hdh_^gHBXv0}Dvlguo!nG(nY@fe4%5p&#{f
zS?O&gttR3V1U|~jJ;&p(n|6p)G5TmSz^hm$VykCaga<FjB8m5UV!ml`Uq1^Q%vGjC
za0atuo!SA9=%mEbD;@S(?74JgR;jRwjKS}Cmn)XQ;X*{1aF#;5vM3IvR3{IaV8)pT
zA2kemJ;FwsUP_JFbjM4gs|SeeNk26EO!_2GNta;XT<HdwrfJK+PPD)k5{xG|quM_s
zFT5FE%}xf^1lItiV###^3lJZS1D&UloA*CW<^5<&cjTXHB)i<Gv(`Lel`6Cc5KhH#
zKA5cDLZ$m>`b{w9rRxJU0LZ4KtK8VQK6Sn6mhiPR*&{nt{*;u+|JV)m=c>{!><M!8
zYvAp2F#64)TK@<f_DTpbk_;0VOL~JNtuDHC?6t7~l5?8uZHV(#|E`j45V>6BTvs^f
zVaAr)X0+8{$6)67JdV4SzSOR#P^h52RN{5shDm(A3%~vNu>x+v%>Ky>bh?piWMkeA
zc)japd;@AX;PhS|m;Z?l;5a_;vpPHCRq$!F`?bjF4-d7>0az(s3tH!XXgd-US2S>@
z<NCtuaCVn#>NQUQmUUAwP(9EqybAjpIWTQ8dduz2kH`LVjge{<>(~+g!1E6~)%Fz)
z-k?ig)tQXd7OG#_+bHUVBd^5CB3DfXL)W*wBPx%+#(^*j9>ERcSw2tJ3_%;FDMtGT
z?0e8yyk&5B7IrxJ><-R=Z?KFOA|#C)*N*dx{kEI%QluC!{w4;}3FL{^k#$QwNtefJ
z)|SWLM`9WhhijT`Tye^ed*|fm@d^MIu3XvG&d%8ss5Ite&cP@KhvKB#p7e_Tw+Awp
z-|#t&rEjKNx1G@bqgPdG{5Ia)ZkuXz`TkolinoI7DHdjK@`>tdf2`5pwifqya+ouF
zup#sx16KMSOs>KAR-2nzq@?AJVizRcy<M;WJnpnZA#pwpb29Z}SE*Fj>Vh@ZwhWuH
zqr?gTo=t~eBuZ4o|0xR4#pwo4(jbn%%ascL@97^^l$@sUU!H$o58`+V#e>`5Z75M?
zNaEko|7>Ffh7T40rxW~%wKhI$`5&DG)$p<a|5>5qF9D5zR#-;-dmDd`*uSIk@7(y`
w75DE-`4>0-4~_pt(SO3izoYU0bu_N7_g^A)J!LRh0{rn@Rp(i`vRTl70gN~lBme*a

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_video_api_thumb.png b/0.16/_images/sphx_glr_plot_video_api_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..649c59b34bbd1a6ec0243386294743cf78ca1ab0
GIT binary patch
literal 84921
zcmdpdWm6qp({*rncb7wOC%C&qa61GixVyW%yF+kycM0x#0tbRiaCx}ys`pR4A7-j+
zs%Fow>gnFyYpsq{QIbYMBt-o3<qL|ejD*^kFAxcz7d|}f=MyLW0?#iHU%tpnh-!G`
zoV{7q*y(ufhQH;ttmN(NPrKK4Y8q@Wi*wOACeTOD8|yS06R9Fou*Q&!B=jc*P|JLw
z5)sk~K!=8>q!tN)j!e`h145*k&?J?Pf2Y@*mr%vj`_Tl)oxRPqdp&*G9L^>1RDaLc
zwq~MT&b-|%^Z5R|t?D`NHm_;b(7Vp!m&K!SQPnFc=wU0M&365!E3Er3(&P5<=Y6fw
zY$}cEqi<p*z;^vVI~`~v()vGBzJZHe??3afbd4~9=|A&E%@$=m?LX5K&<gwG^G&_h
zZfrfM|1F|t81)O{e`_$YjQsz<9O6S2%iaUK|HI`g6RH0zG$^DbYTkC9D&dQ!N}!rZ
z3<Q*`&OB<~ApQ3c{8Vnq9Q+s&wNggJ9IhoZ+0B2{ic6BA${1u&E8h^wtspTFK`M&v
zMIu}c)_ND0r05!+t?d$As(j&JD1#vVZ;U?d*>~69evUoNOOstayx;akw*(=Ul&oFb
zK#1J({(_(*A3dlMjol0HFL6Q^LJz>kv7qsnK`wP(GK(ye;78`Q0qBc@;ffeqoYAQm
zf-r))JnQ}-O7kESb8>O$HK7LH{)-BXfT>|a9=N?v-`Rh7o1XLgOZ<Mo@=DeFHe)oT
zLs3{@Jl1gwI;{G0tz(LgF*(RK98sahD-ItiM8g==g0AVfqN**go_aSV0-xAOuc``5
zrh;K@Dgy|_&tW4sj*&$6AoHki3MRvZ@aSys{}x;Uj2)4ln|o^zp6G$Adus)Cwey1Q
zKXyO90T`@DZPd|2$)9vc%0nPoSAem`HKk;b$&>6M5tJTe4&1_HRbrah8lgiG^(Wpj
z@S3U%C{VH?3)<9@NdXwi`6w`Y_za*7Y?xAAfyoi48A>-^j2gBO{0`qsA^$%{PwM&K
z!u1UEenEJKzS@7Re#pO1eXOz^1O&c9rilo@e|#k@a0m$Qc?uaDBZ@mx!3(RE7x-yn
zKA>|<b#|UIeVU}#<t@O+Ur6R<cS+oEwn~kuxn9sqR~iyws;gO&lA&Nx!s+1VH_&$z
z%Hq5ZL;CBx&*e!tXA@z05C<$*7}-Z<{*wmOO>Er<yZ^=9hbKdSa_s1j*NwkeyB1@8
zZCsqC#nFLoD0RC`I*hPs7-6L5I291k_vKt|_ihM6S7$ddR?DoFbU7?67<wEm@;-f1
zze+jd8ePc4CQ8bX!`kdgR8W2sJ}$cp|8dk+(-Z$GpO%gDu`T={+?@#VRQ<8qH~6vt
z@u+1Pz-&Y~!@F{iwX5|73rVftNA3Ex(c;M1eXKzjPN_@aK@mngNcKEqmBt<Qp0B6f
zc>O1D&N*9&$R?WhDr;g$JQQVGSq*w%V^sq#>UK<9{C#`7wt5QA$K0h45(is*iH77v
zqrt8JQ~XDhG8egWzQBS19r)eAXM7ZhKV}h#QVc<m2_XqniZ!e=%M4||#jIls>k~8r
zTSkMAq`GfxF*x87)RZ<>K89Vp=g>NKd#k?mefpFT#2*xZBd@MS9qRSh!`W|q+dLd`
zY^45x|5ET^wM*T+w73~&`}cK||DEz@#zM>d&(j*3vQQ~l;+dGVqM7}Z%go>5A+f5l
zQ(@@ZJAvQM*EjP`IMvpMX_#W2v1e6>15aG!48Bcvmf2dDuwrXY5W(^LJR#ZNJkDd~
z<>V|peE|ZZDf88`5Gb_$o{=7`a*FHrs%L!T@GO2~r^uq)pxa<Pxm=dJhDC}P5cvdI
zVw#P2Y2nBqLKL!3mV8mi)i6V0`<)VL&JMv?8I4W*dkIHom{6YMXGJ#bC1gv!N;+4Q
z*C4-<4^q<2pKE7ko&*Vc+rXBfCKy-s+kzYpXDkN<vTs#IB}rMG^z<>s)WE45DQE2z
z7!+0lJPXN6Nctzv%rEh*4RY#)CG{!EVUh7?bJmj!#)<{Wd14_|jE+{u8w!?xgR1N*
zvPkVys%m*F+@A*Z>YSZ?W^Y02+&dJ&$|y{lc}^@u3lC>6I4UtgJbVFH9Z?WHc8+UG
zQAA7M+uk_cZl`04FL{ItVJZCH&3<YDx&(T$0`Xc^fddWxkl|UzwE5n*4MX^bQY$%$
z8;k6+GHdG?{J*l|hlp97`xbr&)x!zAf(G^4bU6*4sL`)m%ZdXRKW2xO0P<s!g!EBK
zBs|3~xZxB8!PQknGy7@j7DP~<spF3<7&XbSKnSVvF(2&l=<{WNnJA4?aSUW`mLppK
zU`Q#Hk_ef1A4u#%m$-b7m#+$JDU`fgs%wynj@_sfEbfDTY(r~NpB`Sb{avCBt(c$@
zF*q|$N`hXmR1b2E3SeUyElui^ruHA#5!b=7#LRIleo?yd;Ot_friqbp8%i{Ax9`6H
zmfv?9P}!h;N*|3Vno#``QB)PeOuoZfNlV?50_3wTlbNh#e?zVd^*IpE!J6hpB^ayB
zf0cp|C8w!c2x`Eefz?rbPe|fHtwL}X{E7$<L>9guap$!@QqT`{hAI8p*r(v~T9cGi
z<QwuMV_DrNq^HXKhyfqW>bNL7jaNu(507Y5i_#wrz8mo!AJVA7s462L_3={qdYo#H
z7F-onc7~}bi6N#&y45$OW%xnttr^>dtCbQl`7hb$5Q|uha`=_;Lg7eYt#5|Ejg(wg
zCW{C4Sufsv%0KoqXue9YRX}h>Vbo_CiSP?O$&MLx9*roE*Cup2ApfG1x{`{hqDB!R
zHl>Tg%JyB%GP>++Znv9IW$)f0W^d#2VExvc0-uG+2BXlZP=(O)>k82w{4!liY~(eu
z>RbFzS_X20o*{|^1O?KKf&?7085)FQlybwUQanm1$Z?AOhL_MlX7GCRT=r>_F#g;B
z3va8H3+^9u5y53CCFo723fZxsipZ_(g;>G7+tVYP2OK``5rocI#3Ps^-wQU^*Ko~|
z4*WqL8}?xBQ-Cd1VKA`=#YXSQIjKrjK;1weMWh9;X2T5~2~D#Sc4*6Ccjc1cpQXrf
zG>HQwaH?u@#F@XGadYf++!Dgo68NVj<)Y;aMQR;sp%~u~8;<-<P#*q^ouofQp)nc=
z?@J~t8WE5#@jm>h?2Dle#kx|jj@Leu1*Oj6j-w=B17C5ZZjY%!Q6eFjtHR@apGDAS
z?!DER?9)lYqV3Yv{Z_xev#%8qb!5<XBzU=;e@bHeUVBQCPmXU&ua7G8oDxxmPZiNZ
zO)0zBUaT$_HJpJSDy!)*MR1QDXFHMW6ogv9mTx@m0aPu>;o%&?4ld;m;{m_$noh{4
z42?!*Yon8lIsHo@Xngr=o%4m#l={t@O}&&JWxyNz%8C=6+g~tLL`~96Uj`_lSs4}N
z8MW0KrZ;-M@k4H(;tq6REutX<D3V&wR9LG7pUvI#W}H(B6;eiM$9(%^j*Ho|5gA{-
z*Am#B$I(MttdD|0j}-h6NbSjnEfP~eD9iRI6FmZeDNYJ%re9m_XTrEh#d-LP&=3pl
z|1akcj)3)GN?OlENUnMKTdv^!UjhI$da!ZBMC|wdMR0;+%(vO?WhwB8bJZ${tQ;K-
zt?1~3db>K~F&Ra8!|{zUCxH?`W39!C7O1y!qbqnAJ3+iD8K!hgVdDd*k;;s$FclV=
z&DwB?Z@lXG^@js?#@z{XCsQ9sE9@T}%$_0)pb960lj{xn<zyZ1=4ra~)<8JLpBVfJ
zW&-Dy<VA@ki3JI4nnFuRFmYEhLYG;jXan0mB<Dy>A(n*vkF6d;C`q%*s3=*%eM<{6
zqSO>jS>%d=$gOdRB(?Xq`L6)lh^IdeslKJxSj=5tu@&S(_QpuM&E0W=DvAhyr`)Vj
zv!rE(^qj<WB?EARBE{J7%gn@)QFl4=M6j`-v<qEutgbj$*s_&bL#YTwmyOwY+^&Wb
z@Y;^C2(ht@rnMFXGKE<MV$8)Wx-;mKM`MF%UG!CgkMX~%)i@H*xbapIF&e{)<Dev4
zM_s}wqM9OC6an`BB4$Hh%NbywO_zI6Q740c18ei!c2O0hU{&(RyadBR+MGQeN{3k4
zU!XjM3#8zSA{b~5eEI%Zq;_LO016q1r@>qKbCsG6!R*UHR(u0!=sgq85kXR&g{5~0
zM3f=@tj#W2pM47#{*SKqH|F}Du{v~b*HL(M?^?VS8Mku>S7Nd^7^}NbWQt1_bva8l
z1>&~}D;IbDryc3lFr32zK!yg2t6(WBO#`0Xs_{?RI|`Oz!UhRjwIj`;>6B@=va_>`
zf`fi|3d!0Y^86rzZIX4FBVlu5b(I@-!A>FcWGyO9U_?pAmC&Uz&0#24oyWI#Wz`uF
zoQW>~eHn@(S1ALciwrMQCU@5z+b$jp0?vqbq1-yY>v>k84M9>Ivp7xFj-}6+@3)PW
z2tjkO>K^-996#~}+_tLA;j^xG;`1Xt@r0QNzjb<W0umju3XCPzveTwd!#cCn@Wszk
zWZ8yru)-3Qi&ZJb1_|hkFcF%oVTSgu*<$(MtDBC>ss4a9a3bUoJ%^ld<T2g?`Jj`e
zIAMV@A?Tu}aVI6!i&1gvzFhg9RGf~3HT5aAMfG9FSfSOK+#Jt-;}7ywLF{{g8g>Ap
zc``s#v+?8`I>p3e&W<f&+bim4qoyTeQdk?Z9cg;N2oQxg)~LXTnu%cxnrDTkLjEIC
z8ABI;)*(||O%Wb@t_iK>4BafodcayXExJ+SaS1pJNugwiElHJ7i;s+q{5qepmU%9s
z;JHeFGAim~6H*lLG(FE7tU5YEESIXkI~c7k)vI5KM`JBCpwA~;y9f;}af7NpT#z-U
zi66`c%LX&65%CQe!K;qJ*%gsQi-{p$R4I?@hU(z}=89>s-9SwwmYk%^&)#R$fSIOC
zn<P{Vp)hyX;WA%3{JLth3dBBE6(f!vab*;#gE}9@&(Y7tME*`Z8-O3qMe{vMK<d)5
zl>(hL3r`8E!pqHE{epCR-{B(yol;0bgLAI1BQq8|un67FBF@SxF$K&D;0+y=@pBs=
z<%#DKE;h*-B9=}hK29f|LQnKbN@4>*j@4MvPenCm?CFO}$=}EsC|t?TSL+f$Q$@aQ
z6Ttlo8Cn;r+x5@Y+yqkRPbYKbGpQO+$u}s1d_$L9MFnw4iO;XfKbG=~)_1{VZ^8bv
z!cSvHL$Nw?`OQ~|*rCNPajX(G&ch|EQlnXTwKT<IQI3pjRhj9EZkeTLUwQC>A*}dG
z^4MmSH2VYl5Z<mM%{A>U-XX+vdHROz`Xe$i6a*cr=-ayD+x@n>S8#qeNUwKu@1jC}
zuPJ?R{x6=-n$c`e5X3)09{Xb38V(JFzDKyJDavqvLiymTVStk`8EOu^nfpZvXr@ol
z7~(*581*lcb~=unzdSuoQQ^FuW7=d4oImb;nRnk%Gi<-yg@J;J*J=s(R}6>9UW0|F
zhUWT4Sp3hkU$ulkkW5((8cbjJ7##L;imT+pYKL^vm6M7CO}5bSF=E7G07P3wM+I30
zAOM{`IrgP9H)abIrF#}Z(}u&tRe&BpKXsizyK6pOI?~1;3>&9c8R(usw5bRcjWjW2
zqt`8x_dT*Vsu~^=f3W>)x`2?-5aGys02?Nhrn-T>o$dbbop_?q>dT2Rkt9s@!Sy65
zVp?{>Xn}YNi6B6PwMt5E=h3fVsH(`)Dq@!?$FM<cM%Q*FU-cXOwpWMs&mbAEiWT@K
zZ_Azj%8zz=CH$WMHs|?L_eS?RDEvey`LhD%7)O+iG6Y@wfh4pkq`>sA`8}1<ODWf-
zar6(F@2bi4i1b+uS`v7_<&7~KqYLAWeG6pbHoLlB{}36O(-BsUQUI<u<_(0t@fRx#
z28|fkHu(v~Qc&ly?EP4F0=EQy%zpgYHRH~_J-2@)nC*K%)xZkAmBYIcD}^_>TGCZr
zd$lyjOQn3KWG)P#$`_+VNKE-!lPexXZdOFY_9&74oCCqBxRrN)gH_ijYoHB#S*W~J
z${I3G>mp;ts;w6FVI+0;F;NyTU_f8(Hbf@;ZEyF*==J6GhUFcz?<)Rd(-Kje?su%{
z54Oh?%OMdHrqQEz%{T)1B*S4EXnIG}8oEHQ8n%uFeUWIH>=YAt7t%>Fc-rfYcP2yO
z4O2sn=AkcHr#|0@L?TEXxulq6sdyrPj6j#CCu;JjD6YKfeZ2W+yYD>t<`928Jf8Z4
zz?lk+k}|deyvcayLev@s42TIi+tLSKsAEmExX~jqR+EW?S&rXf)J2?NW94`xfc3{2
zmv0?!T`8#pFYr5R>3OZ^Yas!+K|jX!?_aOG`d32#A?iGzl?$FylKDO2m;Slh5p;7$
zydYKV@a8nzz+Ua@ViEEg2Vd?CjL7U01CAcQu$rSx&|%($<~U^+ckaGjA%M4G>n2x)
zUjyi(k%GLqO(NSd|4jzk)K8%b-|=m?oQ@e=QjZYFs8V;$+8K*?GX(1EyuF9doot};
z@E`01>z!R<Z}ka+EbWt;k|ox&ryu9%dSnR&Bu$$1=t@sRl{7}>QhmO4&3>W0y1UCk
zQ&O(X3W}dqei{X#T>}=vQCUbKzE9zt-r~HONgwn3AVcrYyOD@1fZirQ%MWolFQQwg
z_PfY<ArG<sMsF}F93005u|m5$r<wcr#MpwX(Omv+$5Y%1Z$z-v$F`=LUS1R1HDkP^
zLfmic$9IrL&I6IH$<s=5yywGNMnxibiv(nmgQK6UcdN<T-J?lAoHDW3im$ZBe$_Fp
z#jx))In^XB&F5Cv`G(4dGN4EWx~FD`9X?iKHsP>hCS>uOI`_ZY3;j55?2cgz!(Mbv
zFc`fypsSroCE{K{4UBFrtzrZZWnGdAn?)r4t2dV+6-5xL*R9dXuR#!}PG|xr(E^Es
z*%&~V8&aZvYo9Gf^)O;f@Z5R23mx6#1%bHtYzd%pDyWDP+syFC-=YKg2hyu1`<@v+
zcRSV1ZoTD<#=58HpVKD;RxZ6-wIwxQC#wpHAmj6Loop^i(d@Se!ux4!60V8=#uQ${
z0UJXGu+)dqNDPF@uX!XX?fwXBO<csRJ}_Kp!AybZ?oKUuiL9!Us_Hdd*~&2V423&K
z=6tWeb~|4RH7)eOtJtg<2X(KJug|2Q*J$6nKhXcp3OM|8kQ1%BxBu%U0$+C*Q%!<Z
zA{M^|yMfw^hIarL28jHPSs<fE-fgi|nu!<6XssHsnP>Me#_cJm%yCImR)=tMvM|g2
z&!vFJ_~*AsNH5M8dHFBeq}!J0^tI)f!{iZNbsD?6qM2-T$vl00qPWyA&d8E5oBf;x
z=@KKNVI&!^DH7zO8&C&|Oca7~YNu}O#0+!|cPk%XcVhlcAc*_K{RYcDEvPPcEbK;H
zU9FJ3k8AmbuxxiuEoGi27qOa88;qSplbrrw9Q^f0#VvH>(_+5u?(_N@R8=!}-9nxG
zB{=>Qw`@{BW49H6g3$XWr)2`Ddn@Eo##+1KH#r^j3Z4GC|E~Wy5iXtMf0y#HX!LG$
zeSD43TC4pJY82H)R+a56y{r4<9P2DNjYXh+YJrZ_ASXFyuhvbsiaMFYeNcPT@nhB(
zQPVamr;dM+vN0;J*8o&fB7)&HtZSY+a12ev*z<tRAw76tI%rdrk~83}ju`kx*z4mx
znVZY^apM0-=IG*|-_#`UHA5LeacF$nU2|Oywfnr`?(z6cnr<+b98Gh&Cf+IMI1A&B
zFoNV!Og#ON!Og({SG}ACY8itQ=KxsR!mybv!lpuMtq@;a&fnn_DyvAvu;DW#5!M_X
z#xzLn(<(4-<@Lw+^`d3deC!*&@7C-7lo_oj5WsO{gvPnA2E_Nh<XCs?M?YLH(j0np
z3|3Q~Snp#cVHDB#OmwP0pk4wc>2vpAfsuBYtrZ!ou*JC9kR@ChVU4X@g59yrfvnVL
zalje~xrNkJRlDwP!oW8v4`;($2mkoKxBOSd-lrL(gPvWF^gqO>wUYHACB$4<cT2ZK
zeXj;P-F<Vwp8fW1hmI3xbYD(VOWeG>DrEhc{S=*Zht@q%IbF#f@76lV7K4I@Sg;Dp
zI;E)jZ21NGXZFAK`eEwfAF3$E_%@+-uN8Klvj25L?CKR7Y)SZ?H2(jDpHH{?T*F11
z#Lks_Zq0gM=AC<AF8ETJuD+V|;JQ6P-g+WLvni%jZ~t&R9#-9fG9~D4?9^7dn69V}
zYrWY5&^|Jo;4!HKD%3JLw2Ig_bB)EQy+%T0R8z5AXZaqJwjp_ZS&%T?0<i@p3qOzB
z=W>Z+&P6COh3>bbY+f6W&PQf5E}Hwa9g#C{)CXFDlkelk=*d#j`fgzTQjTiGl@tL_
zr@<bc%OVQOS|B@Dj$v)U@F&bIr`UnM0GVtRY=m~4@OCp_uHWLLSu!}I@A3r;gEmWD
zE0XEs#OP&iZFiT(dIjZWqRubzAuq?F!%u9V7WwOtInMfS<k^L?N23m~!4$_`X<gAY
zs+C?F6hS)6!u&Y7LyVHulyZeE9?r$uM8hQ(#V2(bqoP3gtT}G#B6e52JAQ=_{Cc8o
zEJqodSJdqDqSZUg@*3?wIlnK6{dkaVuhJPL4c;weq%Bc8Mmc7!=l{wVPqHDaq|z=o
z@V#TU6AJ@+ESyZx%;_PZiQ7q=cn@*z`iU>JN#SQ?L$c2n2Vb^4V(>lczaZx{Z0CP(
zsd9ln>};k}XtBSLPH`v^zBAQ)Q&wNLT`^V&RCTb{$+3T&e1p0mQ=HF$8X-h>WUbgZ
z4d@^;?Uj4y=YPJUAz+@xtzqWg+T8@V<nm!Mwv}UDTwWvvmCq-QPB%tV@%=3`{+5g0
z6><-2b9;f>&b$Q*lNz?JeQV!>E@FU*`zKWyQ`!haH^-bw<@2;d?5^2;AYX*HM0b!d
zy607kBV&r<tpqVsn>bTu(Xv*r$N4&bTDM<6tv{1~u~WvHMhW2H7_$<+8k|Hc*(eGf
zv4@R;SYU-Un4#zQzx`69Y3bd*<uO@QV1otFWjJMpn9Vb|aWK4no~M@oT_qY8gq^}E
zfgcSpN@@RN8HnL*K<DDvFx-|1cPqHl>m822|2~U~hK5>?7^_Q|30zu0t6zcO03al}
z_Qj~m_$fekRnaXF!OWwGs4(*+&g>=r;5Cx6)>FZwp_^mLEbw`HCrzDNKL1NHfz(nB
z|IVOPZkoGjIU-SSnswu~y};}hm&Ny)#zhVp?9R`w54|7AVXCO7r%_?yCK}Jaaa;AL
zz=#VMuoXCd@sfa)U1)*WxfwFE5iz8&HeMcDyp2Q4H)`zjk4X7MzGvsY3;14|J}=Vb
zclE@}cgS^LVx14FM-S0n0nlesk&E`A`+>hkUgFv>)>^tUiVB(wOvsZCxu6RHZa0a2
z5#8?3InjrZrbtWlt~zSpf|Ge+8R}laCsNGlTw>{aPaC(}V`DY2rJ0HVxvN6(*%JV7
z4BXw@B`^5S8*kPVlVNNbBMui2P14j?>yi&`gQALcx*TS;MS;qi&!rs;O_rfSGc04m
zK)W2Gjo-<6HQz+1Z&+Kvz_vn($;KT(fC67keh>>*(T4NrwtZ@e&)cs;FjH5>ja>Gc
zxlRT@x{hVPbm5bsv$+h^>%dwA0N5^98N>F3#m4n3zfqw4urePBqFD345AT?csPKy;
zCd#tSMNO6-+WJm_ER@%qN%gdf+f4`80+vHmr)Ct!BH#rFt%x4?9&@D2<tIA}YY?An
z4_}|0XXEp)uZ+VhHIpLm@YBBLgMVwvAfr3wB5G?m1y*qBSpR6T&Me_e{kfK{Oisf&
zI53Nil`4k#w<-X#sDF%ZOy4l4c(#xdc}Qf{Jp`XsHvwho9{Q$hU?reiY&b2Y)L|V?
zX)e9lSKfe^Q;Jj4R)<3-R_nLgdPtzVa{g^KnegNd8iLw%sO#SWoAEBeiHjf#3F`i>
zWtUvAPs72J*Mzik9{XbgQn=*FB5abWOd2?S?!+e?aJn^4k19b%VIbQ+73=Os8CF&Q
z2wiPw++nD~0|S4TJBEV_Vi2oe!V<@W%lpOrxB++R+^b{ys}Swm(d}15IC52dq#_AL
zWItoO#>OjP2M`NNSxITlg(NY64OdfHF0dr*@V8@!SDpb7cqMMU?KTX*Dj&yC#Up>n
zRaU?dvW%|XCMdq9@*RePHb{A`=*wb(wPQ;?q3({b@V=4X6@;idb7wT)iKr^1IBst9
z6DKxjqK|QT#g~0x1CYckPe9llo|Fs06q)Twf4??u>OM)+9fB6G+-+NA^I=$Z=~|cl
zaB}%+Q<<`rUKYOEtpH8X)xbEH>TxTlhJhBH$qaOb`AK)~*<AYf-!5aCrq0qOjcaRW
zlGKJ~)RYOlFLzha>&srkEg^!_qqlhFp;mBos1%x7Hz(oJvN^(*u$Y>+`>RgzYn42*
zIHvD7N=)rN`C8dNfe?hP4V$%)aXSt(E*I=9{#qR>7um;+t~u?;Jntq#y%dFq2P^N=
zxTM1laUhQ1gMHuZ`$fv7?_-ZD3-u|Ijw>(7a+Sp5=bU{`{xvI!qYb8`+HVSlGw7Of
zTSP|B#LWivBy+{8)=PY`f`Uhh&t`;*{&uv`RT2YpmcH<`A(`Ndfre1G_SO4!l$bGX
zoOz?PnOnVe<?^1s=#`vY*pqGgKd($dQ4mZ?#<(u)WHJb;_vsD+jO5{9Bf^TGyeNPK
zVm;v7SRX)R@;#OTG-xA&Pve9}x2%+%0v^hMQLu*&ZAEZLW<x+rjC`jNf3@!v*EY&|
zx7R0GJsWCkN$WqzmI=(*v4QmZ6J^`RKSz9492^~=#G1aRVnddN`EgwBo6qp%2^_t@
z|GYTYj%jCeURJ{|tR(TC2~q4Sp?A9yqLiPK+yqHx_FH6$qZ4YWUGV)_WXn$vDI`;O
zrV30-p6LfgBc+}9cRO_K%67<`$&1Cb7X|!iT#`4FH6hnYZy}ZzFN3CLP|i2HI;Z<g
z91`ZqUdX)`W{?($NuTWLWkR<_iw>q<MI<qcR}tZYN2Ai9g$n29P2+zQ#IPkf3dyVj
zy@F`rBycf4(0b~r5q#HZmY9<NdV}}4KvV}2203;xPnDOK>x{9|eS64q?Tq$;())C=
zs*TbYctSN1FfV~TnE=TKomb0WcI%tR^!-d3QBJRZ0mYrPceG5;p^Z7=7@r&JyhbwE
zq=T@X*h_|jW!&rbPPsUW;F-}Ec!Zx#t~0BJglKUr==1s~$fCOc$HEoe$FQ<GE=_4m
zJH^%crC<EyudDgnaKG31D|IygDF5e+T`<w5A4#fep=N4Y9L2rBrbL)ikb9y<I-Z7P
z>~2nzH>>H#4yo<;GftQxv3C8ug|?tOEatSR<>zl@E(XO8P&Kg4=NMXInyT#*n{XQ@
z!C#4^m-`Q^K%V()n_W}+kEX&(@Nx#)3x`=_U~qF`ai;Q<mjB)EpPc~+|69C|Z>Re?
zKSbfg$$!r`obbo-KrLSg(fypWLn-~GJxqgku|298CSJ^|$BG=;G=jjc{Z0LM3k<C$
zJ#6vGC<qHv(i%Ysn4y-AV~9M*GUpI#c6D_<5rn3+(yaQ__WpPYo`G{m_*sxNJQkmK
zHUAW}Tn75ZvvZ*Qx<&@_p6eZA@qI~`3WcBu9UiGKAkxiq5wuX%q1gxa6jfOR`+&Wh
z4j%S!efrL<fMb@u*H`>qy~*Y|R$lWFh^#bx?_2eov|R!{yp`$_^dMS&H38$wHAXBT
z<4?>u7Tp@%i_6Qam)>X1qgy_JWK>x}o^nOincQQ;s$Z7>{V$_Gm%cA#=-%PG-4!ZE
zGE<K3QD5bEd~RLeuatdL;m`~%xHExC1KItH7Gv5*z?{P$4JDUY(RKsZPeI_h@3zmF
zHeZ+Ynd*3KR*GShpwtFs`qv=)AeKxF25u$CoNzJ;$I^u#zU6zLQb_)cgq(TL+1@5F
z^fvUqN)W!<L%WXum{@%(wfBvl(xhexV3#>~I*PeHq*<a7etGc=X(Qb-6%5(3-z3C+
z%W6u7DYfA>V)F27IIwjcI5bzx#!LoEO+ZXw>j+6ouTCnCT80{@1WBZbm=JLzDQj{g
zBjg!2hmg8WCXu~OZf<qDhd3tWnEADj*{UsCkCAzR(HyjlK2;Pw{^MQBc@6p&57AKz
zRp66@5Xml+hZTsA*!*X<PvF=vb&!{S@6Sz;+o9v^QS3++=;;t4_n}#Ky$)4%TQv4i
zPV}S9%Lo|whmHGzIgw2!PhsGs5p*&Eo6Tsy)itDNFH~Y@J*>Oce6;6)M$Njpxxf2*
z{egQf9q#jboHP|nH!RRAiWyyK0fI}*@^E;FkX3Iy6ATT=kWikRxQ3Zc{Y=`6$uQ!`
z9@iy}d!MLnBp-DNrZbIApJ3qY@AbUTQVDf{dQFD<A5>eE-d2{t-pW4d?#ER+x}0J5
z7O15S6`%&xuBBgrA`|vh(1_Qb@@X7m@q0V<d6}<k)FQb~9cq)gjoTqBW;4kx<wl_l
zn>pdz?RZS!)E08TN=QJ~G}XzsivXtysviK0hMfI;2K+k2*~2n7LobLGJB*peXR6?1
z{uKC?JK#dLaXX{B+uh#xc;}F?cs%?esKe>n5*9d+x8Euk?C59urfdKBzR01Ap}*hG
z{{;CF@Be=LI@#o}WcEqt-pdBlL9FPSkHn15ba3T5dU+0%8aes5PK{Md#(lku1cULG
z+DKmD5_8S-2l3TP*F0zPSQ_Hpb;7wq=Nnf3ggMg(L3v@!ew+KbeYi4kO1uN`cnHSo
z#P84|bNq;<x;Rrh$;$_Q4|F~D6mqe&%2ak|%NBg!sCyd|(Dm=P^UU5b^YlPStg2W;
z@IF&=bNaZN!fKXLm_!iD3ksewQ})i`<%B!i7;mnzat-EGnJqhGVR|p29pwT#u|AhM
zYe&$Sv|o`n)pl_4{gHP!v}n5TemH}2Qyc-iU*3#@FVQ<u7h%<iJbXG$mhDuEyIVe}
zd>-EkK0Xr#YWk2fF^u?Wq3RO7_aSNTC&w4Gy?W7LYqE+^4yL0nFnC~`(_K*TxKdfq
z=UFmSbP^M^arioiF6WhE9`kF<1*`t>@ISmZBI%e?;KVy_AtgMSWR@)7(WULkp?7Tk
zH+!|5o4b3vkK3fdsZ0y#;Kz^|m_Y4d{fPz)j`uWurn%7II3jF!eL=l6s9kHevDQ*4
zCESHE%;wS9`JFX>2vihVYvFQ7J7P==6O*&sy_eBQ^PQ^6RGCIw->zKG(Q0oNTLK=;
zc*#<0`UEuzHN*%?E6n2v_&@N~o$nl--p7n`5i69Xw3PDq{BhH1Oc-v8h7q^2vYlz$
zd7&lC^&1y`3xGRh>V3Y~%Nc**6Y`H@3zOxOT>5?7|1=TR!>lrcxRNbn#VDOjpq4IU
zS^qB!QG8RBk<pq7W30&eaf$7#JSX273O}Zl<}<YNricMGDXYbs&x{5M+I>8)p@78&
zA*#7L5IG|hY-e!WeD3Mi`(<V(_kF+a&3BfkTp+_IFg#h$=Y8tI<MKv}H$EM^wIm5v
zwUBhhVasWZ_t~xgWsC95!CtD*QNoB_FvhalWk-c5Jjx^jCpwlj7mXLpIv=Z}nlBU{
zOp6PZ<08oKgcl8ZddR;ydX4bc5gZsz@>B&cE&+%9f^rw;nFYktGjLF%52O6=uXlf(
zvvh2B5AfIdC-%N3xM_oiRkNdbp;1o>cHcKvwVAPuJi%XIVdcyIg?Lh7FDCVVy-v#D
z3hRy4r^MErYgEs36{8pMZqk1+1+^9~p=JNq>K0LRL*1#ui_(h+S}_d~3b}j8m?39~
zSb3Rqt9C|Q=e>A$ondpm8T0WQT_4RV-zdFdoq^UWi5oUq0c8@HqX-784OUc5vJ=1F
z_*`j-^9}Z7u`Bof1@nL6GV+^<yFMS{vszzc8YCL9C4&j4B<)YczN=#X>QL<>4MjKM
zqMry>gfX+fwmGcNaI_GT_U+e11@Rv##!{#Zqu3K)><j2K3;ySKy+yJ^=~EV`O$Mu{
zm?3jK&cv9Y$Q*sf3=J;(-JYShH<GL>3fSPpAU*SVmhHZw?wp(XWEisC9Q1mh=CjAx
zCa&`A{!*TrXWsh?di0*BgGJ9{NNN^rT@(J}l}Vzm@QwsFT75IXV8883f!~Z0#Dlwe
z1do81-!!o451rdPs}cQ0{Se{0exu$cW~8x+3`zP_TJbiYQKx@=u$82aLZm9a+RV@i
zhW&O=>~PYh-(Ak)Kg7B?BVoO(YqK#&Jnirx*K`~knhv5ipJjZjaES=xqT(wWzhVXT
znN#nVgb?!uW28dvFxj7p_4~6mCT$h)JJ&5}eh77eWSJj+q-k(xl5=1?;o%bK@3NTq
zMNZ9qQ**wDQ?F~Mb)qW6%a%b<_m2xbUBol)yg~-I`->JH$d|bWS4eZrkMFds82}1J
ze6cTQt)=IekemFkh^jEcN19SI2K7k01*gNox#D(n?v-xVRpx`;nQAL24e_k>4eH2T
z_GiAK`FxM+1tfvPz^0?(WU#hYA~4N`jPoFtDl6nu5w#rhBavJmndo~u$pIDW*uIaH
zV5N2~zu&(Tgk654+kvOB1(H2Dal{VtbJ`2DRK_xV3=L#90PXT`LIqZzlO4erjH+D9
zDbI9Nz2yoxB4{ExIiAB`Z}|sw4<Gc&C7*vEA-@=W#6+l#5Q?qXVDz8vf(48bl@yfI
zS~D!DQ^|fDn#IOH$HsV)o*(P#@AJ~iPe(<?mZ%v^ZB|`zr|+9Q<ZRZ7CQiF)KO71v
zYEs{<M~8;QmjE+L;(?`=@-e@hhN{VVbJcLVUYE9oC-nZMXC!6daT)aW<I5A`4G4Ly
zP?QhRBJkSGHx5^ray8d<1=tVZz=SU%+&-lYJovl~#Xp6Dk%TM-yZGx0bwS|}17<pj
z%Gu)mP#qlg6twk~V*JsUDJ(l(;T*7=+LH_O*m36EbYYNvqy_T4w44_SyZ)!jxBV9m
ze=xa!MZN{#%PT|%SYUCsS;aw_nelbBmz)bmFJ}o!rcDJ7rYTe<6$cXlk<+7LcAp=q
zNXKMM%drcKPQ5q@T2a%;g#Pm_K~Z6B0KF??1GcT&*h8NG^ttf>+EP<il{I|~d$omr
zSzw_-qfQH_+vME#3t!g17_`o@nn8dy_a{YYPo{v52pQPl4{VV-Pids*<?Ac>X_EVG
zmd6G+JFqk7pc<gK{~KGUmfPp_AO_SP4!B>dM)G3uz^b&*IH1H2%fQp2!!4(=Cf@CP
z%>Zhf(8aC$jM`enPktj*SL=*wDl|RjvHdPjI=ypneqdzg_>_V|v@z@4`5K_tv|2HV
zK+%G{QD_C{DyEa-@YnNWo<Ine6b0#`HC!P&#7fSq%rdnRcz%!s-tMngzczM5<L?<7
zn-c~eAj=M-Sj^9l>DOqf7z0jJgm|Aa^d^jpEG#~7KLrqJ!^11o;PzvIx3bdR=GVu0
zvIC@hwA+Cb7Iw=YExJzKev^7ezC<?iUa)e@TM+f^8kNRxKSqi{T)_m7^@8BrZQsA&
zOrj8ifA!ORpQ3qt0>sb2DNGajTpx3n55?397h<?m{h}9z*F|3pNZqp0{{^}}N_rs8
zZEDkFWg=PJ288CS5t!+6>$C6iki0dTjh)@nk-ge*`c2Aug*0)cS99Do7Tc8#-@@bt
zejbo`f=yNZI)}DBCdXcRk2K+LpEZNjBqob`2C^+K$+q>xx10vM3pRHSTv?q~CVtnM
z7L9*tWQ2N9H;6i3#}HrRHk3r<zFU2&Y%9}HgpfUi|B#|XhqVsL#=4TmjI||J@>ACr
zu5#fG(X_dn4=)uuic`?!LJtE@dqK?xPVb^)PmrDHe6dg#zGE+7i?eCuY{RFm?Wp8v
zc_9Ho8dNB<{NAzLC#}r#ah=s*6os_{9(WEk)eihGA(fm<3wz7;s>=H)jVBdRgbI_n
zC)D}1F<)7C=EqRlPI=Y|?mY<mW}5z~CwXua*5j3HB*XsJU$3c;B1s1ek_qW7KEzoJ
z(cn4mZ8w>k+YOHVjd7}-p6ss^^qw57TYp4yJF@@X3Vw7^SbgAJ*<K5%>26dqjxl4v
zBBuGdRsUUiak;q=bT29kPs#&uZgIgL#vzBz*A>$8$Jqo-c??b*VT{r;FZKgH153$|
z31bGvh7aoxZYKH{K`7hC9)v7>=iJ@?Zll{ILr;yM<8h}({hM5n^y5-v?_<K#FP`>x
zpJ_2X8De^yiH8?;_^O;1h`Pz)l%=Fly449#uu5}c8RNR(+Qhj{_8B`D<yt58Ps&)*
zYmwez>7i9#GG^JK{*kMgo(UoT7mgwkRtUvJCBY3MAt3{&j(h!46ufU&10x*wQGnAK
zf<jwaLeIOilqSDWG^nW&X^m&vN{3-X5!fOEJI?>7y4Jt%Ad?ZrO^?QOAY9WNE8Hq$
zfzt&=Qix#4whn5L6VV{lcq!K~RSC-V>kh>Pb+yH3%#;fV_I*k2g0J;9!I7^rkX{GS
zS=exI4W0saJ0BywE63IY7tO*iiPMXq(~r~1nq1ld)69}wr8fB{7F|4;L&su35@!gf
z!mt#`5|Lt7MfIZG`UoOOgs);zd4C8e7tp(g$s<h`lMy3aYFvkwQnvC*Rnk?I8nM0W
z>j@#rSiapS4Jek?u<*ydA*TY!n&%?GiDh|3Z&8=N*e7a0GfXzvJfRQR+mm!wM`L_S
zUBG=<_No~kyhv&zwA@m_;;;rvmt%VDznUS!X(r^SQ?(V|bdJJ}Kas!a2uEzD80^}O
zE;*P&JgrWwhk&N|3I|V(ht2tzSSd|><z!49F>n%mV7+KG-sH3*^gO?MO4LDi8r^LC
zFdAhgztr_J3P1RgfhP6uJaQrMs(kiPRexyB0D)50npql6Lot$GqY1Fqh!QCYlmoQG
zVs(>Vp!w0?BQ$vL#MJBn3(H;A4uV<jAo`!2Irp$l?Ip2M=|QGOSe1_+&I;e0ms}-v
z>pUN<$R*6oC?^I^MqaSX=$s-om1~^P?wN%3|KpI77E*7jStXROpcJb*c=(KqxLz8k
zXaT;P!Cc+`K7-+NxY(*sEg&q~e}Yk#$HgW<y0rz&A(742jKei{4a{HAaXchswIO<y
zpql8rkxfgF5cG9RAjQ{7bZxXSbO9ll=L`@@Io|G|=EL<6tnBru+MZTYbP$ClkVNoB
zX|2vZGX-@%iNiPk@HeI}!t7NwnQ#A%%MdI5rBTjpX&Kc<0JSn8F7s!5)9PN?q#1s_
zgxJ1Eul~*cp<s#Aoxa0gH24MbX9(F8`EwP`PQQX!2vDfzU5BLonuF65iW(vMI6DhZ
z4MlsB0d^n-%my!OdQcaC__WL5&z=$4cH_kd8I!fg@zvdzgia2h`u4qb>j6A)o}55Y
zV&t!kj;P$Bq)D8Ty#^_(UsskLdzOy*o6Ju0EIP+T)Jm*b$2hp1Wr%5JmlOeLfh}G!
zEw-H&w$B`5wv|&!`1Z9CrKX;x3^l>hn(JS|8YoMS;bTsd>*gc9ob6TS+Yhfd?z_cB
zM_qGsmXlPJP^!+L>PRL}kM`i%Z#iuEl=B36b>6-1i5@nc!8UjT2I+Targv1BVK_{(
zSO#SHVj&eZ>MK^<MK13cvVu}lp@&4Vc2k!jKhfVb*@gBg<M647c7bwJwmZLVq=h}i
zh>kazwNlyw7pG{V(REP843??Y0}xB$W0x48^B$C$bsiGeXPE)IUMC%IE52DhtDY|U
z9D=V5T@OZM9chC{xV1tO+^WY#<;xMPA?Xs0ViAI;BV8`r1mYpgN9p(K9oCR$#8M%t
zm$KAuSpZF)`mJVYwi-O5Fgr!$yfrwbjX02<_%58S96AnNSn=P41x0P!Z+U`(;-7&P
z&AxXpzw<a4e`Cov;yiwFDy+0EV)vZRZ}((u2^ajpdc20HJvA?x;q6AJ=wdSd^N0Q5
zKm6$K{$~iA(4GlwuE{E*t13no+>+m*aFI3wUIN&>+V3(d)HLvsqbbCi($tPQMs~bH
zy#;v+V}=wY;b@l)-gn$&xmxWfL}!8?*@Gy&7LkOah5y*iLr+|9ApbVD;tEEv#t?<f
z{pLnW43b0aOdyIqSE5yr8<Xt%Q9syux~^a2AbQnf0UyaM8J0n>E(Qva9rSCMGVGbM
z{f-qryayHP%3FOWIriQYE}+9x^BQur60fr4i8^Ku7$$EHK7a~QWt&$qmaNy|OOhx;
zgR&xmdMa$<BCkuje)IKqOZT2(78PH4J7H9Q=1`i9+mKZFs~Q^pk4DkV<8)RiHZYP&
z#setV3^?U)!90nrC;g^@34~e2R&Zm`=h>_D^m@&)vlQhY#>RIHNw{L0YRRUqNMy{&
zL;N!Q$Ig?@%WK!K%;cgYtj+yGJ#yjnOYfdf&k^V@=Lc1-UUDZ&@Iwv@^Ljp)&-*)t
zSz}7sG)kk`U_~Jr<Yg|2^LiV$YbX<<zG|@g6_EzRMyvUvtG28*6+YnDx8Om@!d^9D
z_{T&>^*iM%CGBri@e9;P0uL3ko98?D<v<b2=Ctx`_1UeUJ$~9=i(}pATCh{k!G5mq
z1=-nN(TI+;LI+v&Vo=k+C^Qx&E7JmN!r`>FMlnVFvGysa(ntJy+E1=D=E3|*5OW}c
zs%S06wMR`&L$S6|xg&E`vF*5OayrGw@d1JZblgnmIR+HG8n}wlkm`(9KZ5SIx8h-{
zu08w(7cL->WdUObbDg%KvG(V2@5wpOfU#avc3q)fG8Wr7EO43q_Y*-Yv#M?k{b<T#
zg5`@4-gpv9YY8-<Yo-z3Z)NuHtH=dq(_#3;f1;)iYcFT*SwK{jCfRxk1Oa9GP}D4H
zd&!xWm_p*}#qPmH0q5+^^A0M1EBUBUxoA~kw9QHx3<~m&dI=$2T<e?xVkxT+c;ZFh
zEiAZLT!UW%TS2rj01bO0(2+V)r@__s@NQz;T6pVfgcz?PvT~ip0-Nx7F@Sn((B0NW
zT>qcU39lE3E6Qo`v`)BSU?Cjk;YhuP+UIX1{4aYQESc1Wu)x-HnF+SLJPmP8E&bXZ
z(95Oo^y_;}g$<?<5>XU8bi6#QDH5zH>wEyU8oNG%5&d4WVHckLWZW-4MsQ76AUEnA
zrcOv&8jjaagYye+larh(fA8H2(N-pR&M#}F?5qq}Y8Y6|p?Y&O{v=Iqap`q53D?Kj
zWQmIEAotH2?!;+tZ-=R<wvn?B08bwxl1aEkK$&oi9@&A`(Txw<6Xa=0l*28~*+{b9
zKv(;Lm4dW8-$+pl_g8i#IF}E0%C>b#OXvf)?cRE4<Luf5DS8G$h_-W$LX`K`CWdIO
znP~vY7i~MCVlpTKod5RRmbv5p^n)G)BU=yGvZy~$H%T=Hn~y+5(eJx-$H(l;R++~X
zTIzojM*9DH{aX1BZ%4$FutSbd6y_*@X}dx-g&J=J0*i<>IBYRaK8drcNCuzwcUnnn
z8Hu241-YT%U^xByg9E3bGUZ31>#4qHrns_r?PpVk+hPUUm*W%I;wTOn%toS_j37(V
z=6=F(S{e$zQ)}C=o6i}C9roi4HTa!rByB?s9?vPf4>l`~lKJH8QM%a4n04vS4wFEV
z^?YK72G3EmSF`W?F-ox9R+3Dxt~7LDByJ+*n0njv*G2y)7LS{|jL-&RPWOnZOC$MW
zyrh5G^h}gGVxuw$v;_>NJ(81Zk6Fhc3y!4d_n@xMscBAI&iSzMP@@L9ujt`xnB{@G
zghH!l2T{X*92UB^9!u|pN%z|2TzRO<@}@`vVc;=5WbuBu!(cSB!e2Ko*KE4Jq<Tuf
zmrMc%Q`de({{6iWki}>Y7>Op`e?IWi3I>-YMN$49O>5O_Fd}V`NfZL4_gY|mZ>HuJ
z7kdk+WQ?`)P!Tajj!k1+KmhOKeGYC3{pv?lgZGUQ5H@n`Ae50k_WhH6>6czkM%Kg~
z8T=fu{CB&7!}KDxNLB^xl%EVOb+!hrFq(IxGr-$lk>2`W_<b_HwF1pZix@w{X+@m9
zW|tJ{El}{X>SuCV$uiUvyTPlZ!u!EZ>KB|Gv!J>r0?6upXswj3taFnI0u}I2<|Rbx
zS!QL~hT<F|F|tf0#tirYwFqGM42$t$Jv229wluqGpIwi2d&wOD-M+rxFf*gYErb#S
z1e=U3%ZL-VnGB_5J0*vFfXICwW|cMXnrQEH#}?15<gn7Tf`$8zt&y1yBmA`YbLTSt
z$=FeLiE(55DbSs)4~_YN))}!4cCkp-@qhwXyM-K=VxRo-%yiZ;dY@+u+UM2w{kHT@
zrZU7YSfv2pyw#k$&QpMqz*ZEyaI>azAqA8Sv-^519=C~W8mj4)9x7azkOa3#@QG86
z@h}S?cx;=*P(?Hv8YV3jH$i+Al<b(zvLtlIsyb!`Iza$6avC0u(W|w;K3)NR5bWG0
z2{%3NDlO=p%Q9rcjEPgRx9m{)>b}jsig0Bvpb28GM7Rq)KgL`N_zg>L{oZuOU6==D
zRvpG3CG4ya*LnM>Q6mUR?)cw8MBnF3yhHuP462c3;DyATYzUVT;`<hY?)>2aCU$f&
zjWuL^_E^--V`7=&hP^@T*cfhe6wh8kHWyDJoez4M`5CvB(|W4&B0}b~0^>+?yZNwY
zf1mgRKgl9-mRA`GcPOgm(sPSRrxhw#8QD-)&TZMaLe9+IUY4JD5rt@2gk20hen3Tn
zRhHhNg!ew58$qHLX)b6k*4Iw%cRza#!4L9`xCBb?Lx{F*3Ny|eSENC70$4QJhLkgs
zZ3_$bq4K~IlHmKC5gIl8@0FkvC&^VfF@=%}{E72MI>nJ+P~w~Z<7L9ISgMSo{k#s5
zXPuOIyzKAk@O^-<R*UXD*+bQr%7Uj$Y-@o7HKP09eY1VnDMI@t`zTQ%$OH&6U3-S!
z_YM9Q_1h>bkPyKtfugFIJyw|cbXxeh52=!%O|hxMK8IL)dh()2x;zXhl4M5FQ92j|
zg5;gH`PjQCgew)ZQKWgc3Y5g0H+QJ(;cZ|c2maVrb!_gtISz*;fet<00I)Q}@Uod_
z#J`D&3rli}m=XiWO91W8r?^w-mYaF_i)*WrqEu#@jeTh6S8UP$ta{y`cts`XG1rU(
z7>6@9@1!K|cyOe{=J3ZhZr1-Nz5WLxq2irlXY%Cq8D_{ku$`G+M$QGhbQBekOBL1-
zmT*RJ*jnD+BtX|qQnq@;*7yRq!KkhPdTw3~M%j?JToUg6jN~cz6LI;ZoJK+vV;Z;}
z41pmJp_!S^So}ZM18*@bWhMH}D~q?+R}^!v(h``iY$lX(W94ARm7i&}S|bXGL9!Kq
z2u*5Cbj_GS2ISYjx8*tZojF`Mf`6Sq+)*aGbDUUhczYf*mY*?)O@@zZeI-I9I3o)n
z996WV<K6~F^j)V-mxeo^h~*UcP8n)ExBRDRbEwLUb(gSZA9s?$3Di7Q>;+iTkVVcR
zz;PsFIrvDr^DHBNKe1D&L=idCZV_zG!FiYTssIbROVC?@TwpS#MQS8uq~#4bYS2Vl
z2bYbVaFWd9T_{u}??!?=qu==>r|Akl)j=>kdZvHzHo8X1ma*2Qx%9UyX|cM-SI)yl
zU_R#V<0G3zG}k;m8%jr$xl;k_S9yiLGLpQ5H`=c6rQ^TS6}j&tFxC5crLB-cLLy=a
ze|r9io;eDm+WO?%4YcbLs_}GU_MLdao*y}ycQraeWcAz_G4;f{67DKUb6V-I2Rywq
z(%|&z>@Z_S@xP8M^VVax?@8#;rYJ>E%7X#;!}PbE2d`?LPVa`?tn@98&@?O}_$eU_
zDUzYUlV`H&&gYl`S%|ts@DLUA0p_xN67&MASp?M2s2bpyp=#mxr{kNHI^SFVoQFX3
zl|4|F$o)FYK^UbJdb0bN<5|-$KIfV$Ia-_?10fivM57Z1jhX*r?Ht(a>b_{-q>atS
zwwe>$_K9sawv)y>Ik9atwr!)ao20SroA>|R`wi|VSbINv?Y-xkV~pR_<hN(0{Ux&V
zGn4M5MNCXOy((zpSu5nR!VF`P07bDyia*Ou%sET~10ar_H(Vk_3RNrP^Ad#ev~b~v
z;)<mj7{jPxkw5R3rUR=JHjwK3@z;ql6t<lg1AAIg(-lD|0{t^Ne+zhUS(=S4&11q7
zv5BbN5AmEhL^=RgzlS@l3;z9+QviER7BpeiNDhgxg<&!gonJms?RUZ7^>T=|xIiI2
zum_Z{nC}@wAfw`>ZO0?m0`}na_jdgCrXKh4Vlk}1H{og}j<y7ksx=K24xol!(HTq_
zA%4o=pMu9{S)u7(^oXZb=<Ia@2gf85vz3X?5fDqZ=`8{Z9;?@7^f2j?z3jU&ihxOW
z>@Tb+{lgLrLC;PXNL3>(bcR#&-CS+b4vzR4N`hcW69g}TF$75=2P%bcEW4@|6*Q5U
zV=suA`jTaLxQvQ~NP6YMb>P!tT1Iq$2sFaw9l(3G>-~;wfAbcFHC|)RY9m#g8z=M!
zK%}zgo$@7WFxESWCN8Vb$c4mM2wsLZuF!*Y*d^bvhHG~Q)hW{2ZejB7*jcpmhsp3p
zC*4UyVe?nC7FfpL=YXb=MhQEkLzOQOoHjHG2eQo<H|NHqUYd2+giQ>zA|yr3cu3-|
zY_LC>p_v2O<I*ssM8QX?k=l&Ou^Ap^#_`)W0fXD@OU=53<pSe2k-uC+xAT_3!W!Rx
z-vXaJxQ3n()K^WuMfRadHHiJK*>Tgl4C!3+@%I(R=MbY$HH8zdh367hA(LEkYL*4?
z*!@ggZidA+ftPVa<r~$;tRXW*RA-=#E?d^nf@Eo$?8qbe_nAHR4l^Lt_~9ZIkqC)C
z3L!+DIp>J`RU&xzrl*cx)0JNxN#)?sC*Or~7KSXYG_H)%0f*WAHi|mFmbutIKMvKa
zZwCGK7a&rW*%j>mul*E@s^lVa^hedNpLpL@e1VUpoNMj+gC6L8L-r)mF9!4tf4LX3
zg)vn$J2)v&PDq&nQ0QbCdE~`CU%tsF77s)W;_+^FB!0ywE2T;8U}cOVp!trdkPAo_
zba)YPKY`|io>L=NPmB#+clXg%uh!Mzteor8F6fZpa73bg*-5}iW^w5K#X}k?9Atun
z-WhY6#GcVKIGi0z76hA)*vMGRmD%AiXb4~T=zssP<_MA*faOv<=TPyxzSl#p5Opx|
z-uxB}6fu`WdknTdmmp7Er`8A6+~cd7Aot*FX7u1Pe*H||sC{Ra)<nVgx&0&F8}dLC
zA&fx^hrtS-)&?!FZz=0`#l=VQGRhbsotZ~`2|=T)^>Ov;={J@Uo<eLeCh-zEM%&||
z-S$GZUtaPXAOG)Ai;5H%F@9cPr;Et&E1(Wj_kQkjXE5}DI`nS4Y%E^qC2YOZJExq^
zQWS>9_g0gdD`l-te-Lp1N42L+SbQZ|EE4~|SVV7Myx0kx!{-(BBgMUT?{C|jHtg4Q
zG6_VDq@JZEzl1CC!7uY|#)Q7ef$>2b#i++)zvxZ`E;}VtO6`{)dm8Mx_;~(0e4r9T
z3WyuSQ<%{FF%)Aa<BrA%!b;%Zk~qX#thCVPrD`0s((mc=di<5M@`R3sEQL9gmq#8#
z+AQ?eYu#;xNbw|P^v$tP`>XMdteMi-oKlNVefAFyKEF*gXi3aeb|m+UO$(m18BfKz
zkyyS(ZHB<4&qzjD3&(J-m=b+O2YQEvg)$I}<Qtv!<I0--P0_Itm>iA^AhwkcAp3^_
zeKbfS=!=XYC3X4-%N9|L6|5*!p|oTUlI(fN$s#W+yx_-7?$6>E_}C_$;ts`u9Ss5`
z7O({1&X-k!ow>5<$U@~$%4E$pxhlNitsmb|x&{>Gx)M)>D2%b#?hE&-cgcQ$<9tG8
zUOwxu|7z#*a^q$jSkk|eXG8^ts%Z;Q^+UiEsPBH0SNoAS2-8%s3c>Ng&7tyR!h?lc
z*dpxtw7Jr=>!pwd+*tDw`$3Y5Tn`Ax5f<9i6)HsMgqQd4IvrYm$bWvZYLXOZ7h35W
z3cie&N-U=-FvI(XbLXN+DgU%c;{dNv;CHwy{#dagD2bBwGI3Zmgzt|`b&kLc^r~6;
z0_&Lg1(=xl-3Cozdg^IPl2}C%vYtoz`{n8;QO=t0r{l7_uYwMkQDY2eMS`o36I(mA
zUvHRYviau`=Sr)ch;y@CYu^%YxPs}(c>878`qj|W%@aZ`1S_|uf~~#jzf>atdm)h%
z0ytY5UT9r#c^($y;p91H6Q9Z{W0mw{t8+T&n0=Gm2j7~kWu2V`wP%EDNb$XOhamqm
z$7)<rfIxI{Lx2}D2VY-o%0pYK*ITe3S+6D*czNp+@LlrIdlmDFwmr22TJ_X}Cps1#
zukK?e+<0WwK4zM!G<~2Uf-U5Fs+UkdXX{|+l^y=Ki9|A~)Y#nt4;@^`0D+yS=YP#v
zvZn2d6C8><>2!JvC=kXN#YDMY1uYz&M*~-Z0fHhEOjrA-P4OA`6OY0N?Q;Y6+;?^^
z=*VD|iuBy6-4p3I>VI;Yth;H@lhhI`#^{gfeCO6ir@jKM{u4mUk0W^+Y+53GzTS0G
zQTYohLWxt`noU@y7&TG|P(~MbI{njLsttY%r9SxHQuTM%a~LKAE9OVc$|KN8GznGY
zyqFpqO;!vT%`ITa_z3pXbgso)uHXzuRBhaedO^{I-p}Ftz*_VR0qj<3g_H)y!3XVf
z{@~-pH#0&=)<Y(;SH1dKwQ(UXBX<_OD*t8YZGk&uK}eH(v)ks$!?!nKd>&AhFw7Mk
z4&aD!#>`WlWT%tBgtKN^YJIG-UBVPMR8R5Q*+%Me_nheFd%>0_U;H;H_!As}TNZnp
zVepV~I#XF&`0d117SkxpiGdNK2;wB<%^XE-oW*04zpGG6<ypMfKGWcG<Ayr3G3rF;
z*yLN>X<y&o08)ZtvGKRiY>I&Ksd{~Tr88Rm{>}{@K3weA*CWACV-(`kEnWgDNsQtK
z3I(){B8i|)lkly^$V}7isKtu4Q&(rayL&vmGTl)aQY@Fk0b&OKSnDNz$*1ch%h+1>
zSl=|glx`7I2y7<y8&<}O^rz{HdC^>z;zWk-pmp?)j^gIQh2RI4pYKy}TP9E|Y)On{
zDV9s+%g&G%uSQ{NV^qat@42gT`7EP_N(4K`Gg4vfVbOm7ejp65yH=yD-`O5NcGq-m
zG=o|arG#;EWerZpC{q+<5d%TgUA!sQ7G*_HL_ofa&+enI-%+RgliveK`Dbwz6*H<b
z+3_!V<h<<!Co0ZCl&=Ej<b;?URu$5~-a>eK4D}3bRc@s3d<Muq=ay5=(>XVt^)ZLH
z;t&P#2skG=^(Q00n$tHTsiL6tE3m_1^}eSmT#xqgjBh<n2_Ed;ID@sAXK9(VNSG;N
zgqf}4Rdgx+iXna~tB*qoljG!#KXq_>l45c3QZ`^;kjw0CM6q^txc6)vCHi~a6Dh8$
zNED^=g#qG~N74eyJt9cpek&y=51LEI2SfJ=k#EC~cYC}Na-0OjyvWMTGfOAuW~1@}
zA@dUmWJ!JgvE4BKnEvlm7q9>H(>eWviZ(<@J|v0zdj<rt6RI~QKj|6E4T9<XoI8<5
zO6xBhhGSMT)4N=;xlGJ*Jm$rjhk&UA#A1q8DC=fmxy<9@JPK^@UmK*KsII|@YC2d^
zNtxPDqjK394ht@H@n4VmWn@}Eq;1`-U_Q$WXsLM3GD9<wG=-Il?}$^%JhBfFk;XxW
za_2eu@kR7b0a+!4d;zbGGDt>We#Dg=1s^hsSJVf%aV$**3yI+>$AIjTrrYktCQanN
z<fQgTxF(v%jJxGNpov|1tg5QCLC#ldHCfZhX0m?!AnRQ;tc$Ps_d=G>MUO~ZGkS*N
zj3mv1*cs@x`49tlT$3k;gyV{6Bq|qC8X5O4j<EKd<R)5sl}>-Y<()ogPd7X>lbL)V
zQhW_Sq{EHyb-^@c{~!1GV4zUGgBHZovpe<VaW6|GU-~ZqoNPY6j{=ueq_YSPrahzu
z!)1Ak>OUW3>j*fX$R~YD|EynwYUqm=v1=>WPX9Zk6l#Qc3XMVO2*62D*@Oz=L(iqA
zFL+NIx}{!<)3pv3TFoRf=PW}WrphPIsW~V+P${QNDNW*va`T7bBzB=7z6sSuSd!4L
z;NT{Bt5{Q%m7wS`xwH#^?^AT@#3@;qg@noxDx~_>Urc=HN`f0gk?Cq3`PO`Sj(l@X
zGVCZ2_f|ynlhGI+Jx~^!UIzG0`I7sq(ZqmsxkrvAC%*_qxg;ESHrbPyn_dJH+5$5Z
zpN5uZjw?v_hps??naq+ThhL8@*fBP-9<>am8*~tx<rRiXo}+u9`o#jc8$OoGmYcYU
zVbZhgLx(!5OyNt6`wS8O?T!LTD}Km<@htAdn=jUng!RaL@%ra^75Y{knM_1+y!_T|
zQL^QA+>X<7P5KmEdo_7o7%MnYSwWb{z5-9!23yi)FBs(jazG&Eq3{(!tsefye4~Ii
zc1UC#Qw5A0Cy^&EEJs1C<M(@X<+}fq0jEzYxQxv){6P@J$?%dbhV!P?I&!8HWw0|k
zNI7-J>W<_nHK0f{6akx@i60|W2OajWsiizo$Se<43b2Hu+ZkM1(gP|X8!bV)W)kGB
ze_g=x8)801@vAqa(R>czKr)38g+eUZ39c4=H{;l>RXGD1JB3_R3)l|%ht8ssM54pe
zD&dNnh0N_Sl=4YXT1`37+k)u}fpY}7{tu4_RJLCBEE2u(@%t<;j{t<z<KwN3SDr23
zYkp?lQ!EvU(RFLpJDzPqHZTB^k|L&588?qZ3C9UZT-g{AHZn=f$hm?p3IQ6Yh#75?
zcI5~D_s2boaKj>FLiFD=0l_{$(AxgoAaq}`XX{m2Msll7h#=&=LP&hG)~#}wrnScT
zN(wDQu1Ynm0n(i$UGa$gYvUKlw$I%`lwj`8s4+c=qJNW24F7q8qc6O8c_q%#6ynQ{
zRj1kg>s6_R<H_eU(kVW;4PCB^-e<q4f&{4BWWZrw>O`%GpdiG-r2BexT=+GSCY>h}
z=}?K8rUHT^l%w(nMeR+@Amnw;XQaJXarfEVPlYtU4*shQ45EhU_);}vOm^q^*%Mmr
zn$iQ-_qJDgLFYy-pwy^@HwnOSMz!YXYtrP&a;i@#OgheWuF4>4uINlsjk%-mDr|xJ
zAL`cZ3`L_y#(&Ict%!^f0w3$YPnF3I-f$Ijx`%T~vWz?O`nw-1s5pLOb%tKGeffSE
zfk+DEnMpO9h0Y}zZ<YFT!6*P1Aaxm+Jb8!pre5!gAw2Q9{q%n+E)$t0IN_UF-fRv3
zt`-yUcopE)Mt_SR;`HeLH%M)?QH3jKm~c&?jfjqs(Ys4g0*kWijQ5M`o16_=<)Nq>
zRA_Xad6K=YO!q!)`+fag<;wqUBN;p*_y-8Sm3-VrIp>OyLX~@lA?<B!cwoB!ZQ<n3
zciHycE7YlL^$HjbSQ|@P$%scE6|%aC#2C^JIILIXqom2~#Z5Uu2t+N|6CcHXoIyJr
zu6RDVJ_RV*xA-1Rm4Gj@+&6cAybpGJ6!FCSF2$R*_JZ_t#Pm(;zrK`aI(v61uDLge
zZhLT1kKG-M*;-csmI&`aqqzK}$f_aY1?VCaKlQ`m4;+T#^HGO4YFZq9SRD3AS&hSY
z%Me0x-pg{@W(nR`1YayXcPBXw2*%l{sl!(RI!8ZUCwX1y`+`Q%658@@y_sIXQ65hb
zWb%2mKjVjhV7{Hq@#yMs)hTq*F|v5LOstAVu2Db$P90k(jwmtb6DBX6*ysYz@WB!N
zr@K;R*!^IN@O{qUS+;=Jrb7Y^#3RNC9Lfg7OMGPH0YYKgn`beP+6El`!6`e>TIF(a
zSfMH$azn6FI6kMeS;D9y;la1Y$OtCVqo@VwLBB?=mi2jKYoF4Xr$}?hGVHq?3%J*>
zuSet&ir+uNx17Ije!?Ix@VZL~EY@)Tf3PC4(Yr)8-`*cI88uAWSAf8;Rg}}!4N0fV
zSjZ68pRPQ030s<w)Vf#k^OG!{c5So@_Fz;XrUe5qiR=D>*`bKbeFKG@z6%c?Af@~V
zM_HyKRDiQ=Y%2t|<7sg%QSELGY+u{5_vXH}n}Fx1!JGUTaW$$&&NwO(5$l?riDCSw
zc2%>w{R0l#B)CsIo(>=}04sq=TC^^NkRjVd`Mp~2(}8D;T?Qi0!yKx$Tst}SB^UWK
z@ZL-6e!>6vhui1&N-{G$Q-7Lt7suZLc~a8|snwDRbmLHDg*D6+<K#<d&Jgc1fD=w@
z6ulp^9chT>WZ%WtMKA3l9<j-4U;LQpomde@WIg%FTZcGejUqT~Bk*B6y!1fWeRZ-m
z%INRE%`LyTD!j187iQdYj(S*usxl^?Bt85zNg3jH`U0s{q%oAuF0db{0D!)UK#p6Q
zE@Kbgb{tY#4kU%{h<UP*#a@?7|3nt@`P@`0X7zs-I8CvaSl1TscY;%bk3ZfasyGnS
z9pGdp(Fv?a2Rlc5CPeg-!vnHbf~l!x?Jm?yv7U1?t9huZa%mC;DDM9`|5NMjokCVr
zR6Hy=SbY6u)^3E$=L%$%giA*Su&G7oC(8Hw&0pS|=Inj7Mn5r|7JNPQKAphNjDNA?
z8aH^+@y}`Qx*Xhc^YMS^e$;tg`xwf}V?Yv4#7Ro$t6E6%FhP$LNz)2qL?~7h<~0rI
zX>UHrtmbeT-EI$%-d7@9R6AApX?J!$hw}B4>rkvFdPb+fAq#EMqIMqpj0K94)8hJr
z7a6e={Ui{!c}>x@{`=wa$=*LhRX9MoHGX_D(*MB!^5E&=W6t3H5BKZCdx?L*$F~0i
zvC>i~gm}I({bvP5VP?YZI>Qyl)vrH)xo12$RhXNGS<&8;e}^fgh8HMdKAEVc=G4$U
zJvzvJ;@AIf8rF2Nf+Gwv+e&)O>kn5-QByaml664Chw=M+f8g2FT8Ib#{`<i%>#3m+
zvWQ;O9sm8oe_wDX<~_pyteJ4D`s2Cz$wcr5BUqpr<`NDMD#4IkOF>-a{D!?buu^7h
zP9%80e7|>tJLNYRqThU;wQ_9zL3hVcz{FpmKm79XI(Ux9Z9MeggnhQRA$=#5h#Weq
zs7mHby60p<{cs2O+OFjV^3QFJt{2kHZ*&&4(M5zjejn3sfcI(tJNxx*`Pb^CEtQbq
zHH=R#oJw-i4Ce44+mCWj|73T!xQ+yQJmx(@HxPt1wb-bCN?^Pkg+#(yiau>xaj+uV
zf*5>UP*&(yR*Oa=*e-ncM$fQRHlSrdbi^3K{<*-}-%p7~3NW56s~o=vwM(V0sTM)Y
z=_|8q@we2-qX;EQN%<RlT%m|+7z*0%wYOz-u7LYKKIc4Fg1`5F!PFTknK&&y)*KBV
z4?hW+3~NRv_Q)6ra%43lK{eH%sSAC@YcuYIEs!Qiq~aoxTC?||CUWH_)FTY8M3Rh5
z5d~hYvDF(Ah#+Si$0kU0b*BqVEYd;XA+VlFT`n$GM1UKw|GR;APuO>AhIS5rDyl!J
z3~OMr<J7jm>Qlj<YMvUI%de#00<ptI<RW7ItbHqp#c<dss|sI-4{pmP$JN@CaY%6o
zEq}q1f~nEaKI?_8#|t1DPh3uWA*!#~fYZSuns)EaIF{K~oH5yo52FAy2rz)@@p67b
zKoKQormn7PtYsBL$m=ka3hR}CX7`j;^PjcDvD&zDG~a{NNZBnw)95rAX2|r(4iE^>
zCLiI5jJ~|i$J=gxEi6c%fY++Ohh@~ndQj5+%rUK)eYII&!jvNFCLX@W$z;YEj3x=3
za8L_D8$Y*6$?>|e6=>n2`bpx>-Tg!*JEASg!HX?~WqtkVVk1gN)x6*7B;3-cgiRT#
z)CP$wavuYmdyi3I{3jhfvi#q0*E90jF=$G%gV-{RzU+ejVPUxvNLWm_;!2reH)stb
z$$(QQRftz}KSbUU-@`JrjToF7o0t)#b$vu00|fjFp|7h#&{Z?00AfO$en}`=s?Idj
zcw?x_<f<0*$(klSos(1~1<$y*Du&_aL<2v@jZ<yDS^aB3Q+Jd%V_e*W;w07De}E2}
zfS+B#(KNZ4?LVm9vtT=U!~5Ko8wDajO}W|JpQVw3MjZ-bKtQ;q-3onZYMqhq$@_`V
z`xHL`WEt60wq0ZI^64rjuAUDWpCz)@YWNQZ{4KO+2**gx(N2e|ZqIR)gimi&lJPZo
zGmv#9-D0-A!+Nso<F~o^%J5lDVo1T+*W-GdW=>jJcYM)!g65Hna`4<5SVt0O203vP
zxRzdLNdpqUlc7+o6KldLOpCARhc1;L)nn}?&AB|<IvvCeHYL$w=6IOvYWfPD4{8Vc
z0`%!cj+}hRJZ20dARQAaHR4X@V`INhCZabZKhUV;UWOL@1TRF+C8q+X2+_o#?v&J6
zBS*i5^+#N`X6X=MBRPWTBmA&?pVukH88s6Y9M5ntq^N#WX*kAqSZRRR|ELKQ8=oh|
zk2rJ5tWl3nwpYnDma}AWtY@%G@j58US;>Vox(Wyn`HW4Z*pLXVOwgW0!Vh;;JRXG^
z<f6(lKOJoRiJ|8%{Q^B;sXlJT<`|E(=PJ1gRN83a3;P)&KqZDGoyf)U#XB!Ir&=UI
zc?k;z`N1*oU`C#c=4VABS6Nk%tg+mCk+hVro7wq<I}KocrG0Q?ML*eVNs)tS`MTfq
z^GHJ3=UPBWABZ26u#{2qyY%$2>G!{jal%bVz`yOig)p{30@Wk}RV3Ot<$NXts^0pN
zxM_A3B6W;qkNl<YO98nBQRP*H3WQ<vdO}A_jAE0W!-G4@0qcLlQ@^)MjJLGtM+0#;
zY<ES9JVCC&ZBdFLF}Ym@AH=Tm(36j98i2TrM$HL>^VXFY|IXx5W`5M8zVM_+9m~qE
zQe*h2+GHzkBX#ydc@$9sMGWZQQ2)m4uW1im9dw7mu|GH<oUr7(@pjdEaPt@l-nOqu
z+a30fzothAfKY^`B2v-4?}D@TJeAZt*w>7vxs37A8kcUeUQ9s{$D`kL7XB3X8Kt^z
z&n;Xu<1h++Hg#Qn1e^>B7_N<x<^AkPruP@X2JH0w!#Hs1mw<;eCGwRK%MRp!zfG6^
z6qcu1ZySAH@hC{Aio}-Ba0W-pich{g);OO5KXMdN-Z1`hb&9B(omZEgi}|NL-~aRI
z4$N-E8H>yx63jU!xHg{r{YhN3G`CAD{)fhKU0$1z5e%&?r!}YTR9~8y_*o3V+~Bq^
zD#Pr6GnnhyH=soX+d98U4PY+Y7h?jK%jgjyXc3?j3j=`(^+tB6<WkbJuy~1>^J?}L
znUe(?OcjbCl*xx)A=qrwVYe3P`hWz!d$#8bH5??V(X3HWe|cf#SoWV&RiwPIE;Tmu
zltgf)Y8FPZD%)bM;Fg(%y#hgfhANW!1k$95-9$+Iu9CC?D2@Q7HJ$Zo;0zr~nF0+i
ze)o%dHOs&p4#vD=@8rYYqE8BM0<E~zM4(BtWvr#;5?a0|8mDYTF4_TN2rYN!Sq*E3
zNi4%I3J!z9B#I05Yb9{fS!!g(4>WxgRejLPszIx5@;ZrUbSZ!jyKuUQA4FA7YLOkC
zK&<@NlH;J)RDzpgOxTj9%yx37BoiWZ6k76zpd7Y}rffXaYBpCd%6`pfP9><<$d0+|
z{}L&UCA;OXQA2P07V=hH&8l?2hSoNV4gBmwAp(;ABq<-3=u|jjqoEc5+j1fgpHOGo
zOjkOfYzXID7#~@jK8HfuUiD>m0M)B)pVCZ*;PkwnGe?EZ^}O4NK>w!do$4HdG>3~#
z1U53}NBmdxJ<Q49Ns8BYK(@S11N?=NUqj*2t#@U)mU3cFkRp-H8FF0da;Z3yl;A(&
zGL_3)D<uyCFxR!el)n^?sD`Zaz{b7xi#{j1x3?y2QAZz_fE+>dL$LiG=}JS(zy1N}
zK-^W(lt#eNu3?bgod`&ZfD_Ur^V6;Mrvryq!^v8ps5W^H8x?aNE{wNX39Ey6D2Gl8
zUdYs#(X~$1*s?tGsh5fNTQFqK?a&p(>?=WN?12Lgui=m*sIX%+LZW&XNTEcGnlp^W
z9zCQsB(68K?fLE}o~jzErKYS(i5+xgXi=xAwKx#l{iB@RY9wGyggbqLX3){v!rq34
z@*BX!sYgD`MaPf&e>ro!QW;BlQqOQB8;`<372!RM_w_4ZSQ=4~=j@I+^1E8KbJa27
zP~W$pL5d<;26^^820CRHRHOv2bR>Daujvwp?MMQ*>zgYU{&mAyL2HP#|3CmBnM}*h
z`+B$%SE@QyyJ82r@n#x6*1G0$9OtoToD`&QWN_(HZKh98+fTs^MV)`o>&6boshn79
z+(mT9HmIG-t<YG!c&LQ+>5wOGqlI)H9tQ#mP9lPcx1@WaV&;e}TMlov&@wO>^dro6
z(8e5(3eJ^5POg8)1lhvpO2lAWB|Y6l{~`4x+0!gF>YjZHplyY;_~lNJnt#8D6_pp?
z(t__uHFgmTXINcB3p&4M0+d?TF^_rd99bX0Qg+%4s-a{7$WPpAggNuSm*cy*5~(0^
z>Rte$HilJlgGp()VLoq-K`Wr*s?y)ILtJb)uiLc0xqWXvP1BVMDA0Y|&CB~5XlSC%
zKDNIIHZ@1HK+n2<vTnOqUg~jl>QB2cUe0zhrU=m7yEyHUleDzt76-`0BhhszWgush
zDckSnP&hz-mOYm35P~Cq%r@?#%xy>*p`<-EGnS(@HnQq#{P?l>$&BokS`<;n5o$?l
zS`$}bM-zd0&NHJYBY(69Y<V>a%D6TTp7)UQwlq%`h12@!<|TdWx22B{&)vw7J9Hz~
zK}+)%_MGcue0$@lz>D{LZ$3XSF*{8Cz5MTFqG>&Q*dSc!Gn;EMjR<tB(7;$A9(e?G
z<Q#(p21^2(s8?yCr7`9Vc5fhjqUVrxJLfV-$wV=wczI$y=cbwe8zY4*H?npK=nx$s
zMoTjsx#2lD?e`?$&1rvqcm%Qac8_w^Tk_uf6g4eNI+V*Iue(sDYJNKS%e2j;+{p6G
zbI00L79KE=8<)q1A?oC`)FvbsyzFdUD~X7-J6Dqm8V&lp8Qt#V)bxqrkOQIxq24hc
zdcffnq{?A6yC3+RWmV6jc|3R%Uyo8Nj<`Be{OJz!Fsw(kW$e1MLh;2J^ZMQlaf|vZ
zzficB&XNI_hy)&;JnJMvc}6-JN1v3r4yvBTL{$rWFm=>r%Byofs6Qs0ik;<~^yIB<
zg(58tooE?W_hkrYtLF<c3X$Nhd)FIL-bI8ArT_Cz<0z$aLDXq}bHV|x-ifl7UCa7s
z3JNVEg0hYoDj9?WL)RvBi&|8ufxC)FQuq$fJPU6fyfF(Z2Iy{Kblu$c&JfGqh^K7y
z)BCvTh8(7Z+rprLzrOs=_TTTm+uEysHTa<W`1f(SCFNFueJwYl_VZNR<O&EYHe}`q
z6pjQSV-^RXdn*(SAW{ZOrJ3RUkdiWE#tnT=UqlNUOov~RwA0{y-D&QUOFp<&IMfS6
zH#&>Iqt`B=Ky&(WWAh9<t={*dXkFFqmSynv@S*p9z4bW0?*FEro;sb^AZoIRGh-D#
zg^g`JkHno+21ypgNFOnm_VbZeN})UuJv!LvU$E(ub~B-EMlpar*^~PpO6N1B-GweS
ztc!1qT4*IgOJ1u@qIAwK<wK3@<)hr=qv3eYvvc=^!4t|m{Kt*KTQf5UP_15WM1E`t
z#4c8J2zy9}U%FHVlPu$Qq+UHEDk^Yh6lVcqkv+vQVLCP?KPi}TBce1MP+rdDyjb72
z(D%+@nXVc86|>&y7n9INTFO>gJ>BY%LSr%rJABdu1H-?+8T@%~?|SbjA6EwZ2Yy#4
zw?v-|al{H4w@&y&N_1I-Rtc+1vhbLLGAg=2WjuPkprN1gq^LSE@lwpR7!>ff^Sd3=
zH}}S+*O}Gb?^N#g8-)z4@YK3Fr$AH6l8t0JwsgdhR+qGM_#8qF+!0x^-TBGq9iz$K
z9l*jNlYPgHIF-qyO^FEUX%2UwB4XmkAOlb^sn3%4TZ}5f%yq|UJY<B#?{KNyGi$L7
zinll@<YT1H8@Ab9W0ZkB-9m>F!ZRAlj9=i;rjsMD6fZ~syRh1^yFZ=Jw;wf*ef@73
zx29u_8#?PP_2av5VH5u-T?x4~U5}FWFqbE0xFnO!XzSt8#OXX~fnsq7oSH>waKX#u
z>eLs`_#TV4WqP?$*#h@M5c!%vmgMfw{~l*koy=;}B+j_#Df34}k{pYa@3L5muJ#8t
z?zY@7U)k!v%<+CMGDg9jyqi>TPLNw#9<NU(Oy9sB;YT0AQz7Pfj^sa85jgN?-VDrw
zSNU~(rVdHuGuk@y%gGX|>cMoA*^xW?Hjx9}BpOF{_jOSGt2BtqHG&@zG8*IgOU%8-
zt=iz!ruFAGL{5k^zE}<Kis9q7g&3(P$kI9P>NmxWa~Yleda~nliON3HC%p$j%Vx|5
z;K^pUR`4lw1lwwm*W#Nxut;bPZWsqt>!=pN3Ml=A(g!}!Myb+G5*+>J7&!p{M%<$m
z(ZSJr?8_HgLYm1_n>N*DCd2|sa(mjrp=-%ey|&uFJ8rgmqwEs2vg6DNr>E~9Splb~
zVU0m@lH^nBNW^@+Vp@ZeVPQTUd_!<9zOss8GZ=7USBbR)VxssjY!D12!j1w=EDl9X
ztwOy&TR2D4#7WuM0XAuJPQ?X*qcj$LOv+9{f4}=Z)4z^Qe?TsZI~4%(=PT7uRJR1}
zTM}gCgvk=>3atn)_`O6ZF1H_pQIK<HhlMV@GLM0)B@kR+yP|0<EWITsFtx2I3O%E_
z5)JiFxqXk(kudQ#b@y84F><Lx8Rw;YeUw>n5BAIy!kL+5-cLEZM^p4YLuTAChm0wb
z2$B}RUsQ%p38@}U?{;jz#AMgl;b#*N-kc|tFuY*i;&($~MchU#o`$$>kn4YfJ0bGi
z`8T2R{S^#bZUyK@5GE_*%&-YF_Gy6}Aeljzc-Ay^kc8Q>EbSz(eb=b}TSj+ybb#*f
z1_8`f#LnCui{cBb<Fe>6G-6_+o5xd-<TTjB!KJUwP3jj%#VilvX(yZTmOX#h-E9`R
zShO=?uO|g8^jkvZ1-w6rt{KS>v;>i>WAKi(d)3uu^VYuo$K(h3)^pgFND5S?b1DuR
zaBr+eiIVtz{=wtz9yFh&2W}#?J%kfYqa*B1mm-lW)aWZMUCv0H7z5=K+&pgo>QeW)
zd;k741*7Od`lXDNcyK-peB}aO6j_3);K2EfmV{<3%fa^VN7}DgtHC>f&)YkbSNp#G
z+rbB}8L(E%-a$x((_{3@a@mj#5kc(<fkEdB{Fe9aw+5Nh2Gktl3P>`v>MtfO{<}<n
z2~DA3s4Z+b0)BMgi~BuZL4?Ui;g(JUym+=)2-OaAm8*lo8#w@5(9el@+MOw8ztmg*
zP9b((J+p6q4yQR-dcQX~fa!h`Rn4An)KLYM-IidrKZW;yjz5XU$58mkn`!ICC&2Dc
zsUQ0_UxtOCQ++*ka-;w{!Le?|!KLPJ^jh`J<x2X(D6W<q6z5iZ0M`!9H30?^e>R;8
zPKv|_q{w|i<*8zbx`S<nhc&>Y5dU*t%-ToD$<oEu-VR0}kQOz8iD9=+gwPglML^K`
zfoSXL5m(0Qc@vEyKQR|TOCJ{wbDeLf9vx0va|Zc)KJ>^`;gWC-+qk-x97-6Yv)_VN
z3`cKf0Xg9SqzaCVqgfF`QaEj-$G$j-`TgY@9Uw7c+=Ce!R1)*J;QykrlJ*1E^1hgo
z6I<oJ`y-g_$0*Nu(}`n!dIqn_;;hP2VBNI6TOW5P%hlU>g{g0oS==-?4^JN^f)}w?
zBQ;I|tt%KkN*yn(7fO+v>5vD8D6+zo9BHgjf1uUr)GTM5AQhJ83e$6F>yiG0R3s(^
zZdLPqBlGI8P=a=F0&E$>zveSId4iP#a%uu>d(ZI&E5hh&?=ny4BAvKKUxO)`GfW@#
zxvW!6C0JRo6k<x{#1n;3mDJ-7u<o>EcYYm1VW-Nd%%{+v@okCHkmX`gOVaz@ZM{c<
zY=tpJv#nf;ukfe2rg1|+3-zH{@cNxyxmS|6oxgxTYR)jpq`{(c$~FAlT2AuvTG9$t
z^vLOy+qCuZo&aH+ARw*+fiS`R6)Yj9Jce@jn5CFieq?_|zHZyN2^NSl-*Imx@MEv~
zP4LiHG>OpxQrOE2n9Ejg9LRgmaF?9%FvORcXydEsfO=O`yo2tL4X_4gek$2ak}j@_
zYGByEhX|r+77XRwS$5&~uijZ=f>D~J$X>fME8mJ3(d3nO{u;GJ66Xk3$!PvSqY~NZ
zZMrGl_I$lXgo~8cia&|xZT}?p`u3HU$6AJyk~c&&;hD(&bt&DMkrs6jr6>H&=DJbG
zx4~SGwPfMUhU?wVyPgRey>;k00g!RyMa4wQm!erG?73g!K+<t3A<vRR#mfq0%Jgt+
zHbW45BWDviyR0%ZBK%9BwO{WX#^-cTi>7a2f}KUzMi6Em+s+iF&#LNWaY&%pD6Zs?
z!c>S!S<Y=~Z<n-la>GW@Ut6K%gcdc-9}1rq-lqXnyg*loTSxWgHd35d!slZ2Zu>;1
z1Vqlb%y7kABzJuD3QWOy-_Y83zC-!l%i7nfnC^MPnFgos@`@WV0+55kWl-u4<8A=_
zAw&74&4h1@juu`0cI@KcvFtb6MmH@%H|%MF@w5V(ASO#u5<Dmla<33ITy(y0$sII)
zQ|GG?ukDgkBZt1?J4>(ohny=1|IyNs$G>l%3~SD=-mF282yMmaB^|j7Zuxf^>nkRT
z^ot&8g{7DSZ%upr#ql2NC3`3L+Nk^FAG+OB(MO3>l_NgWAlyxSX37xF<Lxv#iQrOz
zW?g_`K2pCxally49+^YPbm)Jt+s;g8<K>>|c+h9QSbwBElTU@u=QcJ*;E~k*A5=3H
zmVgUa?pgfL^A7FY1Q2ZyB-(+TcTs%Dpxavh3*;_&^%)~g1Eq_@$12$*jTp|Fbh)%n
zfA1AsSjn(QQMmSOBV!dxFv%e^TV?C(Bop1>Q2)GIf6yEIi9T9m9t|rpGP1|ThC6WT
zYRnj-a@aZ#DYOf`o;pFR8UPQ0HKkU_;F!8TPqW%oP#scR<zeO?Hl`V<vfkb$T165Z
zYQ-Ue8-pV{?<wGSAI!m{8n^q*G}%0mjmqzabE@j}#4sM=u-G3|fMtO%tImkZ=yCCQ
z{?Il39yGkZx@MF=&L&42TE#l^?Vbw3<r~(UP*2mZ*4#ey;7Kca+#%hZDJ_uXZDS2@
z8p{-7o_*SJAl>}8UDQfXx4NC#Bh~@!iqVM8lD|nVsAQgoma94R8gCx3Gh&0it5>7%
zq3zXvxRp+^BlVJs9=E-ca#8V)>iHe{OYp=9>Ds9W<T85sqCEqqgt-IyN`At%wq7<S
z%e2E263JcedmJT7co*96W@)DNg=027cX230{P`-`O6CLl%rYcyVveaE**cxfsZJr&
zNE(zCDX_LG{ErQGwZb*KVg_g0lIEBy8AKCj1CEOl>W#anpW13Cx}Mo6pLBKxW!^V(
z!?0qT13nd8Qldz@;dG*<2TL`w+av6%p5M>|buQTpM}V7^Mz#{vydy8XE%G5<DG%~1
zCnT9lW^4{$#a4ALs73Y6Oh!PjidnLj-QiGV<q});rKKmd?d0E80yu~R{TU?UZ&=YX
z&*}oOL-a#UWYeq_NNn;!&RAp+1vjcFON_De;{8olWhd-rT|r{vYc35{q63HwrFp8@
zKWL%s5G^$qksU3h(q^B)7SqWf6T?DwarXSO_&FgxZ@<ga;8_u)v3OKSz=&aTVkmoK
zJez45rrriCojZ*!GL80uru`>jamErB8=xcRoiB8Pc;_bDO@BBscgIi1lZg_(SBf*I
z9VY#*7^0jyQB$U)kj$R(m^!CP#P1tw%^_u<re<Py3K}K&w!gep;rS;U(8}I_^vfNf
zCy+=&hpt9JeizcV$chZrbzyZ4dUKQaEIoRmFsRNaEB~S;G$?i9yt-Ozi!ZUy{+xhS
zbrL>1i;xH`2=znPa5~T7t%4kp4W=~FwH!d;H5nJkw%wRp(8&)QvQD-Dhn47vQ3`XT
z$rA-faOcc44J_|;psMDKK_we~K|&2f=kjnH=xy2!qLySBN=7AHMGOok<#$==C~OFA
z(<99zbLN-K2o<oTkQMxa%-KG^K7$mhTY|1rhNKMJ%c}s6QG(5yZ(wp-{gW9IPG%T5
zB!x$$%99^0KA38EVMc<y0-G*;LzzqI<l{0lTlcxl%jslPq4{s5zx<?#&g4{etZ6ly
znY|rkZh^r)i-2%eWG`7=fv>bK63C^cY37m9_EzZxTO3k$5jY~<gBCgpyd$B#Gs+s6
zR7ZMNfZ$k{f`*1sd$U*x#qMYe*S6ww9ly&<xP`+7qYr30_X7;d<vthV(V2>73(y2e
zN@9!q`7mOvAho#cUG_XrIa2w+V?iYiw1151B)bNU8A*eEMaZi%nXnIimiq_SOV!Z0
zCsHy|A+lfLvh^!5noh2qlZ6;e(_M1CYrpr4q}(u;xRR+BSObERzbQsaYsPxm*kb7D
zD0H?_FOl7)v*6^bKDU?}&MfMy8r+G&m;WA=n8di0Bn6)Nyk2^jx!%&_a>7yOJk{lF
zuL=>9PM3pwnBWt&KXM?|L)_$Zsv@%DG6!--(raAx2jA*9zumQN?$YaG;R*>J5yZ>|
zFfkEv?G{(^8hFAy{5Cai|2`Ujpz7efh3x0JQ)}T#^4=~XSWySQQTE5QRMOM9U=sZF
z0KC3cq_GUU%B_rL8+$$ysWx~q`;)*i%QKww{%_jR+8u>Js=e45?CLFYv2D@)deh<g
zLFE5Ak;B$UE{qY9iDz*mw4iP8EdeXcciTZPAq{cnmZ~OoTI*E7NUffgPSe(l2RxEx
zR)Hj+P+%N^sx(%d@kBXJ9GdMJ2mM8|02kp%MNkF-$l%My$EAIJo}gl<zqM=dTvW|U
zC?DIkdwPhgsHj0;tK){v`+7b7<_Jz!d<^@zU5bdOy7RrQ*ho*h@Tgm<V`NZF#=n=5
zHFus`1$T_JI@$(Kve@O0$Pcw`{&fnEF0#_j{Gl|5<{BgBZAF)bp7)1viq3c<Re@1~
zR=0bw0Jc2$KAl+L6Uh0q>j}YNXR@7hw86~TP2yvRnxIn0N9i>p=VN!h(>3Vh@5TG~
z#~D&O5`9bG@j6@XlaQk+CSt;ztEqF4oX%8-sLQYqpRt>BieKygVrj0vdj+)x4bbK%
zSD8-Xk`yisUb`-nY*!yKtOkE<N`$aNUpZTaS~}X>B00yd&HU%59oJ$UEUkYT*7dGt
zx7jgj12F1f-CtNTFtX<<*2oroa`unU*?PgMGP6f*9DE8-ErfGfuj1CK_S9@vQd3QU
zi%BU-6vwNoHxb|R7~?CUlSNJ;blp&CXUqyPW;3_Dznm@H9D9l#rOcKm-6*U?_rVOu
z_L9{Hm3BIGtX!W{!`L+Q8@$hbZ*hO2O868^uungK`e!FgpAwU-eMdj1!PC6zq@LJ5
zo3EgcTU@mL@-5+S8mTnatY}XUuQ3-M5y<5-32V`%+X)K{4%g|co%H?VM2`CgGSC91
zNH*T`_xQaOUZ1H-QI^Uh{zUWObLTpzfi_!DkN?0HO_)!q29Ieu#;RtRX3`DZ?MwGG
zSf=8=-LI_v$s2b%Hejr1Ss4gTSgN|yrazzKu{J%<uuFU=EfXp3PG*#&=~t%;f|&85
zxw%&y9&dk@P{S~+A(dp{VPK=OZypxfFr@GqV&x{()%P)vNtp6juw;=)%r-L;pXWF_
z96zNse`dlf^}fkpV4Aa;t0&*iT{Sj$_iB{*g9>7Ti=)^&zY?P1Yr5fc_)h!SPvtze
zRx__@<;o(E*4c6f%h6nCPCAV@UO|2jE`gEEw3*`4t>)5O%w+o7DhH$q#C;CXz^a)3
zU;ic;NF|SIt+~ive~MSV(KYSkHt_MXa)GKh&6aUF_@hh29P_P>G_ztQ(xJjz_x5zc
z&M7SC@5d9qe~SIthbVzg10K@EcbqiZ;K0tK%xd3VgpQAQz%YtnA6;}U4|_mPw!=?W
zJAYEi{DLuR%K{vEStTgX*_j1#o(8OMzG?T?biJBT!dB|O{ieTH2si#E{W!h4JtTI}
z^q%@TZOv(SHFkEAnPuAwggbZX??MAFS=*>z{kyezx<}dEzv%w3uGmkUqf`AOMyAGL
zWO^_$26)i-Pnb6|{ODJVzfbwrpq`nv_<6c+uN{vLVa8Qko{LpW2Lg7U!4m2!`Nz=q
zF>Is$t^9Itskw$LcDP}B=NmrJyv@AtoAy3Y{loM;7EeRG?h(Sv72c74llLy-GjWV@
z842#;)U5CG7w;u}6@o%}j-3)w#;R(dLNUqy@8`P17UV>qcb|+(z~k;hI!Dx$DKlBQ
z^x2~^B~wZ68d5a>d^y_pkzvjtL?9k}!~dZ(FvP<|L$4mjLKhFRZ@%`?za~uDUj@Cf
zW-kUTRc~!SpM&1sC?6i~!%xyzd5^bo_S@H^4Gu0|@3(fFkv|~d2TK`hl~KpR`8a5q
zB2d#0*?w|sOR7;i!5(r-NXB&P^uy*Gxh3f@4_NzLCY{>oBdJ(eY5*hUGjwioT7=ut
zU&rt{^#A)2eA5knD?UgkcJ$>lYBz9x`Ip$a0TqKXy^KHEQky#K$~YcmVn-K^6cDzC
z9b#U+zDxYNO?-1^@e!{3-REjGg_as=_Rf;6AiuDN62ba$QSb?gh@isbV9}OHKnsFd
z18O&}s0vHIpvBu6Y#^#QB@NT%%D2J8k+myPcPct=woo&NH_elGMQA^c1k(KL*ROHW
zHRcTyc39+KZy#f)lOW!2?mba8$+doet7po&wf`oWbBUkx$}hHkH>k7TS>MmWCsqTx
z0d12la&Nv>+TYC1;u>^0=X4>_SmUH+LW*m^8mSmo$J7*L+Aif;P!3tRl$$Qln{UQx
z$&yUU>td5R%y5luUf+9&xUV~tk3W{RH$QFeeg%!K5XNwE>&9Q`71)B8Rd8dL_~|r<
zjQX~oH$Sq5?_f@`IG?V68-!i~e&7F^z}}k`+N%F}hw(YsQPFSr3@gsW5L1bEE_NHR
z)2ciBYG<8lkTGM+?rDdsD5I(a1b^M&z&BqP{<GhZKNZHXmdkaltsl(_k2}w=f3whD
zeXVoy5Rtd@>@#KF2;r38VZ<@g|HxEhVt>U;>>K9q_2^DJnVQq}G{5fW+B-k*K3m>%
z2(n@?IQ1raJEvY>Yn8tYg$^1+x*1Ju9oIT#h?OW9T}j)k(39s-Pc#`fRz(vHMTldW
z7b`;bHKV!n+@<@~As0T=rrHHR6#s(WeH39frPI#od3g$wJ1~_{QT=o)(IkdZ)s4uA
zIr%+ah7s7|yN&Ctk5t`AxCZIigFPQ+?Ww8aLl)d*?c5&6qEFV9SV41aS-FYk$}z2;
z7JY5*4f4|{Qvy3)t{E1aSNb{i_!bPEeXM<9$Td<eg*rJDPxsLXE<~y{fJ!6JIj4hF
zCdJ>3VZ}pu<Av!;rnnepwBo52xJ67{t9&~f3)Mg~R(;y@x)SHjR?nQ6X7(4!FfnVb
zC7rrcDg*z04tI(6+pdFBWJ|r8-v<*EChsys?{NnSLF~o6Md@dkoo`5-B_>3+UL2jM
zJ|y)Vcd&o=iK1x&qiS5y2-(F`Ng7K8qB{WH{V>$lnX-{>^G;x3GEl&rfJ|i<4t8tx
zL%wN^y*j#Uz%`@(5YMjTiD%2Z#*~pUi^6sCN4H;$#=V`hSKD8Qeb}zt5lnBR-gBt}
zyMC|voA&}y8I}A4XE0|v5Z|-GsRD2MPJnC-y`Z&Wy_Lq&*K~!)g`>!l{&+Y0q`A1m
z?S+ZhULe05zJhkN9<)uqYaDFXv^RUbGdf(6L0qG=x!u({UZ~~V+#=cCGFabFZJ2At
zXVB>$q~U&keYyv_YmzVur!$HBv0N!zwMFdcQMW%T9c_@q($<vAhS5-Xz1vp*jpx&Q
z_GcuCeCM-uOmIO-hJG~Vni`mE0(PX-`}#RkqPQSFm3B-zU@>{nM}$3nCq3MeKLpoV
zwRt_QbCT1%dxX$r=8V*@(>OB*ErRPvs~H@gH-hL{H4;T&2Jf}CBRcR9De_%2qn4AA
zMuPIU?c@cY*T_W^p;~Ky%GN74hn{3PV-9b|hX=*@N}}0l?iXw0hz4lFT;_-|zD-=9
zyzKRa!`Uj^sA|Pe5C&KsU#7nQbJYnwezUl{3KU`hqd5(lv5VLbLwJ?~8U%Uuy+3+|
zXFT8(`oy|mA{iYZX7ZDQI5A}R+a*SqYumq>-i8kGv#AnYCXG$*t-;M8;!y5JQ%wyt
zJaK)T#K0@liPs^}U?ZF-#8UMn8X(v<C>nRoYxMhp;P~1~N{b!4#FXXK6wdzc-Wgi3
z?RQu}?f`4HstcO&B_IDPHCq4^<YugeO7TtPRpH*zRb=<aZS2KuHU@Qz=4{UNA6+*T
z%BQ+V4^qRBILPPBt~*sV<Q`Ltfz4<r&%Cr4mZ1W)uDe63xm%s8X#4B>D8VUtHNZ5I
zg=7q^8dzd}mrK724&A{>7Wgwet1{Noq06M*rWWxcC3<8$a$_(YXQ!En@bbKF2e{>o
zjp&Oa<utt)pG~dtc-jL0*<c20J@*zVqQsNA!uot1(*EAZjp8dw2c?3(?@?J~rpQG?
zRgGII^4#)V7j0Bu$+r7ruPuO773>hhN?@ehAP<WakwYYAT8*`nD%Oasosio?eVdIo
z+*<TU>Tn%RQ6a0G{|qaxw4}$W*~EZLPm)pn;-?Dn_L#m!OGHsxK05eLTo9cXbe4=b
z56#0RTqV|3v~+rZ|4r8xO_I^`wnG1UJ?S43LN^jlGs>+?G~>oEqvN$u4@YQ`fyA=}
z8p|Y+CaJwAR9q^tDX?ffu6cs}Ot{Q7WSMw+4b0DrLSEiadV?b2^e&X;W%TT_(Bz5d
z%t+4i^5RC?Ev7x3T;$5wwX#{$qel=v(UxoN&CCFDn-FCNgVy`7$CW{1VaLGuKsAgN
zK|h&fH7>rab5AdqFX}kaGG5=WX17hY3vIbK?5unY+g(jp5<6Iyb?d7%I$jRR{aWlu
z@NeereS%664<N3u`{P3eidQfz8E7Vl*C;>56vOt_*cBJ`7z-><BODD@s(pf`Z*^a)
z6%F=d&fd9gmJHPjk#Rp2kXTSX1As^Oa-fen9c=X?bb)yr=X&7b;~_#Fx9=nii_1e_
zMg_gZy}9Mzvn43-SF~i5+B<PKqhSguvO|NRVjBOyzDal`2HvN@TU({Xv)jPGQ^wn$
z2~zUiUhw9O&^F->0>@QZF%KEah@J-<;C#@&aww0}_I{493j1R_tYh#D2|=ovW9L>5
z!)vx?=i%cpRCmO6k5tFEeQzm|g}XTAc%wblrhnz(Ds{M+Nyu;bQ$|5!m<ua4g4+D7
zsG+BmW|7HElUz^vI9`El*hsb;+Exd9FnH8M;*^|2?6^%LUnv<^Mdf;7bg6ZA^8n^-
ztGBjD%a+SQ(bF0*^p87BU!|<4To2<0t`UQXUk}(m5}B;l{_or^#WS<s3Wa%>9gzHb
zek8DIyUdisoVI1`>D3ouz)DD)NOd%!3#q4SLvC3_I|*Clq|hHKYLJj~*I+w*!ntXx
ze>S&3s1Q8KBmFyu8tP$;SGW2D^Uke5f_DIeONS_r-^c0!8c6Sje<_p$=S%2$YcuF(
zvQi#ud;4nm5<7dyE@H9FhSJhHkMT0nzW~siuielCjM`%?4p(p@@gg9slBU4MqYcgp
z6Xz+D!)MLIOj>ajS(CLrthf4Qy)0cMMCH-&Whm@8ivyDx++@1|&v^4^NPdU@Q5=~y
z$Jz>Sq}~r(B&?LOorUMx3QhR%{JXQg>6_mxc9R=5JY4(g@Ba^HZ`l@A_x=wH(%s!T
zqz>IVbPga=(%mVI(lvB944qO^0!lM<NOwpG(w+agzW47H+)rme+Q&Nfn)Cc*R0BnW
zKCNjP_BRsk?k`u6>zmYvC0XZ_3GxBOgThJCn)t#;m492r68}SE`e7}shxg>owoyq`
z1DUyO3qt(Z%Cak7s;cp)s2^LC3K={Bsgw<WCH5);<w+@O5KiAt&jI+P(Z9!1Ir_Jh
z4~Yfp5?V}1o9uW^bW8pz$PeoGY?gh8O-t>9uI{#H&<{u-MZMFS*4ptea0KuCasFJ3
zncL1}mBxXI&zFRsoi%pm7Y-zs)2}zkgI{1d;}i-dO~5Sl?-aD*-zERlNLuRbT6Pq^
z&mZIfvH6)&@)wASTE0rFnD)B0T2reS2ygwc!EK{&J+M#+3>dtHmR;woE5H?OPq>*H
zCd|%!Y76fhmrCtAM+avS$s<&;Ks^|$%r+z3)zF=X6PkHF`&Q;w@|AxPiFWD%<JMf9
zsnE}c6A#43h{aGSE@iX*`{e2AChV9YR*rZ{{n<e_hYi<Vl|sv(l6#^mD*EZagGNs!
z91zBrufwK1KHuRtHTDZnmbqj-5a55QWY5Fw-5T6t)T@0rXd!RFt4qv&l$l_2{Et&e
z+`vpt?trRCXnx(pr?JeN)pDfactYs9Ty+T90_OTmE?0JrG_QW?xH3gW-S7`nS8cRX
zyU^P458Fm{0amsiWk*x|X{<G+zX4&x=+N+$goy2gfWu%@)I_+C71~Anf0G(0Yzti%
zG4QPI0G75jXgdK{oGgCI&l4-w>y+6y@Jw%V66wC}i}d;CQ*Z44Ws<&a7qe(<{fXkc
zF!O>;+8@WKbi##;P#bEK_VyvikC!l1%J~yw)A0ITPJz3^lw1%biqt+jDu#?$Z;rO~
z;YUczHD}Cd?WvKIQgI@nxd%8;E;^98<`|KY9QneQ<0q>##fJ?}SZt==@O?$7mUv$|
z(FVjo@_RV&@oFB8Th+=QE~yw!h4!3pDNQ=witMaGo)x%kUiQw=B})gJAPxa%HMyoG
z#h4)}Uq593xna~@VZAaN_<l%Kni4o{wHdiUPQ;Dk=Vl3Q2WA7Udg7fIuT+e_U^he8
z1cuuzSA@G?w+Kae&Al6YA|uqrO1}r3&#TRgFn$bf;IggsZ8mbR?&>-DPMx0Mp#Fq3
ziJdAYZ6{mn7l7rMTaJmKQke90CgysN?Gq^;f7w53dQhYE<&oBgHhGLOft>}aKw~}Q
zpmjuBwt<|0Q)I)`HU2*_b7Bp3!GOEFFyHU=V+PBM_-WhMUC7fqmSv^4<OV%iD;KV%
zZJz^W9N*TqMg_gC{t}8uTpOb&A`lphi8#L%?L&)K>U(z}s#^7qQD1%oRbKZKx$-mS
zuMfDpR@+811{}*|6?rxWI;%@r+vW-@zUea_9m|@64p=g?&CO04)>yjad=$im&E$-{
z+w*bg3#2K0pmJh)>1qmr&;2Ry_h5c<CY{Ox-LhrL2_1^&HGEyYqG&u=zu96L_(SFj
zwT?P9w8%0H*6StGX!bh1E%i>uNGw^VT6#sAYqBGn#v@uTEl0K=X;@<%L@`sN`vI1=
z@_zjj0b|!|z&bawQhWyBrDL<X{_1ui7y1qN{gqdZ7t9oZ0!4E|1xmRJ8|3JOl7`OU
z(+a+E{IMk`xpiovbCKl|ILS%t<}Uy}qMndzaT};|7B=8Ir!V9Oox&i0Ue>*!(;zBB
zUTfot#V+IIB<oLAoU9zCjPX0~&Gpx<;a(AclTuFX`Laza=FQH*Yeb+zi4?t(xT^Hp
zzoyC!R(aP6i`|P)c;SW3@*D>pBpgYGJD=pkOWo3&tYPPVefmviuUBuX%^z|@t!+NX
z37&SxWGI%ZkbE9lKZ&x_|JpiMj_HmWNG@ch3?HXIuA2uOk&=z8)OZ0nr@cWNgnn<Q
z8R~v@m;toxM7#&5F9eqaG!6ohPEiN8I0EwjH(u?VoPJagT!OJbOa%jsFcWfQ)vpBF
zkv@xiVo!MeR*Tb4dKb7Q;C*9_^e!`zm*0Y9AlItt&zX<ftf+hwx(ElL99eXhU^VKG
zedPW4!t^~!DV2eh(CJ~`(9l;FN~Uc_7W?F~LM)!Ynm^*z{TVV`%1(EAjmb@9p+nIM
zB7Js0{_K1%a5tC{;NS9*Yc<{poV5QTO@6mj;Fm`kE8I=0%(lqkec&2!bwxh<Ar|K0
zXlyo$Va2fHC{Ihu>D(=^i_E}3xCaWZ^Y&S=u#n($wwO_+Fp8YnEbR4-jIDIJO4*z^
zpjfy7cDdl=KX8dTB?#H7!$bC(ns@~s75G;8dqmT1ICyd{<67akA=1bn)eo>{PG47_
z{+&=>-IWM?F^m0WH63SiV&Zm#!Y1V6C8FOvhgNJV`&1LcX!wkF;{=Q-n8{`5DW=6!
zo?JbkUaM$C#58UnFLz2x>`$I|aFjp)tZ*gLX!<d~shChVQo(0IYr!zPY-|05Tk`rZ
z8X+PJ+2_%iIqOl}0<k^EuEC>UC;vo=K15&!cRk_ge~z{HmWS{kpg1LeHaybS=!+S6
z9~BLyvrnCg@<a=6IXwrhe!zg=vQ=T%Jit^GXI)Gk-6^kp7M~8ap&6GV2pM{6;85)n
zRX-2;q|ask-or`MLW{NlZUH_BKfp4zM*$U;uZSMkq8^!^Oh$E`R^E?>nwDVgl)#*6
zrpPOM@kBNuj-r@L`tVBg?;CRz+KV^PvvtBxcj!W5^*M3vyvsWvgTISg9lzYV%Cxb6
z>`*_H?Oof4afOi)rJO_Ft^OFQ&5LPbgm+{^YF(yAJUTIsU0RN<Vk*(VJoDEPOFAh7
zzk1FI<69mxVbxK5cI^5L(#iWjbR!Ch*!2>*8|KVG*vdbFGDrT~%xQNMf25p=Jz^YX
z)~@+_rf;xEgOC{kF(ErU7V7{E`4kLVqMdVihWzwuwxwgW#$sFi+t*aepoH0C-sNtK
zWdX|%16$`YgSU_iHa_eEvBx*!m3WoXzReT(i1bK2Jij(U@6DxsK`5tQX+*)Irw7IK
z##%?Rc%P4>a*BmnJ^30KeYWb?rxb@cijk}TWjh*P4xFbZH3G0*R@ba76^I_lIa(*d
zLUa_EImw}Aef!xB4*K6H;330;E|>;>Hwha%oK$b|lfKTg=u;?!gUL};=%703e!^kW
zV%qo?OrWBFqIh2pPY;MCDF4S#Y1wJ{&(*Z2!~ylyNvR$7^&g14(T^KcShCys@u)ov
zi!cC0i*-B2d$feBsf1rZE8oMCO5Ngj{zQ1`&3e}_&5MbNnO9LiuCe(KXyE6Rvpw@g
z80|;JDq6#p@I4gbQ5Y-@tmI5{v#Y~@sQyi$=1s1qPYJ`P-$7CP4G#hV$0b*H>vrCC
zt0dCh(n537q%^?J*-`qRzl=ePf&ESO31IAnUT*f*nomP}FhYUveWF{3(`=8U781gN
z7;}_kjtR+cQ22y>-U=*rAd(4x+)z_)U`Qk?VCae^C0{822IO*Ia`NFfDZI#)5Ye4>
zBboe#zEWo4?-`SS>c&eO?4oZ;Qp^81iQ{{FmY85Zgl_}bXB)wHq7`0u)-R|#_*CM+
zh+3|40T6Q%SpA`k_i=?u*^w|om#J?bg$t$%D<nLWi6D>9<Vl=BHX;RPa7qD6Z9|Js
zsr(1v=vosxS9wFxjf21%lK<z3pnze@Z)-NX_7KeDDhOSY%g<Md)46i0Y{D}Pbj;?N
zn!J5~4cckGqzSq7qinB_mp<1JNB2n60%$0MUbvR<h+*e6h}jl_BSJ{;fqj}30sOV&
z5S&5f)75J;1ZT3W7|`D}lp*hci!^ZCEn%1?aKY16s=Bmw%&r%Bp}%FnhZXhdZ1XGO
zHH0;wpd4P^cLC|%;He#S%&k8WGiERjLygpcS1g}4CKJPTCx;~X;UonmNlb4oq#gK}
zU`YVml+O7{OlSPFhxUWbguZ*~Ao3O;%Bk%spiHwZSVw;lJ{7eZs?R&Pr_6rm_Y@zL
zlkoNVrC<}8Taln&iA_1%=&r)TBC9$GYw1x*&NOi{?0)|73L*6VWh<bdld>mmE`=P9
zfnvqePmxMc!;_nH!B+Z?(+S)=e?N-Zf%{H0&agRN5B;Dpu@Il|T>jkim8_bdzO2!7
zL!f_s%Wtypnepb+s4^{s{9**fKk)@?%yhNMdEWpc(B^p${_F9x%XDS_xe!AYqkQ7Q
zbWd_6=Y2}%??){#Vfb!!F!?;HggnLm7LPj@d!X7!=g^F(i;(z23;A-9PJG?k^7VPu
zYM=l|A|2m2#i3b~(d-ID2r+WVR*#?KWP(&N-=oplRpP#sHqYsP=;tMu`XlonwA3{M
zB^oXuyC(UFUVlk<{Cp)+<rU_#Rp=*cYKC)yjcJaj9#>*1V}`3KOXCPGFI(3RX`R9;
zX78#M&f`?8CNpvo^Y_hpw4Eiy0tK4?jNbY|o6$6h&K~k|h?>|&%NTF)Pm>cAR%+0v
zQ!>O4WIO7aw!tDLlzHG?DA3ppgf0HHr{7M<X99P0{=T5_D{`Jt`g3-CM4>2%?If+$
zx3nYp$iJlrJjT*rX#41_850gOk6c@6MY}?B+s)cbqx^Ava@=@Ky)Gw%mkKaCuzSq!
z41qmm?#w2fO*0~?=Hq0kCcle{w{NR82FRXwk|EQy)nsnp@7H7S=hXK!F|#Y!u?JQ8
zrLEj&Hq8>xRhYLl{d$bTtZasUHwy^td3wk-earBkhP?ai@2Av;numO{WS(!j^(Xdz
z(GY8{oIUXl*X$c5g?{#CtzASGTOzx!gb@ht6Sp3E#bT5BL7Yqt2Y$rD7%G1H4BJu4
z9zWaqRxT+W7bsInM+U;c!oB?B^g<AIgxb|ShCzG_j2Ul}CAg4W$eF?%&zwkACpRy@
ze#w#S9bZ1Jd*f3kU6P^{d3cOw;VcbK%Ii54%w4G}_$1u>;mtki*Z(v$ZZC7*J@oPo
z(fs9GNIkp)7G)ws3*rhkHeq@ein5VQ0E<t6_>VSGykR{oo|agsD-lBq<yZ8ZOn7fI
zZQm~!POYx8>DNhY-d&Lcq?71%38OMIon55G$W)wFQVE)AjBq30vH_zdo~u1x?v#l-
zZ-oA6>#I@7TaPJf^dQsF&@}!{4-1MLD1;{dAs_1A-(gMA6j3{@u2{SYzD<a?6^Ufr
zmXQhxSuFe@<(uH4;JEVE{1`}<Dj-CxnLJ+sI`gRk(!o>=4|o6iNOT2M_Fadoc~0X)
zruy?mYTlDsTW2KEc(fRw!<Oj5-k?_bNU1XHoq+z7xolKiiQfLyeLG-)tLI2@6E72n
zc#@E&v+T!W{3<bb*C%?fyj`^#Wq~$hF#(Wjn2&Mytlyb1g7GkuEW~opojG6VX`%9~
zR<_7*q>YB1ea@%cDJ1ok^bg(<r7Zmd<WA(_FrJ|OzFWH%{$(au{+#~gEiMjPm&QjT
zARbXG3-e_SgI~f7^-9NJI1Y7~DqAi(9%EMQA^_=}%mNH(PA4U(GSi0trNIH)|IUd<
zs(HHRnGN+HBTZ{Ty{Cd>VC*FhlKMk@$C?aLzaJT!M8H*4Uy3#q1QIg{TlS^N&10cp
z^dnOr!?Qw=b0&(_^5_VVi3&6~unTm1!<Nij6qT42G6F#nc_+C&bBpME=zgSwcP4g3
zbJ-Ty4{p_9jR)YzntxxLQHVss1#9Yq5-@2_Xg9kU`+t|XN?h2so6lX#>T^veFdGoz
zGx3ZQ2YSfU+S4UdrtAt%uxf?iSx9lHSPi|i)Y^X`J3M`|q~d%dM?HV+o#4AFSrl*B
z$+OGO7%|+qE_HRKmiNHbE)$2DEs|KR=!z4q<a5!pL@t3rk>dwU7CR{qmqYl}W&K@T
zNm79gBc~Xmj*|ujGwlzKrKXmtN^4ceUg{(vDt6Y`+$EM&9ta=b`ff{@QOBRL_-5g4
zZC+(A3xW<SS0XAHI(V8f*wb)vnq(-qVD1BDM-A+(fg0Mp^<!bHd%tt|)qW`X$C`5L
z!HuJ0Q>bnLD?sD4FY{gR@_m5Dr%rq5m-RM#=uaK2Jvor>R_Z+86{5x(f{o*nqhMI3
z(4EzX2K`!~??i`Fd(xlPg7r<*mGY~RVogVpgb%{=2bSWL#DGvRLO3g8c*`!kCI|L%
zViNPpN4VIGo2ih;q6gI|%8py+w_K9Vih$?yA1emF7!l{gxtIwgk<;inmS*vzDSWO)
z1_jE#4!bN8-i~$PIFGW`;Wou|%)Elp@l;2DCSO4a)2lbvp=*}vpe<5#p=X2uvVd}S
z)D(<G)R)OhFWJ62_N1LBmH<!Bd0ui$7bB;wq=ixdjP}givDlQ(m>iJFK>Km3IrlsE
zWoO}ayo?m3ij7~uRE=Q*shruUr)i9QbT2^7ih&X<ypdBjL94KdfGd67dzVFRDyL$4
z#MCn)x|ij5AqwtQ4`$76u+yxOyVK$8a+_R<l>=g&B(k8q6NUZ4R>$!1_~Rlc<CN=s
zi;hUK7j<|;6c4~c$br$o|2>A2ij?d(H02@>X&nvu5{HkVb5nOu35I^Y$zn&4r+u~q
zR_Vl2N4)U_mt#Q&!=!F?fg|p>Gs6GjCYf7*f6GN;9`@i+P`Kp|%<2GVVm8IPFG1fH
zukyP|-Sei~zD7ZDWvo1My%#z8iLMiYAYy#xOg?za+P6g45XJ0&!>WTC|BQE*{z3>(
z(MAeEQ^2W5944H!>V2Au9y#;7tCi}YqA*Q(L}+j!Ll=p?FjH8RwoD!NikeC{FVtCt
zfvfu?q}j^lb6&y^Y-LJS)v^xKct84vSWxLk`E`C(J?J!=0G2vn+Fpv?HFHJLqFnTy
zg%VI;OIJd#<khYE>-NOm-&Tu&*1XUSn}89FIof9ZueEs^KVk&zD5A`RvrIudZ|nZq
zY4CX<P(uG_<U9&3LlzAZE^-tLxzGJqOqx=WL46zpJlY1mva;Rp(Cuc2iZod$O%$R)
z+SgQN(ipSu{m#pscl40T?+i^8r>2yC5S*wON5{C^f~nhh1Kff<%tZ;UgWS-6Bp>Yu
z=Uw4`IkPp7`PQ%sv}G+#d<%I<QU#VVE8`s*SC*8csAdR>1~>UaYFSH6p`Tw`#lEjD
z6e<%fx3zgY#`1}o?*F8=+l*Pt*lX#T&Z!vvX!BUisHHwFtBrPy#WkkiYr$Do-Xi;H
zIo{Lk;$a{%8|Ey2n6!HGm-hL}c7QUYGF)evD3wNuk*B9%vfMr{$YN3miHTy>MmZM;
zb`(Xt6yuf+mr^;sW0@XZZA8bpnu-;asYl2wl^qw}g25uhIACdJug6gnXrh<Gv*r*i
zwP%_A%H6WrvrOUECFZr7;=~$LMu@wt6;gEkW0S+v(>iIpwS2VD8$JA7ha`{2UwB6o
zm|huC35VD8)Wgagf61nlrvT8Kw$)&zfuFN#lZP$E!#t{CtmQ+w_YvnOA$Y{j09kl8
zQ)baV5wSov-wy&t*3l+<*%_Vr_f=T!=)3c0v90Ud;}-(*%vPq%bUpacxBPsWk$gvR
z1#cE&5+IC>4O4~st#o-1L`$aH5McE_F9bDGC0AE3Wm<R$HFENx8SlIGk%19JN1(hO
zS=WvULxdfxFV%m_vqR1y4m+Kv72+P(d;YykG{7q$Hkx}{nI34$yd607>%GQ?QYp6|
zEzsC^IcWWVxQ4)FX53I4-a!BelT3-7q6w_7Z&$@>G3&oBDbCoGUUw+0S69KG+d^qy
z;5-r>HE(CCN^gtbk6j}lZHvv0VN;zA6SMGkhN*2D7@ysoa6J9qcG|O&<D-v$AER;1
zNn}vEa^_(@^GC4`KU0@j*Kj%Uph7o!LdOoPiQPhnyrv<*(|yEl0lXYi7aP$0Ww;U`
zVNBcP^@+mzcPM%?QYxDKFL`NvAjJr$MbqlcW46N&@{}aazrQk0y9Y&6S{FxcqLh)o
z;!z<aXhjzx|5RAsX~fp6&HBKM$VU&XMA3S0VUd*@0nW<32)aE;mB9EQ(EGjjF+sy;
z>gvh$9*Z@pFnKp{Bm#*T=?^j`Lv~__G@qk@p|ypXi!-1xK76iI;7j<s$)Aey@dD&1
zvaZ;`hv2~+9zn09FTbqn5b()JiLI<{EebJ#hvss>Y-3clNwT41z8%hbt-)mc0;yET
zE_b$H2K<k-<9~QPIO;^WBW_PM^nvIJ=|85YG{HOPtXDoGt+@rVS_ri4ZnyJx!4^nO
ztjfyJLgvyBsC&Gl!blZyCl^>hrX2i#QcTc?55GS}CM2mv-I@7tHJMsi8mN(HFhbEz
zY4c{TWyo0^b$k7XYyz4^3UR34NTbXaWL{M><yqQJ>lHUJ92^KAQ8~dcVt#%6wpMTK
zyaZqGXZl*@IM^kE0xglj#qWN(Pk;vYYxC0kf`rvQKyBp;lf^n#p51gJ<*;XudE*<b
zNVV~oj<+`CuvCIk_@W1|Jevg(MMX#1s|KBmJ`_<|)LYt?xVVD;PJmrL7K*YE_F0PU
zXBn1~W2X7B;L`&%%_FU^?;p4;5Hpth!2D2Lgi~>+sJ6lxyaJ_1Jz~bPIV8;hpBCr|
zQBioFdGCbJhzmSzK0z0Ck|WHZ=tDe%)hA^RERt&8M%7>;1Lmth1w$9$JZ}y$$O*i&
z8|)?_NTx$YaU}aq?M~d`$&<`tcr>@;2%W(0NmHM&o&@rNlZCLkg^HhRQb4?Fmdxbh
zrdHIkbVpP}0($az^`iV<1S)5uSZxce<gBD@207+SM>)1~z}tT0liZVml`O~Vq##;g
z8J#j6nG{>9^05f<WTeSGER*Krood_P;7-u*9IiMIvf=~e?5(jGXriX+PLx_6=!K?M
zm&(b{dLZhD)!>)=D{tYG^vTNa1m|FYQ+!NFu^goe2{B76jDzWhjPDa=rcNSh`cO{b
zi^j&wm-FX2Mj~<xQD_3c1QPUuj+`R&<4ZX2wL7jgTK@Itoc_0I&}hvAv%iL%{RA6i
zA-JIP03RPX`pM-ywg<4f8^@)szvfCrOpguLqKt~Dpz2b;#HR?Ki}=taLoT<5)Tmkk
z>E#&O2smBvb$9PIph}}=&iF}oKAm(#Ae(Rdjnpe~w7AkCd)-VmyVmxPm@ioIy)8df
zNx=8`TdUZomg57+n(su9bwP@=Z%@R<G~qd!GE><Sro&g;29{!Pelt4ND<?}Mc6SG#
zHi=D1lV@281KJtXqA_<7rph_QbVRP`_A5f{vo{29tDBhDvETY3GGmm6Q^lbR8>F?>
zG57yEJ*IX4cN4YLwkdQYCLyfq2tdY@_JvGn_H3hr@*q`F6tz@NJ9yMf&G>n;mEpEw
zKObS<ZY#8B_2Nq649p;=v=@RTQpM-)0#EqNu;i8oKc}|Ly4|GRPnttfv589R&LoaG
zil<1WrgQv!>+kD0tNq4d*3sV|pic~a0in}=g*@z_QhMHMMyHKg&tR{egW^r`J7c|T
z%KYX?Jy<k-Ll9CdriA`>L_EOF2UH-Lm64Sam2A*zHmoP2MMA)+<A<ZwxSo&7ven}^
z?!&{1GIUELW=cqv6&N&hPG4epG({;BF8!G~OKL99W>$!T_6)owf=1uCK{H#axNG##
zyD5{+=vu_spQ)_xY8QPOyJ1(+RNlj1Q9Ok@1?C;OImjz>PImo?DV~s>e(%_O=nlKZ
z%@3I)EwGA8@{wq&B$9VvpaBPq@wSmIwZLECu-BAyRiR_kM~8MgLJ`d_9W-^dj$g_e
z_NJseCL&g4$cip`tOW6E2u%?>>}dUO3};T$dIVEP<z~-PasUrA`;a%Sq&Xqtve;`~
zGk2j<7Tec~uK>wrZlX~ql{FL!Qa9-{<Fmj`H;-T1-XQ8#x|WoY7U}!g?x{<LT@9g8
zkMKzde93uzn~0`pzbXQ@^aGtDdM#|9M`?jJ>Fi^En3#TH-1QB=d<lUe<9(o0{6#pe
zYymA)`=zLnD3!`YPKIy(wdM02-q|*xC~Jucld|G57e2H^>VmPgWB+@i3itbGL5!;s
z5us@<Z}G0?sQ9m2`k!XUU9>Vh?A<|zzcHZ^NvX^rW?4H_I@9k8{q)NC<^~o~G7HEZ
zWk2;oZ|rnIZ&-0Ok)rJy4epYXhcx`Ia=%aoHxH^3-x*I|;lNY8OwJ4IV9s4V<cz8k
z^f9i$l**Ymg;+I}s%)t~%{O{~BE`e-X8(!$6ji?M_C}L{^+(YuZ^>$Lp@GjDR3vYC
z(>#9<?7MgWV=K_I)SBXxSrcUl(TdRCXsEGXJaUn0ku7VYU6WNR3HfRY&z@2io0m6t
zMGD5#5{p6%e8(3@*&WM<r-0D2Dc(<aMPGuY^*c5oisK1RH@)$VZW9VMId(8VOK$p0
z8Q>`WxJ7Fa^tg5Fd)i|GSu4UIpYHm_11?lWO6z<Vf(GW8LWr)Ec1cU5C(CPulUnPj
zSTc1zrpfiDkmz>)Gtb|62;9n}8CW!vCO6w{`Mp&!%cuf%@~J9TsHJ*d{I3V7KL@Jj
zAlHa%kFhyiB@g%Q#MP`{V?b-gX_0vK?8*4B<~Nlt+p-kR%_8O>v{vE>BDU4Z$do7F
zLMtx&E_i=S@7`0SvFqP*#koXm=u6p22>bO)K5mLLXtPww_`NbV2*sTM5D}Q(<lFk`
zcpl)@D-WZ#7^4A!zs>zV6(KVwb0Y%-)s<nIS}GY|Ne^)|L~RRZTMJ&f!ga|5ryQ3l
zz;6jLWKY94?p|7j^=o0{Sm#>`h)uTAp2@7}NX;ZrcQR^CV4_`*J-v3{RbZ<0vOM0u
zl!Yiz=jGi&B>zCke9rnCVOD=5EvLKxJkxNob<|4$*=hKxFJ1kih)By<e>0|z{iU0z
z$OOrf6T~c3c<qI$%*tomYP&8x*Fr7bhHGpIp7#xOH=1wL;;))5r4~3uz*G9!i#Q%{
z?LwXq`-HbUfBr`Bmo9$7x1cSGOX4>Sdr+XcK<j+Hd)s(~zrbB7(^Gkr|3!d)#v9vd
zQtl1xyP9kS_goO9gh2nxbfX&?u7qVXjTEaBS42<FhE1T<#e5qQxAAQ+4Tj7bvO$Z~
zD<Z2<y%**BmnR7eUirWAV^+;#CdEf054II~ve@})zGznseUZO&ae=_4iCaxp4-0$L
zz~(8^=Ay35+JrXJFYWsU`^+fJAf(gh(|d1Z(s165*`}>06~7n!r;{CxOsjNf|K4G%
zT%178H>Zg4*%Po|Au3F0aNR}0V#xkd6>W>66%D1K%oLxx*>F^|P{)Q~cit<*rv<n0
zNz#u9yUg1eL5CTc<JPj%lz8t)pyu*A?BaL!Qxv3e^cV9_BtK@^Mn4SA`oWaM)Ff34
z->1){Uzx$vG65Hev<7^^V`S!dPn$>JJ@WSN6}E0PeyD>aSKUQ%hRM3R=EEQvKW@4D
z!S`9}l3szW&THErgZ_LB+O4Wi?NvlDM=N>oj<Js(lU!gvUG+`t{wKV2Bj$K>&aTu!
zR-3s?%v}-`JQFg9VnH-9N0ddrodo^&b{-JJslpuzp7jbbjh>!u+%oExGlA&im}!T7
z3;gE*A8@{X`R6dXAaq=$OTjgu$#YfMTc3cz+`D#8^X=qcqp|2sD<#hAwA|aD%}Sv)
z9uglO(h{wN8N?alR}BJa@hlERG?MIMIET#b5bunWY=~D*AVFxH*}rYj-E4Wv9Yh6P
z)*pUftCefNTot^`-kr(RgE6Yzb;exxv#1Ici8crZRs>DtlqO*)H1ito6v_7e4?-Ej
zzRrYzQBv59D%hGJ2PrQfPmxO*OJT!m>53RM5FMnA>3(eJ#3N0pR`0JXnXqp`*BtOS
zfrmX*Or}=7z6!o@v~F@3ZDG9jo|#mVjKR5WjE`--#C2y`a1gl8HT=J41?ce8{Qvng
zzv3OibPE^o>E&eQ9Zq#c>?=8lK)68v@maxuXn%d`uMCO|$Mq_j@+Tk6(uL~%S8IyF
zWJ@My8HoE9f8hC`6gbuPdFPat9;u_{Iq-Gqb>YES?;zIYdS|{u=OIvTS4Qac{%8(U
z;QPDe%ejH0H1jjVFVa3PJ{}HS`uVSm*9cdEk={RFLAR1FaqmvH22P*0H`?^*-_@K4
z6a-m@le$2Iu4I{Tf^Of}oNu#Am{=GSky!*CKbS+Boh~Qy^8LVZ`va0jH~xMv{tE4j
zFQh>SXQX|&4_*F%K23**_@8u=9uiY_d~)EI#p_R7xc@UYp1Ztwzw!q?pS}>PRvr{5
zBKRV)Y8H&f1IWk$7g)d;JFt86YNOydIvV${s(%1%%lWDM;Q$^hBTLRf!h9}`1L(dV
zzu6L>=hqRQL4b~GH<nd?C_MVjl+ym;I=P_z6{oyt@5**GWsW?BT`9z&7QgsT<;%6?
z?%cxqUBl5M>X+AryVF;PR}tU3!w>FioQg-wOxI2!03Hqq+#65+!OZRx-Bqx|58`Sy
zSweb0g)G)4<{V<x(GYlicipLw{k*EGv^92lH|v`n$FLJa$c&y7tX5h-j>|otTKvyH
zLvQi~7OH~0^YbR{i(b>Os|GD<<6m}6o?R{_?`WSjB)xFITz@$yy`zMu8fp^7s3)80
zN+>$idL~mvD}(Ke>QQ+5O4x;#`8XSMJAeOnnryt*LAg+!9T;}xGt1Q{X%beb=5497
z_(nDxksn7e6%jo$HDNZjn<d8~r)$;rrhJsK`+T0%t*-o#idIa5Ve$jd)(B@#%b$&V
zlUL@T^Ms(wp}NzZ1j$X4XWmzyZp<rrO$3azn|ko3qQiJCRI4xs!pRHbTX7ogT0^gP
zk|izrM_nw#YQa;Q6{j#Z#0Uxt{gb@P{#jmo;xC+)wGx)O#bz)~qo3qSoli|ko?FV!
zntK`>7xbCu0#BUfJkn|#s9AHVEnIvrc<p=o_9FQDV^kdS-xl$op!4xj<L;p6tz$Y<
zlap$s#+#Nt2eJ3xHAxB~*2x)S!gYtvIzs23IO+oNNgph;ig5GcF<(kqO<7|;egTTK
zRL7r&wiY*>VWHL~e9|*q^~3I(@AFeDRLM}yp6}3W-CVctHxCs!s}KoOP}}rEJ)X<!
zzEwI%_?>;7N&cNLV^Dw~P!d~;S`(p!A|R+EH&Jys4V;<N)~&|oKrKpuV907F@l0mL
zDKFpWI<2_288wPhcsvgB=Tp;awu<|=F!)3YpcHdv&B4(=TK#~9ra9E+rDQu@*Q+~V
z?Rlz!HrZ&$x_f|MK&7~tgoLugSYfOw15sb8uUrR~zo}#4n6%-558w9TdAi|cOsA$}
zPJFSNsU>}DgfU_)EK?~<dvfo%fU3#PQ2yJhg3a0PCb>^0TaM*Z8bcGg0Y{aaG%2Eg
zM~T1RuPtjBXZ7NjJG=~3cd%XDC$(9Df)0s=+kmI($S|4*`h)#szM+=BWkeXAqrDVy
zHfO63uGgAHHuC6owi>P%DOm;z3UGKJVC=I9bJ(vJ7+dRS(oPYR`k%s^hv#r#JM$Ga
zG<ieCnz#8oq>Q=lpYPFL;PMb^?jY|m%f<N)E;(^}uU?B&U+seNkKZ)LTTTCiD{X5V
zq|(VDF(B5(ObYGr5TS}p1CGTp)6^bo@>x4|W8H%(X8lV6Q{Gr0NiA|Q_PjNjgQh$$
z2TQ>9c&l9p_TC|a+N{(D>xU1uK>Y^_xGR)5|4g?#^hn&0`k<>1G&Rp6?CH?+*V5zB
zPLoZ~QvZSro!&eo5qnT;pWts6exs^c4gDx5BY0QWqPet6*<L33NkAAJNkjzcA;VM#
z6C<1WAE}HDA>(W<9cCLb1ftnmH1r)grAuTK{CFfOJG6}|WDvm+nbosMF(~3E{Zmk2
ziTgbU^e(d}r%Gp^;U|VgIOnXb#qb4WNvNFS1Aw>!Bm9s8GEbIbQKIo^j<rmt8T0XU
zb&JELbHLrdcTgdHlvtmrlMuqcI0n@jkx7HK39c7w2{)OmZrtkeO@Ad0Z$u#;+eEvw
z^RIY64;?5T*L=A&IxdRdOzpMpu`aBoZz6Hv)3*i>Y7!%>%TiL-!0>9P*vpQm^rXNf
z$R^L*E60DBgHCDgAe+=(f>BBv$WAFg%5-BGIP`Y4_XaWp)~1Fi!h3i1XQ)Px%hUJk
z&QOOw5I!@@u!RF=iX^F8P?Fz;QGn#EH2X-94eE8Zb;;ErWI)?ue8%BI_G&0%G+8o?
z2<-Dbk>~jqC*o)zRY{oKQ5&1CVt3%Jc_FC|>g51V=?F}1;(8jQ%17GLl5v%$S#gCu
zTXK#4#g|*mXZ=~p>*aSQFk7dJXOmIMb`k)^i6UsWWs=uXMO-If@NZ&kEn{TI^{)O@
zmq2P+28xze`kPi5xgpA4GpjR@xnzq31*pM(lK_Z-@+ki4D&K&~H&cJF1W>r=sg+ut
zJpKGj>1VU6>aV65lbsiq*`XMu<7`PxJ>McQ2jcEil@8kM20I1Tc9uO!9h4`Wx__T#
z#gJiP{Nd`c+uOmPRv7dh7$b@5-^F%Ug0aZ)9La>v9iGkiJu-V#?Mw6EAi(SaJSypI
zBDCOTmW^x=dWC$I0|zjOHtPd1X7q9YuP*c6I9RS2WvPZXlRhQ@1woZg>J*!P!O|c}
zgQe5&nM9X7MDc-p6wXt2y20sWcA}9)=fBFlPe6XGFa3<5Pq|A591tzf^_`u~j6zGF
zHpIW3OI1Y14D%F&x^CjLof28@-#9hnWaUDW6{G0+Q|0LCD`(S-h4n%|!*QfB!IQ3d
zY5eaLPv8IjA!dWS_E^KIcjrx*kfks@`x-(DBaGvRBBL(dw;a*%`P&%TOOu6@enHun
zqLY5G>^mE4g)1KS*rpsiFc<zV{LLYA6{4NJ8SZ9$F|gw88~E9&KK%m#2%A1gp6Uy-
zE6qn$GCn{xQH-zN+EmqT&hrSPyAwkD=8zmYM5j=P@-2aLv<;kYw2-}4(H7D389zjh
zzJD{~cD9d#(F$xPPk0dkN+Vt^m}pCi(*_ZyaE8C51VvmkU+;J@ej^Pb2{mn2R&HZR
zBv-*gpiOBqg}B|3ZOb(oR_+dxn{PfiPOLSW|23oGOtDR*(S&vDvP}&#;HTQA2G?-n
zS)hTn?Z-9#trtdNM59pidUj&a6$t*%HkRnA*PoavwI6wOITBKc=h<OUcG2UCc2DM-
zteWDfDrc2sNkfb<jX{ejr|as?XZBHl#2C21oHlXBg+!~Bo*ZupWs>$t*|(Has>9?#
z(04+aR}nFaTil1_(}oI7b?cbcuOdG)Iu<u`2Tly)N70)$m)#=hT=-?UP9PC7A(<0_
zs(XUpyE};N8|WvZ4h&NcR!I#|_0xe%WmyAr@5rL$fi!BFmKGQ}9$cd^8Ie0C3M}%!
z$p}n}1|P8yfwiXt;;N@Ph?mGrAzE$pu@^w-Q)09dU4$hD3&jH&46=7)*}G`1lMMbG
zp?pCkHL)NRAV}6tDeW(>-&2#_q{lh@Nx>7LN2-|&#+|0b!a{>Lu>c@-5Hl(0P?Xi;
ziw2rELov<tYzz`mrb0jkH;MFL0R!VpVR9dURq}dP@2QDW@y5@1Cz*GM8SvlH<WRbQ
z;$a&E{BoKfP8_Ewo$>zne-6E{i$hmT8j*}A0VvZeJ#8e!g!R-W6%W~yC}6>i`Q9y(
zOcVM(IoD^x#S_3k1twl#wWtx$<Ks(G5-{V*+j-cPAp`cA6(|Ezz)I?awM)(q%%C$g
zGa;RS$h3FbiRx%F?}TX2>Dws=g8_+f&LG0DI=4x{KhixIuP0U>RN9CoqhmL21_*_e
z6P}!$B<|~*9}3k^)bpBb_O=f=T=<fn!|r9}%Jc_1v$fb_<<SV-M6CGJcaOc<!aOT+
zjSq}=SDyt(Oh51+n~gZnya4b6R*D$g0AZQ-Q3ia<Gd4u#bU5JCoAI30Of^6izHfd}
zr)EXSp=S(jh=UGXlmjY&^kIzR2$wD#>k>Mj81>mue0k|2Q?@MV2!A_E@8fEG(C6EN
z7HsKTs)U{s<$B@};vF`@`uGVLR)ka)5tu<XOH<X4@uO?8XvF=ZQwBkRY@`#)sURCu
zJiXY%#w+Zc-U341nwY==f;tZ{9qK$>k~spp+a4e$3^nS^xEy$Xzi><65E&3rq~U~L
zvg)#RA{n7Ysi69s0!|Akt8qVW=xhfoeR$8!TV<atbl!v4p$LY!qDimHq06P>LCdvb
zc-TpH@b%NpVq@?S!Lm|l8Fqy!&av{PUfWVsOpH><$u(7EerPi6OHm;bZ94NJ1=CXH
zUp}80SF~rYyq9kRs8pGQ(mx(1p39Oe?<2V(DG5}7<x<hSng)N1lORtJf(`=d*yS`s
z6vSKQo{7jx>*4F0)ICw8)uL#vX9rpCB(LJg)vE6eDXL+0Y<4LY^H{1GJ;uMcj8+Zo
zjD9`I)7}zUx`&oG#GnW|VtSMsH`VVf3mAvg{y+Q=2?~j%<w9mM{)Ph5e2}wfcwj+#
z#krHt75n}|Mx01zLBl*UWX8OD*xjbo0p0uT>}AUYd3X;(SNeQTgn-6F*N;jWwE%=@
z8W{~+0kYG@oH00iCkLJm_iT4nE;aQHpCpF+Rx;3X`pgyF0wrv`Pgz)PRPx1yS+4T6
z3UX@mZv|&|>MkLQ9c&-<rLTf3lKtDsSC=3&6=dh&%tc^I4ahDwj2oGn6y0E_5^okd
zhbLuDhl%q?RDLeZjo-?<8;V_i_~v&sIY|V4gGT85`~2Q<{Q)@U-3hw6O%N8>Pl;7V
zI7I$oNDpmBDMklR8JO{gw|NTXKN>rfg29OZok1RgI8dDT^T3(K3R1|f5QBvUxBaOy
zVp_b8uGu9b`Q+l~{D0@Op0*A%^!_k$K9VxRwx%Z22JbwtF0)q{diMz+PFX)Jmo-dQ
zX~qBw24!UGPnW03;criY7L?056-`?ylEQWBcv+IzLSm6Hxh;u&M?|2YfT*$25V<7q
zUUuFR7Oi-sf`VtALKeM+R>q2KsVQHBDygt&lekX4ho@(6PZKYXna>6(hh878dyZpj
zntU@};7H8Q85>z`J-ZdNRl4>bNw~C)>z-%}aHf?yM}3yCl0AGTgkYe552C=x5^k=m
zo7%CxcWua-K+~5vXOyT}ch<hm{?<48fJ*EWx_aNw>(eLSm*SuI3-=4J{;$-zq&h@A
zT$UJo7{#g`1qN|-7e&IgQQ%*=T_{@AMwgu>{8}h$Wf11%zke~(MwBP;QU?Io%5D}s
z`y4Laa$x;3mf{)Y)>~({RcBNVgIKLSsPBVa6n=cmndj0u>x;ve%%A54aQNu<b@?@&
zvh^ElM|HnvQ#d{8w2xGRyXApACO)~?;&%osYh^QLaPCMa9OhY6aI}unbp>vV47H$N
z1%sZ6)Jl1~Uj}nJKkt)V-iJlQ+68{}#8I~Dw@1J`D*>hNmW)wIH(fEm`@B58?rv;K
zKIFXGybOKWnk?6kq|-~AM>CCFqAZ`K=puopm&kAK$SG-HXT^de3JW<ga+SrKhH`jm
z77UGu(nqlk_wk7vmcNponEV)|ecl!|qKpG55D!v1pRURgBj=Q{5Q^5H{w}@IL{e`J
zj%4LBc@Fo!JAwx{?tOb^{SxloeId|;8KGD=;Xxjk)GYsL?_d?LnU0;I2Ul7iB_vxO
zS3Gl!B0o3J-N<E0Qg}P1SUbLz3v4=j(+o6zs-qasaOoV;F!?cRX#}=LX0cq&G~Qo2
zRPu{ccIX_Dl=xzV42L2+Vw-7#ZqJj2YTI6K-krs6JTb=*+<eBU`$weHuCAt68jWGU
zP<+8hxoPngPn%py$NaRUN?}AMb&x~=nVPwIq#;$2kiqNYPJk>HPF$j0pQzvoP+JbV
z7863p8N<f9f{pZkvHxolX3JqhIIaw>=@)7r_m;h7((ZoRhrMsEw=dZ8Y3APFOnDj(
zrND<%R}7v6npw<r{dlvd4EtF1AapH*@|S2djtxTmn~y_JX9tN+U+Y}#9#Q^yq@o#w
z<(;QK&e0TIl3C!rgLs8IFa}~BR7{&k-nLV97#}b`ot;OBXs-rd+T52)-j|c9d5~N}
z^43DCs3e*|r0_IIQ+{(kiL$*P6!R5-+CihjE5}RcY?AQSb$DSPs|{tkSBXc`;6k%g
zY#UZBgEF$UwQ7TI+j6mL#BJ|l>J|O<m};{vFoQu_>hG-!#a3H<z{PRc;nD5=-^o`V
zm)9%GU16gYj;wLc!HVGJYBnB~BPx;Bu*)%y@5NL5yGY`lKbc7M*MtPR;3t#5c8}BE
z;zPUG!9QIIg2g4WnI*-6)j-Gxk5RYzMwNVI<l6T<z~?%d;r+P(WM25S4BGV>hPp#G
zXfpYI0ohGI1kHeyNq@*g%muQ}Y=Z7{#s-mlglp6zljp2WPjg9eXXuacn`ihW&R{U$
zQOnj^B<|haK`I`9?aJndIBC2;XFY^n=Kh_`ZC`I=B!oJ>eKJCHIxCt?aFPiqKi8Cr
z*BuTJx^a>)?p-G5t>HJ>rR_1MZdaT5<$O7$G%5OKh~c0~>OuFQFv4XI*s`lVpE&Q*
zpgB8+2Olb^Ku6H>Hbcv?4Iwya2w-tw?_blnW`upj2~6&rt1ia7E8p;<N%yQb=y<Bx
zk9xM_hc5(gEW+<=%x%2KPN3#G#Av3Ft%QIbhk(Phm9Q2@K|A#U``Gz+>8Za5D5}L+
zo6;xb)Q@m-GHk`#Ck{gx`-4n}ky9}$5+g9rV?M9UT<5n`ANm)BNU|LmkC_KxKoQf}
z%5?iWN|!#PB`wuh^1$?tbvUgpn$>?^nhWeqJ~c>=e6NDieX?)1g(!`6T~2;LP<DU_
zjIj_xZ+{<=)M7y47B9@y8P6i6KnA6}^v>am7_5csw!tkUx<SQR{ww49W#g$I?|wx7
zXSV&1;Z?;Pp!84oJ~=)UecfH~Ia2v{wl4X|<(Nhov4mPW(|l<RWRWzy{JtVs`B8kk
zCz{fJM0QQz4yr8sQJSsp^cauHz4H*>dZrw=nA6kLN3wt#g<p&{&HTU(G>1y(Let{G
z;NVL<!tw#zYs7-r*(AGxar}UPCpag^Ur%2KHy9j=>m2ZllPZSOP=f7*%NG5dxx>7j
z*SCwGwVM$ouTUaJe6#uK7%<9q*uq=z7)Lu~SCYvVR7W8n@+{B_8GF!KFl~Di|5-M#
zkknRthF%2Iu~2XC^Ri`hcB=CXJhOFrcqb(Myro`$-THQ(XY$H*!7zjz8BZaK)y-Lj
zzSOvNa*fU^>QBW*3H5oZx3RxV!206E(|6(+K}_EfH6M`&L)+nx`v&t5ou(<JnjnX?
zK&o?BOHiZ?1$R@Z;;=SI;{=S}4{)c1*&++F3cF13l!3Y*6JBxrMg2JwaA6lz(eP_r
z#cHr|c&Ir6=YFI%htJD&$G`|nR#*$~sy;FVFfie$Y>h2wBWi@JeIsbo`7pxOy&x&8
zRQRSjIts}TeHCyM)or1ru8A+=rgM+x7J~WqkJG{F@CT&()&G`DhCvE>miJk@0Y|h!
z85{Q}uUK}!%v@Kn1!j>7+1X=itzRQG+Hb?zZf`&Pb{I{6r<V4Z&LrZ?k4nIIz-sVD
z+8|^*N4Bj7*A=m23<hO52{G%@$#+I0`LWFAR215U7B^bVRAdiSrY_L!3s00Y&MMK`
z@ys~ta=dJt+{G$F9Ch={PNhUOkpX=SwAzwOjYN}_{vPVOTx!RQ9Z?JWskC`y0^nZ&
z9V|xyhm-N!VL+y?aB<^yN;zc=+_`3<x=#N#PJn}l<r3IH_)s>aj?uUZRg+J|>LHxY
z=*TK4=6OG1YLj%X@jC5t)}wGr?<56kGH@9l;auBJvt|$;jdUO}xa*yu+DEjB>mx+}
z8z4Wpttbm+85NR{*d2EeygL;A=Kge^y~Jm;@7MGASPp@jIjYFuNBcCxhU2WY;trN+
znE2~2dEVyjAf;Z|PFL=GMq3$_F2Wd^lxM#?ymY<(G3m9Ar~_ZEH!aUwwlia-vJ5#h
zKH_Tl{<|hOX$#D|#mS#`ZV8J;=>1*kK>%xn)Om3p5qqDkvz~IUv@r7ulcNL_GRTr*
zd<!Jlq3XC1y%dkJ)xXX!M5^Wu$&5{2@;E&F&!NR`)&<I>^K`00@IGROe~~{E9H<xe
zFkj9Veh%h%JDcOl%Sq#fdLtLbb1J{3=TA?g84>*(uG_5cPK%1FB5&Q_wf5;k&(+&I
zq19zwroXt<DiYl%oEJHq9gShkMw>5!j82do7-IW&B$zyc{N24NJHlz9llTt7`t$lw
zz|*-z3eWzi$pgb_z?Ou(EZoGWA294%OOF{mf(DGh{G<CuSOXKKW$Ex7W#O0T?J1C#
zdA{dp;q!Zt(NAl3MvRdyQ`m3}VxA#Mig1i$=JH27Edw!S)T{Tm?uY#H2Zm_eTA9O=
zFY~y1-|SCTjm|Dj9_>DpScsf92q+w3u?pAC>#&=rvJQTQ!-Us+!q_rw5Az=G$C-=x
zO<J>d8f%~RJ`{%1w{ZQikH5f!eJ#=D2)7d0C_W9J@=XPF;@WLo|5A(D|1+VX;2K;P
zr^-Ga8ae56+Z74_W=-yvynX(i#DS7X<jgQb1cfqY=aQ1*4m>9EgVb-;9y65dBT*>>
zoa(`MTPX?wr<-i7qg*|*kX}7y`CR~?yn8B-Dc`-P>Nif1Y#DFJtzDw-y&2%T(VXk!
zUFhb2j@LFc8!j>7_2dsf4$SjGS4{POe*H|Ni?#BLrsvA|rTw?U0sxjJq(8x%+77x{
z(chSDu_jPE)J&7q+FC_rLx!TvfK`c8gdJR>_lU(+o|bIzWXQ-|PX_=dw4_H?jp4PL
z3<!<%jTp_V!`BlPTJ_{y$Ui^f>ZmxM#@D6%><EiH3{x$AJ4+sVgke;jl}4!AT!E+A
zZx;BOuE5Ux>v&apYO+##k@7&JlO}W;Oi%xQd}!A%1cxGpnDZk)?TMt!N0M^vwRtll
z#bdmV-!+jc-*A6LyRP>2G}0#*S0Rs<6<01mKy88raNX{RDxprl#Tj>KCOq4M%tdoL
z#hU2~PfF`cxdYX25<()4+l0Gu1=@?PX>_lR(wae?u~_nSa(OMh2g_G-)@y=`TvJXV
zX}DUNUzsXvabNIC#e9b$GADl4N=0C)NISb66;<(*hmDAp4WIIuP#*sGp4d_+{vKp^
zctXOT>C{$<1(TzW{5l-V^~OqZ1$D-`)DNf0INd;aJMXGae`c}j;8M$0abLNZc#wQe
zWAGWiwAh5pyRxS*$tu78&wU=&7$NX0rI;Ue2wPj!yve|${>!h6JGLqhqm8U2IXvPe
z(4+#YP3|0ZSz|xmH2t`W6uZ^Z)|p+dtjJX0_CTv{!xwkZLdVv8j&rpozN$)8{>f|K
zdcojwcIj>4q6_wE^APXa{k)jx$Ln9#f=WqHdF3P`=5#D(up?7Fz9<?NQa!?cC?P5^
zcM*pA7M+eVUJ^-uTPSp!DMT(Z85w5I2P#nlA2)P#RCQFH1WdN=dmHqn;5rm|RI`;G
zc`B81fc8ziT;H7t`xiWS9tAEm<sZIofYL|F8vs$twt|#+38azzR6EZ%Tb*KSg&-m0
z&#V3-3aBc_{t|Zw2nq=_4Z@;k-tQS^^|xzhlFGeCar3GAT3orgz0AZP-`P0>oDNog
zY5uACNZEaPCb1Fd-tX9@w5l&6t4~jsFhQ++edSd6$-Cx}LcHRvZs-S2Kfat<Kvgi4
zef+3e2mpLVLz7?Y>)LE>sT{99d3OdsjE+}&t0)MHHqQrBub9sg`QPr&Y|{_)K=j7%
zS8zCyASEF$Wp?%I)AQ6e(u9!~)_Ks+*!o%X-_nWAku5|+F8U>Q&JNMb*9dqy&$<<u
zzWV$!@AVzvl%O1YB0?$^t`e2-;pw}k2-w|uR1kZ*qX}x-T=r@7>ZghB9lY)I@EM+V
zluakT;)a_4l@_J-`+)XoKdQ^dly}c7j*Ay(kdS6vy@91cS@h3gDJp|BVqR?Xr%g|{
zhyy;uneY6@RP+)5_+D>bI(I4lqd%%pE~t_Yb5&)n-F0m=^ev`FSg0}if5p^)w9kgK
zGYePM{0$zeqlhpfZVZ&r%MeA4%GTp1d9ygVRK~FQz8&7~$1@;CUl4!)0anU2fVg2j
zboy$&M{)K_X2aW3U;fmeN#5wnhtW=+!5O`P8=;u%$1H~V|Ksee8ro{xC|%s$-QC?i
zxCSd;oZw!(Egm4aYjG)7T#7pscXy|_L!p!RoB0KEnzL+q_AP5&i<-&fX<zaO^1u)u
zD<7-X@kT_bshl%e2x|EU)43#X|KDh-sD-9(_;>c-v_fZ58E{Y}ZcD8wF(&bIoE@B8
zxb^X3v8`e-u}$f%g9F9Q>wdKHtk4*RVg+x?=RAPxGl;pKsyUk?{3bW~71??h(9%|E
zClf0Hbl;VSty55!V_obsbxxCP7u*9tQMlBgKwaP5fdcvv(8}zzoEwLt<CN;_tGI+q
zc_<xUMe52KWnjDziRic7!(|7++yW#9WekdxEknbsrBlHbcs<Y&kKWh`yZ4Vmn2e*E
zLyGqZ9>?EjEexEB?F8atdXhB~FM7)@5+y`(462;I-+V(Q%{>92lVXVaXVn~33vJqN
z^t@qh%0(#_%m~unfr)i>g%6mK;2_dVOXj??_@|3AIABX|9+BNj<o(;>R5-u5krR?5
z>PhbPnMO|BLZ;|x#h8yGk-zyQg-zUUw?(C1UpLLa+C{o_?$4Z%#80K&1#O#Oesk@(
zx9R#fBSua^LATb?fR-nCxBi{$#93y;mvAEM$4I`QlUF>#XCI&C*1TVPFosx_3n63Z
z_PNtu64hu?I;>(`YGxOsy13=KJ#2C{d^hXC6K<l@it1d@P=RGHj5A5-m+%>uYrUh_
zvTJC8%=TWficcL9{ps(j&{f@`*w!1S64YmgTM-1^c_+=}4|qo~#W1wZgnfu@3>)Vm
z(I8Pn{FA>j#eSwYG5NDM#;(c5pb>+|iD}1T?cQ5t`MBZ(g{J}mdu|!CYF}xYceJ#|
z@kjPUM@+M>-&}!v&O^x*ag>z6B{oUXtewF9p?4Umx53(*H>|Cl>!RT`Rq_){#XWK#
z7t_7%L)HC$68WzR$f_pj!eMIWqB(`JyAQqJoJic9<*Wp^$)6SYE`K#9;^95-F~p3k
z(6$MqI#*N`Fqzkfm>G$tGcUZh4-P3M9t7^$b&d2+rVMLpT;E3^MaHv?0jmGKZoP|G
zf}V`KuN}h@93+iB%oX-{zTRKHlWu+S9ia+5(L;5PlKfc=69zQosb;OwMErw=rf(;Z
z-No2S-rz;fmFAQ~%Qguu3?H8qgV~U7#0&5cl;@*asrV)(hVGW@<qCiPptzJY;bvC&
z`xpBB*lB6m%B3u>gHekli`PkRnKs|&{v5A~kHxngVO{TPiT-paBfO8W)V_I8ZSmuG
zG#DhXsc{c7uSo0o96;3QX4TqXOj;*eb&{f`7njd#TbVYE^^<WIq9D7BOY^1OPbCyh
z0y8FzMywy#HOxO7c&*{y`p1AXT<UK?_jq;oD{(F#DH*Z88l}X}jKc)imjkTph3m_s
z+-KKYpFN*Zxesi@DC4IB&8-t3KYh)P)U?f!?1-!sv%Je`W^*}*ki??ibeJ=dO*aRx
z6hdj3rL}O3!jowdjF`wOWutC`B}b+yghpG*#v^?Rp)%peJ*7z3(WNngzF&LMS}t!E
z&uWwHL~78`eTI~U<Q(rNQ~r$^e@#+3zm|LBQOEy`Qy9Ert3ek~waE4!|F1=Aikj`|
z=*!{Jv6v#+BLC#=SZ_+9cFJIeBw+D2f)yFaVNQ_PW!JO_(woTyjQPz}-`PK~$fpn<
z`KRRaR-vjTHt--8Ds&u<8piU-b=@4aknD?eofBgy3ey_mR|)>zO@C)@d7}M#dFT4`
zIv9mC^~mW3qs&n~)28u0OHw$(Cio7ON+RZ`r4(7c5T({C#9go6q_L4~&_mL0!(rSl
zeWjx}42CjNKYHZg1XKf+#~w)M(KQN2<<X&@H8ac+aCJPP%D;y(JyKrx4@T51{aycP
zR<Ln5Ka;Q$F;QRoK%s%=s@|pW>eci0F85ys(}k6Ek!8dfh6Bj{b%dLZ_#&R=HLos9
zyH+r485>pcEd2S653%{6<fA9yMwWe%K{Lnm6Jl)W#;GV%0V9F5)hVPMewJPUJ4{f)
z!!`3z$Ly`|ci>s})=noAbaGU+Q~lA5aSA<h7|kVCk-ve|Rh)9_;!3b(pY3nSr+-{e
z0cBq!KcrCdI?tpI+roX0!29xC`$n)SCBF0h<z~r%BcRaN-xTluTXM2M%_psmc4Z6n
zpfX``N!h^=ZOxH+@NAs)`l6t>Jz0~=20&6evLGeI@im~zhm5lG9}d=%fHgsFhg;0t
z90D(?mV(8c@W2A#z|Z;7u?h?AJ^AwQci{&tm}FwGXACQ?pWyz>&)wfkSj^`iocVJN
zXel_Gz1(~L_L3Ucw0S>rlC|{mi>Urf*|3aHog)6h@-(~_*W60s)ds+I{EhvsXBCZ(
zU61@`HQVQT(PwMN_>szUovrISRjg{s@tC(1bn2~1WW3sJGrVb8@0Z#Al-oTLaOPUT
zd_~Dy<W3@-9=-YXrL1f0@{{Zxg{E8&0?l-S4c=ixR*lboA+@v{!kKrz%+7t7lW?JB
zkI^A>=lMfB>-Cq5*7m+BJmczcJfe(h*ply-gjqrlE<CJEIUI1WMJ%D~ZXWz)uCyB~
zB>l~0h61{_vOuaUw#&a|;_gu3q+86^>)<DA-VDo5+pO5~yNo)4zeHc}Yp*9;y5wXO
z$k1i_%@cH#dYdeX7O<`zeGewRt0u7z@*$960;tpL67N0$SE7C0IT`e;0418qTh+;9
z%w<4^{Q9Mu$hZ9pfLe2@AtEAL;3|F&r7!y#_yDMums+01F5gAF{az~f8IR7m{<7C@
z`cc55-@;`6Mp5ry3;6Z7i^$BXn{?VTN+sE_fR{kbVL&?G{t(JRp>r`4>+y>mLy#C*
z&WvL@r^XQ|#6!DaaE7Nr%m~==olEwXi}vp!8WLKex^%y+vJ)kpX~BqDiJlg-uCrW5
zGMVG%MQ|^8PQCnx5cekMqVMw*F7!3gU7(O3Az%T&<>3zj!oR!p?!J$YwM=i=dLI&2
zASxIu6O2ka+d|PxpE{I*anqzXFeybzP+`%kZp8>hwy_I;AyVa8Hmlrs_H!$uij>9a
zaEJG+{Ny5KHdZ<B?>}A|UNGJ3Kk<Sm`F^mLeva&Znfy8lz4djU*0-@b-(US+4_bh<
z=)YxFpRzIW0uaY&c&7RDy(2&cS|o6M`4EEahMwxspc-gyCqj6ATCZ=<Pf~8ui`~w?
z96pRs+G+pOxZ;d>0+w&6^Y}Ro$#qs)ChUg|jsg!Zey?XQp(8TzfILj8X|zv6W<?RL
z8$0)0^Zs@le%5R=?i^R=JG}2rv3>OZ^0XZUnMf#Ksa5#9^6AM>LnEN@w$W@S)y7BM
zT<YRy!^df*DMnw()<=9D4B~S&k`%cYjlE*?MTH2wbd8#H-3B~dvh$$;(K6qaKi}%d
zDLMU4UL@lmj{<!cOC`V1E;2U~m$vmG#h>l;H&D4Xi&$}WPpYnPxXAZc6K(B^VSPa>
zk;wHt20fh+ecJb3#DnCk%X^tC!ae+kQ@~Jj{g}hPxc*hN$asqR^Kgoxj6IU|4~rG%
zMt5}2*ysv|7hbS2S$b#x6U?H9{UmIY?`B6pC{`L7VkizLh&S0H#l|SEM<g!Sat3Yd
z(_ak1jp+1Bp|%1xGYU2PQdf_EJCUGKVdE;~Yp6TW#>UsTvGAN<9ikS2_b^YHTtRc#
zH?$CVHv>D$T09=rPK(A6{xI&SYA66gH{#vL6-13{Ye7qtKl1B^1Vi=$_-eLDuU~S>
z#A^_?kiaBHyoe<X@xg@+{G2y$4L%|!dG2L_v1==o{WXX8A3q%iJRbSNnI2^d3A8*u
zj!2Aa3ax)_FYV8F?wyazd=OdsM1Biex!E4gaXH}LgYh0WcqUIAspkT9SY|a4;Qw0P
zlyo&Q-8n**r*}e#0;8(Nea>o*La)KlTGYZoD}BY2C4w2Y?vRr*E;LqLgZlA9xG4Ub
zxwPYCei^rgvT8hH#tj@6JWRMiS!aLH`NmY-TbHLFLXwn>#ZXPn2Y4wAPGhZfKOfic
zzpo&ci7Hr%IMI36CT<6(ymmq0fCKc*O$+=y*2`KrFL}oVp`+kHijfGBUjbr`&~%XW
z<#M-dh9D@28LuY((@y2!$LR|m!JM6sbd3Ehij{OGkUOB;Y=N`+oXjPMQt>4;tY9Qa
z!Jc=_GOV{jtS+Z5@A5nQ=maGjQjDbCG!kMabz$Ofl~onBRpMg8Z8d^&N$2R~PKbvA
zZ*JUjceC7|IXyCr;i@=IcY&O0$Hkf>)qw61j!#J|a;t;Gx?(~zZ6$nhNst64vUUH_
z{yhX!$19%vK<Zjg$O1a=ggA0P4srHvujObR>DA)ox8Yd34P%qb5l=Ztx-~f=wL;Q%
zp*em35bkwE4F)=sn40IXuc)P{GVf?|Z<tF{so7_ChpC;Ll#ivK_5|K&!y>I`r)yb<
z_ISz~5NH$Iy#WkW*(04r2$R7B5!kTp6U=+xfVL!Qbdb}PT!`V15u{<o))^SQPj2fd
zLAu7I)DxPts)7H-X|Vkz-;qCtHMTtm2)10%QB56IrK9n}CHA<L6_OJx*z`7?NbU-}
z)mse{273P^dX{}>@aXRSLG!!Ds!O<pJwiCDb#n)Oc`U&$LY?TuF+Tq4cptKgpzp2P
zc`*`=@+@4jAR!^S`w@%kX3g2W!n0J2Yx<6|*Tmm5R+_38g*ew%7=Ez-AR@VePxq1}
zHWJ+j5x*ZYZt$S^y5C&5ac6D4PSH7bx%P&a(X1@1!n$`o6I3-etuiy5H_SAgvGc8i
z`05GN>!G(NUk3`%AaoSPnbi2WZ!4{DvmegLAN-gnzVG^--ET_HHPY#HzK_k4Q$2zJ
zyB!&bVidX<f)=gVfN<|5Q^IaoKzxb8GDc3pG3SA_q;M*l1iXrM>>kK|CYmmQkuYCq
z8b7on*2(AK4;r3p;ImbqP>keDi&Z?sj68()aCdonbM<DThzDVT*1&zQ>^NxnRZs1d
zvE&2le8$jVwbsG`CXB3h(cRI84{KqHW!FXKL|Qra%0@2tGKH3|hzkE)FC{L2FAG3r
zGbb?id+s|;<5t32jyD&19%UjCTu2;paPo7+Bc;Ty@uQ^e3JGnN`kxGeyp{SgqQO7D
zu>vBW6M(uRy(o5Nih)Ph5}T$?@M0f^T0LJLD;53&g#UMtbEQqBF?RBc7!i>FlAC<7
z>MMdR;)ga*ayAjmW7E|E{@;su(1%4nF`FIPzAB2WcMNBdn4)RgX|{^-LrU?)<D-ZV
zC!)`mN^`1g*+0TG9yEGZmF&0&jBf;{+ms!@T5T?qdCWGFS5FE%2oif@G};b?wV!0h
zai}mwRB6g#53b6AqsP!kt4iWHs06T%HTekZ_O>rD7#A>AU<)AgLDmope&#mkp^lX|
z+;$50agtRZ^1Mk9k?NP1A)#Y>Y9oV3iH3!cqB03hA_~d<dDz&Nz>dH^m$bS0={vs9
zK4BGOPgj3X>>B7^daX)VRMjRDG-BOq29F0wyekCIvsOcaX|!YosfW{3et#%uDAM*z
z4j<kzc=fQ+Xznl^HEC7Wn*!<nA^I#(9Yx-5<?Eomta{dd8NfJS{?#>Xa){r#>sGfr
zGXs6?Aj}Mv%Rhm5>xUUgFDyRiFt=Zn*J@iO7QHG3ur!Ir1JxxvMn^h&*ybGX9>TWQ
z_yPVXqYN@LzM=8O0`m-<{uc|8kb#2i9<%@?-DcV}*(G^>(|G-~lM#>p_5HFNskmE?
z_&1z^9-xoH(tM!Rr6V|&x!TR_il;w3PK`~_$Y9X1wRLyz0FP173Wx6ZAx?jy-4B1s
zz(9pR{u=QYeC?juZyf+cB(Hb$(C|vh|DFNNeUV-3JrKN#^C-PsVc~REowc>$XkKiG
z{}E<d(k2olxO7uGaVA`8?*6IVu*HsCAc`1hFW+G3?soo7sW=g`=N;MpHGVigfb^sL
zD(5jV&K=9;JeC1jd|6qp>(h?av8*{GN5RIlGbT%0@82in?@XtEKyM|MON)vczW|}S
z%E8RB(cH7q>nZ3rJ#0%g&>YU4<ENBy{i=_SKdQ_L$nQ&SK=@~027#UtMWrJxMufEc
z^1ZSFi<>tyCp?<*Mn+PWbKdNG$nzxe8!R@)T45o6Ej<*j%(9d9=WfD>P^jenST?_d
zx^C7c0t6TE$weh#6HyShjTavufZwK#-qf3s<6h9_`9@l!i5ik2-B59@!1t;4WP`KU
zLGUeyDI2ez{Dz&bg_|AGYp6(1evQP*Oglr#*_@yTFGyh@<?=DIY8C9mQF##IgB3oF
z+{+NV(eMnX@VXRLSMrEiLZ@KrLOrjz<F<E#FeIvI4<~I8kH<GwwGTC6RQ&m7(=e13
zeeWN-7#**vVZTm!;5I%+<*_(bVV{XU=go_Yje!|{<|!+=8=DSYDlmIoj%G;5Ct(Rc
zcy?_O7?dsF30xEpg|U~Qs4LG+8Z2ne)W>5Vzw}VU&0J_Z`i4)%RpKvt0;p!;oZtiR
zgb7Acpw%}?GhIY`e!4V-(yAT+KBB#CKQNXG9Pc02JT{Rd6RH~Et3d7YJwtqzllM~P
zPfN>Vis>nBv@Qx6@}A$pG&(PoKW1=KnW+23dqzt63!|*jse?{o=Vb2D{w-Pya%}Zu
z#U!nm)5UH9!fbifxQlJBodVgVocp4bcJxF%ddm-_dPTPMy}ag$kJoPuQY^N^GcJr+
zs=2(K?d6=A7)_yZ7#@!Fw6%@weaDnfn>i-ScS%%vi`J>gY=b6o7ziZlY&8t+0gkgp
zs3e{^M5~2c{nBeR1!8|Y;6_w`=K)^PsLMG9Lz~$AOl@IRa5T0!4{L|cak;@=9F@56
z0jRELn9xRaaW@kaDR5^I<Fkr<g&b`xx~HUY4CKM(W;Yj3G=Rva8@&002$6%m#Nt50
zxoFu7>VPFXm>!y~ifjG)BtZ-zfqX_;3w^d3!7q|N(-F!I1K(oG@cc}x+q58$#5WAq
zPB=iIVI5Eu&x*9e=w-3|bMh)6z+uE@_RBkVsg{D9+1vmVIs(yL&69J{sl@%bY3s3f
zW=^5N@cF00J^y5#yX+-<6Ed0TVgZv*v|@bsZ^Ku3W6Y~BVmls9zk)7WqoM8iB-N)P
z{OU_^jg8dzB&-0z267?|VWV-LU%#qO1=@X<dog(ov>1FIFAA&VaY)K(UW=lMTs4JF
zIA+ZYCqyP~#ZT~&7QEYKVs-I`9yrSyx`WSd>C0FpW>{V}v8#IhWmMyt<u8}WbH8L0
zt0quh!&tVGs}HpKCIHK^PuB7rpTQ?jh(SU<_jz*%kkJ>?nk{bQ+Sa^F>`>a%Br-7_
zM8ioB9X@@8eyI0o2;87IfJ96v2l+B)_(%9qP^g9jMSb$(SUryFO|Voo>x>ogWqQU3
zRX$TP*|%`ffWKjI*t*nCUr1Lpe_Uj1>A(WQ=pTO@XDUWbb-;c^XhW;mK;q|;SxW>d
zDq&_fwHx^ZUMdyB{$v$oUC8Sai?rS)vAc7^QWKBv@q4Js;DAo!W({<o5}MxCcM3hQ
z6;&INd;PqS=Q~yJ$?!8H)V&&2xREO-hUX`8u+!Ix^@t!7>0KXF2U!ceB}9o_Fdu>1
z+P?(s>{KJ>Xew=$*P1)D{?r=?FehKH#s;NqVW$>qBP3UCAG(ZQ7h28TfpYg6()jF8
zZh|Ots)2iOs3p@5QO!K|Ly?qJ>^#gVV+qGVBj8-YH2BoOrE6-1mYb8kNX>LX`W<I`
zOi@ZPXKM@T0FzE{J1`z6_RKi%g-)ZLz)U>mAZr*JFJ7At7M`4(9H}}u+WL}eLHV=G
zQizOu%#)Im%G1MNfh*6_XS_Go(d$kP(T4=Tmb>kZ@L7w7eVWl*(K|?$N?d`2-Eh{B
zC;<cn1)A*0p87hpW4hA(g$KyNnbsM{>QpkJo(MF&rvyykWOvziuDIP@fd{WijQkdV
zBLr>T_PulZ)Rf6|Mp1Zh=eV`!T<`DWAHgkvdfuza5)>PG{-||-<E*u+Bu^w*MK!;g
zoI{9AE#~8d#f1xD1j~{gzeHA@;Vn96^XqYrK6=OH1_JU{=ATQ@OwN2y<9vUL7Ti<V
z+403uNfvC3%lKtYWSzK;WOH3w-UxtqjGE;1tf5p(o`z>w9yhl=#Jx_ILW#Xxl+Rwt
zi%ntE1-hY4B0REt?HtO53XOe{t+{-&VHvaF8mYn8yw!VOx{AyT?Z*nOiX6uU_-+QZ
z#yazv3BTFAI;X|)kB)ikU!lE9^lb2fWiPeR`f3BZk0y3-T}7*4o(F&S_4aYXfh{!E
zl{0yRBIk6v)ar6oU~*F9<QF0+DXD<h%T)1<LcioWwRM3g>ge`Pg`tyFy!s4<hx(LB
z4BljiX;=^bMx16ew41j>`8!(m87S1E?T5$(A-YclF-GNJh?pVn3JMP|WoqEggZBgL
zaK$sY$r!kXx9S%AsHri-$m^VzQo29t5D}jxOKUelLM7=3|H5GD#q1mfbfYZ?jq1uI
zfctUbArAu?rPV-LY|XQP6dx%Zil+(7$d5=uA~lYu%lXDr?tDd4utg_O+?my9<K-5)
zsu_5?4v#%jAq57#_l$5?k(2NBAJ-30>JO9~WLg-)Co^h|9FR(IWq%$1`2&+HnC~E6
z<ek>olR9_|JB`d?j^nOO!d1G8Lo-w4d!&_a?IBSH2W7CyXvpkIs2`XAO53}DI0`ZO
z;+@#4nl;0rX48N4DWi*)b?0^v9;do4mguuCyF2B4By^<KT13;2wVJt0`C-Q#U3xAo
z@k~?bkrK_!BE3p~U5T!r8g_hq6$L9L+NU35>TEwLcG=^2XJh{?zdWHB=<boOm`&jP
zCqct=mK_u+TPA*f!m=c%VGDKfi;`T2`{v5v_;s{Sxu1zZ#}bCIe(fj1Rb=HlS>6lq
zFQB^3#G|Z(n-tJzX;FL;e>N04phb%MOl`C;EbxV+zZSbi9Y%vT_makpU`EQ-L?qz|
z9pSFVCgXCN&Z+QdTrBISi~!=Lf1Ko(o{3xum#q~T{~9dS8)6X=iGyFmSbq73fKKy$
zbiv;lO=OguwmAZM?H4?yP^nr;_ARJuf9PHUUpbTWkVKRiK$4@JwP1vvDX5IAdqQ9L
z1NTjSPP7nj-e%RLMGAM2PV(m`HBe5`L!99<wl4jTk&|ndXO-(QHm-z9Y5krFUb>;Y
zONqZQRKF00Um%?sa^r?gaYYMR3-v?ZiNrcTlrnp+{gs>cz>UYqV@8e0tPUN@@bRJ!
zRf10(dzDInq&{3pKzp-X<af2Wavu({73?BUK_0h1q=FXx^yr3Fgy@0$`NVvgx?x`i
zn`rl1=K?P`BLk1h4-@!2<{F|M>5XiI<a|8F3eT^d2W%z$Qs<Bq_fSJzz_Ug4Ne2jw
z#IU4x;zsgSAK|NKN5xO8(cNb2&vS--PMIt#W8r1bT5X5;3+#5tnck8tO>h~Y#}7}y
z=RP#9Ps2)68i-Jmx1C+!Z7)iEhgGSN(;XraZ6_$N#$oX{VF<0l@hXEZT}Xvi+P-vQ
zKXGglJ21mKip4N;o|mH#Dsq&NPGW{?kelD4FhxE-Ikl=-NN8rvgIe7LLbc3`6cpvy
z;!=MYDh@lfG2%(Hw@rZ71ZELs_xyalhSVON=1Hw~yn><$rmT4nM=kr?Pf%#@ByXx8
zH;<r1_X1+|nP$4(#Oh67bVMMbD?*8GmL2s3YB>ki>~sIhuo6`>(}U`6v9USNGKdFA
zXQ1EcRBR!+Hr>d<sE`7?_4w4x;MKQt0@>THxQCtj>2xIAvVYMla2RC1AsUDd@ZSu&
zlzac}`j4@4dqnkWwv<FT3>yQaI<1UKRK%Udzvx!r%Utd}W;Lf;_|ghLS=*oaNv9m0
z1Cc@?SGr8xd&AAYp1M)Obj<>LwG`^i^FJ4H5O?{Yv0$wajWFhEDT8B7W?}M<0n>no
zz5CvfuzW8!Uo%*Xs-Rri7c&(RTR=4mX$Mfbi#Q3sPW1b;k|S~9G$N<GO%kyTzTI2&
zqAzdJj3v}$3DUe26ReEra>OCxR+wucgN9$)i>q^~eOk6EvI7IjT{T$vrc{Mo>*JKq
zHjpBf_IHZP0`IQq7A@o?LCAMRyIQ;WDo0T2L#5WDdFpD~kLC_DV??Wz>H83K&Y*bc
z(s*YqSX?ouY(;Y~!aD~~HuQs>*x|ypH?-vV>(%QkYSG>z=QaJvVv5tj-Pt^R(S$<7
zSbftwaJ8z+wF|5YK%znUCE(o;{gm$NPPy3)S>9r5yzq}!>73H7Y1jM~jHI+$HA&H5
z<Bv5FFQPEXa>6#J5VNf>#`2^`Od0d_bIT@*_{IM_jVxekrL9>tGj(8duak~Xq?CO`
z0iiJ2bCh>yXhNcJrYc7jZ<Z!e3==VcoGb2RfsULCcZJ>*OI)Q4DkyKLE>2`rki+L3
zXywdL3IdwC+=@sIeh<&W@RT*cW#Hzf+a&EAsNn?XV(VFRGrKo!nZf;3;9K1)LBWv0
zo$CQv&xAtcmv9TQ1Et{wy@!*_6mtYMuMgoIe6WbTXTQ#b%qfQm>mU}Ps}K$3x6}`R
z2pRFyVxFgLEsjN$XOk#%k)isCEupd<`xn9GjOB-(TB#0n{A)3|+gWz0L-hRz`>ib|
zYd%h+w{<^^;(hcck}Twj*&d@3=ThqL2R-tJs_yKt@d0yubAcgJ59b+RFqDIgt7qh@
zEG7jMxP<vC5cS@+cZ&MY>YM0i5L`$Z2FEY=foiwh4n3!=yzt6VCrvjp_AC?QABEeS
zyAUj@KuB{IHe8#*7vXHrMx!JpHyxa&c{fk)p8V0VxRqU&$$O<cYs+iK4~ZRE_%OQY
zfFxG4!jj~e6|@gjbT3TQKhE^>6Df=MO}(H=xs<gHY;YQfs)E_1rl|V&VcR^cMfD_@
zJ4M6s&5?F8!z0l7Nk9V$D?k<kP~OnAIv;IF+L?B1{`;-e)xEc1FyARcX+V_5CnOfz
za^RF5g)NuJ_+aJf#)e}5IZaPhu~r%|7uf?q{4;{-C6V1#1eCDBxe7v4u}ZNqAQTsu
zu&m3VQ=!@1liPk{;#+N(SJVW{>P7$yGW6m6qZ%*5A~m;|5PNsHMHN|MqC@UwPY8A?
zR4Vt?KBU5Lb2$idDnOo5G>%~<DCIS3MFFr@0yg-3tIaZvL;8uw#vN$F(g#^gw<J_j
zU(^zmAR&w%mtKP>^%|%V%qh)JT~-h#u;m;%@78+RxmchZD7B~&LHXf;v(QQ!o7ZZ_
zgHkP*54{X-%0^l0KlqJz_&=U+vLLzsxpx-6$Bb`GY-lRUp3O!cjpFi_a%(P0<RHcj
zR;z}SawkL@T!4G?N$Xc$xVqzVofld3%s8myy9SN@$d1`&&R!P@)ma+FayH?+5-H($
zL-e}!xBS=*$pZV^C}ZVW%@#Hp`p(OWZ1;Jo{A7Z*2n#_=Q=xS!Kb7L)Ho!`(UPd2o
zhou8r0yzHXbz-C)S`HEg1|S&>Lx3Iw){kz69$|s9SJc8)5hizOb1e43NbVr#5%t95
zyQY=WdP$x=k5SlOroD7dVa!)6jSV`0h>8_Y`|kB9PQwAoah|$P*~o}P%EYMHgfx@i
z?qcs~>5;V_o?cuWiGqh3KRViP4TN0B6Vt5PZ(wG{xS=pzG%XqA-krhZ%#c3(Z~2YS
ziICI*Hx@D}mMQ!b!zU8E4HgPV2%!hx_Q?v1W<1vQ>`b$O=@qDwB|4<*TCE|8IdK!M
ztTJNn098|(C1$8_(A>LtS|WPHp;M!&IRjHKbAr2YE(q|WTizWiRk2WM7FnF9ze34|
zO}D~p{SSYP^Cq(US3jND@Ii5WAxYA9Td}pqlEV13xfkLyf=JNmYH-Jampq4Cr-CDo
z(^UEyhGtq-F^z~hHF)a~Ty5R}PW#rHRfG0T#zsW+w|i%iR!9baMg$1en5L^_rF3B*
z4`I~XbnYSj`;DITS$eA$pAaUMM-|t}8RC*p)u6?1a)|97DWC~t3cU!9F`!VQ!E4s-
zY3sH@;+Qx@cT}q|nM{k6pN11|1s_mTVvjopBEg6$${MN6#v=L?L*A{`1CExbDh5uA
zQZq)ZIQkdh@YQKhII2~kl(oi0yR*etfd1jJiGdGBs$i9+V%8}ie2GCo?}B}oj2$5z
zVFwK$asiV7p$Px>fTi9hOW^qfkOXL=BEt$tlTR<AL?G^6F7i=Us*KB>M!RLBY&A9K
zU-L@8PtIsaNu%`_qC{M(w45-KIrl*~g96b4!}%0Z)$f`tsLo)Oos>hj^BF<977|*K
z1lqE3_%U%gM>zPLm~RM<IJq-zdF#a7s3;YD<K;Of2d&dg1RB*k{t0(Z97|FwP~8&K
z|5d>>v%tq&aRiss#-970Qh7G>jIV4Xq<~PFK)IgHhIqGG*!-`}#hIJ?Ki{!y&Ce~z
zZdHSo=%X1<!N90=6}VZ!;&c9_ljvD8$c?q=>>%>meCpu#E>m=^_^TtbY{vlhs$Z6B
zj3tABzK<}%@OxOPbA>$!j4p3Qmttf^%bqQe84MG#Hb*A@$4LK2QIxq;M4U*XQcCd}
z4$hsv&##|?Ggk2NgmN{A<;(evjOBUd96ZRGqgrO1#Ku!?uu~hwXIx9L4(k3vDI+S?
zI0yDkyUPDEbnFbr5MflZ3rvf(Nn#eFxphIQby9h_k#KpEf3c~^B}haVL~VsnpbJk+
z`UyeS5+EQ&12UGDR4{pbT&F+xoZ>RCVwI}y%~=wKn@0iLe0Ka+wpQLmJB8lsw`eh;
z6_7))XwkchAX;G<QxuIbARWi>kY^+W(yOy#oBuL8950A3T1RM{TfVp#t^?=DSEcb-
zJ(IXiKuR-5B9c}Rb*F+&Fe}p|Q^R+)1`f01SU7S#-q0hvgt{l76R!dJEm2vy0VNf?
zd;a;)v0IcGQ)vP*r3}#BNMGn63z<C$mLfK->6NvL&N8ke8g>kP>MR^9&aMc5A)U=Q
z&sYLUFJL&g7ySMEy5*Af@wtLkN`43vw$7#r=TOtP#>7zuBl@YSTfw<1Ie2_lRG^5y
zzry!>@N6Ru_a^~^T%g1672Co7L6kbBisr?VWDg6Es^(F9u)?sHC5D+nwN5erh=&Tl
zK9UOCzV9ZhLWcpS_76qpZot)y>#xA(hvOJlzMUkURrj??c2c9S*YUy|)5`>#rH_HW
z?WOLraj+cXIp2ApG-$&*kdcWR4nQNK`9#9|=Q2oZmA=h1Y@Bq+PooA_0fCodSs<DV
z%csp{thzLSdWqDMT1knI`ulOQ*szTgJ`53pGsam4Xx0k87PiS~moJ_<V+kXkh>bmq
zcFOQ6!qACWO%Xe6GoW!+x3mwGYY^maK#=Si9})ut42eE5{K;zW22mZHGVShne9lnA
z$E70;CrmBZvli1O{xWaz<klw80X@O~6N@F3XXr+JXv9r)MipWmK~uGu`+mFVZX%Y0
z9L_2%=InR5CwI}pfpsbp4AxGApc9ZQMaIF9g`4nCqG%G?`I7%`w(v=E4|0`uq224)
zEqmNL9Q@-#3a!Q|Xr9uDUpPL^#;bJXO{oQvT7vNvfYxb@i0g|b;18LlV*+i;VL?>6
zRUkND_JWqcE6`ApcfaC&NQom($*68z8dzI$0trBaX3x0o)kb@)>os{h65AOKv}bO-
zYRY!KX}UHhW|yU(BkZ7ix(KNUR8Xb*cK(5vK~7x%pi-Ck?Zc6Hi=Ys?r#f(-yqat>
zjRPhoS|n0YfJ!3Mv-|wn?Mo8#NRvE5IAh@eOb9|f@|n6^=V#%ixdn{9;N4Xh`5Fo#
z1*<P7lbW~Met5tTg(5l2cZ>o>=Og-*7Yf#}bkJTbt+f$P1+V{(B}OWJbQUlZNI+5@
z|M$u|91tE2H%Z7rMV{w|ON(`ekP9DDEa-hrSsj=)nussPZ#rzbQP5OS8xTG286;(8
zVyFW9Y27dCdtOJP(Rl0-tDF}H4Siy2HTL!8&&#A3=)@QTm3oo$kj6$KbNV_E;kwQV
z{t#-j6lNyDK)mC*nOPw{=ja-3)mJ0Yv+8<2wxVQ@YbjKRrH^B6m>49+M+{f$X+x|=
zX9l~3dI|`7uCJ5)6cgQE2qo$bUG$?p?x#D=ad9P6{R=|e*GVKu+ZmZ{%i|n!_CE;p
zgd$5Ucu5OV4mShk*Az{-th=^{tke4!vV7twrvq9zUeFHvN`ljhE6SR7ts8m}6p>*n
zX=}EqXSrF;?Hc>hLdjBAiv<smX&y$J<6r*1X}{Sy?lDM=xan%+qY{bd67@Cj8R9RS
z=wPr$kI#P2!?F_)bqGv%ug>EB3D<Rh1xNUX3cz8ogcMKmNUN#{8dWbh3fQT~>ft0K
z+g=5Qra_sZz=0&O158$U$f(W_U@qS$)bexE_pnG0EA+YCXi2YDY*7Da41_A6wY-L&
zKK0cmG%y3D3+{k}fMhG+CF5}lGfD=GtHmgfrQ)O_<~*0KvhCF@%?8c*2`A`qP0+H^
zjafk8I^QTs;BT?|2_U}?@lYt3e|TQmY)U8n+}S6EUE@GKg=5-aI|pct;FPr`z@b?p
zJ;)!4WY4CA-;|KpBSZ43TXnl##njZebf3?|nVK6JOSZ{6M&}{zc(5*(k*<cNB}=)1
zFR37#`l*dCKNJu%+^9N%RH*|9Q<E!$SL{*j3|gKN^PMssW{5hKwQlnwIxlKsu2Ko#
zR^w00kvzl}h=le{c5F3kNseM-PVH8dQhsVw+I431`RdXgG9i6Y-=g;!73IS{Rbes4
zdP-w6(1*e;s~x_)o%~Yz^z$=ir2r)|D<djVQtVEMIL``mIz*<>v+9iBaR=U^1g!G2
z+_{}$t#liIOHy>_{Dd;J-g*j^YJ#%<<RcLRaS5E@+<+Mv$Nyj!zq5{+z>oK9G+2@B
zce|%<Ng+GCl!0Vnd}0GO&pwC&p4Ln}4j^A7kH;XAWVb<<e%yy^(Ntx>>Lo8oIaXch
zUP_R;%&!x22-3NI51c%A8aY3aysZ9yb9IL^IegBsHyRx>&7k-D_EW81G`Ys9j;wz2
zQA9IQAi5f276QfBy&c!0y>Cr;kQsPdjalI;eS1o4V`RT(K`9!VLxL|J|BZ*hlO-ug
ztm?%}G}p)kV_Jtg{&pwrg!tsOA9I@HjEiy}<Kq4P{eib)p9#>~mZe67t|Red0EG|b
z$eB%5f@Cku@KiWO<%QT1<ZUFw8xe*?)7I9-u|ulx6hNLqp8DLYpoec?UQQ-o6S6)u
zmklU{vBM&_)l54`8<)|HP{%zUv6+267lrPYp5Nz~S~bz1<-;Ta9x4!(TwTk)=83u4
z7Yy|+9C<ghatNcZo9I>N#IRRq2}Xp41vWfYxDssw0Nwuac~mH;^!CUqEk8dlL_dK=
z-X(%RD56!D+;6tV{x^@@`SXS1qlg#V&G@YM$*ZIS{g3%9wc=P*-h^$QAy0j&k`rxy
zLImDgWHX)YG*=hN;wS|D)Y0I9#)uQ_yu~Y7^OO2#5$)V!XPaMf(^1k&-QD~ITBS<J
z1?hQ6oL41lLQhw(mp41R$l00Bih3%K_nZk8ycBAd4j79+dZ>k=Jor^?r9UW!rN}V!
zp#kq;bviFeE070#s;#wf?KN{A1J+kqxY{eMu>VSf!DcwnFRsTVjk?S+8vH6!O)4GI
zK0FZ~ks_$IX&GfP&0g5)S4ai*k0|Cwed}+8k^0Kw<!IZ@ENS3vig_|cFbo%Ekv6Kl
zu0tk9@19l9*AU$+LnB4b!4P+z$Brhh54qoDt>om_SV^}&l~+oFgi?IW$|@d^MKyC2
z*9?WC>|ZvQ$+fJ<@==&!1@UCP?s#@(f6yS{v<uD(NLYw5mxoXT@n*t^QFyjHG(Wr2
zRezl>j-;!$D)q4ZINKZ&L)q9TKRjG*_q3bcdMdhzpT8@QS?!re49ylGw5~*FjYxTH
zp}brj5F(pvg5YGW?wL}Prp^Lnz{DNMGPIKLfy<xGqh1E4H*5+x=Y!;_nL%MG3eg0n
zjwf=$R00QY$#PlOl>SG8dndhuOHY69J7ZWJ2zxE3Suu#R)Dotrp<Yj%VCrrEf@(2X
zNz>Adr&?H2B5@l3oLeH(U@4=~_P(Ro|2`o!-_h&1W>HZ~P#3*5FUe9+>1|p2g)TdX
zcP#sTpLjD_qG&v5P>XvUBbl&udx*oCv4$)Licp9-GMsy76nl|qA~KAEnlvlOsgm3X
zDmVm_%K5S1oK$F0(?t&6lt_;|1gZ1&WG)HRx3EuO!g-_1ev@=RZwj5zTbY<KA?_q+
zGR(;0e{Wwx2T??gSwxj5tfpclY11YN>R#dzaf}r;r!aH+`@7rMXq8}6?tU)Q`Tn;@
z>LldLt}1~S=j<?HiJWLx(Nio|*YDT(t^N!i9s}7YU@bu^HJ~Hz*d*8kjN=5sPJYTr
zOnHcc#f85NWNkh}#+8P3{)19qGPR}sy=->5VSeU7(5TJ#ye*dY+t^7?;fYZQAPl`^
z$JcwP_Uk{P6@H9WBJfse`Q(z@@37-2K<*+1!Dz}*!1rSf`jZ;86UlREEL)zmG@DFe
z5!T68;kV;&<+sPDD?u-K(@my3e7mgS_vAQ}YnW0yinQAywku1>rmNV^#@9Q$DOBQ5
zl6JRo9iH2gGq0S~y)?ib6X}x-TbQcB`=C|dS%{%SKe{=uTMg4k#3PAbJ7$zSZ9MuC
zy*{SLne(mA$8^U%iGP_}i6qfu^@%-0E%bEoQBi4ouUJz1{sQ4&lu7U>n9RrunZoiU
z0hWK<iHgB-+q9X=gN~ut0lj3?z@7kbMfKb-+@qHmmV@r8uG{#)QNrmMzl<$+!mF9t
zusF_Ur#*xZ8jGrwdEyqGr^GBf({FKS#6i0(x0oo28YB?Hf%|QQU(Urhw5^m?%1aD%
z7<ds0@FgaHCZ$F%+av6QIOm4gd@uh=l7%vp5*QA1oY1g@#P$pimvsnh8!RRI6}$}~
zfx8sdG%jfrG`X|XjH_<$IyKdCSaAS|SZ2Yay6s0PjN5LL)6aq>dI)`&<d4hm?N^UM
zd#L|@{!j1hvfkQTZQL>$fUR?sQorvo&MgcuIR!El0uq%0KX^X1x`#GAW8B?@$(xvT
zY&9}#?F$nnI1g(B@x3hMOM4wc&&XxS{4Xnb<?PYW)4b7@K3Q->UeGl=%(L{0U6#t>
z=`CYv8dl0yKk6;;y>jN4kHe!x1Ia^IT{OlD29t>JT(&wUG}eICE;m2@vJQdYSbvjb
zz;r&sqf{)Kd!}6be$qJ<+4R$rokLk<{lQ+QjZlpoJGakkJ&y&l3DK_bJnCSfwwX%C
zo`d3CN*1AR#>Y!zxoUp62E`+P@r^VE3Ht=Qnfx)7+RnT>kPCH;oP#}=U}vu7H>DrW
zB{jye6MTUN1-cG0n3CAK#phu^_;s7<(;<3ZO2jhX3R`kC?Y?_vDyQIsoP{>)t0Wh?
zpa7$QM^2je@-%mQ1>y?g*Q|2!e686~T-IN|dz;SGus(CgnC5vg5iQx8^CJ2wUm_gP
zvGN(%VDBBjMezQf>tqaWj?cF+NZViCEh@Q97=-C}0Vk8Dd?B=VDk{nl_G|rozFM+)
z1*NT?@-mJ7lj@feM06F*wk!M9jny5{P^Is0Tx*{FE1k-3u%Ya)xaRerApBEBT0#V(
ziV)4NeO?Nsz3V3(6oOpOdGEK1%X?*JeA{1iIcra1-T&yQ2)%(h&l`W_a{q~a|9P2Q
z`)?!hNJQ`cgv@D<|1Rj?zqzcf=YaXe5U&$$jzgbM{hF)8pX2s?>VOiz7(F(n#t#<V
zxQP>Nq~58qt*o1j-iTkHOr7UhQ4Hi0H4XdiPB<nhHkb%id~Np{s!#FIXQU7pjUxJY
z;@>i-VEzFZCTqw?2>yd*{_XFV0%Gp}0Dt)(IP*XLp9l5-c$#kEQ(<gK9{-;|X92#(
zJ+=Is_?mn9{_v<}SU5k@BCbdYnz9?^*JZ_!j|wgvH>?tSP>G?wGc$OMn?2>X=L`nw
z0d3PP1h$>p2IQwM%$1JFgTNV8^mN#AyHR5e=t`@YoW4%T9sC=LO39mBL4QBg!s?&!
zla~_wTkXE?-aV`qlUnTRel+gx?Y`}P`ea$%$0CFDlNIIPf`ga<`hEn!!WF~N;y9GL
z)6PftXN8ziFfnY2003dgAVPk~SGf2rK;K&tns@)hy3PfhSvrDhxmB3j$E<3LK1N(G
zUbH#?Cq8n0Tw#8e&$wCK#dhB)+3<I%;}hK1ea)+>zdrxMHBI~fU44D*KB)cM{f@d=
z>irv6ok+I++Jw4_U%Af!Qpk>$mYtReMYxup(Xkzk3Rn;A(sMJbjIHyMpl8-Dc}+qk
zG3e2HOLT1v9}xrNWs}#?OQ?T)hQMxp?r8lLc`uuG)bYBH(y3CMIYgY*wfE`e<GVoD
zjgHit)p@|(G0eZ0?iVh+7w}dRAyc6SNLdgqO1os*ylT)Iae7{TxXE34<3SkcH3_sm
z4CGA!>hZF&!DrO!TPrm1kvFW*-x-a>;I<Q%;7Yl>aDdd!2&>Z2K37IBVO{V#>Whnu
zSRFd5SdBoS33!`Nvmc+*|0({PgZ^hIo%Me;X!o1Y*VIpM3!fB|4NTBxLwvWBZ`b6g
zxp2>(^2huFS8vC@L<~sJDaHwLkM$TG)BRCbVGK<gbsOj}qMN?*bPbAX{&f5H`dC*t
zfP0P0J*Ksfpu#dHBfls1<@Cig1%G@nk!|pk-y=o4xR#}s5l+?4*QeTdP3Un@2F(H>
z`tqF4qLdwr=rN1<Ykvh$h7dvMg(z=txJZe&9eze|E8pK2?8D7I+ce5I=%wrKY!`W7
zYTLYebr{2S1vUA1+|D40i$KXMkmFy3Mf{7_)(Tb)P<?6&b62ZR=$`sEBr7=?W2@#n
zFL%VOE&D0KtB{MC14j4#M0bFNRZG+5--4I@zn@F;N0P4lEqeEo<Bahp@~53hr>+@A
zGR0LLg((gpNC>ZU6nQy0*l>Y+)m&Nz6GC5~C}Ts%3_>nv#<Z+Oj9A8cziEhbX+EB)
zE9>|eR6aKaP}T8EgsW(HtStV`{QB_ypYE%ECBxTiuD@mQSVy-HM~<y9lYaR4+=Aww
zBYf(Y;d&e$?g1knZ5HPi8-1;6{WmJT7uG$S)@ta9{c}2IZ2{R_T{pZ`fr0T@Sn<Rk
z@l0Bjb6i%F><FFN4}JogBwS5@4J!Qm7Ln)r)%W}3FP)*x67Do;mLA&K;HB*>!8^_F
ze=9L-KMfRXIQ4`Kjbk2py82k@9Q_o95wuXj2pJx8SA3E<MZl63?;tPKlN!-XND*)&
zUdD`iQ%KLHa<SPCb;e)texCwmjRZEyhJ#~poZ0>I7+i`A=}8C8L@c<-a7s!xs_Gc!
zFi>yGv8%cDix$f4q?#7-rS<IDVSzt!)YKZcX{XubUBY;EGRC#i7_gan)zj=CiqTyd
z!e=85utl_OdiZ`Cz!WbUZu#4TC9?wX)y;i+u%gnPa-9~_CT9tOpMFAm8Y`&OcEq>a
zTJmd#Q!VR{amGkG%It%5BRNMtm3&f=2ac|iV10?CV*W=h17^=y!7LeIy%?riDIGdx
zb-0ZaOOwWTJ#s(yt_TFbJn~Vp7FS@cqgmTEucmDzG4Rf)NX>&fB;iN|Y~jSpxRNmQ
z9#*u%lR3(am<G)`l2hn`rcTS1;LF=bO(r|WV$Vp*8x{7r(%D)ig^ou<W^Juu#%Q(m
zLkD^fdMYDmxxl0;(ffWHc|=dx85&R?XRHu&>h8rmMHXj>uaSxtsU!O$2r@cylVhA5
z<bwH`ho|T^kz5Xlud5eKIFg#MvitwEdZZ(sVXF=tpwxXCn%qArU(ygoG=tr)J+mxQ
zcRHxY%wyZM0)ClmiI!cYfd!+DlX;+7>7|^$79GE>x#cU+r@>AYU8(sg`3UVAGxp(i
zsM$yB?e57N9^hnY0QOA!ZYYrc!nY)RoVii*XiKG{N>jFZ>!%777Ud|w$qH!#6@dZs
zX}UaLLXOxJETCqL-GCaj4g-A?D)kYka9j*}&Qi2pKrn>7674RZelD$$^k()#Acffy
z0yZ(z+s&~%3&co;ZIG}ayN86U%~G*USflsz(hVrlR~O7;UlVnh9SM#-teTMNL0%)W
zMfV1GGCFtgi|X&xqoQ0@5*M`?$OaFWfAlc-Z+gX__4CvplqRW;b>`I)iI!B{>0@s9
zX|(2RYDV|KlHQDK<3~&)Od}R0YtzvyuX&!OsL;oY?co&F9JGI2A4M`DQ~Oy87)_&4
zwR|}+E(D0fBhAqFDHd@A41Ngjn)Z2Sg6QqHZU6ViFsBOYRVQEEixiy2ufGtn*G<yH
z=zRE~>ZJ^j(N$KJA(L~#&%ISPP{Kv1IBrz0aNdAV#lQH`v3+4a?w!nYoh6ZKl`Gw7
zZ>3c)5mmY|`Js7?+b?))laedmN0^%dS5KQc<u_8zOCa9@>%rEPvz@93r}FKd<t;G5
zIP?*ds5cUD-=&8qJJ@)_Jpa&?XdLisos(_ItPYPukADEOvA$THRbGy#_{^%YkQ^hC
zo$0r#g5RXNG0rq#26R_KGRL{78oIEi+I@hmTc4`2lSwCR=R00quvJqKLr*49UR@LP
z0t}B&9Ct8VUpvSeCGlG(kXw{56k1c#nl^GK@sSx;%Ct)<qG4sL^XqLYwEW=_sI5j*
zJzyiaywA|MN$GUQ$2xEe7vtnQt7*pT$iV{;rE;e+MV6nvFt5D-^8SJiCsZo7P65dP
z+maIOfskWmos;R09GQ7e*djKQ(-C@^6+(HqHamLLYc?3cw1C*SBbDTjPvZZpu|taS
zzM`L42fk}Le_y$EgH8SX(l`o`Nugz?2?NK#bgHw)z+0~8al>5FCDR@)`l<o(D;8$}
z9%VE5_Iq*4+)7cf%1m$h5_)FZnNV`;RW`=vJ#%hOBu+yr_S<#qwrVUYPTR7{f*-)!
z<?>mugCosuae6pm;Jd7-J8KDcu14VzCPUaa8Fxp1TU~tJg%6>9g0dPYGo_CZ94%0i
zt^70$N8%WIYu(ovpv$#79I$|1PPIAKjI%9*zWMV?&CWB*+g~)1r8Ctf8A||J>>8>>
z8}aPtn&w^2YjM`FZHc17?L&~7-oCDU_)T;l39nvK;fQmJ_q*7K$Rg}PW-`ycZP60I
zOQ~FpynJZdpOCi4;aZE`PCXtGCWeY{@1w7GeFe-x?wx#&ZZyE2U(m^bY1sffd6e-K
zspWt*NlTfSZYIN2Ofl1zKkBFquCd$8GrcL9{1vM<A%a+uX~grljoxn#mENHsbVdoH
zk54B5yX*sKG$D_&NBbdNMoNAMF?YR-iOA;QJJWbTZwzhpiUQ_nuo5H;8>a+}j5J3v
zWT;@P!?^<P`w=qZ+aY1krgc(GI6#N2LK4wzX_Thx7aW~shR+;J>m6-6*2S#!ZJV|e
z?{6C@r}$V7VCn>jiV}DJxEpl(5`}{`kRkZBsl+Fq#;n*k`p5n%2;jm8QTP5pm(^<y
zZhB7RUeD(l^;4oI?QbQwkTbXDNS+C+C(XBvb{k(@F&P_5papE9b>i(l(05JLn{cNM
zR~yI!rjfhOaK1OLggVh7DTf!dn)ZXSgHL?>_%&ye1PK{J^bkUO!OpXI)$A&>64><l
z(K*UIJu3NQ9uGc`I$(Mn22wVnm3nBAPQSdLyKOR?-3=yx0)#brrxZ3LC1JY-;ZmX%
zHV=Ld-~2K}oMbCzRM51u>O20ExgwgEA|k|63uAiB#eSWDH2aGpx-g2hXn%CyQ$h+!
zKNM}!HdFbu_kTU}+zyo(I5e&Ehwaq@cqy_h>hvXklN8{vZ@0n6a;y2tr_&F`%tXUx
z)w8Aitf+`%yQ?3P040Lz@zA(FE=)i!ZThh^%DI2kiScJ7I2C7+<qf+%3bU~CFkj_r
ztwBK!rs!>ij$(X4)cDoX!sSK1$umJ^MfxzMD;VwpPi0-dQdpW|yxN%v<-9$Uf$(o}
z)WA9yazDW@4_VB<?#MV)qP#diWUZPQ`9zK(8YpIEVLLWcHAlb{!rp*h0JSV&XQ2U1
z?eX4AfNM2-jmabbeT?>qlzF5<cj<wWIo8IU!<h<M(sHJ;a~LT`06_&l>QvGBPC_IL
zn2M8QXMlc!;7^e3bioilL$LOL0M0-$zceI3kR(wX*+#1>-GAFUUHrrv2Je!?4iSw3
zwT)T@qNr!~p(wMQ0s&=+C`{BCow05@lpZ|O0oe*o1qn2+LwUhUKqEApmVVou5tSk;
zO}wgseD1jF!g`L#bH%}~VOUfIWvMqCn%bd7kt?-Tk4rv5gp|Ir1~2K9Y^|aX9&~zm
z_11wHl}^xF8A@w#3al|~!kU@y2_n>smQ4`yV*CCoGO-m#!ETvRl^NDnwB>}bpWz<{
zPztY}6?Z=kTElp(*dJD)3tHDxuNu6I<f4cgeB72{FLHK@XdpzvXtkAHP0sOLmdKc7
zBLY=cq~dr?MTZ7YFGAONHWA1nGs#UAHe;kS_RErCWhjaPMNx5jGG|$@5u*rEa9R^C
zx%pTNyJN-faER!Px~XZ_JxWr$tLW6Jv@J6y4R53IsrHRE&_|!rRY<C1ab9akX~jXF
zrxajRL=)NI2_kG(9V;IwDuvbx34wvFFhr)gVKmIKgCTwEd9r*;>pPUzcqMqHF6Abb
zg2}*Aj&n3Q&U@;0L+ArpR4AR=j^lZ1GR3e2tMT!?&KFZ|Nk3~2(4VLm4YGn^S<-g_
zf<{>d+TgktXELIKEYq-!$#QKt7!8;XvJ_?NAqV?Cj~{-3*6QK_b|q;0`Qs`Ad5-WA
z&f3WJ+k5=4|J=X9|Nj5_@A6OmnLkGGf{}<NZaaCBLl6O#S+XJxi(yiZoEWz!@^soa
zN~aWzQMBIigGZ0q^d0w)PkFLva3K;xpzf3Ot<ovAb9=JS*KWMboxL5(qQH2~jmZrb
zt2K@5(c9vyy5#06z}g9IjQlhIo&OsD@t^<q`M3U^|1mqe6V$nl=MHI`QW34l$`Uc@
z^NWha;FANr$hm8@Xrc&NX87pnkY(R<vR?Au;V~zzLn%$|99{q|1U%EhkZ;_2oi`3{
zP?kAaFpTYpyEk6oq+TV<)aC7iblP1ys3ryfv;X<O!hiJV|9$>@|I$ATQ7}=k2v$Ao
z#?k^<OI{QN1uE0T7$BtFSK5P9>}Et=M3wJ8evA{icXG&wCv!GFb(S?Q5LGHpM-_Q#
z|2p5e^*VQUrWARO=@hqjZ<7y4th+Q=xlIUO+Qz1fEIY>^{KH@8|L}kQ_xRuZ2mcl?
zzH)~c(i&#xZzU;3r4;04fwCGga&e!=c<vwtjmC&hY02PO_Z?2aMaRAQDQCSSMxpLH
z;<ke>sz8xteEIr|ymjL?)A5M0t{B)6cW&Nh;Tm*2k8CfSguKE(2(BE+bHuud%o^d{
zhsWIh@|*m@Klx|*Q~#|$jWRGDWsJv`eA_FjP*|mCoFmIKf=)FKC8?I6B^4OYC2Ex@
zh@R*pHXrb0w#1l<>7YVKSoaIM&_Ym%7*vc*${|BlV7rv_?)pHk6n)ndLPUv0iGEfb
z{uBf+ELwQ)sAYxlGr#ml`LqAxzsTKJZenxAbdoWw^o9KCc@02G#Mq3GQuf=<&9wQR
zZ++=>9*Mq3#Io!?vsKUj?lmL?ns!6$*C<JyI|%3)*vqF>&LSS#CZGkRceKG{qD5^R
z1TMSzG_9~`g?H|sF}iz$U;JbLH2=+i?!QejG>nIq@x)SO&(_eO@F8rc(lz0{gGtU)
ze%lRDCHZ0`;`<g%!SQ-S2szX76z@FiW<kKEmRwQ97$}uyM^7l6#`QwidUBv`J7N^{
zxs4Z{8b2N}!s#aR;LP!K+3?jr@(1{D{PX_;U;V~gXhT}Vfx(<Nu8G1*B>ITS@{~@B
zF<s~FeVN+3pEXVd-}A!<4>(*bcyc;t-A5KphjTr3za*%wmb*Zy3^y54HiC{C?-ixZ
zp?9o93z1LQcjZDpZ(KQ$=Lo;80jRd;>}UpLjHg2;W6Qy{AqRT{%IcYB^_SjwlW)B7
zIwqtCk^olQ<TSR0#`E1>X~n1-kY$F|d`-7;X~rUhDlce#NAUf&ZB;PVG96cV*W-In
z+w`CmUEfeUhtb=PCiz6A|5I?DLMRoSJvyY8fX;IE_s8r^a}ExM?CcCc31&M)_X}V9
zDPF#QgIpWL3sxD7v7hKl+jjeCMUmx<hhu!#vs~8Ho1U)QFdF29DD<J(R(Vrdd}lPo
zYDM&sX5FL~TG!FKo+t*b6k%I4xojKnxsBEfhYy|-(G15!rqcoY`$G=)hm1yPpiL{y
z&Sc6T{QB3}AC6MvLx9|5sm*v>BT1R7w2q^qWLOQTR~uH#ny&Q(Uo);MTHoVC$2J7P
z8q0JrKoRksr>%Q}fO9RKcbqp-pRb|1gq(E*#3(Wa>!UgKswHCCKbSBXrTcVmZ-6%X
zqQ&>kmtW=ge&uZjSq>pJ2Wb5}B0HrJ4Vg{UBbm-=U1YXw$gHI*GC~ZrzP>mlDS^H5
z1cSkM9^VH#QK{wIIr^9eN8@?-;w9Jc3ZETU4&*svSZZF`F?{vTb?(3WU6%9rczXW_
zRKpT&EtxiC+tcWvs2D}fZEJb$+I32krH#fwI(G4_WlMyzEZG^47?%}q-n>N>3-+rK
zFJ0Rs#6C@iiw9Bkv85~uaGp_7k?EYi>j*Z@7kDp3(I~B+6|paIxYI_lQz?GWOS{N&
z#p%6wdGx_=qoO0vGqlo_I`M0v6w}d!V%78d?v$H*JE$n2(~w>AkNTqcYpi8=I>8vl
zt^EUT4R^>T=L@%Pu{W&HDk9P&q)sPPpwb4bBbmwwox>Bbxh9Ij`EBW3k=kM}d3Y^y
z&8-3a?910U{P2gY=0D`&2j8YDbBr}u6uH@UuNlK`HKtm(ymjXWld4KxSKAuR#Tu%#
zxmRV$WIQCZhL>;LVrqtr@&Rw(zD;2jnn08;P4q`kZdB^J$}L2N>pG$ZV>Hf7sy8Wp
zv3{2+q$>?O!;G)roiK`yd*A&njvjr7cC#Qa(i#@pq(Zw+ol{u{qt5a6OSdp+L@PvH
zv_2q8p+v~CoSo?o#h|3jaw<x!vRt1`xjq?Tj6t?+TOooI#~?H0rN!usdefv1J(D`u
zB<N>dZ*tyta)r+mS4H&ah}{fsjD)>O&aeFWzsUE0`?olJG6Q8<uUoSG{6MPN&4<h%
zEtopT8#ne?2|PJFB5b>S(p0-45K!7s<rViIJmJkRyurb=<iWjTRLB^b0=!1YNS<4w
z^@%@Aa&#XUO{e71&}<wgcx+h_qs7a1I$kc{+fr!7%?fTE<eWY|<lp@_|82YrtY&Na
zE?`Yd+@4`FZ5YOkCm){i>PthuaCe_?-E%DKCQVsytxR1uk%Q@!lhXy`%yR3cmkF!X
zwKd5q%8(;Ik{b)w5a<Cz7Sa@WHQ7P;;G{!kLlhbxw}t%pV+{ITFtKoB2)C|P{Knt-
zb(XUyoF1=GqNz6<Oc{wvBLbDp(VNJF4~}{1l@YJsy~eu_p5UV2Mm>pLAgX}1mYva<
zdk>Ge@$y}6?F@OmoTEZyrx>BVMnWLZGNM&@>Bv-x>I0)m1-2k`Ld0S6k{C2WQs(G<
z_UAI&t5u2{h2r)kL&u)~@X!4Rl!KfHA06RBq^aA~39!8n$63busNpndUb!{oY-h})
z)6-P`megSu03C(>=?<rhHCYV2@|Bkv`JDOToSVZD^SmL3K$R6>6kd5st*LBcDj3>5
z#7Cm-QP~h<G9;!0hRD{ws|?oXj;jddc_Ip~i&zucEEfEwU->(%W({w?ew+QhG5(@w
zMNz68?{eDKpqyh;StO=Ddf7g_Ncz)emgS=1^r$0`hNGhy?;Jhh<Kr{D7X*z8uy3ar
z0y!m08**!jWrpMhUi6kKLx{l>aVWk0G?z?aX#sJ9?<2!f_}hQ$+x+%#z0bA1U2fgl
zqnB{OiB)9<)*F1tXg4s;3}wC@lH)w`(H9X=-+CV2KgF+w+{5FOQ{Fjz#C%y3V(P>h
z#R`i-OO7!)!>XbU9y=J|qoJ!E5`?~OQOW}Wlw1tUT?R@7-vcqCL(8xH%CGU{!3i(F
zc#HA4z)QR+gpZ06=?&gx^qa{3Fh?udI{$6UucS^elwv+xarkJ7cAD5aet3MyyHB6c
zwmz{0L>Mzfi(n)tD@uyIppigUjp!ZpO-~E~?>kTqrILr#TrP#B0Tn#H7p6nQJKy<$
z|Igq0HoJo{*Y|elq9bevBCRnDR8I7kzO!VWbf2D2kE&!}jERrSyU6{Ij`3@lW;udq
z;aY-=sF2c$BUO^C3?mk+a|T68jFG&oKy0cTX@l!MQ9OcwHoSQifPC(_av;wUeH0cA
ztk*rtWGo|`E@q5M!`EJUnZ2qagt#Sz(?n%)F30r>8v_{{Vrn>zvW-@hLcFkEw3M-=
zZyd)>L=ARmFtniq4MQ_zsz<~)ASg=^#pBhC2M_Pl5+Kh|Az+l?-G(Th#50sqWS2SI
z3Bsy}`KnJvd1G0-p0-=@({H}Qjh&rs#w}5<3x+6`)>&NR$VH*g+p{iyW7=?6^Oh_O
z(7kXvTk)h0#B#*4cZhg0m9cLpSY^}PfzCLsYYu0pth*Me3M6a`_Bs-xBXNnGw+>$p
zKEkYl`KkqF*=WOTvtm3neErSWm{#L7IWG#0L6pJAjJ7k-IdWx?_=$#7m0DvZ2<y3L
zs7Cm@<H^aI1sT05sQV5T0wXhI#}3hA5oL*J4p$3~&(7$SAVp4e5v>F*{(=)sIA6n0
zUqunvxX7$_)QzV%IStUehTr?Ox4Ax^q9sLGR%IwLpfp{uv~3~>y(pk31JnN+){BnZ
zmsFC`u4_Jea>o0ICmb%;M1kDo?C3E@3=%cyoLSTGcy>k;10m08>INMJ#1W#q-~@|5
zL8N?z&yFhx@*EM-oVD=5;heKtu~UtB`1q9Xf9qXNo;>9C-ociHNNq{-gwwX>{ikP0
zFo{Dc$d+0n4X7C+uIuSG4SRzj5+e)O^7`F7th=67yP-6x9k<sVVuVOfjGATRnAZy$
z-(WLK-!wSi<E7sYjmfrE_&lZb;~`4XM0m8SdH3O*(O^nl_k8zzA9DYP?{oLsHLNaD
zYNLywBfWIoJ3Zs0$1{vrypQMzHY!wPJK`PLtkz8Ok{i1_ETac&IoRFf{`>^3966S#
zJYGda1-t}io1WQv$-G`brZGn2yOt1Ks*@nwZYW9ZUzb@!rC59T_;khnla`&qn5Rbz
zzWv+pu$Y~4YwrLNljvOfh|->o>-fQwBTh~m3?_9fNgALd3R)DdO>204G-WcZICU*M
zqX{a@d9pgAu!6<lR7cPdm7@#7*}CO)b;{aputkQFfbTm(^!V){pO(azbtxMabOMhz
zHQ#%1N;Q}gI>&E)^Bq2Z@B7?2I3QCwR%r?fd=R?Oa9A(+_JhZCn;vli9U@i?T2eho
zLZDmL>}Nw>+}We`krN-uCsPjV8o`rEs(%F)5fyO~n6Eu2s~O8?jjD12sY@*ePgDV=
z(y*!#w!9`+_&jmtK%OH?o0C~h5aHzT3H4?}SN9agQmTLsp8fHNgYgh=J67tHqvn`<
zM<)dD0Ej*fs=1I@axaJgy&_ja*Z1u0@A2lFud}~1q3e5)h|MfP_uw50&8DgGK2q5N
z*KP>ni5l9@(Rfb~m52-l&e?uH4I0g;w74L+uIK3KW7@jIM1_qJE1p57xpT0`V3HHE
zhPglGv|ezwX;Fj=+LI<mdP$qBCfQe@F$C(q<7dDAHNN)cw{b2K#U+N@Tp_Vb^e*1B
zZaZ=_pxvzTz9o?O#8yp%BY^0%@rYb1y#tV$j6so#)T{X!i`k5>ab%*f!I7KDxXQS5
zZ4WzCv}VJsKjv_=V$<~~5;LwBp^kwN!E8H@bkGz^@S(>RIluoGex6rfe1T2fU^FT3
zp9LiJ5QMI4X}n;_Xx2-jcmhJ-d)9qRP%&M<^EJE_6r*yGT7_fe;k^&pY+6iI<PxY1
zlv;6nZ<q0?z*Mk|OHNiNoG#aBi4?{XQ8<Y>0TmTQ@SUerhRg^@s}+Mm#aCZ{lhf52
z7dosqL>-CJ6GKFWz{Yv3Eoj?@h)?=7KCo^Zy3i9PIS+NoF0(8AgW;-({v44Z>{J<e
z&19%JX&lB_nm*9`RPY?<Ij#-7^vVmoaPua!`G&KzIVLs~vkdP7UwPwIUOc$YuYK$9
z()DrMVUyJF_V@O9dUnE}`^$fY8`B;B>#x3n>*q9`!{;4YmNPAfWXkix^$Bj(({>$E
z)u<-mx{lMPPqgz<J#!wH$PXxqjIALEjEe#%4Jbtjk=_N$uA#~lr4kPIclgSiFLQRf
zp{+gZ^$fDe!_|hJYQj&w`4!&%;2rMYe*}d>8;g&T(P+T_?he2FxBfcceei@|c<B{r
zCCy3b8dWML)rg{i+v6dF%;0;^s$1jwz+eMM%NiFGfqrTkz5E*57?~6nf@WuuaoDzK
ztrH8a4+tHj+|mYk>BZZ;aN{~hhfC^BhmsAd*)Z!oZ@=&=uiU)Dul@FKuwHFYl?G*r
zO4!@oC8GHE{vZDr=Ie$(`r2zO7H4$6qYWKdX4o6;pjF39*QQWMy1rwnSEx2%m2k4|
ziP2nea*14O2|<LRPAR}rXGH6%8<0%lypVM*HnU`%=i1JgFMZ)v&Q4djO-H?{84TdK
z>nJB<e&)?@@ZqEP`0fwhg>4iZB!xtIS#ogmE}L4g4=tg05MU)642p3zWL#$Kj|!$)
zL0h-9eN7iVb=z^aX>rbwv(0#3bzuD?!IcAfj!??sx{k6OQDiwI1;s=Y+8!M%P8Kz@
z`G()~^)K`C&Fe|~V=t#D4b9Op!y@B_+i&xGZvGLT%<fb7|0O{p(HOi$=F2sG5XM$>
zb8m;c<2|yxpz)e@+@Mh0*qyL5su<fGl%?4OhK1+h>X2PK!POnUv#8ACbqWWfzT76O
zl%zQY$;ecmPU8n9ZPyXhfZ!}o4;$Y8@)vmHg}aO^i%|ug@0dRJ%)1pgru+Qj8-EJF
z)cjWSebzBh*_<{CG&RR(OTKVxpV?xDH=1!(a(Z+~Ph{2C+^KfCd2NrO*6dc6eko+i
za6CH&ms2Xmx^`%68hBG0QI~JVQrgq^J%emOrV9q8q8b<UZHH#SeATeo2*2kWU*XRF
zF1b}Go6|0rgwx2nZFuX&FY<HO|2Plk_gHs<mVmR0S~*VFD^|;vong+bUSW!g+!z+?
zjz)T%_uSq+;9z%`JoHqBVR06TW#IAVn5h{N`<~uy3EYL0^3q(2>O(0ZN=pnKnI4j7
zhHD1{Bu2VeFqE95(=*<D<2BxX<z>o3QRPE6U5$TqN|2iU$%LPJ@lVmsG>2#3r1qX%
z8Co4!gqowd=C!-K{PJJ_Yy9Tpr~Kj9-)0mPeF!u-ZdN0%?@t+L1^X7H*SIclx;mrV
zKoLB%<p!+_5PiXibxsm`h0hOH4&*sPo20PheIN!HsfzvG5vS`V?>&CXsf*mY`x04}
zb9T01Y%TLePoO7w$0Qr^n*BO|?_1yDuRr+D3Cfe(M5O3c&%ALg)-`Y6-Q|Dqvp>)A
z$&zANp;V5ShCxx{obd3`33sj^u#h!}vzp-`W7RC#-5cUn$+XzvfjhY%ca(E7yibAB
z3T+ZmWa}Nns-n~bZrs`D?8#Fe&*o&49mbO}v)LTmm(;U{W$4M94%gJYlz)|zk3GNq
zy+6-VmROZ12HfbWd&hj!fxw^nnXhv?ix}xRwk3U7LE$NioWm!l40Fr=-D^BKIpW6n
z8g;v7SnklPA=Bw~Cd*@@)XzxxE;+@ljX^1hF%UvzYz9mRWBk~2|L}yv#&PTAS3$vi
zF=wO=i&aP4)|73B>H;rUKf`Z+=ezv%hkpssWLaX??L*6|(=0X}uU_BbzyJIG07r)_
z%3N^d__!g<(=g)w`%k$s9k9tuo}SGZRyUY$=8X3S^t#~Y&K>TZo^08q)TPhoN~Kzn
zLTJ3BC`+^|*u6PsJwM}vrze!-9d@QuX0sW)qms?CXWn<zCo^^ifmbG9;PjE>Kl|=~
z!XloLsgfv)pdzbg&C^+;v;X<mU*pc+3mgmzW>1&&Q6ZV5uo)+Zb2eGV-kobanV(Tg
zMXNg$C7Zm!m?0y(Lx?Ld>KOs&6+Ty7)j*yjv?L0ovmh*+z%Y)umfzybcV6Xm-O)tL
zwY@2`(;3kP+-$`(MBJ*OZCh?u*ZJ=K_xRO^e;K0{U%&Y^e*TL;hc%j_%Ge)VBLq#1
z4sjduOgNo&1j!MT2pMrepD{aXnUn)wf8{1^NGz-6xS%RZcE)=+pC#5;b@_+$T;OOE
z6i>Sr)n<%kpD$j2gF!XshsSGfT-#$gpVN0O_34~p?<pF=_kn3Sq3tw(@!kJBD_QXB
z;8p&sU-@HP8}DH=m=<G7JEjwd2|dG@Y6><VZv~zdp%bLj%uZXnmFM-BZ&H*cy%(~|
zQjSW>qD03kagJ%dZJxfQjoW(r+((X^mcotL(bxIP3vV)OJ+ofen+`cWImLH9-RXi+
zA2AKM-f?quhj*TQz+Zp(m$5qIr}uw~U;5G?Om(o_uv_dThmR+BEkorIG&UQev|!Nq
zHV_)c=}|+5=FL}c(}y%5S&VYZs-VgT1aG!On_QYKQUOG$z2a%x(AO}EySzMkh3mW5
z`0#ke^=o@<)+_qHr(0~Onhv}67+Zi?77NdR_JjW?%edgx@=gBa+kc!Fuie2IVU$(u
zO|B!x@bYBD*LJV5qYcGqN<@;%V=#y_oE_J!mOU@t-e+fLO4oZ#p~xpWgYgLQIYA^%
z+>5!GF}lL%hpPzWIU<l!B_~Zs5r=37%pNVddEN3;ufD(si<;P{;;$b=W^=|0HeJba
zyWqpKCw%khI}GfQm-AQH=`~-x{UXLJShzW+5e^q8eCgE}DF+oN^EE%X|CBl!dNk8=
z!hxD_aJ}Ss<2app`o#=YmdrcL+?7ncrty}p8KTAk9W#_<TxNqkA608ZV6jx(R272e
z>3dIjdH)9YRx2cUR;!-fVu;b2QK9HOJob)x7x}e&f0H&k-WvZDI|vi&dF#bD`QU?h
zDcb?_`59e5VmKNSy2wWlXFQ%qd=#qDlwG^WbT`KbPwg^}k5+V@$LWkkpL5`Ie2mOm
zLseLEljcUaq?$zBc5W?tPYjk9#zUHumWAs1!V53Z9v{<e8e-?sI~irBC{WaBPQByd
z?3mwp{P!4TJG@rD!YK6I-J9a{3!FVXrRZ}GPY(I&YXkHkBU@X(_u!bdGxQ?dm>#gl
zl!IGCPS+b|Yw(K`w9RNj&f4b~$*6;7)f7yof)b6EMA3PfHH-*-fMx4ApukXY_~8?F
zZ<gFVI6wjOWzA%NfDssHmfj09*YZ)l;kTatI&~lT;_z$i>p;Phw_kjn?`_r$!iZ+m
zvfRumMkUHxd=r?}kzNTlFWEIa6hlK8M4Zn#K3>yq0wJ^HdVpwyu?3CKF)>4F4Pw*R
z6+Ty7x#;H!N%I>n3L#kXI6%3ArqK)slcX(OWE|{YLqr%2Cn%NCu6tH>;QJ?!8QBRx
zv-?Y2HwBehvpRaj&S(dW!qKx_cdXl<|8IM5_9R(y-RXVD-S54z<z8!VAiyCp#0)i(
zD4J2yjAR;(%=9ugf3O}TJ&iOVDQ*BrG|*eExyE`|UydF;GOLP!fb61rie|i@S(&wE
zS5(FckL$<%EWfksJYQX|*)<VK9X@&H?vQyu=kaX8qoXrA@2Lt$RaU&-t@w}6ulT%;
zyzC-<ED3(d^B<DEpO(qKng|j4D3~<Ed50egbZ|UAIb%MVaCCHr7{`2eK;b6%x~J_2
zetvyH69)e1^xyHRJL8~;w3~D0u4FcupeE4yflU|iF>$l$cz(IV`++{T7*jGG=A79D
zpFTJvr2?M<i`jx9CH~^sB|lrYTn>p=r-c1LsJ!{y^U=`U(-46W9YvaBhmx+bOwEi(
z$Hz>oDMu%#D2B!27^9ZH?$~smpIkp@Q9k6qKKvhfU<zhMpuTy<>Ef8uSrVSQ57a}z
z`^2m39T&?6jXiDXD9xN%zu?Jq!IRT7nm$li%WOJlx!dvM=huAM1g;0c>zwGbXneNu
zo?#wUj1_zWqch4-5r=}lfd}&i6K6O+IRj&;CUXi`5xRi)o-eMiX?)<1PW}y_+GFM=
zbp49Tz~O8`(#WQ+>4!vFRTPEg<>iX!7uUp?%gc*l$^*OL`;Q&~CEj}uW-|i9FD_fY
zSvsz2W#b*bm;KaAxxalQ@6G6<KOjjX;kF`HBStK#x9rwi7IVwlgA*2K3-E@jPdr>a
z;BxztI4FIvEZjr><iWouY_CxhiDts(%@y;*$4FtoLI}dM%MCZnHJ95NK}&kpG%@o0
ze97tjnAN7i6h~AC6Q;vJ>l1y`@J06p9dj<#VdRJ|EOF)S(AM9MlJabZ{WK*3Y?{z*
zTdG;$(??IJjtY#KAa%#0m?NR1Z#<<eh;hrGJo<Nh#1mHipJAsZO}!&6mrUn#R_7}w
zh2wJ7QnzcqdDhc}DT1;wf!%V;4-O)Y8HmPFP3KHaP0=};rsvD6S9G&uX5pB?l&**A
zY(Mbmo*^D3m)_=ytNRAaP|<9*93MD7d-{a=!Ib0;(g%(vhg>cjLYFWyV`@+Nlhglz
z?yeccqo&~c>YBy;A*OP8@g%KydA(z|dB)2tM@mx$8Q4nV#rcxQvr}$XH8PzuJ(w{a
zAPumoYrd+#qBVzXIt5wK4%bxEk~@9;-)5icepK?^4=`FVG=XmGIXOLGdNiXb=9qTi
zWO@iqN4E=<%qZ;%e|-9{`JR2sP53ctE!$m9czr=RozZfGbC%^Uak1QT{;J{Cdcdfq
zW1!v*v_)VLi|#rm(*+Y#Q}l**-SX4xb9TiclQg56pHWl|G1Yqsps^3+yJmDCACRDi
zK7cir%FZwXueuf6_MDH)IpyO+rqu*-hOTR<eL}DtEehsl!oNNHpE>4t_#6M16zo{7
zS}yc8fAr)5Wm(Wj3(;VtppXTXRVtU*^mBYX;ra54PwWZaSa#OqqUWJ8T&}llwheW<
zVSV@s2eYTd$rg{|v?HtA{&o^Vm%yu1ni4CHUIU-6UvQ?%@$8Hzj~+8CCiH{Sd6-Md
z!MvjO@Gtd``M&<Q{Br#t8FWWidtS6J_;~girzZylcTI|p;1x5SQMkm!2&<H6(+S<?
z8DHshoO?nSJbhs~t2{Q;JiA^I`W>$?zTxTVF>|T#!4hLYi1)7L4QgmZ4n;aU!C1qp
z-|@VD!KZG<!Qlf=rpMUQ5r&o_TAVF7I-If@gn#qkzvF>^%Fp6oV$zQF(D1T<#UDL+
zfUOEz(-8+HC4;eZD(4xju<0!!9rN||Yo5p<gPXDIYEm6|Qdzc}mesbVX5eCVNok)j
z)r!t5)ilAKirQ~RksR_29xD~r7I=w#u{~!NIvyQ7<Y+c$Ud`yd()Nj2G|VRjZK3?o
z{R@64{`Y*b`oGzvEm{X&H|H$uA;0(O_gEH}NGNF9z)c-_zS$7Xls+k2ow9CL{2gB-
z_7jp1G_K~jY?x}`o9h+rZp}C6-*7VfK8NKIi=%<<?iX3IMZ&$-aOC|N9mogdrcG=k
z6t<+aQ&bJgfGV_Gc=G9#g*{@>0i77u^_I$7o-Nm0?Rp+gkEw?ZH|-T&+A*D4Of>ke
zrYuS}X$PrdqZQx#!3nchp8xbkV&D+;h8;V8*1u%Rj5gHt7du=u3|^UOg^|d`@|S#k
za+Gf~YVzE5+oST^Pj_8NTr?1kvalsqY@WVnJboycPG%fVPf5MU6;pOi$E2|AhMs3v
z=X`qfuaM5N>8`oxuP~J`DILU?$>agC0;wnL`hurNr~Gqw%3poevS~_)7buZ0hby|)
z(Wi#GZ*Vb^f+9hv3d{B81&_}D8AT!N+Q<+RhzE1O&1c;umI1_A3X|t~2Mu5Xeh`ih
zPng*Q20vh3#b&p|8o1o7xEVZuFndbB>sdC}Y|{<1sbN+W#85L`%vjnjCQexUg6}^$
z;joJQ_=}cCXAn2^4E$|>&aqvv^ELg@Qlvx}A~UUUHgb9M6+ig+BSa!yNF0QSnjSU3
zeGNlWUT-6Pgwj+P<A|D2<M9C|<pC#)2S`ll1nON+rFmA(`SOM%`vZ&!jjnj!Trepd
zi%CW01Xnpi38gRCc7~H;&Z9$6FFe~>hG0RT^I}*L+X~g5ZQmksKq3$XXD!Rs3#P>*
z+FE&i{RO9I**T_rd`CzAVT?JD4@lz^0U&{oSvs*W!otm%aKP#DG1l2UR7pzG5J+Np
zansN?iI4_T?CEG}W#H3?$NcEYQ!3XpbrnGZ*KN-?{W0rd!8TPap8gNm=>lUq1fgc2
ziyhm3pzC25l(qKMCSZ(Vw|&i5ul^D@pCY*1!Fl)6QHlvV-789{5H)7WZ&$D|M;uO0
zS<L4M7VAoeCeNUKweGmt1Wa_qC`9e4wd0^P{OIXNoK8w41=K(jJzwsWi}--eknoEi
zQXT(GRC)x(t5U~?cJMTV&_>6W$i{eVRid%umtX&wU8tGN7W91{5q#e#qmK~C!5~#7
z5-~<t6csZ$<oMu}swlCx0C5a$B1y$pR~!1afMn1JgSPbA^W<>GkDffGWWcEqC9>*0
zUo=yeals}O$mxHBn>_`|YuF`E6I<#b&<~a-8di9k0t#2qG#C8xn?EC!1;!Rxwcvi0
z*d&D}XdcU}IqZtS#2V&uKxL;)CKbU)tSt!xh$(rss(E(4L5Dn>C8{TAU}}{gJUZp9
zDv7BlmWECPU)7OS`Gk|RGt%@Ebn>q-Rxo=N-!8T6{6Oo4K9tnJlAg(8&M<7aTz-x(
zl+Gu7NI85pra-=PMhEf%xeXv#EWvw*7&$KIJUw_op%p1we2O_4-tTbEk&;E6;}4!b
z<#4W~XqX$vi7a@0IOFVKLMavV*%^H?@Z4XrqvWf4N^Q=VoLZEu7X_dxFhtJ~EGko?
z<nYuk)btwj%<J%qFQ5OPTwI-FF!xrfy^R7T<B*tW2aiSg`0ydqVvcWxz8who+$84&
ziHWF|&rUw#@f^Akv0|9JDQB}85BHTmvtmIxo$@NaVi^aP>4>k|0~Q~Bf|Lzn)_Lfi
zMiK*^kD!i-AjS}NVB=c=LR_(F{))?+ud>21Sq}W$*r?)2KG6EW%vF4P_!yfe_&PCo
zPa8TmeM{*q(MQBi`O(9VsLX&*N@)uYi#ZREW*i?>*a(NSGkoEB=3miJ@O8J~#+-5R
z<RQup*0%e06;Fr*L)a@27>f~;9cLUslx}#z&E_vytuFUkU*E1%I~s$i&<!3{`1s(A
z<Jlp@5cg{sxv94(+35szJUV>D<6}o30#1cPH|1=4%!89D)<=p^Fq<869xu5GJ?nJJ
z&n_I*N59AH>;N$>#u}0YR0BiEp}eT<dy)cm>`_rR%{Sa^{+hs+zHM_NeUvi!&KVuZ
z2juqaeOSyek_cg7VrCeWn|{k`wPdqvdH(VxR~IjFg|gjlNZRw+(_<bzy~IdhW+ptD
zp3x6IU!T8bwONr+s>20L6;H##X+hm>Sgo#+)M0GQ;RqiXyw8G+V)jSrJaZ$7?)RAl
zng#jVu0<t%t6@D8N*1=nhYp8fA|+iISnbyAb}g4TS6p1YL`_fEw(RPLX<>NyxI<-w
z#qrQRppqG1UR-juT@ihtC<@$>rI!vj^^_Cg^{W>o-(yURbtNf9`k^Ppd@~RcQcNTg
zZRofib*R#!4D`N3qyJWlyZ{Gg%0vY3dnTsDsbw2F*6StPrseg;OKz5zI2YJ$b_~P7
z(ZX<awj`#3iK%#0JfiEBFRm_FH#<m5IjxXIVvvr-kudZPH&<7nEye_lDe&IYdyk0;
zgU#ZUN%7I+b$?JGqOv0q>4y$Tcf$gH8;atz;INnx!hon!F`*9~n`X<lZn?U?<khQJ
zsCazWv932fSRC=m$DpRCq~ydMF*S33aemI#ZUZT6udy%;(jjGJda_{K?1-@e+2UM*
zD)c@Oqvk__fgxl~w{^40#S5Z@bV+23*N8}NnH>3U8FL^X7(w{a$*0uaK&Kub3)cOP
zSM3F-aY|L!lv~dvM2sqBRq`)B`-I8{Znm%4g_5h`inF87xZHI-+t!>JB~(kgK^Zjn
z9h^)~`F!(=t!zO9g2Ra9N-31m7@z=+cpOO+c0a2k-JZ%(+7Q0=gZc67gx{S#<$Slp
zr$}rfubNBF{J_z60?USjCSex|>m0xL*(X#*pzY6T<AP3mZ0djmzFKY&2~^jwS=Svy
zGB{OErVCzmYtGNl5qk)Lk%}Z8NtFW0vf~I5hpHoK#M$D`J`i?&&%CVowv(lPc=T}|
z#5E{BM7p8nwO?}Bm$33o)}9oy#Q1bF<wxKDl=)2A?Y^SdDci7S;ZE50k!R~2$I%mt
z*K{8GBskz?cEXp-SFGZ4fAesAr#O%xub~J*4I%|0Wp!GE+1r>`vF_VENBrKmEagG9
z;15r}$II=OK?BkouG$->;gHVv==F|88z@fl?|J(00SA)>Mm9t)XhTo$cXXlTMN^|D
zGP$|nvKbiEVu1&<DeLPUzxd_fQO>>xBt%Mtfv5>lLm`%=h9m`E13q~QSE6b$Mv2K&
zcO8pL?yaai^8Snt<O8xF?SxCpLpx_1JiFkzYHCs`6w`|Jdd(9p_~`T)6MGIyp;P76
ztJi#aqg?xj*RtlZ+vV0bTXNm>ym;|73BxcL4rT?aE3{CI&IsCof^`NFiv$5t)I=nv
z+`*ml>};c1-B%^gw5Yq57^mDztZpBJwg%?O@~}9dOMz|}*o211^D{24ub94G^ZQRe
z;%H)V8ab<!AqD>8moK<%;d%ImNzqYyL+MJ^KJv@U3!13-kT{+gCIq6Xnam&2bZu7D
zl|<ntNXT>J#UwQ487Ia-yq!3tc`XD8Dbe|!`#O-@o0ubUW)4^xPm?^?O-n!Np%Pxb
zc*RGZ@V!r;V8yeT3TJI%SMT_nuNt-<UZhJ3BOLew>q^!`;^OslB8IL{Osv6#8Y-ns
zhs14%C|H}YGR1`3S?XD)-AEq9q3_K7$WF2y2K)r>J8?+?YlSkxlk$*0c$%OrL&w?i
zN32#W=5^qAkB&L29L@@#9M9NpH~jg}U$fc;&f_^7Zm?QW*^*`K`0D0{HY9vXOl_i!
z9v>UZ=@Bt`WPg)zMS(<zL9%>9iWnh}vVH4a<{Op4$3)$AqZ;pj2F4K92gDd5lDHnW
z6t<w0lB9-}babhsZ$0aqB@<I%O(KPsp>1f}#I6Y}o-Vl3XLyulw_~8@_n$t-7X|&`
zA>=s<yUmumTe0j0%BhfKv#*g)Xb?myki5x8V{pb`RN$={+PPycpRQw)dv!72a@nou
zHBy-Zl2FsL(k&?|o9&LeX)z?M40PR=*ui!aIeq$&cK(93UenmXGOYQ*gG1&A$E*jB
zrbN<6N*?QA(`+!M!<7xH`}u__Avn}H)CkE4Vht9_vf_6=9|=TBN$L84xo0Q3n7lck
zw@VPAG6g9KYZ~g*5Pf2~+0b`Al~ZDD82Xmb2yF`wzIV!v{2iTmcG5CP%?}?x#Jh^t
zhdh3kh-yGwV%6@L%pGcW2>aPAF&A0ft|7+aY+gf2Ip(9fcP^p~A>u>2pN*2gr#Ic2
zCiYauVHH~Gxr#Mi*K>7wjS+(ruylCeklMs{xy3BzT$pE^$LF**vh{1urWNz~5$nMd
zLnOtBA3EBhVcV~9(*tG)pz7bOp%#eU+h+~N7%XPLh6dF)gTNG}A0l=XQ~b^uZy+C#
z+XLX0U-P(}qsgMF;DI^8RR?t2fwu4QuE)rP7zg~2Pz<xP176oRbbgHp*upbEJmaQm
zsl7)4A5)&c6yPs@_UBw}J%^9xsBJ(jqC+0SXAI6*f<^>O(gGN;#*h=rDQi9<BvsmQ
zf3ZP;Hg;@u%UrWIQc1xl#UtXj$EV2Q(E(F8$#MHI5LD@0Bxq0D-=Gq4MmRV;q!(q`
z)LCL9DM^L8?)mBGKjHJ2f6If%Gbq*wc^Y1jeA6gxK_U@N0xe0g$Jm_r$qCmt3)06t
zU;f*$j5`{3xfDAU9GU}W)=;l%jI$h7hd4@N?C5>OljtkhbSr$?0ZLI=4iAr6`j)!u
zvu0IHXfoW~Z1}UE{FtlVlG(`vq}XASK_uIx0)=q|O^8@DmIxkW%$uds+%~VO4AI<s
z4J|@Qo@LzPYz}3XNqAD6qPu}sJG=!aC&ZZOJ5NYpW+kq&bm2A8G*~el9v(4Slx!|8
zK{Zd(M5vpA&%gWy-<*HJ(b<$fHi!r&3ehN0BBhkv-ZaFP;QRf&wf!9QyKcuUhZ9Bz
z@|`m}kbea1Vnc3U3k*b#IHJ&&cGohQ&u|l=6hl2k28Co3+url@S1<D&GJ?=?zTR@V
zY7vcDNKSHNoas86-b1nG<<*~4w;MD<VI9^^5Xlqj%8)Z6od(LH*hf5cKhN6iW6H`9
zvh4YNi78U6XQoqn8E{(h)I1?HTTULIP<B1l#L{;yDJG)Hn&V%;Jg4q+d|%Ylh8-_9
z0Y40=XwGE-2d+Z9K;yS8o1d_{*$|WBL@4YOjLkQaC^&^qJ;oVCOg7Je#9KuGVbFAM
zA->x~Ko<rE^1!A}dKMhxQi<_!ba>39ED#cY@I(b1)P2v3^DDX#_Q9o&i+ac9u7T+9
zB*_)Sv8!n78@hPIv&$c|T5m84r8u0K+=(8#nA77!3KZ^Evn(f>!Q9PrX?^5g0f;wI
zK6a_)fjOd+7LzJIwNLQvj-y9Mn3y=2mb86MOfi=>7X_ccdO;ssf<_EVWZ)N<Te_hq
zNy12h0u$pX6KtCeI{g{fSJ!tlOA0eVOrB&&%=w^3JtBe=mxpf|AtkvRe;hR5lt;c>
z#t_yAMD{1ZM5IeKE5D*mD~wYrSI~!ni}ek?40zu$3<0cAPAclobK<6`4aB6})Gg<m
z6>aqU$Wz`ttT8-1t+2+h*)+U<`HH>^^nL(UN(xd^!d?_mCEyA}VJfVcoJ=?S(yTqk
zCanAI59h6YQ57_>OFK4xg-S>19Me>>YPQ^TJ3?&OY&Up~RIbbwH$wnBpuo24cwOIc
z)igwt`$-HAjfUf+io;ojh~fI?lDcl_LZCO1Qi?n(mK@44no!A!*HQ{n0%OenWd26N
zq$uv`K=vO)9=Y1H_ABbR!q~vnPOwq9*sa;chOXPOYibmj78OI|DK(+ilN6Tij+eV@
zwtgVlm?w&<MS;f;CzM5jkAbV}bGojljh<i=jsi(S-l73KVwBQOaJbw@WXwB#5lQ0x
zY}~&lDqtepxMAa$47Ot;j;U5G>os*6Kt0hXl0aD&`LNRkqRk@!`Ve?spL5x45t9%j
zSvfZu<}=6X!4xBgrrpxEI~w29OTtNkjTXo%(P%;$u+~vhqWd*8#_+b0MRBhs9Qg+@
z-atMen)cZR6*j{)H|Yk_6zdKc2IVqs*u@s}@k7KK+TgJm3R9vXP>CfbC5hp>Nf?ti
z1!EjafhL0$m{lcMlbgT{XmDT>rkoH_QcM)KgcM0hK@9~H0!pvL{szKb)lN4Igm{l}
zL~kWA`_Qn9uc%YNi%_H~vF&-*pL6=)6e%q}MuK`=DlkLf!TgZAv|RMBsG~A80V5VC
z1xXABhauwj`-UPp`6>mfg0U599U=Bi#Gy_}UJx-9Qes5uB<vIJ+?DjA(CtSC-`7M{
zVK>~c(j_4{tX*K@l<j)MMO<PZo*>rHhk&Yw;3#6^$;l%wR|)9@t3Kv6G!{caq`+!{
zQE>Y~SDLam6<VPtVQoc-dBTUY`x@6ISScu|h~nuqhcw?l2KbnW_o-6a3w|oOboN@G
zv57sEDl<DL4S|d04Ie#vfLdkSG>EZCviPRu!TbnU9dh~X3qn*@UCb4LI1s_ABWb|x
zFArjbwr_yFV5j7+&X7D4SCZ_lwwh3y2@atTejlCN=k4|)ieB$CGHc`?!stLgAkO6z
zVAP(S_5`CebhN_@z_UzS?4+QuhRRryPoy+pjpN8x#G?niNCQQzbM3!4Dy{OsCAFxv
zpisFg&%!o<u_SVNXJKbl!+;k-#3R<BQE|p$lAtOi&9kNTt&?5b_6);s-(c-N>|jLL
zrWF=XZ=eZR>_VXJR$Obt<m0E5k}Iq>+lInf=F=G^mS0|O&;+H;O@M{usdFhQMjXn3
zBZs1_fkG^iJV_7ZD2*Zdh_OQ4E9@H0PADdiCLzf_2lCb}YrC!QwT9N_e^Uu;(~2O8
z+75K#CD*><roASXu&7F^qJRiZ*JA9H)5BwSQRxSdlf3ySDG-^EG|x&;Vn7s!g{YK;
zA?bcJt$^AI75naM9Z)PODb^Jj6@1dYAmXifVzx1cMANpv)jkoySwjra#0{9hU}5X8
z34`E$&6RF&j~`-9cIun;2InHj^93^um+Na%R4QjlVzF9+SW=2PkuM2@V8voYD6P;+
zF4GVxu&yA*o~(JC4*(`voOO`0x^SKY!aJqAtIh6SYdG@$j3KNK$nAF3t1=`31#c2N
zS+X@dyoG)UTx@GVDXgK6HJjKnJ3eGlPV#9?m8q#HtOKJY0)oe?0!j>d^SC|98WIzi
zaAx426;pnvIAB5!aTH=Hu!I;0DIh>h>7B;95cQ2Z>u<78_eHKo*o9ojpMb%%Eaf$8
zlgo);TwbFoGIdiB!%eqiC7!KBR!vPvk&=Rn3TqPJv3mQSG!KDN#W_dia(|2HK;b-(
z4E%06=TS99B==#I`>BRLh1?61(p$YCc~u5~-_z@ULl)>_{)_|A$eL@pppk&@Bj?vQ
z6e3KDDZK_RyDi%SmP5y`Z%GQ3R1{({CL$Vesv!A23I-LWDhdkgFp_WtoOpgvO!#cF
z;IJrB6sv}kk{Dx_k5Nb|k#w&|k}D$c{ywG3`R^Me)G^;qH9<6i8!otJgHLe2+7g3j
zV)HzX&CpS+XKe$ow>JchShZA?SgSz5#w_Z1+cB;P)>>v|wLg4#sv_~YD*0@B$Z=I+
z$hWZ4OdtwA<y;7wNJ;lab@|_Gy8GQCN8X>&fqXz>NK6X>Yw7ksGbs-#npBo)krRYJ
zyI9kO#MI1bJ$!X@%~cG1y<4%YcSHpvB_={vSq&HxPQfJM`OB-k*(SxQ=lD=KKA7ON
z<A;wv<zGMg5i?g1i4>+lVjzY-_Y>^j*X<;`TkbPzAtuEcm`zGL71I6ytSD+>R#}WS
ze08(oqV*IMU@M-lR(x}_;q%Rst8SaCza*Czlfdr79Y!Noh1IsD>ms4=@omSfN}L@}
zschuoV$T2a*?+~y(^CRD`R@>t?}%~0?57ZB``#FJ-wuKh6Qu=bE!{rUV>I{B*wQem
zveW#V%N1>aiJ3B}<%`Q}p06AJZhb);JTWQ86gU-xgdt%ROw0-XtK|+&Lc6UYdJblA
zdQ@Sh<DWnLg#YQIKV)VLk|@qtBu0YoNhx4WKH2|$?qg%pJX-5;S~4gEk-ZO6f=SFK
z7V8YZxLR}Bv`lSDS<QK|+3>~1HJ{&Hb5XBRg`~Ok)+m_$U|o|0tA-FFgI7Wb?ABYR
zF7oiWVp=K-8~I;9`$K*>dqg5(jK!&te7|4Ay-&OiMc&#swt;--j1J@jvN)VFnSu|2
zK16i?aAKe;471q;v4*dfH8;C~=wZ7`tkyN3zrN(>tJlPR-a;kMl{RWY_Ig1oD9UQP
zLzBlDD6BFmJt?$QGBBGgcv3uNE(OXyY?T5k5kU~KyeY-WPv>o_;J4wN<?zrFlgG!%
zpo-D_U{{4<KAAA7N^W}3FRs=|O7x8dZ%EsgU%YzG5F?~S&;i9`bl?0)E-x(*mdh1&
zQ={q;O-!pm={zR3qzN3DM?7+;NCIOW8WTyvo74EMXM8J(e&70?*|cOf6MPJGF_88*
zJr$T%j)UnG=PX}rYA$yz8WMeNn5tuWdC9Zo<-W&1S11VyNe+x5l1J7Wl=I6gV)EG`
zNMttgIPsLy(-(%1-S=3?WWQ!Euc3)=r*(2e;kO*JWQ^hHcuFEL#7H00T_CJ*hWTVl
zRaD#zp3kp07=gZlruH;l%d^W@`+S)aHRd(M<o75BQG-)x>Xzkt3wXjXP!<7aJ(4<7
z6_}<IzE?g)Q^H!ex5@8~(>v#%oyNCY!;$xAbRZv)6eH_hj!)M?8B*?lwniw7Lz57b
zQV#<^y}Bl)z~N-hgT(<aw=dZ?P3|BkVhhDO!LTn`-RA`a!=|t4+J>sM6h*~)7YNA{
zT;$KMKj$aAuUQWfs|K^Lyz-KW!oE`G?>8;K3HAIYj0oFZr0XL=m91CyeQivQ;8MlV
zXE~LhU0l$IKxKpn^Al=`v@!QWAW9-)92l_>g+LC?U=ztEcJ&r#B`d=QsN3A<@zwSP
z|IhMgd{JNR!{b>8YakNxHHN#$&RaI!TZm6#vr}SJx(H3oHd+f)fKU-4G=0a<udYCq
zSy6H{J>n+RZzM)Ef#h6`#buRVC3loN**BK96op|rnb35RuI~|>_^Zp$`K#^Ex$N3~
zvRsLp7|12jViidy+xIPZRU=SuJ#`aNVB?jTqpg${N=kyyRcL>8bwTYr2%d?XvJ<5V
z!`<%;`21efU{WqA&WbFFXd`vA#u=DQbDnEc_o#Ya_e=it=5P6?UqLc?4K=Ty2l5%m
zWap@F-y+8=$al?{ME`)awI`+g00b49eMq{(QMd)E?)Ty9$WK-`{PD*>z-?ag&F+d|
z5+ZL`)dVI?Fk+A(D1js%hfv2AezT@ak#-391d|esvAm2+zTEsNTW^>Z6A}@h(w!`M
zekd(Uyo;URcNYpF;sY21Dp`U>6(m7r=d=N$dIaJ5cFh$qDTJGP&2`w4w~GD06%(TN
zuDiI`gi1gJl8KyeU(t30VhzKPs4A1gHdgq*wm)Il=Bg|~2uTT=P@{VZlU!KvJ{wNg
z29jnQqYAAm2C!-=?2J|8^JfSA^yZrX^2w)|<&Ky2HLdpV{98UID(esgn<T50Nx))h
zbVswU>3tsGJ9s#p<udJ^bo_6ZKc*fGlX5~5C8R`>h!KOT0rqS7Ej7MURGKaj6fi=s
zipgpzOx%=qi1?H%^<Hl`TpLF<DQIHPDr_;bpVmmu!{kxH2CKP?UQNh)T~4@e&#CK{
zWO9B+Yz~`>5&ow8C2cLJDG<zF@i))>kiFcFAo<Isx3`hefqd7DH;@m=ZDmvLzff9L
z28B{4gnmNr^9F|qm(7md;)L?-V}hH2D0hvv3f?Qm<g2~Q_mC_P;sk3t`Vi@pFa#wk
zcn@7O5aNKCnDYWLk+6iM#1wK@xDR*l^lce;`86^1U$rlDD=^`KHh~p~(FOfrP|bm;
zCInvLIQ;ng?5x2|9rmq8;*=DTVqeyoZ4@G+iNXf@uw#fq<CTygB<TA{*Y+qLXL5Kj
zro8cly?uTNXo`3CecQ0d?RI=S(zaI^5>!&6aY64DC=e{KnjOROBZ|{Ubfv{sHdn^o
z*(ey7n-`4aSs0=bBZ@ICt#28U(E7BuLFn3$6>?3$8G|MzCh(dFF)##A^nu$tx;H_(
z`=NUq^PN+P2ATv`EQKD@?AOp}Nf#qu51xm=^COtfDQCqy_T~LN=e=?W(f|Qc!kS2*
zwsaxU1Z9XyAgd^J%|I|7F=-zHm27|XMD=(d?wm!g=pG%%e=bG`@&QS2*Wv4}Bgv|`
zeK!yy2oo%he(3o>e)6X*n~p_!mUA7qTY*3l#p^)SNEAad3Ph|CQX0VMe!~Ssh$?iw
zC-|%>mE|B3J_URVcN<0?{icv|+LQe9d!6py`8)evC@P3?3`5=|z#&)!!+-kuPx<od
zn(5>r(eLYW_n!?w-1qAHlt2Wd31<te5kkzfOC)745EAr0(RE!;))LVaK@&dZTY(^`
zM0DSVu4?WQulL_%^_~Az&BdlhO1dr)lL0Ld96m(;_rLxNwgXJuaW1`m%f>7~?uiU|
z&E4aYl&UOYUqvPPFM{L}-EIgB{eZdEHq)5Z+W$Y(2^<ITt0~r<qSm#TmRPnZP3B9=
zB4(3sheEDaW0aXa;%c#2TrE-gvLiB*Y4YGE)x%cGLWziTok_Fpq2_Qq=0^MW!H({)
zxaW_rA6}2g`-k^$ufv7)1Ld3F9Zkmz^ABXT{;Z$Q7tZY@w13D^n4`_YUS^pa)%ni4
zg@ru5;unIaziR*Pxn8a!cBs0@1R4y;cIIonRd|$~&+f2LTP4jStMGiMIh0YrgLy}Q
z0iv2${#)5O2EgBn9ox_?UCRItIaaf5&CbJS(W^Bx*~|-RuX%D3>CJaX@@U(VgTaTl
zZzP}w=5OJe-DrIK`>wucU*3y3J__}HYct?k?{sAnY!V&_EM-fy`h*F~A!6*%oRNy8
z9k+mlr@Ed#m767MhMJ5$8&SVg!&KG@JM7^q?}|$=3y_wfSIWqfj}48V;%fZLqmc;F
zn<sJ#dsmWBsm=Y1V5KtH%`83tQhHNgN-q+LSc4D0wed+)bx@<nl@)3HJJz`eBT^Bv
z3Swh*_V#qPJ?e*4@*uU@4-ppA9=K=Vm<fC?ffLldnS!zedp6YokQY4Sxouj^Kb5Mf
zU-uONn4qQLZLwDyGCGHQ6LqI=P94YqY^0)-uEnkjOi&R=nrC+17Pqd)t+d#^&LUN}
zugmOM>H{>&Hob<`8a^Xg6;5kw(h6u_Hwf6Z^HOZmI)jDqV2eD3+REedK>=)(sm4R3
z(OeeDXW2Yh-H35e$lQA*I&uxWBMgeSeE5B0@_J%XoF2t<7e#&PFRB%gCcL{wBW9~K
zK0bFLO@g|vWHKIfyMg6T_HYb+m?UpTM_og`a`L(xX^UV@5^91RZpFkbAFQY}nSiAb
zdur`Tm*Vn^O`Xp^TO6^rexGS|LWikP#!^09m6+ekG$V|(>~8&-=34J@!GXqUQ21rc
z4C9vt0@^K8&)-1y-i7pW?zt$GoSryJ<=?v_5wGYXF9SuqS8`E=Svc>utJEiL&y@JE
z0yF>PXmWYUn#3Wn+h$ecy{<B>1d2valT#pYkT)VQvFzv`ub$dMh!WNx-576L9Dcz|
zpt;|g&_ln5m(tAxWN^Jh*%?DX+-xY*5YZx;_QGzmRJ6R;rO-#*`%agATt|U4DXwSG
zn?R0&5~QGgfpaAY2op~mkDL*2c$BKWQ~i|Wcwfl$&^DQ7joWP+6X`m!@%t;6a6oA<
z0&CO{Fvlmv5E}KPr@kI&yX+gdv%hA?A8ibWzS|YOvUsw8Jn<s88cgYvQc3Whw^jrN
zu2bSNeo-lufg@i~K63*Kbj$y1w4Mg()V<D6g+d(zUU2Fac>?JF;%I9Y5i!QGt@WEz
zif6^R5eHWAdV1cOki0r~3UX~@bMBuh$(l~C>_jKgSYlk18#6;fLbAA-REc_+xi~D^
z_~=R3G{Z|Kt9{INLY6}S4D1;lcDrPhP$iNX#26p>?<YP-4TsrBAy(J09#py0yNI;U
zW9_S$vt0Nh4rl`{5u^;9smn685pJx7bwfL>wk}I|{x5Y=pS-a?eud%%)Ta9@26}&C
z(F!f;r|ottRIeuw;a14<2eWUic)o=TTIu=U7JV|VX$ftBtjaNs{$?4$Kg5skdn)H&
Dd%;;=

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_001.png b/0.16/_images/sphx_glr_plot_visualization_utils_001.png
new file mode 100644
index 0000000000000000000000000000000000000000..6360652f6b800d192b3b9fc046701b82c714d454
GIT binary patch
literal 239651
zcmce-Wm6nq^evp=?u5a0aCdhY24^6FAwVF5ySoIJ;O_1Lf@^Sh{Q(K?gu&gp`PY4(
zTlW*(7hT=8yZcDh*=Mi4_BxT8>L6?kGK_cc-eH54<RS0gAt1hc_Z|xk<(~y=)<pg9
zCgQ4~=c?^s?&@LUZ1zsg#MRN(!PVC4D~-FEvx}93Js+nKCm%bFrK_u>izpYD-T(Q3
z)4|z-Yf4(~@Lwb7j!ICMckggb{^xpME>&jr?)|%WV0jrG&+OwDwIl{`Vritr#ASiC
z`6+%v#>X`Q35kwI6&2ylAmFno-s!!MFaql4yFPZ54nD(=8ThB>m8^6V!nNgI*Ld!w
ziQL~MTrIV9KAFk4X8nkY;_m*!etP=c>Te)bL%kFSAHLanB=PzC_J?}&e;%D){=YYG
z?MrcA|8G;k*8dxW?(65*_>7CQ*R}4Qn;T;MhcAzO|No%=_i(u!+r&1y{C22cx2QE=
z^3LAgO22MyymSN%cRvWd?RPJSKmDI+{Lde6-SCaqm4I=n*Bz<X>DSi{9jsqnT?pqJ
zFK4el_VB)&lFP@xV^YsQq#S+xw(2Nb#hTckZ>9tG1D?CzaG!6}gp9fj9`?Sx^&NbX
zbZJ{tVxeV?V7~L);d;B)eSJ(5TJM(V!FD>h=(>7c5j#4+DkJCm$@ceCLX)OVsik=Z
z@vJ;N_(k~8FQ3(5QMNL2YtwuI(rWmL6CPDgR|p70j5X<kNh)PR=ak8ZW^q?>Cbl)A
zhSaP<7L7QZrlaiRDZVU9ykP=5DZpwpblvw@T6o)c3Iz5+`PcTYaA8UXO*OQ@p2k^A
zJMPDK`7L1&#%hr(W+sLo3@}eKl35vmz+TY7YgfYt`C?8oFI{BUN#n4(4JxO)KeU`n
zw`Pg0X|a_{Sj0jZ4CuSL<m2M%)`avv_WIi%pW)W_F0NPHD37I_Sc>1qJU!_e*Xqz7
z$keEhW*G3A2iw^^-`=o2?MTbx<0qOAaXTWb(R?HCN3F(PY<H~CwdRb#069=}(iRr$
zjEK<F?Iw<b@tr23*U7$YtUtiNbY0xKTNtt~jfW=lt<=>?2rdMg*q#_miH^5Bw;5Kz
zQ}Vi=Rn<sWe?J8;QrkX>`K)l0?`F3ea;??~^*oF({d^AZejdtu`>UI3@bGsp^o##P
z<jRS%vho=Yf)9LElGsQJSi_Z~PY5RZSb$s96evQ(Yp$-WT%l~p8|!i%i&k808MMRQ
zY-gbdyEk6JjrAMGmgXkBtseAgGhYPq{i<u4+WB#HvJ^ZULYMJ2{`T+|h4A+B_8RcL
z-6?`co*PjEo!^ABSf|{rB;&&UZER&l>}*BRhLW$}X$auTmnaR61Tv*Um_-UilAPab
z4Nx@nu8aA^32>ga%6TAqOd$V<QMi1f*!GJTQm{i|zWho<jrt$X08VXSdxApw_?TN~
zt>-X*L8T7yDkVwnR-^s(x(tzrURnx#uMeVIk5Rr^gY|t{uzmh>Q5@o6x!YDfdog>&
zP(xKEd!cCv1A=gDer38C%rzFqO@UQ_JFJP<tA)4cw?PUjIzx4c6I<~=ceD9Li3$?4
zd>Z2Jr_>=VF^ytX;n+<aF{3S!|5TDXp^<0QRz<=<Vy)bYZR`3AH+FH^8y2ldhDaF}
z7xp{<E;g%xC1DA(kd$Y5l3VBz5RKD;eYz67__-=;tCf7|rcdLGq_#^zOfMkk={El^
zm`Ak#J(gn>>rb04=Qc;5wsYrBh>bZol3jD40zrU9A-OoJvak?o<%BsblIAL{V?KNq
zNb6Seo%eCGRm$>CtVm$JS!}rO;?>i_hr*RQf_&VYHcjA`QOJA4E1<jKPW>!+9b5HK
z;x=Jj44l%$V8od+uHRZk0%0w#suQ}AX_ykmxpgu|Q0YAXdo^7(-dArhQZnE&jt)k@
z3`$n6sM{#Z-u{Ibnk-EWCIz$HegyAQ_<V1>dc7Tb`}=ktuzTSjkh}9HMU2QKP@6n<
zve#EOO(OYrJvJS1pK;?X)78nd%95x_u$UcETa%I(@Y<Iq^*V5~hr(<>G)rdZ>dd)I
zl+{j(Ci`jYrrmioWu@uzavt+$UU?XgxZYg3;u|-fE5JIRRY`d_5$gQjLx;u;n1+PL
z0#t_1CX=pPaqv~qGXv6b`7xo>a*!Tl*Z$$y!g98y$1!g|nGb^{JBwPOz@K$>5rOWj
zV*OZ@q<p=21JnKIFb4TK`B75TYb6F!tjNg?n;}IevDL;xdLD6^N*YovVA()zNM%wK
zBnsl9fS-x^nJ-byRH-S^h0!VsVK~QtAZ<J)EsZ1<&gB2_@a5{^7V{mb<Lcx@*MBe1
z@$7y!o}Zg4dUp2c`Qnk!4a`pBNM&iwK+KOm?9{Wed9GR$WGTB=S8)9FPNP^yIs$tN
z#U^ardk?~8I@x;94j{-RDEX9X*1M+lvEG`<LvVW|a>zWLP=pN*Io#)7N@rBiE@bT0
zwq?cf^?n)9_|lcHWE;E26F%KGX9rt0X32p3+fIX2n7VPsemHqK6$7K#Ek7`z>k~jj
z7RCCJ8H~=&pipJ=iqJ$%^V$<s7+^7aQIrM92@i$vMKs(?yU?)#6su!yATrsDVsUmB
zD^%BuqZONYSt7OUDP}AYIL&?*K5jKj!5goS-)M^~i}6EK_-5J?7_`DLIh994S58v0
zKmA+wjTWv{=g<(I?REN~R(gOEMR}V!|FF+zXS4J`a=Wv1a=Q4By(Yp88rO?EIi@tw
zwkn~;G!6cce(YF9c#aM>j~prkc{%R33bvvTcVd#B-cpJiv?s?T*xeQ4{x|~Y1K_Yu
zTmisj$V^d3DJ4cdp~Ek1XLk<i0?Hi8`h|dx1*xqh3L$hC$2p$UA9p%icmRWk=2(R6
z{v>Jg9wzBKbhg6gIje-fdB+{w6?(-WiVJw#ZN*i|$>mLuCFN1%_s*X<moS&oNbKRn
zx}iL^_@ONp%m|gnjNGEUD3Zp}p5d!G$$W{JbdlwF#2!pcgAGZlco?AIEp-xDdXn8{
z9IA6DD}n-ae5^JEMzLnK(<33N(Me>_g~`RGYg8j-&B4yf&PpT{!p!@F!8vN0i3P;7
zsg9t!@3m)4bvBpCmyEePai};R`H45~PRJQwpQ}n`uUfoSY8_Bbi^D2g9hyv7t?LYJ
zbs&H0n)exQr{z7&rLlfp_DF;{M+HeT&575B<J7ZIg4$u*3svmkR?bkj>SBYL;@xab
zaK2Z~AQ7C0Br#Pu4}uaS@i8&cxwKeSo#iy<TWi79^Xs(MbAUFFqfzD9Kj-XZ_x1KO
zffgYs1a3vITMY-|evi5?)AfIRz?6Es<ZIKp-N|o435UTp`vI_T%xj_;y9~d4yUrgT
zWeqXY@*Fe?hXE8I#N;B6RpdO>XF#wy;$PY30Ro%%YRYpVUOIV4G>{^6%9*TSwoJ@I
zbprYywaJI6unVT<^|m|I6Jh4|!Zggx4-kkd!5S?s14ll$;{>Hkf7omXtaRCNAel$}
z5joN)YTQ5-finUVyd5e;UN;-NY^*@R5oFUR@p}67^d52OoeL<5Mug{&aXL?Xodc?w
z+Vc9}=Gi+LA&b_3VCMd>fcAXo_+`e*xP}2IS2}5j-SJEB7P+~Yp^!W_(j|gf#jk}>
zNXi~vJW>Pp$8CRDG+cU~10#<pVyaH;QgFGY7Z8?(1o;;4f)c!lDi?XM;^RA=JX(>T
z`Azzp_t`~v-J#~>4mT}^s8D}wEH&kd7=w&+Y3j0)&)NMUy8x}^{&%7~J+l&{Y@O7H
zCK&;oK=YxKR36?Vl1GVI!N`7deuMWew?!ANLSdBgBoOd&`(|l0-GV{6SMR`YpL|=r
zzV;a>3b{li@8oN7BRrn4;jKUb10IGFV@A@FaIH-SYU(?N0kIQ=6K<+FL|A;p*T#-{
z2^_Xz?!p<?e*x2ohcIlAapeyh(qua7ujy&YC67!>o_JGt{?pAF_O5X7aiJgrDG!O_
zotJT7oFYr8jbViwGT$TG0HS2q#sf~EJbn-ebTE~OnP()W4Hp%=!fwHC3B=3Wm>^yM
zeHUDXuHY!3T<!aCP1v;FDn)f~+1OlX7c-0ciG48%@yf!5*mhkkBYgfr<)aKXz5?);
zJB=B|vDb%&98Am?aEb{nU^Lm~m1>}6EY6-~NVd@4yXSVEG;q*0GO}lf2Qgv~xSqC|
zTafbNq6w|;WFTs>FOe}<s<>X*FCV7oAh#g>iO{S^Ko>eZ!qQksE&}rvv1dRCw*#>{
zgeSAjMKEZxo6Hwg%%Hr5vC2er;I9FbnOZc>)Yzu{Y?edVtBTJQS5kr&=WQb3uhkJx
z@-Re0C7`x5k$*&xkl<H|mZCtk?=`yy`vqCdB{TXv%QzDr#eOvSqTiE?FMJh(8c1j&
zjlra3$b9?pNYt@W`B&5BQx?2$Ep09osFT^pvMeJ0wsBW|95eQ26XyJ{UBL5i7;mL;
z4`M`kbs`B+*Q`1t`5n3?F4%U3as@{|`lkJI?_m0uzxS4Vht?IsuLcSCQwQ=B9?MC<
zhihp?QojQTrc{9UUORiOX_cuwvVLBzMvr$U0Z^fd-*KHPW*z6#_-xGgd^C|kMJt>&
z6f1t$p01Ph-gcw?!n57o2;9q%Y6XmaGU5gnNrE;VXZ0$0-=6axfLzUmy}o#M)RX<2
z`@na;uzZM8=Dwl;pnu0AO`5w_8k8IzDRbf3CS+k7JlEp_N>i>34V+v1-7>XPx3xd5
zzyvg!D-w!b6jnT#CG)T;UW`xVA;vRnt1?JUNrPK(rsY9@=C!ix%EN+J|B%PM{7m0)
zA_IENS?^L<qB2%NM--R^@-_Yo$r{|>VG->k<JVA*v=4R%WC{)~dv4&`_8ih1J7TpA
zIU#-IReW|~3Dqh3IviuN(}B%r<jWU{T|wTeIl`cuIZ1LP93QD~utWhLPP?z7z*?Cx
z;6{n4GhiJ@5b+`Uf>gytg=&y0&l$P*sSuKO7$5F|RIUIOcWMk?7aeA#s7{D74o~2i
zAI>k1BHuO86Q6%FmMYvKKABOJjsPM~)#hWUW}9G>$0p7cdgjHuRV#hc$)F6WEycX=
ztp~vW@KEn27Ye&0pw}K+WGQ%jfsc*f5K4v=xy2T#Sr;jkS?6pyX{>(5wuV5|;|UO%
z@XKYWgC((C;w;>$WKW~<7o@3UZ{DMUv+Gy&76ou1F1CPE*r`MvcKg7F1g7-@H%TSY
zB0Q$vAAx5eCYx7n2f7ucp#ysdbFIQ<CuJd#40!B=k6UC0Qv_KPRni4b3J7CCre<Td
zuO;z5LQ<izm&a4Kh0q4sC=LVkqk*EC`BDb0R7mamU(gOx|96bc)Ma{zFHTC58J{n}
z)Wq<o@9QNWygsprVOYQDF7S6mAy9pW8Ec|<n>32`A4;$1qyT*nsvGu})Kcp;cOkZn
z`VpUK%*u{d8E-+`ZuDW%jg~DU=Bj+i=~0}0jgli$1zHY_Ix&;o2k#tAU2%}ONN@Wb
z5k!D|9e0bR%@KuZK~-+-ld}Jz*x!aMV#;y`ClfZYe(CPIy1i9D%gId^M`1B@p3KiM
z;B0&e6s52$+h?w<XHLr;1y0H?2TD#?M$_xTFMjb$E&)&lXa6Q&6~C9=Wuj}KVagm`
zlr97W=4&VrC*`MipgI@cyPUm9$VCR|@;G*US{xT;>+mF9B`b8aMcr6G{nhQj?_v$!
zH7j6e0}|^b^F;srvhi?xf@5CQ<OpSb@`FXkx|XMn*>!5!m?teCzdZCS+~n1xOz30S
z0XXYgZ;P&OBt&CWs$0jefWJTYW5re%<XWC=dUw{Rj<r`g#jhxfQMLNm)<B?~Z1IPk
zTh%TVm@0;PsX!WQUOdcm-*18<o5KCci6opin5@X;!LSueTM9%1q&t3byjJKaFo!H?
zxc9X=3DlSRp0o;ARP4;}yqS{1$I-WqVR4)nH)7c_Zua-DtMf42qGDA7kU&D>`NI{H
zzxRvmB<87>6ZO+jGbI*E+|JL&TjI$u?h$BtnkRHEWLD<5N5CH?gR4YrnVZJAK_7D*
z!VKi8YS<g<>V4|i+Q6YY!<~qBtDs22y@BdP4)m55&feUGCDb(#-XvvUbMnnF7Q|6C
zenSHn89VxZunA+dXfx#pQoOgq3Q#Q<$rl-LN?w!@MdvWaPQxF;RjfuP(X%e)s$E0e
zj-8`LtIjrt^0~1TC*pk!BOi^Cd_2}C5*myQ_DSafaEtbnRp3Itll;099x=o!4n_l}
z8LNNH&9+uXudNV@=w<^a5TIJls*ZYnFuJgc5niAb9T?wJo=WkP@)XBu9`2nL6@2^E
zA%0fTAyJW4rB%xmqi%H_8*{i!0n%Zy0zw@xKr4UOxUU}8`}f?p_U0|ZNaLPz3`0(q
zcc|1FX-^G*37x*z8a8<^LabbIjZWVo&y)eNZLf54uw9nl#`@K;Phw>M(;$^(6*=oL
zsZE1_pm144K<ct!E;fi^Mnj9XMEcy+-P^tnggy(jQ65<^OBYKk845HHoyE~tr>Lfv
zM~zhCue(|k{?0<JK2}@9CiV8rw=^zrdjw^;9=*{pkd)k!r?SHu$j(k#*&Flj%r9(Q
zwCJ7L?$K5Ns+|M76x|1Pw0L{Nkco9wOrWf3<9&Penn=H>BYM&m-e=+=%mTJ|br=(g
z52MxdnfCa_Ox~FNbsQ7QPDg3WWf_%-R=5^p`~y@u(j@b~9AcvsW;7W<9U{Hg;-A?V
z&tFe+W<@y!_=)*Rlt;B%x$~)`o9Y>2WbNh>We6>4&<l5yyXVIT+kSgv?#{DgSbcP6
zxyq00+*mu`a_RyT{pj@2|32g<>sJs68I;I{rFk-PW9D^V_`Vqq_WVmUQ=2x;SMOq)
zK8``0oYptad|z(JzXbSLY>>C4xfDeNSDcXF-NoFs#JFRnX7+w&FW-vpl*6T}YkCLr
z3F|u9ZE|kRN0|!ECph1$Rh$~SNwl1l$zI3x#(0r|V<Y|n9DkNWAp#ep|F^n`nFhxO
z9s2dngXEAo;h1o5TA&#PIR$E(0Sgexwr-Yn5t*O0c}vFPy8riGFNAB?w`7qG?re5V
zh5E<xJz<^#PSOFH-v{WFu;|IG<ry}il8sLNKf<9Idp&Nb`R>3kHp*F6HPBD-j$N_S
z^e9nQ)!ndYOrdRUv212~XsGLLtkX3<SWxixBo9wye_B+CE19A0H_ln(tQ=FJG7Iq}
zdV;F*!FRS^*x%hdEJcS+bf=QDU_rM1MI7(62i}2ch>e8DZ)PmJZ)#g6%CxOX!mi`R
zMnh=t7ixRYW6cZLBllU1eH~me^!bGq@c&r4k}Ocps<L|dcN3*_j1}hv72?h3yYkl&
z71Ur3Z%3e9hGNMVRP|L8Ie$$30jbnnlFx`p>~OjcBu7IRKXoIbyKpmzwhP$wk<LD?
zdx*r6m#9>ZypwgYMT!We>hBeA<WTh#BGKdQwc0PFSPmI0iokl531%fMh?(LM9kx7v
z*#sOg(l=}>FB63T1G8nv6SZHLHx<!-Ml&m=-h}FN0jBCsqA$ohH|qR1k?5S-x)LvC
zJ&Ld`8f3`nMQz_DFpG2GfEohZiwNLq@}x!cf5Lu*dDgTn%UjeXhtZP7fLJrkI-Q=b
z|1~Ad{YOSh@NsAA)wAE5zF?uf_Fo=aB41$g66s3gR9jXdhJbEF0-(F%NGUsHi_$8G
z@@OyWD+w>+{;8|8k=_fI7TqhBGc6Rk$}m*fx;mI<Z<55Fo|XG$L9aRfttKP1hbSP^
z2_q6t784tTFZ4kUc<KQ$85Qy05S84&Vn9`Pos(l3_*r6pq+hO}mlUs@GY!to>o?@a
zN69Y}7G@E}d}-K}xrVOhc@vQYSx1$8W7!PyKIYS6$6PDKlVJJd=|Q1XZte_CfY1`J
zW;fd%-~HpTmRs%(GBGN2AQ%IzJ5?$&8F!q62jbtcudYZzw7HBQmolgLY!}T!6cseQ
zodrcs6`6rp^C6LQis3M(y{M9Wx;b`vzrv7pIkdzP7qwTrgQqigl$r?hSo+qcC%=Qj
zv00R1p4RTQe$p8^zIdH~;J)Q4$4>12W@*DSAmE!wzbx%qY~VFs!Lgm=H4m~x!Pjas
zawqjrKZo7{F!Lqlduj$s0}OVN5N#1qfD46Kkq62Y56TOmoQa{=8Hx}_j#wugO_)()
zq*>_F(P@APId-})d2m)!<fY11a5ET}RVKAT^&A2Xc@MknwZBeqbw9Hun%}<QXL~u!
z#Fm@;IeVC27d4#u5XsqD1gwx&)2k>BSS{Y4)Lq>Guo5;;b*3AFn>_{!R(hSJY9-4R
zT8%!`ERA`yGys&Y@yS(M6z_8QMbEll6m3%4(O?vwKJ;bsQrU#pjr)30CllEE@aWU3
zapVi$zZpUB-Joz7rD&KJOic97XXj#Zc^$`ASU`6b<ch7KuANK14+!27{(8j#0u_+{
zM--HWq{rk+eECdfhlS##m#VVWKG4fp8PCfp%L#06apzY2UI;<0nBdgE{%hnYD#8>7
zc5Ke&CTAcLVEXQ8$cAf$R4VPN0UlK-LemR@6zZuJYc@ZnXH_OZGRvcabGf}K9%0#q
z<G%|g^1dYcT56h?r9jS``n}Q23|W8;L|f&u9Sq)rjj9=0%mnFYzg^{N*T{X|#=3+~
zS9(`&6{!)WOg?8=+s8}_lz5s0!?{02s5wV#{>(xXs5aotYc(*!*J^eiR>Kbt7fD-T
z`<d5GQ|an*JG-p5?2Y?=v4T!Y1oA$8)mAy#fC)l({G`r@$<Y(ZIH|&bQ?pyZ)!z9$
z08f)jTp~e{6>~p~fps@*oJH@*jMlv_!r5Bjm|STXi&n#mQq53Tg*&Q^UGa{fE)|X5
zdU0Blmso0bI->u)@_5rqoEv-BDd4e_g`T-77r)BX;q!1d7Uao+^GI%hpuT1X*~cOu
z={>!C%!qnU+I4Mg{(_cQc^}8J<i11#6&%z!+}^}leTo@BSMIEGH8e)ED7GXF0Sjg6
zrsed|-#X7bu@xw4_f-T4lS#iPeG((5_Mm7Z>?l&q5JpHoVTH2Go8}hyECbnDSUetW
z_W8IzNil)y5eXSHob=K-T~k*tRhp|3p*j9Pd=xNZozJnD-`&A_2@~tPX*o&ms0zEC
zU`0``yU6^};$GVjc15|YmNT&$moJ%LMh>qe())C>l4mwNUuwP@Ls`}Lom|iH@^9>S
zlku=WHarpeL})_#ybo4dJg9sQUb^Pv!h9mW$%ZWCrm)aO3)U8C)$COu)|+VMw5U6T
zalGI8j-kW^)Pu+>W87<*ItR@BLGk``r=G6JO<)JOv!gG=bc5olF~ljP{Yb>5!M@5V
zk~(}E(jt-EQ>{^Pjo2|xkX1$wo<S)_o}B_9{x^2>7zzAo$=T>cBg%GOuRbDn-^wG}
zDYmj$q<zML0nmn2V4?FY3f>TkaLp&+H-g|ff+(FN3;*c3z?<L_l%FI=%b{>@1r-x9
z(G~k&-G20uw7HspkFoX93kn%cT8-B(PAwz<gG(fDJe@|%X;u(*#%t#sS?A>Z+7vvb
zCo*9o@ehFeKAmhXdzXKcSWKu)#A^rWV~p>~rpnvIoLDyE7uF~<Cz*BO6?{Td5?m6F
z4<nCrucflGHiS^%5pYFVN1bH^pH0v^HOe@kR;p`g!?k!7#zF+_E^|<rVGPavp2y@t
z&QM+DNp|Z9#;8N_V!1Unf<%<#BQj(cTP(8csfd1>w8K|0x~XyQNIz+~hNpq-b>`F*
zizbgm!N`MTP9Z2kTqP;`UWOyXt>U<`?L7#eAjeq29P}^eeYsFP3p83SOPN8Zwas?L
zkW_F2#}l3V(#RlnpJ^wX<($Uu=UXfa7$|}<t<}k5xPA)N<h@e4z~JO}C*&J{j1h5=
zz+UaF+tO+TgQA&w%KPJRHU&anvQ{SwR5Mow3snXKnZO2m0@-ypXj|9wMA~#fevQBw
zMuTHeV~%@Bla2zdNCgq6+t&QekUsyyojvV4wa9FBBEI;>Z?#O`Oti6>fy=h%#K$^*
zY)a1jGLy1DL)}QP-`9IQ*5YCQ2*XqnUtDcZl>a;2Yp>-ykDKoEl>=y?nskIVBGKh#
z_>ZM*?Xb0Ws&Z4rl^D##Oe08!fStMpfVODp7KFH+GhTT3l-(nZ?@B$_!4z)2Ni(R+
z_zFLv5iNs7-x65P2}>2_)s<;%mn+BOz0#G5dT0l}-$|?2OQ$~);o*W5m&afyt*j7b
zwE6@qA-i-B*ALX4Lb1b$<g|yMYKTTniTP29!-RCczDZ(wU3uz;%Vb7;J~KhS^|Hj2
zT!QL3YoQwkPaB%6KFSd$%EAz7zl(g=w|MpOP{m3J0b80(ypmd7eyIVp8vKmMh=T=C
zwP$l%aW3dXZDicTO@4KD)2AtemTwiM$mq5v+2_Y|oLlkvdn+?+I0}B4$ci4#|5*R+
zvzlEOp4qzat<+Y!hK02n8jFfbjaPX69*86B1T4UUA6L^l$KOzWG$k-Vt%+rVoq2dW
z{OS(a>ylUo(PwDGW72Mjt4&VeCseCbL>u4cP#dXmOOrMCa^thrWK=%nRG{(_&(4D5
z*5!0~A9N=}C<(TNQNSrB%vMKqkAT&dV0%Ho%HFKz9b+GH&tGD2eLN2K@AN-YN(|&K
zxNdjz?>gy`gfc*MK8z`{8%4E8pyfmzf^Ay7<Xp|9+VQ=V5*>zcH%rfcKklJkTeL!&
zEZ-x*OGUF$DljCOZZk3s9>W#zOqj@5bn?#+g-#tDjzML(OnYRGU8KCCg#|=Zg8L5)
z>2CVEk9v4<lRr`kwa_9wp{zg)YmQx=P1=M}W39l?1g^7XWUIB`3^|0KWDdHhpC4B!
zf%H_hPwnzkeUQ`#|7i|L_Wb7)tW0FpXbidNsfe>sNw2+PD=H(z(Rher5<(yV>3D3n
zE_e#<`I-n{ZNYesoP`)e`6MVtdmF`A`vgZ4e_N951kLi*QbPV?C*RPpB6~Kn#`51(
zedhW4(PN{@AZi7ohF+N1{q4zTqA@pYwUL_?psq3$km>m^Bm#!IPQ+ssr%9(ndFaCF
zZ6egZqp4IFION?w-0;EMosZ8NhaIlnaT$0QL}vZ+278X_R6f-#kr;J9%d#$g>+THQ
zsi4+dar*vsbP|?C*&|Q5AB?@gs*py?sgUO=sbM%?v7Lr~rYPB4jK1F;_f*mF+obC(
zv}?eF%+Qkp)lM&8Hgh&n3D~ISd^T&gQ%;`AH(swfMYq<&&|`P?aAQMTA<twL))aR0
zLH!-Eb=0(`#t#Zf@>V=`c)V9nSf^%;bS5J~Z-sB(X##&1?>QKRH(ep1nIpL5wG5s?
z>qPyN!k>+4f}U$M<Ml?-Zy0Q!)t;*zPV72LbQRC+Mf*i6!A1d+4a`Kyh;`vuti@|I
zZ}8mEK^emkK>Y(lS$A}x;JX`QkCG6w;1)KI9ja~Xj+^I1>=e4SgLN7lWE@=?p;$W?
zO4rp;IA`-&bxmt*-CuP-gL87>Ck}1?x;&dnLL5!!p0>w6n1l7<Con@+KaNKKb7A*&
zD%_)sAWL(Ky2yJ2-=%hGa*E39>;#JU;%Oxz)kBDXHikC=JHc>~Iv%&wrkSjnJI#r3
zOSuC{fT7l02%$mR_Vkf<m`OO#A%!<CiHw?M?5P7kNNxEPxA}zfO0lxf`E&WdQaj!E
z{N%Cb&N>HjCh;dv@#Q#rL2?sk<EiEiuhoDe%BS6Nl-8Xj9~aEHYY3mc=*Q9^@Btvq
z0b4oNHViD>RCKXS{^X<y`8t^)5MBeDA8;BlZ>J76_#o+ra3Kp`+7!3(t$Fcu>=v3M
z%KC`AZW4`esdo2wN4I%+GdN1QT!-_9uLW-7nff28`{svL+j0+Nl_XqhY-({aBI2C&
z<u;M1%e}Pjo!dvkicevbgk51LPI<Kl#*K=MIxi=}g)671>ptSJxTrQ<dMFKZ8<CC|
z#t2<8y)&Cma-wzqmDPv8`;CiAC3KM$rg1o{0_8!QB|FBWWXL_v=r>Ne2rWFwrCO8W
zQ@xV_-dwb?bD6UFxJ~z@ea(vT#-8tV4?A|*rXPKW=(9zBBeA@13GZhRM4F6BzM7G0
zQ^b@j7$-^%jTl|kBz^BCC*95DE!03WRQuZvDKt&RHrphLy&SDyQfF<SFjJuR{aUot
z+uk%<-^SZop576r)dvI--Sw6yqU|9Ru`q>qbHBp-vCa@LdLaZYD=FPim#C7RzUIsM
zEq*nf(!#dp2}s!X<-5Y=^3bt=gxKs&3<~ezaBK{sO^i71&wQFI_MRBOjUdGleD9LP
zayRt5gaJQ{?nimfbST;BA-hr@+8^If7r%T3zfEZ~bE(HLFp;U-I7WwMIQu{=S}-e>
zH|`vHyj=f?grE6>GZM{PGkl+&ORMdu=m=^|!a~<+_gXdZLy9!U4P1W=(E2at&f}~O
z(IEcrTKRb(Dj~q)e0k(I^BKfHNDBtppznI1!A{X(g1P+#8X?B+eAGYt(d$OolNFpw
zwM|u5>cR<`mSo#H>kYn0@|v0Ev)qo3dz%FOYDiHO)kjkT<rkeey7TxhEbpv=MK$YK
zR&yQ7Ks*kx59)$FeUJb_I9&mu*ez<@FunUiBK^WBN)fpdhDj-uUt(ZA07I$h8<vwL
zB*#)UllWzWmHIhnP#R9@+>=&~>`=!}98Kz;II2>5C!3^+*&!#TFZ~QcL%G^@U4$$P
zI8i$Rq7GHfwA+!ST6bJ`bTT(#`A}#jghWQkvt`rc9eiqQmewprZr=D)vMM~ZHdz2c
z%wYRmFQ~!ZPa5NiP`D`c(ajJuHARPbz;f}t-^=!HhS4)?4P`;U+$IK*vw!E6xcJKd
zZRnqR#(RadQ=xtH@OXvFH&>oS&!yxJ@n8RE!@cfOBNg=qT%MoUfw6U<a2-|wChI8!
zbGnU1vVY7P4fTw9`s86?x@o5-JVuaC%Z6Oc;fqkBs7<xgeesCVd4(-8Hk!jmL5t@o
zq^2s3#JSD+kKxd=l|+86dkD3MYq>&CiXR`<U9dZrH#t5|3LTguk*+^8-O*Ydi?wBi
zxM`{bBhC)@`)?<4AKRAL7?l2admS~R|7N~gx-x_5ZCOf2*dmtXNevIFkX6nI!}2uk
zmGPP9{sek}Ol#%~BbXq%sgQ@u2SOX@!?pL8cjP8*mZW#a#<04Scs`i%zni+O_6gIm
zQ|>;YY?T@OGUmBS%}K(WOqP^2PjUAlG|Xc5za7{xd7=F+sk;YBd1JM~OCn0^ubrzS
zZH|pOMRpTZ0-oww;g699G17To<6Pu1nsojJoez`OPV#hXuFCnZ1($hit6dJy8gE-~
zYEonYkMjY8X&SvA4o|&K6sy!8U7dbAbq?H^(WkG)QdPLWggEuv*xB;zg{()9B@*Sr
zL)B$8YTV?vZd6I!Kur^LgLSasCun>V8ex(ZdGdK+QILfa6!5Zn;Yr{5OVeVOkt!w2
zCxD9g>+SCM!6F;E7cn=lXR$2tJ7~}Eikyr=rQ<I$nm}wdMWYX|=T{ns^)r8ZXM+>7
z?B{Bk_tBPF$--aphQp;XP^I71&8L<Z7&5aefn78(M0b9(NDZ!B&5#$|IA!}E$$p!A
z3|W-ZWmp^CF84x;gBD-54X%eUGB6c`Y9cD=i~vU$I8P7H#Eq!bQiKobqKU~@>~qtx
zT~dRIWm75;SuR~usBFYh!epU#e_fTX=gYvY=YgNpOtNN^#g4fIM&c|;lLRRzX<q(?
z0q?IxptIx?n36$l7wbAd@v{xZ(^TErE=0JEzM-NFPneL?@6^LxUzVwzdk9URUQNc7
zHPF5oH*Gv%_Uoo!faCaB`t!w$|K_7K-f3KL|2MM&=Bk#^Y!@_2gCJCd#orvivD}24
zgV!ZCzkrXcubQbnNvI@e5}CmO(nox*x;9!CQFSfal;!3y57iJmK2ACVWko{QzN^m=
zSwrY$A4DZYtaS%tSh&!0#3gtqdd*cyy&}qLsN(|O^48=1s{sm=5+tbI4>1RpOw@<e
z>vKAO{>dQl=OTi6UJx*BCBYv{-|p>3Tc)a(LaW{qd@}19(0%pl&#gbe_~kQGN=T~g
zU2W5UiC)!9OvP>#Vk^xTo*T7`P@UG~KK&2p1MAV9p-uK!xHq=n<%Lic$!G*l{{kpp
z(r;_?QN6RQyl|`)^^3?iBe23va!x#ro3EAysqS(Z5{fNg=Qk@;EiE}+p!X>*7;z^d
zN9}H>abl2w9aeflX8cq)vqnkWV&>vM7rjrc_(sHGiiH42+N6CII&Gp|0R1Y4x$RDb
zk|D0OaRy$%l$PzSBweWv0|9G;8mDlduCkWVe4kyXV<*Qh>pZ16t`wr`1x~uEr9^VK
z$Yc#Ex8k9#xNU`<6eY@5Bz6BtVf+p^I$SI7L&@O>o+2m5v9J(NT~)$G_Cf>g%;^_I
zrxCNg+edBYbc<(eg4{8?H0es3R^5O<Z?aV^xetF2q?|gs=vKt)-PZLn*S+1dC2>0{
zwlS`Mo;*J+z4cJPp7X&UQEkzn?n8AA66+$j%*vOfzAno942|ra9>;iHzkIj7eIlO(
z<VKG>CWUcUY8yh`nVUcUC*4CgP<C;6m*Ha4D$TKnSg59Duuq}c>cxC8PEep;1QMzS
z_qy~P`#M-i?D=2RB_c(QF8IhZj;7CcVxI}lzq6xaq!#=wfNJgP<5j|mjQN{Sp+B{X
zVG*e9Vm^sqoN9p<6D1Snw!E{yxzni5`kQ{1d-^Q&voNi?X&Cd7#aEZraj(awb<%!7
zQC;$okx*))h$0T+O!SjkNI8$OM?Xl9B8kLQp8oehEk3f>Bq@YKS1@{3%-ZSOUu{z=
z6PpU8P)iP83*qgx(?TWnU2kBD05dksTtWVT>iQ~Z#G<D9%3j{!HbNkDNNhDWY#xC_
z1HA=lTz*v<VZN<&TS8o0D<hfxPi~>UkcAjk@s3O#`PW$;91Yh<MRsW1(9%@#e}T|4
zkJ73Yku_bg93Dq@=-h_;KgwkDWJ#z>WaX*X-=Sd6Avl?DcyakNs4PxmNtqm_A$YX~
z6$omHMtholIO)2&>1v?NG1uzsU>=^WV^1IQ4a}zR<GEZxjW0#`G=o$URy!!~8Q}2#
zKUa_h-|r%{@C4P-$fi$=`el&z@GmK&-C$R4R^E>G`{&_9hCW5-UjaKWD?i1(*~2uJ
zNJb({J~>xxvgFUK%k-0QpbU4xK$!SoDMlfoeMJ%E1f-Ux4G;gPlc|-JA%A(dtU>V2
zrasqhRkF_g^<Ta`sQ^Zx*^2w^aPWGx`+x*qoNK*6S_Ti^Xh^sJOU!Url8$gJ>d}xJ
zw!auj&K+Gl6o~KS-8=Y$6&A)PqE3!``x<M*Btij$ITW-A$I@$tf+?F#jydEtSuDoD
z_`Cc<3DOo+!|E+V0MsQo8*p#aTR`W&U+yFtwLvcp3$0LOt`CxVTQ850_&{hC$fMmQ
z`-A-o`9#}`(rVG<qED<-C>f8uT^_ZNUgG1TXy<N|Vyn{V7|lSoptQKjS_B0>c6lpl
zaz=k%kaYyMduawf6uutafBSnDqgdHlE-~ERWI%DzLTfYC=~7-5Sxr(XKa31ZtlA&p
z@cNnDtu0|ZWyf=(WvvWD%s(fmTBH01+>f_@wZs*$d2=9@NHKI_tY;b=U8WQh$#Abt
zvnPr7KPByq75KpjoEk=_*Q=XnA|}}VHrn|Eqxjp&+xa;4{qrN?ga`r%R|iY>3x4I1
zC2Uw*i#ptiHG$oukn?4^_i4VwgnouH4X-*yr<4kK`+enP2!c^IZTNANpl*kgjJzv6
zUh7~w-?^Q3A4q*CXltdN#v=wYFRn74?#WE<Kl|bwE3k=3Z4&t@COCj4v`Kx>N4eqt
z+`{xH_3UAKUat|AQsCIUVtUdLE#J~!MU1dQ<^C)nbFeCiDF^A=-|;w7!s8d3T#@kL
z&Sr|$P7J5eDoZqeQ-a#(9CY(7`&t4hex!_6ktIzoos2+~<v?KBKfr;Dol;8Vypata
zN3Pa2AY99F3UO<=LxrY{I|I4cD%jauNNS!GH>mENLiXR~F7~2L6DYF?A9T@c6qHv^
znJOzC73`VRbbC#Ma6>o}c|2laAnoHXi6Mn<bXf(TqkQw`MW~)i4ndmus@HOeC65;F
zS8o$H9Z`XjoBmRkPn#E($9D0lRGI^u=)qw(^s`~}j7Ub#r5{Qbf-37N#nz&WhI<{M
zAB?y#RVMjuu0L2xGn?%oAFZ{qlW+>)B!tjG!795bSlPp~Oao#<{HE>`@1?%0pWXjO
z9-^OB4id`Que6#xjP^%5@ITYRpX0?7tv*Q?gWrTUaV8UNF}9iNXBou1QN`tC^S&Gw
zOxbbd|7ZxZr^kux`EAL!Wfwppeu~<OU5UiNjHVJ5^>&q}%b7<0Q|d*&8gr!!Hr#u@
zc@i@6->v<jD_1$I9afj0PygBHb8`$8ElwMSVs3VvF&F>9R#hA}-zRhYOC9BmZJ@%X
z@k;kLyCs3a5D|*(m=j<Dh;z9eYSx+mmy##$*x^5>-4YhdtVPk`+{0lEINIUFH0L*5
z6`)gm-tRNft2en^%AELCZ9GLjDwy%&h(iB^W^}S#BQ=^l&S%@XQGqZaNvjaU(QCoH
z(}<+@w>Vt&?24*bZ2pApnN#g=BgOpnMZoC{`p0a;f_cg<s>~Qk*+wB=cB%|y{4I(#
zfPFCqXwF_+YG)Su7*gUAMfEiruemQoqb&T1exVqd65q6czWV&SXgc+XXsgq}{-1P1
zyV&;>+X|*g9Ps?Sjes<`y(sHwGX)=Kdc6<t2F>@q;456cbj>zn2i7*l;^*9ZY<Fy~
zZD|DDfnF!OUqPCCbrN*O$4+_R5?@zE?Kk{9IeZSw8cWuECq3>ChdWaE_j#WBNA_jL
z|B?|WS(_kfV^wZ+9Y}d!cligslgdtCDrhk(tz^4Sr@!<N(SZDwEjc~=?(t2?D@wdj
z%#a`P#x9bQIH7trTkV_+6{KjHv>~~$n&0A5)S}@K?BT(}qGUpyi-VV7v^NpE;4(sX
z2s11I@2hl*Qhu^v^oymm{6#9!KOH4j-&3A?`R;&LsYD$;HyO_g>#^G?WS47329obV
zM-!-c?;4Yx_xpfM&8)Y`WxyC>NGaAQcm#u?5ecPIoCQG4pdV88+rAI^QamuDOhhl*
z9pR(!T1~o8O(iito{BHBd{DRgQz*dmi!k|m>b12<pLskgyFuy*fkmxg8<osLd*Ly<
zjZu?9h1U4lGb~``=E*YXDb}U7lUn@rv*Qo?0Uj+cMuzPKQo9*t5K(HPML2aotQ|S7
zf~@VhCTFGna+vGqZu0%f#ZVVDe*KL5qRrlI{lCJ{D4Pl^&kaY}=DNB4lZh-xaJaZM
zu`d08wJ#=DdxWwlHJ2fOU7V|Qjw_~Oo{!8WTC<0Hmnpfp!|0Fr;%Ep~yQs0Yv?;x#
z)9cF!)vtLDK|XUp!NP_n$y!N9`O~exk~segMBEmWk!hLb-~^pk-*zXj&WpX9l<^ZF
zUixoriCCqu?_hI9mZ&{1;mCQr^Ja<KCra&-n$VTRHuj&%6knCVTaFQ@bgb(11P!Y&
zdIIcTy)o@_-jow_5c~PnJ55?H9tkHXNeJ>6U54s+zE5z611G}S2|`l*7d&-f^6sU)
zhOcZX*v*{-zj;nS3nvg4`6+{eR7NAgV}|<{CWpkS+iKb4UBYT+EjSgk+6ije>AgrJ
zAVf4P#0W}gRGg^9NOwF&<uPEu_>dPaM_~*by1@CDWAa;lfKY4s?#cfb?P)s&1Nevm
z77t<WKME}gCyKISh(cYa6lk#gE+e5uJ^Yw@&0@V~)4YRdl5^j9$l`{54xyz$7L4rq
zFcO79SNq8zRpt_BL*0e!mzSL8ll$z(1G<L8blD-VnnUsPe;ZHt^E0ZoVcUjW-C4Cu
zz|#jlRdbobE0^qbn+7zmhs(#Yjkk+4O|{iH>=nU(pz$nDT2?-H)WL6JWQBpoN}rsb
zL@Ea-jwKgU{}HjE&SJsCx6|yy`^+g5^Vt)9T(#th)o}j*PQj>xNjQEG_5+}HKT&AD
zMA9)%%=WQwv&M<Easyj*)7l;4Ey`7XlyA5=%oj+P;5O&zeN-uGyj#4CMzwzO{G`n+
zo`jh<xcdBie8O2|(B^>Fbx1RfLLn*&;yghBC~~=u{F9J(U8O1!c8{hTHg+XB{aF;h
z1YfIz^FIBI2=RoHiL4}_M$B%*Tvsv&A8*40yIMKGeK!Z@V>}!dquwX`>??i<Dyd=v
zf3-7>0arLHl0NC&xt!iflyb)o>8l`d;Hx?18a_o*JVY*6Oz9nm-2pI4(H6;uo3}W3
z;R3ev`Y;XJ+qy%^AwD(UubnS{xkalBXs+`9w@W$r+^<q6ki^R~>p;-6X!wj3lOV2r
z+5*}TSL7PL3l3n7-xmDmnJM@hYPSS|ru9jZoqe`V?Hqg#ghNJ`QdCPVbJ~`*(25E8
z-an`RwGhxsn|Sn&JY6R8kl6VFMmIo=wS^&VEXFFiWs<!YIyfA9@t@hisUj_%6?vNI
zGUC!rH3Kdw>Gh+LkK&JhJe4Ux!oUmaEl)(;JG7ffJ46h!0a>h&?+@8kuO?ppt)Ci@
zQb-MT>nyBrY@C`xB-!dZ6Tjv7nmQhI6Fj>^8p=ERl~A)JMzeAsX|H5z($4GC%Ylir
z_kD~^$S$gBS?1RuBb?r8K76h&>@py8iTH5GRQjwpVmI$&&g)<8p7uBZy~IG?^KLoj
zX$*t%EK#6jafRkv1UaR4GaH@)!q$I+d23<`yRCOaqHI<|MGn2+zf&y;;NMOYjXC6`
z2=aumR;c~sDj&6RFh&e_9QleYGqV0fo1DIv8ODy322cmXD6aAO=>MSU_Yo<R?#_)u
z>oT?C{#wI^C|mGHzO_qe3EzlU00ZVe6QY+*1@g?Up^%VMnjGRJPSMct$olLZ$Hm<J
z{6hK?5!5SJ!SU5}a=`##HY2e)YzE$Kmv4zazg-a<61p9hSmnjgzalh}BqyR)BrvPS
z{P{Bv%yPASdcCEEWQ$uRv=AcLL<D51dH)@kBsI~i)9|4m&ITrQcnYM7d3_zaeI@S~
zt6&yjQN8i`6!DFtY!P{KFH#V>YZ`QyrLH@-;4u=tatS8hC2*h7&KGGSFQV)XS3#aW
z_r)F4<|?2Yc+L{p_ALCY;Kr&NUBI&4i|P88?=zj`XQZQnKj-uQG`G($c)?***2_K5
z;_<`7e0T~?QnOU*gWd9G!UI|v-o&(c)7dO;UW)ozqz6ReZaFrlAN)T|W%3iU?^vE&
zj+mn0=Oj!JwDFtG4O<}3ge8?J-H1EVw=(s$5y-GSfDgR+ORCRmMZhkHcvZvvD<Tsk
zVRN>%cH}3o*!thM8K$9RfKG)58d!+&Opb3ZIh8^J{$$WN>eoNvbKbC~M<QL^95qJ_
zZ^m)dKa7NS6R=q7#rzcg6;d>nngz2zQQG2pGI}_6*8Do+%n`f!AAC-2i)Aejr`_)J
zI{uON=O-(l&s9LU081Rx5fO<8YTD$V@OH;eBa6_*5djykPUj9C`Mwy(se7C!O}%$d
zwIVxW1N!6o<tlsiMk7|VTE=L0HnSP{^DC-5-!NbKL}$mn3z9_#qx||NE{_3Ehwt0W
z6kPNQ76>KzFce+>nA^r7k_Ht|&B(lG7%cP%UnJ{?CE)Tk+x08bTTKw2or(|1!0EFQ
z3!qUkvcbdJB64(!tKG>^<?Rv%BQ61)Wym>M%EMBNNAv9H66SdQyD2j$v$SFzY>3Av
z@7HVW#CyRaP^zplQXhms|0`ChOJC@GgH$OYJuu(bR>>Qu@IkX3_06qMm&cglgDu@b
zs9Re>eOLqmta0dR9c_4lC>6VG5v`l0;qw%WGp(Xnj~qWMUl6f6uzDyyPoOXVXr<*4
ztrIl27eOmv&#jMjGef_5?BX*2!21U}lO5Xg>}41-&*f*{g_|e2$s>*E-BgAk{91K)
zQ5+C!kMASi@v>={VnTX10jNW%OJ5HMq2_jz8}sIV`b=+@g%hh}EoUlfX}<q!Fbr+}
zgM~WzXWMPEN4$n!f7dR*rw`kFH)YK1QxOC0Ml5F7NGK+427rHpRjI?3UiUXlKiB5S
zc6NBw{au#9`L2kWk|PV7O?9;Uxo1EeMci=ki&LJ{kP1JkhJsH-+>+bY%?;rfBgv!)
zqIsv|<rSgZpWRNj8wfE5x^reR=d|eZOg_@BZcs#?jL`a<4^NZgnx@<BycYpE+r1Fw
zU#JADTpz*}us)M#pL{*rs+t3Hgji5{i%HYD3-a)xGd3B2_*D<h*KA`}Qoyz3%Q7kr
z@blCRD1c@Nc07pZ<%pJXG)uAI<|_OKS@@K`pbtFByrZ0!&bu!xPR^runI-}~sZXRb
z6eh2^5N%z(6aD)K02uagRa!kL=IN&k(8Wp@0S00Xe=eIo4!*}%FJ={g7ToRtNBJ$3
z`SFuACz*2={HBc`T>KZ)5yfy4bF{#LwoDLVW#26IAFuM#&t9eKr(Yx&m0IvW_58iW
z*QyLZon+q<!Hdo(^n87(O&d!}Oy&<QUW;Fc`l>_a8LRWRY%k+V;V?4uh?#k-lxvoB
z|J1;@zwo<RBnRBpA~*~BSjpwyd`c$43s3y|_FbNA<TpF*NR|Muf9o_+gfV&vWFh7L
z=kV9Y+4JK`%y=Sz-$?T3uFd|{be^YQr~Cp(R1+GJXzAA`??Qr<`3&~v#Xhr`x+Hvr
zIdi2yEVh3qkhX=<Tr5#nX|s?F1Sq7C+|T{K6{r8FwoC}q7N{o;-ZWSNP&`pxloe@f
zZ{<53qsG~lo%sCn)V_J^b;&cjL4&Y}7$31G{O#sciJ;KlR*Jkk+4SdE-ExZ|J+Y}?
zT}*L5<_iM+@q~?J&$Y0&PpR2r9T9xA;3QbTL1L8ArU1num*7Xx^Y!u^0C9SgMv^th
zKx9VJZgu%`@elZfmF#<A%B^ClpSLB6o6=S=5LWPQWFZr58a5>hm~Nd~OC;V7t8u6<
z)>l|InL6LXn{-@u(P+l@yn$T6!g`D4k+zGRTE45U?}xj^FD#1~gRYJzoQO++#!z%R
zQfgXDO2J`^AN9$`60nA>XJ1Uaky){)?35WRER?k}g`cR|cSKZMCitiCZ^|fgu$`UH
zCe(L#Z4dY^zR`olQ9nv37}+PhNrqZMQ}p5NWL=nnnORED4|q%e^z6<hxUkM1v(9+P
z_RnrVV2S4D*;7uDK(@4oH{8+RUUrR&Xnq%_Z7$^=M(S|qL3Jm#0>jKAlU2`A_!JUd
zEDL2@TAKd#d`*}q<u?@O67V)8>`t}ezvKB7<q#*5CJpH1@$S2(Cy^|xAdJ6m1SuN0
zTQc)cid>L|$4<w%U)pXE!;-8oDhkq(3nZ<+DRH{GIOo!e&}kS3O(E8*k{ig2X}sM>
z3&p*oA1qrFJa9t^KH;0NZMVGMammq1fJf?!Nbam@M;m&D?fNok=Zc2SOAsFhNhB1o
zFr?;Rs_#y8fBBrPmLJyD=jQ*@5^MS!h1%>UII}Gp1GopU7CO_~4H{rjVrMoK`I^HQ
z?3#;%5}n~i^%R4G?AqRw$k@oftK$?yv{Dov+<PpSPWng{U>zX;E@rcsKZ{B$M=4&a
z;GZI@nooR+lFv*^v{H;_fyj%{ggi2g)k-)MDIUu38xW5$>KbUk#yRHC&(IVpS?bk{
zaNsTSmItaYn&{(1`TD|DP8xw<K>KyH#1YDyXY`XH{OQ72)qiqg`8Mazr}|-oV+d>7
zFM<_MQ<pY*aSke;__H&Q#hbCRJkI15*|rb=4y%d`I1r#-bHi51&(6p}Qmuw6dJ39=
zsFwMlqW=Y|Kvlod{l}+@@tQ>EJ}jaiWeyiYJf@ntGE7J)-Wz@x(jk*$6l_kIz|k6-
z!dIMf&=@Yx(GKp92k-YNnk>by$9>~|-;h47VKzly^W}VAGL@-g;3>^PI}@r*na3D>
zecbuF9k?p%ZH?r|DH2kF`f|msa~w|B8+!FQG^CPB#kwNzph;QR%DOILg-Nl=wcciT
zU{5G8CiqRhd~SGc^a(+I_|Q*oJPui((<J>lwpP`M340w)oll=HjH>h=s`@cPG`B1j
zUlMhF<@M!fZr9Ir+u5f}L3(HkRG8>YY(A-EuGbgb;q7s!)Jmy6Hr`qy^{kI+WmM{=
zQfj;cV=A)E=3EV;(=zWQ52u6J2(DSeZ864RJ2vk71I?%HCk0GDJw?4m*ZtUcdK)y_
z6Fupp$50Wb0-=XEQ{85?s-oo_*b=Y)8Ju!<DZ`j3*{NI}cp{uROcYsWEOGClT(*eK
z-A89#mT0;r<orFw7?w(lIOa7x+?REYG`t>#!>M@h<Ac62lR#A_w3{^6gGzKR0Bv;I
z7F5Z$@AMJy@65Vlc;sv7h#_0o>y7K{XFh%UnUXWRADH<>6E=El<db_RQ2$7foJ#VP
zS=I~8Le6HyqK|=(5u0rC?+(ZLUOG?9Umu;~IZ&9n7m69kRQXSbK#VbNaAru%t}=c(
zB~P2K>eK{8k$!fJ=e}UX*DR%w)@8b>zOwHJ=Z8Dve&cHxr_%rA^-(3)g|aM>6F?M-
zbJ0biMCA1Q)Ft@&UNd8aOL?LtAwTIeX>W#m&Bkh$(9e4j0Y#w;$T>#)<{X2v5<cyA
zK`)d!0dmZuRV*Uy$B3B{HjBj5Q4thGWFrh0#510Q-u1Z-O0J}GK_s)NU|Q(@&N^0l
zyRuEeVHCpCMvapwEX%@L7Mg&y!IsNETN_CN6^`O4ngKs)4dPsDtsy71=b#dMkoXi9
z;!kvOj1lx&T1Yi)l%hFkW!ZzW0@6}hme5=!{m{~!`?RJH1TmfO-rsn>|E^-7oWciD
zu9t<^%gU0|tf+V<zHR%4EE>seDIlPl3&-d@wmbLtJHLJX#=adh+H=f_PR{M6^2<N`
zfy>KFaMeP1H6`c->v5Q3u{jG7%^D&oo0FO%!`;%XuHzMmuQ3Kk3kY>tmPu{+6l2I!
zTr##s-y5SH!K;b}ynZyKByxpPb*ki<BDRBV+hdG#8`M-NDXe_n&!CMF@sr09)+SXv
zWzd+fU24HvV-x4w{f>@GyX<6QCc*pmV2E(Jti0S-N)a>%cZ@i>WRe!@x>9bRcsX($
zgc{s<YlGv{0wi8vFD%yuEs9GI&Pm}kshlttfTmFHo|EB#>BV-$PX07zn4cr1A#9o^
zgQGoAO}J#ljY~}ot@KpcAB}YiDSO{h$pPbQ=6d_W<#MC%h3n-L?~iZrc*O9J`5C*#
zy}PWRc)9$<FyrNRXI(Dr?;9nj`21;Zun8n9>sqK~icpE57|yOhEW~Cu9TOx}3Npt7
zp6TaC43DyYu=RtrCzf8GgSv(XEf5Gor}dcpXVYIS7P<U0tPywP7&Cz|Mdp2-^9tld
zW9E8we2%Los)9a3OH&okfe-{rJK=sp>p9M;ck`WzEcNpspQww6aI%liz{u_3_WBc-
z>y6QlC>)QCecL$N0mNiGN2iu22f8pCLmerVb-l6FHNtW3Z*2R<(b|Vg<-2tk+KWg~
z^BVILZkG$ojQbtE(b~a2Enn(06bG4y1;Ky!5pz<?N?kw2*KHf4`AjhMXYm^2Y{P2!
z<n%2{9GlNEcMT1L=SX!tVfOP7wm8_`E%@DI2h}MuJ0appmQTdtgmKT#UVA{pCnXhp
zeFa8L&dV8En}_z9su5r7EG%Me2$yB$_VR_CB5^VfzsS#sp}6CN5}Ji0(=nqeXepS6
zz0gMNyL;Oi?RXZZr*OxE8w(wP#u#gz2uLnD96ULwG;y2y2Kk`jfaFR|g)!0XniH1d
z{sn>)ySeyufjRh?>77mS6f;0dsc5Olbc*<d<!f26OW2MTVSoFLaU6^hIxZu$E;-l8
zjB(iNPAcJo6N`adMil8OYZQp)G&?Zwbcbdg?Xw6@<zvsVb8n3O)A|T;@3dAHcTz5q
zyy!9fY4|-TabH1VI_g?0HCGZ&k};8_56k`eKA(}+f6Mi|ih-PjbzLLrZmHxYlWW0_
z9+nOR^?;Ln#DI{?Wr-BMJwl1zw+G+8y|M2t!X=O?tbyf{DQlfDeZW2e?MH_FI0nbo
zu<oILuPfK<N-2rnKs;yyPqF#gh;B2|vgXhcXNp)N&T_;V8eP{IG>?aLY}=$TX2hc=
zdN)KmrX4&z{lL#AVs;Sw7H*3pXVy|fDQa`tqD7bZ{{BE6mRecrf@Vk*O3rj2eA_m@
zy+4rF`AaUWm&T#OQZtvea(!8OeO)Q5hCMCC%bap-Vp$7oe&toKh{OxxF2UCusAg(i
z=Qw7aEEp5wpA~+JHcj=+o<A*s^7KQyg!aJ2ksKXv*67y5g4ZI{>3HC-WG$$AWKL+M
zq?NF0I{UuIT>IW)qF4`)KYdKzJf_f{(V}^F93?5WURdf4p`hBitQRiJg>UlP+<1fc
zbvL90tt7(J!QfMIuT#~lQ;i+(Qt+fO!1J>RGU8`0YmUx#gea8h3MnN=XX!D@*co$T
zWvola+{V)fyo|>>N}3Q0an!aSlVeVVh#|T&W+C7JLr?=l;h3<b1(Y21TtKI)LMe%%
z6Jnj;FOCwPP-q;04w9mK-*$ffc>!0p$Ijcgue`l|W7{^m8HP?OZg#j+=-GM$9en!q
z3%~r~PXPhH{r2o!pTEAd4FW-j_?U2iA#3KktWo^+G1tlwmc4lxouNFDArX!PvyRj&
zw=aK>)=MbJ&#VKBg6DJsov7w{I7ND;PjtY-cBwOYGAo%`49J`#;5<VSPB=L!^fRyC
zrzKa<Fqqi;o&#r|_g<d=Kg|pd_izIlk&t=1a6+6np7Gj2<HZ1@a5?4htm`Zhp2fnM
z*&ok&oTz}v6wIB|JSQ!1qP~;c9%HKNA%5sdyYWOirr5Hiun*7i&51d3S5BsJif%1X
zjL^IUs3PK%Vo{0)s&~#A__%Bq$OK&`GNX>?9Do*f>+X>WfXmb{RcgJ^%NFUl=J@E;
z{4o>a97jAx(19m)6Hc2`@HuJU={^ZYAl}O{Y|5<8TnVRJSRxf>X3mMHF~z8XK-uUq
zi>7t@v9qVd%jZvAJTh%cS%N-FN+O=`2SPOIj^|hY+pgbL4Afe<zFfmTQR3j<yU`vy
zVvb>R@>0OiB`1#kV4o>_b<OPk;Qqd|O%J`hu_WcT7A`gXmGyF?8|>Q_BkMJi#*QPB
z*@WqXPhm}D#AaPmf~8>0p&}kItr3HNTI-}-SxTPPtMJli%}*%x^l$bN!XOvn7)B!C
ztCFTV+~XxMi^+|^RE>LNZHt7hL;{*iDK@H@;~fj$W}$Kp88;i4D}4+eN8|0jVMlze
zOIx{px$)`smHM)>+!n5vLR!*H2g`HnV2TBw3R*fdZFO#tV5?8)vCbBGhUeJK)t>*+
z3BwJGse*mTQ<-WbrYbr%?flW!rYUk$<eAT%ByzTowlO-Sw1g$CmN|A9QeF_<P${@c
zBm(w>qZ^mYl{%rW?eRXx$VRaML90}aO>2lT)I)5SbvhZsqJ;{QXEBpJ5-jJ2=yUVV
zlLryx85SZZ0w6}@obS(~Q|PFMh4Vy*eE4%z>aehH1y#}~4bkcCh~ZkA1Kyt;Y;+9r
z%rcmBT&A@xgsLjanD(J)TxZwsiOji+Ix=Q~qo;*yo@>+ZuQ~^@SgeaQ?Xe$R@CR@$
znyGaObYpwxICl1Jn<;F88lmzWz9KW{BZ`mAPe1*EfB4t`4SBur?fo0Ny$2VY5;|s9
z;@OGA2Sg!|Qsr{{OfHpuI+z~Y#<4#TAgj>7GgLoWH;-pn%7rgK{fYJZg=5-8V_;36
zb~>2{Oq}AawcyE_8$9|vux-*iPmKm=@sj3!n!1fAQZPTt)Cpj7bgf2YY|u*o+_&zN
zga1BL!Q42G2p>BUhO;5Jvr~N*bn+Ae8|N&DP}Ix(OL1__$lQqp$n%f`NlLEMCH5?q
z{Ta+OroAo@5dTgtIEA&}-J1{+M}SrW)T1MJV!s;*ZLT@Vg)<~a=L?vS4~I|6&;5DK
z*yi*5`h4Hc9D|@?6_Id^ZF5$PmJ2?EzXr}g7;=817>Lv?rU=G;wP*43&yUs*I%dq`
z<FpN)#g&tv1=_jJ0>txYjB(D1Y?(fvSTvYBo-=I(%6@D+y9x)!i#S>rN?s^ATql~!
z|5t6m)5iKYRs_Ef8K{}dW#w{NC^g{uqdRTyv^_==??)pqp?E(Pn7v0*Tuu{QHK!e+
zt`^{SUHJL+#&x-%sj_$H=m*Dsu&$HS9|Knj4po$3b;A|lQ^l=Ip%kH84@h?&MC}MF
zqUK0nOSyu}vzjiYgmRS5(atJ9hN-k+F$Ac<GKEu`LUCt<nCK>B4x%~z`9Woz$#g~L
z6f^(ec*f+a>WoY-ISzGIo&g>%a9HQp`#TRA<WDQFGIPR{pecfdf)CFDVO#i+T}K?w
z8DfKue=0E#Jc@x6_V;Ng?2~_rMJ*{t`$m;Kg1<6Hre#mFJUiZaStm!O0{$`LRy<(A
zJ{tQTKU*o4ng>e9<$@bzjj^}w*r&jB@b>m+k_CnE>u-N%dw-7+x+)|O?l#SA^1VMO
zJ(FEH`Zf!IImo9+o<z4OSkL=s?pPmBF;Je7tNny@CjlJ^fre?Hs`G@Hh46_SD2N$D
zoZ&%xMIL|f>20*hvxg|qCL-m;`*E(tQ+R3aq*NmG*iKD`L{V}+vuPPbG^}kdXpK!+
z$D9!7plYZqu{p-!EfQJmeMYV3bcrZrvq)w$CLbMZpyWhvd*}%IK_4x0BuZiQ7V>k6
zgGnY5ku>p#Km3XR>c9SPxxD<s{n(im%oCE<lk@iP55PcWoW$kR7ry-6zXDa>9`8uo
zr)}3io>xkY9?!>PnS+Mk{--a0<a+yZYC)0H8hnPfOpO7Yn|e^25o&cl6Xs^l^D~~C
zEqVsa1fPAP|1zn7GZ{}c&EjMJ`R{oB^YHxs>)I?R9OpP1t;aw6i~*Nv+q0=+kCDky
z+@J0OW}J9$8(DLR<fc!(<lyhe@MqzE(z@%ql1q6~4kxAMvluwLe}5tq_PiIIQ$r`k
z7!48eY(fO_)EAA>8P@4@@6ckDTxJ9HVKoMx(a1&ee4oIupdEEiBAsDl4pS7nBNGwc
z_AN^7wS!h#H(EPlDxhds3u{42&<e*94bFaS9Q!^;IiGgj=<35gFy`>3vmp4GOYr3J
zXUA@S7Rj-%oFQc=0y)M!7y7~2mEAO^Q}#|%!5)gKpt;OBE3jxHNd&+#!6{Brr}G_p
z^EaIDf8(D6pyugG&+*N>2W-E+Z?yY1?Rq0DZ1A1${)p`6HCjQP)sq;kOX262E5Cet
z;rjZ*<@$naqAr!C=0J|lPM=7KOC-h-)I)DZ&QWcjLZx^ZJ`c?TtnA&%5~DcFx*(*m
z|4gpW2!1qdKhX1_`QW#E6eT}>Sy`?t>+K4fS!<06h!TJAOqOd$4}LcXJa%0w+2IsW
z0ptum8odun$xoYJN;+ZR*nnRzIcz+{;WSjRBn%<CLm+9I5vhtwrX2xqh1xw*%W93o
za!hVmP)3i!!{=yCwSc@dd32d+29*>IPyXc8-R7o>MBzaXBqpa&^=k+Xt<4u?R=fx^
z<SFcxZ@yDfBWoq67uwz!t&?0?b48r+1HZrF+lIOE{{GHrjhZu_LYNwv2L(+rAnCDt
zjE3%mbzN95D`O_39c^$}#COJ~lb<U76lul*!+N*|VpHzZvK7zTC8i{bWR!>*7LPTQ
zQ>?$tM1Cx$Zth16?Rma*sxae@PakzSt4{PNo{{y4!^T4OK1-|FIM1R0AvyR6yW|=A
zrE@Z(rqezcbPj~d*d_R4^O-1_GMBotw<BO;GMZBiiIL#%x3_O0vN|@*dY~`rWGOfU
z^-}^Kc^^Oj@<;y7KmIq|ZeM8bjDB!D-Z_r#>0BsD!+J&7A<bgVCP^%p3tv9}LMa84
z*qFx=Lgt)9-7Tu5Tu_^CrimzMQeHoQ;rjA|S%)L??A!KW-*?U&4d!9bgoF{{=tq2A
ziwNIyO5Z%?f?C1j`||<MK4k#R6xMmkL<s#v34v3`@j(mtJmBrb&w<$7N9!@iW$zKy
zKE^yy=0r?0@DXCal;#=<-9m2%?QqJRvWiD6Zq}emhs9@ARqC>!8vB-{2&VEjbNh#(
zvk-`R4*Cp^JJAL8Nxu;qmviV`I)6Tw88<w4g&Der04b9Oj64OFG47FE&ZNG1Z|B;M
zX#^gNJ+Ek45lxZWIQpDjU_cz&_@oUU?>Ku1Z;zd$btGl#%PX2ooF8_&4xHXvAU!$3
z5ju<EJ{!e3l*!JYX|ADJ2z(#k97BfYvzExrNPs6JTbco3u}}B?P8*%wGkdcr5|SX=
zkz?n6e<znpF2XejDzKzAki1F##3DUWufOg3eIty0PNf=+qM6ov_>A|C^;j&)oh8R8
z;+i8nz(+*f+Eg<4K4#2o=cQCW-ELghh3oCYa#=~YN*g<S+c1x*V>+!+OI}EtxWDZ@
z9tYM3OC=PT7~^R4W~}SVvR>i_3G^A!x-3hCrD){H3sLPQkH7}elT&l{-g&<t^bR)z
zQiRi}MtpBw7u=m=9~}FHujbUdQk~0ESZX3q7z$z&FCJQguH$KGG(W>mV!fm@2tz{&
zEo-G-GWAv{mpr>KU_d`cAOI;K*I~xdcjyBtj(bolCv0Rr?osS*ap0)<O#DlfTq!v=
zhDi(rQqntaLXqGeqGN~{=-|6f?kixpZZS(jJi6K#;*=R}aR2ts+i^!dbEyx6S(v$z
zC1EK-TiUVH*oldf2O`c=E2#!N7HNx>vJ{3#$2Y_<4`gIrEBp2lD?B_>CYQXBWNy+B
zv8bvcW}QCfF*?Kh^qJ4i-jznsM)7y&207^k<ReKjI*H)F&s@d<GryhrEao`m2^H8J
zSR+#$b|&YY!DEQWOpv2L@AuHTF`g{uL^~|zy<$Ufo55hSzz#geX*THQtSM1TglYkl
zx<E-0l+=%y%xRt0j+uuZk-RbVNkW-V!g~0=Utd1)cYpj7%W|V1gKdA~x8MHCU;p|a
zXpj4cPC+7U%%sQAoO#t!d3||_IO5hA-MRP1;g06Q%JQ^zl@gmc8d28nMk$q)OLT9K
zcQE7F9^Bu*Mi|#KHS|Mm?K0PG>m(Y<g*?%bc%H`D_W2YRet6KGKwdQ1H<3A4;Nxe~
zq;rtbfAoD%SbcVpF%-<0Ix{>c_TzZcb7SfU{Cvm0i-ryXVERZ^j9J#Wjb|!ilAu>K
zWqd07Q^}lzZkUIfocgc+F#`ntxup#aXHAk4KlktubmEbIn9f5nh?t-mrcPwU80tPc
zCXHNrSSf2DZCVqiGP~_F=i|I*Gg3QMv|iA<2Awbtjz@ItYUlLE85A@JWF2jz^$1~e
zanf=@)|Y4wz2irp;al-Jb!MrYh}?+;#89vCox^8Zc;}yU3wm(z(1(1C8=i=P$VV!t
zL}BedaBEnbsO5<GXfr`ZN1Bqi2RRomC1dsa%$|vNKWyFNajyJFpwoZy`hCR!1otv|
ze04`1)PoDdp^9)@B1v$mF&iBUM+>ig_rdnqIF5$3L0uBJ>&oS}GSvCDy>ne)J9hSV
zpc+tM*uly(R@0KxrBBcK<km)nZ1f(1KuJpo23=y|OlIhW`X!}Mhl-;k_?xhXq-Sy-
z8Yykd>@u&n>)e1chH>=9)(%pNq^a2DD=EQk+OT4(TR^=@6GK9cy7!pRpqeNul#{=d
zNW3abehv+~{_=@m{`eE0UvEg6k<KRUdt@ENM9c^!X{~Ye9Zm59HaCWTcR10XbFWXa
zO!$1$N5bQzGkWU`YbetPFEw7OxzK{Mj`{8<x0ZBj5iIP5%&?i17u*uw;Qqex_PFES
z$rmVP#bh4FInT*&GK;H|Ko4tL$%T?CgBUmZ`tpgC3rBnB==XsCW>`qhnV#AV>IgyB
znV6OcokgD_)y{~H(>}(^juGnnWC5{9p8C{OXe3{%5ERt3vC(}(&R|_&C|UC4oS_}Y
zXy;)#*QJFf#j%+_7*i^31Qlj&OmB8*O7_RW-dpGa=8&Y6gk*^m_dNx1y~lJ)v!DUW
z6LE_|MJKFVO2L3@2^-!qI_!>H;~7(X>PDuU;p}?SGYU1P#(jSIOvx+Twqv&Q?bl!V
zr$7G(zP^108`NATw;lGpGv@g0CTm$yNmC?fGzOa+JD@g5B~dlSprJ7k8>CV~0B44#
z%sv|9{TpLI+c)}gkM;HJ<fl+HrGUWYLttm4(Gffs&^$N#6WTxF{qtZtcifp@adxp!
zSY0tO@>v9&`iAcWU$GXZ=q`tqZw#B-n5lzF8o@&z(6^02Kca*>1HM9!^dWpr5;%d1
z&oC#MUHhaeG@f5Ab@KQ@7upz6*v$-o@o*=c?uqApGa=<jgdOgLSS8RkozA4_uL6Di
zUNEW?at|?Q56kB0coS+#Bq}ipCJMEr5l!CBrkLBwWu3@r=n}?&Zl{JK&L`0+zEzMh
zXwAZnay0V#!usWBbXlWm>9IF|e`ud`U7iBH6B#`ZJw6Z<=cKEGhDhox(@GkSxsReC
zID=EH98*EZXo0SHU>^*R81X?+1QwQa?~%LEd#41a_fZfMsN|G_{ms|!Dh4!1)vHsW
zk_E0cbL5OHHSA6J^zP68dWvSOj|ds@8P?JUD#~?PxL#H?C+@xR{@&OmRF#kY03uMQ
z))Io48Y=g<?ZLp%lIR$ZW1lx?=FtzE;=y?%y9YPK#{)UzX*mTy8kQJyV(QnDxZN@~
zlx5A-r9}6k5tSNq3KG}bEj;XIxOZ|EN-ANc%9-QX=zHVLdvATP_k;Jx0|V+($r77m
zQn+3h`llOqY%zv(tz3V8<<rZR+j<Ge)r8*Tq2~eB(|3ZG7@evbp&#A{%}$w5gujpy
zntTe#B3|?497l^Hz{KcwFzleDM4ClHyuX9zARmiZPRz-c@N~WqGcLMjmgP#(3!@7o
z4OGrae$E(ecG=O1)SNNE#96NwZtD%q+3qpY_KZ*!RhAmblx`qodLKL<_t2@#d<hX}
z+xN-O#b*-|GHs~X41|#=99x?rN==h@n-kRB7%e>3XL4BcK~74`gOUVECB#-UKVxRd
zM8=sYHX8dDoo*?aq^C@8D%LI13S|c42tq#weeVG+N{%$cZLmEKnl*AsESC}y!9$TO
zGp9Vha=Z&0`>y07ENf<6OAKVO&<;$zoAr8OSyvEc>^2XRx%v4iMAbNqJPKzBgei*b
zL|qqBD!jkHF&^*i+Z$j1`d8lG-a>pfVdt~xJHNg%X&8;^gXesLW(Qr2o)nuyXpYu#
z9b8KJXPazn`azN09}nQc=mGcQjgl0Mr)4(s6@u$8DHF+lObtWWfX`wl&Otv>aZa_l
zh&~+&X9qa3;wOju%-@(OgPb~|S(HD~rKvAa?KHKB#3e?RFD)ihYLf|jM`x-ib-l1$
zuIQv4nTd#DnT-3Ba&Y)j1fID9^2{^<nN&i2#y*RIv+8mDshU5pD7Z@q^7=UoVCLSP
z_rd)chBI$^APqBF@J#Lx`XiD&O)5h>jVKa_%oz(rNK>2cj4GMrifIft8a^2N0VCWt
zDl!cRac`MFGnyz%V*T_JtXDK8yaxg~Eao8Sq?=MoQ?Q;s=pqk<q)csvd4FoH%;b4c
zw<H`#qhz6$JPXjXj__%%lqU@d7ET<G`4BesI~EqADNK}^-H>LeLm)1s2=r1>+?3AL
z6#b2=f!|dOtk=wXt(5r!;}MXOG}7@f$2Bo>!g7iPvXU8(N5l?y<FOy?TfqA@XKF1m
zW>SU6+nv802Yh>n3DcLPTrU?&*6AM&<#;=e*uPSuyRr9x%Z@e`%^vhbz)=I8LNiG@
zDloIBO$Gw84Txt7IK~0$)N29@3SvKw2p^iU!>t=RRcgwVwM^cv;^r)6q11J9vT=~u
zh;FryPPZN}atb~yS7BKSq6Tsy7emJ<uFDcZH?>f7p$&=I0>ft%v&MVhz$7632<%9a
zp<uH}8FA3L*c>1d(VJ-sj#LU`IL8?1+!~C|u<($3Es;t#EkIG&Oz(jYhPP)mZKKV1
zF1XzY@c7Ji{ldNtHf<4#^oRf;H;J*)IV#&ej^Ozee*Wc8{Iq@ohx`3k{{6rIf5l-v
zroC@g;WCx*M>`@8dTugUZ1k;1jBd>_-Zzx_=UfG$wGm^MnW>o?!>#C{@SlVZZo%mX
zk8UA~b10f(kXAjX*Twr0W0^ai1QnZt$3pi`?*o}uCQ+GVf2ZTbW6I!{PhS{^@s~gU
zHSD-q!V))({c*6hBi^y$2$j)#guB!{a|-52-??T~o$IwApRVX7Lb^sbN(neMtZ#Wv
z%$t+l#sF-_Li>ckqv$;S(QybQmb#EjWQy4PZ*0fGV}C?}HrjMaOrCqhhK~Cj5+#)7
z5^K>1`>|mtrXieit<QVaMxZPTq5nvVdURht4~zZSXK@^#IX@Nj%!B%*U<}10q>6aH
ztw*p=pDClY#MeG;CX-E^#eh-kLMh+5G0uJEJPhZQNc0pGM(SzyM1=dK87m#J-hK#9
zOgcmat4=){DYa*p8k!MUE(@2Rexh71fqjmS%_!>=k_Z21`UjbZq+5(QA55F78!}FH
zK2zvAZ<>Qo@vF{5@hlT;t^#3p;_~qsPqF8z<2t<??jfpXir+Y+dM)$>-N9Rj>Szu-
zj7&7LPs(}w_Ks{F*K$H#2$1n`eob@$oq|1L{p4CHsgi09^lx;0G{(`eX=T>w0CStj
z^mICiMB%5CDom;@HBYyNGloXGsGSoH2?|1|U~mc~&*CV;fzBM7IEPRp0+yyb!AFRK
z38zP07jnvhu6=jb3*ii63csIp^51a%u43R)3)giGsCyW;53GgKJH$5$!x<8hn4wp3
z4jUY1GmbSn_jU{}r9OH6)<^LD_njT01Gj-?#hN`;`a)FZwPs#kt^vI{9BqHP8cGSc
z-bUo|t9bZB-RR@MS{zpHGfo$IiY+21pm98Qym#`nN#>mCt@F6=q$2FS<C0k=b6L|&
zq??LTk3(t@qNSs^r}{LB&e`+PS=Ev#HMsOrL!DjfjqA0dCc(vNnls1~V@!VA0@_bR
zra(k7N4B|*1C!}xpM{9bMO;c{%_~tb$xQMpxTpC(j)1B>?04P9^wx&m%BD!g-Owyk
z6zR`RFDzRpJmrXF#0xdwP*z5Nur+0WbW(?!GGWm)7%?9r5D7zSrj!dQUl4aLmsd0u
ze8$)&n7sbHuVx%G{?|toT4{1~!Bq<V=pnRhoy0+17D~#L@@Yyk&rL15wJ~S4`xL&0
zNGDK|HD!8N{_@wq;x_0n7p~Wpb*-e+cG~((91MhFbPYS48{>#h$Wna9<ceE#WW5hw
zKY!wyUSM%}75=IHIT}t=#Nt493r9;Fb{cB_G+0(BwM3zF7>_*)0*F8^8JS}f#D#V^
zOYyVep|wWq2is#~-}Wa$A(Nw*DI`1R(0A*I6kb1lVyV|iX51g_$HBgD^l?tr`wm(c
zX6F3hOi+9M@(Zt@|G>5#Y>(eKwxFmwf=1SJZ0_@O_#CUfjln9)Wm!=-+HsH^j=s}-
z98+g@`<R`*Yiyn7GvFsBa;cH?F$Uw_=bkn=Lt@Ucw(e7`Ns0}=WXc-Fz?jjwfjoSS
zon`J@XUL9B2SqN4vP3t3&S??aGbJ$eR&l82e2{X`w36cMTy8gBfBA*=^XI8Y>F89s
zGvXcXc6OA}Bc_zdAxuJ2Om93!K+kR<xo|q3nj6;Ie8-~$RfwT!GBfo(+(oC&6_3w%
z(jX_0a0WS@?@>bV>9|n`9sbYe9*o`@vx`3!HxZUJZIcdrJLr2vE~^T4Sy+~3qEt*R
zQ!F?SP7N&~T$NnsfxX3AY8&I&5F4;$q-2x<zRan+q31B08jYFxfu$!sV-oMgxyG4m
zHA2tw)E&(ObyhxUQsF%rGEV1Rpdoqg3o%hQbGh8O-ab*6^+~h&MA53o&ve=|Qwj>_
zpJ_b)6W8xU26Bi3&n$D8a9Zz7HCph`Igv}kaas?!BW8x{3{;2%I5wWnecM4JX4rct
zi!w&g0Yy?&iRSF?vx<VO7la5W>NXgyg$1J>h-8K>n9qgdk^Mc~q3<Ns7>LviYoH!e
z*-}!hIme^(_`1hucVQ6Fv$;?$lCK=0ER}W1EH!es`v_<d<U^&LLU4A5beMmHkGMGN
zb$NzzoVPINS$TS>&OfGZg-A+GW0=YGsk=e<=+s4`8)zf!U$fn%mdbU#u#}aoc@D`L
z_;50@KxKAXr>#jf1#C8A5LB8LJF`CRgELZ8<HbDT!&AW;>03@Jk$Yeya=9S+HQWr6
zQO#(oz*NlJz{Msc-*|g`<65qivd|8}q@ej6E)-(12;V3ydF4{C?Du<o#&p_~0umjE
zv74fr$+=^i5fkdNa=E=QJjNTxppn=Ndv}gDX%`6?(ll|q{><z83FR;R<$wCh<NhFd
zXd`N=5w$%AZ4YbVSx^>Dqym<Oy&w31(M~|$qmXLOI0jh?ODd$S*u`j{!yA9xAGE!Z
zQ{l0<uz9M}j+u`iAp?vSW2g6-(a~mM(;r6Pgv%u~Ec-UNUIOMFw$TknZ*1FxW8ZON
zu+@CH(_S}y*=E>qzWnr=Km6Ulrg!7<{*Cdt)A})yC1NAj5SX1R*>lr(cgni*_y6!;
zqstBZ^;i1X*!M>ivU<X^Gax8Uy^DuPtM$&y^~QR=K<m%VPg^2VC8q;p9#Y}IcQixQ
zOilU8ftOmz^FV6H@tvqQ?rom?YCy^1a;obZseqFjI1fpfbcCN%1y3J;&^Y1|X2v{V
zm=@FC+6S#L;q8=CWV@6IAv!J6*Uz81e*S`_I_;9759uR1VxRkwS)d%DVvnMxl!a85
zD9pCW!O!UToTpC00--&#E&|yJC|X5k0_GIWeHWl~I%D*-UP=UMCGn{xI`7j6TEye<
z8Z#BrERKFf=;maLbzsia>JXD(otQmh^H8nz83Lp!&<w`8b_2c2C1G~10}C5-+Zk=g
z+aB~?xlq=1QaJ~?Y}0ajhWYq2FCj}H<9#*{CmIqi0H5?HL^`8$W*P{A-UNj*i(EfL
z-<+&Yk;I~zs+skA<$8OK88I^xB}Q3KA!a%#B}KSWa2A0`{cW|t?<)py=h*guwKSqu
zRfLjDs3I-qg-e9{kxF6!*`cn+qjmb&BaQGFq#ibjr3yAfTb7bTJqg%^_J-@+m^+da
zmu2Cy)~CJb^uA&P(FsNQ31K_Oh(aM1#gpRW3=10cgK77fUf|(aH>^9>A(uRPUwal6
zNkYh$HA0Be<Yy;XdfqIKenb&s7Q<{NT&5|Jmx`1K-zl}u?1BmD6EVI)82};-Yluv>
z_O!3G(b<of&Q;68x>RK56dXrmJ9b19HC4#b&5C54nMP=V1W1iacgzCe>8nR(G2u^6
zkLk@%Oo3XMhrjzzY!boYOwYN79jufqizv5S=hK&8cps4>SZj@l+GFEX=O5#stMTpq
zuPkNdh0maw+v`u<zx_%-_UI_xNhGplJS}9cXGu=dgy%?g%qfM&L?a`>r#IRa>aw!T
zK|Co{TH7(}q*O@x!EwJoIZwy9{QMJt|M&kbuj^-qXH@?O+CGq*A{v8rdT-NVB3Kjl
zeNc+>a#_f#WHnqPk<9y9OhvKWkHP4X#o=y9j@cNO>kVsta)-ag`fcI#$Wv(OlZWoj
zpBen!983dgYfi^$4AeVG#&ZK}Gx@QN80r$?NCV_K7HUGwbBRlQ=1-r0;!l72uc+$_
z{<ufyJt6o!dDh+ye>yFKZoohM^q+tJBcDG1!lMmJozbtFW@kQO=Gnb#%IHkt(=k{|
zp{@%smHJ7i7eApnf1nYw+V;tjyL%vI%jG#**XAKsm!Kq+3A=I@xl?NZ&ybljGa$m6
zPJ7ur6g6@-{^LtDkyD96z|Z*|5ry7o2Esr4^Kxd-D@%#yrK~sfa*cP|!a=e<cI*iB
zGmDW2n8y9?eUQ^R`a15-Nw0}2%vjrS5jmP;RJo&p;=~7uB5dA+002+QWo~~GR&PmX
zLL(mqP10vU8QCoJ@avJdCzHxakz5ze#@sWbDAb7xfe#0BhWDUw1?uGzDt;kyH$;&X
zhi^<CRgwco{GGi;(btY0=Jdr=Q;^C^sTnJs{n*HF-%!2JMAX^5`Aj!759HL&ITrDY
zMo?O_KzPzqW>aUwoO2;lixPojB9X1ImWq@c>GDEZ7HVBN1B`|R1(=8!W1oeM2s0-_
zpW2we<@#O4z%d%vbzxl>>Y6#;g}1F?NvWv@w~;&^Vk*&jN(c`_k(p+<38NdE4)!(|
zKau?8B9VOMG4bnqskjPn_XljwvQ&Ql`4g|7uQ<Z-IOx5F3Ut`BTNq<Jq4<8<R{$*~
z6xYMUme)gQv*Z#7-mKOjMjN|2ITdmWhd@e9>s?}57D@`gIzS}S%>foua0J1_jy4!{
zQqBQIZV^YUNvH2TAmw9v(%p^A5(A#ncbiRX;g}GD@7vp;lx4;Wm$|V7Lely{k{U1W
zQ2HPo64n@_J-xJIN={DeM+g#2RFf$sOdTan;n7so+XyS^EV_n$bOaG$JndEU_xrx_
z$aiW9vEG^sKmXzH!m`wk$O!<fJ!1ZZ#~g_mYBShJ<Co<^&EZH;N!-8wIS#rWs1T*b
z{nqIG&)MGYPA-MIR<af@%azM|c?z&1IK)9Bx1jrApj=7;I`_vtrYA}yEABRU{pnxv
zkN^0;;mhqa`Eud^_W%5U@bCZq|A=VfU;Xjl@bdb?+xt7ej=%7}zcG$*CEW7Ds<#Nr
z$$}KHz0Y+Rw#R+np8~~SfBlU`a@cM2!p~p+h>yW<`Ja$~`ZIkP$I;OoFs04N)S#0B
z#x_^1Gt)j}^U4}%PVk`d%*<kL$FS(K`e>8uJvrwo9D8<%Vf8BOg`fWLC)VpH&N=H5
zjN|hFSms*Fv^f;WMxa>Ehh?~|SN0>4LPwv|@Kn)K!={#^u#;ROT*jmDOi3su4#RFe
z68C4}E_#kD_KyvCj-L)#H>0_d%E}pT6MB$1EK>>UwVZ)HAEfwqLfR;j#}SiJ&xE`a
zHkQ-w#77>%IZP>Nnb@~RYxLHFqB+`ZW_Y3u=f+-3js2)Bq~*e4%61$aN8{`J8~6Ji
zJ9d_=-0HLjTXd|W4@U0ToaGV)(VSX2gRN|$8|M%*o4N!SeNr?g(ajE@1t<Ym$KEIB
z5BvOCv+5M%1+Bs?te=8q8&8*v&p9xz8JEPF<a$~rU4q&Tc)33XndYGlyWbzIrJ}V&
ze0vhK)R^s{;cyzQA?A#I=kd66>@g!Jr_kDH(g=XdnOrKCGWB|+UapkoZR%Cd_iL8P
zlkX4RT4QgG(TzT0rc4Ag=@a4l$TMMTN^}-cGO6B<rov^tuv|ZrUOo{4E{4-N8Wibd
zNi0R?S4`ntn`n@x-`DH^zNv*cw@)iSeY)X3_}aJU_}i4@5EM-em&jlDISOMqM<2A&
z@nK{S*xRNu(okHLCW>qLf0M%XQhB-739AWPR<5^&+ozRfscc7kF4{5L+yDf_*sU|p
z;V))$WN2bZg_0JUclHs&x^#NTC93~!f!ih((o)0EF44i4T&8d+ivQjXAvVtA=nS`~
z73GX~)s)E73#z4{?o>M%Hs<uJ0Rvh$j&`u`4@wfQvlEDEn@;P_V;dZ;$Lwv*)KXa2
z@Lc!NIr_o-{lR`bP!}$jg<7vH%Z*wu--+%3j}0n@)ts~fB#)C^G*LVb^fasVQz#VE
zANx~u8BaPo0+P>sfq;1S;{p1WQto8AqNk%H2PRXlH!99<$Nz`6KY5ZY%hLR?-<{r|
zYJ<B+WLDPD=q7;$1l&;g|BG<R1%x{QEyU1O)iq_t;Dec|>U($O;yX{xs{lb~EMO%e
z85QB-Zl?P5-gCZ@Te;zaXG#GV^fRvOBj@vzVyW9;o@Z)I)Fz8yt_5HAdnQGi61Td8
z54g}t^UUMpBN15=Lr_4{p0jOW4G<LCaXdUc^Z4+@XjQI?5%5L!suU-F{Pb7+;nN>@
zm{)Sm{P4TK;6MGZ|B=U!Kk)B<_g|A@;yho8{={fM@p}74+piEEesY|jPLz<@`X(g9
zF7jT~kl5>>j!FuZ>+MDZK7M>;N(*6fl)4e;Ne3s^(-9SVH1N3KtQaOzt0VbgI*K9j
zQSziYvOcKBdR~)#P!3CE|5rZx<MQlII+J1WvyX{sdBn$=TUPVn+vSbxc17J0I{d>l
z!*KXkAn%GXuJ1Vs&N;5z#`yXxxWV;uW#0=7PcoH$h>2!-{>OjgFt^N*TcFo!x%3(;
zVo)H}F$lC)vG91D*HCfa@GX|RT6eDZ8bGGtnS8{X&Mn265aNB<R|wNi_mvp@J(Bs|
zh9>RF@jZWUNwg_l6ldPqH+6GV8zzonK>x8W)AMP?M~TOBgS=(RzVY(%!k3p9MK*hf
za}%jK75$7O3De*0zr|2X(V4o$YAxha2vN^NDTUTd2VlZCAtQ%&M$f^xBS!CbynD7l
zu!wKlxQ9nekaxJQKRfuqaF9lz4!OR%3zp^qBg{<8NoNZ}Fvotyw@SNzhIJutw;SIa
zoY$GY%uEj_e46pgSs_bhqnE0?sTEmQn<A+3a$FqX?hz6v<V4zvVBHeB1CChw_c-KZ
zAa1Qw>L7$cX@hFE%7Y8O7d-Jj9dzUn4Cy_9d0uemNBnZaIY%u!v_dO|=Aoy4M;yL?
z##*a%`yT(?uirL0Fqr4WG$(3-QX5+l)Z@GrSl9z|-5YUo>_w?rwKR4U9J<!cdzn~g
zUG^2IE>MPJ4Uu(GJ@VtjEY5N?qKDJNLX6S){DUF@pC)dllFP<4O-#Y#Yb7~P@_{K$
ztn*BY5l5usNYUY!p<a07^p60T1M2h5Q~O{}o;fYzEbU>hxk0OpDvG1lqKEq!0_@kN
zr25T6c`A5J4~N*5ZogllTr>HWNmC>y2tgL7W=@H1FXY-;(xMQhX=1S<Wp9Pc<;vyt
z4L1r?Tv@}3m`<F}&x8~$mwctR!c?*XUD}aErw{80k+__hmxW$c^LeC-)zQeUYk+Zj
z_+m&Lg*IaxHW^QAnN}LTcxrpYxnEhQLCzar-(Gmjna6eI{O|;$liNlc<?i|J4qTjB
zmIv0;qp5@8^OrwTQ83Otuj>BBgh&CFWuo`RpgQrBk55mW(t;<k<(=EUTYpAm(Ae`8
z?`P)a%)|P?^TP-ExJ#iAPkp;6#`bjP<I@j3tS6QjS?3c^j~{3R*3&1Z<pUp|KJYQX
z`tSqi^&@`_|H$9}{oixF7S_~ANmPIzqQH6=8LNY!;gAR^x6HL}oaRT)PY=xVB3dUn
zo<5vez9cSJPnaF^8aSPj9XgIUC1Q$9&Iu8k2DNl1A6VDObc&oFCe~$UIYmY>==b39
zP4IW;ecjP#@9&L@7=OEO7fRmPZX0iJFO==7M$Q0XN}L`Z@upL`Zda~vZ{HnSx}08K
zzj4`i`lys!0XLYZ1h$PPOT?7LEE;MgYOt2f*f&Nl(27Zy9LqZCGQu&u^$;|PbeClz
zdM!Am)b|+b+8Cqa17^_-MDMM@Mx7pM657#gu^fkj^I9xK1Nysi{OFuGLY6>QPxTm7
zN-35<ztKv)8^edawj82M@0gcGJaw2D-LeL1A#YbMZ?Ek8#y}-Sr4*_kv?)r|ZR-1G
z!o0l?TCex^PwBuatyQ*NnH;4QD>k!BE7)f4q=|AvZh$ckkus>=Qz1J4+EdS+$M<wA
zBC^-#faXmj&;%zlT%fstPiLm{lWh65KpFTN?$6~t_^As%-}f7gf}?Xftt{4BN(lI3
z)N<81<Hq~&=55FWy{D~Ll9L8uSc0T8$^sjfhs8|~?6vJj9>sS9&@f6B-up&|kb%ZK
z!7NQeWArLJ<mjS<)<@#$L~Dhu7NaF#*!6wrGrrHT=)Hp#>5cbC@Bj4cx5W{s$ULRH
z4|)s}`T=#w(_d<(?4Asa09&o(TNc+@8f=9y1`kW(aXs;PUU4yUX^q?w%@93lPWb2<
zSgDAuvSYv);;^=Tmt|}?a_(%o&<27Phr~Lqcao1%jE6*Ys+;r#=IGgbwFy@TldN$X
zUTSaLawDdN5VR4b5RW<dzCAITd6Fx`4`fhsW+w?SJkd9U+6UX#v<(^!CyP*30nb(_
zZ-qQVNwULbH@G#arf*pv&Lu6()66_ic&A9@yk~Bg8}(X<4E!ldq!(1l37%<EkkP*H
zcpn+;IA17LISr@NBhSx2aymaVO%rc#FYK2WYQ9RXj9I7puzY{1WV^+K+{22hTp9GH
zP0BUg+Nt}E>-ECr`i74xvH9@)k=hGo?CfP{R9STP(eZv_nO4?iB?c#F2A<M8^Q`z)
zB@j-uQCZIF&In#!ctwtC+e<Mr1<@v@ccxk5P3HqKCPGMeo;!so!PW4%=16f8Y&or*
zPY;UAjzL94&Pkl-Cmue0;)mb=o)ja`AAjKO!|zFM;@2<V`1;FdmTkjFufFp-*mEYD
zAk2Bs-ZyRbJ~AzV`9Z0Zd*2vU0;b_RA3i<shwTGy8sOpS!vhcNq*lbeGlqh*(vswQ
zX^EOO94~WVIwjUMu`Y>e_T*}^sDs<Y>~k2z>&=bg%w}nWc&n9beWBIPzTL#V_R8>)
z;KYm1%ZYhjD7A{K?v5Nc_M91GS9bh%f$b(;P9208nWB@e)*s1yk(g%EJcBa}Y8%pz
z+%C9UmD1)_kTp#cAx;`1nzr0|o_TnD(lF5#TU&BwnB`QZMBHI>Fd>}tvQ17i!O(_c
zg4ZJ|y5FI2EStbK$kvsPX~T~*O^f%cVjkUopKj$zhfvOjZFgAGyyB6|K(;$W74o)`
z^KSpGIQ`zlsgJJf=IFivT||B;Emde*uya+pE4>x6MIbA}+^OZ2;|z8-t=I)Lz;|7u
z2f}K=0Y_6NPOAFgkHDh_&GZ|_%z6_$4|Hx5lzoz>$u3G!vAuyCjU(VU1Wy{ls2qvR
zk7rZ&){GB9n|nK+TV+^?uCUkoDD?_LiAv6-B$|5gKC~Q}|A$t}A3yMSJrv#tx;MMB
z8{b{hqTZYbXG^;S`~D<KJQ6Rx*=TWw#S9Iqq7`Hnjdxea`)~#N`#v<({JCGhX&7)O
z+URbu?;CqACR(s$weBexe4aBIWjMFJ5}E=u`kt8wtf!gh=ap#+Y`ycg?d)wZ%{rW7
z3b^FhTb1CjHCzbP)_J{L49*n4>AG5~auMHsS{KeI*$(q{(*rif{d@1D%aiOBqM?V;
z$uiY+dFkhMvebF?i7U}7P%NSA=;bzPt32se^cC)tvrW%l8-o<T(lDs*R=>uzO3j(=
za%0~Mp*!B<#$GBdIBHNA%oqwZnnKd|ITXT?t2PYRI*CBbl{(3f>yiIHrU~qW=Da5b
zsZv`lkfj~?@brnt(?`yyk4(!-@1B|quh*{%it*o->2!C8?namPcXUUtm3^yRUo*WG
z&Zmhvd7_)Z1e)R!BM<8n=hG7|I&!akdy}A)4xv;6lMQm-?4^-3EsPK$P0Z&99-lsv
zW(@>`!FKsV^PabFFZA0+a*;NwvKf3gIB;f)iS@j&p3XchkCdu`^0w`q)?H8=3J=Nq
z&J?BLcsM;;VxuR<iQERQ!0Gu1o}NDN{P;u)fypQ45qSIkU-J9k{|i6=@lSkvePIkC
zNZYReP0slH5OGmih<mA|HL^ZL=96QDOzwJidfb`MgCGBDB|Uda@0?E)=VhV~C<ST}
z$35<cBeLi^Q+7ws6esC0ynnxV?BYK{XUFI!3Y8Cjatec@HEPegNP8E&T{5E;yd_k2
z6L=l!^o}Q`PM@1;WZE6>*CA%qlGi2HN+vpRR;tB-CC$V%kz3W-($q#u>-g4<Ht39A
zxLq&w0rR>V6b$Qn;<TQGepKkAlJicjnQ59#3@HBjNW>(m;4DqEl1q*7SA%4gI{qHU
zGw$J5cVD`_mPe+EROFpujggj3t?ye`sM!fShgjQNTId);L$t7DMipu|1qCg&aVXQ@
z;r{_l)5NkYc&EX*AHQyxJ#B!=wrwzJmLJlIAt|~T323=f9j2pTm`(8A`fWE7AiYTN
zX2Ub>i4Q_jbkXUbbN?+5cwT>>LR9w2@y8iz>umT6y^Dq#vX6VGB+Fo8Vdrcpt4(6T
z{YKe$(X@y7+k<!GKExf)f#4NmtUt@&vkJTmbk{6M%-awc38CYCCxv8ml)ux;YC`lh
z>jXJUkev-_a(F~h>wRdTsEw9)hU;JhPzatS1*SMz(Uu`M-_cR;&(*m9^Z(M<Z_^BP
z$L+EaVqo7g`Bv<Ms8KbKM(KmmI?a1!3zS07jW{5TLCclsU^cEhBv0G262%52621>w
zE!_4lE^XhbxiMRfX-bi0UhpZh^^GlO_R=V&k*3ILS$I63^-xubb4sg>?M6&Wi1T<^
z9(-rpHpS%zxgv&h&{c{QLtqY(IZebEb-<dHqf2N{cjBLG?F6EgdJHLpsnpgft#V|9
zAELF<2kN02<Snzk-FSPs&~sxf>J*>az|RT4L?(NXJdxfVwP>nWqux;Tr}sD)>C|`r
zs#dPsg-~}&p9n$8eQBBy!qJIolFe-#53N2xZm8CCFKV}&lZKFJNpNF03+*u9dFx!a
z%H?I_+pn+CJN<cLN?EkZ@FuVd#5A*>A9#BDfb)*a?Z)Z+!q~Lsq!_?#ZnYLd5Y+5K
zVp<=0{^9qm>zRiSA6ce}bzKQAa5+CQmX){5f0Q@~*n7pVlY#J_(|Y1OpP1Ifye!0!
z$a%l3t^2UoqN!$eTF<1Iq<F?@_={efr;9|Nn5LCEPONEWo|OI2Yh_9k&ks*5Y368#
z6hB=ltw^!pVLc`0Q(%M{zXz7nL`WVNDlv%%zik)rjcIXwnje`@ic_BEU=5#6`kbne
zp>EvcXoH$<T|3${J4)v4--YscRX(@jQR{@W173B+ahmbH)3*!0Dgw6F481@v+Sr`z
zbhUME*9-IIjmdNa!7P?L+))T;7sqj&4YYXlTIeC*hT@8k3szbkLF{E0)#dwpB)UTo
ztN36!m{Lfb&(AzReIQMVl8Y6Li8EI4)Q~2~X?r}j{#u)DuC^5#<nK)zX!&^@VKm2q
ze6+TDr_K7<+`RF^$5tAg^*oGsufEA`^yhf9`gLeSseLVHTHY8f6D{<nGzohh>K=$*
zy#uRJadQmNE42MCraTfDC4Ot~gFtWe;Z%qdC&h)2Mr#U;lbuz=<6#eVhOp{_)oT!<
z7O5kTr0KADZ?LuD2KPRc3}laU=)d>284siF>OoDrz!V~*b$pP$uM}mZq?9<H9!PO!
zG__IIyi@lZZQr3uU*eB9!TWnebMT&Jo>|w0lyv6A=$R))&O7hE8=e*47IB2w@y%P<
zm;QGgnwC~+T`9Vz3iw_!pr1A9IdDZ5QA(xUE-=3^+CoP<ym_8k))Vu($`bG11%&VR
z?mv?o`0ndB4Fj!oO6hlcK*a#2L@tere4nLMa_i(F;ovBZ*;<-<ZOX~-+6;Q{+_u7A
zDtM=y@v6kOw{7RLS*b}M)NIwWII1t&rtBhY7F{|zFueV935*&FI4QN@_FbVH&f`-u
z$WuB0y#>l%84b03Imf)roF5*Tlk85tcaCDQgYz$~N<dXRlk4oeMSg}LDAD*M2aI0T
zU^h_qe6930*lvZ_*PZJnLvKtWP}@Z4ju<B9r~tu0KEHl(N23m)ds-d%^!{K4Acnw>
z%4wX%vfdT_S6!&fdV+HVlu{RBV6XbVF4s4lPfYs@p+s)mD>>i9FAj))dM{M551(4>
z1aszGxn2t|UoL$4<yD~>Zec!Mx!f)|W@;(y`_6KDU|LTs%Sx-V@=a+~dYbRt-rkr}
zBySs+>xCEsm)n)$JpcaR|2KU4@PTz*RZ0~TF$C6S<wriSK0NU9^1|mo{u6b(lIw0$
ze1({JJfB$SlMazC-H28dh`K%TL+bY*KK{r&uZ+?8wqN=A%dh;+AAZl|0;5(QPiM~4
zLfJOE&<<67<B~;X?fXWj+w`BT6Mmz1!Pl<C>|mNB$$6>^3_s}JLFn|cQLC&--C<=m
zLKC_uRu#iI&${SP&81R%F}3?3d*x<~=4g3Pit_4PGYg$p+Mn@eU^(j;e0)BinAc}6
zWuuPDmN!NdPw#ujk8Vk}7VI-1?^izm^b@64&JRyg;Wul78+IAMonJq`q&zZBK+dFT
zVVP(ALI0FO&iM`n4+{s>U$%<TZ79lo{O|)meEgB~`H|KN1q}+KAgi5c%>v58Juz<x
zlQwOG4XDUbgKeQgkKZ}lf919JQl%dbr`8&W*-Oy0nn=@#cl5v+)z+P5jF^L>RfX;}
zwI<-5!f8^J&@zQUwe-B?9CJ)OoS*sl=?5N8XI#%3{)SmE8FwVcG48`px5#K8BIk!E
zLJ*=@Tc*oOU_*n!(fx30V76raaIGQn>G$D7ZOFP?8zNlmT~`5GJbLehv=HY-OG<CF
zEW2y#4Uf)5@%82;8O2^Z(KPFFVqH(lKrt;>FB|Q4rR^J|HeA=we^@pT8YacWWamk6
z`krEl#BjLc-VrnfDPi#bd(o8lT|#Vo4%!{{5uM<VSZEsr4hyCa_IhEihue$JfVfhf
z#~H#C(?psVmNYS&9^+qqeh%v3H{9>PX&Bh|%-idQJ_h@qxxU^M-Dq~5k~5`u%8)91
zm_oJXio-M81KqvEa(T}*2e-ZI4Ni$@+3sbRvbAjD_1tib(CGoTA%3a#jzb*p6a!3H
zV3DW`*N5c^NcDafLe)-nj-+|PDsL5kSxcoCW%etb?0}4p?fVeV$bRI4b;<X}E$?*K
z37b;_n1y5XZnd0V70$<08z(_n%@wu~<TJKS?kKWIlf+KbJc09sP=rkM&S+h+x?bG1
z50M>9rgPHxw$X6K+zz#{t|xogFqN#_3<o59dHY7rJKO%sG|%k&#_e_?IEC{}^GZm`
zERUX?GuNBv?efKP+!N6aqKohTS?}4)&fDdM6i?i48>2N+T8Pu6fGUTY2)(B?^&B_`
zW#717Uitd<Gx)&ke&gZsiNF5CU-7V<SWk<rab~HRqoBMG^O+xh_)Gr3t+IXnm50+p
zuNxyQ%yH4??zBOcTG+dyvYQ?9les4xgLGQ?-Sh8=&e6X8i9db&g<H;?;sj$br-ZAO
z{q-x?{i@31Quy}v#@Cmxyk1`U_VSHVE2q<m6enEEOjBS?5kCc{HQ;CPlVh1Cm@G2e
zhGrjx`osw15SHOIKm|v4a#F~Nqg?eRC2~*HQYo$yV!zwW@))!-*sDZO@7?MOmD2Cw
z??^(LC!U|5aeij+JJEYm3{>Nio5AB_y!-Bt1Af~t^v{ja3hTTuY;cERFsBKs?6bll
zwj1vN^17~^*OlxX*IL-}Mk&gm4^C9uxI^UjpW5b^(#*$?zvJWcM+Sp^yODFgS1}L9
z{SvRGT9V`Y0ex`TrAiaygL&-@jdK=1eLvj48wd`_xwrn^*4FRHt+(M&-4%|0c<4o5
zgHt$=Hf9Z69&P?9M&^{rZ4g7gzwb&3bevBQJU@Trbb4Y+N<(ZTFznfm`T$s}s4y=q
zH)NV-(t2iE9{@)!*}{9q{j$e&BPzsdcO=n%FuFsT`cAn8&N%&n2{oM}caIV%FcEWH
zN%M&`Er6FL(uB-?*lQZ4#bQ8B=`H}5SUj<;C!r|}O5PZ~lD7+eze$AHn{DH=OdmYI
ziqM1z>pd0q<k1-!gTD_0%3{$F<h{AiLcsZE8VnQm9Z!b~7TDw@7&r!`<6qyeZmUCX
zow8q~EeVB8Wa7b+WhJPcrimqZI`v<%0e6J$H{9>PX&5M_a=T?gY_(GNMmT!kk6h+j
zZSqm8VeEtCJu{9tDs%8WEEB=Qo(uaJ)G=5r6WyWs)Y2Py-^sVl0veqB)WNY8c|4cX
zMDS9%Pg9h@C`SC~^rG*lTLr9C{Q(!PFTN|BXV`~ucy0J4Yj_#8+jJ}Mt@oa2_M?Fb
zg@RzX7CY{{A*7YcFq_d)Hc>{W2MbSe`tK5&;A-zWfGo#%pslX;PMW0(ciu~7iwRan
zHwe>_k~pl=#(f$TbW~|AE1Uz*ysn%dA1%w>5hK)^>Ah;I#Z&4o(OteU%}K6}-ic0@
z%#aon-pvvK^@(NjTrO`g8qqJD=ZStE<VXEp!3Wl9qV79X9Y9k`JUxHp)2APJet6(k
zGkpER?RH~iL7lBOO3&P`7p~KZ;syoy!|xw?SadnW6e+dnd1;k2Ppm#L4aX_Yv?0%Y
z2rF@Vk0@<*Ck72nZ*MQW-LBlW8!iOqG*O#kbzd&8e0+T3!-r2y&T|Qk&@yvK^jh)B
zlZ&dA>we+$m!Ek&tpp$V^7hJ4KmWw9Km7^MU|MF@#|OeR)5;(&krY>!hd^8=!lY~q
z=fNkXmU-u>dnZjE7nKR1gf+YHjxwYU0`1W=5)=nM1#7_@8?ID69Frr^x!pIxuCNvR
zZ^zI97^4KQQ%Wr7Gs`klsw{R>aLnE_LeS7ygmxV^r)rl&AN?NCRc=>YEil|Y);9W}
z#R2II+ECi){a|#C=Z~NG@bL#inkc3dIz)^IKOUUN4-Im5y=l-75~)5tePp2TYs)*^
z?Mf}%oeLfatsK2;5IqE679w;sKpvrB{+JSMpw{cApW*$|=dOu3&;xG)K!5~w&fUvp
zLPeoyC>;;)5BLB$zNc|VBaZh-*#o<*l=f}Km^h!$EbB}_X`R-1gLC4hV^n4PKw)wr
z6nspqab}uV+%z+M;;7nYh(>VZxX8TDW)#9x?Xn&NToB1rsvJErxcfP*Afq6fYUrFE
z%`CptOgcRfma{HTqtj|uEVlPTG7n4iJ-5N>a}k}gCB)TtQBAGU>rTx#M#~(!hUh$g
zUPXHb33%fqRNzRCY*n<wXw8r#8@RB*wQ=9~g<2njp}l9AcES0eUWY?ZLbSWP-?cr%
z?oz$&HlQ|%MC%AN@5niel5S1b!67kXbp9}+a5#eAg^qvT*KZmIwp*dKt3*PGZ&WNL
zIA<t;ZQ#9XyIZ~EoB|WzP4W4(c-AE7M4vt7%ww8@kOHF&`qrsiGljRuNAY08Icll6
zF|$ky4oCDt5JL3Kv$oB=Nz>39q~r)uk(VQSb3-^A7Od{LG@7emwne-i+z=+Su9)4>
z>+tXvBid^t@A9dq5G<lu`TD~NuCHV0jkoZg5VXw%9Mxq_m(J;fHU^!J6aK_-mf0}g
z50ToNeDHhYC|ufZ8`tZNZQBV5;?DU-bdB}&z|+$s4-ZfBRR_RVMz6#e=w0AtGnmlT
z&OWN8@+FqjnPq+epUl6C!cSsg-!pE=)i6&F^iesjgX`(azHK%<Hhj+bC6Ut1G#np3
zeBk-~z?@WzoRZ+#v@Ey**V`L4D_+)nPjXH?UXZXx{%IY~wRgU}y>e-VA5V|$*Bf77
zzmTsNsm&QI^Mdn{QY)9+3#?}g(P_}DZ48}UGr21ow$s_F5G<m{$Hea+&a(f6!1Z?H
za(N@gz;=1#<;!P&wt$l_Z?9ahuZ&y>9Ueb^;BWrjf8go-#7{r}Gnbz(oX?Smr-?Wx
zhA(>NY|H4~aTH2{w+;Yzk3nU;Z{G^P>B?}hJa!*+$Z(HBwZp1IaS;4bjB?<KdPZZ6
zOa_Tg)5OF1q#`n8eGH=$8Jt5P%`?>>P^sRBdvHy6=G2f*WsFA2+FDz;q`#Ez2Uw|9
zFnW;q7Ux*ck34<+#5Ad`Q&zXb2JAVK74MLIKOk>wm5WKa6s>Z(T`4Vd*)CkJ-x$3p
z;msY2a7++g@m~L2+ovr&CdT@=b~rd_LpyxO;mR-skMDTgG4APghr<8Zw4IX{saY)e
zF$^3(FC67H!;l#huQ`=SO-UQABAn5AQ!+g$<ZaZ#s72Wc)6BH28V1}T#EA-Y(@sUx
z$5gX5_gW-mt66Zi1p)aG@logCaL7$!C=uSV=YkI#^At%@T`JB;TpUtFH?@l@^B@wY
z2jX<PQ=GXJ_HCo)9S5XixU#|M-Tk6B;BK=+IRd>k6{po*+5{V>QV5nma3U=yrfDHs
z!)b7s=(bsQfSvp9y<zasKWH#KP^qJdQl(qX9Y~Qrzu*aBV$d0`-Gz+@8gox2H1|k%
zcGuX?<UzL3+m-Edr7Q~@5J;rxZ6Is*`wiS-TmCb@ej74CWz@mQ6+aw(sF)!H=pI}Q
zg4s}Kdg_DOMINFj7!N)N`0???d7Zg+$M(8Qxm(l|U-F=ZK(B?Gv!>F~scwKvg6LBu
z%oA~%nb$MX2Wqa2RtW*-q|Gcw2%(U3QR>+}D))$d)O4GC#K*+$Gc^{n!JP9PIX^5c
z>jc&g<($V4<$PC*w5>%!CR3VN=7l+eLn&5UE{dE5T9gAp=bXP!ezhpvhOP%OMbE?R
zl`Q24bsU@95veJ7``B0<Fk&R<!tJ(`b0zw5Pkf^*jZbA80T%;pRC3+qb?sP=e7!@M
zcZj%qmh%%I9)Dy?3)i+&%8gnp>+J2hbHq5)n*BL*pq7o=vU1N`XD=DQEJQkO-|1*V
zcDB|=?}<3pc_JoH%T+4k;aJx*PfyR3R&goOduOkOdU?eq!Q!<PY8EvAjbGUE#^3+L
z|H}63FHi@|d?LmL8Af*AE^iF5KpDpr6D%t!EhInT2&6D`T24$!e)aAYcRJaj(2y`_
z&l|1ml<mUHw_l0T6D_#qG$rzq*lGOm!;k#U@BWHuTIr+l4?q1Kb##XBgs2IvfelrW
zUV%Q;LE$`OC}c~eQSSW$AC}%Wy2`C$pkbUSj^4UmXmZ^ME*6C(&b$l0wG90!C6**B
z?&;x7jEP(dxm>AbXY5zvX!sa7g_+F_qEkI?Zk60Bt|?<6rin2md=oPuTEY>uEgIK3
z`jKJ(F5;TziFrNI?NZ9Ma^0`wvZK4wM8H)vqqg-!gWyR?mQ7ldg4{GnueWcM?V|KW
zC+k@-YG5>Hkg>iW*#Y07@II7OCjqjh?^*3G@E$vRK<G`wkK+*2TWOLzB6gKmulLb9
z7@7pEq8R%<_v9XE<eXstp~~)PX_Ci4(7nWcwPs~L^qt@#3}G-KO)9aOBH<W@LZWma
zPNvqk_oWreexv3qqitr33pNl)(=b4RPa_a|yl=~W$h}qjAb|$b6cjdO5!jt{KwP+a
zVN7Ro2UjHUyj^dKz#mOn6Wyzh+Fp}t1LP5|<Ltt0zFQ1;XY~9X9U%>jLpqBv&7@Ts
z9#il*SCw<pvi&+o1crCS;0e}Rt3gBwyWK8ycYs-Xo6VYQNC{rgiVcu`I8j?&c`6dS
zV9&nw9k@eFp&%`Mgx)B7=JNK!oC2;tD*tgLK<LGNm@#CIKXV@VO~XK%^&+&U-2P!d
zsk1<b7(H_q+`M?tL-fo;O;ok%MSY$VpUx*@l#3vu&HA>>#?=_gWDWt*LfYJ8N`l^_
z3Om-*iHD~r)^*XASS4)4S>>Z#f|M+TBE^X*2r($VX=5MWM0d(M&jULsHJYE!D-S0r
zN>$yd!=`mbDCb-lz?L)g!5kvV$)nzSSN&$`xMmzs9qhL896}MQB1n~(=2_>+$J4^|
z)53h7NmH~A_@;mwn|}MjD;<u6+PGb>Qdehv7^a+X%R&q@ab9sLGJLl%o&r4^9EJI?
zCB{;$vn)5k!91OaaWcdstEavW@*+VKW02;VW6*8A-os#W+3`!d+rVzym0lGy`}X$A
zb=#Oy;8rsEc4eI=d=In{SeBJr-AL0+Z~;7Y=O`s}-EZVuWj&wxcz$G_S9<T-%wN9n
zdi}<K{{Q`tl$Q&ni1!OAp4E=ocMT`wiknw$hjCUUwO2FP^V0{WkjTBToX<oYlUH))
z<7wr5I?-C^>+3fq2L{jWcH?rr@bS|R3bE;(7!z?$G#UvB1%uUTm)g1Pg_OI}34`9>
zZc6aGFw2RmAt-l4ZWu$4o#?JUs)D1M_(+dcBfDVR-68Vo=9K7k+Fb;zdrIHKJgdD=
z_9*P|t!j(hulQDoZs3!)0LP3As*w&=HsYDaG)d%DN@1_Zre@Eqmw?vU#w|WJ&_`s9
z&OA-bv)VQH?M5zztyV52vvoB+2HtDQX;qqs7*!zTt>x4?<sXzxYlCav*vn0qqc;sx
z=(v;w-zK?39LFJVy#c)uTBS5^iFxC`SR7p#hhbgtbtia31{`-0-GK(En9Sex29gtX
z<p)>@m~N8(PDdD_66~h&UTt?NHi_Ggd<+C3Z?_vaDz48gF_NOHc=Z`SmW20e<&YvX
zgI275w={a*8AY7Cl3GP$sLMga%;?Bz66K3;Mg009s5^Z2&IS4F9Vqxsnvvn<CLqoX
zA5G`AbG_ZTU2pVS^}XMNp(?!|DT2o^>&zX|C}lI1tz{Y*->(6s8^tPp$P@z0v=GuH
za~W1QU+pvIupwEo*C{H=_Ry3_Z)0cr&@y+p9}eoMSczV(s?+v{);S9s6EY{d=|CPr
zaOA2*impMUmqPJ_5meKDyKc<WObVXS4@AKP41IvJg1kTTe*bMb15sR*!&CO5E@~<8
zh$s84ZXQ=*^^PCciPRg{Yhl|B+B>ZlH|b>utjgohrQ@Qf+2rX{B1K3cvDYSO#w!AK
zURLI5(j?zCz4hV9RTij$v<yeT^s+2GJw39ni<0f?ZXX6@U2EdhYOy53W=(i8DX7SM
zia}NlB}bKBY1Y71O(C4T4yXf0*TUu0)MoIhHaL|y`518rZ(F5gU)G7=e|qNAhX<y|
z6E!AY_pEwe`!&hO`{d`)XsQ;*O4;?G%<IJY^uY4)%=+;M=JO*V>2MysFuF~^LtXV{
z&y-z!?r~tL;#o9{!bfyaZMQ4iR%jL0<-sx-0^UV%p8X~TeHajiJkIC!%(9%Qy>ofH
z(62Xs`RQNy(}y2u^&`1tw%hg2+d3~4k(4AXje;*na7@z#K|xF2+w1N;Ynn*T(f1o~
zfBI)`KmS76DsevX;lm&J;ps<ib?3To<dW}-|LUdFs8)p>VqjWN5>b_1u&MX-t@8MA
z=IMMQnJ#9Y7d4*wL0lFdAD=9#GD1)gS09DT?S<xGZ##j+mv7QX%pUUYa43Ap1!ce2
zVI@Sp6Jr&H(j0AMiBF?Kg{+p*@&hDj(qO5;t8NN;r*(l=J8f^Kyw>N%@a?-=S{6-d
zUP_iQjV=M;U6gmgwO-m87aJi4Vo{y2w?6efXL>F8G2jrsj(0nwpm>aa3oP@>JSXO;
z!m?5dFJHgVgChewP@MykY-^@RaHA1wgHiSQxO2NsSqNAPUn@q%xqrj&i{G8slkBub
z?vi05!rr^ZeS1=f;(&(^^#k_D9~i(pXyd*+itn4F`|hGVzPu&b4e`<X8sEXi-%%NN
z!Q!}I9x()#Mab8dGvF0LTnDWP)$&R1A;+NQoqYB7{s|qMQ_=<1lw;7Vtc&g_@Hy<n
zKzDjphoL#LJ~kqO7TR6J7;NzIK3IvA)0U2`h>V7d31VWPQmI<2U3ORUw&R>*O7UHo
zYBbF_93AFtYDW-J@3aBKzc)#~4<z;ub;}&6#rk0a&PnGoj5-@e@HpNrpwemRXL3&V
z+hfu=#H+y@`h(8;QjZ#QEz&-kSH|={j3WzV7%~vRX*uqF$i_ZIL%XC{dm&80?mnGk
z&IP+?-i3)rcfz0Z^&8`eA*9F{vdpz2e`gCYO@R>gLQp!(Q0tPKv6R!Cd*`h-raj}b
z<I8p9x{C_%_mn1ic{>v2g`lkUZkCNB_aC5UQ42XYuGdVh)go&LX>?kt_~=>JnbY}!
z$Hzyh#q*VZz0fV$u(hV&=eAL9I|>&v2)|WoR$b;+tG1u$@JT6&#{;sl5xhL#uF9r0
zO_N~8T<F=_%s9eHG`-V1HS1?+tuxPw)9J*A=V#(FacP~mQB+G@t8q};5`y=dj=N2t
zrLpgs+g6y)@)n;To>?D0(7JLMLJ0a08N^oduB`H0$$Md*R~rT*r4$vPxSt7rur4!u
zz3II%r^9+aaXOu}t>0eA`^_3AGtPIGIkGG(r_-4<uej3iT)16d`1<7+US7X(N|W`t
zJ8s*R*cBe~dKC>YO%px_m?HDT%F}vgiji;ESITJI@<wu=T#D%jGH-9+*td->cmDGE
z_k4Kzy{v@O%){ver}G1)TVPC5>}_u_ou7yg4?KSOz@u@OpSKI2U%qk<6Wi8!dRPTX
zJIC|mBkOs^Po57SKl0)61M704xz6Qw<@WNzzF%d#%bBm2uUz&k|NJli$mkW<6QhXw
z>IHnaV5~mawk*mg29|Zkd%?*eIKhs>-z6$E4)xeHwHZqwCXt|d*gIgsF@3yO2zw`J
z^1QAr^TKx9C^_E|2n~<*z18jD<qx6Lkiap7fKN(!tgTV9TGc!Xbs2YI-k|Oqhs{c1
zKAvSgaXy`J!+JV$R_bJc8X`SLd`NOs*ntg>d7g=}Q}b@zv97H<cR#x62)Nq@`xxI@
zzD!tWrCE((1HHw}j!qa}pP5kcdSrAN9SDUNZyS%F$u6JoV{_qs5D3-|ctG68KUd?f
zmDixMqycwCp!b7@P|IO(DnoN2iWYa$6uINg^QO6SqAJ|zxZWt61rnW~n3q|Xb07F#
z8BHNzo_-&=Ip#Jz&BUr>4AP{ogjR)crj$%@hX(6uQfO0BRr@${R)Y9+9}{U__50RF
zzuoSy$#5pji_%sc=!9O^NSaIlxo7gWNi6NWHe|a`#t<!b=upOx9?M4sA=R1zNAv?X
z$~_g*i7yp|uajF&1G|TuZ}<VYe%G+v#h?f8-^^8Js6g%DS|=DPAOb-{{Jq4<Ugraq
zkcctd2S(>S!KZgr2OCza0_MDvDD@Z`{@kzMrWx1<!E0WMNx?S9!`9j`NJbA$!|}C3
zbc203@{pzP^V^j;1yYP$>kv=n5dRT_K~UlXeM}51A4(}ng=(-1c-8kh$F?_ad&U6|
zYh)UZ<UHM!njuBP9EnRJjL7Vow)`%>ZVcFNZ<H<5_AHgSpUkd;Ak*0syP6qe(4;y|
zvqG18H=&|&S560x0)N`QHBn(DdZ3mN1;N(Az7;OJa=@1*8IN@mH#I852m4-3oQaB5
z2q+(c!8Px^Tra#_ZtQvI*~)6>`v*`k$hKdo`6e{M{Qfl;YSp%se8Sm=SZiU7Mr~KN
zd}Uf@rsd4TP{zS&ePmr%#k7V%Z<*`$g+1@w_Kn;Nm~b;jXCG$IG9}*1&fCiuN|9A^
z3Xx@&`u}$MMk|H4moGg3_&Ys&IWtX>$J3b)=LaUQq3qYU7jAiD#Kha}jko>Euit*<
zcD+z*Bkv9GCOqnYPa*Q*<L@|~Pxv&ETcg&@?RMc510S9~@$us)K0JS5c9E?We%@Yq
zEpH4~zP>b0r<otlj}n{hnRQ*5mx*(lwMhgosn;|yPAmHx#3`}F!1nfyOBsaGnY_1j
zwZNzj=eu!>+9YbO_Wk?&w$fUsJ6UZ*)af3gM2EN`0i6e$(}CxXehjR;z0rGS-zz0o
z>MqAknCzL3QC{#|Of-dTSVP=2URn}2^ene7vFm}1%}LFI(#pJN2{^knhqFe)*6%Un
z&Ur!{EDt9>e*DBT&*aOMoU;kQsyyXEzjiol{j<wxFz_zu`3Zr#-R_9TL4zDI!?<qW
z`};-ThiJrgd63xADIKm9{X1`+7-|xfkTVTS80CGN^D*kV^cJ0aY@&7`-~V@Tw{nhw
zAa|n>g7-A<@4-g8^o(2fn9%w?i{1kalq88K_=%}hx+~IdAqPq@J2bruC9t2rHxs3r
z>fO(?!jwj*E?Jy)k+uXwTJ#MWjTjQ^>7+#{2MxH+88oWPC<I-eM|17)y2K>T_Qa$z
zwSi;brAJCB8GR~L*qN997)L^o5ILksGm6OfM#<vfQ<QT^kAZ{kjNczr9BDZd*GE#C
zsZ|lboyw?%Rx8n?Gxh+^4C#peJ(*C0x4jPDYiR48DFtees_h;Fcf>uTU+*0HY<lD{
zJcN&41HK!bqXxcPP+bTXRU9MpdS*SJg~;0E(0ZfQeh-@(hQ=HN<)8EQ8;5~T3YOjO
zJ;e`o5;RBNw5_H%5uB8uuSKqdwyUa`Qt_*j*>W}uQ3x!-k%FT+PaWW5lnBavpWf+!
z6(83qrStN-TVs}*?BrsA6c^SJ$h821M~C-!8%-bLD`O0d((Vvt2o73hyWRjb--Q&I
zPc!So8Seze;{?emEh=%fNiGHR$LoX=-`4wmv%w8w3dVB^mKx)z0d02`b~*%p1e){Y
zUX_XOEgZvJ_IY$F4lzD>FNNElx$T=JbeUDMRm$~+&cNki2TNttMky-H=}iZFjFDxT
z)K)kp_Ogq2i;45;15b}XnmYWVhbJb|lyvYq!J9P~@^+(SAq392xG}nUd%@9}JoK`2
zxxUeBk><duzx!G@^7WOTZw$*w-$Ev`GRMT46vLT%WVFKNcIDC?x9gQrE5TM6r(93A
z+m&11IGt7+FfAt@A3w6J51h_t>L`5u_Kj(tS;E5kyz=<)z>+4y9Qk#7;qU*`|BA1T
zK26MSaQ339e);x=?Q(%(nuvVm>^)CUPyE&6e}xa8pMU-vZlkmR`is#EgNQ7*egMtU
zOT#BcJ(pH39q$l5nHY_3L0y$WS5QsRrG@L-N)1WFHO2o1WT|!UX>r?@xm<SgHfX!G
zV4j06a<g)U;9x}jsM0n$hg~7!Hb-Rmf$NGQ)&-~RfoAaPFwtADiW!bxk+E(_h!@Z$
zt9sYMI?p_uAAmti(IEEzuOVJ18pRuA+=kLVyLb4>!w8HJRjzep=rioVHCoMi+q$?9
zD`{F-ma{3*6<r+M;VTb!xc^8{ICyG!zx0I&_<Owb_s8%(bO(J8@1Am_8II5x=Ls>2
zv-aYqTWk9KIva+3$8{zB`QY)b5s(IBih*endKldx^hOxvj!qoT5kbi+UWzt^gjwi7
zw8-V7sA>%3o`>1#7`PN!CgqOAIN4=6w81(<aqllO8U{izmjL)_xeM|_NCdCS`{wlw
zt;@t173Vdu9Znc+$AM{@nHMcgwHE2dl5+U_7=&b0nh9=YF|3M~UnbJ&k+_@*QF$Yy
z6>{0=W)t?&n?l3hr+ez827-6wP0vHO!8t`~+*%v-NfuW*b~@nkaU#t@nwy$&J<G<<
z06f7BWf7QrP4()sG_I$UDBs5?=Jh1EQ}5)Q8NJ}~_!QsiW2B4ubHRb%HVo{Q>@vh|
zK3F|4wKhuA%eJJUDK$tJAj^;iHbl=7^+I5io*S4b!zsg`K}hn!9swVRFb4;vb*@{H
z&GAr>nwTZcfm{<?(FMtn1t<5)W#1UVar9r$^MoHxE`%!Xe7|0~T{mL%eERgndRkc4
zne%y7x>u9<P8r~eS}wKeucG+h<ecnkcJRVrnkRxAB(ETz(fa+(zC(u&Qiwb(s%KUb
zml7P0*06AgrwqDv0i3?4Jquaz_IkQ8xb0WC1j<!|L7F1-JPSgzOacjzqQZ{6Rf;;j
zSJR}Z-iL=L9?l<G<_BC1I6pB)vv^oX>xFH<u<cisrrb7emn%6J=A;9r<ZP~jLajT&
zi@)7(7wYIFgK@)fg4CNVp4;1%Z(lysM`wN(9kJz|n3CxXZfv(J$<NGb=IQ*%<?$n5
z*3~R(o$J@HeEH=kIW<lT-UrgW@ZrNVAD^CxDf038gDR*?<9u2<EsLT~YvaTE$hU`&
z?3ZtRe*MayLgExA=6T}RU;o5U|MHL2+m)A>&%A6`eu#-r&!6~<Pd~6s6CdJ<x9!S*
z|Nr@a^SZs^y5qLrwB_%rV4dg0)U?WUn#4xscs`DBsCN#!R&kZS9rd(22xADuIno@Y
z)|padjHp5;C&m2~6Tw$X*~yzaGGf#q(MPv_c?UyVbiff@G;x%Iv%W{cLbE;q<j3yF
zmZio^ZHU7zvm-F4IH{%YTXK+K)Zy=19|y$X=q^F<l>27L9bR|66Rks}wQ6X9?jU8z
z8)S~CT_NqJ21kGIk@l$C_Ul|Xszng496ay*u|pfSkHP37+u-f>)y6pPnuL48+`&Pc
zw||W9&Mz>!28dGYAKIPPnl8Mqpq>b#O@jo8At<)n$AFI$BLqTlYHn?fR&In=^nN;Z
z9yqU!Jf`%oh8pi3DNGP_A&Qdi7VBKiIpIjliS^-$6emI86!ovTDEK}lrtfkO99ZqM
zC!j^@PWg^;w={cqBr7`RDFTx;5blt+IN~%hFN^31#vT1Snm?x)g$7%cv2T%Lev>H%
zT$+f>i4kSP9Q%grm6C5Tlycip-+-Ck?Mu&0uro9GD$3P){IKsMN`M`LCk99J*ifqT
zI>ebcCz35t)XI@-e|OVx2nqfDGapmr^!$<a@gwP^vrR}}p^T2FF;9!37?w9PxFZ*T
z&ev~?BhHnbO!wBD6f{0Jpj18By*6r43RGzgH%yH@I;SaeS`uS)wkqXpH%@PgGb1R9
z;hu08gDyo(AlE8wK(39~%T4i|w&gqTDNS~++B@e}6~DtfZlw@!8<oMjNGa{dAf!ld
z;|`zR_Cl_e-~!X^S<ee0>40{YNL5R{tBdzNbGvPPdwF5I>@3qnSoFphgFzjRWuBO#
zG7Wl@W8%(L>MSwH*mIf}X0x)jAy0K1f-4Ws?$|99a@57M^^PC1$~mXV&u_0UY^^Yy
zl)I<XV)2_QJBlH&OhN&2R)+s@0~bZhOm&i4H7=yM;)A@&_za`bOQx30Uanl<zHxp1
zO5Sd4`_AR{CJn+W-)$d_meq#V>Q3!ys=Hpkak;(0P{w_$g;9zQm{Qm-uYCRdCu-Yp
zOJJQ69&K_j*9$M#H|T@kJwLOind^4r^XFff*Rw=Mqf;+$eEsz&P7i0+;ED6haE|qS
z;`#9`t61>7-8M=omSx~EFpBeX=EL~|HOyRV;q#YY`SIbIw>Qt{pMT=(moKoZb@EaQ
z)3WfxhYx&se9$ia@WhWFexUTh&)<H*)vmP4?uaObvREE@2vRc_Lk4`3Uq6>DzVr})
zRVj?CDXN=oZ_YR7A*i&bO6!q6WXE*QsU*q`mf4f%U^yQ83{A7*Y%x{#=uLzA5g-<H
zwDex}=h_99EIXqQ1w}CiuBqNTdPj?bko$NqY@I?RcU`{02jVOezT};ntNGPCQ!)<U
zImPqZ#n~+!2rzzL!ZW&+^NjD(0SyK}L~HMG?63}mM&7To<vHGc-If-olsd}?7;=QT
z!27T9_Un%S?$B(FVT*z2pZfwNa^A{lyqr7VB`=OK?$G*C&UIKFd#{WlaapL16r#5E
zDqVy_2#h}45!VCNHC!3EUga7Hx+vo`37M&t9n6gpEa5LQvB*_08l`vk+YQ$maXB%c
zA4&5`e0R^Z(iycfylB~U7yaJh5RNp`7=v}2=p5^%(3=W`XpMlZ#?5=G56qL$%IFhQ
zN=z0@tt^2~uT|EuVL1#wG9@*d_F+4B?0AbA_95adM<G|)PFvQW?L#2CBs$Bf-Sp6M
z$eQl&Lt2a;dxYQJ9{_h0>ph0~Kv#sw7=@9Nl^l66SJrU)>u`$%?`?pOggD`Sywet~
z$+49j#N;h0(LSHmlY%sB4F0UI-!u&5+GwM&537ER9zW^Au4atwRlz-@>6_@aF-m2P
zk*D*_GDY?d-}c7WTb2XDM@|TqOWrfZDD`woR5vIC#wfHZJ4oGjdadla*z~PaJEY2O
zD_r-^?c6z^X3i&vOP#q7eC;ggq6_oPe%tAFsQpX+AE`Z)Hj>c?rFKGVw6U82q)}R9
z-*&d!#^v?O<=cf;8&ilX;&2v*Gdd|+&7@2#xTDBJCz=$tP9nxB1^HM#cBnP;h;uI;
z!jh%1#Hrr9db!(Sbwb0)Ub)>i>Oe4DipJe4EGQ*~u`H7ifDk;nHyZLxs{_5MrEv_>
zvheWmKuWWCx2nyol%3n{O4%=rve9c{+jnXmjCP<IK~Qy9^=jMoywqOE+lALJzmTVy
zYq@c`ep4Ay4B~0?PTj9u-yHGl%H#RWx~zm;_<DUI*UEW0F~<qtVNSB74d*de4t2|v
zeaD+!%}o<&R(jzyA6%=aj80l6&ZmWSl6v`40tBa`s@gftXI@?|(2f}FnKnG5czO?<
zAD?-Cdgi>YB$XUcoz&NV@#!zv!o=5Kf1#C&E`j2VzUmBDX%PmMJ~T+upyWcQl9os^
z6>_VMya|m`0;~|CV7_(d2Qf&)P^wz|jy`~8N~-maQ5)D2Ng=RKkschq6~@p7ADtr|
zao5gS{A^%HPJwY$!}e<1xNKXrcfWJ=+N5f8_nL4~8Y*~kUNzhKdR2sYRU;w&2%-{w
z*lf?y5Mm9C`1{ta)HQv+O3@}8QnV`JW`lscZ}FIRptr{Da>W@fP@HE@I;)Px!1wCi
zBah*I<3A#x_0Qh_Js4Hu4#@UhNO;sl-w_pt3DFcQeSBuT_p&tG_uNXLb%`@Y%S49L
zkd;d#Z#QDAOb;gxE229tc|0K{mU#v*?L}wM+s^2QEVYPR7D9{=CH@@)N{4FwGHE&y
z4e`^nszlF{^%{m4U|w}$rg%qV@4-;k{m@ObQ0hi48?9c|{EE5qZK#ojG1_}F+i-@E
zl9s9ErZmY>wF8d>;2&0UB@5*-pn&M5(K@~C)O?}kVuKwpJbnZ@E4+IbGori3`|ska
zO_Sn`PB}7T-t!NJA<31zZ?uvb-NdHI0f{|FeZ0S4KIkvq9mk*E=q*E5#%61U(TZMU
z4xNgSiRKHwK_8VrcN+L@$bj+N&9nfCph(v!<zkb8*;J{T+(v7pK6rdscs#GfIY{tV
z8`oW32CWOie0-P{(HR<Hh%;;*Y9F+Nzqwm^3?8jbB_p<C`xw}=p!>lf%$~EYc=})m
zqBD5U4wc?4pT5Wv6H=sgXA1rXdm)B#ydU&p%H(caSv{;u7KM7ikfSq(Bkx%*j6SSL
zL>KU2P#Nr{a?2SvhK7vmPA71yg*|5}VeKHf3lzs;kx@>=q24|E2BfdhAQMyMbY4_&
z5u*iXNYoc%yoaCAD%++&mnq39QM-jI*`GP81eW{0aogXxU0&JuH%i$#($Nl_)`Jk3
z=t<ELy&B<M>5PX*M$T-P7cOs~X;b3m?S*gOzA>tJY=-!5TA`JVZM*XF_R9J6Kn#gq
zs(4z{eKboCELIQQOE)fb;{5c1^GU)R7bl*c9++)F8g9_MBc#C7y7Dl~yIai)bTfz+
z^XM<@nPq+8>FJ3P9skqc{x5v~+keK*XMX>e|A9~EXM7tH(Y4MmZ*Odu8xL{e;p6X^
zd|<zAT&_1fo;gIEkJ`XS20I9kWVKq#yTW{mbVHuDt3|7t;<<N&=*Pgvz&v$&l(Qq_
z2n6Oi3I&KdXruR>PKhZ79_9p1g=7aDD$lcq1)6O)%|6>>SbeOZ(AqkCtrnEiITG1|
zL-4GZW~CHoqQ*(VWwzzy+naI>yqYClPy(Z@gN`~ZcGbE20U8i3S4`wcYj?l*2tojY
z6VL28+zf`V+3f7bptr(yePfzux>xB>?-DVG;O|yNk0Uy)zreixF<HsYcY}uh1gRbl
zW&Z)!j&aQF_f*5<>$qT!i!+2sZ~^a|#F1Y5qcl2m2V7+oAqKs5Zksmj=pnchb}}<O
zcpH?HRYKR^RFqZ~X5_r5j>@2sCfQ>TWFo|w>2&;D)(bN-t!DPyRZ+jaFh*zJyrBY+
zh<DZdd#1mI67^wR{tjLC({76<mZgzm#C7OCXrcK&$)`6)$xwEDFQ!?*99o4C`*)&4
zbCDdc_2`V28D*yzqnEIAymtjF9lz9Q3}^8Boi9TE@!AZzwe7!@!k!DcHlfH>vGjSn
zuw7o66T~$Vx`r27B6Y4G{eSxGDCeaQN2`^w?-YN5I7ztMB^Y$J%)Jk&vMx45jsHww
z;I|C}qsu*zg09kt{N2qtdaJY|-XmIx&}fxWg$k@Gv92ek$>GY4(mR);w!&N#EnA0R
z!MzSr>!e&5DN)8?ugxyKPH=%>;Xb{pfJd>kKsA9;@WjdS_&Bq~#Ce)I%?nB4^n%tc
zcKC>7Ek#8n?rzx=7vJy%R|Wg@1CHYk-L)}H?e1CHtWCZTYEd>oaGq$FHi+{&EF`s5
zZreuAotAHW{`Shl`NT>mn>eHbE(G<f+vXF7Q`WfkA?&%Z?U_=VHoYk^PqP~RPBZhI
z2*H^@J;;t`Uhq1U=y%<A_MG)+HCgQjk?3`JDCA?Xl^fsQzVQ0(7jBnt^tR)A#&u;D
z97=9Mk<p;NiPpO`3nRcVdg*%M^DqC*G);W{`i(f9IIqtvOQ#iuA@y37c;_6h%ke;$
z4KL!^ZyR6yH->|6FJJlc<rnI96RhnWPai(=`23M|IdMKc@bQNq`SAE45vMaIa12gq
z;$fbYT<O&r5l|V^>-CjTJ2@A^JoEI!NB;W9-*Gyv{CEG4|0n;;|M)-gyMOl|`0G!9
zV2KK;5}md2)8}9K_2-`{Sq+EnRGFulnhR-~_4hU7Vw4so?{+w-Ag1(&Zyvg+rj#a9
znAu0=cD<5PASN%)G$1Of%Borl)KS&A7bZYyo^zV;IEfgM%VUlKi~YJ=M}yfD14=UN
z)$5Y9;-pd=A!NLBT=)Gh*gFpO;1tUmQFLv#xZ{|VJeT*vwqNbg%yc`nT~uDYC)_XB
z<UNUi4@u%#Hy{_VHo)2|Vf6dqcZlW=<-J3QmnK_rs(!pvb>ca65&UbGRW`p6^g4KV
z2)?*e1b4*2bWrcVDVU+2BQsu<1>nCI5?cRBj7pAlN0$G=wZH2Q7|8^Y0o<;Pf8V7e
zV9-Y+Z&zYKsNaFm*y|pT2<W0-n7ekt@)t%e<kksMSq{g7<D575g}CzGDmC9I`%R%l
zP73@n3VXibg9@(H-X~$5v$R2aqnDye;kF66vTa@nkCrQ}C=PSfG=#h|hB8oUeINdu
zy9@e8E0k7n2gTx!Y=$FQQG<|jGo0LO-~Ae}ZQssb?A$k3=lgSIOJB7F-QFrkhehw>
zUU21npq9*byAY$t_eAoY<ldJ!{Y}&BJru|kqbC>9XKC9==`5{Sm#O1koW${*>z`wT
zQSbj3Ap^-aPS?esjW!sy6MPiBQ)S<ANf5gfy`A%tcvvR02Z}4HK9HSfZ%})sU$r66
zQ^G}mzrg0x!Zatv>jK59a$`{OP-~6U1I~h=71fJ8uUS@~r{%<Xnu$(PmN!$zZaI@{
zW!p1vZyWnAtD=iX-u9s7MlHKw8}GovT6$Lw_ntF*uH?Nk&y8aF@O|jfE>(i8p0ge>
z9~dD)JSI*}rNQ{e(kgq|al2>lom@MF2(4*@Z4K8|?$TRl<VL>j+-_N$Xe}&ZVm&KH
zIYbRNgJG$MM+9q9L5tJol56GL>no*prnJ(^U<|YUq?vuck(U$0Io`GxE|)L7eEWrc
zdzJXnJ6u0h$J%&Oiuh=Mg+R9C$2cY2xZ&EM*39MG7p@`jddob;6TwT@(H|=>FTdiA
zT6tJcJS}Ia7T{D1WxKKOJD1BFTi$qm`NID73!!(`WyS^1x}J$?Vwx77pFi;Qa8^dT
z6YSTjLZcS5VS4ja_tDAw#`W^Xzx?r!tky2~SAX?a{P?>cd7Mwwt?<{s`%BV#;{5mw
zo4Ht`HjL6Lm)nKU-@X#3h10_`m&=<}#nWOmiV%}JRB}~oGgQfHZ^vMswVnCy2z|j1
zPa?B5Pi~dgL^p-th^|xiN|{W&7gWwQ1X#(wxo2AI<h_$i=O|!tZs4Nmsx$@SB>%s-
z$U)7HTHIZ@6+^~5(F9wugQ$-C;W>sSqoLSZmptp!N7A&gUv7;4LM|Ir8z7uXsfM?b
z7M;FvO5h_dO!(zQ0k*tLxYmy_BTH&L4sj<h`g`J^(9PriJuZ=Ba}}TOk8@NP=CLez
zQ}>(T@IWSx`rCWQy?9GQY~DHJgElyR&qK`lz4cIGI_3t!mfzm-LBpCZ)Ic}u?-&|b
zj}j$tmQx@B+mw{z8m-nh#Yl8P1JV5lJkUh$ygpr79VKT%j5LE#guES**n?JfTDj7j
zZRPiPK@AMGX0jEPdCNzTo1`<^PAwa~?1H|#Lf{5R)wAN%yH7eloCMdBR#dAp);4|w
z0`<^v4GoFIsF!Y*W`F2D?Af(%)jJ7VWs(@?`8P_yd((_LIevx>s{Wt~{QD4zq0D#Z
zgM^pIdp(BEx^24&k#mU=We@g-q$=LH8=Y+5oAXrH&6;kR64QD|1B1hA;vt4X3f~nb
zc?j;$#RYyVO<-;sHylUmmKzRN8nJf#7^F$134IK<-U$q*;8|x0)@rZh-YE`xaDslC
ztcQIJwpJl}Qm?FK=Gnn%ip1#I>6FqLzA?>#FzE%yb*83ww<BNh92}=9@^qRxEm2wT
zmg2YdLMxeV-)Xh0Z07CC?Y1-BQ?LfNt>8qu6MV!UFRtsF#_hnX2G7<uxNbYkNg3fa
z3(}MFd5E8@>IsiAaMV-==Xh8X>y-H2^CO=gPWWkpQrYURJo(X4qMQy4B{zznzuh*r
zJqsDAl_&3r=|~6@zc|_J5|X#_mP!8UVao8<$a`Vm3*$+fWo{kf#A#ZgRj${s>@{=w
z_6u#h5RgCHIf)nLVb_G~gJ)W1X5$}~{?^%YH7!Qv^(w7JX_iwk>;d-n!StS5GQJ0<
z0Upn&jqCM|9voj^KlAqXl~xq#yx%VLoT;~so*QGRCG5K2czJyxq(G~MeA_tB6T#1x
z1)&GKkItS8F?e0%RqyS$w-^55pZ<a0KY!%u#~=CO;RB~I(fuG!k?bmN%l!KKjWGw-
zWoGe_>WA7DPpgT2w4sN2#!m|^SDw!&mN~KS(DH`&i6x!L`_ApUkxRvS$HSC3rxj8p
zhMBfk;)wVWDW%=FyB;m~gNh#;%c4X$9|OKu_TK43G1|R2-o9jRmtvw)$FRdQ#h~1Q
zluWc}7n3b^4rspOT%`AcbJm$?yeGqGi>AxbN5{aktdGpkKj8Wx?Sm<9+)UvvPq{b7
zP4t`V@^8Byn5V$9o^jL4wOV56adLN4qk~a`2RGmE>UNwU{(H0B==Un-*55;zoM|KO
zo4mJHL)+5tjfJ>Fc3|JcJIdhPcf2zEn+xAUn0U{D(CstsF!<nIBcv3{c)x&RjM5Px
z1QV|q<+JmqY_u9{Z*|T?nf1|2=|3E_qG;&RY!gM7^%0}njS{Y-K}cFYhIA>-sMX%e
zyCZ5i7{*(pbi6TO-!{2~g4r#d+F9MGbU}YgJgn8k+*VjsC4wEqcZ6je!qEma`?H2w
z4gnYFRIB*KDA9={p)kdWiz*EYUhn_jWYg?upnMOCa>OYTEb*^V3H@Ez8RJoLZnjeI
z(bT5rKpKVPj!>*{DVgnhA%?)59J6K39G|UW$jE&V2!R?DVs{Mc(>yb;C+1m=y^MzI
zjVXF2FLducm~YU-dn)FC>Fc-A1PrNT2UIM2v1^)2F5r{nw8)=dyOY1THf>u`xLnDN
zw{0ia%A69-1*v0eVK~T57i@E$5Igw5vP=pkDvoRRl<F9zS_Eg*!`=ktb>r)24@)%b
zn`r_%@8uTPo9cB(>+H9kJy%K{DqXT*lWsyR8B8QVl7+A_PqXYFrbPE$JZ-dus^DOm
zC;U(Xp+1k5;Rggoj5c#J#1xrh;4eNt@;ATx#Pibw6;Hixyxnb!aFh}qn@qDDgFP3s
zA5|U7{-Cw&K(Q0c2erf<uyTmz?AUM8N?7nnKnP3gl~ECghNITPetVPH23VJcsYUj(
zTZoJ<nA8JPh@94ic~(SfF6y6nyIkp|GIHT^z0yh}IgguQD~0QJ<M}~@<#oSu+jgq+
zgmtIx8@Xj%50sMm`sEkiUccy2swzM_CjWtk6ITZ>U%v3uPk-ds&%da2$b06~O4xRy
zbL86DOJ%Q_K2PF{M`zm$ueTf5*EfQ*vYlXCa;wztd8?U!`O_b{zP@n1UYSlO*^ZL4
z%4^5d>4Ar*CqBQu(Sh^%2|uqqhQz=B!@q;k_}hQ_J7Vno#ls_c-}sWhu(h4GcZKbA
zPZ){lXM#@@FPq=K?WSZ7(yv4{0=nQBW#HOOShX!xXBEHB`T>fgZk1lNjYQ`lYUqtA
zSe}NSZ=53-VLHZ8mVQ?GkfL(AZ@~Q+1Z<OIsLI@>!20+^OcT8*O><ZrceMt%Zis_o
zYn|w{raXfwCepl6eZse+PF9W~{IR5H=mq`E&OQtdNO4jwhUe~!k7m<VBzZ?oixy=(
zDy<KrI1<j14}){KamfDk`Ep$p&D|dMZ~nqA|GUl50nRScBjL_2A$?@^&!q-$Ff@=U
z%ilACmF>7hm?ykXY6EoP{Y<i-qt&Y5rfxO!L*mTtrDYgZ%ZhW(GDnVUNi&KDz(Xx>
zITOy~T~w8HtvYKQXw1<d(CL&`XuC3CLiDEB>iE%ctujVczChDb(yW|Im-$SX&PrAs
zM{V@Ge#v`>??bPR_xOJFKq#DNnpaYqaH!}lCNJA29;QSx&CKm~qwGo{#Dfc-Fe&^k
zL{Ty-r)#<of`rireG!6|G-p=SJ_1wH30!NXHfNS{S***x(;sGhm<6T`E6Fls$JydH
zLZ_u+cd3Sz6dW;4EVJE-uH#1~(3yj0o`NAlmQY&$FUlD>gkjG7oC7HBfcM<}0$L_D
zVBKQSgE;JiX_*9<Toe)#6E1d0kzS&r8>?Va=jFAQNx|WRCk_3Oj()H8#Ur6hz?8Lm
z;ieP`B2NVOmV7J_D&px5*P!i%a@#3al@0aK8D4OvKNP6O$CO$rrAW={oi(+=!#d*@
z$FfeO6!rU<4k5*)%2+H`QY(Sl>c+6lf5)=Se0YB1)5njTP7AkEak&`e$ejWS7HV?T
zU5YzX0^Ae=DLEdNnR!b0jU4AJ+{2Tm#55@^NEOUpG><)qQ%ZO@OoiI0xo~^?ObQU^
ziQDZ;Zw==?$tTMw7~<!gwXPur`V^zVgWVbxRk7Slp<FlieP@Z2W#R{1tz0i}T-I;I
z5V&41+_p@cB1_YTo=#`dJW=-D_*+NGg;ol!w0k9P8{LK@`1<t=?H~V+r@#F~ahd1x
ziJNcywf{Xh$2J=O{Ohl@Quy7|GsT2C+jZyV>sNg1e0+RjIh|F6lM-9WbPu1ueBtHu
z7hE5FczlGW?IOlNw88ZG>4~(geE8u<K7IH^$vck^&sP4Dh;ilV!zcF3H_pq#5@%ZO
zTrV5NtMReqj&~{#3PBE!RvTq2I#9Yozf@-p!#m~Dx53mrarVs1#0Y~JrSNZdd216f
zzq&QvX$S~+Yo2rdo^@~p*R-Y{5?v774;|TIhjp&&(rU%CJgiLXqNv|iC^b{6QU-(b
zbb~{WOHa<5VDN3=;q-u?7p~>Kfi8#}A-IXy1EE*$i=tC3tNsc~%1l9u{TPXw>TGoE
zb`c+&`oIJXqP(Wr3bRZ%VE>Jd#Y0u5XaD~wb<lf%rvmkB`W-cN$9iakELN1{5f6Uv
zFt`hIElABt04!cS;e0ZlKm2PXqN(0pB174g-O%MyiYkYX^XmAshC|^uA<B6$bkT+w
zg(4gah7<ZWTD4)0`@bJu6sYgbRQdt$b#8T2`3qr?9kdlaH)A+_T$oNz%=20IjMa;C
zXgr2{M+FFaABU(M7d>fSELuI{Mg@Za*+3@0*l{pIyo>E{77AqAEANAJ71L}H*&6)p
zCXuZ<sH`Y>Jzu5D`QuD?5S^Ao3{5!%wZmXF@%lmEt4I0|{9QM2d~Y!ZA?_AG-|`0e
z${ai~BrDr;1VqJ6Au!JqOHy|YKs9>$&-waop+iL#ykOQA?{FR3(1%wSoyuxdfCF^#
z8dC^FCqyBSp#qvoCx{y`YGc3Nq*Ar*y_Ujl%WPZ5)7kPaXs}zP=}<np_u_rlHR=I(
zYJI{<TzJbHwc93vrENK>pYwI&_I6{`PI8_lCDwIjUS>|G6H`j`*4VOWh29Lx?46V%
z^OVeIErl_CP)5BAO_H-B5E^kLdl0O#uv_p65`xUrOq?b{h>W5x?chZ@jMkLc=$!F;
z?+0D<p5*mGJI#^DhlOQHG%qwJq(Cq&2L9h>r4PK1_FYJ1<gHAE(P*`>-@b8i9Y4v(
zKgM8LXJVR&G0|OMTrOm{=`iz$bK|fN3GK-_3wj+ElH)pabPA!eZ6)7sT;IOpePFxY
zI2tIG#8$tn-5s1xXSUlb#i_f(2bEHd*4XCx4!QPRc>DaBfBsMZGp@d}JU;W^{&)W`
zeq5he2vi!E*B8Ei{><ywFMRy)k%#rfe%tuzr$2Fh`G#+u$J3e9=|KT3y)m|pUMsT|
zPH8|h4t7Y>!st9atSqMo9@EOF=MVhy%g?N*6PLGF(M$b^#f3gRX*zK}cRqb+EYp>5
zZ!eUs&}t_T2re>e=k41YWp9K4E?I8-P@H!evVY$8LK~5(na@0do3u^$>hLbu!J)&<
z4cp4^Joqu>>oPk2{=M7X-}zVPq(7;(-l?rx3n4CoHqVa?7gULRy|CZjC{=+nF@$@!
zoHy9{xYXzM#Pg>gIiDZdho|?VP^w|=gj2$I2({q5-gcRPEXd{l{c?w4(?KH~VM5;D
zX@Udgkf%Jct`Ds1+3q28Bsecek2_lYxR)AzKL~u^O%VJ6*&1hm=ZW8ctp@*LlhymL
zSK^_LRPXd2J3~I4`Om#Y^m=P190x*UKl2#g27pYj5$}s>B(%W}fNt^aR&7iOa@8Ec
zXVxJ>vrCTp{^pNykI!v&;F?oDKnNtloyOJAFb+UOXhpL+VjMor$}(6~pWTP_Aic|$
z>Ozq7z@n57UOoi%90Ws<LYy>cb@BP57C<yDP8yVuAWh1#=`Jg0AW1kY8UUdYt>7%Y
zN)~u?j~qG#8}f!foFE6n(r7Kpjnn`Ry@5pq>kYU6z8i9+fyb-XK<_iqw+s8rFPxr#
zggNQ6I;<g+L9(E!6m`ErC;5AZ;(zh$H);ly-z+<j9_Hwk7Fc_w#lY)c^r3bGT4R|e
z&hyOUvJmG8*Bd=3SxFa<CZZ7|wM&&bigF5C8|1z4`g*05$ewp{t`_v5Z1CU*^XysY
ziKmAZ7`$G0_N=WW*PYvblf5qll8@@Hw_(1N#<pe3USTwrDR7z>)_Gx_W=gB%TFoY+
zI9Ts=dFEWX+;(2K8%s*KA{TTm{SK{9)*h$epsp-><Nk>9#fJ`4iss4nQrWkrqLApB
zf}`dkq2-YfXAgfg+f!gzqhDk_&796BHdC^uDe3n&7s2s#93P}^AM|F)S%+n6Yn@&c
zv75I_Oo}Lu(`ppJ%(9&Cyj<Bf`FM}*!dmKDt<<4>gefJuhh7R(4dT68W$%@J%Z%>H
z>xq~YQ=3cC<+2w$Z#uQ)d#Q@fNZY=m0+dGeoYs|jo^{b8I*YtfzI`D%$IItm`Ssxw
zbBsb6Zqkwb`s)kZ^^F)4+jima|M4Hl*Bj44iG)2DTpO&*qGZ0Lw74+_U%q|e_4dYI
z3igr`oMWD5K0ZBhYK~=Ic>3^x`14F`g~v}HD0}Ah?LxU-@gXs<5kJj5KYwJMCU8oV
z+-^74Wu@lIe%sk@JKL?Wtcf&vj?$qJ99E1bp0#x9=*Ym7D6P{+!=VE&#CY8BI6Ox~
zs1E9`gUY>cEJsg*mRS=>+5h#??i(m$5W>v)`3Kh111`<7I&C{6?@~Iup=YKmc``*q
z0^A@vNHOvJ;YU9F_<LMT+_p^ZO2QLm=81#51H^R`IQMt)mPgxNuT|~@Z}fwc!=WET
zoMj+rQ<EEJUC*qiGx%`V5jd}d|0skx&M_apqXGWS5q#toNRJ?%ewgU*5H23+l83tQ
zu!wa8@oVttwy8Qp3;mrMh?ez#2p11G0t`kyL~8bDnV!PN7PNcXYm#RCB>UO`_Oi3r
z%$|2T5Yhx*?|E&qH9NCBcf`fh%zsa1#7OdTS&iltBY$KMc-5*8*BCw$d?d_g;=CFv
zbEI81dMkwcYsp=g&_^@4-r=HB6@8Qg%C(~A#U=!bX0J~=hPGf4!2=hPE@5x&jE1_7
zlA%|;AyJB)mNoZ?pFV~=JUAzcEGAI~^OP99%RcVj1B~>PY4Fo$D{#DO2kPX$`)kyl
z+w0GyFHZ^sdpes4l{+fo83Z7DEF$^;A_3uEDn7(}xrf6^uoU#-FdTdBfXW^yjgJo}
zKAu*VWu`LNTIbd}eMofSV*u}@pbk-x)A0f{L0oT_8`~b#O4l3l@W}gULbJg%1x`!S
zVd4U<%dc!AS)$kUD=tl&$cPhtz`iT<;IOp}3n-~o;;lK&N>rR8+qT&R>bTu1m+P+j
zW!D(K;%cXtMsEh8r9lWCXA_zwF_VHO!+Dx;y|YXUG0pfWANgK8y}3K&CWJ}$#}tF{
zcG^%d4|YSbl_@3To}{C2hj1{(J1Q}Tse4_}lpdmez_EB+P3pJXg;pVki8#&F-gunV
zBDk&(tn<m7Q;o|q@xuFu0uuweQAI(_6N5qW6AyD_qEl}>x3Vi)a?gyRORblt>dq>0
z=Duf@LLojJ4|&_^z1;oF!xpYO2F-O*MhNbVF%fW<YIx!2pZ<yU;aRn>TP1Iq>)Q*z
z{QOTmozG0u%yzr+%Rm1U-(FrAfAQCZw}m(=|33bEraAFEFTAxM$;%?8@la00Th4rF
zjj1|D8|?eer}a!|ohePUZ<V@jtRwL4=g-9G871@W%U5a}{4h^EKRz+m2l{@}HuX(Z
z%expTuw_fJ!;^fZ^sM)^bqg)=+EzRC3T+=$DyrLw$6Iti$drud?gcD|fbIyjIWCw%
zGg!9KEJ?3-s7)bV)zaW_k;jie^6=^RjIKK3(Kq^jW0Z_DabBF{C3oF6VQb4vQ{w#i
z#KZZC;3b-C!<s~!vfkaHT?&!p<SObl6I{3-ybPfg+Pr4Mq~uWzD~FBHy&vjbaPd4X
zEbB_}QO>EM2yI#;ng}?=Nu47haAb)cg=FuS`+>W;yI>4Ve0W!q+X&#jD1`g_Ed7jr
zZB-D?{WtOU`-&4kf@+8s5Y?1(;=P9u%-3J-z0+oYM7nz~mw`_cE+ufDhQf39nt9td
z_MEA$Fx5y56TKBR{#hqW@EWc_TBwetM2=4N5Rr(}1yAc%)oUKPZrgfD6EQ6~yu9_u
z8P<w94(#{$$i`R&$$*ddt=@-Z^l62b?O9Xk)@Y4(d)4p~Phc98n3B#vqg7oWFbac;
zs}*kpBOdw)L?1=V#z<|$hBQMW15^jY1`qFr5_S?Daro8xXp&d(kZhtFmL!>Q_KZ}G
zX<fRHvhU>W8~J)-nNXT)@K~s}G(kgJo&yE@bH9GmFwmQLu5J*~5u@k~CaO&b_%Ma2
zYyB=X3rVWy+8tXiT=z;Tm1&*vNp-pVt~y=Dxbu~&C@$>VrU@QMYhnsXg2C3sn?#8e
zy#?xo;Kc2?pi3o0TJ0n^?t%?9c`1P`7x^)r4-D_vN+B#0OPWaL->p+*n-bT3W8XHe
z*Ntu8%pT})sNf@bhwtL8kHC>IxI_GRIPlow?pC2;l#8Zk9|B=`oNI(8k)*aJf4{J|
zJ#)*2TPY;x)a7myt2Y?+uBJ8U3k!O2#)KVFsxbsbvu>H_JUJJxm#aRQu|e?oXyHjk
zYJG<z`Y6g|PB{M#X~rqu%T!EM&N5Fd5qgAjnfUs4Wxr*bwOWQaQFEbmZNB@ygLCYy
z7@s)Af703a8?6^(Ds@0A^S|+Sd*ixYa1N#v6fgvWFZayL=b!m``FqwN*!}J8g|EN<
zk;}_xVlTMmM1E7%_4Vs#{_f?KFQ5O!AO4&FmYvQr&zzT;zkL44@Bi=z;?tQ=%L9{6
z3${HI9BjGr$1k6Gdwu0$TIdNb*Q@BHEz`=zyd=tXSJGf<EYpb}KK>qh=iBE$-rWt`
zw)6IO;e0;b>9R3~MK@RbybS#8l{j~(<a<{}0rH|d558M@P@D}?v`vVC??U$aIMUS)
z6!;Ld83Z~Mx!s!iue&o*ED$0O&!2es_&efsrsf;%x>NH`zFw_Mp%d-d@F5ZGuy#07
zS~;K3tfxo3i<Dg1M<?%vyzkQA*f%C8gA-jcJTXn8ZcMY0_nlFzp?ZR-qv>P_CdwNM
zcZosS6Vp7iF6*5V95#gXx-(j1iivnSG1828(iYjP$YCi{|G+^BfJOG{d$yp1G3@LZ
zR(50C=)JGN9Y}@IPlsr+zyE$?NO&2Bog2q4>8#4vQbp<3!BF)YmLlOXy!Rl4X~uiS
zm-jKKT@IVuzL^H3(zlFvn+X!7kx4O;EGD?sO6yto#o@Jg5bbOn;~i2D-hmJJr0f{*
zhDgX=5W1)o-w*riF!3gLO2!*c<$^Rz{m?{=3RT@7{y>79hQ-zg#D<$EQETIU6ZdEd
zt-a!CL_52W4W2tX;J?d2I^g)@_I)1(?-WKg8>rD|wM~o|(4eg>2Hof5PJ<sB0$uV7
zyBl5Z5JxAN1}H^$k550s?%>a-2;eun3GP4L{Rcic5AkhMxYI2(!KTkXI(x5Nww<r9
z7cQ3@xr}>kV@R6z+;`TX5IipGLOh>ViP|i|FHUOqJ07&w8!>xsInzkgp@VUXk%xI@
zO^Y~4ADMk%@&<`9?(liPi$wGW#3?e(iD^y}WDLi3+sNCETUO`wZ7-D8cz9U(;pvPI
zfl&u_%k*MXx6>vo2f&bcFDbLT8kas$3Z4ssj+((}E{YqY4qB=Az>-?TGnUf1+%lIf
zQ)**QlP1w1=r;`MDP-pff=GvlW`{@AM&*L#st>58v0ZmY8(gn9T7&Z`F-<emI#H^^
zlX6z$V2X<CrFOjU!~nTgiJL;a!(h5VIw9zCZUj?yJ91V+*Rk+Oi9V>clD7?l*U;5P
z<%o+``dROd)+;_>*2NpQZRhp!${39~sbO<n6}^iM6Xoq2Uw`_0Xd5xjT(=9~zWpNi
zN7*SjM%hVpCf9N0Mt*gC{`rr5%~z%%Cx&m0^SUyfW*+95IRsqq+^!p?R9-Gu{`Nor
zXSSCM?(rkTL#dh8DsQjfSbXGq{mS+FM)ZzpN}QH6Q<xZ`lUw7m?QB`O0IgL5gLR%+
z=9!nTFJi{MTpgt~Lh<-XaIiLiN9YQT83hRe`!E4lVbo>^+B>?^IjDq#x3r_~KoJll
zdIYQ-<tXA|XQpXkeR#$NS!`R$^qSe<-V{>fb$G|%mFkIE6P+WiCmuiks4|<7q%qkW
z*OGa?-R{-4-6_zhM{n8nkWyqy6H}U?Hw{g>$SXgjd~ZP1@~gGXP`ZOwm?nMyAw)Z4
z#21gM0IZHEyv@NfO^l(%Bpk?z7v=E%$$8Lmk`>sceh(uu%bX2d-(PUHHQaxv`@rF(
zh|n<57_HJn#W#PSJr0p2=mPEi2n%X<kvF3soI-a(!n+{dNAHxJ<yNv#H@(+QnhlHd
z9sOu8T?Dfaz}e~Atv7{1U`Tuy52r&g{ZIhH!Io>R?_KT_+Fg(voH`ezlh9dpXhI}*
zO@>^-=#|lm<rx&zSD=e~pyh|2X}QwdqK9F)_lyC1ov<teXNc(?&Oa#H6fLG3JzMs=
zOEfs{!}Bq;j{^ZZUdz_;e!5d*!zfL?XQr8P8ct&HjPAdCPldY9v-%dIbNA|OMrU**
zBKmio3IBz!-#iQ)FN!yx<uOr0r`ixuZNg1a;*nemx4g6E!ez_cuDe7X+2dNL#|R!;
zQQ?G=o-Fhx1Rap4nOqzDUU2~xB;`tO63O_z;w~F=@06xnWierq?|OrhcPz;>&w*Uk
zs-{ibsf@+Jk~~l6g^wSkT6f+n7oikht`|PPzLATP_O{xHG4bi~%nu))z<ajqP98`Z
zZT6~=5-y5!OxEa#(}yLOEZfgekWOyWOO#<sC_9|@s$ig$(p|}$OXap@Wdn3^j1vFI
zEl`2Q2bO7(3f%=oRJLwjYi|j2@=rGu#FTu%yQ=redFEx2m1UXlbV|uPZ*N~26a@tJ
zaxaMK%?|3JX9C+?K(Dn=k=9jyax0bFo=MY$PYE9b*)#^1tm2^AY3oYfll)OoQjD9H
z$6a-vN=4gl7q;ugf`lSH4wlKNEbj#Ujm{`HZZAJEyyx|HrPW5jQ?p|qUFAPV@}W^4
zJ@4#8D(U9oM&qCV_%F;A{Ap4+kavvXxW3-Fyu9-D^RN8+%g>zWiSl)$m5rA#zwnQL
z_jjx@GA&6Hewy%;r;jEF!N|CI<@tx-@zdweY$fA>WtwS=VD#=F*TUQ8$hD}P>wuh%
z|Mo*oajrq#&B9kZp+oeQnk(NillSyA+`%bx=ias8n)<wdf7o#!O41Za^O-&zCGYy2
zUN7u<qqb}tp)~<I1+UQbtjBrg>En+~A3j2^%8uXPln;`#pyQ#<xI2BWefR1cSpu_9
ztn*4t6QhgkO(8I9whPQQYabZF3pPj3zan%M0_DHc2(+#Ycm<9H3$W2(mh+8enQahq
z_k;fXuS1xKmnBY@<$IFp;Qf+tXn60#@(@e}4Brg`96AmC{IpCTO@VBsn8Myr1;e&`
zw_Jv%ZT>zG3=Qkv1)LK_Sz99)MWN@cc<*LfrEY~E!v^+YNr8Phs(|Br0%k3;4fZi8
zNaQv~BMguC5r@b5h6|B#PbpNeSL+Sm%N-^5{t!$b@VU|wgA4c+2{B0oST(HB<PK9r
zy0&Hv8F1cP+<B+xP1+3~E!j^{yOnF(BIE89??C_si=j?2;mI~6>1S<*@V2gK;(9!@
z96G4XRI6UU19mef(6G<i@#}SDIf9o+TA@t+{rhlH--lCO`CTJm_sPF1Q2H-?{kCBs
zIPnusk8lc+<uozHs0P8IXyJhb4X2b8HZ-Z=qKroFO+gyDL5z&*=~xI#tCk=JwRWJg
z&z;gc_`$MH_!PxSGN=?qwRZIvl-BX(u<IcaXzT3z&OnO&5Cr`N!v^+!W7Hxs$mnLh
zi@2disEXs|a^>sm8(&|qwAPs|zo51%<lr<jKTP2DK!yQ@4VN7i)n?*7&08{^@A5WN
zs4Xk@*CxJP6{ONFX;EBkBiBYQP2oqt(Rr|KyK%Hb=~YR$);KmC=a{C6NT3%i<fr08
z09SE7-$Q_EQz%m@jmZZ=bIH29&1DchbQ+~>#1h!9j+PS9l<eT>UZ_VKgLoWDbfOn8
zcBG5tR=MtltyG$W^E?w%#Lt1gODtJCB;d5pEY>AZS`h-U-`MsWF{xfxo34sN^kHfQ
zs}${SFoLuU&O6=gzSG)Ht3w-O&J5?-X>9w2ng_3MZ<LbN2$>=er->92uiw6LJ0*Vl
z_R9b8Km7NUw->JaM%fG3>lJr$vS5zRInI3RgV%4LnZm^FdS!dNkZ&8LfDdZZoR)<T
zzyAY|505lI$b03qo_PApzoz7!zx_}Do2+sE_(1B=z0}Sv7a>a{kS0g3gO=rwKK=^B
zOSL^ZIag^{Vi5fTsQjv%^-L6&cb0EqdIFUdncQqB*6>KqBLvbEfT*oC7h2!g_l>+~
zYE|TH^g$b_3GI$uwmVoJpE!N`o$SWrM&AqlcD+-O_lj2x<>`&w3aM8{Tv;C<IG-MH
z!Be(Qw%<P-xgi`QPh)h8xow?6!G}n)912FG>=ieLL9)XFX_UC?U9=1n^>oCTbe<tM
z)acSyxL~b>UV%)yJKy2k;dH<Y7LdDP*y}NPf8FosfFXGI?Q!VbY)#o6#cdpxIXObi
z2sz#}R%&mi`#R_Z3kkBVvbQEbe9pAPdMPjhHw=w!@58`wo~;&!5AV91ZPTD`V#hi<
zyDNqdM9Z!)N~GD(IjrJ1PBwfi<3TNq&VrQ8edZLN6w^d7n5QE|Ko&%WUmd|u0zcgS
zwGJWRx;+WDfDM%H<4ix}oT`cWP9G>9Si?b_l-c3~M)~mFfc!or+eg(tJIYdMR%n!~
z&P8v7#N7eoY>4z87o_(ZSdwVYU{U(c(1RGI5z}XK+=ZOIy+>#Nm%e^m9C4aR^9%^q
zx@V%J3tZwfXQHn`iVB{Y=0xujAZ|IU69Dx_hoCpB7QvOn9-fXi-*2yP#L1K9M3@}e
zXomm{?`bB4TWrHWC<^*uFNL@5234g)r#bTB`GFW6Sj$r}s$$rHamzcew~e^n2-D0Z
z7rwl{@%81x+vUbGB_2*EZMeaa=D>WOh;x)JY}VF0ikI&_*r5Ts=%T$w$WvXcGFeKX
zL2ZRPhQ&`R!7UekC^?46Lv&>28M3z678uk}cJwBAA52ZneQR&EQEI(UqI6jZdogA3
z5Ip8N60|fdD5AO1h7+{aE3K*Jt~kf-dZk&L+F>oNtwG*#K?1@Q)4Q4((MqLtRXC@}
zG|xgcTD_}<2k#M^!K;{Pv`#6FTdvGw5U0QxqFL^>v(8!lYGq=WI$23<af;T_0N3kI
z&T@e$TU{5xz9~z*mdZR$l-{^rFSOg0%jHUY8~pUs9~o~~%C@t;z4G$<l{CHAcfY;8
z(OT#E;h7XBVw5Xp&l|O7rnm};!SUh4Cw_STfrrzB>b1vUa)Ekz<Kf}VvL?nr)!EyP
zZQHbo4OL@%IxRy9hyOU>V~};LciSqoL1qh%iY7)C1ga^c*=NZ6_uPmhTi_U4NAI>V
z*@ZF``8qF=NMg^0l6T6!Q%Y8hwu5Ni?bh|i9vo`#6&EAR=}e4^Z8a7cmW!cmIvj_S
z>Uwm8s_D?xH282jGcOCRRf=6+x)cQGs=4qupmk34A*v~1+qt))s1>tDi`#ZVsGb}=
zN5@B?_pTtBTJDvz4Beg@q`IEXyWWEP2B??UEQ<a8-*G?K-JOm7-yP{jtG&a~htnUk
zi5}>Vzn|Gh7}RmGeIJaZ^Pt}~VZ+XhVLs_G<S0`;yA22iy_Z(?Iqx=*S93s&4lbzQ
zrAtrZnrJ9lJH6-tAozs$Cg}5#$;poHZG>>V>#e*G60WD&bRF-P=)v6|5z$9>zy5n&
zxJNv-q9(e|JkwJNjy>KR(>HKD|7M{Fh8K+;gD03{?AYj!gX?JLELJ3S4A;37wk<Q=
zc05s3<o8+#4$fca`4|r0G`N)Af{V;$p+B!1jryKqs3N`Z>8k(puirEb93rA4DXV|S
zljt$}j@ugg2IC~hKm|5~H@x@C^RO_G;4O#TQvU8Abmz#W@a5|(xpk(~%+N5;nv%jQ
zaIEvhdYXASEsD{tjanMlZKIBci$HLKE?ouR2!7y)IH^NC7?P*Zxs-~(Zj{@NtyJ7_
zoKGjF6y@cdqHTS04b<M~ZV=*NeVEzZ5dBf)U-c$zISzn64yC>#girHw_dt8&6uVgw
zkI3T?U|og6IHxMuUK{jbNWoADMR>Or9o}FD)F9ZaR;j<Mg@mXQ%q*R~lR{vM2FE!y
z=I#5ggFGdD$X&Q^OrtBNnlae+O&SpAq)N7r<-xBd<JjhT0vahMOHnMGRt=eW{VWot
zSsG=lDr{=S=z&JrH%iXbdWco+fUt*8IqH>LDRLD!J4ozC7!0LA4sA)>o@r$W6%vI?
z>y=!KzMs*#<(+za<)8lHf95Zq9&rtR{pp`5pI_*;^Y-l*US5A@UM6xYjMjMl_KngC
z%X;F&hmYj4^ZDf~m)k`mq8b(a5*OTb;_1T={N?ZdpiYI;iN0615csEG{uvh{r_+h^
z>BQHs-zc@*pDDqZau&El&Rp%a>|O1CLjzbYMaY16Oug#^b*C1Nzc;&?ZG;l3bvtYi
zKlEX-lMpwCfb)s1+M#Qz_23<&OU!qq?|B~#c5a476y{|mP7AqYsn{<U^7Te9RlN;%
zJa*p=hBXLu49@ey^TRWXRp-jJI5aE|N3cOhsH|=2$3g1N=G8l`?09?bx+UuorNcSG
zv8I%a9u0C8*mI{8PE{5Qj{ZK*c-*(d<CQ+n(z|8z|6qd~L+B3fU3C0C{|w*J3<tu%
z`#)!l`|o>D6W?>#7;G;7UZ00y=eh)Uh6LODHpY?GsDB^4hVI@wr597(+eMF;qhJg}
z6i^tC(^=*1x$A~>c&~=jFsU_iat=?IzyGdZ@dndJT#9!aV51r$(+pAZ_d!KD5W@nT
z^g4FCB#*>afAIK-Zk0XLoL5Hac7_cCI!A;gsrbqJ-JUHiUe2;?oJZ7%4-}&uTOX8C
zxNW;ob3a*PsX7+?0mF2CufcoTVc{KG4sV+%)K=A5&~b8Ge1|L?!@&68KUV*TU%xG9
zpmkh$@9M6#Ge#ywwcHK+V0#=XMV3WHIm4-S@a1}C+Z8+$LQn_6JSl#$2NQ1<gV@ZL
z;yiiRL^hR9TodyWA?ZyF-f=$7e0Vyut_yFkZ`{g8E%Mj5-kFFjbL8P+Ax6*98||@&
zv}uxB)|vUVu=P%Txlu;rw9c~9xyV?iyREYJPQ=r5lO>M9G$-7Ipu?-;YjF-2wWST)
zI@N}#L7*Og_B6516Z4YTYGX6*ui91<lT@cFEYUwNU0fb0@MxMQd<e!>)%!!|LsTeF
zulMca9(dtkgh3hN`O~DWdHBKbPAcwwSCr_%uc8NBp)=b4yNTQOeK%#hhLUw%NsGjG
zc!Z4D0kbSKQ<}u}wJt8zI4?o9!xA51-*>LpH!inVYrJcwPS#P-T1Hb}=Uy~~zPI63
z*LqCdIA@z@!;d6x+jJCZjx39;N!ykwLwXSxJh4XH<&B^I(|_bYCePv~{_gMoH~#pa
z{{i1Sb-VEG^(RiJmAorH`ODikUh~dh{rFdW`0#<o;17TOH~i^e{+ZAJ{F654HE@3T
zk>|hsD}Mazzvf|m<nsDL43XD*CAY%M?MCT?d76|L5rdM?oJ61p2S0#XAy#?reej~@
zOcT?(Dp#9{al38kHoMzq<L&Cu>F<tQ0oT6^?-A-0LX^-i4no&!ef+xMJ>5&x6@uj;
z*d^2X*8(f&BKSnEm6k=#?AKTNe#Q4DSA=tPng)RbP#uHwvhu^nPn_4Y4p2Ry;*1CH
zm#y8E&X5jgLMw0#uMIJ8SDKHqy}5qBgpZ#mdP4?cu=Ggr#wnN}uIV#2?D;)<2k!0y
z49WX9C-nCvO~=3Mrnfnev?CkfJsaeJuirn@_W_}c20ensj@P<ctKRp*wj)s#FLze6
zKUb-VLyeh+ta&fkemMQSF$A@o4&$eLyF=_M`|Czq$j&?tYT~}@+xKe&*g@lINi=JC
zAGmHg66nGVK|KQpEY7I<%S6`RC@zCO0SkI_$N3=9G_9dxB_T{<D9gbUaECk}0R>vP
zlyW<o`XZ)jVoHga;(ak0y-OQ1h63q^2GW)~RME?A90gN4l$`0qj_0M1_sjnfl#VWI
zd9UhxDEnqn(+$g|yd!wuqeg%3*Ke~H=E88nK8QMadhhglOz47-Cg+%g=V@Iytt%nU
zY`0flE;otYhCK#dAL=wEO@B=jRBctT$aj-xYlZFBS^QvKWN-60QjDY+c|4t1=82LE
zU&jUSJZp-a9~S1*LP~-2svwKw!&FKUgxQq^J*_ionZPdX6htM2sQAfN!Bw+QwF<>T
z9g2_$@m-<{aOzobIAS*lCwLu#y>Ywle0_N%)XthFo=zto=b1P~`pEVHu>*^SGS?Lv
z)as!?Mju)SKH5bd6&+e?rQ}8%s_|^-;XM+;VpKh*dZJ96GkF-1FeTfZ)e^boooSix
z-;;(gS=6GHIBhpQm?c-P*BjG1ksc<Vet2Y_XSS@ivtSX$2Shn1k57R-2Bj84tI(lr
zn~B8~0^*&c>>HQ27f$DuL1UT%OS9g2+gzMerh@5ytn9=#iYNuUbLtI<L>+!rVM{4S
zPN#*_X{Oc6^>PKT(x>xzVwn@Qc7FZizc7%U??3-<f6w*hg*44flP6KZ?Ud5Fy}t7G
z>u3JBzwuxFyZ@SXNla<wKm5&q;2;0)Z~0sPqu}%4`0y8h;Q#*b{%`!%kH06y#BIOQ
zccsPUQmJL9qEMA(TKV|&EE`uFq-Z%BV^I45*GSC{JA?dtZ&Z}BchD_;82fHY`gZWp
zURB12&3D*l=M7mI!=7UXveRl9@KqO!ca9XG1sFzI_0GE<hGA|7r6>9{ftPsgdc9KH
zAm^RBUvMpxoQ3~(9qY|?ECrd<vhwNU2Tt>e+921RDeLpuYB}az;NVeZom4-;0I%GC
z{Ywpuwd^!IfTe*5cbXx31(yXM2vODCF(_G7mNtJ65Yz>(Q~3~N-oFn!Q#}W2bwul8
zwm7>Oyr=sEK{$@tL$EpSNvDJN=O4ZkE!xixqEpq8@{SIaS~WBT4O{gX2Ap9-T^sA_
zWgG4RZf?+>=uqsv^!79iXXjCHxbfarcqlzWoJjL@AI!c-mAhH7X+`J6k@9GR6?!N1
z&pk#}VT0dMUgWHu(@KN-!21M4)zlJ7I%1G6=0LE!S2}GU7Wd;i&IKU@2*tQ}PoVd!
zyF)~O*x9v&v<`6`7SF+mS}s=Pw%DBl@o?k4Rz5fq12>|CwcRogP-cj$x^LUinVhTi
za$}f#&%RIK{{7wmpFi{KH+K`%Qtn)dw~MtNr6Fjihz=f>nGX*SJe<zNlnho<$WCs%
zCdOJRwK1Z{O$myMD^%WfK)U|UnZ1{I3}Q-*Dri2$Xce^+(MLtLCdX-6cz%3jKCeO_
z2$VW_DLdrO>t!Qn4JRpzvyRa-EsF)#jC+YnG~c!x?%vkry<<uf$y;cOQwyVuk?aPy
zl3C&;o~KI0G@PTB!R=DmZ;jPErl`Jv|37VizAH<XrFUY#J(`)jpL3^}a;hOo7DciF
zK~ke3J(?a*56}?Mg2XD4l~u(YBjVnBCU+b5s6W;+J2xARW(Lso!p*2S5#h((%=X@E
zed`;fv$t5QKHs{StHM`%aEx~nySo=v&e$WnJSc>@Z5#8NunwYDUVA#3wc=J5<Ia4S
z?>CN7s@WLWk-^?Y;#U>*X|3T0hl^gwmeYjA>g>0RtZ$w9>VD+z?LF__zF`<fQIV{R
z=!8{PCeWZPqSl+F4{k{aiYd`sB}C6SX*X;j=Z$Sqw8R<4A+jFv&{DCpKQ_)1Mj?bv
z)4(_lj4{egv}<^kG2oEjqxWGL2h!HnGoj;E`qP?Zska7?A3yQ*_{{UCXVRL9)5s7t
zT#JodR_5iAPanVMAuX(FkwVN^Mjtp|E;Pipbh<ln|NcGiFYoXLI+^S3neV^<1E)B0
zK25y4yYT(@-_s1B(a@*Ui67b%DHpQBo~?s(8phM5U^YN51&ouTN!M&^l991<laHHj
zkdZ=v^%OCPQ5M=^cSM(u9-8u8S&=enXX%}6maWOStSBSBb{#3Yz%WkOFi=b8mNu5>
zM`5Pt1TEuD|KecPqCig)?>*;nB6!bQH>ic0H|A}n<$N5T`vzZ8E`)IkrM`B~2L#{x
z;S5+6<_-S!^2qZlBd;(-QGjSMWlaaG2MqgQkOO&9KO4U6)OX&_J_DC-P{ySVpb3HO
zfW7+{NY0;)7UZAr*tk)+_;oz-s{RnDzv{Yf%B?xXo%&sjvJWb6BD_BhEYDX;tvq6Q
zw1ObD)@Y_4>8nTTpV@gzf+SgVEU?4w|ErOR_?#F_7ZPBrB6;>+pI~Wo<#`sptd4=C
zC@`(zniN_tIDC)TQb<kEGqZaU7=vEf1AUZ3-$>9}lM$34lntYWU^lO4k8mA#giH=z
zXVuBGN)qOg-_zRh+*oT!rQ^LI)4PkI_nzP_zFT@1E*B=tXKnkqRMtW3g6P$nWz42y
zk=<6?YE<)qzw96KFL?blWMH2M**{N05{<0UM^fhb_os<z90(!E0oXgvJH{c9TGzmu
zLeH7z%yH1rUl;%Gh=*QSn!=5UdH}6{YZp1=TB@4%1jgXFOeapmsL555-nCLFsgbWM
zpB}GV7m*xUBZurXjtt`<RT%P_jm{K<_u!<GeR&a|eDT7z2Qh9hUTPH>KX$M4B{bDa
zEUIhiq^&VTr-ykI2&&7xyy}d~SbO-^?>50f)f-}k8bmr5EaNDeRflEG8@W}+DBVjO
zMJ{DD?aO<sBs(yOW89v0H4c$hyZ8@yVes#pM>jeGnkLS7cZ4udbC%*_7^qzm0DbUs
zE#s(>pOqLRwIu1-kr|!~o~ScIFw|1Ps;QMVoKAtu-M}~mHEa@<h4Ydb(R-DSu6Md|
z?4ICQgY^bKc*Y^}_T3$q%L!-139qVBLJ|tW+9rvPgKR=!h)UuUf@4bZY}elD&e7LQ
zTZQ0hjHW$0gBUp`YE8`fna>YD^6)bA{m0+&cfa~8JkWZk6md@+$C1kzp%tP0o6eRJ
zPcM(md0~j7z>vd0wFYxJ^X_NAA~{bE{}H1yG+M`ci$glp-c*j|9lhiOY7|19z_dal
zeI;7$34dahD|`VlE!qz_=|CKMz-lqDzf4E;<Xto7y(C~3ouiv#g-xx`^LV;2UET^>
zu)NTEV_9ZW5(!-NMo=m(0&vn5`VP&QV|HI_W2u>xD=n?G+Z9urGzn3h<r?bIk9nQx
z>qkGF+u9E4bD&ppaBByu*KoHW_KI?S>G<l@x_O81dl&OD88ygW*6Grn_bN&cBQSaE
zyH)LOfU{c$OMn#mp6$REUjG>L=jFj(^zHlDevs{dZR!iLABeIqiDkIeG0fg*oLa?v
z{E`3k|Nj5+n;(8;n9h9tvtQ7^dW*fB6{2id>|tPLm4{&aPCnI9fu0wypHa{e?+l|C
z{#bAZ?@*^%C!+!<IUgv|`%C`>y5!%Wh17E0_^<!?kIawH{Ospn^Q*u41>So#Et8-m
zBNtfKtg+Z;s18~e)1%s6@Bdy{VU6&k_Jh}SoEj(Xq&@3Au8aSIHNvy1Rjx(OQaE_~
zYH9qMU!etR1DvAnUB<1u36*mW<Dlk7T32e?gx8`?)s93s9ck^2%wGx){6qqw93&DZ
zq+Z;{DDzhQp@XyBO%wO$li1H%*h*ooBC|P716%9N>q;+GI>D|pqm)6TO`X~}?hD|{
zMTkKd_POB^=fm5!v6Vt@4HqN9?G8~&=<qsA-!f&b++JoLo^Q-+R^truvp=6E;yB@g
zBh^CfTJ=TG>F$m(4y0{ktBqw_>DJ(fi1A)#(WR1?OwSdM%ofqg*om_4$f0hfLhxGo
znxops7@1M4+h6M)XT{|#dhuOY){UKv*lNW^IaGb{nyM(Ni_Uw(5b$1fvGS0<ntq6j
zqo@hIw@hPX90yX;Iig;Fw5zaQ4*4+#$z3p33|+LNv_`CPgOemN=`i=_2lj0E*2Tfn
z7}No8XQ@|mkx(OJEyF0tKpX<Y5X8FHIdZGymdPzkv~X)U)7hER#tK|$ogqZWFa(C^
z=rmFlHt`Sx7E7^(?k#bOOy?8cTXI^7ihiVJJ#^>S+k*9xX*8H7UKQ4$+U<5g!^RX^
zFU<3m^?KvCzyDAC;pe~L?fDHqe*S?^AHT=ZI1Q0I=g7AwzJLA`&3wgmzVhSq17Vu@
z)nETD^u|B^_*>lN!Y_aIYyRr%UouU1oZkO0Sbi%|We~3pYb2;-7zd)7Iu%W-i-SKl
z3Hg~Y&8>0A8sGcbrelL$wrW7kfm~Nvn57})q&-U5S%d4MckW1Q+x2iL_4R<Ptr0@R
zhe5_RRl;t>OsE{lTWc0;2QmL~7F?iWiD+V|7eebo!<C9j3F?NeMex*CaCMW<>zoWV
zoM_U;yhfErJ6PS|6)o6N0C%XQ1P<0TNAtznP^fu-(2s*^=LYD{Ti+?QmLs;8E}=>N
zz|0T-g{}wkZVq(ZpWctKBoyI5yHGj6taj&8wC$g>AK&%$wMTv*q+>;0_A<lg49_2V
z`O|NCdV1lvKRyx6GY=m>bN9^$e)f0&8h<`fv|IPqAMe%oS%AKFSHpw&X?NnOoEyOj
zT6vlVrWnN#&S`>UN1hGS_xY%fdAc+S&5%o{Y#X;9KJ#Dx{=e|Dt*rCR={)lF&%Z`o
z{;D>wYVnQ~;Lh%HzWo{CKuY!obuSLk>u4PegCeZX@+t#Wm(MT8Hap*A&$AF!K^60~
z8Uam_MZRB0q=x&mvwtSzgg;hGrfpfQl6%J+wW993RkL|5_Wr`xPxBe*T_HD_W$(CG
z(+gxxBlyTUib0KWj#3R<tt_=rtYtVwe3G=R+AC~HNFP=Vi0Gm+uo}6$JDYW#`FPXt
z-I3-bEc7wp+$bnPZImQcS+|n!UP^)B=t<lST@mNQyk(X(b6e$tb~_M8M;JU|43g*G
z#Q-M^foU3f`}Uq`n3(68yX%$PeCB+<@aF9srePv<FRoU(&`MM2-HGqNlPb*_OW6{+
zR$)lDOiG=c8e4AQ4gD}bL5!Rbh^v;0i&5CiO^hv47W0bQoA3t4hzSmpEowPH^nufO
zIu_|(!JCe{8&-u2Ex~yP?`Wl|AA`)UcKA;`2fnaRuIbohEE140Iq5)`Z4-FBIBx17
z=)H8T))|bqbT9VLa-cYzfuvoS4-&~do;>3@Fr81hAbf^giX@9wkL%){e{^YLbpJ|N
zbWly98u3eSMgpUZmB5hZ4Axst=M$In3GXboF3O9+iR7xJLbI0OB1VfWu_jUYTCJf|
zNt9YNh0<`E0zW<#^2-xX&%dL$7k>EkM}GX%Z?)RDf;6OsAAa{6*2hPd^#$s}&%XOL
zZ_XEbhYugV<L=FGxj&sTrQw0|X~fu$SJ@@bifY43p+1ZQ-of+DlTxL{u0%2EWTFc~
z(lQ)UI>Rdzx_4bW38`BoRZeYV_oA+`Ly<fC*%58WRN>9Y!6@CMbXE?mqqRaxE4{;-
z<@sn;B((CJ2wH^ggb?Z8Q<djX4HVPZN~7iuDPhV+XjNyg>b;K@=glUaU5FBKE20u3
zU3k|b((VVa5HBS$R7OYMI;{0alOiY^lHS)voxtuN&|mf9c1Jjy795J=R4u>_1!tXx
z_vokql?p4g-LxoZU)&2p&Q%Fw!?6atX~)5}hq>9V5yX8T6SM;<+9TB2M+z9d9?ID&
z3+DMV%i|-{;Q8wAL=3{IdHCaxq+8;b|Nh?)zIh|#m)@|}%kSgMqF^66%Cja}HkIf-
z(ZS>#gSSl4GeyT3q<OKyA9H=%lmk?ga4KFkqGaQExjnOd{DJTP@KFfbsq(vz51fAU
zTfX|)2T>mAbG;Wk(tA0HYevRqx*!yBJ3t{h%#r(TE!b%3Juq&OVF+xwP|ALIE4ri#
z94Y>G;-{WzwUD-zZHp}4)BRo~XrWUDQIe5^1p7@sLd~LF*;XhOvEp`O$!|CcTp{Wo
z-}Jxv`iaFrJ1%5zjb^la-#Z5&I7tvZO#{2h!;&*wt5mbQ3l?#6>z!Iuo+v-i?%y$K
z=#X3!VO~1hWu{U(Sy}v3v;+fQFY+#f>ViKB$wdQ)f^|giX|7OnA#Is06`4aRLp#Pu
zh*3uXp21ly1}0*N48uTco%3|!>-X=uJDquV_g0vvh;1;+$~I2Oaa|#tJ-MvuLAh;-
z%{iXtji)(rx)j}XoD!4(CHJ+~6&9E8Sz4tVBbB@X8v}k2ABbK18jNFz@tCVtExOo=
zht_GVs+FLqRw*UWS^)KdutrtDU+AFG`PuGOu#a)PmrmMw(Sf&CNlCoHZ|hZ&fJirz
z4rZJYD`O)La@L6}nX?9BXBa(Uu#Bg`c#ce`fVGj{#YVceg0r0^btvM5*qhGykp<X<
ze@?E6B`?Pu6<w@iHQI>d!2SJ+^JxMDrDQQ?QW&@9gpFcwGM-1QxA-80(S70Nnxx<f
zPR1XDXS@u=ljrUn$m;_$8@HG5>2*_L-`*WUUal-z#w3HYOkuzj@kSV|$JWNO&GgfS
zS~k|FXPg#r)f#bmvmNUVF*>GEf}yOn3hPUnU#k_Yc$yz$6<sHNhnHLZV#chZwT{mJ
zWj?<DydT|^4j}9faa(c~SIDMGn*tz7fwgU7vs_AH&5MQ!!5)9@)w<TGF^{#DXdL)J
zp)=<~uA=s6X(Lc^y^&kR8M(Hph{2X|l^6$^ziY87dzu(1i88<+#5!lC81POQ<6fUl
zjMWZP=I)AJs(tMjhc#bxWxC#aM^U&}=KC%qw#syGO`!Q50oYHh{as&m>>LF_YvPP3
zn_tuaM*LsC^bs(6E_{i8NC*7Is7L?3W(0JORHkV7)EZmPe0YD)yZg7SYvPZ8`jLl^
zpK!m4{QB<S!O4mI^nb(kZvWi1h~lK1E<QEJFoqySK@2hy(*(xd5mL?@DFmdquk=<F
zMKRC}ecO2W^qEgTJn}Dp_!B+`Vs9)tiPy)v;Vw}R_Sb81pgcxK!LILI1X0|G`qAj<
zOwa)5q){<O$k?@XuD6wWNna2!<(lT4|Ev((N87uMaa}rd@OWb|L%^CY&bW%m?Ka0o
z{R+kyg(#Z^Rj2Q2GeL$Wr$fh1tr60nkM#fdUO$aMSS7*CC_!SMH}%>M3C|QAV^qjQ
zVa7RaR0}p(VU_c0*(t?SW4GA#-Z2En6g}fG(0W&bl!VjNBE*TBn(E6X1Hr0PqvXW4
zZFtkEJ?H|_nP(F_(pHPuWu|g0syGJTyuauE?LEzVwzPsVoK9!P)5tiU7{`$q2VroB
z$lY|}?d6X9^98IG>&Y!EL@D!C8$1|K?{etX)>t;N5Z*Jq*OYjfXYQ7fyLh5IBb|^F
zbC?)BrO89my977oEJjJzh&A;2GBTb|3_7y#-f<oV#$jX(1KtIKH(G#+IfeJ4kq#jM
z;>MVB1f=7%PPEc{F_v@120`ia?kAw?1UF2aFK4FH#2jx7!yxJfy3F-eqUZPA1m|SI
zj*)4Qk%KYH(dZrP5mkn(;5Z^;gWoqwLoN+hEup(FY&TUmY>h66O6yLEHwBBx!z89x
z4scu}436nMayn0vj%j34xaC4E;y!3pdN^IgNUP_>?Rv!-k&D_xawbH<7$b2CjH8Sw
z&e7p)B|5_pJiRq?sRzt#TBah|6W5ZsrDxjP_oTe>^zxC<pMKyUe)IR3U)}il{r9{)
zewJdzTAG1U3tJZHm!g&OvX<IlT{GTVa+Y?-d2n94;oe0SWnP)+y{m6M^6ZLhr5FK;
z9{p9yw8PR)Q67|3Cn0)<q|z}d*s6FWbjYoc(x&h<M?v9~SY@qM__jeUjOPjC0!yvf
zEE{I6Ree8-YPE&IUYZiFnnI~Qh>F1&EoKCoH;O8G4b=jL$ek#aWJDJwQ$&;YS}Tf!
z>|=s{jEeT}(OOsXnf_}{)_X4r(;Eh1l^d&sz1qZ3%e{_AX=tOdztoC%|K}JmMR6Lh
z1O)wsx*qoNuEj_ipOP}a{CBK%xNdZ;(CJuuD-5=BiiYkj2^fZfPGy=#?(WWXV^|(v
z`2IirGe7&ce*>pcKimFkIS7`o#z+DV1?L$;I4s)t0$`7LKN?M|_V)^6SmS9;u2(mf
zR4Ug`KX83|<{$p^Z>gp5?!z5^8hLzrW?pCNvcToUE;HSU9^F{sBdF(r-9Pv~tJhI#
z-@}ZojLCcyj#%)OQahLPiR-+Q@}`$nh<-))ph`QqIfCX(bC-H!nZ1l$TN9LTFQ~Lm
zH#@SXTpVvLspZ3KzyKixhC%jQDLiGYlv4h0o$ANyC#wnM08(bC!X4Ha36$`LAvjK>
zESwl4xmVVbSaK$n%$hQrhVDqFwaz(oc4e(>l!Ny~Z;*86PHjabeMJZ)H0hqawWM5`
zw~bo6b{gu0z02{emBnWe{qnN1t<teN8kaHhZ+`W2{=?t?EtfZEu1n$(sHQWWPMj_$
zLL7-fBr&B3d%Mx_&M^!l&U;G8l$^ywp+guV<2ccp(b=gadbK9s!y7n_jv;hRmsx3T
z4ctyDG$fKN3iY|JwNO*0sUO1>2L=o8-dwo58%6%s#lQaT<-!;T;&c+^#R`&8biU}6
z?Y(c1yH~&=EBeB|)xfdorjqf^iyieiaEcSH6>{FVdviy6N!-7=W12=W7}~%hcBjHA
zkOeq|z%&k=&ru0q7UPA#;hS_y=p3`rj)O;LrXAF6(RZ)wud}X&oHHqJ>VhUR9n%fo
zXriMr`h0p?Zn*A<Ll8wt)Wo*dNI5fanVu@a25N=qEk0VtDUeT*%SCb<sx=H_l+lIV
zH#GTOLyQc5WQb?kl#Rsku3HwfqoR56t<b8cq?Psd#P9y)?{Otl^1|)$2l~qszyHtw
zkzsvidAUMbBos#r>Y@dv1J;~Lt>X9lS{hq}wKZ~*HEg2hJ*p-e?S$w%LF<dI(psz!
z7^E23H_C(EF7SZnKgi}OWH5pHqp##FcDzL!-r9r}eI&*iL24XoMz*JKS}S8PBT1yj
zJ2I~bf?~4t+^MwV5ZMu|-aD;SYR+u)!swmYKpPnc7^K)6f>-xIEtZ0pb?80gI0<b1
zMOTeV+qxn{2QnbTTF^DKKbs=MmcS!qhmt+i{5hO1p=Ph$wE5zH`2u?XQVG$T!J3Lu
zh<^W!U5kgkAaG6&S!3k)^j=<ngE8tbq4%P7f~i;vejF)%rsv8K<(|BK|Bm5wCZ)pn
zfA|yDcE0_$zlL*QSAXoqQ9qt3xfad}662kffG+Qkq)kCo)Sp6&w0-m;Q9iOGSsb}L
zo}Zq%Jv{QefBqfUmmBZi-V<Zs_WZ({Hs+<#wj`;nI&wZ<#$c=#13KC<T7-3#g<2zM
zssW7a*yspxz*tX-f=&&mQBYA^u(luWfUjx}X=s9`r@p!oN+48PH!NKZ-{d@SilXfZ
zfkhv6eFk%q#&<6|oeSudle=gPF}^+v|E<?gE(VnFjC@S_zuGI-!sG&XArh>SPznUf
zpSQww-oz+pt71xn66y>{hP`}A<Xyf~00961Nkl<ZZw*)|K+8pLplNg?<AqjqV=_{B
z1k>3z@e63Jlagq&htX1UqidLuJeY}`HkLWD+%|G4cx$*j4}A0CEtfDdnt>W26$#B5
zhLQ1f!n=_Ulw4S^H|Fb&yd}PVqc9&`-Kb<1U5#N1_8=cwKS)8+yUf0d#J7GpPkeiS
z=6;G;58docZtXOR9KfYia;a>&l2hh7Z?q;0^KOb<?nMRRf>o0sf$d%J#M3BC$jL~B
zHb@L>h(d~c=~M(dcG{6!3qFczj2v)IJ8qd4PosE9jN`~OOw^hfhJk4uIp3eSj3)*+
z662&ge%0hEc`+Enc^G(icj9&)sj13!3PCKcHF9>%;tL;#NSFo=V;8M}jTRPrM5#%e
zq+zb=H{Lnswc@6>YlDM8!>!?KA*l6kqiWYApQk9pTui2nC=E2E#CAjIQj`|EXPkR2
z`s6-sYolgC;nz*ROU+F<2(?p6W2=pda5v~1rAVPSKmEujE1iU^8&mX@<;ME_z<is@
zTf%$85IeW6vu+KOWkXwYW!)0dIFV4+3f+-%B5fIKIw3k4YxSy|q+YKBUKT|_M>b67
z2Ia}k^54kp;pmLbzPYN?p|h6ggAkiAlx`>~6Nng-)%43_y~+iRdhK_S;nJ0d;zadv
zI*Uv>3t*48mZCCMiqjsUP`eNsWfWw0@K5Z$^GaGQ-cAf5;GM_zhPT3Z@kZPdyPN{J
zU<oSOjWHhcYw5Vrjd~wXDiFC7Z@-uii#)8-TUJ(Ir|V|A8vyMLd<o96$MIiER%OfX
zNgyKt*^;GL7EXquYCGzohvt`rxZ}Qp1j<>BZ{{c~uqI$lBio2i38$oGYdp_ep|yp(
zJ1_eyh_8n={Kvol4gT$0{_5Yor8KXKg`&3U61lrSOECt<Ap9pWtpKJ#S-*;S8?ZLu
ztjAeN8Fg+qDMPw)5f94eZjhG3_V|hAdEtl8kBsLtVH`=NQ@Z?n-W$?3<E%7zwM&bM
zUdygt8~d!ip=gd#M?k0JEhB}A3tlr3BDrKzE)3Ja7(L!~T;0z!qiP)Jc+75n&>A6F
zx)qyY(~&T&Zn0LRszxmB6k@T*2uSz=@_eo9Mw%D4)P$hhr6}GD2CTuy*JAI#_4<jN
zf!8^f_ZAl%C988*?+n3n8V7=Rdhkgo&b(!|tqb!zRfO*)RjrTWVwRIYeo_?6wrn&J
z%UL-bjrFuHGsxZJA5%3FRtR6MS?Pk7jkIp0F^Fe+@buD1^Tze*%JcJ;lnbMgx!``m
zot4f06ap>?Oc=FeDW%eixa{3-H#}fl6SQO)1Xhd#IM3P|gOy6uhky-M8r{4zj4Q!e
zE~kn2Z|-?_xnN@iECq>X&8-}XVKRFXsC(Tuwv@!+$vLLeh>yZ1?<g#4E!k9p2O9*o
z4Kc{#5XTh^kW<D*ft|YP+`YNu{_Q<^UAVkCaeZD$y~x4Q8`fKfF&=+jIRdX;+&Dzy
zFc9NFaFG%k!52CeZv)W}^82UG-Q|wEyJxm}qn0Wj2VMeT1i5Ym8yH<=^pet-t8}ML
zzvmDl(MM`4Y|cykY8BVIlq%CS2qNM9-Y^O$!+MN)9oMw}8Z2V$-bo>yK+y8G$xLw7
z5Ep0m4aXh!$b$aJIA@s`@x$njnglXj=S)fMfSI$B4qeHTThWE2RB8q!oT}DLa6s*>
z^UQKv$*IaX!h7Z|lS@^VhRBavX)>NeG|}`f_PolBDyc~Lk1lR`E@k}J*a=XEt|*5b
zD7xOw4n@D}^7joIJ*@YQ)&i{uVy&dwu%_c(AX-cDj#RxIuvQ4WxmDoxFf_(+8Yjj$
zNH#$gzk<V^%~@J7+CiuVt9?yX)jsJMLA0t3NeBezwFnSNmP|Ob_z96AP7Gl@{5!;4
z-zvLb-?{NRrn0Zz1qa_^AA{<+Y`=6z_uhBgSQ*{zKifyAScSy(FhBhAcce_qou>EK
z7|oa5>j7Q|0`hkDur1dHQknJgzH}KoVH;`NIJ@1OByxRvqD0ur3&9K1KbNEm4a@KT
z@t^qD|K`^W=X<u4=>4YdqV8CGL3vrbbKg;zdy!XrFU41I{`DZ%=S=K}Wo%hn=lb~*
z`Qan~@Q?q6&$oqdKAg#|9chc>5IK*L^?Jpn-9c5y+s4o|!djV$(#gwgL%nmX8m#S-
zY^Ml@d`^sI95Uze#Ce){4l}tlYB%CqsE7u}9Ama4n#)RAE6<x;dZWZuK`0az*w+~%
zmo}<kDz$K1R+cTZ<i^;<JPk#Dq%ipJq$V)PLh->7rvO0BUFUj+7{s^RI$;}^TzH;m
zp06`$(M8@~NpQ|Pm5N!SlWv&+-8CA8=mMA10H!kp$93Jftqpk_c^Fz3x3N|wtYv4%
zn%1dXBCTRblaqr|*_Mslys@l_QW{RJavvTZczApZR)}YVm4ntBi)~O-#@0%n7v|f{
zJTF|L$Fw8_BrB3KG2YO*xZG-i!=fYD$W23>gy$ohjvoi&-3c=WdQ{R^&zR<p2Pmba
z8@0^4gaRQ8-Z?9t06}%Abk>|WwZh~jt<5^ranOtWhfb@Rk}IuOI+gJhd2@f^d_J*S
zIGrcP^FT`-KLiYwVTue>Bt{{~#UUsuQD?_iEn7oS*viYs?*pwl1|JywD4SyMj8QxU
zdd{r1$h9;ctfP~qi|_-%ID+#uE67l@4QIp#+aU0EDH7u0tRtt&a!ce~>6a71c%pOU
z(q1>{J+vuTA+l<B>s8_u1}Or9vrJ=PQGbGz5;&3Ul~M&7cl#!-dgFt24(D-T2wn<-
zHF3MGw9*-acvS2to-qO|r)?vp727?x`Gw6{)@>$OBZ+5KHZ^A@Cu59(Q?qz`Bh~&o
z7ey?wyF%xtAr{l>(NVWT)4BU@0%NT^y3!r-mW=gIDM{ec{zYT37f)Xh)ub?RO73bZ
zQ*;b5=z0~*wYNrZK@5<(p&8FsEAu*w)iN9e$z3nqY7t~5m*W84Hz(t~AVjDEj^>`o
zKQ?%?qeB{6i4dL#HMH6}P_0QC-Y}jBaXP}4_V0PzEcFzCSJ~el`h+4ZRZW;=hPy+(
z{k1FYa((uK><ACr|BC#IU1hO5&B-5gybsE85p<;QQ;Pi@I#3^ZK9$VuWPG}#GP_w&
zZ<cP_{<Ro}fni9*IMVtH*CoS}pu^>IVvL>`M^2~0x@`REpMS?M{{4I0Ffo`;Y74z*
z854RZO@dbxz{y{&3mE;Y^HKK?cgXXEcO+f4b%9PJ<&_`)<)7i@g^$;ThGE?{)>0|R
z?Rh(myt_M5Yo^?0%;oOYccC5*0&<O%lVGfJl#Hf6YPY?kkiw9aKh~oATU7po6j~0j
z)d|<DD}T&5$*e(S&%T9&z-tbkgk<?R<AmR#DAX~EvJGE^e7S9f+kEBacH{hVV~jGA
z^UhE$f-UUp<1cvqv@wA`>_RMio$>V|>|5(RrNOd^ZLxEf>%6iqD`jgMj#Gdt0Y+BP
z1?SyCT5{fEgTQmA)4-eaiNSknEqs1?p|!$o6Ve7WJQ+KRlb-XIR3#4}sQj*1fop5%
zm@0-i%a+Nd9U5Hkr0bj4nebv|bDaW%lLWF}Rmp&5X`Sshvn&g5$2)?v^qR@rjKPrB
znY=E<-WiR-)+%zl5D39@z1_%0qIrAmY_*dcRAZTjfq0%+vvldfds@n*IP%DqVTfEV
zXBo@5MHB#PQ5?pB^WB9w1Y%C4IkBXLxBZTTFr;0{A>(tFby)<4G;n`DbALWFjwk4d
z#cqi7vk|!^j?4XpySp>zJ0S~Vg?LkY6|P3D>e)RI_w0d|kA)qqm&4vPnc?fE<Wy^H
zddEA_L|SXfrIK<v<c$po#$jWmg)bcPtRb6?=mOCOOp%!EoHIEUt@34#UDTJL<SbCz
zo|mDE(HSRR3e&*(G!j}BE{25=1Jh|DOat@hXHjl=N7?7cUF<`hwVciq_iry;pI3<<
zHeGe}k_ss-c45r-u3f#(p!Y(e_Do5V(QbgM{F7!XL-a%wxWAm}p)gGm*2FLbY%nyd
z7O&bNR)zNWjb)dr9cQ|;7$4-Gwk{-C!@+TB%6{KBPnqpXW#IO1M~eaD#cXV6MmuX6
zW0XR!zQ!aAMA~S6#Ds|@C5WD~ZV2Jfsk5j9Wq8*w>~wWkBG`Z%9QZ^Cjv-jx@YNl!
ziy@MC>iT9a!Aqg%gK#~Z)9=uXI16gZDEmzR3(~XMzdM~~o5tay*$g5j6h4L!zq*b*
ztofo{7l?(O&EGMH>f_5fuhCJL5wSg;FER+acAtHh4ewIy-H1r8&;ORkz>ccWHHTs^
zcy{SzgI1~rm5B^-OO>J{-Bug5-*9g)4CBP9cDCI3;U9m;hY#N{eD_O?jr33s*`@@R
zMM4<lzIu;!0c$<t1Sp&vW4(Uv*Xt+Gt`TKHgYSR%k*6R3gkLwFQs=8T_cBtLPT1bb
zd6mqU7$~N)t~37bo>cdI88C&zl*`aXj%>lzhI3-S?7EVYyEtdsYH>~C!u7@px5OCP
z!;F!2{^i<2I26Vi4KQ*v3f6SQY3Nl-Dn-E#o=d@K5c#+$8l0wMHnw%<<@uS@)0yBM
zDTz|gdojto-R3XW$^ZWAClv#BpFvq6oz<@F+18H@uy>;#;gz&3Jl-UnXx%gc%yu>J
zuwn^r#QR@1>q-ok7#zdk7%v0E7@6Wou9-8HyDhWcG9-COs*}xV937@PJeJ(LRG?!d
zPElloPW<!N+Q_+6n!y8ux7^>Kcz1WtX`F~dWE=*392B-vDIOi;A}41hW7~Mn!%5i1
zwUD<3V;nhcv@J2w#T*5Q4o)!$S;ok$ZfzBc*!N!7nY*-c^7jNEsjU)%!<fKn8n~?+
z-C6G6o*6F#YaQ`plmj~ihSSL1{e}1M-V=gVrG^n`b1#aFK;gz1Y_uAvQdzf+c@+}M
z5F!{ERrH<wHbsUQm`)?5H!k-V&gV1d(}^&Qx;U#?_gYbwxDbzyHI9zdEoQT{Bq1WZ
z*L*RWJl7CNtaOUeIO8D@YNk9?=G!7%iQJ?}a7xYtn%UjdL{Z*L#UW%(Q#x%a5>MQu
zgW3s5hkOj_AcOTp8zp$fIHGq1Z#dlz#L^jjAX~%b9Jx#pH$<+(z~XKqWlW8&WC(^h
z1hE|+9OLBaZD1Tdr_(^m8yE??syLZ|YWm&YjqI%a;3cf3G;#4OO}rD1Ec@Vz!$cSc
zCg-_(bCTabdb)?^G&wC8ifWfr(QAR8B5e;t5xqKo)Q&aNjih<*L{7KE+D%~T9i=ir
zcUFF9-En7eR1<q>l`$g96R}gR$0l!#CEk5S^Aj;={4hv>Q<~|m5WQT_!{4NLobh-n
zwF|c+dO@EojcE{ng<b^pXerZM!S+UQB4rz5WC)Tg;GHK-CtMiz+C~l)E9xe1?ZK5Y
zUmTbiSyyy*@-*&H-qb2ZS5uV5baTvrt&usfc(y3f*+__wjK+*U3t(EOsr#VA9G(tB
z4Ah>n3EI%B5i&p%RKo8ng!V<{fOv2;Rov`}h`lQx(n^L$U?lekqU~&(AZOy&FNcu#
zo>VuCwdC3et>b;9^~{nB|MU<4g<p;*;{83wI(n<h39tv^q|y~hB~_#)xJq@EZ53Y%
zv-309NqHU96VEgJ^FRI<w%Zf`^1By$ZcJ|{x^*<?$)$+=t_is>`p)%s<?;cpYa(}%
zwLA3~=tdJdzbM4LcWSBqIyN-Ur~)f83>j-JTUuFCqV8~h73q2_k_6iUkL*`=Z*+QB
zUP015R0{W;GQrxTxCVI^Yu^zz**lyISW~zz3r|l!FrH-edK*2X73WxE4Y%8JDEIyk
zEe2!`)<i3st;+MnQZqQq5IuhI;4Qf|p3}zjy0Im_C`O<;>m4?@LwdJA)WHWruneQ*
zqPyUzwo_e`P>?+0-HgLhcEZu#iCBjZk=!%3TkIGKcPHZIjEyoMZh(s`yJhe=42<Kz
z{rw%MX(GmvVVG3zCbB$ZjKtWs4wOTBA~U0?vob^8GAXaTyj)3Jk`aJ`U=5>nOil=R
zM(34Vs>Ei=xv(u8>oT)$D}IVH=iPZ1-qE}v^^Vof$_GyzB^A;K%iH&F`R<!<`ReTl
zTCe1_V45YDOmsqYGgWo}rHS&xSZYbU+^)RbW(gY77~_;PFV`Ebbnz;f1d+(CaW|cb
zVIp{ShC_S^g!Ez#>sqIt1IF1SPr-Ppj2oKET||9Rs~VI@yGa}(7LnulI8(bgT-I8(
zQ<d5N?up-<beat@BU-nOtOUJQpll*3wODMgIOjNz6Q^lpO&g)*KkLNyL)m#JPJUU4
zWzlrT^NBc(;&M4g{21t3#T#ekaPdLh5UtUaJBd0C-ZKnhP|^^uh+Ha36&ynhvbna*
z=mz2tIgKN^6ox3A6q;Bdw`OQvxFt3UuVo01GDca4YBeodd3PuZN)b0gv-2NvmGeQ>
z0=V7vlEX32V6nC-uRssNuEsPNIarJ%`bf+5m}R$K@gYJ8$^q$E+p$(qhTisjYPr8$
zu)~?#mWbZ6tt-nSS@(M)U}=qD9fK1zvfC^iq9T&T(=-viXt9ISLZo*}O4O3DO%yz1
z@JulfTtM>XBi@ZTKL}r<YWBQEJ2-O`3A<UB)s}ph40kPTb1v+P0x6lmSg%dDqB8r>
zdlv<VqBN~nYK4M~(1KMOS=kxk0u%4v6nm$Pjt*+CxaRTJ;JzTeY3V12z1DWN@5P4%
z6iJz&+=bdP)=^$=B5S^NwoPQ%!F!+{o+qVNZu5$Bfjzmf0-rv9;{5%e_-YtLT9-S8
zjPEi&UDgdMlIn+WwT2Mb#_{z4HelKwVJvDJ(>e*y_uoHp`{5(=T=~ZzKQp~~C;k^!
z_(UaTA@a7ybt}|r__RFow_j&IKQGKFBhkWAoETIs)phJ`_YswF+7vec10hI`OB@D#
z4765Qt~Z{q3%7ZdDEHhB)Uq2(_Y!=xj}NOlfQr?*WaK#O8Do@D(hxNhMUWyXZjv~M
zY7U8vk2631@R5$^`f_C$Bhgr__f%`SU2m_BkN?LO13K4iO<BmOT+DEe19zjytb%&F
zueoqtH*yw|(hehbYZSWd2TUvmCPv))jKSck-7>erx`?Wvn~w29;@Gx=qZ4f>Adq{Z
zQLXry+f$@RPYa?6bs=DLqB|>xm%@bOILP}cxv4Z+ygm$oz`xrLYXV!!<klq#P?#p-
zELA%xi&U%`BUa4FcZ|m1Yo+EyS!TBDLMs*LCEd{jy*5fpq_jyivC+=LbW#<Y<w7U*
zMnZyLg0q~?18={2%bPcExSTJfobVwMY{p_Z#WO>U)LPj}rnDm6yR(!cTGe@3F*u?P
z)KW-mV!qy3=7m}cemr5Z(k%y13?m^7I;WK4s409~Yg0Sqo%CtdS@3mG+f1YMB2k_>
zQ%gGv0<6Uko`9vaD*yj5Fb(3-Uv;i7hluPLLK2hrD7~?)D@oHayRzL67c)B3IB_~n
z3}a+WBO6^pqW1Z+FqQxz=~z`1)ZQdtW1283F^r=lM#q*VeKLlqhBdM|uPMp%<h>S1
zk<;CYVGPvXbb%L)GYrwuy1^Zz9CMJg#xXJsk+j}emdcjJw4_@(j9qY?rbxBYc?wLZ
zFu1V?$3~)41?n!Xv*mh38rLodR4r0G!7d?|U9j)ayfN(d_|+&h?gHHfXjZ$S3P#|}
zTBSJM$$55n(5w|hIfzc)TSfxCjhb^HT73Dvaws&6@(Y9{>W#QSTBKw1YAoiB7~AXz
zv$qDLW<=f#Psyl+4J!$by>yj@37<p`%<?*8S8kM?s#+Yiz3_pQ5^EAGWrG!u1t<I&
z?>%@8#L>e?{`*R=0>$Sf{vS?#%dL?j)TrchAIl=O7{?ZGthkZZDyCT8tLJ_{5Q{zB
zsUfnq(mJ^`auFuFbexsk#f`Dr44*y{?+uS13kC5yan6YxR@Yyvovl}%t~2+SlX&r$
z%zC@=_@_Vd?yGOe7+H_4;e0+w$D7G=Uvz|FjA05+er9yUBfHCC(1h`%9{BWMe~aB#
ze)GrAJZuG@D;4$c?@|z;BdDh(@hORY`_I>fho>8#ud{eX=vt!-5xl7F^S%LCv#S|)
z3u^Hh7{-C~<-+MS(J5@p3%BQ)+nU5GzjWwIf;F&bM{taB<bYqcjd`9~mW1t=yVHrF
zqepub@r6xQ3wyluYr$=JdA?EG_dJ|tOq>YD<6^{G&!(vFU-J5i*#zut6unfku!FUn
z$B{Rulgzz?R~S(IqHEJ6Cu^|YixH1AvLL-xBM*7UR;`4>m`<nQy2IH{G)QFaWn?(5
z<Wx9C%l#OLK`z`{rDNO)8`X%tq(F<6YGA8Hc?Hq|`rs&~@HpRSPvRIj#K<r-O3Q4e
za9wBCl(9ITUtV}xR<&6b&vj?egI22I?nL_7cA^kcu(VRxZZqp`#<s>edfuEyrWlx}
zf%7=xjSw%J>4%$FN*lRk=53{0czC(-^fY6(A_v-eBn%Pj9W+Pkk_T}aM&6$98HPx%
znPuBp(nkM6mLy*B5WF~l8K5R10$r~+@>V#X8*APe#z_hdA8|HF^h;}WU3@tgN-hNL
z5^8&8eOqG`m2QV@rz4$3%9(9Tl$5biI4Mrm9mX5RY2<u9bDB=9HM8eov?k#~yZmQM
zn{0R;=~|6F<f{9@Xsjg8jpNAqJds)>vqv%8qafHb=Tpj<X0Xodb1W!_^%5901kYs}
zNn7VUj(QzE9qGWgnmNT$+#$ySyRI19F>YrYi~FVX(jj#Xgvv!?TzA=7a6-U)EQOQ{
zOA^R5cnMZA&Jf4Qh>(tAh>}fk7;QD(jP|*-<cp*x<A<CxFSnV#7D18_B*`3p0!4{%
zQVbZQhv`9Rt2ig!dnux|76qLAtm}H|Mh?)f1zqi8Saw?%OqV3Zl!RBY!|U|@lgl5;
zEvOQDRqJux^0wg_NXts0veV>iE|@B7$vU~OxoWNg!q@Q?YF8dWtIFQjL{zVS`?_Z6
z?F)xSbu4s_oMr5`<#H&PPUFak0VhaAw_*%oK^#9Dq?}o{EE)AF;W|o4m4Z-16LrL;
ze1(n}<|y&68N*fz&N&7r`hBN<A)xol$Y`Q=N#;vMazILMlvd?A=#6MSbE|x_U9r}&
zR?Fw>#w}^etbSi{w(KlP_DSchICUb4m7o9gC%*pqFA4YWSe8W18B0^Uc_$nm-HU{M
zuZpWCMNw--Y=W&2(w+AVL%_7o`s1H)<;rh=e4sXqEtO`ap-d?q24@YJx2*h>O4(+f
zpP%{kFzccE8YkSEA|9%U+4&e^!)}i8qz{oeRK~e6jtfwzd8Mp!F4d~zuWr->%CHkO
zTND45S}R-1IzrE^Yf`RK)3N6+Ij!pN@&!l68udYty0p~F?RKNIju{)l4Z?W{o-J?x
zeK-TCRc^OZW6ENe7~d(%q3o@gNTf{;$JPv{TOAP?Y}XFKp;oRO(ix`;gjNN@&c!j*
zKpYG)2+3?X2g2kSbL1Qy?=BPGTDH=;tsA*p$elDMZp%uujZnlbY?%`;w@skDR%yKe
zGK*=oG0zLn*B36Qk=w0PontAL>pY8;tX3XgUU<GO0zbCG(|nT<B(3tcTqvz$Tcuk|
z>GIu5&a89d<z?mhy5KF`PXnjZ#Q8i4d>lO1Kr&!krKN?kZY=9256`x-=ESFmXI>t!
zO!;7$9BlS2Z>)Kyw8G#0^}pg*@6U7^$&Ng27jD}^ZG}`eYArbLsdT33iNTBaz7+z@
zGACYMZaVuPIi22c4kzy7i7||%UP-!u3{Ja3Asw~cXt~p}p;mz;4MxIlirDs+B7fF-
zC2g{R<K*PK_DZRR(mQeV*wNv}NX#hQ?C?$)$GypjX^N4-JI3=wm;&RLnHPcOf&u5e
z%2J&;B-#L4g%28mJ?J<IeJ0j5$;O8z_f8Cf!ADFC#1O^NFa)eiic$o;4M(?)F+^{)
zJ9HxNbO!4>gA>R(H-RiGP+DP2iCi18b((3)U^e2l(HgfUk&A5f&Iwtlc0|PN1={ve
zlp8gPMe5Gt&$UsEtXtb)O_@?F+m?BJc%~RBAUrCWwXYJkFT#$c*m9aNpd%g^lJGNf
z(k@Oq)3(pgTl)e@5>NHsI+c#Cj}$9NT+NMEvlM44fvv5OY9n_@y^8_YEZ*#NXk_ti
z$XQ2-+Bdg;xSU!8%{yxFhhb4~9eY~{aXy?iYpLX#rQlPK4Vftm%3zu#-@V2iBk7?{
zke;m+jIoTvM7}`9VFyDi9T#OCl-4oblB(gF8n-#KE*oiGWIZWNu7d!aW;xI$#+QA@
zZ4BM{*YSffs?T@Yh)VHdt;INTP;C19w=4%zE>-240%c>NY#To}E89G<*2>4*!m>)9
z0=)?uYLv{{8@7AaEzwJ58XZ0iJbwPj_V^ilx`PfmS5Y?AMq7onYt?SIW9ecz!5(~9
zYf)lxAcROXo;)Yw`h-nSeEd9<GMuJBIG>;z)-92@M6`lpmMR5~M@FR`u?XI>$ojfE
zV|E=y?nsYO4u=pt8`XAP6Bd8(J2$R#TUY3Xo>$085AB0Cqp|e}zl5seCPV4sJ}T&p
z$Q+F#2-dv1`)cKmQIt@|J&ooK)>Vv4(8YG!<V^61>O7P}19JJx3W1+o40QT28+Act
zR1K!li+oDw9M;;SD{7{S2fcO*y*GN56hML4tIF)u3RDruR-tc740A%9DJElw&go6W
zJHy2}-oCjoTF>>CS?0!VPGkx%Zc#nBz^m36Zr6>M>&ljrKySLUg)tI?!<o*u&dl>o
zHr<?fURPNx2DY4eS{72RoP$K@uFFC-0=dm8an;NLs?0j4VUUO1c$Tg3_#!3|!8tC|
zz^F6i;G9CVS#;`UBi|ObWn;T7xNgX~kb&FFLYg-QH^_`XNeuJ0Eo|FLErtK^%U|$e
z`;q+d2mE*;4rksC?=fv4GhnP^TUOa<O{ZH+h(ez1M#7E2N`6AlI3E~X<QzuMaU{Bk
z>lJS*#ub64?4AiA2v5n4m^(dHdfq!qnO9m2y;Qb&;dY%5w8?ZMon$VoNjMJTuI1F3
z&1><dGBM{wX6RJH89dV%8G<8gen9Y|5DDJW5R((91W5^_aSUT11nqX3%+|{$n<UoA
zn8<15@(}5oWsy=nJPNiV0V2J&BR``yn9gV7C<%hbNoVK0s6dPpqntHunEA@Gu3VRy
zX&O0Ak|3FL#dujmHoPX@N!?q>rV1%3$^maI_#oe%=0%NR3^rJ7u~3y5h{)(lX}nyo
zeE-7_V1t+*RT%?{L!Xt7UXFC>l+0_lDQAwLF12;?w$XaUx-Nx&HwUI|cHESvZhu|J
z1UEV4Yo!-K5n>33>sl_Gct=#>Gt^pPO^Q|+ij#sNCm~K%MLU{Wc4IikTtQ)UW5he8
zl!7(lqfyGM=LGu(+;zN3S4YcNLY~<VU>!Z|w$~k4heGa#Y#e4d6L2_1Lb@VbIZ0UB
z%aT}URkkSdA;hn)yr`}RNtNjLLzpf`T^J#a2@)oSPRUukQF_OPpe%mjIF;I%mz8Bp
zBI9gwAUW@`4$AFD8gBgY$A#a2y796UYAxJPky0#cF4Sf;dqSk5>n5&wrB((ntoKhJ
zKk|#Of35~-ft{(iL*;cZ5Uon2t--ce<kA{huQ3KrW5C=Vv5((lYvc26mha}A7%y2V
zn(xnNdA4b+%gn>W%4PDrIVqYq3>dTM=Qiy4iJ-+Y`ZdNYCuC>(t9nKj@IfGzu07Lp
z!S16qyR&JG5Pq$Y<xy^2?=q_CN}hJcO3sgS;<@0x5_<KI(fv|+P0kpawX`75ZqfH@
zrBF+ulOS4*hC=<Nub-wSaONOO?e^Ekz~DWnAutYJ!iDGu<b;!IxY1wD5VYdyrc+Wq
z%p`iy00<{KSJNTa%9;&6JIr7S!7|=Ur0s-dI8O}5QoA8OdwN}2S0P$7BjhS$#LOmV
zVa67<Y%RUhy~CX%{v=v(=RJ!;wM#CnC2M77DXo#J7%8FYaD(?ERWqGsU3j_P#CElH
zg0sAvCg@O#xaloxp=21HWeO6=6o&wGwzRUf!gaZ^%`<gPJiRQuJkNBiw#!Q*&4peX
z-H9dgwyflJr5H<@7k+W~o?o6jb~DW89&gW#sW8+ZISm&+t%h7PrEAVZ7no3jl{~FV
z`s5Jg&lpF%^$b2RxhPo&K4NOYc1tr}VvDU;Cp#1h7FWGgDr+g&UI@0+d{i=H;kpR&
z5xvK&q0F{r>b9}1m37IggI2iCJA(G7yU&ZmVyw!3G!-#=C;kOy0buYVPpeKH@|wVS
z&WM!I8F7VZK&_p!Rq`fez3#>QsHx64_&{z-0Lm4rbf2Y2cU3!-D%QyJvWf0kjY_PO
zgH<*@>5y;h#`E>6-9i>i=I9tFDX=@4FbHAMjgB~aauxoI36^df-F5ICoJeMkZmeRW
zVyN1s?<Ou*+0h_{_jO+R!}lNQ$OxvWL{|lw0;6^b8ne1F9x=WMGrI2gzUHHdm!Ki-
zQuW~6EunYv=MKN`n&{p-#yM6eYKqnxxyT5rtAsNveA>IJL87OXBK(1F_lB6Fd?n*{
zBP64`eL)(<n?%TMwRX0gb=;-vUL6jN-DPC`NOJ_?Ksf91O|H3BMO8P3iXoN8Ry)@%
zF<OUjx(BMXHEoivSLCK)oTs!*sUjn_#(@!4gdiD+ELlgP-57i`dLXI}-Z>e+RpEcs
zZg-IZ-ANEw>79yE(w`J3GM13P4+esR>s<K$VdiNryt!PMf{_0{K0LGTa5jc539oCK
z28<>rTIYFqeBc+U5H2UWlf9*sOxrdo)RbIloD**WMT_<ZplRIr<l)YCN}K8Fk+oSq
z&WlK}he$VuHD`ul;xbLNlKAoQk$GMi#)*#OHm}qzbDmCE^CTl11G_tr-F+jvmCuEF
zwcJ)t14f-n6~gabY=;$=0pqZ?iQmP}f^^V~IS8Ds%Rcdn`U}Z7gouw(_-)P|91*dj
zH^(__te~e=$-&-tdf8DtF=G>jm-^KFg|DABCeTg9XxNNY$q-y%@c!6jKnQ{bCHw^^
zvXHK(5!F&t!PrLKDz$X1e<ip??+MXq{Gz0g-PTMojSzq$(w1f%#aq_i$sUSpR5ZG@
z2;}25ZdP=%ol>M5GN`lK6k!P1(}?k&+TmIYc3bGEY5+Mf)>@QpDyeiK2sB+=!?AIW
zk*D@TSr)AISQC_mF7msQg+1S^nD@N7oVdT7Wwz^JEs2+H<~Gmdd8My~$ETS!6}mUH
z8E(%jWzNdj7ek^}Dkc}Y-LR$d&D%33C1w}N#uE#yc%pHPwGqkuaem-szH&<|0MjJA
z1LK9c-Hq@;?xzcv@q{%_xByL=OHGJImPRlR5t7M#S6sSOu3lwi;7aAqXn1#u*bsTR
zEj;8-FOXA{1Kx|CeVaFGN@6v-h<E?a%(V)E?wiow!R@3!A%968L9eI@l~@H^b;FZJ
zL`bV86?z3{1*vLEY|FW@&WW;Bs&%wp$Oug5y(E?O4s|PXFN~#W;gf1%Nr|WHm1Rj_
zWF%(2ghGweD4domMPn|qF*}`2XEHW8!swM8xyhI<SmGpp6+Se!Rs~j6{(-SlJTJ@2
z$p`Z3BB55b$+*R9#Bpudw#(UsWOZFOetdX()tDQnZg0BB8B5NI-8@Gi<=v26SSVEp
zGMFw2n9g+3^E*iam9=9SCsCuAE@3XY2?EoUlVP1om#Y{LwJwC&s!~A%hs=z9kJL`e
zhA5w34@)q#R@HE8N0;Q)bp!W7g2apy5@pAbQ^p{Lf@BgL-s$YI$Fe^-=^~Bj<gM<5
z(q)F@`o4xgb)AQ6S68w`M;plssCDP9yrL#v3pg2VHk_5&zf+CA*Thfn#Sq6S%q+8K
z)bUmJ=Ad@z-56pN8FNRB%68uYgQ1r~uSK2#V@O$~;iLBq=ZTau^}4d8ETh40=*E$9
z<rFM2O8D8+(=&O#F}-~c&XBfE9ruhB;oYd2k4lDj>s_SaY+goTZ_fi48ZXPrc?^8N
z2$~$c5Ko;Z+=B4Ua@i<p!#PJ74C^*a6W2TAAdaNXYB$@}9Z~ry2ZCwjGroEpD3Vc%
z_+@layU2)Jcr-o)LA=x$?A3vgBlDzHbyGA#dhJeRpdrTB^j51Wl={A}mLBro8Rw|Z
ziJ6<$royMG;!?PC;&3}0_uqW|q+-AsD`Y3l1gCbQ7tO8@YFUcNOK!c8dy&~1>LK4y
z)~2xJjTK0>X8D<X<Uk*V&F}o~>kPGZwzX1{5caYwR9`vU#9$mb8rtBnV-WJh<Z+WH
z41r+?^e!_PN7D-`D25pW-FpZg8=_R;){<JGmMj`)FI8f2GGn&Ea*6|Y<G?gUERNPk
zO6fe0vlvpe#xM-BnHwo`N-m@<gskWt)8L6NC;<qzn%L5Igqy5M=8rjxLtCny`L<Hl
zg74z^IC?{L2J*&MGHuO7(@6eAvjeR==IX?uugT*3DptO4jo;s1Shp41J7;CrOUO&p
z>bD5%K8(kAsHIT06%aEU+a(q^0H#$eFvY0hBC>Jb`OLcs-nb1{UNA(yesf1zPyC~i
z&Wa`nS(__y6v9`}71Pj+i9*kXu4PB7`}BD?^%V%Wl*)Wf+-^xpOARWNqJ>~p8{XW=
zG}cWN7Q4+b6$!M-E4O(jt(oL3r9o~g>C;roRywuHvsp?bZIzY^WliLH=IQ#vk|f;`
z&1Eo3`pN}TZXzKyvI$Nhl6#_=jBy>jrO;T*#++_6quOiRVb4~vuNzFdZOjiV>sEMn
ziQ8!-r^>u#YLkGVZaa2xxY99o_ZW~OU5t|~0g(!eRyoj{K%%}oos%0qgkL)*?<C}B
zKk$qZlN{&8M5c9Uxg6|lXY2v1)9PUXD~Ez~Mf(9-YSS)EEu8frk#1IOwqqQK!-yZp
z!#22<in9{O9fDOw$v%3N-(&egqSrgFOJU=@I7WJFl|X1wQ9f3;&5jj$tsa<G6(iou
zwderOivNaDj+JSL8Cuubc;6vz*D-#OLfc8gX7Iu_F}71nlJDSlcg28@15Vxgyedz6
z7iUdLX;hOWC&_=tirwulQ^pvq-N-r1_t!A5+q%%tfwQX&-jSLx)=I8SW90rcGH;2g
z2Ig(n$YV)Ly`Lg)$3Pw<%VJ26Kk)7sf5q$~p;fjl+InN<S~{oaiW)YVh7>{F5Ut^I
z34HhF#QMYU>D!GDU%lh;j}KUDIS-z<mov>6a6!BytR-~=2x5c_yv$er)dw+&tF6jD
z-kQwnkwQS`{d+ElJ&LBin?LNOt14lP=0JcJI3Dl7cXcVG9h|O=Iru0=n*b2UPw%Cm
zAL1a{HhL|0x88p2a`gxTYprRjW+UaDa4AKJ1aN6#`WGq*esVE@F=CLS^RZ52^j=a5
zbzaqxDst0VhEYpIwdIy-ymf5tm@1uFE1hoTsWslpLg`gVKEWOpTPcOyrBXDlVw~f#
zU5VaPdSi1HGgw^p_`zYlIY>{vblJR1l}<W{HE)bjV71<92BP!IHZClABWYH+Gcx|@
zz0-1KT^Dk_P)gx4Mx2YBr;Yp5nZZVCNxZ$k=i7Jhxj)~ttXD~G%vsGT<RFvn-R?U(
z%@`$;30Z2b!Z@ZHf!x<ZO+~_Toaa1Ecw@;avsSV44bz3@Mrx0=QYcx-O3ehW*>WG@
z?>^k~<HsNPeqJR#ahhbF*n}IgrNXi%-b{A~V|-ax9-bfZ)-lwH7^8S$*Gemyni8fp
zrU0WK8Gw^{=IwCB;>pc(YXzSchP?6XH+TH!%ZYz7hO|`*h3Y%gUBoMaPpppBezvqP
z(pkUoR#(<~EtR|#US3un9v7~+g(a0^MpsfH_lhe{I-S-?sZi?9(<m6@X|<A5Vp%uR
zQmJ0#c-_eSoUZ=j-KeZoj)>*9vds(YeB*k%^6BY?Tq@p)1+;YBvbm&Gj&YJRj&mHD
zd=%(ZJJ{AL-H4rVS3;h3j#}FRS=J`ccG?OrPa994SJqSsc4L{OlTT|VC7Cn(;M9&<
z<CFCe20>Ab6VDF2d&c`9{_S+SGxR2NNmJ!#yX+3@s9p8bz3QRg6}_^U>hOL~+>>t5
zD<MkcjxzHRS*ECq#M8f0zOWrOYDX>eWQ+HaVH^qLiFlguVGwAv6-=qPCf3Bk2|9-=
zTm;43-Qt=SbVaRh3)Xq8^M4+tY<E|Ajsn6M51M}vwX<`o%4s_v0u)gYzQMl1%a30-
z>s`L+j1^L#?Xu2e5M_}yGF#qFr=09xcK@AvNYKT892EJnI%ZP{9;Ye=Rja(B6(S3E
zA&^UD*)n%;2fUMek!xi$U^bb-ygiKsFXN<2=k48@%QSE|dEVcRn7VO$G31vA`ty%)
z{~g^&YP+eEo+ShwpY8F`M)s=RaOwSN;IF<tbN%?4KR&z=IuHhOx13HVVhntEbLOiL
zUvXOs*H!Lst}+^~25$30D~)fy{))@}1^e<W&Zs3TeBbnVQx1kc7soTFqb;$D{^EWa
zaSoNv-4eVZcrh3<-BFvMDZNXf-F5u7yWbk)2!SIf!mDRZ)v=gE(Jx*<$G$FJ?WGlo
zuboE0co~6PYl$I}OZkhB_n+u9VDteN1j9SWd5oMV37wGQTA;R+D<v1Lf}0cr9jR(S
zJkgsR&;oBLGglwVS{gk{0L)G{>&0rmd}EkACU`d6$Ye|~n3KWJhS<bL)buVe^I9ou
z<MwnTty$n4^`Ec3vo)D%_tvP^(Mw^e8!u_WT1RW(jb~jp*4s+DEv&uq`1FYF5-A*l
zX!M<Rw6fxj;eH&s8%Iv#L~jX$C6`9ZO*;KjNTter&Ig=AuLVe@k&t3QBt%L&sbc*q
zDQY5Jx*JD4mZc<4Q4KO|q_#)}a#?A0Aq*3k$lOQXhQj~#*MGx5{q9fF;aDlca!$<i
z%=P(3UJ~{stmV`ypKs6Nfk5YNyXVwSqW!O8xRlc(>~B9ZxyoBNGr2_KLT#4Xvdn*h
zmNsJg#IN36_@A8<c~TehR>}3ShRsRtcg>Q3X^M2w0;5THVY;Xo%2rutA&EZEnPsh{
zs`I_llqFv2Ud8}D!@4Gw6m_T#<2ofL)^%lBGV4+)UNrs9A%$xfSTvU^85e6}O$}ad
z%*)KS-q^N<>ur<gdneHC1xqSY{2YN|R$0v2^$wBC-Gz7eZ*a~qhmC1C@oZl7Ah)!X
z$u*PbKz4@tmU(<wxxOTsgP$v9)yV7Is7=i2oSI+tW~qqHGkP#PA9Vg6nZ`55Y1(H8
zOvBm;t&#Ia%^OXjP`gVE)xoam!L!qY)DjO+1fmpL<nZCQKHzqVp0UIj=~z<EqReq!
z^B8m%qOfW3kr*ej&7LNPagvVDTD%c<zl^iM^(Jm_y<@6Bh>%u0{*sGwB9OVS6SAB0
z;UMN&i(1Lr*IBzpa|6<i>FnF;A!F2ZYA;-6|LlrTXpY1eFs{{NNUSZo9yURgFa?YU
zr{fXlJU#||h{t;fS|TyeR*{U3cy1W|=fcS_$M4?{KcmlIQ_j!4Z8QSY5O64e%UE*F
zjNWh>f)JQpXFLtOe>?HbhqpK?wimv6cjoE&%x&5D{`bElnlolROXshn-Z%^x?=e1z
ztEd5*!F$8WS-yRL&;2FvAOGPG{NXY4Z5a9ObLOQ6;xO^{{=}Qhh0AHg`^uVcU?e{N
z>^!$Dl1Ae!Z{NR{eX$$)oxL6r)kj#=ml8hIAkTb(c)=l&H_oZApeR*B@C?!7x(;5_
zLaV}g5|z0W+D0=njI|ho%B01tYTGtRo(usm`cqNDAY3Pm`LoFDy&%|mH?>y6ep>4@
z@s|w=equ4uw9hu0)r_@_!7&CeTmlWu$hoklA}NA;fF0JSS}>T_#YIdV$o7ZPDRWXt
zUIx%k!C0C3WJ4Gw6e32Ci#k6v6`aIr#->bhEM+ZH(cU)Z>q1U4<MrMttG0?8mI2=?
zXXkNr*1U1uW^!-LeqnTxnmfz9kk^f3I?oS{oMxWF$T&=_TjlY3V}4l}Lm;J1G_$SX
zy0~pM8A?hZO_uN`<2*x%m<llr_*B`9$P{1Zh1vy}Y$~(b?P`Z0T#d^#;Z^5sCr2+y
zU`8`wx?D>ma0A^)b-zZ>2UqzY|Mutnzy9H0*tU&LUBRB89(nro2t{@57)q;L=NBGs
zFO(*P$y^J=IMQn3w%*v*g<dNs8t>!AB?fL~WX(>@j+%uwYO<Ec=O;c~#4#<GMyj&0
z)Y`x~oe9h5Z%up*Y<D;m4M1S!Zq=A$O>A4@x-7gb8)Xv|rZyQN2t-#PdveO8QdpK%
zh&Re2x0-{IOJ-R!DK)mO$a_RDq(y~GQ>iSuP*P^iowg<Ma#&}!Wo4a{6e?YQe^XPY
zJs7Fhc9<w_B~x@fb2^>4+}&}x+!39l7DHJZ&;FU!E(Bv~rP5QyY>nzN>n(A+E~IV4
z;;1>3S7G4q#6}6A(epG2*{oK9%;PX}I$bc<GL0vOaoXbvB=<v;>#8#CooUpv`|vjj
zkkXOE>!GA@s@*vNqM^>Iu=zyqfo}CM()~9$VS9JPLfklq&7PdMV@$A<%ZBkp2-9Ke
zG!6sShu02XWoq(81PKz2ybNIX4&Am6Mm+yZDPSBv$nOzUU{}FNZiaJSiXMz6r3!M>
zkz&gSf$Lq9YwWyJu8IzKJI8)^AQQD71%h*qYK?N6z9^E0D0^ws>ufa;ZTB6ZD>Ge<
zbgVkY)_NRftrVJ5BAhB3b_1apg}|1Jbk|!hOy@Iq7bk9=+s2xtW50`{hMHo)UQV1Y
z6SX9wYrMU`@ci&7ss`uz-H%T^x4>KOu+s(G3aw<^X(Wu3<XAY5a}m>JH0W&OyZ85e
zdl&ifhd=Rj6J^foPCPt3vL-|6u&jx0Ec2S^CNKoYFpk7=WNMuuB|bl0xtu2aIP$zy
zQdi{5h(+>ozNsI^?qLEP=M_dWEwHLu@?P9@K@RsW4UAVJG`&ekSQj)($bb&#<=V54
z0%&5#ys0B=&7#H`#6@!mfw0?@?`T<Thr!#vrsVSrxwR_!_sAak?@<g?WI^;IGON)$
zChr*!!xn+Pvl=QCC3A_JS(i&!W%g8C!!(WpLQ<Q=q08)~3mmQooRb@GLcn(jQH*1}
z5l{I_VcU|V7<&0t(lvNWEo@t6o)hydG8(!4vVdD_Y-^T-pd?m0gV&j8O7sScT__}K
zbEHMIv?Z++Tbawo&D;n+lB?nId1YQRPF%-%cRFzw9P{nQZ5G?#)?T|3F?AIvxETZ1
z%Y3(XN;NbbLzK-aXOU4E)5Se;5c_6v-I6C)*zk>>6U{~FjaF7cTx!7#5$g=8I83wr
zpa1G-%&GHFPft8NU2*2dA3lEMPahu`vDjAVtzb%LUKf_MGB~UEq;Tnlnm1l<SJrK1
z8XUiT>o|{=V$N*6i5EdqLYT8)2eNgnrE?eH;ye$^{LUC^lSS686R&jRD5;CVQ<G7F
z^T;8i(xyEmsTD)Mi4q`JnJM=UwHI=;VivV!(w13rqGdS@rNEZ%tu{4dNrhfAC0Qz#
zrb&6!O0R-o<f;K8z0-0bFDq%z+~$q>B{58+Jn}ljuBnK&TYWf6L6ak8Blpg6IbS%B
z7lwEyIL{a|(MKs1jdX~4Ev!po8#Cu%sku^e1@Zc#Hp4A7)|6RG<<$fi!Z64TI>;&z
z2DBrD0qZ=&=|mU?AyXDP)QxuR#Q;5s0Ztb*&#1D%*m{`CWKEef@?q3$#T=V%ZI#>v
z<q^%lj2*2KExl88&wH%P0#$!;SUTqz#}Vfvz1!E2DUCMwMhlQqB)<S5iI&#rMz{id
z$5c(n1+b;z6vnqsKEn`Y4N0*eq(B#f5IVb|Q!vU$+SM65Ic!h8vkD#T1TrD8YC@n6
zfdDPobe3wIFvE2{`)F_&Wv1&vWev4g@%PxfPu;rquAkjCvCBZRH6%vf$~vp1L6MPc
zj1d<iEzdyb`==`p&x=H@p9j`F@bLJ^yl&jQ<z23Pc=ML~OXS<H-;h$mTQSyg)?poN
zxvR9aa{c|k;GE&RyT~{N%Hpx_K46D|W~}hD>_F=ckMN`(URHj5sRWbw`t6Cyb<Xad
zb_d0Xqv)C{%X%gF!1?|THL+pzj^*)@(i`W~D1}4Fq>h;2dF4%@j<iOUCS6_sdhZk+
zVO8C!Wwa3_Uk62dx)^SCqhlc_4x7$8iXvI!3E~Kd+Nr7|Ctq~;C1sXvVY3qO_2x2x
zjlw%oH3HUvo~M#kpu7Ynj5-Z9l7Chyn(gqHynY&i(2Hejr(>EWT2FLd<XQVXv{q`a
z<TZ)cy?*cxs;iI#t?Q!0YUHo7V0#x0d0w+-dkbr`n=S;m$%|!ZujFJfu3@bsx&ea2
zoS+n#hKbWCUI3jNFP|5ICrhK-SKy@OLS0v366Yj0cG3(0orSd`ffu{=vI~xBFx-uS
zG0NO_-V*CvnO_p~TIfCFTjCER^z@FjtbG3TNJ<rJ1i6SoW`Vg@Oerj-kc#+>?P`p1
zik!wk@Ia}|O1$vK5~Jwqo9X1<C{$jSSxt``)~UR+SGM^^j3Xs@te5<QZYOHJ@qhof
ze@p*Q|HSVeU)Wma_rLp*Pd`5K^_x3#Y4{{_;&n@;l$dWb4^B4LB`0dxc)o26ljDE-
ztMB;wE|6?wA(HZ{2vE~aQxeHKkaNN#jPpI|ZI2k<nd;&P;EvAPI&%=Mo4UxMR>ZX^
z2St&qUK_@XZ$ZihRwD)^&eO8+0!pa_?~bl)_ZZl=!m@1ICCH+yxsXy}%SAlJv+y@e
z6}@oH;^Cgw%yL~>UN)Qylq{dEW*HG|OD3&VSl+$SQlX_?<zn$h%%8+NUpjIbl{HN1
zl-!ut#%-;fw@l7n=KY|=wN5ITd0AL;QJ(?ylu&Y@wba<i6-@w@&TyZThalv$Orh_D
z(%O#Be}|XbzOK==;XJ!1;cgBk|9B@%?T5cPiV5^H8zY^K+&6^a>%6}A>!!Br`r{BW
z%HqeWY5<)*cXg4p5>6%aeKCb<`T*_EnRkZT`vDir%wJOmt$rt~900y>{ewTByCW!%
zHLCOhq4GDK)$6=|XR3=srqv|PSJ%1y;8qBMuIdaS**S9GvkcVppw|P!7Ck!0d+mqX
ze%*R+f+{&VTw3c#A?n;7k0Wd3<KvBg{{E3SmkaOj&pcc=^7+KNZsgMW_%zdv=dZte
z&*?Ps=6=L_K}(`{tXbBfv4)4|Ct5E258vJMH>Wd0EOhfk**<ZLfqM6qG*H1Z_&}i1
zP3QA1VSV7={OW7I9zH<ZMAL6PzxvrbKFx`jC3BrOMn5oLXSOX921!kQd3j+7mT?+H
zu``Ucwjg{LcOXsvIN&)ffLm)&8>JPM7^0G0J?r$@H`>6+pN@0=Kmx3jy~P5PwYbvR
z)N@525E)Nl)IUQAv(6ey9dym?L0HgxgC@_O(I#4obwSm;j>UEo_|7x<OI|;z7zl%5
zJb9Hw88OOX=M)PhMHe`>SIO=cy}vH@zCWZK8_53ZD5O{<n4_Aen`dP=QSwgARE%wu
zZt2P@_EyL?1X-Bp`-yWL83#`<RovR#%l<*7Ta@~V(h5V((ur1yimk;GhA5u?N!k3M
zvs19%620X-M#jM_(y`J^rCNzJwRG;LfXNF_p9>F9FZ}+8&n!tuTVggOgrlt_=DD$z
zqJ%+#;Zti2(evH=3(hpil_&2=TNZZr;BaH0SSwYa?c8oJ<lY&j#FY)gc}#DtFOMQ)
z^ai5G_zBl%E*}2Jzxy@I|MCy~=Er9qAFeFRMm~+)mYLitYpHD8ilbv{BQGmYYPw`g
z!uFN_>6c$~_LbF5tZm|YyW)yEwN>??$0&j!n`}>u6k!T~HPi7}0#;natP|)?QKn`j
z)C8mhY1&P-+Nh~eia?%r=eIOrqGx@!Dn=9oNujwda~h%)8>LbavfZ-ETwk2VWai$A
zlV>w4r==pMC_xY!8w0j<K?Ytn=38dEG<DT1lx)Z;G2a$aDhHt~&zbEY`N?75iVzb;
z6#vS3C`mW~DOb^=S0OQq8b{2Lto6_#7eNb*<{ikvp^1n_^yd5c1UN`&waUn(>o^NU
z6J1L^VCyf5XGR&>){BE?tzCRP)Uc^_K@@j%z<aMno;r~B-OLC<$ZDZvky@Fik=EV)
z8PRJe^$AiYJ7p~r0wy-FuXEI<_QFn6=Y-U(XpFVTtUv4#%dO#z@B>0Po+M2qltRKd
zO_o#J<lTO|)k-Y}d=S5w>hN*IBn|9R^wT1$D5p}+-Rx@T!H7}K>w!^=NF9yeDP!{U
zrhXS+76ZoF*Rg;WiNQ;3INcp+8teth>-}t9ieb{qXf02-%GY0C7}AaN6zKgV!D_3-
z8b4icRNiBZgtA><t_;IS93sE}_{ira^YONF36{V6`B(h<XJ_6{4r=A@9B}@|Zyx{1
zi<#)B3$2Jn`Pw?qtuR=}>F%Bn;~ih047IHEoUvH$K77TyCG+WJ<qto6BCi{_muHL%
z4AwKx3pHonT~3^b$n~}m-nK(VDf|$%$X5xv)B9koV-M4!SW5kZbZYH!UtZ&*MJ<p^
zk+GVQd7Sq_3A%>R9LC7_K*Gkl*0<`}VeRpITjLl3wB9AcTMt_s<et%mKVofD!m$)`
z{YzWr{+AU3@KcL{(Q>&D%FN!`sB(}uofnEmTL>$5XMD?bJD}KI9wdu@-?aNNS4^qk
zJJ>mqTgMF^=M7^N#^%0=oU=sl8N*2T2I>aAF@``GA}(0glI3t^pQ-4Vv6A>!cOw;J
z!5dnZk%U(*Y#gFkof<=LB7uq45C?~kfj9<g2b&9KsMu2=1h^YUzWeH)Z_XpNW`4JB
zJigr6YR2@&0F^*$zo6kIK8Q+T%`2%@wRMGNJJoci)4-ei6Sfv^!=m<<;)FMx2ka2&
z8dVxk5#ue(wlKy~DpDu*qk?oawx>@}H~gD>K_9k-;zqu@f5-pj*T3So-+xl|0koEx
zmm8^Ot{TAf{&L6LVWJzuwrup2d3TQd`?qH%W^x@_jiuC12m_^d()yWNR<MpZ%Gif4
z^Tgbt?lzuEj54KRO2c)7(`va@DKyPan9(l2?F3a_SUda*s{Pl%q-{;|{DmOT;sM8Y
zO-VDfDurj-GV`*D`BAB)Bt|bLTWS*dp;#>o$~1()6aywYY=QZC=K1Nyye2+$k%X0!
zaHf;f#=3|!(|JQHaD7>LdVJw}eS@(uoX+GVg^KfH(VesWeNI`JJH67H_@M8BGcqaC
zj-_=WL^e^xbsg_DweOX4rVC<G6n1P?)~2pu)b-L>AwLQHt7E7t8xHgX>US!q^3Gwr
zYTOk`6>kWEqYuWtF|Qj#*SH}?dUi7xtK`}JP~Vx`vS$c<-<on4Fb5iGoz*(3ePLFl
zkdj5+`3l1@92^9r{DoSiP;<tpM3z?+vmr=j*X9Qll@e?1)~fI;ny8*!_v|i>)>@+T
zA{R7LWDGq>!{og@&t2?>t#`+@mOtg_^7hfH+_QeHjjkOi%_#9%)|2-^*!eUmDy-FL
zY1xat7b3OrC8Wb0oaLLZ-}9RvABd-k`_qXbG}>UeOe1BJC~j{JAqGMm@TT+f`3d``
zLv(!q_{0xSFF0fQo9{mGfBW^fe02$!T8LiIoQKCJuD2JSuO0vPD^?7tit?$n#xLFt
zFa>_MC9Yc}Sch$$QJrq5(}{b3ruvcZ|M+``e8C#ex=9Ah6bIhFJ!6d{=R(6C`-3@D
z1N#XlMWskv<sMUt9>!Mt$eq)s9lau3VjnDxL&`<Y572Rm3Im&?wu-l%RR32>0V`Y-
z=e-!p8NMK?a@=nPr;ei0Du*QmJwIEe)QrgqlZ&ER{~g8zr#BOK@1xqtT1MO<9kNPz
z$+;eIw{!wRrZ9T3b~1+31MDDb^sbhY@_^Pgv*d;=jgf|_hBY^Gk)k9<$KWj<gEvly
zc-DXwLWlQ4mfB-&OUu+&gi)<V9@WS!xI3uUU~QmOaZ@weof_w{9+f0Dbyc-gV@VDI
zmaSzL+eyJ-PL9!t6W8$W!f+nhmIWIF&3fobI~u7fyA$G)bK<VpY{zw&H;0dwL1TZx
zYG)$Sktqho7;!=L$(4>bV1wtDuiWJkJG_IYI(B18^D{3`p9#}M`1QXg4i&e4qV*F!
z1pemz#J~RbJ^%Cn_n#3xyCv01EgLx}F4LK>&jatLfu%Q0%lzuAH~j5ePgp-w?Lak=
zhCs}>ZG~kay*xs%_~|Ugmvgu#8S@1vU3QlPy>(c(Oty|xD!m94zit~-5hY0QJ0VZK
z#5Lm7b{JjYGz5Z|YP|!d8d_>pTM1#14ZwCv%Q#~s<;@ssX*3LVEo^IMTMHE7q0|Ph
z8W}ScLtbT+&`l@CfiXIU^9Z?$?ewRaZ4r<D*2K}ybn#AK6#?+xvX#Qe=Y{j*!dFi(
z_#rTCiL_<5H4$Q@6=CYvQmCn_0!Jjb+nTrr$8E`!q^x}-NI|Jx9VV-4qaC>j%3y-0
zbg};}I#$U!OU^;9>b@u8Otx#5`xngfwyOt}echeF{F1ubFs5N`r|JQ3z9`_Gm4dDI
zcEF$BiQas7M%;gHb+)ZvjNJgH6bUslwgL{{0^%TR=(f?SJ=oPbC0Y}AO5-e0DC<Vr
zR(h?B=ZkKvP9t;e;bG95z_7=hUN`5apGkZ-G9?vKN?;0R2sH0-M0$l{1aY$7F!qQu
z@o;ezN4sz!rKKbpPjpMGjbt-W4iyHVV+-+s*gHW%Wn~=&E73+>J6Y$0k~E7TFGWTd
zVlw14AyW4vBcq#TxiN%9?v?+~fBF~xhkx_eoKGWpYuukAm+*n;9NV^v6_+tQJw5Vr
zy+N&P-S7{;f8r%|{^s4tFTZ_H?uMsX*6Q2y$R9p0{9%Ttxsq(h&xOqzwr)vX6b9Q8
zc{2?ZDw7|G!QyFn(;4Hy<uovbfEmt=U;TpQ9+;nA2$SP<j=WE2&VyxLC8(~qDg;JF
zkM=|MIB!w--3hVNdP_9%h#Wp%1}I&TPF(}WYG#OVqx^no8*#sE>b+qU{SxCYHR3Zy
z_FPT!l`)i2RmrQ&Q6}v>LS`|vW4(afyOIjVWkN{Qzq}UsX~n?Zn=|M85lX@LicJEC
zIHRe9I%7{&X6(B&l2t3u9bVaUwD&K#OK^rUFx`#NV7Ur2-8e(bmTq>>_)JcgHU)gM
z_?~D*yT6o3TfrhQA4b^!^UIa>wkX-jFnEbMbxw>=y4vs>pyWic{UCf9Cy8L)SxU;{
z<L%%jckX+{8_Q}8H)F|O9KO7@Os5Gu4zw6qw~eJ6a_zFQ8u6F!O<>gS4SsU?AO%dX
zl{o}rw2~*@E3JrEziFMz7`dDWhACpbk<AK$*s+FN$-JbE58er{!U(~zmrh#}%e=6D
z^+33@*mb2}?y)Wq!pOh=t6%Z!Z+^qFu3YX;QdH3xoZ&J!JWU8bz46y?2LAn55nET*
zoaphyRsi4WwiAQp_NPzem(P^mxr~!IptVN2K2RQhpxBY;Wsz_h3?&z?^UB#EJdK=L
zmQ_f1DG7|$y>>(Gm_0jVxC{edT~7S`{XKv9{SQLKZ0dvFL<;Mzl|xE1>!TB!Wuu+}
zbRp)=w<I0`dT4=~X5}jO&neMz!5hmk3_|j>o|l&!^P2hHpMGRpHp*5=b0V*aLgxAT
z#?!+KT@Q2+d)Jri%<Xw*JWp&(BCUzM6-;+Z3>0{K%az(xnq&-HsXWgs%O>gwnKKI+
zEtQHYcmZV>H*t&{oELlMz7yAkX`ZT%{A$NViE!TeKzkuzt)*lcL-dkm?u`?vVU+mh
zeRnloyQwB)1n<3u=yV+^R7rtU{}RcE*gLHzjBDMbTiXkQru#})@*^0;SgY2fbrN^O
zTm<C+T+_g(-r24<N=-tr+U0;=#-38BNU<q#=YlW@`OhAb$k$?HV_h~$9Q0D;p=g}3
z?%4PCqWtw*0@?w4Un#}XyQIK+E6Ro@IW&TlR3WtOxeB|)(weHA34!@#6y&71azaYA
zI2q5iO1I#=5k7|UEB0$_XiVoyyqoy?bm8MQ^IyOJ#P=T``P-j=ORZVPOu>tU&O0cT
zbemajGwYi8!{;lXljU(vjNb6GZ$9w${vPWs-+z80TgM+#;%PPf=O1q@rGu-SE8_WK
z#FWd&$PywCZp6faVf5UOkueC$gmp~Q#CeoV923uQZwY?Fx0Q8!p|3OZTzGz3neN|k
zUmKmByv@P?*zYBJ|M&Zg{yWy#!)InE{(-C&B@Y}2vOG6CIkvxYIGXUcdMgLAw#Ret
zqaLfi6xJH{DH=KeY8vjgBV)oUc}3pQ&s}oDgiI@WXP*2;ub)&54AX#(L84qeL>xhN
zy%?#a)>>FfReMVDUq3XgMjqH5;@w|jE+nvbI62Cag+ov(HCbA+bQiQDH`G+9Tca$U
zJVG2c!r;ldkdg*@<cxKe<+`%ItkhJ*W!)O0mqXTDBXC(0V_gGcxIt$tjY`F;rDrVy
zA8%VCn@V&82}5y?rkZ4W(TB9teP;unO5@X#__S134c+lNLlW%JV8=i>1^f_!M%l6y
zMpGbIkwMMd#=K-KhPyE`je%|@X1Vr4H;vi~7{g`_w><N{tVDMqJCBW#;c`cwpSe9e
zaQ*a=H<xelm-qNL-@)(}#ykGqzx$T|=P&<?ZB69bd0B4EDUnJhuQM&J)F$`x?|%Lk
zzkcNU@B`a2Q_qfgo_PNBCw}<xNeYzb2ZFOqZ$C&vVr%reV)G4J<#FEFO2-Gs*eV-_
z&;%VZ-7sKC%S=fdC9MLRSAjXLm5t6g!^sa!r$)OQ`1Y$ej8kCA4U?t1_rWlZp6L=9
zqgckqU^JOds&Xkbg*bS9S*f)$xF8#6lW65V-Ez%|UL=3QT3LjxG_Ko?KRjId;qwcv
zRJL2<_O!9QBzh~nJl?oHFXD^PH1f9zX|}9|>?(OJq_vQ@0?pAf)Fi5xSyKoNG6ty?
z)~%6}Y)boBuCz|7Vx}YOuTyF!;)&MbDo0?>4te+1s4eT+U=Mkxv#;;lbi?dTJRn*T
z=epoLF?vbn3rZ5~T@Pp3G>vXtO?A^vWmVn=jOj|&Q-U75g|Dnjqb5f3QdhVNF6dg?
zp<g3&)7`cK__L;CR~VGEkxQZ*L&{0hFJD~>?T$j|2+=43aev>Pud=^q%UQb4gRCW%
zvBn9v!3A+@bqa6476bd=%K_k>6jLG*)LF0>Rmbkg*jiV8yBreETDcgy$L{<s(_cGA
zjNDThvk2O0U7)FJ?B1M>-f<c%w_flB{^n=j@Lzv;;_>m3?af63keZB^a+mR4&WTpC
zMDmu(=Xv8(_S9}Qdi=ukvhmyvj|)6*ndTbTt?_A1;5;QQ+&<3?F4D62W4Is+nwM?~
z@x(HX)YB+-$4>Y$R~xv-0UHAEzWom4Cz?rYSsXLB3b&^lpFdyuIu2Y0!?Rj0SCujD
zFR>@j$6+f3Sv{|6l>@vtC8PHJE+6>;&PXOf@5&wN5>Vu=#<{EUnT+Zx)GTIdT~avr
zGf~#TZk??hD|I^E?b79U9GC{yIE-zywkbNWlKg`4DzEzsUO%ZA@NvMnK<$~sBiWcE
zbfLA%s`~A{8`)jgoV8SaNUdgd@BZS}S@90In;7o{Ee6tBRW6m;R@7@`cVB{a$;{Wx
z&9B7Ka=uJQL6UnXuN%R8mN~JldlH&nXlpQtH@nAh@qznc)Ic0VSr>B6^x8G84f57l
zX0h$Gwo^`?N|h{xBfLpQ61_Qz60RdT7k+qr;p5Z7nlb|ZSp5t##`G~T3<Gf*Fdf!*
z!#W`^O)e0u%%XDXc<VV017i$gOKZBogl2~cA!At6!u9b-PIsQsd!9<AB$40cQuyKb
zf8^&s|C;gJpV8w58%CN9cnbHYfpWTF{fV!8$CQleE3IzyTIkl3Tc^yA^!CE`{J^I_
z{gLro3(4@?zyD``^Y{Nk>5cEcdB@wYKj5_@_ugWAqPI-#mOp;H5kn*x$7%F@2##;x
ziyZTMTUk=4ZVNpvkSBUJq)k}oLmZCzU2q1k7SE@1WIP4pR(VM)xh5G2xXyVB41=f9
z2}O)#jFDMtp|h@ud0u&Wndz-_z6;!+BWr=|4XI}4b>(GSX}vQJqOfUQ95t6^<LPDL
zwup6X-5R&2g_nmLt!1u{H`eRQwuwt;R(sM~8f7b3pIL6J#My2cniUnpmYL_3B}oWZ
zZ8EPnLMs*)vGR9dpEq`#Hw@m>*|Px<S38xwnXVkg_Ugxgj<H&6$Hru=6t7js3HxVm
zRZLNwB6qEdIg{1y>DB62*6{wmG8Wvk2HL?55HjI0D5Z6A0DUb6jBb8PwCl3BNWq|g
zEO@L0(dZ#wnm7^e87U(B($7=Mq0KgO@%H0^edH>EOY(b~oUE@p?wvbOGCg>PVPqUf
ze26j<^&*3eLzLomuTu22&#jw|B2nLiJ^#TO$%^P*NS%_RnqUu4Qu8N~RH}QUem2)R
zq?<JGF3_yN(>{pdkF{10^9><@XS{cerxW*E=C8kcOB@34&L_g?czAeWo*Owe)^)|X
zJ919c-dRgkRHYE(z?*kp5zhBKE;Cz&AD=h6pD@kxd09zW_U49=ChNLjTc+0xIbjDG
zQ>Ld+#Pd67@99IN+D<PUSZL0XaO7Tj|IIga@9?J+A$4Mo_$^RI&(A&xdGV7&$9MAV
zec7B^UQM9<p5yQZioK%%V)<(&d&lTL(z5V$x@5NO@0*+0sSCMO42rA~31HU*(pst_
zkpEj;20_#5UyKQqJg^t~&N*BYnRIPCiBM65{~lw4lpEW1qunyGm_uLO)hVgA#_hH;
z&nvwORKNS~cO&GMia?y*2^?dJbRcH*3~xq;i>EusFwB%ylm%riEVb_z;9%YBoRdz;
z5jyxmq$0gG65{-}+!E`YMB=CkSO+T;!~Hn$^*C@hMRJE$3riPxv{gw)GtN>rNfj)$
z61`N@wKQ@`6eonaRt=_WjBkfqnn}0B4<DbHuPecN#?jy`4Bj2%Dr=1dGuiz?lmg2d
z%PB^jGDp``bj~XI2-ZtnsW+VpB5$ijjfFa^`9xY9W_*u{LBcxHCdJ1o5T`*NDC_Ai
zK$ptbxRE-2LyvbfJSJZ;&;J$EH%y3h;|R4<F=RV3z5fOL@FN9SUT%E+@qyl93=!)+
zxmNH2;|E-r@MV*%g!5aTUtV~B9_TJIjh?%c<L&(!XB)v89u~1W?(4>o6GL_QEDIr8
zgX;}vJ+5bR&yuRyJ8$nsmM!zRc$QRI>2wc17(#HARvAWd^OGh^2x?nec$r_gtt%l|
z?k<u0`-rb_%|bkzZ!@=Lp_YXAGGFX%m-KaVZfd1$1Pad*)x0i==Z7onnpwBZTE&5~
zD_lM=2@@PGONdQgHz<wmc4L0Ia=Wc;N&U)=p&A*}i2!#$KZWnoLA*dt0q<m9Z+d4M
z2ky@UwhDl_M-yWm#s{cPT?&n4O&FPd3eQEH7j<`1Gbrhfx)BFHW4lV|&@NnI*+T(R
z4p>$i3SPseFk;ZMZ-9tJI9?2hY|~8<3#OoE|N0Dr9D<T{;jE|QFh*sS_4R$t&IM}W
zFy%Tdd5_ubtNH<>>48o<_*O;!TNEYHvR*%q1ha)vGX*r3R?KJwl5f2o$%?s%*F>$a
zW-_!+v*x(oty^(KEMgv%YY}51yYm<Xk#GusH~l`Q@9>C>Rt%C6Lj&W503Bi!UYC>1
z&ej?~{`PxrFIU2MUtw=E|BrwBEz4FIN6XhABEekvLtOyFXigX`x9h^1vltdxPxp~v
zM8^EEWIUc99&dExxdcbYaCd)?T~{Id?cck18Mkyvt=(8ymzfG|+e~X4?ec~;j`+R;
z9cR06J3enz3g@5y6$Z~>R^I!O%D{J*fuH~Ej&~P_=|a4<Midrbs$+K9pi{*A^<eHw
zXjRunBeLh<t&VjNgRuhCuFiup0%%?KuvWE|?_hdEhr8kus!9@B$pZhkrU{}g%jnx$
z;VJZe%}MO`j%2th68~(p7!Ww&FFxOYQZbNA6LW)Nh}{rW`sM8F!M3KvvTT&vban=`
zelQ}fnhQB2#wg8z9X#V{&<7KmA_d{P3GAQp!5<K9Wux{^r(&!?mubz%rqVR!u6IkG
z>+9wWnl5D|ovc;{1JPN5fsMEphAJs(!5Un!;&fGZCU?hpN3EGPC)f-b%bdjNtsw_M
zE|pt{+snf3mN?nK<>I7@^oINM$aorXQBVTEORaQ6Dou=qrVuc#5UnL|ni)_tIAqqU
zi=EanrBPQ!W^Cf;mv`i}k@5p$u;4syI&u2&mT$lNK$r$GU~vO<OQ%vxqMDKB?=bP6
zjv*{RVrro~r^GQK`WZi9<AgoG<?^$C#lvmm_kZ{!&!1n!K3coV^M~&UZ@z``LKr8;
zawUICEcL`?G`u+tRO|Wb{T*Muk=Wj?Cax(#G^AXyYhucUu{V4v;#^mx8aG&ST1X{R
z@`ACM5ARPr&kObep4XYDZQ-kWrkO^w!h0}kE>&$I)s?MeUbZWru1`lXa5;PK?+1G7
z++GsPwlemPt%Y;2xbCIGH)zKEt3zaWV&>zmn2?k#c@HTU_1hPNo~WAimw5#T_bEwg
z<GPT|%I*1uho={wuN!F*gs7_bLG+H3H_#nRu?k~KyYGVG?i{(F978Y~lNz``Tjpyf
zw=7-UKC3*K)J8X?FJ`JuMh@blD5S#*vU&bl>mf;!y~zB&>z)BfSkK<VqpW!PtBx8)
zCUzJ>n+anrA?$8_`{A}12FQlnPy(V;a^33~4*g!ORFzSxfsD@ct<Y{@x1#NPPQ9QK
zy*cm#`(vbu5Y)tBjhFlG9i|%bMxZh{!!SnmNwCNMxFe}X9U)5FWt{ssN4|HddsrAE
zV>IAf5FZ1T=<ZXKzB@Y@8O>Uwk?~zrAAL7<>ia&-LFAS_PvW^li1b#7w(_fQKhV<3
zxA&d@@PGOlAD=VdfB#Hb6VJ~Jttmn8a>5k}{@T_X#s<2&IXdq#PR4aLCv1p}QAVMX
zu9;K{>^M)6$ys`<6fYL-w`HYfK_i7|*;t=H3t2c%oCawxw^kUu$5zAHWTv(-oF;~N
zPq?nUxu3Ydi`;#EW*98L`1%bm8$7KtR^2;ziPqM*aphd>v!?yIKq=(Giqqr}BhgvB
z?ev;3uF<y4Ryo3r_Cjy>*4b4bws$EC_7E<;A73Kwr4>T=BB$;KjMy0)U3FhNSz0|`
zoe?Ct*9z`0Weoh(VqmM7YH(GGfWbMnN|axiw#2e#mXzqN<807H*<pt>dk5Q9hNLjd
z=!wB%TPG4|-jIht9)pYuk}lvbY6D?Kw1atUyUcV;Y6=}n(p{@vgMBj8O28i4X6r4}
z5Ez1|VOUFNu7x=ja_QIxqnC|(Q6rHi;DW%F#`3(ZJZ*_;Jh|yAvrw$eL~G4F-&U5j
zFd55-yNOfq7zgj~&kU15{k<!bTOYi2T}f4fn1mRl+V&(Ha_$wpQAwOYm$is@|28k=
zHSyFszq)_V-~Rk-o{H!5rw4F>`>(&{tMA@%KB;AMkb<DJ1})R`iiv}G|Mx~;pD9nj
z!<QRP2}j*{30HEFW=8sO;qtS;;qhPonVcV4QX_PMqf4zU+lFs&H;#;EBR%|)ZUXC?
zIgcZL*f7BP95`PhCIptPQUbI=3b(Z-N=>*v5->vKDjAgQRY1l=y<i8!-F;-7ge1Cd
ziKk^|t(mQE*lww&(T$9{a!u5lSklJx(-V&mFU+^hWpJFQz}xdcC$rc_&oe`SyK|&Z
zC$^lqZk1dO&RE9anY`s3J;?|HG#G<PiL%b5l<~kfcmIFd{;XM+EW7W-etUBFh&a<7
zbE*L-6akPRIErXW-PYD-)B{a2^{Za>(ey1ezDQ;=HJgMug9TM#C}d^jbf+^#xceTK
z9@g3(CqWuf`Je!9W#+y2oD<=`eebpY>px7S2+6~kq6S3`(Y)s~ZCyB>9sziKxMzNR
zL`nvWzU?H!%OUZ}Vc>BsEVXdAWSh?myA*hNHS+p;AVp2$OEEA7&wP?C()BwY^#((>
zs;EE(@|qDNTuCivZ!3Bj8+_jx?%u9x6C-1FD_<jFiFNQ@O{8SBfoS791sQe5>bfrl
z=-AAsEjqfjZuWzrI(<|Lr8Mo}k&T$8P^`)cyD63TQ5_4Nawlq^$=60+SLKYfM2b<_
z=TAnTQVpTAsAun-#wIr<SEr!nPA+;43G1_DQz<mG!>i5WnT_}ug42<o!PqKXV=HK$
z&a<H-z_#IctvY(p0J>>4L4p+vXdzE)I`VW5my$*rN6%q8@a)wMLppJH_nz0!2A0Q1
zuCHI<<HVO=yyf$6K5$zme&?fCT<k75KHhV8%;e_C;vo$<@uc0zu-5=IoHx&xqUcai
zo97{_w0ajkFLsepglRu81;^t_c>Az&IqjHp(dMo!l;Z~$CzKMo83(2i`1I0~rp9jl
zz(qN5IZk*PSlctw)iuwrgx%%H;V|+Czw?Y7_WbNuZ&}y*JW}b&kIrpG*NzH=5U4Sz
z?*I^;w}i?>iXM-uTWA?d*0F#U?4_%oI*HJXfr8UAE(XmTz>uA8W1vW-IvIvh(+oq1
zHWJnXR`VycSm~C<-g%s%5&ntY^569OenCaW=38sSMNiyChLqG`K}6^K>q^aP&`@h7
zI5j|N{V;d-u#1}Sc&{#cDF$K%mWJd4trCYw8l%aG^nle1r39f?+ZI@(-<S=y3_;=6
zW)5*mw0FeLRCYp2+L?TKJhC*$-LmLRc3ByNV@i5Rp4LpUS(r{O9jCQYoTIu*b)szi
z;FT#~bS^2ia{st6pEAisVsIQRkN$E$F^tht;=FkctO`TV3wbT1pi#=DR31+&Wv#k+
zl4h`L1*^s@xi*c_UDcz1E`_hY{+gH5#Nov!>^}XN<?;&GX7+~(+6jEXhXFwkT-PkQ
zZ3uwJw?=z-M?T$K@|nfdO2PR=6Q{oL;~w#ek3Rd1ufKUD9QGVyB>BQHCcGCe4kPR1
zEw>+TiSr$Kx@0}xvBp3K^__^G8XVOLIeFscp2IN{(#$%*8Zg+_n7d7{b44f)7KE$`
zTEsCD21~_jN@Vj6S}S_^8?pqRl4q!m<#gh7f6wv55wehWf#igt2{&Ve!^p)HczF?d
zei%U<A08d+I#Zf<;?od041t^dK=FagAuxF%)J&prJv!pc124yst6gMBkrV>NKy{EC
z?8ku%AS^2)jUlgGIpJ9fJbLXKOD#AY``~!Bi|mu<o5#SzX;rA&c`jn$<1zBdVN&RN
z7zkdq`0uLVEJVp!#~Nb<(<&=7y=*X~H%}34{o{J`t_N63Y4dA{UVCrf<9gc@PJp+L
z(J2}dyutdx0)0BO+78V#|IJ2dL4UC3w^W1O)gM8c8s;e6E2co|<gIGvPj%DtyT19F
zb#bE?4s|;qN>xsWYYnN|x%d4J5i5oi4GhldoVzXrDLB+BSogWGaS}R@s)@c2odxT~
zRoSa5M>-dw1HEFX#M6SH|NW;gC5G1XHKT(u+BEe%gk@dHwXkbDh8P%vV+>xU)b&J?
z!W2dx-apVzLfq|{eB)}2{PE|Xv$8VXOe#<Hfm5jzA9<K7VcPTT<qP(Q1M~5yW>~=y
zePWm{HBe`q2;Py`nb8ZEqvI-$>_iC5k=M`nTqWUtEtKMMPDei<G8|pzv=okHUR{s;
z?Pn7gQ{t<KN3Pt;!DZaIr^%kB2@)$~@+=Rx$l!=xV}rZ0Wm9~3%O0@x-0zK!4!m{o
zV&}<uB&P&iwJ&cu8xO^-*tf`D<<}qzSdC_qmj*bks~J=V<Nb(@FswL7O{#q8Y8E-8
zJ$>!;``@2Y8*TiiukTe1+}=JijU$)Q6A-8x_!C2*)y86hD@u$rowsSd)jYvWV~ng3
zLIB@%w$+5ZWY&4XJJTKSoOLIj<+#$MlH3!Bj$niwHR8b;InV*oo0$sgk8YPrO!QoU
zIcJWinc{@^r^0cm+99(tgoyLPG$c;h^Oy_8^q{5cY`6tco`G-pR%tkrS979V8pnr)
z<HO47aiz(^vltmi&qcHurww|W6V`QQ71h|ke^7}I(Q|CV@p0zOhX+b79MZtWZekpC
z4(VezH0f?;p5`JYXa3XYUvsse`2EWZvIjRt4q?LO8L115lX{=KKyyi}Nk*z9qGa+?
zX!l>!))OuzQfRcA5j?GB=-rsHmJb&<{P4Ga$m`$!J>Gx*1;=l`;Ns?n>E$OJ^McGL
z_A%r3*SuXbcT%~R%*g@L$TUvgBS8=!7!Nx>dM%WB$3^g@UDD>qfp-?{>vXL(Zwzl6
zO^jv^m4atk9icW-&BRs-B@?jcjZCPbakshhyfr@C4@@r`H&-Jc?*=ZD8pG@liL3p{
zi;I!NkSMKkcWfvDP7Bp>h<tQ&!9F>NftSyAJljWwG;1O~zr;O1a4{tIgE<!(YrnL{
zQZha{k}QPU%z>{GhQ=~_-Ug3|x6zw041poQsj7?;2i3CK6T;3po(%)TpsEOQj=5Gf
z&nXS6oUOrOgVnnn$|!1z1UT)A@e${hWQK^csJ*R0>uj7uQ#hkND^dtVyib}^7|aUR
zIZOGo^VK>iXWe`VPP<$~1-o`$&p#{j<tf=p|E1y0kjK(dO~%y^@qSPSuMVO*V(c6L
z`JnYCzx1uUB<lO{rhe$eW*0QfDNymtEwb)u)39+;TnLuAU~Y53=2~J7h}+!SX}&^;
zUiIxyI?D=bOYQr63>y!s7c%->ID@)_!S$*-Xrfl-*+x~ZRc2;H26j=4{SYFjT!?XC
z*zXv;=iU1w_wOEY&hgpxj$QQp>{q`gdC$i$ZkX4V*+*_`L53Z}Zcmyf3XNg6qXw0J
zrVt3u0~!j1z~CA;yM!w%hu|5S<NETF;9)m<K7V)A=Fmr+WS+T!cRBE3RnE!~BWVci
zN5{<q_4W<va*x|zvET3c_U#+~@|!!}zg?Mz$o}=f7$Sp5ZJ_&JVB-`$NNr4pH6UF9
z=DaGMFz*1VG|?tdN;@0><jk^G*4Dsl2k&fMi|E(-;5kf_ojE!hR+YmU@9`UV#E7)g
zrI6?GMK4ae<+XAOv{@G4<AVR@*Y_I}%$eXKxm27_R#iKjQNg^d<kB?uanL!3+GGmJ
z7P$I1pOe0Ds-s@#RSRwDjlLs|iCs!m@9`<nPAgKY$pa%k1_&_hC(<ri#eCK;*VX*a
zyCT!Bk~m?mm3POP)yR1tjw`3qEb*&S1|_N$8cP*w3zVwz#I-nL6WvJg)B?F=y?aRm
zPNX$z^{lHBFI$E8_b2t<9wJ2=xmA+FCREp0*23-m%$xT|=Hr6f1te$g?oZs_odES~
zA4i2Z%~nj|etYP9Fwa~Xck{}h{N>mD&X0b|v{z_9N=Dj=w$7w+$C3+fbht3!gW!^*
zgi#CG{79*VVRu9A<SYxV(W9v0q!GHyn^Zpf^fmWu!YwmnXiU#PX7}PHZVc?MBCXu<
zt6aDXj)&lQFO9oeN#1j!GEg-iUL2!~Jipw-m^paQ<us6t35=~PJx~JCX%jYig=R}|
zTm{GAl;xjl<3Que!C+Oh&Rv}blsG6Mvfp3u@&}RSqpZ$?`$>5+5~)LC7!sow0+m`a
zb=8@s4~l@q<O%zMU93#Ik?YGn`v6G_7o%sFE|mBg98-$srWv-m<*o+n7Bit#g23d3
zVGMj20wFZiVoDGHTnY;la}#niV-wVEZ*Y#$L9D_cj-e^(ux2fOT<hG)Gt{_|r<@w^
zY%@ti2ehLVYEu}su305<QMVG;4``ag_fE~Qx&>@X8kO8f05(cjq(CyQuBuq+L@Fx^
zpfkPwws}kM?n6MH`oXw!Vw?WeGZ9d`XeBxZ+e2LNYAR*~uYLe{uRbI??|;&LZIG<a
z<y))Js&4QeoUD}jmM~37QbZ+&r1>i$Sa+v|iT6SCRifG7R=u}wGiDN;-DfH0>?(bf
zt;JyHR&<^E*_1{PI2Zi6C~(e@51Y1%s9ZZ&4p%#F2FIJzBVT><mN7(rwLbFv#Wk)p
z9^Ty{X(Tt;4Ux;3c>j3M4}bWYhtoa3x;>HOh>rt@s|(^VnryDJ+Z_lVqEk<hQ)>_f
z=PE-QSWia=sodAbVQ{#0X1AaC=JtVQ&FqH6%|5YDBecw?E^=EUzkVn<soWoD2B+p<
z!7ZG={xzq0fa_O$$neek6EDWVtBc0#{gEL(Q?8rzpe@}4wl3SBolL)LzkU8YSYw*k
zLai&_q@@beTXl<U2X{xsJ2ylPxS|#T!LgLavaGt+ZJl<|rz>#mW~(yCR4>xb3Z1^r
z^=k?(f0xnoUweJ8V!(+evk?QX$+n3EuMU7~Q5gl^Th4aWpp0BdG2o-8R^hamY{p!m
z%2JrYeAFQXM;x>Zj6-C9IpMuyZjHJsp=Q`6rrp4}8z2p|XkQ}&t#Udprny87RIIb`
z!D+BhDV**ycS|OsPH=8W#32#l2p+0WWbDBu;G1q;d2Or-d~CRM-Z*P%RCgv^ty$x5
zyCvjWMNMOBW37b>Od+D6yu-J~@p$C@-92T^B$vp*yZ1*<k2)WZgC`Dg+fZt^KlMs{
z@hKp|As&h}{_>kQyt}*QbxMSosLMUct;%P3cp$Q)mB4hZM8-ydL5-iJ&bVPm3xkoA
zDs6qlm!p!OmLrszR#ywwP|1uhnX6Zy;HDiH&tEW(4Zpl2l*}zbyZ;!e-f?QesZ>Zh
zXKt+$if(j4V022v+WW{MCMDZ7XE3s}!$3RUQ5w4#$SA36ha>jQGYhE-I~@Cv?7%Y3
zyJ%{i3yI6LM{YVBS39k>XnJL{c`P}z<f5{us=KNSeReD)HxS3bZc2>FQEOpA4<9!K
zMS`@HCpCv}#Yk)CSz~biJTrG^tL~DYSTj2<2uiK&0(`VjTt=1Z4k7SrmpJ%<ui&cA
z&^!*eRxl!+X$fotvN?VTxY%WQ>f9HHVbjcGoPcVI8^dUneILMSC8=rzsm2FL_#u(T
z5uYZNMs`D}j@n5BDXJ(iExu~J3lPk^MEApv?6AoV``-J+80N_<gn%S#2AckPjA`TL
zq!<aMsui)Jbx$|e?)|^13{d4!vxsshmEbYo3Bi>GmItbgw7AE|p`Q&1bwe05xz&CK
z(3yt%S8d#BzWB;8ul>{N^>6AX*o%Pept9@H$!JYP32TP@bOZ?dnZY|QBBaxMPH(>E
z!|fw9C>IkMxH%kH3;gVxcYHm=BElE~1>xtPe?eX|rwZbQ-JpK_>skOU6!#Z9$eDy5
zTw{uSI6XouEbB~?1_ADHeEH3LE{4SV{*l{LrInTI%Rng)e0;Iz_kQ@2<KqK68K@49
zSy-#%{oRS(7#X8yJr#cS&7Y8d{JX?-!`m6&+};t~1vgB5efz+&Eay>+6Qp5ouh{rW
zOySU;_y9C>Y;;Dz)F3QsHEFLS#6%n#z6!1?hetZnV1<Adn-&coz50x_V(U{awtY8}
zqHUx0;QKhd*%+XmxhPMb;r?pvH@6-7e#L;-YQ~{Vdym4A)fJX?VV;d0YQ;brBW|_;
zB4^c(^SIR<!9pM;)44VwFX|y4g0s}9K_3gR(WGhA8~}-q;d002)t+%0SxSX*WS9o^
zm(Q7cEEFx+RW9YUs}4asuU3SmRF;ZW^o|sTVHg;u0cS2-UPSeyHd9B0DurbZgdhxF
z2+lGUa>2=9HkVkmZb4cl`d|>5>V|#LnRb<CdFskWUFJvT<pdUCysVAK$Ay*~LyU~0
zx(=>Yb-{5Bq-16kO`~eXH&Sv$C!F%ipZ(RJ@$t<>ihI`O5ov|6yCh5_Uw!#2KL6_9
z^AG;$zhSt3rkhZ!NM6au2ONRulljpLvaYb);SnUyly#=jl#HbldZtvlmx0}W$B-XL
z%RS}zEpM9V-Skq6t6EKaTgYoBif*o<6qT|WL_PY*E(G>L4VaQKz?FAm%v{?*t3VOQ
zQiYhCS`ZgKbT#BhpbCQ_S`Hg=h-@kk>4{^g9BW+`{Ol;Dn7+T6VNg}u<TX=@jw@Po
zxS-HuF!wo}r#Me@ia-Q|AhGY!9r2_#T3)x{8}R{gjbcZ`ykw3`;j}ih7*$!}5F!^t
zU_S(|ryKIRnALIQdN*(wl-wq*(VVi_>AL3D>~KB13)+lNH<lQ2*1g7*a9%r0iDHM8
z$^@NQh>-(4h!nNHHJj+*0z^+pQHh2Ii5F?+@vhklzGb}m0Z8v|MBCt68?BVh7Pu?^
zgLPPfB?F4L15aH(dxtMPNf`8Y(A#VyL?hyghC?~Sh+BtW4T7y}-a1I^nrfVN)-_~=
zxSoDTS&>kv6qWpns#t>bIWrrw5kjDP%TOtK;{o(?P1*leIM}E|TBWvv6kRVGjc(s}
zJJU=<Z@i?Zgi%mF4Zd-)pSYSL@!bP)IU!AWm<y#rOAe~=;W+ab_bXpLHXfIi%gGVO
z$irzRrh#cxW!25~HPik=$1e@mJX3PwB1I#>X4ZAqXGFY8TFb(N#k;rac|2um@Wi;|
z!~G+F_T~*geRYK^kNo7*XIxzD`TWaU4&$C&3#Y8Izfu;~A#riov3&6>PA;<l$=@UF
zCc<)JZk~51;ogXOT4?G;Q^e_vlC{>1<f)R+X2EaH5Ev1Zgp55lf+{zLM8q))wAGPo
z23NOhuNfV>TWSiTsu6sy^z>HQYBfa`oYPUA?>~$u2n8K!&~amoXTHmKAhW;W_5BEh
zR>KGHwHR<vb78KkeJ!GiP!&wCu24U<DC<83Tyr*SgpdMhKPW_Ihp9>|ye7DbW*xYo
zvXW+nf%BTe<&AVjRC9>dnnp%eVQuOdC`E5zj8Ut9%Z5HxjmvFy#XBWH1r2D^jo8Cd
z1Yaxbnk@>nnMsRErt*?WDd2*28o@XWHjmV~lD0Czt1@9si6JR#wYH{)AuZ5o9M_q7
zU9=OZjd>Qzszx}{8YyY!cL<JB)oOLk*$!|E8WAyH0`>21(Q$X2DUTmGz5f!{Lb#f^
zkdZfUGQaxud%phW9bf(G&)9{Z&_X2cUz6J-vOI!Ra$R}*<{j(lo@a;1#i+#^Evo{d
z7OGd4dGJcCIYc;}76zx4@?(0&=eM`y^+C~>Qi!<_b&diR(we!%v-9zZZrU+AZxYr>
zaB3Uf+$N~n=E1HtT&c8qWj(GOkD0ZI<s>W|PYd<5z_e4{w(f(fA$L04mF}>IY0jPB
z7{-S4>72GfBNA6fJI-WLgn?2Sd|(`P9Hiae_RMbH?;h>iXpNF>Cam7T8)?pJ20$%F
zS92@V=G=i7rs#Nf7<hKEqqWLf_S!8tB`8gU&S$AmouX#v88jP|dEQ*y=DHSwqA)H*
z^|0@8#9~Mm>T{vd)HXxMpz^9?IiCVTR3oD&&1iRN>#Bm<3=t63@CT>QL~S|)uL5iH
z7&xa~1Diuj6_iGy`4q0T?HcvZ)QbXKA6;T20dGF_cAs1`gra|to(UqjR@LS_cyNJQ
zl+f#(kmA7TJVVOFP&d;h5e>6)W^~pNLm1szHGy?>PEDOkDLUG)g;T3K@^v$$zOOiK
z5fSXqw>zmk6!D52`C9QJWIE|gJ5K#_9V>-asOu4ynVMI2L*o8aIL?*(MfcwiYvcBq
zxj$;YO0JbyDp!}+gc!JA8<*EF+3)vM5!O;zmz9(f^J(D_G%a#Sf#fHa7%20sin84R
z7YA-{@69sYu?l?o<~`ruo+u<Pb_3TJBlq`@jMsaXIkO)fhe>U@2PZt-Kj6LacXprf
z*|Qzr&hIGq_w0s=H3q(U%*3OTZ$nCot{QhjIFl>wsHRyl)eI5Rvsf(S23+U7d98$<
zOF$%fwF1u4$SrFFQ98Hg33M;|J%iJ{9B=1FwVxS%K)j)2;e1>_<&m6;qRMfJ(egX$
z88P@xUf-)2=xl7C1~VGc1IT&Lnlp1NGz--!2-&>boh5RGW*bOQrfhdd3uDx{Q|DOM
zRSB0?LDeM_V)sVxH!8B0#%a-+YH31SGBHJV<HUSgS&s{K$xxKsm15YeV~bNO%Umsy
z%4y~6aGcUic1}kT-Xo<@@`{L}2Em7IBdJ9#V}~)3l7=>gpq)vp8VIvhgXkg{V_=Mf
zn(j38bIdD?WNsghJf3DsDTD@jb!xeM)JR)5dX<I^ks&HkOGIsSdsHu?K}Hc_jE>bc
z{>{(-oX;*tKHF8w-6I$N2mI;hANW81U;aJsC2`{%fBO%AoAvmfVSlLyL|Jg_Bg=eb
zEt$`M@n^jG>eu}Aw?5;u&pu(l8)$7rWTqJDEjW;zX=NdW8OirZ8=3dd_~P-Ahlj#$
zn3=2_M;w<Gd@j_JkdBpogKIY$Yg)U`;LOaaVRLA+k*;;ya9h>S-z>&fsxW%Tvq@pr
zCByM?Wqq7khM5|b|AE&64wQ4CB%Dr+wE4Gn_#q}MNHs)d%$3DI@Mad~wV6$~XB-04
zpql>D)CDZr+7QtLp;ri{LaBO44_=+~kcOul{sfxx{Ftkv6Pu%)5%*Gn{g`m1O&*tP
z#d|G+wLm>Xm90adpob&<U(cTGn**T>TOo}L<Wi_j;pt9x3qIh{BE>u7kvKiHVj5^J
z7)sYDdBHVZ&x%&)L1}LV0~f5D)uOgFBQT;%phqqDc{rTAq)wvS4u7!Cl8t-NVRJ`2
ze@#2<!l84G^lN+JRLzdIdNTns2!2SB<b=t2#?XkNZOD^{jVln0?A*s2Pp`97{Z-b?
za+-C&H4aoO73ne4uKOZ5Resd6S?YGg0xbUlyKYj{jj{I6Z6R*P-SR|At%Nv`#>j3r
zl2YLMS~E*3o?pE=@;+yL43t&~eqhOk!+y^&jI6D)+aI{Txna%4)Kun}It-*Bsye}f
zhEhU?$X&~bi`?Dbao9~<U0#zyW!mrf`7gib_F>_;HbN49{o%;X<k<&xNtM>P+>gW@
zc({KcT^vZc@ZsHC-rqeCq;eS}pMUc?uYULMLlD+^WsyLo8L89IS<z#@8S3K1kZ-5W
z%@7n>iNTYCBUuumvw~i#Gvr3&>+zmYg%DR{R_ksthroakBOBt=i_>88TIC|xnnTRj
zqBoaT@R>@d^BA74_u2y9D?%@2^DO=wU*D@32;zv+eGHyFd_@zxDnb~N>S{52Q&Y+|
zaX1<=CZf{B^x>|Jh$loP<%vXUuBI0@x3XGQ4;-~KS##n3{-{xrgKi8>v{R@lGp~g*
zXYEc5z8hm?ObNUvSG9o^%O-CkG*@&cha=>QA0jO!0%}t_9wx2Q?DM+Tt9I204`U*R
z1n$W^(dNk>7y}%pz+s9Sw1IG(S59@M24N|g$KxaUlo_hQ*154fX69otWr~s`_e10&
zMJ3Dic;6;!&QuCg3SKI{t#}%Xcl@hA`<%<){|tFJvR@b8-rXVd%Bw5S!>#c3uYb*_
zS0mHa3x;7wo)=0!(bh+f4>N!D%dhxX|K|V8|LfoUl0W>(NBpn<o4>`gSFdm(;W?7$
z6Cngn4|mM(zNO|9EkvIE_TS^({mh5E2X@UfRv2^99PSEN&EZascc;whAtOsA9^5uB
z_ExCXVg?};h1W&sKJd-p`(`%1PGBDsKfFFbZM=J2aKV#G<MHmqE{zO>9_-QhCMtC^
zwZxNZI(8V(3Pk6lLf9c8p)l?W!+1iLg;E-0@a%U3yFq8(OLk;~<fB)^5HYx56=5mr
zRX(`JE~v~_{qdEMrt7V;t}9E<TbN2p5luJ?+W{D3*fxn^opZmA9p0`sUfDXsCp(2&
z1e+g!KfwHxgGCxSFU+$#Svo@qhGD=X1ZjBkW~SmDq!d`2qvb}W&}M^3rPBxPQ1zWT
zfZbc0m2zh82h7vo-dnW%c~B04MhLg>%+AFS3-7%akDQAg7fgxcR87!jiQ4Ud7yd0~
zy{vo#PaFcnIN+$brYfT38qr}vO*%Hwupl!V!x8vy(bZbDLe^x2t<kdXr@?p=I=eP9
zr+vMuFSJmAW@D|M%&1O}B8r@Z;H;3bk%0>x8K)Dx143vk`Sd_^>i#)~0p}8PZk%G|
z%i9Ou-DP|j0O94$H5a=b>soQc!1ePNlqxK_kh5Cvrsx^to{L@7_tXlzL8Y)|&4kMx
zkH-_wFLpfJPns8l@b!lWO160J+Q6-R{`P^#(~;qKUh{IU6meW!TyyD^dv$TxaoFuS
zoig7%oJdai#ru1{Y%8aWkNLyj`z`Jt9=JOm8OCuN5eW%#)x2?n8CO}Sjl)N@P!A(1
zc#w*tli>S!z<~?faqXlMjHl&YutK=ud}A0juuB)bj$fTq)vT&)vYnS=IV4Jz?nO$g
z=OVVXrktgx#jjO|O7v~NX)W;miUBE_FqV2}Vhd9}wS`=@iq}C46spSKjO?})HF}9D
zkX(YQPJQ`!BBc0TyGE<aMc95{XRxrY>eZf?%>H7B_nuN4OKm&?J|xO<p)CcPjUz%J
z#y|nqYJSnyq4{u@CYDl=8J%I{@BzkgVBAew%?gxt#aHlXXHrBhMFllNswK0Y7V@l{
z_bvrKxw+!gm)G1}?iiA%if}wF+#NG3l{GKabs^@;O^T%8IX=$hc`?a~E;{dpA?W54
zt!wE-!di>^H+bEIrXhhi_7g~mw8L{={_YEQFJAML{R88_dc$A6eaDM9a{KlTG5tRA
z@RBg>sHgWhY0UTU`NhxwivQ`~{*3#R@b1lfe)+{)KD_yY|NZ~?KQK;7&0N-QHkEnz
z?i;26FFyVu)9!}#+pkEakV?gy49DT{xsr6y&%4vYSMMJ9!OH_z`#mW}?f9Luh~u*a
zXwzrgOXqpAvA(_Y9CkZCdEVF$k>k8lr82~X!*M(wS+kCoQf-vjzQgEOLm9UPg?L*d
z>NyUCKuAgb@8yANf8Y0wzh}|6s;JMpl0wj*@B1?YYbl&^CbY_ny+$j$5Os85odgIa
zTfrFu`)T0eK@D++6gK0Ur4+SdMx|=oh@m!mTyVc`>VDWQYwg))EAO|@9)tP;h%_Us
z!MZFgr&*0-1g3G&F-qsYG(B8&)@j;mlrI2|npd5@+oA5gN(1%hvW4q3z0uq62Oj_p
z{^{1oTLL0%MbgtirS8af9v3{B1>4$jCWEZC|Gxc9BLsrJ4JR5@Q>&=A2Am0in0bu1
zLPC_N8N6~C&XP47wAw+gcFlEjI}|Hum5f{rx3Cd|A){jIC;b3t{y;MaJ1I?lK+HZl
z*w5KG4u-f&e=j&<fc+SVtw7B<X$(UmH&1SkZ|;x$`RxN=+-I5_X?114pST$JTueJ^
z9JqY`lI#Na4<}Y5NvD|DPa}DqVLFi4%;2h8tcx~^;}jYF4qpqeuJ(9@`-g>jf62{N
z<?Gu=hIOQ^GpAB{a}<94Vd2^Cs2o@qcyV*ZU>+CgA~G-V#n-<=5dQS*d)}0V`r-|L
z=O>?Vv7b1t8N5m_8%>)~uToFxT%xLhb2{RMaTrL!lLAQdl(H(9K-An0>)^#ZXw~Gp
zoiONhQde~!_n{G^7E0Yq!cG%n&+Td5sjw?x&RTRun9BK(?>|Gjy2=V|=lIP_f$v)k
zh!CC5t)q{&`Kbe0uA+zZnn5Zd47gEeg)Ro;7bv;UJ5?I^2#X!;)rg0E9@#n0Y9M0q
z#aZWvR6%kj7va><jYimlc)WAuIU~7m5N8QwEro2hqs|^QA0u>)euwo<YNXwfw>?dv
zKEV;H^cs+stNA174JkR3mxXegp$gZB#K+I~e0(#p+wZ7l<uO-I>%wDRsIC&ITnAw{
zC6!Iq$~W`Mw-xI~)N~>`2(BqZ8*kb3MaKi)lTz9st|3J(E|fuh*d2KB;$yD%d)&C=
z$_@PU|Nam7fBf-Z@Y%D8|HVJ~hg`n?ZQS7_o5xptOx(ZwhQIjJzu?b*{ejDiksrK@
ztaIa`2p_)uD=sc}l%j`eSr$%@ceFWkzZ|)G_L>y6tMNrBOC~zV!PA0cKCR?Z@V-*l
z!khb*+uKKOE{^Q?yRA!5e|yfxPuA&fK_ErpTfE;M8N6e^-|427v(DSAhGey--!lY_
zOjcr!-DA3sg#~^%HSiJB&{uQ>lqGKlPq~=YbqvHbPz08&9e-Vw*HLO?t&JEWh>|_m
zs+;B%&8Wvgwo%8Liw4G+-ERz@XS>M5GIBo`j!U73mz1gpyw812n;D8INsmqrQ~+KR
z4J{bOOW*T4C&J)ubJhdZI7?PjtF~rpR-&q>;iPVmIHoW_%%7kYmiZCljyQ}2KPV}3
z(GHwel#?Ldpl&hF&V~^vR%fV&cKBv}9?t21z;`-pjVf2`>`k6$qY>YUgZj<&V{c;$
zg?zo&&o(&iOih5P0c)MX<+K2CA}R;0DvKm!@D)eezHW1MwC7kyAG*WUT8%8-&fT4%
zQBqaaf;dZ6)US21vt|1Kwr$)a(l7%<eQ#81I~PUHsY*n|8F9^<tkLm!T462tapK}C
z5US_f`-N}k!uvH4;vVmyNFoh;Qc66#yx{exFDQQCJ_k-qrsT>nj%LzR*zYH5Y3#?u
zkDgz1GdPyhO56{$vNBvGTFxw`@%3Y+;8}6(FD}^acU<^JU9VV9k2o6NoC+`Q8=t;-
zhB&ym+!Ipbcz@4yxZt<{)(yAIi9i41*F3BR8AjfH@k{<s|Mv&};BWtg@vvtQHGB(d
ziAff93)B&vKIb$o=J0AQZnR>%Z^E*yta+uOe6bYFtj+v&yz>^K)lH=I8E?`pibXwY
z`dSxNDMYJ6TNMD<KqtR&`g|*rr$j+eR15K2LBA1s4K;Pn$_w{hb-{n_^}ULL+6uvz
zdGg*WbM10F@ugWYK(L1g@X>=)!dxwtQmW0ptU@l0b<xfy8A-x>oz2vyU}39`GB4+j
zp*AFI1s77p<5db}ka^AOJm{U;Y>pw|ag<t2CL-V!`c;;5u!FT3v%8SNzGo%1%7kMW
z2R)p<76Q2xWjU|K<~J4B8lg2rDoHAXOI!?*%l$x1fm0SF7kplsq!1h&yyIpXx!R2k
zL!>O3{E$hUteSL2Qq+{JAcF5kDPFS`qK6>_st?GR^nO_f_Sx&txV(7@PNkSWF8t2N
zdw%%8`3LOw2VQ*iW4QbbzrUsyO~wnu#Isi)^LszJ<6r;92flp}zWI9PkN(~#{PX|*
z4>-sX5lG`ioJNLA(M|Ytq9sQ?-80n_AAkIkFTePe5(NQLq*g~M0;p1?A$<F|^5y%L
ztBXgxSzRlcM!6})^t5_zx$^-~n-Om62Bj)oZX2c7h-88{!=etWIql*^D|c0?`SUC_
zBwSObzXzH#vRvrAg0{KUg%A`j59*si%`}U4_0~N&R2|UK4G&B)a4|)a6{4kRhCwNM
z2)L$5(B*z0csQOae`RQZV6J2$%Gj6Q#R^chNN82LEHT)j0jnsnY3nG^H`UP>t!aDe
zTG#n^8B;N?ipCZP?=&3^g}f!NG|4#MhzK>WKw#WmAz>i82C_oQG*o)%T0#8-ng*I#
z0Fe^6oJFC#?cO@y+4xVQ&~5IjMlk15z;`v`vK_eFIsqkgdT7QakXDHnM;ow1T-18l
zg=p6eq$!EkQ*dpi)tS~-Bzc5H!<l+O0g_sqrImW?=%29D(L!fR136||suWqN4J!!U
zwl6AT`dt}ezl}|__zyutk`T?mILF}%8GS#_zE{q|7jKXFc;ut7BV9ZvpYHkU_JPko
ztbDt8G6B3IxpCMrj1!0b$Y-xE*iEqb$o<+q_5<^(s-Tz8E*PVZh!4qevrAkgM{;UF
zG{rrq+jo5Zc;YYbgkQdW<o0yWa=Y?jIr8{;PZ;-HzW4~<e?wVU-mQg7;FJrGOJ!aP
z!){M*4L=O*FR%E~@BSWt{_Y-0N7gv<Ae9e4|5NU7-|>(C=pS=E3<R`DD@~h;Ga7}h
zK`pL3+^!TbN6str>BM}RDN;8@z((|UZnJauC`YWM#9A`d31W!RHe!JDn8zt%p=Nt$
zzD%uZmyOdk;A6nmYQ$7S^}G=txA*WQD*HEheLpopZB_^bwIL3HUY{0Yj!G`d2(D^I
z(#8QdsY%W-CJptdjk<PN%w$d~Ql}c_Sjj=wl^2y9N+;y?LmN8?vRZHUDASsYst9Z@
zn1ZSwnpumgE2O2}#fVQm2trxJ(keCUA&?!cizlp=HCHY#4(!K~7@f`-OQ9|+l!BuX
zoTv+-G;(rw@WI2JS*zfJZm6!V?4@vsj>DL^o(3)tBfDuL#K_~riOD}2Y0f$$=a`1X
zI3!Y3zi+%*SekBJgKxUFSbuLEcf7uNP1;}6+(^v@;)s`(>%*S?^H=!GPvGhm&3nq}
zp0>_Nfs31u`Tf5e`9HjP%{TAg^XxM4+plgI-Ao)lVtDo`DFt$#aUz7PkJ&$e#p&Ib
z?A%JrkG#0~3?DpqrxO>$Ky&)mD{8iKISpLz5}CroT)2N+7{(J*3R^|pP)p(3l_~yA
zvgwC{qB`3nY_oqc0-l_^-`-`q?pUdhlQv$u26TE+R3>QdaiKQiI55Pdh*&lTx9!PY
z!d9vvYa!4`qQydUmZ2f~z<Pn`)IBiRA=IimQI@RhJ;g}!5Q55Lk-#)KKxI_H3sds!
zrh(n4U3W7Mi`ZN#7<thL4KQ0uRaBs}{gvrQ%b7eM?Rp8>JKDIzVfLuTPe7Cqhv+Q2
zwaPqKq7x_!QVI#%T%GYA%#cN5XFEG+ItL8_JavP<HT*~Y1ZwN)hS=y(|J?P8&Jes_
z7<Qtccm7F{qlI5Jh<t-ly&7lrd+%(1tII<S?a+HK&Lg7Tvxq)pdYuZsW{VJR1e?#R
zZry7UV?~X#579)GLalY5fO0msGmgbD4x4(w;Vd=M(ZmD}Pv-rKl8JpzzkN1rzYx1$
zG4ydyM@Lk?*&5Wf;Cx|!*s~iF$5ZCr`p940Rlay^EPj9hb<H#vh*2|LKEB@b!)JT;
z&#(CpfAxXe_w&~5t)&r$p!}-cgw&NOjsz+xB^rsctbB8O&->d)e)VDD-~H;A_aBbD
zxOvH|fAj}@`pIiffBJ9vkN@&tP^5Br{+h?zcjV>7pTECjZH4O>D|0Tyy~j^G%}1z}
zPe1vD-~FAR@}ItVM|=KbKKV!g1MXM<iMN0L?+G6AGLupt4Qeyno&OXWGiOjyK1;5J
zV%M(bd=8}xNlo@l(YCn+iH?(oI3h!)PD-$^wOEm;`%UY`hd+}|Q;14FE^R{)YH8aZ
z7h@ze^NeVYUJZz?0cXyAF-F|_)SdIc{`y|Uz<}y#b*^z<h^iIt6DZplo6y$6^6&_*
zs%GCF(uTsbxnyPCn?$0^<Fv04q9{3#F1u=~HO8_AI=-UTo8C0sV)K7}FTrUZK?t5A
zCfl(?U;WqvLyFX#wW`&Lk$D81-N0N{MsIA8E*-vkW%1{XuNjfb5S;$<9(*IWKue9a
z2p>+FyT{BW4yi)&qRibfa+ngAqmB^bI3PZ-&X$;FH`3urDX^OcEyQBNc2Y{y^hOaK
z7dRKT&)zCr?XP(L>=kjkqWVP1A0Y17U%epht|-F`#><zq5OB3nmW=N~ENREZvzPqs
z{f-~Lo2bY4<hwWcX=MNMCF$Y?_&|y)&cSe<XwN?8@az>lyg>+DA140p-~9vrKmY1q
zBf03?_G$%tu^;*Iiv!NV+Ti(N!1+MVb(7gSuLn|V*7?{b)}Mk_mH*)|n9P|lNSr0P
zo!4bY1Y-20q%f%#0kT=};@tM&^|@f^Gk5Rw^AauerqAMShejd$Rw~{%rWgsWk?p<I
zYMV}LcuO11r5WVk)N3I)hH+G$fx18G=QNp}H|bUM*qTVBK~)FY)FdJbH-;EVLqwbf
z;#jfcEqkK^Wl^3>o{xH96qSkQVIZX)Atu5w5Mo5UsHqNg156OjN|2KAwa~m4VGbql
zm0HR3qe?hwxK_bb2i|#t(@_RyYUB)|@Ih54Ra7?W<9RHgci(NZcV~r99|6Ra%stTY
z6dmXpCB1M<K}UqO7OHnz^x0mf-XBjC7U+j|sP(iKn~Ie55x#Tg^yiZrco|}Yb%L%R
zKGxWHV~(i_qBc0*t3rchPKQ$H>59Q*+0KfQTFX`pc&}WFfHOsf-<qTpy(%V5JCD<U
zuLUBF7gvdwFAmg)_q;!zc(-Ie6vtx=<eCv2!)}k%#yE~#?FN4E;)3f#VqIq@U%9^A
z(Nf~^bgRsHD;_`G-E(s>(e)Klm0wbF<=w+0U(J=jcz@)>@sYLN@p~^nA?G)I`?IeY
z=Ob4)*PND>*FX3z!nEV=uYSp&y}hT1<Hh@1ev&_8DvdPmHpqNTiNE)E{x-k(`j-Fc
z-~Nn}-}3z8iuU3K>%34~W4GVw-X4QCj@3MJV$}1>2V2Fx(a=mB=L5sECrvwC=*gGP
zo(EmuDpRcX{EbXf**OQR@ol`RRQx6qrMnWwAf(_aF&J@FyM0A{f^uf80O={8Qgja#
zD{Mmy6e|S(|AGvRF%rEJPCOV}R{4`Is1`hWN2v<GQb86dERc#u6bGjri83m1Rg?&4
zhhD9!k(R7lTOYg`dzieY>YTGxW$_YBy5q3f{+Tl2yk~HluBhpNMyP1bT>nfjYF|^O
z=_YR9tCKi38AtG}r<J2WQgTptHP;Bq;hJX%n!Xim-sD$Dc5o8mv{u%vk~Lf)`GF}p
z4pU+rlw9U*-qdW&QHqeWF2d-Oks#gXMef|WE3VRzQcd2~XSR(PMqcc%8HNi8lWw8j
zQ}7Jq1v%^(o_&g+E|99TGlm^yj=1%x5y4(ayFI(Z$g)%dk?H0WWW1!9<fZxn=OZa;
ztZ_(@u)83vACTpNPd|Ri`yYSG-B;i0{r5W0N>0sq_Tz*!9dnF>-p5kKx>7T@p*we-
z=fs`)FIFi#tLSx2T36=UHj_pA+V$r-YZH^Rn&-Cz=k2~*v7v_^h|C(#JS3KRWjU?n
zvQqp=m-*CE@nYKH=yA0u`sF-x&Xi_uaIGL|v@Oz^&R(CXWNTWwd<+{6h+1*)1{j85
zk;6(j5OXfX>t~H-eH)|d^QCGu>@Ip<Urq$CxZa;}(z2>mjp?q1CY4oGnc;lgqQBLT
z-{zDyvqm~RtcNAuBV;XF(88}Qk4RZ?VW3qd!g_Dl;mOzGR3GnJQzpK&O-0jZ?;&^{
zi@1J(HllO1s1i^Y{6;Q(Dm7H*wG}cpDo7z{@ri?Bz~zRk)p$!@Q|FqXF%I5Hh5CJ@
zE_f+KLj%2Y_z>~K9)>-Hk!Hj&v4+Wuv!JPZE->JXHW*qFydeug8vzIQemxD36rt3@
znrB0R0>c>5+z{PkyFGF6eXP(-;;8Eo5gUb6$c6b)ka=b;mHVY|%4&mLYSRYAqL!0)
zT#tb%3Xi7~E{$9t_Wb<Kd%k*m&yt(*r<&>bg|ZeN9tz)5I8Bk9D_?(D`HMF*Uq8&e
zeYf!9<!8*NOV-DCOv`(Ar_BB8xcTvqxh8S>{1tT=aHj`8eEn-a98dh^n>W1r)eS#b
z8~dv(cGEzU1XZ}Wy5yhzxBrxfH8VdP*}wXT-QgJzkMFpNH>yf;jur&x0?`hkVMq*S
zhSz1NU7^zDd>wh}k>6X9+A5Sq$JGubMEsyiv|SuwIg!^BLAtGYFqB3QR3|Ef-;YTf
zJKENuY#qL{Lf;39E8Dt}E|oR+ROx8mZ}|FN#X$1f`6km^4!)Cb6wV64F@&&j0l+CF
zwq#{o3l>)4iHhfS@sMsZK|@wngL=I<J>Z-phPVL)m<2D*SiBh5QUL~>^Hd`YPARZU
zo*_bBt92r}$p#TZt58viPP2<|jOgB7E3r|3R+(oWYhfKDL-0%y_Q{i?BYD^dhi~ZY
z8P$%qriDPMVv><S^o=2S3#&0xn`RZB2=i&-?r~u@pM;)P)_tg{>c%G6hc-i$B#=ZH
zd?5P3{^E+8o9DQoNP?>c7bEqUDQ=?dK4O@z$l_?LTB}l4mGad_5P_T-hs6HsIYSCa
z+~I~xBuwfe*BUNPgmI5g14seq%^>Im4lee4e)OXs@WrxndOTXE?(oA%KsEUt+IBoK
z)lhi~I9((C;lnn^r>X+6?yy6_J_OzDYNb%m^JTh9#mK2PgYRPo)v-H8CVFKpKx@>c
zQClW^&yq9qveK%`=sbXUlf_o5Lwz^A$9X*f)>6n#H`E}GAQ@c1i58LKjF00`l%`bM
zfKZv02-7rh<pN{WLu1XAIb`jy`;6E$%fV^TkF-WWpU>-EV({?EWx^vIAMSV)w3~7Q
znKLQLdq;@Qq*iR#9qDcfz2g?^xT7KmTft|x<02~CEGFS~;G&~SfaJD#V5RqK=Of|>
zG)3ja81tebOD^^~dgt<2-EW-x?&b{N`S<^lBb&F)mFm<JMJ&_^#ct0*PR&#lPSKC1
zPK$axw+-g?G0}WdRKRONK{LW;^ES#kr^$)I#jW*`$XN=iQe_)~x^7S0HeO54MMn_P
zzRrly-NZ<MT+I**ZzF_2OdcmIE)>nq@K!V-3gbJqZ5J9#UM*5u5i5_5r@VI@_7e|F
z<?Y?Vx3`bnofa-HF7Toj&$kapA`Lm+^Wuje@wimp-Y@*>+grYUsJxvQ);trV<B`h6
z4}XUb`NZT_*0JLKz%Y!&5V?N+DTK)K?H8nBBn08xyCZ+`)f-|+q=y+dOk7@G^7_@p
z%^~rFAAZXJ_|N~h{6GKYzoT3}XFcAtpLQG{j$AL9Fewop<H2~`clH-0=A8-4-F&Q%
zV>ai`elMNF`G^no97`M$saB@EDiP6x6kR(VUUQuUDxxYO@9kb{gm~C;b$ZewTjX~U
zQ#G`zb!x{5Hg(Kz_WEAMz+p)2hDeHv4n!|ju4`zEM?1S#^v!vjzqwWkqTc(UjOq{^
zL4pxB)cQ9~QC*vw!ZBltszMN0mWB1$SVTX(!`r^$2<A_sRg^)L&J(?ML5DFg4W3h0
z5|<azzzk_nGNom5F$X`>nmShu$_{#?yu;bNr<Tex&nh8141^H1;0Q2!?NVxMtfB@K
z+SOS`fx{M-cAuL17`JXfWo4PuapvJvSaUa9G3c^3STcU_(4xkII!8mb@JV3so?TGl
z%Cqa|Tuc*93L=G4HS8p$kuqMffAJ&oT3MHSGawP%d{3;E;3KXql)O?MgkdCJyhP%T
zh6b?YyyATz4UrIrjf^J%A>q<aACj23xVho#*)t_)b`mGr!6&?G;0s%a?3`D#6xTLb
z+QWAqOy@n#Tqe7`aFdj&zHhU(?GLyQXSmtGCR>J2$T&PH3UF$a(`+PR^Se{doEDAj
zPC?ja&(5<7_yz$b?iJL>-Gzuy)n{Xqz1ah7C`+qCbp>iUllKHlbyVk>#zga;F=>~#
z<cjyJK98VE7mwMw27T@}Vw7V)CiX*T8awt=V)T)#ivvq8%=2tyOh{2luOW7Ax}gF6
zDmm@iwOdzYCSaoq*R)dyV`?i82O|VZ#b9@1`lGXUOj&^_QkIf-yK2(X<IB%k^PRV3
z(9MPhblKsysoO^05903UAH2GC){eA^j#Q*pWq+H5IG92MsYdz}<wCSV^_37BxQO#A
zqts5a(op0Clhi&fVEaq`O1jw<_Gz?x?9G*W1UT)qYdJpyh7w}yJ`Bn9^7;o2MZJt3
zLP~+jMV6L9R<2UO*TOtk_CsVE6R$5vzP?@ffE`RC#1M(W;^v(eBJV!j^RNhi`T0BM
zV_`pxNGXitKwcNttnP@rUF6Ge-m}!o*YEFndq0!KBQA2aKk)jn<8XO^xRS;LxQWMO
z=Hd2^<NJH1h^!ie{CIz>^7QB!qhmLYjJrLjs)c80fwzw<>7sHoMSlASpW*g9{`f!r
z1*evHb#>s|w;vQuHO`gl_p6~#=lk>Ypzod?Vk5-A-gDay(7%w)LOSYoYd%t~IIM1_
z$6JefVC&iB6<vx^m2Rzjs`#^71hD2%pFgp2rWP|=Tvib^!21nH1n@nJfoX{Bl6J%?
z>Mw<;fU2G86g^HNr729>M+q?o{NPAwWJ-aEBbP=g`a!25GL4a8983f17?aM)D#GJD
zaDTsYceijns#kfRxoO4Ri8E@-(6hf&^6CZ`oaK;P=inTpeTLGsU~zUJ^aDKNHh8pQ
z61PseajGYF67b?lI7S>J!W1pI$$Nu}JFlebhG{O1KB(zMw`X=<=iGJ8%yZ#k&g5Ek
zrj3w1Od*iO5v!V;^t=b0-pu4ZWAN-^B8Vs7-t!Q?COvD&sDiNQVDTfn7e7GaKszps
z<H)bS`C9eELn5{arD*t4smS1zO-sNLC?&J3tHO@Mjuev>e(F9~o#<f@MruA1YGz1*
zX*cRUlC$9o#Yk0Ic^c03htM`UI-A@cIhzxABtbeENBX9SLT?6*HhX~7!q+?lY;7tp
zz{U~w=`i><1SPS8Y!a~AOb1?{%;4?e*Y_=Y??rv?Lsgl=QhTbP`ckYV>*3!5pn+6~
z5JgkoMz3jqx+ayAP^2NwGmU}We&R3<?0o~TU2&fimsZqBCpl^>3WXcgsL!K$@zeF`
zLW+sqZo)+!$<(IgNN=!m48g8%XX8Wk3Y)tawBR@M8}<2zT35PlvLm3L>?-y1ebaq1
z%kMRYK7nF<AuR-zuh30`^qL5r{jbEl+HCW!9+G-WYz2&<Dx+G-Sye5=FzDKD73Ugx
zS;*^5E7`h!(YWQ9OqyuYvf7ZU;Y3apEf56pdf>|wP6Vw{CF6>@-E~aN-k1I|>2#@@
z_h7+C9yNigwJSR;9Lkha-aE!|BslHDV@gCTbiFa$d%^6VbcnU;I*tzZ`-uT{JT0Yi
zx2O#E#l?<ihlQ^nSJoP#1;$~dIZu$n^)3-&<g^q%+%6;!pFF?j?J<)};rRH#exDfP
z!01MNExfzE=WZ!{d-nilz!f|^8!Ny69Det+#5C=AC<9+V%)E&Me?B`((fzueKJajy
z2~JJkuJ;qa_3;fKee{yu<po2CR03Jx!*Qm){lK~k`{~5T&ky`}|L}MDlb`<uZF$58
zM@YJ2>0Pv~lVGyd?;r%6?^l0yq;V@c`x2I(r`nfY+s2QoNJ+TRY`j-3-#{AyXJk(t
zG?z$gpcdmyNhcKc=j@EkDMGA2&v`w!S`$;R{ARE3Jtml9w0Wj=*&YJI=!A=?Fk2Y{
zYg6ri435KYV44PEG+SV=(zsI8t*R4ThA}biCY)E;PRWXaB9(PFa!fN1L*nr`s4G<`
z`q&MOF<LdHL@WWO6uH_bQuHjXacqSlt4-^E(meSpp1ae^JZEUa;2beIGgs2xr#4uN
zx?VLr-Z%DB<oPafGY!1hB@Sa`@>&orOCi*irK}WL8Is=k<kS~Is-gnh!sE;iZME`v
z$~-K^=01rr1uiCqhKCRsePjq}D~6nN1i{|3J8Mi!rW}vBixmkgrGN{8IPA$q_~l>z
zoa5aC(=>8?eB}Ad=j?y*DYuU^AwOtBVQVx3G3}JVi3aazQesjiMqPl0_=Lu-^g<|t
zL_{*dJ7S8~*(ndCnJ;*)-BQ@#Izx7Y9c0~H36pv~wQ^P6uF2xMYcX0KG^wb`ibAm7
zX-XLV(3i8R&3`6AGmxz{O}XqTeu5S|rKxv7tzw_0sW)(n<Q!OPrYxD-Oe*LC;vGv>
zAAwR8K{^@>-Z{@ccqZd3U>#Y&#zR%0!Zai<b~`REMkZI7Y9^GT^T%^<qGZ`%&VBts
zv5rLyam!g};?BB^4F)wSS_qNgqaG&CQ=P|qC8-7PNKx%o5sll%HYQP)6GiH|c(cMf
zdOcLT6JHZ_)YK|CbDs28MA*m#A0Q-sPg~J%EO0w4`>z;H+i#t{80V)Mv9lC~XJa#0
z#HNKySr;u7nnM4<cf~{8MpjO=Lepf%st5{Ch}5l6^8%_Rp-Q27@J$s@-aDi@yw}4W
zFygd^n$`70%h}ux&8FCpjvjW@nbRQzhBRmZn*EAaG<8!a46TVV*etoyDzvP51dq!?
zNP_o`<GfJjnLMw|%gW^x`1sj`b1PCCwShR{<zdGzsK?2N`v;Z<K6`b;o4GQtj_f0?
zc`kzEFgS*`FiTdlU23G~*SvdNXaf5w^5T%VIV7g!nU|S&x3@fG&+S8I3X#L!lYC@~
zj#M)Hn;RmT*Ed&u{LyP(ee@AS8klooIW0Wo#;FRoYvJy3<)h0z`?7FxeZxQg+dtuW
zn%RHxC390t_|_%MXv5fDE#Yit^&N@of8q5sKuSyV4%a^^BARGp<03AI76y`a?UZ6O
ze;v)}S`iJ=3Qi6Es(O~B=n-!lUH|$0J=j0>6KGuYZ~FRP#X$5PXxo%7I-Sw(2hZ!P
ziR($>5H(m)M&)5QaIqV>*pFKAG|j!&in=i`+AVwM2r&}JM2MXrCPWX-1tbPqO0>%Z
zbyY&vd6QArZA^jay>)}ds-(bf)H&w!%Yjc{>?s0be?ORHXI>Y|taI|gcMgErusT3=
z<1eU&bc~+;nAi=0{Seuw$Y_Pm7z4-0g_a*U6=5y~FOAv?E~v#VUWr|rGafdyqO6s9
z6&|v1Dmtf29xkTDVNC4D#26D}92i5|y5$a|s~FE)iKxGt_k=WPymAZ_9H-;LfB577
znSb%e|Bm6XL%d@*jePX-1^eB=%ZqD{U%w;Yy=NRF?JCj21gN!cB@m{B!>enV_a>Q9
zQ=~+}t&dPNB~ftDH2PX{rDiSWT0g{15~E=7nJgnV%kPn|`g_*tOGgQ`K)|^(&%i09
zrW>F!Q>;+x<hQ^6a6iq0@AKzoh=5pjM)2z9S6gFgg{2hc+;C2VgvLqpE5rwG=Y`v~
zXcm1fO2BodP$-3#MpkjGxo(o76pip|&u}vmUu}+3V9o`XR6P&_Dlw9#w#m-A70W;X
zonfz~n6yj`6#>Mii5=CHN-K+{JX+eB4{D?A&g^@Q+{I?Aw)UKUBOg2^>veZH^UpUq
zWiQ<OhGpo2AT~=^Zw5NiGpJi*2St6b4vWqk(Zso~K-!!KTigD-69_kp;->b!Ij`W<
zOiClih1xQ$XmQF`sAxg01-mMw#P+r515q@E&xvRa=&Pv=GPITE5)vZLjha##!ZvnP
z{*%3DD=ZL22G)7D@EkS2a=wo{I={dSiku>gDW)@G<2#y$6#{NUA#h$zx2k4eP%|m$
z-V$8kab0NQcsMCX;N{hxV_v9-#3{pRwItC5YhL-{%X_9_WWU?<=I()S?pGd{#?94%
ztKj(2vw`3K_=5Gg@b$Yw@qw$G1OM)qZ&-^GW3NY#D=T;RkG!7-2vJ9G5*dQvh@>$v
z#>nIXGYe&{Oli-n7taaab3844_;61y8Hod##9U!LWlnQuomch`kBncx<>QyvOqU0K
z?<YUvc*=b7{yldOkG97=K~j1_;M_(Ybeu;8{pD;wa`yN8I<&26`&;5s6dA$C3Iv?3
zcWF-ZRs@_CCf<<EPSP!FrlRBU7?f+JPlh`e0)M?=QNQrt`1SpY0oNsAs1LamhA?t-
zF%m@sZjz6zxsYpNh>62~VmC#`A#E;N4Y8C|g}7T@sd=S|daidCw8CbM+$yKFK#Ghz
z&k)s<UuV;L2zY0qEDq>VzIHQ1PdyYUVkAUo#tj~jMk^~6^~>)pYkJk_bUx#4XdOZz
zg~%AA&23ao5aLsK$#a=27pt&4t=s%iiB%AMKgiYZ-l1LM!>RE0apvtQvx=rOjxKOW
ziR3&((CZU@B=}&X5A|pWuKN+3<tieMalAm{i1+~)CZrT@-@oPm`hWhv@$JLR?%6ft
zG$Ez(%U}G0>(Kb8|I0t(@pR<<H(xUh1L^uBN>Z3h1Y8`6DH27CJM1BGt>S#8F1N_(
z9W_57A&>^gayoH;d&_E@arg3WRb`3MiohyQ<zPzc)TPq3r20R#lfj(R`)Hk){xyAw
zdeo>^2MWc}dRUOHk0%eZ#(*l>Z(|UPY@35P!B>UfhHlMGCFg~8U8u6_+;1M&#@*?p
z)jYyQ%tTObNeGOFh^Z-EAtEHv@l*<mp6NSxthF(h!ehy}S-3e2xZ0>CBdy>}b{9gl
zgQT1H==`v(S&4EEf>OG+0|M_^mzm@J2bSt6MaL0w7;#Qu#89iYjb5bd(`-aTv&2A~
zX**|~zQOz*sqfa#ibB|7ZxA?Df8RUEwctZUoXPWw*|I+60hkQzY4_Dt13jmpGx@gy
z65UWsDJ;xp-E7_;Qf)m%Q%Sk)fbfP;>2(XLUTO^=2E6kOu`$M=ehb0VyrVh?J`l!<
zI8GYh?35@?bLw~E&L~Ono?-=7tt#&vhG9E^!p2-TKZfWxw)l68TCp*Ot!wn4t4+vR
zxm}mhu}gug-N-(PG3XuNe(^1(I&y;)9WSpgn2S&(gAW|@!sEQ~&ATHnuP;bxU@bGr
zdmSxY!Xd#=US9Ikj}8R4@K9lRdBw%^1CR5}%^-YwHE>#BKP09goaTjJef2H5Hj0Z#
z+A$3~o?Y#6zVPDmg0(c}*>SvAa&*g?_shcF{jAo<`-xg$l}4^0k}2LZpYEt_<ziGB
zqyr8Y7d(IUicg<kaWN&{et2NXMnbjg*rlF*-8}tgfAtYm>jrQevItMe9C{`j>NUBM
z5*l@N8balCoaRIu+1l;Gn`R&&I;vgG@p765+&JRR??-bp>UYexwr6CUoBqwO?>8n0
zS`dsWa@dVH8qqbz-N@BA5TZiEIV+^P8%K6yVwgtt{}+&^sey!q*xcMnQR`U|q^gy#
zR_Kkg6i&;GWSxiZhQu^QWk_4~<-O{cl|5}8BUTtih07{^psWyg&;brzpH3|*%aUGT
zV60yy4|*3Y_{GRlA#JlHHH@i7JkbiTRQ-7|2F4+BoK*s(+Hkd`)uJdE;Yc$VZcj5`
z-yiw*?nG-ahQ!djH=G$X8HuXtChc9=9{Aq<ikOmNn266_<Hl=Jny4;PmW8Dh-n{#k
zHxCQ#{TmLu9g`2ZJoEOePburH#Jgc69j>Sd>uKTY`i9f-k?0!DWpbJECRGg4sS{h7
z3CkVr^qzHoBwajXJ!Rg%`M~|{2i^F_k=7s~xU5D|9w8==w}+xv<N8E)Vrb`(tFL{t
zE??!A%5V;$cSFwZM?Z$1c-1$}a}XTjtQ$qZEPk~{JQo9<rHxY<FiInZz^G(+%}}U4
zF545q#yu-JXIw30?-)bikQ~!~BuZn4Mwf=5vfLrU5NsYOVp1}o2$japK3m8cIc9ng
z4JcY6&39}?A?qgU+*n1CIZMCu!4X2j#{usL>U<<GGr4-yfme~9ltyaFY$Jv~-tql_
z(T&h`31lnOnrTuLtw8LsKf47+(K9_n8)fwkO9XH1d6BB9k*f-twuT$@vzMH86kr9U
z^O3Vt;uC1M3EgH~pC7R<JJJ?I6k62+1PeooK|Lsf#hv#KuXV>iEkYau-qb3-(rLcl
zpQD<c_Yg<oFfk01HVYflWgE4H-Z-f^s&TOFkQehA*cyzs1<!111J0X&M?s_(-4pFV
zQnIGE&e4i8Cyb$yT;}TXikKqnGPC5B`{RjW7&%;A;!<V2UvU&Y@M@)bNW;!NIAFKm
z!)fNT=OaUOT<%9cdU3(v1=kW+R|{7ma&>dXuirXe+)NyEV~!4?aIu?6q6SP&gd%G1
zTxWe2E+@|<S@~m5$JM*tj)%ubHKvL%gvfEt%%yOug2WLYAxmL7oe-(qJ)F3`z2kZ}
z5#t-u=U;GfbHU}+HJ7`A))rHIspmjxYWe-N5O|u&gRmjf2J+awG3jd3hOalDD*tUX
zr)P$Qh-cK#Z5*9`gF4x~H#Madv+{T6k%srgm<S=;GxXi@DSQW>_#0o}JHpsIAqE&t
zRgeM<F%r5Dvvaf*EJ!6XMo$dTjAAY>kkUwti6#ow58gA4BhZ+ScLu#^R}t}y<G|t2
zSmp%~q6?&?6?LtR&7!lZXMew_4!3pd(Qe>ojpXFaW_RE$v;L`<vqb&#4$EzGA?rXl
zLz89%Hgtik>&j`#oR&;$mEZ+&9v?IcHjRVA#JNyf+k#h=)Z$4);^s=@M5VzHJ(pAB
zu<K5VvZ*j?GqtKd>&D^+vEnN1uW?te@Y4mjh--q6aB+3XCm+A&FaP-8g7-YWe@}Fd
z*Ea`VzJ9^O;~n#&%-)Ne7ks#V<am6~m?HDtw_F_#)T|D2F6?wGtQl8lm_IPzeTB@o
zj3MGm#vKk653Q&RU5imAowjjRUv0o4+PSv2Eof;%FvUfG5Iv5b%u}W5_48^r)Map6
zhwS^<LG1UtW*Fm`s3un*cDJExwa?P&PK9DbGPZ(Y3~H^qT6Y^-Ww#r-+>L5zRO{B2
zu2t|9h(SqWE#r%FN0dj?2$sOOxW*W)utB*XSJPlB3CBJOhv>Jw3aM?gYsQL!(Bpk=
zoj0^VE=^g-crB`$v*N2(tGP_!(DXBvvg)7r?<MNZ4o1#wdZ_wVRdFNPpzv^JH(R#@
zu&x;&Ac~UeLU3Cl*YBP8e%ln4w5e~{H>h5D8o^Kj=P;9)_GCX`Lk6%l;dM3Pq9;l+
z!lhj+LzKK_%G=hzt(|ReHZ()?4z!@}Ba^KgMAwz`9tlzT4k*V#`k2&p{HW@+6HVOo
zc93*;SMPMR;De)zS}|;Fcd;Zz=eMHa3>B-?Oi`_{r6-Ot((;OLnTw0Svx|{sp1HqU
zdH?<c52ur^kJ`wsu;i+SS|QO&R?=v2cKt_c%hUpg-NbItG1w3sP8yGoGXXg4Mh=HP
z!3(afT<!wb2WxViXH1dM0;PCrTPaIso>y>*tKG<9ci{5slH=ovhli8quQ*}aO-R;)
z1QAvd?oTUAtGFl(qZ(e#^TK?5WE>JRjd?8$(G%9pw_kqE>yKWte|E#=^)v2IE4fxh
zZrqt9=xr?1*Tz$Rk)lOfjCtoa&l_imqDNDEb7@<%(Hk4Q`dsva*xJi&jMoqSo}yTL
zLF$~^xXU@Cqe%JB3d62O`b}Qns~FI*oJR66CeMCKYLJp5(M4s#0xl+6)7hA}O!97=
zv9?N{PgGWLL1EHy(C#Qk#Ju6x)2tu96sFyd{S@&oa5_HHN+G0Vh8lj;=z<x0^tq4C
z(Nc{1;FvTAbZ#3AEHIn)?Qq`)w&?H0s7Z3$OAK!7(701#oO6L%H8{s5wL0EUx{HM1
z8HZ#hFcF7Am(u8Y_ih@wxorH-r^4&&9rLPV5^u!2!!&Yr*t6SB>dxlbV9|;(^eynI
zH7$nrX^$H&Anvp}RS`VSd;ZBE{Ud(yi(fF$GmlO1(en@g{_pd{AAU-y6?`J39i_p)
z`WOF>%gd4HR|%;z>kv3xT&tb4cLXwRejuMdkUxA)JKo}2W_tdTHeOI33-zdyIzpti
zK7Ui$SP03?a`b?zt<tK_%?RGqBDT?D5<{>9Pt-@jt_98>B(I|;*Ut06e*c`o(emWL
z2)&cF``(%Ff_C3sz2Kc}&X&6vxfm)^tuXMy6g_Q-+$7I?@2F-(H3iT0F7e_pa1nu$
ziB<jYL-1Dht3FwC)Ey})^gKyrxSV(<0!|o$6}Ujjg|#S>;OJ{Ywx*yf0R$m$-W)-V
zx9pO1B5L3C#CP&!Bj>z*FO@wd#4LohKK6jRUqUlNAvz+kgGdz$-MmFnn_7kDM5U#p
z>Ame0oYPcObRC-)<yMGwX;=Tey=(8x?oD=Qql@5m4=A;$FNq7Z=C?y!YreMqcLTN3
zeKIQ8<{j1*HnEOSbkp`u(JANE4N`nRKx~E8MlN~V7oGItvQXBUysngzRiWa*`)KoP
zcji7Aquh=p{jA0<=tYD>uYFTCuKvv+mc62%b(b91mwRrmFPTp>rz70?NE3&TNr~Xq
zQbS#&NoH9xgIC0~h~wdS<nZ!_S1+GY*O}mStRc?x?b~}otz2G?TwGn?Lm-#TlpL>b
z_B00WCnXgx%StVkm^4q}v{nLz-I%!A4NS@NaDT^|8$;9#khNRbJI^o<<RTo)%4uGx
z&XJ;y(42!(R)%39*UItXk##9Nd-aTGFRn?gQhij0KX7?<!ToWjl|nJIs{U;BF~FG@
z@MK2c{T&c%nr-0lq%#*Y(4uvsZkI>fD6EGT*_SvGk`L6F*iAYDX=pK6n=m*fvUd`?
zITzXfrTmZ{+x?rozE?5u>GKO-KD*$s8<~cv>4nK;gFYhFHZwcGXy~D7RisY?ZCO~C
znKfsvfxL$?(UuYM32{zS)r=%qiq5JA=ZIrsr~_gtaLH>fw0by~tpo8<t7_ftduL>f
zQQr%N?oGbkDXrkp-QcOTMCWz7QEmEkd)U<4HfnCQ51PJ+(5UQpX(Kp&csMak6FZA1
z4t`L0)hkr!Lt+;v#&O{0a?iRfngEHjd6YMqS0sgqdm?}75wOzwVW2C@hd`JPNElVW
z-8WO`@hR}5AHL#$_s{=3KL6$CynTPm<@F{1;17RF$tT<rsq0Fe7k>TaxBT%R|7-r)
zKl{hjx^n&OHPhjeH0~I1NWMqb2ioxiK0o4JBaRcpa0zi>KfU1e@?+NZSGe~xwcP7{
z3c8*75HVZHe#;O`?2@LAi+Y^{U02@g=xB%v&YN7Zn+Nq7a@!t|elHc((eK$i>DC%C
zvW0{QVj1v#{}+PgxYt^V#SxrofP-hhHo_P=q`>HeR#!e6V7QrZuHi-FlrQ%qyWmJp
zXr+7CpL53@ATC;@Ds)sL#_p}5t28)sOGM1R*E&?M!r0QsR^9er*a~KcUNhH(R%TjR
zwMb0?BoUlPTu|b!uwB0xBWY0I0kL9MYE}|naD?iiIAz>g{Ic^N7Xy)?9EYk>MrU3F
zE+{(C33g(KcZav5Q#Z;Wj_MOGq%#6(g@1o_;I7~MZa`&sena=1cdF8e-uw{^R-+<a
zt<u;EOJ}2tr}aSFawhch=(wYgLkbm7^X51zHgak@PE=|lwt_)gq2-0TES6*`#^2O%
zwPvcAQp+Y~Rg<zOU7wiR#!?Zr`P83TMMICA^9(++PmX6-J1#FT@Nppch10_W-+uco
z562U05keYROVOszsg!#^4op-o2FH{fubv%n&hhU3d#?78aom%0VchMsSjzW2zj;OQ
zj*_+c^G>+jk9>G+JbHLMEgFjE)u&=zRu20eS0S<+CR|lgcaN-I?7%;*RaITdYf&kl
zwdh`w#!(TavNF#n)@5csEnHk)>A{;u;`MW$|M<7K*pD1`BfAv1zPvzie0aF$?KfYM
zj|(D-HmZD27r;|4(Z(wfOo^bVh@ydQD<*njrt3g`Kh#9d?Bhd1EGE0}ZORGr+GI9I
zE>NmwKjbwNz1v1(0OFsd?z(<%{00ZRy}nm5@bSw_K6-IO49VhJlZA75+k_G|XL4R~
zO@K%h+q#{4vxgW-IaBMZ=07PCiclU4E{#kVSHxi=9)xvXSfnYrFojM2Ra%8ra85P)
z%AjoqCCc`|11hE2h!nn$V9v!r*I`-**!|pP>o#ou(|Z%!e0&I-rKlz>>c%2LXTS<Z
zbn;Z=#q|YM94RFwJq7I;qeB;H6Np_e90|LT$xKKbZ1A*dQPsjyxe=O-)i;Fx`Fr0K
zOai0~YyqhR@7V2kT;5#rJ3sv~fy(F`*B5){+qca70k@tw-aqiOpZ$Wb-@N6Y{x^S*
z-}>zzFzzmh(_XLbI+M#0Pi62PA0wqg8j_(zo^gMT-;M0{m;Ce}{kOdN(eLr)&;AV$
zfAtIUaz_m6zPB-QU5D@0VXjHl?y`Gvn|6EaN}a-|s+sR)0=Dt%1MF=C)Oi$ra|iHE
zPY@T+DV%-t1CUzP7^<%`X_R`NEj#Z?sgagK^A6uKWMOiJmy_e#4F>f$opnwFyCEXA
z(3*~c)>_qQ3x{)&W(P|!32zF4lm<h5K&0w*^PX1Da->~Kgk<)*xp10i3oQz2?OXt-
zW4$3oM5@LK*P;PeL&8a*h!z9EMN>GcG%cmbV8uXRZz6;0@@s~o8@`w`pY#+-^EL26
zt*6zr#h%p=aV}bMV8xy>=$#J)duDaS;y3$c|D;Rr`fP0m+V+h05rY0TMUZ?*I@Yj5
zv=^HC?KGo6sAmMD!^9g@YcGeUU=pn0wDE~lTJtoo|GkY7(G=6#Js?C6i4H{zLN?_}
zskVOU4u+<hs$w1nZW}XfSrf(~P!lD$z4xxN!C^iy@cIzAOi<^UQ(n1!`<7Z(E-v=G
zxyuZ$vEL1dH6ptqa=G6zpN>q@c(EUNbv>~k6QyNtt`2<s@&@lrQk@cS-{11_vrB@9
z)A6Vl-D$*yz_JwPwXlk3Db3tv8+pwXaqN;lhePr_e|Cirjv}6U&ODqJj%#6A8{P@N
zK}@;_gs7v3=Dayd7MA(M`**iEAE|QB^|NOjUcBbzXFuZkC!cV6*ps|)wVN2dBGoTH
z`Iu+VuldtI{TX++_ZlhP9OA@W8L{Haq=U$24CMqX22?8Oohk*?SIB7HX2=Jg7J?Qv
zEg&r-O?@vsVvULNhjt?@xp7=(CCIMILHcgv)BnAX!-Z}v_M5!ES21vNeaX$$B{2@V
zxq98?dY-wsfCQ~(^19M!1fR6p5jEi$Lck3Pmj^Z0@R*gUJ$Oaa4nqp~RfyivN~WGb
zLL?%#X`oh|#>DKKQ~%hZ%WSN(JKKNSG3l)<cmgZ-gWZXyxplJ}g<!%H*_r;V&O(nN
z7;(|8jLp!)MXhF4mL(gh(1$>A4vB%DrO@p9c%Sru@g9j7+^<{>=S=F@5fHJU7k@Ti
z>cpa&O-_|hc&X4T&3A=?J~;b}YhHf}hs#TT`qSTGJ>Ik1X(Cx$?>RoaW!fkH$shd#
zKKb~C27#oB5O;W~lw8!_xSpu8!g7yaPe@w{Zlw^Z!zFFJrZqK5`S{0wi)SBw#xMWm
zU-S0g|4Z8P-sWzN);bJqb1|!oy(Pz0QKX<d^+n&48TCQyhe2(9Ziao2Goqu8U2Xew
z?3x8bczozej7pI6VWGrCPoE>#s<Y;47=WVpv<Xrd90l(xhao^zqNA=C4M-Ydz$Z2A
zX;tUXxv9$7hotKPy{Du1i~%$vSd6%*1*3P4>MFq{<qjk@?Q+hMJp@UtMdf&<IYRMf
zYNB&Z5nbQGpA-C4o~qXy=W(Ki_1TE!43ln$u#Mhy9UB_an^uQ9S=!uub6hk8N6lx{
z$jW;q<th<He?H=jkEFi--uD9K$tb|kJ`B>|q+tDh8f2^Ph)>o6s^#u9A?*S!4U=>C
z2WQShG}))_dAe8VXR{`yLB&QOrQoZ&59;Auw?l*&als>1`@n3RkDLn-z?;{CJ9{xy
z<DYnY4^MeaXGEldcw8sUI*dzFR5jv-yCdY|BOg7xB1W|wZ%bx=JejkdV>umZ1e()v
z?8O)<kCp2~;(G76*g3qfyngnaXEzr}$>hAS+wJ-G?vB&CFvdOWa#Do0HmdWSPA6_p
z@Xd#rx3@=}^Tg0J^i3MO!SnL^z{S;`i*bi{j#E~<<1gPm^5K}7OC$Ke=wKhz&t@1C
ztwC*`wRuFe2$i<2R74zWh270_Rd)b+S@0=v|9IlT1&RpQE-+rc;CFud2mI_m{YU20
z1EA(WWjkuV%i50$8Bki&%n`{1onvmDzW!guJ1d68Tsyr}8L!hrT&2~@2kFG(F%9f@
zO}$H-P)ep`J;%0Ukn@O33=r)+o!{j3y^4Y0!xMRnH#oh2NL#!_EJRCX6UNp>Ds4Tf
zWC<aRQRl^Gq@gKZjl2fpIB2ykm1NPYQnGd0`m<USvQ@>wTb11^J}UdV3AoxJv@>eZ
zt<JU%|B0-u*1G+-rxPY^W)$P*Ue>!tXAe%Rnu*PmDy<*RUAG)!WQbknrfP)dL^A+H
zC<gQ6#eTMRQzmItj)^zp8G|pIM%j9=ZjWbDV?$CPNTaMr+Imzd-o*{J4RO%THzdaW
z1>^33l;@Q7k<<Nq@^WH04E*TolEdW<X}@RKU*pHUR^d(A&$U(j=>yB@9qZj&?!I})
z_01Jwgz?1{alE1qSA^jb;zV<v>2To3fA~*{z=uEif78lJEi1Wb9)WXhgYv~zbqI=v
z6oc-uJK#IqZ-qb;r@REQg3r25@6ImWPi2c@{ssL|>rgtNg&=2HsjBPx2B7yN#=ua8
z6g|0`eXKnT;-jT<R)P;i%k)qSYUR7AQo*yDp;3rw>*Um+Cun|#DFej5x9^g&W~M}_
z+GXhSR-0W;P35$uHuiWz(nr44qEN0h<?VEfUZ<{$;uAg$D8r!Z-MjfwGxVw3pLPgU
zuVWJ}nvkd?t4@-#Z82DoRNc5fz4!>Y5FtdHb=fbcL}fkb`!UzH;$7cc_lba7&*a|L
z&LLC%@RS~tZt_OwFm6sHMihDKq{}A_z_w>Nb6%7lw4@ujk}4e)M-icwm5^6bE8r6#
z#@~?dd{6H+he7O6SK_077A=6-K94;4I$fXEn|8=rf)!Y9n;>{p86vIHN@f=Ze(>rw
z!3jPH)dj*l6QXBKp2rUloJwKZUqNVisay;ZCdaF*Jt-8t3ta4XI2!9bL#ymAE_gT<
z{^D2P^7?Y8%~4hKgVtDb=3%Y8xv#2X!sDAYl@4yMF1Wth^Zc-*w8AN`+~1ve_gMJy
zZso(=s6eij{iMZ|I&<oC&>EcPm1fD2eZQWUnRPAf4i}`$Ytk@)^PF<wVVMzeOw*o5
zAy?toAHHCpBCnoZ@#Ek5J$~_@{yj*h)I#Tv{B<d7#rPwwDlyop3eEX;Mz2Jlrr>nl
z;7o~t*jb|X>zZA2HuefGSi!E}f0-9W9-7kzu^ERccx{Fhhq9vlyw?7v*Y_k4n!a^o
zYZqEo{cErb0Zv)eA(BPY3B3#2nbILK%{uQDcxy_!^VPE~>$b>i=~Ym}1xE~GhAM6g
zmZ2?5w(8^=cRm<8SketsZ1m1k#lUwS@+Uljz6mKINIMigpxdU=?lF`ijVxtj^!AOK
z-kEsqe4GnLG|{PHp+SsFDCtmB?NXIyvcWDb0*BY8$$6@oxu{c`>dS*M>Y-7^Ls|5o
zEURvaCPA!{l{qd7&Bruqu#Oj+9|+ThI?XxV_|mi|j8_c%XE+zgcW+tlzGm>1e0)oO
z_YL>&-yv(}t6zQ2tJlx?@lSus!H*y_%<q5(A!$Uji|nR}&wl?8@$1a#%b!!`d(A*-
zX11}7c5IV(dg%DB|8H0rkAQSVfQ@aoApjdJ*Inzn=DfA0sh4P|kKbf)04nKn;y0oq
zbi$Y(c+Tr48?ECMG$hJdF^A}!G*?}N(Q*Q!ArRU$1iGy*HVedf0po6Bzu$v%)cMFd
zpG@l5RE^QhNNLoL!rLM1>>!OviC`k&Ya=u<lA@B2s_O=|Is%bWo}9gw{1}t1wWiOF
zU574@q8S0O|9;2iy$5tHdgpZAId8=sxE?X>b>v|~9#w_-9<pOf2%86KgVE>qy-@tm
zH##NT^+KYn5}xn{tjK7fV>A0;iuweEy1j>|-~Ye-vpl}@7DU%lL&L}fL+?yB7P~v7
zt-+^!B%D)}fHQKZhkO53t!c6XR{VwtoL^50f+pKoP`M0IgUhH~j&K+Qr{e?9u5a+k
zQI<@Mf$M!@omcLIBTJx3BS__=XBXVv-Lt>hGemei&ZOiSr;*e=^O}_a9R_~+#TP8c
z%!`k&NYSwt(Y=PqH;<V=`}WA~40F-3-m__7H;t4!(0t(A`-QjfPRvcXJ1&&wd2{#3
z+8i;BisXbqE|u)nq%Ap5S&wQFE?P?p!tr>dltNxtmN`>I7_YCmy4bPKGk5o%;&e3k
z{{B77d_-zxS(In+>eUNAdG(y#{=of*ci=1{{k$liTz?&dJJIYLEr%7BcH4-@>4-oq
zu~7PC*OYG|8mZUD?^Zl`(Xp=+A&eult&NQ!>pc=M!XtE9HuBfr;@|N4Ud4c_BFq!q
z=H+S@QmnJ;8-hlU4#9{<w$TKzTy4kN4<jSu1g9OGlB0skaO^?zp~JKau~)CN#@58<
z<~Fxa3X!7!|D*%7Jt_r_T26yi2D-s_7WUTt^qXrBIDD6(9D+Ktoe5c>GB<@WpR%I+
z-}f$TovQOXNAjY@hPU}n>#TP(UFwYX-tqTE=R@Z)ID6m9-{>O=EjAodN19KpWi^Jn
zq8DXdky?~-T#IVQrILmSl5p%uqXrBG=SkzpxWC4Q5x3m)?Jxc#|M5Tk2QCIjeSFV+
zd(RjhFP>lV+28s-4p$Syvsd_Oht!3*e85xD#J3f9_?Qws7nj$3_7DD7eC-|i+drq~
zM~YZiYu*c8(gxsz-<Y&LDG&!#vHM}su=Fmn6T<3N=yeiTzc=2U{kxkXNPS+c6}_#o
z^CSup^ic1d5L*jcM>I2ka;)pB;{@w`yi=IH$H>N*z}aV63(gm+Xt6Nt_QY{gWG#*O
zswO@fV=T1IlGc-UkruTWZLY^QM|{+S+<QyDRA|Fo^mN^*L@6S&*!Q)Xz={@JpgR(}
z4)fdJQDpSYi-9h)3rZ;TSP>S^hj3LXtTRs!yD%!_G~!k7@;fQA-~ZOsbHLkJVng_>
zfayQu>?`2U1<Hn->K^7g@5Dd({^r@=4S?Y3NTH(y)`W-^nyzOt==@e4qn)}7`t$lx
zgycMS(S`;q0Q4H^o})xSsSaPV6(LW|p8nbaw7q$pb%ELpIRSh^#Nk7t*2dr+ql3p~
zWgG^EDN(Cx*Y_#122U)D?%l({Zc5x-9;i67lAAc49vR2T{$fwAD{F4l7Km}+7eD_q
zTCTjlnRtGshx}<NJRBEhseJy;J%9OTQ7AtEtw2)0SDFvJJ*}`b9#0EH)Mw)UII|Qz
z*iWlQD5n@{tun^ZoK78WZp2tQJsycEaCJD4>^aS)AU@z?z{fzYCn$k=S@_kLU$Pq}
zzV(8X#{A)y$GZ=F`{rw=A#&L7sUmze4g5(feDeAw&tE*>$fDi<q@x1TRNfha#gd6=
zQ7K-zL85TI@;k)>qdY`K#}FtsicS{J=cTIrxYk}VkQ}%e@x!2MD^xbxq`OM`t!Snf
z(SNP+{ja^gpU;5U1KVTXEVzSeDNe?Y6wE~lt%?+xcT}m{K^uIax~KM{Iwy@iqBiM8
zKQn#B0*?~iq%m8{VDGXvB9MMpdq(8UAMom>??c>Vea7t24#BIWYkRzGqu8A8(7=@7
zz&ACOv9aD((QDp7+zx<_AUM+t$Ecdp4)_J1tV3j*am$lgV26wS)0?WhFWs)RJI{4w
zM7tb^L?x(YnOX9Tv=LV`vd*l>JJ$I~kiyzDq0k3)m2znSU>qjvdL5oZet65*pZ_^O
z{||q{v9A32&%WhyZ2bO@Uhsp@o-<65=O2B>xI3`Fy1}JALeP0-p7AbF@048XuRo#S
z8TJ=^^!NXWcRx#%_rE6AY;*QN*Gl)3I7FQRipXZy3cZVZ5)1pZ>q3<#J(y_oskcHg
z8Ye`=^z2=y?(~3?b}k$cOS6-6BA5XQdOwO9WqJEqITsxRxsFD4SXvL@&V+AG@UalO
zoXLBJVPF^rh4Wiw1X?SsS)J*=%W;4byS3>VN^Z_U3<P>1s-Gd4Z-%$=+AyST{hiMi
zMKbV-Kr&Kb|HD2eLkUlM;u>_V?0oMW(YP#P1X<lA4OQz0srPoK=>gego*h~h%OFwp
zf|dv3`vGZXi6KMI<^jEK_Qv^9PJ=f^MS}*dJ181NYgngsX>4oYJI;Q61ZO0#<|@qz
znWCFFdSHrp@KN0feK>z+Q@n^%gh~Az9H7!NRaEGq*E_sc)$pl3?MSz^-S-SDNJ0c3
z&F@6tO9%lkFeJyz%f!c5Bcpc=QOJ40JI^=`9Onm|b4<IDbe{?1$jh5coMiH%<l_4e
zGt;=^u%Ed7aL<@v8ZO!I4@hggeDQ+ixPV{T?e^TCj@;fKaV>CnJThms?!J9Ekqexv
zX49+!aTxjd*$wT~c*u@1Cd#^KLT&KmR-t*?x)M{`_QP5$OKYS!ay%WGc01NJBi^fr
z3s2Bpv>-2Y=5b#6>f2v)|L!ffzy1}c`#YKwt}m~6{o)0uby0Hix^T1qd(@jt9Y@$`
zN+s=2jM3<bS(nxdwPc!9TGP>G6CJg8z7Hs=Q<`WTy;i|ELj_zDN?A=Bt@ja!5|y2v
z8O{f4gBlyTcG961-2L4AO<&&+8CVPDv;exHMbnt;oI@Y2O)Fiq!i}~dTT|yeu?H<U
zPZaR}iBtxKgZn&It8Q}(w1+$lgOQIKDF%Z8jl|O{H4z(UC{)(}dl-f%51bwnHD^6M
zZ7x@>5(`oF)c++|&wH;VE9s=La7F`q2WSLE%}EeE=fhg{!WG03(_jHhwh-)dcVq+C
zZBfsgfb@Ubd+CaZzA2nJBRa23<wSnGCC3Yvwvg8&%W~2~(t492YG*P`6BxW0yb`AB
zdSZR^Ik(?@#k)7(@WbE#UAzbv^?&9^pWN`#t7i;j;&A<p-PLm=X$7JOYNiFK%aOrB
z3>i6njieEFFK9Uv;>gFp`wzJP>H<D|X$P$yp0%mg8|l6TJ;AM$r8*oY+x1eD7tOdA
zV`%qhqRR<o^CPfvkMulutJ-x7x39AjQ{>E7=qID}_kkY9A{&pUS{GhQvA3xs3*U=E
zlswp)Fxk(^c~$c)d$zn)_ij?Xf2hsWA)av@RnuMzwPb28<XrK!ZmI~q9AWFGdw1W-
zg5JS6P3D!aXNY$m41r0em~i5419p_UC~p8Q!gJ2HJ{?$bs=-F*?nEtrx68Eg1MT6r
zztaym*+xbEAZ&EV^L(&&VxAK*V=qN7wsOcjy1S=#-Oa`lZDaEHf?Mz56P>p^kD9PO
zOT1V6XN(hsd3(^iMW^=3KH@?oq{;kH&JtD;F~tEmI;k=cd?KY01VYe2Au(^1jtcZw
zFMKz?gmd_mh{K5Y8eUe!AywFok(>QMa*d&`+`YT!`f_B7i7^et7+6k~X&89)ig;b@
zMuure%PV8jaJp6-WAdyO9v+VDcN6=4WWU>MQGufrHS?Kv6FzwEA7_63?#TUJ<=x}V
zub0AKy*;sHXfASUf*VKTI3h9e_PEkY!$ps*nk<={Afg4Q*c{#W4bD448W}@UUlb=i
z%(F#f3&(j@-vJTQkSL`xKiu={zx)MPH`nBcdt`lNez;|sk2L4V<G|^3q}0mYd?bd*
zbU4t%(Rx%c9VK?>yc21S<{KV!H*HyqMtLIX9(44m(qKiYQi=$UwPw~*iBc$e#;r4L
zIpLcq3D@@a&@3Xnn-w`UzTTUulvv@~gyo6!@Nas3zha<OgFUS3Xy%=52OQp;IZOAM
zw=iI@8%eak7mQHkLfBluVvMNEyB%;Fn^?AsFJNBeM)XP+7fMs#?><Wv%?Z%MxY~mg
z8OD(q6W%UZEg7jwpdx8ukr&Moh%pgU+Padyp<vKVM;Dx-89tuvOTjka&eGQ3*$Rqk
zu)J#Bd)lHo==E;tiJtg8B6xD>afW$8zLU|uLE7k$BLY(HS&w%Ju+A$juX;bxd~;t5
zwQBx@6Gse=HqV^$%zXEr)4N}DJRbSMXP=VBiTkg<;78YqPoC}BA1)a8d*ZlLj=?Z#
z2Pa0vR8q&*Ia5~r-LE&|OUNs<%CqO!%sk?1Q6GjUW0<}%HVcO7o$PkOdBP2G>U+^z
zF`}Gxm)1pXq%f<>H&_iD-}qo((>4OBYz2o?dQPkKXyhJ$s*$`;;q>p$ar@8Pa^|b@
zPx|~%foGff8WORgQ&qLa{rkljEYQoV9L<V|Qp)DSmy$Yz3g`FG;mr=Od+&(DP7jpe
z5g)hpjhM{Su1W94e6y?%lY@4*K<|v@Z>Shn2-9>NS9<EBb@08`>SQ=A6x2w@Z47)n
z4AFbg0Y9<gN%saNf^OEZbTYH;S;keUCF`gPG5vNg92{6Qbthas9Z$hoTdvQp6--Xo
zg39|Whryhj9ML*;hYLzpY^D|{EdzxZCOV<f`^Y#<3~A6jma>xbd>%E~9?+YKr+b5f
z01+1kdbm{Ss4c>fJgF9<2k#TL36IAkAqqk=nVTZG(>QQ>@En(waf<B60UuzUE6J<B
zLNwE}X&e!0>~=fm<4N-#mX%?=<nHc~S{$W#zPvl}i*IlF_EdN{>8PhT$Km>#i)YXD
zAe?9JA5MA?U1RcIodC_hWf)auRYVk$kBM==10Tqz6V%MKA34sNusA;+nb(ye4UFT&
zJTFwKEX$G8LtxI2toM)H-@Yg3$~0Xv9QF)RM;LJ&2&M9O|Nb9xaeb|aZkMh0{ny4x
z!PdWyXnIaSfl`bsr&8IlCGooLJW~R?o*`>qDa(q~OvoqPy5MWpycefun8Q(Q?{P70
zsAKe=otc{@UzojysHr;TH+y|Q0-+Ctl!!j49=F3VIPJ`1v{|7eGU!aqDrV<B-Yh`9
zSGK3Gx`{c1D!rwQ3E1JT?B!YtJyl9cQU<jPIH_v?*zIUV--{2<pd9llcc>vn8VO!Q
zJ8CU>RPNO+L-VBZp)rNc<Yq%Ez^X29ke_xM-Zsmy6#&@bekLOILt9$g-c#>A)X}XI
z(`;b{yRw~-XkE<4V)fo)X0<b_MQ^$qx7bG7yEUiUZo<l`I37N{;o+Na)DvMEVc2uL
ze@8897~+6+o!MVpu;wFWexw$`4SUAjKz{obr?38!T?mZhfpNDZ3`wPJF&IP|wU~<o
zMTD3}aM6mtf|Ezu{g?P*)QIHuL@Or-SEy8cP{N@P3D@$*V%I~sSZCM~Bey-A-E;>R
zX%cAFJl>z&hwlPs&N@?@ulbJNoaf~F>L*v{7PM{L;H)r~dWO%>TKDrPLt0ZCTX!x7
zdXW&U$`>0gZF47w!Q9=GysJ*^{o3YJEy1r<Rd|R9ar4(duN!$XPVqKcOVeb9sm9-i
zm^Q>JSgYJeAFcIF2R4+ZuWf~6-KJ*H&t12|sYx>qgP{O+p!T9EYzJN0p1W2%f^!D*
zm6WJxWW%I;XoR!giV9u{{N6`YPKtHg)vyE6ARk){N`^k)zw`2N+lV5stMX4AfCdFQ
z=^O%m-&Aze;n4Nzl8plzQW`C<tm~>i1HllFD%8?!L>l$k_W;$93{ypH>-3CL>hC9Y
zZWTvsu*{WdNW8e#5FY1*i^GmYW?g0~jbR$iSE8{lC!}Srt}iG8`TdFAkhr=S2u>KZ
zzaqy-!SQgKIqY|=S@2$Xc)Vpf9tj@CsE3+sih8Z56LZPjEh~?wh11eV(~e=BNNHd>
zo{(C3xP4DqEp*E}X4JmjrHOHxC^c$+jf02z#BPMPL~1=DwXhp>|GvF{U`PYQI1z`$
z@$pD5l{kzPgj^e^<B^GpGz|RkZ~Z>4Ixfb6`R+Y$-+WD}nIHY+@9^SBKjf2}3#4YG
zs0<#IQ<6eph|!u}L$GSYRcn5m?M0%w6Pj(KUK?6e8j2W{OdKuzt~zgC1>m}xq$xF(
zmURni>Y7?5>+_P-i!?aTXd!kf6k^WXnDRGyeXn9**zXw+7lfG1GE?D;-rXh(W)RcA
zH|HqjY|NvK$}Wxa)K*xR1*9<yBg1Y_ib)r`wHi*KTC`|qiie~Qe0?n8eSZM8%GNvr
z)Bc*JBvvpu=TYltv3aKwzPyiyXe9l;&|@sc{ySO$pxZ@^kv?axt>RE)oNl1;)E;!x
zQyb)|9`JH59IP|)hL&_VbnD{0EjVdg58KdA&O4Qbd4o;0XbCErb1Eqc-V+*3DYD)^
zuu9=qfBsjz`8PjD_5(}OL*;Q5xY={_(G8c^dtTI!^x)2!!w`9PdBHw5>eQG%d&xK?
z#)B4DsgI0m!?j0R@q}rQL*YxjCeW$+LK(cYLY<9wA(=eiBg;wUGe^gCaYG0fG&g{+
z1gG#Stsm^#D)wvCyKDV-U5mjE?k%67A6%TrKR%e`(s&Ap&}!Laq5Wyvx-;pvqUJJ&
z*t}lH0Z%t;`D;?7U~3Km@v<Fqu8(`RueFEHb?IxrrvN<+M@Ix(;bI3~CsRh>)dM|?
zp_!{<hdFJHrh(4D@16cexHR%$X~yLDcKzfG={K=tN#H5jwgZS8N8`yg#P>rMl);aZ
zrR*ALg4?+hR&RlzaIAfwPFn1vgnqA_>->bay<Z&{tw_sM(KJwJ-XXOb8Cb-|Y-iqq
z!#okJh_6M>r9z5$uNJo26ci+DhJrKMEwvF^LvmJLRjsV+N?r>~E>wB6qQ=HJX|R}M
zwRndFmg%zQ{GDRJxid-_y<Y3JG*0Wnah|zhWEd`)ri5>qyk?wlWQ(X?Gd$d{w6Y?l
za2N(!E-dq`83Qp;*O@AgX?H=%8S(18n%6>|;QfbtE-xZf^$;0Tu*@3II3`ZVd(t?O
z<wUZs*u@D#V%YCBg|B3eAKo)R-fsn<_YoHd<%^X}tdS_DjPagvn(#i5i>`f#@Nhcu
z;civ_f(WHl7IV6dF<H1+HTWEs$5|0G5uUyJm^a^kMLph8Wo7h&pC8HhckDiXMV^JC
zup-_@h%tJG6p6ua-$$3At|4{h6{N5sXwH5%UQNw%E{vlR#_{JIoxTrv=S+#vw(JzZ
zOxJY(Z?(IiT6(AL_d`_wg`9>b4b6Z3^}ULL>*p`IdGQP%k{#|kWA$bWEJ3x-bf_C=
zvlHDga@NXb7zR>G_HZm9LQDf`I&1}nb}zQ*jmTB=Le1IACf%@NFegNVHEXNe&&D}M
zv5nuL!wv&FXB$K<I%+TiQjBI;;oR0`cWrrR@b^XOf+xDc;7{v_LLk}<savjk&>N26
zs@DTz+i1>kWbX)jDq5d`>nz(@N5Zxd>sMAb5tdk)U_T!R8V#XR7bgvmFrOA)|K3me
zvw!z<?!SJgj)Xt>gzMjb$!9;jrYsq#ynpj0F-3mww|~T|i}?Se?azKB-?lqH?6are
z5JOH?b*4L#BAZQhvztQ8lHryNqvb)5{zLud0T66J@PiFYGP+Gk-Oc%4Ufwg-l$jCn
z8}{gjwSJMsHX!mE>;ewY1y5C0W<>n<Uh6y3lj+(*9!K0bV!bEqkD_`t+xsoeb=QKA
zW(cm6L8w8qUMU2Df+>lbW~?ccb)w7{$ca=c=f{ui4oAkj*L3}XMvrUuYRy5l?@fb%
z>}HqKAe&|kMqzhZ#*IEpFEFdjcCu+=B~eb?FWY4Q^7(IU8$K9bwpzWNsQP|ys8!V@
zv5n@~aPF6j9{n2nJ;>7}|4H9$YyI_Sa@K17uTcc-McLnU=()Oln%I(B&6rPl&a`da
z!gtFyzxn#@)$y=hhMTiu6HJ*ZpCe<{=2o*p`YOCsM@FHk$e8OPh0-V1iUn_zT3}u>
z0j$;MQ@`HU!q>1`)Z^v(GB2<5_UzPJ<-ghO?~><SQ1A>z1)Yk}8Ko7s9)}la0${|R
zw+&9EWKxRS;AnI^+^9*QsyjeoZc?Jeg|-H~^Afs~q=7Z%Le7;~WXLf_&QF@qphe4*
z1Q;^c_ihsmV^EEh{97^(1U(r;;LH0XcgH>Z-s3IgR9NFms)=cufI??_+~^sG0pG#t
z@jc6VW~->5PA5u^+&w(tyk#6aPSb@_3QLS|J`+6TB+<lU-&0~_ofn3F#~Ks4K+1_@
zZ~6IWH+;#!FW;RxM+q6K*04N3v78=hIn#8(G0qb@L8oj8bitETl-54yNvW{gjZ(bS
z_e|5ovLtrH$ULn~)5O^K%=3b^P4gdMx7$%mVL6{w)o(~S^4;?zeQ<KFq)IY|ckkYF
z^TlV_TC}QMx(%%HR0?As82e7qti~D>u}PBTHZ*%HR9-EOHEly@Mw~`jqWYa622IcA
zt=B`4fvTAos<qpekxf)vwpeaxu1YqDb?WzP*Yd!>>#yH1bVy-rD-T?7YF*j3302B(
zQKDe+&eLimD<p3W0ugl`;ONGziGLD=<J{)iei3H1TB-P}#SrD&uh6=CK?+)y7eX56
z+!abkH<e8ZEz6u7Tsake$TyLi9F|3WoA9#b2q|hr%!L#yHMGr24<=#jDp0n@SPckz
zfr~d4_BJRP#6Rn)w5m?|>mjKLg1Ys{cJu-%m+kk!3*4>WYtug?1CFSi*Bn_}<{$i*
z{|kQk=YPd-|I>fM-K$r8aaVa@;@!KC*uLZ4$He~O3x0Gz(8UiB6Xe9WJE%Lu<GPWO
zW-1PA0;Obp7;%UoEyoq?OiKmAKp1uyB=*%&Dbo|_@&Rhb<V0L%aN2W`3voILl~D^}
zIAHyNG<AjR#QUk@+vpTzjlJeUY>3IGLD+(f%#}uwk>U!=BbSc#%FkVch&KJe=8yrJ
zS4g;MsEB^N6gjy5Nv}MiS~?Z`x(;?lEqac29jpL#;20U|wu?dWp|;uAjM#Ip83@X2
zw$>=wX*9T8a@H%w(WDl&)=%6Hx^YrBI!6BvWboX!abu)7%SjfJM386g9cT`tvgcCk
z^-|xq9YIUBe2%6A+b4(tdY%}y<>|wkbCxs3y$GWIf2_m$GkUWP?hR9g<EP3NS&JOT
zr6FZBSv3F~J;Y|^6;al9ShTWf3dGXdQY5V!&{QHO7h0}J6l|41F=i2AcB5ok)T+s#
zS{R|{n}}p#ZQA8x>h<$jN>%}?=oLySc)z(Ygw|@V1LJAAFbs|#zdBHXm?G!PimREv
zW~S34xnw$T!0vE<V4fFBoQSh#lC;M8d?Kxh-PqChf#58uWrpB+yG}e`CZ>7jFa}7G
zi?{5DAhpjlmZi{I<aA!i1#WK+*fsHfmZ)-@R+f1ow@jL6YAJLf;C+YlmN1M`wO%Ws
zR_1x4?+36}Q#J#&CC<-}8Vcs{F0ihVb(!^b+uAGTHN#nTpGo|2iIJk#c54cE4|jZc
z|DI);@!g0G9j{(L(DxSGBnzfix!-b)?1#X93>YgH`w}xLM@mXm3&S8Zf^Ci5<nCK@
zxw@ldS!k?F!bn<?#DbSX%LS9FrtK*W*&5X%&-wP5SLf2!YtpL0IH<N=MT?*8GJor@
z-!ODYE_d_+&d5b(n<OgvED(v19NJK)hn_@f>LuPXB^xv=PAq*T(@NGPz8B>!Z4(q4
z*%0DFt4ZJ*W5GFE%T(tP-bNmJ32+ez80<^Tt<jCu7;=(?s+<zJ7Myo<A-w#(W()g~
zp9A&tZ=oqgPod2+WPw)P753K?O9e(-CW0OU3gwnouS3TsI^0s=yxXdL3we$5=a7g>
zS-jfSDklS_LceNz+Y;kIXq44@rtX&4KmIM=eg6%YHS#-u@Oymu<!c7-`C^FV*m(8%
zYdUKg?`|<+#4*$L1I8QjY9(dMdC400z2p!0Akbc#sZ>!@yFkMd*BLVyY_#N>F@7fz
zsbyt-dWXpq>pbaeBGsc^Kd@4HdVZ!<xc}k{YFPwM>-KU8;@I@#R_(3wPHAjs6GW+&
z?ep{^1e2l3U!iKF8@IC-+X)5K(DD%!JTRNERpIYeuNf5^?=74E%dkh2i`^)9Z%SmV
zXKuV$?CSNRkzv^CXSC^1y;)V!LqtB05-haE)ojD%7J}4jmDlN$XGn7JWNGhep|*nW
z27SKW711EYPuwj)-83*_@eIL1Fq9}m1=R?h$F^Ztjo4-r+9)KhIKj(<b%nm2a|*8t
z7hg(|sOFMoyQ>Z35;+T>TonmH->2quu(M5^EDF`<QmJu)5{33q17j^sQ|gN3k2p<w
zyLQ;i#V?{{K`okwO0^oPvC^z|F(_Os9R$J~*BIsNuMx0UDza9i1E>Cd#$AVkl;l!C
zw@g~OrdUd&B{&SlT3i_M!x2otn#%3y`0~|(-C$`>D&nWf@$|!YEODWxl~yakc$Ovd
zSTezhz;_t;Jbrk`8dq#<^ucmG?&*3Fj;_nZ@pjMomv6Z&EBm2mSu^vp;=Lv2k<m-v
z#ac4+`9f))w=;bGcp@6hl;JcbmL(BWBE^MPGF~azh9rm%0h;Tm)>BjygW4^faAB+|
zk=KZ&(e(r8X~J4h&|H?1R$8f)Dritim3|zA(s71Zx1n5u<leq}hqH!Z3=F$HUw-~M
zKl<#Bo5LsqY(>Fq&E#y^?*?|`Kx>r?aH5do%(AR>&Oq4VgY=SY!BKIEl%_#)O7XT_
z$aR*K(N?gjP*cLFu%6b~6f;#3V$esD?GBJdqM8O)HANfE@b4>*IMqr_nXOiH!xBZ<
zWUdpCJVuRF%k}Wxws7w~C8w(bdHYqXKu2Y(L2fVF0x6IYn7!q!oSbe0){+)p%4lj)
z`azlvF%IWM<4{}GZCW;*Tq<kKq%7Rx&CLyX*&yYC%-0P|l+qO&WePL@ZdIyjS9HKB
zQt~$y$T|?9aL>l=Dh;urBikThtf%igT5F`3Sm*f~Asf2QXZvC~+=dRL>ekI>xNWom
zbv#o1NZ18l|HyOu+2_pjMSQBJQBtNZ6UGg!{(!&xQt}JR36l-B75X7l<An1a7)P^?
zt{Y@4REv)>P?f{VF_Pvd+~q_!9(exnj@|tmZ~;T1EGO2>Gkq6GMT9;zC+4^?j3eRp
z$ouynF~)HB<_kL4(P|Rc3I^{-YI8IX))e8cn+iwu^>?~4*JiM)ySVz>n^7C+hGy04
z!({ZJ{?yH}{m&OP;w9up-v)WF>lPOBDWu(?Zb@?6wr;L}ZhIiAZsx`qjPEd-URP_D
z7+<wffm|q61H7zElq#?@3N3Hnhkm!#UMUIrUK>G+r;JgNo(@q4ls`8z(`*#Umc~iL
zxr(s`>qNpQ34Gh3rQ=T3)K2-~{>IO}2@ahKZUMC|O1y8b(&8st2Yr6kmi!{1llQS2
znZ2B~c%xD1`g)nR(HV*kvC$l2F|N5$oj5SObwbG{$r#D{oI;h(k|u*mDU#$_l!_4v
zi^LJ5Hfm!e?8_KbyYVtqnzju|+FaS>ds&OPYb3fBkqwkA!&2Ylc6Xw$8V)!=uud1m
z#s-(uOp1{LxX`go0{xG>9eHgm(*<V@({#a%J^P1Sf;B9cGyCH$V2NuHh?*w!LM$AQ
zN6u^F{CvjK$Yti^T<O9<TNBUI!hYy^IP7Vc#K$S~&GU(GPL&T+qB$?kdvT>*KN<QQ
z89@ryHFF#s|M*95=!56=t1tP}ub=tF$1_ubE@$S;S<0p`SRIN>N_w7(rX#K^UIpmV
zk<$pBh!rL3S<hfA+Pf{_EQB8G9Eb6i-~IW|`SF{3Frv`k8~`~dN=^)YN7s2$mP%}j
zpvxsEDw<wtUeLU4_BE=tgT~fCN><BjEtp*Ne$yd9myiu5uZqkWO)7msN7?v7wu}rx
zJpX-#4q53%H?`rM*D6!k{?7(s$V6Ysy7hh<5R*&7Wv5e1y^3={F0<04s+A^fZO%Jt
zZ8%@A-s#*#RAFgs(Z5nEBNO(P!ep((ShXT*+-cJ?oZ28&h|N;;04;cMy;x~jNioVB
zP$al07`p)rY&nHRw+83Tb>(WcUPWAX+pNq>M!I*7Rz*y7-9{BU-2`B*F3hutGE2_H
zC~9YCEhfv?+tLi>gZ)B_Sgm>84%epKVi}G&-w|xUI7>es=*A-$OU(<<IWgLSni33q
z`rVyCrX|oyFHX<2Kv<}nX6Kw$*f+_R?t8#cOCn8Aw6Zdvo+zgi;r530?OT@fBg5eU
z*3x%@5ZtwzwG|1^F(opeBfGoTJiK{BS|V|pz*&5&v>NeY5J=5+*s#agQE&p&@PZA?
z;cMO1Wgr`2zM?1!QojsU8YY7kCqvsVR^7Ze=(=u*fPTK@!F)jrXdBdv9QNB#N#MoM
zCd9zDA#cCe)y)qN&04JH+S{5X!$Iw|f;Jhlk>D)nM7P8CJs?A-o=bXIY+@#TO)5Ao
z9{YD~tY&+-jl!7vvzyj%6)YLrf_0H*AlgbQl}eztZ5UA+tLV?%+SQUJQ{(k7(yjS=
zps>a93XCt`Ybk<S)LO2w&N6HmXhj59*5iVa?Y&l7j5?bs@}k2~t?i0f)>23XL3DyA
zs9VSAvtF-a(amwNVX#VrfC#->y9V5BAF5BSoZfq`6wvm(>w8^lBA1BCQMN#1G2Vj@
zgyRF@_=;|StJ*BlT+NGI{x;6IXmGXXXgY?&j+`UZgtJ0<)>7#EfU^xB0=vV3+xrJ<
ziG(gNJv~t_6Z^Ycrsoq&OdR@-esB!^z%PFFJ;56G4>vSh_~G42HeDZ><H~ZLIWLK!
zb6l3phfCpyr-@4{6847V%?MH~n|T;SfV@l>T5ZG%IadC~fBFyk;`6Wg2j9QtFMj#R
zAN}$Jzj%C7lt2blZwx7CCXMPg#)y8YiO8|<2YJFaCs-*^gtG2}qwjjW-v&ED5llbw
z`Hz3ZPk;6$$FaYj*Urg;kz%9xCh3}{QB1|wD18LANDG9qSCnsi>>SWJOS4>QEtCpE
zhf6e|&w`5L9mc9Y0Hc}%MYjEx#e(WoHvN+7!}QMc-tx&K{XhBZHyZ|==mk>F0;_&f
z2*Ng{D->IsY}?Ll_B;`_x$6sCHO^ZPV+`J?_|ED>Xw6G^IfkvW*g6T2*aT3*ff^d}
zwQ&`USgnlp0w2-iYSlPbnchupxK0oYV-5W<(scs;lv=1k>>}PvZnQv{atX?W22_Fl
zKqoTWsMTPbSBsOqCJb&LjMn6XX*(~ze9b5TIRLFv9ZsQZ;|-m*a|PcKyay*2*fx}H
zv9Hc|xDfOLyb>HZ7l<aXCJ4>Y^}PtLdKtuW>bP8>n1OCIjCViQ1I%Ii1FZ%zZ*@CN
zUN9w7^Ga=nR>e=d4X1H>Bwe0K&mUM$kMz3()fUqEjNK1{8017426BTi?D5VM=b5xz
zgy+qX!^3B|eoyEIcK01|nQ`l)n`9;CL>LdWma*xK?H@41olef%R$Q-}urZ$sNbU8I
zmV2{aFNMv$QJM~Y*Nqa{@bv+d%dB1D{wg0weZ>{JH&R1uG?T#SXXQH4{|$l9%ZJ41
zrMR_ff}vRj3{@Ag<<xIfl?k0(b|uSD)g&<!xDGP1VA$(Lh1T!>yNHMRgi@0X1@TY^
zCmVZ7E7lZjO@g*KhjT!#tVz_)Tdiv;^@T=hO@3zGU}}S$3)WW2^)L<Vh1xTpyl<Nr
zSCE1tL1bq0eT&7A+DJJQ$@JD^TBeqy6r?<>a+YkvhHi5y$pNiY4s9zM+tN!V_PNxo
za*rlj2!otI#=JcH1~8v=ljuL@8bs!=n2Df*m7LO*GL$;_9bG>X#v@^O#EwTy*UQ&!
zMu&weT8+5iO5^S@^4V?R;qJiL56tVz`=<*I8C+#+&Cn9B@9%kdcwm1xP-_$TIz>{2
z!~H!uMxGy^RO98v`gc09u94$SM`^G`saxOeG#a;QPD^c&HJLGIytQ~+$Z63!dwI{T
zs_;YK$#PUOC1nbZAAXpLC2;rU7aZT*vA2K4!S%d)f5V@A^BvEZSz@;pMt25c7J(s6
z1nVyJ#6{|UMPP0yxw7APbYVxyLch+J6T{H4-}PV&ySsb-@gM#!uU?H%EY`Jay6xsz
z*&HsVN)&&~E4CJKrI{^)8{NXRdfPy-B6i-a-?XMng$%q7E148HjOJ}vM8ux2Xt)ZF
ztw!@V8DMkSv@zgJ1tSRZP6z$J%dg)|Gk|)zLAGW@dA!*>%m!8J@3F>VgLJW1g*~@T
zxo#FGxcaj(q9)xk2GGN#ZPStt1TJ{2Q{Q>3lGp6CepCg3*7sNYp49-O+NvbD)k<gG
zm2Z*PR)UJGYgprOZXjnT!X*sWJAw~N1yte1Ke=2eMGkJYC{?2gbVdtXth@TVMc7p-
zx!^TP&Z>u&E&Xn@pj}^kxj=#sk|ht<VJn0Vt2zd&0Z1aY+CBtMOE4PYQt)iLh#~-V
z%>?i0haI*dd~E1(A<&0jwxMdVEn}+%Q)$i#Cs($E-;k#X(-QOh@3=g^r}y$%uwB6U
z5pO$;*SgveFkaFiTN8NMdqYwP*E)~u2F9Cv2s=tG?1Ja=^iim!yu<Hy)RM7opp=Dt
z`kGJ{7#>7rU6NcrO`c1PSRXeC+^BW*<#X1q+4r_)sOh@VZ}587x7#8c@Ah@7h%_L}
zgIMe4I597X!L}hbuAhBWo3Q0^NZ`)Zr7u%I={n$S(Yvr|1B@a<%DXm!ze|xu%dPK1
zzncD?RRmFNpE+G}^`O}_2irEQqldgc`udet=sXy*_^++JXSq4zS_KK%V#}*qv$jiC
z1)Rc~s88QPvyN&FHCM60=y@PINkLyo^4eKvuAXg4e3YNPW<d{fMi8XDp^lAeyuR*g
z1^dKCYMdm7hG6M}#d}HL6b*rs5~V7t{i;qUX%pQ9q!KAB&+10O7_Nmat|)4$V%D}%
z6(UTm*3cauINW`KJ>KI&hwpk)DlADIB5@|K3${UUUIO4!qVv#)hNF<yh2=6cO&8EX
z?BmCe)LQ6F=8K0TclWm#Z&=exnip(oSZ^7|k>|%p;+il{((Kk{p#@9HqM*Ml3#ByX
zb*1k*xr9r>J8=z7)5;Pv&!-EQIg@gurY&Y$=of=ktrHx^Tkh^}dH?nu>%8J<#8~*|
zkN=XN{OB{BrW5||Kl)w1nhd}9?R);aKl!)(;@fXoN|k#1QnnAUyq{KVghrha!8v+s
z`SO0p?|gB?Za?yW`|tjg%d*mHqVGLoF}M8Q&ws)fcRNaol!~0KPTuocvi@E5x)hpP
zFs0y)=q!wp;l>(ADf(F`ayJIE4c2mTm)byusWdseB{L_e`dS8^T%_1f+;H1)DWF<I
zvL#BX6dhn|1>bjceuoeJzn@{iDyMc$_hP&BKk-FccZFUh(8udR{E{eFOT`9zosMwr
z<?pc4TVNGV+eBHQkRAft7-z`Hq`tY0H*OXbf$%Ti7s|1}@U_YrUb$PZe<o;9OD30~
zLrYb$kGuN$*`Pzca7~bH<;sQWrihjB9pOM<0${w?0mNx7wA$DjTerZAD3I20x*=^D
z3Rgp-QJ?(grZ56zdgYZiJ$s{JP173PPQMR5h>F0t!C68Wgp*T5Lk?m`t70XsX@Rmr
zjgV(dURlqNl(-P*Gn7c0Pr|K+j$wDyuV-jZHbUp184>=_4AjVcK4VIf-ng6@jz{o;
zY8)Z-1Se6jL#?!RB2y_b(uV<D6$Gd?%KQ;WrG*2{bfU1;q0N56)uEr6rqG*pGN4#>
zTmTX_qC!A%f@~;ORbO$rZnzs2Q8y$)FJ@EK&49q!g-=6z%#};rOp@EL4MA$3{Qjzk
zj}!+9_HWpl72EZz$YI4bC67<G^6L;}HjM;o#e6|Zwvwz3S-Oh;td;sz%9Ve&PCl<P
zSg89&hXx(rMa<V^2$xI`bN$|oa`4U@d?&4oRmR2Zz~pr^!&tg)sCO?~j_nL;Ao0IO
zVSg2m<t%ltq3a16>pI6RJV?$8AEYFv>jItkgy8g0SG9Jo`s!|5e!FHeT#L(^2C+3o
zAhvu06^a>?DM1R?j0+gQqxZM?H%G$m7B}oM-d|t$%@XH;WtoZdMS^auRKKTnCB;l|
z1@8uqH#b-dxh%3Z=E5|uoTrGhjr+TSv3KMow16|xNE+vvnhWc4A$Uh?o_-uDYvTFw
zi8pUv5xnED-|;-JeEj&#ye8JXP->&^1IOLK<#Hyjg+<$Jy|IEG84*4DAlA21A!XtC
z5Bnnz`vV`}zE$T}<2kOpdw1gB{6GJQpZ>w`^V|RNKV$irDXZtRuYSrG53l%F|DQkN
zUw-=we)u@6P*$GHmnwIM>3Rk#4|~hczqsXhKD*_|w>|q^;M;E>`19{h<d&gSQe64{
zfAWv{?ce^2IA54eV>pIS5(b+-vYUHBK9fsYvDPwnJ*jU*mjQz}R^s<-p{6W$z*3+T
zK~?K^_OzEJ&IY^{g%GA*pYv<7>bBS*U#d1KOfMMC^bw1<ZD50!`>N}>B94E{uit1G
z@J1B3USU*ac`;2(g0@Wg#?@Md(Jp!;Q(9|cwb&+lkPTNhAl;svS_18sGN-U^$tucI
z(UA|>_R<gjsVdp}aJ^)&f8M-g+iwT-RbB4ngKEuYl?-^LKALVcuuc7P#ct&&60s}?
zp1dCIMK7QpZetAAD@@u{h-J328ByLF`y%RUD!$xO4Kbfy5Tc&merAx(+-y`sldvUp
z3o*t>Sc(B_JAx17EwN6+iLOJ^He;xTwAKbCLrJinNtgH7_(WOH(swXVl(b%Be?|WQ
z{r*6+9v^yY%-22$17r$5yueo}&RE;XYoz8Px&6-K`T<-=s*+l1d{4i>XUUmr8l8o-
zOtR@WeGh7;oL6kW;>LSix1-^u+Ej>LHp5zNq6Xg%c<22U_5~vp)+YSC!stS2)to_v
za`L;?Eut5q)v~CnHrSW(S+;|=ArcxjdmS^(HnbSF%mKOJH>50HmCfyXA$w@*$5*st
zYffyRYz0keHLMicm@Rf$g{E7X*OmqF(r;kI?q=oZbclV=n2hg&cE?xNT*)P=P)<<8
zUN4<kvS?(Go{rGT-(Q-+c*oF*(6<>&E)`pKBWg`WhDK9J(Pv22`}&#ErI4TqZ7vP6
zmO8n>tGgTax338OK&_SM=QAHaKJu6%21oE-Qsvb5?REIC^<{J0w!;^DCs$2WQE_!H
zRnz8V2@<iVq=ObCLf75?0e^T!+uu-pz~n}*o;7AnUQ{&L=z;{WrF9|AvlPXZDsHVJ
z(eS(8^J+I@Tjc$_w{(3+t(9>Yp*7Yea&tUl>&(y_cDtUg4+LwO&l9JoXLds;b>POx
zU|1dFe$Ur$e<ieJ=t=9sd73z#FRbe#84DpWFAENMdVHd(8^Tz}Zs^ITaLyGs^yFs1
zNy8phw&GkHDXx5Zb4&UB3rbu$UoNbz@~aOMcf-oVuYO?u@dteEaQl01h6nnwlbh?m
zZ~UwL3m(rW#-W$ofOTQF+jD&Qj352p?{fR$Pk6m^>~}l*5b$G<Gb{hA|LC{)@%x4E
zA5Y*dKmDzr@x@nP(Q1*(<JHpkdl92L3A-xRVf$Xu4HffQOXxawz1)LyjFPg63TI1|
z81`J{98HNFSLOe?qS@jWYVs&nrT3Z$A~#KvG+j0_f&^qTRr5DW6Eph@sVHc+*W8}J
z%dg*1KqwvZax)oeS!%o-JPPpx*9zimf}TMYs&aveUsmF2n*rBJqE}PAK?a0+eegCv
zy@;@o;lf*L6Y<@4Ic~um+Y|-p1TT})_C0MI*LIi~r*4F@4J1|+!rHvHDL1P={Ib={
zfxK;D0tKs}u*sf=ZX&=-Sc-h3@^dQ!;(gEo$jQZ>BU^X9Y--Z%xM?L!2-k}qUKFI)
zkQn1$s*u~pUKDaR(&n|jIBY+ImMT>VZmFf9lMX2^m^5LgXX@oG>-imdev%J&GZ=*K
zC{_912DLey@zj)%k{WxZOI)WJ>M2rFqArW3uE}M9LFgVtaOyh|C4vO|`0+se_?F~W
z!hR&4o^Z{Ot-<%BWVFX;N}gf3!|orjt`|;!D~GWL&nD$VTm$-dZc2FYIt&OMwGDDc
zX&lLZknP>#h0@)4>1(Dzd!dlzZ{j5{pkbV0<J0B;v)N!_vjn2U3RbPh@}|)~8Ai7u
z(invM*2ANH`kAwiYQlA+HM~UW%H=P}p7&llN9;@HhBIPyl0mfMo52|`@3(bOGqtQ@
z8Eptx>L3J3rn4*5oHa>M)%4C1Tp;wl2%V5HrI@1p-nwZ5w$He^#<-T0g?iBQ#F{O)
zKweX<oFiQ~aQE=Q=U@GVn};`qejumF<>}0~fBkEGEj*phQnqFUz1gZ~*Az)QP>A+H
z{1Va_XZ{B5SDT6@i=uEkz<BS)K<zxYuf8PQz9#j1retW@;%cU5pbBK&N+WJd0J2o&
z7)k3&@5M%4bjvhN90xTn*b#1S4s=~dj*;L!>oRi~9itB%#~p*O^j#0l@c8Z{>$2j5
zC#A@;uH=%~-`;R@d(XF*Z`d6Un9`V+6;lh#vS6L%Fb;SZh^3Odk#!Y1E$aIv&alkb
z@9<M%&6R!_h^5i@JtWEQD9&*{pEz2}o86xJ#aDd1%%nW=jOUNveW2fYzWYD?pZV;k
zzr}y}pZ@3gqvQVb*R+_KKAwncWr_>q&@(+f^WpT2qw%{x`4PYM`~QIUZ@=O7u2HPV
z_5<EI#_@qyUE=mfKVvqQwNz|3QW3<0swke8qy;a*QG(!Y$ecK(Jcgnrc;KnVl5&#4
zF=?j9h6-fWc`PfXs%6z$?XYl|uHuXbt2AO$$!e=w{S6tzS{ktwQmvqYf0B?Z2+Zd8
zlR%=EP`khF*Z&`y0pTsBnp0tHD;KW!0T((^_U3ZEWI#m~+SX-_1uudMV-u}ayzj7K
z(1Tb0te|8sZUf<fyuct^y5B|vTrc-eE_j5isaXm{HW80-;7Vl}?GxBWGUfj$RhL8M
zXttPDL@Z^EKw8^N*o=zJ>Wh+FvfkZwn-SDP=m4=#X=Z?RimC+hpISrnPB)=XA!Ru{
z8wS^P81uqcT8&!XTFz)(FQ8@}t|n3mXvU1bcU)D*#z5$H80*w44NZrVb-Cc;iFWyj
zuLa{f{LL3Q8>n%mq#4r|+LR=$N8_TcYCps>CH5cNBv?nvYqI!(8<>*iv(ooC>u~*^
zY6Z5J!`R}TRrJB&yN>zvk(4HU5I-`d&{D+pJ4^`Fx=_~%vtDqwU(mvyX1u;XMvx~2
znNptV`dWw$uu1#ocB$)TUrLt2sUh@?^E$-HKO6EOpPQGE5V_RRS$gBlh5EAFkX@V5
z%-iroQ;WXULDkW&!COX$hHV%S4Tq~1d=UMTkdgXxej3Ch5at!SR6&w+mcH|hV-Ik>
z@cj1N0Ig-f(~Rg2s3II&Ql%870lc7W!8wMmV+;moV2wqF2mP}f%)Z$gYZVSZ#udj)
z_JHq{HZ-y=r7X>f&M8da&_vOS={!Sc>0RJ>bL8RkFM0FjkGZ>l!1_R~h4b)4To+QD
zFeS0(B9w(LJaUL<2^%UBmmE6jR1shN){H5TL^GC(m2<@k@`3X#ts~n&w?9JYd7dL_
zJ~J;1I61?HZltK#&RIuJvTWp-$XRT~&OpnF+$y$7lr61Mo5Ujz{eX9d%jrz8(Dfa=
zn?2^Z!_t_ZKeEgV(`BKx%3&P1y}NlS^6D(c>^R=ta-L?o5ID8QGEH*IYr)f@_n0O<
z3c*^su_Jbl_qFoF(?V@fs^xi#w62lLB1dVm^urG0J@X=hO&0=>=V$)p+i&>8-}`;+
zFk;@l<%jnx`~8t+J@GI9>RWz%cT4{`$+8u^bfN^$XP>{~=ch;PVb9ID<HPsg^6tY&
z)^*~|?UB2|@<VMH9OD25xpU?<Qmw}t(X`}TX-2H*C1qMJ)GF2MtreWNa!(jT&LY|@
zaiuOZd7f2(D$jNwI*c_^Dpn#jZ%aX;lthe)HHjq^gyQwy>%Cei8F@ZSksO9Kl4Ftq
zpc$52SaTsaC`M3NIfJYqN-BD_R?=W=N>{Y^Z~OI|B@lK!&PnvIaY3fz4XQI-g)!D^
zOlJ}P<fTCjtd~LsqYtw8k_;t^dP7QEtXal2ft&~(vS4MlN@MRm)_+n7T(`^CVayiT
zQKfpdHOVD!qQ1XUDEbCUnS;kzQQgY<sYpTFgk?p_m8`v9($})0+OXPi2vtRGRzi6c
zjli_)03#t0`6_JOVCi<fqF=NVAn@dMV>Pl_iS2b0EXmN~oTGkHOe0*m{GO`~i!D`A
zyBY{$69l@hSGR?1ookxNX{O`|#$v`>a@}KY2Aa{B+~tCqo-yT-&d;<K3EhDA;cDSD
zRSr4}<RqJxvFbX>k>H%Da)G|@ab1rI1IBk?EIF6!2IHyJoU#4L-it`Em56DE7+3Q8
zNH->O=oyc9l$vPMds<m><7b%tD-l1ULble}tgKb(qpR}0Xx;={ovyO-#^Qo<?%G3-
z>Sea*YOdBf4XIH9VPhL!wdz|bGyPq{<+ttApj5)PMexGmUq73)tNqbnbwjS#P1RXp
z>smuo3^%s~if(8IfL><a7^$J{WVp;(%Y$^llttY_6J1Tp<>mQn62U43E`kvCA#m95
zq!b8)bE|A%ieT6vwXTEagydYv#bJ%dlqM~Dx3`3ylqoIC%IS1sU95(68O`&zdU0oh
zbBul9*!PV41LLqGgrL#oUOw~t#PRNy9ADFQhLn>aDi|7G91~fU=?3~>uDSx}J!aDa
z8Asg;p7b&=)v)6DI9Jj+GR;ZS`07GTQa|nW>ulU#(d0SXTSAb2g19b<ddko!(d~tr
zrLy@j25!fZhvPu!B7GltxV`1L>osFRJoqU^>e!LinUo_#KQN5FqFhjOBE`tf{VgeF
z;u6U*({fYSLLsgzz3-@b=6*j?YvXX()5^+pPE4oF{n%lBU|AED!YMU|Zlt9wo9wdS
zf@XoxuG*(R`{7%D=gk*<H4c>hp5{H5W#+TjKjQN4d;aw8JI)3E(Rt$58P;{ddc*yz
zd%ifm;nh!nMzEISzGr`X$G30a^4XVPF~9$gjz)}`-Q9uv-Hs&{o}On~=kd+3TxQk+
zHEIq`LxQT3xYTlz>>C$^Hi~s6E(`1F%sNd{{8Sq_54bC>W>gO(XsaOQxfDTV*#>iS
zU06+%u%V2E@6}R?Yoz8xj#-C*N-9NStc?L@u|9zFSA-D*%{$3;Fn=>n{crpAo6!Wm
z1K)uUIKAjhE!2|q@)4fD)~4dY?kcVUFiv34QVKN-#3ia=wan;?Qc9*}QBa4z2iMDk
z2iHT{tEK>iM-%Em&QtmS$_Xf!raaWmia_wwa3TdCkT_Y%jt;)lmOUdO5*k1xx(4NO
z^~J9mSKE?NTac(nOKI33mPDZ>_~b((;u$HhFiIuJrcAL{PT4k<Xez8<DN;z$Ag_xK
z9fDGr97)=vzO_GsZn&+IsH{bWUb>}P;{;CKh7^4P4VGF~N{LtHeYHJ)w--oO<sf1M
z6t%mvjjkK8c2NFH$})sQuMqfF00961Nkl<Z?c|az1&n-6Ayl(mxJEW!*U2DYyHS+i
zSfQY3<hT$l<Q%ar>+t6=-eO#jwT=+>xR#)1ocAotLchBKEHzHRcUTgpKTu2uNcHFD
z+1P?Yu4$6!oxZg{v<8|6XcbM6iry+0a#IpOKbN0+>z&~`RJA5wuV~TA7EmRde$~&e
zaN@>k+GTAo;?<3^*kFE?>s4;QX(<khK8W!4rTX?7A8w$jJ<A#h-r!BeTd@!t1KxV=
z7ugV+B6hp#WU#AScZb2dz}R&X!|Ftf;hbaWd(|&AtXErBvl#E__dABek$yK~ttIEm
zvaIA3am}zBcicYQ;e%XW({y2)ZYg?53)&)0sTNdE(eTF6X)slanbY%$TBX<CT1iXn
z$AOzyuh<=S*8$pE4=%`aQY#K^BlY%*LOA{TWDt~pUn`xB=kYxA_L~p<aGH2N&CD?o
zYm-nd@9|Yd)V^Gw)0`!+t?zqwyB*VXVV!5mtuK=DDlR0U=8pRvKYDe*w}pq>p4;QS
z^bo{_xJ0_HV;DNd`vWOvo^0dtbizB&Fm^H&86={&RJy*W8OP-`^ZfBlD~&EVYVera
z=$xk-L)h;GopUe_u<!T8!Q$h}T0PU6IfaJp44f)&Pc!qhQhg6KB62q6;Z<XH$(gU;
zeaHXow|>I@^Eb5bzT@K$&s@5|@#cXg6_#A_I6M-w?3`i0A9#4U=gsGzVU3~d9K-RB
zoBKB$Zf|%v9Qo$)8`>mJg~Ql$*zYmM^I<AX(@ZXvd5**~gAu!A=j3P3F%sh<j*8mI
zsW49$=F^$^`OG>kLTM>7<U+vua1H8m&QZ+9=_7gxWAqHDSKGX^BBISvP|}<dB^TBx
zAx0@@QmN$H<ZErotJ)OYoS=n*0$S|QL*Ik%Uhb>E!>`|H7zpDY-|qm64TGfeRY@l^
z1}ahxQ*x%{g7scJ?;Culn~k^7f)RCkO>CB`rZ1FL;h#Xx119urN3!!yPwDM2u3)5^
z!Kmm7(5YRa?jnDOaRH~@?Kv)NW0O&#oS{;S2nk(3AbJ{0%~l10O}CFt*rb=0Y1fj8
zM#XR3u~HuBrgn8UfQUtF!v#5fio$hf<BD2Q5!RMm*RE>Ytsti46;tKUORA!_7iy3C
zR)W+@rsPDdo<@f;(yC=>cpu2^B_BfUV~Lzsa#~e;AkB8xdTLs)ecH}}@djfHrc`{_
z%V5#+)wUV7oCzCfwg?506@kf;sAsIh_XEZ`s*xI3<H9wMCNCF)52R(ql!&27(xnkK
zI>uqO4NXLZ5+viPEawk&<E>Bt^E1R5##`v`v0<b&ceRH$xhY}$_ol7T_e=&}p^$Rc
z_n}0pkq6y6DgCKNjId3;@&Y%C4GJfzjiLj2i3#2=YLL%MD_Qazj8J13O|~;!T?`w1
z+l<AyEvCA?SSpPr6{GqZf7KSihW=!IK2(Ty6q;`6JI3HZ-7e_xxTS;&`jv>Lk+S~m
z))QREp&tobO>3(-)daNGt`r0`n)8IdV|Q~;f4nE`c7i6=MoO973a%Ntu4A{~T|;)(
zdD0EJUJJbByHnhxZi@!gl9=X&^JU^ZFOn`8JU&=<yB*{H$nJQ9ySZbVa3zH<h!9g9
zNXEV-ByR7QI7qB+WQMq|eEscP{_HP)#jl?xo=y|<w6Nwv!%3tv5~S95y_}beyhgj-
zj;`y3OD=_FnV9DZY9Z%LN#e3;IZ<<>AA0&uhTXfJXAF+CCY~;5ZubNG-9XoOjQf#N
z3a8VF=jW5a;b^3^YKd-cZ}A}zyyN4?XVxXtIZx<4Ww}7_sj0})&>AThoHN|t9@y^!
zrU24FEx#7dA3yMLzvDb*N^3kXg&HftG+y70q*j=+*c?-t!FpV`BXqr_12)Sae|XEo
z!SkQ|2fxcte*90l>pTAR`yc2&|AO5r^~(OH6L>yGI%|mrZf|cn9uGL@>4N2O`#^qr
zWU_@dow)ttE9$!w!#If8&%-zb4%YKAR?hRp94l)o^nJ(P4-DNvuPLuJXQs=Uby+D<
z5)dzsA6PC6aTRSuvw|3Q4LT!3uC(%Q%c2a-xs)rCW-%1gB>w=3Owan?)|h3`&XtrC
zYtE8oqYGOzR?q--W?11B<u39bH|%iQ?3iQvyRrnoX*Yqv4?EEe7$f|IR;!8?tq0Rc
z7?MKf8|>|!rvjy{7^EuLbiG6d>o#SKq3cGikBwlQUf$bABoLGFolZvD2&ipj)@TZw
zQA)xXnpZyGNgAOIUaW3L<RDv0RB0J%!_#AIz_>tbS;Tb4U-{CK6Q!*hPwa4gPc;&l
z1ae8E)PbND+Tho+@T-1%$;Ah5bkp<Jf{_i^7=tS!qOu4lRNW>jkptG+PP?#W3vW~v
zQHlT%16ffTV?EWX_gg)#Qmrhy2vf5jeA`7KmsjX|z|vHkk>jNNSOwQ%U5^i$<^pAH
zEcr~|k20Ye`C5YHY<sq7JxZ%kvs%o8*g1pr8#F99p$AHg;(Uk;Hcb*#S0as~6~2~M
zz}P@>(sI@sk}=U<;(O=miGH^ifodt#^LtEMX#E4WzoDt`**dE(4(*dr(AXF!sF+}g
z<dm;WQz?~Jl7^B5Fj>_4uG835$&@#$gK4?Q@MdH)+X{-dh|#7YK$~NwHFKrZnrXDs
zUWOEe3aBb~sis1eGAYxV(k%M<P!V2jRRefd%DRXkHbou|M}GR*8x{0cfx?ZW??)8{
zn``e!D}|g^V(bKht`Pi4H}s6V0dEX3i7LD6I{YwDQ>N6Uf~kP<f$?yp-`_GEZs>=-
z<Yy>ko;P|+J36+ZcIbLR^o(X6uqAu8E80}6P);RfY<FhOh3C_m)A>xL()S(r_Z8ob
zbbXKW;_tSm;{Aa2z0{<e7xZyM%wFQcn<~Z)-@iZc$AA7y{_GcD^ZmOMr!|WB%%~~T
z23c~9m9B-9<aP0mnuSMhtr3D)e%Ccqvsh=zMZ(f(7z)lCb~`EQO6$U_ua1P^cz*oI
z!|e?>hmo$+q}H@KSu8bIyc28k&<{9g+3oiF{#rh~d&lW?;=|)dj{QhpBfGJ~#D%BJ
z%H46#JkP``Jr2%Gdg#Y@AMwGGGL#xcJg38WsRi)djXgePp6AM%8jXl;R&z!ir4~--
z#2@|oEkAwujGx-T@Bi$lcx(Cg>5(_LH{8Cy$Gcv_t`J>KEsgQ87s@U;_IA&hGqDxs
z5AXSJ|J#4XAO7JV@L&JyulZ-c|9kug4?Vu~SOcZ&DFnWIo^j66+m2l?gXex2@oG<u
zYa+%d1A8va=ZVW{CN5G%-9TRtr3jqW@TyA6nY<=)idSTyD5AHyz-aPwTt$NVPt28+
zL?2KznULDVp6!%sH=b6d+{y)q?|Qo3k^Xp3x7*_!*y(bmh5xo+ztJ#2vugkOWIK`d
zSzeS{WHa`T5O$&}-h@z1r%^Kq9M=yR7x03k0u9?ls8MU8Xv&uD1`X_q(iD~p-c+pb
z)J7$NIL-=G*J{<IwnlBbo!LOcu~oIURwS^bDwo?f>aJnDyr|9va+E{adW*50K=8F<
z{3Kkm9u_|Ic-QH`Acrd&5m_r>E3F8>6Z&4&z)7t<mZrXTx!4t|Hae6jc<*Y}SX7Bz
zb>6)Q;Ea)}KPO6DAxCWJgnm%y-x;}*YrV?R<*hS@W+Yt6TZ3ttg2i_`wM-SLk#$`}
z{a=gfHegw2Qhb7C!q;|fuRH97Qxs%F{N~<!Fd_o2c@-Yc_t>rz<R{2xUz;K01d*Vz
zOf#J`#LGnlsx4DWx^8BrR;*RKV9i)X6a+p7*YB||QPV1yqw6F-G+k66xnRc!is{%2
zvdSl+J{U_i?n*h;T(8=VYe7icY^E|e)Y_m%8Ble=XtgRWkdYao=;pVI2Elid$)RYF
zah={ldaxFGw&ZhXl@{>e<T={nkc+<NRby?9W=fb%=c1qUrb5%!3XR78xaZBA*Gh{O
zY8B=GZok9zy<8}&F(^5S!aeqCQ56?QH|`k51I9Q~DWqkFD)z%#DwrnO`o4$I5&99|
zj|{s5!!TSCl2T<DxFTtqESS?Oh>v&FrrMy54%5JqYDI`;T9u1$+;MwbvCgx^g&3ot
zhs_eBsA)?!ct!V0kW|iw4BIVl8cRV24NaQin&G@Q{_5-Z{11QnE53jC$TTlZ>nfMH
z3M_o^;^=UW5CSfQ>oAwIT-MIXXC^n<;%SY}iq<KmC|VgOF!0d#<XFL8_}PyixIGLw
z@7aC+Ih~bdrTReEbr=Ju(=$&`k2vQUc6(aQBEVLpqf$A2_{h_Tk9>Upo-TN(E4Q}?
zF6WOJQ@Oj@(+5k=7H4~gAu#rVWx8M*bfG7wg&|lTMoVXVa*KRFMN%z{{lMWc(1pnN
zk29%MN}Ne#=yq5q&XrTl{9pg_7yQSUh5zgi{(ztT>}P!Y^{@Ew_(<wIg6g|UN>Cbg
zNgRLlN+`~hD9cLA1@8h~Kk%>r^v`LBTb|43{M)};_=kV+I~;u^nvvH({yk!U%Xc6C
zFPJrBjHh#!K8W?&v`R{e5=Ce$t(A?-WfpoN7E(^Q5O58=4O+V#q$6V+YEz=-a@Cf}
z$3md=ruX6&5|*>(=Cq~>9Zk=GEyF_)25;0YRGZkzy(iop3AYdIZ*SQh4)k3|Xqm-q
zgU8?ZFkm-<1XSn5f}@vFEmbcHD5>DPKo~~7=Y{LaB?AuEsn6L3g7wO`2?SOfDh%x8
zP^%Ri0?zgrCp@LJp>3fi8@FpP>Yc740?>ctgWojxhbjWGS~AYsYdz>@58Uh+qJ@x)
z$2jR^@W#quP(*itbyTdpfwsLwFS`H^jiHn<h(yMPUgCT~TGW~f1FgdUwW$57Dwkx{
z;-bzB^l~VcT-_Db*izNDRIGJ+!gnKnILJX+5+)~HahT%32h~Yv*SEFUR!Em;rt>2;
zt=m~9(4G!A&IJmFFpSi?V#|uHE43Ekc#Y64m&-)Z&By4)WxWIs;Q~x2S`{&j^8tKM
zF<ve=<ES`tE~4trnUrTtO{6u!mQ|trYfd6Sv<8}$)JLmzwUSmjV4Z=K$qBmMEu~l}
zE6x<k`b4E+`UjeES8J-?_Cg~V$Tf)=u_6BRBJg@mk<uEz>jgz>GO5qAsIa@xN!SXI
zOQu%ISTq<~isX{8zN6Gi&LS$SdPsP6jMUm_RcMt~Rt=dF)v|O%G$>It1_=F>bJj#N
zt;_|bnUscep5l8!m~6m%&o~a`sQ&2es0N0i!*`t^3sr`q+Ih0~<Ze4Y490ha-3?(l
zf^~$PS$&|a;`WeaOh?y`Sl3aI91Cj<B}Z~rL2{O8)l|iszh!o86hSLVAzZY`XtR5Y
z6Qq@hw3@0|xElH{U~Z0#X<$wAd3DZGQzW%SEJ@I@CYcTWZYP0QO7~Q))=m|wF{I}C
za7z67uio*i?;knMt8T%S&O0(4DG3T<T~MK^AtyO&yo!_ErUG~Jxm)I0KJ!H^ZDm<V
zX(i~wBNvM3Iezl_EpP5d_PaqMuS;gWoG>+Gsa(z%PS4Mh=xQD3R9L1(7O7U)jRPe`
zK7M%1`Qsz=d15`ENzO3b9xx<?m}<0^m@W~EA>~LPMktx(bSA|_DHU(PwalB_9k*-d
z`8;#qd!E+BZD%=*mYZ%zE=F_&Ya~w@w`3I!cQh(LoL2td>1(oW{PW-aJ$~z}ulU&)
zU$FF&$XH1d5;U)rRjr?wNuG^q!c@zCzvu4mj=THU{MY~0|DAvI&;Bug_`m#5=s)|M
z;1)tRu)lxBU;djv#rmG38&t0&bnI4q)QS`~)tpsG3`GaHN*1wo$@nlx<gwKNIlX%~
z3v5d1`WbcJ3EC$YClzhZ%$1buH9u$T{56W^R@EnoFukMO?dd{~^Pchcmb=eCW4w9b
zaM;s1i?0=vRxTx7%LD&jhXLCS%B^o+eNi;x9Ic5OS26;M@cp*C3Z9IYZCy71LRuz5
z7=>rHP)efas&I>VwrQ2TE`VMd#^E|gODiQsT8^Tfs!jQ23pI<FqN0#*(NsQdKveIv
z6cc4BMzU7A3XJ$P#VWN(&`w3*ESm7c2(;g<BXk2*6Uu5yI{Cc}A&o+7hTICJM#@^S
zA!w$6^pCfiU+g``$^l|!)2n4GErN!_p_vDWxUg*<?XAk8<9mz)r*+URN1?vUg2{>A
zjJQ_G1xglF#ab&D?d6&2=`HJeVS}%nv!o<8E{rJ6`~3}}6Ct5D1!ALfFsFrOUDzFt
zpcmMt4tK$+-`(N6AP-S(;Gi@^uwI3Ya+wDw7jcQgmF9KElojI~DR0oHA-5z^*Cny8
z3)MQTtF)NajnR=y#dSe)1WIB(Kho_E^1ihOrc#$jY>Kr0UWU7-Sq_59;Cv?tbd6&9
zs)==(<UK1@UUw^;d`wZZ_BVCEQ9-Cot<H+bjanclyzeo-6LiRRv@l#FuccO5ef-5i
zTU9t{)VmxpHY@Zm2bggIsw5|(u)BmZiBH}*&GKj<rC5p9%FA$5s-RO&MTS;wv<6jM
zYtc~fqOjI*C5$7CJKQje^)Y))5mA}7(!XK7{M%TMshJoPCC%j3Qs%Qj-&qdNE7Y$o
zbfz{;lZ(#v9mYDc2G<nLG(qV&(=uS``hi>_<x0aL<z8|Sx3sdR6{wWFf+BC$TjF|Q
zHy*Vs!IIX*^JV4T)cEds<>NW=becKOkytWCn<&K{(^%()HAN8~I!}oi?>)mXP-(=t
zicOk^HJ+5x^)(WRy|K<SB`!27or<2VG2Gnj`0*Ea^cFhblUoxV#WFDlPuKU%^Tc$y
zaC>)ybyk`TvsgM~T-gnt%X#8_`iP;>a$>m%Qr56sPS12fh8Alqr}IU0K(#Uq1Kv6=
zrwhwj78>j1y4ejK));D<`O(deRu<j{$Dwol=+%+u)Nnq~d-(RTG0iKvM9Q=hQipYS
z<hbyzHvXGm{+f3me&A;h_k4MK$5+4oTew!JMn3yp*O8Y*niqVo#LI=Y%vc<@W`_Ga
z{^0k2pTGXgU-I?;@F)C<Z~VCXpL6rOKcheH`S`;-zW>E9F~-w-Im1gzvRomKw!B7a
z%GY8u!EV(WUS%op8cO7?)xbKV%jt_{ww6LlLRnfT%c)TzvgpYSSb6Sj=yCl(tFk~_
zJ7K0thFl{lj>nrDcJ~k1q2sVWa(DNN+gGnR-W(*%t5(XqP}fRXKXK3feGUVy%Jxyx
zO3o|Z1)1oIh_Gry)!n)==Oh=WPE{C3HI}?g#LF|$JkgJP@&1-=8+KTvqEaq!M@~tg
zd{ecm*ip+$N(%=0fQ2wX6Z@gj+<+@TLsLJg%v;_%3ANF39dQ|WYLj$4i-las3ipLx
z#3v<3Iiyu^SaQ6=Kb5K%cfqy_TgZ<0bl&5d<}$1oTJ5OKV5pjMhP2ZKQPLYL4`D44
z8}$M(PEd_B(ef$>HkpQq1}zO!6OdKu-XJfEGz+}%d}OeJbDT-0!v{yn3vqg8K7FL+
zjQ7Kp&&ELTmA1ZY!p49Lmbxs&r*Ei_-_r5|PQr}lX=3O{6}u=ckym^#B2Z@)0ccG#
z7@W+&#<5xBYEJUKmniO(Cj6<SJj%fze22A$H7_DWvYwnWp&43}{_+~c%1TyxrBzH_
zS@VJ)21NO){`q#2fj1l_-@<?dXfz6<Cowh2kXGgUb-u?0y);`T=arT+UFfiWyuN3$
zbjZKGUP{dx2tCE>=8pj%@V?ies0Nz1-zcP*1qR(H5_P6{y|l4T1x0F6L$x$b92>@n
zy1&I0D4JDJWusDS*>bsDa|KLr^kJYW^(A=;TVIACQO<i2H|oK*S@S5!&y65%&LCT6
zYhoM5U`m8IQ{s#@5SyGIrHDvT^erMfTi2+cy+-f`h92MV<=~E+HB$*m9a@ahpqE<3
z+E&qlQ#BW*Tm_&duTrb-J9^U#&ubTgfvI?Y^&#=AcNadK6VqBrxsbCcyK^e6G18EX
z2kV3j?>1p*Mn0c4b3UJ_)zJ0fYKu&9rt3W3IH^=$6Rjk?QW9xROfeCx!3G(mAD_<j
zM^B|ti#+=eH@A2LA3waK*2?|Ey+#W+y1wWB-toitKd>%|sWlceu}n{xnmC^)Qp^ek
z!tL?ys(Z3dtl!J!nQ<6tx$^G2@96u1S_(sNWiT{AsjN$490o9%*Rf*i!tG7Zt_xUa
z2*I=Kd&cm@hfCt!WhSi?IW0UteqdiR?Qp~8vGCpd#2@|jcYMC@`QMK{U;XrF6k|ny
z5)*lui5Ew~l1ips7MwTK-qV)EKmT9<7sS_Z_<#H#|119JkAH<Nf&XpS^Xkii?|<<P
zwQ2aAceGWIW~V8fO|X7)%A_nqg2V#L^C7YPmM&=M&lcLHcUad2O7O(auuOuSwkp~m
z>jPaUiKnerES6v#1hC^CHy!~?S<ZB^keNjfQya!QZVpF|w>OM8H;jiPhud3jjz@-}
z$Cyf<W?Gsl%S4%HHlh6A_Uktq25L%jAl68U6KjPKJBDE=>c}c#4FZdK>M}vwiXE&7
zyNto)h^x{tmCgdgyU=60fl^l5lCJ)E+q505WD2SvHqeYA_>Pn$+7XjQTaab&=*B%(
z1!b*?|Jd1H)Va#rG%9JGuL3Ddqve9LnVJ^TdXZ_b6<p}#XE6@r1KvuDSxTauH%<~1
zF_K<Jl^g+QDyC(eMPa2vZL4fPrQo}P(2XJnF<L1re1II9fNA0xZbn3LIWE#x<_vv5
z$eYu&eUTTA$e@mgCOV9kXLdD`@FXLxis}5sGCfnGeBgc8(VSOtpJpPMAn#R*OwUi`
zHPPpVa{7UKek-Cv@9D>b(lvr8xR40WO8`}lGzZBThW&xd>7D2Te4u`kjbAlOBFz)t
zW@1W$5>%<Fb=FGw3UWE*6e*=L9DAyYv#ix^I48-v*FvjY3Z3&}2h!Kc`<^XDGM8r=
zNV+4{`t8ocHt8WS4Yn{K@APwCr6Dlo3bja0DOJ+=O2t{pH}}?vmZeHtTixE`9LYH`
zr%bGkes{z1_6@G<X{JCfl=Z^$^q%E%Qc+*01yx~FlweUuLP@k7h2AOw8#+v_q%`Y=
zuC~Q3m|qdJoD=IZ%Vnvk$`;$*Hq=T*p`WH#S{>+W6ED0mQ37hJUf$ZMdDR1?1<)wx
zDmAuMJ2Hw|?u1f&5x-djn*N~AbHi&$Q*%Z(E7f=y<yK3RP0Kq+aRQBRsHQa%fh-Dc
z2{$Xo(mPAnd2BPRYvJp2=C40oc=x;zHBq)zp|7k~Z+qwH`VMH<Mnz++W)##beUY=E
zS3Y#q=2+L2X_~}p=&d#@)~mjxDm<N1l8o63!5M;Wbis4~a8J&KVYj322d2|A>on0S
z?8iY`?y3kdQ%ZdI-S?cH&vXW!&lfJ|GrQg~PiN+7Vb^u+4<pW7e2^@a;4HZ`auNO2
z<I{zw(?ai|UlwR^KF_?lzoXJPozL`L&z6kYd&_6{Bg1I8zw7A68~*IukL;J3Kltfu
z{`eay9ed0jYt5M2xI8^lN@ag@11|7kiF`Oe@xSvs(S$4)(xz9dh52-r=x8qsi_`Ph
z<V0@`|Kh*+7reQ>=a+x+7yOfd^pDwp^_tWCmYOTay{E35p3N(*tErxdZMRe{V*_#5
zcz2EImfVRgFiX2(YQ^|K2pu^{Bz`SLYv?6aR$Nzmaq2js?P{wy<LC}Igxw9PK(3KO
zW6e=1CRuhpu=9b#FwzB2pkZ^OOoEoj^NBo7#A#uEdj1>F>;KfR-)I<EPLG^>LrPJS
znk?k$L|abyFp#UjbDO8Rq>OViC5hGIq%dQnrl^NZrbVf=?Ns@^maJkmf$&9K<m7<2
z4)h>4NW+tHo?L}eXjK#HLLh`AI8WBeKdq5&7-Tzjj=W5?bp~FlVoQ>Ie2D{An17v#
z>!g=QfGo&`aZUnrY@qs4c=1+A%Y<uZ3}PY4>xwlE>w<<BRZJDWc}WqwE}|sXOTzmg
z*$TF!#7K*mE5|5R!Rj&Bs7vQ%Vy)T?CS!njIcx)Lpm~F7g`KxlYgl4p$&uyqL|nwi
z<b0syjB^HQ#}i>(#o&A|v7DvQg@M4zS|tm=^A-~vU2vMYooL1p*BL@D#WwSW(5t=D
zcZA&$*B>YioQBniP)BkI@_ND8rt!}i-}Mp$-E_k%;=$bxyDn$&LDCMh)QOr}C~>Bm
zmzeClrSIiLGmUv&7=}^oai&5(VSK>Yy>RfYVsoSAEYZ8h5R+`KatO!<Ts4oOly)tP
zq5;0qsv)YY#Ha>h%jGauGwhcO^Z7)&JYv&I*d0XCUJ7NIuxY}Th1M*kMtt96t!y*3
zT2fYhO)HeVDiYE}pxXp(an2!vT8UXMC0{R|rmwTLayVIMBsCJb=&BYrN%XD+7fHC0
z+B_T7G(!{g%o$7P4JM3oxK@d^t~p~!fYiNu=Tu~CF||>(0LWEMXfqP^nsXKbqD50W
zx1rbRG9W0TQ~Pa78c3vvx6f7e1v*@}M#+X%kq&Dd-a6Lecw8OdK4yM6%`8#)ToILO
z>~@paBly60a}cdVm3Z-WmExL|63e`(Xmkr@Lq8|7b++0xN_Nw#Ai}jZlHy8PXN)!<
z`e6C|vj@I<bI)${bl%YoJ*6~G&yP|;oD*+eJs|GQj792+<22EQz-~8C;>vtkF}CA*
zjy#<ry%Wty&5?1p7e|OSESHt%=d;pxhFBo2i9C2NbENZ*W7km&T;_#&&Ex_xWohg)
zhJD`?f@MGSGz-UF$8Z1aC!8K1>H5HuGtCq(i;;MCV_D<O<+-sx>=_4-?K@t-evLEY
z<O$xf=9QQt6ia_RP}UXeWC)A%OfZ(=cE{aN_}~1O|2a=jf5^@4k#4Yj{}+|zmw(DQ
z4lFgX#HKwq4ud`ym@C>vk-^z|i|?e$-un#|kn=Vr8Tg#DI67HyyAIz=ro;7qDTS7j
z3RxW{c$za3Q0F|ZlP-apSK>Ug%nP{`>CNZ|x~|8yMxfzZ!RAOlB_2;t&?IrN=yNu$
ziFcQ$Yj@Az>oAa(nROkR=7rtOE5djrpFgmcL=DjVh&PT_q9}inRI-!{tqB~L<Ek%=
zBz<`=54#Tn9ak;-$hsAGT_=@}C<V|^p_wL*AtgtfEy-Se+DSHa>o841Rf?)Jt-dh^
z=zB>vYgJtUC1P8_+Kzs_xrVx!R>^Ue55-nqVvJn0#$iJg%UCUvH(oL|Ct8t1uL@C*
z>5UEr!VR~$P|{3ms|+bva$LbJg#9Z~r&dqPvPD_v2%S_@*0Rdv8s)W?9s|VXfDy|I
zMcWpgY+xnRH(-_RYg70OPn*a&;*#_ppaDkG80YAQBNj`?ii;6zGTA$dwG8`%@Zz;n
z)&&zhrCCxbgkfZvFXWW)U6c>L3HUHb#I90LrWI<O1UjyXm{ee?!bC~o)y@RnSWAJD
z5?zeAP2h@#xELTIalxVDd`HTWV2pUTy;w%qWyZT+Tx9tPZzSGC5QU;y<C)e*Z1a@3
z2xO<1zO_yfA0#)!Sb_Vk7ZGX`0h~dCcV40kG;hIrcKygcbTSCdZz<^nVUM*HlQO16
zg159RTX&6_X0o6-E=cV!jg+J26Nq>&E8-)zS_uksz7vXIxlnRaH-n?^M8wyex+T<}
znX|bHtV$7!;8to>O2&H`7)mXeVzqdy$J#-k-AXHq>T0sSPcqC~jZUo_#?iz%)3$U$
ztFgZg;Z%)w1n9*ogS_Z{#6J(MN%@Ps$W0FJuBRG-&E;$;q~vhUkh3IjKF@_;et^Gx
zf8jFeMV^u%6>%cRB>JH+Fbo58t?Um6x?xajG#m~Gtar3pnYs%pCUQ!|Xb8bEFY60>
z*>!@bYIJ+enY@Z3KCTO-EQznyG4ugT#nwdcJ?HZYi#U5WXVU)WhIziQw!;10fi%xJ
z)7b5Lre!8~hR*lccEjVx6YoBrxh#?U!$7Ht+i~Q09C6NYem?W`@g#$pHw?RhW*w}F
zC0CwKk(*#Syc&7`e4-Sg?#}bfnj}PQ*9Dw4+}_>L?+5yudv0F8A}$HH?|8k;+>Vak
z1k6p($GLII8RIJRvXE!Xu-oH-A$V-Rqq&JRT_`CD1=z%|4}i0HBWa>$jVwRB<KO)H
zYxY0;Ij`<t^Ue7^zyA6=p1=7Ez8WokKM<=Wl_-ug723)<TroA!s-(ACYgGs<t&>jm
zFEn`ua!%x&<@;2vO&bQr<DUI2_GJvDC}Q?nD{G3_RxwV4j*OwiNL&-s`NI7CjA@Pg
zn_CXUj&a-(hMxUy6gs11OccG{nj=|VakW;~T6kP1zMszjeue=TI`)TGjC-&_a^(Br
zb6QRU1-gI_J(kLHdB(IX!m3hLt=kKPW2^)eIj;xiiuG20U&$C}B)G>~Y!zNp#NA19
zuCL8#BE&K_(25i2uvNUX)S5`k374bVvIHJX%LP+b>M{{ak&E2OXj0-LU&Hr++B=N2
z3X1(tE>z<zw8TpSVFul(Eh?C7a={jB>3dO&QyaBi)WhvHXh;xS>02;j-E2tk%JhhL
za?yxL$zZA+aJ3|{FSd5&KdUJ+HFgq2l&jDcs+!QPNMmJ7yf)_1WU3XxilGagOvWmJ
zlJ~yi!$9aAcAiM*_vHDB=7MnK&Po)J0pg0S!Z~K!sEcfmIY#<YAx43htyd%<ivK=G
zNRs(ow}i=3a6$M{w~64asJyKXU!^40>6y_xn)O&JC9M>UaCdQmQm9_?)nh4|(I5xA
zLG&N4+fmBI`usiJkcoZ-Jhd#?tYS{5@zxk}QTrj97hvRqG*-$+Vw%ad;*2!?IWI*_
zRxi0!1l{q$iKrBbXSOwn1A{7E4#Pm#Nq>D-3ZT>`n@EFH<l+_$fFN2nh`=|O3{`R^
zVl6al83v6yHdfUA-s6JB^-_5)?vP9|7EB?8fHgrxw{?}!9|i8M^@Oe?gibE$Qt858
z22vA6!$8$o>q&TMKfKTs*1R~Bk;B~?t3hU4p^^st80j}LW$TwuLfUM-4tfyQqR)hp
z2yAM@_8p;<tcV;1C9uxYTTjW2?@pCpz0Lga_>t$+h4blIsJkjCNJ%Ssokief17%%#
z`u=OC{f_bOHFl6>zDkySj@DE_C_MZcBXL;)NzoKUq!6P}T0<}5yIGwJISI@^giWIb
z=gY)p8ZoBw`DeFmT`ewHhM{AfF63op90JqRGqwuKl2c?i29Ae;G)LYqky0|D511UK
z&iOFV_X9a+o*$pD(6;lAlq9Zse;BByFr6a97#KRo^ISQfFARN0S1QXa=S@EdT{i9q
ztZ59p9oAdU&lAVn8%zlJ_mBL8hmrH@IIo$9W5?U)g%6Xsm*O&$^TOEe2+fi%6HRTN
zImx0|8!XezZnq;&6F~>v+!~iP@_+h&{Y$$4{!iFx+u(W4{OGph;WL=?6KjOac_s9)
zA7$}uG_r)&Fbraeb<Qen+Q2AM1|l{V5fK*2#OQ-I(YBF1?Aaez>2t7+xHzS$<+IU@
z0cR8~GAzp?{RNj9-x~K1cRbu2*dLDKx^N!X$!FO#MC+3z0L<<7c;}&J%6frC^+bQi
zU%y%CkQ19#*9B^cl;uQ!d<9)knN9++$AvC<oKXP^T5_=!tAqi<PApS7;V8=eoP?86
zB|Sz8ROXb~s(uY<=K7YW4qI<}shYrMM5rgrzxcPUmIGNY0vy@^7_ue+7Qobs?*@tg
zv|f|HA`Mt0Vo~!R+a!_A+N=tDyCrQaTA(ni5g}16MK^3sUaOU5Uao!cpp&QZlFx69
zTAB*W`7D>Y(j{ADs2u#V6>OEyqTp93G-YgUV7IhFA-_wNkR|5~y|b_;4KLCm&Pc#Y
z%+f;EqESfJ<EvxHiKP};Gsbk1z0kHgU3p-(!?`B8fKzRR-HzN0T|W|Y!qqIj?M56P
zH7;5T6R-ACAA-Q{O(HriM)M#<oi7d$8Bha;7!z)q)P7W`S^iwopS!h&Q#8RELum-?
zU!B6|kk$!j2aH|F>v!1l3U~YoYaKCmhO%DYV`D5f1aUaTO&p|F%7#{#3OZGq<TZrQ
zsqHV3Lx=4~`a!npRwAW{Skz(!DsDyF0|&93<^(uNT699GdFRPhkPxZ}2E{5kD0FRb
zLEhKWk~k1c0g8w-yMfj!&Ut*_X;ip;-?=8*nyEF&XJD)JZBk%W6seNpq74SG6cfqD
zu~jTr)hwj6VzD?YZVPe53Vz<I7C$vU5lHH}W4V?C)oSE@)ZuH(*WdaiB#CqDOA!=C
zTqE{co83V8WeS%9zkXWz^It!4ex8`mXG~l<n#Os#kQ;Q~;nU3Nhp$=Jl`xKsy94vn
zd!D{;-2ddaxclO#<eX@RYp4*QQPWK|TpG_jY$Z^Ythos}XQtD$paxZX5jtmCmlevw
z{?K!oW=_v%9_}Ssa6BGKi^R4s%gXcP6N7gw(<HId&N1$G91aIct1OobZs_Rx0Nye!
zk;`Q!cOAd_V3=d(W*;;sLtF;b$b4DoyAEp{F&36(<@WZ7v5k+97fQ3N^TO>o5{8cR
z`9fPNHxKtL^TJ`@lh#Np5WJ+VzWd<^TuHn>bnHS$Zx?*$c{2na*TP@D{lFA8b=6rJ
zl#?K0HAz+i4JwsjrKz!%N;Zk$q>?$sm0$ep|B*%M#8wIquU_%n|NKAV`1%V<uC((B
z!_51~M`FF>vsbq`D|D1=f)>i>v0bU1l#+%kN!gsWLXi>#Y4%PMMsbF&^OB%x#p1dy
zGjfDT*FiHPVBhjg`rydkQM*8I9e2k)x3@R+`w^q=uAF6|OD(gc#FVm#&%;RF?XDD&
z^(S2e|2+%?sT7hbfSdYq!Qw-pq?tI+vSoRJ8=dRODYDFG4eSuO(vO~IDrvbunz1HJ
zLsNSZbk&kHNl(0*tL&;e4Lg~ptdlJ9X7NEpBrfz+EE3l%UfI@~Sd$8L-AF&~WHVY8
z!Z1*pB~E9V*s!E1scKRt;xM%mmlIC6;8F@EM+rYsi>0xhhAg3*xW-0$OVI<*8YKBg
zC(P~8YbsEyB}@JQ8iU%3q92(Q<&gIMC{fJ{<2F>ZV$d{4i?vlNG!rzduv8)E0!Egy
zoww|RAvQzvf!0-ulYzoq5AiMRM#O}d=SRXY(Di%#%@+`#sh9UU012P&tQ^ei`H47B
zWDL7@kLyPK;f98zmPiO*l<`eB$C|HM^M2e@8+3Y!wJH|Am=nRvwpMc1d&FGP3pvAz
zTEzrG5U$y1_(x`|g$`X$PDv_1jeM?hTxqV-g@INr>GCZ&ha0{G7cfX)y-~54wFAa<
zlp3WmQ7xt32mRbea*5Xgrj=PP%gxU|jD-Fu)WCA4q=lLmX-KP*CRj_-vJ~Bfjm7qz
zAZ4YJ!zjTeC1P3;i<9?q_*J2ZjJ4z<+4t5fC6*(lCVZUbGpGJ$%(m?qa*_A4wjyqv
zmf3c2YoHoUG^|2D8Mle3)Jj{$#X>{qi&m*XZCN5z^;t3Mn%Fc9PQCH)@_eXySpID`
zdO>5tt<nAj*xDc^fXrNZ`m|QO7jC>|d0#Fw{MpwtzkGLQo-TxzcysJnro{7krr1he
zGSjr;(u_+pOlQ{lk+$qrlT!HApZptIZG?WLT1N|kl9Ft1)(P@~K*z!_XSvCc@Bef<
z(f5)~&{86=5$~k<C@w2E`<^ZYE=we)g0qc>yW8s-GS3sWW`<$J<Sa4a#xU%6%*#Sr
zRzerpjRW)LLaAAVl|{sHPp1pdr<u;(vqsTS1zQOSHMuM+aZM~aGsh^)6|&?I0x3rN
z&{LFi_kGXU_jFyL?>)T+O37!_u`UbFG+sX(S(b(KDe>xf&(n0_65w&U@Y$iKrH+0K
zyqy!%<%#g-HN{#Lt4jdSWmy>dOo}VFH%E2R6spO5c>KWS@=R@w;qH#)47>XW4mUe4
z>%tg3UwrWyuIs5koOnF1IDg^p(9v0mns*@e+6JZj8Xg2KMu*s<Ju^y4nO311ER7I6
zb(AHu>pXqz1*x(Q@3gj_P4tTN%*fn9=RM<YM;JzM;+QBgv8Xd4(>TYKITeC+Sk*a}
zQiw6KE;A{v`dRyX9tNJ@y=UGHSEYH*h5Ye7UFR_x0V~LE#syKiFG;xHv`+Ni(*;R)
z!Wim0Q>*mVcjHl1p`oYf<S!n~ib*eNZt@ISx=z}fsu3}tbIRdZK?7Q8lJ4dV)&?13
zv<rO;Z)mM@dU_)44qL>7K>KOMgkCmBWU_ZOD2Z4iP_UW;n3KR_PPw9uOWkw`W>XI7
zR#mbV()dnAl~p(3b`2k*ZT~W0JjQ#h5wB~_Q3OlK*R488c&q=rq2W(LB^cY`!hobc
z3sOUCbk=a!1s=DEQDb!j>&RMbEFIpGM3>@<^A+DZY`??w14H`>*)-~MmNYdhlXF8<
z|0S)IW{C5|=w&N*z2uQM4a#XnkO0}-dik^6fGrhl6|tyuV6?m4fN>U&)h$XzY+gz`
zvK17l)u#2;5+&NI#IYJ<u$>G?wmM3RGWBaPQqBqM<#S?MBYyaE+8C+b;YHnUwQ{>3
zXtpDi#AQB{l0wAZ3oSv#xDMw#Nlq(~wmk!0E)3JoGQDT94GNuOR3}kXV<5HJAp9a;
zW0gy$5G81<WE@q5V69N{%yRk2GCkwNK-cfEUQ#VFo^CuqE%F{!HRf4^Y08aSi&wj^
z+^9uqDcRoHs#mSFhi+bT=cKVz^c<0t<5lHvx52k2Qdaf9b{`16UquMJIf-<$w^pi=
zoog~2^U@-TKDSaCv3-su9ITXj6-b&^$!WpaUg*LMzj{CM7vFwhNh>=q6y58+C-fIi
zW~CnnmTdX<hZAL4INsfm;>e2Oaml1q!3O^HufO4kU;h>NuRe!vfWDKe+nnk8Q7+wD
zgf_62T18YCG#9{nOG$}2B}$B9fkn{CRvWKg-4nV_29#2{IUJd%ndx#N#g*MSD*ctQ
zt<d#>-7u2kN?i26o5s3K%+vWQ)ZGo9Qb*=R&e!L8=7XYihu(0%?}+nEiUqJNDbcJU
zYcgr75PCslX|QIoeh)*yd+@<XOKtGlZxT44FWlbW(scvp(+75AVAoI3^_;(+@d*w?
z2YJFdxH|-<6nXddJAUU!Khx4EN6YeA>$^^JT6FnweGjElUE#ZTKM+$QbUov^=P(3*
z=V$kPb#xrYj%l4)J}#J67=z(MD_oYu*dd)0#$X(jqLtfCKmXR@-9}pp%9N8tP?wS^
zvAqP7H5pc$_XLdE#!)MxqYpuIMHD2f=&%FLNO#2=#Z^#ZWLg%MHC~fzE0$Cv%lU}}
z43~}^XD*kIOiz!*W&XR)0>5DbVXee<#`dFx@Ti)&REuj6`W=24m@m(y=?Uv2msA;s
zJr0Mlg=#HpTrf3j9O)e)jIwb!BiYN%5JE4o$Ra*elUC((46P=rx3aC-K#G}liCDAX
z3^?xzVW7EAo0}@3--~;oR?g2)gdmsFWxfz{W<1<uoh3L&UQW2s(}kVv_~ryo1b?+Q
zy=<La=w4u7oOo0VNV=ktxX6+utK0UY3{WLQNs^&p+Dkg35qp@lDngtm(lSfXL^l$4
zM>)K6)bmf3uQ@AKB2k!ezA#N^f|pGs^j-v%8?R~{SdYUHN~AZH<h*Rgt!V*~lTEbB
z7A@8-+1_o8bnA@kca-i9*F92G#E@`JOE+r7^*#OJ7Mm*}bYd?o1wV8!>?rF=)Z_9~
z$b+ZJe@)v6C5cU9XNv*f@1)zJ3O_EQNI{5lj<S&>>3Xeg7eT&mP;0Dop$9y<G`e8a
zYcBs-=jgkEiY1IAnTE-M_52O(@CDHx2~GsDu10)2OWs5&U_9Pcx^a-ukesM#Qc<VZ
zt^<!#u~SKjxUP8LX|Yw2&r@wmhZU^dfpLW4AkLdKWAw9KYomFIFrCkjJU@NJTEp$}
z2IB<|zno6wILX&F9&2PE3!TMwo&3xws#9bL5d_OhUWngHl4OIeaS?q-YZz_wbDh{1
z*#v+_1__ByRlZ;ORIChXrD=eWaRT=mp++QKT(bZ+GhverU#woTMai_6kq(V*@RQ-<
z+CHdMT}~2IR9mBznN~gLDf8<ePJDcSV(^t+zaz$l{XTHx7k=`iJK|mA)!o2&xM5j;
z!uz)$a6a(uckie*W4e)l`-|^c3Z(pszxw#h(+A7`?v-RwxE={eFNChAsq-hNh^b9|
z>WDAC72+~ca=s2<B}Qm4ct@v8P~Z35-`|mnT<j@H?!}sA5WGF?Fh#cHr)4GPl@7!6
z;}fPe*5!N^O5g4Rhwhf=bKvdMh07dq*7CH-GDLN(YvTDdbD1;VTYA^HzZvN|C%UAD
zM$ZRXM52VMt+jH~2ZmmR$T?MxUp<f)$?ETik;+WXg%AvOzhjO`h8AB~=b7EjftnXO
z53hU6`?r71n*J%aY0{*koZ~pK%oB&h9^<802<LDZVom(~C$AZYjvs&dz*k?q;r=Kn
z_PSbvlk9_58aD^ad2O86mB-0&@Qq_<5WxK<Tf<s$ugPMd;JC)xOBPFk)F@Wyma5I9
zGC&(`)09@kK6H3*@lISZ5~+@gjuY#=vZl;h3f0K75lbcsY^@ZEI8n~ek9>Uh!=@m_
zwuUwNH@e_|qhVk-bnI?l<BktZAK#Hr?}hv5cDVjP8260(8@l0!kMDlKbe{cjPdALz
zD4UNn5T`S)7IuSV0i;@Ze0)zSiThV?FgB2Bv?LEt%n9c^IpmyUSuJZRbjITJ(7-sk
zxY9ye7Fu5jJ842(<0QHJH4{stnLtUA-G0Zz{VU=cVV#JrV7i?gQl^sPq<IJG8MaRO
zJ;17mTgR?eMBG?v!}%r&cR6ahV56o$0ZK7gBw^8@ZUm(Ltg7YBdyjP;&P(7$$&zPp
zZNLQSMieNv$p_yDDUdP}?Yhhtnz4*Q^3^4TNUAhJEsnm&RzufUMwc<UK`F#6@wL~w
zW#=eK2LoeRrVGw@xNgMtx0w5%5}cv5!aOgqexQ{`EpWKG1&nz8oyCm<E*xcAZeq`q
zpP^F<Nz);%h&oG^B+Hy<)se_FE(1Z8t<=epYmAXC-%0>VPFaFb8iWw8aC51JoTEVJ
zanUj+p;}@o*fCsn8-5ruRX%fTSv0SqBb7`oUSVk2D8un)BZmN_Xvg?#6m1wJB&Vzb
z)3z#Gylq87Zb}x*rE>k7-K_Ew#M4?6XMxEYoh?zZr6xfkY)3SX&IMY{4C4`Z^NMOL
zsaiTCVzZi8*0@~lPVzm=uqb6oGHeNb;_zN<ubD<oo6S)Ei0BqXa7&Y#<!$@5)`~FF
zT74Z*x})l!mCs{qMVk+|dTlvejnZ4v9GR%z;f%jh5u!<e+60kv#=P{Q7_o)9)})n_
zMyrkwPcz^A@Wl6TPfSx}aGsoI^7+K+ocRy`;TwMb^8>yA4DURy>j~Y+n=ei*YvlD;
zU$M@Upe%mmo9{nxzaRMg^}>JiKmH|H&lpC0ZwcKf6iE{LtF=mwQ5+--^y8=<DU}#C
z7)+t(v@TdI{W0KeqgaXH{_g#IhSBhF*wJ~%F!q$18T*lCUN}8|Ag`-bNsk@R(?nb(
zFXg!JS<=d3*E8Jgd44((<H~-&qhXj%Cz95>zkNQFQx%HDdAjDQILu|`;a1QD3*79F
z_~7Y>j^(m4uaP+wmMTM}HR8<cL*V({Bi-#tsTpG8aCbvq7D@WcjkvC~W(dwOFB7}Y
z(}%#rE>q?YOplM;93BX!!}f;UDt_GI`VOD6tkp$q;cH&`{I=&m`-88zySwA&_J*$O
z$+fX8GbvSUa2$_EcDsS2ZQLxGh}c61+mWlGw_Y4=ep^(GG^=ic)z+lo%J!Jzm3m2%
zVWB9qIOIex(m9Va;?VF`I#@&wZAc{(Q)XQg(;B%fGndQ6nxuu)I<JYp686=cxN<@_
zV(o=#7R^bphN0^;2jN9~^Y=Oo<h)SQN^cGQ*i$AEZUrpE&=d1a{J7w|0G+2hyyE8J
zGvc~(nt#B=l^7?|I`i&_w^(Z#$DTDu-oAZDu$~{SEBoUex+dUUkQSz4Ppb*%Ep|Kr
z4%af5$9Lk{w2s^)CbH&CPS4ns>2^n~g*eYT;mLs*<3cphcRho50$1lqD;=>U^0dfS
zW3~Zcm7tmHC21+k0yJt~B~#ovS}Ww3uvRy6M9j2Yp2--s3K?7}xUQq`cT^i>+C$24
zQd(t0??;La<c6d@dXKk8xz&toTMMCqDvek}Hxlk;B6rrxAy|sk>UMidZJbVLT6rW`
zi))3!$w;+k;e<m7qSjTUAOuHimStJUX(g9LZIxsO+U_$5H#Ae|^GtgC=P<t`1V@gu
zgt81B_V9o?e9k3#<_}*p43^IKGy_<#%G{>BwQQEwd9@PCMdE!R)~M5PQJdP9X;G+6
zc-pA8v{GI~X4@~}kTcFWQq0;Xu~ikry3Y81ASNVS#&(2#k1v_ISms&EawNvu<GrYl
zOD&{z#WrbsG&FoS(tJlQ4IetCuOw4l0!KRW+6PZd3F~`#FRjo@4JpR~vBTDi2G}$m
z2*B7t*X{W1vs>YCjlh^`oeAR}te42woH0(e_fjj)2%T4JRf{QX4U;-UQ4o~CDbiqS
zCAWg9PPHUrU=Y@CtJmIMYomcgqGq;!E6tLGNTU~a)Ag#=dPNkrbiUT?_3IU-gK(b=
z38o3!V5(YRRkOh6P*7yU8CrqU)57=f&RkZ>|8T7kXmqX-_JQf?L@tSkFTWy$4%3`i
zC;Nf>VaGZzc<(q*6EQ}9UNW|GJiq_QtGgrLz5B?YeE**Ac&DLc!jqRQg*B}tQd)^I
zYMz57e$Z!Kt(y{9B{W?nQ%Ym#9dj0&r_-<JgXiYB*LZG8%6oo%q{ht6Vc_O?<eTrm
z!vozgvQ871X<@e;=tECl7Zq6=N?fot^5)@)v9L}vQ|)Laao8W!86jI~S)ed6bb-SV
z#5&wL_I=M`82SF=Gtbk)w@)iK#{+v;d2@fydVZ#s$l>NeoEotZZg!+7gSmHxvG<%W
z3uEWm_eYFoGTiQad<SoT_$9~VJ^g+s^pbak!=5zHtRLR%r>jcWMOm4bXIzDKT1dlz
z-EJVSllJIHs^dH_)Y=HW<MsWKrCxYEO%&@0hZ}~QJ>w`XkELoOW)-vwB!{I9B1kt%
z)y8!tO%wL~M9fiWW@qV~XXtvYx~`mcQe5SPQdri+`=>L{=L>60G91JxfqBMYYr$48
zOO($HUha?3@A09-hmPfZB3OgBD9tX<;@|r)V64TZGnXI!3KKlWTL!<QwoFPB2FvC7
znQ5MIyOA0K>AmpNHD2(wlH<x{zVQBhVvU6#ef|b#Ja1lq$-2yZe0*XpjpOYtA#|i_
zIqdG)9q+k3za`~GDtyC0jk5-Vh)UdGSe8g!7d#o03wPs@-FU+kBkQtYeItYsQe;|Y
z+BkxPv=&a6GoecY-;)@32i=~WgjvLReOW`Ogs!9Hh^3Hyr%s3fn{WylmX>)2D#i~)
z{b2M~DpsReg!cuFl~sj!6T0JY9ItwYTC#+$=mBYgtSeg8Drrf$8YKU{HPRxv))~uq
zcf+_lGPfVFDQZ-7f?DZ=Cl<MMWDBk0r$u|Si>Q-hluJPJAsn@U=|*ub^ha#iW8;aI
zCVX1tpz20UcZAk>e*Y`-(|d++&srs`+v%Z1lMTH&MF^0GYg1&e>#BgAd_OHmf>I3_
zBbBw*s0dA)AdL(QPMsYM$u)2Y+_)|?K6JXpio>M<%_VH#(`q7?LTD@29B{UiuWg++
zF<O>o7F%q~k_g!}Fh?CEalMe!%-W#F3EMKJ_n6SjVbCf*NX~wqo{{#6Ac8Y>T$Ci)
z^4@j0&|yrYrbQ?j^*y(SA_kspRK)88z4wI9;EktIFwTqEtrT2|5`06Z>w5X#n}`No
zlX5EWr82krfG-`*R;*JkgE3-rH7|JqEf;-GH1k3A4O>jMZB_>tY3USEpd^cK!&}aq
zKv{D^tWm3><cWwAMG=6FVzMg0GFxC)mCADOdHQ(a@xzIh8lJ|!H~8vtP6qIRqkFjL
z{@CM(d-~%IwJf-l_5Id@saDwaJ?k`ae=~5oKQcW(vs@ytACCORyGNG!gjfAXDOrbs
zOkQVdNpvC5(10Q%!ASD?bJU!#?UtrBIa8WMI@?w#abZ8aW*j`JB%Uu5HD|nOOw+=y
z3*-d9_={f=#-3rnBdwF9q*}+iX1@FGJwphbFAI+!pEwKyZ|?R?)0x}DK-&+z`}jz3
zmX;*}w(C5nsS!)%amu(wBC><E9CibxB(c6($xnE9%KX`PCw}kCBWuajIWlyfvMl6f
z=5m@icw}+Rl^7=;jyrOy7z@rdOqPMW<VtOsr}Nh=({D2lH}aZv-8D^?f@sl_ATLu~
z_&8lK-m=UqfBmb!=EH|aUcGrm7dj3%M=*wYUP&%sgDicwyMa<I-MC|Xc+Ji0R}6!Q
z+EbiK%bC2)l2on^8|yo1V0AL2&eKdZj!+A}LGg($bc7(*KxqsGw>w}DH?*!Ntt+v<
z<??vq<FXLv1!t@#oeIJtT0JW<XU1Xhbp4LdckGS_j>iM(e5S>PT19j`FU!B9ulai#
z26p$K;}18anCW-7<WvbBn)QVKAl&WKBQY(61@ifs`Qt}+<4DPwR5M*SaMukS?!Khk
z-%`@Ve0gTSKQd1%-+%kpIM;Lg@D=^d1Iy)^_51<y!hCsVnNJMEfqr+SmPEhXGx~e1
zwd8eXn!n|AnefJNbL_c1y~7$<aD>CHqzigaah`9#{Tgp2JG|=$&Zi4;`JT=j?r!f8
zSZHQhC%o_29gZ05q`_>F%dX@|7Xm>Q>chAv&Sx%<?@2ikd`GYzYb!T*599{t=~?K6
zCiw%^K`?<9Coc0$Hw+Xb6o+l?dN|O;C+>{2`Js(<z8kQiXI&;rYdB*mIkBuOYmwon
zmBjP=@7SlA&Qx-%k{@52i1ny+R&ora8n=i7+{rLtG0LxrEzE}<E_666uoz7ab2r>#
z`n^yAreUg-$CS9ToIlVo<eEvP(TKQ;{C!&ET-FskNGMS7L31cXPg2S(yj(3*E2&*s
zLu{H4nmJ*m-Ws7jjJ2AV;I6Kcv~D>E+u$R@vz+~Pn{7IT;5^MbmJc5wER4r5m`bNV
zmk7v)u9uRNrh=lBXieg?X_6*uo%+*5AjSlx;7eg`vQ@VRah+)?(Tpe6gf-Ib@iKHa
z)>85Ph_lkV=YeUOiSdJ2nrw&bWI(S}OzrVr3Y)S9f3;e1N<nP2hEuh-(Qp(|{tNwP
ztQBnmqUo^&ZD_1lVnjL%WEj~*x~2)mQDUZ~B5s4N`yI(B5PeEgYoS~O+dz;OQ9}En
z!<imjPjf*#4zy8EvL3alO!W?EGNdy&WRQxNiQOQX@E;yeTrOu~hS3{-_LCbr+eque
z7jJHn!kA8q*Sx6KhtOkNB`piyfT<Pd4L|$wXS6l)=fC(GlPkx*=R99Hl%i#6m1Vh*
z)|Hx~6yW5n+pu(^P{lVdISu9dd88IvirkJp-#twPs~VSFiE|?6$ea?vTGm=ArSbgq
z#D|o)y*V=Uf#>HZY!e5|`;Q`4ee>#pQZhfh{mA|O4W0Ltni=|@aXj$;<0GvVZf_5C
z2KHTGO$qNUIl;G&Pt38fCZX1XcU<OK(V4*U=9UlZ2mbnTCN_f&o-gl4q%Nvptf9oh
zdcJVHIdEBKau$6OiVSt3!*`zE1xi;brlTJ%&*xv@_$l4+T1c%_$m<G5BIHe)ndURU
zc>IA6^E1v$i($s{S6_e6v@E=N^MDVQZWyo*(z=M-VmI)7mZ5cbyy0;Fiv7a_{m@fV
z#9ka_U1@b@vrv040N_9$zjevJAFkRUV;!z0lF5W%A%pQ!L`H4ok}<x+z51MRbB75X
zd0D^&!gQf~exxReRt~{SBVq_L3^+??9Bw?ayV*000q-q#+;cMo^38!X$-pvC^VQP&
z_dX18ctzdc6UIu|9cgPOFCTD5-n?Q(-Ff)(r(_eDKYYjWc*8jE$)zx#K2S=d>v!DT
zJaBw?O)V>@=SS8#F%F(LuilW>h4pgb=HV-r@}ATCUt^9(A)myY%zAm0>Demu;yf{D
zR<-5WRtOrwdwzOP7;dF|I(RIN-SL(;Kl(Y#vVbuh4@VBi%;Vc{xt!j!riDC%?*<x%
zaXhf!AC!hru_H~?@;S4fp2^Kpv-r!`TsWUD1UGXWds5Dvo}cJ;J3t%&1W}DTA82%}
z>C9z5F|Q}OVNcgGYl-9}ka*53##jk>$xWf~Oq*ueo=cXHliU&4g>{`M7@RpUc7gNx
z#Il^}yN(tk*%*dlCyLoR)71igptizVlJHU+O%i3hbJo%iqim<%V@w58sab81pxdkn
z9Gk|_=4E1;ClORO3D0RwvKWL%Zd%z~NJTOns_|gj)i-W^rwCBRHb*O!z8kL<%OOZ+
zKkCZZBCD%-Mx_oH@tnt1v>C=o84gWg@Qn*JTawuzNKFZWIDHg&I=tqzTI~5m=zGx^
z=;deC&0w9E3<U2fIg*wcAB5iy-5%$>v>SRPA0RID!BK}BVuA6nC#Oihd{Es(PhKY|
zk+dKoH`XyPku_IJwOXj<$z>8A(gm^ArA4ELoo?@zbzR9hYj%30xr)kPL4OX@)T3V|
zKG+}w-$s`;1PT{}t5mm2LTJ<NM_ePdRviwSz@1K-3#Hjqgnp$~Qd)K6m+Ixz&?;bq
z4O%nHh6Hd$-<rB;tZgz_Sov(!RxlLKA5P@+O3IawPbXq-l$e>PnQq@PT^z$<;ATHC
z1S_J;3fAqxW^7rZB-LmX@P>!azhqrjo<F{$?=1VV<2Vi+#vLEtJ#l*fR)0Rr|ChEu
zYtk$`voo=0k7xQOpJK=XKmkOti$#iRbxV|Pq+g&Pt7~1ztnXx|Zb@w-S%s<sG836G
z`tVI>+><We69Hbb%vcG71uP@N-Qj-D-tVy1v&iynW3&|Rt3a1hzWz#*w8FHm>Xd?z
z0^98#=QR6m%kg+-9%mk(FI@UTE(&j`H!b}z2(DWjatX{!q`&qk61CH$l<3-)y7J@@
zn5K!^d)i8b&r1MP`xVQr`2DYcjUnTWBF{6YQFEOZjLy{FVY24scqBzj&PS}X49m=J
z+wuLIH#FAqTxNcHxpLdUeQUvhP4fOt<HR_Rv~9yQ%*2prH#OEMVp$hT8m46y%eWO<
zBn(XRJB-fQYeHEsjGMWq?~fe&BY*t(fn!)`HyyQiEOWruEw_g~){5n}Kc88anYxuc
zl@KD+GSh7<iuW>Q(i-P$toKsntC3g=F=gzEj+XdgQTxMcZ4OenySkRiUl#Y50I?!U
zRoHq%wcpb1wpd@`<_cmUULQ#Pm9&XyFT_J@R?N0F7$?n%vK7u4tP$a@(SWy@;xM^l
z3IV4LwX6PRGy_kc-{BC|wsjisJugo^<LSinsYh$b^N6cEjPIDQM@$aHWkES%_m>nU
z=ROC9<7bpAbk3q;<lLX}I$>O9xP0K{xgv~{2%<u!+uhRDTjpsbjTbh}h9wq=MMOX$
z&~A6^_Zv!FXquLsMYKd-82U*r`7qITH*EK}w7bK~4v^i<LBA%c2-=}^jd3kiZHQ^%
z@zc*-PDi%e1MOx{-E6ShvP@TjaST_+{wk?ebxpU?glS+HMOCiM%zOzb4c=L_k;7~0
zM^FY^d3Ns2dXWjIJ4VqTVB7V9DvXN=qAKUGt4BM=nQ886+s!I!(@-*ff8}_5V!Pc?
zISHRx!X)+_Wkji(6PhgD4hxMlkUeumz%*fAx$?r^wzd*+m9dl(h(QVkwOYgE=~U#n
z5T}W-jFfp`>5mXcVwNsU8%MU5kdk!I+7L=%S>zcDYucsGvImrlOrh2hxHd)BB-)5(
z14(xjx|d$9g|aKpMF20XV@2sc<p7Q;isK>6@0A$D3K|u0QJ8jz**>6_h0s5tDyW;c
zI9o#uyqqtR+_rYrzHX#TNYc56n6b`-ag4*2VH{}dhNh~CDGJ}9L{c<pFMjwqUnD|1
z%tTg#;4Jy}#w(V2S`SP^yV=PlZ7qsIN|L~-O(k7^5|Lq11#E?KGU-uHY_dfOk3EEG
zErb->tTC&I^C2ThZlg7_y;$vnMXi>rm7B2^07Vsw%B$L6+?LimiP~P_OiHg-Pwh}n
zCIhf$4=9IGhIyGKVNr>AtE{UdzQ{Gpm)XWD5KW1kW_cbds3o)X196^6xd^2)FDMd+
z#__Mdy~k@wd9+q?S8@o%xyRM^>*r2uoH$D0Wmd^Zd^n1IwW~ee6q?HO-FM$H`*sc4
zl3WP+xdfDQe-X;Ar}P)iN{m4i_Ume^szeA{dCSgdhzmJerg=f>3Rh`%+l{Qcw4&W?
zxLz)3D{h%31YSKH{*qC!%;Y#@RiW<%t?s=PjLvzKR#Z)muWRwbm&`I=iAhscj<$9f
z2dKhnh%9wY0$#3io}{8Uk4KuSVcWK}wdXhl&Qs?7bI-Q1RJLLaGf&Sa?(PpP)66t3
z<W$&f6_qoDaY5^vx~}M#i!=&WXrxxdFnogCb3XnPA-Bw7VO$30xNsULJ`QKm;|%h?
z)Xj#JGn-ArZogyOd3<GXmB;#;q!e|#!T5$@UYKJ*S#c=k6w$0g@hnNUaaqvbp;o|E
z$$8b^NbW#dX2F%Ox{zY@8e^1^K1ZXybo#1Dh?^(lYs=>DfY!f8mo*!qq_s&YE16TT
z1j1GtrB(AIS|Vx4zCNo!RagHqYl5dA{)tVTu--F0eZ+f7;EKkQPLF7#v2`b-*x`lB
zX-XIvVqky&7F%z)o?iI;=>tu@Vqo(ZWweZJucx!@qm^Uo2XcypX`=Ry>-k8fEX(!8
zJe)|e!&f!)Fi4`_IODu0q=GdS%C}UmT4`DWMX6KE`S{Fuev$Eh-4T}vX@QifjOF3(
z4KYO$h9xc%k*qyoxZo%ZFOQr)zNfeb*KTN5xX5_zWh58_uU~&l)9ys!92H@j$&4(i
z;OiE@J&=qggn<%el%CfVKj)-#iZO^l#~HD(<szX?+KP&L?uqjt<JJ|ZV(c1rVHJ&r
zdb^jf6K$xv0|~{;hks%@JyX{WMp@}rn=O!0%N1oDMrj%E$JIj{B|u0kaK@mDbUdmE
zHYVqlT!-YK=O7_LWmUls{fXy~KT+#U(=^1GFsonP={2T!jn5?s)-~%j@iL1@Q#3m=
zo>r^n=gXm5)+<30!9;wjR;WuZNiM8)=O&uGyjr=WnlC1=mlJCNwq8tHQvkiWLx%-k
zX5#n^?!c@o;u4s0;bwP0rOb3a)2zTBy;@-tiniOLO~v`+FU*UiIoh@TNjcKgjU*fz
znKI1RGfs(<MdbyBrLqQVTImed-9GEZ4KObwDMh^BP;GW7;{<O^L21i4TySM1Ylk%o
zw^A$(BC1@L39Z#y<dt=OA&v!M2>8_<U}P+>BuI$$5|^Kjc+0;WKh_wM6@h3~)hHtr
zP+2RaHEvjAtw*h*My1ReSS0CyY8B@yZN!2KKv}WitF;i(9{^tuaZ)B+CrZp1r|{Nu
zjaTw;rp=1m`vbeZWwSY8w8A+pHn~^`{fS&6uGygw`>%0+9c4pFnK@iAy3jTi7dy~a
zvZ}g-g?dxryN)<dEHSVwK{U~8(WJHhY9&-!GfndvI%X(wovM_?X4lfRHQ_pm_NoYs
z%DRTi8__GJgx3P%>8F_#7iyEZzuR*-Y#92H95c6v9o|9T_h{cRTn7>|9dfm22@Ae*
z%=0K(5#w2=er2^NHm#>BP&XAJ9SOmcl@yCcYYLi}6qj?4HJaL3HYPF5fuEiR-tHWZ
z#D~*~%P?`THBV13TrXG1&~0msmg!Mh?ZY`lk$o%0I8Qt$S;xgC5p&|h`8~rNu->DL
zB~jRQTb7swdX)^9mnU|+J-%|QEgZZtt3Wu3OG(S;Tb9V!Pt>}G6qx!mB`>6TVK_Y#
zt`{ttrtWZcD=PtOCA2R^vQomNLBTrR$t#;cVdRVq31<U2EW~;GN`7rrLANbiZ^>&y
zCMi<FA}gj+$VOA|wzS<rGzvwCh|4rF_9w=${6)lTw1BVvC5nM=(_!mMFpeP+)5PKR
zuc%(tgwKBvScdC_nzjtsO0n4<DD@4-*Ib7yLxBa&GETHrg)v@68qQFBh1+XXj%>C!
zSnqf_p2fP-ZK;)JxIRh>S5?ctHwLhlP!NTw25&5lZ&BXkHwW3f4??0cRn3j9n8qvO
z1PEwf(HveA`V0NbCrVzJ=LuaOKv|~YO4T$dZE*FL%gYN-k1u3%q1{g$DRK}M=(-MX
zG}q%3-mX!zwgV(tQJ07_4y!#e7qT|g+nsoJeS>ioa|nbuQ(`8m6`qtbI!E%<%lu5q
zeop%iSJgOIVV8=qO!V^$DT!~msu~K4CG^a*quw0oQ{d-!e`Z%T+}*#$`v#|ES`mUK
zENh3D6GYkjr!WgaF)LCGBmzX%i=arV;CjHN99L&X5<NhPEaM<(Uc@qJF)Sfbn}n*3
zK#OD+N6G=crT~^4Sz)?z31x{^l_FA9@3O8~(XS=rlv%sRyb2J7TA0z4qC}vX))?a~
z40*|pSVNgsF1a$Y%TJ4}lIs0EB`?_iO!9`sR_F%4x&M~k%>&bT6j7dYQczVb<1jK`
z#OG~{=CJ8#cXv$V$asDt=d^Oig$pm8m>hy3g;lCAHb7&nkQ7lEUwsvcmK?r<P{uKk
zQcqa3&(Q^EB&f<bFP?M-DoVt#_9ALhNq<%K8|%M1I!X?*e_w5!`fD)KB4c<ZnHo}f
zt(H`+!~(e5$}kepr0lBrUkRBdi#YNtR$EE8l)eZ(wOZjU7G)fznAP@~(Aclq2BlVe
zagm>;UFmR{g7O(_74<4c%hO1#CAIP4u48xD(e1YKIZ!xfNl6ahbeeFsqPSYn;$fXW
zluSu8WuD37$bQqXEQR*{5pONiX&|2-iEEe@IpOV^LyUaQaTN-(3S}t;>l|&<u=E4h
zX=1;rdHwp9YSXf{6;3PWd15}DxWCzQcY9!7Cfc?`5%~Ok;W8~8HlBxF!~M;cB?Od$
z*KhBIZ80qr3^6ZUhaRWku<uwx#9GTd2im43&WU*rcxxEv%+0PNtBMjM<{D@l&p1sq
zwdZ&a%qe43=3&>+c~6)Er$MN)k9}kuCzhq=@ig%1bmF$M4A+tKI8iBur7&LyX?kL!
z?*~ep8J2}5!%jmnnqPi?#MK?njucmM?I&?rECJ_hx=o7-5wA5`*OXG&ZgzxRSQfDr
zYbqcMFT_f^XBbDalGNRCT8QP$rmL8(CXORvkh_7LsoEWtuSArLoI%DqT-Bf&x&Lxb
zEMXxHBV{F@uJwUT;bNf40*SI#gyS1qQ#&KDsCg!ZOw4iz<h(-Ua9CxistVT%yC%nl
zWv=iZREorDCZ&w2<URhc{pX(-R8()Mwl~sc)*Ua8zi|5efz8bgNyE+dj&{4_d_Hna
zGk8z19XGFk&G~XBalu-#)}^7B%c|Q^Sx>h=(Cl6jrkV4{KjHF>&Vkld>~}k?mQj7q
zP;GD7?{1j-3*+_3GLCHT-eBvF7$(N^GuQr1>+T_YP>Fe27{?K99Zl0wH67#d!u5LO
z=H@LqEa<AGxqZ$2{EoJ&NeXUn4~*BswZDKe^ux$?oQTTe>Mc!oC-$u6%9}UeV69~d
z6H6`xTXD1Bb9#Ox=R(_Vh#^o}$MEuwPtPCO>~Co9zeOp-5=N}jxN1vPZ#lobCne~*
z9XZU*{Yefjm00E;=eIQNj+`RH<>jmJC?`=>Tcw2WP=t7;R{!$bSKsll*%K#G8GA1y
zPo)bu3lcd*0iqI$q*h-ne1S0ag3v|w>{$sU%2(Bz^fnX7U&m(sr(gK^r+;MI*uT*D
zjHM_|%mue%XyYO@#t;^qka$o<#`q}d^sA~tX^S&fELp2hUF)?TSDN@mgybDo=Hw!*
zTod4O%CZyAGEcBhI{2KjsQpS2*+G^GMVNXZqgAHf90+kn#U8VL%jGok>HQCk)0wL3
zaJvWMI)QM$rAmoVImxD93d%Vd*QS7Vo~jn%(aZ6L5CYq-rD#PqkkQmtg()*}ys*q8
z3AxnMik+>!lb=f#oq{Nz5%Qp2$Dnx~ZMO$hPBiU?IFFPxgL9-+bZ3oUy9&9qzi=|N
zHW=-spjf*YwK@(`j%z_+WC#AG^R>ci5I9Q_@$0(d|6&1K!?8tcAf*~8NrEhSO&={8
zWwHbgt>;=nf3+-18r-T6SiRh)l0qT&SH>j#i@NUCy!(vPiEg{${`CX(;Xu_kq$Qw=
zToxtG#5|&%Cx&^=?J#7mMX+kDOel&7Z&gV+t@!5kJ==FDYA<?}FWDx3Es~coW1=R+
zAfU1gO(mQQ=RI{*5thKv_Y(hZ8w#1tc8jr6#8nz<8#!!h&c`D$CF-t06-VDs+}&>Z
z=B8u65rpxQB_wN{N1Cceg}^vZ%#)<YIwx`NMvJbdY8v7+ak&g!uOpp@U;Wj0l#+Nk
zT{)izV#&0vBd3P8>6qp~nj}mv#>{!X5{%>6_voUi+J+@(j?>I}6tVF)ThF`e6Gjzw
zm1SF5`Z5agIKz3A?;T2W9i{k4fYyoJN48DN+izbH`^a{;p{fkyI5G?atgBOJJ$*lr
z(tz<R5qB7w8cXFZ)>Twh1x2BqC+iBb;^szj8>VTZZX02fgn1PVTC|nmrg;g(Qm{=&
z<!Tz=V!h~)atxTUmn?&M7SW&*aFjC!-?Y?Ci>pMFpjN6cDKQR;c^aA4bVj8$rmiqW
z?1RQt7$^J#Yb~mH>dIjaSTxQ^K14{FWz}E(H!lWmzyH_l?^>qQXY%zkoj2rhAe|;k
z$;@#gJdeaB@$m31?(mwI=SQBNpGnh&!{Ig4I1;XpL=|z~Vtoy+Vjd@A{6xu_{`g49
z1J<`F(=ZJKLk<l68Sg!JZ|+!@ffTNkG;_UPC{@dCv-!%>4{_n?^DkIi$q%PQ6dAi(
zkV;W_-^$Kv>M3cY*h<j7+H*O5W)4T5o<Gqw4bP8{U>wuD@bdD4bB=9OfAvOdQ_-{=
zuH(St@ygAublfQf!Z>g~eWZ)G7;72(7vb<L%@SdlGu|0ar$^@LXR3NjF`6)4@W#<r
ze%%+ZmOtlFcFPhI*Ow2ZFjGRHsy%JH#kzHrSrSHDLRvVVFKFAd+3s+z#kdx-#9vz9
zV6-8{8Ka<RQDG~qm`LPeNY2p@S7MkL&d=O^kJxD_U%lc=2?Z{sL@An-VdzJmpPw1}
zGnMJ^n{MsW6>jy{e#t2RDmF`r<x3*nno}VV8}n6xt&N3ZQ5sQ8uW%Y`B_-23vHxi$
zIqk-X!(zR8Fvbv)Su1v7;cL74>=k@r<$n>#x#UUHh_V&B>R{-}^BJ^dyFJh}9lpI~
z4l^MH$vMw~M4@V15)KpP@`(LWevOHWR#4lj&{U2k$z`*>-Akfo$)xMcO5X&f9ol+f
z08}-Lw#2yL>jqypq^n${jnjfmMM985iB#<?oKh_F2x-BXii9JV)frG^MUZnrDG@@I
z6v5ebDwCP#0g$zVa0R5xRq%BOF1F6NVu@ETHu5}XtygjBm*<7m4=wtQFX@YlHFzi|
zVkqb@SsrWrab5#rq_9|ntk(anj6^+`X~3j_YijDIVwe`Pf+Z#HHXY4=hp!uqaVV#l
z`!gjD7!?RPP$7}p#n+-!PzvWObk2k<q`xmh$*{~ke*8dJdz_IBdRyzSzn8xVDvQKc
z=M+&HV$@&t7deZ@XPy@f;#m0R%`5KiANbRs{!Hy1m&*&jIG#Sfr`{g~RHY3eM&7)>
zM=w1}Ww!e*%Myq=v)yf}y(2D*FgUEk`%LX6X>yupYAZmSm@~erP-@_MzVN$$_gmUc
z!^aPwd3^5q_;|z^WF-<TVHs(?!y6AOuK_=jSKt&JHtpKkD*7?-bRJ0wE=wS(35TMw
z2CFiaGon>+75yTj)uys6sqpRYmT6wtSVw2UJ43HLRokMi;`Z*A7y`$aGih40JW6J_
z-4RPzr%W<E9LEu@3x}#EhCp3+Xy;hgL|9a2x7*0XWwmsYg(PXT5?9H15kg=tRDiBJ
zYQ$!!jOYbyjH~#tV6u^kk~V1XX}S$<+oG+L)X)`GU2-C(MRIYp649XbxXOdFcwghH
zYQ1+>Cy<uVw5I7W*5jLw`r0$j6W8nIUq(RKUtf54`2~|kx^_d;-LPnb&XL{r6*q5x
z&GUypvE(O`3e;M0_wYaguFHw@(>rXt<D3;)Y2glx163}>X`<*8Axj|3GS3+2aK7U*
zUF8LuM&|30+IwOa^}KT}53hDarExkjlh8(Scl!ni%)^nU-eRq1Mqb`|nm{YMW{YzT
z<8mcnSgvQ%^b0yAQi_bzApYf2XquKOL=O8shy9K`PsE&fd3s0A8S5MV@Q?q1GMc(=
zNnzyle5Ow$#D%-vfjKK;EMzpc+VJ}Dn(SI!yAiNZ$WR`cE}uzR(fAfkVd~FNBHpfw
z)B?8h#CfK__FS(g6bW<DY&R2iRZ);#l6N)NvFCD~@G7F?Ox517+aFM>W*QTwYQ-}@
zNQOjMX1s5R%fdKK<TVP`8q1O+%QQePRCO(BgM~FXDKm#C*#W)+6z$$~pCcs>OxH)r
z5Xctm?V664D3t^Q&C`T2E4w{Lv{C|DTGosSrHRWdW8jh{p|F&7L#I}iwv3lW@V5GL
z%L@3ga)}iZ&U?IdqCN+)ajv7{FS!-Q2zAUlzaHjbRKippCInO(Xx)}^jLhdA<2#zF
zMk|YPhMX5z23(#=VdiojuqAPN{=jCtrP<xfYE4Pzgf^Nyj||hm^`!z=vD+RnDWYA2
ztJ+mSCl<ar2$;qK^E5E@7mS9=Nj5@QW_));j9HTQ_V;KF*O$*M{Y03r>+YY`vMHlt
z(GMiTG+{JkD=6xcGcgKuNv{#aLJj;2CR<`&HKok!ssV9oSifpL3f8#wQeSi5%NH@H
z6eM393vs>hbC!TKlz^y=@uHhR@Y6D3C|`>cRSJ0=Fkzu;DvWQyST<Elx4Yry;SF_H
zQM6_nE)q!NEOD76Bf=zVt4Z@n7)~hXQAX^lx+s=qV(Kqaykwb*71gtZz~yrJss~uE
zSLyf=L9fJ<e}M`Kcxjm@jP+DiMT$cDlxI;c!f#%^BE-ne{XN@l$8ywMj~5b}DHgW3
zw=B~{yKz+6a_u9Bn>S!Am;OZCby#bemIYl3ZP#(_dyFx}W#)W7bGzSAd5f=Vs!hvy
z=_w_1cef|yfpI9jyafL4kMD6+#r?kH=Fmcjl*TbmFYMX|rP68%jcjU9Z4K5pJY6qe
z#iv!RnHI%q5iEP749N?R!_1PZ>VgO^os-M8@tSShaM<l=Zw{EvD^y%?%Y`Y-%)>-l
zCU%>xQ2r`KR^oiB8?ibzT}RBCx@oALBF2UD^}@j&C`DK`zG`vWl2srid6vJ}y0wz&
zfks1^n9nbqo?p0xOua<1s;FFrF<_jCkW-9n4vvu&RwY(bg@}Dij-(|~vPD-FXfIO$
zV}u@9`8DocVU$6+N(9|V(<Ds<U#AL9(^55xstNeIp}HPPVd5_ep8tl$!0F|g!%a)q
z-LN@4FwHZf>Xk0``uHR7-~B{YNhrh1<?k@NTePdhPak_Kz2V#Mf6MXs%n~Nbcw{_(
zqOLl&HxJmVCATdvzx;{e(sR4LXMb}*DL8)kfhmpzppOad8n*j8`r~JEj=Vf(ISj2w
z8^tn5!aP$oTePmpoN>M)rNGpmNO`7dI-2I5tSkEd#E>Joe_`9yG`?cNK%8i*hBx~=
zS}$U~Yu_^tJ?-^^^%a}V24yt!5{cPR)iuu4m|zG&_Kez!ci20J@t!b`jKf4sk)k~z
zDZFdxs)_UInI+GdrlZUv;<8p#RV^r(`xg|7Z(cnx?>2;-P+IVvMr#q4DbU)p*&cZF
zjg#GXT!^{Q4;T9Dh3R^t9<Nl*hLQ`Hi-ey%Y;N${a@*8c>lv5Cm?FPSBgR%MAyPzM
zC5zZh$X;2(o~D^J&n)vmoJV}SXIUmF8e=W7h!PysSKMX@f+a0VLlHZabut#uC1b24
z4K8Dk>o#InQ)ZnU1WHme_Ct2oIVbYEPfo1sfGCB7Dw4dWd?g8aYu6@3L9kBJD#?<m
zAnAP~c&&9S7tDOYSDxe2^ZD^J?fIF*{zkr+RZG(}4AaEN4}az|FZ_HOxIgUp^*6t!
zt=e_}4PdCdn^k4J#nv++&cv8`e*8cR1Fs%_L*3Mbc_c2g&>OAAR+f^ASm{u>s=_)$
zRuh%!$VEC0HED)VKM<1uaI}JXyb_}9l#NkXvv%uRLzXOi`8_0)U=?+#)$ZwRg;EA<
zAgaGy=s64gg;ns0vgrIJge#HKA|KZ_6kC1e800ULf>rNf(5ui;Xp%VvNux|rm;(hM
zu2a8NT$hquQhiUEXI!(VsDk#E{rzi9)ygFv7nb47`SAm~K(*O{%G7N~DG_ZlIt2l^
z%mca+3uc^q;ykVy=^AA;%Voql!*71`9iOK6{BS*ymIc?<l(m@<3Y|Rfy6`oV!CFUM
zOL)>eP3zA_R$ssR?tx{zu(`jd=_*OE(~4=H@bgR=2A1uXZrjpsTb>^udGqZX+IGX`
ze5RCDsF({?RZ~|LPmhl@RYjhqA$t4z6;0zA`wQBF(U}lu+O{IQObU_9eC5;2ncMq2
z4qZ!IX~KM^Z7VM0A}$q`DM;7twMFMb?JdJHGKR>$b=X|^RehkJ6VKOa)$9lj^2-?A
zt<$W^IjYLzoxy6$G)*X*`EB<NAy@Ri;;&!5;)A!mfA=R|j#Aiex)!YpbtUhy^D<ql
zz2@%C8;q%#mH=dyapZKm5K_Un8m9weSQy5LF%>g2CVjEQvPKSPa+rxx99GK^2s{Ip
z5(c`qp{gtK*p@7IV=L*mAcZk$UAHa+Ie~FC)pn1skd(-A6hwZJ^;ldcu`B8kbdiRt
z+fubHDtn4?IO{RqW3|FsBNpB=GoRhR%w*tU<GI~8%+t&~U#WaenxUk~G!2C7kyktE
zausqoG<A#H+@NpXay`E1(=UHuvjkqf{Vl`LbNrm}R(2n?EnJ>|L2E<1Kal5v>(hHm
zwZm_3u-y&Yc870vxYls#uk_;!uHB%~4Cg0mQ;2C{R2sBlnr7;Hz}7po+i)GOSe<dE
zu#9I?nDA}Fv<B1c=xl{*Hl*=_^`59Jrt6840xy@5V~T7Vk8L+J8%@)7*s4LhT8N3`
z#5D9QX@+V`)RyMvAbk7dC-U4=RSk)ZC2=`F^V3g%&)utU*xlZf!a`M5yt=!SZit9A
z1z+tD^|i46tE$6lg>#OyDynnN%uB#q%{caa`0$S1ZqK3XM0DdT0v<wS8YfbWG+ifO
z_li*7>~EoDTvdx8r9@&%=v{|Pg{StJ`tFrX5^`8o+){ifZA7Fsi+JXG{KU(<ADON%
z)J=^lFfW<9(Uc_nV#yXzgcU&>vk1bhbChf-Mq^#Q;%W`nRm5e%%G{Y!3Z`mN#gpcl
zn8muI<hCKFAeO}tvBqI-MJ%$z`x+mLd>=(T=O|xC&}xk*{R<t>l(>q$JY~9&%E;kx
z$8K|j)`eJrIJ1PA`SMK8(B1ut+iu6-9u8cGE4s*w7nWH>OuDS7ZAV%L5rOtsLRk3u
zr$1uqhN3F&4tu7MsN0sgK-F!iT*EM(DMoR7|1Dven2+y;i*9Q|3Z%5qRGy_jGS3rP
z8Jc#3_nz8Ya+tB!P<iRtV~U``R&I4Htd>iqB%x4iiU#Lu!TnP3-VnpQ2D4;243S_L
zltUT6TA>0d&ZrzwNq7N9SD-Xj>=mTdox1u!vC5!LMT+Wc40Xx#nv`i*SA$G(GRXh;
zaloX6BFoxHwF3N^2CUYC;|&XS)rd!138$j=4LQw}JV__!B_dkIfHt0FD)LH8L}}8p
zNXXH4L&RXs2Zrl~DH)2bXf_+c{vsCV7$lzsB*($580C4I7{@iNs1)i-M6dG{INTig
z<^5;8bEFvX#xVCiwY4<bV6|fjiJO}P4>t!w-!qIOw|6%*P0g}IcH6Dko~=V!OXoUt
zsd1ICFmhOM+CY+3U(Vt(nC69H4D5F;zkB__d>uHRj>7pT5S9zxIRJEtbZt%F4=81r
z5)4bCHkwK+oD)&*w)PylhJKn?3uF@3hmtG_YeaLBN@7~Xl6e|>Qa1db|GJ@?E&Xmw
zl@oWnd%}F-`RSQym_+xYgeRi41!Ztv2&0vEYz_yM@yzo~-(R?1u7s2+OF&lwn3|`F
zq7rpop(-PKDXDi>XG?|YZfR&J^Fk<!dG0wrEi7Y0(>5zrQpBttFF-OBYpkp=MAjr-
zZAWrih`tix;XF~6N!a>G`f3UbSuLXLiZJDYG&y43;OmxLHN;p;YABM^V)BCjXWa1r
ztYRR=i8(bS1;@t^)K!Z$(p6<L({v>gF}_9lmU)_Z{PT~z{!Vjy_nH_BO;wYop6lZ~
z+IGh`-~1MgVH(aXaUjIR5=MMgQEd-6ZBc&9?%`XkHH7}mbo#`6eI^V&W!C8K9@lR1
zb<Obd39l0&1|FZ!Y&Lt2%gp}uckH)!%!^p|tlg8RK~@GP;jKiuhGix%Yr2+pRQnBO
zo{3Q+QLpDGK772=G#!Vo;c$0P)oszcJG4@a%R6$JsW-1kb;HxApRm)!&F+S#>rh@g
zR+5a29u8aft;M^7QxlU8<dg^@(GLrzYM7Iz3I)~&$T}G-PAL$4gX?xe@G43?=$9wb
zaAi5a@OXN{y?RBnz2%zZvh!7gEfHfSu2#7UtG<lXWrUVi&-2`0Sgt4Z?JI13$K5x-
zWApF|XALR^FdnNVDbQF;BD2gjaX8ao&pdtpg*Xnx%?4LDEHUBGazI-T)=(78aS^}y
zY8f<E_KwomD~UJGS%h%L3KKs{*NakU@3EEaigU@NEEdO<g)phKrLfZPQX;OqdH-d6
zy@n&f3V~Xu1ZkN8z5;r*t4L@`rrh4K^j8)Wv2H^Q12JbRlL_M>-J_8}lDG^!eY`Tq
z#GAL@kwYMjJ>D78GK#O<?^sra|GsXR=8=E3zo%|HYOirvyn<MgK&P}}4iZRoJihS$
z{ZH$7TP}BH4b5hU)-v+cx!^WZuvlN=a-oC;l>Mst&`QXWN$`2Hwvk#-yxvO5B*m!J
zW@y*$-O0tfCLDsCeOg<TuE5xJqF~9+p>kaH0iw+SaUlp8NxJSr`C5c|2T{p{Cucct
zat<rRsKVJwbO%a<UI<IXTgQ0qQCi`e4a?9IVx+_<&nyLPh44lZFi;H3S{wjc5*>|}
zgua}M6#W>Mp@(V%{)%r3FT=>o^O2^i*tRX_mn&hK)|>_D(1mT0S)u@pym#`Bm62&2
zNvlv7Bwr<`#H+X8h@NE~CscJqe?4Oq%)^zJ&mXbgvVXW?_i)4Ia^>;eJ02eH*zLBY
zoVkBE5T=Q+EHrgZ*VHU$fwpZA2bOW7svDdW@@0$xUpZ_ky#IWn*=+gtX3KWJ=jZ-H
z<;9vhr<tZ{==%#HC%%2SWr~^4FOS?FZu#`-nS|l{`yGnRzHPZo6T=j^X&Mf#i29DF
z3rm3|B%F4%wI!1nLS&AS>o9X2Cw99B9`0ZBYI9&vneq6{X?UbR_mZKou7D3WJ7UgE
zOXB(E$gX=UXF-(rezVz1+NL%5s%4sHrck&JBUuSQ$5ky=Rf!0Bosg-OT3X1I3{5WV
zLip;QA_lCnSYr`kn%G=RUe>ck3V|=eXj4``8Vh+!EQ|aZl`}SHlonB~w-%*61zBS%
zol!(g36zjY1XLP98-ausX>oE+lr+lc{cl|i6lItrRL*mA<HQqNG|Lk3ZA}$gw5h2!
z2d0?Ny24HapFaGVWjIrlF}5S-!0_@jp|3?Gh!W)D_70m2$B%#JbbQC#x8I<AO_&$L
z$3LT#rfE0eJBR@lBIfHZQlfI(yE|-In5ThdoY^!Lb!7<0UpQTk*jns{S!qy(6bGj3
zDA1R*P}0J@M2xSPvVb0%!ws{2VLqPe>YAIow=}o+Tpxc~1wTKsnFp%vJ+9eu>945D
z(H(A4s*ui)<nu>jy_M>I?pcNlDJFsCq{uKNrsqennoYNcGHEE<V4Dr5YKYfN9L|Kf
z$CgfLgT)c&D^6*wZ}GOG4;@~?Z-3V?_7_~Y=k`~BO(}(GxJZCfiKMuYiYS0X7)4-d
zJk!u~S{A&Qy_R|X8}!|4ylJ@AimX`K>8q_zi{(mbgE5W}6Ztwae0t*id}JI(LMk*V
z<5HsPHpC@ieM88AGvqaNM(Z`^*I1MiIE>0#M2%jiAj&v#Mg*Zvk|Rdf%u5ifT}jMg
zM%68aLJSG-y#&OplL%!DMNmsv<Gy;oje^3;Cdi^o_@Zk60=vo6fK>%mZ)plFA+soj
z-`rv5o}oW6jXhOW<GX_Zm*xrIbSRU#ygcHpqi!2Au*?&#*`S&ob5x8=U>*Z08gBQu
zSOO(X6z5Pz%NW*+@F=B735rT<>RMB<bZ`DjM0rt!z2h+B>v|pQE6kU)zZ_ZSE8}>@
zR4Y~*$#zghRwLyvzLM1%B^nHgl3iBSe_8Rs8m01T@%$pfOlTdkYE4Mg_G{6XOIR`6
ze*r)#B;!CS8FhYPKghn`kYfVU7yd9>8=P6a)T_OF9%rt{E2*mKm!5}*Tf#h%l~@@K
zMZ$TM##WWM1qx_W<X~RytT`>D36vFBQd*}(1;E!>WvLUOkCfpv(>N2BfHfNLEGZ|9
zuH<_G@~jp`%n7CP%J2|}MObrB1mq-jr1y?j-+aT=3sJJ!G}zjs21)g+tsyNl!|6p*
z@RZ`|^Cu3w4NcdPfx51k$AROsq#3qNvz{%A?d}HcJ!3!6Y&M_^afYUD$RW_5&cqP7
z-S05EaDILvOfz-mIbSZI6k!Q?XW498DsQ-42EKXqircm!g~W{EZrgG?pID}eeeKzL
zi*ufC+Yxi2b%o2gFon!GENt7BuC2Mw3+HKOiJ9v>i>=jL^tU%`(bNxn!avRY^7&7~
z$jF7RspuLJ-1TGN)K5&)jCY=<sqoekmW8Tr<Y7?;Z5+#aAf`gqR&<*dUv*SfCGG+J
z7lwtRP*hrpnraQR3X<j-)z_R9l)N>$h?|L$DOp79S`~~kQcPP3VJk6`li;_DQWOJ1
z=MZtAwq7Q%MajoXP>K|^2E=xrgB0RlvtBZ!`M>&p|7R5gzTL{mMTxD+ISGG>AyOHS
z!h&-&+m1ZWgwvJAXojih{NaRd9k;K(;o1*euQS?0r4%t2mJms&XO5>MmABMx&p6Kb
zI8aDjjw74xfhF}4^SR74RZCWeWgHm8OwooACv+SMCF8mcPv;Z1a@5um?FBn&{B^*#
z9p1@9xr|q){t0C&xv0v_IL}nJFkg@46c`>q(6k%uW<x0n^2E)nZy1g*OydcjK5)HU
z@y&*19Ej(CK%YLcC<rf~NomGjURGSL7D0$Egc#|zJNjWDE(47R!^Akv%u~dsz#K>H
z;Q?zjrD<6vNNdueUS^ilk+@9M?S`BcRo&p71lB0kP<h3Cd`B^sJkQcyJI68&ED5}C
zz*b}p&NU<_<FCdPLMgb-ju;a`M@+F)z9Sn*vmAWJI0@NFi4+7<^z`%-fBx~GxL)UV
zH*UE0Gh4r54uxfk*k)D38ZnmX_0U&ZNQ*KJh-58^eb%h|{hZMzL6WI~wq6wQB?-jC
zI*e@ut!nL8Mt)xZTh?yUIJ2U4g%h8$Bqu2hMvI`#SSyKgs}sQ(`S&>nSVp0NO(RRG
zD9(Y_6b$w5fl@NV<;3y$MAdXu4vKB4s}05(-1Zhzwdm%BYyZM@e8O2rDhplPGENDb
zXOu2z?NPd5RFMK00~z8pbG@DkDRXms$8b4w?LROLSDM;ktfOsON{r0Qf-_c*G?g*R
zP}vq=HH35_ELV#50vlR8_wuDUPz6j&R#`%WiX|+=%6E|b_%Ex;Qp#6UYt9*`3mAtn
zZWUr?iY{NL0HqYjMp6ma&R8pfBYzpCXQbx9manUeoWvuJ5)XC=JtY)|mkT*1=~&Y&
zR@QkCAdD40e2M|B6}6ME%NT=JqIp0U!Zf0CK-)m-fu>Nb#Dpt_I8~elUo&NKC)G^_
z`YHuO5m=bf7H5p)R2Yp)Syp>-k!MJK39w~N&@7}7*d99m_P_fbb6mtl6c?(dMpa%c
zZSw+MVd*EH;v?gnc=zrzRb}a_mf<>3d&AT5#MJkAYq_~OFwFut3iC{;D%y5Kv*{%0
zDrHJ4(6vncz#JpK^2E3>Tm_!AKXg1jy?k|$SZjIn&_K=%<4h=-)>qWE<2T>lk%|()
z*nY?HbfR&sG;Wo_=)kV6$z_A}6XURum%=#C)V1T#2`6A|Jo8e^IsNj2+V6>n4wVW!
z-BPjy)K%V5RfcWr@U`dW?tp(h5vRztAJ{Gvb=hE4VHyXFt7w~+=jq6$@2R>S##q{>
zqG_wIFfXmjU+l0Y0eRKdAitVSB9nCHOGFo|hDH@uk+Zl-)cQ<YZN3In<&udjv7me@
zs@HQwrc>%GXG98&L<xG;GO<qFzUZ6cTCm7)OTONJiOB$pW_K$cby|qZP*g;j%=LPu
zsysR^e0uk1d|gvYz<EoPJtbLmT@kNmysfzT_BZ%;Pg!Q>m(R?XBTt_`ad&@5(^M!t
z#yQUCN7=g=OPa6DmkE_JfBNH(y!!Tgn(bEhli4ti6O}PoU>OJc>xi}$(=5P;p+8}@
zrqY%z5v*hAd$iHAPak^D#}hO=rg<img>pJln?e{b4A&!d-D16GTX!g1v%P=KJa3pk
z{s5OV$qAHc9L}7dKa!Nk<e9pvz*|BRQ6oh$3|D;RxK1N6Msg~IF>=uo<y+M)AuLRn
z3*+-Ue6tZB@Ab<5_MVtBOVVf(VK}mBuhex#7_}VyZUxbtkHmQ*#K7G*zhasfrlF_p
zwj@)N^NflMB}6XjLUZ?)y4&#K$A6$}8<v;PxcwbCk0FxAE6G}{5oWj0*^HovV_-PF
z@cj7`<1{hO3))*k)?9|3&CNa292us_##Suj$g&KyO}%O#L_j4ZxYgsnrvI^qfGDjo
z8sdskjY$H1BrK_r(}J@CaQc#NCwpci7lNyB#)=YG_BBvgyE#$+>QX3H_O{Bbot?E5
zt<kO`tFj8c3RV~T;e~m+FwH$)YifT7S`k9vbnf}(rw{C2y<y`#|N39=@l7WU^)m7K
z!#n=)PydlOw>!SOf1rsottpsrWVw!*W`k<?6p-)lS>Q(AIfiMXVAyVNiPJ(_t{9cs
zc#STFm#0sh3f$k`(f1de)1(lHVV2I{)SS;3d=*%xk*Jon2mnwx>jAmERV7m(UEJ5?
zOmP-ShmulS?M}kj|AN{5GHK9CY`rlhN?P*|)@=B~N;6zN;$Oaol(Pg)Y5DzP92QVH
z2u~u)w1*-&_9+y}l1~xHY_~hHmu7`kVjEq;%rY$^^vZ>}OepWjc|n!1PRJa~G?Jqv
zZ))qsj%&O)I#gtd3n50Pd0`2XMIXaLoP(rHx`?Y4Rb5H^Z<K^kA#KVt9+N29o%fOo
zIxlDxZ{HlKt3-c#M!SkK&D7rU@T+fm{`85oj4ad4GA8Q7fmF^kn~r9?<@tEVn8Nk^
zg36h?uD~khIWW%))8$IjG#p;Nq9{W_LXuL5n95^Hq{<C<5BEGhJ`$D%4Rm!SYelu@
zlW0v&3vV6{9FG^SQ_sy!C%Omc&?VD&&&%_R&}p4#7)L6n@wKJ$j-p_hrWI~gXd8<%
zFij)N5*UY>QVNZCOmpOiU;f1R)rNe&Qtft3=ZQ)?K8J|6>q-a3&CP+R6u<iJEg#>1
z=H=;`u4~X(d|jigBQu~SFoZBCVj0EmrzB9xthQH?uE<5cE~?P$B@6|nGD<=?(M57(
zz971lQqoM2golyW92E5hHl_a(b-xx7Uq9w*twjM@*cL?rIgrcD7xHIGU!uSxsU*He
zb(KQO;lKKR|L182`WIB#@$&M7CgC?XXzfIunF`Cikk))?;*3V)s~YVTQ<%t0By2s?
z*wbiFG?j?1rk?rw!uj|};}t4zDFs~9;%~m;GG)5TlExR={iKC&zyF?YyJK2r*#l~^
z<D5@dVTEs7Y*V9c%eTM!EjcM3fBXZ>a;6W7oX<Eb4@L&&xhIEFY$nqQXFX|LXxo<a
z^CwK#VN&Ax>4h8%qit#IGsE#2Zxom7i5wE!-Hxz!pjBNF$BFZJfnA5|T2xjHr=DuF
z#kMV{7r`PAaUo4JhAioTN}&)aNaY;`&C~n$^p`VL)3V#|>5nIV{P`W;S?&%UVID{^
zaCdu4%!OeTrLH7<%Hek%2R^?4nelp+$YNWu+wZ82qN^%QRWW~@c^W6KmnY)9rQO}4
z(+(FS)8jMe=O=D|{ae`HQ_vDOs|1}H#-7-pn8$%(UMR*;wB<Sn#yK;NGpBPejE5*e
zIP)}9lxEXyP{vYpVu@noLaiKwbu9qm3P>S=vMB2)t8GezfSx5r+3$Wu6pEP8710@3
z>oHcNkq)t>Kq?wz)JlvK-G)|*q|{f@X3-ws?!=S7!gfrFG?gW~4W_*neD>u9XDng5
zvVFMa-+l8f|7sA1!}a9@;nN>^KJBQQ4ojp>1AqO^_vli%Jbi`|$!=mCP7K!*uAOni
zOxtdlE@ze?1yl?(#yIX@f5-K4<oR@_-gGNtz|wXto4O+B%y2p4D_NC<Wv1^(Y~4_I
zd$u>PScWT-SolRWnNi9RNf@Ijsfd0;387G{_P<amuM>pwmkva!f?ZM0lr=9v7f?dt
zlNExj_e8KJ)Iq&wOXMOycd^JWtIbo(=Qauz5e7xhP~@_57M6Y{%o8Og>dw<_Hz;3G
z(C7lw`NDj?N=8CegA%|CC8F|z(<{KJB>DK$Byc5VL3xs%0V9`qlu6jU1TL3>&!3)n
zd449x8E@8!u9E3r`MR2sRf9EuAvS8G(Z&#hJa<jyshfsRA3kx|>=|>x6h*t~*vxy{
zt|Jcv(==dpjm{BQ0yfPYA3t%OW>kK~e%qoITrOv}-G)sR$J62Np1W6fpfxFq&6ZJm
z(>aQcz@m70v|O)Owwn!c8f4{3#?=kib598)P3_mr5`!~_ruLk#k>j<;T1{Pf4mSth
zz59UCin_KqBb3Qu7-+keu67s;PNgr&)>OV?7-p>24A-797E&?%zkm7%{=?t?_jK(B
ze*Yc%@iW6bQhSG1BL0I7MJbxLBgMk6fA?$N|MH&xe8pJBoqI*miWoEF`NH$dK#B_I
z3n@+*YOEw-eRTvV&0mP8<;$JDI-*Km(d-$OgSeYgSpTqDr%9qfRI6+t7b6ZZB~t+S
zBAQ)OL)YgdXQ6;fq_|rrm-4xCk-}YcCQ2;X>wBwKIM^5J<$v>HpxB0FDzvW&!<p;5
zA35B=p{^>jQK(#)$APA)xtz{S=L_4rTiVTrwiQqJe4RxYR*qa>P86$A(?AK4fBo0L
z2UVCumNC6=FiE3S;lqb_*!haL8SgypX3wR+FpM(FeK}tk=aIwVK;=Et66xAKw%PIN
z^O2ix-f;W3|A81IfBv8UCr<r|yVtK6=0(uP(?n5*n82$-36b;5i>R5KhU@cDz!hHO
zYRB~Q5v=3o>6v!3!PYHL<CQc9cAdvmj*u37rD!(?+!`D-EfXmQmi~(J4fe2So@Z*`
zF@`|Nk>mM7N}4=JAo1q)E2e3rSk2q-zUOouIG>JOo<DKhSz>`@S!i0JS*0T25Zly*
zapL9qL_aLpC4)U;6x`h02&F0pVt?jRGRG1iE9%P8FVfxgKfa^3mT&*nmBZU_iT#nf
z-bwWtM)Gtf`weqO)+yH`%Xr0G%l@$EI*we23FQQq6hr3m@tMPo9Qgfp0Ie9O1?wW#
z7+G87SuAk|mPG(W&RCXZ7U+@^fmw)2KshPO@IV>rrjikLSjeT|bR*<EV`c27SrZ<$
zFx8DgBjiJ^6=k&(q724YtANpgb`&dC%N%DknKWN0d4@PLU7neT5yD8RH>8qD<AfRm
zd6~&!AY7k`6zUs`sU6p+Cyu8xZqqPadupQyF*0fhA!Gc3fTgW#9zTEJ<NJ5q+`Z=R
z;d@j`<TP;^dW>`2+-&iids57#F!A~EGs_g1mkH|>d$q#?KmPm+N;zJ=Jz#U<^74`G
zzN7B;>($P>n^)`MQdXV5bVfPNq-egnHcCp;wUoR<i$pz7DYzB>#L6NNA?IHThc6i6
zlom=!;>k9(2({KflT#$81#9XxidsI`7fYnXA<I71dqd?c<1}%&lT^5(9iW(pk?DM5
zx?V*H*mNkRs2gG8=TNwwPk0N?I~kit5hJIpQ8=hNau&PZJT6TAh&5hz$Xe62HFx_=
z9uk+5)&qLgOMJCWinu{W#A~+3TI{9}aa!@?kMH>1>kUtzK5_faTdecUb6~&SVYQXN
zQ>54^dBLM-ZN+(<nMZM>oQ_9QOnmdrEm{qP9Qp27-?6*71?SKOl+Jv5e3a)FxVgJ!
zf4e8nGw0Kphu5#Eyk)pRiVMp)b2^{F)Lf1)ynT4f`FKQY&$jD$Ih{BU3*LGjZZ^ze
zA*I51-+jyDhmUxt@ZNGAMs%?>DRS6%_{#I~^9#f0E6#azPVDN6&TCBB^8TggI0XLm
z;~)9I{eS-_{;&VT|Ap=Gg_m;Sx?IsEvDqIu?4?K*9h=72j;!FjUw@CvnagnI`Q?f2
z{vff!X|+$Dt{l9^>Xwow5jYz~S>al`h-*r%>#&^48sYs#thbKp%Q~uGiJ0;qE$`1-
z5Pq@Fe(CI0StG&LqClxa$e9=i(FCE<IS`T)s;G5e*AS&-($^aNFVEH2_wZi~N30Cv
zFoA8@+_V@zNiKStMbK2BMtty_{ejKxorq*|!Ju$e$1pDhlUe$i{`ryL{F{HnlxOm!
zSjGXG23^(kPe;aaq25dkeNWppEG4r{ktGyf&hIE%v$?rtyWKOV%x>E;jswLgD&v{v
z5$8JU&4$N6{XKQFW3v&a_YxdSoN2aul=0ZgqqSnEggE!^&+mi-zl`{{rP(wr<3d~l
z*=LqvW<+uY+WnSgUdYQtaT-*GcNIY?%-nN2UFnY}?j9b<#!DotvGmsq^<ht0L|IOW
zR8_T#m!RydI0w2NZl5S|B+Nb9INpEwM5`0qO~;fn+x;GEEK^t*LZ-x-bh%<zNKsfK
z;}Yb+jf-@teUEQzvZ<-N4JBtj{_u{MWnlksPq)9NpFZ(;zVc81_x}gAcf5Xk%isJD
z|3{|F3yrIokB=<jJ&UfnUXDz|h5q>yl_|Ws-?2}7{`24eiD`^fb<Js*IQ4;UFN*(Z
zmY2<L8bR4AL)H$J1Q3OBE5^2fkr8l;d6nU2R8B-B4k(6*bG2Ohz_JJ+$2t#5(8<P$
zGFT#GlcX9d#PbY<c@Zm7l6c`fk7(nFN@JWQEu-wizu4!LC6}H!_Vg(+jWb>ars0V~
zpl&*rWk#*m&!j720)0I4>FJS6-xD%yH0-;MafwWbkh0$GP{8Bo4>Y?yxfJf-e#gz-
zYnG{Jx*jP?GYnS<3%3ssXuoH%8~S13<<IX(X=2GR&l9`thPG+QO7@`7Pan}QABf`>
zb9YBo@4u2Bb6y32>tQJNmqIBHN?4166$7n{BtD9>TRNK(R{Nd7xr&?v^K=!;q3htQ
z3n8wcA%*aUwd^~U1o)^@NU1<skPu6;C@NLRYe&TT^D9a?3*$UfHIBMzi8-(8@632P
zF<sA;0{gpLd|eZl1w#a@AZLuyq-i2egYX7Qh7>5Omtqi2F#se%+rAANg?A3^3|Nh;
zJ^R~RimlN0h<O~jJk3n~g}O;tR}25c8G$0jNKs!dh%f1P0iy~b1xm48uLI-IK#WXF
zpxW#h`-!Xp>nKIhhrsE>NB;FUdv^C*#;I_-Ybhace(AY;xM7?UOPKi8?Jd5lNih(@
zhyxx!yeCW}O|#+f>Ym;04fFNF)DKi`$9}(K=+E4?ulf0>pZVeE&p2Ok92a)Cw^Yt>
z$_b}4p(M`3L_cTRw!u3?-8Ovs{8<9cjN#gk*g3G-b@<8>LgYNW5VKmRLNG5gx~fpR
zu-!Jo2-vlJIQIPGyJ!BF|JVPMczNNsKm3Yo{~2;bYemyKZeHKvoDny{aKY9MN{c3>
z+O!k{!*$^F{47}Sq@YN;<7VGcDNuP53~(_BQoDR{`K`ifnRw-|kSr0r%0Hr0qO9vo
zWma41vTAR{rYp&+veFQjN%?x83xzchqD;7ywo*{8&5S~x_gqAT`sFXmO7x4aK-T*I
z)sTUIZZYu3|NOu5@BaP&$isKvaXft{=1e~hkP0{ch9yXJXxnxut(a4wL`}|cy7Z_B
zT_-B&c6UctTc*>Iakzjk;(Z?@cQ-p8UTulv0yYu19;Y38yJMaPN>1!=-_YK^!uf_c
z_84U_CF9yHb66OzBTmCOC#H8l62=Q&Ykr<)eBFHotSm!Mj2Wv7o6Uw8BSk5^j<`)t
zKlJFT!PQ%^6JgNgIdDA=Ow+{gfB$<N3RP8n_v>GY2i_S%m<i*6sWiK8PqTZ4b_!A?
z&jZFPs>U&Xdc^HI&i#N!?2C21p{q69yL)tZk2ap^@<g5oN{kZ1q~Q7e&$u~oxVgbs
z6+spD^_HsbnCF@CbQ0)^PuxFrC|A>LZ-~9beQtI;(m0TlcvqFOy#C#9dH=%?xMoY$
zwEX(l-;ol`-8cO5=}5eO;=_mcY?i<Q#dLZ>k@@>S{Lg&+^FLA76}y9Yny=?8*Xu+W
zW=hdm=lJE_vsBetF^n@Qi%UScnk8oH#)^NtD8U`CA{(<h77B@R8fm&tcGP)A{6_KV
zXKjc{QtV=uQFT~GRNbOV!eqpbn8FIz0lCT4dTIP+?3O~3IN{YhpJKq+1y$q)TvCvA
zmZ@<rkggZ{^O?S6DqllM#Cf4;O}i5jrq-T)yJb3lVvG~YiiV}CoIGpwhTs1BH#{Ft
z#E>~Ye&X@dds=5`bz+)MOggj7l5enm{VNV{f6eL7Kaj3hl>bDfA9;NL4qGB<$z1SV
z%l_4CZV$Ii!<Fm#LS0ofl^4cBfiLN1r6kZo_p`QIPpD<pWO!1VMaQ95C3F(-Pf-#A
zRYF>Ylw}o9$w8H4UcZ+LUA5B9OXVObm@B7(tk!LU>`ZeAl)S7$N{uoWw1uKp4|rZF
zY?d%j#2k40?OS$tw-{FuLSkAbymJKWa81qO)q{j<%{_7IMJHe_);Lm*#A&1iVG#&)
zNhW2^c&HRbfBBvjjIZbpH!C()QZujd!k;fk##xF~V+vn7H|srdk*MdClJt&xWpAv>
zgb~ZnA79vSYVO8~rrXlgEr0*>f1+rOa+ZkZI?jZ+h;}DTY@DL)JjZLITXkDGN}TrP
z($k+$M5ib*F~)@w7P^}o?(QCFHk~l>`ztxa?VE3y`ZIPJP+D>9BgJj__3yrC{)Zp9
zyV-!6+3fa=!;CeKrrpply#%&-Ph28admf)ojMGAx7q(rEoy7TLt(H17DNg;!rfb&R
zh9nc)lvtJtr{UYXj#v^b0X05y{kRacSoA7aadW$8x8F0(6IES_jTp^5&cYtZiR*RX
za=v1X$5)w{72`0|7)$L9?WRVX^i|IySumnyDmg(eYYv9~s&x^Bxl8~sc8zkE6|Gv+
zUr}G)11S(tqP;+^RN1_yCn~)vWmiGrT0CeY&I1WG&0kjtUu?Wmkci$u(B7<TjQ{%k
z{hyaj@ZGPN<HE;3|B0rSeKus;ZOiHOh|Y=acF!{Q<ecgDTb`ysj*-jD6Sc1?DU*gC
z<y-3QRxD-KaXrsGKfQ1`9GET_KK%Ft&z~Rp{&&CU?eG3I)n<>XJ;%?VdH>T7m}+LP
z4TJ@gW)54+GLD34rd&s2fjI@9&%fX+%U}J?-=ekQ`1qM6!#G~~=JhMIR!rkS97jHV
z`b@QY-*ZW9w_CKY7{(bqyl!i9D43?<=Kh{xUPx9mmjv?!XuLJpEMp?O>9`)xH1`kG
zO-JZopcKZipeMu2^O2mOYb)+vy_T+KoY2N`{`?6&KI88kQ|_6U1+{YTvm4Py^XgZ>
zff&%0*q5w=Ww>HW##>9O9K|T+v8T3<MQLW+;vZh)a$@ThdOVY+D;fz6$|dvqn{V0O
zy#ni*PR~HktG(y-tKXyeD`J0P8m@%?h2ioH&hh5;Ek0j(d_MB_`(IOa2R=W(@DG3Y
z_uTpe|J~pInr&P2{-@78K7OXIV3`KA(Nv9>i@-T_<xs|=s)ow6s2s3%y%cgrS;I0<
zsA&QcWs!8oSTfp4x*Zgu^f`xFg+DopfNNPMtX9OZV4WxB^p(*rwnP!>efcwKnI)Y}
zTeMx3@N1%;TKNR13auCB@k-luU|XhTV7|=c5YW~#mn%z*($x<$r>7UR0cFL8H%+sk
zm!}v0_#Zz|B3NxHVd2~R8<w%hSWUIPV+@g4EbVsBY3iB!k?ZlqbbTTG@`>BKTb!#%
zMYG=>_*ZY=5_2HT3;pR#Ntw24n8M6<w-pUWUYM62C~HJ<Mrn^xUdEBEVMOciD>TM6
zpp9s40$?oGcWcIgjQw-TgfNS<#Z=^wg(6szKw*GX;yR@fZ#Bx0Vj*QIOwuxvmL69%
zUw!{tdnj4<?Ozh%N<k?J336?PZ5sOP$Z)-)wM;IIcN|{5qUt)Fwam+vX_+L;Kx?$s
zxXMwom&PMTQcS41Q0zk19@Q0!Ye-52r^;x2(=jiB-C<9fBikmRa~4`+b|R?MW~GBl
z@u^^K;WG0)f1MI2GX1ddcwAW$Q_psP!$1Dv4~&;HH-{ajWHK;*dM1&1858E?GvB>^
z#l~wU9T<xyN0?1UX_%*xr;qRH+KR?|&gUbq-n`-F{vPWraT>TDk7%XXzj=f86`z0k
z0dFmj$16X+d*t8$@Ba_%HV*T{FHF<IU;WivjxR@YKC#3^-L!<~3&uF+c@{p0R@Ajn
zXcY)pOq*42l(1T3lw&et>GakSLSUW(ZSAR=4ZnQ+Og5JL%?_0#-L@sDj7plrrono{
zwyW`VMbV1$@kG<K7+csL4ovgJF!ZDpdH?Z=lntA%qup%L*3ot?&P%|R_q8bfm01f`
znFPvv?^Z#fMk~FBoqf4+)qnBpx>#HF1i1*Qx8&7oNI_ZTY{^O95iNE@nHE@_YspE9
z1e8R^C*rC*NMD7v!1{Hr)4U|>WqH5;g){?=3N&^?V~B$y&ON?%5-JlHVpurEKtbcH
zmZz7Vvc07;TKL~>3vuSt=SRw^M^!D6h;LdRzJJB-+nJOCm*;2tR8jxtZ{hxqMCQ8m
z)XvinlW@z&k^b|0wp~jYj?Dd;w%fBT1J6T`Qi?;@V2xta*5v-mb(mPj9<2+n4?AiD
zVVO{2f*fh;ie*VCBu`ptp|V|%Cw9K#>GLOW4VAHsFE7l~OdliUOj}#J!=AJRQIu-M
zFiZsNQK`Vw3&)@T#AesgZMF=j3;prR!>_)jHl8pC(vonM!+V1}?5J)Z2*qIgEEYd&
zL}_1FjMGBfZP8?!wq=?|!tt4D8K@6;sLd8r&4hm7_GTxCQC4K_nX4_%)}%O6e8IVf
zIVGyP!RCUmp4c32q2AJMZ&;o`V22m3pMFH+2+JUchfcVvk_&L-xoLjIcfb8T_3c~g
zo7e1qZFu<Iza`ZzwQVTF6R&Q6VEovTv?EmwpFcfg<3e54v|UH#E1WYpU7_=WE|xUU
z$Y2cmaU?Gb%MxW95QC5~RUx6)SnF9tfKhZTr9d%8D)y2kp;&7&iI@T*%o5F7q<hdx
z69g@UwT^iV7*~;_u={nA7fh4z&J*W|F^b*(z|%Ew|L__;ywKDg6hmHSy6uLhsX6D&
zJT26l4wW<8!yc_QF-XGW^)%sK&F0~brD(P(;GE@n9oST!y;pqt@Mq?6!FC-tH+P)J
znVVn#70dFCU}&`_#|4{4=I0CVrwea>^*e@`C<K<2=%<O{dZw+Ej3k$tVR%L<gLkcP
z7``|u)Y=s(EhMxY&|2e^MVks$Bu={+N6cTe*81x%HKiHt47S-4iz0@RGz};uqF6$U
zSa~7pYbX+Y5%P(cQ>(u?MwAkoozm(rhjA$s49+=$=+rekU(<FC*Yla-bY`>LvEA>{
zMloED%$E~+5!7-n3wex;S1GJ@5$a-?q<BF^ayrshJFM3fG}buKN|O6lCyLT~#byu8
zqkL@Z8)2zvSw|@IWydT6T2@OffKgwpwGBW1@DpzzwxqCdK0fpM)h%J3x!rGg_p_Cb
ze1xB$&cvnWci$-*@7Ql(NSTkviOY1Ns%z$<N8PuCljbP~eB)WBndeU*F~%`m&ls(F
z{i|P7?{~cW(~k_tBkukc|M-u;@amgy`Tf878y?^Ngma$0zp&ZuNx6`d=zbPtLJ(5|
zyS`7_(0Gl;6Duh~rfG)aX=+F1l_1+wq(6-`zLsY#M}~gF)t*h$5aNv0nY!|rT&R>|
z&IRid4|h8Z;JhW3!Z1!Oi^PeaE<HDQHyEu6>v{RdA3t#!Gp}9EtA{(f?S^*K(R2-F
z%_b@TrL(2T8(-G7)R&ahQj!!3(uxaYODXwxS=bpMc0!cqOB8V_D-(q;o&L(XDMgB1
z<pfq%R;R}zu05lTtOM3bgfNB*q}8oM`La3tFV=kj0>wbQ9(f8g+SlCO-r!9mHp85l
zm%!<A6|8X85_Mo2Pwb2#QMkLiVNs6b^T_M}?%$)Umb%+;`tuLe3buy>D91Dn_?rjX
zSKqUL{VQtMaQ^rso=EDir1J|lMegpu<>l!kX*$ziFWkO;gV`Ri+4AN)Ay|zsFYI?U
zjTKnW_2rD#3fnfM7#aF2fBxY|tTt>mEw}ggOd*k#VUB?h?>=G_Z1!8qGNS4|=ckY8
z!=5Q;a2kywERn1nC5fNi)Seh5jnCJfN+qg8he;1OXFyw0oq6^9x5Udc*V8Aec8Bf!
zI$BP24-bszk*fCWzW)v9<1;!1`pb#BuGrt+;oB|aykLEWRz0UrPizmb(VH8L^R%Ys
zvMgkkz(D`_$o%0mYWG03y`k<J+S-#$OJyw@;iv2QLVvliKfJ=<yh8hy`SgLEmxxtM
zD@k|LmHN<$5NCzI`2Dw(%>&7`<f3V+JB*fssTe^pH@@Y=yPw$@&HmeamN^in2{)e5
zWg#coKQ>;mgn_>A$<qMF5|sh#(Z<tmb|_t#`yQhe%M=NzmrR5(G7dd9!JHz^W-AuV
zdBm?#q~koJl_a3Wu;7iNs%nfA2oYLQw;SXhm@t)No<{1nlVrS{)?IlbD}}G`*z1Nk
zo*B;rrmAsjVXsHA8!D(a4@{k*U#?8!%rs4mmop^<4*LW9n;qjkF<wUe+wXC{CZ0Y~
zo6Ojs(7`Z=fUfrJUwwnK7S&IDe0;=K4V#-=Dg(xPe*Wo?93H+!m%`MasOp9sC#*_r
z_8a21V;m>?>p)jo;=JIzmw+V<Majyhq-6!BXqIUv&XYiKsu^v)9FA+T6~jyli(~{y
z>q|<J%5@mO!^A|I2jVgdjnd0L7c@p`v{Hh{Ee4~4IAM||A%|ee3ngll@hBw;i$#gR
zD;EJC-96k$Eb)5TF3U_^*OYaIr;R2I7t-|wm6nzL0Xbz%QG{{eIuCfOfh4QL80Fxe
zGa>y%{qTldZ7@~}K3x>vTC4@}MHITOWz6e^k^)*cUj>tDrDz&s@xF*72LahoL~Q(Y
zys&pQMmt_Uf9BA&G?nMD-!fh=JbiwoYdca53}Hc8N6IqQ>a>M$C4?DgESKvjAzLNT
zdBxqXW10eg`r}XRwmXJtW}XKA?Z5k5s=8);Ir7sV{(-}8$De-qnGh2H{r~vyXle&x
z;cma<)6^59OgZ+qH*8Yj{NsBXFKK|vYT;(2#G7y4@J~O!V+j+Ru4Ne)be6&;<cLBp
z@q#9X!jfjHO7`%l>lIU=t{s)PsI0gS6W4LUc+ac0rE5IC(v%dayN0_rZ~6JhpLz9q
z$A|Zycs?EZtH1gcRvEU9=Vrg-*k^`WCM?cdaV@Qvb4qaqUO943Qrwg;c0wph91&mW
znPqjhh=xGYI!pNrMYF6U{G#Nn5VnQ@!@fvg$bw$hs|G=~b~3HaDMJ+Cn>J1m?Yai5
zA<mF9ne-I}E|K+!5c<DzSMbj&28MCu&EbY>v!iImGR}gSO^NrPK1*toH)vnsHydJ#
z+}-XOFGoKA^aq}gSN3<Wm~z1`0gf+}6c|dT3<Cwnryu^Fhua%MKQe?!)S78J<GPNy
zAE?`!kpyX>sVphZ<dQjkd?LMn<o!>-@cRB1ziY6@QEj&9(9>;Os!hi@juJ&#H^i8@
z|Nb{r-VvvP$LD97wxjDhVpcT!y^Mm47Tm3?@U<uBNM#gT-}3SEXBh`tK|xlV9mN`y
zS!is9QJON3sK&F|?WE$bDuL{5?s%C84262z;Ty-)_q6Q>jpPit?SW|=IlliHQ!=zS
zG~fIUbyG8Z{3nLT_vo9~sA|vd+ut#rK4P9e(x*(b+2Na-=f}@1!-ytR`GRqg+9gWp
zS@81x$boshqP?N^7Qb!K=g-u24TlyVYwQxy+Zxr~k(Zg{=a1sqE(N{0fvQDUccKr-
z32WtWSK8szNV1BlyG1$2ue`_1M{=C$b7qN!wow$D$+c(dFNAp_jAv9oaJSo&Y{gV8
z?cHl`zx^Js4PF=C{qgVk<^4zIY?-H#=l8#mhJiWEa>4tGq8|m^Vl>_vG#PCQG39k4
z;h0jS=`u~%qLgBqXO@uhRZUS)8$}8;{WW5JjWvdGzGAdQV=u}P&I8#QiMP%JyD~99
z{>1s`z|w7)t1Y^|XBc~`!<NwZgkeBw`1!*-?p{66xSG$8pD`uU?e0kDCzi_#&KgdO
z=GEW)2XwLg{NsNDRb=Ep_1xavaC!R3)9EwKW>4!Z54R6!WBB>!A8DE`{{CC~PwyZ^
z8t=)@<EjQ@z<MDzrYw2vN{jZQ#3+PHEwoV~n<?gbMlB0yFvg?wLSa?Mm%<keQ7%e~
z;}zDFQz3=OGEI1=scJ6@c_SlI^4k3u$rh+8OI|NkrIZNz@=RWJA<C`}g;jWEjM%K0
zC*nLaozDV~G6m-iR>LwasFYAqQUsw8<067?YhY2#qgZ`qEGq)gm@_#%vdk0p-CL5I
zDZV9y5o1McHpR^I=|oOZBGZlRhhs_@EBcU}k}wCpFt9<+8LhF}vMhlVX6(0HPRA25
z25#<dY1@WnoS5f@w(Hp6EX?zSohGhB;Fre>-ddctyxvt9?|FBe=!@p@6qthIP%A>d
zQdI_HTK@FsKY?2K5C7qRN44pg#(__N{+UhJ^7!)1vSfbu``_~V+t(BlyKT#3Wmsb3
z>E+Db?T-C^&onLk<Ij)8RPfHy)Ru8v@Lfx@+wkh!xBTP(?>}>Q*rAffYt7^1Gul~N
zUvuaxp3mYYNLrz_a3<zC;H{+}1G!{2O-l%gL}Hm|UccItQeGVy6UG=`y?I3}&>v6S
z-rn%^^1?s-;ZJ<~<^f|gH-|0SXL5vb3QW_?cE4H0ohy>r*K6GUO5M;aX&@CsoJn~R
z(W&u5)-)0dl(T3<&_*IU$m?A5zZCYFFBQQ0Tq784zDybP>Q-3oje^{yV8tC{Yta#e
zk({DTC{iND^c7OHT6+Gg@ArQ;flxT!iOccAaJ}HV9o~Bmhg<Pud&{^)O3ZBT-k{n8
zRyi)8pXqMyF;$HsVad!Yp_-bi6iz?=LiUcXZh3t7j;d*xe9KT-=F6343_Lx4X1m|9
zySu>{&3=E&<L7szaX{$><vqJyO-Qpqjx=mH6(qr!Ds8ALu`StE1)Xx_<<yg=2}@?%
zHe?MUjyQu@qEe<z3o1rx@2Q+;PCZ5`?!Nn$;n<@nRENEE$i<?4{WZ6olrbTJHq2qB
za$*%@d0|@@()J#S#j2?`o<b&%fwzDCd+Oai*I)j~!}q`9csx<<AJ8P`7YRg(+3<9^
zvUzhuafYzW90!3M>9HpcnX2hXSSsuBmBnxGaMm#o0w;-6FTpfnX1I>j+YL9hX9-sx
zfB64U9UiEgn)4Fy^&L&KBaZLDdRnUqQ^IJ&5@yQyLTL`7ps(bCHOqtwJ!KvR3S!Kf
zisw;P%XS0ddXj@zTeNRQ^&S_dQ^Pg9kbFyhIM7xd%XMV@^q#nBM0mCY_S+7B|2;}u
zey%q>e)`01SMm7n7k>Qkk(4usuBEb>t~HPnzLMahWQ3P*j3H~wbzFo@SPbXmnRX+)
z>3N(mDq&qk7-y^!YgS!Vpe?Sog#By6JaKwHGMrCXE6;=V9a&dY&6e@_k;hM;DZeN-
zn~t__=?-_y<IKmGGcFxDFM)bkNYhW498t#7-QIJ(oEgq9wAPUOC(xD|c>Dct_~v(i
z!|CxA%kd*BPU!i<kN@d^K{s1Ie}3d9z^gZJ`R?soC~Ha{8Q(S-8>yQP6e7SZiWnm)
z34A4IfuLw@a8)fZqV?YwWCS`-@*Y^{B#ya&a~_MhGp@&%)iRoxmU-=}oY)18LdAkB
zlD3&OGGQnJSJKvyqO1{eUI=j^gqda4e(2SkU6R3Kake6cNJ(kcHp$wr={BOU&WUB3
z$g>nPF)Tuw(~_}aT}4rlmziY_qWxe^$($Fi=YfzNZkw6Sf-#P&syMwI`SkRHvX&TT
z=4p|&sn&Fxj=B-lZ_4>=ylqwzl!cOlBcy;<Qv6Iy;vawb!0o=`?KiJjrio>lp%ixe
z9caVYk6gx{q70wU6MyyAu-R7BMZ<Yvm=&>jUZ%p!HB#_QF%ZLrW&X_T!-jwNx4)yU
z9cfwk^!&)spLl+`qO_xHY7RG@{Op#pvi37EWyX-$9uBnIhB-!tB@(k@({=22o__eu
z!<*N{wR8UFn{W8|@jcU&x!nrUc3c*`3Ur$Wt#`x({dpi^Xe%dEDB~DYU|u4+%xpIs
z#$}RqU&+|2A|%L{9v2Ha6~1|O&rjpPFbJRjd>-igceHIy$cA~$Z1&Bn2VfNj3xw&*
z1mH_ot=3GAw2D=~blieKk+fXB5+n=ZYcX6_Om<m6mJ}8e-K~xO%GeNK*9zj&4rLq~
zhmH<IkOWl}+BO*5%4DfzK_V+H&r^<T@RwVi8DB(&{{qE;GFEoQDUeFS?{Dz!21;gA
zdCr$B+uZ@J454I(^N~43iCAn7>^CjNIbzOy{^b`o?T-DcJ*Urqrqux`j8W5ddz!l&
z*t{kz8GEl#^Oa3kvpYQC+AZz2qi$<#+j6enp^O1<xqb5u$IF#ZpFeSHAsh=+QP@^U
zWRfvs$p}&IaJZwcDy~nTAq9qMrsyPqmdlm<yIYdcn7RR1GfyK*TiU5XyN-H)OR0NQ
zyb^{<()>0RWtmZ`;OmO1A4s~8rUmAO<IlffD~D@4ly!JrFwHG#b3<nf`ScN$7wkL{
zULIw4t__=R%jx*Y<>eEb?Us7Cr+CkBxiWqDg>md@n+n$)NFfp9!pqYWhx>bWolHf-
zQrH+xo-Rz!pQ!2<zulv(;X2M>G*f@!hrc^f9S($PVVU3ac{sAnfo1t8Hn+D}m8ka*
zsJca)o|pq_8BxOvx;v0^M*9j|w@fb|7?r|zdzK<B@LFloybvuc+Tr~M!Yr$b5-2Eq
zx5d;O%<U^`lDww#L{}H)dB7M?ez@mGTW(TM!ZP#||K_iMjsNfe4c=S+@E`vZ|LcGH
z1Ly0NX_P%RQ^Xd<yd<p6__|>pBmMbGy^*L_RRV2YF<yGYybz1*<&*~R99ApxGGQyr
zc7KES8^+@Yn#~TU6t}g<vc(wBG8Cr1M^z2?55J|WKVWsH(T-#!1#OCj-~OBb4VRC<
zz;x!<_Ydq2-_c)RxP1PJ5@zbAVH_@89)IEU5UHCz&U)&b8-{6OdU}ty2DQJXgo)uf
zk$mR$>$j+sxP1DVUq1brYJWqszvFbeu)DdzA}Mt-%$OC=Th#|bSeTc~%1oE;-)Lbj
zSXYa{%7|_ttr-O&Or$maF~vYJa^aPnh|40j#LViWPB>Q+f&^q~4P_BaVpfW<jH_q6
z!Z+eRC^->Ua8g(XhU<CtRZEv;RE^eKmSvF@&Ke=C3T$r=Y`0t4(dUG9ju6BNp^Yf)
zbBxUMB)_jU)_VyJ3V|4tgwhy`?>emagq*0`iayBYcs-9?`bkpmRH39qhzm(6@Sb^I
zsJxIdzly*R4U1B&#g0I2t+70v2FBsa>o<SH<#Hv4gi)HX1a^lThG|9{&3?Dxd>)w7
z8d+RhJemjR$l8$~7v>beXpY0g<1jE!J$L(tO<Qq2U-;wo#G5y-h%xf{=@C^lm4kkn
z>BbEzid$^#uQ+3gF-s$r;#Zi`<I9C5%H>)&HM{+ms`Bh^Z<)tL9wr`Mz2QIo=RYzo
zk+*kSZf|bLArOLyc-yw-ptlT{9%l-hwi7S@T)4zSE-*|z-L_*+iKgoqr-f>>XS|Ml
z{`k!9=7w~bxxKsL>FG(P1~Kr*A3w6)ZrSekIA2RrrYg+KOwN%IicG|m7Pm)<q&56Z
z?5+lDJxXg*RO{YfeFc%p$C$)5V86`&SAf<!+0rZ4+!zTy%2~*PDXyytYcQ@xX^Slo
zmxW~>iMgOnjl!@D1C)rh@{DJtaMoe0m3s~$ul_~U_}{n~m}27R51-lX544+GcCX$t
z^;g+*FB7ioIK26m<I4-<@rg1`n0aC@aCiF#`}%ufVjo|a=E&}C$B+}%{uVP|xxAd2
zjibGLh1%WYlO~TxHtmKbOqjO9s(^`S@?~Heq-tz82acCBp+BLOVhWkI*^pA;7z*Au
zDCNm9GxQhgZG$xy9Tt?*Ov9P7j8HYiR0z{dwcSz7%oGyNIMStOoS$&civ`3t`28)#
zw^a8XOL*Y&^qK1@;20McN*pLC!ZeUW3&1!GIOp)rU~*)8^M>m7Te9<#rWOV!47pfB
zj+i{5UClC$l7TK}YwEhi_y&hzYaC^n@v0`FM6;o0`gx%~JkZ>}!7IsU*IJRs3n47T
z`VMz^Lr4S5<%Bkp)nI_xir~=snylgUX(r5ps%e?}6Wh%e9Y)6S0?MEog>P$^C#u>}
zv|}31q$yH+Nj0pTqV!jo0$Z19H@8r=oaTjTUWnFw%{A9?M(0S~?j@Nnh`q?<C;*|5
zD9SRNj_4qc0^fLCTNuud%wZro&-Zr+?*4~=&yq8mOxB7eCi>%<{?c=LInv$Vb3I)+
z&l9JZXf}>%9=N-4U>nXKkG%W&BQy<p8M$pLn%3~J-*W!&!neQKv%T98w4-Zpuyw`d
z=?PP{Z0}z&of~epXa4S=e&%vIbF<sByMJK0jD-HgIG^bbx19g<pJAEt#&G%h7u0#g
zZ(GumY1<9En>)f1C?&GnDf;nDX*#aOavqMT$B&$cMAtbEha2Ah?f;3x`I)Dm{t;7G
zY;JG3oO>!`F)BeEc>d+g<@ii{qv(&H<)W`P)P6@*)%4dF@;t2B4u~x-MbHsLM5lyn
z9ZD;#uZb}TKt(z6d&fvj@+@hs$l0#@RKqx30gdh&0r~)84pJC6C!*SUKo=Q(N+f~U
zxN?#aZ%&F564{7;JLL(b9nSP}h))aW&rj4%P197Imx(xxxVn+4l?}vePtl;QAuM7i
z%rUICF407#EZTu-oUy*eYUs~b{DX*h=V@lnLRzS-XP%CD@9~xV-4X+79x-eBrnHNh
zlrka5)!u9<N@4VBXG@6>PZz#@z2$tplH<g$zWIh20{fd=G6r2WY;JC_P7^|*pBJVe
z#gDBjT&<Xw%+oniIm>mPNky^bgs}!+)qHxH_|Jd%f&cBl{hq2ae15+2@#7P3-`uh2
zAd}$dn)&{Q@p2?BfzO|w0maKOlVW0-qpUANTBj=>TerOZezWo-;JfeM5<?(99{Kg#
z2ZkkZx(q0#Xlh4F8Sg4$TDaLZ=$J4_;o~dMu4}O-F-5uLtyTh&3W@uhE!WEl1q{=|
z<3Ih#?r`9D|MvInHXHu%cYnh8mc!w|^UD>jYrc7Npz)r{ulXZIx^qHa;aGz570!vc
zT-18WYU?p#o)z2ivf4Xylrvhd@!wg%Kt&4`&^QU;(n_xdLKIP~24y7DJLio3|I_xT
zJ(6YXnI*RNct$hxF(NWDGVjo26<I8nAgCok`ul<)Ku}AnR#kTw$vZHon0&C|OncUc
zZ8xEhB*02!0(p@@+;9)`bM}7U=UEGZ)TWwba{`}`ByyN2DUtnLSskLGYA-BO>#Mvj
zBdJY-oa4WEQSh%i41E0Z%6D(?*_N%2A)Y?}ssLTL2vKr+p6P~LauS%E3^8Emnak5B
zPG5hfF%8RLv3EVEpMR$92Rbb=M$+$Z&@rN8!hik3moJ|Qe&g=$j;3iahda#Swo=v3
zPi*rEVd{s@$5ipP6u$fJ1Lv1#(mb)HKy(`-1=?ZYF!rpk6R-1yactSYd53B{!s}P0
zF$}vN9|Eeq<v1rC8mrsN&+jeM@d%~R4!7iwe?WEzRypFwH>BHdNQc5EAbvX2NrP4r
zEkN1k8b0f=T2+U9)3bm47I}Dw&<YtH?xlvfVx*=>OF^R0O^avaaC={s!`*P*NrLWs
zf>I=<k;b5#7INfv-(&CIk%UE;is-dXy_LAriI4@C*CVEEG^#LKi<OEPQ>}rPMDPyX
z4!rvx{v$bT>~0QpV?#VoT%KRAYp)F3gwzAMv25E++xEz~B8#sn>r(Jn+n-dT%2vE*
zCj{wq!KH%edb-^Wab3w&E`AOU<=5-amZY;Ck%E*4-R?*^Qnu<FSWaIdINHNKl-k7?
zM$qkBMqzL%vABR08FM3fczmSkdQ{htg(0mAAw(RK5DP7m_56kWL7;AaB1FOY{7MKL
z?O;j$z}Os+X5=z$6kSkl3;mAK3{1x_oL-*j8_Q`4{N}?qT#_I-k9(cycRfbk&|SlJ
z9T6uYAD>SYWBB>!pV&Qq;KSeiC(h?1m)C``MIP@){Ic=#a-<soF>}y`FMs++{_x8)
zfA^pM4r?rC?-&ku+&(<=^wSw_6wNRojlqY2ZCb1aZS-~LSkb<vEG+XY-LNBj$F^(;
zQD0cfw4^Ms-B9H|KqnGPnClur*Tv}dVFu!Q)2<J>loHl9?8ZBs^L6)|GX_O)j<l?(
z#?lN1r5aXuBy2Mw`Rimtqm%)uYm-}9N(sN#^75KdI^(scc+ctc7qWA7M&eH=Y};TO
zi?*6HReNZ34%=FiTals=gS&b#Y7U{!kR>JOYrAA)S_A!q!JRLx@l4YX6+bTvo8J&X
z)3)T->DN^Ewk&AVg4T6%ChEaPDNRhCkm9u>b{#QaCkMx~<L93~@qhUr{~mIt8%CP8
z$7ex*xTTv`luX>l!t={BUr#G<A4VQ_l2Yo;^ZwQ_BiT*`Eix)MkTmDn@$vJKK`UNP
zE3d~hUtTVt43{6DF*))6cI55b!prAh=vu?^rII$kdHaT+US26z=ES<J?D`&^GDa(!
zagS{~nzrG5K9N)4{fBSSQqo#c-6Kl!`gEi<4Tp!8ZS&Z+PJx;tNxnV<o4#XO7e>>d
zy`*m~L*Eib;mgaB$GaV(IGm64jpcIQ_{Tr|%*(v-?b}DBP`o}*Og>OhI3GzlaX54|
zy&<Q7+g5UtSAmp5+YY03T?t64|3s845$^gizt#xF_0Ll4rdm+1Igt<+LRf2VL^qTo
zuhOmB9>~E}u7?mvuJyr`@P0*=DuW8!R$*}^lfneqqotw9cx7Fv$~DoJ7*|qs2ppuy
zy1ZU92>-^10enEji0mx7Z$JpV4}>h}_qQY|IlOtty3S-#@JSGTqWHq__&_d&*RL<6
zw#T`dU1Qj!<h*)J(6m;fh_uopgyQZuAINr~%xAn?Ie&aY9XGmh&pdhN*BQ5bqA`Xn
z1b2slH+OraztFefeT7DNB813xKC$?~{{Eg_r_d$R_8tB08}@_ca+z?sV5}whi8r@9
zl3&437!=!SA#Mp}G(I>QEy-@d=YU*hifL$VhoI2)4JOsmVWALXqU*0}Vx<V7vJ9rr
zf5aTWF!uY}1@|ka(I_LYHo8K;yQ%xlQh_b@`jC~S_E&?i(}EZSMk`V-71&b}HY{XO
z)v9CHA+%wgXZ&(z)Ru=qk+H~YrH@!s#YR?(+U@Nv5ljE@9pU(iUW%Fz>30;N@u51k
zjkUPw=-cKR4!dAft&4TTo?NbpX#rU^C{v+4&*upj9OLemEo8#7qO_tt+>?d60+9l7
zndpX*m<mcO#?}%`s2E`+i*poPclTO#$daHjg!NnxZ{rSwpv{>aJ=tv(sYnPqBWcGL
zb3gJA|I`1I>9Wv0e8a=v{0IE0A)ijX`TKuQG77uD#mx&Td1Ows0)lrSU|m<Xr(f_{
z@%8H~Sr#7e?lHDt#~meZ?Ds9Id&JKHFEzui!^gn#eB^jJaes4Q929yO8Twmp-n}JE
zU)UK%|M(3*`!j#|r+>hr>Dvw;9Y6p46CXajN3|_~{XhH`QewHBDe4nq9GQ+sypKFS
z-eHX6r;ktk&ENiARr4A{_y71W2(vQmZfg?Rv=otX%L!Q$S}Rh@ND0wxtn)%j3DfjM
zAFfaxQSsqYlSl*?iArL$LP<kP?pH#Z5E^4O!??$78@Wh)$fPZ>o+g^z9@7hq9nj6l
za(-ofnrNG@G7gPF>Z|Dwfr~Y$SBj=&fpd|VCpuAKKF!coc$BNH5~VH4ZKP$w&lgIL
zY?rwngi@x&y1REVUX$I|n;fEkAB=<$nU61MA29nH6a_-${CcdPSs}PwF62@_uQ4Qy
zlxR^fvQ`#jjCJG;L?0+66LUnB3S(-dhLV})mB+{L(OR-ON4q=J=TtvpvxIA%(MUnx
zT0Wm<zTfV+6_7&W@$nXW*>KU*Gz~AO$RaK4<_IBSMnm?AfB559e2AbR<^@qSyWS9e
zCh-UU`giXL<BlJH{K&)Y9X~%$)$?&BDNggkalWu?E!NcE)9!G~w9eeVdBgenh2TB!
zzyFrkRxzKBhywfHyhlhF#;y`xDeQMUrsIW=zkEh3$>C<l;r_t9Y#dKB+vYGz5JTXn
z&nKFu<-?mDqJYeazO{(s%76K<Kk(_7BT`A8=L;bStZC@2<Zj>ca5vB&23}uZSm%k@
z*emr<5wpNiD}q`pLJr8kD7}ROuGUR5l*%!Ys(MV)wH@N!g%BJ~yF&pb`?_XN4!<n~
zH$zEC-JorcG93y*EU_w)Q>?58Wqz&VA~XU)$`xNPv?XOh+E%<<_|=veH{3S=i(l{m
zs>8tV|NcKh*sx8**p6hOP`b)2G$5Vh<<Eacio*K(f_nQFXB!^w-yy7Ed-{kq1@AVt
z<C*8DBe!qfad>!B$yl4?;ikuLo>T<3m+0k;%5d2>G!YdxmUCqH@PHWyj-P(Sh=TVC
zzigb`g{EzJ`0lrG_YN-<r_Y~A(a|?84^rWBp^PIYd*X5=#~EY|D7h%m)*#|abRI2h
zlfWy3>Go{rBckgt!w#`b$l%x!Se93^kPNr?RS~+L3CUrbp03}K5u}_Ds%9pHwpZnH
z!1=F4cSJ}{LZP&!l=^~Q&lA)6M2P`es{eeMR&?94+l_!E3tcgzxgd>UJHPPs=@-8L
z&F|PY&vHI7P3OwvR`xowH?+e@f4IZX6Tv%Dk_7KKy}r=crrz#TLA3(HgrA;SJ|9V8
zrWr<rNJQ_D(vWR~GMX)J7;RBPU$wuQd_6ViD{eVchY7|QM2Z}L`HXX(EUK;1HZ5+E
z2;E@1k<yK%l-Y7kYpamaIS^AqcOB#14bgc*Xdp%`kr*8mNs$d8NTx#qje>OwO!JIx
zT6Ej6Ue1W@p&w}P9(i4U;q}WG+Wig9u%k4EB3Jyj5MN*D+YT3@ZF+RSqp=NcI%4~!
zB5t#wYg*JeFb*T@*B`MGy2CwWi5Ulqbxfxtl7XkEBZoKNFl-(t1Kq=c!@KXWrbnrW
zl!?>FpZJIW>HkVD8Kni*wD^+gWy|0G?SEoDKBKxD_V@Sv^zjpGS{V<8U;g|@w&{p9
z5^Z}1Y5C^8;^pf{w6TnLk0_-HIpagdG!5P14T8dQ{E8q~zdao#7pCJgZaL!P%Kqko
zlqJ(?Mq7*CYlNwj!h)=<b#a9l35hipq0Fy!z^?CWEl*iu%H)`_O;;&w89-s$11y#P
zw|GqqE947}RrGC#C{>Uqu33jtKomiCk>}$pgH?3HfN3oKVMo(-#OO)O!g@R*a$P}W
z9|-Hja#`w%!!~$VCj?S!v{vL?rAZ+PRMU~92BRsZa5<isw?y)hp>1ev#D|y43_zk4
z<dg{OMzkH;HdRSms>WCuO$wEPfv8|+A!~?mw;OTJ^W%>{@%?Xp!@4;R_Xmnppd>eU
zw|sj2!ft43tmJi?dA_W?b358*U~?;0Ds~2j!P1ltXBEpLAvbszc|I?^EGtX!G@?!}
zlqkF$&$z9rP67QD!3)9&Y^VAB^@X?ZzhPS@;<~bJj-S6i^B;cyTZ~cU7_io2f+JIy
zSI;_6487&uhxaV|k?HkJV--bIs-Lxn%jHZ;kzv;%b#+JN64A278q4MM5n~#TzOE&v
zY2|<ZU;oV0*B#$H>=+x#>QXiK$->WHUQxCsq{z^Aw2kC;Z+Le<@b29McGu(l!nU3%
zN})x~HRV!pZl&Zz)3w#@l4_V#h?*y<jln9n%2$w<G%cbCQu1|;N}*{6$dTY?qF;#7
z5o5UKacY>9E4v_FFHI%ahJ@08!4Ow?6h#uFCl`=Z_DqNi+k7JDgl#&MGK5r+|MJMd
zztT;h4G-@>pk-lw`HVypjUrBG&hwF+GEbkra`W~r4>$L0Ing#9D9QBcGi(#v@r5`0
z8{Yo)U-6Ut6UGQ0Z+jj(MVL=?Ltpo1wxPtzj}K|1Z4Ju`P1n$OE$ii#u`%qvc}KRE
zr{@c&Uw&YJ-!r~>kJvqsl*KPIAw=fW5&90T6q}Er6m2(<L*VBh{+VU==)1Sj4VaQS
z>_&v}#IWIW<nG;jViHU*pD7NOd8+FTEw58Vf$)yJPSxSPEmfi-1u@nBXy5lB1<ret
zDCoAYBAb}-LQqJ=ENI$Fyz07voBh7trb`1kR{=-LSUaG*Lxoxdhd^Mp!fy+H+t|PT
zhUVs;*Uw)G(qM*>nCtCM2!#}e>2$7K1Su=dS6EU?NMo^W%j3JZ^!uA?_c}eZZjM|m
z&o39u9MDaR=|{HdOj-(Y^GG9E(-|WTjTD$}Um+U+IpVg6wuT%sS@$&iJKp~FfDF@h
z`k}!{43e*(KeNB_w2$9l9(Kr*2$vW9<txUv6sggTC2kucXH1FYv{AxFDH+x7k?qKQ
zJ`sGR>W1rqY<%}EIbC5O%|H}|%gY5JHSa%si?x=bEkb10ZDMolHB+Hkqi5Y#F7r$q
zGl#d|f^NvsfihqWTUu!5iA~nbidF*Eb(HM!=O=6wm^l%`#CP9*a}`}JOh5gIe0aw?
z>{(do8_D*|KcEgh`?kf#NOSXsP{5^(8E$F1o==x2`my8b(=S|}U%9=#=j%^DP?W9$
zqV>cw&A9o@n&9Ev?>K(`ME?4<B7I{b1&<J-x<I<2s(`%*s##ArJ#E|34n4tp+;XYO
z<|`d-I-fvoG{X+5s#`v%jCTR&J*|**+FtkJ0WB46*OPOhF+D*VN)tg=EVN%YKwz6)
z%_wxYWVe#sE7=(6M~sm4%^k^mlx)dqMU=w0-_co3nvQ5uXj_A@;QfqWX8d$vo2MG`
z6Pd6sEa!=JUQx=>c9~4=cl)_Qwu~{PRHuBKzmSlq5a{kS&9DPgcs(z4x-t`#RJ2y1
zlVoyheK40o4gsvK4|pXYasBt|EvP0(QTYi*v%0{ae||!Ld&}$Vncx4rH&@bJVi-C;
z|6FmwW7m?hW$}@(ml=}5wp~44WY5h(kPzH{_XbfS!Bslr*XK*6ZkhvUb;iZS&~+#z
zQ%WrB#xGBT=i|)nF!J4p9Zyduw2-K{@cGkcw&Tpv1yZh9Z__kT3iI?zSQo;aans6l
zS}4(@O8m7_mr_PqgEkVS6kQ%z<_RMut!)^Fo@N+0&z@yjk)lqVjMaqVdH=8{3r#GE
zKmYv1p)>dd|NP}#k;X+LqUW}2XgbS&?6|)fdAL8&_bqm}=jQg7qo~1gB@IF;Qi^EZ
z)(gmAG}$@Vtcehm6o|>8WL@lNqie`s+h5rP6%tg)*%6bd=<XOGmkO-X6`Yl?4IchN
zBfM4+rK-OsSCz3=mD*UF2Lh=DDLLY{RAe%hPRJEgY_5~1fAhmY5rVGk$iX8_ONtIw
zM=LYRDvB9tyFIsW-(tF!olnfmh0~A!L|P_n({MQ4u}&9E%G`{Wkf8XD)oo<AVPry&
zcWhf^J3e#Rb(qEwVx(;iQe+w>K?%|}*YHQyynlPoYz0~t&QZ|n4umB7P@yADM?zI%
zT4QnZ8RrALn_KoDen;yQAD>=%I2cR{G+J=`&3Ab3`TWZlUcY>)<KPs~MX*g1VK<U>
zL#rExU5oJ^-8M9)<?L6c(}k23!;MB}&-6N>Y|X_hp}54Ern9VDq7eeybXR?9!CY-{
zTGzTtqg37Vc82cu4c(jX$jTu6hI{>jpN_T5niKsn(j5-yQn)?b5M;(rXWTU7bHJ#E
zwmHxUswY7Jw#8r5yl)=gu$+%<$1k+fA=E%?C9&`8EgyN6A0-G56{>Ki>l)VVNJ$Vj
z2U;OnYpVRZ2X$^E$NJJ~nyyk0ZHv;H)QxD};O19+i0sFnxGvb~%oa1gfG!?lB%8{1
zD7pOlgOUoMEy5Uz+bA(0kaYdPGS66}DJh_&z_cA%S0ye26p3}2DN<msc;a!e^?<Y6
zV{h*$A<{K%?J}DNL1j>E=O<(eC_8XIPpE!GcRjoNdyJ86&ri(LnI>k|;}NN0vmJ35
z7!L<z2yEvI{m_wjH;^3bm!Ak(5|qNdKEvyU?!)idrf0N(?l7QV&zwL0k=*qRHv_Fv
z9QHd(j5Jy@ts8gW{f^{TF2}FP(*@IwjPKu|nwHzU#N<|7t}maq>uB2+zfH{BQhoGo
z%lULdimQxBNOm{h5Mr(`f02MhaDip<tWnV2++d8MZ`(SN5P}pl^IS)kB}0jUa^1^n
zYY8!7Y)itEuY#6z<-v=R>t_yy86TlLpo}FqZAAeOwc=%R<n|%ZTwfGr*E4@igy~XW
zPFj$qK%@jAQbHp5jTjw%U0LUa&DGnNHkNkR)f@#7QzFI!MOHGUsO>6cH0^#QDT6yr
zbe$z2Nx!^ejlz}Mf@!T`4c9UAFx2mfl;k3RZC(g{<=UrGof4IS$La&erEpp{{`Rn|
zrd45(Nao9h<N}S++}_`ytl{ZA;}*};>4I)*Zb=B(k0b4Fpzj-Ua7^b{jw+Hu)faD@
zV{1pIEl`vt#9AkPK1W_oE2(5+DD*9qazaXjn<ieKGV7)A`F!F1VL<qq`7%ND1V3{*
zA8Cgho<F^?E}rviWc;fSxNYHlK4Xnxki%byyEJ`|wuZ2-2qkeoP^9PO`ITi^YYQYL
z;+D9*+2i7oEhPG3U|s8^I%Gl0g{RAeRt=+tZa?xcHoSdvOJ@|p`(Gh>cXyBYu+_%O
zRTe}MHJ`$jd65GuM@-YzdU0@+;_G@z2!w3vi(3_P@(2NKH`Y*}tS~9nb!aVq&Cdu~
zDW@sdfNTz)5Lb`_+Z|}ydf3j<BXX@TR)j!3NEQ+<x|$;}hJ0NQ)I5eltH1CJ{*4a<
zr<X5$%z@`mUs*3FzWe=mG);p&>=4djjAnTM9-}Qq2r>b=W_J6LnRvy83bxBcTvj%>
zflwS{M6?E(1Ii5K<Y=^ExasIy!}L0HdaZ#bV>Eu9q0|$4P9Br%p2)ThTuEe4zx+^7
z&(U!hM<{|&xDu{bgfetpixRc=(I58w{%}*}LMcJsI6r;i>E(!$7C}Mf$l=}h=+<C%
zQ$4w*Oi8gir~Ssf%qY_!g+})ySqMZ<%+tiSdYapw=Kc+(X$aFqx?Eu0a7N?S4Iu<^
zm2_Qqtq66s7L)6^RZ2~E4lzwAY5DrY9}saN%S;NMm@_x`j|frt^yfdayLn)^y`kt0
zxtHv;W}9bF4dcTb;`wV0v1pAkEoED;u%DSPFQ578kN<>^ncw{GcVuNT{k{TJyk|L|
z2_bO!?yu<j9_cppdS=ZPIwDK8KE^AsMJdHF4CGkNb0rtzwt#o6F`#$5Iu#NiRRzi@
zkvYBo!g*fU-`*juq1hi$wyFQ<sumW20YZ!<AyBGiJ3W(DhiQA-u0v}}+qI-+!lXzj
z8&V5m6!;X8s-es)es&DQfXNa&j`g-T++a#U=2UBkw&8L-lZ(afZ!ua`QlV`b_9HT0
zQKPvglCu<S>jrIu)B@f0y#3}o+~vaQ^@;g%Ms+>Ihwm|6huSuB@&qK^@K?AHk-4Cf
zU|BB&zmlRShK0>9?Cu}fLSTQmsTIjaLn-{{|NH-qzyJIH!256B@_Ibtw~gKF6DJZW
zi~5->g_N)P7AX-TQDT*=NM#wv1M7UotqV$9Y<G**1M^aNef?a2hB8c3;JhZn(}|`v
z?1!HH*kiP!ZJVpHtsb`i63R<OqD+GpvF@+Cf#4*fv2=fFm&8PLku{g9-Bk@mYO<(0
zV<9DlOxqf6+dU#DO7JXU`*loOQXpI^q{b}^)A59eg`5L1I+AzQL`flqfL|TDsp|pT
zSk~i(HP#TWRN@*Qt8cI6d_?zqyz>m>z<E2Om87!<A8U#qE%7ly$!yDvZQHtP5Ry=?
z%>^mZQX#CaLQJlj;uu+6CKkmq&(~UFr5L_GJu^=Wht~4;-6Nl$o+%}>t>JpP87W$!
zjbXSMxOsR)0B*i8j(fuDDR+f&Y)~aLy`K5_<-*GvXotG)aWh0=Sb}47D?k0>&|0z^
zM|2kaFn=PX4j(tZyS*VN$LEiqQKitdg05+pFNx(c;Wy9C-GM*+w?FaCyGNu{)m(VK
zaQ}FRHq{{8nUM*>xKGUU%FFYK^YKC|g}b|3wAL)k%;HwYVPJip`1<KYEE}c>T4Q;8
z=r}G3sV%JplO4Z#c*Ff|$8gwTR3XJ`6SUU}#5`ZClBK#h#8r?M!v<llV|t%!eyb!>
zTqq@=b%&CQTq4nJcsC=8LU$duZR@*FC`wB8Tu=&m+mNzmV_MSzu%2IPomLy#Zbv)p
zDLE3?j7Cy~L`w@+l0_jVhmHZ=fa!LCV!oX5@l@AO+Wd>#h5xF<Ksvti^?c;h=P$f}
z_Xg=!kdQpM=Og32hLTuLN5Z<Y%_sa;En*2-UmiX&-t5ui$l>jKe*C9D^56dJ|Ce{~
z9(jCwL)iH`UfdSWmou*-)rwAE{qRJOG#cCV&@`m5kV-;{LfSUwxh8qrrhXHQF`TEB
z=p9`@pp<6NkY&R#_7wn<s!XLIYrW4%g0y&_=}k}cEA!>GKFD^rw8I@YcMnW2N9Os6
zZ7kBXT%KOpT;TKb50sSn-QWBT?a<+u6RPd-Youv+i0&4U^!o;1JX+Nk{^fM0A1dlL
z<%}v-J)5oq5DDZEaLZTRaz@yW=TCnoY!l5e&<#CO8P-rKaCSJ*9quX-tR2Y`jIxv{
z=?;&W{sxf}$FDyl@<uxxFnVNLBDc5q#D2%q%PUHGj?+prw(NKJKtj3=A0yUim`|+J
zD@J7WFw$s)jSGhiTMEejKtbZ?`8wKyZ40c+%FwrLq1wz$jz|IRFydqNfltfIFz$GK
zdy7ztei&$nJwEtaITM9cLQSieS_4W*Ot+)c4aAM$wp#l!noKoJM%9s=M=Fby(Cu%K
z+lFdeN?4E*%&uc|8*SU*E)z>|bXrl0C#C@XKmxyoUKiHrkrcGCSSwJ?0U;90<(alI
z#FFWE9d_T56D-G9N=|fbt(q#^VD~!$^=H<-e5TyJ#m@^ydUQJ?#v9sYVR2_}KfEI@
zGu!hEsX(*8Vc)a}q1j47G(EQ;zNeF!`S~;dhyUI0=yok43l6(Gv?`P`6QUyoP25&4
z&(Gv$<9NPQnN9(6sjB9%VQo`+0Lxkr-zrc<p=n17&o$|+Btlir!0a6B3Nl58)?UA`
z4WkwF8j-SKb%Sk3iXA9gBUA@E;fti`kzfpM+af6#1hNQp2*mgo(_bb_g;ElwBt}5q
z=1QeoF68Ax_8XhqnCF?}>kDpm1n+A_S3uwQv_p^gf;D;swJUCvg_MY!qcoZ#B(`aA
zeq&iSq>{D$F<5d;;60(tv~8U}mR#5kJ<YVDy<)f9k&&$1SG=o~Mz;-gwyA=q6fs&?
zqr-Iv9b>2l$W)L*vW3KD*_h80T7d0ao}a#Qc|9{P8y_AXkWy5s5`xB5u+C-i+}-XO
z4m-Nth|o2MVA_V@SNgGI7+QSrq-A0|AJIneZrC%79j~vFgBHBbk>@F~>pQk(p&)rZ
zURWoOP>C%#9&QY632eFW%P(JOjV1VnH;?yxIsMFT7`fXGxOHJS4m^E*`4tFtd$X&_
z`4CY}15%=zTH~JP4HqgYR4T=3S~#9hzfSO!t{C^W*UZi`PYXgQI;*+6YwPXb3f?^&
zxEVF~cRl0np7wBqlobq=O11r|U#oc%sniuYoDmeHHW*V6h(#oF+z4?+Nr^TVr8GH3
zQdmjJWAuPA=9&$u>)9*;C9b=E{cDa)sRoEhtuVGjH{Gw<vFIHsY$(-M{s9GP6jB=+
zV@T11P#D!AlpzhW2FRr(uO9K=V=_>z;luBL$Nt@0c8x~oh%%BKGhd&c`TX@2`{Pd>
zk4OA8bJ*<|Z}!~2dtmqed#3Zk<>{Gb?CHlHy5Dp7Ch!mc@FPF{fBsi&tN82RzTyA;
zKmPA=Ng%W)`+yP(=RNB(Gww%n^32<c)-9(`Um5xaMoHM}zEDV68Qogq*HCxl#-fZx
z<V=j7R)WYI#`cH|##%5PsR?YSC*pLfBh0ozHwx8S=F7x%dI1vcxI>6ciGiSPP4*6Z
zOdnCZftSzE#NhExp<-gZy(h{FZOY!^L#P3~btSl!W*lgS5wxxh{*obw+8-5y6g^Uy
zN=YNJEf<PgY4<(D&3LuvfnPVK=dajq&v5gA&<b~Y#k&P<4Biar+jlhI{vO@7=>08w
zJ97T?&$xA?F^0j~3f1W=kSLZ?DReO)))S&YYb={q_;n**j-;5#N^-dSp0*uWKm7ye
z;~BHxVWl9-YL{CsN4E8X5`tw~39jL=-_@#`)@<h!DZZj@hfjs!_6}ov#>0W+>q|YA
zdMJvyhOa1fnh<qMYH!#=qR9f4EB3Y&L5cNbEsX_H!(i<=Rtj7S<glUROxtRjZig(n
z_PFgHQyela2(3UFgzC`Ql4BsaObmg3*MKO5<xF%7Zk>riq5rZc&V?d1N-I)UOv{FB
z7If32yFCMnuw6)^A<rkY5Uf72UM30)t?;Df#5%9^eM_+&ril6?MzcCk1lD=s_U$*d
zDjOqayrH>!q;D<j`HWr`((Kt%WY>>eguoifF$B(+iINMq`#tJPwkw+>*+@wlx5g@Z
zG6BSf(T2ObZ!lV7tmPpq+NW1sNTd?cO3?1_F>QYZ@B~yNNl2Wq2)zT<6RuxFR=J{D
zrJ<C}#V54Z^}u9Wtf-@3sF0o_>V7>1k1h!jH`enD?)4-2bV4aXXBu+bGW%Fp8PkfE
zil(sy=PAkKw-q;U)vjk7QsGydq!x-IGC~WoEJP&d*Qr*{WJA|B*rr7fI~EsF6vkmk
z*T#w`22Q6FwjZ#KX4^J`FU-@KSYh|bD{Sa5sM{)Is<N9HGe|+mg7fS+T~54t*mHV*
z;men2v@mpy;o<QCpDIY{=CI?EB*)qF^Yg^vc86^n)Ya*cbEe3MZ97T|IOplw9n(~`
z*83r{zuD2XiW@WXI9UGp=@lzl+O8osir4c*Dw5Z2BA3G5&B(rQNhxwZ&m0$rcNeTN
zoc&jxE(={V@!@7i*Jze?xsJaxAt%o3%6A_g>tr-k^fJ_3%z1GXq1X-e_PvJ0>=MWM
z!p*KD`&2`9u0qQiqv;z%Yb1xRWgJ_SNZjo^#(hV(>u7cZ-8iC}3J_AJsRscODOY2s
zko9sXb%U@CGAAfoZ43xqR~ntbD2uiOD2qstvqb40+uk6ht_>91p;gs?i;^mL!*GQU
z**evY%ax|sV%qUK+4h96;<uG3LnWrBg6TSpzM<q)xdpaC3Q^lA+SE&qkc7O#zmP)s
zuTvPmdB@@Ap2Pl*5<Fp@@F@|5<ncGZXMXy^;o+Wc=+IKq4Li=ujLVtM`)Uz;e5gMp
zQb3sb&G+Au{*V8WuRs4tNQFaRz3zH{AV+~M1?L>c^Qp=p<|9TKw#}23K{dKk(~va#
z12i3ObExbvTH;kAWmU)4N)nvIXhTSzcv%?lcNICjZ1nd#l9YH9MQY~dLJW@G{(#bk
zuCM!>7!#M5Cu}p&?+(?LHyv5l6WUmYVaMU$-J-SRcs#SS9VsS)52!AX!;JSUVO!W<
zzS7H@B*j#+RCbQ8Z%DSK-x(4aVxqW}?l7WnM&wdctxgG|xbD75WO99x7tqbXba}2X
zL}4gWV~s>^6Y1#_?fyV;4k<OZ?J3bw!a_qvmWtdkRNuHz^+vHy73@*8<@V+l4z~oU
zao%x09trcAv`pj>c$yY=LZTOkR}d2<?{U*a5}C<shW4Iv^A@WmWj+xE)HQ9PAII9&
zE*VFmdGj4PWu`Ab5|$UTkf^RgiAr|(>?w^Tc#oeygP&_a&{+DzEw*W^tx?r#P!@t5
z9EeB=mA79cwHKTtrSE7DcLaGvqzz^J+CSEcG;auF$;*UuXJ#oXLOIm2d()2yWhj1u
zk};xCN<y^_i8|)Z*L^*8ok)?I-Qx`=EW{KbS-OWEtyXNOXI559hM#}>NS2bvx9>?Y
z)lgB<tZqhzS7eTabw*l?D~WV|CB8%m86P~;<-+5KZ}2XnHXucmR2b9Xec*IFp{2r4
zGdg8T$aO~_3(<SjjietsQi5UFSNK)a5eu06Z}A1Bv{n2fBt>ZQH8@u!pfn*DvXDp=
zf|f1%`Z=`bN_#UDk?Vx0_=<Zr2B_;Mr4>SCh!H7rtw5UV0pD+IZesJb+n;l;b=9^S
zM1v2UUe6d^&@!X7plhx#$x`?9Qe+yd$tl%Cpa5wMMQ7$%&^gm2u(|G)Utf<1W9hr0
zB9v|OtFbNwPcO$h&YoA?=5cFno|v}(H4G-mq!el%C)D7kKvgp1HSd(9W?47>@ee;?
zx{;tGH#dj6cXof7q5(zm_2m_x6{hLxSUT4=htdXeLTZO@JIrne-s4S$xZOYOxxKxC
z>@e9=qUWZQJl@<@XU7(h(c_Bc^XtUBHwWI{7!JFR*Vlz<^-PPShzgO4YjrXxVBS1M
zDZc*v#C{k_(Ib@N)7NL_^TPXY?$K7zHipf4q%wpI^Rm$OJ?dKZOex`Fs<hjXP(mRQ
zXd!62mit4`zSCqEF{044wjLg(KsOd;45+JlFxMJjjFmR1?TG1a$f=Np1J$8)gO-x)
z&!jS;2w2r)J8`w4R=AMT4VIfLMN$(}M(P%2cW7l1(Li>j;3yQdv?ytD@l5m^O51ut
zRi@??gre<76g5<<>c641LDhyUMQTzGb!8zn2y^`$6;g2j;`jT%jx*pqItEe^I3J-I
zkTG$y8^8y8TgPeRZe$A{-42){xV(PFO@a1s&orIULeL*-$~^z_BZG{z?;jcOzo{qk
zyfKGF+jbOXktMO$hU^x4CGlB;8c^MUZY0usJd$}@DYHklEooVhG1eFOm0w*F2&HN=
zGzr{v;dnWtg@O>+K5p2yrD-&=#EPxe7Gd@*muJE{GmInsq584A+c%Wp*tP{@HJ3mA
ziQr$jeN!-v<;{1$<zc*M{`@1;Wnx`7&=+Lz6f)7x6cV-@Yjr7R;^u0HJJ*Ca%Tl=q
z(Ia!ly}ZzkJ95mV;I1D!#pU=+zdKYMv{V#nP$FZ;1FN_YQ^a;{txJ`}<@298egSf#
z>3Y)T2r;p)N2chIF;H}aeSBXbMoLnY!771@D{F9sZN*4~ZVLU_bNT7dJpb@#_74y2
zZy%6vJ|KkPa{P>l0WSsCcAyn*i>y8&LjpUHWWu?O=@0nn#MKZ&a1o;n^XHHFr!PpQ
zaBd;`jYLMQOFf*#NDP@&e!y=R@_L5k2`OWm9%)-{-+rJQ4yb-lySuGGnXr(2y?J0q
zq^axy{7PQWtlLt5$fc<J>a^mvKwKxf{hr-^M;~kQT;zzg^`(*v6lE*vQV4R`m`^Wc
znP~bEiNLQD1cz=a;K>+;2urmB8cQn${`qHQ-*e~(w9;(Lg?oJmrlCC?==&W7Mb{YS
z^D{4BJ|Z=Ay8&57jP6J|VaJ}NH6#Z^kBh1zy~h#T_ehsfvae}oUut$hwg@HA{YbaF
zp*et`kMz48%4(F*bu?)@q;AM5&^0Z(2ufB&3O*NN6d(&yHoVSjUE=_FNX5}KprwLP
zFfn4SB&A5-81(gl?Y#%eRYv3yNg+v;up(1LBnR^=xS1?7MN9An?;NMs6FDZF^EDvp
zBOzv-_w4s|GEp_cV6XaTfnc5%d=X@+$x_sSs!(X#kW#8_i)AIg99dK5{>_f*vf-Sg
zq>SG@Qb5->v~5G)0!qni<1&%GN3;V<)dzb{9v>slyL$ViwvhyaLf~cE_|O0N#E&1(
z7&2qo5K8cO|L(6SF`?TIYYdwUNUc!Ta^3<ze0<@zN^;P(?i-^gr%<1T*ByO~k)#B-
z-@L_k14b*{bfkpH4#9U#M^33S1w_v{HvIGF8QUEAtMBg#+YGrdFAg6fhh5Lje!#CY
z&(nq$&|AeEGS5$E1chmR<#s=?OcRYYJUt(gqHwtBX}b$j8lGNGNNo`+bGj_-haRI0
zkGFUH{P~IJ<BSptqY_37-rVdVMLHw6*>`AF>x#CqATmW<@z2Iw1yZ{5{i%k+k~PK-
zSltq`M=MJavL0l$K{tD(>`4AdN%g<S=tyC%#zDP9YD*4l{og2rN{F05CM1Ov1AeX9
zgpzX2fkltHqK-=eqN_XuDe4+h>%ZI>Xr%(7h`}L*#@ZdxFQm92Rb4X#?}%X{#pPGO
z!oTTZpqLibjieB;+K@$t>B9Q*6_GM2I<%C;oH0rfR!3_aHf{LfpZ}RZ{q+Ch^)mDK
zfB$#<-T&!ZdfPC6{fKR<DQ$Ro&$e86{pkyFTktox>~3$#q0nf}cHB5_j@|G7J^P1u
zSf%;$kN=2USNdJu|Ci+JNJqD5;VU3$b1W`!_waB9NClj)-uSNHvAK<`HN!BXgv3XO
z&4nTrr5!Pcdwg&h1a{aXngOdjWIN(AguGz)2bAr3`Sg)(oA~B;|DK{diuaHcE=2mn
z4au#f9BA7XyBm?a9lN$gl!#w8rkK!;r9IrBq-2@q+S5+u+ACI&B{>Jq=c_Nk)!^Vb
zjO=$KIYac3yv$@%Z?c6<iUFl7n8lP#*d|KwOwUi8o?p0sc*GY&REltXp=&ad>ZsSE
z&>BIwoOu59neK3hX?mQDh&3>W$j~+DhdYchEIwjxA4o-D1S~)QKpAQsaCrED-d5%x
ze?(5d5Wh;w?hd;<;KPiywyt|bJ;krfg`y&nO!ixK%a;nk%G*o`fe;H)NxWN#!GjQW
zLa=VAb;X#5QsC}wq0@%2+3RO?rWDXt;@30Y)!Tz=T9mSsZ6kVDN5#?ClfKk6{h^K$
zRYvRTDMvsl0;6hLIHmwOvToHFxvmp-s0TkmO{`ztxfpA(W*``YN{O_tG-T}KW1U<{
zL2CyZQ$tAT2D{%AivXc0IS|$vku%yTim+9mJMKxxBUWe{p*SBW#+y50%m`5kzWOv8
zMuy!Z!Tb7>QWksYNg1LssJjP@ROHi0G7@PTiqJ?ipqe@{Go>JMMrDX8lR|y*Tqei1
zt{BttG&wFVVwA)-9rHSqInfx4R2r+`;n1UUV%<DLZxM)EFE)mu(TMC(){v9OmV`(F
zv0dPbPL`^Qk;1lBq-`!N%SuXxIc*$ImpYjtbJ-kiV`@k&XB1T_Ym5LBSz{&?fzpzp
z?+{8c?t9jCt5Z(}o09DB-w<NqJT2s0vE%DjFBeiO)@?&8#r@rZr>DZYEJV{1y~EfV
z3`;ptC|72^xK0-8ib)7qLuPqenO)(&S2%y-x8HrBZ!A6rZVm^0jx6hjwwiu7GM#5G
zzWO(m5cq9F8+m0IT&rm=5}c#o?`ehsq{eS6rmd=BLd3a=7(9;;H(0CLd}3${-`(ul
zH-%FWoaUL;Il9(xe|JFUh-h!v$_2O9YOWLtw>XSe<eYduo-j%yL*(Xu$Fe!h*6{K=
z)Ac>a(*>!!N^Vt>m@>h8tTwdPay+jXo#=0N42?vj#Mm_qouTa-+RieJ9c@z&aQ%Ky
zp&kf?(n#HsGKBP(aA+Zhxi(sIMhJ~oRRN#Fk+21f9T38xuarTpp}3iXyygSsH6u{L
zZL$Q?dZY+-{h4z_UxSM!C9>9(QW09-ds14lT3zwov4Vw4oz5!N)Mt6}NEF7{3i2vI
zs_m(YOjc;!qE-ACzu*5=hk@HS-_i_wOmNjlzD!K>L^}+emyNW{q?{S<9tg1zt*r@o
z-SF<+1KnZIFP~od?%TKA^drOFJ9ZCmIDPq%^@kt0{4f7+wsqq$c1UB8Qt|ZDPf#*V
z*ARRn7Fi!YuD%c+-aK+XE4<s--QEzl1?MLE@jxp(q|s!T&}_`-3*-KPR+*bOcVsj#
zPZc~8OJ=5!Q{{+<oN4-jdD(D3|1&8#c72NylDw^?6mdf0oo9Z1=JfQ1VQ7%weLxC9
zzdKZ`+WD2l6<i`Ds5&Y)eTQvYBvnL_l!T@O+fuA%-BwZv*V}<)xhypONYnRxe0s)R
z15(B|47U&D9O$%yZACRZyi#})$whD;NkUnJu>!v?xXT4Q?yC0mPe0U8!mDKW<}K~+
zKx<l>5|LrUEfHT5rL|aN7{7U60WaR8DNO5y%{ed)df1c3TT&L4C6EDWQHMvmZ{8t&
z!k?e$+J^Vvyv0&zvP1X<*-6}dL`yKL?uCclfw*~ug4KIUh-~YOo1bZn!E_`2uqUU)
z`Sls^wmJ%xlAO6#At6(6>&&)PC{+k6xzv}TYC59VEK}Y0`*mVG95}rFfKP&T4X6?s
z+d83XI!zLa<g0tUs4CfsaRVQ5enlxLO5vR&`ash(JUl$2yB?`3yWQN}A_~mY@%sDr
zC>JOpKnS(_ts7!=6a=OlF#SMY&kVg_Tb?0HLaZ)x9cOO$4N@hxu%UHF3M-~FoQ}_k
z=y2YH(QM1YVeBem6@hL#>~KrF1GlWTDKm`M+el<`E4Cdd<2}A)Q2HuIX%Na_tVRim
zF10uALtu-Elmr>Ym*+E^dm$8AU$80BT!UiEx?IVA4OR)x=N_vJT4!!=2ByoIm}@=p
z!{aTty<unygU(nbF{U9-5&T98f#v+n`ROZZ^^nT-Au2FhviTZ%z04~@2|C+WT`+~k
zd%~3-sAZsOEm>=}RJdt(6j_nP#%hY_E9b)48jcMKcRF)f76cX_JU#{**&;-)q&`^~
z`-T@$NpfymDW+N*rMgyJWrZ`v7%8I4oP-dyj!8k=dN^OU$avo~3@uO3&%FQej-24*
zFTap-L>tW@G|Lv)a$UE$tzbKYwjHT}wT9qUatx#pkfu&~h@RqnT`y7_Ap6~}k{A^%
zi^posfBN^oA*aaVR`%maHyU2&#BYB4Ey*u<7dcKV%Q7=GExW#>7Xt5VH8=XLQv0q5
z-Wmqfe13YxrNW1Pq#t%ni$h5T)C2djt!yqLq-GpC1d&E7zWe5p7$cY0S2{c3LqHpa
z)^#!<grsTOdXbHh<Q=NNngU}cUxiCDJ5WtsDdvQ@24oWvx4s}{#tyegs&q)?LevdL
z_Skk$2)-Tyu3iDZEhrVxvI2!dh;U^HWC}`1jBOD}Qru9Y2EeOeD?s*j<yZ<iCzLWZ
z97w8-A_4-CO4rX#aJBX-JDR>a5&q2&1JlbVUgDX{>oa~`*p0iY_B`D2{&(NAygp%s
zpuN2(WKB^TzpfnaJeRcc*jOIkJ|LqfUS7zJ#ee>RvK;Am3W1?%B#pJ`Qg^EJmrvM6
zGw#1-|C{gl_~TC;KmEus9I(THOo2nAX~r9JO60Zy)3Eu#e0nB?NZ0l3Z|+e0H}!=0
z<&UJ}vEITO!FpUNxuCTHBt=RZZP0DUu4&lLCro2OYa$tu1Iy{e^7WbL*E3QU`ljWZ
zZ{LyjJLczS#C#-(Kz19%z_7cmINl6h+md3!%?r!<6+wn>2SVag#v%wV;N8OEa03!T
z2$iJx`7@t>`AE|?ynp+qTF8zk^0tu9C-(1u%i-GxO7`G4^l+fuR;rs+kdvz>wd0v4
zCFavB^S1Er`w!^rLumW*5t##9PBbpiG`2qMb0YhQFik~jdPhl-2>4RrDW8A-iLQG_
zE5*z65v3##_Xmcqr!yKKGVA#jsRYB&qJ=<{5!(^J&WPZNqR?1P^egA5PXt#-ucjMn
zm@3vAm)067WqlzFkn-xB_mwSPR7ZQ*6Xy##W_+3|W$p-Jor%jz7O=a$t70YRS?3G$
z^ApSSGbtsy$M0!w-;lN$x4w}5`fJEDQ>%<+p2-was^&ko2~j-5evi?HR$G$y6kovT
zdKgfW6g^5nYZPIsL_%#eMj4v6z1}h^!y~4`I-jx0qe{X%PmT!%7$uPc`mw9yZ?}=A
z^{OVe2-C7VG~@(*8L-k);!4N?ilV6c^P|T5dYjRXm?Eleuu_uzN^lFhDX4Y?ZR+>C
z)G;RY?<=|&(p4dQnH@iWdSPBSj+c#ftNUS>GSjlwc8)PPw^jbSP)Mb?G!43KfWqx*
zM~sebiy$=r{L>S^`R1O#{q}(B6n(F+6PSdERW-i8p7EC%Ej2~Ju<Ni2a*P-i=o*WW
zkVHYr`l40JvW1AX3LgtD7u@EEB2*-J-*VfIG^Qcs%;q8?U5FXTg41~>OU3yz(={Wj
zbJb@rp)(p+s*50|L}Lx!Ia0}Z=P^yFxhJg=r3#GdL?G3Gs5B^3r%3`)e~sNhzZ-bH
zoO!$*QBw2y(`UlwQCc(fBQ6%cz8;x3&uLm{Z+A#3SS}O&I1;yw<X56wNg2BRzOF;U
zOxo&mA_-`Ro>5tf$kpKJ&qyoknm1O)0a78N#x{m=Y{^mLedcwEv|Yz<zkP?wnN%cS
zUM9Z2oEcjUMG$>Jw}x#CxUk|=#CAO)6oz5PW$}cVYIdq<`fgxzX8`)X;o<g<-dKjA
zW!!a~I!y?Uwjb%o9&H<hHWi{3BGLs^<0&cB@M$}cazd*H)oDszP^zKtZdkV?P+>^c
zwy5z2+wL&!uJSL0AjU+$A2H^tpBJYZYL<plVm0cPT%|-xqOjzWi7C~U1qyL}71!XO
zMu}=PB)8YgK&jzfAu3#`2#Uy5NK7RGlNI_}KTo-E)nWbjsTQ7JzaV{!k+Pne4JaX)
zU!Pf5Pru*MAMQA>GkJ4to9FA77X~Znx*pRHn676xb_m&_v?TQ{Ejxq)sU`U`vn(@>
zR&~s$A*E0yGL2<-xS?dp{_d8W+gs+-v0BE~2})+TdxPwDq?a!wB@jvAmWiex=o&)~
zo<IG!e?qPo#y5|c+Z&2)D-}u^nzpO;o&@VQ)m&fK)o;Ghs9|867v>OfTSaI#!@ziV
zPd^@*x6e$MGhAM2<{p`04UxXzqjz_dAfTKX`+=}IOw$tm#`*b)X<lfwVR!q1VShtf
z=Gqx{j<)Z~fODSP{hnbr&<{O6cuEYci({G{-FFQ>ZiLe_VVW@qPy6s5X~!B!Ndi6I
zuz$=v{ro3_zwq5}KXCi)x1<EEwX{Z%)|o&4;g5*1VlZ>o?DjiC48(b<RWhxyyFJ1?
zWGB!~%N8TsWns5JaR2rVN^4r-5hAjF`2{HjUDu$M!fHdwiI?*kZCkp-9&%#aR$`tp
z#-ep3r$h{)j%}wiMuN31B?R0y5sf2X2V_zxf?o(S)(}pn>P_|a#j~yi@9^t<?GHyz
z=X3SS7nsf$atK^bM~=rM!{cXe9^X})9}#?op0w8#xLYqwUrxxoNA_>NM`XwH>7V)f
z$3OGtam98GN=mkELtD#@QrNbKQYv10J_CZGYgo^P<#M984Y~x3(PUM@Sz6ZiPMS~1
zrX_2G&{YnUQ{~!I3aJQ+wCGOLjU#cLNy{rm6~?=J3YNHdT5GY^kaUY_JG|Dk;|^I0
zrfGi-(`hL{n_7F5l2RfPiB`4cl2amu%67=tee5+PPeCxd!0|k>Y%Aib6iz<i*R}qB
zl^~_ae3`231&EDB>WVF%US8;jDtL*d;;6qqzwkInM!O^CLZ|8>&~Gcndxow>-yI0v
zBb2~47AFhm*Asqo<RWRTLfQuBJxU6M$lMM+TP)R8u&vcFXC%fLhOV+Ix_+oKD}SzE
zq^vNckm|v(PH8qy%S@3Hqa<TnQ|2*7ln``n%e<^8ZL2H67^IR{{q0|*OLfnywZ7_s
zOI0fCTBWqM<#<{soknZJ>&q+aQtf`@en;Q7{Q0L}xWB(8#YCjm-_~VD$-?DjB?pHn
zk(dHa)1tMys(drDvB;w6MV(y6957ZRN=1u1ztstf(KJJg3yGocY5SJVJ0!Wj%n;n)
z-SX!ChHZ9yemQZ|cQnSb`9#wwwsj!^uhT|OiEe0+*YM%AMZ8aJ+g2|KIb*aY2}q)P
zMGm`;n|>hYfKol*et6F`ow>QYq3e3Al{9Tb2%)ZE$33bUYGy#nYlB5srF=;w7s%P+
z!})sQxcUjR#5N<^4%lYINWr$wSJidRrp2_;c6T)(Um!+T4+A;ZKy5`<ky354l$>f+
zR#e$hJzv+UK>?&R7||lNx&}Z!F;v}hU8~u8h}P=b#PH-2Nhv@|_;vg5BPvotpoP3v
zKW>mh;-%z|KYj%9g}?a^e}j|-EvR6Wc_X=n@o-x^@Lk7pJhA$XZXbzZhSMv{<%k{{
ztnIK$5Gm}(9gA&i2N}h1`^fI$o|GkR+ar`DhK%iZwFlbmY5>ca$`r5)(KbY72r2XO
z;~%*E^aFGQQ%IP0#3;-0%L{k+kC@#JC1!lg<PaG5dvx50>x{M?t8<hrX*$q3)At=?
zFBn!sx7+jXx4%b(jpvVnVm0%$k)($r*<#|4|MX{ad*<=4|B4$r>{W<i`ySJGG>2Qp
zd109+ocA=Tj#5P_q*yDQ!!RQAL>|HJ4@}dU<TjdNL^drh3T!*#UtVx-LJuvMufO15
zzao_;Y>pBgb{uPm+cu2%?`S)NG?Jz@v<O6Ugc!NGxnl`|;8yOtLp65^g-Z!aVz4b%
zH)vx>1$;_8-rr;2J<?i>kEwd^O<Qjn-Xn@flW^W6bk#~{HKOj`(cRsWpFh$mji5dR
zQwo$2X}bYC9+1jV!h$<*#1Kh^iU(D?678Bmx?E_7k-NvY2yJ+L`@nR3;c|RMg^i}D
znnMtZ6f5?&h=Q>V>$YIEL26Bmp5^OD!tn(=jJ*B!eT6H@f?H06ZDYHf39D!L&0nML
z?kUHwm0|BA=a(b4Z!ks^OQF%0lxvsW`^p^`F;zx+PJ|SQej^fbX+=m`!$>KSa=|tu
zL0Mu<IR8r9bST>q{1tIZ0j0=ABDLn}^@uEyRz#kE`9xSfckjPLw}#{;&TSZL5fc3L
zOgR2P*LGNKiKSw+h16tHRgbSLx@l<p7K}xho*eTPD_e6vDJ60$Y%bK>a}0RzDJ9nc
zEuzALVycH2Q9Im5E7r9_SV|5kDUnhUqvE_yU|LMqQz%^4!jGSxX+>af?r2O#l!Q_e
zuLaG}(GB(IE2YSlc}pJ$QdV_$t?C4Do)_BIvaBn)Xxz52c~952SEa0~&zhK7HebKL
zsZN2mm}{^$2F_c;IY%n|O8@gA(Hg`3VNba%c%|4}L`g7B_0<a@s(Mz+8XilLlG0Uk
zTn`ZGinUe7)bVjHEUTkyV7I%$r;MLgw5%_#aTvK=PIO(%;u7a+=6GIs-1S&le{Dq}
zZ?!d}yOt~@&O1^JXxn0JLtJNah-`7iG!-P5M8Wiyhr^E3X~G(dl#<Owo}Zt%xjkTQ
zL-d|?^GGGQzq!RZhl>HJ;hVPywk`1U^F)XqttBx;LdtZFMk|HPHDoCY%<G1+21;UH
zJ&Ox$A+pR1ce@A1t^tuSN?id$nuo`SI;Eqs3#`?&<5;~JQi4_}+tlEzkO-}4`W-nq
z!tzA$3n@q1cBGU_pv=YBTtJDGyy8OOe0rg2I!Y~yqD+gjEh5)TOce}ed|a+;AYBo~
zXmSw<l|fW7pVpR=1LFVRlvyGvdt_-5(gI)_UBO5(5WPd04y`(ng;F+%4vNPw7yPpQ
zda(RAJq#c+L))=TnQiejyB(|F`0o4P(oo3r!n6Mw-S$Xr>F+FuZ{8x}!u`#G>2&7u
z^p&7B)A0pk8p7tP>QvV1Ti;up3uLM2_bompOgpf9_YJn~u26~!1t~d^Gc^5BUEtSS
zk|a<P#8lWlJfOxsUcit4<-ehC@6d0*sW6lj(J`Y%;r6iO`1FG5dQ?A9!dknZ1xkuF
z>?I3sKG&DCv@GW%(Qh;oUjFGnW4j(iMmIg9YUyrokU;t66PM}2)5otM9PPVDjHd3%
za%MM<bk?$2!#|%-w7Nkmi+6P=FN7p-ndeWRSf+{h|KV@BfB!u{{crzstv*RfYXnW-
z5Q(hwg%TY@+mTA<%cq}t{`!^dJ;oSx-!Kk4+IGO)9_Sw)Yrv=!@^r>c69vVvzd`3%
zN1D1P*%3jYFA~!?$Xr8+b{N40h@PAh-bd1O<}$C`JiH^b)?O`<>{m*b7~2s|hi@&-
z;T`SnhL|HE)v=rHdm1I#<}+DJLR=B4Ww?JJ#l*I5Xw%a51GcgFWkX8GI5=d!L7E0q
zK#9W5?tn>waJfLrxSU9-U>l9wW~34I)_!x#em@|jrm+@P68%;B(R4jR=PT%>zVwRL
z$e|}k$Mo?h%sO%T`~$6(y#MgXWm%Y(g|2IAB3d@Yl-Q<oy^&mNKt6eLjKl<{9g#-h
zykowKbPl&S2yH0YliiH%Iy!5KIk2v^o1d-+hi*6^sk_OX3tJFG({S_l137rkFV6_m
z);+3iD7o?@;x;qCd}K<AZamO5BLYQh52!}c+>UJP3e-$P3<A*xq-aoDqE%(fdsmpJ
ziSv2o^|%pysC!c9vF7S3NE>~l@jhX+!Fz|_)|&NDn#MK=3T@NWIFt|sA=#Fdrfsm=
zGIW-^!$8+HsJPUd)zGnfb0A-iB){U9^~&&%*Oawo7zILVrg>(XwrVwX0g7arSDcHr
zc3MCbXso6iIue5C*9$piF7t*90goj4j9X^TP7qThB$$2e#%Cc3G0^orW?2b2F}4~>
zk4uu6%JsHb&`Q<DO4H%piqM)6LdD|3Rcq^!LNN3L8Q5K?3Ll>@SS7KtaIf~v^BE~?
zc4MA4)-AAZiFsKuoknQ^DPF%X65S2xrX#owB`RzuY-@$F<%G1F;jkm-TIDoC;(V%8
zfxg29Pl$zYzWJ8%c25e4<LfI;mr+f_&@~LLA%;l)^o8UT>*~2oGqGe^eI3Iqg$t3G
zJ-N*Eqo<(oDKlSY-rwywZ;@YKFRVUt)3p_dr8P>5>uw${QZRIFePsb{)1YPbGzhL1
zs2s^L{^|-yVIzc!98M+Hh+!$X^#l@>syU+=H&R@P{>XGWvo4PP{vIzSLP+}k18%#J
z!iu1d`eX38Z9<!x*~+R;$dCyi98xK4)1$P-%@;~Z*rq|LYR^rnl6&(t-vd`9d``G+
zVi*L1ks_e6E%~}$ENwt1_1_0N)W*#D<%QJrlx|1c^b93pm1cf<qDaLYUfJE<*PDuS
zh;E0<jy2V8zRZ!}6T$l$kW-rR{(=5lx0=rtRAFp`X<K%88r`%ADG9FrFvy$<p+01a
z02L(4+PbSOnV1Xf=Fm!3RHzFy)_~Ql(+o<}w|Dfrf!)}nV!+QQOe4uDA=^484SGOz
z7Oeznn@}Z^i@=Qg+J#?dIwNtxR|IYgjkQRb={iF%6UjTg&6sx&y!q{T#j+-r=>oFu
zdcXYgBcnASYhL1fJd%Osc*YmO;oW<}birT8p6lxqvr;6#Ac*ATkV-RuJ+WRE+OCD{
z@uJ}83tv8c;mhYIwAS^XdBX(=Oxzbx(z4(0Smy~H9Zf%?O@}HlSc_TDh@vWYTY9X9
z=vU+wbcBS<%QG%Tgf%rNDKbOfBN04({Yor?hr1iJu0%JttxU^C?nXM}S<go<$7gc(
z7^$jYMKv^LB&LX8&zyexnRMq-<5)YwiTZNcHcIq}=;*AWeSCw|79Rpe!eABpuxEeR
zU#n&>G*>d%aJWOY1Bcr?#zrGk#P&VPXr`}UaL(0^u`#3^Ng-g2A*X`(5n)<}+qcyw
z`ScUAc%*31+OogtC^v$VYdB7~17TfiJvSv11=$_0$@|Ri<_<(5dB^Fza9$&Q*RTr9
zZa7d<B`pfm5h*ozCpD@&pxYKGBp~?>ks?FilDCSmP)1?I4j)q`2MU3(mRv-wp|*Qu
z|Ax3NY+FRB$j#jy<L%A$hFm!ll)|>mHHfG!RU%AX8wo|Y#;9{qOy`;Ta>i{Nq)ej?
z`<oF&=5kp%o@<zDY+Gc>oaePdW3<F5MccN-m{``8EkqKEzOjsr<@<*nH$!t}BnZaC
z9(IZpH`dp>h6tO-Z@!vASI0JQI2TFT;d4SaZ5_Wi4Xv_-G7*HUp<7V}WZG()t_4w7
z3TuGb!^<3TMS@hUF0f_Eym{784+1{cw!$>8WFgU7v)7WFp<{NLHA-HmjaVvMASHuQ
zmQ+d&hg}Z=Qr4HV5cSWXn(`{^UbihS3!+!#=-4-c?}i=(%=5ywd3=b(;MroVQ#z3e
z>k6roSM4x>G;GU+h_SA31XRdkuIr*)2+k8jK$Og7ng|&l-@ixc%0W2XCr<Oqayb)1
zo#?eq&*~!cyrAfy6uP$Kb-f_9;bzz4VqF0kZJ}gb%)Cq+&r2lGup2sTH*k7>;`8g7
zf}w~j_BRw_NL3Y`A~(B%zVFE~vaTzK{f@>OgwpsBF}a|Wtf~8A;W~LFWt7xNnaEMn
zb`NA5t~J(~5SKdgcMH+akTP?u6H~1k*69VUByGE^IfS@Tu2Q2MHYf$7Y9O9c(cMzl
z%5lnxrW??<M=FWQsRjVGC6@Y(P^M0iNR@|>qbG&Re^5%|{EFBvbp@BeD1{c9Y<Fm<
z{`)|Of^G2YjK1=fV_JDUwAEE_RTXqBTMe4!i0V5qhNn-TS=P0F2&H0K7jlR+eP4Iu
z2x2IF{rD3`!MpFjr#tKkA=C8@!#I*+BqfiK62EM?;4#`jj35HuE0pQ3^^jVpTfHI+
z!|NaZj8&PNyFITjC*(Sj7t4|yb4nN?X@?OrbR3^f_}LRogKDen$6cPu{(>>}4@j1&
z2HLFV_T~ZG_Vxeg(=X@}(PWZen74@{6z$=TyYGLGunlF|czyj!nrE7Rs7&>1s@kQ1
zi=pm~@=Ck!FnR>j(C;$i?>;b3XQr1|M9JJ94wTp;ZI2#zoX;zN{ICBDTMTF|*bN<h
zqdA|?6e$U^9-4$P6w~nShi}PS<cB}}k-LWn#{KSUUuy^{0d*=;ge1==rps5_wndwY
zzpa|-o>B_FWQvjy6ME<&M?m6ZVU3DxET7LFTRi&?{5)}<9rjkRUOpj3tyP((!RJie
zR%DUb{VnZ&PX_$?mCKi3klTzb0@GM*Yp;`pfwWO6Vx~n|%MO8CF1VaP2#nSkt!Zsd
z(@)b(Gwf;h59|)N#LE-Q>6J#Qy7x8~Emf@;0tg79avB6z$WVr?8-&e7CopOt245W)
z5_D_OO;6L0gi`(T%8Zy+RSdJ=$fEA0DY=G{jHQ4426g)__Hf5A>`)~l!i35Nv>^sZ
z@{uGPbbp6X26uTOEH9{B)6BXZ+3$7)w<2Y|iOQxyn3f`-XoFG~Z5kTWFy4PtNsey9
z8iO1UWZhk%H`l)en86~tdRr+u6P+jK+RT`zjpyTqU!G@X=O`s30CnAoTdC;#j(yi+
zwC4HcMDmWYZyAT70)nKtCf7x^UjnMh<Xjl7;Bnuu@7vnK(T0B4VsfS|iQ*k`p1FKI
zu}llzN2Jy?eS<U^ktIfHk`D+ad48@F%GEi9vE1$U3^6dxD^eNyu|r$Kx<z7vHD;dH
z$mc1N$snpgEL_<uTP(~W;(Q{;M7-%RN^-LwQ6ls1?!eP=;%O<o&NVk7DU=v1fa+SQ
z6+)ntsxOpli{LNQ&yoxF`pl5p5?x~3B9AvWfZ%wXan7-B6+H9i%_B-%=FQP5Ny(A0
ztTmJ3BlG+PV=8UWY6GegT%|HK<z57r)0xg1oO6uBfVLf%<B38+$i(@wF>i&#!y_ON
zrKuJ_QNJ&<Sh2<sQ^19~Zu#x^-w=c2%hQp|wABjfwfT?}{PcX}J8g(Au=qfX3G&K)
z-}2bot0Gx)bF-^AVPDbUrfslV(RD3o!Em^#f9A4&W=ZbKk|<TPtqdX;itqrKW)CHk
z;)cLt^S-v{#Dr}|Y=6VHE|BXqKxkcgB3cmKUw8wi+7GERAyLg%NM)*tkXk!5SK(RB
z{ivF?xwZyk45XM)0z%qIIb*a!nii!k+w3WYt2xwMc?XFiCEjmHX=>&0U-;L*>M-#1
z^~n1-hZ@K;1KTq3X}vICCc2@=Md+HA<@k#89=}E~b@Z_*iBEwyZ{Dyw++h1X<{Aib
zE^s;?8OJ?`-B6k3!k`;LjD_iVq-pBD%*{uFtJR63G?Kz)+VIXHZBIKM(AJU>%v(hE
zmiF!r)(gEoFmD+nCEM}LoCCvd1l=;v4wK;N^JgxbLyQOZH}|kD=x|{-c68&OX`VTL
z{X%q(58r>FX?jGK2tSjY!*mTzzbE+!y8~^zgKep4P`@JNRfj2{ZF-!aIlsK1jRC3H
z^#j&u-n@THd)V`GTKUue_9t)`Y^P|umN*R9wj}~73A&=``yKCp_dA-d<MrhgDFy31
z<3d2T4Zr#RI}Z2v2q8%!(inrkOoZULIgH%i+;F&mAf<?3FGTOL#**De%z_+ZJ>^np
zrNIt8;<~R-IkLNdtYiLV#&ly104+0LK7ZlU*Av*LCW76}<K2$V8jQ7U>&))<mT$iM
zmSx#UF{4xg@5xFr>_&{#1R-e01Mzfby<CtYA(5D-MX7+y8PyF4RhixO^c-p>Nz~g*
zbPicE#%R<q(l<3D5T~z9aUr_}x6OR|@r-Fk>`-ffqSUaC$cdODNvc}KYg>}@xFle6
z=nQ^2Q%b1ZfUZN!OjswTWuhpJX&Y42P)bd8yRF_F8>8v1sW$-IFb*S~8PSI~=p2dD
z6W)7b^k~!87lg4$(-Wzx+it%nyiN!NKBXGQYDbjN2q~~aT!VoLV;iJW^|nzm)>w8!
zrQvzW^<}KDwYEaII@Y<~P{sAOA+BTYTuMblONEP>X<aEKcH=<PG^}&AAg<ercMFYG
zG)D1of6MuJWSVP60#OfPO4m`n)S6alq&AFw!@k#yttF*Q(>C1S9@sSne>s!4h2`lB
zeqG7FW*3ZU5JD1MxDpN}^XhQhMmvnyYyQX}37fBn9<Ath1HnavvK0#*5*L^FbUgAh
z1zuK1*ER@ISk{1#nV1SLW)>fbF{7lSi2614t)*=Z<F4njdG=$ZwTg91<QQ?e&{$Jt
zHz`*peXL`ALH#o*x!Ux!QKVcrohKxM=p&*;wh;K?;}>M9$#b8mbV5~g67%YKI}S)~
zkxfP^!SZ^f<iL_6jcrKU5L3eT4b9C+s}zgKbf#xJUGSHMOE^)k*`d3)4~#cAG;Lz<
zGrAuT2sY;kp-{-&-tO3Urhaxr#`%Qv336oDw)mlEnKnwPlhrn5z_7){;(fI?o)@%)
zcei`KIqVpF%P?qmL&x3C9w7_U@j?i-vZ{<iD1j8Ru1(tx+qFooDY^=xL;)q$RTEq=
zN#o^;koDlDgeE3XXrc(%W&{*SOV=riY3d|L*Gof=3w}K!bc2!|#&)EVh+(Z^KvC<4
zluU_{tU!uF$^O?yOY)v|nXp}tZF{r?x1LC`ijHCqAgd<12#sw=n)ZMw5OV|(iIkd$
zP=+)w|9uVv`^Q_dE?m|Pl|T%8obxR1%(A^;Oo#3y`s#P~#Ur(4_w9Fe{Pg^U_ld{f
zzGe6Lf%Wvv^zjFlc}ACn5{79$GAGY)_m**gi^z`P>nTi0%{rY4sh~}RB#>fdiwC!X
zyl|jmL$~=tc==3zEwt@Gcl(C%-S60c@bw0k5-H4djiN+2e*MIcfBr+=qX|WP{misY
z?An%oY>>8wHyW)t3_Vj!tfwO-!Pxewrs4kmw?rwyJI=2!SZ&zmnK+;E(LwMu<3Qgu
z6e)=7R6FU$lBrUqrfXTwFI+xu^!7V$wc=096Co#Tf1qI4-9I2o#;+5>hkDw5kr-on
zyghIjnmT$;_iUR31UH8p4v!CjWL;PGH+xd_OvfY3>nqPc|B3bdjP7>W?r{CkMwaD*
z%mvdJn!_8U>5xL#>Qr>>Op9(02sWnUGmYsm{l0!pDU;$?+;l`}MYNV+EkAq7L22H%
zk|uacaGal?Ft$NUjc7(p1Cm_V3DeJ<fBMMo_7>STjJG%B<%Jdx%R*dMD4G7Szq%h*
zgm={kP@sf5mKQP9S$<hV005_KNkl<Zk*t*1wj+gzlp52t6?<&!EnZPjwxdvulYYL`
zJ*u18wuSNLp{|{P#W_AdU0^#?rdRy7U>d{Q_wT@a7Vj}y)$zZCRyX)-Ft6$N=%%kD
zd+I2^$Vx<$$y4NpoSs-NudL@M6p<7>UTCuEC~;%?@`T;rLDRBrGj4vVtaaOh=?E#~
zi=fdB&ea-OoM&2NYj0j#ymM&V00JQjLYlg#kC{S1Dpg;0-lJ7@$){BFAt88TjKoq8
zGV8YDL#VrF3gnE{5_=dSWsIr}`OCc34!n!R5U+u$DvXJ#)O~c*5X*Hp-L<^EKhQ~N
zWWmp8hBt2*EyU@CczI$yJ=Yb9HQ3t`TFW|B?gZM_0IU&|5Nj>jT6VUrpXHpfQsYyu
z<~$TJuRLECvesN?&zFni<IBPv;kY<@FREi>+sK(Z;Y`&ppj5Tpg_sB_)Akl)G`ro%
z<|At?9B&4Gd74;#KsWW}l)Q)Q0U(C@y%+N8-_ROs3_?pns#6Ie75&g)mE`p_;hf@r
zZxK>(v%jU^jclu{nG_^GC6tiR7+iD+CGer<8njZlWhHrEgRyQSI*-;0L_(#)`Q?=u
zBDU!WxiG(+E8uA8IXyq&FB7}No|p>L<TxMCynB0(R=Ud9q+~a=P-YsVam$L)g8i<;
zZvoTP^{~6rNtKc{>?|*638OVw4XxGObPYy=mRGosQaI-s`i{n$ngu|TgTt3pQN({~
z4+v46ByJ%IQ|o#;VNHuf)Xd2?kxM40`rK%3N0H_l{Ddqe2#IbggTZfttUS8equa68
zf++|pn!2uja$Ucah!h=4X`Gvh$x%uLisj_k=7r!p$OXS$XzU(J!C#)hSW16~(j!fG
zQw4sxKIahvX<E#+)@Ylae`&YyU$-WR4%PL%`S5{ZcR+R{PF^Y%YFW@W(ICkb_K)At
zzWq@7@6HpaGk5RqFr{$*@(aYm>FG0<<1?qlaWn1^$#HXggQt+znHfn;j(I-OC{?RX
zxuBbRx(!Rc!J4L}8FpxEF}4MvF=|7}Qe`csVfXlf{{9^Y(_F`!m19m=Nb8CenPq;V
z1V`KU7~9bdBW_)3TZ?L2$Y69s9u71}*5%0NJnH6-cnm~k@Y0Z^q0Ady*|4{_9I<q|
z@}osgMCVWf(l(>DWW0N<6{Qkc*M*Q0KY#qhalX*CJ^P`Bw#W2)3Yu;=^76}1WEar3
zroBau#I0(gt!Zew`m!3wLQDY#M7JT8MqfMPAtto893I}%v>ges`+J(ZZ&B+j+vh*R
zwh)BHXK2wJOovpGMm3}@kx{7IcjVI-bXYOnKzDObC<VDrynXjrLrWsFq51jb2q=cW
zr5Ox1%9_*)HKeGAJ=smnbFB~}1=%mS67f$@WEA_GJ4{M=x1dJFX*wZGp&wdun32(6
zS?&hsJt26K4;ZDORIG5!5jFIPB5P2}*Zo`5c2Gcf)uQH>R}cw-r0rW8tFVp1Dp_lt
z^O?>n&;whV829f`<AH5=Lo^2MH>6SQ+V*-&l*qQH>Gu^Y8f(&CDph^vB;x5zSdQc-
zfZvcgR408gNTY~CR=)pMVOAk}!gfLViRc}=KY-Gt7>Ism-2%jpZCfcR5mO-KSThyP
zK!^d6A|be6Z|5;*qT9fU>ZM4vldqJzHtH(<OxcDIGb&cf-@0u~)5LV1$W#n*xq?-;
zb!APNk}|?7LdvzWo55Iv>3h5nG`}KbHIIiow9zcF@bu+~TrWJdlE3-82i`S~`1v1L
zKK;n{`a)P7F=vFSLMYu@#<m5aKwMv-N~6Ys;3M1Si9Vs5jua!2gr_ji4k6*`;&G)A
zOXm5!@bP)#d3J;(@VSP+l&Hj2u9J~`H6q4X51JuF&hvs+@b29`+FFLb;cbT3)56c6
zo+`A)6-3HtT}Qj+njQE{tDtQ=w9*)(DU@n8G$Z@5<?epZGC7uYLg<ENtE9lT?O4~1
zw(E#7@j9P)Gf4JUkXA?WD{1-v(e@v`vSn$SCiX13rI{T&;tH4H5s{)YOLSGUy1^D`
zHX03rJ>q~rkAHz25(FpBK^M>jSh^}JBSX2o>IhR?F3G{NhyrI_K&FozaNzpA_n7Uy
z*7v>d^Nc8wnGT1t&jz9kxOpVa12V&U4c)D)<9edvbeecRo>_kTBft3buekrb5J^wZ
zM>d<b)K7I;k4V7Bh?J7H>v{k9!n`<K1ep?*RXp5oIA0cqH4**%$J(M#5vfeEK*}Oa
zTb$$kVMK{LZu^RLoC{S^2)eGrImhM2(KZ#ydW@BbO5@k5WGy6D5@2zCC{Z9j+7#<<
z@{Z^W1gyklV)-)!kF**|DgrHOaO<_!d;LyoV-c#pHlZ9wl>nbms>q^Lsp*yjfTM_C
z%6mm1ad}3Fh?~yL`Al6k<mAiGB{-rR(W10>vPi7!rND-~BLzoFk-9Osc_9(dRrSXy
z-rwqBpw5|%G3cf)AEF#lIbrJxK{;vbx?*)}Nm`qV%jtw024qUu+Ry@iI`ZwuM?~Fm
z|J9$;Cs>cq)NRWg9K-3r%i&ve)1fMj)RKpX2ed9bD=jtJ)F@ko4|Us>4{44hx1uUV
z({EWfTf$|)Zf?<ai_Qt{$_FhX5jhf^Be|J%8VReXYiio<j?K+2-L_|Xd10MKj8yoT
z38q8schtip@gT5LF+0cM@sTwqf~o25UV#X>B_d{rJ6y=MA^DZAHYlaY-cxUG&`piF
zhEb;BgbNXAYLZQymPC+(Zg)pk8lg0Pb*Sshfu^NyTm1E961+o5Q*ffr({(*jCREds
zLtvR#>IOtioSw^SLLtdQV7eVbNV-jrv;sxPq8fx)P*q)Qe>pN8pD@xOnj1{}fY{ta
z01-7JYGj~{19epr!USILJCrtLRdKU%tRc{L9a&YRF_Gu7&;*S{T1ATB)(eqTIQ~^l
z|FfTwqT+NMQB`$q8YBXWe!oX$iU~4DY`rC$hO7)Ccx;NLGG!`)_hdJca<O4bWr<-T
z`GxESLau0INZulILPd}68+5ZFEh}-lT)hGlAqFZVsk$2Hp{_qeH$Bd@1W{l(GDVOE
z+jIz3;escisFXpQ3a&OtZ7f-qaHY(WWjUh|0P{c$zt`J@MBBPl3`1a=MvjLMuugRM
zugW?iN-`@#S5S3K^Fp07rr%;K!7>fZ<5WN=T43rHTW@HZ4yg@po$$+0eos=N{;&!V
zqBz`zpl({4wkcfq{D*@<;RLwFA;gi7AD;*zVzgvh3a4HOS$^N+fRd7%-L@!@tBOo0
z^+v66Idglr;nk*LSW7-=oo9|OGcTtTYMJ=v#`4YWEpN6FIeujP^q%S6j~w2e2p;UV
zM)wv|Hz+AsR!{IVZZ2!5EEPWGf>}<PK-p>glzAFg-W?~F2$LK5_QNwuYj&Frm&?M%
z1;(Yk5aTtdEVTY(bt}bi#nAr8hi_e19ES-dGi}@AV=RF}CD}9;eQgQjf^!Sed+N5O
zwZ&K`q$~{rE%^h4$hkmXGrMian};2*9y)AgI6R+An9;5D+p1I-uPh95CdNcWl8r@g
z3|eQF!I73soLAg5v&;+P3a}Enga!!_$QWx;8G`d%hKb8C;{r@R^67_1c5Thwm-np8
ziWZ8dt+7?6ZY(Jk&{A1BMIP?={QkRl9H$FmSuhHW(kyG@G$o!d6GB;nP)u>5GNr|^
z**4tY^jM>KI-FQnPi}3QW>`hYrK`W&bjVy@pl%)zLE&;j*A2KCOu;=1ZIPRMVkiVk
znKUsuh_T?fZAA#~YAY-msv_Qs2%+%_gsDnhIJhfR$RZSkWh5t0yoUdZNzjlHr9vD+
zX(q%14H5{fF0O=(0Z~-RDQ99Xp;x0dv|G?su^FxdQcJWpWTD9<TnOmo5u!{b{<p)3
z{}T@b(XT8oM}*OYb;a~GL0N<~<lLZbMar3~>ktxNe)yJPD|WA5LvYA65(QLudu}!X
zuPbc7XL$ExN%Y>oMJY+H8-99xW_b6Ws;zkQ`P)*7GZjfFmSrNk1xdoE#5_;bx}-!y
zDC2Vw48sMT!3D>7IS|(wl|n(V=6ogSIkdHCpsibM+t6=zEaxNh@d-B-H^96`oNj5~
zd|6b;>pRjKk(H+3?x=3>8OICbxMFPuDKHL4R2%8`x2)^Jm{#QBLceX1f&$x^M{)>|
z!yj?Q$O38dF<tk=veaX=Wr>kxxX^cPaega<h0>=^ptQl%HF*C=1?MzOSlggXT?kb+
zqW~cl>#`z^<?i(x=6OQ;i9T4Qmc-o-Wdufpty|WRaLbBAQuTMReL!}F3YeRYxC}f!
ze`IGY%GQi8&s^TUL*eP}9?)&3(}Jz75ys$bMWjfdhRd1Suk>kUyWO*W`-Tu!f~d$b
zz#5QMMgQ3wAfoW}UCZfoE+js!%L6rM*zch3(aORy5trqUWWIGB(aF&o3Cn`g2{(@r
z7OJ`>g>v&yK4WDfuG1e6Y;IXdSrC0dP=an~Mf4L^2&~ae^T_o4L|9j>h-4DYt2d<7
zqniy?Qy%n6moG0+w8p`s%P53s4E6pM?sB>Yi^}IC$AUj5db&yzsUWPZFONhL`)ywU
zTB-siY1$5?B?$+%rZqLgFpz+@+0wQfY|~(j!I2P3BjlB4b{&P52Y3Odgy5LxgjfYi
zTT~X;e!L(D4;iYe=5D{I>m1X(ayk!Oh6OF*X1AehTXwrGAt#Qf5v@JW7ocim6qOQm
zZAH^JoDV0ChZn*+@yosC=MP`-^Ut<?^Ti&W2IddH$BhGSb<AsEae>{oVYk~55xDbA
z92dsZz-5@q1;v(?Ulf97cgu7fnSEkdJV9BWE(fNNY1$g24P#aupAPK14y7Zc3n?ok
z5}|Slhfzdm*J~tPn-?jU-T22(M;>-v0r3D@TcnKKY`64nMa+rA>FTf$5~D3m+hDS$
z(u&KxphYeO(VVF(!>d<&_WMm?*#~fIB)AMYu)ldl)AyW@m$H761-OORtUe=rV0eBe
zxfPi~cMaB7Byl8PX_A%J<QPjOQ<Xqk3Xz*_O=}by!EszkLJ-!#hwnZjwW95|jKhFh
zXX;Hy-&xjqK?MKDwbIsVzPP`|t$~;Gh4bl*^B$=!qhFbPIlO0?i7C;T$j*R~0xL87
zzTtM$k^PDknYL{ZGO_JCLdZw~UBAH^iMw1Fj!$Tj$m>$j;gumvh1qNovgD3Zj-(U;
zQFsofDzYwQdC#m9Mr)j3i4=ZH77}eX_!#kiE>z2uuQ&9vMoVd=sv0B}Zk-S!7TtR)
zVa1#%lK{7nqs99`$_`~~l-ZE-f=y-Gt3-nS7Hw-{^jEgWijQOYS;-2eDw3Pg%A$>B
zU9S9*zwN`o;yq{Qh&iEzq*Dr?BFC4JdG-i9VGZ2Ay2DCn`kKSzGnZlD)!Vn&?Ur?#
z+1|cl{`j8p-FM`dN0hSkUwlb>bBhn2n>U~FbMH7jeqxv|IKS}n@<`WhFt(ynnPncX
zD$d0C^nzCsQ`ZCu^JPLQL(CGlEX*I@F^mIJz^-c<hYR!NOk8F{@ciP>{tRg=wl_PX
zlK2>TetBV?OZXxrfo?srhIn`+PXq4qp;XZ{RE?qUHng=wrDu%Jgsx&;B<BzB@L3_Y
zH|Wid$9Es_5-L?-8u^+^Z>owMGa)SH0V5PzYLfTF6iB*gZ;ghO1jbn6b(|Pt`4DSu
z3o#3UHCI4|wJ<LXrx)tlf)|J!|F{;I&lhB6+3tF>o6$K^#j#XoG>EP%AM#iLAllfX
zSQW*yE?huY2Esy|j!ee`Sy*Jd<2*TJRpVDsS<xfGO;|gT)&=7x<Wj;jcGJ@~3L#cZ
zDn0A$9Yl!r20y=W`S20dZD_VN%9b(Wez#$~jDR2pg{~@8-C>$7wyFtD&oT|1jxW6X
z^<Uwak#D~Fnru6KhzM2mw%T?mZLcCC^~X`K$l#}ib#kCIw&{^l;HQb41#zAU-jQW7
z?I8vJctB(@ZB2*{=NB3T%G4k=!LO|ISk%T>eRYfpvLol^k94tho$=!tYs-|ud55tL
zn{8hX#yJz6!@CtyzSeOqAxrAr4j&?U8PGXnY>hM(A%V{sWh#6Ou*{GW?RLZV_Mv1&
z5Tsnjx=?oCS&A$ASrS9Sxq!}vw;*I<Js&w9P8^<}Ih>}l0zgn(LH~M(QDs+~fW<8=
zYY8si?>ipu?`RrJTUoS>>~}p{Wc+f=e7@!Qbil~SXSX}vyzWpkus(laeSTy)%~!;C
z5z)kTW;vb^36d|jT4fccu9#QPJWM3Hg0^I{V2wqIiB~s<)@pv%-&5I&+f7T(U-Ebu
z`RjK_zI$1C7b9cUv&NN_1U?IL6xS#ERep1=Ui?8A6GLX69m*Ik!<n|-Fs_cpIZ}>g
zG7tmjOYvKz>k1%6gP`bj1qD&8l`aNORf2ghha)$Ai>?ITEi|>{{_cjp@3>qpT!sq|
zxA!PxLBTQ$9!~>XZK;}?s3m?HC_SD;({)sJ#X2vf;IOsD7+G34t41k-1jh4#HInV@
zI1Mx6D&Tqi@jVZpOZx2%Sqo(Fq;<ugN7iK_#6-w~c@1<;&9*n#+A_~8$IDDD6Fw%E
zHG(d-zYqdRWZF%k0|sB16-G;TyPgyr&rdJh-|q;^h%z-Ptat}0NM1fYv7RpA6O9!Z
zYe<cy>dW_-obf)AMX7j}&uOVF7b0d^)krA`p{yL6evej$oJPcTwI~%3140!KK$NO5
z4C2D&JYej6Z4(r5Ta1AaR<tz5VoA9?UWSpF6Z_lGuw6rPny^HIKeISTy?s?!Iz}V)
z9_5ZK^FS6Eq@`_daPy3;bunN5C)5f4)2xMS;CWhzQt|bdUsBbIyexPTu(vzr%LPTI
z)|oiI(AOJQw=fScJgyV{rbP<D_wPR8rvckol!u!)ZxNd<(|E)$E1C!~a{Bnf>{dj~
zXfiTKRJTFv(kyWET*zi3Q|)f?0wBQ((;DzeFdi@Dc%;3#pp9ZVX)fmi5An|7)|K7!
z6TAJsBz2n(p%toWu}T+!&KejmEBQS0U~6hC=(-9`W|;<Fo}RFEL)F#D9C70bDax2t
z)s}v{CClPQukN;l7#Jr<mZe90Ii5Hyj%8W6y}zT~_GG1yqU^9lj-*gPOj?N&vXc_$
zN)544%8KV5D6E8QKUo=#(GoX%+~SdyrD__`#Rg}4$$UBR@rUorST#k`e4*}pGOEaw
zq$IB^mCE>-SuPjie4*a#ktu+Ste<}1>9>D_OOpFP`A2Nt{*1fjXZXkO@#l|ZC6Otx
z4ks`Y*%(qKSyLjL5-xiB@g3{5VER3;zxfkv+p~-wQYLFn-PicEvK&X&RBA*@X+WXs
zx`aHnB6^1uD^VFtRUsv8wk<MwHdk(hl9G(X`+!sip-S>Q>4H%$-gHxLd_t7KVZL(V
zQKaBm*NN4wXrqxvmfNx|{o!R^n9gUi1armtrr;p2_+={4m*5zqE5H$<Q4*9Wn%<Zp
z`EoFbNurF!N{#T5)AKuaclR{g9WF{_RbjL#*1LHjhEVqUS)jB+C|i2ZF`$Ud>&i@-
zgq46)D8j4FwxrWT$qy)F$__jy1PUq2fhEPca4*&strb2v&d<+0eSG0`m{}=1*skrU
zo0^ys!+By}SNiRazHhkMYc^fY*Pp%Ov$wBkYKzgTh;yRHFC)w3h;7Hi#t_$)+g-<|
zh4s^qT)zJfcRI5U6B&)|G~2eP-L~}Go|FV#%7x|^3zK19J>F+f2Hm#kx}vHojNVfb
zP<}z#J51Z6nyT;#hK2vbKl*~zMGkA?+m8qS`lkzj^<m|&zx|2hv=9X7>-Sxxe7*m3
zr70#-a>Nu_LuNjmd3-)nR~0@u?sgr>%<(j`dQou1%q7fb&2=^8F|sPWfOTfmcRW8G
zP(*HTHax$)u-VjTnfU69S4_i*^Pb99=*rUH+%Q}UU*-8c@TxO3eM2H+o3>0yeNj5k
z^Gs!m6h|m%Hyy?r+&~Ewde6EnMLJ|OcYVVeBC4)%A#-_tW*tV9(Pdiaiz>P8JE9OA
zPh(*iNWpyl^@NbPkjOsbq5x$|{vt%wb;n;D%}sC7qSREYYt>TPnoox#)+lzn29ehS
zTv`S5;y8SICQX6lJ)1A~bbUj$?J-@CtgCB2rZ^yS4y@N_X>Ds%EE78^OABC_PAFZI
zgJkt1-jCF(MaT+M_pI}XRt2Dxq9dhLL`PTZpbMUuBE~cb!Sy|+NU2=-Vq7p*VPt#d
z4%CFavbqbkv9#@`Oc7%VJ1Wz%PQHW>m85HLFjYqmk#!!Jmotm!?}k7)&yM!y6?Z@X
zQ$G9R3!dNqo^kk=*PlJGdHn_Br3Am6%(!u)3Id_a3o$9hC1;whMM+89Y5IOcbe_xc
zKwM@>k+iOypPzAaF;!ho7gjgZHg9=)|ACWv=JwUwqUl}cViBuqblq@%eB#6R-%|A(
z?DmGNEYetlpK#N}6bt3e`B3)c#<01$C8lDu^Da_pMF@%P7HX~Onhl>`&Mf0hEh7H(
zjO`m#^q?du3!V=HPfy?R=4W4Wdw*Lv)G^?*CQF4*fM3b$Lhz1xDs(Ti?eRI2keueC
z%DvY`C=^95hk^v&t@vCjm=dn3*Mb1n+ECYZ37@5O?IrU^>(x3hNTF!jhUi>5w6vSj
zAJ&%L+fA`gUe2uZz;JpYWZ2q-uB#HrN`=>+qelpZ)R{D$$<IIHNwijD^2+vhM?%xq
z6?#3BrvVWILMoQiGZ*i`)VS^r)7+Gi9TqJ!A*>vi$U3a(<q2gB_g{R5lnu*xVD*lE
zv!y~HqAzll;NcI|AR$o_mf=ha33Y40E7E#le0tCMhu`t$^+REN=gcq+aD|X;b~~~Z
z<P5c~%MCB5GRn85L!45vSqd^?z48HK8Ld}Mhb)+Sw5{o?25Byc?5Ua#LBh|MA`rP!
z8j+H8ZLSC)xsmKAX20UDbwwr9e@yu&NItMmBYwDW^1dAWu0We|i!b%b<R-|0w&_p;
zlq@dwC;~x)Zf&X4$^1w3t@oaB7!WxV+{!v#5F*nw7B?TjMTCgtw3N{^masT-RWV#9
zK7M%5Ff3HMK~*g&mqW$nJTfniloMUovcKth_4*atn;qLt!?vrbs~W2{AqJ$a$U!1w
zLh2pKXmmZ(7{R8m@y8#SzWXi1hi8^0U~u&JTbf;mFa~7|F_goUNC}Y(%fH!{gQC<5
zsV%CiKnTL>S%wjjGs<>kDR5a}l0n*@`c5G-SR%bheDU)){Ff=Ps*b<<cOUqF{<B~4
z|M~Vk$IDp0b0oBsB~PX#l0a>0Qiu%m!tr?KGR%DS)fddugtd}wWq7^6!PuIo!+`$-
z+doIV_Xs0d*AjNKl_mIy){<rQoG&HAlB4HlyJ4Ni;ua9HG?40=X&9Ksp^WKMW=#sA
zCDv59dBI;MmSIF!75!#|$Qf(?pjDqpPJ{DUA&E+}*=>m-F`frHTeIma_Mg21QxURa
z933y8o;g3CSmqVe)?^`QHXU#7Z+Jdjm=?!0IzGOfkYw(5JwKjCE~_VB{~o2xNGVEq
zHhE-{q@)PmquYv<0@^C1(L6mL%gUf@kR*neGl%C3VV=3!v~2IUR5v^9ZbNgkMVbb5
zO(2)7hbYY#EfrnU5&gn+cp~JCX?jwaannH4*QB(P!iDG@>wH1_P@dT#LarbsM2a9v
zrlis>Stu%PSl0pX0!kO?nMBc49jh>aVq?3~TnZU~Rk53@17V1=LdgnuS%@Ox-A9x)
zG@Gvq#@u^Eo{7td6d7r3;gbAs{O3REFc1*@@*n>b?tb<sxYZG)Wx6c9|LKYCr;#uI
z^5=BBuNf~FKE8ZUwGM1Hd$M+A7o}<<lBV5Ynwn|4u&$nQJo4T5KT&B*MZ%;=cYnj{
z&mQP@4f8UyIFAv<<`#r3RI|k)b7DH2c>3<AKl-8j&v)$JJaBmbfwNx`QZS9>%Y-UK
zs+=P^25hC7FK5<E#&$h<S;?{JgJlX-M)T(3mff4rSWb^HEo`@+p;Chgo}2xaxw=7B
zEls-tX^BcQhlQr8(L%Djy=Msl_wtC&&}=MAT#?pdZOyu@^fA&kcbLksdROXP^8i8<
z6U-rS9#;g=*VkM?OoS9ssw%d#5HPl&bG0g`+B8fo^Gr+uB}#CAIiE<42d&6{MYkJv
zw+}31g}1iES+1g(lmbF)l&Q%nBa)}7>%yHMPb3_|S7>XgyPkPjIQ{rrv{Cq2LOdd6
zM7O1~HKy4j+bvbIh2RO7BXK#CQ%3eXv}q7-;dFXoSq9qvhO9N+_6D0X2#u^Y)>^z<
zi>OCP1eqKJX`OM)j1+>)V=Q{<$B%UVhPyYPU1cp7f?qLhRe0tkq%<HD#uy|5zfL6Q
zkRp@9O3u(VEjcD8H(-n*D1{W@oMTxg+NP#46;f%8wK!u~<`EY?J+KbP;$m+axDv^v
zt_b6J<-jk*5Q#Y#Zn@Uj#*kgWt)8SbdUu10r7C$DCvNWUsI)5j)tEpeyqnRoj0lwy
zD49VCv{8U2W(ZkcgGr_8R96*kw<#+U@3@4S;rS<`m#o7lF2@f%+<t{I5VN3adzPF~
zw&M2HhV$iukb*U|9Q_r`J1v;X(rsI|n;tC{rq&Q0^X0_&@`-07xVgQ<>eBS6>Ix{n
z18KL&kTG?j6$0GA^!^7Pzx#owk0V)W_FF|0J?Ux$b#viM<e1TwMK_jgby0{bMIw`x
zCP_({J=3_N?FL<0VhTvzpz4Y+PsC+KR1LPh#opdSSXfVw$je7+7XJP(cD#A`$Nc4Q
zzvln`KmS|aeL9e20V`#ZDCe_8Z4DnkKGJm!&rc`fb?hvqVAFNTEXWju?Bk)V8*(a>
z%a{`?QYMYpQ0KbN<z>0wu+9#hGnH0YQ;^0<K*)*BW`i|`X`Y#<u~g!8#l;D{7o=rk
zy^N(PqDqw$ipD+UOl2j;>e5DVnlMjPR@3iWvXI<7bog;aR~6boy{WJ_uMno;%hmDm
z$9Fvb@PXsuj7S1MPI#>lDRDU+n4Q2S!QuiTW<pliYH7e~O=C4y3T|pmXC$rFG?gU<
zcdY_uau)1w_q_Z52OdAYu-*17OQB6J<HFs3!)CwX_TiSg-(so;08>|px<_b1P?<Fg
zkOpK~(*)-Vt3Vh8nX2j#$^sJa%Jj;Z4wVJTjm+Mmv?}=kA#iR*5K4PMOI+}PB<GUL
z(b{6QAtq3!qLz}T-65pH$AHw4L`2Dk0G!TGXqg!<j(J>ZyB3gWS+lHyamvigz`hx&
zRB57^Zi|um@1{+VQK+0bKR@DQ;{5o5ZC|syz2U<T@A&w`M{a)pr)+-y=h)X@FungC
z6$AHg9~h>A)2AP>_<~FokrV}o^T0H&<SYoXKqvaH<<;kJNlMWhjhYPmSN9McDtZpj
zFHA4z!Z@ikKCCpmU3tlm7xMWTFFZREORaHhWSujf$jyF>HkM_Yu~J@BodK&QH~oer
z1=BFGoKAS>c=bxqZ4GsHr6rLw(l#hrBi)Ru8gBYX-0uh}QB@6IIEI%;wBBH~;_&W!
zn*I(Y1hTGh%UljNZBJcyeE9xXq<LV!d0Px+qI8OfVJv2|wkORq&z~L(l;iFOw>XHY
z*s-dHlv4?$<q~WwJE_33E=<E2TU`x!(Gh(DtufLv4JY(cBpOms9?D2!N{}dx1Fgsy
zqK51RMk|ac<iE5goT$*v1{EYmRmA8E+g(s@Ki!sY|A1~<$dH67?s5TW0Vz<aB3L=+
z5mi_8eW6K4p^&lwXx3#xvcQtbvS@%)45aJ<LA<iu=V>YkR9)a4H*dd0X-$xts;*&O
zP%)#k!P>g0Sd&2!$nynbEmCHDEUHi?1xh4TTnosCNZVGWXFbhK!@%KqV5%!_V%hhP
z=aNgwBGYyq<I6M4ICDClsk^p7f~u;dw8@B+Se6x`72R%2Wh@~^WR5J$z_K`W(*vUL
z7<2(2DUlE<QE62IPHTXwB6`oVEGVr}iZazfpmRp#Lh1BbTn&RoTRR_rpl*6X6rdsa
z6Q^aMX)Bbe7?%ro-J6n_*P6cCu*{AyU)b+9<d}GRH!%9L7u~ikUDu(s#=DurdB9XP
zk3T))+|29SEw{HDtWk_H(r<29y<>F_Yb#W6bn#4=kI1y*9-sK|H^1k*?+<)DIIK0~
zT;tN2eoVBR3g-hc<Pw;Z3TZ(lfryd$G7&;V+8XC2$EOn^OQfDb3yjsg`f`h`JEDuE
zFmpMaIR5rM^8OWn?;rdn^6nelGV%D`?{RA;nuh=GpZ*zL)AN7*zx~ez+bbiWuZZTd
z#v7-ZAAkHn43WNXS(g=KDs+OIZBL9xzIb)R$CrWm>T<|Z{ZU=aA3#EGrEMCtl3dPb
zvYWZzc5Jq{bZvz(iWFCrD9#h_R-O;f#FPns#iz_LuS9zKu7_zROM&jnGelPgV-=DF
zDWZ(Tb~T&Z9l_0r6j8?DQz}^Bo9pvRX-wUeik~ji+MoT|&v|(JnvWkIIXoR$mU2j~
zYeU<0yo_g_jsw%OAhcm|>FSb*C{qpsySApc#r&v*05ZEx+4n0c|3E{hZCd=Y@bSY7
z&PO(zEp4ya-`vn`w&f*lO;Ly&O%fSF#z>9W+#vH+YwlMvWtyo~8P^Y&BSIN;)tB05
zbgbE<j3qix%$cs+qeMm|jgSd}hZGQ^Dj4zTFwKq>1%`C3vP$Z<D~CU1sj6y$lvSw=
zuXD)_IR{B`x*XYT?zww-h19aN&T?eBTsRI3FQ=LNU%q9t-<8RO$i(x^-w9{n{f7_y
zfB%<%##ev-7u<aQ8SVC#+&ATiw{7|9{Ue|KgMZ4l-LbBv))0M3{omd_AeMm-zyCeM
z`OM$@2Y<%R{R3;1q<rG8?z#K?ErP(V?U{#x_2pC^XvSbsW%u{;!sYP=bj{sYUsLaP
z%*zC-qNys*KYqvQ(+3VuANlOFx3t@iIvYe1BrVZROLk?<t^j}CwX3>8>f)Vt^UO4j
zWL+~Rh|{I4BwR^lkCy}17@FEJjRU7o@5$cL7|U{aW||kOs=>B9!a6g}C$epk(jZZY
zywdG=?7#e+6rk%m4(Ah}K0UI%-Ll!=mQ-lU)OB6P)gn{Z8j%xC*N_CP^GZw!PcBXb
zGV|nd>slHQ+E91xAI)@C+t6>f_{9~GO)d}+aos5iWeBMx-HkTHm<Tx&*V3<zQWGg;
zuBz6k)fe@w4Fx}KG?kVhDsqg>^YlkIJ2~O!iDf*K^HtL17t*?*a-wZJq||gxhrP;S
zgfiqL2=j$`9LO=lbRqzz-O%lJ$XveEMMJ7+HhZegEtz5(lS-Ez+B6W8BcvIYspKu1
zf-t_E2AaB~uL2>Atkz<inz}Z`q)Be!a(qU{P;M4E6M`&!jd5W-99hSS)d!-f+1);{
z%7l{+Eh5V}Gy4<cICD79bfJtDL)X$aHHc!x3qDX;P321hSP4m7CZd}mM*MIBYtXtJ
zkhFxf3<M!cN?s{cj>xKFP3e!eLzy$lEl6Cs9m*2EEOp#6RdJV(_~*YNtA<rp)a{PO
z8i<-01iPEJoGwqy4%*G0m=kWAiObAzJ~F#N4uZB4*t+8EBdEyX{EW~RTx8c<eD>@%
z732-IlDOf*V!5ikC&qcD>3WQSI*n}o6XEGMIJa>4@xbxJ3-clfA+dPRxOhT7^YxpB
zx1Vk4+lFa!q*|k`WE~yjc_f8^n^(L~#4I^a0jVrW8eCeqERG~IK{tHe_o%MOfb^XC
z_VK{K{FncN|BIIV;;YZEq`ZbWO+24Y+<vY2@BX`g#BYB4f&cfv_z#dvQoQ6+%1yJf
zigB1qNHAKSA78k;y(K13TWR{PqG?(_KA(6z4a`a~&MP4#+Nvh_fVfVajFJSmGR2wJ
zY_VFg>1wnT#1L7Rxvb{qnbYM`yeUefD?=+at&yw=Zt9A5)3KyL43Vm~Y;Sj%raWiF
z6_xDPm7F5xDn1jYqR|#V4}|Gbh@L3?>_7}iEv_~5mfd|%yQ#Q;*l~C|ay*`x=R{>|
zHoc^s;JA2VD*TL)5|!4pTG1Fy4vucKp{vW}rnEHj{~xY-y@J=4#a$J^itTn&nnRT)
z=kni7Ul#(XDXT&i12RPf87Vc1TnvHI6H@VQ#AU=-bHy$Tq_71vlvYym*HuP^&=r0?
zk^PLVnu14;0j&~*vNp(mg=JzH2eQ$$H+Q6~s7OeSHWBB>vR;y9BGGmm=oGPAkm`x5
zwzQkJ9N_GRltXDfsEW@&tGL@W2ve1*gI|#9?`SeGua5U0-l3X?VH|n*@(V<D$Kk^#
zMAahn1%G&Cyu2{KJahW+J-huL;U=;hQ1ig@a)4CE7%oJre#cjDz98g8o+skC;*}@G
z%<0n;H?LoD^V#RPdFJ%t1D#R4{^D~s_phmLZV4%q*O~eBOy2bD?(e8?_FT?KzWd>O
znzrNqi!Z1@dqZk&P~C<+orvq1^UIOl?xtWW6C6%wtSq;v?Y5`6|D3AdvYtMntYw`R
z5DtHQES>QE9#M6~l=1V*@zV>>&(FO5;tOuxeva(#5Y<X0E#~%t*+-W131K4JrbDG7
zzu4Y%2>FH>9H!Z#?KRYs0x7uiuosfrXkrM=!*CsuB{nxVR8?JSpKirFhopdvy3Gwa
zIMQ%NDQIegBI9BaMFsD#s8Wg21|bS4B<IYsc%n22U8BRmc)75~i0}&{SfYewnTg4h
zyknhbG$mk{LL>yo<$NNgh-y3BmGU;9o(mo{MV!B~z?DQ<gHX1N?+`>E(UdGeh>0~x
zys+f;fKnNKJy85%qVb2rk4mZ3Q-kMtDf?I91FaM&(zT8RNeH4YU1C4IqlymOY_M&6
z{cHvZz9i?Tf%EZ&93o1b%MMeN!_%?`hU1Z>H2eJ>cVB(O&Fe22hneB|Bg^rL;qe`D
z_2{PMq1&?EZ%g_-1TqrSmRp)zXVNqwq^7PK+`5wH1yZ827Hw;EQ(t@I9vwYO%WH)x
z5o5xwO92~6N%RiCURB6K5R)f|wfwBrA8@M?nG5$HStF|*!maq_O!A&(Sg@Nd2+MZ!
zrab77<tF8dAgb1a-~$f9n}-HO#lQWV_t?7TzAYN^AR^l)F}qAH3k2%%{WDEf(bW=T
zG^dwO1m|gFqScnInuw?OEKiRN&nM330!6AT%QnHVCMF*^ERpfYGtYD4?Omj+P5E4=
z86itsWE^JJ*)uK<pCL(0Hj$(hYlvKyg7a;Dc+dUo2dYg+jG4pZ3-dVe{-+Oo{Prhy
zm7(i;%>Drpo{5(OKEi!v`5*q9U-I4KM}G6}xlG_vKnl=8;C&&>36W^5#??&`U8TSn
zBfI^c+g;D+PY0w*Jf9YP3{<tnr$klTk~ztV;5}7k3p^($Hk*#RQY_QJvU;Xvp=~>+
zYlEt(8<Hr>W~B^^o0*fvnhM#~h~Ubh`kFo$SHFvpqCk*LS>4FGzUCq%$~5GdAV$bn
zKgAM|IT1xf8CCS}+8`v{^qSk<8(er`xHw)8XVh<h;?t=(Lxv?YyVAzd2pVf>jHEV}
z<X4oe%fVKu@(i4d>Umilr}KrTvfSO>V6-Iq$YxXV`Bx8!;8=a2ik9R;S&K9krd3E4
z$bLXxKGJMou}VWqa2d{Y*8}f3JW^F1s0yn~qatR3){=D|OE_37A_-I1qzIr2o>+ii
zM}(=Vnl0-zB9g;ydSulhfMq<<G#x@$NEAvKthPvL2n3KCsWfEAcDteQtpY|h5LZN8
zsZ_-58tQsSWi?SKbdJOv|F=4d{?iWwKmX<}KYcv#t6%??pSi%>pMOR0fv5K$5V7E3
z|NX!F*O+<6YD$k)Dr~<+H5MZSxBHIg^T6rpV+mF4UQsta^W_D9c|l!A);_{dA0F{O
zvw!;~x$98d9nP=l2v6_c(VdU1^UOS6uua24!4zj9sEjI1`*kL*6EZ87VWRIgnA#GP
zL1@J?oY7ibb*{2#M5RDWfjd1T{mMEG2yJmO@$&IK?<UXJzxV=tBsB}?=QFyJOb(V6
z&hyOayWjJ*v1nyD9xgBkbh|~{hL^)LFCTtQElX9<wwBGNr`z4q?)Kmvr|E~{-cE_*
z<0HCi=$e|~Jj-Rktt;K8Ly<^+Ci=3Ua=8R>!nzQq0dgd|nbj{0hZl^t1#F`XF(oQ<
zg@Le^)3zxjym|F#+hUrGP!h2Q!f+;sm3dXT6!Ff}^xFc|k($-Hl70_GgyQA}-Bh4V
zSsh#tD>;`sovk~x)>u=(8Ku%!NMI^O1d%KR2}MAXsutPp(JEl-hLp>`t%yPjNif9~
zH=av!->uioMIv4WASwyO=7GBIkxSt!{K+rAV1IK{MyaXXrn6rO>r57&gh1FDRcY4Y
z!Z;3P#Mn02e#7R~9XU(dzGd4r)G}eW4Xt^@NfrQ!>^iEp#o7WGaVe5@L`h9RkYt7F
zszOfGkfkE0qTa3BEg)#?78gA!MYJg6$|Nl6DwHX=M*~;3hMOmZ6qu?)RW&JOLP+H$
z?mS6qvbLpIHjU)rh!27Faw;IA5RtYcQ`!_tDnu^fu63P2L`-D}Wctdo+4jURGK~ZO
z<QF|{TXUH_Ne8N?C&k3PB&-%BIg_`Fw_krrwYleXdc<8$^qU^sKqGbtKa;#iNki}y
zAdA#!9og=-v<Z&$%KCI>37N;UV_KiN-)^|uR0ZtSR7|tue3-aQ3zy(=Q6WsgG>Tja
z4ufNGj$v6*O~vcOnJ;hd$;PAWhRHudinul6mX+@I6>6iAhmo7z7Hc)$FT8!-^WXgA
zulerdGxJ(p4oV3kum(@1OJgKmiG`QT1tA2Rw&iBGhaBiC!_BrOBvJb7F`~6%zuT}Z
zD<N?mmj|rT*O56KP8Y_@nXYLO5;D2KbW$Q}dj(0UqKb}@VRE>Q#K=N548c>i6}o9q
zLXx9HNJSP3(^aGpt{e(MiY25hY-z#dun>G8CeJ#gLRu9dtwo6fW75_T0i`vSF?{jL
z^4Y7O^B3RT@W1}+5Bx8`I`I4BndD0?b9F2G-3BXCdG1}sX)#8ymSGt2wipp(Ow_xI
zH*fB-Rx({C$buXa+t&~Ham2Zm7&C5OS(cSHCh9h$OhN%s2)g=~+}#xurcedXtSbWP
z%3XjtOlTQOU{}|~?1}4uLK3fT2}Q|(tjmB<_UaDEAR@Z&$hyMl9+_9Pf|LtiM4K8!
zEIFmL;zNY&sFb{39u~B&5jjFC(;DX-b=@Fr1NiG2tSybA6!FWU5EB2k4+Bk87nbt0
z@bvV^5C7)hl0;^mPT02P_Thok<0HfAi1(3g-?F*skZEOl8SwLp>RM{4d3riBT^#qX
zXViRQv)>W!_8fm28K0l=($X|7^X1I%{?)(Wi(mc>DKjp4hVhI@C852p4xbY@H#exN
z!i_T~CAv!CcU$~AQ|HVwkNoiCw|tQocH5eapugE8{7j=ImDbd@Q1g_Oc(;;bL=do5
z#rbe1D$DI>Uvc{I0aw)|W%=>1en&VB)U9P0FVwAJb~7?(#+Q#YZAaHyg4RSLewwf$
za9LN5=Ob_4++h)DRdM&_pVHsIV)*bQ({O}b9Nl$W7t&x11m{Z%_zxa#w~$B(p$G|b
zzDjC_p;VHLCZ&W6Fpo2Wg?6{4Y7|CSNU5o`#a){YeqM?sCVJ-KKv)K1Wv>UM#2P(A
z@Cakjwjqo1fFTRKD<F&{B|#dhn|ta_NAe5vG!VW2W4Nn?H*0EbX>CPv9-keyYf(Aj
zmg1;SLXg}_iV{;*RGW^<DzqrK7+YJEG3D?R3up;t5h-C!!i5A1R_72ok+Wmd?%8O?
zyj}?Nk)&Ey-;!00s+!U!5gI8CDlBEEp36!}#)vEudUV%eH`|gmSsg+qmgf)T;fSqU
zq6=8z*uQ>5Wh+vcna@X(59rz=q`}2ZHWk(Ft5P@g3szpW%sHdm<|@r84ghk#?n@Qk
zIp*PlG1ZkTB8s9l7gw6GC9xcXXNj@oKt$<Khv2|AXxoC)2pUiwF;SpdLZPb`?;S$?
zA+M1kDf)d)x4mVZH6myBn~F8e7};@n8fj|9e0f55x4eG)8s|K^vW0-D6IE>pbJ=MN
z9cdfQGHa566c()6AhIXS18y8yrWKb2-X~TU2tr`C8_pk(2r2pc%U4`nWF1E8ZO3tm
z7_kznfSwj&%$#S(<O9$1#2gLMB)0X2PmeF0mO$O_ImC)4Y&ph?o6%Ft3hJ7+ZTaT&
z2ljV2^!rz|+Xn_8aO;Y#G+~|*cRP}o{Nl|m4>ucT?+c8JV&SW_E^9d?F%`+wFiyl6
zxxLvVfO%Q5wgk&+<JdGc!@RJ&+q2(pO1M-A=6R;AEq6C1ll1&_Bt^%jZxB+_v_;Ll
zt_yYDAVvB247k<NG!0E#mjwLk$_lY5COl;f`I@~^>PkMmLbj6kBvDL>Qt9#;5$YOv
zHmr*)&7)Xuq-k8PFq~`ME@!lqXrr$5Ud5fM_@Dl>2mbS~I{x+VF8s?MKJop_KuQ|v
zBv}JO7OP|utn<n=t$4T6)HQWwXq(ceIbSYpwjJC1TVjN1S-8=L_IA&Bc;Rxq&~8hB
z&@YbJjwox`?03jz0{zTJSz014Sf-c|BB7HedWYTAn4p-(32RFWz|AkD5J_T1%FMD%
zWWZWU)jyCBWGRR-Vw*ixyC+LU@FQ-`RP_caHPTxAbY?uiFF-4ishZbimE{65Q=a8l
zDcC$ukX#`(I)@Mzx6T-C5HXYe%sPDH`NQ{Dn9|?!Vc^r#GnEm%d3{UU_x$er@2E@x
zN1WCpF_dv;v)_@IsYn;x%;FZFJ|56k(LLPJJ=B~(JtCt+`U{t*iJMK$cpmxghYth+
zKEv&sJG4j$H<QN$LO@&BxM^aY9iLtXp3fIHH(Olz$o>7E{kEg73`mI+lGCF@N~kw2
zclUcXeNEjoh-$<A?v^+lNb3NxfOszF^Of=z=&Yt~H0N<41y6Hx$Nj6%xVzb+nvMgC
z?PrSZ-@nDp3+?bozPtdDw(r^A?67r>FcxVy*xjC|AAjKV!?*aba{KCzyEkvBYlBJ>
zl_O~xAbPeBU!sj6E(7ayLL0+0O@svXc0;cX)A@*-E+p>@q$YV#g$tl$Mk_(^j@3n+
z_spwDN<l9qGK;cH_70cQb%$J%yh14IN<oTvzu*>!ttzxq<hmwhiK#5v7x2)U0vd_b
zg2j1sR@7Zj)waYOIK6ygUDx6Sj}hB7)T)pbb-zKHnmnG#^UNA9tg3*K3M;{uzr$SE
z-Xw|4sf@o;A}tHTxsvI~kd-dTP^rmzB}G@Nd$ADE=3&5(BVFGYx?<CC7-oiWz_wdN
zI{tCACi0b5DJ0!y!!$4D0a&-?h7E+#Bb6YI3(>EHwlocly_)qlx7>dIHC4Z7IXyC5
zO64#*hg(*vx<z*l!Dm>Pg2fh@s_79b6&t54a-ZOR*?FpBwp*5&WtuUjOk9LgXk!T}
zvd$BNjL?C672cF!WC1D(s}WU45~ldoO{v})mAH%}&L#ToEm~WQtvDTC*lap%-7pOk
znKRxE43`T=DVp6asZ#iqIG@jiX@a)J-blvFg+^?ttYMjlayU?;B&Wm5@cbi9)e)>=
zU1wzWBtNm94n&^_ro)Ssc^z4T!)&*xU4@wX0#mY*YMcn(^X1?B3x4xgzvH-!9EY*w
zE|S1YO<iv=QSovq0-opbC%XGL+`hi!mw)k>{HwqI5xd{=;p|Cbpl>REdVk@#1nREg
zWtw;hnNCQuFf^MD=V9XG;}iC#=lS^=nOE-iJAQaR5K<zhP!4Aq1d7TSnyzP@W?o(n
zRJOua6|*b4=c=w~>xPsI2SX~!>-$^Y+~1Z=kSmqarmk3*5-!UjvgtSMwq0o&Xo>fp
z%38>Y%kh9vmT_82Ragq<RS<$g3lQ31tLx{=Iig=tN|U5Qi4xMwF_DGD)^+)ArSj~E
z&H)0cD{dZMGarvEm!V7#QY`rET&m7WYmyR7Ls_3xmBq*+lDg6G-~Y1Vzx|V6@?nVl
z^Y0G)+n<g&7a69x1n`7LONkFFMwMyq%}q~bHNmZ%jweJ+RJx|!Za^#MaiQwU3UM80
zmd7Jq(;yJ6+FjQ#P%T3V0IE{eHC02UEu*l6rbg<Dv@C>>vDPA0nL?%<iVi%jjKjGg
zj75!ZUXisYBu7;hj5F4-<^@el5&~^2+&UwONT-SrEvEvk%AVkh=`$yfTh3$zQr9$f
zQ6SIjf!bu2>3~iP+H~aPA^C!#_6u&h6e{N5_F>?wpM8xMFsvTiZ~2fCew<k@6T><&
zAJ6RW9{Abc`%~uU7u@j^>*Cnn-*dmcqmrPtVDS;QEpaVlsPS~-yTADrFUNuRpPu-8
z|L`xE=L?m(hDo}fn_Wv1hNqVk!*O6fPt0pZ=^|nnUyjr%u+fRS-Sd1rB5cRQ{XO^l
z4cS)sDA+Wb`&V}$Yu0OJZ@paTZZ>RgdgkfE;pIql4zWrq0WwAQZ(fs)=6v`_oE`J?
zf$rfe{^Gy<-?M-786?5*^nv5={yoFt2Ym31=M(dI0VPrT$ksb_Sm@e{O<%LWy`|||
zVot1X;r*}w4G5^Up}o69>z2j{DyuL}&3Zn8TS?begBFsg4C6Q<hzL`Iu?VS{#v?>P
zN_@!VRG>57d4}VGu+A8(i!??Se5bJuo6VLiuJDsw43vJJFj8Tf3KPpHT}oLhK1yJ^
z4Su-bmx6At`&*=KNjc-k;fl8PMOWN+G;NC?7wUe??%@k;U9&uWAo2nTE~gVQI{LoH
zkT6z5_Q?4JGNF{lYKuUXQ+X_65risL!xXN&PKzMneA$Vv<BXpcg7(n&Y;M0IZ}zC^
z3~pjPpEy3f$J!cmtuR?rKqqyjsje^h5ECdx@@0Q3tCAhisM0A`lGO#ATS?Z@hJaB5
zy}RZ1?H5dsABZ`WI+<-TH?L5eTf{lBMsOsot%)AOd@5C^;7K`QQlhS_GD^&utQ1+7
zs$7bRNO?GM6$&|54jw|uBCJvdv?XXo^d1q5p4b}6a5<6v#OfAo)fVDf6fA2%nVQSE
zQb&svf)vY|z}C$*<B@S<q3ueN+4Xz6&6X4>BTlIcmp>&^&bW2L);oj{%+r}lmM~aV
zSA=!qu&l&kV0D?;+~Rt}I%L|nx7fRTQrFPi16gUD58T{7AT}FzfAVwW4+lEC=jr#q
zVS9fIO;6Wv`Qh8|Y1$ip{NdO9=`X(K%dfs*iUOB3aX#_)e|b;J<?rR+{Pnl|{1?CA
zoGhbIw0Cd#a17jECiMA0Ya6`O{P6L>7!<i~`R+I0((ZP=TqZ&)!DJWK0+z|p_nY!U
zUmb_Ti4YR|n>#X@Wm&k{_8g99y1plC#S#N|Z{G0sv)8mn(bX28Gtq~F{LPlA1ogHp
z3VF98q$Y+$UDYTlIUQbzA=2$`8HW*DmucRx1Uv?-ED%{kz)~6lq+Id=swyw_T!^R$
zc~xMq<+-aR8bfd;zk~(R)JV4yW>?m2xvVTQ0`D9?m8_AF0$&F<cYCZ5#Cc)$5mPI^
zexvyMn{W6>?;Zc_-+s$?KRw}nV$)^Xs-mqdR)fBJ6ml-Bl$)D9))=ysBq3<Jj&&GW
z$C0+NRGXf(uB>ao)a6pBtVY<5kUe3YP*P%SLX**MW}_@ACw#MK6^5#|rD7;TnQkgc
zilCLnyAf$y+WH2qEFr8#Y?mA%2ip1uTh%P{h!7Sb1Bj(Sw!FjGmXH+Iw7B(%P?}|(
z$#KOm7piVY$c}j(kXfNr%{n_ShZmYYU>ZqG5i81?HKpQ!`8x>^?rv`Q?2EU&`|yZ&
zS0!xlY<3-&Wnvme?)G=+oKeQoZT38T_JHYXuodob!VeSmrpK8DH=OzIrxQ<~PJH>*
zJu$=WeuIggwyF>$_Pd5nR}+2U{rmUKKGE%OnH;3?Oj}!gNQ4j(O5uIr_;TWOxFD;M
zS6{qg*EeWw*?jSeHD~5&;q?3tRn=&1h#_EYsk_y6jZ_x7*_2*(Thm%gI-FTA<;5ng
zp%NXhzWxRGuRll0!t!t5e1T>m{(9v2yT2g>PhD3y?+{tyPS14Po}0TJ<7FVlP#*Lt
zlZ0TNC$e`WW2qa>8Wujh`-wOWRGTfU_ZV%#d%O=U%Zv+wtTfRF!Z?uEfUR4Wbs?t$
z8_H=Vg@|oxvS>=0KX|f8Xd!5uO>w@v1+6rwvR_=+Qb()W^4~?yC?Uu(ke9V|^o1c#
zBh%qPj*;D)H*8;hhN^l}2+UeznwsJ836dj6$N2n-)h8Zycc^v3q_qgCUcEw^7Tq3L
zPESaYu{yKw4c<lOc_{pCsmenuQ<(UwfMmRz3CpE;<f9?Sf^yXYLJmbRQW?x<kJ^7h
z-n~V(JEW{xKmCsAmXbLkmmR6JpbXwE1h>#``oirWXQuN&6bhOiEn8Am%x=jvgnma(
zE8+Pg%i>wbfwpN7LSU6-sWe76*iA+3?$G-;D5Y3V<@}Z@X>nUsn9Y`yA~F<8nM}FB
zMWjI4ioWSls-&G`0$tSzGTs+laqymqLZ?9T4uK%sjwlUl%mwip1BQyUlu>81*&(u`
z?zUJA(|93e*xkNm8cMnv-4omyZ3M;^xY2wW*xc?5^h(vNAtRClQ?Yu7CKXJ0oUz>o
zqbiQ4XDU-s)iq7GWgSk8r)TP>0o{_eJ?8c;!j5#C8|tnl3$VAh+<p<uv_vbWdF9Xc
zTcnZv@BWAXn(670n^$*ye0<@XzxS8C`q?+!|N3`~rxzY`;vfF!{}~Stx17g`*RS64
z?#J)h+-woX@O(TWg}~U1-PHV7|KPvp`RRqf`S*Xt;e0{s$U#*6_IL0355M}JS8u-L
z@o8dEd;ax5|BfF%yytGWCteMPMr-DI0V%oJZTR$f#IF<9Rt0DkLjkj0L!ez>leDJU
z-|^<nXILxfD@$cHhr<gift%YNYb<ebxZtR4Rf5E+9Daq&tj-Z)#u&rtcp@UPZNsvx
zJWh`Dy=K#z{}^UXB@>aeK(&RYsH_GoDwp3|0uZYFjCGOn$W#KkD2=W>v08~Sk+Vlj
zgIgR@2xjj|Ys9${e3f-WS{<>ekee2}7$Bo`i!_$M{PQhuzj?#|<Nx)q_|3P^7*$#k
zwz9O1C56JXP+B6bq-#1%U6qWCwYU(dcKhp^B9oJ+>dJ~zDn-5Nv9>gGBm!MSTvnXR
z*YB4?`W4{<d7e?-4PyJ6T-Ugmv9%#Y$9#T-9I=~wL>5?Gvo1#@D@k?;p=r8XQcf(F
z7lz>@LIzSS{s66Qft*z>wz@AcGC!bI#3_Z&l8|L_iqu=C`2{6g+RZEcd}5TAuHTWh
z!4C_`jYSmKwDh;H%IfrQ`!FzECcMvVHaGnGSHEEz7rLh5;r<q58-DuDx8!MJ{Pf5)
z&1@P?V>G+A;WxkgJ%9D@e#1O^{_OAn8F8KX@Zro)AC4Fykt)+xmTjwX>j+g%RaIzh
zm@X5?mjjo}gcE|-pMQyL8Xi>vY$ycN(@Q}=TFZ>Y$AFm^S_L{KbWW(M*y28nEAM~$
z31uqUz6U=u4<mkAacgmE3uW0q+)-Bszc|LWA$w0;9qTY*x*c`D1+HP8YpAQ+?zwyQ
z1*h|oWuDk>I?ksPM%L)cP&JCX?G5wQ-%XCp)5tu|WptV%!77^lhPZg<#j(1CcagU3
zXquL`Z5hXrgd|FXZ7e=rsI;aa=OWd1gAkOyZdH+eL^Tz*>2Q}b%Q7Oake#kAUGRBz
zI3H06lrp50sOyG)vt=DFM0c$U#_RqaRrIhD)-1tP$nLh->|#JBk4&EB@Lan62&6K2
zB2}Fk&M)}kz@}|z?!TbEdyO#-Mpc~Gf$8*!NEsb1N@-MI5h=H<e7$)IBS}JGWx-Wf
zA`(KxtqW2qQUawbnqHEX#i$y4^A@|kMa4+CJQL<ZU(_PeZMw2A)e2=Qa!iB_Lbj;w
z9V%z4q*+~2@k&#Huml1r1gdI-+`UFD7ot2sT!~@C4T5>N5Z09(B3T-eg!y>HkL7SD
z=8}amT0(Gyuw3z`8f_|K&LuSHU2z#mQNmuiggK3^(6Zd#5SoO*RV|{c3FCpBC!}o&
z(h|@Jt;ji%GDxjhW1z7$&E~$0L>EWXchK&*j3-=h^nJrPUXV)Q$1^Tus=mN*j5Zt(
z2P$jHJ``rb;@EaA!xGU(GY%)l@tM!R_!E{j5K{!1na&5=%^qVc`sN;Vi|lUcn;K<%
zc6XBV=}682<%WWY-tSONMaYo{S(klnbbS7^&k3(@aMS1f0)wnt>fN5Zx33wF2ma*0
z_@~sh1`=OZ73)0lxqJoLv8)$<`LjJskOU;({QQ9s633T?w{KtZ>B9?gb&PZ3`*&xK
z6Wl0^Y<e!w1E(?2wmtXzJuio6-rR1ufAtx^`~C0ObPZOjvcFFP9|d(=s-jwH#$jT+
z*$~L=_IvjC_cWEoT7{b@PKPs+Ol1|f_xA+vakg+7s;XujN2YTLC3+V)9!iaOUOeY<
zW_$O5kC~U3XEsg68p}FrI84N4LR~|)T32XYnhT^t_Kervx6%fs3&g3|8#A(&i0BAQ
zVf7;=y0R#%aMOa466bxnNJyxgs!&q_zXn21bT>QXw$yBek;D*@F7m2r_&@)j{%ii{
z|Mjo<>)(7&$`YS4lk;?4i)tG7Zyu;NJ#qCYBS9%<GTr?xb-zPvi;|h;d?b!1Qkrqg
zLff^3B!K|QbG#hMG0|?fEc1%=l-eNznI`h(jF?BbeM8;f<9#vAn(l$*XKGaw){$j-
zEUQ%#=tRn?g!-f>Oeb<!SpABSnW}2>A=0!v-1tJcyihkaZa$M@ht`I9Iblshv;B-^
zo-y{Grrm)^NG;jk?(wMzsFc#I%R<*T^qW`sc;fGbGw|u-k*0eg=FH3UnPpt)C3G_4
zOs3z|7!i4V{{vrs^A)$dp4)v-Q(HcJ^}v6Ke+?q={jYw75GCbWwT91M-%{0zc?krc
z>9!k^sX<p92gl*#M|53N-@hWBPMqF-Vj3rErPzJ(7HKU9YY;MXx(vMe>Kjss$niiN
z2ePRPXFbMZIQ#fm{@_BC@s=#3p4H6^mop(`R)0ba1J+pV-IlvIpCR`5Jbie_@Zmji
z9$7CXL!oV5(xOUYqC*S8rrXhI05(C%zKax@+uH~1{sA2wd3u6rDkdy9GcArp5sl1l
zB}Gqg4Ih8|9nnXszUTJKugMvv;Zk<us(^u%sc5u8Yt7(2jh5Gg0mv%}ud3?eK=+X#
za<TdefvK85^7+Sc!qyE)jj1i(mn6E<8q;pbx<gjE072Gsc?l_tPm*<62||)>L(^{X
zK5~A1kJJVuAWSF74pW&DhU;&LA#!^9!0GTpUKbvv;xl{0-Q8ye-JD|?|HZ(vc!pu3
z?sf=m(L$9#j4J!XoJ!6iMF+mz;3EooZ70}TBd_C4R84Ge(fiMcF>w0vceu+7DLSkK
zFEW$&7;9+z9%Us-3hK_Hn~rLGOK=NfS*bS;<s4p=DpJUqrm{dHtt)AskU65YK?%up
zI`jPDCxnpHeNPC9c^&!P_wPBb0)*o0&+fSG+biS<mf=FQhIX?<*@ox?2}m&`1Q@HY
zq{_K`-D4oF6G;jL3MeatNI_HAO5<Y2#f4-9y0WDjmLiwq3*OH(ZG+C4*-e};XVm$C
zs(PkrBKyo7*CKGzaCv^B-s}-VG0hWw)071K`9h9?yMBkRYr?wn>h^&U7sBe1GIBaU
zqwR(_Z@#8nH(l|+YlEN$1gffW!6V!Yy1PMCir@lDf$eLC@kAU`xfO3ULL`D;8Jwpw
z5?vX#H+LiyQz(9bwl}nG%{&i9;+2a@pxbV7)0r3@$EPQ{ZC}=OTH(q3<)8kNm%AhH
zzW*KM$Zo&m&;I_O^ZLs#*zC4gBMIwF-&K6IdCjZ+4R0SB{`}8>$#%EnyPw^1yWR2p
z>421pW%e8{6G}z)o4Vv$MCSGDTh7x&yW3*5U{fiaTbM3K$Puk%nS=~yv@#f@5UPd@
z-aDie%*(?0bYYriq*Oc~FQ}^KMp=IMo8OlJp{a;jAxy!OuYMti^|}rynq8DdQ^=5s
zBJhg2@GBG&rHeBo<Veb~z;kk$ZE3BUrj=!y5s3m#N*S#cec$6_C`3iBbVXyS_qW%2
zUc4syGg!;*uHpaqKmHeNx{km8n{V+sQ`v@kv!&klh;Bpw_A``FEYq2`t_v?FcviO*
z`d>;|W2p9b2sabPGpEZ1L8hrH(lWA~4s34rq#&73Cq!}>V^CI6Hx)TAi1R118%Xvi
zupP16BkL_L1;*1mw5|}UL*Q7K3%csc1UL%T5E)*+!`3x7``1jfE7B!XjI>FXyboEU
zs~sUE48>0ph{TYP8j>&XEfFq^r$^j6)AU=MU&+F-`bg6Rd`$Sb@ONS@{O*TO_#|+C
z!Ai;9&4$`txRH)-Ur}*~@d+x+dN~norZ%60EFA5(Z|-Pxq!xzXzW>CyuKfJ%9%W#1
zf}rXW4yr67d4}_a5EHX7Y-~gJk#+Wb_~{8DB)4zgAZ*KYxiAhhcb~r^i^Mq146|pi
zuf14NtZP1GF4IU=DSrMZ-*6l!<~2~YJ6uYnWk#jQx_V62@|+~dn!VkF2kAV?dyD`f
zJcplt!^^jSg<l=Xdv0%UOPcxVnZw6V_!x0H({H!zOik6bBwJzXn&>=Q8OX)wZtFnb
zwRCMmN|Dp)LfzK1O^r4h9|Bd|P_-@dy0W?g4w3@U65F(dbw)?XBytV~d0ADZ2OL&}
zk|giZQebU!J^2fgP`Hp#MwZawm4Mb&8&DNq*U+fKAt00df(x-E?7bta5_%e69$6oM
z!ZaGwv}FY9Tydj^L{)WE-InMbrmoR0Gx<zdR$ktHkF6BWw&Zn!l#uO~EE3N2Xww#*
zZ<xzr;L1tYf8fkVNr<s1OtnU6gGj}$nNma`%Hc0XoEvb{#QDRwtS_I6(zffVy1sO=
zO(80};L)<AcynA?&(Fjd&@v+$OI=rt^NcfzZNEj3P<}?N69k7)P+wEp&IMMVkSfz?
z!?dg%55+i^!qt+v8q{`A-S$Y4XjP#fR#ih3qWI~h0WA<x5JW~xO;#QOXj37qMK=vT
zYobdCzY^wCNqr-cS%Gb-cCS!%k8lgxHmvi6u3MyS@Xlk~4PtaGOJKk0a6!;*Hw=%3
zeIFuhZtvOLJYd=lbz6~BX8Y=%>3D{)V(W&mBtjS&PEV-Ha`*ZzDc$3PL+Os-CQ?{X
zbpz`NafNGQUaz)<-Ij^z(<53re9YYRTUZyC`Ha<?;dmh|1@mi_LfIaWNUeA|JTgoR
z&Hfeh%X?}S@KKPmpzewrMM*)^G^|pwfB2kw966spvfKBlE2TEZnWoOXde}0L6Q<Uz
zCUbN9x=f(ffrr;U-+cKQKl`(v<1S}j4$r)P^Oo<w|CUWH`26i2krO}r`R5#uPjtHn
zv{Y1h?(Xlo3>TV8)764C2ck<j;YdER-*y}h&y3T|W?S3<0K<8_B8Q99LMq9;&J5E`
z2$6N2S?85XS$_MQ-y)@8UCLVFco}iLDxG(AB&TrAU{R33wj~J#6nU2bVn_ukEd?=V
zl!%avLOfhSRO8`DbRL8#xL&Roi4dsk0zk_Vv{FDWzqZoIszLV~bX9U-McFM;#!zh)
z|KtDk-}3p-zT|Iy{VmhHAZ?9M<?tePkqraP;|M8~3T{=E&(@U$nWex_Gk#q0^9sSU
z$cZ?dnT`icx2NCi0a?cF)}o9iYkeg_CUi_NPxyI4FDGJqhibRDlnec`vRK)(&S#E~
zpHNazZxljUjIkKqQn!0tJP?B;dxw^iaXJ>C%o3TW6H?ctsTgFH(ZuD1mNl2l2h!T1
zWFn{Hk?^a>dyk8nes{|_o|vY@>c+ntQPG>9y@n8Z{bo=04tLIYC;0X6KNY-fW$3pZ
zF-9IgJo4G?hGjVO>2T)bVWNKh88QwWzWs?>YnrBFziW9poSEl^ZhupflF_rAjx=4x
z_Vt%UpE-a0j>~D_beb4v&&``xeEE<5Ddz4q?|=30sGDbGZE)kl`T4}VcON;4Ky(vt
zvgAfp49m)#1WqWlUy(vk-8@itTioRZ=g+L;LO`>7_?+$SYs_X_eBR@k_2WDIP)?p_
z@A>iDUo)PHy=|RlnzrW6XK$&Ro@t!9Omq1lC6Ajf$j{z#v+HQLTk>$`a)!3+Xu6KR
z?;%DWUcUjYnT7-FG%$|?^E4xrrfC|MX(~0qun>Yrid@Wu##~8UP?-vBU0&oqvCad_
zJYp(~)CN)2rPJyIDf&Nt(M2w@jT}qZ@N#$|*Ecj>sV9b<kwRgc9zr554pxs4p6ndU
z`Gq*oG`Bm%yx?3Qq{w#DBXgpvJt{e3TuL(`#ZrePb9(s%X<@tF&{iFZ!Xgmbg4FmB
zNxndHa=Zovbw!rukJ?z%Zjq)Uc}LC(Z4EhH)67zk{K|O#4sm%#n~Ix_#ibTeHOOXz
zs+uw?6`Fu#z6_Y`36(+#h3z(EK#Ghpx;%PBg|0P%Kv-9Vt*~)qNlOXp#mF?A!F!ZY
z_+?=nCd6h(Z#zD(8m4I`#YAlss}Jaysf<R|4LT*F5BO^`JBD}#@3dFjT1jcg3_+TL
zC~mhn7Yh)myP??_;yf~pM|9P(dHps0{bwj$fG*YUo-7uuZ7}^kZk?#Y1zSn#O<Pp0
ztgLH=Y&vdkI;M5yZ~pq<^R=!}>%w+_hscrf<%xB5SY33@$yC&pVO^FY>p7LIiKrU<
zIuVwM<#eEb{RXWyZXWRKid2@iZ&>}xcDuuxgjA8?<-;HKvQkwf1Whej#tR_@l2q6_
zGsl^TC8Drx&pKU5(Q{b_PL}~=TXs5;y~mFuOIs5WSg%-?iGF*_^LgY(RX8^?oEPrz
z@7V1(RHmZd^u=Q#YmVcEy4vyZ<(DJ`>ol`27w-2x<LSWbhg+6yjnIOx{`6<ehcmk`
zKWG2&8fy&W`AAHG`};lf@QnRK_Z@-2V=9f&ip2_^pI+E(H^>wSE+Jy%`E<a$z;?UE
zIY(2~{4l;RyJ(TQ-EW!ZgcW51nk5{MXWqS&@E5;CX^S=$K01WH>a;^5I7e0)+Z4Jf
zxg0p9(gn+|i#;`qL|i>_@%U+moN=o|WP!|(L^-(R6bp_#lta3#ENF=|mU^>AR~BhC
z*al_maxiowIeC=T^li&O{jdH$Kl{a({OVWV^W*zZB%vASnfi30X}jX62#(-N3uj3Q
zv945ggRxb)=(vPi0x_i0DmY(QP8Y_@#9fGlSOSkpE40z1tgp`pQ~u1VK!<Dyh529&
zUPhwrI4|Ir0Y4l#e*8$)>=33ZZ7MA>wny9kYWy^WD3KyFPb2G^skK68S@Hlel|!{@
zaNgtmh*FMqJQ0#$QyY}234SS+VH8q0;u=}6xe;k^K>QtK2KwEGDhAeRWx9-Pw>5S1
zie*{Q+R|-my4wx!e|kqGG0%>F{xALm7r)Tm@3BosSQ=KPxxe4jcO5ZDl(sB$xn0Y{
zfe#<v6BozNe*QCRR;GC<d+uSzNXgxozvTW;{*><3S2Q;}_~uU#^8mtuF^~nn{N*>u
zF!S>Cfycl3J-bgQgwfQS9oDwQs7Mm_uij$2o<T{*!;#5l>bAz-?67T(Ulz!bWjHdN
zpE239+23QL;CH|KeMzdTQblRDH<)h6@jUV$e)l8OvT%QU!&`e-Kr%uiOwG%AWEn<K
zh8!bW3aY+`oUl!W(w581BkQ!5$3X~)m^jWeT5GDRATg_|M%fx|3`S|JF?cs!cY8&6
z;num3^OUB<%w+*oR2!s)I6Se=mx5Srx?(k39Dea6zoNS%vg(L##a+(0^D|Xd(QkJs
zYcRRNuOs6+(`*|?otfQQI?+B5uOXzqwv3k}h=h=mb(xq>N5=C|(&<*?*A>55T*`>6
z^g|g_KqCZ%s>ni<;sEPX#^Sm{nhHfExEZ5#aURHGf6TJb8&fVS*)-2c@91v#G;be>
zcRQTV4CBmne1@E8`VCP;*5Saq&NN*^*R)7$aG}U!jI~8o>wOuk&oj$7A(Uc~hDDYU
zY6i6Qbd90E-EcZd%=Q)i>(7ueFet^`uSj(znZ>~DLZLz`MO|0K5X#TQErfNYYC4wH
zA*_W25skJ6sSUxUGNzXbsTA$aEw=7Befo%8X6pWies@Pxw`gOb?x^Yxl{U1S9r`MO
znwQ8pI&PW`-Uo~^c<(XBa2ZB+w|BgL`$ajR<P|00ay)W)dgT7qTXKqwFYlSAGu3q*
zJdF#=)Kr_@by5Q^!_Dno8BM!@*5&ZL+1?b`lWEHBv<k%lDK#!entn@FH!O!Ip5Oh1
zs%o~kHymG{N^*UjXuBH-kvtCsY3SN5R^L(CBj@uAr{^b@Vc`7ygy}l&?jK-{EQ_bA
zADEXD!%shuGu++Z@zcBCP)k8=wxqBSlPAi6%7JC{+`oRyyezEQb9=u<iA3AiT!w+R
zu4(%%QD{O_QP~m_9WMv=w>KP5Cy=7(piht7K77u&2CBNH?|PmeKjGGe`@4JQd8Dxw
zQXAG~Mr%n7nTNYu9-p3Y&hyoqS6FTEex>QR{Py?nsOknOBy)ge^~7~W#+gLK#nKo-
z7I~Rp%M@mN_o{?ygC`4+T9eWh2W(8qx?DeNDS9B4&qR#)rPLs+x-?LPC~H0M14)6_
zl4jEv09y1gT`)EyCFJDE$&uqkP7$3}awraiS9g;BpZtvN?;AdR9GTadWm;*Qp1N*`
zuKYh!7U(Ny!}*ff*UFHD#B?33lI$GQG;=<lK|s}ZU=?zf#I-bjWK#@@GGuaa#foVS
zS(c2Km=<(nF}CKu+Y*K0JZn;S%c?TAyF)0$$H(vK8;P-z<KdB<BHPU^IThq{z_MnA
zR+gLvZQbC!Bam5hAh`e}balsW{}%5CNDDb+M7!he?ltee{WYK7zvuNguW9<0dA+7s
z|F#bUzx!8z%~!8BynDWorKYVk$|$y*9hYG&ym=w%_BT|0&q>4YKR(koHD+3I?ve3u
zVQZ^G&Z;VQZA07l45tg<{_0n(=L=uGy{By|a)|iTBg=8Z>Kdsh`o81!Kl}%5zxfNa
zQN*wkqGwqhZW!opcIcehZyWCJ-|$mc@#|lI$4`Ir1MQ~emw(=~X*9#}%(TqdwkNI=
zX_;u7ip3g^=OgCR6WaiR$mCa+kMHouBSJ*habgYypiwtf8TU7-m>B1YWjwLlG~D0q
zsEwpjU=^(A1EUDo+7cpE{SG-_EqgAMU0I04=s6w^SI|%-`L$FBDI%x3su5Bax|fh3
z1SuzMT}FJ~t@!i6ysU_v(8l0Hz<GzM8>${81wU+63Eadh%RXfgRb600apiFQNRX1a
zF1X>yI-Kdc4st4Wy=fpWBdV$Bw;j5vczJwbnrBo&p^<!qN>DdN?-^3Ud5^6OjWvWA
z>DrE)n?04)B)^iJFQf4g=;{{Jcc^AdR+hXLOs!idBob>|WcCGx>Q|!kWC<z9t46kr
z*c)q*x7))1H@b}9wp-R^=K13XhL;!Wwgy{nJ5{Ann+I&yl}1AH_;n_Rz~SYYbiL&)
z^IW*-rP)uGhU(!pRox<nBi1B@>@iBxG`8ri$1`!dFijKZ#ZkA4)>;Imf1eSg6d9I@
zc*V3@V<33kvJ@baY>2il4`RPw4}L3g38<<@Rvq4ZruoRStmvvOp~#f*^N2`A+JgYU
zOt^U@hJcbVM~_g6b-l3N-y(HGWw%Vz1-C%-fwynoFpnc4C7z$2=xf8>?JXfjgfRGZ
zVYAtgQH;kIjIr$Q9tZ)JaUz6r=<z9GfZgtn5FN@}%Hl&(H*Jx|h=|A%ySZbd1qlrZ
z+IGiaI#3d6I@)a{BY3!fTP_$lx_*Q0x1<m`ot}^?W3*u$JZ*nR-`=q1z`VkK(=x9o
zhH>Kl{*Ig?e)X6f*;vbXI5Ve(&%XYgWg6&rcho`<q9k~Sn=iz5B$FA|g{tikx%~gu
zX+~O$!1MU{%+2NoNn)KeIXa|hP)f4f?odLoX&ZLiTb5-gA;NWGi5}+?cVB!-mWuZC
z*LYW^P0r064llIZ4PV^7p;C%v9(g`4R7P>T?FcSlRiYDty9$P<_gD?#a>k{EX__*2
zmlEg8<SAuHDgW{N98<*9HO>_{kv64uqiRdFX-i%!!w^S+Qpdd_m4#GbHT}&7Rac~|
z^FSFzWyO_g5Q;#Hh{&QiNy3VViR3gwW~?;)#V<eS%|qhf|LS)<KOIOR(sy-ncEns*
zD1fc&LPd2e07(jL+k!5a1OTL<Y3dRn4xZupndoLVcNUo<^XY`jh2Nn}O}oEES2ccG
z@Z&(cZLwX2^b4ZiaC6sD?YFG4W1L6U;gL8DygV=Xy<@w-BMZkojihX_N^y60U!K2l
zB?L#(hTOk`Wq=6F^n|hvcW-}zUsrU!g9KT5y8T;_g08*6-uHxA;*!JI=I^Fj_?sUe
z`TXsM%36-+flnV{)7NNGU=OQ^ya;I8hGiPaYZ*B|Js$b^;e}ma)3=J|;g0z{Anl5Z
ziN0=`PiGEK2c*n=wS7R6`1aikoBfWuvp8QucX3@YVZp=!sS1GT^y3dqhXeSPYO^K#
zg~P`am67<6h#}E-H4hJaL=61&+wW01Quh_V{`dcem<(NCbF<sgJ-nvYmYfq$@4m%N
zBg8<ME^Mr(X*+Dw;-`feO7+fI!~K`vpz5CU(<3}TvF!|XTd}*@6;{7rIX%ya<k{`_
zSlb|tMZ|<yO901*!0G)@NL5a#w(ao4aOKRG2dAJUeu?Gf?qfpR!Yv=q7p7@soF?XR
zA%?(iyP@qHLW+35QpA`brwc|as<y?{CBQVzEAu$vfg~(BNwW8>r;kLp;O40$dFKHs
zGe%0HpE+L!{N;>kN3syOX~9}U+xA>eXQtzszOJeMf9?HguVqP=o{2p+?(qzF^f89a
z$gE@)*<B5S64WgS62C~lx}guP23l%RYKp}oE9aQp{kq@bOnb6XAMChiR+3UD0Tu+X
zd?5lJ9`~NJ_qIKmt+n3gZB~FF21nKP^qWh@&A_yKEDeUbF#omIOg<1}LW_i3Ml>6A
z+aXPdl!|41!YYlf+M+EEvAjT}BH#&Uuuap{7_CdUyV7*gfl_$Y;#7sniR2uO(b$_y
zY*S-QkuU1XVlKbH^eg=1dzPoS_$Y``vL9wrD4PD^G@+EC?tApPkHORyC5x}YDbIxY
zglbkKsTkk?$lbf|i$PI>F-R_}o;p${S!YMO({e%?No^{uwM9*UEY51<Oco=uQ&grR
zrkS`*q%a_)VD=8}A<vOH7VPv#7>yeT{BR-|jjbEfJP<;lZd&%c2c#55vMMrHS63h{
zw%d@Sr0;u<DX<L3;+U9WdvQtCtQhwNcm4ejZ`iI^H0vIBR@uxUqGfr3FUyRn4Jt^y
zFP;@h$r3^7j=JfJzCC!3`$zWsJ%`Ej{JN*#-r#4?G9JsMt8Qp6ZZT~|fBk~qkDN|B
zR8`SeEj|R4QfS-IY*yI1$4>{g7teSmTh<rX?C#$I0d4D|HE(KSh^#Lz7*0pJs$v}W
ztXH>ay(PqvX`TxZCP!S1)Xl0i)QL!Pd!C+l<@z3u#ndPSob!lqVwpy6whh*ZGWm;+
zxXe6_3ze0$t1XN7Os7XQp%|zo$JO<Q@87-!B$d%zKD*^~n)vbCceHhfkDlwR3#2Xt
zRiPy7b;sq+1rQ4TQv@#inpKW$W#s<HJKTJtl?B#SCFr(Ww6REGSbRXHfDsv=N?F)c
z6*=>XQkan>StPrDIU($Xt}M1Kxa}Mh^DrQlM%EV7H|SM|(VFG_|8?C$rN~khn3z97
z@(7tBg>%ZhunwfqWT8lM<f7m57e9T;|L{M3;QsvsL!;UBEh%`ax}vHK`3wOweNRdm
zEiwYI)I>Y7-t?#xP+GB!C-QNRSSIRKjj)o3H*b*?Ng`ynuQtRCB4%{c5q)GXeYm8j
zJw_bBRMeYm+HH%qD|T^aclb5EZRs~xw3jc*^P!m0br&2S-r~mtF=Z;N@ymz^3m#3H
zPt4=Q#pN@EsY$AVbiyx*5HcBqn-+4MS@jLO<vntanC<JrMEQs48OT~9tm5X?H9|{j
zEr>Z1U8Jw{xl4ATX*!y!#`#EPbdgyY!LzHLO<z%4#XO9t;P^0(><>ro?;n}wh1V}G
ziNW*Z`#nG0J@T`@<KlWt!tn0RBYVH2|M9o<x7Vnv7gWa5)P`X=Q3=8I?Q@h-JpAV0
zbN{d>2*LJxgOZ9>*RX%w@#By0+4L=2HVmf)WfbFS!PcIeSGTn53#5R%@4x56ci-WT
zM>ZENQ7DqgXqj2B*IX|%y6I@vTdrSxO|{x`*j`X4&pe&*VMa;NO<QX9N>KG1l&;7z
z5S9UQMi_;*8kq!136uh@?HS<HVfA$}_wf@@u(%?5mT4}X<tahcuy}`afoYnVokMGd
zi;+CfNC_er10ckLF=h4X#(-@aq|`;xFq}%WPl`nEXquWtpsFfzt`V`A{kUl;SruUr
zn@hAP4<L6uqJ+e@J*H{6GzO6(+f|2%fov?c>5$8SmV#guuBj2)o|O$pviD?dF;zk3
z`jkl0lm?+?VA;PXxVadIkRZ;5HvlMWX|yKHGg?M!t58(~DHW=rs0c{1tQb{`n<v6D
zW3)tgi3p)&TVx~(i}V3KJG>7_U9sM7P`1LXw&(8iK)p@~Rne{rR7Rvg4hz!HRI3W_
z1j&>%Vip2bCA2T@f>ua6U9(wrq<KPWK@JOk8c2(yuQ!;cfJ1}_X(bZF<I|4m?j6;7
z%li5j69ZyC6<L?5%8OTmcjI|wR!R`bEF9g8n@)%jF;&Is>5lE~Gpc?~y$n1)-Lt*C
zA%%pk8eGWuWn!LYob%MS#x`|fRfN*Pu9Yet7Q(RJtWZWEmEg2Lkle&+7`do!IP9KC
zIb!QAb!ABh&vaTaZ6e^%O?6hhXc7*oE$fRls-BtVg}499{|lKk#zbCx@fA5oOkc`L
zkeSCL!{a-A)6=#aauO)ha5^54DWQcbrfM>gePTR5()JBnYeI-LjU{DAI33VJoxA20
z?P|-k1e(idTx>S%-hW5c)Hpx0ZY-wj5K1#?Maq)Ry5-sJ6+T-c8RtA}vms?qUENZN
zz%tFac_jEfwU+3*CC5ncGqaCq3zHkDWy|6Q`nKWavm5Ga&GzDgX&6yL^2MuH2q}5k
z9gD{bnGf$DxV+xdwuXLV$Y0;`@BZ$${N-Q#1eC^x2{G(B9S0tE2QIE|xVl`Ch30rz
z2z~-3sQZqpwm={`PY8j_=eNwEkPl^sfTr1ApiG5xnQ?T4$rF_>HowzCPDMsoUv20v
zw@9NAne(%UGcB`1oIMpFl_t@{v3z&`oaE**sfl2l4o_(azP_;hcmJDz#=rTS-}0N^
z{SK8iU03nbZ(gyCGhTV<1kzHTl`#~x#PP)Z`0&ZhX^EcQk8hD$(YG~KQ?q;jp5yx+
z-KxhgGum39OiGqzMhQuMd4ufQ0u~aIIE@A4{`4M^0&RCmb$!F~%`cdT6Lr5O`I+~>
z{D070ykdCzP$;Is5SB!)BoFuZ++1CAe0s<6xTjfN;GAb!1Z{fDa(ZO7z9FPYU9~8o
zh*^_W#&j)hr_i-#nwEc<20|e-n~N*f>kY$Uppu#x1IO`*HIlmPc=ut4G!<Q2AtikA
zla~d>7(FIu#^Z@xRTPbCanhRRVy;uS6(NDETkP|O=h9NQE#t6IS%Yc}_lJSz;fea!
zzu|Hi=+-TpzCb8Qplw5C;OgcYpCU#TCU|n5-QyF%2Y&L+OU7y8?)?X3d(HJ1U!k?&
z=K6|$y~5avm@*gF*U0n@<L(h_Bq}G25-jtA)`_mJNLo=h4KdD)5AQhL{YZ2Jjn(vR
zkE&~wZJEXiWeNnSwB@SO&6*?AGGc3sZF;P=2wl=~QUq4(j=H-=*Bz&)_ee6PuF266
zlr5yVX&~j&j96C<Rnu}hp0(J<u=u4QV6{Z3vf}b7l6@}OhN?oy!W6Hns@y0cBQQAU
zNPwx@;zbbu$og2S(iCHg|F#yH*;Ys?$yG(S-WG?p;{hcp1UzF=wnZyRat>R;rdlKF
znj99~FyW&|Nf764Wfu_l9}v#r=YjF?fEJlpw?!?I3t>@dg_MF2i(mLO><h2E%f&);
zK2g^V%GCIIz^8)n4LKlXNs-p;3v9D0{^G-duz$;Z5M?_lkmnOAdW@Es&4!GmP2z06
zB+x0MWkpV!oC3kkpex$0FUVgZ(5AvHWn7*2Sec2ZC;V`v=^Ko-<lxWn9z|8xq_E&!
zBqqV)6Z7I&<^`pXEDs6Uv{Y4#&<Y_HN(r(MSZgsk<Cc;+h#@ggBYr+1vY-(?F-6=w
zv448x;ll?)PNY<9rKjOo)G4AxDa-ZM7Gn&<xF<?UP7BA!JLX|14SM@i$%@ztAvAT}
z5|svNS+6#P5Q;~FU&uLd+V7bsSDfyunve_lYZ)EM&(u}R92aEtJRNqt`0AFcn^!ES
zu{@ByV|HZ$5|2lulvH(vlHmP9-L3K2v)ydDfBQXdKCxb{u*MQX=I+BiKmGYPyuZ6E
zCO@KNQF3KCo*n>=sT-^kbQ?|G-H>y}%bL~tim<=O7==+5+iqyvElSk{w@28PO@BcM
zj%gU_*K0xwv`x=+cu!+0lx-MxZ@IdxKp1xWGTCcwu~V12;V>Ku(Nkz%zW$2MrZ5K9
z5+3g#2)Q)my|}&M;c-W|YB36$+S0EZmgPWRj{LX(;=iI_U-9Pt0qF}1V0w7KThF`W
zNVVCZl0+!Q={Rz8dqI|h)9HlDrH`jsZwNlIj3e{1u(+A^`T}7zB0H+4W2$PFVId{K
z?21Vd*0PAi+LSC^oZYz`QIs9%25HxXoG~gP(!w~Om`{oB`314+kabOVp40t%8fEEj
zUXV>kOpfPmPq*zbO~v2+yI-I0fg)pErl~l<dWbII<^ioT?sy`Gh^`G;dLG|B@#g!t
zyn1;}|HapE{UuFxfp@<vfHt$nbR8a9B+fa44@}1sjRYwoPDo6*#jY=r!4-&@cbF>j
z&Ch<${oPw~4!Cil-Mk>m7A*rR!!*zMGf1gDmxxi4+m|niqC&_8Eoy4BMs$G?5?$M|
z3=@F?gvY3cetSiT6ZnZ`82;?{`+w48z#{m~FMs5+zb8(P`{`6%&Ze35wn8XFyWZdu
zJiE9mh()6q4~O!y$cbeba4z6n<Y|AT?K_Muj(|y7etdsKUcI7Sb*PvSS@7NWKk)p;
z3)=0L;fEc+{MC;f4<oOi?$}&zsOy&1^(CsYOmQg=Rc^*F3y6$}nc7x_I<mR2eDlp~
z#>w&PU;RLR`%}L7Xa9BK1*Cvc#s0N)uh(#O$!gW2QbOiXYUj=)Oog#4nu{AYH_ymI
zaQg5b9|G;>0%Z*2G@`1G&CN5K5EzapT$YFoD#yYAKR39|<3fzZCp{$?_6Me%G4@2a
z4k%N=9;f30sWbwCZhE?_4MA(>;Yf@gDK&lH0hxI&makRSAes(DBBj#RZj3>TV!RV$
zala}Zt{PRfKzYFh=TTabrNmY>UELt&i8P<ERwEnBoD$=3Dn=w)V!I8t>*?1Ua`42{
zk(5gZcu4`yh!}9rl?Fd0$O^m*D67e;A;pAM#d_D1IDGh#;luZ3GE^C|HN`h!QRp)S
zC6mDif-9Qx5F^S3vMnjndACEW1FOv?8BLgvWT9!Bnx?JJ7QQtpfkPpUK}-%GV)2P^
z3)5+jP@4YoiuU3Pp=*x&J<Bu{6-IC*tueO32LURx*{(|GK9pv_X(@nKkug%xsDf?o
z40M%ZS_*(lnF_LCSr&p@2tE?BB4lV*Yx>QGN=vjVOm&CADvJ=MPOEKcMv5^IY=x~m
zY<Gba0(W{sONH%M*mgy56UW^XRnt;SiIEC7%{-kRs0}pR=O`)g{>V5i_!#KA4a=N~
z%K%xSjbd>Nwy9{Eie;J$4mU?`ub*M9rSCgV!-42#mg7R#Se`w9&Tcqy8b_3tSVgh=
zZL4CO)Ryeq=K2{dBW_%XQXrKgIEOJ7?-ztN)RiL0L=KM#bxYg!#3&(VR^5tJ7dzg?
zIc&XRTpS5aQ}rAlpJ<v2sRZ*p(XTtI>Vm52*zfNN!O`}K(|Bb2@(Z#6WgC2)na4fb
zra?%HscO6%a6V&$V;F}b3q?{lE9S)^O#@QV8-W`SxM?5@Pq%swQqy)V<MBYXenxyy
zEaL$wGIdj$%EsYHV-3gM1ASj3WTvibZm-vvx@OZ^oSV<+XOGgB)8U@g<u%W4uNj_p
zTx>VEi)+@yiFfbbaf%sZEZfT~M1YfzY`U6e(_*A5hC4ZM|MZAHO9lhQaqs@eAF)=F
zrjh+|LQBImI*5s?vCJXS_bnO0JkE$%ygk~!Lp3X8cSSZ0Rc(le9ey4;og%bbQhyC(
zqAP|--R2s1c0kotjT8oi<>G41|M6cwV{>`I!`*wxl4)^Bp@<R45s@<Kte8?6R9&Nu
zWDZB%?0MLo=vFO1yMD#y>%Zji<_8|%Kazu^YFgBKgD{#zAt-7gv6W>B3o<zlr<u#I
zp3!Ws@TVOj%orib^AV{HFMjeDq*x5cviJoq!hZLLci;V<>}STq1KB4w7gy-EqhD|F
zsi4_+PY)mk!B?alF}h-L3*&gATPvp16S~cer$`dMm>?<r57RRsl;*eJe;{7>C~erC
zMvOM+COWu(*wJ6!AdCjBQK=Zc$P8f#h)S?Nr?=eXNCIL?C<-6nu6smfal;;ecqGm<
zwQbP0!AnWMsd@EcLlQMV{mXAyUu@B;Ky@xnMF=OB{Rie@W;&e;j86*YaiLiyR+}w3
zd3H}bT5Eaz^=s<aKWDwYL}iCKgEvAb?t0$Mgis8OWQ?R`q^cW)P$a3jxqZ#%`IiV=
z<EM#}egLacS(25+<`%6CB4?VWW!#?_r-8bz@NQyRCc4#{s$H?N7B`KAc|q4T*Dqcn
zR82VEGd{jUH#JskLRI5^!9a(2KuASho5J89Pbh7%RgE>0WlqfVh&Gn1t7k~9n1*9P
z?FvfT_&gB+3Xi~hy48xRY0ljLOiGcMGu`?EWy|(ww-AUpm67Kc4U|>c=HiMFiaZG)
zh#{g>Mkz&b3-0g$SrUXmTFbCMqOwMah?J4+O25Oj4<O5ahQ%QPtu&@Cu6o`DraU6G
zL>r3|lHfdvL>5rh6<P|scTBUx`&cM#O9We(4V$YKtIG{SsIuzV?U?rW#O4N3Um<mi
z@(b>?BZg92m*h_lj7AtF0VPs-8HUo+;k{>BW-43giHK4kj>|#}fvTy|l|>^E$rD6G
zXhjHtaToy!0>;yc;5~KSpsgjRgcX`@v!SgkV)RVQT+Wd%j+3ft&{~jlaT|<6AnP98
zZK&EUDhI3xh!{VmXr)wiU61p{+-Vw)L{C9GyQP$vV&viO0Y8njn-$(P#p+rr`t2n`
zz-ozjw-Eh+h=J{@AxZFI<ZxWrZZFwvHu%|d@DGR#!*HVO*K99uiatCnSdkcxN3O45
zlVm|O*S6y6xFacuMC!Vu?>B@H8K*sUWr@q4IPP)tgx!>`|Mz!yG?gW1X!;FxU30j9
z_sP|77$>9zTS-#(c$cWF2CaLLhH0L7dVDGpRVC3@WBLom)4*^zloW5yh}{mPLMqGV
z>Lob|rs>GyJ*&1O#z19On935Dikve~?|+A~9n)??%1n$$*4u0LPakM2jW(IeNP@_?
zY2fig;NtQb*$0;3>HD6#?FxWOmgc%lBKo=TJerQn+vjZm>T8Z4zQbD0JeAVMc-)hG
zAOy!xfA*8ofVRCxTg%(GZ&69}bQpQ@`jVsf_;KLr?k(PxRsFi_soI|Jet662;SOEW
ze<%^zbREMI80Q5YBCEPXwjCir@{YPHMAjfQb+abx7O5@M@c}vRS;m3Krz4x28-g$>
zKaheWxPVX%N>@C6`)iub2Hma^^$KB&^5n1o`G3vtfAw3w{pGJ{+lIdH2q|Ifn&IiO
zI4<T?8X=A5csg=C98p@)Y*y%I1J7S@I!yfj+k5`)_aiPiUR@h*(!kZKDbL-eDROLS
zP*#xV%<q5yEx!ALpWePgRn;d;b*U6FI4~`y?s2CbE_mvC#pUHoc8`xtyM=i^(deFO
zSm-V_AqGNRAWpRP6<Hdl(;Wzd>MDlegqD)WhdW|)Os5lV)6w=l!|{oSho^rS*Fr57
zT0vdcOku|7kE?D&tu4F#k+*mEy#DGnsjKmB;qBY^jK_hh)})xZzUa}lWm_5Ab;~f!
zR9!=s0<8>=fn~RYl=04CE)Co34bo~flIOP<>`#elm}xc{Z40nvdwIb!THb#59gh$9
zNTpbHJ%`<a;dG*zCQ7-;b9%^JUT(Sl`9H&~*5v4kA2F}aF%JW7J`%h?+kWO!-%cQ;
z05VdPi;EQSyF2FHp5ft+9DLD0i;QX8;)#7a68sZ>oCv{lrWRrAo|phvbg)V*eDZ|R
zVRAy-1uY5;NTfu}skqN&NJ<ei<mk!qY#NkOvAI=_#h2Bw8%OlIOxTpx<m{1vlDf>0
zgQExv6dj6rS!n8tZnMG7p6D0cGUKNOH=Hoj#P016_~C%+8^(A*w_BR^RnZ_zNy-_O
z90qX7pAC{0QUW0>RJAT35hYMcutY}`3Q0ny8N5emfsX;11x;O}>LM9+ODy%rm>>j1
zEWHuQIWm%YamBHbK!~0(CVU<WcfqzOU6ZWANXdG;=K95RY`Y?-M4krx?tvjDM0Dg7
zh`KJTxDd(90ID<}O~+$F9oHj*RJ8L_VVjy1OGBKh%9t<CqYdN`aLYi`7OUChJXsg(
zX;n2qVhRD>v?yuFLNK~OlC}^Cqa)=^+cvaqM~(sK7eJj6um)o+nS@lD-~)4*OJB$6
zsPa&B%rb&Zgk{1lLxHDsineK3U)|!y17X^eD@$8RoSV^2k2Vd&fJEUJ$*3m!cEdam
z+&{ddR++l)X!;)SPmHHM>(vU6r0p(P#v_7AjDnOBLL27c#O1{e!ZuV^vy3N}X+|o6
zl8*h;8=CbcLKx<8M@}BQzT)`!hKKzJR+q0i%_B`+gI0)Cf$13bCkzfBMw+gnsaihV
z-BGt4DGN-urEYqLC7=ZOWk!&|G}KiC%SaX#x@vLr!0qizf)8j@>}K6?WY|40%_El=
zS2XRK#RsAfRF$S~JA4MT;^OLtc^Xk<#>0-PHcU$~qpIpnDPv}fbFTCqEX8|5*Dc<8
z8f#Jgiqy6!YkBtNH-us4;qgb5G_?JiezRp655$y+%Z%v^)oMdd8PinMO-0<^P%8?k
z%+|<pM6{W#B^b?SyQN$8K;~z^_$hvw2=iD-eYWPu4^KRP*z@MyBR}3f@Ut(T@mGKO
zGk*Nt_bjJ9IR-jeh<yP=5Ry4$mi@$jNYoeCh~V%`z^*z}+aPTPrXo_*NkTPbP6)s`
z$2bgJ<cOqV2}exT5Pc#_#kfnH4)1yO*Dp({!3_x0AyvcIfBBcZd-onGB*QczltqL{
zmI_@Nq$~YJVV*cV9hjGZuoa!D2vx^uEZ_d}_xvya_V>Ko?Qx@L37RQL4u_dnSB^Cj
zEdot{#s;4ze*f(qlm3=3zj{rKWzm-1f)4?iI_5Xu(%n8sOH~ly!vWK-xVX5%9UrN?
zHFa~r>hdK%dHmvN+cn3>2kPz$(=@olBgz^=iUs{011f5?k+^9j=D-phMpay1UYCsQ
zf9}aZ@Q$m#V;lzdi=$~;2yi?NSaZRvtB&u!k9_}kKf)4u{``t%o+)-Q#fSarc;Zs8
z3Cn`>p1L;ltB#l@^KoD~l*TD31?yGM_UZz)+7JaKH&ZnW;jrh2@4sjN^nu&UORCic
zH{bk>=6Zu29Kr`GU2JyS9H|?Pl>(~-wp}w#3(j;np$VrWPMol&fOS9;rDSm#zZ3?4
z3?8XfA=+78kdZNw=Yi49Jbie}?%@Nwr#;42tS>KUn>D&=Ng4c7R^HlZmU-mi{SjMR
zHtRJq2j=|_zk5Vi6>f1X<HX(gmbz(JZB~rq38bKI8l)%`Fr`gdp}L7uT~^f1iVzcS
z31<>kg<lqIU24@~SujEr{<qXrO^uNqIxfWEh`MAGtP(V}MIt!Gh^*I$a3r}Ix16vn
zbXOOIZUs>ln^FS77vm+C9V$b#6+&sW)c6=kO+&qXMt${)T4dC?CmbIcm>HwOXhUlY
zL}{Tg7&f(}RuMs9o*c_E;!?zj3SFsE#}}n<Vey`r6G|$CsF2E*l8ln&p(6xJYNRn_
zH<uUwd?M$fT=4T$%z%U-%?JE+#HC#7-zgOv=e9u#K?a1jC@Ba&V5})#@Jl9!vu^=K
z_NjG|{tf#_L=0%DSbRn(MU-`MPSb|C%<!>n<Xma8HwGji5<WOmF6x1>EF2CyvXmtg
z5dv-3m+#D(oKk7*lahI!ao%y<J%C?xb+tjsh|huG%DGpnr)gXIi%X6ovG_8{6WU_*
zfRv^5k(yP>Z27`)kJEw@2^R*Y!;X20tS>vHt}#mS{>>Y%zx)NJ>Y1mBr`?X#rj#Ls
z)FsOn0;^5M?&$#|p>A3bnK6zWhY7oq#p59cybm<p3a!6ncmI}QTv+uD(JiPH8D~et
z^7@;<=EK7ym%68}6V5Fx^Tc8INE=F-ASPK1q|T}UA0pOjYOQ&Ce@_rctTj|sg%AzV
z&4guQJ`5}_GwmLUek6r8QWiXPb|urcc+YB82#ZlPgpjD(nz#h~645zxczlN!nakU+
zx&ERjC78V@%`;j!#={fYM`D76EGjMU$bM#VJ1(!k!c7O3agXy8^L(OiTgZmWUL$nF
z!`<&mIk5OQyn6A1hr2hlUCZgTu(`TsnU2_|#<m^C7ULab73-^>aj;Bhg!yW-W^vEB
zzP_Pf{dmrH_5^Gp7>2;MZD@qVCr{EETep1i>I;si6W@RPmLJ}{=c`xO1eFnO%R;7B
zB^zeWSa?;hhzX*sxqL17;kzGs=@$;C2_F=ji(As<&~=4=Df!jk{u^#phpCz(t>x@E
z*!2A5ul}01zxj9UA0Am;;`a3`ns$X!rGF`FNj8?cyCBGxF*t_9h&3JGes|CR?f?3J
z=flpi*=j<J?B_t<jqJ2wj*|6~dHw39uma*jQkuJcVB=>_cfWxs(OnPi1ttZ8Uzm>X
zaUb5&zW$Q->J`E?ge>USTMku&>^InUP1Sb{rz3USp-n~A%y>7`*-O%@=i48Dpj&Ub
zxVSA+Nq;CAQdwbiO->0>Tg=(j_dok&VAWbyZSjgWvq#Cyc$}!DMdrk+tNG`D-SfA9
z_XFSm{vE4U)3g<*!$f~^Q4DZ`CndrD!((A&+X_ENF2DE+f8~p`&AS59u?B=>oE=#y
zw2)|{S+8pHX(ITUWgcjse*v~5o0`q_i?ifyVYOM64%B5P411cYLRAf`XFc`RbFwiE
z!vTV)@7J_l2SOo(MRgsruFxW~jK!b3tt^evOs4|o`S9)y#%TOJV2q-!YkUZh0^~%B
zfg}`JTe6f0S-OYE;Y7|2i-HtADJFDPquLH*Y9fktyT%32;4;TKg9Ni`sM`iN%@~o2
z^<@m?Trii?7+7Ww#}kY9G;K}W_pH`ET1mVsGP)RCk+lhd%ZXYm5}9S5(Nds|!WfC0
zie0a&Te@DbJicQ&?pgO0rt3K=14qa4?mgOS5YW^u?H6CuU)^HrRaqq;?orEF+#q#J
z6&7llP;x;w3FTqF-!c1$UscF{Q&P0i(=-Yh3bNLP385rL8!BT_MxjJPhEyh-LQq?c
zl*N0YKvx3eG!n-NZ477-N+3igEROxdJ=SQV^CSdWT5P{5GkWic^F$g(0E{vyqe@D-
zuF+M4TNY9*FMKJ^-O4GQOD_U#S4bsUrUk!DSZxp@VXZ+`Em@r>YRQv@L?*#74ulxb
zke{<Wb(#o^qi$PbD8$PY0@+lh+#?m)_K3dt-uvl<ANGvLLosqv4MKq*4jj{rv4)&L
ziP9Wd`a?=jgm(+hFRc5XAS@~RGel}3rA%-WF$QvM8OI4kz?h0$_n2mjttteHJa=>}
zMb&OdSyJ~~>fAF9CxWu*W<>*Mqoo<Ez_|sjEz3NSoabV_!A(Qy2@!>dfrKakC7H$}
zes;8V#fP^GoJo0JYL*D&G?()(6oO1`1vySsTB7ib(~)sMW341C1^v4`2I6qw>e*}j
zen+fos=h}X#bI}(t|~$bT&#QQ=9-Ws&X?x7<Ke*d<rPv1#^WPyJYuw9y<TH%O^A*x
zBvu=??VA4b2CW5;AMUYL#qG@%Aq%EqkF*u@G+|Ri^h<dev@OGaN6v|h>n}JS9*`m<
zb%oM1RkdbJC!}d{KB3f^z8QUy&85KN7H(gELGpnZ9X@!*X=GcisjcGv!&_|A(exXL
zOM!PK$Mf3@)|(Bh*7A$L_z7))$uP~ldi8?G_wOJ$wtZV5NI8NsAT?X_l4g6!hwtC<
zVdt=wVsNnT%k@9ZBb$Cj@I?{iLq@MIQ9g70?K>Xcejp{!`syWF7*wrUyhn6B?~k5$
zKi=``o1Y?0$2^R<bYk4^+1_k<{LrJ1kF=|vi)Xhi!-=l74AP?vq$s(Mj|@yGEtto5
zJdF$Q{^sBDaR0!M?+;|T0VgnRz;+Fj7l=iX2G1+AVrIo~oN=S4Z!}>z@b=p`G~Jr3
z%{8Xq6fF3moVSNJzo&lrRUrnF$wFXEMSp#R85C)O!@E0l+n_2<-K`-XNpoTvChiXh
zPUDEFyVAR63~_lRdKhPq(t+dYiI=Z`0YdW+;!SYTYOZdsx%%P-%Ts{~-QPVh4=0*R
zg0Y;AN3Je5c*fGr+cvCK#l_VPewi^TW1?rh-q5r))vBd^@l&pT@|TbUZhzp-uYSqm
z9eEf@+VC{Y^w+nvZG}jg-YQhzaGGb5ZctTAl8Pi1vvc_Ih&%1EID&V~`#rT$C?&DY
zivIdJ_2!Bc95FdUPI$kNf<bG8m5QhhO9*5nRnrsCcuODRxlyvkxrHnxvhHZxHO)=U
zaM<GxPt5P%m&T{n1*%y=EH7T8RXJ|0sEi`2;-*(^uBf^-$;~A1%L&wpX?BJDg#f9k
z`z=yWNI&DI@JR+JVkkF~cSw<0tvcGS$7qAKh8#j^080h3Aw@~m4#xw|XsWt6hH0x2
z$}lerhy4!c62du%fpMA<$})%s;U!PI1NT3^W!&$ubw%CxG|h_I)b!m3U3Cx=@$ker
z48R`0co<JKm7%i|5oe~Q$b5|Lz;3}d$XK>1WD8c8X2T4XF;rSnYgKv&$Osh}$Kt4H
zv;kdSOunk{OR1g4(}{VWshW<uZm|6ZAq{hwS&jqQFGZdvieOj?N*6ycjU#@ZSi(qK
zR~V~E!O<y2)l{WE9|FrXGEap<s3Zs_OICnV^X}FcD5C(Vs*bE{hSLGJxMEvMET&l^
zm*zC1u?qY$f-;Dz!b^=XEuwEprvu(iD5YuEm)Ok>w%f3b134aW<EbE$jYaFTT$RW?
zP6X%Z`!%+*MHWcLFC$77t$e#$lZp6wPjCa9i|35<6FEd`TeI8Uu`Dg??S{!sq$F5x
zH`u0U3^QRK(d~*CFMdLnwxqk;4byO<YtCj(D3VpUX~200DIju0D9!5XCDXCSD0{Zb
z7MQwbJU-#vgqxOvyABEP7`sAd7*7MMeua{n?dFQ7-JXlf3yihJ#AiGfK$OTxRp@?b
ziK%;}%?LzMz_b;zt{LZ<N+!Y#b>FhSxI#)z)fPgevX+!eCMFBT>(^ft``)x*t)a1(
z#h@uv#X-;3bgL~P3MwPn-+vFHG&uWt=Hc#ywhh52+P=ru731NFxJ25@GA|?ZGI4$V
ziqrf7+qUG;mKTFa_?WTAqOC?*gPZz-8`i~utg0$pOrUi!ZChq;Zf{xq0wg|sxW`mI
z+uN@(O@k1T{rk5xRfAhjJbXB^y}3k~ind*0pC0j1v%dUk$$D9Vn?~HcP`4|5EL26^
zHi)`ow>X|(U*dE{uq`oX+&ocD18JTZU1B_(XfLk$aFYD(W8jDX@o)L7zxa}YJ(sFt
z3W0~ad%TMroZ#>N?sq)8fKn}+bw^qze)-)GeErQop;>QPk9(fI{E|=^(sINv^SQ4l
za`$e>-~R3m7Z(@&{MAcl7g(I*<+kVYzxg?$UGYEv`+v)FEajE!n`?ge=8iTMO!LWk
zvR8Pu<#%s>!;~evhl%g)$g7`S&_;=orce#Lh#Vbn4-=c+f#zB9YY<XFE*SYS1m?-3
zZA)bh<9I;o9;u<*tRXjSuWsp@mgqe5{X6PPpUYf=rf!HRuCF_^i0t-H<nuHCKkH<m
zTdmPDvphc1H#OHcmpr?@;{L}U5Gf#68>)T-ha)c<ORY4zZ&B-tUV&aU2&D<96QXZ1
z+G6`FE}y-ovIaLSEJ5%v1&;e8U%Y%qvt2PL#cmim97mcQ3qa`kGupY-npIsRGX>qa
zOw7kaY1+t<w(H0_Ax(!`-%{PYg1W$Aq%LVnZ#+2`E6{u<_SIAcsp%Yqgi4tZJ;}{X
z;}Ibg?P^`@JQ2mfLtC1rBaEj~k99LFBUT&6hkJ(IJw%T-23r|cZH12+CoQ&VKxpRK
zF$^bUjwq$DTH{o)<=sC#LGWl*41=Oi%;SWL5!)0gUPyteuBqyVuItd!6oVfZ5tYT%
zO_?}|gvg1sl!w{uJf<p5bulD{!_4Fo!8=3*sSVZTbCR-5LJ;;l_8;Cb4o7TT)2%O9
zJ-ecAFW5i6XL|ZTZ7tqKcK3IjrX#j4#~r7IcRxN6mYK@x0yt3yyShL%9ipmXnlQ4g
z_Cta>Dw3>_X+a8T*K3-#1~=gbPf()pxU+z&CFlkr3|Rz>HaNeqECFd-qzBd1BvX@w
zA^FlX;rv4K6FNtvHpR9zCZx95*5bz_F=bdh^EiSOY_FfAv}Fm2L`Iq#DT*vqX<evo
zJ|J^>01Ii*=cdAG9`W-`ir`X08H3OoZCgx?CCd;N0#LPWVf{-*ZZ1$clZGQH7p=F<
z36%o=aL;nu<E9e`iK$v_)lgLxQe?=A6iXH(`iK-oQtF*UONCzsY_~>MEz{|UQi|2Z
z3$iTuT$vooZfU9&VR*uHmzd_Z7%s&~$bqhHnR38rS!j!HLMwwGClrxP#M&BxAh4ou
zHiYESRmEZVNZs{?HkmV8>9YFK6{n}Sn7SuToo84R-@3&QMSAZjy-Da*LNgdzK#B$e
zgz`tKh9<oT5>Q$I2_PK=rG#Dt1Oyb6sx&b)sX`D06a=KE-JEltd!IWWXZE}^pZ3h&
z?^^4Jw^;LkjoI4$`Z_wsIz{uNT#?;$=(XV}x9F%e?ftQLK(^3&&4C?ATeo*-bY(ry
zXN|v0(0}&!iemPep(k}JH!|2$3Q}S{b%Y=e-J~KfxPWzZ-RGBni8Fk)sk+{6-?-3c
z;{oOGV&WOOW8vmD9-UyAa98{aF+u^Cb@D?ItM@XiV0Sr(k0v=eLK>H&Zf@@9)#5*d
zCF)aFW7*WRY3kac;(#mIc-&z>N&NNV>KB*gymq{bSmH>iyq&)_O9m0=)UHq;c_70>
z?Qhph+HnoiS>~=b!UmY_2_}MP6|phy;p+9>bez*+tRdby!sc`U{;1|gg|S{SOUt!~
zD6^1@4PHlv)!wK9Tr4^bRl<{uix+=j#5T@Mm?##0w|Gbl7$G0Kl2gb0pY7i0I@rIs
z+>4Ke9XC@VmOVPJbuBbqtn|>6W}DZJP6tZK$9;rtTK)^IzSw~^!ci$1OFmWZdPGc>
z1$<pBV3PDub&ffd_S3g%+j8&uSZ1XFKWyu}td>>lyZF(3@sJ!rW<4>U=p)(pZCr0|
zjyh@eD2i^d(-Yz=ib6Z{#iSE1*9VDGcXP!SQhV6xe_b+wr`vrcQU%n|$%P|5^c|p+
zQCg0)&ap`ME<9V^HCsvT{W8dhLzUPuZ~vAT-!(+xNvMg-Y^?}%dz6VxvIX5Map{rJ
zs~5Y<B>4<HyBFsL`(2q9$H2Gy0@jivK^v#vG+`7e!7eXFHGV^r0xW{yK0bOgJkq~)
z%3L|OaLCCic^pWI`REGrKKa3#@A?=erP@o3AK2+9;z~PvMh9%9YW1}`eLIRZ*Eior
zha?}C-ro{j`mH(VAaU<fQ;9au%~K>}90$qZtYDg`#b5Hb+bh@1^FursC~f`3<A>>(
zCWR+#Pp((R89brz#i1rKC6HIZ+X&y+uRq|k!!pftWd@==|1mdjbvc1Z0`-1(A(8CR
z3M0LFnRwYOfT|X6cfc3SupKLak)Xya^^oU0q^|Ek3I}w{NhmP)pVBR5qYt*4nd6$Q
z)D)*pF=Am}6>PQYiZWp^UJ2XG3B*yUqBo<I5UUtT@jJI)AB3feiL0~<WTkXecN3}K
zMwwXL5j`l}IAA2wxcFweShp>@7hi2;Jq{Wn?POtzHh0<24JNaGhg*7dKpNAkLW6F8
znBx^>7&9udZnrcX<{r`aP_Qg<)8jp$%1VK%ykIj<vjorQh1s-qC1DQ-)n241WtW?p
zp!KyNbM<-QkEt+9mXAN?%(U&KP>?Z+Ihr!#501XI0NgBGLdP7xC(~f>IJ2kKReX0y
z)EM#wGg9VkM4^L!^VnFR2vhLl)tuJEDG{L$QI-Cmz!kwJ*RE`v8B7k1fx{?Gs8@T?
zr|e1_Ni>81aswbpC8Gf!g+<J#=mb4@R?Tc@?=&jcE;SY0W$f4e3SZ5KMcv8ylxb>?
zQdJaZ+=vQua=xbTCU`g&H=R^Jrg;X6y%%xjr1TG2n$m$3+Fw%>5`nwA*uWIW4r#O*
zEqX8rxr|u%QY#2tDz6$F-{H$367$Cs`byYN5g;w+l>d4*FX!`ST|ji}C_`?o+}I8<
zDCmR*p=;|uCSh=O(L_AG(&=Qp!k}m{^2pg0G+?I)wG)v1WbnDLPU)Rs$BqGFLr>M3
zM&f=B;Q6N#HK_OK6<y(t!NY;hzs1-5+N4d5mm0P8mfflbo0?Iw2|f;A6S7-&o8g%2
z7@Y72;e`WGCSd2XSE}eFJ)qO`fa0$elXh8?G_v_J8t_4GAV#NLdK`gKvn*5@91S>2
zXf3MGcg<|OR~X))^!}bs+lgVyrj!F4dJ`CQOxUjch0yu1(={*y1H9`({gj5<$zt;u
zU;K^=3K{@DQ)n|CP3``@Fu~^L)GjMs*8Nq8-5Sye_g7Z!a4!;e!K#RqE;Kk2%7IoH
zS_jODQ(hCZm>*}NR=sPMdP-=*8TokajuzeHuY_38aVU9E8QF!quKb9S1lT{&6#23@
zChF6~aB$D#50pKlcsw?4EN{<Bh1GPt%_--rgD=9~DPc5H57QVK7B$Ad_HUC5j+k|y
zva_47T&DU=Qu&{EVwdcMv_2AJelZ0X?KdyHH=DZl7*8ik9dU*xY-v$Kg7gl$e$saK
zMNSy6sk5nLR22fe7eE`%{rEy2?*P~37G1?i<rX_GL!;sGmgof9y;JJhj!^#yyG0mf
zy}ijTcU=P&rE_#N%c?-bXtr`{bWk|V?TGHu{_?Y^dVO6q$>5yuqpkcy7&DgVob+3m
z*o_-ge&2sXv3e!mP2l3s$~C+Cyp!20`LafnEXg$ck?vQwzFr`%?iMR^5;Z9*f_`=+
z{y<wHV+A4m<we~_Cn)y&Y}nM6$ziB~b_riXagDM*@z;agTV|-qLJo7w8(mc@%VFhp
zPJYM_KkUi06mHXPV>V_v`fR6^RbFEAG$q9m%0*)Y;)cqyQv+_#`(}1|Q2}Zf*>Gvj
z3UsEaf{jw2d5{;ftg5m`spHAlOPKwd<`Ts-ufY5DbXQmPTi(7e-~BxCJMT*QfTN{F
z-Ipe4P+Qv`W&fDz%app3Np|L7Tdb?Ei=D=OdQh;0Qq#<pr>20kywsHMT}$fvj}3-d
zhM^H$0I5s?G`;PU6sb2(zfSl=J6-R(uDTcT6-&|s%-9cp;_3YD3-<W$oDE`21th!S
zuO4G~H6Od+zCI~7ZJ18^T>kuftB;v>n?;D=0Y`rnhKQ@a$*x-F;AaS%homNv+ZBW|
zOpCCnIfkN+Tk}=Eu3D_+K5m`&mk&k*z}Ft<xz7D0QDZg6Z>mjjJ3gpvm0|(h=He9B
zkYEhX@H(pI1p^Xz1hBV)jemUQQ6%rLQ^982_xx4ix|S?>)7>VS(&964;LpzzIo_qF
zluQa`Aqg}41cj8ZnPA!jRmT_VRzIX)7=KF^;Kr%&v4=L+?537Xz_+?M2v79tukwj9
z0Bkz->s%yv^Xd>^hQ&%eJ_{mQAXS-lGxu`e6tum{<;pg{owP!a!IswT*VMCGzQgR#
zm0-A4neujw)L%TXu!PN~04pHlxU3eI_KBkIuSVN7me2MzR`KPraJKQFduB;ZzY$9r
zn7ygNc3YYFST_-?m$Xp$Zh)mZN<h*?Ua7e$4(fxoe41q(E0?<w2oww1I@)Me;N!E#
zHWzx2%(l%}3Z=M}t0@KuicJc@?ncNv|E&a2XXi+T&E}ifWSPAckkkClm&G-LOK=1h
zPgZWQi}RfLByy;TSfzYV8AByUtW|jG&F^vVok&~G2I=vDGFFKCvpN#EgdHV;ky=P`
z?k&hI$BZg^iTojI^YuhS=@$R7-}7s66tnNY#YRsJ$Fjd&8==vf3UDCA53D;~?hA_F
z<cg`>?`Rm*d@_F4N2W|{FDt#<_Bmjt5HDmkb~H%Vg_w4hemuI#|DN<|DrmK><h0UD
zjM;dy%1Fgp@*BGM9lkWjEsRKQqXo3;jcob^pAO|G94#G~eLFs>XTGYWZ6VS~(-&)>
z10AU?I9U#lcO?+CBMz(wrG73Y)!&|I9j9$P|1*YK{~380Gqm$(G_Woc>QR4HoIksa
z@woIUXxd*YDK)%p{q%(5a&CZW6#H)d<~vYkBn{fTy1azhd!i(cHsHx~@(BW6W1LmA
zWITD_HIO`j_ES<~#BB5&#BP!)Va!SObHNuIz9JFV-(Mc(*}J|UHe(D>c%JY*s7K$$
z$z!4(`UooMbkd98?_9na{p^k=n5i2BE`^H(EJ&&^oO0PT)?7se%+<8N^iPXWw^#lP
zN-r|d(p04{+LJ#OJykH{3ODr$F{Nk*BR@us44v>oM&W^xkt5zcKQwed{z5T!7vFDi
zMNaG%BTe4D-3&Yu!me>-SJY`d1H_d-U;*`lT8p_vGia)(4++7~5Ka<>B{aufFJ)_;
zb0yWJ^8_9&rJf$;?SGpuh;`bHHs>rCOE6+$14x_uj%Ep{;WX4OZQyV#8n9sZY+XUs
zIk;Qhgr8BA3gsy<{8a(NV#&9;$wA#+hj|o!d+klhaBXH)jKt8#Qftrl*2pproKe9x
zxY&m4j`1T2OT;B3Gi-JPEL(3@ac8(RtK~<0DU?b>6PEbeGN`SX?n&{KPVhwVg6gyD
zw5gXHd~Q6JSX};Bgkp$0iqn_H?5bZ6<{gnSJj4tJm#R{dd!`gv@Hvy}2Y+VKpT7_$
z?sj2E#IMur?Wq7xZ3=WX#v#)jJa<L3gnCqRtl_n`B&`CjKX|bs-jR_<kQkIWKl4lE
zOelkX<8W|2&_Fi+W4R63PVsm5u)u9e+SnU4>IO1UpXb#G_t1=&atcfy=5%5flz#fF
zjZsX;jK52q@Y~29e4BpzojmUl_~lHt;UHu*QBlk|o(0A~-@eBSpaU~3)D1hNon-~-
zLG3lC=^=$YCTutqmaA57XscYeSpI{WRhR152G-Wvg5}XozH|u0L)G7Ul5?H!PiYCY
z@qlcc!)H#IAy=!XVETs{R=87WpUhBV2{b53{n9Uzw{V*Z;Zt%ZTXol*|FdhK++_G-
zc8Ot!mPUt#qm}3?Cxt367_OT%asYSnyKARjCNeo0&bQ4p7bR4hGRB#uNW8rASs-Ux
z=uEd^x~5L|%!lW=(mUi?2dhc5!&X)#-ED-Vnd-|-sXj3jOY-$r*9Sd&+3_fKSkF4#
z>rlL9f3AH+T9rNt8!flw;WH4kuue>8QJ+@M)LZd+#+>a%!RC=z<*r=TGj}MvawW*M
z!<O8ONx!}da!TNmgt3R5OE0R@F+(*)2tD^_-aui{344_k^FLZ%>5VmVA6l=$-<=oD
zo?guo(&iuO!_*dT^>1zErBGJ|;xfL|)NgTN-_Z7bujbZ?j&TrRJ5ZUp0_)q*IDlhN
z0b!^z6GE^%ql#jSV_4V4Xm@j_W41aF<W^s#bEYsr`J}5*XntMyLlfV|b110taW|CQ
zUbC>82^;Zo5dzRRPD3-pCQj<hf*18$9sMbxH$*Io;?!_`4L>1l);rZ%-u_2pKHmxR
zDdceFtT!6;$GR_SXrjZdhbxBIz5ZiGvZH%rp9GNzd>n@QQn(|3XST?G>Ji4ny~s+|
zMRk(9GA@fcOVhFq%&7|A7>Ne}8xY$2rgvEDcE7;J@DZT$u#IoO6~X4eJ|Dv#a=M9Q
zN*9;g-Y>Uw$C%|5wWm9~oHpD*erP67q#z^d5nDrJzM&o$BzsUGMHoov<#<l7h>H76
z=XTXyPds7xue~kIIE5l>4H|or!*rl8x3rPA&dtBXde^(R7%?wTkj?DCH0g#{W6QqX
zJTC(Z0`>LpMhJg(+-<sFApH2s!NJWhGkF8arU}*!W%?th`rQE#=z3;NM+L%uW<_%%
zP1#{h&-5pS%PrOUGSg!0$=4^A)tTk@9a*4KA`!Z=UAnJ*H0C?`|K`sdm}#1GnR`Mr
z6b7<@FKQMuPLBDVA73zckKHf8jQ=!Z8FW)x{%va~VqB)20?&MM9TUf%H58RZj2?Cw
zqUfb_xKeVaVU`&vA*TGem^>`is#$PqrONa4SFL#yj|}JXIMk0R=^|DS{J-SBdd&wA
zI~N3Ky!pYMzcj-N$W?NC#lrX{*GNkkoX_@GVdyQwireMZHEs>aEw-%hCc&R;x{c4|
z{bu{4u2@cX@nk}|ua9W3FuZc;<mLLK+A0*Wfavd#kNGl6`$@5qXowj8WnZcC%8WrP
zIQ2H7eY{Xey)^(bmd^Jw_%eYpPtCw0a=RtWa`uLR7X{N!rlcf!&NjbpJx9&MjH_tj
z+wO)X+^Ua~qQI<7INDs8t(50Gu<sQUPJTwXSdKBkFuO>_(qh2CjtWt<5el3-$CkDW
zy|yd2eU>9#kq9vsd?IUEXC@JfQM+2`<o>t7x#h#KV-bbuO_vvsCB@Y`<us+F-7lg`
znA;3bRN-|4s_BaWe_2TZz6CTnyb)HbcIU3=1xPt(v#_c0`4?@Z=3nuQzrk0bxS8PO
ztyY*L#l;FKNxLW6IG33UO0;FdK8%B|axHRweBxvtC7zfyIm6(&6IcDO(&hERN7A1@
z-rG|Y_puG{LeL=?+_nZ*XueH#1-ZX@FBP49jm=_=MZk8}rdggtP@2aMmX3bNRO9y;
z*2)adbeH}0Z1WQgHi;;Kmnc7XWoms?C~OXUAPu)dE8sFpgkp<o8)cLvvD2@1>4R^~
z5cw)}g4rMl4hUg~(IvvRUS4GXz3FF*R6&0AHyRrCnwCBBwoMP6D9qBKp(Ek?S>Ch=
zS&z@OuE1sBb7=nplgr7>WAsd_lJ+9z77ENFl#D-WX^!^8R)iX-7EbKRS2i)Z1ffzI
zP}<qD4(GC|7K_oz(5n?ZX=k>L-tf{%zOne9TWw1fw|0w4^or~A?`!|LW<*hz{KEqe
zw2%UcfsG1zAEa>&f%TvN>@K=%Tr^9!(ujCb{m?1<tkSDrgVGE2rVZ->%RSZ3mnlB!
z_AK>F+}BwIa#jqD4qn#%a3#5t*b^Ai%4&+QzVg{MxudD+N+xLckyoyjf_lJ+Pg{=+
z?`syU1%}~^(gw0{ICekfrh=I7+x42U85RQ&tRQXHH{)<?Yv)tNyLsCW|7HZmyxh~u
z(R4_g);#Xb#8T*36msLgSIz9!Qdv8AUUT^*9QQ3HSr&88tpRQrN%7pMb6=G%GfO#^
z#Jxc=?bf}Gy(pr1eepZKB4ztO?OCM~HJ${7Ot_u-cGn}p3>2NAXb}EPmS%No+bl>1
zCfxn|ONU8ul9zzl2MXiZJ$<z6d&MAL&|~pvgLLdjN5aZbB0`AQI`Z@X+Vl5b9A-G)
zRzhA9mhU}x@HWK;)HJzHul5?j{BPYa75U@-U4zXT<RqCc{wth;92cs7c&NC<mbCJ5
zrC1Wv6d4rF81yQOB0~2RGF5umglG;JHOFC4K(F?Hwcab@7dcoHuSPB9H^V7m_W@ex
zv~DGz04gF8LZYkBiO|Zxz@}=R_i^Z~Oz7*ZnW5KvT_7=A)m!i_!D!&uPGKWgpXJ-%
zLxms=?(^L4vKn2F#fcXJW>94CWh!_bDxnnb73APwvvjpW8jtN)Luaa+qiK1%>zM9s
z5uWfec#dxh32sAo9ZvMA&HiTW3`n+x7|Yannz5qU*wj~kXij)PZP3{wa^s1$5H=5q
z8m1~m56Z?mr>Gjf)4IW9;Pf`SQQ<gQH{)TS;a<PbidgG?L2rohqfTUu*oC04eEOSA
z$sW8#YE~4S0Z@@+;$#!pxhh%pvM%+urxl~Pw6>XhHt?Q^3<k>(Br`mSB@J$-Nv`A1
z*h_}90%pI=@Tt;O3LYv2E{H<F9&b#HICP#ebY=b#S3}cdhA4^h+C~ho&V@ZcxD^yW
zmX+y-W+h57nt}PWU0D)1#J#|>aRLC|;6sL5h+M1CDTTKqqbh<fDJ^ACb?a8NbbS>(
z)Fo8d`f@d(w((Uh%kho@1C=-{+{f91rN{IykniKtxBHTt?$OUu${M^3Wse30+=5F5
z)Sym_0^-;CoN|*>vNyj^^7L%vz3nRO8G+V0`ag=f*%EzzN6%z8kHh$Gg6ELRsz)*S
zZyGLZC$fCE^B3CG*&#AEI_*IdP<x<6I<3><7@Q%ZEozobg;0>A+dbC5zC0NC=xMD}
z=<G{d)#V^uMnWZ1>d!6ean8WV7;6S~b03S@>2`g5a&-4_NiJvU&I-L)Yh*Gzx<SiT
zC5hrBLC?Sv{U1p&@2|7U-bNBrqfQg54b2*d4HkqshjBJkH+Q-R%!USVr3m1?9I<EW
zlg}XH-wxPq{S7&ah$KdODeYi@^i3c^Ef^@2HJn}fy$ceoabCm8<T?Jp1HKhka5|(0
z)UjCMc2985^^J7S9Fv54Zegizl{?Ps>8(76Qhu-r=vy!S$8c`_tfmc<LK%{EJ^~r)
z{QOfvw##cohHh{Ld{|I8|LDk<XP}GXo4%W!!;*Yv&EWBgL1cw0m$KXy8{}?Pm;#De
zX%Ok&>TcvQTHXWhqab$Wi9@dynCEzsE$C%k#2dVs;+twmG2Uj>*yf8gWOuwqTi3_N
zC>=H<ct{C#$Gd(DnzZ~XvjvA6#hN3b2D{+3u^H!o`pR3M9JW&w(&J2ZP3{8f_>qW)
z8}mZRaXBr~k3NP{H7X|KAd{)d)xu|J>zWXmKqRnb-#&8j$nQ~%1QW_Six_J<O6BHM
zH=*9ZW3l>LA>PxcXoA;z1a)%p9^UH(G)W?`ISI1Svp=H$fYsm?q@lKYj=S8-ixOD?
z^;T;pRVPQECJSrR<wY44<xPYe_ap`VLvr>Kmmmw~nc#_~zr*pUCD$*hoU}ivU?#;S
zAwl*wRfp;7ekrzg_S*L=<ua`AKmw|4<erB}3>#VO5IB_9<c$%3Pyk6U)#-B_DkHAd
z{3eqAzUNO<nKmFfJjq(Q<k<A)@3Rr*OT1)I9`ldZm7}(7F8Q_S4z9_I=khtgfd@d*
zdC?@h60$aWr0f?GamkH5Ds1)L=+!4Zwa&#|bgh6^x7L>CL6?uky)NtxtS-`P3E24W
zScA$EW`QleEp6WLCqVp*GWvDwO{5SSJ#u`w9CP_i^uYJx&qd#06k!T?aed_Pj`iVT
z@Tcu&-W*F*-`TrD!j1<xM(?#~^Ha1(&QuwB;1IdIM!qLOtj3Z1!pBOQN6;QwQy%uQ
z;Wz2GRX>S#M!(k~ZKja|6_5}6CP$8w#R+(F87Td%dJmRjjXS9xQB!6XtZBF~d~29;
zyx9<Qit9ppa+T2UVK4Ya4{pgmmfzkxz*@!jnw+0Y9wHtbmu2jX5GO>NgYK4I7ws$t
z-28iiZVD0~-HG{4`~=)mk2V;U*c%7ruIqx5&(AF~cVm#HBLK1+k`zdcUHnxIT4x6-
z_eqje-s+bj0kCh+LGP8O`t0AYqxT0uZ@tOqy`;l6&_BA!h0f!)bM&eH_TQal<nmm#
zhk5O<eA#^@GN$S1;_@=|J!pSj6m%U#(h)qoj?cHb>8g7rRT@~PK3IJpFDg1xtqNrC
z<)~6P+A;WSxTm)zdbY!U_GiQc&py0?7Wq?lGanS7itJ_^(>zSZZyu0|)){96!bRi?
zQfK%0f`r?vib*Emt-leHadNolH;EUA?@51MV)Rx)zmoKx4ntFs0&T}_mCoBo>>dT`
z$sVfmi@(GDZd@bv4VC2z0!7gMf>hU-ACf$PWKnzqKIHu=QkvwU_k)38s&%eS`mk{v
zN%mkQo1UF)p9_`Y`Wh&=kU`gJkapf;RfEM^4;>oqkT(|pgv4Y_9ctekp^PGvpOvVO
z;Ii_ST}gNCwm|71U$sB~y;O#yj3Wo*rntpo|8G2{>cY96;J?fNw+bn1WcyS9f1dZB
z6KK7g)^z_V{htxY&BLhwSCAlyo_ZetzYj0NiAKegvi!fd8O4!l#G}HzZ%anJ`WNLe
MGqy5n0^fV`Kh{f+)c^nh

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_002.png b/0.16/_images/sphx_glr_plot_visualization_utils_002.png
new file mode 100644
index 0000000000000000000000000000000000000000..4a92cb04180594b7b1bece71ffefa570979171eb
GIT binary patch
literal 247376
zcmYIvcQjm47w?#1^g2Xm^cua4KBEQEdy6`HwCI8uz1I+BqDMsUgkba<EzvuP8bJsW
z!Sj9Vz3=^T*1hYVyU*G8uCw<3?Y)0HURV1GF##O`001CXS5wvp0I;I}Gw{F<Pbk<<
z)E+KUzAC1^2A+<-0k$t309v-bUe7&!pS##W{T*KTxOjSq3P=iw@<W|{eZ7371qI#z
z_kw`u3n#%zxYGW^BKTfvW<CG_sqKFTR)xZQ7XTIjpstKC3`GC99k1mW&ESZ8Os*%s
z@{&0qwLRDD*$+FTNKqg<T>-<qN$oYD;OaGig@^!=j4Et{l*K7*N|}P#`lgl~D_Q$`
zDOy53B8+HJlZmw`3*@(CJO$Df$4Xw-r}O5lxw+Tgx9wq;3bo7&sf6KosiC0%-X0DA
z-uWxI_m`aV|IQr^`#n>L=u7?Y#Dhrf(G%VOFY0%TNAN$nj#_5K)_>&xyYj+sHtv7_
zZ-=9f<&!@pahQ$!+l`&0g*vkIo=ds^?SEAGXH4_o-=n%~llwz!#BYUXr2lG*POo0w
zUtLSy?{55ST3h{}0(Zk7?~Z$J+pZVx|61R3-Tw=~iLSHm0Rry!?#J(`g54H9iB~)V
zW&QuHZG`kMcv=_Fdlzs`O4&LFzWii6Fi?2E_6k#Y-*Zp?>h9{*(9IR7?PdSe{kEsq
zee2HMcjA@qUh~KgaHC?!G{+pDZJ16-i0s*TVb4woiL8S5ggu;6$Y^5g`uc9*#`7en
z-&?Z3O)x<AGR6N0E?FeGxNckSJU^{Go!2{&RuYHTF0%?^LQP<^H~fxirG<vS$H+m|
zF6O;@PTD<XJXT0Vdo@vbX=;JkC1I)wAiTTBSSZC7+Xdou%tiPOd<a`a=)FA>vsKTz
zXqdi5V-Iy?H3+BV#x8$Qv%eK=3Z#gmAp?At&O1&stPqPMpor(Nk~Clkl5;6zvx3(f
z;mTwrnuT$`$hDqIgfh!$l6|Q0)gS69AyIO2n*Q|MMXediMG5(~bxAlCeD6FB+-$}7
z7@9Xi3oC@|2@6fe2azZPDv9T%Ux0_a`-Y=ZQBsV#Se@?9p2V$wt3N&4-5EpOoNExg
zqWXI56s`mT!)UGvkNsQXwj1i^Rwa@2FxEbAdE8kLZ-kT%Qo8jvCA5dlfE*vaGmsR&
zH(YmpcPlv|HJx7)fO75YfPbF0=%1&Xpxlbz0u7&on5VcVcdA>418nA@C2?xX>>{Xe
z!<aW}0TuYH45AdY@s5l)-AaSg7O~<gs&>}{r}1%7c+lCRk_U~9Y#r;%brChVmM1fB
zZ%!AiyLT2&m|+E6tQ1$F{oN<~HA+UpA=^7UAbCa#d}XK{IW24mjx0HmF?U<^kZG1|
zUia>io31W(i;T1R8}B14ycfR{xGhVHw#P&sjpl9NwQn_Scy(^Dus)oHlvmfsjQ6K|
z)~C$(|LE6y6udWNx<ARCzON(wcX2NHsY|ZjotQUbWB`rj7V^p-_V@GYb;*9G%?5D|
z-vl-<5c*M&RcIGG<E850bd;N<XyA|-if+1-1t-VN8UP1YCd`McZNfNH`n{7NWB{Nh
zbI#7Uyu3*{E+EP-W#CMB(+RE_r22z66rv1P)l3pSJqz8wF1Ws1fdKp4ZV3b4$lQ6}
zzq!BhE4U{mARrVutg-L$ll{rKQnPhD4zlj<-r7^RO~O8vmG%#a<GpM}SQeGQ*#}S_
z#4D{^ttahko`E+nhHsBi?khH~r+7KEUx|icGbu4~;Ul=pl)MDON(RwT7Q71X2|7Dq
zIb#CYh1mUQ`ep2MV!MQO`aE7ks<!T8i~f;;3&0TE=0!i;hT!>4OgB$a1R3gyqUiM3
z`V)F{xB6Nkqa2BS>A+rM4pueV(;V6BbRki(qeb80gOnV|hxn1Gs3Jcl6O>a?J_2$o
zY}3f%3mm;JS)Q&bJ(g^)LC`_-yqLq;4Jj=;5eDfc;hWN{D!nhYVqfB7lX0^G%6=7!
z18^u3CIIYi-n_5f-XMAz`*D6+ek%X{G2Qt%N@sYUz602{?sX(HL1zOF06~S0!bPt0
z+=YlWy|1yKvp4cgNI3tpF%lji@)eFECNhYUp}=3FT$Xvh;k6<O))K97aUToMv1y7~
zmw^H+w*7DLQ&*<iH+uS-S?pPNF&C)RtAJ~%B39TiZpt~R#{J>P%IZ4WK*`>->3jF|
z6pO2DV7o#pQR|twNj%l!aYFmcOKHpAyWzzwnFvB!=OLxpw^*oS1r^VjTrAR9@&2HK
zdORp3bp>>_0?LS}@q^>vE~D$EXo-x9=l@t5rjOpwTSR_Pq%`>@v<Hu35uN!P(}lZ>
z7Y8=q6SbTYW-tDHEd!m#2F)_SHId<?;XKs=&MOb6`s}<-q4#Szwg|AFY8e~9-yg!D
zBxG6la=FuXBj9;x|8aM}ruQqzK6TM6z30|MJ$H7_c8hJqH=#-=LtQQI&O=k8PKuN{
z1*Z&){+yj{o!en2ixOZBNR30u2w+lJxQn94P|O+U_#3T3Z`s^SCSjp97Jc<Pf(h*0
z_~YI7w-9@}%n?nrfpCm%61cE;ug{Q#783&7B?1to-K<Z`WPZx!8p?%jP)uD7+<Nca
z<N*y-qV5KR?i9GWtH(7P++$X{>{nzw+K8`)K{vPHfMKcJvjhTcd~7mNp;UYYr#0O^
zqQIA(J7?{-J)Z4rKgqadfk4y^wk#PF-X7vD*2IBVz#WOuFBVWf@8wO(%ag?p+ZD+l
zsf@=BSX#oNKnk^jzkaR1gi@4m-{R>Y!OfyHL4a~&+#W!$tF41>E-~3wabcGC2ZSLD
zZ$&`tSL{?%m5$9ICRnIl%mON(j7rRYQ*>`=Ozrx0l_DI-novuBc?UnDAE2C8wo^>g
z-Rdpb>{2!o8WSVSov4%p#>1s^qQNwu$i4P#GY+VBBd+4Y=_2`EpQ(e{v7REzM7PMJ
z&6V(|d)hIEP$!HYN*BB2t<kWrv%wt+^+dWCu57rbEgab7L=|}rPx$pQki*PRftMii
z#j)lvLz6t8WPi7K^YB}vSik32xZNRtWI{s!C9Mo4jfL6_p}ggVcQG&>P8Q;Du5vWY
zQfPM|CuMJx|KHtW{qGoe(Pm=Z{8N~XL=i6;RnA=TOK^<LEN7Y2R|sGsIR?(2#!@W{
z(#0}A(zG)K=uObZVTb!|p8olozt!|v=Li>AW-KJpJip1buy8(w6TZO0pYz4L03!c(
zN;m?s2;#I{@X=24!Nl5kl=8fa#RE{q@hIpRG4BO!&v*uQD%Ag;3?)awF;#Q`?9dhO
zNW0)jTuAE8(AB*8aCblj);w`P(K3CrUvsxoNiQ6mRfHz$z3~h75yP<7-kueY98uw+
z0r!ox4;V_d9eLNvaBdNBT49wP#)-8rU10To9))DJBZKF1`@912;OoRj%q6r~7_pM*
zMcOFZ(EL#b7rE++2Lfx((AX3;HP}X2>exh2@)tXcBr(lfH=ek=<g530H<N!9C@IY)
z6}fIwj_1#QkJDeSfSY(fYhl%a@bC*EKClq!SuD{$ywo=O?y%d0HcwCIl?Nlj5}pwM
z;L(EiahuMmT-Z>j;U9pk`{nIxGCyxpp5-3PcFWgyxlg}Ts<hGMkOhA4#{p?XS(<ev
zu<)9{6ao!lE8|$$&x2?F27CNCNP!E{i)L@LCXM}L7}MIHXm^^A9k0j`na(S+U@Ot5
z7o~#pGT^6Ug2JG7iXyg+O3UCD#!6z{Xj9fk5(qmx4uJqWl$<KCGKFW~nN_LQDJ^|O
zSB-a-Rw(E@FEIdN<k5#QiH(CpXURqjwc;YE;&oybZCQk<1ApaqG$u^om*I<`cF@kY
zIV6Tw;UiJ3g8S}WxX_@qi7au#GOUAmDTa|u+`wwUNdIUvmW%UjQLF8dwO40oZ`jNJ
z?lqL!gwTN3fLb6?5^OW$R95)=H_>EB0f`BKZa=jD&2fw)cxJ#Ka?xe0DKBYYo5dNC
z1eg^rw_l;m)pVrMtRYz+K(7wVQ&>&)vb7t(#pTzZeoci}bgI@3sj)dh9iV8<rpY4t
zi^zl(<IRc{jKeF8DD>`Y=3{XIH7#RZ8S-2G<o2w%Y(X>|)JeVXJsBS#?giN~#2@Ex
zcONHtZsu*0P~`x^?N&_BxTMQ<kOUpP$`5{|wFV?DjI~;ZC7tH1`y*{>O9@(>g)QZH
z$`FA!%V<v`t1}!VU2Nr|=Ds*NTKAC*9~F7iF<0;<iLfyquo_Jjc>E_PBt{p7X@6oG
zrX<2rr3eAny#7m|Re>{LHmC?_8EUbnYVa!%?P}0>D;F}D5K<$F!tcWkAx-+SX>V8F
z^y^w~@81b2K3r7kgm@mii`Ju!F?6L?DT0uPkOF@~iZ9;@vv?WfsSw)R=4tKyYW4Io
zuxR5a0FZ@(4Dc<Oz3%PzG32BGC{y`pJ6_NVX|0;~Pg#5HFSBr7|N26YG*Hewq-US>
z`HFCeMnu+iO2nH3d;1TP;2r&R02)_P%1`f{9i5~56)P<TEFEx?vZy{EoaPR^mZA0;
zh)P1mVHxx~66^}$Q$@DMc=ggkN_n{pCpEh{l%*%6mGP6|nb(gwi3lJc8woE(KUJcZ
zH`N@AnfQ53-lYNLZ5ONU4yR2}iq<?*-i=318X`|;BcMY$r&8qheZ^5M)g+}>bJ&z5
zA0rmdLi+5hT+737O?Rt=F*m;LQ%CiMOgZ)dJLg)OJ!S1D&6p1$@=<muq$WW2CuxN_
z!bm3gl}(Ra{f5WlvcD0{P7Z7c!bXD4keiZ=tIstMZWwFc2M;WkR1&q>AaVsB#k+%+
zTWzKJ9#b&|lRcq(dLOMYK6d%}!C0bGq;TdzQ>3*lh1G$pOZ`~Js#faBxM0qg<Tl%)
zYg@gP4dR!`;1d%dTjRPMXp+z@PJ^=jkqF3Mv)hgGH6M{Q5jdV9C2ka{2H^xB#c0u|
ztwwSub&K!3B#EYla!wmU>18R_=&G^YeZU$*e5tuq6B4Edu!yu%N4SKt7zW}^2c`%c
zYoTU~kjvLFXZyEY`X^nDcG^8Qk41#6tEuEv*)5^QlZ+C{52SvAEwmT)=w}ZvZg_kR
z9iJi2<{yHI^oqT)0S-QwL|<iXcFtx=<6lr;4zj+~E~-Wh`yy0Gcmh4GAv?ht?74YX
zNn^j+pZe;qsD90~)&+9$dAgFpILkZ=s!G%^Ofr!Qg5jn?mML!B``hZQ7rBF9Zu<Aw
zpGX2Q5ZSV!KM!!$t6SzD_7LsBnMfLwvrrs%2njdv%DrDFRc4(@lAEZJwuGlIiR$U;
zjqc|eDO@|8$U|H;@xtPSt36y2QcY&(5A9Ug8FSBB4-0I(2S(m5w-?(fD%o_{GC-Wf
z5f*NX%c_B+lmaAFDic|uH1zERV4Nl4a<(CWs$#zo=hFxHU(M5S=H7qn{{*@D71SkB
z7;!C=w3x$hsgJW~Mn!|o6!qxQH)wrdOtdLdB9J6|e{`y9Kh3ZVltKIq2LfROSK}Ay
zPn)xXi*S2PlCeJLdE2_#-&wa`F+jp#+OLKaq~#gaQ1KZ50<NE7IOv0v1IT-sNj4j%
z@CMdEI^m1zZp2Z{nn$~*q-<;5;#FZ$=y2<X_r`ktRdSpuV}<>v9oT)6)J8fCPhkk<
zfptuBTrJHO5!-=%=_q8o@#%yhPO0*nHI^7f)LtV(OD$x7Ao{Z@`P4p<BjkFH1d6)4
z!mlHMd<aqKt2CO~#5wzT4!2Y2tNYQDiy*5f=?euy-`2Yb0X8!e<>{9R;;1yaqThj@
zQDzhPJ!_F0qM`YDs<HUD8^Pb;r{pQy&g4m7u=8zd4t1cD!=bncC=eFDjmj<YB%Y_J
z;md1a^}|JG_I)MP*0O0z@u84!;jGWcN}julfEz%v?27YymFwMNcaSQFkaNnUz4(uQ
zNzVkYS7wWfX_)qFN?nw|RR{CZ*o^jMiVR$tye!MusrZHpGv*xLHu|q8vHd#WMC`<x
zPXd9x;tn-fJaM_yC_q*k=b{%0d(o(0v){zl;uWJ5ZeLSy-Vd|8bKK};W=-WP{?#<R
zK+Pn*_E4~5TpQCc<^t+ln$YU*{1BBi^*Ptr9uebjq9H$1Qd7CYZjKrEhEZ|Ko(aS5
zO6@yRiHwHts>S<5dv=b`jA1r*V+kYS3k&2s3snT^V;QZgs&SRQa^4g1BS@b>>2;2x
zjI<Tvnln9N(J*nh5p!v?B_svsmjM`f7w!Cc`fDv`XB@^LO^UoS3&AUflL9YCadZ^5
zOdcW&gf?9(EsS_APMm9ua_Mq3MzlrDF0FE*<9o_DalaWdu&Uc+%IG5%w}#jX;}~Fj
z%F^STbI%jAT8v^M)l7R;^yf-aF);9QZVoLCfhk>H!}iFxjvNv+6KoQ-154|zHZ-c_
zae{*!4WTMl$@94=2#S<hO~;h`E9ESaT8~Y-d(T$beL`V(SpS<F`perCN5&H3nqxrt
zP7s+f$ZL{I`yS&RHhe)^zpPQg(OO&oEb!V?sl4wQl*c(wNXrzo=plG%CJ;x0gh&-K
zW6H~N8%Ss9CdwL@2XatS6!v&2>8DpHsXt-6Qj6^ps7}k+d&O{L9NePdXXf{ByGkl5
zbCyKLMNo#8aRiTuI?fF;mz^kiifBSc<uYjhnOR|eANqx7k<r&BbO$I?Xgu-`WD1v|
zqnDu8<nLHi_(WRviN-7E(^MQ^=cTXXRG0lzl?tk`4WSt7_$WkFm()5EV(U3z+7(w{
z*MIx?$IaaLZCz>#f;tW3qk@`(#}FMoMMOFzH)UR$rKB>94J7YGt_M6FIQ(ACtDq~#
zvbQc`%lcx?n!u}fqt6Nqz)u@=Nip17C+lB7>#TjidmdMeE9)$IIfMMpdf`pj=8_Ed
z7@gs(6XxsV8<~PBkdyd_GW-D|JJq2S8a59hkZcNSBkbk&AK7%j>1I1sUjx}M!n?{7
z!BW4gh+{rWO!yKB&F->)iE38AOR09uFFu7qnmNhRq_e#+7?3o@9z)=(#??<))GuB1
zfREUHQq-$2&_Zbb@0?FvKCU%*%CxRG{I)V0B~DL~D#+PkFUNBLK$K1-XSbKZH`gSE
zAa4>ew{}l@j|b#QYnU4<!Y2O7M%J#|vN>GHUQLAB*so~5z6JO4m>56{l?OW#l}RWQ
z=tD}X<rj~zGMj@vd7^Czi=ss#VN%{<XbhIG`?MB5DRyDzQ_M53_MK2o*;Mro>(e6E
zCc8+nXLA1Fmf(Xgj7~64pa7A!9k3y)IsCY$NU-01<Meo(Iwa^OVqGFkJ0LF5!3o3$
zmaGT%D|k&>3-2)ULOO&4p8&cDm&@zRoog1R>|zSS?7(SrGAvndvJ$?zz8JE$`QxZz
z*`Cf~KjgUUt7zO6(5L$rcl<15_>A9iVV(v09Ka6){+<d#C|8!*3$~JPnHU>BN1TS+
z%5D*6S7-B|P4kURNZDkMEVf%E-)58}j|b;U)s!_S3|A*?l5tJGu;O91yy1;LYtT`O
z=R|l+kh}<A1n0r3g`|X+X!?N)+QyEPK+Ij&a0=<x{?YDd{LYT@M2Qcf&5@TjXJch`
z8Iz~XU#%HlcojaM`v)dspeQFxmFuEY8X1PBkna&I2Tiq|RkQ8eCW={&uf|PuR;TC6
z&{`$y!8Ee&!iZws(S_hQRGpQnS9UJ(FP>%+=n(WYZ#dbU;YKPtl#eRiNohB`IKR#L
z$i0L(IpJD=b%VLukU#wwW``aa9eJ}yGSw%rR=P;|?H%(JWCtx9{uye;rT2)?MLGmF
zuoRWgwH4)~nS}NNg){~){gesuPHcb;x95%P6F*JGPsM5&=+2ChPB;Uyd~xo8&)Uo;
zd=w%#N)F`Q&7tX4s);8b(u3|QOZM;hKRKa)Nsx=DfK^r95X`_DQFGG1!A$)X8y3Pa
zB0R0!(4xveW3}`K8w#x^JEgv?Kw=BSv3wrzq~dF%m_snRh*X}tu3nP=_G`Ov4*Es?
zrT6T#dL{bRj|3-jlW4{s1HHl9P%R%O{O3bp)-Os)+z)_Ves^G4{jur;W>uzi_itgM
zC177H8&^~0!7JnZYZY4C)k{|#d)&GZ3S7&XXjW}CjbXD-6A8MDv45%t2IAUDr<|pD
z-vPg=@Be_{0|4x0d4K-tpzOXI&dYHM^dXLFcqLz~1Yut9Vk0je&*$|W*1OIr<BB9e
zNI5#+=fJCG<ey-u2R>?-neWu3OLUMN_^G5B^YTPDG-VZP>!UQpuQKJpqMhASYvgvP
zn<yv8AGp5J+9-r@Ke<;DwmG-rWnzNM*c~P(B^Q)8_^%2Y<NDPJ#)@~~D75+2JHV)#
zHH9*=JOiVPRh(Nye{e8t(;|_IK~t?pg~o9bxB%Vox$EVyTM5lU&E%I&D$!Hz1&Bwh
zt$D+=I4#do6?*BB^%Qu!3`q#TpZsfSMRcF><7&HEYhRfpG}ba%vjS!`c#Xdg_e58C
zb%|1bEo_&Q7QxxM<`c`iaa`Cw$72{aTeOa)i7S4&9fXDz=%AT#3r=*9-Vyn<>^MH!
z=)G`pkiM5};q~vffKLy(P=+YJX7bnh=^Vc_pm6Y0qepD&woiHPyo}Ga+FeGISjJgd
z5tL<(W&z4<9wVAjT1S-K6`Zp0#bzwSoUH~yz2p^MkP~cD_>cafLmu7bdhQ%O>0tSD
z?u?h(5?XHbY5cK9R4e`uNnHSi4?+IT{B+qPX98nEq4Ws=06Ae^4L>5b*f9M*iJHKQ
zVp-cPHKowtwe%!?zCz!sZH`_>;h~bTm=O|$i|j4mSjKfgMf>scxQ%KEd0bT-e`YMc
zJi!ym%>&rlStwqLaLzHV1~i+DW4($mC@u;Yy7e4L8k;(*p^syMYo=_Whv>BHabNr?
zQ-Q07N7rO=3dhsr<U%0IMsUqd;<S}DlXhgtn1!^W2uplvfRWf_0=k)YjYSvnW+fFA
zW>le%Jk*LB{kzZMy+uYql?IHl4u&vao@5vX3gc9b|HDf0dHE>U*pcRGV0@PP-drhQ
zk&NVn<IybZfZg!vdXzA|78VkoaieWmcX;?W^rNZ!zo2<43tzQ>G{+~#mU*nua6SHo
z>J$dr)2fECG1+8kMxOHdIZ>X<OX+`xR6KRqnW0=@0mbV08o!*h<Q7ljgd&d-kdWg)
zWGT;cr~6_*ls~GqIt<K5uLWiaC-bBnlsW>u88VQKKp2s!dE-$~&$B0b$5lo28?n`t
zp<d4|6jEcUmqEKD`r`T0j;WCp1e83K4*n2jgXfH=&xGGOkN(vly7O6JuA=my-<`0l
z5$cqgif<)0eN=_}53qMW@JH5LR;N!qzvwC@l&xNT593X?cYl4286L(hVAe_U0jRaF
zw2bch4tRa5=Hz{?JHstYqOw*g<YEW6CVZRRcwVHK2IC0&;p2EV^s$(oi=8vrKHKD!
zc>=yF4BGLXu$w3&8-kAas)LB5E&2%T8Xbvj_TZlMPkVmzR8za50+FKqVJkC|e0fz?
zW)RlE7-dS%F_1Kt<L7tR>`F<M#-v<RwBSA^g77SodfSclQ^{9<>rQ{KOpQ=pvV8;n
zG(3<7x_eNGeXk@CF$-%#Oib-!!GT+upF6mMj`$Y6r#Q=XLKF(F9Zr|taW@-%H^36$
zJxV-)>A^m6z5s@4CY>vTrPe>w1xx;Di|M9~>|am3ugSp*#3Q{4(*0Z;o*g(%I!wdp
zE_#uiJnGGqMzilqGd5-bXnu;!QlpO25EuQ}+hwYxRx{C~RECIBkC-?A>I!(jQ!*zK
zo=I}cfc%Rd9ypWt{v?xfg~hX&iF%Pu2*CSb<r-pnv@>`00v-LBLyZz;=u9o*a*d$*
zVTkB8BM=*{4$?EkSNOai=l2loXw5i8s$(3*#f)r1rEtkcoLcb2*b94_$mu^gZ_l|w
zn~fOBP2D&SceY%OmZAvx%=FYwSJkVn<}jLMDeg5Su22zE+c%qn8MrlmjYWSwOkU{C
zVvm7qjJ}T{MXOR-)x`MXT7&)K2fY;aqPi-|>*~xj<qF<#d$Uz`nC6fz5DT-)AHQo%
z$x4ep7d7lJ69*dx6p0r}Mo(d}k22l=K=ghk%L^=ku+E)6)fi$KFxBFMT5+&bzeOst
z&R{RTR5DSS#E6oh$ziFOb4`tXu6)H|^@FR20>qRQH`%E2bv3q@P%4%J3C7>qf}Z86
zp8AZ2rUYhJUEAk*{672~qJ(gM>fBJ@mf{urn<C6%Ns-bWDGsn7WKE|)zYey*5rt?G
zefYLY8W4iki4$A(^qi>wT9gOZr12(i7N{pCBaG98GwxkuF1WU30q!*N4WMR(bB)H+
z#;{;fM-5&+sp2AaEIoqnA{3&Gq~fbTt!Lb0qcF>0IP@|F!;Nk!ZA+fkK5JZ(G&Qu4
z;2&1jV_%}Y$c>{7Ki8dh^un{Gv`rA5Uua@+oL4({XFTA_Bw?QvhU$fcm6n{#YA5YP
zso6kmtuYXwgh6iLs!bPURyux<B;Yy_SGLIdr%JqE4)(8|_#gsDdr>oYqc^|TOK1Lq
z;@1<^#;qhaPBhU@td{ASzN57Pyg!aYo%Os75L<7l&xd<|);Fo^Y%$$m?W~=h(QlbD
zTYk|$Rk6N}7(0E&P>|J?S9~HPVLxR*&ouVU$GOWuau%lxzk#ASa5j&kPN(sQK2E4x
zlv$G3QQCkoXKO#yHC*8A9hReWphJhOBnImKw1y!q*O_cWEzr2!jaZn=P#T}cb-$x`
z8dY5<$}~cr_y`K;770eYdcIx~G3dTw+tnLfVzRT;+)JS7sDsb|aFX#=@H*G9wD!|w
zN>T{C<{n}-9{{E-GE<}<7G{vtlKF}$1Kc@FD!C;-FUVDMwtWf)`i<08vW#Uo^==}w
zgK&x_wbO+Dj8iP-@wToD$&xeeze-j;Z}kl4`1Wn6<(Kvcr#@BvUdEDO@oX`mg(#F_
zYA1YM8mWm^Pet#b1H)vGBFU#x(6Tn6AHouYqbqi4^Hpm!zKy^aQYaNIqMlOh<Xpql
zW%36BXS1?Inqme>9BeI1p;YyoD0|>f$fPV`_#<YSaiw^(ZYoWOL8f2O=C7YZl9!dw
zdq0CjO0zL4M(wr{OXacllOTI^j;IhD*?Ck)sk;o*mE`j_`esRnJ(K|eAN;h>k~^LJ
z%s;=`@-*XH+dSb4|82z0AfEZDxqx4#cp#B8PjsD1BK2F}2`oGBf8E>(Y-a2a`Qa}R
zw(FO{1|Ie6ak(ZS5AXxLr4o%xY~$pBA&X$Cpa|vm;<idndhcKEkjktUQPGo$vJ$$T
zFN`~~#k135>7(oMoHewp7$q(GB}FSbR6~yP5OTFq$O=IOMle5*^LKaVDEm><WXl79
z$8jevWuD<%Rdf%Sdw<2Nm(2G2RsG|YIGtd*Hlqk{#O)pCcAT1%_~%3A{?9**ELRMd
zE?p{9ycV$>en9<`l7YTVD;>(^gLj2R&(uT+3nLD|K0jY}J2|eJz4Z0#su91Kx5h0J
z&!I`NhdyZxI>O4$B0{BEOLk4ZQT{D~6fOmhTd|VEZU-(f=dx#aXISck^Pbe-E&1y(
zNXLCeId$WcR0;74xKeh(*OzQHSqP=%f!IXU8<YbUD7|VXWXZf}5jA%o@i?xk^23qa
z{jch0%LHF@+kN@}DJ2-FSoif~XI}czRAub)Q=E!jqhWS;wLW*GIMKKMkvPTCd%a1>
zmL8LNJJ<c?`V*O5f5pobn6|I@gwN*9FfP*bD+M4Y5+?lk(RCWd(@%D43Y#U)TEbf7
z8qzk@(M2;C)ielll-i)CV~AN!g$!vv|L<CvL8F)N9h6itPFcOZNpjjT;fZ(Oku)LY
zf7OBdm|q$7l3ScA;1HmXCc1HeD<t0HsD%u|f~wZ8wa56pC`EG>JcP4>cQJe3HoM3V
z)yY=y*%@DZvrZ$hq`caOMMUY}&@Gg=HbZ19k^Z~=?qlRM-sY&je4VQK5-s-Jv|bi~
zzS|I4i-}2!V7&&uqZMn^u$|u|u{*cAkW-o+t`slRFFx|w3~Jmk9HNBtomoaCPeDVZ
zhA`)mu~Wz0Cav^~66TT5EuKh9!16g-jlXvTnNrSSTju=PL^;$^X~Zmp;j|DvTse3t
zrs-d+vTnT&tI|y7rx!+pjxrJ}Gl{>@S=eL>PP%2g+n-Do@qV%yMQheY7M~vJ+=3{`
zBn%M~_gs`B>klcAsaH3GHaCo4@S~ZgxPS>G0mdO#eQ?d^FMnkkZQx1O=lxB9eiGFQ
zcAeWItk<a3(OwiXc}c$@RW$gOZ6$@wT2jecIty&`Tf~uZvg+RZ_Y*5zR$YQW5$mRZ
z(u8TCU%b$elVQworq>1<(&j#!Sedzme5?C?y>;QF@BN;|QN<ld!4s_N_y@Y=V6dg`
zMxq|80qkvJ)oU=azM3LPzxrTT7naKJT|^zUGpMQmJ**RGzct82DF2A!X+p>6rufjR
z^`y#UZy~eK@VoAu)8UA|F?kEmv`pF`K}5lk&Hf1PGK}7B(z?)IINK=$=omJg!r17^
zA7B@R-xVccPUv#ND9p(T*2G>T`U2Fi&C;T>XmS${!e)pUg`BZ}5`RKA)VCc|+CC-Q
zZ2tapxXjHX9~i&!g`?7=z9Dnoec(sdQ01*K{YF25RW~D8nkbnY`Di>3TwB2KSpQvs
zF$a8R2|ZV#ps;D5_ShD<Ms(X+>?zsYipZJ=KNtwQ6wz_LmgkI75tzO6czZN#5(`eS
z;@|AJ|CZ+D0y6w(z&m6(h@piAl{K27`X(*V!S&*!hCWXVozv)DEPT^XZBo+)w$)zl
zpa+cNH@CJ+>5UB21*mY3)LTd;R<bH@g=PqzU|<^TxR+INZG_mw5hPz2wai>DvT4o-
zXmI;Ldq{f=mz&_MUFx4;ksUjH=BMO*b~J323j-R7&n0mBidyYg*1bCA2q=oCAW?ML
zar!vazP;=sRT{OMWbepQTwAG;T+uF>byk`c5eyGxn8L6-2-B4y{HUEL*R94q#1mjx
zK(iA<P8D%@%wKM3U105FQ_dC&4KaSP8Nxpb%`m!kbtB_LnSZex_)bMRFQ@MW*CNZX
za^xLy4CC;s`ul!YOz43uP%2*eE%!o&!?Yr*^fF++CsLV2nvRe!tdyzpn2PfCGE4YS
z6m1o2#i6k|&dY>#)3+mYZ4-woQ82|uUZHiRDkeAquiEK3H)Xx+_52jsv-iBGta$u_
z?*>Ey+JS6M`XjSB>`F~^wsnWfU+=AZ-=;NcyjI;V?7q|0rLhO_ngqdy2D};e*EXB-
z9x^uFEip4hOW5m_PMWgA@%c~8<frvFVYw}z5e?M3OF<%uOBS(h@%3QpK6xb0!)N8u
z&W7tY5`%%I_4^kp#n{<J_|dOgYZn6)AMXGr5gk>0OY+?P;&q?mNX=cwIOA=4YX4a@
z!e)z}q|>qx2-EC8jl%nePOHHxZ1nWF!Qge<1k93%{y7O1#U`>_8BA|m&$WEC=Jm)>
zNMQV#0J}Nsgme_1-tWnw%k#kD?h`Vc$urd68ROJmz(f!baXC<8&bmYsHT?dw!Ma2Z
zTXaW~DLv7rw!I+z<PXjD9*g0|^q}@QZ2M^(<4mG@$L9ORJgaLd90e;e$Kjk1|6$>?
zkb=clt7KV;ll-64t^E{U$%MRwr06}!Rx_>eJL0oGZxw2^fsl@a=kIhO@D`&;o^H8A
z#Os*QPFpup-_9?@5LEcN*E3d9p&ga*qsJgyLl1**syJ>FP&+Ew{dtxuoVn`A1v!E7
zA4Hete_L%Mu!7e;;4rDbyjc$CQr9qqy+lHlnr*pCJASQ32}<3}WH3sUT!gP(`Os5w
zh;ZGU5(#Gi8LQbBu4W08Hv7E2{Dd$V>u@uXuZFEQRmY2u2m<szq`d65eg1s9d+lWL
zs>@fdEYJXE)@*4;*iuwG_Vr`K^~Y>eADm=A3d`4jxu5>3s581Wqszt<Z;<_Byv<VG
zAFiu>jWg&tJBIvF3W^s3Ag=I2M20>&SM&i!=&koG&vKP#{tg}LQbhDbe2%uvoq5{q
zP?^Kz2f&@oT=K`}He8XWqp$}o($R+lLOsxz<KvE!tcMs5QN^(zRf*eiaH8PkJtT<#
z-9NLyfSg>|E+##Gnj*A9sI=xd2l21DEyXdM`$(32nON*GPmro@N4y>iY2abW!?C9y
zR5^`(>Ux?`<oQ-h#0`0mmqQM|)eYy*I!nYM&lo|du|;k(fT>mJ00&vQ=oW7XS?_85
zdEk3ovPKe^NKkc_zQ1-9-mXjgG5_Aw4rpo*Oyk9%Bdf6Pjn**_ev{!xav6}_>7Xi%
zLEgGMYt*P(jjj1PlRkLu5iFGLg9-C8wp7&aaN307S^T(3ad$f{+4e{-!#03ceUt!0
z@UR<}Mz&X|PZ9nTRY~j1lXrScc53=m|IWNI!||bfdMsB=a%I_%*do~R$<@A$N2SnK
zAvm`r;<74^Nsj3CiTfcuiZU(%NeC+iTP=d?l$`-tE)gn=x+1dXox5wA>p%rBhOeg+
z!Ux_4*$Yw3WAjTRU+MSaSB|X|hKrMFuDd;TvGaghI8%G!H|-JfXv#7E=lLI1f7_1r
z5RiXwP}(TS+DtH3=&w#%s$jReP0hM>#3?<fI-HrEIxP~TvZORLi>J=7V$)t8H^F3A
zzBW{sXi(&g0k_ud>=jD*u6V6H)b;3FA4pt3Vgnmz#cwQKMjb2=_a&d`detajvuOgW
za*%;>Rn|uCpb66~6`Lc~reQ_S-RgqE?z}O|0^<di`HszhX_^QLRN0#^BD`K%+VAqd
z>Cf%@c9$l8FuGsWOnV$ud+UQ6ZBIlHt{+Y&#yz=}mI|@@{#llhdAO`Fjt0qg_wm8w
zbe|y{XusO7$x<i8Z?<OrnhkRP%l_LBJ;B~TdZ(x73wSTHJ{(?d2ZXH(vroS(jmIP#
ziZSE7q?8F)@}l^Ol&N@hOkPz|IPmx7aA2^oymW9WjtA>jkzBL5B^~xNX%s~vx%LZ1
z2i3$xp8_1Mt&*D{_Aib6vW;Pp{&2-j#ndbs2=kP6)yFkqpY7X_UjV|!$d0pmwc@o#
z4W(}8c26uZrid%|;ZQfYzG0?=WxL4N)59xvs!qQA_p_hWu;*>x?ZGxr4U;HLLqlYb
z;1;*6gRL?W8vwpb71748B4hE{hfbaDh1``9!XqvUI{N6TJ^UnPYW@35$3+XlXzzY>
z{1-?+;w^SF<**;rBzy#W*UMFC*>$zm+$lvbr-vsJ?7gC1yfWBY!a65ff1qt*MaqsI
zh#HN#ph-F4qCnSqqd_OkPnJq34S9&9Pp0BG)^XPho6)4CV#3Z}Uf9+|nzjI6+h2!I
zfKH}JHQ7B<Z|2@(Imb|k?sOi<i*ngk3Ms=T$*E=g5*BiZifp>eto2Ra9S7m@Q)&fE
zVb`sv8lRNMQD(&t6D2(D`j@XvP+3BiWFCp{`UrtYFUaxo!ZXU<e>_r`0r(7$+*C7m
zTTu;kb&I)if3iZgOi|(&{@%KIkz2fx-KAJ_`P%1A+qCASnVy#;b~X+yLD*T<QsF~k
z9@pdDbVWUu=I3S<r|h?F5rj9ivdpscMS_lhtmH=Uerhdu)K_thmslYdu$YFk&;oDb
zLiCVTjn-*93U0dpVT9Ir26h{~Ry~9Ut=iA#bIN&ZJLZyjRj4*9E_zDBc?`+xm;9M@
zl7xds-VkHAr|cWOLI=JgksTh8ir+{ssxPmyn+TWHaW<&T6=g}Z=YN2{ngjgC6Q3%d
z(HWiz2?A0`-w^_d?MW(26CYC;KwEn;yixHTkC;EK5+o;Oy01)sHO-J|B8W&dqqVOa
z%r#2f<#4dQGg#~7Jy015PA5V$pPULqD`BDAJF;oHiz_KB>(h_q>b>mJQ$j5`UQQd^
zvi<v}l6;|_`SAjiNA3Nw(lOgB{<Mhto$-&s=qYuDf}Nd#GuhKS?@b!W-hm;&gyRjb
zvqjYqtLR*}->OHzCpo6}H4l?76MNjIGs5Uc!#=8vKVSbm(Q5duZ5P4M!W??Gp&-AB
ziZBi2zL8D0s?FxGvNO)FYcXkJArSAR8zB@>Qtd@(3v^E>*#3>*;u347D-v&3Wnd#g
zpTh?RE)edqb-b}A>_TI>hq6<%x10LJN@EAjIecjUy?@<7c3)55Y(3$*h<{sE&)l@B
z^<quW{OzwX{lWOmqjz*=+*|K?g=ZZswBPIUZLx3IN~aks<-$cUEfTKfN?)b(zSO;%
z^g{ZZ@&&(e4@<?ivsr4cpDr&&&_xEk^)rQh$4hMpT9e6N?h@4Shubx@<OH%D$Zl#A
z6J^}wOm=WwLSE1So7=R57@PH<v$HZrq_tCCv1k&AjO1M$=-TS2pyf31YT(X18H43^
zHeRG<gwvFzP}ZKmoRa~s`HG}`d2tG61l~thnWY}OP}y!3c07Qq!29F)j~u%Zah#8P
zhS(W3C&G(5u3@<D=$iC5p8Bdm_=}D21YRC{nZKYmWaZ6?bbQWizkNH35z;Y+nk9C&
zXu~wD1Zr3q7?;{gUeh%V!79pe7R0v-p(<47)#=HEt|!edlkdOGw_4F?{L##KVdY(h
z&n5SG_=^>No1|SWS0y6xyOQEOWKXF+Doh;J4tR8q{1_o&_~=<*$$8NB1<GEUVY95Z
zRgZ>C4E$yLE-}rCRPE1t+VY3VB(}HjY6_i%wJY>vASV;*g&|L4s-9x;>30w}`c9@o
zHeShiF>n=3>Cv8#?VH9p)wBMvc(Nfa3rcO1lx6b#4|jh>A`n|Xh09Juju^|ypxSZ8
zbyM$P`RJnsV|cYP=|pegm!#y=Q<N0xWwGxogi=C@<u8CHJ2%|TQjohVg5NEIm35II
z$w3MVvF}#OqAIVTDmxv>;gX%y!IgR`QKFyR7g1}aHb6zT%%v5PC2VgPtx}yQ|4gc2
z1bM0i_P7$9t|<pfvq(H{p!<xI=u$qm-p|h9+L%{NR5AX7-hZ~i_`o$m)0B}l11|lW
zLeuWiaDkC3Z|xp<v*XHvp>4qI<2*iI=~g51{ZrngX@jW#&|U*U2?nZOUpWs>lag{R
zma0^US$}RN+KMz3r>-*F0OS(YrVpymg4a8>SbcpVJ&nY9+b2H(<0_+3U?-8%Z+R}Z
z&o?ym0#g@2n+1b8<tz6K|AG*T?8B09&N-$^D4R_Y6x7XehmVyhq;KRFm<b5bmS(hu
z+>_qhu=s%AwBJ7pGQ5bG%fS*j%+>9cd~5LDxZ-3`#pyt;W4;(m>Z75fG25J3@@j|e
zQOxiI!%mTF*nG8tGm7VNk`hFVVm=7%*J*onVZCRLWR3(O<|oXvL+7bP+7~ss`_f&}
z969bYz@h{XVTmkwfO6`aM&Y#YD3C<Gt1#PWys=QGonRzce;@T8ZL|(9C1T=*=7f6n
z+;YqVi*v*CyPDb0y6=FeAsI`Qp=Z`<imE|^fAT7fN^+zzY=8EmUjI7GRT~rk;GrG1
z5v0E*56;iY<8GjK>G3W5=F_=G_cK*SnaH|3bny0=ur)N6o$HIO{%~5_I5S~<iR2_5
z4aJ0Mq7rG)Qr&<QmiXclPl4Smj%csDU=9x+VZtE}^D|0Cq!v-;8|Pn#Tx3jU&59oy
z#i-(7gb7RVXEMcQDw9R&X1}va=0o0RB`De0?yr!vkg^xdX;ydZ;@I8$7U{capUdjV
zbLAU@DmlrBr@s%X<bWfIpFEyVH;n$2K~`&jy4mkmF-t&+w$-}K#h-?L@BijNs7-$~
zPh6QS)KEKy;_v!i>85CLKL10o;*IFMZ&UQ``KI{00T^6@T6Z~5<I%OenRrp&d{waJ
zYi$({$vNlR;P;hZbWY4FRA#YH6g2AedTefSf{xwIeCIpra`cSIA!Ckk#{>32{keiC
zT1$z``LZUB%6$Zv^SW~6fgEjfNGl`$HBPWV;bU^-q<R0V=%Q0uE;5i0_T*mCrY4^c
zxrAQz*wbgU5k8mU^N7F{{MQ>{qI93!=K`F|1s$w=xdvXuzbl7n5?M3Jbgu<GubI0L
zUFIHUCza<CN?2c#CJo3+$%*(Z;h4?2|MJOE@;hH6GGIsXWJz$IQCfOqCvhxC{!e{_
z91-0ZOQfcQk$^T{W0bQMlg7WlQrTKwQbb+{CbS2G@Dd|sbi5dbqkTUr=ZAY|f5xN0
znt`W<O~N66owN(gjk_|JY;@m^DitpJJ^BLrp4oIaiFEuv*Jh(IRM#^VkubTvn9u7Z
z+Gb8@w|^6J^G0Slxzw^pmT9q#{!)^}LW3td7<-6_!KZKdTN-ORox<W0_7m|SDk|AA
zi5*##&-A-yp6B+ui-xC*<Im<{MeH*N{~9Ee++JZpgySdr26GwIIDibZRubcUY{n#R
zW6Nx6J9(8q@`3N)kUi!jJfSD0JNdWQJn%+8BsF#KH&fm2H{@W|998yqbL@^rGo__l
z1p5s;K`l@Ew*Dt`gUfXSS5qQ5fV%U8er}bWHkXR%ShYSvZkh=9^~Ek;9iZS{bTQip
z9#B#EBLQRd?O=Hc$+ozFaTyh9p=vDIJ<|Q{LX(~gHDOxU*bA_~lS=pV5AP~<%s#yR
zl%q0-muk2AnE^@Ip}V@gU#zn7C^|k%z(xEO)&RxEue+ra7P@Kdw69eO=F`-fw<bav
z1UY<oQ^;3Tidj;z&x!|9k7z@Fl~lI<cD*=v6401b@VhN8@?ruF)6m4)7o_Z+W9E;-
z)$EQq`2+H4g+K8&*m?OD%h84Z+IK<n?>Ay`*PYIVC_f(dh_NLW$gHLmU2RRN^Yrw3
zC?9_~-IF;cQC4Gno3ct3x--yp(~Vf2ZZ_~XH!G?wucp`r66EvCtU8WY5U);vogaF3
z04@`-PZB|)P1$PTcztmbiDwSaEb{rOyJ|@CE&u(<Jf}&C|AS7eKMbh;Jo?JmZJ$hS
zsO2Nxv##f0RyFT@d7*5{FmRFTMlPT+<a96RcT*GHdSWZ=!=j&beeKKmzMoUVXR<mB
zK2XsaBO8nU8rt>N65hn(WBS(6{{QsJJdhil@3>UM{v}luOJz<k22r%SUMvrzzm1pv
zCI4%F5#j;i;ywBW&nuZyZ5!|U5@x4WB~pc%BzW9VQDEFWHfVfyC`n2Geqv$b^>R90
zbh`+_%=gVv>fj|xOD{GMNzhe-I~z9dBJ>6BUYN`=dG(gbpXgR52L}p-{<rTp*8hU^
zQsOLolx(LD;Q~&R<APO_6g=1Nj8uus`&5fT()_746b^G6Xj6%UE{O3G-`1iOXh&8!
z|Bb{5Z1st5DMwZ!qrTzXyJ@Ra+=aEnJwv|IytAr)FD!1HQGaFL*+rc%5&tIb;bk-X
zKQ+=cUS<%1=N!1bw?sZ#`X7FEio@p*caC~5{yHr~xnVQA73yY6X4`yF%3zb+0mhT1
z7s74M`$rDZ^T^{{q-sc;;`MI$VbwJM#`F!b=LehNun$(Ae3?KqtXj;y&a+oZp}w#J
zT&nj6X6*~TT72fNUFMce5>Aaf0xOLvo1DwiAVo)&Ih8ttK6M`YNvYzKC=)s&11nur
zwvDz?!_9!Vf6^c1BT*vxkF>!DH0CGVgMbeuI1qlUcpCbt_-~))UFu7)8ec=2^KCml
z_hs@2XvMbmzZ|t^+dVOz5nFrXpyWsCSd<K@i6InDD;f|Oc%3RNoQk@}yY}fDNOw*9
zabX{i(M-4w?%7)Z`4w<stTcsKl}lVX0r{jfPC=VFIm2}ZxOPD|GSzjV{9t|eAC?JU
zTCT_PH$Zz*-r2W<-F`n<7mZFA>6mWbYPjIl`+*zXhr7w8e*v}`Mut#OHX57B^8p`p
zT5kIlF)wk!9!Iyxj7cOpzPgCI2xUQ(&=*iwHFbsa<J5QV;(w0dHN?B*b0!r;X4gdK
zTtu*;*^>RN$Fmf9autMk)6_CC`0M(;f_L5UuW31gT_~~Vw<%J@ZSq2?sAK>B9X_v-
zm+AC)Y-1&G`rm?0?r~+GM;cvRDw{2Mo@w#Qe6ZS<vV6qV@{!P3J-u-SkcRB>dW1{*
zIt`<YDbf^ib#IEuwHy}hZB<nqm|0dvD$N(qyZ)tl>tO<H;^jg0*;Zi*;?LP+)!+x~
zlXqK5hMDOo)wjIeQ=Ab|fsg~wX2mI^*r%W3y&<O&=B1sBC}kX{8+3wF`D*(`n_-fC
zhMK%3G`|K~SOYs%l#*wokkArf(9+<3a5-Uykuf<&(Ku%<UyT^Y(fa!fx410Wx!dsR
z;moE=hZ>;Ab9346R`_ye7*YzZpzP=$b=}rA5Ig1|V-bf$h@=Bu(o%7sf2sPwFHZJE
zv)DpugWlsNW}foqTgca|4e<JTByEJD%cI}7bI@_hq#`nbF`hJVC~?;&kHIXL79Ejl
zwWqct7J)GjKL5lR-!|^Z)gv#}qp%gJe=6-;Yzeozg8nnzi$X7Z`NY;;bqypwfu_Hy
zjX4aX$`-!JW_ncbJ>+Ao>!qjB65V|`Ech39yypw7l*m8pbsDz19}g~0Txz!F=`C@b
z(*EXsj6!DS;$n|`Koj4x`qe*sp8tog$9YsZxu!v}-K<wU`OscZ=`w~N$B+F~R%c=_
zbDrAbJ!u9+zSIv~OQ2P3QmE7v_38^sAZG_jHQa%()`9#>bfM*CBK`id=9vVpKJCKT
zgzOvB$7!k-0IW(oU8Fz{-N)|2ue!xJ{|MNrYy^nt3n%U`nufJr<7O+uomN6u8XR#J
zXX@XS3dZzTks^KHbVk_1>yMiq(;`tX#is<b^(JNuu8lZPCnhL#9y=2($k~g}9!&+G
zOdFp}N{uT<IHulazVn}Fk;eV~gzU)?B$Qx7PJ<NMEJGebkzwSg!ae!)#(tqNw6-HK
z(QwzA32`(EK4?iy3a^x(we~vv`{|9@ZgtQ`Q~77An@P7y`tQmKan!!fswHuKE|E>u
z=SNopGmOZfmQm~KSWS70RQybn@Q~iDr-_Xh3PErA+`2EU6`g_J#%drgp4Qy7A!UnN
zqD_TY4{q#qT(IX1%Y=HvoNr=%+TZc+X&;G`Sqm>9-SyX+M@UX@`RknGhyH*K{|Be#
zE{f-x{USLnN==-MG!C#gFN_vEW8jM2kF#5mnZ~(o-8yUia$lb1Rz53jZVUg6+!Q6q
zUef0WVXs<|K5HJrrxW#-j(rhj&r2G<L$i=*#wR*&_ztP$>kwCiUn35ucNyFM;~|Pd
z1EczIg-8{p?dT`TM|SJQM2zb5l3p~_4$-{c;RA4?g}99Ne9%RUDTKb_o7|N=2A%p@
zbqdynqZ5mkmdPEi(~lXlJ}dN@eWtvcuYc%X@t%ySA<qp}5>v|C+EIrcp+QT;GE9q<
z)9TJ=XYEaP44F$GL0ytRUOqub7U-$Q#3d>T*$N>8XPkB#XFX~zSowvG-()UX8u_3q
zhMol-*Kt1ojB7mSwkg+SZtGBep|0^~+)4G{?I8EZee~!?gN8EQo3;AiGQk_R$+>@z
zXM~@T%a6Mv#!jCHu^6f5mI((K73qnu%CvwS2Rh$W%7v5|zKVJ~Ps{S67f}T?A2oD~
zx>SX&Z)^xIKDPgpz~Qc$p-iF<cwG(<r?>k{LAiSrWY`b0Hw!cI=~+ie6?uJrB(^u|
zC<xK7A~G|b`SvLoo+tju2=;UpLC{$o=G6Em#_aNo_I|n-O(=$ykwp*G{^@7*xh5-R
zbCHe>5E$#IB~Xnpa5JbIYR|L0*!j({2yhhVV9gOdx;3`p&wh6JzPm^!JM=Rv{PelF
z+Sl+>njgfqaP!#l>VfTtz}>X;&rbW-YW;(PxXrMY)4<O3Dj`r#_}>!uU%mSZ(nw>@
z65hqD|L|@@-3B@7#OeKC*T;nX=61OoK94_PW_IR89t?#0t;hNg-ud^hyrE}GWwc1{
z!(I%ZtOO@y4QX%y>6N67ApswFjnF&rm;VRtKoY;dHpwe`mua2=tBtb`gUDnlU7r(m
zh_L@s*;<leu<%tj@c;|~29P88_d2}Rfm5~}9>9zsmYlTyScen6Tl_lc=Nw?~mDuJP
zfLA}@JO}3BjO^vFLzA|wh8(U8l<%-cNeY#yX<D1SgilpX(BV`m;!;EQjv*9xJx>OO
z5ss!Cw!)+Zv`8`A;hnP+!kvOSbIkdH1N4VCF-Qlb3&(x7>eE!?wf9C}rbfE~G4{VW
zrZ5XO@Qc}Gt3pUcU)2B#uh&nGz_qcV%yvIlQ^dysuSV=jEY=!{<zZCKO|y=z-OB=k
z2Yufadl=@&f$2}E0o!e(>03B3@UkvZz#<`y29au=%Idbm8eKv|4#3YcWviY*eFg$I
zF9SLr!Yk-4K_B~a+rGiJ4mk$$T#Uep2#z>?O_-~-QppATHTA5ne7-XJ$SbQ45muu@
z7&ax5qSeD<^1@hQ5Lr_jYc>T+Taasl8|wEvRzARc6q+iQjZ9vBzfbh@rBTS2YI#+W
z*ZGCgTZxbcPCpz%O+i02Qo?z!#OAd6A-F0YRt=>`DfQ}NB{Cn*N&>mp*FE^vpk{(0
zraU7TGLftfHl?-xs)T;e^#>`buTsTSF4=oG!I#I5y)~R41IKR%88}r&lhYeG$3Pq7
zYL;lZWH#&Aw4*KgdeE(Ku`8wzj5(od@F0f${{Dd`iM}bUC$t%KSJ-a&^7VnYuW#kd
zNE>enW(-~p)$^zko^wDoanFu@XLhBj&=C5WObSxEFnd~uH<M+Gn3}x?(Pu9E3UC7#
zR=OIK5aRGWKq#z}HMn60a#Z#)0Sz)7RI6`chR}kgA^<U13$!eHqH-exOBlrJ>iK*!
z!vGHvN!tG&$H4QTx;7^>J!{1YFPo3QaI36fDYcAE>3j(!!Wu&63qGkVrIlq!B=OKA
zCf?5z#~?Rbn)3Zy!=~&PwZiqcrr5Vum8nB*!|5HN1K3_1|9;zGoyzT)0%KPVwD%YE
z-l}4-n&`*1Fk{xd4u^n&PE{|+V!?q@77Vtq@TveKU)e9#friyhsN=7)z6DdQ-V>xm
zQwiexNe?3)fD<Vbkq6oaeEE8VOwL!d$;M?66QR#e&40)ch;6vv->~m@^zAEJ=Zu39
z6EdHyZW8soDhp>P?HbPspejGk36}&W0{ebL-#EaCQ@`+nbs&Ut%OYem1G>RBTB@u;
zIqM*kq$PwThe@e4_J$UTAz2|p*cY2#W_g@M1AGcT5!`cUHFfE_z<diS7qLSeFP71a
zst_FK9Hi3*E5@0xp+%lWdI|O*mPPrTQEGKi)t?2GM@=bI-fuauT*2Ui=dcpgIy3wg
zbXMXj#b7FGV`*0c2)VwNjKHOaF4(%@)ftY{C-q#WD5bzZdHvxbJHpP)b<pjb;mhNW
z`@Ug;?977rbtr85920p8rti>M<$Z7EL1ftbhGi1xu_$hf6=FrH%fbgsFOiP68@9JQ
zwzmh`W{@Vhzuj=ZKk7af>pVem%vuZs&y!wCX5`b~bfZ^IWd-DwN<n7)I25t5hq7L%
zHAVO;(}VXN<QFfx-NFSu0;!bruSV7yKRJDJbm=5sYa+Cvw$=FLqC0a!AZVRs>Ed~_
z@xB%b2htzsNv%U2e5Og-oMeg2PMpVJAja~5so_MWN>w34M^RjSl<q<)rixq#w33bF
zf!%=&hv|+8X1MRifnSdU-_HZ7gkm;fR?eB+FI`85L+U_BH%vdUw+(N4hwOndUjf8Q
zlYunEOpGo#pB(VBSD&i)4zcpkFLh^D;9D%5i<)y)M8!yI<q0IOO-uu{7nTlrZTQba
z*>!!REI=U(ig+=2#yQbCdpVkl5Ql1@H-SDj$ks5PKm^}Iv&LRd^^Eqmv3fHOOWOAb
z9*=Lh-`<LL+TbxU#s^j1{yI<>SKpM?^NdN_q|&SE?e>L@>W|;>zJC-f)yU!28pPDs
z5E_n5b)RHqZ4G_zH6OcZlv$?RIhna;l9{GVr&(N+`i<4+a|Y2xKVFFz=Xf*GJ13#!
znSoZeBTW?>IN+(nP^J=W20Q0Ok2z@79;|#q%8c%_m~7VUi)z|37P1;AU{VB)Ae5S#
zP+$EKTuBAjHLLy=6jp_q=@|7qu6nyF7t%UWHXF4>SG9@Px#A_TToq|VTy`!%FAD2F
zb5Lm2IX4Kl&dv6=Z`e#h=vAv3H6Rhi>PQ^tq>b`sc+GhwM%rs`>kJxSjmaLt_xBIH
zzrW-8d{Ww8O~wpVH$BizaDQ}s`I~R}^6eW(UoO*dCw8i0h$!aGtXZpBxrQ%UI{zc-
zUs8BK8`vgR4JmN?m4KzXtd{<IOI6~zbEb414Ert+BV=St$d-)u91F^-0+H}@ezKbw
zAl<DB{0PL!$y01{k=EFnVB3`QW!L#5z%?#)OgxW47gCVs$YC&?_29!xE}sRda6@^y
zu3iGCWdEo7;j9FV-ZvQmFJ$E$Ck}V)3dEQkXe1OY9_PR@Ps~hgofZW5+m5ZXPXaA}
z7R}1AL32M<Xb@hp{f4$XW)RYARlP}dq&XhcFdw|X``eD&owgp@mGr#YprRCC5dy4i
zS=PbyYP@PGYEb~;=PR>d!*s-ja|6I?rfNN}weYIZnhX%0j_F4o939m$hlybC3RfVc
zLtvOSGrPBq>!dU=rj0DK%D~Mw+-_g6-)?M`1+Z@q^!^<-o;Yz__qNhismh-4i=b~i
zZu=M9wl5G-9Mf^Tz185FcLXLg_p&-sjw^#gacV(FD(JnKhD{+>&m=+i+>-GN>Hzi2
zj$H9#>X3V5m65M%<`caZD0})K!9X|%Tp_q8*@rV4(T!5-&19vxD5z0Az6AH~@OdQ(
z>D4b0E79iF2IS9*s8P03R{H|CM9)g>fJ-2V#03Yh^&>!0S&NG5)VgTYqiGBhQP5`w
z&nnMyFs_7iAr`Ydn%4BWm9Wg;St|UopkR8@-L{&y+_#3dHMGvrf8$A`EmP_i$RS5H
zrr7jst>J#Z)rRJ?KgYoEf$z@`JfBZ||Lr@T&jYgz!B%ER34PHF_lEs;<M1Elyk5<s
z%GCpCt+2uA@Y6Zhby96hBd$bixcV8QKumdougg+FSIJzb;->>)UZo~+|4U&}cxo)X
zbtR$oeGB`e8gwFRz|1)YX3dN+Yv7O+aN%Vw>}#Cg>!8Y_Wkyvx?p&;63_PD_`L8It
zHDcz~7t+j2_(n)&;5bf<lN-NTqoB`7c@-s;rM3BEDxL{E=fL;lgp7#~HGFRe`Pph^
z3O1l3xb=qZ&H+jH$A<g8LpuNsSbKdIs8*wzN?!=D*&A+;jiVW*1ZM@>`m*aNricg!
z2~i8spluC)0n5#Wn_sYKjn2GmsY&}Jj?7(6#z$QLyoR)-;DSY81LuqOEX9eJMMb4W
zJCwnD`bqd_H5jWF>uoDr_zlrJjwii<mbC~b;Ftwh=B4UI!Br9zhfNXmzGK^O(4)~l
zp#-G5AVE>UCYo;8_dD+U7j(UW^TTQ7a@%%1?hkx#zrv3}c9Xtz4r&a#EQ9kB>3!DC
zT>Y2z%z;a%VD-Ec)C#SHh<$dh&x`!Msxsu2v1@zXPsBt*ah`!t8f$8a;LPl*(jZJa
z;9lZXitvV0S40DWaz`#aNUG8lEI7)qJ_2ffGtFnHQr8QWs|v@YbPZ*ySTJHp{rob|
z^Xm+EvhNlrB@@6>hzmXsVZy3zFNA`XRYoG`U!hXytuYXkhd!e}Gf1d1({<V|$C9WA
zArE9$zlbJKY#^US*963X5tHK<$Ha8z3r%QRrpmHf5HL+_eJLDi9j#;V*Q`wJ^CB$A
z!0Ex!iDl1VP0*xc+YS4^Uq)k^R1KOK!WfL`cA90QOX@49B>|CC@{%gIgiHx*22s*Z
zS7pa6l+k`I9DpV5qK+1lSev6Vk6Bs;SqJ`v)o`F?ThV(fgFkR_p)khj80Vndrg4yZ
zmiP%7HOra7Z-Sd*&Y&J8IkhZW!w#y-B^**JNmGS~<NbW%`}4r_c|xY+yB+l2p8@F&
zS%XsU_l-J%d&9nOc-%VrtzaEn=m7Nw7pI^5iO;hMx*BZ<YO`I9KM%MY*PT%1+8VVH
zOO$qHr=Z<JEF~!F-Dq`8(oqTsWa?*B>B>+Sr4tfGCR!MHUa4e)mIYZ%Hwc^1>taRs
zL^!a<mKx7>>VMP_Njv#%Gh`~xV_qhA0Oy*dkm(aRaGXz^=Yf`h_N^Mgd99VtL#@ne
zuClSWlZj$MZ4l{5aF%%M9ldX??3EfDPaeuxJ+Df3#6a;}F^N{yw4VF2SC<5<z&iN(
zn&s2iR~3G>k%36Mpkg^1CkY*5)G|rDhHXhov5)xu68!$C7f4-7X#*rnvjeJ{9Tz0T
zSxB6pu>gW70UT;An0)F*RaHr|K~!>QwT5WS=gBoD*AbcPqY^~`AS()ZViseK1ppIm
zCL!ya>e&Nj3$E3taI|ur_-wxTpS%9xIL?lJ-^s=?x)1i9Ru9KHF`fr}k|~l3x>;U}
z#-ObcBTf!r&($c&D3%KSbB^bU=f_D!@9+zsHKAJZzBNJ)qG;O<_czX@Wv}3*=~^Qb
zgrgfL{p@KAf3YDXtHd?VQ6LvtN+FCA*%(!<t>~<4LCo?(FU>#<9`U_X2jRj+V})1K
zh^(^gD)-{kw7G==t|o0(15JwIK+ZEa5LV9vQjIqr4h%|hZ92AfifK_LC(Yq1ldKF(
ztt@?OQ%aUeoAUdw5T&(Q*Zl{pnJ7GyuC}xWXu^9v+=(mSH*B{%w#QB;&7GCzzL6m$
zO)5Rqq7N5TS=5U(8i|QHU<nSm;-*;H|C-s$=I7GFC{K)PsG_U^JQM4$g~6&?9S)=-
z(lh82<SBCX%3fnK830J7T=*@6i&axp3;T_#8Z@crh@zK{Nr=roVHRZJl+A$}Arr!P
zw0MWGe<EP8IfJmr@Cg|ct#|C(hAv;LS1>Wgf%AM~o`aSqlMb%{u10Yag%<{%#lT5T
zND~S_YfW%%5WYyDDo1LD!Z{a&RV0eN<H&L(<>AC+hrj1_ErjJ$n;we|BwzvqoOpSS
z2}t**Cz!2!aUoUL!FN@zq&$EV?yN`%<8u1iS`QHcAEUh4)|5LdHSsc;<30mWnK-b-
zfXE6)zy(zY7AmmTEHG;f><S<xV)0m)6f}AHPH}j6UVV%}T<cx+6Es!Va$9Ah$HZDv
zFf{Rdu0IAJ^@i=Xlg-iTNz%855ToehbmR!S{!Oxbd~#7qusmKvLWo}$X8RQ6X5`}r
z9SY}j;5ZH(=ZRUEe>cUgb$q+s@z`$UWA_eNhXZ&H!XAmhj)FX#7R2tQ#8vhQMEJib
zQ?*(jSe6^~Jra0md8!MnFkcbHjB(M)!poCiTd|$m)ktjRHLun1Q(<Ne0OM1}w%kj-
z2D&GJ;gcdYYY;WwcO{`L2?ezO)oO$X`P!Lq9~DS=xJaG3S4Ld_&Sz0oOmezb^p40t
z>xOW-o<*YA7d5<G7qzqO0hXmlc;fq@mSD??8ysJ5hI?Z#s+$50XbT!@0_}X}K)I8q
z6qd|`FGQf`5Hr|T2Uos?S3uoQJsavuI%7b6YEMT>&7rP)ez8GTQ?ROqszRy{TMfJV
z+KkKJ2A>YV7ss5zni<<m(;zFe4k{)!aRnF80iT>5&PvI#Z5y_hKsZD;xadWLH0z+<
z$AQxiBpm1YB-GPcg>@Cod19Ue@6X@x{@X8@=YjKJ)yTEHDr<Tm@$&{wKXLez@e~a+
z2WAW;l44F%X0YXX)auX<Q1He7wp|CQmp#cU)Gl~PUZOwusGbiO<N}yjG|=NLiodU+
zMd1ucOj4thzstZRB?_{;6wlS)@w$(--$p@c9R5QzqT=PJ{qjnLQ2S>k2`g}AmLsdR
z>Ma#)SV)CbrJBgg7K3Zbuf>+hs%aq+%X&w^brv2~>BkkIOkdFE3KFJT49>=2J*xlk
zd(Qda^;LECwqa|`?)tV}8$kftXivI#p67{lQqE=uEcN<DYtt;;o6dw2Xe_Zp$||?Y
zE#i8_>-J{&)7u+vw~nnf&Yw-e$H##b6#U}VE2zURD%&yveA#6wS{?3DBDOgLt?{Kb
z;{#13olc`L%~Bv%x`YV&^d~5|x+YK6q5w+xg+<Tgsr3;rj&xiZD^KYZszk?d4lokO
zCUk{u-JrcgTDz=1qL@?^)Vx>+D@)*f2SrR~-waNq>S{rHQilQY?_LLDB7MR&E_;a8
zoNRPBWH7Ddn;tk0tLC>#nTwr?FrdvH!xOzMfm^oP2>kuMEpx(h)P<Era{1j<ZN0E%
zrd_K6T>X?}=BCm>?@soK)`mwxK{H=Hs)Y$%f<0kXXsbb2**u{suhm1$faii%CGZLe
zS2cwt7F=)>lb%)VOJ(H(aKB7q-?IYxKFFj(8WiJdGKRCEo2h7Gn6(QPG`913;yAwJ
zw;-%B#=!f>cYJ*RjWenXOTX~_VqPWmdJd5jr$3>d$U||=19Of`3BKMNR==(ZnS;P(
z2#NaqQiqn$m)RScGOuZcQ9Tj$EN0uw2MD!3`JRcdM>#yLLtCL}OYfrwS;Ms=A>mkX
zE@ZL8mOoKk0e|&ansJ(9x}aDBR}b(KbCxJK3Z^P73zxE(tCec#T3^?-U}V%BaKx)}
z3ho;ZfwZcEQPoqS94RXqVPMgq-`aY%OQ<OYn;~Y=aL?a$&i}ziGe;!0tH}tAqF;`4
z;QTlc$BFYfcv#efTtXp2AGjPF$D9R00K}V6l-A_GW*zD(=zGJLeZ#k}ckI2P?Hl_3
zGgR-SgTkS@Vc%I&0a{Fo?ZVnpnSC~tZlMN#EMI+qWjU&XNCYAb=?w|0uvHCW9RPXp
zj)BYr9@ICOcJ!UtK5_*GzErdFU3VkiD8ynT0BwmG)FB*m;^Rput#1vtI~_**eMj4N
z1hn3T6{CV>z5B~Xc&uSLgBhA}#gubp$KDz&nYAuQ#<lU<q)oVw0as_hYwZedRts1}
zSQ!+CZwB0Ug~e(I7VR<!iKJ2a>mn#Oe|RR^)o5Jy$qYE*MFq|*CK~`R=7uF&*Y#$x
zCQZGXy$mQ93!i*qXv*s#5l}6ha`mnFPWW6svkuz@7hKJR0yx%Tz|HKXTvuBp=xY4g
z;-Tl21T0`=Wy2AuITpY?317?!oGl<|Fd=nXQaE|e9Wi-8ZmooGjkHPxFrEkA5319r
zPij@>KvrJ~st6Uj5*Xn%F@qM2m+6@EiL<hE&Xi7UeM8?i?0d(Yj^3Fp#Ht#SU{AAk
zPH-Qr_fxU9(x5d(t>B$C&me-cJQP=uT|<yO1F77@DGdw<hM1$)a1Fg-$pDCRFIIs1
zY@p~Rj3rqBpVmcHOiFPnsle4cu+R5Ib`kpv)zD2&EVPx#RN~7yTI)A}K{&H2ZLTgW
z79f|{f|LEKgsIpkB)6(i)t9b;{hEc+t<O{p%Su9i*E#?DzO1fQGGT30F6QYcp3j4&
z?_A#DrEW~pIGX~!*^AyN!Uq#IWejYldV&56Qki2wGH~B^ygfLGdvAueeZ&9x|N6g&
z$M2gv`q#V`03iVWoB!GW1OMCq-Tx7y?Fs}}c3YpTwRYjrS{rWB0mqy;rsMkuXH|T)
zj*)^#TlNK946P~lzLEXa`5KSPGUm)uF&8D&i&0}`3}=F7WR$GF%BAMWBtymnd*6`l
zQ}zpRgB9IS8$IEduS+w7v{ro(9*`aaG~&^1GqhWSbh+?mpk_XYP`uZcTV^%R>fCjA
z%$%eeTU7I}abc}KO41KM)j^+{#n4!cA=*5DMzXCc6Z&>>{)9=<NP}I*&!h!r)XYr*
zHj1NI9pvA?vQ9Kq)-4zH(IYAx(!*(0S{Jn06z&E;P8`QOfG0Gqpve=e9rK)(Ig-87
zmFp?2wsR!T<B6GU!h1NzJP}DPL~E<stTCAOs?eHsa^kb-$qqn_kQ)MOg4=z^dG65j
z#F`H=1}V%TS+rQAIh{840E&?nHQ33PMSesnk5PR#THLJXzTl^oX-f)1Se>iTAWwjr
z5l(A)K_!%{=OpVeUj3t@epZim3GVbW)I^T-DgX*@OF}o1*910TvA(kWjsS#LrG&E6
z$vPzRs<OO-(sh<<S>jw(ZT;`XbmR{Uui_<8y(yaO%>>%^8dAjHmDm27NQ_VG{y%s9
zp`b9P;}|C#hE~ntn2xiCkjHRH$;Vf#$N+3cW$Lzd_98^lU@+nQU8^2L>KeR0FBy6y
zAdzUQxb=;L)|*0lhkO4oA}D;mGI_W_wcwlfDF9ZMDY&R}`3Z{*FspV?oWn6^B4GS)
z859dO!cna$71)i6ZFe$^&N1mJRb?(LF3h$WG5vk(=w-=3fA<%ZuyAuukt<8!)$2%L
z3H8@BGun1|1w94>dJS>Y@BIec4Ew#K?}mPB(2Z;(y>QfJ4*(#Gt6p#m@mF93hs~!l
zAka2Gk+Dn&SxICj#PJtmvxU8`%v6i@$^HzX!rppKTGQ9hs>;BF2h|)wXiDsYm#F4*
ztQG%yj(!E%$ojKb|6WAOA`1{^EZ8gQuX7$JbtdP5n535O&9U_zOFzR|z^rtZbE~D5
z9gGu73a5Y1Rh2V%n>A>w>2USEt#xeMji!w;;YmA@)qH=!O}t;%PwM$k>qmt1yIaAi
zRzrJ$xRlTf4skd;N1vQiX$`_X6JRClUI$#VH#y<e6Rfp&X<Ld0jOrnP6!zCua8-cG
z5>|D}gPR#Wr_dp}H`k#r6>YNU%vdsy>QBrPLULW!&kG=e{iJ34oj@gSH86*1&jtNx
zeU7ZeIy0|*;E}9eBq8odS7bA&278(I8bypsHbzjnJ^ZiGQ}A$l<}2AB87!?AHoeO0
z)liZQbWz;aK~!^7Radihxg0vqGXWK5$_HoE`XpXy!FkO4#^Ef@6!SAi_g~n>p&+Dw
z%CglCa3m=L@I|96YjIN~ta#rE4rp+Qa`fKqwzERoZ@Aqy=Bc(U8|dbGjDcgER7hLD
zn)zvt-a8_NjrzXf)_0IB@}kuNuZn<Wtsy|s*$ls|ldU$RVQDbdn9-oqqgjy<nCbMp
zVgL4qZ-4U*U%ote|DEd6xT-?XmRb`kFAYEzWHbwBpU!!<w6A6l3t@;l^qJ8C$XUS{
z1tq;`tvt}Gp@#-VL)Tt(!j=6mdjqOY6blAj{esm9T&4OFQt{6+5}^r=Kp632C^7I{
zsi*MknK&2>9fF2Th4{eUH=HMjADvHHKG-gV+N1$x#<*yu%SuC(5bg5UT5#gZoSQL`
z*jk5JD>0&?Z#O*d4{X~FoN036Jin8sSwhBzsjq~icf0;<1%qm^6h(l2A(gU;k8we5
zb2wBN^NIdEm*T%5<^(fnQ@dPUX#~gMW13gLf}^I`dtCFn`7Y#27RbsIxC_Eb)L27r
zmNJ@s#K0_LJF0-?dcMG?pXDp511(Dc?%%9xBETV53$`rVp2&ocz$lpTMNVL)*go|K
z^4g!ZzOKE&tR{kMsx`wH6P9*;|CJPFfo%Vnru3h<{uq2z*EM?;Koi0Pxpn%1xFe%8
zqL%**`w)s6J_iMG)`TYk%As)4*GQq9v*tj1U{k?u+i=@RV+`Rz_4&F#e!;)^pZvcr
zM#;()v=%C9p)ZF}ikza#Lgp4KufG?r=MmS2TB4B8O}dD{|L6bxzd)v~H&<1nz7}j@
z;=rr7;LeeZ0%K<Hj^3yM?Ok!ZaVEsJ8E&@?j|T&SwZ_uam@ULrMF_Tizv!=-i7^ss
z0uEnqFg0v_$F^;J2Zg61g4s+Rn&Dnhmg9LmnPF6*6Dw!;Y9Oj9_I=0Lcf@%j=inqU
z<)Yrc-f;ivj<4Sy`11C^)(k0vGaM6a{Em^B;+%-4%WsLn!ON#6k^@hydLNQ69{GZ7
zcnVZ{lAqy+wL|zUu7<l<{ktM3XH>B5MgJRj&M0VgaGAAI)#h~=Ec|xPS#VgD{#H7U
z5SVSyI+)f#AIe$EeYerpJ(HNRi(k}XHkLUeGkqYv)A$WctmxhUpu1`{KFJJI^uDu-
zaxb&Am*9%Gzys5I!|is*zCR$~5G)19fQFBvv!2(QH)CLR7x?MdpS8vSdj%`C7u8!T
zU}H6%9nwjabWcJfNGj!}UCmz!07kSRA~{36Y^|4NJOE5##n79=gLA;m^%8{5(h;%O
z^r}`!YN(1$*_0pW#Q7{rs+Inr_U^i}g06=b?47K*gvki+m1Asu)|?CL5YbxGiE+>v
z(j$-(<<nL3<!io+5ohH)m1;KETF*eaq6X$P=!~+v;SjJTE^J`(xZ$r56s&964sc^e
z1`M2EycO0AGw4<&by4W08d)L4wzLkhCF}@~GNY<F2G!Ie*H#O5qDy0I4NpHAVL#p5
z-~P-0C;r#}oBx(09Np1`9`IZ5G)>zYHtX0nS|faXeBk5#1IIalps&-`iq%w0V2+9B
z#}i)kz<s;ne!Bssh#X_!`^Sm@)BoYW#TXS7WU(!nqD8*4UT(;RHm*`PicP?Ag{>R<
zuDCr~(LomnpL4CV4r7|)Wl<CffovPO;7ff=OtFRmiJ)6Y-*4!>mWL|lIpAXgn9Slb
zF()g+5d-Hu>6>I+SPjIX7rRSM(7WLFWk==-KLvX&?A8qJal`#~$F3ATS)<_Ya&C5z
z3y&->Ap!a)TZ!OvBzJpF>60QS$V=-XE1{~=JDI#d^wVZu4MGNj87r~nrj1Ni9WraA
z<gEMP&&j|wK!@nS3;RtFR`8rf7hFbo4T1#|&@vrD-i+3iFb;;f1KEng{T)x8NEp&7
zzzX9*=2ru@TtSg&`f6ZDGJroGU-4ys!*aVlj}Lr&{DwJB9#-lQX;*(lg}MDw?Gp-P
zlX{GC2_%<S(lUKZB*rY)+Ok;J)e}kpV&vE>p}GvGm9?VkOMKGZu%BgW^?6F69Gr5!
zGV(DP6s)yM4gw<wzb}9=ubJwKGzPg?l;WgPIb~78qGqovecgh>HJ_Da2osd$e!JYB
zSMR}-WSe@UaG%%nTuiYH($A%O1#`0M^_j?#6cmC&Iv?zd^-a-(PR3$n7oiebiqKvO
z;moXty1*>Z`oz8k_MO@=HZiIs{~gyK3JSM757Z_7U!Fl3Wbkl0l%l$TST#(VB2-Cf
zG4l1rmE#{*nG3;L<<zL20*XztMod}_9=G%w-2S@8Iq~oR<F8~ixzn5JcEe}{pj*?j
za;DePm;+vl-32ei3P{Sv+OnIy6qp4QhzAd_2+04R=e#J4S&gcf-xC*;<Qf4P0BUxF
ze;Tf>Y#JdIYN$rxgw2UmHke2~B`yguG|NkHx8_`%(XN`pKgHH}w7!F*IK2*pyfEm?
zbuw2IEXy@PiubJ53}r^sx<Ok4(&-~)0n9t@6~t8)**S>5+m3`{Ov?8iY$_`BqWZ1^
zA!~TirH0L#ON+u=i^ah`{%Xbr*wYXtY#~WT*yZjj0#QOn#FSUg{mKIU^E=7l>;R!b
z9EU0KP><C(M1J~<I*|Z}y|XDwTsnNg5d1x@DO&GP+0okrxBV-Cj&V3bPedxl85qxr
z7+xM+?76ArrCb*zA_vmtsvLT|;isSf7C-OwErKV;eBvJz`1O~6z>HGnTk)WB&Hiwx
z_IjLxoJ<hTll={9b+llWTKLIIUlQo8GuRZ=v*w(vJ4F^@z$rj%iJAdHvEik94q>uT
zRk!DBZU7ccS3h%&VSSF18X73ND)lg(lxfo!<4h{}(*<}9vlqgt9z{B6n5I{6Q%vaF
zbZJ=D&&VPXR7(S_kRJC39`^?@6VLYpzx?_u=A5w9iwU*LOdpQpd0sGb^rFVv1|`Aw
zn#GXDB#bZ`XRgnJ)Vz9NtTq&J4;Xf6GNB<{^;ZZATQl@+E6#dF+F7EOVtPa!d{_+<
zp>=Ht!6U7pz$OwYY@Tb<84uy`f{=z|1BZR|jY6364SX%iy5<bL|N6W<U9cw*ieX+%
z7-pJSOopV`X;l-;24M~Fo-;W4Y*G+X`5`i1?v%9L_{UF)BM&bXCzir+RSLMw&{hzb
zbI{|d0-AshnIf&=1y%s(S@T{^%KVE1?o~UQ2JYJi-R!lwuIn%VeHNVH^F#)PMyupT
zQS7|tIWHVpO`(N%_RY|`V%rp=8@5ex+Z+6`Qvj#!V)D$=0;q9hVHPwsYo4=IbIq{I
zbh!-r*4zfTlxRC&2Xk%ih0$km#ibd#?gjY>1X@|G@nBD?)TK7&oD<as#`2DtIgrUJ
zfozk-xKnr%=U)1Smrml<qzB4|fXEI1JQmcnawP+z!{m-`U(nkZgbU7iVrIki8Ywy)
zEd-m8-pOn^AWNMZS;-YAc%=f{cE{uX6HGeSngwpnuz&oDUrX>sil{@h6HbkZ;S;_5
zay$aZIB}e070yY4=H&y40)k}C8X#Oe;RP%5*=h>3HC7^`Fx!|YR#!KjCURL7+^UMB
zu=qt^7X<@!u4XNP=xdOD8jk109JI23+;-fsLCa>vs>%DRhD+Qj0Re5tJrZ*;L2b&h
zu9<R)YT?O#sw{;BnUr3mI}v1dz3p$fw>#7nH@#t=0Y87k^uPwLbuW7L7=wM1tR6&C
zASfb;&M0Wgp<t~UqEQ$b5u`6y6M0S=`l*6$4<XQOh)J`|7>VCGW9N^G*E;SsigNL;
zSJ@o12BHkk5Lh>!Ny~;bDRmw#1V&{HGZMqS{EjMHg_ao3frYuYh7=BHInO~uu}=U}
zW|A6Mi^fOb@aLsaR#lt|@`l0e;dXyzw9ED(GHFP+dK%M71I<pVS$Xm-F_qb09=MT-
za~?eKq19NcAcV!A&&u$)@W+`rpA+XO23N{;P%TD+M{;g+B&-W&I>rn@Xl`d_Xd7)r
z=#NyJ>r!?mtYD7SEcH4Fv#JGDKJN8>>*+4|&5Xgo7Q><X2E7|pH$-PSuc_d`7gk(r
ztRp}i9Uk;As&#Tz3w3~N`RxEq7?NUzrEEa`psS~&`kJe}*^XP^`Sn@K)uNG8WVEym
zY*qtIFct*WT7!lZr=<>sYJ>{()kLfS!d^D3sC8fgcr~PhN(UweJdjlDpRp$LH6BWn
zO5AfQjw5hBIWD%PL7azBLipK=(&*u3yw~b^(BsRK6(mKV0TG9`4PqP`pS5SCAe0Cv
z&9d%$C9_MxFZFu?lu4q(%l3s;HY&h2^s<GB>4fMOcCWzJH>lmOhOQu1y9Z17wXCS!
zlbCv8jUobsAab$kcG9LJ;Q1ZT_dny~`$@r~GlRbG8~1<B8OdBp8j4y9j0@5w`Jgr{
zOh~o$pm2wYC3ue2c|ij^K@1AS_IBfJ7<6PfcH8jt*Pk#?P8n3?G(8T*T6P_J=^4UV
zRmMyPB?JZ%Z6<O@vF)l^J&BMUqvlCN2!DD&a{z9bN5Wd63Lk$3_R(Id-AEMNQ4QTJ
z#HjGyW`$z`g*)ptRGtBaSxGMZT@9(lVR2%bB6cxCqHx6Ld18}>;eq2H+rSpAqbfC5
zK#A{0Vw5ETJ%4H=G(+xoeQ!*!&lUCA0ui&A6@rM4nMO(+c*ru-|I1Q=42p>aGo_3~
zx*#TT$7a2zh$#??7!&h!oX?5p$H`HE*P;!KJn@cq^d7_q3m)_`Y_m@N$KL4&MXLuX
znj{KSUl}=@yeM)&Og5@A;$rYDFL_Zx>x~DODYm^`BOaF#*Gs{xwWTV9N@j6lik1|<
zxl>`vQLJ(K0-4dZ6GbJYzM!-&Wi*#y7Ce1$gQEHZh39XZ6$Ns`);W)x&XtZV78JS7
zK;b^zl*yW*Kv1%(%#hixgEz3Ir&%QrR`>7{0|`)xN6{<`TH-j`EL#oK_td+vq~}y4
zq>vE8fNh8MF9`et=HMZ)Y)(X5(RFLqa+E#2Cq!6vdQP0<iI3+yw!LAqJ1e$1an?a2
zg~cy2g;vmA;zSWcGvK~G5a$U$&k7RN-vRpuo`GEbz?nGXBykQ3vl=jE0_ogiCJp<3
z$F}cC0nYQJ%C^>?7VH!;;7$rWVOTQGd?Y65Q>50hZ(s4{{slP$!yCT;`H48b<0$hx
zNujVW60wx+QR?)g#<ngRs4izwDZjlyv_#Gd>}E*U>Va?zT~%r%Vk~`@!m!v~T!;I%
z;qkcRmNy81@qAu`j#}%`$q9sub)^b;TQBH!-&$*FS^NzjIM14uUb8lki79{|z<dgx
zn`7G@2*Eu53gb8<Cg!Q-=t*_2*1_Idqj)AAYlyrE%7+6mSy(7DDzhrndPk4Mq`w2b
zT8;@HJRE0C9Or?LXM<^CMq}UMAHid`UgQLg+mu=2DEHLJ1T-jNS2Ch56|FE6VM#WN
zPRT=s&=aY%g&G5i8t(1zw7<9ya?@WBlY77i?H(i**@gUYA?}O-eMts4)|_-sWM#yx
zUFL+U1KMzDKqOykW;o0`t-k@R(`qR}QN=Eys(Ia~N8p^}I$Qu?O<NOD^uARvsSLWr
zhkWlV`(LDk)tIYnkH3L<=9+O^?*mweekoNE@R>MsAYyG&&az&c&x3XZW0V9(%VWQ?
zZqp6THfDOESEiCE0g*co)vJFLmtJ5U;Ovvt=ZIGY7~w!!aj*FsL?V|T&Zmj(3nN_p
z16jQeV5Q)yHKU<%np(inbVn;H@9oj?c>5Li`+vlF4l0AIA7*BllipSy<rJHxZSfNW
zC&C2ZpTFSNcH{(l+xc;<q3;jGcp{71iF9NZwL+LFuW5gq?Kq5<_^f!UdNsq80#KGo
z+&Ubq>Wb=F@Kf`z-Z$KCceK7?Mv$H^hRDyfRsbNKd(7(F;YH7kbZF<ih(G=OFYx8|
z4afHb=Q(h*9h#(OMoXBq+WTcEQEO-o%yMT_S0rl6oM3N6CSsoK!?m4z#`8svz#=mc
z@`*|4EM{U6nKBdI8d4R{aWW8s>req862E=_C>RZ_Y$_Sl$^8H;)jUXYm@7aDYIZdH
zX)28naJ^H>Kp2Q@I3YSo2G!ql{bBK%$DrHkw!xa>`2G$b6D`RnW_7NdpR5kqlrx0|
z+z7=HM|sUR>1ZC1K`b(|xbYqsCoO@`nvgao{n|$+BLDHN(?2IF<1hoXanA0T6FG~%
zR}Pi;`;Ps-!J1%YeY8EOhG#v$*P#1XMX>rBf#l%%&$p5J;w+0`Mm0?<LI&7u(`HyE
za_jzO!6^!=udrs!O9eU#akR|wT3JFjVQ^S<)C+5`p9_3~Gs-Xzf;tfSc{$>dqO{gG
zObo!K4lo`B-8%hQYIA5-5JL@b4_0a>R-eCN*-(qHdd#v#k=hJCCM&;_)+o(7Zu^ej
zw#r7&%0wsKE62c0pqFL4NP}iah9N=kEV6;P@k0`01_f;(=>XNw@u+8^7w7|kHWw~|
z^kndZl+A`y?OjUI#Ofy$8wCK;F6)M6YhX1iK?@0yf@!3zH{1hF4ZZD9B$^y}-0t}D
z?N9hE6F4XO#u?#@)kBIU70SGaK<0#ib2pyP@A$<!Zu1Mi+$j{j?O%YehL87OaejP9
z=D4gRqMG?uYfdiOxz)SV+E*O{!ax~;-aC40@bloo+8g$5BNcT9QUfm3H9-}s9oBZh
z328Oy^gfdVg1m%}GbY|Yp72N*STT83Gq~Sy|M_3w=Rf^V@%8=<$CnfDIpF{Ozr)iV
zF(w{zBh+dgeZNCY;Ln3dz^uwGNS8g_(HgC{p&QRHA@v;+rGYOS4kvuI`a5K2MTzTZ
zV}3jj?ADRB$5X(HKpW<HU`(njBLe61#2j3sku<mKwHC7=O(_#Vk`*}pG9ieN>POVs
z>HyLq4km~dsC_Bm{jU%d5*&u!w+?NBZR@yq#rI73_mlkUM%rO!0<kQSm8=hI>@%R~
zRY_*T8VIpeu&519VoBnd9NiZINHGUapMQCC!bX%SuyxW{p@L%!<g-+g;5~o&cE?ZO
zzCe39etL4Y!=Sw_BDrv44Z+bX%aFS7H7EI!l>g;}dhv-1yBssCAqw<uFIXgAJ>Y6o
zs-O<uVnSGlX~!DrCxU8<D|$2ZzFmx}HC$dY;FnDRG1O~@bFqA+D7)!_2`Miog*1wr
zOa$pCH0=d>NKx{NgFT#sb0VCRw7@18tXCf)uS}nn7jbhA5a-t;6z8a+CZc%U9-OFV
zjpE{POp$_)yvpo^hr%)tphuE>KudFiY65%A7u!XuEm?vo8lthBR{LzHK<t~eNkLOq
zM;t%NKbw_x#wreUVbx_5UEjN&^XK&`3JSu424&~@1n_}oU(t;e+a`*~?T&9h{|sc{
z`0dwHk5?tu8sG)vmbi?SO4f+d22;B1-gm=ozhUbQTi?;Q8~ZPr7{~WZsXS{K)86PX
zAi%}6k*W--Juz06D>*4>_;v%fe#6!_?E8&s`0C$H=L`-nLPJXsifoM&(45bZV^bCJ
z{TG-rV{g{53EzQd;y6#V+ZX)&^S{L3{oTL9PmixSeZs}@AAb1-|J~pJcV!9~*tQ$~
z<=_6-xV=3PnfPws@t(ioIo@$P=aPF>4Xt%-vSDwxn$$)ELI-sbW7drG+OSiqay==u
z1^BMNfB!(kKR`V2tm;6m<LkFSVY7~Tp7?nF%IeWtS*Ej33XwAH6V9Zm{i<53_+XGC
zgKD80th$#K$z_LMx~$&p>U-v2!#W$n`Nxm@235`u_|U+)8~h2N6s^5?%yd|=!OW;e
ziMIP-l&rEY>yTfA;sx|>w`#xyD56@9H3_XI>0e%1i@0~gZSUB3I-^>Xt1MLk9*+%Q
zzTENl<wl;q2hMrchK?!&$5p0XD#T*xq=Svf%<@Q=fAokbQVB_Nj9rkE@Q{@0*yq9U
zz{>L0(3oO)s6nmMrv_{oS|jGISF^F84!me;mz9DqHUmE4rNqTQ+YzW4q~S6Jl*X)b
zLqL}(2g^Su3X3O;$dgpr=>ub)=%#470~@Q7ERK8?5tuV^oD?Qz24M<Wv_XY7&ifT1
zKVC|lbBuxKIBS5D6SZw*<1k=0=n@7+87ih<&Dca-9#aUpg2M1w4QjHxRy75XSwq)Q
z&Mgq-`p;archP0ZPN@GjS&1!5HVC_+DgR9TvM(oXdEz+V5xAjm17?ny%t)Kk>#A=Z
zqcb~{##z%^`)kC!W|dnb0YK`YUe?L`{=n`2M!Rx3xd(4I#JB-|VvG|CT5Gt|MjBNO
zxpVSf?+vF4D<x4#W9Hnpjy^W*`;J?`VQ)8>y~d=5d#UWpBED3(Se+ta_N&_Pc--)K
z+&TXt1kdLaTOUAjtmRU%%l(G^?N9jm=fA_>{`9x_vfXjU#Pj$&{N3OFOSHb@$X}2u
z_?y4`ukc_0&3^-HhT#*tKCs&jzsN7}Uw(z-gdYbIfNctM#TJIg{ee?X?%SeG?|p|z
zRrg+EMe;yv&k-t*I^aJ(o}375iu;2eP_6CQ8qtZ5<9Si^*AZe?hLz>gT8obVw=@b%
z?=e{|7zEQP3kcB9jDCmg=U8v(TS2;i4V^;khTEg#>z6whxlbg>T3Eym1~789*TE1_
zvWzARZaQZcUY@`?X%p@4fG0HQ2xw~8>;aLQf+b{w6rtyJ7^|Y~R6O7IhJ9~Ef8~Si
z)k}E1?bvpXMARzl*MRzF9DEYKaQFHhLh3Bi;a&%1qQ)J*96<rRf|?Q{J8=^8wstXO
z)>JPMTp25yy9p>+UeefHWGj@D?iS56OT4n+7@x}u+JSW#x??&ml9#2%rDUzZKu1r5
zC4-(|3GZHu@G>H+ClZ(KHz{y)%<9`Xwq0QXSkXo$Cb4Ts4dJPb)Pwkwffb|j^Yy-6
z;-CcYX^esMIPpB5SeZoMcHH|7t>2*ShTH9q+kOXBiVE!290nw!<^+&Ib79^!PeFik
zN7h$Gt3XHD9;aT7ND;vZujls~#Tk{UCtlorQjQDb)f&4PORdel!T!#PlqWup?||%t
zG^$54<|On<wgzn5u-$IZ*>KDgGM|Vk(B5n2#S$$F0K>LFaJ$jEq7{S9+GGV#dxjZ%
zhqZoj{#lVM@o0_HG=a}GYrLkuDJklY#{*3pHr;XAZ3tmTOu3ra>~A>|v#@UN>D%p&
z`~3~~+no@SZHP(n-kbwEo&7=)OxyAH_7lFn{e-W#FSxf2JtrQwFZlBH8{WQs#XtOq
z{~p`z4c~tLTYTC71Tj)>t=%B9VNS!GhBM!Bd|$?qHRIVFxdUB$sjbPNTpFbdV&q9p
zs0$OvJZnc1z8&s}ap0zRY}*ZczgIGtFr$fO-+O8M{37HSmmykXl@`Fo^0Pu|kO~-k
zX;IB@_C^Go(JsDqjx63E4f{?Rz+VMEvQ|Umo5Hf;40gWHF>s7YXv7M#n2wp`+KVVg
z0MDX#PABv*2Tj(-NuI4wAg5P07}!hzYk1o>Y~{2WNVrIiB79vO__N`Dzv1iKhHYch
zPM=x+^9gH$`<>ZVTqE%+V+kHqs*2WkJpGB|JP@F(>9$efETFTdf<82&{W>zaWN6Nf
z<Y-G%RxKRw=Or|8Shi~+tb=4Rujpf^m8A#v&Wd4W`p}Cd%_f(4m6E6fI11kKY83N#
zygXK3L7}x)4b)l?MI&9CU{y?Pv6l)RREYXPeZ@G)>z6EfzS09`dE_IRam_Kym@9zb
zP$Cg}vh-!?-3j}7B%bGi_m3yOe}CuUAV5PymkpcUV7f#54fp#OE_$%JNL>lnX7`#j
zR+ZqDEs27ITapfwV$ki5wlOf}9I)i@cd14+78NjO1@_lSP~z|X(lE?e3UR@*)c{Dh
z%-T000`G?D@5uZG=?%^9kU8-@pZN9tJAON!d{@0=-*4FWTQ!^SNX)CLX#y<&RTG-s
zwSL34Kd^0g^hU@gMe+Ul3;Tn`;%RL|>(%IPc9A3^YYstKnXGj)24Oi*+ETy0y<x8`
z7)f0HIZim(ui>7eIBm?cx-kuE8@Bz1y>EDIcf8%dkfoG{Z(sj}fZ#lyh>wHzIRf<i
z8@|5%jK}Q_Ti>v?1|Ji9zu|U!;PLp15s9`vu<zd>w&Q-kqir3IiTnF^wCxRVx1aF+
zmp{YL#Pc8iBc9KJ77e{8p3ea__H85*-8yO7`KfVUBnbvi_KBj{U(b0!hTv^`!|lrh
z_b(4x(7@nx@EtY7moInRertGtSLhT-1GEEu7u;_g#X4q9Y-@_%Nj69Xv^Qirs|0J*
zkt7FVZB4P=8f;T+cf);e=zSx_|DQW3+-el(oU<ld(Xr8;B8n^P0yEwq?E~(Dc~*dH
zhSMjjgL6>uGA7HR;lK>KqP8_Rw&CmB4fopy>kUJyeEY->0k|n!=V(VdPV%G`Bn(OY
zlb}5NTe%m$GU}B@Qz<AoJ#Z8qE^8cKD?Gc2qN^d*O0D_%K8QoR2P75UI!rk4;Zn|n
zrTDUJ;6^*?@q>WJ0WmKqfNYS5D0=VL@4%-+dxau;*_W$nmFug8x;~$Xlyx_=ebYo}
zu;QtpG3TuQPE9>q?3t|ar}=8w4M&_#@tru($T-Sk@|<J96F5EaanKgrBcal;cLw2J
zV_JJ7Oaf_$<QUcG^T5Z)2gE0~1)(i!aMIbc_niiClGwHeHD;$O185yX2$M(#kh7ZV
z<Q4X9!+z(%dB4A*^^SDM`TT%5nOUnMUNzxM*O1o%v8q4{-W)mk`N`5E6*|JnH1o+y
zBPTKj5(B-RnB&P_+A;9`<5&Fp{wtny;Bmj>?b}cI>Fb}+TElUE!2Lvm&1Sl{a>Qc<
z=O3h4xt{Lm`;N`n57=TtOfiy*_TH<9v}R^BPE_;&L=3Inu<sO59>;<6JgAKjgO6D(
zDr#l6)*<zIPoL0aFfq{*yw;wq5vpc%;$1V}H)Ai-tsqlrXu2VF>79UV9rp?(y>V{!
zk;~QBYDiTLsgaKU_J+sTuXsEj5Cz`GiMOBr2G3vq0srvt{~eC!i9i4L9p~vdHHrqy
zD$T1$<#S*RFHfmOV`qZoPB#J+<=s>{t*!z6*0J5#us+8??+VwTGi+0Q{q}}G9p5;Q
zzQmc^z2Ww@;c?plm7<#_5({!>|7F`d915L)?uIcZw&42fO>qCR!@9sW#cktr`l~3K
zdr9l(B>dqu((lTKK&gF7f_!r)u3mqQ7ziEJD0a+aFc5fo%byc51}p-*Dz<KTyY2Y$
zwsV+=Dvkus09^i(3MvX{uSJoQ!?8VZjKC`Hrzg%aDdqq$GjOg+aWgLSsdKRU@hFM8
zve`qJgW@axQe0X1x-tPDOCj7K8kmtdOPB{~Wa@bRISS9e4xGd>9jO6pJGQ-e<+AJ`
zD6fuu`TUb_pRmlUH-qW$j8Ek$8B3<^!tt95v13_w=l?(2{;Wxo?8?$ZzcZ<-nYl+q
zCV(2cixl?#|NrEQ@P$uT$jc&|4OC&sj0kr(Q`IxxhqYBbB2g`k6uGdH6abYmn5&*)
z@3q$eR@%Q~<2VV14*F^D8iE<4tL>Gx$rP;YjXlcI(+_A)KT8>nJZzwC!}-N7@04_K
z#gK3~a+9h%H<SPGnNSB*F+ZfcP)N)%AO@4&Ijl9;Y=Ktg`xK*Tm^-7wXc?NERBzW0
zJl#H$(uHNYF<ma?vNL#Lf4ouqgWifdvmU>L&&COhXf`6%v}G+52}z8rK6Jpe+Q^$=
zQm?tSpq%QN{7iCD9ri9$wqdNKr{%`;^30qTrPBpZrLoo>Vq!K1ZW`cQWB7<8G5MJ}
zMg=N)UGUNJojZ<P3n2z(>j0cH;=+{V63Z3n^s*w_a3Nfua18Rg5}ho~>4z(_STCeV
zs*^fPvz+ir)b9``Qd|f@&b@mn#MCwW!(vXY%PQ**!<c%A<S?d*X}&N`Ghvpe+7vG^
zUT8?+@2cZ5V40q{efY@phYwtrS!dXuxrCV<FSrr-dVlcK%Par>hd=00-=*9Kqr^AI
zmNmCY!4o}%H0ca01;@bk_fV!VIhJeS@)Vh`f$l47z7s>GIoW}9Wb<*kB|hYtJZAi&
zndNd>xIJCC%@e~7B{1eDs{Jr^{t^fYq7TdjkrqmXVu4u66BDE<kPK2+8s*<((0mNz
zP$15>?PN2j!fwsc6%SXMlgEnJukW={tzz%a7#yYItMKPOB<b%(56%Y-jIY;;d71ED
z2g+Uw_r0>0>ASCXc>RQ=Ha~Yk6}8LhR94f0J{;@5o1c&>t6>Zxg`~VYbGANCirq3N
zrg<TSK*@z}n(=o^>G6zoF_=CV=TIdcg9U+%lh&C}8wrWoMj6ICHcJri&sl;stzu!=
z!Fo8lI(s*ETbNVrame$|sxjt}>%$lf`pVuMP=~QahIt>I4yCPGVwuDBCLVOH%F*63
zr3k*txv>S!M|>%C2L<T%^A4-5Q=CX%%o4FAO#R-+iO1^M6db}yYpu%e4$a<;I#V%2
zin=NeTao_$T%*#(YN+qlD6TWyV45yGKYiq~JmLMs?ec+Xxe(k+or{L)t+3ZuMjs?U
z5Jl10>U|gm`e3CTgXjZwwtSGL>a&*4wl(s$;~XrLrMg9rC*cXEb~Q*S$iyM~$Ys7V
zFIUo(n7J^5<Gw!d+s2%vBJOQK&h%h`v+fu;n~9)5Tx}Q2x}G2=TufZ9Pt41O<$3|<
z2}%29bX^K>F;aab@0s>kp-Ool33T^vRYU8iGlAG}>H^|2kwW6>dLzY!(H*%~_EO1v
zX5Dt2Q&>==8=F^FAEjhch)f|eeBgSyu}q5s=`k8-nrXw4izuSY?E_CAK5|(uT$YP2
zg)s=jk)0=X$EOdU_~Fx!{P@Gq{N<1T&bqA{=+g;7>nK<4frI>U>{agrE(Uz^3@0en
z8Pz<^k!cCc7Y%vG=+v>AUs|KG5+>zAUq3{CoGx58At)gQZcF4kE2Xgy*~8Q#qdw=H
zv4})*R<MtpSi^W1mNJ^Yr}HFlF4cdJLG!*9w#R1JBCL;1i^S|~hdc@J%~35JM%lrY
zZj`bzx;dd%dg=IPyy(%TB{<?S=ZQH@EX&L^C)_YUqHf;qJ$yH5R?XZQPF2MbIJ4(M
zFUlEKsLy5xp$=S^0?linS&!DS4Wk`4yqlO);=0Vtmx~xPILd(0+FyAvNfg#>j=`DW
z<pPwwP)8$Sk)OS50F6L$zXO_+;V^xJ+$Pllq%+pg>@~zOwc)1NLk9*t%t~9HMfPWz
z?XO1GCTy|qEKJfAsWvplppnzNzBL^-LyBy=-cv1&yk)7d4?NQcMPav&PhwW|;1sHZ
z7Nzexqz`DSwQ>3@>3y2tQJ0D}I7gN7jZqrS!0+gaT7i8PeUXT(1J_ib#E`y&$~sPp
z=P;l+F}s|nH4NfNMC+u2OL#99*<QtZ8qqEQqh`KqSB?w8nPT=}KWRo(o4U`kRUU6U
z`??F>yhQwH#Gv&u+JLTZ+aW4raGek*=4Ih>xiZfSI9*1U<&|yEGL{=!|0&Ki@4gWs
znk9#agF2C>nGh#RDWv5}m}kN~v)rD!TrV(SzAQ{hk(=8#)0|@uiTu+~l=4-He$Fwx
zScTmUQq(SN&f$WBoTiYtOoD#nH0c6%fqgIZHb^naMKm{R5-Tm6`n{u0Q|5h+y-8%5
z;zZ~$a-jxC_Z}A$3`dMJ49{}8>A8i(91}~j91>%c<qq>cbD0-DK7HWna^vgg&lF!M
zMY;?fwnrMIRO2{DuR6R#bhr@cL>isi6dN25mtWWL7IFoi)_1mir}x6N1U!RVoJh-H
z**)22fb$f%%yM@Ie<B>U3cqdL8$7DHqNv{1N%Ni@=S|>w`ZoPN2F(ml$xTp-6C0ry
zRY3A;S9{+~iCp@iqrovg(3}~$b+ul-cF|UyDyGr8<<6?g4Hh98yt#rh<N#|9hu?j*
z-YI!w47Hzm-=)~i;@bBHt_b@}Q?h7F%Uc+OY6Dsg(Y$SJ%_>q#EYnO1NmYD^^q|G^
z%f~!=XWw^v@629R;22uuwMzw^Ld3bC;cY*Xy}C(<AQ`qAV`NE57lS3Njiy6cgJIKO
z7>snx15P~Dd0fL{{Or<joz}CdCx`sl$Fb=3Y#XxtIL2zh<^zHbJ9#VQEwev1*2kvJ
z?G<dIbHT)P=Y<rB-rGB_=5wS9sMSsn;@1Z!Ur2{D6FaC*hppIW=8bjGYP;0MrR>z2
z@g<Op@bTay!OMP3bmT#I`J10hdque(g%~Eg&{g1QCS%=_L0aKRDZ`vs501bX^sb?c
zv?-3m8MU>>UOL;F+3q`UUmxu2Dj2dR96NE6mzTOAqdK1yLPwex;xw}?SLS&(s%Rv{
znR&S}w$+$1$MpnsXeBGi#d0%BE96qhx!_#jdi%(azxfA=&C<kty)Z8q*?FhLoF*Ke
zyxh37C;a8cx@PV_{e@hsQF{k_&CHAR55r-;OOa)|aG7S7G(qy_>>AGC8Hb9qZY;ML
zHB0ifXvWq$%k^sOUwh(?d(Q!JVn$&?PKM(;U(7EFTLYT^9fcK`v#RMR+!#zYr+D>;
zX6hRz7Zx%^Gv3r7(O5zvq(Jnr_{8u~YG=MA;uHxf;KE?IVoWa4tXnU%QSlv8^2~9T
zJ$~@a23MykknDN&F$k>_=U}vctyw|Naz!{KTP}Letb4;M4$rOLWx?{de14sxxhS*s
z@TC7Xyrq~)RX>I_PGepvcq^qB301mCX{BnG5XONMo3BS#-oo+UTQyUv*4T1k+nYgF
z;jg}*>yRIr$G%Z&R;y)6T&;G)agC?$EA;p5ffxeSRraPi$mm8(ZED*N$Mj>c7@Rr3
zeq$U5ykA7lllzWGQuo#>7+sRCpRd(U%YsE>h-MfF$)xu4Wnr0T<ri30dQ|bk$lvt_
zU^9|l*=wQIuEW#IiBXQ74Id(Hbc^OYCmHtb#dp>r(u{j8kNhztt8aC%??s2)R>|4y
zKqAx{Xgx|DJDd$FBPm2VtUAx=MgBWT5T-6f#9Qy2zSmIacnBNif2E4(7-Y)qu8U4v
zio|VW5JMsvb<?@T7~!;f4<XRIEL3W5w6YPml`%Vf46ON1Xpt?i<no}DUE;iPcmo|X
z8FzMJIvBM+bA4we$x@_IN+p-hdf$1xuDrcG*d8mFX~vD65Hsnrv*wNNA|(%8my7Mb
z@5D)7L2;UymzfwMqjj5Mh&j}KlP8&P<cHiiFOw9<`?j%98!-lQRXX2R3d{AGfB4OB
zcz*t%sN3a2nkNnYLy|rrB$j2yG1&SOX<o>=@bcv^tZy$cWd2ynPVk=H1?h7JOtXIP
zJT3ap!JyfySqV+79Y!PiL^N9lkAD8%8rRE>4<A0Fl2J^r{r<vFpMT;vKm5p#KRlCS
z0)w4X@O*lH=8`7Lw$tl3)bfKy-8-dK9%UytxejmJX8vf}ODP(IO;w=^1+-7Q9}pu6
zb&Mf-rR+7WVO*z<MjJ`{*>^fw*$G|ej5EfD){iq*p%_j+$}ZpeVX!ffw3iPLxH<*1
zUNOaT5QNs6;Mp;B?)550W3iOL+P<HE{`Wq=>Qbo21_50R>tkcvRE;Qd4jnXJ*k~UL
zfT&}ffiSg;>wH*Ggsv7Wg&+!M@5)Okl~N?mabCXWC5s&Zru_Tmr(+SG<~G<nG-J+~
z(G`f-U%k=!7$*ZLI58NUR|xuXvDVfpZIGJmgAe1sW?1pJIM$&K^*FkB7WNV8wQ@MI
z)>>s#-7KA{iAH#M>$!1+uc*R4D6_{A?y<q|01gqH_YAYUF15%~+YH%4Q4FXrS*mi(
z2^9^ij=$sJrXL?S%yz>eO6}+nR0+!^*U81S4Xq>_oH{R&UNpt^G$p(-T8gQo5A6ML
z_H1UU-Rv1h?`c60p**Jg`D+e$Sz6QVVl-JHkAa^YZcO$rFta#$%sUq-Rg_WZ#rRs?
zgW4OpWxCU)Qc7muA8gx$ns=pmnId>f35<gujYiv=G6!;AwQ0|6jBkhthXU2w_%3=6
zxmMmD50?8Y1EsG~WtGs4h3Cw^gae9n?i99#J%A~p-Ql;Nn+mZ~xZmGwE*02I=8-r0
zfZzP)xBTWezhjP*i9KiH^e&8Kz%)gs=<#%14Lnabe*EEg{MV;v9=%fEUTLGs=AvdL
z=|zuEb|G9Qks567rNb5T`O&<u_C`vbeNSQ?hQKIU$#a*)Wx1+I@QzY4Ps_r~%NKsW
zzwo>ANrra~YRSwoF@+$t_O?=s65(2J^7umY#J9(t`}$zdnNllXUtd_a&8&FDXmj44
zbF`qZnAFikkB{J7$GlSr-ZS;2&cLfc=vdd?s|oKMF?f!w85cCWJ81C9E}G$>`zFIg
zFBaudgyLa|kuofvwjMCG85DLS{&cAO`|N8kN)73M5BO-yn=r(x&AD$o`60QuKm6<+
zU<O^KRq^JB#BITA!S${mq}hY(9foVNpl$j?6nG>|J7>pQ3u_h3F$FnBe*5XIvu(Rc
z&$DQ*-KtI(Q22_YdDzRqH9Ks*EEi<#XZgL+kq|>i=C&LnfbR^o4#T@&Vn|1u8FBPh
z5c$t37SDS|ud=;pf@4}~j5=_o8MW7+hwyO2Xs+Jw00&Vp5o5?Y!I^DHKdKIYc1k{Z
zahL&@q8(VnzK1T^5e^5?_iG0tKB)SM*71$js18yJddDT(rK=$OU<z~`O3IsN1<80x
z_dBY1TvyT=t>T<6%!W2|axzJ5^-#Nd&8cw2DQGjfre?l-XoYHrvJZjDjOK!qD%AV;
zJPg@dRMA97gWlL4FXUYguqD4S&y!XC#R`+1=o2Z5B8b_zi;|tJdu+IA!n=vqpyy1f
zg%D;Of#f@rFVrM1e_@kFNtAVain`GA{y>RvjeUO*hKAuG&BSG9nk7QpN~P=`e1z6%
zdo?J;s>IQ$`$`Fkk|OuJsHXRQW3Pqfa^r`OANhE_kwRpiXVNU4M>jtHVVHQ#bwc#a
z-t!?|Se`!dWzUq?uiP#(OAI&~gl;rz8DP3G@I#s-(U>z^)huY+H?6(VxsE61n3S^I
zg3g3i*#>-C7JhgA#1bO8Xa3L6f8Y;){CC#6%TnCjS$*<+cz)*d>#LF=Yv$+A|Hj*Y
z`e!b6CV1_ctrQ;H&f|XPzTWxr`h{xB?%)HrWsz9Vd0O4X@=1=7U`mcSdEy*#)4(qQ
zpCX=4n8D{lDNeZ$t<$n>>ksi+cio`(qX<(}h`L@+jJiDf=s1HTy=B__gAd*ZgebTz
zNH0`erH4VTM=prLh)zz(qe*j8>MyK;{{L|)w4yJNv!v_FeYG!&dmrdE4a$ZoI&uEp
zcgK1BXapOS#Ne55vhPK0{56SJ>#VBJEt9*fFk(>k+>!G0IFLsl<gHl7YSy7?77dQh
z2o8c4Y;#qxk{5pGOlakYV~R<1MLBq>l|+^m%?5%BC>eis8Gsf*NdAOC2AC>aXm9YI
zV!lbe_A}jy+6}jN_U@I5;M|dV7EWu0Q$*w3=@st`e>_=62rg>1F{-zrpe(YDVsEn<
z{aNAoz4l1Yv^D3%YKS^4WYu67<x%c4$hDG+(%78$#DfMWii8d;`_4uz(R&+a3fSHm
zWE?sR<B6-gsTd;6=#|kU-9xW>uG2i*F!pAGJYoLoC_AN=(+p7G;dsuSl6MN3HZr9o
z22hSvz&=P`{!<}L#HgI?WP(DPHmS4ACg+{qJtXGC<6UA(4i@BZf2&ZNW(o*SSki<W
zgR-wgKY?-hlMh@jS3Z3F$j6T#c)mX2{lM2ud`#5VxUY981+UUmjH#kdW%C>JbRmSD
z?lbp&g%G%1u3TpglBXy`J2AkjWlb9cY;@&<JDJ-}1QwsTr1{icC~Q8Ek`5|&2)T^E
zKg9BeKG?UF`?^wUVavN2Lq<}Z)I|*a{p+^zdVhuG!g9MZ&lhGd-NMKIiBH?gho>i^
z_ta8w4q}SbT4+rI)6puu6<%Jxu%tx4Xg;)-o!9k^&tE_D_VqK{+bhupre&tqK)hTu
zwC)`rCjvG+UINqNh_i%&J~rZHmAnr&%$HIbn>n@GX*gpj=?IX~pqX4cHIPS=ppGjI
zisdopB<Kx8VP(PN!Z1OtG)t{k>j3%*dpT76W;z+{@->g7zv1)i*hjlE270X;1ae^I
zI#Axx^<mkhYLUIc1&1SMZ<WX(dB^Mn$$?E63Ep!}fjOzwDz%{cPD5(vnt9c@zxsZz
zIjD}d_wUquDQI>OoUAT7MmO6s-p|Za<X{y#JSkzmLQ-)ns9e9dnZ33Ad(0s0+wb5U
zG3gKULr~2DV|WL{i6Tfh0aYJRMT`CJj<J)|nM>y|2E^g5t}ek01doH({+?R8SyQXH
z#ePq(&8UMGd}CstBaFuBHsCPf8g}nWza6M%2jb5`u%`}LVb>-Lhylu!4Qh9XGl*!s
z+dLczdb*U5XZ`Re_8WrYLTjyxXw6w`&TEd)YGwGHKoqqyBrcbQ>(evSbOEcXTyTVF
z@6xh19O1h;jI1bRnAd72>vp0Ro81J<k#LFNA~K4y?rezu9HP?MT8A+_jEHxM6fdMW
zGfE@H!B`wMH_9dvSS-iI$I0MxOO&TjlFh>9a^b_X*fCRzI2v_CrYYeJ{tQOfWI!<=
zuXkwf8<=wpX1fT2ocT_g>Nz9_hI8z#QtD2a5}&3;iG-!7!p|F{>d<Yq(8okf7p63w
z?|ILggkEXZW$xTb$!xXIYmo_Kt2)d_qdCjlaDtdsY!tZ2ldI&75EB3R<8S$RxvGQo
z5!V}g-g$j}#W}~6q;k&N#^ddc`#h_&?VZ<c<MY=qyu5y8dws*@%H%wkWuc7D@^s_*
z_Drvp`{R|o72-T`y#$tPAkIltT~`c7Qc>BgU=!REj7)(IG^}8DNB)TWW`B=kvl}H|
zmHrS<1Pq>)fYcit`Cb$rGqy3B4bzQ&JfN4V9Pe<f`M>R)|5ca5y6)-{YSE$^ozXRj
z>_}~Q&>u8uS7rJhv@XmrIY;z^#d{`D?r!ja28Wk8xF%7vx{s{2A4R_oL#yDv{fs_n
zrPE7?io{O#@HzvcW(Nm8H@sD@2t%CHK~Fp=or;T0F)+`w$@~RFiS=Xo$bA2#9TJ7P
z5MhW<)H@dT0<snNvhwJj5!(B}de~j(6>yWL$qsKDDq~PmG`Csfo)07_Y}R!g!RFI6
zDQeW<xG|7<Tkjh7`r~dDwUD4HPmB5RgnS>af6Mr>rw&6*DDM^IuA#0%fc&Vq@s0DX
z;iHEkow_JXUa*i5?_|ujZ@d24cbUGGqRicH%HWD<kU^^4d7hc3nQ5AIz?&ym8-?Dh
zZ5rh*w`Skx=txEv;b071ILN%ug}{<#uFErXx=1O^yY_%%@EoEJh}O+6JQ`6UD1NeA
zU`rp$i0~G9+C>5=rQ)J_>^RGIK+kEOXKt4ZQ-~yc*U@)Hv^vd%re$LCk!DJCy2u^e
zlwmN>3&X+0!g9Mxac|V+T9qnxo?uQ^6p!`6Ti(d6k#En$*4Z8#KX0#mefgRExa-hq
zvb;}GT~xOjP;G@RKMdxG^dhJzMNey<jG#$p%)%{ayiY8PY&>v|oHKjL5X8<I(bIOg
zEf>OooOjZq18>Wj`+BFfLh#Va&h~g?d%W>Add-LWV9lAw{f!hnJq4Qc)LL*c^1B~@
z!~gQ1{)OA+Mr)0K`{VC<`NLOpVu(z0BrVE`SJ&KC6oeqEwb87>gr*}HQNl#M!}7qr
zKlc|yvxT#Jbw{BhINrvFI$N8<gYf2|WfzW9dv>nrvnWvc%RzkT<L_W|(Wo%Y5_fdV
z8WwjQ#%b!;y|A!>CjD_ZLP$*7AM#T1jwJ>rf%CHab%V*l^L6I=x+t*=PhSh`Cd+~0
zoMB|&WBI)g*2k)Y(GHr!<KFe5C3xJ>(c?oTq=b)=5E5aUw0K)5<;EBj(3MG?W}-73
zFDB8;>QRCH^>gyaKp8oDdM~Hg;I&BbL8|Zq)d$fog*VBD_>#C>XNmSqhal10!GD4y
zzqs1vr~^)ym2>YG-4qkiIO-mWxoBQyYq?l9u5~<7X=scgn9o6U&O}M|grhXCkqSK;
zOfx6PB{`;OR81#n#+!XMPBV|wT&}ybGVSK^1W@df(xqF;dm-nlL@zJ)h&%A}-ick`
zJ(wX}2-0qNAL*@7TV`~1wtYaYa&(mmp}c+EF*~x&)vDR7rwf<m15dY4OmU&K%03FM
zXZAf)Lt~jP40q&3%f2J{grlGYvOx3^MyGdqM3q(uDcGDtdnAmG3o?2XjWpw?`Mi{|
z=MkebO9$D`U*?I>1JUWij0XAD0Vz&AKYieG)gf1UI|tU?{&&wCw4#JG2N-)!n0Vo9
zQAJHH?5*<k{*}-F{(H*X>zOyP%vaL1;QA3lG$?Jii_<HD)BK-;4@57>j0kBmcU334
zsv|N5=?4ab%jKeVVG8v5dZq7~<UKwHLl2zo_rbQUw3=a5_Vv!=@rv_<ZJ)F#onwib
znT>hkEjsqmnC6Lp{>Ojf$ES~kppNH<Pe1a9Km9*!+nqWhN6e7(2vQqx0KUNo88S9!
zVpy?@4o750WzfwGVVuKfoMu1_EcNezo8?3hdNB7}FB=&&-#+)vu)^&SfDW@zF%Uu&
z^u}QX`L}$2ole0U#&8f6440;BfanLB&buJvvfydey}L1L3atSekb);z_mpB_@^G6X
zKRjQ#PRckfV=(!FGci!F5^@{?C&x#X*AY5$_}WVuJ^VO~j~Gq@q>v)hJaM^Px!f+Q
zyax1MHEbRZg~J@9=A&)1>hiG5{e>}~W-EHpJ{*Z><5ZWXD2)QGv)e%LxX-<Vd5K&u
zlljefTs&Yo_Z=;@w{cc_hp?@;Mz5kPPElek9|BX<VEC{rXK>ChhOxr%(8l|VZC3e9
zEY}$3x8M6=@1QD`D9@s&YhYOf`=u%1qo<jz^wA<87Q@Pg=3GaDE+pR41*T4-)%P?h
z`e2ksC>17IJ;VWr;o^a3<5YnwzM|=Xz`ULMJ8SH?kkDT*mQvE@t7hw+;zCLbG0bKS
zBI;{tnX>PAuV5f=HWs~#(Y7oXK&%yo`5bFWm(sR9M2~d=-y5+Od@Y0!$t6SA+%ASh
zsd9<!OOp-#ur9n*XpK1~{D>lH_-1S``S?c*!#Z3D4$)~YMWH?^iN4GrOp$p`Bo|JG
z)w-=RBkY~l3irD6?|=GleEqjS;5H?{4YR)1!#BZ~>3y`aJ+(BZBco#sQV7Il0U)I-
zDJ`e>)RHEekJM4cKr*~O#7J(+c27ZGN@<>mKC-?&Sl87^EX`1i!Tt4>$NjE=BFkd;
z-jkf%7JLf0DbT&F{iFBz=9tq&ZdK-kF`B}@kekRue#E0R<H&{|?htQ!3dTe+W6Wc{
zA8Y<N6R^(tsBVu+zK_F7<;bt@=ihfg+|et0yOw%F>FP$RAvkiTKJqfYTkrpa&#&`7
znu94Xb-)bgW%qmlH8E)T7lYKFEil!^7(AZj;Y0FVg2!8aaT(EU3&gv<_{3~%8-gcq
zJM^aiT??%^*=&#XyYHr))J{C`f22%~ArX?`qUzPU`w*DYM4D&j>%z3m@`f>!w@yCI
z<XVT(8p(TH74*=Xs<~F{IWRmv&dX)BA-;AP?%iz8u=(D)S@Igze;iUl84AmMVTqF}
z;1I;A?*pkKE29<J`gREcyLCs^7VpTC=}#j``Pl&%b(xuJ+`P+SG}#~oqw021S)KRV
zEZ#0I%a|y2W7x%66HLzW!;<*)d?C$=aG8jUhNYzs#>jN1#IQQ_XM3+qhJ$?q$GqB6
z8?>6)9xMB{QFEo`%%~NYU>*VLz)gu3X2Ru4x;?=(OEvEhZfzD7e$+GA=D5^*FQTya
zcY7Ay`QEBsD{Ooq#R=y+y=9^w_*}{B#<uTx3a!ZTQ>-tk>AM@GWfmpa2kQ`&9(O3U
zYiXpZ6M_yCZ`*#`?~IWswGnD1?;D?AKXaXD;`B_eom@JPw-<bGEYk$dLxV8Qv|b34
z4Zf2cd;>B^Qp2WBuC~lQJ4#oe{BNfpSzJ_1?QqPou`O5bU%v42@|oIp?vJm0`EUO>
z_pe{6d#2&Yc$}YE!UskS&@y|s7}jd0iM`?ove!U=Ct=WNl#!X!6Sv!q+eIv-ns;8y
zN{EpqP0Zf2_&{%gr}@GXJfEIFa7{DLd*1xu%j;L(?yp=3d`UsjQ*U&qf;GfR@1qIj
zNMT}0&7w?|$aZFMv_h@fjPOK@-%2CoPWRP3(Zb0T3WrV1LF_n4DAD}K`XRE@98q=^
zg3T4s@_xZ+KOFxES{g=Y4qi?NE}C{?bVjhVuUkUmkUK(S;GDAJg<+4sk9f@|C&<Nn
zT9G)yJ4ZhPAw0O`nWsQtsO`@k*X2(?2gmcASb~0!%${UQ=f_}95$7fL>dqPFZ1xsa
zx`k|>L`nny_B0l@9D}K$z0+py74Or@7Y`wxie???7PDy8ulHUS9kzIU@C1tvbf`ja
z_R^<m;&#2Am^olraFhw%f#Y)o&5Tp%)GVaqsJyXiBAIzvH0onRg9ceGXFJHmsH!wQ
zd+)5v5u3<+M?6F^E}RSQ-OvAEE}34S>AM{Y-ZoH*ow1s{Lo)e3PCEuc2cCJEd4780
z<MWmIGLe=UcUdSnwp`im`_`u6eDLpwQ!tDidp7Sg4;#2OY(^r*?<Gwv^Nq`8Azp99
zrzf7CJ~B;{CD3V2^uyMA94xVZ3{)6H6yR3G+Hklt*7Wc&a@Nwzwn8qz&a4~+YZdnG
z!4Zd;;sxhqKjA~bjUc}$gX2mq#*Aqu&U_!Jr)g%HFHG}-^RktHJPPamL2I4M%NM?U
z{+Vl9a9w^jTh6SHhvqV);o4BAU^9U_2E@o@T|$U5=c_td^uPU)e=VxE^0%Gbe}@i+
zqq+}cXvqBK3)`0$`kHBFu+y358y`RXmXEhjbQjp$PA-{s-M$&HO3k0^PVkZ826a@;
zop>ip#CgKSB)veM1*ybH@R7V`2A$i78@FY$%N0iJT<3-B<*EVqzSBJr0+-v9E}A5#
zR_8s3_lz4r9jy7m+vAOn=f1DxxscpUbMW%^%Kh;sVo7PV;%UK?N5i|JtQM!kwl{?%
zH4Ek$3}V&$3c=Hdr;loD(~KOAyy9ZepJRK|TeipH#M}Epl2Ac)2-gk==;1sesDn6+
z7i($8HWJ-_h?Vd+eSX!Y@J<;N$6Y=&j^VmqgcB?+_rigTju<=CLF}E$Ip%;3d7`W&
zx`JAIA6!$U`9K{`L#*C;-3s?Dvzw|}o7`ZmUwnqa5K8e9c3Ov2x_zkoNGUN{gk|tv
zo4q!M({QHN##$<6uZZU5i6bKGWX=Ijh=CM6Y5IjRpZBC`hTdrV&M}mxJKShg-UpmD
zXId$^p?vEY<(gVbA?8eT;yIVv*>a)z_fUHnlwLU$n8_|{XHZMD0rfGEc24-VJ5t|p
ziX7}F#<Skmb^V>aS6ug|X>g_)8G85QJ#ZujJ%ih_aJgQH%S4X>XYg3J!E`lDQwNbl
zQAxb>wa5F#b@(lH%iO40aje&nxL$8uu2<6a#{Be=kY?!$_G(NaRon;ICx4`N+EtGs
zx@pN(-#@1Jb<JU8ZU?x(6XPP^C@%=8v@BkGD=;({2r-<*fEYbKJN8}hz>(P3nhpeS
z!|~{qg+0#-37VFRmbGsL-(=m<Dz7hJxIKN~)6*vwvtZcsPOGx~*!N6m?}1dg6}GYy
zm-$>1K6nU$5C$R13uv~1eAqX(nrYpP1&8Jhc&&@z<#M{OmApPE>jt&cy2H;iKm72I
zJU{)&5-)<GLf|bf7X6vPLDnGA91PSQ)=}k$lrH#`=s=hzF4r43l+E*&cXBB_>dK<y
zk8Rs{x=b)6Hf=*rvY$SD<a)i(UBgAka=Y;S@gvVybC(TCl<Xl9Zdo_>e8(XJywL~u
zw>R>K-;k)R`N6;c>EC$VAIg_j3kW^n`VpDw&N=EaFQ}!Nhg!EJHLZ;hJgqB9&N)wr
zycd(TTkz4r5E8EH-EpqzxDx^{7#fa&wBheSZw6zIYIB6|&&Bcg7&ISALn8uB{zJW3
z9<6tDW9d1MZBDOgIh{BH5l0&ug5f<QB!-v&dt>m(m9OhYv%|Op_fmP(N-@rU4A3P#
z{O+si02mrT1y!Kos7AfRC;6WM=4})3HBkKF^D`*9vG0XFtBnli%sQj$yU}RH6v)vt
z&(pc^k54em09U1!bQrE~0;<*qrR?Nwr|y}aWhCfu7EWJf2jMWWi93_M>?PnF=6`>&
z;hoqF!DD}hsK7@?u0mcUg+J#HW|uweuy5_%Ln)XqQ!q$lG)=8~>1<nO+lvB+JUFia
zkNI}xa=i$HwF}ETL3PzKl<PU@lI4hFXg#6-K5$l17K%A&DvDDU1j9MvG|9#y$t2E&
z$i6qjPQ}{jU6(-bf+(A1cN{9?gYgqB!c&lnvMY{ms+3G$zk=_$u`|sxqU&ej(rE&e
zJT@=0W>Xj8<T2_H*!R0Jp9V)Ja~stxzk?VpDKSlx(Qps#P9S+l>xHsE*wzPo-e~n|
z3@L+Co#XL%5QoeiTQ9ubU)Z*dPfs@$8X+1<We}z$n8y5wymu`7#LIS9m0gQqtRXRI
zZIwD2Yuk8vyl`J1!q!XWv1OLa6F>a$52Sd(yF{1*t{0Z&2HunANO!RAJEM0A<)$P{
zl5K;O7?&II=>xY9Kl15^AGr95ksI6f1Kaq_*S8m(&$KvE#~{a<y<~2;Cv{t1UHudj
zPft&zCE=pudcCsDi?SEy3n4|ey0h*(kM%))yVH7BSB@ZFYT@(OpLl)!%DS$+zI?^o
zY{-GxMgacsEt5_I58W=7S}OZq1pfucG)+PYqQ#lw@Bw@m46Tk>mv9K~>{dDFh@1$9
zH%jV}q}ZJVaYu(yhhVDSJ<X$H-S+q4f%Lb0ex1$5uA%bakZr5g#9`J3f;!|U?{u@C
zstrfOac^EiDOAPpeW)`S@W_R<uDIa1xQT&ds|J6MaZ(n&=ZuYf|EP8LLD|fT^Rr(c
z!x@gDL71#21`?793s!#co-4UXd^bh2>==%Y1V`OW`+}*653$JcQ370|j3!?e<d}KX
zuca#JyOu&P!kN2$hr>d;wubLwa2*kRX=p|<P|8-c!1NRo14lEHth{FhxAaaQX3cyI
zyT(zyY8ZI@UB^K<Y#Qc+V1#r&Gx6zbRXEaK%}Xjey0AJ4eWpnjPa6^~xo|SYy78o4
z&?=X@(-~D)tefuS-?Wv&x^Jxe&bsaFxe&}W%m*(U<z$0P=PZ+0o34~h$@1ZMIJ#MD
z*P<EFINp5~G?=^&P4D0Hs}Gn7nwUnl;H|@)<}1!mc)tLidA^X+%rqsv3!@6kLu#cN
zbhPJ8$(h_T%{5%~!hp@^#%LsX?^!Hq@a^r++sjuz{IFW_sCmQ`B7IWeRd=xM4{}y_
zcYpmtubF86P%3sZp2X}gEYr+w5ykb`OWhCN9xHn-1m{R0u<x1swzAfpm&Y5gZ(k{8
z*M-?SwaODLgh_Ll<fTOKj`{M$a=CJSx)RbvZi@NTIx1}2ikoiwHMb8uKRxsD!zV7w
zg=r|hcdiMydUDGYA1N--9DKgN5`*K1+wI)btu@M?@!p-Jk?S;(a3=&a9CYXDKEO0F
zmd5>UWvi7qCz6X&$8RgS$gp!U246KRkLUy5)Of=RMK`3-KiFr7jdiQzgshLwT(Tsi
zbT`?}hoB1^?<8XL-%$HUXZUXp@{Z{CCSY*G8%%zlfal>apCs#Ais<_7KcY(h2cKU@
zNzFq(C_xmsIC+}OrdoFAmujU`awGU8-nN%ZXKAok$=;)fRt9VB3YDo%2gHG=p+M?3
z2E*{rFa{|Y&ztWr1*h_v;l<x}mh|MzQ0=|Y77u<cl|eB^fZ{#NQPH-FfkaGcQf1(D
znK*NE^q#Zo{^j@Z(c`k1Ny8ZjdHCaQl+Iw+AyrLNa8OL07j8Js6Vv4)=F}L(feo97
zg?+qx<F_%`aqMn43{$q{2Wx5Ijv1>qP<g+YN>wg|^Zp=~foVCS_m**BuzV8*w)RdP
zoj7zTyI>3s7i`AV-bZC#r~T0uA4@z$v(@lH12Ch?OHEsao9{cTSw>7Df(^Z)+t523
ztYdSg(#!$2o?4b#s{)hEp9|bc+dWjt1KwNw1EK_Dj78!8muD`^mDUIA`oM>*WV580
zN@q-?!>auF{iVo4q0~ZY<<#yBk092Avda;bDvzKTy;(Cz=(gpZR;$=vAyT89bLVN|
zvEJG9O0SK5yQ?!iJ10R5bCkes3K}Laani3Lk%Cmgbu=FPN^qWSU0Jt>X<yXAy54cE
z<BELFa?a!})1UO+hgG@Baz`$g8}qzSJ0b?O!uoh4%^Pu=xIR7c9Ii}r;&yuy{1zgN
z!7X&2`pWPB_?7bZM!r0gd*^HKgwgph&%_^O{FX~$U02q1Gp{H_fvug|C5-camJ95M
zn+Pne^7Q(d|L6bw-|3%!QqFY>Ou&foG}zOj8)W<vES$9;b{;y&yLE5<8)hN{?fGtS
z+;C<^KQ!kIo@peKS#Xao$6|L1+;moSgO!dM6tNS&(P)LkG6NHZ_Cd}?nzF4@YiD#P
zH|zX&xD<LGlHhl7*h3iP1C#bfW^rm?MpxxGC5x!@+RmkAYIn5mRn2$o))d2Ma<aW}
ztx|KbVVc@RRf%G-{4$Qyo3%o7=EXCqN(W4Vs7|BqlFigQC~Z@<@4O9xEk{620NHi6
zD9;_LvSXboBZmuA2K|>FOhYf%$G|)V^5~2q^{7J@&W?ufmA0*{>y8_aa8Z0;NRBW^
zN$aB{r5T3VbPw#(IHd9x|JcWQ0H}KA94r~Hfziz82+>kS+tei#oU_0iXJI;)zu-LG
zb+Z-GMLAkC5gN=Pah(%BI-H9Xv#Q9$vDHqq&=2Q|7)>TzGeW$&y%254R%u<cgKC&{
zIRts7;N~P3+>*&G&@27^O5fk?u<xf($hm=Q3ghX0Fq&g@FY_?(?IIbZ6q%-YZn!up
z|JEgi)DS)Ma^vaxBe&&=Wm(9zF@|Se%$CF}PI8)(It14lUd$R<ZV&3VLD}hLmr0v*
z5LM~R9IwfBw~f;>A&fz?j=Go3wyu=R#?w60owN<T4`vtH{9x2fuREa^8j-RU%3djZ
z7OYF;$&1L4y!n2)AUwTQ>i%HeR=_hFl$-@+0=H@5H6{iUZdIH6OgQe3SKjsyq~r<3
zDIjMIe2mPCx~PxWCvKMoA003EJ8$>9f_Fm1b<Z>>=4s~Xa*<tk>*`i|;q~=}-~V6#
z%FloNH~MHy*Bd4LXMX;?FpY^29c$b9^Xq3GFR$z`Z|WG)Y~^jg6F+{!r<sqp8y{}!
zVuI79_TlmaELZ;gU;kAiDj!bYp<avpAIE?WWgnKQQFZy?0JFpH-G;p5aQ&gvz<Wcw
zrN+m|AZ-j<@bZK@){S+k9V;F==75eKJ$OY8lv*tQG@Ry=-8u7P6H%kqPA-Fz9W^V^
ztmeOi_fg4NlkA`e;+?fSXb>{G7HighY7u}M1Rd#5I9bCUH{&VF8l4maF*q5SIfcsm
z3bYJp*cUAlwEqGha}JVsCq}F@Q6_hN7_{V_wep45#E!Cm^$jb>SOL)yyV-MeMR~Sv
z9AM)Qx6#PGd_UB;s@mi5W>wI<yptH?9RB8<g+1nCRYzhvxZ+bHrO2|(Ec0?!!bkYL
zieOt<Rfj_}h^F>NY5MiS-p%1}<ovr`h663^Xp$kj6wDCc3<70o?#;e+7{26y=sc78
z_FSe37lKua<J=&}9N@SN4mz+p1WqO9X^%26%vxO(mM)iq^17W9M}N;d@)I0vU7gl}
z-^M7*5@!|~-B>Ci1a&9)?{o^MpvNk9@6G?pIR(LAZW3q3nV2TI#JWIh<r@gKiTPwk
zZ#l2jvddG+e7N?U+1Fiy!EyF)8blX4ycRP=tflC3@-S%By3@*bLTpE&cnI5$UR5Lm
zL05U(dAz-{#E4&J{AIT6iNeZYo&fL6QCBk@<yb42CHhEQWGdLo&h7flI=`GOt>7GW
z+j;%@XYQ9L;^<8Mf(w>U?mQuBhH<@Kd0J*gLoN%q+m&V)kL!+ko|va(mhl?+Pa%?A
z;5J?P@zZY@d1G52ylfjkfBuQ!<SM$%GkY)GU%&F@%U`&^e1TeIZ}E8Lx%{ZP;574m
zyYcbqiQDa&;9$G2{O;qA{J;LE|HAh2g_l45krWa!B(}V<m0f$RH;EN%RhD>qhoXmF
zf_-Ss*P^b#9p|FK17q}aox6is*`!?=Zk$E4;M@w{k%umEA3R~mMcMaeYfgnw;$V|G
zDta#|2Igs^=1iebi}Xvi{2jz=iWC$U<_yYk!DwDi^tN8Rf~ehbDGZwTCPeY)@YEa2
zV_4{Q3XaR1n4)DXj4q5dY9QRK#rE!?N~Uuh1ER*t55_JFQCoClA1}=Y)WbJci;{dl
zFf}Ch!NUkd=cz5Q<tnOR3e4gCadj^*CjC6`*TV7Qy;p3e3m)GG{lE;J4ywFoa<PGw
zZdB$aYCv|?1`09Yq9+jWT|O|QSR&fd`PikfS)}1(-Kl$#{j>#mXucpgN6BDf;=~Y`
z(}el2suea?nfTTEy@$PqbB^dd$@D2x3}UjxbQ&JYx2Bu6!V=Q#;B_`66TiC||N59^
zs9-q6YDcI~@&S@3dPi-A^>L@u2x(?G1<*5Knj_7>tLydp!?8H5P1g*pWHS}?vcotm
zC1TWkVVWk$5Mv^xh2SUTOPwS@bO~3o6);ZTOl=^quAO~*u<t8*+hnBJbf~QB#=d1-
zP=|07FTHnqDddtVmgc8ONJkl+oHzFUL2iZ8>bb@b^GRw|C0GpVsgSpo``cHd8^m&f
z<axU<Ou-YbSFzB%XU|$U>$b7y%x%7LTb`Jryp<l0J9*EfX|}<>`6Ts5%Ny(0&pbYS
zkS%lycuNtKcafNB%QA7jETVy@L=1sk8k=c@=6TYL$REiGVgx3ic)C83FCXb~;WAx$
z`OEM5{N*#(<;GK*8NKngJ$Sr);pd<K%>DIMiEP`(wmrD#!lxl+^oQFMx66gwGDB-j
z%Y`4V&-{FO0`GatJEQ)E-~weYXVPD7gL$%b+jV&#^UVVSlvT?(-IG;}Sr`;51_iGm
zHGe;#J#%Ir7RJ?vRL|}3tr~>M{NVz4bJ0ybFh*dE$b?Xglh%#?%or_O-Z}nmhInPK
z(4cv7_O;W|LJ6EU+072x(K@3Sy7vs<88My^jT`1EZ;qXF@?7V{6lLpBTIa1a?xmBv
z21Twln7rYR-@Bwb@0e30#CTZO3*+k&+qGsogCWDXVl07f`ON+hy*aPpWeKLEC~ReB
znkR1aAO`==+pmqj@Ae|zKXMq=f#jS7V!qqZ)fQE*jbhw?7n}mYgCj)M7u`B3-zvc?
z7^n44=@R-4#7@dN(^_X+H{RavtoN1C2J@7@(<BV5WR77|wCvnDMVw)wqw6voL86>v
zP6V7*8%NfGcV>^B)YXI#PGdE@zjf8P_e$xA3LU_dB6xA?mBU+1+$DX%VcGod3Z^1r
zEjxXOP#gG29Y=!KIJ?*b;}7Amnee2^lagg!d+Q*gbuV5+`w)ZKDFo{{6n+w7BF32%
zXM9WwY&T=WUeqzvT*>Xuy58B>H}-8MuXk$M1bO8`-Wz4_q=i$RF7GPShsbhR^wpAC
z%g&>A`gEskckcH$Ca>Zuw?ZvDqZ?5HXP2N;mcqKRg`i8JWya)5QzXGD?DQe9*PX|{
zvgVD}J5!puU9L<a8a2Z6GQZ$_B2F`Se;zK;L4B;ey?$l7T)2%7mhh)@paUUAE|-PN
zWfo%4465xrJ_aR@O(G6NduM{50%=MtX=YB73W#au`NIc(|NDRA@wl_kS6Z!n{`#41
zz4LfHG%Ff`$!k{BydzArx|Q3N+dQ)*!OSH)K0JNk!}^gJCVC(2Ez6GFGU!_y7V!ze
z>7qQ?fU*YhG<5OT*7OdI^$23RwNcG6Se>ZlO6lNto!;b)BtI|{i}q@0mM0yJv>Svd
z-H=Nv%8!|m8$}SgDcz$RG8MfjlxZ7{vj1)8{I6pl^)6*8NbKbu_(33uyY59fRi>$#
zxOGH-3}(_~H|c0HIg5Cv7`e=WWlqHCsl&0g##`y!_d*>8HQ<HM>)8D6y9s8HK4Ew<
zo4_masF`8k7?fNo+YY5!)hyn$Yo6u@(C9cq?1a+UOCvX!tE1Ph;E@y|bPd3N`2-Cr
z(GW9IinIVdygP-8R+dg^YVuCJ>*#G$o3a?OacWk4rVb9P<a;+?piC>B{jss%H_Bdg
zSH^h$d`C`iKTNqCM$tV<);L77JE{!Ek((?v`Vmto+lHp$o0reNSQx{qetV`?EgdW;
zgIP}}NbM^por_&wV*r-Jdw_$dsbSTitmG;P(hqz%Fk62P(!KdU9r2DSB~tLDbX*Gh
zGu!Bfl*!5j&uJ1jj7#3epgY-eUoIC>jh#0=OF6|~Qm7yALe8P1W#tGMR9MR~fON(h
zQTH6pcG`7mWomPbs#wx)@K<k@Rtx*%m51S-WymgL+aJ`L1uaBfw0$&rKD9)vk-md}
zOl+46`{k-A%~0qyv*wL;-)Mc<;Cv!FFDrK^*ln6;d<<grcUii3dsk)M^vr^1TI5sI
z+#pSg5Ot;*3T9g$hTz@W8)bB^Q(|7^RGfmkH9+<gn~I7lBu3D?x=a_%(3)jE;JHp$
zQkV?ZbNC_+K_hT``oPoEv#6<Ql3>+quk8-<P$u;4^2C4o-9OQC=I!fOYQC#r(~+RF
z%YxHoCxi0G59;SIKd|E*I3ZLGy@wy_um>kKv^PPA2mI9SOdVFFfjV{X8i(gy@ZR>E
zofQLevlL4kd`EMHjv+Bm3H(kphvC2F^Q$g}0jI6PF@X=srn1gjpkpJC1Gdpk(0w<o
z-3;|Cc`Z(f+cNQViOln4RqJ3Ij=c_Y8|<YD!*7A;2V<mDy?T6{dlz*L3;faT<+NGM
zKG-ASkYzy0vezW;&TK=+P!%N>ktf%|UIw)UZdXweLxOl#XWx7eI1f;zu(kJtZH#vc
zsC%c6h3JXVo{P1N&BD4BbaDvBhRsT}pjx$vz}hIeP;*g!b2x@zI)sn-zk}s+F4`gC
zoVSB+3>mZmw(!E))d5CFR0cG`W(QRk*U7BJOV@By9f6a>r75n5{*d>U`dc#ZIna|b
zG#DqlIbAL>MA@R`%JBY7Qo{{pXfKPg!u$yX9-QtNgN)?todkpST6>4M<~E&%gCa7B
zmY7=sJFJ3L$y$qot(pbm9Msd+3%wQSvJW|6f)gUMpHo9>W0`otWCOu&g8q`2J3*J!
zzHOBCK{dPct!7>yFT8#ItRb^?frod#x>e0CS}p9m*hK64AVujjYVWN3rb;s$ihuJP
zvSR6&pLyqPJ;H$(_pV~?vRp~aLY$`a_f6Ae?<AP*yg!+3>g4rz&hyM=nVF9Y*y-1;
zS%E-GW;egB0p3NXICGgV%+taYCqkV0asMrsrw{D86ZTHt8^no^A3pKn=>sW>)zrtJ
z*UadRy;t7$%$^I^X(2p)qz|}G7yk06f2Zu0D2e87e#ky2gmiwV)~nC}j{PN=6w8#r
z=4w=M=&%NFYp1hsy9zbyZjTG1wvHd&Y+Ed{%PXVA1r=efnZvNzcg!K+gJX)5L0!_D
z5MYY3ZTVY1ziy^r5CFEQ!Qq=JRlRL$=Yf7&L14*bE@&=Lh90!*gD@a6xWvGm5@AmG
z(V40|W~v2^uchEapbhY=c+AVvzJqPL4Y{M2!nn3UHacZ#%08nzMm197K?g*KUhNv%
z8Yz>Jv*h+MFx$`QhgD3W6Nl%PW>X@*f1LAjP;9a@aLq2op&WqV0;OiQJyS}>4_Pb^
zpQzPQyW0C)E6IEMaD>AL<#?v0F{;=$1a)1a1j=hi8@8D~!t2lHz%aGQrCRQc$u_EW
zaJ}ks^nQ|WjQx0h$ewtwx};le?EB8PZme%JK1>YnDZR0ks#Y3KMveZC{+=)4`@5xj
zZ#Ed#AXJf}^@N*(_jKR2DSRNHL9+Lq`MPku&UWy}Q^9*wbw21nhYR35W(J|dxk1@8
zEp_H8D?QHMVQz!!ysW=Lni5w9hgPfHd!e+Q`+ld_LMto#erI$GYj=YfJux_2XA(q4
zuf9wUyq<KxSW$!PePqIrFs~KCliU3Z>+OQ~h3Fk`Z?A0oPR^U4BOfIwbb%>OOd&E#
zqmC6fPq@C2OXY1{xn6H#7u7}>oyqw#H@lWX_SuHuU6;?gvgeGegSut>7)&X!_r`~R
z{2lY{ncH;b>3WgXMQMC^e&)l|g=I>(rg>2G`g@uOqUR}0e3&N|?<6{PLK-}QIbO;2
zm3zqyOK+=mwvusE!c8-NT6p^OK{*V`@wIpUum9s;S--6Kbm98X|H5y7`#<x$=O4J?
zadohb3L#K(=j)pJ`O9a<y3^MkKV$}&o;{EI8{59qA1kF~{FIoJ<rM_So_9h@Y?eal
z%xPfQyhVfy>3c#K65v~a1fZB=u>d_~ud=Cdx`ab?HXA#LE)E|&ee&jn+*O1aE72oQ
zq#hHqkAyyKeBp6!a7!~`c5wZBbPASgBy*>5;>&>}+ETF1F(5gru7h<XTBQ`h)Jk;X
z{-!8D{AHf$wI~X)5Ax{jO@|D%k(>=aU8nW$?Il_t+#fq3R(y&yry!TzUYd;H^!=lc
zM6a}};0FXbtQH%dx2DjQVJsHBw^rKYcaQfG*9OsP`0Rf1oSapI%`nFE^0NQGGYTw7
zZ0EeU!%`~A+8VAl28G~722}+$?9bLEd{|MR!5wJMP=_KYr>Z2Vce<N|g{8?a$6M8g
zb7mB#OD9HuGRmBnW2u7VH+d0Nv1DHF4`H<P!k9H=Jo2||t+b(PaG>{phKS&lLtu;E
zbp$;%t((K=s+4BtW9kH>&5V*+N=)-)ai9L=b)SmQ)3VztH|q)qgS{%DPcxcEsCq68
zXVhf0zI{BE^G3>|4M%R3y>8^ZvF&&6ub=6)FnXcZ4UfFMMn6Pw?=aZA7g0K`O;U>5
z+|CoCBQApMz?nrxr|moUmoN0Z5<=kX*RPb;iD_X@i$!}b&^_z=KzOr;`C1tYUhy50
zPbS*4N_S|Vwo+NwRe~(<*~`ZL@!;j<XZH1>kQ-;ALL1D}h1>0!<#Ofr^hEE2I7Mzx
zANlczAGj=w=1>0c{P!F_RLKP{2IolOt@YRqf(%aIdwtyLc_&)DZ!J~y<`8)L@WgL_
z{0+6Kfclq~`Op8$f8l@o)1Pq9H~!1N{1<+>+>}8c1Kos^r4-hE=gaF?9_xcIU%&8p
ztjx}nj4C|Oi*-#7mm<?NlhS18S2ZL-Of9jfRQ##h@26F{E(uvZ^j3*oj;{f+imF%f
zx|dyESScDqP~X*1A`||3#t5c+Qx8>}Vuxc$2JOqPK6tN;7?ptPo_{g>{O^5!)uj-<
zs#R4}#-6av>|#_B+i1$|ZRWS(oWx(l8o>i^9-S#AV;qIkf`J(DaX^T47s*u0FxkBu
zhf@h#{@x|E_0HRt2~h@iev0zB@9h9k)a7&?s`3Z&fc}oqRM8#jVh3~Sa6&v<^OOrM
zY6yCX6W27q_l<BuMxJAMbi{XdS@7mc-YYrlvWm{K9b&}A1BM%dg}Q#`;B8<#hS*v<
zUBv7;lB<lDU(I<mwPr(q-k~aM_p#QqF`yUC9F9xMTQK=IKhHY^A{8qJ6PDI_<emGz
z5%wT$$>{7=rg}I<gpQ`%0_Aawf=mlq@q<G+LCf0oxim`A^KPw4)jda+%gl0_<ux?v
zLO3M<&Z&ZP)&Yt*a15P!$7*{gx5|B8O{myeCh2N)UdiYt#tEM;=2=sX)-2c~x6D>n
z*7cS9+ZXOHpXu!o(lwHoiJXK??HgE7P?!tBl-<f1@gc~gi(&Q<;dJ5+HuGtvvfW=8
zb-_7!`T7bm@e~(=n@MpJuHMmw(2Ddyh`=yU6Hk{L&zC1ApH3pHG%muVYstK}7q)$8
zczAof@pyaT@%0Nm?*s<RJP{}P5Y5xf_4dp>3sOyS=IQAJx91zrx0_jI+r=1UH$D0w
zbeLmgF|R5g-XUl?CN%bK#SeKMy?y<{G*B>B={!Gw;OXNt&p&+R`Ff>pJ7p^Tusrjh
z{^ft-fBj$o-&}tDfgdj)V9S(sqhDsB10F`{<komwAN>5~Xa4f@Pvl(q^yw36y0C9|
zU8c^mAe;xYK><h3?HnY-K`oU7s`ta}=*`w;IN9HuIJG<3$iy_^e5Ct9apfdh$iUKI
zVjE0re>Pfmnh}}S#|4k`>0F;duin(v2OQq1d8by`bNSoO`M>E`*(aDCgwfUIvrz}G
zc6_v`#x^8IJ0@k$;gT&uGe@J#s$ou%%M`hoc;!&I5=7ywK=sHq)kGiE0;5xWIR}&g
zd#U{VdMBk|K{YeWWnxMG-CsX=N)w*woHk1z#uN)M5%Es;7YFlZ9Mz|DxL%>l0{k*f
zTxS`<ecLJY-dVRzaE=Wbi{%`YBEI_lzKZwU?Ob%|=RW2wMT%ft<50OCyl1)%fSoSN
zX8iJ8<VkZ_bQBY-h1wLX<E-p(hvjxt<<Q=Zw+_~m_YFcITD69Ce+)5BCR27Ex)Q<B
zrdBkXWIZY!yW}+e9yWk>v|P<rTEd!E8fD+v_o4{Srl9fG8fgp-WT$j?J_;fSl4y)B
zxD(5TL9ko^`Cw_V3^o(%gPb$#x`Xpfah4-%5$x$O)p|_y(pc|zvmhAkCG*(c*ta*z
zey8P)-imy{u*;jH<2XW(?7MkqhKUyL6q8~3mTNH%iej|6QMXJ96;k{9zztEpb19@a
zQHMR_4&FzW`9jVQ9`{YXiTUYFF*2u_$wz9=DqcpT^~!!wUURn0jaPEnd8~KJwlemO
zo_Aaya`83M)HKg0)iH#~?deH6n-CaPuxL!toMYtxTS@^ZU5qAx?{lSh>bP84*KOtV
z=b!odmp}6~U+_~ThQM$B;dlJg@BWbwPtQ`2$H>J;p3=gHAATTI%@Mb^mHsq{DKJ}R
zIAhSf?5zOcKp($j40Mdt5hs@0GxMY@jMgE<nLZrby0OeUDpQ&P(NXrA%~NX_D{0V%
zDB|5%n8D^zQ<yE4&=E$$wIrri8BBhnxWZm`GDZ6pZ5baVo*iyL?s#8G&SpMd!|ySS
zZtjrk6s#B(Rl2rL&9LUbhfd+-j~^pK6KQy*u({TuG(sB|u_=u7urwG0a+M*S4?!M2
zwqV22;`c6mAKAorhish3aX49G)Gr+WTJPL*Vd|Y_m?S+#(qvso6l<c?!Jy-UXpwP}
z%-n4n><V|*p?A<)kEqEhIF=N+E)$m}F;9UxeLuVzL-T;%2fbu+m2=^+gJ?I=+P)Q9
z?T{>OX-LUSu2NKbF>7R(D{Sz{i5RqRD%)d%%kghymWI<nIA;<VKitz@RaS%>#t`GQ
zbx=}WXZV1xp6H^ruA=oa1}O^jjluJDnfY|P@bS7ZPcy@Na#uT?tKpqhi}na3Iy#az
z6bjT@BNv0G>;f%S|G!eL(#8hg(VHu!x9>Yn@!4_bNpBz{xMXac(<Snt&GcsPqw0Gd
zo47S4q4j|ak!6}$(yW8HZnCuAH`e<X%KFAA8Q&$&9Oi??FmuEHZi{|ENQXsAoYHBH
zeq10&a$YrPu36s0UNYOZ<DH|1KzHS=1~rp8B2aZ19-Q=Ai%byh(m8(4Rx5d5@uSh`
zJoY>H`ztv=7`4zk7VNXp_Dx-)*@TR&EP#PLuLgs!FJB3lt8_k5P)hWH5(3Fz3;{Sg
zPUC-=iCk@!PzN<Q!qCN7jC|3NF6X|j{OM1BqTHT&{^0}j^_kmp<q{L=$QOy$Z6wF*
zb|=-w+v_WjQANd_#i*Lhcr&HM(*kevOz=*W;i7?a@PhI#PtUxr5AvRg^TOq_Fiq(c
zEB01!ID$*e$uYVjNR?>#p8n{PCrp9GPxus!Etdoj_CmnY<3@mH!Boe&a}4wwXNTjc
zq~TOJyH59_sk-23eNcud&L^3rLEi_Dd*StS{+rJEUr9+7JH%TxRN62#u0L4=VK~6k
z{a~IG^E}Cu!zSvrN{JgB1gj*Y4cf-&<l5NwooyH9>x1Z>!;9Y6Dmu7+0p$0tK`3rw
zLL#FnMlLDQ-5@tdY1$|)9E-Hdsd9=Efelq@-Y(R`zsH$_p{it2UxVR%t^IXN)(7Un
zZu5~k&aPqKGbJmKel%eAnpq%~cJJW23HNyK5Dpke%H_%-xC-Wv@52$`J<L21j=8jt
z(G<&&rpV~?8Mfo?LXgFDKPCEy>auqQhk!hr+yIvXw`Jzj)0H2euY9;&Nb^D&KrO`D
zbcy4sb*CTdYU@1YDI}}sYKd;8h*WSG@71bd^~u`?{rp{-z)of6$;dilK10-c2^P`m
zgLOx(3V*f3)){OyY!ja7#gO+^9FwTH99AM#?zVZQl^xd`(K)03&gNEmK%=4^ij(0i
z#9i8zX-X{f!m^02d8C(>E+OQ;7r{uo=26?e^Ll^7d(U;b>RCpu*KOaKuZ`hE5k7>8
zrBs73)uDN_tQba^C~XkCC-=(R{e`bDpQ-x;hs^^^qn8X-2jAh6dk#LW$k|O@Jl$@5
zefi2i<ekwiXCns2GM`>sjBeD<VI6}+caBB6Bd_d?YSimGI`_w&b$wXkpZ3#YYmh7-
zrXFBKPfw25w((#8^asYC`SV}?%)Mu(rz^|lqD#St)82bbj;H0q$L9|$)6Cjmd3(DX
zt$G-w2NlPs=VyNR!w+}{d)|0F?v$dWyyx4*peuN(ZPeE3g;S3e-LSMa3D0nm=#z?S
zKS*IBx1v2}^zVMm46F$5G<*}BRBOl8#&Vf)$#JMsJ6+DGwNty<v>OcJYM|U3<?ZhP
zANivs?W!gP6$m%0H1qKH%I?Jzog=v>oXzQAYPI89R5k7bK1HRqxyEQvOJ}Q%)$9u<
z7wKuxePc8!p~qpq^UK4yi*;0i9s+&}#ET7X8l)cCN>BiHP$w`LW|0s%ERF-_L8H=>
z)?`&2oidA4IHEh16!zU`K(`Cwu}NVnOY252jk*`wfbMj#<yu*K!uf~`F#N!!V{kkO
zqY?cvB$m0`7=vgbK?l|^*mE$=QBjg}WLhTXX(Hl85^#zf98tmJ_25Sz1@!2mkVf&U
zo5g50bCc}=!IOgL={ED@$0vUC@y6wH!Fk8t24#m(I<$5gf#E#OAQJBmI}&GuY_lI}
zjZy`dh{dddnGcYXD4WrkJJQ1(gUM#JPR$g~41tJ5$U(H>qk=B%V#7IEsfQppSU02%
z7?zCG%TCQJeZGna;-OV}XSI5G)k^?(jmv4gS21HW^y^24*aEYQI9HgHbS~bYGpWa6
z&yB}g#CWNd*nrmTfUitGnIWV^n}g0{y)#{}>`RAPm(o^N?vJ~=uQr&xCzeLdJGndv
zQ(*EJ+0kqd^5a2y+$m*K=0sO0RxL%`_oQrcqU>5qtxy`Vd+MHfdwJpKpZ?5`|L|LW
z{_-=|`6|eBPK=znGI++(ho_FgW>y}1DfH1v1g4neoi@}>Z(HH({lT`bq%nBDUGT}%
zoTHh5D@0Feh2#@87k+wu@Z~RGdHnQ++AGNiZeb?6B${&^)TmiN=z%s5Vu(!X!tKKc
zUOs;&CeQ8Z15=9B0rS^a+Hm}*fBMh-_VSSwBR{`==BL*O(K((Lx#PA{c(lOCovn)v
zMn7oF4o83yNlvk@F(g5PXdg5N>$cH)V{x<TU>ae1&#X~#1AYwl+Hs>2X9sbJNP@SO
zhZX6kxS!gJ8FT(^=lrkBQ!uE)8Kc45q+QVTkQx=f!H;!-r<iadl4~=#Nx1gGL5BD2
zg5CyOsg%~`dg`#MSS=GFNb%XeS2PbW#jdKDaTGU0>cY;u#qu!}8mKaMX3;5ce^0U@
zrH15=rZjm~$_D`N9L|rwg0a3$I7>d@oT|rRX<sc@%9a^b;q!nflODlEUiyIG70e!-
zF+;kbjU)Hqut*4|u=Ybvn8C>|V49Pn7432vN9SObbqG;|=itG01=$$4-q{e_*&IQ`
zLOtl<MIXLiF7maxJ+aJ_o`0$M+RrJq^bwjfz&Nc$f@N2qZ%%`YgASmM(`oP@ljD+-
zTK6dBbaT!Gi9;69;XNlJ#&|-O2SrswL-SuU)i1p3N%V-SvBWyiXmzk}54M<at|<j?
zG;-0U;Bdw?!I(12#MOj-m>N1@Mwx~W2V=DO6HaSQgoB(Vev@TJrDlm1w{2As5cQsQ
z`HNvya%Edrw&@MWBIq&;=ialZtHJC>9@|EEyx~$JO$!dqqr^6gv?ifvukWHxb^h90
z2cxE$A|3Bo_l@n#PqgH@{pl0a(}f>CedNY74`0}6@cc~lis^0M@wKdceSM>r!gZN-
z$#4E-8<kSH-yiIcJJiNyT4<N6goBnH9aA!9VTPb{<9>hT<?Cm9srUh(K7J%P&KXt+
zf#d~K4}(FUo-W*OSB8`D?$eLI<+faLqx1aq%rebHBLvjm2qUmuu6+3Lk^TNcFrTF0
zg5H1Y-0u&9lXXq$&D8qcieT++f6Pb^sIjPEbPPt-ndpTn>x}A3UNl#jlQo!-$S|}(
zoE%eh%AgON{!}<;>+gLq{y+NsO7Elojx*ap(H-IR-7I0f%@63JenzoG297c6*{8Q!
zR7vWj!lUTSG8(xxwRx5?dybRGkiMZc>i)jDkAFAp&yuZjGg_l4Z`C-%y4Z{^h^k{Z
z#_5MHv5CbhT6Cnhs5>3N<|AS`z}40F?}xrpa1KH|FP&N{dCTNAQ?@KRD+eBIhE`E7
zuW6cyX|is_m_I`_($k_M#*-5rLtvgGB`20?qTAp&nH4mKZ`-+RKci(}M@L2%b#zHi
zu%}_2x+HaGHVf#7C`u5`Jd391Lo}+X7*7M)_=X`SQ0)@%sE~MvbX;(_cFeyGVsTzB
zo<0!VU`at0S&|41yR;Z+h??SQv*LmdD;K^IOu1v8r9hv9L2@L~y+2E1sg1_yRro%R
zvO4ujDM_obKj_tRdA)>4F-ln1aQ427SaL=&?ud2b-4VF{e$fu2LHDY;%E4F~y;541
z;+kO{>bqsS2DOd4@9g^<W9;+}dD~5;k8J3js?63a>%Otna#|?#F<6!b=jD{zb5>_;
z2Ua+S%0op&FyY)VrWbo>yFVcQ%pd>oUx~4^K7J%F3)kC6{^dXa7d|Xk@+h)jm;!sr
ze0h7}{`$fkBI}1wJl&p1(KE-1(K?TH=l=T2{&+yCL>zNWEKkp*%gy2~MNu6)*7a<d
zEbO$ohH)A|(vS$vYcA*fV4osso(O5?!>1ql@yFjW4Os7Ydh1*+7w(TP;)Cx%rP5BP
zU+tA(<N;}?Zr6?AC-z;opsfw^UTCJkj^4{~u@Cmvxxc-!uN%W0hDYnCP0l!(lgb|N
zbV?cE8e?;$E~aK|&_@u>+{-h|rD%dd{|(gqx3vnt&N_RboY!v;YIFMY*s!l_&d?bo
z=g4i~YvV}R!a9oRrB7I<q)_r0G-UG50^^m9aAXN{Fk$+@w?_Mgb+$Z(GLFHX5?jjb
zdty(4QLHU3S@{L^z4GNa@Wvo2cabT2u9q3_gnwp+wSlT?+i2{0XWch3V7_5N9hcI<
z+7XSkQ};~Xb`3X&bpYOTnG@HG2DOI@G$k4P9d4?*)YGN$Jlt&nKD;W@Bj2~xE}6S2
zZx5p`hdU2=X-9&=aO%X4O8TI29&-iI)%*VMrM?a!m>HrCF&U>z>u^dB3(iqTH@Kx+
z#n~9$<CFgWV6K{q8Puf^?S01pQ<DA3<vKIZiR#Q-%etg*MQcZYd;|)e3h?8?Kc3y-
zy;5`PLc{re-;0J;3fVtsy%AD66~GLbrkN0vT`W=vo1lW-IRA14)+Knl_o54-_m1zf
zpl&AG>%9}Bo=3Bpj`ND0Oi@n4y%lP&)S~Y<+Mq1kS=i*9IS3n6zSj#`NRfkct`dU+
zY>ww4D-#_|DJ3q;1x)Z*FO`&%ITFa%KXhO(<d+xLpZ?6(x<X2{-~2QA)9={scP`h+
zV%|b-FN9JVd*-dYQS;8WZ(Ns!Wtyp_@bdL5fBMUx`1<uTd3#7j>^x7`8$ASFeA5MQ
z5H_t7W8}JAV90S{eXOE+TkUwfe$jMAfSBi*+wID{%)|kg<-#q^Ofm8K@yfI;ygu%v
zX(r9H()6-m!o3!z5^+w2TmYuIalJk>$Az*KQi|NyJN-!h6K<>w=MHp7*(=+c$$J$V
zGR*EK9QOFyXWkvvzI*qd%8-UP%tJF#7aU_8-c#y+Tk-gz*xtY8^D7n2f=l{%57ZdF
zpD;!W!C)L!pT2c^F7yF0i2_Gg<+Dtaf<Js<Ns%~3O7EgrIaBA_4R=Vu8AG8nJ`mb}
zaO;^WlP;6BbuE~`9D9)sL2a_zaEQa}h7Bo4w@NZFPm$YoA($V|W6kVF=Mqb#p0Sa&
z!8ca|3U;r;742{n8>EX-WAw}3Ev_%gep-~lN#<X0s(N!NCxfY3r{nM>s{m@P<Xl;|
z%wyk;-jQ%!{w(4*+W~2ZqIZt&qY(fG2Y+8-Kku0br!%UFYGj@<PS_>bfL!)`7QSPZ
zu{Vn7=#J1W4)B}*tM;>akHLCoPhsEFdFx7|<|E6Z*h|;D`3?<&8S&|2P<No>`2CF>
zT>>24EQryedDsbyg<zp6Dq2P}!@>sMXF^JJ6B+rCNFm~xZN1EdkO;Ay>$R6c-F7jI
zd^wxigD`Mddel;=)-}kUBd9h^gIdM7iXFU@?ZN1k+7%sJOT~Lxe*_=FM3G}i#8`|%
z+KRG)b>?(R$+lPCzt6Y-8_+O6csa{fK~~c&r&J#_A1D|!HZXc~hHX_B+zRXcl`mgE
zGx<b}j!_2nz7a|%I>+7{>*EzhXT3*@)-2rLUijlre`0;ynef~~vbhh`ZD(ArioKmC
zyYLQo-o|0ket+Eg`u3H_zA4_dFD$Bn^uE1fIAe^m%t#9t(;Us03(p@u@|U--xEPpk
z7e0LY#Hfw8yz*FA_AN8}gp2(&B<wh*X~D%rDVuT@Mm6!Lph3lDTCun9#s-s6Go@e<
zlL;4_$O6HNx8MLL&o%U4>DA6;2HjXqeaNbRsH%W|__6&RE`?I{(AyZNU%nCENaMY#
zA9TD+l!G$qnyUL+ajoNr<GL(7FAJA>W}fE54usqql>=^g$AwQ{klq+(#P(O8er$ZT
z?R(C6ns7%O*d6pgcP<V~$}$CPijFC%gPD>l@R%lE?i+cR2^!x}0so2yg^BsR4w>Lp
zK_0$GwKfgeS|<igg1oi49@J^rKh7!QFvLVnk-W)||6pW=5Gl2=?wR|xvDYe<EV9@b
z&8*CehQ3F|@9^{>=&8B)XK~nHADF(BmoQ>tiiyUNdm*E_Xy3s(Pp<i7y{OXErNuGM
zH^VQ@PtA6OZ>rdDU?G!n2|tgz7@QL(?;W@n@V=j&ONd_ZMtCnE+&7n+I|p4hbUti1
zzI|&$N^4nbH1e*4vpBJxk|k@^nn{x`yb!}_DIMd4+to3_kfy*pyL4?I9MoR>o4W-T
z;Uz9}URl$ZWr1`3LEm>U$P!DxI~ZHhC&tj_dzkYb>%W_rvE;&Z<Zd)+Jd}#)$t9oW
zeldvE^j(>AK#6>-78fP8c}MFi;6l_sYPI0<&ieMs=Py4oEl(`BC;Tp(iM;K!T*&**
zPhWqAyzz9siS_3_xfI^sUikMv{*m>u^26-|%hN}2P--UE&SjGO!89c%AK1O;>BAHM
z_NPDE91*iwfAI49mA9uSZp+Nl6RCGxvsl*FNFnkMKm0&!gC$5Q|N8Qkm;`lQmJ7Xh
z9{0O^i+ZQ#!sGs+<-(E+w~x;}U2k0CjXBNu(WrUWQ*?uE-%duI_(jgVs=6GLi@FFm
zY32|Odd}r=E-Qjsqs?)(J8($kTKgclNb@$>HlgSNd5<<u2HfBB`PF#}?VCX=9F)zW
zl&yhqB#uB7C6Kjln7HyTnY`l?J<s#RGKqntO4zY?$6IaWYE;AO-vvrhte;vNdHnVF
z^Uj7V=B1N!rd1_-)iVoO1DP0%eky%7!$%xy-+dT_0CUu>8@(7e#cuKBJ#-C@Q=ESB
zsAZlt`ACS3=qAdVrI9Fuk~^DiyctKY5)gGipli^ES${Ch8;C}&o|=NFtPDaFW2+Ah
zvuZI5-fBKb4qVp^X7tWzhn7&6S#^qqJX%(U26Ce{;gMMS+HgcCi;3PFk-!vZoKG4F
z+rYFA^Urb7P*7XEqY#6qHdW`fRBG^~pmh|SW+QzFf9*JO7c*)mlETE4W}=Vhr3vO<
z>0pF`3$o@o=2^}L%7T+M6b_~(aK5V=j~=SGu1wY<-in>ps%8hR<NajurYc5y7tL_E
z4$d=8GgDloV8=;qYR;ec=uZMh@Q$7WBO-yNdEF2L)!O|uMPQgXursEBubHiNN-nIW
zLXjU<Etzgs3awXK^QH_J)p(e2t_`VoLn&NwXP2ir8zMfmjoNnl5G(IU9&-*GF3Zt0
zw@hy3cm*0#hbWQYa8L(5NNcgKcl?*12-D1bxscGmUpG0%(i-{oh3)=@vEP`dnG};~
zsc&z%ZR5+I|43aQFd7?$R9<1e@jv%L8J#c%K7_=6o%oR-x%}fRuNLsLOfyeukua=_
zk=UBFK*4fD9(iYbJV;e@#gU8v(;Kf}ztGo>=Xs%gz4P)`NK+)bM(mxs?-J;_K)PNS
z-m|0ylIBBcn#r{S>Ll)u2c_D`*GGbn#5mJ>V6Sq1@OaIul6gQI<mjC@f`Rhp3GUdk
zvM3pz(L6nOT(71#vgE{GnSv~V4!z9(==1BmkA^vBe$xUgHO;VRXIKrK@PIqoQ^QX&
zVVZZ2G)<&wlG)S1AS1QF`<{)ORY@%or>KBz24yeg{rhM+(Gm?;9+;)TkaMV+!HtgT
z4I(Zja}9L1$cKraf?=-?qQ~`t58Wt_Iz*#gCcSm(C%$_~a0EJ>^AG?NPo0?pW6;t>
zh*Gz<KG<qyj+5xGF`SW@Zg{E>bd+7N?^*fIy|Zs(t>l`it#IVVI(xOpA?bs0nR^rC
zY8+VmdrST9G|O4F9}DcL5(CG;${?p<_kwfA)-ongQ0v!wP^HqLSJ`%_pbJl`+HvYO
za;fZ9aGBnRGYr_)nL8L9HY~Ia6P$93kBkC3f*JLbyhBLV*?J5<@b7}IlC!e2duQKw
z_I+2S<;9WL;9CqR=e)VY%6ZqWDckIgPS?R7y$=3y1Pz({u0<?r&gM=?Ef{-hi5^B{
zI1NL~>e#B{6g9wh5;}I}4)of{Bb&OtnqzKOp0=%%!O%PEQr1V51zoB#q{ooKVlm3I
zH!Aj))uohHSo6wbdr;2K!aC_QI7xy%T8BNey}fczKl8F&Nuv{eQfFJ_r?lR`QudWC
zs5A>eS$E3gLEBbbtGxXD7uwr{ueU3=Papa8FaL#X1>zJZQjECZ`0?_@zkK>F;fE)}
zW#V@qe!vY4=oNJ;_|`kovbR$C`R$Fj^}%CXxyDIVwik<N+q2EP2K%<teFb-6j7r|K
z*~aMdjUjSdp7?lu!U1L9SnpqTo)`nFmcqU%^W^#Yh6DC}qtwhd)IibMYmst)bb9l0
z`1LxIyfHG51jaK!Yp9?)*n=MM)-5{oSxSf8`G4^Fb(GX^&psR8`S03v_s&4HR(x-T
z-hvJ{gA6o`)qgh3=&E+sJu3lf&r}=C9Bo&uN+KIb`RT^@Kj4wb=A#`<&M~El5&KEl
zm5;gcoI~WuH$8^KV+gNK)1(2Z9f);QYq@28JH^O#nnj2Fg&nO!EC)G)HrrrQZ6~YH
zx_yts{y{HPPO7?mJTWNlGJ20QGq7yoI(4)Ys@e9!Rw}txVszXt7p{w_a=mNt-%4fp
zLHAPa9$qtw-2CR^f|K?cg2kp*Q8dlBMDN$RZ!QAp&Y6Y7@t+6f(%x~Y^d@IqAA<&M
zc7P0IQL^rZ$6na2YLAB?(%CRPcvbpc&sXFIn;{J(1T?j^EJlul)|t<cT?EI5K^J3h
zjj|Uu=ZQ%NX5K5kDPKHBPl$<WPU=2}`RSV%e$G3k7N#gm4L7{$FCZpQ46qK(T9Wsw
zstq%C-V6yFs4^d-;RY-aUbCDy$(^%R@z#gY3;oalHrS>+)YHc$S9>RC9mKNGsAgAU
z#)~;q^3IX2SDW&Noi%{1YpC5!sP<SNl(K^}ZO@R=oSu``{^&p(w2~?78(R!~nLNGj
zc%Nvk@_4-Q`OBYKA77Q)*DI~;vZL8|T&qL|w<+@S@{q-J&v;kyPapZRz4ANvPfV7C
z==vbF&Zqgp^4pJudEyo$ama(pcUAA(rb9G@z{|Sv=g(jHw?F-f@yPTvYZl}d5lD*E
zzuUf@=Um<%yggo79}m+yDL(W%U%AdVF4M&d6?lF9%>A+IJTq3AAitefCO`<pJc((z
z<|1uEii{yQ)KW~V;}MCWD^uH9E`v<?G(`5px`^)IXQV@@1+}SA{Cgy?eRufT5cS*t
zEzDmFyeX<n0TW|TyH$UqTO0nKRFu08x%;~Ddf&J|c1mfIc@Ih7=|kbw2i#Eq8b8u-
z5*vI6=6NPsb>F%w($*S239`b8s&-H*d$U74c)XWCj0;B51G6dcOu;i-Aj$Ye%8vI=
z7Q03-YgLY=r@Gc|_k*#;=y=yP+<sbEE?S&vnq(9096k<89_){e$F{TPdSdi-$Qs-M
zmu2DO^Oeh#<h*Jdw0K7^%Cu-+*!;mT@y0fiZ@$?u#BgH#2OfS}F_;y`F+}h7@9Q|=
zC@=R&?@hPj&kTv9BdJwRwg)YwSrwN{rB#XPTzFT?<CTn-%!SpPjY{xAs$-`NfZ(hH
zsRy%25SIQdd1|#F2vG-^c;S^s4OJSMT}6qYp-G2wj7(FaJEvUlE-g&0g>8LkfVW)E
z=h#}Iw~8C;R$CPWladOSJ_dBHZxv96#aMfPgu>S$MZfcr;hp7#_mk2)j9pf$)V1A)
z{kas0&%9Y-7Y#HuE3!fSkl5{<`E+FihoFnN**j}pK1+}F&Ym|ww%H>^dQgG2%z+f#
zc^t_2veI*Bf4~csvabXqQSAFl-tTxzCG5S@s>Epy?H5PakQJDsml<ctJYK%=WqnmT
zBI4!u-c(qQj;~@iPRq>oD(mhT%tFH~-}@N6+#h`X`jtQY;lJ@=z7oga)^GIv!5Eck
zinLnz^7Bu`HmFfEtCz2z1<A}}2*zaWH>4#HN+!sz^g-<!`WqinbnYlGsCK#VIz5Of
z!m8phBq0Z-7VQIVBSEc48Dw)!0EU8wKtpGRV@boKivt3c?mM-7kFNapKEFC-2RMgM
z=iu^aS-dIfhD2y?==Zd72%+S~k|qtgYO8EH)1%tqQagJQlu&C|wOEa3Jzz9{4)0nS
z-+m)IB(2(Z@uXJHQzA}5;UmuDS{H5_)OGYO^R{9`IGn?m2ENdZ;nJE0-_UeO_@EZ~
z8=Hi09t|mKs8hQ*)ZJvyM}9yGy6h(NPP)z$&$la2w}qIJIgB0|(5mVl`_9+L%6;7^
z<~oX92(@=oOx!Lr&zFVkWdZNWySXp6ARLu<tL=u=xFJQOziN?*Zu<QYUk<V5oXHFu
zp0=S`5mi*wW*39;=L8?2?^v3_R>n}%JJU1Yvs`0Kkw9+@TUSmOjJ~4=-*I$;3FPK!
zB8EsbjMn4itlBLn!@^Iz??fvo@YZ!WcP6bJ6>jZ~eOJY%L3%tzbJLUzvKP~YAy?7i
z2jSsqEfa$b)>_MFfgus44S(Sz9QY8<mj7L#FZ&m1e>@}T-R9P`FN)kS&kog_<wffZ
zr-N~?vJ0uLF&XrB5FVnb!X@U^!1*L^bj`Vrxy(`Ajov6Zled+!?~G>l9wE@(z$d{Q
z-ZVUTG;guwKIM-WdcHI2%33myZR5!GXuUAJm$9FXQ0r(WgjX?epb<Vgd{@T9IM$an
zq$R)~Uq4fR`pU&enB+QqM8yXCxms=PTPELEw)=z6KmWudd|~?Me__olqv`_dHS_lJ
znLqy9?+LXtO^NGuA(x#Posl3CE<&eKd!rgFu~Kny;`ZSq1K9rZ6YE}xQJtDBXB<<j
z$_}?ZQClO|Lf;3u%A7DBo?-j~q94cbSn$0E%Xik^=r&7q-4jPOIPmY#DKw!2wu1E~
zz4OEr&duY!GkLdV%{FD0MlXA(6oWC2lF<m&?zb*0XIqP4>?T0GcBh!iQxMUAX+&qT
z@(j_2jH>V(K17mv<4ZPfW>uK>!qmc}pJ%<e<E>kU>M-$%Y;WCwRvOip^JN_$4K?*5
z59{J;!lWUNC@BPbD@3P(@^l1vxc6+<Lx8vKnRT}gVapU#luDbUY=ALfo+mEzOiZG1
zhL~*a48rUgzENDki@<VHO^*e6bPBll8H4w7gKS;)+!}(#^MUuSMGTO;<yMStj-kSF
zj|(AJ#Zjh|-s#0dESi1a+4HW7sv1jb=z#Ib<5N7{YL!UlAOz-RmLES#o~lFsIfo;v
zV44oDg43nk`cVOQZNgFv?lVZQ4a+qUTy)_4$;c4uGUSFi+Q3!FMn4Q8kYfCXW*g70
zX?)Xlo%lZQ45%{`!#1?D|15p5m%_Ri)|?p#uA49Kvl|!11}dW|E28OQ9IX+8_S<Ei
zrT^%9Z+k6l+seMJl)O=jSZ&@cqqPJK<}<kL_~&5i4VtYLbEop0mf#E(Am03(OIg*q
z^@bY_*VWZT95py<E#$nxaO`{L{<yPkE43E(-nqoU^|Ih+3F><7<b9`lhi`DNh5zmU
z`v36L|My?HmdrhGIGd5=eJ6!TO0pI|F1G7D^L)GUczvamov$xn7^Tqi#`buy-5-=~
zH~mkh^~|!&EOVmv#^rh;T`uNypuJMMD8|e6%BSZiwsj?!%J2X1dqN7#^Q4RJI6u9`
zyEg0o-v$KE_3jPz@O`|WLuMYxFu2-Tw=jlzM0A7RQ67v_%vSpk$T|Pf=hs<h$M7B9
zuXl-#=F3cqrjiBWvB6P><3M?ib|&Gt1dB&ZmP*xB;qTqz51aH1ql1e{ShZE;hYEC5
z|81k7&C*rH?yj@$l{iNx7f3^llX%1oHuE1mSV_mCMmTkAozT6?kme3!Aj5Sv|LAd8
zTln4AFg(^-r!+Xs0JE8rNe|(1&7NZn;5iMyx&vG!kILKn;Prmz{@8iBZ?rB~*(HT@
zNUkqE2q_SzK)OWYCE`L7`zSXV{EcR+WPM>D;wdm|V?fj(%LU9}O^0%)GMt8|y0ERD
zGmoHVmwvOfse|Tp3{acR4kX4l|D!RC^<{&yc0{5ce)C!}4Y6Sc=t4)HR8uk*jj_Dk
za7M}0+Bm45;%RqWZ&3AaWjkQhSp<L8))kz>z<2Y$au#G{LO$tpI??M;ZoTSu?NP>p
zxWk)FzrGC309q4$(>jFKwbGoayrq?~M7Aicfu2<}LqJ(Ox1j_soVv1$7kx3ML|i6%
zZ}{enmMa_f+y~FWky23i>W0A^c5u+BLwcKBG7*K!nERk|)a?@O;}B9+g}5Zdwqnqo
ziB5xlpF=d_VIo$XaPZozb%Mi+q)rP#P?s|*d4+v552!<pj9Gi28O_tvGj8v^e*Hqu
zFqMg)-`@DIfBt`HpZ_fK!n(8VJ2e-!Z9Dg&A1Z)a@02ZbOA7=Iw<98<^nuy*%N$Va
z(Fvw>j*udkbm95akK8`}hUceGESD==-pN^eCxw7-vby=FfBFYrzWyw4DPS*hlI`6f
zohH3hEtWmeO)=gF)eU0k_y)ZyM_%L}+sDFh^^CPpwbr_37!K&BQ4q!FfBgAXmqJPr
zZ^NBK4H|>kC<-y@wh^_NoOjUOsU7gVizsH}86N@Z#l&6JTHklC9fjeDmZe!ltIhb{
z-+uJ#TkC9nX8=CJv_wMGP_7}iOB?2|GY0s;aDg^FqZLvZM8uLgSXL5@31VnXAuS^V
z4Lb2$xY&UZ0>*@iGO+81Ept69IK~i{df&KjGilptN$F*s!Xp=6){XnN^Y&PI>>7M|
zvt3R}rglDS&XFchT9Tl-prkk-9BC4yF~<8qS??j7P!P^E0QMda2GX!fu^27U;bqw8
zk9mTI_{~^EtyNPlOBhxxma~t3F4}i8RyV4wS~GVrgS@v)=tIP2gEJ!!blwGxhvkTh
zRU<GQ7mO3tG@i2zDWQ8U2S}@|@!IgoV3q@~9Ygbw!O0fW9L9ok=t|6U80EEVDBBF;
zbdGJ?C}mft=EK>w9d;hAH^}?hor(4=b3Q$P8?xKara^E!!-Ecsj$|ErYka3hALID@
z4h_h`oC|h=S=Z18wGP3nKs3s(%KpPQ%Ir_Vpm|SqoI*wVp}e-4A_|4_UReq!1=d5a
znVc)ty47JcxME}m4aqxu-p;sCy6mSrLs6V>Z)~qG_<!uI>rT!pxL&^e3BP50g_;X3
z?`+#E_xqck2PkC^FP*{T&bqGBNIg9(%`eGfI~O!;?u9WbrFK%h*o@EekZ%`0Jbz@G
zE<_hdaUz6dj<Ak71Q^QBo>Jm+U0CmLx)5un)7Wxm-B#+6kz(W%tG86-RH(TtCDM7k
z>x|@$@+~rq9}a~p=$$kUYPDd?Wm2BKcYjCr(YNuB2d3_90DK&v{aa4#81Ln@v2n!e
z2oSubs$uTDrsy*lv3KhQvqXTC)$`qFaI#rI!=1PL3J%u25~j!q2M<{cgxR|@eM^BD
zM%|hM^PEh5IL<SF^v<4jI8Qw?Jd_6Sf2%#82)=i~6P(d<Tem@>U67{0+*@I-m8n!(
zmL>CEJ8$botqtcRX<oS6Ghe2_ya;;J;7CtIcH)B(^iBqi5C?9FR0Qk9mZW7U&8h}d
zE@KA>&R`9zrpgezX;+BPoMt#dj}FF!5cX__X%;_vGGSW9xxqYZE)b(X|6T7ATaB&?
zE;z65zz7Pu3=E1fnDN_3-TPn^3HH1@tcI<FF_>g9tP@jrSX+75*_eLd2nv$*$b0uv
zij>olcn;c2DGWSwIO0TyV8MROPxL(YeN}g8B!J|lx|X4$)|4BjVvbP*OS5&6)2rS=
zv?8ol!EllHz31G4=N`}deJ?t9%xu~@=vUP*-A1pKnj0l6x1v<d5Bw;Qjz80%f7Y?5
zkNk_hWJ>QO(g}wNt~;tigP1+i;Be5f8@Y5^Ra-TZV+8&C;IPg}Io%A?(Ri8OiKV#n
z^S}RrFaPvEbN%7Bbm!RCJ1;-~nQ1S0V=U&!gSWRY?E76IHNjDi?Bu=h_WDY$85aVV
z<-%p2xn8g2yz%;Y<)_bo;pO!!rDl>3_AIl_7d@STi!;~D6Q4f*$fu8=7#^<IMfTRE
zGxEk5jjxZd3{d3ovPhiR2VY)ZjMm&Z!<cM8y7SO+xFN)1)WH$2Ho{PMY^0UW5W0d2
z*J3URpm}YY_^VmrfA90FFNKqedO$^8mIXeT+#F*cjPH-iV+x^sVe|JeN@(e_%g(JC
zPNr3Op!M#gIlAwv#=rV*#uCEGdSTo29k$YBd^JU@^6V0`D92%y^PC*Z5-mh$0fups
zJEIOt&H4_#QPzqJ;KKaNZ$a;e(H}5mPar<L28RlB4-zW{4cWHVXj@^gs}`@550LCb
zF0?tp7(_3g^?~3V>1oXruC+DgNkobBT!PI(?w!yfR;iqE2OJn=FJg=$br1av9j{dz
zxVBCR4eyh&v);!$0Y?w&<Xn62FjQe4u!w02f>m2qg&#xMLr37Oi&I6c+yyQ4X0xYn
z)|7JuFZ+qsHDqgT*rT@qo(8@-K&gSH$v)%=`$_SzvbF{1Elavq3#!rjJIn~bWn_VJ
zl}}u~AH!ShyEMC`#OY5KN!h#f3d)xjW!8Idp;U*HrtCzbv7bN{_2?=vMi;zw?8n~U
zFIhpHjSwV;ELo~$iQG)|Rx+*b48yVy(nTqDS%z?~XApc=d$SHf^8oBp>W=R&B<yhS
zqtUu27=&o|wa5U){tkBxw?$GJtqKho`dNc>8k~=gPo5DRFF*e~zyJ4t1qu$&{`HOf
z<10V^{@=)tofrc7@y52j@OJ-<i`FS*<vP@yd3(ItFt>AwGuP{l+tU-=5ep9RQAUqP
za8RE$qi|`$%@c9Ca(Vi|hmW5~X`-}ZBD_I!(nrmg3tKL1wQ`##a|4(-;#o_T7*+dD
zE#5KHaiC6*fa`WS3^@d!(Cp}dTnvWo$1!B$*V<t(FRhmge}_&%MsM%)goCQM?`Gx|
z<N(=Q*R5q3ulAY_!sRjx-*WbYp#$f!ZSR9rYqNdujOZYG*>59phX!K57&Pk}3ga2V
zUPp(r7vbPT9y~tEV@Kc128Uk5nmHw&t~1v~QJ$^IBA``OjXFp;A(!Ah(Hph$yYHeI
zgy`|6<={Xi%<(b>rprv464f~#yKJZTTJXVfyDVH{lD@+Q#*nZ|+fC2i<A^RuM^S}O
zwx)|QMsqtIqRQm})g2LxQkoAt@H{pv@<+O!D$8tXX!Qu+0is!FFKJ?)G)S$jYhGg2
zVsEO@ox{Rz8gMYwbm>=y{;`9mjYGVD(Du@qoYsd8hg-8l|9zGbC&y@x^6ljUU>E&S
zJ2>k)Xet=m;LKo%DVUMiu!uuWUindkv{ujeEV~04kSSzJ>p@AhDr|<qPEyr(<!LxU
zsWC><WoDY=={?ntTRyyZ0-(gaX`YEOF`WAbS{<elc^|3=6O(m=C_r#)sf9gnY}+ca
zXbkjTaR&7SCzw`E%zLM&hiCNZV}|84CurSN-OZEieV_1{ooK@W?sz8>rYbjFU2TZ!
z%BLw=zV$F`_&_neMRZPVtRaeb-!}5+KlA55{42c`mM~Md2mkHg{u{sl*MBANJN~jT
zTBYWd*T>Jq$@T(?ZUgPN_0DVF2{G~E`oz=IjqB}-kAdI)<3ID*?`)3;uYdW>y4^{!
zQftPCiOcQE?Z@Bn^qb%E>36^5`Sy{vE4|ITKyzY&^?j%G#@n`XUw59(>@xa5MVg>v
zEo*NCgGQaPIBf>hhRp#XCKIx`VBIhbL;J>1SXOC`VN!rnD=E3}I?4ah=hs~dr4f>+
z8y!;E@t`@~dNDM-^EN+tx1*NZI7~6Jq{uSM4K$`i>+sk!_ig9CuH;-fTt=Pu=2h<S
zG2)zb4pLQHh4$^wIXLGQKD-YyOU~5R2sJQYW~Mab!5Br#Ex?P>GfxxO%gnOOIuui3
z-PA1{`2@i^=F7yC5`Fyj?dvcOxm9qj)zc}o+hG(_WLjq8GUI}xGkp`AM9FUAGtVJe
z73XM86)4@E-ODK5s&jd=3mO{s475ALBxAJ$mu|hI;15;w7<6=O$|FH+-!hNK%DQgE
zq2bvwB`(v%Wtvz_jh=mDGmiGrEqDgaDrm<z9Xi3;oPlqu$M-7pu%wPLC{fOC)7Isu
zB)HEYGv|oN6X;-39h)n-L!GWFJw~hW)tlo2gcN0uks@abnjuu816RF|BR#Mm2jB6Q
zjacARSVXi@>o)`1QYv+?Y}*D-9$(XC;_`GMO^SXzXoTK~9Nxb#Z0D?-G1I?RyTX7*
zKL@70f5Qy2S&bNWvUX-b$(e0^5L04WW}DxLf}EnK_jlIOF~0-BH%F+A_s8t;y)bdT
zs~Cvpnrkx>cXYZ=-H|#ZjdMgV^-NP>vUJ86f&_t8*%j);bji$fIy<^`U3t{c{3-q4
z?EB6VC)T%D{`ddm|K#(Z|3a=6-zw8Q;b{cdaQ06tmAYl}`ry9Z`Qz80`0+RYz_cg{
z@3PEXuTR7n*z?NgKmWxJzF}g;%<_EY!*71WKmPN7;vfFuANlczA9<Ru?2naPGc*}*
zKGr*R+u8P2`UnQYM=;Y(#^83nn)9j(&S?@G(VIQRz=||y=uGb{n7`aN-E|+zl*aZi
z409cj>w>quhv5GnGX>*Aw*wwJHUySQ>l{wweRTm{okH-A(Hl!}+@{2w^pbr@?~#nP
zpTjO}OE_zlOR4x&i7AP#)Bj_>?Kqk)^UFO3rkt3jVQ|*28D5|RG$p&KgVM2r;tgX|
z@7V=Sv&8sr<$PJF?R|K9eDB&Ct~(9z^5K)OZi-``at`5~DT-j4l41Kg=;|0mKfamQ
zc;hBZ$THA(JZ!ipA1utL+4t4Y7Gfl($mS4jb4cZ>!E81B_Pxp+14dH^_}DW0x^qnv
zVaT{`nG;KlOu;j}$Mq(k6~mmn4bs~|UpzVzJMbFd+tKQBvbzpza=cttLwvXL7G)1g
zk?LI1W*zj02yhrYb>P~v$q<C)E#N#FwjZLW4|NhQhSO{hXO(vExLO$u3#fk|6ayB4
zs>53_nH1~Q1;(}s0%@WK?t5k{h56}1xF(hl7cNf=*ZCsau_<1I&CR@b)`dvuriy+L
zBjo1D20OUoAWb4?<U(FoS{6lEMM0ENpwlJYTV>x@oNL6Wz2_Xvb6}86VE8^)vd4NV
z!VMU-tPQmS(V}2MT~6?kZ*$!JP?1o(3^8?J1)D2IrfE7ag}m>0KftRynWw}YJ(nff
zTuB4((kicCzUqu=jnAJy^Y-U2s?0}Y3Jy0pCXeDUU1gAIy-}+2C+=^r{ORX^=lbK1
zxM+Ldb;5wzN1m@YKK}4YbZCP^y?4aR%>4AsZ$AEpe|-AHZCbSEE;H`#DffaKgL%?^
z$@xKWx((|97ZS^IrI*ad=MOwRJ>lG6bh%ecH}qcc*^pK#M2u9?ab1GY-tE6zSBo12
zr@az<5S5-`7s7EF{59nMAANqMA)dbUIK*n-R1`;5d`xAP7={@=4JOkkJYB9lEmtnH
zTB(*fTj+U8%oX}5cKbC|G)Z~Bqf3G761e*P52y;~y-IO-4vUF_dk?Pors_;d>0`||
zP4|<k9R&kT8a_KQpQ01}Z<;0+`#GlTLayIG?AAN?J<Hr`tK7GpEgOy17&eF9vo|bR
zmz};#_qs@%*?Y7otBb?$r(;Z?D$HLh<Zuouy^|G$<6w-)0(k^a8J#U>`L!e+3e7}&
zAy_{*I-?Ep-q^QB&Cn{iq>42LVcaoAS!uYz5&w4#sC9hkBU+`=8SV0llcQBArCM8i
zoJG<V1ZUa+Q&$J`HOe`&8C2=xq2%m1H>@)|U@E%|gO@lc9^X$hfZzjDiUx0a6F5a&
zKWvK)tMC03^?i%m$0^~Xw|@e0QjnB)o=~(_agI_&t-SA>3Wemk-mW}9J#oD*EX#%3
zIx!d4ZRI<y2;hR8M-SLG!~mP}5)ON0UkaH<H!BfAjGdT7mPqM<Bs2tWtx|Te!$8`b
z6e44iaI{;&b+{5A^Amk9y;UCrX-a5#Ix$VkIlo*MQJl?`Q4nMnTy-?}!I8YTuo=DI
zBM?djmOFpGV31hv%?@RGN$C|09&ZosZ+G5a@2roVeXq2tJ)dGACWrUB6nf9pmdRyh
zUEg^5`kA+v7uNfoT*OukE)kq#GC|}rE3@J7QJb$b(@pW6A3yxS<PvRbxFmhi+x>+<
z{_*!_>%A~f39c7@`0#-xC7S6>s``3rt4uMG(!{o}(3<ot294@nIAaBx0MBFRm-bSr
zN3m%ou|d@*fo4GIq9KR>PBHz<&#w&G>2gqXheh!>W@EN^k3;#uhcz7aEHQGKCze^T
zOR~zmBQ{5f68=oHqyhV$&F>zljZx)2<PPS_FU*}zx}#yjonz1|Tz|9yQH#4DxF&}~
zrxw}&kG+3ek}TWOG@);v)>2h7caMmD%u{6*U0oE_f;7?s+|XCF2+z{OJ-2<q3-BI&
z!E*pNK!FH#qKe8Y^00TD+|5*Vt$F0)A9Ja>@64_pC2Rx{(48g`>0xfBy4GBCj`5HG
zFob_FycE=JIAe)8qK~X8ab8v~Z?M4;{b%QIAA{>&G~T!sUT-_Mt+E?@<R_cMxkK`v
zVua{vRk{zC!cigjE~^a*pw#iWCToDbh|8@9ONUGf$rO{*LfUM{ZP=V!F{Vt%`=o0b
zf>CI7(Dx~RDf((v*C6<rXr_*e`2f$Ez?j&$D6Ydqf%7HvIuQoqDUx$BR@qE1)0$aQ
zU|n?3&8SU``Lj=)-&wx$BC|wg9*-lJ!Py{vR;A3IVo04JJBL6Fk9$yClc$qa?9S{k
z<~&2N*C3!l@wzUYE@#%$syu@MW#pp*_90MPW2*%WLR;3vx-6E#AVa(mx(cLLnHOP<
zS(oddK56qzmKzbA)@Ex>bk8xxp5{xm2F~Zi=`5=b;ndP!^bWr3Gj^vLk~%smktf~A
zxnbg`aSViglvpky{m^<!ynFY=yZ7%*NC?fz9IuVWm;ydGGtu?@#pzXM*Es|gT_H~L
z$F!P<ZR@7S*7cm!{IVCyU4qOhlYtre>hP}PLdUtr7<X#j$o0lGzwq+<k=w^tUVi+6
zPuCwQ{TJXowHC%`#Ha||J?}UN$%40*6jg8;LNX&i@7s+}pMK<?Z}Ky05$_|;2TrFG
zIM`dI1cy7HIXyk`aoZ_w;9|6W^szS)%<&nRQK-G^BTkF>3B`r^{PqbeYWE#O2lVjT
zmyRgm|G?`%+WW}f?U#`LvGLV0y=gQ{csiYVT31`tViPqwb?9qW^Goow6ea1eYLO@v
zQh?^^74geU`h1??#=)MM`L~m&eVD4bN4q-*G8j`tLzzaYjlD`wkZYw43B6KM2#B!F
z<*3Tnm_ENG9dcDN)&2F#$CoRww>x{T%zS0guyRR>^XWufR}B~!qs>hi!M?X)o1Gm%
zAqGKs+eUK=S|3i8o3pCoV6+VjlbO%9dy7{i13SaUv}qVc(}6Pxs8t(i^?`E%?;_4-
zJ%Uw4?QOO(&B6LiTl2PXqZ#eZFqxQwK_(IJ)x8Bcs>f*o;GK$r1Kx;&#O8qzq7EtF
zY!)?VDsI9abMjda%}g-s^d$xrDIqcPFpCa56ztmd_DGHkAqetV7DbY##F`dxfnLp9
z#yOi;C{wuAPF%cwresQYRl&_5fEZ;t{%~EjnViiUg3U>S&6}`m-<xJDwN_&!X>cBb
zXFWwur*w2mhZ)YiF5Yw)t;)XL{&%<@Gc9av?&IA-IU9axqkKAXzMLL*;08}h`3xa?
zmc<}gb1CRU47#)?(h|+=kb`aI%?75_#u|-jXP>c@%5YxJ#VU}s{;kWJE&{V^4pD+P
zqZiPr<ht|v@{#@Z$`9ZEhVQ=rOK!jVCG~P)^vZ32rHw+#8?9zSKp86`kj9|pjoday
zd~YrcgVGDR7jkP#pG*tuauU537$MTv#M3w5@QYvlvAVOE_`m<nUsHzd_OlbyWdGq7
zGX`Tcd-mMH4ATsz9xSdH&C{%z-p7q&E`jsMf%_kN{WL|hyrYjy1wR97>;_<eewbAa
z%@fY+%G3G8<+Perwknj;VK?s{FihPC+Fl8vP^nC7<B`L0lF~bT0UvQJU{yqK2kpUY
z02aL27TaIVCDWT0>k>2^&b4sQ1@Eg3Z4u@OQeRJa%7#3jpG-QDbFQ@9xZN|?+s17x
z<kF0JBCh^vT{xdt;i=1lv%_+)^$@doJJ3Q(tmiZ9x)OY#)k;`b^TR1esJD-Suq1|;
zjrB0De%&i=FO1UgecBJmObpkJ{t2_Sa$Om=V?5>qitv2PjG1WuLm1_QCl2xVHAL2Z
z2u7=X+ZlCq2w_-hvjy$}k7ZVy57QjXDGdy>0{144a`vUQX)EC&<l^&LK<ib^8y5)C
z2AxYH_(&c781$CXrM!LvkAh|+OVZzoMz8he)Y^DJKa(NkoET!O504xU<Bw11>@C}8
zzGrZU*2nvB43<N1jG;K%=#%s>PBEO_D2S@)VOe$X>M+tFD%%Yi>}5L)RGifV9@E`5
z+WFgAia~@pJ!=H(#-bwy!I;(H$LYSCl9clp>#o||CZtF@B~ItlVZ~m|kS{HfxGdCm
z*XOqaspib<%MBL;qjO}L^jaiBoY$?Z)1A4;&gLQBsUuPcu;E(g<<k%R&9{HSZ-4kD
zsXNjd`Sj@{KYsk4@4x>ow89d+VuO8Pb&hhs@|z$2hV`_t#)aA%-@p9GkFT#(H~9Kz
zKjTk-`KK&#<=gKU;`e_|Pl<Q$zv7R6^{4#$>t9Qk6IXuz$Nx|M<^T0xNF{DeHkX2&
zh^@f#77@IjPOPWJI!tZ&(}v%tMcslepJqe*);poOqtpL=UOzccVT^}&w+$mz00961
zNkl<Z&}cL0-MR#OtHZoo)LsRWlMS?tJmiL0>Y&oewX*j{b&j|sMjY%`@pNay=04~a
zDJX~o4vyq}9hK1Mf2^c7Zy21l{y4dZ7Rm1EgLGaDx5qfr&e~Tka_tGxD(}XyaGv%h
z&1*<dRn1tfl+IonxprfQs0B}}1WO@4Oy5Ef{jM|_pUv6Ba$0$MzHmOBHIU1h;Re_1
znd|M!2$A)2;qv^%>2#Lj*++$M%mG%>0V_63U|Gbt@#ZAz?cc$&Fc=Rbx6u?SSci2D
zs%UXTP=S+$vfw2&2&2uLa?E#n*WnUFptOE$IFqr{YE?xw{aT_`j!~N;1|J3sWp@rz
zT3FIT2=Sl{Gal=5&aRJ2wTd&Z>|-&iQdAdMth-QWVxdetD%Cj;3c(msmKHW4H2X&T
zy*8Y5a85~o)(3l5enIPvk*zyz=4$FV;Ki9n=WX^xw2fjeaQj@&^O>WmGxe~a9YgfQ
zBrBR_S*J}Z(+IDRc`#!_s5c@wCV69+Rl0abGip_vlX4$?l+I$b&hYvf(}sP%=fhgv
zIpPvoR?X*h&~;iZ(skfiQ>5)?LEk#X7(gkm_A^v4RELtw=Cx;-zDD*P&D@yXaa{)?
z$m!JuN6j0zmruNW{FWa>BHRPpcIUfq|B~;&`y0mSw!S>hjP+`{@x!-&#S%Q1(-Yfv
z<2T>_hL_hL`S8uJ_}R~Y&AaFKgb?|vf5TVb{DSlQcYOHjE1uR1p}{B>*9One&qSMV
z&HLsJ?sYc*3SPRP2|G_0<6NcF=xsnU$lUqQ1dR?stzv`y0pO#F!dov2Kx?Ay9YaRv
zm=?g!4N{7nQsR6*v7A?XSj_~g4_VBPP8q_^gL@1Bo<L#0n1bPu^uegc&z@K33qw5T
z(6BWG8+coL-MejiXh`Kc5wEnZEN96vI_skLJev8G_a4`seH3Sge6!uogchI$bu(|@
z2Tmqm^dpF6-xVN(4UmQ#5Crwa6o^XzvzhQNK$W+Sj0VNdIGq;ORXGU&F>b6&<nnZ8
zbjRiCiRY(h&Zm=bdK0{~>JI2d`>G=)Z4`uX(AvD0i>3q3RJz1cwaHYlk4~+FeV6)m
zuY>5i2IJlzxTtp?7ot@;0k7uS^d+WS@WCvyr4F5`SPrRz+g#g|E*c!Isl!QVK<Pa!
zVFs-zS?Vok-}W%OC))6O=|>FWguZlS89~iZZ_$oD-Sq?R()afRQo&>4P$EQ#gyG&C
z28vyb-mwH{Y?5Kh<Vr3(RL@?DW<x`>A;LVI2ea{LZb<p>1Gu7?RLgG&-cg$1H0Q;Z
zi`G4@r^tE^tmnjXUWi_WfNW0&#DMQG2uD|6O<y$O%lSyZDfx7<x<q%%My~8V6I~R9
z$itXTV>%}luv0f4QZNeWiPhBG&Lh{^p$$FKS-gWL{mq1alcn**6qFy}11?1HRRjYk
z*sq#MG1tP{8|$)IhZk9rW+5Z0z*q;=?S(p?p(pP9Cw~0&8@B5UtyX%HMyfc+>YZ}m
z_dBoO|CT+xz^H_F=fh>;XJ38I^Kv1yfg6n_EL@fgm*vD_n|bfeW?PNu46zh|E)CTf
zBIp>6J9v*rV#Ty1&Zm`5BkU3=9OBI}jS4Xmi=fygB{9%`zcb+{&J;BLvg2T?mfwi8
z?0X*7%pe6Xif0O(%{s!_P&c>AUW#avG^^4Zr7IF#QIVRQQ)kuz2910^Nq5e1Hs@8@
zUAW^R6zfF?S|3#cs3ysMwa7frAUc$9+%#bG!+c1(EGy;$pB_vadW|pMhn!%ubK*s3
z8L(vwx=03LO;4DV7D8NT4r;H5r;YDoo6Mo5FnJB=$ABeC>+#{;zzxs4rwi{cPb{lg
zQ_jucpmF$5%>k4irWyG+#YWC5Exl?}#`Rtbu6J_P0R82@^M0=^tK|zPtNg6fP_^Uf
zetH(!89s~>Y{r9AL^o^FHV3TMF@|L!SPpcIiBdO%lGJG|Q3JdsEjnZ-<TfFX{!rde
zKcLpcLTF7o48?>7dQ+f~!?G8`quQU3>x6prZgrKq3AlPGCK|MqIIChq5=*Pt;XCI9
z(l3ZQx^H*3vWX?;63#`2mwMeMnn#6w9Ns~UEIniH0U6nWpB5ViuYq)mEa%9&svB}H
z+AL5uD-$wuqfr|4G9HHl^0%7TU61rQ3N0~-)#>DF3_&{-XOHNdHmEURI8oYDTDUx|
zTns+dt1F2yEr>viB21fRiy>-`g@<)Lv78nSPu(z*g$^ODHDz72Q5mBVf)ga=9XSh9
zTP_Q64V*8DcTb7lZshw9<l%Vz_>q>c1gC6<V0JhXnaPy0w?f%&z^=-@cLHMQ1s51q
zc0Ba1I9nWL-`UF@&KIFPxpDjPN7}YamlOoCR)--F&E^Y6N=bor5foUhz?+G8Dn#6Y
zWVAjzIB?#i1pN<?r+^t)xw)B)aaaSZ5*^S*yL8ryuTjDw$9Onm+QYxs%HA3TC*e=4
zG@4a1%{F5JH8O5A*^K+bTM4z@OAMr#9)mhlQ$I2fls+X4TbahanO9~Xllf0z7<azW
zTeZkMPfN%E&nybZql`LUbI^(X4rN1YK#SA7WkiQtBEi0tW8K~iLABw<B^nKO6-_oq
zI}{u#tt>IoVq*!3_fOAURtW-UwO4vRv>whR^uvKUmf7ZZH223)O5QfAw0e`xajBKv
za)Mv)o2i*kgb@_3<AONqzT<qLRb?-v#gfMQBl)Y^{6OklE$Cw)9leBsr<J(q5R^5s
z7&pnNswqYmpA^G+z+igK)H^UGeM~m6>k7mfPL*VFv>F+UY_grl#B<urTJQL#j;2<t
zlx6f!AI?O8mLY<k2j|_x4*j$$Mm9M7FhQQ#!Kk&LHMJnV;-o5`))``!s9Oq)4eFC1
ztV5mZMDc7*p80W&INJ!idr%i;H)7Yx1W~~bNM)!y2|2rBIHzhKn-@J;T)JQCkOmJC
zot`T!AU@0~snbECyp5`BoW#O9Ush6Dk50>Z!H^o9i{Rk;BdcKyS+B3>#By0N`W3D2
zTyJ)S^-k-7YBMO=`J0PD;K;^UPWn0CLs~rN#gn|AzZBuL2+r-T(p)oNHk(l}MYN6K
z)G6k=v-Lr48;vg4VK=Dv!nR*2dE<4v@;Bf8hIi|NE0u45_?GX#{VgSD&He3gcL#JU
z-?Yedc+m4P=a|FIwJ1B9>Z=vPnyCo(>ny7nXfFH#;G@yadey7xp4-V$J@jc!3xlo)
zJOxfmJowUK+6IhHu7%PY-D^RXKBx{;@ICYklNvay5W}iIdvD%3mY|c=IY+gPYz%*R
zgp4uh)d>r4ozY}m=ltB~A2tbXG~@cp0|x@NdFp6VF}pWX^fxaJnhPZhADxyD(;_*#
z!Mo=(KmYkx{Nk72@bvsl8IJqD)4WdoXs(LOy4WHX1vaGx@5R)SSC8WSmLRd+X+5#5
zs~*56+X=awx`PCObE>{mWT4Fr<X?ny48*|7IctF3^#CaEQiJE5Gh5zSn{2CFR}rwJ
zL_qe}wJK6{kZjS~L+`NRcad#@*#;*QkDZptyLXA__kpJ;&-pY6v9hkRt$sc|7+~w_
zAy$J<=AkBggziN}Ov$k>qU-hENU4#MnGezkrn?An5L+QF>VkT$j8>((T^b=MGd#wE
z_q#0^Rj)n<k`Dx<xV9#S*1kt>*F?WMaDnINz~$^%*G5W3hg6eeW^0Y$WYD)J9SX(k
z;l7*)mnYA=XUC|XC3ZofA+Ri}mWkacy2FZ(#<F&nr4f^e8!<o#*6~;;=|jc0MjMUL
zJ0Zwc!J(?S)lLtYlq$<oS(3VoZ8y}wm{4>J3wpD;%?@KYOJ1}&kvDbrvM5`?Rd-mJ
zw5INUuZ><htM`PkSQqJ~&QF14hoozbvTvMEi6uqKCd0~<<Slks=6IuA`-jQjvLtl^
z&a<YK6y%LIyS*hXgc$WaObthC+6TMLD7y=^E=JS6WL~e|bIU*Hv;>^*yxf1_<IDHl
zub=qt(?|a8zkSCtGL*uPKmN$=`Voe8%*K3UsE})2#hf3Sg|yzOqp|o1OpTak3Wd&n
z07E1RiR9%7F^nVC{}3TN!^9Ml!`nt}m4F*I92@Eg@T!cW*TJwZ$;3p0cuFZp?(kj}
zlbM@R*Y;eG4Qav}80I$FKcl<P+`+wfq@dV7RX5_HE9wuc!gbuw;Y=9gV6I4XV?Iu{
zxs?3y-VebU|IKs*)*-d}Isa(%PR%>5Wb&?q!H+?pvn0<q?=JkOfAS~%{Xh8=zWVw-
zK1H^@@M*xksBMhVxa_Bs4eMrrk$JFJ4aiGVCzDK=7flEw12WkQ)KZBd9<WddfwanV
zsdwn(VMzdD&`5YBZHFJQrl^N*al|kPG+b{c?h-aVjBXo@O@s*<eQ<)%Y6YXrjV4pF
zX`Su8=d`SZ4$n^u|Mma!U+`c3Fa8qX4ciF=9siDM#~t7FyZ_17&pf~1dceH>`SW-D
zZ9n7h`u@k;`TU*DpYx^P%w+Mr`NqHF=RN-Wf9KD7^JD(jE91U+XMfAj|J~R4-G4h~
z_=Eq`|K$V!tN-So$e^u@*&+gg*4a#86g;XSXX#2#b4GFq(c}COGs!_I?pWKU6!zR{
zx!}`6G`xICVomv|uM|D~6iJKca#=Z_7q+`PCKXZ9iYR?a0H^23BU<=rJrTn~N+*_h
zp@&9_yJnuXv*$w3gUflLlxBOKiad!U<T&eUq28~2`<MTYGy?YzKO=_0ci;VneEmpj
zmG<c)Km6st=Wl!`I=H`H;dWJL;AX8Wlw!0-u}m31<W7)_^iHRwu6NGICdpy%W-ywC
z+ITNQ1=dlFKZJGmB=<-|Wpa0P7!1b|bdOU)+3KTCdSh;ANIec);#@0R76r5$gQFI)
zP`dHAr{H2jytAwQ61Ewj4Bck#oZjT=M!Ou-=&L#LvoYIbZ`4Orq1*@Ttw`zI2YVKL
z<UG_K#S6DI%;>#!O`pGyT041{EpfF@Av##2=fj6*e)iRS-mPcOUifJbgC(dFS-dBv
z<;Z|ei@wL3Cx5GrvhU>kM$Vc2z7csSirbL(!XK@(GYBEtu+AL<C~exRc8`$EDX{C#
z4$;9H1Mk+v`%~iGDX}Dnj~eU_6E;n)$FMGLV%5V0c(tixDOKIZmg`Y{d1oX6=eb;z
ztd`R6SJnLg_Upgn^!fw7Vi0Lz+Hp^{(&rTQX;EVEO{f+EWEDP!1o$EA7ViM%3lE&R
zu__6%m;h@a-9z+VR^%yBd?w%sW3UFTA-O3ChhJhk+-VQz+BRt2F@}fXaRmLI)+y#g
zM1$Uj3B)W*V`z_=)MqKT$Kh?}`XJw5`1svlLB13HV87i6y>Ll^#XIVLqg+2h^px#R
z&AT@6C>sy&g}8|1p#!M6h7U^N%cWAH98gPbv{tNG&}XP--r0L6I1y|@54NZN0YY}*
zaWEVc=keIE^$lXxEFo}-3+E+~g77tuHt1fV)k=Po%RIbiw4wB&YQM)x);_6NIB$^5
z-(D3}n*+!*rfRUoV-6u~vp#5sK@`duz@Y`R_d(B-8W+g9hyu#sROAR+OQXjM!4vpW
zye1>L>vbb<Sz;S>fGkODp_GCRLuK3Hy@V%cl0;pr25u=O(y~JLhn4W?awOfZH}?C+
zo_G3sR&`+%(%|*rF_aPE%;KW-!Md2%z|x&uP}&z3k?1GyzIV0B-mxx$_oowIKRxm8
za%N2nOOWlxY;`S{?RXDU@YN1ArV_e}4fnUba<`~MiOpbLBI|iEPoe++>;HRSR)H|>
zZk!ocK62YN_zp~D!5g{|a0e?)sAfAL%t<a$0c4H|mb@oXy3L+!*zTP=e%HWv<!}V&
zSW=X7-rf&3e6403Q%dE2FXY@<gTchPvd2m*m9k5t)QbN9<Q9c%RaNS-H3@2m1f|1P
zzcayED|dQXAOr#(Z#3SNqS#b<hYH(vCl@^@OAM@w-tS)Z<8u}%MkI`A#29H0`Xp1B
z8KN5opBn1I5NoY>=ylll{2}0@@rJ`}9O{tFF55P4x%YKa(9Ky5E}DZrZ$qxqoA~=>
zP^;4>-Wtt3<AJ04bq+C7VxU*)3r73fHV$)aw1selkT-BB=v<up2X9$<vzB7ig{_vB
zjkXx+T``DzR)t=-mAWKynA~Ci`8#Nx?YeQlZrm)N)gcCfHy6r*H%g<5j%4!?PdF+m
z(Jv96Z<WgQ(yy(t<(+NYWQd2^5%k6wHk6HW;EYScheYm!eJCDKv`-xvA)32rx8Z7;
zEv(?5;2kMMPUn@Urwi{dXHKUTx5)7klba3KX>Mj)SIvst6w%bjtfn<;&E#4vU_=Ww
z1o;uIr}d41OU1^&`qzKPmZgwiVq)5=#}JQ>aS9iK1KNZMy_mM5R<^uz&q_pFQsjI(
zNwa~YwMwp)oSV|SYB%jgIJ6Tnri1z%4AOJ+-Wq$W578)dDazw0#wxQ}jx03PsZG~f
zj%wQ<OrnD&WY;k7N+X+u1m<AUn@`S=EzGpnURTkXQ%VoplrigiTTi@!rN(%4;A0$>
zb!~<J>>vFVA?W*h@9cXuMv|^}w3$$!&n}N{RkkAv(<;t9z3e@XNpmn%z!_UD7)+aD
zAO+EbXF%4pgf7U*7Vil`>mp7qK`m_Cepr-vqk8I4$ST;j%<}Y17@pA_Eq6-k<h_!2
zWplVAXh^vPtrc2NB3Sf_AMw~1`{)wGnt5JYBCQsy<{NA9EN)QpP6hT&NpzzPLXu{t
zzv;AVQ*J$OkfNiDrfH7Lf)q8k>2~-T)HL`0G31F=loofO0sn#5Pq(=+HEghec;^I{
zxcOk3`H**zf<YCw$lX9}o;JAcJJ<WhwiN}r%thT0!^2jV!`w2XGw!-0BF^*M*INZ?
z3Z9sP=znRZCTZM47iXR}WQ6BAV5V+3x%Lq^_eJYiyyW!FDv0RGyQ+h?W*?)=74yDj
zcFST2LChTIA@7;%>nr!C3#WA@+ThjqMsnETOBIzLK<x|)-b6gjg2;2ulv?mZ-4R{B
zInzsM9*8j}bqhY=Vw4WToG-nz;CByxc!jCwat!7j50fp0*+-VsiS=?~q(Jq64Xh=S
zGb){PZyqG6E(HhbNYuvk25!QeeBUh3zqq&m?f>?_;@|zYb6L-P^Zo<tvT(kfIbSY#
zuY+OIb_E-R!#TANMTzEp=jG)SfA!tB{M*0zYsTpO@@HT3_y6pV`1-341mJ$Z^4-Tz
zeD}i(Km6t+w~rf_%Z2mhLTMGe<HLt{yn8SI{immA&3%@|;us?yl-~8}h3~%mj&Fba
z9k(Ao@no)>@7{ml{PaXxReio*U)YVG??7Do%SnD%LmTqG-TCzC6TkVJ-*S7oQcGpK
z@50X4g|NzOty}!*%Wb1G_@iI`oL~Ry7ku^AdoE8;#ATrj(WqOSn6w?tMkSv$jyC>{
z*V~PcpMKzf^S}OY2|<*`)&|>_S(4V0^cbe)utV6GBI;mmo+As7&dlSsNA@^CLNF^A
zH4232Sk}lA9NC`V)|GxZ=!7LwP3YN745?Yge0Ha(@Nw7z$6k|s;Bq>v&_G>&DJlxK
z`_5=muSddRL;pZ69<w(w*|K082RZL-`weJp`;C3yVKmkl2p}Uv8+?zh*+#Oc&)_jq
zjys-%>Ha*%G08m;5N~wV2^r&^ltJc#++3sQDtqKF1;2mC>!-~WoY@guhafKvd2KM~
zXI8x=<CnWBR#Ov2@37a(ZQr@?J0(~786mjB86CBoF_-r`fPC<W_eu<rW*GaiAG^VD
z6`bg9m(z*C;CU?6*12vQFWXMv`-3&rQ9Px&aZk&xNv$2N((JQ&lj^r3b#M!wBZ%b9
zweDk(iy{YWlg7f2iMKj(&iwfCBd1jr@v5nj5}?*dvZ^W=yGx2<Sx@9WQ|=r4edB)H
zxZiiq$%`p6nx&SBDJ1)jDnFhs9Y%5z-uUrwYnAtqV=+Uzea{fgOshLHS~Dy)czX1V
zCE}MziXN91>KLRd`vC9U;m0zCI5`XV3>Py&^Ozm;9B6mCVUWZBEeqF_H(S}c?st6j
zoae#eOg(RO&JZ3Oly@Q+7c#YA3^Y+?V@Nz-&YV~I7<ngq?&+sA?X`W-I`6*Av~gWz
zZ*WTT?pZa^jMLPtKWL`I6yzddIP1)k;5`=@e&9{9I=VX?I#XDP-qRX*8qr6V5N!i+
z3@=OY8SN=+ZGU7!%-B+S<pfbp{d61!H>5r2e$aiS`#0f4&Lk1Gx#v<y;qz@mT=)S)
zPPI#rIi#OLs`04k=WMA?!p56{|D3sdFCpc!sFR9IG#6XPx{9n)_f9c?BZVV9v>K&k
zV^TeIG1Fjgt_>B8+Hlu35f;y~%3#ntb&-UqS!Y&3$LJCQ^00=nr(A1w2)?Pc81=l;
zy`!~4E*pI)+^E!DzEw{9->?twyej=AY52Y_N!coN)@9?qOUn|xGdi>Gxw7UbjP!tn
zOnIa|Y*-IAB-B+F$nOWs{`8rG4y|M29K5H6j;}Jo6KtcH$L^F5dfy9Wz%s->$hmUg
zGdUM(soI!_h9=%TcQkYf$H5@EzqshV5AdJAb>|&^@tjt{9Osm*rFGPrSrWqjnmLx%
zP9IazTX9_>7twpdfan6<JN7P$Tl8x0w0Bv%*cyG5&)&zFhk^}u)Nz;~XtvWKP<vzB
zHyBMezfs;eDLh~Z>m^Ne<QXB5@$_7{-|v)?xulgfMbSTdqt|RSUr)=8yx+;&onA6=
z3|8j|L)1FQAbA}KX-Sk?sWb{r$<G67gWLwC$*W35r5?(-EW~x87^U%YK5H&>-zZ^+
z<-}e(Q_OX_JaN9P;0CvQ#d$r<T}m9Lf6|w3jxk1-)C`i@$u%LH?9btj<8*dn#7vuj
z5xX5g!e4{uus^@7D@$5%KC;w8hzr3lVt}ok<-90Zss+~5%H{dY`EurTK8rTIoZi?H
zn9Jp`V5L?CK^@SMEVkpa5aU8CVzv1u_-)#>`#101^+2=AbZ$(+98PuMDDhCvno1-b
z3<*OmT<Wc+p^uq=ZO*&V9Z=4rs<y;7=ZQ`S(ie+d_hQdxA)y)Iv}l+R(u8itQ1_g3
zR;KfK<PlHDTMtGvPONL<d|q*-u&hy@NtTlxV^r4h<%Exc?Rqz=a3ZA&WTw@&m^p%q
zRg9>kJfBZIzki~(fe(Y21T8x+h`ZJ*iZn1<RZ(hM2WKXNt+yjIYF@LscBc(Xt(jI$
zh>M6TGmD}l7um%^j!sMD>Ad2ir@F%Flvp<1%f-wgT~MC2vJ$$&&b$NPZFW%KbQPMX
zPS07!TRh{x`}LC!1<CLI0Vnikb`pKIjPAe&gAJo<8`L&r#^&`<Uhg~iJs(5IR!tIb
zI)S%=EFBZrI~l$?r=hEBpJVyXInpVyo>$iMLJFE^0E|3HqjG7h+QUtOExzjawGPc|
z_!^>4W&!UGA0wsrLthb%x#Q!5C;ax+%^spRs9g`xVor_gns|OX^K`jzKAjHr@?uJC
zg?&iu<<Z909ise&1bvW$A~Qx4YvUAG6L2{kQrC_k%i=ydqcwUd_}Vyo$La%dhzS<m
z(CjBhQV`8|dOuYL*#t_p4fayak|DG76Uli}(&jnM&YGojjjDt;4XPt91TN<jm*<sb
zN#vebPbYlJ^s&i%Dkb?prRgX1c&#b1L>p9EMecCw=6XM{@;NISL5GC|Rqlw!EVZ#`
zStcmxDdEFv%q*{K?97XQO`HZiz0;;o9jq5wWuMPy&Zm>;jA^mMz?#Nx?Y)~mU=0@{
zX<3!i90N-Vr0A)`X&p6x*u;C(F+}GXeo!attaq(RQRna-YOA;<9{xo>di+qgR_>K;
z+sV07yA2$>F@2oFxxmatcFu%&ZpuWpKHLCW<9C|3J{WtWo*Jc15!(YMng@07)9*|F
zoCE0O18Z1_!5m|qqYY6^y&HtZb2=voScXSnahlhyOW>3OrznO<ikgd+z3CYmFuE6e
zXEcV{$)qJRT0h)(&*z1GUsx=2;(R$Xd>|%_#wjRtxInef6oTx#<?`!~YTO4eW|i0T
zIt6|h+@)NDL1~R-ho(1wsNo$xMbav<Cf<=2I9(Q&%?`pONUga}r9-gfyrizN4F^N%
zYppdImcoHRxDgD#`vb9$X7GkExS>T+9s^=!{=PXmLr`F}lX={hcV4!Q>z?Hal#7Mg
z*hA&a#K~(I>r5Frm=>@Sn<?$-QBk_!NT*0x43iA*I7F5R)peHKIIRnNs@%;W&UXw)
zeY8N!DdI;EOpjOSg=rI()5>}|v7V=(X8DW?HNZBg0X`7vnsEfj>Adjn-4o}uvQBo<
zx{#tY95G5PGhJP08ks7vK?D{rp;wH7HLaXaXPJbVUruiox(0K#HA>cSlGbJ6f#x*P
zIiiQ?WImRH4Wn$1Q+lUNHE$D+d*ANt`OMi2$cBGJTxwt~FC^zZX-VW>6$r$T-NH#h
z^)Uu~vV+xmyw_)3P7BY^Phx4jc|EOXK0F0-$@<x0AqC}E&p~Og^_VlsyWe3Mg&I2U
zJKMf<-*$^AG){LLW((4(Hn$1h5rZe363<T;oQIgzg`U>cIPgjj)V#(UjNa*9s@+3y
zX~SIa1xKkXlnup<Uv=I=(?O>WVHz_|#jrtNZ5)ftTL(IYVmOaCFX^--7`vErr5u*<
z&WfNJlJzzVMQL54VNQ>y$=dnSQf7={?3{6Qjv<(G-R(M8oM53PsWaG0CX{d#Jl-B^
z?;Oq)+S86cEs<r_;eT3JymRE<2o5e!atcjL#LwJ{6mhXU$Og_jC$B;>noDe>3J!`=
z4Ts?6Qk+C`Xjn&Khvdv$>ElfzA($DUhTGmb+P)t)E2kxKGDuOjJld!F(BF$Gs_4W>
zDo%xGSQkYOd&fFFm-C79MaGHCl1PhJ2v}6!JkH=OXTD(u4>fycyB$~^G%Oz+mgxVX
z*M9{2NDodmu)UT>uj6=Q(F@8*cA-P`!l;4TJ9}%~48oa~+5QbIKl5H^UM>ECA>w2V
z;D;!1Z(q|=AtnXVcRy&pOWNO~*+dkvkh;TlSrg!A=Q9Ts8n618)S|_v0d0t|MErRr
zo=?P6viG*W;kwvR+DEl<#sFCswZxaph0FQGYPJ<}_*{?Fv}xKEyxPXv1a~o15xd@S
z9S&g9+a_YZNZvevS}jzY58Q5d1s0k9K(imQiWV1Gd|(YyA7|?<XkE}=>vCEh<`q?{
z{9E>oWjhg1a@}m+IyoOqxTyJqi-C+76!RJmt4Nn~Vm(EcRRcEX;q-Lkn-5>HuIp#_
zHl>B<bzyg&E$>7$Om;X%mEE-(2XVV;c90?%WY1!P-8LNzdo^L<k*A?n|MhyK7CEEJ
zqbhM(PPm}ocRHOpFDDI^{X+2R*pQtwSj|7&W{(3)11rs0QhXCE;~#^<VBKNh(EQXc
zznWT$4TO!_i$Cnqah|v&Ga&5v0PCp{m%AopW6~D;;d(nEork)2(v-!y*E-fSTbklR
zKgS*%45vdOn^pEi?}j?<lp^PK<#akxYk7EjP5b7@`Sx(8b#vexOrO(a>(b?YbP9qi
z<D#z8$0Vzm6y&h%cWd1HaVQ0Ek)$DLZF`KRWQK*a`gh(G;)!KV#As}#-f(!TiEFEI
z{ge5U0mSZE$3my8aEzK$4Rm-e(`Mc4z?pFJ7~0_LP(d=hS&=le+p)}#Xv*;EF(>M-
zx%3j~ZMOqldq(SXo+Z&=ZCyqCA41XW06a{iIl8mO#?eXz?^r`(wbpKK?lT7SRtnc`
z=eq4u58Gxnta5XfYcQNUJd1+2SV=v3@OKQ-=RmuF20&#eHrdM#=M?4vVp`0h;OM4q
zgsIbRrQ!$@t9YlZ+a)rV$cUaE9X;5vE2Ymm1$$nYK&!%Jgo7>?VJKm>6xYLfoRfdN
zDvvpspZ|L}#r8H1&!CRe+-6KwvbRdX=$ECj@5X>KUzzBpgiwO9<Q;JgeCv!>>3OH`
znS9?VTaoG;%o7R%o?e?6P&pIJhKm|}DJRy?8}VjiQ+#N5&b|i2Lme8R2j^I#hJH`)
zpLl-%OiG_?KfITxRVxkO<i?r2#etfVdJMbgy)#D9L+!k<af7|~eP`RUy2>Cf`^>1W
zwXxkdF-58vYH67fh|Zv&X#ZSDX%%!fyCti*$C$;F{kySEr;*}hu1p@kaw-svfWBv1
zvb5f)mSNsXIqvN=HdM>(k7!a2M29(u`Xk3eX?%~t|9AtnO&YHT!|D=`!^7@TDZgu^
zHoA2|$Cp7SUB-6jbrrl8WwS63g=p{TI6UWdvuihJb~87nwrX}L4ZrB`#bseREwa6c
zp0osd@ftJ-=uTaS9B!v|x3d|5j3;MG-I2j!g=dV3(^)f%kOWKCCPjWOkLb$~6du(3
zz!$mUHs=q&r6mg1TT-;7Ng48u*4TF)EVWD~9t_V>NQJILXNf^Yus5GMZ^RqZHl->j
zhDqU{>gKsdCN*F0zfsE{0zPu8@&=d~;(<$gFOF=Ck;}U9?sVqqv~pUO0}dFybIlu4
zEQ&WV+sQsRQ$OP)-J5Muo5xJ%5@-MOR>=4?$m__|)ve$_l(PYqMqC2Rd8L#)^iCj<
zykqsYNumJk(PiN*xvjdFxF%UOIAM7Nu_nHJO-wJPWJR+-*eR_yUh)pxogSQp>jYNR
zMt8(BS65ZUtQd~q#fI8yBM%rsltm*->DjV_W1zNCL(jqRswSf|eA=$7cUk~(bjGMq
zGNTrn@!?%pr?o_{L*5DFO(7CJc;o)pRxFe~*e0n#U#_CfHmkVyCP&&llrd3?W>UdB
zPUn@&y9>){wU+zsx>MB-nMmYmy>MPGtm~pGx;OSx=ym!`dAenN^e!%YtyM{TI8sbv
z^9(VxTB(#=*td<m?Tls_!6E8_?GG)*Lq`zoS#VZlJZP56n~-&SYmJA$J;t@yY8}?n
z&unp=aO3FBmDNohLAOo!K_%^z>giCfxDQ0h@ZV%dCl-Ue)<wQq4+{G@Xu^j8Ed+XW
z(g2hqN;ks7zZ_F)(=nLMhOjs}&mL)ZDaxHR1@_t$ylaQUgoPjUWwXcd{;<1ntua#4
zd*}^c&B7vTqitgtR42E9)1u9|bjNvFI3;!CGe#EE2bG$+-fwhFcVdpb!|rW}QS7oM
zf_Lg9yX@m#a0VSY`Fsu8VB>VKI=j*8yiBhUgXUt(LMersRX{1bUF!@d+j4J(@p@i3
z)rILP=e*^s7~P)UZ00kR_i_}Ypu&%1#?ufZwT{p4;otT8kBj3RL#UwlN4nqe@;zEY
z;4~|RWjQtxhc;!pL>N<@7p4GgR!=b;*xK-bx8b6HaPPf`{>7kqbjZ8l7C*v+n&zOP
zAoJzC&}wJj3ftYY3}pddmdMlc2)PM9O4&RfEQ%TOGArOd%fR+|71eE^BslvKwdjLb
z9;Fsuwma1gDnoc@RQza6i8gd7lwR3N=DzRLK8QZiOX1~u<<tGnE$2h^4eAhrvq5*S
z8akoFVfcA!%baYuSs|&ri_QkMHM8CB?6-|xl!BIm=dvcA&nu@Ss^GFJMs!_P%{_~%
zx~Vals`t@5IcND=mBM}B7z3{NJD*-QUSBuZD?J95lbkE#vi#1VxsJhZ8^4CcE2l-3
zXf<J3iis`klx8{0Os^~scuxaawcWwgwsH;abW^f!+b&UEX`Hzzs&Z&!_c7SaM_=9t
zj%;Kr_cYLZB+1n(?<u__qiBx)#y0$+1<_&J9)4mDoO{@9*IKEzsq|76<Jb%;2_D}k
zM4-(wCV6O`Y$#6jv@#ymxc$tj@d#MuX=`4vzk4vej+YJ*=aj!5gGEPrPr_m~gYf>-
z=|u0HQ>pA-j2ImPkLUULsrK&@%qf4Q8C)~bsX2FWh<=<xl;|sZ&a0Ssqrz!fSfaYY
zd6>^Ul->umcYI7oz*StA!v_qeeaqwXI)j&dltX~7p=k82iy7Bi#ici5Hgi6nF}^Vb
zw)>rZE4Ua%1SpR#$_FtlyFutN21fVVd)zQWh`OzoA8xgCpVE8xjG<`OW}m?YPwk!D
zez)WM+g?A7eH4;liV5wEHW7HSS%Nsl;#xba-3330>LqNub|sLxiC0_M<c8Q0IOK^E
zCiTwR@2L$K<1lcWuigjSUP)b)(hwXz3<$6~TR5wum4uHr@Nu3sInJlZ*|2S#7aUS5
z9i@tm-sy4Bs(j;pf>E6+vwU>GdLLs@iu4xE^cI5(BQb_d_K3<VNO@;3h10T7Y9?pT
z2>Sh7Yh1U?zGv~X$<$K#@pj|oD&M2ivWVB*D)&-w!<$`qXEQqD>=tGq42U}1hYSE+
z9MmG}T+4%fE4<z|O07f>m*_cLCh_HRI^rFdC}-B;<ZP=<gAT3WT46K=3r#ii7@h0=
zE(r0q@$%`)cHIaW#;m}djL!IC59+Rw%f@TI@yqu=<In%}@A1w1cZ7f)Hkn(_5@+?Q
zL%UXCx^<XBJRH|!N%5coPac08%I~eU^ZL5eO3~~mD1P>Iz7YM&8dlC}<&suHh}6-*
z^@Hi9hts1B3k!NO3-VsO!7eZwbn9vjO@!ekGNftnJvya}kKV?h)yBSQ7E(*GLt()g
z1-1{tBBfTEiJ-#jaf`>V3gt<oQL(Dl1%rI$y4l9e=#Grg?Ku`{;Pbhes#53O5#AD_
z<u8nC^l(9@Yh2mu&ak0qKwV9Wf&kN-`HL=$4y$deEM|qx@YF6+gw|aodSf3gi}{=t
z!Ba63Mzeu^G#yC9<E2Zo=YgPo!qH`LX)^|qV-$JnTqiiweRO4jbcL=0&WH_~x0TW;
zeJDzPbV@J!Oy0>!!6~e&md0Kxy);=|R57ZOcUZ=fY%OLsga=%Z+kKS&D_Guy8`3?@
z{cp_qfy4GK`2o3Mo=JZQ_(;sG2klf03xjvMJBG`PZ5;FT3&9&DPH744_&K%#7@j))
zQ7pRmjdj7J)n$m9=D%o#WanF`HLx^@i)V4tE(}D?!b4m=%Q=wl4q{-5o~O%+_vaG<
zD7A7g16Mn(3^KyZ-H_X>1VXA>Q;fLDmp|Tz1V63IOK4UUlhP)|wS3xd*DGrj#cqj?
zm)94XgMwvE<Ra+9e1!IrdD-r4r3hn0lK0#Fu5Kn6OJcr<(Ipbhg)n4=5u;c;9kIbC
zt1IWm?Us4DZgwa%>k<~RDJ<JyF*q=KMPts~3y&Ftm7SI~$lbTiY#A&bao;Ju^LoE?
zy<RD|OsbxKN+xhB%7XX;swlPa@p|Rm^34DEAN@m~`H^`2#2A6#3Ll<-PFUaLww+jZ
z=#91v_Oj8649;SVme8ekey&e3zf^r{L(;*R^Ty?Vmw?Sjo-UC!ES%!XIh<I-f)5ej
zJKk#pZ(}FeA!BAM=1C(+Yjo(gt^l!#8nRhHzBRpd%Gd?NwZ^Ej@$R{>Z<>qvIS;e?
zxdWwlT+={VG*N`HM-@}f)Z!udH${vz0x>4QF-<5#ba*#dVmds#-X7G>Iujg^dEDe_
z*f-2W-YPuJK7wcQHq3o9e4TT%VeqG{Yzo0cZG|yJpa@P_ezN;EqB;`iz2WN4NC(}-
zUE>jFIj#Q3BQR+S*<wsgXV+yB&3dMurRZ@Z8X3^cI$IUpwJQ?yzHdraoU9`6WKC~H
zfs*G`<hn4KO*#X;bF?wLQSd2<MHWOla0Bb?A1Q3fdm4%h&LxYY&+LNw@mvd{)Baa`
zr_?{lOu;(I@SxIZ7OXZUS-RSJ74Lq^wdVm)TO(JshTU~VQo8x~-GR$d^$>ff7GF56
zY-FAD)=l`|d3zePI%?IU;=Lm_M^ByRs^F3Y%O!HI3shL0qWPXKD~k`5T!`C{f-)C!
zRTWrV<ra5p>*Okm*qj->J)g$Juv;TJXM?)|9>S8O1fPs5*QsS^-|nQ~!3DNfxa~Vz
zt(NAcs;t(^y<~1#?y@cf^Glkpvzb<P^LvL}3%PafmYyd3yb)^WloUcD4S_iNrNccJ
zZd)N&UCS6e>k>IF=@I@O=EH6ei7{KoXiy71XX>71Fm}(l)&x(P-XXWneJiwGb|Dcl
zamL_qcl;9iY_)Peo%qlH^dAs^{A&h3QqoFXzaoSTH_!9s7mVn+)qr0UTWzESH5T)|
zG1vy9WKOE06)sCk2dvbG><(I24n&Od++RZC6cVSTf8qn4Mi|oN6dx!~xe8$pQk`5#
zq?d5!NHU1tL6MlHo5;brV3r3_gpMvoUpL>QZQI!QtoLk>jJM~b^$t~ntx_AcO<lla
z4iaOaHF4IR4Jc=!hpp8)%p8XA&w=V^+kkU~=ngfmG#^q+2eeTK>(1u0ADo7<F^ZWe
zwYcU8?d=?<G+ODFA)vjXqp+Y5MQ;x&5@IwGh;ZVZtC)~Bq^?b|r`-<9Rwa^bwNbJh
zlb7|RAfyz`8e<%;zi$3)@1TuCJuYvw%C_G%7uYtg+Z~2rP3?aR%d*%^Ne3cm577Iy
z&)Q5>iKEh}G<5+1+;k$gdo^u8^d%BO8`3~H?+=EY?*nh^Kum`QUNh2K{t)aV{}_nP
zec3ubJi-^37+9@iNXA_kUC?e+Z#rdtKV)CSgn`ox%W?P{>c5XctLiYsK6C{4`Ocv^
ztNu>yjnN#wLTrvUgn4=ge(|iA$QUcWdtx7wn4ecpA&`<bkZZ^U*BIui*s2z1?SqqZ
z;38u*g8%&V9RsrfKsS6ou-@U9KsY7RY0+TLTGu){dzbw}ZjG&Io*-OV7p0D&;2;~=
zhA}x6999*h&p^n*w(q>|JA0P5l%JI#1=d79Y1pM)3N5a^cJ|z4?J?|r25+scL2f}+
zS!;z}bpV!{+4C-k#(k&km3=SVZyR;bEaz35zQG@Pt5A%wg1Jl%MO3z830`m4T^szv
zU;hceI2Gc4!^dYbkscz^pGc+RUw^}g^}<*aA9oFu$1wYpIU8b7hp^Q#9)^GpK@CeZ
z^SHT*riYMl!lG+PtU($DZ1&;16-O=*@#bNpqGsk&h@mq{68frLkVC1Bfm}YLclacs
zXdhw#l_ofF-!%Y^bF(!lb<QwGQKym1czFN2M^`@S%N>ZwAE-C!er9DkH@|5B=%>y4
z+xubcGUXFUdm-p`+ICD4*)XjQhyPiaK4Ml_sEg9tZ1wS_-{X&GN@JVJ=9504PbSrn
zkzFfOy)EWev$^2$_lI)Uv)Q&wI}(q=6>o)36ZEJ**PnSOJMX;|Zrdi(f<sn4y@;k9
zO=FRg90m-w)g~@QvP+J*gya4UH}F0hBd_9GXSje1gAuy?)+SVEq?Cg(<<w;zET1lF
zK1=|{9UTxTBL4Sz{Y390DUa0}caBouHm_us%}a>l+j@s{Q40u3u*KSOgrthUn?V<o
zl4|&Kt4+hJfQ7PDDQ>+^#_qoyn%Gp?!El0nN*lP`7*T=SY0(BZlE<%sNF)vjX~8X#
z2ISgt$>WmRwr*H%Ejr6FkrgJK`ts}@U<%T1RnX8)TOw|W_*HGMPe~m^i0n9eE{A7&
zH)fOXW@YVla5OU$lgXZ9K~=3KPBB@UT_q3YUH3VdEDa5y%nxNrQkZrdl#R#WzgVT_
z2D+Cia9$(lLolRR(b6id7q%kT(fzj3a+b#7Uijg*@#(s$(jLQTm5%%CM!i>hmMI}o
zn-Wd~#&X|&hj090z3>md`kMafFX+*85AUcUv4)ksB~If`Y8%c$3TIOAylk1=c5=&<
zVI1>TkD=~_2QcAO>s@d`NJJNTTF*SKXL;}th+#SS>>+5-tTxfnygj+7D~k3kcq<D0
zWZIxjax}THmZrnFwuYw@T_g4ZWxxmw>lSlu+;>^Rl+rmTmC;i(B)gc#s2Mtxt+L;X
zb#MkT^|{eLI``-ueGc`<Bl6LBD^{qvFTlJxFjq!%@WH%~B7;Vlo3K3?f`s*~yAf~%
zg3~n}!>ec-N|<Y<Q*)#4Hh&2sP*hXt)~07^=0{`H(%KtTsl1a9WY=^Mo07N>(V|Q1
z?74Ei?cDFXQME10U`hDaS=QCGB&H1rUe2VgvF(}H>qhjBXD=mki~;9Xy{9N<&~g&Y
zI*`^UE{lGyWoEDYLMwyrDh(SdFG2c<q?saB1LBUd;OQ25%)>rJ`3_LJP>|%45ppb?
z$Ju@LdcaEkgXk25*-d=Q#~?rRG1&w`({URP!!Tc%iM81RFm{CZ@_A{tU|Thc;s~my
zYPW}IFp8{fha09~{Cuf8?}_UJt{S7szrT0LVsw-?h#vjiA-r{Tap;TD0IP!*I?WF%
z^7J_@!$kr(PGFh7eL<fb4>)OJrEW?1MTdb~BH_GBW6;Nu@|J5M^nOsw)PgIl$azmT
z_)HCmmp)?R#5TAK#y)Xza5=5q^UnRQYTcpt>rnJ&bdEKcYvWvG7JxIk|4R&9PAkus
z3+MAmNmfyvOS6jmo_Fs1ri|fS@LiSI?Y5JPK2Iry<~n0I?$?d76<pJS7lJe>!<h|3
zpJvLB*VoU_`1>n<tn9r}PO6U6)@fnqrF&9*Pil_U58j8yxAn$#zp>qKGB_(mR^m~1
z*=b!xC-!1&^f62`qCx2C6ge+?*cpywY^By_*!_dWGrcQ=SC^nXY+D0E*uFLxQ`0(a
z=&lVY9cp8Vj<3!qo=*`MBc%;~ELU!2$U<QRY8!`u4W=QRQtXsNQlUF)u9RFqd#}eD
zw%OLyI?O6k%q)ABjJN-m0{ht#A6OMG5$PP(yI}!MHcOaR@v^IKwl=3wO|uPYA~45R
zb>$@&b>;y*hlUbN8t%N#)tF4Z4_dC!x&n5*HzA(of%k@QvT3>RJJ<W2T04^t>rF)1
zy8f;-MT@ecP31lYECyfuknlFt7M3snGZ(zjfEcV+?#EYmL$KEp4q9vMK}z!}Mn#3U
z4rswmt#rYL9iG;OoC_i9nX1NeY-5lNu8fv$=!Ty2*(pvo)bI29$u6k`uS(iR<=$zr
zEpfHsowq^NtW5RrG(WhG%GSkIpG;bBgQkJ<V||x8(sikq>0PmZLx)2f#_;Hsw*y{L
zKLgIsg<|Gpec-CKffcG7rv8+OtI=6vBL`jJk&>(#l5OBiCrmNODdE<DU&7&`;=fQ<
zs~t9_Hd3pG=o!fzMnk|ai@Ken7SsoCheH+XVK|nUSi(cZwS-7Y3&b6lRK+~6XP#2x
zvP9M;sw#;=i?DX~<;1=gdTwHCfDId^HE9f0%B>3esj@3^y%Io1UCC>v;4IQ}XlPbi
zVQ-mSc8QC;{O?=I<lg1fTg?}zb$s#El4YV+D&B&ULhwXCh;){Kx`Z!m6+WC6+Uu1X
zPc$N<2XKKKfq<t_sl(Cngx<Ks&My-D=KGzOZO8Xr@3X@xde%jLH*rxH)?2g81T0*B
za6VmF)5+k^a6m<444GPmVj2}aY_c#PN&#~+sv8#ka$25jm^pAx)u|&;yCeC=FW)<U
z{p48tO<er=#PaUKr_+Uho_D^rV3h%Q0)BWKL_2DBl+6w)(TdFCy=!BwcF?wmqSc$I
zZ$c}_GlVv>W8PqgLakr)3PbiVz6*kCO{rxJ4aZw9jHTh)6kHlLxf>PN={l<8m;>fY
z?UmhPY+LQLIBbSDxYtUo)AC~+Zod-bcv;|>HAN|!FSMf1-dX_`dd`$`<7F%CwHdWl
z_7KebX7H=`fm)s1b?d;VAU2{Mh|!?LT;}`+a9#N_!5MR`a^LUlbr-#6A}zMzt2}N-
zaj*0qXw58BYNKw2ax1h=US#)O)NbEAwR>_Elhb=q_?v?sz@;#K@(4pLx2f{}eO^Ct
zrr<5D%ujG%i!20EP?wNmwA1O~UtU|K)XF~EfmMo^ZJ1wI6*-zkW?B_9E15R&#%9n}
z1;*#_(PTt;&p{`f9{tWaanBtx-13^pmpw}v*dTSa{6t5|jphfWps&`1v9Cd}*VHqF
zK=hw8GcYW5;<>@9{<tJud~C+Sd*b2_c&N3;X1r|Wz<P!5m;vE<{CIV0R^lxw9Ri^x
zEpp6kvd(_ZSt6~=GY&yGF-6%$r$xB0b$`QV7TLNN=bX(dOsHj*NH=?hK4u6@w})Mu
zW339RDVD5PN5@r%A2K2=d!^=1sDotJa7jtL@(@%Wd@*Tu&e1|-q%*@UV)P6s-O-vS
zq?7jP9;p#r-}v>rXTJaRiPw);Uh^&iTFazmWj(ESt#r^$e18mVxIb7I<s)>{r+`(w
zIg`Vt%`?Ua;{>!Jx2|BMt}Zt?&*BP;??fMQE~*lw@H77-&tC;r=U~4hZr;G%U)XP-
zSf4-ekN@mfeERgl_untXDF2>bI~lnThY?vX5pbqckZWlSCcvz+J!!RSIP1D8i4FE;
zR`tWW2J=I*+t6y)4Cu`*nHw*Cd?wWER<(9F4}(GQqDYq_Rc-VS%Z_PksROJ~ih^<O
z_nopAO3<vbR#VJdgl4zmupgNa@+eZagW#%Kl#p0+q3k;$B=+sj^?v2k^(I}9m*QDW
zsfpE34FboW)0?0y=e*o=qlgA6MA}46j8rq#;6AGRExoW8!I<{688kQK-4r|{Vp=RA
zB(-$*d*S|?xxQvf?o>Cpt%dzu!NXn)_xsLn>1@M!aP8pi_^BN+`m=-I(f|Biub(Jh
zGpbpbwiD30te;a9Q)Nlfr27-b(Suf6GhYsj{b18ACI)FI_86K83@K)TW6@1RFo}ge
zU#QdLMTu|@N09E}cn`A*g68eC29^-Wqw?DCEIw)&*$2BT^f2(r5$yC1Y4YfY%|d2U
z(&z7^8x2#YTY<PlLQ1C8_QxhFKNEj^rqV096=Q`so6|(iUVwujUOTlH_P9~%nNlmx
zSrwcT#W_+G{PBFg(9)olOiv0G`TG5PzIuA%!_yNXIBqplTNPue%EZpLXW(&oxA`8N
zAQ!xsMYq=m*atM6W;0f`VVwOo2Bl~q*t72C)Uu?QIIWA;%$L+7;B;Eq+L;la=p>8!
zl$zTOWz!BH*@_OVOFH4pm4En)ulV%o2maM<<9gem%S&fnR$_`eSi3fyJ!fidMg#7L
ziKBDs;!3UD?l;0vrM-?$9La=78vLru5jmmTM$AwLbV`vWu0#UsC?vo0&3Y5tqz8sy
z$fHtfA+}0%gYx<V>lpllpMS-_|M!1wzB$sm48zBRlb65u%KLZ}B>teT_GbEsq5rN5
z6(*tl^Op{U-i)nNL=Jd+S6UxrQ_*UmY~PDA!iT|JtyYTnW{)pcRPuq*OrhP)w`!QE
z&+H^hZL|Uu^b3leP@MyM(ridpF%Gf1v_I&c?YWS*Ox+7LXL7i+?>pD~&Zq0f$ucxF
zC(^;dF_`zp;*#}wZ3A-UxvOJ8MP;am;Mq+j>&y;i+LP${>Bn5=0aMvw-MXG3C)elc
zS2Uasg}gOx*UJ61uovC2`<l6*3(GmsOFvjutq;MmMp^aV%jmOOmWAz$(S3QZ|Mu5U
zrlih)2v>KwFm$0(l*~M(L&h#&Bf&#$K&`4)dwqyhWX(A20*Qijw3k&|Qg8iWF)*fD
z@JpO?XgC#x>BSIDWoybpWHDQ7GnNFhoQ{zeI~$d1lrtX#eF4(IMTPtfrixR8o>Ch0
z%>FwjOgiM|!5E_qtscSOG6r<5%4`maOmfX+YoFx{genKG%1lGXO5HPg+i~8r`@!{g
zWsOkl!iaMCEH)VT&a*BHxe0$<QsCYB#PjLI^SW|5pK%`c{dVBQf(x1$5#1g9=GugP
zdTX)gz;+JXwfF|zd781MhS?vevv|yh=*ToNCbN>(Mtw?&<xTkeaYg%J_(b!OHq38m
zNV&P^JFdZceu5CmeGrH&@tLI-{_&sxIp5#^4Zr#JN5=4nvQ$5}A1d>G-=zb%=L2Su
z&4Zw@TiNlqR}D3t<1!j+U(E&7jPXpV+*+lUOfOk6sp*7^E6F*|zHsq(PJUw$C?hi5
zVzlFpD*svioP1~G%vZ}-{Oa8^|1xO<toE?y-0;q*%flTGxV<ayV9tmpiNNJX@0FRE
zjC*wZli^|?oL+6s<;uR5&$_oU5L{=uz)T3py2rts5j0#8W2#imtXkBe<7Uxkg=MYs
z)5^I}%Ao9x7G=vZ>Lh#&#+2!sPGSs<uU{CYfx%hzzH_~8<b9*;5d!Q*mi@Qe&T46E
z4?W3)7Tbp)uUfiP&~3<u2dDX|?SH|Cqd*QzWIdg@oX)Jvq7ohR?Meynt(!9yx3cEN
z#LRT^Rw&UE9F(nbyH#$t!u?)FCH0P$2Q4>>FO)20yFFG6Lli<Z%6lKQuII1~gWAF$
zpi`JavSyYB%5|5j@w_Bq&(R+gwlRho=plb|%jE6dPUHy{4e1cx-soZ2=)75&(8HPw
z+?QL&z#u3r;mIBCp|f(nH^->@#V#2%V@hd)D%{}$ZuNv3i957?Ko^6`JBJ@~<}6#G
zj%pOj@!9(jvm=;gwhp_}DqAfq)rQ`=P@DvIbTi!Qm3!GKG_sc(IHwgi9A&TE&9JTI
z!ibTU745nB&ay1zu#;=SJ8#UEO3sCC6ALJLZ+vkykVDYcDXJ9?1ncUA*$y7Vy3*Bd
zuW1p-ysN?qKH$7?=TaMWbhcW!7cnaw-E11h;SJ-RQl!EhxCO@=BA4~Vd0BqvP-wLf
z19UfSh?SBiQ^rASnNljtdLb>5-3IaR{Fz@oCH_Z${%ijAzy8;}=4NY6wienn#sA-K
zcWy5?%2r6;Tr~||E3MM<&fcoJs?m5qD(}XH(ic`VkaoSX*PXrYl$z<S+5?sdqjUBf
z?*lBp;a%XyI|fSo>aF74LTk#V7(^~hpzWVX@y<6-@i0F#BU@?(AEp4LKwH1LoqVkZ
zwz*Bchvl6#>n7T$a@KT{O~?V8V3CtjOp6J$@0sg%|AK+^u$Jy^SjE~7k1K8T*x^;!
z_bg#)shnaH{8Q_ZQ8Jnx3Z<~^m3{A&+{vkvP^I3z-dj5iICYrF7F@zrn&{Mf8SJBR
z-8OQol$NFbt(8_9rOYTAGau~ZKs07ibkNYdIk4m39Jmr$(&bm=VqiI)xI8`a{{4HN
z&nGiA^t1}Dd2arWACBY%vwJZs=tY~HE1j`b>ge3Bg_n<aZm+wbyRNJZi5a_0DQjia
znaQp6OF2nnVmA(TcH4)5N&n@O^|!x%(xITh21FaR6SX_&K|(c!WW1rh*_)_MPL;*5
zM45#7W3$Ij;4$36RD#)>TEbZ$9J!`{OVf+9c6dzogv2vp$Y;KF(cI{cm=a5z?>Q2a
zgP``+A%&q7h}yV_YO|%>l=XJY4Th)X-@Q(w59tz2!HjK+)q>+|^Z43WybYf1pe=c4
zFL&;};_&QY<D6DW^xI(1%_@Xq!9cLNJ1;iiN<Oh|J26IbZhYG1Mti?*IMXoXELdaE
zG~B8hyeF@gytnZXM9l-C*26a_80{E0C{bNuWYA)#hDtA)doJ9!jf{r5GY%1^&(An2
zt{BYEC8(=8rS!X<LaX$iAy&L!#A0((p{jJUbzNnfQe-KxyT%e6OMc}~FOL80&;OYJ
z-M{?zHWw&jxs(E<a=%{LZa40)SK3}#PA64?HaqEbnW*iJ>vlKqd@)vP)u1riEbU%*
z_Oi3*otiVfHPUEYT;*MYrzKKG<XSzgM{?e6gZ7jHZk)i~89tK9a)+%u-@HFP)W4Qj
zESRgBz|O9)b*_H0puiO64IhmDJqFIt%xpVJyoJPrQLu&aSDO!gc0UY;YI9IML?f4L
zL6=Cct23zRU@2KD{|RG>86}8!n!d@=R`*5WPMusdo613fU2`*M=+G%PgwL(xyCBMH
zGorm1BwBU=O5Szvb2TE2u8l2CV+uWlWrq%8LTSC5g|-YUC%~GPMWIWl6U!punj270
zgO;J^!5U9Qccpt}L;Imw%BTiqm%)fCFzenZyE>;<o1kzLG1jtEJ=9zeS5&(NYWLt^
zlxg;<qHQ>c>G#|2f7+pd+K_olhbSRa@^JE=lbKT;i|rU-cJ|uHwcDwyg*{B<HH9N?
zh%GSeL~aK~@o~V|u43HV$IDMajq`*N2}4yeVbHp#d09J1aN`KZ9!g%gy$?{G>~ZT3
zt-$bs8Z_MY!^u%GX_DV<$eN4$-Py*FXG*6@>%c=`bH5jnR%&5!kkK47c(wf0f~$>=
zXAgs3IuTDFf%{&_x6G(bAvms5iEN|N9N+>=ys@N(f#-V9+_ubpmxJj2z7s>>losBd
zPK2(iWBL=-_MkD2VAtJd1;ZFMjPjT(I8O`-?7@{`q$6hytbKsAB(nr4#)5E+UI{aa
z%<^fK6Fn)SUoJdx*!$RNUXEHGM|6V`Bnm5~9?q7-iQY)O@Y+5RDEycI{lAYZj(_&U
zcii_ow|ybHK+c(u_ZPmqzViLYD}66KpBDlZkh^YZkbC3P{S^kpp?KEeI+xR#<~ylH
z;&9xyE4O?z=m?4rTtep;?-S3b0ApdWQu<EmgS`&c6e)Wq*UYjkbk3wvD1Kr0XVU(W
zUw{2eex5wPA<*l_7>#>wEHO}rKNxk@EKIsrA>+D+t|5V+qDo|i{?_zO>4^veYo;u3
zKo7M2^EIw@Zrz1drVUFM>m8Rx5iLPrSk6Uu<fY=C?NBlgHHI^#d6g^iaH6g@b#v8b
zMs@T<_~!?>-k{2GuC_t5jt)Qf78CRBnRZ`jv9RUF{SI|6&{U!fXF@xpVvjnF3>&=&
z*=Qi!@@m=J{y~d%&T(E(Tuv9(bSC+g7%z-5NO7>l%n}ppdSXp0ZSOV+SJ%?^LUYI#
zepRPF>_BXWH;*waH{DaJ=N<;Hd*hZhrxo(<%}RgBfOMiM5lVoHqD_CF*H804a>AX3
z@hN+EjgciLk`D)^%OhA|*tp<0eN9Z&CtJWaQ7o$I5$WfSYUB_u+eS1a{la8Fjk!o4
z3tO0}*Z`x8mz9<UFYf?i5^YxdjhGV7dpwb7iEVo>R5~u|LI)qL3Xn>>n7>gy+%D~X
z%(vton(U!dRAM9?WfZ7I>RN|X!^sZ0E{Xp@I*meBs0#7miwhPTI2yi6{_Z>n{jcRr
z8-eRBbG;X~()4o~{H9mJXuLlq&Oz1}*Xt`k{`irPA79w^o#_wbovej}Igw8Kz2(>(
zSqZJ{4mMNZx)3<6@}pm3AUP#J71ya_YE%U8$%n`*jcT6zUs#ybZtG=8|E&CmN+~=0
z?aH#8h|ZJqh6|3<^O;sQwp=**g<2e!b>+YKAN;4>|9R)P*Be*AknaWO2jBnriSNJv
z#HUX;B0TJ=TcvkrhriO>9FF_GvECn+7<k9JEcDUXwoJZPuC=oFJAd!13;*cXPdvp+
z8;%}UwwB0NvB!|u?mJ^Nmaytjc8Mj3aIu$4y<Kt3mGiQ|kYG$<MY>O=PW7_p1a1nv
zEcHt!eErBhm|^eJq2S?ED#~Ot`l?jMvLpHYHLYhf#4NSqtF4D-K0H;OcQw|Vv`AzA
zA4jC>>^cm#$-7{r&}~TGMK5nAXf4*ch^I0RCR8^Di<@Qh@S2Adv3EwxZ1<h*mMNP-
zXf}H(rLx4g&`~4r=44>kK{E|)<i|kUW&}%0ENNw3R?;H7Zqf7$!AGWNo9m*9j_zm`
zN*2wu&2_T``mn;RwWC0u8%ymHrw$Z+RGah<f()BXMqMPOBnYyYfY*1SJ71pcf5+=5
z9SSpgP)Dl=tHzT_+cjUXfNM85TzkMwlrnD~<FTh}%=`!!IsS0Ar&@=)5%W;^V&fQn
zP_i5v$LvPBhG96b?B$T~He~L|UYg=?t{sCm*S(aU5k~EVOCaZpcft+VMbODo7rf8x
z`}RB6Nkg~(S*zP>!{<Wi15iiMx>Va^iAScghr>^4dgyBo{|GT6c-H7RFOH`r^0bN?
z**lav*lQ=3M%gQ6Z&Hai@q%10{OB9MT^+O!1Yo=0xn6JNT~VIihP`skl#h?(|C%kh
zD!Y>08r@94a5h92e#f!|)-`fY5nRWO&dq64!Vddjl;LTdIB-WE?q}B%Gk|jq7sj!`
z<-OI|Z!e6r5>IDZZRCAdB@Z+g@NOmN$~VtX{7?STKj44=&;B(ZZ+F_Va@}6}{>P8}
z@ZAgdmyM@&J-Sq1tyt+&)z&Jdc+Q#owt;ihE*$u@o)}c_w>w)F+wPBl@yvhz$G_t1
z)zkc-gq0dk<T~gTg*o+Zhn=SuOZoFjL=D@ZeXF=$S(kJi28zw(=<<$7gvmZ~5AU%4
zAZIwnXLRuikv!&N<FKwd2CrX;*XG@sv!dD>K0`K0DTJvjvcbE(M`vUW9R&TshO(}^
zl*YbGo$W1GSlt11r$b>vMm~Cjm{cwIQVhkIkTjx}*M67;mc5einPxsot?FkpoEh;o
z^G6!eN|bK<oc%!at?K3tHX$2_2KQlFe9Xb9eO_nt1I9QsB+5PS>_vM;F%K;JwnqUB
zM}l58mRui!Jy&*{i@6vMX#MB}XMRK*q$rB|msIb+>-Cckg|rA}(Ew8IYOo99?3AA7
zTDDJU0nK5tRT0h>(*c7W_@v#`dBMMUeVG*XC2Z46V}xLXEMvvk5?1#DF*uy-hdOu|
z?a;=~*4;ExTxp@;{K9g&0FKM)!jgoMk3Mi=<9h#G6!Ku@%ujE4Ybytv$~jS1M=5Ia
zgJYPUVu=9;>_x0~heAFCSz*j3hjZfjpV!1`os>Ffo+`yE5V=LxE*o#}Dr;Iv9`4tV
zd<+#g2CuJozWx4%kFPh`Z1kasLyH_tNfyMdS8jQO78TRj%&K8@b9EH+r~ylge0VzX
z?zGTaAv)OBo$S3TTVVydCrLbSq9;!+%V!HOzLBN$V=aV`SWhe0k1yQzohLU)Dbcn;
zSOYzHhD)UN8JAxOb>q*z`oR4^`3L+T|HZ%Iwr6s#eEfLj_2o|8DyJAJHlUY5njC9$
z$r_Hj0Ie&S$7iNR#<uN*R(QSNxZSS&@$<?*`Sq_kxsBW%&0olzxn4KgUT~!nElW6=
z{%6~EmXPSk`@b<*ymh=y&bYmqg^6vXcxfS`SFB`js4%mc!mRjv6Vx_~t7&`Ezc>&u
z>x^o2R0S9P?rY-jG8z<(yR}Ytj$pF<UTr2-WC0$c_J;Z1L+iL<$#-p#OJ~dV==gOg
za4^@VSz>Rpht`I#{ldA)x&e|hS0bKKr3$7u_Iu%W&D1Qov6;$RGryUMpnvZ7Le1l`
z!H+IDY13h$8I7K~>3-WM+|#u`J_=lI_A|AxWff|<G)fh`ED_*1?v;CkmJFMt4s<XK
zWJ@!Z)&Y);iFq5H6<nCQnm+z8HWz7uxJX7_i?)4j7~XL6NlR;1&GER{MY;lyTmeUK
z$56T3pC4_k^L*)l^LDh{yTV95U!)v{W3^NPQr#bAcMRR!Ne#Mifj->PLLW?};TS1W
z>6D<0emcoJ|Fp_caatNrE9^HYmfrrsZ?|~@-T3zdZ$nq-pgGj8kB;x1v?lqf42fpP
zP&MX^&&-f&Sm{tw15f&`!pQYmyp&wH=ppHf`|em=<oUAl&ASuNOTZPi*B@U#@$C<v
z_;}qYx*5b6IIWB2Y)6H`JJ?GW+sEb_l|k*5)~hzts>*pyfe+6s@7EK1hS4kM82RKJ
zyV2+fP8qwWgj*tg!X00%yVB%g@6D?o8&c0rB1TUsJ8j=^rxP)qwXl1`r^pxqkGze7
zbNu<Qe$MrJ<DdV<U-Ioev)?s*RP`!5;_G(92RVP<YvI-f!L4awb%EGn6s5VfT-ZZk
z!O`=TzxQF`AN}kr)|wTjw4AA&$R&{X8?g_9QwDghyHu(n;zN?a&~y*e*4l?aYn8l9
zYjNPr29<^(dS@8&{#ipJIUTMKjRNK}42Kh?d&~-E4!4`|QGF!5xi4>sbt5vllr1_y
z8=mg<4EEY+c;!5ps4;pKUej55^hy7e4aS};Im?o|BZA5>b$r9#Q|-boonY)y`x~1b
z$C4Z;7uiSSDS3QDUP`U7-3zzd25x30H%iU4n(4vQTBcfT>-|0n1zM2-r+}_|A@7+|
znmoyhQGSb9RkPa1=A%<wClTQ|oR2Z+4~Ga6Mkww^t(uMydj7Sic%vOl2PF%f<V6gj
zHo614e*_<y*`;7cN6mkLLt#AymQ}f=C3Z%e8K@i|P#w%Lb)VsmHV|^W(Kj4JDqX+>
zC9;Zlc)PF<p!rz5hkokqGPT9!vvn$^;*$KCoLzX{WQHHPtbWGp1(uj}SVZV;XI)OT
zswmK8HoEdi8I)Ejt$a>Lee+(%{4Czt;b2(q;5e8jMNEBweoh-5d?2k+VMetvupzm_
z46h<7X$}#C)XZ^-OueLwPG3tPj6n<lp7(3wXJ0+@^*QmpCSdUGOJOgS`@X9k9fQ#d
zeM}!KX-+~2;5&7ws;j-zomeg*dI|g-1|I}Y=fu<b#3@F+OLF%#-yaOBn-QmRNf15V
zVQTr$Z|-f|SbJefD-4Chc<-ze)XzLE3D+C9mrsO!=lpb*J8Sm}EiakfHBRY?=kvmU
z?~i}Qr<Xf_@!KDnylL;X+FHxxyyIh}w!uATs#9+6(|Y3RbRt;>Z?U0d^PUm{|H;>1
z^XE^H$|nk*-d0+4-1iJIvZMu?W4}YMpD6o|qmz3jVTOF>CaU~i!Jjx!&iTOdF*qm;
zFUoaovb)Av*FK?{VOaZHh-t1<W3P-i0@C>$zMT*Lj_662+=46{IZJTaLzJ?1{+{_e
z-Aqh0c*R6@tu=;sHY;cr1XBb%IY$+Pm^#c_x-}DR2|CnZcu4=^Jw9~K7UvpjW{pvE
zxw|OBxBJHJw$n@`D;(a0kiF8o($iYbmeW2I)+OdMB^Pej8`qb}{dQ*z%{bCt%t%s)
zLkkiuR*Peuv8+nY(_uYO0n%GB&7%g(V*F{ZkRWb6D$Y4)MW(0^QlB8R1FAnTDBCCP
z-u5r;O8$=5PdXG%&yi)-FR3}wOCaIQ7u_&Gv(s%2lWYrn&UD%lEHA8WhPO|87)+&~
zi9FsgjK-VJp&PC3?Q8TwzgPA=$h$(mmvh3khWFxhUmK(rh>Jvb(RreeBw)d@`UE3T
zMwO*NZ%nU}7(6j9inH`G<JxOuMxDHUiIt#1ZF5kDPI>2Lwx<Ukx{qoDyT|#&?!84K
zCh!n^p?Td|E0FQQ6YT*T!BJzUheZ!gZ|rW+mOv}rQoU4-dk^bnC4D#%&+2%3Yn=Lu
zPl@gqql)h4&29&l7lW(&!W8esx2Y-~RpYL}?r_PG8Y~O=B-OJYQi>0Q1YB3p&?zO(
z7QgAEC0@CH9FX(%<Nk?Xy9=%*h-Z3OwNKt|3>O6P>@U>YtLU}sO6)(!j}PQBc=j9p
z^?Tgq#E6j*;AiXLfB5I$@X!9uU+{1C!RVeq!8?fnO34H#*rU`!2$A<;<?B=6>qWl)
zx2`TJ;dp;u`O^;zzkYJ~?UUe?^u&ls)ZApnRF^^CvSwaiulOp5%Md-a?%+MiEz}tB
z?MiZu%EE9F=ND4X<dNtEs5clF=>e)BhT6xWcJ-|*?5JdxUKwMd;aF@)I_$s@B~+w?
z6x5w+?&FD0PM}}D90S*YD#~Ru;odB)#F1B5-)VT)rRw$;a^E|yRtAZYq1_cCwB0kM
z41&eR_TZ#baO(U=8+f7)`{o3}v`!aMUp6~5;wka8BtnO_7q(kwd#PMs_hTj{=u1Og
zv5TPOs+bE0BPy(GFIv0Tt<gVjq~|BDx1BKro%&L^bfq=cJ+s^4*PCZ_f9NvGw(H&$
z6YUOXmLs}P_dU@&Y+EPij_YDe4!c*abZQvHAb0833u83K5E;fbPib<46_DbnsJkx)
zf&M<PpL8gkpDvuvT0F%M3b43nj-HlbYQ}5R+ep!LRs~eK2~M|0Sv(B02r%s5L@HDH
z_ZWiBO=V6izSxRw5aX<qh{4gDKkSQpw}-t#Eb=#Bk|*HkJ~Bl<eN<{KY6)Al*+aD7
zHyu+unBiIbBF#vQ1Py4TK|MHaqQiURq>Dj<{jAEs^{zl5Cid%u2QnCC7)M%iP3PtN
zvo}W>o&tpK*!$q#8rLG9sSzNaW#<sRXF09JbtPG_NZvEu-jiG?ZLxYw0_**u9G@1-
z9`k)d*TgFi;pA4*HH%uhXd{vHsJe${qC#mwlpBzumx*HXip&h|aHW0wD!uVq?>sFR
zmQ$o5mroxOu{Np4iS*LAzr65jc=s<ZPl8?YD=zQ!;25V&%Xiio`PsW?{^6hfnt%GQ
z{ynu<*?Gl83z6(R$<|cwozo)6*z>Zmt|#gktf5o)%Fo|F@#nw#8DF0~VgE>Xj&(hA
zcPH-6(W?ynMu^l_`S_dPkV?VjEBD)#5IvXY50GSsqHvmyr%|?70u@i<)9Y352dQJF
zZD?#cv!zHJUb!1}SGU(>abAsP>0`wah)yxFs*tso4|6!l({W?41jji=-mmKyo%v+N
zIMXDkBXPlo$TkpANRfhWg7*?_neS3>gMp`Yqkp!6Sf;A5Nx_2XO+7y}G#!&h-8yy4
zl)cc4Xy)s(kd{Pna`O#7u-$i}_q^UVe)#yx{l4qk*?guIyEZOMQC$kR`%d06u8+ei
zy$!RyxHYb?H?FT&+*rYjh*I`US&hP4DkB(F)U_wf4(XT$*=Mt-h*BH3Ei*=ED`H^H
zLRi~};Prkym%;&Ct2%TiByMWEX~yX>Z!_npe+c%`>9Vq{n*Ia^esp{_G;9nTIvYRQ
zV1mY!S}Td68SX!M?)Kdp+RP<4MK@n|17jRNH>5!M?7ulFmb*}*9;99#)2!cldao95
z2r+>_Ic=_e;98?3v6pH!pPp<s#lx~6nmMbs%68A(Uo-dD{j;gQ_Huuiy*WD&2wqjU
z_a<}_bEUcC?V0z9@JDCDso?~BjaF?jNMN@$W!B~qDaaSU3{P>M9K5W+?Y#siFQjx@
za4ABHv=At+u;a*M<~Gm8=dI<Gs7Rm7DG@4KSWq-C6(<-X4sG-y2y}=kNc7d3zD&*T
zah#XL`II=H7nW7w=@`Y(g@^URe02t&O2L=R$&1_|<*YGinq_L&Zpl-P?c+zv_6iGZ
z`N~=<{RD(WhzlXDqzJ$I<{LgNp1L?Z+6aSFr7GB^vjoQ}dX}KsNNp7t6QN0v_78sc
z#2;M-ZWOk>(Q9FZmE2&<mF_c_=O=dHd|C13iPyjUOX~KCvG43V#M7D6<=L#dVbnst
zzA!$0N9>i$vhw5WO`Lgz^?T%$dD{z@REXUKS*?LFRrXyLC$&_<(lyg^{(wy;wB(%%
z5|1FtQ%baTp<GsazC2)DTjVwtIM_bHJF#G#BcLjOSoh#?>c~Sh3v!Omx0{8_z7@93
zvOk<X2a~mfD9yGbDqF=_c$I0Hg429cY&k@uuATdBV+_Z)Kfdto4<EVTH{7yl7O-pR
z+e(vh-L`SP?YzERMH4nhg-I^gjAO9p!oC%j;BYaq<Z77~RULW|<kM?6TNh^`S5CwX
z6P!&lfOC%08uvX**I_KHHVi*=m?vAeqQLgiAtHo1L8<$)-?dn`vIN1(vKjgV%oJRT
zj5zSE5n?ALhaV6LW>oc<6|~+D+M*&jVGOe>w#?SI&;E4kocrs<Iv?v$T?K%Tfpki2
zZyJVn`=4?BH~)t>KhWr)U&5ge)bv5b;BWcf!5d)c^Y3NCe3LQNFQ5H}6g*4xHl&?q
zY3M1K*9hIv;WS%ln8%8^CpL((Yk^wmY6e)tNi`b14utBQEU^7xcr=)v(?tY6`UpPC
zUUwg&uU>2CQyIKg$W_nZOvswtYNn{kJBXgRLa>4O1T}l><l0%+$g%`#?K&LXpw!HL
z-?(nOyn29hN=NpykH}xBS9~ihapB9!W_N>abna#6L(eSfEKar$bSH)dX_57ZcWTG4
zH}=;T&Odt=W9#(9^8Oca`WluqHLfh{nScC`Kk#3F_!+HKZudL2H9oz*@bUG9%>sx@
zDFomgJtsG4`HrgvM&nO@{uRG?PK?*@D5F#MPn5k-o5MXn@pOKNtB#M~|23cPnc*Gf
z^%LXMN4d7fNLn<b(4eu?N(OGU`$v4)l*yedulF4nBVh!3ucY88ZP52dUm7#nA$Cu~
zQ*)+lO2n(Ru+&Tsfku?|e6LJlsJA&!blx@+PdI1T7Xr_pbqdF`;2bfkKv+^Btckb=
zqI1M3C^&dA>O_dp^DmM>qvB-cl64>`gQFRAjNnY?Y=)=~yw@)8FZW>J^wx-eYF!?i
zLfs2_@7!|bZ@z!w`;S*jX(ZpMTV=Zz_IoAYD#aJBFIR5YjqB^3x>uZKy~w_!i~NE4
zA(c+u8@`M2rNgBQ;_-zM6o1*OQX@+?r|2rUOP!t$)6oOHJF*$y)iU!s9K-9gh*B&~
zO7PyyMTcE0&35qMT{?7QemEjJz^7vT_j&z9_R#{teD`hOvBb#glOh6vBSs6%`LmI`
zoHsQX)Mhv*>Y!oeG{O5ECgQjtb76!aJ8QQ@KHtToK)~ZK{SLtNJFnlbnfhrjr)I#a
z=wbBLuEF;rrL{L4W{C>}M>cuC8y*>(JwDeM$OmT2<XwZX;yOKg&a0B$8k%2>Fyr~0
z@OiW0_Rfh-H7c3G)}%21xNUrY-S~8qz^#uiA9gq19P&noo0VYlq-Bv6xC2I4uzYHY
zB3vRN$b2t3Pj&KADn$ob@`2NmSfV_S<TzXDa#f@KEyH_dWH)l#*={d9zk8A;bxd^U
z3Cqgz?j4KkTxnce<>k{0=Wl=DtM?b;awa`paL?aBJmKR7(wQC>KD=A_`m)mUPHo^q
z(miR0!8%8-nX+%}_bYk3)4Jl7or6F6`Fl?NmAX6jAHL`M<F|A?oWaE>dZ}!mZv4A{
z^>4}5uHmTk`<2tWk}l7j*E4BZ$-R+~-LxBx7#%JpLRk6jcUK*%J|5vU=OMd<52whx
zHFAkFD%FvllT&MvJinKnUL*8CP3F#NITD^gFw#fxjx`D{?a`5!$nfhMf!lm;x<pZ7
z;QcVz&~rL11V32TK&gY+p*u&6&aBNeV3pBaXWuu4AnlFqCZjj+0x>wjyTMbcqYNpD
z#~9>VfT7@|)(I{WVtwo*aBQku&V0J={P=R?ww0sM$@j{3&Fr^AzEpg$#JXE1-z0c*
zLxzx(;xB2RF4Sl;r|V80QF*JBY{r8wKpc$jG#9AQvRrNVJi~FuV_!q=4sF6iW*X@p
zr!N?C(2UO`8y;nYLuZUCgF@^4TsJGC+8zVx->*^lX@^3LQB|$dJdPn1PHM?!ib;mV
zPSZE1Gz;`VFpnb~!VK@^5!9HSfk6;&<_WWFn4+7wtSsmCr@Zz5556=+_f~1UqasiR
zc3DpHP-h%!=vD@`+JZ&&)Wa#Z>rj}UExk9K?*wOII!;W5evG`wQDyZm*28d-J_<|F
zMv%Q`A)jt{UamW@_rjhFtyLbD=Bgau!UjG}?Lrb2*?H)lh6A?={_%r|K5KALoT3cp
z=tV5AsGyr@9Q^6!?=-LCi$j6Y$-Pl)=KlJLHNB^~RgAD0`S@w0U0)g7owDunnA&%~
z`|WS})vtcWyI+0H^8PDYT$O3y)n(M$SX`x7ZT>C<OVvtHE31zz9!d+eHNyKRTyG2)
zAS89b`;~To#V;$a6<)vnEu}S1Uqj$T4=b<Vzwp=pkH6shapU!V=c{kt^YgFYi#awR
zmkOm4o=$@M0+cP|TcK<_@%$D4_Amd6lopG<bP!=Vueh}Evv+|H=MxKqeb3ZBxaUSM
znP&5a(Ibkl?3Cg)-)bH2l4b!oo0X|+@ZOO^V2PgQe|MXZXsl>v(F%iP2SXr)L0W_c
zgyL|uE9A}(gK(8_NS7sfDTRI8x!q;lc#7&GpDzg?BHP}nwcw&7Tkb;Zom<%{tr21*
zxO52e=APl52@P}RZep)7#N6o(@>aNCH*Tjp&JS*{H@4eOzRPmk^<X|n?J$KLL+iQM
zPWOOUr?YQ6WuIO}4Mt^VXY2E2_C1s9LnYnX!*@vyy%p2_jUzM@l`t<RL{s%a_viSI
z5$G-)^M@rU=Ovrnz%&>AeO^E5P<Xi)1`YCH><XuGJ^;!IsBTb4B@XjEs-4&C#_gW9
z0J}cGoF8-q4S=xDZrow-13m^}YOO1tbD*4spkVWG@jU(NpQ=0P@(yW-phJht+-N?7
zhTLOJeT0ed=tR7;5FKk<waOl&HM>{(J(eu>po%$`^nX)3XTv*hL#B79g|FAhH^~#%
zL>q9mMU)4)HW<zvB(qGuLKI`AZc^#$aFu)PX_;nw%4dk9q0Cz|Ol#o84bvbIUBf(f
zM$YW_%EwO|-~V{i{pk{ZdAn{Yh$Zq6DGdb%MNdr1^7USrXsrcjA|vNCYdA$Qo}_wp
zlm>gPBnFF*IG04D^~#$LK%^gmHg8|8Raz<3?kVFVAAgg1`s$Ckk2Bo|dh(RE!G7cR
z@}g9_%LUgfZDg^lT4y2P(h1UP_eVqs=R7r>7^kmrE|BAU%62E5JGI_O_un8@0iu1z
z1kST)iby6>-9mi-Yr^R_+`j(}zBkq=FPq=|)wkSp<+^XI%fh~G<a#0OcjD>9C@bfX
z2yrE(6Jz_t(j`o5DN)9W?|$gKJ4x8)qnv6}=$u#2^V5m*`9v6n`_)mq<8|wV(Z#{f
zJL_pB?i+qh_$u3$F(5U^;+@S2J&PMeIT_Iyj3Do%&t6Ui!rmNP>FhwI6O)j<Wl^5>
z@SSCij4((}airL$6&>uo&}$**joW_bw(kraYl8D7a6Sh@a<B%rTg9V<wr$^OZE)Xr
zO5Sm<5t26zm7e_+BDEB99n><Y*^=jm4k*_J#g?s*w+%NOpT6I?e!8*UE2R#(w+<Om
zwkr4~#z3!?yzTfQBS)&{W|cuM1u(PEnkly{W!t!K8+#L^F1G`bFp5J7mX?DsT2s{M
zo7wElxYo(xAPkl@;>^2h^k{of<$O*&T~4IcLmjx?z0s!p0S3*rHfk^QZO4~R^s+gg
zlkOPfVUaKprkJxL7**x=hp`>=P3pU4u=OS#Xja)l!!tL8y~h+u>%!^D^Rxf?e*-i|
z>D<1%asB?4{kBuHwM?EzCTzgq98=z!ljEQt)&a{Rf1|W2LqUo~4u5^=COoY+_G=+u
z8)d7wz0zwV;<zjcM?(0p4v^psM^A}0CDGNK@ZgjSaX=Un{mG$pnBV;4keexIgD&4E
zI7C6UnD<eL@_}#DvC=R!_Xf^0W5Vsa^YQhS{dUK<!6_we^a#>9ErHd7$-^=H9R%+V
z5z@@K@ZMl7V@sqM_46Fml10rem7E2Ul>v3kT#K3RW)H>Z0&xJdB`9o7vw_|kfA!aY
z!}-In2;Olzow(M9Yn|nCCYMY+uYCQB54`{CD>+3vZ-szemsw+V9hf~nG14>m$XGv+
z)@MS!(r>$YPX)SH4DAq|O@{>L?CT@x{KUJT{ZbxDK2b~Khwne}{dYeQ3A}rH#^HGP
z;fd$>&&vKdJ#qQ!7xcK`UDN<OKCx^sw5{WwzT&%UXI<dk<wR)%pAx4vaatWur^wS~
zC8Z>m(||P(!mZG#jMDMBk!oWZ@{aOdb}C~)=oQyhJ$aNQIBKRhImkkzkMi05m>K1@
zRkmh1B(}M^ptU%RQtBKJVWoc*fvszfw3Ut5?Z$P#Q)(4#v<A-S#N{FqLaVSeL7274
zb+y$-$s4`OD#V9AHw_G=I&>a}${%ozgo44CP_m#I9rC`jZyU8_6(UvMUbSn(AKv3a
zq~qA?09QH108KH%)JETntaZGnW%(`LZ#%c!&c0R3E|%38?szWc3*|^&rjrl-vR;#1
zGfZ2Rg2J?-7d<tYu4TB+x}11_UN{9OwxT%%%<J*{yndQXYLm84)p-RIbw#|ru~}Vf
z^Hm_d%*15B>mD20XEOu+op}L7M>U(^;g7b77>J96T_I?u(59zI(tqo_XE<SYm{oT3
z9hx&(#o<hhVy3f47cfi!(ux+6w{~$D>!5W(Sgj0dwl3k=e8(ek4i59c)*7@x9CD__
zL?pvDU_bPGqOIxZh8P-Nfi>OQO<5RsoQLSaBg3}ZD!F@FmE!VNJ6n}a!L09w9Z<D3
z_TuQJa@&OCy6(BeggePCat^N60OvIw&Z9^0yr14yg1{KsSaq-T_m?QQTt8&RG1Ink
zE;tvtm&^xNFi~8X*KCZ>Lht7*-*@U3c%f3C2k+LE*RAmK=|`R~7nTU?`NH+}1)mn4
zo-Zux!Y_aQ4eR+tX^rS3-Y<l>z<lgXJXD*b#Q^?HkBJ^mxUoUG<H`*;nvw=X_{=M4
zbihYKJOQ13&#WJQ!DVl}Y?=OY<8S`*H{7o~KYsfIr&YNN+x^b_ubx<+K5+TwSGdzV
zid#6JpNJ{3Ch*sta=qi<{hF8i4?M4MUnBR@2r2OXY32RX!n=!LI2VF)1X{r_ju8hr
ziv-|vC6&TD79DR!3m!OJD>$c32^4RN^U5m78?9z)+uv+a9BjG06khH(u4Uz1qr{wU
zmA=49TYyIyByKRg+)#U$cUEnMTfOtT-*~;>*>hpI!IA>&;y9n9K3DF@U8J(-Od3FK
zgwZ(#M_K|)2+|{!VIs&!K?$>Pbuf%JZ;)<u9zO=P={e3;G*zS7HZv4g_pEq;F*>(>
zmo~%?E+unrRo2!uYmPVuW!t&lZoJ;^T(?Z#8#SASNFN8h)Nw2U*68J2%OE#zO6PfN
zCNw+<$#FhK&I#7!3~~%WC&k3OOCXq0U}^b)=G&L{F@MMFCmjl9Z@A#_M1n^_GCmNU
z*;MoJJMXocH@bOzlxo%B+iqakiQt*t(cBS&4NYC|5W6gHqZU9kW{V%-U4#g39gNe;
z*sD>&iXPz3Ls~xvJ0{i<&^G59?os^?kB}?1Yoog(PS63p8biPWFI;c-UOpr4i4y7^
zYf#{h4-(`x^R_8{9CpD0=_HcTADtJSH{(p;t4%iyQ<|EV;9d&F0*d+wWIGGn=-4TE
zJfx+9L%u7wy^yousx<`8F|kYrhj%Q|@qAucPYYp*0%FHFl$l|g*!4!qnZ0O$rfOy$
z{OwIJhy^nR4Gnux3+280?7EG)F5bNIedFb$*fk%Yp84hT2ma(&@A>wc$@#{|A6`j)
zus%KOeSiJT`wve%Ulufk^LU@YM_HS6fJW=ee;CL+YIvCe_FfqK3;p^nVf%qmZZzj`
z%VJfVm)RoVV#4*6(JC&TaF?%m|3?eAzxr!_`0Ypby-FKn45QI29=P4_#IK;?=^f7R
zF04!96dnHdBiqN1Ea}V-FB`6Q-klac?uES-LKvJ*k;^&qbXjD_5j;DE76vsAdV*30
zm39RF&?}+KE`H{7k3OJkox35Y(>Bcnd!<^o$lI5yZBH(h*ZZAM_Y=?W61inc%P6)y
z(0r$R8=88h8rGJfNqe&Ao!kA!>+Qz%y0dK}8N?`0uXmSKzo%zbKk&U!^2Tt3vzIM<
zKb<u2MWVCO8yGbnV?M1Ra7vz2isT_i)EXj-4<s+wfU@n%->w$Y<viy#r~^m$Oou|?
z>AZ4YwF!@|0H$6tId5#cVDH{L`~A-Ce&cq(vCW)t^F^{65NEzs&uilA(}M4fTh&al
zZr!@?$CvaNo_Eg+@6VC*B4SAJGU`kz@qCJmTG@)Hx&Dv<{2^uvy>><gUpt|5RC#F%
zwcLEHMdO=dkZcyT4(NU2WN+T%7{{qN5XGz;dxs%gYZp8|Npu#xlh;e1N@`KYoE!KQ
z2q}oFmZuY=Dr^P2=n!&a9J2=J9^Omd!n4OvQqg2x41=@igPx5mog36DOI+)MyluqJ
z&o+4Bd|L26P}<;@8)bC1R=C&7ecL4>YfY}9J`nwb);onj{ScB(=U2c7_Nt__?mWXq
z*)t1;G|aJ?#fI)tZptBkz3pOSl&%M7O}w-4A?X_&Pv@1C0xsyj%wbn;@>vPU<0e}H
zFgpbo!{OCGnDL-yW;U8ajEo@_X0EK+ozFPOx397n$~LIp^YZ!%-(=pse<D47#cdnB
zE{WR@KQMga^8ShUUw>d(lo#xgZF-0ccwzj`1HCa?meHK|`fo!81AWKsujJS7iS?D}
z9k@Viz?c~sL5IWW^x<j4<3gf^j=x0S|Ge<y$1CTj3)lR;98gE+>9laZobk?aefh*#
zzT^ELJ+Z`q>lxoV_uu|oS&Xkw<b4<X5>L3GHR*%k&0zPYg+cX$;-N*)NI@1Stx+S~
zU1b}Ur|}4eVTfhihpZxolTXet?;h!1=3L=RKyP+-(xGHC!t0~kL89=SI+|vXPMcrv
zN*8oMu9f@!&i(bq_0yf(>qaToB|BK6XSE#oOZ2=;dd|ZrtS<1BB2Q<J!&8IjUg2(Z
z(Sf-rd$Vqr1sJEG!_e5(1UkM~T=h6RG<)4iuJiRN@_t=7C6DtmEu4o!ToNIMDN5!k
z1)f4+0eUGCgSJlJGvthKm1QV*W=98Owa}h{BaF`C93PfI8jch?_fmM>-Dl>Io<a1U
z=ji!r@qA^6f)43qxh#P-I`-1>UA7y;JfwclGvOy%XS;wRFE`<Eao|_w-lh;ALEg5B
zRm=9BxVRu1C1c6ea~>g&L9b&nkBr?kOwi<|m{xpaAmVUkfU`+{7PG<!dGYiZ%#Cw;
z&AY?Jbsk8449jo^qRlRxJHE^!1sH;>Mltc7Ou77cux1?YunwOp=GF?(@m?F$eec{$
zWovND8`}&i&jrU|4UX2mSpy8V84K8l!Cg9dgA<?qz8BJvD6S2)=FSW1Aw*0Kp~Y7x
zLkuY<%|4hm=wYf_VMzh!rL$O<M6%gRpNe(!CK5DQX|>wHUla=&1U+OVj@nyi+Z2~c
zAA~5#F(jG!P38K$FZ!%sbPA=?T42i?-+lKzKRjy?oWA)5YkJT6vv2rtx+r%bG<=LK
z$y4){j)tNgiQZJ5xnt%vTC?gu(4D+{dMU77sb#}Ct&`SNL<Ar7Ic)~Qq!EWitvjcu
zcjUbD)z{x}`{kXN*T3QAe&dgS@hg_0+-2|Kd^r;?C!U^8e0cXv*>BViAAfvdeEcox
zn_u!mB>Ouf46^gop*)Jk4pf4p`)Q>DEe68*#M<Dr%R48AL5+?QV0d%Wod>pK;;3dX
zqD@M?{Bg_*z62cxf?B2l<rwB0Wrs}b4O};le@AB0<51gzA}z~yr)-)1wz1#TK?KXs
z_G93h^e#_JqFhdl)_FcB-klddtbunI8Fu98ad)f>(dOs8C(T14I8MpaMj`>}v`UQV
zwIRD!O{56#)_{9=W*zYLy7IirV=RP7ocBC<xwH;t9IPqv6eB06KXkd_dn2^QItI_)
zv&BeHf!#mm2(62z?L4f`@f-uE5O8Jik}H9I-YOg-a3tq=7b0K9$Tv&kw5T%=X(26&
z-}8Xq3voCV&UQ$Cj${14=}@@7+&HC$glCPa6kV0V)p=Pdgb*aK!bzi0s~L24O4BcN
zKA@48;oifTE4F#dbj}80Au<~HFb;^y1Eaxq+x7oDeefkDda~I@a5U%0n{*YU4b~ud
zq;<hYs$6wg#2{!ub+x*M*)<;;i4>6`o3(Po(JKdUd<+T8oQuNTebmsmHMaXkcd+%&
z>)y$=(T2o3qi16D)Gh&9C*p!<D~@}0+)EZq#={I;uhkA;r?RK?2NYJ@Akhzlv{n(b
zDSAOiubI4c%DqwUnN%AeLZB>-doL^m&TiEpIGPm+&;o?s=vkD)K49B2uQ!R)YObW<
zsm_yqq~yvEFPTrbOx_D_n2TqOJfD-+OI!#^-bVu`51TJuE+m=@^nK^I+l~Ko_-p>j
zpZ$VA-W<(4a*On|uzvHN)bE`9z?Y2c3NR5at2sz>gVF=hf%ApdR|aRu8F${u?L^qF
z^w+<l?=S4m68Dx!@XZW5hw^~C9V$akl`gE{gx9-p;^)8oWB%%|zvsXBFaM|f$*=#E
zfALTMDea|mes{tziRJy*y#Lj&sjl(qhaY$?8?NvC@z=j*^9!H$o!b~}qjK|w-9h#N
z-xO}soa62SBgmkU=<#l0?VWE%rJVFRFH6FolN@usI>?{}OJfM)vFfIyE}>%qO3vYv
zh!e;Ca1Z0cF+8CUQvH9?_HWOUBuSbd_ER-8_lP*>WM<u~Z!^<#+uZ>Nm&F1EKoJT>
z2tq&#Uns(VkbeY)4<LmP1Z4rr&MuY!me{%VPEYrAcU9LdGtY?#cQaEysG576?3tY|
z3Oi(KDy#FH2oHBNQ`Jwu@6$?-eI8>A{b27NoUE<eK~L{YYvr(3j`hl&2(P9>IS8kt
z@X;~xD#KCm^|4&Z#bC<3I%Qs-GN)2FWKUqk8hHsSEg^sn1YpX_%hLh1&eQdRiQBz=
za6j?hsc<)EiaK2UVlGTfw4-Y^PNi@xzFt!TN$x~2Z2#BZxz_5#q*5cb3o3<VN^g}U
zm81(PE%X8{36H8kaneog(Vr6z$HFl~p)qI08~mQASs0yU6AC^*{iMSCQ)ZqE!>Cgt
zB6Kr%H0b6qcxxW`e@&snw+aeRj~DvXxi5Zb(Jc<xy<rG`z+czJT{q+JHD@$r@O~gi
z00=H_itO!Hm*5LfykXVc8{Wc-R%_kd05#0nM1B5p%GnRjex-}jE~}648yM<n3MWOA
z)45W3`sl#^MM4*Gdk0$B54%0bXt-^DOfvS456)P6<I*Zm>&hgatzMRer^UDgOTj{T
zlA2J{5RM5cdIQy94K;HOTkBOjIUADMs;}0<qPthmgmcHT4}13sTpB{SmFuMj*6sbC
zM>F1}i$6aFon=^2c{4t$Hf1P08>>atab0=3Rvw=QN84DIg)Wu0th|2v#M{R+t#zNn
zT-=holuXV(h7;LsQHE&l*=t;aCLXUVKmNrp`Sj&I-~ad%PZ7*nPSfO`L3O=QyR&C<
z_OY8aNCTP=AenXb4crAPp6Lr--7S|l*mB);G>*r*sadxUlApQzdf&6rLghS`L75A`
z_lLjFHQk|I_`To#0bl>rC^>Vye+BbBZ{MD|*b@(@39XeE_eY-U6K{Lt(K;7_bMKt3
zb2VeJj^ch<E0Kh#21Cl5r>8iy``pmpIcDWFPwtj$Edr!7={zax;levCVF)5mXl#7v
zY>z{MqI(Z6btCmg=_?bHCw_w(9z8ti?%1T!L+j>EUP%SHo4ES`t`C*Nlz8veiTCb~
zOq1J&uT0Ew%HPiuhmx6d3cU}Ab?Ul=ua+AQ4lk#Y6LKi1961zcFwTdWS9b^Q=fs@c
zfk5X%663x&_kSvhIeDcq4X5RRb$j=*!c%Wd%?%l|`aKe><j@;5fs&P6GMC9=ousjt
z@g%MV#k_Z;sXN|YYUgQn7h(sH!?bS&ETeev6*DqIQlA)UAu(#LqumKC^ca%V{}x)`
z-|_lZLBWUL2<uX5IicD4$is3NVj<zfUT$$_RUf3Hqu*h>r%np&z8iEKb{D<~dkDE&
zj74l~W7YMhru8hR_wMjg(GZIHz!%a=&k0*r_k;<~zKFQ=J|}9cXcF>VNK>ZFIh<Sl
zLp1skYpe9OBI<*{&4R@dG2yl>_mPnumX&@M=2lQq9-po}R#<A=bQ)qtN^Vh+ySSf}
z`JnP<f|0{?%-v~PS69?_#}G9aR2)xj&4Y}PLA(L&jIUq{kub~d9bf(Z>nc25Dmh(|
z5~EZxD(<zCLpe+9NX{6p30bP~=4s)ZHy7T#y`s`l6tzwT*6YI4W#xQvvC*80bJU}k
zASxR!yY+Tdc{e7w$`dZN@z{+wmn(ny(_ixYKlmNy)H#%iDJQZz?J|k6u8qE~lrlT>
zX<d+95uIZuMRAWFOEjHXq>x&NUa9pA2IcV5*A~J8Brzv8L<QEJD-A^E9o5ryS?Rq|
zO5yJQM6U&_Mm|j3y?TkxNAkl59N+tpbUL6}nWZuH#P#wGzf|Ro8Rw*|S$U%Kw!>p<
zoLfUJvsh!X2BZiI&L&b|woIPDq>`+2(#+jlIF#fHSTB*c%rTho5aK5Rhh<e=kvjro
zycEb@-g~y&I|n)^G42G1b{WnEQQeD@&66Gxqp8mlKF&v8yi9!O<dX=qCJu+0JWsJE
zgsz1e+NQa<mO_c_w;QaLR#)GfNSG=_k5Rnupk0|Ilc&Vp-GRHq#3?Ju8rnKp+{VQ6
z>{jtvHY3SYVg~kL-9_T+8X;%)Ic~lpO?@qNC;%$tBAl~BvZ{W_TihQXR_1-Wr5lJF
zT-F9A!WdjEk@%`Pab*%nhWkEsC^)UZ1JZnIpj2;cCspq`{wCN*7>?275ReQz>j*;b
zYZBbw?s!|rDPo!=7%RTGYOO9z6LHV}B%9!E3IB03+lJt%$Yz$+yV2I>o-IHVC)nl5
zSFDKZ8IHj~XsNPT^TwB@jb2%bNAXOe947K%cF3jd)LB19JvdbZqdJZZ(IB@lkz>nt
zy}=rqfL1xLm8nc*Rj$`7muAlC&OS$1G`p{h2o+@|`G5>F%oNYn*2uDqR*f}vXM(Lp
za^yVB2*>^v=#DvueM$`burU0Kq2KtSy!Os@ZOpZWBdbGyr8lP==Y6114slIymCNPA
z>&GYFp05-&pJ3PB0rv6jl{c@ioSzoQ<TIoUxj2PVHWFGi)W<!1H>J@SO=)f>cCC#+
z`|H2qpZxKK-@1RG_DWL2)(hA5HFNT;y{;>&S1$cE>0x4)xTn2`GS!??Iv-q+W}vo`
zQpXM_EESrKK7><PwO6DqZY*cGYIwb%)=+%-Q_k*XV>HU7<ZI))8h0;0#57ax?zsQ(
z6Yd@kF?8ojsaM*1K@85StmbfuwZtmI1-N#v77hYUj80g)4>wH=4Lr7}izKrd$CNmw
zObOcfNGYv5h&%3kiv7~#z)5PJ{JrB*C;|RDlg}-*OBW{Kh?}uR*&l)|7vf0-Pf@%H
ztb(OG`pKO>&7u+yHezkn*F_RcvR6U~(ki{dnp~{bTaDQ4Ye%E6msG;2GcyT@xgXi_
zFq0ElTUpjl9X%~|3M}gPw~4v623`$`HO7!Ew;T6)46XI8A4;(AWYZzI)9M&Q>O+r=
zF(MTlF&VS2OUWLfl5m$3tF@@6GezN45)U~sNkW^UO+3(=)9zKAP3171ZEM4+)W3nC
z;8xLw45-NVVc@_v7^K!3d|y9cWHgC~rE=Q`dkp2UdQ^eP!1{ouVi`K~#~Dz4=q-cN
znv#SoqT|sbMpC!WEmJgH+vH8|@EDP-<kQUY;Fb;3;XuRP3T?_h1r7g>(p`jS*ARsb
zj=C2fi2CAB0X<n9AYzGStyF`hU0q<(oWU~&`{XG{CFS=aPh~TtllU%vSb2hH!JU@X
z59(UOL@&w4d}__8*Xj^bcOMoH#tuWR@qD6X=mA~a^T#J`$!;wY-)T&%a>1H6*?V2M
zE*GxLmGkuqlTtfw>{X4&^TK*=?#r*bF=kQ;SWa9(*uq9&V<^A#lB8IMX-bstc<Fg<
z{N$IP@$o}J%84``z!ql|(UBNic<nw9d%8UF+1GF2c*h4Hf0tYeCceh{Ncd}`u4lJW
z)|s9rM56JDLB0CSgLd`94RghCy<)vO^|PBR@^OoWT$Cb8pJt8^FImL{BlW`htFJhw
zN_+g$8~^h?Px(L%+YRjky<^tHYQbmwc8jWSYuq|fA<khkes~s9CRGmVxOvt@PKjX?
zZxWLd&DX$~Z?sjq;{65HcOFX6dWT5XY7B7koXvGh0WYigsM#J6WP(l8WNG$U=w?V*
z<%|)OYh75@n6F}fKqG80NHZ5{w$_7T<JSS|*Jq*lcm6S1vrj6^p|er<WNP8YkkoZ3
z771pz=qXzFbBE|H(lF~jce)Q|kMDmaU@IMFmI_u`*2-mBxh$1+wcYo~{C*1(d)pa9
zcNspppSYV6y}2_0DU(%rbu2tgnK?esAxsZaUhhQ}7&i68UMR|qU+bLzc7}MRxY1j0
zK&Qk|_c6+^i_ZgISNA8WYxB|yr+Vg`yje>nHKj|B<389yssWMgeTAGeqVBS&p{(tO
zvuP|uWmVARW^s8cj`^|&9B+&XX)5&9MR(~iF`tec4uxqh<WjgSt1q6r6Q%f&^xk|-
zqqI)7)yujr0SiW8gFPsyWA`<f1$A-WG|e-Ogm~6eJ9&mu9A8J=Zzh)rxV;Apr*DQ<
zzAxP9s!>@ap^_+Kk%@VBK*RlLx_UrNa>EqyzD#e;n_>8*w4BpM4NaoKXvlFt!+xZD
z`RpL5dlA*;%33e9wqQ0RIL~`oyz;o5uUObB<dio$UyI>2wF=AXsC7=hu;0ySJX`cb
zD-%s#GybQ)_%nX@y@`)L`5tqD>*_v6E|e@xhdavMiCTrf{?U*4v%mZ+e)tdnAuk{9
z$#%keb(@x$jUfo_dco=nk`ZxZxYm3aOA3QITh1}&5NMUYUTEursjFp2zv*~5(ECbG
zD?Lqob9p4?#DPk_yk*WSdU*rqulefniT=UN=hqzbw3FMmjKe3`d~J=P^R3dWAl)l~
z27xnsC0A6LlXLWE95&JoCh|kkH0;uC%=#VJkwZWgwt+j(o`=P(fYuR<na-_aLCH~E
z!{iBvGYN-JmKt##4iom=qeqvwYNT63ua3!9@wv%H-?LqFSfw=|c)!Lt&o#&!nw-TZ
z-p=&j@12r-eze6s2%b`C2E2hgG(kE5dOA(|7AzEtGsmk0BEd}#pAZD3u(h_B;o`L2
z>)KhG_alaBj^M1$DHo1&;cjwv<^6GX7l?ZGzN0!F_Ef}0rLkYf^XP=#`z<`!`Zg1#
zGpg{v|Ml$ve?Wl0DuvJtq@)xPCRK`nnBSyxWWkO_hf$mD*+{caQX@|aXv~BaZ}w)d
zCIrytmKknpCFIFnQ8fi^vAKB6r=m&N&k{rH@##zsU`9i*hq}_U0mP|pNu6CIIZ^2l
zLxM(=%Gqbx13~UISy)1ZMQz&vnxvVM!^>TKn2wq7=xYXB>gEjFvu7-7>NBLxGnP?T
zuQB*r%we7y<>nZ(=Lxps@zy$ZS;;9OYKVHfu?6LK7zm0v2JV^wKm3x!3}7)zWRg=#
z%sH~Dlzh@0u(V3<PQP?0ty25yob*=d)onWZdS$sf#dBFZ*JW|Y&$<u!k;JTplHEOZ
zX^m1l>AG%^f$*!8RG~W4%e(?B84Sls4|jKz<!fqxLwj85*9%MO9HxmmpS+|t<9K&s
znos=X$3N!Nk6$vq_Z>_dje_Qv2*$ELv0UCVnZp@1D9evePqeynoP-3FWJp_Ct}8_v
z)Qdk~>nu-a%6wp5tHT!<+AB#K563$eF^a(HFrmvgoS%#~y~o@1g3sPuFa()Az)P)Y
z>!fC+bwygmR&U_bR%T1wch|L;8H(GFXxJXi-N@=bRw{|CY4h+OI({dwsb`#V(2s0g
zdn8>}<A97q8(5m#Vpy*%)knE5*DHOQ&|^n!WUfx7CgxvA@&r*-XuU#MY)A;^`l#GU
z-jvPJlMG|J7FnKl4?3;;gozLvxox{Sj^0|snY%u~sbnA9T8g@86d&s3vyb!Em$BX>
ziw`o4fZ~<f(xR8rd)S?L<t*m=&s)eNJ6Ei(v}L8Q)q8#hXpgGH-4`U_qK8R&|9&Rt
z<TK9~sLjzA0VMeZzf(4*?7bN=hl+ye8;qkK5UIPmZmcYx@2mem1%-~qSV>g259whq
zt0|m2yC1^MVQmK}B*2R94W^_W%TtWT^Fu0$Iw4dPOVC58Q%WXJ*;7S8YQwB<nt>Kj
zmNS)%rTgab06;Xgy+redv<1}Sj*~%1+IrXP>OGB|hMT5Ohq_)atd|8_S1<J>JfDK<
zDglWwFny2l+pz9|mJjAt3k^py1B<~iggb0_;ID)KFQe?%tpXZltXDI|H+EDUDFU1i
z^ntC>9Mw%c&_JdA>NGSUNk<mqnWXoK465JjrB}|&m8Z*<%d+}Jy?_qa>a2+M3hUK3
z&af`gsH9TTE%1)hrqef^w!c#;DDL`rwbhv~+3(fYuV3?LfBs|s==&e?bpDF=v@p*z
z`re7EPb`;}pM25y?|<|Kpa1li6iqyQbmH}!H{2axkcw+&lKY#`mJ8Aruq(Y8I#sUc
zCqDn#FKEjX(yn~;!96b?UXaR>sT`@7N6-~ruiJAS*?o}3=&2*AaY!rDpLqP@Gs?p~
z={PgJ{E#o+zTxLze8wV)!>s7)6ZfPwO6%m_DXl|`AhC4HwJ|m0pvsA|RhlZg&125$
zu!%|<xp~-2`+lMMv)XVc3zo{j@w!)K?!$j`pqLHb5|;*Vubrn$WxZBE{H_)4uIT`w
z)f!33V8J<I#j8Sx`C4o7)`L*N*6SF4Wl0FL%3-sjl-5XP;fa-}_T&(9WXEFO%TQHv
zPRvs#mkd$mZR=7clEm11u`UPuN(JxFH4m2it9zYc;8=3N+;OHkGtr^0rF(y|_nZC$
z!=m0iQB~nEWsZk~pQFd<%XUw&jj)kPK0Irid9png+a8-mGH#K0?>p`9e|^iIf*<yh
zvVY;U2Zce;YrV0oE0?;mTEk)zS66WpwDqPOP3}u><_$8_@u0kUG7QsHUDTpEN<Q&y
zd0Cc#nQ)vm8r2>ly9>@nwx*sP4Em@OgBqxNiK5=|?^fwy(srCCa!#yEW4$)oQnA%-
z8$3gZtStP1#)0j5o^3dAi{TbYlxDuMTQ`;v!mQoBn;<Y;^X6F%+N}V&l+B22z^miI
z*R^fUbxAS!B&NO%_{`~+&fX$2p%EN>u@PCNGN(d`fI9jowYoC3^~zEg&etn%uUD3J
zbtozXc1=JnVQZ&d&Ed1&Av-*nv$%D3c3Wmw3;SlPF1~nY<M|b8H?sI-G)u~2oiE=!
z@!$M^{*)J|%!hYy{^kp;OicMBR8IWspZtRV)BpLO@p+$^A71g<U;Z`!$N%y_@<ihL
z7J4Ti96oZ)Zz1)Sc6q~6JJ%<8{ms{W@$<jor+@Wt_~`wY9OP5733qoVat;N1o?-R=
ziiQ#3<}lkkeSIR8j@Vn~cA=#!!c5OEc)K)y{-*Nw(m6~SJD-^kM?Z*6$R<=0XA?^A
z^rg{kMbC|CF{W-DN!vj)Dh2CK;t)%onfV#ymIsQ%3n9Xc4YciLA4|JesP@iqmaM|d
zLqdA@K{*Ysm&STo{GYO5)vMJcUd2mFfLGz&V&<}nBhirrY@Sj<-p~jtqKPpjx?xUo
z(d6cI-AL2I*`9d2cG7yGn!_eZd>>CaF=_a4Wk)I29wRK(ZF2@gawN2yH3SlR6nFb|
zXNFD|R~M`Ja~^#*k#K<)_hzf=;d3>&_837SsnSHjTAXp>b16jJh6JIh1-&%|`J)0(
zlTIo2ctBj^eFv+lN3wgHjs8X|1;v$yYD7fbBe~;@-Y~4SI_XN4G=+V0U~O5H+M^l>
zJWqlZqL{--mOX<B?5(v<=|<K>6E~$(vG62sTZ77Jn~`z&5=WZT%e+jnSbT83F46Vx
z<h3HnZ=^@|@DiCirMSE2a$cg5X4`?Pu|ei=Kh4D#Kr>@)E$E?vAKJdroXG-RuKvB3
zh6X&=Q>KVRG}e8Ax%=#mKTz+U$;P@hb2gDk3Ue%P=nq=2|E_FY{jF?+U}u;QFU!hQ
zG$v(v=D)0sl}_sdtFVf()|Jb3;c}_;#VLJGeFN=^)y8sZT+fwW13fkWe=2UFah$UA
zz}4qui@Al0j0bf4YVOgby`feSTYUq~6JIPV|HF^}hX3>*ecv4}&kO19fiqw6)6c)+
zn=jw;ogaM2Uw!^ph!%eSi(hhi`T>`cIKG&}{vuOyp$UjpE|<s10hROHC;s}ce#SRn
zFZ}$opYW5<-tvb({2qV!gAaIdSC|e*Dg{duDP5_{LQj1gx%zm1V*UEdX!OUZNu4P#
z9(d}BAOHO4yuK_P(}A=s%!k74Eh6XkcUW8J&^pr!*Ym=g1+M3YtzNwid-mZP;QhJe
ziB>eEEWqwA&3&<q*kcwMrzaRaW>WXa*n8mlx34)Te(>_d>uKTKjN_ch)_p#~+cR@2
z$l>5lz_C%Z7jk!kfmskJ#IWYj)L5&VJ}E_lxFt89-4yk6!^ENB7CMPC&&-ELK6^8_
zws2%NC?#>ovcaN$6pY5UTh1r-X7@NcHz!?lCf*ur+^wyfQ=r525sL<|4JM-O#a8ob
zeyMH>TDy<fOerx<LbbgY>3wvs%r-XT)_jlWoRZ(Wajz^Y#DEaFX_SKpkYA5;w7<Tc
z&4n3-<RX-~3Aa<ST65TClz*aw+Ubavn6o`1338STvMYzpMK^g0SH)E}^rzL%vQ{K(
zJTxz9J-W9z?>&Hk->9~>XksZ+h?2AJqb-|J7M~}aCm+5-$9n<-*Lr3Z_r$4J#|1}o
zP0SfRIYp^24s&>x5fcA#2-{F^;F}M(5ukRU`T@{-^f!VjWKOv)c_Po*@1uD@sI9sG
zR4`*Ue<twKyo2>omU^O2@7uhIzA?s(DJgd)bAOn)KTaH{Le8!w=y((F(&4%qrrk$Q
zt}EB&%5uG8Ya@5OI*I<$<+`$58tXOmBs!owKeS2~4#|5MMWY!Wf*_B$?=+fi*r0$|
za~Q>7O5Tk9t6zM{cRqT_?|k$rU;g~(yt=<<W#Pk*UvQXa?#jeJ`Geo#7e8P5#m_(E
z%g=tn2eUGt3Ob$0$A>L&D@Ly?S_+rTnKy6V@^o$d^MCat-drko@4e-}`hWgQ{x|>8
z@AJR>$G^vipMFH1PawvWGIhE7A#z>F39e61EKiTr^@;WJM8mlI_;<KZ_biuyA$uja
z?q-M8fGe1rauhRWl%pha71nB8>&ol%mCJecA@Qv_bPIb^5xgJ8O^l7?t-dqJ=yo{G
zpn~82^KN_cP!hlU!2^%ig-dlCmxno{-DfXcuPY*c*trtlD;{?!^`s7Zz@KA4eFOt>
zIHwKRC?5RqA=Bg}5)&7mni^7DK$AO6;;x|nHFHiJCZ&Y-U|E~HpvHa3>g>4N(VxA!
zsBBQnSB$<^)Qp3A!i(Y+3qTL;fjMa0J5K*z)@?rdxaB?0LF?w8aijZULhGJT<~+<n
z*bewX#?e4jws|m-oVV8K@g6naU!q4c6e+*X>)YuRnvt?a!zzWqEBt@rGe?wig{R*f
zktl6CK7=P{4-(M;YhfACTct?$vYhN9n1Jlob!F*|RGh-ORS4{5rwmp_KuWzCsxIo%
zl$`F5&rI#+G}9>;4#(MXS83anFo~>5RViim5v13ZzBFX*(WG;Xwq$QKr{q|7>yE8&
z)8?=!>!Yz8<tjMb#u{1zkw_8!&sh_tWYXm0PRFL!wPCrD%-=(k|DE4Qpa!Qtl%1_L
z>bm&!wyPf`t#^vRi_?)0U*7T2s|W7yW)6|Tdn3?TmX))0&QzL2)>Id#<kd><zNwCr
z_cZFdat@Zv$N(h*p<tw;AjO$PYPL<BlJ1%8yW==xN072|m?jjVWN)HLcczP#JIbp&
zzW@G596$LET(A7redQ1S=)!;X&;L2oq#W16AAJ9wFFyZ_Kl<I@;&lHWGToC8cPw>{
z=DW~ZWw|`^^I!auKmGHc@K68bf8bIxFFyQ~|LI46#r^S)|LcGAueiSbg8%J*{-5#Q
zhxlCO)#o91gv;gPhfNZG{>>L;I;T1F-opz%`2KIvr#r4+eok8}xmBj_-(Qm3SZqwQ
z-k>$-q*p(n9$Vv!H)p=S%e=UsxtnIMyd%3G`*8&S?O2b*X=n268Ru5*_r608{+G^^
zzj;k5@#)JG%Q`uQ8&^`#(}V~C>$c4mSd%-!wq&TJa3F}a-`ObbgO&|x>yZfJ*4upq
zi0%#RoY3r(>r!%d(ORoFWdoiWy)~zx24T~BOy*nN=Gt&Cwd`)s-jho~B5}&=wXR%y
z!`8|vCr$?+awO`9G=|&b1XD2}lxWd7TPmrq-P6IiUc5)yS|e%NP#1^a-DJ>I$dN4%
zKd8~bZmoHB5*0~pW6$`rZ81;9Y(E>hJ*WSk*SCX@0=G<>NV#~48YXGJ7?s|XwYo3)
zT8$b_JhygDXm)^0Q+CEdZGpYwO{(sgbV^E^@>X5ArL7-k!-hgdgOMW}1m$c8UubDy
zgt&(}-G`-kBiA>f^+uD%(iit=5g;gMbzMDcNIvAHovX8sR&NN0zqS7{?vY3O&JsAS
zi&i4d)p6Z%kjH~2Fi8uDshMZ$x%gZH!LuUCIT~TB+dMnW@6bVf|JIvZd$+zpORYMs
zuU!_}WuY$}x-b{z<tg+2ec`=3<^HZv%7nEV4vvZ18t3)O#oVN?^%bFU#8seq%IGZa
z1-~|)s_=+o(X3$6<X47ARWK$t7*L`6Vx~KR;9dAAO_^z$VhuV~_Sl1Q5*+iF_jf$J
z{~gLyXgQ<lp3=YJKl<a3_?usT!ry$kaF{Cp&Oi8Ve)so2;_ibF(bK2c{2rDSt_xY*
z`WxWQFaDaJ|C`VFr~m9v`N`|X$M4_q(S7B;Kl+&C^nySCi$CQrfAJNc{o?05=#^S#
zpFr1EtU+#-^V1V%jY*U@Uw=cLGP%r%bhLCX=L@AdEN8u9*3nC&C*1^IKBUS?2bNr@
zXjOQ9hS%rD<$UIPJ#B~WuzUe?tMXWfgPJHYE8*G9WeeJE_nGRW0c&`2l>76_EhXjs
zm-jy7BNB{(X^#e|je;)@r4}2d;THPV!-XIcJP`>e`xg7c`w0GhMt{L81d9gisCt*h
z>F>+xP)@?D7cDv9qQu(W=2|TJCbh9HEi@TI?Vw>9qVx?ytL7}ZR&Dbga(CHa4O-uV
zX>C2Is&46`CRB3)??K_^EUc|?SsPa|7727jMv%0~^gB}5=x_(hAbT|j-hd{0n9fNA
zw-52^(>iqX+3aFOrS==>DF|+8R)Uh6gBq#ghGt$C)V*01JqGE`dGzWHC)a~S#W5+S
zG5x3w`(4F?H88NY!NGS0p+{o9v2WNqX$<mD&{2nlMhh_qfuq{s@fue%YIR|iXY<aO
zsEapty0&QgI``Q`HV^Zh^rqpoW3r>H^GlInKse59qeKlTZDTCNJc5qUgQvbMen=K^
z>s}xflSBA2GK^3nuRg@Y1EnvbZBWk=%&E4b)|s+$cbIrM&Ad3y9OnWl(Wtbw)0WN}
zT7=r&S6}H&3dby*a*BaEnXJ&XqP6plCKMm4WY`wK2U}}6y=Ijd7;ixX>`^tmo0(>&
z@NsG-Xv^`SQz_gZ?)c#S51Eb+u2_~to@bWF*ZlVPKIH%KU;HQh>%aUtYL)-s-~B_r
z_q|UzzW5XlFUhBe5G_`+5$1O@Yhcyyo$mOnFDhSr{lwdEe#v1L#C}Q2%76He|6S%J
zy#3;HT2g9t41AhqKPav%>-9qFLeWgSE<Ann4fW~F!z<za4_@)rS6|Sq(XI<Q;DNOV
z9I2hUI?U;f&yI>Wug}ZEn{(mK<HG59Ml~}{4(v+Yn*pT`2Ao3}VYH^t51HO=gMJ4a
z&6ziGnB1Hvs>&3Dms0VpdJJ_MW|8B)2L#mGZlS#mC_=W+9L~bs3|jNyH#+)o5N?pr
zqxaDlhM(86EE`L2d}fM-HRVDJMr3W(J%pNP^{TK`cQPg=s=02+NmniC^v=ji0*92C
z!@7U#<@LSqFtYz}*+9hS+8?^njy{5SsqiLeUTY-7vHv&PdFz;YWwrO=!mBDCMGZC*
zcyc+OrRPXK2x@6=sXxp=zs>7g>lA!aSVzq^i;5Hjq{B{Y>y)?FU=8`Y5+3?M9F9&e
z12KgbqvyL*@z4km12+p%mQO0Gm~ca#^g}j@)7BaiEG5U&#-QvCn*)c2EUJp;4eQiG
zT|BahAxS4X|IPWym_@%tn36ZYk~*Vd4J+xou@~^bl$>cKfnRR(5u#7ft1p5|b<taO
z_EBCd=`eW{T!LBB2Hh^2_1&3Hy*txKtKpcfFIe5OKcTDBb{mdiXJAgsVM-jP<ntKJ
z!oI-me<uu0aTu6!NXlvQ!6452&W_cdugKE)QWM#NnjQz;ShPu;S)6dCxOStf|L-L4
z>F7Ik?qMplxd4;<RH<J}H)EO(eD9<0@anw}uyk<0w=^L+lgdo{`U`&e(UI@I|3gxq
zc<=pJ$n=sWzsG!fADxblv19(;0*AW?KK$gn6rK31&))Fozj#a22WTfQZ!i4dI}`uq
z|N0;EM?d_CyY|HL<{NUJD2GhSnSAuYH~Dzv`u35>{tNE72U|#Vu5Z5N)%wUMpM1dI
z{LN4J`pYj#k@=zu#fJl_x;h;bMBExA=S*ki^-}rbsq)_2#&JHA%Y@+=^=LK^Xra}<
zHJ66~Eb@y@YvX^0p>`hgw!7l{_nmXVVbaQ&i7XM|hgsyn<_G1I-UGc2oSOk>1UVxp
zsiF=`w`hW6PHu6bYxrV`1bM~R$zTm_<2E(9F=lb9Zf{-zX$Ur3t<c@wb`y%mcWAAR
z?%#H;MU{eZoDzqUoH1t>)6*;_BRT_bz{-PRRf?jTedamBa;^Nr+%2|Sr-t)n76}7J
z*FEVCJ<uWYzCEKF0Sgt1Iy~LtT8UoDMj{^W2$6*TI<Ie6DfF0DE6!&xf$xunG8hc4
z`82M+Ivy&an2d@CE1_hTpqpj|a|fTH5bU+>hpO3*k2P;lnFu}YLD0uYMrDz7D`^?O
z({FBa52~P9WG|7aSKpMI6>>b1)?Cg`b1{>XPBnxv!!6RId>_rGun`#Z0yZm)pu3uR
ziR&9NDy#8b0|KbEp=%?@;_Ka*rzsj^IexzrwIa4ws5Mwtk>TLfJOsLQx^zs1RRy{a
zJ6T$%p&%)S4|T*U(i^!~?!@gba#l`h;#d-=IlHoQD&&;uW}F*5riEbyX}IOX$f}Y!
zUDk&(l?@WW*m&qSwchrcbEZzSXPn&!mXAPlcY5H}{Y$3l4oL^3jb@#p<$%yx9>3)N
zy-%6$?$PN;x&Mgi<tK2sb10%N&=#y!mR2z-+`agKgz}I7eK>ve9e(oJ*C>roKDg(P
zfA2f|;FEi@UBD9M)%Uo2|07CKYFqt4RY=E|=;_RSI`aC<&tZMyn2ek{wmvf7eZYIK
zUh>nw{;994wJ{y0V5oG&VjcSeCuQYW3MDm`-gvWA-kclf%gWPbMZ_6ogVr!A)lEp+
zEwb|c-tLEK^t*O2IufH8XM6K9Mn7U_x{Qq1AD?XdN}`uESUHT}+WHxN>xH<Dy^p6<
zbxUr7S(8R0R!UR=(PIk*_1+lYT5@qXdrzfRPx5q}W1))Akyz0?IhDA&2xxIGBvGai
z0ghQ5qbIWr>iAmsDn)jEgATf@7`+t^B_r%w9KlFxOj(#pVk(&_`xLbyBDAO&{P_)L
zQPA}xv8FeK#M0dTcnzY9+pfe~_%)jf-zq3%fk_?b9zH*ttL6C6SXgcFn_Fo7%wZQ>
zEA_HKjMQXy=Tu1!+ZbWP(Ak7cp6%}x)KN!`;oyFc43oY=GaDl(C|yI)<KE)7ADqN>
z^*Cs{A71V|uQ8{&hpUKfR?1`GPs#3C-+NdOw7zZ1k!`6u4z49{0ci9aoT}{bN$-M5
zrwv@VDVjQK3s|T%^w=qf14+E!ve5=(Mr>Oa?=y(IFJ^UHUln(A>{-z{Z@x`W*Uszn
z!mG=Iq#TXbPPYnuCDX|YQ&Q%fxSuDEbLJ2M!8fVTji}d3lASf-b^?*@sW8RB`H~Z*
z%;a1`$!(icswTn_a(}4`B^402<me%I%DHfVdf@JGPf9a3A4!&|PZvxYRT78O9n0Gl
zEeCXdK*|F)zho^lDG6FD*T=WC$2X+91`{rm=X(w>62JF{FZlSoZ+QFkhLVj#sT|XU
z)+g%v#NqxUrkCG``9v=X)&*G?1S8)m^RY9%_aXV=L+ay~q~*2G*{G{e_<H$*d7k;=
zmtQjBTy^iCnA=a_s{2whFQ&vtcN0r9o_b@7+qX2hE;X#N6>sC~j@idy*_vbDIU25B
zP3*oqgBW1ZnjQz=yU)Y-`0yuV?@x#h;ApE%2VK%L`0&<}*us$rDL_za`>2>&BlY2%
z7Az)l7|v&u?`KW7+a5`z6$4(jfLfX-ni<FYrK?XS^B?vqBrc{$prdy(4%@1D;?<f@
znFFt&)^&Bmzxd8Mr|_HtRo{a?Pr~7%zD=U@qHxMihjq2J3+$#TI~ia@IUH^qzf;2Y
zKjlQq<6Kf8v9z|yZX7J}=O-p1r~Wn0`Csd&#t+L7diW=9x_~qq9e%jA@H!b~EGB#A
z-JvckCLOFXwc;1e{xe-bnr(f6VIZ|ptjx2OXU2NTiCxLUBMG+N!7bO<1`IUrPyFa8
zy;PJf*NP3I0p0X16Nyp}KWsG&znZyFZ8X?!7pN<r{lKg-uXRw$gaO+|gD9|IFW(i_
z4y~ghW=baL!4P%9S8wh(tKB)}V{BhZ>Z;brdb&kZstHw{KYgi<r^R@>x+WusPm+pk
zVxN)|a|wtkPs~&Cie)Z7hd^DfuVc87#lUnK(1#26N>Qd#oZmgL&|ro%e1`8c@H$@3
zps{#9!67%8rioYgFL`))h2}eInVrmVUFp3-J}{qNx=`-!1s6Fom6sguUqLRhhWl38
zX*BdSWvmyZH16h!qQ*(#{N@|<vhrBp&~4##I*|@{$aEkd?l4VcJwo!Jpt&&R40Yjt
zzQ>*(;QTdGR}8lp`K|ANkIS|3-~D(0Z<fbLL^L3vIE<4A^L8kF_~JkkC{yO#jE7Qq
zb@bpog85?J$agY_2L!)gvcadf{l-ma6qWTB6e4*U28qvJ116GKn>T+Uqk0bIdd{Pd
zG^+l-SLcy1AR3=Ps^z}n?Zcj7Wbf3~CnAb7o?2Aseh>zGO8vTr3#r?`o2yf4<k_?F
zOyW}OHdnlMgXa5qARp*3>(EWrKiE^&8w?R#%5FHAGuezx_he`2rp!qohQnJUFbLgA
z0!f)-(qYQV7>hb84hzi%;*3qV{dZKv1BkNLfMNX;ozk9^DVfiekzen)|8_bBb$+oc
zHeGq^C6hOiiL61A25l9Z3cdS+R0|u$Ca^-6m0EX!7Ag6mra&@!bqK_2^+;$E4L4#`
zrepD59k>X~wngjN5RrwuCrI_u(PTT6#WYU$9v(q4<R&EVBS`h`xVj-|$7iVyx*gpW
zn*m$3xGyP8z}y?9qursL77KhhEJFCZ5V&98_8wN7dl?~+8+B<csYXfa6@r)t0?m=d
z7-a2CDMk+BRhcEFHV-CEj71Z*yRAoE-9dIEGT@#%MU6vAOp|9NQ_2+0QK_WOtWP&*
zRRL34^$JT+R6G#m1Ua$M1;=YUfFc;a8yD2P@*yhE)H<Stmxm+opB~85k(Lhb^lH8q
zYi)EXq;g_Xs3zF_KzjKJmUD;@6V)n|16J;+%Zk*qH*?Lsi%_8)j<ofHT`J4@8)P{%
zAMcs&K7jcJQf9g-bfqISanGo%le6w}K~DljkokaIpOAKe<q=IYAAkH2-~HYveEG9q
zkX9pUzfq)nUqR-Aaw2KsPzt?6Ntr1nrXnHc6la~`he_M20eq8eH<V56v?*b2r!Jc9
z>tK66&tBpOxyJs9#7T58h}hsF5h1jz0pSu&=A9%mJcnGFYuo4b-Zmt0=oMOBa9UME
za`r%YU0JWyO%|<!<t@|qlGwke+r(t2`)c@pb)kAXhGDEd#@~Xo;i}KZ$hpvyUu%g3
z(h~{aw@E_GCjQ;6yS4nV-*GU*gR}3e#9XrPi87(;Iw=irDi6law%hR7aZUtmG@O>3
z`I%CwPTs&MC2@buyjLd<v#*KP8lQju_Z;$k%b?(cOa}kgA8H<d?qPu;9tl;QUE)Hm
z8~2UTS7;&G?_kLzW6u6yh7Y>m;8<{RK&S06as1!?TA?$_BQND_>mu)!!amI<;I*-E
z!iOb=YFNj^5pPKQ0v%o`9x!rBZV0G;;Choc7*pM+V~y)As-CS!lTXF}Ju)H*m_@|Z
zirZHlWi}zI>Ya625S(78(V)$^_k9x2%={2-Oj(&??u8RKT)gK{twVLSHm<8HHk}aS
z=6cB*NF%ER`^U@G939Hmn;ZJ62e-C1)@9{#b&KE{*tDv%)RLxcV{1yzfRTh}>uvLP
z=?pKXn^&#Kb=lP937B%>ZaSiA#=3^5556XIW?d81Gu9gmnX4T*K72qv-XW<t{c>I6
zehZR?ekG=NMFwaU!xNVIbfT~KsCp@%4i8K(K7e#!ZG&dpqjw{SW=I9oOv+>3D@kWk
z%1DK_ULZYjH_v?UlMk8JE9bA@xXN91%g9rRm5ODu6i!p6cfSt`XbIs}tS<<A<~}ld
zVTd>fcmMXF?~!f4d-RWMxDDVC*OIcEAVz-<*o^jI$H6VmvnJbC+$#NXP!7@t;_xZ4
z4!l|?7)JpS_2#N<t@KjdDRo(bmRk2~iR%_q_oAv!EYX;5n8K&cmyA<&!MHAqk6cwZ
ze+*27Zf?q`CO)I!29gG^rqn!Pv;k}Nu}{@aJi8l%YEtGYM@8=B0PwzHZSKuB-g*Rh
zM)jrp40sKQ*oX?L8wnSNzL*l+&zwr8)yqcA_!ocnE64h;yuKZLq~2G^64SFZ#`1|P
z4(nDq_!r5cvP4@fkZ~|OJY?q35IF^ZI7ChXd&qDp%(4y9@bW&O6>&eK8WV@2+ViZZ
zj****V)(&@xV^76&z6E7=dgrxkcD901ix3Kyv->&jg7Fj$jX$3IlBn2SKo{(&cblg
zNJqNFz>BarFkJW*Pu-hjVLOa}35je&f~Ra2<3fWawFI?Jlb45Gp#)7dsk^`WkZ+h3
zKTJjG;)8rzm~RpHd$1_1$p$s(Saj+T(jmL$wM3*Eo7M4Dp{<qc)f=8qiwgh|*#`U!
zUzQ#=(AK=xH5@NC8JkRhrBWz(d7lSbu9?&Q1212^;&8h6o$d?9n~AXjESknwn~=MY
z(ZfrhZWVa=vMdx;dS5-*HfPg}<e)_lD;*}L!wKu}!Rd&kBRw6Ea>UdRehbG+)r_Qq
z=9yIHXo$C1_6d;*v6bH6QpEK-hvSJ4KKzh&KJ!)WT+f%SY3b`6BxTy<SwL=469=Qn
zy^friu=V!>0+PU+$Bd3ad>GH<?|O~T*YT`xUP2fynt4)oQ)M{5Isn#P$*wVY#|Hz2
zeqx|U0pBnzs?Fz(NWjTk1rq^C%`ID6T4>$4T&|wf^l%!$wt-Y;J3~)eV;$MII1D7o
zM_^X#EY+{Bj>r}CY3)uN=mx4~I~PP8HqPpk9<#`1CEkM&Gww}m-N))0q?|a;6RG>T
zGvyR(7Orc<Sn+m9jCLK*PIlPZ0WAp|W@mO~4U6>26X-b$?;kSOQ@F<d2G~afW6o(T
zHo-?Q5+W>zGivpc&BbHk#2F9PakO|~;nFxvq^vpSCmP#70JR|l_x0v82ei21Q|+#b
zZRXKBr|i&=cUjyZPkmBbtwQg*F%r^f`if^cI!ZS6;9!y3XuwgUPf}Ewb71VnmC|)5
z7H^GS!xss?Vyll7b<@5Um|^o$Hs0IGw9&JO%ul=)kKTwktJbDLUxY=Kxj0`Ob%VJq
zLP=26d>UG;=!QdE6}IPK4E%L7{ACMQB0-VcP4;}R1k(!`u3K$!E9uKpSyrP4j9^je
zcsAwzl3p9KtbTYTWlZtfE5qlfP!V;SA&^k!!;zOSU-90nS7Cc@TP$gG8r@k>WoEew
zoL+Ez^)a;z{d}P=)m5vtu}Y;~AF<1s9D+D?UUO(pD!nR96XkFRc7!z3q<|Dmli$;z
zIVL2$Z1O%^p}CMqL_ss8$^Q|>Ql^rSX=c8E;PCe1it{#1=z__jO3y?E9!PtSAl`x|
zJ#O3Riw)={z?tz640zSHKg&>L#cNAiWMJwyuaUmqG$*z{-NHtABiplJhmVH!#Q1r=
zeOr;eDv7{6zSH;}uYlsvt->141k2iduQ!XiI?)dl*DO>?R9l@&Z)3kfkGTpR=wcq!
z=^N|IWax^<J=+7CU*Rw#3>Ic260HRcZoyXSK5xB?lCp4^g4!?U+K1$4<x)LpmJ~WB
z!LN6~eStXS6y9b5gF9U3XIV)~ni2;a)ZG~obpy~}<Ml1!BW0dSOfNDI5TOFjagYrG
zm*LG$)CW@g#dlI<Fpe`0Y%FHOCV=4n=h&E{^fr!4P)STvrj%&1h&v}*WCl}qah<CR
zglny=S7W(|J2{H6uKx21&UzXPGIEpg=h4s-i?UCmkroxwl)PFnwap7^)XOO{yetv}
z%N;YXz0s;{eGc$L*Tg`Eo*+D8VrdJ}e~7M$5r%sK&(I6U<h_F|aL9?G9;CB%(#@vU
z0hx6#+q*!s>b!6Tl6XMw!Mh%5(WWew*t0wLxz@^hT|6jUE7!GCs|WiYlNk}$k!17A
zNoz+nxZ)|Yd~<X<y>kd1Q>KbAi(3&)T9{1Wy3%-rbO16YN?TWIUD2e}s&IVC{YO6_
z9qzeYme8cEoa+;HT{z4WvM!`Sj!++r-a7PJJvrz-5}kyoo8LKZ?LSkShawgKdri*B
zasqx0htCEvry+@w@`<)SGF(lM$C>`(p5@J3)~6?yWeH)Ii!}!wTg){=gPBsB5hJBG
zsAyD|NZ2{wRt7g1{cTWi<FxVL(aiVl$QE60UbcsUXCZu>#~uQ&aXr0Tj6!8J9sCzY
z8_BkIvLArAA9+I*>0tq40e4n&dy85}Pb$s{zYC~9H8%nDSP`>D%!5HSA0|}svu2dG
zuGL@*a{hbXjP^+mo@hz$^afTNSz~XR_tttqz9wD;3{Y7Grjke`Qi5ZiP*v9EYr`p|
z(L)f8XVKhnQr-KgnQeh>jNBXvSPFqL3@BP|sH}hg>)TZdA{@fR%yD(+t|N9+c3bEf
z9A@!`cMh?OXW82#RboJlpvB9olqRO+SUu*1fx1?Pe~U9=d=N~Ti@SlwgG=J}xH;xM
zyQy8}x)v@^mA6j|mviNE!Bx{kFe4)vcuA;2I4nwhw-`ttrFmAL@tm_zid#W!8^BM%
z4MIy%U1<`Z$kY`L%ZTwAv5mn(fM?+xDjwXL?IJp4z-$B^SMZ^N3F%%6Yj3D1$spS(
zPvP1sRo$u9an;~ec*^e@6!3n`C{R`0x>)zgY`aLvG27Y=tCgj8E{kzpSHIQ)>tSIv
z?9|pIN(Qn;g=R7MCGKH<Cn%h>(9^`E-XyfzxP1K$RvYEzOZ4;-Yl?|mGf(R3)Tgc-
zUcATr@E%W>)xEV+rc5(m|LjYy>lM`(9C9Jck<#3KwzYLr<2%-!e6J(k8|WUyYh78^
zmC?u0GLf>cNfm+x6oEUMfv&Lu(To|?dPUdbi9}Rn9pCiBsWd*>D@!bxc|P*Ax6Ud<
zp?q`D!v^&YC{hMq13US`v#G7=3_I@ct3qxVEjKR<wed*u#{WD{<e;IBzp_Z&2FIS}
zn@J??$N|2lbYxzzDSgNEv?2J!EcR1y`J>wOCi4Eh2j0F0n$WvJ6;f)b2y19sDzZJ}
zj&SXrHE80i2*<3Lgnc~<*0zzP8nayr4hJPSU}G>4-KAmZqLfsGQ}M%pnhM9H9H>YP
z|58=PP@*9=v~F}&pZgFnsatLYiAUHI+Cc-3-RFU=BkRBQc;8;%5<W`I$p=fjdeN}R
ze7w$!%&#}5SRAHGRloT?IMgD($jjt52|B9k>|^PQiVqGKkF3j*Mnd6-e)gdl^TCbi
zi1NCG<-}A9h6~BowsKt>XT5S#<D8YJEY#){F^6vecKkc#6j`G;+P2Nvc1qcC@_7JL
znEK5nGv!XsN)Jj-%9(j`I;RH(rK|>A+`-Zu7RuuH)&lw&^Ac2jqqN9o#>|4KV^EBu
z!Xe{>Hxy=7jyZA2*)=6%9I`k5&tfsL3a;(SY7Uo7IU(X1XhxzNXIRhPJ4(gpMTCF5
ztHu4<%fO0T<+>P8YvbD1c*gE2pA`;SIAnXK0537n$7#U+QgLkd_LXibxIrD3w!(7d
z`t(GeW?w*rzkW_+EgbJ(QBL<fo-e$4^TeC4zv1C{;&gZ7@y%;Y;Cw!Fc>jH>b<SUZ
zhSe)2i*w{9?y@BB6Npgj>h83n7z<`Cnv}}s{@E6+H}|KKM3d(Ao@CdLIQyM6AL#-q
z8_6=Y_=s2{JKPy^1EcmNrnMfXjsvsbp!_XhCv`RyfGJ+7<LlhTvbF_~XRfI1@8`2n
z<Q*^%0(6Y)+1~?VFQpq8$p11RrCo6h-i#gbf9qjz)58$g`fTn9HX7Iddt<(K@8aG@
z24=#FF*MjBf~ADz>@ESVdKOe2s$JUZm6&4~bQM+$wpvs-lNik;bR=<Lo3B-8_rV;2
zH5*7)dRtwiAWqNrs*gCMh2vB><jmnXah!zQE2TGP4DB5Y62?x&-=9;hG2kl=X3}79
z+14<(?a4@}tv$&7I<IeMh^Kz|r<6tj+Mp7TkT56VZc5JHkZ?3Ls6Bf_(;PQG%$egf
zaX3uOQv`;%!7893Vr#9lB&D|QUPF<kk7;bB6c0Yp!w?bX<PoV^XHFA$+Nqkj=E8bt
zT#u`J%EapPe~o6D&!To7{V2tUO~%S7(@Rp0bLN<pu11~&Nyc=LZGzfVoOfK(Fc=e-
zYP`8->Qd=3rObN(KKCK(z}P+88Nwr1DUN*J|E?rso)V|IaDOP=&4s(7+!qg0B?$#%
zieWHMdPUm>Fo$@0Ow7ai)IOL9Kj>-B>PfOq+5^wR(x@OrFppMTuATFB<r22UM7C1W
zm?q_x7%DzBDob%kT+Lx{IGAT?=d_PSs>+IP`Uw}`dWPgKm^rd=ht)2%Qi^AL>Du`5
z-~1&%{_)>%K3{xoshhAJj|V>f_#<B2zi{)jXk^nWkKcTaN~6pN8bT@)64&4dr*GUm
z&8TE@E}rFCb&mlxkm3Ty?woa{p{*iqL3$+#)K*z97a%!{=5leP#?h>dz|caKyy*>Y
zfceqP-FgTC*$i5CZ{tt~*Rgm7dVAWB1E<5r@fRKXh3Dshf8MS{-a8e3P>x|!15WLI
zw~-(4((UK&ue}-G2Hd)??u%x&eai3;GlSY2*Y2wPYi&NiVX3T3b-%1BD%_MHC9c+3
zTjkn?W}P*xyN5ZWsP`$?x;jxo6B-1I;jV1fozhCjO5&hye5Y|u-7H4(MpL^Rm<wTh
z?PwC_?CR#ZcygJ#P&!C+uQM4LJ|Mvs^L5aIvY)7-Vs=rWuSbo!+ZY?UfDi+>cl9QJ
z$Lrg%kCM2W)VwtD^2MUD>sj;tA=B<>UL48>In?eE*>N?GaQDZFxuiHS=1?d`_7NF~
z1Wj>S-O<pen@QMNxZ|pu*p0@r(rY7`TV8hrOP={!PS{+q<_@TBt$}BbEXn^2gzzQ>
zRSaO*P*lghBV%w$x=&MkaVWgHo4K|zFI6zn;1b6|#yyRewb0j2U!5kG1By$@83a0h
zpN+UjZEn8RMUa>|J131rm6(z+7oV{_rJ=$M6?(~t38@<E<cI%dZCt8yZ3e@~CT2LK
z<PAbsXv#)o9E=d~Ow_tjt-8m*Dc3G6EmW#OGfdJsO~n<_C3Bc^%uz^`R5osRoDhWO
z=h;hwTIuJQA)$V45Dncz$495;FwAQvESHrpzxpZvxBvV9gD<}LhQs|GmOS&F=ZVu{
z;_cHThvUTezw-(AFW=|si@)LN^*8kE#SP{}KqpKl|6V1$rdp+;XfB9|Z@6_u@(!`J
zR-@_bhH6Sh`W2~HXjd2f84&c`IcKIa(QAvK5Lw`WJv4A)jd8U#;3OIARQB6=0}VZo
zvA)mtTW%YC<NxY=KOqkFcb^B?4&^XlssY0WG$nFF*}VD9n+oN&ud!WST<n-hzEeyC
zB8E-T-kPy?@6)Z#mH$ic*s`*Q?Sw@V=e}Rc)nHj_4Eu7UOX)^Q$!Bj5#lq;R`5h0F
zLfe923Nz5IPRW!wgj0AKb?ckMa|pCOaJ5Pm+#S3KqP9bgt&xCKQuONv#n|t%Mc*rh
z6K!vu#VVOhi!lK<gu*h!Z(h~=ZT(GD3c@yv*oTo+QWi8+#4=?#7NpJo=9*&gePyZ6
zTtSqnBu>ZK;g{@iPYSKV;92hnBDiZ{G<E9ASc;ll{-1)<rK+s822D~iv6zBpSf~KK
zs=?`A)i!)GS^*xNqxo6#z&pKM?6E`)FmbIy%E_<EVC@7&FSHZm0j8Z=*ILEe;+)mW
z)_WOQhN4>%Xpz}DlgZ!P(H&<371dokf(4UDP+aY(p(J&=zmLW|9vd}v=A}|D9;otY
zmWcOx1gGmIDbesAl(=+5IB^X2x*F%D^0*pL)mRJ&YFr6?|2UPXh7xn0wo!s9=5i##
zBS9bZfae*F%I-rm-~$I_6w>Tu1X0JQlXoQQ<;uD?{$Ky?pYSjK<X@oEiM#h+Qtl4k
zkbV1vTII8!{|v2HK7Mge&v31c>(g7l{`}`mrSQ>rzt0?u;%+_!-mK9nX!b$s8r6VA
z200q=UU`)`%NP23f%OsEBdM<})*Y5=iCR17$1{(w-}t<W;gN3;tc<W8;Tdl1_e<Y_
zgM6*(CU^=s#~0ydzw+#NZtnN^yZx|#ehx^yr`=?pV~uzau^YI`-&@1w^cmwx<k@xf
z-l%R&yjyITKmWZ9GNj9f*C+L4H77-@<6gJ3&6}0I*JrRcc(XLDJCxLWr(n#fGp7tz
z1F3-2Atl$@3})cyQ>Zm(1@-Qh8OfE~o7dA#uxr*?TIJdrm)5xU#&o@KEa9%)tIy-8
zYrwU>hJ21i+A1=$(ETKLLhYFA___zc#6EjrODaPQC`mR_&ENO>c0+c;fJ1P9mLx$B
z$1kWH3UgNfvg$)(RFrjXVNnZ{CJu**X(~)*H`UXW_N>TU0A!IB6Nj7CmAafm7~(Xn
zoOGjgl7!yM&WLG^>sn!To^(#a90R;nykTbq6HjgyYTVqM1PL0bnsbcz#T?nI2+eX}
zud)g0EDjLK9-Jf=8sU+o;YBj2gDvoV=>sY-OkL+Ol+zA<+)zS88R-i(f{0(hGd{X$
z@b~AHvKERpuX0KO>4>#ZS3BEgGw}n*vwQT-!sQk7%H_JOynedy<>M2dy?MizPZ!Q%
zDmZDPnCm$b{=L*f4A=~Ah^I~uR?x5#e&-O<oRI77!QefK`Eca);yu=qQ8JQddQ79U
zYh}4U@hAW6PiW1!t_$_e8|HO&_snM4a^aihiTU!#zxNM*kM}?RnA$3r>xF8<_36yr
z)0xBCT+o+1NvL7Sm^FiD)EY(Ili#QI`KH!^e`}-HE3A*$^$mS_!&0v_O&pF7h$WWG
z%H`>a^W~A+s*^7?MX$gGk+t~&1EZP^D9pxT8$lC4m>s3@U~ja&FC#Zr?>Cgjj<fr=
z$RB}6)4QdcJa23V?Q#oNgYJ3Ts}S7-2m3x6ud%0nJmv^)w)Glz%|^^DH-}HHyR~|z
zZoSTKjanCKUnz2M@|=0#G83-X#(7;?>WXw{Zygj4LA=Nccax(4vpTs(!)VZ&4+cu=
zt__nU5}ip?=$t}1o&7$K$&yWA=`9EZoqD}uQ}Hz)lylb@jRtX?FYJMI8_YGJ#5nlZ
z8o=jZFlu6WL-~;tR@}elasRCi@ton4IFyN;#3L()N@GeBxg<_gAr;SHF}ED3wW8u!
z+B_HX$nr~JIExZcX+vV%-R@eLl3=}&xF@CT6-69o>*CB3GlT4)?|Su8(?wFg2upUj
zYAVinlID7aQ7P<>o_ixlf@_>q$zgUkL@%CMD{D`i&}<xbW`3Cae1zbMi$Tv|*_)LX
z3xBkPHW5<w`w@#EX^xU{3^bR(;d6YVW2{5k?;40PmeChhZI!yZgnw-6sU$bss|_=U
zt&$Rx@d!MafM$HCl8Hv+bLCnaPiy7Pb>*w4mA99Lx;9t4j-aA${o#<0-1z!iFe?0W
zMkNyE<g<0>T?`E#xtB97&!mU<Y15t00?#v=rUCfV><N;xww146e?`j^%U6v<$s7&`
z-^A8w*E5&9g0+RZo*jQK6FSe7<30E9y-&wj&X1h#UNOxFmh)pkLyd$F9%^lgDz0G3
zyqZ&SYVTTMJ(JrN=@;ttk+!br-Ah2ZURK_G{gyYcU-R_#jjyp3wl4vmYuGYCRKgav
zE8Af=&>o8#%A3Sv-tTygiWJ+uo(5A73B`CG8lgx;<oOdBmBOCAdWDKaMX-I~4l#MZ
z;%0!)IMljVLmDN?CRBs1p)8ECzP|pO4zoWANEs&WT(GD*cjl5i^(R^tcUrZ8n`({8
zkj^seBJ}X%nxr!cOj=wk)aoYE9Z8uAOeri$;u<7QD<@y+_Pw#7v|}8TGKn&)qP=3*
zN>|8XTsKOZ-YZ6m=5&t$YEkL=UU%fz6G&fcqrWpW4t`E(n*1L8Cd6wwJ2RwYAtlGE
z{lGMG5~kwOlk5*-l!iWl#ouEpIrI%)_DPhq19LZV;ZM&A(TqEXLN1A9#*}>umjUbY
zL|ra^Al4MvtT*Wr!dwc|RNTR{`VA>Ld#M>Uq7Y;P8)FAYgzbR^`yn)(2DeQR0wTf3
z6OKt$cC*n^8b0qg+ev}0Tlqh>+bzDjm4J~cjq=~NO2t_%{&P0o83Mj)5okuAq_xKN
zdgXduS(eHgb|C80;k{3h3MD(<pG$GwLO>WT@cYb88?}I%Q1OaJ6FCca2gM{ilS!cz
zp_J%rh{ZwG8EiSsp}5)xGiCcqH!uINF2)dIF`tkZAJPvmqA8yoLNc#zG)?5G@DKmt
zAMmGt@t2$%lxd<Vy!op7&>~gtj|UE?8Pm@7^2l<%u(n2-j~q_-<ikCWPfvXP%_E<F
zXJ(ow9^ZWJJ)C@pR->*LR2w<X8)MD_MzOXaeW5<Rrk=mTu5X|&K0fn&hWU<kI-<+U
z^@|Eid_Pq;?d#Cix{(i5HKuOucSi@j)OP`*sxx2)o_^becE}=T3EX6&f+tj>$*U9q
zar>EHc_Kns-E?lz9oRJv4)w*F)v*^wkO)O__iJ#O)ffjFV%Dh0ef_iFQyXH@sEWj`
zrxPC3q{A5UrLB>}%yC&0<r;`Y>ktZP@G9>@C)J8%SXXFCkdl$!NG;Hhq^=gS>I_K-
zPux2UU-N2%hCGr%hlo$b`)5sHE>N^FsYJ5v{~jisZk{*hX$%v-;nE!YX6x%n1oa`V
zi~AnXZ;k7zQAPiKuWuC;d>m%Ny2CFe3sadkSR*GTNgfsJwungy%RM^hhL61yz-d||
z+Vw7=87T^uw()dZcH@&?0xyHXA0Gms-W~cGB8@Z^?hgmL!TEgQdbv>7l|=SYm$^7}
zqixGPh(o>IFg8Zy9ei&M+wk%IdaK(W?QuFA&NLXCU}uDQPvEvRe)rF}S;iPE>KkDH
z-|9`4<{Po^B{+Uaia2|u^-k?Z6}JbtF6)LtW>u;V-=R3^)s@k;k5b-o+LD!&Ca=J9
zCg;Xs&O97vK7DoNZhBy8u4cC0Ljj*T<{a!O7aishD?)I~<(A<!dN=!P4Q7cYpP!tj
zBh%>x`S?K3vnz)sETX&PXY=8}o7Zpn^wUpy@6}86?b+Ab+E^-Ft#N-)j>pVD`lEk<
zrp(j%N?qI<BbUT99}r3W^0ROF{PSP(^8N)y7VP>proxrQ-7#|_=YnaWN3zh`iq;D4
z3YXXD@|N}TmVSM5=gMg&oo0Gpu;%NcBTx>gty|wpl+YM}H9v&+WWqm7!~Jqt%lkUm
zv+|VU)V_gYNNnT|h(2NO0bAv8{*8T<c@~7mxwNlq7&hIv&vN|i$Z$umXyDwtLT%4r
zb&z$~XG5#HMk9d6ExFKWswN32CpmG#WIG{9687=aMDN1kym~)0W~^*ba3mXtDRDn#
z?kDAth1A!u$Z^9u|9n^0tKC9yo{&n~ZF!Qo`gl&ti{pVgH`Z`UO-Y!ta!4s?%3d*9
z^Rb>Td}|DYz}q22lJ>K8^mWAcbC0MJB%X^zLiwx7*5Cg6c8zA7Q=irUnX)rVyrlBX
zA*X^Oo4-%mnn}?(vAzwIAAk$$Vy0DvrFFNKUVYex;Ssu|gq4I^am+twBzZJ4gY|NC
z-GL^0^#;<MGAS2-Fw^9Mq-Ak-NAis&117r3)N395YYTgVy&N2ZAYdB`-fnQvv=xzT
zgEU4a1hy$Uce?Aov4Ce7>SirH9=89mzmwVCjEw&t*W}I4y+PFtBt{eFQklBBmrzo2
zt~_2B4pZ5Lgl^N2=xOyP+70U^Mb%PB9IVjohyZto%yn5=sxN8+rzWKk9Qsh6(|q7C
zPj34gxP-$XO7HD&UAHwF%dJ`TT#Tfdo@OM^{yPCxLWl0t?+p08-~S>1=pXz(fAS~)
z5=~Gsnu6p)DezDJ@gMWMzw>>6R_cd4xgfoh^28Tke#5`~7yl#6^@$(;_NScnO1+%X
zc_N!p%0!+IKo}CLA=c>if-R4<>m&8`7qshJ?0TlRD{VS(INY&JcUTqdO{2b9u(i^x
zI&_M<pO%PETWi&E`(URHN{~Z2Y0DU4C}MZkk>=<4y^J*mV7mf2TO8~W1-!p!h`#$&
zhN<BY=iT}##h%_iBc_=dgYfIuG#ce$Ej@G%y*q)S22IqHsKLZB2(5<e0WJEP%B;q*
zI8D>7MqD40#3-?Un+nGSFLGi{4pEDEud9f1D4F{?aX)8X95RLOR^7dkyH_IG$J}+l
z4?Xs}MJC)7x42hl2wL+#T`t~h?PL$+Dh^4FGb8qM?GpY^(a@Fvq(EE0ff6|6W*AY>
z#r_)5VF+4d&#MkmC=|H~&cDX%TLp#p?hicNA2=N*4pR<>i-ZbK99~K(*dkHR-q5PX
zy4;*cHB^r&Bbt%ksm*;%)_(RV84aAClwLB-6E-`3?3A+RT;+9Txh~uL4{X(9hJlvM
z!46z8$BVWb=m)7P+dzxf`j$bMHF9L4Mt^2bMfWmKMPe4}4R1z0liBhk!5eZm8hgo&
z^u_~@ptC_!H`RkzO12F(f)Lc*g#XR`GMe@(iQBom^AwzUg}dV%xMb()vhe17Wtt}L
zk5dQ$1<i`Ja-){?l7GMkIZdRLnWu@n(}8xqvabH@df#b|H7cW=d{kvg+3#oC&DlHy
zi=K)-SIBNJ5qfuPU`ffF%4v$==d+z1g7o{6a>1lC72$vJpZ$l>&V2FNFEQY=uU>P$
zT=)<EXa5fW<Ujml+)pJ0QLPWRWia6{{^Cdc(|`Kk@_WDgU94Yte0$||cgN}B1&7lE
zX*xK=$P8H@v2~%>3+?hqzrKNbrPVX7t{mr?dA@gwb~o<ip7j3ve0}YF`Sdf+ZDQ&x
zZ(H?YOFc9}Dx21Tk?99Z@3sjm;u!mY4Y$U!?gzKst{FxquA3!|1dD4ZfyE4dB`C0c
z-tB+iHm*lfvonl*|M#GQd%hE&ciWtfDklvup;&V{CsZ?I;Fbz8AGtASrBCLH<CM5|
z(2_aj#38wD4y}>f!aa?T4$9n01Tg=Ob5c%+nd9Wm<x_EIU#e3-w|<v6HZ#mS5phS{
zh!YO8`g+}<9vw$Qik}OLvWicRQ^9)N9F3sRgBs2_H=JfI5`olifa5qTXm+!I!p0l#
ztvYL~{_4K^TVCHPD17wZ9S`>>O3oW)(<z3|P#Ac(i6GgHr-m)>)=X-yeyzQGluk}b
z>`5^NNkTS-wX^iq%WqX^f^}uJEA{&1jEfkuE~${F$=y&x(%!2NvL8(}MTQKUZszs`
z>IanKnX#DjzH?@a+?ZlI<~Z1vq0<eC46IuO1sq!P)`5SY?SYMjlG5%~()UgA?&faI
z%m17b*j`=_y4sM<-wrDu!D`r>o3e?)Qu0xaW|Kp0t+_a;PCQ;`Ofz@K$%S^2S(T~K
z3BR2*6qX$`zh6^v8uIBdaXl7SfYut!u$Yz=r@_d6RGi{xDY1L?_(9>>Y3JRbkdv^|
z1JCv0>WAZrFw5HB$9fCvYnbK(^I_)l^ffOYX8zOv_>cMOi?8|l&wjxl{@(XF9A>`%
zz3=eLU;G3<{0K?e{cM&CELR@izU43f&5!w$fBxU|-g~e3_x}AqLbUPl@F8~(uPE~j
zA>u6JICAs!l|ccdWF#w7QR-!Kvq(|OVPZPYE@qofJbdsSK6&{&$o!|g`RR`+=P$TC
zU9j-gZ*7SRu58c1aZ1}V?C~;-;Z(g-az-_wF~ee`nnuDFq1cnPNJd7#2$Qg3F+~4v
zBefxr8ja0Sd5<OUe~!Q{7R^qJYSaxZ$90=+xU-+Vz_63w?7pqjbEjKkYRa00Iwd5n
zl$x1K=i&ku=IrDWhh`d+Rqm5vhs68o020B;I+F^Axwx;IIKEGB^yGFVf`sP4T~vGW
zB#n-1U&4~c3s%qOC4r`lreV>oJ0ZooGwzDds3x3^k?JPSK~44Q&>{16b)PNM6fF5b
zB$`)$J>GZBfp@D5VO`F&Zagi&fuQi-!yWhc$6)C={E&*PR3&Vdbu>w(IQG3PTL)gE
ziDeOCw+8M;Z}165))wd$X=`}2nP-?HNOk_VYGKL~Q<<IOIn7L(Xv>NktPsp3&w^cz
z8jw&l<XY8e;;^xMT*>R#GCDe{&Z^j&tTuSneRB?#G1bUMh87dsh()e95R&^0c-Aj5
zS}&z-IFE0ppf##jB;%kQ4Yx7P>NbFk>`X<rJ_LziNI2wYmV=Hsrc2G7YNvKWXqQPT
zJ3Nq;K3HEVh3hBLL5wVE%k=!?bSf^o63<Xl2!Py0*V~D1wI1v(*n*zl@pLmYBfdfi
z^W5jj-wlb0TCD{=Mz*HofEMO?=HcZ_#47dr#M9#=M&t48Z;;cGUrOiYs~7a;p5@I~
zl+c`f@%fkh>p%SwpMCKKr_+)D<Ujs*`K{mj)Q6$uiH4W{w2HgdUa;#U&^gR=KoyA<
z!}<bk^~$+b6i-A-IguV7FzKYW^5Od*aQE;3$9(g<f5^}O{9n=j`bYFPUsBr>B2aV+
zb+q3Ti7eE_!gWEme~>}ka0;=AfM((%q6--ZPTY8BNQXwm#v>kq`d2GA3*1k2CzR||
zZGThQ-n>Uc+53)&y6dit>+|>A*fA#5n4eHe3OaVsUQK9SHV0v^j*=S#CiCQUsdOMq
z$EsVAB-}Chlp)QDrcRsz*K?>ifo3jlO{1{E&;xZ@Tcfs44&sFetP~7G|E{j@!TYAT
z^MRY(1p%RlsBEk3{pR;!aG89vV#(h3^!wv`WzC7}_39Qm0-$u1CQetj?k=x65_T7x
z?pEM>lU)AWUf(JxJUpDZKOIS_Y^AJ^iR)W-IVz?$8bZafuim@Q5DtM4!O$7YS#kd#
z@T|+$)nNj2x|IedrlyQhfSLV^T2i_tnyk2PLg;O!U&F&ZCFtQ88u+u9vaV^kJ?yv5
zGS;+t1>=h#N@v>xg!|YJxXItM_g&9$EA7AS0SuOr5TclEGwuH{vM2TnlE|jEz!ERl
zR(PKy-9S)I873^311<~L28=mIo_32xl?zQmD>7wz*C<#0LM+@<O-6f8gC^{MFEKa5
zp^Gp_^mKF!Zgs=|ZKL>kAbs9r0XWPPsZXqHbBM7owDsy$h&9Jw<FhTGwp!g#tjq^q
zzV{JxnR$5ek`F%mm^WX1#r5$mcc+Pm7blK$CP_usGmqcA=BGdT32z?X@LRw2eSY}6
zKj6Ecz8{n44qlNSPLwiHFBjU{Jma@Yvd;Rnc-EO5?y}~SyTn$6)eY=Alyt!I9jb*^
zFZA`y@$P-f!~48={}aCc-Vgc3pZ-g}{_8(O+v16kh(kHCw;e+sYYu`x9sA1y7ERIL
z0_lO0Z82Rp;o!zrvt1$GBQT7n?ejGlb4`$K9}LsIjbgpQRRe#F&rF2dLGniW_Pk|V
z?kVp?4(~w@TACoutNpG@6R%>XTv(;~mlU_u5izI>Sc>YsQxXo1CO>$I<R}G0%Pr?a
zWv{*_i9Sc^ooanED@1b`8;&Z+k%A;b5o1!fukoGZd#{Cgp{kDn5D~Pyx0w$T^D{4}
z<bg9RHq6ZDiCE&g7+nYI5%)$6>7&m%EOY#f=;qMi?$GdWe|<|zYN3=8Wo$faO<4DV
zFM_iKG=wTCyPIiU6YILN20rbKol)w$q`a+HNZEsf3NmHd#v&LcbKi*8b@fSa6NZ-T
z4cXNE5K@o0o>4UVvIOVa8#{yG?Pgj%bOFyHnAU0tfc6DD0^WG_o66(%AlrwPk8(Oa
zV+q~<8}^c4gW@e?(+QzM%v*0@rw=+*Ca^X0)!!J*XGM>n9}trTLu8{jVjC>cxAHSG
zy`VB8^30ubE23QL<o5;kS<NOfVQzen5imoVAFh_t2K~_`e_Qb6S9=O6d6Uqq6~9T+
z7Ydc$OF&0{WQ@(3%fw_2vqqPS<caxo$H#}6?|%0?)aw~~#X@UvdHa^yR?0#7!S8&^
zZ~y-9a=N=`o(qx^mfSMCn$o(ku8uQ{NX!Vxus)GxWw|`^^z<6LUR+p}q1Fp2b(&^u
zKG3HZ)ajm-jx;IsE~MrznRll<_~9Q=rz0tyc>LL)Lx1ay=dgh8O*ZAc*bd!n_wEL^
z+i-pH<Ee)NI^hOzV!j8rgh%$~W$X5At={hYq1qiiF#qhaUu_RqThMwZ;R)EvAgrn*
zB51%}uthKj7NXiV*{P@acbrQQP{f^^9p>@pudzRqc2Z70Ml{T1Hs2%NvwVppYOJ^L
zx5VOE*HCMVYLGhi9-n!A=HksHK7!c%T=e7;Jc-h6WWZoriKOVyMxQ6xd#W0}OMk8^
zVa07=CbVpnrc@gw5%E10+jN@qet&<B*S87^W5I0)Xx!R1ss!^U%CNr`ahl!m66!ts
zITDbK2EgK`U2S6`AVL<4C<(2tTb8=rTy+tZg}p#dg!5={mXwI_9g&#)CK@o-P)^%!
z6Krgqexf(wo5b-^&$Px>aFd?xe;iG*-Uk%*tM_hH4Wl}_Wo_KfwG})AaCqjqZebh1
zV!pA4V90~i&SBp&n0Q{#^bf6J;({?(QYH_;HX7b(+pv8cG<`4}I%Z=~LevWWgG4CU
z*i||-A?$0%q;F&)F^efy{?EX^-?@3-jL+^e+FPfdpOEugxGQLg9DQ>`-&{#`EAd4W
zk|)GGv(rO@ws2j}T+WZ2pU$ZI8q9NX_H#1IJdqDaro)jk9ZB=igM@WlX6gS%I+=y_
z?N>;Dg5{BKKKq>I@ybUZe?+q@X%h7GimA-xGSl-3D@Rf~qUD67iDr-r9Q44a-~T;k
zGk$vg2EJNo$)ayksp1uqLyMtSzG>JF4<pf`$r+ipxrw0=CW2$_X~cM^_O{j|&~A9p
zv*$4!2*%pp`(|TLN4?Np!HdLx((yU5pWB}3?E7R_`j1Lt9NIO^@$BZa5%EMY0!>n%
zMyRW~b^rK|Vb5-s{JV9-!k%KlvBPE}Z4Lq+oXkLz2OEuko>xm_%<LZf0J3Hh9a<nE
zJ>NLD4pe2V^&V%7L#*SB7~={28Vt0<&9JME4;|X7{FuuiLL_o6T#bD_wElPZhHhWq
z+9lNwmL0!t-4Eu@IPg3mL|YOGpLj{W8EY6-34!B}16oFAGBCVQH{G1thU3VsMJe4H
zm+RRVzIet|tE-&VoEF;=G?_mjiKb4_y=qK3!n*&CSYYu{@{Q6PKSY!4S**Z*4{D2&
z*_+y9zuamMZeJVSbS!Gdp%{nfO=b<A1xqx;{m(n2Ygp7-+^2?72W{`>Gq!_Pc1r6A
zKHdO}l6ixD{JS^DjnM5->Dx7JnrTtr^y=3+;D@on$HC+D$Gsx6fj1jlg&@J8dbUWX
znYJF>!lEt|QI_s*qphvp>xll0?=b^~dV{SD+X&Udx-@EEFiVKeNUl&OB>UWBAM&1=
z4<}NYF`dzJN1l#gnZ7JoKhx_O>oqJa-eTu(xITV^)ho$5%j20J{l#DM%g=tshabMr
ztB*e5bSflCs8#=ch{}5MswpdLzmj_)rNqN0pYq}F{X<^=xzU$5SiMr$E2=9g%C=iA
z%x&T}Ai}Y?Hi~K90aB5ed!pmfI@_k>_t3Qxdr%+0q&LvZc5w0>9(9`J@Un^wb{tw!
zwQRj9_BGa{x3MQkzPnl<8U~FG80)orqT$1uyt<DX+98@71hxf2Q6o}h&07#i$>(7Y
z3yb}pM#4AFhSoM>$?cw<N@1R7r^|!e@?0-l<3Gt@2Stxz*yjgBl1A8~4Axi<X6CNP
z(A>AGxS65NEqDxJo<`H&0fW~3S-2k1ouz-B*S87^x6#>g+i1k#LW>dIY?}%;nk0d(
z2c(24(j6c6%-$QJm^(}kXs3-bO-6R=fnj71t<^0+h&tP)$I5BA-S)`s1!sh~veF~0
zM$_9Da5S*vdzQ(K+bkZ4W7d5ui-kb4p4}ncOTV0+-M3q+<vwb0zn4RFCc$$yjYgsD
z&ZH{p-Y*i@V-}FjUgB?5LUBxSWKzWE$jJP}Gw;!#8Jm#E$|OouGxG^|47KRK>7#Az
zziWh3WZa37`ES+9cyIrK+FHbV4@^T}rFq_h%6OIOfHj|o+^!nKn<fOuma{6&RbIX>
zTcs{n7Zvv1^MA{tq*3O=^|~Skc|LID6|Gl9T%q1;r%QIB-0^`j-;?r0Z!7ijHBVpu
zl=D}=;NdipwDR=LXVk|>KL6P-d3rp%-;;GNm&$xx$j1X|zDLUelTK|<==pQBoUnGL
z*9+2D%IO8FN30~ebsp{>IQ{;=$4}=6p8oWo(d!d!QQDe=?(I66{rlZi^TSOu_U)Ji
z2GHv6dE|i_XZV>>$~aV-D~Ho__(+ftV0zNOF@s>USQ%?Rjj@poWxiL@BQY84*el#P
zC(MOm+t1v@qg191mhyo<36tj2=ca&4<IM3r=T(Qs;k~XawN{@SVv(2<l<twxrjdYk
zG$}b}z}IW*hNcd|vKzrJrOY(X98M=v&Tf1eYkj$10LS$mfly49AYyrS96oUgg&t!7
zR5o?8pA9JztLUk;h9ub<*;&JIlaaLg{mX&4{CaKXZ)=DrW6Dn`c83xx4pr-6?zA;H
zz=qKqj$vz3|G7R67Gg03#1VLBt+t861aWA>Ky3^yCs;92y_or;W-m9BB&@py$E^z4
zr=BTD_<>lDgWrwLe4vbH5aT;~k7jR<k9L}io?@s;eD{=n9AmG`hCTM+(vM9gk%5hW
z`@Z$>=npjls<8#G&Cg`uutqQob_JsctY9V)+3^Sr@GNuo{>kWlX#D(6OLS70Ji8t*
zaRWaM+Tzeg-1ZJOxa^itaJ#wHH<ZUGuPc3z>#tM_sT4||0io#$Q8=LV^^x9Q`2nCE
zpPwc4ItH@ry^U5|C_VSAZ<;3B3VrEJha+BmcBcW)LCc-*&~iZ1?CWiLg14XXr~l*s
zz|Vg1Id_M|tE2Mfi!b={voCo#9T8hOohE+!hrdmp3VAZ7SMQM@UXo6CzTSI-dJVSS
z$Sw@&3miVC&&Titt4w!CKK`9Q;EVGczWM22@=y}j>l?ZzmKuGvU<CRj-sznqk<)R=
zv@w{bZx!6u3+THteyq3LJ7{}WC}>iq<Gew+BiMom1-a2GjHD+;2ESL2agE&v={O^B
zuMQL!KuT<jeeLRwG^l6JsW`pUdvn9;BZ5LOi@fjQ0j$<II0J5E%iiM|sB094_D2U?
zXdr9b0#6@~3#z24kS3=o59rb~gVq?()I4zwE!2o~qXJ4v$y0V{vJO2|z_+pA{QJ5?
zs)a7fx6u#+b~I}w#SSUn!m-lcfdT)H*S87^NkPTHoD1$$!eBfQHH+IpjmHu~HGj~(
z#Td#s&~+Sy5Q3+X1+!(aNQ;cfH}~)p>Sp9zwrmKTb(bTJNP|H!hEfD>o>HR5BJiv-
z#8x9?5m!%-j3y06hY;T_N^{-&1HE^@4t3wA<O39<B=?G`$2}V&w4G8IH#ahwF<@q^
zU>KQs1iO?-7^w%QSX(q2Z>o+_nFQRI)M<13#vXwosyH3=SKYF{ab3Ovb)z089h*2l
zFG}iL7S!{myKV>5?f;$_@oE1%|0nGMY<s^Qw$$KmCxhA-x-Bfd(#eQbv;#GurRx*R
z^@8<EmQKU3I}h3|idE3bnQ|(!&C4C|R+GdbCtAIds9auu!C(KY|A`;}<ga;K8^8Es
z<>|{Wc{nH^+#mSvZ+({+4+YW2ba&vB?|zSFMww=&m+z4#uXsd@H;k<XWp%}tM{2#|
zfcS@KIneO|F4J`6qks6v<ipHYKl&dzJWSN(%5jRuspc&J`^=NR!8szX2iW0Wo@Mj5
zRqW_l^_Vhfcu!7lqC!fE`}@1?J#L@<Eg1qEtJldR=_Y6mTKEt;iioq_Y-lKKWH@Zo
z#6r#9qXO8p7sEb0en!}Xn??dKD6OqFx6E!~>G2F&CHA^J3k-Y26JYgnxTRg($3O>s
z=$9Z^^XHbb|2Nj$z|L0O3D)=ga8ua&7%fpX+=>zQk*Qua(4Bc?@!dzj^Zn~Vx%Gy$
z0pW(hrG<{>*LZ!apwQRq!)p8v?b*#pQR87?wrB5*;LvHcZayxup^$<_bRQ9Ze94?M
zR)asSHwQ=e2$^D@?S^Y^5D9xw=sj3H{(hP=g^@vpBk5R}M5emI7vr1y0@awS8+T)`
zNN)JVDVO~q4AGAD22GIUIHF(Bj={=UZ}AlUP?l#S1sO3*_lBy(!WwJ|m0=yRi;2<}
z;P%Rdo9}PtF09+aN;lO>{Jlyd2h@~ue5Vlzq7grC+4KI%zUjuZ*$$w%#@=t3M}gzM
zJ9%x-n)aiqo8!pVpXk>aODE`^)>i6zp)HSGFK73ucfJj4UojF%OHhj4gUBHzzwT1z
zO%-gSP_NkeYp5%~{OSw-`p^G0Kl|}ddH*}#;gvLg{AWMqbiCu^@7(i)k50UL@4(&N
zfz$nwDj7W-IFtirJ}@0mI9*a$YezG*x?nU4g2?KQv~HY(=!c*)W7teR9C`KIf572*
z;`Pt|4dwirqZ%#8j*@}Nw_uv(IGD%59pXUt?0mP6*kQr#&^BkM4Pokd<(zr(;?-uo
zvC&0E7_*<jHaA?aQrIe1v%XujgwRn#$=f&hY44Maz1X`2<l@014LFb;(j8M2`$jrp
zJLHM)^)8d|c0b%Y3v3C}&Ch8mlnAOdf~O~_?xmI_CZPe=70DS>rS}Tm$u8PG(M4pN
z@`(7g_ukN!H;eA(INpY+7;N)d_I~83`UO)%N!wX4s;<$m`*mL5Dk!vdrQUE|huPNF
zm<)>K_y<OA4c05}*B=?5?lTmcxTe4tmO6dZzb>A1-x_TBG%$uUBzsUm_M7GO(;Zi2
z9FQr~Zx(&d1V4D+Ja(hB9UF2qGLsI?!E-r(&c&gc!40>;l<L&FqT-o^q!xh(p3UJM
zS$+1|4Do-V9q@yC3`g;-M;%|hv7Sibw<7N}JGRBKUAsdlJu15q6vqAad(()$9nZOJ
zmJ7&eJ_aRq9QqMF?dBP{pe8ECK7vA|1M0&_guUvrz?PrAM0UC&VCK(WtuZf8EPbLC
zp;@Ea!qTqv)#v(p?^N-oDC8+ol2d0z1eqL;6dACdJLw8SU9S{bIREk|EKhIv#ozo4
zyFT*6fAo8N^4;I!&6l5Z{o-?8e3JRjryuh2;l$xk{MyQi_g;QLKAd1(oQ%;bIY-sg
zSHL}udiSY<Ir%tJYiHW!D<sdzA$ahudLpC~FFyVrDGBFa{+P5r1!f#oRs@CK+pl;H
zji7K_ArSk=qaeR}P>OhC`WE_2iM#s;dd&WnNLU7pbX$dNNeCkuV1s#jz_zyszcDg>
zwtX;?lt8$)RPEU^*~FM}{eAy_dlkwSbm*N{+wM;m`wODa5`&<|9vtUPiaq0|nRTl=
zoVheq*M8qL7gBNiZJ+SgNga1UHCJ^{Q$e${=X};mz*8ZVliEU@>Wn}MUD)k@-tOPu
zLQNh)K}E5&@aw$3T~DD^dfSKLm?~HfBeg-@+DpcfNxNg?khX^eUWiU@C~zf44l*{U
zx;vxl9@Wjfu`-6mn>V5M17&1P5;$S}+stj3#SdM=vcg$Bk@;HR_}<P^8B+a*#-aH<
zd)Tv$lrp(Y@poZqt-fe_>!g(Fy}4LMa2Hr`D~`}eXqd6(ocHf$<{t3lP;Gu-dJ{@F
zy#==GKL!T+E)0bXcxxOK!~e-?sq#)xXl?ZaTLf2K=dEwxfiDg#@k7FHyxKQ7?*=*o
zo=>A{J(|nU7qPq1KJO{Wc74HPRbL~DklKRUBj<H-VUe1TT3wqfjsqre_J<EyQd|YD
zX@@1|G9hBD%LTEO)++1yEtkhHdGq;Cd3yVXtipGH`v>U#Jt-MeUHHyJ;?+_3;Kf9l
z4;=6AASsdxhR<})QGN8*Y0{XI2TQSn`wTfjqDw;~i<j#g20d8P0c{J1#k0rSR$jjU
zf=NGO`Q~#nZV)IpQ@U=GmTkNy4%^<IrN5#<c;^|KKe%V`k!a#@JZ_mc+$3h~@X*N6
z%>o1WH{Q+<U<=yA#BZqEw_X>%7d9lqJvjUTALmd69P94{hJdL>#`hV&8v!P?JaJ)+
zYUI||-oHz8d;T_%2j5H9V`g)h<lP3h{oj86t?TOjnJroL>LQmE6>_MWTlHBT-2&=z
zy8cLNhr{p)a^rq)=cDXB$6I7tRmk$22ny<%a*MKm7|(4b+D)m|d*2`jGpCECD05re
zkYsO?;3AP0_<jl~K{UC7wJuwRnZjIcSV;7N0mk9AsWIc)#zWV*Mh!tq_f%wCh!&Xt
zU>?Yr;1x7I6^&_1eydwsx9_UzW^ZY@vZevC&8(|i9MF4sL?xdaZEi(g*SfO}a#Rxo
z2D|yU$Zlt`bDVePkz!l0ko2piugGpm5Wv;8QJzCeW8urTIV_+SAHH+bNRCbKRYF+%
zzMG}o1ohj`NjLVtx7kju>obfRg56Rku$Z75dv3sNnU(hRmR>746|R?+%k_!`1<<0>
zNF~#km1SM2>q?ojH||<TmP=>=7OJg0KE7qSJkigO)TL6Ta6H{H-9ON-EB(#SdHls+
zFkKt@a7QjP(_wa^Le3nH2TY7C4J~)hn(>RqH2Yyv8`E6KsbJ!>%tfG^a-h;L-_Ru?
zhX=5R)hkr@*<5SoR1z}H^m++RPo}le`x7LXYPWDnmp#)2Hd~t5jW=t0@2i-;!#)xh
z2g=RSv#r+~7??;%1lgY1?R$ubJ0B!P#e!DdH)^!rte35-i5}*Fm<IcCt1_ax?IGgC
zo`2r$Iv;EO-RkJwB&GGXRj;EW+NzgGqP8BumJzevkJdY-#wb-J++}0t%)t50lfDrY
zgrLtxKy{SB_jS41$D{Ib8uh-{_UE^MpURbA=k@KxYpO^}+mM@9D}7xVfRW#jz&1<5
zl6#1Fhu^<QZNr0Se^2+~5j1r8j$?XuZmq?a_lRGTZ{+P}28%+|kj+Q7(jzl#&EY5+
z)HDAM$l%FGgk@ND+*S(yGlNz;ILwAedjf7~;;yGTDuzL;ZBZhEtBz9~vh8Np!Mg77
zLCSkWcL?G9fsX#jfM#|Lga7Z9amWt_(WV(0nc9sH&Tao?GYf>RV$yCAoM2>2qdzgs
zxsaV{WM;uzk7p{oILFlONa6$hGrAwtn*2au;K78=?>j+ZDidpWjmGm~Lq5rElFqAH
zXeF_jQZJpipZ}b;u5g-ZYvu6ro=cx8Z_d1V{EYL{6E9!APi>VcWiB1Qe)Bc2-v0oA
zr>8d#d0MAyrebLJ;qLwN4R1dG1?%gtvEIq~B_$QgG$T2aN~W3HwU<1Ff_P+@KJ+A&
ziQZS1)=ATWGR>^l1*sLYD?Mk7hF!nG+GR@?T3e`VR5LUtR_EN?>Lz@mbV-zQz}gwQ
z_k^^^oO1-D-mIUoZzIbLAa$%G1WcZ@k3?Lju&s-2jq!H|fH<q4C2-*TSt3x4gM83O
zZxf?&(DkT*k{QE+*%PB7{&L2VHUB#meReR8bvP=i@frk;_D)dPAfsSQdKEr;89~Uf
zoe=Wu%ZqILwa2v%%6oL>o+W8c=dC;Z_6`%wlPN!&Zf$ICzr$!WZ|o!;LzaZ9zk!}Y
z$`fgF%Va}9D>_YaNGHI3KQtw8pr|tgto7|cu5ok6gr%+A={w#zdJZYMhk09^TMTYM
zm#2wb=8eS@5FU)|)dp_ae0D-SAT@8K`fLF&*&^ioLB^iV>d2DDVJLyubJHM=z+@85
zvaw(+sJqrl<{}^8Wa7RkY1$WD@<lMjGy7pzqr%$DOJ7W0Wo(d{8NFw}&lZhg&0Ft7
zQ@SbT1D1<sx|hXKVU2#k=qEJWt~pkAl%@MK)4^(jpnKjz3;|UQYin-GdYsB<vN!hS
z<~@G3(d>N>Q}2(&_V->+5w!V4zTP{Zzd7^v?Q7<F;!pnT|CW4i%w<Af-O(OSeD>Mr
z95SreMx~&qiNlKn*Y(2b;l#O_JFpt=)pfa$%(y$veDwZHid518`bn8Tdd2BDljlO7
z9lDvaCoM!{v$aldXVemCLUKma>?-ndbY)CRem`(z6WkkWQw_I;8vQtt&Qz4k+Y8J2
zEmQK(S(i?pXQufcRH^F%6wDe#VkH783$2ZbeGwQU_(#dlnB>L?{_j|4iz<0KY>?qC
zY=rJX;TGQTrm)`^!F!OmfyR1A+Bgt9W0rz{o`mI$M)lVN8j3%6gKS%q*(qi{5)g^6
zw!^8IBXhT_%*<hCeP^b}fOZ*UQ8()KAcQ>s+3g;UKxh*CWQVdw^6b#Jj~2FQjGOJ~
z%cyupMYIJ4Ve^w~eP@Q%RvEn!>(L_>4MRa?{acI;|F+k+3JRx(7o1K9QpzAPdsjwe
zO&g}2BKo^9WMm$`$h*px!H<g;psoABm|ZyytydSEtSgw8NO^KLhx;)3&uue_MYi_Z
zXcCWqZ;f!>H47p;SARF&8rb@NJ)@BgY~}dqqd46))Q*8y`Ulx2^S3?*r;lvP-+OqK
zj3(Wl<=>XFy9jVBsBw=5^TMzwJ<WwMVNLD!z7dq=d8HsOu!xNsnO7fEq;{xk+)KX~
zu_*$o^7|#fB5(rOk?1(YH*9x@<@~$2>eR-Xd8bkkapsmV8rtLi{6H)@dmz(B7I<@c
z;`rVx=I?yMpZ;I|YZ8g4)VVtyc{)FF_wprgoA3>d4}SOwFAu`$Fq4l5nz*;k+qbWo
z)Oi2l1uu@7IVVn&l3V3kFTDTW1O4vc&u^MYheKrL5^S2letSdBCk$4D<J~>AdDYNG
zd>*ta9YafTjfJ|6w`^=T7ldlBXo7V;Bi6`SxtyQKR$11@+ix0qI&%N;io@|9(>n~s
zaz+@Fyu6oVTc#Yr`lccs0o}8I*_FK8^H9=s*yxg@dfSkYo1kFE_A|Tce6Zvct6t>}
z7}?iH+Sc1`Z$s2Q)3C9B1hGLGk7_0aHWpBgyfa{ET#rb8#z-1LQ$_baNrVRz!us8|
zYCCS!^F+wDD`jX5EcT3vAgOHMo14EI&t>najG!<WTsuq56U%{-c~!K%Y9xT~R1c~_
z^VZ*hb@tuE19zv>W-A~_JXE4Rpkcf0maZC%82@|O7!X?|i}I&ir{=*O-ub@1FgxqI
z(&`EphiINjWsb(If8kc^rrdm%X?CMqrPr0VR;<-1iG6VoHFOTPfu_uG9UOGC@iQLv
zg&Z0AGYH1(NjCclQ7^mu@V6h0Gz(SZUfSJy(T@YW_r0k#7VHk^>|f6h{`kJCPT)s3
zmJku7#}O^#=R|OeWsODJW39@D=RU(=w|g4Tt#64X-~*O+is@h*HGf~ZVWACdxb?^?
zee9=LSWL+Cj4OU=tNv^Vj0UzGHW^}V%f#Acp+7Br{||njpZv=|<@29?22FW-eTLJ_
z`r;)oKYhg?{MH9N+$r_4QkRuv7tXJ*eD>zd>2BuJ?|+x?eE5o!Hj*{wWaz3;CQkFr
znpY$#htnN0O>Tb?heGQ{Qm0eX9U^SaD5X$Ip(f90cuTfuus3n>k;K8~&}Kt=qh6n|
zEQmGQa(4fxR;l&s>!3HzZ@%XI<{R$rUr-)iay&hd(oE}$s5iXn&4Yp~CB{k$ydG|0
zM-b%QfG;l49F5&~En9AJdW}6Y5&%S^x{QP%=V0AfWY!z59IGvsfDp!mBj`jz7F1%N
z`{CcBgVTw(%3rns)dP)4xW32Ch7W|FBT?m#ylzxb12Q#s_;T!({kDwH+UeEZKs0UQ
z){O;5e5A%b81FG4)&ZG??AdP57KX1;B%A|&)xOU&xk=_EW>!f04Fm;$INt12{DX5&
zTSmB*RVJho3d@ZfA6Wu;d74HO$G;Tn*`8<$Dp%M{n4r1G9OR+uD`+agf(gt%4!->%
z*%o-Zd4(b(J}(bHTvSGvJ~Gj05^oR^0x?-l1lH~YX1uBxy^EU}%4XpkZy2s%-scv$
z9-{?c*{^x8a&9-&ZF6>HbK@X3*uPhI(Oi-eS;@eGZ?YF;3$A0OTP&owOVIk3<+&G;
zZ<Mh~w}ad)wp>&Sk<B7L8SFVrB(5(OvJHHdP=>~Ag<*9~R0_4NyD?(CdR^#s^~!CF
zj-!g{)Y^AeMsmm8FTeVn!;2UEJOBIt6<_@PbAI%vKjQA?OFsJHZ}aNYkGX##e0mS}
z%M+J#<>z00$#gjIi_gBM=nH=5_kW-7eewZieI(VlSamj57UTZzfl?+?SwR#n2eiy2
zQEFX1V2M(jCj_NT-dInK%etajxLz)}Vg|abbWJ|REoXZ4PFibZ75aKbdu4g^1)?#N
zpsn=lBkT1_nJ2E7Gf4}RCch`kBeE!0X-wsgJl`SeZo8)9nR499LH!&_=SCDz7<AoV
zIS0D!)!tC&8rvZ~-scu(v77ha@Dajg7Ef>_2XZh1%T{f~?*^pig?Q{W+v^yIhjwPt
zfN}b`W);jd5{;2Kg;vO-aqIu#*?nB)M%W<w7vf^E+jDPcd&S;~Yu7M-@HgImX<Tqr
zWAQ%mix##)@3I2HS$s91wzO3~asSorgSOk;jT#fzc3$Ye{q?O4@v=(X*f=1fYwRXf
zqeS$ktXBj;On*`(u+zwjJu>(l;2X<|Qpz@d4?I{{E?_?M*<wLCOJQfW>^%v>Yu`eo
zWP3>*bg<hJt0TzP(-;bgu}DUwHd@$15Bmv`po!Yv^kU1m{J{@{K8sRtT~`<V*v=3b
z4}BbT+aW%Jfi+Zn*hcUE>Z6?8f@|~!Hqb@{lOFWeolkDJ8^*?Zb)au-CJzcb)ooM(
zTh$W%ij>+`*=+v(F>oB9wd;1j$6oNq=K`M|L)c~9!(Qq2c}BBdFZAWS9ULa(gz;<X
z4J>7@y<?Vnc>g`V`RgBX{@Ei7!m?gDe)z!u{D1dfkkiZ&;r&yi*w@U52U>s4Z+-V;
z=F=UYeEI`kKMIEz?~#(BOJ&M{D9h!-ayc`V33_wZO!BpjxQU!k;8N!K#N~29Ql?2k
zbBvYkjiiZ=xUeox4a<dID^k+7_UgJ)=85%shI&P9<>}2^PeiQK)&)U0n^DSvx;D;F
zPlz-Q#}lRbl)Kih)bq-8Jd&pat<StmR`BcaLwY=SjT71Iol1d`aYv$|d)05-T6<14
zj#L&CyZPMxEI@YqYEOV#D3|wY)sp~E>O8U7*RslP=kGpdejRC8bH;(cvzTrP3vr<K
zMz*?DKzn{U4&9r*VR5D)zw%v2MZ3k)D8@%q+4rxvnyni?>2s?~ys{n;Y9vBCT0ua6
z4%y#MZN!X}qUmjHdv?%)H$>s@d40>E;IvB%ghFteUBB631>JWaj*+EG9Bgf53^&=l
zbK;}JZiZN<i&IUzRL>@3akxR0Y`oywRtWXx$SSardF~BS1gtgsx?nMgTQCSqg)Jht
zg`R_sD(20f-f6gUv2~~Pi6nO~MB>`V0`QgUhgXcGtnS=eYu(%u8(s~R(HF2W4>zRw
zDQ!W`)S_R|Nj;HrC_m#KYQ<+0;0JXlDi2~|^>nj1h1+FON7)?sBqK8litDzh2xdlW
zJANE}8SJ`MCz|%Mc^o8lSzXD_$i{rE<~9_^_Nu3MKZxo(LE&<~(v~Oxj>Gc6JVP8q
z&mR$%`Hr`_aCq-izMCh${^~1UzkSWC?|#C^hu<PCPaG!0`pS#xfSfO!8ysG|WTH`0
zqB8OD!AnXK#IB_I=vtAK+}e7YNY=w=W<qkoQX!>@rbnM`kurTgf=v`wuB*>gSX*bF
zPqYrA1zRpG3Z%qAGRxbyr2Ca>6S)X=xp1u)v|V}p<_n}<$O7lfBc|@#Rcm9O1<94W
z)66_4zJC3duV24rT`Kn8EAHR_fHW2Q+FAQbI!xZP5}|hmk{_blDFzz;JY75q*&(KV
z-_l(JLL7A5V)3?Dzj@39xFM13K_Qx;$hK}f15%l`<jO0WTh`L^wF7KcFF>p#zs_-f
zn7QS}22YLTPol480qgd4JBuvVzHjjT+BcY#(c>CbgeSw^^BP`Mx}mk>v%C3z(-74%
zdVpJ^>wDgsVTnx%?qX7}<g@Msl3NP*Ep$Grqlpu}ch+C?xc^o`VObZhOJ-nnAb|_V
zO}V+iIwa=f#))Sa^%*R(N3#}kcOT4?H(IDSgCfd~q^3Kjw&^6s!506%#piRrnE~5_
zi4oYd;)jo@&ta~_L8?1VZcyb~tE_AD;WY-TB^vH~x#(|ZJ2YeNwbN>2S>4gKn?Vwe
zha;LbdIuI_rcoC6EgSKn-aRs@j;?*1(8jhHe8G-^Ko%9%cqVC+yzd(Xi0f%P<%OF?
z)!MVltA;g!2Yicu(ZHbHXTFpY_Po1;ZCh5>b&O>7&&@fL^!d`et?;(7jPbf$uGGuu
zV*#?Sxh#opjj9DZ7P?N5jQRAyhx36~^+T@9h3oag<$PgXI)^E-EDKXzv9<bm#@=aG
zu;~SQc+UfZC5jYe6~rpjEYR=h>jmpZE;BY7!Xixbj3n{F^5x)%Ywe(gW(l-lI#KIN
zuM6#ZhHGa!H0+=(Pp?_uoSE)F^7-S4_Y3+$U!GVmk4!m{L|B$9hr=n>U!&E9H|Hm&
zdFIoPKj7<c-tg5gzw(}oW)AnAG#xn{@4^GhNP{UQ3R+?<x>uNrAymge-VqamAbE0e
z>7Y7;n-ZFB3DGU}IeGfau#ILA7U<iVdkZf)4!&p4Z0q@U(QtDh?-j^17SI+HT2wZp
zAGpIZ8&jxv#Oh5oeY*~2`|R=edk~6jeCR&{0tekIack~AfV2Mmxs2}?GZF$e9RkB)
zS@TTX$9`>gD@rt<9yi`}(CG(T(0XU-?>@)B{q-$fQgs`evmLzYx#8LNPqz(T+#>Hy
zR%q}1w;<ul85#gHdQ1Tsj*bJ`=(z95ps5b{VVFbRs*m^i#ql(|2L*37`D<jlTSb!g
z@tkyL4z*U@&%GOFLf18#*v1+{wg^pI7SgQJJFKm<)JE&R_)5;?39=2?$~UIt>7!z3
z+x7Sm9RH4l<+R`Xy)52?;l6Ps?%B|SY|o<K`rm}LwAsz)`ItuQK9z1L$H#ka6WnyK
z;=1@?@tpIsUdi=xWnCA~A_I%frK3`Rb<G5X^=_ltr+>*d0`+$DS8>-*%NbHZQub_}
z6;|<ya6><$*=@M%nag@cyJG7~?{Dd3+I&ZO^&xh6NwEc89_jt;O6>3jw|b>pOw7uD
zkh*nBk6xFuwuNpk%<3kj`GFQ$r2FM~<N2~aVXxm(%WL>xWqJDz=b!(ahwrWaJ<3FD
zH3ETS!<X}w`};fAg$N8s)^(v?uZVP(r<I(9!<>0|^@7LOZz=jEQ=S0fNrY4mOo!Q<
z*R7#>B8yW-HP3YMwV^T`c6S)ZkAXy*@7M-FiRTrAYY6*!;EQg6CbF$Xi3(n<Z_QF7
z5k$s3g#94iK6^{QV!e(`Q6iC(oqaeCdWf;Tw*RigfGW0q;Ypm>n6dJ_*}PS#<Be~x
z$cAaBvoSr$W>kXqE?~Wa^-!$e*n@a8%~kg1l<zHICQmjrE_3YR-o~EYufo`$(O@d>
zSLiOy<@y^KvICBxTQ^8CLv-^?zrE)Jx7(Y&EgpWe9hTU}L=Id~cgmzj_VI2vz7J-G
z+2-UZK}{5VsJ4IhekmIE8XDP(A2L4kK^T+2_Dp0~gn~Cie1Z00(s<UMy{PVnN{dAW
z)+HvD37GgSV;mN9OwXH2Cdn7h%``NuyA?K0D|Imsc%MVv)pG>k?MwQi8$L?I7k-1#
z01F?Wy+?AxEssI>;~EC7&O`#9@P5T$q^K&A)0SOrV?R4w%{`vVR2(<A&e|HSgFC1X
zO+pH^O7;aDm7v{DM2T0LCg_Lf`035p&zk7vK+6+4X8+m&Nrj|JY1O@F)SJIzopW1I
zIg*Oe*DID3DR-3VNRvcPQzSP9%IuU%fqs30bnrimS41giviQO0Q^wSxQ1wGLmx=S`
zf&lIMM7=zrVzld-`uLiq`pD4B<0JLy8@~Km=i%imG#9RQh0`6SWJHYP@xb}Akf*}3
zEUe3gtcmM-;ZQQ?>y>NH%y%!id+~z1yL%qrJaK+{;;v-cdI1rZOGP!4%ZyN|>sv?(
z$w!hNKxc%bKMH+rk06@J^9gjog*Uwc#uUW?X%5a?{3FjHOk?$|R=oM`^AVo2nYR5g
z`gJ!Di~px!8WPEy_2GZne-%j>aVsv#ovi(E9#?zw*M7*yoagPG$6+Uttc~6#!Jrdy
z5=Iytc1uV=MzRqw26P)fVd6CX4R+6YFXFeK2a)~EjT_*M$-XCAjD7=xYW6)`egifa
zgK<+^-Tr&6zxwb5?~prAXImq*V?oQho$1Q<5N<R5t#SO0LE6?VF(|2o_$uDVY!_|a
zypJ~ki3J?v8B=#oI7C@1F*?snuWrNqPH?+`PA3>JC>H+pDGBa5;|Fs$_Ut7xb(*0m
zmRV%OrOZszVY9Ve>k2i6oslox;XNoKNFEcIY+I09cHoh5OCI{Rcn0&ro6uc$q)4``
z%${ukiEFH_(@9uTGzGM$A$*I)<_)z5t<$ypEL`>MvEw}pNlwReW4uI39x$}*S<M*g
zqbj7V<Y|7k?(Sc{Pd&Z&EBL9IXYVN`nikF!QdbH}lY+?Xztbuy`?aN3p()p=##)6_
znwUz5a^&#x6EFJ*;ULd+OKy`aF$n08NWEg_)vU<X^#<bfz_zXdfklr^6|Ay6euE@o
zdHkBkufK%0B4t9wSZ&2#e?wgt%Kd?caJgJbrIM$aj&WWVro%m&5_bn>xn7w{p)FT>
zRCZ5KZ#mt+M5Y6lyx(!WzhioM$@P5U>HNq;c^TRXuY%X5lJbP+$*a@rm7EXc=>gK6
zPrG|(w5&h4F@|zzUVPnH2N8Q_X+PHD{^0s{i1wgd4uXnBfZ1K1jp4fL)r8!h@9YOi
ztYg~>buf#*R~*`7gLtH4qYrJG2t0ckKXS7l#=so8`Pkc^^GGo0=rCN9$4@CikO@ZS
zfW!v@gh*^hfbFX8nT5kij#$+APCG;Lxt?RpyVMPKf^j_9-~Jr`me>CeHW$v|Y37GO
zx`AC_hbKm}bMzEk$Y$Fh@mu%-WAi2&a{ggWYhy6>O&5JYO{yC2w>N48Zm@OYrWh(2
zQVu<VZwOM*BV{nCI@@EBL>ot?i7QOY9ND*equEn_REn3|-5i%4%f*Bgho@h^Uyvqs
z-*wqpME>1<aCZxFp8~mf?_x;Xho8^rb&U7entXBJQ*z_C-a8haMV(Me_n3b@#BCDI
zd-m#1ijjIuvg+R08qm<Q{cZj`ZM_1|z4Z$-gThJ&4d1FH3S}R@qpJSDXZMuPxjb8U
z$NQJm{*s|9L0~kRz4w%;2(1cvS*a#&BBuc#qD<K{{Zs|ZnJ)id+WxgkmLo~e1Ru4F
zyGKN3o&%t&x|<x~%<Rg_{{LUIUxu<XWRvW=062Mx2zR%O?gv%#$b)Kfx@1`)f&x&;
z6B*&|W~Tb;dr8oq*!OqHfu2sxr>`_DJ&>3WK~+e@9$bhTpzicqiDHmMuZ5<an1tG@
z5%RhcV>WWZcH`sMKhbhTw;S#Hiq@U;%kSwq5aYxgGbu!L5++U{LY}f!1`_Nu5Q&o=
zbg8K)qZyPiAi8Oq38xGB@?!Vf#dS-bX!jfU>lMj`Y0jK37fLNQ08cYWBKDPf`<ZgT
zlAr%Tm|y-1&Ks+$_c6~K_e2i8$b%d*3bRot4P4)arLEaLByc$Mij^{-A2RU{qxZ)?
z^ZmZ#Y`tU64me1B%?^mw-_o^6@TfTrnfT4a{rPY@x|82@t2h&r@q2vV8yW(^zmI>u
zF+lOpW415qM38VW`vzpi`1HQ+ZO?jSpd8m>xDTimr;h_;sJs6T@R40K!w!3EthF&O
zXCHzz|D`r)5^Wq&lIm?WhViOGPMMqyV@37&J_j%N6AU6=i6KL2$an6jpRF4>F!o0g
zkYs2M)Xn02*fm%d>W4v&3~LDW>br{#X+kz@3hOl=9f|#->6nuY|ML(6lS4p*>S_<i
zpqB|Tbb>1`Gl{w32STNH&%zG&S5EQpJF!J&^uBOZPToxWLOrw-{gXZcXx+Vwj-j9Z
zp3f1X;9XDcsF+A5rNlI4QhF#g#hD;>wa>Q52ae{8>9aGcZT=29`n?)t0uw$)a<;1V
z^DCr8Y}4bxKOhK$93%!NQEF&}HW9*%ZCV|P7MgTaW%lhXph*PbcV<_OP`X2+c_PQ;
zm0$FEMkR@$EkUbBb(8=>X&6hcHz3R)O50Xyy^>;J|MoL_d*f}tbDqp9riO&h1|O!B
z=qk+9V&7My5jCdSpGC|6t!1awLJH;}+$#|6x$a$AP8Uk=JfBX)<qYX;5MJ4s=R~3L
z{_R&5I<-}{(wWa^YKGQ!qzRHIdaVR@)}Q|)=P&<6o1cHP6Dah;7&LXp{k9?MLD~ES
z<;bypj=7xemPSkVd7^ZJi?VEE8~#1<2;dn9)Uc#c)P;X?D6g$x_fDl4Y}mb%Kh}2`
zP}C>>|DXwtHE^so4hUHXysB0>f*b@9ctA+DcbiDi3S+nB?{e%F%iPI!00961Nkl<Z
zwue2arVln`Vx%<=T)MSl2dEz?tqm)XPluu(2u9I;bW|Ef8r<K*`zQvpdX0|?_h{=r
z{ktJLV;JqCA|nqTRtqtiMMHS>iN}L9-ut_GMeopu^B<1HjN#SrC-R5l_$;I4>yPe5
z<cQityc=o0%?YlMeS~)VKX_FdoXgE)*p5{lV-t>_Di}KqOf2)L7(PKIZtgb})Y^^K
z*J?Zl^I!*M44>5DpWvf$-Hul<pP|nwSHaF@yQV{M<X%;Sku<2LqxNady!|s7%dL$^
zUc;ajkE7S-HKUT^Xyvtca(GlbWB4CJpby&er;Q`T2+rvG{OSQ{e$9hJV&mEf#~~Yn
zWqn5pNc6?74yZYlITEYv&qi{ZZ9X&(F*p6|{;V>dpssWpT6d%CYWJbP<%HxQl=TDY
zjlFF2y7T_+XHLsZxxS%V2{F<H(&<d-Akmcjorf554BHKIYe=3*G1GcyIW4UBEBk#X
z2NSPH0^5G)<#I8MKV`PkIg5FG&1q&iEtGvDw9eC)ue9xs#6ZOkubeK&5VKVikXm<`
z*luqG9y!-wn`^0h4558Fa#Z*7n@ZX@_(2#6ZtsqW4Py_e#jmRqO~w!waGQz8ZyL0Y
zgEp<|&%m&Bus>_W4P|J7$8+Fn3@bQwARPaCCG98~gK>0Nk*jUY-xdGI)BgCZehGXJ
z0~iM%52Qk@fYKg-YQOMd*yCM+Ka<D$_<T?LCkn58f{((X?tZWp|HrStZKfbb@qBoL
zc*|%zBO|!vVej!^ii|%HY_+-D=);G6L~dHO-9{;TV0hvS-QMSWXevk`1*_bKLEFKJ
z4#BEt*CqrR51<&Xu3iDaA@cfsIDOu<Y@p~H*to!|9QY~^iNAO|{h3BKULqO!v_THv
zqaIuw7sgQdp|!B@Y2Wn$7a4ss9KsmDQ^v;i7F7^E9?#%Tn&YqWej<-+_Grr=FyABV
zH5|k7?kAm*0tOX#ED7H%K5gEQ`GMd6(fNQM_M@Zx{xud<7crGMxXYxVG^$!&hc<{}
z8|@HDjS6b8&=5K7xu?k$#i&DR@mM(H{l{i0+R?tF<xcMf?K^0NzQgTGrE`1z1+-GP
zjTD9b<5yB!sQYH!;Ie=SaXyh_q!&ds(pn+Uk!@eCo04EIs{WkiX=2|C>vkiCK<$m|
z?M4o!bx9(u+nw|C3;A@0oasakA8i#R&*Wg9Wz-5$&?u%=*>@n4m*ic$SfP`%#e=T<
zQHa>HA4&OP>S&LF!pLL)&E{i=^H{OtP)Lv6*8e()4gn6BNxZ9N?1i7q8&z>0?V~^(
za8@_O;(%S`GfClJDW-hEd}i@H)+=V}Irf2LPy5{eeEwmM_i^r!6{Ov)HO6;4uHX1`
zjPBbJ7D?{H+8LDF;nMv1lONwhZ8qcd8Psz~Mz;E}yZFC*{cW8BrjXQ8-3v68$EmBM
zJHQ7`4;RzHLNa{IF2ryTv<>?41s|0@2PUb$p<Dau=6T~u3HF%=7hk;1v#u}LL)PVJ
zYe$gA5!P+33_;C0HyrQ;-*3t>AAe5#&y3$OKI7r(gt1(Fw>=IEL<n*E&UkP9j^qYt
zJ$}-1ct<pz0&xs-q{-O;+g^j_7Q>i3SeL}7&*eAAH$S^(AcRAx!%?9fGmsE05iZyR
zn|vN(0AE0$zZH}C`?sSK9NpM>7QU~T?Ain#*Km9<YsLTixeFo`KNwUV*9;l4hhqZ>
zhXd~LxEaO8&{rsr1GIO=7?qzkfl+s~?u<EwR?vN=mMiu4jnM8$UyVJqUa8xSvhASV
zXEu!@n8WXOw}UljrqkJuYzau5%oVhCQku~I7@{|a01x4%nw3h~R|w`RK#EW*+rF{P
z69lXaiqYV=7;Uh<7xH{^$W@rm&+K()+wV*<vP_l~SGGI7Hyd8pf`|<?X&b$jPt~s-
z06E!?Fua0d2xj~@HaWkR_B@zyETqR~D)Eu_*S)HJ6eDBC_uZ4@&yV<_0NNUj?$3C5
ze7@uFauhoUqsNQA?qXZydz}^iQ9vC99X_o*XnMX5Wt2bzA$c4KpC71$({%i9p9}aq
zKj5nbbfh~SKMX3kYnVPE7vCRSzOKe$B~~Dhxu<%;rhoOP{`J@2NJ$;8s18efp0KMP
zl5fNw&>>(M9e|zeMo*O(500~dQ%@h2`R8Dk5Q5Dff>mxYC5}^MT>8(@LiYm$8*r&F
z?!$ieGlUR>hsaxtnPu>TA6Vny-P?B<hk-M%Lm&)}G&s&b947nmpd4W!Vku_h;e7ZY
z`9N;KJ!Aj&YRU=*>ty<POyyB&k6@gmb;LV_kuVoVb%nDo91k52nzN&h4v7JOxlqk3
zI}6(}vqGct`Mqkz;S|4T@iFf?Her33464NMA(Nvs`)ilfj`>nacnmHNW|=d_h625I
zrL=y)S>sS1*LJ{?tu<;b#9*}50j~vDb{jLM+H9<>Xjx%<r>);8>l?aXDeDJXc0kZ(
zHs@U%DcSqa^L#uv5NQ=cHkEtR{h0TpkSVp2^MtDVc)44w_D<PWXl^ZSMZW!x-Vx&A
zj5|n+B4SBjDUqh^#e&_B-j#5=Ahpo%D@<m&9Rlp@8=*x1TuNMK-&B4BJM@Z_511~*
zkU@hV267Zw_=aqovAY%6-9EYriN?Ge*#4-VO*}k)e&8|>wY0yBZ?2{-Fac?|wHW-M
z|Fk|hI$>Yy?XTDNk;U>?-M!5sK|lRBtZ*6%>wg6i7|-^YZ>Y~MP19i?%6@zlPxf8g
zDB${|yG0%c+%Zf)-hXu3(W8XJ!KfUt@LzQ_|AW`x%szU!Cyt8H7qf<=V)k;Cp-Q&)
zX^5sEjZMHmW()$pS%_cQu^|sOQf=*6(8NRtc@(LiE{?U5W2kL6sdMmwLiql~<*^~P
z@!3-P?ofD0_uY}xf2Vvtlzj94e5l#KKMu*U;RYYF4nOpVPmB$Y?U#kaD`hNZ5!-ld
zAUpzo9!kf@3@6wjG$1x-9t7X$Z9_Jp(x`ld$L~6>jjAiDIVf<=6`5ASx(jZ!<gy;1
z$M9Fz01q;^F2Fea^agph!eDF=xKn5dW_tGdHBAU31KTjwBd#$#yq#Q0=Fi)R8@0o*
zx#%y1Kx@Spj1d8PYe)z*ZBG3)$g$S~ZHIk>?GE*hw9U-uT1A48<_XIk57fF7#WtGS
zJ6oy5l!=(DXj8}`5woJb(rAR3sMSP&HpfXs53%a1NHf^8t{aqMFxZq$Y?d=JO=xYF
z1?;ns!8EaFCHg$b0;{UeDNgf=+uIx4+Z*X}=Ja$X_QHDKEp%wxi1JJ!kx!@ZZim*L
z_5Ek$+)2wb8Z3I$Afy2!+LJZ9odLUcFG|Mx?*s1nv>uV*eqmm9j{?!HID_z=tHEG3
z0`8V-i`2+rqk{KmWPAtTV{8xmY|ZlcJTlw~<kS0nz78m=4Mhl_pyK0u*rzxk@}?vI
zT`~5c6-nY28iPo4>|LJ*;1UBioBY%f@Cn8SLdsUC@mc6rkG)Il-@zqSv>|Ft3q*`^
z*NtUliAddUc(Adu#VFbbr8?K&Pru=1K8Bb$o^Ez3!F**@jq#z4Y4WN<g26axwv*HC
znTPYIia1jAfQl@UYE&)ZP+Q0J7~15Nj*V=jyu~>F-74;gXCZq_-<v-KNzwRlSRHxK
zsv{id5v4bt8-q6*h^ckIPg?tYvfJ=U-I_q*;D84gv54Rn#oj{%5HYa~>hP9#OOhr+
zOnk04#y@}V{Ce-+aoC|MlE@)L8bYP`7<>mMh6x$8TY(Xh(ruagCLYN?Ya*JzjW8fY
z%m!c#3qQT4=|sbrIq3*T!m6|=q`px}Y+Y%BbqPmN@Hntb-4XSmraU9TX~jVR4=ZVI
zZBkeEeTB9Y_jlU;7h<~;kFMpJDgBO^X3D;!WheBVq&xJTsYiN+DP>6Z9j9nUczK#o
zW`YDt-O*m@ZAaZ<HxvkYayq5Z>kd>jPQ=jcKnaS*NbRtkPIM8b(+i{%l}OCl2kVVe
zlpX^)MMCT7x)D-fdU@jZ@ydRCN6U>g&BPSg+lKa?{rXN!GqtUc5*@5Mzx{%)AJoh5
zh)-Y9JX88alz1>?MAQueeWUtKC2s%h-E@Ea|Ft!O79Nn%13Qg$%ixv=v}2veCn2bW
zqZpOpTkxA^C+Oqzv_9rj&T<_4-KXLAFtxP8*x?}iWyjz2C=`u%BzRJzIJ?gaGeY!&
z(UKWIJ3kDvPVWz=0-KAO7)`@trXv16qZ5_>NOnUX);rRw6@~5ZKuJwGlUy~5nCD9y
zLmr1UMpRu8r(-&sA*O`3epI%ftul`4LOdHl9MATOJmo-WW|S5rm`y~Ev{tFw;9x8m
z(;(tn1;^F~^FhS*5K){-6OYgF{Y`nae8ZmkyNct`ON>^uj(=keb?kv|cJuAcX&%CB
zcGHFvT6Eq6y7_H2f^W(r6zDiH9LD<m^Svt}Syf>wI_sFa8wgtbW%5vYuj)p7E^^R4
zAM=b)9f%LG#E+6utk^a6dmiF<*U~fs9M@40G}yzQ-6-u~c8mJ8`n2HuWyQsaf3(VT
zFpEMAB=PL%@aPH$WEs3Ni!mZh#^?!wlqP#NKC#kJD0QRNLT|eVsGIUP%oln!qe5MN
zG%;Yvgfx+RMfZ)q-O+YK+Krsd?puNqQg-cyb)-Z1*n73>8<Sa<G^G@a)r=0?s6X3g
z(VqM{hf)gjDKjs2SXu_Ohg?mWrV4c0UM=d=L#lEx$FEu&G0!}`{6KGRP2MWzKBCZ+
zb$ch~ow$5Y5pKP+)fHhQg@8~XKQS*q(W7-MM_~ZRM#tyPua6<{(AU_1S5@n1wDQ}P
zV5`BPsKd`IdS@DwVe-C4f>=&^KTP{%w8T6H(x2A6aIAGcufs887*)Kg(@oWX?$c>A
z=2*uEHwGDE;x!x&w614pCdPXEtZA@WUy9N55q#eiwfp9EjHSvAW~3gp^1s^8e694q
zGB*6zUw_k~&>7xGZfhMAIrtE!nNDG_M}jkftmPlRB!OD_#Fja*n1fj)zxj<;5sS6;
zV$*!V+HdLg(_zq|nHN$J^VBy1QZxl;dt@q`jZHk-vJf7Si&;gOJ@vS_A$S+k%~Wet
z)%XCX^+sbHb_SP>TGhG%uQ(jr)ux1?-r^S^kYe`A)hqsE@M|5GsJpk03K)(VLf3Bg
zBXkh=TnXN_b>q>u-mE$raFh`%jqm;WfQDm}?9DnE!Sosi9@%{vo%pE2zwdM&!}Ad{
z*t!h^2d(r7llSl;GyhV=;|afe$v8OV!2Z34A~?ha&K?^bki&PTXzj=SKC)}$fSDeA
z?^X*V7^rs}7^6y$nYetR*R!XkEu=8n=GZDhce+|iptMLJS$={DN&&IXL`BF1O0yu8
z7#~JxK?KPVWAOvlp}y8=rO<03PTom~5T?lvSyh(vli|=I+8u35oT7Vm*+!9bX1`yJ
zkkP%44FM%1F*D5yb+;m~ceDjshf*r?iOc|YNay8(wo2RYw6YSWJ6aSuKSQ2?K>ut9
z4Wn)QWIKU(z|OJ%G(QOA-{t!fp%ZPG9!+EzTk_}6yWFu+`yqcMr5%(=Tbq5biQIO=
zKAs*M^I@gahw!w0SAhL?s4V`@GHh+c?p1HQJ%_&DhL&qo^y7K-(SnKm#_th>9pkY1
zaaRC23ZKR({!O=JZ!##_<MYX|&;NTI3UQhtP7o84r$_sZf;6Pp<8(S?-V&^0^frFv
zH-kU0pPsjajnF&270_^OLSq5-=7UVck#%)!(H3z)2w9MnA!MW0#Z1VL4eKy;bFTDw
zAn2VC^t=D>bKyQ*7VV(IaX}L@shF}hr0j=*Z|u!(locihp|sX<GkCkM<C?^YAf`I*
z(hf!s4jtjMk}<1Lrvdlg%toYlaGxWK?K3Ti8E}n?#eQdSs^aI$=rcY!W)i(Q6*LAr
zV1sS4s5z{P8e>R~L%}gqht2FB8yYOZyMbVXP`lSw5e8>Uh_t`ri}&`uf#TyJ8izyY
z-UGGy()=`&GFV`ICPk9daFK^j0ztYuj8jk@c^5+KLJyNAr6uc3WC({)sLcX+s7Tl#
z?nD}yNFWoU8K5C<^^qev26Iw%H__wyb@;W_atNBnh8iW9&r*z(-kGP#bPmgcriosg
zS#<b!>pGHFBlqh&r+G59aq9?;R_f#6RyEHp$rf@GA~9EDZ)6&^UjJ$@kgz~VMA>NV
z2K|b(NMCoPE{8=(2d9gV_3A9WQK`!yHgFtr0pK%4GuEeHi&6O>_g@X$7w@>cc=6@W
z1_Y#A<f*F9YJ4Av3@747t<CqeM|aa&I}GSP7fz#VwIb1>S@juV2#5v{H_GdnO@?s%
zOov1C_BosW=&+92tskDfyOu`X#^XVp7!cC2E`ktsxX_M$wS5l({tsS%)1i>36ESCp
zj?8^B$f)*5aXK#ia8L9F+N`2WIX|F~_NkTZ1FA55ypL*B-MnttDeHKs`T^3I`IOr>
z1dBQhDI?J!0=!*|j`@#=;xop1SVq8ihsEcgk&q5Nk8ag>lz<$aNg}3E!8KahAL?kv
zp#-7f8)QSf+l1!~X)<@xg1M)<qH0v&-p*ifmeA^>QV(NUX-70*y^amehBbuqB>Z|<
zCo(z`TO{FF5W_0Ps%|@2#YI%=85V<4ZLml=uIYE!akn>$g9a=9y&d<o!`~@)E~xu#
z<?SE$yMy}p%emjRH+GG*o{XCf$GsNZ<33uZxE}BM*qFLI=XPy|@waNGdJa*<lnEX`
ziH^^2n|KKBYt_{jKyO40>It3$90_(X_$HH5!jK?4{8F0Gn5eA~P(E8Icei^F{wq%N
zOzktI=yhMDwoaZFn)Q9el&pKLJ1I}(0A;<}-?x<%9vykB16ni&5y7Hvi5cyc5QW^o
z-zI7)+&|u!mM2n-D8ae~Rdl~2^BM9J#DyT)igbrDkL&L$WcAtNfPIej#dr1M<52Ko
zOj|ceopxvCRp`#N@k&{aZYa1(r7MT=j<Yq}XuCncy)`6#pHUzv_7T7Dr0qZ2rf$k-
z57Tm%p~O%5<zT9K-{_nF+@aAL)*wE2Gt^s2G1BWJbuuVk0OE(gU62@m%kcDpyT5-c
z#S7HG!=O2&Oi0PAJqz4%$&usI_>FLwCw!vS?ZvFFy}C7nHG>fp^=tv#a17pYNJ$mk
z`t`uAWAci-5P_)9@5Kh7goyaymk_-=`jiajAGVJNaJ=srPGjw3a1}2yD4*$ognG5Z
z%NR5fBogvuhl1n9bUm!agKsdscY1Q#>0pScJ#;Z8f*8MDdw}+dI*Y?sRAXu(#!VD?
z<Wx5f<GKTe(oTw2EsxgS(#^&I>k|fgEKF|(eDMk4z<VE+_n6VxgP($t4hBYlxZKK^
zu-G9#+IfRbh9$NSy8Ze2+?~NV5Fein@Xgu&vMhwrLD|(BobLlu9@9KOTgSij?j2bx
z)O`bOq?C<niR(VHb#S%X&nSZ^;8idg?S&*6)kve6k~>C+?W{_R!Jdom17xGd4lFqW
zcCaRomTk@KmLpcEG?d!ynTy%1Rl1lbl8V7Ktu}hCET;?WdgJZmgRft{aNq8>@t2Jv
zwh86|Ux7cPUVWx)AZ>siA|YM)d{csemNyC;vOJ?|gWGny6VhgBds9QEPRJ)35z+Ad
zS+6!IANSGT_d!Ls%KWqXd<;~((b~u6>z}>16-9iO;xnvZbI;L-s0_j8ATrp7J77BK
zXyvH%hp1G??(!K`@bA#aXZ3=?!70z1cPLjlW(mGc9^a?dN)(4FwOPgJ#FXB|jYX^J
zU;Mf5_J$5253jiP_*{-mjP+6PcL2&(__w_NMw^RBS}yhwh*)(cz7a`s$5M>@9~N)F
z8%1{%-{i$NHc58jz@`3}jX-a95rYkmBnG6P9@W!;CdAk)5j#C=S#9GXSO7?#?SSeH
ziFO*-UP&ZE%p98NIBvX}0<;<*TttXiph@>iU0XX&W@4ZYEIcJf^<m|6Af}n_Ijewk
z#5*w?n?%(`VR`mL&J*W)vqdk_@KZ^TG$$8F^|B+anX+=Qv2yl}HG1{#)l0AT&?f7~
z4vm8jldSs520#dT1P3+7tlp5^eo$0n48Uw1<EY5nM(EM1Ep@AH3Dj{A0V9YchGZLA
zu+Kg!3hVkFs`TR_je=JD?=5)6KjH|ZSv^n<s%mN&>X;+=apWtzS|u*d+F)>$I$-gd
zDJ=)2(y=a%-l=7?Lj&6gx<Z;3P{<yk+4pydk#euB`$i0b`E>T`(YoK5BHHojSyK#K
zcUEnV(2b6~ZyN!#8WGnQEc1lKKwiuiTx&B+=8!?s;h@{sD<PRxN@y$fs+>~hcKu*U
zwiX8D(QEZt3qT-bXmuk`v(0;2`>XpwOr+(B`|V0+Wm*>EbfL>`*Jt0M{Q~ty&CkU7
z3#60b^d2-d91YZ2Ijwi9KDvY;PCreVAU0<*X7lg{(*qt+RhlNW6(V-uy+JjjMi&br
z0g;Tvi1z*P78;whQGhFox#IRubB;0h7+t%u=G5jbz5#daxwAa-0L@_afsFKlihNQu
z2Zw<>(8WE$28~<Hql)^-Vi^115jb^pKM^n(8C62)hGy*Bks!IRe+>ivYp=iQP*AUK
z?A~>6VIFu`cbrzWky53(r%()mv`lmz3=&(EV<R{eeG-pBB9e)DGEeaip_+ItMz9|;
zb2U*L@P#x%uXcl?`NRXRFfC{^IHq>n;DWn^stexijw=!%c-V!RPW6X^+aY%&N|;hw
z#ojl5lJ1X|);yx@hKOxlW+phGAboTsA!d?@F+rM%1Va#Gs>H|`@k(#B5TF(l3y$z>
zv7vs7*($6<pOFnGcji|V!&bXfL?rk<6Cve^G-Yb*hLPhCk=Q%|$3^jqT)f>EQ8Z1i
z6EMAi6~jVOKX8FAZanB4pG3F%?W0=@RHp}iezpA$x{c!u)CcBbu#EW~$q@iFoOrvM
zA=$xn3ue($s?n%Z5G0t_eMo0=@(rfhwM((n`_8^wQd-#yF^Pxx7(A!Fal5?}`(~a(
zwGb3at?WSQa6T<QD>77l7#r65!1cR#9fr7hBxEhbJRgp>lI-wTNsP3&L8fS<J5sZR
zzuoSf=85Y~S#MYJyf9Chy5GF%dWbX4H^;vZ8;G#CX2n{69G#zEaXN7hnNl`-*^DDE
z(GL9>Q5srTXi6^ynV*UC#pZlnXbQnA?cTp<-&l8PHutL5u|!84dbI;~z{~^s4TeTr
zR*KV$ZG-ikg@O6EVzV$8a~>UM?`E&#ZoyVi`drOJw!jOykciRr5Um+>8^s{T-UbO<
zZBFI;nW9FDIiROU_p9o;3&HXv1{ni~ltK{y+_urH8|NA1+q`oR$GSi`AYQw_BLwX4
zKuH}Pf(D`1Zhhn+4t&Ov8LhnyRD+tbF2+Q6&+oAzs<+R{vysIm;2~#-)+&XZ$jKl8
z#p$P=2P6`GVHGjjr`iD(Q?iY^1vFH8^`TKWwvsrtmhpMV1D_rclYQ?`Glww{LI~pd
zvxn(a@CS5kwz$d_tI}KZRI3M7lYxJZtlDOqw1u=_?<sa@NC<@94o-CN&uj(B!vPXZ
zWo_!6HnwCQW9DH*C?@0zZqc1G5+W&rre>Rx*n+HMrZNW2aQL`r?KH<vgG40BC+@Y!
zAOa&#yK6&ZB=RXg+BbCZZsPN6yS~!;M4lExn(g{~SD>vTw8J9Vjuay|#PpeNY}d4P
zJKLjlLa)?)HOg)<-yc(F+v4e^ux~fZ!w{jDjaIF7k9nfEM%nJ9dZ*SKLCcXTF}U)r
zEBm%H&y#H!9paphjUpsx)2Kz!_AUe(HCmclVMmgYI=XiR1e@=OCTAQ)YK0VSR}C@R
zAitZVYHfvDHX6#D0{8o!kzMjR&m*TuFD8ymd7+d>p60)X^MI~_IGsRkIKOR``TWJI
z{<<5;D1h!r7wTHAgPUJKvRBO}P(we#hGVlEqzvD%L%{7e#+Lnnoy4ClL5*=ILonN|
zqf({SYOt0<itdy;{%ki<s6n}ASwHe%dbf38Gc$7|Xu)jWgNNuel&TJ!OF9P7oeI7C
zWKYqy+vgk*cW{Iun3elM+!2RvN&Y8|+TL0vG>nikdR%>UlN_silvUPsNN^M(exFnF
z-+TRSHW!B=WDH=%6Sf9kE1w1_%`2itX^q&0lp}GT9}qUS5!6~J)oo_QG!2G}Dd7|U
zc${`hr)#Bj_hOIZl$?;Lo(9&8#u);FThT@fIv;VIPB=c=P@8Leg)YJoH^_taVwJ<V
z9M0PJsz)62)UIf=E@ccTyXz6e4}pi(uV*DwtctI8cqull(qQbHB<6qPr-}dda8B*m
zAg>+<c|Hf|SUZ~Hfg4BXaF3z!@Gy!o5tBn}eWb}bd~j58dUT4t`*6)!QR2{<6-9mk
zfVIBjk84n%f#(4nj4%e-fTOh@uU-qS-ih)M!KtcKl5NHkH4=ItsJIp*9aUg=rK)(m
zr1VN%Z|t`Z*87!Sw@=1>;`hWhiBc*_%mA*s%W9q`J9v60#z4lzn!RB8%WWsn$T3o?
z(N;?-<Y{KgiG44e&lhJ1+4F7IFPV^pvR@BoN$VCjn$v_hn@%Oth-k3NIORmIl~OHV
z*}A4qSyxiup%;4=1C)jhF#E$fSXHUTHn2Ql@M$t~$G#~k$#?g}oP5hcO3>b|yJ{7t
z6NE(FZ%D7?lxW?YWb0~misTF8O71Y$88UYF18&n`Yw1YvD?1|DM5n3;W6O%Y-HIcz
zkfs#fC(ZYxvhS?-yUo~AKC(el9C`*H>T882zBeeeY8EcNLrV5tX|!Ura<hSl+SI7P
z8B$0<q>I~o^oI0KuhkBPpf*f)0h<M2bwsOgip}A&?1h(+8~U*NfycAmszFp^ehbDP
zbLGFm#Rf6O@OPLgNVN9b4uwI9bUuABeA&p+3eoX;jff0Bd^`ptqePHku*$=HE&42?
zpiN1R>zRt${!DMxHc$QZKoRJoXfT|zmx5pwosJ4JCXxht^?69Gcpr)luuS|`tJ=j6
z7&F07F=>vC`*55d4jdf(yYjH}jX}^_sHK?J%s2I>-OL3)nRwAUMK{!05RcC#1nE|h
z)&BULcCm<4CQ76!)HW8853Bru`o0qq#^o=yq8M&jTCs}N>B0wH@z^+Q=uGnpZFB;D
zILQb@QuWHqpq1|IFWb=Ns0a_PI=22E4z>N!S@<i)h&Jf0(rP89G0?MWSb8JrL>6If
z%}w)0mPKQTB}26PUe^!S>nr!`8>MbXTmRYeVBd|iuf5YG(ZI5d=b1Du*12hC$_duh
zD{ng(w{>N&JF1<SCR*v%CHr+rDcgZKNDl*&^ZU8qY+wwJ;cM-kIU8iuQHNfH+8t<#
zk+N^lt(%Tw`jwBjSDr6t>d>3?iizT^Kxv-+F)y=cM|VnHiK;A@XG$ptOKQASmA0>F
zub%CYk<eM+zmcAlJf9#YTD8wTOxb8Fdhdj`6Y>+P;ZXkasZ+OC@_AM8XS4vF*38kh
zwMr|S&)FJLp-V7Itq-V;;9?@gS|4=zp`tz<aL4m5;subc8*#WMs9I+ht&{E&ETcf8
zbqz5Pqq`>ri`i9m6r%^USF@N1*$5;EsG%g#P2d<By$yPOL0chBC+kkU8?!z0`|j5|
zCZjadJ=1yYnU0V(cW8LnX#9^~f1~%&kZX4fCQ*<$;16qqRgDf6;z&jFN?dJf+1=%^
z*6O%7UKJ0)Sg-()^-A3fDXvVX6FC`t($qEvV{c@lWJYF#YGEWGwN?&X%{=nmL@x+I
zY#`HGJF3Va!<rgvxp?4}>Q>)Fc0RNN>Y}!;(2CN6I2*;e`w|bxO0sP>t#@Y`IdtSl
z+UFc`31Zk{p6ns(^<c@^0cj8wqcXrg_ef^ztymVZ!*s_1%;0J(s-v>-%3U!RPQik1
z9JAL^xw}B_I1QZHksNZuEA!r(-5-h2or>Crg43Zt|L4(WxBlohe%mG3b7!+7!+x66
zU#hj^)s?z2GoeIL_Ev2+2ZZP}N0iof>UyKx-`VdUR>=yG7dR|KoW`mdZJR*=63y0p
z&ZP5+fI(TU?DTpgbY)*RqHaX@wJF<1tBv#13)3_`D2~3#w`R}`A_x46SibdDO~(;c
zt!oZ)JbQM{@U7JTci(qr&u|v)^tuzXrM{&YxZkhD$>FQI5#wa*SD=<k%88Jy8@DWL
zd;Zt^X2yH<ud{vC551O?|NcnO!uI|vz3wdYM2OBV^mtCX(#zEf9~C5>eLc0KV>z?{
zy%Bn{84BvNBO&K_tnXog9#f$2&64}_<OgtL+g7@ZBXdqH%i`<I-rwq#M-e9i$>%%j
z+7UU(CE4d{4)MfHOfxYVBqM6sh6bP*gsV8*+x!slp~(X>9rKK4YaYx^mWG5*D|Z^*
za~8T6V7|`ALBJ5g*^JlbbUu4g>%42d0kD_9$Dp}2LZ0ZovhORs6;hl?IYX>82?t!z
zc7Z5|zfkwnvDcks0HHBlz}rT{<~w3tz`jA**!LT>8#*oKs_6e+0GT=xAg0->L5HIy
zBvQ7YYpq&1Rqbvk5WHhC33;o#ZB!!kn8<k|#bipxHtd(3V=Z<tbbqjXR%=yTuk;Ru
zas=*l5+het_v-jwM5r`{9^5h2uNlP@&ZVrxl$n!33s!A6t6ID=*8t5oL=ms%)IOVH
zRZAQ82u556f?In;jIC=XTPbDV>pP*Wh*o+?#$M6p6?FC)hYxaNCghorW)lbwCrAvZ
zAFL8P<$Pe+HYj(r?nmd>dnE*=Hd`1$(__&buS&}1_DNsSGQv`<Yie}Rl^kd08jP0N
z3oXrrkP$7kvQxG@b$#Q0yP{sP6Eeyp&eb4G8$#<2(t>TKtx)xb%qQXkLZxpX^xL11
zk3Z3_ug3O3Nz+7{7t2{*@5X8ifjrOVqALb-l>N$bezFQPC3@{7iOlnb_pTPrs*r*m
z(v>mvH}Zw3bJW48vd+M3tq{a!H!TFT6m&CMVhDkK{a{%xsQ>w1H`e{eJWa%DW^V@T
zre1s&`qx8+CZJj^*IXOhW*c;9m3sTetxTS^9gt=aXpGkN2H9!bH=;IrIvtyrgh0U9
zS!5%D6s_LDSEMREn5B%M1P@3G!NR{ljj7gly9P8`Ek>Zo2EBDx)oUFEs~4>Tbb}!y
zPA-uIDT74HU=ul<suB(p&_na7k3_-IP1o9J>rUP7219uw<>3&FJ3os!{XL)>52u0F
zDq0IRdJl$HgpCpJ8%E~5M2D?6=*2N@yN20S*`a>l@%{H+f7?u<H)2qt8m+CA%?4>M
zl9}g)7$OCm*sF*sdV^8804Tdn1zdq_f*GTXrO|0r!-=IgqITM@M#GHBy!?af59st>
zES8YLVlv}8gu}HHwUgZ1<H6Op7IMxry;t|SamG>T3>VJk$!swt;*$N;wjtr*Ewx&h
z^VzE8*3b|vI?o3^y%kgLhDMBuO7_-LZGhz)sFsaZcUs+$KGB09Q}Y3HaY|@72J5DW
z=+tWM@u3emET%urAl$woj3hjPFoI~1!>)I1wyL`cgnBc|ah{m6q34RRqR?cx=%B}0
z@oH7EGeY{9vxo~4wb}Kq%~)e%ol^?F89ekp$14OYHa?5v#-LWaMK~&**p*o=C#;ct
zj?rmlqwYJ~dZVw~ppV%mk`@p9u|v|vBErbuR#&KtK<%B|^^LTxOmU*$-{{wGwD-Rt
z>(%asLaGzfd@=(%hgB7!R+~}CG*Sq0x7CL9KxsD5nwEu-C)WMLFm%8E-*N8UCZkqM
zbnA|}=b&gS>~**KS9ReM77}HRSa9Z3uzRD;AWk*N;dFUnFO|LQ_ME0q9U>f^r*(Gi
zfN(-eNQl0XH6#?D^#v0?R)=F^pq9<{tuY9dL=VD<tu=dThjd)5Ip7lzFYe@+H*&av
za#Ya*%z3ye`?$BOS?%~H8e_2egV?mv7?q{~9<~=K1g(@>sol6m<dv{z!Zxv-o{U;9
zf=WB~8*P=g-Px}1?CTBn8Cpu|Fwe{>p+T$**M2~(HZLe%44P`bjlFX0ABw@s>azk3
ze2zsvyv*EcrvE)$Qro^#H=)*zwwv<ve!r1oglV!WGrGFBRf}=;4P0Dl+V<TSnQcfh
z3?Ck_+(jk!-6|f(%!KAx+fHj4QX)(XkwmXMYuOC9Yx>v-QFqid)uv@E4C<!+-He58
zhGcl4P<`kNIAh2!xO9>(gpkn~yc&1@a8QP2b8uKKMCU`dO5JzLwwb?uOoyA|fZ_&h
z)ykb(H?R7wGf3MVjlu7)*aFhz18_Mk0n~*^5)&E*lfm2Ix`Xak^`n*rkC~;70Wdl?
z!cMQQc=p*%Ll~hoF*u~v3a#(Zf_<kWM^{ZjSvP3b5ggu1KF4XbQnov#?4%UPEwjug
z40d~PxW~%|+dh97RhSqZSJLeMG9Bra89^no`;f0U>s-q1gR*UC-3`tXiyRFRn|-+j
zeC=w{5(IbR4YazmZa4Pg8|@^-EM$iek#2*-X4q{n8#2u-(>PEZ7Iiz4ZClAHqjjUJ
za+xMmrw3CfBf{zVnd^H)(n3*XPQ&!i=5no9uOy4z!&WHU-3;P*kUebUOB6SgG()*@
zCYpn)x(!NSJEv^sc_Bp7DYNg!{nzjAhZ>}bqHU~&ZkZ4<**vLQ5i;gkc{-8vM5&Eh
zHd08mey7)&I0ZCjFC6XOhF;y}SLvl<p+Cty#=7{tYN&yYLXFzIZp@@c2hayYt}E3E
z6ge&CjI5nj3?2&zz3g<Y463~<hZT2F8>X-KJMXtET>_V<ue^NuJ@e_sd_|&fKef@z
zM&EAq?Ml0U_#Tr8nh9z_RjQpd57oCFpo5;<Yejl9qCnYfHV}ls=@2-UtsL;)AYkZ#
zm%)mq;M;uvJ9G-yk2j+3wy1ZnEK;OwE4tt5Ig|5DNSRWLqX+KsK3ERBEu55-dB}T3
z+q~KkgcQZG>2lzV<6-QrK}$s3=@Rz}YW4Vet0W>Kv+F9n)pClbDD+^;(Sf@MhjJ8Q
z_u=q>w;aBw*Fu`2|D0d)W|jKB???zNr!%TBO%v7m`YA^GCr8#8ojswr&wry~N@D?s
zGORLcz0&F@MRv6B??|F;62q8KB9byKPCl?Vv$bOM3S;Zg3F3x#y?7@e#`5yCykWEC
zWL<iz?DxjLt>l~!6>Rf4K#$Q|?HR&kv+&l?-Na1dEUAG)hzKDi66XIiUS>1=z26kr
zLTShUX`&(<oCSl{VnDQ!1dFol!nSrw-)T*$`^vuEsrzmnn=??16i^6V4@-i<sIhsI
z{f?ApF4N3nlrr+PkW>lxJ6n5naENd^Ev7k$mP5X6mfDt`byv%NRD%(rmdfqpo%!@+
z15*#n$&(4xPERkCZVWY|nk9V@nkd=}qLsGoMCr7$Qkw@hsj_Y>*>fv!51%k9VS%y_
zR+8CjuiM5&1e$H2k^|~?_n%+dZNoEsJP=Ymf{S|SF2Zv{gbP@2A8(w`7k~DR?wh7e
zt2a+gbldsv4hh{l{=?U3@byKgYVHVQ2C2QH!3Wku942<~)uC(f?owRT80fZ{9teUJ
zpwvRG_MF(W>iU6^KccpdVz*GK@}aB(l$9{ekIePfXk~@Elc@CI)Ko3>z0!Ln=0u7U
z($sDEt;?;|=+o|#OK-&$%~l5XQmvEnL*<Ceb;#F>(_xfZeUBL`@c;YQ-#Ak!Y#V%B
zZ-^vbzWhMGJhOhhQr8c(G^*}|ZmLl*$(CxPxHYe;^@h%vqyS@rUiKaB%{mXnsG5q4
zvZ7JRCL1_Q5OZdgXpn|Fg|c_2WIeF9V@NwNM5h57qo<nSr5nX@go;aV&e6wkd(^3E
z4+XB^Z6b!>(gv%+Tu;01bO|=Nbw^WiXvOH91?j~EDGD)6#AM3o?o>n@=vSJHSCT6D
zfLnaS9k{1Tu*xEJh|z_0jo7L)6gn!#N($z)VS=6khe<$^(KgkYA0p62*?nMMsHVD=
z!J>fHBHCBN)QPjZ;*xDhTgPzR`P5>A)0na`VR}K3!<Wc7*}r*sm19un{4d1gD-Roi
zF=Nn9A}~{VFU4jz)dzAZAU&eh^%Z?L`fpFv<5xKWK%B#mB2k#;lRqO$Oc2XTispNi
zbEImg1baq9o(#|Jt}`t=Eea5*y^vF;?n+o@dh6^yl+M$_-W8b_r#(kPoJeAaY>c3_
znp3N~sa>z^w-3(Cgt&!<F|;DXCv1cl_0}*>fK{31P2sv8AAa9&ch1k#;g)LM<M*(G
zv3j;cteJyRYc&K)Eu>%vm$pVAnhm%#w)IYs$UL9ySt?iNxe;?h^klmE-c4-Up}OMU
zOy)98u<KfP!`a2V(C+JDDAC=ry{g*W1EX69x1BKP$ikJZ(|E9wY-npV*dYYNEhW?q
z7w0@NJ-;wK+o(6{ZAJE(5Kq3gnxiGfNH0Ser{2MI(9&qyu-uj4g{*0{qL`~daT=?A
zKjSQFS9@+{WW8817>_j@-9e$5q8lr4Rew`${dZn}+e|^tylXibjFFcMIz2NzCzwz4
zwvl3DS{4iX5Qi3eA+9$+nN1rozJPYS@vXYd-}*J%L{mn>#0a^M!K$w)!{?<SX)NAs
z8$O0n&(q@K5bHFmzprh^)w@)q5E_Ncr~fe#g2Qp@FiLk}Ot(RBtkERtZt1X06J3>k
zFV><3rR=N0ZFRRcP>kZ*eHc{t4Q;#834^DA5zIR&ro$UOr)1U3e#bF}!!ko_b>W&3
z`-VSD2+4E_-9CG?XW0cqow_@}-Wfx!RTHY!Mx2`2X2k5BjaAg?6NE%pFP!4RhN*3(
z*JzMSOfINNbeh{jbjIL8^Z_X~zsBXjwEOYh$IE6V@mQ?t&ydZ96cu&&uySNmv_{_>
z%@2)~5@m0!`%2w+H=eUuQA*i1i<k*{6cb8wc!!Xv$qI*Mnwexr>z%&6((i91EjEvs
z-8s~U;+;;;8Lfq0>Vfh03M!RS4ZaANcnnqcZ6kO4ZZYMf@&mKFs<l!v3zE9siK+%s
zjFl`xDLaarPX>$a%)yE*ZR%pTo#dV2urW_5vEFaQlu66UM^W8pNsrf1!Ix%+kEpuS
zi%`ps2Fr49rJ^|#b8?sC%)VP!J<Su+gtES~mz{Kike=P-u2?ZwO>~DsOirt$TgE^e
zwl@O3qmtZihQ}t39u+nc$ibkX-a55c>!`59AO<0&+0yS60&cbLep0nmBxQ2Sq&yMm
zcI{hbuXnl@LJ!0+5j6N7lIWt;Dlknb?I6GOzFTpHF*8-w?J<Ua=P~aw1i{u>^G>SO
z@g0SftV?KW+{+Pe7FsjYP88{pH2pP;>|cBRO^3pKdggNZk>%pJRT78|Qbwm2Q<tjI
zODE^_#$-HRNt1J%3n@)D5NejtKyQdF(5!;C@D<zmjiF=Q`C83;CWhdw2elz%Z3OM8
zn0`Q8rK~q&o|vqXR?8ua7iR=jZyC-12P9@nD<BqdVG?pB`)s1>PTx%jAs*$}TD7XY
zEA-C#@y0YwlwQ#FM(s+IhsAjC&ZF&{v4UcvVS_fP#$y(7=q;jQGDRzqT2>z-TW2K6
zQCJ<bKRq^nOS-dRvY8A9vtKCIDr=0yIoV<H34g5nofvmCdMJ!<7%~Q1@JVkVM|QJv
zw(D_!N9vtHz3fADBUUMc9mcI1gJSq1g+p21emiGIP-kKJ5WMYxVecA}Lxw{u(amrU
zwe+rL@|~^^DzEtW8Uwi)poux8vRfB3`0>_~TZMbRT!^8g>kX-Q_WK)c`=Du|PZv@Q
z<ReKdASt<mTB)Tljl*6{=`7_=tIak{39Ppp%d$|n8%D1d(zH-^bIGiAJMd~jDYp;O
zoQb`&-R_50D8_U^e<20xzPsO3r#59;BKLbEr`cGR(Uk4AY^W+_zY<fh%3OrK*6+tW
zKEv#__&g%ASEGbZ0^9v=F`*LZf(Ke<W7#BQh9Q<nCoxdhD@g;AD=o~_l!<*f(N>ZM
zD1|~J`BRjGshw-Hs@ghCfQOhict>gwpygkey3u>3b|LJY9G!Tg5uz0Xy;Y+4xOtdX
zj`?RCP^cb6nUK$B6XKmU+Kph7v(eA9cMaWOP%91%BHBwKb<0U;3hEX++AEO;t)Q(Q
z&duugq?BTixCE-&+|alYfiTTzpH0Nu3Q-a{i#c5X8(x3gpc#dL4U@V}#@CN;EaxZY
z^D}g%u2<^5kx*(;Xqk``k--bLMQ)bU>H{fPw)&uyU{d~|HiEW+w34XQ?Z#d<-vk3O
z6|XuWVne4Pnh6p}+0@bEdWBl+0ac9g*=QlfDQhUyWbjDqwvogbF|*!am3U0ZsGcRd
zT|j5scDF}1#z`^mpy3EA2m8s4zHEL#%+~p|Zh<O$^+Bq^4RKlsQHYuj95woc+=_3b
z$axWu<0*^`ficSna9jg0?s$m7hJ`+SHBX+7?ze*lF$7M*?EwN-(F+aKs-5VZ^nJUt
z-a8|b5w#rRlrv3-)<YaN;;>s7j<_03ck<nH*VRO7L7bp~cM1w>mEa+M6w4!?vyj71
zFV$c*H{p9g4uLfJu+E>s-rXU$mIFGFkO@^tc_#Q`?4=^b9ZQ34VAM*S7Hj9N*SF0>
zEsr9ihA#`$%~&icUMTx+*(5&Cs=Xt1J0Jt~3ci=_gaI4!s@Ng7Z!4`UO)JaOlUMDD
zS_&~HTB$xevq5Ur?)rsCK6WXEDNhJc_MNg7V%*61JTU&IbJI>O#c0I6lbl;p>%i=-
zlc>Fsr^)mL?jCHNOEvaUi1fCw%ro2l8;KQ3UpeL9Q}PK_%cuz6-P`Qo@oY#4CW^%7
zO2aweyA3XyHp*;qyn1T=KmmHOKQnY2tyZWTRWRzWL|>}|`n5w%hE)v~8GKN{tIg43
zN^as3{NOAE0wGMt;SstC5ThSB9V-@0_*rS6a!`y;Klwf}e#gjvwza-(yR99?*<?n7
z(0umHm%jsi)ZB00qR4HCF{$xr>t389CZy2WZ*SDvs43BNMlnyB-Z~*Rf<$U9Xs_gG
z#!{+Sra;PQoYAhdeWR6?6dJUh?fQ<=!12LY+#f`ZrDQOVen-(v<CGm$v{;qfonNdN
z1Igy5)_%trU0f8|+@2x~x|syxXF(EBYd~UR%9%WS)T4lCA!bv;9zor1Igmt+PAEd#
z>oF?{>g<?ax!+e}*^dEl!0Ds45~-$OjOp-<ioyNu+enEsuPQVDGb1`vD%;$rIqN7m
zbFbB6@Iz|{;|XuEec0g;pR<XiemayI0b2mWpWfiRh_;ECCy>ZzH7Rf~iiYy|x2;0&
z)>*j^7^>qiGPn?2&8>Fm1u^dYoC0|_PCaq<gG`AbIVr)d9$Ix7$DlphO${zonzGSb
z-(J7bwIkZdaUq_6G#F`ZhYh@a$3dO7nM>$^blSF>LoR`>6mkkS)D?qzL=vTSa_@v_
zb_uF}Ad8hTpp<3=0o+0+pH7a24@I`Q=0ranf<qEP67!U7xIDT*+#xqcLeI=gV&7M`
zbtQysI)?rmdxSBF|1`Kv(JdVuVjQu(wG^l1HZ;j$)v|BqBAfEF-IJ7wLD}x#$O!4n
z|4QXT-4k6lQ!owq<e(%A5-h`{i%`~;wioY6tQaiCqqw}F&{}BwPB&_Y(LB3#Z>2Yb
z=Ui7p7Sd@ZFBf9S&SEMC2ex7dfUgDZ!y3f}b@OZ#Z0w-vwfKz34E`9IGRWa;W>)ex
zLkO<nK@^c>j?igxZ?NR1eAW*(4>y~JoqDSdlkJqUkuPS!Qf?prrhWcz7O&O)PTOH*
z$I_dbG^ss=z3tR(rR|&9^7dqHxfrWIrx{hD)WW)&pra;ZrqtF+sgb(Sqfrxk*(mFk
zCZ>@<gjyA?J27T7I!xiwoKb|3Ox#7cw%vSM>`;htV1>TnIZJ1Reg}!1PA6ANW|ORo
z-^{u{+!CWHP<vO$?49wHvYCjbM6VfSvyElSG!?e37^dB^)*8dmw{0gx$6k|}qXpk6
zYuzZlBF-eS?`sdH%~RYQPOW+{B*|bI@y)_c89mrp1oxirn95u2XOBE>0B1HFx&YrW
z4|h$sFt+!LJ}+`pa>K5i7jBjjS=FFGu|ec_&zAqy+h<>RfBEbOhoWo{?Lt7W(0c1O
zx2U6ff>R1ya%AfmTSvysB_MgC4zHj*AM=tifb0%s<|#q%+^_E-f%$wQPA6CK3)CC5
zjapWw6pr~<_v;x$ps5Kdx7!DGyVDg)H#59xKGCbem84AXVw6pRIEl}{<WPd+l-=gm
z=@KH%$qf5m$dlVlXPl)a4x1QTXPWY8Bwcy#zq>!Kc}m&MEAipI|89Sch|001`rz+(
z@A!dAB~5M&S6#5zF#^M|d+7!0JFt-Av%yjA%JudCVoDQX{+)@lLI5K4Kr6-IzTT<Y
z?ZEN8kdc?5+7Aa>RVBs8ckbQP&?AJ#l_^Fxach$SKZzP_net4C3qd<Q>~wI)1cR!B
zn2T`?9v0jix~{ZpmMb>P83S`KD4jXfYN51>xY480h<(0>nnOZN28D@rX{N)nHLJid
zKb43%2&n1ATC>E)c7L~`t=#`j`~2T@D15xWa`NF!YmJ;Td#SYRUr_$UG?|J~L=4AN
zn;EEw-bgf-Ojl*Eo!S~@+t@dlCJXkE(9p6HoJv*J8>OsNy2<flCeLOQk&;Q;oBGC3
z%_3ToJ(xXID8-G-oWq-<4P=9mvsu+jn7l2uO2)iQ>;VJIZ9oLA)w&xs@BMKo)YW9<
z#zg2u?Mz;ohY;AdjgAuMlR-k-2x>REZS{Z!G!BxpAACAIKRa5L&8v6Cyn3Q?Ktr`u
zui%Y3XVrM}onHLAeB$_vc!mF<>sem3SL?}!(Bu7^<Nsa9W@CrV2s=q)Y?lMRa)y`X
z9GCAl{NaLWo4p(bL+_Sl@woy8WAET*Wd}_;fOx)K>qCra@PKxkYvlk>QzDecy_92^
zYcrTJJCLJYKROi`XG$<N^S#`V5Iq|~328bQMw9LYX{_(Rn24yE-;=7+y|e7&cviO{
zsb=oH-EO3q2o6=M8tyB-(dtIbxH2558|t4`LU2kgZg*m~&B+gen4|f3Da0w8dOJ_X
z02Dnut)jcBPB}YFVT0M|45}W?G9motj8-*bK!}z#p*q-X_C2$|pTSg_y%~vwAXc9{
zyw+N!_RVL8f$I4x+A90c{|E7DqouEC1cUcX@jK417$f8f4T+Ryq6kep>+QqtHD+EI
zeDG$jtj%*zvM~^SBQ`6UPWR0A5YyqX+G|02uv5)_iF)1b5Og<WBe|GxuC;2)X@|rS
ztW$A@k%}1G%@<Q?JEBJ784a}1Nn$gM?xYvh9jzPt`i_P8*!;+HMIzdwALa&Ig)}k6
zOcZmj-S)qOA>MYq(w@(R7}@uY6fFL)wMytat?y=W;GQ_DO6ewKkyuHwq8+wk6jhM~
zGO_J<dT-8DYRJAidj(Q3aff0RUCfL7@)*}0#ODd>NpN`OXf2!8V$6!18IhhMnA$Yj
zFk7R0ymuQ?*4oJ-9U2A8A9gN&1SEO8YnQJk^MA2nr?C~bzPc51=YGGkzrQh`pUCq<
zoEF0o1$r?WZWajhLe3L8O_Wk7`-ULo7|rMl&-=x)4Mefn(O~lT0qxlAARHV2U@Tb#
z>H}61IN&GL*i5|@n#s@spXp$<#zD8my=-EPo|k~ZAgyskBKo79rim#{Hsf$XkliDP
z@sz{Kvb9Er!@Bx-?$$fKeGs%;bx9<pKnPBmlt2g%&BNHkf+}+`m~ihd!ivcjZg0+<
zu|aDRr)&nZ{|=$Mi>z9HwlizeJOhRWr@RpK3$fl=YGJ>>*|&B&V@#2<7sTH~&S&a&
zC&X@Uu7)|7vv0BQ6{eZ#{LKB@j^|p`)ZY@*(d`nW(IydNJvm#cx@xv}pI52pn@p2g
zpoHPU15-@YQV&RyU;<8c;KPJZj>&&|i3ruho&s_l3hB_|*t64}_%L8e%PTQ$%{0&S
zl<38*-J3Mx9B6A~nxClKH@0v86;6c~XEbJFoQeMXYz<oxs*;S@;qcF1c4{f4lx;?%
zW<ZG!CuW0eQl9Cm#+d3IacxP8rlIO;bmsN?LEys<3#@>Mz6Pq((9PP$#C*f0SJFgE
zGsxq2kTP+aO?X%eT?0|^;2-O%%f3Qu%qf`zuC&8Zc)eeF|MkE6oMz^9eqvglh&q||
zO;aQ$@;o!mHe9d!ilm9W{vE_?A;7k+$aJPV{wd;>kelL#kiFf?Os6xFBG>m<+IEFt
z?b&^=>|MB=pNV0D5Xr#|!iGy=Z#z<zG}%EAr!zUvs2Vj-1;XSv9S{}Le6qHxH|n<1
z^=9qlUQm(+$T+Q3x_zOJFpc!D?#`jTvh6EDyA9+*Fi0pQ8-&K$e4Yl~D^A9IQj??a
zn+?%&UQlUF%gpWdm45qR&J&~)DoUI(dA?Bh8?COS=|qgC7@ntqh*^C{7LHg#=aY}c
zry<c(L}WthZXEV#VwTn|fP-Lm%SLiDr%rLmtrJMI=RV-jkpKT9?J9)PuRK!Gnv07J
z!xvYyI@Q+1C3bMelg)^ZVWD&0<CraA)ioBw+K)=-@mkj#ZBNbuv+t*l*DHExKvXf^
zihIb2DpCr8&J>^z7$RG>*1XC+28F(18lvWhB^WEK*&sTVowlvky61%!Pjp#qfEZ?!
z4J?4ty4Ildw%Ie663fe1?(0fKOrRzat&O^EoG;JBDIxJfa^{Vdp@}BJpL;!Eoe_A|
zJciXYX->5b*{jQL3pXTFZ;R7U#jLAiaIvBb+C+r1Z_MuhCDDxvzq==n!*_$^21R=B
zl_@VMHvH*L&6OYwyBV8FwK{C?%>b}=A%xkG0FCVTE7Rp4*<XJq*I$tIdu}xmMLnI+
z4!YJMj+ucofy_#L0fc!<r1RN2yO_|jQd|LUbDA*{5SzCN1`)MN(M}uo?jaE4WH}>a
zzUJD1<gv8FZ%jtwqeRktCZ);d3|%3b$JG%BpwVftc1`y6o%{WRkN0<a?>s+UIGs+0
zvWOiD+q!`ULdfKKu_Vm6(35@V<`B9DsOqZy<K71TEw8`nP&l72%uheU@<c5=+x-J6
zcYMM}h_-=Cn#l8+)8&~o&E$0E{oAkHuCJuf$jh0OCr;-lV?3;uIgoNjTVr2u^cI;#
zI87(=d?L*=4I2XA*E{>ZGM~<*X+rUk>|P8)(QaLf&k~4^%jJx;Ky_vW#q$^r9~gV!
zVzfby)}2<0Rk8cdv|I=|5Ry^GYT2mT%t%a>y){}>>nOT)0jKlAv@9I?xup{HZbSQS
zT|o@d2iNLmUPfCJ4+aN(c$bbqdh5WG>ly;)sV>gfFSH)KQmfY7co)%Iv7wh&!5Yjr
zhpukDjpMlXn{9p_K#ba_*dAjL!bYsNuq{F`A>v*vu92u^cS*j`J%SSk)@`_OM-6Jr
zg5-&izNgOJYo)9!r+HzXXM;C<5nJdF4h<L!Lqa6XCnABOH!B!4kffN2oIvfM6zdRu
z!ygCl0183%zOc^*VO@>Fn5Rq^WxL-|g-gt628|Qp`~w6bX+yegFk6MP-H5s)LAb3e
zQG}<b3-f8AHYKLTx>{Apw#iKM6G}$TPt@(^W4rF%?rd`majz+dfecDbCJq(bkVHDQ
z>^`V`sI?J;@w!;v$6&L$*3@P!F|utNr^^$iR13KJ^fHvl<`YBT5M$(iw~oKo-HQyR
z7D&@{-1BjbOR;%aOcO@<bH>&*k!gr3&o3{0TyKQDk%7J!LSAUq)@09}<m3>m5dtY^
z@;noGvL|ku{JAv?6CHLgSdPoE!yb@oEvusggNkkH^dL4rsI}N<4Yo(5V27vr`~iX5
ztHZ07uTjeGIxK^M(10X=&)$ehsHITqMkzacSpl=***at|@_M&u&(qUS28(4oP@7sx
zU39CPfMsd)u-n>jXzR0k!2js=H)b>kmZulyFTbajo!38C%KGL-c7Tvc^BF>-p=`zY
z(Rp6DeEmuf<`=&_z0ka((^`l#NGsH8R=+7lYBMg8d9_qNKJFB4yuH2h=Rf@sRQba{
z{3Ad7@Dqg0-eKEUudWQ9Xx*rvlVkas0(qX$Jh5#?iA*`!0JEFlNlM8vZ99!gg7(c0
zhqxLx9}EMpW%rI|w~ehj#Vpxy5zG%J#>DgaiPkD*+mX6jjNzW?`D9o)g_I4p*|yye
zNvmdRGcTHfH(G{;`$egF>w{zVy*UlipfYXdAw{$t3(;mj?{rnfl&57Z0^@nuLlmOX
z-aPTEc7q?f4{}p7vO)8qc1SZzOtiYOm*P}XyT;MY=0-L}s|B}owYiB4q+?#3Ir4Z-
z%Y`H_*0m*19+P-DO{zjwH&g4Nr5H2K;jiROGno>5hqfENQG5VvkWAtc8gGa2zMI~D
zo@Z*Utn1w&xxnSgoLzg&Ov@J}&!jX_%gsD_N(KEBy8S}b$eb?(ZD<9po%3=+f?AyB
zJQGe|VR~X~Grg?Tx*2K074RZ~(g86uwc*;U!?I+sV6XxK_nd0I5~K6=1GLhe*=A6J
zVb%tjfgKdXzp6P4t@hs-;^FhN$YF0Wx>p^Wt?jIoeY5)(lMw;BJ--&iX>gCF@Odbe
zV7V13Q}>Cd>B4O-oR>5A-Pnlg9$vu(t9IbocSw1LXz=p=b|q~q!zZlw?hL=#)9g$?
zF=eyG?yb^puLkS1{eY;W*nw+|AcGY{GIm*)WNS?VlHGoN<ca$X%{ne)lJ%;<fM<s|
z)}H;8GndN~)3l%%@)2S*=uw1LHs*PuspXz@I&0bJ(wRLAB}qWr?wR!-s+DJ2??(9f
zH@yC~nZo_bc726pR<t=sbV}@Np=>*0YAiuWQ(~_hx9cnG?VUVLG&)5KA)J}dUznCN
zr4&BC{mgdXNPRPNyM0Bp@%s9W*Vi{<h%C!QjDgY$xBJSkzx+xb&Xc9FoG<>x?HEC9
z5zLn_Oy?KF;#(m^vqhd#AfL=4AxWp)KlpgNITbY##lk;YtyVEK8)BDQkh~j|wVa4z
zwhd)lT~rbf3{El4!Aj|k#0Z$_q$o%iT5$iLXs(ib+06zZM0&?WVeYFxSVrB4p#u{|
z1HEll3Ft8IYd((|)o$|*xG_@XaDYa!424oV)#-f$7uH%#@k%5_n5LO5ks`qZOS;cF
zoZ(eA!?i`wFrm=~>Jk&(;{ru2Bii`p4r`FydfbihntJ%i<Mn)grc7UKP$y=aZ1}DD
z6RAWbFNg8l{eENr_zR&clY1ONCxu80!juzrFH|+rkSSYXA8xhS=b5KWh{@8_dMBkw
z*>?+kaZzN)#EBtYDCe2lc0v?p!6Gf+5^ep3xFmCxs5^z7pu!XuBOjbz;QS*RFRTUD
zZ@*B=m6w+%Qp^qk2_wI^t5RBHiqQts1Im;@h;F^$MtH67e(2h@uB?QVsl|x{z0<bM
z?C+hhFrcCBekbS5zOM&U*SBE6)YUoXgPj&aa5cJhYzH%`Y~)V|(fHX~?EcwYK&ZVQ
zP++eH%T75G!b<T&k6!J*i&E5Fng{*0s|7}BGaHC?y)!2huEl7off%z5dy{RHd#TJZ
zle4?H>O(*of;qT0HJwUM0iw@s{CDCTx+(i@mZR=OJmz>NRO=XLz{8O|VCM)QlZ>Ux
zC6?3dOe#}22WK3%ZmEgWaykl_kb)1Ll^BGPhB&N#wEH#bBYXUR@}WSl=>3(CePcSE
z>1|`L6{0Yu83|D9jwBzp(JA*kZ{L1~7>K76RVyFYH=ZO|t=kKcYMWyhuG@{)D%bms
z|NJli!t2{R|I0u9BY*haPk?Y*o{%4trJr@U-S6Bs8vw=_nX?V0r3dCOXO`#RLGNt0
zH(FmE=S_z96X|WItFUgXak`_aqiut+%DXpXzXZ(7Bz58I>B796sAXs0SN3f;kzO=v
znxx3>zL{7~thzQYeW7fXb>EPvOp|+@IKyU|7IMl=)9m7&PVbu!vN;Zg7#k_O(%03W
z5i{NDVffx)8;N7f+TAs>tEH_K8!Xq>J!3YxMfT*$Ue#!;8VIrY+ywWUQPk;zF$U<F
zhCK^8P4@75OFev8#(ctBIp^}@v_~FP!1fzTYDkH|#E8?>S}0{RWx5DGWb*mK>G_4!
zluduJN<Cr=OKs71Ac=JhwcCJuWEAharFa=HzaJJ6>KjRn&i@XBA1_{k`v6}JN{TTN
zqUknx`FD^{$o3NnbB@(=XEtP-Ei6puFUWL3l7;rXzWoJ#{e_r8^i1n!u@HiiM!dPg
zaPU;c4=C#vhBCZqA>gTX4ugegRwyAvRBUFUbvUU8sw-rhdm&LPP4GkY@fx&Viz~H)
z1vo^JhPyW(MRfLk!*<T#%p|j-sns|5Vvfslbb<*fcPgO8M61##=*CbaI`z}xRqIwd
zr54)#&WQL;0VG`<19cU?&$_z6zHjX7-55Rh#J26kb~g))V3s$MW_k$ZXo-+9Btu5r
z=BBs1&6=E8GgST}c;FTX+bV_GI%D@K8cxXOYi97$lng#?PD2*Ai>bXiF$Yu)CevP-
zByfuMtOlRow4Bivi1UIFDEALSGLz1K%j<8ODLg-OStdU28(OVuIiJo3sfbWpW8Lm_
z5vDw|-LG6f-np+=8l6k~Ld-Ki|M{Ii|M9<anP)@`z3gTML1mgw>}6*@Kk*;__x~3X
zC!WqTNzCggr_9%%GWYw9Uw{6Y_1h~?FE9M=AODFwpNKirnzC=kQ#R`ZgBi-cgI0o6
z%V3C^-pt=-UM`%@;V{DkC6&Vcc4ONN8^0`$jkn_M-b8Jk9jFc&B=xi<_cnihedT`r
zV47#c;)AfRg^!O9T8Amm4w=D{CrDOd-EMb(0Bp$Fnx)wVF$dXg+sU~yFSGNVEvLe?
zB1TKKfz||0p;K2-8qZqUw}SSaWtn+7zknpxVlI)xyG5`KeOvGBjzQ;?NHM@(P_dap
z$hK%gh)yt=4ytJH{nLQX?n!bu%rY0OcZm{-=cm7dk0eHtW{1msfTrd%TFcH>3%Ltn
znr())EQIhQDTw*Ml-=1M)mcymg-%#3X7hQ%$SyWS!T|Q})m2K7T5wj24L4JoOyJnn
z{qP$hbb2?zoXk(;<${!dH0D%Up=@qe*D)u#Z1m%{QOgx=JNNrL=XgE_s-^DCQ!-oU
z5KJvACLXM1N9M(1WLx!25<4(jS5v`uHGiRQ_p-Xyp%(MSYQ4g?+KkExBYngLV;Rt&
zrwZNS!kF`+j5aF>`!i)P4}9BUsZxB+#D_hM!A)*mGGI`*m$6<wvVOd=uo3fx)AR!c
zY^Xmf|GW^g8SF_I?x`@Rz&vHDV%7i9zX%5Rs1n=8WfDTPYcK`(hU(RZ%(4^RakvL#
z8Z|5mz@XrjcK@)$?vs-%&E{PilN<jPb0l!f4YO!*>zroBnCj+`Wv^&ifk03X5=z#|
znzqMH`&wg4MtrFrGZ><IuGLz}lg%n?Q!^2DcILn3^*0>~>GC6cfD(nv@`R*``?p`v
zUYV!p(SbAP=dW&`-C(_1WM{W8Qm$9BM7C{3#rV^EX*3E`ikwbomeZN#`~q=ie)>x9
zJ7v3ZyS)<yUS57;-YP1IUw`?FSq+4l<?;jZ^lSsOw_jMV->}I9?Dre@`#ZfC=J~?u
zY;;cbsLT0uCPcFv5R_I{TEDXG8@Jn)eJ}Jn@$&LaFB{+f^xr7kO3X7g%$&ZwSdKtf
z)U9?~%)DLi)csecIg?^$UTnV6T4P@~Zg0PmqVW85A)cS<s^n#%)WWvjjS@MXIbF`=
zWu}z|rBLg}UJL#1*4w3$b7BgK5Udhg&d<b@DC>>w+PJUplvYV8lVj$*oY;4Z%v{n;
zo=;3=<F;NYWurF8;;yGLGUW(lk^psQzkg6`Gn_Mw_*Kh`h{+DKT6UY6_(Ig?mg&uy
zHR)ic*nyTuTmI=4CAd(oQTD=GDh(5C<#{59jk<pD`j?%j=NE$0@V07RIjBOf1ouJ0
z&CIMJ@xf!vXt(zcDLS9t1ewhPTSn+@idJ!kxPUW5qA0Berwp6ThlrwBZm=Sd&ZN`X
zu7?|&SqRQf*{*E28zC*koTyzXO^MOqxfqmbn%qL0gFRJiG1EOvjFx;j6I1rVb3lU<
zQU{uK$`0j(Y*p`+CtBHUBbIDtYQ5R~$%n9?PX&WkJY~_13I|j%vXS>~XPy`D&<xrP
zLD*Jf1n&FB;xI@`6Uo)MtyCli3W0n&(QohU_Z!^q$d|-Cy|58jcV&`Bkiyn2t7J+R
zNL1?%z9Ej8S$Bo7_J$O3V?diZ;u#e=%rsH@W||GN#jv^Awlz?ngVLz`YQbLwkjPd{
zCzE2Zc?6*Ie85pO(<AJq67pmg`_-V`9E3=uHu$){(^~mtLy`&kL<pJQH?$Xeo(wir
zgOH8ACv3X~O0`;%R;;5ZpjsWq`g>4PpMHl1B~O*<`Kuw*d!<}|MfR1P68XIF{QQNT
zNV;EH<`b9m6KQ$kx^A@FD{-3m@yi$Tbb8=r+m%`@UQ!9<`NWss{hsCN3(#2aSLXNG
z*&l&rx$yk-1?ftyh1XwxVLAW6)60e0ckZpQU4N$S-*`EHLHC{g_5ned^TK-lK%%>M
zVw~rYqNPV^qSe5*7uIz(<!A5IvJrS<T4uI&=lb@I<@7|FE=<v)DUXUiCHnIZeEq$J
zs`P3e=rW!8@yl0Uo_Fro58i(LGj+e&&88TQZ*zxC1Hw3+E<8QGu$)fznQk|_MnZCL
zqJ0Bbb=F!8y4xq(v{<#*I<<E8+IauCqP_EcdEz`}QnCg5{{G6vkRpe(3O#fr%tS+X
zeA9k{+Su#PZCzRKZ-gu4`HAQjB|$nhSlCM2SI3u49}!cAJkhGr{q{j|Z8l?&wea+h
z>vzxHzTIj2!oIES`)V}gl#!^ME?+p!iH~15*4J<Bb!VC;a@kq8jlC;*nwV0iMbCif
zcT%(GjF>IdzE$?(76r~INx}S!v{fIP8~m51*=X0vpfQ{6dKKE-id>zxDQX)>;K&CM
z?2zoO(CTIa&ULj6lGY9qK&?AqT=Q0;=T7U;w(ZCX>1tLDAy~GAih1hC5ZtS&9*_o#
zJMfw>(nvoP5zgl)*4v#B?K|X@=*<LG{Fce+@R7P>Kyqkh4C$9;;p2KGgk-rQd$9r;
zb212GK117f$J=2H^0KZW=gG1hdgpXHQSU<8Z}itckxwg{zjC{E?kb?~md&!9j-f6(
z%pg5g?GVlkLp!7AMmpQR6Wq33HMqWGa-nQ;XK~YSNKkrRE#gy@<$MBtVk>sf7k5Hf
zudnoSr?&w^2D9C-o7sTWO506GVc)Z|?Hl*4aNo_&f11p-RhuGuC#H=ga@_Y;jVLoR
zrU$Dlh@lbM8~b`U##QU&oc?N7_dn%O*s;MJDR>(@A^AkiH`q7U+Xs6A!k=K8*gjs7
zx{=bEWj>)qZXbVPziynSOcNo73-dIw)s@mJIWL?p&(wwo>Qqt|o-b!kQ|9CCo!f2Y
zetkvT&eM5fUmJaY=i}|qEYHt$t?cU^Y6Z-sYMCd(^oOIu)ZWo@CyG#;Qfnhm0n4y&
zwxN{5JTE*yoecUYW;YyTWH~Rirj+eUmc)Bmts2JWr&G?v`H7$Y@Q-YJp%i0eq?D}P
zOcNm|YTtSNwsT!q!gNBSP`AeE6iMmCJWb4VqLr1m%{mrU6U)rYi4YU}X8M4--?`uK
zb{(fX^T&-CA}N^pV9Yb~qLgiA-&Ud;GeuP?b?16}C;M#XyjU)9o=gw1*Fx>p!||+&
znwImyjSndqtdxur*V~44Opu7O7Yk+zrBin2mJcD8x>uDdDVlH6=a+W1fSkIcZ8w%k
zxwGBh+1IOiXKi=mp<fGncX400&OijW;b;WcVI=h|4tL<ypzQ97iW<z3Q}phpnyOlj
zotB6DWiTbFH4uG3=(cbAOa=*5g@8i}!*8ipi)icy?bUrVv$eXj?i-8eU97kJp|EYv
zifgqp<?Msn&hB1ark1zCW$h!--8S41quDJ>B*tSHLGZLekw(awZQqW#$RL&x9Yy%<
zOC4M?A*quOx`j%ObhwexAY$*)N;P;>Z5{{?QT5=eVsj)K=5xAlL<!`}GiWn=_FE$?
z!t?xv>t^$fL`Bz|A6D*h)@;63#8mbnXQRKm2sI|N4_C|}Q)=_h&Mf;ahojSKgWwJZ
z?kl9tNEls&t(sn9I-NP4Pqf}htI#UoIv+4&Orw5RTI=lcu-a**QcFWp=JNakn8U+q
zd8W5QsW)>QP$9<IlNhbzt<?g?QcC9U6#_A<uP;ojN}-_?{X29Dzy9gJaGqykOlaSD
zetuz@A|F@fmv7&=zP>V_o_KrzFp*8UBdrnB%yivQ3GCN*(9V`48t>e)S^h>b>*ICZ
z$l0I_dg1o^Gl;2Gr#vyIfKFoRLzvmuJPEzQdi#afUuQy^>@<%PL19~WPR}QvpMRuD
zqHLSl6L;n5@`Mt&Z^i7m>1=hS)!k3|3pv{0Tqlc-jCo?;3P>fVMCm(SjbX5D7O9y+
z;QaCjF3&G!k~huApr@3(Sr5dSr>{TIsYahmCqkY{1WxD6F+aH8KG0tHc>Bive&^-o
z2cDi@=({kTpNPvs34!7{6sfr#GmWxuG!d5NWD$@vB#A~6C_B0}rfDWcp_GlyeX3G0
z&locmY#ycA*txsw>CBX$$n(X9*Pba5c?7lu1j#4ELnWZW?6_@)xf4Sq$3zNPMb{e^
zBl`U&lwzdigvdneg|=^8?{~IsXI|bppD!%aY|u<FSS6-N3g$lBMOcgFWDmlvd091J
zV+xVHET{^#L-j797t^{N^9c0lB!UN-#NY=<S4wj_W^cy<FsxF1C$$*UT8#2pcSOzT
zv1}hm-DqV$;6<B-bWWEiW4r*#hcHnB)pZ)f{4wX*;G`)~%1(AYN%LXw$Z8h1;Rp`-
zOfeheAZ8?&bEei}R?9kUIv%e9yWyIXuAQxH9>v)xdwo3Xp4-rSXI)p$=d)9)A&Ob4
z93fZ=LxJ6@2|(lNOhwUBXdk~4v~m0PM@~OZ<a}YCFLo%D&bRkBwr0OONpRY%7+jh&
z66|2z_nlfJrEW-Y2i#;<8|&?zkB=)M2F~-sX)&KG`(93m)lRFJ)wNrW&{sabU5))%
zcKW`0lz<frwe03-+nTL$RWCHmeIQS+`ggkbn70t{!^{Z)!E{MBU+W+p#aLXu8@p11
z304D;+9xS!`8((ozWtZ~M7>;?=7sVxvHtpJraW5`l4LYyO4%seo$T$GN}_dTd;8TE
zWxbJO<bHjnOJq7dbN=!p^KyY+dHwZge*XDSoK6>}Y)aU{ubt<a6tmH}T4lRlt<}xZ
zPUzU!?l+bv<Mq~Bc{-mFI_vF<rW4conZB>Q|MDmEOc!iG9ZhUhyz<)i%|$KIRnCnl
zktT_h7nXE#^>yVOY(U)=ZYlBc-dOK9diyg;l=<>(4`i`wQ2J^jF%_0P5x@Mz)A^Yg
zGZL~TP6aG4Ax3(g*w#DJJ4+CByYaqS>n-z%IA2Vf9|KdK5ox}e-MM~Ph{^f##Fwu>
zF-@}zwam<5Kt%+5&_3=Pig12<VVM>P=AC1IxkpS=is2!)F~&rUGeN8sE@h=P8%Uek
zST*1Kl&wosX#H#(bg#7P*~!{Xb^IIlk)%Xk7J5tO1CuiI`I*)`bGDr3<@`(tiP|db
z^^IQMp*AaETC=LUce6K`&mc;#cciRH$`EEZuXCoBnlD%DZkKWx^%=D_#cb8T%RSIX
zmD+uvZPpIPxS=^=pl+!1|7{O!Z1*c~KmU<XH>4L%r^PDcQV)vfUJ6;vqpNqT5L<1;
z6g+9nasbTd3u@b-g@a^Jdo?q)WJHW8!4uiik;z<pHLAC7+Gg>wAkiw%#|^Q~P+aL9
zh9htqDq-s&luobR6T(bfx$m1L5k~v(A&7l$vF=0F{H#Pl7SNapm!FJ5x7|%Mw7t{r
z-w4Z0=}OEK*N<=f{Ps!<!XKWW2r;wY47Sblaxj!^mK|Io6CgyU^AmYGv8^k=zP|G3
zKmVCO{Qmb$K9|X2GqtVTa6ILiK=iKB!h_cJ%6fe_?towrpxE;*BIGCp7n~82cl!gH
zd^jYFc=ymMwb?!q<3ygHnbL&DWXk_$_9-bZ?AzTtf6wbC8e>o}^YdW~^4AUD|Nn<V
z1eRn<-|OvzyI7X&ay~OZ{Y0Ifk^Kg@U#N8_Pcv!0Fwcwo$!{$ALJqKQJ6knY$7!Bf
zn+;v61n%2Tspf?ggWbe^-+2A@&EW`xaPI5Mz8k)JTF%UILA$WsS3a(9Twi}iqVo0W
z3+rBZ{pC-b)+=?p!+xXX$o2M)Uf+4TT&#-Nc3SIf_mwWf<>_J`NdzwEXC#}gL<$p=
zm~(1fH-?$nbDo)|6Zh)}Wxvv@v8#GDo8pubbRqO^##+tT1Suu%_bWYQvjN#QPN&F^
zU%oQuOp6Jb&s0gQ>&pG(-6(Y${GpymDY7h!GmBPYw8gl&_Tw@yJS}Hrno$xV%#?NI
z`tiZ_e)F~;PNy^HFF&$ezW4)wXTQIjd0=ya6wwU#jHYBxj~EV2h-he3=_IYRy3^~1
zbmLSSZF2@p-hW?LTG_}sGS8!Gd%RQ|wXViW3Bu`gVp>j!23l!strFA1a`}PUJHNcW
z^2dMvzgVwVo-b#<zWl&E8B4C4%FZaIwIFSG2TLDFR#VfK)tpjAT+oHtLLX{pwE!Ut
zcyhN`$MQSLOw`0Pud6`@20c_y>~l4_D(n4bGp>3kN0{cBd79~pTOhd6qbtJ|W3Axy
zV*fzqN*sjscJ*Rn|KtZ{c)NPo)V}Y=W)aM&u2+Y{#C&xej<Q4l``u))`4sC^06$dB
zz;SdXBM!9G>NyL+MQ+VG_3CcJ?%3-Jh_2l<4b5=ze12wr`f3>tAAe*ybwn#7o$Gz0
z^_fkTuYdS=JpJ$^m&-HSc3yw}3+wx@))#slZAh7US-9`j?Zn6Bvl&keo!|ZR6F>g+
z2iDsM@9(cnDN>q(DCf(?(1K{T<hAUyVl#pq5{Y&sGWOP;mTQV+Vz9_zF?h5(wb^t@
z<|1I<U)_hTvTr*vb}nINo=rCsv{SU&d`pshC@H$FW?iEO3ocV4&=55SX8L=?ai)2O
zoUJlR#+5CtGE3y;cmIw#oG9;qLVvMoc$`jDHHEJ9Lb<=wCgID=6PK5-T&5Q^TC!Nb
zeeia@QdMZviRST;r+G3sDP}b8PIJq2BB%Mp>>7g0eB%7{m3%t$_VEp&^TQ86F)s`A
z(=(?Kx!<nr`<;Egni%DLHo1JQlyygPLSupebBJbV9n9u=nP+mEX+eq8!m>Pj`kR@8
z0q$}dn3l6uDqHqNWLeKa63h9CJe{fQm5-nQ#J=7L!CL56H}YxWhabLjzg;Qo8@;S(
zwAqNHi5xPmEAQ={`&uXnX>Qck*viVb-;Ksv8dIJ*UrwZCVK47*zgo6KnyIZaP3FBa
zpM;E28hh&&J=i*R-Pm+v_r$x?c_D?1E5#?K`GjJa^nLw6e5j>9{H(3n{Gd0YLexep
zmW|N35@RBziPPl^tyS`rNcrjT92&3N#|QWK%@GT6__AT4MU}0t>}_@C)=WrBkBQr6
zEqsa-olMLVIXtR;Nr~D#HZpYZd4mm_-ILzExKd6gsOsk0T6-brL{vQxqV6VpR`ms_
z)hVqe2E(bK#+NrUzG$=|Vhm_E4t=RRIlBs1Efc^7)W+~q!RK$8R;)VQ-JCAx%)Z{4
z=h<N+S2as=-HRQpCf2H-pdf*0U0L+1HAtYe`d4f))qdnYk4k^UvWioMB?_$$N8EvN
z8=v2j_Zm5yJ$FZ(qHMF15KW72*X~ZeDf8*f{N*R4-z~E$B_u~=c|wUyQ6Vk7eEA*o
ze5SRP^>&3;ZJsdAOc^B2NS<l;mF@Os*nBr~i1f;jFJCx4eKo6Wgzfr`b-mIR=H<kc
zve|p6S)6OLg0Oh{V|DZZRk}7pH0%77EFi1xmZza2<ax$&xCOM)OFPI7(>zgIU|$Qh
zZQR#)wz8rGqvwZ?1YyEDhv~G?i?G*jfkx5hfXR&_b57<B_dokk$n#8?7edH9pPpGx
zC;sxM|IGXKO8E8*=dV8zQm363p3gt>^zwzOo%fGdUcdd!{jR7aR3_r`g{MFK2euyg
z_VXX9b){=VVrGgn6(RIW6vG7P>4^m)Oo^wLpO~M100Q^-H$Hy-Bklf*NM(+2o@RO}
zJe|)xKRx?mv<ZEwogQXRPhWU>`NH+>l^zS*dWF7|r-^*|LQI)$+YmbY{ezG9D?BCU
z`GqJm!if;O!9=yN*Gg5Bs^?@>NeLJB?Z(^N&jbl9=dWl!A&U^EGxOKqBiBs3eM5BP
zG%qOd!^<<j|M4r=w}p?_U(sORJnObo_rkvI?5(k@AaSOLnR&X9P`34oDE$8S|G=^=
z?$o)mm7T3N-qxKTe|+IGKXZG3qip6)^mINQ##~cMBs#aZKl65drAz1e<p+YkSk<9s
zrq_BwWu}q2JpBQoQSKkSzx{$pB*haUDy?jk?S?Au+^9q>zdDF;xqNZTol^D>^Je<=
zDtiSMM#>ef(>F8TQ}Ia8TKM>QW$q`=rwjkj|Ly-x`G<dRv_S=N)oO}~;4`X_16nN!
zOa!JF=snQ1lVhZbvhBl&*cctHtHsK;Y8{O8;9Ki13=>m=Yo=w(<JD5YyPDpcAJV}3
z!RufCg`fZHe<o?+bTS5ws8aTwJX}1RWmFHCNY&K#kD#1RDLYf1C}p=9jOU1`<;Sp<
z!jxyHn1;i0xvQBb8k5DoUvx{9!Rz0DGx#5l0pbS)SO`!|$rzl2f!sUN&5%VA@?`g3
zVxSbW#z?_}=tW(F*VQ&jI<0QdO+Z+q=RCapXq_QS+deq|@Dr=fl($#je))xU{owNa
zVo<Ck!t)PQh0VL#oEAcwSf(3o&XnEkJX#I3x)YZRxhpw#n}y_woC0}X$jP$UfBxlH
zZnqnAHbO=R%af?p1h5iph8H{1P3>RT&1mW!QVt$UW$@aZlFhK}$TgOj8k96IrdTg^
zXI*c0t;588IoUx^G#!GR*tn50Ov{-VBkT2@wd~B3ArVdf4uj_D`S(2k@M2upwo}%Q
zh~@ge|LLFU>no++Dcieyzidp)Os4ZRpLqI*XC%+;t@8SMqnxf>E>D0Grzf5#V_ZaS
zY;C8k1yOUf)R2&r&=BY$8lyu3F&maG(vY_D`twRv<&;m%c_9#3J(HnS3&i;P!;j3D
zFVGvyGV$^4gAlsW=xXA}FTZCwJ(1@V+KO?|OX1t=Ur6hX^YibJmkFVpO~k&k-d9TN
z%*z?hrh+veq{O=22@<$mzChi%{_@|@I5V9tr2G>ro$aF&L*TmJ`LF-{f7<*c2>1J)
zR;zayGw&Z)-d|tYix~@^pS~FWe|kbhxTgv2J3svRdrqe_+rC0f<coO*P0NYP^9#$A
z=|#D{{}<M_BjGC|f$e^$hQK@}KHk6amtX!w#f<_(=CVxmURc)~@9)1dFE5-gKOizu
z%g%j!=i~Z~=gSwK&SydlZ1=#u?oLZJ?6_4E@vZlFw$eR$47TeVU-$1RsX1pNC!jHT
zY^Rf;Xx-HG)A@qNOqWOtnGiBMpD5eiD8b$0Sj`gI@OPmz<wOeZv4xOgqBP5mu;Fx~
z^=gAyl-73cw|D3U!*o1*L0s7!3{r_95V4t93!57e?r4YF3)_9=^^gC|FMs-He*NP=
z^TU@Xo}SO<!;eRdqK9XG^;9ndar=QQk0AKLLvLj_nyh-jdv9oUx7`q!Vk9J^gi5qw
zyhYA}sC72IsjIuA3!c6|6wDKAqPOan(bc*FEV8w=hU<rI0NRX|l;_D=M}g?w)v$0M
z<bsjuq1p+xn6f`c+P0##q9Kx&3)}mz#ITasKr>AvwRb)~-e>}+^H)+nGy9Nx*i9f9
z%P0mHSyqU%HHj9q!2NpVu5Sk+U~~q1t<-9H2{8uZG}GKZf4<B%J95^~lp<1B>iyk)
z-6~BhT^=crwKmt0BqV2~C65aZEMbObW_&gu-N-R>IbZ1AuCJEL-tM$E5tk?Oaz=q=
zF?3*xo)saMKU1zh|Ia@ZUVi!qE+=7qyK}$3L5!SoWcl)p?i;Dy5pBffLOLyMWo2Ej
zX2KORU;glqgqJ^1whwN9{tNH#@6_$e{q0wN_}xEozFhdrzx+4qzA}XdQOUDmfH*^-
z)XM$+mGyeH@S9q=zW$l|@n=r+L`(~5I#c(JkJc#r#J=CT-#^e+IbEJO&j}Ib<N7PT
ztR@O+)z~-dm3aEq1T7+bynmyvMhirg+w~hEC7v!%v|icwJ2@q$c_OPYpHAjouZn0O
z%EZ&<EBm%`y<MrbF{KW%a4S1`%BTe1t{b$Ks<7^bX_|PtT&S({>#x6}-`<!{XWm}l
zczb=N_r|iEIbEJmZEV{Ox>*Vs58guDUq7z=%b)&*)A@xjUw%iPF|VfW&i(xx_m6Mp
z%BY>?{DcJMzF*l_3od&3@*|ouaeAWk!0t|y+g5mg`<34R!ZMxFydcu(ZKJGrwh&lz
zrUTpUM%iv|v0-UhDL%8DLFvBHez>_He<iX>DNw43$?9<Sj0vT)whyjvzo5S;%X#K>
zdLpG0jf6I(bzA5boHP;gOvL#0;?~){8@1ECu#9b^szk9Mrq(m}^#<ChWu+C%5=b#Z
zNO(@26))O7qZuFAiaX?XnwtGZZH=`w_GZqV5(1@K=5d(vLq(0}L<FI4)q|~)WBLz%
z`Zy5wtvG|l3<m@1p-^VfGpg7e5{Qzg*$3+m#~?^>1$yruz!ZI>{`?XwU!*$?6nZdA
z@95uKVjyUwcQfxJ7*;M4$SG5#TH(~y8CrurYxn=*8Dy}VcAKAQ3e59``^OuJPQAYw
zwLLHV^wU3(vd!4y^h67Z?e@;LTb%9WQLS~q(_FKWL!`?@=y;MHNH0X`kl~{=RP&(}
z4`TjhIoKXzCVs-SWKNee8X|S~2-IR$-V!Im1VSK6GNB&OeT~ND!rnXEUWg$xpU<40
zURduR^xF+onc{>7wGc0=+g}9Ban?dtSk6z>y4!5Bm?k2HfJ|QqD1Eq5{?9xVe)-FP
z<H|<6zY}P@JU#RL^n2>OQ`QecuOv=9{qT=G|HJ=+?l->u`JY+u?<h*8FbllAd|{eS
zW+GSbZ12Btef={rDuvF^@9#+8SaKi>X0N>O%o0he+~40pgq{r7E9DMN+1ER1HkZZH
zl(OIW<!v=T@mA4tr>s||n26Jvy06Armw-$se)`=LYi+~;b5df8yuWp7FDz5$$LSxC
zWL)a9ey}WO=I1ZmZyv6K@zYx~j6VcRZF_n7J-7Rn`})qUT)A%vFxqQ25)X4|ysl=t
zXUS%E08aDF<#OTI*LPG_rg`Fb|L_m?eVh$+`*@`kt455~5&1v<<9{Sg#yc;qFwK?Q
z$2S5uHmS6`St>j&iS+zCsz%oP&b+uMROqC9;(UHVrVA}3rqhL<CbU;l51e~oyME*L
z@fVs#qYUSn=lO}gH?Hrmj>T`L@Wl+*f(q$8kwT=`#(lf<?b{pI>nmD2|8D(0Ht_xJ
z8wS(*p(lZyXV!J)Prv*N>#ZWa@b&ANpT7LW`SJt#^up=1kaJ|+#F!e(g?xJQaP=!%
zDs4A)s)#9%X^r)EXYU<##){2i&1*+{rBqML!@8I$i0e>>;h?ejYu%}}JDxlt0eYU$
zsWY9POgj@HN@Kr%v-w;Uz-$TfJkfTy=xD{@Bj>hn`_7>(_D*IT2!Uu8BWhL*)0EAo
zV_!)jqRCiMwd^jyvkllHB5lJT7VWjQzqTC_u}+~I+bM-aP9AQhjn<24m{KBzNXTXZ
zqTNJboll-vbb_nkRhwCWc!!jt=|NQ0_@AOgRhuV%`H}to73yl%I7EU96hZfnrp7c)
zgyjrGVoszSJ+P>d`c4%4ZhPwvNj0dIKmYNc+17V1mlwixw%J)X0a#hB5S(+OSA*^L
zZKbJ_=7lDalo!4{JrlfuiFvjZw_u5iU5xbv!d@DCzu929?BtMW-OT)YGcoTxWv1ke
zq<w`XO!3JDsg>o+GnXI!V3h3MxW4|v$G#J~vXzbP`bT4amA?ag^!n=`IqAld0`ujG
zU4?ersC}n5OE-$aDuY%EwHAJUeFfdQoM-mi8`2|vn%Kpvz0e9Go%Q2S)N<$Lr+?4+
z%MbkW%b%$0m9@e4@s4(7O_TYT`;ZMLVmc8{YFP#Q746F9`GrefxbHTV@B7B~{*7(7
zinZ=HLQGuOJG~1b&MwJUmPycPaiX-!93%PZg=L<YB%xZktslI<{mT8e^78dZlt^i2
zvn<6Zy0`Vp<$PueX6x|v%U3SX&kzHuk?r=zcKwC*{TFoKn5T);JVWo?x1Epc4c<OD
zFB7Nb!sY2Jm#4EmjOxy$>z(_y@o`)ED0kvyk&$_t4Q}W==kvtr@)JM)@H<LrZ0j32
zROUHQOF%T3HG(9H8kIF55F#<p<oQBeo+x4lTZBlfk)Ar~q|7POmzmq`gMBY%{^wlj
z7y?;3A>l#e2F<iCP;acI+E80NPp30kgqVcW^8G+;IiIQP%p-4Cn5Q#2pPA3kh-9c6
zQ8%mnnsqE8%%G;K-q$O+Y}Ns~OKdk*xzl9>q=f0iEQu`yuks}#^NG?T>-A=kP*WS)
zdd397LXCn%hg1w^2|?`HP_<3OE!($R=m@8$uV~wNrBj;n6eH@crvwW=YNa~ks5=BI
z=Ho@Ts@wCI%{sj!&5ivIOY4P{GhNNsD(z-rzn97>Kat#=PKG7J&=HI{%HJg69hvB^
zNg*YJ>zWOnkw9QXPbx9nfuY?xi|Sre!XwDYM1bNVyW)qR%_X{b3Lyli!YWffA<4)G
zm)x1t!Y{vm<Kup%w$AeMBftOm|0_$3Xg0B}(Lggrcgp>ZZMzXrety64pa1w5Bxatc
zObnelFD&OL=F^j9RCI&L2!Z8vvFl!TvPAoA-R?mUgCJ|IBvylpnCJ=SDG^(xx*KbX
z5y_FQ?Myi`pPqf5RM@v0N@tm8LI@ORut_uv^(h34ZE0p}-`h?v0nyHsjfJSC&|?C*
z)4KBV!<Ya0heDE>ua{@0D17<pACNS$?;H2E5z_+QC{$&AV}FHTUw`G#|MK5>dA{&`
zI#aKA%JoJKnfc3)mb*}QG{l1$vScA%o=og=JDFceyF=T!zrPVBa5_JmXyx?8JTHV+
z_(dBZzudXLzZ3JHdH(4i%&l`e5l@BJePz805;IjItqIq?lb16;{_Y=%G4t)0KT_5=
zLR!pJ%B^TaG|cZ_uk=>A-Rw{~T^6gf>TH9wlsKPHfKvB2_UkutUkTI1yj%#;d`yZ8
zbvhGUA(aa9LGO*rG;zv_wSlAwtsAp+@^WH%{>qdRQ@|Cg!rS{BaoI`Jozr<CPqz4<
zFBjugm%?5f+petF8!;zdp1+VO^zDvv24L?Eq%&o4hfufs2*Q4UXS;r|H#=<ea$&EY
z(W{NQn~zNBHe{|ri1JKI=J$TTR}1=yfq7YoX(5D!^g`cP3sZS|CVzRR?1k(7<5S3u
zy=<m%K`}Q^%a%y!?Z#!k(8Zpkw%v#!pw+~4Z*Q;M?>DaRuV~$w)|I_h7aPI8t-QT|
zW4+(_`t=7+UrtQ1@)RxZGo^`Im3yh2W8mBCuY~w3Q4@KZJ<q|jBtkg;j&fuSv}WdS
zeJ}KVw+KsD$CukHb-$yM2>FD_*!&E4u1#qjpPAvtc7+_Rn(nR9FsqgvCsd3PR9mww
zj+D7PPwe|nNRaYumKyy=(1q!IrfH{Fiwo^;5pGJ^=K7OpUQOS<)CbgNbXa@yZpeJY
z5)-Q8Mu9#@iq<u!VEGH(hvzXyg9Kx=v^do5L|hf#O|5G%W20`}gmGaePcvn^qqm=_
zNwCcN#<y>OA>FTRd*PfD%jKE0SOVMp^vr2KaW9R({P`E^wsP;vx9c6IPMQ+`{`ddH
z4=+F1=Xa{CN}y{XPbaJLLm;KaM5yx<+AV-*ua&y)+~0qtmUlv!`FH>H|1frt=Q?j?
zBPc9>Am(Xen)6}#9dahBWrRqcJy54mi?j8_Xv4McmK%S&(nR?9^WEn-RwSnoS=T$)
z>(6uo&-@+Wqn}<b{P=V--+%0+OQ)F_8n^dXL>r0Dc73C7H=a%>{<nYkPwb^q*Fua!
z*;nFu=7+C8L7FYZyj9Bj&fA~=6aVS|`~Tte<@fyfhyTEruRro}{ot2B{TH+<dCFvg
zd5%0?W?H%P@$=8-xwK3?ef@#DUg`UdU;o?xNtmCAr!Qvdt&lHYSmv4K<wvH|3)A_T
zbz5nDV>%_CzkdDy(e@u*mSty}CiZN3<Q1me+&vT`fE3+XRqWzoQ#3i_&*i`1qzCLy
zc6Vi^sDudN(!{l^j@pugeIo!QC*)eFJHf&r;0UwpXYccU@B2KALq%{i`~88&YBsBu
z=mrkMp1NC8e5(PK1zJjs1TQVhw4@Xm#}itD)shfL;^JAnB>EYl8+0A;q9F;5Xs&2=
zjrJ4n{8E;l#t_i>ESSd=GEC+F&*4mr0xMzNwRD>e&E^Kbj2K(9BA2pYE)7Uq*VMTb
z_{MR;xid1CHf2WPgT%HSs%~*^B8(^eT!4edSme*ybuuQw?&*;a-~Th4t2>&0jjc<2
z*oPUTC9AG4OXwJ}wlL?dPMCUA9wI4R=EI<L!An$C+1u1rds(I>l(GC<ZJ@}ING$Wj
zFbp7I)vqDLavrhWhQ8}bm1XgPAWApHFNHc;e7vUIh`q=ASgcvLX0^ScUvKIAzF<Rq
zDClq!St?|u5lNt|WnN}X-GHc2bz3?JDWH|OxQPGwm+og7Pl^5{{JfOnUv%sqKM|&p
z^XaLSigdXhYhzekqN>XhJ>|>y_5x`HE=%E`(bW*bg-)vZg@T!)xM0vPh-RS?QeJ)r
zC9)brq$>A$eDoM&QB_R}C0!=-QhdKs{DK1c<G-7{ER0Jt-23ngW<!axM+za7A(|AI
z)^#MS!u;3TkVGuIC2jbbr6gpz=}M&_{Vb@t5L7SCZA&cp^GnP#yTo!nU|wBw)oc-}
zV%=^qMiZc1KyoB4BYqwU)2ZxRl%?Ok2BEp{R{U+>BaOkikrWDsMdVy|OzFa$@?{zj
z(~MFEZ4`bUNipD-xoBfkVwui7eEf;gE%d7^Qc}FSyCKEQVK`$dOW*dKrWx5ZwDr1#
zOqLTd2DG-cn_En~Cc262&%ZR`L6+yGUs#q2A)s$7;(TU0?0(tN7+WKo1|?y$>Pvxm
z-2dedh1x9m=$PEV@$-(1X0^J-Xn9#;6}_BaCWd9AZd>YZ#j0ycZ{bqVKx@v!K-C$#
z)fOQ&%hcjdkGSbr`V4--+Lmp<!DPWSoeMuDg<nb?^KfD~9}B|QHmJI#GNxbyLc}3y
znx3j(kyFCY11bkZhVi&VWWh3=cz%9Q%uqKQuCMO7y1rsQx14s59ABO>A~Q`R=izj5
zt|k_D;(R=Db8|~>Z_rgw@H5BZfXtD;Z?Vmm6v`Ma%mLGY6`H0tl$wi#P@`N$_KxT$
zvWN(4QNp6MyOf{|DHn3*c|37CK4JGYN>_A^rKv12mwtp@ZHcm=m7`k-$32yfXe=%(
zW>-9}Qc7%XF#U=ucf`uD97}xAUKWrs71mYG;Fk`SJvvQnsx{l|o|qEDX+cXvBWvbq
zWSsVdRLVC?%(S|rvIe6y@d5=T1(GlrZ4fyX?vE(6N!6@MfZ$I)!#I=uk>g>{_uv1H
zVL0>p^&8&Yy`r<4$<0iMOJPcBq%!1Ces(Gcj8Qaog|2E68s{A;fwqQ9wZ!O)JA9fi
zt#E^z2A1;<j6rrAvX@xhQZ-xFn;Yu3M`=|Y7NIbcem)Y!&-e>j##15$7VmIj!7U^E
z{R7j>GxKnu>pEoHVpS2vhnT5^!i9jnH0qO3C{vKjF=aAEsDmhCnVc^=(VvO4QkG(z
z0J1<=mm96F2+?Djwxpm!M5#z+H1jx8_iMDVTyjqkBi@HUj>s;bQmz864Ay9zbEH_7
z@<Q@U#~`MHThYcME=c&FQT8%l+TS@}Mq7VWTIQGu(V^j&_VzCtTExZo07@EE)nRjD
znNMI1O|zoEZis%w`5B)}A}XasbPn(G1>KTSRfVm4w5jOYmJkE`X`!iG+O{LQg=si3
zPbFwl+rsvW@n@d~WEqJ$Vl-r1-mX)au<eSgx8JbZY^j=tkcu0<?RrdA{(e<$i|IS6
zs=w@B3W0T+9l{nU-*lchKYk`I=Q2&H8qf{GRJ69HOnwa^MEdoX5F$QZe$J*w34_WF
zR)AQRzx<(aJkP9GJymy&KMl;&MCOIIY0K<0l^JJ_4v_^Rmn9-DkhP}mRs?DI{P_`e
zoO$&oBaOz|hQof(vUpbghKS_6f543=tf^`HElvHZoaEl&#v>^&1zT(?TnNNGW9u4Y
zYGl<=NsW~i-PNnhmt7%3WInyHoL?X=%=3&$j;=Nr@@0cknf?AkIZTC5A{BIP%``83
z{_qn*8m_Nz5IHa%o>=!Anr4L%4P8@~yVf+9FW*8I3vOOu3CwCigowLXpHlST7t-R$
z!DBKsU2$C#7ThQ=3?c}A;&^;!7>-nRLuD*eWw^e&rde$YXWv#}n~MQW5srKOd`8KJ
zdUJ<bt=a7!cz%A*^L|IyKC@Y`>5RryZF$J0Sc*+>5xhvY>yA}xQKZsE8DEG96oC{Y
zXD?WmnZ-wf&<I_FHi!tz<RCcwJd#schBw{1L_uTGcABQ6YF8Mo`M1Lgp(?^@p;C&Q
z^%|WbX&xC5FPQZPVKqLK-h#0f-ImLTRDyI7^N=sMYFW}R#qcFdVJ`%<Q5aQjVTaQZ
zHxGpK370d~y~0|Ju|3ysI=ZW?;*by^NaPqm+A<?9KFRXd5Cu|2qBR#CqbTSM@0gb}
zep#4ASyFea6^jqZ5PoT1$CxiT_Y%u4ny*qa6j~8QLTW{dB~Dx0s#wy<1n;Qo%YwAB
zETP~_gw$nUluK!AnU?}`3l298=&Io3m9hof9Xx;P3I~3U(w7m+rLg(4V;L#_i12{2
z7?2uuxp5Z`pSaM7FG;0e5>yGRR3RV#i8=D;w*_MiC`%S4ft7_MnHr?V9rrk=v0cl{
zdBD#D#wr#!(>8VKKtzX>in^(oy`xY!J;U+L`S{4RlxNDSX{jp1*-gZhsB24IS%RAx
zPDen{^lLH_pAsPnntnqCEN)_5wXE;Ip{hGl_B55DlBT@1YC#e(k25KGL`<Y<EI>|E
zzGvstj!z#yvFUq^5(N)$*BEU|V$1euYYMR-7QDGA*<T9WUX*g2tx>7$dc(31=6{<a
z`wJWjRl6zNc`51F8dFthEr7!P{-dNjUMPMNX<BUE;d5s6nT$cz9rH5t;}5^*^TS8J
zc=JWcAxCgQkY&c7_hfq_>Wbjb2xZFxF9)R5=yr{&da~^}?w;7aJfbDszkY+RR?JbN
zRK@vx=6v3hQl@E2CnBNAriECNC&6c=)#$258AWR>^O!J1Y+s=%1jV+tOye2v1y_BK
zF%q{7r1QsLSWc$ZCEX+qlPifVMI`zGksVrPNG-t)xbbwc%*{n9NM=bOBQ#w@f2Gk{
zaylFsPA`ZQ=<15i_B+;hZ$U|p5AQe(W082rjGRW=?G?J&pp~XsZ<*ahmIm8YRO=15
zmBA`OoCjD=h;dJBRzSJc1vlcSBkF=)F0@odT4u&^Kp9Pv#a_mR8vFB;!d4oeGEx~#
z-J|P{94<9{Auc<j$dY4Gk}A-)K`TKP5fzL0<xf9-zhamIO#@xKV!S=#+)@}c+gEIE
z?uj9k`Fzd=&mCM@BVVYbA(z5(v1Jv+LqS1g!ZI@+%TuVx+g-75t*&V|H?ZEKb&GNd
zWhKh~Y`e?IOO%zLy#riITT&HFvo<VKD!SSw;`amYG%yb%hnE9QQxacNTgGKVDv7`N
zm)#Pmn+ByTW?vwSS`|-#L=#0qnUX=zvh0Cy#R$g*{^x>2R)(rJgd8r6uFFlAGUl15
ziOMK)bdWU4Dwg1i$xTYczv(Ib{JSn#?!tCjoGaWVp)ZQmGBy$y{B-%z6l_L$UX*_>
zhw|qCGcF^>cmc$jU(!|ilIZ*SKSi9-IU=iytSe-@p$neiB`$~Z7E||_x+ZdDUL46U
zq~MWS6ScwG9;q)dLO&HjQ&W*r<h*|@yNZ<1y7(r1j4a-fguv7-s$CJ0xFs^jfJ>HT
zSy<*s(=;@uLm&};#Kj3OYLu#qLRCttssI6%u80)Al~D$f1btm&Wx%TP7Eo=j3j53Z
zLO=j=^o(<P!|nSOLK|{YW&Z>?U&zccMP&4NH*-GzSLi9c>>k;!8rs@ojHKxlx@rg^
zk`m0a!0Ebni_WSLCsC}fZ;2?*<3I`%H(kT=d_qt*?c>X5wp!5LY{=R$jUy~0$rZu9
z72y4m1<FmY>QJ`9$HeJ$=H+-mAn@~%<)E-qlb6hHdIXtRZ?*_wa3W!CRc4sF)c&iw
zM<%GMmT4F{JU#OI%~yz=%1z6ae7g4w)97jH6}6=}4QxpQ9p8V#wk=W{rlX+0x<l4I
zZEf(QD+{ok@Jne;t7^nri)boLwWf0(Z7fong11;E@=)wfjWul6eSxH<Tv9JdqJ*HT
zTkQQ842K=gJN$WH9?;3*S_i5wH&<l|8K(1*95d~DgEBQH1PEvRd?YRdb=~5E!h4U4
zzRXiEkTpMNw6$nkk$`cU@zK+E6|40MEiC(&Bl&`YP;o|kfnIH?x{gI_j8;si#QC)2
zetk_{7o<bZTr$}MGDo8O^GND6o=LJrDnX3Cq}FtxYHF&gqF=AbLb6N))A3jsMRkL?
z+<r@XMqY03;A7@AO(>~ptClP@AvzZ4S+82!x?%|dvAIKE-I1js3su-ms(|F8xQuBc
zmz%4)Joxfu6q0_Sej+q}8HwYL;pLIt!xQ5)G5dtJ8J#oZG_iz09bt9>5liYw*?4g~
z=W-ZbIszFAPCI6dDKZxyih1p4#*e=+jK;;WTCMTnVww9HPmw*!=z{Fd8KtW-gGZGM
z!8{{GLjGHl+GPijQb|-vEg=<Gg%lAX3m;3W;*-{*5LvY<BQ~YVcP!<6NlEEn{x^MC
zO7YA5+<Y%qJCTd9s4|6s?3R&6mxi!16}qaZ^2PnV+Jdf;Mq>!LG2)lRFwaz#C944~
zHEulNr=g%Lq#+4`R{>#4BiLA7?2NiDJECNeHlmvbp=xr7C|g4q2r=V)W^u)U)>dUf
zZ|jE2NTf6<sd3(u$y9Yy-n6Q!AZENrq`>pz6G{oXenpCrX-NbEE2Eg^0R^;8#WEKs
zx&Wqe!WfIS_OeqDB?YHSw~C9XH~iOVG+)1ZOH)bwG9i%o7-=$?s>UtLr3e)8OCW~w
z#niSz63|Ao5E)NLq*iQoVG*$`4Cg&gC7N!<o3~#P$-K-9iz_+(M%f};j|pvUX%$Wb
z!#Hq09(nWX7B>$FxhA1_+7GM%{y+i0`Zf1=H|!6OJb(T`+w^SZmSs625e1jv9F1+r
z&J(<+Qi8+zz{{gSTEh|@d7jY8qckj1AiIWTUU=F+p{=E98@f%;IY+b-$SDv$ekOhX
zjP82ss?7aEUE0yk%?ML53?rs$kkS%U;_=fnDQ8w~Rjg%mi9YTgcZ@Gj^y@3y%@&~`
zWJii;CSOD!oxEe*?|_IkhGjV6Tp&qHzgpwRfz#ocl~FXS4aXOQ&w+T@u}mXMYqV0F
z4=?n6&+Hw2w<>dJTVRA)O8V`Vlp-(tJvl?$G-#Qb$1|1bI3D+em`PdEi-I8-hZCFa
zHBy>V`l%{bVoOMg(>Nnlf$7zCiyvqFcx0LW{7|qml|`xqv_LhM%1Vrph~>!h5C2j~
zr@%6e7-J!|R0<RYJ)Y1sZBG`Nx~gf~7Huk&P)y!Kw?@qiRtp3Xtty(PCCn2+II^vo
zG8m<>RRIvj9P!h9i6QdK!@4wrOR*{X(moxh6Z7$rcklkd`8*U+kSYXWYinYHX%1|q
zEd{p_a7#ee85vWV(_esdC9+!Z;Kgr!0jZ@-Wh?x`rHm)g*5X|#f0ydI@S?(nwInFK
zt&74|sAAa*vG9~qDqRRJuIfLY38MHaWLfyelrC}HlF<3tNi20a1Z-t~@hANJ&#Rxc
zp;97saWWuba&syCi4x=}luwTk_TpkFf2O~ry0U=iCWNVw^FWwScn`wVRI9GA)Lf#f
z*0k%_NG&<<Ux>#8=i!JCfqqkikhqx2ZqjrFa`B?nYjl5wQi^_k!{P9R3p2y<5keVP
z_O=uQudX-5*m64VF)WB&ilWg;ln!W9-t>%VimxFRgi8#b;Ah-YtcOY)s?`Q-GsC#&
zygT#q_?h#0Pv6#9TT$0FZP#5I_{AABPG#J#WKo*NP!8SfEuC58+kgAU_ZK)6lCm_L
zE2<O_BIA8w^-s$}Si)uTm2u;M$&RjHF`mvW!@fY8v}Q>&GD@`2xa9EJQ}-=5H#em7
zKs+5;)isOQvRz%FRa5AgDdCQL2!Z3vj^use1g$!1RRB}k)M#6=o1b{ueWWr9tpw|)
z!dSun=_9o@^mW5O{pWuqX36&Uj&8jzl*Za{dwY!(g4vG+9A-5#0x}U^_Kd?s({!i^
z%P=FdhGk()iL4ze3oIF<B@Hzq!R~2~Ulvj-WtkKNGQ)H{GMvtsZVS3X#Dy3=!*Ify
za+vt@fxJw7_~Ds;yQS?{NUiAg6*l(>pTX30uioOK!yTW{(+TAr^EeYvEg@!Bm15XE
z@bdW`s_m(o4%~>F9sPPkN`jOG!#E>^;NimqB4<v+k*cu>shIpl1|t;F2bSoG8V0{$
zs+qQHsG5$`X&}auqtB;jp7x(GRa<85La=+@(QmIx8Yo8EZq43JW#+%k1?BxGA3q&<
zeApMSvemS8O-_O7JRx%~{%G%L*ITN#Cr80)_mM*kR8@_(rraDWjm(iSPh^*H(}*w?
zx?9t3t|0^t4<DFMC+enW**$~SnAMh?3cyXu^6DOzOcGoGjj~wQv?_<6EKcImxk`+u
zJ=6KfaraUJ5Fs$T8D(@?5+=bgjm7Rpxn-Zng&Z@zsYy~ZEh8$3UvldaAc_m!)l=6O
z^mmF#ZA!H9GR90w{BqkCLcnF*XN)dsz}kY0#qmreIA6RCKR2_Z`?KFiDf$&Wi1@iv
z5lbK;U+91Zwx}<-jZ6FaQtB&X8C{m-mqVgpqYD<M^a3J6|7^Y}&H=5<^g&9AGNv4S
zMi8SzsRnCIK~(q&l;-s4S%ODHY4h72genw)^Tcs~<m1Z==imQGcYQ~{+8{#Yi!Z<C
z?VH!70d4DIj|+i$9Qnup{GWJy{y<wPzJ2>8ciWrNs-7H1CMua3#z%x0i>lWLgl$16
zmN-LY3OPBJpNo=;Oyc;m!})-zTdKOpFOi&Qa+qjrp;Oher7|@i9zGRC>!ts6cYDWv
z_k#0hHtWrWkwu{s6Vb9IX2HjY-CzDtP+>;I(y;W)R0xcwLEDzDTQi>z%;zKX;>nvE
z+SL_~0WJ)bMp}BC7^aa`zrhMk^nv5?%;Tp=t~VX~=|m+m<7wjc7r!RyntT}pkON_w
z5pm(4|MNf4bsfL?&2M-)ABzbql&0c!ZyDx9-E??YG>=I@zv`KXGbtJRzUO6sL`$fx
z;y9eyJ-x8mtWhcR`20w>S|LNCG7<!++LFB^8AY?*(wLgM>x&cM<+1QsR1wQ$QF<F;
z3HW6xvr!@GR%>+IbAEY(>^SUC$Z1AYE#58oX~8OmkAdCs8LI_)yXNlZ3Y7(?<ALBJ
zF)NH}XsS$CH~4v?&XIAMd3^ZHs;Q~1MG=|Y!pQ}c=@BA;toiBvGg`u?)r{jv)Ag99
zA%?(kJ|U!{X<KSpFat6r((E{#Pe`RuLSN7z<&}J#XL1(UYM`o$Iy2`)t<6OM5~z&k
z>GLP1^FS1t+uK{_ap2+MiPNyq_B|#Z`SriSD!lAIll+?DT%d_cDkNZJ5#6|X;Q8s1
zS8oGrW!N7cI6i+yNX2S>g|-cI0F<DqG&&{>u!KMql6F18;ehv!m!}V8KlA3RuR&-s
zf_!=S;Sy2==3!Sj|80w@da~4bG$Q39wt>sTG!mRcM2{QK%+tu?-Gy{oZo4idT+E!O
zf@1I?;(cVlKhbm@E(we*OeZ%_WW>cQqA#B2P*OaUh3-5rRK^t9yC|b3DN{j(xTQel
zv?cgU)Bm!3mO|n$N>DD4J0U4r&h+QMkDr-LLX|@ZsZiDu7RS#vJ*{;SE9FS>^rZq)
z%%w7@Y;jS<m@giRh`3l0(*+xyQu!xyE{$@8LRs_6n@$q>mk@!7j>-tC+9K)}VVg^k
zWFRjS^L$_lb2-FwWKn|c-7D(rJ9eiLQ#Z6-OP&VSCevw(pM4mZ&qv6S`E+7iSG2Fc
zpsOvd*35?;MkV^LLim}S96C%qz59%*dtSZ%lB%{yQzM0CStd@09d4QMZb6c`xxGV$
ziCSnHE73aR+=xq#c^nWjVpYNQw{^u=Z{Khp&X7yP{=DCzMJ7tc@pLF%Eh$S7C`DAK
zk!D3Se;zIViyR6%d(twpe|o?#BVAk5Ztt00z)Qt=K0z!tvUqr4bSJ_zT{i2Mc*%HI
z#{L4;iZK^Hz5uB;c2$!?=J@!?@%l4P8>*_pC_!BrK7ITF0o>eN6N6(Mj#z7H>Yn|6
z&tY2Fy!wK)y5`++0-?EW48#6`+7@fp{xp(=;>&Nop{h&oLJL7{ZQ+-v$m_dTWGQ+4
z^oiXxa{uNY=S%Tv8W(K4;#a@=b-5Wop9uRA<OG#s2$4o=o}YKb7};#L1>_e3=V4%Y
zdLa9Srna<gMOK**M!Kc}52Q%HTJ!uku|E%Vw|5NFL>MN9)0wyglCG(xWs!Sc=84_o
zCstic%8_B3nCAsVF|Hv5!+D_ZTC7pzF!5&7vfDr5W=U0TND`{1CeI^6Bt|!3yBbwl
z4lf6mabVqaxSa9Jgr62fiqy5?a5&Jfdz|xx-I2;#l$Ka)SW<a74bzU@(-YBqZm+NL
z-XT&Xx)~!PPoI7Osrl90+yAeMjaz0!?(i2EHI#zav@B$)m?lTFx#8;W4M7Nwr#+B3
zpUylyKB8?yjL@w*Zf~y0ZeiVabahP-f>t=D^9hj?t&Z#t2ZqxgT{TFX(7_{BM8?SZ
z@iU)4yyNc6FKO*PSsIjS%K9Ki0`9WV4u}*9Zel#XP{{%tT%2ci{<0ghWT_cl;CLFj
zxxQwZ9WFexd*1OyyTbcW2%I?)Qf8bNVip(Jphhc0ju4_iibTA$w1p@*b}I}a{+w?w
zZx8a9rf?ym#*`6S6=qk6zmx}sP)KDjbLl@j6@ETWO7mGOv{n}gCKsgk&#msV_>A5c
z##9Ih!!HS?(riwp#9IE;@^&FrNlT$XE3#B%Ww6cW(vFVBh9@&3<>FUJiLPr9bckxr
z&Fc+r9`UEcrB|hkGI%-DzP;y_shB6v{ZOzKIeGT+#Psk7-X9-{s;2E$C~Ht6pqGVB
zr5Tq5ODX0#cgFjPw(d~E5W>uO9wC*txDOvc5%VY3n=7mkOy@Js&qxK+d}5kLo*zCU
zq^4b8v0C+fe)k7Hef&gwbwgcO)RiuJRS_;2^b!|_a#$|&fY!QP#@xbYb5(YjAyZck
zvygP_S8Utszx<&vC#X<JTVW+Qx3K^IzfiYpHdnVaO@m05Fi!|uGdaf`C7Y`&yz?AR
z1Md7lv+9|bk)|!;lhw8>5Vdh4JI%5z{Q9qcMZ_~tp54o?e8EX9KGL-{_itWvy<KxW
z?8!oM8qP!kZPzpTOg4tTw}h~;dwL`<1JlV-nIgJatu|C$$23ilGvhe^5{;Yo2ciI7
z*J!D^y1Buph(yu1J)5^TOiSeb4<E2ii@Lj_{mp;FN*c!V3s3*>_nd$HEt8gf{ms|R
zA>e%|67mqhEf{SHQC+YFk)~<!!SVF@6LnpaM4>P$?>Ia@qoqKJMBQ5GTWqcI`$vQ>
zLYH}-X!?S<*fcF4K0PCqK~#da(zIQTFp?+(Brwu&bGI!VBVBA;J_d&2Oz;cBd)zn>
zQJAK|<;06$Sgkjh7zt`-_xK4Z6H*C2y#GksbktSN_rLoCuU_4;-K@|8o?dpilvr=q
z+`W3kVShx+a`Sb=iDj91bF)Ee#W)Ul_vbg$x@|Z<&#bp=q%jBtZdy1TPDpdH`w7FR
zrzh%4)3=tO1>I`R>2zioMzqMRYr}cB!?~HqhYx6_shgIjE)m6Xc2vf2d^vDF&L|<+
zTwindtwDE=+6esgOxM<Q-MTE7V;~NDlu<;};4i=HB3W^`dBhE81m(tjb92jwPoJsk
zirG059;6{8E<7QK)6D+mh1*wefP$+yACKs&BF2RDfyI}UO=T@oT9j6o6pz5CfK{R_
z&%G$xPfU4$EORNfX-y8LExdS#l{NOllM<;Ad-Koo(hDa0mowoajt7d%SxRIsK5nUH
zL55tSvsztBX0gEaLM*h_5CS15tTDwd7t1&}goLe1oH<_H!{y)-2xUpqVya?zl(`hy
zyk83SHf2(F#4zAy$8z4Y3={adC~=#Pm?FYh=3!=$0fnHFnKT_pVM6&4+=w(9RduvY
zgG`S3c;<9G5d>r<*}d#oH4U1`JTLelQ8M8CLe2`MEVBn`B-b}rIOq87_kUnKpV@4$
zxxT&Q=KfXbrc8$t`!$wGNglp_`wG!)*lceg1aj~|WL!p)OXLXCJQ7pHSWP7bjVum@
zhsP(LUtZ|iHTt^Y`uY}a44*##@RvUnn)Mx~@31kD<{6(ADW0)LU0z?x;>fOh@DW5|
zs4J;RDq*a}YxhfIG0YRjTHN4330y9|gL!f6-o0lTCxS~fw|CgKDaEyUCdNR!E;)Ck
zM4F1=;rQ}`3yJmX8l5C@nmLU};`v0CBW@m0eTy-L;j@e<&c`Dm!#FS8y?VvraKbAX
z#sM20ecdomC%V4FXBZZTYI+WHL{<&gcQ;6pNXAn2*Hmqb`s<PNr@EN3gh1&Emjpo=
zs%}N9T6~JsrXmbONeR>y?&ZL7f24ig5ZzLAoJM1&foL>K^kq*XL}`#~fk-u0Tb9WY
zeX)U6m0=ktQaTrGd3%L!8!FqBIl9d(ex@-kijpB;rZc9lkwP(tfUO&b(?Hv;=r$`<
z-!L9d$Q&=a&6$4PP*;}Y@c>5QGK~8ZS}BI{h^iE=t!dX=It7(h5azPTv^77zd(UbA
z#Qp6xwyGFs$LeZRb^w3!dE6f`=?3o;DLa}AfmiCnVfrus<+oIA!~L6AbhTzYykM<m
z-Sr3;=<5cT0^>ANNlo3>_-qgfK7IU5-&E9H%`$s@h#bcm??y~Y93G$Wv*-HND`I0|
zan#LKd24h5-4=Yoyx);^h3>CeT<M?W3ufE9iDjDDu2=Zr*`I2X5KKphwk0gkuba}m
zkCF478C+yo7T(qsE<}6;gT(u#{Cy%bIZu{~%2=$D;C+$Hlkh0S1&JMF#F)~}@ZOc>
zx0Ij^@lqg5A>aV53vSta|BDpJ+6w;USjZ{<QaG$kc~cRhxU_{<T(;h5Sq>yE1o>xg
zwk^8=B^5aoR7SWsq+^JeJw_rdBTCteFG8WTCT1Y|U-meeOcn)znWHD>0uGc>(o`*T
zD7WcEU>HWK7+KteHUf7#0|F}*=i?qNODXX2ct8lrO#{{q42J?p9QH>Jrz6r>?(Sc4
zwdzqKlVYan)_5Or7p_qjHMZ-KIg$jp1oJY}v^6nBf|U65Gud{+XaidE@bHOGpWdUS
z;?0*|B2;7=PE4sZ-Gv*G?#y|ZenFln30fw!R#dMQwXNv8miO=9)0hS~P0ZtgJSO&^
z-~Z(g1+h##eK?SS?RG=AS{MF}vbb?1BM3QBH+9J&+mhY&A!4fv9}_oUeOX>zLqKW6
zIL};NZ86Far<qX74y-me2z}t`rysbnuleGOx14v6=<S+Hmumi}rzd1pQ`w5J4CI{I
zeR`npH@v)j;BYwb`prF}YnZ2rrtO$!hdYhfY36i15>n#+%dc3kwrp-+;gck|ip}<h
z5Ip;r9h>b2QCX_YFYV9u`i83O`TYJz{NWL)YhJ(ox(GW?FBm0Ra7@#}tL=)mX$rVX
zB#Z_lG_{6py+#Ph%W=>1<7cYMLJE)+wJpN-`gTiNBKhD+F_3zTQ$<#O9#7O&&Hb;w
zVZB|k%rkCwv~7)+imq!)sZhv*)+W=2hzW_^;}gg8z-hPR?U!E=QS6UrGzGn2`<|!g
zC+e`!^eb`*2ni}DocGjqDaqBg<*wUsnikB>72aJJABAm^Dl&Ud4hs;N=83lJIGzg`
zb<_4}Wbt3m4tu39Zs$LK>bhY+MPzoIj(b*3gGi-+5Od`1n|pjtOv{YG7HZ}^G9Y<;
zdZKC?q*2sePi&e}5)zW>a3)NN-FZ(vRcvlwaeeoSs_xi5?{F^g{?jx6{QLirT3e)!
z{OUJf@gM#V|0mARA6UsZzr4AGWoCSNrkZD#keP-P%XlQE#B_LKJnYb_NK?Lg`--Mp
z^S}M)e_{7>MoLe{vIIdGM#d=sC1K>21)~+iX+%kl_kp(Uev$4}wz?3A%Gf1eMnX#I
zVz>*b2<$J&h?FuR1nQ<`aTl|oQUwd1VgU-K;$F@%qw~+};XnU`;6W*DvZb?8mM8@c
ztu!%C<-cF{1u-SEi<fcH1)&bOi~UVfd<9Y!Izmc7F1-RJe@1y{R9Glt+tSR>LKJgh
z7MIeWXC6n+`<Xc;M1sC8MAGx|M9zWW6Pxud=kp0qrt4~!c|mEz>)VdmCFauv%HcS&
zc+V0XZP&8+z<3^LDoa(j?2iM7VJK4tW3k<quI(WuKD_&$uJ5>8Z5WPwR@-Y#yJ36%
zin?xi**)OKGg>HoN-Rscz>LQOPw(Dw{q|ezYK5{Lb%g1#<MH8{Y5xhOC0QB9m^htI
zbbZV5Trk@)1$12zB^-t$r}0=ubnd@Gqxm2Im;aeJUwy|H-+o7yiKnN}_<4qu$tiO>
zp18ibMpJZtFZ(^p7}~b5ZPuGLE=5vEXjPzDN?WESl-p^g(K=#vgX~tc{hD=KA*AB;
z{D_;*3~oZ&j?qWzX2t9RDNl&J&|TH^tD0f=fUYd{w&rknX1%@wRnifrX;9YEbsf$2
znyOo2n~th#aBd+A#eAMhnnP>0+iQeYEFm%O_cYd!v*gND%;SmYcRzwVQCZ6mFP}&u
zQ8x`;-(%d&JWU+lzbp1EZJDNt%1Zjpy3jc<Kii}vtG+_Zn&3U-SU5#>-4mn7i{g|P
zLgHsnm>uWiNRA0#dv307aOa8QB3HbA`-bR?QLI?T3f>{k3+FJgYAV{gM%j*jeNDIN
znNvnphCF#xOvuL4U$0RShQ}v1cQ-7@BcDEgLI}x~w4A1~e37!JT{Se@HPk)2+R)Tn
z=HWBdr6g!_D3?!DBlU^KNcNAPczoKi*=)GIy~S#CDL(%3vt6ypsv|p(a1NCU=c%eC
zQY#t@S!z_%(6>G3-JbPk&5{!9%P3<R2By;hNz$xWWT0Me_|<>B=F?9<Fz)xPC`_i!
zX2ZH`@F8*g_E(Il<;QoQc=_}Lr_u59{LFEjd8-t?R2Xf^F;PoFnos=hAOD$`$4?j$
zQO07eM5Kk&c_gPq^qDZsG}@A!XBiikSXe-(^O3$=6}-sBArL|$#Dvhc{7#i2rcw|~
zIh9+#5T#5=Da;8ekfuUNT@FR1FFS?O5y)b~R;8@ved&IXDL_l_FR(gI6p|%`3-+&k
z<b~xQeDNcfJwQQ6NRdg=mGREcO=$Hq+%Ep2Dh(GgPz?DOWOqiAlOXsqRfs7Su}%hY
z>0J0&Fx*0vX@gIJV!)7SsSzU4G?FX}!TEUDaojB!t!e5Wr3E=;4lmDy7%^2vRtnuV
zgap$PP*uZfwIYm1QVN`hQ!&da!M59=wPrXSISwOrf5orA_zG23JUl;@Wq4Kd{`=qa
z!w=t=v7Yx-A~Fw0Li9K-==3$hoe_Q}NyFy$p4F>2gmIzP6?r)`?;nVMKx)G<OuYN?
zM-<Q)Nfa>56HCgd+9IloW_!(cy+$d)JdTXRSlasnj1(LX`@j65u)4nE)$7+(T5~w<
z=xW8VSbU1KZNqhKS+CXv=QtgYND_Thm*!|nkRp}JJPju*(^4P-k@r7-&u|#{`kOCM
z6q8ha<<WIhzKkhwJ`Cg>NK!H^j)#Y5VmctC<gdQ{hOX~GCQiFUNx@vLX_Q7dN(Md^
z-u!X@!r|~pUDwp>Yx=7@w5bT*)6@;-_J(6F3*F7?it}k<KAo^t#r5{8NPUKp-~RJI
zfygC>w%HV_>oBm)o`>_q>4y*8-*jAGUzN(aE`KTGlG*e<F+@TLOyi7Dk|h@^Wkw*B
zWEv)H<1h9lO}FakZ?-%<JY!?Vq>St&=V2^V&74{F9a-dyr7IP*hDda_DgZK>>6?m}
z71WlMtY~D5sw~b8w3S5z`;YG!&Sz>}Vt^)PK0obPE>iNVSFbSIvMi2jwPw57l4}j~
zz`E{OgvGWUNoT@2{Zh1gd3rAJzHW=K6JZjVz9-BISyf1DkQe;$pMDI2m6o_U;c}#I
zYL*mu_wgOKx7Qe@$&zyLS1o3xNMXVG1+}i(_ATS?8KcXM%w5pT)=0)_Aji!8+pk#N
z+;ALcB$--CtcqM;b$s*Pf8aO&!~d1>>4Ciagf$YC_sniVoevxr$MgFi2`O>+`VEht
z-m`!FK&=G2Zh7<ep7nZzu{DRoiBIo-V)yjSagv}j-+cFm_0@*=?;e>?19epw=vjzl
zp*ft+*tRDsLAB}-RmIsYG*yML24ZPIM<iNTC}Z(%p|VvO)uJyfqCzH(DIr26+sv}e
zXeC&d1#(egrc6l=NrjNA_>2D(_9sh=J*RL<0_nnE%BgsZQz8Uc_**eys|z$wmeOSi
z;evn>7o5ihKqrg48S*bCwgg3FBPBW;VlIb3(cNZ5DywyuGFmEZRTF|^o=!}s38g@)
zhPLa;DU}^!Drk}fY^t)q+n<iK{fhOg*R<V=v`jplUvO^0b`3@dPUn%w=V$Kk?pd$b
zjPp#21JPA973gJPI***IhB%+N-4tk|$$?6MZfXvvg_oyKygYu!xrLkSEra)TMzgtl
zO<X+x`uFdM;|pJ{)&*LaB5mIy?(T3d;G|}9p01W`YlAeBtIe9%uf8OPSRVd1qfr<w
zvE7Pw-@*Hz{_=-{X=>iR`<~<-ukUZkDbUsp<9H?n$7;3W;o&1PM^s8&U*F(^V_9a7
z=M$^dn!8sw%&UZ_0MZVRPprF!S2s7beaHFb0hK)S!+V6Tv0X=&`7%>91PrI+fb$8Z
z68Cqn7)}Qse%kTw4?i%T&Pb*CoB!}1c=h#n)a?q<*0hzyFOGSMg!sffd(ICJtX5mb
zVZa*AvJCj?M5RHrhGjS)VnpYJJDzxcdO}}rXT#}?Q1I<nZ`pqPH#AoB@uwdlWlUA`
zay*g718uuPisD<Hmzj0j@$~${;dCq|s;bQKvobihu-$H1Lf~$-<vh;BT;v*G8jGA9
zove_xL~mQRswPo-B8!CJCYJLNn;~YwYSlwdSSc6}du-j3tYzArSYAf-)h+g_EHC}E
zFrH52lsKQyh#09t<ior7XrW0)()TN3h_r2st!fsxu*?hl;Y1aCKL5*aS=QG$S#$GM
z#Dx>5c|uBs6@vME!c~F&SahzdzUS`tmi@~Um9|v1B{)Y&f1Y+Z?w=tGtTnW&j!335
z4fk)}V5}uAo?-Fyn}*{8-PLOZ(4|D*RwOr4+ZJ~|vauS3W}FwAwnfY%N-OHRp<Z9J
zty}yub9#Ef#hG=LiI4B7nsvE_e*FgA0?9*|!3{LlvRQWwF0k6J`H%nc-{Xg``RRx6
zIgJa~uU>O={}y3u4#$zV-|hJD(@#7c7TT5Jum0vMno9ATU;mot&wI|NJ)b_psu9HH
zgqsJN%AiHU+R~)e%1|{e8G+UXzNhO3D+St=w_<fsmzMs6LK<1dIo@$Poe)ZrQzqtw
zUzQ@^$)!vvwM7YqP#ST;F#P<<IiqB81?QAV;_|j}0rOQC?dBXxmOdtdi-G2{%gZ7D
zF`XnJ<ic3eQeMO|nc2Gn9MlGo#jzvH$jSSKnB%22o-j%vr6xpyttw(VGmmGit})#P
zp9(ScH13c&ao(S4n+hKzaeqRnktKMBap5*<+;QUN;Un(6zhLAQUS`^^W3}FrV&rf*
zmbSOdy#K=!r_-6*)bvfqG?q7jw|BRUOZi#PA{Sr7^;>L8I9GswxoX+0CA-6(4=+c?
z;~5bHE+qb+fB(Odgrx5~?q9zqhKR_SWg7VMtFL+U#g|Ae%k$TJe)z+03BHsTl_=@I
z<M9OJ0x^{M@L&2+sN_OjOO`2;-OMyO$c{`#w*o&7lndiR3W4!lxFh4Tu-#q}mW6lk
zJ`<WXqV1`zWjLLvWJXjKi%?LtJpK3wm_`-{uYdh(l2y2zXu6)w_LlADju496M~cjF
zm>3q%ZUVJ=jjjX}f|NXGe<ID1PFZ|_ghK0zm)(J?G3?&|$nC4wq-8|Mnen`18b<D3
z-Lpu~@x140w<E-akqXn)msjja-*m*33B!@#D=MWK#skawg>_v~-``=4qNysDWg_Mx
zKQZebQ<<V@43Vzyc=!1uy{$+>7Ew;sFz=5@EqQf!%kc7!lmu0GMQs$>jm*=DadJ$H
zE3e}zQ(J}7qNI}iLXMFs1?r|^SSGZP%sCUZqO}z=X5=z+{P>yk;mGQ0Lv1axku1YZ
z9zDL6+<*HGN+`Oj2t>xy8DkA~Rh8U)Oqj3V(5|mho7YU^NE&9cTbLIQMioI2iDuO?
z9wUBQaLUj%Jw_?&+LGF?Fns>xBefuihD8a~#*$q?gv55$5~LuC8Owq)4We4(ZHH;s
zNFT@qCO6XBmTD!5&N1hV>Nbd`#$ITx#%j9l6>|VB3|DssKXzCNl~_1G{k{MahaUWb
zJngVeOPVrKY!KFRvssa|W4mdg0n!_U?P;%Gp}T@Qx7X0!4D`3JxzBKS_ZovjhLNRy
z;O2V6{8{tOZHw(z(6l_6%+u#5D%I099Vv^FnzA*@T+%%$($#B3^a#obGlh$uGbNNM
zZ|loEF^}hhQeGAy<CYN_3pF&P@}^le0xpm`7neV$q^5*h+%Jx5p%mUZj247g;B{O;
zcPST{QOJ}BeUODglncIs^2Zb~VQIDj86_ls@ujgX4Iw3Lrj*>$#lV&>MmH@>A1}ce
z62=(Tn>Euc*&WWjdh;FIyE}IKkEj%J*N)&l%QCaxtbt5jSA-Zi%rky*=%%8st{^3B
zQ&TlHb={)1VH^gM4~P^vAC6SIp?iIWZ5o{SNFQ0tasmFlJMg?e7Gkt4CDITR`)MrW
zGB1#(p|=${M-Z82y+Lj^TwTvR@AkNu$WcKEkOEuf@Voc?{_seYmM_2gj%7(q^FZIN
zFsA0=;gN9gyu3Vfe|y7ZGRtuI%O48TJ8ZqCyIPflAOw;x2wYOaPZMrgShYQ;dEoQQ
zj@niiUP&S#inQGA*;Wm>h3(B9RaKE}3o+34J+q5MztA@;#OjJoZ5a<c$cg^;maI*g
zlc#b^p3O6x_LhIYc}sPDNB!mvw$hv*Kd=&+Fib?#a(I4X_wjqSt;W8($C`>|99T}z
zXpssJJw(>GH%v&{?Jcd7v`e^1LL}?9p<8CUm8EHGuGcM(KYY*c{`I$XZAabI)LlcE
zXJqn7TX8%_c2Cb2_pD@^XSO#txZucA^Z4{cZEH+p$iXw8PI!Idsx^$$iT#g9V$Ph-
zXKH)kYP%sUMZbAoW@=lZV_;rphI7DNU11xGtSVj}9%!W~R<kTQ?_L1Odb`5AnZsc~
zr?LR<)-|iE9@DjqFMCe&h!>JZOVaU30-la1$P%k9b=!b}Zr!1UBJ|g!wkM_$p**Ym
zH>_>RPHT}#Zel*4s4Ib4ugKCC1htYRzaXt9cRiRtM`xScZ_wPLjb$80#^YyjGn-Xc
z)SguXs^WAwkn}>*XKXhkm0}i|y6JFXrs`W_NEjh#sv4^`vGI(@2VR~Zsk%GXuU<2!
z1oKS(9Ei)AlRGmn6B5bC=XcEBadmUUFb3r5f%Vk_qDIG>FwPiV9-&uvZ;-1SRDF#X
zifq<sWvKNUySeAaSd^*pi-Y+@TN#d@e=M<LTl0K8)0m34w|8vYj_KtE;|7$iu*#x!
zMCkJSkwO(oPZk$r86_2z0wQ=$rvm}O<Q6oUR#xnv_NYeFS_LtYRYi0Yw(eP?z{o%x
zXKMTBev2F<ZWxGUT#6`dsBOhK%t(<Lk8_Ee=duj8wPQRTXllze6_!|6Rlg*ms;Vxd
zr_0-eu?AJw#X%A>LKbJlG!0l=AygSj#uzUAv_M<;q~r)hDpOHcEvi|w=o41K?)^`^
z>>iOuF)g0$?Uz&$=HozjeNDH%CU{T%<qfiGS)M<z``tejq<Gt7ZN=lmCyY`gU(}+t
zwVbDcrZ0k|zS}T}3SlkowCBUIyd~e?-_f=meO)8{ncx58|4ee8=ac8^?Kgb+<u`Qe
z75m|g{JO&Xz%WcCZMpiocsQP(o-mar&S!>U;e1@!Ufr<mDinnSwz=MtX3uHAqiJho
zy+wB&$NgvS>^0WbfB8dU|FYxpX$MMkb$!LvdQCE-EMCR|Yc&eX(_zo?ctTei?*f&n
znF83RVj50lsS!(GKu!v-wqM|gJnkMDr-7T-UtsE%tPD-xGdw@zrkUPqjGNI_OGXox
z5dpmZ;w@ouoEFda{!3QX7P&0Ej}NrhS8&|HJYiyFTea+;p18GJnpMMbzvJ`6KwDc*
z#}jo^)3ht<x<x397cbnsea!_a$@8aA#hgWE8P1F;(ymr~_t)QHjpcYcVrq*r8f9y;
z*6j92e){}?mWns`_iV0e)~glweb4jD5nI*74CiU!-TTi3S7MiK*V1+!qE=M>1|A;>
z^BJrmB8XaHYm4*lVusS_#*(xky8v@QOO5OsvN5>RnLwtp7C+BaZO5DMz9hxKp(}0x
zAtYgm)Jo&JHNX1muUVFb@%aNbI#h)FFTVh*2{|HCz?~*SUvqPJ&+&0jSZ3BY*Ywv{
z5ED<w7Z4duB8EVWmhJTwLK|$`5u#w49kQ-bF{0(4e=U=cEW??)?x?lq{{9um{WFe$
ztSZ9dIlVmd^1R2`2Gy)sx-(VVBDJF46giTQ327>nE_v=uKq-OIil(+4_D`s;M>Q+j
zo3|+26UHOU@qw$Aqibr;!%#ql-HP++j319^Z87~8UDfy$NwO@ytzE<F4$<GDvchXi
zNRVu4cx$mB(t_2QXe2S$w0B>j*PGHRAAh{$&plNQ-K(pDM;S)caHJwLJI^T=L3@sl
z&8oRL(G!FBxbZ}G4qMd-TVZn<#q4#U>aW<W*GOIb({7$w(oB;Jsn^dFlE`u{P350I
zZeDozLy^doCUA$Ag5h*PiSm}PYCCdHe13f3?*5+Pc*I_C&p9Q|0%Xp#s})l$>7=Tv
z$YkbaroDKkLvW;&X!@=|AKe8BPU%ZPE~P<Vl2krS1vw&Cq!g$nct7L(LQ;mN+aT+J
zO3!TTfEE#{PpH{}&zPfRVjvA8>u!yank59>^A4FZ=i?4Ctookyx?{ENIFDzR$@B2x
z1N!=&)n-f6_Lw+ve0Wb<MqUm(zTfY8b#sO3uQ{A%?tb-(n>TN<>n)RWyzKVO^US-C
zAMhda_N%YaDbQHSSFf&Vx}N{?FaJb){Y9B%Of%6hoDMH!+t61H!6*8z!N-UYf!8-%
zPJPF88u{^u{|b@R;2oQ*Ta>L?Z#Tr03U6JPf>Q{Qm&1X_PoJrDMb#9ifNfja^@d31
zwBND0-q6*SN(ACE^78QyJU$=DLLqF!cplIZM9if99w7uqDVFH@^!&hezoqILKEC^&
z>G;g6n;S$DxZRHA9D}ORGNZzt)5)Q$azjZfQf*d*(;hD~IeC;UU-pzDrZOND=TFZp
z{xh|$dHlnVym@_#P%vi)vFI_i)x_z9NQt^{L6&85R)tzwHBE8*YK87r{O0feJ+&<k
zX5!4aEX>{)q9rwp^F$A;tDDQbHD53e2GcgwT}S8EyetFrJfWo_E{?vgnUdgTe<XQN
ze}6~F3E@4~2kL%}s4bohrlL}Yx~dCJF$I?CR1#u+j~hpXR2W_0b+v(NbA?(Q=iLKB
zWwzIAbY;m_AYx`Zc~(u2+_v~-B6!E?G+|beCc)DW-?M-IOk)%g$&@wQH*b+QZ!p<m
z(u6xd;ZG-Q+R&_CQTnof{P^jJKWL_xJ=$2-+YLf1c2CbpUm%RZIhxKOR3?a-+GZ*h
ziP19+>0+a*N^?F7=HWzAf~GPUk*KyCWK$!SGpo8Ki;V3XjC6=up_WJ`JT#J|H)z{o
zjA7K0#e32`5ppCbjgU3MSdv*I%$6K8kWjf4w~DtxkTF=55y6p-V9A!GE22`=%JOYx
zU_M}Fgd{oaMuK-JR}NiKWl#dGIz*1VJRg|nKxJT_M&h{~9Gj~xOGt%4>>{RK@mK%u
z?=W?T%@SK#5`o=^A9+|t&c}hpC+1-!B~MCIiM3un<2>{3!vjhxbmci7&UiN=W1+N?
z5@F?8+{m)b9G)II98OeK#d^IeI|-3XvPdaRWwBLNc2v$4nT|A=3n|ryKqE>!I=Kb6
zWL$7Xi7EwT!7|NgD+q2uYDn5K&3m?5a2$@Tx)n{`ltzC{tfMFFADK@Fgsn0Cn&2Xn
z_zcw*+SE)-V0Cp3#;_a?oL+WB=ct>C!(q>P9$9ZUD5+R=Yj(T$G;Pgsr!l%lhmohx
zpD-d3eet#&&u3(R&E0SQ8;r7?haI0jzr)Ql8bQ~!jOP>0c7qm~m(TARaAXAC)g8-l
zroDZ|{j0Y$b<O*qe$Vsn2^BM5_Xwleefn_Ovk3IHB*%zU=^|?S3m*z?zhbrOFQ%VB
z41v>b$7*{;M544}-EWz@5o>F7RTE?4JkHql7Eh+#-0=FVZ&|O`oDVzRfB(Y6rw^Qm
znQy-RHFeiA&1Wj3aKn-49imy4Jn?!%KyZFOaC$g0>|cJtBS$GwP0e9)NSVmXgz%1s
zcaPlM-f=#@5d4C*hTC;d!%SM>?rOt04&)S=$A!~wN7t{(O7e#ve&qY#{+67-<@WxL
z_3aJXSd_Au+F;fVI(i^;wOQkw<D4!_xRg<KNgVkQKuKl?%QEA~Gw0(Gr6k{c^DRx&
zlmd=cOv{LKrGZ)M{F(jZk(cM^ODYB`DToBB&eC@c)w)Ksnq?S~LQ>fZ+qH<gMj(kk
zp_HcET;b-qEV+~<I?w58N89(*R<j6!Z5rD33O7!Krx(oW5uH4(0aVXKfO(*frl1aT
zCL3_iLrF+94L5JTX0zE6rxQ^rHa9o;Wu~fnu5Q1;UcE-O*EGyLegB@%fB09V$SlVr
z$1f~z<)5dI4v!DGQ%%)0C;{VeqHP=U-5t;S6Jeg|+co*JLut2H<bI9T3h$RvptQw4
z#xf(Nph^Li6DiC%AL(v3m|7A~C-xuq+<g5jTu#J!q_vWq4O0NKx~ALS7svDQneluo
zz_vJ%juW}+skSSUG%O)wgebzIl(0FXFZ>w+AS@yUWQd3u2+@<#NTZQSV{(PkJ#JZu
zGE%*6=u)OW?rDrBct_uK%uSDT3%A`@gk@wHk1Re>iyEZ_ZNH&5mY^+-v<NKI{)E2P
zoTH~%Bz1~(DH4Y>i%Y2WE!(><na&3u-u>>+b_yZEX^u3N=5!oN!7MngSEW}o%@a}=
z@LSF>Ee>fcS!seOlLhCN(r3tm?8`$dri2iRWm(AiOx?B@GHT*@JR*cZ6G$PKky|R<
zB&`(7JQ0IJk_pL?vx98;U;pv9RDHwiw}r~=#~D9O+%^rpmi+MjCxom(Khs^UVH*)X
zaXv2`&nKGAimU{}1?KaKtL+vS0}sa&{br4r7IYR&!;by=$nQUY;{J;-S@&zQE2_@(
zvLKD+YJ0<J9Qn=P{vG|*E9U)<zyGKIk$%;1eZ4_pI8Bb%Uw^~(7hiF`eZ}#3Vto0;
z)w*LkALy*$u7daf`VX}0EvMswX_&ZM-_YLwHJi5M)!PGZIdeW7IUG+|U7kO``s@Gp
zFMlZb1kO7ur5K0Ph5fPc`1Fiz`vSru)3jZw?VAc0XN<KxJ-@KJxkfiVn3{*ffKe^&
z?HAN@<o0bt)iih?v9>}IsoIJl3=U1xHkg#~FDDN3jzwnfU*9tykL(|IjMJIB+k2A9
zh}#=tBMAEmr6Y;N%khBKie`I-ZY!jU*lJZqUvmLp5dxNsjDegbU%h$F{dS9Y9`6GW
z&pSl$Y*sy8RpFL0Ch+6T(Vys7E1I^u*sY3Lik#W3RveB8&cjeBkulRWEpJ}k6~w|a
zGc1n7X~$;OV^ENb=5O%<Yc#6v$^oeaP1`X{BWW2KmOyoNg>BaeYQlK}nP_Z<G=>y2
zbyXFPy%wmhK*dDP7*j;*TIn*AMshx#ins-Mb+smj1@J7bA(@V<G2|o=vO)+!(=14x
zQOV;L$MxMkt1rGQ3*GYztIY;$G>5~1^W?BO5#xg1{hsIJ3o-@zn^%M^nM0<D{xTN$
zlZQ}<%qgLB0i>OeM-GQQQfapBierQz6zjX!Kqf28=IVyD44lRTT4n@^d0x;$mi<ul
z1YbfBrnZQjSXDh)8a^C<<o)mdfogM0yV|03&&ym;<h9zCrQdR59!~fSwqK)FCXODj
zi@?ac8Q~|yppf+yIfI+0BEL};LMv)N66aHiVTYM8dvx7_(Rl9(<C%(pG8UAkRf;%I
zsH#C4jjl2wLyy2~TewDb%ODI?TiiJF;p0#A+Jds>&A=Fz^FZ968RLxX8nUhMF5`WX
z>b!Y#3*8m<>K4*1*Xsr;{~y2nKI*nbR4pH#cN~X-x~}M|3bSOqFR8LAKy-zuDlg)m
ztRzV)f)EAaA_ORnPl-4!<eaIrK?==y9w39REK*9&<AgPaIW9C!_e-!Md3OmgL{c(^
zXLV5nUv01H|A+rOpFe#-)g3RN4-BUftu@CbVw}f*{T)rS!U@H>(o|JP8qbVjV3^PB
zjyqQC6>VECOR8zPdG(s+ZQ%360}n4pHb!$E2eP%?u5a0HHcY2yWQygP(_Nu#jcGc@
z-Hw0!PycVOw%2r8(bb0OxaWL0lBDAH?blp2Yi_Q~ckfq!`|t4Qw<sd`3Ear={J^K(
zk*dE!NXdtH?--6}`lknOwpVm*L%-?iwiRk!^E{leb;S?w-u>kdg}Uu&nuckbFj{gL
zP9!O)`W96;K&ERNs<vjg+cSGl*Z0&_gQ{wL&g3}r`ST~Fsj2D)hoaqF(bo!FcMP*D
z@w2Mt)A!#~p}GClD}+kCeEN}ELvI7yrY~<RSz>I%wl<h`i<d2FQDo;avZA`)k`>s-
zvhFu*@4rIW7Apfy)uNRIU9no9c=`AqjbeTKil`MrDXz3(nif7iJu)1RtW-wVn&rha
zE`gLJwr((0#o!&?YQqu&OH8Pk%2KweSxbd6rZCfY&(x+!<i}yaUuyhq+cBR;T3c~-
zbIq{KBq^}n3Ta!s^SE&&T8l9iCbT4%(QPTzjQ#|zAt}#v9I2~_scTeU5f@h=WU|Ee
z$Qd^;WNV1B<kUsVNFO<V_)K&I^?HNuHUufiIzXDKa>HU9^5zb#r41wb_{3BtYMHR!
zkq-xA4B!W3isUG1nzj%@PkWZr9-#pQ>vfMMp;IDHGx<Cq!k=^B+sy`9m6m9TfgA(-
z-L8DMtYy_TEI#2vL~BLc7FYK&1(cr{Pdn<SqwCiQV}ZnYo{Hc{f(VY&=TD5EpQ*35
zEXoj$XNWt(`G6FWaTpmUPxJgpcXNXXGig4fsvg;Hi-uAh$q1IX5at2pJYhZ|x)+SO
zELp1>)2_%;vm8Gc7^AIFRgE+aE-nPsVoZhik;V%AJQ1WONK5vSrs@mMIw#t;MkLP?
z140Q<72WoRaTpM$;%#5E-mdZULiT|$j;t%e(g?EmsHtKx7W?K7vEOqe1Dn1j3&Z*F
zOiY=#ch_Z1bNO^#!>@kxErSmj1TI9z{fWgTDxFc%<Chba(QMZ%j{AwOw<MA1+ZIB?
zdB@}(rfrJ<g3H@ej#zC7%ZXV^gj+yL+`N>YgUomrao*E44a;!mm;&{x0VHC*L83{1
zVhqn1yJt<L_L4W76(8M%X<9@qP(VLT_`ER2%$vXZE845C$ioA*7HF$juL_D<S&Kk1
zE|K?7d+xvcYyS3&Z+ZUyx5V=^*KclVZ(p;y`iglNI2}K6c=&+6cvh-PGL2_sPUtl8
z{P9OV`f|y+xq5?l9@{tArsnD8iCQTtS$a^i?Fcb~kW9moAT0Ijnw$GK$Q&Rz4lhSc
zBiZdAnYy0g>KdUd_RmN9>pQ^m=G9k!`9onDPPkfQnidxv&Gv@w<_$R(KvoKlVH|1e
zip_S5GM4L`JDwgM8BUMn6ib{;o=95rA_U{)+1QSS7Ikxt(K9l7_AgHy_j~TY{JOlI
z=!Sr$>T1U238pjC%S-Y4t~c1Gp?Ui?vcDk*&EzJcJ0Q<H!tNtmCDCqX{`3x2Rk%E(
z-+n{Ay<&B<VICw+*Kl}#A!~(9fqc114vQ=0n6^Wv$TBBlR%F{EZB5=>)7XmlzxzER
zo-tNqD^r;7%S?77jjrhG7L;NNjwyJAHXM#;a!#z*YlJc^aYX0Dd0tRei`EJ)6(Klc
zj$jqiYD7-tWDqykh^yDEtA=hnU;clOoSz?9mWg>7sK^+l2+NFLMw-4uDZ_l)gAiqz
z4LCQUgg_X9v>Rf|sOf+f84C#0VCtTf1n#)UR0gF>&b+T%Qi{wu<BpG)a*{<|tX)<Y
z!`Epz7LI-t{OY@3qiltbiX1&G0e3#IOv9gj@A|IC+8d@}gcvyPPaqUczecueO!Umh
zkBs{ntqnFToTn4x;f2#Mp=?iEUl9|8dBIqP&k<P{|At73{o#cWJhhOlHyc=$8|~@1
z=kt$$plf?ll6?5XJ8WIC-fXGbhRsb)IPNgIjDnQZL;<GhP;y3vg%Ag9rI3{(_%kX{
zq%blv$E5%Yg(dodd7KcsqOUZ~&6erBCn2et9^G{e$1;+$Idgn|AcU#hU}Zw508tX9
zrGNDWGmxi=^Z7_uZ_q|E&Lc@A@PXhXiijLfxQ9q@Dpso-&eIv$Rn%8kgmHj*UzkLf
zkJXyL{q`HKHygYZoQ4zQaK@cZXeEl(Y+jhoBU8v6$AxhU+<$e4?m8adedhV$P)3W}
z$o2M$d6>$qzpWXv$4@iM;t4r3%}Y5bW{0qfwyJRR$k^A^MseQ1V26(N)eSBvh9&aw
z{3BJn=KA^;WfY&E9=Tqv`Ra==FkMSdifKGCJ-_Gt@(~v#<KdnZJ|LC>f#Q0-Mh+8c
zoLQ_uv@MmesOgM6JkY+n#rK9eTDqnNH#6-&vc7sl+kMCNcEjiQzeQ%n>i#=6IU$VZ
zc-V7&bIpFYr>ia6n6lSc-4IQMGO_&5&Rb6B5p4v~jYy&C-~0{x`UUs+fUa9swPsmn
zo(>0$P&^-Ye0o0cMrfLT&GX@z$(;$llp+7(hr%f&_U}G&eRYG9irc$8Y~3J%({W$2
z(bnL-W62pK6z%m5ZOXWvAT1MN95JF`cirr`y}Dtwxuw2;LsSC)ROX7u{Y=wssc$wo
z3h}YtZm_K>$n4X=L}t#BuB&i$MVLmK^TPP_K!`Kly2on6DJ5)IW19}st}xbum#E>y
zVP9^oQYnH9;4_EA%SEIz)AcKI%t)bVl|iJ)!HsC6XoY5eIdiL9ro#y<;53{u$xw*~
z&LcXP!buhr?)1WPK7!InTcedmH#L4<I8PJp<{Ghv@o-?ZQbizh9ta^aEi*m@wwo;s
zBlGO&u5M_W2Jc34Sdd6!l7!R|^)=^Vf8pU2?6eGt7$f5_;r)rQOjLEvJdP-(h$(Oy
zMyxS(O-<cYNT~?YvVL_>wSI+Y8f;4B7||(`r$d1w&I3u5*MF@gpo&*__Lm)o!YWNL
z1s{^72BnL(Q7epu?RE>RRf#kH$wz8S9A+v7i}M)U&|KYNw%62618!n{b%T<WF_<=F
zz;<hz>WXOBxPFUH4u5=Lnr4&|RN9h4!YIYf_7)#Y`bwos`?s>xeNVT&VHifbzT=DA
z8_1E$*kZpjn)5g^FQr)?5omcCdsLd31GaAIR-4iXd6`JjGcOCRfYqum%p7YAwjy|P
zobiLBwvzjsEy~m+shDEsS%R^eX_=`uYpl`al#3dZfJP%kMx}tt8QT~zeewF{NYge*
zDXFC@Ma~Ra8paqo4$oKw+E@$<lOh<&=)#}(6k?z^61`dxb0)Se_U)GifTtBv2+sQ>
zesQD_*}uH7-|gwIH`soKS$R5VIE`nP(C~Sm(e;L>!vkM#wn$-+b<c7fdHCrwA3l6$
zegBGw(~+%$ZP$>4W3%nJ*{&f~ynFw`7ix_+iPhCBZkkurZO?kOrs-Eq)4=}m5w{<Z
z!_4*DugFN^F!1f)yyc(%<#+t2|JVP-)lG*+^6lUJHM!lwT62DQ#H5+$4-a%#*Hr5@
z5AT1(y*x7yXZpG!93~FK6H02DZ)T<#`1JV^krQ{fYlJ_cO@k7KzFV={tWl<7nvRSq
zAex$C9?&`R@bbW$yA8kl-ESFPpzUg`f!o{HtZ(0P{q}48aO8)7`9~f;KGRm3`};fU
zZi|^bo4)7j>Z**;`U;XD#lZh>Yv%ty4~64-;??z*ahRxW#bLLnNwJh2rGP9+S>QuP
zIC$PYv+8c>RvTvP&`po4%Tm`UNmDlj?>Ihw;KRE=P&GB3sToDWUe_>W4j(=iL)vzO
zNP%%IP0VVu<@J@u`H0P$`7q&`czpQ8a@dig<NEFuo9nxR7+!6t*K16_;dD5aLTWfL
zPb0djKpT*ng;W^zxy);0h`fIDmX(jp51)}jQPss}6@w?vbD<DQNxy1no0c#;4i7K5
z;L##dH8rMNp{*s7nWm9S7kkxedxdw7h@^e>E$;b=@tlZtxrt8k88<&scQs;89FI?k
zl<C%6WKIlE??Gy0)#9<tx+hj^^7@vb8l+VCro|mTqNfA$;@>q|V^xQ-7O51@Ibw{g
z*K3T?q-7-fjHr8>eoJ%n2AV69wNxUZazZR4&W#vjif5O60iP{JO_?cf<En<7shS2;
zH>8v?n=NRK50R82GAAl)i*8Y=Kl|R9sv(Pn$d0DcthQU!>INqaIXZOJAiN{ygsy7T
zWyu-CNJKOF%ygbHc_GgW)|Ta^$bc_07o}}^OOi02PlVuTwIZvAt1rH#UuJ|7T;1GY
zD}{)WG&#b&lwFTjfWplqLMDu85Vpc=MfMIkooT9?R2yV+v|1CKBZf;$S3u6?K*_0?
z`<7+q=>ib!`ZbITaXgb^Wb}ci?+^gvKpel;f(y*Uhzya=fL4+m7Q{Rwn-T*JF_1%1
ze`a0&j%IDp%Ao6xWgZa8QMaptxCk@z{;5!YoBq!*7XVQvR3ov%;Cw(Tjn5Gu63SR|
zhPtZo()0G~ulUt(euM2Ayo6V;e_e!4FAt<{!|LXi<2dr`5V1KhACCw>Qhm{q)QZnb
zCRSVg-m#Ahrm@V^LNEc-NqV!QYBlFE^6lTe;q%i0VJq&xxTV?jL}^e>ODzm`8F9;r
zY5zb|*H|st@AtgE-q5!V&$BCy;=_ov8d+C#s}?`P?d?6bt!TGBqHV!1<Vx`Hvg7PL
zNP!zi9zXs>v$^H<zyIG8l;+{(iSdWeB<E2gv)y*Zc_CqcI+By%uDxOzM%;Pe&8ova
zKl6|O@;$a+^X;#{D_YT55G_-Nw(9ueSKo4ev^;(MiOOg`K0aaF7NIoHyB+$XH;}n_
zyM_GozgK^eLm`a|z8<OS3S}i(XTma*fvRmu8Ro^a-=A1@Yrg*WEw=4(KPw0q&+Yb_
zEEE28Vmus}gD1p5Z#D6B<avH#zV2Y%V^yY_W~5L|hdovo8FeK!eppz-iE)|O+`Xcz
zYZTCJJL*@rI2VZ|l99Ca6+#Lgo(@!A($pQ%C64D4$K#Rnxg;IdSJzC7BQ2ioszVz^
z(^<5JT1rwY$dhB-9q~ZBS)poy4>POvn#DOj{P+{eMZ)3`D$;FMtk)Y>ufN1}Yj(S5
zR&zmoM^V6m#bqmadH*~1&(Ex_=KSyzsi}&DXS<=TYfgJOIR^sv`zP!;P$`M28)C`~
zIS_0aHwkU2n+l-}$bt+LQqt9BBn7aH=i=8bwv+NRl%fc7%!+9o3Cn`*J$c%rN!ZXK
zC~@9pI^yReh<M44mzjSEzC0Lhi?o&$J>zgL4{cM*j7e&QRA?=zv?Tp(5DT6h|NJ#q
zYDs^4M>xHZ#se~YEDj|FS}Ib^R8@09GgqV-$uZ!jffOY<o|qPg)<tY%Rbk|;w`=?~
zGEP%@$SFgnM2r<R^O9(<uBp2%Im{47Hdi<3zCw{0pZ4sZo@rKPq$i_5A*pO}63bNb
z?8!L}PoG%oiZGo>e!<ioRkJ3EBCz+){UTl0RgFRrP{iPvBgAQ9nhK6VNJCw>w9N{6
z(L8SM-eMH^5Qs=(2uSal-AKwFTh+A93Z)c2dPK@tB~ezRj6!FNk2N7=yiCl~$UL5D
z+7(^1`BQQ>LMl|#vbYJQE!VfVOv}Ra;RWX+oib&7Xbo4JYqYH{3?y3u4t+<ZH8<BC
z`F!U1_<`NSj#qEKX0y7*t1ZL%M|5TRzx`kSJN}RV{eM@G8X<Fxj)xB)P$bst7LfyC
zoM^NphJ=a^oo0UjPyb9BUbwz~h3Pja+tO&mZnr0<g&_#^Z3}kIhuxX0O;5X$T&=H}
z&u4ZYKeFw6RLX2OD`*-<zc3w+{P++5#Le9usxj1kj~oM=)rPCPTR#8vBO?O8eT7&e
zYbBA@iu?cb-;w8m<NNPf)e;dRr*Wc2@vz%-eEBalO-*Grva4vVrIL}4KmMLCUf(jf
z1&LsoJy&nO#B0ei&Ahz45K}H}t*T<4J>%lId;JD$N<lD7McdRoKRy2C4+SMD5H2kV
zB(64FM1pyqIh{t7wzU0<FV<Ve=*ewEy<KrU9$4M18TNagUyi8SFdYZ-;<@|cExsy6
z6)`)OhiB~e6<!*KWkxm)x@uUm;4m+^ai&Rv+G^@rQ8x;$Gn#;Rfz|aDuIgATi$5Ki
zrUTQoaN6$~#)+-V%=H2}ayZPyMG%AJ@c6>%`Gv|<%sG+V$o=((&PdYmQf8V~5`yOX
z_KvJHbN0j>ks-1F{76LLa^U8xH`HCr{JbMB11SfVX+UVh{mm8bR5&Y3%!Qt+6j=!B
zM)TD--%uIN>98wm#{^-X`TX#LU&?L1s_PQ@ZECcnh*wfTri?sJh{F>>Yow{rRa4rR
zINH8J$Ax(uNf#bdj0veERaGMdq)>9=Ng6PhKE|a>B9*3ZHZ*lZ%n9cvAft2vu`SC?
zSUl4_gHWt&%Q7!`H$o~3MK@0c5>{(SCBOZ%<UYiNw(0_L`{Tzg3-fqJrHoP*1t>p*
zU+{}3q(sy9NF|F4z&VEFkuXm*b<Op5LuIeX2<9X(RfSZNahf0$+=0*<qsw^h=4x9U
z!6MSOJ+f*EZo;3BcrB>gJF2QiT@|<YJTHV0$vIN5YD`_RxEU8c8zs?QgTL^8y3K|Z
z0wD`b(-d&vg@vPZ!AgV+8ba4SMnLon`{Rk=mr`h4U2z^J4v!C1RRu~SjV37z${=+`
zlBMm-JRp-mD@zoLX?DdQ-W0)=kC8B)E;|E32p*p^wp$@`Ldx<6{imlwNSX0mu-x<-
z=ORW`eEI4Nd`e^ztHvNM|1L`iAS5a*+^l<KRj~SFRQzv${}07iT(unCk?rfRF?G%M
zDKqc(#KXDp#`HuolDBW(^5v^*PWuDMvddh(u8}D+jU!4)zWe$u-+ujuahk|d0t=si
z`oQz!j?p{*>c9P48j)c<uw8e^S66s96PJPE=vdv|qx$dgqbF+3O9-rN8Q)#qe}NJW
z?;bw$>dhTW2;Ag&efJtSd4%lv?)5#*x&s?{cznmz`U>^*5!-gWKfdsv|NfuotD4%h
z2nk<(`!(lj=KCMM=X5;s@BaPY0hznEZ~43b_}??kGpp4Kr8Ubua(eiLY-;ce%P^8N
zj4rV1dZcpv%fJ37Y}N9`+b=Q7AlkMp%j@nhe<<`fH=O2)Uaz>ldWAJD-aC}7(L$DD
z#yr!uEv+a*68F5Px;Sd5$7i}XUl4T0D@niEVz*l&fqX79lBQ9(v=IElbRIY_o^}QO
z`Wk6FqznwxOk6zux<#bO>1mHP2IpZdD{#+6ubCB-wj9R;B*%4Ep`+ul8(3}D-2D32
zq!{`9!|za|LdSp*nO4<kAy|$F^7)07BJ1sz)%u1|YjnTHBZ$w>CB<L`+p0yzgwzIU
zEnZ~mn>+GcA#)&{pCQkv*p`Qy0Alg+YFiOgfSB<Cq7!94E+OT_A|<pPs%j9Tj2fzL
zRc6h8A-f5Y1FZ}sof$@llob%Ebh$N(;P7t7Eh8aBbX9mRQW{*m*uq3c%0jKoC`{b|
zCDj9^ytrRJ68avcbQu+BO9<trA3{J%P*O5X3+Zqy>7(Gvy;mFJ;yKR&og_+70B%}S
z{s=`B`XYb-<0oao;lm#|9Z$S{{f7SbhL{r5`NZk@nI!~n?_U#zBxNqx>B4(5RZZ2c
zsp^I>ja1V_n99P>yBU!p$NdXgfRY9!6}6BAw_xgu5M~;=Vi*sIdC$|kM}!EZ)f%Y`
zQe|RvgvBFFMXqXO)uQW`JRNXhDTiy_7TmMQxG)y8SiT^qg)F1CoT=&tp9<ynXJW31
zrHC{<J@L3Ky#DGNqJZQ*QIyWfGQ%P@Sy!}cgD{3vuL&t2UBac}EbvkzghpjmV0N*z
zf=yGSnwmCchH)wvm1>2^p#P1IvoMb&9)(`52%;3Ew6-V$ZXQ{DF)3DcU&fn85to7U
z(}}9<kee%%g!TQGJbij(csyc{9wiLhx38&GhiDD9PY@l!O+0;kPfUv2%~f%R1AVii
zt!hS}Fs3DW$Gi+^DXE%<s#~$%Y-uXXS6_a`h~Vn&mq_QBpMN4O6U4;z=9*!@r)}>r
z^#&nkY#LEHkgZ16D<1#wp7}~qU)^%oT|*8`HgWyc7kDGNdHn@9-!<T8DyvX!L)G@|
zKD^`j^uni4kJPncRhPGA+tir0W|>CR68QH14f^YE(AqLN&+Yw}T;0Dx98UZ%fB!%6
z-CzF|_ix|e4=?=w_y0<zMY-^`4Xf=nUDp%#C;s|x{|=E7@Bi>yDx+C_@eOs|(skRv
z{Gsrh|J{G&@$&<f6v&(iheu4c#cr<%O3<p}X&sl5a6WK++7&{hF|6tqH#z3h3ud)}
zn|qp@JF0%g^B?||G#tsxh@2)iO0Z;u&#>D)bNJ~!UwrosZ@&8##xy+r_=Fz@rb=*q
zbB$eZu)3Jr62Ley&l9-9f!8sEt#H09eY?8)|FHL8%aUd5l_vI0y@abRA|f-7L;<KG
zN_4k%^AY$Gd{I6Dubb=!yI3f7WM+og%H5ZmnitbEaS}zE0mcIwFWSxGA~PcP-m!f7
znsfeRjBj8>h3Sk;kq^58ZzR%L(2hU+;S=k!AY7*Cis=W;@saWFk*Y0ONW%4sl>&^Q
z2mv)=odp5s*Jp&%C}**5#Ar>pzHxni;`04U$cE$w7G=RWdL+a^4GZhr8>4NteCl|I
z=mz%3k@4`rmyh2upP$g(fIHk%Yed$Fi1G%@x)r0BmF?vP>pGlxM@WSd1Lw;N{cb=Q
zh3g%NMxYc@U~c^pr4(u{Z1aRvm2Mo{oL^*A&6L#U(>}(AZ14fCHQE|dF7(45Ap=s(
z)DjV0N7!;Z@l?nZd{~j+y&}6FP7mn5YqRUxMmX{xEDCYCl4C&c28OS`;yka+Z)da=
zM4^cya-Gjq1V;C0)1i$+R*6AUmMbVpih)`aHDzKFSnHT$YPbz87`hRuG-h|eXwyPx
zej|MOEp@vR6Zoz0{qO(4u-`Ww!Wc@<WHM^LAhwx03>3iol{8pFo{=bW+Q_MO6$qtT
zsk#(E`LPGN){N^cVQZ(tsvXi;+;CuIPdz^|ZyTjXV#wqS!a9z}BYxdDKRr>ljVc6g
ze*)E$t)<pNS1KxOEMX&v8)iCZ%B|Pg?GCNeL}{YmFjk`gQyMt$hnLWn?)X4>c_sOZ
zu}#}cMv_sWI%;v06u6u(?CuUoB)&+@_`sYJ@v@R)<oNDA!Z{A_S5m2rcLyqx^Ljyg
z&-(IA&XN7;z~OX{vxf8a4J8%ZJhNQRtiynQ_?pAxuK>yV`WYQ3mdgv;^?ZCebKMd$
zG(&{yTI-xGa6Vtr<A4}CI`s%)WqZ2v)w_F&b$s{j3nFGld*S^cP%hJ(%DgN@KNGHj
z-SHj6=_}BCoG{qq3F#`~^?Qz{(7oI9KmYzW45p1{zy7Pgp!iI<Zp2tvec|}Czv8Q3
z|0Pd<_qTla@Bi<(hkLZv{PsWo9k%b$uC*)r%as%p`>%e^^?Kp${4GZ!<*D#+*;C>|
zuQSE%INYDOeEOdA<umiFiQ7Me`{?cYm3QyI!a2=yerCR|xcSVso)Jp1yq#M&t*qE?
zAm_sQI<bFv;PCDpvO-dlEw^1`y}a`B`Fs4+C!B(~1+JGXyVE1@fA&kh{`If;<9|K0
zYE1z~8{XZ&!{$m_ugyszDvOVF{SjR%H5JToBx#HHE7}-zt-QXzktp0B_Xs7)wQzoW
zWk2-XKi)CTbAxax!@Di<`8RK{8!&_A@bHfAaL<%=L-dw{PvBGGGOfgQV!s>lMpJ>1
zny6E@1?xJ3b!^MR_4S$MawZmqJ3U|ycjRT`k|X=RV}Cf&iB_y0wwagLXC6O%O>ZP|
zK9iR-`1Y{9E?3It={k!uss+x3q4<b)%|W4!V@;V%>&DgAA(51r$*GmNuYSU4g|ZH(
zW&3cKjIavjEZW(2f|&y~dWx(>ksIf}aAVN4U1#*ID$+TJqNPoXRP?4fVrnTE*D(SX
z(~h(DrhC*tiJp{Oq5Pk|wrxfCdxqly?RpfJW%i_6ki)=O9mZ%<Dxht1z=entlFPi3
z=9$9}xaL~PH{W?Jl^jCr4z7*ee|I`^I2}-~YslwyWj?>M_&{l1Q!th_iJOa#W)~e4
z>-CLTE2$LH8mPITL}rZ_DiWs*R?xH%Er~HwY6D8v+&Tuf6p>-UC_{>ow1Kh~Yb@ja
zf&HmNrj6)5F;;{$WL1baUe@(|=JThIeDjN6Lf50UL2BEQzJ6^bHR(#Okt75ef$BP}
z(Ufgr&KX%7mV=xK!IR3$PwYoRw5GAXbz2t!F-58el|0mfyE|gv-BYuG3{Y@pXxNvU
zndgb%H$;`lRQUP(50u)fE7h%^WsQ;OELL}HIb(#OP>Horg<`w)M5kp#i!am)))>O{
zLd+4R8`}GDJkpH^KK|i5?0DenBhm`?yFEKmF+<-@Mx8(o68y%;?>^$H!Ym+F$6JQG
zuYZ9Z##RirMO0XsK7U4RkxIe%iA#}4>)2Vz^Z)aIp+9QO-HC_$N7Qvgmx>D#IgEI>
zBUebHryCvV{LJw<u)M!xe|Jwm9=QJQcYOJH#$P9fZ#3)qjgMcxW7wZ~|M(ttJR#G@
z(H)pmV4GL8v3&gWnZNn({~dB$*uK0`hmqs^umAjuf{>D7H$p9F+tKY$#OXrXJcD)g
z5>g7dejr<e+wC}+o_Ak=&2V=|PMMultRWyo<@NJ-T)uq5N{JeJv@$$CejugD<@|~e
zlHN8z8>Qlu;&Og#G5C~8>q4#><pxZ*<MR0n%iAkMZ&3;|6=O6xMO<fFLlp(94KWr@
zL*LjgaYNQjO%bCF-+cX!Q?K!tSLDz!XGqtD`>#JBjUlDTbh)&J#(R{Ix0$kNM^>xx
z-jl-0wr!ZMXSd&TbRFaUJ650h{^<!_1s~qOM<s~?l#=Yfde6|e(w{&6?hi~iAw1_o
z_Dd5f6_k*PmkCh@<Z(wuW38pv4uM1)gK@68eWYf3yEzvY-?;43V)q&$GP1&QePNpw
zw!h84b8U0lFk|~3I?I+@08ZBC)<HD4k&qHC8`r<)w);)DZeVL#3s^3NSSr@_*nYr1
ze1jQwBq7`Q#IJ1EH{x{uk+uKBYv?+tdv=E-MhJB0h=)UKE0YOAqy|qGl~@x-I&9bB
zq(%=z6TT_PQov~R&4saT8(In;A0CLe>l>Dd(rJ`vBRHc4eebAcAcW9*22{g`m#Pp|
zP>X075i@X|t{AK7_8n{7h@w!Eq~^*RH;NGK_O9I@@rQau@nc<P2%Z?W=G+O9)b$wa
z==y;w4V6|wS%jvj7MvtaT&`DY>KKQf=r^``;?(WwM$Ix!#9XL3wvAKCt;3)sicm-a
zB^64HDAh(;Z8#&4#-OM_5t=0xqco?x2SNzV?yQ=fHqA4&7MwAFM&26Tq!O*&jf%4!
zDF*zufs%;Mu%$pqk;Ca8X$)~{9VZ-5Bqb@^hLMK->Bzdw)KYQ%4rA=C^at`bGfe@N
z1LL?O7Rk1(<d|7~Vcr7!ufIn3mg8}U+&t4Xv7E2;V_}(QYS^eHp_F2)1z!q_U-<B=
zU*g7rurB0Oc=zEQZ#5H#Jw{3J6X7i|hs<fXBT-l{ulSb@@!bo%Z~hYAk6fOgaLd9t
zIEt<eXkw{czxxfY+vCg1<2Ub6MN(=;3OMaYYK+8rVhxe*c;YYq5B~-sBwLu7UuVp=
zVNznedr$6rzWQK^RiTCB!#9t-OfyfHe+Hr14UToX5|)+a?Tr*Rx>L{bZe+e*=}rf7
z?s3kb-N?iJJB%GLT}M^29hpPsZJsD8V6EZq{sc9XN~IeH;ue_RE~G6{OXG)2ZBP}a
z=TGb&?s>aB6Q?Wh?vGg0c2w)KQcJ|>%zo4y#sR;sOxFv3YYrw;Aucn93OOTEV!d9e
z+lnz7Q3X;4ifXgp<HI`uV#!n`+2#w|w&DFk4l56b9pk$@kQN_f>#3Cz9|CLIutGA7
zJ=(aIBJMi0F$@{(;Rpy0<3L&_YS<XW#;G59n=efBiqsZSz*@~VUmGI2-_iAsQWDCw
zIk0v;>%3BP#41UyjsAFkIX92|@x)=j<7rt~{n{1<?{S9_tyJq?&K2)hVpWK)M^r&Z
zvMd3UDt)J0qg#w^_bD@?G|zep6_LOgOD>s`3tAgWPA#KRD<!x8FO6)cC?wK$7@=Em
z4Z&y~vG0-p!2a0vJ=z{{TCu&o5?9Z7yyLh#5w?k|e_>jdHh-6u>=$xTbo)E{-JTRT
z()AUs1Q?Adl9)2N&{$DMqX@iCXI?MQ#OoJ!yFFWit{-uo!ztQ$QEF-hEn6pX4Vmvh
zf1;Gi-PjXS>-f;lVUh=tD7GWS3N_)AM{OQo0^(K_lXE-8v|fZ(K^bauB9$CjXHQrc
z#(qFZNf8pM4Z3qwV~{gw1+^rsQQV!5#BC!y|B?8rd3t-pc0I#xcWX#%g)hcAq_qg$
z9;~_>(6<p{%(1<12!v`2#1F6JJ#U}BN7}w=f7^z0j$s(-PWRLpv9Z89Q6ywpQNsbP
zHB!{}V2ufL`(C@QqtmTkQ&tc;W0Y#T+%O{h0aZcAOvn+}^+*bRZ`sxjV=M(hl?rV&
zMhs*WYby9@!3=`;AHLz?;T^mEfi-%fUs&HR%<~4KU{zqyhU+w=ZG%waIvyV1qjZPx
zo|kX`h|ZbI^-R%<et+Qb%U=^yAjLo@45Jb(+sv#4!!Le`%8`meJv<OaWTz_r?MgTF
z*j>lR|MELz4xGPy;_>l4b680XI1Iul{`z13HJ`u#$i5pmold+wJ@fHp<8S}3|2OuB
zkwIs4NW6Uisg3$#rB{aiaX`z;yZ0aX!*}0N`#Z+d{hxnP_``qxJC5Ul^UEvW{_gjj
z4m&>l@+<1FyRG;Rd?l@sw4E8=bqHsgh?X*xf)H>Rd$zSubS2<euGi*7Ntu*dxp$5p
zal`1wx}l{eZIMz7ofh=_5oru=7zw3N^U7f`=>Bx;hRx)-F`eItIk$qI<B?nio8J&p
zG(zLhlX9e_h!&EvR+KepC0jSMj|d@{&u9E)LP|;e;e#*R!zGr5R4PgtQf!t?g+OSF
zHnxpUVk^Q)K_U7*_Ye2@rP=;eO)ZcVJmK>fUS7_4ALzQFjW~Rv#*J=&&-nNi$|z!7
zDYhr)f-xOxkcbp1HMNJ;&HG<t;QaXu)^$j!=nqFKt(}S7P}Y~%SGL8suIj<z#(QKp
zAeE$FR^oicU$2O5MJa(_7S^d<3n2ta2#Qv?zQNMG-cYLDTFcss=cQCwr+@4MPEViO
zH8LHqpFi{Z-R~eQ4DaqpDg5NVGp@(TBiS!pUZ1(XUisSfsBt8PgmFFl$5!|wgkZj2
z$u|sya*px1W6mozco3DU1X}lut|zQB^RkjcBY~Eb`R<P&N$HQgoiF_S7r*4)hY!RG
z!ZzVhDMiV-jdxP(vFKb!IDNzJ@qOD_dykqYQd|gWWAlMS-{DL})(BGJQ$|X4({U=+
zT83eujsvBZA4_g(F0HS`ZwLX))=&-ZcqIA-nIcAOjMBsqnn-R9TZ#B}!55Fw4q>5W
zLD!v7#voL4dP%8~QZ|lPDM%@?x~JS=j=B7lMOA5211Tlre$VcBq*Q_HMvRffWr7e|
zioQgus3^e=bXTYagmTm($n%-E=ZVf5cDo&>O0@0~<G}ONnVKpcs@d>_q%#g55+#QQ
z&CH%{xq=;#{hsk~LX=F2nY+%B_dU9EjN_i|a%H)^HMG4HWTQCT9l6d|))0xoV-7>B
z$4!yWcF0`md&Bwd6{%pFCvvR_1SwbaZr4JWDI%7Q5*K9Gv8<k*F?jD;Lx6cfZY#;J
z<jwQ(H@^b~!+v0T-53r#UZ*o?&92)MVj$H@E)6SRQ$k$c(4A&*mB0PF|G>ZfcmI~J
ze*QI-BWuWf`uH7ME6Vweyq)<HJ;@G;%Y=^+`42u<f0jidozML4+i&^)<0q6-eDn1q
zV=0t*LUbM5Hq#+l-`>bVa`%g0P?aEsgp-P}EW~Xh=fv{%N=ykMfs|R#7ry(=AL(~{
z+|Y42?ivK_bVrDh*FS#Z{OMb|acH5G!J(`sOAtb!h^*%;vJ^t{xZbwX`o71N2E|*$
z*18;w!Z?F8Jt}mBZAQt`X6Gd#X&?GEZ3N#&4@$HTLZ}68G+n<Vgvj}N#faFPC8k3l
z7<3bqiZKl#kt3Acj-^tt#ek%eQ=sP9eip4srpGTEof$Dz5yD3Bt(azT2B|xCLZWPs
zR1#}DwrOU$H0PGJ5^F6~%iEXF#1P4O<I9&9mUZR+;T<1-^$W(`K-^ZM-^ejwT*o{I
zd=~U>kLZpF)#kDN&b1C-Hl#8P#<4C7({lYW)hiHfjFK}-nN|QF0;Oast<P7|STi|C
zYD|cld3z}+J8*ve#Fvl1r)(3>U-{L)TYqxj`IK8axCpfAxPR<$!ycc2YO&*pmSD6(
zSMc+Sk%|5$o|~>iYTdd#1YmlM8(NjDYCM@?)b!So)|FBX*J(oAgt3m7*E2;L-v9g^
z-R`bgU)KfY3T2rY_J`K*T{2zRNXese02Z9h)S^fMj414^LL-<~PY{lN+>>HtUS?7*
zC@q^yD<?MJK#lvMF}gxYkP;<%(t2~{)Jk`2F$~^gjAp)GnWqWkc69xLVK`9)bY{Tp
zJFIc72YhSS+2}T|OT_l@Ep3F9V}hKJ;-`y(b{)IN_YG|WR3SiF=z6kWky=sgK-7};
zI%7&3cL^<<yQ}M2)|KKn5E*^w@Y}-XG~?Z&K`_IF9S6czxx8G6e&IMYm*WqnhwXaB
z)PgaRHF=CWv{8}}SXv#a*gc|+Caw!%oiR#bg=UH?!@g(nlE>qY5CSzs{5+F<pdUK4
z)@}1BDyDOUG;kXGhMv$p<z_^PrDbwdfy#;~p02lS+9Ht*A>;iOHTJl7N5phNuO7L1
zN=!K0gH~uVgB=NLLUe|l1^qa3Jl!E_&OmJp!k_s+{_CG}JedZ`Lt=#G{oxL`8%cFT
zrYmyY5V3-Vuy}s+@Bg<y|Ds?!%dfuv7ua3TaW`-rdjf*7Z`8&TA|YhHe0kyB&woL_
zEX@9yd74<SGut##Q(y~`_m2<lL0v23<5%20eBjIf{J(NI-BGk46-hrF=^x$^{l<U#
zkH6=4-+yLrc07Ie%;`9A{BY#!pFOZY9Z~LtQif%EC6c+U6UW0HM)gQ1*}Nx)h*6a)
zZ;qibqqHL>NiGp1E7~};vH0r~#RsnEnOx!i?npN37P3;jTwaN(+(KiO)7?EnHinNB
z5-kdQT`)QlWhF<0(uN`nnoQUVHPdd05=rYsnir%p*kPa;LtjT~4qQ8n-#q8@3*k5a
zp2v6Zz&iGK?<uZu313iTQkI|^*GTG>G`%8>Vf^qF^Y`EJ`#=1aB;arU-Ty$A3IFy?
zx8HGi_l{y6K7qLL=4H(Uzfq04*%A$+D3aFbaYtVTQV8VGBdtMJNvy5xCEl_JA%y1W
zk&;9OQ4uK;<^>-&>g5eRj!3Jx`{o@b9Dh>g^22ML7UK1}?V8QN;c%ck9GmsAtdw;j
zrbKlktFSP9K;Mtpci$kiCKZo+{4197J@NVl6DP9w#IVCu5UMTERx9+~1LNTiH_aSt
z1M_v(v|bC>V|7P~3p;BFG2&~&$wIt7asK=*=kE;5`GOSi{?~tvdH6ZfjEwz-a{Wl1
z7Gln*p<_25kzG&O5_<C3QpsT=q$?7Mj!>5iQ<!m`!D>s4l`L8oq1K9y0kKU~X~;2C
zDyVKiNQKZao(@>o5#tXh$O5iw+6W3+^Noa?sY)SAfl^6GA_if5LSBEOGgC?%$%R0k
zR}_M*B=zKR#vzKJri#dw&JG|YQfZuRIS8rSSXC&nr(d@6rFF)NiYbX0H?sFAG_Esr
zu@IFZ=O$w3T2ZQ^gg{wCMWJFQ`2~bxcRZq9N68Vf&EzFg{f0;rF;#kZpi`2P0xCy@
zfT0_xwiOFGv!hB$w>wf1x3Q<g+L5PEAF0uk>Y2Pw)Xk$m|B*Ss5<GWzN5Zz@#y#Es
zj>FwO)foKKN94M2I_`P*^#{Iu{>WM*VT%YU5mm6ig%RzcXN!S3M6Nz_{1^X{oDzMQ
zdHL;srfSEp{`KErcPCz7Ua&%9t7HoooREmEqPiY&-=Xz^Z-4hkPRBd;$H#w$Md7gL
z^x*^84v`~W=a6xQnvqpfrKHNr>-mi@FK_(u4}Tz0T7}|uLIDrQBW4&_Lg4st<nhA?
zy8HLUE%Eq^U!unW=?q5e1|hql3x55t{s!+ezx%uY#K#zsuL|{>!2k4b{)+$Uzx&q^
z&xF;Z`yNe31wm&F!3XBoE7LUNdc*FxBNa%CCjrB-C$2Nom(T2WJ>Bs{jv2oN`pz*N
zntf1dNh%3z4AhEs9mcg{^Oyo!nMOe^k<D*Nz=uHDnuX8((0^MKX4to#mDGfF#$V5b
zt<@A}*@U<%Dn19g!C}~gv{dgA+Tv3og&84U5Qr8ETO+~<G=i!$W^nA}9&HVy6c`1|
z@jyQe#IUf;6QshyHtdA87HKrpLP>!r6PsU9Ss+U!Ojq)9X0KW&uhjac6%=I5#N<&*
zl4Zdd`(w|tkl?okYb@GGvI4v7Ahy6_i3#rmNQoVLwCWlb<4<1md?kxL+BmvlkL^bM
zws3iVLQ?R4#X3t&ks?4?g*on!yZ(k~?vO*r<1ljlXj#wCE#X}SN;EB`a}9Hma=~cB
zoj%^AR)Lh2u&w0l6D6;dTtQomwHWJIQ=(c&r+SR-P;Q408>=!@>#(Ln$jIa~d0N<Y
zUE7Jz6G}>IPDl;wwo*i9UKT_tSfhwJk$fb?!s)2neWwiC7?StQTSUf|Sk|_~r${c9
z<LRE5Bvtk(-9hz~ur;kH6|$d*+uR_PQh-p10JTK4(G;aA%7GavdEieqg;JpuiE$%T
z1-12X*kOm35~(vWHeQl(4n#8=lu{5P5toS^W@L_t(xa8Yii8l7jG=UUgcFq8;!sO!
zo_wQ`W^8tEnc)_SOc^ARR3+M4q%u?iDHlXs$uUq;Wc8t;>!%q~Kv0oV;pdf@&vgBO
zQVN7*3xVzPXS#mi?wuwJg&p^t-c=NZ6c$2=NU8DL%GaH1qveu#d0j|i;PdA*ug_;>
zT(D(jiHY-T>jE)G<LnMZ##+O+E>v2$a7%O3AWLR?ec`KdM~;#8*1^(6!TIwS_S#`b
z$HV;{wZP_|8Tt;fESMzm=L@FO{PnMX!Sl;AZ|CPf|Dw>_k--e~k3HKoVWglcNs5^&
zB-7^UyPn6N|B~~^Pwd8lem}B>fL~UQ`#sh=LIodvYcz_HK{PS<_@UWWmC?xCu}$O+
zJM8(3zxXvGM@9?NJhOy|t?dJRetqNka8Gv}@wpJL6N4QfXP#c3STAeSDO4g)8DlId
zCNh<mX`z;ibQ)uAYe%aRvp+H%b__xhLW1DY#nX3=6w^&>FbJvXt;KbX?<1lHkx9u@
zOCd~&VXvu9l2XJdMP38LX{0|Mxzy5HyKZihQW8&Jz6Zgr9H!0Tjf1Z1o0|hc^npwK
z(rkCP51+Cv%$9>E#+6tjwSd+g$NPKUy?>9QFfSL@ZRI%Z5lWL&Zu4GQP$eU3B&9;i
zD>1ZoygXe{IS|$jl%h&Ou7x-UWPw@&u5%z2a@;pTv{pj)Z6~>{tvF_IBq=%S1M4(X
zazkUpS|~yg5p5;;)BCP-_$WYX`rVOSGxPbC<X22LB9tTF96Zh$geb`Ag3uji+(C}y
znA!ur8&T^*$*t7BB#)65Cl!T^Rt@7IY9+>-JywFO;Qc}g{`T;axZR%Juw%bJkW(gC
z2~x3c3;BA%I7f+roF+<`sJYQ^*L7v62UJ<f(KE$B$*pNGZl1Jk2w9+Nd``sd5b8uQ
z1NLx02!TuyLL>{(^Z+3#0&LeYx(?TM1izB%$|^hT{d=VDC@~V>K2x`e6lMw)J?s(E
zpsFO5Oo^$P3WT86j{FZ!Z?z=SGLeLYo8YSKfa~@saWfUDmVc<)G%>vfpfvDeYTe9Q
zw2@;eO*2DEvbCtbXXtmN^-8KQEs<><h}*;2bx4}7P(*X>`L_$PRHV}680nltN{GQz
zbGo_Uc1_a~l9D4SWjbxqvXFDYDuL*`rVm;}O1Ys!)`l(@qHTz%E!8bu&u}=g3o|MQ
z=4B>V!T#MnX1wD*Cy1W;<r8W09LEz?Xx?7WoG%ylrxSg@XD1}f8nFGIL?PzH`Q?oe
z19$hD>(giE%Y__fv}l(9EHZ=0EY~w)x-giY&N@;pe4fti?sq)=tG{HKJ^t-Vyj}^f
zZ-4$pVgGoC+qYQ0bf#G`w}s2oGef^)Kb&x`qwfxUczmEAda9Ice!~Zk)}}?+cSrv2
zKm2>*(=+<@jGrRi{(#fEG5=#g8cp!w_MwVcW7#|S%U{07`;B4O+&i)+^k`YEC6$U+
zjlh|NBBqS(2h44yC3}<*$XqEtlh;7hn(@N}N*^E<{MJN3V-@Gigz84-b;U~1R8ox8
zT-vE1wG^t9ig{V^+lCT?UGF#^_iW2VSZCH{MUN+hlFZi&IcApSN_RbwLV&OVnWCEL
zxXcq)TY5L3s-cz@?>#Z4HX|QKqyz-?yCcrFhmVh*7!%oh&YwTC`i;ZgBi(3_x>bDc
zPA61IXtzM{O!LgJ+mT{H=!~QxK32aWl^|!&GM&lmOeZC^RFrNX4ufWz&ct<N*mXCz
zOliqxZBXha%G3H9VvuwiRDUDsZW~63mTk%Hx|iCL(|`IBQWB%5RzU~>BWdZ`VIUMi
z5t6R!F<Q3=!M5=B@=QNZ98Y(ou#s~i3eh-CBuZ&!fatFjO7rZONI?)$Bs4-AxXsIj
zbr@~wToa3@=}HLA6Yl7cMSua}4881F-k!Mp{<oO!iBwwe!C*A&xM!Ou`r1a3@p7i`
z9c#_xToDyyj>H(aUL$LD$nHqz?lBKvA@@CbnMm6+xqvX{CQ1fO6?B6E1b*A-ttDka
zm}hcT=)R%D{d}ePiLhRYB_p?hv<;n7@`lK%mG0y>lzXMNBgBvIN8Dzfe)}I#wnJFU
zwm;I1N5tNPkYpi|wnrI5%57vNOKe3#RZ!6sdqi1Fs)-yHYFrUDP~!!W3MzU^SgAQs
zq@+kgkrq=tB5!QVL|o38TLIPBJC=1tq!l4DsM122rlHJ*QmybuHK&s4k3ZVLrHLrN
z-9nQI1UT#HY|D>?>5Q}%rESyryeDoeUEfjXLYS`%&a&@&(v~1@q?bQ%I-RhiVW%E=
zmd(by%nQR{S-nDP$#~kKjYDO)gp9I1cc&A>x=@H*-p&XC`~5&E&0cSiY(=oWO_Z&n
zfIdBaL`%h*JgzGITVN3rxg`3pKJfT(hxq*Q&%Y?t^QGY(N+qRRlhuBp-|Z<CI%^qv
z$Fwd;C5S0vonv<s$_hXggpk<nPpsz`0H)V7QdnNDZ<H931>DeK#sNxUef}dVM`SHG
zhsnTjya(gZ8fske+eQ_Iest8`t|63bX4iFW!6W6us2x@+)=-fBffOskI=bCJ%^s~Y
zVN1x);r+_%(=&JPHO6-MxG=w+`GLxFcYlvoqItb*LpA8#9zhdRLrfGQkV0|2PDJ~P
zm8KP&7(<Mn?Q+I;4l{H}qgd9JkRszSpzVO{jsz5!w@=g*(Mr;dJ9eiNtM{Z3!46dG
z2q6+P$bN_D59pUSuAjb;W99HraDB(N6x^_9y!(LeH<qVwiR;Yy?TvNYP}+bxqfHZ;
zL6TEIQjpfN-|gBs07zRv3PB2?J+w<Jdlyo&tQ)z4wGK+f=SIa%IrH}XitQake?Zhk
zNX_~vYC%>_i5uk~xFimDj|{sbr7EOu*IL>m>!iT1XlK#ZwV+i9IA=*ta^9{ipRUC9
ziggaxji_4imp4kvNF|xxo}k2r9f=X7M(s~@k6&Y~CEr3&DP;r+Squ1pObI14DQ?Kx
ztgSLtj4|xTfim>`*WdnEq|of|@8}1E)*alvNAK=X+OS?eGrvvr-V-PkW!Uw5Hs_F4
zqeMjbieD#|>oai~=>|=$ZM<2F#2xO5=VxjK=Uh9*#5RhFXmZ$?&o7AjNBX{}0BdT^
zWT`aX5wT?^^8D7`_gp9iV$2|*<oPFUcqvBCub;8~NXdoE^JjK<kL3FqAq~<FEln+I
z>vqnO8aL`~v?aBs+=6pj8R&W{6;TsVNpT^)c}i`<ySQ0%Rjrh?kfXpaGpg3shhZ%J
zxT70KK-_u*nj26Efs}1`FJ%+Oo7?Jk!m6d=W>RFjUJ%CO#vM{f1i3jA!-}W{r8QPd
zw6OF_pp9-nw^D2|qeQ`J4LLHsotgbc9!K1;W52&6ra)IKViWx8SHA{r@l|p<-Esfn
z8+^$0eUDLsb-A#-J`<LOp?8?hu*?DL1}vK1bW|W2!*JZe@x*XE;O95evXWyn3+Ni?
z&4>?~8va|o?*HpWp_GbHf=WeNg~C#)IBS|s@$y2PWb+g&)kuxj-c<dCaU5GYwx8J#
zJ5-50fBC}e%lEwfFMr4HKYfQ$n)jzW-v8`t{>8ujHyrLxtkX;mE7qA?_w%0JhrdAe
z9rN24w)2JM`Gt@&<z2yzd)&BVnXl9;T18)uq<O-Zf<8UcKfWWo4yPRX<-&Kr`2#s`
z>>eLcl$OY?kk)|fdUBLF+aYVafojc2p;+b%%8eY4M_T+LAf+V4$TUxwlo`i8{ceX}
zwr20shTU$D78(R3Q4qQ^AZ`p9gW2sdckdb+{QB0M6TL%sBUN{3X;HRNqy^;&RpMi!
zKkd-t3H$s?{LlY|%gYy*^^Jbm@%Z8A=x!uSORk<$3RY{**O|T_(AKc}jhs`PQ44{x
zmj3>p5(B$o*D5^`_^DZ1Q;4A2PQBmnan^CYT<|_1ghXpeN{MvKaHuM<!r-JN37FQo
zxsqaLofrJJ^4I^MWT)!}w$kzelthlXeRzi*F-N3oV-Kk`+JF!W=LRT(ye$;(+leJ-
zzC6FMo-ge8N5=h*m}x!1DHAUXS{nAlBT4rZAxSwgIEhmN=>z`qnU|-wOYiR<P{xvU
zYNLoGSys;kP%`%qcdbCjw#L1+YRu^Y(P>hyba%O7uiu`j&LFhFhk(6%k5&>hFU;S4
zk3T=LH=t+BCMJZ`bX`YA;8hzd#u!`Wst_ox$tB~{8!@h6bF;wemgW{UqO~I{1GV)l
z6FoU@tUi;Az&cANw}4&Vep0eiYbB)I3V3qs|GpxSMl<v~oYLgDA(g}^O-=#7tw<qS
z&$56VR$^M3NkMBcj;gv=(HR3ZCTa$xf}AOSMpIc+L1=?en!X#+eb1Vk#Z@Uu&KWJ+
z2oT*yNk%smz63&U_E=FW+p<8-7*Pnz#Ov!bhsO_$N+ARYS^~b7O06yID78i#2cl5`
zyb|n=4}^7roN=O%iemrGHw|GTAZmr%jmUXL0-fveA)$1~;dDYeM*`e9l1f1WsT9`B
z%=~tx9|qQS#oC_rx=_A4()Byid7|V>4uyA*@0h+kljjMoH2d8h%eFPmBnDE+3}gT2
zUldM%`ETj_4(&H;ZRzBYSJFCBmS&d=F|#ZSC9jAosq!5~DXLZotw@rlQ~@ERmQWR$
zls71vh~n$N{#X3X|M=hY@Zmi*wu>jVM2f)W_y2|U=~D~B<-+yc9+)|7bi=?rO_Z7`
zVMdgSMzCD2tWQsDmx*8uAKw28=N@srLu-LlmN9$c+>*o03vt^RcRky7Vg9tTyL-Tl
zC$vZ$okqEV%lSgsLc?TJ7;o4OL_umeo<@x9(6WOp!FDXWOiq!Q6uQ?W@2M#=U9X6$
z(A_{Rh3oao&g|NJ-AH<~r+2M~H^oRskW*#1H<;ex!oW72**<-XQXeUAPwa-Ck_|!u
zs$+QfHCbuOawT702=k1t5#=26_<+ud-fwKlQzRfQT|ZE&pgZ2vDTT;^kOFGxsUd<=
zbVG;fdxX*CTMnSs_ABR1j)|%a-SI?0Fo(ppY{YdVES_~<s6w?8B%xc!^`E?CSxFw!
zmWiRDjbbAr71U6uQX&*oA#q*T+)sX^rUHtVmamn|(<@&-J)^Y5<$|sfasM@H7*MrR
zKYoiV8EqtGexoSOw#*P5{mput+(_*r#Rs~6k0=6LK%_(nE6Xyqeg<P1AKsINM%}`C
zF&E_R`eZFg`Gi-OR_1ct;f4Y06RARPJ8G#MP6tM7>5ZVSXQ&0Se?%NlsJI~CK2l17
zl1X8uQ0e-S{r(-x<w6RDVcZd7hMEyElB;65zLK^JLeLmkN>j~<wen^lh~zbpL!y+Q
zGIMGv#AQM#g|iMRTL5uAU$J&*IgJ!xt4L!Jr66whS!Fv?Db$>Ze!-t3IR~^+n0}z!
z-GLej+l9C+g!PJZmTnx`q$S1`D+Ib5NJ3J4LgfUq!4Q!eaiblGq!B_wO|4%75V=s4
zB5pI$-w3@%<90hf?DurTcnfDLYONUe!2Ur&F8FPw)JQIY;c%cT)%3F1aPKADvN5{x
zqf+$Etw^9Q3(j`*`yG_PX>T#zKq!Ii16v5JrHyz~Nrd@=5T0ND<<DBcP&s_oWV0iR
zkvIj`%azmUSk?vq{U7m9pIKt&{_C%q!$b}nqcMPon)s=)_|LQ`h~$w`;7H_KijQq$
zyPjFD7xwpenB&Ow`ih(jB3FXn*grnvwPVSGY#gP6(;CyQ)D-#I*z@>J;g@UXyKg@-
zh(LMz$R-ry!#A8B2O=BO%NI6ZkyY{b>3hCBJu?hF?;q}2=i3haa3Zb?U!FdbqUUsX
z0<BPoJ8CTK4hQ<arzTH*ok=O7wcy>*vygc^zfhIo^x=rE8M98zOY^-86hbJJHiXcG
zs&Ticrsk4JZlvoRHLi@Ep&J#Txtv#CKfj^{SZi2gq~uK4JRxi-2`MLvs&uw*N#SYX
z^7(tj?ewu529yPIq6TrRHqC5rmF28R!Q+=J(fjsr+z&kd>?=x&t@pOuBWtDJMAr~K
z(rSv3h$zs@(!xzfptgc46{`(aDoUxy6j*0ZEEQoJPbY?gbOTwpop6ky{kv{qE^T(C
z>{*tTt{0S&QOa_iW-ilP+Y#obIk>K4*zL)3hv|OmK9a&=hX<;Sw}N(vYiOj*G$D*^
zgi|Hj4V@EZz2cWZe>`FA!0Xc+-~RC<G0eE3<MlO?N}|8N$4a0G?0%2Fxo@Os`b7zW
zkXtotXB5_G%=>$4X$3;jZx~roG5qNKNjVd0zzC1Bf;b!z#!^$pS4o&Q1QO#aRY=Tu
zpnLcV-SyOE!+-t*fs8YX`{RjdXF9EM&HxGG6}1VX-$=`XG!|<$u5$=waY|v7MAFEp
z^``UW*hJS{D)V%O>@ixSjlo&hg!x<EVY-H94NbAB>_2%!B%&XXwBRDOqHDqX_7Jy5
zBS@6B5`dB-AqBK;EsZ`GmibDFEvK{07nFeA;fNF(JqU7zttf=u<5R?jLP^)Q;gc;4
zS!5e!*IXN3JGTr`HK2qflmKgDi(AVB**3NkMH9kCj1eUzg4_FQT}!PaT4#jP7}KyO
z)Pj_i=qoXL!g6h6Pa&wKAYw$;N=%U$X!=Zw<Qze?o)4icAwJ&)0070ANkl<ZMM=`)
z5y7{9n4Af|O$TDwFvcKDM9WI21R{Er$i!u$BtfqoMMjh|bay*c$-KS1{utyFO7r&o
z#BP7!`ROz1xTn(%FCxWHNdbSBMIlaS*7eMMJ+n?%N=od89lg<9OCjdU`}=pe<0J9>
z%yvGr-;a#<k91hPFW^iIIOWvZ+dd2Cw=axNasTceeXr0Zaryimr-w(9KU3bGsOyC^
zy;9;t-*+f!>3;s<7TE{ecSHp1a>b-fIKR>LhA^KQKYU<(xFeLvx_PFTFPPE87TN{V
z5~T&b1baNtb$xRV`HkJ~z_342fKnxCnRtGFAr=^h5!*SWzN6Y5sz1<MMLB;#`7_}q
zkdy|gvArg*p1356s7P&YA2!kc4N~H)YbEa5VqM2}y--pH(>|=e6pWH+MH9|bWL`a&
zY2$Lekdo)H+o6=;a5~bBJ=F*N<<iCk(;G?UmIVkiH3gLG=z9w(u}vGuO6M#^2>j-6
zT(rpBbb;J73^j}x+p%pPl%mLntKOEGb!~N?c5jhdkVA7eE$hl<S*QqJo}PKVys;k!
z?v4js-$o&C=LS}c{->HksEVvzvk<0CRT_I3sZvwYkt!8g3xud_xsAnQ6|COVm(uX+
z=Qmzop6LhMc6qhUm)CXSI!}aU<uEwL-s0B<v3QhLj7DLM0U8@9<%sGBiWDTjk;6vv
z%|a@iWs9vlKuXQq^-74ZsQ#71;XqA=%VlO;SDbYWcYCbJbX`x^8FW7|uaO!(ua_50
zXBdYbtrg11#vBrv>aPU9kyAnELdIiUOX-%H3EM;|iLl+Yg3v6qeLqlgAcTPJJ7iS|
zxKUeWYfq$=$qP}?QjkPxC%r#;KZI_@@x^aY615ckvS4f``jxmE$~F<Um9qJEJ&oaT
z{{T6#g@nHa=b~Sk-(FCqAVNS&MMp(;cXXzwltNyv<k}GNQV5hu$ZFfqS}R4gKxl~x
ztu~}-PQP`!w)Z6kN?a+<V22(d3d7Jt=a7ZQvGTs<m!v?*2EG)6+ve9}O#@HX2&>=N
z<{55mH7#37mU9lJ3?XHdtgVQwT2ij9uR#d3F}EOVW}fCYW)A@;1x4hxS$38n6RPi-
z*M;miYKVwbv0b9n4XGVMwgu~SzbD6~K|9Y2$K8PxBm1$#ILS8q7LNT<5&bhR3V;0l
zf1%f{C?%{506sS$Tqy;a3u$g`Or_uB(CqH-iBd8zfuSF%H1O2(moJF4fe$F5h+9Mr
ziC_Kt*L2QseQgErP6@6hvQB55(Y$+nKv~@q$3C)b8?h$(T~99r>t$kkeJ0Frl%RO{
z@Qz*ztkt~#`UBtn;ai@+|Casz%H6vUJUl*<V`MtNp#Y~eLkVm?&})NmBX)OS3x(_X
z3o$h}ln|Pd5;4R6-Oq7{d!+4=Vd47pE#Y!TY=LMr_Wlk#^juS7+h%&#fs`22p<UPV
z2`OUH)G8X{nM_gjHpA>&b!AQ{Yly(p`OG%0Z5*=%Rv%EhG47Xn<snt>_9xcz#<Xrt
zBM}i_n=o2)CTuH1mFz#<la(aBTrmi=YZO_lEjeelbtMFkRf3WN+u2hPY}dIp;fWqG
z>=<@?I_Fx|r*p06SSz}|=W@AF!p1Z$ygogFl9*fKyX!mFb!EOzglS=&eyS;WE%Dnz
zxt{TFuL#rO-n~Z-C#?L?v<eDZRBBvF(ow2l)Ez|>uCJfb()0eqJ$f8?|BGJ`Q$bX5
z3l1ej?+AY3`|rMmDtWknY@;3_Nwr{|!&pU#fo)na)}jSqG*t*fN=PcG))8&pKo*jx
zm+!IXGiq80IdZ;C%?7G9#5cO*o|sqC$IN(qAYQJGYbD!hy0dL}X_O*t8<`vaLN&~U
zHi~ImZ|QNUC`#RcYeS)@KuD#9%9JLDwe4QpFr_)yN+pGW)DUjzd!rOVh&HqS2Ziqq
zabZ}NGomDH123U;b5Cu4BW??ET@j_BRLGI|IihamPas&=m35hz&u0jc^UEva^~&*b
zrhE5uhQnP;#5>J4w?#&lwyANV*;J)aXe9_FYFr6(W=RRxk7zLxOhgI6e12uVym7oc
zp<QzzIc><L;C(<UiIi<2G@Wi{2CZ+MCmS(s?L=!dweKj~g5TOgB7|(tCIOu>%}!72
zF_*HT={Fxq^9H)1MY^u1?{{n=AyOgbNaqYSw~d}r0%bcowWrp~_3g|S7Obi4WLr!#
z9@Sagpvc8hrQv6%M?|Wu*9k;MQnBOMz=!4^_niNyVJ?EFD$C(`pzd}E!0dX$wzZom
zY<&NR-=Mn@x9i#OP81<o*Olq@nL1lS2z>eShTj5XZyAh4?{}DfM7kcST0+{op03-Y
zYK3vYsSk*4BW)XGP};D!9Wh4cX(HW(EQuhC%5ZmLTNi|soS$Bp)(tm~4C9f$hx>OQ
z7>*}&H=s<$<$|n*=mYceMk)nkdxmj`as#q>jMY@D5uzjIir*Hr5|rsokpev=*7GaB
z{g3|{|MCeZ730GlJL9O<U`$8uDpg8~)MO#C#<mTbX~*Don+a*9A9{viB&JNIr8wO%
zwi1>tpp>KAOHwGr1X3x4QJmg=Kx$1lju<U@URI`MZs_D|qTi1=Z73;oofm3W?2acG
z56InP!xyJFloAYw9mzLbg)y2CR%*;lmn+NWA!T;EhFIwL2aFV?bs>h99!F$i_H<nb
z2ttnNgJNeLsb+Tjeakl7Fx!1+(Ulh5`IA?vg*3gAmbp=Fw}iTCw5+I|qE<;R8EOFD
zKq0@O#s$c@VTVnLx6j}5<;y3uRGjV)xcw0~^oY<PfL+(oIYqAp^X1GsO|1h&Ld=PJ
zTR5~eRAGoIlTtxRNiMA|Zg+Q2ClxVlDBDwoL>WuB+u>v6`f{ac$KmdQ(^m#zn&;h>
z4N5?3L#~zO?YR**2S-=!4fPG;R>E_gH%d;}VaG5wr(2w3gD|E{UOifAj4>#s8jd^X
zTeDr}>iunVr5O93lA0MLY+G}x7*FK~mta9kg;s_X{7=x?wN_$E#M~^*64}IDC0XVP
zxwW)#3cj(iawYqg6REXf<)u;$F1dOrrDY9*Cw%_Q{C2^9dExl*h;@ClVQYzymXfBn
zsFesUYC@H;USIIfS5{vcfALG)yKhj!BFjRV-Uy#QasK>5j)~*l5pL|W7~086=oUgf
zNQ1R~13RXQ6ct&U&>eHdScfqR=M2;JjQ1<Ou~@B;d26Fbr9jDs0Vx%!482i!pQt$_
zq$2o0Wkv{vwzlb2DK^WgHq@M1mYGlrxd6&C96S2s0a*(1mK*ZdiER~}9`4coj<7}6
zdBaG75|U0UgzZQ|HR1fvy(n~jt1+DBmQdaAcQ|XYQm}0s-+%tdWm*y1@^Aj;Uot*E
zkSMJcOc&DhO!Are<%tjr$Nn9|&>>XC9`~)*aP}k#!#I-jjQ4M7@A1!Xgy}-<1i3nt
zxAeOat29X}cCN#AJzGqK(r#$At_V9a4g<S~d#Z7aBkcMevhP}Tqi>j8?FNLXWM$FD
z&|5=Mt++d+6)7^-NKD@mX<)xfCHj>w|LK3F-yLwfJ7O;Q)f1DV^a`gm=kt{+701&(
zeYZmx)x^gNn-AoiFxoI44|J-K-xkW|F}7LkmUSkH$leW5C9(*{ejpK%gW&FX;5toY
z3J>oe+1<a#q{#O6OafGADY>*oKx?wDq?n0W@q9ivHvI5_`udj)cL#>mGZ=wXic%}q
zIieI~0bSqPy~bN!Bj<#ViLSKyYO3vnnsXtg&;oHvvfJ-iw?=54Lt{l%1naUA!iu1x
zw5Gpr?5m%=R!@Fg$+BWjE%Y`$eWpLZq4uv-p-3?`{CQZB&LFxG88>RJ4C4Xg`iA#b
z5W|X+8mBc@fg(4JC`1msJ!=TeTPuOt4?V}@0b>m<SEoo6ws}U%R@9?}Mk<Ty1gREs
z6*Puj=I(Uj-Q!(56%Av%F1g@r2hubVx0dz*ss$k;$!Wabh`x0_7-`8VQ-!A54s&b;
zRb=1bhM{#LQ^l>pt$puScv0#50XObi&P7y2ZeY3;BPj$zXiwK#3$`Cx=SC?%di!(!
zsYGeC=KkFy$I~8JGxPP_awxSTazLS}+2gmFDg;$1N~+|T38Cp2uIt#=h2#_380xKp
zRpvtW8~*7NPp{7?YZ1or;b%W*cRZqW!dio*5`8m-)STHa&*V9w`vdmw9`*R1k|lbT
z_+_D>*jCR~Tu@RXwZz(n!LPJrdrFy7GnLwe`<&Y1m?Gi^Kl~wg5pyQzrqiUF&{dO*
zr)+Ihr>&#J)D|3}QKrKgN2rx56gfq%*E2)!u)_(VG$FMTuyx&_v=Lsem6$40Ia0~I
zzMaWZVQOVmf^fcay<SMm#@){?t{d9;t5kB>m@X4`UqCyG_{Tl&|4EC2MuN}Clz4bJ
zAq{Nn1~r3n2s`rO>qjgVEj2!DR8{G^fo0l|gTr>2UmV`CEfWvN4pAbz!;vf{{xTDn
zg+D%hVZUqrwb~8nn=rPn3;XUs+B`)Ubd+cV$U+sB;jl-SdaIX=?C&1vcSlZ-@9Bnt
zS}Q(AFb1g&Ii#EWqMbZy6^y$bLeOY~B{zR^Nr;j#LNdohQJQ{#WYC6no_O-V<^1g@
zj6K8s5&iWe@^s|x-94RdgvHD2E2`^w|J4VS8(E&u?Ev3t(webClD4fKwW}b-M5)Wo
z#Uon~%&&+Ni1U@;Jy~i*ZBCAoBD>uI7d-0iiC7c!^o9=`w(E&8;D!!qI?TAI#)w!J
zB!<i6`TX1ez~SYMU;fo!pp8Ng8sjQbwt}Nf<#>08avgV45&g>B^Ak@`&jjx&t7j91
zWw~OM#2MF0Ph_DO)n>knXSqyNBiaA@uj#dBefmgv{)Dj>YkHLXDN+GodY!ml&lF=B
zOhXu~+l-&zcz*eg5+mC@-!c<|ZrsuD4(v`R^sa3Rgw*syhjJQa8q&hl3foG&Oe7x&
z1?;dV%8hbEwj(7u*W0dGpoPH-ff)vbl~~y<m+knf=#6XS)SAe(&}ofTs-^#@3)`|%
zq=r7DZHEw=S|Dy4<c(U}VjF@-YD3i)FBGX(%<hrmI?TfdYTqG(#8(BSU`&rv?R%Hv
z@nLHli(x?A9=gNsgy@eg5VpMGmjxj;<Nbjmn;2?sNAMn*GqN8k;)R%(pS~pmbY~b3
zdvq<-Qn|kVp=ln6Js1JV*4=CjQndte4zaoIO2!R6+E{WeEx=wY)>!DkI>VsckGtru
z?{T`I%Zhb9wN{pOX(D)OB6^XQY63BDglS_x&G`9+?elk(u;2*1dv_1FF6Sa!Pli;2
zTm=q`w0o**FhVJ6bI&1>SgIBj@q>nvTNebXxvh*OiQ3Zdxsv^gnAeuBr;_5<()6m?
z^-E~Hqr-mJlK&|pb0ZqpS_v_<7D3lFy$L}nfDl|S7p`AEVyr<P4kSC0#l~{E;D7sH
zdH2oF+9F(Pn_lz-u}V5)vHu*Jf--``c;xlvl|TOBGyBtl-TU_}*E91x@w2bLW^|6?
z*X9ZT!ynN79@`IS@-4Mm7<L1PQ%Cj_T3QaL1MB6=@4kE@hroUu3Ety<K*=jwX>v*o
z!-(q#e99<e(Qd$WEprq?WZ44B8f@2*s8p>Q2HXA#3q^>=H=#Ce|BdL05RKg~6;ibB
zQ)shwq#_qV$&yl9U+(UBKn^43c;GUf`TX$({Mw2?PJ61VAQbE7$uaSGy1Rj^z!`xT
z&3ZWWd-|?N=1TMzlHc0Rbz7~Z)>z$0u#`XwZS<sjgOQr`^2RdHC?GBqPv8CqB@6xT
z(7g1W<C1dwnfH$V;e_pbdLIzqeIkSfpBGA=na`hj`=l^;N1Q$oHql%wx!n2?GD9uM
zR4@YOQV}t8*dOVRcP&HU19_WCsZfO=MnTy;Z%<FWeR-xx`1&t@#a9oHSf%;)?U^cd
z6N{<0=B}T*T^0hVaO=Bht`Kc3+qUug{K9m-K+b5<#yo4xT>Z-B`I({LleMIO_rPFx
zST{CYG&H7D%J^HkjFL?t-kBY#7JTrCDkyFT!`ODc)uZc*vI^5#WUcsRr6h^j?I<ae
zx6n`!B@;_w4*ti!fXnrYvIF;SBy0;Y<Qv%&L`m&|oiti@C}XM05|yU(26flCM|Rjj
z&g77(<(49^o^CYQv7^K`iV(v<kuWc3Y>fyd=!ZR3j%+ni%z!8f)s0wf&_<$UBZ`*X
z95B6>sBu92$*NFuf@Q`@h0%(gv-rcIjT%#CEh{N(2qCc6;$y<c4P_*|VULi4ao6+i
z-5tx^Aa>Ta_PiPboz+<50EyNX<19!hHKELbQYtZ5lu=MC(pIXLNST<<3w-=Neco7J
zKCxY1=?BN%alj9X%Q_Q6BuooZR9xS+ky)(>(Yhj%fU+c%F(4FKC@PW^OS`VM{Mg?h
zj7A!Xq>z1JnP#?SK}k5CPS~!)SdG-p2AfjE^*vfSq$t#qS#xZsKqWymh8zgwctDtz
z-6*LoM!WryA|&;O80vOMcDp^<H!S?NY-D8PI2nhQvRMaG#{1>ZzbJ@#V~dcaVznLP
z-2>8%?FO-yu``6}O4?>hX~^ScSs0E-T-UK&-%!D$V{0@LIdC}cS-yN>nXed1t8h$_
zbzPAHL@9(DTWLHtq;ubCv{ES9R?bpx)=)qx&HiwtDuXlz?b>dngwVPNv_T8`W7sG7
z)^ngtvxHgaC`H`L>t!o9FO|gy>gI7)6T`yw?FkzL#<;fdbdK}ul{_sdslXU^PSN)r
z_r0Unn&g+}&z|2f7RZ8<6UuZ{2}DCAQe6X{Nu^P@ppR^~A(@IB_pNp^whyOu9a=*4
z8_SXyhJn4cH*=F@nkTBP^tNN)+2&a;fvDTu^l*RV@cw=C){g_7mc+Ory=-Ffjh+Ab
z<43L%dgodhOHPc#fL}e~)At<4fxLO1zq}x9$Nu2~Z4C3v3+K1yiI++cww3Ma6Ui8s
z%PX}6N-CtNkvdcBKM*$a57xl~(Nq0KiJtT83m?Dx%(^UeT|*tqP^hf*`<_dxZ2k+X
zbNt*Gl(Lw6gS3WunFz~_$)O#$`;nSk%GVf+H43%Ol$fwxk9K|QS5{zb>j{WH<CI3w
z`Vh2L#Mv{S-$*%=v_Pm<Y9DIBNErJ8(~U?Wxx7B}`gTET&Hdep-l$dvH8ia(M^8wJ
zU^=QP*bZa}q-ABfzBOz_SP?0qx+ZAWCe6~{Mvs2EAj8IXnF%jf?!Njtb+>CL(VN&D
zW1<vD88l4`$YDcifi()F+v)l5R|V*_Wao6-yzF)i!vM9iYzy<$=G=YPxAL6=!~Q^(
znr;~Bx}II{=xsA=C}YumhtV2;KI1PJvTQjQq@qaKlGQag)Vnog)w(9U_XsH{!l3&D
zJJCT+Y_Fe4(=*O0oRLs6LMV2-BLzv_XL3nxvn36-A3<v3wjfeK2|<`9axGM4FvF1w
zO&G2P*SDlHwIXW9uQT)3oPeTKjFHSCL#ODwp)G2`H)%#{s?fyToOwm4_W3s*Qn?$y
z3xsh9t*9xt(`YF;t#Dl{^^#H`X!lIlb&Xk7Xx;4nNQ|*z%thN&XaCQTXns1M3H~z=
z@4x2$a6rUDTo;aG&+cJQPKCD`sWIU8d$N!C&7&oagy=N2CNh~q6KYFo1tHV<jd-)9
zEo+<S2Jb03;kq8%_ia}fBAwD)FK>ixCdEiUj_meh>)k~XOJ#p_47*(ueGrt`L>>g>
zSQ{8DHcz^hij+J>7KEyRYK6U&#%)mwV=UWpZV6gx@YBlc#}|5~&=!uzk(v^S%yxO>
z@NiGpx0EHdGKURWD%W+T%L;iDQ`dPzx+Z#6*^&FkSkYQhQy}`8=r@#XBe_^BSvU7c
z44zOkZXD>X0ge=D&~{)+0l!=^q9RK~sET4Ee3{81w1|98bjKt6cVD5po?0pvfe#yg
znep2jNY%QqE7)7JS;&z{#+ODVotG6=KO(n|b!yhTZg)iQA5pF{%8D^S*jSc{^?c!D
z{JSPji$ZtYwNagsbmO7z(EjB0{JTGJeojaw8HOHx`+zOe<yH<OSwaBc9vE6_T$BtC
zcN`Bptg)!7`1HGPxqSXYHw-OfQ!<CaVaEefS)`WeQbB2oQYaH42IBk%c}2<=QWK&H
zra7ht14RDV+2F&4d6{XD5QCAnIbkG_;!TI>D7E5aL1>5Rd(a(Owp_uMSCpu1)0MYr
zA(<ZOMqJ6Qr7lLU(*>bHJBzB8J{O#|WNi>tHI%l|D5DT*q;3n_vhw!w85b97%I%)1
zs*N>_Bv)=H@&T+ultiiZwuAlqRY7Tu>qmrEt&Y`5dfR%Ahcq(wJ=4V!W5oM_wT@vp
zAnd^Y?j5^vM3r`;%)&RVpu2Tu`^Y>kpcTe;Xk}aKKPMVrZo$YNBPyX}M9o-b(ZWFO
z>32Qby5P4favUk8vQ>dhG`g*jSYs%?!&DkQz_s3Ozpbd8DOs@F?U|Pag=p}{)C7G*
zyB1Puw3av9YC@!rq9dYfAWPS`@we;;a2i<?o$e@Aw1>UY4f~M=+cHySMXM33d#X@)
zzmeuQoEC^2ZdJjWY%K2Pa6%MHO1DZ_3DS^CLR3jE4I_dO)NVi_u=<}N(cItNBQIyR
z<=P5k))_T8x?Y0p(9$ptj+)!*9YgaDlQK0#q7Rs^kt@UI@i8*b6Nmk=DeSUk2BP0c
zIian^b_a%Whv|BJXuz_ZE3a>7mdhKb(}`h!K=nORDq?8Dk=6#IE!2Yd8?_d+kVq0a
z#GAbjglN_=YC#uRFALgqXwwr*J0>d8`iGr$T&IBU2ZDsFzj8i*raKHc=P-j~nO2-O
z$ZcVI`hqsi-saa0B_yZao__xTog>c+^YfMLuXrIT#-X*tpor_%Qnh|Uh(u9_tRSR-
zm4f0|mU%%-$!<5&_an~rRHYFpo<Dv^%>~<O`mW=0IfJegYK!{Y4$~djKfLGo)mJ2?
z5Fu06=2mL){?Y<HLK4D;R)Vn~i7B<OlUBA6QA*HT!?+u;eTPdlCQ~V-)zfRm?*7E)
zXH=(I&J*iAVf&Hsa6pqtAy9;-(*s)nbW!-*|Ht35Ee)~UJIj7IV726WomsbAQd74P
zgmVs~4M{X~`N?&h4o5D}Z!F8wat5|z+@BbRfv&qj?`lh|N?O;3Fy=OHl+-Y@Os@zf
z8HbT#y0(bij`>1Xbocio7S-A2Kd+S%B1yTM!zfX5MOV>qFHw+`BW1;WUeGmRsEjtU
z?|W*Bgs@?{9d;Zj=S$oDDd-RTR;Q^AAtr=svuS}qbq=>Xp!Peo?HcjauWYYhxP19U
zJ--36jpDRw_o&gxTtG^kQiPmaO{H_7f9hIMQh{hW5FsTLC@E7?Mo{Vc9-|~Ngl2#!
z&{>7F3M-qNC;OG@<umK`LJceZxVz;}61r<8KvGC@&Fx`MGc05uSk|>&dm-Dy*IKsK
zld41_K`E4RDAi%KLX^xpUC618#A3~8Wm_AdR%j)uq7Z8nl9jS3+u?*_4I4s8QVbMv
zW1Pu$za-yEiHeYzuIFGzx;v<*Ll1j~Vc!6pWow{5)20(B)9k}SAhZEvG5rBG+@bqD
zqBj1N%8hs|OZ$vfYAf{VEZSL82&7W6#t~y80;U^q{Yd>m!_b;UYl{2FMc4j`i$YjD
z`~9AEn@J*}x(+|RF-664d1DK)iB-n(`O{~{-I1CF)5|N?+2%G7mhF-dRF>xp%k_+W
zENr*3i}xQsup4*8m>9-A+PUTiP#PT*%QUl0D_VE_`Y-+p*E#e!V)`AyZwRI6O^=U>
z`8uI!$9hB?OKm<-Vr;#*);3C{s3>5)uEgLeNNm5O2#xD{P|bs!3WO}#l0>?W{nuaP
ztwMwiX&mF}o^AVt)f!#k{P8oXWcqQyx{fR)!S-ySU=0MHczb<ii;<)idVfzpN^*?E
z>nmctU{z*+Jm7{sNW<IP3)k}-+p<t|;j6E|Vn2*%(^FBTc|lW%^F#`sq!m&bv~U>P
zBZh&l>lwQ}d2S!-loK&}tY}(`xE1ASsc^Q#I!6i{*Xx9zR-|b&Qc(nVy8~J&?5^kW
ztM@2p3Ays}a^=hCXVTl5APdF`umU$Y!m?58(B4}i*<z)Z_JQtg-yms!^7{H$Kcl7x
zN{p<_1WK~T#Fi`lVaK~etAy-^ft;H5A(g;X$@cWj_g|g}InnhU504+Xdwj=uIH0s<
z+g6gFDcef$E6z0jf6b91G&MJ2F#8RmC96=FsxV#aPH12P+Uz-Iq_&KA_bAhNV0qmz
z&SCq3BpcQw#>Df>g)grcicn;!$T?F&>pb}8t|y(qhl<daswL-jp|g(1pMAsObYPhm
zw#B2QVvq?{62yehnZeptu7e?C(53@4whttICitxh=dyy(IHfVJL*~>raZ+NtjvQM#
zj&b@Yj}##)N~&hnm8uOjQ*OIf39jo9x{Y|9YdMH@oyc3ji^4Vu$~sZD3n^?oeLEwB
z#`Gg*98gxble%bCxLV2<%*zm@Y&7JXFm6nT5a4n~NlVHZBV;=@l}cPEg5NM2gj2T)
z(H$Z;N^i=EZMsmUXs8bart2wEp{&Hu3n{Mnb#3W$*A^Nh!AjAJ>J(`0=!YIFJ8HK_
zk4Lo9kRo*x<hbE|pi<jsL9`}vtrc$Ep@suVYN8Y<6n3}6TG*!4^fIkjV<wl#zEfyz
z&_W}0EANsHw7AJ}qS=_8ZCZ@0%|RA&`tvUe^p+{COfRn-9v{*D*hVQr5ayNFkIxK)
z!#IN$s+EKH8c`x82Xache)){ArEyM%k>UQ4IfRyaEFj)h%-A!$yJvrTB>Pt3xq07?
zY)Y_&nXYpvS=g3qvy4eYkrLbW2&u`NZ#@8E=6L@BB^4oD3F}M<Ee@`7YQ^<}8~Jr>
z{k|cBG6>_ifB$tWl%W#Hg!xLjyr6~-N^LXc!=C=|NS!uLrR^k)ap+-3N*lvq+m5R?
z!TRaDZ;@1%dE)i;6`w0rTlBJ$b_aI-fXo@w^$mrrZw&ZbsAwx7a)n$FIkKz^4vng<
zNviFjMpQpw_akg8RlBzH?G8vgvM%i;qDF_FB{c*#*`Q{9?@)P!T#>aEq{Nhn-jixU
zL`_A}k0agCQF20VS6n{ewB@)PsUh;^k2CYjN9<@(Mbl-6lPy&`j1uP*o1fY)Gqf?l
z6lYS7Ke_Gx;+uDPziU|WrH2a3vS545{lf#t!?8W&ou#D0dYy4yhwEI+Ol%v`Z<H8F
z*B8!dCCw*n=UQSmryCNyM><DInXt@%uXh`bvWFi>RpQnqQFEf^gfpfwaB^)U4<z3E
z)*0Y5*7aZpjOjtV5$1*I^2+jhZiPr)4?+-vr<Ba^r@)hljJD%$FbKagU0)E#J48Al
zYD?3X(ndK_n})W|1)UStTJpN!Q#*mB7+SXjl44!sONA9HB1Q%@_D7hWytU&AsW7`E
zRcY!CC;j8A-U3@v-q53sl_W|*)k5CZHYSl0Q7XkZSR7Pp*l_&-y93cVWFKyYHy})h
z)DBr|!<N_F@)kFa&~|<Wzzw^W3>NA(9)=<ZLTK(c;is0iu9YyqfT*ZLgJc@lLy@FZ
z$T1>Rvyg`ZyVD8h2DG*0oCq<ZfRGE?S`hMv&QF-$wH?1P2-CIVDbb*hrGTvkv0d1%
zFKpLmMDop|oijx%s&o{ksU=ZEVqLyaOo!VYvHgIkg%lbhoSH~llS@Sz*AO+ggmO``
zkzYkKrxYKEVL>U0D1AHerc6fuH(lF*mPNrhgHR1OeO^}T<w`&7P)d_+!R~hmQJCfl
zryY?qLdq7TFovo${lk%+>%n!nzEL&L&tICAD^<3bIUJAd9^WI)Kot<KPh8JWY|DmK
zg5I_=nw$gEGE>9W91?cGsRkg^S|PF|`9j#HHd@+Ruqu~|((QO(Q}ZIn(Bka5Lb+ut
zL}`i8QWDF$P{RuTO4&AQG#J;$O#1jWyZ1k5{KnqgLj@@fBCdq6;EY5|(S*zpc>4bL
zeEj}fWX%k{r5C#Gn75f!0zUK%)?#~$qWoA`qlCa}#c>=MjAFW6@l$)K#@;j90b>+6
z!}0NfC`5yhnGT^0;c~_mkFJ{Qd?ii`y{MemjkK=V&S8gvz8|T^VVxv~jb-&W*Tjf*
zmi4wkkNZ89W)Ga_7nDdGAMbg1|DK_)*my;dP`bzU1J?9pVQ~8qSu#0i$WU*kE=Wl|
zp9$Xo<i7hq{h$9Yl+r+LwYKL|yTL0;>DB>#^VBz(+s)CVe)v0;di#Fmmd=&BmDP(M
zJju70l;X#)0nPQ*#_8hr$HeXL{9ux*&9$Ud`|sNFD<v07$?ZX*)U6ze9|L9oLkPs!
zQpjx-(r_pB_C2I*MhY!yw#%Gv<^$Chf?J!J5VzK}_SIY1P2PUy(p+P;v~rs&#4Y4^
zV`ovDnL|prof3ZxI*9gA=k|4_`g`<<prTWxgyz?$wBTn=5uX#>=(?`!+XE?Av~Atc
z#&tI&`OUzfERu{?25GwHtce~i72EZVIE99Crx3U0)(I^+-l)E%WP}hb+m%0ld_gNq
znO_mh6;fdHEz6Nh<=z;S5Ul<}NwxK1JB#iF8K_!Og(SpGQWB{<?9PE35Vw$ADeZHp
z)eqN4AWB7)h$yMO4>eORD}I?+r#Es~uv#K=q^3wrf-DQww10bECgx>Bj|X%~%;zUc
z*r;j2Zw(D&Y=<xgEerNW1>R!DxDI6u+qM$o+Q#(rgv=RdJA@EK+0Z20KklILPg@kc
z-{_5E|M-A)J%{^8usuS;S}IFv?t|c0L<gd_W9YO6hQpEm@IcjV{+0lR#IGx<WOfe+
zdgCZ1a(()OF^;ZJOrJmU{O#}CuKF_3bpszB9}qc`gXZz>jxA>LwxWc=FDp?BVp!Rh
zYeWA|1u0tNlaNg`#yFG^jn%R>L0d`F!XNt7=GsEG5l0nh(?s=>B2wHM%5tSU?Q!>C
zVfS}vHzIGR4wOOMfNxqPYV-||6+AVs<mB1r3sy^9H&BW|t9w+}5mO^BI&0~Mo-7oi
zv_cSTY@^I7jY$O`Cf3V_m}b=d1Cow)zED*2xQ8lQdV0%<>xxu?<IgPas~vAH)7uV0
z6WyB?HxvDKLnajBHd9hSDowZBW4j(Vwk`reWuC8uoX8>Y@#9C%*E8qy8A?Uf#MfVc
z1)W3p16udQT(RSh;%6>jzCiRy2}&y>1ykF@>c<C=fcvjL|5I+$|NZs<gI7#}a7(U>
zQZQOUsSS6d6g8#Rf#NLI*_(I1kaMB>mM2k4A&qO;7^%?x9(5y_hHWC(M6J1j6pduq
z4K0-`4NA)PaLl=NMPD!Ik{I_V5`kFO*5WuGkj|lYJAA*VI7?OmlQYUR!^ciZdLa<v
zR>Wf)e#W|%Z%MgfI!bBoNTqaZi=$9#CM_FmOJie|0LqbtrY4WGidfrYR!W1inp7%6
zJ1`y5F9@Yus#yrG*E2P4=o*o!g)iF)>UJU&0%|75jcr~?(WAzZ<X58iZE>!JaHHm?
zoJl$3`W^lssVe*vDXEF^utyzjoBM{)>L4pEdL1HR@kk|DLqMU>&awIp6-SQudt^5@
z!Q^@-goVrJPiRw+rx9x;d5whE&-~%(8I-+|HR1H|75$yWZ;j=mg~9tm-%C_2%r9@~
zv9RfdA}YB=rt2HdXw0rT81fAPY-~#yidqq(WfY7tIM?bV(+$BZrE2p_W7!>gidIlF
z>)R_OS8TUK?(cB7?g5z-MoYA8VXq=3N?K&0m7k|XPO<s^OF;^5JK2IzhTZ9o;dp2f
z_sf+A+_D&{Z#e-7MUm_dv_eZoo-3VzQU$SQUY?)0o-b6T$kI@B8)M~C&>~?)L1&K^
zntYkaBRFT#Lx(V`HEFE@B_%Nx5R!3!peT*A9ZELXnD<MAb(MnaI($5`ZWG%anZh&9
zI^3{FpeWijqUa44H})N&Msg|CT<EQ95U^U%x0Bxg{`&vlSD^{yYV1&bCWg=+Y}!z2
zZufQa%}r$*5K?3)Hy4eRqIFY?3@SH!`K^B?r--f<DZz9G**moEP+E{Si9w)D%W$L=
zDJhYHXOQsC!#$#NY`SL_dyFx3!-?t!vh69_k+r2HkBZGqfwOdOKuX;*Ib@_Z2x$=7
zp^QbAj4biiG2KRI$OaIsRI2eTDY8yeJ9*Juf<j1?>F^=8)0OSWDN$-dYR9xJY?n8z
z7R_Wb3|PR40hLy^b0VjTG+oQ9<cdh1SzPGH9Y#x1SefERjGmkXLP8ZFdfOW7N<nNH
z=^P^dGgwlO*5PhPW|0E%(h9JP(P&jU-k%80+%BwcG{vq#$%K~VnvgjVQe>Xzn=#tq
zdX2TRHH!@$H3{A>E7Rgh((v%%8^*`4=#gB$e22)1{qc^lElfTTwuL|b?l)ASn+wE9
zZ0}lZ|8xSSk;Wi$+o_}?5LIGy6ON*9kT9uD6R%P`;>(h6_}h%IUE9q#hqe|~1l_QM
zZfNaJw`)Q!jlWbwbFkGyNgkqa*D9=R)66o>^j6UiL)(G(BfG-^H*|=UsD5QzCQ1P9
z6eSmO)ret_9S^OuKnP+A3|at+S}K?I%$71aWtOm@hdqbg5tJY&!Qvxo@`z58eP$~Y
z>$OnAhVCqf$45G6P*nUDG2KXaJWweRA~kGe-++k8d#aSADiF525tiFxsQUx9ABiDg
zyB=jN>tzNaak)@J#5oH_V>^qn4qZxfbtHq|x*PY0+wSPUy&s#VCVu>{AI3U=|GU7C
zgO%H8Mcl?fKf!+fy*H!Q^5dVq{nmf`HTXfzYh$1GHGfd~+Cu-s*J^)V{>jLxAr$H@
zEcN5h`2PuOEadI`{>iT&exx71_f75=Km6Q3T&_R)qX_ZeT*4pze5L;6%l>eE>JL}=
z_I>J)f9rrgJhp#FP8iq57_|zr-zeDw=>}e`G!m<!sMaMSN~Pp_Yb=x$DL^|^ry3tA
zE{NcpHrIDZ(^FGvwX|_VW>EQ-Af}<YrByg9DR`U&bvPkRhqfI-w9#3O?em9UiDAJw
zi*&kOJ|z*tP?I3$N|6<Fdyj=^BghmRTS{3+6{3war9ird5+^K3)hN)latpdO`f)Dl
z$J6(+&cx};>(dKqK6ARe!yN9AwW6%0_&{D4l<iw)!4AX<+j_+*F#7}O4lN8)C!%j6
zHdPX$CyUy!F3zD<Pj}yATt}ML|M&OUKY0Dq+(&Ek@v6>1t@vrCQ<)egp+MEG_?la`
z00c;Z8-@mDih=F?!g=*bH*mPWM@xMpnl9iV6^XQa`uiPDbp&Z~hdb1GV0wPSZ!^2Y
z0s4_{*tL%4>4JTE=F_*|6QAF(gJnGIIh>C4!-&$l@y7>62mv1>sa(lQB5J1G0AJqw
zHV-YWa66ZRK$3N!q^(sCp5Mq}rITIbnxx3O-bBe9kvY-u{1T-E`8GnTIijSXMvqaF
z-a7REi?;vRwJb}sHL-8%CAK)i%-lUh-V{}rw?P7Y2R8WJY_LOsWPt@Xs({SM$OxBa
zM{Kdwlx+OVEDHn*APW%<hXW3CKgai8YtCPc@eQkSp=UUputwa;+bsTs<+{6cjb^&u
zSToqe6W!?<JsgS49bGCl7GefxH9iE6A3kz0-xE{9Ua#VabWY+Uv8-!FEs1e(oPPKn
zwq&N;J4;HCD|Ok>zDew*7Dx9yEz}Z8%S=lftre_M44s6WSFP~I2@<2^J#j7Ui!o2=
z1fQUDC2kAz?SfU7!#JRfIFz+E*w!d5^FRJy|Ihpn|Cj$0tx9#(4+AkqTCId2_>7VZ
z%ND^81ZALQyf<{-LM_s%)h6aK6nTr3DtQ{dABZUvw+(Ng>yKF9LlKV8I$tSyMeDXF
z{dMHl*p`(NCBe;vBmUuc=<a|U4(P6<wS;Mv^?pO=mCMVoP$P$@@99oYl0SX@!u`iT
zvA)0H$hcv^4o6hiA;@ioY8FMK`HmEh(owB<^5rwzGNYS}Xu^1)KR?kOpHO*2<z)}E
zw-Uls5;_$u27E^eBT>sP!choqL#4>NC8lYHlm(_|B{XMU?-D@u_yN-&Df@v(Z4|vi
z%V=Y%Wg~8dsuVFMTHLTDqpd=<h;cGQkIO=dvrt%lKx;#ZncON(Cv0va_e={VZrDzY
zrmaDaiIyu>8FGr~+A!AbE_6-tmQoX{HG+oT1d=k$^NlrEa&F|fVDg5kNiL+vGiK~)
zHBqfY`#=air3%H?d56-9oP{CfcByoZQH-(9iP5zc>@d<JyrHyivg2xB5f?)6Ev4d}
z#_Gm0%?#c^?Jz(~E9*rl0H#Dvox^rLMLU!ifMZS@edjQKl+5a;$VJR`!|6f3m%EIB
zt#%AnRFA{M1KL<(`CmWV|2JO>%Q92vSz0Hfq}nZ!<M|OvBgVpfpDDSrILW<DX`?@#
z`0)4<V+^lfzF?Xnc!N<5YoLF6!qkCvnK0dfr%%7bY?1fZFT~|0iDR`=KL3i2nGi<0
z{>X4ViUI5Nz(0B?X=tzSSZ`S4DgdjTD4V9`y)85bouX8`b;o5mODUFZo~X-&F%4%e
zF{W?a8(EGu;+CLl)>bHSA-}&6&6S%{vL7&pnv>|7V-$OOPBg7JJw9VYzz4_q{6L|h
zyr))Jmx-5O{vs((v^^)nG9<(q8Ai_iNGzb=URhs%B3*ukoKa5j!C?#x!xN_K*_J!j
zY0{SQM$>!;UNOZxZJDUAnu1}Dg%Wy7=$Kw!S%3PC<HN{!=+P*aX=Yv}huQgn>-@f;
zl!ve7h*1p+=+*?$s|+b_7^AV)(o!N<sXUD}=n7himm~|+MdA{*LKCpslXa%nD(oC3
zKU<`xRywg<P(WU0@;;9(1=b~^T16=%-RjPfipwTRS>1u|aHLK*^0s1fqod-;U?3Yv
zD4KRS-LS1mhg5;8(I!v~RNwC-6TvaKT+ptER#0jo$4r`LmfIUK!Z{{l%*>Y;mbWW;
z5`jdQDhK27#!44b8_pRqxRpZN?ii(ahdpTRQ5weH@7Qw1pujj!A+y!ptDpy}HB@He
zGE-}$WQ`dP7{u9cw8nJ<cE~iXNlX0J(I_+)3)<7%2;Bi~y*T(w1>@M3ne}#Ke!oz*
znOq{<a%UVn&c`EFD+$O^LK>iEw5mYG8ilcrloAaq%BYg1kZe1pHfGdq0yVSl;sfn_
zymL@R9HOklC_~v33}y1GiSZ^Ke5vH?jp;sPwZad9;2kw>q`Z+zf`&xxI@uwZS_Fu<
zn^EPE)fi)Gr3gf?W>iVL3UHHAQ8$WMC2!=!&QB`Tn#p;iH892zLWc`IN*Ph9<%m{@
zIAM41hyFkqI-1tvsP+yYBuUO0LrEL#$PMF!-au;Ds?e>8fo?pawJ4?>NFv_f`cf#W
zvF1X~MNC4$qm`qT#{KoiJVll{Q5!U`>A%{wONDi&Im_YW1Lk_?*MIw0g7F;31K<Cb
zf51Hqtjk?0C8Y??pwmWv{mgHle_;qc)_0sgeWW{|@#8>lm1SG8sxlr=eE9eQZxr3o
z@zXCqfvQ~ZZ!C*r=m(q+7;EXgCDzPVr2=SCGGXJoN&rdd$N`jhRI6CyOwCBeU@WZ&
zV`Pm58xzKfjr=$alrAvc?;>s8B=Gz8{!ZvR{)?`hpFe<V_|8*mrD#Rn-I*&|q8LL*
zj)`TCRF}yb$~F;huawJ$wl385LW&c|KHxh`Q-xL&szlgC5;u6wean~{F)pyFO5gWP
zudl>qB^8LuGJg8Vco-?u!nQ_i2Db-;tTFD@HlPt0oE3=MmMW?WYFQci-poF3ko`%+
z>PAf~w2DEYwOn}G7|9f<;{7e9kW<-(FLE$_RR^^udD&VUxeT1ecY%~M%RI3y6W%*i
zAH?jG6W)7*?tntsHn!<bu9d#?n6j}g3#7O^epj@qf)&x?YYBFDPM5r!*uYw;T1}Ba
z_fjHCRcfhFo8%Fv1(he-dO;V#cWk$bI6kme!(9c`c*6INHc#MtsvFR*$CN}X3yuX_
z5^@eYjCQnTMQ@SO2SV=|J40;++6v`P!Gf}AWvOeTv5=1slJ-?5mdk~8dzGs3;OK`A
zMUmVL?>TinQz?`z-BrvP<vnH?#9eG0$^@JZC=bDTTz_PCj%Ay8`Q^{N{`4cpD*o`t
z-=j}QXd=2O4YbnI71|!FHA>4x&}`e4Rt6g6nrS&>mBe9g>qc$z?>Zl_p+jrwz*TFg
zl4&)Q)>+29G)P89Hvm?nTc+iR?K-@%q;;dzh)%NK^ID^o-BZ*AE?H70%Wo(bz1gV)
zP06ynLMtLrDK*h{W?IgfWx7Dl_#iOAoTEg<q9JI!5Av*e$9Np6Ia5>GyWt|lTY%QE
zN<1tU3Gs<3QrAg54z2A9F7YBYRq@g8#zVn?v|4u=t&Hye=9fZ_jr--!Rx;mxH*g$|
z3}c7)9`7bTZxe)prY&)usV(zx9ys&?;!MtsaC`)#=(dekX69|-%_@e=fNh1CZv6Pu
z3vrpqaivi3&On0k>50=1e;|8LObc;~l0ara)q+)u>2hbf-f3II^@coc5DLj^tWmNb
zG6ru&RgkxhvP%cIR>2y&;{#bWv@<wo2~Dxw7Qs$qSl5Yoy9)($UGZZ;E5%$Q+g)+L
z-?`r>yp_x1%ln0{TRFbJQ-ClGqK(%Y>l`k4wq+$84%i$}XlhYl6;)Z5Wo4cw(1C6|
z;GAM`J!}Q5K-1Wkh5PM_^Bw&-V62gl4+?R)Vxkj?qcyBfA{d9h!`P0j6)G+~44z^i
zD5|kct1xg{C3Fr2V#b<hIp|gywH?kUafezfTKBfk#%n_fLt83*T~J1r<H~514I-Dz
zf>J1@P--E?{Ov-pc9(JPvvjTHa{7iI@uFpKB~wymYdcoiNN9WAHREVYbDlKcdHwZg
z*6Ak3E0DCYZ6<Z%Vl7#M=v@dHEu#vflEghWMa)HT$uVKVil?Ak7W(C!(RpHf`IYN0
zf8on-U$DMs`1A*k-~FCrf8-%$+Wm!cd!gRHP@2MZ9i21OybzbWguZl8m8V*fm&Q`a
z2x_F{$orQYak*2fLiwKV@XXWqKj4NFnVGm|YOWZiP~H>9O4ci-td!f#wBBhsLMwzI
zsPd8u%6ah@SR;M}D+y#ZB`Ct=H^Je2Cwq}LQqBi{qvBO#90!6II;h>3&tFmMsGZ=Y
z<vYr`P_-8KGJ0Rq8k)A`9kGEn_^#ib+liVIeveCSC9`gck|Qxmm!pkB8HH6AZL~O+
zwZ>TswUO7A>-8NIEaT}BXZOytWKlNNLQboUi1rJiq>Z^1tP2<$1k9)q9+#?^)+8nr
z*;}+KdDyuqobedv!ARF?m12H>BjwD8Pv2pOK_WVfNT^+qU1%wV95W?X%D#hYyQ)hg
zpPAAIl%usO?^)Zky0?X#llZ7Yz$q=Bn*4V#uf-*CUj?b#J^nyH4q^z~HvHgt{QiOd
z=@ZzF_4S>Xm(R?3lXSLXu*za<qoQ~?9ay)Gq!cgLH{Nb<^u3n{Kv|aUP9Fm2hYuXa
zo{|gQFp^7VzOCqzDM=Q)Tgfc<8@Dg7++VJ|U9OZ`xW2vf;lmThr;+>nJDa^T9*%_2
zQyY}HFTS*r8MM_LA0HVWKC#Rb)9sD?zG9j@z-ua4XDCQ;NRFa`MhS~(x!+O9Ty<F%
z>Tr;D@%TXJz3}saR*M`qK8T+}X<`=3yrkI4msyJ9?fuI2I^lFf*UZodv<23U(He}=
zOw&Y;nd|k=>+4@I&T@Qy;Ph~oL$YQ{l2O<eBSp)y4DShjz(%=TOHE=_s*UM($2%|K
zBPmf!BW;nmu1vQ((>!xJ9XOm%IBzMcqP3NL0pqB<$)_}_w{mSzv&?{L<TzuDk&09q
zajsg6R-!!m+AZWP7<6mo&lqdTDWi6jb>H_;E2SiPv2};<drauj+M<=FI|YUhuuV7O
z{qpUzUbmIHEp$7gqZ=w%FX1aGQS?rS%`&TQT8o3VN@b_k3BBGiT6O>i#G7(?B`q^)
zTM1!ce0;(m4;bGwmPX1eaa$-|&$cb(`O2+k2`E};k!iX>^&`~>wAQ3hQQkt=6Vu9c
zo0wi+L~kEP#wSgGJfnjH?@>5hE5gUBneI69@Ngu>OkHP@9!H_M)>5g+h$rmkJnO6&
z|CUL-JFa1aX!}(aIkK*WwBD)njYH??zWXR%7cYr@N@-}7m_%ASj)NFMn{*FK>2KyU
zKr*(ir&&X;8NctFlorMkxrx~;i{B%~jgsX#i>=THL0(v6_Pv)lf1}S}EZeeBQX~XV
z{|(|-MRJ{1*`eiC_GTs1RHJ11If^DD$*-P}mT3wZ50yg9jRMrX;+&-zBL#<%XRuLd
zNzx9N*LPBi+|Zmqexe%9wyo4s@J>c_b@!N*oW)&VGPPB#l>jA;K`RZa(ORZfMWzu<
zqgX9~y_#h|r6nJ|l)po#5K(;pFaMRtKmH@%{r>m#q2u!9M}GX*|CzdM9EO2$7zoEl
z;xOP(14e_Ta=%~DTV~1csJJoLMAhJfWd2rf1VKBDSYsG_N6_FxPg06CChng<V|EsL
zD+N_6dE2<Xy|GLa+E_k*{{ho^K7anq8dsE85?nD&ye|v);l|^~?{HmD%9*^ba!8em
zu^y8l%?sPxg|{Dn<n7BB`tzCYbRrziXd^Se&<|LpsZ?qzq?E{|$uhB2a;~KN9p{CS
zGmHbR7UH(hib!`&mq`jruEb?UrHJ<);~mCzpzhq}g>LK!EwQ|PhP*KhJwX+UDk$wK
zN(s<z$=v2hsSET2;cO@wT5b}Yq8h~pwC-5q2BvUtD`{TQDM8D`eOz+6yfGe6I5$w0
z$nUIiWHcsMdadaXBQ_Y6mV;8GF+27+?GnS5BDpFtrvVyKb{T~n9=rD2S}QM;{!ax@
z-3hI*J9ob>6>}!1!dH_QJD^S*LpKbPbQJ<hSt=050n_#Lha=l~WZ717+Su+_smhEY
z>44J$Bh*?(7q$JTzNys=O|T9Gi>(>8O_cZFs0xXM4Gw2C<FUiKPlVG6>jItDwEKn1
zJ27r-*9+t%pG(T5`6di~>(Qr=)WZY0I|<ZIfo&Pzfl-xaCH}McfoOYBGUB?wf5Ewi
zs-J|T(>yWWn3qKkw`%CzC=k8V8EPhnK*|woEVUtOrmwpV_=ZBSxnNEPV|p}+wq#mO
zEVnm4|Mkz@-oDU#NBH3fsaC%R#jv6pG6u0bLS<J*m0Gdd;Jvu1C2XjD`yIr%CnK4Z
zBmy_=&q7TyhyQ9^Q^w$sJ&f5e1e%Pfin5r_5sX6_asE258INbQfo;B{^gcRD8}Tb%
zT9CfqI-ak%cCDan3nfm_GATvL;wXyL1QX?)MeUa#semrDB$)P;HV}U?^ZkPI4nK_0
z8pasu%Bo1#%UAPZPSmu@d*aRh=Nd-s7q78kU@HZ0fFRhCTES?AR)SZj{QtIT{{QPz
z7$2Yc=l}ixf%6~!Q68RLXnsH&&)fT*&%eB(|Ko-4e)k8CKRog2yYDHbu}%}W%go!$
z7rJ5P<HH$uII%2|tu}nAcz0q=Gp-r5>2ODbqfqn4{XRogq%HB|zyFD4jT{a~Op~y0
zRWi9{j?d4WpN^Q$5(Z1!HPQF`j4BX9KwC@MU92%>+U{^Qty1QNU*G;ju8JBXX`XQ?
z4yPlJfA}4t@43Cbb4!t0lFSQZ+N*^o_~BZ~O|pqwE0h?q&fuKcL(T6}$!NpSb)=Z6
zWs=2g&C~>{6+wWHJtlNi=c&tu`O6D7XO8_yKL+Z0qt?oBI-~r^Vh5hqLfKXh$A?`>
zQE3KJtt1M*3$n~qnhzg7k&R*PX3FhCT4$WLbiE_jh6xT6JVpy(tX4^aI$49!CxW*q
zEnQA78H*yXEA6Y~ur?}?iga1Ihzdh1IE_8VNi1k7RhGl831T2;T9P!duWFbOI;nU{
zrf5xEHvulTCJL`AdoC)(xIoTWA1De^PIM`vt-}pR{BR_06W$tXFXI5`!~P;G=?-cG
zU4`*u_mKEfP*t!NszFZBwq@^z61qC@9VYl4IZ=seWBU9vZNAc4k+Ek=QXyAy=c;bN
z_9Lb{V*4W{^!VW@+1w?=wouxgR#tM7T}`zfeLSI>MyCbc8n??U+j65DPWb*rG18SE
zk7umXgr)$C((K_k78RYu!S3&bNLfX7XHdi($=fPf;ntGo8{0IoUf+28?Ps?6&S5+-
z9?!VLanB@JnWnpp6Zd#Z_Z5ec^DbA{7-OUx1xijbuH4moS~Vzn7a}w|>jX8C3u|2|
z`?FDM+q2x|DBHG`6cg5Jxrm(n98(czbPzAZwk>RNqgjh_dzSSUfwnJhJw>2aC}UA*
zS{6S*S|>qwG?BQMEZuOeP>jfIg<w#>jW$cGdv|VWt<Y9YQ<{01$=Wa;4)R<np{|xv
zsWecHQbf4oy{Mn`*D;xDXx+rYZnOw#Xt<q~luFy(^+_E1?eB0Y98M?3lVtJE)5QJ#
zonQa@XUcSCu$I<o);aOZpZ>(~)KR;hC02^-8Nd4@(e>oz3bx0ej;I3|t9XBZW4>K5
zS`f6_YP2=PZKcdB%d%3HgoUSOxLjs#w*}oQkA2T*3~mS<&nL<H?E`)Hk=yl7HQ){b
z(<(!7nBLQD$F?@s_p3MnohXLp>CSSW#FXY8ZVYtBN@|g@=p3<erE8fl^9`$I9(j4c
zf-?BfF&>X7XV~&aOBt*5Zk7`jPT3;OXx!l_S<SUl(juB~V2c}}3E4Ds1B30^Oki~o
zZ+9VqK0IN@5i<qYW{mZ?(=%;&Kq<p=c_Ze;JkK1CCzK0Z-`~h_B^X5?0>*f9E*Rf)
zJO!2_DyiBkRVli|VXyiW3XM08+d6T1du4Er)6+?)m9@yiyjGm^k^!)7DC1--WE5MB
zyj|{WF*9@>W9L|-xL8#K_Jx>I3hf;^XYmWvim`#v1&noj-Z0;QeV5BSPzio-sZ}E`
zljQCyO?NzEyd|y+b)Injh#5w7YY<ngf@=LY7eY?yf1&+H4g+hU?$X1uqHEg2*e%Ue
z(l_gXAy<QH8r2k3aqL^|n5K7fS>)hVN}lbM&@^=Du;EcMF<d~SP*$)cTFYn=C7o8A
zS9B8tV&?;lN@5`aj6xfB{6V7W3GKbObFJIsB_p7BKZ$&=##mI%auBhrgc?xWw<YDe
zF03&!UoX@qW7D**<ecaq9tp=Ytvg`D0i_zbY1}?{cWzI*(;~H00NT-%WLyimREsQA
zHENFRmN1!{S1FL3cOrPm|B2`DUD%5-!IEs-#=6cd%Yt_Xt--5CE0&TAd0mO?$}-Oq
zf!lR-!$_JZG84DUg<1-(>*&V=rODpF`9MTNNvJAmZ`4Z3^6V>PaMs}v493@ttX8YU
zoo3lp`mvYKr;oc>qoTCKI*<1rx9^Lbk;Qy#O&0Zg{4ceUa~9Nz`Cs}+>h6hAIHk~9
zlw7(>Pa>aJ#hSlEMsr+O^8L=syi$!=wsqQAx0S~a5BzXE@%H-8Pk;UkKmOa#9DYzd
z{^LJ${_qKa<Mqn@m%pI48A@X~Kk)eJJGMXlE6dxj)RZa9Ldu0i1=Y~jqh00Y^B3Zl
zP!+nN!*!nV*keki3xUVy2To7V!Z2A@QjFN%GxP%wKYSw2cUWdjZG8FhN2be0-WD(l
z*9jh2d(Y`ZNB8haZs=vA&UfPLuiW!Q7XpK^*z-s(QI^4}5kn-kg7Qd=rS-51RW%q(
zsfGLPF5H_fay%V)czmQ9$tj+<m2F;F?^m{MqYFK)8u|g6UMb2D@9)%kCZoww@ZHMp
z?3Ai7IilA^4lQdiN)dcd7X+K3YNf3+SP6+q^Tf+<Kcn5qIG)IH#yd+ha9>x-x?!Bh
zjXmS(jKT2R&!4$`c}JZNa6Y1}Jou=#L){F^7D*}7^<qSHMsdrN1dPXoZ-r(KXw{Lo
zh^p-ylgKEIwqEr0d!<!M0V*v=axUK%Xer4KAvlRmO=Y7}>4y>5b;OuKOU|^eja&<5
zzQS_bTXl^#QqiFdwJG^I`$F3~VJcOSF@o!M9+E7vjc#;KQEOvc7ivz#VhH_d*SB}9
z^MVDsvWWzDpP#Sm$}&%M;~C!_=?*7M=tMD+_l0Gx<ZZzjwM!2(t%}^%T1yu?Az!A9
z8#;!er|Y|Yb}aL6)c_^l99ko;5v#@d-jwi;jI)yLRV%e-Iq+H&=Y18Ni8f#TJsGVv
zeb?h3M#d*c2%eNOTTXN|{0@&SIYWgKE5<9#5J*K+OG8<WRuxw_RNk;&vMjV|<SIT2
zlw1N%X$%b=7qnL7+|VWBb)yfK(i&STOfgbZBrYrS?apR2pc!)7-O7dOdL^z4B}c3e
zXf%1<m{yUFU*0dw(~1pYPh_ij=jcL^cM5?9QXAX05p%+YASsN>Vyu<Uw`DYzoCRq2
zmHm|OSE`1#@5pQjU_8MaTHeUJPs3UZMw443R#7WCV@N6PUMjiF?Ke)BbXTp39JRHI
z(h40MrOEztpW^-PFNOd3_dhc)iNlAFeD{z42%Q(!NL*24U{i_~JoeD@*B^i3A09vO
z`@@-WI1=X#r6gK#Sr>T#eIVDy<A;w7y7A@b|KQi3U)bgu=QaP`fBWz04kx;?FfW-}
zZ)7sZr)Qpi_>OVx$!X>Bc!turPBSG%wrS?REYw!%tf#IC*JR`{FEi8o8_Rv8I>*D~
zM;<<XLVHh4k(4&v=>x%8;^mEd+?d|pxLn>by7I?A{sV3p>Bb(VJ#mVJZN*!K?mQKV
z2BpXx+!{;GO{ATuZ{rBp1*UnydjLW~y}nQ0HkNr3q;ZuRZrXWHttOUfk4JQt)C#xz
z1?4O?MLMTx3f|t|xW2rw&Ucy;W7+65WfAvu*;ah$>BciyOW8yMS7MX47EwYeDwGA#
zj1LbSKYbvK2lA2_#v^`@!TkK?3n4g$^N8=I;;Olza;EFVaAuv8S-EO-gUCgp(hr?T
zN@J9n`WC6Jil#kf31qT{Mk8(;IVHSx6reP4))2;CK3{98Eu-w-f_I`I%3GBEk3zBn
zzIG(73d_nmnav}ehu*<@!sjXE1|^AZ-UN@bhO7(*VM!^Ccng5EMYgn|RoQjv`-S5S
z&1iyllGy>|lqfX|tWOE*gjOk0=H{Z>Q@c7R?E&=*<cP^KvQ(jy-^*C~-S-kgpzHfR
zQyM^_QY3CO+t$Q{Ip47@vE~fg359nYWlym$%4KmW&4MwOFrKje=|A1Wwe4h9FH83v
zWjU@OIwT)pU1;k}2AcKMtVoL-q_rrz!P1!mWh0jjvcox#@g5T#r5Xxeau?FdGT)$N
zymKhy_lv8NmFxmj%H(-vn`XBAL|!&Ft;h}5QW?e*$^_Qti8W%ad`>wR$SbXg>g{&9
zai1f;FKlU})J)%Z&_+UVRHd1-sDg49>Z`UgrdCGpqc1ts>$b_*ZRaDUWg@019i395
z8p2=<-#RsvfEt?ur46Pv3|0P_R^Ozh@(jx*(7s;gDGHiHDam8diap;8r}o%V^>^?o
z{QB!}%=5}WcO7}2xm_=$4E@I+P)1&k7Akri(br7h_jIko<(+o9!E__OeL?4y<KciC
zJZ-+vt42`}wBy&$KXQG)^2dMr!#6v5ERDm%Gw0_|EX#@tf$?;Nw29=-NiM>=W#Z*R
zjhVwyf=LG7VRDn@@$C+(5&D4k9Z$de1ILe_u>C+q;CSz^U(gMS5>yRcGB)(g+eVz;
zdAZ)`auN6AFi6XvGtCFg@rZVox~}AT2HVKXhSr+n@j(`=G0_bJF&36Z3^l4%oYCl^
z<1jqXKRgkF6Z&gy98M!8CQ)pJ4x<Y3_AZ_PH{gbmRF$;qAqWI(S@`w!mHRYft)m-{
z4Cg1p@kzRxX6eyX<LI6~vTYmJd7;IOvjhDJ4v&uv$0No$@)qev&%=i&$rdh^nlc9)
zaA5#MG4!<>R!GHZyr=C8vQYOikCT9uQY)!suv#K8gXGbI<YsT%hEj^03L$tI|AY=5
zJi!?39#-N^M>ojm&UsH-BO5XzNM)~nwIGI-cA|+lhFTiAR^e7DMQh>{NljpmDsLix
zWuq#^78AA=47;#|2DK5>#uhics;tvZ9P|y;{&PzPir_Qix{|gi3&oO1X{9aF*5_>z
zm|@8>B8yU_lqBXR-N1JP4OxO3=S0YW5O~v;u-2iK=;>qJ*p`{59oWD+E!gc&H5%i4
z^u8z3`+={upi}sUfi@VW!CJl!9@cG>UEIC^)j8u^Mz@Ae3uRldwPJN;%|(n?V~0j_
z|MG?Da_8avfI0WjHk4|+F)v~jxUiceTY-{E>%w%qQsRcQ78gdGR&vm8T2zD$%C@oI
z?`+dTQ<82M`Vo70<nZBpTsJ^?WxBjUT4`xv*~!PP)O}3TFr8;SALvKvy1cWpyJ$@$
z)xIN~J*`&Kx}vI#G_^82E~G)J^87BMuoJ}{7-ulM@I-0BpqM?+MInBdSjvuZca)OJ
zCGRRD;j>}PzRO5*iD35bxnhirnVOOud8)ynsZB|@_4l|GjL}T%#4kVp$n|#PaDL$T
z|NJk6bI%eZ%l5`RM@cRV6Vv5|>o0$$(~9-_PMq(2`RPXvV~@fRw8zDj78lCxj$S9~
z{m$qer@afnSaPYjVPH6(IB3EDruz*&P1F?Ga+Rbd1<QRREgL_4|B3PAM{H4CF0WjE
z{Ry=$eE9Ip^WlWH9mh`}(4muRu@=fQ(=Kn6+m-v&V}qw3dTc*%u%7FeU-;$U|DE*n
zD~}(Z>HaWct>x*5?<np7<5_0V+Z`2!(E46pFj~uj)dk|Zp|$1p{hgXOrdHY38Lbq<
zIARY+#;1?eT3EJ4kk54w07?meNSGoKd<n(2_4FS!1;r9G)B6>G5;MR4{3~0E{NW$|
zNZ*fg$vVLcR4v9a;{&xDDYhr+K%Va`+k&l`V`-E*Qq@Mf-|)IIoDU4=ky6BQ?0Z>=
z)*N?}&L+^jZ6j?PrXR5=EZ~)fQkmYb#AzYrOskcC7&r_gE(A&v380w9k}P9uqg9V}
zk|^fAXPfWjZ4=oY23<96TiDi_Wl6-Cc1fQ&e62O0_|_q)3f@}0GSu2ATV^harY*+x
zU>#N)@;Y(9-}cU7!?wKd6`*Xg#NL(}>%2I$Q^8nqo)*=lYs}(oO(`-@lYHMs(|3W2
zRNrc6C)JXe?^nKZxMPa@o<n15##AR#$bC87nvzkFTovE`uCzkgfVL~yjubmWFUpp-
z<0;UZQXA^4>8r`_VfK9QT++8un@q7JySb#Ydlxc6!QcX4r`O8Jv)Hsn8AHL+v_~oM
z29&Fewla7_=Z|Rfbp~F<H?%A=PAr1rp04kErT{E!#7rB`8d@z9b!#lzIJRwLnP=8z
z739P)aD0B|@pu2g@!<o_25cBPoQ}l%jW}JnzgA-0XsY4I4u3qMwV^)@gf5V?NVRJ!
zJIQpw^&_F{$UAYFm;|0_9bNF!*)A*Og4>G^;|$h%2{LlBOK7!9=66Yg5C1x<Y%*bJ
zC`yj93qd}k)|zBM6i7RZ>Z@!QEpF(q<4<KYBtf?4ve0rx{XNWUhx5qG`<-7uzu@0*
z{P+EU4f3B*xpI4Z=k4<=WAEvX2bQ?;|Nejd4}>ZYL9K~(p3op@S>x$HJ`;1qlIgtS
zd>WWy;(mMMd>>KDQL1Gej`ZV+WxmnYm2!Q@#1&f#CRi$t)EaO1JJXkUhQ}xRhiBB`
zKwAU;bf#Ft28`Boz1}FT$UdPq?zbCpl0XX;dg3Gm&teJ*g&#CpHG*{%Rxz-xGedtM
z=SVCoVR#Tnwlz4@QQ|5&4kdHHzBAMjl))O&HxJ`TO%a*|+N)OCrmN`djUvqxt!8|8
zKxxCe-uV3aGc8qq_vsVo=M(8x$iAo9!0kGT@}loZG}Z*ISz=YZT;6E6E1yt2e)vdA
zmFxSR*sk~~V!9D+CH?Acx-ozL%zU|X2p-i6ttGUTYQ=eB85N~CoF1{-GGBkAq=eRn
zq3g*xb9uX9I>8eJD`)~$<uG}FzkrfF+i~b<HIrkLy;9EmXvhnid0)_$9BFYAfLYGe
zk|1x)b7a0Maf<F95^X&VNGair_z3F09Q;}@S?^?wQz7j>6tl}3Rb@9r8Pv8?woRmv
zRB_!>nM&Sw8!(=GLf?z&OG_bB>OOL5m6r3K;1;Px{4N-eQbuOLt?i^ssT8G})Hv;A
zyrmRI7eY7emyY?C6``HQD4BN~tw1#}iW(KQIdU^(ZBg9-c|^Aj*AH0dzg;6r8`#~k
z)|!1hAqQ7lXL8MQNNuYiKa8WbMl7PVQ%X@&#aK;q9cp;Q8;>d#oh!7&#}AL3e*cN_
za1bL~l{m>Jx>~`{qzI~^RNIYi1`~RWw&D$%?&KxyDQ^-AWkbLX17kU(cGt5HfngXJ
zI*-$hN?{y4y?<m4o;#Iop0HANNRnPBJ1|ffXLg|lXl)5D2pu+ORBkBkvE2wt#)jVl
zhmd`YRR-S=R3*<*7kaFBV%98@yP#FufthJH4T{{<I^l*XZ9xmH@$0j@tt&<&xSpNG
z<o1iEAP~M<GHHXR5-3AahN=|p@6jnde|Y4#_dA#OD@G|ketah0ub9;`U9Y@<{+ZkT
z!t?V#(hpB;De=p{{h4;TaU23`Ug(`cC&SU58G1+8cci>wy~BpU>EjdWHgR7Te*Vjk
z9M3EM_{7j1DEbcDf-V{FET(DVdM92R`sB!|al0?9)57KF-*|X<WE@Av!+@?QRwEL-
zbiMH9*Ds7yM{o}3jHoQOO_r0V2SICVOO6q;<RQ2mIdq=EKA}9AmQgtouUCG4{h9u9
z<J0dyVU?sdQ47XdQr=*TC}$YY4^Ru<I?ksDmT8h%{2qGJQY7AA@FAdEWmywjyo$5<
za3bc$m)FmH@i)w=rzmKz7x{B97Y@Cr^MSNfYH2v{xXf4bHPR|rXXrF+^C~V?DsjHi
z)(veWV|%m~e8>CC8^7LfsFrznJaRY>XcsUhFr1$09-bHvC-Slp=NECfY9nT~TrgS@
zcK38?6&FQ<D%y1!r&wokPI7c>`D)-3Wei9q+iJy>OniT5x!>vfPO9o!1U}ecTQ}Av
z$!JJxT(5R#v&{9I78HQmm!P#uWvaE>L4JiClep;Pg0q3@9mXhvRg_p*^F}Hel?_!j
zG774atd0tWidRk`Zz)OFq#)7uu0@nWCiG%v(}tWkR9z@Z3Z}0HCUJTHCraFVk9B+e
zWzK@AHU<|wAq45n)|G0EEHjmsJw=M7vYV)MMLREQm#*8RG^@Y}yC9+tsu-n8{nk0u
zS}9Qse9l<ul2B-8sJW7=#OqS!BJ}-<;qg209c_AN`}`xSW%^Uk;WQA&12Grc2F5#b
zMcx(b!E6GStOac%J{Kxd@pi^?+ct8Ry^vNK;~ci@2&W_Z&|~ilc1LQUGWmWL#g6y*
z{bDf`ywegIW=~inout`gcT<eiEW4)A^|V&0RiK&L^ti(rZ4G6a<N|3TepuH{E;p-D
zMiK_AR<tY$a%%4cOU_s;UW{75K99&!e^(tzBWST!F6xpJHAjp!yJT3QsN}TKs^DJ)
zZP9inS*A86W?3uJcK*L&ZvQ5i!WJV&?UF{Td3k-|U;iKfPkHHe!wWi{t!*6bL6)eW
zp1IYG^A?pWeF%K_{YMU`BQ-Bve)*C6`xTv6mL$vH@!^4Wy0dMG`0HmLwhg`9h~vn|
z?|+9n9H_^@^5rx2J~7?y=&XpT@Nha~4kPO{^FRLI|8M-u{f+a-N7i+rl(>WHGQP9I
zeXoUON=(bbRx(e|Pc(1Q&XCqfn(o5;!Qsb|8VlRJlH-c6Fj&LfGV}e7Ic}U!7Sjt!
zQu&Ut8|l*tTUV%svFq5DNFOBsxa5pAXJXvQDHEJXGwr^3Yf5o{zY&$;@c02Yo`pQ>
z9QO1GzT^5j@z?+PKZy5<ap=)Me_=}-xvo5%p3z-kSu%a+u&S~xHvo>O9%TbrTgncn
z+vbVw?G23_+J(wE_MD!d@SPlZLx1G>^gaE@@99Fv^!AGD&vexQirOlf4L|faZ}HyA
z!J$1hXS&dF{`7&kMdG$lznS`q;0Vy<vQmm<0q5FS*C=BQBQDvmN{X08v9e{6|5+<B
zp*V~7@;TKi=!)PwN)!TaEpl+nJX=OEUyX5TEi@A;*<ihR60{NnR%(T{hzr|zhjoH}
zKpUZ`u2K9MrIK@|T1_(+v;(IIe!z`AE_m81nb%URO1IE<5L?Mb4&fjjgGR`olB<y^
ztx~M)XYwZa=$vJd+m!52T8wB_&^2Nxbire-?4GPqvgc?*;&sNLtQOE*%A_o`Tcu_H
zpp>DN2s+VHqE=B!L1W9?w~M}%#^`$v{)Ub-+x?BYUMN(S>xwQLE@qk$_pvq}sFVto
zhO)xA(m8@}@?B~z=FnO&Loa5nQbpd|d2}~WwZ-{BkH%ENPLxeNAxX*MZe1XiLQV<W
zD(8m>Y#2oCQ#O|Sh4pr&#3V(kcl#wNj_+n1dN|?^4+5}@8>oukN0VF30;p6P9XxL6
z&{`9hg%l;kN^3115v?&wi(`M=$az64Lk*pDK}zJ_&ij3oC<|+2z~~?&BcwBsOJ(1Q
z?5wby%l;7=bg(qzcFx$}_)@svZUkrOhQRsZf!ZoRef}&jV&8FmI1!xZ`hKBSMR1<r
zJ=<XvcvzFgRgN3FbmVp8cAa^DxnQlvA3DZ|6UWfAUV7&D3u#*!dWR}A_qRKbPiGE;
zBjw8Lmv>%Y-kG;Zxn3zvaXcIuhCo^-=IcAQZj|{(+*U#t2z}35XQ-AMD{s@xmzP(z
zT<FGu#+@O<`E<k^Ll^_jXofCOjbbeaVy&cY62^%J*Y`X<KZEo1eLx!pMq~Oj!{Lmz
zkZ-TFxT37b*h23G=oVwdTa6at5zdIO!CT9gGiymQmgzelKYfSOhV!Av)P^~qnXIEt
zmGgJMr~7Bgm%U%!aAn5i6+;!-W!Ld=e&GE0#4wIDEh7VCdz$U=r!zTcKL7dOxvv}d
z`;C}3LLWFjK5#gnaluoj8{Q!J=C*$hamIv^V|SF<XikKFAZ;@s%jS|I)I?Pc?E~HM
zNa!;2U6YnObq`E2N>TZW4KOlq-K#ijc2J@cKS#=yyt_VADl&i90c$N*87y*`el<;H
zgjDIA<awtZ{h@XqR`42}qkYxwr-ar*gk_h4$}(|x-8zS^Lbp^}k#?+y)iOseR8p<z
zmT)fMh7r>RLLW#avt9+%_cgPiHCUDf<t$wvuzTDiTBEv-=6hP)P_?3Rl^E9ANLoG{
zYNJv?RkZaoN^w9lxFC$Ome?;<Inck3h}70Fd**c6HaT>Sk$BAgcc?|vnv75+3RE=k
zZC|#wR+*=Xni6%sW38dKh}9y=T;s<4`c9V=ZczAKQEmP<25YU#GPLHsItODN{Q#_*
zZHxGRq<I<brdlZ|w$e~@!ni`smAuMwURg!9;!5vgk1--mMQOs}%y@jlpFaqeWxW$i
zCBDB4`d@iUPArQ=gWBM+z60fi-dYlQT`B8Ars9K47L;qWz2Hcz?AOw|@7IjQ?glh*
z%vVxMqB@N6o0C38@xrkC3QCm%V_jD$iO?BT$xsFRZ+6+R03>(Gk#1-s`7qUWprph2
zn_mj!&|{4uHpMtfW@tf)gtWr1KYwOfXIu!}-)<PK8IDIDo}W;qFulC6tP9IuFQ}3k
z$5HgcU5D*FAHI7g8=1A*!w5N1I~g&R99iyn-d}#BrNn%n`P0AtiI>Ye-~Hit_|7xk
zC!E)y3SsaJ-6P}R=m*bu7*N`<EGzfh#B{rnrb5bay+&d!H0{axo$WTGo>tE1kq^KB
zMArv$$=s%uF!W@UaQMd#9SjHez!V$GTF$2f{V>q5sQ!d;y{JrX7nF8vYa|56;e2Gi
zFO-~^*F;(uO4`uDad0D+%(~5Tfd-H5Jqp2Td`&FN_laZzzB}+={*V8KkAM85jBjG(
z_Wm2wuRoIBK9ko8<plBuYR?Zw<SZ(IvZDO(p3Zx^zNgj9$sRZyM*5+n8wOnP<eJHG
zWqSPutz-<7?l;)mW$QdfgB}BMx=Q@2b=+?^-qU5Dy((%-n0?u5Y9;||%B<TaFNZRK
z*;`&od&_A<WNLnABFPJ0_B`aAMMuA_oA@8BMOhg{)lyhc*x(t4VV8_rj5QJhkGQ3U
zz3fOuNSV$UOb93+_Iw5_@0@cuYp7dhE3#woqh$H&eH2(yX3LpiEDiQdUrpH(X`8v<
z?>kpY4hOUIpStsrL+Ge65lh4LBmL6{LN`$6nS6VPB&w>iCAL=N(9zKK9a<T5P1w+Z
zX}Bt=8OlZ}s|XB~+cD-A5GgUGQc@(x9R;l<7%R8kSf~xnn4Lo<XHP8^F&~SxI4=wH
zJdu|>$A<^P;e_@9H5^deN?RtH3E~^LeIed27-tBf7sq;z!pO26N>#kocxM?0S*UI?
zQ+1_TFtv*OxX9R*EM~WOU#KbKg2S97F1R9+&YUDnON0g<-^m%&yl2cw%#XJ#%4qSP
z<OCYDuGk^aY!Ee16>{$uv*2`$#}5J33@P3y%S_&+b4|P4SQ(2gqNLMWI)NA?DJN1X
zq9!ad5~V6s5zj_#P>Ni{weA4ExXV*(MSF)24(q}<1ys|PW-LWpQ7!J$-M{&z@K68r
zUnsS3xli<`GX+gcmEUeRbgdjZIlNPf63>|{=ci|$fA}4B%Xs6-^955PKJ@5r;PCNG
z_xnF$O5)&!ag<7<8W?&<iA`>6V^LM}<DdS_^mJodHZFGw0r~Xt6J76k)eTRkGaZzQ
zx0Z42==wlYj`!=G={}RAz^Qg;KmB1~u$tp>Ag6U-3P#?)%p7`yc{mcc$nE`(IX^<F
zJe?kRIuDq!7lzFy@<&xxIyd5cAeD`{X1V!OWVyZ3YT@{Y?+Lv}YsGXkykFmi<CGh7
zUU5n=xu7|mA2|*ODvjy#g}h~2Zp3wCSr*PuzvCbN<-c+K<3BP!JYjZf`orT#?){l%
z{0sN@AIT-M%nP^Ih1hk%iq}e1D9VxVS9}Q6<%+YF7Q`TRJRb0Uhc!|mtxX)T>u*2u
z`IkRKsRZLVe|jdjg=QsBwx)$PU(o$QT;t2kdU+KufZKy_n&fS?)<BtWt-Ur5V-?jv
zQ)btu@0kd>VKfqolrq5u>DT~TC2b4Kq-Z%{y>RHE(Q*-&ERn1cCQ%T0UsLQZ>IO|o
z7Pf*|XF+0X3r?2Dwn6?%q%1;$71qAN`Cc5X%PcCLoEW=~{`3(OdYTrZWnLFrZ1`><
zbiIJu_QT*f4p`%;TV%f8vHg+b;fa2H#C8MP56rEStHfR^Wh5^k&sb~dlq`vj(s&J3
z%Y~`tcQI<3CSc012oIy67&&yBbVf4Hsi-QQkuDKs9M(Dt2IV}Zid?c}k?59GNYert
zs<A|ESzM1gd?Xy6Dcyic3*25L1SxOK*H@<7Rs0sC==5{k@ZMsyV5>J(3C`g|kU9Ha
zH8)WM@1wd#Y87W38Q9P`-=n-oJ0ZZ<CSrmZ3o135kNEjc*I#iQq<|fdl;@8uN|Cpj
zE_jR!w6-tLPam+4--(=exuW8X(T1W8%~+bX<Rab-VOxocD%M0HW13J@K@c4ZVSDLa
z>86dizh%cINcq}??bKSs`hZeiMoT+hqm?4d>YCA;bQ{Vznh%t^hZCuaR+@GDJE)Jo
z|NZZ&RNU(u&Ub8U<RuAEZkq2rJwKq0B1dtkwp{3)L)V6~2ID(O8xPOV9FILXi#r}M
zE|70G>ayUl*v^qkk#KrjxZQ7D-&fvW?-*0?{m8l{j>iL~Rcy`Zt?>{#sD*<U+j-wP
zj)#F56R)qY+-?&UOV{^QWw?BPXTHxI4+n~doF&D~R;l`Fg_sh{v@lN#%e11(tBh10
zp6ELXB`>LBpH7rki8Qt)GA)Jm`i}CwaNwPlJYy@m=UT))8@e8?g}Y9$SSr<Ms_XDw
zN7*7eWeiOWS=!QiM^!Lb%lZ4?@%+Od1=d%U%oJ&a!wL7@_ZSt)b>i~dPt2DKH5XzR
zq$+FT3^vA5mx+GtF~%^Cqs%2sCdG*Jjx6K`+Rh1DQ(|2=Olv6JNVluB{at3gZESCE
zv~|V|1I~K^v*m~}JuU9(VEZMIa+3eHHej0g7uJ<RLZ^%i1FA-rZ6hh+BWdICP7$|R
z)H+7v`c63aO3Pto4W$aj(r85porJEpCWTAMC@X1Ktsw@lTK1d-CB)wS*;i$s`J=H)
zk+w~A=4$88C@AZ~cAtoM`7DeNbp1eoe!z4`wDHth$;+M97`z&&d84*Q$Pv{tDQ~oU
zVV!5TSU8VooY9!H;m8sI<2CJc5cQJrXg7ehP@`P(wcxB`2nKC6+mgibon;=acUP0D
zQaQFPaz>?*5s0yn6Ddw|p)_z_#zQS9YRz(>DH$W?ny4*-X`r<bY7b}B_<=S&(S|d|
z4_MQown)CeNSGB$mdi?g3)oT03V6>b@hL<Z1vRRuu8c;f#yZW!RA^tLHp3vOaown1
zzEd4M7)Ny-(O8;WP~50-BSq<K(z4);kwtu!QL8e72RF{6y{73v86U8RN9^Gw3b$1p
z$Fh$ylD`lm);q#*Ae$XEp%qPQR1xG2O}f?$D7AYZl;k`ZW274c!69gkwGyVJlK=o4
z3f&61NyjA6@kT3x&2HAyYye}aSiz$-@h_ym2PM^oAQCHmW1bdn?>DwJqLii^0$ty6
z)|y-{OxG(aXX5q7^72Xn-oCujsY)ArhT(wqmQ*rtuP?0Eh1;*Mq*D2Z|N1X<-9Ws~
zyuDBS`sFjzt?~J{8-Muyd3VNUo*o8LimdY$<vhm7B2cNyIP`SEleUf5-(L9q>ld>8
zPcVM^o}nMu3dB9EJC;J8GxL36Fi@Y*JUl)!9FD|NSV|Smd0KfoeBdz3(oPpa^!kHj
zI4`BKD96j&h3hXrW4cj*bu?nm1aIZ`E|r=^`Qf{cQlKA?41JJT$q=M6dwt{W*I$`#
zH(F~9#}hUvjDqg;z~Q?eu;Xc8JSmzIZi?2({Q7)mn4Xy4FBJVoY(>NZUC*-2Y%$S!
zyQ8%=!|}wi8wtH<T^1R`>}hbhRkmdzDw#)zevm_=R8;2~A5IwSc>QhW<>i&-{X#e&
z1U=zBhr@}`%ZRD&>LF;PT&ap(Vy9%5kA#iX*eGQAoSeg2L)oEqjh!s1B{N_M5=di8
zAx?Md-Ztk_@UG)<IAB~rH8ID<7%3@ZoWr<)b&e$`;<^z+N8cZWE0dB07|Hb@|Icxe
z!`iwXO$^$I38~adN`=r_RDY0gmk?;$LIp>m?TgL4MO4nHz4EUqF|8}LWNZj1-B_jz
zdA`uRgwZJ9;l~p;9MIb1x+Bgg;_Z#R$G_@Y>3oN`7F#{hSW?*?$4WYbngwO9nn33o
zMn;KFW<xnHw7f#oVghV(VAdu|7o+sP2(3aHZU7oPoCxQS^rvThH%cx+&6HZmt*~wj
zszm%CM#Q$eZNDPa*-_*|>F{!Rx*pvHY%a8>iL*F{tIqQ7Yez$p6^$2$P_CLyd+gy^
z!h)77DP@-X!q`coTB{orONxaQSL!-b^H&Czk$eNwNypk0ZIkg@yxmx)Nj?|7|F$^-
zBGGluQjIL?m9dzdC5bHy4OWUWDGZHuSnIzLfm4#`UKKi=bvs#BI9IjpYA{!+UR=*A
z&d{PzV11zZj-tc25o}W$1wuRiTVD#l{qlvw<C%{if8hQ7!k_-_FU-@*x+I<-9ylKl
zJUx4i@%*~1SZyIj;{8U`8eJ0I;Gt!Dt2y+6+jQsO{_-RL_NTw{<+lsN;RG(=ow$ML
zrSkLViBcS4=w+9oGN)rFPR<~1!>whkwm3Ip!${7GxBEMJS-IS1nvsVu=E745xN+d)
z4<GSP@(3OeVxD^W<qPG0W!)CG<e4@xsWjiCxRK+Gtp&DCT(4Dw6MV656=ywbN-Wnm
z-d?XLyNF(0LEUDYZuq<4d0Ul><}3f56!FJ_!zi5hyv`UZ$KU<Hus$-sz0-1GdA~5N
zD~D$%pF!O<`^wUgiWZ|WTGRI<>fr-vTiBE!#E(NKv;50Dp+C}x1I}r*Nm!>C2QTS#
zTSr_sk+~^I%f`CQlq?mH(=vB83aWp`^_~_BdD}2P&_6$;yAI`p7^1AfI8Q$g)FOn*
zrYcSEz%b{@^Ib-N-U%*Xch)xT(5h#rS&M?kJ93mrMCB}g7{m#>N>*{MI}Y7gobhO7
zq!QGc+PqvYtx?*4DsgHPbU}YOF^)$;Y?qsaBDI|~`8DuIDROI+D%cE+rf9>8p<2gy
zc!F?(CgjkD9JXznD9fGg@*<I-NREcF24g)fSF}JPSmrz1JYj0)G<LZDO#k7G9iFh=
zh-!+UC1;^b?`(0VZW9$pT^z}J0j0SPZAJIbp78gT$D|b4hI|f6?Ytq$aW+BFAIq9{
zB#1&c>3XZuINxE6Lz#r_3bi|756{@{fc70VZm4Y`b0MddGQE@1B820HETe0ctN>?Z
zzONKD7Wq4%oJSuX(4j}A4e<0;Q?=;+l@%V9(uP=IjgqlotR?hkRCl0Ypc$0*VyY_}
zwXWpt&e}HCX=0u3lIKyfAW-)FhArJ#^OdCEiA|GJ#MOkYt6(KSq|mjH=7su|xn(V?
zX`vq2PPBX`QB#y>zsXfBuI@|`EGub?-w3i)c}DA=_688qMAxCiK+58QF;&nd)^t>a
zkXrZ7PqP!7%im$AFi#tP5R3KY{l@*iFi$HrMmf-0W?Q3JJv9X9$Z_GPzx+GT-+jl2
zj}M#<J&)%D<KPM2(pt~&fA>8<|M=UknQr{$PydEDdp>U%IGj%0?kk@@Jo5O|%UrlN
z)_Ic|uygdsULGFns2JvL<NA6dO*8s*WOx|3UEZ;!5N{LL_bX{$F)8tQK6Cu^0ow<h
zv#g&TR>L$+G^6<V!|(An(55TvZAI6@cDrE@nlO5d66ml7(z0-WeP?@r-9I%^uT;d~
zQtHm3$&s{eOw$BP%e*sPuxUkIWD(ker8^ur93E)f#?4rkX<~^l%<GI@XVP*fFINn`
zwBjgqt*ETfs$g3cY_jV(4oAE)bivU|X6Q!L(?}SOm|9888)exj>P|CO7UXeb-6FX(
ziUbwQ0xm`FyE?~e&Gpw8Tx_&dSnmtraKMivKJ?UFShrQ850xN;m1?^=%woO5gn;il
zmMrO7))>a2r)foLkdeR?0F)TjLI|?JtB_k0PePStJbGV<8iU5t(9F|<F9qkkEDN1n
zs<jr<p1a#<xDe2;qi9Fdplp)->AXr>nNk9hGnU#EDOb|6Nmxp%G#lu;fv!K2b7j5X
zc09GgD8X&lZQe~!O;kbF;JO~)4Gc{a=PUE=oq4)RM_G~>^v1}i!$%HUIwp%Kqjvev
zPVbDwtuRkBp;u@xutM1%DOsjli_uyNk}BB|ZU1*^g_bu|j^vt2wZSe=Z78|C^sboG
zTI8w9psS*ekJ!^Q+IO^)+1_7?*B44j)Dj^_tg1BKxX%j~L5!$9IL0~l_eDBK?}E69
z4`+-Eq(xC<0v~XFqHU4@Sb=R_SW+eBgfki&oQ%5CKHl2ZJ60*c$SAfnw%f$*{gwH4
zC3L-X2il<a&t<t!{Pf!k*R7GTwA%Ric;LII5o<eYt70Img)K#5$v7V*|G^66v9%_b
z%r2s+i1gMugSIj@$|=eLZ}uRcv<dcFeKojiR1^8Lm?RCkNh+JOVo=MaQc7fNdk5t^
zP)L@--y?bLU;pi|44+^3Ir~hoj@El(T(D{3Xezye@Zm@>3X`OICRGljCwR@aTp7N9
zl0#2}vWBO}XTJaRk(5?GK0cyyC6&bMO)9kG&@lwf$0up0eK>Kwt$co;d3ydtcNq5r
zp^{?eZA+we<^BCe%Y_dgdvxE!?T#++{+sL<T23hAICO(#EbPl2Weg7wCz{&ldM==R
zhw%f)Q{ylUBJ)H>7&#?cl^OhWyOFnr-+z4K;mNY)@(l;AoZoi~jgkwRDp?R!WR6B_
z-d^ANb^eSFf&RN^9)~02R`JTv_XG3&hPIlgr!&@R()`Z-b0+Td$XbQTGY&mAI8usS
zzx>MEZ@+QB-2}HV3s+|t2Fa(j4sSFqFVtnFYzfury*d^Z&9cQ^))?596>FV%3cl+1
z?<+sP{W~RBy3>*K<0ty_1HpGp_X*-oUF|yI2ra8%E*e-P?0aVoZ5)M0T9c@KwAuk+
z4W(osV_hJGeW4vSs#FY(qME?TT9d`3+KHPeN>j94AqmA&>93<7EeBspw8}199;nI^
z>mD82i!_p`(a1O|)r`|1%fLO^E^CbQ3^ve8Wxl*(wPvX+#_qCEXHjKG9B7TUnyf&1
zkL^d?aDb|4C1Tcqwv}8WF%|Ap#BR_(zu*o71PPii+f3OhqQ2`T_{l<20c!%<^_1H7
z3vf55)k-OvTqLMQqolCW>|LCosB=mv>u`OK(Q<(}EBV@AO;$xIjFIjXD>0bpMp`#+
zFTZg4>%U=I5g&rq^u3JJVv^CB^MT>eVO>Y7YWI~GF>o4#a#q+tIf^-K&k?ZR?)VbL
zv_`gNp=rzce8BDTyKTLL^>P8^jgnTNFpiH5!vT#VZJ87c>ng=WHy)_kN)E`j@P4^)
zxm=i&M4!?swFHcTV1q=vS}T)+k|2s1x9bO7ko{%ZqAZQILpue!h|9w$ifE!qrIAy{
z?-Em^lyJkW5syUCl#1+VYoRPR%C?Cw!&qptfX=N^jYgY5vmHjsxuBH!Uq9Ucw_XaD
z>mA#5#4WO|3m>1JI1L>y^Nr9lCQb~80izUaXYf|hQs(L511TlG{PnM#4*^>%!N^U0
zTNi%(`4^nk{P5kGA07sRa{Tso=ejJUEz|e#{QO8$jW6$a`az?d<N1&OKz}#^72OKD
zRN}f2Q$bmSJ)AMS8ESsN(Q@PVe&=$%i}NygK7IU1czDEh2j=UIw_jd}(~ZaTiQpV-
zPQ1UplhNGYZ;Yn0#f&p*Pd(CT=ZGn@E*o1dU^QXr2>qGced0Dxl-3AAs*q({N%Nh~
zC=3u|7D-!wU=4=+<mgUk?0A4Mu++$wR^s(W*&-??YTC$3b9?=T>z7KNCxNX+(J!A*
zCyq~#_`YMEX6Cmy=KGD(B%!HkO{C(}#(68)<4H_X{eV9lWOQ>^)C}G?Y7<+%afTQZ
z-WtZfXFT+znprLrHEIrBC$K_bon{Q07+0KkGSjzqhpHv=?()q$EUxAR#z@f4T8Js*
zb&!X-NhM#ZK{ts$Y^4xak^X6G1UDR|(sjm(xS_Etdqg3mj6^$X)%J^huaHZjW^q`H
z55du#m<Lg!&sVj`47U|>TV+`Y`vIFN+7WEP0{7{{{eHn~aJ|O_Ps$ss()&Is<F#ZL
z=dIwn9-|%Jb<|RKxva&8j)J39jZ%?TE4R--;cTTG&Qdwn0^0<&Qf;95MstSZ29oK;
znV%!NY~&<!?_FmJ?CcyP3udbtxg@OhRP9koG7z-$yU;)`atVUT;^9Ei>OKP57P7fY
z?(jU5mm6)nlS`JoXCEj{DrP?(C8NL?$+_r;UA`H`5LPP%A*H5uM(vmhYZT5I3~t}^
zXi`axgG?5D*GY#^CGeh{QC5?vl4UaVXd|PWQY)ECQ4(iaQBrZY%Cc@O(?VP(tf_=%
zNM(gqs8O-pz#WhD$5A{RsuIizm(!j_E~1973@K}jjDiqZwzgssr74WI*wCZ)&d*tg
z@_V*(kcz)%XsMx7LgkGdS9GZu*RiQ8q6no?))5X5xWg0p9;Y;QTS!&??Jou6G|PPF
z@%g~<WbtTfEI9ADuZjQo=?&v2w%u6r>G?sFA*!-nXKwHBJRE;Ow?bMcruTP#`t6n5
zw4slWbf=y*uS{<@rt8Gu97e(E*wOW#+kGZ35#1KTp~nvfZ7N=Cm{;;TVRgftM*7o7
zS~aBg!us|b^KAxYdAlww^NQ^Q{m}8@pZ=NgyFXH$W6OV{@mE~%n9h?*qUA<j68E(6
z<IkTNE!diQJPkZ5i|z)jcFe0VlByMDQL2Tu++?<C9CQP=KcKWC$4G9FQ(;+G=CyEq
ze!!hR&>c^x<B2dHB(16x@_Z%UZ{%g7t_$86stQc&!usn^)KXdQceZsUtx+<wz2*FP
z=IQyFZaC~X2SZ9lfM&LlmYK3N+P0!w#CpSF^bF&Ovzin)=6gjO&2SKzt?>6XH6@1O
zz{ArAf$<$CN?Lc8Q)9YaQOXe4l~M|!>rk~4mxZ|Paf*AQRBoWk9=P7b>YvfylCPZ6
z7~|1i)HX_6YHDoT%(^b5xM7_YqmI(lUD{`?ld8OIa`0ID?GTePt<h=|RFKx7GRy4(
z%HaKpPK#5uXf5+}z-Wtefo2>rW*NO~nVb{K0DF*3jSJiTMpcTUM|x}7t~YX)@m2#4
z<AAqjPu22rflx@qZ)Q2wB5g-`D<xpP>3%`|Qc1@P-g$!ScWj7UP--7hsJ)7=dC&VT
zkya#wTdA_kA1xTaE17nrUTva%Hvt`G<nmQA>%M;fB~x4b=6Y5nNfz8D+VYO+59nO!
zZN(mY(zc?!#}CInovz1vk1B;!D^-h8Zi^djT?O!#tfXQ#g$g~@+C8C9OLt?<jvj|$
z=rGQLF(?wHBv~Z;4zCi_Bpsr)m?}=<Wx8RSypO6>Ow-gDS*|yV^EBTP{J>}eZd`~_
zd^8Hj7y_*|ww&>$ib6<hwDYJYU0KVO7#Av%0WTM)2rTR_v&}VAvn<N2)ldy3L2R<f
zm+pE*xd2UyL2WF~ICMCnYlR#sQGl7Y>#^el<Ma3U!x?RZnA~z?<L|I&{=+~0fwatg
zczz_V3-e_n*Tly^e8+sb@?~CeMUsk~2R{rrcaSB{I%BoLDNEWiU%vdz%jLp#OLV6v
z=z3f?5Z~T;x&F#FFFbzwfOnQKcxiJ%9IdUQVxr3ly~6!|XL@^wE%L+fzT<Eh>Bf$2
z-iYze^!7$dqJ>utTsY7_JaYK%Bgc<_;PByhROe)2xh=HmjeZ<RaV2M%riCqUoFC5U
zvT<8i?#qqWdFFB2IGxUHxv=I;K~rlJMVW_+W;i|I9zNm1K+REPk1;dd-dVPSvYs%W
zB=~$dQk6qhg=&S8qKxaV@5IXsPQl^f%;*DQ-uUI0U%0+sWgMlow4%n*tdm6?mb_-(
z-!A<4*PoacQPw<vco1E9PUHqdiaZ=2abX}=O<GpA7^z6^G_}c?$2raq548QD@5Tcr
zbPQY1`^#_K7r|B3Qbf!Cb$pXEab3TqMF~$@*!)HrVe(jqcQTeTS{8b_7O9?$yzo`;
zBM?=|C5zY0+HctkBE{2!PtXoZ!C<IqVcV7+&8=|WOGr&j65+VJc+<9`&4d|ON+&yp
zk`g6ulx0Sx6|};Io<4LE2bl_;b3kR?X4VwRMOats7K!Uh(}uWh(00XJtrXQbk0Z3o
zx=!3KuX{S0jF+gD5@iV*V?yuj{9LrqbJNB+_LzZhqZzFfo%b@HtyMBBl;o7RCdl$u
z#XTNsMK=*VRGmeffTC3iFrqcwPGnU(*v(o|dsQXLW@=ZoL>f7+JE~jL9aciuGn@w)
zJ`tyx(lpw4<fbLzFRtiXF<NA_HEra1V!K@#oq{CU0#18^^(beh8_CjHO1EyPO=5Me
zYAE(+FK$ppoo3=prBSw(oHO1yS{372=QP%8x}hhg$hs`Fq&Pf1VUG_u+i^VJSZ}Ya
z>$;yqdXFbHpoDqky-1pkYG`G|^k|H{LnV_+BF9KAf?QF~3cPNYde@YMa$1CwWZVu*
ztcdBXLQa{SBhGg8<B2dFQMEF^y%2Mxnu_ZN`r{+x`3cvLdtvc4V)pOQDNqu@D1QFy
zFI?WQIAu9L2KwUxv?3f2G&IIqj=`Y4qZ>Oe_dC~hq3=3UtNiKDKk?_k{zS7He+c-~
zPD4%7&iccNFa|z;|6vDkDYg_DhLNY|k8I0ITqevWNpP3XzcEdBI^$Wd6AvFF<^Dpt
zPt-|bGuI@ttv)!0<42zU<$uTF`5AM3rWzyp1rKL#4<ESR-}d0~uWhd9=np)ekJPlV
z+%A0n<!63-yK&hX!<P%q7{WL(o*uA1?3|UvmI|YFbbXHr0jg%324-bRO_Qsz&y_Z`
zJxM90MUu!Gw3>+bcj9`7R`!ZlW4tF9B*DpQ$x$|jtTI&_#?z70;}hfQ#565jZZmJ!
zg;+9c5=!SVJmY+a3o>FbM)n<-+nu;dRjX`}u#=pnN;3+LRL<;CfF%nq`1*DsZmV$b
zT|gD#<0v3*8|$(Vf-H!of7jxEHi913MkI;WJDhWDdrFqlnx@5zkarg-wLmW7-c^l~
zE44+GX(;Ex88I1QXnLa7jkYclU%9J3Ol_<&QEC%3b`}z4z27k74efgwCu|FOok-K2
zIA5{Wa(XmaB`fUC?IWQSNqNJx!f@ya2id3OEC`NmKNM=IdrpL0YQ`CA&O!igS+XBm
zV_8HIGhakXX`Q^T#(10ycporY#2hGvc5Z+6cCCA3hZIVzs8X6*GmRD6OfHGGX3&PJ
zJy=h0Ryu>Y$n4&FK#S~GOYnVbMbIFyE4C`NC6p3F9IX*rW-ys<958ky9}_u?q4jpX
zvfggHM@G^01BOB=j*=EqoVhPqrUAi=x~W9Ul8Boq_q5hDAE;WqM7l`Fmuq6aPl5{p
zxLgDvV;pf``l1!KRA@z_c6%>l(Uwr&WBLb%hwriDh$$IYB5e(Tp=tdMjjj}=Ec+hS
zz%*H!JMSfdZkG$yS_MH;GBsuF9>W{A3|=}~Y84l<Z34S9npQO>7HXbpMuY;+3Kg~+
zM||H?mPnIeSF9iL{Ycjj;%^Yz1t49rbN%1`Qn>x}D@GN*To*3W#>2xS=c-sH`2x=8
zGwYgIx0U8)>7lG+JfC<@g?{KML*Uo9iQld>r_+&u6-4zkvu%m(I%9I?={VB&-JU|F
zI3Eu<FY?0c+dHp6XQt~7RpHa~8T-RCosrDsy4)!W${b1ejjlg%ohE|Q*lytTkN?2a
zfBkQgDrE(C+rIKvfMwnGq_F`v3^;4id1D+rLw{g<JWzn2fBwvMUis;#3)(9F@K1ka
zI5)ap!qDg2#IoE)8aIqGT4)tr#IIobj`e=ydcWdczwj^~(Z~J%&6!kXMjKzguwCA<
zDUy<6u@kN+;ucY@F$@DKiU}<^OWHPBPAOQInQfVH1|A=d7$+L=U=^kn@k8haX=8bR
zLybMg8d6L&459BaK@q1rDMi$}FT`@9rOcu=mCEh)mHYKZKlFmyXiZSPrTi!Q*;<14
zc;|%YgM`%x`oeyz`f4pWCyTnA47G_7vbIg4AD2ZGUs_R=6})pwH1$=*G+>p*=pY#B
z*05S+i{)!-)gA*X`2m4yEJa6JD=cwA-Ct!~p^Wft);npL*yf3@3&dq6ZxXXvQzWj7
z1h#0&4p!ca)75IG_Z!<PK|D0j$oRu3A$@8E)(h>FUH)5gCgy}T3h#ulDLL4Vz7x+y
z-**_Jsd16fk!pLIT%*=T?|ZB>xNboCj;fuQt#X9*MxEctNsu5IgLQ&;Zh1u^yA#>9
zhzl39SNN@nU!t|$O(59rp%3^FFh(&3FNI2NSmm*OM=6DJzp~tKTwlLH0~ah-Bf&Y<
zW3{|LHk@%?hc)J#C?%#<-0Qo~WSM6wvX`N>-FuQ)mQ_0EeSbvL6dwpr&jf#9DUCy#
zX<CXL;|y9Wf;HgxOo$w@(UP`_biWFtC>OR|>4t$WI7%%jZL!Y&Cwg45kCh@Rrd<iA
zaN3DiV^?{lxPj}@+LN`X8MBY!nvA}(kYelZ@4#ww=p_$i@4V+}X1z_UTNKunu_)~X
zOP&)c7n(DuFo@LF{2e-lX<B(ac?msfip%xF(JRKG0}5pmq{F(b*c2pd*NGf3gdo*S
z2;Ad4+Ivb-ynVT|&I#u%*Y^vW!gtS)IIW5I8G=Vw&FOe#JRYg7aeb*=uUCHi`4`5a
z=O6yzcYOTxOiPismn-jYR~{bE&<e&n*0^CTxSeYG;pu^)w<rfzEQZ|QFI;~8h2`Z7
zPCN3nphr!vmFfD9j~n-(7{Ygi&=ZD{re0W=g_IjZAD|VPt6IZ4Nzz-Zk?B<`Qf1wa
z(N*-`(!8fToya~g=Zt=NA+578{%O>fS#CGBFTY`nn5V3jDmtaaGS6)5f>MT*C5dSq
z4-DrAs<y-mm)BRe>pL}NMz1)WI)-t;J3}iO-7;}nxW}1ULwkDS^!PxqmZIIB6y+!>
zQr3lao=|97ZA|ON{qhC~n6B#s<LS&6BWYepDSnHgG{z7@5cQB1X{6D@B@!2LqEgUW
z?U>-QcTF4Hwh5Uw72+P<c{~h+t|QikoWH(&n$Boy5r$2k1#R4Wi&b9()_WRCum>em
zQl=<HQM;?v2U=R$)){qu!6>y~01IuG;<i?B&f>jz0BYPw+d^KY6EmqNj3YWYyizPX
z+PeY$&_jWFnn)?5Yh&m;9v&a)`hk)Qu~te=7#BD_b&Q7x)_GyNUa;0Oc7eX{=(+$j
zlnTU@<gW^*HK<@K;?C6y6FQ6^WvrGnILlU|MB`Y8aUDMNSf|NU3{cA^_=+T`ZV*|r
zF&g7!M`gZdFevGqGJ=kFzN1z}!J?bR8Z9{PoKYpy1xw!#gubKe#A&YeemM6%_ThWn
z&`S}u&g5+(mP|=<QO)zrbiWIS${Jd!tjk6%Rq{x_jS`y}!+ggu90<d}e7iBdz2kI2
zFP7AcmuB6cxpKcSO%pnJYTU?MWG#(pOQfQBcmkyb_pTH|Gj6bLtL$y6WM*i<8jJH@
z{0Q2B(vs<|WJ;o%1QOwcd^fGpwwappHylW5g;pS@gt7ta8Zi}8Q{+v^)iy{5Lv0c)
zRie;$brN%;vX&N?zx}1~Wu1BaG%_4U);JN<!t48md0808fJQ1hZ!Ez$ZkH=A1csrf
zmdfq@O21^{>jm53@_ylZoAAMN9(y|1czixG9(ziz%(n@<MRF<F3brZMb>;Tul{l@m
z*68|xvzok2+^%>2<1e3S+GA{BJatsHlCK-K!F;<BQsegd7rOHk_MB<PN&A*F`7Uvx
z?{BYy%nqKe>*)Fp6<6MV`weF-Sj~MO3s__M_<hII)0y$mW18aUKmW{hoq2v7cz8JT
zzAXIP|MWirj-l`A51#Y!j0+v(hwm9fhc=e9PAv0{``a6&h;E1uzhy$_8T$?s9CjEa
z)k$gI-rp%@udKU{?sVe$cmIG52VzWIKmWpODbyG#>_kwj=*EG##ogr7h%sZ0W=n}_
zo(axlwED&<YTDvLhtZ9kC95H(jb*-Lt!VyzH=vzCWr56Ty`u_}faY2i`L0&t<}AD7
zD8@+LGosy|7rku@^D+~o>^9alkv2(Fb8esK8j<WtB94r>)=d#h!5Ax}DFuoS-3@5x
zaav@maoyN<epIeaH1O4+sw7)EG{l9f8svqRg|MoOp|r}nEwolemfJE=Nu_YR&!}y~
zE;7;q#kv(*sSJmK@o*3cWJ=U6Qd7pC#n;gI>Z_0fqVpc-<l?c{39qGF&<1Kvlq!1z
zhx}fuiP6tkx9j2~IeQ#cJ}>8}%3yQarMk!n$_I?~$cVh6sl;9WX`CFaB}z1@QHDNr
zj6*NWd1Tj+c1ouuQyXOEs7@42Su47n$m;^RFb*AECsCqUgE113Dn>9#elx=pb{wc_
zWxKDWIg+-7GCjtCts0xeGo_WH8cR!sO2N8-?|XdL$#Ps<s@51XI?Z%sqN=o-S=LCZ
z30I+NhccG3ZrrDLk`ne-@PV|4CWYZ3&$+UDq-GJ-SF5bI8*RPgtc3q`JIAQ8E3*(V
zpV~XbZJAk$qPYb-h+jy9(3IDil*NQ;oX}5GEa;{&DzoJ~OD?jE?mF2`)GEn^+bWR4
z7}-?AFO%e@Pk)C?A?Hj}g~Q{(Rx(30IH#eBm@CIh=N!&7l&L5h+ZypMa6Fu-xj=1<
z!$`c`$vG07AsG1Z*l{}ajJ;>@j?^N{I%CU5u8q&XMeftYvaP&*x$t(m<Gtf_7|=BC
zmn%Pip7{CmmD6#c)-%p}#*?Qt%gdKL^DVQjSKMzua(sSb90I26(8dwEfpHx8Ok!SE
zOc6}Lhfm*OyyI$r;`Zae^ZI_lc0C2faO@dI$$ajHfHju4mkU4r_zNi(x}e37wM1xE
zj9SVNm&{!+97D(B`Hbs_opW^J?d1#8>pL%>Ux?Gh`Fx}wJ*8E0D^L`wCi;G$HN*LQ
zrnE*k9tq=t?r`S$)9-PIGqpAjeTPkvvdt{p%(ksiMfGE>Lm7vzh2koq_qhImwgIJ~
zl!S^BiK?h5tI<YbgQlW!yU#>R&x}?Mni2A7a~M6Km13E0=(fMSweHNCwyP#0DeVj)
zVI_l6q%4(h&Xu^Pou@BpTa`vE1r;;e0yOj<C{kjPOM)x`HC9Wi+cx8^MLQ>M+?+_;
z%DTvlQsPRjatT|lu+2bLIPY*>2Po0{H}=nO!?a2;BNa_mj@BR1T9`St<Q<H+L6d{j
zcLP|5Q<@_O+E^L8j?>tSj=u<V>ua!l$s4v+SzKzx<}|G+sx~wvv2C62q}pUpVg=xa
zmaLFkl5q5`AfI(>sItL!BhFWewA4!SmG=a)R{7T5DPuv4kHi>D%@Iwb3mRvYTt3PO
z^Qt6t@EFz6fVCNLfoKM{rfFFu;iY9UwI(1JT8ShTsm7uZlOoE~Qo^SUNxD#sqaRP`
z@ev;a-MSDCSMny2yoE;S4{WhueMfgVh^)6ZaiP~jts81y=yW3-J4&{~6LKD9J=R(&
zfULpwoos-Ol|Z7_@KrM$&WytVIZ?7qIBLtZ7@?#+Hz8m{z_<>x0cSgy#G|sm3P+6Z
zP$m##W|=0o_bWv;y5pJp@GJ=Kb%O2oZM><GJPzw6EUKzP+9D-aaqy=`+X|)0bC7^E
zU&!l3OGQxf)-rF~-~LiC*6{r4JNoem`d8-5o!(iDQEXdeE0yaq@!?4p&%-dX%nRCB
z&L;^~*VceDR0HS7Bf(kXI`eow;;l$9*ZU-+CM|*P!FpmAgm28k;L%>w_YUVR%@|he
zSe3%`9o?ZPgg`BdVLVV%!+67-1XFM*nUYq@bf+0dwes>E$H4J);I|)tmGEri8OIaG
z8}!2?W}8Kl={p{W5f?n`EL!|l8ZYlJOt*<{@F<(OOm~)TWyuZY4s^#8##u_v497F)
zhesKwD2-DF?E{CyiKh=A(5*0EE_A`-{jT_^^5W*0h+2N`)A>j$MY@N+7c4KCw#~Gh
zDf1o5hSiNRc#ea>_q6k9+X?w}U9rJ&IE;*g$Rt~BBAs1Vkq)*(j2lIZsw{X9U7%Pw
zc#Y9048};c+8+l(ccxa2c3(v6t#>lDacHL@J2qp;4p54MWCR#(@O_7Ij+`>addlU3
zb&lYJfWwR?rc7Sr?#$JDR;|YDrnPO0<XG8Sqqhd#5A@#Qv_`dtQsDNCgq$Vyt=7br
zEA1*v;9)$_4I|EXVl*s8a0GQvCPM&xRa^WpqQ;8rg2->D8*!c_S>9QS5f|z(9<V2<
zDPoaBHzyh87zNsoSgpmdwU3f=DWE(29yMv4CFhNKEwm((TUEhoCre4CsWDU64YzK1
zFVfPuOw6ypk=6-A#XGkvXOu)>?jb&AM`dFK#a~FInyGOkE|N5+0~i~iWNJ#%811ud
zT?Fr()hej$rr1&<t#?#P<OJFox^BdH;bAozdN(w-=?hzwi>t*2)g;2w8zoAwl8N4-
zRF4XtR0_T~^t+$NcLA*xHAb{DLK01KFl!|Q_S&RF41JIDB6ZfSVVy^BY2S%j$;wDc
zMxYjS-iY&E3WK4?dj}Nm+eXfX(1|+91&?)3;x?5ynClM2TT|wGyE9EUjNPMEOCfh1
zj)rXt<H9#whPKj8?L-!=G5cPwk=9B4IjRXteVHg_BQ7&!5vTYN*mC*XUkb`Nd>Ewa
zv4(q*xt8fXam&P1iE9LF>4$+>;Lkt(2j@~bp3YRIS+|wL>BQ;b#JVn0(QF=zV!B?b
zDbjZxA&9$i*|MN+l|*UgT(I8J4;`&muD2C69I+ps_z)iGPw#lu(AM$(eq}5TqZ|F9
zC+Z9HR&Z~3^q>CB`SX>>Pao({1LN4^bY&a{`p#19SFRF}61rth=abAe`Tw-{raO`x
z*|y$ivkwuGnKhB@Nmm1)_kWoKNPzC?i#~_loT@4_BHZ0({n%F7lAsG60o{Ovn+6ah
zK!TMS9&X!g@3p?wMhvcG1-Pzwd4Az`y|SK9y!-GDrQ!3}8()9<jZ`%2!<qNr{hqK+
zs9w38Pdq+6NN@dH(ED)WKE8j)JX!X~GxR!)JQdT0Oeq`QF?cJDlnXr-{Hjq_bL0c=
zck;IL^2=XI#|>+w<zk%}X0)~VDPWuvMv4OGM46FF5!sz_(&IjMXq6NXXc8b2mMmtP
z4y|iaOqi~~DvZ;3d&2sFF`k?vDIRpyN3j((L@siYUCsj1JGLtvQBLGWV}g|#iuZ(R
zX5aT$x2tpF5^oXdqc%j_Ug|KXNem}iX3eU7+xQf9grF$EdoN%zArr|<Jmi=tJx<H|
zRZ45LL7~u9+HG2kOH^4c8x06<+c^jURg1(%PP3@o))SY@MK-nTv$SPKS{6!XFI|hf
z+Zj#E(w|<nWjO^1GL)=2jn8r-bws1D<?sNK6WnT}-tWLqwEc>vllF~czly=nYI>6v
zhc#Nn0RuFsjd(O(vl3F4+0N~X>X~U>(R0ICL1Ogdh6bitW<9+K20PUq>?CwkE2cAV
z6>WCt`GlWOLcuj&q}<1ie1E3ie<OF18#`_26sVo}{F%DlNUdVDka4xq_%5#9J~9Vd
z6^FW0O*T+87&HF8cr~;WA%z6w7|??<T$?znjkvILDNw7Z{7RI>z}r=Tm`<`5^e8sk
zS}O7^t#epo$TdMPLYk<MN~N@pHWnWyv{o28DsL!VQGUiu^Q#&Ur3Qwij7$!RXAN4Y
z1F2VuVeKNJZiAR06aW!YxOw?IV*Vdl6w;pg^Z)kmoM+E|+aOo&IrHV~SLm93j~Ly>
z`KaVHH=j=A)@fREetJhd#NBGGp>G=~3ALscrfK43-^DTOC$>Fuy^92GUf1#ANQK9E
z?%SQOx39F*JKlddF`vEQkV~bb`RV6hneRK_fB%W)azXhoygWa1|MSoM{PP#yomam9
z-6tO2UpPOUIi1er+BkoB!gz;!Sco-aaweq%Ta)neN9Av&F&QCCIv?;;=IPxNfBMt!
za9(rz^$Ml2C(YCMKk)s3_un&LR%|QKE5X^<L0zrFRzGTCT4#)E_+ZdouoZe>i*ri!
zl;s5Bgdk}htWO`fy!(J##1R|I^_4q6trx;P328A&&hYGJ=7)E3vfibQQk5ZhL+c9L
zk(ol1YqHTI11!0*-FHfBcoRmDi>B<E`)w25x$VrS$m!Av%Y+(nko7p&_FZNIPJ(>8
zHdt+$mJ`-!GFJ((uN9>#3Ml}xnP+i-j}2OD%d(#7tx9IGhSnv@(3>FW3<JFsMt+zC
z-^87iA{_~}X&6DJ!|QXcrHmjRFM&ixf?&iDSLnxvsuC3Btj1_Zrc-ky-VbT<)GV}5
zV{wxhwwBYGbzM334LuG5t@qb=tk(uLi%O_gN**a^TFHFGDw@&c+-MY((>OOvQ30Ga
zOwLkurFBP2iS6sJq-Uf>G%|#(voa)aMPh2pKwS*7XC3f9YNLINEgcDI3<vml8O4=;
z9F!EvInmk$j2Aj;&XjnNl)>r7GS7HrXv$;4f(g^WkyLT|8jDtPW0O2@hxUP5We$_m
z%V7UFPEYR$>j@J)9nZdPq+=g$W=WuHwNiV<T8AFdtR;53GR&t{_7$WcDlIdbBZ&c2
zYpIH@BovBZG`3f2-tnfh&VevlnJ1(JZ{=B_g=X6anm$HpD>XE=X>x6Jl*~QmNwDr#
zOG2F!v`X!YhLRi^C2FHimoO~r&_*P$tpdhkgP}W*4=0Qb<anpXyU1kcg)q<Pe@33d
z^L6K6{^ehJzdFwA4B8P(=a=8U@OXJgN||K>dSg2hep>k-|EK?%&o9s9)_A<UW1bgM
zj##6)oG#p+uf&=-oup!d@IUsL`11V1?Y<G_Nwm4M432Jl0;OS^1!Y?*@#~G_{!BYI
z7C+InROKTTV(l2`xIE5mv7)r&yYJrd@zWFM%ZfIdBQ^RhQk5k<Jm7+%lDI$rimH`k
zyU}lVvGyB<caD;)^aC#w5AWWyK0e|M9LK>F79N&~?|=A6GG`wD^gr<U-S>pa)6#}&
z8PjCgMHdKGm%F8$@S51ytyYjTsyFHXD*g9lGUGX&PrQG4$NAkS+_VVIG)B=k+bepR
z@oK__Kxv(f;J%Gt2-6~==WFBT^B1<;op0#h)(iPKh{rBbsnZ$L8`FFtkE^_DO$>iV
zVUT{Svw@O#(thyU%b#(JWKNjgX^6b>NC&7W>?!N9UIyo?Wod0yMb3w)AcqOf7{xSs
z0BWgX%Bq#q<$`gmyu5vwmU<I~6kP^{DbW<TESUi*j*RBkrT9%j(&h$@Qgvw;Xfs08
zjUmhn(|Q(F%dwMNr5pvLrLEemCm4zU+_qhamfA6`E6e2_Hq4YFbAz-W<ULYTrnN@z
z`jw?mYoy!fSGRS_g{B0roXVYklTi3pGKPw=3f<t?#X0U&M~^F9tyMPaco3U_5slS?
z4$)#{Ypsi^uE}$>%A=J98dPW_0~jU4Y&QjL7f%c@Fb<uV8~e7i?*}m#?6lI96U2Gz
zxG-~hxDfm!Mu~H`)FKAL?S)br)jvov@n@6{FtVacmiAQ3kv1|9tWg-%F<LW)h2AwK
z9>gw8F=G|IX=2_fMNlP1sS&|iG1fCLE8Yc}S&X_|(!sXf$uTl5^Jvw1Y4JKkiwE?K
zwP2lm4*5tN+fG5DyzsG(ZKtG!HhSDpMNZ#qLszgtkon$Qk=@e8@Tio$r*XryFraxN
zZfUfU^CCr}c8Md^ev$WReIQI{tg)p1nYuYlYp5_`y(jno>yG{pC<^z~=uUdy7|Wh3
z&o3L+$@S}WT6s7dv~GO<@)d1&9^Q+?xoblznHo28jwt8JaVKpX>*>UiBfS*X%bEMj
zjTAG>X@#L`0EjW7v?5G$dI+sjQic}Ue*IB4!<<QMgfO$7&gdyfgrMtu{J4rjC0pKo
zeBkki56tt5vlG^MxxLWX?^jB_(b5j7u-_xMW2dGfr%$bkyJRoW6>BeCE>D1BkCB(>
z7kY^}H0O1pp5C!OJz}hrL8VscKAfekQmULv<93jfxNBPn?>rLVs7S|M+}Lhhvr9!=
znCFSp`GGLcG_9pg(vafXi?nQdC#5JGqZ7BKs)cwLCXF=?O5(V^P)9y6y^)6*t9FBJ
z6>UA!>C9wgFuQ+!Cha>8i6&hypHOxt?k}k0nfUoLsTO*(Snug5*0Z3JOD^CPwZZ-R
zm0A*$w<ss+WciR*k@sWhsk@xITN$U(F62lJ>uHsCR8H&)K{<3Kp)nomH^x!#vZ)vu
zyyYmtCEi%(brthdmI7V!$a=_$rj(e%ycd~esifl|ZFgERRO3kcK>L7J7B`*ILk{WY
znfdfEzQakNbFIsFXk6zdk64sSITcEL5g4HQR{RZ$PD+`cJ61`^O*4WWZ|R_yED31o
zV9OE!gc3DMYhocBUkf=<DUpiTRwUeKe3m(7jB}C?T_EQ~RT6O1nz-hT_?{^xu}9&J
z^;$(imy*CH58-)*W#aO1VR^W)K0bia^e#C7$Mq}ex|8EhEe-r5-Jei4pjsjHMvfch
z{#7z8dM6JBn^PJW6WwgIKITbHayU>3u&*|%^@0XBmXZtckQ;yynwvoRw9H^~k!)_=
zJJT>(8fP&Gjn=3}>~d>UMp1hsrX)1w)^K5>D&auo5^2>ECOMF4ZGlcx4U5KVjrXE%
zw5pRzCLe=^R12zP=pxlNR*Y*(Nfc};GViE#)R5b1KVhdCjJUw9Q5dT*6>^E78a@9L
z2F+HBv~3PE-W#_1!k!bSWntS7PRoR~o?tvL&o}PJSDvq5`SjgKR{sHxNJ)kJ^Np`t
zWV>yn_t)6=#NrL^QRC)7)$snZa!!eTKR9A0Obg0@vX*&X$;Tl`=l$T8BF=b@ZAWQ^
zpFQitB)!UBaoTcT&XlLl=V!_K-Jb7E%Zcu+7(!Yh1dF{a<fz$l=gXI0cz%AtdPgdS
zhv`gHny=4S_I;<<PH73FoFE}mBHt~~zrOJ4(?`yqKG7dPVVwt>xP*u8y_b<Z77ucc
zv?7D5K8B!qWCZt)B$ZLe08r6d#voT>Wd#$U@&+-Xww?U)nU)UBC@QKBRJC>J;A9w^
zB6;5iFJC+bl?Fjbmr5I4IHATSscWS88SlQ)D-ER`sCXqXV~!izMp{0|HPgJpv~P$7
z>Dp^8Lpf`qcP-vBE$&dXmeXL7emT99YeiEjNrt?oNd#ot55hE|wI&2njzFaZhsI%*
zn1Li`SvDDlNzNEWP=;<qX6daVcrO&&T<IwdzeFT=0p+Fp9X}G?_AkHi^{+poDAaZ4
zUMsaLbT|>#M^SM3AjZ9}$tX;3Sk-WYLzlWRf2f5fA-wc1+fP?B+FQC2C$Y5_=M13*
z4xPa}daI~j2Wu&#C>W#0AwxEAA3SrIMmq{LhM3AAZHp494yr6VW33ieQ%dB~{zy5$
znjLAG+6r6i)Y`}uig9S;z<7F7=y{@9X&qZ_nCF?xhbNYYcZ7MBY>ZaXIiu4=yS|WG
zrz<O&04)hs^i`&6=vgudx`{AmG%XkQ+g0AbF~`&ztN^Oy`(!k3a<qYc7d&|wXcU21
zD%~YZgi1{ZG2U@OekZLoC1qidX~?}(kRH9!P`eoBy0I*mMI6aWRBC%nq5|^^*7=dK
zEW9Xd4L(dlInXkDskI4Dgh5jE0?G4U+G~x5HiE_|t#IT6WgV(I^qz4Qk>XDcttzb@
zFyMy03xg?c|AeC8jpA{g>9z2D-MQz?vI<f+ro{8h3yVJyKAd<udrFM_^}qZ^*%Y=K
z!uOWo9M{_T>yLj$DT8r}`!7G!t76xR)(hvSi}ZF|9cmTLG<lxhKcc;3yYJ-vK$k{4
zlKgXroDYoFG9NgfP<}#rDGI$+={1Ldb%on~<9fexzu$Ryd|(L^bj5T&@%Y^b+;XOL
z!}0l<YpwWZ1VjhI>@Nw`D<N2%HN<_#jZ^N+^(I&5^8>CcX6xvsP)fmQMXSO?p*0B<
zZ*asy**7u0XboLjNJ&eNH>ZSCjqWVpl4g`rg4wNw<M}i9m(OfpKXcqSzFuF5F-l1N
zG@*?Qr>qaCarN&b{9QI{jrYQz&&NSlpp9V)6V6&{72sL=T4_#BU%C`juB4JVUT*yQ
z?|&U^qz8=E)RdtsF^RQ2GG(PjgE7)KA4aQIGVNPTBpUB0YHzfzC`C4**2IWZU5BF&
z^n#Y$<1DP8VBNrm_^~l+kz)3W?giaN=4YLx&6VD9*5a*}7+7sMQj(v|cre2Bp(o<)
zg_h&U+D?Rd!OSbRgK8od)ZU^jG*wZmlk<VU4@qP^M7`xFXpISor4UOZwTAX$=CTjp
z$@f%)ZUs{lEk_QW=(<z%gtlNQ*k<w05c5nOCx727v$xDaazRv=%_yf%sXGmXOgmD+
zdTZ1!gIwz*I<_^~V<Z+)A}J+pDw;%>Hbm6myvO(;8$|139CHq(fs%1nG9)YwT@$U!
ztfYO*uc-2OTB99TdP(GtNNJrPt-vPF)L6oKkqv(glw-+QtqIP)2H6{DWlq<{$k+SW
zN0f$<$ssxd`QG=MB?n>KP_^KlmjQE5g2i_p>z&LByoToKRSIOc78e}7D~Z0O)2+d1
z@n+>(aWpWUlq$U|rrEz{PVe`f{n%cI!nMlJYOEo6H|Au5j;Djw;(=&*Qmq{OhTK5G
zZ#SNvF07{sXC?e87e$VF+-~pxgrZ;#UN!b(Bh}9JNGy|KInDGcAv{XSRqSasO!^(R
zdA_{d_|O0ApGh(E$3J`rwQ|0k$xTsJ<8+!(O;PrZSSoX=)R@V6XU~b?J<eHrEwn0J
z9SUr>XO8;@RV?jVE2ep3J*`6>aiUpEEfLjZK)FYWmb~wG;=Z%po-ui&SdR`T?0O;M
zSf-iqq&aT~dDzz{qX^ESd*g%Es4Ay5JLI#c(+RX@-;-d7FCQo_aHK>zBqDK{f*^HU
z8-pytfNLKhC2Ac*)?Q`PQX{6aH<UFfZ74&--#IvLSJW6F?_YoA{_=%m-#M-~;*l`D
z;+;nOg=rG~bY)y|$KgbqCP?U<Nc+J(9#k6Zx}r}V-L1R?aVPFyP)3YwXhYf$aqG6q
zzU?v?(~=^#?;A&qIPaL38EcI=m`g_E$1_8Qow-SRo)Pn!a}!oc*m^ESh@?nTmvfdn
zd8RqwgO@(+NZZm38l_5TWe31bdiJ(MO<A%nP~xv?1%*Vo$_8eMn{b)5b2vRy`wOlY
zTH5GwW3i3(-37m%$$mx|PcCq58>)9&iPS8o+FmOCP&jKw(wIY|rPXIWYNW#j=kccp
z!uo)I_<_lLOgYf|6@7GaI&eKgsl+CPP-7ZFb@^vI(Y1%#2&(b0SR7@<h#FE#lp4uH
z`%Z6I<wWjCM>jie3<ho7P_oGP+S=hgK@BQtA6N|~?OiJ;WUUlh4L7lBlve5KAa5^Z
zqfjNtjl5P`F2sGKDamOT#ZRKeB-oPLDJ}@vcAT!YSuj2fIdP|t90Fq<s!Nte>lLq^
z4ANUC7r|oG+USFItBnDx<yokqlDfZe+;4bm!8u_%4W*lL?$zX~fL4-Z7IX=T(#A^m
zL~YUvRKiS?WIKtONJ+dw*3I&cbV%FA{rXHS6&nItbwP)>Mwo?1)|IrBI#Q4s3O>~u
zy%ywbS!GYy3qdEmKFiIw%e<12Q{L)7L49=c4rdI04g}*cd!qM(v+{yX-r=pLmyGkC
z$M0u;`2K-^`CtCbzy0{aU;fKaJe~s9D6BKw?gu&qrt=xC4A~lHXR$VL-@X!W8^&6^
z3#i(;zuY;0ICJ^%j@BEgR@(h0+I<wWb2#tte&jW`#&HN=-ztQSV>PGqj4vH}K~aQa
z2rN%C?%@%oCz=Xqr+9q#iD0bQ#+71O7mS7$4{UG5?M6H9c;}hsnWijudB80vrl(J6
zE3=ez9Q0D4&4fuZi><YTBw7V0u`GlsNO((^mnOxC(FQkpmPt0s-aFcfcKz$;-vm9O
zphS`QT4(Xj)2s9!TdP<jAu1!;o`$Atk<TgQ#7z}U$G&fLB}CeCD9SDa@mi~JiM&Hk
z0c`_C88iy79d5aZN~CLYQ5;pXXQ<ZTO<-D9tcJGjXcSfh2xh&vPH!6RWRPy0rGpf{
zBXYm(C=Fqj>vWhVre$WD0>+4nM}9qtCatZ}hLbh-CTty}p*3=qTU)CYtEJUp#$T!r
z#R>{-En$(2fwtc`ZlCG<opm;ZX-0*Kt_@XbN=(F;XUgp=*abx18GN9>`-p#hBupoa
zpD@-7;$nG*xZ`_68He!;#!eDjTRN&HRF4=_@KYxsMzLS-H{w=_7Q*7tVM4nZYZhwB
zc+>I0K)w%e3o^(Za$s#N-a4%JXd~@N(;Ir+#=e2vtTkvQ4rik*#yZ?EndK_2&w3=Z
z8A1u8X-ZD^T{ToMa#*<D=(WC{j%)rFEY-mV?5fd4ZV%Lg?Tub!p4W;b^=Tu2zE<j$
z*y^T$^%J=^T1)`)yBH%(sMf|z4oYi7jVGcL>*Y2<liwqygIp4B;JB4h(5<978YgoO
zV`Q&sNOGTcsHQpM&T$-oWOwA+u+DQ{R~BQaF%q|3!k-N2UdVeS?z^C{wIn9iT3=&M
zo3b=_5<ijpI~BLY15)C?36|Yzndds&(N#>E+t)%3LdvaG-g&J=sPYdl3QI72`r$hs
zzyHMb<%PQKOlu(CcaEzt*e55-fLOuA9i=iKKdfM8K7T!Ue)-Ddy9;5NFsTyOg|MtL
zAg+yd9R$sBnw>)6w9eoh#m>C@;ZMx(KViHUrb+8ufBQ|a$Gy?^9qm1Bk6e$8nHNrv
zk8}f6iOf7to|p^ti6ic$+HliEI6Y9bLTL*aQjr;eq0<NV-z)*EEn80Hl%UAq$=Ctk
za+c}xggZTAmlLWtDh+EjrOB{dYlHJ*?-$^%u=<q|x@A=zXFOpMd_*r*PO;hm9qT=-
zd8E`vJP!8bkREsg?Iz~1VCDyLO>Q?zt{nHB<4CV0O-w+UfOQizF+ia@TB||h>=>g)
zG0~JVhF$B#biNQyXSBBXV`ujcr=3XiqNreMg{CdWO(^Z@Mv}#*FrnOmDhb_(iLe#{
zi+N%BfXFo4kUrLuQMzNsfY@1!F@|}Xv0Bqhk-=vbFGX+P%tBR=3n_9h8SfnPvY?a}
zB~a^x$q|CwpkRbF#e=E^8&L;9?$~Z@_g(NKZl*Vlst`-%dfd5xe&O@qewG=4vlu5g
zTL0h)^NIEGBh&c_<0rIs^wKz<E&b&)>DW=N(vCZ&+%b7WIY;n@;2IX08)ym(5f^Wi
z+-crXXOFSdKwDIC7b`_4h>C}elJ2B2tMb<3yu}+U;qb$trbqIhJi8I`XtdzR>1Y>F
zK7jM`9gUG31qZErA6WucupIUq3O%5ae*96Yu&_#^=7Jvi7bua$XrlouMaMdfR-~LL
zN!pI7i(zY;17^&ijMgacaYj>|!D~lvO$tvE`F4zll!e|zx{UD7zKK_4P9?Xsc@+Xd
zQ;H5aKM~dwrOQdbl|=gnvnv^8g#i{*l*B(JpI@(4<_2_XjH4YXK}B7G+)WML1~js7
zlq%WZ)=%hZ#m#5z`90Q)A1lX$o`=X{+c>sO{4Czc^Ry0a6|}-}-^s@zJ_%>>cBbh@
zbpM2#!aC2`Ff%=z=@hi>%%=tOc&3-azHKa%A$X0RR#eVBf4+nEXnim(YCy!md&lW~
zW_^5OJ)I?=G&e4nGrc4(mj@OX@RLUePxp@X@dFRv{~ohGpj7AK>CW@=BiolR^ePTk
zRA)Xl6o$BM*baM`nI?}{a#A!@f&mpgy%ster2|S?8DeP#&d{Uc$dP=zQm$WVIid26
zd0p6EUf8cMSQN`?fqr0|XPO`I%Ne&`@Y6!CiP|Dq19<4;x|b4W;CG!FI{8}V<t>Fe
z<fleEn(%MQwa|+EWe1*`@&8cY%txcIJI#xv%RPVP`0^XueWUh@3lqy(VJDHuVGSj1
zVs@e9oG@KLvX{RZn^5A$Rbj<WXOs~PuT~jqV?S;v-34RaGck#Is+U63(2DSnzG1R^
zE0P+98cK>xRa)G}#=&b*C`f$bg7bc0fgRRJU`{WMrob2t&eEHOF2z+3o=$6>7!$o`
zv`GS9GnO#P-%Ep-<!8RXTsihl=22Cig=!*`Hh#iSf$ALh?x1L{Df0Q(-?;zsh2wT3
zOrF{cP6ej*jPU_oDz;Q?DJT+J2_sDH2l;v>ZZ9ZR2Lhy^?k_MDFsrwymY`%3u<en2
zkGN^Ygn${LH7JEn2~#7+2)YHOXey1wEiHZz6izE~Gi!|tUN*NXsP-5Sxz)B2@}ngR
zvb%}uW|%y!cLC`Q<x)Kk!I{@g=o&w18SdMF)?Pdh=|;;t6a`(9>~gNCJ|K3|(8Fj+
zfJ?<n>rKh`vXw$FBLdYLl!CogYHzIPi`*Kjp`=K!g`5Sko}+AzW~iQ&Qh4t~QmYkR
z$sy2>TdXpojx*Yf!;>X$J88ekp-QRORJYb-Ft4n^D#>A1qebeBCguZO4cg0XqjpG2
zup=gTOl`z{BX1JMrM37H+_Yk6xg{9u<dBl(^HuqfHk1)nnpOfo)d+A;qfsr7Y-ho#
zd*`XO4in!$xG4BJP~x52U;oBB1<bs#%oEGHGA|Qfe)@^N-zAH5nqXOZI6si?8=?Ko
z>1k%x24@AcuzD#VQ_W}%)6>c{PmnEbzhf0xZOB<OpPzX5@%LCS+I>|;k+PT0wRCRx
zot6uamowpfMtj4gJ6jUv$bCDQjX}jqm^}0O#2QX4Df8v!1${oFrwN0G+Nq^bN`w??
z$1axfmf5coq7mFg)OQ(RYm1&AV7cHv{6GscC_PSE0s4igBH9iVanG-+4r{HPxaUb)
z4!x5G!BdU4hx|-UW(?^d9RlMsPPEq6`+@Bh-FX|h1FNC>K%FLzXK5`d)!FuqWtz~Y
z)8fwl^(*)1-=IipnjN7v-YQOu!x_!}C{*R>PGkL0Syb6<@7tY{B4+$P#yKWGar7)a
zq29T@+%YKfCM^x48me`4@ARff>5$KQIb)UjJK~{Nk@P8tpC+bh9+0?U^cn<7oCl|{
zJ;-oTNp`Z*5b#jv5&VPP+??PT#8GOgTc&i)7MnPPOTvE3b=C?yFQ|E?*+3k6^Lo8<
z|9s{4%NJ7Gaf>BPa`O!jCr*!#7(KJ?8};A+l{o~KAx})X^77kHJimM;mxKu-z1E}6
zx83jj{PW+aBRgUap00&9s*JR8?KBNBMHmX3gh37&;$>-M46z!jnb5sRE<x$|c_myf
zBD>TbJpyhzouZ{kwtC3m*;<MVR6Em^m$s{_SnY-SnG<P?*f3MWh?jL>T%fu@sSVtT
zUNzlltdg*p-XpE-nAYVMY%R6P^VCCcto@9(fy5nE5~V9tgI3c}v~}F%ncX9xF;)?4
zq)`M}ZH2$3wL!Tuocl%&B)Q^S6LdreWdp5AhJ+jW9<?OOu~Sn<_i?kRb<k(+P_tRw
zF!d=Aj#WCyWyGv{k2PL`rJSMr6HI4JN$42qqFa<rwpVnocm*~c)I-u1({?2vcWIH1
z03TxnAnEJ~JZe>1!&;bBN^2yTP2TI4(m%c^tjomZd}6!b`1R{o-akF^@Vie;>%!%7
z!5hujUw#H_r3It4P%W+J`%gcJ(&+08HY_ZU?^(6Ow9HJw8B5cFk_+kBP&rGFaduoj
zeBku)JJ$0FRWof*<g{~r`O0==p101AKW$>v>5e~kg)@qEo^Wg6w=XX|KR+`YgYSil
zJ~O{xSwDQo+&kv-i1q<{K7sd?T1cBDa+M^>S*2vO>Y{{b89y(apWdUi<@EHP`SOIG
zPPp|<t%4dB<CnBNj4`h|`Cg@WDmvsYZ5u?r(zQX`A;(Oa5+$^S4n}pMi?&WFW%P(G
zT5Gv!C@E(9wzF-U<olL{b(ZyfqIbze-j6$}G~B!sw~cgrkzQr1C^JqzrC=wILcaVy
z7z!3OE%+eD#bet@aR)ly$;n+3o|<G0cSt$O>4nCA-z0z6Il?*<(?O0q$FWmuBMj!g
z)?xyyReHdBFkWb7@bv3YdlYgy(8}SQY?@75dW!UVVD(U+q(qMgrHP`e0cLRVza@#q
zlA#-d^C;h6WrE7h*oP<la-nEXQx;PrZNHAJZjpV?vqw8km`(yFJg+RL6B<uB?tK0F
znV+{Eib!f}Ddd<qVx*MByiTHmTPA6B48=(Y3ElZtz!bocT=puO_TKTq(VL@}ot}5n
zE=r`D5;aToDNWQ@#u=>EI4$ZmdZE-TCOL0e{Rx-D2sx^hcu?7(Dr($m<v`Vf?HScX
z2#|ZjT8D7~Ydv&n(HSM|or<Pu2L_6+;_s-1)*F_J>q^>ntx<-kz!=$ZRTtWAO#%Td
zCCkq?%x#k1C#J4Gp0U;rA3QNAiG4jbVNJOJ)?=*PUgNP5(;=zV#-K2A3)Olg^NE+G
z8$rG+6si^UFs}732UqDhG->tpj`l0T53;ZiG-HJ8gmf!<j!=?ttHun6CUdA7BiY<&
zM$-CHjEIa|(-8CvO5?(epB6E<^@bWMt8X`S-3ru3Du-l`{4>}`lebLXpljfs3(udw
z^7P#k+KT+k%?p>0-%)ZRgqau<fBpBrQMZE+@1AhM(oh(`vOInSrBK^fT#WQIZ0tv(
zrb6Bhaw$xYPn<t~#6LVxN+EsuiCQwL6~6rTnU~wa%hA}oC!zT9^Nk<=_6y&Cx-cyZ
zE}XdSnd{4)+kT)+LOIVFOECfuJB5j^G;W&3YF`T}j~K)#8NfMZytne8jWB914<DG$
zA8?mPX%!hOs))FcE1b9vl~S(?7;hb_OaHgGzbgV1(xOq?096cbN{z7eKHz{3qeY38
zVwAxdjn*n=0E|>0y8hZxUF3?G32ha}zKP=D`N|Q00Swwe2nIWbwSBa6dTB&Dx7$vh
zJUR%uQ)z=zA_J{i(A(3{&6gslD5Dg9Qmo4?sD^<UpQa#jo6d?uRx7p`IgSHA@;1KB
z5&m8bq}`@krsPJ}Vw_?mmzA8bR!||v`ym6|sQZ<iGNi=5@5E9u^ZII1>c;`S<E_C~
zM{OOHp?6Jd3hh?hX~vwM@E?AMTA!roltR5fQ?x~I8){uKJc=xF3fSp_3lrLEfj=Sx
zSZO0=ZucAe%MGI-RVngHE18!V^BVX*&2p<O1B0!`$<<j)8L^?|ej85zfG(X9@8s=;
zt~9-**RY-%?}92Be^%9SyS-3a<>~Q}d6M=D!DG+@P^%SX4aa^EW1$>ZdOYNy)&%RW
zt;Q;trhuDI%$IkB<wVzp(iABP56gN%tTgTE&dIz;2Z0bO4JAUa$i%Fpm7-K(6De(n
zbQH2yq}wwoC-(cDRz@~95RWt>GDY2|j1kWTkoP<LmtQeTv0mQMl!WNiB3KTQy$;f^
zk%Md<h?tU+AS<kw0dQCJ?!^$Hlpyy@9qK6=Y%A;Kb`V6}rK}Vb?Hnmq_Wcg~1~t>t
zE>=LTr1*jC9r2FXS26}kb8Ee^<t$QL=g?us%qNWZV+U!{PHv4JciMiXr-P;{R!L&x
zKfEYh9xtR#@YOL_a%ueZ%WvHF13v|R`2Ks=%Z0CBzS2@A=R%5!-+uaq?Rq0D6JZX_
z@7mZ{I_cPnw-@SBIQE^An^fprvD1t{eZZbRkd+an^nd<WUcSC?bl8rA>vqsJgf)=v
zil09J%wK+VtdnQaf&2Et^UIZOPt+r!%7I%l&-+2&4~fomPUM1!ZNJqd#-(`BOJ$k@
zswJG&xFEe!yI#0_`~%Co@906eOoFKx^4BbzwO|ZHgWTIUUeOq&BcQ8VB(GQuNur|B
zD3p_vm1&}uDNPJ4#!q;qvBTy04S?1WAei@#U}O`~#?We^caiM&k%fC)uY&hp19N!5
zS|?}-B?<-I5Y~MDLYM-VcN!CBRLYobBjwDo-Ow~quy_xtpsg5=YA#sYdAO{^m{46~
zwDhh^tMn{*Y_!NTYnS7o*58m1qsVDRYofl;D0(fF1fb;vR}0wTQudx254M!Je*R3^
zHcS%(mi7~il|HhyqRpo_;=Yr&jhd^3$J>eO7pBLLxQCB~%X{4U2_0s<(U_XpjN%?6
z)mXf@sJ_Z<?cqHv4=8P@$3{)^-89pRYKmSu{$iL<@0fxkBj5SnDpT+RQ*##6z|AMl
zZWem%@VNAnh((HK(;amj#QRlXiQ3Ti8-2f#$7~=>6H4`QNJzv~IGRLhc3@q1CaVeh
zEWEgHy>?|VbeeUT5Ye}d`)#A7T?~;LiOEF{F=3j}Va2TrK1}4&I8vk_2DVm2-rAL%
ztUDsxY)S%gTF;bmYtUNeLk2xuynqI6gA7v3LF^6VBm+V#DLzNeXe(HeFrP(ooeC`$
zw(A$-erG<N!B}!{l$c%vY@GL?%?Nus$XWPV&N(R#O4AU7Vlz&@lO5rLB|*YFYoSU}
zu-=XXftb-`sNKm~h{SP^V}2siZTqbqIfArWNz{^Hr~;MNayTuOlnW?>^D{QA;3jZE
zhRRkHRl0ZTb|uv!g;{H~wix?QxGDH$qV_`CBVXcywG$b`mzNt;ihO+ko@ts{=Y?zf
z!V!0jbIj{H`gWaC8|gTRw`W`klA?4!s7GeI-JmP{vf`Hu;qe2@@BS0v;~!{fa<Lrz
znXk7SN0H5?){4^*3|zZBz?XaGr(bW(M<W7Xt{c~vUF_@DpoV)@3?ZGI1dn0pI3MI>
zO2g2l$2ovxJ~)vSE@#5!BkuGBHjE9-ptRtEzx5s^07M?VaM2M-=s*?-;7h&$z2=ce
zEQ44VJjMiZt)_!*+sHX#202mu1q!PzpolR_zHpa>Id7C0ghmU-8p9F->U6^Nj(3if
zitvb9C+AA5NOa~fh1Dt?7SkZ-3c8?*1k>j%tsH5o@oEUOR+9PRcBe!!c+DZuN+TYU
znr4ln8#(b>=drbn0l9g-5q)d58Dj}y5-8np0QZz%=P$igdd|fAl^A#KFJIZ7Ur43X
zjY9>?bP0sf+bvCKmT1XYIJS%(W6r8NdUDL~z9W426K;LPFDottPz|dD66pMdc6O+>
z`e1e`bdv(zdxs;^ZdY)M?q%55movsrtn<v%!v$p|uQ=`-ApaldEWI$Yjzvb8j>xv%
zDcKVDovsu<Omx-B`$oE7Wdq6y+cLoiiC>J3d7U|(7LvCd_rJ4=M5R@^Ewg|by<}l1
zrG!7F?fAY+2#;2pbh}bU8^anc{q>UQ#-n<pq(rTiSPEOrG%Z+ogB{)oX>HV!>E)p1
z153iUB6-kyNP|&wa~Zh!E)QP<Wx|Bl7G<a5NvUFjYfTdLY9DP!VO-_wO#-BrN2bR|
zNqj5?RmC`J!X$xE#<6XiJc|y&dD-}Dqp9I_XoZ;SkkeL-ZWs;0NsAa=1-N(|bOpvw
zgjtGb$(j3oBkl*PRmoZ&7=Epd91Kg8*x6d7jikVt38S_6b@GU+ombqniW+RRoQ?BB
zj2)^u)f;8ML)xXStQ3+3_YW=#&$lb|&Ub(OJ$+jF@^z!3@qT9CZv6Of|IRWmasoSc
zwq4S*9-bbVg5w?!g7d`v#>?klczpMsj^fCIqMqiJ+B(zak%u3C&-~#BrtklRd;Ca4
z;Y*<{E7eJE?Ri?+wnQs;QWv!JvQAWG`22FP6iB`B`T5569(jLx;Cwj~<{6A8*Me8l
zk1HjTqrncDS#K3*3>nFTRfx-QIX^HzeMflsfC)3&TABW907YsN%q|nn3cAZfLc<6$
zDW#Hg<cL{XQ^!t82Ym$ExM@M3R*aFDKdnTy<-B8>f^52br<X2a?4^-o674;TTsy4_
zEUb<eQwTv4@h%tEdE$Jzh+}oU$SDq`PnAf_GBOF4DPXOmlm;Y}>XazB?<IJlG-5z0
z$pFw!!pW_nCOOF)XNDVsS9h$^nq>-XDPpU`eN(#hPK*b=gCFso##oG9q#qwQa+TAy
z8VV_^jVQKCLghrTmhkk6syanA=H-Mlp0sZxTt^s0W-xKO(?S>+4qxXM{Pc*nf@L<U
z4WX<dDMu#J8OEj%pk*<c=L3@*9ZTPC^q0?c3u--qakM0@#A#Uw%Od#gi{%Kjv_Vt{
z1>Jj~Q=)7x%2teXBl$}rFpbu1$3{tsmWKRLF*1$CP&Q2?dNp`*DrjXS5NJO}`?0;I
z&jlrY`;;0=7mS|;vSy65wn`(>$=;22jv~dKMxkpZE!})0=FCwtr7OCduo%1&tO`w3
zMCI71@k)y~3<VQB%R1w{D36qU*uxhiJfhZ`+{P6c>ruuJI)KdPbQAY*ABr<$49l!Y
z0~qM1l`sV;ks2edbzHFHHYG(cB|)ms(<H<7EODE<cS_pXkIwT}i8<q(;dGt}mql(A
zt%w?~H@1D3eG!kEEJld}t8_h7aS9Yexg<A4ZKTk4TToS`#cg0&Wai%G8FwpKo`KWo
zO^l~~M1u-yTce!EOcSmwsFKBhqW@#t{r{_?khjEgIg?vuIxSqV-1fpcOEv%L=U+Lm
z3ra)09;DlySzG+NFr6>B-eq`c1F<#ksp750pB94gtkZ>RCfw5p&fopOyk6+qQDUUi
z$=ilQqnjhg9d8QWR30B3CEfY-v=FO@l<5?nZ&$wTjh{a6*cSOupB_11X6DCt*szF#
zrHO(?GOdNcn2K;v+J2+lU&I`8Iuo8guzdUjm-pXcrZX<gn1OZH3VKeY8Yy|Bx6G^N
zzE%{Xe5k!j&%R1BU2T=rDzyo>Jzig+{!Dpz;_=~efYmChHng)CKTDs}`2R!iv&ay+
zRne$7$VH)ylMLc2gF#~y!Fi@(plaHn+(c^~<09>L8_4KyI(vtA=Jn*B(uOmES{sYj
z^qk01+{`Wn>23BdtaulM3JH{2=v2b$rQfZj<&aBZF+;-VWVqL}6wGf06)Ut;YYuUH
zr(FhWrOF`P2h6hK!z@M3X!JBo8>@F>&0sCPDcRJIMBHwa(plFNUN>y-bYqCM(T|N5
zZ<KV$J44AG%!ESGOQzNgFJEvelB}iAD@Ijut{RQ2m0A;#O16U7FfmbkqGuVpwtc1s
zC#{hphZ$)#io_BX<%Hd;B+k`J2$M)qwaguANi@~D-=D!KLI_M@7I!`1Cyxsrr!~RK
ziym#K)F@#(G-A!fUP!&r<BrZ7R(G5;=w7h;GBo=g8!U0xl-My%2D+^)&{j5R>&Us$
zwe-P{qhN!=*oX>3HSSqLo62^j9yc@vW3>!$zu};3Lun8>fE=P)hi05?qV&=GmzIqp
z(5hr42t!v69Nv2fiol6QJLn2sHn!W9ctk;N+konZo(uPQr&SSUR3FIm38OSE?v(u?
zZH4DAFWjOO!D%|Pt`n!_gjMRb-BYxfFI$6xreN7K0s-_Z3PrlVpy`;dFil#;GW?dk
zgxVyoHOJxKkU5$*GV?RqOY!cR*$1q5;*rQ;Mj%($20fp#%2JOP88X*@!cAej@6vZQ
znwUGk{`!LIjq~r{bH1EuxpG<-pz+h^XG$)-`}mIK-6z81JErx54im~ca*7fSm~QA<
zPPsJ%)(_ufAD&PlNJ}m!>encsv@gFhwa8>M$MwdSUtcJ-<HN+erw7*4BH*XnmC1DO
z`@z2NEIM(1Sb6&Jj>k{mF+Dy?V%wms8X8s$YeWIHB<|N|^7RD`JUqQ)dH2MJAO6VW
z`|p@e7Z@Ng84Ndp%N-l(*eEHB=6Uqzg+niGqdqM6t%`ZwjuTxsmV-`?2QOc*P$d;B
zOw$<ZW=NWD2DZV9Cch6rm>;1twKlA_xB&&LO=9{|Nsu}vWnrv<7`VE;NZxx&2&8e!
z@}oy745}u#73Y|zWnefO#}Vaa=%93d4Z6v-;hh~+(Jnx|Rx#Gf6;|odrjnsA@R~F2
zT)<}|(T@YpmX65YucTwA=8SVr7(2b8JG4}&reS<wcA_3@MN+~jqR!DuZa3NoY^^A-
zamzxZP)30+DRSHoj^=on-=XwK_EHgIB;K#2eIwnzU{w)Wt+Vvv#sD^pkbp)>k+^R-
zqeLx*crc7rs2Ztpr;bfOr6|L17|=o^UXd}2(pvHloKghuG1^d)5hGY~bSpW+QwX9(
zA2SA1J7quE_DU&*c;AV)jW~=_x<lM|ZdRazE?Bz2l7!QQLphOs#%<&F4U28VxbZp3
z54Ho!Xga7yN)D?<Wmk?(5En{OlgveWD<dt>u}oo5X){_&rbDgb%R%WO4;2<tSBesM
z`Ps+(Nh^(_(t8_*w!dRdzG2Ue38a)HIk2{oCq7}kmD$*R8#GuX1umR1^TU`O7~)Yl
z0Dn61?yPa%ak(t4rv<CWJg>@OYMSH#u^(~+&vB<!dFNEGqFz#xn)c0$qv|kQ!fU3p
zroEp0rOVw%HiX2iV#KtkTgUW{8`&o`BS~{cqrJnaM%61OB|(1tql&_}0FS@^?Kd7v
z=l%B|$))nkFQ1v6p?rM6EP-0Hq+?a-PdgXDPT12U{^<wQau)4!>y&<{l!hvr+${07
zQGLT2N9j<nS2k|+QZYGW@&Wq`Yv_FU>4dYDAAkJ9{d#a5iHCO=yph0}^W(&2a_n)Y
zH2I=DJUy{oF6e2cC`H*eoRJ5RYL%;7>)h`<*V~=j^@Vb~;!#v(u+xeOXEKJKDyWVk
zOTR8>;&x@ber11ohE~MVZ~k7Hdy_A&*7_BmC2ew%AF5%$H_nqwV>=@Ep_)nQARm#^
zJIfK7r-{iqa!HguVpS){%zi)E;~}}Nr?ccl4BU9nm9p=YF%PJ%!*DlS>m<nxB}sE#
z(xHx&=)GaI##@h{96IOM(2xDTNl&@8SDD<n6EUwR=2_&dtu;z5uj&_V6eWwo=!g=W
z(`pqbyHl@d?N%Ep6=IB}lE_H_Yx6vb{XZ32y@GWt4^OO5?+K?B9SkZZ;+DAH?sO{7
zIC_WNHO-3YuNY5Fm6#)jMl6+=mmB4ph~~t)SKPW{=p0{u<ND)Yxc~MOF2MUw@0r&L
ztvlt&(vC?HYb9HEzi;ff7qk)+ng9u9@Vnoot=4NHmMqBTCT2TpEvh!~0&uIXlS-lX
zPH>u{6@^S0*l%U5$VMCF(qBL8UJhY%wL+^&gfbOk%*0$qf$nT~k)68W2+PE2op9CC
z;*Q!Sd!*OKec#!(1C$eYeQl_Mv`3qO5o>EQugFERwd;mDL_Thv6Cp-zQp}Mw#atUT
z%BGlxiLKTm@tEWLB%TuiH;OWO6!DSyF0(Y*@FeF!VYIY}v@<AQ$uW}eyBI(H#4RSM
zmAKv6o?m2h_7koJ=AdvAA;A$VTW$iyo17dT?NCD<8>zKHHdYedX(W`Xw9cM)bS@a@
za7GO-Tqov2Yns{=x*{#OR>jQvtyq*o>5_ogx+11RF49q;NsOo;Fw0UR!CJI-81DzJ
zrOW(nNRd-YZ2R_q+Ry)MOZES!{{r~8fBQH7_{Tr~N9xDhdRuSnZN07k53j%c<uCl<
z4}bWt{`>z>QE094<BvabKA&G(B5&($y{)(Pw*LRKdhcAXSAO@q-@US*{_oa*s3^Rx
zxAnH(*4z4TWBtd*(QoT*y{)(Pw*H$~Z$;s4y{)(Pw%*q3dMgTV>utTQxAnGO*IQ9|
zTW{-ay{)(Py55Sy+j?7X>utTQ*Y#Eu-qzcCTW{-ay{@;S@V4I8+j?7X>vjF_V9<Yt
T78Y^Z00000NkvXXu0mjfXRoL&

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_003.png b/0.16/_images/sphx_glr_plot_visualization_utils_003.png
new file mode 100644
index 0000000000000000000000000000000000000000..ff54ae0124e402f7506a84104e54e5d509355b95
GIT binary patch
literal 202016
zcmYgXWmFtNm&IKM3xm7s;1*=?32p%bgS!R|?ixIhK!6|v!QFxm4#6$B6I_Ek%YM6O
z&;ICs-KVQwb)D*db?e<*F&b)$xLD*^2nYzc%1Uxt2na~Xuh&-?Xs_qd#B1MIBkCcq
z@1gBt?cr_aW`&??=Hcq-;^Am#PU~gm=5FWW%+D>%&Cf||>*3+*F2=*-^nV4oUEFMV
z#-wHUUY%gNDjB#VAmE$*_kma}Rb+>Nh=8ChC!_Nz>*z`Oi`7EPF7}ho`??#+fswWt
zzb)(c%5UXN&v#t9!__;J=(N$|?nM02nEgo<XwQ;-iPKi%U4`r1uc6s|6Ag)qUpvb%
z!c&KmE=e{Cn?8Dr+Xy?9Jd2a?^LJF89zNW6{ErcCy@X$*-~R92e+>NO9*W~{vNm^f
z6%g=h;Qy}Lf&Z^W(AJ?J=<Gj_{!`+mf9}S7{r^06JjYj^_I3Vm<yZ4WVY2o)<o~zX
z(Lsv*f9`LtT>Se?E;9Rpj{`5aG(LfW3UbedQq%wAdx4K1YGQI%2EXUf36c_gylLNh
zZRz>2(xlb<vDAEB>UrR05dPr(a`NIrbMl{;cv+hby2XD9f!}4q@2Z|afmin(-v8Yn
zc*2#*qIjS#=<M+kr!(jV{%CS`cflWYyT$T+StO-pxb@WW1^jnd-Q>0U?#qvx9O&KS
z9L?R*%fd@Ie$eA!ZRMpTFyJh(_20C5)#J3(<H)z1HP%AH;11GmUT(XMd$X<cJ0#Hw
zUzSLvSTY$_C)ZSPSv(-kQ<c*JEZgarBy0KMy<7x<{@s`@#*7~8apy>9_$+z0T5K+R
zFqqWdj&gvCBAdz4h=(TTUAV{|EWb154WP>-SRhi@&>@aHj*d5VwM(IAFB(GznN9Ay
zr+vxgA{mUqJWg}ctgQg|kMta>GWO3{1$ot9G!a{Yw9Ey+0KxZxyU$**n^5kMv-*^8
zxR@cc=;X*bk`fn>m)+NoF;dTG!bWrW_0V5bCU-?ax9T$w*|0UB4$ej~75~Qf_ilPq
zLclGBXa$YDtNx{>v85&P?Y&|&phAVlwb}$uK2YJ1s!RWCyr5Gr0BB9w!0QOGFgKo4
zEF{-Ku$VW6uif75z7Kr9i`T-Y`N30=-h7IF^N>J~LUOQ|It$R{UP^;MAHug=oaT8U
zt=a?G>0-<gLgb>1T0@);KkLP%);i9obKtIR`1LHa+uhqO1lrx?%ZU)Gu*4kx9LXvP
zsR~X|vb%!(!n?u-HJ&g-is6)Iqo+GrG}eOLbR7G#dXRLS7Dv<5Sl}7~%Aa1_w2)*s
zYhyzCuy)GjHFT*LpWULMPunlILAO$GSNqn6#iUMO?t?DiV7J^w1kDyq=8!DXA3r2{
zRqoF3BX~@Lb^_Me8YWxeEEYIH-ehtFvrKiB*dV#DZRR3KA5*hoLj8NRk+?95OY2KX
zfHask=n;7?F(@}$qkkQWw;0HB6()iuCN_!2s}Z4#{28o*t*)+KP|sUF34t}L9fk?T
zP0FA)I++bquql8siWS^ilBU3-CvAaGF^gVc45OxaAPVLXAp?O@zMQTarM#{tK|k<v
zasR@PEH)1ZbH<sn;k;vQ;YPe^DI2D<dJ^%WWH}+s5TBPWQb{R<7)exqsy0arEU)kO
zcmFOXagGH%q)!t!yH{au!K+Zz6nYhTh1PBXoAw(R0bnYrM6-@Oc&)Zh#`|#`-GKA>
zzfbxy*sa?c7&PK)%7ljC^cQ=qvy#+x@ouFs@TjXxynXwz*+^g8Qj-ojImFbGARo;<
zP6-Ey2m+v?MPpbil*Q*^*~ysyypef}cMxb1<X3dvJ;{Fa&~<qDKEXm!E@oK4P){p%
zW|pnNwRNqeaKaZx)&F#6LlDNL2<6Tg{%-I!4Zy%2icsMV;Yh8W=G7W%lL-2IcNdct
zz>|{O3wT_NjQgoHh#8M9Qzu13XU?vKVXL7v(wiF&5J|{>d3w1Hx-nV)?<nBf?KiGY
z!Nc=jGU$2m>JB1;<%xq7>VZLT#7z<Ok2L5n&+~zz<nso)T-wspSPc!OqKq66jhsjf
z=P=tz_3_MOqs6hDhG&oogyQTCzh&;0juy=WRiC2wiJ@E}n~)Ibe^VJEki(#>{u>9_
z>xs7SO^v3TTI>QwFrmshI`<Dzcqh0aC_pQXD#tm$kxZ+Vc;{oh8%xJQ%NFul1Oh{&
z>5Ziy>*Df?@pK}JAvnn*I3$HpA%mUenkU)w=w)mnBT5usNjcH%dFt(pQOs4Iv%kg%
z>M)|9Sa~qyUh=WRcu~Zvjk%2)NhRAJLyFX;&IhEPNE}+KX@swHpqM@rPDzf^0&-@@
zh-qm>{!Cnu)!YwgWvr00h%5&6acp`B2BaiDK)B;sXW=w^GGHqnl$0Ojtg}ylxhfOR
zRN?pyTa-mE;EP4vT|F7hTkG5p{x?mKD;k492ECKEpx13Q9oPPl)4_ZxpEQcSz$}{a
zE6LPm=G5LZ+%9}dW=<_S`)QQb<$_v4<_j$%m^2~4wUxB;tu}57yDaT?uRF8`??IDQ
zk)c;Q4tr4ABUdH{6fCWjHp?C#Qj#2_wA_g#77o-CP&AJV1QgdaLdHnwK#@;%-v4fD
z$HqC>Bx9G%LUW^8)yH`xqyp9(X_PS1F`{V_cmcSB081RX@Ki#OqJaE)Z8G@MS~eJn
z0mLyCAP>dqD}WJpijgz$@X-8f^IsQ2gHacIf)i1W!|}w(#l+|^N)?Ec5MI4Lp7rDa
z2n@wvKdYw>ScB{#RO_=D26<`{{X6LG*?ID}A{vp`Ddmc!X6*3cY?mHQ3#*rRjWnDV
zqQ_KB10)ZnF}2_|w*lf1sie|%(`twfel#_|T>ae)LMvb3)s;i$vT;m@w`wZWeX1cu
zl3_Ts+SnSvfBiT=j9^SS6lH_})4MQmA|_F^X<3Dk*jRLsD{v2hro4t0aUl@e@VjSu
z45ZJv<=ELWGo9o_hu?oY&?{)bV?557?L8vNMCE@p#yhnpX`7E?f55SfD-VqYMCLL7
z%}P$?#e^1LCM<k<gwRdkdKl0u_(nw&IObZ=)=pHo!*$&BR-tPm5Gx>t0!BzZXZcQq
zN?E)Gyk-smyWj|AzRBE3t>t;jpy){0%Bg!(a4I4LeGfK_lHLKTKj--dIn<U`?uFa<
zkpRE6fN8vrE~FbEBd4!L-=~BVYw%)oy7|&{RTdjE-~y#*Lv>34W=o1OcYhQ}SfZ-y
zs^yXh`m}TZ*z4Jf(0$x^adjBqR3*FzusHi{*KM1<yUUSdBAF|*uPH-vNCl|c3rjL*
z-0cNc&kc2)4o(`lnHs!73dQQo^$Mcnm1)tkr`sK|Y6gH!Ev*EYqQ~roKsqs6F>+`S
zwTvvHx#OXv4bld0#_0DY?fj9GL#8-%d*B2P9IHzOE^*JJy*A9cXayC5V_U#byS581
zmqKEihjFsJozQ$@|ASM}A0c5BroWx08h)_)Zb^0*D@8@7zrh8A2$kyN{+|Ljw(pZY
zpE_IK!Y~v)8P~mDSnT~!ofz#3unQ=GvSt_tbXEYnVho|w)%aCZ+?e4bsD<0*j#t^G
z*H#DIJr%;$k<1sjie^C>?CrwFl!8(n62D<Q28>1on$}AKM}cUXsjLK+6MjVvSk?uU
z`fr$cu#u(dd+-*6ixFYTZuB-xdkDm#Q;P@=d;ZX`?~q8q#0QeRWWabJk|;+Zx%sn%
zCz_u)5T}Lgoqi{6xROf9s`vx+V`l*%6kd)|zgpa2ws{&)sH9nw7fj<O_E}qJ0;mtp
z6{O>eKYK{Vd<-&*G-saY{f-OR*mtuUCB63B2$W)Fnc_q1Y18k+p$MY|r)MXGcN&N_
zuG+2OV<F1PM3Pdrd-exUiUmj$>+`7p^}Kozjawuiky5-to&`wVFsDqb&QA@s7}xyj
zERIGqSBfUi?K!jYHob(_tZMKrZ4V7kD+(|Jsc=Ikce~iZG}&gW(xww2%TFJPRNA%C
zc3zH^ej@HX6vT!q0AN>2e>%VO%~CdF&=-?wk}!dKKeHq9!V(j3n0h-eF3^9NNTz^M
zC@u|)E5xSw@P#PcPY<bf0t~oGxz*p>J1nny`Brq;lmK*|PBLG-DMCvCn$eI6C6#0*
zmd4@!My_I9OAj0i{Ts0Q%#AV0Y<-eDb5SoLgA5%SQ%Pt@DIJ<Dojq8GMT{=NHY++}
z-A_mK@d2oRI?pSBq7z<_kt%%NQe93F^n5YCeEJ|)B+kZcJ`&BWaDYNlucfa|%O<DW
z$W+G}l9~!Y(PZe8o(T4cmMxJnlRiw7B#$Pn6AtFX{%h8QEraQCXzSf8rxC=h(=|d+
zpoeMhF#}(!(ubD-=V#LpDI2QiLltZ*&aZ|kq+TBT19d9Z39{2-Y3pa3SMWLleR{sr
zl!_`;t8t+l;)<cf<Mat1F78=OppQ^@Z<Ny>!SW`f3G7|kj=kl~VleqnA@3??7BMXp
zRRJM~E&u9`mD$X^Hbhq03Qf`2oH^&>0rVA=mxj%i6o)PMqd<izX?|Hb4a_*Cfyq5&
zE&cg(?8EbLTnqS0uK3@I&SDkTX`pp6lGDUV+LI?sW3Z%O^ZSG_KTyX>#1ho((_Em>
zVl}1<B4xYgo<52(EjPGBMB--SF6LVisUn?kVMEx`F?M!Xdwb1X&V<G;B15@o(zkH3
znJK%ccqoc)G+U@VB_Bzu+*pj2a&K|Ho~@?-tQrpb!&9~|@mHq1&sLEc?+ZW*QS-X{
zqcop%WRO<;`AN_Sv+VcU5urwX1--1cfVEylD<yg^JW~8HA!j*?VoP(p+-#v77%>Ao
zrmbn<CtDpoV|^`C$B>)3Y`wUP_SngJAuGQj&JT>+(lhI)l-cdag8*Ku^w7?93VoV1
z7;XV&bTR3ebuLu_oeTz{w2iQYumFLSmbFP7noK^y^l2-3wOJkPYD~0WZU~zh%oef5
zdemn}Cf9`)%U2$j0efo(UZNu~@9od^TBB8s1M<IXn;KZ3$5!-#H4N>AnL!ryD#vNa
z&;~l((G#>D4U$~-P@aj$%ys3G)?MMnQ@#wHC<S!k(imZYxT(+HW^{z1=N<tY`{$tm
zvWBU5d2`@XTz&>y+Mg<rJDD$cnVV70H!KJ!I!Jt)QfFwgf+7-}tGS$PykEoTO)tTs
zsj{wys+Re=>jB3xUvag@+S=OA_O}$ZRG#bnio8#c#>cIM9)ld9I4HDL*mJuqF15Y`
z<%rPUK0wTch9E`UEY+mSZ+zb?W)_}gPlI0)Tt;7&-c;)$!tiBgu&w#?%xrIGuZA8S
zOAk`F<V;X`(-v4&>`ezbIUkLZ?{bztPE_2w3skkUnu0MhtXfkP6nnsjfw(7rGCDDA
z3?L$yxfyLcy~mH)IMagS;+*Y^mD!k5G`J+2DbA4Qj^K4DAEP3sEJoZ0am>}2Denvr
zqfk$PGW-{<HPNVRsePQ96EZ~zXtByyF^GaM3AtzesqsrwM8e0^p~G+QJDx8g3$G9p
zkw577>P_}dllSWnwH?S)&Nexa+mIlTgjFbDPnEh@<DCAfA^difMo2^Ow0i)d?60GQ
zo#~}CMzCN%VCbVMlHVz%9}|ulW=zrHEB1dk(wkILkun__kI;E_uL6l##cyoWXeE%R
z8p1?0n<ym^OmX`;(+!N!3tvs$_BCtu-@Lmw>yQYUVsAf!j%MbT>I>VKK#vpppS>nd
z{(SZ)ShV#Ikor^+zbVdAs)nu~HHkKgcL1%HR9e^%K2FhZ8o90)lPe*Z0qX5Py-vB_
z{qq<#AK~zPuN?e#fPnevxL;QTULnqi>2D4(x;`Lb7-F4t9ruj$QQxmMa!=ue4}EBB
z)a&SL4;4V5J8Uv<a4(vjkiM7@d2m@pO%gZ6`irvx5(jqS8%B{UTDodZGGs_q{YP30
z5jh~Sh5#d=Z7UHXEFVQULid<T$bhB4<8@)b#H&SkuI)8sNis#x$(GMNf7We>lbN*=
zer^@)As5i&=t)~zeD9ZsMtCXXRKjin%!h`XLQ!xp4u8~DSH3QNGogQiJRMxUnmUjv
zz%*euPp`obdgFbveImcHEb6J{{8K^Wv{Wm$Abn?N)vjW4M7(+UD#OVSaosD(250;D
zkkp={?G+;Ww&&An`xW8Me}LgB(#oR{eVqTKA^=8@bQy|tJrW<KEmf&p{)b6&I?^ro
z<E#J1eh-cQ4=50$g2Z>Fq>19;fuo&DPPKfAXz99l`fHb1#nEc>qFilfReFXA2r+Y|
ziiW-WP919q7+Z+ZKCP9%GZ@I4S4t{@IO4I2?<uDiet0#s?e;H$=HF(?dtZm>y8vEy
zr{MT8(F@7#-j?PQn@f<UimW3blu%dSf}RzMfT)|rT-+`iK#&TgBDs)4cDR&v)6c>9
zCx<<{)&OW9MA&Kght!e@3zZ2y^qNUd8(?`OZR02UI9>{r4PVUOq$FxUl3~iDV<7)c
zXTFs7;I^R~*srkr>y$RRFSV__05~{@{Vo9oZHSf?=a2~GlxU->i+P}$xGtq-rUeM3
z&Z_PpbxP>YjLt#LpLfMI#)x82Cdb%}pypc>p@@6t(Znf`6W+2-dn(@hE1RgrT6l*@
zlFG$oSmICS(gr_mYhSfhv=Zh)8PX(kqV18FEYvCY-ZkoLUb2wH7|^WAqJ-jyYGR80
z3J<ng<3xd_bq8LUclHW9M0+59k#>rII3C;HT>g5n$>Xmv``Nog0oK&RaSQs%+Cz7q
z3sJ%3q?Oa*n+Kn!S^4Z-L`l^c|Cmn{zj;y^!Y;YP{-YS>{Je5Aris^YJ9E=r90^6X
zp!PLcb0H2(BArShIP~DZ4k)@&lU`Umjnh-c{90ht6j8K`J=~}j^^y43%BqqK0Mk`q
zkvsgTNK=|pvXrPryC-xa#4y|o2;!7mh!*2;)ok%H{D(X?l2^P0y}NhczADPrG+vKD
z6Tq+#2$~5?x&UKLo?h`z0WFENl(?@klXIa6eFPJGN$<(xwJ1Ik2(W!?t6h)si<|7r
zXInQ8pMs$NQyc9QOs0GjcabYpfcS<?cj2&0lAMm&4Km-ytNB6=1jSyIy_4?C8wP$M
zwP%H8kV>-ySZJs%ojV}KXgk>m=Ha#%<wP&1UBXstdFW0Zmlo3C7^HVz(f@ARLeX@P
zB13@|9;+uVGeRPe?Wc|hL^H^#&3<m3Qmhg}n$ir${NI28o>JHsE4uBp!z<+6%C0>H
z%J}q`&a1n+3pngN&|x|1<IU{qJ{YB^L{>8+e<psoHl<f;^y5(hl#Gs6nzoY%?;w@_
zTM@qh^CiR5G{3FoWmjO^^kKFI9AwB;ghA)J!0qq<+-Vha2`xq%2$T&05cd%vGDB(e
z_xu%An5}4OeW{Ns4FG~)3+e^kQxzyeiAnYFw0RnwMk*!T4fK4EiR}cI^8P62fMYTH
z_E9JzRa9jS^(w!`XTfN6oJ<2I253vGRWYQ{=ibDy!91_A<m2cdQe{|~Yg~IXo@~lE
zkrN~?)GGT<fHT?Yjei%N0_FHB#<ZpMY6BSaa)`YD9GGLHP0Hx5SsiC68?F>O1<B|*
zyz*0f>{#j&igTq)BA%YIL65sp!$NkK5yhbp#lD*d3n3u4DNcf9OtXl)fNB1iHS5Hv
zHNh5=3iAC-QO1yLkD2w%Dh7j3Pt0g!y?812LF(NrxQl20<=D33j6bg+|9g@yGmC!9
zb}x7-Q1K>>?_}9IVpgmNtpTq;+*IY!bcn4MWj9Hez5>61Vw-h2@gOuc@U^FXzfl#f
zRG)+C`0$Nv159D9zgVp-Z;@xezqh743QsR8F1u6s61R9-4{ct7x6n5eLj~Mcp!zx7
zM@-uIyB_{8gT;FN9(aG-?#&jS{1P@IRA!ZucP3_zgzPI)Ve7}Y2K&}x5Ibm=BPgE@
zxjXRFl})}Wr0663bjE^3KJP|lD7&agDS>p87;=SINJ+%!U@HjLo=V}RZx=8Gk>d?`
zX^QoWNJnCmeA`>j>RCg!#S_gildUOvC@Y*aL{4T4yci<IF8?+%j330fqauz6g}1Ob
z)r+sta+SdmzR-9@PSW{ff)}t0^|Z9XdS59gf=Lm@uIM$D+2~mxI>bY#$V<Dw3!OUl
zn)C^Mt~1Mq2?jp<7ExyJ9!7j4wK6~gv3yOp30vo3JczXg;<(EgMjsp;OxhGx4bA{i
zbUnn_jigDe8zeUtFyY_*#dZ4A16GebWH9JzIIAn+l`SuGEpoJIYbZ4`hw)lig`Nq9
zG&8Hb*_mx(<y>72RHDf&I{2$x4XeBTxC5HaYu=J=KoFWoE1;j{$1I0kyX)oH(gVjp
z;I5ODh|WF;^b<j>G=C=t6p{7~U+Qds8<is7_XScW?0Y&`fIk-hUQQ+gi0oi=dv?d8
zKZ{=(=`xzP;yW>n_$vk8h{jxeD_Mxdyvo)OVE*g0C(E@?42vd>&*#&8=gA%L%+2%W
z@pPZQ7cf8Dm&}{%X~yGp06-!)Mws)1Ew2FwvQ3{y`ho3B*&1yz(+!o$n2ru!Esh81
zz7=tLn*3VP<!m(taytK{g6vVv*Y}>ER#y7t_<Ob{uKi9&^X4SOG{s=}I$1&j6a{-2
ztZ1-d4+9KPiLhfAHVtXw58tGJBEr&Hkw}XF7SG%<nF=DR-x6GGUTQkT-K`$T#g)WE
z@e)A1hs6S+T)dgpX2=RK9jO7wDS*Z8vjAG$>f__V8cI~fn}h%xV`0bqlFCzH+=d$g
zeQrYTfgJgGeh85~pSy8mHq-da?Y(NBG?cJ}$<X~1xBThBS1U7P{%uyuhO_%X21#Se
zSoX06%<k`mFz?vX_X9PAp{@px5ykE99>vFMCE3n_^Ls}qQ?(Nm^)Ss=9~M!cy#Bv8
zI+#SUv6|UZ4)MMMcqe1-+u_!<IKKS7;ZI&$fkB1w3i_#5OG<VOw#+>Jn$DElc+B;Z
zzt+4H42#@lV%Vnoy8>SzqF$tfpg(0Wi-6$fYG>u-5}S_#NWZA!laWLfe9wB!r0|^m
zx*}_wk_8od<BTJ!{I-WReM-XG%GEOV`WuaYE3;bdvU;zsd3XQC`nJ*Jr|<1&aWmuw
zzW`)lBo}8K-iHkPhM3}zU3z(IrU~xlvYI0r{sRwKzo%hfQ-G7w``>OT{s_7yP>nG>
zp$%!`k<xUyg0y<trm<E^m9$SBhBQQu*uxUl4grj>fX&LV(Zf)kl=MN)jL~9(M#9P=
zzJMD{^|U-_OqGDuKvvkY$l5>XrCh+24EEQc68bY^Eg}&AYqdMvIzG}wUGPfQ2=3_#
zy5FG}6V6dY+|CC7$bK{ZUC2=yV16m9E|`ZTGwtG|`|^him%AP;-cE_Xck#I0*r4Fh
z@ex%u5NK-}6Zk7Sng%7kCub*H>$Icmc}waExip04<&3o8S8X|I_cp4IC=~vh97zRN
zkh~*1!>x-;5X8XQOu6m3K!3j7$L~$|fj_CNm~0T&Z=7mXjEeK%(uM%&=Z4dA&9G+3
z3#sGus`ZrVr_YZ(G>;WT@eH|&vv0&wE<z1=%j<R3I@at+_}$;?VL+o@4|u7AP+S_Q
z0`ZU>bSz^iQ4`!I3>kPh4m}wP23ga#=Jz*^17c&+3E$zMq_H)+cdWRk&P2p<&rEMD
zhDrQ6mU&GVYq`EFP<`L|n`B{G_GWehh%vmGv$jr0PD;AZXQZ-RI*kKsjRzGB0%Tiw
zBgMq%E7>r#)n$_IR__RtV^p%`Q>Xnc;(R7rT=}`Nho{lcx5ZW*6DY&$tN+Fe+FZj3
z3$zq1W@LE!3DCp{NP}Le$oAS*HIE9sN2mz9g9k2Lv?Gw(b*~??+kQFY65PxEiBj!#
z1d2W1(VhBLnhbsk%>qZ5Gz4mcCPwcvpIFNkWtcSHF|1SBm=HyPP`NbT6@G=3d3Hm=
z2{DB9O`zOX5$;l#e3D;H3W!a4pMFObC%wkgG+z{Z3TY3k^Lmnq{i-7mum?@cX4-R#
zUJ6P<rn?)8{o$&NaL{i*u41euPc1|O6|_PsN~a=9@e0!PCeH1>s7UsP6UqUKLQ?2j
z;Loq(?Sh6SakHY=J<a9>ZX?Aq6?iY-7DY}>l+EZ&gcWpJ@E}*+3~2JmL|)Us-;x+!
zfp<j68$<|cEoUE9_<}aBx7M!?)wJ}jyO<_hu_*$yApL<JhKK1gaT&wY5x}Zw>o_&8
zO3+?~_BEs|jM7-d`zZSf$`T~0M5k6f-YVUM0#=L3?T^}Jp?_WXQB(-~a;4u3Qo;Cq
zy9^bhk_EL2Z-26kwLWY}ak!&%?Pm{M@n%MZ{*uPdoyEdkKHseelb&{0e#2J(u)cMX
zRy3?I!ne_Vhqj<YkAFI-;+W4!<=UHf9qYN4uEVg9FOC|#N?w`_`fK*Dq|hBsI@c>;
z6?4uipc3gGvywJV&s{m%^7!@rmB~trqdJte;fWD}d1T)>(QjdCCRMse-}PX}*M~kz
z$qBvT&HJXUo30%+hue$ioagBmX^u4b<7fExoiFP8lb6k!H*^i^;}>M3B+XYh=<eW9
zf=fna%t!O%N3%2O9KFh+1I=m#ZoeD?)z37xGhC1bYr|2X>14@3Nc@Q!qv*z=&l!5&
z&Ct^|#NU^`4}tA&e}zNe`%mzljOET^2q<UReM_pf?<1B#)LSh5aS?+2%y8Ru7zbkD
z?QQbaUB)cAV$h8dz@417w6^yw_;%E_C<2RoYo;-40Ur#Y{N3Y;BvaI&<!Siair!R(
z1+zXTWeC`7xI!)FZ$eDQO-y-GZo8t1STrF<XiUUwH~^L&v18PXnk#Ewl9ihv7L~ox
zAyNI!qd@91QSe01AVVwX(Ms==k<@W<!wwmG%}<afJW#kiYz&%`@(_N=(-j%9QvhfM
z6SLHctN=Q!Lj*q?%0z>ZRxRhW*apS|k8C|ghj@zvRET2pe-8S;G0OZCu$Dfqq~+oq
zYOTNlKcISgopLI&q3G>69UZHp-!*Ugq2gcpHuk%Hm3x7dl~mB*@+59UV~8|q7CiNf
z+2bjl9YU}877-7PO<v)R;YF#2V=iKl46%SZo*yy9lY3XzXi#(IgTJq>>cTl#ZB;ks
z`+Ljwv2V*X)cjF5OkhK#S^kHoG^v-%7jxFO(*)LXHozA#cor!NQ{SMKf$f!AMVekI
z18HG%T^@157A;_qjYz;<e|w(d)a7;PPwJ;)YTen=lHRgfU`RNTa)0|Q8zs$__V6_J
z6c9BvAL^!@LCjat1x0tD>JU|{F|Tln{hLoyiu`^vQ4IsHqABoK9a7Nx`Iq;{PiGE|
z-@+E1Q)@4tI~a&YP8<GhAHLGdGF2;Y&)Ht-K}oUG8}Cc5j|4oD=CO{43NG1Cv%m_?
zUhlWKw7+u4a8i8dC{<#}jSuONwsd5%M~!+kGEH0@UE8_C*9E{1Vb-niiAi^)ws6;_
z>8tO*%{K1tz=kLuiwo&O#dP{xc?SD1f=y!byGiL-Jj_P>nW?O?%shiGlzenRhR@9h
z`y0|$Od0lBF)HfuJAjoEi?Xz$YxM2h*P9o}TuZy@&iAZbKtHkUXuKu@nUK5@?Q9M8
z?d`tej~q$*2XUD1`h2JEG_&)|m%GPw+-ummy42I#pt~X(gE^YwLNT4zwO?!5Yw=wX
zp&fO;$_u#mT%2_0`#mgLLvi~;=LutZw^6keb8!GPqD!;O1bx`@DQkdMK174`*FpHd
z0QVcXVrbH=9C4wrcd9K;oZeF>6&2Z9<rbR|_JNvzX4W|{Y*D<unR-~wPf!ItHA!dw
z&G^7CXw$ys{@>s=q8P__mdeS3Snr7CC{-DP32j3!F5FFOl%2x&oOzBo6Lu!)p)n}v
zYn|+&L5+)IGcvIe3~L73n|l*86C1OA{L4y~oqP3@8t;^Z<sx97zBkN_ebMrYe2Ry~
zf6yW8dQe3Poi~<P<PCa7_;2ar>Lv2*KI-J-?;(@)D6sVQUSEbdIJAu>v3P5YJeZ*S
zwLg3zKw@b5z*-70M{v(8R7p0)`0;Hrd%Eu^{jzJlV=a0c0<HQnZN8&Z>FFVm;^PIT
zi5D@ZPViHCi%3ft0}|nS=9qP1M_cV!rgVnCKe>S{8ovz`UzQ~Ks$#^~KrwS-q~yRU
zN))Q)u;nRrd*0!`L@|1X%{8P+Q4v-Sfg-=jT|o=m$fJ2ao?zvqOLAXl&0bPr|1zmL
zBkY68s9Zp7@O31GzM)ljz*0pqQw?v(y`=Xsue71A*sS5~stCTThfTF{cOJ)BBxgQ!
zxq~i$n!QMI{FJT7aeRhbHn%xp^VY<=s>P3{zWp`T9zQonanl}pmzOjPSzT`V)|57$
z6J+{9_}Z;{&AHxUBGxA){F8{hn^Gp?sf$bM;%^Y2uMC-lv7VxmdAdZs1DDQr-nmr!
z$Z7+yu-$zoSh>HTW{#raBvlJItK9WEAUKdrGg^KR#=b$`-5rO+wo|ETMOfm{)%_;K
zWRk7=W4j_CZ(as!%#ufBtbr>RKems=ht`fID_%OrwWwp<v}oH-5OxtP^vLVnK!1x(
zV2Sq2BZIR1q3Rn`iY)CwXHtTvdMdGYIoXt%$6g`-<}xMBL{?PP`D%C2apHQ1@_@I+
zddk%aa_aO<?(CCkE1&)qjEmPgdy>Z%AV<nUUw3=^wkp^lTcTA0kF`tu6ohQnOW+*J
z%0m;hes__hlz+LkGdJC)kFha=lpZzhD4~-&p7T8D=5y`QxyTyDvV%$+JF(sc`+k>4
z7<T3#u*vg$`K>pdY^Av2BW0`X7F{>b6o6{-I+n38u3xUHt2oUtA%ktbcy*!+Fy5@h
zFX1>ZExUg2ab<0VbNYj2G^QmK{$8EW>J!43vBO=O)fZ2<1q%<usu>?8&}Yt(k{Upo
z#7}@Pu;M80M3h-GE$b(XB$N7L)?zimPmOW*pQ~LD8w|9Sm>9ldRj=$JHibGxcJu@!
z$1qXI-BeA#O*5fCuu4BXR|wxUh?@9cXRdbA@>>k$$f<vS=aN&M!#HF4fScdB%BX<7
zywr9{d<3792+4_%Wy~4NNd;Wei&>DVBi!1;bidZSwM)Q2HAa<2KTL{mDD?0Yi_8{5
zt=jW+wVW{kK0+G=<LCsnd196*7kR|~Wl{+nm}il5f}1IwSm+NzP13nh5&U%@^13r^
z(FRiA*=B9imw5I&nh&pZ(5HY_e)2Z>u?H=68n2U!y*38ZkfyjI+;31(t$=b+)Z%6%
zk5U`wqrrR0w@!a^kJ>fFf9U*7uD&H*Js<i?=n!ie?>vn4V@!}SKhIZVxcrMJ^F?;s
zgy{sc$JBr{!{zU$ViowAE#B1{{^H6+DuzQ2Rcx}h^)NOYvzA+;y92I;#*mr|_s_$^
zsdpBaLJ@QUdi#~Urxn_<3=B<6GG$@95lm=u$a8}3?UnwIw=qTENN0C#d?}U<^YCm(
zU2a>fuLC>QlSBRm!Y^8A!kNCnTz@AYS}{0hNb)053Tcficbc_;#F&%5{>A)A4yZU9
zcyCir629STA)v1KA*)J}e`$5$R4t5iTFSA<hu`qyoSZ1Ifv(7YRZVuLz<b6i+;E8)
z|LzJHy(@UB#w(jya|5P9#RaQz*nMm@Rgs|A)5fBu7GLr8p4Va^Z~tPyl8i1j*iM^u
zkiGEl_sr%HF|_%%;RaGg;<TezypNo)r}Bq_A88I%w*fbWywRFZjpXtqGml-#%@UTG
zZ1i3@#)JW_@cXWFH0wjn%YV?DeNSfCWYh<ze<nTsP0l40e9J!7TpTn|NmHR;r>KxZ
zm5{s`(Mz1$7!h&X5PY7T+l`CFOS-l-Z-tWO*L>t+Rzg<99mA%6c1N>gM$k(U{8c#<
z@z5cAdUWG$C@u@%TsAPiwL<$Dl2rS_qEFnj1_Ynu)By_Y7p%)?la9oswy%$hN(njH
zsEt{6@0~SadUoINJsP6~(!IIZ-jtcUjGLsk{k$s(3|Mp*#h>o+q!2>w5@q|wR9EpC
zE(=m|3LcYS3SxG`xN;oJT5@}uP<~jS`5b*JTvXX5zUs0ms*Z5kK3MTRA>TY{3J36A
zUi~=CRvW!jPBv22XlznUAQ-RQFL$mQja>QGbH9D^pbH}-{fm_FMSfJ4GK%9Hk9H4V
zt{+kwF42Nuc6~^AUq$LWuaLj3D5O-2O*W<Kp!{jUFhNDImg2o7R{ZW2(G_JVZyjf7
zq_Q$u`w`_$ruXgPfe)#jjQe$@r4U9Se9IrtaxvXTuXm>9@%vR~Y=~5lPf?o80)AA4
zA#F)Y+aoIIeE@eDD+IFfk4dxI;2#%A^UucB`1p81@#FN;`>J(^qQ)W5_A~z#jnAAd
zw>@OUp|TF^L*v9=hL`bzl?0f-+v_(O(|&dWV5tgaD_?pxFFVjrIX>7OHHDq)CFKFy
zU1{aMuImOt?9Ro$Sa6~%a`CH<p$b$G+y6rksNc}(syE6HS&a|iX*(TYuz3(w<Us>l
z`gELMK8{4~+Dz5*-LDr%EPZrc!oQhNh^o*MX${+W^9{7;1eoi+`u=HEvYK&RK8f~n
z-$-dvx3S^<?#?2@KY^}vi4|2fo;p>{EX-)aZ#bKK(FdqYLbNap7b*A{yh&9!gjN+}
zr74!tY=>*mf?v{|QjJclOwxuHC3~X$bFY0_!m~$JN21|Atj;(hNrD8VuFqGD-f=sg
zuJ^;VmvK!<*1`>jO`j}Xpr;-&Bk&HvU1{D)K5f=mN=q@qzLz-0?$f-I%Ao5^Q9Z;d
zZ4oaBjltG^#x(9ki=F4dMpHnybEhwP>A<@3IfEa>@GWTk(92J;P+hKHkcvIgqW9Qi
zv(3Zs4fk-Y$JX^Wwx^VBlQ2Yl414#@rqAu5&X`G~jby+*4t-lk$K;TVI9|eyw|Zrk
z6ho=jN6Qe9zV@bij--3L#77QPwzR+?8>y3-6SjOYQQNH#UBG<ruR_hfx)1V-LVXU2
z+^hH=B<jM(Ze9<Zc#0of>fLgY?0zR6`8LV;A+bmw3QhAr83<W*@K9G}TKsg#u9IO#
zg+VYdDH0&@di(q{ibWHg+mBq&`kvMnpU{u5qKa_!)Y4ks_C{D_v6J8I4eUkh#oF12
z^!IGj)&W#Nw+75&pV2a!>7B4t`z|D0b@<aidyV@cfE}KH1dG30o(U2|%R&MyhnCA2
zwFNG>=kl%9X*MZ;$Mh62G)l}qf9|)nvPQ*-gS=gxq@^|(`J^F9mFdjx=>ed&E2+>u
zNYS7B+ssX8_T?)mgLtk7l`oSsOACwsQmijLpK}g%GQ2pdDE=f=?@LT6sG^YXr&{-Z
zyyP=#dBp4bfOHu*h|WAq2>5cH+uh>mcnBFIsEK4`DvueqqT^0(n#+Fg=iN%`u8oVB
zrc@!!2rQ-eFjjt;#+X4%P&PKH;F)LoCHA;SyqD>tPe-+JBj$>FTAA-(M~e8x!}8ho
z9_$NLD(lq6<~6MCmF6b^#ZT(cddK7}%J2Jii{up_Q^hNEH)mLmE;<))R;W+{CO!mF
zehQU3YPj!Ad<Xp$8-oAAYeNyE+7djnoY?J9ve&}9lVb-<24CY=dym_{->aaNuVFpf
zHl37te$5jtq_u;BhkC9DiK+Yc24jjj8a>_KiM2np`+Zy5p(76U^On^JWo|si^2(zO
zd){a8u-)zq<(1_%=-u)S5DN3VQCB~?Gq?vPB6me{H0ZV;zR^9{(|4B=J(#RLeigqa
zYZ&klBn!xjdDvq~hpMFO#Sl#;(oIl2ySEO2(NuZ!o;M6~f=RKR6k4bow05LZgvi@H
zz-7G$lgn@QdP8mSJ{5u@pp{(3?Y{Ngb-b9|%(aTS*d3Uom3Y#TBEH+YWEHb(=cUhO
zA5LAW{vPq3`zZA%>DQSY`sy^2oVS2ffKIQ-l?Ny8;i(^(mdgKqD~#XU;X8v8NI_Df
zy%YGVIbDdl^nU^`-IVmdk8Gj!j>bo-&3)DV<w*JeW#UPlExA(j^Rs%{><ORxzTkGA
zK|-!nxrnUm2DSdkw?d0!$^6590zJ#Gdn#4}?Cw7}m!7H@wF+!+ambCxC_<oXSuJT2
z>iYzh%ifgWk#Ky&cv5Nbl|;9R_w%^pxm<fjkp=%>tn}3V4F5g9RBb!%Wl4UZoRL&-
z!N;+&S;vXCz(Q~8RORoon}Ny$>16ESgs$ZLlfQQ9QF%c~iRFPt!WRHKtEA1gpB7mA
zp>!(@g#jimi96j5KK$!A?s9D84Qluu&1Y&l#;(p2rSOZU8wv|JT@QMPorm$OPS_WI
zC(2s~>@M*ZtN6(uUOwe7Vj%T$ta;Cd=et)>9PWA``bfS%^QM<t#2BGn90E#^TaNR9
za$N15Pp5o(<0WI9MLxEnu)g`-q$V)adi{_hHbslxglLJ02u+DBaY(YSv(f1n$-8C=
z<6n|4fBzz7n^j`1BqEo)OJND#cE%q2FiBbckTP0m#6GMQ-JD3}J61$Pq44ljVzy!`
zfg`+wf+(nUW(L~yp9nlDQhEesJ3ey4dRW1&6x+}k&7@5bT&7o$F+k%(mw+_ML00@a
z5}j18!?C-mY6jMfiO%duc-M)>3ORjBgabsvsf>Y&ZAxo1Q0JuTG>-Vw6ln@8kNW#Y
z;+k1IG1d>Jr|2lKsL0ZR#8{lKf+;C2p4>4}nAzPFx{I#2oG47O4Ah%6^BaXF(*0)%
z{+Ej1U_t;)&O68$sSe-VZUkiGEoMgC`ZX^|smt_tLA?)#_R!Hl7cfm3&8^?1aS?;9
zcIW8Xkl?%}cQTu%VN`X<8tJ{TA;l%;op)xv`pX+EzGjX&UbE+BjIkH*tgT}z>`1`A
zXGf|`%?^Em7Kb87{iK~vX){wTZr%;=X@D%I{za;i=J?x*b)E2qCjYSXBSe1eE`DL&
z1OE=cQ(%w0b;I4nPe9wnEmf5pF<9S-xG3Qu28~;tV%U+Fdr3j~_WpjTqQgDt1ya%{
ztD7&3sGm-!KQ<>Gz?85aqcPm|aNPg!AU+UiBT$6y2E}~WAGxb*#6$54r_^br?&K!7
zOsj+w=Bj4qcrG6(22CfFT9U)mO9gblrem@ZX)FTqf^LR7<v)pWc@F+Axd<MG$Dogo
zgbLlE@Wig*eXu$5r2Tmto!_wWoit}~w`aLOz2>(HuK*TijxH_xmnn~TR);nNB3VT{
zIUgqcD?<mG{n&rDw05Z_=c~$Xen=LJ<dxggM*bW1$#kUK$V!NQnjYuJ$#>aYm4i3@
z5<(-Hc$nHKEayr;;^<PwLR7yPo+mf$>;#K`kY>53zg-2pWKC14QgNMBsI-T%pilW9
z6i=2kxKfHQich`}l7fh=&us&KDXXFM@k9i;Gp4Vdb`>a&ppJ`SOHd$FY(5B}brrFf
z-pVBKMSP(O4G}%o`HtW4<wOi7a9pEO*pHmNnQ5!@C#*hj^xnxibjl#VhxD7^MU=Yw
z%!k7I)?GElEB0bR^Ye*Vr-k_`q(vH<dsg&Fb-#!G;rEhloQ~=tuUgpqe=&(PCWF84
zju+!##LhQiTG(^%HVOI65y_}I_L2HTM8%>G)rfO1!Gfe~y?@C=L7^Y`Bh4!;(3J>0
zMoKx@<7HfEovLHiHTel3Wy>zY*AG;H0$I?;@srv@qT<zy#(AlvfA>d}uf)l~fD5VQ
z<W!N7U#DqRetBj_7eofgK8u%@qnE48hQDLER{BzxAki^iKplqN&e~=*+0cjeL_^@{
zvq7;k2DMo!(ZONeMCgEMd=~tYf@_O<y?0OA*7hJ5eVm757RisTZ4F1k0y9HUamvlA
zHPC$N2VKX<vQs4@TN@EHelef`+IcOl%t9sU=DC&i`0#S47#*u9<N4-+X(pAb3l#~C
z+GghL8J+CUnb~DMrfJMcG{3jDLi!g-){JcwWEL6;YHXv}ZL_v|Hmfo)Z}+l%9|WvA
zjoKaRU8eC2q&k3v@DvEGVT8DML6sgC`{bSLS?6TRWCtcLKB8Z5!;|EksIc4_y?Zo)
z7JX3!AkN8P&}ZZ7YV>pKiHBLXW~sI0Zm;15rW2p|kbCHt&ZT1oitMGp-D32@JI6#x
zG+D1uPxDu182Zcq6_bt!SuP1B^q{s3Jj5?8ad=TOCL8#V`4NOBo~tEWt@sl{>&RNR
zDuI3Yr~Eh2nd3qg$QE%(0?4-EK16KU<>ZXJuIDDVhWHD|v$|r~0>iZANN(fD*q)Iq
zv?7t|29X#h#RNAILv|8t>fwdr+wn9PDRiE{R#*x0p8bG_jjAtO)I^~|nkqK61?=Q7
z8~8luR1`yIKtGlg7T}+lZ0Oj25+A>u5h6{f`xAS@PB~jGBdIk0Pm&qeds=xzCw`I{
zj7>lX@6QX{+MNxRvrM9wDTR@;&l0kQ@tFh0MZLBI1%z!$b}+)3=$0f4OS0@nO8(mz
zMLJqOPmLna56gxNKcOn})x2j5>bs7pe)>*SsuEGSf<+O6`Jm~3*SDWb6`2MGTYSk#
zz-MI+yN_8eqs7lHhU-lv$ZQyF>~n$_ZUzQg1rD-1VOkKG#7;vFjPhY{-0z<7>+GAQ
zvA|DCqqfu3GveQNdyC+Y42PbG5$OTz3HW3Lz7!|zoal%2w)2l?2Vh__G{D+NgL;!h
zGt(Vk4B5hXHfoU~v+E7olWm+<o?k7Uwz7T2oys1{@Bz$LHEv7f6a2WX!xyjf`$}N#
z8n(cT_=lUBHg{<ZI$us}SHBGm@zV-$*`uogD1O;td-oxpzPrWA@g4n0mxK~ae`hWC
zrBm(__8j5PL{9YUUJM1%^}B;cx={k{_jHHh)>dv#yVglx+>=i<f0%<Qohh}CNWq^x
z+={3USiR-8>K|}_G)hmW)%rC`b46&OvvK2urM?D($7H8=fW>A9y!P}o%Y9Kj08FYs
z%OWZ9k_@AkFopxRq+~AZYpj&x4p7P1jkq(i-Y$%z=2LG|oNYR(NG`A-`vL(r1&0Rn
z4PmhNZn%!`n7x~Lu0bM(R|Z>OewK`~{D|KnXkM`feD?X;N{ip#V_@4whPCyAgI3WR
zWO&b6PpF?>eOv&XAxT)RsCT*SJN3jZ>Zc;(Wvw6_Q93kT`Jq2c;y}{V_pQjiZ~D~w
zS3>Zmf<vz89md5KlJ$|*4ZUe3*3}(nPPUc{vUm~h&-KO?tDd}nJLI`fxpO<od*Lx&
zViEk`v7NRWlbT4!34yJj&1tfm5syjDW&i0P`gG<P5+?Vi?-z5oP^>M30<6*3`N3nM
z<|F2Bhh}6k{W<0`#@?X!7>(>Mh6CTrF$`!<xJqAy2KzXzd$8c3aXWwDQ7>PLpzQmc
zht=N9tkSD#N&#Kx>uEL0B%<9ViiK_2fI0d+k+n5I|AeJPP*Yl}x%)-%Ev_(wHjgbP
z*vP$&sK`|2QfN9v(H1FFQMP&!+oXjP*}NBBM(AZFV!ZH%QgY?OtPE%{(_Mhaffo~X
zLdBo>e}$_IWiUhDva|6FKJ5sEfa-v8w#hU6aoyEB-NVpitmqsUA?wvn4+h>SQ3INp
z$>6Zz`_P{w6YFC2H4hy=vEAQB2;0rF?tWZ|-@Eg9r<q(OJU5DLI&HU<JAR93v(7Oa
zeag#qs+e@x62Y4tm8G!1n#ikiF}XXxO7l(4nApCqZew4vvM}HmwcbsMRs4Ou$lB_&
z69DLLtakdm+w7ewD?lkm?!=oPB0{NK^i(j8UbV#&i$!fY`911+8UOkzw^6c;BaaH#
zR(qcBqO@W>p-=a=X-VxZyWh~?^dw1Y&6b<3^{3-VgkfTA<x5J}_@PyDleN<V(5)y7
zx06P1w5CDh$n<dk#V}-0L8*=&f9!&VZ)sXlO#KVRd2(?>^ORUon&&7tS{Bb2Y;GIQ
zO475+qO*e-ew)0T9s2P-URb|7XMO0o#?P-U$RYY)w26EeqW%PQyb(#b*EoNmTQ?3J
zXlK@fBGMcX_7G<<TPb6IyNbxgX6u~Z&Jg5UhIsy!jz}W5vq7zda*mQJkCg54%13Vy
z7cDdgsdar3t!Mq+Iy_6EXOs+$m3?F}486UPgcQ|Bddfx-(wvnfRyAz+el=655!1#h
za?_8Hbn*;{Mct=0rk<XRrWK=BidPlakCT+@pn<y}ZY)@3i(wJvw7j_vQ_p$n1%@p>
zWY1NZ`1tv~a%05Vb$h#Y5|N3a37U5s7bdmP%4KJl<-{kmW4E$RpLN7fy>$(n7a}A4
z68XLxCc`pZR*T!8SZj?JdFwV6!+xD#{!@&`wt(NwkyqklT^OH>xY7JM&TMo|7&(5`
zR&}wQHoyIYjYN9T%zMsBjZ&nieyATi{oSkdz=p-gD|L`~UKNE-PD&!+cKUADy7Nzo
zLB=@iPBRAx&D;>f4l%<bph%;$!QAG2Ypc+WOB$<JSyYUE{id@I?(~NLJeP5UQ{iVY
z2-VGtuFi%!%cGD~&V~b7O%?^4Iz<3ARXI)buxS}n{Bs>q{QhO2+V@N4!~hTVYkty0
zEJ$32XM|Tt+Z?1SVhY`OIgPQ7HpDI_n;g-!h1(iLBN5tGvKfh+Q%%OwKDwOPf9BDj
zJ5A^$d1IBXE+yE}w<PQ^qg8x*I<spdL}3(GtAZtjdAl5OHqLqw6@3M-e&yuPF8u>=
zijsiM$A31+wg_ze7N3U#H4-2dsPxOu;!QVUP)fI0cB8L@EtZ=tgy-F>_p+&Teit6?
zV#v1E)qN>^<`7<nfoNW8(US@<G+h3meehtxrK;6Qcjg(uI-RUgmq!|q!!Oz6sI;TI
z*fSu?^)^6dw=-*jt78%+B3PmrtEpOa%|W|;(Kk%;*Prgg<Ag>Too%;N<wS`Z2^U9o
z?4>u8M+bp@QEkyBe!AM=FWA1+88bWPuv|Q@F=5k=Ul#qvv5MFqWK}8cElQXD23-AS
zb6cteZv}fsOxR*P4pZH{8JRgzNX~1=Dn)!)=q2^-dC(i)Lxk0nwQVb^*Vfn`WXiO|
ziC`@JKYxd@_c9Uw?8(c7Pm%AY&lpVL>_l76eI>Lp`{es)ocnK9nMVKLeN;IyHZ0o$
zlBnAiJOSGtlHR~i;&UdM>QvKq1?F$wNH(A7B&b(bkHZJ3MiSuz5fL5U=$S$cVcKlY
z?!mFs%|sjNJ#iu%XKVXfS78HGB-*1t1B2^l{XK48_o2475<=f{IsdY?(c%~z?oYql
zqIpT7dAV#ls;HhjAGCR4ma{m)7IBLikSMx>K!prxG(N*H2UO&VPmlJxdoXx&3W0ny
zQZG!y+YAG7`lq9l0y|y4)V-3V7S_HPtEkUN_f#@9+HowQ*!~Dq#R0P0a#sB_V*v_3
z<@L70tBNeFr!>+79_q9m^%KMuKhiDj3H4cObAD!N#?tP@nZLWR>C7BOzRA(X;(H;-
zPnaU68qGFLa2Keh>Es{F%!qP5XSOlpl`5O2QENhVA01c0IeU78=%A#MSUJanKZ04O
z?dsSwjrB0%1pMM#=d<phvq+iX-LHT8XMpsbQOyPsHF|fF^gs}t^s*>-F&neO?|jW?
z=0^mh%-+?qADqNCpe2(!x_CrqtxkjQr>dCByP8Qu8AG#T*O{)TK-p2$KjVB8^+|3l
zeQIMF#s5CS#~dCU=3Ereen5X<O#X>`XxH$@Zx(^bo11PgVJyvNe!!dy_pmK=r$pu7
zYO#JkbK7olwjHEgnZH4<8+(&Q;{qf!ZX@VT8-%IbvY~}?&v4gmj?i{Xay(M*0FuZS
z3BxWlqh(VoZrD!FCI>xBuP<tAoUvtA4#zYSTaLFkD0C<Ho3CQI`&{#_c-Q%671vX+
zfA?~%7N@RKvuU?`2e!1k7ZFI6OQ=6{D!lG4YO!wOAuz63MI{>Fp0$Lt-Y!>?SOm(w
zPnfiI*Qw1G2#Lk^AjZMhC-3pO0-=#-LqR*%{)VRM%Kp*n_&um;H?2{7oWl_XMY_qN
z`@HUP;uUWUUHk){(2FD7j{NWTX+%%F*F1KLM`?d_{ucnnKsvv%w;QQ>k=I7`&lFq}
zaE6u4yuLXZ@wBUXP)$&u;AYiR{~TfVT_I>ASRkM%UZ@3Cnhm;m=O*EddG()H=7t{_
zBo)qq=)HGB@X3S^-?MlDMfaE7rhU>W3~0U0m(vSOk7^KP2-3VO^a5{oPh!O6MZ-~l
z-vYuoq2O-qaj3a&C2;Jw*_k_>9AjkQxg3a12;C3{DXsj?-~4<2>TmuVhA!B9do;(s
z$$sm$&q7V-eM{?wPe1+@Y2A3cy(6_yW;j?>ff%6%Wb_Are!5UTefkrZr;pxE!wjYx
zk8q!<2=#8!>a-&2qo9_FQ4Tvh>=}8jZZAE9zd(p~9>OymXBG=?O`Jss=a032_HWlE
zcmY}_T~~@*V+|Z>X&!D96eIyN*8{wGZne<sBueI<m!fygr(oA`aa_uoW%Y@qGR1;s
z=2Hx(VDLPUx_RD#iwt#&yuC<_GeX?wkubUb`F=Og6L9PAAjUMqnI<(nseu$G3a2xX
zbU0Ed8W4?M9EvA;))^4jd*^=av@wvSfG*#hLuL>Hn$x6Rbq*csu1PtL2l{#DKD6Ni
z`99{ngE>EA##=nTubz66`R9gxi#?7((e9{IYZa+Gr8n+%u$SVc^j+v*1c`|dqDRha
zK(mLmdHr=hXTL76>^BtyOUe!f2ZyEa?|15Lr`-?QFdvBH_hxpGg%rSsm)ItTHOMLO
z!)4<SAD+2>d}g~mVL`ZDFPQm58l#Z1ZyHe|{pb$omqCiYxtKCxInOW97=e|e`W)$;
z0%_eir^YEh@S*lTEk1!3lrkDWy&a4(_~SLPK3|A=^~qh)VXHPewR)min*k;gly%K)
zG5LTBPw4S+vlGrdef=T&XE^YUep>@NqZQj2boHdKq2FGZ7?N)$8fi_iKI2-8L*hC6
z=;hvrBL@$Y-qZ>h6Ox30zi=_d+5L~lpA7~sXcG@Dk%1)lE{C9g&`pG4Vg!%0#`}He
z{k;<OP6|6>iMm(l#v1cX)N9;cztV2+?!>siyU;&pLQ0O>gowsK?+5q&%M7N0^>Sgo
z@5B)N`VUtJFEJ6sMWbU3Li8Sd%$;g<=+!4vo(X>g+!5kZL5DY|pI4&IPTT`Zj@H<V
zzwF>IY6{LqwmyI66eNl`f+Tp#n&(b*emR%k$Hz{#)CB9DZrv?}U7-2ggOrpJlwg2)
zXV|52PFMTAt|euLf|2|_t7EF|DD-m7hGK`{;blQ(j;oy=tPFAt{MBFmJO1KNe@lo9
zrBurOow9q5fZzz?FwX;Nc82c=OWt_-_yg;9@fh2()5?zF0yQrhPKCC?6fTMiPcI+2
zK0VQUgVt%sPTB92ay+yFh^yRPh})4mk&sj5_@67f=VmQ@6MBi@R98G<afkQAq)OCy
z2+dBh7<@ZmWPXmxIC${YUO9G;CHKzM^#r}+6JEo>e%aBFL9K;aAxfMDh{q&{5a{MY
zUJYT=5%IC7aK<6`fsPphr)RS^i?Ue=NOZ&`*i2oVLhHx1X$_Na6A_!4hdgy7#zCI>
zF$5(gc?cVUDUgnCg_MA#4b5}lPj6T)^jc@b(WmC8(eCdUh1!i;4q6+mmka6o<fxG~
z%FOw%rTBEw7~$|D_dHul^|8d?ApbF09M^?>yJr9sUD4>q4Kc#-gjNs4tCag2H6->X
zl&Q5t19I%V-`>0+OUkARR0&I(LeTL*(|*nRO~pWIm7^AdDf?cz-w&*KR%<X=Lgcb~
z{I5UjV^)h}#?yLNoJDxKUb$XYwrwLXnXpCna(AUM?loUxU|Tk#M(*L@cH14gA1g5@
zYBOr-s7AKy6$ueJcK75*@#x^3J>FKQ|FpF^!q{VGcY&id-tX13?7=N=i!1rJZE?im
z*eQp13^qJ6Hz)P)J7)*L$@K-NGbV@FCxjQHr>6yRF=m{0sjzLVPYdhQLfqWvk2z25
zgLhm_-G?V0TI*<+`7)pWc<4<y=HBE26ITXD0y(b4I6o6!O!PKwfRbEA-&}y##V|X0
zKF8Zz?+$g_0CTaAN}?T=FJIp&$5)oHvgRv_f(?&_4$-@mW1j=4Dt%OPTFA?WM!!Ef
zujF;5*@?h-U7K@anTcVg93JDUt}jrD1W|_hWJkB;bbOq&M|z65AksHlrP?4!pWW4X
z*q@MZ`?Z;P9PtRa0n40MoFOMZM9V27n}2qkgrPYNHX2j$=H=5~>Mul@N|+4jDP}$G
zhP}ZU2^&UY4p|W&$n|?(ajMKOmn)i*5C5vWAIi<?M0>m=b0dr{Y8#HV^7+#r`26`t
zAaK0D^X1FW{Nq3U9ryRIQ^X7iEYZcC=iwGLaJ^nxmc?VDd*i)T4jX7nT-I&Qu81Ce
z9)RLqX^fGOBBd32DU9CO-`_a)8{ktqABXw;uD$!aRizM2NIET~W{(ci4+wi4<USf#
zXM(B=^lbD8SFWl=`R9ehe_jlP5Qrl>J!N1qf|Q3QsEs~dDn6JjI=lLyDq=XP1E&e#
z?AU|P2?@~;<lu_?bFRdRn7V$#eY+<cAhQs7Oq7k$(YIk}bz7y+5t$>B@jYwIe3NtH
zYpacJ1+5Ls326aS@q<s5F`pTkp89cS6~u((1x>4?5oM<oA0TxK?tG}4i+cC6(`Je+
zDWaEWbh!{QtlsE-pL-yWnB`OB^DV0STlyfJnzJdup417+-5Vl2u8*fZI<**PW9d_3
zi#I(@z9nzAhBW1VyOZKOFUx|hmx%>T^vZrA^!zpN=x=$UkJea+M?)s@K{6&_;ueYP
zl3B73x@f}(xYSap&Dcxx0Y#;AO^NGeW4o+WI`{i4mw4fJ?9^IGF*<jW-6SqSDAuSl
z4?kRB;UPYQnEZeiRp<Nd95OR)9y2+RF{jprfXM)vs?`WF8ZipjrwiBXg=JlQ9=5pf
zqmMz#X_C+#)rBP{uIt4GMics133MD5z8{5VE+$%L_>W9S#=0bOQbJTNe|X`?A3yWq
z`RcI>V^Aw#gG48UKB28fZ3pTzFQ_F>5gRkzs?2yxoybB!#XIOIVW!%3>L_5o(WvH`
z=H??dM;~;9u1=%$)(M)Xy$?hJDJP;VjHa~Q_{xpreuvOWC?=6U2YX0xWwEIb4idnp
zF-De{7!pX!#`E(hLP#9#jpKOpm$LalshlJC``s<8G2&xgJvs0c84UAL&>DT7z`z*Y
zpOxUb(u_9khdN(08FM<D^NpVUG9E9Un2X4IcUQ`BKd+g4z-<iw88&C3GozlH4{@^Y
zg2~>y&+|AA5c7EMB}}oDg1YWu40B6sAM_q*y^$A()8%*WCRmota=Eaq8>KbI-Wk16
z=4R>LtUVkp8pJy!ltCut$kX=ZnG^3{xxatmpZ?(=`SaJGXswdf4{ZcEd3L_V|EffK
zZS?j|#W=*+1%eo<Cz1*|C+u+Xouq_ycM;voK^<-fE61Hy?`~_H2Lbp#!<Z9Wr*6S|
zrHf)I!3^(tEc-=qVD{j_U2qDBTN?G8QZx^qUyANj^^reuri+TYYTm`KUaW-BU{Zpm
zw?{1WsiV@V)<2yBVd^tf-S;2G=X&^$o{5jqi-FTJSdWU;%9sMlA&!)tJi1R3g(SgU
zCA~k=1O2N{k?0)CWfSpZctrJ7B{GInyjBF3NK|MO3K{Cuh7F^(ikXmu2cOmEQRkjn
zGR4MxE5@QktTh}h_5Jqp@e}F!!)%m_qjyZ(UNBOc5yC+|N*`tcG=sJ9`x+)<X=894
z#RvBVVOiot7G(wrc`<XQ5%ykPpe%}r&!1`C$250^(1RzlPBgT&PBcecbka{_H28J3
zz;7!CVs`bm*&r!gVxZ?dg&Xc0HucWZ(QxrWA-(x@sp4wgZCSW%D;fj0a`3err2YO-
z+vXJ6mX*C!_O+2y;`MfSt~f-Bbtng|^l2Ay`-fTcSpdz&LlPz={)+)4<Akr~VIb03
zmcY}8gpxhc(#%I~_Q9wFkw{)PVwf+#bQqB}Z>;MDM(4QS>Foeqn76ls+i`cHOH_0o
z{OcNd`Se7N#<nE#npt0-xUMTXtsLEX{x&AmJe+ngTv6)N_NKUIjt#a6KgkJsJB|bv
zJ$VbE^ixP!DQ$PYHZC9zR-Q`MHkxy{>QTg)(5QfG4e}b;@|7jO&<^8>4eLrT1IwpC
z%5@mv+L0lH4<A4CvOP0e=j)e$WWV2Clo=odFIaqE4U}5A-QEElic`0L?Z?hl9p-fb
zb(l*2shCzly7_2q5dt2;Y9@5+wAv}fqqMtqnsrJ{<mkB=h>t~O;1#&}#7)7&scP|J
zG6gTRP)BR(u!mLE)H&YTpp=HGxcKns9B+lAS5l5_m&Ce8wD?Sg;NlE-E*zEHO^8Xj
zToO;$<rGE#Zdb~g%k>GF!B;w6I;S|!rVzHbHaT$T{Z#{bNvxNZK8)9&|4g;c*ROx(
z?du!8SB4ErA4D6|fA6BdfV&a=5UBKh&_!t&Ed*@I7{XpFU4@hvPv#sm8dj9nI=6Ah
z>_%^1tOT(+GFKmm`It#}IYg&J?3crXI$3?*fQcV~;TOM0WQJ6oT{=7z=fehp3FALH
zY?*1Dj!?=>a}6UR)tyj`LBv=V-&0G~nRe+70cI@M3(L#POh{!GD6U`^rXDKH%=@W}
zL7k%Q=Aq2P#TC;!=|A@@k~DmqwP`?*fCc+!sEAB-)YBwqq%JseJ9Q#<1~N?tfYpu-
zVJN;cb;Hu2&BDfu0Y^fLy`v3GeQru04Qth(6`j%?J+&AUb%}{>UC7sGhI){iO$V9|
zaYQAYsjcVy1t+wWfA0Aoy%UB!+!d`gN(;1>*w#6s9(jZ~iOjUunF{IWRcA;;nA1@W
zq))^-1doyq!Nqi(;YY4l0pa+Um|A$<7IIF=FnZm;!D8kIN>^os;1#-gs8BcKD2=^T
zTJcF&>$-Y?j3~$b;LCo{zc_5$dt=*H_FSiSVA>=mWSo;jz<V2%z0%rXNlp?EFDEDW
zGy{(6{!qx;tT6p0Y}8t4<zQGRFM*)mS?tF_pT7E|c7kRu%R)+vI|qW#&rZoFUe(rU
z!zmE=sQUyx!>h-bd@|m;C6-mli(sFh=iGzn-c(&UioZ~1!w*Oes0CbmB+xzC&?{tz
zt8KzYAw?1yMQ0dG0usF$r4|(DPdo1A7E28A2Uyf5{O%i-QyX)@#qg<M;Nrt&xw6Dh
zT$Yc#*PS-LaNHYO4b720h9dwX)K>hE%#lx@{={><IK(y@$9<ougKx0SJd?aPPp3SN
z;t7#s@?K-kdWc?7mQo2!I}?hT(wpa*hv*$ot;LH&%>88J(JwN3=e}27>z(VmaJhJd
z_9?#cAx;g8VTvq0hPl<sc)Yky=g4r1Ma=m4;e|`Sa*KEV?(hF$&aMz>o<&i5rPX<|
zn|DQN@n;+H0^+DX=2+CI`-%>Iz@NX@b;8@NHCijwQv8~UbH?YYwb5Ip<HLwjT3DAW
zqRRa^IO-ew;q*cu_0gHz6o8)mY*uC2u56bV_qw;8vhNHH5a%z)$o(K2#xXx<)Id^?
zpOs-Vw5KsnetU{S!*q~*JJgvYynT?_aU2uDtfX}2&+DY#{BsZUz?r=G{FOFg{Ij^X
zUuZ={<)PSD5z?Y0oF4EAot}?xE_^}6v-bB?&``d)0#wPDmFo|GAZ@E3!nM+i$DDUj
z7t{U%;&-hAm;^#hRP*#tHJ`{RqVs2unfQ86l{_;}rtOlGdc+&0UnaFqI_PWw{2wO|
zhPc*7`pn?4!KlrL&^3=vCw!aO=VC=XTKHD6xhKWs6G}BWWqXSF&{NA0STvz|b)>2m
zN<W|;v{K0)6G&9UhR;NaVItV-fq~tI7mP8|PxPqvZ?go(43CNl70*%d9Tvcyc3qV!
z(@i3V>Kq6qOoVq3Vq)E#{?h0Lh&R2iRXRt$r<8uxJNny-fptYgpbd}v?S}_^1ocC}
zf}+W-bhWw9|L9(|+YleUYoH;zXzpl~y)yPfL#BdVX?t_8znzXM<FaLzoczJ)-RXzj
zqhYrt&Rl8*e01nwaMay{h5Dqxx=d)y)!fIuaeKYfkLHjOZVS5Kk7+j?m_(K(u&jv?
zeGKzMq~;=JniysNejU!Q8$L30*)q$Lh-s?xgNF&x&0**%`Z$F(K?4!bE_dC#QvuFV
zSK6r4rsmajN|~HiwtRWyZJUVyom152ymUpUO|jdUG>QvX?6hu8<#JaK43Tq#?Vh0E
zFD=BhFop}Y_7W-gPPB*;IMIOCr<D${=7LjOl8byo_8=x>a*ch)(>lzq5@(k(2vfXd
zGdlRV7l(XHqfv<=vgQlV>!%3~wuh#x_u|1V;xVe7fm!6Ted1XH`~5%h(@%d-TdQ|k
z&sPXaZ$>Lq5!`)>Uev%6I(>JcP}70nRFibU4o7I#u(5`fe2E;-u2%1*_;aU$QM}8j
zaS&$LQ;RV+r!J1#x$Q1u!eAUbeg76<^m|=42K#-d^~UXf^I~wcUvSnZ2kxwADF-R7
zeE#82gpk>9H*WhY_v7Ze8N7%+9RrA?L+3QNC9eGVH~$_a^ZNDAwAV_v;`X}p+<6?X
z0jyJ7^9+KgCl(dzegK71J;9I3OAir-x<4SC2b#>;9Wk;l>m%>L#+b0b+bg3giX&%Y
zgA_A)TZxNLFXWt(ICDU3rW2lGJddmOXxwEwo@@+|=5fWn6Lfa$_(9+eNpJ+J2J*J?
z{KF5d&mUZv*DK6nTnhO1<e#w*BgN>k&mj@w;t7Mt&agoe7dg9YW>D_q(fN;a)`<DF
z9@9<G$2WS!3lx169e(inPBIWooT?q86KrtKbT|ijSsNbv-TPxSFyiwn+VHcdbkD4a
zA##Z>R(8Sh%m^`Ws?L;JGf!0<z0umi=!L%DV04ztN?cczj+hYh?H2{2KUh(HUeA~`
zN$dSVu?+JNt$}~<7-jD7{@u?^ji589WWrn_2pJ_IjG*LY;rjG}b$cSDJO!-AsEry9
zau7t~)HMA%Tk>xz23oCi46Wa|sDVC*@SGXpA$ZmReGqaYEgl+k9InpPX54RgzP|bJ
zkJ39icy2(mF+KI3|6hBbQIy|CYT6j|+UGQ;Ixld0c+PA0Kc{$_S5ObPDMx1%VM!}%
z&OTUY@6>yzm4QU}?jFaEN$}})>eB&p6mt5eRy|d_WB$Buo>s>=gNGn5i^m0@;Y>3N
zVg3ca5jjaV#cby=p$U_9qcjiqNhuJMZ+6G2x>grO<`h}dMj*|X#-Y|M=Goda?vz>G
zn*{o3UZwYrE8(Z=cM1j{Ui}GK%7pAR7$GTWMB{~RyK;THvu;lycVb?MDN@Sa4|t5y
zeU$H)_dk=<!kRZMWYTh>7q|BKbqgd-Sh}#pl{YWh2~nqI&_7>q(?{GomfQ_+y*!h*
z%_md#gW3u{{3Eg#zc1c+3@#sk;4gmspJLVc^3#pKKmMLd$DRZ;SL@c&=+!*qLOSab
zSksDy!V<k;v?jFqEc7<$b#P8QOffRV-7!mFsJA<}_ZuMo?4ratsA`<Dv%`B~kEw1-
ztFCY8`(RY(8Gp}f-*@DMNvAdl`g9na!aCph%pj$eKmPbveEj?)`|f%1eIj}x%)pm<
z$jB5Kexm{5^AA7r`44}^>+ME5Zph7xEo}0+s&3JpV{}ihIw#+=f*F>52+kSH%WN6O
z<Cll)EIG3++e5@B>VuHBZKJm~RrE|+WDdoeLP8BXwGP5*0rMjH8-4O{8aSuoB0Pym
zhz~KKQxfCx{Xu57gAX8!QOPNhKP=>@XVP|cyIt#)GU&C^TO-QAqHC{uA2U|iP>rPI
zGZ_YkRv!m_W;u*8h<TwNg<ek`gMW_GPV2K2X6V;M2hQn;k9-K3qt=m!D6jWMj7}9Q
zGbGGh=&SmRvB2tXADP-D|5-wvL2)s5R=YlO1pH?@l5{#hVi0N_l;#75_T!+{9UYw*
z9XW`L2dwo}$oD%jeL>iNv1cmhumR7FXj2E&=bW3yX*aa_oyTyLMO}N%tdAdsau6e_
za+!_avTbbJlNYkH5UaHjMdy&bN^NuCoBRvV1U8Jz(?Uvtw^A?_SIujnF(Aa58x9hc
zNFq(Oe>dE(U)<MOs;AKPli!Gb17q?jghx}x`;Ln6{Cwrpr)M+;+TI`DQznFR&XC3i
zb@T~~8xk{gZ%;?}x!9_WY-vJSJXE15J}){3G<X<LOp#?-SYjk-oN(C*;qePkzM~zr
z8K{!BNR>O$Af`wPt`4uYavX&^Iw{P!(2y}2Hk4jHm`F^h^`NxQ)AgAY7gPpEch38M
z97HujPFN4Lt}Zwlo!T3<H>_7i@MvVW-sv~wMLhXWXX$bVd#Ij@SLVSs6CP*8CSvsF
zSpiG*+1t;TD?j|jU-J5o|1hnN4$qzbe4`UIa+G)8j~h>y7e0OXBPxlnfBrjK-6x;c
zQL`zi^GSqJ{XV3e$$23!nKf-(wu@5;=7KvDC4`mrVw9SZT38pKjP`cF&FOzawZVr!
z{4Ia`xBo2y{P}+4_`m*NNJDwLe&mPGe`G)Id?`Qsyp00glo(f1zPJ!I38OWvIpQ=J
z9LGU99GUw0&p#7q^zrrS0}<s_e}X^%L^&Eq>4fA5yBA|F4s4BwHlbCYey3wvx{NcI
z=y$u0)(GO3tTCGV$$jho=W+-MNEp1l{E;92_}BRO-kDm;KqaCvxmdT?IoxTUp5li~
z&YA7`qd(J}|DQrghz)<nVnWUw5}OuggS_N9>CSToe7dHRmSt*M+&=xViHRa15f>K{
z%R~<p2r0ot1CoYGT)LL#VVZO-ol!kXw@pjxw}wu?9SZy+d??OY0@I6LOPyH_v!R$s
z%UKZQ><D67E|52l>H%+W?>F9WZ}j6Jt8qzzK8)Ud!e5KkkvTG^GKakRA?xq=OfsFh
z7z3e;BNJ)P0THvAWg<X8#6`TOV`c#OcXErN>mcGRj%+kyNDRhfM0yB|<G7mq-Jebi
zhatR>@gYU#^9YJN*?RAk<6w!AI%$X$vri@r(F>r~U>qFxJGb|PS`TFSgkA}A%IidP
zdde*KC(?Ey-0qIJ;^UTwoXMmv%JzY09{glS;bk^8#Oyi^`$hHyII62o3lEGiI&n$F
z%LnxN6PmIM2WgbDQzQ{Jvdn_IkKy)5;9n$X;OXPW51*be>$H2MzCpk}rvqb%t13Ah
z_8;A-Q@j^Uhy)s)Iv*w^P-xhU&Wr(`wuSAI*yk+jl$4J@JaK*6I7(xTGGi`_Lul#`
z2yJrAeRwA+O43BsM8Q?A)`$CB)zzoHdro(VGA9}?1WjH6hrqXa^PX-9{_lsscFy4z
z5n|4B{=7e&3)OiY5wMPRysJNUS}(j@mMIGIy{a|1?~PKMr$`Z4^1_mmi$+@IzVF<R
zJ9J}7i4<32T+xu|)*0ru6=4doy7QF?bH>5+Qk&}|g7_q_-rN`4d-uX%RG*QJE8cB5
z*C!aI?7ZE+5Yrz}+lZRT%R*RuaK^H2R55N}?hF~E9Nn@N65IC7)6)l9^#L^X{+0V^
zL<ychP+h+frtNZcM{I&+w#x;Pz#6v)2X@plr~GxGgSl4X<I_jhC1XR__X^Dmjg}Vv
z^rye&umAYhm<|5sum2m*p@{JB|N4K<$LCMn$~#H^fq!WKKzaM({9X-QuNyrl%DoUh
z{&DiO)oGB@=&HQm?xgsY%k_y*PoK$YaaH#6cUbTjGe++~RK&Z7;NsF@f?7xM4#FW`
zC9RRP{Js+i$LP>y!uT^-iL>3sDG>uxC9PMk&mXD7BZS}Yckai5TMxq=E1Nu&s1-`_
zZnTdx#3t~*@AT<3xZQVJ8>HwoqsHLc0r_^&p6I4^qx9mMnbCaQxvGn8nY3AJ{((_K
zNRL1-l_~IQ1DUhxQ%d9z=IB`?#zYK0-dB~!0o6luZ+SmyuBjOin@!IdHYDc-b@xf}
z;=CNi5v%#h9@kZbm=e$7g_KuB0=GK2^_{oZSN{C-7jFBG3Oucu9KOv$XftK6_fG2-
zEsFIzmHw`|>Q+2|p!ZoM*khlWPzI+3@%(Hi->&(YJ#gCe&TEPK5F*4a;X}mtBb(wy
zQ98i~Mv0<nBc|*+iQ$py(i>L49k9JOzP^41>-2RYYzt{!h?gfcW=1_c*%4^1deUki
zkJ$*P3rW90#AwQnB&CE-gl66VFD^!>54Q%p9tIF^9LJnZ;HafNgp%j=I}0QmhV|hn
z@bwwJJRu>_Yr#gPR-*+bW%g~IX_$U?{g>BoOCVe`m&?NKcH?M`G6ri1bUI_+_*-ct
zl)Y7I=^T5Z9<BgQB52NBmW%|YAyf%8Q!Xj7ZHbRB7q%^O+wX`Oc?m2__5`hA->Th^
zct37fYjhRL?y;0<PWrp7>pZLjU4$hoVdg^Hn6bMvu*h#+q&Fk4i6yTyLCkdybyVm=
zKP@Zbo04x%o@1XEA1A4tH~4%39L}_1)V*-LA6TzMT}aUw(&yOF&V6t0j#wfo1-5M#
z0|IYvJ72%PV(nl{7q01vxIXjr^wBHUV3=*BdeDk*W_>=`2v|tO?a6JMwE)B8lf|eL
z!t}^jd_3w@3cIerJD6Sy_X@qgBJxjc`4ji!#?NoB9KiK*<@)sOo`vpyUQvIMHLYB(
zAGlm!7`<^TH)?kgOkTDr#+=i^g0OBY$Kj)W+c4H`<N5l;5?4xX+>RX?jwp>$Im(R?
zR<_GCpI<)vO!wj1kn;BCwxOqIet7u<FPAH|S6-ezvTRqPD%Yn!^7-SBe2&6;{ejE!
ziT|Ph_x#=8{T=uB#${=s1M7wmy-;B$-ih<nt#$Txu=m0bPgl>rK;clHKU`V1NVq9+
z5w>d}rAX@@T#}as#BH5=K&+i5M7B$0x#l^s?f1LRE_U!u!=XCEYQ}jjBOw@sM6HFl
zw=a~V@cQ*<>VESgKL~kUSuf9wVWhJ2{`Shjj!u}o^~TRX{~39Eqqjmm8ZmfMsF=r?
zTa0v34|~zcP>;@0eEKM@PV@qnWh2BHYi@By;|}uT(d?*O*O+ubYaO%ZgT^EgMOl(t
zL^UMVW%EH=tv*72tOpm#oyB1gu+yz@(g!XYR&`Z;DF^%gPThA#AKwn#+4UMgP`1^J
z079hJj`c!2?!3Kz<#v0c4X0B}DO81#ea1%b-IF+5=P;lBa0Vhd?A<(?*=A8w>Wpo7
z{vDYP2%XLl@wj%eX<a-IJ-jGB=Xf}RIYXA@3<r}5sRuejH<^fmG9;0<7nXHpjK-Lu
zKOvm&%r_5ztCjb+SF9J76xh@YONkL}%~9l5VKj%6MIMN;2(<3A7La9rKc^-h6wxsH
zi5hWIH)kQ_^np!S-;0_PP4c48!@9lL@P@QueSmsVvR<A^*Jp-x_R;)Y?BDV$zG*LH
z(hpO=^()qIn_4($%OjEDR4-k8jAd;DE0yCFh{C;A%DrH<!%?9$mK3-y3!h$|NI7xq
zonwGSLz9x0j7GP)xo7{}NSycI%Ry}w8Nyy0_tpqeiIZpVV~}RhO$v!9k=6%M5;+E}
z7etLVx~Dx2-@t2c+>Y*1#4GMg)){s(1A~n7#b^+i+`5utnq9Vg`LV$<lgUP(lWxr;
zf7?-M@70sRQl)9275^Uh-nkX;Zl5m7x~}Au7^73h;C{PP-wt&0cS4+E#0&!KmR(pm
z|3B@O!VaVJ^z?z34}aw8=_A7iZ(o0+?yr<yy!)79C!KS{Lk#p$C*0_8V7*sbt5ej{
zU~81|&fEJ}-uD~NmnS}d{*i56L5;oMh#`R)#~1`%*p>_1dSMKA4wNxSQ&pSOLfRHO
z3+r`dSt4J)d~uKdenZr4TWeZSjU0WDCUu}8OpmpPKjoB(q8?#AloS?+@760%*JtvQ
zkm}(-eF#MY+w&)W_=~?}T^2rl{*fO){t2<bKmPQE*Zt0VUr}{I;jw!be#$;dxAitV
z2O)3D`kat#+}k@bt_<yb{Cwpv|N0ZM7%>Mvetafp<=7jod54rX7lds+QcEL?vTl*(
znuz-QMh5JJ>+Kvm<Qg3B43#z=9yYuny5C+I!#M7D%CW;}m_~voa$dMRJ)<JT`+}H|
zLgF|Mj<Qo*<M#g5JKWwK;aeAST)xdP2$7f;(&F~o(ma)}+;7;vquupLLH+rRITK>^
zq(c?1+Y^_|6CRQ7FujGTy*Wc>x?$lQDJ$fdNXzVCRXn?(d8Bv14+afs`un}m^xO~M
z%suK?C(#VORcth;O?!I?#f0$itVa=Iju=Ag(=F5*b>BJeH>c-hj%uC}xMK{6?izq^
z;Z3dkBvuutC~P=oTWjMe#fzX}$h0F4j9}lIBfsy04*4b=^$+Y6%tQqIVD;cM>C@3<
zPO#|YzXxEwJHiqYDKDO4s*`G+NG9TW|7Konop}bm`B>$YGciQY0aG?Qz4~NbJ*k;f
z06MLEwHA6?JyQaC9O!4~CE~6oL}mft*DUCyfaa@wY6Z}#X%c%}j}QXdeMU;383g}%
zQl(9csNi<bpbrAeuhAHfBIwtw-&72YK4`V`{&w?R`##uP!$dj8^yimC#VEaFNAtj!
zQlV6?+v1)7HB-mnZ9gd8*ft+U8<R2=-j9QOsTj~rczt`HmW&zq+412Ub+BzK&({mv
zx>4>sOax(cql@2QwK%=OLNh~VyzVzovMWe^&`v_<KeZM={M5WkYz=F!83^KoiDpu&
z--6)c4I9H@@Y<&p256PzedqRmM+UsU9{%z~rA5I7`@UhsIgF&YLAB0a3ZpiXM9)^J
zV_IiiLlARzp^XhfN-QDJYK35ES3bP_z|;1T4^Mxjna3*M?r+@pJ8AN9;KCI*Kzjg>
zic(9bl*apO<+$IuZc2_qBxBXDa}1Gfx$tuR$mM!PbK>W>SCTI15JGDNNq+DLrJItM
z4N0yqNtY{6pMGFjSMs`GHh6peN18_7|LGqIlWXi9dhks%rG*?b+tbR^^@SyE+>h!R
z568|(*;W7Dt%^A&Qd+pKSN8(P=rmR7eE8urA6`E5@#Q01PAnmlP5JpR{+iF9{vZ5@
z|NH-q6oe%QN^O)jSb{@`%NmhnH1R2z>r-NT3WVHgyi@8sP2O3a20#AImF1JsF`lmr
zOOCWIE^ak>q*G$hi)UdVuI5g^QW5;Gbt#jcn**4J&BQ%*0rDa~cf*>GD(}5}LS}I!
z%`;q@&*wP9#3cHN@x}e^tv1di!1F*pA=ptXJqU5RxP{DI?-WyJZ%#M)gw=-hF~eJ+
zcaP6DjAh%Vm@RU>K0gu-TlcK`R?8eKu8%^*TBpa3W-nr5oRj@vc8}f>T9~v6!vnQ+
zisG24Wj~?bITQ(*nEN*ehKJqggL^(F9Qqh<AK~2QBtC4AVj%b6gLr0h^O%?!lx@3^
zf*>{%1<ftC#>^a;NrinDB)vCBTIN(hW1<yPYhg3YNRSD)p8`3VhxQbS`%Rm${5gu6
zdBoYnC1taU2pUL1+(|TThvy-wI+GQL&T-@2i^z!xozA7vI`ufH`<+^RMox^5kf<0T
zFo!A4si^AIxX3Kra94?mTL4X*2GTePy%WM91{Wmjq%vwB<e`YTy_!j}ocBEh56tQw
zoZ}9Pvcn#y7<CY1V9m*OW%K&j`OZBooBsmE!2NcBDaXB14<Av>@J_o7qu0(*<rss#
zR;b-m;9G}M(aXxR_;}rY=4G21!4lFtrR>yd?E9VLsANrODWnv+Y#SN_d+*%t2c<Z?
zo@3(qa%IUY(#>z^XpGy<M|heK&Kc@i<$EoJaPh>dnN}FpMM;;uu*?95(YqHGQ-?Cl
zg<`e2oyNMyFqg8s*rtEWqORwJu$k)FdguPO^Zw<H``ewYkv#Lp!(}1H$fCh3T@t-P
zuMWw9_~>7CO2y+Sr+BVa-fypH2^_^o17Ef)8oZJZBTdM@dz`KC@Z{Fv+52RfFJbhp
zXiODv_xpO3#_Q{ypa1?V<$kBWC@CJ4k8?oGa8K*9UU+)>z|-}aj&Zquq>NXr^#>Q*
zYVoRi4gpH}%ID9YczSx~<>e!nbz@ssqDJ0dKGD+3*MIyo^{6yD5(3Kt&0yP}dETz1
zHL>Q+Hwg1VK{j`fHXOCOynOWPAMs+%e8i|qWKA0}_!#OnWiIQApd9;74v`N}FZ@46
z=X>hmAxABq6sX1}XVz<E=t>=fyv~V|(P8F&->K$;#!t&L%Ox@}meohL`e2hmZLTs_
z%rywfH$8@TdjB~e_<LJ%cI4g8sA}J&L;^7^ENO*ur`|j>o?c*D-o*Oc)Lhi|{`F_p
z?ZR@oPU~r)oV@fr1kOX|A;=n?(JGdF&kA8?{tu(o>YZ&l7`+isrTauOdM&*7O07F<
z-uUpt54=3T_`HtphhS?5A|6bYa`GZ%h7OJ1-AXuJI-Cmq)8{^|gYpYS{kf6bbRV3w
zqsmDDcyLiv#bRU%_nFD$kM1(Wh;^glymK!*)((Qek`kqMdUt(R{(rOgXFrmp>6tI~
z`&Kh^_gFHss+!e39CC(dBxu2b&W!|JNq72>bt6E4t|S3k=U^~1+-BG$ySCgS!rjfj
zT^H|b9@zs7*(HEDEZ{hhWL0%$WQ4n!eed!-zsFjF6*Tz$!y|{~i0KX5CL}x5@i5FS
zq-kipZtEQvA|XupAo_|@HWAzQp(&u;cIrl@lzTvua~>b3+jUWE0o!nTxC?K=C>uam
zC^!**IuU2mMBrR<QO&3)h9RpxFKd=u#I|=>4jkqqa;X&wRNGcs+AvD>vsN0#eU;&?
zV~&x-vJiu()JhDFX$n#hSb2ULwO}^`zB#JK89VoyV%T&l-I4qUq4@USo1g^mD855!
z(ipB+%;hs<NbGSluq+G9;UxEibJ|P@C=YnO{{N7B^yl2rX=@l$*tT*z>{^A~gej1_
z=m@qZ)(lE#v0l)Rwb4tXWErzvx6GESkojckL{EnEw#t~j4^h2)xq-6HA|BeSWz8OM
z3f^H1<uQvLDEETRBEYE5;bNe-DkJe)>ABL&5L_klrk)N5f|W<X*vCX#t<~PS=0qut
zXd3=%q?oam+NIiaRv;nbch}u}6JgV~7B1IH+6rdUyl)JXi_CG6!bOfeS;4dcZqUZi
zpb)3Kx!3Q<)<&uqYO751f#aRUnVAq}m!7f&YrLGF#q$~@8v43EQA@%a5k@%;3G&YA
zsP&@Bij9;S>r3M0^M$-_csKKWK6AP};p{@%k_0kMGjW<F49NvznCX54>sXWUyszgo
zFY`0CH!fS^+h6}NfAsY?9G8XVu;8teNatndZ9idr;JfdB%ZHzSM=2Xqh^&B#k>hD$
ziVMa{dYT;aO~afrqX4b)`0$3~B0IK^*B8F~^b_Cy>MML3IN;6c#C6>?5>^i6Yub2T
zFXU2qetBjXBx~I}oU07eDWl?aC+eanVi^_ckcj0@hAt1DVP%C7Y``A{GKJHsUe=}5
z)&)pZD>Zk#(?YaB-8w1#;Wza^$CtPEHW>gbJ4QXYAMPK)N7`ku<y8uzCN^e^c1!Gi
zb7#Gt`SEvu&)u6h9FKR<5{#y*_1E$9tvJxb#|~&YGp3nkUNFwF6|p_%oJG4}t=MtL
z9tzY|Y}`qs@%Hf@Uw!or)>^i7rL}s)#*QviA6xe>*XW8-EW5c>oDH|HJJj-OkpsXQ
zbIULQJ<Nl4gqpHy<Oh?V(OS_#yvho1dwQ&p>A*^0Q@0d-mIT#LMDGaBQmT9p-WU$^
z!sEl&ym@>}u!CMVMk=%+L3CyehTW0DP4Al~2j%HRj0Y*OTE<jq=WmV=v#)~2>(7n3
z{clDEdS*DrsN_`ereo<E)TIRlmWCxPGhv>^&6cx<B*}FUe2}%Fb&S?2L-6&^dFCl{
zIGhN<(`trhsClEV7i!uB*fiw%fZI{ZhTtU$*IVIA2zx;@Av6)!7<4)#)3Dt(DIm7`
zdlMfua^F@aUlxP6>B(*P-)QN=_VOOKjcOcjSy+|@=Ob~RIe1Age7zX7u>A5r{6qKX
z&$*$WU(URoFRa%@y%wx8z<cYYQmG(5*!^(VTz7{SzdFh*V{j?;wl8)rP;#S;LfSTR
zl48JF$8d&`n?(3dVkHdT(I$^?GIDHH{%mXG77y&bqdT-R@3evfH4G&udhXO*q3+yh
z*;hFQMp)~>>b}pp8?0o_2^?&jz3s6&zz~1=u99}nOGK!#RHQR%MJrahmGB`jFAE_?
z1{P;^46%<$t$1#6o~o)^YnAfV3%Q6|)o3of*ocM*(Yl26m=2foC${y>_4*0xtS}!+
zVxAU!oSEi@DNdMf@V3CPtm}{RGueqbRN({Us+@P&u4lY&tm#Vhf$4B!IUbo~z@&)*
zlq~6mV+^+Ine)qMoO4`j!GyrS{N`7@yL)6gEJE9Bm8xB%_l~#oiNpOne*FBAyU8;L
z2d3g(6io1t>CGic0lDVH6ldlTpiJ^PhlyW({Vj|4<RAYtFZs%-okHVySV&3o!aqI#
z#AW-6Tr!`}FMRy;kq@8Wv!zuI?P<oEL5P+KPdHfOA>t1fKRMiDvAM!9)ZD2~?3>Q@
z8?31rusFQpl+Dl_cVh@nf{9w|B%AS0SP;!aLbN5n<d;=FjWYsq!ii?!a6EE<e`L*7
zVpyFeIT5=xJ;bDAlHKRfDc!c6anA$Wbu`=>W{4Hlfbu$|nBJ2M0i1W--9K<TENn&+
zHB*y=#H#yeXW@(KQSMb|9jDVHU%mU5B`$0#G~Bi;wPtw$TS-MVr4SYID|LxX>RK1D
zQ9uuHd8FvnL-*7Buy?#ZdxnvY(heOb>!jc{r^8B4%n;DcZn1Md3U>sta#}0Tp%6S%
zhb<MH6Z`gP9q$7750Bj4-HW4&3f)RE8-k=e?)&|o^}rI0^I<*^4ktiV?r3I#HM(zM
zpC+g#!N|{Qh800z7q@`MI7v~oH$A*+Fr>mZ*yRy7-C>>NrnH=7r8I`bzH2<W@8Zg7
zwc-FD9MiJkbwycoV)RPhE{t^Dlfp!->Vv$$Ui2bUh&qJ;hD`D3f}!?$J$E%x4E-3K
z7897NO{-?dV7SpDZNRac5agm5W-yveLSzpuNs7*0XdM!P;H(G;jc8y(2uw~0xch6z
z@c$vNpSLDB3(Te0rdiaMjvo50%E+e<Ms36~n1f?N#%=TDIn1)lPPI`-C+@iht<yS;
z(#U5S^I!y2XN{xf%6i#wZjne~XEC#gFJiFlxY}Bp?h`F>auRAZb`&O<Ch2KHmTN+T
zt8#IPAW$a{gWI@qSBcs#qw}3AHfp0;$q=wk#3{6fA0h&CPIAc|hMudE%7ETUrBHjv
zMTgUrwzk8ZG{U$Pc@Ls~pgBv9wi!)mVP#{OPMBuyAKo(0M-c!<qa(dNV%LbUuw?p3
z;^HVmJ_~NayBX&q&P^ES@IByc#vnI9un`W9`XKnv-XsMsZyQWr=w^QAo3Fp&&BFuU
zz`U*U{SN_Cjn>IElT&6}H*&AMIo&gd$P|KDle)Yo=d^)$oLr!1G4t1w7y;rWG_}zi
zTe@H-=`OFA7hbLx)@|eN{+@F87VABiw93ygPXzCIe0<A%cM=czfa`VzI-_QuKY!%I
zho6Yk#800;^TUtdasBw2!!(f}AGrVOD|&6D=gMh1aQ7GpOT_qgLz`M>V2FruROz6G
zLb!}&-|7nWZGdfZPz=Rx8!HKSO;<VAdf6q{O8SKh2z54$vySLJ))|h=LR=OtoF%$^
za+YK8Ok*NvSt+!xXQj@+owbB8VHs@a7o797&h3D7H!E4wFe%m68Oy`lcf5J~wGc3C
z(>(fD>z;%g8To9+tLQQKz{A5EocC-cv*yHgyRxnq?CzU4V7h#^tu}1%>Y|VVLu-Q(
zq|3KjBrtYdFE>Rv`#+EUZ8KVH$N{pi0IW5)c<!Ns?xDqiWIf2=v9IMVLQD+K5u7AM
zMjzNrVi+Su!W3{uES;?vap81?^D$4_^>*yKNsfwjggE2EBz%D`;b5aSwIOze0=>Eg
z#66X1L*kk-hExh6XMxKZfr%L(pu*@`pSz-su<=M-?yzxE`&<!ZbV_=7i`{*PAMJgy
zK^nH;J!3R_9pXkSiB=`a&>Cq7eTew^C?eA!BJCLbZTefY#=3VcdQ|N$Ff6@pP<M!C
zA6P4nG-G5^AfKiFf9u4A`2~KuYgU7rRCH1;f2Xm5l8d-Wve-(qH3aK#lZibW;}t~p
z55Im^F>t<|1%QH;U8Wlh)9EzI5KxEH=;6s)PF9ru&cGp9-kl~~jC|NCTa(DhQJSjC
zEp<z@tRE6XTn5I10&|=VB2Y6{QnGqwnj>?ZRnU@YZM<IiigW}gRhbi@MH_=_hQca}
zSMQh(lZ@4jRA|<@n~b!zPAlTqw8k;(Zp(DJ`ExA{BT0Y5`&$=+?lF67boIJBn(XGh
z0W+wh;t;j3wLAH%l54#Mij-V9pD%1{!YWH3=R|OpyJ_NZIEbyr4C=^?QEsZ{KDw%F
zb!S+GC4N{Qcs#zN8fd+eN@k7+QRJF{84-(Pjt9&Pm-xcaadyfXJIzGXp$qT(v>b@m
z;VeY&3C`gTM@$>Iv!mC}vK%-aPt0_dX{Oc2niJU!jFGSV`Q?dEmos-h@Nzlx!}q_X
zt{Y>SnWh8YMRKj2ug~&OL`HQ%cc<QZa;Z#{A^1S<jbRPLS>n8K^pV;nBjbF%P-~{=
zjpxrF`2PF9!-dF}GeH&Ioxz{{>A&V*efuwEZSe0uGkRj0J<}4zCS>GVi`LUS*x_(4
zNGQt9EM#sEBx|kOwkmx%nzKme;!yjn{LWXRT-OcjFZVS@Cwfnep6EQw;lQ#;f?>%U
zy=>T8m<)tzq8M0>!5Kqtm7+i`V;o-btL&oUn(4#ft>}N`K`8efMi9Qn7^cIK`FJN|
z<vt`Cvz6B!z20wbKfHA1uv<T^H3E3aiSv5l<>e!3J8Q^T6!(f5H(knWzQA5ns7>%T
zDz=(%sNLY$amw=lB_jJ(FJV9h%_3fPsuXvlE74DF%tn~@z1N#U-e~tag&^TMR>HPo
z@WdGCMY_*kGIhI9x>LPCC3=}A#CZ|$mI3P`1%t0b1w_erwIph~Qu9?-J*tE^Ml)uH
z)dIubR&y9bX^mBzKpSKlD({IP;bjfe!A~&X(~V~j54v8j<Sk)FBYOGyY!e<#)5O_Z
zhDArkl!KBwy(UH%N-1i44b~B+8Mhn>%N@%!VT`<oMt|=f=H!jRJEP%3B2a9jV~0rj
zZZ=9Q&xmez+n=D@jT0Pw->JE}KH9HuYYo9!nh~9V(R)eUAC<JOoS&XqLclFEmPzsL
z14ffoX8$MP40PMM@m(|))<%n4CQ0-ICL4$rW(Lt1Z0{`2^3C0WH>ZW%J1<FpkZTpr
z_ZS3g<eO1=&K}@1yu}93bX+*z-?JPRIas^wm4XwR*RHZ1O3(8);+*ozwc>Z)5#vND
zjn$~P)4=ICb9Y*ZlX&`#eki?4<RnNi4#P5siOEN%V1?8*2BQ?25E#R7O-8`ZNH|6*
z8hX>^(3&US-cOv4ffyYP`CtbcU}EGV>Z8-D6gDME{BzC~Ga5cu;uHzi<3d!RkUS7t
zSuqGjP=bO4kTM$B$<u3R3I~pdN2a*Y%9VAyh(KkQG_W>0%W@Raj*h!p%cQ(9oG>JO
zt6aC0S_X$OvuVUIUe+O#cbH|NrOG_dlwMetgAgm#lC<WHN@Hx6yW<gWEJr`_{ZBvg
zDP4L0^b_Cz#ebl01!Ek+FEAXn7Phj{8yw<{w~oU!bC_o0a?j~-$7#94V#!muL8zP$
z%yD8`W?r7pY}>|Y5{&ct(?{MszU62Pvo!({SpwAz?v^75H{*;a#EIvZ4;-q&c6lgE
zm=ZV}7;WF{%AMQIz>JRRx(`N!ne=n%>Rd5Od+d9`)+L2=G$beE54@rOxxS2n5ChW`
zh%OLAq_xhdm6BIVT5&BiS#U8>tnSR6$CZMul3p68nSmkKLa9~KHv14UnR5d1>Y54f
zJWji!<FXLvnbtbjO*kp%ZDnhPzT38}!%%Nsn4oTvU>jp7+se5%a&5e<XL4Qf){t_-
zV3=n?gd1-NW=|7Dp*;%ksRi$*o50ZwiQoQG=g?bc*l{Zct<h9HYh@B(Uw^sL)!VK`
z<X%kZFHxyWhv}`Qdx<&@P9o-=6Wu{a4v_VFAy6?@Vx1@Nbc!LP`!Gd7I@~xf*iz`V
ziEyzNTE5WAD$^3bkHRh1OL`>TV8>vZCPh-e?#zeo+u47<CU=O~{Vl@}Oqdyd7I0nJ
zxUMVf^{P-FSu@ls6WcZhweDatUC;I=R@Da4hy%fzA&vr_`tF^5oEGAIlw1eXVN_a@
zh-Q0RAKBOZ2TcbYW9-I8aZ9S^1$DQy)+kj(fvrgrs>KLgYABUbvtn+O1#5MRB{ucm
znyd@9jmzbQIR-3p1_hlKd8yHK>imztejaCFcJyM<DYV$oCQCS4=FKn)%+~NnOCYqy
zb*o&<V4^eW=%!ay*XF`jI$aT$-H9sG>&4fN#RW%qo@ttyXW6Y==b%GIGy>g7^*p*L
z@SSzsolYDNGs7CTmZ(htpi+&BdbqcC>m1@Vv&0$a9Jw`$c5QUQ8@JX-IWt-lcnAhb
z$rHkrDPYy-n|q;m*-cvSpjWZ^wT2It$K!#2@zo>C!x3{haM4j^X;Sw|uDB>@`w%qk
zw#t4ouN&4IvDD3X93I{gk9U|5X$)FTv|5n}l}^ctl10cSosK|PB>l=Mdb6=zF5=Lb
z8q<8#Ja>mPp4x;oHfqB(VG+zxI-%Mtm+J;D@cob9aep`xW8nGo=Np>UIEM`mJ4Kuq
zx}&9If~V8D<drZ5X7@mJ9!BTKnmK>`k@vs-3(h}%Agu*I-SOt`k7&lQweKV-r8bFL
zj>3AVqhjn}Ii5J&-3v^~8fG7`we#k5&zsYc$$7593(c{28iO!TA`bEabcNM>;kun!
z^GfMLNPIb;*|x%Ul@rbdf2;cGjaCN66nv;kQIrRf_wpde$(T_&EK;@iJ@!=NW5p&e
zqOYnC0Qbul5blnk7y=Bz5<6=#tw1fpe%DE!jOwFYJKp0;U>OGE<lsm-ONh*nE_{#4
z-qZ1R!Ju~z?+?sTEW>RKt}joVuP<ca+B?M<jQ4ownS2zE$0mU|!FYl-HxX!BuWaK=
z=|Vt-2K#DJ)&9m>=4lZoy(W-$BL#Ntl~iR_DONPqF&IglGtS8>Nv8;6{I@s7{f&0&
z+>PZwungy@RmS(?q|w$wCsw<EeOQAv(rtQY@zyZMiB>zC$#it`;W~lNG(`~?TT9D@
za$RvQDwbEK3}$~u=FF&@kYMQ;?{4~nVXfG%hlJ;d$hgTh-(_y*bmwjq1ZC_iQrYQy
zJDdj-r0J=JZQD4X&y<u2-a~ke4jxKWm65tmU7HlxwG7;_bSG3u1ME;6qs3TCnPtdo
zz&c$WYnXu1_ZU=%Vcjj8$66~oDdia$6<)gid#F`K?y@Fn4ZKJ71PJs9T?tq@+?}%w
zD!q#Za+pRRh3?#IIG0{GrB*<6NUck8^=eD~hhIOd81OODoh7X@(yR`qXqgTH7c3(+
z=7DhQorSHj<%)}*YwPrsdATN@*CdBu?-Cm<SG8Qc&M-SKna!%;3?bmG9KKuc<lMMi
z61_L(s5ej-758D9Io;oLe|Jai71EXKb)^^Sz;iC-lql<lQG75uZmMRGa%1aVh*Ngp
zqGfgw?<{K<Rll$f6rqXi<ZDt02(q{9%0v*+WTzCFT+GYDn>UY~?vEsEN#~W+g+OSn
z#bYF3$U5p6YFldLRJg8*<NcBQ`!}2(-?7}ik>H*#R5G+;tR+!vrIbS6iu#-7rmnS;
z@)IF0jME*pCrZn-UNPP>FEjHzlagag7fP<T2j3hW)4Xsx-7zgka;ZF*XP!R2=co5S
z@Xg~_EYn0uiCi)xNtboruGk=-qYWNAd!ZH1GpQu50Ion$zcrIeVpJ)dtTnWjxP1Oh
zSu@Q<zWL@)xI4b1o56Lx%AR;MDu!w;b##Ug_~VIiSa^K%mT5ll<K>B`^D{>q$zHB0
zhhYg5Z{NIO1USSyzIyXD$K?*=4cD}>y}VHB3axNHKlAkP%*T%(dHVbb8(~z#s0v?k
z!nCNhi_@mdgEUT_KB`>f(KEa(kY&H9<U+&+6ExtaJB_FAoty<ovwP-$$**~e+#OCD
z?YoIM&l=e)w<^kOnSlbbqS-q}=Y$eBj0TwHENi;mL8Sj&a{{EJ7u`<aa5yqg0_3`G
zE9-ipTd2;_y_cwO9m!%ekHj8mIbpjz?DutqjNjFoYIYu(aNzsx^Ra@U28ltJenZwm
zqho3%*CG(9A<PrO>WjX|S$9*Q|Ca$i`#QrZ&;RC@Fnj3Eu0c@HO_!;tGZw6_oQ9AC
zt&yT+HnQ5nSW)cn+BPjDH1HeKDrN2BtU27>qU>o6(<Q>)^nx+6jxcsV+YB8iOE6=I
z^TdvV_d!XXW9Za2N>Ee~@KRSX-!Q^(iE#ldBHQcrdJ~HeM6a`hzpSzfyb%#)7=vsR
zB_-Lp?;;B2D)iSA7<lD-v>~zdVOWN#H&7ACn#JfY-x;G(#?tUw1=V%6#O|33`=9Bp
z)2j#~OB=X0aARmp`3@}7S;jt9keQ+&^o_2Sl}%w?6u~=sJ+`(hQ#IghUt9mfub)*6
zv@$Tx;)9{vCP>WzG}-MB2csLFi^O2A81k^RHS_*_<$(h)xs#gQI59}~+;?G&br|m$
zy%L<qD&eh^!nP)Ik$B$gTG(=tm_{3klhdlWD^)B=Ea>PIx*#}*L{6rhNb5#f6RlKy
z3_3;<wD96Ru8Rsb23aeN0cUHex$1aEFg?+GIrN*@3A-wtyz|UXU>3D>wk>l>;-5A?
z5D!Pfv`9jomF}^tdbuN{D}xA<QZ2k(uRL9@TvB4OM?#FuG2*=t7F(0dgVAJUw{3}%
zlL{&gVU~%Cb56uBwQ${@h)X2K$np3{jE%$Lj^pvj6eqlM)S5WwjdfeuYJvbhSj}da
zpL2<xPfwq@oS(${6e4pJyXMQ~Gouz>o<8vA>pudZl)~xmz}trhPSZ@TnfI3`u4w}w
zd08*Kq$}@L$)9tjR+!=mX9IqAJUo2O*YmG&F_LPd)y#FfFf9k(JbuMDzxtN@!yVpw
ze%zk<oSta@z{~TQ$NK}1hokIGjbS-V%t(H{^Nwl_)jE=Q<T0413EMIsfBc@8^O?yj
z4Gb@={UKFsN0JoNmKj}m2i<j@UNsp9Hv<l_%UW)`M6A(xZS69{c=aLoOiGo!<)3q_
z{<C|{Ua-7dE(|4rigLRV#A2XhQU$W81}Ruej2`0yYc67s11YNX!QBQony8Fo9WG8B
zj|*?#eTD6V%jHGIgjJqd^2W2{;Kmp?Qd<m@Bmye=5MIabL#>9!NJ6R6Z2302*z4va
z-bWF^(J7<fI?vua!%D<?>pLJxOlUh3AUeqmvF5b^cs=Z8)GccRiD@5uf#AeO*RA{v
zVo5{emUS{Zb~@fJiWEz+v$B5arV{BmE239ttn8+{(Mgri`D7*LMAJ1dVvVg?vtAmd
zivD1rSmGhbULR+c<-n+2xC|igrF?6GO4nk*7(?F!;S7WjB`nN(woO)&;r4rntReP~
z({!AROd;Yz5W=H1*c(CC;)7b$hs3$dlwML!{Bpt{@3GcV%Eo9wEA*<*a653jMzD*<
z#A^}foW(X-`8q^Lzz_s>3DPj2jjo{Z4WL1Jvh5X}f$1_47|1%w8eN$TtnGrU_MUk;
zG9Qlk-~rK!lvV|Z6oUNuKTv%9^NN8qm?ukchTMgMSF>3Af)jFHr}I?Ac2Sa~HBCsd
z&{`)|@#at75xpc_Np^#m(Xo}A0%cUU)~R*ya(<S|Tq7wPgE?B}*^*nq4kN{gvl83r
zjG{3OAvi&Cs_)y5PP(p?(r#Jb^I>K_9szlPTNT*&5XiN1F@j^x`^aTTI87;qEoGV#
zs~mesl0b!IZ`G*1i4Z4xy7g2XX>=O)HoCLc$%xp?%`;jdx5hO|Sj(0Z0JTfEo-a@I
z(a6QHLmz5s8o?{0qYwh~JjqS32buAXc|LJ}`byR>;VPPj5EPgqW3Dl};$Vf!)s>kp
z4@8YAHQ+}huUEG1%9}+9Z!tth&#afvwCzeSf^0P|^WYsz2+VpTo6)&!FI=0E{a>xp
zF%1}{)63v;xscyh92P&#+}}TPJl%6R9!X{6<#J(}7ZyMBaR0#l@xZdoe5x1z{LlYA
zKco!v%)vQ)GhEWj$4{RatuicZ=}I<m|M1AK@Baj24Zr=nUvVx4mny*+{E#HJZX|T4
zm4*#ch?XYw!&WsS7iGH_;fMF$iIWpBR_>B-VI-hPaPE!Kwa~)dGAq_=;`~y$y!^6@
z0p~2GY}z?`?5N`9w?;<c$OD}*#12;rR{iI}dzyA(qiKYBQyZX@12t8#E%qT%;MB_9
z>BM0<P;=Iiyw2@z4175(g?q2|zafB{;5}5YJ9nLT@13_%g}-*^Mh^uaBFl0T=g=t9
zo!ZxZ|32L@+PLh_00}v=H>9k!zn@gNgJ|TRU;q0)1=)X(b58Ut+8s+5E1H9$#fY`?
zd@*g@Y@X3iOv{3`15d~GCb|KuqifxP8*2$cnh<9#ewr9zQtRA~@0H0z+lwnRupux{
z2jaA7!6La0W{6OGPd7v-!TyW5cfkc)nQXd<<iS|YZi$RhNqJZ9D{;>{{Va}H?-U<g
z>E%_>=!2|)Mz113trcTEewm2Jd;ENmpR0EP66F=4Hfmh9T0nREhqL?V=Vk2AKDB}9
zrAe-8t=yd?V}P-NFwF!9wq{JXjNB1yt&(nya1A5~s^blWWnp>y74zK#^Ri%!Fl17b
zgwGwx|7xoJ$6r6I7|7R*ZF}09h&#5@v`em%q~~NL)H@VS&|9Koaz=J1&SGjKcuNcd
zk1z~e)G_jYgR4q=Ft}`q%SF-(MPX?%!DCHiD;1aQZOoW^<&rkWW#7LWrYT~FrR7Ss
zXguYW(FgC|-f=uF%!iq~(?OgHwNkWOPqk8tL|~?z8DkK4m%vccpR>d;<Bf>HM$-df
z-_;`@=pjT-%Yt=|3gjvwJOd?gg?(4{nsuE@7OZg}>RldOwhR5)ke=n@m>03mS)|JK
z#t`P2rna7e?m&ZQo@e|R+~2)rneT|p0&Zq>t)2~(vaxM1Z0m)b66@u{dR;LgVvM1d
zEFh&)>Am3WAg2o{uQYwo*)G25Q46CsuFq#aefWvfyEmlrmUUZk){$CaU9YUy3zJ(o
z%m=t@Jm0_N<NTdGuv6mW58v_Oc;wh0Fsk-_eDj8P5BC^ndAVNMwv4xq<2<uW69LP^
zeB#~xJJ#n<{PgsR@597|Wi5#xzyDj#A3xBhnNOD|zA=_}Z{G3kyRVtN=U@UKpFi^N
z|DXRi=$-R+zIAoi^Tsp<S`-C&r)dbxtCh>>h}j@uBx45FL2Xr3>=;7y#2%@mlS5<a
zfq4oyo7)rv7gH(em+dAnS{%vy(uEdj3?AiiAke5*8{G_iZ&GxjVkx+ZJ$GgO{@!|P
zs458F#eX3LrfH_tp_)1&_J!aX)>0I31G_~`uGxVU6VB~uULz7u!_9>P<3^d=sesl<
z^C8&<7NeQdUNZ*vm|nA=9=&4}V|(lNbu4Uk?IG{m?~BhQxTn89<jr2FNH_2CIwi2W
zY7+FbK1`{7?_j}?E=@v+4D0dUFj>o}nNe0uuSm-(Kd%o_{Q0Op))<_(L^osnjt`b(
z$D9%<Mt3^gv7GL)IQjfcRWYPfthf_hfc`kLGJ$9!A%@!urIkXf6+1f441%-F(MvI|
ziH&`Ti)fl><+n&9Dgkol1_<&oh$UDlrGqS!>Nw%&1Ko;H+4PNG618Sw#USy~O@IFJ
z#pfipMMvr4?bs&(cv%yPR>Clvk!cnK<0f2O#9d}BWSLx`VnCzQ9JzihqgA|f9PS@E
z+`S<j7OayMdTkXuD#1H}=jk+SzvTY0*Uu^jwl#|{pMmj~;S8k>S|9XWX}&UrD7$jC
z<9P#j2XPy0wQ;SD+QIfNqaWw+NCg<8h=!UG1$EC-VdmPnTvtl2;;r=#lSN$BdS^Qw
zILtvl%u?{{Ml*)PGBE`&RKI}G>3V0&m0TL$z%*H=C1RWrWxtgVx>P}kZYi;?iBF%N
zxSm&*X~NFxZ!MyVk3O(WLC26{CX-E;xC1QWpI&0*Fip3~hwK_<-?(=sQe8__aw%d-
zG=1Q$BOnLy)5{ZEEp($6BQqcby!XTqm=uv&tGs{14qR}!;Dznb20ko=urSXjte1yu
zFNu;jF6#@I=TBT;p2%tAa@klfnZsSwE~-;Xk|^SuNF~uP&z#RsJReTfF1m$YvI;vZ
z>-m|_AHJuI%ygKTouzi6?meB)l-Bsw!#zu!ImemrzyCYpJj*Dp6)qp%^Zd=%OvfW8
zh(+gcSa`TQvh(Q%qJEyIfOm#w4a+q1aCl(B^OROTefq@P<(}vBGao<vMB571PNdfG
z^UPO|Z+JW%Fn#d$?t%N$BdIptU*40qB=NM~VNIYEacwxq4Jgy~p`3E2<FF2l;#(`w
zQh#N+_fDvtX7rlrP24nHOXMn|y5r(lmtQt#AOu5jGKw8z+|aiU(Fqx3=U>L4c7X@s
zp>>I#H62W=^4##2)}7?X>-`hGm(U&~Q^pVjtz_~r)ST(9h%Up+Byt!_(W1NcuAni4
zGtym`-fp2i2AMGK9eW@BwlBsgC%<;mdS#j-%@~;^eL<hwoe36h2fYX-b+oKIWs4L7
zA|O<)0`}G*8H?=phaSFr46cs;Wt!=4Hf4~Vb+3hr<S2OO2wovY>Rb0t_z#nJbUWx&
z<zUD})P;ZxfsT=tsAfxq5U}17;v`y^A<xah8PghJIWZsaaDEb{d@YJ7*5uaUB?!vv
zltli|?a;Q$OAzdHCg&?u33l>M=(53k5huGK0JLdVJ(oIjs_+tw+IPLT#Nb5FFu?kN
zbpjPK&VzN70ktOTcA@7i&7Fy0gqu?Pp8X&q3+KGLUo^L%_uHyou6g-qV_tXoRnIT9
zSajagtGxw1=?MZ>R>#g6Lbn8GCF4SI+Ji<e1zKhD<|b$yT_V<B^(_DJ>*wK!voO4!
zG5FbocT{wI(FZkGMo<`0DY(|TI~=$_N@VBDTKRm<q}o`zXPz396>H1n)taHB!pe}M
zpmchY?lPA`S7^h);A7#k7A|S<a5r&(cfdv{19IyejU~niLGZrS$;j3iM{5Es5~0eV
z)Q&BcTtxU+#~`P|cHLMnE7#|X<U)riqAmlySA|rdoC8sc8@)*ORIRRg1$!ql6yaHO
z5kIh&Zf%G#QPxbV60H(7ER#V)&5iY%Xx-vs;BZ(pSxpI4)5J1~cevKVRx_m);+SM?
zIMjn}4Ts~AWt!<@qg6>I+%_5a*Sum!6`)0FVogNIfFXDsold10hq038c76WDv-6~0
z`SkRY1O_1pS<?!wtX$4dJUxHn?sSKBo@-jkHA^+`wNTOZj1-!BAmoAzk@;{Ghrr>8
zkFw$s@u~bw$J3G1B3|26iy*C?1LK+e%vKY@dhU-8EQfn?HdJp|PLI5O^G-AZRtm#w
zZy4kGlVAT!uFuc>_}%YdBuul|D(!!-jpiESgOUeitChAD8lEXQ29{bH*K@uJd-i0`
zUK_^B1Ih-cD}^FV3TrtYCZ-Vo;OhRL>+3XoCcht04u@g#UW9@}(6(hiU~8p2Pwz#p
zodNrW;4HXq@lKTEwN|<3Fa+o2X0#To&H=BZ^VY?s5uC$fB`W%h0qTB#0XLzY(-Xf|
z5rI1EWa?#f-3GT#e~eDvHqI}f3BePL<1j7Kfm&G|dM7l+eTumsCSI+Juek%S2-&Wd
zQmX*5FE4~wn9R<h=y>Nv0dFmR$YiuTN6`s_D~0F-!$_vcmc^^yN?{6jqNmWwhqoT*
zWtwQ6lfPjM@#a@O*m@+|MOK3`LO)R&WKq@N!;G;WdeimhMk!Z%l^FG)@#O*v+Tldz
z^%<(yMlBnyZj83li#o4{Oq#=xBE^S^DbA8$%1)on)O@AYq*!Ba>6pTAQ09V{^#?#3
zO<+XnLd`3qL9h<gnP9pBXTEs<<vOuC{n<a8eR3d^WwDq`0nlg-Mo#4IN-Gn%06Gs=
zNHM#0!B}(SN({Sq`-PTQ6&Fk4;tk9$%Jqdh(p22}hrE7PF;H~0IgG(IRizk(@^lqH
zAQ{Zm%E?;pmzm{wkPq~lS##!`3#rOq{csPyHU@@h<R(cSavRhRy$-AqaR@_l2Wpi!
zV65P$q#+;!H(4J1k>EU2(Dc6{s7;3KYetu#ncf@LNorheoh?hJ?RFV#lfzgddxubh
zyQus3kE?f~G4|f*QL4XQC8r=x5>r}R<(e|rEphQzUS6)^cQ%GKC2DV=G^)OPPMyW*
z1uGX&6+bG7tp|fMFi(!-NyDFf5Jjz*--Fezr**b1;hdKX8cVJ_lVM2eT1i~iGndO}
z*7Fm2d#1GnX6Oi4>^;tk{$h$&7z)PVs!^e1<>l!oCfC^3mFLSv-g_fa%fUKYPM9gM
zt}E-hvc!c}3c)yNgOnvN;pKYax~|f-4?&3DfBiN0caI#8CtRF)_xQl%gN*sGG;25=
zX6~oJ61<LgJzJ4{1~WRfcj9v3?#)}~dE$RQ3jh88=if47<l8^_pYW@Px3~_Mt+1xd
zj~_pgU(QT+;`HuUY?lk`x-rIJQcGc@lX?=bxRpaec=gKO?_FqU15y$@t8+q{WFTtd
z(5Z8$`AVya-oYEsJO{irOu=!tOh4yV{b%<&1;?bLcNcV&?HvJogK|YHDs-161Hu}K
zDsL$(inuF7AG$txopt15RmmM_eUP*8HPYposs{^-J~vkGpD+wTp+8=QYQ7K2Pw5TU
zi%f0+6TGei#O7!9=-pTMvJDb?T-i3vn6Q2@Mq>=G(*apEI3s`FDWBf!_vM|}f<V6J
z?NCtx;uxwq=;kK48($O#c6TS}pR-P^!PZ)ubGJrh?gLDLB!cJYEF~qjv=SVc;}L5*
z#&%FpSBR7P+$$|hXD`nNM==|Ht%NYE4H1m>glW-<r=d1UNi5ryv|S}|V{}SgN!v5t
zL5z~rDnAQ1foAJna|bp?&sb%@59KRJet@S8bJGL#Aq~_hl~EF=W;~T$u19Z_v}u85
zZ||{uj1oT8OJdX}@3)x$Z`l{9i)0w&_q3q_a;Pm_ERAKiy;jNONu^Ml1jm&syZ`O-
z%ykZUCV~;P@lX$=eD1o27z1yOpqp!@T{n!olABl#Vcb?9uTjVHy#E7hfuC0lScJ}G
zjnMa;akL>fGF{UFdlT?VZ5-~79Ofv1iCXD)<=Ww#Dr=T;QmU1Cn((PHrAn9sc8cot
z?+RxTFK3KF@(0E~UXhX2x^2YO@sKBuVd6AREMXG6Q4s}ctrBOOb0wt)2+tX3Wn7;s
zt%~~78W~FsjCjXeqt?#4C1P~URd~>S7#S6pN+?x7K!cxV-7E^zpynzf$aUl6%L{%9
zbZ6y3hXp$rM!pXhgJgKuPS2H;3hS0gxiU|l`7m=h3P5F2Jg9Y6P{cc`*@>@tt64_y
zwGpR<-VA+MLOjS0ZpZ{=%~w9Zyyx?W@7S)-n9($(q)He`7aN$5vWf^%J|7<==3rH@
z*tmTDfT<fVmlYcp-rRk~>EJQlux)2y$2-e$S~&O_hm)JEb=Gy|hx#LBG(LU&iSwrq
zSUPd`#5xIbc{y@CoOt{7SKJ*B^5AriQ98j`<`4)`Fx0Jy2(9MKPaodnTW7Sv-B)k;
z)gS$e#X52>{L!EODY+EBJ-x%UPItg)JYTQ;@Zm>(`0*!>@t_ZFad8;qnBq)rDoh*V
zk{Ajv$xUD?b^yII%>t?_rt@-r#z*~})JNQkynoqSaB@u#Z*jRw47U&JAO7XOyoEV>
z%r4%FaxmKgty)^K)Mmj#@D-o4=vQi0=ilz9*4S8MuqIGVC(IM`;Ycf;>-j>fRaZ80
z4@*~NWKX|qe5|7y&lG36wbWV|TT*vW#g4|_hZ_VJEuFmC5ofAs6i@#?@;Caw@7#V=
zSFk5#dL3V*LcslN-gG9$sOn$;+3ek>x*4<pxOLEo_)+S75fkpBKX>c2VRR2Y=<he;
zS=W0)3fsZxg;q8x)>J?ptlH(QK(@?&5UcdvbVDzNEmwSqf@POmLi9B^L1?0+GrblS
z#Tkt0)Rx%Ng)qzd)R;kQ;IOowsdb|!nF`PfdtzhLz@yqaIN>1}kZ0Du%kK?ZlfpoS
zzhm#1t3218_fq7kByeD8L)LXx`2c31NMI=$j1zu}(dqB5W!tkP_Gf~`aJQS~9Jsm9
zdhhgR$!X<sIb$iz0cJPE?Lk*}%bxLIbS)yfl#(|5p%a2*sEZ;vr!JH|7h#X7_$OEs
z*geYC3cSbHj%^)RL>LsDjOR*i1Ovw@aMTgg=$&RA*^1)7w9a|UvcuXkCOGD0Vwock
z4^l)FW2kn}Mj-^tFkU-zM;zj{#2SJTG0bTSJRBD8=2^sdDX}6+Wn0eVEpxtHIiIi8
zHZ*hFkV|Fs4P%S!W~_K?2eR+ZIg?5uw@$8t%@jPT)0~4Yep|f{F*}V)02a;|4pZPX
z2bL%*)-t4PFASNe7^|TohTJ+O71DKMy(ZR_NjVc{&$LJc^q^CFV=@Eh4c18{X$YPh
z<*ruwS!?I<^eAhb4%0kC>s&8Sq!@U5c~820rlytNE6xmhH)3rYqVjd#6Q&6lA{a~S
zh0^5VRBK^NiK2h+nk$E}=y=q#EDP)PNeU6~S;9;&m2JJUmMbskPo(Y2r`ot)&*bZc
zayjEVSnshxTq({umct_5KyCPWl7qNbRUi)uE$I>qTwCS&{KDV-?ceY)A2>Wd@-M&r
z4X5RR>y6Vivn-Olm`fqXz~nt`NY?v6(l;MY_kvfB3&a^agw^i5V=Wu44j7J>i&!CB
zW11q%a=-_VGZ6~-?r|fK*Mjx(K#h|D+gZ;A+o8M8u;50g87M<o9hXb`Ik)ORyVvd>
z^xhGpBTjpl4y`$gEs($v<N-{h)`7FiW*__5QzmAz`jK7y^yX{gw2)t}v{c#bg*{H(
z+8qgAsHx$4AkH%`I6^$)mJ`=nxU?o(g5E@X&=mu`x9Z5@>b;j0;SdLq#OtX0093>#
z2iDFc5LZ!W7&$<!*=2$9%YI>@V~5ok^Lu0G{k2ah9IqpFRNcN81|5YlX-L)>BGyAu
z>8%f42Y?R~$J2>!EZ!US!=w5GgAejKHEjw;+ojDn-J-;)tG#+SPao3Qq?+lf`RHY%
z-wyNrhu!s$5?gOD%HSAyoiL9ok;~Mq*6A-CT1}c6V<hXs3|aY(QS~_nRs))xH4qef
zw-;m1ycRtQQWA^0qgh=y*>NLU?l0>cJ_h-$0prKb8oCQXL5PpT3Uo@=L7wV{x;|w6
zw5}Ho7?Pl}uImq3K{P42;5M0+fGZ8`3xWA~C%Gm{vJDoZGfd79yiN(~KY_KdHWBtT
z6;06%zDaSgERhgBwZpke4z*lNVzIlX%r$4Mb4>Fjp*OWq`XCp{EH6fgZr)msQ>2VW
zE{Rg4TAG8W4^`YZ<S;kl+dc%#GD&6GN24>SeNakay<P=YvSqH<jcqN^(QaH+{l*jo
z%IQS^(5v3KR>nzI)*~W15%Gw(FfWl}kWO9?AxBfEf!sh-^ql6vw{IW#%~y}ihXXIA
zac#2pD7EVN@zoNPa^|{iWX)R7syQ&b?tpzSCAO_;lJ|d_A_K@-VtGv$;iMfhJ}b4d
z7~$x*mRZj)q{}m=Rbdx6Bay77Q!CH}-eG4ixe(ToYa^A)`MNQ(e6OFLJ~R0!`>Wn1
z5>|67a?Yep#%^=U^is$rQBCL5(|bOA{GMKtRMFce!l_!Nu&<qxGC%zA9p66v3G4NO
z9fFswrBF1vEBDS?3&9zNccjutrSWop;c|Xq!f<r&2*WU1r8vV&OZ@oL51gOQ^w~2n
zGtoJ)o`=Jc$I~6gTF#dX4|jLO<w*AvfBf~YIK;pY@4v?y$GeAjq;=&fKe29wQYw?1
z2za7f@NS~@N+wXV5DA-xuJkc*9ct~^+Hs4;1Vf=J8d|0dW&LGO27+^<at{6$7Gs?}
zq@0?z2jD80(rHb`vqtB#gP{ktK$)BwV29=n+=H=Lm*hb_cJRso<5h`0l*w<QHG=aT
zmIKWPvi|pm`8|cWP~-H!Y3Jwmq}8x@;UYx#&dR~prC>1%0TQ8T6>^^*Fw$-6fr^Z#
z8GB#)$`#nz0=L(`1~9MA0@x=Cuf#t_zrA1kGfWsG0?7(7fU_dTwH{+V);bk3s`H~}
z%`~twLChfK60#&2Mw48nZQ8|%LQ&Mee?T`~3<=I)y1QvCb|`?%RfI?>@HB(>mLh*M
zw##&K+`8<pY!$f<yq_>W3c^^2HPXrJl((y`Q{7#|$OOQJz%(7y)+sI%$z3q?;0e~U
z*_E_z^u1`ag2oTigpXo#Y)C_Et;EKA{TkL-x{-6jdC#a2ycSNiVx3%%B`4PFm1XiA
z7txp4VRyU6&5bs^BXkH=La@f@#25&1=6Hxw%nWf8L}yrn<bZhZaQ_sV0i!mB{b5#W
z6I`NoI3%U8W67yNE;_4_@Rc=JuB}Q{Wi5C|FYbl~?xlmR(rsFYkFxI%(ZMDcm0{q6
zBeWstNADz!5S>aGi4{(z$b+^GT4}T*fk<`B<f|yc?T}HQ@uDye!Q+t&BNyQZO!I{E
zQYp_-Y?Ft>%oGK~39|FEAy7u6_AHJBXQ_$?txa~ucgLAuef^fNzIubPj&xZu>jqVX
zhkb}*R4hUAAlp;d>@Y0BahQXo{CQEkIs?uNwmE7sAjOxVHLn-2?D!Q~!5D>nePNr9
zU=F0RNj$DGOw&YYp34f|6vhy-OfZfR1HpTFe^cRnO}v~hj8xdpS4u7%ykvd1A&{5z
z^~@Y2=j#iXb)!v@B_*EDpNWS9E_ced$(ZpfEP_@fCV5zimmRHBa^`p6{T&a#{Y#dI
z18?7bMTv%Q-B%c6c`2FqpFit0Aa=Z(3m-pyqO2Q_r#nep^A6)Isb*>!{P4q%tj{lk
z$u(Z8?;vX2wgm3)ADB;fJh;Hy(*xI(cwVnO-oN88&B7o!f9AMM6u(f{%GGS-R%p3H
z_e4A4ePGLiA>Gz=OPa%WN*m;C@Ub(>VD26_$*35|EoA4HP9XGNtg|k-oAph0n__bt
zLvG|uyQ$cBa1KUC&bl@nL#)fz2IkWpE=aOzZ-u%exs6y2bFEA!=)PXo612f#UI@#=
zTD7!R!KU>NXCk4C71Y43;FhVO#90d>w&QKo<4i<~wj=kA(F4DWh|3u9lUpk`I(hJR
z{k!)O^ebd&f9?P7%sxR-+oIXk>QY1*JCN8eqjutP>#}z%rd7UwZ~<>-oC_E$5%Si*
zB90kd_R_V`Ta}+nSEi+yn=8v;7`Ew8FVMJMy8vjh=NMKBf^H0>7Da%IOmYluypWZy
zwT9khjf2`v=@Qx|xgAxze}fM*;c!og2c6JLRI{vu6x<{wCbkPX&^k#&oTdYDnz0Bo
z#ap9!G6848ZRIoiAa6+=4&D)_hz${}liUkqXo$8d>e=R<6YFZzVhpj-8tZP?vgA(4
zTQ#~A6MNHXhKO8u+bTnf>Fx~P4{Yu9lCgDT3ZCEt(P=OrhG1cikvWRh(T;zDVnFPw
zLfo>}&}-FE@BkZd&Wjbyp!O&TR(8{=HJUN_Wzzb0Flr}XS9;1CCD|yYvTd0yMZwLM
zD*JmKwe8ho@Rn)vQYCeJ6S$<3UdL=GGa5@RmF?w9z9#&Tapg2mggJ6t4ulXSqBbY#
zYLN%9bCx+qmJsN!O8}3qLb{4coC{Ph90l6GJG9XH?E{TL#>O$u!q!GqqQOgEdE2cj
z&R~bVvAVsnOwJL#<MFsK&p~Nz;G!2czLN_{tT-~+kR6(g66L?|!8e6mUbr|%DMwN*
zl#+F_Fo_x-%e9~7Cf_~A){66?b%v?rO*{EkNM*y?!Q@}r=i8RJzI-D4z<Rw(HLr!_
znm1Y(5z#bFEb~HHFG8ZUBED=!D{5K3asKp)-~RQV6GP+l=8-@9H~%e<(~&p~Ys>uj
z{zuBX^78g89!>&&`RS)0xjcO)01u~oPKTq6FG~?za;uo$@!D;T4i?MgB_rYP{>1U_
zfnVLfVH$%NqG$#BptP>oSBo)$`4G9gFSKfi?u9jPl&p{>>ltluemc`q!}xI{pkW-f
zLCJ$QGIb1Mu|hYD5Q2=t>A!4ip>s;f(_edR?|$q_Q}HE+SDQz(?^!FQ@3zb#E1lyB
zKP}`^DC>o7J(EfXBQd>IH5&#QAv-(p!E<;2hWooms`FfPVzdflh~M2tTo+-ip=)>=
zx>w$lP-W8KjNZRpI%TXNrn?eMrOUT?Kd~I{BxI>6P{ueJVVhT`$o~2-$%DHa;Ra3`
zdk7IaJyANL#4KWOOkrRji6f(CtW{W35R9!05(n)y?p!I2!o)X>?uFVn8m+uq{i=$-
zg-9DP&Vh9_V<oNAfb*hJ*!TO!R<QPFrLLpP^m2%$TL4IMV79S;-vcy&uC-1rl{Ucm
z2{#>?mlMuOhKd<YChIEL?5!#oirb+aeH2RVENe_N8e>RQ`mkVKU|5IskuV@kVj~=5
z;9|u4D4z)iG)~?Oi9QZm4D4~ncK=+Q#oS`9bz<DAv}aviCC=Y?3i6pr11B)2{Th__
z!y3a#SJLx)PH(=!&QXfMA!{}74AEOD6r82H-(NNT*IqvlI@BeRZ|FA_yvL70wO~tQ
zT(g9Uw8kNdkne6;xYkOu(3BUxj|YchS4)u_oYAPca=vU#-cWmGONn0PFSWy9?O>TL
z$HT($5Xrgnauwx!u9@q$Qc95v%LalMNJj62ZA+xJNWxVNoDK_zd10Am3FK*&R2sR+
z{(qXH2nSQ<Wm_eYtXjw-DBMGTq6>8TEf;-H7PDTws~saFo8TEnc6PZ|u+qi0T&cYX
zbVW!1vd0nzgR>B#<8Ycdo(?3W7>QA!9L@{uV<%n>740y@>0#Bw>WpDDQMNwUPq@of
zZX)NHClPv1(*YME2FT~DFcIXeyj{et$pdMMk!A*Y@-QkEi7?%AC9jQb+L&f}aHO0u
zh!xLTq<d3ewNhZ^z4MNFnyAKdJRLEb(J=aeEs;L`MCg^5cR%sr;cI5^sFy1)6+WJy
z$mzo8`9N=tzyJPseERT#dmF`hkSlh;Jk4^m>LlUi{KCuiOsbhSj?x8K%VAmg`rTK&
znU7fDaDRvW7&*OpAQ;E_cUP{LE7nEkd16|QeD%&TO`eZWpCkh#7si0RCRSZhEr*B?
zqRDW^;Ddz83@oj5>gZHsa3e5EW6-Gi7Y-dVRs(KSRM$m>yQi~>uIsfE7yPPJ+r3Lh
zL#sk94DrbE%~yowKp%#_t&DA>Ras4WCrM9M$%b+e3ajJp?hWt$=-2ppA#Iu3t4!a_
zQ2)J^V!#5U<8E=pUoZe#)0LhP9Ietu%@FaM>~DvTpCgCE9rL{0)|$qO-E@Cec<+S@
zX#eYzft_(c-&ZDkCt|el(0^wy4kQgx2ADDz1U;aP0po6aZttxK;%-F<`n9ZZGj!E3
zh88%(XjHN|y=<JN5Ky<*rWW)z1};n(wd%Gijh)f8j|BA$(a4!1X_odiM!qY(QFbDl
ztZ1Z|bn_8EFRz$q)q0pN(}g{CQepyRf^9>@2GKoPA27X1%yLVTgU~cB(aZBeToJRb
zh8up*u^7-tWt0r9V#mN6p*#lXZUy78UIeBxB#_a*arEwKJtE}Nam%=QO|IS71G{fU
z^lLk(0t3Bd*5@BNe|#VYi$5-MZ5g8!TvPxl*Zv>!`gxpzDhkK3S10<*VcBX2Q^j^t
z8eiWpJRTQfoH&=pr8P<)c&DCOZ$+Tyyo`?kGxozO@#(V&<XWu+uc>d=gH}HC<hVb~
z)M-$fc-fuN8`!A*!dvDMvB8t9r<Be$Wm2syBfw~K1J{ammU)_)f+s~!E}0NLt#{6s
zjV(2N7?2I74O*6@MCTkWj2ouc8Br7Oer7q?B<at2sh~w<ra~H|V_^((sd!)YLR8i7
zfHDRrC%_=@gm7i;kf_w?MCrWSWK@i3jmJ2zX<;&&0?0XYe!1Xm#K(zY49nqwb&h4Z
zV~Mly){h6Cquk&^gOtzCbjhTMQv@>zhsYsVo<Dq&iAt@)($AIF?o}I7X|<7!I6v;1
zOplB=>h#!NN5XpP*Rdy3BD-8Gn9@-y>3ZSEAO4=>-CL$OQLc%yWj_A&J)^7~4=3n@
zpT7T|zyJORe*N{Yu!o6*g%ARFhl#t>%-?+VmOSM1swwfZZM>_EQmyPJbK&db9dnG-
zl*#MLVm;>%pP8q~$eE{)pJ;vX&9~q3@a7#ug@}DLo}Zs-b%^+NShggwqZR?~lqT3^
zH)Q9$?=^dYwhhXt>T3`5{D)Qd|6E_g8YT0ops!2X=sp@B+M$in=%Xr-3#u|eYS&=1
zIP>t$w@in7!OX4~NwCZtrmH4k3YcE8U2J@`LhzR5aNz#_4Z%gmkOzFJMI&B?Bx$W7
zSY3COq~m$**hZ(+0(9X~7@15Of76*uh>#XG)_IoYz&tOYWYVreJMYAqBT1f8(91|z
zF}3^qHpbrPvjce;iUc3>3>k=iYIiBfs|x{i?3x5wTN(p~9UnYIaqcYae}B_%&x#;-
z^nMdmc6C-^1hUi2z)urK6Cq12q>?#rD{D#AT8MSSyC4v<ogQfo&U$&S3~?OL#OYBj
zI0VW@7mbAhu!i9rOdjVWJ}lDg6#1GpCq^qW{TT9DItz6Oc@QhmBp;%T>wANiRh(Ev
zFy0z{eTLW}Eb<;XCjot38bs5DstOfTg$JTsZJhN4g*UCviz8*|l)!rd>FG4cJ4Xxw
zjpU(qjk$Ka8$~rB?aMMxK3BS|#7eTH^%MDWWDajdlwds;r(^ss1@`#E8i}8|CeZQJ
zkX`$(Xlu$6AFYY6x^;=ioF>_y4@jkRy{;IWnU)zp1xe$}iQZ();<P&Lt+Uom_C*7N
z^%7iUL|kXBBsR%SD7G3I@9s%j!Y44+h%Z%lZj3>zos<g$f!Cy3i3X<NCAYi>(iAzb
zR}t~8iFK2)_G~9rbQ-K5!v9x-*U*dhR+VaDBIGC;k!n1Gc%+@fX}Xmw723ExPz2{$
zxNezC7H_I|?v-SQpnipC?#2Fta1RV!s&$O=vaPH?t{XiB*6T(}8{Uhy!G#INc%9A|
zgZF_L12H-o4{89E^Ill``cPTs$mAeWNzabY&lh=cH*^Y6S!*M$8)@6<qmz}>P^-`+
zQ(7q{-Len{aM`YuQhB~SOHt-TxZ|v5RpiR~!%zG${}s^;7>>`Mf8y!G_jof1qcdu!
zonNSzCw}|%nWv9G^2h&+{}sgymPvw`{`Bi_aqsT=s=$Q9^ud}E&cb<3{O+e8xn9mp
z_L1Bw+qMc?dEIC!+_VxjI34c!=HVSFUD+;An4yK&`Ai5-#)3+eYo*ca9y%hOrBju<
zI8%%YuVh?qTZQ_|E(W?jn1&e=Sw7VAv>$+7z20Lel%<LNZAcVr@8G7H`**+MaQ6mq
z<ZY!~FXBEanSoI07&`9lLk<WRIUJ50j`w&!QA;D0!kQ8(C7L3H_p{dO?$rlzmly#^
zkhYD{Ds>1T$s!XZhOFseM;XgxVqOkRaRS7q+q+E8QWI>pT@LgRF_V??v(b*fYj(ly
zrJuc?=DTfibP2#3T@>rLu3p>6eJb*Llt}21Fk9S>j3+YyIW%Rp*yWHK+&cP_Go|dQ
zeaN%Gd3=aBF>_Of%QbBhmeUKpC3zM|Ym9e}IZVVD1wCB{U3vQSPO-9X9adH>cF$`t
z(m*=(&bKaW&rZknE)#*FRzRds8edLTG|94$foT;+#WY<N>icSqp_fc=&!C}eqF>dC
z2t)J>wP8jjyk0PvTSB2zHjSt`%$J|v7oUU>C&g7aR8&}Ng{UZU>*ZR|^LB?X?OK$f
zEAyf6gQXFCAOwk0Ho77j-P~Ls|Jdti6$7_RQx%T8|Fw1GgB$``2@J8#waQvDPv<M2
zo-d?J5`kQ)_(?`Cqso{?{J_!?PjTWf2ZkA>RArAIykH<(p|y&S7Qe{|dze9Vfn%IF
zh6Rv)*Ww(}TZ%?@#t`;r&uxS<A0qRjBXZ{%t#N*N;gZB|w&hBymAliz*N^uCkSU#V
zO*GeTyFMFa)Z@K~?VL4KMa+&N9WK@gl2(KUIHV#j3Keorv<(=;mMZ6KB9*GBRDrbk
z81MzrJ!lo^jF<J*kOSUF8FzXsBdwY%>vbdB!u7i0edKtUndVtg&0UJJm-7>QMo8~P
zY&AnfTO!sJ5Qtm@lAIMMN3RYWBtfy{3Qns2+T`L*rBJqwK__L25w*ARZ<cPb7o1KC
zQZBr_oVl(Sg0~#!fS-a2A?3kRuFrh_@wbe+Vtn9yexjroT$m{7itPqRXR?T}bA92%
z`|r6nSt0GQxiJKe!E=}*-auM2c`JN;e&#R!`mc!HFoi(L8_@aq;XR>tE*7rW7fH69
z0&|!#2&CegD{Ibz%We&C42OAPUM8NNo&|ZWj0ynXgthJ@IY<25)zXrv=|>;*U-+6}
z7yZeISZ%m^^XwuimBLGUoT7k-c(?HJKwOT5`5;=Al4*G(uUGPVm4n;~<xheIHCfO{
zj*oAc4<{VLqR;8dHA@G3i@`N=NJj7R))1{__A^tQb$yxW#Zo(%-sz<X^)5!}P5=D9
z^Mw#+i83~pQJYZhTDj%f*H+mKEYrjolkTBKD7}yL_cv~IO*2Igx-WV~IV1p@9wHi{
zvxoo47kS-3v!jlM2h&>-1g`G!e{Yhoi@WvLP-VZ_e^wiAM88}zDJ8a?WumJhxz_Uy
z>&&p2*<)OwvnS2U%Ba>V;^fhxH)7Prr+4~(kn7WDfUIXmmYshWAdi;;hiFYOI#n?^
zA0z@BM1R(@&h14DVeKpI$y-Zlh1SkcCx%t+6-JpW`e(+t*-rPfQ1V}bDEQSe(z~1y
zwGU-3^xLnS?R19|^&z<{!Ru$P37cAsSc5jr=9Zlyz$D3tnCIyW5#!E;koEf?^7>iD
zK-Vfr-eASjI!7<Uz;9iWuq2mS#v-X?wo=(rp{6Y2mez&bWDRZ4tR9`(27QQ<b6TRn
zFG8j1_^Y-;YT~E#sbc(w^PbiRVRGWrt}@2*)-nYRL+WV2iL4T$GtAL(f1G)^Tkt_D
z&8KwbvR?W0eCD!A!dfa~`+Gbcd3-zxLic4Q3_~?pIoy)1(0Z43hGwz%(aAZ<XbJJ+
zuDYgZx(DigkZNNqGI<EvU99WIx=9T2q_j301=|^o#RgcWg`kB%E(I8(Hg)c{8*D=~
z53`Rn3axhLWoDix=4B#RL1UJ@a=BQ>D1_iee4|vYK3*TtW(awUq4-T}sTkH=xNM1<
zJM%H%f~V=vd`%mrRN@?E0;8PvQZglPG8q_R_i9Eyt8KjqSi?f}78eX+EjJ9?wL&hW
z%LnQhT-HsH(_?V$LV;Y%M(YCY5LIoXp^9`n`0EefasBQG`fq-N^Nu~#BW;O%xw4*L
zc-c0NbEK{-*Ow>0|NGx!O5t#r@gBl75f;&mY;|L6m3P1Tmd@a(k53o_Q;du*D~j41
z=j+N<ry<^ppz3CO8nSk?-O#gos!NAje$K7>&+uhlA9yN?+CB4QuMGNlogj4Oxf_e~
zvP<@1mK4ZxCS6w2wo1gT=_1mU8Mp+>G|3ZKPLJGw{jJ!7x0ReOqG2gA$z*`(P>s+;
zZLJ(^Bu-Hr2XSHaOs^Sl>}$=Uhbi6Bzi>Y|D}p+&S|u4{3u4-|sw4KGYllJ3S4x>B
zGsM~7V-wu^d7YXF7p2{brd=Ndx|YyVfnCp_yJB<u1mtV<Arq6{8@U#$+t``;qUms8
z480awDKgD!Rm8)h#>1F>Eh)`RN=ZW1x`2K9=r=Aw?^VX#M%F!LN3b_V^=A2A`ha}f
zoL3=6n#RtPz<8wJ4(bdUw~pLuA?SCf&XC_}bv<C5r<AI?1>yMbVm!A)pVU2IbhO^e
zMz<2;G=>qOqme>sU)S8OZY?IA_e`3!D8XAYy^vx+&aiz7ut#D`(rlIjYM@reoi!j|
zd-T`og|HF4TpP7&u^^cU!3Tm79m$R=e#IvL!>^xJ47eC>0HI}>n4*IrR*2LE!3c(f
z_Z*ZvFj{AASqNVWr7*hF=tW3S`-rgC!F4UX%$emba<Bov1PPfKVs~r3<3REd+CI7w
z8)2>uwv`S<?|8U7vZFv->!fYFi9L)FaZD+h%3y1Q_fMbs@bQ`RRyan_ahB98A1%{q
zVmd{z5uYo53{uOYA<#%qW5Gv@&~l8Z!+XIE1p$rrdczFy&Xy(#iM4h{7&4mcP;wEc
z1OT0((Z*Oqw1LT5CKn|Cy$)KgG-Uj0oV$&wi;@>xmCz`M!ctVCYR#b8p)sC;oD8Fh
zn!Z&jSn81Qm^M&4mME`{vDBt1Wv#MRQ3uaj5QZhtQxYumkeh9pg=HY!Q6ZJYbz5;x
zcm`(dgs$CPF~Er}vRmn#tZ_1lu>u&$xoWpq1Up<eYU#YZT*xUSa5l>v9IhFDeE*&?
zEPwvnzvBAicU&JISl25r&(Cu6H6ceH!_0L}Y?rfG98;#IjA_ElpJU+e@eS|3{bTO#
z@7Z!<V7OZjxUat9{ntNWMwg6)TB$iR+F)%3Tcv2sw&Js)_)c_QIsDQIv_8mPpjW#u
z{TKT(5<SjnD)5sIR$^{-2&)~b8AkWwvQCSW7usO!j4hFKBBf1Mh77Tyifm8z!$XQz
z<1ME*?+D9L)|lf=-&P6QX<b`LCy(AHQ@qkEt-yRbad(tN%(SXtPz7`B8a#dAnjvnH
zJ`(tVwX-VoE4gg4Ql_b{N4i&QWSl%0<h)7RtJf66E-2%EyHB}w(O=Gdhejap5zE7J
zD-=L7Vs1Fw{eT=pZPmia@3pd9Au-|(utvs%`}kT${yOy5P@P>FKf10FYPm80N8e+e
zg$K}@imQ<bc$*W!HbyOMo4|<d=r}BuLAHRA$$=Cng8Y?QPYTEo16CU&Zv;{8wZk_c
zU_2v}8|TB#y4@AxW6W+{-B$_wbt+D?ece1-6FtL!IRzL4EU!hIvEY$)W^iAO>UR(l
zw@H^+L{&?mO;#<HlB$p#ZNLpl1IFkI$0E;m?}#HZsG_;s61^ykVeI&LnYIjVqFejJ
zo*h4{7!aYFAcu{UrI`?^(2HNOo7OVV5g$Bh$RWM&qpcF3oO9%s)f0M?MvzP4^YaTf
z7?#o*PSX9NlU;QkQW?b{qi|8p%fa__y<m%_j>i3Q;Tu!&aUymrYm8DURRU$Uns{1Q
z-1Ulyf#-E&ONE9Z1k1bx;^d{{cHpALcnhX89|CzaLJss=ZXdeQ-IRO)It9pu!z9@X
zgLWGcwkn}ZrHG3lP7d#_?1EHv>o7M#qKt+#nsg|fs5Z@85GFzCP4LY|ZpNl*U3Lt-
z9e5+}Lo0=xGtPTz6Oz|Fd)<d|%lBT_3kE?!i={{sz<iW+urDDpAS6c#50T`wT?4YK
z#e3hYU13IyQpu^3a>LSb-QAS5PVdfK8(S>|zjrNqBdZ8?7+F1RsnRP9s@S_)V@p?R
zEw~ujYT?u8k9_+0J?G=X`g|q7{{Xp>&M!PY{mA|4NKxU<)8|hZYq`IF<aoT}<ENi^
z|M@*RuLSF1G<tK)%N>t*Z~5l!H&i;8%LS7%Ic+jBG=tzg=W>x^VeIO8srp8-^sNhz
zBMwzROK04=gmabBZtIy}?#tY|>JB$WIzu!BhR$F7_22T{5AP)qEoIhR*-EG4XgE@?
z)RJU6&@20-)>`>m(pByI)P|oIUM>UcClOq&XX^DzX@xo<8$&UoX{tIp4-DGjK>K*+
z{rnTX4|3W_+m(`%Ucc&4u&?*L(ZkC*&ou4SRdL_!YkaX*YL0?h2EBK9C&_(QpO0q1
z?y`{4@GA`H_utW1=fUl|+7CGuYibRE&i}PzmiMXe|LKN5wyHO<w++MI`Inp}Bc+Nx
z5VHpt38gdDdTR=Dm$ib#>h9rk`g3?E83wwJ=(VW!L8G9p6<FOD7SvrnIfe)ioeP8z
zWi?==xRC&;aU%d4W57AAPGJOHTxlv~w2D;T-{ZZ|82VnyibHrhnp^q$oaa@aF?0f<
zT24x>tk)~&^M%VQ_TtvR0F9|8rgyQ?*QT~|xAOFK;oJ5N_NzN)Cykr@o#v*eu}Vdh
z=W!+Fd@GEHuB0XXRIbyGta1L3*UwuMbb(!r0XbKU@i2^zc%&$h1U-ZEhHhZZjZ4b6
zB0)rJPSeCJty%hLxA7PZ(l+?`;hA}>_$A<@#m(XN_3uuR$A=@{JI*f`YA)bACRtpw
zOiN&zg;dyk#UPQFL){S59Px8t8-q{hmDFB{!E+3e<KrZPMl02oABL?ol7^)$I_mDd
z(Ne)|1?w#~dO>4a5dfB)*|tJ1mE#OkP!FxMl(8pTiEz>fhw~B@dLw=*TfvSlkP!9U
zms03McI)0pY!UZCYl;{hO~rX)UmOD=RRy7sH3lgqLJ<8y??cuBL(=n%>9XtAl*n4D
zbnJ+NI4+Zji@dCg_N;;_Mp_@jh3Ng3uC?#PYOUmw*|sYwU0Js?Id8Ynk~Ruk8k!u|
z!N@eCXpy6!P^|>Cvl$J#0Gm=Scqal!?>%`-l-5X9MzeNX>6cIZr@#1rL)&QAjlcc#
z|3JDX(&fyjr|-GHJCaMLmCUE-&s=Na+c&@B?YnnewhLiiz!=iHk!O!TE=+IU@$TE-
z@Nm56x?TWCeSovPY#ZOcd1PK@czU9g`U={ydkOF8Rk#ACK~)i@QL<YuGWoE7pgriH
z>q|!Kpo%MTsc6w-m8kcB{6GKS`9J+%{vZFtexv{6>wkRx5BRd)^7hRUoJ`9kIIl@G
z_pNZ*r#6i`HrDG(-XwXp6{|%`zZEZkD6s736$4fGKVwjed^65kj4DYTny_|Qn3tJh
zEa!FQ>E%MvL(4iRO8V#ppQw8%>F~Rb^(7PFdR~dWapw;VO{ntTa#$uF?@s_+FDtEe
z0RzRre3*%|<NkC|N|<z{txA@69}uU&ayrnhVO<MGGoX)C#2Sg4t*uM+V;z{<X-P$X
zIv)3v<3cC5M#N*fBqz28y;asV@%(bew80W1cZ*~ubZbP1Af+C+WI!ubaMi5rrV|(E
zEMbz&fj%0wO5#=T6&nhbCMaX;@X8bL${oOh2^JRuHh2kdF){)VIkSu-H}9}^6i(IH
zHDv)ct%<xQ?%&;UdUN7%T!fKe(Dy>B<Y|h;=(!e+=-fp_RVvon6m41y=a**<L$P>o
znWCk44Zt!(92j2Sj}EFw>Dq<uV?Bp6R*=0-z(~^^INmKp?|DHE<W46}j%f<?>&D;x
z<-aFi&s?6L`H%no|Cco-Y%q*^rL`+*Gi=Xie){yD)xw|s%Rl4c?!@%3{|z5Mz2`sw
z`~Sf8ywY0X;p?yY(|`SE{N^{m!DCtTN^OPndgZdNq}nh(68*&8@ubOn8Vuu%V4Hi#
zR5%Qj@;1bFt#B`+E3N;sEBvLsI&zIOWYV{51kiD-y>)=_fB4`3Z+N&nk!zJ-yQ`oz
zDTZ~D*mjM9*{3NQM5CQ6nq*j;z_CV^_;nBmT{VKet@<|~=85~|z`;k*-8DldX{{F7
z|K3Df+#-(IyJ}-^UFLpYS)J%P4Ji_qLAVj;oPxp(+H5InK_~M1<t^sM5|Eeq;va6%
zp<PR$yZ>DP&OkB02k`nDx5>;tArPUVbe=m5==EjTgLP_cv{n^1q+|FYnwA}_-3NN_
ztL_8od`11=*aeUKVRjqoBQ3TFly}{Q;;L0#zOxmqwS*9f-ZS}lvy%&)NCTsEGIIOT
z2RE9lG_E^0V%*-J{g5^K^>)!?8}hU4sf6;m$QkhaiVSr6DIAyUwsKupvXV#J5HQt#
zZHNX#&Y4siPtO;gpU?UXcG!I~(A5E<t8`;UI|YJ$Z@rLfRx+pFSNqwre{cU3oPnA%
z#(O0y3DAbVi-Ev0IK;r4(~0BZi1l8^^2pdH*GezK;wO4ch)^0Pqrx<WJ{YUl=xcZB
zy@xi`lN*?)NeTq(amI0&B9C`Rj*oX32Ej#HPt-!WB(Cdrv$KUM;Da0xc+Va?TYD!4
zt70-C1)CuOK$CThDnJh(JuU`vm()eygG63hfp^q8xLyjEOXgrLbBG+~nK?$X?ns+*
z)?05PBy(2K!q&2yO&=!mkPN{Iw8%M9&a7JoJBSQI^cW3fan2J%koC#vgy5OyDDTG<
z8A{_SRmMxZqSXeO*0|u9mKmBMr%Wz|^Ytpl*8P#U?;bfGkF2?{ZYw*t-RKG-1WQBw
z#yKVEhP+BfLWrI?seijyw)IR15oZR+W#XFT0v^_=EmS^#8ytQL%*#ZaJo8~vFSOuG
zyI$<m6qu*Tyvzh|xn2z*??DJo1T|)`UM~FYfBGvv{`8Uar)Ns<gvl{^Fh(>Mm+L2f
z`ssJ1S!j{oITmktyni6hGw1P%={WK5_1FCR?H{q=8Fuh;e&%w1W}YTa%Y0jH`DtRF
z7mkM`A3nTbOT{#K_*&;6*qdF*fPkRP9%!R01M^?ID}NcUp$C*auN1Z2x;})xTHECJ
z{_p;`{~iD0*T3SjCeB-8t(~kQ+1&!$dMB5rpeVIp^(GX}FiCWHt#sq4wbN_D^i0%!
zZN;$~%N58NrV#o1;hs0g19vXMXqdE#&PngpS_`F%0M!`?YW>6-T@Tt@n9Lq-W315t
zwuN~L%rSD9j)WM9(~OUkLW}xs=T1kQ5Ki~Y!(bH0u8V&WjSU#~b$w`5b0G(!1-DZR
zcbv5xtbHGCD?->9CkNT+d#b=qdQhvxsITkFwr!MBL|lk4X37x8OGgp}M^8L>i^-MN
z3O}uhXDfov)+?=7p{MG8-&w0G1F237!f1(eyoq}ERmbR^=sht`EOFvAADDf>TPfOT
zq9?P~Y9qKS`n#-UtE@4s^Tgq(H7Te^p0nOsKm_W-RH!<o6&tWv!;OTe*#K7lj+-u{
zOK98ma^>Uak9>G~VqJ?gaHETOevI4KuC=gknU_lf`xDR4XGzoyUe2Ljq|nt+J!rRv
z0_6FTTcOmf1(>X0LCKXyo`HLY#Xt7?*=qs^uY0O4jA?YQJ$uK);lSxIGsTHaryDDs
z_%cy0GA^bRMscqSb#{o6R6czmw85_b9jQQf0Zhlxw8Jo1rZ90@4xA20<`Cs(ubnm;
zEoatG7d}0&q+D-`-zmxgJ*+Uyeek!6*K1e0Z5yQ*CNFl8u;-y00mLMw7RsqDhoV)g
zUd@fXbq@0vJ3;Hj|K7WdN%!iy4W`KvoX1Dc_3CdOgfWK0JP~8SSugw4TA1fR)L#tn
zYB%x0W3BMR1^4UmPJZ5QL}BfQlO3AIeFo=PmW9*(Jt0igoQad%jBCz#?>HP64$FaQ
zn$(WirQ0&xY?Cp1qB71~ZB$ceT~;o~W906BCPc5u*2)-4PNZRMyJ1|#e?DaW5-f|G
zcz8VV_;^nYPIi%9U@y@*nWS_n2z(F-Q<%g`R9Y3`s9r;cK-o6hCLU*t1f#exur`y@
zmGkzQ_doqWa)FP}Kk~<4e?zSs#tu^2h~98_TyPd;v3IWPl@CwvskQKUcqD2hYYmRW
zSKm;ar^Nue1`=SEEI)deYsfoK=~*5uMy3Ry9FIQ`9P-cgWv$U-RJZ}HX^{g!XV0>i
z*WRIt$j~wr-7A4IodLDu2Et_MO&vuC4)<^5z`H)tww25FLQVx|VG5S+qa3784oo}5
zPBz7;zVFI&N|>5ymlsAVIOpYhGKQu;Ajvg-ov0a|BGApPf!QgF`@h4!|96_z5N?yW
zJzQrPi*dr<zcF33i{7`|qiG@|Ln4Ixj>g`5XhWxAhhwPrMxL8PzlQeXM(uuP=uX_+
zGFxumcAeUGb$9HdM6(l5hq)Cg)>xu}hxEd~`+xmA{_6W5h=&tze*J6On+Hr3%ba|s
zqJFo#&SC7+0;7kl>WlQOF+(OXlXEu<p)(!tRbMa$c+DjkJEuXWPo@uvF&@%%e)#bN
zKmPUyrr`Kj|IMGUoMxI9L5AJ!B4SFf1xxqxjHw+3Jn5vy8pe?4xUtH%F<qVwT4*@$
zZ`w9zz}iSDNwjx+A)-{?eG+Kx{v6*Y5HevfN}hB|^K7M2woRZ;t?Jjax1YmXM{$3c
z;^^lE73m>h`i>P9(jO}PI4v`erz3Hi7zVZ?Vv-m=$7SNOWm>M3Ez|X)8o5A`j(yMe
zmTO>72=f?g$)(UVw?5axOWNqQGpr*-fyj8ZJsRgJHB+`sSu3AEzwqh#N-CA;EXy3Z
zI~`aK2VC^zR_Tb(+s_ksr#lI)iIJD<MX;JtM(v;<z*;bMsL(|U8G8d~>Frn>jHp96
z%Wg$6RzjAHg}zF67<^!gUeKh@+@=QsLyS&Bh@7Z#jd%DEB%CNX0dWk4Ht8qDkkyFy
zmg6!pPcub>Ge)n1HPz1au*@ve#H8cPx+{y7i5{KTktw$ai?M=x?L(Bp@^=dzQaY(Y
zDVdbAc1MBb5Q$SDP5~Dk-HHYxw?b9s{Lrx_22n>l5zI_WWS&J6P)E5XHM(IaPB;kj
zap7=S7^4w_Cr*L$c@@!RZJe(OKXsxDN)eOCVsAUDbzn@Tk-1)8IG4)b{QZC6oBIcP
zZ@mBTJ$XBGm;(3C)2=W4@cbR!e#OV9Pkdg_{Q6hF;a~jfkNMl*88{yK=3o2?|LQk?
z#`BwR_`m+!e~bO=zkPK>=zSAoV45P{d-~0WDc8YJMRxy%uL<_=zU$%A^cpchMFLp%
zv1ZqKtWylGw}auht&EIfaw*c`N>+-KARrYu4-BOgYAy6$<QoW+0)Z^}fX6Xo=vE4G
zu^U$GCS*EW6Q*`(71uWDRIOuRZenU<JG-H8hrE1=6m}}U-NkW&0CqRL@ctf*z1s)6
z+U2A>-@Cw9%3*7wg>s;fB`r2yZykBhb)vDa2ZnjQzjkEt&;kbJd+Y{w@9++p<BMx=
zZ_aj$;{N^Z3VZ+bndcw>hR+{A@>hTV6T!Ua{=4sa`!9dPSAY5^ggMd}c=O6+7~}P`
z7ir1RR8L)#VerPtRKPo?7&t7GJhOtA_ZXmyAh9<+s0R!?RR3kDsZ!UK-~GjZ=HLC*
zZ+Sc{1aJA*|FeIE@d0{~6_C+`T}SR-rzyr7aTx5M_}EuEMpg@m<|;<3O_;;lrl2SJ
zer{sL*M+`_!}s-7tF+cICSc7DNfL_e=&$d!V3Q55IdLhajA`vgGc{%}F!vSQKmPi8
z1VSBup+4@N7}rO`InOapEC)%4D@}@lwKYr(L_MqqgHgmYZLPsH9cws?5B3WzZ5Q;g
z?RW<HDhJEyI15#;iS@1RsefI@)3ygjktkEwBpx|!h0B`Ra>Y&#KRLqWWfT=1et1eN
z#ChU$I`R1MfHl%39Pdv!<9YM;EywwQAD&t>Jr_nF@}WmZLpt-|Jfk*pl|6awm6STU
zbhg^5#$rPdPrWl3BNuXO4R2@NgAYpIDSeJ%A<PjQJ*HJj*t3Q?%J@+pD6c5g9hcl{
z2mPKV@3>e=Zj(HdT}dksHXmeNQfs+Y$d<tpBxNn-%$imjgG*Xzz1=1#)*0}I?!4NL
zoajEfT#d5kaLk9m;cjML7K{_NgXji=0EmqI`%o|XpkthL9Njdkseq#yiP9Fwfn_)&
zSrd0hPRE7bE1y4p7N&$1=+bcRCd9N#gG9q;a>{HuQF^^;7?#EJoL1IPKk@15Z)mOa
zcYpiml$;35k>MStZv60ff5rOgBbVzl$2sur*T3d&Ir7!pZ+Lio!<)N%d>{A$hh>&(
zUD^Yz7j&ufhB!rzOXQjp*HuQ}f*({Ej`{zK5BAUIB}!*6J?Z%Jppinlj#plu9Iv|3
z)x*tM$7HQ&2-?0Q79@B_)^oi#Qr_sjvz1K(bCgxThae3jt40%|xN?jXzhzWf>OzD|
zD`8Z^=s5Mu*CB_UwL|v&-V6H!gXu>1)>b$e(n5=v+Bv2ei7}{)z-i;-_2g5;yD~iF
z(bsSEygfd7Kk&slq+WR%rhCCyJBRJ!c^v8_=-RRGe@@?gX-=FD-xY`^+XN}_2UfKO
zuYVq-<E;IQUG&iP&*(BXZ&%{_#QEa~qO*K`e;~$%oD$#t<=>EA62JLx{|(_V>$F2Z
z|M5Ca0p0Hpp$S^2YXcuxykicIB?b=BGY3ZqQFLg!Ch8VOm4DWeB@9Rdo<4tO`|y$X
z?>|y%<Nf8zpa13G@asSQmgBvU6!!wa44D#x!FsDZf51vW6#aE8KWI0-h7Sg71525h
zm&i4VV-B<lH0Cz75X7?<RJzwMrLd*M)ms+7uN5R94%MN6yeEsF7`v8bOW3w+D!jTX
z<QeMoKfz>RuMkJ?GN;kX)fj`fhGh!ObHo~f-CS~_m_aj*I*RPUhVIKL46SOt^Tgmd
zEK#j<P-+u9-l()<D5;Ti!7PDjh6ISTp}tmBL?nlqafV<$(R+$9lq>~7u7i1T*yssU
zBuo=!R0+q4l47_l3)3{=o#(JDynB3P@)PeK9&y3TgP~;|!^=*42QlptqkRb0cFUP9
z1wNfOK3@`H4w$%i2eK#iPP^+bQmpC?+%e?9SZ3lp$yhOGj=`}^6Nh=hTd&qEWF%(A
zx>T$a6@M#~5*g7e6k{Lj=`Kq4K4PIN?e>6nZjVI{w+E5X+AdtzvwE$i6En_eJg~X(
z){RjbTpUb`Crp;<=vj`D<q)w*)myv6X9GqkX5Kk^>v->OH_UK|+n{W?u5chMG7uRI
z(^!rZ5BEo=X`+-YJn;bz=(XUiA;idXTIj7YO<vX!ARg(_@ZJ)AV4ecY{mh|m9408+
z2i7go%NaXfg+<O9>Uv?T5+)RuJ1qv~!S2W4`TRsClhTFjr%!m}2;P$o6dJ)fLU1f|
z5bv~;e2K1yLu(D=!24geXkfGoO$eV~(${2&CD&iq!(eYXrC7BXn(pGESqrV+EPKXU
ztbV<eHrBLB15na!d~U{2JF}+l=~{B~0em2hK}%WG**Rls!O^Mep6Xr1($+Ps3oPDa
zy!|~JpopJ6A%xq{=B(AvSMmq;RJvX8Czdav3cgbMM8aq74Q8P4$xwW6fzfWA=s=)6
zqX~<|px~nY&+qU5KH1YkWwfTc9HW0{*Y8WU4tw{iP7iAhw;(ij0_{$Z6i>gUaheQI
z47OT$`{seihX<~gjlciyM}GX>cN`z@`1;TOm=O)5+naUn^&6F+zttv_jv=DxDR|kX
zM?t0g-O1$TGZoaba3BUw^=#;~4a)U}4<DcSo4@*7p4XM*apAnKT(*^bxpKS{LHaHt
zG>F(;{>9#=G`m<HonGytJsY?^*GCP|4VP=?a@oZ0U8NxC3`))lWqMu3>|K9OiOcyS
zESyR5bEx%JEW%(9xSB$sRCzAvoTX@XLPT}y65Y?lKjiiECIhu$gMGc2yM#+%40CXT
zja6YtZH;Tm;0(PD3gWHoJ4j-T*YRvHEYWkEA|ZH6>ug!zJiS#)k{kfj8n(*^f`%C7
zw8@7(Ji|=#fp!@-wXFWwtfLnOQ}lfEt9Sg-AAJoza@np71IN1~^I?%-k!c1j&IOif
z<}@$dAMbD>kV~SZtVQ)u>tmn~t5qLtDYI=!K_;U1JYO!{-AxjqX+2Av$z8_T!-zOj
zJiUchDhvqTb39C(?vKpNEEj(W+|M(IdC@pd*=5#RX`{*mC<IC=7-L9VqSgtFbfwNp
zg4r07CWw<8C)Y~QE<L({GjMl*$Gpq}*qJ4lL4_LvL~0EeJl;FvEQO#+O}ZziTm~<b
z6Rh0a#!AEMoY8D~OYMfx!%d~FZeRJQ(dF4UWSlP}LwB1>lv=RC5@*lhxX905tJ)1K
zDR+(`g0=YBb2v>nV<_vD&mTS4bHb_}ao^>NxGOMCfn^FTG0;jT29Lpzl0s{G#q>hO
zvgM6bH=ego#QQgVeEy#2&+qy8`6rI9asK?7%kxKl;D}u)<th`NU}&c6m{*9I>sqiF
zN*M}G`ehdbG`W|h5NJ9k-P1i~^nRo1eKC4AM)WWkLpM$+pT^Lv*kPUO2)0t_y;6E1
zDf|guJ@>-W--D7wP_`@wjF0pxg#{|&Gp$jz*t>bnhX^}efY>6vHq5jK7#XmFKo8F8
z$X5=|pi@BaBl9#76xH3k=$Fjkivyv{HLk_EOfs}MQn*jANjh<Xn$uNe*p1qi?dyBE
zN8YG^;8n}8e}C)~fm=<09wD;pc7{ImWc+fpjT9?(h{t~zIn>>#oZ`Sn&(<3*PPA58
z=9$Mg4}{Cc$G`iY;2rP&^jrGu^>ao?==QZB(y@?`WCPJl9PeHP>?>su6@<9Z_V7Fj
zVCuEgux!_rm!E#%yTAQ`zx&;L9v|+xf4HZ-?`*ZOK0oo*x8Kke;cRpbGInkZG8Nj3
zPjP&Rm|P|VlI7r{G(z&<EXU)@{BmYZx=QL2KGu*m8CJPF*4#KEwG`ItmC1RimEd(L
zrdkWW&}&G(l%e#>wr#>xah4b-yfZiLM5*-;x<!B9WMCg<8L4NLrj5Zcd(RR*!OMgC
zS~Ay?d096~Eo@uXZvS;xf-zdnSVbF)wJdl`u2Ouoj#Rw;b!2_rgnwO1XH7|Vt)pvK
zEa5t4blF+28{3vKqqBI!pZw8R{4f9HPkHz5E!SH3(1rB192O3zlX&-?mvEhwSX1IO
zE%*=!ni{yR+YQ7sFSCgAP^+6BSiKE`H9Q<APP4<?E|H=p(!1_)sxrVwCzm3>)<mrp
zXAJj;18?3Od3ZQ*I7Eh#V&?vMV3`;E;V2IaYX$$C%S~qiIu=C8hmsTq*!@9kZx0~J
zD@Rni%few=*f!7UbmH#uo?tA;<-jt})Ye3!pz8o{WrTN}X72A6UUHUog)_2_=~kKE
zYF#pbHJW}YJOb-wRkO1ya?X^TN!vy(vW`H>UC~8dZAh2UXfg$g0qejAnSit^G_N%^
zg6TMKS%M=(>CETJb2>~syG<MzF{ofuf>0QS=qJ3Jv3A0mLT`p#MDkXue9rVr%a)cB
z=k>yW`pbXI^Wz&nzyFD;W<J0FiSLeqmJ(yzWK>}~y+|098HVTu=Nx^Y7A1>yL#v&f
z1i0kmFIzx(lc#HEV^xW5b&}NE7pAuHIu?g7iEZPS_J*<4*r`Jb3}-F9i}K$ZvuiJJ
zx)-mhp3XTsgQy1#&f>aKN{wac_tjgY?Z@eU;2R_#B_)z`%+BD|ePod-z#JV8a|7*&
z(qHmcmgOKAU84uI%;JB)I~=d?LvKx+c?BueD%l3eUb`Dax1$zR0nz*y8Uz(>4((#U
z^qdb}_nI%sg=YNTdd6yzW3@Ptk^bHo%YWb33T|{vFU-e<Zq97E62e3bo`=Ur<|XrT
zzVM&^<6nS<ul}chrd_v{1j`!*Njh(195F<ht~jgQf3FiB$%45R0y~QzXjBF=QD_yO
zKR>ZPo%y@J`yJkUra4OPOiEY^EoCWuFtTFUbr)uLiOBG0hh)L-hA=lBfDKl1W5T4m
z2g7{5iWbNvO0&|isStDM8Y0|+iT1(<YN_Oua7w%!rfWK)RAu{TY;+n8u&yg<OQKIe
zM5=o!7{ecSi~c+T;TLOxFK@WXd+w*1$!lcjkdgj%t6Vn0RMsRMaM_oEb)H(K+VMK>
z&8;iJPB_*#|7w$)JqC-Vvu>I5wa~jEZG{*-d7G)Jh%m@IT9L@vm#3?+2*%*x9ryP~
z2_wN1JshV=EOk+WdyEOxF7dUOrxz~Q3tt`YZxp8x<>08sP>h0<tR<*785AE`8^Jri
zx;ygi<30BW?X3GNx4YI#Rqu3blBRfB6W28XgJlVv?q|Xz<KtYb?5r#dZwX7Jfru;(
zGBq($qHWumG77^C=EKDCw9u)9L%==D;c2XNoZefG$A!~zVVV}^I5S2iM(Orv7ns9B
z3=75<DVXRu<AfhR4EM()Z|?7CYnFpM1pMUaIZ?6{JKlL&!^{)8W`SgB!lAK_5F<m2
zx|9pO4j6{sUiZd(oUC&}e%XRmf|;nGWgjO6z2(N|+`Ans4=&SWk3V@Yl*b{TP3?u0
z64y(n=1wb}^^(~xnX(mHX$<GsY9ZH(O@$D-xnk1gBPD=oNusySAYGo>)<nvM=q+=!
zylg{kM{N*;C8tUWRrjThUL7TiEr|i4{IVH@`+>HP$(<MFymcOHBcs2LGIziI-o+Xn
z)e1?|N!+fLZat5UUOJ1h1ZxP!kgd2g21e1}=Jnt(axfo{M{EeBA)TdD4T=e#R>UD9
z92KYPZu{Cvk4`U%lq}vJ#Cc=nv$p%!((l<gD_Jd5lvM#n8vtD~upfZ0MY=lBZUuro
z2P^b)!&2`D)=ta&Qjj)4A#NQPg4%EQ^@KVE^sj~?lLmG6i1tB71w!(ZLd@wDWB>g6
z*k=gH!8~B3{7O}{o|l(Isc<^zG^RB$gDp4y<A3;T{>8ufSJ=ae)(l%&MZBpi3!EI3
z^E44d#JfoF5$`6P3%3a9pkmCMwbFq>mH6FoIiD#Xf8sy?=fCA#3SYf>M;(S-x>(em
zq2?7+#m?K2wH3xNhO8Xqv(d0Q)rB}?rFm81dk7NN7J^KbTN}*F#9@g7;j7S`lh7xn
z8e(L{>U0VOx@xe7P6Z=^Zh7zZy4jf{s$Y>mH%MC|t(h$g(J-LyHiPgQ{(#rdn+zCm
zK}N!D3|0QW>IFjZ7zVu+&fCh%C6QM3YpMfp81T`FSHI|3%)u}c0uh~8F`IOgxd{T(
zvV$KQMWH~g8Cn<KLERE*-4s<E86~q_HlCl(TrMkZfSZDJ``gA{D_GN6oFRBGg0wd1
zra`IXZR30u1gY&sM3c^n%_vS7>*=a2zOe_4Sf!f!eVsK-)5P8ViTj5;di3-V=(SUe
z9A3jf&PAaeO;D-(0pR6;lsWOBj!No{Vml)mu!6dcQ@~q?H=5C&((U)VV0rWQfxCwj
z-8AkWPOMLxSp0&;MaQH$5@8C2Dcn%gpt;-AM2Hb*1K@~Zq4!2~fypmmE40S(bYMGQ
zS+ARnu$**TX&AsUyU2uRQlVjMjY0)P=U|NN?>E2Q#{Ac7B6xw-MC)}Awp-4mQz=>4
z6}vK9tb!KrVTz9B5GkWkssupoFe5Pd;1sGN9j4eKE437sFyV33(pcBbb}0nwHH2hO
zmlIY*shM07DF;f+q`WeEVX;PNh`BQAU~;ni^iIOPOq1OqU6jPq{aUe%8L<_PK4{x7
zJ*Wt*@p8ZD^R=r|d{2V=@^ypusQngD`=O}AP4ATMnX1Roq&p3k9y~=+#SO<+GN~3+
zo`5x$IZgyG#L!j+3){XBdEpJFJ4};xnpH19dH{R%@0-3K{)RaQf*GoDkmlBhNSF`I
z)A1EFBvs1okk=pfwPSa$hx#7c)JFq^!udMZw|c-zx9^lh^Lq+^XC=r+gNKH(uPZp!
zBXq!a7)B=v`t|pT1ghxYqj`ntfCyWS#TjVb+`vyeZh4*$Z12zH3g_!bf|tgk<A`zM
znl`@s>)-O<{CEFru5J+Nl=eCqu}*jba_~p23s@I%E@0KF9dv52hcfN!L~A`}SaPX+
z`tE!3#}EAF-~W!>99zz;waR%AEN@SDw2~Ru6%I#4j1FLE&AbZbb;=^EP9sf;aweR!
zn437%gRz7#ahMJq=L2(iCKu5e8K<8?7dJp~V*m6(j$&!Rf=pwq#d=wPNGOvO)O$ha
zjl&gblv7E(T(7)bH||ooLEDf!<`23*e^xPItr0j)bWj`3a3exl7bIXuqjaAx(vjD!
zg1b?{bU_hEkBbiDh2L$1g)lIob39JmEen_<FV`zSetPCopj8R5h)(?F2Czd?@w}fH
zt<!QLSwpEpHrzIuPUI@$7;79kWj=oTbOZS;lVf(4TbiPQUMlIb^78y5X=Qd`dZo7{
zsZ%L2YGJY#--q~Ly=9*Mf7Ja+uPs@Y---QpYi6gp&Aab;A5CUNWM(k4ie!>SLV!wu
z070m!rkZQ4kDxE0&!B;zfiw}Mf>4468dO1mR0OCZnaqaSMR-Jb_nO<B#+uz~u(qB1
zf}lwEK?3O(cyJH*>-U~(W_$0o{_Ed(MrN{>2|y{rk}k%gx{iL^gLjl!D$0@zopn5<
zB01k($K+-zGS633lvDJZ0oQk!l8IsFG>`1B_S7bU5v7#IB4!vRCrT9y+wE@4^=`{%
zvm?Yv*LUI;*m<-wZ1x+r`wiROz_95tR#?uZaXb`ClCZj7(2pE(#!{4`vmS3e1%)?~
zRbZXQSy`M$H)7$6A~jxxF$P<}X-j82Ax2fvQpXgS{EXL@yd=UDQN8Gry%B?yC5szm
zjFEYkj(bJXj5eaO7&`h*he?HD<H$?fsJujHr<5?JdA>LHwboKd3`5Um*Q1L;Ync_x
zOfFrD9McMdQj28Vr_}WBk@Iw5oKG_At4$}YDxEiU+Ogkl7_;Kt;oQ<c4H!r*RZ-B#
z7W3P>z8t^bV*+Jti;-3BAbf8QUYch9x3~h$MX$Dl7zNg?<Pmfd1Dfs7)4AMc&849`
zg<?EadrtF&o{r?0n^Ht1oYW@cO6B5C*9a>(>jjCT6Et!q*%g_Q8?_<{qEc{L(^>Hh
z(8e?rL}6Wz>pCH#HF#IfQ3V&IL1c8QS<cdkl65rzs?R?2f}jDYwx-qtT~6=SS?_Ys
ziv)bhNalS2qIJFv<}OP73N9&!Ys=4bk#do<)m3t*E(Aqw1u0NU?x}dHSYlSBkQx~_
zaJ4U7@3-u)K#iW)zk1DQpT5R_^a;8Svj1xFJ+R)>c}a70PQr$ab|~%8+GCu@SbuS8
z6l}T3!KZ{JaXgQF^Z75}cxF<LT_<5z%6U>w490M?-J&d<k4O656|v>$h+?2X<C>wK
zxlX1kayvPt$P{xUN<|kjfD=iq#c7Q*lBctxRWcH5g`<|@SR29ey|?s($kCN)$}FSX
z`VfahSjn~->nzS{Vu&1%Pdq(6F>JhyeVxV_X*R#j^&OD`-CV@Pk?C2|Cepq+WU7`p
z%BS<lJTGF$NI~g}b0U?h8nMV#P^-o{jq`@SckHe<?7E>rcCe(#GDcz$d*GlYUw-He
z)o4r=nA<xihC5cExeya$ln|VpVQ`ABGu&M78HNF+<oh`9akgt&@={G(=jeeFGrjBZ
z)<I3gX+&v5Tqer0u+<6=SSr0UY`kTeA=b=N3MCcBB`_{C`%>6-J=z*#j-nxtk)`iZ
z)^c^VWp}-0N)ctxP=df@!@$kW4OiED076Q#DAzZ!R@qREq3R+|UpX<)fl?~le!#m;
z%k-9oHS7m=+by{kuC8}nU0t!?@9FwMWL+0!gRq^uz7zMfUK}H})6`I~Xsi;y0F=1P
zDKw@iLd|$3#iMna`4AYVnb?9kR8?)fp(#fG8C6m&b*VU|#fW6dEQO|YP#1rT+T>U@
zNk{3m#T$>chFV0OvfcHh3WN7#quFjc_M4tu;i(Y?RaIg?-3ry-dxpVtbv;l644V$u
zKPwMP>u3ee)0cxwYe901rYTNjKA0NoyHK&-(RWgG?Fwx7eM8$c*;H^A<&>aHxn)FD
zs1#f^-)X7-y}N9Oas%WnjX~>*Ym*7q%Rgxja<)n4Y$c@Dv08^Nl@w;Q773`aR>o|l
zRo^m<#cW@Yb%&YH_-P<5GbIO{u{bTMsVjqDBO{1{JlC%CsM6z|VenRDk*Qo#F%5;T
z>D@c;@y-ci*bi9WN#~Dj<YKLD8`Rr==Z1L5fh<97D>Hw!_${mazo93(YVkCvZEg(D
zo*eB@Zq9y91R!^{t^rYN_r4k-$qjGZM=??=u+ojFD>@_SS5ZQe(^dAKyw1>##wa6e
zq9&F(P_h&S@<=k)I$|S)$DA<Q;(bpHnP2|-zvHuRgWd0>5kXk|E7{X%dzp*pT4i?L
zxE3iTQeAx>8UeFZC<`GQe))^fIX%ANo3GE*RJne!CmTyKmictLSkrs2C4j6}#wAL`
za;a#QsVX-2zcomSMq+FRbE;WRTQO1Ux`MS%+$y7(vYF<+l0p>Khp;j*$}~c?Zt5wT
z-Efn37Eu#9BiS%jRgsz_1#v;1fiad?BggX-PmgcuI!BCw-imDe@1z(I30xIeHe_<l
zbXGQA>mAA(QYjqfiKl61X|r(og%#R~m9e&_&1Qsc>pRWRTl&F)Q&cBk+q0oREjTlZ
zG_m~;l-TP!O;k#B`@4bRdWY$RU~*O?%AQ3_Y&1jX*=+}IuCLg11N|`2HM5<*lgQDS
z6O@7}Rh*}cq3?Q}lS(axK#ZC5@k9y{+XzOz)(pn5aZWt(a}iI249gN2$B}uO2&IMo
zw2o9e1I`dAxJ@TA!saSiyrtW8eE7jDK7927!={%Q&#Fl|<E<BXFBM`A!bNED(#O+@
z!}-KK&6JeMxw3@FIL#O(j)u2aSCa0y+c9(l-uGB5TH8duKuo0+v@zm<=nSTc&w4ci
zd4U+9JZE&vCRM6b_f{1$1^qZ(f{#`tKxVP6>p<r2717#;kjPP70x5+QCAwH4PGr6B
z*bW0zm`OpL1Z5*p7a#a;@c69gv>|B4cE6z;I&uX!c$AwiBLibaN@=Z;up^_;ox4bx
zeCI$zYMp1wLS*PVA+v>K=uO9V=;`~8W$IdJmuVZT!2VcSD??Y<Y&uX1*GZZkO2#s4
z>}yP!R7&$~_%18;@86}2r7FYfAi6?1T3=A@%ca%2$vT|%l-P)L4KcC46H;mw7ecL!
zj1_YxXL(<*Z*D23!;c+v*E5YrQk*Xke=cHR=CzoFWkii&l&0@`dN0<&-nN@xD=93L
z6woE(t?1o*C+Z+)EZz?oKZpvUxda-c1T8>46Ni<$G*VQ38Oe#7NJfNf%B#`_g~)8z
zKdS;^J+NzYv@EGy*y&a^gSd4Pp-LM?HUgV&1w@s!P@~%nT&;4y)@u~i%xg5lBFL=9
z$fZ_NDj227HRDo1DZ@C9(%r9!PF<zRg>i{=-nY-Bc=Pri|C?X)!B2i7Bau{4&E>Dg
z%$bFnK@-txruUvSNcM{~+ZB~Fs%Df_AylQfKP0~Tn_sYuk-zxr4gJ+ExfN2eg{c*#
zSaM;?me0R=;D;{*?+z1FjLq|5Z9D{lUhUV75tPyx+YV_#4ZL>@eTVBhs;bQAGspA9
zI7#GoisiERna1-nO%<Wrm?xLQv;<;IIOD{!8|a*6&2)%4i>fWl{vbICny14N_tlr=
zR2VkB5bm|*{C6QOREowtZ_&Ykw}yS+vl%+97me*XM2^!!iek4|Y9nqLvGFz5T;Md6
z7KlV^ly*em5NB*!Fo=fVG?Ij-(rNk7Im=6lx0p>w9X#0^TrYAvlOkA4QL@o@ooCoc
z>RhZ<j7qe_`2nphxfVjHLKMt}B_^si<f^E&#J^OSm%unLvT-{{q2M%h6-*42X=FJ~
zq>$NYgHvLxlw)L>XR;Fi>e^fjW6gwS>l~@9kFEEPs~3CjUfpqXb%S-5qdU{th{=JC
z+u)of=gbl#OIj#Rn?Ftyr}IR~;FKk&$UJF|ha+JM7k0Hq=BM5dc-M<NVXInev-n+E
zuQb@!Xp7aV`GE?w9ZM2_>=wycg;j*@44t;5nz5b3cO9J*4~N*G^@~i1bw-|vIddLI
zA?wvDMNpABF~HDu?6w>F-ZLyc#fT(QT{<N$3n!OCtOCnf?-;flmKgE9ZL!gc&RII=
zF-qLwa?Qk)7~Hd*Q`9CsMb&n{1Ij?>Ev3@pr>i(gI>*p?`o2SRVj2S>6-reUt*BPx
zJIiM9?XVZW3Gd`kskydj)6!V!iuf&KNdLZG1x?B=fUB;gGm*JcSAPq+o@HGYY$yM|
z6d9dZg{hhxC5Bc=Zl)n*Voar4F{pb-slBkOYY|#?oBzg=<aty@QNkxFTBC}1KL{Vj
z(OE+^^}<uZ82K8lguG@XlK{Qy+W}+*m`j#1QBACls#-N}Ld&0DXXK?;<|RlMSEW8M
zT3s0Q+DZ4_e0qdnSu!~$;iqIN1Xj|RRU$3B!29|c5m-@$QXv$HUGAN24EdFmwq_Wp
z7CKgmsWdY+;YGz<B;-jM!uiCtN{mY;R!hu6Y+k~2AvlJVN#jhZioSECu&~TC@4oty
zyN^C1qQy(5RBUIk-V=0W(sP@7%Sg6&24~b|j8=q%fhr<dO}gj)>vzQQ#4o;j<n5gJ
za9hZ2->W&bLrHO*!$ryUYMglcaN^-KQfs<+ThuJUYwtTHimRj$nn5~x>nxk?hTVRL
zv5I9nae6v&92d!gNzeAp|Gu;1dWD#HI-Z%QAgJs0UL=*xUACk|NKxXvn>mxXPb!ur
z@#ghA#wlTjE!KFf_01^iKYe{?TeIhsJB6~9q3gKWZrSxcUFRDUSP`Q@a59$_*F9y7
zkaMiL%mEdUvnB?`tc!SRTX0&@cZzQ3>32QL64^MzRqtBVuVKuEvd&_RBGn>!^tF%`
zEGZGQ%wVlEC<W(b=JnGfID_||zHjllF>;*FgjA@x@b-9SjERl!1nx}|C%Y~H8l+X)
z`D1IvltdUu#^VWHGMm2RX4|vryx3=YFZ28)N(fa8AuNO#IWIFw74Dx-oQ^ZQ(mL5q
zk8SrZA-~hE+i-Vvi?y01M$Yp@EQ!^4qgF9Yan@a4qBIAl^TgqJ#95KnU^LeI#@2P>
z&)rO%QYwTHWuA&0swtP}#4im+HHNJcvQvtQWfmWXVr2ueR-<&qIL)vb*zLA-eNSD)
zDXx?v*$vKOs$yAYISkq?&S*m^Rrm--qm68kL*KLAZCOgDOp|o<E3r!nR4NNF3T)SY
zj%zNM-Ws;Or}rX@>|ECh3Q1mDQe?lixW2=BFMJuRn-LaLq*@6;$|;j#6km)Lde)l0
zb0{k3^UU04k=6=7z<5JH_{Mpm&9aR+(zSxBt{ZYLJS&mEpV#4ZCMGH9OUaVLRtswe
zJM}rKGvvygYwN7#U0!3km68J>l_DLx78#$|HfM4bW3$sVLXzOA7*p$zRfEl20$lrE
zB`0!fB;hKFomm;8s?<;}^YI)Ls$@|$C_}E5B{ed01y>D)8K`2elyhqu>;(ds0-#(h
zyH|3W)&l2F^CE0;YuI!>!vMXuc-{8Ts+yoX5pxy!>=X#g47qu@G_M!i4z5CeR%Ns{
zRyEGRlB9U*jg@?pR&A)}0Z~yxw5+1`ia}CS#a%U%k-{z=?%BM6`v^}{<~W7RVWpIe
z1Vhe@%PikB1xiXdt$2L%n&aEIxa*gU=S57yPRKc-HO+;hHD-;VnNuTKCTJ=nv}uM!
z3*oEZpvNPhzkWwf8l4NJxd(<NkW)r^!xYmcX*SGf9v>fh`#8zSvaU&s@@|#pmb=yj
zN(t(0w5XeG*VCUfn?u0ZN=`Frnutpj*T}MxF4e_4dd+@Vi{TIx(;S$lh0gj*Q?M3}
zg22jNBCm8C{~##NG|kisCTEP@U|s$XlpO!z>$?#MjR8%Yr6>%BjTd7UYouD6VkDJJ
zslxZJN<AlUnR;PO*0vB0niLZyYhudusnGQXGYEw2H&&eMX3xH}++A&Cu`h`<MFI+|
zE#o+YsdQPY@aY^loQ3qIN|pz!#JFaeXPyoRuC@cGv$&d_W8^$e!cho;*G~sh6#x3e
z`OKICIcKINFoi&B6`9gn$bz*nkBQT9=J9DnE4bNoY&RP=o1VV&Vy9X%wnCUk@;oz5
zvuNVyne#I9_T3}n)0ttjkrz>cGKP=>bDYQ}@$(;j%*T66p|I?)d6>2=aVF--G@UTU
zQV=pq=N-;?Vs2cHrx9mooHJbQZ`s-{SANUj24XFQl1MS2s(~UMPKkvaD<vYGPia!T
zT&X&eaw5javPi*^Q>J3X7Iw8}2HarLy~A~$TysN?M4l+glwQU*&eClMy58Vt!xA%2
z$xI%lM6T8MIOCvM`u4t~a~?}$Jm*4~BQ*x1(j=;+Pdbmb7Hfp$WSy0FSzGyjwK2=J
zrgH}49j+xODpT<WYF16(o=H`b!{#|qbhCgpvPo1Wh{Uu6mM9yVG2pD0S*JnFYMeYL
ztMq$_KoJ-aWKR731_pTd_7U$#ERB@4GM&{b2Zg!KHlI!-?+#~~t-D3kLm@z$TML<w
zi$qs7L}^u;6cj4FBWfja-!WdsOS)EaO#-zR2(>xFjWa3>j4lXQrj9h^qLBxsdcZ0u
z-V#^$yVh<yKZ?2~B~lavB<q}T6tr!Gz}n(yW&Bl2GtzR7oHKI>Qn*^j)$WRYQD{F<
zb0g$dLG37-DO3(qVmvR*<4BHc$d!C<-S*znR4STAHUu#8-q-f^LW(udT7Rl|Kxhli
zhIomId5Of>5?5;#g@}syrvqwhnPTS6Y2-8~a*Pf2ffx$wv6OOQOo`4(Gje!%V0`zM
z-R=fTB_!bh&=M*|?kH+9jD_aVxhjCHm9RSBX)+mmJm98#?%$nQ7T9b%bl+1FjAcew
ziIoS5lQ*3c5(I*LFHuI)r8K@o$(KS^5XV+jse&$<Vpbb!Q6fwsKuDB0LkulgN<AB{
zt>jGwwMt;ys+h^Sl5%Z6E&xp_qEw3lkb>A+gK2wgL7`0oBV+kFM!FDC+LN;pKGwIm
zzGF}k;>T^}MtP&z4IM)-S@v=;X66(rDmRImeE3od7D;U@xk!|xGHBb#AwWs6%mq}%
zb&jMG+QH_=qjhDc47b;Nywcdmh12QCX<9fngQX=#`k})Z&H1=+I!WZFZ}Je^S((Xu
zL&<^jbOxn3g@vbSy3DLk^UQHsn!hu|un=+}q|9-iIgB%@&7HQL<L*j>ax6LWc$~%P
zMr(G1XXrhB=P^o=b0Wq-3Nvw<n9nogag?x~l$ldzI?lu;P&Wf1ghoV}8OM?T^Z)Vx
z#>49e{u^)5w4^fE6N_#yMJ4A*2!Z1?L9OhDfxhb~wQ!y$rfES@Mf>bM+s#f6$ks7S
ziehS7`;o8&`o70Ei)Mk|(`8}Ev(1DM<m*?miq`c1eFsSAyu{3(|BF9mJ8Uq<T!Ly~
zHK38rHs{1NNs%;<Vmg#7JiL3&|My@1iELiY?_Fz6$Wh)!=kP7DGMCIajyxVt;@qXg
zF7;3T-~Wcrc`B9hG?GK)haY{)K1)Hf`hSNdQz~qBTXBfgVuq!SZp`UQu9+!Js1uy0
znK=fouJ)4H*?BP(GGa!8lIK#jBGg1skzzr6k#!j_5=X18pf5_Hoxvo@wzxpU-{b4`
z>u-b)v1VT=2_dTO;DX9|8s%%>y+c{u*z)qu#jsFn7XOGQX{I*nTyr$lD;J>=u0koK
zB=S6KMG~rjltYsfN>`S0rp}@DpOxHsQjSz3-G7YY3aXU+tj<|dayVm}sZS+@fI#cD
zp%L+d5Up{#tD7K2=Hi0=$oS1U*HTQS3su?@iy|qD>bDHqVe+%WB&W<g2Tm<_Wtp2t
z{c86tw@H3Rt%6AA(#UWMWgSY(Sj-rM(;8|N$JUgoTAb{>@6bm04&yX2&PzjDYLmn{
zbd4|`dG}87i?7eTI|XVf-0cUJU>TP{DpDlnTv?XXQdXtowa)SO-CK4a{|M_G7S-Ha
zMN)Y&Wl~Bc$SuH2(rFV$JG}P{ox>C1@HIXL-X71SS{RI`#R$>)j((8)JWeA|PYZdd
z-0lsoDUnc<a8cEYB&nv=S7>G1n7qh{Td!kObMc5MrI1S$$KJ~x_w3t%Rur>l6wt!E
zR<X0>$5bsNC~JkBx_)n^6*{YD^FFP~s)#jllib%Ls8k3Gp1-4Fp!p%nj7%x4ZX}w{
zwXUs^N>oKVg*DH4UaD5H@YI?dGbxFdw0f}K31d_jvW6uU^peoMrtb`9V;Quj!{By3
zDra18QMwXCW}1V9`D{Ga%ch+YETNEc#aPHnLGRGp4nOo*W0^`Nhrp7h!_$p0lBxt8
zDP(Rpc+X0qjbY(@K9Qo#K3#ipYtDp_m=;k!IHTG2j_szyb(UN+=lRTWmN7z}7oLt2
zkB=t`FfWnQaVCwTmq*Jc6{ndrPvmLh^S}BPzy8fv&Cg+db)NsSN|&gso##r<6E<vJ
z8N%y-w*I?D;A>C!`FpMCM?3U+@B2t$+q%2f@d%IP`Q%a!zM{=3?_b}u=Bxkzzuxof
zUWrn`G@gH3Hr4)q9W_6HKKUM{KKp&I*L@=Z)he75YUPI?d`!uub>sq7pT>o8T!bUx
zB!XI7A^B;eNg**#3s0v}79*vFUm5zLyBJgXp~r57$hgEp(yB>n5zA~>NlPLZ=@P`Q
zTagf%rM3p0awExozpH`YeDyWXxwbJWK+DF8C@+*MoW})#Z{FNX-b0l!Nug46LQxu<
z+rnjRQrm_pp+Ti3Jk|!SO3;rKoYojBjhNYr`C26lM2OC!_pU{9GjyS?ja7RtNM>$X
z1(aufe{Jw_E)bLWUW7Tw3#t{?c`ggDh|y1l1~Xgf<n4Qoch-@sB9zJ)6NH4TvJPuA
zDqEH)uCB_ov_>g@a)YDSB1^S#Q!a>fQ41N+nKpVb67-Z)A|vteB`Z{F&U(toNKiMg
z{zj100J@@!#@k9MipO){@i?&?2CfFfJO{qHZ-q#a@aH9Fl<6ANUW3v++&}Q65b;-A
zw3E=eu(ZG~l-Mscr&ZJP7sPBx2)a&JuARch#5CPgs^Vc;B;>2}myt-{4QxBZ!_y<j
z!wJ`QD6M%qjO3WajEX0T+eVs5-7*`byBE@+#k{vaYHAfSX3iO{n!!?$GesK<s$f<=
zR+A{ES}9qOp0yYN;nyi^@m-Jeon*bVhDj-n(e<+Lt-*d`%4cQYYZy?iVqk{)c1oo0
zI3~~-FB>DN;9Qf{G&u`eT=_!HG7_*>NNBmuM3Mx~lvGJeMjMgJ6rlG#zIS3|R3l5u
za1Mxnj)WNn4YL7*B6>x(qR~_+FfyC-UIIaKE=|TE%fWOWdXFA@^kyS)O0CRs7UGw)
zcr7<m&$BpSIZNMphR&mu*o`hRa5$dHDT>E}7UwgC_)3epj0E9a?>6i<8@Ue@=4IhL
zorP7u1WxCLdC4SQiKmg%(@0zrPHB;@P*F){;&l!{<!ZO#fBV1rUo-5kh4`p6gR=}a
z@%iz{uin06nofN2%{#7kTYmbZA8_~Lj(M8->YH!)`is}x?yvBj<L2rGKl=Cw+}+&a
zyN-+yDCcG5%eUX~H-Ga>UVrnJPe1*H7cXw<{J=afm~!HS4eXpo`<~aQg@5~tFZg$V
z`3sIuCw7~TY!tiO9W`g>Df0So=AZrX&-v+(f5c|HlWw_f#)7hnU&8rx;>)kT=8Mn2
z=IP6~+&>BA_)q@xpYr1$ek3ouv)tX@Vx8s9@e!rq<%_!(Hc~ksp7^W3{sn*XXTP8`
zGO9|>@DKm!XMFbZC8ufP|MY+Ozw)<#`MHpCVj+c0Nb+2sjz>|0Xc>*^br$J7A&Ti$
z&H}tT@6jO;b3%LZ;3%~a(n3)g;|<1XjMapasN|KVO-_aBybx4m+%I&>G6yk4(N3h5
zu2_j_|DMYU4(ExkD=q)QgR<mmP)Ir`RC0rAr#axqAThZslwBo(&?wnYv@)cUs8ohQ
zVT}+-OD#<+F36L!;<Q;xCDhoM;iVB7n=ZdTmyuc<As{-Bbsq76w=GUK%g9K#h*(zD
zjlwD?hr94ysYpN?Rgf&LZTs3Pdsclm*HPMnHYoF6ggMWlXX~LZRkd&~a*JoCId&Rr
z$t9zdpmwI^Y$ydK1z4H=+TffVFs6}Xl@`avH4~xMB&J?wO*D0sQbcZDD=Gw1jPi4h
zg&fIYLDADY%N!!RzQ-!X)wX8|k>&9uR?Zo!wnB`piq2YswLBh9<nxiu%MT%DK_4}!
zbtbQksIVF@h_b40ByDxS>$&SK$A<$icN^whG0Nh-qW2PWT&r$m(oBj0twhp3#Rb<m
z9-{r%mxc$7*gUjQrPVsvC_z$llM2R^8h?s*Sg*yB!D@v;QL{9d&Fx&|nN0z+Ha(K!
zD$0Sb^B89@oQPb7gR&MUAgF`|ypyvw<w8o&vVW1V{#}?3St}n-wFpJu(Rp{#Tc?yb
z=RjKXxLZqDYo+8$R&w~~C{~|3%PLexW^u+^v=%vx)=H!pQRF!x2eMZ@qVd);=|t@e
zy0@58u!A`2N!M(eb+S~WMk#UJ>zr&Z&F5Mx#grD7w9p$bGh>7)ik26rw@!>8e2-Q#
zgM1jy45p*x%=K<d=Q@<N<RU7BQiV9_tsEo*<v=JhtISn`fu@uQQHp_@#P+t#3GW=e
zcWk_al3C6t#`7W%*H}LM^mpkuTT0gQJ8R3X&wTK~iNeCK-bVKOflogEfX{yLF}GJY
zJROgWVP-r{eDuMGSZBD}-*EThhF3Q?bi;tQLN=VnGpNM<yL*U{&pv#~^@~^ZPO+;8
zcGhF^0wtkq;s-D9Fh=qE%{^r)IIBnnoby-@8pAfj)qcm_?KS)B8-dd<g@C~DP&xFT
z<LN}Lh7chnnV)VqTdsFIGEkl2_C^#Iv&ul>#mgJ^*H@?t-V}WAP>9L^rP1DT`{D(6
zcQ^R6mwzr#n`KGFITFTE*v(H*#8j}_Tw-X2Ar4_Fl1rcro%L*8$CNT_SV<Pv08^X^
zC0ypW%8DtJYMva+7&+dL93N+tHJtV{*?GokA;d&!Hn7?obdp8>eZE%bI;Tww4f5n7
zQZ<yN7Ii^CBwwR-yGmckTT08BCbrcIooSo_BPK7_$_8D;k#Lon>bAM(+*AW4JtJPt
zO^zz>N<VDz{easH(#@AbjtS#D8gRDFZd;~^SxvQyB#u%=Qccn6YXAWN07*naRGW&}
z-BysqL~Acjd8&;wDq0-tT6mG~I`Q7uqK+s+yONGj1IWl@MLh(nZiGPNKxec~IHkn8
z*IKHQ>=lASTTn_IRITqY#$s$!b=3?>2`{>|IBBCB8Bt-}+MUY&7#df?8Anu!lEeUK
z*Nd@Ku9XlHL+=^9<7(5Pv>_eOxK433I9^;0e0V+JYGOPqmiw=`{^SP~ziH#Y8R}|m
z=VWyDURHpr6;t45>-hL~fWsrpbd+M?>4<Zdz4zQ*@35|?sGiPQP(tPtLnz1Lbm9*_
zzT@U<&sWn~o>grnSUaRwNMAK%vos8K^^vT!=TVVoq@;PaTH_qJYEjj+_f?!~V-apj
z2<bVrPaB+d7^~^KPGsDbMv`5%UOW?E<x`16q!L+b&KYGzc_HL||E<>OcjF9bF;-B`
zU}c?&n<f)Xr4nKx#e5-o)FhjIQ38kP-2Q!o^Rm%0=SotUxlS10WMQYrB+<IA6Ikm|
zXEEonmC{vAZ&1a^C?E?Hvj&(?3vr1uFKqtcRjCUjUmMYM&T-)sMrM<!i1AR2nKUm<
zxp3Tg_PvH&={t|rnyc-Wb3Wtyj*o9&@apCT&S_p(DF#ArjQlJlFi8oPN<S+_ER|T1
z=+rhxiZPLb6hh8QMDe!oa414dgqRx4smNxaC^42$wNSi6WzUV>@SlJ7DZlyVky<K!
zCk0GSS>}DGiPSXP88|HyUq8I10@qp0n4Gng9Eo8jFSA5t`ktLGT<H<7BDH!JrHOgO
zmOyv^H9y*Jxb8h~a$%a_oHM(S&`#csoJ(WfCwVT=Mnk4;c1_Q_Wajh2(_!ZR7<d{J
zLodXp919U~ii?e?x-0;)Kuy0<3W;3AJ{E`}h<)%}=t@OvVG`G(pz7a(&2oYBdFFgP
zGo4RN^UVF>OiooyL1Z_kR<TEpv9LtRO7FDA8|iF}g1+x1bu-3Gpo%dTs-mP!ESY7F
z3`Q{?1CI|2$EOA79rH0_`$}9gDM=)%_f}-IB@w^n-0$z}Rcf=C?FMm=GY(xqDTBvT
zazwADGi~!%tF*V$iUo0}V^y0ptGt}$P!xMui*XgxSo&*+oU-^sRA|2NZPu&J1$lJ7
z$M;*h%?8_?0}T~pOXHp>tf~@-lpFc4&5SjWqJ*QNqzhHrUF3nfdCsrKHKjDpTC)mP
zN=SrjEbj^%8!ZJowK>qeH@;ghttDW|8bj6M;i9Nmt0_tjaI18i;5E6T_f6}r+Q>|t
z1f6O@O>HJmJA_p81F=@*%+IcklI(XnkDO+QwTL5U&MX=7IH6VLcH2XpG4x#D?0Ipu
zW!HNKR~hV<)4{QvkDTB9hWhGvK^gHJfaXHzP;Q-{16Aa>9q{8%@3491U;f$OQ0Btj
z)rOHkKWylSjt_3H`QU?Bym@!xIEg`(keCbOG81Fs_VxzX_f$RMoNz`I>QXd{?L!+g
zwtKGH*fH1VE?qU5v~=)@)2&lD9Hkm+3L=4|`9-iwIOTHA(f0%EKZ!9BO+aPgiIq}X
zg5t_&kRqh4Owji1Qc9Tixvc&>7!ydQwh(E$&am5fI=ALY%i%W%VVg?kr?7r6<jY(s
zsc{F^1qsiUa#rNrQtY(AKOtn4(Kszp!8n1JRR^}$be)xa1Pp0ToTiaDCQeT$;*v1h
z;H}6Crx-ac3n2!A(pV}}3>=m-MJeWP#$q{-3-fuRuZqLtBaXz$pBVZL<B~X?N2YPY
zZU+nnqrhuJXWT`WmmB*w<igg9X-%SF?ZR9Gp;Sq9ECs7&UaYIYM9vyIXW8`wTi-F|
zC<tG>&nQDKVryCA0$H<QIpcUW4E)1SUhvJ^cchevOJJ;Lj*m|qAC6c|iwcFOdE^*o
z<`VgY%sxl@u49QK<2(|V8Cx}59eL4>D5V+m6(MD6$xwQ9%M0l8#O<~tXZih?ES~S3
z6=`6}6{;eoOjv|$rL{$=MucodVVK3K@aa5qIM0NTvA#JgrX<o)rNoq|6vlB9h%Ab=
zElLsM!Z^)LQ=%_o>QQ}5#;cM%xROPuC9+Hl$9ZI4&V)HIj}g<zk`$2|#+Zm9l2aun
z;Up|gPME0-+buV@H{4ynU=9mmsf>>kP95>uQe&aS1XCi`%Hd(;@OUQ98KV?w2?Po`
zWZ}8wmJX^xS+kOzzqi-W5A?%^VY8KuzDhK)wlb5?(OlkdXRS0CN()L@#g%dO^B3}|
zmRV~q#E=L!3an0*(N0VI@y_FYN6bZF1#Jj%;W;!(y57FuVr_@*I)<Ugbv;V2!E2=@
z3IaJwn2uQ^m)Fh_stT0$jw@}kzIk(4CnyCpPYEL<5X3c542GT$Tj#w*DYqH9)(u9x
zpuOty*;gs+TKJ@Oe6eN<81c|Ac}<)Z!#3yL<*=(jS;M2WSgebIn2ZgJF{r99MaWtx
zSm#jI5+>o?_|1l0F6_5kmMH2aLuKC?yfN&(VdFPAQ@Pq$K6-h@>FE*EdG@;<U%WeV
zvR|@GTWmj&&a*fKb{(!8z&ezX{2N^a#rVPPj@!ZV%ilck?yT78p1*wv91fYPov4Jo
zXPN{3W?&47&TqKVj-h74y9bo9>~3yYiX~#eS*)%t3u8sLpRKdojEGg)WR&oUY<q7_
zZTT5BOG;vAunKAsZb%ApZ?zK6gLUE}2uOsy?}%|}=4%Tzb@bL?jcOw*^?YX5_dZ`<
zrRB8u;=TFTx46Ee1EIVm+9<4ojdi&8;n13xGt&}DIg^vP6{4C^Ms6?tx~r*HJ!#X>
zsP_M{I4$JBq75jS({-JvYbjMB1wsrIt*O3ZtHpO7kd5Jdo;i;TF=aX>EZ$NpF(!_u
z5nU2ft#nr8NoiRK719{7$}%s7I0cL<ET_n0T6oZw&Ud7&xIavclgQ|<hJg>R_AE)}
zXv?xRIgC`)Ql*#XL2YHuUlcjR6tn2&y%D=zgaPcGW#~Pfw*pazh`6@{h}nq2k!nKf
zP}vA#E{s|6^G{yzbowEmpU*rU&S=hj_4X}q-#&16wQm)sW(k3(@kC8xun|jSn>U2{
z%xOH5Qe^KvAMO?2cuv_d#!QLpdtmWOGilg_w|sK75e`U*Oj*oH*1U7H$V+3$L@(na
zYfAGwU*$W^p)geDQ(y_1kP1T;m@b!0ECrTISR{ZZE(zbzh1$MOq>f_r6s3xyhD6OG
zYb!0(2pX)P6JeTJ&I{*xW;`zJ_f8naS@H*Bh%AekDuq;<${`Dv#9FR)S8V$|eZNC%
z&(_aux-FgQP(_gzA(W*}q9}+<BE~EQN`^TI6L<`XLP1!DO`~2a)_%W(ihMV)+3fL~
zp+WX?kQyrl&456p&^qm=aTkwp@w3QCXHk>P(MuyE3jCkM8rK*hRiRL-V4am{**3d3
z)?<RcK<#VfcRvhh?Vyc@ypwoitK<+Ysl7K<Q>&GDt&8b|Y{<rGRH~?4Q7VhOs+DML
z$z~A(nDw17%Bw=PE_=m(g;yX(R4ZRaOHC>g%qsA2Qmctjy%YojsY~8Vf#y`&<bkRo
z3hTVONg8Q~64c8+u8oEyqk~#xFQ;g7MKUs4ifq?)XzQ8MJC5_r(|Jalo~vEY!+GK9
zbRvYr$!T6)?-}|Ho8Ge7v~jEwPKPv+g~PPq4o6D(YmC+0Z44!E(ds3p@5x%@U4L;`
zbYz_K9F8Mz-W};}!5fFwj@?%Cv_x{r%%QNv5v?sZFRo=&vJv#+i?<K-oj9+)J4}S4
z@Ya%zs9jK8qKauMj#@+}Syp*vQ;Y~-t;w$IGr@A@bd(%1NqAn=(nzfer7UJ0Q7d`o
zHCL8pVTlVdEaYCS)RmR7votE|q8`cZvnZ{bPfKkc9*{0H{{Q&(9kU5UVq&ymuud!)
z+l|OoBxGR@Qu#Dyd98?5t`yh}T51ePW5j>~E){;)i`#!8<2sLb2Gb&rQz$}WRF2MS
zvQ4-x7%aQZ0IG00%%pcCVTtnK><Y6L@)Afu&5M{obTTtDjX9iCpwvt$GVAFKY^-MQ
zgk3yOBd4iwd|DXi#H2F4N?d!%CV2Da4X4wLLJB1By%ZZoJolHFNd>a1#9DFQ(;4Wz
zrpC-z3mStnhOUztT>?@q%qa*#2q`W~$)qq*qCif$a~P+H%A;`n^FR0r`7i&LZ|<L1
z^1`ov^9{%QGgrePBuA}9J~&OBoa61|J*j5)X(3D_rz!CA*7L`o+%b5?Y&J}xnW?1G
zIt7vWq?B2vGrggCr_)S|azn0UV^bJGGjdDd(yPglyeqjCk*Qe`r<5-$h0rwEOVeQI
zkO;AvhqTR@+GLI~GmVRcq!=rea!f2CN{63vV<{tE30W~ufn`dZPc!3TX47vPXGO?p
zVM$Eqg`Aqf&63Db&<?E)o$KhmggKQgq%&$oH#BUVGpBQ8w+&K+mLi2xm3uYL!hD`X
zY@5EE51n_QDx>>;n+|o|fOXR4fg&tXxD>73UGv<bk+d$Vljc}j9jR$7b>m&E5bpZl
z&^lRVjC9Ac@P!O#4ZSfaox~@lJ%80;^RDxNTlILY(-w)MpqS;drD`oCLirZ*5Vh7A
zq_Y>ZH6t@^EvhQ5<vFgI1rp5C0!iEEu8@(Q#37p&ezT$y@@JJIt{wukgK^#bxiH*?
zXh;=@L6mAHKy9t89F!`O-W7OST{~-~AqP_Y3lh?7mP^5S*Ek7s=v8g_vtN9}k3ad4
z>s?2xjYBp*a>|O7k9_j-mJjwDUfu0v2Cu-G=B<%4&RGt}CytK~{KHRn{IJ`RRYHe%
zoP(!ae~3}woE)y6!jv-8n7O{%^2z=~_BK;vKx6pP(;e^5Gv|<4mcnjhna>k*NNjgI
z&gU~ZNA}wdAr<yrPpOm8Fq-L-(MAgLHs;PrwEU?QVr~-4^&QocpjeC8##c&{A&VkL
zkOU#9n#NZtN)b}Ipkhk5VnE5gjzLr}PHT1>afww%0;42pQc2)vJ5!ZflRj5a;I~Ig
z|FP>k6$5=Q1Raf*IYfIoSGX)Enf+^|DwT;(T|>91?-eMrh~)s#(U5FG8JU|0Q>aE}
zE?sAu3z_Iljj|Y}DW;NK6vV*y?1!O|kxtydK0cFgTLqnKA;%yoe>G8JAe)M@ny&NY
zkce&eBJ)j!bDHhYakZ6fg2(g1(<o4NCgV+^H-*FfYaVOnS6{s2aGvE}brND?tl=DH
zatVwvvgAl@#eg=N4_|EPtwP1jL2F8=XajC=C~sLPEJRLWVxHxDtFk(qImXiwRTADi
z(4OiX#&qmV=AZxmPx$Zt&0q4@_b2Y(9LQ5*2@9w5nGBr63@M{hW;sth<ji<FQ^U-a
z%lzSoA7U*mddFdm)H$Jwbkc(Oh!$KUjNh8#MbDg4!77DQ2Cu*=>C8*}qgM_^DNP1g
zC^?gYm=Q^Lhj0a&yIhEgRF!!3hfG)^F$hDq=F)r|BJ+|Mr$8x)fd_>Wvk+GsDRtf1
zJ8$JVWr$Pce4H6ik$5Anf~9#wEc3#65<-(Il{6>9C>aXqN{WS&p=3loQ{?BQkVL5v
zG9e>YuIu=$)(bHyrPK=Mim5HsrnZ!{c7MOmmtA>9DIrol@e?UU9N=pC?e1kg<ed?J
zb)zNSx3ql#s>m~0GvKx8v)2QZ3XSX+qa=)KC39w)!O!}0oO2foUbJq{F3BPN9ur&S
zan1avk?BeibjTPh0bAZ*;CrK_5U@s==z?xFrnwRXLh6auWj028&6=uP*1h)<o)#rw
zN#JKOt}#Ze+S6r@EvhPA+PhL^O-LMqelD0wq11Rm$8_UcTt*JL6*G;`B<nUyA)_=!
zJs&V@XKKR2p;~Y~Lm%0tN-0~y)a-t9<>7eZ$3J+9(~4zYC@404&%39$JT8gX$C;~6
z^YhPM@sm$>_!8Lc9Gx9`JH6o>vn4fSq8uTV%DEIe*KvL4_-uDaN1()js+E@?eZ;$Q
z;q_tWtJn9$Y2<V~;=2tgCZ=g(>n(TJdvX0sg(OuFwW&Fb6V-*4jufRWx@91c4XN#g
zy2&nO*|zUd5YLH}#61<U&UVfVVUsnJB)vBRUT1^CXF*GhxyYVZ$+&KHvhY|`ts+0p
z1*L_&E`eau$nbaJGqArA)^AN>{VM)UGRLb$>}zY&Rz7Rn)nz2Gj#@4?OCzVGoFRxL
zW>gb%OqkB#t-={$!5Z737-Q*OPtt{wgs8I_I&AMGE_qFJDnigug7c6NxK}BG*|Juo
zDz?YkXnfaEY9wW;wvBiAPT@L-?}dD*bAdi%a>4hCm%ENnUS4tCS-!fz=l*cwyaZAe
zMSyb_>n*7y=9r1A>zsv1q3bQT*IT@*oQ@+}YqF}i-s3kN)f=i&nBK9e!({^!8J^4>
ztx*up%nt=qZ>YPToEDNxyt;nLKl#aL{Nl^k<XD>h@yu~N5OQV8nd@Q0Z9kxuW*%qC
z68ZGH<7YdCDl<#jGcN*PRb!dP2U0wPw)jCpTEv#pip6nbe%De2GL0=vrb;YxY2-?A
z?2_kHS|LE2(QEVt>Rpv8F+}DmP@QXqrLegb6=x)BEw{P;GzZ3UAva6qT0wQ15{k%M
zm(rv+u;2D<ww@l%beK3ioEfJ8wUAPfKQXY(6Cs3V*$U%%=HcPU%i|esHJdpI^uI(w
zH9{taNSI@D6wJ-`R!Fuv2|`vce3^Qgk3YlG4GB`;zPkVWdet@x5?4=6Y}_N!V>R9V
z3c0reQ4Vhdh;N57hNL9pq3Vh;vVS<|a8{HW7q%yYvZ<DRP)cfJx`sryx%avW%m0@m
zu_%OpQjvK%!VyU^34g!{)-Gm4i!ipEYQcA%tmBk~#AjA0SV)6tAw!ln(gI<Brzlw+
zP^i{|%b3z2-xX=J7H?fcQ!S}lQW8R6tWq3VYtYI|0iB9S0BJL6wH76{rOq}hWS+@y
zwV@o=ShOk^<povt8?~YzG&81F$iLig7-Qzoe(@zg`1k|*&XTiY-t-&}@v{H-!+^D(
z$KyL{Ex7%_>&F9MJv?EJ;rBlKfdAsBAF?&#9iXjeS{9zpuX%hIF)uzOXv>l{2_X~y
z@NS@b#T+s_Z!y*5wd@Q1Zp*Dt6x;Ll7r$XZOU6tH7-jhQ<sI8$5F{DZfT2~sr?@P;
z*2!Dtpjt?Y5~?6XN{oV}d3iQT5;>#rDV7j%Eu_lYmc&@)XM}X|g=F<tk?3O6xJ&st
z&64Lvy4Lk`X|!HPUe6vYC1+v?C|2dvfBgDR#lY2VC%FTY!fTPhtWt}V3Uf>>Nunz+
z`d!^vkF_x{--ETNGRrqch(d7;q*&-mK~+OEm8V$ngTZ$OZ*@xw)G`lI&EmD?6J$iP
z7gHpc)RG59<DHeJ8bvmslxQ+@6<FI<EgOYv9D*dIz=9!YBtRo)&J==En4QBZ$L4BF
z|6(uI|8yoR)8=+EORNg2$y6m~CTN-26&%KEdS%g8?w|NEh^yna?<6JAirI&<Fl;Tl
zIjAY?QO1j;3YhL+vy5lz4}XYPx47_z;-UDSAHUe}yRUBeSAX_PI$bFW<}ATNPFY@V
zI_`#^0N9ZD$;Y>RwkfFTiN)+l)q^s`RG5xWl;Z<AF8JM^N}*bfnI}qxB~@x}{DxGR
z=1B13+*ng2EMoQC<s{H#J0K*_LgZQo*f`IocQ~sEGHRtnNFigaZ~T%<$_cBrz`t#t
zndVF!BhwVgv54vfg~^JN3ME!?YikVj&O5Gm1E#`sn0WW*#568Y1riKPAS{7#6fXhm
zGz7)Nab|OW;??7UZs^&~LCA-56sz3U?WT~(N#JmW5(h|CIgLSP-)$_u&d_sdNrhSN
zIT;k1SiawW#3^M|h~g1oo!AGfme{8CC0!3>4dx_4VTvenH&sxkUcOEvmafWb>AJKg
zSIHsK3aoL@xhO5v#%O)<n+Gl<s2Ee5^Cz@+*lJssC&=ac*AT=+NIQ$tnVK@wc)pn5
zXxEIXil!9ZB#27p?W!d&)=SOX4vI{gf)tKn!Mci3J=Ixim3KBHk>|N4jPkNyY6H0T
zGgi`|ggBL42q{7p5?@N|xJ3wsR4Z8~s$Pm}s})kzx71JT&{QD=S46F&)P?9)kzY+w
zsc~Tp0p~RT=JPKx#_-{*7dV}{-gs{NmuMgtDNM$3;&3=nO5yQv;?I9`&vB{z?1L>I
zzq)0NFwc?w&=Eu7Zx17H&zZR-?70xEX0DouLtD!-4)o4Y73{hJXDT`;hQ4RFmGnM_
zJ^hD2AiR5^j%T)8$93M)8yS1fZM~{eni7Q8vp$cLK;(j@O6;{)z9}oprO|Cu(x&_n
zbSTD(A~quYItFV<fG%jY#v02zv65RW5<W%Mjd0rT`-P{%vsqCil<L(O5JBNs7M%J!
zkQNSC8+4YedYh&0R;{wmwX&4leEJ1uSr2Kcu+*hux?BpnZO5?h$ssc?3%V9uG~iVu
z?<R7Hl#P?0nP)m*$hEL63$=hz1s5F4vT%MnFrUVYLsjn#-nxr-07e*}OG@On7*fh8
zvErv_MG#6nMiutOu+fI7V78Xz6V@9#Yw+8i>O1CII2OfR<ZUoY^rW>EA<lV?-&nkD
zsen1tc}=kr2pU6Tnj^X@t~MRpU5D|SstVOqQq5GezMQK^e={X1P*NpNGpE<zu)lIF
zVPZGjqTL4HZ}_91{fyW5XHtx4XUMt|jKUeqU?As+(TbnmZTRWFkfsxJ3?waXYdJ0Q
zBAy-ykFUun+-@g~?--aLzafqfJf6<N$v{#D&*RM5iJE|z7{`&YEW{`scFBsGbL(^}
zRvT`Ho{w*~eDd;&*RSt|$JAaKk%(HvLOoHd*l>FLtm0Uf!1=T=&p}{l3REL;&ml35
zv*e05=uD(sp7AvCE*$y%%hxQ^LYxB2DX^SpPUD&T`vdcNVRaUnL*#Ux84n{=%{(rI
zDH8h@pb~*BNo_GVw=^_!%%pK)T3QAINXA5z!#+3HyIh1U)@dX2<8Nsk^7r+MA;`?t
zW?WUok|jh-$su$J(gBg5hA3T5u7Jj_$mnHDQRw>2;HGuBDW~Qwpahw-rZpilr(E}U
z*^5-$d{=Z<Vt0!wWSx+oB=C6%Q4;l(LYGEpTmtiW5>l77bbXI?;!vj%Yhp11k{?>u
zsNt$Fn|jHa7z5Kdk#hhHzUxp*qm31t@C%txU&xgjsLo<zl2MY8032_$jB8cfhvQm|
ziibZ9bu(H~kd(&U3JNj2(U*=?R{%NeOp|)@UNNAQ!mG+~)sfEEXbb6Y-|)rjdv3N{
zVvJ21S<zKwh-sRbi7a#A*Y6&9h>o|X$lxs>fAEs+cE^+=hhyPaV<0KZukQn2pJ$W~
zY?3(Z<%)Qj_@2piXft4X$9~hX^MYzreZ+S?yP+3zFuzA#8(cRK!;xv8IK_$M(~Rv0
zZgU3IZ#S|$=OF0!kQEy0d~so|^_jbL`TsI5P^y^=H3v(jt8fU4aIhprQP#iM&&P%m
zwFVk_zUp$8YHH0ga%*F7Idi4)UR`B#id4N|Z2zs+=y&rOu%23E{?QxPkO3hpEG03<
zNGgaVNUiC1`aP$)_bqNUw)1rR9=8>Dul4Xx^FneoGLE8zLYNn3SD1#3?=`-2gwUc*
zW2E<<uq5WwLJsoaj8=js*fxVr6^qOkz3-T^FxGQPXd~pBWy;LsA`7z^LhXnOina2<
zvM@NaRz#IK<xE@_zCO$xCOHTM7L*qXt+7K#x9h~}G)ZwYY`hqCD8(U0ra6iHX6Wd=
zA(KcYq6#ust!*Gv3wb%S>90uEQTqXPeNCK>JiLCx2cNve-@d@yevImGsQndx^z%D@
z{yRSwhP~2^%PcJ0c_dCJVv^LgPi{KY;Wr$<`6VG$)SZkRe)$)_WR3-s7kX#ezWkuU
zR|Qo9u0~STym>fdon_F4YaG|!vU7$MGfRayj-)V?myu|IDloWl4pm|6I);*o`<|C~
zd%E5+r9xF=DeawRxAhFYpd?+ABtfI)9;GESj%SX?nQ01GYv{b8RE@GipgNyNDHPU4
zt7Kd?pJ%>&IP&`8D5%CUaD1FO?`KY@Gsg#!0mfWNRsNJ*NpmK+#Bz>=IT5BPbJbJ{
zbL2b+LKLXgC`C?kDCAh$;+9Q2meL##<+Lj`p=yUiqjJ4)9KNqtNr{YXXt_vte$j2W
zLv!6oQ_3tsNWR`VN~+-7Vc1M}E;kP@{DWF0fe^DI4(*Ss0<~tD#VV~?<E~e|w$KI4
zWgfd02esx_Oz2DFBSb|BKq@Ig&7>6Y#XnnMU-(oq@{l-O`8n$$lGhx8A_qb(8I8Ej
z8Dmjx^x?gyZ-zuGx+)gLMNp5%AJ9!s2!e=OV};Ao?3QyW0$sQJZm5)|Adrt-+bO>m
zT^gl~xI0?VR@4@>!>!iJ)?55wc|8Yy^gAE%@%3xw7*H{?=`80sld}W@IirYS!8*kp
zGh?be&4E<l`uc``*zxW(6I11zhd?l%B}d+zX0V2w6XU~M2InbJQV+9}$ofzX&ToiU
zSERu+7^pchYQxz%Ot<0o<4@4W6Y7Y?qK)VH^u)V|Ctkd=T=$xjRh$sgwz_aEWdsSW
zt2fWY&C>i9r14X&V3iCLt-+8{H9~3oMO73n>76wTpCzY!*^hEcl7$oFC4WJ$1nkCZ
zsYP9KH7?}QioDN_zfuW}`w|Fh`c`Z7I~4=5Xh>j-%ug~x?$SjpM5j5l8I-(Wt94`B
zc2U}rOQK2$Hl}y<*8{^|y1O_=<}nak$V>LmkY3E0=^Pl3BmLmm?uQl}RSBgq+nKOr
z#?wMgLLxHOUS`kEXs*2H)i7YJ;kYb>QP}0Hfe1=5kCGsmOQv`IC6uNrbAkC%7AAN|
zf#V{2@vq;V80SEf%s1E#P!;DbT_?sFS+DGAT30I!`;dga-1`pa4cf>iMioM4%+E~W
z<GU{zx(!BKj>}9u9SO@qSQf_Pk&l1)3Ahc)^+Nio3FAA|b`N%k4+|!rQOiBGUW0Lz
z3ggook{gI~;Bfzrp}XVZVd3+?_$y9_kr#JYeDLv0TugW+Hl;e9$T_fN&-oO&8hW%f
zeC!o>Hv{|5@^lzEhd>G=gb5Qn2$fPKDbSeqj8q{T<bHX}cIP>TNJyC}MzTuuoo2I<
zupyts$fY8iX-J8Ao;aLNgb;bL*>HQ)5fE5$2@9vua6FHfl(153%aV_r#)WATD6_=E
z`Dx^^JM#E=;Pf~$PmvI_WQj+q%)=Bg-Z4$HNP*^nQHn55oKK^Kv}EbNiV}ER)iz(O
z&$8Eo%zHPTzHo5*&M<V2nktI#_f@J80A(}F7dh5Cv(M!jHO#p=l&ZDs5^|=@-77g&
zv8rAFU8(Ik<_l3zDbXpYvU<fqCV=Ts%Ai)g9_nHoRExSCuJUiR#7b8Kh#hV&O)?gu
zX!DJc=<n2g>EAzCk<Tm9%fhp&qLh?kpl#$b`j9XyZF>6cfbTstX<^rOc<(Xqp+hp_
zKxX;xlMjt<fm>~#QHE4SwyI)6<$^N0WjzRa&}vI<g$q@rZc44A6^WfkiSwfvMj3LD
z_dTY->2PLuHDGPW@4mbuslv^6;P%GzjfJN};<PLzB_zx#N5&W_O8oize#7?amY_Xv
zr^x+rVM*fM5~`wTO<Dq}h~wcoj1U)0&DdHXCzfSm#**`b%C|&oa5YeCqG(Mt2UMAO
z@zJOF?G_s<8#D8w?ih0B!%wcc+Zn!oCr*;W#xEBbUaR-w(e<-6F(8d5Qb<|TqT98`
zDupIdQ$cBi5;+}fkWmTZTc8n{<@&DYoWwJrKD%VDqxMpZ*b<{IoI4TO7?cv?XQ_g4
zC9T`~NZ)FWey3vKbY9R)VviGsvlTc_RU|>{Jd$F1Ch64XzpYd=%uwR^B{XT=#?$Qv
zY%k<4=Zr98V<DUu(h_CSYc1v+HIXkcVT(hqQejOf!jzb&ATyl8W$~0E+1@w3=h|6t
zmY6cWAfEDBBtE%HzQ7VfBR2^=G>3&+tQ-Ofnc<*_IrDHn@$TJ$@i4Nnp54x)jbb;*
zp{}&T7$pg3O76w#7PWO9oz+;aBmy^yYC_xQ3EpO@0&zcgPPInfeElVE`vT(~=aevg
z$BU~CFFv>vWJX&lU}`3pOtCx4=0j?Ki;ZXM<F7y&igm0EW2G&)4Z9D1$Z3|*#cw|U
zk{k=0VPLo4VK#fZt2?USV*3q!cp_QPeW~1RJG7~6wgb0Up6h)t<SQ!X6euy{=fEZ>
zT#nMo)Phk3gM@kHuuw|CIe2+Ha2OZH!^|-T7FAGIjB0YJa(|Vk>O!uOB`rLi4?LcZ
zC=I)v<@%<R`QlWV!@?;S;v{+HaxW1XPi=Gl#$k~z6yh8>9Y)TlktGC{m|3zIVU?;l
zj0@UZh?%qm;xrO+;rMvq@#)Mo3TdzcsaCue;jC6t2&h%dY!t)b*bauy3aYgk2Cg=i
z<6&+a!1wnu);@Q(wYIswzOdLYAzQ%eYWLj5bLpH)uH;mtK{d@UpjAqhbz|m&IINxe
z3a2V1GB7Lsg1kR9$#b}T02xhaZN>XuQbNS%plE7y>{D%>sal@ZFYlqBB|kSatCcub
zQ&N_7t|c(2AVddkq(HJ7y57l{K--IV0?_2eLQIRXc~z{)YiWEe+F_=RgxsvWhmwVP
zj_l2<eJ(lpn-Zg;0~ZumZk})ntSE03RnR;jj7J_`zvhRx9p}e){7?V*GtNuk>5#bX
z6vxhSRGI6S*BC9t%W00njK_d680A1)j*Iwrj7c*L9nM-}%=D!vSx>1MQ!2$Ma?O~k
znU;WKW|}9oapX4va-<9!wyuIIq?#pP;+SE(VfXQ;csH=sXSNtV8m_qe?1t;DWpL7j
zi8sSDsTHb_`mAZ3jb{@k*-ILnuMN&SI&1L;C`mkag1A$bDnxd1+)d497@ncO+RTmW
z#o>_lyae6J7xk8utALg<{ED=#JRGgW`$v~=2P=Lz)1hdvRc6hdHDYmzEV#L`xKm7c
z-@a|FZ6V3~WSzy-Le>@K4P7s}2D$~l^qY<}MZ&bSnQKKG)iTd3N>|A-$dwZ7g;bTI
zL}TW;5<{V;jH79l6`*7jc3M-=<S2wKmBD4`twhL<G4oV}{pqZ1LQ@EoQza|Ql*Mux
zrN~N*Vrt49&kL$(UT!xGMq-)ot~LzYt~snVb}5R`^0I|hYA^`0l9x<KnS~^zs1=ed
zkU}C(%f$dE<;2G?HayM~%QRt!E!Qt^dHLZr+ieHRpiNJ$id<T*w%&mmq=*?GaN}D_
z2~_2%XmV9(H-Oz>u3xeJ7(V~}OP*RsJa|J00qqBh-{QAB2B)Zx2Wm2euyDQWSn|yE
zzT=~pJNn);rbrzX*;nCd%nL3@`XUy-(v8UGIxjLyj(NfNaC6(!z0;(}%)8T(!?I(^
z3mulCD;Qbqu`I+880V39$44HY&g4?qZp8om5HiPkBur;?-|=!V47M5EC>ncO(^(09
zO1k2Wq0^cggg`ilj8U3>Z`k#ck8!neSPF5ONnvK0M@Fj%OW@(@iNk3mhb-fnA_vi~
zv)p)5$}A<x`$WYUxZZm{yzLphYBtcGjnzDz66sqm=)bR*HqYeYm1LV!V%1`ooG$Xh
zifYAxx^$>=m<fkK3ow8#vLDF)qML+HT-<ac60U2nvKAJq5e^+WH{NPR_yb5hw<;E;
zS{_9$tq^EXY?aM=h4_`y7wO=d%v$TJt_rPllOVM9_(dDBq*W4NBw<CePiP8G6$UGL
z8cIoc+FA^yQk%L&{$7ZuB5kWiOX1&;5~V;{jVq2-C2&dmTu}zqJ`eJ{saC|8hVHJ&
z6;1W!O2QdK)tbI@T=_lQPw%+x6yfy1%YEgek3Zt;H)rnePyFJGd$8BM@>_<DXFNU<
zvLY2l)*TpyF&^9XByEW`;<}DnE91Nnav>~%3OK8|+jVrhpku~Tc{)v;7pT3Ly*)+d
zr+4U@sm`%=8b#!0uh_W6t(xe|iT%}{toC%?@Zqkqz1gtcdj8Sx|Cr+6@te0#l#(vU
zz$!o0`Z-8QZr`dv5cP{Q;&|A5hgGGa)Z!%)b8d0PYt(R)RIesH&|D$~O%j6b%6F+O
zDd3X9dW*;s9aV|OqqgF*R!NaGEh1c%M2X+du<X0p5r;}i1v_Xu>u_fE<c>_!!m`Ac
z$E}{DJ*7mIMV}>QOk0cT9NrmhuDGPAN#<X!w<5ENGK*Vhb$ShH&`r}>gm`r!d^GKM
zSt6<wj51g)a~9_`o8FUh;r={x3X$V7Gfy*?!p6A^`KT1>&P{_Y&mnP2iKHwjdyzUx
z(v)GDW)8=h5)yaY4KKGFoHN{BZRvU^3IbZJX-Q1;%)Ce<TUM1&GV>CNK@38a1>;QX
z2E>pf#>6s*X6svcdbsDuuU_%N-Am@7XSRl`n}N;FP)or2fvSugASsgLf*BeEF^r_A
zuPMt3Wvp06lR!BNhN3*G-?RI`a&wxA>NR#4Fsjgb$1<I`(xCG~cz8n|pIEF%nE=*N
z%#7^}#%q#;pcHI7{H`a5z*$FfE$F5Mp;k&LWF@?Xg-TQv+gtjLxW}!<W(HD;gdBt*
zMaH06!bB;B)6;>&yC;rM6RH}VQ4B_*J4@;vTU)pr6gS(BZ71)iRSK&#y>|>&bJbgd
zRt#3-YDR?xQzCm;xp}$cZsX~^rnBv{8OfWtaR!$Xd7LQAf(;8hDo?r+bBmg-aOE_g
z?mE0t+@EJk5@khi3?KHEA6*Z0USqw-8ADp4NLo|<4=u^>&1H<ZM3(pT&q^t1w$J4~
zIA1C0%DuDgV3*LSQgibx$!OK=VznRy;$GNv{np`(1Wc(aalchbr@TgZ%l;%NP|nX?
z-bzKt&7jAaN>MAcD@Ir;MlmfbNxnvl&w)51Hb=N5dsrz#+{<m_Upux;zWlDLLC~gF
zR0<dpwz0w$YlLn?t=ibNw0oA@VWN}<Q!YqHrSV|^w2%uGtEIVW-=|)REmfNlmU>op
z)TR_ujnEo`bhdrhW3{4phMk7(pb5uEZakdtkEm$b@3)*EPsAAb(=UI?|L7n5K7-qk
zwdF7cinc7JQmtcud&73WBgZJ5gjyw4(R*~xbp1dr8C5fPTg#Tj&|6#yq|Ly?7&y+0
zr~s7ZaSp^XF^vJ+J6`Q9|LD$vE{x+_KJXu4Od)Ky%uA2nRj&3HO5kSGv)^{ugJnh6
z*U`Y58?ll^S97VH?ZxoPG>c)0i7st~8RS52%7a*%{=XHCTt-ELmTmfuDMcA4Dka4L
z@_b^7c<061y2<-%k@tAz5Uh5^>sX_HOH}bY6$5FCY@9(Afl*vz>Mo@)$9N&j$egUb
zv`F`-m6EEq#T~CNXIZkyv1-9vi`w;)y?qpdh*Rn^1Ei{!{ViVFs>-}iYYi^5um}UT
zwy{JbF~%zS*oX6(bILqTk{M9)BDn>+(iO{TsZ23mn6e>O>LR-0qzc+rypqrlf%+21
zr<v&-3Fnzf8|L0&eTyKr%|BdKmKcboayXqio@Yvm%u12w$lJ#=DI_+|v)god=df<|
z6i=<|kY`Y8<=_6|H+*oj<&)1G+HSc1=oJGqIt8*2GAPnxpvmOwDMd@Qd_Iy-Z%8%Z
zwXlf=Do>5DVo=&)ZQ;XDe!x%vFyqXYum0Vivc2B1{qSQ<?|6Fq4M9zG!;W|J!V`&;
z%1kI?EzCybIoTQdt1TZDaqa8-4%Jynpwn$utVHrC-VcS`*wfzG7LTnUD|}9vl2Iv2
z*H<gLWH~HjW>^CIocXACsNVA7VA&hP*1)j$+-y5u?FP1;$D3y|m{Q=a=H=A}t1{IX
zKDzF?9t^#KyNze~_=dq+u6s}KC52A9v%;Jc&MSObFfn6uMA=B{HAiQ7UB$RXDcE$D
zol_j5zHG*&XdIO*qxrD!=z52>B9%I)jLL|W^7r`CtvM)42v?@5iyBtXniCH#wa{pc
zX{Jj`EXyrmN=BDTY{i4oZH{g2WdpR<T@JZM=$6r;Rstr;6hQVOZJMKDZu8vcj#-KX
zn-$@tlr~5fQ=rBKZ{J&z5RCjcG$wf?Ppjq!5_7?7F<BBaVUc%uRS1|iDo7J%d8DM>
zGdf~5%1RWjY9!3kGCEd;L2W37*1~(Jr8LLE+P=PuO>WySwKc3nlxO9WYFrK@38oFT
zEu}!p#GFYbP%HG_gEMFh);s2u+1~Eh>>W>!Bh$MRrdB??-taGe^_u_g-~K88^!I*>
zDv6*Cqp}=p#`GID`zr}FqhgV>qjRE-_bCe!lQLJGL*<42VAv?fmGPvU*!Gq$?+b%b
zSnGMUbv$IleGv1Vp&NLyvv{R=dydq{Zy2@%Zhy<1j=Xt#BF_hLoqcKOl*TEDG?Yq-
zxzlQyzg}|SS9vB{AbVW}jVe`w_)3yU`;;SbS;Tg{g{^5-nsL^1q2GJYe&}Hlw5q+r
zlr~zVs`l&~)05}VRGV`tE6KLK>nlI%KXrYlVql&k+XC9Pm!`=cmJpd+zO`#JmD)hD
z74FWhYLU@M#Y7`SL4<Q4(1@nrZm7mm%}gG}Ax~38LS&7=UPF)T-8#wwD$^7s`A(=X
ztg?<w<rE`t&J#oKI86(OIU%)krt`A-gj!g#B4$lY8LME<&9_@wVuF$a#aB8fX>>Ui
zmIdZH6XwhV6P-6WZzxDgnr(B-R3hg&@%C}#{&A$Ya*)mEg?Eo<sA6f{cj7OA5@Pe}
zovzl+w89t1k$?B=w|x4^Pw>{r%)KncaK>BL>@q>Q4s8G{kethjI6un#Ogj=9QyO_n
z`Zhshaw4aM?>m0)AN(HYQ$){C*nY=w_aS)8&C4w%XTFSyC#`s_na5H%mCQtDt_fp2
z1x2qtAKYG{Lgd;x1~0r3RjVX$Rq67K23w%hiJdiE8O`2E_pVcBD49K(4P4ykbOqNI
z?AAE8w^#hY$zgD{-(tO$%q(l^d(U8HX1~ssg~2bC_4Te}V>4avd2zjAYrva|+4a2G
zb~vZ$oWZ|Oe9M__ZUtls*eV(JgV&_q&{=3^q|L&#uqe%#R-01k3{0(bO5sXnz~HMU
zBg9gbo|(<x*Xxo4V8n0ZA|=$i6;}1a);CtTHrCp<-H9YnBl7{<A^{5WJ=cW8=S%|K
z4)}$dLK{CAR$qadvyhzsJwl#oO!2Z33)OpF_<K-&W8R?<X{7cRWzEIGZ*_cIi-9$U
zc9oye=DS8IsF7L(1(kZBQo@luKT6EA?y*()L+d_JO3N{6Gi;sA^Ji?4%sBzg%&0aO
zI#nyK8oVDa#I&_=sqe$yMog59F(Pd&1(YCj;v!keDsg=?uqiWNK0fhy|43IObs12~
zaJqk@nhs|*yWX=@czb-}(+_U>>fwR6GpMd-bG66ygEWm=(Qi78R`gagPZMZyJlt+P
zDlVv;33I}GSaLx-&G|H=Rpn~8rPmdwG@mHTQ*wNB7F*EM6xel+t4^WDdk$X@)E6Cx
zRQb)D6R%w5)s^MdaAp|x7^O+P_nz`x<f+vcW91BRsjDu3BlkU^lDumg)$I6%L#536
zU)10F?>cAMbRK0DODrrQ2-%)hp{LvE>OFE{c_ysZX5Mu{e3WY3mH+hh-8cg&fwi(R
zT?jDZpcGQ3YK`x_xbDRWx!|2tw;?3vv^t9+QMGC&g^X@+o$EZx$O}=GVxh8(3tC&c
z!7*$+e(1qj@a<)7Au`jHFWCpWN_DA>qNu_$N2ZkUK8SQtYi#c&uiPl2c4Dmyh$daq
z>B3U6zGA!(Lu!^7-K->^KIK9PMWkau%#~?LB8^IsM5VJvx*(-UG4Swo<l%7UdN7bQ
z4-XUPgFJB8X`C16GuPEIvH4!N`v?$9;V*vkHUI374s7=ZMWUt?H3z1YS&j!(ari5X
zH=3#)xQ?0<kYIR0)=nzdIHA%BO(iWOq?uYGR%=p@cm>f9T>ap8@w*M1oyUYHbXd4Y
zF=|8gmPJc!@stY*VRoCAO<+r<H=2Iavp1HF6H94|_)}D+a9RvcJe9#&hGJNhVyH;L
zkSjJ-HYl#OZFC)knXcr3*j+j9`YXx@=`xE8F-RfWMp`v2%mk63A+#Tna)DE@^;?GC
z(K|&-5o$qOOEpHk@tYf<$t@EiB}$6zFp145N3`hFN=mtnl;j|YSs=o_Q+&Mj<RW_V
z54RoH&SEm?R8b{K^!1X&!&v<X7UcKn(yfa%oe^1HE%IJ_$!teGj~0~{>DD^0c1lzM
zsz!847&2G`*5SGi+x5-LIj{UT>Dp_SgSLP&R@PunPz|N1)c_6yLYjN8@smweU;crT
zFOh@YHB01LutrFFXi*;2ELO(y{oX%Jo)sZ$46Vr0qKrLi5u&jcN+HXMVmCx1Bt_dk
zBV@5uSI^l5NW;kbB^9}slHy4t2#YEy=cZaH8m-<dVBWhf_a9n=rBOzqlDZHIz0vGU
zf%~s{c>9j~#}jFhG3d(|*Thiy^ZR#vF^cKe(0Se-pGY<FZd{n@@K)yEF$#gxIm>q2
zLyUM0s_QtNPvnw09Y$PE^t$p`HD5kFvG<N+2#j+Uw?sAbs#E;_XRoj|a33AYB$iY;
z&zZPHc0-5Fi8uFu!}+-2U;KzESpMw(o<FEt_II9d4ieI~!tAB-Mcyw8YLV_gwsc0-
z*0)g_qh!28$(Y7hvCiRq5zF8#dqZ6*ZSNHvS}S~KDM?J#D$+#1@03}q{A|6-JSlCz
zd;X(-3&r?5Kn9FfXsT@7jVzN(Wo}w(l)`w6>m9V%PZ<juAxecP<`~XiW*lJ=4W=<o
zcUwKSa~K1*lN|C&#qS5MZg#lNv!sIA4q_0J6S0oOJW{fvBw@>1D>K7Vq(cv}GBpyI
zb<%b8n?a%{+vtPlS+9y5W?@MfTj;fXK9wr5#3luiZYC#59doKkm#!t;LR*XR5?Uf`
zP6-y8=QAP9B;PTn%JDRl7D*TLeFp|Yip<N>NLN~N9vUX8lytUF<C))l^*R69PU23(
zctYErZnx!ddg9Oj)t~Z*fBa8)`SL@mYJThUNM2@?b|?%fi5{5gC?kY4OZX904oC$M
zBrvKD3?$s~kutvFO@_DK3#mv$qP7`nTp~F|R7$uk9g6{aZP{4MMmwCAh-x$xQ?V<A
zNyP)Pf=U&=6k-soUUDXeOpOJnBpXAmKJ`_)0p+3kibnFlrASo5*GTE2&5cQ<JF>XA
z91OWiK+J008e325w2Y)^jNYgvTVmDmlJh03rM5tuk|N_AnL;BEBBQa+TCRJ`-PZHr
z&XH39W7u_$>)y2hoHd0`HHa}^GSk1$*J>B6twiD~WAWZY^UYt8fwkkwDT^{m0X0!;
zgl6JnwG!yu3IUHQ1g#aKjWMH3lsUZC@~$=3w@C3KMP;bZ1Teco?o|%ds?_DcT<7hV
z?X@<=fsqtBZLRPINUi8@_OY!pQAM=s3ZCcwB{$QVh}ddtyl)Dv90k=t()E041ynT`
zmcPU!%b25Tv3M2?s+pVd@7q|nL#*VMcV8RdpcPMQMVZ&}O;d!`XGl7if-;4xZO6`5
zrs)YCBQY1gejF(pls@n@CI0$x;mdPo4v|~uC{=Tq60YybQE|1u;%a{-<jqo9!a^N9
zgEi!qVp@y1_!%vP<8eN7nhH5-4r74MlWX9sw+Fs_jQsJZS3Dj^e&^$Be)XGsw%vwQ
z5>pgS=bPokhc9lZ^F8NZ%|v%CQS>LxTnul=${2+Ql1sBttIx>N8W*e(iMfnoO3RCA
z_n3^i6lY3IqkM<0impNli+t~f7OkieNHJg<+N+k-I{VUi1MRuD_fHBlE}v@+yIM((
zwnZApZ#iQ8Zl*)X49{u>Mcm#p^^)nH+l!>Mc>1Rzq(I{gHW!HnRnQF`y0g?$E~G=y
z)J1G!)f`1GvC4*&7!YB!&~B_8@<kU=hFBHT6eyvY{y2M?H_O~Fw_-)I(oxdo3yDwQ
z7?l%Ckd9I2fHJ;_bD;AU5PpKyjhH029jf@r3#-`BJCT)GV@ah@b0+6RC=2IhVqV1e
zewi0SLHzbh!Fx|<9ZJ(8cq1VSYc(Z-`s7`!*w)H?^X`e?eenh1e3JZvz2cj%1Aq4U
zBkx{6@Z#0qa@85|17??q%MleulG2Q)6Yt)<X5Ycx%|^D4l0=r1lf*|iYfNt<m<;m?
zJ9wVl4Zk|vQ<oFgIc$u$m@#32C1R$)zQB!=NKd1rEAO<x#8z2hKbN+Mkgl$jg31|{
z5@`-BV_;eYcAsP7I4*>7rtSjDbutqxBKe{DfY*{K>Y0#cti5zI!qgR6nX&Tf$I_xd
zQ^8wJ=MCOF*8B&(HiwcdBpn%*5S#ob%V%z@QVId3$x&1zA?4;{BlFtcD(-eY`%Ne0
z&^74A8V1*x!#RsP<|Wwa`|mR#bI;mX?E?Di<_3x~Qk=DOODip*JY?2dKq(n1p%LDY
zl1`s>lWYbs+Oz<nDv5HX`E=;kVHFaoK&Va9M|0jYhAdFTv%b3)D9>t!m6-K@*Hx<c
zs$8ZB3aTpYeF4>olR0HF67=ORl1lkpT2;L7>6J)ot!@nQDu!FD(x6b$86jl120^Q5
zl}3#5A{hj6v&%W5nvz9S7**DFPNc?1lHx}!(+jmVWffhG9R5|To-bn-RZ%%e7EBCS
zz+1->B4e_=KEuP5IX3SOYc1olu<v&4H(REZ*l(}6zPS;1$1Dzy-a6)KX0z?l3bwtZ
zx=!Z_l*4LI?K_UAGn<V_+aJe;Uw?DYVa{|;vD@?_xm+X{q`Ok=dd(EYW%b?tBaXs{
z`vJ}~?;c*`Kll;0@A>8&u%nEwt@qT}GAc?DWyLBzxfDrF(wXZD&|ojVRdO&>wb;i~
z@LJBVP!eg0%@0CQ@%M{1jVRx&k?{fu0o^EtGNz^Rt`L}vD%a7Jx)}adtd$scmhi3C
z=yy{Slx75?WnraSNhLF-NYbFaB_Xi6QA!T)Dhe=d49044lQW906$>blVwGCSv651u
zsw5qILl2Zz%u7O#0qwGQ4yc8p@6gJyoCCSRsah$#H9|&g(aNcHh6<obIz}U%pLT}M
z7;28>lrYxH3{qG(RK!V6DUIzd&Y26WQ&MgLGqMwik&tHa9;G3s%CrcKG{%XC<49bB
z6bJxOv79m`7F4Hj*0AXuy)|gHhNVb1%6l(jXJF2SfAN=p%|};`tMQ(0*b*1`-~HGB
zlF#Q#91}nL`43pm_w@FabgpS3jgKrT^VJt$^5=i@=iJ@y`T6htm=9j;z;@JSf*c!J
zrLdfjzyX&NMOh{`{OT#NjD=0Nu#uStu_!jF@bzh-9s*sc>{XMFwqih61-W(DRfuF_
zS=pQwTef>{q#J|c<>2^e=UA#@nG*A1Vjd=nm*PkpiBio-^}H^yHAz!zgL4|^B`#EH
zf&88CF7I{98S6wUHF(J-4=q2z8p(fH;ks%JODvLxm=nFxSZz_I$i}WUU>GaRITM;?
zsIeAQ7R1Xy?=00dtJ4N=uU4?nm49OkUplkz_w{}UW3+Vj)>=wcO@sYB+cef-P;HLh
zV02_3vA&Z8$WogxMV0WP+z5PXMHpzC>FQc2$X{({Jyp@Hi=TsrWQj>9m9r=wF5`>$
zI0Tp9sT7b%5<vpmL8{sz-+pm9I9gP1QA&^mE5)++mQF#hE1fA#RwnEx)0z+4)J5yO
zVpU+|l;!)TOqdsPO3!4?lrB-l7t<V?Inip(%u1wmmuHl-z}}K>*ybl%B6VfNqgv#!
zA$!9xc!rSaHdhp-cykE6IWIy4ud;V_7&_;v7>qOQZ(dM}A|^Rd4T}AyqZ5-zyoO#I
zHdi;iBU6g!@$r$}(9>Pt;#?t=%I9A{a-1Sd%p61HFTQ!Ayl{Mcx1$34?SPMwX)J87
z_B@?N9*z^<fJ%w;+b`MPeZ~i$ej;8XSjo6B23-XeGHMMJYn%#XFEQHC*(|;=#^7{A
z2io3Qb8fdzM&U|pY%?IrxsaA5E|;xzxU7+eL`n2@W6ji+2rD+)LY94&eU@6}EJ9u0
zSvg+iEd4gucPa*4RqxHNWFrePl50jclN^w&=ekmURtUvgtdf~~N%B|KwEWtbmK*;p
zji|YzyR|CHJRQywax?fwqEMvsg4Tv*T&OWiYL{Nqy(~p3ViFd=+`x7-$<*^g%oW`^
z@D8I2wsV47G}1+h<pR5u0;XzIOUf#_kU~r0tChhTZnhoQyPn?5Eax;WoMR!W%(%>q
z=MfV#JFPLbayTt4V`yd}!mi(X!^RpMN|>ls;KjVE8~|M+PQ#KF|HCi8;=lai9oagJ
z$ynR*Y6zTy<5z$CxBT>@J-ZKo7wdcKxssQW5CUI)@eTj|zy4d^o+B^*+kegf@_+V+
z{Lvr%z7SG!W)5dGmBZUN#M3=hRbKq=Kjt(Srt_IiuJ}}OHDGFCUtk**hjZZV)5LOK
z;706EON+CX&P=-G++fdjfvua+%4{}hZij(C_~3>YonwrdeQybK;(R=l`$SdXe22?4
zNmVAND6p&&0~ceoEQUs2tJ<K96pFGFK3jVZonzC>&K|NDgE%8yqtdO}i-kE!;W%j6
zScf$-7NeQWq?`yLUOKhf5Cvl_1}&q8oG-DyK=aUVMmep6PE}a*>*oJ`zDg|(nOTEy
z?8PlD<;*mXQv3o?@y=m%#Z(!y=_=TQUY-2fp_{`3A+SpHK+rX1Y9n1fJN!vD!Aj_R
zmVm8>NT%tu*P?Wtzpnc2)u-U{JC$-t)m-1<l$-T5lBg-%78w=A#vjQjEE2w^(BgMu
z6iz9OuEH``&$@N1^@W7iTuXl2b#N`RQYMU(jBB+Z5?nraLs+WpT{XAPk^rkfRKF_4
zgH?@4Y{5twD%lD}SnbW&1_T}MtU|?^JfCs8Vx6HV#~d?Hx$vuT;oYgAjmLY#$1h&s
ze9ux9FJ8T(7|Ur|1YJ&*q4TJm*=%}ZSVTH%G$~AMydbc4zu|5xMs3p)`Ql-e&bUI(
ziQ{?X-!2jIjvu`Y%&{<R_Pp?p^ZCf8@6grq`EP#1{havan<Ia9II!2R_@Dgqe?)(4
zdGmC@_k$FjSW%XwB2UtqOQ)^?XEeR-gcqi1CPL}5<}0v5i?ucdfwfKTB2JMpX8e2W
zTZ(~<QX(UjQFoMHP+({(dWCbIugWDr>EeOVlvw3kju^jFF<=$A7W%WMH{~eNkuerh
zO|rl!n65$Iu~K5B$Kkxi8d+@Tae8LWSDGuCx^zN0Rg{(!GQ~`t0ySh*mKU?+!Z}2&
zHsZoHii4kTMTs?rxn|~+FVNW~=)*L&e~fseDMm|Y=sd3LFfH%AgoG(lti+NfdbY@X
zFNVl+o=9`VDER1h%a1<T^YZozs}-RX#`DBeUI;Z&bHL}s#ydKrS;mE@5Gkp$M)6u*
zv8u7>3sj1j29?|#9OX6KZ8sPr)#P@+=lZjsvAx<sRs6Gms`%gk-~R<y)^dBbVRQ2V
zX7`aG8>*7y!WVz_7yLi|+keHIbL48{n3v4|<^TEr;o}zr+p8@lOyrOlmx*OMa(aB=
z)d#P*y1C``7YAG@=p@co)kxq|ULq<>cK6p0GxrZi?yh&x4`|gUX;~IYF3$&^Y^wFb
zi%`wcQlZ%IHhiiSFLpgkkb~7bLkLkeX=BI*q7fU|l2@!&2y4#!OT4H?GBlcefcHCX
zYxE_EB^T*_=4oM$QDmOYY^2sIw%9RYOBREtHj2tc((g)6?ZD4?qj|CIIEKQ!L^=GC
zj&^B=J5p?#rZ_2FVydgmc5_o5@GTeg-`8tps~cl*);DN7LKH{AI4=TsA_$}e@6_g=
zS6U%lWn;_{Z7iMZSYjZBAe*&T7z4U$Vb3?1cAcv#Rznw!!QhN<J_Y9Dj#%3PxL&0Z
zqxD+2m1lGV+J3AmO4HWz3K3T{2}dCp#Tt^dat(yT(7s10KtY>`s<}BDnl=V`ep5^P
zS*vm&rP%yf8X1u`ib+Kz#!bl}2xn>UDy%b&+QNfMDX1jKpR#hVbLLr|i6CwjIoPa`
zaTjqxm4bDS-C$UX;fuq>-`t;hbFN?|L!k2;ym9O|TXwhCRKMl*<B2#+jz>Ru`c61X
zIRhngv$MS1ZHOUJ98e0n*KEBbWwBYl+FL&P5Z=uJ<1}$j#8P=$Du4ZUV(UCw8}9CI
zFOkgKPi~k};>&No;+v`R#lwM@Zy)&L>#zCb!&g#-8p#z<HG|UR=J`Msj-0BRipW8|
z^ElIF(#X1wVUe{A&&8rZO3(^A;TH<0WVI`otC32zAY)dqJ`c6+i=xbHgKYW!T3r+{
zD3Lgp_Wmn`-}Y^;?^X<GBXeIZ<~vmxm;)(Ec6-VZV?0WGY9kXG4S9*gl(1TEv?^7c
z-9iQ<tk|{Y)eVZPtR1l|mYOm#6;3H*y%Ve05Q$ahakXW?TVtB_uOfy-YO_eCl@J%(
zthhATy`)TPa-1auveoo0AVjxHu(rc0E}_lso2keWAO^gKyIs$#>w&9nM+|{^iG*b)
zr3IrhTMIXxXWM(MQaqH(Yc&aBQ3<KYYK7Gm*XGi-N(O@02JLlY5G$^(w`_+FZ7o0i
z@H2L|A5pbvv;W|CUh=>CU;LNkl=<U-^mF>VpP+Ylq`06JxNbmYDRLf9iDjI)-WvY+
zAO8_$K60L)Kso#{2<#d%x+u;MkC<iT_Ik&czmAl7!7D|zl1dgrp(yEyo~FPz4-+qM
zPM8-KCk`s4Bqk{-$;-cPz9LtWe<tU(ES_4bL^&Hp*E<;@@$A5+P?Qt}(o|w~x^A#a
zDPiI3HkxXSvBX4e7Ov~pS!;0K%OSlaatxFt+=-<Yk}()<FOa?}Euh5;th}V$tYKw7
z>rKUb%iXSHS~BmBf%_U{ir9>!YLVYrih))TH3NPrk`-u8D3v8OPQv&2x^QfyKc!Yd
z{HUV7sVQAL9kT|$0JRoU3e=Ktx>9mv&L_-pjqx3|DG^FY)GA~=n#!rR?tHZx7JKx9
zbi>aby;8s3@vAmSFG-$9$<IojRTX01M+?^9N;kXlQc${JRl$<UNeF04i43x=W5$wD
zB@1~s&g3*>YyxMVq2s!OPNg9R;)zhJa7tF0q_g5pVXPJ3`%=-ZgA^e8*|=yOp^3Ec
zee&9B{Vc_ZYAJm=X9~~FR8fZ4PXSjaymIu{FR^Oo-ErpiR5?VA^&4W&xNeK<dam|c
ze&^E{xP8yWmbp~U%M-O~a*j}t@oGtY^x}%&z1yPW!mMDk_vjp0rkSr#na|%X9KytS
zN__S7Knx4R^-FBGkft+lmc+Yr;^w1k=2WoOasA>2<N3t?`j+3_Z29su^6;w%vUA*j
z{R{q^|N7Vb-p_x^Pk;Ys?7h1fJ*~-;r8TP}Sr?95s)}qaMN9D+LNk?{=aw~M@y+vM
zm9nm$1<qKOl>~ixgQQ+4YvEA1Ncd|JB%sg&ifSX|Rx$BgN7Yg_Kch$dma5=CetoxM
zpn_A1&PlZ9nnk~ag}g*4iQG&r)X)n@U^U~2rIBb5VpI|%Ae0hEr#4rWZckkIa%_-G
zn=MxbIc1r(z@-98F`|+LJ~*T3v_>`4jT9x$Q(H08Sn~{GERe<1K9t%H@PH4QoMGt2
zB@Q5!ObHPclPnZ6PAQ5mq8E)^LQ1UBBnm18hMMS{V&glmJIBqgXXraoSxgznfm)wg
z%HA1#n^9Axuvsg2qno3YR$TA*Y&RWUx8VmL{Q$kaC1p(^qHE-*A9}WTpVHrain{)c
zYCUB<kYm91dw%@W-{*h6I9~qsH#nX7=YRMaKlx;fEsE~yCB|xsDhwggU%lh}<~O)J
zQkO?Qc=a>>RKq+kXy-1dO9f2R$i5pB-y9<!KTi10(>KSi+N$7`TPL)V{@RAO&L~s+
z{V%iDCbJY5IqCFPa-S4PG(=#nv<p>fib^Cvn}RJB<7LCV_{g)8ex&%zsS;8|<%CKR
z3(Ym|Jvbjp!W^7!Mp}gsbBpK|k`M~I&R}%MG(`;SjJD8OP3J6~bF7-+H4{CTit#d%
z$VH0Zr6~#4IMwg#l~X{1UvxW`WQ52%P*hy}_mzUM%tBPoff@tUgw~?kk#L`cQJQQ$
zw&a#UAiN!2Q#<_Ru-2+^FLI%->4CtS_E!~@#t)DY<+Ga{4b_6@dHVX_UP6iBGJ}`B
zx3&Yd*$Z1m&5EiThsqeVMhcf2Q6<V}RvCqaq6q^<5QWmlSgJOnSgq8OP}COx4c;13
z*GXemDyb%HGZ9(~Y<Q1#uG;9dHD9WY(^RvVUk&BtKxycLlHxZmiNg~4WZ%)>yyD^f
zmM=~dfAcQz&0?soBZo+?3gbMRq37clTRy$*39jRtFAtm!BT6aekQ#B<;jAWy30E|#
zSgh9!ro$<C^YF-<r!)WhixXempQ$L^)g8l+euh5$Yhp}nUfeO=S3;aP6~*hPBUg8~
zoTi0gzenYPh+=<x%g=uQAMod2JYvt45B}MI1z-J=ufF(Oe)Ky(ZuvT{?NwRi&yD16
zt0)Z$VxAXMlG%KPut;=zDg|YY+%LtY_|qUVdE1e@p#?qFXQ#-^%hZT~6s*?bdT25+
zG?hv!Ey$?gtZfdZsU7;y^3QdXGunKsHTs>3f!-RdeJ{#WwGBLH!FNRLOPK@n>5S4n
z+GxBnG8a!`r+E2gTIG^x6*xj10-EMa3_?m0*RiUR?yg8>E-!17rB!j>vPR>KCY2<4
z&s7$)G1C2Mt;jK6pzG44fdDFKLdcYuvDZ7g%|Nb+5+XGOY6^Ix@J8W`%+PFa8AIi9
zOgx+e&WA>ROI%s#z2j=raWxDK+X35omXNUeBqXNxQaNkb_8pz?u&$FZjy5|~r1pl$
z^YvEZqi?Tnx!&KQH`mlq$jgy_cY`VsHM~IYKO%P>q)0p;Fe#%=$8i0M|KgwS`2C-}
zr7Q<b36%Ln|MAao*Plvv8fNs~(B1uz-K$TjPhWsGY&*w~fA~ZG&HbMcQlv9d$XRcB
zvF-Tjs~x*@U_V%Lg=vn|oSR#ua4!TBs~FRuv05+O3#iS<e`W6q|D-hMH5yx1;AY*J
zxa&sUUepGw<seqRB5jqpXsku6XZCJc@2hGa03qR3VPg$TE)*q(F*ysD!8N9PDTO5o
zxhkgxrNFllkXIGRvf(%AM{95f`p&Q!JY8=|%~?%8vc_wP=0;0#nX)9PX$8)Hzl(uz
zK9P+UK85dawii;dcQ|W*tASAn>>X34rUji7TIHr35X;=WOxSJ_q8Dv9t}<`Gr2833
zwWOg)0uVHUmXRV5YC{DY8nDVhwE-nvm&C`m&aIU9;8z)0t_D4YLJGnTH&wq!AZ&$0
zUdJW%S)r06td4YAX8occH_+5d?;FaLws+Qb4cX^C(6rBia~LIRjCJgEVXW82K41G$
zWg6j7JvSZtnZpv}vx!nIg`C0E%*$KH-K!nbn~8VR!q;Qwiy7wH3LfXlM$rvDSA*vV
zFSdMeJ@9TRSXJ3>2F5rNV#I1ojFH1BaJ>~@|IT@whSRd}=IOvUr@${C1E0Tr<Zyc8
zfBc94nEdJQ5Q?IH^qOCP{RP$!+<y9F9)JC7{^W}{9PjqLcsU~R0E%wAr<Q_Mil6-8
zGybzb{5hZhzy1~Jul@%<+6~-qd({7rx<7liB+Je_vESN*59gda#gs!;)`S8;fK$t)
z7DcirYj$gRkY4mh>Pb&B$@)Pi>y|jFjos`f2?AgN1+c0zD>EZABJRCsaQ8jwVJ-I?
z1yDrh1I?<uEHWxHZk&6rpY40E^{sEHnalA=j5i?0qbNmVpS>ov2EQ^wDy3D;xoOaF
zbi<C$>DM}^Hqh!cSZO4Dh;*T&4I^{Olw~62WgDYy4Ss8;a&ZJ-2|iOq_X-^oRg=tG
zU|@FIm=%?eDnKCG0Ehno#ef&n_tuKIhPSy=@PXik)b!6<$<#9}wGldF=VRRHtT3%d
zr7n6PCz@5R8rut!tFmEp)<U3GqzZUcY2b`$S)oZ&XEKD|d2<n2<uDnKI^KB{F0QR@
z$z$F*)0{V``k)o(Ir~X@3@D7(2i=uB*BFB%05v+?;>gbOxi5u$RTZ}Lk*gr=$H>qJ
zhR_jR#G!uawKj5A-)JwLq5{#_<f3RnuB#_{#TgM3z^BAGjC^?goPKvh4I`ya%oBv+
z8s9SY@FDK-44NaiLW_IK^ne)upc@CS%bxQWugRB3;*0OVa7T4Y(9AIsVuuf&p&yXL
z0e^Xi%V&Q4Ti@sL{vD?;zt$P8Q>mxFy&`(g_omF?1LF`BKAVyqxVGarr@jqFx>^_i
zCM8|RI$jeMORHFv%GTAd#6Tw+E8FS{g922AjN4OJ2hg}cErbxa0x(yd1JX>I)|9vt
zeE`WUwXw9iaZgHX&^lVwjaqCr7rmqJ0uiVsnN(5>*XRR7*QwmqN9_#JQHc6dC=tzj
zL#`Z5vxaL8(ed(X09X0qi*NI^{(4<GEtI8FTcM_z)+OS2g>Rj~&vu}*0j6pqWXgGk
zKDB#yBBYv`9$zD|rxpRK5)Q@TC=dd;pb4I~emsq(JQ4e<b^j|Ov04#}bGjdSw<#Vr
zmi;;gSmk=o>HZfYsa0B^kyLP{7#gNz*;GY^Gd2qrsyfu6WZ+<kg%3Th+tI>Em7q|+
zel@P?UX{z!Jy{FQY_5dCd#clUw&5XWjO3NqQZ#yfH6iM{5m3{$QWOOfQS0N<s;V6%
z>cN#2!}e4niDib%G;>@skExOiEIEUV`kaXJW?t+9`(Ai>JQBjd)e!j8uii3E3yz}5
zoWQ&k7N59(IMKDt7=$ScfAZy-pWRP<`qhc?>N!3Hy5pymKl(YBxzJ{K_1nM4AdwOy
zw|VC4pPzVrocNQ^zu<!xcYKtE-PILCADMF_jsySMfBe7V7jGVT^ZIMXZse2S{hK^K
z9=S1ct2C38dp#z-_w;?VJz6#-b0s}X*XjCoYrv4nP-{W*Y{bF>vBNi&T9;*FnodeW
z*1rs^Y!)SJ4$)JG2+jOa#FB5(ae(q&JXMq&zd97Ia_Dc$Hu`pI0`KU8BY5b%9=2;m
z+k5|Hi`X=HC|?$pwiHca3L<n4%?!6{Sv=$>B!DI=HA6{0K^#RqNWpXz7~j>3iB*TG
zDuy8hcG1xX$Vm-`G*(D$fqgC78I+8eDH37<rLsDDK`UAvmb@s06@*<7VsP|U47dtq
zv#f+ADZ#K-LB!*oLc;sLBSy7VRq0b9U1lDSGv`^^)84+f!i`0VD%mzdBaa0u83ouA
zc(K3c>gpMY&ZL5KRxt`oxMO$yisAyxoHT!+W`#maQ9pkw!{G|oJ_ZTMcuNa|MqWxm
z;)v^aYJ==O!F6~kv^sOx4gBbbpYZ9Dn5W4Ob-|C0;3LWI?`nm)rgJq>vYhj>Jy5_F
z&%0K$-gqP{`6e{w8B`08u_jIv+nkyP<@)3jrmBI}``^T6YrxETL7Y?Kl@5|^yA?$i
zl6s0`mbk4|QZ7tOW{n3eEh`(ndG$yDq7$OYiAv7YR>-+A*Gk_9u8Yb}*G<ovSBG<^
zb9bhEKr4GsqxXU5`$+HieD~V(Pyg`SJgvW8S8$53V8kjd_T1KSuaG-f_ikeqFU~r1
zMTA1IWHy9a)EA_#aCZ<24hg+kx7(c7>nH3*M+mi9hhH;J0$$WtLhXr*LBNWDuF}Ib
zl;(X>9k6Zq&Iz$OWs<d^#i(^?%}_}FpGs;g4a$PFjHBXBJ>a}Y;y`suXssB#yjs3X
zQNtiu5reg$Uf&O<N>OI{nzZNLnrNtg1SMxhQKKOvR;;hDwd4X+M1ns5;%PM@KC$;|
zrJj<anoTsHridcHiv>gV!+@tT1SKBscLQHOE`0Is%<FfL#D1XfJG>7(94CSlLcZ|w
z>IH8eFFal{fA;x3UtJ0h^F?_jyB$|Q_-lNfj)b<5M<EdCuCBPbevTgw{NnhI*bjJ@
z`TYLKPk!+^IXAldGjZ5)cX!Ln;fkxfJ^$zb<Nt~O=Rf>sOv5#4nFwChX*CtZ%VsJi
zHqLeL9rX4gu-$mxBdPz&dIs?USBtVYI)@J(L)X*h&P=P;VnFv>S=SdZ4&7$hH7^V6
z#IT9_@3g8so7@bkFgfQ+G;RMO)C9Xu3xVhbFZHRbua(Zc`g7CSy)><!eRT@+(dc5J
zdZ%9U2r&j??<qiP+F^B3p}Qp&=5u4Z2x&1$%Y)$Xwpn5oP%t=W2#);_*!PYpE9C6a
z&EA(nU1n0%h2Dx53nFEM=R}Nfkuy0#UJ@54T<top4uM@%uld2LtKLFoYWn)QnO=PD
zfKk$VjA}sSoY~>#%zU1BJk6YEbD3*Tgg2R!Mi_g|td1gx1VVetOuy?#KD>Lu*hdPP
z+5)L2ybq*!#qQ;gDAjR%cvRLs6?b{mg2)BhtbrrlZbx_Z5`0gQNNGY!Gd^^5p~L$Q
z*K~8lM<k94b%u_syJuWKd&|4mUzrR`&17QKjy+}6kO$kSooL}DTKv4f`K&=P!YbDa
z#*%ef1eUfvRDgX}MoKJN6s%-gxBl+dgT_dEwQO?H{Qm5>VX`$npq2*bifW93j8G6H
zDBmG&vwP85Vl~}y(W0j2Cplp*jfDCZT;u5t`I%BuRy*2xIT4CEF?zQ$;q?K(8au9r
zz?{2p^QivXT{qVUo53b@TGJ6bzl{aomrg2?m?xMO2+}sc2PJUnJYCGmO`M};z(pNj
zHY3nEXIgLF&_fW4p^57BSSZD9q`?hMvf}%FqUrmD!!^8UUBAtI5Y%`LWBc2|Z$-&!
zh}2r%I&4=p{jjt`@D1@^1B%eGKq&=r#UhX8SMC!l4=0z+hD86kStHRlM5CeRW!t)U
zPd|_Inv%H|JIy(AYg?FH@Q$JHm|H<w;<|IRw6H7{-*w#XJ6`U4zB(p_(^r(sDS0@$
z#=K<iA5MJs<%!RJ`HtRq&?<fK)SQ@?g<LbceaEMtzvcci@%h^$_m>2&!+Xbf@1ApY
z*dbk}+uc%I;qf?gxxZ(5I5EF{;P%BU&JS;GT-@ncx9bSQ$SD<K1g1*3B>IEngS#94
ztAFsH@ZbOIKjq!K%!|V_-rOI_t<d?Xik8}plfqxP|7>eR>)iSt6;Lmu0N8(3`7>S(
zx1x^}=gG9EV!(MC9uecMcun8+LB~)l??W3RePw-CK$RxdjG}Y|x}rPZ<oZ^{z!>f3
zMo07l1+7?HWe5&OWGNcJA)=kMkB-oJ`Yvcbd=+&AL>RijI7Yg@Q`bT-^ug1lF{Q-g
zvE$wS#N*LAxyI%f?TroGs!(l~-bcp}9i4wKvQyMod+79c?Ff_%qRwO*VY|NP+{{?R
zIz{I|s>3x$hhq@OD2=}JoHJa!CwSYOL~STFsA<(OtGc0~h)A9bX<B%^WTs?YiWns9
zJ)uHZ*g~K{XOF=LR^P*&c+%T<oLs=)HCoIYImKV!;dVDn#|vZM^UK$-*&jv@p~vO3
zdXt+QSE(yt8Z=K~neiduW2fXrLlCt3ui!gczR=3T&;^E3GbB`xT{h&Qrb0<d{BurO
z{@PJl@Dmkj<}g{SS)|sDNh?+Zh|>*!Bbu#6L|ChHmDv<~IP@kf^HwW(Wr3HHY$Mc)
zIdA*~dvNXhO|>#Nl|VWp=*+n(YZ%9rlcE#B(L`88l`(-v$_t&eEoBpFI`eNr*GFFN
z0{gLJC+aQVY~)a?9!fDNv>Khc#EGtT4eMzXzF8*uuj@68qmr?TLhjB7D|l_z`5vP`
z#6aJ7sJDAXa9hG#u1Se*`al82Mm105xMngeJ=53jRXGDvO|9Ydne+}L)2*L{sal%K
zm|JVa7-{cy&8^nW$Sj6{cV3l7Qb}oMnJ!dQKsXV6(C*iOR9o{JPNW3Etk0JNss`)e
z*6uFRoI`5d3Scp=md^RrluWZ=o}`JiCtg;q1?N;*B|z78T40&Mntl%c3Ep;UdS>(n
zYE>nbrzx>o628JTEtKg(7d)@7M}B@kaauZZg1+mi;^`>d>^drq`{Ts<c;UkrSNOBz
zyv$6OGrLiFRUJKfnRtCvi2pPvTNjzzq35@58h_)Hj&U5gUp!x48nHO0GdPL3GIM<W
zDWpW-1#b5}Km6z!AAazP{ml)%bL8eZEs6T>kyM4d{leAFj=%pO{SJTp)1Ps9e5brP
z?@i{e|B%-HGAbZX?v(nk@|TN8y@gcsXi(jM+qC~qJkABg=~3K7rOKL4S#hgZN^^8w
zASZ7!Y$eFoqRqFn534;@gnCZ8r?~9D>iSm2KyMoUm827-(R;WG!j%sc0!xY1rU&!Y
zeqbCTJ~*Ni0!kE{=CBnDeIFUd0i2Tm`VeWN#zk{KaOyiw`-LUh4`d{m5N(!+nLRW$
zak<%b4AD_sBe_Z{k*<r3y%J?EOXcxAQBtK7Ax6<CU#%R-r7`D9K}QBd=eWBXc)ss>
zK1L3GV03yf9{q%r#AShHgEiGKh0uuc31$>K+zLvndzjR$$@|C<Jy&C(4?^z)qmQPG
z1u-dRuvsZs8Iv=4o{_3jxKt_<21Gjk^}qVpeEs?jT~uPuM<0L0@SRs&&KF`nQ7;qy
zI6xS5!$j2sYKYM*>XIf19mNjuU>m24k<t>Yc}y!N?KIovT9xhERII`3eIC3wqM%jU
z&MM}nW*OBE;Ra?S+YsyAw%KuGD+GCOb6#h+@57~QDOIXMmzz0@Kw@=PgH%)vFpl72
zATNPSnwh7?>|86(1u~6G&Md7_n`dqs<?6(-51yU3F^;kZ@u;Z?Mfn8Rhdoz^j)6>H
z7F^D-k|dqCj!kf$%^zP$+={rQ@|Qc3->j=`To|2m>mjQAKKT_xpWqcuk>+VBZO(!f
zr#ygKHWdmsRuZ7v9<4;Gf;f{aIxPfnnjNB}uRz0XO+`+XT%J1T)^tXXe^0|-pIih>
z%4Bn7Et08DsC6MPGih0%sq36owKyLXV`^(r+i=Z_4lzO@m68)R&D7v2M3u~%5t}pA
z$V^DgY)21;v`o}xrslG}&vo2Vijt|DadEn?qYFxeb53oDS1*Mp!l4ofYsu=Ym=;{g
z&?frW@X|P)X3}|LnihgIKDrypbzxed6d?(`yxI{3PUne-#}i|}qZ=ZR=L^2~6z|xF
z#@$Zn+srw@NqP=_<cm|{?Qvr41JC!K>(S9i$C4J_-CuZ|9m|xs>U);nap*ec(<A-e
zHBx6jxVz>%AAZ2Im#=ihGA}$%6US7k=ZSac#0Q59x67Wx%?-c*gO6CsD}M3CS3Jy_
zxs=V^DvCEma^;s8^0Kb(|38R<{o%@dG4!cz=n>7_A7u^c^14=Xrb^Yq&dAf=DT>pw
z=D#fSLi9>##-xYq_gxc?)z15y?H%8$7(fhirV^#$qwwO;(Q{338WlU|1mZaC2d;+5
zZX8TAUlbytv80NqlM2pzV&CC|5oJn6T%ZCaM%WKfDmCTJ@MG<ox~Rkx^`qCD)|t1#
z2g}Hh@(l%ry+Y83KIKH77ur(kY;GCt0jM^5NUhOCS(v*nGDOD^J$(rDArQT&=R$iV
z<qHqf!o%D&2~Qd!2o6<VsMDIhPbw^F<}tx3HE>!aT@R6c7ufe5eeCE%ui-cbS2~mX
ztYFXjUYm=EkB9`EcU<PePyXl+_`m%(|1EL9$MuoEi+uQtpYdP+=YOA@Vb9|)eoi^R
z<F5aREM}Th3WZ1*21Q_tnqK6z=;uR1(jzR7)Rv$`TJhAB$th9IQq+(%O-!qqRi@Nz
z(^_G1J;?R&TPxXS5-JR?RbTi9b%kT!gZ^t($mYEY7}#dJ>(Z}9H$ti&(w;_cs{3BQ
zYEFr1nmHUs<!v~}+f*o*%#stKHO5d-SzN)1r{J{UX_^x;N<iGoTyI+;lv3lI60>)>
z0Nypo>QCRr6$(~1xW1Mn6(ckP0VR_vH2#}i3@pnDu+SeLsI3q?Lk4R7)eWoJ3UBHM
zP|6boTUJp)6RK)VOmd2Ol&CITr07h(WbLe71gCC#&eR}kM;ur=NADMxE4;3$blSze
zA5mOVCZ**)GnuAcd`a7T#W`BjPJ0bhswGoXHI-FFy#5SBfX)@1YtRxJ%%oWh5;`R*
zIvs14YN(!SA~&K~S{BkWQ%l`s+G_#*{xhVcNB=9GuDB;t+$UIBtRk)WR&aG@*EtTO
zDkIXQ5!(-^nX&J9Su4{Nd5E6#1g;evg}3()L@!+L_Z)NOv)A_=r^;bJ@SSVVk6#Xa
z?^Vy~UF7q(jqwKqbA^BWN1szmVb=*gjpOmcFJEg$&Et8Yxq(9;={rZ3!riVX4UPE>
zuJt^-yCJ2-S8wjgDJh4j-;;362`&#OmMQW2{+=Oz!S_CVPIuVxy$@b8m&Rvr-tg}6
zNU19#Bb)!g+KjL^BQ`2nd0iNSwEjI?*!ndaYT)oV9S^Q!0@Z-%>oqo~Nh6&hG;%5|
zMu3mO5&A(nRz`+xVyNpj{`_Xo=(j2cRN~~d5TG)2f&Te5FK<SokK|NHIWfeJy`cgj
zDnp<)4Kl*%OswTZNfWhcD1!q{j>|;@FXoxt9R1L-8-hkbRwH9LvG)>iE7Z0f7LK(J
zD+tvdel<4&N`_L9V(=hsGxexWt1KUs5Tp=)Ac&44sPE4V8^fH~EtPSYSxSSNfr5A7
z&Go4@^~0yZWtn;XaOSi76OT*5hrlSFAw+@#L#Umk#*2D^Mp3!Y%A8vllM4*jceF6z
z!VZx@nI8B@|L~vivrj)G?Dy;r2VAZE@+Uvx$1iU9{lEFwIGs=2fBlAdb%(pEgsxX;
zK4{WntVzAy*A>(<B^_y(ugRA;;5@Or;`lgofB#N72{F)`5Q-zGg!hd|&<1ARV9{U5
zlax!>g?2Db9NwEf%BmDWWm-=pDSb#BvQ_5T1FD~kK1|s*aaM$3MGiz2FrD|CK3i&G
zNfTuyh&5qaDqlZdSW=>=axi4;s`m_iWM|nC)IxNQK~!!Uq!Dz1Qkro}ibklGd7^nG
za;gqpw%M@QJz32}G?c{r6nrqG<eSNZeqFEW_^9DR2)^s_Vf9(4+w6XAZ8pOcW%?-x
zpj1G|R;#AJRVbG64(J?p-FHpZ<V2k()r?5dfF=zILi1s>8+Ojob=_t|?7iP+>?>hW
zoA+jObWbV%$vIJ=wZxiF5xgS?Z-=Y5bt<~ac~59Tl`B3fk0w-gvh$ASmFrNP$M*wn
z*b!pH84|L3<Ad2;i)dQlhKy9@N|`@~K2J6p4Q}HLY)D#D@vb09^Hyx@-nAIf_BuMQ
z`p9)J+#GuPq2u+J!sFXVmI}cIZm&8X=gQ-o2f8khmdfL0<{<Fm_KG<bf_Hp!ci@v3
z;HiA_{J;-h?Ug*3GKZT3H+Q%E<mX@T@l8+XbgVK)M{Swc@7_=fWEY`x#BShrAGjU{
z#=hs>xiR^o4u2_gnid|<iJ|M+?FKHFg{dlX9Quye_eWBh>4W2RKGQqlaCPA3<uh)F
zK<}@4^KfFxZ7a+*pW(gmu=hres}YbU4a(DOzoGm3YLca&2E{?=BQa_*AZC}YMcz7g
z(9t5>_(y$Bq&3QJFqA#eby59!+7rLw+t`?YJ2in5#xAfQI=oc+&e4r~)%QEM#rY0>
z&(H<BzEfgX6{>h@h%||a7lH^w7im<|bYePRbfc`56c)AYO_dmQ1NNe=XKkyALfsAq
z9W84$EM^a@MS;%5S9Y!5p|JfueYRw)_*=u|YELnc&Nl37bHZwlQOlf}rbMZYE-LKT
z#M604-**a^*FqKb!>3i2QaGl}yGg@w53w_Kg?7`mSkO+jJRfH&g9;*k-HwzuvA>1e
zkCARfe21@qo7i9Pc>Vg9l&{opVQ`MCzBB3Jk>WhNXRqk5Zu#n~*S!4j6Q*fq=sY#g
z%xxxyfz&j!T$T&<{DyXVO}!imyBkUrj*lm%$D?*TA=0vHlv`7CDeuwF#P|SO>1s-v
z{!(Zw_&}qPsYF)IbHfMS+1JHfO53`a)tPOR*=&+CU2C2-2f>gPC#GP*ijAizTBmF?
zsD*Ltn1dn@^O9^#P$=RUJPbr!jC7^Y1xE~mmqI5F-@Ivm70#<!ja{xg&kpEY!Y_s3
zzy(Kb%B#?O^F(<_SyKv$f|SO*6t$lIhL)ngu2)Hk=9<pxv!X^R%~k{ecIfK=Ei6k?
z5^~Wb&0Lep(450<9h0F>uHk(nbSfa!4cV#U$j|~%R(_?B=ZaKtaUeDDp(&K_bd8mw
zg;|X8y<%22gwBajebdhzC8vrg;m&)9534@hTMh!7DH5#2S+V%-^*vTFhrx!fqvj(m
z4q6DHsKBrI9JGwIjI;!;ZeI2p`|F>qYEk}yQBe?k&mi0WN6EO>5K%st_norTMf5&P
zCiu+QcQ^;r<;>-L=HYT?UJARt(8Zo${D`TT1Th3ls};uG!09}37%I0{1G~=i(W@(J
zSy)oz`f5)%^tjj<d&l+Ou_Q%thCZ?n10^lYDJjS4^oXAmyC8JC$lx8L2y<(6L(e>G
z0-=Z!x$mc$$4e%=fP@~GGV?T(<_V$k@a~@L{Yd!g4dJK%mg~DK?p{1&HymhXA(<If
zm9o~TfXRc`f3uN&Yn8J0Zm29@Z1f@`thRa0X8!BtT*OI7mE~!UigBr|;i(i&ob+yM
zcD!4CCLB3yleMA>>VWxIU*9gZuw}e!>;{Eqf>%BL>Y40fP$xbC@2V1H3^FZFw30Wu
zP>ccREx%xyOxLO%cZeN#`#tlVIG>KRT<C^~i<+FJ50e99*stRTXS!f9XnG_4n0@E_
zq>i<WUc1J1=ImDZ#}0q*?ZYs_jyf-DfJ_3GlJ(H-^uc#NsLy>Vv_<FW&h_d%)py)n
zjr{nd8(!Xy<gEJgvFmtt7}$@|I$=G)YONX#?Npb%LEEKqv%kabZdGU9L?wg4^V`4u
zW3I>l509ramrG^wj^FvghrE3ClJjMz5*WsT<8k4S{_x+h>pE(BBn%^Qw<mTH2>~dS
z`3UDXl!q_L$FJFqBQAPM(Ui&RA}*-gpYtjw)tMJYfbr(1kc~_j%wK@*b8zb%(>_BH
z-C%zuyeO4Ts-vV)Gu@E&>9gNONjEDgP>nlQl%YO7m^?vu(WIpwCBfe8cZ}g4SF7<P
zl#qHe25xqNUJ|5=w15w)$8N2%<f`tQof>^a*SOpDnodYla;sMatmZ<>S}fAk)7&|A
zm-JSsw1!523+ATyZJyL$ze~{0NJ&`+!7qvG>k}GepGj+GOjQ!DB$dLpyp3RtI&GaA
z*1an#;F{Wc`{=i{w$<#s6wQAq2$YPFDOJ&&wKH=z7cQ3ew-FLq`6}v}lC4-+l>{QT
z;Y-u?zabLVA+quU^lntl?*uDQ#o0mOtnjxWC9~OX#(iOxTiQsInrKO_vlWzb`W~Gb
z5Q%zSfPvKZo~lWISMHClIVByg%p@N?L6jrbyUMfcf#)|nl7)5E3dNAm3aZj+4t!6_
zCmoCC%v>Cgr!$9L$A_=(G<-@5F%CRFUNrqPdb;Z!{m_#u9D2tmAKVh$W#c`xT&be+
z<$O6(GKxZvk*mX~>nSJpyOD-aVx@BdoM%pjOG=zBGsSy`QJcJZo>(pyy!V`znUo6G
zS33sa-RrkF95=5b``yUnoNR5b2-XwP`!8|{R$1S=&xx}%0o(h6I8AKa{D)Ua-#9H<
zyF_nX5-h1OI8PrV%d!SkJ$;5@jRYsL%1$@0gKuzst772pFmM>7jt4qVh>k9HIzzL#
zU2$G@?WIt0INw2VoGugd@k~XCPN+i%yAc;UQm#x(VmhC-QgV)cjD#*Sl!3A=c9=V2
zRGUb3Zd2I#eWRWg&a16x_i0u5`Xtt(@swh-6?v~Hz~F4K`SW|M&k!`%z=<biwW+*1
z?1)|9MX9_vP2>i<KGFqG^^R&xeCO1iE*NU?!R>)*npG!Hqj|lLeb>>&PUnPvO;@vp
z_g>z6L1=MD4Fl@JB$`6zJRg4eod4=S|6lV*|N0Mk`*5Nkd;Y<H@plMb(GzKoT7`e~
zkN&^B{^FPX?vFoUe>l(`uIYAH1O&+^+H$0v-cZi($ftXT5NXa6yJt+@EytRnd1J&^
zvvn1m1zD||r%I#3R%p2`mJ=HtC~;H|Lr-m(Sri*V$D1^8E!Yr~;DKqmZN?tZDV!Bj
zxvZbDQ4Fe?T8Xo$QMlVho{fQh5C#v@G9PvVhYqL&8rNgb^%%H{4p%djW>9~I49zv#
z`rc6qnU_j*j>8c3b)2jwd8DSFyR)%KUG?a74HtP2*#INj{bq9poYS~)#8bp?Yhw*P
z>rC1PiYSRU^baLbNqZ{v^pLYnxm3*}T4$x=?I2Vh%{ns|luhMVk9V&{?@9_=n*|JY
zcz*+R!SR$<r-!gK!3W#ijW6ObE{wFN1EG{B1<f|f(v4cz3xujeGek?)QYx83i&|8O
z6(l8R)~tGkoz><!UCt~yo5ZkLA+A4Dt=dJdK{#S-4=cibbEsXL&A6)`Jf&qy)it}W
z<J+5ShW!YSkGwq|`QnQ&xy&;aPpX>NQkr_MNE5~o*hMXU?rv}JF_P0nh>jSdsd^ms
zGI96e9sAv=>Y$wI1KjQ+k7MRMC7g6*DeU`!$HyaAyB)0whuw}&U^!n@0&JmQ=gUNH
zqNuE>GVHu$#5ww|r<H8P;>E@)o@vfFAGto<^6Y~T7%iUp>ab@RM*4Z;{*7iKv?fn`
zkTsux-E%Yj5gRuHPR!6szevkL^SY^qIINe52e*z2v<X!&8IN>A-ZfHI$JCq)DV41$
z78~(aluOgZ)8s#Zn&6WUpJ_gHR5vx}b*8Ddou0fbrrCCyNsoH>S5X$En<rXIR<X}a
z=|b}z-TsQ$DN%5lCURBoK@3r6XRR<VT4;1##5Je#^I~a!O=0?yGcIVSU#d#9oXy|X
zhdtO#M?_51yE@ky4BgfPTuFu5<T>7Kj(t!Y+~8s8tdlAFey7|R`kv@Dw7%H*#FhzN
zj0|1JvhQ`XH-@+vpF(hW7vA%suXcd6UzrJZ!4vyIM|%MyXIA=QU_A8v?Z5S7Zg!5(
ze(@>KZm%iRBgc1N(|0|OZ@%UyfAUkl{OT=#|L^^6y0PbQ^PJcX>MU7igshS)FSwy2
zut!Rz-`^0&5&!J_eDB%s^7->W=JlWaOUnHla@JHnY3l*J!mHNlR^>>9=uK{@gh3@+
zuUbOgj9t?Ma^3M7n*R=%H-z2?od;KATSJJ-_|@Q}ytbp8TcNZ{uAZ1ZF{lOWv#Wt;
zV`T8)OJX4L(J0)x5iiP`9=pJ9>{Shu752@guoTN36m@;m^G0>?eT>GG_t1(;SDh&g
zXh_yrZ|79PbeSe{(n`Mwq`W#C2BLRdmgU<#s=s#EFmw=<9UolWI+_qXF?MuAzZC<0
zKQN95BLF7aGDB5JytOs@T%G*tcDNal(0jxMjj3MIB$Jss6X!0aQqbA#`%<wLe)op{
z{X6HaB3swK5?QyXY!PY|YSZAfS}asXv>*^?aY2TpBF^|O&Yl-D*Fw})wLvS5s^l9(
zU2TNMN@&#B+`P;r^BT~xlG~!WOD@(NnF_~D37j$5SBvImvu=Ra98Lm*4?G{>)m29)
zDoZ{dA1G;N7)N+mh#^`WwJOB+!+>jzzAoJDA|KogjKPuSh1>ms=eO77tWKbp^O<q%
z+3yC<r?Vc6zN7DZ&NXXdY)ed+Sq-OZV@a8Q(Bg4FbZYDt0y$Mm6ONaODOL3V2^#vA
zDume4&}7|O3iEW~?VC5ubD|$c4p&zkp1tJRcfZdE-~S$0yOG|*-C<8BLJf{LUw_G;
z{`4mtj~AZCK5i@A*9K-&8q{WzR{IR7PpDB_lQo};2}F@aSECxm<!oD{Xh5KY+8p!H
zX!HCYf1p|u)wX?aZ8F}2r2f6vw<-o+++B0n?R6e+dCTjD*Fq(UXkbedTwKjl8o@cn
zE+UJk)Iu=FU{@PBh)w<DT^}`+q-4tFjOzwmDFmCFFH1(mvs&10GaK*phFdk9E#<A+
zAtAEP|J?@TY;I!KBCZe}>kVZ!+|XhscufWLq7tUyfT*3t3dgOw@(?1$J8JaCC^d4J
zvo0b8Rz3(|T7L5fZ<tl6ju4z3Fe?*SM2N1@@&wg540iYe><`zt?|z?Q9C`KX73Ff`
z_I5`v3+42{=b!x?A3Xo*fBL`V)vG%~zgMTh25DYs`H|QX(h{<qae1QES?_t+(Z)NZ
zW)9Z}e)M<#OP>Gm_xSUF_TO>-^e5UW+lEwFe?C3v)$bv&2C%G7l_F}jst3B;V7QgR
z-I~Ms_(b><*bFusNY$clC2TpHqcyM@uvr&78^4sY<yr`E7&U&Cy1=3-b636N5Igg#
z&~-jWwcK_Rp=u2AQq8T-+iY3TweHnv6GH%76Jk`vqAdE^1_#~H6M~Z1B9+NyDnc$=
zJd1N$csfYs8&ZOQO)v3E$W_U%Q(~_-bDE%?^(HM09eqF0()zVP5#vu4z!8EaquMU8
z9(>|FA_0e@1J#OtM3Vp=4x=35d@xd~?Zl$pvOZ^akgq6#6{^a-*Dj4s?rN3?_WG+N
zb<4jfRY{9gR4ObOKi476Bd$_W-~U!Ij}w*DdYk>O>Vj$`2Db(*RkeLjML86R5TYum
zDs07o&=3U?sUq8Z0yfl5c@eUCca)TgMqd5utFNfj$T)gPn%9!186P4+8hKgBP0hlt
zhK_VTay<r~9~{>^CC<Hi@tm8hku=YYea~DQU%h+Bk6vE0%op7!N+UNoo+mzkllb)Y
zBbT|*cadevq?8EWad)+6x9_<gcZ3*toC^2nh0os}IW3i3p%=$)=jdFc>$)dmDC+&P
zOp_*!ir0~c1jgsD*d4BwG+qnUJMK>xMImmlxc%_E{D8o}`Qtw%&6g+ZUwJA9wmCjx
zJU8`~Xl>Cw#XBvAn~r~lvhCl_<b&R6gHfE?t-D&)e8@S57}<}5Ikz^_k~UXDWg@U~
z<XV;OocXJ+Z<kHr<5TW8%Jgov=!5LyX7p2XRc3vyv|{EPz}R`3@725AI>Xi)Y4L<k
z;Z%`I=S9hgZPHm9PkHe)w(t>U*7;3(Yc;OT$eybDRaUra{@*u=a7`g(XXpi@0Wvw~
z&2?(k_tuo`0W;YX7`#^SQDui~9HsYB9iU97n%7gZtws~7a8L&@?Q%<5#rfJDJJCYQ
zxs~jwKTnK-HXk)l-0$>VIERm&dYgxi-SsWIt7}?Hq~*xv{tdOvy!!YV{dmRI^&N3{
zjq4B8=FqG_)BB&kCQWZRe)$Ei!Z13*^$mW!ru4VC{svr+bB?=@zJvdJ|9k%JKW;2v
z{W&#flVdf6vXL=YheBrtOy#|Ksc%i$vWVp|=xqR<M+Wy^hbCGeG#lYGgW|--6s5lB
zJ>W&X5uEzBD@j^REuk+r#>n+BkSq+xdFX~nEh>i%8#7#s4Zm82iqj34rqWUOWPjW_
zhZIdQtlME>1kkEV3tt>vP_Oe`3XYzxiz-Pp#K9Z?rIzxW&l%XBzkv82#qHXr^Vgs%
z{o2jkhxPDv)as33B8DV+gGLpZabDMy*4~;Jhl`3_x;27We?|NMt*vBKGk~!oQHvC@
z1O16WY)I@jhi#}!sB(O41SydatoyEQ8-?gS(4SA0Gbmxt#%G`&Cx|0ebnLp!gdrgz
z*uC;k;|$j3z_F$pmeQ2FvyPzOn`3XIInf3~^IP=&d1d=rm#Qd+A{>>R6D@gOd~nOI
z3y5T-Ye)!$;P5OwK0FfRKpgha3j5Acd&jHmfj(AZ3><bNP8!QJ;k_etk)Qn8r<4qR
zaONeVXUC;z4EmRkiHBKb=#m>XXRZ!A?yh&dyxLKnjwrwW>YiWT&wT!vIp@Yw3SA89
zZrIe@MQy%ygi&d@E|N<oEen@Rq8mrz;g)_FXyQ1h#N#yKeW340rd+A-PV73*^Xn_V
z_oLtC&;HFHKuu&bIeS_sYa_ajg9_9IPt`r&5F%?66@gsG(<p~&v&1$ODj26GI4u+!
zLaUj1p7lH}=GD>K=3d*ZP^s^2pZ~t=TNVRP?9*BcG>3Djqo0wrVu-Ys^oI~=Qn!P@
zQnyN~wn|FM0-u*z7i+83RMg0(H9{-kgVs(}52{uynp4y)15qg7%Q|0mHupodvwuNk
z`Ti_$tN3lsqF<}BK0!lOoahE#S|v4Ot$QQ)U=N%V1h25$iZF=RC}R<1C7e|C2N0);
zQyXE-X-^TnLtLX)L2A+96~6{!s9qTnVm}aR)I8(cp6c}dwpuoWk`Q}MzN^q&j~fn9
zJG$10p(pOHh`VQmIFgSKEcc%gC6Q0BxxD_I^Zh;Z<0Hr8JAV8-zs1YzS2!xn?{qVc
zJx&I2ft%;g`Tqan@AJjKbfkwbNYlweET2e9pxulTrz$Q7Ef7%jML)BK6D7uM>|7K%
z(GAy(i%k9DnnJ7II~{Alh8P7UNo=#;iXkW)ilpnJ>jLB0F%XmhhLDy-DoWl9-uxoG
z*D;7Z>9D2L4TW;vGwye~nUqXAA64F{%$dy-MEbP@t7<{&f{s3%CpI-f@^w8Vo2gio
zWcZtWJ)w937nI;us%A56a==YZkZq1?w9mDhs(y3I6Im^(6}HC;`_fF3g%xThWoGrf
z5Gx3qY=2iXC{}_o=vve6t+gl1>}|Ygzw|R*y)%?VD+F3mRIk2wK<MjRi?S94&f%N+
zh%_xOYSUh^8iBGFbt;sm4a@u9Q&r(F9fRsmr~7PaRNsKIsWn=y;Gg8F>$9@WxYseM
zXrUE+M1Wl=-0U<^#|_uSe&D=lr+&NhT+YJ9#t*e*p6z?)r7+y|j8QmUCdQ$o%Z09Z
zE+unwd&d`V9{Bvrcl^%BcXUBLIjR&O^0PM=e(`qUtH;D#D!uny?MHG^{@I&z;_bs*
z&bjh<NlZ)StH%oxI#Y9aQYj?wDT{jIr0J}Uf;gOSoGyvWG?Q{>nkJT<>8~F-><2FA
zlio}5&@$8Xma@#`v@lI)?rv}R_=D&4{mAig2C?Fe75ca3jJfeP4M1~F4|_3<d@Fjk
zs2{_Mb`~4qXkI~dg68^Y@u*{Xjf8i@$kgmifvpgzRS|@gi=MgPup0PwY65gVU99TO
zb}(n#M4KIaomaCBWzs72hI6Fcs1$o4@S<`o&HWZ*Y3iJZs^&f3Y$27!|I}Uk;0b2b
zu%7RmE1A<-q<BY&YHNyUK_Os}^?G`&8-uf7dU@}S++{9MBDQPkT;G{=5$kT8*KU6;
z6m-*ZPesu>+VI|3^G}8!+eYJbLu-~CCV&sBK^1w@2M>p9cAeVeHU~{IC1pxZw32oH
zSJdPwb`ffaldFvj;l$DHZV25T$un>M><{_p|KmTSmqI<hMV7>V^nCo?5BcPij~QP)
zCtTgp$oS<Q&NXVyNL^@GAAu8IeD?=9=lSXn{t2AFB-!cboMugE5y<xQma5EU?X;Sy
zPKu^Rma=tnwH6xXS2pBo?rMUD&v+->+*3JF)n>b(pzFFm^jCD+bp!0L4qRVfX;ERD
z{IZzypgG&Eav)uE!Dhn^w<N@?rF4jqZnsxk;hZR)dId;p7CEbYfDj$gM~!tB4Z3R0
z;bT;I(G{Xt$JtcT62#N-o9#1D8!Dx9VnuvFiFm~h^Cv9t>437?wkcD@D_6(a;qFAX
zTngQPtiX5r-^B=V!5a_9(A<X2+uQp}6XiM~vK1Lx5P-aQ@7K*<yYH3wWyl?lQnQ&J
zJ%w?t&$;ejPn)y2X75>xft6TZnrQN%{!}*VR0WmzszrlV0yJ3*9k!+`XEoRXv->t5
z<V~8k!p5>f$fC`Y*z4J-g*0;=Bc*1V^YlZcR^=1zy2x1?t}b}XF1S97bX~{WcSrgm
zay}o~?FK?0NqOcnRrb3*Uw-)&Z{I!g;%ea4UC%gnOmpFUo_R<MKmX#9Kff>9$(P1O
zD(8IR@iNm#<(S-`E-bl`YU8q~ZE<vs&IyP*(>~4`{d+#17{`v?Fc7>a<xFl3g2RVE
zt{GQ^Ic0wN`IkiRNiCDl6X!Qy^XAQKmifZ7n;V+I&tu>ZTj9Ik`H<V^&$mKMZ|Rc?
z%G>-%$%A!k)JiSX_1vlEE^8l2a6%Bs&f#3d`%d!_N;2}Y&XPJo<F)(FB<$*Kvy?*C
zxGwae`VXWgAZJ>3+XEo2(hA_cF2vS0mcKO2e_pF~5rsFhD^Q!wZSAEDDqKC7f;fY=
zRPxj8q0BiMBV0DJ(G!V7_33V%`FWiyiZjbsvoHk=QeGbji~}K-!l;{x4-wal4R3Q}
z?1T1e#HAA}2Xq?g{nLM6WjyO>MdwPEfF(TbeDWSEwY6>2SYbiuoc*j%FsF~5zMpy~
z&9lDuWg?yLna__%&9s_K$`_D+hYTamIl9=}XP2m_ulVfKpYbRE@?XIa`RPx-;N`XF
z_kQbRK6rIQci40L;zNerfw;S-wxFHHa#r%z<E%fz^+#lc{YRhRJv{vRAFw~Y#RX;e
z(i&A$WbANt#*NU};A*Gz=J$YYO!k=7lAqww^`Nk?Tcv}mrglviE5^ORiUxf^+XhX?
zt^<MV+dFQbJ);?vD#$h_T0b8x)L_etw~i`wnV5@?7@Vi?dL=T^kbWS9K*|NOj<;fr
z#&EByyD#d=fHbP8f+WPSjaF!yd*H+o({DC`@Co|2nP}(;G^kn(v{pCb>?ZH!X-wdq
z@^$?Byy6VrxAkMhthLBr$91d0m9BX<l}0nakk$+_Fl>D#t~$4Ve{B)Otw65sZKY%@
z^7ZFil`&f2SfT2kSpA#vRRGsOD8IFCy{Tnzl}aMvY>!+IcNcn<2wEe7s29P?T-TmR
zkqCXxqZPttL7dZuEIoxO+0SPq1n<0}Q$AAHW=MoCsL|SC*Ym-RXAA>FgjzBoc-(H_
z@%%`Pfj$QH1UT&251z|BbF~{dKU^4xkr)E^?;g3?_r!f?arVOQgOSVS$ko-J$Mc1+
z-#sG5^X7Qwykzc=GxJ<IErm<1ETs~<fme4ov?X&Xjv;vRJW<8QGDY#AC}}GgOKoI?
zisN#*&<z9goG4-eXAOJ^dOu6$^6<#1B))q4C2v3bl>5&=Wtt|&-3TvVfe5)3(lqhn
z`86e9lWlEBv&ZLY#pyjqL#$eBnw)BdXlV@*PmFOp1A=oz^PE{xYv&qn9aE<TCz^um
zsoL0V=Zd~Y4vpXoUJ7NEFnxpTTNMLonn+EpG-HUH+tTW{-RkPFwnjY$g^LTU%-PuW
zIG1S##fbqVvT1Brr!MUneTX_ss&#{Yg|<QFnlbJ-FKyNQI{JQ4qMlRlgPa%pdrBHy
zJ-kDBa(`Ooi|cEsL!4IUF(~nG)eIvF4LcE&A3iAoTCJ3-<{dF~n>FY<idZ)YgiTv5
zC^P!W5>-1On;vbI{kqh|Ifitk9^W#F=Tc|V<wDNcx(T6=y2!h}V>RUoPAGXH-+#`R
zzx*>!4-fp-U;k~Sz+JiL`yapH_U@KA^z3e5(DgeVo3x6T48@V_ghTMLQqNyN=%KsD
z<;4E_j`6quI_F>hgkS}UZbF&0bXuiz>*m{R4Q-tvScbTx)<nzsS2wl#<UZI8QCSO#
zZJyigz8NmCA~Wk8t?LH_;p+C5yBDwQGk+R`wZ9Y<Yuo@U8f$F~R&}Se=$Op-1TF;n
z-HwzJ(Tup9Q#VZ!Psxd#5;<iXB^Z(=Pb5*+HDLUPFwfuSQT?^M8tV*Hs6`p^%C%S(
zAa#2d+*ar-5+|-%3~^({0T3~rJKIduIHI;OJ~`~EjkJDEt!48%04-FtD8P6?YoFJ)
znNxF$@~r@Ht%|y>`^j3_E0Uq<o9kkJU%ISCYNItRBoHM7s#7Y8JUW$?Ip5<=X55U_
zT&pG0IYldURN{%Drwxr5R7O~9ru>DRg|%2-$<v6+fMY)*As|giO`+yzyN-PnVo5x{
zn{d0paL~LIZ_ira=@?;%f$N>RZ%R$<y2!F9@2l3zJSWC+$9_K&f+yvLd76mM(+6#I
z9xsW{-_3meHuKAe6F)!Ae085#Dij~_<4AYdBjWh_I4h9jp{AtDv#N-2DTU5^q@spt
zF%r5SDTQ2SaB#fLIPsj$XHr_U7>JR1o|#S$eD<?HW!#U<_wPuTBh&e$%~r`Q%Zzi5
z`}2wY;fnp$wVL~w`B~0cn>TBGbsSYuJX{kyU0IahVQV`?9dT{~G8$db!gE>*d0r@G
zMwT<o6Ot2N>^wo^-K}h;1ua0MbBrO-x`vl*a?-#0`c}n&?opMz6r4e$tu=Mn3>p^P
zxu7EgdnkkJ5vPkVgotmDvUWLL3`$mE?En-$uVx7fSf}l`gHj1D%|-<JeVo&|ZnG{@
zoF~MNF7`<93e#?~oIrD81{Fo6f?ZrchwWf-W*w<B!=g}53|p734|lKw#E=qm4^uK!
zv4h{~d$q1bA;BjT6EUgW#yno<MK+%<b`Y-LQ$+83j2%&goZm5DW~2zy<D)@>g6ksb
zlqh*2^buDDT4t^br-!$s<6ADLGtXbXV&@ufKl>@qL*V9W&$vGjhu#XHi0=pMSX9Fs
zV$|7YY4|k3{V!nu0^bTO3y#e7&4GM5;(bpoD{1CwQ_|w#FI-I_`DUC85l=3Qb&v0N
zNVd4`fZs*};%IC5O)c8dSz)<$rPZu&W7J6C{nZUuH@6%7yKdf32g#H9%&JCNeGs;;
zQ&0~Fo(jc=5nhdvzRDYEXm?<9@zwq|grH$g*o?V8RrDf){Dv0;&el@cng|Rz^BXeo
z-kjjQ8NG5mcu$tMX0(@LBTFmLRnf7o1+o1=A^WnT0d<?O+EnoUV%!iXQQr<NG#k#v
ztwAWYQu0DWIV;Y_iPl=arx|}DLF$jx%neQzKrVPBcv|y19&Hts5*xxG-ei&n?UzbQ
zi6t)x8Sgt*c;EZTZq%YX<(YZDuq><9^OJMux|UaNKodvvI<m<nQ<~5lO4~=zu4}YT
zi|6y@Og{+T4zHRsT~xE5`F_SzxZ3U1wzsKg4t+-|jlPS&!n3Nia(_H=z2EWn?E^Vg
z$neF(nLqmCE$^nrw5UD&aL|-O34!T!Qa^#55eFBbh$D18UDwn1J=vn^8=C2m5axMd
z2tw#PPWLBDx>#}Q>H41YG|^}*Y3BU+mU%gm&l4$Eu5Vv(bN8H-E|ik#cROC)J>zfu
zjla&gKR{J((Q1_Dy;n>7tsvDkp}|q>*<j6M@WiO$f9qc8Y%K++MNKs$CATD8&bU%=
zt>Q}nWHZ#O@u@+IPH?d&y1?Y~Hw6a1T`>^4o^ih?<zyYU9_FEoDi12z7TxO7q*_*A
zRc;Y1yWA~-R;wEE^!>I0uZ-(zrb0@<aQGNh9@JE-=&ggo4vp1@aZ8dCOtKV_CYm!}
zb&+_}As6fXLeRgvx`?Tt_j{d;h<5q5F$EhDRNDyr`kt|3$+W*z+r;hX!aYUmzF!PD
zA9Mq=`KYS#j8D*%!1jJI0G2vt_f1M;^pQ`$K63fR&-naL{w>`&Qmzl2AKsEv!R<yu
z@9D;#tD9SDIZ@_?*mWFs5%>5NT|P114D@|R*Y$*6ZABf>ibfveI`Ezvdqe^)XAqB*
znRfgV=OffinI4e&gj9v?e18DxG!p_XtW<9Q-n>;5qF<$3Sn;94+PdAZzqGE!Z8~ye
z20~Tm$ERXLw%=M)g6is9&<{Jt-QJWVy78`X@cPeM%jQL{j&wX7WX^15c`E4GiUfs?
zxAnBb#K!Bu9N$*tAb3NER;ahEp8e3~-|Pru^<@uM5O@u3YN`bn*ER?o^srxH@%8WO
z7_eBQp_=WwA_YxHe^shN!>!_)evqXoy0iN9gFd4e@~54>77kic6nh`)Vc|-nIYBya
z1xuxrg!c2YEv+?5xK<;rzGo8E<f#})d9g!6Eq?{5se!0bn^be=D3yS+;p;k5S{Blr
zNM!=?%7&@xy%9{V+_?_Oh{e|as+!>m`C54xrNWXDr}ND7>mA>D`J66*m!if%tx#&F
z8wZ5O<?(@ZxzKl>tLr_l-<^m8hkZcEMCU*nQx~~h3a9g==<egpG|f!s3)IAZh=fiM
z56vl`Yi^ZPD$_hsCDIRjVm~N{=Xm1u@Rnto6@7CKaRC=Q@~WqIDpOr@QlE)ZN#_er
zG}7wvcxFn8aTut%XimYjXjTTwYs!o6^>G;Z;rD*PZg<7kKl>@|abZpup1*v-cmBrT
z<ipzoWtk``<3%<p_z;4IklDJ(xoz@i-wF}~H75jBxoPubYaZPv)?%%Tk-E%Jh*@>G
zYA9+=y2Vsm>u8bXqj`8Z!HJz!<(p{>zg01C{p=;zSBDKYb>ayzZnJXzJaqn^i_S?m
z=PakAg+^tbCu+{buBRUlbbY4}ORd|8pw>doiJG;*5U*o~5WEqBtedc#hq!Iy<h_~e
zI@51kH!52P>dpDin;*Vf=30<adryieXlO|jD*%E?K%PiK%}8dgRz)XTGf9*kWX_rM
zp=UGXd2d9Z5UTDW)#jwGfs<%<zE6jhw*)jn7@W%I&X*HBf5xBvKmOmm`Q>Z1V?FFi
z*9V^c=p~0Xa1~*B`-b2P-+%Rto8EJXjj{+=-}@LJJl)|+J<gp%-Ys1yb3ulY)-}TZ
zMmuD2G#4m%_04aMI-T%RsOdtPPPFC1ycAp<xVm{k7_SkppS6^przA_>JJg*O2b%G{
zI(KdNOLR@GO*nS1WGxPBQI;+%8o+jsbu-udzfyLbkw2S%l6ckiu&^<dcMi9XCK@!q
zLci-+>pjJaXvd+4zsGs`mEvGENoqVv&o&fH^zdItCnY6){|yq;w|P{5?Jl+3)v?%0
zqM{XAwPWl3!?PA@QNMk+$%xgNl690;jgV`ff!^Dv+5d_dIU6sm6<jqh5S*wQg|?w}
z&A11pG*x>Tx}aa99eXZmBkF2wv92-w9PI}XJrH%|`$T#zC6iOOeXFNy%7tjo)5KF+
zHkVKxuSuFI0#Z~~H_bCkD$FU9>r;``8_MMkIn@0);9S{|#rKm4*JCG_N=teG?}xzE
zz9$6e`$)MgoTf>01C|Bf0xd%qgxK|@>B8}NCb&lDh2?Z%p>epq#f=^=iio9@Ih`-e
zzH+-)>)_q6rz|>JtF<xDNj)Q6<-@x@`~5X<&WReLj^t9}{N`(x)6qz_PD9m(&Nx+i
zaOP#E)?lcyA_9HiG0zk4zJAMLx5N9$<vf|`&_wJyYDz}ZhGkkv^Nf=->3q^AbUy26
zSqta4U-R~7f5s<we@i)VVI?f9T3{F=y}|ssm^8B)&&@f43rNsrC^sls*X9%Bzv;Y7
zRK4d!KUbYtRa!3ZHBYDpTW2Ixr>0Eeg8`NfFQO)R-{AUI#XvufbmK@cmb73{c????
z_8O?O&Qemz;4>ivhH=C>&zLhb4HgRBfX0Yg#SXSeDy1ysd4g6<zGo&SU9S&OH94J#
z*&KIvAh-9RTZLOoYe;qEl9a%>6#%RB=ZR75q-`^(hO#!rk_BNcObmV&SjmFe8MV%Q
zo6YEIt4IOY0^YYRGXZ&0D{NsU6>yHC1uEHyk4m!B1F0+vEiL*jw>?z2oKO7tzy5pt
zn}6{~^t(MTfBYl9^U016W8#a~Z^;eb-Y0(h$G^kNp%T&qp)R<?wQejv;D?c#CiNIs
z5;EQeHRos-2m?~kw8Yqhn2&pzsi$`=Wx|!je0%_uFOuh(`#iC~dqFqc5W0g6$-3>F
zTw!-vyEJ1pt46nOE?n9SS;y$!vy`@9;+|r1F*w=Ia6=m!vS!{2-~u&Rhh_ysH3%3x
z(3J!vCJTMK=gk}xoqLj;fx5>vD>!AfyESCM8Qi_{9=vfC*0o%6vF{De=k0Kl^vx_q
ze{C;u7CpQwT>#V{V?C_b<H~y>IMtl5-VjDBRV_bk#8jifugLtiQGykXUJJo>LtQ)c
zR+>7x8JhG|Hex6sIvO;hvs=d)HLEFFNeb~7ZDO>b(}tx<HN~HKY}h)?DUq`tf<+(&
zPjP`(GUP(dwjY&(w22~)eo!msR*D`}r6?z<RC2KqhK=5I_WT~IwC+Ep(5&eZP0Z2w
zbEgJchZy+5M|XtasX1|fKGLBD+3|Q!EtxSMxY-{-D${h~`1YO3;+$|iKF~_y_V$(-
zbOaQ=kaFhneBm-r-0XW!(nucivS?70y5n%VWR@lI@$+l0uNz++E0avzA189oEc1ow
z^oVcJ1<eHM#vMfU1aP9kNp5qURKoJI%;c<OyuRO23Oqi%RWhnm#mF>GG!Z3vx2k4o
zre?~QGkq7hzPjRzFH|9QTngX6xguKrgy78Vsex3k_8rly1McZEQHqWP#Wj3%bY76+
zSW+gZ#!?ckC9*x=DP_u%be~EIUn`Q89NCKP4R&?_y8atlydoMYUn=4|a6FB$zQOhF
z#sq7ih42K8x<;VVO4InufK~dUw)tp4LkOk?7l#C`!nCqc!b^Lv>enAm^_kKN&7d{J
z(VC|@sLpSYYc1-`<y^3Ce4V#G8HKE5t(4SFFGTIE!DeWz_c!aD+$(WWnn{$_>dYX0
z0aiNkM*3?{qXKF%!jC15iOK|52q=WO6$*Ko?KK-)<SNbb2=p*pea>l(>2yR>9#pP2
zg`WQQIpOh0c1J$^?N9i@@BbcG!^n3IGcBFi?*>8~xKCG%V@IFg5Pe6wWSV)w`))u&
z2jY=_r=M?GkX70!fiz#BHC)jfD}Es78PbXT_=Y?mscBI@M%U5Bk$Jgrxu1!hGVF`1
z`0h#<EZ!)Rs-a!k+*)8+Jhef+;+%DqdN8aW^>yWAY+XaQ!TnB4YCEjWezmP*0wM0M
z=&x?qseqAS8>^+15<&~Xa_YCknfI95_PWkGvu1u3KA0_)wfD`&P$HOazT%qkWSVW>
zVt)NL5?LAG{Wo<E`L(@P)GQg{b|rc`^Hg{`#QQGjp;DE&RT_xdvJ~Tjtj)(dc6)!r
z)7h;eK01<c8{$%$`nh9rtz0}#f|-L=Qx2^BoTjihs`)RuQq!a+FUCEnt&(!lV%e!5
zg_G6yVkNRF6JMkmQLJea<9tSZpp~S-W_8ui06CM6N79lIlw_$6r=UrPj=oc#R%?p3
zwpvu-Wxq-;_H!4+fs;UWo)8CI7(hH@5I(ruadQY%@iZB@f3t9UJW;13B`-u9!JaN^
z3r(KshLMMNZ#kbY)RH*t2gZHR(D&TGd&}W4aQW<>`{Rkj&~u{lcswz5o}2w1-x{eF
zD2ZuqJS4}b_eZ7*Z!U%Nv>xV(Wxf!cIly|&h#9&-c0rvMTO~SgQ^H0U18JF=&L^Z*
zV(iJK+C7Vg4r^%Dy1(lFuE~#cN)!=-7uq~qb6M!dj_0>`{O~&;aD6op;HlxOYBhXw
z7zy4nr_3~+Nz24^ndrT62qV!4#5Gc`q_il<0@<=7YA)pSMbEuj@F}Z>e{EJw>sntk
zbJj}>PTTaV1JRbL{so_%f8X`()C74YJgv}~L2X5KjUqyw^+SGQu<C=~by1@t#mGZ8
zN6R@`CHyoO_TCX<wC-9f@s$kLoTE%SYggE>=z>XEH<r+C0YmHduo@gqvpVAqyODhy
ztZG)_$!eX9fOAz94>rHvH10MU5Mc{ASr3G=w(=M+X)Oi}0utCp3OJ`Te{}pXpD#A!
zZp7GaMG9y!QmTHi#lF}1c+})Cedpxs9ksva`FDQ6;pGQhE+=EdFOX*He8T&YA|u1|
zkBN06RB*Xt9B-)eT2-qFPT%zuOG;xokwr~NQqH8yBkp{n8}^*Pe4xL1p<QpDI6r)i
zN4)|<=%_iflt=dC$eb##fB6}&5O%i#-xWlPl1zP1MLT(i1yI?jNLn=(zY?CPs{zsK
zDZj$;euHFN!_1$aG}x@*0c$a!q|vy)VYq&w-?B9$<W$zM7mNh7ex6U}G%H)()|kn~
z^t~v#tF6UMHFB6C7n?cMmUFM<*Od%pj|ZEVJdi{8Z64KMyK9qK*$530R<o|{CfcYV
zdgy$`heonlZf$G#U_{VTsqQZq0&6#}2eM-wRko@DK-(VHifa6w!+SGDGc`&@s7{~#
z;8t0(@(-jY-KZOO=sxMn^ODr5G6rZ$WX+30|G8SRvF^>aQu9nRV<E9UqB=t`3W)cV
zq@8?mYMqRci+!sbc~PU|GfQ45CF4a)S&9WEc~NWk-8FGM(C==^Ne}gX@93Q{ohPzA
zOKF}!pzC|)qE@$;%S5Y*7td~p!@zWUpdWYib7r41<JePbW}Yu9jqL*W_m4=cOv{<`
zSyfFBmx<$)xEcm-WZ?ec%$MhdU)*P2pAywaa;{9vY(HDxlt!gi=J~|F_x!aV-O_af
z!+66VeRj{+bHde%EHg(t_@x*G?=s6gD`K4&TB*v1NQsoQy5DB=4xt*+w`lXGi<j%G
z8&Y$8|GVGg2OqywIUr&Vl$Eq7mtjBjfT7Z@P}4$A6TW$*^;?mzU(}3PsAuT9E?Qb}
zwJ66bXX9EmYZhf|5T2eaVawR54V;b8ury40I#2$6*S9JL)c+e|x03GO`=M<9nVT;B
z@*Y!KGs2S|hOHUN&JIyKFvX_0YWW%nx!{9FzG^yIu>;lgm|;y8t6QR<6K4-zwvJp~
zy>crosj?KIiN`lj5oNfRHnS`X7}HoC=S08Xdn{ayZ(bi{`x&g@S)oC(Lo9ebw3ei#
zuQgY^Y%}79oGzPR_I#Puwak0vHHb3LoruEHZY8AX;pgIjkE1D(0<PN;LQh8{jyw9@
zm39ne<}qi67O2aDySii8JtMT4mIs6zXqgsHG`Cj9ioCR3sN(S5o}4pvx{#+M)A@n?
z?w)va#q|0s#%EvSyPmG^@KOvqca*H8!<J?)X=1#2f%wSf@ks0y-iyPaT|89}(6~q3
zj#a)Xw2j@Z#h)F{@B0oQJVEr(O0uu-6oP(_+_#FuYFjZ%khsJ3#)C19xc-ifpRhu;
zZpe>?AhW><$mV0AE{9gUG-dv-#xE;+w3;gU5TD4StlIaszM|d(Q`mIB;n@T$`bKL{
zQl};~DR#W;eFyL?^lqVrh8NYCS453kc#@;)y!B}wD<V*3dsbRSM>cicpQ=q2w1dhy
z!?NIg*~W&of>g*AI<(VoZBaSlk}O=zB*@m4Z|G#sEJaa(79(*}kW?d#7DN;QUH1w$
zN$Yza0A|9pt!r8B)Wf#UvG22vwsKi$d4@I0*arw5@#-1f^$Wu8$`VCA^Sm%E7i6Aj
zQ-P)9>d-M9cFdPEw4!t7nlwnttM9^Yx8wTmj#e|VkKBLV@IzuJp6PsI^qyK1&z{}z
zlRx_uS1T{>u1GEM=Ixna>4vArGmlf^-TA_tGhaMpe(~na<FX#Wj$9OJNjcGZkMBE{
zd8XD%u8v*s{Fi_C_qaaX@%z8{lK=USzv2(Sc*ncxLiD;`g6Np_a-6p3n^F`JSQTS7
zdoQ)nlAc}B(q+BJxMqdoBDZ%R@Y}!h1D;*)5HzpGdo5U!Z?sm3zEPXtTf*gp)K%&Y
zt8~4ZIx1>c2{el;M%P7cs%9~=pcB*pt*OL&HN&eWsZ`0k+s0(rpRa`cZ&nI?Thk$d
zl=J2<ug`WfM&s)0rRE{4kM@cf1mgu%s*M@6^9lA#oX&pyMg|iKf{PL7qXl=MyK$}5
zrVpxCW|`&+h{yRr1bi)O^S@R|T_YMw(f1vFKk75pD!wVf$2+%4ycB}hO<K&So7V8+
zz%@!WgrgM1@*3<G=}fQ`1htx7bI$=>EoKN}b59EWeg{oG308|`=hSXl_2&*G5Mo5U
z<NAY-xw$$3N~Afq$T@)v2M{5eKEC9c(@7{!ox%F+SK57r9mGg(nIKgQi<)U^QkHg}
zZ8YMjW#Rbd3;c5C@vG12hLI|Tc6`L|deznjPiX-cI)>Qmrg(bP%{I*(o_&BH4jlS{
zhGSV&H|xEzVw<Y1T{`0WJ7l<ff`27HEk0R=<SUfBdbC+rU3H!+KC+GR>~qt<w35uW
zmr=-Vb;{KLU{Jr`#z3-$imZo>`-=y&b-Cs|=WHI?M4(y*1#1_J>foJM&P8jqZ22AL
zW6|2L7DDg6x!>4p9i_q3FKKIzNVfU0c1NyEgc@~p<$+w87mG1&+98%|$e*DMV(6Xv
z>1z~i2n|Y8!m)7_l;@;-Os*Qc>tQ2gDL<r9Q(<Y4d>~M)qs;mitD*)K{h4ygl(ZN}
zVKp+c?o_}9vpmKb*PtTxDNehI$9aQ)0c`Aq?St!D4<T4F+mKo)NzJiV$4%c2#BLyt
z*L3@9WVk|l&4Hj+NSTrf%Y31w%<b6m;>z>#*%f0ya6F$lU(R?DeV=I|%@a4z54?Ky
zlEdK&aH?pS7bNstJ%7gi*I$#5P4{0M=hKn%>B7}@M{dG$QFwkFBQEKoi3rC<jggj|
zm9#9Cw48A`)!|o@aC=X`8`W`gStxiu|Kfu04t(&V?{o9whA{~r+}-lefA%?FyuD{B
zg=U-tX_Y+B<h<x8MLZH9+gLI9j*>EmeWZ_jmL*Y3;&MK7bA8bFHthKQzwz6A^8II!
z1s{cI-XPV;qG0M35n5W1n#?FmP4;xX>KgPS8fzZ3Fx`UK+LPx>5C>k<HnII-<%6uv
zigw86iBq2vDy!w%(A=it`>U>RRSal#-&hu9U#`))j;Bpa-<!g=F&cB+4mYi4Y;L>C
z9b`=}GmGE)v{~z(tI|bZD>YlO5w*f?qH(YINN8+OyebRY2CJ(+SZ$Ek&DsYoJYpEg
z+1ngZtE?D;MF%T`ZX*cRg14Iv?aA=O`OREs{Z+D(dZG{XePkuL>APH$$kwrfcDW%$
z^+B+E9YP?6PQNugb&2~cy2BMB9v1>Rsay~7)LQ66rw3#348zWt+)9+|x{l}r)ChG}
zi%aRi2~`wPp^6fgF89cC;`HUGM9Wifu_N{!-gVpD-S@p&XX<@+&i1ohwbJi;#_MND
z9H})?99+(iw3Hb32da<wI#6pSKYm6mGYofn_?L`XC#Y*=Lq|ufGv>CPu{3DT>|x3R
z7D}=@*6Hkdt<v@2sd=L1iB(!wn`R)0SFB3fUp6no9@Zy?hio>tEApkD^{`gyZH4%-
zU@GIJ(5kJ?YNRUtUfaRB&Io_QIRo18>W3)r6*poBYAyANKwLA<skx5xf!sWE)91gb
zwWKzAHd=IMcit-M%Wnl-bKdmd>xjnLxKYOgl2tP5*R;!8t+^2MbXWPYh<)!?RmLWr
z&5@hUdYd_`HG+?H(a}ZqK8dDCDz#B+Mr4&c>H)vXQ`h~<wg<&ZH1<NdQy!=R-0DPV
z^KR$xy93wHKBBvRj&waC_N0`VlX9xcd{M@{b9!K;1g#LfqYpyKiFcQo>2fAG&*^yN
z{^5ZjjpzHo-R&(tdQwSTjuTRw@v1V@<C$6;-OxjA98YJu;2HZ!A3b>~%qbCK;CMOX
z9PG!QWzLjbxj!B`o-#|)_f=|D?*n6-wFWT`yn68wr{g2j>4Kv%O^KiVi$CV(=8C*z
zKKS7$y!hdPAAkS1`JeyfPxzny_>XxroplUtGWD%<mIh6gWx+WH@A&TZp1=0d4gJ{h
zfB&a{%)BI=)3KJr@zIOt{P?>ssmn|WFbut#*QK(JY5ba5QY%t2K^lGZK(Aa9JUIiU
z(Q?I8E!nUTRt`WZkdqdz!IDx#qos;3nOZ%~W!;}j+r|Sca#l(vH$~e@SqMGEZu~~i
z=(j2cx-K&G-A1@DWBko)e}yU>>iw=`KH1=OK`e0z#6RuedN_w|2ByWnX|kKb&JsL@
zVyM8A&%MrE#j0l6bi)eMu55BmA5#Xu!@Iay3L|xM>uPoocxN5wHt$_Gu4=|7x)Frd
zHZvZDf9j?a58>%RU&B>uV?#hTO?qn^AHrLOx3V3abFe~#R+y}RXO%&<l5sR*7&X7a
zG6`yd($prmi=&b@SE(dAQmLg-(izeT%1oP&)aAl_d`p=p=F5>F3K2@|7!M=reCDX4
zhw_>a(G4Rx&6J!-r$@D#Et&3cz;y#v)b}BHO;ubS7^l3@<qj9ruGl+AxxB%*1l=`F
zhK;wOM}IMQLhO*#QIrqS{2WxKMOBx}_3%b2n^r+nl<eVCP6}dL-&K!pMleqZ5$lvj
z*DUg6X}zrv*@|C5O_`)-WVN3)WML(t;(`?k%|=bz2h8T6C=OhJkoir%R$J}WN8<f4
zrt{9abNzR5Lg>~a*vOT&Qi|R??*iVzQZi5bhmCHW7G_-+h<#LYDJJi0ZIctPQr2RO
z_!SYd`clV*f)&TvTq6s?N5U}RTp*Rk=`x{XeMjdVT^H4>BWhC-3?Xf(^l+Q|%IeZ;
za@tmGY^qrdpt9Dw?FNb5LG%GP?D4}h{B6&$S9MPFPCH>B=VIjc%6vLe(##Me(i&-*
zmD@m}kC7NPLVI2&vy-n}rc6qW-9B*GceIx9GU9Q#*pn_3%QRDSW*7(NWkzb_^)J8V
z+4EbRc*b$W-5+^8K5&{cOV*#``fA5?nONq;dC>#83xSX`UGPvG(a?-kRCefn;N{f~
z%o9EaD4Dm@h0lKZj^UsD0e|Zs{3ks7&L`Y|dgSFtdtQC;0UwQf{>h*GYkvCq=Pc#{
z;GC$Ns*g$>4)C&v?|pE`4_{vK(e=Q7@O=8q7yQ}l6H}ULCG))>{fPhQZ~rxf#PN7!
zxCz@ifwObX+!=%S%GDBK?0Rm7#^h~Wj8Iah=1k2+(bud5!DdE$tz>FRT5Orsut(7p
zl@^*|l=YI;%(1Ggq^c)WX@ydYhG4~@W9x5leXC+1ilZ|FofV>UnAFaDXx0T)b?fp5
zPk0+4toE4icLvs5DI0uqBg1C(0@;FKoU*gmPGXhsY2>KK`DhWuf^*(j`F0SP6s=e|
zjI%^8J%Bq}cygLs15Gwqvte_}Rr73#;#O#A-oO*QYqw?V`a_I37YMQ20MeQhYLU8j
z23Fl|dULnU3pd!gt{tOVuwqupicVNG@any=<=x6Qt6qtZ0J&sbNid%%$1kX-*ObdU
z()ocr&zsE6d;Ofde!#^+Gbl>Ku@+WOlDS$7HO;iL;2q>?A$m_UTTpx;w~A{4i38pK
zhUxf@ApmJ+DM`&c1VkF;G~-etjJF!IES+`UPsC8r4bR(qa6a0B>!5i{X={cE**v%@
zqkWqb7+SDe&91ujTJuI4-4^RQ)8Awx`bQc{)qeGT8F8_Zw;AC|eiRq%v(84MWBdK-
zHI(0i4{Gl94Yk6*uGbn>9M%|UTiafWxeKlt92U;ik1C%_nVhS2&H5U<q0>EP*3?Lc
zjv-wL>PQ#0D-{<gcEC}d_6@gXhv+7&L5H??t;L?zVYRBpzm>*rf8h4{E3WQd;6q@Z
zX1;p;HD9(bd8irh)J-lpg@L^&x|q%SyyUX!&evvT^$;nwGB24@8^J|<SP7~|EDJph
zd))JnknslMPMt7wBc)24CzTb}M&ETf5tivfo@b(yt>C33IxpM~BjlOOIk7+N==(^w
z-!o4$)8#^RBZt0nwIA5+R8D?;`;OD;#MNO>%O;cbp2PJGpFezx3z4oHDJAjn_{iye
z#tGc)N0!=H=9xTAOw+{*ZlUiY1$bNv5+hyjQM#P^yDYPCna*6r$gA&tpZR>|{&=MN
z%Adcva6O**^Z(&raC!H@?(hLOFJ3WT?Pwl~Qy+zQr$@$NP}8QgaCLRf@BW>?%XRva
zm-UTiiu5C4=m|dazx}(v$7dJe?fJrZb>PK^A3#IXOlL(4yl6tNt3r*1&PC(?KuHk0
zj<K&u3xs)A<k_0yS_(B~q!dVrGAVKc&~l=tWF%O%yOyTo@>a6$Q?~EsCZwvx-=eFQ
zh7w@4uvH$ObN?a61YX&b;@!3xIy2MpX1QC;1VLwH_CTyUTv-iT8o3ri>~Nvyy#cP8
zWLS5-&I-hCfPYUu=4E|Y%yqC<HU7bh3+M4{V_!fB!3bG4ve1#89`IY0E}m88AlAvZ
z*7Qb+d#@NMP*vh6YZkaX5vA7m@2q1{<RgMW$=bn-dvcpXj3F%=M6@!@*PMr1ifx7v
zhCOlEYh`HTlTx0P4S&(6U**=7nj6le4A=Pxr+4_vThjS0bv`R1(vOrpLvxg@*$v_&
z;sPEhu2CqsH6W$ysSWacCQoPGY=Xy)gWkUob>Zej2&$lv)+jy_UEuW9S9JRUA2ewW
zAdUw@KhmbR<aC1J4u5z-^}Qk!upvS&7&~`m-W$|vkAba^VtQNwbE90l{o1OT+1TG#
z)|nMV)zT}C@C93OxfTz!zaZ<`eseZY)Gkn8&x%+x4q5XMeDJj9HqQiAoP@2&q%~Uk
z&2}L4!P9k-aU8aBAR<I(4vW+$UXAZ)5~vhXPUK<=3kShN7lq(dT`{pBYcVd0DD-`=
zaqBp0Dx_t$lt{6ns<9?#w$@aZx9%&eT(}nSLKlUIV|Uo`@}rOW_>&)Ub^8qGJ<H|7
z^|&Y0f-8wB+w8b15j8IhOU~p{HYY^THMJTVt>~yYhYt~{;9aEa15KPV{JW8>7avlG
zYc9EQJT8dO#2~aR+PUMHlNRvBIhImbE*DzK2rFW!0a+y{mTBSkI&yuj65@235bub-
z^5OF<IuEWWL;msMLZ0u5-m}{coQ{u*7<$Le-5u}V-K%-e&~rSU)P!hR82XM<D}C2<
znHJnIayl=hSv0M)Hja-c{IJ&|znay0h~Od*$1{E3lTu<j9eMuzIY0d9yL|QV!1?mX
zNdkZP<!hF*@cGYQ^Edzgf5y$L%Jt0+?Sl_Ee)s#Fj}u8MLyWxn@=M-4ykj?x{N4{g
zVfu36=6qz>MY8jB9%6q*p9(*C{)|_(ktLE0rD+A4YnD3$U5GRvltkE`O!B-D|9l8^
zog+68r&%PM`)F;3%x30kCM{MB=&VaogH~~<+wF=bl$=Sqk{c|wG3Clq8k!gAh@Ic$
z%g)ZsZ+d+@p8*G5>{J`u%*e;dQ)ajic-M@iAzNkAyqf7a7l;VrDlR8N><C@2j$T!}
z<)(868y7g694Tz)R<}PZ0-RI3PTgdiEhyEGden()mH)sB0XAdyW+h_ORidU{fU0*^
zsm7DLT&)#?N1D^Q4uV8SjJ+KbE3|KBM!BdFg!9T6b*p7=!*x$Gv9%lV5_Hz?93l34
zfS4T3`wmyjMyw36v!W*8o9eJFedVcm5CBg%jI<`3Y0<n~NXrG%gex;uA{;)%T}OiN
zRAx9mAo-CpCqh)70N6&_8YL@wkn)0Lhqy?oi=i^g0FJ{>p{`*MaiBUie5y?|x1BMr
zyQ@8QnY2)GQ0IxZ%#>kA9Cn)cH$BqQ1$XrdiMJ~I64lTu%>EZfLRDdYwf%h0tf$$$
zSsbh6P<%i_$C_`iLufJY0A&U%lJ$f}twBZ`db9pLLkE;Z>uf&mOqGHvHr66nSIb5?
zsv>yNF@^F&nsswVdR%_9ivj1H`rQv5;zXrwfR^mdZoNK-8JLsGfK#qCJbkCgfkPNN
zjic>D$Ag?H2*sfeY@*GI_XgQWrM8Ubi+F~vr$3B{c$Snn9#1S~&5+kIp{kP#wK|+c
zc3sca^$pk0UT}T)jO&|Q<udGcEGhB){79=a=kr;gby2I_S}Gw2i?eOS=+uioY9Xy_
zP`v`GH6kLO<^s)o9u~)?c(M!JS2(?X3*Kv3k$2P(2riIQ!WloSwMv>7YR%ehmcleo
zI3rfZ;0e;$jU8R*^!cd5s~5NQeWwUlDV!hgxjdeDbAKYl$bRVA?MI|lc6Zmz(}n%j
zLH9F(RvO3ig&_ozQ-p5}4nGdG+PHasOUoC0NgN-Q$n681XSjcSRAh!iOAFq4N=;U1
zL!t4nKKmITy?Vv>Klnh=mR7kMN6K>I55D}G53e&HzIn?hokuxYnP)F=`QH6Y_OD*j
zJI@z)2fq619aBl%?E-&t|At#Cq4VszNWpVFpYZ!DN-NBlg-No?*|Q$pK{^}5D*r<*
z?_rwG<Y^|I&&=l|^OAKBZ=yv5))Z%LC1@%~U(&)-iZ&#$N!%T(PDwdK&EzFfmYK3F
zOsO#C#+(XsE);PzuT2FOvEFEyZ$G7#zx+V{JFai1Ch#4;S2-Bskeqa;rc9r$B57_-
z5QG(Sa}7i?zG;`>jc4Fmq0~gn2})9^Eso%#LFqaJ+>-ymiU!?09L<mf{aPn&3ntQe
zKCF_ZVzH1SP)kzSGL9%~KAWts;3e2bu4u@rOZJ4#kR+!?xeVTngfN1gd(Rfxa>)_3
z7A`dzuVT$X&|;^SN+#K13(In03m9=44)o;DW(Gt}=h2{JgBATy2s98oJ)pb~TJ!)p
zEtEV{(gF?q?gg#g;eAJKPQP!7G?xjn61@_+VmBaWdJ@r49ck*Q28~uTWm(io(J9R9
zL!jGTQJu!4dZ#4K7(Hpa5L#B<b+^NHBT|wTYJpM~TwYja&#=FyxsGyqM@<*P)dIs6
zRe~W;vN<|7TMIU)*^Hc64{-z^I;+jB3w62y>&#e5XNq9zLn+2f&>~CB*=_Z^U#sqB
zYpiZ)QeLv56N*Zx;z36oT6ZYI5w`VWg+xh|%C^>5KZVt&|2Op#X$05kgBF$5>%COe
zP|3TiB}C1Il$7x6#Ix%>F{tgi07Kui-|Yw?(L@za8KHSk7<<OUj-l76?qx}g<3P)q
zE=2C0-7)M{b#pmiczk>$Efb|w)@XDQHl9c4BfBmTV~=Z&v@Fb*i-xq-O4oPXK7U4x
zx+h?tZIh^yXyX7xW1oAkglI(xy^Z%2G1P7@@a{bG^EVSyI&*(Ma!!{g=6lus(>e8=
zpO%TVBpkwS9I2&nIUNbk<E@Y^rI3<JyKi<K&kup?UEt=>a~LCi2n6(Si(SOKf!*M_
zca3>TbY0+ZwMT%I66gC9v5(|ZIUgq;-<^;KeF!)lHD!j_Bek*bJDk977wJUUkqISH
zq~o&8biMHURC#mC7CcpOOn4tvZWla30uS@dXP^I)zx%@<(7$@cQZujbANb_s4>^6+
z`1!kgzIgqXfFq{`?}h8zD_(u@ikCn5A(X;6M24$-rcxNa@WbzX%;jf)$UJB6t{lyG
z)Y3R!61mJwmzhfyOBjUUJT>XTyv!GtWu%M!=8|}MJaV}|GEEn9DS9UQ06uIeuXmz0
z#8vqr)gZ4;GD=f>ty0K!hNg;_Wl5x2IazZq%%!T`anrGY_nOn7)|Y5w7(3`YFkaVR
zb$z=rLBB)dh>yK?;?B{D8E=R)D5Vm6&8;rwNp~)y+4Gb}ZE5p?UoX8UK{xAe0I!6X
zb&mM{TwUs#-z*y;PBq3l4-Tz?kNP?din$JYBXG6WOvmpC{fLNA(n84zYy{wHGcv{#
zJlt}^`PHq^4hzgmxSG7ps=hTl-G?50FxRq-Z;slE@efQrVVY1-2&$B1<h%|ip5hkm
z;EltuN~<UuS-O-C2;0FS&QqF>SDu`$GC9xWG%JZy0@9Carc<&JR0_^z#Lom8(e0E`
zYp#EN7uA?UL^%RYi>_K0gS8rd7?t}GI;x9ii>WXrwU93}zFDNG&2U#+;m{ul-GL~Y
z4ClRLxtwtQK~al*q(1x{(haS<MSQQXz4AM(Fj$F#PrQJqd2y?SvMN@9yj&>Di8V`J
zt9e^fM%pRaWIay#EbD8P6(JInHI@t*8f3CZajHYch=sLkxRV5|3oWJ%FJzI;rEo1c
zG)qB#<HrPt-ZONLKCa`qVBsucqrptD*`}f`S_tokz`oyU_)F7)O^A^YJG^%~Udmbh
zLt<oib4!1HN59|UtN@+O2fy=y!}T?NKhSFBaz3%l)0Q5&$_oWCZi%L|N-IqB!sFwS
zDP{VhCw5BA>W3Y>>ubjSo|SAGLWc{TqI1nz(_n#WPL&JFanb({LRuESesj;yK7Yrb
zfAx;nUqAAAyigkCT3Ko#cuyBhwPerDI1U>Dx;7;O=A;(4H6<mTrG=I=!`N}P8;FuQ
z41wJ^FwYnAbm7_6z&H-<4m(mR9PjUWd^mzRL@Frh7{`(8t1Ds*JRC1f=S7K#yOHU5
zPcR-6;_!V3RBo<E4*Q;RIg>geJU?)p3tyB%?}T0SlpUPr3#kgNE>xGnb;NGK$3U&{
z(>JgA&h5bOfBXs0ZhwdW;Sc|q^XbBPxT7`C{`LxSl|0RODnkt1-QDo)`5l5V41ud>
zA5fMHOMc+)gYV$p{gj~#)LIyKp4vw0BD`G+mu2BN7jkXvhk-CeMaF9*r^NYmV)ip>
zS~xwv<NWTP%V{RfiQLpEE_NMp=)s7q6>Ut4It-IbrL?y3nKh(M6(>twD7mmK*|;;A
zDJ|q;KrliT<wjL;<Q0ho{eHuM+l}Up@mF8pjzHM&@Nw8W;Vmi7N9`PQVwq>wF3X8(
ztZP%BYwt9+lNz?p7jbl<qm)HEW~aq~pc}Ua$ppQLqNb!`Hk&?Z3R*SCxRr6Q(y&12
zMnoLN^1DS=g+MVEG>=3xy>Slj0wQ`im!w8OP8>e=rWalRJSp+8qOA2+X;p)nsy*m5
zb<T1tfz5dYt>}Hrx`CM?Q(YPCtJ@x^_Smf|4I{0A{(SyD#Gu)DgSyEZfnUk6ZnaYC
zLTwh^E2}QGkmt$J9nD4eb%ZIBmkTZ~SedE6TC?o=bsS-b*4jCHMM}aj>P+25aGty@
z^r7SQ_>NuQk>&|6N~{YuKgD^fi$=5z_@UQ%XH2BaJ6gZTJ5QZ2)HEZ*gd6TC@vt4x
z&gNt9b53+l91V$984%cPj;1*b^Teiy^bzsBer-ueEsA`!LNhyQ<f)iIU&YpdCFxc3
z5^+MwW_r`K$ZV}@_zXgktZbu_aW2xFH?F}GgT6IYj9iY#Z}X`B+Fc*Nc*Zbv#7-gd
z*bQ3fn;fduOmE|}!?@>We?<sAeYX<Wge4_PDGcMz-1!#9{h-KBhz$D!<JBGG)h%({
z<Gm6!vmKOP99`ELXQ8m)9mrXI;OpuDhh<mPYDIFT=FIVU<g2f~=FOY8DiQRK!(rs*
z%MaM?uLwhrcbZ-o;(!mm`9@gdvN}*=Nr_JROoHR>!<nD_^i%%ekN=#{zI@<e(Y>(N
zM)P3{R?4m6n$Y#?vRbNY&V!GL2=hF%%t;~TQjk&+3Q`NP6ZZRo^ZCNnP2l=!&+*|M
zop(HYcFTS=LUJxf+%%4bG~-~XLSPt1LI}M3`fDB^kDQK=>^cn{yE%;9zkOurA|)qI
z(?Z_`hM{BV0<Yhk2t&ZtOg>-K%qJIWE_B}U{A$O;oH<T~T#Wk@I%EI9yYtNd__I&>
z;fq&%bo-3o|K4}`)6c))m@^-L_d6V(JyY~8C!7;<Y3%n0Vy{Z7yZDU5VIdtm{>eZ4
z1HSWHKjQO?a3{jEmm@wp4&qrnJW3_K)$YCzp8bB$)&4*q9NtILlF0Y(^n95Ur~3y^
zkB=;~*<Cx8%&$#~ILnt<5_wjq$XX$r^W}RKB(m|9N^N8drka;TN=5w>nij3q>G|qi
zu(4Dj8Jg0?o^E$de|_^FU*+F_eLK!T>@|YX?1xo`rO;@zbhmLgT0i#0(e^HA;Ae>V
zu2&5(=0Bi3VoF)mo>M%bAGGVrljc+7@KGJ*8m7l?PmF%O;W~$6-Bf)rIJO~C=YX<8
z!G<XKe#HBZl8ivqMVyN|-%m3#Uz9W81EK3w7DuI_CL_|+p>CBL<Al%;RKUBWGDH`2
zY+|sm2sG(c(NHyj$C;y=x<;V~qVX|GRq|L_PD*NOsL_#inrgXPEizhPU%d7=r4)UA
z3gQBxw**8AC0$6%OiGK&ZbHX$nJDuEq#2hoX`bo(UKzhtM*z*x8ZrLEv`o~JiT#L+
zJ+2#&I8cjFB;rNms+Y@|D2@4aPxOJDW)R5ln>pd27TrXfDL#Cp+jZnLDY-NBxVpgn
z&WeIN+;~gDo9uK|hRD`ouH&=1CVZ_*0C<mVX%B8IP$wO?n2}LV8FAVrLQ{hobY2)j
zq&BSk>$M<lmPFW~)ue@kaXczP37Hm~txjW^Ru>2z2v13!_Wo3Ew0xUK_1EtD@YPHD
zE@;728{=V5=mx-T1!do<mFw;G4X>U(C&Ym`?ilw6oOqU$N%Q4Nf?F~!dO|lKp(hSI
z;<#trUoi~BhTi0&MVc2?Bjr*^Mg0SYzN0judPUIIj1<g>sx{4u*zFJ8=8E${q4}Jk
zY2ZvqiBc1tb2zD+Pe5>@JeXW4re<&`H>S~;i{lrc-}BG^@Xz?{i?4V%UN}!PCCygE
zIwYukIl8Fvw03x`yn&o`Ji06kC2NGRG_}WVX;Gaz&~-h_<;<&_j_-YVM;ATM?rzz2
z0cxT)q3e2DYn+ZprfFgrMxysbF@rZ#K0V&Q<IPuJbN}ugT?{xf-LB&_RfayYA3LT?
zL3SPe5E=T&bUx9ANRi4UnSFHB(US^%ew>)Au!}vrUB~BdFWjG#W}ci1Qbt1DA;7y!
z;lKIgKjH6v@B946fBmoX;~)KqPk;XByg8q_9(roZn)X>5Wlrp0JXhqgWZIJOrO`s<
z^}9#@_#gjMXnX$kpS|HnzxO*_jfErwzx~PYbMrXy^*{P2^vw~R`X&s0hZj|)P3Ht=
za#nKT@!`mHQ6gXyC5(&r#29s?<b;+Ix#*aomZGR_<H=YITY5$L9k~>f4x(pGHCt{&
zw@cLmw_$FZULZ!o?JeW&9s8SG#&JiPPT%Mm{gw`dYh18G9b$g+rDU8yEk!p-bNBOA
z=gPG;To}}(#ziyIY1EpiInl%+ozCNHQFz67YU)sO*}@>~P0(nB5Ukr**j+bX>s&=U
zEb1B+s|8EWDz9-yGFk~Y)!4q`Rn`)HR1!zk5kL_ogEasjlnfbcj%057%d)7XvEQp!
zewmS`)}YD-P)J=NO><ggb&RTpAXI8niJ*vBikTS_ZLoJk!$KBXS#YJ`N<dIXI%wJ(
z;*eIzmj}*|Z<Ui_t1ahDS{6cxlqQ6}$Crl7XM8>zb2V$=#=KBVA$C2rLWmw8I;%)M
zA`g_DDZSvsKye+oPDxP`$hqj8Y?*O+qUM<;FBaF^kPx<eqBB#dErO_1n>V{<Cy<tj
zet$=8LCHuY%H?Y<EV~<>t0GTc+)X!Sjh{|NVilE&8LvZ4DV6AxN@hAg&|0JGI$dKx
zPU@8A4AM_&Mna&LOv<VWUZK$tI$9NKU65R8>7uAmP~u#5$_pVaiUKrIsZlF9395Xc
z>6#JcS~MT<;hPzU{JLJlFtF=Y9#vXp90z>Y>vL5Dw5p!|z~OM<*|TR5B5`+3H(rDD
z#3eDGA82{e$XV}+eNX5{^&>FyY$;hgOMB*%eNUwi>GM8c&eXgxbQ)@v%Tqz!s*VI(
zSrAsVP9b?8ypp~aEoLc|JY7g+=8}{*;{*L}PwWRPGKD-}n3tqvNAcXB8-MzXuldto
ze9irN(&ndC3up?YQV5}AO@ghZ!jiT%89VjWaL%#J>J+GB3TVsBGS5U&!kRcoEtQ*n
z&nF*TaeF=D&6sRDKZ516JU$#b9*<kG|9Cztu{+P~#vW?rt1my}@$0vo9*<m(CuHoo
zzPq9pl{N-fna)S!|73Rkp1RCj9*^XtW8)}|n_cADK9KW0uY1qCsc_dj_J@vL2>j&B
z6Azb}X*OSkE)n~VGnG#tF8srv{*>b2@BG$p^WpFRJ`a}*w;z6B&4sEX@?|EMLOWkz
zS*Yg=(M5KLJ>UDm5BYEZFaKAbfB2IB-T&dgp!?vvbVH`^2gbW+eEF0AiE(#eH1(2D
zsi`Omm=$fQSv^W>(?+8qYDz`L1y76`suY4Zes`mlN=^&8m<y#-ciAml03jK1Hq~NX
zzq#GE2J@nFa+K^Ex{hvlQ1{xt=lbOfu3x@le|^i?cf{71kMftE$A8!Lt%`wG>$-dD
zt{sAQHZ~iTT6JFRb)kt9h;4)0r|E<fPuK4>7oaT4xQzjk2p(EVq-inIn!|NHEiII^
zP?Mnvtx;??zDkrfSWnDPUYyx6sug6a8soV#bv$Zen)9R&eG|k5B{b&AtYyJ>9o}hY
zT#I^>%Zf(Who>QlnQ57EQ6W(0JuT<WM55Vyk{Fd#QY$GXn-!^i&505srLMX53Lk~u
zOoP_yF(q7`sriC0jX?%3Sz)KfN=}4ydf;^bC22mRK%35_=6R;pOsWI@{)RXl)F6oj
zb0bRSG@VGmuv=mHwToWom7d@&+E_&AUgGfLl>yuu-bNo~Nq8#D`Gj}vi2&&wsaAEf
ztc7`=RH|82?V<h(J+)M|bQNKG_?qr;qwt)Kh05tQF3q%fZ7}^Bcx$0SPAF9gpE(!O
zJZ<tIF~5doRupRGp{NhdT8t@2q~!(jqM78rSEsblK^%<}-?Uh4s3(I-+5qDQ6hyLy
z<dnsD1?c*7QD4D`Yg(K}@YQ%CzuCwDrO=u;rG}0by%9G}XsP!Z7{-zP)q#d1jywFg
zqd6gZ2e~7yn{7~L{yGj=(nL;Y%IqkoXf@ZOE_6*J)t7nJ&76X#l6vK|<l0E)2fyVA
zv;+vT8z@aw+0e9zciR1sE=S4)xy-bzfoeWD=IO+KcSVRDr8dszg|8kHzkEo%eq4C>
zIB`7BnvmDjd!S_XB$zJ~DQDe}oI_k?7zQe;)mMK8Hj}g9%=@LPoUl%6E8fMp?FFx1
z-SX_Zqc`H>WuBPNXNI7oljGx&k`p&KS4e}`U)|H3CoL156Q;|=o7Z2`+Co_p)A2~l
zmHBe!92)zp5izUcJYQ(7GVXRz8|U+d>3C7Or-;V*`T(v#QNPGCU3fWm%-4-CLgs3S
ze0aC#*<Q_|{_WSEFWx?qbLDb*B<H{=*QBN6^V;}-{n5YS=fC_JKY0F}k6*sz?vsxY
zMrx@<A1FSvoG-*&NT(BRp2$AprSZEz`Yk^G;2Ez!`+|S<-~D608~-;v|J%REcyq-s
z|LxCt^Z6IV&}*TX(k2CzR><>0os;b$4VAw=ndJphbEx2hk=oTW!LR5FLS7af6R%_6
zRo&!NiL#j)^&=8B^etQ_ayu%LT^k`ru5WJHJ$ptN2X3zJxO?%En`d{7yFtnI)6C;?
z{zi}Jw?hV6BhMFdnu%SHTP1L%YE|wX6-P^p&eE&dKsv2-)8$BAW^z9ex?YP2aWt=;
zt&Uke6`k1!7inI*y;>_dO<)9_*bQnuS_|Sm)#O>q+*4>hHEbsD)M&=o|II^c1=rMe
zH_r+S4}Gu2M4C+!rdFM;Efz0p^XQU|sjl;*5F^1mA|T6!)`|tt6w|3IReFWcRkl{h
z^JSB|wOVQOQGMzuKrz`)P}M-Q=-$?lGSO1d?uAP499<Por9$5$WoCLja(R5Cx>r+Y
zV4GNTk-R9WvQ^F6Zp($~&CjUsKBJZe7XnMl_%PE#Q%5)}o<iS24OYZydZak@6Hqq3
z9&)v0NRzr73KS}7nh0JDdC>XooF-xyX>lMei5P&EwVLj{%AK;9k9=dk|B5gSh-G9;
z%e3X4S^<Zh6<$g>tD<g$vMD(lE_5pGLO0ctChazT#94?E)$$4qqVqVaD-b)Xcq42X
zG0X*A@T?M69W4}=)0viM-4Kj07JM^gDk>zHXN9CGlziFP=iq4hqH-%INS?mcqx$Q1
zP16}^8Ht{L9KqJO@^}iXnHX5qx9?aJJ{zcHrdC8MUPq;hqSUN~RV}cbDf3wox7M~`
zu$(h%^7TB=O6a7jPJbNe`n~?)@-$oBOvFTquT-chkt%BHvK9?hxg{k}l(gV{#QTA|
zWQv<;&GF@9;ZMFi@%mxnIB9~a632v`HPEaz^<Ht#5&IDzE8vt%LgD!MNJHqyesfE0
zt+E>j8jWREX|`__kw%&p+OiOw(2XO@vhZ-se0-;bLyvHCbw!}@_RUv#CtTlLYq(g0
z!_|Q$-ShBf!Y#sbo=KMnTA4W>&(xymRcnrJ=t%QSh@K@E&gZk5#JRxN$0IT7-s8Qe
ziyF4G%$XP>hrQ$ZQt_>Fb2~7OJs}ER*U^QJL)Y=;apvK4q$~@M_wUF#v%9|KG*y22
z=4<}@Kl_y5dv(W;|N8HfeMeaq+O$xonOQ2u?C9(1g6s$CGI00&`Tvi)Kk3md+0yl}
z$GY8pop%wDZPlr|Ey*jWfdC1dNK?%;)JPg?DE%(YsDS`Y1VMlZK@E8c61<mHb?bDQ
z85t4YeBJJ><zSinDGCymIS_DaPx72>+&%o;X0~jt_kCXe7ysM;4gZJ#-T$5+e*5qE
zzy9C;hW+85+xO4B{PCmg6^b`rj(yc{)iF{CLG&h6?hfMV@4A+zu{GmHV~wu%drR9i
z#Ky8%aSg3&sKr3lg&>y`?HXLQp4!a~rrVH%$EipBb*#zmqT4!bckFI%*z6Ashdukl
zf$gTpT1}is$~sc!U#2(ub)A8*kPG-_W|^-HU5{%z$e9!%8;in{gO^2%k`N!`AVo|J
zsO-s0WL-xQ=5-rVm`Q$moAv7I>6Qy({FB<;I<d#)A|BdONHO4=7T0enN{gpBzK$&_
ztxjvA_oO(}cGC7Q(W6zM#09^O7^P4t$h?{Ywm}>BX3fdD{yl3cCQ-tQ$`PZBY;vXI
za*IHHN{-|hFil6(4C2L)30T3nw;L>BkwQU1N`ZA5iD8joq2^C9%Wp$&`6L8VVk{}H
z*f_IQhE=tM9P!r^^LQq&YZVJNC?~>1kppUs?TBGvoo8|?^d*v}XYz6+g@A5*nzk2`
zBSjS1jY6kL3JcmQf>N~GE$7+_Z=Aql-hRfE@beX`WR6G0OQ#|SXQ~@%&XEuT!=VGM
z2vJnfih_<SAY#pwk~k%5ilS0g$3j7sM9N3BwrIC`yTPg2F;ewCpo$-<^Z)3@YH5{{
z0wiQ<(V7NZ9UjsRPUIZx{7!7B<FYWWo_=@B&F$CdMl52*uY}8y`Se6ig{EzUR@TVI
z&Z~VfiGVQrnWDh91F95B(X8(hwbK*U_17%bKf5P*kFy$Wc+>fmQZEaL2=-N#`HFWf
zRg0AULM}pVvqoXH<2CCfg;mh5MZ$YtM***pr(6YUIZM0WSqUM_g^H(DSX&`T*8EA!
zRCysWN}<x!keFOs-J*RqkReoENY;k#A4C{`xnzF$IPvjiV)BKAjD)l*I43_#+d5Jj
zC={_sCV9u=D}<)HS(as%E<>~|d65Hg7zCVD3)fOz5ygASRH)AC#tm#*$NlYAz((58
zv@Mt8k#!oe#?lQ9sR*E`<jnE;g{LnsfMQ-(uICGFW0<ZJVfE~Xj;?JOHZ3S{&auv(
z5Hl$&#&PC!_4JL#shM0d8v{3o9p}qgy1qqt`c5nE_Z>}}_;A+~a2$Vl;$MCLz}@CT
z{^K(_G@MhxXhoPt#!}cEZYaud^nu4@QLaa=3poU`Dx?&M*FbiSB*Ml-_7cFPW8km8
z{lNe7fBnDWU;n@UQ~uds{S_a+f5&?MnXEpt-!)7@6B0PrpsP(;WJi(?Ax^V`Q>Kd9
z#d*=T)vjFQtxHZQE0;!B1*|!x>Jadt4WSYU(Hh+}WWd^ncDF;jo-oNpbO|0GgM>L%
z<nYE?Hf=*gfpsC2L`a@6kA&;Ucs&22HTtWHfz$DksaCF~B(A3?+P15ko6Pcs;Vtt&
zSKCw}1YSMcF~&;!li8`Z!Vixz5WH-7P218`7?D`{1nFC{+(5<ip2X`<!9*ERn6@J(
z2zh1ofqt{W*;e@8Nqz=t$x!tQ{!I`k^Tj~&i=aep)iNkuZ<3-|R<^0Pfr!c;N<?J@
zaH9>bk!WFY+9lTMf?uZ^>{Leu#-W-WkV&Bie^j87sa3R+odBg$ZU6si?vtv60M&rg
z*pk^g#WY2hC9=*}qF>0V>H@-o#@5z))zBHsu)Cwn39B`3KC%*p#MNp{CxuV&3#M&}
zL5Nt5HO%9crp>qxjB9cI4&$U#BLr&6YESD-@e}ct`iI>5MA%pp=KO*&Mu=r8QK|n>
zod$7T#H~<0yD>$<uBHeY=Ftm&RXdcPQQ?F(4XGH(1t<lTqKsL{M8AsQwTdxI^pxOb
z)+p*rDGXI0#I=gsGC8cow2BZ^OZ2c#0UIXLDKMoK+ifJ1BY0RROcrK)Spu{TTDMR%
zSs6mCc?B_$Qb1crT0O6M<<2?@T~bD@rumnNfc{(`r9fLjl&4zU{=+20T*FmBcpNeo
z(Za7ta}~>4*_gFtiuh$FMvp?)s#XFg%SwUCa^M?-b%tE^L?EF+64xyvU9E&(YHe-x
zYE{!h1#_HBK|3j+o3@bZ0x#8;>-B3AtqiB07t0T)%%{`HWflfUNR@3Od|1F>tz|O|
zqF+Ml)0iUKXr^&u8poQbndxZ>;Z^q_NsO=Vi`NcCTx8AVbw&{HhQ4q3?wj|#I}B`k
z(Pf30SmqOcSx{QDzuBX5pmCNGBjdQxHJ0tJXPIU~C}`Vqyv`iY3pbly08J!@p=IbB
zmTBer=~zdb;>uapNGwvw?mEkU+b~X^X-+7UYKKJd>wVWU3@vTfqH+9p{@GumQs$;N
zTznvAPuEf@kDA2?rqc`i!+~b#c(^^#caA(yn0iT!HR{}MdcwHiI@yE$G-1zY%%J)H
zuYSY#|MUMDN<)Y%zyIwCP2py{BV4Ac(iaPP$`O_Sgx}Yd@@iWrFez(rPDFzWMF^tJ
z!pf*gTZ?TQnxUiZJHc$HC?n{qVTw5ua>h0-x{>oUg#|x~Mx*3H-*#-e0oQjpr7$I<
zQ$Q`5dAf4Rf1)Ag9QiU`e$g8JRmFf(7VTQv!vknTn$HxiFinp!ve~ZF8Pmo;ExaTL
zX3j|jc1?%24r3IpbG(Ma7-K0;QUTkx7b`}V_Ft}>bdmlD63b|@8WHR$q2gf-q=eOq
zV&#7q<w<^dJFIg`q-BCADtcuaOgjkQUmHk)<Y&1-r&x2Yy^J)L31h4rPDs{5STl_k
z%V7#$ZXm9c2vr&io|Gdw7Wo|&LHAl^vGsS6e2vzu^g*pPB?VT$$atVaonAM~S6h+M
zVj;{)W?Cjh(ptUys^G(lUnbcQZG$#eBB8Y*mq>1Utn2C23gLu{qXdCyi|aa4s?nR#
z%V&@hicVlt-V?l+jCA8<oIxRl1)~ainMm^$D1^HCiw(4(>z_$(`jzfgVJY42@TRO?
zmjz=Kp%w$Ss-1I=sG_iq!&*ZK5o@&+W!Ep*Y(e)#(i`jByy`eYknuy7N<7!XuGC_s
zX*#*>t*dmXvS3AHaky#+2_d3vL*H-NbOR-Ne7vHTgfcBwOB-A4k<v~qA(B)Q9Y8H?
zDTx@@%0G}pFyw+Z5`B#_GPa0D#CHFPAJ-T`VV4M5-c7+Vo7#3=-m**9bz)hrSZ9U(
zp%vD4;3}FNC@GWDN(>7|8_X-XM}tw4XsS^Z`&#%Fbj23G;>~3ZhqujGIs%oPs;c0t
zaym#EV>C_IiDM~8Qdo&maJ0@kMh`!|c>exr;fI%*@p_fHZA?VJ5dAEz-)_TZ*l~LN
z%)HsqZnh}vK$W`wpz9iG^@hlJy-IgWXF~8YN^lK+U9njhj?qh~PFOu8*^^wo_?(X~
zeArq1npw1T5_C!QeT%ODd}u8WOAe8~aa_uo#zA)&Sd8KG^u#!?EU9pv7aoR=o7*kc
zYR=~i<7GrCMcXwpdNdkk0+(rFiG?m^&WmSRJbl~ZmzBjwjB#iy@5gRGu)W=(+YNDz
zgdhogcWvP>4jnl=#%TF6E@+dPW{;m{vQi@U){bNxVVnqyB#{<0vtMyWm@&4xgw*Q!
zzyCk}3qJoR|CSG5eT&tKA3i_P`Dgz5y(IpXn3$JH*L4`JF}5L9&6?5DK`|91CGO~o
z>sO}cP(;D7mlj(=QEWDB_Itq(D=nXK%5OZKlBAQRs@874URjrw@q9s-!iRVF>^ECB
z{eW#-hGBzq78N5Y2#3I{y4PyMQZkqL%k)OSsu<X8Z`f`(q!KB{(A|BY_yuWO8=-U|
zo_nRN9KhC~+D>k^rm2UnCdawvOM^~Hs;v^qKC>(%#=5r|qNuG!r77%r`zL1uqR1@k
zjG3>X^ji$=G><sv$bKdkIZ%}XodZ@uu>#S_-U}tqg0gbpNLN4<^#w^Bl>J*QNn+>J
zgdjH%V=Ry=0OVSKZrMn*F>leS68_^@$0L3eft|P`>dd{!j8f*z(oQa~UrR0?B3?q;
zpq1K=u3T~AQq%8b(@Wx2A(TZ5g__NvOoP%L-M%4Alh{v!r&M1pNY|#2$YBxDW|8RA
z0$3xw0>TPTs@@z{D57tO>qPb|s%@nkQ0Z*ybv#lF8>Ow-KTBqvWxlAaLz9Vq0@qY=
zO{iQ3DM~OFgF|Uooe817Gm3EeBeZv@_B}}}v=Q-N$y(eYRUZtM(eT<GQgTC!VUfA2
zs<uZuiuclCQtnT}E|HH4!(R85AugR1+cuKjuH_aXU=f8w?4mhF$exk{#<ig3RtR$Q
zTjwC}Ll8d1KW_dY=vJc)gdmYBr38|5Xx(6xBBhm5-8X){vMeW>($||&?IdYKtP3Kg
zfcFzAdX(v^6Qfj{Ruv>7&Y@EGiPyv47$@4KlH{fy0_oMxsT5F=)3t;2x&~v;Xp?b`
zrX)iS3Som3#sz-+^Tdx&3r{aE#BqdmWr_06I%iq0C&Ieo`VDHnGCqCcsx>#?{sy<Z
z5o@)PuIa1TETzO6JnOQG9opzRzCw!Nnw^n?bXH@Gs9Zotu9p$(4otJ>e7@3eZn4(T
zbv@B9beo>b^E1f@)*MK)2=7pE*zefwwuEKn7_)fhU%Lugb6%i1&#3E6NttzCPz8dI
zc<<Tnb}YdYVr0MTP!v9&W+1bs1*__Q8Ux#Yz!+UKLlo2Hf@=rbp=J8%3*YQJ2CX?Q
zfp^;SZfN-VJn?f5Ea8GP4x=G23uRu3KEA>3iq@zs_VO?-bgrQo8GZg2|L%AES3iA}
z@GL-E%m3g%_-lCB;k~e6Rxe$}!*;+}O|quq<VAm_Yey)jES=>n6vbR|`OazDrloZf
zIHqEY?|b%#9ZoBjRue;}dz3Cu(>jKsl|4giQi$l1*tQMZ-G+C!JGO^CP1}k_sfxKJ
z=*qL?g6ezpb}uy8<%+$We$g8JRmFf0tAy-CkIo5fgCY^83o#~=b5w5uAc#h5WOLK)
z2BllMJ<nHkmWW)G#+pWM2}NeRDav1kE?AlHTEs-TNM->=S}ABt1jvD?XKAxbKN{;q
zs2D58)3z<a2UId>Bd&v3lF-3*!TXWoHH{N-V~z=<5~_N2m9cLxtg3J!qvcnal@{wH
znXs-Ed{OzT3Q8=*6iI58X?XOc=*3!DAU9&QQm-m@;J=Y>mS5+r(l$!r)t;Ga46DYe
zLa*gxRZRz_Iu=qAKD90aXv$O(pXYRX!5T#~JizpX@>juIYKiF7IkSX?FkJ~L<2E~#
zYiW1ARJ1B%n#wdMQBuJBMSxLtgk)>w{A#VyuOOo$N^-3Y)^()l@ym*93n8xJmtGgK
zK_S1FSPKG!sC+fjB}zF_)`)8c(1|!dVReHZzM?3Lj+4w+bq%7f>fP5AxtLy^D%2aF
z7lwGwq5+6f0J~jJzqygm%ul3hQB!DQwP=~#kq|tsDuA;Uh*VHobTFkz$4uJTQY@<?
zmeZ8u8&)~^nHU!Q`j6X9;H;#I>Z<*SsesBD6)D<C2dq9R1tDb`nZ>Gtw0)HZwMqe>
zBEie;&nO`f%CD^2jk9#_G`!J9Umd5?;88M$<4?(;uW72UkSKA`$j_CNkUu4PQP>$V
z1yb~!CdI$`apaGmPo#OI&4EJ1s6tGEWi6C-CS6ZlpS}=MWP7tGtt-nKxK0<o{U`sv
zI0}qKIXSRFCJ#nijB#=SQX-}fA=ISBC?wK#A%#dunZ{Yxd8BW^Xv1|JfjzC$Y_~gd
zkkaz_aw7UbYit$l7Hrp$Qxexj$v7j$GzL~5nWmLxjf~fon07cVomoqUxU$Rvh2nf3
zg`k)-))|8Lcx~`&5IdaJ<T#;~ss|hNZAY%yXr<wNJVID_x9>2yuxne^6gYGSXJO+s
z-g`{j(jcHFvaI3PQP2`sTylhIqHE-w(MX8Y^XH#<{NWF8XUo^$e9!K%!ESauSj%?j
z@jpHC>3Ah!xH|}N$y$q1MT#<2-BBr_gN&mjnJ~X~mD<*cXqbZ0&~^=~Y-yaM9~xpz
zY(75%u-|R@@Ng$zQ<-W#7M{SS>)7pgm`>=H%d)aAo*X?sCobzk#L)L!Y}12<w9W#R
z`vuRhDh8J836mDI*2Hi^KYzhCO|@KPN_1$~Vok^5J<B+uO2S%0n+nPYf}f;K?qp<O
zl(fcEopYc-Sx55gpQwXI=4;A2&~j6BRyGQZLn|$ijSAOE!PoJf#x@Pw8m8+gACeN!
zZBBKxpe@?86zY67Ecg^rs?-ENfdj={FizJ>OjofYZvw?qH=7bu^@vwga<#d|`eMW!
zDu}1ZeA-H|PY9x4cnu?Zwd<)Y$lLhJS}29M%%rxX7)K1L-qxjcb<;ao0;#0xc4#nG
z3AM2Zvms>>CC%f7cC)43+@bmnuFQm|KTwbmr!06w$c1G*6JjFf%y769Ern~z>x3<t
z5|-+@Em8q%gLVR#(x#A=_-Y{$q8FQ3(IWQL;vg~BLDL9tyN1Px!&m0^sdg45s3}Q^
z5s*|7k!V;cVaA%4l2(@Izd`#zcDq_+X>k@<UFWuPG>qejHkPjI>+dIt!^1dA4EP+u
z3MyNv`ZZ*DTSbM{S)wHW!IvU&%5@Pc1UsC(Cpx*YkRw(hMJ0(8J?nUULvkzQ>eo;h
z*J1vV7Xuhw=kN_7n%au5x9}AI8~Rvd1+b%Pf}k-5tqUm&h$g?~p;YX5iKM7_BWnWE
zeW(F2%E*{4BO}GvL%&qV&TAonoC~p}x1vH>MT!|!k`Ug!bn&c_8|QIK{AYi6=Es){
zwnW-SbAPuZofpohneVoa=gY#&^CP=v<=d||!~&-f<o(+A{O;37KG}w0cSC79RO>`=
z+4g`U_*lCFM&8YwY1+0rp1_B|x~vo*X{-b!r4+fpe}`!tr_(|ViA~=#U9SZ1$tlt_
z4p6K<uo)~v*O6l4dcCmU?ocW*U$2C9t%n~BeZx4dcm+?F8NW<81BXu2n!+^CC=BE3
zNfc5}gd9*tW1XfHN5}<}py>vz25Yq(kk-772xIUZ4m%P2mPF?aNf|yqpGh7RBfXB?
z?^?7qEYp!?p6G2$be46ULCfVWu9>(lY#;8?Ia6X}PMM!Se_~n|zWU}HinIL1Z$9vD
zXNhrPULsB#Zg(v&W8&pH(Kd>$Q}q&9-U@ix<7<GNTpKH0G1UE`5Th5{b&ZC1jU>|=
zE8~W??*Q0sH@v&Mk#T4#7<4_$B-2A#OUzQt%=5&&ct8=A!spB!JaIHKCeMNE<%#ih
z{6$Oj*A)Zvh_%AU_I^bZNvXxS4%-f_VMT{C-UcS0vCTkSMhuG8FX;7(HimB4;@S;b
zTWJp&jYVNyOD>tPj3htTwzy~ov=!`e!4hJ``-qJZYl5WwnMMeQS~kyRoyjF}J)g1G
zv4)kG^NHQz0n=@8Mx&Pz<s4nJfs$F*tI*7pAX3q$dc6(Bdl4BKQ<Kqzke5P`NLnq2
zR7eX%DP&BYF{8|z`oGbJ9Aw^}v)J64%|R%1>6JSmbFJXL+<ae?9qUM8T^8atVYJ+~
zsv;1SqAi7<NKlYnBX+wYnv57D#T3kI3lt%h<rpaKfa>4T^k>3y#8n7RC;?*)?XV-w
z3)WaEEMlbHwj?XJ3RRQDGUcrpQl=5*c8*XCN@t3#zr&nrAjh95Tq#9t)dv)9r4}fO
zv@WlcVytnv#*$J&J0Z-eD7Hc8TEKv_7&V|MEXO~<_5;!EaoSKwH0Dpr^;i3uZ5m4L
z>?kivUS%4Y;VXqUvKYfWN=CdDTB0O)N&zP`@E+|tTqACcLcz5?IfFKVlrqb7=6X5P
zw@pRsM&|j-_4G4Mx24}dh}bYDny#m5T9){ad{9vF)k-T7<SH4<<P>GR^bgV{UDMJI
zo3{hnXj_vhYtBY4$SY>vm=z&O(3hz$5g8HGTia_<rt(`FXraD$CChlC7M>+n@?$AL
zu1Tqy<fZE)T7?;{2QG8sr_X0rKe1^*WkbJd*_nl}zu6FP1NXOEbn^`_&kroqg7=Z<
z(}`xY<-hvfXV$D))|pSwKckAm?W#+`WK0gYb|aUUm?Ty`mul$+*H~f{;E?PA5WRGx
z8m$?+mXH*E*K>P&!?Fg#>Y3(|&Cuh=D~HX1bq0zE2iJ9`(}wYKWt~S60*8Pj^4-IZ
zOK*8RjVx<m4T;m_G1;R_&*Qvsywr}4O6<2CuK}&v7`o2j8Y{A|te~+t1AXUe)UIZ@
zy<>X5fXZxkd#>XZ6(hTS&l)0q*W#1s&^vb4K=vX^y^cITf8yQU*Ke05tqR-ChGm-A
zZ?~in$l74CW+Kw{mapF*`1WDLci(=+w;y_X9dWjiLpo&o)^W3IcsvI#Q-D^nbpoA}
z{a8lEN=wwc2!Q1sOciCFNf1FR3ExXWQ1nVGjI&s+X=<2|HHM*cRPg>A-(U%Wd7AL)
zN<{8SYm~%8WhIPA8OwUTa6Ucr`12p7!yLt-_Dh~$RSa~yJKDSVEH9r#5a=3gx5Ks@
zc86PHjJ!O4LTOEV*wc1BY1xPvCwkH{GBh1+KaiB>ay}AMV1IK5%96BrWkU>P47T17
z7Vq&fiE>gGxv`=p+j*WbW}+STC}Tu)rV8Z98fC^C3v73L5%5JX!XqcMY-5yQFRMSd
z6yC6#IfHgB7=u+J>dP^a3Rou_jfnrE6e<O2Hy5ecvr+<+Q34{5F=9-qC|^y^D>*4l
z>8e7#ib!%2(2kR^C2eI!^?Dr0t<hFwu9BZqYfQJLc#n&j4g;yMri70{P{&&HN;LFd
z_}yV$n677bt|xUjn49l$7fCx?#tSAVe2Dbh9o=S!vQ<gnwzzJOPLb?KVQed*X}#XM
zrGnO^Ah`)CE#mL5qQ{tG9RXQ161<e77hMA~F49U`=dZW77;6HXl}><ltvEqy>Y;5c
zD1&YLI<H??;)E)eq@2*`k}Ohyrt3J|y%Q^=t%5c6nhBwC-5{1SKMR6dDRfbC%Z~wV
z8=Pt7_iJr+IatXYPp{(EBm!A$2ytauCzf><F=tU&#L9NK{T|ggLN++vU<~34@ryVV
z{xLu0H5)`#hmL!r;6(+w3Mpk{Rao1iUM~Wb#F$d3+MJ4eRpJZ~;i2dRUUMQe5)AXY
zF~7POa*m>4FB1O?NXM**tyutS`ps`&RLz996$PsxCU`kt`29~We11HmvSMR2*YUzQ
z2L8Q&cE`W?%?*wD24@}ZW<v<@`1nFW^ZfKg-wkx#o*zFwakuOF+mBED%@04Z)fQ_y
z?9fyOel1dJYdgn?_lkbAku~7GhzgO>NAN4ED0W-X(nPI^CG+v|i9%%8H{9$F#2gtm
zTPT^A;|qSC=^BS?9oNeXIeHprv07Xg-+pz^^)iaEIc?B-=6pO7Qe;h$-+y^wye_1q
zXq=^Unxa<j4n003tjg>*JG9j}<2au#j3L#}$zYTS{2FI@S|)D3eqb3#`nIJw!#Xac
zkcm|+>@@fs7+Qx?8nZQg`S=IEy7`8b6s_%uIbk*(8|QF2p_EwCvyT*&`5*tQZ)t~?
zhlhKHVIUVcU(PH(V4UT!Kj7NJZOn{5v-lOOEUmUQR%2a*HqyP8gk0IT3NjWph(3W7
zNm=m1$xHH}&~i16F&LC!<q^(~)ec?2M-frZtLHi|jMKz4&!j4b#b_|5iZF$`YfZ~(
zJ~GV{nnLfSbN)-7Uk4q^iLPr&tz%vaK4vy;M=^!d=}EGk+lFp?$L{tW^E49XnY>O!
zKXZC{;)iKryX^@n^5yXns}0|M_m}jW-5ZkDwHvy_Jqls<_rsP_B;e<Id=&VM)-o5$
zg|JSnG10_~wYp{vh|o}LO;{!xt?8X3N118onkz7m6RPHj>pDlbwgP%gLk<$JEbtg8
z#!ylu)|-f``B7nA@Y5_oJw{`4#x<6v-H?<e=8SRDnhbtLR|kfVnQR=+TEX^43CGb^
z+RREh(aV`u@V_WbyAdrxN|IXDbj;qfJbj_57~<Xvo54qs%Q<WD!N1Ky+qT7fNip<(
zC6fudqwK$>ba!Z7XfB^gFTbVLM#cao(Y6k=yMyglTvx?>{K9tA$<4iXQeu)2sn_fZ
zZEV#ORL4ejndBsH5MFIHF_MGF+NKr^f{1=CV4&iJb3kWHSf%9_gN#W+ZSUJE95SZC
z3>}*snG?+ODy`#siwC(-g(z9!ZQFuU*ruo0hL{R%-wVDtu8<PON_*C7vBWAX@y|u*
zAy9!q!j^Aj;U`I>)T*GhV;J`Ab`MY)`za+%1I9M^EbmsMrQpq#QWk%qBKXhsQLlF5
ztfYHX(apkwh(u&WR;Y%dEdPyq+g_=&{Th9)CH^<pF&xzrSgPhm+ySq_WVv>2RBgXz
zaWxoM^#zGiMTbSp=ni3lB7jqc(T-v@B@~_?FMNKUSl3ed6akCE^p^2-qU0S9-+hB~
z9b}20?o5lVhwuH@4@~pSKeZa&8ZOT->^D81U(UQ7pSU~RNk<eRtj0BvlP7pttAb*O
zX}n0|vHp2WjP<V)Hq<o+zlf;W2PuAsVR*x_r<6D!Ux@3%Vbjug9iP5@rgsjNEz9cJ
z^p?ZnP}wOmQpsVZQHlGz9rmc0rpz2m<)VPIhL|F!IS@;s?;QJ1!j*b!IPABe;M4P&
zpU*Qlhdl>Xcz1itaz0|LLOaFye8JR2#Q7@DqQ12(=ZS6Spf#9M$XRjNHk_V+U|mPr
zZcDe_R*lbwFi)&c&lpvzKy_kF6Rj3xep(ZD+Ypi$r_1VTwBkBWbe&_r?RbcWk54C(
zGMHgUci1uvJz6W#_~Zzc##?g1@DHGRrLe9GVH^qhMeMR!D2ttIv05U!0V#6Zo)kRe
zwD9zD<T}l)A>xCtc6!h{V5}u;hjEF%=@^Dq9BRgJIlj~}w-$8vFL-`kgt0#}efa}L
z6<phZGAw?gZF`pMh0FDf?pl&D_!O}vk^PKHk#(84E;FBxXFga<V+{`vU$e{;k55k=
z3T*c`V*k?}!_5c6d}UpZ7_G5QNAk1W>Of}5DKRfA$<Nr}+1QTF?mbfoEX#~>&{|lg
zE7!GB?kHR?BSno8zM&%huthsZNdkOXEA&OHm9`ti9Z@3L3Yu8i2GUyR>X~_7NMXe^
zt$;<cqScU;!n2To*Mvr6OwE^ebh`n8Mhbw&YQ+1jL?9V+5xlJu-ny)Og|aO^WJ>l#
z)v$l}KyDMh$QG7EfE;P9#W$dn2yJxLI~Zf7yJ0MAk&V(?L(_<-R<}m<b<GyDeTNPc
zX}!u#!#3o0M@f;>>35XN3q${o7zHaVs_r^R)8gz<NT<OVTmAKgRJ2S8f=7o~yB}KE
z5=ts(Yi$j0lCgt`Dzm^g>Or~86HOiMXrp8VlO;>xH7U%b8KbvY-C=8z;wuY$nMRiD
znYM9KlqgaEmz>D!gpY!TCCrr~=;|m5tm_FW<HvI)O_quiRr2RAMdtm^$)@3Fa`s|_
ze4U4DBi$V(I)m0aob4o`v4%F;cEI>wCIb3%eH!Oznod5ylF+7k%P2{<!IttC=Jguw
zoU(91as{^lb#s*)qOPM8Rcayjs@r*^8v;rMi=jfww2@JTF_m3XMY<xQ7ooDm$cpGt
zBXBHZEXrD*k0-{{jLDkLMpP^9?$&|MneP6M`<nqbyrbO=<hVc$IFgXlfYFNiI&#=H
ze7HX_pU<et^Zwz$;~$^!%S2P81Dc{l*M>#7A+Fl7{jyYxq2)D&F(jE&lQTwHtP;Gr
zP63l6o4&zVxQ;XHvf>Pk<3w*Y^Az~<>6z`#4z0i^>C}W&c>egg{>(GS<C(*zr)>sK
z=L^mKo=ayro-ULqw9l?<xJ)xX7M`boTOx6j=vv2a(-NZqsGKz%as1nlXa4<f56s@v
zse&~Mf4&l9;CMdq;k&P~ttEzqn@z{ML@;1f!4$|@#JMrg9FKqG-Q8aS9lCMiUQ?p)
zASLhu?fG<mW{HtjX+Hk+k*DVq5APpn`;P5?N1?E+f#@8rwQO2AY<Cpn*xbG2=EFO>
zz9r{`Uq<|N5%-j~Xl2m0c{`YMmbmOy8DfrDWl*`$v>mRx*Qi}8wZB7s_$&N8v4)>G
zrDu-gL=KV08tFzLHb%i5E0nPq-7swTZ1xAX`yK1)NSr51@{H5}KZ1&e8&tccbuCkR
zq;VS8^z{3Crdj5rIYz?yMBvOYY*?2Ct1_F-o^Er?Klu+HC^>LBKhZZm>l*p=@sAkW
zar5pk(Ou8==||3AeoqOJ%jKE28|VfR4z-&t*2ztDI=}Gv_?dZ{xWC!6T#p=;0^88-
z?<JPhI#x8#AAiJJ%dpv!Vq!dhffBgAxs^jkjV#N|usN{V?+9yPU1s8ZCC7j>hDIBj
zwxjC@=JS!u@sXShZ4DIJbOTMd;XIvdfK5ZDplpLFuw0%cX|3&Ko^D*VnR$vzGS5Q&
ziI@kKIu_(bE~3bvt{2L7N8fEIe&OZk-?Q27usM)o5+7_9zJH^N9Hhob`k|6>gE59|
z<mP0I#=0KUbeJeqKuSaziyrQP1E!SP$`=m9bUh-$9ElLCzCkNek;624sbEtUyOK7R
zlqw7+ECSM@)Z3YKOq6q|k|evqiS<j@TzhRK%qABJzluu!UW$RW9^%e5Xccj7NAxSW
z24fqR<420R(jLC$8d`J;Z`}y1748)}6?_#f$~;-L0-Dx2d^R#^u462vYCNuApuidg
zsan#CF)|(@2Vu<HMjR?hX1<rxNvMa`qHQn2$>J%w!MP5V1dbHLD(^u^_;vnOOZCs~
zamKuL+{~+WS8H?;O$Mg=umNj3(dyKWfL5=GlF|vxA&I-9D6ksq99oxZ$+P4n-1l5k
zwF_!-h=d?)inbGxsS!<$a6{^d0fBny`rq-Y8Bv-rt@v?dGqj}Z%yf~qEeSq+x8>db
z2ER;vc-W&=mK0Ab^i@#w6xQp7s1&WU_@MdzyLW_n;d1>WP0?aE8!v2PBx}VoPxy5q
z`&Ga~QDWVll_DV31iT!BICyF(k<N)--*B2l{0EA8lw5|%uUI2I{gevl@j{A`o0}au
zL@w)$rm(Inrx^Kge@7v4K8?J)zmX1R3f%0sgjhJ8&+K*st=4SYhLj@CYSuOL)60=1
zDSrQWK~V@HVH<<7j;`O*`YS&_kF0+zSgZK{VMol7tP;)|mh(jND{)=%KB9FUnI`E{
zI*g>Cl2KMOuRk+RE!*KcOc_u<66OU6lGM6f`FMWjkH<%%0l#`&H}Lb9XBO{y|Nai8
zGEFaq>oPAiT}#)sn9<|58#cFh40m^QT`SJ9Qt;y_0#z*qXVWyeRzCk+yO270f(kTU
ziz*tdqZHB2lvX$*&CMOd*WZ$+30pMX`AByiN!rpl2_~{m#sXqLG`L~Uupej|aaNiB
zKpa})G!mv!98JIA`BlY09&WMQy~GJmSHk=PSnLo{IdOM$5ABxgmmk>f4z#X8mBcz-
zF<nR74D1ejoYNe31Lxy~7%~sHH^lkEI$qc`4nLmAQ)C!KcVJyhTql{$YlmxkaxTp4
zS{c@g#%h|z3I4X-5(+>@$A!b)JKV4%gh1D}IIS6mXC6QOfyK{6rEqPJYkG#hM;EcB
zG;NP-TWRsEGt=>fd5v`araBB1^E6SM#~RJ`dZsHf;WJ9wmMLYl%7m<lC_-GYy3nXX
zT(V%#bH@8woD*6qf33vRYsQ&e0y6Uy-@Bhq808@iG+GfZXZ&)hm3*e_w@CcKLX#5M
zUhG0S*ZGp<2`n*Sl%lD>V_lOV3yBgX-$4=-wd5DXj9<hJ;+K)=CGbaVNEIKQBNl4A
zG6)`6Rh<AeTG%+D10qUrxxtGaM+kE=4^vpwtEDnm9B7hGK?qT@Adt3iDdOBP#?~95
zuB--2({EVM&%~EMpxUn(gIJva*5$%79;*#h2_m~`#TV{p;<{jrBrzJT$WfRE;7NX^
zafawRtd>G^xjd7@igAqunIvDe19I#2YXaS1JIB!PNO^rDMWQ6iJH9gRE49+c$hwAK
zvsC}=o>CGy1&oo;MS_DCbfIQ=)NTo2t$9o5`x7>xCX80j1t_Q_ILR`wV!_iIqdOtF
zg(M>#<H#lw*FX#r)7JmK+6vKBFp-o5K+T&^kP5j(OfgI^XOz!Ovp{xAhO~J4)>3>R
z%&_0~G^XH}k#_fhG6QH&@$;Me1YOwP-!trQdHnPv&K9<P!_Zpxn+?y`6XWrP)Jx8W
zRhncBwsk`5tmxVp#a38qmsHg?l5=FU+2gF{X1C$_@tNcCOxHAM4gIF$u<cprN%k6J
z2{ACufpxmF>7^)dn~u$(@ykMpB208f3zm2ZG_GM@BHw-W6<zDlIgr<t$w!XkgfoiW
z(6I!#PBT9}&z$DWngUrV{_30eY?>ZpCq7<get5Lp4G!06bn+}KoR1flamKHKbq?$g
zJ<cjZk`7YNnPpvYU4zz&rqQhH&ln;(-P5=Y>wINiPMj{!{PgmL-#tC!x}NQ(mywgw
ze0cXj+ghwvgmp&gLfh{^TT;vzr@6b`veYhwb1k-!5Vc$&RP41gs&=AQ5*b#>VadUx
zO2+jAr5z|m5lh`;i$m99pz9hA`wg}+q;<xB_zI`A0JEGDI<65LyRuezD^{YFUjx>Z
z0<1<kOEw1mcD4L(e|}Xl@a2c!k?wBU?{A<ae*XLg#S`x1A0cNU(YcN^y`aL(JdKiy
z*LE!P#9$29FF%p89LkL<tZ76ik29Ie@kf%1)&FZqDRXoG4a>5yo*&t4w!|!OiKQN%
zo9!M|B5mVv%|MEw9x|euou@0?{T<uG1A2b|M8vT<9LVKsN-CIUkSb2WG+nu#zK~O5
zyFEbKV{M0aO(g?Hj!&O3#xQL6;xG3jVVUW=p7Z5MDurRQk(*@jgk?lqhwC>q!%koo
z6vDcSR>5`XDqKn_3%!mk5GD$9g)qIgJ&m=55D5N=+U@CX?ghS*1J*cv^zwC#fuhl2
zW^lsyUsHi33<wEig2swi$r_1D^+|LRuhd9mEh(*}Ws#SBxiY?dLdO|ZV4X+WeuEG4
z{naa^B?PPm4{uoDEWV&B(v#M{z+wuvwuCW)P1Qz2h!oPFeB;%wsf~0{UISa?_oIaj
z7iG2_!-{Qt^!@{7UGUcz$cAqFy=067<9H$G_XO|hhM~GH1e2PIX4vdVx$xtUf1qny
z?)C?aQWCsF#utZcWrLWWe?;dfGh1pjc1&nzWzlD0(l-?|s+7W%Mm)han5X0j)zhw>
zL1~Mn(01nk>$?2M@OVGdI7QC7w&x)yaoxNEw%*=vBVO)Tt%!z7Y<%@ziw)R1hjC5S
zYVeAVz^KaEP~s~vRx}3IHpC)&9JCgyW{oY*O2mVu2&^mTBzE6Qr%Y2|K2J=e6l%Ru
zgfKB!_^YoE+-zHn6QprYGO7#H3AZ`auq8#zSz5XVlyZbL;{<HMe$%pP1y5{mZ^->X
zF5(<2DT?EPI#)JAZB%t^7;7eh&IO$dt`Q>YFmx<o!sG-ovv|*Lw_&&G1^!ern_=K|
zJc;NvMEarS;oS|(xUft!M$1|Yi^S7+y(5H`zTcoU+}z)Bd_K|`hxao+t~8BfzwI$0
z(=?XP$C0oZ$T>5ulGYhZ;pZ<$e)CPsrfXUJ%+(iu|M|jKHysD<_~VbCDasJmz&x(Z
z<HByYrE3gf5{^sPG>qp9B?{#<Od~jhDxT~02Tmi*Arf*Rqj+9MA_|=nr_%}7cC<ya
zx0b%!a<kvkb_Un=V1&5a3|q1?%<D+&s%FeDG)@an+%Lp+LFJ5Awu-wgT30J*lAIb!
zK^gP5|5>H*NkNP?>`20};yhw<qU#K|AKtS+?Ca=Q@;Ovavd0>Q>o>IR1~fH1N%$Kn
zt`&eM0eipX`E{LvW#sZ@WqSNfzunTm`wek?Ax<y2evfMhRQBv|wkQ;dLfbYhA#$D<
zF4IJ7Em{>Gy+>;&FUWYIZ5nnPhhG*>*AurpvHM+~K9ao0uOlHWv~5RBiDmIDF*5WW
zS~+aD#q<NMGNeTcg>Kk!bti(KMGc-J#u}mN1<$7IS=I%gBbVdEVcXIgN9%glIPvKZ
ze_)A;cDtu(8##o%e-jP1*3xxbXbfSQnU<LhwEc!|_a1Ge%kdfxRFp$mSqP*|LPArB
zV()96!#VNdETbfZb=y0(&eX6X`BdA!<Kg}RvLckiaQMLaa^!FS%l{aY6AyPEWR&6#
zLZMpEq?`qylVsldDtdX9yiqba2yb00B_g|=CH*a~LPv{{f?^skoF0Fsv#{OmNHwEd
zJ54c4X6&!RrWhazJ<uR2iPE%`4Oec7!e+=6@grMTZ|a$33?*eynxdqQYii_iOfnLY
z*|HEB(NdWg6gtWx)a{1$?pqm6M4_U^EcUbeZ~mIY*WdH}@du$BmPFI^_?SsS1dKT@
zG}f@$>=>^Xl#*89ZhwPThLXIDMM98t#4Mc^qm5utD}Mo{gu9>>sX8D+DC`dR*mi&{
zK`{!&cs)v{g0^%{;qou+_5GPXA*?_PMLzeO6O<_PRo7yj6muXq8}lkAteO*TrBzSK
zBx|tQVywftuI9<dH|woZh?rUN1_=X)GKxk^RC&?TSCVh3tDn2FHK^=?Qk@D}U|GBE
zAoRYN3DcD>F5Dk>+}^g_KHSo{2BWoTlM?vpL@9x0a}P!eJJ)uSfSM9%y->o+?aiL^
zd7;sod0sgCGrL~&Q(8lc5&^H3SkY39l8vGy_{e)t-*u>5SjUJ{iM#tf!?x!-dz^w2
zS4@s{jb<2Hrn%rWEIx3a7jy~SZ#usD>Os0Ms_^Z1UkgNRT`0zKE*GZ7GxS~6XB50&
z=~_vhZHgrsO$mX+uEUjrn<Lj{C1oe#>6pnGdTZEN!!%Ch2pea3x=ex#kAcfY^W*0e
zAND(*o?cicPn=iA%SBYbxe%5W*CwvlE4$s6oIF2IBTX~N7}Pmlt}9E$rgz;y+qUen
z#3Xj)q6*VA(zXqmEUu!WNhyfCEJ9otrdT+gF4%U<b~iB|p9r&P1ul=Dz<Y-6o?){W
z{gYBCE7sR!9l3feQ;w{wmmaBBl%xngh(#Tw@QWS;?3|@L>`5DmWlnJ=drw@}n!;=7
zhn~&;2IB;T<^9S$UAU@DoMzci+vXRo(O*>zm~Nok9kAB1T+iIxza#YxX5BFCZn%sW
zj!#EiXSsd%EzR9`TrV$N=M%Z<Nb8C%&~5hEt)?5cG@DxyOPs#2+iV5SvpsM)9B3NH
z^?GH^(joBcBC)^5u(`RXi6ZPPVL=;5)7{{VWla(9C5Cu=xW%unBon%h&GrMzIIih4
zX&uodcAEna_YaJxFT^m{S3)!S$m|nl&?KM8)5LDKr|&yr@VuPQ+_xJ{-CTD2L6qUz
z5T`Te@j^HBY<4%~0=DbPs-@&YT1S>;5vx=RQax%YX@;WN?Qg|Log*Pgs7KB<+kRaL
z<AqWRfkZ^1oZ;d9H-zylLPcvaT10!R7h#?)l@z9mXc{QQlnCogOdeAtx)-f!`hnyc
z0XU^9hSLVWjEv7;a9TPUA$YJ3A2O}2?1GrF##HS=jZIFGRQ=t8GFDjIkV^t(CGWf1
z#bU{rBwtHrz6Pipq82Kb*vI7La8B%IIlsmO%lug@P4Wvx9clXw!|i+G<%nrija3De
zBU&lC!<~q6T!U)|;`xO*UpPIA!1CeU2dr(Gq9>OiheC*~>y_4kkBR6#<V@oXd7W!Y
zqr^IEE8`if6r_T+Z7qgZF6Srm=|$2vn})_Xn(YlCMqJxV@Ra(6<^7-OBV!Wj3gmiQ
zQo0Ca$p~TZ)iIKDpyVKhR^?>8=De#)B`vj!AYU6^NpSLxREr{(L?MJI)?uh^Rx8DM
zRSM|(eFcABvyPRfX?h7d3xTjE;=JH&i$SsJ3}uB69}e8VdryDZiBPQ+jFtl0tP!6h
zuG@pPXzN7#QC{tj8C^1+)4YFo&o?g%%@0RNk@<Y27>BC};ur(oi}*5CXMm{%!e+Cf
zX&Tn+NQjY}n}M6V1MOzu5EM>BTCbF~<NoF^xL&U!ByL)spUx<w`R3t)hh2xWhIxtX
z4?9BgIAd9SV4fGQmn*Ha^c%?&ag7yjKv6uuTq$9tYa9OJ-5s-En649RUf2u`kg>Yw
z)AP9|@@X0c*UQ9<_oR@pZOb^XJdK8FS!hhbD9^|9%D!vplwz6|)|?qb=IJ`qI?MCb
zGZZgaXzvMmrE7a`Zw{QtnwFUp<2>S;7Oe}NYYCy|)7Tc!_<1309BrfDToFo1uvX(*
zhGF0`EllS#eJ4@LVLFlhjB6T<6W~)w3&j{3yP@rStQEdV3Z8iyNl60xauJGQX<S9j
z3oX+%avAEnhS+sX*E7?2CHhsIO<IY&&NzuiZd;nofB|yy%$>z4P(dzHQU9{8;;$+O
zZr*=`EkQsbX<_{Mw-hD5&$iLD6h6HF7Cjtz97k?_B*en)-3R<UlCp>`Ow&pGw(T(O
zK!ag<{*m?iNXF538&rmu$1kLuxVifZV>@zO@M-0I8F{zAqu<@Ijwi-><nxywxL1z0
z?O2zQWty=~M>lNHw!vSIn3SbeY%JaGz?=g9^vLz`vk=d6WE!uycHrm7CzjQ-J>0WB
z+@qDouUEeL_Agod$~e#T+gpbH4JrkmK7XWb8=PwiYam`$9?viAZr{_o4W%TWo<Gs`
zH@I%cJiV|^S3bP^mKZeCI#o)cCisL;CzOUb&u9&8*Hq<E{Qv;~07*naRKFor8|P@O
zA*M_z4gn(?zIyi^VK=cZl9FhR6cep;7z0gFxR_at+{~tBCa0D8^1{8Uxdk;K$7+RZ
zTB5gvITOpwJWYK0<8S%-w|~ps?Ursh)ETa-R7wF9Q8F>qR7uH_mTfo5=%F-;ezQXv
zgVl}9E$b*i7NM($fW|tD3>scDAWThiQx#BHq@7;FX(;NAQ6SxeK(94{g0`J7Bhp30
zkia;eSYxIcZiH`936Q#R^xGXtHKY(_Zpugq;xNetf)`O=)DGKsZ13Jl(U$`L{DrWL
z<f5_0k`Q=~at$#=l<hGT+G1GOmA-8RSdxNJDmS;7R$-0C83nfgHB0r+?y*J^8x{5T
zlLAUNZ&AfDtZzw@DSG^Rg(3%DDH0Y1S@<7yG(^s%I9KYRr2DB?i(szAPV$>qQMwVS
zYGCr4&9x{IDdsHk>D2CmadIg5c_u9rT4~zWF<wV}Sg_7;xVxp>9ca3Zyl=}y@)Mes
z7$VJPPfF{beB{<(FqE(|ElU-xW)TI4m7Cqb=_=zS)ilI)t%ZP>L#0?k@HlNSZCe3M
z!dJPRFJy^Fr0*Q>-hbfu@>?1$;W{@%!}FI<*lxhK^1JW*o_=H4w1)l7mKZaad7*C{
zecNz7U+_s7C3kl>(rqbO0(rFbNK-CcFEdUh{>8uiCtNQl&SU23G)qpw;Ao9vUIN>$
zB_s)16OMy)7q+d8SyD{=@O<U-IkD{&A1^C08xF0ZGm7od2+mqrj?;p3ju;d6l4%=@
zR*HSA(JU<PTiUH*J6Im>4;)`!7{^t*JxbBKmNkgWirmYsden5SqaQZX<!M_YjwKbE
z*0I%&rfJAIqlzb|glk(l)RSjkS5_a<!v>mxMz`p$LFs~tp*qJB+OI<0)lww2oe&7E
zbu_KG5F}pP<NZuJOZPw-L({c2_n@We2f=M?EmTY^BWR;u#r+q|%gZlYqrYZBpPxU|
zv$EYCxVw8GEmw>gxcl~>5|%UP=bvc@$L{Vu?ao7vq~(fgZWz~<)qD0sPX-){^>kuB
zy%3YfUr$5=!~UM@c;e;sDE@vw(^?DKk^Dkq8#dd9=x6%AWty+FT~Dt&O1OfNTVJ#i
z$tnJB&F$^iT&FXa`68_Imrv*v8K;r!GUHrJ-}S7k=emgE(;0=)ne)phiD5RD-~Rsh
zgcL<3nia(#Da)0H2+870B$tR*j&HvI_bJVW-0drxa^m^t-xE`&Z8{ojIlp{?A~z+U
zB5v5z?pmhniN{Yrk%C8)*=~1l?dF&B39St;muJqWiT&2{aQlIN_dwfr<eX^wtq3ot
z6JZq}y|E2po{5#uEnV@zGM><-(Dwt>^gcOI15oXB-3F9n|Lz;c(`URtv#|!U!8j*v
z+LRdjUhIR2Xv{A`UeZzs>x{KVl=7~{HN08;RAneoz1;=D(xMF7<f>3^1U*|Ku4%B^
zNV^P;GNvYF3MfX^WGM5NLoeZ3ZgWEzufh!gI@9A!^R}_N#*s@vB~RZNmXJ_H=Hm-~
zzR+!N(EWhZ3LjUjR`~0M5CU^Ipk2?VaiFY3J1av$@W;L=<~5*m#wj`cjDpTt>0q=S
zU{H-=xP6Zzb9(%lvW}cC)2~^oe|C>JMhYQ#!ZJ%c)->djNnUa~UReq`c~V^D-B$wo
zvDuOHBIBY0)O?9Z6~)#<Rv{STl9O;qs3V7*K}$!;`$<MU+6wYnn@TrTZyf}xSeMnK
zmXZC?Vw#SkEXv-ZY=d8B`oj&{DAGEitr1AnI%2H|hvR%jk4G5f5Y@(!{mMFBC`sHh
z#>lwzd_Kzv04G7%z9GXrPpA-}WQ;a+oq(-k6dglIl8D!~QY3~LNu}m?=)!Nl{T6Ku
z`<nw@<5({jzWn%+?fZAwP2leS9WhM2yWcWDJ+s~KX*U~ApT4laxuI<=>oU`H9j2Kj
zFh>{8=M%f3rEeRVw+}t1`56V=-fReI=5+G>7ytGLmYBFdY`8mYIX)f9ZNomQ8uttH
zRoE>LLr>QaoagZkz@tszc`o2HCTT9S#u$Y;Ng&eef!uW#*EF=lmaiY~$Xc_#eUN;D
zv_cA;Pa}pz>sq|c#G;8Q(GCOm4>#D>k&9)s-QsNKI*nv&*}J`1oLdPhDtbZtQnM69
z6J)ePn_h0^>%{o-!s$Ar_BY&`j<(&<bQP_gg`O$h%!Z~4X;mpSuc*e#T`^XpjYZ*T
z+J?4kQ7KZ;G);rDo!AVkwkE}h#$YuxO^dBshj!SKe;IS=*A)X{yl{Vihi!H&>muIB
ztVqk1r>8G0(}>ZU;}3s>eiP2DOvj)3>g&Je<+>*>BiHdrIewvWEt~xv!)Avsnak%N
z$T_pW{Xo%{`Q;O*@yNQ)cp}*}Y=#@k^i1#z^Z7_;D^orPuGb4nNo5twg;rU76?*N4
zfrs0BitU(}nd|dUg!#;-X)sFT(PY2y;lo#KcY8W#@gZ`(oH?FPpawR(JxVKl%(!-g
za}LJ>O+zw`h+Rs?DT8S`lrn@E2{{q6Mt6JmB&O3B{B*)?wh%N5vEpdkL-8z^GmUk8
z^X`L0y()z@?v28?Kisq3-_mV%eD$>@ua|0#n~9-te))o%&h*<GN{n2lnVb@bVNaV2
zy;h{8=!XrTQ)I1Zc%_vbT-92YkO=3P5^3>5-~)DdA2>fJQMhU?Y=soiO0%pLZ=7T0
zK`49-bgp{)wLx2ppRT|_<D9T1awI27>eJ3)loSUkcvKBoGS;CoB9y{_RjBH-7Q&*8
z8MHNW^9a?kkYX(&GF{gTC!!|gefsn}x;FFvtG}S_da{v}vFr83_4yM&y_`9&iSOUt
z^ZxF=Fbxn7csCrReeFHQF2odx>&naYOlx~~hkHy3EX#t$P_(9Lj1;3j5_6^-22u(Y
z)oTQ+8x{WinUV`C21?QR`TT2^>Yw2wx67F1z!z|k#1Q9{WHeAMSPG4^ort80pj%V)
zV$l<~AKFMrnJP)XhEf>YffaC+1Q2D>rf4LeqtZH4wa;o+jA^7zT&jzvLX1kzl6^7F
zD8J%18=1fMJ-2u7(MCeoO7@H|pYhi-{o#(1BTd%>5o5A!;LDlhT9j>xVI|Dx3QH2v
zV)B8uZ~6MG2mbE(8Flq<$($e>7|C@|a>$h|bWCe$oD<RS>P3`F=9{nYaazWGyPI2f
zyDjnjgkKis^O?SBDJpUE@WADC;&6Llx7%}kxga5Mg%}gVc1u{LmEJmypFMXs2R1`X
z@C$8cQ7Tf%+}#dn6PZpIK0RG2uHozZ9XFenbvbc+yJPx%MCpQa(*0Bl8e^f8K&R0s
zcCBXU9N%{>PZ!U)daPAqM=e5#bxuUeebdl)Eyik`f)G8s!+XBH{VRS76MsDzo*w_e
zr%!*tfA-wn?Xk|$8;uVNV-?oG&93M6!+VTriBY<y-Y;A)BLfzz8`iwweL&%eA&|4U
z1FV2CfJF2&%XDEnoe6Uws}@xv>p0N1jaXMxL<6R2YAy+oc%uZa)0OFbq}d&4oNy_^
zy7C&aonl~}6UIg)TF0RB9!igD8ZZvmIGV=ca#KUAPQPe{{;Fc&?q)CCX|33FE%SAz
z#FghC|DL{WC0HXBLKxZKeZcm6hTR=6KmIMB{`fb%`~Kg9!#z(w{{aJ7rx_Z>bp3)c
zip|X(X`MKI{v&Sx9-4v8!?!fM8yej(E)!9CY=0m{FFD*!alXD_+m2<LXqukFfVKnk
zGGTRuxUgI%Y`dl1-lJScfB!v0w;@j__J<aa!;dfIl6gKav^ulhAL#Zse6`tNT!U$P
z$q`>x=J84<q1qiZdp4T`emtSB&$LY|iDGHy{P@JD->=+%_*RM*GB^7@ei_j^VF<L{
z77*2^GLF90^751dtJup@s4b?P`QeA(QIfRQ+^|RWTYQukcwJ_+61?X;j$*AF_80@M
z8@OpZatb76>21q(p0T>cHVtHXQL~cvDM~{EKQH+C%F_?OWjubyH8TJ8KI2+TNu}bH
z6+T7`;$KEdUPDQmVl-V1LUBzazVB3MvRDpn+e3a0sM2IKPK*CP=RhtZx-+Jc!Y2v=
zQi_;AWw+Z3mC-=1+3kf&ugf0jXc~vEH^BS5ulVla9VyM^H51mE70AWlcHi(Xzx@wz
zeao`UbTzmr#7IsBRH|HvKnN?-G;%#3`Sj&y9^QY$Um8t}1=Cn^DY(s+#<fh#1-t3l
zZEso6kEG=ST0*CSpV<rp%gbk$bwL?R)ATg%*C9dv+@981j1k1IZR?*!$-%F*s>ysp
z)@%C>-LNAkCkdS~*3Yq~1{#ObZKYF2Qd}TOG_r^Xn;KWHt9Vd^g+?nZLK=+`5ya*w
ziJ~RFfrcakybwbmjaSsV(r-896y*IX8FIktBGJMq+O~rdan90rgWO7skVdm^P_HOu
zZAi+YRN*z|eO)IGH#^>~8TD^Iib$?2Y}&5+#CS_8ti*aG7FlkS(=>{`H>C>m(aht-
z_lG^l(?^`M#1QCg$9lao9B$ck134?YzUQm^TjC_4HxKXbsvXnNHjWqnjCPJz!?tU1
zZG$z2@tg$sQ`fmR3WetBdB$k?SO4O#$$sT>In%Tq>%0KDj`<?3vji`K<ybV=O8x7d
zVb>apF&u`T^W=#+<GnBzP$IFlMx(8g4oT50KJa`U`R{gL@!euE&4!z{!99FKjx$e>
zk6bS^o2`~{tTLdrz{!-NYg_jB51<U*drrp}*41+vCpLBf66?C+<4W6hw4Jo^V*y*T
zI9F_sJ3KI$4NHP$y72h($~+7Vn;v5gMq7ci=>kQ<?o`fIAP#;W8Bb57=n1YN2G6=o
z5Ce6_K#byE(7IrZVhUH{+|YI#nr<M+o)|ie5?iDB2e}{rZO^aTVPZ+lVd3fL6Wh&}
zT-ot{J`qxj>$WJ{v96il|L{k?{{FAoZ1#Y`QJ9XO7={C1fA|Y9hGn_p!^E<z#1v_}
zo^G=xsfNwnH+0(_d7YS#Us$e3*73r+Sa$ml*lr-rSElO;s|r7V`U#~Q^15(%m)Y;%
zvHGiskD4uM83}VlH-VySKu<_SpJ**KyB*6q6MbQg3!h&;;=VL&yO#Thdxpau_Vztm
zX{MilK&wo9_?C-b7*C(s0sF3_)3;C(F-CMLe0{egcuVIzXw9l3vB0!OuujB=5y@2d
z7hK!otYKWMEpyn4IIqz%7cYhD(<eTySImbGgw@xU*9KDxot8WWX=$2@?-gWb3ZA$w
zESDou8`}Ht*x!9edv`0<cJg3aY3pkX#t8F&yf95yp1%B%@kK12uIpK2MmxEMS=&%F
z#GJ?q(2i0fjd8^6$x1|-wvi3d_QD<rUMxZph3Q&sW2CE)6S<0NP=>a3VwbFXf*56U
zb=JL+<#eIa<EYrwP?hVl32D;_6v;N=#MkTkohZvoLX(NhBy0tvX{{r!Bjb3i4v`x!
z$0v*im4j#`nmsEC+x2Xl7TXWpefLjjJ4fdlX;o`Q@In&|sW1$C9v>fBmIW0S$Ps58
zZnLM~AF6$_;I^4qBn@q-wz7Y$kJeRKr!uH4IB{K_1L_qqEVm<Tod{r6gGS<8U&Dsb
z3ay*k8PI?uJ4Z?3O&cJz&Ll!YB_jmsI!J_cW2~t3k+y9j)!BO|h-9tFIzrta43=da
z(bkX#j~~yH?NN#d61A2YaS^yu3IX%lnzfaKpiDhDEls~?Emvq0#hr0o=4l+co-gdT
zTZ}s4gJ-&2s-r-^p_P3MXd|PK7(8KJnWsr^n5w!eQsm+NS6FRPD$C8eX@pv6HRJJ_
zFF*doJs-G#cgr|VJbwJh*I$2N7<!a8^j**8_<~09{{1`R^-S_ByPE^+G}Cnhwy})!
zL@Ak}?-;LF#+do)yLa5&-|^$`exPv<YZOzO=-ZB$>B@H3bGJY6d^xf09ZxS8d{TV(
zu*ayvw(EGhj5JQOyWOIU;c^+7Lu3sZr!7P0h$S+`$g+Bl<Ajfe|H*&vp6~zqFL`ej
z)A5<h^vFCe9QHTl7)UWPY{V^eI-S|}9oBB^SShmI?{Qs+G6wHM4Q~oUX4RTu=;^wi
zwsq32(FjLHOUu8@ifxn;PEMW@Jl0sO7Mpj;3BN|sx*(3Lk|QaI{>bQp&MRq*Bwd{?
zDd9>+n~c*M)%P;4LZfs-kx*eF6NzPoa3LXtQ&phMmtV9-e_d)}NZjA_44c;$xBy;6
z{9x#IcPzPJ?229{KK}Sy?)N(yt4V6WU!I8L6K%HxCFC%(dw?}@`t(QK<(UuP{01Lk
zef$}BdZZh6XxE|I4if^#^{jqj;~IwTmZmq9xDdRj6vM7>2*;l|U60tVquW0aas-v}
z(-BQTH!TDa_~@=<&7wl?Z*G|{FSs%CaQ_vXhi`~sVjPe7@CP>QM7RHd?YE5Ug0_ac
z!&fva!}SY3Txpt~99GupL=I7^vlKbcSLPTQwmpq;XqAY?(sl!Ob0A)i_~}GW3%2i3
z#t_1UiZh+<u&%42J%-=>)jy{-1$TJQyYK%RKTr5&#;8IL3o&I<hOjQ=l5nPB8ZX3@
zX`F^`L-XM;G5wxfbR`kSDx$W+YcH<|WkZ-}rq53tKR<CkpYS0wY%@t|xzWdjZ3aS)
zSYw4IXf#$g0)|NnRT5;eMo~u#PK#iUTu>!q)WDjeRLUi?Mgg1TOieX|RQcb^7>d$>
z5xlReIrAyL6(zQjARH>VNFnxEGi=^W=}~1EZm=Yl^Apo_rENQ!VNaZ{Pzrt962r{(
z{DNy7$~8=vGq&$g{hl=_mJk`IiPmko>$cc(6$G<wz$g(nIVa&-i)Uz!d|u6gezPSm
zBU&rwdB(a1*N9-$R1HVW|G2XWQuO3xswO2-koi6YP=W~;!5^gLNpTfFbp-=yZA43w
zGqFUWTFM<6V_KB$DJ9Bb2o+bJu}1K)Maf?E$|5KxQ?wMFDi_cNYot4oYvGW5<otZ5
zwT|g|<o$<ttkYQ0$qJP-A&P*^xE7}!N+rOPb3{`_3xdL#mO>%s04Q)BdgJIH3PnGY
z&kM^cg0#MM#86!hwKE{XTumw>lng!)yu>QUn$<!9ZN$#~{@ZU^rb&|f4m-4SEXxR~
z(8j=YK4Kb)-F|s`<l**)rftR1100_p8P5VLQbn{a!+u{ourtG9PfP(D8~VN{tt;2d
zmD~F}dS|%49P!Ia<1ClzUI^ZEd)ToZ9KkOfHaot)-_vNznBZa8aXeor(er*U84AN@
zz&4JTjie>6p>Vyf>^CjLaNs=6yi6<Cb!Cl7=zQ;XY`c#3;Xu@$A3pz`r0kZ$e%G_x
z_n<A$=b3q4u-ejfEsb-85bAlR&}Jnj#dQ`7?0&zc?Y4pucD6bIN;S1)iV|>@%7#)5
z$rn=OHP^6eR%#+<Nx~E<38M;HNwM2Bmd(bJgNXD2aSteI3}`DXAFYiLICH99DB*ON
zk|1S?ywCETOI&`@8vRwpK);ojO6h{th7czd3R5D=D6B0s36{&1K`Sno6PM-0&2God
z;g-jvCo03>I$})tknoQ`GtCn>`z^5Havo`#mJlYgcBpX0Y0Y|(gSIGzpDwH=5mUtb
z2?!)(nWu%*<IHx`5%Lu~&$#o2Zo8v#(vF;`GcIkgZHF$7%k@gv8kX}jaT%GOJ~15b
zFpY+EVY9zwwT|`jf;v7jT`t&m!*YHhKK;bH*%QKozno|qY3XY#8&yexs1;p5uw2K=
z6qt!A;MbMR$`WUo_t?s?Pa!hL`m%&8>-i`iY}1I1k3!p8ww=Y6h4YVpOVKh&RZT|;
z3t<hoZVS$VZgH-o40{R4h%>6&NJ3!nU@eW_Qk)_Ch}DuPmy_5JQ;Zy6e&)wN{4K}h
zl_+l{mvO;uZ&`gNg@hRjCDu(+DG~E2Ik>HBr442rKEkl=MVO-zjHps5+mLdW&>Fv@
ztgYP!@md#@{PQ=EOow$=Zu}@|Kb9i80<H1uMAjPP3@Qa_0UJy7i>P|DwB^cl!p~<(
zfQyOU?boE7nNBZ!dHPJdyTxQnyV=q<1I>1i+uowOBcDF~0X6=>rr(g_$Tx=_<9Nar
zBcwu|(Z=Jnc;S<eOxF|F@k-Ne82Xm$>6w==pHW!=Q@fo+El<Zsx~7qXO=V2{H7oVc
z@A2!E&XtM@&XtfUF|HUz(PAYP?aM@tGe%X(Io0-;6DdaV5?X_<cEpqtS;>23Y*Qgt
znb#!3*S#3(ds!I<*i0$D-l(Mj$T<<iM2vyuJd%81To(#aNOI8=gLE5GkQ@MQEv*tE
ztw!R<QJJtzkOSmEQX19Wyuqb3np_OoHjG{z4z9K25(yztkQ@_h8&b7W>LSj86h*A7
zw7`d~k?uiUJ;SEqU;WF!X7LL#W=zi5u0i)L%a?_=afEs0n4VdD;?t*RhD}f3bQ0Q@
z0+-7fT?)JH7Ht)No>}G@>l|IvGVFI4XYkV^xMtwEdQ9U`WX{J6DMxN^xBUF`7eHJg
zo1x?WFz|A`qLjl}McX>Q{%}K4hIRGa>^F?l%-&hF0cWH*%DaxN71xXCqUL#_X$<?J
zV-1m;O;3sw&yPP5fBFpHyrUG@=7yYLo;{nPq3bOBO-nNjZ1-CpKOZ@rFYM<7?J&@2
z!#qW_Y3Q1cXMbj#7xryW+c-8uBazRlj%%cMFBc%P*kz5N+cTM{b1eoCwm>ClmQq=l
z<y9ji|Bf@z7ztf_rEY2^G?=2abcKwr_E;Goy+OGY^3Ep#S@Al)D8Gze`m2h8rX3`?
zOBJGSXf}@V<&imOw)-t#o}Osi4x<Zg*P_cF??n{m*HLEHU;h;)1<qeSvYejy!|#62
z_uqd9#nL(l7Ut!Q)()jI>-9)1g+KiE4?KMJ4W?~D!+d=qmW)jaV_>?TF|McSw$QXJ
zFF#{SA~ikE!Z=D5-*0!U-g7=Z6Le1m;(EbwL?=(tnd$P3Up)H{-=a)M+cfCip7Y1w
zVN1fb1Ag&L>q@!I%*(>yKv~%&rppDb9m_Idw1U$~Y=9|oDG}=o#<pl<n8%Uz@`Q06
z<K;>yu)n?Ie3@B7rd12Avsj}kS=!zoKmL*1?FIvk=OdfLfo6Nl^Oujb?SO7KsA-}z
zpi}0m4gKv0qMtZD{>0t8ub>oM+oMxteffmy2VA#DBv%@(A*Y3Px^j7b<a9Z+hDao{
zM#VVK93BcU$16?Sv(XLHJd#S0L#z~v3RvfAbaRWflINXcz}S{j1f*o!jt~Qt!Bu0W
z1E%dz)=8YK0;LKm2@15aR*I`4Om=0ceqK-%oNGwYV_uyJ#@4ZndP6J&s1nd6Fi$h&
zg%ng>x`k|7e)#gt&7ESOA|GyVgcF}1Ie+=YryqXD-ELr$SN5vVgv@gOiOZ?5zxzsb
z80*UM=@GxqG|m!IrtkMcyF4FB>zURVhJMF+ymEf}Od1Aa2#h5Xy=R$6mYn}TEy;g0
zk6&jd70FqGfvjynWg)LsQKhN}KQ)IyX$cvUj)5qqsU3upBFY7{vLsc3Q%cvXzg(kU
zKjZw0w5E`<!fGWOY~@|(<cY<T!U817z6!Uz6reh@-`}9LC8vr9jvhZRD5dFoLHj~R
zm4GDz39BuB9#MHERjZe=EhtMai5NXGCRXp6mX&#0@XHJ_zisTcaTo<5__w%b^BU(}
znILUlV;ar7cL%IqnT}6n+mP3l+xuJYzWti>=g;Kmxg3wQ+XKqXblVM1TfRKLph<L2
zB8aUuG|oc+Kd%foH{85`FMNg)B`R4fOyh7A?%qAfWhG}CTXQCA*LYf_lhZo5*$=#&
zuDI5)GcBcZBZj^cx@kp<FUyQ;4BfWHSVQoVd}|bJ2P?Mg=vfxeJg?-K>6(VW|LOOn
zam4*m(R}^DItQF~^1I4FRy2)cx8KrdMc?)O{L?3%e}2RnOW9@;!UoAj5q*^8#~3jx
zV~7}vxK0XH`ysg&>aE{ZP^F+XSR*N`Ig5S?isY>*KxK-`GCnG=4y;(=W4ShkRJF^B
zWi*pe%E(&L^6%%o3Lgd#oO~hWP=7vEarXbepI_%DSVpGPRebGkz%^YJnR(7npHWJ&
zEEj19YvCtvOa(5Kg7OQO=g-7a@bk!YIdXfmBgchedybct{_ZRMW{^OR=@Fe4R8D;L
z-S-&hSl3nBEFqv8%jqgJsIK2)>_C6>fa$krJFv`0o<9D>@p$3>;Q^%$*XtQyefT=f
zbWP8A6`0UvJb`OS%fdR%<flibX=J&aF_qhJJYCuE_rjTW8lwt*_3dX>P*&mnjBXOe
zS)N{wG>v#;=hG2m1dd{~CFjCnza{v{>H1M}7P_9|1;6~|`GuR^hSayj7;0|29E_`9
zWf5dWXBy7e7rdW&dVa=-#Nlv*RRw1>-7w&<&rIuD|NJw%`w#594QZZu`TUuOzy9ZP
zYcZNoH=XPk=F>AO1-uszzBZQY8u76(O$(2YPkj3qe@zI9B}78XDBaLDEy~E@R~!7q
z3?(JjHPIO-_)cv^FJPQ(=*Udp)V7%}8HxcdBCRklSZnI+Sdl{!7lNq(87&(%rLxPF
z!nh7hgWKFnJ35EzO^0r1aos?-xg}hVv`vFwCz^NfIDGhqIL~B1ae4ZQr{6F1hxgJB
zyqtO1Y)DJsG+m)&tZtab5tLyLpK1Fm=f_8^Gqiond6~pY^>xeRPoK%AVY}T!Nu(Tc
zS`%_&8qW;<hIP$cuOnb_{r1-^)jzw(7>9MOSag-4lp1H9rIn5rM?vkTR*PVHLlO)0
zN<dfg{}VZ^^>+UnRa`epG$~6LO)eDzwNi-SEaf121ltH*ut+FQ;nnI#EieK(1p2LG
z*w=BTw!~#&JiZVYPq#bB7|XqmyjDV7(K@3_mRRAiqH@9OmfR#VhMd8Nh@WOcNL;To
zPmj;c<CVr`k}`srt={yMvUDbLlACEsZ}6m;lC<QFAs5ZZk6(D$-;x?f+q5*QXS?0e
zHXV~+@G&x|j4Pfd8!k^Xr+J~Z--=aENmfD(fwu3sySwG){!XA!B@-5L>MYYt(>FZ4
ze_%RY7_V0$Zmu&adFkdA#S#;QnfJFF=4oc=d!Vpn`0{dQ=p1d+u-R?#L9|x8jU)nI
zFIP&=9JW2%O~=b=<U9$M+*!vmd%QEeoEFw4@%iZs|LK?C@jv<(|9$K?ckuW#m;drY
z>oi95MshV)QZt)Q5_rSBa(sD48%;B8h+*aF>CETHQ6h!YLP;}5i8iU!Y!##By(*|b
zWoXo5lpLW%$U(yBDm4-%g_2TL$C8!&dmyf=9BO!%t{p82XG=BqJ?B8OO-(Bl5@af3
zm9En1+7;8nHsO~6%YI#i@o=DR8p8F6@iVrOfCb&OZ<XRYi=zG2?{6AQ$cpha;)Cb>
z`yV)b{WUg6@PYfgTcT0KVmO~((3^u$l-7}aIblkmX<K53^?W2LP1g;aE@u&JI3Z*y
zW!dfq##Go6*?sjF9Q0p;_r&BW#*(#_`CijYYnj4y9l0LQV#Vk>*6T=2k#1;N*DEp1
zq>w3<Fy%vL^?}1-Bc9=~U>gk?q5lm#QPgKpu4VV(9pNgoyF5q$$mudcO2l=-Ir%{M
z`z>ADa#=Hf|J%Q3*Bb73J*IC&+*jM+g+ei!SQ2Z<{PfdjPV>s(<j>P8gKdSpr(4Gq
z0zduDPxOO>rsHy%nScK?jrm)?`u+!YyPnP6*W@4=%=za}lxEAE6@Ixg9v{hm;p=bS
z^Z0z?=Pxf56Gb8H6HmvHhi@WljI2w<SVc&Q93s~BC>G)>V<Z8rh%GSq71y-VGOn#(
zMJB~49@)N?dFXt_G##xO(A8h8jYj225pglr{fQ-zk>Vq!ESdofMO!dV+z7T2;hCyT
zf-n=O3m1Px$BB8=QdKV#WU2)+a++qg%AxfWFPAIF<B9Ff0izWuD|`Y<L^U^JIqe#9
zT=6+jRL9Wl7;H;Of#3h(xAa}l?fYAT+j3na&p-dn^?W3z3`A}ZJ380!>ElPX+gsc}
z?gV4n;f7&15Tp3_Q;O89T4fL*8)$jm5Mu?qWOQ|^loFw6j1wqJjH|r2O5)T@NkAc2
z<Z&rk)aTm16@SDm?gL%0O4RtJB#K66u&SWY7%RzhF)Vcbo~G}}HE?S_pYfNAylJn6
zijvo=6j;U)L&P-#N=h-2uNQQ&sJ$lZPB!f*3VG`qP_Vx_@Rw0jnk&D3_FR|gEjHU(
zE5*TDWTa7qBudG`FiD<(<>~Q>WxnC*@iX84<~M|#XorrW?<@aaI=h$gLhmB`yN>74
zV^!gDK68Az@b3K`#%Rho@|*Yf42L~Qfm1M^jy!(+35{ldf5Yv&cND*nO5yhI9%BoB
zZW%9UKK}g3b(&cc9QqDpG}AcKwjH0JpLjYgY=(h@Z9z%v|J}nK&tD!1A>bNE+gQ2j
zrhsW%w%d-!mnYh$qo~M!Xt~`rY+J=oFEgKBuKY*;=l>&0!#~p_KYsoj)-cgp!`<DU
z+xtDCMhb8C7V8|l!=C2BaUIWGPA3{4FisLJH=Blt2(vALxSr9ieS29;R<DgHN>Su2
zvG#SI`43)fY41xgQr&xMcj5p2c|E*Ub+wdS#6jR^>3o_9xs%SWa-t_F_4~el4JFww
zs)}=e$@8m<fu|pS<S+ioUvmx~mWlK83*FGu_dRn=v|Y!r-_!2*lpw^ojFY&`PSM`H
zqj8UHhK{gYG251_Un#CbtAMsJefgOsC9$@29WlTf6IvPE;hwwO2mCbAEkfi<BqHMV
zo-SwHkoo(6_n*@@4OUy6QK&ex<^aaATryqXVCxMd%x8u{03yTwfZ7|v@q|x>Y4yyP
znfr$aHisQCCLZ41OZBXUIiLK3DTU+cX}bnpCR8r;+m7=%p+mqL#h1rNHp75l6A#-R
z?O~4|ZWxAJ{CXm-6RKne3(NJwGLLLGJ2w4*X&X`&9B|2n%hL(F?df~Vci+B4wJnXd
zWIuCFvj8Xzbj=>CG`n|>FF*c~Z~y8&`<n-*>y<C3$ba&0f26E0{D=SH|Bxjm?!W)1
zgz-u|J@a>e|9jT!3tAPH>k%D2cblH`apZVhXg7w-{6vV7Bp7{SSpyAn0Bdcf)fSZ~
zqOD+;wUHyBgeao|rG(|5<%a)C5G+}|w=os6UvNz$wluA3=7AT-Ky4i>T?3C|lDTTi
z5CnM(lCvNtWw~((tVq6*pRSxvS5la1Y>TxDX9KAeY}=5EV=x<Z$&AMr&esb~*Ht$}
zL^T_>cRQ}jO5Y9q-GBL?(>9h5H#_>a<#2l-g}``zq1oQB**=iY&&<z{xI<w*UwQud
zM@m|mV_`{&wZPrsz};705kq47g^J)m*T>ohr3|iVDMd@Gl>$Wwd?m*>0M%=umU9$1
z5RGhhydtQ@aS;8?yo_kuVVa?)9!jUcC<{geu^?ecrM#-Q6FILWCGHcYrI@Gwg_;2{
zEfd?@JvaAvSm#JZv&<7=U0H+Y=EK+AzJCX=Lcx&Hs?anoX^pIxGs(|13NVGFPAKCr
zTT9F`iZf25n}+?}J+^6O%_z&ua^ZAdS>{Q)p0Bn{>qs%wcRg17WRh00f|QX0{V?!;
z^=Q}9bv?(I6S{3kO5=0l)8}V4W=H?cJ%fe)cEfe{46O*2&*vF!oD{khg=<RYe0;%m
zEw^{?xP5m=44!cuv99C(yIbbt6T_zC>3HSKaptS9K5+fvXWqYiN19G3BjG*9I<oTE
z#xe9Au5EZZo|(p#Y4oTP>ATE+KhQS~r}LHhG!YZ*wp)CS4o4YM$vhl7ZgwqQ1IP2o
zfBOIYKXLr|-%uLik~Ge-J8ZbQza!VK$}-OsC7l<x8%oivYa)aZ<AVI}HE_7UW4|BJ
zCRcq40t8a89jQo830?=Cbwu@c5G##!;sn!Li>N#YViYT^)_;oc)=GmaZ*U@t6stK0
zxuhtuG%9PVXtcbAxJl<&Y^SddnpBc>(tpVs{WS}^X<EW~q0xm!Wx8#T)tc#iq;FfS
zmRWDtI?j^_VSfJM4;*$oR9<ky0pl#ocxGK^rfKEr>Bx3_;OqN6Cd@3?h5LW<&uF?m
zrtg@p7e4*)2ks7c+<y2D3fL}^2Lmayj4L@8vMPLf{5?zbeE9YoKyf^tm_laLDsFBE
zO36e&GcH%gB=feS6|Qmkm>H)P>n!g+yyN+J<=`5eZm?}5s^pR}t%zha6@6+guIc#j
z;enJ^l**j_#5|AW7`dKLG>0wsUwuu^1v?b^AY;JkmTWR13%)m{2(DvucZ<JTy6skG
z-AKFqHR*2CTC`KlVWw#vdb`E-cO)O!=}b!`O&9S@XXu83ci(<be|SgEk)b61`M2-*
zSO3$0#(I81tyh+1;{4?&UOxRqGi<m%{lw+zXZCmZeEr?`9M2Pf_iz3y&X<MX{L`<w
zy}jXgzx_mZ3#DYv<Am*d+D*?=3T^Aa7;<s6&Y(yl2rLrPQi?>6YNH_qRLSDSZ#ys&
zG84TQA8?VtA%$cBXob+}s`9&W0s>J=(5MAbS?84M4Kn<LP4so{EDOiEhOk~xC8Av?
z@T=1mT?(%2&`l<5&HelDxIBGkTxOIr%<Do?ipE-Ux)R4gzKmRoW-tj;3TYnEM)UOP
zBQYg5n+>Y(F{b77Pk*FYX1ei8-wzCZOHv!&-`%4u41G@y$P774aqy4%$t93W%^FZn
zqPj_t%s;TtNn6x48?+TMY*q@}ZYb-OQYvjNM`@ocOUX)-N^QW|7EtvaD^-96Rk|gc
zW>|^q73;)8Xmp1*4Oxl!HIpz(6T^a{u)V*bKOAH>JzddS(Y6if1BdtT*xuff{lxTq
zCWS@z38OGgLy7|X^2wuBp{y&Uf=-~?EeT7?3FRE!{s6}C6_B&#@1FzuC<474Q;iJK
zGAgOvh>~Rw%0Ys_5}ruMA3i?wn{V&AUe3Jx`g?x+?|#dCz0hwrEc3+m`HV7_H7j0@
zXLh?C+ujf+Pr`~Q%~(>(ynOkYvJSM?FwGNP-*fljYlh8+ur6GVCydhEeE7gTT?Me!
zw*3C%3t#=_pY!$m17D6;P@4B&eZ|M0KH<H@z!xpu!9roYTxpw@(=rlLBKS~Af0}We
z#97evJf9|-USYeI5TfuFjN|1zGp>Q#{f=GJaJL^QdWE3a4;|gma({orX1C?}=^3RG
zO<VExAuwLA%=5wzKYYg8j?HF6RvKk3ZR^mL?B|@WSsJ$5SUsg^UK1ph(lt9p##@p$
zh>X)qt~2~sU96Hk4a#ftK-Nmmm7`*152xm7xK_q%#!1mo#kjfZYW^GF6W(mqx%@J{
z(XT27((=S|T4bSRk2Z>BT1d-+PCXYNXuE+m4wT&udY-tsFSub~{PGjefBSD39=^uy
zdQ97~yFbuwZ&51q^Or|{_jez0hb@~qV7i2M9YqIn)|?(6MakNAtmn_@xL|C{=j(;*
zI5V^j(=_3_rouW5v7H%u*lY&2LoYB7t?;Q3Vn(-?by>LI?r<B&^J(P9S*Dj4cAEn_
z2~cF3Mm}B7v>JxpRxEyG0>ay#Xz4>{{`8r3y3*}h;u?AS_?f%!zh&qKiQR3Cn5sfR
zH7!{?=1)H{EelR%X_{tmwqg73EBaQVR?{?cK0T3QWDSx3=s%X5XGk(j(%lx*Nm`bs
zK$uZ}7ST!_FX)^w?Qcmb6Xq*@-(vLtg{(EIWc)ItU1gm&4Z0*=p1z=ecwzU)BUxE4
zm&iFQ!lx_t^BF?mdO7j0|FeJ1nq<T6hXIrThF<57h9aOA@|)cyr$CHJ@V_x&l=`pP
z`>$QwvUI%?Yuyb$&2rm@NplRzVu@gtP!cFW7t#yp8T5F11HFJy0#PLvZRSaz5j$e<
zz07pOb+3y)BZ@ZyGawMk18C$;012#EvF03O^#1qnYpv|Wt75Rr!6V35U273;q$`hJ
z%CZ}hsyw)iRzL3i^EG_-L+2bJkV=Cn>1<^AdH?Xo?|!`Tnx<lMikpWo@X6w*32y^Z
zWZJGnXvulJ5WJ^Snsyjy+Kv<x!CL0aiDh*(-G<-;Uw`=($8jbByROFBnaea|sup74
zcs#MWd&M-4e0aRD*@EznR!+o8v3q#Sc$rAivo4--8u9Kgs|WqHUek0$t6kwlh;TJ|
z7J8jXT&u0u<^T#QDv-runS=pht`mc@Xbw@Js|HzTf?sfM{c(pSC?K48wPHdl9Qc^X
z(Ur4(C4O8GDp?`L4-e9W%*9<Gq(U2wX<B5}Af;d)N4yX0Z*S-}11Sa8@lyB{F_E$*
zlhBn$M1e@fIv7H+;H4m_)PUA~$+!qjRVgM}1ZldiG1}mlrGyWI$Cw6H)nzA`%H%6Y
zNeU6`3wcm#g&+}JAkpyj_{7)m?^u_K!_#Npy}4%^8n#1+4}U}mcs#F=p7<aB*@mvt
z?1vhAfybwrC0y8VCgM^NFKwy&ip2Y$eaSjqaO=W+Ibux9?)_V2Q}g*(|B=3L_~SRv
zEI#vp`0xHDr>Bnyp%~|tzx(A^SQ{C~8DRrbo&iym;ohO7X6OuUr5I;P2oCEFO>L-+
zz<baN=4qlbHBu-pml-7$S~ujyqE)2dlxae(Dy&P?72IrVv@8`(sWs2f2fBVhjtkrU
zmJlLBBtnJ{AD-CXL>^w<b9b|+-*j|CPgP$v@>%>i{V!kHl41JcMJ!VJ?!^>=_~WiW
zrLt2m&D8>($`smT3CEU}L6OrHl(NxvsTj+mIWJn`Ow5j?O59sqJuCnDf&hLj&=AvK
zRz><ri-GwvVzi-eOD!*WPl}PrI!^OUTeTz=u+tM8sgOBw^Zv^szIb>?_v%ZU-7Wrf
zB)f_In>Qc~i(RO0UbFkzKjr5AFY%WH%3Vm)#Q6A;8qeLUFBy-IM59>dk=>g&MJKr?
zy1PhNW~y3HDR8G}K7D>hX+^i`$(7>i<B{*a`Ih}|$7Z*s-ZZ!fqEkG5`iz#6Zs_py
z%)ELk_lS@h8$2Qg>~+N}5LjzbO@&=7L@<CM5^7D`Zc%l^csQ|p|1<JBb2@xayV;>?
zQ^ts0kB}8P&Ak5lXDrJCer3L#Xxbjx4k%L`143!e=NXy}(o|UI`M>?&{~tf)WBdPK
z{~s<T1$I4CRj(O#WhZ(*91+^kHip(s^u0!#mhJGG>FEQPkAEg3IX*uTt!2B}ULhs{
zR73yf9ofya`7<I#QjRR+3*ITVx3B2B7Of6^dU|62=1U&l7R>13`4J%^N-L6^Y5SIV
z@hBtMzIov3<@o>WKiB_PR|=MNr3C&!xJ*~wGg+|C3*MICr=5;0(@;kE*%0ILYR`-X
z7IZZpn(hwMHf6WJthks;PDvF`pF*OfMCqz9{beL1Q4E4okdq@vg-|u(S_Uf%%A!5q
z?gqA9OTJnVRaMdUrIuTci4y+uOqh@5MIeg?86TOS&!p(6j4a1!wj}NI5^??mtq`Po
zgHkP(G-#nPT9LFMr1CxJ>KYpo$vdj9<A($0EEQ5!DCNlM%0;-gG$1kCz{k&L$b#e3
z6ZiWqO{LjvJD$GJ^j%8|fk$ugr~**al8rHZ>n_Z*!zRH~2sElCtTVf|p>Auw{nK~!
z{h4`Lk!1e<pZ_!Js^-tX|2>z}1sftCK78a~{L6pI?d^{7^9OcAi@Lr9tET3@?UARE
zrmD&@Usf2UNn~EVdf>NzD6k^C&e%DivOqNjLF_`r`qCx|IZ`Rfw5)_2sAZuh8X;L%
zhjjr}RrFoW&8|b2Rw>)t8=haz_?TG63m+dJ`S~xt!U)BF)A4e294?OST?rQ|V}6tY
z#pthX)zXZLF<p258D!}agjkv%<;5P6GG9Zpl#BuVh<?vxqR1qEU~I%6>8&XSBBgdM
zg(8U}ax2alK@_f-3r$Uy4O!SyQI7H2t|`Y5KUgmROV>|R7;8!0)Wje9i$Zg_9Jm}#
zZ1!8St}uO%kAa&_$9#I?`+xX1*eGbWw^aQO|M;1h0wH-W(@d5%FW>x*+x;H9toZMK
zPjWL>mL8_j0<Sb8dz8#rw_ttX@ZBTUeaG~C;QsCXmBP0{rIoI4`05v5mM&^jrVu$x
zzWBR;ip+_1I-`Nju;;QmbT?epryA3?$fjmFzfeiRrfxX<3&e=5%IK^fc0{95!jwN?
z&M4v8-`(Jr71OljEE$G74&VNXMrr!p24p6?K-C!5$q_{%)E&P4V>t%sZ|T1H=NREQ
zefL{}bF_C~QQf}BUOw}G{lEO5ndgPN?T|)uczG$mm&l+rW^)G!gwkkJp`{?Y1tSGo
z33P4H*&w%j;(Ug=^y_86hmbfQo^UaeRE27{WIZ6d9fJQ=q>+V0Cr9=x)+g{2l^e-^
z!a0wfi&iviMXuUHJzPh^GGTJ28@4#1NU~;lc%Z5p8YQ_rf8^T_-(j=Cxs}tWkK_;t
zDbUwd36P~gt!1g1OMy}lQlf3!qA`w<s;f)uz&V6e)J+RQTpjbi{Qjm!2@oW1JB_y!
zaT<v^vg><9itOT<>GL0%p8kk#2W)*y$OhAP<eYKSM9zVaj}tfd_ZU_2?f2i%r$kp7
z#-~R}fl6yOU;H!f+dUsY`~xX?DqV3N&oq6{;oIL*SB8Y*_1C{(v)%FMKmVS(YSH_9
z{PSm$D^|mQ)R!=%48gk}6aUFOVzlHeaL)cX8L-R5JfF*$R2m@jgXl;IQFc}`6J5z0
z6xVyi+6$E_>Q^l_LRAP=j;O>{OkNpTc9FUKOrj;Hm6!#QauzJF;uozYnN-s8)-j&X
z7*lrgrc!7Ui(OGyB`qlvqboTFqX~Xx8CSH(MN$(JF+0NX1?Mc)-CI(#BkPXf*RpGs
zio-ZzeUbQ7#$25a5oJsvafY(%6hG`n5kGd%qc3DV{pvNx!;zFC53gQPR}Jetk#pw$
z;SO&DMola>^4-&g=E_KTzi-i|;=^HPTrz+7v~t^Nmfb?LuIQ@gw}1GAHi>`rul@&g
zZ8$!D=JNPN5`yQa1Mj|g$Jf978f0d(X=!T1%k#*%SngiGDovR`euobr8%ftTtc#_p
zEB3GM*j*Nm-+yE?Y=~>bEen@<qHkLEeM1Q6A}vb^sRTY0$y*eOC}7hzcpr!<<K~4|
z_g`LveF5(m>Tbs`{>k6*r{DjPVY9&<0>Al(Kk)9&1C`diy5Do2!LE^cw%lCdS1Ede
zUrXRq{;#Y6asB5Q0x<<rbVyOYyHZvBa2!yoGUfaH!~eL#-hRYR=UAqmQv5jG{Ndk|
ze`pEhYpV=J5(cH(5;R4U-AGQcXuCzpO8kXof}hG6sENTbj~AQ|RQ-mw?a5J~>O%4p
zAu-&)EiaLf86S__>~G2lELmKdq3)@Z;{4%z^3~t|`P*+8wtG&#bbQn4#A*xHbN_IU
zF`A~n=jHJOVY-xK1i;ji#V(*ELub&cBKe4rvMdCy)rzVzoX=;X_rzu9^}~G$nk7r6
zGJ-&ep0q>|ioWg2sxBvP-n`~~n##*&w*{%uLep#;oVO4o!benH81zz0+NLJE1AQwI
z%`I7Jw5kx5BQ2iI-5tZ*pR;`WhP(G)az3AF_OBr<%;O8|Vp)B}`vuuG_(R~U*Ly1S
zK-+bMl#qY>74vNAyAEXx!`)k!(=$qILJ}AO$*=fzVSBgd_IAtR^CPlp={CP)J)cRY
zXSjX9A3rjl&uAs^ULd!(tX`mQ-V%kzXpJ&8ItG-x5Zppk70J)2)AwkZiCPoa1>923
zH|w5|i)>>(owyucxOsJl>NjL0_(-+iFzA}|)0V&e@`3vPD^iAUfAd=ohZnY+mjCo`
z|1IZnVrYA|jiIi~gSe^`x;AuuOKnu~U0hw@N(j93G;NEj3bI(sgwhp);&>MdS}jS=
zQ*B;xo@S=g#Od%%Wi%>P7(I}6hf(nI-JiIO6T@bptsDBA*Z7e6d^jS!WyumfU07d6
z42fp5N03?Pkt#>x`4OZdZ*KYeZ~mTF@4n`9Q={WbnvWd*{A->}$Meg9t&O~T^OpDT
z-jieI`EdLxx9VTr)%F{dsgSwAor+CVTw4VtSXea0psNN`^?=0M1@e_Fr^@-MscUl1
ztkb3BKw@BZSL<Oei0`C8gzIkkTHQ>+7k0lXgg*gSTcX9eBAPN)g-DT@1Trcr)0gpo
z3{<A5l676tbY0OT`?3HQf~$2gp-mY@>+5-Vv;_o4yoMxWW*i?eL1FgbbV1f8;B*=}
zj1vg(cEMdIkNwcoHVw&p;tys~c?A;*RqT(!gH)uHSX|&YzyFSRuXn{&Fi!;MsGEj|
z*A?q*IZY>0f={P~U%ZoSH#JR^M6oh0lB^7e$&v-E!I7iI*(cuJZTWBhn_nVGtkcBH
z=K_wKmq4WrcMp3co?s(HS8&AE5rt%bdsBkjDKWXo7u}Zqe#^)8Gq2vg!3Iy;wxr#b
zPmj-NQE{`WNOr;dz|c25+-z|Hrt?H4YqniYOa`Si=Vf7<Et{?<<noS4Iiaf>>muQJ
zrgDkR{+_q5ANc<J?@>x{KF?gnZ%fEKN}`tt(HA~JP6+&uybG?GfcyiAQb`IeJC%BD
z{_$0RRdnZqCeD=JEc|Cb>q=C;hGvWWO<(Ve622D7l%S~8C_EyR&tK{mWx6Z>!k4T9
z5Mm+j7MrI=ivJ3kfil`iP*n{3`?AX-(umT{)I#BcXF5Eyt`?u58*b@tA8;XXIUR8)
z%d4NgV?2DMQ;y&RE-Q97547*TAnP4=Iw4I3j~{894Ry1p8Ctr{K-IOJn~#XPx}LWr
zK74+}tSkO>WYq@M?@Gr!rE-=g1Wnb@zurJ_1@@9Mm+^wBD(q$C?)HYH6{hQPsUU<+
zh)6W@?hTu*N7yrdzMxblcO}iNwIZZMAd=lmvX;v{7an+9UecARN$m}?AE@n_u$<62
z;t!7yR<u;~-N51L1IMQi+`oE7v%e+J6aMM3JfuXVd-aA85`JAc9}nEUeaFx>*kz(q
z8(2r&=@}t4-EI$3u||ii45z~jr<W6E=n=;Y$4?(wrwi+vX*N6BTGQP;K-D0WB}7YQ
zXZUo+v>QSURKpf~Ij}@eJ?yb|#wblCWpQq+8ZYaDQ?(NnWmQ{s1FG9oJ-nq#1um4M
z#mh*V9qx9^=Jl8K`wdYV)_LW>{ii>t-u?@cE91ydhXd#1iIf7U@<Op=KrNPjv*mQ0
zSeBLj-40UG0%vbo#+i^aIVPk8qZPGLL_5>&cBGpJ#`TG6h}6bVDS_PYQMzV11}^6#
zo1544yBkEX-0wEzoXAy!3z@Hf`L~Rpzei<DfB%lQzvX=Rp7H5hhRq%0<-$C^@ZEEw
zYIdlqrrO=IoR7F+py>xvFrW;}lz9E{fKiI$(?|aN>036h-jHO)X`Ft_t@>AY1@EY<
zimuxe6D-StoL0z%oBzXu51_7khRwq@*?+{@sRZb>B85m+C21W)@k5B52`RJsfYvq6
zOB4!8Mp%c=xd49LLNc0GtLrgA!h4IiMZ~qt6I!b>)<sb<U{{N}1{{+QXsy}p_ZTUe
z&gb&~=^|7T3UbIqAF%6EtaeJG>WbhTF=WDO%b34uQB`%VF<02t<3i?ioQMfhEYpnW
zJ#os2y2ANDQ#aS3Z2Y0zR;Y@JE5)&3l|R3X1i$k2*Khdr`7@2Kv8$yYHb@^3O0wGy
zoG%MY6xRux!DPvMV~APfp8~dp$yv^)nROny-?#K_&9{I4j`@7zX20V+O&pI0HtmjS
zS*ZJlSK*%MB6gV&z`O>I=ZP;<sehg>6Z2BezpJ|D=JtlBuDRdsna?Zc=7!T{;nU{>
zF}1wD-yuZC`9Mz4bv4`Ekz7KFOg9WA4;4HqORTr_n-&`*Z9f!sc-?Y&I*~Xr9!Kup
zzQGvH;c!MO!SDX~Jzd+g+wEz)4OLa#EHQdQaD^Qq<n`t5uP^K?|3X(7Ew8F#T^I{7
z5n`d8dcPo~L0_@aWuMQFsO=w`G=;}e5t1ifCj>&9LR^&!?E`k5@F5^&jfes}pO7M>
zwIB#}4LGWjkwPKme}%=sH{U<<`tF|FSFdR{o04?S5$6+Cx20+trq3UV)}m7+3B_>t
zme_3w$1~IHXm<mnTQU8B4x05c5{)76_oN_LKmHzpr|CLqYnp9~l#VpcT+UC(N~286
z`EtfDGj-R~)*V_YE;&$%3b96#OvLL4L5Qdv%h};Nk(EI-9bsOm>kZasT-K;8SstHQ
zoug|?ueiN?O-hNbX;`ZbOF9yqrSS?iACV&9ohzwwrC8m9mXcQ2XeH?0|1IYB9Z3je
zu!NVDm?a^DOcr!SS{5V%qclPrOw|!mq*ofcK(dzAIjX8bv9c<~{@u@L_csVvI<&f~
zh;AV_J3{-wDkFY;Mynro^$js+R2F1y$hl=aTo`scs#;Um4Z^Q1%Sb|C2GCti4lry6
zrpu9JM~uoeRZV3?2`bu^zJiDSJyF*@9!>;67)8jL<R(lNsq2=iu8_Gj2ZZdYvcyL4
zK4I-dqbn*Y(ONLiCtUWB0{`NdU$eURWt1kt3(0givd$~>JTtu7lRmw$<Rg>Zk+fvh
z38P!=dFDU;hu`A0M)-w&W4OKP*)$btI-@pUliHsB{afmO;PmuFVy3^nMQX{Wzwqn-
z^zV5dI^KQt1=Ky^d}MxpVYq)_`txr{%R)qO`26pAb$`!-0NIhHqN@8c!V`+a@d??s
z92bk5&zyespAe>Izw3Gbvwy<vSAWmr?|(%U(BJI2j4Mcq!jbLxQ*PD2x@$UoMkR+)
z4Rycca`O0Pye6j$TjGamlyeIqTEq_zM!#OEm=($-oGYG&k`I;StWm1Q+6&|vwds(#
zQ2y4*qH_^H*?A$#TmWK1mQk*dq?n0GlCCdBA>*~~YN^$X4FPwVm=8yWn;mU!SPsu9
zAyK9f^5R<92vRG8x1>-s#H9ls38@4h*L5vM%+|16X6k;=G@c1QvigKE6)BX-X4|%`
zK9F+6Ei<}qi@h-xi(^QM;IGaHKxthnlri#jbadUGZ@>GVUwrwF%4piYqizOH<IHe-
zM<S52Vw^^1S1NH$+aOJ1Xe>h`IfRmh2sxuGjZ_uC`SS~YJ@LQ$?|#nW7JmD?@9BpI
zH-JeIQ)!md3w|B3v*YRW3r*W_Sqr>O0n==+khKQW4ZPm%ky;aD<g2g0;7@=2miwC>
zeO<BGmCI#h=sT2zwysGeUf=9EA15T&&6_dYZa2)%vD<HHYQyvS%($-H?mMROM9OeH
z&V2LXiLX;+x83sR-+y3U5;waWo)05c<+;CW>6(h#=xg=W6JsiTh^q^wAff-)DZEq?
zohx!GmBSc|7loV=xge;?<?~Cq9PLU~DuU|AkfA5XqHQmeuu_HnLCv0j2qof)?nsP*
z94txKNMR62Ox2^by!tWZ4<``+#p|aX6DUI_v91%|+hVSBfs_<6Dl#>?8rW_3gmq>)
zoq7KAZ!zB~R;BQ}9-a>PaiXm>!YGWcN#0|e#ZDuWTgi!{+#POMgu>{K%k%eKUOp9;
zemfAYr`C$yZlD*E<?$2K7)Zhrgrn^{qV+7N3n@A7Ufr{Q_Z9Pe!LK7(H>kQHQW0`7
zRQ0+-brnj%`S^^f8nRz7?S@Hj3H?26_6VVIZlYP9sEsZdOG=HYX&UNE7jv1AG);|G
z3Z*38Pb66!_91$fbt#C_w#6?K!)}La2GANEG6+Qy5bR3RwKSV8eo@p4>isQQ8nP0^
zy2j{=`EtU$g~fhmCebvSzH6zjD{8F`5Fo<l{%iKHenxFz`>-cmCZ50lgfWKfLOBjF
z3Q-%z<0CRTQVb~5mY4ioZbDr*MJ=5tq+D^!M6P?<;f54Kp;RhO3W2<iWEE+e0Vy>$
zW$La&=n79DtBTX<GYaT#-;~t2TS?N=_Z?CjUd|Up5_Emf&D|ZQZqRMb-~5vfcRu3g
z08>G%zJ<rr%<I?hSx(PnEN{R5B_T*AyRcl2>^2Q`PH0&sMt}P7ZBeSmjMOb*nRxm9
z$k4T%J^b=-{|1|2K2B_go_g5Q4+GQjNUCe3k-S{W1m?FNzT^Gh{5K5UfRzgC9;vE8
zUOn?==6o(8y4y|7fBfxl7;YcXyyDZR@40*RmTt3UeEN>2HY6cIX}Ye%|0A~+3Q<}d
z^L#=mLH4!;1_`(|lUcCqN(L_D%=ui%of3f*=2{gkc1ooTb}c<~@5@VI7;do6;hZI=
zOhV$uk?cJNMc=nbX&BGX*x(AbTk4_#mKsEH9gvH`5HH9sGOLveSWag|jHv68SJQWh
zMw5cYUW!rGXif6vVIAChWnM@`&IKfs3vXeyrG+CyV)2nwwe<TvYl>L6;+#WkU2+Jy
zq``&vBxez(xD8OcEb4Xmu~wTy$yBJ49pm%Ug};5fVLNO{F|z5mBx$GzO|zZxZe$-a
z-djFAUwHd^L%Wk?m)LAH@1$nR&{mRjwge}bR?9lCY@3Q-{pNdoh+I~OIyzE#;uo(<
zeedbJ4{WzRpFe)0tt*xs2`M9#CRm4ej?K`}7>%rIhMPN-(Tw8><pW=Q{RKwCdO0B&
z@U{S_Ztn)x)uO9n0<01QAE??Ev}9aXs!+HTL*KFl&-YJ9?rygz0^S9>rsnyu@c;ZD
z|06&D*?Urgaa=gfmY8G_*onm5Z4VmOdBW%jDUdRltID1f%Xe8R$gc1J;t#^D!lFqj
zp|wU|QT61=8Z0q*wCc%Ommp#)+!ME+na9rvrAvbbMG%gZ&q!BfPbmg85Ys}=j!GM%
zu;d(3s=7{`Jl>UB@;d)zQ^lWFEsOzE*R=f>=N$85Y5NV$#?bUzND(bFql?(pF>Lm<
zyPox2YFR#4*d%D%mP-=kC9!Q4eGS=72o(F*uh3awr!ybF{~jq6s^1}$!I}-@<98VG
zf-;8JyPkOZOx^d$?LdF?z&Mw%T3B77Oa0;jKaEHcN%q34n}P9?IWL|p%j)UX!!5zi
zkQ2*fvCE1u9W*_{C{lE&bRv&Og05-%4WvL<EBrhnjV>^VEAa_{x+mqt@w?xl^*3z#
z0UIMim0i4@R;FpD>I%M7Qc|Qww}ohyQ|ZxG3ZV^1g<V(X<B?t0pxnfCnaXN@*x-G@
z&Lgc#yt&^MAWqDnBz74&k7u+t?1l|eX}V!k&fyN<p>{3pcEDQ4JS~W(!Za<;O$?0z
zBw8qf3+T2-2!XTisvefq!wtc&oS#1-q{7sZ<?x89D^MDtYp66?WUw<H4N=gxCHs)0
zMbr(}dCX=HF<nXDiij?u%$p9SE8f1@;M{_}2(mOJpw26;sj$7__y6JF60>G`xMTbE
z7lfA=x~gLd3+?tDgh<Sh-c(p?$<g!A|NiHN9j`neK7J-OEwA@Cm@p!SVxhcy{g%lE
zh~>=q@N%Hqz2Q91^p$3JJD~d=Z|jDB-*Y-VQYFp2jQswa-!jxaP2Z8T;BY+PM8{XZ
z{9C%sp3~`ww=2(o{2g&Vk%eS;`v9j4FNZUmreoOM{ghkvukJGKmj31rq$b!2NusW6
zQZh(9^LWAAqH(o$<^1x3){?&8VoZxrf*2ARftDI2%K5!&+H%iHN!Jh8)V5|_1HpRy
zyfU-{2!n2RFt1!rD{E!REGQ{3rW`4xP&D5qoTdm7rpts40dk`0TcoL|OhdQX<K2QA
zN8&spa#;W*JK>!tQGhe!c`4Pf>odZ+D>hasQi@bsBHIc(FHDz}d7LoD;ErdUb10?o
zDHYazis=8$e5aHqUX{s-^aGF(qpNH4Lhx{R!}H4{_xG<@oTuFmRFy(i2c%SJ0h_Mj
zv^p+JU>A$+^F}ICwEXN&aJG(KHcVs2dq}EbSuB72{(|*^rqeiUF;egwf53W2@~;?A
zTRuM>`11W5{{8>_d%php7l`DUrjf(>LXbterUVR|Eth3vclW^O58si=eErL>IXr%1
zIgeB~8boP}bbVnA>~Ho+k@@)jBaMQ7+q1pd@#jB(B!<N4c)~80KYaT_2#KG)yP?(;
zWxX!biGTe+{ff<Q$6>mlgrRF%ZZ;KPzPsW5yL;B)IUPPT&$*aWvmlGWyv*bjsOk#s
z*B`-YQdcNd0mZx;qbEqpLNf_c4lMHtr7P;XhvW%iA-EZUbvZGfKcRGsZZ>F>i@j3<
zu>?nzs>u0j^HlmrGwapiex(3k5EP(M@c$L8g}X0)PTw~OXQ{glq{uRlkOJTT`d5gm
zqBS+~W(WH_hKJXjK7NBcKhuzD#0}q{M%;Pfi@*CD!m^U43)THCQl^r;Tu0(OK}gg_
zlCr>vfHW2LZbPTvBI=rtA09cM&IoT9?%vREZdr52=SWVDt|`dj>2yXn4W>3+9>3@G
z=NV@PY(1d0!X(c!o{>_ZcQ@?6`Z?2OWV#&D+dTw_To<Zj34SHajzK>VL*%lYF;zuv
z2Cic365hMo{(k6{dp|SIp;Q^AL75UR>a!?S6;%LAF?iO`AE>*5;r708u$4m9;Jw9O
zPK@(~{lfz<hZA<5h|#gw8B-1AyeTJ=FMyt8SA^Dufo?6esW461iO2bby&SNY3-&j^
zq1$W=vN;D>EfDc;<&r%DN%j$`svmp&<w2%UN{~{<^jlgCm&0e$d?rN5QlL}jbofBG
z-67i^WDO$W=QClM3vi~cX?A<+y2h^y2#*R2E(T0D;3Fhqu(8<c5)vB($cV0MmXPq5
znHVDXU;lz-97%R%85cr`tn-MeG<KPh2$p#!HZ8;3uZzj>_z|5vQEF~(Zc!4NTH}^L
z)Ag9@mh2txhK9@Yk-Uu5DN=Ve!8w-k%>M1yY+k=F{pWDz^-a%je*HVbW#;|+cjO?L
zeW2aFVfXN!r$77y?d=_><B`XgBVE7eysqV;9|FrfqmyH|?+H0@K7Rfwx9VTr<*g-V
zLqFUyK7L!md`crKMXEdOe7UmPf#mRZCPzasC08J;ri{VWmHe49LQ0ythAfB#nx-YA
zN`E_Lblsz?_G%c6NLit)Ju1v3-C^xSUXHX)%`o(+x-6ixYLH12lcR21miZYI16|Wm
z_cgjVxOF7hiRFAGT8l7(;2d_GnJ*JUXjD}d7egT*ly*YOtg9y}O;{sq%v1p|4I(8P
zrPy^HnGvNVQb=N=@aJujBbl~Cn+oSwq&5ic`Qc9|)fK6%6-rn9?oZ$I%P$^i+J!ea
z_efRp1;f7Q=^Mv()6lhs&&M<09Y@~3xkFbKA*>h;Z*FR~i{@=#u`UzQJ5HB{KR;%C
zF0i8<Qpw@C!13a!RhepPUU>U%$9XL9Bzr#a{3tkl`pjWmnWLcY+N*BbBXhzo6PM!z
zA|nUQ&D||VYNB)KN+OjgM?2%hJeKyz(06S210f5%zv7=`0l9?`=qk;}$1|7H#OvFZ
zzSeyEI1-W~B6vAX)JD^`HC>~5b=&d!Zs6u_k4uShd?_J6Rb8j6h4lD?Nie2JbSold
zRB(hJyc8lAhe|G)pz=%|i$G0^;u;7ZJD&-zEQF=h1({AR*4umyEMBXMS{L4ikmCC6
zg^)tY?@2+5D}I>?el57-tM&CSxqi}O;QsYn8V$?w5%==BRKM(k-fTcAHaEBI@9)a_
zTnK#n)xRaJGe!bZvAfw(NwAZpGaB?jlmQVPs@=29D=LDhYD_ONBjhC6YzLNQq>|9>
z2B=$%s&R);RJ#qn8c^-Nz<I`*^>`vQD^=4F5-f*j=Bh?1#T*33^MPuU>9wTV?YRBw
zD}q}&9nQr3$b31I{K9-bAb@_q#ncU*snI^5sv0jmX%;Ngimp4vHMuH``Onb70<y%p
zqBJc@--ynV=sj)QkYYh&#u#XNsJC}iyL*yQ`0+%T&*bDF1`zP@?n~NkAlQs(YJ^)$
zjm9XnD9P1znW>a2Dc9?+Vza+R=g81%s;<S_$SMV*9Z1?R*-TX{s<y*kPCvpl)@3Dm
zPa`DFpv&$p7e4xV9#MrJL)#6QrsMAQE24A6q>0%QoF{sVUuJfj8=R!v_%Y(cf~Z?s
z)e*;$^aB|%x^f<xKuN{0+hd(0Cr6xSWDNALUPD$_j(0>R&t-f<D#^pcEz&e7BXE_5
z{hsI}b>9(#N2Np%lIiq>$cb72LL*JZXfNE{-k|#yf+vcE3%(qCUC#J9AvYWL_xmek
zL~#1UBmVg#FLp%hj_~w>=uULEUy?Q&gTmZy*g22&6GBNI-rgcq%`{%vXhpc35t|J|
zKXCZ=%)B_1uJNX&?K?P}e#))-S9jHQjbAQ|AwUdBGV^%B7>%wvnzk>CXD9%coC)6Z
z`Qsx#Yx<iTw65ry`U=<a1*RedQs^Jk+O=s&S%NVnYI3j$65d)!0eLkhigpLx9p2p`
z>YCJc)H{K49+48tIm|xNZ9DGnZdg8jOY*)n!>_<BzbvfdRBEOo64r(3TzdLS8#X&l
zRae;6;hZDqsu=1JMC-2>O-Z{MiptnuXex=(1+(4Mh7cg+gbR_R0!n~U1t~0*BKlC!
zxKjLhww%aFAvsSA-#vPEL(R|s_Luma@vB8u)s-2e*z^s$G92fX%VcprQ#FdZt{IX*
zRyA+m)+mv&(}lzH3yO%39s5pkx?FG>nyO=&S3aE!hWPO`@$P0%*D9VKkMy--I==9I
zc07K1Avi~0R|Idd)^hm#fn^>^LUVk4#syEj?YZ6eOw$N4(a7eSt5Vc`hhLXsgM~t&
z1)iRs2_amkCz6}p2A=_4T1L-@6OZ0=w(xwJQB^~uDsI{>yT<VPVNX|Un!2W{8tSU%
z?(P8}eR-w>F(r(tDK4Q*4l7y+s;(txaBf6o4`j4z5vDFd^AJ&5({|g!KFa~AR5ATt
z7a68%==vQYlsrrf9&cBasnAuAG1avan^4A(bFn;@d_lskmxA^dlIShjt>o}ufio~3
zo_RW+@NVYSn|nlEQFj$pRrB!nOX_xmFbcm~-0_*V(JT?Fy2VZ_FMs+|ku6BY_*7a6
zaT>YVZHtIPc-lJAnTp3xpAkT9I^sH`V!V>46jg?{(X<b*i77EnGuD02x{M69;pW|!
z_*6Ra%XC31jWLF%Yw3133=wjS=-_C?iprY9WdtN)nK4SS*}o?yhxGw;U2ZxN5piWX
zJDPsWDl|3(nh>aDX1<JMU1M$@kh*4CRw_KfTF4-;G(eRCLRV1L09Se*Xj4pTZNJ0W
zB4H5%7Vk(Yvf1va>YBrcZy0uWY}!4?r_WTTgr#!I^!IPcZUQ@#OoR0ar7}iEE{{)q
zbNWmcl3{mS&e`g!R9SrB^zw{%p3U6@S{bUCh$x~Igcw-o2_MT;Kq*zIf!-mNWEoFX
z%Y;&jqz$5P5z_)GFmw&U&5R#@!?;?uH#f+f&`pc!dX(0vtLL2*3A!y@gk@wpJRwY1
z0=PoZ4O{RQJugHNk?n?rz<Xa1p*|pDq^d-@0kvU!^FW0tQyMo>8-;V>8rC$3EU>FX
zOU>86_&G`&qFcDwfVC6Vm2{|@hTAXSVRXg$yAO=V1FGxT{Op$qkx?NMa=_f)p<R#M
z?8tV)P7^A7=JCLCDrU@fw?#<7ez&7D6**;o`=5T#yZ3L|^*yK48QFB~?mA*{MAQ6~
zTlKH*(p61u6n>pb2&^4QG2_;m6f;`4n5qL{njDw&1UWN}p7Rpn=|t01Z2OL`si-Q0
z_%WwblS3j?Dy3E1<JPr2EJX=rnaV&4<Pecbv-*H&I#gRDj3MdLXrL8TDskUcm{crt
zKCYzNU@r?{nXduC$Z|eor<r6u>pBsG#jTDU!Ffk$J$0>V+lHJI^JD?}Bjex)t*vt_
zpa>zb+im#v`9j~;jNV=U`wV_6=gfAQFil-7pEBW<KtzE=Ba|%Z^ISCK&IjgDlJ4^~
z5;Am~J;$db(|JTI#hbUUn5{(!7)Q&8Pmlaezb-AHwq}2SU%L3qh}DARiY|EQ3^}x1
zUNn!Fm7{f3yNdG!&L>=syqr$RY|%>6^*x8@%=t8O8a=y8)3=((kDpmjC)&1z>#BCg
za=9R-<mIzP=S1{@b+LpL*_kadLOt}<eakG*tTtk8;N@`S<#=MZ+fmmw(=;MfqOA;1
zFEg2v->8&fs1mko2w78Ui6QdKS3B+>c66JT7%YNJT{UcWJ5Hw;V)VqTSyf!m|ASi*
zVMa(njG|NniNv&C{RDkk*oGNrCq&lN?GC90MU16L5~`})-%1&pZjY2TDF&AF0bvrl
z>FI_$w9)uwEXG%%P?ag|29XG1MJYvXDljc>U9i)sunGQ>d-SK=(7*X#{wFS{BftFm
zE8J0`Ha!Nx<?)%};SIrC^5n4d1$Q~&gCn}ghws0o>HE?Lzk9{ybm33`_E$W-y667Y
z4p|!{4xD553%=H*t1eT>Oe9hbJ-ITZuwea)!VqjM7}$0B(O&0MW*U#Bx2!a+6v(=x
zsSVLtvWiqygR=!<<YdiqSjej<xRr4{((i6*H(w(()bql4cmb6`8pd(PuAa~T@CP>A
z9dEw=62gkJj?3yWn;rG`4NBK^!-{+Pmby0Bbs>09+qDFd(V}DkLX1>ZSvjT{S)V_W
z?ZV|U@#%NJqTUWPZJBCBVenZOU5jb(KGWa4LTiOzR+6*S{RZ7P9KRb$(Ou6rE9`t?
zySwFbJn`}KCmuZLw&&rCUtsDQyUb*3X}S*6ZivyNWeIPFG_$W11kc4Z2-V`(5^Bsz
zGtCP!CxXnZQvsXk<w8{#1ZaqfWi3Cu-IOUy>75r}fRTbo{87XvQXmS08t!Oya}9MR
zPzKp_AO*Q=iK|6OP0rA6?}@4bb-?<S&2}ig+yriRSZ`^Kf@slkg&6TcQSJ5^V7tFT
z*M{|QKnD*Zvc22l(}15>mgz#HYqTs>JK0pIu4DkBchviP@_fM%5X}x1cFf~}c4&!4
z;$9AgYroxH!$dVv8@gSOu4{A({4y<U?rJJsQ8is568Z!4#j{-gaaRkcWkoAV4n_67
zhDcR+<m-1s8_5qol^6x<QZgyd7kusXa-?b$>FySGTdY+ybqPu#5{;>;>K>{dOpTTm
zE@nhkUcPial@bMK9RcSFF%wfFNeBqEQW&L>!IGUP<$xd0gz3z(%q-)?JWU0Tw%SrH
z%YvbA(M?613vei<fD)RpI;0V3Q>tEwz~wxb1iF%#x{v_vbRn*Rw(VJ5z-Ue1)|hFc
zsw(QX!-r>9>#@sBGi)&0lq7jBKrnHwxD~9gy1ujo<Kj7AX2#=@fih)y`TWfBIC9fU
z`l01`DOh08EI#3Epx^X#`z`Hm04dRRO>h-m-_f)+F?ihh#GluN$trH|_PAg%!Es(9
zkLQuDtvQ}&+NQ?&3(G|paSnnn9=0?}@cGjtqp5Lr;?=`F&*LX<He0k5>~>o|fB(pK
zyFmzf-4#ouk|=G^M&TBVl$w|0h;@O@W{dTK%RG~F<~U6lQ&Uw68xxm{LucsQhA+1U
zr3{sZSNoQ`t>VpH$8dW?vu#N!6rRhEBl#ktDj4)kh#8SQLV_^{NhD{F4<jj9nr2(j
z(kT-BMDz>BY|92jDMVH!>p|J&0|-?b8^-jcWN}qhrUG?e7Pml75iMkyiVB4^hO87S
z=yLpKDoj0~RSQ`Xh5L){(Vw&!`10?5Nmv*5eT@brnelk$@zXQ23mi`;rsIjesoCD_
z*}b}B_shS-@h6(5rfM3xn|q?}dHgs3!0-OYe@nlq_~-xREB@I({U=P`<6|T&5hV@l
zIFsXytPL)Bq^h_)9qEUToFJ|<cJ`RAMk?^Z;$tQPwJynEnFIax1Jtr3N&!`kRyBS+
z;?8GG-O>yLZEjGNVxATbAOB3V+0ZmS$*;s@sJ2^VHPe{LASIWVBd-%I)6C7y7TNZA
zS#I!PJ%WI@p?uhFTSiq)Ljqzh^(9g&;vQ9Y#co}h*NLPQyLYe4Xi5rVj2vHH(Ct8X
z^SYdspPuoq98+Z7F?{tmRClkDVWp*E{P+jlG*TH;u$oOnzu7UodQ)}~WK8yGmr<3$
z8ikuL96wCBWubd`OaK1wsjCCyhhNh*HO)|wbWNHU=F0(RM^Z?vt790-s8^YqxGt=A
zK~*h2WvXFAGwc}pj$oJKk{7Cc;bSBViOiXx8)9>V4T)ALvRKMkcCp9`QUh8yq?D<L
zzE}^VN9I6hpn3HQ<t<VhmZUKTbXrMOF0l^NqoYKaMDUce<*p~%nJ^t#r%R~_Nr6lW
znV_x>DYk?xvDTxi7Sr!(VlH&W`9icSLP^HilEhk|Q}cyoo!M+Rh^|M*T%3+V;d5e{
zR&L+Fq1*16zWFV?-3|S#_h>EA(}<pD;_6XKz~(32s(*FY^YaU>lGIg=ooC{*lB$~M
zBX)^}4tcE>_M1Jo_pdO<a4!tz(+jIF*0!c^X}9;Nsw#pj4N*uu6;f{yW`m=61+pxY
zEg>XR%=qg$epR(pi0ccJ;x7=nOb%i|dx^a~v5uedFHa~@>a<=P&W97{<B4EBRaMdT
zO=-P&i=E2Zcx?<>$m{67RQ6C;IE|8=&l6<W?lfIj(&<&VVR4aK8@i^WZHf}!IS<OB
zYfYRd+?p8ACrl_~Xa$8`Lniq0_e&I$s1O1hGskgZo<{EOZ~5b&{zUL4_w(lM+fp~y
zn$6HNMfh}B+4c+D`&+7}L942Oa#YFZG<}D4#n8GQ4>XnLVc&E6FyL3s>qhdY#|tJ0
znx?^3it{)#yTB>Tn7ZO&*HSCN`E=nlSw7nfMk&tj6CaNwciWLSo1S@|7&blMeg8<x
znRc4_s(Q_8%hBj&J0O(eGF#5Gr80)DZCPC4<I9<8nb{8=-j`z?L8dkunKG@>bVhQw
z-=IX`;cmnJrl;F=bo&kMZeKuG+Facx#a$qTEDW1yQOY3n4Jv!2C}B{a9J*={x}|Ei
zplU?+#dN4Us(K4tvv5^|Oo^ll{T9l1L026qdb}?ahpOrkx=ae9Us;zk9)&RtF@tG)
zv>qUrYN}RssWU1?ySXPT`xo7#Kj{f`bMruBG%5z>@k9`c`u2ftT&a!W=4Q`+w`JID
zIWH4YmE^+h+c!W)YlR34uV3Bqzy06;OTPKtKj1?`h~gU%{f0axYAJX=zF-%J4$y>}
z)q0ZDC<RI=6pHrdp6W)D-9n`y`iO`UsWnO~f=y_pnbwtBDoodbb7&(mn~w7|l2MG~
znG`*n-5yi7Y-F)g&6hLYI+|`EW{C3z>n3znaesHu`*-gO<U(mO3hy1+M6#b*oyD&c
z_T@7}r7>Mc7I1z(U=*0^i#ba{2rJoTnqAMP+p}C2Qi`lvl2wJ)ie)^bWG>Z~c3?a|
z5q%<|&{AMjA|0R5RZY4IE=;#2yM^o*T9L>i0Fw4fa?3fgoDXQFNXl@tza{T(iOO)f
z6n3{gJt4x(x_Yb&RMz8FQ45E4#m*-JnKcbm^=rEOJ;IinSd57}XWF5sZdzi__^LtQ
zzQZ1$@h{)u-Gb_Ng&dg{{OU=S#xEEA@=S7b!3Vb;P1n=Z{dMeB66DFRK)_muN||6S
z*-az^`sM-Q7JRf+W&ml>3gR5d3f$$0xEzrwAafuGk7-(vhUBIazKISc6uGI12nd#C
zx*&u?*EN~MaQ~VrEX>X!`wdk$And})CP<2JzxfUqBM)!hlR`kra!j$hh0~`Gh*%y}
zA!YJd9`rI~mUYHD$Nu3S`;R;`kW=P#I@44YZk&-JlByz`3Tr^5a?UM)+uH{o-u;5A
z>4`|{2ZzfbW!V`CWr{a6hT`wg8lNFbi9(an_~5UVt{*IX5*Gu)dTODm>iP-=(pNB9
zB>U3eH%1_YC)t@ipINPAofmS7G>st-9adVV%Z!_6w2CO1Y3hnB3wkz_5He92T}+xn
zfR;#On0z9JjOh|p%ydd~JRZ??i&hnNb=?6AaL!_C!=~?;#|gVQ#5`i^8fhv-2Bpe$
zDgoYma;9(@grF$Z6ao17-A6ut`^d20A(dbl2Aucw!<IMi;1iD=j%T7UXrs%^CndCQ
zP&$*1LAMQNGmz(*+EmnyVt==z?+tFsnB>^Eg3bGV$!IN(Pbn}*&C9s(<*OaHL#f%F
zFDv8hu)(8c<YkIzqsx@l7)IxLJj~QalStUl&zz1UU%r2!Y8Q-_tV^j|PK(1w7&aSl
zo|F=6No*5+KnYcvB2uy+8t!*3IwmR+X}SiX1VRa_wnkM2WBr5hEag<_lG4z0H)Nqn
zas}0*b&VpG&c9gDvO-r|$_g!AwZ%16wYx%wG%;qB>Cp8K5hE$iKLWR;tWdIMUOt!K
zx9-Z~Ku9DBrCLavtG-%50;LoZ@Tp7+e6SEaSr*)V4F46{1S$rUQrM6X&4yZ)^ROys
zjFw2HSe-@HJyU|?;|udS5TqomEB$7m>KgpV?=aExi?80&-@QX>uy!Q-i0%dw(BCP{
zI?+;Ico>ChI;4z5A4p1bdVHqc-{O5>IbW!%8dbGdafd85KBWjm#>+^pYnT_7Wx-4h
zrjmrH3a8sNpa!PncZeL>_Zwtgp_?t0u9(LYLRXys@O!+q+`M`Pn+><`f5v=zVLU$<
z%2kz!IgtfKHzFidZCCWDA!6qR`UaU`nr7;zEsE2WSYIsNuqC>+j6YMxx-(WAw)?w+
zwiFsAGSz0!bU7FLm68N!nWm9xI*{B#({_aM0wFL>FO1Pc@W{;#&HdY}4!K69h>n)U
zuY`5ILW(4}L(A{}mw$sc4L7e|)7`xxr_3_G5YH#PlI3rv4a>YT2hTPua?@ci6QUkq
zIgzjZ>fj<W3&!U^P&eO~`cYVM-XYeai%lUAd?w~U;N~N^nc!AXB}*~fzG8R(7Ts=W
zc6Sgm*ohE)>EzZOh=`8~kpk<}HzYs_kU1m##4=fAa+rS0?ajWVgp)6P1u2U;QkpW7
z%o(_pW=S{DRxQ!a`0+s1^~fA(ltlQ2;4PJ|!8zh{X-}wE?^x#%p#|ugW?#r=QW~_(
z96o)^y3DjgOAdz3H5+02hGY!Mb;!lzQ!aGKf5ev(9l=X%k_g!%4VZ4A-Q7`bR&t!t
zRT-5w!-jT$hcpeuK-adIB*;nONVt#*A!DkVd2u|=2`wbjLP(aX&QwZ~ZKP|8EmKG6
z>hf@OzBud!i4+}ml;q?Qenq7~UeBoIM97Y8B-v)%x^Oz2@UHAKtE$HOKnRYe(KNN@
zY#ljeDkaG=P$4*-M_hs^G$=_D5?wX;Y?+;dX~90d5S3=%?KmDTfWR)66eDd@(KZb}
z7buyR1#6s~!&F@n=n(NKv92qKR2Eg&2zktr?+-KUfA9=OaI@)nc{%aV{{H7^Q38Th
zRWr>NRU0sdPsfRy$1`8Ny&(rj%mv4cl&Gpkn~HT=kZr}ApMOQuY)Eb@^|}~&rz^Y<
zXqCyiW3y@b&4)Ap?dyU2T?tJ;pB5J9>Dr37n;pqHKA&eQ3AHgiUY6_P+jG0=iJ|C<
zkH-sI!p-eK)3(gBV_Yn%YFNGFGOZ}7xZm$t7RSqBMhZz&DVj=gv)!NxY&yfXZ%|s6
zOb40kqpHwvRjE{$LVzJBjj1}SswIX*QWaSgzlPF=s=I|e6YUwTbOBA-3o)z+Js@>W
z^eZu5ZLBF4Jhe!WBFPudKy(XsIgvv^D~*!61g^RMZW6TC#lA{GQ5PDgL+cil!7b(3
zr!wsoeDs%XB!1FjzzlnoRy1UyUs+BEF5`jVR-)3Fx<MiE*3mW%^C9sMfA~-Q`i~#k
z-`w)A{!jl$?q7e!<?$1haJX@#8um=*1LsdK2q_tEZfM#DB_wjS9FIqapZ^`(_dmlZ
z#i!r=1JOrJDmvSB9MNq{)g<a#<3i+gK5=t*ht`6aJTWJR%?(mlOv=*KEp=7n{fZ2c
zS{YK^BDxJ~>Zz1OHa#efss~UScRHYJjnXy8&!6zY@{9lWzb6SrbfqGq+5wd#Zn@C5
z9nEHo-rvw0Q*@k{iI@}Bu%#P%#`B2urNhjXJ{UvkSznF^LM%Ij?d=_XTS7y2@x<{$
z))FNsXxuBR)uc%B3x1uL4+jp9&)nTV5Ohy8;2%G;>lHRZ@{XZv$iedT{YRR91C7T;
zM>;>_W90V!4x}L@5LE|NL8z|N0(FDb72E9%oL=xBKcg?-;2s^OyTv;&p&(g>5SXT6
zx*QM{*lSIb=)f-|n+Y*e<XQ>4PCuT9x@95831cckEIY}KQuw8!(gwFoASF^}?D<p_
zv`SKU1NcOk&%_)M#^GZ^D}!lwq$JRFN6LjHmJ-*afk+4|!8@E?kt9Njbd@Q6bkSf^
zP0YDSYleZSHCRWo3r*XS+nPv1RTUe(W7C&9r|LR<D(>;PTH-oWHyTyfOhUkBPdFWr
z3A78$#|xX=2cma{(f^P5YBmq(9E-BI8%S9aj7HtwQOC$So=D;<!_pn9-Iai6&d8L|
z34926>u}z4IGoGk)C*o_M+Ta<Wm#83^4CC?U_aFCyPE7Rx4ZUgm5XSl*bfbcK+qZ)
zEfo<FR+tXN`HUcuL%_}RHJF;PizR267ti@p@-bqjxGbdjkqH1$H;|=g$r%?iyKRqB
zihk397R;-~PZFy&H?Q9^FP3Sw_@G#pnHUp&*W!F(rrh0a38A2lgKbguN>w$)t8Gzh
zMF}VZ<N{9;nadoRm&E<H!TSq$`=0$~i*=UW?Hw)#7Q0f_75#R@akT`YsZ2$%9$n|_
z&Q+p}BD$3vJ)7G*s=hB6;5j1a1!)}qGPABL5BEE?(VQp8ySs*0`<5&-=SyT>Emftt
zzuzKaK(sYuDAUHes+fYmR-`4z%Y=~vqZGHhEvxmMCx?8#lnKLmqG|ggdQzI{DrBmw
zn!{;AE5XgSr>_MnCx*76ZEKport2Cun}L)P?RH!8NV&j~gl_Pm)DwkB#I&$XzT}!P
zXcG}S<Lwi1&6s+?sIEwcL?XEoT2#6wWv+q3as-JVanT~BAcjEj<y=2TOAHGcMb4^V
zt;rZ&78owN(pr-erOP5FMSl&+*N_d#_~NO!ioJ~K{-S&Ir!5AQM(0Sd3w|2$&e7dH
zus%PMy`#Syuu+y>kP!6K#4mm!xO?}8@jS5=iMAeSzi2o<eT#j3=DUCRcf^$Fn~J10
zVX=Jo;RF4)N1GZaHPY0i^F&I4*Y~%a&M!zQX@-tv8qr3gbaw@lScay?yF|U+QiX|T
z+cMk4>FE(A1he<dLXx8==5qFIXNUdp$71N)3}o-fcEPO+tIu2xPvqxEZtixp&CigC
zG9?+01UHkl#^g+v3GE|_OszFJ2FA;ZXj;6kk==mvf#vZLa;9rr&c}0+v$UFTzWp=V
zuT)CV4qK|thR!JRJRydAL|e`%olm3~(OMLNO7z&%6H+&T;ZMK&2b2+T5Nuw*E)|-G
zS5#V&r;*@RW;at;n(ezc<RtLRjLwPK&RFNm0-&yu{g&iH*%2Fo+1}E>|AIKJ*q86h
z&Gq(yzBVW~A+*Mh7eop)ZH-ovx^7r5ibO7)_LP{;2PB@R-OvrU<eWJ?eZo5nshs6T
zX%MoUg9dMz&M$Obi(L!wBt%cM*)zMyx?EVt6DlX}-hYiAZdp@l7|7@d(U*>R%H@1o
zlyFxR9y=|t&S;eprXu)23K3mbJlx+Q>jtGOj2X~MvrZSL@r0XBrN<kJ&?LlEL^;7D
zbVLq2(&}iX$4}3QTvVnac{a5Go$$+%S~cJVO>0=rM{-!oXK*=S*9n<_%ANYxceT5F
zDkZVbBKr+elrZOMyioTW*gTMQ4FZCyh`Ka@jFcDzK18y0*l}gGk=aH*eSE}cfy<KF
zT8uI1$}rC}Aw;CEcr=Rr(1Xa_tpnaISeI}v@~|6tv+ua;HQP#1SEZ_&odq9QP6v)3
zKjIfx#_>i`o1(I1^)!`cURS!NE1rXt7_Gzo07tTss;Tfo79W8zsH@Lq(3@hg(~68_
zoECgqnI=n26{~Z_wy7W?ks_B3fh_ESd7enR$a$nL$lylTgb)f7Ka?qMO2sK7wPLZJ
zZBsMsHatH)Gi=@yeB!(BKag|AREDjp_<mVgLntq7Q&G1qSt+Wz#w{Z$MZ8aC10f}G
zIuoX)P*_E2$cURg5Lu>$`~8l0Z)(<g!sSR+^^9Kf=8GL|ompqc=a)0{GIPJ(vhQ1@
zaeO*ntCDWT1%VQ>&>4;4<KuyV;;S#-({>K$BRPW)WrsaZGeRh~n;s=1eO>YO7Z1$S
zh?^&@)zr0N*ytjsk`k><!L<ihERxqyW=^qe9#ljLO<iwsensRGF3i~z-CW>zszM0?
zWK`9cI$*uQxw&KzM1cTVYcVQ9ayS>RTH+tfoSLd`$}>=kt8r5xM1)k1K2k2zPpyh=
zGUbSK^EF_ph@mV9O6K7&GoJpW#lUyJ|Bql7Ak%g|S!&vPz`W~`Ay93$$f!UH4o@$b
zx*^*OukUw!{j>MTm=IxR{`@U!IkCGfE0%Fyih?%7^!SK9J~6a{-P@m$Nj!h}BX@7!
zv%9;)P8V+4mPTfL%52|%jg4@A`i|)`Qq>*9!#moCFHt!%{qEl|zKpQjAsfwn8fp81
zs%}ZyP$v+&q8k);o-kF9Q-WFvj)xP=VmX{<vVfs$d3g1TzN$EX{)B(|j${mBnW=|8
z&2C^iPc(gvl&Vn0PS1tZH;t_GiqaCL2XZVg$}Ed?<@x!E#X6oJpLq59f!)m=(>O7o
z&O9g2`HR4tfAX5Nu7vXmefJu3`vxf$N<`9nfpsM>6U+GpZ3OpUet|X(N@>zEvaA!w
zWo5tJQ`eeGfQXS0QaNgfo;4+02I>~;1JA$z4ZC6BJWj-vxV!1u4I9LKCa;!tI??qV
zZClP@vn}u+m0<N2r35ZqSeBLGioB}nI@Z}?>Kb$14W>|PR0W(-dgMYVWV=N;N3=7<
zmE5e1rx)B<LVH<2(+@@Tv#gw7o=D5ec)oD^v%jIeeG6JMpFe?Ht}hq?QsCy1`QVBl
zA}5ly*z*(n+da}4Y9(>=86^^p*66y0oN&t-BQv#t`Sgrbib~hCU4z$UXHl{$3-P!X
zOs(lktu%X#l;j*~h8`Cqc03_<qDqb=HN&n1(Dt$=uK}Y1)B2Znh5uSF+25h-3PPdk
zX(dU{5L`JgRi-bbO+ac>7EC{2eBoNpid-hkpFVtM9A_@GMM#CunQ5F!DPfet+7-91
zAQZ+J&TT^kK7OtkhL(An$-?mAyC;6}?vDS>&+h4!Mj#O>6sLf7<Ygs0j}jS_VA~F~
zRk^PPpoO5V6+#srN2L{QFjQKxgv9C`OIk^?{2oF|w!1AMW?V=tHsZ*LEIE%Wi}S>g
zi8*3pye>WzBtaJCMb|Vn)>@DSU=(vgNmID>S8)!cD`+O8mHJWttaL@c*|Jzm${8&+
zpFe(vn5mnZhlefSeD@h$8P3y+kI;37Wtym(hGm&aam7w!(JA*EtQ{ei&nZhu)zoyl
zD1MXoM4MQxW4e@c>#C`UNz=47h)C151++6dZf|e+<@+}T>xg+J*mDsm)kWMBa$;En
z>v<+i!Tu)VT>_z)=83v8ENhucYrPV!rLHyaUq7&I8oIvX{{Ah8rx)6;=H~vEu5Zv)
zSwy8+8ci`oRE$>-03eEvKIe&?1==)ZN|#?qLECPNiMQFKO=&-c5NU@E##F@QNV1p-
z6$)i)Ote??VMARN!z(}*s>DLGD&Qq)uI>U2LXecDsEkX_C?!uy8F5uu%Oc)@Trk@)
z#sZT23$CBE81Qao)AzJ(gHSL}3%~w%|1<Y*-|+5>pEIp9>+yy2Y2^9&!t1*YyZd`o
zTSo8wW{bcg+X}fB?PgWCIOnm`L|xZlH1&QDLbHw&{r;Za-mtm9qi$P-HU#JJ%Nf-b
zPJCJ|U0YG5n$@n%Pv0X`#yQXL{@t%>YR%4<)VS$42vdQoP@=+^ie)??MW$(btoKyi
zzUX@EnMP>#LX(x`?a%*~jZu6&9JnkK?sQ-?46KX8OrEW1NZWUqbYkdg+-f23vFpNo
z9&y$&YzCU`9^oB+otdTuZ4CR}j%HnHw5Hz-1&*Vj+YB5I7w*~}X<f0W=aOFj{Ac<*
z2dXalg0Rqb1LxBV^D^=J-D~=r9a0*)x~4Tdj!#cKI>+o*GKQ+DNkXu$7DT3QI=bDK
zI#?p=YK?=o>3I0!3s4#(G9pKkb2K_3Z}yNgM(M(P2$}tx2V_`ajLfUYXfURzjFqmj
z&f}e<scPIZmF{%4UR9j}B{h?qXf^{?U4xQHS<Wq2XGw=A)@3B-Oz?^L{DRgR=RD3j
zv=n^)@GZxuXNH?Qs&3%s?gk?=#FaRCg11~AABokLn}7b_BT!7={f<w+{uN<a=r(1c
z8e+r+OWh8I7D+~|7gq0!9+$*;JTTM^(wh?M5(4idjjFIAlZ0R#&s24TPzGlgWX?D{
zvBrd`I%LxjVnEt6r^A`M_g^BbazvwSu`vE)zFIBl8jVjy45XDH%eJVny(>GP>%&)2
z%zqxh7pY6}y(D6W#rYqV#Zr{rub&q}a70yhWo|8jKQ9DsHVCP(LNXdzHYut`h|K4g
z7yevlDz>z;rcRL*0_!rPjpqK%-H*6mr6gnaOqYPO<*1?Qi%h7%Xad`z<-9m@3`ha`
z+AQe0j;3i*bwzDz?Bju$Gbn|RU{he4Eka0+%gpQ?Ra3F+T9yc__ednQF=%7ZbxYfI
zD6J8qw2g9%g;{dFVhjjXEOw@<kWz8F%mgnH?+q_62ZGPkZO#3wd!})uX=`q7ZdgMk
zAh6CNk|Wre<W@xVU=-RkxOFOZZ7r#W4!KH}G+~`Z8&zIfB2m=_=Ota=LZw)3M%N_-
z>Rcei0`hsdy(ugTr*Pi#_g}x_<vg?aNK<LNa~vm&6q(JYL6Y&#F^)3`#W>EGO5$T=
zjgiHBx~`*d4Nax+Zbf!A@4tLUOrBx8ft-<A(^OUYY`QJF+n}!ZYL*(^cAzzJad>-S
zSw2%$JxXap3|G2jh16w7Z<jMEII4CiivSdDd;eqY6fa0hT|X;P0*3VfKwY&(J*?za
zzg;FlrH`7CvO#K9CTM;IR>%p7LJ5tOSJihez@iu}emVa|_vlZ0!ssk9K&2Hv#veTc
zbxwFYp``&;(eHLt!@#x?4BHLI=NFEj53DbbRQ;B;ST3g*s-Y0a>ROY$W`B1N#^79F
zcl(Ol*KdfqIH8p&Jx^sCOtV4V+#`S-J%)f25;bh-9tIx2{WG8cm;V>?<_7)#3x?eW
zyDZF?i7$WlGrWuC0dq}@N+Iy;iqx4D62UIu7KjchGL<YH;2~M!8u2Nh+YT=jqTNEC
zxy&nn{MY}6yPyA@FW=mg+j6cWjN$(FmRIiu&!0bWI35YsV{IA97cim3x)pmK>EFGl
z?K+-*|0}AdrP*$|F)j6W%lYMiUryB9E%VbO$4}p|P7{7!(6wQ>y`$?l46oiW?6y>0
zOC=R@nVDaXkR{#jhSgh=lq9K;LZM}b#*|=63^a9HTnEu}f3v4nf_1qds~+7B1(p#P
zl6PfSU@Ah@l0`?=44C@|-0?GNT`;C2=ZIZPPu|oO{q_#GyTctmk(ZVB<{sU&kU=?%
z4-p$Y!)8x&Z6b&m`1bd|B{+-G0%r?jAXtZ8SE{xtm}eA~s!&SMH4T+aG%8`TqwPw~
zDEb-i9Wgn4un22eUp^8~kL-3m)>+mUOVic}UDK$RkOJ7LFy}FZ<Vk+T`G8avNo1B~
z!Jp4m*Il4g0cC1(a44<H^dlsK4|wk}O;;+2Sujl#m&=*1Z&+R8^EcnHfAg*cBbEMf
z{P~adnje2n6<;BBLka@XHV85@2I6=^YKe%Y0+m8B0H!NA%=>_y7LKFkbeaj?W0#eb
zB8y$fNQ_qGD3D}A@P#@U6ULOII!Q)rjf~*dl_VpkGK6F}&nr1^s7-}jW)g|b{Vl`=
z860j|u2j23yY1O0L2n}&!L%$?D70)iPZO&PEY=fJ>HlY-t}CPxJRc`k?=i|S&NDWF
z^ATr9F0Q=1eG)vLEth%at}2y3WgR}3-C$QM=C1ValK>+XmC+bE;Otz&x*=e#Ek?+k
z3zNWmgejr8{dR{(@OT($>k4Ix0H;j11;^8bCUF@pbygTHkRp-2FZH#qM>XZXmZqYr
zJKQvqg~bQYupK~^qcSNKF$pf`i7X_imx<?#=kCotP17=8F3j1o*|gm4cNmfBh2%E&
z$Rv0^mA{$F5h);w!f5az<7_N#5(LlZGY`9-;2j^H&x|%8qUZg?j{VS+bHqrlTn0n%
z8M~~dHtZwy%?7PCnM6z;UA~<~AS6KRnvhG~Pa}vz5JRa5#$<^|l$DI`$m!)7tr~K)
zq!g+98<LC=R&p+U0JmCVTv4XOXvNiQT&mJap>>Vany^j<QLf5wWa9}fYEUKQ_ycdk
zuOrbr`t1Ws_efC_WVxSysO<iiuAjCTkdo)iO5WZgRbVqHDg<F&`1t$Z(eC%;l#nSh
zv^^CG9URNyiR=U8GQr_U3?;153_FaeuJxG?9~1p{2aCn@9Z3}DxW0a|f^#4xZCzcN
z-KA=`*a%saDo@wa-QDr0-+u$A8FPC}Qii5iG$Ny~ddPq-)w^{r&VXh>YC|0yGJBG&
zXm)o@rz1HfmdlY3(}~-MSF~--G8^t6cC?$G<|T3P3%B<jH4Dj}S;k!GXEAVo`pi&Q
z=oF}hCZvq?nZw~gx4$7ymd8JTLl%kKufD<vNGX;x?yw+(XL|lzIOH)QLL|k=H-G$w
z!}Ec<s!&GLo=0+m{{Eh?yN*t1)_Eq+BTCmOquBIYniL9Aakz!xsB;1ZDi;Hq)QaVB
zz!g9+mQ0R;RyLfUA6Zt%elwKPlBx&^)(B$Q(CC`w>4|kb5?6<;n&M#35WHnMJQ9|b
z;mwy=(-3uosam8`gw=vy*|ZgX^Qy?H+)CdzR85Wg@++LPT*fnX-B49EOI5Lb^@?E_
z*bhA^ChDOFyAqZKgvQi0S(NHplCm7lXi%FSezJ_e`}er<nXaqZ-|kpK#9LSR-&F@f
zvYt<9Da&rI>WMiKtS>>uQkUTP^u#ib^nJ_z?jBiHxamUO4M?F_*O{nlglW;uP;v-L
zlh&2ZFp#7~D7bz51v({yOH^Hj=xdDn%fKOjt=BjOa(txk2Zrq}n@x{0l6<=@$W?EP
z2nH}J(V22aFQj3K2?w|}a5<iFb|pBE6!5Sc$Rab23sV$DlqJC~3poZ%ZRnet${4gV
zMIjthc~Ke7&@}w)^`5(3Pl|z7Yleq=kRJc?guBc@CWlZehLgp+fcM41H(Q4`6*hR9
zrek$;8Pw;(ua`zrcQs8v5Q4zE$P^+k<H{VkYR?^`6Ik!LtdY~a5+ls>O1`@8JEQ0~
zEkn~{)5PinA(llGskElN_#tzCt(!?%G6X5ob)~6Nw=JrwdA>N50DU>~%b&d=MEhfZ
z-&qGra2^+4ZFUfSnMTHlX=-#=CYRB!q!0=kIbUDQMkBN)=Tz!$)-o-QoBLPj%CLWx
zxnB}V349R5x}t8HQqzsT0M@kRGEK}EOYi}a5;g?dx}s8oWpSM5h2!!>6b(rfCH(Vw
zWN16aHSjVo?5YN`WE>}i$n=fD+Lh-=%XYiLC`I3we!f<^uqkpxi893$(UX!S$4Cqj
zA!|&%MF>UkGch@|?s3cEYFIsE?Lx2tZ92ko!c;ZJG}y2fWHz#xD1DhSN*M^KADI^^
zB~05E9)aNMF;M8LAqI;w4N_I5u9^}dI+AxJUydE@dcj=DoI*${T~l?tzvv$QNsEDx
zpFZ>O<_&SRYzM=pYdAh1NUEfxiLL}HM8<o^<@`dU4Suy(ob5u+nSQ&Ya4!U(zxxhy
z;MMytX@(6hX1d)SRb3OjEi4e1u(qHUvya3zf~u~GOHfK;V}K-i`j5Xx$;i#^hSOz1
zj3>mV1}#hVZ?W{dEq?JV=ZT;+X?TT^0yiIt>x|Z#m;?fWPl4_IJ$2oHZkbL8M2cuU
z!k$^qPedWP|MDyPhc}3jn8y>B(-GY?C;`O4v_=+RI<@)f8>CR=x`E9db+_<VeaUn_
z@cH+@VS9JW?S4bbnP%8>urK`jU;j@erI0GI9Xd3T^ZCN;JWEWpw8$X1{o);c+w-Si
z{g#*AKvh+2`-Wy1%GgOLB%16`*q86oT2d*Ezbccfp(Dh=IFATZ#$MVO2mz@p#^Xc~
znu~{uOsy>E=VwMLjOjS8V*w#m2Bj*D)`Tok-4@-{Xty#So;ZE~Cj^3~t}&{dvo}=*
z1P~44+FS^BMaxpv+}_?GrC>L0nJ#DMWo7s9mj3n)+szh#`iunn{hno>IA?*WT7*=@
z=x}zX(i)K?wble#({%UD(-F1V5=17f79$PywlCzeZos*f;8(g~%j_1Cw?rWkN>eF`
z5E<tca&t$sc}=swMRgq-N88^NoOKMW_JtS(&9El}^W}uj7L_BWY8d*K7+h(W2!jx+
zz(xe*_Ad(o{k2}(x4)n@qSRvd_o!}zRHYAZfCeyi4+!dRgRI)ZoA;h&UO8SSo=y|v
zys$3wk4WFPGHiy1*ZU10pN~8~9jS!m_5E!@iRX+~WpW^-KpTUz7NV!3;-*#Hbme;{
zIz!tx7;o{b!+XbioS4pI>9Z?^5TY<P@-=I;F^tnhM4*(W>3j6BW$~VQUNKEwjyI(w
z<^;OpWn6h&GP4gL6V^G_Byc`(v60C|?BY?B3FgqXbdBcC?Uv`$!ePxU&&TT+x%fT3
zFW!Ze3o1Ou2m(xHew6D7A&{o1k6l#E)5N>Go~G@X7tb`6-oKQB&2|UM5HsZ9@zYe;
zLsB!nyr5;i@-YNLUh&FVP^pUXa4eaS7!XSG>g^js0_Rp@3>;r3lGY#%-dgIm!@0mT
zPWbhDWF!Ouc>DG>h{W^Du^gl1lGQ!Wj?sbq_`=P$Cke%2oVZMurqaC8J;_-z5pCp;
zn&-Z2kpgO?sTzZ68<f<fERb0s2?c<aQ#mG;8f7YCSTXgUs@@X(ipYi-9l-~3OeL`|
zYJy)-(jW>-6{RFG*pm2{x)1@ALsuF>u?|Y5ODL5BRMV;}=wvdcsvu?jwQ(SXDEX#T
z^w@}s2*kKTF7GWFLkU_+U5?VS`it(-pR^e0Ha*T>&=y9Q8J_|Og$)rcYwYPlNP!fK
zhyDEgNZr(AGI#g4^xGS{?XHluLgIWLk<zf=-O_CKWGSe$#)pUtGs-BETbV9v>5FTP
zA~TLJq%4tjPu=whDOsZ=Rt<Tx!=0XJcQ=fyr%s8dKm8fg80xx3iyL}TF<y>5y&OsX
zmd);#JYCT4$acFy)g9B}nfdaO?RLX<yD!u`w{lq*n#z=pz0}D4TUu37wKb+PNEw+a
zDCM^4SeLQXA5BF}iTyCJx!JIP{lH`s-~8z#b~;lxFzmOCFK3jgiuEicx^|2HYC}Tt
z;lp=mCAmxsIR&y-eDRB~+3at~DV3l}OxV)_qb1wj4Yv=kh$-UdGtqgJE;!@p6Jb1{
zjfP=Q)o+SBUN?xmvVHX$?=2z;bh{xeGw0KVkB>)!GKBDg@H6ip_9PFok$53^{q@h0
zMq?L;lo=U4K}zoLZqapL_~=!`%eQ~Vg;3;3O*t--DUxMZZVsa%rSf1)1vHhsC&gGK
zBKn3+@3G#J=NCpd;!aQaW#;s$r5ZM9Q#j#*j1d_hBTgtxJCtCR3xt?yOhpWtU?+sm
zRJBH{NOTj6oyl5Zniko#1YdN!F;gZox}5E)y5sJP*Q91oHS7^?A&#HRLmJ4k{0v=h
z$W2#vdwBe~fDkMp(0BVQnXzbUl@LWUTi1l?A9uBIyW3M&s`R0&`pR}M54kK#KrV|C
z7F0^1rNqTZh?$fnm)UZdMx-v7+GVnMyRh07nIm;2>FSDC5BHp2Ua;$0TpvQvG<A`O
znG&k1x}K(~*!70HZDD}4ZNtq?Pb(~bafH)>^>ko9T$rb&nBovf1$EPsU0_<)Qv4Q*
zEJ4&Y-W6hMWhA09qzHA}bD3vs5<EQ~_<XW_9@k<dm73KC78gkwM(dfaqZ}uzvIrhp
zgs9nX2WIQpZE7A5B^wc(Wl0I6^%atn%W<mKCEcGuP%;Ik06{^%zH6~p^?W5%Ij)L{
zkB`qpAILf4J>1;f&{T$LwK%64j3UVr`kbC$t~=a9yE9cy%#mihVc56K<EfAzSBslh
za)Qr?54e!nzuHqZHJh%(Y_^!HW?fc7%$(*GqX#x!MNW_@VNod}H`^YaBi~w&_vsqg
zRb(M~nM$_fG%bwdh5e@Ei`zZ>zNTpv`|ZG+H@D;zS;rD?u9^x{X-s7boT{x#z*8$y
zi1-kxC~zxPH4qYBP6!l6S16T9L7_CHbfRttPz_QyY-RZwC{-3=$z!J{kOtXq%FaHO
z&54l3uqmbdG3y~jhGdD(qEv;mWdYT6C5-6T5$BgeFw_zu4N})+p=tUXjOj|YDVGg+
z%Bfg!|AOnM1qelEnO3U1duC;rAD_tXj1rP+&?s4wnCHs@*=-o!yk|Zgi1UJOI_|#u
zWeLc9{3CWgBfP~_hG{+%51Mv=Pute4>k+$N=!YH4<%o3_Woncwa0R<ecw68`x3At{
zt;L;RaO;9;TKby@`qy8xd0SGy$t`GEnBR6<_`|P%#WG#6F%zc?OlP)DP1|?Kszymc
zYcxqI(z-Anp6T22a@oInLsE*i8<@wDy6(BWywJD>aeTql6>U{hF_7$vx0aMLtF;W<
z9ZG;Zen#(WUJpCI9Uqye@&8TTfA!jyrRka2Gwig>Y<pa|jC5qPSVgkXqR>r2Qs1<I
zh9F26y3wE04-g<hR{|u+G$^2(EsE7utV&jz$S}ulx9oJo>SFB7)@9}efyxEMRp6Y6
zbM~5Zj`@A>`#ea!p)ob<n;S%q#TAf1o_5rA%Zuwvwq5nixsHruNycAY-OyiNFb^}j
zGFYoQANEZ9J;#sV6UHN^=~1RZOGApT2%ogZZZFVPgOH|(TcTrCwU}xHaU@I&^=gIP
zTqD$h!{Ht0-2+)lqOmxo8IFP5zGa&u(a$X7K)=~ytwGj3T7aSmX-*%%XW1VMd}p)4
zj3;UqlwXQ&)O(tBaTa7Zmp*gKC`2)Dc@JU1%>q-^h!jx>loVKN&==Q)&Xtj$sS!C5
z=kl<1%Ne)KtS$@Y76ea+Ls_F79|+SKMB?`KOQcCfT%;sR2pHWXO~n!tO4ry%l-(W~
zX$@JG%8<&jT;!(*PLJ;k2{9BoO)?EREKCo3s_jTNhI!aA569vrs5?T81Qf<v{IlsX
z%`<gXGfaONQuH6{r_k^uQq^S&<pM$$qh|`C?DkUvGV^d`7><M}KpLiHDx}g}!mCCK
z%&I0e2K&51JPi}`5}B6gtU(s#ppi2fNnJO9z*YuRSKMB0>5ae&aMM8ls;5<%aC~Gr
zKQW(nWC8Y~huSio%7ZOt3A&7WSqg-wZ+phY5mUr=JuXFb-5??i<HU!jBT*=x&I{k)
zpEx?WKTXuuVq{{NCSrn^1um3|oKz4qI3G&QEIZbl6{CyXy2RtI=J7B;A94&i_|j#U
zMcO4qsT-1uJjYr?BD45JN&>8=Tdg>a3n4nLR;AW?adnL{nj8`_2VBUcN+aqDpFD_6
zjGnq}5gF#wk>Ff$B0N{xgv?~;Iov%Fyhqy#j3v2<*;MpV@ZsYJE^e*~Au&#Y{nMV8
zx0lpL7vGFEtou@>3~oV^S#=fTdBK<#Q`ammv805l3_SB6YOPt-mbSLMxa_c6l<A!k
zOgS^p3!6<3B9~Q@56r<qUZLBr0M4W;eiS$1V?kwWsS(e0^zh6<Nzs#|r>Qq2RB6G;
z!Yxs@LurFsPQ^)3Z_!m($dpMSMVW>|eBSJo#J`e-lo*r4XHx>UVJ5{HWlA_v%L*km
z?(_iKqU#N|-lD5DG5Kfur63DMRrmP(hj~r@sKY>2*K98@>AMXW%XBz0?j8^#<0&Mh
z)%8oRZ$Bdm2-AeBG#Be@+;CvH|3L7O-Q9Z*=aIhckSVd=Zt%;58wX+tOv9n1S3@8L
zhq4XQ)VT3Xa1K*<RQ(2HD}>T0m5?N4SkR(lefv4x%P+Y!wn%+KK&F{A4=m?B^SDRg
zxU3trX{h=hKQFXx4aOp~psIVaGK`0Nc8{N6T_6r#YMipfEXk)6n^wWvpt2+{Mb8>s
z!Fl`fgjR~x&C6$AxMLhg{PDo~7iZR+4UK{3<_6ub5wgNomdCf>p@5_W$uA@U=~-x0
zKR?J;>qJT=7?r$7D+(-02(&h=FK@86A_97KiQT+J9NsbB-IHTvj-ayOqE;xaserfy
zqO|CXSLFB>%Zldm7SV2qDRFVR<;Oqy9CBbr^Zrz-aCO(x+JuT8;!@tq`YUp9Oy>hy
zWM-KO)4(_!IS&K6>RC00u$0>7aaf4ULep4S&c!z%G+9XIc_uiAf@g9fi0DyG3rJE-
z#NQFS+6tvD$~J`gh@U+&N$T2SwWKl{BVir~n9tNr%QOkBStGj*%leXNEXK?zYq+c`
z5)x$$s$H|L+VW-;<8zItq{zczPZ$s6=<!JuM5z#HU6-7J)Hpwa$V}r27iaQ#Ac;cO
zj42WP#JG6s(3IVFD6NOd&m4~9A9JbxgZubU%z-f_qH`?XA(SG8gwg8xumDgFdba71
zAv2CM(=>599hrs+o*^wt6%F|5JdjgHYRT*a-X~Jb*xD98Nenc$!dS~{-O+U|V@MqK
zXU^k^7p>qg{`{6}abo)6R}3G%#~sfE7l=}zw8k_QvZ@iHyp4toHi6wNOw)q%2~#Oj
z2t*=F)J%gTp}0GFqLhq_=fl&9_fKc$AXq|RoHJULT|Bu=Msh0Ez?{o9Kg=_yVdTxL
z;tE)=YTo26A3yDRcq*hhH&ed*x)|&vC2ak>CXkX6Yb`E#jFB{r;nnSyoFpIKf1s8X
zlk;3(T~JjOAs4>JGDY^&OkdRn@#`kyI3P$2Pfw5`<xChS#={ehF(edOB)ks=g_|U&
zr!)7v1Lm8z{QRH&m~OKnEP=30<yw`3b>AWYA0ky%;e6yYPDrIuIU^<9T&~%xf%B4K
znvtqBk7CYPt-0K+sI?+S&!@)$Ei$*;1~*IyCD6uVjplqf(N+r4Mx-k2fA6O9Hve47
zbbbL0s@o!^L}*+118yq#_*9w!1!5JDS|MerZ~A3^e$Weq)R=mWFf}nnaw-D2EEQUr
zf?5|6`)no6xrF_cG6*EWC30Fg?;f#rPl_H@TT*a@>4a`NgjkcZCxn5N6485NaF}{U
ziiu_U!|ok_)M21AnsrwbPX|O(Q(1#+JA@RNrYU0yT^=sy{edt}1q0bAOz=E@|0{-M
zxO(#i-J6E#>5i&xnf=1!rw^RQZ&6K+sSTU$hPE^1WQIoR8fg|pSZLawoDw1>qFboy
zhOpU?+)UMPQEgu^yUR!t#X+5eC%KupjLgG?wH5tlOSiqGU-z6J9tdthNy!o<Y~Ns2
zf!$QOjFWeF_e>#>s*d{N3X=oVIFW+D&yM70noW<jh8zO*>Jn8OSQeIPU>eV4sS(YF
z!Oz4YvR+?<QV4C!3D`)CwNzFYEGDJ(5D_SCaX!$s9a1K;?n^g*o~VswIuE69DixWG
zU0o0nbbW`c6uIsZx<Uz$XgZ{X@wCSx5bY(le+m5+A__#dh&UD4jnpN;5<Rz@LLSVz
z=5kX|rB$~sVa~%HIyx*7V<afS*$<=~sG1JjcWf@NS>{}TCcP!d1uqnJZ;;)J#XDq9
zRPB~z8juQAFUS~>Qj#*H`CO8?QlW$)x}j)(5ygH-S2*BqrmiifTaji*98VA<ZW@Vh
z=5o^zg`mE<MOH1|G&o@(EXWWLZ9~<qNk~FWB%x8ZK}AJ$6UJDwGH4+oW|rNAdcHww
zZJ)=qnelXFzq=zZ1KrI_Y}24zfZURFi%SAIkJx^TscYtGWFF4A5D-RV>W=oJ`(rNE
ze{i4ZCUR|1Qauye014VC+ODN)n(_t-thQI&+`eL298&K2boW3?iOMR5aVh|!l$ocA
z7(Cm}8e<G?0~wF4EErA5@aFc4+9)O;h#~NBchBx{K>LZWw-rCTz2Ze5iTfXzKD_1l
z?K=*S6S}FWE*gw#ial}hxMeJgVr__s;z!8fQ)c!F7ZOjC=ljPKQxxn+#}9W;tk)}K
zhGC8zCr3z>@vY3I5%v4=e&(5zTq?Fo;bZ3G!y`Zc=@;c*C=`{}Y}%Tx(wxQ_=RMI>
ztg8wT$VM`|<+);34i>|3Mr+OW#fmpCH>~=aySo!vC?@CF_HDV329MGTB_%#4k}&AL
zMQNB$3rQs6wBUx3X_`?+BT_<X$r6gFNorZ(I?gi;6UWm;NSX6#<mtOR*8j<h%c~V6
zXspH75?kx%pK+;5+Dh@b@A-7M!_N+)qcVoY!(j+KoCc(-SP(3S5klmmR$Q%HE;l_-
z#}m$Za;jOz5oI+YK~q;$#$b%Z%_CZ5mK-pRFAN=9av`Ewl@~f!SSiUTatsI*w!I(*
zSNi)YB1Fl9WD+SWvNRwi%Q7JdrNtrg^YjsfsgTMdl>w>n&mm8M@Eq0@LjHaY00yOP
zVN1kBhzpfzpx=Nl!9us3vC)uG&;Emj94cyU(fnb=vOnrDFg`rNG~(Stv)$q&L}{K?
zqXAu8w9)L}eot;YHZNbI;!I96a<yi?(J;+a-J0{m6WUtZ+s_L&xT?AP>%Yd2Bkj87
z`sTWPDVpvXSThyeXH1OeBb6#F`<P(ShU4)_3=!#O<kKDQe87+j$2$b#GNEk|5Usk7
zBr;96W_)_&@Zl|fnXy&Fe4H^?pVPD(;_d@!T0l$s^_uSH1?TgT^LZ+}GFLO4M(WiC
z*+?b<Zg*xA%a?jMj>JQm5(pBNvCnm<@^_|<CFV?ROi3h1Pgo{4+YJa+#%?(iyh9^O
zPhFRXqW6LEG+=FoRT?3nvIdmGEuJI<*RNhNjRV?^^tC0`nz-2#<3e2-lF&@E#wD=*
zC3*V-*|cTU)pm@BJI41vu)2v%`zMYcK9J`VtJ_;_pJ<iDYD>}*FE!q^g`RhOM2pD!
z>J^)pFK~V$imFU{7LRIHw3{_TMEbsBI1kL{GeRl+;z&v(x=Tdep^>=Zgg>2Fjz^rE
z(Am@12Brz6O07mxl(>HOIG3pG0)9kDNmdKXIFW@S#RM@Sr6L4}$eC6OoEsS*?r`%=
zZHmP%H=mK%V%jxI$U-9{J<Dt?F$R<@{Bl*-RO?IT!vV<T<VkUXEYLEuS#@|@#7Ay-
zPopL4?Ip?7%)X4rm9Fr?A#6o!8{+wlREqWLf<LSr^bhwVl|gF@vOsX8kffBzBI8`4
zLwa9UA@g~_yOHEP>3QOFx!F)R3r%C#@6Y%UsclD9S^B;&7}@<9gdxO86q3#=x=M3(
zxn`agcK45jVaJa*f}g&8!`Gi}`0CA;Jl->W`iPq*T!<_o;+^NByGB^Wc$`S%!ni+k
zI*(v27)w@?dbL4X!#tii3=VG$`@u7(OyAU8uY2z2Ky(YMeuZF9juC+_<MdpbPz69!
z4hIN+Uwyv2KhnIqD1kjd8;cO0i|vN~{)rF*htouek;T?@eTS9BFLEAdlmwAdfhKKQ
zZeLswVqh9)-0X=iA&FdGZqZH6>0w9LH0Zj)8qE?jVez>A4nK`3KzB_MKt20t6wiRD
z7_mkYyelsbS3O--EP;$3DFlb7GatVBo~ze42)kl9AE>*QzO_u}kvs={2rTo$b=Okg
zO6JA$@&18fawt<VhQQz=NqQ23aq;v<q8o`3nOaMJ^zyn`Es4}khcS{(T@zxasw{0+
z5f{br?hchad6^0mpthhjR#zx(5LHc1zD%D~2@__eK}b}JcsG=wo;xFS2ST7sOI8&=
z?9gOl1`!;YKnjTz7n){6%n<xY&aniC;W_sa@qQ$_a-hgWY;#eN<_ff`&{81U3zV&j
z??4Dro(S;-5tLbxqa&+lLEj(n`J)a4hx3UgIGU=40LQx>$CD!fZKJuo*`UJ0YTNPg
zx8IXKed5*EUy+3*3JD=Ge|*RC@hz+EHLEYb#&#Xf&1`R9@cCC?G7Tq=<B9$Lfe;qj
zwxc$Rd0OyMkbK~{JJ2)@s;zM`5T*$qz%7CC!+TES87(D!+j4k%!cPO|!=CG#8$SEd
zk7&9EpNs1FIL$26SoEJuCMZRt9~g%-j89ksN?THLG`c}`6=@pK!V_D|c)#a3%>+|Z
zUtVyY77pV=<re0}A(f)B7Nb<rkpjpv-XbRAJd#vVkXj|4agv7cEL4fH1WO8`F=M1E
zUD?`lKA$;_C;F}-5V6*lL&f>b>>O=nh{0o&KrJQhY%7B{ZAl2njJ4HskjNp)NLN-4
z>HLU0Jh4PD{S9X-#KjF_&_vUsl!a-JLE!v=5f$-#W>q(IjU>f^INYPE9)&8}RMX;5
z51b#~VYfYXUl9rPl|eX55|S7-rtQ(TDNb%{(Cvz&_`vtS`8#&+f8eVheSv7MiZelh
zvQ-Hbp+M(nxkoHZ*AVbwKx%`jo<SwkKnfl=4V0;1B1?gA11fj~nkD<vSC<;03V*>(
z3;tP)owOn6BA?><tpH#=Jt1>M8^z)mq%>G#=-2Dt*C)L%cv~$?zgSyEjEOWIu?VPY
zgsyNt5S||j%lxOE8Q5Ii(DxnjdBRhynShcg^?aau*0{Q5WIP|44o^HiJn?k5<MB9=
z5lE%Dy;_wiQ7SfgA0lo^^li=OuU5Q%bxYsYSS?Tl`liCoC(^Rz^mvV1W^UFUFJH7|
zcVhbV1M|Zp(_umhRUT-`u?$DlAR%O=N>rU8in{FCeIQCrRu<FtT=X@sFD>gz@jv<e
zbE>9c)fHaHr-viIxu5v;rxSg?=EHvAG&y{JUJK+@(&W!9`IPv5=YN_OP?G)WNZ+m4
zoz4XBAxAD(MJIeX70Hed4(IeUbD=m-a*9+<jd)%m?sj|Bwx@48mSv>xHLqV@V~yeI
z@e#7X)(utLF)fb6aO7d1xoH}t68LE-Q`zTQUd{zCZj8X%n!2g*(*!A?q(F;ATU~NH
zdrsqwl$vE2*nbjiZWN|62)7W%fiO%=b8(2wJ|U!Lwd(2m%z2pD9T#eqSVCfP31NyI
zlAN$jLt9x$r50?JV7=;)Idb>$ft#E4vw~RTm&hy<=iQFSxA&aeimPo+)740ssGD^O
zuZMsao}9~}A_+nE9)*NdCOJxJoEyklQP~y29T7UCss^bg$t?&~R$xn5$T<-G#K}2?
ze#Ri{{~_R)Qpp{qv~W7_7{-~ZS#f=HTQWB1ndnZq5NOu7r09^wqD{*(OysnHYN`86
z!s5z*d-;z*AbdP7+^%oAeQ`_l1Gdv#-dy9C!1?fqRGBy*sH+tlEt%#s?|=Jin%ZKj
z8h1L;bS+)iP;b^`B{}atk=#O<M~0^ZZ@+s>5*d>OB6!j~(e-O`Ot{67NaVi3Uccge
z+!2<@G6$lPjHAO3XBs!6l;lY1m}+gAe4<%Ar}L?#wCe_;B|<Aq-6B-tILsU#_gp2(
z<;#~;t)`9zCiC(ACuZl_-d+|@_jrPkNy<>IYqo8NUuIORan3Ov7lhRWA9&cGI2{fc
zqxt-+FA-W|O;r-|VJRKc5YSQ<iOus`;&&ugC?Jve%o|r)7aEy0$eeKVM88_0+Z9r2
z*6ju6d|<!dK}wvKk-2YaRvUz_$WoC+LJD7|Eny)J2Xwba3W-dP;r+MlKfL4eNB@ZJ
zn=iOh26wk&-o3>yBg=Wuaz5g_HFPa#gSQS_86NIGaJu`1s#d)E$<JBcT#@Hg_}g`<
zqz1Pz>~};r6&Q>(n5KEI@st}eXte6FLQsiFYbEyPlB%-LZ&I+#Gj&s!X+|!XVIhjt
zBt2t?^|RzC1dkz?!=IZ;N?@PspV}B~+c2EYocB9yQx}YISjcW9E>lrdM^A7IwrVg+
zqjCfaesu8xDKZIBfFMGV7RTZ49ZlD>dGV4M1J>A5UyYHt4CLUR6VhcIm~$W|ffbP$
z0^Uu(m*4!UezZ17A@L!AR?j%y1sN1EFC;$@-2gcgPY;auEsu8(JUr|<?iYmVQA(5*
znUYM?!eM`+={l}&ZZMT)vugPK^$jm?uBj>uB>XZH+{|>&%;yE2d(y19+H}-nVS4{9
zhwtCwP9r&Zs%}Mn(V;6t(>0{$6_5y}l`_r~A!c%xsHQ<TrFtt;BDonEJjxnOTWEtJ
zL~d@@{Kd~9A{e~q<8I^^kAi>wZ-2$#{Q3i@5P`x|$uXmZKnPG0QdrPR;$r60<BrGO
zk<F?{$i&sUW1bhLIdGaB^E8uUpssAuJCjK=b2=`V$|6ZT$;7IyxO;e@X)Lu7Ty6@Y
zS!9q})AWU-A*5tj0#9eb&8kJ@h>jM29`Mu55<J$Jvi2citu9G@k@3^WGR-XWLS<`E
zk@e*ojKcIQ&g0Cq-vcFd?z~60HLI&jq$$&%#YMa?LZ}c+P4BdLW?#&Lzk|jZEzu&=
zT19PSaiLg)c*cT{)57^wCd=AnQWl(tg@=##xM9I6iP9b2rl-BwAgUH=3`z0~HA;o@
zn!)Hu^b;`^_eJyz<9SEdzr@WWem)~iz))c-i$@?b7^N`U6wP$b7}J&<S#kwv^*qX#
zva}FFC<fW>_PY3Hq$Q=myd0>srs*!wrXsk76e1yKTnHo-m2J?vCWlm{b%*;u>>B-J
zF6b}+>0k2IU;J|_t@!4@|2NFbJzWD=Z(b2zzaY#bpYA@;G@+<JQ-$f)#HPZoR&<*c
zmCP)&<M`nn$8W#I8bg{!#^ZseU-RPiEuZc`kwj!XoH;)2xOyoN$`Hp9VnS%kZvMdT
z!+Tr|Tz~x|v^GrIJl8KAbBMT*@G&Bkqp}ubH0S-Uz!>_3^N~1>)YhOkUD?Z5hBO=q
zJ5ARakOgDRlz6&(hu&O3Ro>2{NQkN@3q?Q^mH|J{gb<m7qv{IMQlYRh(bfi`ASQ=V
zg^BH*#D##H%E;PCL0ws523u7atB9dsMwQluYL-$V@F9!BR4TNo%A2NCRI3&3W{aC=
z#$k^;JP~wFkhTCbLM-Qu6p=Yo)f(oR@bEnWPuJA++aA<s=t>bE-m&<JWjK>UEMNMR
zQCFW6HrJS{FZ+3+i8(Qwmf3cw;JEw#SNQpWX}64S!B#bG-IA<D(Xq@$F`9)00V+6_
z)1jzftBUdA#PIlm$9Lb+)iw1OUlN_;JPajxV-01lBGFo7Y()ybz(b5dir-NxOG0~|
zW+0bIfh~Lem@-CNs=6ghiO8X7T89(LRQTaY%CX={DGE_3z>Np;e8js2=OUi+v(wsu
zFCjyh5{S%rK7h5<ZHH1tG!j!P%z|a0>pHB`C>aqdlQpDD($)<L#W?<H69|`OM$626
z+~KAJeP=oE-y=exvW83`yOMeqxl~}sX(Gmj?WSi|72ap|ha)k9v6kDH*K9T`E-p7*
zUS3nz8lwcxJIIlk5<)-}5@}W#E0Ns^+gRF4GJpJ*-LHSi?%@nM)3=gtQ)B81lqQ9;
zCq{s@5~($-ev8nm{9`OaYW(DIb7>M3>m3B8fQoDl)uu;kix3GjpXja*{Or#f{^x)G
zhX3}j-|~O>Z-2=zzxzar2_bTslt@rAW3|A~3s1Wvr{kH~1+K3y$wZn;bJcfb4DQ1{
z!yJeTf){0~WhBcym2^B7=N#S{Qp$wvkqO#<RbX!90_>^_5_}m)mU%>Jh0D6+WwOAJ
z3+^=HrkQTLL7Fm!sI(xsg(W8l0TCljQ_-zDmc?^U0WB5X<%;%l&8zM+lIqw$?D+Wo
zJC?^i?zAxJ6Sg*Vm0>uKOmpUVT9_8cx~(`I7Y65vsYrsXQD`Bkv_$2AmX>K8*sfc0
ziik>4wJje$J&@w29F#)f@zWlWGuz&B`RaoH=8|f&#`GP!>cO_m&yy8oMr3H3h7=aY
z(-R?qX*x(Ai6W&)@<*aCS*9=@SjO-SdRmwJj9*A8VCxDgZ4sL-MTz}8inJ7(rrDxp
zMQSh6x+?YQ<d90Uz4)AII^eS=E{XGQKv_%lp6>D$IY;K<Oip{!JfK7>f1W?!^G6*9
zUVia8m92QX|5S9bN$~Nzw}cqD`1~tqwp6mgouB9~FR;dPJnUc^Y3mAQ4a<1ucsjB8
z%y++iM^2G#Wm&Zq+p8^ED%!p#<V;l=8Y>CXaC$hB-3%f!98ajKMF?2+Evv33O(VxJ
zF~@=`jXAU0UQj8B&QV!I%8BFCj@9J_k^Fn}VwguRwdU2$6|*$#-hE5kuW4>?5#xcj
zYEc(u9PMMk*oGxcxar7t-D9-H)GKr@bu*C?*&5U{s&oIar>zW|O^**{pZ~j)v$i$P
zJKlZTQCY*)rpH)FA(C`LS2e*E*hJfR&-AMWcU(9h4;Z7Us*3r1WF7{Zb<Z-MsOlbN
zEtOK(^`<0KV?rd4R0?UU;vZNBv@%4S@G+26K$=|K3_386C%m6aiaV7troDMX)2>S5
z)wD=cBWzWaob!>UZtxe^#AP8{!R~aTb`f3GR7R7%N2-F|G^QrUK*$M+gp}~h#QAjO
zw0kN(f?>qg4Q6{qyV@{~XK)j?s<1{Exs0*}FhT+eeqMlBlE6UKc0?BlB4fIiB}Uw`
za5xsi<fX_MTjA%qWK@nvntp{yk@Mk5%9*BLl{a|X;F3qgf)$P_q3fFZ*;XcVWEuvd
zRG9UKm;!`AmXfNj(4qvzqJo7)&WR8_*4iSN5eX@v(~6KXp3KCb_F<suI{JQvBr@&4
zV_0T{SlI7B^78s?Y@>mafH$?JsVc5-ugE#^@Dw@vGd`3F&h}zO(^drMI314!=lSsc
z2ds)*UaYx(aa%YaO%D=ciWp@fXBy?HWgwpp?7#UnKYaU%kM|R+RRdJ;bfoS2@&@ON
zuGZ8wRo@`BK^aZ*5jQ!EF!*Wa@qUkKT9gh5CFr)-$gX34dLo1~%hL(7x#sGNFHrqU
z;@vOVfBb}x@Mk}{=Iej-HUInn<A1|{^S8e$$Xrn%K2i#d);v5u(bP5P(?Gv2P_R|s
zla_#%8l_G=?F@4)%_0{g(kMjA&s=-KGSA3o<LU8mrcto%Te{Yk19kRfqq)4eAcnx<
za3E)J5%zOrnlmd^0&KRbsq307D41t!5Cp2Krrq{bw!)3YTB?kp6L9r%4Kb6Sg*wfu
zLw8#)UtaP0>(97*`;m7)d?bd_LXfgZr_RF@has>V7Gg?Ft}Nty2z0fj)rwWExvUL+
zrKydksVzB$(yvd6^}1&|9oape5K=M@CFr>A8#Y&$Y_{9Nt0~|Bgi(mfQfW!9S{7dv
z!OlB~5!-asre&T7OueS9)`(QHFS2Sex+S;?sSHYhkp@J9xS*9lWJOA4g{X9kl>Ybh
zO=WC(fi$uF{IsG{Jvj%KX<~K@(>USgBi61EvP@I8Z3!;19~a(!IB~NHyt(Po#*)K-
z1RU`<fBUz*aS_{8=(=Wexk5iZ@!fCVqF;Q)7yrr6S)b1Azx_Mh`N+lV*Qlmv_x9I(
z`tB`Nt$F#`XBZEUPaj!rH*8<tQfa|OzXCGLd0<=?t}n0XHXZ)yo~QTk8Ar$J_J(G&
z#%?dD)>owQ%<=7ay#Jej$Mxj}mzOKfQ_K13h!+vnXhf7~TOQKK^MFd37ppaB#p!fr
ze}9jQ8DvGXc&2eAxrK2$Q_UN!(ghd#^h6Ghu5QpNGcFTa6h!m+>5lO5MB8@Qx?wpV
zsZ7oG%^OI9?|%7PLR`=ansrOvZg5L+SsPn1&I`%U#2heoi(4FSURaieI!3gF7!uQP
zE;PMdz9e100+p$|ma1+Ei!V2;m<VpbFK4tak`^H~N<OPEpN(h2Gz1?|M$uSB@Sfx0
zK$4cKZ7@lpWQJ7m$H(IlU3XN?3ZX4-E>^dcp2%({PiK@U;kwFrWK*-=tUyR|Dk{=x
znqgVUVIe6^-LFupD45f8sH|#wgo#BjUH7z`tAah9Cqkvkej-d0B4ns5mc>)onkEm#
z<yc1ll_Dfnh-KrkxD}*9$AFe4OL01#7<UI|@9`<oueS{6k@-ARS%n)%=KX=ErvtMK
zRBeT=G@I3`;9sB7txeamHoDY{GQ?%Z&u3B+c())W*t*76hB%*5N@8t=QUa_&(cqT_
zU)87#*)PZtk;tN~wHCAdVc?K|sGsxq{|4V}G3|!RWa2WQOvU!%HBaY9Zrh)rbn#O}
zx3Itaf#6)BQsyI-P}sI)UL=9TekvqTt%>uD640+Y(lRrho>0nQv}TwW+HQq+XSQ)q
zGki~WN6sI1y#M77e7K*vKRS*J>`pVQ`-RUwOSGNF&FL9{1z{;+u^a>QFp;xh3Qrsk
z6UsFBJh8+?R`An*+;O#9W7lgKM~=gZ|J(oLulW3D_x!7W`7cPaB0A6TaN-yxU;f2w
z{&)ZDf5qedk&lNb!ZSrsh!T=cQE`9&P@MKLa@d{N^gU7~>RNJpy(-4TcaKa<@f`$T
ztf(o*=Wub2Rg!5KxC=91yu7AsD!QgLIJ|c#ZCI9t!|BN4Jjxh!WofL%A*gJP+^krR
zM@WgfZRjt0gv?0FeWFl~GWB{@M)*}-cnQOi>>X8IBQV9dIG;#HQ`I$}|MUwkUS4tc
z{*I@IC%p5tT~EKtJd9_K(}^TCE{k&bmV!zOY9pzXWYtxr-R3?0bHLFWQzk<pB7|a^
z7ea_=ZP{L}=&Qn?2(hg8blV`SqFGllV6-K-Yr=5AFCMZ(!0&-sQkJHgP#P%}hCqx9
zF}gAls8`73%XeL?Vi*iiH!UG7pdv~&<k6AmVs%uq0PBz%@>C`~q9)~p)*7vAtf{%b
z|3ux@Y&O><A<$LNJP$a|TwK8Q%?0gtg{@n}VgCnRqkq(4;P3zazsJ-y&E|ss&6kAW
ziK8c1dvuEAyAOnid!|qCSgm`+e4x#aqmTUd*WdHi&%Q)xgK1ZM^^b4xIbozA&I8zn
z^Zgy;;}aL3zoBkA#)l{3=|s2gxc=fR)~{ZZrDA{ok(?vlc7qXLaUKWmp6<}bvAKOg
zZEBKjQC&|O4!HSDSR9+HEu<1a$!;LJnWn9|yn035uZZIwZ4@z-QStKhk=!)s)s`3n
z!g~(;1K;isT;0Cl^3CVuwnI}0R#Yv!!vSeJRu>yq>owE)%+-r4hU0-LMAr2RAxp1X
zOCU#tmb9%c=;au3&(e^(YY=4I;#ii2;e4iBZ)vs{&u_6)c{pvifMn!cek}z)qO?UQ
zO{CE5e3k@?tR=e%Cw<}fM~&5zWnOT8L5-(UiOHE1ikDy-NBle?tznr4@_6RFyT`j3
z?_FWbXXyGhO~0mY+tPRT8DlkB2o~oE!s3PnraiXp3!Ft12!~JxU6<Z{uN2-nhG`~w
zhYB8@9BE9*5Q)hnY@ts+e*D1I#RW|-$x`5*D~F<#iOa&_{u9wTtgTB5R|+DMVYg>I
zoT#>2n$0zrU;LP6@*M9!;7?DScK77qL2Fj472T>skjO%iY)!lEFwVg;gD7^f>^#df
zqNM<(FpWhjMa~JGJW_&p&q1*eh_OsBv?{O>|2qp^a)3Zti%jY{oA4+6k%c;#!;DCo
zv}7^@ZF@RZBUBl=1H@&bl8Mm|BqzDOs>q_@-+yz*s%uc%prodjk)|<7k*Sp6{Pc;L
zM5SKQZZ9}Iek7&Lw$`lFLOVS$ef);$>51LLNc4fq)U4{v;1W+`B0eqbmdKlzTl(6f
zNJtf##xp`>rg32&9kUC#EXlee$AC+4oE%dKi0XS*R~HDQaV~OrI`Q`76Nh===dZ45
zJA>U^V;8~vaYxLGpZ)kX|HXgy6aG*C&%fck6j>5EBSqoW&$DA@qOyu<Y-p<nEhH(y
z?d=sxYld;;FiIZwXSB5-655m(0U<Im1dNf?)=*cL?Ybp;hjX5B7--s#<Nk@11>5x+
z7t3T)#mJa4wrUXVNOl$pXe*JjAdiaxs3rOMoPKYL*+d#DQ=^0;jOXGj7$#y!NCl?S
z<f_z4`zo{k`WFBBEyvTy<Ku}XcpeWURbBBkCgxBk1C`dSn`aT3mgwgRn^N-kSt5~B
zX<Ho6Xd&pjrXZ}VhRd4^jLIw_P)CR^5EqA5HBu#n^rUI`ynfXzDVK?>sSw&QobIr;
zA^J?++H%0bkkbs=VN6XD0-<v8U}S+-25|UgM43_<jHf4(`yT>+%QTR~f>ISJ7oEGU
zG#W{g32ADQSKQv-5@L}mDN`YGz{tdARk7V}Xsa5bG+HXse+*I4*-iZQU;Q=z^q>6&
zm#=PkxVz`+VNbhW^TW5l=I&R2hY=As45WC2Zdc$(8XW7^;O8^b>4<E$tS@hv#smKT
z1A3VO$>com-o9t`;uYd%#e8}qD^1^agk@%ayki{BMH*Ar%ykRO&|hph@18h4?dex*
zY*piQL)~vsm1NjOPTrG(qi$NfcNiszF;H0zekRS5c^r^Jb3B}Q_x*cbUR}{#Zjk|Y
zPbX4v9FH@H<4m<)@OPhh^P`_I`^<PgVOAPZ_w0u=yKjEr@%=|60j&i$uU=uAj`ik(
zasP>&B2sI1_xFWyZYzTG>^|L*Vq|-DffkU1`+ZohBv=zNXJU*5H?j<l`FMa_5YQn+
zta)Z#3s=&rMl(MrbgOn#>M;VQ<BqsY2rW71atNurby2*c2+@@TpL2}IBR&VL)Wnp?
z!QtIf02`t>zO@pV+Lkx{Oab+Xj1-wDB}p2bY)D;&B4V1Bj37NXD6}!il<{sM&I4hd
zNQ*C~z?863mzR%vg<W^Zm@sMK=Rf|8&1M5cLVmWtEfX=gBDFz)Z8!}B>A8Px+78p!
zG%?YxSKQp*uwD1`y=2u^>^^9+Ur-`bwO#4_D#fzQNTEP!gzd3a26Zm~|5h_kGnK8$
zs-&xx{+)3!BZ324m288lm_sDzf{I<H2@^faSn8%B#YpguJV&zgg!xbWFyKa<J3}tg
znVb}rRM@)UfQ88XUb3T0@Uv-anz|$78HSO6{$)+u)(neBHwvhUNfLa-<b@#~dGY3R
z+J4P&+~KD)ZQG(<Lt7`}bVNj0&Yl<&XoIzhi|T^bN78BJX*clw<HC7<<YLuwyJ?9&
z(6pr%ynh-wP7`Cu_$-lCMt7Fe5;!l8!!V(0!^?*~FF$*O&=wzJ$;%9a$4_@uKmHQE
zeSthX)|-UViX0aHv%ma;Uwr$4|L*U<!+VcZATcc77lDlugb+wEvD+W0tl|1%%XYiv
zd^&S=(V?ti_MVs_it<pO<`Pm?5+n&FGF@9UFAMvphtlFuMGA)y<e14JB7x*R%i=g?
zSk{uds*r6>ih-)D(N&GirMj4fLa!Qf3}mHCHYOHEO42o|(}d|nB9f<x6a$OQNMnkO
zF2_PYl#0d}zIfI0`RkwX^RM?@|GV$_*S~n;`{P(1vIv6rTwJ!)ss!|6Oh{ec*2iha
zKYJ~l^SrpZ<i+g;q=1_}A}2Q2*W?sf#)05#mh;3iFLaws)rdkqJ>SuEx5Rc^M*kAD
zGE5{~@>o;jyoVexDw2gPsQ54=lto&LwJkE2;Nm<TK(DCU4L&-Q)#%j<-Icw*dS(e^
zk<q5Z7+;`%A>rIYt;-=VCx=xAB{Nzk;&LcI6WgGwl9dwLkfSeCn?K<5$4SzKWPd#I
z{`>Ec(Q%$frsd4*FMq^5%{+bc2_qw`uBPoRBL1%5Em*xQzU;TZ`i83)FR{VXZB~R$
z&*A-n`RS2mSddEd!+-b%T4t=#Oj`1Icp@#H@$ks9IFb~sFE2=063=H^tGHNoRAg+9
z<nhdR@4n^ppZ$ccQiW%*T4A!IuMA#D+O9*15_*ZzfiG`%r{jrgbA=0nA!UTAxc|)$
zoZjElwwBZRL{(e%ryY$O+4K#>Or;dT&twPMIgY0@_ivwg`SOaUZ)v(MFMjq<F-jAb
zk(foXfEh*IHq?zRls#owyhAHNn9c=Ti7ug}m^dGgMdND>DMsemv!Bk`s-oGhF>MV}
zVoceomT=pQkfp;PmI7R|mBxz57#9W?K^v5`Bm}`ZQYdfxC`n8})zS4WQowY4Bsxd*
z{&^~xv9+OZTNHvgO{luYNP(Xwf`9}hS7?e=+ftb;>dGK;DwW8(Ci$hLc7-5jfs_)f
zB|c?hibd>_5^cR^)vuYK-jk+1=3-4Y8X;<=)CBL4(jd%IcmgRCLM$igEXZ1+`<}Y%
zu$#GPZ_hi%uwa`G;R1~n$k(4?wIVD7Za4#(x~j?^Untt`HR72mphaex7KD;is})i!
z(&9^qiTu1)a4hE&*47Bb?{%tfiAY_ru{j0iX)FXeAclpQ1P=2@U8?`kJ}G9DDp`r9
z=?TV@b0)>ol1cx6ccLsSwl{B>{fHJFhsXDbT4LujeZOLJd0DEkO0mqIs%lAYq;Fqg
zx{9W+nWsG=IFzthlSzIij04N$a8VGWWO0$CB(iNuk7w4amhF1Qew<M{bBY3cnux(L
zP76LHj>E#qEj&yMK~(gz!W~CG9!FHwvkx6=?)XWxtQLVZHL7X2+^l$WyP?0hqTRm1
z&ja51qHkYj^18ta%|H9`OaAKDZ%dO%K+GADGX*0ip7-gB<LOM(G<^2DVOh!)Y1Ot&
zb70f9oTr)1^(E`JEl?YZLGW_B;$qz~jWg%Np1RUpZnxBJT}YbV(Y8GyxT24C9<2>k
zU7?I2jVF*`Oog@;%39JiA!M0GrWg@9lViEwq$z<~B@_ftatkrTaxN<uDFlhB*luO1
zKd1EE5HK1P+?b00>R-I(FTYsv|M}H{fBW4X?+*h?8k84EEeSc1ePW&*-i1=B)|%RC
z+O91PlzFCGwOnj3IP6a(rD-=kvz83QRM;YbpFP@~sM?zDVj&JkYFjM4J`3h$K}v~C
zf)qV%)8U4N#m&@BPgu_7?VlN;B=dX(GTIob=8~u_F&0=+y?TYLTGTUvP*n}8>5;a=
zogbKoCvpx{^_IF`kwZjeu(}vjV~lurB4>}%H91DSTbSk%+il4)Fr5A)v<X_PxV^o^
z=fL0p@|W~|hmW57_wQ-yhVJ@;JQe>kVc~rDk;nTz{YB5!>sNHAk#YBgjw6Rp-?Hg{
z%5WO^_U$K<mSmxL`PCcZbRwM|5RIj+EQfK#EsneQPwdWTHWyoBzT@h0!^OI#sx;J=
z#e0+h)i+#Rtmzs<-8AT~pgi5~BO*tdTC=#Kh72juSDL2Q>|D{CUcY?B_3aDrj=5W5
zX3OVqeoP91>i7xHJ2Hvw#RYBGA&f%V9<#Y(Je~ON{ad`BxwyIJ`qe9pf=ZQ9R-8}N
z{UzPyOR`^B&QB<#8OM=`V0CdppAzH#2|tfWAqdHlWQ8;}Qe?CecsH~7$g+5*c}7`_
z$`S?S81T7-tst|^a|!J<5~9Z~CnQy|cU3hZ<wAc8rEfn6S88#hbcpM^rCDE~tm5$L
z9n)|w*4`Lt`VK86LRaYZ1uP2;C;aJ*o5tb~P!g$&s#M8}s;R-xh4ckv@5njf=c!0a
zRC)8xQIq2Q+?Nh`zpzXb)9J)|e}}RXQ`Ow>j?_V;O%2gKGua~|2LzICwPKu#4%W0i
zJ|>nZaYKL*nTLTek4#pTzmJ%yH`m;J_7jY)*uDRj;drEyg7GwBw4qwBiAf>c$TAE>
zqcFB6rHGq~VRIQv($^S6%9(kdn5PMET?usQVk}#p>u*9D;xjtc*b1qNAO=$*(x3KW
zKxs*iGt)TH>0;8;#&X>4*=~v!*1584-R~cG_rq_f`xSMg@F{RS9|^;V+AP#Iv5XV#
zdP8L*%Y4RaK~-6TpBZ-_SaluqVMh!OkrtNIj^%VFsTwCMre$Cb9<y1K+6Iv;vRmjc
zH=MtJ%k`@l%ov{De$R0{mP2lqIHBmeOVa$ragOY!2kOlQRlngM|MS1#?|$<MyXtwj
zUvQk!Iq=($XG~>@N^o?6YpF{Gtm~P>fxCwXa;4clJ#o3Jshb8r4EW&5&-K7!+B4K`
z$KiD5xIbcFl<&wiPc%(~Rs$ECHCafcR=m2o;`U+#0_MeGj4sIEWoBL$+PdLlyJB%O
zD2ex;wr%js%;Ey7syPm&w(J_k;xerOX$-cikU|pXv9S0}U4T-VvIfc#-E^Q#;XwpP
zm=}c9H2oHdW_;R}Nk=N<{G7|1u~Ld4%Y@HpMO`h3l)2RKzxa92fARAT?~Z~0?zcPs
z?)?tw3W8SyMrsxxuu9N2E!&F~b*1sk!sGoT8c9?2Y_BeHE}&dMXoEYRnWr6n)3UIj
zZTU4)B$P-fCD5u!)oNXFmX^@0k;>x769R=&8ge9O2O^Tw!gL<d+K^;}?O)==k*ZeY
zFridU&<P{p88=063w}A0ogfT*M2bj}@zV*UA*Y02PNn6bi_OyK$UL9XDl(m(s8-iV
zQC=jJk|g(sNzVSL!@xiK`H#R@qO|<*o9~E=XB<vU$1_da^Rs{c=lIi^W%r5T6Wz@X
zFRov(vH}+qDkXI4NLetQ27ddCU$8q4JRMH_{Ga@c@pPauU`)&9MT@CB9uH^6)5z)R
z$h;(sX~-#ZzJJ1oh3ozeZkX^<ve|BVx$O{5L&}m>qq)4fBw(3cAT0|ydoJI+LL5%)
z@Ad@e5l&zMqcu05y&^6r<~gtoBYKGZ{LlZAt1rGr3dQ;H6T4skHK(T^nCF?}(>)20
z+RzRIN-Db6(AWzuuC8dhaxy=k_MHFf|BXzrzP>=q4YdTLGrBTN`vu%gnkNM1i??LK
zI8I1Yk*z|hnrYaP10V$<mW)9Rp5b&v<%Dgj(r6&l_C2O*S*_R4BtwtT<p3me#?~F$
zWWsaED8=&crEPGhJz<%t`z`hA5~RkDM}$-;ZJ5p{R;!-6vMj^G`udjD?Q6s`akzVn
zNQv<@6tGFxQAscsWSkJALrE~EE?VjkNPZ&unS?~7;$=6+AS#RZuHZlyk2_5T5VTse
zzWj>3zC@iKNp9lt(>tc~iMC&nV(CCDA*hVN+S2*<Br+236QoE8nH&X1HDs-seZ(y@
zxpMTuVY{B~?Mvc(#*ZgzX$qqC;sttp1IK&jnDG*H)f6tkG$KT1@e4Xe>gT{y&Si9m
z)+8ay-jqZV0%<FJj6~<26ZIaY4A=&r1;g^E4ISD&+@WPC-RR(nIWjNhA#xlGYL--P
ziFMnds+JTbAvl_TgVct&eL>x{EVD;jjSC4812HGYX{M@5h09GRP`UuCl;F63M4X;*
zb0E}f#@@1wg08=&zJ0+!=IWJT|LFm<-EeIiF28)m?qNh`IQ;M}-F8jAxuU5W9)5Vo
ztQz*qBme3@|Ch`$BdQfl1OMd5*JL!yJoDGT_?EAK{3TCLu>S0GcKair4h!w$9T#1X
zL-FnVN22aH%#q#4PdqM~aV~F^zl({4m{|2Iv=j{E%s31vWy;@$TWG8#i%8uz$SShr
z#Les1ym)a#Um2R(FwP6nd&XhJi;9cBrfF(?3@BadhE3ZLL*#tgp|!=wa?e02j+0|a
zExoqnC|P`<Qi2p5Qfjnq3wA4)upXl9yU7bmR|r)mw9#j>Nyw%_H6v*(S%%*&iBisl
zWhspf?=aTDw!GcvdB(Yjt^_~%yyEMh{*3?d-HCtwH^1fWy9sL*Ri{`rEtM_?Tp<c8
zz*<A!UZbjtEXz7mEP?amk*3y|z6bAFyvMc;rfbl3MOIBQ^}4dckO-7oSW82S3D;e+
z2#u-xa%Yi=BtY7T6qXc=WKY)@NKvepN>nU)KqyV$-xS*B?|#@4V)Ud0s_LH)Se_UH
zF$SDl@Z*uHZAo~h`Gm*{ty_-!BYr-UWd+heOw?6F%70iB>5n=LeDTvCb2^Re9(IV7
zkQ$n{VgIzJk{Yo%Qq0uNitEo_q1(DptUf%Ts~TN3jJqB0Ki={2{gE%fx<X3H)kTYT
z6C0~(jo@;##c08YAHL@}E_B-q=A;-u++&Slm;*vfl+nzK<8(MPE(>3L@rq5~U~R?b
z^IPUDI2_LG?mp3OE^yvcn=;B$N}+9wY79+XVY>#AGKahSBHC$tOkK16{3~wXe1%aJ
zQcCLEm&iEa<~{cxKQd{JF^XxNSnUsNuhv}MY&ji{r84JSfkDJb@G~+*y4gal`S9T*
z=es+a=hJ*sH;7n38`D&F-C1eeJQL@DX%<58kR?@9k<)?_g8t&-8DLa)`~ui)H_xkr
zg)BV6Sfo(Q^FUR#C@B!J>;+W{q-DfvL&D;Q6T@zgX&U;g8`|wlKrtUJQb|O<C;N%$
zCJw_9VOm7=i2Wz5lB_Q;5!H%j7#Z&0k)xy1k;}eeo+g&ykh<&^Nr4<ASqh}7i1UDV
zbD;<dNzMWFEQ(u1LN*4~Zn0NiA~!FPbq_u;9UjqAVCn`)B|NDuF?z;np)(e3E9TS4
zcp6EzLiHCU(~@#UG!<fVgCOAV-XfL-H=L-A1~(TJab00eL#VdM%`H+XX734rN}1?N
zCMD`Z7Q|P`oJcX@Ge~Rc+7(Laasbn|7z&*uq)d!~C@XYy<PgZpkW@tw2LC4>Ak6bj
z`|5@qO7B|en&EsR=gfM0Q>G}o+>UbeXkB5n#+?V+&6-@dco9Jv7Jow54dZxZRo9R+
zy6P~t;q-Jz)3hk9xPJYHd3VP=E*RaD)|%?-CE{A4uWqrGMaaym>1gVX;j0r$D@;|P
zROZdsg5lvark{V!r%#W({LxQ%@#R;1{LMGK|L~4C!<j$-AN>m+9v^u1`ZGSgdrQ?=
zR9&$@4{SFpsyg$NKmSL(dG(tA*Z=)ra~KyszI)5D>G;jB-tiy){w<$<@gwY7^6g&n
zAO8Lua*llV;tg-_KG4(^r}IEU&~+W3?suf%=zD{arQdFJ#qn^W?>fd1`Qql5tE(HD
zTGLvC$dR~vBwNe&az*x$d73cR5@W(SkMkZuW}4?hjE#xoc~5_JO^A{Er!)IY#m$8!
zrhrTd?;~MZFttWM(-T!ym;HE%WLa><sABX>s)UlI)~E!gZVBEa(I^CNoXch7Jz*-m
z{>oYii5MfgZ;-ZBM3t)1O7SoL{0%QZf64#(|MKs6+7G3DqZCcwkW(TiMQtpm*3`P9
zYHOq|@-f->Y%Z^hdmt2G6wxEKqF%2tMv*BY$~=uME*AcSkdPe0?=ka$?yig5#5SNR
zq=_u^vDCKP3nT(971Q{LRg#cB*7hjdu#9`A(>-o}1j?!<ONCMfsWi4(F`e#FOhlp4
z@>$9*Ep@eK93QEw3z~XEc4akE)iurqq8mwCW17_;bdCOT3S$RNv*H&YeubMoZ(d$w
z+8RI3n3y<xc#kxOepRz+Dk`h_w}1N!{`#+e$<^&G7u%L)T=?$ek-DldMwCuy)XaXN
zYC3e)GK~w5$0O%CviQhFQ*-sg76MjV{-Dd_6Fx<z2)SyAhbOA&3IR-;Qg@M>_Ye1c
zc=w+5><~ogYEHX7ZeEZ|5MrS1F3SGed-_+e5DyPb=aJ(uarL9J$IU4aQ$`C|t+%+B
zUvQphw8%&j)hg50E!ui6wk=&_8Hb4w9K-2c;2&f}id^bT6chw<w<4MrV>DIQ($qD_
z!;vfvrtT1&5mI6+OHvBDH2}7<$i64XjBXo5hSTl=sT4*^j56gy5{mH5<0lJ*(yX@E
z<ltG(dk706mEB$0ZWna)EG@Rb22~T6=~>x2u<RcYe#U5nt{PPE1n;?f_bn#{>#NuF
zU;T`#-Js2hbbR3S@IZ=?YK1iw(rU8Mq!`gsf)<1XN)}KMqR{+8Dm6$DG~EiJEGBF8
z<>%;jMIO(L$0x$LCoTg<P{?xv>b8UvqnC)L!)$Lz(NhP7&rju`AvKFj_>gH#i*C0h
zH<G0xMn`Zb&JP2p-BWo;Nlg-&!{Lt8`zKt?v@&3YC<xePWS&BKn3$T{Hl*kX%S?K1
z4Il(Dc%t`I&zR%iX?7RglDnRG+F|nmy1^$w)C!@hKju>X2lw#^s%C=-f$6lPSzTg`
zVVp<ic|Z}$L<SfR2mIog#+lYy9v>d5bg9f%)s{+2n!2SjHDNw;czWXc=9YQ(n5sf5
zL0m?ZkmY-m1-fdeHE3P4u`N$+4Jn|PBUp_xg3HftAW5vQ*gt)u?^m>q#$B)Z^3VSa
zUH3#~`02m=XM8#RV^RnRnYjM^4Jibq^aOWe7<RnAS&?onJ`1|_6{pjgAAR+bFgf;5
z15YHK1y_{c+l7A$DBUd`(me?1P(uwR5<?0|cXvoRbPiq83@~&E2nY(&B@7`YDJk9k
zKL7Q80$8lWbI!f@y|3L}&nPljMZAxz*@7=sT$x^pCn$W}JB$}}(U^Z+xg6NjwzrMS
zjx3{RlcS7j76ZY~)vg;ehYDc*fA{%}wTmq`1AnhCsyUI*+h*6W@&b<8Jb7Ir<8wH1
z?idEJYV|*Sv(9wVMBv>OVt$5~#BsqR@7Wr-2y>h_$e6)g0VVVF<n+Om^ZdP2I@Fl?
zm#-MPcnYbGI`faQvcf9*K)?>E)EA|X7XFIgFS0aE***t;4Em2Ga3wZ}0B~$aA5#Hm
z_Fk{!{(+1cZ>-Vg+kwy?#e!aoC)Gk2*DI{sxQV6(ypm>H=9qL}aMD&ic11ePAyzz0
ztp_0(56a;?75qVr`;Tvnh3ifJz$uBxdG<ResIA&JE9Iwr>}FZi_gNcqt0|dZ2e!1V
z>GB4G?u@<D$wg>BRp^bh6J5E<R2G@k+&(+h2q|!Oas4sTR9lggNBtJcI;gz<KeE?t
z$XO_I=|OgOo6(E<>b$m70761V{AZ#sOP9-}^EZo4f8d|v-PrNp&a}FfljnDL=MH~>
z4}O}m8juf=+c;@hcSmcAQna6j=}_c&E2Jl*H)se07Qt@Un*_r>7#h-n=G2Owu==l^
zeF5XSS~M;=3O@DWt07}q|LBTPBtjw#KDJ`VYZseY?D;aVtKLR@n{>M$TO;kJpOcX0
zx6?LNVpLPmKLX3eqP%5FZx4f_C;rOMi-~eQ-|S)dYe5o|1S}|{y?E|cS>33WwN}!z
zP6b(;_>1e_?}C6I{vku^oy)7mpmPe#A3MJCQWhTC{BY3)FL5sHo@ivvQOGY^qiK+n
z>9)Vrit@<1JWR}znW;xy0>VE9vf|lvW8$#Ym6V-{LMgHjtS2O?ga}#=g44-tr>h5<
z>mIKttJz_&&>RvLmU!fWf7_^N5Ut<cSpwj(qQ&d-Nu}F@!sQWLMQ_sP>o&fV7c_*2
zWopB1in<00@+5bDjQ(A*+|!bS;Z_gjcq=mN#J+dIVLrMV70|!%`Onj-CzUK_#J(|{
z7ZKw0MhB)f9yB;G^sDN%R@lAkOKXJ|`y|X0a8mjc1fBXqap~W{->1Hp-*3!6HAL7^
z116Bt!CQtxeCeCJ^!?_3>Udm2F}^7kZBaf|Xh|*!Z4!>vkop9HO_R<5In$@+2s(CQ
z)!ZCr#nhRf#K#k?>WRJ)$-p{w$%N(BAeUgC5@A0A)%EjhIOvc#h^4gw7cois<qVtx
zFEFmgUXMx*YqLwLx1k*aRT8K=``*v#@ea$NeX2&Ek%-`!KDBQUy#Dm!NGj!ytjLh0
z*v1X-^V0?WJO=5a3ltPl<1%1oE*@z<-AHRe@ke9C`Q;Gs9uj=hz`LPpUSw2Doe*lm
z9f8JiqTk!cgCZ(n;}_#WLoLg9C_%jBRSxOL<nc~_bz?Dku@9Cjn{8XB_jAER536Lg
ze-_<8di7=qysl}X3Vgf|0B(^)jez7|XYp(Hutr}cqgok=mS`C8cyrFid;5kr9AH<_
zH~2i^qL_%ZffRo55qB5Q9pLf+)=#X*&xi9&A5Uhd)UiHyD?w33yJEaSY*?mGgC$mT
zdvxi^W2A5VZc0mvEFXtu#c^=75Jmm!tae<4NrJbN)&n$JNV}LZ$&xTTMck75%+Yk~
zFN)a5bTVpDsI}>tqpq9U2O9t<db51obDKzQq**Jbv&S$2@@#1dzi5dArCH^1AWumw
z96AcVhw8p);L;uKd59vf!~SvQn<$k1utwob3q6GcVBl(EFDzyzqQsx43>Ek_K-UHz
z(pZR2>VsOOQ;q9Rr<%P5FOM8}>pmgZI+Zu_eg%A2Gl(E1K9VW0OiK2F8YAxFG?V?n
zOxP<#{}`p-%>orDtxSUrt+Zp0LgamyL>f)NAFq9_Bkt!ays97#fo!pLnEMvoKhr6@
z^>fwL+(-PlSFd(1YT#aSBvPa$gQsBtysx-B8ga8mn#E$rK&=gNsX=^RKWnuSA6@+D
zUvTiI^FiqmesdC0W_m~O=H}SK(|7uF{UI*kwZ}AU%2=b6yqHLh3jpwa+4mLCBe4!|
zN<UqFvLk8crjtYUlgj9<R}{q8<K9P~C@3&M?t61&bkOz4zJB}@90x4+)|hlM0mIwD
z_w86ye{YeKO?H^^gtMOfVb*h{1#VARVTt)t%UKS^7CM@47}Mf)o($JKQjxRQRcduH
z?lhD)hKolvoQTG180S3{R=|M{$>~epj(=jIn(+qS@SJz>$<S|p#W1`~q=G~2FRLx~
zqJG-p4)G$qD?i9I?q`<7<&R2E6y8fziS4H#-4tzP?dx3>=7k!JHAC%0FfU>8@y5lK
zO7HI!9r&xgvLUHjq)bXWf}9Qlt)m!9J@9R&tmbZ@kM)Oz6cxbOOK&z7qp!{qa)tdC
z5&e>F`%qM|y<Ta<T(v9LryN|p#c}x8_rx<$wOMk;uTV>ES$^#I`!*2Vaufs*-V%t1
z<p*w~Qu6F@s9E)=!-%OS-I@9vc57tyDEesQ{H5+n;RAo4*=EONiWs6!`Y4|2@HOnc
zYwp!x*bg*gn{qNJEswc5!8-5d*_3wE)bu%iDuDk5%h=9ct*Kf87NcU2?(uVm;_0^7
z0!z1nmu)Gritv?I*63u2u<e(m^%47T-2oct+GE)jVB0;{wpkAw9Q2w*8^R?A6BEbj
zdB8!~bQlffI3Ur{<ly@@v)A-jM;9_@3Hul6b65@U{*7~ly~}K~xjd3odDbUU-_|P6
z^gSTEnb~q2$kj88qCG$4y_`=o7e5@BJ6gL)UUii(PwpEp!ekrFe!N=~FmT}?_)*_J
z1*94YluC}EpRfMyjbF1Un&<kR{?*8Q%OU^XwvDRQ??awq;Mhp#zeS3yN~iIK7L9*f
zWDCPQi^7c~3|->sy@fDMVmzEJGs`?<1A3Lg^(9FNl<+hzHQk5t;3~G()WY%8gK?rs
zdSp;Fh}o)v6ss>~(xbcvZX-*sRbBvJ*$fxwi;L8rE8W`a)8zjaI6YI6N(}k#Q9Y|e
zfZRz^k3}uGk7MJ#W{1~<s3;zXGalKyuFs8vj(xg*gv%q3)E9;2MtWgFxhj;UC}`9g
zGjUiQW_r)}K=I@}qMA>p)Cc+HVv=Iby-@Y0r%O6X>~lI%ip)(Vx(dPXIU?0qQ)X>p
zLN{uYtZvHq8qX;iZEq>lcxSvrWKokZp%Tf(7UEK*5YCji)Io!l#jruzRBI{8La)RZ
zKu#(fsS8?)7&3g@6GDVQRcH2vqOZ^Xe@W^?z(oCl1F{D(x{n1+2O5{G>Q=3Ru>ani
z3d#)JZDuSV9s9~u^S=`?Ila2zOS!z;Zp`>S?+T9BClUHnIHttFcrd|yay=d#2k?)i
z0>=*?5WO}!->mPi^*XNROg{+JMDed4bjf9nxEn?l{`=W=>pQ-5JhJb{%!FMONTcBG
znZuL4AOH<AW8eVf6KvJXO%YV=3u0OX5YYSyOuNQByTeo+0gus<?a%+aTN~zTFcL~g
zFn!z%Xh`Y^^ousNMj<!PH`xtrDScjQwq1N&Mqf&%eQ~UH+0nH*kiH8srKFT*N;3-z
z!k{Ftu6Tud*W?&ZdGd9g8-ZlT?d=$8^<)0a@@m2<5kz1apZV5TngYMpHuL=l1G=E>
zXrV&gmH~X{tmqyZc3cY8p%6?!&PO76$;#$5aWu=LexR#f-wX+VCep8r5+diEIR4(i
z8%5`Lbrsqg=LiJ4lBr8$Ya|ri-4!@fnQj|i4;qka%Y#j^XE?P6#$l|=*zEe79TtC<
zQ!Xyg7F12Xb^dhXRi=K=ZT3<&^csE~C;cxfzHdI;-a{6ONMhA>NQQzG5l6gec|{g2
zlH?N1SUg+BxMUsm(I)vBF|o7SS7$kIBG8D`SgltqnOibO6mv<_$h*MBFeB@xX0~Ek
zmB2VQnS~fO0!w9FMCQ|T`yOwv!KPsI%PW%hwhyN#JZ+=w>qn97VULADB^86y*?+I{
zsE5w1X%nte3k+Fvn&|>Vm(0f3d!GLZe4s$PZk}DqEK&MEU9GA^UKpeUJkUQL9v@>-
zFDxlHi0kHMxC<*%NuSEj*3y*WSlQpeVjE`hf&Dvtsti3&3%4@K15az<s&^I_$tUua
z{;)@Zqy1~O2NCp9;ce*pH5u9$vN2G%fN(Eba6DeuhXgP1bUt21JH^ovKCj^ya2?J=
zp?;;6qWC0UD2iEB=BaA07jJb)*Uy1CD-6ylKGxbTP-JEUmx#B=C-UfN<d5eHZ)vL8
zhusTy_+_923Fp>5c3nLKn7Sf)2>^V^pzk{<&+U&}{a1N-+*lZF!S%}=xu9;+qV&J=
z${2Pdq?`F88<1#5z~6WF^tQeLFbW2t&NE3*wO$aTGGB)KZf-qK0z&I%dC@<}%$ax#
zF;C-opI7GMJr9dJu>Xe=`0q24pV>y`vJ+pHRf>U`TW5Mgha&hw!{lUlI^yxcrEAp_
zcSLpKn~!m#?#-OAJruI`Sq?K%C_RbUGFlM8UT28^Bcw~TF>#uKlTUC_5q>BAL2J2G
z;DiwU`<_saE&kmv;m~S3uUOS1PtMgtXD{X8<K)OCNfnzLXBcl<i0HY<*C`>j=%F|B
zou&2xsf@#L9*8R<)h3rJvza^OOQykMSWQcb-`c?vea;?_&Us`t^<#NY$im^(?V;jL
z&ELQK5u>A`%$D33D3m7Ggs}p{;>hLWs!^MIzk)HDH{K8S8q-BrH?Ppim<2HxePp=C
za&Y2;^u!Q!LHDoht9<CzY8EBC?{Ts8`px)*_K1JDXpH~2)t&b6=JemN{&3jhpTN<B
z)dZKIQw-_F$7Gb#^=_GTk=?v-*m(AXj~U45YJBTdUOJ)NpV#=kv`A2(YVNbRav_Q0
z(gHgk%w$Xa{fnoC1nW)DnI{aBEea~y&57xh5`GA&iy)19z=$vmHmQv1-iPB*zxOd8
z>qb7g{wdCWW;BcKXtukD%Z%l;y!M%!3%#AyFI<!|t{vj%o!Q(|Lt9c!?Dp)j>NQ1j
z5|$=S2fkVl$???_C!@ukXdtUP9drIFE$9fGlo8q{oSC6VM_AUNmm7Rp5MQqQBzh;1
zB{{1MK=r`q%&JTG{IiklH#kLrs>cOCQJnSufUCnnl-Rlxg^jc1ETqL$Ea1Y;m4<eW
z3%AS^futTrr6$9$VC@n6>}bIgAbvOu%8ea2#%#FU*b)o2Qhx&CD;)<*9_|(4#jXDo
ze><*C9{$_0LeD7EnwESCRjMJWI;As8tHAx>Z6(xPgmT1rz&Tc<L;wACoJ*1<fe5p4
zdoouuI4?uMb!A=zCj6k93CoE|jge8Eqg<%^rIKR2gLvc$?iQYSl@3r%<J3;GoWBG~
zeW&d^4k3E-#8H5MV*X^O=ec2JXk3y)7*P1sZ#^x78$J<M=^Wq0H^EVK%0!Nwgc_8<
zv~Y!^Ioa+naG42IFQ|<R9bw|n=YF*~0G}%}Z%#Ynp)g&Y%`kH2*BaU_Amd1AmYul!
zSCiA$0P|zQRTgEOJa3y$-+FVpar|}pWgoSk=$d^ubl!E{dWYrz(n8Leb)&Rtl5=>D
zX!k|NB#94Ief|0etdmJ{EbPyP-FzNHRrEk!Kd?CD&6c-htcHACzkOwC{i-u!el6v#
zuBuO&3SY><8!HpZ@#>FE%(pcf_CLqv6YfVx_kUZz@c)EqPKu9)@H`z}B~#Q4`Jh74
zX06gHh?dg0N*=$;tZQD(cQhu;6%gP@(Z!^}%2B`H-2Xe~N>;cwFfyvaSehA_XQ}4b
z4xuBlQ^drGW%DDDU`m!p!bh|#lwwB|YOfg||4Oe$UUGC^A3RUE6%kgouZ)(M1QBSZ
zaa-c&`5BbQVF9<7LpMlxlx(<RMb7eLed+ECWP8BoeRNQ}g03LFH`IL4>^IVq*4&bM
z4CJKn#ol%v$H5;pD<ICF#RVN(kAj9q`s8Eh<n07v=Jlj7vy#2<x{Ik(<IEa_B<0`{
zQ{Hup|4DJ{2s~w#VJ06qjnWcbn3Xo0mvjq?DY3kgJNzH3iK!2AFuzZzL%OSi=-@CK
zeG6&d{neK$Hjzqwm_SFwYRJYv!-p~eOmHvl#;?16xm0-!44R}t7ifWF^vL~v!ss+@
zDJFz`{IVBmF-_Z~hTkq3=_3tTqn<U&Dk+f|8Q)6nUy~;HO59Wf$W)@TCt<94L_mvT
zQ5y2lVfQ>@s)l;JtRWT$&n)ZsmcyS_mEa|2hY@jzPT1ZoP&ksXj%Ia}^6;2w_^PiB
ziy+6?>JK;x+`Q=}&;^1a>^4Wk(6@rT3H|Lv*wx!&RJ^gQua$70RFn5*PBv1c-oSW~
z-F95x&dkI`@;Ev?s%9mdEA%^9`=IRK`?0}w_S{v~<&~%p+GR8XY|8dpWB8E)P2>u)
z|LESLCW|Rn`b4(YNi_Ago5P9SP9Y0Et%E}U5}NuNKDrqZ(mEqes;{ezJpuzECD3J2
zoiqjcA5G#N9=2)kfI<6wlQxS&xjUI`XlBaG=+9<Q#tYXPw$gZHjVjiJy?Uu-h4D@V
zlYP~HB&@pD*|A}=&AKVoh3y~JzkFjS2tY0=QRb`ApVcXWHQsc@{%#wPifjIjlyeyM
zY*ESuoBfMwW0CK_HGOo#0fql`i>`W=@t*Vk_OSoj_d^_>3ea?(`nM09NXFt&q8wEN
z)Rdkwtd9KzrF5_fJlQuJ3lDthlvM&i_2|Z0b00EQ&lwkO7zoQj=^r=~_Rg`|Q04P@
zm@#b+#;-7;0Ebs~sRmf-8`1?lhUnp8)%kPeuYJeOK&b}1S!`yK2G0c%lEYX=)<IgJ
zaUv^Lsaqo9RwM4oyh>eXs~OB-tPDez(=k|ii*@=YVs_z0SiTH)9LjZA>*Ci^L>%XT
zd4JU!KZe%$%a4t$Gc0*3lw0#^Im|fS-a3HW;5$6e{CGcb9je?aSgRS=^RlKTfF*<0
zJQ=>W20fuEA<oqZNQLpPXCbxD;vqXa#52sv9qSjH6peUrD>gn@mf$XYhFr}*V%($)
zd7vA3Ia+JSt$v0Fu|%QNSa^!ReqCbfx*zC(v)m;oFH~>BSbYO;0VtR!6k1%rm)C5{
z9Im(<<Cm?;_VaW@(?JeGtSuD!k{o4Z))TQpuMREYC59x+{Tvl(u8>tm#R8!-M6Mie
z^uE8E_KZA>O+&s4C%7H5huQG!?`;9Tp<&S3W*E=|>hb{Wh6pDB1gXAshe=(enq8A(
zPR)N)QBdXzOu^YIF-et>sj%bKyuZg!3Yi>!?5vV#aSr-d8+2_CbYAhw>&Eq{X(fZz
zr~~9@_;RNCa&98ZdRTj$0sNcWm#^~nw}GXc^V3-9<N7NTEvCQ1e1s;<EV+4vXG_fm
zi<NmGdS&Gz$5@?+9Z#mna4G=-iGLT^3-|x8J`#V~ys;Xtla_R#uwoW&0V*Ef=@A%m
zc5nasE*Q_rTWezC{&jzVS@@q1zEUqJ4lMN#kgCT`Ojr9tT{;6k<N^^1g33zK_WP`|
zTgmgwF6+}>ijy0PG@*g*-_rZ~qP4!)*Lt~sg2w?U*8sBg;Ke*_*!$b&QPWn>!ka%G
z9W1uJOUKYlwT-rqrxBIIFG`XF?+#iz9!JHn5A|JW;ZFGqGLls2bbj{vQFgKXkLrzI
ztuG&*A2eBJk6~jqqhcvR&pa1WCZckqoSz|tMjJ`JbT+V3XE>bzL_&+CbM`<GzlA6B
zK#*A;Wn8L`asNNoGITV0=e^=*twJiLBpL)6m(`}5e58yMg`4;PkhGt2IL0C|vYz53
zOE6pdOQrENln?v>rq`QhR$O&Ql}wM{`PhxcHMF}mWi@!TK8?Ew8C}|ivu#1Di&mqf
z8tlkTKVZ2-(>`vhfLzMz-7_vsHI7AJ?&p}Lh`C3GZ&j6(stLnt+B|JFSQXL3@&JA6
z&2YA6OY3o*sr3z#ixjET)F`Ankey30X46ntRa3hi{?q&N%8GZ_mT2Gp)0byiPr3Y4
zrIX&s;nPIQJE)F@Qrjf!WFoB>$T{HngcU$m<#aU!mK@ZmI2S|w6k_1XE|EKiPox2K
z-c-iEm<nVO&{j5=Zhx^TOv#+gXEsa6GB|p*Ga~$GXM}e<8gaUsaDEZvH`%n5oy7h}
zV8nukMq247ju6wAl6}3OD}#_<wE=MakhSAufO~X>S?z)ET(A?+<~8mqwQ8fmj%5&1
zkpt=OAwiNjiOvy}lwV&QK1jNqG*z!nQ%SQp$OZU?R~WhD=H^QgyPJP&;|}4i5GUf8
zxsMpViLe7rIgR_BOsOj@02#8dpKs-(LT8$DH?!JysNLmJ*|l8{5?f_N3Mi!_k>s{R
z93jEajL{RvH@#Suqgcg>aN<_mW=oQ<ox67AYU>buveOPGD{cGCVN8``&kfI@@4vID
zt({$3rqoBhKZc94V+MG5DXT=gxpz#pz_;fP-&XMl=vKybU7e0!=LA2}579cC1V!X(
z==)eddRhX?WI%vXE~%!Ql*IeYip{7R{66et0lwn>eNh(wD3VCF-H0|P>^Z@4!WPUT
zEu9(Cmr8GGwpc4G3C_LG>Hg!JcYR{%Gzf?*x%7766iCs$2{TO2ofQI2tvbez0<vI#
zRu(m`+fDD%89RejO+^<wf4kT3<D5d4nlrazxVct~Ss4idESS?Q1Uj<*-?G5emqO!n
zfaviy-&@B?{XOOUa4d@p9s`1P+Wyl&!v@OVsZg}MuRlb9;bXn)v?-Zs$IA#AAHVo!
zzW-R1#u5)nv;_Z2GlyFrVc;5v820G9jyet79kCtD=MrYx$0yQQ@N(loDA|DK5OwE0
z>22D30Q!_M+cE_keWRz<yoLVWkt&drL!WrdO0__ttK6yI#Xp@x{7y)aA;Ib{g<XsD
z?Rs}IX*1Uwz1f;_XUyq_q~!0|a!b14Z|B=fq36KJWS_Nj6S3dS$FSGNRdG11nl76X
zzqz)XEXYUa9RP{yxIRf=1-t(VcvahF?Nx8K)bieBnfzH7@%!aYdBt(}*Y9YeNvs4m
z`a%IhH2d}}Pl;M<kL2kAq4I}DdxYVK>~!KESpD|jAbTu>$7uPth`*1se{#<C)2+As
z#>Ebf)r9`OGm!HUopX#yXE!CTyMOL@_2(J>BgB-Z6Xb|N>Zni2q>`!}`|m5Np;Fia
zjLz$vpfUwLF{VaEu*Ev{1*ApeTjq*$T>RNH+A}^}f4SS)L;dL$4>1$8>7#mk&?rbP
zmr<JKzXhEug?FMn1Aw+u!`<AcH4Q{?aF{XsQon-6AD(L2rSMPmJdIDKhUu@*7zQ4d
z<8!eg-5{KwDrnu~<lXH)s$%JCetA*#1=zG8z-U1IG9CvGe-DnF>;bn0!0Gf(-lkV?
zDg9<ktOfX|ndu}}Y(Hhxa5Ct2SJSP<g~3Z?86s(4pqSkmZn2ZNg_sgADG$XH$oSb(
zod(c2#ca(7QK#_n(&~R}VLCLhb5JGyP&3@BJ24>r`oEB&C#zKO$th!1bPJqqAbPTy
ztuUO+!ANL@ozpP8M#(;2xnM+ogog2HTJpek)SG-;JpwImEqpeZ8H(0|KuDaFGy}5-
zBtmF|i9@CuYPCmLk$+hS^g6h&oesyDzPwwbaqrXEg3R1W&h*^VJJVF$idIf%TXD<s
zC^zXFY1htzyf8w9*jILgyqScz)@iIjLo=ME6?=s2Y@eH{tChJ@$Hi~~r=kUObEW)N
z^qYPP<R!m|Avd>nJ&MYik)gHC#KY@8eFwzUWd0tx7Alcnr_0Kgnd;_@v=3m3uCcqm
zsLa<gP%xSu3EDik$lKEsOXU?b*e>xJ6$c9?1e8qbI7QYFcm;%47$!u@MR!X|YtqC|
zP^bsXS_Q;aYhuoyvwzm2yU=iKe)Hb-Y(M7s3NiL-#3q_NZyNd6cgVLE2V!2Lkduyf
zg=D&)+)pHS6;#}kY6M8F_Q}L$RmKATh*EY<S#u0h=*+n1KC<?DqWuMbZLJ@zq?u0n
zFF{EuVceeniDRVEYRYOLb$M0aS(qCwLJq>MyyVl$zP5!hqXW3Y=eq<JCd80J6lK1m
zdGZK9MQi4aFPEUYam?!b+;fA}C4YnJ=nzdD5+rF78)(LP02W`)9nr63nQ8Ft2JRSO
zd2)Ak=;bINcDEjUHZBeZUJbuI`M6SZQ#K?gloo}H+$|;OROK7Ad=Ia8Odpb$YSy2d
z20o$6FW4gyNrgij|3H?=JVFaf=R{)FEcjTgivIi7DeqFFvgL6Ml{`5(C#bBk63-v(
zq=w6lY;6Cce%V62jxCR>vC4k_svR$#jkvoQ7EVH;YyeLJsmJ;-XebV-wa;mu^Y!0f
z2N)N3X$u`ScHLy}d40-%ns^zY^|X@IfP2?@benF}Ix?Q2nhkvnpG^N}6VW3-mfU_T
zIRvb0@mX`*`2`28#pa9~m>N<mUG$T!0x{4(E7M9r50yqpiK87f-@-7TA#IIRz^8A9
zZm$6#dKO6ICs&fTfoZJQu)^|DFO^H{w~uQJ3y7BByUVYEr?@mW+^~j2Ikl4J)wRv+
zz<SxTsHkzIF!S(hi!Q6n8+r&&LJQ;Yos*9;Wtd-{fm2O+TLSu6CkV;*j*E2hL~Vi;
z2viU0X0}<;DZbp&#9fpy=f;?DlcLw;P#wzxs2X0&B8>S*L5fV<di2<t$ghD{PKilR
zp(PvjL*{WZzDg?5cC@h=uos@j6dW$l67q?@0>@G0{?IH@zx(SMisDn=EmmQU-I`*P
z!}7Mg`D}CEIGhU1pHNScK>#_3C0xWV_3T|KG)9PAASA@f!1wCL##UGb2ZekIk24yx
zw20Y>djB2l%=}Y6$oVi?bof}_lPqVST@wbs#vZCOtgYBcsB(Z$U14^=8mP~t{lX19
z_mnDd?A_gsunzdqAV1d6kt|>m$Hv}_I)avvRL#8b?Mnp&>`N<YBOKc%rnelgKX1pa
zfbex@nAGRkGCg>IFn;xv-wjf4G10aQ)eYmdvQu|ZRq3TNi)a5UKSBX(EZX(nAjj!p
z$Osyj#9(fi`zNSHM%lmL!<shenUPOw-Zz?HCl&f%HN~)}VoqEI$~}JCVRy)Nkx?S=
zt!?q|YF*$=x#3D{<Z39*54nE)INEdk9jr0+g{x3rv7qEBTjhl)Q1%D<tiqwH+Aven
zn|6%IvCGa=!RrnTaonC4gL4ua)GZw3y`g!!B`5S>yUeW1@8u<}+0xrI^O6Ij?2BYx
z;mBK#-GgK^@Gs+KrY701G>jaZr({&>v4WJPDnt2~q1)ExLr~_OHiB3W0HHh*W-<kS
z`#k+PD%Al{Zehl&NI=JZxfJ|?A(6TmK(_SvE6o^>?i@vt(o|D;5zGQ<MC1A$d@0{M
zaIyNRFCFns%xH5zm+bNqF@KG1qS8gAWI+CJr&l^K04*jMA-`N(5zn3+Epc>o7Ntl$
zXbBXPfs?VzHzf9-6UX!_X&p_)TPcSj?T@8WSBf%ivuT}Zr*?7w;TqYktwsxS5N-|p
zwaHIj*gxfNEHW^zQ)6{Dse>7*Z4|Ac1-$Pva(EuD^Y2x4@YNc7w!>I2n3>I4uA|f#
z*x68a12vuk3H{MbET-NWjzLEpQ45E@9Sf{loYWUJJe7ee1vlA4`Vd0FQM$<JSOF8X
zBmYtVb9wh|@+fp@Y>tnk3ugPi>3^=*H)Jy|z(!?GE%ac-)%WK>jG~e50=<M>B_G;X
zi872DAA&=iWo4p|q_JZEGLajJQ>eN<cR6vWgCBUmk3m7j%j}j-XS`Hr2>VB*q}Gl!
zc9K!C45NE$=3OvcVe@6lU5Ws^7&nAf0Zuv9E^6-<cgYw%j4#CeMd3H@61CK=Xa7U$
z)mmbwrH@zqG=ZV2yf#@nSsBr?8#tdyJ}y0O3vgt9A9%wz{QP}OYI#lXCK&62E)$Rq
z#F(u#ht^@2t$r+}YOt0Spj#Wn7kde6SM@&Zl>b6+W+<)_TJD);pe)PYo{S3r&fSYf
z=QURz|ET9BbLv7a85hT!OVLH0cH4g|I^cKf$@{yLP)8kC4^*}W2MK<@BpR2vCEg6p
z6@j5Gnm(`yYlx_Kt)1LNQ1>)e(lC)~ci)C+k;*DJ8=3way5oO2=`mDj_XenOk_HRE
zWtHzO!kC=W*~cw0Z(K5WGR6<-l(CXefO5H_q2k~q1Y319bnyXB`jm@xZk+8fOY1*`
zG8gejCYFlbWGe=UIjvHzZg@UFdC;+CmTbCfDIGZFg&Y4$S@_&CGb+LR+uGV@n0LEk
zfomB?f4fnx>c%-o)RIl+irbb_w~ktoA$GSg*(O$&h%lAD4iiF~<qB|KKwel7rV?Jh
zaX&>K1(+BJ0$ezTq>#aezk=g{MXtT(D?aH>-{+kqtZO?zn%>NB_sQyfIHfzR{+Si3
z;prD%OqnUk=4GufFXS@^y4@JImUmWfTh!IjX7SzQ7+lb}DN8ZdBupv^`h%gs*FFE9
z3S~H6qnr&~SeV7Ns3*D`jbY_O?ysepEs+jJMHS*V@`Ry&2EQEz?6M=aRWqy@-d5ch
z<Jb<;m|@=bNLtF`Ck*Mo+QYkujO?BqKnUh%UOfMYWW7NwEF^lW%SC04PCMMWevfLv
zA=B=|0-P|WHAqR#>o{XF5&SaNVX(*Ju~<_jWn)k;=y^h9k4U@R5wKRE6Jh;P*?2nl
zEk}ou?L0C`Hzxt9z42~+sXOS3(5WS^3`2SR_ghyfsR{k)$><?yG93L7+FW6`mu;0@
zQO$QrYVk|af?M`TgzY(vBh2mkK!XXI{Dy=@C(VrEtvn-?>!_DV>UpU?P*U3=-LNcO
zWUbNeB+k?82|ha7za=fJ5!K5ffax7ij|VFz;JwDv@_Y}3p~v&g+y&p%%<wcgWp0Ha
zVvtVIA*MYF10QUdP(EW4I@C}6w286xPB!}OE3TDAxDKV2QC{bR=;%#?y$22~Q4rxh
zeN^^1mw%G+_%GihXx_umPj(<0WhWsm#gZEFPg#O_0t^PPbQCXn{zfWv1ohbIU(E?f
zod^W`!E>+fz5r^2j8-jeRYeqpNk`!#)bxHAqvk%9(d3(=p<JHzW{<@&ma@9XxgQQ2
zh9|(2=rDV99rt^^8oZFEQjlg%r%?r(vL*nmJ{L_Sdk)qXWyuH_yoja_Crw`ZOA)!V
z++%O1ZC8>sQW*Fhz4ogCJ{x!%Yg+%U!&2MM{q5c6?D9%FmJ`LZALo_yFg=f@8T=k-
zqah*<p5zv5;~yB<_lH}T%E;3A6TC8(UMNojocFHTd<lcvJ7YFCk6ilkrZ!4wQdD$!
zLxN!(G#-&cjsZl_T%;d!hsXbrzy6#99T!`-qs|-;DDwF)T;VMQ4rLF0FI=sR*g25w
zxtG+qWK<SPM>oIPMk*%L<TL17>upZg$oOV^6L77UJy9@vm%1*K`+RK&UajW0-77;J
z{!muRmK7t<+!n2`TWmfie(2UYGJOxuNU^rUep2E%F$AIf<p(0+!&2tl?41x5Z8jCk
zSL*N)KQtJ_7<x{~xV%2Dw1KivAw(-@pA$y2ML<U|xc!qF;v_fgzsK*2dY@8OQ>e(N
z&xve>2-1G0JaaK#;WBRmy1)nMu%G0{ma?alVG?PkQ<P$!Hhds2-*DTx-gEp^PFqz&
zJX28e;hRBydnZJX03KqfIQE0WyT|>f>L(DTqv{NWA{0~b^B7#eJE-Zu3JM4%*pZSH
zJ%wH2DpjCuPBRA0`qjn)#xM;dE^lM)H&3sg$$=EfuLM3b{zQIN2kg!`&8Dr1V7il9
zsafD_4X$A-5hvTc%?hNg=hupk63t=~nfTbXlGIh*7IaQ&L4I%*n#-r|bo8Szhv5@j
zD~VCsoVAm~jdQ>FXPQKSx0BtrAfy4T*1c<?Z#JCJmmLd#$?1->kzE!YJ515McYLbz
zn-zjku%u=z$I^lI`v%ICK&PEt6A!;W9uX|{zHrCY8AXy?HFFFU-i;|Lb^S3*Gm`*Q
zmZvv?ma_5iuO*6nj`2_jJ9?5q;I-lcC+_U#vg!c$x7d6DMi+2}KzXfSgXE@bGSkEB
z#C44$@A&LsKxfKRAR<&YDS8}!RJKQfswA1Fv{Q4t(<5lpx?+E46l>xH;3B4b2^RJV
zwt6sbBj4Ct?08dM?Zgq~ajQ1i?NkyB2P!H@AH4Q-P$gjkA#0Qei*snvUO-;b38d%=
z3O*rANMw{qw6UCS$OpIB^xwKL`aR|t1)cPWb=;oOGfIO6Z=>TIdZ}3XhIHgJGBXCQ
zy;q3DjP5{YFH~m_2cF6=k#f!9Ox7s7vqVQzi$wP1RmM8>VKp@96q%1jrv`N1-XsV-
zoDC=QkW&824)Xa?CjDUXaASHW;c+z5C@`g6t@><OX9RIt!4~5WYOa(uR9rX$2$Q+N
z$6H+++t!=tuwMja>>{~T&d#qqKaTptEYyb^FoVvgB@3@3#--YSsA~(+wS|=?tWvi4
zc7y**B}l?-k9~2sgXs8X`Q;XKp*k}mfO`QrYSd`W>et=ZGJ#_i?@(1)&aZfiOqr}+
ziZM@Le#*Un+bd(k-UAjAK<u||1&XI=Z`xbgP4fhuk4grMKBO|<ua-gC2l|R;`TX2?
zr9yy`MtPVXHT%SV%r#i>bL7c?HTVm|^h>?$KlI%*D<JaQB0epQG>Jru|C+0$ubD?j
z-(=?`?ANVqihLH0BL97$yF!(X>YgIt$Nl|Y#BY4(!l<oHa5G%z1Fy1Fi$^j?D@g_U
z7hQSbSUk0U$&Hbno%H*4$=gUtB&x!tX6J^okU)w}xIlR0H2Ri^==V^=QVl{v7$2ii
zc)QI?JQ3bYVqyNxSVWynNNFZV)KJ+RYFmx#6Ttlei8`3z5>g`@CjEuH;>zAS%B53b
zF;v46v#<Vvt*md&hJ(N`lm&#0K@vDJnZWJ$I+2g(>krHCMF7?1xtKufvHl%qafy>r
zRmDV~`EDL31?r%@rMA39OkbL1{8H^8RL)aP$V?r%>koHiV&|ugek;NyK^PgDG#ryx
zVNc$RUD3qu4@@}~qPjCtHgOKg?CP2-;lk+~FV{S+=avzbsNC%`5<6aQUifNgFMu(A
z1<G!N>>XsCRwL#~KeMLP&7|gfQRR^gk=lO}iuoI8xhVY$#Mk?;Z~)~>c!Y(+7tRKM
z_v#HXr`0dv478(~ezieFDB6vvImDmz-IvM+0y+Nj_1tLZAJ^Y~{pHKZF3jTx5adgT
z?iCN!+QzWoL*GjTM`_hG@`mq=Q4X)#U!<lpaqii0X&McJ1n_&lBj+ZY95(MI`|rPV
z%$hc+VVr^xnM1jYuQ(-HJ-EZWOo@}9PHIM2nG6h(h-_@F-8AfxoRFmxE!Xqn=_)Q>
zhg8`20a|9P^ya|>iG?0^n)tyb-DRR}v8ocURWeZ$M&k>pkU0XrIsn`X=bIGqBvW2W
z33*oFwJ{NAei<Gf5uw+qq*ykR8KN>iuZfqUs$LLBwC!X@u#PEJk<&$xE%iF^*Zx_f
zOavkS9p6T;uPe#J3-IT0EeVr^`^zq=+t<;I5B-b5w|e7+h6&AknPpjC!K254EaA~T
zax2q0=KSd_7P>r!6`5l|_NW*|Yh+h400(K+zJ~mM653`^*tCh(Sav&YG(9@l%${}m
zBg9W8mj9S!1$LhRS+$sC@TTiYx2^fUKH1k}Yz+{114tL1@fjyBeP3Tw6d-<4oB=S@
zvSopPmp0z$P_$sZ-z^RF(;$w3VbJ-Wi!TLe-yaqh>295=(zf5*Pn7E%-1A^JzPn3W
z%u_|e>!#RDph9bC&k!ln<0p?jTpRRp`0=9ak%p1M+N8_=Na|SJf9#LQHASTBTBF7b
zeyNXgDhZsJw?eLx`Mx*1wFp#UyI;IxbL*<i1tPNR$|L>bV1p1$Aw4GO%ijqd$I1pN
zT6YMHqVk;3Ms%wuRM#Xaer>Dkj^<G6_DOrCm|FiEOWNpEx+rSNodBuL;4<@--Oc2s
zY4%}3nHWp4NgYyV^4ZxKUfsa^rJvWWW%fLq62DB}3o>f_ZeG1xeWL^SLWkIWeOTjQ
zV+lIWJvhMiSa^PLco2;Q684rKTI}Cxt%Px?g#U4r$(T8}{?_wXX8MaQ)<el)Y*95|
z!>)mGXW{&N((W)HGC9|(mKRkOI3~<{&x$D&DdGG=lQR@QWj+A~#G6(x`(&1C=du8h
zD9)G=9np3h7MjUvu5^5mtXP24q-(`OW<JK+C)7zd3KeR@l97mD-!or!m}#o?+5ue1
zhB&JY7KOPb0`|}Rn28myXBP0<niZ1s^xRd>TvB7)vg!@ZoO*@F^)2eW%-*zGbr)-D
zty_EmGw!_jBuMcmAfY9Z?toG7OYeMkj)t_&>46d`hX*tJcn5>-U^vr~dSG4PU0Q*8
zQ4I?#1|y+JjY~a1NSa-d^Yc}VW~Aey@=$l;1866tI=s}{>{#)l*rj;0MqKG=9I;&0
zRxOw?bt>QHc+YB_%5(E0P9tWucm-w)HdTFObZ@}ZWMqYkmcBM~^^fBb2Wa%lrf~ix
zA-u$SZ!#2z!77GI##Vi)CI(LmPO6_A^W1_ua!W<J!JzsB7gsV0h0V40anK-22oc`h
z#4~jxIyBR|GjewpYS-1DQa??r4TJlcKe~16Gz-!NC5nC~j9LWVvt72($rU8Dl<o9N
zr%#C)5|Jwmp&}ZzcPek=Q?|Qns{8tY>GX#9nb9RXuPwt7j#^_8mEbs8Lj#+%<koKo
z$bjSQoAiOxHlrZq23D>Q(`sE;X=502KdTI3<u)QDeEq$D!ALhpRDD6nFn*1Oo?qmL
zO*eS5Nt79D>^6WK=xeot>BS`yHBmh^GRoy<6|N{e)YDP#?ZEsmRY%85mih$^@L0te
zrdc=~IyCmgwwP#GYt+j}3;^~_YUXignOy)<x$)y^`c7b<c#kmt2HSue^UL2R&UN<*
z&aK=+tJn>Kw%V^gI!>>YnJ0LzhmdDkWqobC!Q>L{xejs4wwW*d>*@Q!kh11gt-fmS
z>M_PdVpV|L67|Chhx7)=;{>Con9)U|PZF61P&xahmsVM2l}0Mf*TX|uSF~k0Ffv%b
zwO%p<ndzu;+T8}U{;<5%kHBd+6cT+Cbd5GmJ{Rk?*kqAajX0HgmhkpuTA*t!Mfxh4
zjvYSmaK+GQQXg<1bL=}#6m*91%EeUZGoe0vh^Urop7Zn0JjI()cJ*&-4-cVd50d%s
z_o@4d;)Z41v)zszT})1cqzkGoakI3`X~sf#2J3h2Bl<Cx1GU|*Zz#UlA&{=LCm?q^
z$t@2Y3e~T0K3MbS$vy`VmKeJgX>*>?MI>zlsR3GZIQtLp_8wadA@ingiO}Tqq?Mgx
z)|~yyL$CM;XNxeT8Bmw_t#I$ntIwqgc++kDy=pK#?+lt^b{M)qPjq1#*9PC%YGRDp
zLPnIneZ0%^@}jXps(JqS=s3YFuR`pFIX3=j7Le6c(dYWw*0CYZ7iAOSnVu|T4ra_e
zkQP2<cUpGy_b;fYwBv3IQu^|>XiskheJwG^y_Sx}ULc&E%WH9jnXfqs@$N-PXpDnE
zMsJ@KQ9rrOY0!1=xKD7)w$(*k6e(E#0OLpH+O?+-akllX6u-=(Trg#Vzudw=`^s}^
zzCD*Whw`qgIS%<Nj}gmt^O03N$r}Eo<)FN9xP3#Rn3b$cjrY`Gqs>wkTeB#MN@Jwx
z7mHK4N3@%i`|HmnhO7{_h^Rk{g0#bP90xAaM_2nU@1DY8XO*2H2?>#;Hz^dS`_B`I
zmebhksiU@@6e+PMriS&FL6Y`NSa%t6<+eg59l%-aX(``cYTBtUUqZ3`vc=o7Ht6T|
z%k|*is><qvb#eIpWUre`+RApkv9-w-v4=@$SfyTL<7q)lCONGJxS7oP>qps=0d{5G
zS3iY?FIRrnS=M(QOr_kFm2H+&DWvmLQ(m9mH6#(ISxeDnm3{rS!1e*ZjVrb3wT+f%
zYxz54RhhzUK9ndtk_dhcui!>GfAS;zcD$a|q_%-SqtDGZfy$gVKPyiXI=TIF^U%!Q
zMuOnGzerxmNd9NTr^UVKUQqeEiZ7`GS-mrp5qJ7dSI;S4gFzl?;!?3oF}zkcPaOUo
z^mHLzGpVKRLqWa>A90w!N3^qEi4{_W0zxC7DUJegtPr;NWKrWB#2l-qIxZ^}RR(9p
z^&MKk&o?k6Hw!r_*%9pfU6+FB5cogQkm6XtHG6&I969pT0YDl?`Xq>$NwJ12I)uKG
ze5pKu<HYAwrc48$9N1EO+}U}9YD1#9H~rZ9LK+B+gJs8n2Sf|8&8~BN4psW`*<I($
zrgYF3t1}T5O50h%yeMTsg2c?`ZXL=(it;>hb+$vd)2O6u=K%FM8Xk57>zQ`Ek_o!J
zJXdGykM+!-HR86O<nu_~p7di|sMgK4D%r@6Ao|M^J@-_1&v6|<-buevNT=mw1>S2O
zRsN-s<{wxZ0rh_k4Vvm(L8syp^GGXkLQFFm{5X$|b+k;&pW|U7jK{0|`?zAVz?0XF
zc_QtVctOXfwbxhs80+B1yGqlO@!)HE=&<zFiwioY|NbMlQ<j(2S!X;Ny*<3D^SR>)
zK8dWV<+sauMovu6AL7AK;c$@9lKXxGJ(}QWkZHcLl;hBOz5C*!{c*EvCHPwUk?U0G
z^rG$Oli;{KR2LI!<j-??pHWlkhuq8l0fNwTE|wXD^92+pJDoCyp`pETq(k>8k<R~t
zFjP=Im*IeTI-OUztn4|`%=*i#-)+|3PmI0<o(#sc6SWN3$qz&kkws%w*h}4>^|RNX
z25kFqKf5s$3A`Kl`xi@ZR?l&UhoS415=9yMU2(9iGGRsXe$)uA)nq4IWuI<m7X8Eu
zf1xG6?prGh5l6&2Yhr(v>7UV&_3r?2*Rh#^%T_9|wWn{u8GX7LDP52e69Pz-&L-6T
z7EcIA$r@hSPSk%3lKb{&!nE^8Rr5wJ_8(1{=nUm2Jwo<hid><>6qA^c5#C{VTpk-x
zITpsXOIv}9X_;lc+504`&E<G_<oM{PAOEVXv$*Wp2aqCVUUPHIhn%M-S=Q-({QDmm
zCz^}w*w)2Gh%h4QcILCK@+9(=;H20Vg1`n&=qwbWT4@xLV^h*FfMnqz8h+QzoE}ZA
zikqoix3BBF6|MZiQ{n8%grgn*GBHf<h-?`6ejHouWH<3G9V7>}D6svwt$m>NVZc7^
z2EFBVw*Ip7d;ir>a*2PJCDc0$lun+5Bus^=@-$e(G-7$KKUUv|5NkHv1;FNnQ($$5
zmuja2!U8|7UPgz~(Q)$*Z0U3jy~5TEBsIWEu8MxiY8CE7#v;z0r94RXUdAD|p1_2w
zC?A+CsGn+awVl_HTWS^FP?+%-M;cZqajhetNGC=idv|fU$cMoxT8h*3T<j%$Oq5w7
z5<j~Y1#uUef!JHWJOrXbjQ28KzW}P5VCBI6Jc-m;RA(s$xle2<KgM=QY-UCq+$D+O
zZ18J7tR^dx6=r3pJNr|RrD&_}F@c&+q9^!K`fj|#?-mfA$YVYy+ZN-s$U^Y5c-;q^
zu@iF1V2uz?^*DCW%Psk{-r@;Q2XpRXj0zCMe$lg=Po&v@)4W6;AMX^zni@*?OQjJm
zW1Mk3DNNPID6uM6Q1jwKgXo{PG+*XIDxzTn9Z5|RGQ*yU$Am1kglDEFk^T_i#L{~3
zPEicwr2ls8YkHurl5+R7jXO9aW}9PkHRQk4>Z4)ELJf4>B6h_FzTy(7-P|C4h&`q`
z%cr$48LrLF@J2vnSU5!lWNb?M<|AAC*G&IP+9dp*Ct5<($c;gv!aa(<J4Q8hR`b3A
zQba-?&ae5mD;%iOjg|1^#K*EA@N&wZI6&~mNX$k?2}P{XX(T32>84Dmq1Dh+Pz2QA
zcppuZMd}k7x_<oqmdK}Rh9!07#=pP+&^gPv<nXoD$)faUVwK~b@joY?)UC_;SEerB
zfhQNF4qjiSZ%$0q(a`0~+(>tss~zc^GO#55?|-lE<^(z|z@M*o?n5V@cRuW5_}m&t
zv-w2l-=CYBd0Brv)_6B`a$dVEk~daG+cT=y#k@IiaPriFC$;|P@lT+U@bOY3nBbkF
z2DIojmqCj3t<C&_XrIpS(R6ANz~p|ld9q02t>ELQ9*<G6eNG_dn|C0c?vxeGxPYmP
zmoDKJa2d7ZFtb1~n;-ncm!b91cf3XVks@a_Ca7i8-s_1O3wTlFvts(3kVe8%`-}M;
zAMk$S;{BT5zp|`_TS<}ZTZJP%>BorRD-LNkyU~rOw>{BSph|4@g$euk8QPY(UWGkL
z2yhVY5BA#`(VaK=X58+cCD%~@S80P=u6RxS)N?C?LX@cVR9lNA7HJOmYE|<Y4`TI%
zr|hHy@K=<4ZxVv6e@iz6--&t|)&HqdM!ip37eC(bAGkRot>)tRwHl?#l#RO;7}_vJ
zq05EI9>~!Pk3~$pF<iPE4RC4VpPS`c*+P_qf(*)y<X|lT749~1xm&#cO6FTTCV?F-
zO(JeI6Rvt4Pc~2q2>gBl$F!{M=yl0Rvydf9W8)%z6IOKvcadngt^eINaWmT<m#KD(
z=%fy4NNV{#OUctMgGGoGI2B#(Q-UT7;tYovD&B!t2jkFjq|r;>;XC%+R(193fwbZb
zW4t~A^6t=5Ar(NzESxhT(^MOdqr{p=cq);K3cb+b9<($}d(o)%tdPR6HY<T)i<JxW
z;_{L%kt2H=*|ckH>D$3a6zbgzPV^NkZesa60@je(*(|_U{@sx9Qe<w_R&(*^re7aN
zrF(s(por8G^OI^B96Sk<GBPuMUwdXYDS3E>&^_UFHPA#hcP%Wu{P67LP%P)WjwrB2
zg2dd2<0uk9%(JQA2Y!2Um-vX5z9zb}H$A&Ww%_(2o_ht~H=5p$<_ABV1}iP!2|Uo>
zOUoSLonFg;^UEDnZ#U^t4zflO>mw*rk&y)I!Y{OYq7r;(x-1^yj<2jnHdxiWdNO0y
zkB`1}UII>=Wu@Cp-POqbgW5=zo}S)M;>`4lCV#DNzR*>m=~bAHkYmwJ-nM6VY0L3(
zoAFoYEeYDCbF;ln+JgF-J7E)2#-AYpuj_7?fdgy#593Z;1gfu0o&tOdN~QWN!p#9K
zO?JEy9X+=J>A47bMe1a?vOz{L4}U4L3U5%m{sE+{Vc(a$tdO)tj0I;X`r1!@q`*nu
zGTH~|;v`nlhQ$bKU^DCcNbXJ(7^CZam13|cnbUsSWPa=(W5{{lt>e6@NHLjF{UXfc
zrxmjL#ohb2(L=ULxY}a9+ZSLST@~$V0ygOI`^CQ!h{f*KU_ezO5&C7PyMzLNxO*xg
zl$U6gHrzZ_yZo-}UTD-h{uM4OMZKM-p}CWtwqYq$h=hR=+WfD(rJV;r)|Xdp1cJdi
zdE_iS+2lCGCmm9vW?x?N;w5%nqbw<+AbgEp;|K1eJIe0h&zTwl^YdCp@u=7Z(}nD5
zSnLFaTqg$k#?LiYh-Kd?VlZH{s?Hz{{78(K7&xzuH;$TQSNZ|YPYuI(f^^>DCn(Xu
z?ajm%f^b^buo2u}#N^_UE9<e(_)!C=T05(a_^14b6Fi#?P28J5>F-a^=(fpOzU?0|
zdx#T6ZBKZ`+r8?|s2eGLlQ!Iz<Wa~plWztx*zL|Pr=8SV?`NI752C_Y^J|1ilhGFo
zRN)DbSH|jD7GAkL?mfmx-#-ry(Zw~Mjw(44ngoeDzh3K8nm>P5{GuaY2=>Y2uTGY7
z>yG}4$-StGLz5Q%7OPW>xd|PY_@c1cZYL%cSG=jUxvQ-)AtuYGCef0$o@nQi8ZoPB
zHJ%cup}u#}UTB#S^7=_7w-)ZzD*PAa^aXr1DvH7Gh55)omqYAz8T6|v(<Neemex?-
z*h1K#%{vYJku5T$ok~m3b5yg%d+StkwVphbu)a|S0aB1muE1XJ=c<gyHO|n~#x@Bg
zdF}ntFNk#Q(lqaB7C=~U>CthL*s;;b<cJa>iPefB=73-mQ(-50boa)A=$(v6-v3a;
z!{oxb8N~{s#c}2~7B%t^C$g+?)sEKh3_8ad_T_g>J(h=wf-3RgscQWki(vB`J}<Ng
zU0!)k1o^7K<4GB|)74pu1FG|fdKMax?A-v*X{k8PYMQ3`^XhrL8O@mG>x;51<nWn#
zVr(9fhKP%aF>Ah_Y<Mzu1#<>`W|DcQ97Suz{ePXE`9IYA9><S54wZFKw!}2{C5q@Y
zMkX`{ji&5zOj#miO|s0PAxoMeCL<BDMr7ZmF&SB=F}c=B8H_A5ME2o+&wbqc58Tg>
z@Av2N{sZ2R*ZcYOGDMFhh(QrHL@5hR*xsB&oy~w(wCkQ<uhKKky(}Ly7t==G;g4%1
zT~Xiww@-fSSLaTMhEnt1x)lz$9cp1{U}#vi^_^X{8uPsks{t9D3FM^N9muSC$P`O3
z7eJ@vXFye7104EGB*JPjAwj*m3h!JmAD!L5S-IS~eP|ZYtiE4ZIJ9Xiba8W2v?=IA
zszl#@|HeYLZ@tBc!F6=07Qw$~c0=l6^Z+O2OxcLT?DW8FzFJ7Z8>#%^nHL810eO@+
z4A3+7B!y#t0_cysHDa{RYCcmvIm))RNUm2nWrK-t+MBWR>ok{pL@o2U{Fpx&(lb;h
ztWNaf>(tdb#{u^4_zDM2o%<)0kwf-3TwTVu`y59nVq)1ka1p*VE5D_HR?2y8wojKf
z5X<=;4ye3sTz_`)ph`srZ-u%dGYpi2uKq}m%QAI}Eq8uhKO+Ri%OH|;QI5)l3*cU@
z{I=uF(5#Lt2+Kw$-M_I(n((n2xNIpAxK!h#i962!vCg_EUiL6rkT@?`V(fgRd^lPD
zq>2_!Y_;$2z&MB$DBhayrFLbRV=%ZFCsqdG&YIOa1A`JJ$4r>4%dZwkZXhw5Rukv-
zGc^_sN`5*C_4WHeZ<C+Y$F)r5mrI2?uo7B~USGddKt-NwB>{}<o{V4ZH@xn0wh{#o
z7gW++Q;9`Q*6N_!U_r~{vUV~Wh17R#0*N5~8;07Z(q`*IsPOr6)5W7Ek&`uw{`}1B
zrIVb%=^y)4NePE4oliMNQnHwSifwU$1^8Nn(vuGNbWe4(wcWG4<`amdZg-*A5Trwi
z0|8lLDx1P|%O_58{o^tq3?p<ONY#JwDMY`oO>vG??z{b66#IVSh2|~CqhiPr%AM9H
zP4B?F$%`XgTliU#NH|E)4oH4jLiT<faT(T9a$iVjm`Y1it4wkdpNswGZQa_1Fh_RO
ztv{D3;#Z%oNmEp2{7CyIFlkgX29qOPHXAV@WfxE7FDA5fxMQR68}BNTkYv_!q<!nL
z={B+U!U?zhrDPVb(R$Y<<Lzii^do$%=O|<SxlyNksHqW+6-y4Es%dV48){lJ0Om4j
z@XNo>K2gp76bxnZ$rftK@WBgz9}P&Ul>Klv!zdv9Ln;cvv<-VM=ubl7sq+Xgz*UjJ
zSk09`(I$t1fy=d*eS1~|Y2{o?wX332uLw659pam(Ri>lxYoz_@KMc)**n+2QnVCl}
z?aP~EBJAdF$g@&w^Tw~~@R=2+1MVBWVS0L*?colC0cPd;7B%y}2viY2t8OLlpyYs{
zC8T%3?|N)?4%Y-7e$s^Z<IIXqNc3=R`}pTZVyoy>^%yhopvfDComtp9DcWOvhrFWy
zS^AOQu@>jGm6h)D^NtMn7s*J~s9>5&Yq9UhmRMGgI3vJF0ky}=i~0x^ZHwN$FfN_I
zQfOQ{5&-T8E;wnFS>?jk*8vM`LOpeAm##cCumsx{x48Qhg`fNy-SfhbUv#<MJM33-
z#15i%H;1ubvLwo-KP0#1$b7sq3Hp@azyR0hyDcM!iu0_*iC%Ra!Js|u_iIz7Tsxte
zn+3M_xSM|IGcT$-X9e=U2DbZ%Atvh-Gm9dB$uK8q^e|dYe}Bov9O$Eprl?;DfT{fP
z#hDthJXkRONJ38<fX>W*%MvwdRnb<hpmyIoZvQJe#jr|`mpm{3=&x$DZmz&a&*_p@
z?!5m11xNsP?g%-B#ebH9GFFqwyenp*vvtCrxtjbi2pp*hDKF1bkY{aQr1Fc+Tz0RY
zxaJf8m;Bh3IgE<eD*^6Fb=9GmzuYZI>`~NVWaeYIhNhOi+<1f1u`b5G?HmsMEb|Xf
zVB4jLP|p~{#JF8|iL(5&LiTlFj;OZ+M}fo^S5~%T7R>#f3+4?A&rM4O$&wUc>Fj3Q
zF;dIln&|GA-U-{($GiiIux85}xuFMhE5m1~tNmURV6=yfS3*8cZ~4?IU35kSaM}vs
z>eJ(ba*w?0e(ZdkAb3-d0`btNKM$?VDeu&a9DGK`OF(+`FK+CnpAVk|vICzhU3T8i
z>6_zTI?j<;`!~UNYA*?<CX#ZjJlIgttrhKtfcFA|B5xq-<nK@-9GQ_gt0rklVlyWY
zS1AbD>zCAR5LS8^M@p~}2o7<|da8OA6<IT+KVfjKY9;2IBx-aZ!OnAD?y@u>)Ds^E
z*I{zyIr{)43N1OQW_0t<*hDoY%d#wZTC$JGmY8Wy*=;jTOXuUn2)L!`Ml9q@=Y8${
zj`_x_O<s@c`NgL%YjSco^fS#Gj9KYfWJHly!RYBshyY+-!TbL!hn?n=8xb86kIueu
zqg(QbK9J~p(wkF)7rdRO1d1hq22YX_(A@^i<P0f#YCi)}C~Lv}<m)mhp9cex<Mf1W
zI=cqWJ$U>s{hLJi=B6SjB?LzpbH(;zwnR*5>pUqZ?Zh=c!#p4r8%K$l-cO{`I`A0C
z0N-<{Z>_m{pOLmT0|&1*qD^}dDVc|CWp9DxQCEKI2CXhgq~`^#Q6DJu8Eg99Y#?Y|
z%bLcbiG*|Cp=Wa~ZrzujU=@NF0TRS3V@wVe(?#=M2qUjHsj4Jef@~GFkoJ?t9Qu0(
z<Mf9OYgp~r64j2zx)+nGD+N{~dMdqC$*72Ei-aQr(Z5NNS>is6>}oyJqo)~0`5A{?
z0U)%!)gT!cPN_A>K|Q%PR2AN@Q>~Q-IqMuzr+3FG&GyUkQDb3`yW4bB%wC(e!K)Y9
z7c6vzL+t(DPAQsc^}a?j!tPd-CUs@%l$MUVwih+T`K&-WwDtO8jxUF^K~N4m_1SJ<
zH4vwM%VwbM_h90riSai!+!&hsheoBKX_{lu!*Pi`QyW0OPB?TwJbw77q?zZdfTjm)
zM7qHFR8OozX{!657mi1(%u4O@4C^YDu29QD@g;=jwl+-mpwMn)O9T7+#_1liL@(Y$
zw#DvI-5cz(^B(pYb&AFLxzCG)@zxpG*47UxZtDfuyU}X^1}}1xW3t}2_4Q$Mc-y==
z_Tu`zmhh;mTZ%>9jZ-mwzV#Xb<0snZ%@6W4*!@33r<W%hv8>bE(VzR;S>|VUMBi{D
zxm#|O_y0`%=iix~gQ^(pQ|U7g>HjTFrMM->{-*_$_Ma{mNk8~si`(@7wf0Z>=^J30
T<T=8>0ngI>s@Yq(d)&VP13!vT

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_004.png b/0.16/_images/sphx_glr_plot_visualization_utils_004.png
new file mode 100644
index 0000000000000000000000000000000000000000..abac5920b907dedafd5fc9472e67618100e7f7d9
GIT binary patch
literal 9826
zcma)iWnA3Q@;1)mF2%J#(Z$`JVhaU|ySq~;6t^Np7K&4>P~6?6IK_2wE6(Ej-`;zF
z@8|bJHaRERN%G7yaxxRGp{9U^PKFK#2ZyDkC<}svLqLKZ2LVuEzlpxf9k88{r<|Ur
zmW#EgkD0p_oT{0ptD}pjqn$a8x0Snxor^OMhX4l;8;z}}r>loBC#Tc@yMV*R-G*~Q
zQsw|A1kF|PqX!%uj@iEr{-;EV9UL6vq>`+Zwr|$Snx8AhzTf`E8MJdv4At6|8cFdS
zUrCeOO9jlxTDzlCW7xS(?=~M#V58OIU6=Z6%W`|!v5f!1S}W6N-_^=zdoi;ic~Qt|
zNr_?!OhlbTKMZD%PO|;>1CT`Hh8X(&@bq>~;u8vzIy{4<;iN+N4}d8OR_Wu3N!JiG
z<`z<r0#Yu;8ZW<sXD&(s+3(uXXehH3rJx|dM#}#k5Rw64$1LIhoy!lFgq;UTRwuf+
zPXKv&Wm^p8Wx<c>vy7Jz9zMQ-?ryk2Hr2fu4^yX)@g8tjqIwSX>>9B`sKkVTh=>SM
z3JSxdP_4Ad&A!N`TC35d_Ou*ChZ#@~V%~m_evwnK{%@h`JV_0XH2CIFe>4E+^Y7VO
z<tV?z$Gw*>_V7LbHNLy;MbOIX{tPG#gS|%^i$_T2@(FKhD~;7zXaat!(X`^@M5E+n
zaDy3ezdHm$s?LC!h5=U$UZ-R5*_WjansH`TKvz5!;;XI?i1zoO*;r8_#8!|UU@<JJ
zwUp2`Dzp+r-RsJ2M0V)sZbbfsU5*zki0TQ-*RL;RkHo<&&Y{6>2LKeaHp&`O#YJII
zkVqj0>k}&BJMu~WMJvw0jhh2TD^Mtkp`=w_(Y`B6X$6o0X18X|kF~Ls>H>edI)Z!u
zpc{AZ(=+HzG5>@AgFH+C#SCerGRU$O0b*foaoI-{MbJKj7Ou?FKn&AD>{cvtI{;Mi
z4VyA@7T(m_f|TkHEn-oMenEYZE_;l_u+}>0tptVXkHo?@*^Ig{EiXqOrngF-slg!S
z3Mts_e~jIbe1}ki4(fLXW5=#O8UpTX5<#{AcUq|i><OvAit(6sMlhZ0Bl6<@l>5H;
z#`KJ8yWhhpB^M+8<eFwfT85Vh@4fcLfT(YRb{QE8j4E+cLP++Sf1O;ni%AkY-!)(>
zLa9hXV!Kxc!(`6Bm7<?6o{vzK!(I=0h6<s9IoxbAjk=Re#3Je~jX;K#vYV*`c9u->
zJ7Mj-6p+JL6xb!%kH6~UBGLt=@Dr3oJAsyo65Qbuo?+TdMDL1bGYWzw*_E$4%#LT6
zdTd6qiW<I7Dr|_oOyL=s{gVKrYy<P~6fnIjfA~PKXD={-w&anNxoDQGI9Sq%EG>yj
zEP^F%{}UDr@d_~*B#clcX<`1x>B#&P&|mz6TABYbUZsyPmZtW<({zW4!TR{EtuV^Z
z^PkZE|CxsEe@y5<R`YjNHjxI-g0tcP3?R+yNFzQ2{Yoc(3j>8G+SKRF@_3Kx2aCe!
z3*Hk{!L@<Rx9jIEaDcAX3oaL<W=LL9J$@tE@;=_EG`8@c)lHbL=nI`gj05+^Z-E*0
z=YL!v6O{=K_@eDWnjeHzITSvfaO6Dqh!+1+JJR)@er=L=zgpy`>Hv`_Ai{_+Zo}rh
zK;zS~2ZSs%QipD`SA7ceY3Ru9!7&OJz1-=|IgcB-y1f`Ph<!Re5}&v+4a0GI4U8A}
zLH=_=ura;+YLudnlq~}T8yeN1%}>%&cVdT*ugQz28%`S(usw7K!AYske5mz9_TWsM
zks!<EVV9FPhD>(&1SPLPhnDb!A@k6x_s%%qG;#-K|HAaiQndGt%d~N3+Jb3coFry<
z&u|214GLG_gKe%K1C{<Fv@OHcLJ@f8EpH+*`q<Vbtt2hY7P<-GK@4Hp@e0Y;_u|Ec
zX;>5=X;wIlLnJg+OVv_M8nX6++jl#k_+s3@eBAfDRD%4~`Qh;nVxauAjFoF;11E$A
z7~nQ_C~RX@v_UmLL5`O$9d-GPieOPACYJu$ANY7<E+WjE%38$1hWWEeN^>HIshU0N
z?(nE<Y;jyH9fYBWF;6#7BlQIdzF5!l2^<J<ofNox0n6+#Aa&EE#O*R>Cg9&TSXF(n
zC;HeGEE#6HLY(t_Z{!)UBj`l^X74_?8GWTCN@GedQnZQqBTK^zGQ8Wbf`R6`uCQJ9
zT^En19O1*9ilwApdFF2YLXB=0O4EZ6EN5q*68Q<t1m_@OdX>!CG|sUc`!0pEA?4eW
zrs->IeGVqqfLDtno4fP*uKkH7G3o2#>~!hP`e6y1E}cX!y2aM;(k95*L7NM0;O=$0
z`~Bq$wsPsH3Q{zkp@Xz<Jf{H=6G<rW5FE78B9P{)>nK0M$bea!(8x_M;C0?gPHn@*
zg|x_H%;_1<Q|Nb6s{L!S6${Bc%-*IOt4JnARW@@NYO#jndVURG>%@Z}wvUDQHm&a%
zGVzkT)SK{9(zO}-dZB*sv_ZR->FuB%3_8l<++(T#fHpakOls=qaBl~Be&F;tkF(km
z%lrmHV@dsq<L$OJk?!Lu&MTnJmD`gj0EED-3wB@QNAshcG6}vwAFG4A`-K3ogy0QI
zV^F*Frza%wXm$TM7DCsB)2;8+9Op+>FT5@+ln8mtx45rX3(l=w#2XH+fqJE0N|==?
z1x;QV8WYd$mIG5(=uejWeaF5-TwTve{?)Iwai;S;HAB$4rWbl$1ue;!@&(~qvnXCd
z*)X|TQEmIBiWSL?Y^I<e6UjV?g%zj(By{|XiAycf`+4|I;XwM#7nWXX%E1MDI7OIg
zK0D|5nRK6ZY;6aayfDZO<2|O-erM#c@j%a5uS(H3=?5qY4M~pTf}s)v0x~@!qY5p%
z#S(iwy2gEatG$q3+$uJz!touoOpz(V$CO2bX~pjH6>R+mdDv5Y@sXX(-FcB|`wsDx
z5E>4g<4@jCdwBV-{xr)>-Eyrql5njhoMdjHUNp3|F4+6p`Y^Lfs;bFLi>)v)6!5Dx
zbqiR+_d8X8KPkkjS}?-z8{P+2lvp!e+q=UXCO1Lp8YIH~&YXvGh3?JL`*fBxu8jIb
z!l@4^_H!SHg1finhuqdFcQ#b<D8CKSkcI^sY?O#l?Z^3ij<qp{jKAnhr1C152=+MW
z-^{qsOAo_qK!K4-iR$7$1Z$|)G;^B>Mf2>ySp#+h4|<G1C1$Y3#Agqeh>*n+?SPSH
z9zBXVXDN5Q_v~}QtX_xnE;MjY$82lsYWKaf;x}I6YHA_!Ws@9RV)+)jf&*d$(Jc$4
z=pII&(O9%NHaQ2+XkN?(Zl>xAVM&aBQ$JZ*cJIvrk~f+ISdikwS?OJ3ylceBJd$0d
zxBl2^v@&~vmsr&9Ov*fy{Qcca8u!U|U8nN^ZVV&S=BfhEp@=9E2~<7x<mr0f#bVSZ
z(rugI)5W_6<7m_6mGtK6IuVK<*Rs)95bw(Ob^x*Zl6K85X|BjZuY8ftL@PFuFOEVy
zK|sM+kL5+UWCUnaMt!v}zAL+jQuJ%pLfR%xD(F`*678WWXQ(c9!O(9*^-WM1LMSa0
z3jGpdxR33ecEjKDGfB8+<#33CeSq_&Fh248+IqQ3AlG5H#A<|?t9IWPm{o<r&#c6(
z{_!_;#=iui4d9!Xej4pd7KzGXr%C;r^nzK$G1rQNB(6Fv;gd8;0Z|yp7ISnJM=^WH
zt|7d2k*YkuGPOAT9_E%5Ifiz0Mmu{kN>uUd_e&M3^>mQ-u~6T(u~1*$mB9jTYx$$(
zm)pBx7I|!Mu2bTL>ji@M_#|nE%_*Q?k$`DX;q=)dnwpb;yS5oztyyU8@F%jqHzlm#
z*|NG9M0nmCw#b2gc-Dv>)1*TuntMkyI>=$6bkRI*4e`R7e7!}CsyCV5GS^f}T8g=z
zLj4=fUsDe_NOX#&;f5~Zmvl#wx1_YK?)WTewd%H{8ums>hVh4#R;SB1Y-01-bt4-f
zW^~NPubNH*isHFgGR_)%%N>iqZp*LCe){qEFCE1CHG0Q?GuRo><)sz4p&o2V10ViQ
z0kKZ#y;l(F6L<rezvps#s}F|2hfZ13L1vyge`hL?cKD+<mN@iHsJO5DMO>94vVJDD
z&%we2OwWrxE=T_6bl<D)aaFwsTu<~ybl9fqpD|$mJUb}b9wacNRLg3xx&YOf|5obu
z(#(GKyJlkUcm*&S2O&4sSy(i3a*T`5metTANLA1eu@~(Vf6gZl%Q217xA|hb=2kRM
zZ2{cHr+>U`cN^Uvofnz19MDLNawtfVSiVMcek^ryKR(el_xI;FWCk>1;X5w1!?KvY
zCEVa{Iq9|v!`bjal^0wE<3pgXqeAZhi2y7S0@0;2_hUJmKg|nQaRlwR`{T0}S*!QV
z=XF<Ie~#9DUa9i9LiM&1Lne<B5sD$FG&t<jfI3D=;*JTIi5q$j09hl9%(@kfADKk&
zM^9XzQ_0X%^l@NGPrs)RFu;LT!t+_?7<<bq^Sv(zN`BwY{ca9w%e~mT`Cg+DdD&+-
z9-d}pt0EkkaTp(hZ+A@6cT|%7-Pi(CCDo))IkfMO+JbUP0tYm^9`)aA#u6|$So8Hs
z+K5>AV1fdmZm>v?rTn(3ybl@57o^7jhij5tnA$49=S$epr!Jv}k&0JIIM*kJTQLv5
znM!X>PMiUERYLdmaq6CzOSgAb%aJF4kqV%&Cw9m%Gu;E9yzjJ#Szm&t9b@*5avQpG
zU`G-~Vl25r0keN=X%T`mv);aHL=SCmtycF8U{X6a%855V60|=FbWZSe`7*xDe_o8o
zW&m7C>)1CCNc93{AcJEnO*~gJUV;0|>NT7(X--x>v393r4=puE2aCC|<W4SLhGmf^
zxSI(`sm>HoS4$8n!A{8bmQ)XTg~}+t_*fNincb!EkUQl2!c^J2mT^t0n78&>x#At2
zsUtB`jGUv3K%Ic|_Pu~m;L%>!R^yW(@<&K2@_y2%$UZG7aFT)cR_p>_Sv_vc0t?B0
z*D+lry`uQ*TXsQ;;tO8anm8EDl?fC<4Y!ki?l%3@?T5}(FRS8>uWN#rS5m9T<vM)Y
z_Ka_0!(fSytiDR&T7`BS#`|Nyn>L}n_4ONMV@*M2C3F>HlnjCU@;D(rg}nQh@yLX|
z+kWv~*R8wX7rHa|sT7zAyf<<>_EwHxCArQ*T#SFFEd%s`^-948?(yp@mwug3E>xQ&
zw=ZyruP)T=CK8gu!j*CYDXm!<$;bxiQwuUyHmKWeGS-d0)174D(`+lkk#{wnLoH3a
zO3P!@;lagwaa178;+npErW+P#i1WJ|AIzfO+_;XfRjWZ*x}u^v<IN@qq5V&^%AfBv
zR4gr!5W2%7+J_q@j7Q|VWAf*A{EXbbvbww?23kQ4NFOBp4_=<&SAYMEM|;<>n$ZEj
zi|_0&G33I=^Dv#hQuzr?efYouOGDW~C4Xfu@p*8^rk#FRWy+>+xj`)Y2?S2(v^a{k
zkYkycaQ*LQ`&I5^A6;zrZEU{nCv1A%$~5%y%hKRFxC!+b_Sg1>SDROpdWT2goZazK
zKP<JibiEQczMP;vc|^K3WEA@rfs+8{>>y^E&ue4HCuVnhU$#`4{1)Pac5b$KFc-DR
zGlWbB$IX$2RVAIjS&fm5m;!19h4>sd$UY;Mk~msq_Qf<k?FN%CdYimI@26dzo$W;o
zRx{NZwUj|Z`1&QXrcm*CW@hjj3}OZc8tg1QcJ_DPQf)S&K)Nd>TzUc@;hc+q$N-et
z;n8Ya-~*4~3Isg40{lme1Vo3cyo0iT6#$o2rw%G78`;%6nBF391MuAOIaU~Y%boDp
zyp3saW0I0$`q1_A{<6;@R_xqLsJBjW<w;4h*TvxhvY3KuCj`_SZCdwx%72}UCi{SY
z<lFN2L5x3vD{cc*a#yaf`Jac4V@&)rJl*wWf<YvsdhZF%A0A0NDsl;`sxAl?p6PE>
zI!C{#vM4JuhtC*KQ#3UJ<Z8mq(>lmDpU}dWa=|~s^pbJ6OZ_KPMhVo=I(3HH<j*z-
zZs=lcaSHYE$?0)@6<h;CNOBH$H*%uPNt2V$>mzzL)5`O?0F24oFq?XP+>_X4D)8Ap
zNv#(>Cp^6R8i7zmEQ6%QZ2GC~d#5#l%(_j>_QctF<YyC8-2?K4%O%F3eciWCYUdc$
zX;#sj-#Zdp<h3#%NKq+NmH}n40W*85T}xV10YOIJ^T59&07wu&(l;gBE1ACMm8?rk
zH-bWFoMkB&u{BVR=^wI&By;zRUevp5XOEih79w7ElBIHqT)o~t++IgI*j>aB&m;OL
z?;(LjA{~~@133)zH<(c(*Jq=@V_Hm&k~P?9S$Ilqf*PnP6oCm}4FY}$cr{$|%bL-(
z_kDQhDGnShF=@`x9I{jS0*;je4AinK-gkFSdTcoPJylku2dIrwv3k)C!q!BJ#+?Zz
zLC&bUpEumZuUm1TJrLAp+WiBraO5y}p`)|WH4)Q+th$n#dk=Q$Shr)=_Af0KUz2^D
zdpsLC1O2Z@4HcD~Fv<7IFWsa)eZvS6t}nDZB4-~5k4>5iH%$Yt`>b)o@HVaPW!m>Y
z;G91ltt7HPFTH(BC#<Ph%JW8LYJKQ@<t8b>G$o<)`S#ug(PZa=Id5)f5*<Nj4B6$X
z_+oO`vvS)`oF-+!)xtUIwe?Kv&L_LD(FJ#Smw18rfGb!Z1*UJ_LC8}-L&v)3)AIqA
zc7W*ZZjmF~v7KBZQYbAa{M!6w-wDn|MAPTgP?xhlOl(v47PdnNW{r0iY(4Yi6ai=6
zkf*cG-$OSqCe9N@gbbPVQ|<SYT@xJIA);+4yR}!?yd8RiEE$QxG(#2lO5)Y~9u%KH
zV$b}kRDZS~aLVGD5?OUjogY%T>n-5i6uIUmBFVjWie0<1rLA(QR6T73-+BhbIh=it
z`IB~S)+!9t{FyJw0*4m@amIexAP#(5yS=E(PN-cX-v5NeaT|y@+LNg2h*_P&FcjBf
z!SHD9YE%*R(Ug2D%kXChl6`c=Gpl2M`y@ZMknk8d2siNhy8u)+t7L1dE&?+{odRR+
z$@b;e+H0gPAuMIFWH#XaIFq)aGIP9uXHe|6kDz^<fK~VhVv)d~v%;3>$b-f;p+4<L
zD(y|zXv%TI2qe?(^TuQL00OALKnArH5cB5iw-k@qBieLobkE^@eLB{lR716K;}C@e
z?ni;oPa`}o+vE+{Zw)(u$s1P~4*7B8Em?24zXI|BRu_iJBE1GK_iq(kCh1QqqWG3o
z^7#4jEb|g9s$%_2hGgn9L|sp5L$3RyO#A6cX2yK0%p}=Mul-W96K?N#2eL`cW}f?I
zUmOWuH~;vYyH*^PZ-(y8bXd3<FbHHX%d%uDrlfuO5tgFb{YBN!W?$fUhhvy;9%A&5
za32!bCoCmg?m!P}2|sAsL0yX%9s=>Sb3X*gt<|q>PKltw>g;6VMpE!cWg~&2;E`1(
z<&^be9j7BgCOZ3w(bcz1Jb!C@FtN86<m+KD<56K~TWSBr@lZ3N^rH4fW(;$O2EEQ@
z+8Cih>iazYknyRX*8K`+Yu0(d#Bscd9j7;Dsn2_%$;D`H>#`r2<fTMSm%8J-d(Vc<
zu1pGt0qzktx3^HwRh2m=4~rKq4y#_jGKf)i+Qw7vZOP{`8(CXmuWC`~d*>!a@vfb&
zVvs{b&D6qprvpyS0EcsK@Ggf;`)YB3q_1dL>GxD|!Xke8qFxLoD7=(^wj#(L5^TBV
z%ZI}vt=yO<gfJc-0_D^DxIO#Rf@4WE<X(l$sNez}US;I^e*e3l64t^R3X__(geo=u
zEDF8+&s(ss+<rmqa?;(vJH$$F?8)f*@LsgwrLqZB7-$wK2xst-RSVJaYDq!mY3s}#
zIsQ%JJ@Kn((n_SNDuk-p6`O#}V}|TT`OM)x1i_$(PSkxekS#V&)*JK;%m^xj{XHX8
zq0U|O7<GdaZf){p(k+mriD1o7rk4f^6dDI|>>(q<h`L6_($q#EBD^I5=8@NGPoLv+
z9r`fh+G^LyuJ!}wHNPUg`syj5^_Sz{tevuWi`AtY-nIe@&+i|Xzb3kfju-kKw@0u!
zA%$t>iag7&V*=#xjF;{SgG8$%!8E_gkUuMHkjLQ#z?GUj0jznxiqz(JTc54z12AgT
z@13o92DX9XM$bdP=DXs^0W=uM4P?Vpnmcz~P6^2M`yNZ_fgHUg&}LB1Ck2uOpPP&8
z7xMHIXnzMV6IH&Y_H%lcs%;G)|NbB;rotiGTRVXd`V6!wgXdoWD1iM4GBu;8%wDvu
zB+IxTGhBK6ZwdU(UEJ>E)LF9DF?_6XZBNZJCNMC5$<39B@;8RBMx(3wqt3mnu0fPa
zzP)4UVM5dX)h%n03m-~KTMIVCowCNn4_7Qj)^*-=JtDfBN+d++oVNOYC@_X8JhJk(
zx$+F)uOG2QmR%^2{$ov>Ia4@a@Nx?+I407<jl2;Zw$>v52A~UJ$SK1id8(~*d5SMT
zJ2|F}Y@zM=K!^@H`{_HGzp3V}{&jO+G9g0$l@~WF;dW!j(atq#!=XXTgq#ww)-M7<
z*8KOIeEOI~b9mup!TFBR(9iy*o(5$`(j9#SO+|`8+I3J#opyeXz8b|}(@FZ7=KHg%
z*-s0ARXDxaru`g|>U5?@Xn2U0%9rsnbmW>yU;eNRY@)I`Da4)uLFkqf5^64kMrrpj
z-p<E-TXuG-FS2;DX8eSBD0j^LL>Q*wm46v7pK$H6Y|Zc+UikbW<>TNjI1n4?5Nna}
zeb)8Kew!bscevDFF-2lh9Tn!E{)k?o>YmM#71(3|@aV@rTr4OCPV}%aPc}cm9uZe6
zgp1r8L!D17&;y5l!M}2pk~U?F@im=C_j!y68H!T)$7Dk{FbFY{Ci)=si#{JOuAwv3
zpQQfai@r-na1-#gScr{g-crn}g(7Lvssk~OX*+t&%G<<0oyz}WZLjoJ?;Df@s(Mja
zD7Y>vt|P~+VbAwRGA~7+Y>;k#LbW{=Oeanq45-A#PotX-XpErADz{*Ah`Rl1Y;}JT
z961~-<}lHHzlY1~L!7<hf|LMF3WpaWLf=cn^??uwyyR>RSD7+bn*#<mD^MNkXo_IV
z9`qTQ{8mr8sBdhTj*$NTYq9svJhQ^0TitSr#FXOaigbkW^%{a7kx);XDEDuAJ;Ly_
zZ@}X)2O=@q?j~@HS^7{9%*K_z*f_DvhpWw?0mhfz_ETLZna7pess|6yA1-rrl+Tl3
zAt{$K(%)0miHHaV#^Ro&TjqIXEWRoCMVESStF<-jUR{#>?Seu!-emCxS4@6lX->kV
zt|@OxB=M8_vHIGM47{rzo?}6VRNYWsOPRop#x;s-SRAbg{T^MDR;y=H6o-(`uLxKH
zYj(Ptn1%3XlRj6@Cw5EAV#OrAUM|{k41Ad_ObA*rV}6N8%*mf<Ke;vzL#yih2J17)
zuM$osnkvL>dyyESv?+UBQZYmei|RYG#*DE`GG?WPfa1zz^tr$P>SA7Y87gR@aO*lh
z=HK~VG`E`R&>5O5Ci@9I&2F@7IF`DpE@<*EDPit~41{N7KvCX_=N!GU#U?R406TMq
zEy`<FA4ji7;R^8c8lY)$Fk}%bCZ!6&6;0&*lRaE}kBD~-A+|pW!J+W&VMNJAFC{Ij
zE7=2(UR;4oT8gfPgjFe{CSLa>r8BVb61g*qF^l9tn0PF51X^GQRy>aVt~LWaX-sQK
zt*k<0+oAFmK84aU3#NZSdGLQY4J&0;eRR{nOo~lE!FHoXnu(Z26TebK=eTMPhykQe
z{h1f;o45oRc?-9(Bu}Wq2yKw~us0~T)ZbHfJ1a_|q$2u(yxT9IDc#)P_p-huAd>!O
zTufHY#3}Llc~%+mOY_H)4em(>nA4Y?q4K0gY*7AJD>1^i)28UxpUyHhk9r%%Kx~*&
z!L8_#qx~DaB*&6zz_g?(7)_)wXJ8M(+XetVEFOu{l9E)B%6(Pl3_xK7N~}!V+<erD
z=8F_xH}n34Zny&EGe&VICCsW?$n9=;KnTZH22P2e6#imC;N%@>f5QAjFzD|VZ|i;s
zG|KI6gpFMWs6_i?J_i}mr5klA7w&EDlul)JgP^>D2JLl(TN<K0A*0j#2ZFy4o?T3A
zTDho>o-MQiAZdJ}fEV^c+s}tYyaj1ymULm-C5m82cUnCPc}vLn>+Jl_;p+VG(nM%}
z;UY%4r=DWzPB;oWYs6ZM#d|}u14p}HdsU3oQlgT7>UG;gKNJi=uLRM)UI85&GDjqH
z7&sY_^XEBc4QLxtX8s;>=-A@0>QOKyac|pwtIf93;*$C`3mCxj?`vjF<agf5FRWM0
zreMhzq!W>%K%uDgachRiQJ@s|=vl0|DIAC`1u9W()7m>qO|0VCK(VF=`}<jPD<ppp
z-}@LkLXOodJO%j*fw!FyPG62*h-(C;52fWp4P{Cvy0bgJ$wFNG?8rZN)5;r1o?bPl
z7!hx#Fl4)ytI&&U@-FGB0ocJfIjhs>$>QK{?jl6i1buR3BDfZn7xh=}a~zMGa%P?F
zu%dIKl>;yqjjGMqmGn#`rV-NT{FntWZbpy5DfD?yhvcupCew+d;eS4Ssr~~sJ=eP+
z93W8gUImU}$)$W&wuA3FHRO~mppw3MEG+buW<&X?p)Qria;~8Jgxqx=>$26B+mXa>
zk~MfL@ijAJ<VI5SFZVwurZ3Df@ZlYoNTw<;MGaCPy{G89^3km@sff<j)Tva?;VLGT
zI}^JU^7;8m<yt=_OOpD9S{|giG$KC0Ax`xq*>El}i@gOn(l~-~jZH<q);kl)PdhlX
z;!k8Svh6sc{)bBEXUmuE2JQ)RbpB$#^df-_a}{3Yglq;E^T8Rq0z0%&dJ{p}$d7kw
zQ9ADnA|!S8yLm%2rCam<M(SW@I(@@&c@#N){l@-uN?B!n6RQh6&qG0M;Qm+O0_WVu
zIrV5wGejJ|8H-N#6kCwK?Mk_bTf=5l+p}JJ48YFCxyVO8`_^yuNFxg_B-A`w?O)(`
zvTj7((uvF_LhtL{@IWg*jUHX2I5^#n#y+Q2#N6afCp3g}L;FjK%|CZ=m{nLMIPP1~
z{a)-xm#EUCWH`HbmSMq1Lq#sz2!zl9MICxGU<a}ercrUXfpTB_<#NL1oY?25n_s8-
z54XC1UUy+7Oqu5IF{e>w92o!H`R*J5wK7Ax0;OOZ+Br`e<Z>Egt5!kn7+)AmDoulq
zs2>XV0~f8fcytID*oZ!%$raJaJm#(KsOEb`1w?Vj>r7`lWCoo?|Ma}R?6?t!o2C%$
zY#VD5X>?a{PUlC=*UM7p)MGd_EdRZVIbxm9I8jH^SGquYK4o^<-s+m<Q2C-I-gWX$
z{Bl_&k<&=#RrhUTN$!Ga9yt477}%WwpxOj={!wuaj(!ra?7Dl@ow)O_i}``a008`m
z0@13sBo2|d-%TZX5()}wZX`Pi)NC^{HNW)<^VTY1i_oa${ta~?<9&k^nfq8L)G{@}
z8o#qQXCi9CmEQG<mWkE?VbtFscN0rMeE8Yl=Q5MCBMs8!k4A|@*+)q}xNsz#k_Z0R
zQ8D|#C*TP$u9{CppDJ%(z2rS^MmQZ*nEYKv>}hZOi~*vxxy-K6TmXezAi@Tkme>l!
zYlLZxt>O<_YuL_9+B)~GuB!LMjA2p3g)KTF_X>d~pCtbTobg+<l-!9mu0H_98y<;I
zAD6GMB$}h=p9#`sWC{tATj`|OS^lXh&dgX4jz;O6j!PtVYz!nCn3SCWb`v}Kig~)B
zUbQ$rdz#Xj&D4_r2oOPMI>2~q+eD<Vn(ED?vNX%0JkL5+UT!P*o3GJqf?s_k(DZBH
zLs%A)hYCR^(VM!zF(n96Jia>v)tnjT24@l;`?nGUOjSWF3E~aePC*@!y@=`?*47+H
zLDCJK|5{C)496Lh-6*LS+-7E6uLK85r1h!TXega#1$@pIC2>rSfFH%eb<|gLKz}p$
zG9U%*?b{!3_9AS*3HvGiR#ILntdDX*3OTmX7R^$G?l9&%TGyqbWtH^{Zp%1YC=C8!
z2umW;YNV?tg?(Y5KYsIssebLWg*fNqI!BP*3|p+QYN)_EF+j?E*+TH{=J@_Q%uZ60
z{)AGvG@4PIJFb+B*f^(~IJbraV?@A#3G@^8x|CjFv2eQw_vBf_-@b-<NnsQ@#~VA}
z_zRuS^M~sDW$%$JD3r{V2vkTJ>(kz0f`^v=aQA|3zdHRvYdz+z!AbY@Q)8%zA_yR*
zk@vi{Y_R-KJ-dZBlGL81Qz!I(4H`EbQ#2$qf>ST4KwjA(sqaWX@dkSzRWO~PKR8;f
z@$?|8e1?ZRWNf~cJpn2dz%c?g)sGZl;h;_=UJ;sJ8s`zy<!&3+B9F2=$r%Ro6@&je
z%04EC%x<MVx)jVa!YIJfptt`~*)GnsVjL_+#?!Qp2-%(?+^?0$uj#LC+3JQR`g}{4
z{f4Ad;89w3IYWnUg;`AFZwxKZ<(VnI_v=m=g7RmJK5EdMdk{67b5#4K>oellGNE>b
z@g!bT=ox^wDbim9KT(bGymrZ;j|I7=dCUTBUnG<!p<d|*?}pRm%7?-4QQ-0A5-~0s
z(wq51Z-J29AhaKi$oBu~;OUv2C4|FqUuUEMx2_f!7uz~G7}VY(h@sAB6p5l{oS}ZA
zVq1)D=-Gd4Wm<`FiypzUHOM1&$Br_Qbaee+i&D-nxMCg7GcheKElf;IQ?PB77wrGw
zT|rdI$<PM*6@-R#4^t^v9ELT49IZ@?3IC@XYJ3_NIbf|%|4=GQ`QPB5ic&!6|DBWm
g#}}5OFRwVeqf3Ww;M+M^zZaa6oSJN<v}wry0jONu7ytkO

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_005.png b/0.16/_images/sphx_glr_plot_visualization_utils_005.png
new file mode 100644
index 0000000000000000000000000000000000000000..1ea22cebbb6a3f06d46dbf9a020e58d708625658
GIT binary patch
literal 9191
zcmeHtXIzub({2LNJA@_#B!E<DN=IsFD!uoP^rE3xr3H{6MNp&(Qk5!5lNumMQKBM9
z??h@4c!JdT=J}uBdEam6obTsD+MU_i-PvodotfQaLw#);%IlOM5Qs)s2W|`k5fKC5
zkH|@Z_p$VU!hsKYgr+6J#NQne;`GQ3r00YP@bO3Zcskz*c6;>L)8Fs5n5@`s(HkBJ
zM8IPOadF@OEg<Iq=%M%*RSg`Vgd#x4>M;mJ=XCV~SE^Kaf<U~cx^Okq(EOdbut><-
z5-I*6YHZhCmR0@+55+ac-P268aFTIz36Zb|Ny<P<Y>Dr_YK-ttFDmAJt#>+V5QveG
zu;2|6W{CW?HXcrjC32ql*<UWvG1m;^RT&#kcSlbnRk)S=cDKI}-^_Zr5Vipwkf0`p
z!{L0l%n%?L491x17K=pcJQBTnnCZfP0|v|F904&fWUN|}B9TZwjFc)MEuRAhgg^hD
z_%*ag;1v)cV_;w~gz%5QLgFl_(W*{PJeir9#ntP-@k+t;$!2pd+RekN`ueoS#>ShX
zoD38C=H@J!Sy`HxN*OLiq&j*41)UHk29$teM*^XUsm+`03}&xcxgyiPeu)&%BT0wC
zoy`bSMm16a;h_deH0^!<9wo(!L1YXcXqpeE)c7o=)1m9~e|3OSgM0ty5F4oJ85nd?
z|1Y!p|4shyEoOIx8Yz_lKTMalqMH9>Mv$SsyT<10>xXqei4%F`?D+5Uv(LZx!Ahz#
zUctz&ar;!!<-FiKeQo0T+0>dQTfGw`bC9L)b(Eo{%q=mN4u^Y#MKj_9tKwwE1BXh-
zA9omY1R(o&z(8BZTgqG^v^J6E08Cux(rD~9`lMT^!%HDfzbNOE`@V==#$mB!8DS%c
zS;9hs93F0<IuY+msd)uvgzh81k)ElSP&4Jr=q$H@E)&YmKrea|aP>%n%k9#J#+|bd
zT-`T6aW0@vSnv7cIDaU;d3J>wFN29<`xj&#&K-;H2d~8v9tZ0hYLiovHVu>n-+m*s
zaZ#Cy?K29th#qf@SGaPEYUzRUmGs{W=H4@Z^>Uq0iD|%^6am7j{@G6t_iZwY+p1^F
zeYbfkYpaWLDXwfnGBz1^k+9r5N7PkRK;UyRFMYu}yPmLGhHvq`ju&<Bxx%wrR%M6L
z-u$`YVUOKtd9xoXhC1&tcE|LBTq*975e_%?);e{^ZH~Ujdb`V>lb>=^jpAaRsWd(x
ztg&q?hhcqvQxeg4XThJHWUkC%>ixH77U$lLU+r~BH7*EU{zxd6^|Y<9g4nSAqcq2!
zYCTr-^B);a4c*C1g)v9-t2PisOY1@XR&BpRl)+bK-~9<+lSOmrMD*A>G)K+A$FM?v
z6nSN_r}`$QN_UUv6l@|?Q&Tm&H@dYN?LRnOdfU!8w&n+A_Ksdri4O8f?aEY82wEYm
zZku6jKX<Cg>#SYSOiiug<Vq&gpNI(y<IM;$@m>u+C<+tx>PS5K-kr2`MEO7q-Z;Et
zb?TC?hsAF%%D>zX_YWmNE00b+8jd7<Rx?w@A`!}gc~nUepx=9w;e?zXDMv<YpSzr|
z*pt?v6|j5fVLDrtb!?rbH*c){+5>ePII6=kvCU9fhi_Z{j|zHtW4w?}$<5-#zhXR|
zs9;1%$U3~!FCX1(I4CoHzjbzDr3nHt-eGl5&m%31qEh>O1ruP>5eJ+&&onQ>$#Ez(
zN89SB`?MxU`wn0E6yBOf(d~ZuM`D|8i+Y%`s2UBPSW`H--!MbmWdLjZ*t?sBjFRm(
z&!R}}sBlv6ybIRhXUzdKjhkd*Rs6O{TEZ&!I1Ck<DSe*%`C>O;Q$DloJ{qkXa`W?G
z*CXn1fKlqE+@jWRF43R&2GE0y+JmS4<Qiw<DE?eJv*~)-=)x}XB<1El8~dxEUf*ze
z)9vsZ)Lx;Xs#2WL?@YuG#W0VPw<PbtE=TJKuSGSMw@b&DSBM({y>^8n)6T}{n{#g~
z72Kt*lc(e{@65;{cG~N6dU&nj6or{?U^PWuBZ_1jR(=HsT@KF8IGsfuNGhBxh80B%
zg~<bMS9yBQXryzHP$g}>Sn;nK6xBbALs56%3=EjJ&0&q~8aC5S98H?;&sQ+mfGcw^
zl-&oV9EnbnrF<&!4hY59B~I(HzXdzG{C=L9BcP`ONBws{urAVNXuq2kv!&5+%1Yhf
zg;k>}m}Fe$G{jdL-I5IDniL^+DqFeJwIqIAA2<V9Tki(yJ=l}yuzp}#@F~}7|C>4!
z{<4)O>qFhZS^0hO_8n&0#eLTKzVGvj3?D;%YY0_11*xLVlxB31;s#~nfh3~z=P3;K
zc-Y(KVC&@FePF#Ax|v%pR>(vno6fDiHityVd-siO1vs2Zprstu&hwQYgfFr+H6JaG
zDB7<~ihPuC)!%&9jIJ<$M-Ec8oemTXR($rS{gjlyWSnJ!5YR)wN{MpbatR@cRB2rC
zB+8m<s&?O4?e;q0IV5osY;TT8yUDyslU2XrFD*g@dfp<EqZx}FdTUAg02Y=&_G`Yx
z%fC)Yn8jUmFth5r<E0XjW{WN90E_t%6@;Yl!fxDo$;k)$&!e;y)wZQ4D}vqGNaScr
z1;vBDj|k%V;k`tGW${n*J#}qnWXfMGpO$Ql;HS{Jr9U?Ms}rD0UldL`W7J6NosS`W
zn|>ha*JiI*ZfTd&(Z;3nFn!iEld~8Swj>?x!>^prLA`BB^&h{ZeQ^Km%^x}!bBkuP
zn)!&I$$mYbH`A@@GyYcjS}(4x-v<BKDNBI8#R(W;KT0++OBcDgF>}sHQ=wyhHF0()
z%aW1D0p|Y4DATClJPcOpOXJufm<5CMhqmW)n6t8V<9h0sI_J%rQ*6#zhZnCVuYca#
zW`V{bahbD{0e=%V%&a&^Vv{Rh2d|8|SRhe>gIop-%CCMVl@h@jg}qjv)+8RJs=u-k
zVS=g&0^YWto9%e?y1;NzoEb`@cz1y9&*8NnuZ<?oh+&P5=s02e+Tl~(X)nOs<rk+j
z`?xpr>bH-nRcd1hb;zj4Y2%!5zQn>=dU4x-3UO?vNr2p7Zc+MI^AP$xr^Vf^=gL8{
zj+QQe=ouzP8<eYXp1C3Goy-l?FAWD3`G*KE$bW@A>4sqJeW@M%YM-AoWrSj8K$6yu
ze!2S2z(k*aj8fShYu6v@(%x>PFiMr7`X$SdT+wdZ`f-7-#N|g@S=%kJ#ykfLpz-<M
zs_uLH@9INcE${t^VgehUM*-^Yh04{Y&hsf~6)0?qXkENZa}4b)DJ2K6YMjOru|)MK
z^@vUR&V}vnh(FRb#MMyYM6RZJ8CMYDr?m?6X3o!ZW|t7!_)_}P<mC+oQU=|@6=O#W
zAE4_hUykd&4ls8v^GZiRWUq_(coTjfD~Z&(Sxq`Qp$)Gn1_6IGNL!_8YTtk*3syf>
z6XMXs{QFdLa(Q~rt^T8_;E?Yg^`>&NAfl4p(_>E#=z|0Vtlz+VA@h%O&lF0BA||yt
z0fq$B8L9!ty~_ZNBZ_tF>Cdz!QCrv;dL|9<-L<H0IN$tI*B|@vhe4vmro6nR(dxRV
z5MV5VLrpK)sP~3BvaFa#Vy%MwlQ4`bbSHi;O##~-MieK}$vW>FYxVX+^QA4g3-B|S
zL~}FAC?3d2tZo<#J#M)%y$FX~t4`d`;POb3j&Xg4QbuyPNAd^p)URJ!QC8Oa5>(TW
z^Z@*x_T%PBQ^UK26|_{-(!TWdu#W`b#P7bD>Vsn_fDj=|Sa&>O=k%+tgKAKq=~Eq+
zj>KQs(RD^B@1mAAcYifmp~m`x!hH*J071jIS6khZwvmDekH7}DsJ|8^F4y$3H~Xg*
z%V7sqUknj}-0e@(RP-L)GoisUAa#h`px0MER?+}sW>;cPCK-QCEh?8DJ$u?nZJ728
z!o7$<qSWBva-0GxO+>AlYj#3kL+wg*$pq`zMp|Qb8yiRj#LHgqHD02KP3*oEnDA!m
zY3}{=a$#Q&D;b{*L<@>R%+}o>sI0fvOIEHyMnDrU%PTFXe$#Mr%v&KKFiMD~+I%ac
z&`nzV(QHzV=NCl}JR~JcO0Q7^qWAK~(!$aW)Ydzo#RET?ph?Khmn{YYk+b^?B`)#L
z9$tUnw@YSY8Eni02wR)H=I99j3g3-cE*`+7qBP(^niBz$k&@qh{){nKrq~2I?$lQa
zwl=5Y1cA8U(Kvzs&Fv@tdYCpjDPGd>tih5}dTg1oG-o7UO7!<(4vvpU;9gLP95zoy
zrsi);rfQ$4KxCo=f*SPVF?rYC9o7)>zeN0TVpqfR+A?K0el+UJ=aDZGpnNyc2L#!g
zD@t(ddCS@6Hnnn#8M;EPXA~Is?%L(?q_-gZkh-d?)?ak{(uibMhTu|J`~<TvRl$m%
z9pE<aq+jEI^e!ap2cqn0nN12KnjgihPQWodC0^gGrO{@D7&I(^!M<BM#>e`=pFer*
zz&aTfU_S@FUT#p3Xm?vd^<B{o1zHBX_@C|a3ZuVJ4>$c8q5N;dsJXjdPC_N#w-AW}
zuG026QgHu@>FjIJ`BQ;M9Ie5BX<b#mF@-x5V>6IAr5`f4#y!87vpn&#wB(G-ZV5?7
zG0=+=-(>mw+x=5iV-=3gtWoItZrZQ(qZNDAI{pZw{TZ>GPOaRbXID)F|1no8s0WtV
zB-C|I_4wdG6GKi8Fz0PyVLT9v6bc{ncQ~W_TcIxr>r`?Vy4RJ?J1if`h-JJIY)2K)
zDaykP^)xS}S}hXM_;5{aihtXFvfs9XwDS^o=M=Z$qcC(5TVtLcEwn??Zv{3W^c2O3
zNiJU(I+ii>b;&5_>shFORO8xafR5R6z?AJk8X!a?<M_QbH*n_<RIsy(1JYrK5QS}-
zNDD%3cs<)vxT95={iG2r;#6+@7nG5IgXxY%!&5+@P@>#;EQFabxQ$5@*B^K>GFUp1
zrM)y)#IL}TQC2C&nmza#-_}=uP$&`k<J1gMKLeK2)V`gz6j7(~ZPpH-Ov|V$O3Y_l
zy*v2j!r#%HkoZ!5U#peHH+G3)>>VFc{S^>-C>`}J1$oKIjS@c$mH)N{MISb?8-n{i
z#Xy`1?~>m{oR%EY3PiU3yeRIIQswC__c(U>;Y1zwn*`B%7X8*>7F7Dl79m<PUhXxX
zC)<lH4Ag0LQQ`8xLsnpL{=P-1+x9T^KGo#lcxRN6+#Sw&QEygotXr)5Gr%uao_n&*
ztz2I{7oxc+davk|RQd5xOGqwMgEEtJ2OT9u6{vUodwJAMWosCJZc&U6&V^_KEL}{$
zY7Pyp*r$Au_np!^u{kb=i3*&t8KA1G-(Q$PgKx2!AE?<S!e)qi+h|e*cTQ1eoK7N9
zjCVwdA&rH|g%sTnPs~<2G%mxl-0JrD+v)GN@ubWMyosC*9QBar_$_l<rw=x?6P8tK
zjT&~ykO*nU391bMlix``t{Eykx_i=nal^RqoEC2rrYw&A`mAs#V-Azh(0wVc1&<Eh
zl<En<Hi+&eJ)%jzpWgSd_ms|-d?faj5am5Z%Y)#&?(Fv+4UY~Xv&XNUkX$<nAq!vt
z_(bXr3B5D#XlUnbP?~o3YI!?{&}(g6s`Y0<m|Mu%1{=h);*}B&cXze`zw8z3CAqN#
z|LI$*eY24sW8d5IC9P6enBvd`0{nDO1pjR`x1kJ^?*IXp^sEfUe*RtdqGcj`jdOXX
z>RtWvZDzU$(6o393!6XYC46*1#!9v~)KjeTz9)W)pWFi7V*=(Q(k6W%=6S*OUX+Rn
z%3B$(c>eNo(Ur5af=ruyDBqMu8*>DoER(l9&-Dz2MQ>HXR7;p^E5#L_Ub*VV$OvXC
zs3jG}fY9^x6Akr{dZZn~3QefgI$TYZlj|l?jtG#tIpFU^K`yz^v1WUo?bfqOZ9u~5
zpk3TAzQ7s)@NYWztEaXodPAfM0@^iS@chWbfcn^(H+oZ<ELmaWnjvm6A57)ZD#0<s
z)$__t$1StEV}A*4BB2vTPf-_03r=7eU?JUG^9=}iMn=`%^(@CJpHB5nGA}X9;@tpx
z^kg~&s`2v9YmNi%yk|`4(BW8h^s9{6Hulh{_0*}K-ZH76ggZH45&Y@mnG)k#qBerj
zDQ|if7r93Yq)%{dp3k0Rc`$GL$D=QgU>^@(#+849kDLB8XWVx9!3W4TU`wxEj%#Oc
zDkK`#$=Jrl6Hprd`N9g6x;p-b+`pJ2mP5E>(r?{ezwzQR(F4zo<~sCI83%4&ACX<`
z%wA8v@X;>Rb>jxGmqh>@8D;ratT3jXimwpJmM@T8Z1raXKI{bHL9LO%T*Y?)9j?pT
zrPO|xRZT@PLIpg4{`+6Py0K<P()~~1)~})PRVhJ*4UDnl?=<6p#ZLE^gdY$fHV{;x
zop_TRxK+RNYW<sSRlk%YPX@Pd5598}2r6vz*ZYRCCIO2}exlbP5^lupPW`ze4Q&`v
zF{qE)k{8>$-6>_Zm@izLf-(n8B{@9OKq#OFYAEVg4Ld3pdo<<%t8Uu<iuHSWXGd>#
zQ!KU|FX3f2w=J{HDS)vRjb^ra5GPLO<RmHXL=?@%GZcL+g*6&dA0|bh7wkicI}4(*
zMuOy82utEyV&;fvu`cZOY(RP8c;qPqo?Xch{35~ePr>bCX>Y);R;oZhp^?)zt?FGO
zU*Aeg`rQ_uw!F-A%K^WWq`nhzPOp4xNq+QuG^>4LK?tx-tKGB+j*?Y6Ih(<4944dG
z)z$-TVs;ruVoAbo6PA7nB)Zti*NMvQByc=b_bwDuAbpk1Vnf#*wH|}x7TfEyZv?DP
zO9#Z|yo-Rjh5yV+x1PS*$FrS5h)q$nJ~4chUUC~IhGS69c_O+2KPG70wT77*2O=v{
zQBkLP7Q~sK+y-5K<I1M&bNPl&9Gk0@b)-__Guy#;q(k41Cw!f{vIH{1AIeCrr!NT+
zo6-{W+TFwsvt0oT!5fp@0!(8?)7^TOBLyg*OGeWS-VNub0>}35x=fu{0HNd%-y!Ld
zWN1Gnic+Vxb?78F9WP?bn6X#eJB8Ur5B_<CHHyF5ML9kpACpyDH~RUfB&|e4fI<FY
zQ-1s9<3*pFC^2Jl-Um{`1J3kY%w<3j&ip3l-iLX)i9)~*aoX==?J`_Q6sq&g{m#QI
zM2Rz#BmHSe&-~y2$f6``r7d<;wZTdX_P=N`K;v<HnwT5CT|ystL|ZKXUC8H#sVaMD
z8-cX~IBRJ_sa%G`3Brl2uIYdj9&bT+^u&LJ7Ar}F(uTADrNL6fZ%Mf_8892Z04hJu
z?NA`976dTPhbW18d?qPngP?x1ThmRA6UNl&R;_1Ca12UY@hFfc#6mHQaG(yB1EyfU
zV^r2{S{*lsLU+R8m}0}S;q!AfmB(Xgj8GoIV`lL;xH;wbLMu`OEzfLeYq~gVioDkQ
zV}G{J^Am9F1+dRnW`LG9m1A0_3k6coRBb2Cs!t9?OwO|CnL9=ZYD%>=Bgzuf3LG1s
zWpX!>bpoev&k4cp?E$3&9-0^R*6~qlFPowS7r&k`-m*>;6l>W@F58xJO@LYndcE<q
z{rR6793LE#QeNZ%CS;C1UsCtU0l#7=jhqSQ37Zj9osdv@F9}jjt4XS;^SpNE+!*Eg
z^l5b5&%WAlNy!V2evfsBUZA>jJZ;^5NN=>0eT-fd&vKbC&jDAj5_)62xSm`((mF2;
z%RkJ5eg&n}x&$noE`G2mv>)o4h#O1}W6N#=^YRivpzRc~^J|(?^72}VzhNvJO;1Or
zL=?0P>w&<W&0gHc;Vkw$Ra5z8%b)I|vL;MlViX9}krb~{<3I{T5A$0R#7il3pWB&a
z22QpjvZVQ<^AZd$H+%N_MgYouLaE{p%FPh^kG=xDyBo39iw||*Cry=sP@n5TCrx6x
z^Fj@cAy>%mtCk-!NqEbw;WkRUZO!~9O`hHuC2oO8F_0*6Bd{}%-(O>gIkOkXN`A5r
zy-fgvj4z@uf@>5wL%#&^gKiK8*9B8e7I&<X9X2)UDLZnc9R{Q#VhmXJ`}ax=FH7g1
zufYHdGY&hLlV+PmpH$1b1g*tYCk|K>HNE}#GSW@z6YY&P!YgDmtA-i%nZq*U9VO`}
z`^M`xvpKlnF`MnH-NZn)&n<idduzdpRPgXPO5#j8l<<%z+?)ked`Po+$S6@7o)(9(
z`&(x7W1Mtn-Bxh;D>}?G1G8^k`t?ireHzNyYT~s-GPgul(D>0-2llYE=xwVH5{Xl{
zBppIz`^&(#f9!36d!BtL`kIdHd46d0@UHof{<OHs2!^)LsD5S=P;7Yzq0hmR4gbki
zIR#DodYSn@t}*oJEFqyM@z%q;>(ih+r}cjb6&U)}SJ&avFCttX&*H3!^4t6pho(a%
z<ruJa5Pp$*5m6#OI-oRM;$SR2EAiahG|Tp?0vi41tfe3zW_z;4g`R{Ns03q$Qfkq+
zmFbe(Z4i(WV6VCXQz+ZsL><390t)EJr&p&Kxa>Hz67qX1J@e-z;3s+Pyc~1P-k%H{
z8(cl001T6YYEHh<g6E&MqEY$1dW%8%mqDK$ILU&!W4qt3zQ>p)5i1r}B>}ZpP-uu=
zCLPf2t3(#oR1J;SrN@6@pQhaA<-g8A#A^dgO~X32_b}lS(i<&Pik)|tyX3E3X4oF5
zrW#Gx-O6#eQ1YP^pv}w&rZNO+nhLM%t|ZzMFi5beSuTxP%P~{_KIVs>ZCV?Oac4lC
zp0j<35d4SJFH!loO~$YYxsEeXQg)@BUl$5h+GlQ^xPSQuTAIMCMV!-Qd#&kxTmbwg
zh9Vx2I3<ne9UC;*gFpGV4t^-|{X=cMkr9uA)_6TNy%Ej?5^5KX71DvdS}5ytFu(b~
zo)4MK`6>On)Bx-zdUd9yUYMDQEO*id$Wk}^SY%P@Jrc!W3XB2rr?D<sL!6Ax0>T;5
zsmvFX&ky*}en%=&J?}afic3Ql$d_F`uOz>rXgt$dIcfnaANmGNThP_B=Jm>F?S(|U
z24I)_k5Hph;8N?bm7oG?#u{*vWHj_`a!>0W-R^*_o2^UiQ|$yW`sSq9`IWv#07yFB
zF*ICx0d(fWA1<gUa{NBHEIfx}C7-7JJLWC)9)1X28aQOb2ZT30<sLdaE!@>B>LHjW
zmbO7P<WLig)wfR%MS(we7JGxKDgN&J8mkPWB4D<ea?;*!>>P%*{!>e4bXn#Pq+0-4
z!=y-JbSX+6)+lf6=d;H0?z11`8ZYFgb+t#{1uxY!8PJy|Q0eHJXH<4Hc=WkaOK`#a
zMt+K4Q_@p~mv4-WCAeS-I#HW8*@q%MdW6|@D}fa_29|aI=iKpX@8_X2Z8{)=0}Zs_
zhkT%>&nROnq?BWcO8-?_i%~g{GV}|DYr~lCPs~jzZ!^yx4M?q~<7lN}>WH_ojsV#2
z8+XqmFOT%qCaYE?_jMGv4<SV%9fZ^piCf*0<!iE9(;KDZ!P9KBT`hFL&%yRbVY?3v
z%)UdI0lKeW_5a%Yw~9DHqeJ7q0Tc9{zEGkQ8|w^i^>Y?7B}LtTmb^gZ2?I&NB;Juk
zSxrq14CCe1#8iDWln)kcr~S&jWAsB}dmBST+#u4psF+CwlK8~cBT<Y~D6YhZcqTtR
z%p1EqwB@)c+xZa$;_|@rThH_r3r(EP9z`2LxNHIxoj-yE@`&%&olwYPY6hggWO{Xz
zk=VB_^_DDKDxK}`W$X_+9S83*L#ysh+zkf6`wA#;$qU&{7r*afl1C3`*k7`_m&brO
zAn<BhkCYU~7?NQc@tj<l9yIqk7yG3$fh_f>6tSC-(Nx5Tr-@Zz$04a9<Ep2B|2m`7
zGdlb<oZ32vIomfeTesG4Sdo>OJ}kayH$JHsWHcCUCI0-iF3!kaOyP?>fLzgdx1KXd
z2lQU*O$h{Zhktsk_+V;loiEbH0d(X_@#)oR_5MxDCi5}Jsc0<eSu~qX2#~RhpS1rW
z^>d{)%gD}5lC>UH`?Rk8bRD4U*W0MSDU!mE=}`|1G!Fa5@T^6b+1ca(##w;8Hb^%}
zu6Pi$1W$Ke4K`e?+5?|a`n0Jy_$uS4CypPUr!A}JjT<>gYPr#B?H&#y?VCt0)b|M!
zmAegbTfp(H?jW#{#F)#>TKx*>rc=Eu{cVv@@XP%m<{S3#Qo-3^p=BWc&YNpv2?^|}
zMiONZ-n}+(lD&7@OKwzOi--xLDRuc71P1%{H|ilX7=+)MN?SK5=#NK3NfM;$1SgV7
zemgj?b=&vuDO;tM*mPLv-=lE?h7L#AcNIj`uc4?Bwu%sv3uA2H15CLG8%IIH*K+0b
z(Ta9;8oD_%Nh^TLEAMsHC>mX^Kd@Ea5x4EH$|*4w_(@xovRVh>?EXO3gcHPOeEIiW
zn$<N_`1xU-|9&@60+tC=nvM=pbT4?OA+raw@-?bvy43$mdcP<yrGG$q^<1*)9=Pu2
z4Joj_p6VMoHb~IPN$~e1seG<?>+enJ)|*~TC-PCOF_K`=y)$OEn0mUL8s9jUx`J5$
z5mfAUQZ5Ul75NA#p~+;6jdRC{PO`;qrEx0ixcm61{@^mKSyW^qfPe|%iCRunaA-vW
z;tWOsY^0Gk!8d}bVr8Ek1yZ!u&UtJ@X>qi4Tw8ed=#^PA5ONaHYL+zK`lh_-dy<}p
zJ)C2v(?iCo^!W?we@&tP7qzD5P?+>=?L|ysZY2A<LLCUi`P+v3Z;de7TbkG0Stnwu
z)jM(}6UK;oi_W)=6AAB}jy=ckw}wx22E^d)#tC1qPCUVDW6GiHguo6d214q_`d|~u
z9cC@`HCH>1xNkCL$B!$@TBn<#V?K{=z2K}(?A@ZjT~`E<CNh41ZNuVX-IDA21-a@3
z-ob{E<#01yqjMV<fmTrZ<tV?riRYr!jv5I46QFWTy_UbB&`#%wIP5+6B#fsGOI=io
zeiHHE8vWJLYyxdJomLy&$)6w{E86Xoj>I<>vs6W;k3O$P$#Ur2Q;wd`)rx;nYPD+g
zUNPs<{05!Bf$F9+_o1=)j~5(I4n~W<>ieXI?bN+90vZVG+*3@H7)xRwTQ4pw@2#2g
zj-2^zMFL$qUe>lnn8$zsN?<U;{W&eJu}uF?@bvk3Of)mySFb2VwEO8`f$9d7wOc!u
zFzyr&Rj@5ZK9s7uKG@Vrm00V%*P|24k-PL_cdsO)Fdbsox!A8;w1gR@tJA99n7O{w
zQG2C2gCC&d<3C?iyyzj(+-E9&#GpI&7JtX@f3^U(Cu&UIfje1Gh)>*Se%twpxWwG+
zijrZJ&%b90tW<ETo@st$ZgXYI<-PgeS3;_T@)i$M2JAy`xaRR@tbVHeWMV6APDdy7
zKW=>hcR@gl-jTt{4M?0ZgoOnrER5Ji1WN)U35lst#p<9^SocDDem*B~cSF_ai{0BN
zNTLx5>MKzpA&P3R!VeMutqxcd2CLjKWoO`4hmq>ltp3lJI|!F2S#{Gx%wH3LOE(~0
MO?`N+y5rOT0^soUEdT%j

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_006.png b/0.16/_images/sphx_glr_plot_visualization_utils_006.png
new file mode 100644
index 0000000000000000000000000000000000000000..6c30c800e81f11f4a507d3b369959337f8c21263
GIT binary patch
literal 192949
zcmYgXWmFtNm&IKM3xm7s;1*=?32p%bgS!R|?ixIhK!6|v!QFxm4#6$B6I_Ek%YM6O
z&;ICs-KVQwb)D*db?e<*F&b)$xLD*^2nYzc%1Uxt2na~Xuh&-?Xs_qd#B1MIBkCcq
z@1gBt?cr_aW`&??=Hcq-;^Am#PU~gm=5FWW%+D>%&Cf||>*3+*F2=*-^nV4oUEFMV
z#-wHUUY%gNDjB#VAmE$*_kma}Rb+>Nh=8ChC!_Nz>*z`Oi`7EPF7}ho`??#+fswWt
zzb)(c%5UXN&v#t9!__;J=(N$|?nM02nEgo<XwQ;-iPKi%U4`r1uc6s|6Ag)qUpvb%
z!c&KmE=e{Cn?8Dr+Xy?9Jd2a?^LJF89zNW6{ErcCy@X$*-~R92e+>NO9*W~{vNm^f
z6%g=h;Qy}Lf&Z^W(AJ?J=<Gj_{!`+mf9}S7{r^06JjYj^_I3Vm<yZ4WVY2o)<o~zX
z(Lsv*f9`LtT>Se?E;9Rpj{`5aG(LfW3UbedQq%wAdx4K1YGQI%2EXUf36c_gylLNh
zZRz>2(xlb<vDAEB>UrR05dPr(a`NIrbMl{;cv+hby2XD9f!}4q@2Z|afmin(-v8Yn
zc*2#*qIjS#=<M+kr!(jV{%CS`cflWYyT$T+StO-pxb@WW1^jnd-Q>0U?#qvx9O&KS
z9L?R*%fd@Ie$eA!ZRMpTFyJh(_20C5)#J3(<H)z1HP%AH;11GmUT(XMd$X<cJ0#Hw
zUzSLvSTY$_C)ZSPSv(-kQ<c*JEZgarBy0KMy<7x<{@s`@#*7~8apy>9_$+z0T5K+R
zFqqWdj&gvCBAdz4h=(TTUAV{|EWb154WP>-SRhi@&>@aHj*d5VwM(IAFB(GznN9Ay
zr+vxgA{mUqJWg}ctgQg|kMta>GWO3{1$ot9G!a{Yw9Ey+0KxZxyU$**n^5kMv-*^8
zxR@cc=;X*bk`fn>m)+NoF;dTG!bWrW_0V5bCU-?ax9T$w*|0UB4$ej~75~Qf_ilPq
zLclGBXa$YDtNx{>v85&P?Y&|&phAVlwb}$uK2YJ1s!RWCyr5Gr0BB9w!0QOGFgKo4
zEF{-Ku$VW6uif75z7Kr9i`T-Y`N30=-h7IF^N>J~LUOQ|It$R{UP^;MAHug=oaT8U
zt=a?G>0-<gLgb>1T0@);KkLP%);i9obKtIR`1LHa+uhqO1lrx?%ZU)Gu*4kx9LXvP
zsR~X|vb%!(!n?u-HJ&g-is6)Iqo+GrG}eOLbR7G#dXRLS7Dv<5Sl}7~%Aa1_w2)*s
zYhyzCuy)GjHFT*LpWULMPunlILAO$GSNqn6#iUMO?t?DiV7J^w1kDyq=8!DXA3r2{
zRqoF3BX~@Lb^_Me8YWxeEEYIH-ehtFvrKiB*dV#DZRR3KA5*hoLj8NRk+?95OY2KX
zfHask=n;7?F(@}$qkkQWw;0HB6()iuCN_!2s}Z4#{28o*t*)+KP|sUF34t}L9fk?T
zP0FA)I++bquql8siWS^ilBU3-CvAaGF^gVc45OxaAPVLXAp?O@zMQTarM#{tK|k<v
zasR@PEH)1ZbH<sn;k;vQ;YPe^DI2D<dJ^%WWH}+s5TBPWQb{R<7)exqsy0arEU)kO
zcmFOXagGH%q)!t!yH{au!K+Zz6nYhTh1PBXoAw(R0bnYrM6-@Oc&)Zh#`|#`-GKA>
zzfbxy*sa?c7&PK)%7ljC^cQ=qvy#+x@ouFs@TjXxynXwz*+^g8Qj-ojImFbGARo;<
zP6-Ey2m+v?MPpbil*Q*^*~ysyypef}cMxb1<X3dvJ;{Fa&~<qDKEXm!E@oK4P){p%
zW|pnNwRNqeaKaZx)&F#6LlDNL2<6Tg{%-I!4Zy%2icsMV;Yh8W=G7W%lL-2IcNdct
zz>|{O3wT_NjQgoHh#8M9Qzu13XU?vKVXL7v(wiF&5J|{>d3w1Hx-nV)?<nBf?KiGY
z!Nc=jGU$2m>JB1;<%xq7>VZLT#7z<Ok2L5n&+~zz<nso)T-wspSPc!OqKq66jhsjf
z=P=tz_3_MOqs6hDhG&oogyQTCzh&;0juy=WRiC2wiJ@E}n~)Ibe^VJEki(#>{u>9_
z>xs7SO^v3TTI>QwFrmshI`<Dzcqh0aC_pQXD#tm$kxZ+Vc;{oh8%xJQ%NFul1Oh{&
z>5Ziy>*Df?@pK}JAvnn*I3$HpA%mUenkU)w=w)mnBT5usNjcH%dFt(pQOs4Iv%kg%
z>M)|9Sa~qyUh=WRcu~Zvjk%2)NhRAJLyFX;&IhEPNE}+KX@swHpqM@rPDzf^0&-@@
zh-qm>{!Cnu)!YwgWvr00h%5&6acp`B2BaiDK)B;sXW=w^GGHqnl$0Ojtg}ylxhfOR
zRN?pyTa-mE;EP4vT|F7hTkG5p{x?mKD;k492ECKEpx13Q9oPPl)4_ZxpEQcSz$}{a
zE6LPm=G5LZ+%9}dW=<_S`)QQb<$_v4<_j$%m^2~4wUxB;tu}57yDaT?uRF8`??IDQ
zk)c;Q4tr4ABUdH{6fCWjHp?C#Qj#2_wA_g#77o-CP&AJV1QgdaLdHnwK#@;%-v4fD
z$HqC>Bx9G%LUW^8)yH`xqyp9(X_PS1F`{V_cmcSB081RX@Ki#OqJaE)Z8G@MS~eJn
z0mLyCAP>dqD}WJpijgz$@X-8f^IsQ2gHacIf)i1W!|}w(#l+|^N)?Ec5MI4Lp7rDa
z2n@wvKdYw>ScB{#RO_=D26<`{{X6LG*?ID}A{vp`Ddmc!X6*3cY?mHQ3#*rRjWnDV
zqQ_KB10)ZnF}2_|w*lf1sie|%(`twfel#_|T>ae)LMvb3)s;i$vT;m@w`wZWeX1cu
zl3_Ts+SnSvfBiT=j9^SS6lH_})4MQmA|_F^X<3Dk*jRLsD{v2hro4t0aUl@e@VjSu
z45ZJv<=ELWGo9o_hu?oY&?{)bV?557?L8vNMCE@p#yhnpX`7E?f55SfD-VqYMCLL7
z%}P$?#e^1LCM<k<gwRdkdKl0u_(nw&IObZ=)=pHo!*$&BR-tPm5Gx>t0!BzZXZcQq
zN?E)Gyk-smyWj|AzRBE3t>t;jpy){0%Bg!(a4I4LeGfK_lHLKTKj--dIn<U`?uFa<
zkpRE6fN8vrE~FbEBd4!L-=~BVYw%)oy7|&{RTdjE-~y#*Lv>34W=o1OcYhQ}SfZ-y
zs^yXh`m}TZ*z4Jf(0$x^adjBqR3*FzusHi{*KM1<yUUSdBAF|*uPH-vNCl|c3rjL*
z-0cNc&kc2)4o(`lnHs!73dQQo^$Mcnm1)tkr`sK|Y6gH!Ev*EYqQ~roKsqs6F>+`S
zwTvvHx#OXv4bld0#_0DY?fj9GL#8-%d*B2P9IHzOE^*JJy*A9cXayC5V_U#byS581
zmqKEihjFsJozQ$@|ASM}A0c5BroWx08h)_)Zb^0*D@8@7zrh8A2$kyN{+|Ljw(pZY
zpE_IK!Y~v)8P~mDSnT~!ofz#3unQ=GvSt_tbXEYnVho|w)%aCZ+?e4bsD<0*j#t^G
z*H#DIJr%;$k<1sjie^C>?CrwFl!8(n62D<Q28>1on$}AKM}cUXsjLK+6MjVvSk?uU
z`fr$cu#u(dd+-*6ixFYTZuB-xdkDm#Q;P@=d;ZX`?~q8q#0QeRWWabJk|;+Zx%sn%
zCz_u)5T}Lgoqi{6xROf9s`vx+V`l*%6kd)|zgpa2ws{&)sH9nw7fj<O_E}qJ0;mtp
z6{O>eKYK{Vd<-&*G-saY{f-OR*mtuUCB63B2$W)Fnc_q1Y18k+p$MY|r)MXGcN&N_
zuG+2OV<F1PM3Pdrd-exUiUmj$>+`7p^}Kozjawuiky5-to&`wVFsDqb&QA@s7}xyj
zERIGqSBfUi?K!jYHob(_tZMKrZ4V7kD+(|Jsc=Ikce~iZG}&gW(xww2%TFJPRNA%C
zc3zH^ej@HX6vT!q0AN>2e>%VO%~CdF&=-?wk}!dKKeHq9!V(j3n0h-eF3^9NNTz^M
zC@u|)E5xSw@P#PcPY<bf0t~oGxz*p>J1nny`Brq;lmK*|PBLG-DMCvCn$eI6C6#0*
zmd4@!My_I9OAj0i{Ts0Q%#AV0Y<-eDb5SoLgA5%SQ%Pt@DIJ<Dojq8GMT{=NHY++}
z-A_mK@d2oRI?pSBq7z<_kt%%NQe93F^n5YCeEJ|)B+kZcJ`&BWaDYNlucfa|%O<DW
z$W+G}l9~!Y(PZe8o(T4cmMxJnlRiw7B#$Pn6AtFX{%h8QEraQCXzSf8rxC=h(=|d+
zpoeMhF#}(!(ubD-=V#LpDI2QiLltZ*&aZ|kq+TBT19d9Z39{2-Y3pa3SMWLleR{sr
zl!_`;t8t+l;)<cf<Mat1F78=OppQ^@Z<Ny>!SW`f3G7|kj=kl~VleqnA@3??7BMXp
zRRJM~E&u9`mD$X^Hbhq03Qf`2oH^&>0rVA=mxj%i6o)PMqd<izX?|Hb4a_*Cfyq5&
zE&cg(?8EbLTnqS0uK3@I&SDkTX`pp6lGDUV+LI?sW3Z%O^ZSG_KTyX>#1ho((_Em>
zVl}1<B4xYgo<52(EjPGBMB--SF6LVisUn?kVMEx`F?M!Xdwb1X&V<G;B15@o(zkH3
znJK%ccqoc)G+U@VB_Bzu+*pj2a&K|Ho~@?-tQrpb!&9~|@mHq1&sLEc?+ZW*QS-X{
zqcop%WRO<;`AN_Sv+VcU5urwX1--1cfVEylD<yg^JW~8HA!j*?VoP(p+-#v77%>Ao
zrmbn<CtDpoV|^`C$B>)3Y`wUP_SngJAuGQj&JT>+(lhI)l-cdag8*Ku^w7?93VoV1
z7;XV&bTR3ebuLu_oeTz{w2iQYumFLSmbFP7noK^y^l2-3wOJkPYD~0WZU~zh%oef5
zdemn}Cf9`)%U2$j0efo(UZNu~@9od^TBB8s1M<IXn;KZ3$5!-#H4N>AnL!ryD#vNa
z&;~l((G#>D4U$~-P@aj$%ys3G)?MMnQ@#wHC<S!k(imZYxT(+HW^{z1=N<tY`{$tm
zvWBU5d2`@XTz&>y+Mg<rJDD$cnVV70H!KJ!I!Jt)QfFwgf+7-}tGS$PykEoTO)tTs
zsj{wys+Re=>jB3xUvag@+S=OA_O}$ZRG#bnio8#c#>cIM9)ld9I4HDL*mJuqF15Y`
z<%rPUK0wTch9E`UEY+mSZ+zb?W)_}gPlI0)Tt;7&-c;)$!tiBgu&w#?%xrIGuZA8S
zOAk`F<V;X`(-v4&>`ezbIUkLZ?{bztPE_2w3skkUnu0MhtXfkP6nnsjfw(7rGCDDA
z3?L$yxfyLcy~mH)IMagS;+*Y^mD!k5G`J+2DbA4Qj^K4DAEP3sEJoZ0am>}2Denvr
zqfk$PGW-{<HPNVRsePQ96EZ~zXtByyF^GaM3AtzesqsrwM8e0^p~G+QJDx8g3$G9p
zkw577>P_}dllSWnwH?S)&Nexa+mIlTgjFbDPnEh@<DCAfA^difMo2^Ow0i)d?60GQ
zo#~}CMzCN%VCbVMlHVz%9}|ulW=zrHEB1dk(wkILkun__kI;E_uL6l##cyoWXeE%R
z8p1?0n<ym^OmX`;(+!N!3tvs$_BCtu-@Lmw>yQYUVsAf!j%MbT>I>VKK#vpppS>nd
z{(SZ)ShV#Ikor^+zbVdAs)nu~HHkKgcL1%HR9e^%K2FhZ8o90)lPe*Z0qX5Py-vB_
z{qq<#AK~zPuN?e#fPnevxL;QTULnqi>2D4(x;`Lb7-F4t9ruj$QQxmMa!=ue4}EBB
z)a&SL4;4V5J8Uv<a4(vjkiM7@d2m@pO%gZ6`irvx5(jqS8%B{UTDodZGGs_q{YP30
z5jh~Sh5#d=Z7UHXEFVQULid<T$bhB4<8@)b#H&SkuI)8sNis#x$(GMNf7We>lbN*=
zer^@)As5i&=t)~zeD9ZsMtCXXRKjin%!h`XLQ!xp4u8~DSH3QNGogQiJRMxUnmUjv
zz%*euPp`obdgFbveImcHEb6J{{8K^Wv{Wm$Abn?N)vjW4M7(+UD#OVSaosD(250;D
zkkp={?G+;Ww&&An`xW8Me}LgB(#oR{eVqTKA^=8@bQy|tJrW<KEmf&p{)b6&I?^ro
z<E#J1eh-cQ4=50$g2Z>Fq>19;fuo&DPPKfAXz99l`fHb1#nEc>qFilfReFXA2r+Y|
ziiW-WP919q7+Z+ZKCP9%GZ@I4S4t{@IO4I2?<uDiet0#s?e;H$=HF(?dtZm>y8vEy
zr{MT8(F@7#-j?PQn@f<UimW3blu%dSf}RzMfT)|rT-+`iK#&TgBDs)4cDR&v)6c>9
zCx<<{)&OW9MA&Kght!e@3zZ2y^qNUd8(?`OZR02UI9>{r4PVUOq$FxUl3~iDV<7)c
zXTFs7;I^R~*srkr>y$RRFSV__05~{@{Vo9oZHSf?=a2~GlxU->i+P}$xGtq-rUeM3
z&Z_PpbxP>YjLt#LpLfMI#)x82Cdb%}pypc>p@@6t(Znf`6W+2-dn(@hE1RgrT6l*@
zlFG$oSmICS(gr_mYhSfhv=Zh)8PX(kqV18FEYvCY-ZkoLUb2wH7|^WAqJ-jyYGR80
z3J<ng<3xd_bq8LUclHW9M0+59k#>rII3C;HT>g5n$>Xmv``Nog0oK&RaSQs%+Cz7q
z3sJ%3q?Oa*n+Kn!S^4Z-L`l^c|Cmn{zj;y^!Y;YP{-YS>{Je5Aris^YJ9E=r90^6X
zp!PLcb0H2(BArShIP~DZ4k)@&lU`Umjnh-c{90ht6j8K`J=~}j^^y43%BqqK0Mk`q
zkvsgTNK=|pvXrPryC-xa#4y|o2;!7mh!*2;)ok%H{D(X?l2^P0y}NhczADPrG+vKD
z6Tq+#2$~5?x&UKLo?h`z0WFENl(?@klXIa6eFPJGN$<(xwJ1Ik2(W!?t6h)si<|7r
zXInQ8pMs$NQyc9QOs0GjcabYpfcS<?cj2&0lAMm&4Km-ytNB6=1jSyIy_4?C8wP$M
zwP%H8kV>-ySZJs%ojV}KXgk>m=Ha#%<wP&1UBXstdFW0Zmlo3C7^HVz(f@ARLeX@P
zB13@|9;+uVGeRPe?Wc|hL^H^#&3<m3Qmhg}n$ir${NI28o>JHsE4uBp!z<+6%C0>H
z%J}q`&a1n+3pngN&|x|1<IU{qJ{YB^L{>8+e<psoHl<f;^y5(hl#Gs6nzoY%?;w@_
zTM@qh^CiR5G{3FoWmjO^^kKFI9AwB;ghA)J!0qq<+-Vha2`xq%2$T&05cd%vGDB(e
z_xu%An5}4OeW{Ns4FG~)3+e^kQxzyeiAnYFw0RnwMk*!T4fK4EiR}cI^8P62fMYTH
z_E9JzRa9jS^(w!`XTfN6oJ<2I253vGRWYQ{=ibDy!91_A<m2cdQe{|~Yg~IXo@~lE
zkrN~?)GGT<fHT?Yjei%N0_FHB#<ZpMY6BSaa)`YD9GGLHP0Hx5SsiC68?F>O1<B|*
zyz*0f>{#j&igTq)BA%YIL65sp!$NkK5yhbp#lD*d3n3u4DNcf9OtXl)fNB1iHS5Hv
zHNh5=3iAC-QO1yLkD2w%Dh7j3Pt0g!y?812LF(NrxQl20<=D33j6bg+|9g@yGmC!9
zb}x7-Q1K>>?_}9IVpgmNtpTq;+*IY!bcn4MWj9Hez5>61Vw-h2@gOuc@U^FXzfl#f
zRG)+C`0$Nv159D9zgVp-Z;@xezqh743QsR8F1u6s61R9-4{ct7x6n5eLj~Mcp!zx7
zM@-uIyB_{8gT;FN9(aG-?#&jS{1P@IRA!ZucP3_zgzPI)Ve7}Y2K&}x5Ibm=BPgE@
zxjXRFl})}Wr0663bjE^3KJP|lD7&agDS>p87;=SINJ+%!U@HjLo=V}RZx=8Gk>d?`
zX^QoWNJnCmeA`>j>RCg!#S_gildUOvC@Y*aL{4T4yci<IF8?+%j330fqauz6g}1Ob
z)r+sta+SdmzR-9@PSW{ff)}t0^|Z9XdS59gf=Lm@uIM$D+2~mxI>bY#$V<Dw3!OUl
zn)C^Mt~1Mq2?jp<7ExyJ9!7j4wK6~gv3yOp30vo3JczXg;<(EgMjsp;OxhGx4bA{i
zbUnn_jigDe8zeUtFyY_*#dZ4A16GebWH9JzIIAn+l`SuGEpoJIYbZ4`hw)lig`Nq9
zG&8Hb*_mx(<y>72RHDf&I{2$x4XeBTxC5HaYu=J=KoFWoE1;j{$1I0kyX)oH(gVjp
z;I5ODh|WF;^b<j>G=C=t6p{7~U+Qds8<is7_XScW?0Y&`fIk-hUQQ+gi0oi=dv?d8
zKZ{=(=`xzP;yW>n_$vk8h{jxeD_Mxdyvo)OVE*g0C(E@?42vd>&*#&8=gA%L%+2%W
z@pPZQ7cf8Dm&}{%X~yGp06-!)Mws)1Ew2FwvQ3{y`ho3B*&1yz(+!o$n2ru!Esh81
zz7=tLn*3VP<!m(taytK{g6vVv*Y}>ER#y7t_<Ob{uKi9&^X4SOG{s=}I$1&j6a{-2
ztZ1-d4+9KPiLhfAHVtXw58tGJBEr&Hkw}XF7SG%<nF=DR-x6GGUTQkT-K`$T#g)WE
z@e)A1hs6S+T)dgpX2=RK9jO7wDS*Z8vjAG$>f__V8cI~fn}h%xV`0bqlFCzH+=d$g
zeQrYTfgJgGeh85~pSy8mHq-da?Y(NBG?cJ}$<X~1xBThBS1U7P{%uyuhO_%X21#Se
zSoX06%<k`mFz?vX_X9PAp{@px5ykE99>vFMCE3n_^Ls}qQ?(Nm^)Ss=9~M!cy#Bv8
zI+#SUv6|UZ4)MMMcqe1-+u_!<IKKS7;ZI&$fkB1w3i_#5OG<VOw#+>Jn$DElc+B;Z
zzt+4H42#@lV%Vnoy8>SzqF$tfpg(0Wi-6$fYG>u-5}S_#NWZA!laWLfe9wB!r0|^m
zx*}_wk_8od<BTJ!{I-WReM-XG%GEOV`WuaYE3;bdvU;zsd3XQC`nJ*Jr|<1&aWmuw
zzW`)lBo}8K-iHkPhM3}zU3z(IrU~xlvYI0r{sRwKzo%hfQ-G7w``>OT{s_7yP>nG>
zp$%!`k<xUyg0y<trm<E^m9$SBhBQQu*uxUl4grj>fX&LV(Zf)kl=MN)jL~9(M#9P=
zzJMD{^|U-_OqGDuKvvkY$l5>XrCh+24EEQc68bY^Eg}&AYqdMvIzG}wUGPfQ2=3_#
zy5FG}6V6dY+|CC7$bK{ZUC2=yV16m9E|`ZTGwtG|`|^him%AP;-cE_Xck#I0*r4Fh
z@ex%u5NK-}6Zk7Sng%7kCub*H>$Icmc}waExip04<&3o8S8X|I_cp4IC=~vh97zRN
zkh~*1!>x-;5X8XQOu6m3K!3j7$L~$|fj_CNm~0T&Z=7mXjEeK%(uM%&=Z4dA&9G+3
z3#sGus`ZrVr_YZ(G>;WT@eH|&vv0&wE<z1=%j<R3I@at+_}$;?VL+o@4|u7AP+S_Q
z0`ZU>bSz^iQ4`!I3>kPh4m}wP23ga#=Jz*^17c&+3E$zMq_H)+cdWRk&P2p<&rEMD
zhDrQ6mU&GVYq`EFP<`L|n`B{G_GWehh%vmGv$jr0PD;AZXQZ-RI*kKsjRzGB0%Tiw
zBgMq%E7>r#)n$_IR__RtV^p%`Q>Xnc;(R7rT=}`Nho{lcx5ZW*6DY&$tN+Fe+FZj3
z3$zq1W@LE!3DCp{NP}Le$oAS*HIE9sN2mz9g9k2Lv?Gw(b*~??+kQFY65PxEiBj!#
z1d2W1(VhBLnhbsk%>qZ5Gz4mcCPwcvpIFNkWtcSHF|1SBm=HyPP`NbT6@G=3d3Hm=
z2{DB9O`zOX5$;l#e3D;H3W!a4pMFObC%wkgG+z{Z3TY3k^Lmnq{i-7mum?@cX4-R#
zUJ6P<rn?)8{o$&NaL{i*u41euPc1|O6|_PsN~a=9@e0!PCeH1>s7UsP6UqUKLQ?2j
z;Loq(?Sh6SakHY=J<a9>ZX?Aq6?iY-7DY}>l+EZ&gcWpJ@E}*+3~2JmL|)Us-;x+!
zfp<j68$<|cEoUE9_<}aBx7M!?)wJ}jyO<_hu_*$yApL<JhKK1gaT&wY5x}Zw>o_&8
zO3+?~_BEs|jM7-d`zZSf$`T~0M5k6f-YVUM0#=L3?T^}Jp?_WXQB(-~a;4u3Qo;Cq
zy9^bhk_EL2Z-26kwLWY}ak!&%?Pm{M@n%MZ{*uPdoyEdkKHseelb&{0e#2J(u)cMX
zRy3?I!ne_Vhqj<YkAFI-;+W4!<=UHf9qYN4uEVg9FOC|#N?w`_`fK*Dq|hBsI@c>;
z6?4uipc3gGvywJV&s{m%^7!@rmB~trqdJte;fWD}d1T)>(QjdCCRMse-}PX}*M~kz
z$qBvT&HJXUo30%+hue$ioagBmX^u4b<7fExoiFP8lb6k!H*^i^;}>M3B+XYh=<eW9
zf=fna%t!O%N3%2O9KFh+1I=m#ZoeD?)z37xGhC1bYr|2X>14@3Nc@Q!qv*z=&l!5&
z&Ct^|#NU^`4}tA&e}zNe`%mzljOET^2q<UReM_pf?<1B#)LSh5aS?+2%y8Ru7zbkD
z?QQbaUB)cAV$h8dz@417w6^yw_;%E_C<2RoYo;-40Ur#Y{N3Y;BvaI&<!Siair!R(
z1+zXTWeC`7xI!)FZ$eDQO-y-GZo8t1STrF<XiUUwH~^L&v18PXnk#Ewl9ihv7L~ox
zAyNI!qd@91QSe01AVVwX(Ms==k<@W<!wwmG%}<afJW#kiYz&%`@(_N=(-j%9QvhfM
z6SLHctN=Q!Lj*q?%0z>ZRxRhW*apS|k8C|ghj@zvRET2pe-8S;G0OZCu$Dfqq~+oq
zYOTNlKcISgopLI&q3G>69UZHp-!*Ugq2gcpHuk%Hm3x7dl~mB*@+59UV~8|q7CiNf
z+2bjl9YU}877-7PO<v)R;YF#2V=iKl46%SZo*yy9lY3XzXi#(IgTJq>>cTl#ZB;ks
z`+Ljwv2V*X)cjF5OkhK#S^kHoG^v-%7jxFO(*)LXHozA#cor!NQ{SMKf$f!AMVekI
z18HG%T^@157A;_qjYz;<e|w(d)a7;PPwJ;)YTen=lHRgfU`RNTa)0|Q8zs$__V6_J
z6c9BvAL^!@LCjat1x0tD>JU|{F|Tln{hLoyiu`^vQ4IsHqABoK9a7Nx`Iq;{PiGE|
z-@+E1Q)@4tI~a&YP8<GhAHLGdGF2;Y&)Ht-K}oUG8}Cc5j|4oD=CO{43NG1Cv%m_?
zUhlWKw7+u4a8i8dC{<#}jSuONwsd5%M~!+kGEH0@UE8_C*9E{1Vb-niiAi^)ws6;_
z>8tO*%{K1tz=kLuiwo&O#dP{xc?SD1f=y!byGiL-Jj_P>nW?O?%shiGlzenRhR@9h
z`y0|$Od0lBF)HfuJAjoEi?Xz$YxM2h*P9o}TuZy@&iAZbKtHkUXuKu@nUK5@?Q9M8
z?d`tej~q$*2XUD1`h2JEG_&)|m%GPw+-ummy42I#pt~X(gE^YwLNT4zwO?!5Yw=wX
zp&fO;$_u#mT%2_0`#mgLLvi~;=LutZw^6keb8!GPqD!;O1bx`@DQkdMK174`*FpHd
z0QVcXVrbH=9C4wrcd9K;oZeF>6&2Z9<rbR|_JNvzX4W|{Y*D<unR-~wPf!ItHA!dw
z&G^7CXw$ys{@>s=q8P__mdeS3Snr7CC{-DP32j3!F5FFOl%2x&oOzBo6Lu!)p)n}v
zYn|+&L5+)IGcvIe3~L73n|l*86C1OA{L4y~oqP3@8t;^Z<sx97zBkN_ebMrYe2Ry~
zf6yW8dQe3Poi~<P<PCa7_;2ar>Lv2*KI-J-?;(@)D6sVQUSEbdIJAu>v3P5YJeZ*S
zwLg3zKw@b5z*-70M{v(8R7p0)`0;Hrd%Eu^{jzJlV=a0c0<HQnZN8&Z>FFVm;^PIT
zi5D@ZPViHCi%3ft0}|nS=9qP1M_cV!rgVnCKe>S{8ovz`UzQ~Ks$#^~KrwS-q~yRU
zN))Q)u;nRrd*0!`L@|1X%{8P+Q4v-Sfg-=jT|o=m$fJ2ao?zvqOLAXl&0bPr|1zmL
zBkY68s9Zp7@O31GzM)ljz*0pqQw?v(y`=Xsue71A*sS5~stCTThfTF{cOJ)BBxgQ!
zxq~i$n!QMI{FJT7aeRhbHn%xp^VY<=s>P3{zWp`T9zQonanl}pmzOjPSzT`V)|57$
z6J+{9_}Z;{&AHxUBGxA){F8{hn^Gp?sf$bM;%^Y2uMC-lv7VxmdAdZs1DDQr-nmr!
z$Z7+yu-$zoSh>HTW{#raBvlJItK9WEAUKdrGg^KR#=b$`-5rO+wo|ETMOfm{)%_;K
zWRk7=W4j_CZ(as!%#ufBtbr>RKems=ht`fID_%OrwWwp<v}oH-5OxtP^vLVnK!1x(
zV2Sq2BZIR1q3Rn`iY)CwXHtTvdMdGYIoXt%$6g`-<}xMBL{?PP`D%C2apHQ1@_@I+
zddk%aa_aO<?(CCkE1&)qjEmPgdy>Z%AV<nUUw3=^wkp^lTcTA0kF`tu6ohQnOW+*J
z%0m;hes__hlz+LkGdJC)kFha=lpZzhD4~-&p7T8D=5y`QxyTyDvV%$+JF(sc`+k>4
z7<T3#u*vg$`K>pdY^Av2BW0`X7F{>b6o6{-I+n38u3xUHt2oUtA%ktbcy*!+Fy5@h
zFX1>ZExUg2ab<0VbNYj2G^QmK{$8EW>J!43vBO=O)fZ2<1q%<usu>?8&}Yt(k{Upo
z#7}@Pu;M80M3h-GE$b(XB$N7L)?zimPmOW*pQ~LD8w|9Sm>9ldRj=$JHibGxcJu@!
z$1qXI-BeA#O*5fCuu4BXR|wxUh?@9cXRdbA@>>k$$f<vS=aN&M!#HF4fScdB%BX<7
zywr9{d<3792+4_%Wy~4NNd;Wei&>DVBi!1;bidZSwM)Q2HAa<2KTL{mDD?0Yi_8{5
zt=jW+wVW{kK0+G=<LCsnd196*7kR|~Wl{+nm}il5f}1IwSm+NzP13nh5&U%@^13r^
z(FRiA*=B9imw5I&nh&pZ(5HY_e)2Z>u?H=68n2U!y*38ZkfyjI+;31(t$=b+)Z%6%
zk5U`wqrrR0w@!a^kJ>fFf9U*7uD&H*Js<i?=n!ie?>vn4V@!}SKhIZVxcrMJ^F?;s
zgy{sc$JBr{!{zU$ViowAE#B1{{^H6+DuzQ2Rcx}h^)NOYvzA+;y92I;#*mr|_s_$^
zsdpBaLJ@QUdi#~Urxn_<3=B<6GG$@95lm=u$a8}3?UnwIw=qTENN0C#d?}U<^YCm(
zU2a>fuLC>QlSBRm!Y^8A!kNCnTz@AYS}{0hNb)053Tcficbc_;#F&%5{>A)A4yZU9
zcyCir629STA)v1KA*)J}e`$5$R4t5iTFSA<hu`qyoSZ1Ifv(7YRZVuLz<b6i+;E8)
z|LzJHy(@UB#w(jya|5P9#RaQz*nMm@Rgs|A)5fBu7GLr8p4Va^Z~tPyl8i1j*iM^u
zkiGEl_sr%HF|_%%;RaGg;<TezypNo)r}Bq_A88I%w*fbWywRFZjpXtqGml-#%@UTG
zZ1i3@#)JW_@cXWFH0wjn%YV?DeNSfCWYh<ze<nTsP0l40e9J!7TpTn|NmHR;r>KxZ
zm5{s`(Mz1$7!h&X5PY7T+l`CFOS-l-Z-tWO*L>t+Rzg<99mA%6c1N>gM$k(U{8c#<
z@z5cAdUWG$C@u@%TsAPiwL<$Dl2rS_qEFnj1_Ynu)By_Y7p%)?la9oswy%$hN(njH
zsEt{6@0~SadUoINJsP6~(!IIZ-jtcUjGLsk{k$s(3|Mp*#h>o+q!2>w5@q|wR9EpC
zE(=m|3LcYS3SxG`xN;oJT5@}uP<~jS`5b*JTvXX5zUs0ms*Z5kK3MTRA>TY{3J36A
zUi~=CRvW!jPBv22XlznUAQ-RQFL$mQja>QGbH9D^pbH}-{fm_FMSfJ4GK%9Hk9H4V
zt{+kwF42Nuc6~^AUq$LWuaLj3D5O-2O*W<Kp!{jUFhNDImg2o7R{ZW2(G_JVZyjf7
zq_Q$u`w`_$ruXgPfe)#jjQe$@r4U9Se9IrtaxvXTuXm>9@%vR~Y=~5lPf?o80)AA4
zA#F)Y+aoIIeE@eDD+IFfk4dxI;2#%A^UucB`1p81@#FN;`>J(^qQ)W5_A~z#jnAAd
zw>@OUp|TF^L*v9=hL`bzl?0f-+v_(O(|&dWV5tgaD_?pxFFVjrIX>7OHHDq)CFKFy
zU1{aMuImOt?9Ro$Sa6~%a`CH<p$b$G+y6rksNc}(syE6HS&a|iX*(TYuz3(w<Us>l
z`gELMK8{4~+Dz5*-LDr%EPZrc!oQhNh^o*MX${+W^9{7;1eoi+`u=HEvYK&RK8f~n
z-$-dvx3S^<?#?2@KY^}vi4|2fo;p>{EX-)aZ#bKK(FdqYLbNap7b*A{yh&9!gjN+}
zr74!tY=>*mf?v{|QjJclOwxuHC3~X$bFY0_!m~$JN21|Atj;(hNrD8VuFqGD-f=sg
zuJ^;VmvK!<*1`>jO`j}Xpr;-&Bk&HvU1{D)K5f=mN=q@qzLz-0?$f-I%Ao5^Q9Z;d
zZ4oaBjltG^#x(9ki=F4dMpHnybEhwP>A<@3IfEa>@GWTk(92J;P+hKHkcvIgqW9Qi
zv(3Zs4fk-Y$JX^Wwx^VBlQ2Yl414#@rqAu5&X`G~jby+*4t-lk$K;TVI9|eyw|Zrk
z6ho=jN6Qe9zV@bij--3L#77QPwzR+?8>y3-6SjOYQQNH#UBG<ruR_hfx)1V-LVXU2
z+^hH=B<jM(Ze9<Zc#0of>fLgY?0zR6`8LV;A+bmw3QhAr83<W*@K9G}TKsg#u9IO#
zg+VYdDH0&@di(q{ibWHg+mBq&`kvMnpU{u5qKa_!)Y4ks_C{D_v6J8I4eUkh#oF12
z^!IGj)&W#Nw+75&pV2a!>7B4t`z|D0b@<aidyV@cfE}KH1dG30o(U2|%R&MyhnCA2
zwFNG>=kl%9X*MZ;$Mh62G)l}qf9|)nvPQ*-gS=gxq@^|(`J^F9mFdjx=>ed&E2+>u
zNYS7B+ssX8_T?)mgLtk7l`oSsOACwsQmijLpK}g%GQ2pdDE=f=?@LT6sG^YXr&{-Z
zyyP=#dBp4bfOHu*h|WAq2>5cH+uh>mcnBFIsEK4`DvueqqT^0(n#+Fg=iN%`u8oVB
zrc@!!2rQ-eFjjt;#+X4%P&PKH;F)LoCHA;SyqD>tPe-+JBj$>FTAA-(M~e8x!}8ho
z9_$NLD(lq6<~6MCmF6b^#ZT(cddK7}%J2Jii{up_Q^hNEH)mLmE;<))R;W+{CO!mF
zehQU3YPj!Ad<Xp$8-oAAYeNyE+7djnoY?J9ve&}9lVb-<24CY=dym_{->aaNuVFpf
zHl37te$5jtq_u;BhkC9DiK+Yc24jjj8a>_KiM2np`+Zy5p(76U^On^JWo|si^2(zO
zd){a8u-)zq<(1_%=-u)S5DN3VQCB~?Gq?vPB6me{H0ZV;zR^9{(|4B=J(#RLeigqa
zYZ&klBn!xjdDvq~hpMFO#Sl#;(oIl2ySEO2(NuZ!o;M6~f=RKR6k4bow05LZgvi@H
zz-7G$lgn@QdP8mSJ{5u@pp{(3?Y{Ngb-b9|%(aTS*d3Uom3Y#TBEH+YWEHb(=cUhO
zA5LAW{vPq3`zZA%>DQSY`sy^2oVS2ffKIQ-l?Ny8;i(^(mdgKqD~#XU;X8v8NI_Df
zy%YGVIbDdl^nU^`-IVmdk8Gj!j>bo-&3)DV<w*JeW#UPlExA(j^Rs%{><ORxzTkGA
zK|-!nxrnUm2DSdkw?d0!$^6590zJ#Gdn#4}?Cw7}m!7H@wF+!+ambCxC_<oXSuJT2
z>iYzh%ifgWk#Ky&cv5Nbl|;9R_w%^pxm<fjkp=%>tn}3V4F5g9RBb!%Wl4UZoRL&-
z!N;+&S;vXCz(Q~8RORoon}Ny$>16ESgs$ZLlfQQ9QF%c~iRFPt!WRHKtEA1gpB7mA
zp>!(@g#jimi96j5KK$!A?s9D84Qluu&1Y&l#;(p2rSOZU8wv|JT@QMPorm$OPS_WI
zC(2s~>@M*ZtN6(uUOwe7Vj%T$ta;Cd=et)>9PWA``bfS%^QM<t#2BGn90E#^TaNR9
za$N15Pp5o(<0WI9MLxEnu)g`-q$V)adi{_hHbslxglLJ02u+DBaY(YSv(f1n$-8C=
z<6n|4fBzz7n^j`1BqEo)OJND#cE%q2FiBbckTP0m#6GMQ-JD3}J61$Pq44ljVzy!`
zfg`+wf+(nUW(L~yp9nlDQhEesJ3ey4dRW1&6x+}k&7@5bT&7o$F+k%(mw+_ML00@a
z5}j18!?C-mY6jMfiO%duc-M)>3ORjBgabsvsf>Y&ZAxo1Q0JuTG>-Vw6ln@8kNW#Y
z;+k1IG1d>Jr|2lKsL0ZR#8{lKf+;C2p4>4}nAzPFx{I#2oG47O4Ah%6^BaXF(*0)%
z{+Ej1U_t;)&O68$sSe-VZUkiGEoMgC`ZX^|smt_tLA?)#_R!Hl7cfm3&8^?1aS?;9
zcIW8Xkl?%}cQTu%VN`X<8tJ{TA;l%;op)xv`pX+EzGjX&UbE+BjIkH*tgT}z>`1`A
zXGf|`%?^Em7Kb87{iK~vX){wTZr%;=X@D%I{za;i=J?x*b)E2qCjYSXBSe1eE`DL&
z1OE=cQ(%w0b;I4nPe9wnEmf5pF<9S-xG3Qu28~;tV%U+Fdr3j~_WpjTqQgDt1ya%{
ztD7&3sGm-!KQ<>Gz?85aqcPm|aNPg!AU+UiBT$6y2E}~WAGxb*#6$54r_^br?&K!7
zOsj+w=Bj4qcrG6(22CfFT9U)mO9gblrem@ZX)FTqf^LR7<v)pWc@F+Axd<MG$Dogo
zgbLlE@Wig*eXu$5r2Tmto!_wWoit}~w`aLOz2>(HuK*TijxH_xmnn~TR);nNB3VT{
zIUgqcD?<mG{n&rDw05Z_=c~$Xen=LJ<dxggM*bW1$#kUK$V!NQnjYuJ$#>aYm4i3@
z5<(-Hc$nHKEayr;;^<PwLR7yPo+mf$>;#K`kY>53zg-2pWKC14QgNMBsI-T%pilW9
z6i=2kxKfHQich`}l7fh=&us&KDXXFM@k9i;Gp4Vdb`>a&ppJ`SOHd$FY(5B}brrFf
z-pVBKMSP(O4G}%o`HtW4<wOi7a9pEO*pHmNnQ5!@C#*hj^xnxibjl#VhxD7^MU=Yw
z%!k7I)?GElEB0bR^Ye*Vr-k_`q(vH<dsg&Fb-#!G;rEhloQ~=tuUgpqe=&(PCWF84
zju+!##LhQiTG(^%HVOI65y_}I_L2HTM8%>G)rfO1!Gfe~y?@C=L7^Y`Bh4!;(3J>0
zMoKx@<7HfEovLHiHTel3Wy>zY*AG;H0$I?;@srv@qT<zy#(AlvfA>d}uf)l~fD5VQ
z<W!N7U#DqRetBj_7eofgK8u%@qnE48hQDLER{BzxAki^iKplqN&e~=*+0cjeL_^@{
zvq7;k2DMo!(ZONeMCgEMd=~tYf@_O<y?0OA*7hJ5eVm757RisTZ4F1k0y9HUamvlA
zHPC$N2VKX<vQs4@TN@EHelef`+IcOl%t9sU=DC&i`0#S47#*u9<N4-+X(pAb3l#~C
z+GghL8J+CUnb~DMrfJMcG{3jDLi!g-){JcwWEL6;YHXv}ZL_v|Hmfo)Z}+l%9|WvA
zjoKaRU8eC2q&k3v@DvEGVT8DML6sgC`{bSLS?6TRWCtcLKB8Z5!;|EksIc4_y?Zo)
z7JX3!AkN8P&}ZZ7YV>pKiHBLXW~sI0Zm;15rW2p|kbCHt&ZT1oitMGp-D32@JI6#x
zG+D1uPxDu182Zcq6_bt!SuP1B^q{s3Jj5?8ad=TOCL8#V`4NOBo~tEWt@sl{>&RNR
zDuI3Yr~Eh2nd3qg$QE%(0?4-EK16KU<>ZXJuIDDVhWHD|v$|r~0>iZANN(fD*q)Iq
zv?7t|29X#h#RNAILv|8t>fwdr+wn9PDRiE{R#*x0p8bG_jjAtO)I^~|nkqK61?=Q7
z8~8luR1`yIKtGlg7T}+lZ0Oj25+A>u5h6{f`xAS@PB~jGBdIk0Pm&qeds=xzCw`I{
zj7>lX@6QX{+MNxRvrM9wDTR@;&l0kQ@tFh0MZLBI1%z!$b}+)3=$0f4OS0@nO8(mz
zMLJqOPmLna56gxNKcOn})x2j5>bs7pe)>*SsuEGSf<+O6`Jm~3*SDWb6`2MGTYSk#
zz-MI+yN_8eqs7lHhU-lv$ZQyF>~n$_ZUzQg1rD-1VOkKG#7;vFjPhY{-0z<7>+GAQ
zvA|DCqqfu3GveQNdyC+Y42PbG5$OTz3HW3Lz7!|zoal%2w)2l?2Vh__G{D+NgL;!h
zGt(Vk4B5hXHfoU~v+E7olWm+<o?k7Uwz7T2oys1{@Bz$LHEv7f6a2WX!xyjf`$}N#
z8n(cT_=lUBHg{<ZI$us}SHBGm@zV-$*`uogD1O;td-oxpzPrWA@g4n0mxK~ae`hWC
zrBm(__8j5PL{9YUUJM1%^}B;cx={k{_jHHh)>dv#yVglx+>=i<f0%<Qohh}CNWq^x
z+={3USiR-8>K|}_G)hmW)%rC`b46&OvvK2urM?D($7H8=fW>A9y!P}o%Y9Kj08FYs
z%OWZ9k_@AkFopxRq+~AZYpj&x4p7P1jkq(i-Y$%z=2LG|oNYR(NG`A-`vL(r1&0Rn
z4PmhNZn%!`n7x~Lu0bM(R|Z>OewK`~{D|KnXkM`feD?X;N{ip#V_@4whPCyAgI3WR
zWO&b6PpF?>eOv&XAxT)RsCT*SJN3jZ>Zc;(Wvw6_Q93kT`Jq2c;y}{V_pQjiZ~D~w
zS3>Zmf<vz89md5KlJ$|*4ZUe3*3}(nPPUc{vUm~h&-KO?tDd}nJLI`fxpO<od*Lx&
zViEk`v7NRWlbT4!34yJj&1tfm5syjDW&i0P`gG<P5+?Vi?-z5oP^>M30<6*3`N3nM
z<|F2Bhh}6k{W<0`#@?X!7>(>Mh6CTrF$`!<xJqAy2KzXzd$8c3aXWwDQ7>PLpzQmc
zht=N9tkSD#N&#Kx>uEL0B%<9ViiK_2fI0d+k+n5I|AeJPP*Yl}x%)-%Ev_(wHjgbP
z*vP$&sK`|2QfN9v(H1FFQMP&!+oXjP*}NBBM(AZFV!ZH%QgY?OtPE%{(_Mhaffo~X
zLdBo>e}$_IWiUhDva|6FKJ5sEfa-v8w#hU6aoyEB-NVpitmqsUA?wvn4+h>SQ3INp
z$>6Zz`_P{w6YFC2H4hy=vEAQB2;0rF?tWZ|-@Eg9r<q(OJU5DLI&HU<JAR93v(7Oa
zeag#qs+e@x62Y4tm8G!1n#ikiF}XXxO7l(4nApCqZew4vvM}HmwcbsMRs4Ou$lB_&
z69DLLtakdm+w7ewD?lkm?!=oPB0{NK^i(j8UbV#&i$!fY`911+8UOkzw^6c;BaaH#
zR(qcBqO@W>p-=a=X-VxZyWh~?^dw1Y&6b<3^{3-VgkfTA<x5J}_@PyDleN<V(5)y7
zx06P1w5CDh$n<dk#V}-0L8*=&f9!&VZ)sXlO#KVRd2(?>^ORUon&&7tS{Bb2Y;GIQ
zO475+qO*e-ew)0T9s2P-URb|7XMO0o#?P-U$RYY)w26EeqW%PQyb(#b*EoNmTQ?3J
zXlK@fBGMcX_7G<<TPb6IyNbxgX6u~Z&Jg5UhIsy!jz}W5vq7zda*mQJkCg54%13Vy
z7cDdgsdar3t!Mq+Iy_6EXOs+$m3?F}486UPgcQ|Bddfx-(wvnfRyAz+el=655!1#h
za?_8Hbn*;{Mct=0rk<XRrWK=BidPlakCT+@pn<y}ZY)@3i(wJvw7j_vQ_p$n1%@p>
zWY1NZ`1tv~a%05Vb$h#Y5|N3a37U5s7bdmP%4KJl<-{kmW4E$RpLN7fy>$(n7a}A4
z68XLxCc`pZR*T!8SZj?JdFwV6!+xD#{!@&`wt(NwkyqklT^OH>xY7JM&TMo|7&(5`
zR&}wQHoyIYjYN9T%zMsBjZ&nieyATi{oSkdz=p-gD|L`~UKNE-PD&!+cKUADy7Nzo
zLB=@iPBRAx&D;>f4l%<bph%;$!QAG2Ypc+WOB$<JSyYUE{id@I?(~NLJeP5UQ{iVY
z2-VGtuFi%!%cGD~&V~b7O%?^4Iz<3ARXI)buxS}n{Bs>q{QhO2+V@N4!~hTVYkty0
zEJ$32XM|Tt+Z?1SVhY`OIgPQ7HpDI_n;g-!h1(iLBN5tGvKfh+Q%%OwKDwOPf9BDj
zJ5A^$d1IBXE+yE}w<PQ^qg8x*I<spdL}3(GtAZtjdAl5OHqLqw6@3M-e&yuPF8u>=
zijsiM$A31+wg_ze7N3U#H4-2dsPxOu;!QVUP)fI0cB8L@EtZ=tgy-F>_p+&Teit6?
zV#v1E)qN>^<`7<nfoNW8(US@<G+h3meehtxrK;6Qcjg(uI-RUgmq!|q!!Oz6sI;TI
z*fSu?^)^6dw=-*jt78%+B3PmrtEpOa%|W|;(Kk%;*Prgg<Ag>Too%;N<wS`Z2^U9o
z?4>u8M+bp@QEkyBe!AM=FWA1+88bWPuv|Q@F=5k=Ul#qvv5MFqWK}8cElQXD23-AS
zb6cteZv}fsOxR*P4pZH{8JRgzNX~1=Dn)!)=q2^-dC(i)Lxk0nwQVb^*Vfn`WXiO|
ziC`@JKYxd@_c9Uw?8(c7Pm%AY&lpVL>_l76eI>Lp`{es)ocnK9nMVKLeN;IyHZ0o$
zlBnAiJOSGtlHR~i;&UdM>QvKq1?F$wNH(A7B&b(bkHZJ3MiSuz5fL5U=$S$cVcKlY
z?!mFs%|sjNJ#iu%XKVXfS78HGB-*1t1B2^l{XK48_o2475<=f{IsdY?(c%~z?oYql
zqIpT7dAV#ls;HhjAGCR4ma{m)7IBLikSMx>K!prxG(N*H2UO&VPmlJxdoXx&3W0ny
zQZG!y+YAG7`lq9l0y|y4)V-3V7S_HPtEkUN_f#@9+HowQ*!~Dq#R0P0a#sB_V*v_3
z<@L70tBNeFr!>+79_q9m^%KMuKhiDj3H4cObAD!N#?tP@nZLWR>C7BOzRA(X;(H;-
zPnaU68qGFLa2Keh>Es{F%!qP5XSOlpl`5O2QENhVA01c0IeU78=%A#MSUJanKZ04O
z?dsSwjrB0%1pMM#=d<phvq+iX-LHT8XMpsbQOyPsHF|fF^gs}t^s*>-F&neO?|jW?
z=0^mh%-+?qADqNCpe2(!x_CrqtxkjQr>dCByP8Qu8AG#T*O{)TK-p2$KjVB8^+|3l
zeQIMF#s5CS#~dCU=3Ereen5X<O#X>`XxH$@Zx(^bo11PgVJyvNe!!dy_pmK=r$pu7
zYO#JkbK7olwjHEgnZH4<8+(&Q;{qf!ZX@VT8-%IbvY~}?&v4gmj?i{Xay(M*0FuZS
z3BxWlqh(VoZrD!FCI>xBuP<tAoUvtA4#zYSTaLFkD0C<Ho3CQI`&{#_c-Q%671vX+
zfA?~%7N@RKvuU?`2e!1k7ZFI6OQ=6{D!lG4YO!wOAuz63MI{>Fp0$Lt-Y!>?SOm(w
zPnfiI*Qw1G2#Lk^AjZMhC-3pO0-=#-LqR*%{)VRM%Kp*n_&um;H?2{7oWl_XMY_qN
z`@HUP;uUWUUHk){(2FD7j{NWTX+%%F*F1KLM`?d_{ucndKt#W>w;QQ>k=I7`&lFq}
zaE6u4yuLXZ@wBUXP)$&u;AYiR{~TfVT_I>ASRkM%UZ@3Cnhm;m=O*EddG()H=7t{_
zBo)qq=)HGB@X3S^-?MlDMfaE7rhU>W3~0U0m(vSOk7^KP2-3VO^a5{oPh!O6MZ-~l
z-vYuoq2O-qaj3a&C2;Jw*_k_>9AjkQxg3a12;C3{DXsj?-~4<2>TmuVhA!B9do;(s
z$$sm$&q7V-eM{?wPe1+@Y2A3cy(6_yW;j?>ff%6%Wb_Are!5UTefkrZr;pxE!wjYx
zk8q!<2=#8!>a-&2qo9_FQ4Tvh>=}8jZZAE9zd(p~9>OymXBG=?O`Jss=a032_HWlE
zcmY}_T~~@*V+|Z>X&!D96eIyN*8{wGZne<sBueI<m!fygr(oA`aa_uoW%Y@qGR1;s
z=2Hx(VDLPUx_RD#iwt#&yuC<_GeX?wkubUb`F=Og6L9PAAjUMqnI<(nseu$G3a2xX
zbU0Ed8W4?M9EvA;))^4jd*^=av@wvSfG*#hLuL>Hn$x6Rbq*csu1PtL2l{#DKD6Ni
z`99{ngE>EA##=nTubz66`R9gxi#?7((e9{IYZa+Gr8n+%u$SVc^j+v*1c`|dqDRha
zK(mLmdHr=hXTL76>^BtyOUe!f2ZyEa?|15Lr`-?QFdvBH_hxpGg%rSsm)ItTHOMLO
z!)4<SAD+2>d}g~mVL`ZDFPQm58l#Z1ZyHe|{pb$omqCiYxtKCxInOW97=e|e`W)$;
z0%_eir^YEh@S*lTEk1!3lrkDWy&a4(_~SLPK3|A=^~qh)VXHPewR)min*k;gly%K)
zG5LTBPw4S+vlGrdef=T&XE^YUep>@NqZQj2boHdKq2FGZ7?N)$8fi_iKI2-8L*hC6
z=;hvrBL@$Y-qZ>h6Ox30zi=_d+5L~lpA7~sXcG@Dk%1)lE{C9g&`pG4Vg!%0#`}He
z{k;<OP6|6>iMm(l#v1cX)N9;cztV2+?!>siyU;&pLQ0O>gowsK?+5q&%M7N0^>Sgo
z@5B)N`VUtJFEJ6sMWbU3Li8Sd%$;g<=+!4vo(X>g+!5kZL5DY|pI4&IPTT`Zj@H<V
zzwF>IY6{LqwmyI66eNl`f+Tp#n&(b*emR%k$Hz{#)CB9DZrv?}U7-2ggOrpJlwg2)
zXV|52PFMTAt|euLf|2|_t7EF|DD-m7hGK`{;blQ(j;oy=tPFAt{MBFmJO1KNe@lo9
zrBurOow9q5fZzz?FwX;Nc82c=OWt_-_yg;9@fh2()5?zF0yQrhPKCC?6fTMiPcI+2
zK0VQUgVt%sPTB92ay+yFh^yRPh})4mk&sj5_@67f=VmQ@6MBi@R98G<afkQAq)OCy
z2+dBh7<@ZmWPXmxIC${YUO9G;CHKzM^#r}+6JEo>e%aBFL9K;aAxfMDh{q&{5a{MY
zUJYT=5%IC7aK<6`fsPphr)RS^i?Ue=NOZ&`*i2oVLhHx1X$_Na6A_!4hdgy7#zCI>
zF$5(gc?cVUDUgnCg_MA#4b5}lPj6T)^jc@b(WmC8(eCdUh1!i;4q6+mmka6o<fxG~
z%FOw%rTBEw7~$|D_dHul^|8d?ApbF09M^?>yJr9sUD4>q4Kc#-gjNs4tCag2H6->X
zl&Q5t19I%V-`>0+OUkARR0&I(LeTL*(|*nRO~pWIm7^AdDf?cz-w&*KR%<X=Lgcb~
z{I5UjV^)h}#?yLNoJDxKUb$XYwrwLXnXpCna(AUM?loUxU|Tk#M(*L@cH14gA1g5@
zYBOr-s7AKy6$ueJcK75*@#x^3J>FKQ|FpF^!q{VGcY&id-tX13?7=N=i!1rJZE?im
z*eQp13^qJ6Hz)P)J7)*L$@K-NGbV@FCxjQHr>6yRF=m{0sjzLVPYdhQLfqWvk2z25
zgLhm_-G?V0TI*<+`7)pWc<4<y=HBE26ITXD0y(b4I6o6!O!PKwfRbEA-&}y##V|X0
zKF8Zz?+$g_0CTaAN}?T=FJIp&$5)oHvgRv_f(?&_4$-@mW1j=4Dt%OPTFA?WM!!Ef
zujF;5*@?h-U7K@anTcVg93JDUt}jrD1W|_hWJkB;bbOq&M|z65AksHlrP?4!pWW4X
z*q@MZ`?Z;P9PtRa0n40MoFOMZM9V27n}2qkgrPYNHX2j$=H=5~>Mul@N|+4jDP}$G
zhP}ZU2^&UY4p|W&$n|?(ajMKOmn)i*5C5vWAIi<?M0>m=b0dr{Y8#HV^7+#r`26`t
zAaK0D^X1FW{Nq3U9ryRIQ^X7iEYZcC=iwGLaJ^nxmc?VDd*i)T4jX7nT-I&Qu81Ce
z9)RLqX^fGOBBd32DU9CO-`_a)8{ktqABXw;uD$!aRizM2NIET~W{(ci4+wi4<USf#
zXM(B=^lbD8SFWl=`R9ehe_jlP5Qrl>J!N1qf|Q3QsEs~dDn6JjI=lLyDq=XP1E&e#
z?AU|P2?@~;<lu_?bFRdRn7V$#eY+<cAhQs7Oq7k$(YIk}bz7y+5t$>B@jYwIe3NtH
zYpacJ1+5Ls326aS@q<s5F`pTkp89cS6~u((1x>4?5oM<oA0TxK?tG}4i+cC6(`Je+
zDWaEWbh!{QtlsE-pL-yWnB`OB^DV0STlyfJnzJdup417+-5Vl2u8*fZI<**PW9d_3
zi#I(@z9nzAhBW1VyOZKOFUx|hmx%>T^vZrA^!zpN=x=$UkJea+M?)s@K{6&_;ueYP
zl3B73x@f}(xYSap&Dcxx0Y#;AO^NGeW4o+WI`{i4mw4fJ?9^IGF*<jW-6SqSDAuSl
z4?kRB;UPYQnEZeiRp<Nd95OR)9y2+RF{jprfXM)vs?`WF8ZipjrwiBXg=JlQ9=5pf
zqmMz#X_C+#)rBP{uIt4GMics133MD5z8{5VE+$%L_>W9S#=0bOQbJTNe|X`?A3yWq
z`RcI>V^Aw#gG48UKB28fZ3pTzFQ_F>5gRkzs?2yxoybB!#XIOIVW!%3>L_5o(WvH`
z=H??dM;~;9u1=%$)(M)Xy$?hJDJP;VjHa~Q_{xpreuvOWC?=6U2YX0xWwEIb4idnp
zF-De{7!pX!#`E(hLP#9#jpKOpm$LalshlJC``s<8G2&xgJvs0c84UAL&>DT7z`z*Y
zpOxUb(u_9khdN(08FM<D^NpVUG9E9Un2X4IcUQ`BKd+g4z-<iw88&C3GozlH4{@^Y
zg2~>y&+|AA5c7EMB}}oDg1YWu40B6sAM_q*y^$A()8%*WCRmota=Eaq8>KbI-Wk16
z=4R>LtUVkp8pJy!ltCut$kX=ZnG^3{xxatmpZ?(=`SaJGXswdf4{ZcEd3L_V|EffK
zZS?j|#W=*+1%eo<Cz1*|C+u+Xouq_ycM;voK^<-fE61Hy?`~_H2Lbp#!<Z9Wr*6S|
zrHf)I!3^(tEc-=qVD{j_U2qDBTN?G8QZx^qUyANj^^reuri+TYYTm`KUaW-BU{Zpm
zw?{1WsiV@V)<2yBVd^tf-S;2G=X&^$o{5jqi-FTJSdWU;%9sMlA&!)tJi1R3g(SgU
zCA~k=1O2N{k?0)CWfSpZctrJ7B{GInyjBF3NK|MO3K{Cuh7F^(ikXmu2cOmEQRkjn
zGR4MxE5@QktTh}h_5Jqp@e}F!!)%m_qjyZ(UNBOc5yC+|N*`tcG=sJ9`x+)<X=894
z#RvBVVOiot7G(wrc`<XQ5%ykPpe%}r&!1`C$250^(1RzlPBgT&PBcecbka{_H28J3
zz;7!CVs`bm*&r!gVxZ?dg&Xc0HucWZ(QxrWA-(x@sp4wgZCSW%D;fj0a`3err2YO-
z+vXJ6mX*C!_O+2y;`MfSt~f-Bbtng|^l2Ay`-fTcSpdz&LlPz={)+)4<Akr~VIb03
zmcY}8gpxhc(#%I~_Q9wFkw{)PVwf+#bQqB}Z>;MDM(4QS>Foeqn76ls+i`cHOH_0o
z{OcNd`Se7N#<nE#npt0-xUMTXtsLEX{x&AmJe+ngTv6)N_NKUIjt#a6KgkJsJB|bv
zJ$VbE^ixP!DQ$PYHZC9zR-Q`MHkxy{>QTg)(5QfG4e}b;@|7jO&<^8>4eLrT1IwpC
z%5@mv+L0lH4<A4CvOP0e=j)e$WWV2Clo=odFIaqE4U}5A-QEElic`0L?Z?hl9p-fb
zb(l*2shCzly7_2q5dt2;Y9@5+wAv}fqqMtqnsrJ{<mkB=h>t~O;1#&}#7)7&scP|J
zG6gTRP)BR(u!mLE)H&YTpp=HGxcKns9B+lAS5l5_m&Ce8wD?Sg;NlE-E*zEHO^8Xj
zToO;$<rGE#Zdb~g%k>GF!B;w6I;S|!rVzHbHaT$T{Z#{bNvxNZK8)9&|4g;c*ROx(
z?du!8SB4ErA4D6|fA6BdfV&a=5UBKh&_!t&Ed*@I7{XpFU4@hvPv#sm8dj9nI=6Ah
z>_%^1tOT(+GFKmm`It#}IYg&J?3crXI$3?*fQcV~;TOM0WQJ6oT{=7z=fehp3FALH
zY?*1Dj!?=>a}6UR)tyj`LBv=V-&0G~nRe+70cI@M3(L#POh{!GD6U`^rXDKH%=@W}
zL7k%Q=Aq2P#TC;!=|A@@k~DmqwP`?*fCc+!sEAB-)YBwqq%JseJ9Q#<1~N?tfYpu-
zVJN;cb;Hu2&BDfu0Y^fLy`v3GeQru04Qth(6`j%?J+&AUb%}{>UC7sGhI){iO$V9|
zaYQAYsjcVy1t+wWfA0Aoy%UB!+!d`gN(;1>*w#6s9(jZ~iOjUunF{IWRcA;;nA1@W
zq))^-1doyq!Nqi(;YY4l0pa+Um|A$<7IIF=FnZm;!D8kIN>^os;1#-gs8BcKD2=^T
zTJcF&>$-Y?j3~$b;LCo{zc_5$dt=*H_FSiSVA>=mWSo;jz<V2%z0%rXNlp?EFDEDW
zGy{(6{!qx;tT6p0Y}8t4<zQGRFM*)mS?tF_pT7E|c7kRu%R)+vI|qW#&rZoFUe(rU
z!zmE=sQUyx!>h-bd@|m;C6-mli(sFh=iGzn-c(&UioZ~1!w*Oes0CbmB+xzC&?{tz
zt8KzYAw?1yMQ0dG0usF$r4|(DPdo1A7E28A2Uyf5{O%i-QyX)@#qg<M;Nrt&xw6Dh
zT$Yc#*PS-LaNHYO4b720h9dwX)K>hE%#lx@{={><IK(y@$9<ougKx0SJd?aPPp3SN
z;t7#s@?K-kdWc?7mQo2!I}?hT(wpa*hv*$ot;LH&%>88J(JwN3=e}27>z(VmaJhJd
z_9?#cAx;g8VTvq0hPl<sc)Yky=g4r1Ma=m4;e|`Sa*KEV?(hF$&aMz>o<&i5rPX<|
zn|DQN@n;+H0^+DX=2+CI`-%>Iz@NX@b;8@NHCijwQv8~UbH?YYwb5Ip<HLwjT3DAW
zqRRa^IO-ew;q*cu_0gHz6o8)mY*uC2u56bV_qw;8vhNHH5a%z)$o(K2#xXx<)Id^?
zpOs-Vw5KsnetU{S!*q~*JJgvYynT?_aU2uDtfX}2&+DY#{BsZUz?r=G{FOFg{Ij^X
zUuZ={<)PSD5z?Y0oF4EAot}?xE_^}6v-bB?&``d)0#wPDmFo|GAZ@E3!nM+i$DDUj
z7t{U%;&-hAm;^#hRP*#tHJ`{RqVs2unfQ86l{_;}rtOlGdc+&0UnaFqI_PWw{2wO|
zhPc*7`pn?4!KlrL&^3=vCw!aO=VC=XTKHD6xhKWs6G}BWWqXSF&{NA0STvz|b)>2m
zN<W|;v{K0)6G&9UhR;NaVItV-fq~tI7mP8|PxPqvZ?go(43CNl70*%d9Tvcyc3qV!
z(@i3V>Kq6qOoVq3Vq)E#{?h0Lh&R2iRXRt$r<8uxJNny-fptYgpbd}v?S}_^1ocC}
zf}+W-bhWw9|L9(|+YleUYoH;zXzpl~y)yPfL#BdVX?t_8znzXM<FaLzoczJ)-RXzj
zqhYrt&Rl8*e01nwaMay{h5Dqxx=d)y)!fIuaeKYfkLHjOZVS5Kk7+j?m_(K(u&jv?
zeGKzMq~;=JniysNejU!Q8$L30*)q$Lh-s?xgNF&x&0**%`Z$F(K?4!bE_dC#QvuFV
zSK6r4rsmajN|~HiwtRWyZJUVyom152ymUpUO|jdUG>QvX?6hu8<#JaK43Tq#?Vh0E
zFD=BhFop}Y_7W-gPPB*;IMIOCr<D${=7LjOl8byo_8=x>a*ch)(>lzq5@(k(2vfXd
zGdlRV7l(XHqfv<=vgQlV>!%3~wuh#x_u|1V;xVe7fm!6Ted1XH`~5%h(@%d-TdQ|k
z&sPXaZ$>Lq5!`)>Uev%6I(>JcP}70nRFibU4o7I#u(5`fe2E;-u2%1*_;aU$QM}8j
zaS&$LQ;RV+r!J1#x$Q1u!eAUbeg76<^m|=42K#-d^~UXf^I~wcUvSnZ2kxwADF-R7
zeE#82gpk>9H*WhY_v7Ze8N7%+9RrA?L+3QNC9eGVH~$_a^ZNDAwAV_v;`X}p+<6?X
z0jyJ7^9+KgCl(dzegK71J;9I3OAir-x<4SC2b#>;9Wk;l>m%>L#+b0b+bg3giX&%Y
zgA_A)TZxNLFXWt(ICDU3rW2lGJddmOXxwEwo@@+|=5fWn6Lfa$_(9+eNpJ+J2J*J?
z{KF5d&mUZv*DK6nTnhO1<e#w*BgN>k&mj@w;t7Mt&agoe7dg9YW>D_q(fN;a)`<DF
z9@9<G$2WS!3lx169e(inPBIWooT?q86KrtKbT|ijSsNbv-TPxSFyiwn+VHcdbkD4a
zA##Z>R(8Sh%m^`Ws?L;JGf!0<z0umi=!L%DV04ztN?cczj+hYh?H2{2KUh(HUeA~`
zN$dSVu?+JNt$}~<7-jD7{@u?^ji589WWrn_2pJ_IjG*LY;rjG}b$cSDJO!-AsEry9
zau7t~)HMA%Tk>xz23oCi46Wa|sDVC*@SGXpA$ZmReGqaYEgl+k9InpPX54RgzP|bJ
zkJ39icy2(mF+KI3|6hBbQIy|CYT6j|+UGQ;Ixld0c+PA0Kc{$_S5ObPDMx1%VM!}%
z&OTUY@6>yzm4QU}?jFaEN$}})>eB&p6mt5eRy|d_WB$Buo>s>=gNGn5i^m0@;Y>3N
zVg3ca5jjaV#cby=p$U_9qcjiqNhuJMZ+6G2x>grO<`h}dMj*|X#-Y|M=Goda?vz>G
zn*{o3UZwYrE8(Z=cM1j{Ui}GK%7pAR7$GTWMB{~RyK;THvu;lycVb?MDN@Sa4|t5y
zeU$H)_dk=<!kRZMWYTh>7q|BKbqgd-Sh}#pl{YWh2~nqI&_7>q(?{GomfQ_+y*!h*
z%_md#gW3u{{3Eg#zc1c+3@#sk;4gmspJLVc^3#pKKmMLd$DRZ;SL@c&=+!*qLOSab
zSksDy!V<k;v?jFqEc7<$b#P8QOffRV-7!mFsJA<}_ZuMo?4ratsA`<Dv%`B~kEw1-
ztFCY8`(RY(8Gp}f-*@DMNvAdl`g9na!aCph%pj$eKmPbveEj?)`|f%1eIj}x%)pm<
z$jB5Kexm{5^AA7r`44}^>+ME5Zph7xEo}0+s&3JpV{}ihIw#+=f*F>52+kSH%WN6O
z<Cll)EIG3++e5@B>VuHBZKJm~RrE|+WDdoeLP8BXwGP5*0rMjH8-4O{8aSuoB0Pym
zhz~KKQxfCx{Xu57gAX8!QOPNhKP=>@XVP|cyIt#)GU&C^TO-QAqHC{uA2U|iP>rPI
zGZ_YkRv!m_W;u*8h<TwNg<ek`gMW_GPV2K2X6V;M2hQn;k9-K3qt=m!D6jWMj7}9Q
zGbGGh=&SmRvB2tXADP-D|5-wvL2)s5R=YlO1pH?@l5{#hVi0N_l;#75_T!+{9UYw*
z9XW`L2dwo}$oD%jeL>iNv1cmhumR7FXj2E&=bW3yX*aa_oyTyLMO}N%tdAdsau6e_
za+!_avTbbJlNYkH5UaHjMdy&bN^NuCoBRvV1U8Jz(?Uvtw^A?_SIujnF(Aa58x9hc
zNFq(Oe>dE(U)<MOs;AKPli!Gb17q?jghx}x`;Ln6{Cwrpr)M+;+TI`DQznFR&XC3i
zb@T~~8xk{gZ%;?}x!9_WY-vJSJXE15J}){3G<X<LOp#?-SYjk-oN(C*;qePkzM~zr
z8K{!BNR>O$Af`wPt`4uYavX&^Iw{P!(2y}2Hk4jHm`F^h^`NxQ)AgAY7gPpEch38M
z97HujPFN4Lt}Zwlo!T3<H>_7i@MvVW-sv~wMLhXWXX$bVd#Ij@SLVSs6CP*8CSvsF
zSpiG*+1t;TD?j|jU-J5o|1hnN4$qzbe4`UIa+G)8j~h>y7e0OXBPxlnfBrjK-6x;c
zQL`zi^GSqJ{XV3e$$23!nKf-(wu@5;=7KvDC4`mrVw9SZT38pKjP`cF&FOzawZVr!
z{4Ia`xBo2y{P}+4_`m*NNJDwLe&mPGe`G)Id?`Qsyp00glo(f1zPJ!I38OWvIpQ=J
z9LGU99GUw0&p#7q^zrrS0}<s_e}X^%L^&Eq>4fA5yBA|F4s4BwHlbCYey3wvx{NcI
z=y$u0)(GO3tTCGV$$jho=W+-MNEp1l{E;92_}BRO-kDm;KqaCvxmdT?IoxTUp5li~
z&YA7`qd(J}|DQrghz)<nVnWUw5}OuggS_N9>CSToe7dHRmSt*M+&=xViHRa15f>K{
z%R~<p2r0ot1CoYGT)LL#VVZO-ol!kXw@pjxw}wu?9SZy+d??OY0@I6LOPyH_v!R$s
z%UKZQ><D67E|52l>H%+W?>F9WZ}j6Jt8qzzK8)Ud!e5KkkvTG^GKakRA?xq=OfsFh
z7z3e;BNJ)P0THvAWg<X8#6`TOV`c#OcXErN>mcGRj%+kyNDRhfM0yB|<G7mq-Jebi
zhatR>@gYU#^9YJN*?RAk<6w!AI%$X$vri@r(F>r~U>qFxJGb|PS`TFSgkA}A%IidP
zdde*KC(?Ey-0qIJ;^UTwoXMmv%JzY09{glS;bk^8#Oyi^`$hHyII62o3lEGiI&n$F
z%LnxN6PmIM2WgbDQzQ{Jvdn_IkKy)5;9n$X;OXPW51*be>$H2MzCpk}rvqb%t13Ah
z_8;A-Q@j^Uhy)s)Iv*w^P-xhU&Wr(`wuSAI*yk+jl$4J@JaK*6I7(xTGGi`_Lul#`
z2yJrAeRwA+O43BsM8Q?A)`$CB)zzoHdro(VGA9}?1WjH6hrqXa^PX-9{_lsscFy4z
z5n|4B{=7e&3)OiY5wMPRysJNUS}(j@mMIGIy{a|1?~PKMr$`Z4^1_mmi$+@IzVF<R
zJ9J}7i4<32T+xu|)*0ru6=4doy7QF?bH>5+Qk&}|g7_q_-rN`4d-uX%RG*QJE8cB5
z*C!aI?7ZE+5Yrz}+lZRT%R*RuaK^H2R55N}?hF~E9Nn@N65IC7)6)l9^#L^X{+0V^
zL<ychP+h+frtNZcM{I&+w#x;Pz#6v)2X@plr~GxGgSl4X<I_jhC1XR__X^Dmjg}Vv
z^rye&umAYhm<|5sum2m*p@{JB|N4K<$LCMn$~#H^fq!WKKzaM({9X-QuNyrl%DoUh
z{&DiO)oGB@=&HQm?xgsY%k_y*PoK$YaaH#6cUbTjGe++~RK&Z7;NsF@f?7xM4#FW`
zC9RRP{Js+i$LP>y!uT^-iL>3sDG>uxC9PMk&mXD7BZS}Yckai5TMxq=E1Nu&s1-`_
zZnTdx#3t~*@AT<3xZQVJ8>HwoqsHLc0r_^&p6I4^qx9mMnbCaQxvGn8nY3AJ{((_K
zNRL1-l_~IQ1DUhxQ%d9z=IB`?#zYK0-dB~!0o6luZ+SmyuBjOin@!IdHYDc-b@xf}
z;=CNi5v%#h9@kZbm=e$7g_KuB0=GK2^_{oZSN{C-7jFBG3Oucu9KOv$XftK6_fG2-
zEsFIzmHw`|>Q+2|p!ZoM*khlWPzI+3@%(Hi->&(YJ#gCe&TEPK5F*4a;X}mtBb(wy
zQ98i~Mv0<nBc|*+iQ$py(i>L49k9JOzP^41>-2RYYzt{!h?gfcW=1_c*%4^1deUki
zkJ$*P3rW90#AwQnB&CE-gl66VFD^!>54Q%p9tIF^9LJnZ;HafNgp%j=I}0QmhV|hn
z@bwwJJRu>_Yr#gPR-*+bW%g~IX_$U?{g>BoOCVe`m&?NKcH?M`G6ri1bUI_+_*-ct
zl)Y7I=^T5Z9<BgQB52NBmW%|YAyf%8Q!Xj7ZHbRB7q%^O+wX`Oc?m2__5`hA->Th^
zct37fYjhRL?y;0<PWrp7>pZLjU4$hoVdg^Hn6bMvu*h#+q&Fk4i6yTyLCkdybyVm=
zKP@Zbo04x%o@1XEA1A4tH~4%39L}_1)V*-LA6TzMT}aUw(&yOF&V6t0j#wfo1-5M#
z0|IYvJ72%PV(nl{7q01vxIXjr^wBHUV3=*BdeDk*W_>=`2v|tO?a6JMwE)B8lf|eL
z!t}^jd_3w@3cIerJD6Sy_X@qgBJxjc`4ji!#?NoB9KiK*<@)sOo`vpyUQvIMHLYB(
zAGlm!7`<^TH)?kgOkTDr#+=i^g0OBY$Kj)W+c4H`<N5l;5?4xX+>RX?jwp>$Im(R?
zR<_GCpI<)vO!wj1kn;BCwxOqIet7u<FPAH|S6-ezvTRqPD%Yn!^7-SBe2&6;{ejE!
ziT|Ph_x#=8{T=uB#${=s1M7wmy-;B$-ih<nt#$Txu=m0bPgl>rK;clHKU`V1NVq9+
z5w>d}rAX@@T#}as#BH5=K&+i5M7B$0x#l^s?f1LRE_U!u!=XCEYQ}jjBOw@sM6HFl
zw=a~V@cQ*<>VESgKL~kUSuf9wVWhJ2{`Shjj!u}o^~TRX{~39Eqqjmm8ZmfMsF=r?
zTa0v34|~zcP>;@0eEKM@PV@qnWh2BHYi@By;|}uT(d?*O*O+ubYaO%ZgT^EgMOl(t
zL^UMVW%EH=tv*72tOpm#oyB1gu+yz@(g!XYR&`Z;DF^%gPThA#AKwn#+4UMgP`1^J
z079hJj`c!2?!3Kz<#v0c4X0B}DO81#ea1%b-IF+5=P;lBa0Vhd?A<(?*=A8w>Wpo7
z{vDYP2%XLl@wj%eX<a-IJ-jGB=Xf}RIYXA@3<r}5sRuejH<^fmG9;0<7nXHpjK-Lu
zKOvm&%r_5ztCjb+SF9J76xh@YONkL}%~9l5VKj%6MIMN;2(<3A7La9rKc^-h6wxsH
zi5hWIH)kQ_^np!S-;0_PP4c48!@9lL@P@QueSmsVvR<A^*Jp-x_R;)Y?BDV$zG*LH
z(hpO=^()qIn_4($%OjEDR4-k8jAd;DE0yCFh{C;A%DrH<!%?9$mK3-y3!h$|NI7xq
zonwGSLz9x0j7GP)xo7{}NSycI%Ry}w8Nyy0_tpqeiIZpVV~}RhO$v!9k=6%M5;+E}
z7etLVx~Dx2-@t2c+>Y*1#4GMg)){s(1A~n7#b^+i+`5utnq9Vg`LV$<lgUP(lWxr;
zf7?-M@70sRQl)9275^Uh-nkX;Zl5m7x~}Au7^73h;C{PP-wt&0cS4+E#0&!KmR(pm
z|3B@O!VaVJ^z?z34}aw8=_A7iZ(o0+?yr<yy!)79C!KS{Lk#p$C*0_8V7*sbt5ej{
zU~81|&fEJ}-uD~NmnS}d{*i56L5;oMh#`R)#~1`%*p>_1dSMKA4wNxSQ&pSOLfRHO
z3+r`dSt4J)d~uKdenZr4TWeZSjU0WDCUu}8OpmpPKjoB(q8?#AloS?+@760%*JtvQ
zkm}(-eF#MY+w&)W_=~?}T^2rl{*fO){t2<bKmPQE*Zt0VUr}{I;jw!be#$;dxAitV
z2O)3D`kat#+}k@bt_<yb{Cwpv|N0ZM7%>Mvetafp<=7jod54rX7lds+QcEL?vTl*(
znuz-QMh5JJ>+Kvm<Qg3B43#z=9yYuny5C+I!#M7D%CW;}m_~voa$dMRJ)<JT`+}H|
zLgF|Mj<Qo*<M#g5JKWwK;aeAST)xdP2$7f;(&F~o(ma)}+;7;vquupLLH+rRITK>^
zq(c?1+Y^_|6CRQ7FujGTy*Wc>x?$lQDJ$fdNXzVCRXn?(d8Bv14+afs`un}m^xO~M
z%suK?C(#VORcth;O?!I?#f0$itVa=Iju=Ag(=F5*b>BJeH>c-hj%uC}xMK{6?izq^
z;Z3dkBvuutC~P=oTWjMe#fzX}$h0F4j9}lIBfsy04*4b=^$+Y6%tQqIVD;cM>C@3<
zPO#|YzXxEwJHiqYDKDO4s*`G+NG9TW|7Konop}bm`B>$YGciQY0aG?Qz4~NbJ*k;f
z06MLEwHA6?JyQaC9O!4~CE~6oL}mft*DUCyfaa@wY6Z}#X%c%}j}QXdeMU;383g}%
zQl(9csNi<bpbrAeuhAHfBIwtw-&72YK4`V`{&w?R`##uP!$dj8^yimC#VEaFNAtj!
zQlV6?+v1)7HB-mnZ9gd8*ft+U8<R2=-j9QOsTj~rczt`HmW&zq+412Ub+BzK&({mv
zx>4>sOax(cql@2QwK%=OLNh~VyzVzovMWe^&`v_<KeZM={M5WkYz=F!83^KoiDpu&
z--6)c4I9H@@Y<&p256PzedqRmM+UsU9{%z~rA5I7`@UhsIgF&YLAB0a3ZpiXM9)^J
zV_IiiLlARzp^XhfN-QDJYK35ES3bP_z|;1T4^Mxjna3*M?r+@pJ8AN9;KCI*Kzjg>
zic(9bl*apO<+$IuZc2_qBxBXDa}1Gfx$tuR$mM!PbK>W>SCTI15JGDNNq+DLrJItM
z4N0yqNtY{6pMGFjSMs`GHh6peN18_7|LGqIlWXi9dhks%rG*?b+tbR^^@SyE+>h!R
z568|(*;W7Dt%^A&Qd+pKSN8(P=rmR7eE8urA6`E5@#Q01PAnmlP5JpR{+iF9{vZ5@
z|NH-q6oe%QN^O)jSb{@`%NmhnH1R2z>r-NT3WVHgyi@8sP2O3a20#AImF1JsF`lmr
zOOCWIE^ak>q*G$hi)UdVuI5g^QW5;Gbt#jcn**4J&BQ%*0rDa~cf*>GD(}5}LS}I!
z%`;q@&*wP9#3cHN@x}e^tv1di!1F*pA=ptXJqU5RxP{DI?-WyJZ%#M)gw=-hF~eJ+
zcaP6DjAh%Vm@RU>K0gu-TlcK`R?8eKu8%^*TBpa3W-nr5oRj@vc8}f>T9~v6!vnQ+
zisG24Wj~?bITQ(*nEN*ehKJqggL^(F9Qqh<AK~2QBtC4AVj%b6gLr0h^O%?!lx@3^
zf*>{%1<ftC#>^a;NrinDB)vCBTIN(hW1<yPYhg3YNRSD)p8`3VhxQbS`%Rm${5gu6
zdBoYnC1taU2pUL1+(|TThvy-wI+GQL&T-@2i^z!xozA7vI`ufH`<+^RMox^5kf<0T
zFo!A4si^AIxX3Kra94?mTL4X*2GTePy%WM91{Wmjq%vwB<e`YTy_!j}ocBEh56tQw
zoZ}9Pvcn#y7<CY1V9m*OW%K&j`OZBooBsmE!2NcBDaXB14<Av>@J_o7qu0(*<rss#
zR;b-m;9G}M(aXxR_;}rY=4G21!4lFtrR>yd?E9VLsANrODWnv+Y#SN_d+*%t2c<Z?
zo@3(qa%IUY(#>z^XpGy<M|heK&Kc@i<$EoJaPh>dnN}FpMM;;uu*?95(YqHGQ-?Cl
zg<`e2oyNMyFqg8s*rtEWqORwJu$k)FdguPO^Zw<H``ewYkv#Lp!(}1H$fCh3T@t-P
zuMWw9_~>7CO2y+Sr+BVa-fypH2^_^o17Ef)8oZJZBTdM@dz`KC@Z{Fv+52RfFJbhp
zXiODv_xpO3#_Q{ypa1?V<$kBWC@CJ4k8?oGa8K*9UU+)>z|-}aj&Zquq>NXr^#>Q*
zYVoRi4gpH}%ID9YczSx~<>e!nbz@ssqDJ0dKGD+3*MIyo^{6yD5(3Kt&0yP}dETz1
zHL>Q+Hwg1VK{j`fHXOCOynOWPAMs+%e8i|qWKA0}_!#OnWiIQApd9;74v`N}FZ@46
z=X>hmAxABq6sX1}XVz<E=t>=fyv~V|(P8F&->K$;#!t&L%Ox@}meohL`e2hmZLTs_
z%rywfH$8@TdjB~e_<LJ%cI4g8sA}J&L;^7^ENO*ur`|j>o?c*D-o*Oc)Lhi|{`F_p
z?ZR@oPU~r)oV@fr1kOX|A;=n?(JGdF&kA8?{tu(o>YZ&l7`+isrTauOdM&*7O07F<
z-uUpt54=3T_`HtphhS?5A|6bYa`GZ%h7OJ1-AXuJI-Cmq)8{^|gYpYS{kf6bbRV3w
zqsmDDcyLiv#bRU%_nFD$kM1(Wh;^glymK!*)((Qek`kqMdUt(R{(n>VuRpe=*?A}S
zTkkC*_Rh?6tE%o|ix-I=!JZj-><=>J4+e~Z|K@=W*ai&PGqzz_BT6%*D3KDIWOrT9
zB{TPqi1+FbYsJn(0wZ=AAi&<h>8`FmCr{?y5%I3e^ZXv?h)&Vq53gQxTFyvs&^9I6
zVUC9(w~(e`@w%;dd`!eR5u)lVO4(Fw+lQrqa@(nkNhyzjB<}+uO}Fc!)&j2K?QmD#
zf*2dXRw#HCetH#W(p2DFaxu+lAjP4py`XEBT-3IAT27qiGkU2N4OH7!+P)!1^>a=e
z#bcG>oo7yo)3T6apwvo=o@t6&5IB8)idnFmKxm$7amL=iq!>1vN_RB>K`FldZ%j~P
z@RZP@G;IvmD{}eB7#e%r1eRrCIo;{~;Jq~y3d#dsuK(ZU9{njd^u9Gj3foq0hh3|X
zn=%D*R~^CD)S5x*EG{Veu{L^Xl&oX6>z3JaRWhG0o#?ypysbKB??YAZL2sb!vWka}
zYT2^Kn}K&2!+6YU2g<$RvI;P&_xKd)t?Ed;R(h`VG89*dzNvSo6Vd6TU>sv2tJZ4o
zT=PyTjU<h56)k3*qjs(Kyi-Vs`rU2!-c;DMt%b|AvTp^MEbkj(@`*VuTDa(urz@B?
zzz^CO78K%bH}~@WxZ2q3h1x39eBylX@G=vV?$Y<H!5UA`kLr1i8V!A2KT*q$6BR~z
z3keF|+o<(o$%>7AZ>&!{PaiMjbtCwh$LDA6E}!sjVc&KQWSVBuG;0`=kEA%$!v@Z?
z?#lDNK0os`KT><+vh95RJHN}%zx;~xvap;Mg7X^byv)4qcSwl*`q#hY$M1hdDH~Hv
ztbnA%`EFrK3*t0AO%M5IVNRJ*fL3|^>J8^bcWm#kPyG7B_k8`cF9>bmfj4(|T-VJa
zVf8@1?i-Kmg<J}cPmc^iv(|&h`^u0`85M6kQJ(?{$EZ+;Ml5$Sbbas)rz?c$BH^r%
zDZEwnx-O-*u0Wz%sksxp6`}>|*4g*p{HFeg_zKS5CIjGf$7l!ltB2PR674eB@~Q<<
zQya6xx+RXjxwBrM`R-SL&HbA<oX_{rb{Nf6>o4QyTXA5Aj~mc(W=u27ydd7Q6}3I*
zoK?HvoZ4~65en2zY~0yL<L&EreDTFsIOo{*E3MTVHg<HK`nY;@xne6uwe03n@h;xJ
z?l8-%Lk|Eaa?3CPJIteXgqE@@`h)4u=$z^xUSx&0Jv~mzbl^0wsXK~2OQPu~Qt(9Y
zC{@3QAi`;0c>U^2-n@QGbc0?u#$IScgXm-ohC7hK&F-6~2j$%zDV?;yY8k28&fgp#
za;$>J%g>G6{y#B+o(#_zmAn^%bR6A+x~!nU(Qw3NCeE|E*>bj!B)tw|h`Kg(p3ypG
zD8AnNz&s^Rr#oT{w3?v_HE-1QLftn7HVu6~;186t5Q8S+2B%yJWiMDJgeBsNVAC0$
zh8?y^1+mq?G4U}Y_ibhJc`<mKp4@i-jkaIdo_>IBql(8b3(K<LLn6&Hr=aPDFBgLq
zmY@HN|I|JDQ*P+Tr)Qp?FRa&{dM!9(fcMtfOQnMPV2{Jya@{>v{Msn5jKQVU+rHTQ
zNXd;d3j4N^cP$3I^9(PH+%&>>QY&E$o;C$S(~)DV`e$1kw|L+XJl$iJd8ZW&s3DZR
z({rch3iaSd>%PhxFv2+x&h~xY-(V$c-oeARx!WGA0}S<-A1Z0@gGPjkqoSQrD^{`U
ztwe~Cd0B`lF>rWiV~Ar!>eO?K4^&gtI%kxxUdTn&s$#kDY9ksdMC%&TBONZ!AK2Ds
zuGbGZ@09sacIIgzq?u`6n9_uFA-Dp=v990g&*UcNSd|Zut8w08yFL>_W8JT$5SdPQ
zEax+GipV}OfRZ)6aE!rredhV;Bi?(iwIDI_dtd#GclWPZPK(m^TBTao7=q{Re8=hG
z9p8O?&;1mbqX(%3pA-{3bb52ywSZjn&Xi{67@<u1Imd~gefc#@2;}enBTxCtsGUON
zd|KFd%?tnV_&t~H3v$VPe178nhxh#W@dvhj)kAxlaWaU>F%gI-M>-|K$q^=xUmPx1
z7(&gR>easKeZRq)nt;O-45uu^a@<9TUW18R>+CKQys{u#fF0GA{5HRw>1n(u#0jsO
zfz$cS!^4?1SB+uyj@_%+t=S=_9h2@phfV3W?HT_#aNS13tsz6Lr~=08&|>;XE(Gww
zbN}#)yVJrZny9%qJxH9ne-0MDnjZCD_0Ds5_nI%>ea(^<HWM0d+m%|gK7gI3qDm<w
zL;Px8qLaGT6>Jo-16&^|HubRm^fBxmFVCJK+EKb;<7Ar@yySE^&50Qbx;ZR%J|yLi
zpjJ-j^f?q`VCt~#1@G0qJz6LD$iu7G+}}T_qlyaMX)zn4raK<{{hsZ>QjGIyJ`qoM
zfU4ZF%mOF2Z{e6Gm?lB=XO-bp5ZKi%U@=Zw6dg?uFB%N3um!ig#!vS+uem8L@48YN
zLu219o;-AQ<+NJyK!~1cSqQeGta)ejO5QGv{dy#YsaQ2ceSd@MMW&c+3IPnA;?WgD
z9qfAUW}+DOF?cH`kZMh<Wyc`gXpt`BIZOz8Q4AT3rjroeLu-<v_ZM1+ok;Xf1%#p+
zm>46IR|4+wIxzgd$?K=B37!?^(rdFUYDdQo{Z@74QwO6q(iqIqGoj<Qc?z6n-DU5!
zQAZ~oxdyG%I*iiD&pPHo6jbNL({g3KY<RzDq_B6$tm2Cl9S5$qmS+1zN1D8b8jS;m
z3DPt@P04aiX>e69E)@jY<Y90dHy$cc*L8G$P{l@VG^ZH?&Z{_u)(AsIVBTvkxku=^
z8mSEEoxK!l@A%~LmeSS^n3F{qm!j`MvJW(8&C!<8lonPthIHaI^YH2|^L$nTU=kbY
z9TB^t!org2V^<ePQSw>z3&GELpYVP{yeIUCcNsx%fansOJoT00KYP;@xV&vJ1*MyX
znXkV5iZ`!b5d`LK)$e~Om}<06u9>`Nwsj--%A30f=9rjbRBKY#_vCrsAb9S4q-Qnr
z*RnGrq)BONqc^twf=t?7UN28PT`#QL#{I(s<^C-$1TOokKf^o`L*Vu6x6Jo<>LDL+
z-L61q)Xd|@_x$+d_oQj!`;Q;__PbwmegBcuG?8Dw;^B)g=(VvwR_>+~_pc*yNl0in
zw5js~Lq&|EY6mq8!X=JltE<$v0j}vmF$}vcP805$t#X_Tx=XH={kL2|XtQCw^P~`P
zUN|ocX<4jr*68xdJI*mMjftFfrO>*al{WwO&Jp8;W3W9x;eDWWZU>~lS;?A(Ntw1z
z9IxKK<IUSIm4I2B<<Y-b_cYu{^s^Z+qQ@8_uU@^uhrm`cYu>qTSJw4{JACs3()F`#
zwc%ng7ljTOS{sb0UB0s-fpOq^xhcXq{&^g4n^>)(2gtDka87RV+`|Oj!-@gTdeFb)
zSj#(<m>9h$dQFH-A+pKN5YZxGig-~=XX`~>I34AD%#(G!9k*_pqvAX<&G<MeU!ZF^
z*r?5Hh~1z-FKz*KPi5NBxMoDyOCjd0a5+(!n1l!wM$h)#Rc(YzXVP+yOOx5>iXx-;
z-41WHyC3kQqc0Y;VT&O!Mx)oEZnUz~swNpa(S|U@M3~PiGL0(IPBGl3zcpK|d$*#;
z)b0wyvg-zOhgkN3bLvPF(Mf@RmiFJys|oWn{B*ah1~aMHq+0(@agmaXx=FIyO0yH9
z3%AL{k&W>JqWVw2eo`^;e0f#?3Ql*KelVodX_TR$4sX%JlXKiTRr-5@Q*^w$oA4>|
z<5t<4Mn;a(OjYiv+fK{&Au-ftAPx+e<6TsNnmA3#>Xm6u%xN}3OQyB)a^V}&5xiDq
zUV#>E46Ydpt14cDXF5$fRuiqzocA{wX=|NU)UWBpGuv)Uy59V`7KUij-w5H>g<yNk
z-WuJ!?w+Q*xga2eIw~Gj`#N`!zbd)bTcAkEh3DrB+q&b7C6Mz@^p5*!;&eKxtwsiQ
zWX32rRdXNR)U~!VtjZEUEw6Zeen%B(y|S0goKC99m57Xl!!xH7GQ%Z3F>IW@=Zu?X
zl62_G`@UOFB<JxCQV2xv@uxG=2L9R8YiC(b+@0^3=`7PstBp19WEqI)*ZuhPi4T`&
z?nC71^31p2{F1tEjAdq;P6VIGweoy@)Q2KEs*AQe4K9#NWtxN-BDpt)6NY!BdEp!q
zwQEMk^YuclnVvTuKmN!!-~1ImCbpc3rs(bre*X{t3IF)(-_y0hAAe-@ooNnCOH`YX
z=(Uzypm%V?<9*anl$%*dZVx2qoY}T2eR!I8Xy)QD`>g)X7ouFZ4eM|3Ym80`fs_Kt
z2bR-`Wzht~k~ezUaJ4WA#A%`ktRi?3a;p>rYKeG);a53C#WmB15S;3N^g*ci9ij+d
z5n(!=na}q+R_;TSF<W`r(cAszkHgDW4u|#AIZ?n%-g#axJUzW<-<~aGEU9}%#!Z*<
zk}q)76lN2=jf$Nm99lOx4xF<7w?<^Y=p_V9(5&KBZ%T1Lx)J@%#w^OT@4em>@?zcV
z6r+acI1SrMF_2QE7wtZK$<*ya>E84Ll@xTEkmg0fTLR7}3PPw#1;ogAwd~aWO3hbY
z^_UV~#4=`vvjW52R&$6@T4S{)&_$hw>U*L{cwNJE2ouZ?bO{{cLD%b*yzR(nq@X{a
zYs!OZmN<LMu;`eWa!~e8uREhFr4+NhM(2ssj9<>g<(_4l5YhKg?DrmFPC<m=#lnSD
zpx8#o4VCiUY?Mx)5nT@3pQ78vD~`S&)LdI19oM&WLiCO%suK{qm(=}H+1HilPme4y
z;+Gl6Wcc<0qv<NM|2=R9y6fEdE*1;xlEW{PCi(%BizEj#gCv6Moy9x8x<B#eZXx&1
z({4Y=wJPU(45AbLW(=Nl1o({La4|5Q7tRk4ET=^e)~<V{=#}PmsBDMP^MXrwZ@hA=
z_<iuCG*L=p74voq+?{9c?-tUeo_?_prB{ud1P#VvIOaGpg~Sw{lDfuVl%f*?5r%I%
z0`{We7^PU~P1i$fo_PB(ad(cS<Z0*!JFox~(Tk{$PODnjlwISWbFRo}gj`8eBD#Q&
z$$&!oKxkz}P>P@i1t}n9G;mX(*Ul7AoKLTr(n2d&*6pGKm08ok+UP9HSw%ZG?rJTw
z=Z)c&ArV^Tx~<eQIK`RGB8CaN4w-^SmW8%g=6R;{!m^x{SZS7|HE&cJV{_b}&jfLt
z!^Ah=f5(UY$`7Bu=bL}`C-ki#;)!8_;i<K-m5tuulxBkSoTi!6G?SJG?oRjIE%!Ja
zd5Sj(l@F0QO-#$o)8jMSwlSIp<9z(^o;R=Gau#8Bq7acKQe|+zoH_X!FM*gQ9-n^X
zR0Y@dp{!v_;9X?2W3Q`sZkK_Kj&$1xW5G=JxpZ@`h|wPVQLuGQ;T#Rk$@mR#=zpj$
z5r`==O_AgyDJEL$j9MvqrR*!dWhMtcMT)bXxexeKa8=Vw(=;;>axIivHEpvG6_a_d
z5U*~T@IK(JD>^R=X`X4VbKR7a^1Q8Vt<VqKmh%Yp)`cnR77ex$p=>M9wUKM%X?-Tw
zmEeRu?+C&?D<WKi5amb{#GpL}@2Q30r<=f0hQ@Dyu5;+EGu*fpgH9|}&pDk0xR+mk
zbn~`b5qT67_Dfaj+F=Ig=s}~7qt}Re?^Sou(F0_?UWinrYOM1VyiGB5bRVY#Xos8T
z1y>5aHWe<`LdzFgS#??xj#0S71x=5n3vLXiX)+}B%g%h*zMbReYkG&M-QO|7K;p~@
zvx4i&#&um;uUCWe=$fHco!GW9sPzDo*?P7&wW>Bqq7DQnLmdS+^*uWKG%ck0tho--
zVN}|#5zX$lK5{Sl50(x%#yE_P>XtOk3+8TVtx>9q0$bB0REH>BY8aK$vSMzN1?Oyv
zr8f26nyw4BjmzbUIYk_L21T0|1+CF+>il=Vei~<B_Vgmy6k2U)lOvuT^CpY}vlD*4
zL}F`Px5~8)COVUiZhAF!Z7yu3(+zRiy{a;UU3^^}K6<(jOw-If>u%k74;?b95$K}T
z^XRI;_s(;FcgOiOGn}xso!S%ts@15fhkI+c&LK@ROPcZClUt)$*G5;oachk|XGUuZ
z4?)nBJSAL70cSqn+zY+yZqfx0y{gTxH9~Z}KA-qUU%X~{bw*AnE;gzxP3u0*6;}mq
zABv{kR^3nLb;AWwOWl0W>D4>Z`5uXh#-P=mRx28z(kXeTWEHY$r=!pnO~3Mn-fV1_
zi#jx>#x$QT&)wrCP@9s*Mr}w_7Qvjf6RN#(xo+Si-+cFL9!_UciadV&ctf*__qgbB
zQ^E(OJ32ZN1D(z#uf!=b`&T3%V06B#ndkT4@xw3w9natY$i5cBbkCdnpVLIx+OIVz
zr8bRPPRe?yqatpwobNc@KPXJe33G_J+Ie&Lz?-`>lMh^DP?}@!GzM{=R2&o{*b1xn
z!gYIQ%`2rVA@S+?nQbdvS3Ti;47aME-e_eYQV6jcMNuC_A?Sk~uVY5zuxQoZkJwX-
zk5!wvioU8n0Q_&afbehxr5Ir-me@Oov;ws#``spaI;u~4?Sz0Yk!1+t_2AfZ))1MY
zUHB1`eWc?Zg2CWDA)J_#T87&gT%SJie0?GV*WM{2NC*TUm_kyH$EJZeQ3BD)O$55H
zSGIAbbS0odgJZR5YJYK#d0Lc7ZwaJbw7{;tvR7SIsuj(245Ep1;=QhtY>J@9e|J;d
z-)N`a-&p<w$MBw7b$qW*8fz_VVs-e}hZCG=w;8-6IAKl`t#&rabZqkBJCV*bB^4Jt
zN6UqBUGY8{mRF|?ay%n*X4FkduyiE&o4#N;r?%^%;W;WYZhFmkott^vxr>3I#JwPu
zy}h@?2au>uPc3ZQ#`E(tWzWPAAihKg4<oATNZqEcO$+Q=27WlYS1P0c2h>Ka7~A*E
zGITZIyseHcOhD{C2Ge0Uf6L}^&Z$nycm`s^OMiS1wVKFX*Cee$2$-G#g&tum0jGz%
z_l`lOceOwcY4lO(-oJ!%*>zKD1yqOBx)xV2w$%Uh>n9ZhAtkzZ?5mD6tA{B$rc=a6
z$JiV5K)Lna!PeMv#izivb^4xpy6!x#yB>bMYizJy)q3%IVfH~Yn@zzPW5hc>e7D}o
zxpBGd^xl}0-9TMc+^1>g?%{!l`+I7yuwS`eS9;M7Jm*5*cgngUh7ZQZO;rYaZfw0P
zamo#Ra?CyvykqUE>Q~l*AvDpQeBBKKg6{3QF%eWW*(pUQ7xS|4=FMyF9?tB}u|Kcu
zT?vHFIRc^qL(WsjFxyfi?}h8SbACAU@bHGa*Y8;F-)L}8S1K7+G1jtEYo(Mz-irB~
z^`@@1vgc34v@q`OsePxkOzRa1j(M4x=b3%?Z2N_h>+QieC(kr5+}+(XEoXA6JeEg3
zefWXzfB2TKUVp(dO_Y5nm(19;%X$FVKqtT6uDGb5ql*DI2c;FxGke*&*34dZM%BW}
zIia<k%g2wDHPa;V)mOjI{rMeT2G{kfd*ack2-P|2=!_5v=R4wQ;q{xhO!JBFE}!`H
z{K(lQa?oqaBP?;^?VC4@2&Z(<7jM4gyxb!}xb7R<(-Wnx&<fAbk9>Od$ou#2`SkGv
zF2Sh6s0Ls0%CxAptJ9|IgEUQnKB`{hDKLUAkafRk<U-U16D{DTdyA*;otza&b9m-|
zn_u&kxIf)lwC|?kJSVzWZdH}nIs*lCMRRnF-YX?;hy|GCtZTZ%L1h12^A2c7uezPc
z>2zkE6v%bmR@U`GcTl~jhoDj4Hj+gwkHnp5dB=5q*dOZ%9lx73RSq7Ra^U;z^Kpux
z28}`2enZznqa(GFYf*^QQ09qZ^;O^Foxds2|8{`SvCi<u^S`+z<Otn4GzbQ|={gnl
z;=tL;X(&0+i54ZZ=xPgbs@OlYZB|HF;5X7LW$o&$Io;o)>}d_@8sRRzAfoFCamU#v
zbi6LXL`d_*fr1awNS<TZ)HZ2QR8;U%H!<H3Ww@lYfK!p}^?JRD#Rsa_Ily1eSOq~;
zL>b2*yPdM{x^q876vkEPFDEeY!uM!HW9j2?45>Fz5$KvlY?tpvEXr6LUT2`XZkE_1
zbK&?iy>)t3L1bwI-v)jRizz>VMK;SgrV2V!RD{0RTG`kX&L<VTW7lJAhceXy&W^S9
zfBN;4ih))J;vFFh-8DsO4xs67e|Q*Ocq|%&xgzA@Xlv$&=PR!`@svAz(;Fv6?VkG~
zjBy?bp3y7O2b>Y!N-1pXPA(eHdtD1#E*jJ55^3^Q6?dbGMU4d=okCXx=g`Q>J@4%6
zMp<`Sse}}5jHGDcB?Np|6>N&SRu}`GoulSz;~B;Dq!9GbZ)zv(rgREEFnfhr)Y93u
z%w<>qw1h}Hor%+;33X1p$8PH7j*_kf6(sGo@N~WM>2l?=?=0?2Oo=%qLQukDYnpj5
znvU$YZKvej1eHRZb)w?ESMf_NT(?i8C6Q9%e11(zjnnC#^ZCq_CW80Wy7Qbj)@^00
zJ46W4SvI@=oJ$IP`1Fy>^Cz`F#l)P{uK9HN$f$*<Pe1bJ%bx>KO5yJQ#M@V|xSM8j
z&HQlr#C6{wB%am_Py3Z0Ov#^frB;~I9o|L4?0NO-OTL_c2cHsqZM2%XZWpHI#GBV&
z@YT=0=HYZtaDnf(M?UVKXyL@u<1?=xPP{&yb#E%da+;XY{CXce)d|&k_Tb55Fi#V%
zW!``H4NuR{Ou=Yi1ZC|HtztWxq_}UH(Uo`5eYfdV({XSU@TgtZaoZ*0#NxHB%Z%g2
zhuky!UddbjDYxoBxYryM%e&>mFaoG5w;NF{1~w)&Ad6{`qI0AakPumOQF|O{QDqPA
zHn`D5brkFIY2tidc>C@PTpwI6PbMa;`pnWdo&yIr#<-E%Qk*mq(8!1IGIk$kH5Ac=
zQn76LHoDsD<})EA6~WOdqu)Bu-aEr-#ChunAW2PV2NNK9%?xq!QUJUh_B!g;wSmU8
zkE1~FYNP8;e+IRrVR6eg865{5?+`_*rPw=NzjUc2I^L=1)r-^JbQhacDV<MOVqP^}
z^P<++nl0<4QL5?>28JabitG((W?4>*+Lg-y`d;d{rl@o)21JB@1i}f#m^3WR1-4CB
zlJWL?hpr)xk5f9{C#IP2F)HEF3GPOab%baZ^`UX@I;Gc?ld#+o&JQ@}DP?0cpcQ(x
zXSf^qLnGK#V-l>0^WNc_u6#YJBVZ^3yF_gmutrx=_ySl^o^E?pXCPfC0s~zqIkA<=
zz`3rt>JXThGxO<8hyhToNNH7pNHOZ4{|&{*Kdl(p2lM2JUdUZ3cr~lFFM1{CbvmDl
z+AhkjX-yMaEVS0yt9tXN;7LIft~9$L=;+w#O@T40TkF(1czS-+%G@F;8-qDH=Gl>3
z!41)3#5;{`^kQgCAx5t#PV;@c(b=ynrL<et_k5a}&u2g%;LZd#Ax3hoTtsoq`53tj
z4W}uku<e;<#467blB7@}-CH$kZz{w|f$jno&ptX0cN^U~=XFFJ^yV3@kXz%rYgo&c
zcK~YFZarT<(MKZ};ebBW(ky~kM@KP6=6TYa-VQPeo_W6G;qD7vzr?F*8e%kHijKL)
z=!Sz;DpxmVx;_vsrc@w|MqaOM+m$zq65diwjGkF9A8Ff_UKH6XKJzMgmKd4sMwZdJ
zY)@R9=>D%(?U)9P(&=Szxm?I^D;`IfW*#11bH01vbUw3}ji<|nWnNgq%&Ui2Je*G~
z%gl#*;ZOhckNI}bFwdO4Cp6)*ue|^8fzc|%!M0z?0uQfV^E(f}k2v9%fAuq-OTq7z
zC_)&T#MVVacUoz<sD)^0N<VDXB62addr^LP@0~Px1!Lu|`4*x9MT&E8jBbS%{+3y>
zUU#0K3YVwf?qa}uM=6_ijsZ8SdikB`C>(vDGltsXYQdTRJcdBCE^IW5FmGl9^m?G~
zRc(uXXcRcLa({QnX*p4IwvoKe?QRTwJuH=bZ}z{TfSMQr)L=Vzn|B|bx0u4;x^uBZ
zA;iS8+^KVD6zxvk%fA1Z?ua!mhciG!j@%6?>)hW=Dm*|m`p+-_{+NOszsGy8`W5Sr
zwTo5F!LVY)IeorJ8#kM03KP?^;M^e4@x7^Tz}e{9cHrV1F=`Xyog+*WBTi<WJMg_a
zd1yy*B?A{D^K>Fjixn)I+aN=Q+DEz}HVF=&#l0&o*y&``Co~VnS$0cejLM!5<-QU3
zoVU;7j0@iI!IfTK1dSo;3TX7I0@PX&7YNHlIzJHRlm1-2E08F!D7Dezvdsc|*gw2G
zKEI%2e~zgQq@Yc5HEZSJBpCz5MdCCQJ-C{Y?ijhF*jgjqigFD!2&xkV;<B*3{et=a
z74xznq70e6X~O4$<bN^M{@t&iR1D;6#<e4DO~oBsY1SoIP15r^66zg_CFm{5GkMY7
ziFZhC#NbF#;Ss{XCmSOlH@K;^2ZPJDbGc|*p(-pL5(7>WTdDZn-NuZ$S1$X;xE%X;
zVVV+dI9je$hs9G~8GZ2X?LFtah50mde|J)+Lah|**7sT|MI$ixoEc+~4wt|%(w}#v
zI1@z0V58ZAaO~>Q5A+lhcguqFo(kluAv^;kaK&R+_L6nIm#kRhKFqs3xNH~tqp&~f
z#W62xopWfF8$^inOfy^0z;>W9FwZk#3?A;^vds6SWdT1kx>e5trEF~56We+r?>p<|
z!g^him=F<a$qG^`mEH^94f1|r&nwLybgru}dep+GjqBqxAAbCvyLWHc%UjlM#XHa5
z3hR1hy<V97!f8IiedF=rE$`=F>jQh=dH?ON`SE<_++QK4_I>^44ewq(Al~tGy|QhY
z;5_GfW|<};j#u*??;hT<K7Qc)PapUuPE0t~vh&?Hf64RvA8FIfhs!6v634qY@A&%N
zmrNmWlF0kV_x$m{_<ut0Ja5mpuI~E0F-?({R6*Wp76S8P<q{h)3kpUO88{EMRaLPg
z#1u$9QAa1o#?mA66mK@SDMc<)Df@4?n?S5M()Xn+EmDMl@i<UuRI80HgV386U6@!3
zZeq_vS%18@!3os_!H4)S#>h0yv^q>vr^LP(1H(Cr0dC;1i0L&u&|<>-1I;Tc@f2<@
z6c{(k+(89&qRoe97dXT+r-Nk-95KCeoF2U+hOxbM`!*JKw)W8X?Qe?DG`Od~Jmlpl
zRJ5B91e+2#TQw>A*&e3Uesr)9M%N}GCWZ@yAWY6NYG#xb=@o5R_2&&SsXrgn$BN*C
zBl#H#2R>Mn9rM0}VRWa{J<Ht#4zHh|R1-sb!-{*=1=x?HD-%ePh$-GyD6JG)t+>(g
zGKk(Wr=Z2SB{ue<E~06kjo+e;s0PT%4G<JkR7<c?N(Wgd)oCKkC%RLivh<B!c52Pa
zib3O}oBjObv(HIwi=NWe+i^?)2)ZUxt%NX|=roIg_z9mDb(c8@Stl2m7*K4Qqt}mP
zv`X-v)5B{{_iu=&1?P2zUR%YDN(^4%c{a^DF1f$!^^=N$ZO!V-XCT2byinSp^+C^-
z7AjLrx+^z3UJ$rHsoP+yjcaYx4z70{{diA6E5MjkG}J^D)IDp3nQP;6T`9S$w>Eer
ztGKH5&USa=G)Mg~OChWqO@z}jF~y)%zlhT5dS}a(TpB@OnjF)T5HG6icltq>stD0-
z-`Uok4<9~peO_6n2{)U+wWumSg~&2R8$+s@OgCBT4sfV{dP#}XG~Fg2x@**Z<I$OD
zb*)j!rKlxQ`XD$@L=WOmPoLOop^I6JWI&9B5J)jH86vY*egB3V_~`L5DBGb8LR^S(
zVV>`BK_9Zc?3BE5S)aH(e&G7_iM(%IE*tA5bGlE~MRiKqHHx_I>}9849(jKL#N+9X
z+EusEOEzI=Wqp3+<B#7^MrJxq%-&JEQujVRKT}%cXRjVu(#&(3`R1FyBF(do(pus2
z;}1N3^(E8!j6}8QoK6d`?#~>2x`C>n=P44r(44SLGp|msSO|RDS3Z3Bz}w}4$LB}h
z|M)#^D>yf?w?>#}zIgqH*LNqR58mFt;^FQ!du{x1`GLIc8c!QMP9mkKYr}hPK$&h2
z<vnLQ9_K*}-&%pT_ZOCX@5I_^V%JP>>Za*dB3BjFofpr#{C0B&Vicm+QS2DwhQ9Tv
zPDs$5e;I??6&^%@)-`%oI!LSf+z5`=z2?W;{gXn_&>qn#V@#1&GI<C!XL>8D%LqD&
z9O5Webhq9OG$we_?z;4L3+)kf!gzG-ee~PD7%@(M?d<E7X-YH^og{rmpF5lh4sHj%
z3MFl{Y&&I#76K|DG_3;e)*y*Pcl*N*-y;UsM*lj^3^$uH=+3&=LPc{Fg7?H=kRtP~
z2e15xDR{aYbgFSMbRz0w#K%ZSbR}xp5-~<xaKtpJ)@A5(bMTC`MqKWg&-eH+sZzcc
zLlj$bYYZ9$6>LhP|8YCCo$(SB`<%)73RQ!hf>*k13_-=oJ}LlhnoZAT&YY^e1Tp(=
z2#yrJ>KO*O5OH3iLgE8BPZ?0_PTel_oVB@=1ftxO+K=o96<K&6%>81y1-;)^^?J?g
zKa0HV@T;9)Sh46spjUSbda@G)oUV?&7h-or?=<7WaN2`LE(Ka;3UU)Pj;;~wFM5{$
z^y{bLh_f=hy$E3r5IhwdU-UuEl@Sd_R0_Uz?oTHk&KlYIv{pV|Gka|;Jupv=$*Hwv
z3TDl)QDJ3hQBXR4*X}ZxLN{o`KnSUDSqqna@alfz;r@h6PzL1IIg2Bu1X1z6)$7Ps
z#M7Dri&UsGD7E8CB^MR`)iKC>VY_aumzC?|MRTEJQc;&c@6{j`80SEh;$j!ck*d`#
zui)rJh9Nv_F6sxi(ya{@ChD5WRH9X*hIKM1)ZAFFJFPo>ikwc1C94^MYMNLk^$yos
z*lMP<LK>5f4TpKKop3szS*DpjHd@tG!fn%Wf6XgyR0Ud;rq)E13>ag;)9F;2c*JR%
z+x77Sk3O*X%7;(iYhVzHkd;<wW##hxiBFFoxWBu{`M`Bw$u(;=AFNQZ^^6vpb|B<}
zkBRwoR)@gpOh~%oQSqt%Oy|2ZcZ+&$_gWNb?LA0f3Nu^Xi7xPPe#LTnAZMWlVYz$F
z+c)o2BjB_!y!M7j;P-#$_qaYj^4+h01!G5=!&Pnndu=q|P#=^&AX}}pt<VTe(KB$=
z(zrh7o3Q6d=Ipg0P9IP%dRr+JWl}iD`7|-b^cz?A|4?6db6^U`0p;-sQwS;)9E!Fr
z#{pX_-3NLvdhHB2E(GtucSrE59Iv(NJ%<o|(3{aYoH+-Ajm}$Fmqzp+htsI&&jzT+
z{RP~FcHU0>T2%z<o!6<C*t!jFo&Fe|ylp%`eI&*}6whf|v;%dzIt*TEipLc5I81_B
z7hiG*UJ$a~ETzr>VxM0KFEE*dL(vI7r~=+O`q0T}_nu-C1YZg%M22Xl$d=Ws-b!JL
z_o}C`$wzPjA9R}Nyw|^B4E5$$JJ@<6xkXolDM7!ZG{~x|A;cMR0eZ9b=0+)3des>9
zXz}F=3Oe9K^70I|YonHpRyRgl>BXE^LnqB~Xps`)#FS=DFy)}nW@^6D>TXzLZt0lH
zZ!qRU(DesE8%<$E`-PfUMuX@)m@`4TfS1qSf4xqeO@EHh=9nDlWLYidS^zX!gOPXg
zcBPdGd<2^Z8>E=Sx*$$&T#4b1Zokm-YT{xoT!O&tlU`q#BhAE}zsc(-6$8aan?nTO
zOqC)A<>@AVKr@)Bl{@EnSZ0><Nk7nQX3d%BT-dAr)vq2P)W$$aqBqIjA-6&8(Cfg7
zibEKhJ5Z~>0b_+Q?G^$u@RQ?JI1_zfikAL26t&6Fea+|^G}C*-c}<P0t+QqAwEZE2
zZF(4MWbaT)a94Hz@p1L8G{)W=J!$pVtL7A>Nn=WDt6cZYb=$dwD^E{X^*f8O?mM+N
zFd9`qJg43vcERceRMn3PYU@Gp0`ufK-&y!mh^nX!`g?HJ^|a2m?RX#bg2s{S!DJYk
zx>j~B>ob?jN7m<0<n58xc93BsT($Rjr}~R2Ib|q_;H#L>vGVlkdnVu5)|JQ0Mc;eT
zsO9K9E$_$_S=W_yU0KpXD}^W?+F;L`m+*AGa9vmJ+J_=UAHMvOhx^x@&v*DV^X~O4
zrVw?^hod>+?lki-MV1(Byc^hx<}=9X)ZR(UiTgKindgcBd=&omfB#2}l=%Ah|0zFv
z^%mdZvK7`n^WFOo<fmsQH*xpwXKa@X>$)+<U@}W#qqFy2z2Z&}1?ANnd%t(3p$*t~
zwX=G!q)7&<CXSstcUq{ly3;!Z3CwdOIAMyO`(^qmx9UH**Io2XHhTBbM%lp=aW^Pe
za;if2Su-G<XjFOIv!RH)G4!F^gV#GxF3yzPk=6$}D_>*3JW}mo!O-X8^!|xMhz9)$
zCR7W3Xnsm>_+E5!14s<E4p5t)v!nM|-Rm|;>2YP-kTK!HV2s8X!KMScYVe|e-W#7j
z*!LBDu!2Cp=Iu~10pb{@Ip}f|+>Or)19!L+?9Vx`)?nux&HGy;GWP*eU{}F&@{Y3a
zZ2L;|Am=kqI^sGQs4J$)eD0N&wX@e}gQv(wUn?=rW<vyVfjBKT@f2#)l*F=K*|)0(
zZj4T;EBp3H@Q{)wwd&8pO`zF&x7>k^(KF82@5A^Cnja7-LvDJ2KD2=vr83Hn)J&jq
z$o1%rvTs&kx!ZfJAESm(^|CW+)Aw7=|F`T5%tbN;{XK11fE;EES4(3#Y_C-_dG=B$
zO@re~)!qMgdE`1r0uxabZ9L4wsGqy7A;us$QFL>ywCjfWE4iuV5XNow@e*~c&->r7
z7Wiq!fJ5m_PL#gq#nXn~$aG5w>`lQdwQ;&XbDEO^CTgX-m1~FRy|QK<C+)Q|PZMEp
zOnW8H5jQ3C`gen~sFyRPsQClq7_aEaYTY){>UotX&T-;ynponbcB7&S(pojnHs{K|
zH$ZvLc&Fp~z0#_xKb`1UY9Q(zZ;e_z>$a1UXRgYF?n87`Tq?0t`v8S7&9+%ora{eB
zMUd;p`==+u66xOQgANC7Fhsu(pQ2`X*G|useJ`xr&Ymmt6qrvlr?Uc7Cc}d|?-WIR
zpqjn<nzx#D1Ya9zTIfya!x7U-cW^@|AZxzz@#zOX{`hOQ>mxFng_KkcBk5`b3rSZI
zG3n<MQbJD71dEN!#~+cp@pM^napBGV7u=l!(uHk%R(8C1oTr6TnDKbM$y#S!SH7*^
zQAXp#`|o-F@FR{+nggj$id<gKoKJVW{qhU$&nJCwde11GD2_QsVp0rsYbrvkIrIIG
zKM-1Hw88xsZ}{2If5zfGITwEZ2Y)~=g|F}4A+6IrFdC27E8qV39p8TUJ?C_?hqk&n
zL_AZPsm+9GLtPTX;3c^!OvMeLccxiERYf{a*GEFK&&hnmt?2ugg9ESE^au{0tHyAL
zX#U~f-dAuir+^&dt)vID8_=qw6-R9j9K=uwIjer9R&)Lxerk)26~RfQN+-?}^XW`0
zo$K?3R;#US^d8o(N_0<uXndTfN?=Me-8pJ4jBPh}P{obL(T59)i`GtF4#b%$8r9Q(
zjQqv^|AX6)=?add%wXedObB><&700d#8m(G&vJB|=4P-0;MPH-;zzCXRZMt@{`{@e
zhS3A`Xun_7vu^i<7Pf=Y3$1KgteJp1I<w0;g>1=j5S#Q}y3k8u%ast5V%hbUP<_o!
z5V}+LXL>CriW5XSwe4*Cg*fZ_RAkT^cpR-~YTc;2P6cR%BeAhr;8ATIyz&qP^qF<+
z@_U2Uv@kH??>IW<s?T*Ff);ru2^<J*=(^4sA3z3%1lE#4yz*1TroV@l?Z}cio(USm
z-ENk1;O0K-z0;c`?<<$fGmgR>VfI7a9&~fJ92pN{YZ29@lzk(dIx%{NxhSId=0Z7g
z5ssLOzlSw}JEB~j!h1sPxYqGSg+bBlc&^k&6gW?jvyGTW?=<JhP8I*9b)L7ZJFG1u
z(K9a-%ba-iN{fghLUn^S3Nbo{1nbN_X{gr{Cqz*(%-s}uby~QeXBFS=J1d%0w&hIT
zGS8PQ&(BwC8<x2(<Wd=ZLtN3_j8kvzK=<7_XZBLat&^)_GZjzjE$5)C-`4I!<e*V0
zz`~1gnj&{|WJ#)GEknEZ!qAC|I13dKa_f}6uwOUU>(07o_MC}xU|KW+deEu8G0DIO
z!Fi1&jWKYe+|{Z-Ywf(gd#!7n4%0kC>s&9N*i+=w(+}*IkJNpo_llQ6@1oYWp(<Y=
z0&$w~F@ZQ*FO;Sar&<f!zEkY)U329W7aNZTmSthRe$qlD1eQ3{OJ!THtmVqn^9T0r
z%7@yxUZ2U=3+3{R@8CkfMRlck?^#ZZb_2B$=1C9YT1|mGG_<5^EO2d=$LA;h;xGS#
zSM!O}>(~6=*T3LyIpKTbZkkyZO<v5UkWyp{fi^VjeW2-^ukIccubdW0Gj1rWJ#^1n
zHd-ApJS`WsLbk><C6?txhygDN3WOf;Ba+vG3;IA!lYr~2&jr_^`_6FSN2dvtp{$O}
zW&bI+>OZ*G;T{aZlaeP*N0$z*d5SBrgAnxrOrzF;cgAKP$JkRRX1e;(UHtUsOVYHE
zpRTmMvbhUKoVarb61-6NhVPLy&-mzx>5O0QxYoj@HPsUIrrLvU7~rE-M-SH!g02XM
zI)F4@$J_^CA~rp+4km%RiaJB|0C9520`-^s%tFTjs}cDdW9Q>_Oej1rBXvyOeiQ~B
zgE3i1))*?*!%*q14_gO7h!f|#JGwYR5c9)h`U4?E{hXRL1*09(=9_L&<J8SwJ$#@K
zZEW_M>8APUWuxB?^W%p-^p6@_FNiTX2EiuGqiW<bHJf$%Gly2wW=2G_E@bG+cZ_P!
zF>n^p<efk?=-p9_d3h;%45XwMbx(7)ZgS&Bv^-wUdqRr(SpyQr%^G?LK|zU+!zpx1
z*Fk~mhq*p<{j{zZ3mDR%vTo}ST|qQ0xZpOK)qpDt?2D24e6P7CMzW0#k{2c~#9&i`
z`uAWhtWAYI%|uglAv7%xmL(BWpmun!nnSG@lUnSqd*+%m&U>bL($Je)D1DHNW|kLG
zqFZo|^OPu~k;_ggS}n~n(1$7R8+w?F`nFHeu}oT7_R;7JY9EwRSg%)wm28>obz@r#
zbgUa!Rli6<p`2dz553xr>vWvtbUmV?6BUmH2lJ9Bf_CzD2zi=04fF<@Qs8cmeEs$n
zzxd)c^XbG>X<VDGJxZ-MetfY6?KyMZHnL@{XVV<Wp*!H%OQ~&Zn)Lmjro;ep)>vNY
zDx7pf$7iKh7Ez9VYnk==iT(13w5sd^FB-{OI<*2#;T>)cnhW7Pxi<DvdA@Fptl#U0
zPal~=(*0HM8VPH;6**`2O~-EYp6R8K%TATfhfhE7<NI&uW!EZt+f+DJs}}aPQ})ca
z-~O7fKm9)I^@1CUm#w8xEV(Q9&RPr63nO^;(%4Jm>G_Gv^Ai)o*}o$WVYEu|!c*J%
z?)z_f{`5?r1M@PIyayL}bvpC<?jCWD=gWmx_xGgbOb-*k`{mC#rO3BGe1j9uyI1en
z*OgEC6YEwer84=6NFe!z;3rzIWFj>yk+50lN*@E?q1KM89ltmv3WaKDXq_^Y^|yO6
z5WQEGa}2kz80YmN<;}D`0AE2$r!^hVip^yQ!wzbNGI<%`hUE=BfH<7r^+7xi@X7!Q
zro<k`<af{-(Fab;i54Q+{(VFKhC*DaarW=5^Ycg2YCO7d6(R@k^x*4Su!uo`R47`N
z+-C=jcAIvfqN8cX(U-n(1rD~r?RBgH<i%M4$3)?U_{ZqC_v?6uDPu$-SusWMPQ|z`
zATHpXHzA`rKWes21E&+j3|cOsOOhd$<f?7cAx2b+qW%2?x^y)pctN_qX)F#XfX-Ev
zNU88NAp}R!zZutcIyr7#b~m<)UI!seNJxq>wqcES@;2q|X6rO}*ASfm_!ya{li51e
zWumzYLXUyy9GhF&*NuJ@ZBEhnaheE8ZH^6XXr0s8_+VediKC026Fvk+g&3@Gsukz;
zdMtToy<S<Sz<E)9i5u>)YuwyuBY0wmSTzJ|j800Cm}btWq{YloH$n2k5;X@T1dspw
z&<u##6pn{ksZDW-&g0RP!j2>F1#+=jg@&)JxpHk)t19OpczW?SEO0L!T-9#Vc|y{C
ze@q@Wy{Lpih@RMnq923TG(v1DAsQ>Zm!c2aHfW{MiUuOpEt9XR2zNtAeG*h*9AhA$
z7e+41518f&AGA`QliDUvr<o}!h7)w>=VGLcM(tT03Eoi+4O*M-jPK7gKl}16UwrWf
zai0CMBI^cKg@=8pVpJ_b`XD<}*W3`6=sC?%Q~rXgUA=%0ifv9-3~2Esv=;0F)*Zhg
zD;T4YuTN~#8RW!XHjT#>VVWjl3tU#{QW!(UGEqD+MPdm0{_ch6>(0~jg|Qd5=PM-_
zPC>K2+fc~M^YxiIC7!QOT-J>?C6;~X)AL8t>4eXnvTZtMyef;J6^%(APUB@q>y(`N
z)vy1GSHJxCEU!+yefI?=319m!5OF+}%nu(w+BKkdyqXK|KYXC98?W!~HE}I?#5?wy
zsb%o(x8JcoJ}D+wf>z&A)wpemJUqN&zPsm@kG#Em#dY6#T(7)-c*kj)l|k_Qk@GT9
z!a`juSJ}v|&~k?!NNys8$d(mDx~=;yX%5#ZZIH7NQfHLG+yj2nQ8Aud$j)y&fiMKM
z&id$Y);Ha4s?BW-y^%BRreZ(9If#v%ZEZM)T9=)R%y;+rsL7_i73zWH7PT7YTA3u;
zzFyZ7w83d!h|9uSt+Y15rVAb~iP+T&DsU^fb!w<_){2Pj1efeMQ<0+U=zSx0;CB^q
ziKw64IknO0gSXq?eT<-AAVbIN_}^ZR34+-c<xs0@5hZS*v0Y+z;&JP;4=biKzJK(Q
zAT!=a#A$@Q3onRcM%TS`E%a9P=dzV)DROgV3BqvAetLz*9ohxJiapP8S`c&*MlFT_
zSDEA()_9>SUFU?}b&Z4BP3aoirnw!}x_=?WnRt32rjt!*HL6+HK?ZKp5);>z9O%5J
zAx_hYG|f1anG&2>o=n6`ysdmjALMOUhePnhDdA!Q=Qa02gobLXlAUeAd$q1ME5=Y8
ztvG+XmNj=m->T8Im^hkF87gu;Y^w|{rn?tH7`WQ$CFAPG6az6tlDA+!gy>*Si8-m&
z(T%@{VnFSxO5Acz=(XA?cz}y|AJhsam^})j)7|vm8cl?-OjiF6M(yP5O5d|ZNj6HU
zY+GjAlj3Ge)&0GV+K%cm2FEl7t&+OC30(HFzl_;1W;BjkD%;bQeBB8{$CcAO5$D8t
zIT2&jh}yhsSBpM?y?4whvBXICT?2S*71B*i;(eqB<tWgO-JyflZy#uiIyO#eR<<^(
z5{*Ig%G+U8@q!!f#_A5@n7k(i&+GHTJV&Fofloo%_+Bq0wc_YxLw9I8O4NUU1m6^L
zdE(+drJUJop_FWsg-O-$IIjI!Z}P)q?3_9uT4zWlZ`R4T!d^C<8%*JaeZFlw*QXEU
z5LvHRt>&$eT=PcjDk7SuiDh0W>qSYFPQ{l^tfJQC8_yp;@XJ5{Q&Mc)y?M<a{%8M7
zUQcJz5Z0FY?uYLv>&nyHFL-sQ;Fs^e|BlP2k3`_r-2->0vyLxIQCxDXNbdyewnhhs
zV+xv)aQ|?}`Ti9@dw9b%1}P@h3iLr~-LS6?k;r^X+&>gr6_S5q%^M{fB*_Iv8$5q{
zrtJ+0<3>P3Jhef|gElgC3{r7QH%t(tj>74`-PS_yjgn`-j@aJg*wd!sa}2LFk8<o;
zE2ST{%%Llt^BrMY$fZ!$3)}k4UNVTr^j6bs2s%RcZV+PN{^1P|_phlwaLqfTRTxA4
z?k?fG3S)(C;c3`j`A9<5NrM-=f4g?dI7Li%BbaKJ?+9UHIo)f>QZt~8cpYKO3sdBH
z{cZB#;YPTDlg1H3giTM3PN*@9xEoU#I7Z^=s2S%B))W<E>!Zd&2a7v5N~1FIh0%jj
z`$nUc7pq@2(YF$5BjP<cPZOtUodQ0nM&a1+i>u(=%}QNI*XiX@OSb}$^uTQ6_`U~N
z09|XHS}JXTgb6>LnU_1f*9;XIO(*Ln*zBzu7>e7WoI+Ac?W}7|8I3VCDt$O`J~Eug
zg+v?>C$$ldG4Lr7LekF!0gIEDq0z_Dih(1}*d3pXcgQW~+9t-WYJ1krRpP^qr=Xvi
zHgF1aI<7%|Kb#0-zp_96z}=g#aC6ckaOhescp(L+g@Sif|2J1n|5LA@1|8~}$T#ep
ziXjljpgM4+F|Jv|L|WsNRLFO~EL>})IcUZUKgNT@u&br$4bEuPTzS51OhKr<vh6#)
z>R;-H;M`!D9p}@+`IN}H@^n?@d#;)5wo*#b3(G}fP)J7agKgW{*P;nmDROsOIL!;o
zG;1JFtL&wbi|+rYDXDO9&pd6bCX!VLSp|hh=uh&IPQT@%AIV}asCTs^I<kp@A-c26
zwSv<ww&hCgMWHJ;`qw>{G8nvrm^`PuiSylw9W6#uQYeQH3i~*SSHna*40U=q^RRjm
zMpI?$WBq`?T=gdMo_SKC=QN%0F(E*HzA6(z&&u0H+?qa+rj%$h$Wwq(acG3;mMeK}
zY}3Xx>w{y@89}Xh&Y|6#`KpZqtM6U#%+o{_$NBDzSVqI>1Gb(0!}r8qd3yIfKfd~s
zIe6;jir)+GpFffJ3m@kby*2*&n_uza#~*ocNu39|;s(sqtT(Go5}ux)c)C8b*GwB{
z?E;+Rv@Cr2?hD?`XB=>PxW~Ou+`V~46wmXou3RryoKMX2#I&6G;+<!j0`EV4&<u=R
z7z6UUv)YPkIVD0=O@<dCL=BS}I9ln{(WxT%5gDa1Xw>{$4jmF_0XHV9>ng%M(%Dqk
z_0oweepRdO-Zi74Rizfjbmsi#3*vI3520@>W7}v|S5v`j(vvf?p&o?F>bSpu!@HmV
z4q;x{w@mF-r*AUMfA6#yaKPyJTO9Fc41m^br6-D`HQJ~Q6~F2JcG&niaXQ^I&&zGC
zDNgOC$Fm}Ypj1HjKRp>Z7zgxYWpZ>PVugqOZ;s+X(-3ulsdGWF14;zM-}c-gI2FX*
ziV*B;Ipt>9s$mQ(aE4e^vO2w7nzaxxx7TJC^fm@QPKa4`Th+!+Z0(~#Jwr8eQZ&ud
zy~N0Oqc`eK#F7<7i%CD93G?!Tc{Z(wbe%37sgoKLpc8Bt6E3Rm$%TmYrZLNH*Bpdq
zX^BCf59*4TZ8h8oM~=mSJ}RSRXcadGL6q_sy}uQV!vz(X>X1N3`{L<6(0W42rQ?=y
z^O9V9tOpL?it5)6P6YzJWY)*;cz*wi6dmEb=(Qzc6kJRIsn`B*^7?6<fvO6}aa1Sv
z%j4K;2dQd1DUB~57G9qhQkr-!jZ15kJ_z1Cv%#rA%?BMH0WyxmYUjg8709((iNR9e
zTmY+l=E?JLnyJ&EH1)E3u^U*-ei0n=NVpi-U7(cCb<gazvWy6$=?z>f&O7F5W{QD5
z1#-#66llHkeA(FcMu-ElAZ^gHCMA0BX>r^zy<SvJeE6C5VAG^OAGCs2k(mi;(2j*M
z$fXiOwF}WyzXQq`+<65C1+RoF=Y~e5Ca+58!zN>5L?;39!P3HXGzE}z=K1M@cZrZD
zh6u~)g!7(dxo1hU^48BM9+TeSN`utTPP%4Nq$z<6;wf>8j>jKA=tQMfW$EWiYY(Ok
zskGY2qRx-|rqd${Vos0kWh88uejP_5CA!OXiYXnXvR^NJ_w8SEzJJS<CdzfEY?=4p
zf5Rv%=hGeZ!S~;M!(V^%Ex+^S&v2)SlY<x|_os>byP3cE;w^dT=T-Ner)}e1ZItSC
zH<=4xp6{7cqVAczt}HI_{NqRFDKT>9)B6v!KKSbEuX*+69m9l(eKZ~)A8B={_;on8
zU1LWb3fw78vCDqw&iUAD_5y7ilu^ys9_jfvtM31yzJ?Pc^O>NpYue~B8Xwl7jnU|%
z8juUBF+gg!V6!yy>Z`ArP7jKiT`!tonKz`HCSZz4ueh!@K3X9L$8tLH@bHG{6JzKD
zzSLq7uS$}1PKeIdU1hiNymf4&Q)&Ua@+d?nlM-$^a}5!)!p8Z)vYeRb1&mDEO=uUq
zI&(D1Qww?>2^*&Nc;Ci2`g{%`55th)L!Ti7)lVHR1$}iTfR00xpleGJ5Dt9sP{p}-
zaQywvx;>|Y+|m0@P}$8{g(zgF$skM<Vu_HY7WR^P-d5JKQ)?mBjo_m~$PRj>6TA!h
zTp8*(psCZNI`AlzjjkFC0dT_b9;Sc~i4YfUc8Y$@ns-JkI{g^>S$YR`D0vX4(4-KP
zj_Z4amQ9>kR50EeeSL(~AujqJd9ML|T^mH{N>zo5RONvfR~zpF(cn$159&x6HYEr_
zL3%a~3f_}q#3FfU-D0l&;YKkHNc+6Z)6bQzE3vXW_Vok#a%PTiRg~ZY4sT=pt_Ak^
z%^HcHxF)di)X-h~p=fKy5+ALpuex=O$DAhJpAXne=XzZcmzkECFhxz{%R9a4n8jOl
z+FNI>o9>GSL>Dx;NK{<soF+EuO{lgS9q%4VTFNI7C+bVJof~7&YG==dfx>I{T1f&^
z44PZsBm0ziUau<RTX)t?$J(=-OwlR0FqHpq1g~Ki?X4=c!bHhYIwG}r1ocRJkGFIy
zUn;b5d!Q)JvvA!qm#p5@;Qb593`PAa&pfF82jw0JU8{A(c-c<ZAJ>f@BkOfz-#0=~
zZG(>!jPWv^5g~*~N|BVjjt4CO%7>sVeS4@Zb7Jz4sO*oPkB=99a5roUP+4nZUpMw`
zqmNEDPD8Cqlic@}QubRG!T>JYl~O8?mq#tiyb5=`v#g3-dH(TxzMcP!Bm;)$<Hzs$
z^y4=K8N|^UwbP!TsFzRt^3z8?z5kBi{V)Dk6d5d&1~L7?mtW)GJ@7?=36J!_y6<=g
z&+E>wzW<Kv<(bL7Cb!DAt%_D&H`*R=S_vB5ogVn=)jRh6%69pL3@f~zpNY}ySkQ=a
ztu%T)LPxZ-^rli*XNs8cO2_4{Rj9w+#Xz?Q(~zN&<-;sb#{t;Q>pg}+S(@74hDNdW
z4t|<>c=t0-_iq4C-d4)>qVAKD87P&GVdLID^nmb*)9K9V{6Gj3wKVopSofWM-)V*r
zKF(TayVnraT_OsQVBa=ItJI-@B!^Cr7`mo|17)m}iFr9Or3p}*Ztprd+nZvu-EyMG
zgiKDy&te^ax9o!3%RYNM%@5n+=o)}Ex+>OhUA?uB$5iCyDACX%Ww!Vm8Bb>bdT8ou
zvFjl<xOMa;XG%Fz`_N~B4}_R*V&-NHm+QW1SWYkWw(GM%TVsOv%yA;6r0D57=*H8h
zcZ$<>>u|baaYtT*XaniZJKwslJv$xWyG{g#Spm^PX?#9au_Vhr2GS~wiZojl+WYE6
z=q1zJBUtE~>Q`+d!chG}ZOEv^mkS2DB@}vN)2Nz5KL7ka`y`Y&sjj+VqQY7$BtwzA
zpx1((w+DRb(4q`mnGbs(9E}(vF=~{u*otU$xw$_6uGdd025y(8DI5>~YwPF-ITW%I
z8ETtrm9=C(Jzx3ocwxWnDv&FcFzLu;R2{RZA6Pr$DNUT_$dJLlSKXt>pcu$jXsr^G
zBWyat9x_Nia!wQHxB$BETD&I(N3qDx6yx#ixs6cfQ(|6hMD7ElHJ+cIxa?{-+j3>E
zmHWGeFJC_>K&Eucb*K4u+x5Aqqn;2{Z0DU&4KX{0cDOiEB&`Y!@MuL`3@YTkY8wz?
z+bhr4oxM~;r7EN)q(~^J?!l@+FG1H=Ll5|nble%7j<jm7tk;d~3fJpK2#ND)W}0V3
zHFqt_o}NE(WQ6oy)K)W8w54KA1%c=_pvhTj^7QI)Q4<tPuHd!muT3xBy%fr}G3e}B
zV?^C;{F}8K>;-R=f;|_Wo}Rg`7h-Un=SY~M2_f~tQm&7D{O*^Gx*{R+{QQZsKjGs<
z*{`@Rcsi3qg`MjYKmPCy*QP6^BQ`h1$T<d1Qz8iLYbI}n_m7YKyFdRkQWvHeDR~1r
z?|=M(SUVR7*XxreTTYQV&Ik&rxaP{5v*NN_BZzRC7v^Q+)2By8UK^tVAT(vIdri(!
zKX-MsUDfoX5BhKUn&1%q>4?~DxOVd#A}N!?YkHiafQNdw2<b#x&cykoT9uM%c_Xh^
z@_N;S+$-fzg9R;FFeJ{e-!Px<@F<Hu?^mu_JJ?$cuIM3|LLfLHImaAkrZn67a;FzZ
z?I69=OHt}wO3<7A`D5n`G0hrfERIo|Qteu~<=NL(*#wqpVvNc5&?=NZM*7DaH@c;n
zVh7!4y`mlx0L=~&i_kg3fAovI?4LPM$I650ttbN5_V|Bel5mK-?bk46zZ^fSi#MWQ
zE}4Db*>cv2u8HJY&o`_y!y$7(e57+E&Fadi)~e#<(V;g|vc{)(`f-rk(`SILXGYeY
ze^($+&;f^PO%R)^2tGuOzy{TywQO^H)j~M;0(%P1QCgw3XQ&gynf3}X=8FB9F>bcg
z<1Eztm#7MUbBy$^XGHD8m<#>(>t;JW;6#0B?n(^ynOnl96(dftrde*;8466&jEH%j
zJ`*t>ObA`S|0b`WR19>hg7ghGEUowSq73}jH3>^|sdX%}m&{fw+g_-9R&h(~N^WvO
zJ2I<Br?x>Ks^pxOr0|PaX*T|<t*|%sQ--}FVZ(<&>w`FX^=VffV+H4!qJ<%KEZ{^}
z3CRm{@;sbpUfnN*sFmiY{mNy%^5OBB%ccoydr{lp>$@|rU*9Q0_h}^#p-MImw`MD}
z-gTW}S?qmua^7{cgnDsTTT?XK1NA=GYhx=qd5G3stn0?QX$<jXv^JL%+nL1SA}rHF
zv_c@40t{7~dVkvuwxODbIV2i|Ry*@DGfxxqGLfsIF-u;#TpVK*Vhk$2F{)M{FAr!L
zO5S1^e$zQB!kP=0ZKvkWe2(}SX!bK-_l;62X-+zUF;06anUXi13=Fk<HPO#%TQ3UM
zaFBw-M?tOS!f@Rx<idXWkvayKbyMW@7+kwjAlI_dx<WftRoiHoBHa!C{KsE&{q?u>
zU;F|eJV&U<zU}18mG$|Fr)}drC+fO#efq>VfBj3O6i%m^5FkzyaZ%04RyVd*dH1ug
z=?uPq{|OP8Qet#nQPkdezOGzt8WMsEs&2NYp=&qSg`Uk*T{_hAQ*PCNfG>G@;F&1u
z@XU{+GU(%Fg3yiUE)E}bmmK1(DUjuv{j#!et47R9SCOX9z%@{&X`aAx_nL<<zgAoD
zwvzXYYFLU+G8rHpswhp=)yl~w(v;M3kQPSI^qL9cUTPLQOzEEfnft*z71RaOD(M(o
z5!2GDjo72D9R@jHDP`8o5byp5o8Z>Z+tft4DD74>9r_@!wS<)l9C`-Z70c}t(67;l
zPE2}l<XWhH<6!2iro)2>y%t(2I?ZZT#lxz`L*!UXYBRI%yN0TD1^e{TZ(M@jtB$)x
z*FEJxus1{XX8m6JfPCAWS0zSD<KRgk0qwV=IYY*+BX?E^hQp~d^mjU24~P$xQcbs@
z9REX%=MU(Uxd+5X>#b~bry)*bhzcD=3#DUSbGy2&nDikqS<<2gZ|U?xivc~ujw!$q
ziLFVqSqrFvSs4%3fPU@KU#1tzMhJRs)M~|oW+KE8iK04^16BNjP5w{6eo`^uQ@jC$
zmSti}9)?;W_O1v<5KbX*GVZ`=owa2pd>NENY^TwSlAw+eVXcGfT6mf>%YEYHB4LRd
zGBMQd)_TW-<{`9WbfY%HTpMgF9Z13R>i*1u0&T6cZ`)1mA)?}#QZkjn)&@U(`pA#(
zA9>yi=M*^4ntBzIW4fD|?h?2}$dxb#d&{aJut-mF5RyY_Iil+DUhu;}K%>3fFhjkw
zrD;N9t(_5vj^;X)T+}H6KxbIAF-}M>GC9ZOqvpTYLCckfj$g(5+nBl-d9hUujq(^Q
z#U!du2F(qN@r>kM7){mmt!lwihla<rfzh#4c`f3o%~Hx*Wvi+Vo~<B^OQi3+Vws2D
zY|E@H1MQ9qd)c{eE8Z*5K*mAnI@}clyxJnW)6U6>*GY_1z(~&3y2YZ{;kr>v=jrJ}
z-ZKhkbIj4>oABKaKQM;lPk;GmT)+D@*VnID*DFttk9zYpB}bm)%yr$_F3)Ol+%t90
zNK;<^oFezH-|+71-{t<{fi3S0g!|=$|Kclt`0`t1bj?Vpm6|i74c1n0Rg1>#Dj^Fc
zbdnFo;nz-}^+E0my*hm9zqK#X=y66ffuDA;8gpwySnEh-i0#F7otCDcw87RH+fL3q
z`@ZSQkfByomF?+%cxcfo!EyKI9dSA98go3;w^f67TDKO`>!Y{n6tDD3D=^>PaevlC
z%zZV%pb6$UG<f#Fm7#8tK6VHZ=VnvpS8~~OrA#wjk9Mz4beudG<h*IxYp@i<t|;Sv
zyHB}w(Vx$JheaSC5zE7ID-=L8Vs1Fw<A5B)Y}Lxh@3nGRArW;4IMMOoF}~K3zYV=D
zROe8}k8W#(T5in$(T`YX<pH#2;%YPk-sPR>8lx7rO<_cCbUcpAAUi;Ga-hYDB7e2k
z(*iQ4h_l8hh$5<kb@&1b#uJ^~_z-W_?QResk;A%rtP+mvRGntWx_PvwdWOF}1sDSy
zFGZU;2<SR9`p-u72MCGVq)RQLrX{c@tCmXHtCAdD#1BmaMr?)S&}Vz_q>&g@)!c16
zy%>vO9Qb&hwhU{cTl>wP9Y3iUP@$S4hsEpCOo>$3#c$Y6=a}b2h=G0RA${zloe`hB
z_vDt%6MB<IkW1m?;}b3lOX&=+>3+%UuDTAbj8fE5xT@y$;QMsF;EJP;#>08xE2)Gu
zk-F10MyZslfihd&`LwS1>lI0n$8}@d3k@Mg$Gk++6tv^^;FCjw1L@4CNFI%tBfXZ}
zhc338(htC<0J(6QG+SZNZX?20HFT*IbrGb=6N1xSkg0AxauX!#Xvm^Thrx+zv%Cdm
z5|rK)-z<7FHcRVrV-VaRh`tZ46mrh^5U5Q_Uh^DmAI2@;dtEOGih@>4ktTqJr0HOv
zLu5cnjv5}K$!UiM<WP(EepI{4j2NYo_sX6dj*jpCrlj?DcjnsIY9WTBYq1;IM5se_
z^|0-gUSUwx-qjl0ex=rePl>G-K74%7hxgy`d|p@|ujC(ogxuJlpZN6YJ09-N6cgTj
z`uG8Hj)#ZWoX_{X|L{FOeEfl&SE36r8ohbu<(}8~Z~5x&S5!Ke%LUmpdEazoD1+z&
z&*h@U!Z_6PTJ??M=v!AFM;fMn*3P(f4d*JQ-PSX|y)U_S)g5k%bcSjObkEyS@A=Cr
z(mT(xME34!>t@;XIt3H-o>2SPp(^hN`Y9~u*My`o#x+-VODipw=GmQg%FW($?4VrI
ziTm>d({zM;X}9PGw|%f}{q&w_74h2m1aaDRI)z)+T2v6`<;5iwR1E8bYM}(tX31>H
z4(;ldM5ikC!`A(`l_N)Lo*uUPdWo+6{NLGs4EroTyUw~51I}xwZ{z}-3cifA;pl)>
zm!X}kY6bXP=gq*Hn$~SSq}cQ^tkI$9bY<A{%Zfs}p~SUP#L2J@TzBKaJ6sfebR-{0
zGVlz<CV*m^F7NCa6ej&yx&qj40!=HV<W@acH*Md4A6?;gLpd)h%sxK(gBftRG>&&n
zbaH-xsk}!|6t$uP27q#m?se*7?`tlF%eoShmPE7H#!b%_6R3_ODRJ2IxU%PbD~yM|
zpPGKE-&aT1IKRp3r>zOP!Y;;uoGT&$hS-Qlivms16MPW5z?vJEJ>QB16|H$o6LYlY
z=%d}nV=&ma!TTQ{nYT(<A|W~a9B*I${x0$Q)tL}H&rcU>E)Y7hJA89YOJtgrRM>k(
z(8$YSZis14ggLT}!H4ISy*-g);2aa@*OLkwomN+22wQ3F7M8NusJr(@+Y7Q4TyVG)
z6piUr09bNn+X}f<&NEEWJha|X#*t{H!pRUlK4?_vjre731vk1vLd<htN}&(kt%s0s
zMcoIj8Dexa6X&UYaSW7H6_q~L80`B_jH*BAeds!1XnLMV*Il=zMAlNZV@DOlX_-`9
z6m(T|WED&)(fUv>MDMqBtz#!vYbBS=wq4oxE9>@5&f6`tq>aM14@(Z~AUcgGR^%8c
z)G7h%Y(|5wz@|MHf>(iK2!Xuql-Ag*j%MAq(k~zQ5C868LfdH9jlcZUKViS_?3ZUg
zeENom`!l&@TFHEP{K&NyzJBvF-oAUsWxEjP1w`1_jXVd!d0~3<j(1=Gf>-AUuImMW
zy$|q?r)}fwH?NtO89sfYl==eNaYqU7=~cM`(x94%QjF}Di%ve=Z)gwthx*ddI@mF)
zE)_eB+(Fbse{PL~H7cI#;JjV~FVAiD$~&<yDi+*hJou(Nd>?0xBHy3cw=4PS5z4O6
z7tu*V6Vbw8=;3=lop|^9EpwbzP<BjehEeTAt#!KDXfT}a*J}l14s7d%n1hn<dSy6w
zBm710NWodLY_a1Z5-bt1RsG(b$HabO_mRDcpRERtR6jjzY=STj27!XrZZ^fyZ(rxg
zI2cwKyd>>exBJ=2-h0)K-#S?RQk~&!&Drz-60>rKVY&$$y)#tYji^8zR5`C!%zoRo
z_x7TT7&aA?jtM(onu+1MK6|3GkvO=a**M<D{?49NE*@4u+nXEPM8q5-`qNl_q>4$6
zXa^aJnI5S4Q-$ab=YlnbN3o>up>v+j&}l`C79JOHUIo*=_uJ3!y8glDyERA(RVsF`
z*V%)&qDGuH9%2&-`){4vH0sz`uPb@e<k?o7{mA{+r2eMBvY%EARNMcIK`Hvpc;^sP
zl6ov*?X)m2GeaED>&mC63&jpC=e;WFQ&4=O?V+^8?>5%gOn~e2O6raKaAI`*a6)jL
zmWkI7cK}>3E3I_}1Et7(nn|<g;qGLVFujOc)hzEmAWf0w?nHOOx)#JTpwCmniN?*=
z);0RE4y1P4ZX!P$kB7-~p_5x95s<FQiLF6zm37^De0oOOU`dJlMKcq+6V)MTsmCoD
z&}oNLEi1d(#Kn6@oHTQwk4CMUc-4Ew#X_Yi%Gh~=@dSc#2XG+K;bX+bpy4f|Bk-6r
z%Q%1bFtMGdsj=&x71Xru<aOua-92}2?l_$nWh4mpUg*PQni44muEioc4-rw-inTUF
zo7Tef)1!r<I6`nt$<ez7V98JiM$q@8gXvMaO<v^o0P*4!+1nJ1G|iFo{Xz<XC-gw>
zbkgLRrbxeT{MEnzWAgQx%i|;e_Fw-i)_uoCVbm+FUD-Ebdwk{q2|@P0`wu^`I{3ri
z`$Jyc-!c7@|BUw^e&9d;@t<&gUTLlH>dP<rgMade{NfkCAmCW@N^OPb^~z;k*=s{W
zB87?j^PMH{Sul(j#WwejR5%Tc@;20VZE!EKm6q_^T@2{6RM(2_K(Cov8mIb`;l%c_
z9`7PCbbQTJ%NPh=Xu&atH*B}k>&CHD?OiopQEj!Qb)GLzlw8^O3vIg)IYzcx7+0Fn
z@Pxb5nJ?bHWtmUpHt75Q;$J_y?4j4A4n{3$)JD+9lynST|B1L;4!stD2M0eI);N$R
zB~+>#sL^c)4#z42)0C*z;`5qtv`*J%R1W}qKZo5b-Hp$RfnnF5-0;CxQoIxc<1>DN
ztqESlZ7-_#A);YuXfDX1X>iWrnu5~A1wv{#5{LR<ClC6Z?_=P_4rU+p^Y0c<KO8<J
zW1MIqVeoFvc<34=<R#uL$%3wyj)F~fTe@yh)L*qifLm0uliQnMwpZ)CZEe+>T6>4X
zg6YtSjo6!xJ~PL_07fv}xY<TSRNFVAPDZisZwz;%`2v<?XqJ6v6X0Vq(9Hp1@3A=5
zPJv?ITQB6Ajm+6^1>^A<>JRt#Pz=<Zkr0fiq(B=T<sA3);FKb7?(R6B&bSbCERWs<
zxmJ2n7C$K<F+pia#)N4GeGuKhA0VAEtYd7$Jh_p1nzTT00WY4@lz4rA=KT5|VGw=N
z^+YX{%g%M(Zg#dfMMBgAf)F@jXKU}I=uAweq+l5u05mzzsCu!66!0mMyQVIN9yRjP
zDZHcB!SzzOTrwx;m}BBJ&&(;2Z5-B&v)+1BA(?lI7IuzRHan#CAsM4rXp#5qIkRpV
z+#oTCDd;H6S?3<3u1`iM#=tx$eLtq8<Hn&k@SgMS26s`XH9mT#Wril?J(Ek}`Fho2
z>*37Xcdt30&#bwyZYu}3U2KIAqobjI<Gk<Cg}iD;LQH`)nSZ-iw)L496J7@AW#YQ)
z1w5RXE!1wj=m}F~UMA8Mm`{^=p%q`+?P8y%$UG(HWhMs4^{QQCj6o@WqPDj6a^Wxk
z!=Lf~`}aJ5c%<}BoIFzi5!F~+t{?dR`(LroN{bBMvjpMw!z<D}^E^H<ohM#>`6a*e
z_U9}Fh8sLRKXQ3~WS%DOmie~Y3e&_qFPu+je*EzXS1Qu<;p@DI=x%l)0R=(H@ldjr
z!QY+_|CYUmiPQ9XWvK1et<woN_|#!<J49p7Ij9{)+>cv&VC$V+RBO|Q4Qri{(g_zQ
ziw@uDxln7T*B$BQwyzx)x~7{|b&TAf&Uo+Hwhcx@_Dyw8{g_<p#Jbs9(+#zL9)}f}
z<?A?*5u|A`0N7aiSQSJq1kz;CA|30>F&N|X&Q8BiIrIRw0_a13X76q4dbj`(Mt3xS
zPzw*7weJ1(92#g9#%+yvsLuz-n&CEWw&Yd~;X=enq`G1OJp*d({QLjmk9q#^QCBJ+
ze*SyEz`uQkYbcoz>*8-ky&|HO1)#^5_a5;<6G@LE%qBs>QbkA2j1X`(Rlx}~^&9B)
zO^YLEYYyoR&)|96IN#lohNsruKt0$BTM@{*X3*k-XiZ5QW(^-;x+P^Gyr9BeS8?@m
z-yX2HfzX}P3CU20K}tyrAk=o>&}efXJShe}hkDV8mxby<yEPP`&yU;+rDnBS+6q=D
z6~|}aIWjE%uGddq6L?-mPopbi8ry5n!Sm{L;_fsvrHM?Zi_=bgnJ5<>7gGwO_!osb
z2gFD#pFU99U_Zt-s;WK&Fdd;;haotoIB~a}xI3MhV?spiA<<|#vwpbn;c;cp^``ip
zk{-~*X{8*Zsx!rD;Dk6~+crusOhN4;@yJ6L1;p$|Ej$*8;*N+@q2@;3I;Ytae9-FO
zdG)_{eSnFfIa?b{lPCH>NP+7$+#VPr!fBpJDdJsV&xO4f<~fq=7XyOXO+vIPyXl=g
zO@ZL`=k1~jYd<`rzf}3rdzNM4?%{zLCu+{5DdB@>&6yB9r}M&TIq4o1F+{9gtoHM#
z6iCOzx;Cm5TGy4!`JA|an29MEvb8dXRqr&!wHwA&{pUm1FVV61iC3@hc>Vf;6g{n0
zYFF4x@?Iw?T?>Mv8jq7&iAt*~9JOmmh?H%kZR&A$XfTS81LrdPe&u=l$PeFt%kCrZ
zAHU;wzx;|?H^dG0eIo_o{=DFw?QBQqx?cJ5(+|{Iczt?JvPjk%J*O|eqWC~d5o`^l
zz$#sS^sd)X@PX2^o*UW{f{jeuZwL<ghx$4?|6$yK)~v_@prcv##%q^D5w0Rb$1rrS
z2Fi2>)Q%r0lc6_r6rDIdywQj6^%HGdxol75z2F^8(a}TFUAk9>zZ+^Nn^H30cja;4
zk(z0jC&pgzKIru%LNgzb=9<1t)Z{oh`!UuZH;UqIhmY=zo%7Z?#@i(BSVIhPh#!n$
za3)))INKc$2BTRbBSRyE`+>$D_B(`5!wt`{4pyI=hZey0-ePt?(LbZYI4h9G%lECL
z?3Z1;>R#+;9CD~Z`161F=ltt`@khi~c>46rb>E58#P9y&-{YVDv;UImG&8y#*rtA0
z!k|9yZ!CTjR_a>XrX)jGLzDM63!#@z2-;}6F(9bkM@GLDSC~7X=EDE^|MvglFMjn6
z|LEsG<BR3YtMkl1{-6Cm%iT<~*@JMnT~tgNM0FxP=rg8v4De)jO{blHA4AuyPH@uo
z*<gi+58<Y5^8(H#O4(I=C)V&e|5-4Wt|Rn0?read6M3=8jW?QSD~+;k3UzALzMi}N
z9Km^t|IHLfKP{-p4gu*0R#ZuUScP}D%)GuklctFwu$4;gjg$iCW#Y1BTCS8W)9s=f
zxj@m5{mAw{#2UvaO`Ic_LbKfZTnkV8Mz5XWJTWOmCYbm}e4x}!*)nCVeEj&thsP^>
zsU+`M=EVKoiRE;{r$BC%j{3aAJaK<_uc0+5@pQfDSTO0R9qa>G3sSXfbY2S?cLQhH
z?N}R(WbzNkVMP(AAxq+*ui72P5Sdc2J(Jv~2N5A9uOUQU)wmKoAw~@+ie5n+!zya~
zq!_vy3Bhq*Cgy3TSa8PZ6%Mhd_nelQWty0R+B&PvC&ifP$p<C5*|h0+(7B-#h&~Ra
zrqUs`lNywg+4qb%VO}E3DUqf~nj$`Vx>F5AZiQ;h`61S7iLPVuDwvs;#5}7eppJ4&
zYV<=WUO5Q!dEs<g7^4wmAWf0y=T$|Qwefu22~#KeXcRGhEcSMwS_dMPM&^2X;<;4*
z;;;XNuO42}d*g>6zaekWoTkWw547tO-#-4D?!Mstrw@EwpZT4i{epk=v)|<}e<g4}
z^VL84eg5$;{*cEvU-94l(I4Ud{4ZbJ5O&|B^s-~@Hyfs22g4NE{kMEgaD4aO4wq)v
zhyWD}U=eTZc1>#ZhQSSPFg&-Fkr*bIq8+YgrFabjGJ*3zD5X$q(XTgvI2jPgL5KuA
zGeUP-h^yVO;x;AIxw<2@L#z0<>BP`^267Wqi|ZVQz60{|Ia1i$bi>(atq5R$!wVnp
z!RLa%V^wg|383($<KAvf_skv)vSh`^%dMmDxlJ@U)&lZ!e;vr;VFe86_t*sv@9+Vc
z<Fjk;XwD9c;_>~JQuw$3!=FIT)P1Go%w^kgTjt;X>p$k{@e}{-fA`ObcMFX{kQXL{
z{vF2#UyBh<^|Un^LJ%PZZ8oM9IW3buvtrQq7@(`bxi>qgZET-Q=D+#B{y+F1|K!i;
zt??Is{+HxZ`KSNrYeI1R;s4?vBOyXBDjXC$*lpw<&<0hUsKem+#6O!_U}RpAPD;+2
zFpqQ1v>*EY90kJX>l^RI95kI;Gp)^tM4TLuB%NXnn@AngSP6n}UR_FiMq0bkOht|Y
zQ;Ut?<n_}CgbX7^sgL_0#`V$gK5$MG%SqGWO4DLsZ4F70WQWyYFp7GnofD+lSi?I)
zbe~~qJGen`oxmVp^<cR>&q~#6YJF=*>R;FKwCj;kG|JRBjYr<M!ez~Dx#A{Im^^U`
zq&eY}CyYR8g)~px-QDr})hnE6mvDZ#!%N`J+qazO6JZ2u&GcLtedvdtJPk2{P+-(X
zuDU0$y|V9}Tsm9rRB^Z%)l=^U5xtOGYXmpj9(+*xLFw}h2XRig6p&UmVb2M3(vTm0
zpuC_^4_tDq9qfCWg6HBixlQv-tP>D@u!X4Wl3L5HLUs(6s3~iE&aC@NV{qA5T5q=r
zit_?N=suY3$gA$7>(!`h4$pjwobG4lWkI}-dQ~?NRYW8DzlZHd2OaThCF;_sQUOmB
zjnY=ffn#{ltcm+GcjtxPD<9u~RHlSe=+dxVtrlBKgT%sT@}Al9PU-cgVOW;H<G!+f
z_?{1+{({yzfAyDtO36EMIWvMs>c+Q!^=GUf-gCJ=a-Jh!fB8GyFK51Z`xUQVzv0dO
z1ECMXfYUN-wXW>}E-1Rx2O&*~^OCsco$IPxcf}7H3`hR|;)DHzd8yJlXiqvmKWMa&
zuEV_3ZCLlM9g#JTc;}g%BiBl9-5&Oe1n=m2uJ^{CH+t`EWz(^=vFeWyq#?R$l$g|&
zBU=2HQE9vFZugZqDsgnYjhX9EJDGDsbL&G;_6LG=eUJ-IIT+ePtC-q*rj$r25=<1S
zjZd(X&k*m%@X$wJztQuK_~hfjSLcv<<!ML{s<rSQ*G<@F*yny*U6>X?-+XOOl-;g}
z*yp42VX?x;Nc>nCpdDxJpY5WDt$#-6um0lKJbw6yYn56u&WTM4J)?F0>OcN9ZLj=q
z{}2D1^y+NWjzgR9GED(}7OG87g_Mf;jV!@4N6(TXrxci@C&r{Yv@sM4<Y3fcx&oor
z!T<69^55|%|NfWC8OWKO3u`X?oB#NiyuQEV7k}_I=LaPz9tD656E7mTpwm3vtY|<K
z{begZXtxzkh=OyGrA*9A;<~G24y*|jxlJt;@ob+hFj^&V8{0l{4UQ!oYXuF6!}MLC
z@5vG-MlH%R*mhjgby1ALgp~RW4f*e3GH_Igqj#OtSmi1r1Sc$0WS$dF6n1mTJ4FUf
z8g&#z)TYsTV^A1Y)dn9((Q{goS?8eCrgpqhX+<b|Bj<uFQSBtH>n_E-n`*Vy!%Vyo
zT_A-(5us!)2yz|Fi^ruvoDy-GD5GjPPSO;^Wm%Y}iQogLW#QfH*GysJ-K*F581%u=
zvW?+&r+t8!j)>7d6l=TX%(h29Ja2rw?8G@D>F5r0Pa3>+*Po?W)f>2D=z+1!q<PY@
zV$PhSXPG8W^Fnap#d6ej^$>iZIIk-HRwyMgQZOjSG1jwPl<s}#eUMALJ)pfmVv*zR
zL8P>{3)l78c8%JJi5H6pmK$$f#AtA7FfD;NIi_=9IVYA=!l6}f?VgYYL@8#$tFk!g
za!|WGkGc)YhVKRk!l46^!H~vso_O_eW|}5S$;uNS;DKHX-U%@!mb-=C8q*YX9Rcc*
z9*qzjDMaQevOLV3>c(k;vi-=q?ey||+d;O`l`^?r*s6vJ#pT|LL4B}?F?f9bL?*NE
z7p@;Z5JatIS)kB}-V>u|nWK8Az2-}FI~-bTh=&k<yF~+|RcK21{G7g~Lo6|oZ{zc|
zxE8App&4ctnp5juZ&lAW=WzD*_I+dBH*EmQejA_57-na-)ID2EP9Z{w>|@aOtm^EX
zky`L{s=23nSFyD74d)|EP(`_OqS`DCBfz;pjPdrfdFSl&)%<}YmF`FxRLhrA1z#wA
zw#`3OfuD8H%`{MqcI!k33gsD1StNpii;h2ky#L2!&kB{%n(1=H{?2aSmuVf2?)7kb
zn5#kM1^QVWE0WQLzy7PgW^^5o7xem7{ZgQ{&bIIT>Q~?L|M|cFfBEPC`+vr`TNvHe
zL40;?+t1%>(@DqBz55h{?$VQ@(!=3o3c>7^M_Qz3;7!j4zx?Z8^QV9IOM0#3QdqAm
zwKaV3wAT0!-+W8HTshyXApJl(3o3Tk|KV;^nnNtFkI<Kl2p73M*GG-eh08T_xom3p
zu38Xu1|?^+Y<{+iX-&_K(Md5V3un^&9BRE4i!e9>uBI3%RiDc_=bPRlO;Ohryqt-@
z$?K<025Q4a_i{0J4VOTKIeNv$8u+2M#<gVdD!wRa&_zE$5@WoKXQQyBz<El<s4Czs
zE1ajdO4&6BKw86f{Xnn~qr7kQ4rK&}O!|R#9X7SBOzoPhK2#Y@De%?L-tqIFe+eOR
z*{%$M^Zl9mv}mx%Gy@LrBg-^%H!nP#@9{B`%TC#|71hJ6kC8syO<B6_nQhw*WTI-%
z=gWor`$;1-U0_Kwx$8K4h>A0n=2d8=!hje8=hMX9!<l)R_2Q3_hk52SFBZqCyUbcE
zZB%^##Yiay5n<nUYMnr|EA>vjv14eOAYN~rTr1JK^yDL6;QryBd6^ZkGixrx@uY55
zp-DalLhz(n3qjeNc2C~83_&L+IK8>WX~XNiShl>Qb|Lm?Xia-?-MW&l&%UAKd>tA3
z+f<^|f{Ttc2TtcjfA(6<ZdloK=NuC_N0<Yry9qBsS+9J2AGkj6II|-jyWE&`l`&0`
zWr{2*(n=;p^@Q)cL2G(NdZFUj^2S~_9=8vqhc~=`{D#MmKk)wJ_ndv>`Qt|}kMHe)
zqjsS^SDo-gp{YU+M2VT}T5t%Z41*^9b{7LQy_dBRI27|ox~GobOKZBm`i#6LqIwua
z=;D>~DME8<hxMi-*h-=IO6i5&;7?%teCLd%e*`6US}iQg2?>c_wXk4Vq;6|eEB0E~
z30EH~?DP>*OZ3{1=?E|qa4HmxUMFe#&{pbX^nnl(^E4?}Kom%%Hbrg+!suPEaVySs
zl7Wqg74FmPu7Y8On$tZtIWxPm6}ZFJDSE8++rPVUE)L~@$_I{`06Ri-*X@Nq>}33W
zw2e6jhC$Lqc>m*jT@yO(fHk7s3NNqEB&m(R{^hUv(|`4E`LF&r{{?*x_BmrCboWva
z*;q(PvXK-tj`t`6j+HW~3PN3IV&Qoj!1VO#!hie!{v-BWsCi>sceX97iz8|gcv)92
zA3yTN*I&^M;ViZW83#88oeCYrC;iYCIGYe?mV-~)2<g9boX;!s(=+RCtE8^sV+~!C
zVH$zpM01nOM!jCIOg=!Z#9&J?(^~MEUPJSxgwiY9wkcD^J5rhmUT)fnQtNMYi~h99
zz%j}at!Is<jUdb+u%ti?`k=m+%(Y~m){RmN+m@}{f7z8F1{U$o(8g*li@}kr79Xvn
z6>onTSzkBhUzgHZ_g#0bqgz+3;W{$9?yT31ZOh2$EJ67FpMSxB`TKvsyLWH7*20fn
zNngup;dFPW-hCf5TxZ`|_no_GA;d_u)WB`sZXlj{nN^&JS>5cw>TM96@ajBqH+zEX
z8YwD?-ffRlP5nDMxfK1i?$lcGB0QW<ym@ox)vFVyQ(}k~GY{tz%e)XyXMI>Wr}*Do
zZaNFFu^>u5)TA)L9eXEtdjM%(Ii#eMk<+xWZGpSHJMLdU5XEs`PAv1RRkc_|tn&n?
zBfRr8^KifLl(ViYyy!ZnD-XVRvn~nX#L_R7N8p04Y7SOK&Y6-k`?gVwt|KsVSMteL
z8`>o_nofaI#CZr&Cm^jV&1>BoQ93?2mgtE|JM(!8+?^&K{iYpEiYD09AQXm>!bI>h
z&P_Ne^d{t@lDAUzbEa2XcC>xxdA;x-{{0{E`1%bWfB2rMW<LJ#J-<Fj+P*WkO-B{d
z=|#h+WC$rJ&N+ohEk+jWLaUv;D{v{K-);fnt>w4QMmu!D*(9kq4Xoj?9sVuR-En{&
zX>W+5rcNDNV0h=~U6ubgK7er3y#!13^j`ho$qpF2!*`>UieuRK)mx(-$LVq43!0Cz
z@9f?)dm)(n$e~k!Ie9!58YUPhO!HQj<)j#0u>)FX@t+j~FC$|x4?oBMZ!of2t&1OO
zD;#=&-rm#kw=@VQ+8ox!e(pIRw(iw4&^{jn@7%g$D{`C_2RhO}8e{$U1Dkl*+Cd}g
zt)Q_<NQk<gtF`g3|J9%H=Hblm|F8dZ>$;sLSRPhW8_A1!QcOBsv4SFm7Y2oKD@}`G
zEPkL-83_OK-~1{6;g`Rr7A2H!>!#v7gDdsk`S5h%@_f<42hkP7p}UX+E~LYs1Cj;m
zhBz`coPdi?b7SIUx(8vtUR4X^cS>{Gu=RuWpx0)!V4~hT-Gqp>RPw&#<<QY|OGnhI
z?D&kurqKZFy0UM(>Jw0r>QM@&_?z9LKTSaR*;?T98*U1LhiPUC78yEpq<`HimrXI1
zbyp6!?#saWK&@Ks1RM9}){S7N9P49L-kaX+DLNdTb;~?o3%v{bR!A|Bx0$*Z6$S-Q
zD;hcb^y#WBf-yJ+&%?u6!$=6E0Ox5^OI_079+61x8ee<)^u*<Q;fwRbjp7uO9voGK
zV!l*yju;Zh8pR=u+K9pP#r>JDUqA40vd+4{aJvtyj;4WWn&M^Mxvo1fSeD4$!%UoX
ze4K05ot1+T9C1lBP?4pmwM9frv~7E)jKYw?e404lEp#gJ6!DLGc#6}GGX%%^yl{73
zn5Km}&5Th=NxS{oN9MSY;)1xM1rr@FO4^bkJe<$GdAO&oSr6_Q36rPioszZK2|nl=
zW}e73t8K{=4#jz5N(?LN_FU+7z|gAs*c%^lvffAi<qEFr&Q4X(x{uR>-f`n|9^H;J
z!d=rn{uF{z9*2H5wHNk%=X%N1+-aq=UNYMyQ?^1Yjp038E#z8pdm$!nu9*Guo)SUY
zuF=~v*e{Q4>&~7FDLCfjc-jW*TBvQ1q9gB>607Y?8@+l;E;_v#5X*12_xp$XIu5jB
zOzwj!=hb7L7+nROj6;|7(#496YL%qvHE!2wx1PsFFP%jk(Fsw6?9`nx5JP{<%fTUf
zFrUw7T#W2PJ4<gG6p4XW)FGlA6>sWp_tHs^PA@zA?g;Lr&KuFs+8tlZzGv}Hvs$L4
zs{%}HHf+Uk>_o;(k!}vOTY;d@!3w?Hu++zabpVDw7o-g^D`dyVX!hG<Jz-;M`>Wxg
z?77>zLu^#wFvf)_3H3#5q^n$KU*fO{(#^+j>(W}A>B;!4xEqw(_?Lh5r~K3Z>VJ+u
z-O-w`l~u)?wz9zMK{-zoDJFtX#E=MnQdpDef1-&oZ`Mi=Lg3Tm6aV7B{}Y|gmYwyw
z(Z(P}v-%D6nsS4@U6HDGUWF9u&p(E)9Q3oXusPF(cyZdin(#eF4Qq>0r^>Aj=4Ilv
zB!%!*Xx{5OpsOPe(G{!jLz7L--Qk?jsURxo*7x47n}a!G`W0P~4EAkjUo%_o>a4-s
zZGshNfA{OBO$G#f)RAx-!<7FodVv@N!l1Xp^S1JI+1b}dYl^uQyNE;b>eVkc7V|Jf
zLm-k5CT7!aGB*vEa40r&&?p84YE5Wec?Wgd+1Jfb#fece+hybN(=(ULN*my(sNMdy
zao-9~I*S*@gF1(AgDwq<3OSyySIWNQx}r0kS2|iu6XF8hl*KppfT&ffncvqtVVWlH
zAMSW~bx%)$9wWVWYSF`M2;^KW45(4sX9qyg15)S2gE}gEZxq)VNx&)UHcjfa4PqJX
z`+ocVK04mKeZ~E&JGwL;Ufr>N+SK9~9X@#`%aMpvBu?>$ng+|=o+e^ScozXrii>(>
zePjv?xC*UtzB{o!Us<o4j<CFTTl)~eGyB9uU^1a$YmGt$ROcX~`}-|yw=w_qx)X!K
zYLW}K2Rkfh+NqSR?21E~tyV!t2rwnja!QoZC{+WX4ww-LA^Mw#y7z9j$Vx4RB~Ao9
zwKUc>vt0_&1q&fL(&dyDQEDcaojpfN%j|h&^uppqX^6Qp>R|G^`wU)@xYBfYNJm4v
zQ09bmp^MrIM<2B9w>_u`oalpC?D^VFDSjlusaWpiAQZ$jM2CP)Ctl8h^iJu4sRnc?
z?HJL~W1w^;Z#Fzz$?UaY@&ueX<}?w55<^>6^4mez6DJS^=^kmiPCGy_*Z~~Ozc2kb
z{DnD1q72hGXmcB4BF-n~=}a`Ijur#l4td~jOhNE3hx!rP)JFr9!uc}R*K4pDwC$kV
zeXj5q?=;B9f`^8<mld4p5jx;H46#Xqef?u1fhqcrXkMi?qQX{jc!3{&{G@A9xD{{C
zcp|;)v-)sT4Es#qzyHL)`4|5q|NOuHuetg`qEp(-WW@PEO2@&Ua6aOE!uyCbt9G=h
z!4b-ItP{cU-~6BcTRwbzqU<{(XD-`{#$ju-dQ3lJ2z>kF2gY@U(-{?`0|>3ji%{OC
zEVx4m=*>#5lWn<)Llu-1<HTt?ah^}i@sV6qXC&S}gRX9X=>1LKA=cC#D;nn<F6jD0
zLz%RoJ_@=Z9$&Ok-pkI@^~%$A<9;tUXd8OR{6_cZPbvnS6NS?x54F(*tBUl)QWmvx
zoj+Z)Bd^&6ccX%IMG>cfPaX+gM^({597yb(&lC5{0&?c*dgZ$hk6a40ssR?stG`?T
zH#8M5go)8QEf;ntl&WOIZPV$*|Bt#qdzK{2?>n*2?`-Z7ac^c;Rd4798U#TS;7}Yc
zvxvhPGs$esq&Ge4f0Q0IW|EntG0|iaMS=uKh#G>8r5g?OT1#f$CBog$s)ygt{U*pU
zvg!e5sDLG^x-#?LxE}83{LXjHRmCw*JbBB!efRDN@;ObO*;^n))9J65%646NdVJEh
zGB=Q3>21^2scmD_!sHyG5B0wW$2=u&tTfF@Q8C~!mHjj^UuH;=zKR7|bfyq_SG)Fn
zPZLXA87QAG3?vHk`9z#2d~cMz^1R--e|66=3yeCY9D#syQMuWn5w@4R3$O1koX>aU
zQkdpR(*mc2i=OlSnall|%iW37dDhggp>e;xYp!+-kvPqYezZdLfiA)n63HhzLh{;G
z5TYXneK@@{Vd0J<nXbZnPv{U`U<#88qlS8^OU*24CAmP`DtRf$tiI&rJtCg1X&PiH
zg>}^v_l}}@=hbC#nwZZMzIIOMNZXXV3`rwOYpr;%53`v1+BqF0PN$jk-HhuV=aef}
zPOct9j<<oJk8T<Cl|2*xx;%1QpOxzy<`b5|lsuD*+~1wKHQ|fLtF?c6@KCK)QPD@M
z`L$DDD_?J#Kzy(oS+R}q10KAaci=c3V-T&Li2UBG9imMU4OcrU1TLqUDK_JpyP-Rs
z?i0f$p4SDpJkd%sr-)iO8Q3AETdsSz(+b3p6p3O|G;$Z&b?YaP9Z67^LUfKPXl8)3
zpqw#;5NF~vsSwTJUEN2;0&75%CI@1XvzSnFjBZJRFFEsypaF<IQ~QRlyLUhI0Cvs0
z7JTU@{2>5gCtriPyJ%j)(Q>#S`FR;q4m;b}EqAtS0Z&g)EbDq86-pT>0@rrfAGMFh
z;FVWzjsM@j`9uD*-~MgRzxsXLIO+Q9H1{B+#FVtnF-9#L@-8AS;#|VVgb!&CDAf0g
zo{Im;|Lb@8gFpOJZKqsr&>9V`bRP_(CdQEn`RU^m*QY1uyH^zJ(NV`h2h*A%FIp!*
zgl0RZ4w<6Yqz_!z064Y82BLFBZy`p8R`sTXpi(qC9LFShEOy)+AL7lK<(&?K1G;rE
z=z!b^F%W~J<igX_0}l@moX$yU-{|mO51U`+^%ao;XDMP`f1tzCrq;eSH-^(X%7^QX
zb=}C>I4(DcQLR$Lp%J+PV>n`T#N?Uh$la?mchkurJJ@PryA?{-c;M_bE$lRThIja(
zFn5Ykd8we#TrQPc(Gr~2;1q>v^1OL{&*^kRM88f<i5Mp9lGkZE#K;Wv(wO5!3K2%7
zEH}h?%C^wAjf)c!5E#tSb54P6g)$mj@ATTaZJFD)a^E|5(~JwA(u#V>OJSR5Bm`c)
zx^VaU!cq$ouF$i>WTz8v-n`+}>w5rlt@=>zcj6ejpnA`6T}wV%WnDA94=(eG6wlV_
ztq<1e{>0toLK~e|ukU#E>J|6*_e}GNj(p&3P_^?kPt4{Nq3G~7awp&jqIm&`Cddg6
z--SFHNpyG&(Xl>eZp%t(-4;JFdc)Cuw~1X_EL|T&aT-LjHMWj99puQj7-q*BRq31L
z0?8*r@QkkRl*`>r9dJsC<{g*w#Qk}u4R|o2pbXLYX&Y4MlsKIduU?-R8BXVkIKMa#
zx}9jf*OStRt!FW)rBWRy$`4jkeRl>SMdnF|uDcGG``OSoNAp8tbfd<+wq_j>0|uQq
z{8xHZ|H{3>gv1lHrUwH%6%Vrs?wggECx2M57h#aAomfM}^+C-m&Zz~|hoCfD?=N}-
zarn!}G&kYbD``1Vx0T*9(FdZ_*3@0W?@$V&qo3<EC8W=!=sBgJw#eF#R!mQ4GQazj
z5-CPS4CfOeO?vW27jli(4hHqw&uph--H^4YP2qL*?rZP8+5Zhaal>h*!7$z!UNSlC
zS38pZ4Dm;4Hodmi-!S$dB>f0X`=~)m9d@}92RftZSC?+_{krzFokRET2+ENi&9Hm+
zkRcqW>kozn=Xicz`MrPsulO%7uL$>ddJsWX{JYrGyKwO0ZD@bQuC=JOQpcY2p!Cw8
z|MZvq-~S)~RB7tlLM@riDC&@G7!|SjUOnLc^4&9!*C*l_+_s_>%YERaF{D`fpY<L%
zOvGpbb0YQ`YlzY`bwY^LTC{|2*IM*O%|%@wic8?do1of6w8Jw-0ru))brVHzNry7j
zs)-K79sLZv50p`Ox<2sm{&S`&Qc7kHY8(Gc83xn>Hx!o5jaC{{(1kaohy+jVoyT?I
zVOiOX3)df5a8aX-UGT=usM<D7j?)~NPZ6RpqW;;LGxPIC^f$F8+W&!Qyl!+95%ulg
zojAR|!%r$OxylU*SM?Hm$7xF3T~54t{fhH+Vm_UiENo|<v~skx3ccg|(BvuanWvc;
zH8`l|Oeu})(=#;}0+zh!94GHN$Ecb3t!t)1gKf*)Za3CtA$KeNu@k9_9-=4HiRVde
zgqAAUQ(!tzeEVD9<lEo;7N_%Ug(;5O8Yv`&_iCrKtXc%CmwtMF=JEQ>x~%ltXl=0N
z!fjdcqKSqdy?UkXj(2yQrV}a6grHvA$~YjV-aF2FO+btuKeTajc!j(`15jQ%y0IK@
zo$iBlb&bDYj^d*o2~aN9P8}%kKF|P@oGYzpN+7*ci$;J7nnX6IiOcE4l2>ZhBtczB
z40-TMf21Z%&Xb*Rxj!?VCfa~_N=RG|%D@M;QikBQ?8pmliiedcO%WVaJ9%m>D;jAn
zj$A#npSYZ6=6PaUCM)gowy-MfPta%OG<DAB2}FpKS6)x_q_Wq(mfEOh*J@vHU_e{_
z#Iq-e?of_%2OzLZI~1Cgj`NWi61|v^&JdH3CKaU)O(AsRl~!zx*7SaT{pKUOpGdcf
zb(&djPt>v=5P$0$V3wSQgf-NS;KeadGjoc}DUpI*{4uC`qt}e<jTF@HK1X#2MIT7%
zgij}RAy`VF_o8S4@**4-ISx_`cTjTbCen>-Z+R8x5v$DEf9(#0eZwA>XxVE&)ak)Q
zX?raFkZx3@3D}$s5JTHSy>my0!UysL4N<bsL{$V0OL<VSn`OO>-f<SSzU!-Y(dY5C
zhN2*QJnj(q!4H4TcYpIw`R?EPTS|#o^z;y6l*W~f(ZEq~vNEScJ!$ug9<~S48ZsJr
zUHL!$AO26=o^O;o*lss^>pZU;K199G<mK9-G9eysnLqvc7fipB`QmY5DaA4$_JfBa
z(0ly4_li<_Yn&X08l)6C%@c8&7&2I|SDvm5w?!+rYw5?@=ZzBZ=8AB`Q5%$!?N=rG
zsFB^k6xD??#-OyOuC}J@gNSBBJUl)TfBrLC?VQfD3hrIt`j=rX6v33|RyYBq;JKe?
z&ZkKyuQ9mh!qc)*i^eVXVZtr1)4mV>fYT6%^HE_l7b(DFSqXK+qaJ>5A_>P}a{8uQ
z)0>hM@aKtfO0?vOv)b)^Ef4}-bfKT7#ObW9b7c$-DshpRPB<56qm#S(J!ze-REG1k
zAq=eXuXffgb6YoExMQT#i4LX#zGV7xV|!kxxp8)$C>pHPN?}`9nhXv0vDCsc8rfo<
z3xoY*n^WY~$M<~l%};pq>J1?Vp5m1$6nx8^<Cz!(tu?k%*y=_%Z~kprc)l+522ld7
z7S_e_^!P+x@}aJFD1VyLi4^C<3WPzqyYh8!4pT=$AULr+P=&TjubPits&>u~)XFz`
zCKsrq5hl%IpQ2_s6ob|en-Duk-YBhcz1>u-r&yylm>I8onkMcpXXZI^+Ge^}OQakp
zC5{KD^-dWI%Y_s<UCwN!kmg{u(ZUo1Q%rc#bhtLO=Jpg{?40U0ndvg@dS}GLlmdNV
z_0vO>BxB?>CFXg;@yv3|<l5;&=uQ~Hk*2`;lx(xtyonTbQyFcjCZ=~&rJL>UYy9ds
zYO?~afvUlW+FTj-Iosy3J7mNK2$TMP?@FBn!4F3(MI(yQ;4X~+uTl{k2HhO#V^&r5
z=qjr={<rk1pT`iosGg)dhjh(;P(4g!3ZCJ|p;N(o{WDHfyyiWo#6|t>07?P2uC0b+
zRQ5zwC#FrX=DwXH?_;oT8i>Qt`M^6l6!b27)mYdjrWVXbD^>L=O$UKpG!{gTR=~aB
zYexijRG|;Zz2kjgj=>c2yC_Ymy@6Qi*r3#IVQQ+2Dy>t2+q!A_nG9OhVQ&mYI;tfg
z>OP=i@L+bxF*?^};YWY`L%#jp?@`2QmP{Xn$rDl{yTamDW4)D<ouemuIcT)5A`Jfc
zkAK9EfAVwc5XxHUt#aEcxeum&|2N8qF$^K-T(`m>{q!@w|Jw_nzkBB0^Nlg;k!>+r
zJDL{v^+i)9RnQEoBM32YzMQ$czrzP%Tb_A-c;@N0X%|d=iEsY*lO6ZIj&)NT`nqIA
zU0>g;rP9)6d#&VJ*m6CZZ8d#T*tW`NpMJq@srb`{;1eMv3yS(rzrJ#;*@u;x5(cMf
z;?3p4-8?f*Nx8c&l%f!v^0Jz`C*G?#C-{R06reRn$-?ToW@-nBPM9ZQx=YM=Guu`;
zN6)J{TGg-T);fLXV!TjC*FO5uX#!iVl%^bPh#nERZY!TYyocyXDKXDhpIZt~%avR^
zqjf%ix^gR(bD9+Hty(9$KL8k{)ouR?W8ixw-)`KVo^ic#o+sX1X3kU6_{@})&#&r)
zkkQF`(=O|6rAp`R!!yrMD|g*avges#*KVziAO7i|^ZwI!{QRe%<Gtgfk6!cF|K?w_
z?D{>(LIYD`h{sJ-*UWNVczk*yhD0s}yd$J!YTc;$+!o?gYbWQTd@8!B*47VR(mCj2
z3R_Wlu$IcUY7RpWx&VdX5I6AAaXO#4ySp&WGh<s1drvUYs2{?%t$HdjF3vko@5(g>
zuPtyfMo#n0<?h1P8+}>yM86BWM4?iB0E@zQ_I;wIV5Z=?%!xUvEpm#J4HR0jy4Aw{
zB@pL{V5K5HI19o;ht!}2NUb(%DVmG1L(k4JPZ42oy{@drMTVgI0G~YbDVgR4<FdUb
z(zP9GwFS9#!=S#tugB*rrRqT6d(#%S(b+rLccehuJ9mz|<GtR?do8z!ZUD7+J+V8r
zuxV^_qYVu<dtPo(wRozOYA0kd*xVZ6>}&N_Y1Kr+LmNArc!~^0?gu~KN=158SA%%k
z7;M$V%md=^bXH(!xRRCngcuG8d>DtmM|SVxH0Kn~UDi$2;K6gAW=<!VT_8DIKZls1
zyHeUv`|Og*+X~GxTrBG)*v8dqFRlzb#)@eMwyML^?1T23*t8**0WlC2S`Kx4m8`9)
zny#vik3atLk12WME<zqqY?$4Jp8$Ipn^KS3E3gk(Km6g3`K{mjTbw@mHn;1hAz{zx
zKH;3U_c82XW&+JtQITZ@TO)qwcm5}y7tL?z+X}6+Y+32WT|o-=eCa;c#$eRJXYU@#
z>y`KK-}CwVMJdan{JaV^o8{ujfiP&}qG?BbBuq2&wQ+vT_%LX7r7jC)E1DYFchRLB
z(b0SNL+hO_>p^o_Gt06ug>)PW_Q6pRSY1o>mX@Qwdvsx0R>lC|8a|u}vGL_bfv+kc
z^d20>r3fC+IcYG9R&xqVDb(KRL-pQ6EJbVx;<XS64D#z8kE51KZ;nzMbL~vC$Db5J
zPUoP>b!*~&3VibFtPlHEsY@Xv1Q)p7R`7$V=@h<PGmlpleTh|rB6=fi>&nC9Bd;ze
zp0Ao}b}fbLZBdOv&V2gtNG)o2dc0n_)l6%RWy>r%Q*9y>=Ts~>I_s_S{Iv4^;f51<
zbDp@I&z#RQ^OQ7BwKu|myxwT*%CfBLiC<T)+sfx(yyy0C<#ax?pOVCTo^MzF$N$?u
z;Sc}qpYZB5^OGO`DT9GeGi?a};2-=isH3th&v+l`KuUorMWRoXW?GJi8!@cJ=y`Sj
z5tne`Rl0DBC(7vLu5qF=Jaj$b^wMc%&<lF<>1M@igW(FbRZ1yrn+_bUHU?f}VfSce
zAf5tlj>IX^Ml<9{?TOk<nUuDPf$4H$nmuXtY^4!JIeA3Xt~Jj@AHgDh=QJ_JM8Fi!
zt&`WnD48OT%Fy=7m~bHwe8f3(t{68Ny=p_6;<-%G<5MJBbE5b`@-Uj1zr9h1HixZi
zraOy(^}0w5QAA?dGF#CF&3lN!Dc9-Iu$ri!QxNw7BG45E1m%gp-oOA~eEy!4ZUjt}
zwJV)vw}a|h?VLxP*X|Mo7dLee<qXDKtC)|&qAP|d?M{=9Ao&Z56RrC$<)D|`7_?Cp
zYVDARCBofSq$!N92v=j=Fyz9-gE~DB1bPaTJ^jv(+cDi3Lj}yWXaG`(Q8fxKm>_Ui
z9j($|TK}skV{45yXB}Ka<kj6P?z`a9iP21uI~28}JC;0nTq?Kg#(KNa%3gA%@7vkh
z+s&ooFtH(k*ZY3hpI1?=fARV=Gy}o~a2i9rmCCvmO0mY(F>n(2vp@R@Ef?^r*3Yd`
z_qgc+`-Zx^Cls;c5br);7k>OFKjLq``55{jS49&jmA&#<P&XMqfF*S94xkVmZ$JBj
zhleK}a(Y2pRw}z&R=@Yg!|MjG=m$qW&7S*TYh^1+N%y`Vu-Z{6uZW`!Rnl?2(fuAn
ztxkj`XULUaRw&turQ{`WZ5M9}jG={Xtt;xHgVwMd7rl1gk`Q;xt%&Ltysfbv!TAba
zY5BDjrkoL%Xw4rjoByfTS1c+*^SFavsN@}Yr-{?7UG};$Hr86`(#)b}5K-?0w6(3b
zPHt6UmSEx#pjX(|4l;;Sq)Np_IKN3aH@FkeN3ZWm;t20Io}Zt1UN)XBU}-Cb`846Z
z<NCDm{Jc_1AsHkUrl6dB^7NLuE>{rYId42H%fVScuPaa6X8F!g@<wi%TpLg8%HwUN
z7LAC!Op#AsIgrTK3h$p*4Z3lTyHnycC*~>PMf<T!$<(}3mWB1Ya(lXI*-mY&wXr;{
zlr1yPCvwgvL|OU2{$Kt_{_qd}giDIVmg%Et9{V=<r~l9YH^22af0O%<UQ>$3QZ~^^
zIHbrlO}L`20PS$7lu`-exulu9bYcpV4l3g4E)rePqKu9P;FLixLxZ4-#>RR-a8Y@n
znlsnuE4O83@*WwgAE~9WZk19S-aAvz2U^p3*4|a+ul6V{aEdW-Iwhv)nUbfB&XhDH
zri0QzqN1AM94UIz6b{a_wocuea;S{M(&JMkrKHFVs^{o^sGYObBmX;ZjXvOfchQ7M
zr(symtJOzEUX3L$$TM7*m9=DEy}H-N&M9f2kk>F0WN5>q6Y{8JDf9p?sjbT=wKxVB
z^h70siymLK+v0$Rzs9dmKmDcZA@=SIai$Hh4VOJdPGc$+6%ml2uzud?qiKGGS(>rK
z-kxacb}fPlu5$0xs`k7P)DkK{ZQ1OEdMc}}G1hG7&q13KwH1cf(|;+N3M!)CYl?wd
zBhh;c^%>-xQD}V_8c_~}NZToQ0UzqeH>1Bg1QhMk7E?WiiVKu3R0+s@;p8H|y*Nx-
zZLDkN*?L#D)iUb$xM#gg`aQ-_MAB-9L68s;r?i;&p6DEmqKU1wF`OpZr8MEZ>K$&&
z!fo9QX&DTsVM2PCtK=*!TcuQ`f>O}UYNs19y}};I#85Q^;`qZK{+RFo*54+i69KXG
zR<%?fp-iIv6=DTm+D@w{?w|kcXUe@Zpl-O)xwgs?d2NHpOSP}U^0jD;vk#7FP9$Ve
zEvoEDk{m%wPJ+?oT`6w2)2Nz9feDCPFPhkEVLW@At^OII`@N$;>64?lYiNJWa7sai
zpyJg1Yl{<Hlb7&5=V;P3n%FG&qbn+v^G4z?X&A6PMCHgt2+l+_Q?yf?iAqE_iQr%Q
zykrcG@N|w=8nvpIbR^xdoH%fu=3uKGw^iKim?w`v2Tsm0;fZ%M(i(9Nh#QpLSk|m%
zKIcS8x@gx5Th=Md2aU{Rj=0N2I?V(hSo)yl%+~aT=S(nChZY<O%G;e%VwcdCym7rg
zQ;Xj4(Qe$)8aY>%O`Q+XJ1%qNa-N7&ppC|Lz4EjwjnLMOho^=2@1N<wx)q+ER_aat
z@;F_j%Cb_|h4-I-!Q=yz8!&__5jQ+;G?wQ_e&=8O3;yx{`X95_mE5%A*Mk<SM9-)~
ziW5^v>I|@dpIl@L6Uj#e<#8toL>H`Q-EO2lXt}cF%x&3l6lM=qrX%g|cFUBlFwL5k
zzO9*F3hS-1+%kKOC#cK71#eKPLHl-pNzmY=6e1xM+LB2=kV2&NjtpnKs-|Fi36_DN
zg(9sq@>(fdBMIsjK#{HB1EUn>h;z3n&~7Jk<+Pt~8@Jo08i}ZX#|0JnIq#^sa$7bY
zo^Sdv(yjc;GoPj-m@1uS!dZokTj^9MW~oIZv*Vy{mDcr?pmE(oMQxTYcwB8JlKXm3
z13&rsFNiVP!X$vxg%x#Ph;&2aUrKhyP@1GO7_B07Q*#G6Xp><Jrj`bEYj_+6t!j3G
zs`rEU3{045!~7W91)@S{-P}gk-VE0<HUI8CS0qzuLru)v{r$t><JQq(yH@g=G4h-c
z6NfUdYtYXCgPDV#<iiJvcOlY-ko(|PD&$HWN`DOtD%&=_k43z-HR|xw436H1EHS#N
zb=11(^vQ|ds0TYaw_0gvJ$!G1R7=(q4a6DFvieOR>3I1VoNc(JbxQ44+9tH-R7ifH
z1RPPzJ5(;f!{ZbG`gebyzyII-BhKdwr55sLg<XioFE~wF_13?jVMCsWckfwlSEARR
ziq;APC{696FB%99tqM-aRr6f-GNNVOXr*cx6%SgsjUFPMbvEefUPXfee+((=mQXQf
zYYiv*PrB|Y&Igc=-}P8#QLJOoo9VNfA{L<fIthU^&BQcm*Nf@7L>%6YLv*_r`)Md&
zP=TmnKx1ft8S>?vNMDgAaQLJPBefHw*=o$rg40yKJepDhA?Q@t3^J-(IMeH(ZVl(v
zD${{E&7?VMkkKe?wZS!``8ir<lpU-dJVHrA3+mAn1iW%{Dd~x`wQe>JeH{FhU{1Kx
zj60tdP8ox>tSbD9f#mdJx^>kAtQeT5#A!;1Xx!*lGLKJJS}mHf;57M6KzZX_8C8pL
zUf-R$JD>GBpk*D`<*KUsE%SWcShq%XgYtai`Qb*{Dp8zTR~SfbJimR%!@Eb4K5-hG
zYKvqOhaZlA`)~dy{>I<_b#9|_U9XI;83WU4(g)t4sTcxpF854vQlwxEdad-To+3XW
ziqnbR#6{{2hA-ZJ!T<0-{GYi!KQQ@0^uos<f5d<P-~VG=#Uf?(o^G|X-7=$A>RKq9
zx&wxJx_3LO9q^9VKsWD_JEQ8bH@YAwZ6QM!0<pN>ZQZMgRS-q%)KXa1RWk~Lqe!KU
z-G-wb>b(tWZhHTfPR&ghh%NK*^rUVOrvrp@>ZS-Wk&A|2wWh#pObM4WrBz(g430iJ
zxo&i6_~h}?;e#Xh%AoBsZCdRt*NrTN+x^BQfi+j!&_&n{A@-m(?O$^{!Q*vdnzU*-
zgai_3!y{-rC=A=kch``f%=!*x4{adyqU(thPwkb#;B*p#kH$-P^NQ<9D+W#4?0t|&
zF=co+A*1>7U+83|DZD4dM2JZds~sg9O(`R16|wBD8zD&4&0Y1b474DPp){X!!T#*f
zHEX<t>xc{BJmNp7Fz2Q8Y~R#j7*F14R?m#V+&%<aZ-^*N>aAx(1bWS|XEiuajJm;i
z6UU0v#KpZ6VT`JwUVd*hjr88tZaoH)Gqn^GE9ziAy6Re-Q>TVEbS8MAcW=Q`<5-UY
zI|b{yRBGk3Pk+hp|Eqt^-}{IEg}N>ris(6OWVK=3iKtl1Tp2(4^B=1K*9kQjMI`LA
z_40v@^QeK|`!B@y8?B!1ZI5SmyMzVs8k1n9N~)JqTi`$s4+BM8&Gl)U^*$`=@Zk|g
z(*vfZovS_595B1PBsk9q>KvG+gpYb}I8k9UBFAt7ib@0^Mcr#_>(u&U`*()4eVNx+
z8wNtqHz!u1m?owa55IM-m21n?BTa1B{X6Jw&_p-?Ry69<HGK-jD;MWOz&W+UI45ew
zC~D8iEz^?l9!Cm+#Z|`SadW`;PB>|zo~9sri%wQAu@}+Ay(#MAVma4Ngr#n5bz}BP
zIb&>+bb30SLnNiZoMxOT2l?)F<>V)NYrMX@FvST8K}8^?s^!VMqry|69LQa{$~LqZ
zXsMN4bQl;_V{iLfNii~~$T>ymjklkE#%G^@!FifA^>7~s?4;0!@cg{;C%^v({OWIg
zkLT+RTIKG1Vww^TJ^fU>ai{sDetd*&U0H5d`slczVneAN*L35ay>N=2Km5TT^E<!u
zFM0UlQ-TZ=EDj#tzU7~c#(({f{-G``U897T71AXuzgjz96cK8<(5;b7p+hr_S^`=(
zk<#P)g?A4NkIxIL(pF3BK0nv26{XvzzT9f=1Q|-}TyHm)>#EIP;%F*S*Sz=Itr%q3
z*Qk`WkZ(7(b>)kP2TJV(=Z_j%Rft30^yDMrnL^?cCzjgS%SxK623X2U?)l(vC1@y=
z*bubc3QuouJiT9$;Ca4ZX)$qIHgc(SJ0ZK|aaAAkuk&k9u4kJl4*&rG07*naREy4A
zX;7kdwW=XO7h_bkbu79*(8V3%EpfW1X>7IN8m1ZW8uAi?F3?>Q33r>Rv&FqNa}D%*
zpay%I977!`^XWpGPsH;{PxHOgO2tQgj$$zG);dM}9;(%~ag?E!+S<F?*+3f;LpYN1
z#OREH)5JO}FG|y-+51l25nW|hdLncHx@qjFheFjR2=ozL!-Yx|(dga~7^2-PWWfax
z(FE0yCcF>$VCdgyP(^joUC|V+zEdx~kNfFV*T-yHiQDzY^V8!oSOw$al@91fI8!j|
zY_HA!x4JP>^ePT)LeDF2Km8@Y|9ij3-~N05`GF*Auw7K@?1QcVdgm8E{Tb`^nbQ=Z
z6iY_+T0SKjx|6~#O0~2?cIX~Of@|ZGH?Mf}>Ykr3SN*JpjbLNfIlO2FV6TVP&t}*^
zHTVksj6@I5!8u}tI07=f-LINdTe@mfa(?M7*pVh59P>1(joYY|J*-RiU&lep5+X%y
zsjW36jAIaTF@33L^s8wGoCYh11+46xxHp?%?SovDC*DOKqv|BsZHjZve&67{=5>MY
zRe6p$sp4Btc$HGMS~pG#YdwxtPj+F@r-7e6()HxpT2;y*!}`2YwxWEY<%16qD(W(D
zUVZIr*?7)3R$m!t@K7m@x^673^K?$!&kovPo)W=1UR^F++m$p={K`ik^UXIO6QkqP
zp~FCKrpRw90@IdYo%EXsr4PzLt5c0fDy34h4ni?#Me${xhzPf;ryR90d4sQB{KeuN
zb-=&;y+2Up{>QHweK1ctz_i+w?|WXT=Gkt*^S1Ddcb_wW*G<EiVhHqBD0!uAE8aVv
zo^Sll|L33c^za2iI^gJ{)4m(<qaXe$f9E%Ujn{MHv({M_xVFY!u1Zl1ouHe+vt=vB
zY3`P+y--CFqxHJ+@VN5!Df4ivoMsg^wbCgl^jA!%x@|=ViPkkf7AQGueDK<t`oOtP
zcL2CRN56y7bs#`IwRX0xQrlp?u3Vq4EZ1k2b>;2jl~#v_g5dT88C$ND(%Fg*B9jXw
zAIKr#1?G9y*3D9mLKW`=3|$0EZ)|Jf<b~T)=KZ^kr-zN0BI{Ej%!9HuYSl{Al!Dr5
zd!>9y-tVvKs}GBqolcq_=OeCzh$oTso-3*(-{W;<xAqPgeXj@~cFKc2YrnN_P6XXt
zeH?gG>F+1x+B6?xfaSs)*X#U&JjOJW<_pvLOt9oY&%g%_G;$&Y8T4L^)7Ak+?BGy~
zmZKq_G1R)e+XJ2VN4br27H&6Sj|vu15#e6T+s#Jnoet;>OQ75N^AD8z&S}9?@Sfo`
z!$lZ`;7o`K!A|gq*%ebVue-*AGr$E$-1Qp9$zz+4SRP0SYG>Y48%+wJc_!TuV)XsB
zp;gJC8#1*ambe+s{o@6^1Zvd|o3^cd_UUIV|NM9O=C{7hH^21_(w(-EO5FK#U<jZ7
z>?i!_PyU$GJRzg=u-+I=ok9&2L^`YbO;aN0s=`L}usOz_2QhR&nkI%@h*32of*eDm
z#y$vpWVSQArij?V2(xgygKl<$Cf!CM66wP;YDT)H%Nl~TTR8XiVi!3Alu{`^BhBM5
zr_$I5N;!s*Zl5jo+0|O{_P&DqOP~p~Q(J{J)8x53C#JafN$cjfW>uTg2UJ!49^%V3
z=+(3X`vVCtlX4SUvlcrS7^9POL%bt8tqLZp4MQdfvtyb9y(@=P*UIy9queUb56_gX
z;+@yYP+%#A=WQdGOcqCAu$0W>cBPB3PAh@H^|rBI7v>>6zJE`o@|>PI&1Y^~<@tJJ
zxh;gt2~WohB<GoYJZyQ*)VH~HE<r<UDxDBE){)75Xrp8A1gCtl8_NB};F)6JZa#5I
z6W6eiriA(Jj03bqp=(5H$5DVE{JS6VU;S5q8?l70vSn`L%G3J?p58qX@KzNH59^Jm
zyt4MfcWK<W!aPlE<;HEjk++3E{J{_T;){Pl*={;mjPTOlHN0k;_r~-4w|sQDqc#2c
zx2BozQ&5p$w^Mknjl8MY%DJFiv2$QUVO}+<@Zoyn@w$?8BP2^!tW~X}A~sETZns4t
zvZB$ph>*+1ZCP2C%G?#Y97&t;#?a=$H5aa1rfh?y4W8B;>vkotne|riUWIFPwHlUE
zDY?+<pj1_zZ)PWK3{IB|Z$5g%C$B$d%^P_e+}<xlc_O*MD4kv^ER`|_kMC|gzQ0n|
zh8Ll3nM|kUraG5q?Vt`Mc-=U@#;?<SVm_TYoi89>vl+Z7PmZG%J@E}23IQ!;8Q_Md
zGVXc)Dz;JuxFSiF9E!r~=+Nv8dP*^o(xm*c_uxFaY`l~v>8W>~FNB~Q!s#><rx|g3
z@tU5<b&4)prsMbbxt|<iNT%C8E-ny~Wpjk6C<Po5xI<Lg-})Xf^s?E;l(b6OICAG4
zM>7xra=iE5%DN9eb*C@(P64l{;@<YgSq*H9DIS~M0J>Otl+%cH4G`1ZmF~ULbGGK}
zgF=W(HSGW0c~2OMg!G{n^M*!}XgsgjK%@%)IR!=YoC*WiT&S|*hv$7ReDTXq`Pu7N
z{M~=}5BbjbehorId_d@|PdEPbkN$`$1fmyiOCdKE-`$FKznDDp`1zgpNUy3N2~lmX
zxd6gXK6}e^hJuF}2yQ59!;WmfL}wdE#O`eJqC<xs06czJf5vEp5SS)U5ExyxAvNo@
zb)uRFP1!cRh|xf$Y!<$@G3vw|Ji&`mRPwTCjt|~n+pXo;eenVR^(9_kF@aET66b{=
za1N1ZZw}`ut+8x{+8V8DS|MUV8O?6`eN_{io}4!{%KkShUK2Oy&VwjVH%&=D_~6Mk
zlS`&M$4G-Pf^L@R!f?H=TyGntHYQOOZy$qFDo@WhT(2x+Fa@=i)NLaVsJB9pz`AwH
zlJU~no(u2m#yc07(nM{-+sB35qBgo$rxV|Lb<b9nN87f|;4R&#bxQA+K^-*Us0*#Z
zQkwealh?RiREU#e;4~+u6bRmPf4=b1N3Xd(K0~+WD$$!@|E%``;oZZPfA&xRHGk*t
z|7UE=jfcl8jw?U^{Bu74{2iaXy0;0_vE|Ie?U_+EV59WHrJc#^m8a_?Kls6)@b7-`
zN4&n9@Gj7-=;Om)bPWp{w7T)#Z+=TT_*__;hLfn?QE3+$qCVxMbW!l#^3HerjwKZ4
z!TOxpawFHyX(&wBdZY9X+aPaRKvT9#GIU|spHr)&20ay35HWIPG_|#9a7SmeG_ck>
zd0E-68`pK^_Ox+-A5~G@w11%FVn{+KS4FqlXsRU%fme60xXky=^BvA5E@|aFU6}lY
zbfIo4D68j67bsh$l%@kpgEcFSbIX-ZM^%RAQSXBgzTQPeX*zK}-;>TKgY0!9^+5%i
zrp2dY;wyG;xr60dH1uRKs`BXF#E1(2SFIy9O^8xYoe+ao%Nlp@L&9fwK<#_w_k23x
zTm++rQq+3mU^(`^>U}eYV~n7J)FX64pG6-XQU}roB2Cj(gH~f}H;W3uLYh=jJ_NB-
z_Ky9ku0VsRcD;%Y`?^*!6#lKUhX{>hAW%qM`(8R&QmNSk#h%-puQp2~wkg3l)^X<@
zRHY6^*R`DPXanuYuol^Ankd^BV7VqL*i%}qSb}CBohCwx24#CqIt2ZBoef2O3`$px
z)VSRc2gwW9cZJ{om%qy&|Ir_-P2G@^ylrf^XXY5GwUSG<%3pPi_#oDRsE(07tY;+!
zK!2}~!J0K1zK_A%r<L4=6avkg9>U%uMoYA6wn{sZYhj-{MpUn6w(CPPFb+*eZv|fi
zO;n?yqE<m9(1r>bg+Z;E(FWVLv6YRIH`=UG>JpT8)`N;1?vZBiMVzyomSGtlP!AdZ
zfBf|oy9v}{;+^LdqDC^>g=j;=SMr*5^6AE;3^XcN2W%?@F$JXe8o&U@gg?!i_TOp5
zDUqVbTg7qB+Li4kG6hEqm3RS9;O=|^={!HK)Gu!2t>_0Ac9^ZxwoC;^+cX4XQqIf|
zRSvhB>7&uRay^rWb8y^86{M`ojpwEF^ssSTD~mK{sk~0wP4L-gpYi;>BItmWQqp0g
zYv%q|8nvtKuZ%%ViOIv19HTUDqvP;I@0lj$=qgZ$dC$kdi5j5#K@(_j$8Z+LA&#f(
z%EM>x_*>ukgtu=W*xJUQ{p6QCy}k14bW)L{b6VBAEIh}^=kMQAN8`S3Y_}`_;&=Zo
zfAE7p;>|he18X5G!%%@8;zIO6`t<X+{MvVa)w0vEkfLVDl`d?8H<6(=aJfC?Nbi+q
zLu9om;#wPfC`UFA_GTXJmMghrS~a`X(34y5ja(YbZKLK&_dzG+QrU9R6Mof-?QXWS
zfv_%_ZK+(JS8k6h=lN`!6&0i9t+HIVBYL%Ml~xq(aLzNui8*PRQ*SCdV+@?3VYjvM
zd@bBvvJRoW>!39BTHV$Saj@oMi@xp;Q;HyiTl{*54o%aE5H*DoguE5iQrK~Kk9u|c
zLN{L6Q6D^RM5^puud@@-OEJ%$EZ18@yz<aHWbIIA#_T=fs^+BFCN2h>Pg4To?#Hu}
zw%gB^ApKZUol_B^ehGcA&N)1Kve&RRkBo8f<Id^lxOWzq{9YYD_A?7g>1iFZxAL1E
zmC#?i6>-lHz|etx@#h%SE#POU2pADdtg~lj*Psl|PuPK{J+)IoNMVeomSX<_tlm35
zDW&n*FMhtu0eQM@oaaf?IQOrW8&w7)mm}VJO5wP6`{$-7WlR*Ac{K-4PLW61o_T&-
z8F4ze@?(P?iW-LCnZrboT|5W5MzXntdLOz^)oz-uV0GadMm7MKX8QA!$}lW+$vdwD
zxzXIMs+V7fAG28Q_o!1%f?5RC&7n4A6?-d^paNBIdX+Asxe9Gp0t{T&9G_-;;^<wi
z>Ss-f6|X%I+B7MQ#<8dF&(I4f@E0ki|Jds*4FmJ60v(4_9>Q+U9WHBCxqpXFrQLow
z)0#0p7*O`y>ed>rcxvc~SKd7P&hQE+O_R42GWD5y33zdIKd7-NVvuI;PbU*2J@fX{
zCrs60!5YIXXBlw=H!8hknjiS!n5IO_m15kB@|!}8j>~D{)kV7r-d{H!ZVFX58p(HN
z-+6rdDep()&wly^kJnYN)ubgRK6tKqrS;6M6t-3vHVioL`1Z$V<{(IEJUU0u1Lq;0
zB9a0dosGgL-}#syKRhXvWM?LY2~|;zGr<oU&-%xI@?*AAx&OvT<mh?(*&}VKY<c7P
zdZhu^yef}Z8{2i^U2DAi;vGNuH-Est{=I+0t4S02x+v#5dN<y(zbumW`qt|+uGhoQ
zTx%zY5XF-OqUg!I+poRzd>?W&39eDIh7sxM4%G@+x?C=mIz+Sjb0cqsl2xHQTDKgB
z!n##%OQv@;z=KXNO@*r_N*!HKDlrJrIede%6s}Jzx9386qp5=3vLUv0<MzDjCfx^h
zt>l|_DBuRQs3@^F)IHPn`_x?3sgN7Fp;4|oeP#n4D5awV^T6+kX~SC5P!ZPG`4x6u
zQLU&|PxC~2uj<UO4QPMun|$<|ukM|;`*vFgV5mJ4j?ySE`-#Z{ZcOY~iZ+ZoF(I1;
z{OsQ+#(3DzaLztEg_FkzO6+c`Z*TTIJuEg8ss%0BO6h>_z33ni{Ep=9o_p=t@J`KU
z*y+AmYT;;(;jHUEB`r@YTCk+>v#RjD*Ib5r@Yte(SKaQFq0fZYA-I=;sdxG)N6?Nl
zy~RN}G#h42&!o?7(?LcYU0ybraU2jjwJNXW-JtI^)m}H>KD1=*B+2%S3PVfoa><lF
z*s5B6?@x)#95im)3tk#q&*bVEKC%Nwb*T2qiMo1b51u`(4-$ji#z71A)@ZdVdSZy6
zR^EH1xZ9u_6+6t$;f)FjdLl(!z^VEeny0pUAl$o_W!NM?I+{6AtEQ*g+8kp-cksQF
zMA5*uQ&D0L02>-xA7*>DYi)=HqB<5C>H>4EBTmKZS_q~G8U8YI2JYXe>UUIydo-U(
zp_tJ%zIK?mQuaZ>YBrC_MIS-CS9OZ3twB~x%uN>7Qt^`~1tEG>!TMlOj1SClrn*kA
zD%3fjCc+%GE_rWrvH%BBoQI0ulp+eVH7C@evB%Ck(lpUWp*EeieTt+>h*KoZDn4|r
z!`$$#lV;%?cN5?J#w%V=fuFy9%iG6iu3M%Kbp*s12q{o|Wi5@er_Kc^bf!7*(d!FI
z2G377oO3i8#5s}96C-(s7yKOg{%?GjKl|f9XXhx9m-H}trI&#a7~V=-#^BF?@(b1<
z{y8gxq>20aM9wQuw?}dtEUoeSbmpV^gy8tupZ%QQ``v%T_5A~vpwm<*9B{0NQxC6A
zuI?x{aa=BEjW{;t_vl?A(_&>ZLu;A4i7Pd+OFyS>g8;_SJKO>F?oO4G3+s{@G1_3M
zYHnc=z1kw1@%qb}x!pFJMJkU0GC6wgYH!^-YQJ^7x@h0PjAMCRczk!|wq#q}vi>TW
zZC%JYTVyNTt}E}}J@JjFD=s+BYgXugD~f95M#+V|7E2Ut7JI8A+g7`MhPz&79Q^nT
zEZuDA`|{KMU)9$zDo9g39mTXq>c>*weQS-;ss+mIo3{rvw?n+Aigr8<H}GE9j~F8{
z1oMpVsDh$wV%-P5SEK0+i8S8(NTTZ<k*>mLLG>pC<;x9;tEH-bz$@0SVMCiLw#~JW
zrb(aUS`k#dtWdCu263o!!YD1k!f(_K1Dz29#D?z%`R+(#(5wwbQv-DbYSj~N89E$=
z;Bk_4fUat@6)<$*g<kCK#g*WA5x)&XIf4&3>4)=zp>V$J=m+BnR)WD4G~DLtwsM*$
zwcqVk-}MCzcJ@5?syl~xPcD_MXmae`De>k!;RSVh*ny$8jY3D_L>Hx4c9QgzybWgK
z=o3(*fKn1s0>gQ>bpgV`sf!46QUQGL!v+h0-M%L;1FD_mgG3ksYNh9)2$5hRXh44n
zNz$Aa%`n+=A?j47_EB+Q)IhG<ym*8w8QtTp3~67_+c4T$_Pup~Uk>MX%F9sWt|(c_
z86keDXY?x#1Ft@spbKRYl7-UjR*PCYYpHCtqD+9;BqjFXd~hT5m+7q!g>Pg)xn*kU
zOuZw+Q~cnebkfO_CQk~^ngX3|NEY#G{R9oQ?3G$*y;}32dc2!Bh8LPgIhwBUjSL|U
zrwc_i4M9~=V8@dk6dbuVu5@w~{9PnU<oxQw{PDd`|I3vo-grCZ5{JO>jUgIlf>X|}
z6Y<G0OTY!aewxRisg9R<(w4xWVILCUd%yM#?p~j`y}uzrts?u$yyZ=u7^f3%ij1xy
zOnAo}9Pi&g@r!pq=k7G|=`T)v|F3@!A3fknZ(j3Q8=pTs@z4L||CcE^?vjI6=n|}<
z%mP(vVbr35AA8oi2$P53{*7PZN55DYt<cd0U|9<}X>#MJg}iCh^VBpyeBU6n&qD25
z9yrIud5**&WTjf^MeT$>B-57+YO4h26#g|nv#yPDD=bT)m9DN61mAQ4>SeI4N*9Fa
znWscd4nJUdT=?R%XO`OrLm|PuW%8CaP^ncybdZI2Pb=rQ&wTU!BhzW-Zp|t_TnnQc
zLeP!!s0xP*!m@0L44!XUxo<n=>>PUQ)>PQMCb=~b9HoD~^N4G0NG_TY5TeEhi#4`6
zeg5@ys=jaSw)Y4{?QYT$KaLhYhx*ecI6ZYa4V)64{Ruw4^rBd)jd$+In+FaGs+4Lx
zf5o{-7{N|?!=PmSXU-ZT<YGWvW7NiSyB^^;E?O{Ecl7Sef+)(jOFNYJHuKs>(Wpz-
z!7*=yIPfwvVqgrtXB*0Ew^8wubiH&Qi1s~p(V&(%_14L?z)-7QtvlVKf<i{1bbYY)
zp*lF*LY=jKn!v6#J%Q79L83kP*c@jE!PYep>S@V{2vbT72GKjBPY0DJViwP0!9b<)
zv}S4_Tr57l4QS>98+~xAoz0@6rE0`L--i<?^yZHt&pWS3$mkX_W~iDAxOAf1fyH|v
zjIOyAc2AA2QPJuI!OpBNbP}jtun(=hPLj-7(T{`f*yRw0pn4HX83@IM@CN1Yh?DC&
zbd9F8s@)grTF~+-hBLt_2mONGOS|g?*=e@p)6wB_+lca)U@bhoI^&vl)rY2Yx7cLe
z#$fBsa{3i!**9sOu;iG0j+el6nK<3gwA{FD8Y3Btha@K6t?ErbNBy2{WlEhkI@`7}
zx|UwV9ND&w>%$}K^X*8enxiL$cw`6QRq?siO0$Qd)`n=N4c!S@CGmLa+;`8}d5XXq
z0yR}a@=U>#E;A!dtfTYPg|+M5;6?qUqjwd~B}Y03Qm|IQ)|ir`2Q36Gxw5PUH-uN`
ziOb!DPg>(T{Gg5o0p5J_nuoW~Ork>!HgF9)DYv}Qqc*R3A3>DIy1%=yS)TZ|Z2aNB
z{ow)i|49f0uh|iE5H10Q`iDiA8U#21nr8j$4X+C(z4Q7awB?ES&zov8(3ZjLZRHwu
z>ZVj~w;OreC`C`$y{Yxc+oT(u=gn#6SKeIs?l)fX>8Ed1$7HvRT0{dHp<Wq7V>nZI
zadB*0=K8#`u32GdIt;IM&$+VPR%$L#yLQp1NC=+W^THSTi68&$Q?_NJESc>&vt3u7
zZ&%*FePq3E?8zc)E<9gXZjUz@j`g;YmqMAXK&1dpG;h3FS{jbEH0o_**{lNr=$Ks*
zqjE)qTNs1MdF7A4WN^q|)mO<`IoHsLLnysziI(Ulw5#K5>C{}1fu3^O032aQMvo;$
zYhQN-H+I6Uty;E#C^8qk9YU0++}C$qi*z6zW9OlVD&eKmTm?I0c0)aLF1p}$6Etp_
z_4cfyt`L~!nGiLh&LKz4?5<ttW(Dr!g;VdXQA%dHEwq*aPnsr0MtsnA*F#Kb)-|&-
zju;4~DwX8504I5;bgkGrT=rp9GyE~s%{!r^Yelnx!aEH++fSrUA9{t;qsABW!7y;U
z-gtWVPMaO0TK-1QtXm=aU`VVS!wM*!*G>Gn=0YAquBuy@QY8358(NLLw9Y_SN@uG|
zD|wry_QA?IPjiY+igU;2FQ62tdl?pS0SN-GsXer-D7n@~!}0L&aNLW>K^6|Hp8Eh<
z!BFQT#%PRV;CcDqN*9P&xR51SiW{mS=tD!;T4mJAXia0$&CLOJfJQ%GIrdV=7)>cR
zqrrE3;-1@rh#ubDh7a?XdPcum&Ok_vuKdUBqagz-R@i#wRtmLaYWKa}&K-i37Z}7;
zwNqlcpNSVu_u4oAx^C2niE)IUJ9*t$V`n)v((Fi6Bxm!w78R%Dt+GCEw5%VTccN%Q
zFb=j30?I9>G_f>QthZiqUd1)r(pYbsKA7E;$3zk6A?R_hse-dPp-AJoHOjW}i^rA6
zMK=P41#KgBj&Pco?k0^otvcMC&WT<dL4?OrSk_ANp8M0plst_}?FH!^x`6Ne>Noh|
z4}Q$(4H5MS<2^n`MpGMMniDQ2Bu<En-~;z}iM!KTuRUU+EM^<pYcPR1Iw&i3TlA(*
zp3#Nd^P;NRp_8hQ)4@NA8~mfc^XtSk@!7k#gcvxv&TAsCQ{XOoYH4f(%I!wYD{Z?`
zJTP<;_Aw#@E^*@28|8lH8=u@W&5@;chG?X8ijKQW;xq^K4_9px^iHo)-5R&sm8Ykb
zWyyq~-3EO)B&dMudcCo(S$W3bZPwAOfuFs5;?s9eifTM%p5Cuq?^m9muROg|8(?Xj
zI`mg>gSs|ytZdgpUMqPi%B$8vUJKVXlZ!&FUPLQihw2IJut6VC>}2J<?sk1tWK2XH
z(#9S*_O*TWT50Hl*33tJ__l4++_fg9Xxd<FO1Jf@_o8zSvCy64;=!Rm7(*Kf@jK#R
zzho%X+LVhG+dTH=Tz$_N@9T~QT%7(r_rYcZ=&;wc0IAmsqfu)n^>py?FM2A~#OMa4
z-)G-M+TJ73cY37Z(6l-417g%iN{QJ5iFR~VBZ#|Eb-GzNvlD_MqQM8%vRK@5>s_I1
zyY8MrSM1*2^3GZBLzhFuYr10q7u0QmZEpJ-5WKMU&iOPm#cTpt<>On`wZEG{uy-B)
zci4OP{qGt4xoLRQR=U+JYcnKHuVpI@Z)$w)E3l+S%t;Xf!)uMRqO(F1we7`uX3q%_
z(*)F}x}PB%*M~27^~tw*Mn&6lXe^b21UuF50ZK-1j^p4bHn3`kiF4p-$SBZly%<7w
z+TJ;u>RD6--5I@5iwYETDeV0VPE3Gpx|Y%95Jev1(E&S^>wRF5P5Y9`vV5s$^eYVm
zr8}q)y23LJ6?eIg)-JBuI4Hfr_UJ~Nl@J$by)r7w%g>Sd^@-EHp6<%6uv)&ocxdT^
z@P*b`u9@4@jrkP0+@GvCYLI(p4J&zT+@3c^9ndY&(75O59j{X2o2L^#c%HV6d{el1
z4-i3w^`;GitvBX4AEh)y{4tHzrb_T!Gf$iP#eea|Gq*KkmU)BS02#y-n5LxJ>~2?2
zW2ahG*ypP1<vC5n<Z)gXF@_2<b8CF`@f|S*YF_bSMlgl2`CofiFRYtqv?a$PNG-Ay
zi&b^KBcp-OKs<ymt!9SvD!R%U?-HeSmg|*T+VOKpFkOgq{L04{e(U?+WNV4*bK%u#
z#s$x>B;k`cC+?@f!{d!>&eVK^yx_|OdC<Ey1$u9vk#w3zzfTHW?h?<rkZWTpg(j7G
za-7c!L8hufE(2XmbFHlF!sGKZITt=YpZVy`M8>fWr{c+*=jnRGdm*`KmaU=P5VwtG
zQI55j&h_EO<J}YQ-#_yF{>Hi#a%tKnUUV|gOU9?jvaD(aS~FgRd|S9a-?(jAt+u_>
zMd583#xIT++iOSp-t*)R4bD7yPE%ym7Ypgv_0@+80Ch2IhaGF@?Dt-X9pM?h9eoVC
z$?rhapw*#KwfpaNg^>63!X0LHps=sKheifJA>xtUuZJALL8HsDx$5sYt(6{1sfpKv
zjRVL<z4_j2^>?*g`VTi&^!;k}vg)jcIwf@&u!UTy4=uBDerCR$NOJ<S7EaSdN=Y|C
zdnv_%a`_*M552R(En8>AQ-|7CrBtM8X_t0AsCY2AU{s*e8v8**sTHlAM>P3S1EV}G
zuiTz?TD2<5;~bZydV=UZF*-J>)T$@_ef@0ajv6);Xhe0Siz7EJdTIkp*he?swwEtC
z9VWXGm~JQnXuTd|BpAEJ@1-fRI|nq<bkcZe(~iXy@MAE|fj|Ax5Bd8ae?m8OTP)-U
z`@Tm#x_gNx2K1nb4pPBeb-Qx}5ge6KJK{WIL(*Pk)U&SJir47RqgC8hGXus;O6E?r
z_t7<$7&$a|YO?W&sPMB7MYyVS_I%VY^^AU{Vc_|?;kL>>Q5DWX;W!y+gVyy%E%imD
zV<%20Vu2Z=iC-#{CY}@1-H9-(xGP4l%GlD$*NwUredwLjaE?)F2TW*n$U`Tr1tBk$
zb;-)%bdHCockT9mGbdih05MQ%BXu-WzNy8h4eb}$ayD_3-so%I7(M7FP*DyCp|r-k
z>oZ?`@yPA*#yKSJ?h?)mcPHJ{#fgeoyf;2mj<l#toS1@!aja{mS9MLeV42{?r7FbT
zm}B5_nR$M9BX29>)qdE<P(CUIQcA`j?E)C<Vw1pdiQ$vxyq85!F8hLuVWrLy(uvx1
z&|8*O71Z7ng65~YkidCPrxX91|L*Ve`Xc=N{ou`I!ui4Za^j;`iP!ft%6tv{bEcO@
zS~KTXiLK}ZFgjj39xd}|d87ACjPQ+*PCVW=ZjUQZC9_FKf(F&JKJ@yEW4KNmg{^Kp
zTpxM={De5Ty9>O2GqDawSvqUpcy67tECjknG7r@~Xim5;vil|_9m<+{e!OvgzOm)Z
zRvKH=Agn%w$J>TW0ZOB8nX=r-t@HH$k@pW*mYa$O2T;c#IW^8Y(Lq3VFS8d;r^w~x
znS!EP=hKN-=fKnBY74;E_vJ%)IoaANQ5Q~koNf-kycj2boZ>mSqTU9rc0Hha%L}kc
zX|ONM9EihyQs3cJ5w(E@Jum3}Gc>yPm>P7|!nt6!GkxA0>bD=3*r#YgP=4ArPd-3D
z`v>JSf)&;TN3UA(s}Qhm1DKH6d!<CCIVw%y!jYW-imC`IOe@8*Dt6?xn;r|>%$(Pf
zn_c(Nn<~%IwOQ<a>Bj$orSn4+S1+Dw2|{2>31Q%4JW$25WzNyjO6HT-vlTgYqJ!!j
zm8Ky__BkC$F?J1gkyli?59%=P-wC6?XwCM|cUZoIU_wBw;ZPhSZ;DEIZ@iL0_9h<3
z2f{R~K(93d!gSI$+u#PLhz<%)*P|C&TZa~J$%(tzQm>gRLsjtXi77ILK=MGZnz55L
zol>0@-4b_|5*U5gd<>vjxY0Ng4zbTm(Vd1n+(l;0wKT;YY1?%;PBi<+?Ye^h#Otet
z4i!%rLiB+tc#XJ3AGp<2+_h9vvM)P^JxF>#g&6Ro)7(IkXPULwz**tTe4eOFAupS8
zt^?=AI?r_q6fe-UPInho)lxN~ZXH%|(};)(%>dCwI66ni(TWPTq(N-Tm51oLmBvF?
z^=Aya2rW6&p9f6>OH)`25t?XF%&3j0>xOj4H!f#RUTc{@d3EM=nJi(=)TId7`m$A3
z>hUPD(zZsfjg6|Js2!5*P;;d$8+qOM-miX>AH4su4o+6{Db5p9AWre9OT}VHu?%%<
z9yz#wX^P??9Jir}XGnmUR8V(YHC(3+gW?B<Wpln<_}~BU{<r-0C!PBAK+YSl?<Tgk
z^7?+_JKwlto)foP7&oD%p*oFqBeufaHNsa<Bf7Y*n{vseZKN6Ad^9tC;i&H$UpznY
zxZSa}jfp^a1N;E#l)h1N=C<DW;^{r_AFj0Cxtt>(zdG@l8&B(vyj*ef#5Yc!>QB_(
zS@ItK;h2Ks6a&=_lJ`u`F|rB>*R1g5{p`7$wIAcvITGlUZKdXwZM|^|Lf$g(9v*mn
zzR_~i0Hm%P(cKhylhi4*b&W@bf%owGKJo33W==_A-zg=|!SV21sbBJe{;T?O{zW{z
zi)>r1>|X5NDy=uOzwZLBeWNm{){-pUu-m!oda!RQCe~K*o9<u7vp3iW3o#Bk(wga3
z2dW=n(0WCBK*aheMz=xWC1a!2dO-Xl?ywH-&8#|vbh7CrWM~oEzWw67M$(EF7-?CN
zt`m+<l+GzQT#ATjdD=b<)vg%=W9YxD@KmjB!|SPeM@mFMf+P0G?h@EHm%}nC_I}Wx
zO>BtqhVJgj73TViy%N1=ILFU^@)Lr0+|7~bgx(4%fD4n>mY2qD%^;D<skpM&g53gu
z^$hIgK!yYyXi@9t4BJoVLbOIx@p?6@4OYSa{urHBtq{nx5xX13%OKKlRO_)&cgf_z
zN6&eR{Jr1$YcRj#C!asidp*AQZa>xZ!tjC$h?oOGbq~>Ng5;bcLAs&TIslcXu7JMx
z#2DqZJKx;s$I#;3o=~{!T?Sjt#Oettpti)w5Y>JS8!pEfxUN*V6NmJwqRB7u`f72+
zdC+SooE%e#S{aF+rk7=7+lqxi$V;`Sj;Q+3H|4%v2>2<Iq9?RLtiq_uf5ln*vDL=Y
zjvK`pw<OLyon3`j*kp@&zx!5@-tpoIPNAz99p^bwTj%X{<vACgwv}~R33Sde9^#|k
z^>pqn)a05g&$Ut|fP};9q?y*9ZCQDITIsp+$>q#9E@xu&eDvzVG)Hw1V6~>bvaBoX
zmW|6A<lb1fLdhBsB0+CVJq<YRHz<{D&8*K0clR@wbEGUCmlC~c-$9HTR|);_Oz)Ha
z=v{{>CHv@QL;8LivO=k2HkBhvHP%kQW(~Oz^)Yu7ME$q_<=^9PeEY)s_!G8}5T7AL
zx?c&C$0tXPkVW7!k?v+%&RkugjedkDF$O(%nkY25F(@(!b6`GedfYy2HlX%GZdnCV
zG(3(iFZAAdet6{ZiwB+_7G!v$7fxPqQ=ra~OXz%Z65d=UN)EieZq!on;+Rw96dbSS
zKz70@IO1qX-tfI}9|v!~amOd;#FQLUu=nQm`oD>uSS#(e(6^0{H|`ibxIt-FH9O!{
zbbRk_B6;EMb){EzR?Ob>?K$x4uTM<L5mF+0Pu+@ITI=|qdnCU$FYo=KQ`!B`s(A?q
zDsb@awg{#eY-88bs6LuymNcALT&z=sz!@bLybwYpdQHNUVaj#Ulk%=;*Y(MQXxhsu
z?>N0$K#%u>E<3cVVl(A+E2CHaGtM6^oX`%st`=QSLeSddqjpyujXmeUvD+y97)Z@{
zD#29Y!d{_ksFl&K-L6^NHxqH-`+;-}R0=TwoQewtj<rF*&NzlG85V?P5zv64hPu8s
z8*cjO%+pNG8$bW)PxXN1g;5L9Y0Y`-)?rX|FnU^Vd=&014c1Y6u>lKnT4{jMu&8lG
zUBu#fHR|0aF_KrUgGaOm+R_e>A}?{2ruDt8EB%<f@H#32*IH(BlTuFJ)4GFz>B=f^
z&Vl>OM0gDB==x3#?7b1YNOTXEYAqarPu?PiwI;e7g_#c*Lu;MV&Hry}^Dcf(!O>#C
zInPpy(upEE44|J+sf84iMq8V`e^{W>t|8dtig#LLd`VRCD-8p6DV(E6y22>Yl)78*
ztfd@c8Rg0BrbSPGPDH0#XX%4>6ZWmC9cy$_2*}+`AJCpMy=nv3&H-vjg{DPy(Rf=;
zx^)m68=ER%htU!yVk~1&YT@yE<yspLi*^R|wppXqVCsSAZLpZ-MMTJDFgEoSSLwJg
zNTQ`53iVZ<9#)oXAzxP(=UL}SNLE1{^uy>f*eqD%@%hTrb*0zBDneZgpTECSbLAWp
zcjt)|BO&gY;*}lXXu~m@;9_KbcwqJd;)zqjqkPmJfULKVEg}}<YQg}9p#shs*LA#J
z)I;JN(>!yz?}!Ap=O<NePZPiX{qOOc-~A>}@82<<?)YNec%bqujRm2M&g%7a*P>^B
zb>TZ*Q~Tz5LZ*P2wxV@S(&~a`KXjU@r&A18k1f!I)GEF=q!zOeX<C<S9ZKW0W$s(!
zJ99+lz_(9<d+)h;INc}STqeGGcj7W7t}gSqh~>5GgZPbCXM!|__k8E|%<Ge9_VCF$
zar%`voI>FBoS2ihOm?Wfv$jf1Lh2j7G(sy#DAd{U6g{5~4Q>&E^Axy?!eepAV%)nU
zGI-@3-<~I?Sse~T;948fRE7O@emOfFbg>F+a~By_kJk9At0#!Zi=rjot2@W*nvL%b
z*9XPmaPN%A4&hh;Lx{&F$FOIoX(YH;irMH#-EcI~uo<6g>6pf8*RGnBZbY(%0_jQF
zfHkxULA5pt9=wRB#i<3Zv`%muvZP|duJ`cn5b#D7)CIr2r`J29O4L3nL91}ZM3~(+
z&fQ^fWHUI`J&fKhad6n59~zTu>!k|@qr(TzCowID*9KEVZF}!f8>KaB&x`?I{PLHq
z>xw#<b&b^)L{aLZXc^BO9PgLL7f(;Txm>_Ri;bP$)Mp7BJa$w-?b)c%&aP{4s(%U=
z#I;97dnarRLxUX23uOpJH=iS1Otn`X%BDzFi1eaS#-nwX+se9JnZywV_HuL)4R_ij
z^Y@o-_(MMvr(fR<U48q|BEDX=(!RDr**1+Grz=1t-GZ}JY#P1b=ULIhP0=d5h3Q5u
zGR%5wX`K6d4Liz;-kZDSbQrSpC0<`?7+9CWrGtxh)0pjH%Z1hYt@k#nVW8L!cQ>0_
zR1q@p#j8*d)f^}^;+QXIh7SzC(r%iN=jgf-1+TEzX(rBbr?P;-vSehRc6Z=gpfPwZ
zh0m@Fr#A7tY&@=&YW&QU^pucCXKO-fj#3*zU~QJ$ErC*@*Gx}?DQde^tDS9wb#3Ie
z@s5Qld1CT(v?VPVZ&`ccx>i1af8*`@8*|W&Y`t!L@%{=!BaP=t^92m)KeE%sY31<R
z2I<W!Za@8;c|IFgskh6p)BYHHIK>`O98^JUid7X&dAkMw{nw5r=;v{{zoXTGw1W6R
zn$P^N{_%go_rLcMy*=`?Qh9K~`_XvcJI}qb&{#*s`$R{WUE*6Gy+U%~b&Q;n>PBP?
zZQvSu%JU9Fhsjm$g6EZY+<R1rQyZt=xTkT3BYn;di1xr<eB|=cD}F8NX7K9%LP%Py
zR6}5%6Q`h@JvCFystbtM$@=x(#5pvkIq~u9GZzoZ5B%NC$Crs19aHqA4~1{7acMR{
z_DmSsVSh@FI(w#oel~FAw{POO)jg(EPX>MrCJ|yEobaUKXsEDM>Wi}Z>-su+0KDdH
z?EEIq9c4GIH~P@pV45~R>T;)+K&QI{K`CR^qIj>3cJ9MY0%sfi#;B^Q2G5==Fj`ZQ
z^S>v^^P;?c--U(p!72O$s9xWUAgDFcrGNzgqIuZwO&bRG8rq{v#Q0q=0;4dxqM$kt
z40?W|t?#ww+0BHZ`k@d4;yt}vkBM>EuKJhv=(Li#RlvfGhNaNS7{uX8>2wIw_Q7S>
z4()1r=MA0o-rO6C%xQ|`;PcNu$Bls(SWBhmiW`H}bx)|dD#9Zg9M*wc3-grNN+ByB
zpzq(i<M85~=_`ao3xVjs!O*g$-p!dX^l;hCpC7$u3{0YTOdXD`>SkM8*K6nmX_>lt
zh@LWF*&4|WUggeqy>U9d!i%Hw!F|fhkSC_)+Vvs(E%D*M=ylIX)qBmM`UmZ+F}hp7
z>T$0)=kY13*fR)Q>1;VG-NEkAb4Fc#ATI1Lg4JOziv#fyG40BK`t{W`1GPd3x-cCA
z7)?;hwK1F{O-WPtN`cl%QK#FSD{I~J_0g(YS*f|<46ch)LcB1aCx+;>x7{|J3rwfT
z>6}QX8DfBBH|x+6nPq8qv9;pSIYGS8rL(PtrB+hPYF%`WFemLR_d;<I8P4VukL%h_
zm$yMk1D{kFGMd)tHqn0i+B!LRoh*UU2Fq5dwXxMgMU%Pw(1)#-%)5st-aTG<eezHp
z@7^t3AN2!|lhYy%r;6L%r|W|TnhdW|pb|YJI9gMWv5tTBxDk4-3Nb}zx=mq-Kc-N8
z#2O83n$)y36-F6sC1T+kXp`CLj0>LM{+r+9?|%Qx)E{|^aC4rK0-Mv?;-z&es_gdG
zO%VEE_Kx{HbMFJ^sFBii&8HZmk-#`1F*wD*se3jNP6Hhn+91@y8R4}HCfC7K=ej}M
zy^4G?zoLJu9$Z{5SqD+0v_{@o$(A)R45HQkizskO7fy3zj$%3K8n86H*KB-C1N2Q#
zKCRZ5+FdCYl%rmB#!wHl)uamQis`*^ABA7JBwANL`L{0<uVcVBaCIQPYW4M2HN*Hn
zoZ7xtFK4G%KY6v~jjs1~noR7aRj>n%0`9#Ye03C%-jPvoz2a#QJcLM`Cc-pZlyf_T
zX=8wlrW<Vs@j4LKR$+L0mpuRu4>|*XaPZS-$sPYe^pEJq9xW1ibb?or9!{MHqv`wC
zulM0*^5O{DYiQM$R+RP_U4_O@C53Fy^l+wl6)6ULkI*zyU0%8g&;uj;9~tNXqAi|!
zF?v^5QfuZ~=z8BD1I!06ar$+4T6W+NFGw|sTkrhxXFnszfV@$2CRfcS>D`l3VD!OL
z$*i^ydFRMgY2w_P5`3bA+n#W`rxxz8xj08{YRhe{nRpOJ7uH$`7S&t3iKLAZIL!&6
zQ@UQ8-VK55O{vrdIbUh9;nJBtgy*dg!o=1SzkGV4)^fn?-Si?K4h5s@>0eVA^}Puy
zTK8h=hu-j}R|zo^Q`gAgrfY*aTK|IZ&wIgzK(89AHqb-#he?@Z==XNJ&695H-ODfe
z3n<240y5y8;264an^>mx!D?P=MDQsP=ctf%H-0X7g+6l8aEurZjw5gC!SvpIw<8gz
zi1!dC%fx4p?oYgWb4Q#KTkZJENdrP!rHmV;-RMo|Rn@W~C}-Hap3rj{EG7aAQ8}&o
zeA4Pkqdq8#Tu9f=EN>MbI<wRFCv{MIQ8-BKG*h*8EJ`N@%Sd-#brDHROH|F-sCi*s
zujIT^)5NU~o}O3grtM;Bp1|t_y>13&i96uF;W{A-DZqT5SZ)jV7o7~%_v>&WP+Q^Y
z@rnD_uSrRZVRVXB%a*El#URnHHX~ET_OUCak2npB$@T=`Z+-KQ`1A`tYw-E>u}-8}
zLo;fzINn;2T8T|hEFPF$;2Z*H7l}?Qs&UZ$AnXt(g9MZTQU`A9l&n#`v_{E|Q94ny
z8$<fOC=a4IB*91n#|aoVB#P=aYCXKUrB=2rQ%a?^gp0Z{v|$jnMYk3^lSZeM6hjrg
z#c8+1?&GC38=5@B3T=8V+}6U9O+1KFV^avco&%p;65qayw3@+t?xx7=Ia&eE-a;n^
zF}9;K{p<YN<APn#N?h@~`(+pQMVeDRh&V0naeHvtsEkp-G%LZ0LgzsRcnlS2?GSA#
z4cCkE@XqPIW~y(s;uX3y7%u{_utV-cH`F0=Y?wRWek`v;xd!L8Mb3p_Ct!URz~W<V
zGLf!cb%7V&zc&k=DQK*=BPDZawW6p7Roe%<4N$`$Du1m<R+?itjd(T#9u~{<ff`<a
z#?ei{Rx934+4czSbiD}%70TYL(sxL@b=7{AZ6ye_R*W(kM{pLRxf$1@$$_m8_MWZU
zRFyxv=E%gjmll~&(880B-i{qjvGu6+W`$`&EdxwIy5+)JJFn+Rsg3g-SeAkpEjw!V
zO}5gt{Htf`+Hfvv4e<#4wcq&be6hf-s1Brci%=Uc$dbB5^nR#qr((MGMc6f_u^;>g
zqM=b~A`D>~(DSBW+t8vhz?rs8`z1VHI<i$e*_#uBUERfT59reVL4J=>Eg=MgRT`Hs
zNilx4&>@t=3yR^eZJEY6I^7#D&`q2owOi3d^n}*67F1w5O}Hs&BA$s3!7(-sT#&V>
zU1GNliQYJPr?Q)%n|ya2#8ZZ_ESa7y^e2XcZ&u!~*{}jJ1Q0#t?|~Fs=}4<=IV(pG
zCQwmp*<`hs;vhH^Ch28YUTcV{iam2wTS@SN+B>6ZcuMXY*KJ|lH1~d6H*!bw_IoF#
z#1tapunOKnE~;9swbFajj_skAosU1h<L3|0w7lwx-f1uUwq>3djs8vPrk|B~Bq(1g
zy)m$w-z^h7q@6!B$o@!Yvjs&@AOjhV-~P?-v&5M{dwk2-o>lx)3b8bN-e4>EC3D~5
zji}$(i=N6Sr!aAlU<-mhMChrm_l~rN)Jk15+by$f3cIhR@^stCx0P|1kvQql(sgmc
za==G#bQy=m%7<VN&8`nnTUlizELUo6iagYAjcT5hB7<QT9UHJeKujEiT+IH{^gZtb
zS4Z_Mt*C1x*Je33%Gb_8_~dTp{yeET)Kuv{cuvuj!>wsL=27hQ_0JhFBFnTY3i>-s
z14TTo^<Czprz_`iH1=6QL@6a4sv9!ui+z#o8UV9H&KlaJL%sUA&ALdXcF4mleOU6I
z_cVnfUi{soLw|8I?82-MPhEXz=ElJy2n^|Vzkrx<veibTBHelxoYD~vZ6Zx+W)`*7
z24@QKp@Cbw)1WhO4G^_AD$4re(kP`ImO;=4z}6~aP8MezJaz!6rbp7@M<dfa!wzKw
zH@sFI4~=@>X_jC7^yjS0m6IE$=Tb^?^iCU&ttv{|y2A7Oh8Ci?>#S>ZjPamP_S0``
z?XU$ZPp?!Qmd)Y4qgD+r5>YM2T2@Mj$*EJMTOe5P9Tz;ooKPZAa={s5OYP*c;nIoM
zAAd|uXFk6vbsbV-6zizyU7Z!X_2d|m%p%#wq3a4qZk3U+!y!SDisW?v%DqyzVtEiI
z{zDH<ph~akw9{fAKTECpGkb4sygLMD*XOX%?sUS??1m0(`Aa>cU(HRh2Z1=f&~*%I
zZ!EP?9k>*zbnBKD-Moi7z?}E^;51Fn3zH2Ch+47w7_?&H6jJrXZs>tHVcjb3mT^(-
zM_o28#1PMR%`}5kod_v-6(d@;a<!A8fM%Ze!OiQ*&qdGVJ);#`t@xlf>F!V%Xp)?W
zBg}yq{Q>Lr+N@wkR{{+Za-32^9F#g(HieOHW#Qe^jk0AO5CDp>JvVyk$Rxy|_Ley)
z_oZbidK#s7q<6yTBoArw+-fFeA=ZL3&-|bL;(?{>Y5H<^qSvg6fdIX0cR+`2TX}wb
zBE`Vn<xEOJA82gS6&1(yR$(X~D9*7Ec>9pqZk_YAaaK-)(uH&F{Nj0IJZ7dmxEHgI
z+A!dTXjSL1AgA6@bFn^H&AQ#2iEcc?H%^i7+(ot_Y)fT*Tv$&FJt<G<yeg3!Iz8_X
z*xsb+oQ9voq;;X<6!Le|#{25{e0!$0Mu?7iiky=6l4t8r2;T36iVhgwvz4xGh^;bv
zM{oh@T^Dxe0MBi}+8Ws+LscZz6!G#f$H0h~dqv^xJqq?^@;9~caWeaQzdk&{c&DfO
z5CVM&rD!SC4q5x)@rd#02BTvLRhqN`vJcCp7+PFo!8E&VP6S&__tZi~{?w76IS<}q
ze)qUvUF-EdG)vjrolzfZ2#%krcTmy0A_2CMif!`tgX_l8EF4G_3id&V<&*-G!0ZN-
z?`A7g*xTFTBbd8r=PULNwAH3xw>I*+Y3SH4)~xlYDn3GU*i%RQaYw0;G9I7N4#{h~
zq46h9D|LI*xaeUAz2VfE;Jpra0J(Oy{j)HLdKcoUTS*8p>F@XbPsp+P+uUq=gGe!$
zz7$QrP0<~if!kKhVy$$+^Hv#An4`XzU|NpW8X-h#ZIrqT(lytFH1n&!{_9-ce1nak
z9T(o?hW=eId!d+VD$upWJI@rF=E8VSbcPPt*B-6erBf<goFiBOQEQ#LRZY3Xf}h0b
zDH}kw`dTlDh9F_hg}Yne#rD}p*S*MwRV`lvvAy)myuQ*f5QlugUFkxWOQDTs6PN?E
z>sjYh_X44$KoI5Jd)0r6dHJ39)*Js~jTmj<rhO`EJUm{t<mQx2L?P7chI5|nwlPZ6
z)-JcVdj+}(rK*ZwFQ8pa8soZA+Q3Z_Qp8IqOi@t_6J7Mu57?!5@WUZ!2Dd&G*1|Ui
zr{H;WnRtCSGbf#FpO=km=~QXlww2rUhA)jf=kR0j_`I>*vW11Hy8e<p=irHmDp7~R
zi>(wBK6QLAoF)yoc)V?Vd`kAP=}BSoo!W)l?Z#ye8jNxTFzI#P){Xa%H*y<9|H${g
zanJp$JAEvTV}wE7GQFv;cDlS#)S<4N+aT3J92q}4_Z==pc)Vsle^}VA8@$o@({6Rv
zdNR{fax>U-f57%dXje9S93P!d{Pwrr@bMJ6mB#%X$ZO^LbfwM8DaSMso1IKXqt&7a
zhq$9P&3OkVUK>t>leUiOGhU)SPgCSP>&l*+27^TJEM3hGd!@5h9UM;%&LI-K(qdT1
zOl_5%%W+a0hA4O+@HnL!T03fe_eIN@pljF(x(-#%?~DJ}`PEJQwJ%hAr$8U-UtiW6
z%W_qllyjU?#JhnXO0&74*n-=W{A|;0d4rtUEqb8nn)qR&%W)+9G4y#>LEno7Y!5{8
z=F{GX(w%?Z{o8v^!SOToeza=t_wdefdrOp4ZHrPx#q=W$A)w`Jg44VcFGLZ%8>%vw
z7vFkt?hv6_YRR#|9dY7gFlr;;7Nu*QA`%?$+YpwaYu9LYvSjG+5X8Lx0lbc(+O5zT
zHVo`&;b*`6g(=(ho>yc*H~ZFyit<|5k}-4$nUjh<J3_Qzv}Vb#V_*)8aac$NJz<Yw
z-6aAN9BDr&^}&{vmQW@}4}Fil!w8&Hr*+KrCaSM;j?;YNwk(w1S#srCwEXGy-Gxs+
zdc|iCkEHozv}jPIw08X@gFjB{0z~ha!=$<}hlL2$<C!184lSx}5?l!8cG0BBQW|NX
z?hbY;SCKI9#Hj8)n_-w$-R*GB%WJ@1pmby)m=mjiNs93+4Ff?STItW;-qeb)TChqT
z+7&Hy{A7@K8T3+^iNq8LULUsWZFy164=j~Tj+0QU197@T*3uYDX5@x6y)k?1Tyr5f
zuU3|uCio>AN`m*Sqp{X@KxaqMhc~r<DJ1Xc-s#COCgL>Vt?#_&itjqCl-{&@wky9^
za$&nJ)V1IRzVp$AU;oxU-}vYi!3nu{Zr6o}wvmV0`BJN#Q)Kdv?Y8ld3%wfD#vr&s
z5L3^07<5`!iU0>LIAV(2-JO}I%D2DqieLWXfjD3ES|Yr8BmCskcLdP~$A<|C)+Hbo
z5Osax(=W8fIC$L*fARS{rlcu@nzd~apS|jJN{J97+mcD6<EkcG4X=eyZ7WFA?(Sc_
zTY3BLiBDeN!F<ArF=<U7lGa}~JY7_EL)a;%y?Xlo?#%Z@`1o#S%et|q=*d~DQ=Rv;
z4#jH>Y;QYOYXK!G9rdCP?a<gmlSYS}RUZ+8JBnD86YZ?a##)NnJS}WwjG?jEr4oAM
za+!3I6upL}RC=qn;Wv_Ze0-UB%AIv9y7{9g+RXxYblCLfIO(<^?<c=TGuHuMbIri6
zR`=c$Lo#T)K+#0OvTn580HYRtQX!U|Vig-1x-hna^MNT&Y$a23*2UTh-h&%X)$<0^
z?!2ln4%d>$6MeFr0)M0<4%-0tS2tm_+XuJ)f^LAV$D+C#eU^8Kc;Ki+I<1QhA}?A4
z)iBuC=-qzqf{bQ~hTdq97fh(z_u8EUwHC``F)<=W#net4RjEU5^<vQNUWJ|0&<#p8
z6jCc)0uuCExAqrxLmCIH2a4PEQBUx_v(t#dD>}2b#;r7R!!|)h=XR9M#{e#{m1<o0
zOJd$}ea0XLNAg}zymkSS*JpXZ7g`YJ2}<jTSeoBZgy5R{QIwZbq!#@mVUDz}X{1|&
zrBqTbynA?L04eTmmnuL-9L+Kx7^;zz;TWhLG{r>pW|Kyr>ykHpmf>YsR7eSIpi`Zf
zYD#KrwAvs9hSQ7$oFZev?Kuy_)<qqJjL6on@8od6AZl^!cK-l`KzqLns&D%;udg-?
zIIp~~)9{@k9@ebTOD@z_@IE0fF-%P89ki`bY9%<m(9(w{yX6L6Rk8cb>kNt;>?dM<
zScbtn&$SU!)Tr28C`0+UVcqY+dyD=RO0F74F-&X|gTl3#SY@aPvYH)d%b6A&vlWCm
zn*@h#R%Ofja109-*)o(&a`4IB%r{@3cy*a5S)04^wo>bcm&PT)n<;Ub6G4P`eekI)
zDp(YqjDr(`8${!!$IuRh<UAp1;^XWCtxY(AQsMcj;7ueb0(a-c_rLQAV+>xuy2GV2
z_@pN@KuoPGLo?vJ4Fs=Wzk;q7A&|eY_{q@vVtV6wS$GwouNz5(5w!0;=S~+r5j`xK
zU%p%T#+zsS$APE`D%7eWDYZ7MJbej!+5g)vqS2DMp;gXanC58t<}V3sg3xPGB*Oxf
z_61f%RE6)1nzDznm&z!OZOi1W$y_0LVoJJ6Z<SWECZ)B?HagXNybA}UFJ0?aQ&3oW
ztIeWem7h(1kW%22yNP9MeDRcdJF=oT-4@oaKeG-4HV|0=zYfU)&XKznwS3KKf<t4Y
zccH?*wZTyL&EBu!+&H@7b#v&{ni;(k-JrL^+Me;J*Z4Fs%!$x*WegSLVJ@d(r}I5(
zSmV)c1M@E@dXc|);@=B+dezUPH-BgZc2|i15G~k$DyKte?;UZSAe}&?RTZF#s1363
zG^1&F+-Qul(&~y26{2T1NE@J`nhNQrzBe?W^}{9^gJx6s5H$C`51gGK6^MQbF4`$H
zwHCghPR{Mo&pM3gby2#c+wbRPC1<S%cc>WI#e@`1)!Vwp2|MQxO~Wo|JI@gbj*d6f
z!A{Q~pgy41?Q>G<tq(nTDq0C!9o(`$hrJEfss{rf6Vg;OwDm!5gG-1CuZttZr0775
zobLnc`%Y<v;e@i?_?>_DuUPWnZ~WHZ<UYkC=xJ}B>~>gH%erdZ>JVB8+I!FkIa{dQ
zx?0yrAX(<cZe`sw3!)Efy9nB7fYv+SIwj}>I){aSbVUL>R!}rdkUKQQZ>OqtQhGs;
z_zSLr|M=^x4FdzB2vgK*&Ap3$%NuPg&@0VCE#x$-Mqm%)DBVOfs4%K(5D+4o=w!UA
zp7u0#uN8w_j9VT8tv2Oa;Fy4FDM(cTQuM;)9MY{Tyl9=L3mT-c_ZgJZp=qXl?!z|l
zOv;Vc;52JW9H2K|E=Z~RpwNgSbZ5zag`=b-ct@opHFFx3DGKK_@p_89xtlr7le#RH
zM&!|yq`IJ_=t;&=Gjy=oC->+a$*b>rFJ$@M-}@2sbV9rdaqMhxnH)aNxHv<c^)zgG
z*Ukmbmpi`M9qGE^Wbo$h#AQ~l+S#dXFz~6<N<y|p7p#`~{M`euzWthY-Ec7;sLKF+
z44kLL7q`kUp9;V7ej!bX*%G^kO~INLrq-B4gwr`V%4$FV!L^#rQd8vgq_>Ov)MB+S
zDEp*sjb~D+fb*Tu2j1$6_Z<0sKnh-mzg7pi7Nk|A76Mpm+y`(z+Jt$E!Gg3pxiqWj
zRgn+^)8z4PQbqHw!=N(-#}orojO<p>y%W9lflo?Fw64SNW=?{=PW9{hYBi%7@@986
z3Xz>q6s&$$2jjf0Dt>L5QL^e*oVs>2Yrf*e(Ly5hX6c2h+i|1X=BJyr6VqO_&T#vA
z3HG+XA-fyL;QscKCWm~$d)sHV|J|dM2o4Tj*WO_ps>KTjVKhzOYh%|bpo7aONH6-H
zr6Fi3nkrBfQRqfv$uObV7@8@<u=;Nm0oQ5LgIOQcQ3-b9-Un>>Ky@xgo!a3`j7}4a
znBN2CbVD(8K^#vH58R%wcm^ZrCR<u&ZH;U0Y+ZlOVer2=rRt{WnWLvW$MdQKpwg{F
z5e%n+N!r|~)QUJzOb_y=_Qi*_v1Y6H4GDj`Amx$X2R<cQ&h*hV&9qiRQzSqgCpu(>
z7`WWs@w66P9i)%G#rb^U=Rf^9zy2G4&H8nstyQ~wK?i^^)M?P|@KtJ6Zhk=Cv^u@^
zjs&mQK{$p#2WpcK6Ky)NZ5ySk#jx*ql0pBf+0tx;7y`jJN;a2D?N(&ei6K}*X|+w?
z0_S#K2rHZV60ff`49vk3!Ut8JVhea{iYuwvm%e7!=PTl7oL9JKbfH&`Q#^hcn_Mb3
zfrG)_C<J8zA-!tq*l_6SuA9=-sld<;fuYG;!8@XN)Lymk*)~b<_4MbQ&`LR=>u%OS
z0LhM=H9af5zGFI{Xrt0|VdTuHndH^W8ohF8VGi8#;Qg)g?wW}yoA|BrD!`l~ug(*%
zPA5*66Jbhhxf0y7ikR%CiotW4C#EzJ;zakNc?L2NRQ0PHp9hqpaT16js)t(rZ1|nO
z_7Mp)!ztOVCuqH)T}Zrn<GDLewy0=ek)JL)KmcN&jf;bjW=2^6$7f%>=fD0te}f;s
zeIVDu<aLk>De&=S;%|KOj=O8-{uF2fmQ}S5nilEI^Q)VH_Xfv~acD1KSdRa$-c|ic
zx8yZUEvs;|E;Re;=E&ACSgmD^@>OfA)<qM7ZVtU2r$s~8RN0>^Njm4?*;=QI3Ny7)
zPi>9Ssjz6PDz0jELky0byd%i~nlAVT{Wu5F!#sJ;r^GY|YOM+n+QXlmR&(QYxU5#^
z>I7o=dJhBn`b_ic8_#JXhFL|$DH4O~a5QZY+awyLHb&iWt>UC%SgXeQZVO@BROp4R
z#zXn`qutMY3#&0?!s#=w@!UED!Z1|8(16_r>O6>^y0ku4ufN(1clcE)%RQi{)2Uh2
z;eNOe1cWv?w4E*)#|fNPU{9oTVg)_wu)y3(9|p>FTRX@842739(A)bUM!cvSW2c?;
zInk42Gi{3wFWv-)7O!hV_O4+mr5{T5(L2BV#m`|3QdHq%%Z=RBLb!HR$O%W3@Gb;D
zrN9&w>DqH;no+oZ*r9Gz>y0q9*baZWR<_o;mClk2wH7|Uf6bWBI^<26bz3y_?sTFq
zE062OIR#>D#|G<z9s~tA&nI5p7nZk=jOzohU)?cXW~5h^=VxMk4PrWq)-<%udv#Cw
zT@li?fw2wSXB=rhlbrrp=hP;tNrRn1!G}l+i7}tpTBYTcQuDE9>GLiovej!Bd?)xy
z6J0BmCfb{5Ex4&_-LtfDh%!Y)55VCsVHogY{@&gZH}J+Q1s@0&0V!Q=EByw!4?;3^
zKBkSX3e$Sj>Y^L6!Zpf2=?yEjs~Wae9Rzwux`4ME%$d@%!;qoLWC)Y<mLjs-VJaR?
zyz|(NH^$LC=A8<<bZ5{9tvS!luc~7}VZ15ms+`>56dVEQ(c!YAI$<r1+tR4_U12*P
zxevnS6q%>M91_t*9GaKj`=C}aflN|VAex$76fLNI&qUYv7scV7A{xmj&S~P=>*Oxd
zY3SssKiGtX$k9FEf(Ec0QJBh;wwxKQ;?n#QMkJy<skkGu74MN=`TUD7xPSeI-}sGR
z<NE0r%F#Ntdipo_MDO|jTA6*|GzW#x)?yp3t@OibZiDQ8VuLStE9p+__y}XrM#rjD
zj#K?E4Ae@~&geS~3ZT5(3tFPD!2(@X#p|*u&D1r2fMM3Op~9W$14t$JK_30kp0qJw
zBu3PQTKXO(>X=d>0=*TpO6tQMePB*W?XEuR$q=PNG?zk!Xx<xg<zU^kTx*VwZ{E+~
zI=}qoS9x0hYQ1X7w5&N6y=;tBh^Kq}NT%iKIN#N#3xGnt!<_R9ed_7miBNiFefkWE
z6TJ!0Rd6T{hw`tupndmt|9A$uy$Jido&NVmz&(msoYVElyTkEtsO)zdV7Kczr|Vyc
zl+hVeMM}f9W@wm-WlI+|&KMW019(O#L}%+Nf)5ieof+Xomtf~&4Zd+h*Q(lHuE{=V
zR!b!eUQgf7%kD+7U?i`)mZsI~dk9fVCqP5jT2mBEM5B+}=<0Tm$kr-(Exi5gGg}u`
zjNG~+TGdeMZkmYQOW;IEPKA-BwnNxka@LZiu@7VVY`Na4MLSBgNmjUSjiq)i3-boM
z*Ol?S=(F1)UG7!&FAk<oS#5yWHl{T)PdW@ZXUcr<`1s8m)?CRsbG|$A-QW7VJUu=0
z+Qh9g%u;SH3GY4A6m5-`1IgS)&yo_LXvkg)rLMGw)Xjv2?I?nqoK``v>s3pNL^T|U
ze4w!rMO5YpL#vGuYraL-cGbIhx~MpQPbl2&&|g+<^wr!1-jRY+Z@SmbcAsb`ujN5H
zeQNPgU9wtRn&#jK5t2hY!);m?4|NCyV2E1HP?1j%M-vaCK6-4b)XpZsjopW-E`~V-
z&e1Uis6_*Yv{p!Cfqid!GH6x9cKXoi-%ba+b!Rl44qMIYJrBY;Xk*r7!+`71j>t;L
zMFobv3nCux6cWBni5L~$SL;)wEE`YH8@Em2E^psk;l>6-1tkNE;~bp;#NL>O#}in#
z!Mpc2``N`#8anR}7w~a1*lFjIc6fE{JErBi&&E$jkQW6qq65Ib;D|V+EtFdM!H@o&
z|MoxsO@5jS+q&AOF8Grp_(-wqyGLQ}?Oa1tET;l7fAF+>J?}ovdef2Y;+rs3XV9%Y
z#tw11m^kQNdoRX#8BT}Wv;R$)v=_|WGU6P!+sc+T@XD^BA`3+`#R=Xx)lR7jP3~f+
zHY#H{-0+%_07NH5vlF#i>7!BWVC$V}3cPA+I~9Q~TbNGoIA=b0gHtt&T?g062R^w(
zCV$~qUwQu7AAFUk^{>_|IB(Rj!Rcm4+fUpo0JW1lfYB;moSn=S5emVY*${fuT#&xQ
z-9a<&B}^9GZal5mPuPo62z}TIzgL_DylAe3#uGP#fSp0yFL3#O&zgDw;8^gT6XI~H
zBpX48Q9Gdx*?%r`GIhJhT1G}yhmg4kocBna>26ow4-E?I*6-3ZFbH<UU>~UW_k+1o
zR9U_^?RmF18fqRvt9lBL(Kcfn*7vWKtBN^Af;w@;b9;Wz^TRv5!zesMqGi`(SSG(A
z5YDST+1a2Iyko1~*2-#YfyEa$B}Q5Yf{MbIKDezqkXO@5;5^}cN8S-lv)Sp;mJsJz
zTR}|7zShQd-IR{?4e_27qw26?;8(u=Z9aSZz}u&Vk8f9A&+1O=r6FF9P$@Aw*L|?j
z+lRn$;rT$MzN6=Mr-%=@-WUijIebXWX<}>%oIA?nG#`EJU_SzysOXfw*>YxQ42Nj`
z&ge?5)nQ15a;}Habo^zw3C>9ef#?M<{pD2OJIS*8>)69Xbb9vPDa^+p#X$E?v*ZzC
z48+OPfHL%im7<<(b7^e1!MX?~8>HnyaCloRu?Z*`oHGZ<We!{>$7-3>UeCJt*64Yo
zbiL6>(_ulR9q^n^W&*WR3)Ea#oN#|m+}{PxQM2Y}r&DXDv)Ytghxx_#6ENUB#Hay8
z&RHC8t!%fIr|ZUT>pD1%7r~osMJtTGc2-9bL;{IHa>B_6<}mTpyuy9*!Uc-xAmGwj
z`T43b$xg9BdvN7YMB_a!Xad|`HNDre>O-o+otIOH!if?B_aA-CtB>FD{<F``hNTaE
zjQV#=MFV-Th1!V@Uc!r2NOrZc!uG;$*9xYRb^36$aXeIjeO4w)Y&sO|Vp_NV)9o9L
ziS~Ls?4orH8wM(LG25DMP<g<uqaI_RA~XaEs&|M7cQ48%cJm!K9cp@gu@lxdC}?iM
zD?F1SKWiz7@yQB_3eA!jz1x-W`hee`68CdptLdves(<BPuV3962U}3KCJ~cAXu%^@
z?|`p)fmwkdM}W2oxRg&9OL7zE7!@08_b?!50#fza%g}bZW@w@&JvLgi#eGMq_JP<=
zeekmB?}Ed<ylB6Fhvgt>@EWG}w~gP1l06V<j0$~NeYv(Dw$TWFAl@qyhtdMAHN-Wm
zJeFhA*O%#VYQyx<wl@tsNQ?u~7-)FeL5aGTpT~J^$=rvX;T(00qf9P%$CM&J|MNdp
z4TRgZ47%<H??~Q}t6i%PzP_{1uXVLPu3KeYiwX3GdH7YxJkZMeb<5h}Q7WYkqXyv=
zlMawFa5365*6F3v>qhsL5GPtGMyt9b2HRQKkqjtZ;q)me)$z@5e$2b4D?j>ozt4a7
z-~2Br9qMQ#AJx&)2YaLD|D475$e-8;-4A|maEy(pPNsonGDII~&Qlo_5op<N>xPKw
zR=l?F`k*w_zSir3NWZcZm4&I(q+3vwiVb&k=gYjl(lBs}NjdZAh?emzz2in_4h}~o
zH#yRwVsQBA2+1?0p#AV&Gz}18PJz=YlBT4og<hC~#z(HD@bsK`|FH7(?94*Lkr(Yv
z4LrJF5!6i4F-N^IkIGKbT<tmOKlMal+&LOb62u+vxeiNxvy-B8Al>1HBjK3EagxE5
zJhuvq_XKZ?lV}X37B%f2W_33V6cMRgqpX>yrLq<~rHDbo-V-{c#!&(V$~`9Il+PjJ
zalC)JQnw8v9G#;gsul1-qg=gLA(r<#_*kduhoZmTfU<{};KUTtAO4B`u2=oRpZ?iT
zasS;vU_NPwg!-{{vz4Zri&j+lr|o05@L*|%bYxh<<Y8m%wuoY?Y+X<=0pY>z3P{+e
zb6sTGenc7Q;@<~*@dqNbR%{{EiMh6}&V+6!ANzi5?`%UYkj?};TOFzz#;J0-HAZj@
z5wfT=CNL;9lVlt%n}{t`Lr7EPo9DpglsJoK%MYU*TGve}hDaM4+v*rb<%pH_3>sgq
zll-shYd)P+jMWr!cRtv_Yh0%xL0(Kqh)1*CJL*-}!$s}I)aB^3O*5+IMI1NG0i%1m
zqGm%>Ga%jE8cv@{??B`O%j+;#%TOEh7(<itJ~%ax-j5)&7y{mTbs9;hl#RSDbaOyB
z5qwB^)AguSKs7pyy(zJ3YM}#q(_d~s-4%v&NRP(=EMjgbG*Y~^Ze1`h)<njOF01#3
zbLy-TAf==OmIbWo=ip!9ZKq!T9c}Er@&4_5e)8iV8(L=$qi${n(RF(%T?d}Nhi_SS
zfQ8=dISW&YNYlTsiG!joujod&l}0w_(Y_8vC!E9J{$w;m7jgqPgxd~Jj84l7D&Bkv
zahs+|QBXrLYai-fLCcMN^nCp4j_2FLzx)^fobUYF_xb31-!_}O{y!PxKSTv=k9k35
z<Ubr9^%By}AF00oad`ioc$^DJNJgRVVq=GhUQh3B|4D&byxFi-Am1&y(|vF#<b(I)
z-Ai5ef9my>hJndE{JTge$YAntAB1}!XasUoJI53q_m`Q|9Pz;soe)r=*t&&dSeT~B
ze44?cB&QG=qQOPmG;^I2*Gnevxz`3kg=kzJ7WObS#O3ulF-J#pgW@`+MN*2KCKbvo
zx$|^eX{D1yh*3zprN#-h4Yt~8C}l7w$4B=wpIjzBIYsWKz{%;gc=9XK3QLB3z?y>;
zBZNVWFEFFn=Jx-m?%#iG+p_yU>@((Lt-W_uopbKF_x6Kiv&p6?Qlv<U6vdat$f6`j
zi2y4yqC_$5C{dh1NPzqY3E&?LB)`P@Apry=j$uibVOxq6ONOn7B}*d3_nX~KHmQDh
z-+tW3Ij5@j-fPXr$PZ)8wX3OEea{a9x;N3h_nuR=Yu8$9&N;?+d>;)X)x9{ISyPOe
zU5?!B6LS*gnApc`XoA>mOS&>S+ty4a0{8d#Zs+R+J2k6a2;AP@QPy?qBzE)6ZZ}g}
zBWSe+>`Ddj(E@7oM|U1X8xSsG8jOj7@R`ql!Q2|7hc`LW(wZq2S~cSgJ9}-6lA>jj
z_vV#$@QPX4phYu;!fj+NH=*}-nH}pGBMc*t3nKDRrU)>~nbwW8x|y>G6go?9EUmNB
zSW9EShmaF>O`OWYa$eoewG%?3qMT}FRi$fWQHyd761Yl{L-hHlu?C5lsR+&Z1dpz-
zxVfI#QJLFHsFvzx(Xo#5AcV-~k8dPyLtIMx7dw)#ua|D!zRkJq5H)_EcbG&k2E^pS
z8qrcVXTfe}8}>ZtNVK!Dk^tTJXd_ZJBvdaDtq>qsc8J6XF`<O5si?&)x<}`%YBPG`
z0}X$_xCpjVsqV<yq|$@X`$}CG%DRG@>zr4$gqRFt8Z&Y+YjM?PhpPZ|S}pXl&{L$5
zO)}g2_KedNHzC0yJM2&>>zTeT^xC%1X<i<k){IQ8YJ1Q$O-#v%a3Pot@$gc(5Dtw%
z*juHk^3Q+!w^^12i(XcX9dF${4Z0DXcT+&uKo?gCiK!d}BQ2KNoFHjq0fG=!h@#|Z
zOoJxS2&73^TA)<tEkw8H^#iw-PIaw*3V|->LC{+z<cz4<-_N_r@-O_JnmPd1W>&^?
zjO3n(IdgNE>8gC@6CY=HIPh)X`JIOBsFEc*B54oVK!wJt{O>~y{4WRd#nC6-&?8ik
zkO{$yUM&QM??c~6-y)t36a!i<|7BfPax_9SE<GGScO)E})xx{mJ6@_7K)OMjbh0Qh
z3$I;I%se6mi;7)p0SR1R?YP-x4*Q*J=9@tzC~N76IjIn0B+nBuIZ>u{BqTb}a)zrN
zI25$O+uk)z*;vTtM{gHxa&Lq8u6sUIAJZ_{D<xa>r55V4qH8x|OAKUR0NqCqMHLle
zVIHQ;E(dlwGN;6x5;=-_F6b?#o_KLyd9k>4uP7-A0aIR>)0%xwI;>^kw!nQ+2v#IL
z+GVb$#ML}8=ZQJZ7LMa^<!0y0We)W#zx^48#V7vH-j(~iTfXq6w=Q|)BFt0fwO5~#
zL!_QhXe-8uCtT))h{+_k%F>|5t+wqIx}|8*J6bBem?2TG-N{4t*<!Sik0Dc1-?RoS
zZU?y?exs6Amr&twt@*+$^Z~~{fc|?o$mT%>3~aOA@#yzv8zJ2e>4;L*bl<hx=5%hI
z^LgR=dT+dq5O}9F+NrYELQ-X)8YYWt1c{c#UTrf-5%yvc)5cuatq>|jIhMi_10g{S
z3f27Sr#ztGu$iT#7$~i|sR$&DOlpPN*LN|nuJ?e4{=`JrMxGoQ=$+v%G7ieo<%Aea
zsN@iA8=?Xgx;?=zIdv8`rVAHoHj{7FI_r=j*o-;28pP~~6N7W~u(%xXx~0-t7xyrt
zxRpvN>jN_xwJyGu?XwaBs@7?bK&9R)-8<+hW+dA0uuCw7MhFU8uwW+3T97muNio=1
zvvr|$>vD}~VKaiVu9S75x4y}=M*;nC4H@Y%enqD%VRJTk04v*Qu&fJTeB*Nn-AvKk
ztE6FQUadxeog%a-)DSO)JtiS<B3mxOd9i0aMjN{xro>?i)Y@5!68APbe&_CtcG&{*
zT5T-WkvNIch{z_wQ!|hY$^0AC5@5BD91)+Jhqxh-&57J=t%UQ!ex3*^u*->0|JKLI
zG4L(l_H7Rk8QJ^?Ml-@_MtoF&qpmoDG``PXmyxTX27yShn0pOqD`-_3t1AlZnX_>)
zpccecOJj8cd`^Kh?~G&RV-XccUAOV)yIZ3#RScNKDOw>wXE!D0S0C~E<2^ZMYH5^O
z*yV{UM+H(ghCo*fG9uVaOlzT)GhHo|A%G>v)yA?Gj>|&Tz`UC{?2<)6b|+&5@z0V7
z18Un2i@>PE8bbHQZ{|iot2RGx4iBQ6sYef1WyR!ik|~-UsMsii{%~ff?3Th|?d*38
zYx4t>W?E@?eNq)9Bv3f53txV5<nzz(xm_DECH4~8rA#tlXzTogSeK+&a7Xa2T=2tG
zgqX6irN!WI?UlFReuvZf>{$>dr8(c<b2TO6&9!ZCYe7;*k^_1Aa3<Qfhn?|IYc;)h
zGgUzEjk{;hc>esE8O7v;s*qZsmO_k5CRqbBHdx0Wkc*Vdo(t<>f&^l8dz7IR!6Znd
zXiyz9NZBg$4b2cIQDP~NrI;=82%QPR&4(IHF*2t}4u#MfYdM?YM~sLH=e6_J?TNJ%
zW;)l7Y~9R}-JChNB`Up<LtrN+H%+1>dqA!GpmT1Or7CSXqtPODO@}VqY*_r73^Nf6
zC2>E6m>em2ck-aG)z|s%*20Al;xyZ{>m#5sySJm<u1z01b~kLjw~qCUouTR14$U*(
z0Yi$%zU!){Am%)2W<-Vts3?*Ijh-s%hvGC%n+<V{ahtIZ!lE_r>T~o+Ytft&eX~}z
z5S-yRlD#V)<0`Z84Mxb*WXhG8jYpHZIoZX4M&mlP5Q+1Suse`)CO8r@yz#+pu0`nK
zvShn<<4U<dhOMU|D-PQ{e?!vL#JhqZ8ok)Iy=yI%ciw)+`FQVLs5y3q5Lj2JYuzly
zb4t{%uG2S_Pi;nijyd^wQrJ(Cy(5nM9NEuyokWcppK|8@y!hzZUeSx<wG=n<6ExVh
zweUPEEK9&lq7<~-%)B;Njl|&NNG~8ISp2-t)mv+RFA{mm{MLs*%q}Os@q^z&8{;NN
zw;u*VZA}SmMnJ<|ahvU{{k$+tvg~fq0!%TR%k3GIO4sB?-WWUBXpwFFW4<P$N;~Wv
zWlu~~Hh&)fJP(o5ck_BVH9-*eQ{rkj5v4OvfoXqb`u-5MIN#kovzrprJQ=a83tb{T
zWmGZ}g(T)zt#r!yp7Zf!8)fg5w3=ma>EvV^a1`4dx(*YCz8wrUTDEFfg7|zeFib%Q
zyVkBy+P=@8DcLH1Ra~xiiGd8hVRxSsc6W?goe0yqX>Jii6Q&rL=gHvm-soa}_-JQs
zjk{8LcD8Wb>pZ!-!n*0+JZPtTo{#Ho4Jopq5_Q%wF{62B#g#j(A*QKc*Rvg}V%O#p
zi)%!pm?vV&tn11&?|efP8)0JIuZ9ukVI7v%#ry(NqIY=b*$d9Mch>Esj8@YqtD3n~
zj96#ld;qI-tB)G60;a1m|7{778JcG$CfnJ^!`zx~UCeN1+hjJI<jkJ6$jCu(WF?3z
zSny)Qx|lgAM`S}F_WOw?8S=2Wp1HL~lfW**j!ejzsWqk)$VrIOm?RK2y7t%Lyk5=N
z<<S&50J9e2+DHLH3TBK_-1}ZGl!xN3kYquc`LD+)=)1HO{aSssQqZ6_tFML{m9m)M
z16Wd{RzrTGSa+|MCQ2*D6`<Lea_g8JbqY$1N}5bSXd7}c#gU^0V66O3qb?ok5b}<q
z5R)2|AM6=x&5DuiBsKLzY>?i92tBIZbBvrSVq~}&12GMKc=Q|uHd7>En_XM{_V|n)
z++k4Na7TTOr<}b>XkY1bpcT=IXaVh;m%a8*lz6eK#rOy482OFt5!v=Xnun2z@wsB0
zP^U5c^N;-&r4%L^$^i3$XdbWKySl{K8ySS^KS#E`*bs%*EchoUlZRU(tob{bw?PD7
zLn&G_htKPy^Kry^LcJMZB#NORRu~z=JJ26{ncHIRRv4Jgb8uquC`6&)J_l<-jwe!z
z{Nm649H08sr}_Qg{{wJyV+XyoQKJGb4<7%e16;L{2iVs|!Mz8Zn`U<q^_f4x5*C#i
zYN5+|nWN%dDsOl?%@QYL*qR*+i;AA=Xie5Y6<YmQpe9hQ#Go8@nRyC6+sm70a>(gY
zU4?E3i9@C>h*k0~7s@#iqURT^XV<k_$DQ)TD_2)6OW}Bbht|fl%Y<y<7PgqeMid$2
zhTytjacFuY{kSaUA*o}O(W7e|Gv{IWfFJ%b`ps~{jyW${fJ_0_RxPk%vV|X_&7Y(-
zbhSBp2s0rh=6T}r&7R-&{>QxjWWT{{`)T6o^^U83_D<LqUGLqZp@Udpi>f)OsUmqs
zVt#-DS(kiububr1pXGJn>So6@&t_JF1?JS+xx2mR@e@N7#vD}!dk>&lrh-=c-(VZx
zACLU)fBFmj>)-Xe2uV>fU9);3S8&ybNOV7iQGF-(7ht;&Vazf89*Woo`w*(_rEO9j
zBaOQ0hK$uWet#p~I6#0IcMO!_!gP)p8nsW6`_($qh>>6)Ume(|=Y;MZO)x_0<9*`s
zAu&sV(h*I>WO{7XZiJ@*VKSqS9F$iMv!xTFMsAJP>Gawt)rv*bJk3K8liP*i(T+uc
z3+|@)GAs3K?<H7gWTY(J5UL*g1q~YVTHmB;h0qG28`ouwU>M{ys|me(H&h^~*?Pwu
zx3solcHWxhKQshdMKF(vq>!_v_645{H&6T92#E~7ig~6~FBXQ9K*TqEQG5P3M8b<|
z1~0&_MrX+Jyg-%U2ZgO0!Gn~<2)OFp7lz!@N1F747PHPaP%7AG3~oRq=5+xMQu}A>
zF8#INX(VMD>ClI6#1y%G_KYvR@p%HC7m(xDaoc;2yX%J_;#?TxS*W9Pt?8yp@UKEb
zaJzVK>|AQ7nD4@V&Nl90bk{a`kbs2;si_Y}1obG~VvLTwnfx4aWZFGjOyWFd|848y
z&F0xhyFghhB4@tz#uxbcpZyuW|3`j=bak_h@stbE`*(5)hOBSwa}qpF!1ul+#)r^>
zfNhYzb6To(iDrUpsx!gI(>b%QBcSSX57USQr!Zuvo7cg+c)e6H@XGa$>wUKIz!XV2
zFy+Z+XdbsKAtK!fWt~8X6QscX>CAF>q$A`Y^xXu9Jt0k$+BvU<^YLhvQV3k-Oqw#g
zwliZKH8?Qmd<k}N*!iJR&w%sHve)5re;~l%Mi~;C2N!Bciw4;uILmpkK2x$>gCLRO
z74gmWfjlK%Yn|8b&s2rOoS9Oj$3Un1=Q4M?<fy@WpImc3FQ$`6(P-Cko+hR|*_<$r
zbhS~8*lyQ)vpn~Z7zPz%A%ZZ^6R$meh5Nf()@q07Yp*^rXnIWX6yACJ8SAoebv1AE
zOo_>=R5eDk`e>k6OZ>7$G|@xm*FW(EzUk2;zWM#{aah&=-h-^h=4sNXv=v$*UKXPJ
zFK(`5;`6-b<gh`|i7pKs1smcLJa8?y&)5?-h4VtHwsDW0V$j{xN&<%|@#JbkLZ#=#
zRT6d)L@V!`1g<BblTaS*XCCbnH#rb`g{BVmSE!&t(ec?aI5LV^NMG-=y-$!~l1Fh)
zM)0vnAA0mYv>gwS4RE5}yNCO}7GEI<A)B*alej$_BlK*VCPMVh)P3i1ev)WM^it@J
zy!+PGBFO{WdRB>k5GFfVMn{DhVa<tFJ!Zs}0Hg2}uP#P%>h{@Fa6c0#W2vcOV)D)1
z`62=5!iZiDgw`$!nr)P2>(HJTNZnv|#7frII+ccID=Nebl2$7ttKMsWT3sopBWtZL
zF;p+aheF7@$PvCI{_NpJxOw4w1JwtDU|t%v3%~K<4^!5aT^bEsq-&+O#<DD)Co&>-
zom=6scoMUh7EEF-!JJd0@71QL0a_>T%*;o<k+U&yptcbl<t0i8Xl<@gu*N3%7)iMx
zZ0*4qaYGVqzTgU{2YZZvUGGX0bEeGmY@=dFc<%4+_|;$dMSk?hehlNk$z>1n&kSDt
zbT|F+q9X!9+|bIt$gY)p-SpPn$H3}4kqBXo3VaA7RtU%>)Pqv0Ii}XyD5Y&pu}A<3
z=8!0$5>feApeFdh`<`-hwP(uarWT^jG|jd%QrFcr+rcvFG4K8^#)1sX87;*t_Jwmf
zp>blmx*<<S6kN||CwCQc$~HUG#<E(WF-@7EmS^3XrTN(a$QDaT;`1t#Xa%3Yjm4gP
zrXwP*=?&XW3?aB#jz4tZmaul%<>*pDH(uIJeh@eNyyt4n{TwZ{zWMmX9}}jW*-aDc
z)oh!+GsMOD6ha_|#Knic`vG!+yE_J@8NBXG#5?2Q5apvsSL6ui^VxhGx<w#|5U6G4
z?*7QS7M{NH$Y4!R?6db8GCqHEmq()b1{;yQhkV7~`{-x+mM^};ANsCuvv_Ntx!VCe
z;8pK*yKy8^j@x4FU7c(_w1l>3!<ZR!S<Jx$rlM<(0iOqVXIn$c#`v}1qdMBr&8<+h
zQ+p)W$dqh;H_wf=TOklzVMpQpd*PL^CyEk9*-weXel|5sHQ2Yd#@d`>DdzfS=Z)#A
z=bW7>AHic-CAh)>MXJSmhhTmNr}LRwtkQ1+r4DDqM2>;ey1vX({hE92b`vCH$ET3D
zjwYo@&J)vaCR!N0k?h*!1i*r>3v`3TRY&x>Ir;VNaPv8Ej7YG;cC2l2nQ3tEzqNH5
z7G?cVDmLKv=;+`7-+8OZ*0m2J>lT$QLhnLX3r_3JLuI6GZk|?b`??i(`IO-6#m%)4
zb5&K)rgS6UIO^&nGzOv3Vsq=dP~2<4#!6v}<~G~k<_Cqi&9`FK=6;xsj&?&t0xzE5
z@`X3vU<xpg`Ms+Xv|DN@wQroAVXZ#A@Pk@GJZFRWd++3&m~z^hyVk1hTOn-x3>(=5
z<bm7X=-vf&g^e4M1t%DasVfxm(Zz#Xh9xRW7#<!rI_kYs*46f*=uVQ;Ox_=u_j|jq
zV#<~vmOb$HTVLk&_rHhZap5xd30vVl8kkLK(A6caeh=su)X2wAjQw2#J=j?qdPaMt
z*Shh=Qb?o#T?5Nb(PeqSAL!mh>9+5UCgTAl^)G$BR59?{E04H7T-glB^Onbkr>Rp!
zEU-m|kjI5L>BoM`$Qo(Ak=!vjbcFy}ZPp#;Y@sBr(oRRhv?H`e^11oCRzwCI^k8Na
zgEz_I2Z0@}-46+oG5-%6jI+6k`35u~y5ZF4wYtr%#16U?O$XftQUJ1b76XnOy7H7V
zEe3jy&M0+qSnw_)B?cb^Oj09m=wqBIBTB_!MbGe75S-}PmEIjT{faweq{7|KuE{*l
zXm1|Q)6sTu41D_ZG4q@^LMQlwb4N65k!)@|_!cS=LPk`LoH*_H#O*@+=%@Jc-}@ng
z4eV@=%lQ55pf|sV#0X#+PL(3Y1hxa+I=MDrVr)|iF>iq^RxfP0C(^B`8-%Ukb2J5?
z0lRf^@bOD)TSb4>Ax4uJ(pOk5W$tDUT<6KXD(pGm=j76+4BcXg*Y0k0(PztI{EU4x
zr%g5lP{`R3i?-UmO#!Bzp)$R9GM#g%bcEWhc$N@+whE=aOG@z9=u4sza!s-ujMy99
zf;o9-?SA$tPiRVL8D9%DasHG^UQ_a9RNn>0!6zXik_Z?&(7ot4cTFScI^~EEjrh8u
zE@IuX9W;KB50t<QRhxlZm&PV{b;|>P{~<};P`%cTwAjU@!ZN0N0nvzrPN&nn5j2kz
zlhj6^{SI|O_x?LX55ehX`(B!HC=elKQ&4r-iUC29Xf<~i**+7np>D>DFpvD&J0JV#
zZ&KF66oqA7Q7?KJ{vgfPtQGrxP!LH`(5B=lOv(ORMA%IeIfae1*%j7OILuBa1?Tk`
z@$g(a=Vb(}4b_dILpi0bYq!8KvpZf&+tLsN7ujhDPGJ@&Da!*AvZJLcLYT}GVs!8-
z)`5dX%m3=H{2G7seSeZ^4wU6|vHq1yF|f_?5$Cy?uY~GqdrAye469gifE9V#gNGpb
zKSQ875i|5|=0hQ{%bBbF&YfG8vX*VKD|mlA_Uta(IrG(CFPBXa^CkBiV|r_E#9&&<
zVf52#cPUyosJT}Hu%99tXY+3N&QMh;Ya~qur;2o@C`Lxqv&-r(dGRP)_=sUDag*M9
zjq6q6<(M3@!lC(pXc7?|o(Ya#AQm7~Yu#L@hQ7C#;tR~@RrVO!HOMAA^l2C|W^;gQ
z=4oHYDXGI~!W0hX0A^ioYeSrG-Ej~rqz384zB!WA{2j*ck3)<@>0sS_wgcM>hgR*H
zPrJ-67-QGEC)*H*v<q|_57c#WL_m;a-IfOXo-v}x^N!En75?!@KEWUT-Vbq@CQ7Le
zSGJ9eIUEXu8!)v8^U|nlY+1zf80<3Gd@zy1gOW$AKu{myjQOAVn4<Lu-UCrEV95O2
zjU;VdQw+g0&=QzcC?19rB22q%j%aFxH7hi%*s%GkbuXnErz3h7st(glAs}sajq*4w
ztShCP(jqqd`SWaJufWVSWs@X2;t-wx(p&rb=L~G?FClTpa68>}{w6671Hs-pt3g2P
zJvza}JG>Ab8Z~4lM0=*J_O`@0LN??wj0j@;U-o}j9T2h`z&LlviWKn!{X!rfOK{9#
z6>|v%D>|)v&7|brch_wcV%NZ~g(+t+!k&-Mz&uWnK<U`nbzMlif~4fvDqhAJTnKw3
z)vz_2p%1WR8;$mh3!N#I@cb5a(;qWovM$v!6+ZLHPxFP(f6m4pDi$Rke4hvd&9L{V
zwYrka$beA<jh5~iyh%&GZnh3T1Ww137h`N4t&L_)I=I$Ob-u>1)lD&4j!TN$bF!4m
zxtc+kZJ$1p3~aj^V28UG3T8EK#a`>?DPbc{A3<sNAVAp)7cC1htW=GB^M}5TPkihn
z&<nNec8^E#*xh@<8oUO5CDLurcZA5>M6;r%d!b_2bhE^E6e<NPU`#dP5!b!4EQ_6|
z&914aZtk_}g-U<0eg4Z{FIfy+*r%5I<~Q46rx+|(Tea9PA)wN?gTK?aN~&w86k~xe
z>tYY9c6w=MWTQ&b1~FOf<mM0BJepI?ECVqpKYC>r{R!wD)6V`5k>$f#;8yYboW;J@
zWPO5#ssynOylJPXv(}@Nd+bdh2tqX2Z6FL1Ey`E~8H5vmP$k&Lw-LsI^%N-vBq+TL
z(whaZc(BbN`BW9CD?jwT-^DL};tN>($^gF9=TW$EjheHWn4C#8JaE!VglL}E+A6J_
z5$R}MY3o8O#Tmne!_|Q~O}4DMMP=Fy*KCYa&V1_GiT~su{R02SkN!SB@am0+SYAj<
zVBL%nr#hh~D-baB#qOCB#Gt~BoeLXMVV&b(06U>lZG<yslLIkI$rK!8*4t4}9<%Em
z)i%cHamPE75x@`&Flt8LOVRxzVzdoSE;?*!eM6x_jO?!t%)1@hD&=@*@<wCMY@Q%8
zj}EL8QihOh^bsOip|oyx+i`TKwkb8<<%<3qeO*wzL`XK4YTdFKh8(b0v-56FoJ5`W
zIl8GHH>W(2VM%SUJzm&1KfF3#s14+6cwUGXgeu$jj)URhOy8OYFBp8yY~w}$U-vH7
zyfchMD<rfTs@ESJ5cd9lFpZ)h1fs*k7AR(jNoO$pjTngfOj%?&bKCnG{AFWM`|WI>
zZAy;{v`wv{y<67EgZnXN<_~R^LHP8iKf^~q@)28y!QEzi^Fq$X55qCmdl$S)_s-v}
zpkokfs~i6+sD&TpoLSexvQ(~i$%_HDF8x$Gr_wF!1^c~ojF@MGuvVp%LRGW;?&?_z
z;*#INmrx@cYY=<328<(kWcV13m`Wq0>WYVvl`_Uw#m6Yd&i{x1?ce85|G9sQdERq(
zyV(0Z$cq@yy5erUY5`~=SP3kyk=JHt%S9gVn5RDzA!Tw9EZ4`1M;pUiBz)NIIjf&3
zuoVKm8-h@3vorTyRs%1mCIE(ACtaz2FjwD1)epWYn%RcINLGar0;MXQwv}s9Opax_
z-*O>%B)do{SvO&@m`g>xyN)T6k}-hC`M$Z51)D`mw7hM)dH6Rs12M>YoE~Fi2>w5N
z8JWc^JY#kEu4Qw5Lq}}KvYB+;*svi)-*m#IC>o=U7@akLG5pvznqV86dUBWmF=cYj
zn<LnS(EE<p3LJ9g)CI(dDC(@#jd`o?h=&p+F}6PhM+G9HopQY8`Sa)2t9jIMjKXeb
z&V_NBECCTkdSPS2X@|6PgkaCI!2kCrf0;k`<9~o}{)YD%km21zA26T{PKPm<t<zFh
zotRzeC3C?UzO=8{kh{C92__|q!Al`W6-G4nC<C^|=^r~g!BAj}PFC%kCH1<x^WX!B
zQAGTiOAaxrNsyvhN~fHe4p(L?TnlZgZUL<xIct1?lmj_tlWB?tU8x#xqpKpJc{-&P
z$F)|HL?*tzeFnP2umtfUK4C<>=7+f*W>z>o2v|{dS)H%vICT3qqaV2xw*Pp6AMDS?
z32`Yp4@U<K?elhhs5CKPnSrh`TrEK!T>G)vTld}kUtFe4p!MpeN0%_JvCeJ(x@^u8
zbkn<|^1uv7EP2q2eDU8)H6YQ<cI-WZjqYUdp1P}diZurG?GU>lhg|0AG9YBJ=1Bry
z_~MuNwO{|R?eVGx`9CP6@uDUL<F*V(#SynFAyBk3bo<HoEmiBlQ%o%DN-5Ps;Bxd3
zB`ad8c1}y>RCIf88n|ac+f*mT6(n}ewkN>Xz2ltDNm{G%)KUnSLV*X@V2pEAIWLQ?
zZ`9l-%j&#>Mo!jzzVWHXKlqz}oezBD`*`x|t6L#vZ+B5aMV}uTd2n!>j0E1txzjyc
z2SJkMozS6%KnR%_C(B1@#mUQ7I81`YYtNHQ*v;E!ZH;PiUD%@fSEMGOR<wp~0f=^U
zt&PzhMAePuFN*umTeWW9<B=K~bgMYScq@kr^Y~CQ=1L~{QD2m`7Qeyb5(;3u38qgE
zW9Aowr&zh=t6P`?4k?d^!MkV?PhqspB&KYYY;)<E6WrA2<uO{So%n+w1Y*4W`H=C9
z(TdHLJON9#j}7vGm8!aJ8Uq##!7Yv5G(*H7rf+=DYkbrDp7NWYebc_TceIaALboo)
zVATl32VJ@&33cUsK5_rz7J_iRUzoCReb_P2W`vQa*<kQ!BM?Q6OCYjsGQ^!#!SkL1
z{1^Z9SNZdQ;tAjWp>HynR+VlkGJd!Q=SEm;aQDx<`8@!`*a^(p@xu1LM26a<Bb$1l
zZLC}~{jsjdaT6VO$HT~I;L(^2%Ar(2wlUGTKUP$+AvAawlcq|p%|;9%GEcJ^ub{}h
zBc(*C4XTZ|a?Z|h@20zN=E;C4T}(lea@s~KsO2663FPwiO(5KS{c)~9nn>eWA~>Yk
zEI-2u$Z#1GjNF1a)+<iE#-AT2W{o0$bamr-xBZGurD1TfOOzR8LEVU}A&h;+S_JVd
zgR!@@R=188tQ-&@IT&kPw$#n9u*~k>t*g!4yL1YL;Ctja+(Vj8BDg{5=0z}=>(&z)
z5@C|v?1iuuQLUB@SsI13p#TG-vT(MLGQ>>Bl01LoH-6K73Ls`nY#rRNv8N1INb@oA
zLLfzqPGGa`&BRNDQYtw);u?fVB+il6Dk+)QzU9>f)>=Jk*$NxW9>ELcz?`!qwc@@5
zj+u=y!$5@n=8az$jBqaQ!|)Zo@_0rRdNtFYwJenLnX;T2CRDhs@m*_t@?*cr>#shd
z)kjo+wzJ#gb9v&t@URdoRkh?)FGQOvBGH<?<a?u|+Pw`F!~HMDn7XW6V;8Z|sJNXg
zc8=7fq}YhkXrpm^7q6Ep2Fm$NQM1y_DQ|8|qdK%rwQgN+2+82$27@`9r<oAU*49WX
z1_H~bu^mob)-lGEQLHi?(668l{rO=naqFB-^Bb7wo#$7Z_d%_zf1i;Chlh9a&Kt&I
z9JcqcL(~q=oZ=RC;U8`@y0FUhq6E<1Y29o$a+)@4&@qY_8w3I)OI;8~+TDN@Lx>DZ
z;E`{j<EsPw==XdFAN%ARbT4G|0ci;??@btBxpz+5(Xy~EM_R31-CWsWrIr0|+Vs;g
zPYm}opRPLc(2-z9D_U3UoPu?FF=f^cf93E06Mp!6e-HoTKlX>&?`C?hjM8aH=f>u%
z{tS&zuHYHsf!>RSiagxZ9?FIsQ5l8AHqX^zr7YXMwN12dRlG@i9I|^b(lBrL-x(F_
z+yKAUy(>Gfx=UF>b>kC+l$Z|(N-1PF;tIjsG({q<7HTQfQhk)*NS2H_uxuUJh}BNZ
z@-j>HYwl%CbL(|`Ge&%j!^YW~iBcCkh+_0&zzPEi>M_L1zyT2JW}A11V2C*mosTs(
z<aL8!)qC5#4#37vRutHuT>y984j2uF@(lpk^yqEdPsXY=B*W7|N0Bhr_dvu{QLP)C
zZRL#T)2nroghYt5tv@dwFZ1p0V?{>~C6d#O?v$KNM%a6${f?Z4Q7msdd@&htp7%&f
zobQkP%uoFc&)<26DMyccR%+=;a?>tH7S#DacFn~>Yb%msu=`?7uR2pdUYaET9m0Un
z#mIjn+IT6Alk)`(O|k+nI0`(H5qW^Nj{85(p)judP;hA28kgRzk*V#{P>2)DTP>VV
zM?;}YF*bi2ehHWTrS^_YGgmi{%-qLS4}&`qz45hiRLAgeOY98I)C<j@?UaK%);hY>
z`^JYjuZ_B_w6-AY5tcJj%Vx3!7VkdJ1Wi_$nTJY3CnZXCIq6q_y;Lz^dsL^cjo{Fz
zs+z-QvasMOBpVU<qD)~%f<1&OWuii<*5yn&nVf>r0T_HfW;lR9L~%PPQDa`(h`@eM
z2sYPM?;<5cQl6OdjHGBV?Iz1fXt23(ug#=_lVcLfcCc6{9<iBWBZ(05Fm-U^Ur4?{
z94QfV4>Rbw`N1FTv-(4Akl@9{WDs#4LRI5ui_fA4`_}dqXWt{d=gA|oG?roqZ7pS6
zC?duM7>A`nA;#HgwU`}rYvgGn3Y7EFf}7GXPjcH**Z*%th+m@t!%1jRO(H6&D@e?3
z@t^#{ukzcUd4nJS;XlX^{l4$!=4!ufN>&{Fj+Z*?*Ets=k=j=89v^l{<1lx9?-&V4
zKu7pZZ<kHMH+s`UkMJW~q@ubNQRAM#rIWp=8U~}eZ$1cH*C~d~jACbe#^QMzLz)_*
z@mXpq7V6a6W`CPfvM?ux*8Wn_iwN>{UJQg3i7At|CIUy!;)V=>&v+SWm+D11rbxlQ
zXN(Y9^O2<&=*E#5&w|UxeDiMu4d~m1)u)0Fi*Y|7#Mtn*c9anE)*!UrX?3;FsOp4R
zhX~Y?JAH=0CNF~xI&SV_8d5}3L^axYR6B4%6_DsMNr(1Zr<B54SA<H869e9lF>~0P
zT(8uHWjV2~!|M4W!yC`?ARej&G$xM6g`fYapXQx+o^5WJ(e?48ifX+P^m2d?d0&dS
zWNr`~aiq%%S#_iixvWK0DXo%Yw0+py_oX!MmO>c_hlpi*3=7{@JqJa>N6X_dA1S3>
zH~hxoe%Xq_5X@p4_atBxXq%kUgQ7;<c{gQC&Ur&?=iQF?eapA;z2Eyi?60oD`7OgJ
zEyie;_FF+}?`6SJ<81KeF-3BYE<d+5A48-!Lsz6jUkkk!!deKm5?Ui@C${A4#m?5A
zFAY)-LdY{YB+jwETVUYjih(rE?60mUwRneZhk2T^!9=b4LpKZ_Ov@U(%>a{Umxnc>
zy_*rwJZ~HDU|bKYOCw+eVooLxQUj`oX$5|03>(HRNlI|ZQbtrfkfPapxemE`=a-Uw
z?{E<_Kko;fjEHslzA+^q5p;b}@giQ(;B)H=-8XUny$F{`-G{|Mh{-lEpO2av&-nyu
z1U4%MYz)$Mz(>M&{qAq$8(x3P=fCtO=e5$$XU@l?eQiqAYRP)C9qdN-LkBEwb3W=y
z&_>K-QQ7QoOx^K~&^FjXkaUTWf(`svw^vm|hnjWOpZmfW_-lXX@A6N6;g|XI|K^|K
zTi*XZQc9*4u${wEOP`yG6(ESo!Aw}&x9jyUJl7I79r;i=(9QYrQf$cf4^<<m+Q#U7
zkk(u|A`j4)%m44)py*MMxD~^t+j&qx*v}b7f}=3wc^WXW>$FX?-8baG>{tEJ(h))H
z-QDu_9bs&~?8ys)XrWEo(JnSS=Q7*Yeu2fu_u3e+d81*P?LLqKf5y9$Wt(<$o^v=K
z#=7=<YFM&0AJ(Fy3<gkR2-CWDLNfxIOwl|4PHP40=Vg1eMwD>(uJ>WEl**+TD0QTX
zn&qzmy(owowMlnpj@C&;BNUF2(z;TXLTP7^Xl$5n-W$o~%2Swtig;}8SE?B<$k*Uu
z+}+*t^FRGFy#3}|X8k_pi-AqRq-p9zOuRYqd3Z{ZwN#hz`Oz%0d5837DS=spI>vxr
zEX6R^xm_z&V{UZZnUf0-YZy_rx|$lnei@UEiWio4C6OQ=g5=47t!_=!FlCB^d(@k8
zOB^{J1k8YWD9rl<F(%g25mhS^-uDgP$oKrnkMh1(ZfNV7Rw_|slY;M3vJf(VE^5_H
z{v5R-InirD7#R=7b!pWk@gw&^BrxRwl|X90OkbA`;SD_yx}&Jwa$=<WvyL?pF}sIH
z5Q6ww)!v=9@TH1@M^9hp(arS+n+A!boVQsy4E?LINKr1%S;0mNN@rQl^jgW&%)Gy5
znkQQ<y>BCe-W$D|`<?j#*q9-u=!Bq4H!%u>&F5Smkn;Q{>(-6R)`3QMzKiaMZ<e_p
zq@)j|h=PTdsCWU8%+QDnNl2Y!roGz(=rgWCR2w#v2JKzsZ}ylD2-WtGE_QGPg&>()
z{ehN1^!MrrQ&8UX^cBAF<+ltHoHEBRojhAzkQC`DG96~q<O-Bxv&8*0S*&z0|AV}r
z{d<x(yW1{jn_~zF#vNcBew$LmJEQLDb7L;*kYFu^PkrXI{9pdkU*`M1`#boPfA~lF
zmT&k*TlzZg**MgL7Y8O`9L`<a_1X>#e4gNvno%6~W-MI>8o+jqZ8Nv)AH4+<b(AAs
zBwl?S7CwfGArQuBLP6tze#cnrfnr4jaZS5fGs{;L2g4*uxk%486in>!AET32ihX{C
z!ue&E>et-M>~`5aGt7#)q1}(I?LPOfn5&0PMr?8~#we{jA=mE#av8JPTE_N)jUVD;
zDfa>)h^bM~4Xv~9HNAci&(mOhjdkp`S$wRv5nF7}SbJ!6Bf)+jKNL~j5;Us^<G{?M
z#)Z%jQHeCQB2ly$>R1*d={2R-%6VBFO0U#@DU#-Fv+U?FzU{|}5ZVJ|ag@fRpf+9p
zZ+-MPdFSo7iBaqz5+_wUpT+KnUiM9I#CmV;UtoWxtx>UPd5OW&BK<zq))0Y=b3dZ<
zxV-Gz2$iDDIgw&wZEDDc-<P(oR!A_=AMw2u-vexINgpOsUi8FB?-qV%qqrd%)$O`;
zM>_?vqN{^Qb5seB%CsG^ozFb`%qRKgr{8NFxHJgMrWV-kGPA?{wRPjUsb07hAsJDy
zLscsRaAn=HtnBuYIAsu1?*;J@f?}9(Z4a6!OoOcot<giUdlknXelbmYH?Nl}2Il>q
zX}>2qOWJ1H32~dsP4Y9~*w!hnLad~e*zNa(5ZTv?&6g8t+F>!GUa^B$ZJpLu>T(8c
zF5h#Ll4-U@<P{?h2jwvNfvgXpTZ3DhD$)bB6(cZi1;CL0To}c{E7j)nqL32e$$~Hn
z6NjJ0A8FsPHmF&x02z9CX%I9Kqi&fA$VIKNg_U$51O^YVI`MJy5WweXejtC}cYKIn
z{mtJ3vCn(;`YYVOumGXy>cDOnnS-*fD+;VdxxTt$3hE&zrdaURXgmoWQ0nC<Rrd`r
zdsN)^)twg9H``V>$gmFro<gsM>JipI^Gm<VuYB~k`NsFXhadaF@8?Ip|Mzh?>^In5
zMwce0eKw75-(0w?8S;+NyYh=f)IR@+$z>caI+};*htA*>7$>8<a-(~}&>aHC4|F2|
ziOWJa^nPP2hj5Xcfg786!70P;){%kWaQEOnM8CJXbwAC2HiTHW!%fP&vlRWBefe`^
z#JT3z(9ChhVQnEl&`P6M^DF7@4KYSzqcaqjnK})AAoJTs30^ctD+I@eIy!W1YL0Fj
z;`IP?QfJJ2QxO6LmxK8SqqSlXzuV<nL1zt%N_WNI(5TyUSWBT+<Cipnk|Hf6v{k5$
zUVT4m4bd}A0`tzSn6)+5^J*;mZm3ahKEm+PTMYQ#<*3rwe_BJm=@CoJp%MAi$3MYm
zKJ_V*2!~x}aBXU_eKV+fQ)jZ?<ok5*%|?=TOZ;oP1uBgOXS`QyNblqn40G{k-R^#H
z^u{{!-IPcj^9#Vf*Cr^2rrF-=rcK^h4ZHC%4#LriG2^!}uQ#b`iHahZXJinTM`ZB0
z|A3`l218wzWi=PnG;{U%hV1zhLU6Mu1?k*eP2^|}xc8?st=ULWf)aCJih?vtORPm%
zTS2whUSAu_d0__Ir%ECAj?`|Wjplp9K!@!Y@FL9+salF#?oZg|mCJ_pRbDSQCK!Ph
z!UY-)N}{7pE&ej$mA-%lP!X}JGX}qa$yl;h*%;wPA5{JJ%ZbU3v>8dw%4}348lZ=`
zJtV!EGgk=if4$9HFGeAQtX0Z_s#*o>-S5RqAE26HPc-79s7sVab>@)10E14vk^Xcb
z!`-{%ZnBT=K_<hu-+2V1P^jzT@9CD5I?#;1!OVDNIeZMPQdG+6NNt^vCZt+e$$Yiv
zaC1!xkzKMwCdC~hfvP5p3>1QfjUZ#QN#ks&j6)EVzLD6hC}`#dzbzVvU$xTfYQ>oj
zU3J)CK3jwl$$|57&&NOW89wp3&+`xd+0XIk|LmXPcYWxanL^yy+*V+@{??&h3Bfx`
z+l7WleII-bM`J>oHn>07h4RDN?N@z4jUK6wjUJo?Ya%x*2JClEo>RXa&OBgh`}+pp
znAH6$Vj9H~_B)Be)wY>gF&or*GU}VRxL?1IL<R$Vez(pcU$d`)nw5(bOpMJR-IXu9
z#koddJ$LA0L^#C@H=kX@)_mZF_rnd(X14}2+ejj8h)YxRbI0XcwMBv?HwWvk92jId
zCHp*=R;)NF=Zo&#o1Rh|r5O5|T=`?7hEa45xLSP#ua?BP98IOuiUphXp`QV2rQF?F
z+&D(EG>22Lq{F~G8Ba^q&{lom7&_8JyQ3@;AV{K%9Yi1b^<U-Lv*!f5<;-j<46~JQ
zt!;<A_}<stN*lF~yRLTdSRofTr_WLfDF%+cP?{Nc)v8pC<e8(%jmJ71UGjOYG@!JL
z3u*VJY8xXCnQw4H=NM^>-CWqzMRqQX1jEt14iwK(z6Ya97-}xd9uTqTZEzH>AKmck
zD^K~>Z+tJ0ZgwQNG<@Bxh96(=Nimoa()mbP&zw(Z<|tgJJvk;MD5Z8vSy|n7e8|mP
zY1DqQbFVjIsb=Bd)r)DnW+O9aysY5#O|J(KYnl2()bgvmUQSI=2jOWzXAZSB(>00+
zeNoCFHFyOJ=3rHp^JT^(b|Y#nUI}03!ZD87wp!KP>P80B5NKzcwHxf$sI_=5wMmvn
zz|hzphQZN!F|x|;u;(i8ylSq9L$>0reF)tY4-P+F3^_d4T-X9m#(~gAE06P%Mls;b
zbAfH7KnP6pj9}x3<#_TLx03U8f#|VfWRRrBjCvdt0pWZ)nPxR6rrm)w**&rcy}M4u
zZJZ?%($4IUeN%~>=3N*w_mLha*nE6+-@Ok$K(k_@RYEd|67BR_7>hi}fJdwIIfQjN
zGlBUQ_*mw1U;HBf*Z=2#%>Up=f0#f0C;m8(A6+pPYWK}8dYYm-PXR8ZsL{P|hl4sq
z?8OaZF%0m4u>sg0ZW>(Q7<3>pT;ON*;HHat%s9w&JA($SZJ))+T{?=H!A}@PmXA<|
zA||+ua*%<1d2r!rKpofl>+~g#AdL|kA`G*xZQTU7;7$s$6O&S~wcK@d4^9MaogV(h
zLSS^`b|42oyX-k`tzY&qHG?Bd`5PAB?jDEGeWsnlkR==cKzgx_+Sk`GC2Flv*J4hU
zIe{9HwXO#J*Y3r}*qeK&*TsUsT1AG_UT_4XS@+wDck=EOij#|@AIOTkwYJJyS6Zv?
zXa_V$1yD*;YM6NA)1T$bUwV^uSy0^J6#VN1<U*PaKKB>w`7U&(oC#vm&k!R)Ey_8@
zAzf@+qg0D^^A^`_-8s#Sf-GprLf_WbIaM{KfVj`Z1C)5Y6s1=S{klFF5jBR)-BMUb
z&P(s6)kiJ`{vd_|_}LZY!ZF(jt>YWKz`V#(;c&R-`uYhU{HAZ_TfX6SlLI2|K-nql
zf}mXOX8W1lJT!V)spU-6i0HiS-Q#yt_e?=YHjbjD5qg8xkZRrn;tj}1)x8)AjUvfW
z<E{{Vgof=co2u-qyk4ppuxbtRG;aQbU&`m7RqerV517)%*erIKsXEEd4^clbZ3uv~
z*%PTYqK9KxI$84r)qd%eMzUD<4%*qPy;I%zqjy0GRBf!Kvo=8`5;fAqG~Z1Z)^)Xq
zq<43m6Z?4&ShyD7?7y#n?RaPhG$?+ECHL=Mo!vij#oIO`R^<M~G)?s0Sx$?&mc?ki
z2KN*QA_hwbBOu~-nWMT@0h#x1Y!NJiI7UKrX0~>!t|p7>ill6oo7zDGU>pucOz2n#
z{vY<=o(0j(U%R!6mP$y;$SrxY18|dAU3C3+dpb=UF;9qOOL1$}<)k*hU(QGV!9V`T
z{Mv7PjDPpv`M3GbZ~yj<kTlrcR{Z(l{Lps**$!s_ZN<NSG6;J7x#}hk>rRMN8ETM$
z78sk`{)vyD@Itk3$d7rse&rZt_*j@Zix)3y%->=BGSH)8suWYckVo-3xWA`=>P=xY
zeVu0$4D=1vi_}R4mFCC0jqy?d(S<pzXi}nd)@Hu<U0oqCI2F}5QMqXP>7!^6FJy;`
zU=aKqB33KJSyy6g+t{#okPfxO#E8Z<(-~chhl{xc*}L)$ovf9$8492|lQ#v)kgzsH
z`URCzw$6KGA=vG|u<tjCoX93^1;BsK7;V+IqIKbXe@CxIZv5hx-lP%f5_oZYZwQp{
zz1@sHT$>pTxsA~Ymz&;#!SMqpWoI`Z3xT?>Nbkht+?r_XK2MqB(Ks$sHnMf`Zd<XC
zuVEt8JDh9hT$))AJ2zvjgYG;s**IZFD~%$&`pPvqPTZUqUfi$TF2gz0M-^BBVcs|X
zaujzwCIb=JPy*jqLh`1oE7Lsj%ImN5<kk1_?Z5k5c>3soU>3_U3awU3R8$)|n)6}Q
zLZ}7l*5r*nX8axkW{W*IE#mluT4Soso>{Z2#-yEUCUMF|O09}Z-ov(M!-X%gAtvwQ
z^>U^|0>vxBOC6xjXmmplHAd&~(H@AwHgBT)n87-q<o}akPJMABg9!~G<V=X!gFCP%
zvv<1MLbb}QqQUJbV<H2wH?ztd71E?+M=9ocVxISak&>bs!6*8BX!vLk9)WZxeyI{A
zKu}uuYO*!N^BTqjh<6HtS<MdJK0xTr4MFTMgvNYz05wm7VX+*7*)1pc&ku1T<q1h~
z>m(s?^ZB=lDbW<1UWf*l%5)Du!Q9(o@~Sml<XqPLdK|Pbwqed>%S5SVL3*REXTSH>
z;SmQzyW7tQ7wM!ysrFte%|-@HVSA8+bYCDHLPVSLh0lMQ|MP$FAMkJf8-IpB{ipti
zCbg4#dH7@q$p@4hwvEH!eVeCB%xq&kzi;+UgJfoJXI9Na&G6s=7Ce;6JA@G`G7g#W
zI}c{>a@~w+*t)6+bkCq*bg>v2Gekn3ZF8!g-{C$M>W9TBgdTjjuiMucrHz6SY@@RV
zuo8le@+NHrt%P2TfgOR`SyqoRRxY{p_QoKx*=%<;*TxVtDJe8_y)$yZY*4X;K<%AY
ztF0Q54K?a0y);&Z5)+ALBd}J@WUV`4%~7ygDy^)}QLxbmj5I$8$t{lw&Nb*rzr<;)
zL_&1<7r@6(_&zwE^^}qqvx@XaE0tOb$Mc!zFJAD4&wr7-`;&R?r)1tA{w#tuIKYNX
z<q(*XFim3&(Jm;V|8Bi^rYSK^6Umc9)m0NnAWf5n)7486+iIn@W*H3#xGzDB?x)b~
z^HlF{FS1_POoVu#GLn%SGcrwt^J3BH^VOa-B|;ablzDMpSk|&}OomEi<AsgIKbjst
zhq`u;QokSutu=0Lu6gbCSD6pjeAn;&c0TaIr%;8MO@?jtr)~vPYi8H3Wkq^%qb&8Y
z2B7u~MzO?2NwT51imAnh=Sq?Q(b6{Y{bBGy<l&|j-4mx@5-Pjr+Mr=m@qM+|OBDlF
z-7D)_%o$Lv8?ulE+qCSn8El*Lv~`H%BKID*2g>gGNH24XUuOxU{v2b{#n?N&da;qM
zLsl_MOkyUfUl%v`G54y=?Am6{Y@2mVR(RyJqt@tiM60rLO76JpQm0XJ^=9#$KG)t1
zPl5%oj1AAa79;iKm@FM|Z0h!CvRNyGb*y4(htXZE25}{)$(sd}W`$@6`{>&9{>r-O
zb}4m#r!kUvL8fxCK)!6Ewy5!c+G4an@2#3lZ#g5QEtZQD;vh6`v)k<hu#FYO3%!^u
zA<?9t)!#eVrsbbSwNuXb{Eff$*ZJ^oe3bvqpZ~Xc&(kON(Y$~}vC$Dc*t8y?8Nuyg
z+6tvP>Cd`2n?0-7lYJZ6;KkM0XhSffl@J^ixWI*@d(iRk1DO~|%0N*VknqJ((jY<)
zrgBaS+Wpz=NXHQtxg4BhM)-Bk8L(&AZsH(<f^XMejH{DOExW_abwW&3jVx;GU(H(5
zV1FMi24itrQ`N?XDrktV|GtFd*tpT2S*a!&jU(-{cW*AlJ>8otdHiRq1?Q+tPEz&}
z-h)Cq#mtlgQ#S9D?24q;mDY{CX5F`~pE3Fy`;~7GhLu?Cg>k1M&=WXoep$}E@x~W<
z<BczJe>zjU5Yj~0T@g2^-JjKM^mQqCM$sF53|c&k#4Mhj1A#In@n%N8;vgD_v?&f;
z>a8}45iz-=Whwam2(qaPE;4XKXGXq&4g<1L^o#SfVql(jW~k%}hTfI)i+gkZ>&krC
zGZDEuG+x}_^WyHFb@+)q7#oZ;^wL8c<9<Iyu6Ne7ees>U4H0<q<S_x|{ja^s@A{_K
z>FYvEu-naMURT;SrirfDklvA2Nuta-0<&>Rh|~%+MQbP0J=xHn!`2n5H#7(#c~WXp
zv~*&t^d8YrZGUoEB(Nc8tyLpyqqMe?X2{e2U9HiVDh8%0vzsR`t^_oIUdRsvst8!(
znvMBngVP1^#3hh;*}?5_PTLI3&IZ?H`}|o_q|qD|7?!nmzy=Yonq|`s%O*=15UqEy
zVsoZpDU9^ZtxNqNh`~F~ZQeUJuI|Pswh<&%w-1dM2+kkslr9JQ2v;!)%O$G-f3LxZ
zh+g3hwnGRG{mZCi$M=kXzkI%j5uFCT^zFe9;x2+)LE^Wss}ug5^w)hhUW?1ST&pim
z2J{MZ08bC)k-=T$dN-bjZ1>hw0qTclaDw7E+w4#d!-o<?sq2{!fB0AUzx_X6@PGJ=
zf06gS{`%HED9zmo@k7!^Q88urb1-fM-CQom;f;*V&^Rte=<>j+U|mO7-3k_`F+V_v
z*cvQw!;|09tv?)(2<EgWy&{8D>Bzz$p%Ri83A$aEaf3GpMG4m7rOMayYuIWJABl%!
z%n%*ZvbGz<V~m0BdVk{|4f0qb23Q@^w*A9LH^B<CX-ecd8@Uumt5i2R@sP5%%}q})
z^}2dMl^B5xPyMW=kzyw8c7%{9O}Rgvv9W$&ih*g$%=5&Qt*J<ko+$>2Z*yO{czT#s
zaPzULWifyq%6;1n3Tv`w_l+;U$)`X2c~0jA4H*f0R|$+Q^|H6NX5{SX(q$X>U5cpD
z7z|dub;P!h<tRSeH>JdwL0_2q=0+R3jf~f2Ay4_@J5Y@`P~E^uu`%!%mA0{W`KZp!
zd)QFauEsYINlZBzLDJE>(i-P;A-r?Tqt~92cYDgYGVc=eJhO|5ckXYwUlu=4M&sq{
zFIqs-6ombpIZTONwoH)oGO<+U(aklYjc@+o2l@TK?>i9+cXxN}9;fY`Ao#f!6EWE6
zpw<;c*iSQ$cgi{XxEP_8O0Qnj^llda?jtcmrZm_HF}l@o#6}P%Ep)>u?<Kom+a5Zl
z+X>aQ(R#BGtepHFe)ZQ&6$4oUlN;!ahb|x{wTltdyCC1-qaPkoVK~nLJnRg-w=y>P
z=0-L;OM&V^Fd?{qvtAA%hXzB8gqR&l5}T`gb&SK<;NH{{2r)Y>WyT+xE`*g4XtKeY
ze(-KuU7sK0g@)w~T;ScY7Dv~gawdedZElK5?aUQxbO!$L_n|i*owsaAhGK_R^Y<Qh
zlpFHOSlC<&f~^$x8#u1He|G7OUKcYAaw3yDVlfD!z~#XFoo}?g{O9)W*N1Hc@vsQl
zib#osq~@&U2V<}OE6{MQolkz^pYwnJOMi*K_`m*N@qrI~(7L$?#89v*JmOtn%zoeo
z(C8^`>P?2vgVh-A41WXnF;hm7G<YI)BaLpKYRvRE*@*w@V(89Se6Aq_rCvx?-bG4C
zey^*O=-B>wc@N{a;JFatU228DR$n8kIE@%+e=ee!Y&c-!LQ@_=o@bNKl}fGMyJmZj
z!)~%YX0g=BfQ=zTN;EO@@toaU%uzbpW#6#E*}G<KRtt;oU8C6ZI;?k-$!b#$SJymw
z^*!9Y@)|KEmh-}!Uw(@(>6duXD>0axTnYvQN3nHK-TJ)Mw&~7Cvr=niaHp2leECzz
z#54%1jZzACr^4sH@Fp+rj?`|?hUT4t5~?n-WZM%3LtnKI@0PJQ*?${^+^k@A^dbZ&
zX0}ciNI@@&sUq&2*)9c+BYH6<mU{NR_wsvf{sQS?7Rkj(3%p3g#;4-X!RXmVFw?l!
zND`R#2Pcd-2+FZEQfwTb-!hj<OfxA(LfUVVZ|zn%Y*KVJ3;L^9ULo~`S<5yy5+4u0
z_vs_P@B4l)`|B(AH`l!OzW0J+X{L*lqQw$=Ll=5(Od&h>2U>wVP3-56Xd*3(eP+EF
zFZE(XY)YY>Pt?}8&@FY8JcuDGqBbtqRv8KkMW`y2Zj!{+o>~+mz>1@@0o%Xd&FkgH
z1ku=&62rC`2ET~WEq9xnAc%LWQP>EB$S`P8YHy@G5z@?M3p9hG(Ovrw@D9nVO*LpF
zig4-t9hwaT&^6X}WE=YeLP}1^@{z@`)<yfJN_WCst;PS|9_|DioNmQHgKiQ>8Cl?R
zAxe$UAG~8R<ipG)T8*Zl2qSaAu4iwZE@GbK1G@GOqK`Qv)}fC?wUDz**&b-62Xe3#
zT;9pLF21|MKwCk2@%g^XcS7_kba)~7=i@@O8wK3xuE5yF>axG!@Vhm_vK5XOmCNOQ
z+q!N=9H@5&$yxV>kAM7Q{Ga~)f1iKr&;2=m{Kx+&Q7+NZL5%35<KBpCn8n~fMgWd`
z71$2Z*zAadct?hzTt2EmW$U1#b0jtnuostlxwu~`>)HNc^WyGZpuL|LSoPv}E@Ekj
zs$o-Y*!UuCMJB3f`}%euj43irnf-pZjRO%Ohe2NL1gaTrC7Yvn8l@KNEU{ahB1~CG
z!PFIJR%8_8BBW_zo@b(ud3tG-b=hX&A&x+^u#FLy<;-A}#~~N)4Wck*Arm-UA9(%!
z-@rF~;9I$Q@{|xG>*>U!{S~P<LMxo>g)}Rw)OBU`Lvc7ELigaYQIObgKSpWNA%@H}
zCtlp%@#VMQ<};uDf)Nv4c3is?T1P?Z&*m@(Qvbic$#B_zmU`i!KuF^AWbw6Y&+I&z
z+;SjewKTMs4ZW<J=3huTQ%mt;+{UdVuO$uzmRN8RG16+M^vlPtwK;dn8SrC_DS=+w
zR$R1NnWl-uZcnL|+A0m<c3Es}a$2}~`YL%gtNPtE)AoBqIWY96EGwlJVu;-AceJ&W
zS~cRMU%3z?yJ_Nkzw_G&yO}DP;wX(a(4d|>15?UqOhzKqi%DK|;$KXOX$n*Y36@2&
zHC)}49<5ZCvlR=fUYLm(YM`Bf5ikdu&}yaBPE}aDa;}}VDV7%)$Wz?p%fav0yM4W!
z&p?1FPo|AkH}VN`$qbK)81!<mcx9r|%ydFXWCRJFPzxzfq-i!suWsG4qMKQQYH*(d
zgYCj_`|~2eyr{)DS+%aQE45onRE&us|ABGdsCTf{=3AR%m30B8-WgJji@RL!9a2Oz
zI1xvX9LRYdOA#1Vn4-0r5k!c_7!AWRR|(TaCN{c}D9Ojyfs|)EKzzZ+IC;|A#cSM{
z>ygiV{tYVzzM?H1ALjv?ZToo_cR<mG5+Kcfcm<n%_5$5ttfA{2F#mwC?M%Et-xrL>
zQ3q$iOSCb{AQGvJPQ~ZADQ<{@+6SZum*|B^7v*^SjKA_%{|dkUk&p5h{uh6NM~@%*
zL2Ct5_?;sm!x!E5^DCLe_L*(`64SJi`9{9I;iux>0mclr&FO+hjUb~9y&2!<<(QFZ
zZG<*B6E;<oOOb0&%SJfrB1Ew<g()o*#^0%;>(_TN5Q4?k9(I$>kBe=nR<c{+I!1XF
zGwA8v^w>&do;<x1VK+@2c01-Y@j{L?gcdLdHnC<UMu%*q)3svxB9YxRGhgo!iL9k?
zcX!X)MiFacEmJ2odJhE29HyC@M~`{*^fey6@{~u9pBR_na9}NkSC6-7UpO9*w$>#W
z`qMh8CXcgKat`K2pDjSms$=sC=-!B+!kh2B;5R<@TYU1<pXXfKR*KeUp;n*|3P<;=
z81gt7oRzW94b;=w=|MIZL965KS)3;01&B8ity&JiT1{1ze9QxCHo3hKlgYcsXr(#l
zV2p=jP|P1fnJ2SBFTLZ#AflL3tk$}r8df03Z8SF~9=FSh-8^%cXO6hNeYA1l?NXSt
zGOvaG_%NsypYueVCj!EfPpoU%Mj&fh$$+GbZ$WL155502ZuXhCk7v$OP0n8J;7&4O
zc>~7(Fw1*5pO4h@LOC8;j(03;bq7TwHY(l}S8F9u8{w>FWo@I8@-aqM0;WzWwL-7d
zwb0guwyvB@<6M=cG?v<E63}Q(1s<`kDDK-&(+7LtzvT6DYJxZs=bdTGB_OrfOwE`+
zTScN_BnaBW76gKVNF}Ou3DJ25K^v`Ga1PC?wv_h}vP0?OE_oNI)QSds-~xkuVqY6X
zw_qZh=K}-2@mNR^=&g8VySFOT&5uMHQIc=uhK3AXawKhsV=b$38KN5rVF$pl>a-`>
za>)_17H+-RfvY12VQ73imExloSl82b=v#+uh=x0s7-vEiq46yr_$D4qCAMy1bnF9;
z#2?0?xdqIn%D-o@{6&iuA;F<;aU&mW&L4au8t>=7D4W|OY;bOf(L;wERq*=TU~>lp
zWx)B}5e@r{;+t4VwrQ^C6F>7)KgsQj7yR%3xBo7$yz(lDBSfP>^k-qrPKJ>a;}A!P
zX(EOp?{m+4N2|=o!pxSsxZKTo35L*!JG%|P`>_4~_KfunO{!~kbfS4DHa2NHqi}>H
zg^OfLvFEaNwr711K8H{L*Xv6}NkN&DJr_N?Eo^t#*w+J_^Pq2z5;;tfoXqxIfZaTE
zbvTexLB$kK6+vSp?PvDa2X^xe2<uwd?{{deOeyoq(^uGCU4sax<B8kbTgrN-xg;%m
zK-l1UOws1aInM+Qlyzk}oh+oScO#j-`jnh)Pr&c3N;X26qj@=x5HKS|8%k&mj)u&#
zF8s!C{3gHji@(N~-g<}UcPG}W1|Rp4)hO2VcmZm)ZQIuOfbUziV4iK{HV#>Y9L)tU
zMIoo)mbW%4vyVP_8A8Rq`hy=d0@^yoN2Y^cP~9CfjL+5kh5*NCYu}KXNK|Lbiy<>1
z!cw6OUV-mxxDo?zLTc}vmou*%4uol@tFqS0ex6}nIG4(@mhI;wW{o_}6Z7E!>h`|L
zkmo4EVYj25?&w{~8AAqZpw`amxNtr#oVs|zAfyoK#SZ3mIkB#LraV(?WnE63Ztplf
zzvFy9QCqV!F(!z{4H7ZrsVYt=wfqDt9_=&X<rA$@`vPi;mvt?aWueu=QX5O_W_PSM
z7KkwsvNbo>fh&?Hm?szxvtQ-)a$|z|faE<f&-PG;z$9+GA;F=PPM$64sky7xkl4f^
zW`v=I;R8Qjc1^5Xi_;EbHmJon#fNis=_9}SGEX;+vCV^%D#UE>qtM)SF#5x#s;=V?
zq<N2s(8@}y1$+b$x;h!-0uO5~#-#Vn&5zJckE#RNP}7qev2BK!9Ez_*4RmeJKXCPg
zYeK;wsMgH;Jf+DXwy+{`CQdWRkjdBsvoU2|pF0-M7#Unf2ba#-R~c;ZoPTDzcfj4I
zPh;fa?=qO~-UQP}H9t5M_#x(F0xJZ<`0@5muDg9GUmUQ%M6;|616>gDamwIiR6g?I
zU*-S!@BMrH&;RWIgdhFEAGB-xV1Bvzmh+$kADfIA9NHYBV+fFZRHG8Tn)heQNjt5&
ztj9+)<GtEIh{R=$tpbh)x$IGb>5$Q(VyIRKBqv?A2Cbqrs&`c(#I4}4ECu<x9}`^9
zk=+!SM;MkQPnZ)Q4Vv3sLs^R<RuUqIUE*p!Son)tIhk@M<;m@f?fX?CF=q~spD;gq
zh571$rPMW+)hvLgn7DrQh<V<jy>mL=vo7Z?J#xqkg`uXgbXL*Ed0Dx=z2jUe^KK?j
zM$DRb2M&)OvA??7p6iq*LYjPi1aBHVP%Vr^P*dfcmz7`p)erL%|Kun5%x6Bw{b}X2
znu%JeBZAoFggzPrHu7qWk<EU6pjPVfziy;9(6X2;m`ubD=@5gdIH+D2n5wI*SrXTw
z&$lV9foASC5X6FiMiC`3(Cfn;-MCMvlCn7*YHRcWyD6C}Lp|TcL5(^b7&}`*5Q3qq
zsB&Cab~$ph+mZ7X&u{PT=ckM!7HH@Dqj+&YWsj&A<BFuLPE4=lc?b1K^ui{3k(5%T
z^~Pykxm(Qt;A*!c?Xn@`T}{$?f6o#Z%6aAf_8G@#&pF*Ml%-JB902n)k#{pVakblx
zqIHWEsLf51oPHLa2-;d@?JKP|*0nl!rgAPT)h5(-`gk+dB~S+v3HI}LJHlb_-WXr~
z^>PHl`9RFOtrOmo;$lWZG$+7iVRTtROk>@((Y@=J6rrcB^F;zvnrLmcjyYH{AU?+m
z(Ij=<pU8-sB*<Z^n6kmL40)Ur-eMw6dz(wg3!C1Uu+3f2B9bTnT>>#A#9_8p%m^q*
zAm-WiqT}z25f3{ydsj2$L81FXN6l$1Si9@I0z*%tR@=ZF-s^+Gu6^r%R0h&wix=_A
z(Op2g8>xKPcYX(tZmxOz`HQW?we@Qo?WLCK+kF`Y!|3QoK@mJe$)uUhKD&J7Gcab7
zV}p2T)L~yBJ_EF`2K%fr8V_Sw!wQ%0Qvq?9evD@%B>($PZ#EYW2I=tjn_uET`pbU>
z>5U)x!5`vbfd_mA=fn_->_~Lm3`oox*IdJ7=&>c*vm9sp+E$R>&|YcUP`8sFDQn|3
z&<yRKq}Sa`BnYi4!*2nX;x>m-n^a>PwMqzi^MBavi3gI?DyO@<mszS`bFXiB?J2w6
zM4k*H&(qEdeV0S2Gxp1JW+{^=xXsMdj>BP35@B@$+irhwcfOVV)y~kJl-XTfv%h(T
z{mm1X+!|4{y47%$z%)(HS!i4xuBo+vjN%XsQ5ubft?gbbcXxNZ`Q}@^{q{R15sZQB
z>pic({$37OH>BN643=J(@{X8h_l@wzWjIjcNtbkf=@);6|L!0CQ$F>n&vJWv&+Yw@
z^HQn4rUqxnCw{iS@PL;zirj%1_6N_*1{d0Ug<vz~=KB8}Behh5K%O!wc=@asZoydk
z!QED47}4s6kko4X*hm3P$=)+cpkp$`%@HyL*0K_!a~eD>XzU+BgegU8H;2LwpzhV+
zpQW_U^V6BbG_lJQeZS*=S*WV)b~{6Bw#G(WJ(4yaLYOkCDZL5rJildkIIuLBu<K!d
zIKsDm^Lt5iro3a_{hT6KS6AFzT{EXZjG3|)>houIzAT0N=P$Uwy=7en!BMt61ltom
zUt%rP#hfC0w*tt>>0`KIoDl7X2vrZJTGm1-%|;4pMQaaszQ&MzEY&Ciq{x(Kro$uV
zM~@%yRsQ9#m*WiN*&-O-ei)aNiGn&~bAQ-BIF9Tl#*i%VGiBm5ds3Y1TnPjst*vIy
zDUmeqtm~?0YHLISF`J{D;(BboF#7F+yEJWVsxdhnt4Ovv;HVbW5rsJKiE*M8ClF1U
z5VFnp%YrPY&Dk$a6DiK-seqY`h?+y)kQx(&H1BL47K+IbL$a}n!@?q<GMl2ITL4e+
zD!DssmWDrryB2CI^yQw~D=|+XE~6B<J{<Tx-|-#%#LxUJLzXrAn*kpVsN808-syXc
zu{yC+#BvIZFJNTFrfxa9&EZVCxE=f>Ui3j>C&MjKxdbXXoM|JG;+C3@42;c3#7uhz
ztj{oU3cjeEv?r!0I2~{K5C8ICfmZm@AO2Aq(PgJ&MmpyC_F1CfKI)Djs2%9Kk+FQ?
zYJ+|@7}QCTwNxZnmk4SGGuXT^rEKF92}qo+AXQHyRM2iABm`r(cM=k_c|~Y<r?HCm
z71CtL&q(UTQJ_}oZGD-g`Zf1@-+NwXo{~#emHqV<Y1#o{^Rp=HN?Fg;y3(p0s(F84
ze|1d|i_0y`=_0{xm5?K8+97Es?+)bs75l3jcDvn%-c<Lfj}C{|)+o*V19tO7Q=!K(
zqxNT{4O1*5u^q0ic~UzeCeG`bQVXd0{HIcAy)cD9kiPi@q#%TV5x?5#gq73z$ba=;
z{at?Pmp{yl+k0Nzo;WWn#Z?Cz304HgfIz~gbh44xr5LxbvS(>5)se9g#!~end`^M=
z&iDb-WTt2XAsgzMS}VP|gI$o3(&6jUR0U-<GTzu5EQ!zxgBS$5t-av?-CNy!HaZrn
zG(~GxmTHKQ#nszr&Bg|X^adRyZKzH8(%q4#d*#W^4Tq}(=hMkD5Qf|1rn~EAeeMV*
zTZ3Tt?e6V&fW+N#<(sarNKsLV{K$|0apvnA-ulU(V%9(kk^OA4`Y1MHJRgl$rLGdW
zeR0S6v`|)aK#j@VU~|q^qy#}rp|<Lo4(7r*_P&8?nkiG?)EFimv}Vbq>zBQ~bt`Zc
zv27;`<V<?<g#D9OxO)7A{r*5Z-@mIh`jQTWBQ99H`m*k=HLEatYqmkUyI<^11X49m
z{FESMH_}mhFQ#difJ`=z@6F(qIGL$Kt8EK|@DG8a=7u#6F?BEgZn@hk)?wMKwmVsl
zji@qUtuagM&i0iUX$U!H1SPmEr-_ll6d)!eLnfai5oo}=t|n=muZSVg)&<defzH)<
z3<hbs(;~w$syl+v=_R6G$y$n;9ljA|y?;iW_N3jl{l0A_^~Se<>j(L%pZj^Am&2Ga
zqIS^KS{S^EQT-3ek#!o;4*6l9`v6uqf}?Dsp0P3ETabu%;u!j&egLa)9D_mMov=vn
z7b{R5^L*^R{B6zjXd6y!8-XE3*|;FuIiFAbzy9i9Ls5R{2Y-a>9l_?st@d6PKV6)}
zD$dL;v{oP_ax$V%X_fQw1*-ZLj%(FppMyjCr7TEF^j3|XoXqhM38XZkUFdy9YD3G(
zK7TUeTo1+zp{#}isEtCpHi9H9(%Ge>66|@7iJ0D<amd%|Yq#5Rm`xtlw6ovuZ0x0i
zc4Mx$DN|Oz1}cy;`S6Ho{|G`PuZ89K0<9}KXWMM&nKbRqkHBH&)+*I4%4@9@|C#m-
z*?M12M|xe^`C&0;^I-dk)`GSbVW4$lpE*S%eXWZ<#H~}$CkmCN6-2FIm=9OvdFMr@
zP|im_^Vu)(**Ct(&;7!$@x~Y5<ajFFpI3AK^F!X61$of%4@A?UWXJ2re15F0i+mN^
zNBWTX;@=NBM)q?gCr5PkVWvu;*45bkemI#N&I!(r<*)16$ZWM5$`;Kjbp$dAx=F|f
z^sYYl9;kzWa~A@tg(cXWKL-!A6XAYo)PZz#MT3)aQO-5+_VI`cJULwP_~x40I(eSQ
zd2hP?X6d8W<fvunxSV^xKk(M)-yqKu?|IK_gn35@3Nf;~y5@^t_!9fWHGAT~H|eDr
z3Rs)Vva2Vx;^P+aQAa1FNY2TUd6SWXRSglTWu-P(hlOCxoetRFu3xF0(wlev12#}t
zRx`W9lO(5!>2PiCwe!rQ*I(n&>+j*}(G&LbL{??F`yKG^f6wcsiUI9D=)mFXl&rJC
zW~0)(X*J`-W)P8egWJ#NdxAu!`Cz#KZ8gSiPKacPpsi5OD|y=4Ty2`sveL>*FODXt
z(j1N+5~U5+^8-#KxR%zeAWOFx&w+18KX7V2+oD&4rh_(GJ-fB6h3dqREvB@YH@OTn
z0>6O@s+IG)5VAp_Ax5-T@9Hd<UL=@%VocDxxqexNfwAI^5Gj4++B*?u-a`n)R&A3i
zD<LSoo`_BPp5Of;_PdF@`_p!>M9dPnmGyzHSGJj&IROUq(RUb7M&&#fv>2-6-|BO8
zKdc2rMjF__4PXo?9vTv<Z&cm$B^bqnKi|P2%KkuwvK0!#MUn@!T0Ez~5RUV*@E`uw
z{}e4NKkx%T#Dz_bjb-9waD6wzXRVF0oV_>_(jb~GXV*oKpc^)*)_zw->k4(XO!7FR
zF<CdC_s)q|D;5>AYIaE*;M{<Qh{Zf>TMgyFiaH7A`94w~tvJmP?NDp|>l+y`eoJ(v
zhK&`YD>htcX&wJ=HfUMQH8SrH#QgycMt-ACl(E?+W9E->z*^4Ka-=PRcAvfGS}hDs
z?QmF^#Wr(u?WfEW1cRrG`@wHH0&492Jnd*Irfg6v;)8WRl+zvU47Dw2wLrC)0?YZH
ztHTW;Cf@ksTl}-1{zZQ6qaSx-V5hiuLUVHIfacW?udYgxi=21ZyLTN0vR&4`ISV@c
z?%k@51}=xdJX_Hlyy6~u|Dh}xeg`?*+FI9Q-Ky`ey>+u4b^>y91x9a;7To!ji_eVW
zd@r?5y4Qhr3?8(-=M)218J1?G#W_ch2ZqxMOR;OxJj@Of(i@Ayn|F7d&L<9&NeQRD
zsiMZ-GdM?ELC}W$T)DbBFwYa~dEv!dZ?ltE$d9gxDf9VHev0SMZ^>!4La~%h3MksB
z%SvC0OAPIP;8DqDv2V$YDJCbi2QR3cIFC@*)f+OLhc@Km-u(|Vqj`_yiC!z|bf)T^
zA=zC?IrHf86An+GlJ+|u-@L*ruf5LWr?0R-?2KH0UbtP4?`ny@95SFvT~5@pkf)h2
zByg=;RUQMKfR@!}>D`mCLZUQVIDK8H^Nuvley9XACc79Id$?mW`xG)72LruRi*4d5
zP2_23)}y^4G1A54S;pKmXuYVzV<DK)jI;kWLeWN0L0h9P1`F@z*@}s>Y_MFb1=ZE#
zWxWe-)tTxxKT0{1VlbJ?dNK%J1kvU?b%VFVv_taFWowPPoHnVO_Kq%h=2K4tEM_%G
zLV86-t#VaC+Q~Gp)tLXU-pqW@cYHhl?B{>UNP#}*Y<2a8;j0T@jEt-WfT4|U^!0?+
z^L7CCZW0%QZOmf^X$#H>IfJS<+Ir_CJ20UpRos~HBN@&P_`${>vqjt(#KsJLTW5oN
zFr3qRUELtW%8>K<#DDak{SBrm@x9;ky<T`3;jD{w1+pnQ)NEUWaWi5Jwz%p_j+rp;
z80m=PaT2wsvxf$GqKmtzsT0FOBBUWt#77H__5O&~#Wn;djHRfKOl3o(>tc{JjaE+^
z`y2vVPjsw^m3n@urTVq@I-ie-RwPH}{T}=ox5CfHLqn38(ggU3$qB-^4Y5ck+NjhJ
zgkG(%>J8Q-Z8=&otGWe)8SL$i)v_3&lTJ>8?srV{mHp*<nXO(5(&lno=%r9PX6iDE
zhHl)FR?oDu5@IIC9raxJ+*_aH7k}+z{K`i@&fB*qmfA0Fd^Xy4G<%?SR>BUC6u7Lf
zS|2J{1`lN%!fZvw=*}%Z-RtS-G2&Gr$8CKlS5DN`9P@_PzAB`c=-m;PZiRgJNaNZr
z*6tu12SbAGkF8aPK4GnOTj$0j<FdE7vr~54<eY;$L_!D?LCnZ$tv2e>-l=M)O))98
z!?G5h9nW0PXAU>lwh#5b87^s4v^meIH3HT^PE+O^zx7-B(&xXx`FP~*ckW1eVrqrk
zFMSc&UE4P1emQCIw|oRqN-@M0Lz%%e{kz@FG$qftal&S>x&Ay&)G2XJHZEP3;>AE>
z-bIpU4+IHhm!77pC&b+qy{u$;llAW24P$NG@%Z|Nn<r1Wdi0pX<3~Jt{Fv*jJt+n1
z=^kC~(bK!*8@(JdP|kFP^?c%dy5q3hlcyc@PAkx3LP+$o+M-1)#3yEGl~@bXSNgfK
zEcZst+g;JhiMAfMS#R*UbnBqC+6@u&?yqGvHh3upbJaqgX7c_Tm0+6UI>r{RT@RM9
z)#{0PXRyE4l^_MJXV&GOD1x+Nq#qR0M2OiRZkuzGCS#SyhL#2C)wHxu7TiWt#e!#g
zEyQWZG#!wToxHX{%$sij<!pt5Kx>8Nd`~TBrfFiH9>gpM9zg1tNl8q!x{&*sKlppT
zm7n>=U#8&*#(?E~5i%AeKoFuvCw4Zf8iSv1PK2uJQc*YJ0Y>^H{<()4Q%H$g23j=C
zgZx4;P=N<pXH^E#agZjB!yq=l6nP+71ia{a=y--2<R^X|0FI|4f9LQ2Lq7P;AK>xR
z_iP=UIASFJ92gYqg-thiJJn7S0j>1Xef~I^j@dWZ=Jw1szcU%?{rSxOva-K<g(t7P
zpD>xJl&%Zq_8F(=Z_%51oadAYQ?^2H$QPRt2<v(>RD3=ls+OX;T|z5L%Cfx7QvI5H
zm36hwG{R;S<~bN(hr@VgV0eTWqSDthUCpsAMInXE$U13dF>6;<=bW%{VYDl~Hv3;U
zxId@pg$Is67&NV2A-TbnT>L_6wdZ&8jk*_~#RnUU(V;jmq=(2Z1(s#ypZ?S@^O4{F
zEH6%#bFt7gtG)d<_Cwm#P=m|k9vS0$Fwo)dsv7gvXimP1VaSI4y6HeR)NS1DucD77
zr#!69?EtTyM_|M`V?zx;&Qc2NTCDi)9{v<;Y}<R^G8dw!?ioSOG9y%l(mSPD)`K)#
z`>{hviM7^k<T%C#Qx1e+Uj7}SoC`Pmi5wfZCmZKhwfi-cHql4LrFT}fB1*<+LA>a4
z<D$`=7@xon^uuoEn?CpfUOa!!)#1RrpXg=bH$L%czW)clhowbIgPbRV%ao0FWMNj#
zuYH#2WOlnA^Blbyac)v?NV29eJAt*g<_dw;8aNrobb~MtDNW2bH-x;W9IengmzA{^
z0%5ZpzDkL!d169gIa9Pz+DbXyQ||7$fA->Ct<je%2A)5AlVb?vd5>!2?)h8H^UgOn
zdx*@#yOVRWxt>5Ng`wH5UMU8jorYxo;k6dZY8&1(&rEJcWDts3`kLO}l?$zz)*mHS
z8&Sl0M{Q8LC0FgQuE=S2a*cfk3)#W>3hM^J*?h6k*0WiOCg(Cp@J`Yg%aSJV1}f55
z(28`-;3h=!WLB7l)HRmlE$edhU?(3H#EeWgK&O>6(si@^j3Xk{cLEXfG&g5!Td=MB
z;Vl#7K+?{4e&~Zdxjyjbi`#7mjl*TC-N`euZ7_%sV<rwX-eZLkTW}+MXgm0^f@my8
z(df^x&-B9uo-@p|Zv?y{x3stS)<YcD91_BA^Sy@~n~cI>o9_6}@V`fJzKV5`zx9bv
z@|XVO|AhbjU-(~gef`*S0aTGzk(3cSbzL@NmEiAh8_m{6Sz#!JQWqz;b$VH-Z80Kf
zu;}5?3aK1v&kM)4kaky=$x&8Vj>NuL2YD_qPlPao2D(^|JK9m_E`*e6%gV?r&na8z
zl0=i5_IF1D`dWR7n6qF^mzdnGz(ehMb%|=*Vu;P5QrgKzb|Klu94u30J)fv`MX<FR
zyxZ*}5KUfZQaBrD#W0|3d+uTwCOifor{qv?5Kjv=o8zuZ$bp!4OtaE_for#Jc@3r>
zIiJq_`fq%KpZm3s^SLiSXDI^-=#Bzh3W31|i1C8jjbZ^Y;-k&k?cxR8h-aXA$kuU>
zvO%|w*vD~V-RgcfGbfANHH1wmt!!&6c>@$e@Nuglb;+fNHYu=Ywp1$w4E$4<BiiA%
zuEo~67aY9}sbKR&SfRQHKMcPVyBskLRKWI~o2xydN)F~Rpmg&GF-I;d1Zy#lS`17%
zTXEz~!glauU$5<A%+y;W)&d1~hXcF)d+Z8oXF1;U^B?{gUwnSYANaoC$CFoHv(Yf!
zR3u$C`~A>u?@DtvGpQ8G*+_$eP)en&D<m5=g_KD1#B|s(?{{Xi-73vT#>=u2b4C-S
zX(mi|eztOEy}#phcaL^uKkvBO9mx9~S%j#a&<b+yoQ`+g_KO>0@0BkfZ{O7#eW_wV
zBoXq={P;Bpk#>4vp$gNU7;Up%jxUIFy)3+D2WIch2zJwskTNj}^N@mclYnTpRKR)O
z+u_m;{_DO;s|V^>EMxNQ^KM2%busgr)=Vao?B7*Z+Irp&>)smee1y6hi7!sX`9N38
zG@upQdNSXJR%yMVWwqsZJ`!Vs?)vlAC`)HbCONeEd^h=J%p@k1mDVf0s$C<6x*SQ8
zNPaEax;o5e?t>u&v=x^14C;E@lufP_617(+OJb%tt?=}E$G3m;`}y*(JRhU>Em3U{
zjfNS}Fd0(J94gR33!&GIS78K8TkzW^Bx?4{UR{q2iV-C|x-#(%V8|^;F(AVe!ge?|
zF9i4e@2xp8b99RjVC2Ccu)kqc2;=a?|Ky{;`7!?TfBM(?i~r953>7<UdohG2#E6z+
z->ulpf17jK<a(NR*6l5eAynOLMXP4A*alllMbgB6f6dkIfYz0@-XZ6P#F-Sl8{5rj
zGdNL8rAad$fETuCqt?ZX9y6sXT?x@y-V$wWQKOOVzK&mcvf{C8g|4h+VYtm1?0UBQ
z9qiyq9_QU^F_li*L3V3$KwGD^g<8(U5Q!l>tP4?$$Qlqd!Df@~eiefmZ)><1AILdE
zBfKY!PI91BtrKHl+U-oS)GMtl)M{)0lb`-PfA7EhIX?fTcQ~$%Q#I1ub};!f1|ZQw
z(ps(Ns^}dq3@;D=9NC>-tI}nR792edUO<O2n(EFe=vyH@C<C<=Gzq<Lx`FNR7uY1f
z(F=tbgj&^Qfsx*XrLA_I6l!lwF?zvd-*2g}9yP4$(7F2I4<br!hOosj5^OKCeH#xY
zV)k&SZiYu)l_`K`v+S-atgX>x7=RhKN%T^LXloC6QqHUJ=3~ogXv(u^x2(_am}gU-
zoR^i*3)hqKsn5K@vlmDH_>cb)ZmzHG`lN{_##xKOoKvG4Nr`Z=>zC*`6h_`o>X`{4
z!qpX5j~-E08(XIosLkDUdqdr;L7Z!My1QdJFWkSlC8+Y=*I(t))iqc918JT)9InVY
zA+^$q`4Ft4Cb<uhvvzLlyW<<ZR55UM{e<i5E1Fg`Ms`o%i>_x3CW_|Lg=y|Zk{!S)
zBJ<9A-8lK78|c-ldhHNevuaB#eeImj_r#R9O@0t+3`Xl9X@k1(hgz2t@pNZZqZ-rl
zc)BO&OkYn_HJ3yY2)&R5G-Z3>)|HOM8cUkl$aPZ9ECUbPP%98b?7&MgP>L;(7!%Mu
z0rAd1w{0XLMh40sV?fC)_s>}GFEDv>;|!aFx|?Dc@S$CU5X>IfW^{EsM8Ou}kNv^#
z=VyNHH!wpm-y{ZHkIh^!+yP5-lax%Y(FSw)(s7872xQDO$9sAAWZwQA2E0Gyo?M7b
zBlzb50=&&O$LCPzd9+p^6fI7s3w8k5h`<pAZzqOK^wL+^X7EeD{ww^q|KxrAkstfl
z=|Kq5$bZ^BR@F}p9o^KJy}G8mma|2Yiv2#39lEZDFiAF|X&6({ZXWhw?;hf3;yjb)
ziBQzK`xrgSRmhlJwAZ?gWn!KoKy~<2yRw{$Jr8BIan8#u)UUx;m;(}_6pK_*Ewr2o
zVIqpqnz^gBW<`6-nMwEF85s+_hv*u78>y}9kyckE?p%qXF4Ho!K`_?6hT$-2hir`5
zxJhco2gTU&TP_5{yvK`LUxT%GLhR%zpe@piaDR8t-}}cu!LNPf6TJQ6%z16BwV>4y
zrRJS)2!Yy)9pJ4(OlB`Dkah<%&e2ZW)~A0KwUI$@7uoPe%vBq&ePzLMDcoyqX8zVY
zr8ed<zm9hB#2Be%^|3-jn_GKhXhHJpP@Q7{233m+wL=)rIjWnKwtGR6VxqYROG;LB
zEbdg;)w_4BxQQ|O1k^T(R;x9RG0@J9Ih(?u^v)#0+=5y0YNx3=eTrCtXz70gK^Y@U
zAV6=H5UXA#-5rnIofn3T7ZD!c8~`(3dH}LOO~3Q4xB10i`E`Eyhkk&^*H^?8Xeqj#
z{3w>N-@kQ3mb)becC`aO=fFJ8%-I6Rq|U7SJ&zvWkb`iZgN<38bEhH}R1_mE1X`&C
zZCuY2*Ed(Z{>ly4j~_A3Gpb(HhUh_8E9b7reowf5WZn|zJJRj*ceO@esu)<yVj(+q
zCG-YqLTHrZt+`R<>>IrRX3?0UZEj({Lc(m_^XZPz&n$j;r#M-cpjJtbiaLZ5)+0FW
z)v^fG2xe)TAtWk5FVt`{m|Zb%fNVs?8Um?9nrF&dkQNEiR0ma?xq}B~UGLE~Fl8e*
z_S%Tjkmy?5nB4x;I%jdCBESfgDFL0bEaq0|hbjVE)xtGop(AKrX>~Per9z)fYGr$o
z!)rd@HchD-7ovBdZDScD3Ryw)ogexD-|?*<<RicRNi--B>D{%r0CJHG;tR2kO)yT-
zR+IlJ+iX$5nb^ib5I6A%aw*m%utE2uU{gl|1{_Lw;4AQ8k*atx<CXR>;JEmy_s#xT
z5SLf6@d)T+6yiE=cb)ss|IXhf<{f|N5B(8T5}_WAWECvtGffB6{FbA|j<%sx!8^Xy
zx~Sfb15m91&NqAZS5NGotw)+m)Pxb>(usXX!|Z8Aft=hBNlnpJE2@T=8H{bkvZ%qO
zt=G*~xvVE@IkPTb-)e%Kvd7)x+J6P<M5$<qhJewb8yW`|=o6F4u39Zr2_TCfj+L^W
zQPkdJcxbo_+ek?bu(^$n1I+-QF|T$cAdG+<N2(mc9vx<HhO~)TT)Oto$3OKM{>Fd#
z_xbYMx7?lzr`4opL6wBjfK`3A+&fw;ZCxp93aZUmrZuxjR_s|1m#BC9z{2Ij57odZ
z%<VZDz5*yd$}=s$b?yyji+xVc#Q+=Ms*nB#=C(<$+r?sdj5V5iym1#=?Wh_m;=>0U
z@wDz}gBg4&ILd1wY=#_;gv53kclfi9j?cp;iJntLw0VKyiU6_EOz*bO1p0<GUiKnr
z+ARsOT?otK2+O)Gtf$kq9nHJFjZ9)<3N}vq?N5B1*WUk)eCs!VfTxe`?^8-d@oWS!
z^rE*$C^n87>5Kzan&<3NP(yR)-Gt~hQ_k!UlZQJ+FN%Vnb4r)!(qr=zxZ3Twd2~bE
znY-orys(^CdR<w2<90byiR|~+q-hTc+H&&s^wnQ4RScZ(o)g>I$tTYVZ-1FIO)k0U
zXw8H?lj4r^x^lih+D%N6xhb+1%6c@IJln`1ib3Pe=Nt$~%CvR4xDG~RT1v_gCPaey
ze5V946C_wN8D?GE%IarMAWah?MUHp(c9X>CV9iGhAraz?;`7;ZW^EM_g;to7nFH0X
zW{BaCmXQ^G3Hg7EV1yN)0|s9NqpJX@y%5D=Vu{2QD78?^dO1M7D`-xN0x`%x0|8oF
z5y@!G<_oSim!Cr9kNn{G@tgnr6IAC2h+s&HckI5Ecr{xIahTvlgodss#ypp?sQhbt
zSAyca1}n~J?&i1=0bvvXLmzAV?&#ogMZrZvXlvhf)gie1*aQ9m<6r&12Q#WG1fxB{
z19wRYtflbR{y%?%vKD^qhkwv3_+Z6D3TT`x0BSjNe}7MiiQR5*UJ$*R!Xsw17S>)N
znN_xhISq4{?xaFjmI&0cvZ|UN!hnS_+u^pZ)NI`sy^_Q>UsDpSEceeCmfRAvzlRX>
zj`($63=o6E$`hqVF4ns0EOk4WgInE(m>AjhF-AgAS})YP*a0m8g1UI9(yAj$oBO2^
z6q|M|W;_}SDJDxsFcePoQV3v=R@Jr@6_V`W6DQl2)lj_8e(nqW=YR9>^47CkUK}gu
zwK61!IYzWLYVAl8)~b}Ul0s*<i>O9Q?+^ny3di-#B9>u~XOO6ebpe~Sc2hVw66<HQ
z%LI)VG%^EJl{`&|cbCJ1r8Y+t^&&kqa%I?JervOsVgYx`^DGrcv1&-w>Q1l$NKu>w
zec7)@L!+BBtS}NxjKCSUI%g!8j>BBFlT(5jzpl22EUT#q1SN_xrQp|~cp#MT%>zqQ
zHZDb+!q%{~uBI&Mjd`~N5%!0fIoUo^s*e`B@`+D;k|$4JWuAoVEZzj_R>0dHFQhy%
zESh?6v|4<BP-<O?G7y<yL_#qRjrzDJr_7YI4RO?pj7YffMq)&gSvl8r<#al7I<J6G
zMOb_1R95PJv@v-v+}*y#{qtw<YKgvFF>tykrO1>L>$(uAv^Ep-jx--w%0ehFSX1G+
zcI&>*_e6wcJrkBYi-<g2ljkc!NJc)1frOCqOxI31-_zEU!(GM|2q{3!D2Y-l>sm>*
zl47xxzc`u0VX)2fyqxITxO?$}lrl?Mc<04)ZXUlz++C4lAe`@!oY_rRpq*v8GdD9a
zi&R2%t#?G%)ksFsjdfzS#Zrnz(gr2viPFwct&oX6V@Be}{+~jmmzA=vcDSaA>FTk;
z{%!aUnEPE>R|3J39b4smenyIRsHQm(L*8a|xP0(CzvTnG_V|V`z4Oc-%F$Kp1!~{U
zc}Et4f4{^DiROzCE?cT?+|hcSK>*Y-Hyakn4-fC*lqMGz?^X&3<_|GO1=|?I{(a2;
zEQxWT1L}u`jVHtxh&V*24j0Ji{v0|hwep|+mw%hJocY&&=!cp7cbKArj{>Uu(x+)c
z9odo9n7k4vo8gNPqAkX9x~DD+IR)d6#AHa7k4Z%KguEk9J8H4cHP3r`hY$;`b<W2Z
z+}*xlKTmG0TRENXxO@HurrkCBN3R*#ur{XMo@tsn*RS(Iff#96f*2v!1EPAX^s<1b
zOU5t}b8_c6@jpujUppbp?5#n><6YGVaUn*c7mYR|@UCqXWzx4aP|MT65dcdyv~@So
zFa`Q%zuQ(reYAD{`6oZl-}u}An0H>>b6$-P6@wQuF_5G)?-Dd6Pi9G6%CxDHN~wet
zxjU}zVA?qs+tdwcOmx9QeVi<LVZ!?e0o_Qu57jr_JM$p8+5lpR+qhtO1jHC=t&?5j
z4r9J=Bg$5rTQvK)kIrsNq$sR~ZJuA;cLfZM8HcE#B=_UDYwun(1L08IbHbzPZQ+up
zOkFQ(f!d5?l9F}ZAqIkWQV^1i#LHmh#<C!J$F?5@3&k@tytp+uL4=}8h{7zH-4xmF
zcI>C<O15A{g<8LJ`{EtSx^jD*U>2^kna-Ijs+%o^Yz6!_5>=W@ygLml1hMcvE6!}>
z8*{QmNELE2wM+6)Cq*`YgL5gIjz`vZN5z1&rCQ>lBnu;wm{{&^dGY*h-u%L6Y;UZl
zP<uD8mnsH!H&2<LzK8QWU$loNPo&)qX};p-@hj9?dFRb9A|Wt8e#E@n)6Q3RHq@1N
zzUMIQnD+;o1a4nEqqf4MCr?2VEm(|ksfA7?rA*A3^SZLuX3WW;&PENE?R+{B<B|FB
z2#Jy9e6}dqURkQmc&ozo%_Af5)zyecc_uoMu~tim61ix1B?Q~R^9(VP#7KR;Hkv}p
zwit~3SFKP{v&$4UQ;;rVW-<~Wm=j)%V}S>HU+7JUx^ss5=meTEdb5Qsg*2f-xWB(;
zih;xaK#0k@fNsQ~5CU;`&68JWe(Z;SfWP(k|B;g%hEa*To9V@PX#BI)isXQX40*qu
z<-J<BGd#XSOcx$Ou$Wfde%I!z7dTVAZ4oa9w!`wlpXpFI$b-436Ql%(0ObL!JHF33
z14HRx*KBYXE+k12Go$$jKk?K2-f#a9PoF$BH@en|sUiq@w`bm8A#@};VNFIRL?G`D
zhB&Mzv!E6ss92X@3n5M9I5QF!wN>-gsFo^b<kn^c*%T>t;k+C<FDL7QRY+Ku^!Swz
zAXBFFNDdP*LS4`F^=t}-ujiMJ&en=HBgC3HSrzDYxgbU%kkX8dM!=mITPx1Zal2PB
zDA&4Cd&WmjL~tm2Y|I1gGVMG>O{6j6^UMhAH%I4Z->ZIDk<1L{)1UhSfAhco3Ep{r
z&+%MX*J__fY`oc8XTF-a-cJOX31Xa)u5RRmu&%449cRu<u>;_|a(_OP*dsX_dbA+{
zehFPJi3pL%$Wa+?gJVrtY_5w0Md+x7HQ|0Jp66m7D28SkC&{i^DJCxtGR(y=+h$k5
z^klKjrFeFODyOw<o&+{a7od}Kbe@5doEsn#+!fTv3*IPL18hY_o=u7!#o~E;8<`73
z0b&JP9K}Wwmh*`@?@%3lD<d-6njL$_r8s=$W=DvYSFv4-TDRvg<xE65-o2odBj+`f
zBr%7?lmaPFgecy;SXBEMxM)=(6B+}wO6%Tutj3Koq$V)MNW^j<rjUpsnT@r=S}Ubg
zPRq*O>CFA{$nkWdsgYrc0b(=`Oft1uiZjno&p4ir1eATwZZ!4PUoU4m)Em3qOq(;O
zv$EFC)x1Na^8EQ*mhC)G?5>}3^UCX-j`x()iM|}E>xt*@yv1jaXRfdJlverjn{Sdr
z<eR?fgY2(va6SZizGC<IRRr_u-yg0~wSb?yXK!xd7$dBwEJv2wn5vOgENj4sh9LyX
z`N$Lk`<&^uu&!si=L($ekBFEqG(@4-m6S~07ve-O7O!mPF`zM`tx~-c5zmh*%bE3f
zLZU6W*gJVjO!E~@618_?HfXY}3&9l_Ywh%y$tgLD*TQi^GMHH;(;DPB@#M+t#;o>0
zA#Kf4tEL^Nb>;llmzgB+L*MhA{KU`x0$+ai!el{SJ&&*=Fj#J|==*R0lz^li8nQ!o
z3tEm2Te(ech&HZCGqsziLQH0>J9duYB{0ZX7mflC4t^Mfq3s`DAQ+suZ4Y1Z2nwU9
zurQyn?Sj~p)*8ACQV;_#?(X@yU;ik7?1z59pkonZmb<*QG|LK~=NUvu(;iI|wJGy{
zZ=MZxf!2s=#@sElU$2R-!AMCVf*S@InljnKZ<a<GL<u2tI6UI!<~4AS{ni?Bf|w@O
zZqL>ftl;h4e#iPQ8NpwxFBx*;ZiiD4GZ!*1h)U(+ylGl_FF3dD6`$>U^m(xO-|ld<
zJ>p_{BsyZJ$LOdtB&>sR-_;6uJA9gRR}>94x-;>vxU5cyA@iv>zQo`BZ+?=u-hRPa
zoDC1AK^J3_*J3|w+Fu#^;m>)J;Le%mJJEXM%Dd%S3OVjQsc<FdiER?EI5O55K}11p
zl=d)yX5$Wwk)3)bfR6;aABHj7H4WbR8iBBrqOszaW!Yx;t#@KH_!_nR09)Vgy^tfN
z8o9Z-Pl%aJ+I23y(~G6mW)Beyt{OmKS*>BqSxA_pUa!R(i6Hb+d>uu)BMx!!A)@_$
zK4B!sKy^h*2yxTy2S*0FdnOpt*l6l8&whPF3Y!wcGGc;7LHl^?dOz8yN+R)a!|w4l
zhr`~y7rIIKa5rA<O5}m5o+`_7rrh6C`a33<?Iz3I<(U*L|0R0zCdop$?vH2QdgmE;
z#}iAbtYvjMJ%mt*DbYhF=Ei>7aX8GTP>Ye<XYcr!JD7F$yLi1^gz@@<<Ci~;ijwCE
zBywJl%=4b}-7U9wF9^Gt79(q`BsEt%X_e)C<nDapOV3{L-jtY9<h9q{&-rxZ&9~m-
zvBLGECno<3I}T6YOF7-KoS&ISb=uL^lMz_Jn7XZTIxn>KL@Fy+X~)&gdpMTD`FtW~
zm=m0jcib&!@?-?Z+uM6o?ri}`Wq-IP<cu~md|BM|CCI|OJD55`D?OPtv7`yw;$)i6
z>2#)*g*eSLgx-ZYKoeo*O!Gy6m@l^FOJ{aB2h23m3}{low9gKa*RE!7D}h!zY2H~k
z-j(%yMEgpWiAS%$MxPt2+7|Zc(PRGXpZeqc)xY&$^Zfou*T8T&8^pi?H}YU3-8}t+
zEmkHU9cc6Ux6(sEoP$uy5vnCPHZ{YN3o}5s&1|RvHliOmdjUrl2BKgajS)nBbALd1
z^rh^d-QHzvj);#!dbbrGg1xUQ(D-LR|I7T)cYg=3Jb4Tbo0eKR-`}CFFi+W7V&2JV
zZ}jEJTF;bPp`5I9irGg=kn*0=I_v!l?<~|2li^>WHt)|C4Aa&Vy|2cu9`o=J4c!sr
z8O$k@(~js;Bhw}yi0iu}0e!8$rkt6k9je_n;y7&?B`r-PHB`?025|0%!NT1Ma)58H
z-4geFbRz17*x>DK?uLNRv5T9Ojqsfq-Md5`7EX>w1v{Y42rLi~S~q?DfAbH2hIgLb
z(yGDrQS2E72c^_Hu_;GFob5T%hIAZJvmAin{_{Bo<|%u`bz_<m>&fQK?q|w~9q(@c
zJG#8~z`A?z|F$S^y%bO8@6s&}-nkJLfaAzuYfd=qtn0evacr_*h4ZpFf1rEe5((MH
za6uB)s#Rm(sv!fdnUbQkPV3^3dpC7f?{3p8RD`>8;d*CdjEHRoW}4P1r8Dm>r$(xh
zeO$mm3>hOvyYHo(nDW8USuv;7t)T2|>vE~Z#tuPQ&I_d!vLgi1uR(W|MMU_-$3MXz
z_<<ipW9I7V>pXey>+JS3y*JkLJ?rt7zAmOTl1NC?b};vDaoLMRYOmJJb!D1&WLK|s
z?S%a`;pPcC?O0DomhyR?x3_tAf25bnl%k;#Sm9|l$6`{|Fmbqk#MPt6TtB*DdH#%g
zI-+gm{`jvVsOaznnXj4i%(1=6lmmI%vw!p|$CJ%Rd##ig&ndj%aJXVQpGnfWx_ZR!
z>J`52yI(_l;r7K_?591;Qu*Q=pCP6lPhS5ZVYlb*t<Uk|%b!9^<@WY%=G}qa!H7fi
z)io*a5a9UY9o~HNOPr2JUVZY2^W8I^i9njzJ$ls^P|7R>-hShAq?9;ZUDImg{>7I;
z3$HwR#SSOA=X^eKxO&XhqZ`UnSk5Qv>5g7)0Zbt>%{zAc1E&|yxPA5}y(@DvqR!Rs
zz_h#K#qos)*i3X3NfS}weEYVM_U4_>=d;VqR#X~27-Y0L=3xnjty<bu2u7?tozD~%
z_PZ-|J@d{NKE>7b4XGEaW>xv_Z~Yejr~jir#{c`T|83_Kh|4n{K=V2}cic?pLa-T|
z32CxfK4ZmsW!=iQX^hC=Emcy2L-^xIj{?Lw1bx7=Y(>EMj3aK;$V8(Mcrbn#KW`Kl
zz0<`vzu*Xko2M9|EYR15ciwuFfAq7zz@PlXf0#LehLxBSIhq3;-X(3UcEz~@fjMW^
z9&OU>W2`ZPsOuRNmePsZ(Y9=4EPA0-^CXzEc@9WdZl6D=729}2+8c39SF}&$yhECq
zBh_M?Fr~3Bm)wc3$5)PVL)@aebT<To!_mY@3=^u&WW2zJVoBHzIqga>&D0etkOC=Z
zLNL@S#zb${?RFgpW-x_FDWoZr<{c?I&q&pMLwv-5IUfiAKAbc1(ck<8Z@lqkdaWi8
zR*R~wZgR8Fi8*C@@63}S0Ao~zKcA7fktKr&U0}a60`^@yTQM!rYqLxjbjuLXZU?&h
zxGW5Wtq(+?Z|l^{aFY84j9R+4M)8i{3JCF#9U~QH4;*UkGD57iGf!5u)w(cE+1^a;
zVAc+$R@2dANsX#X360p*I2Ed#+-T}zLu|zGps=iselkR0KP6n8Y-6BmdxZ|Oq77xV
z_YoYCaMYsph1LQIGs7F^P&Otq!sAj4r?O%Z0G6k7UJ9+NH3g^_a_R<5*5<@0<m)T0
zUU|yl=~H&Q*_30dtoQduP!0CI(=?Ii*-Z6(uM2IZNMYK|hz3X*RpZTQOypPJ%k<<`
zBu}*Ck)(m$i)ZZa?r9-0WeX-s*~S7UKaAwVBMy%aOp|*x>>pDPGxc~+Io^-Q@2kFE
zsu<{pS4h{7EKc}%M>)L%B+{WGz47$Pt1w@4_vO!Wee;+(TUWUp?})n{^K{_m@gs5$
zJia;b;@K^=bY6Sq3H5Z#a(~O!lv(ee(~p(I!FUHL&(!5;v-yz8)1Kay)3Pv46H~Go
z*)(O=(;aEPrYJxs)H9EtzD_>eP)cDp&*Ts|9Ny;5FMgWydZLPu=RJAabJ*_*zPP4&
zPo8Ikc$O2#XYX)YD!cuaD-49w@rdS?6a#m6FW5OOKe~q2Izc+6TNPaDnG}>slzQ%F
zd)_<ida?x)@;0w4wGr=6bS?Jax0$l8tmivpBew%n5X$Wf*7GeQLTR1d{u)W}dq4CI
z{K4=3Hh%WQzm2wGKy+bFS9P+Hb0j2)@4}-avHLzc4_nNZInn(<A4y!}dD~{Rs@vin
zQ7Y;qm0<_0+xyCcdf;MdEFzm*;5Zo6|NJZLk=3`qZa*o2`y~jXc1ZNP@{fM<r`b<C
z{<S~w`^o2{nN)$Zd5r}e^Guv(a!xGkiMpJLae^=z>{>g$7ht8W3sa8NydwoGG|#th
z)5}84lLeWy)wu&UPh6J<VIu7^hy4w$FZAZ4DX}Q;Iy~;(wQ{YMWhpPSRKMn4s*T=?
zxy*^UT=h&)Ji}vH+ZwN+ZQZS(rM@P+wlOpScY=$_!oOlj9-+IV_D(6LEJ-obW2G*I
zS}Jk&?`Ozftw2RDMp(w8Z_t$E`N&WF<S%+)4ell4NrQv8(#>2aMRP*51IRKK5Vyd!
z3jqRY%0$hq%h{Y0ou`mvWNn37Yz%3i)DF*wwz~FK`W64a6$7dpDKDqQTI}~;f*~g{
zMspV&k~1&7hl!6R4H^m|$-ZuA_r>`DgSp>O_PcqL&Y4`&EWHJDF>DIGRF<_-n>!Bz
zhuv)EbkaE2#=Hts3_~d5gv7#B?Lh5aNonu}yvZ0NA-^_uD;63`(B`H|2IH5qaxN>U
zQjC}4;Yxxq&yyQYxvW{Rxx5&vk+Sggm21x3_MDt&(ll=ljm2%Ra}-CS78)HEPf0B+
zqMf`ypz{G$p{lz5dnV2Y^34_3HwV%*lcq>pPOR^JA6bHpHM5OLVl)}Mqy*8_M^OxY
z4x?{M0m+FTtw?(}ua_zYzWmuw(w@G;qeo9b8(;X+ml56~zwsI9oe(o2?`X$&kaFU5
zyl1|8OrCd~Pe%?ha`)xW)4CnXlPF8OC$yCuJGak1N0T8lGPdiJSHFSt`ONa-O|Gu4
zZBBhYbN0jY>iQ9)m3hkK=|HRQr>!f;(><r-9oLVZa{c%<!lTE4kvOg%Kc?&Z(Wb=d
zU{##J@p#AG^Dom|<NEqBv`3`0Bjm}Q0V~hm`XVt#4%d$m0(HHooKNg_dtTf=qp5PZ
zy0T8Ptd#RTA!YLZis^9Uu765d7UL@9opl8v7`aU#(#?bk<#fI$Pc!@dH6a+`@pykn
zS)U;{kJvqV)y!9VA;rvES6Z_`r$6&2{wOy$H~jQ3e}otJM^h{K0>ULf7Y9L`xsDm-
z=<3zl`OQd3Tcef5Wshugx;7wt?9d*w{o#AY@CDzH2`-g}3L)smE3iH4GB17ze~!P~
zeA$(Y*TMKXKB7PZ<Qclj_kQXZevu#gzVG4j%?<Co^%kdAxj9@@)|K7iK#a*dK4=;^
zT;0&R^107_n%z9}^wIy1x<C7oEz8pUu-_ifU}kPU+&MBbvodRDw>XG|NQw{}0tA8&
z1oR+1>CY1&KyP}G9t3Ct1rQp#sp=|rO*v;o+_<B=+jOQq>0xa%k7_~ejE4rYtP048
zxPHxTIA`y**ZRKSx5X$W!8>G@;&6>D5X1d9LUD|#W<-eBT01K@eLa2oiWyZ(VTu<|
zu<^Ci5utk8wKFI!@Rz1F{9}GB%ZPW1oONx_Lr&tlIRm!N^g+;a3a;HB&C>~3yw_p_
zw$5RkuUZZHssW6$DBo5oiNaXX7+C9xMf5mmEl!&=8n<5O=eBarN$kEU@-P4DXMFw5
zJ90?%)Lb{DoG?n!x3&h-oJU(yVCEDgm!%fE7cC71sX$U45{mL(ki167thErxm7XYh
zA?m4^Ki>)p&MAc;|M|SBStIf8=g%_djI~k-oI45D7(%E>@LELpD&9Ni7M!7BN=v7#
z<a`N8a^4b_AO&8|6}D4tq_qn$heR>*PE&sB`OHh8Z}sJM#scFKX`GB`#%Tsr6**^8
z6w9M93giHwjaVmD5u0g=iL5h_iE$(*C93}V`GjhV)I~ZPN^1;Ym4pd7E6=a0JzQ02
zn5&k*##PPOGU1&TocJ;k=LwZFTG=Y@wslQRl3W{QOr<Ky8D+3au_PI>7%ODgw9Jy@
zlM`)gxOw@4tL;@iKuJD_%Bd1m6|P&+G%L{5@Fd}Hq&Pnnt-o6>@I5;N(}}~!nc>q1
zy7ijw`HzXG2jcL6>#lH3kID;I*J~7tM4@RsQ;6)xiNi3^Sc_JLPs@VV(pDQz2O96$
ztQ^ZUu{-X#*@)fm@cCQPvarl2LYQcpmY5RLv@pd;-?eDvu<aVt^)$+mCMgu!e#6o2
z2+JsH@FX9)E`knTwH?zuvE<0%abUY{XpEzA9djIb|J&b6#k#)2Hy&dxVUdh;C^Xj6
zwrlW)FpUh;NCuj2MZ0-{Hqzy&oJ675SuGT`lt}^r6)R#4G|u6i2VgqwDQTu%-?DZ!
ziLfM6i8Niyo9#7ZvB>n>m+TLZ{Oqs)8%$0-yL~C66t|`8J^$r@_8;-1H!u0${O|uy
z>`$Y#^r^~R&U3eN-od4#Uli1zk@oDFE+{9XxXe#1_$pdbn5UD=Wgo8{hRc6@I^~~7
z16=+`)jHuxdvJMOJn4>=dO8(d&T{2kKvYT^(iJGVpsYgsjxrbSA9wsWfAO#QpZ+KR
zBWPE=xO<6rhWl^7miA>yc;B(aEQ^dW<T&B2VYS+DIv!9;T7jFZYqT<yv=EmC9fG7I
z<|JxjZE!}*Df2uAscDa?Kum!U3Y+a6w&@{9YPdG1<D+CMXiMvqB)k11euOXsjkKm~
zXNeMJzUmsRbEMh@Qi|I8g|jVHTbYqkGRYcjwixShu6=5CSJs76h?u!<*m5m9lu>xC
zP}+)f22~3PUH#l=ZhmDCl)~vW^8fl5e?zJEHeE5@5{;X&&S0H%CC?53UF|qB5-3u9
zC{@Q+TsHSo85Pc2Vo24epJlFW>g(g_Hc|DtpS=6$lW`V3Ksjp(r0CSvpwp9u_sl^s
zMiV73t3r=t^Yq5hwEj|9v|Y<txM)xUO-eHNKbPKOtCUql+)Y3(I4iG5=TaFfPhn#P
zNm>!|WTIq*Fc_1x^@}D(#k2$(Ygv|ff%wEaW`VP%Ql3M}t;<yDg=doKQtg~)A)?k|
zW529Xh<@YTWjsI5&Zm+gS8L~Px5qYX)|-LT<7dJsT7ko-_pmJV>nr-z720~a*R8CL
z$vU!^bv@<CJd34NX+=p&(8NsSFjHbfDUw>bah7&_MOt;_lu2<WFAH&+YYMNS>pNCg
z*O&^iT9%n{IB-;%IE*Nxv5o(o?$Pg644Ag3-E6VeF&+0@-#sUF9y71#H`g3a2Od8^
z;#$ki^DpqXKj3(H;5hEcz9r2wwm`eO!mc%Kzs9d_L@crU$Y!;c3unD$yWQfw<9IwW
zXXy|u^CYpq#<04+!$%SJl`x@=!?)KsW0_NAStN#dy}e<X8%ZW~EvxlQlyMx>2hw~(
zlh~}bJiB|wY4?#BM#NRC8J5VnBuuYKOCk>go6Qwn*Am0R!+y_Qv%=KHWpmX_`%D|+
zuxEcd(Dogx%{94zZ9B4RD7ld46Vo(_RVoFk9yOFSLeXrlZcxh9_Lc~sreUJwARhN>
z{|O`_3gryXUc4ck_Mo(+XlbzovS|MF$8Y#w{+IuZ|Ly<b|H5&Qf<SRzlmuENCkSK_
zGNuah&UU|g8m|FS=Lvo;HR&uv9lh6u=V{b%(HjU&sEz`-_@pnLfHP<Pe6w@z4p4vF
zxg7ZJzt_=1eH6~~-ue?ILln}V@tzVRfAO#Wj{n(z{?9ORL1hs^wcA?}<2a9NdgA_p
zI3C!25&_q<=P$9gVTcR41lqQf*6VzvF#=$T%K|y$ogvR7<RrvQZP3<<7*Hul1#6pH
z49^_)pUJxiN#FDyZybJoO^6ZKbj;(C_?^$;ALB#DB+?bg^|YjP5y+CU7~^WIeYq_1
zTA8P!uUc8p$$jNq#Znnm>YUW4>ns@*mW25FsUtCIvsx+6Ybkja&Pu1LW*sYy_Z`3b
z>Q{WaKU_|}TFF&swTu^<#^HT~arSa$$_AcLNnVpOPicYYVxyM6GN#ku8&4ZDdXxe}
zEUvX*SBu}A|9^P{P@{;ouJiSRu?FXzj00T7e&h&A<84=S1mwO6aRCLb_jqfhASqA!
zf%8Z%XJPFdT|c*S#_`Jc5f=H3wbpoV1<RWfIp(KiLkvp@PvZt9q{2`wlS&(sf+b~A
zg&8RdCCC_7BgIlK64thqh|!8AW=yqAr<}<M3fMbEh-Y9C5@MtUb$$lIor^RO5oI<{
zBf|@YwZMP)@BcY%*K-&qhW(zd^+J;zcI0Ig@u9QilnH5~7=yPfnywSZ0f`hA#^FRt
z64;k(XQp^3BvLE64Ax63XWM#W+cF&Y45uS;nMGKsl@tudNi=fP;8#5c$Z270Elz=|
znB%B_x31#%Dh94!yup?rGnh1S`u1lO6|q*+cuhm$<%=)S{gzLs6E{mF#KO((OP29O
z$|ANfzQr~jx@j>@kH;|Gf5UwIM8?szD^!MuPajD+aeey=V_R~ZSklb?aN_yZE#2mt
z`LyFSp7`+bYwnbzX<Fv##59ap-_rIgwDl~<M@-5r#|JQ$cC%$nfo1oJ<EIZ8=TJE^
zoQ}Ar=iR5zO!LBed&hcvhgO<pKJw;^A2BU6r*WiP-_T!Oqf+4Whi_>bk8=%S4#dOE
zr~Lz)n-?^0MJb8T_wQ-DYh1fw93Gg5BQKwSK@6H<9xfGhSQ1Oxp)|yCL~H2!UO+U(
zU{%2=3AiZ5;c9#J)$<<^HUsk{DT&rdG0`}OG4Mb9(?7&p$N&0&_kZL#jHK$XPBGv(
zw=k)T5TG<g<xBclN=XX0;e>NWTA<(Aw9DBUz*AeY9$w^8Pi|V3i(g!y!;3Dn|0)<e
zFP2Jk2GX3By)s5R%N%)fK~UG(OXonyn8sMq1=jgwPBW!MQZ5`14`5&7`)lD_R05>-
zj&8jnDNhQKWts?_2tgbsxj<M%)EBkGb}j4M=Th{gz_R~Hm`>!P1*3|<bCmPM5K*?n
zP-u!_o@cto%SMxeP%2k9m_}x<&M2_m_qkPn?;h4jVxs~|62O+y{t{IjYj9Cafip>x
zQac2tRE4+9=ks|(%$XF&O6tofC9DuRHKuw!<;!?Vo>O!LO3Q{p9gS(F$ywslsojN`
z6aVkO`1^m51bD_>=W0!}&PnvFwwLjnSnq1fTvZkab(Epb`k6|)th0D=6l9#$gjlcX
zGsgFFFR6>(Lf6eoYInssX1LBHFk0c8h7bnMgF9ywVVR}SuN(~#T}r@?Y+6uK76iMN
zFr_mH%{zBFO{?0XJ2(Ae%4iDfbuR%uiE6*hSR=N%l&X7%bGJgSjhr)9i%=jJSVCk8
z1!EL*l#n>9h2>z*xMV~hCEbXE(r2QglJ$^8QZ3pltWkLH2q9FAa(xe4*UweH^IRlF
z&}etjTxeYt`%1<qYwhsf%ik-)XK+m;n|fLp=b3ql=zayh$Lj{&dXz4h7%0@P(jxT3
ztTR@3G;PObv%xl9jWQOyp|uOkGLrUz6avZ^eA}RF7Lf0H0WoPAw<s$kXrs>J{sZRW
z;di@7zt0W5zyCnT%zCrs_VyWRI$})E?H7MYnD*@N-_i7r&Fu@C%_5xf>4@^zoaUK%
zS-9#uGT>0myB*{1ftVJS<Bmw6zq;di+VQaaB>w(oq_GyXBP|o&dRA*sTt>RCVHl4z
zZAYhDN;ra1IOm8q;GM1hZq3cjYmUR7!+4N(+QWNvikyZM$7#g5hOX_H=Y``Wice=0
zMrZa9?<I!WSbp`J-w;w1m1tI!<&iQS@kB_Lr9^UxXythG<)2ghirijRH08kkyWbE~
zrtvM_TJ{egp~y*TNfFmy(QF!q<Bm`7-;%<DCUdp9qHQ`*MeIGc;o)%4em8KnwmiFe
zNw;}M)3oHA@ZDO3m&1-Qi;vz|&yQa{=b!xXkNJQ8%fFE?wG@F470}kyom)M14NBD%
zXd^Vlrj>S8j+o1d{yc+3XpQHVb`=p;0rZpiyMkd9Phvq(mw&wQ2QFel5gb0b9V(B!
zsDIG4IFB4Ik9rf!OUbBWP{#7`xZ{_<{5gO8<Da0k!5NQf8gec;@5m*f(n9AAQ%EQx
z<KqL%c%WThqq`ob6-%74TCp4tgb*0p9_>0--hr~XZX*eL2yc8Tig6C;oN;QV<b+Ys
zI!jt6O49-c<qiGK3ly2%r+1Y3#O^SBpIi0!?r}cdhOiK(5k*EBFAMM@IUT~o;L_et
zVY#sJZFb~5Nr75Ggn61st~SBC!68MUh+Jz+`g{u1a=k3eP)~<ekk8swx~aN!5Ga{X
zpFeZ|=>bx#ngA&ttdos2=On&xr79By5f0{9-$$tcvA_kEr>X-cor{(8bzdaivdCC0
zRV$&Yx(TD7#wHZ;J(P6L|3I`nmjW8PirBSISSvY3=4r+@1(gL=TuP+zMr_O0qOE0_
z=jzT-7h&yL2LZxCIqMsAou`K+ok9T&359TfoX9!joyA+x6%;Kz097>&DP}a_jVR+|
zI(GvUs6t4YApc0L+66)tJ=fQyk@1l+R-6DzVVu8whZG5HlE0(Lx}dXM2cr6S_5Dfa
zgjR)^6D29Eaui^*S@Yt>bCmJurPPq1Dz?-LZ8|xX&jY822X@C1b#={+X=$1jzI7LI
z_E}WzoX6L~PL+alUJ8($AXlA>u_zqA@ic9NN|Ay<l9X-5W@yjJmJyA?YVf|nx(45`
zY5Fz!cViBHzhWSq4&2?{V*Q3`p2Uxy6=^#1`SVAn;e^qe$FKh$z1m7;%gpfjj#saL
z!o%^3IGs429x0C>@vdQYbxXh6u$0W<!`I}Txw?5t(U$SyJ-gE*^E|Q;$;PwluPMVl
zVVM~BkF>Tj<#XV8JfM`rH!ZOoXq08CLa$BV^X%r1Vq3;(<aqy<Fz#7-k5QThO<pEm
zzI?@cb4BYcONbl~dmi^YP(7>76{@zST(iPChhqZoNybx@A(V_$2IE^K13VIPB4mwj
zuec&H>^`y#JKSmwL8FK`tI@WDvM?R?c<Xrc{G~*_Dus8xMkULRskNtFZFu$ClIH^y
zh+!m#!v5hSZrIbUuPHHd7)EkRZ2K#kT=-x77ypc}zx~L!pYECKhJS9rpG|}pCBbFB
z`J@}Lt|2UE+n>IS22fl`VNdM<YUN&a0CJ=dagzBio8YN4pw1nNy2zdX>^w%0Jwr}(
z!f02Q*|OaA^*L*ya4r~1B{jyJS&sLRUi0?d2lBhGc=76oG;K#Vk}`HY?l|7R=k3Fu
z$2sxk^E+PLzQFrdx)Qn2_FHK?E(^v^5_~kzJPh|Vwqvus!<4`@O;`*?YrHp7j4qLw
zGi~3KQm6ttBUs(2SoR;}bBlpeG|RaEKDX-c;X^i-n1taWr&5U_&M8%gMv?A`vN-E%
zpp2kfQ(PcL(3NX%B&1B0B+sD~#<pOa@3hxt?wd6jt5B8JnW9)@H8aL|loc&Qsp#E;
zTz9|t<*%8Bp|-#4$=(>z^_(GIB}@2PijuU7xK~h5`4yyCvrEpPuXwrEf!cD{G);pE
zGlFHRg1GaaUyA0ZzjsbBy%gQ`&)dddaN^cEtgUdMEIOF;eB602c1!25*2y^K{4Ax~
z>&_0H+I^6YqpqT1b$LC`34kqy$h<^QineL6R;+W*S<+IoXH-4SQ;T>bn{&!yhcsHz
zd0WQ?xgI9eWsNx(_*(1APB<5v8EYF_*E5yG%D}Qr%#-jvoD-}y26EPyLy9v7*EAQQ
zWT~B{6y@`Hes@FDc#cCr;fNuSv$zB7dCo&3E+f-$VA$;mV<4*rRU-4Lr)j(#f>M->
z2k&bx2{M^Vk?Hg(`U%(KH(R`e90OsVIY(@#7?{U|u@Q;ZF{pe6r9*iS#^Jofdxy&&
z?=17U`<?F4?^O)kUSA26K`T~m!+0DiapwM;U(z)VuIV5ZLO5}C`x4t-(Qj^f_~vK4
z|LX5~{^g&+_Kwf*ek%%2r|G?7IDW(!#p?Q&G!N`Pe1*Gu0lsJT><j$n8m~R4X&|Zv
z+igj4VgL9*?-cvv1GZ_Ih5_$83O(BPjMISC5#q#j7_iNnW_^cpE#2Lh^zDi~?6}%C
zEI5|a1G!}G_Y;lItgp7TSJ%8+tuW4Ge1~(MWttgJM=}Z3Y`|Z!T5VZQJJj)>#y5fi
zO(XkHpLzeAnY))?NYO&(`s#{hI-zyK5NO&pAgWJg99^U3;wc4Yv6qz^7_uxAA3uKJ
z>T1inX>k1&s#~+fbh+@gG8m;9Pbajt`2GrGz_mTs)!AX4$N7esYRs@c&-LJPJS%e2
zmSCsIG7Y3KR~wPKc!uSGQxU>vy})Ijf1ZpA4Xo6;;<-I~UQX1Rs(4Wwt0%p{dEu;G
z3UvMlS_9M>n4~W4OPSlAi;Rj@t=$Nv6#w<V_?P@g|G~e<%jaKUR3W92JZHi@G6T67
z+~y5``o+J4>l&tMq^-e4Ax2R%t5mrVfe>bf;ly$O$or4)c=qBAKQfva3&vY=DY(@d
z?;3{bfL(QL);G-iPo(JpT0*D7GP3G>riTwq^CSsszQeokhXnb3do<2g%Np3Gz88gj
zA6k^$s;ao1YbyxHk-|iZv2M~86KfnwHz;Fd+!H5A5{)e4L2($Ip)=IY)o6u9NTV?#
z>5(}~qG+kWsS0vQIq~b?ek&rtl%?ZUN+n*(_~Ge#=d!^HYv0sDG^3Q{cc`;?MpbuC
zB~_Lp2zKi%J_AhB?l%_W3{OH@>c)Ke=*0R5HS^$12$fjfloG9T#5l?LDGQ4~C-MIq
zBPoyC7#i<{W}0*52&mePP%KNR^L<mhBh}3io@|WYT{o5>r|`|XhdR2@`V2jLDv-3+
zlEjMKr>T-Qjh4|6sG24!-`$y{Sj&R%6aexZIYdHCT<MlJE0i`^tx--k_OL91Q?^ea
zv-22$I)1Ru19DmqLE@%k{ajK?%=63=XPTy^X{C)H3)t$)Dz?LIpV6BYQ-WzY@af&m
z*!T3S4r7ft7AhBmB4KwbOF<dtQN)a{A;pE@JTWZHlfd2TjDZ-%y`Z&>QigCOj-IAn
z;oF`ZJ7Q>s))*s(e+#LFu_VSY@%i13^=eIy5v^bucZAg7+BM3y%yZ^9Uw_5xFMq;n
zbp;q4h2il%eZS@P%O8R<Ow*Ai3{2BZOp&JTXjf~J@~m#((5^S+d0>3}$aH*UJ{_1R
z%jW7Ow(Uvdk>R+*s=~YXZ&6xO=(f)@S69!Om!o{~eodNAgfXIhpy*6kW^zcxCDB;$
zn+?-E5|_dpCq6uUi~HzVw+(mC?&!C-*qaw<r5WCRjaHdv`vr$(;<Wp~2DoZlT73gW
z#5TGVUf*sAi=}l7Xw9r5vA{4#uujB=5y@0v4!EYlS;J|rw#<Ib5;Bdp%!dcU>A>mp
zd&c_@n3pdJ^FmCCre9%7q1BS7V0=qa)rOM=nVG^uoG1S3um6U(-+qg-4bHcy+7^^e
z_X+xg)?$T8iIgOgI)_D~HY;!avN)-W+y^cV%X2<@zSvvDI;K$O=|lsic71Crw*14B
z5vWVC!lmv1WP7UB#M3It)4yNrhbT}IbgH!G?dJo({OT*-{NPV1VMc6TnK%s;Eg0j*
zI^uldbb71~k!uc*pCyzgix=GZD`pb5?O6K;+x6W3;HNaLqjg>uF|7!T&;&y&^!*i|
zK7C@ECRCUpN1SoE)fL^<wiZhTx6Z^OX=pv}``oI(cMq)#Mi-LGphW$i>lDb;EC@+u
zGeRaT1*1F~iEq7l;<Z9+UpoUDaAfBwDO|JxDMe7JTI^(ma7J+}4c=JUFp;)xLPiS4
zw1P<18Zx~9_{cdTJ5`%q$@T*2sU@ovqg69-M)9i48$TDd>av+>V<o&s3(sC_SdyGX
ztuZJSIA?~O3)^#n{ghhxL={aHd{pY`xfE$-pSuaFQWeh{o)iPiG~t`(k^!*HGi@UT
zKI&K^27#Y>--ym5CqV|+;z-1WHDJnUVegkh(=|Bnn1=Dv5iCW7)k`JmIcqt_fKghC
zYh+xelrR^Tn9v$*#R{h=&wX|VoV5d~IzU1d7T#?gUEk3t36V=#X7tv%3w_WS*@Qz7
zfRt1ABFwbT3lOM|b7IOk=NTtS>+SoFw(V#d>9{D3=t8s<5p7niV<g5P@T;(3jg|LN
zG$kjNIf~W=s^dw<w5k-W2J%cgMUs~9TBn378ErC7YgE@!l&ytyLXl8mA`^*ahHxO&
zYe5w#<KcI@N55ZcVMyFvcl4`s3tYaurj=P|ySZh`1!HIQH1O>=zvAv{gSVQbdX~dy
z;^{L@y8$KSFtd3Eb7J@YE8O9pmv4T|5@G)I4!8S6+i%dWMKvuZ1dQvLmx-10^y@Xg
zGn6<HmW5Ibo6Zv+-?BSCV%wJX>KP$NP?=?TL=(`yfiMfwv~8KQY+Bv*4ddYfcZxi_
zd&TP68)6tZogP`jZ&}R)?bS<cx8^iYXluCLzQU^v$B!)Gi1!^i%*?}%9Aaf@NA}~9
zF-H1zhc^zb60umCw#Tlv#N#8&up_4l+jS^o2w_0Qk=C{nrQCE_Zux_sG)&_Ww|&9$
zFMq-^4lL7%QH2~PV#=fpVV=k(;f!ZE9f&F8o#x~2fNEDLZE>~T4t3|%>awAp7qOfY
zX%56i;3d{cqFSMD;H6w_G_n|~r#9(jj$i6tF3&H&N~~h_h#}pEOKV>QJx@-7^X7j3
z*Qbud8OeG{RIGphQb0WYC22Jq%NR0Wz5B$=FEhHo#*&!!pBaWDP1EB0E8=j3Qs|n7
z7)Fl!2b^~(=NS%rY}cZ?E9Rh>LgX|IG;YmpyT+EI__nPFqiDQEJ3-HeX`ye7d|!S`
zw^|dY6Iv_AajadNPG+H5AUiSt<IW~XaUmxIIY~R7v)?{|a!(sw%3{3*@l$g?!AbF1
zawe9DGq%E1G{!V2+fhoCjT$PhJY$XEVT-CEIrVZXCR4P4g;Xw}B}m5@Td%=VH4btQ
z$DC*yPd?i}>$P_F)t_JExog0=C_ZD9&w#7*wWzeDtQN!>6q9)H%UR>_RM!GhRGjal
zT3DQ2K<DtH%a3Y}Z`+II&RARB20^+3Ig>&l)cYnZi_FgJ1`pJe^wUT|QP9qbJ}9Nj
ztXkF0wPcxFJMSd4sh+I0l29UQ;ozOExia$IwBDkXSZ%%2HT<aHt;8FjL*qh}u0t*m
zqu3j*(V{CzS?-T48X@I{`=Gs}44H90Rl#4uYJ;_MAX)U8wx})|i1MA0P*xHc^yh#2
zhiGNEx?0n;YrzORdl{3+CzdHnz*Q<sOm$icDV}>ss%loKR_r7WkCHG-P|?<ieH~rM
zK}7n1xChQ_rWKZt)<!zFSp?CP`gxTk_fSfgPEv`}?{trTuVSEEZ|VghA8LpL3WX^V
zWfayHe1hq4q}PhWVaH+GalP4ay}jYn<3d!1-nGP-SVCg?^p0U1xV~BgD-Qb;-ir!0
zYljL)oYu?-*=UPWEW?4hBw~sz%K!wDv5dpS?$gM6)e`a%JC3;hfp)#YJ84Ib!ycDb
z*rr7n$KiOSZ4A@?o;aNtKEJ2m-b(g$I<UIBVYZI>@PK;!#BexZn-$akf%y3?^XiHa
zCYHkv@1>=$tt?a}1)^58UC(qp$>JABVhSwtOlD?^BaBzr%CJu%GRAtbgd_9*QD%Bh
zPVvrZd}Fbu!<LEtH$S6j8Kf%TQo=--1Fl_zbD$fXYbpJe1Z2dK*FX3X>+bJ3PIGOI
zO7$)6Gdbm-7hYKqXQuI3Z&bPbW6C%yGuToJwjRLG9Rc;PEly_!d@l6{ufMBgb@`o_
zu0pl^C}sY^ig<o~P;~{hGjOqA0cZT|6U$!O#OFD(2nmHb{(t}Hf5o5w^r!gEGmO>5
znAqIBCgseqd*I{e4>X$_Otv(uHI47_>nq&)2Gu_D{{3%Jr>|LcD^fi1X1n2Z+F^?X
zt7M(gX2EIk!lxxN9CsW~M|``Y?;4J~dmcW1KxF|;Z8ldpZ5SRu(Rwc%o64A{Oqzd`
zk7Yj6x>D8gSp;=SVq7r_iKDQPA|Xx~Eu*e;Ay-PN;36|GIsgFi3Q0skRH#S=T%e7+
zoZ7Rhi+b%U6jik=XB~+GP(r+`Vmxf76tULjO6bfv5yQaW{_NM>f8J4YA)f7q<vhkZ
zlK?AwLn(9>!(PU0Afu#<en$8PXB(URzoOK3xuW{?DfPs!p04Ffe+-HC&fjVKR#D>l
zBx|kaYTaLKh?JV#r?u#GYS$?&5|=*Di?CO;#>GS|GWQ4M<g4mopep{AHW*u~NB&GT
zwS*Mpy%mTdQV@IMJk9c6i*ygXGlY;hkMy)ww2cV$axS=<h-i(>^tGxDph^pjF;TR<
zrmPjN78b7NK!jzYOf$5NyoZ`|l9N~*wGoHTIqrIy10h7Lx9D8g!4xkeqnso2vY?dY
zNUVB~byX9>SuvG4J4lo+RSP8+$EY*s;Z6q0dEsfiA)``_)%ucIQc5OVU{gXb1x-;J
zZK$+PTR*e>1m!wklc~VDQ>LJl-;G}Sy@~<f^c1Tpst~ouuN<d`PmDRUzFPC~^Jki-
z#ppuQHt2H2vIGi>Wj@Kw`t=`BQegk_Tc+J-e*5d+@a303fMRK!0}JD{M{9>tnfdrg
zEQR0x>bE?5^#;>4pkX{d5KG3UgfTE2_ZZjV+co%x>ERuwB$DrN7EY&3h>32!VO|#Y
zyL*D}h(MeV7#`7Sq3FzTxM!Ibu3mnDGA)hw=*<=TZ-0#~3D@*2)50*%l*7n4P4o_w
zm6qgiIH0v-ng#)K>~=EyN{K^>SZ6R+f=tHKiTUsu<62IKBcZ_6%`N-G$P_Y-nrNH^
zJGE`g5~Xr>M!*!O{UfXG7QepX{^PeaO^<F?s9~Trpi}0k4c*O4;xe%N^p>A~@rwV|
z|NOt;Z@>D)zx>O;Vz=ALjMdnu>7NoAQ%*8o2q{)|w%C7C%2+S0Ho>N9VhXjwK6eAM
zEWGD7{+YgZIq{zt$!hVbsL!()2rZ(kOF>X{o!gqaP@YfyvOuf4;EO8#iBNcMvr@q<
z^%wB&(*wKxBQMum#$kj!k;2oIHk*d8Ki+eFtGG&$mp9ie^F+>{*nfP_`>%h^?WSjy
zXRcJihs?Bp%VAfzx_yQ99_E?H&!1T45$`M^WxDQ4XqWp((!8fJhQ8adKONbB{y^$`
zVhEf{BrXfncw)+DBjrEN$1;x$D%LiyBo?BwjEkzM(r8)W)Oi!Dv+wiwD1xpkXiX92
z0$N#;I!F8Jr_J`PFM-R(SSX|{HZ`pb7=gp+v=GZe3KNhZFO0)b1$D9!TPL&h^Cl*0
z^2(eyAmKtrbPctrJqG}tlNQjLvx7#Qk@C7z?P8?#WU0Iq09sdVLVoIa$>9BRU+K%W
z4&GZt$hTZFI-GO3t|2TE)h?)Gi&9CzL{x2nyu`}skn!UA^E1{-INf;*%x4#m2=ZJb
z&JcAu2&nqm7zvzGqPK8XiUDnu#KJ2nfQ&JeT0~rM<BCfJajv-nN(osJb?y=gTR_H2
znqeBHb1^Nb#*<ThzEZ^8S>T+Ur~#wJ+Ph3kJ=oR}=lPTmq$t;(Q5i$ToY$}dN)^#M
z)$hIfSt<q{tWo3~>US;0lRmFiDnD7bi>U>#sRcrQTF+Fe*JTo&nv%nmn5-f>&yvL=
zUr4>qLw$XU{BNTe2-AsScN8C|>v6u7n=dWwKYu_e#WWqn8li=syfPKIPzq|9INX0A
zmclZg7!HrzTyMy6qS%hd!%TPkie=RkV?qy~&}l;D#H$~CiE)m3o~6wa0?J!<$0O@a
zN87D2wx_#(hUwO5+cS-ieE#+=kB<lLo;^cp!|}LhsXlz2Mq1x-ItonaaN2?Mq-kOv
zM)K!R0PjE$zYN2P>99Xz?0MWBxw^U%&a~4QRp^Xds96c^ZdpdOPZVeQ{P2kPUJ@2}
zj~G*G4!0%e!uD!SSR%XQx018ac9cc%%OCF_xZbQtUBfX&p55I^b!jd0GBb`dIV)P@
z*&iQRmXXi*_bef?-Cko=!5K~4_bkVIhIy{<f8_PuOaAyz|8xE)|L&jg&;R-VhuJp#
z_0N9|+7NQ#{`0*+V4_r*IhPueq9k0Y2+Mp~7TT`Ac%e%vIAtp<T-BBsa*L>3t8_`D
ztJiKl(LWUjRb^DKy87SFC#M>sa#0bLClaANA>Uc3&h0tbBvj=hoC}CjGQa(B&kug`
zXSAyu!r_s|dzN{?KYz~l<s0HSl9z$Q=eK<R%|y3-A??7!o@c8SX$tIyBb1EQp5b%?
zWf;Q;n(oN{(<iJmG+o1f8qf-!y>9sQ_5;~?*6S-MiIgKwYeG&8r#*eQVxBX{;|Z|1
zZvB03)!(~^G1A=%A)u7uoH<V}q-6q~&;==y)pvkcb@MAl1b}A&TPcYg=6br9)0Au$
z0yj!o1ch>_up-Vy5mF@OAbJGrA&V8!RWOW3V~nJSQe*+wMqgU(GB(RKnmP+zu#VRX
z=dM%cqMoUv4+@fbuVu42w~NJnQ-~?aX+1`?E|fDJpqx?O<zKt5xpX_uOb=bfiPOB`
zoRe!_Yq10yu^5WaJ_-Joj4p~8X7TLTTTk)i=hs@}yu&$n8I6FPM8y%L@vbGsC>CWU
zHcQU08#(dkl4zVD=8bvsx-Cn@7!mTBb0oC_TZMixB@A--u|^X@BGzJGD-j;rtO!}$
z2y>X(Eh9JUD|G8%F~np1PTcXi$dZ5~t*N>yDUvRc$wfd)eb@8sdP7PRC5;#*;YuZo
zo<iyp`P?rV8s|A^j!+`xAYpVV)-H~z%UtcOWVxPW$ztJFx`ub14?^<VQd(-3O(0oc
z(+UNml*(K$3C($J11-Ybe>iaZ|NHoU5yt(N#(To?5wna!N@^Okm#R}~#XO3l{p|NQ
z-V(Cnv^%kch5c{7X8ZazHb+<jcegi0qlm?@-#wsLTcIe;C-Py3DFNR!#0>NPk)$+j
z+p|0DMX=#4(;_Ky>rKxo71l&HuYSl@{|J_am==n$WNmRxgYg|2g`qGUPaGfjV#R1%
z=HrQ&B5mI=&qrbyNg-1zVQLAPc?oQ{E3|RsFk!t2j>2gV{e~q*Oa|o|HZPwOjssd1
za_>+jiDx_|;ymE$4DIS_P1`ga=FBgD^-DI5;cnAmI*)UXY%PazqV0N&F~ntN4w<)a
zKd>8TdMEFvQ3l&+(lVhN#}EQ<|Nbpq@4&Yl4kP1l-r>#9c=hGiY&ISL<Nx5F(!KbS
z|LVV5vmEZ3isP?;@k^Fz;Bfzu{fBRO+&%EuzxayX{={isYBXmNK~O8go-<S6oMI?J
zAObd$kwBd#s0vQ0$XW8a{a>WBrJv@n=hmjya(aA9;wVqG!+Gf~MJ57#M>B9P2+m@v
z?-U(51-|<F8{%-_usovUz<AP9RZjzCScW4I*$pFW<<R;w4~HX<k2}`aTZ~qutXL9I
zBFbOONv!qcIJ4wHQ7ygS(A$QP0>Am~ujty2n-@0(x8^uU?%%!RxPPo@*vQRxL+d>6
zzx|f=`UdxpJHeQyzoy^x#3*g^lp=Csegadayoh&RNlt~#c{5r&IYE{PMPr;q!^Sw1
zL!>B)Q!k~)Y**y*8QCjNg<8$WEbaqcuu3qkr6h_*X0WPocAT7DFUDKJ5Ld8xj*%R{
zW5WZU0KFnulS1UIb19S<Wl^?XTK$|0*(Y6$b(XF`St3IyLgA|;G?a*LOD!Irisf1`
zlq?<CWg#agSsPs^r_}1>KNk_Dw%fHTSf^PkvN*;_2$80#_~aPr`c7hv>zhk4u*?%6
z>$`1yMF@lQUhe;-Nij0dv#9Y=7a?G-=-(v-mXI*U;OhHNISCwSSuRecr`IgNDPuKD
z4CvaR7bus;YQ+>1b4a}V^Z`X-MPizd7aOE=7UwM)Imk$6g+ghEQ%bDCF`&=iNdfOH
zt%p{FB@pL5y0P{1FO;m#jVMYv(;CmHXln5x-<5!j>R7gZZ|6k8Iv9HDVsb9}WG#(x
zA}*t>TqZ(paZQI-t|qnTi=g!UceT+{9Dj$<`rr5XUd6!YufO4kKm7^&u)s91zki_Z
zJG!o8jESag>94M6uC6E{*C<l)rP3;z>*sj)iB;bcrXyzEa9n1JYtbs8Ees#uF(nD4
zXxf$-V2%l`3~qbJ?aecmVW6EBtTiMO5%GEthdr*({PGunMdv+MTbxm-I5Ot|#xWf-
zZRchFr4(V@)Aycb8tJdLs4GKw+_9v>FfWXUk-KNlSZz1Nn0WU5PCye{nDc3wFs1N#
zT4-92E(0nTx^>I`G@wJk8pX#?pIG%h%ba+&-q38X(ET-if5S5GNb`Uyncl*5JTQ$X
z)~gMxuE#VoJGQ=Mj)~LfN1Rr4o#h8#JV!MRUR&}qa!jKLe+;z#3ad1m=Z=rxe8m@k
z@Pe!BXAH+9A9s=e_HVzT%n$sh|LK3oloEGe{vl<aFmX>iKBA5fbiS~@Zh5h~<0mhl
z@vE;t^7Y3(rfJC}-o5|C=6Xv=k^9g0=*#g!3AwAB|FwdHzH@<>3njDE1N(V_J5OHE
zLaBTK4yihqEjdf4qU5K)QlI0DjC~RgN;_5;>e%ADi)mpQj_h_vQW){J!CHm0fm8~%
z@ucGD%?e#Ir^g5O#{<4?p`IR;U$MU3aGYkkw&&-6{Z};J^748^*EDQzwxkd^?H};#
zYgX%Lr2ReP{U_YEFz=7tzx#%gX2w{UQerM}yWMj8>J>30hTl;U{QLT_)}xfc`G!)Y
zp)Oz8QcvKKg2Z&6H=dj$N)<F-7CU4<ijwSsxQvX`32j@9?<<EwIt4~qFvy&qb1R*y
z-cIB^lN42Hug{aF8rrEOD<Nl$v$Dx)%~P&|EWEC1YwMhgMn#0PGS|;RXo~08e4a%s
z<DU9m_4nKQJ4!uehMb5YlaZGCg}kYsTJ%r%%UP)-m;4eVK1#7JwoN06i~8*N$xO?F
zQ5Mrm;+=DbSThc&#e%AK)|3)d%U}pWKwaMByd<oqC>y+WjuaM*(Ul^YX_|&%nrT{J
z-+wLK&+9|YN>Ex6LS|06YAwVH>8!zMaU7%|Ns-n_DB5|IC=|Ta!fMfyAh_FoCd@}E
z@({7IbAo9;PYQENvW5W4*m})Xa;cKT&1ThdweGmO>d{6{8aYwNEa#3?tOZ~lOVtt8
zWg}J^>&&IoVw8*y&hgS(8?>?&Tm7AkjQX8R^m33xiV{nss-FwBE0mJppMh#G6$ec!
z=@QcPZ@fo;r=O}1?;FDDfY*gdW!iOz)tX`dNY^wqKReJij{Oh`A@J_&-?H6oP<g`j
zTa2?zr#<sLG7K}HKR>cwZ+U%pg$X0mapLZ$e}r$ZFkQ=VJn;VO-*UUX<>uuNpnz>7
zxi^q9(`lAdsVcnx^c$wQ@bZf{fa39S#}G2BMsa=9Q%WW-Bd6)eDb;zlRygliV&*i=
zSZ8_u@;Ud9N4C!6w8u7H+H56b8Yk^FD#p@UoNsyg@);@3D3#eS1LJri$H;NN!*AEz
zy?RZ~1>1}A-o${@4cTNu&cq=0JzUG`_J-wXY1iwD?FC&jttzz6;2Ot5Vhkg`ap?6L
z*WHqqz(!{pB563t$vZ>a_dNgNOS<iIa*p&R@yB00=b!wuKVsfLpyngfG_e2pmWTIm
z@%@V9=eHa_zvJrmj@Li<lE?kP&;RD{I2>pG=|BB3H#gV(`d9C{?izmlM}I`tMo_(7
z%hzw;)%jQ9SHJods`_>>E&93{pNWLj41+VU<-62CT~V{=w&NLsa^|KB-uNFbkQaeY
zt?<vxf?Q91b^Vl9qt=A^fGQE~T7h5fj_6WwZHxAqtTlHpe!$`L2Ts$7a)xo91O{a-
zIUR|oKt7x}6isgurWDe6LL1HJ_umpzVzpYKx(;I+KD_-3KaI4fBVE_icMVCccyW7&
zve0)OIlwYW#=-o%-Os<bk6Z$|&P9uI64gzDB(NnBnJSF!P{xu|poEF66t-DW<|Cz4
z+FFi;7%0k8vXZ1yD^yNtn<q-$P|8`mLQTI5GjTqOg-lD_u{NHp>Wm+W=Kbl9-*A1^
zb2tnbYh^T+L|CV_24``NWXqEiDMU#KJPTSY)vn~~ilG(^rPL0W;mr6)D^ZavMGJsp
z2nAJzc4sOf>O2`xm-|R5p6t1$P%w;Bpl>`W27KEw41>(Ptsy5vTxxd?(fmk~p?p4P
zdm<GBz09+q$Lp!yR6vh05{x6Qb#T6sQ9-2^j&q>tSG0{|iSj;t(=bnqXb|dsEt?kP
znChmVYG*N{w7qmO6f$}iy+KW6ywFW`9sfw^lhqn;4b-uuGZHZ6yr83(C2$@)8f_|1
zKth<NahB_%uDd_^;Tzi4qbu3ZIbE|f1T&q|LMcynY^Ahhx||O#wZlMlPRR`ad`>xM
zyHqEQDkviwjTlAOq3il(jHS3n#%rdAVpTEjnP&M9ekVK?uH~s?@Zb0NUd2F~J~Qnm
zxzX~1Hi~JONYjK)9fu{*v^{g|DVuBbIB<PeaDC6|<6G{3_BZs;UgI_$rfJ#SZE4mw
zD3y8l@e{xP`M0?3n$;LEZ9==2q60Z=cAq}c`i8!3nfD*ial+V!561(?(@5WVhGD?9
zo>kL=GQ?)2?_jm+S@#{M!vVBnNrezIy0OgD#NB#>TRHA`C$61kcz9s5+M?5pMllR0
z-XHfg8v4yz;22~Aig5}IG;|>|zW=~H9BDTVagKcc_5-(HenH>%GJEt!OjRMEd_&fb
z@%>wdX~L;2P16j{de+Zh(KQmanuZg{-91KY{MD_XE0;j0HPbjz{EFCK<L+*0bD#{L
zY1@_9qI^r26PxA4>h85L!*};MHIv`{4ZbLd3py?2QM20gw5y(f|KIy_?myh(uAZ}b
z@g-k&SG@j{|A5CiaC38unfLt3PrhOI_AB&i!(mAr_nGbW4L|wf2duBJa8|QSBRU6u
z_VZtGJRHOhl?&r2w95DQkG%hQkI{zXI9*OR!h$cCh3~tw)2ESX?Hs7H+8aQOYZuKS
zQV`_2@ssMzP$7O|3T^8-Ery%tKOp3pa5@p@h*pKJ?@?B>KOTr7;EknQuSF-3GI5?6
z4m-waq3t)s5c$!UKjiT^kb!OAu*?I8(}43Gq{!prj?LW*PRAqPd_1tZ0u>fIeIlI<
z+vl%29!}&Gn8v{IbR;aln?30F^>I2qV$D`4eJa+5dz9}dBSNSYP)ZmA!sZtxd{z#W
zqgoIoH8Ux!Z_!O5hLL5N<b<Vb5~w=2*`-^Ma+3LcT7*epR2?gT(Oz^4N~ndpXxes#
zF;8)}MxQ(9Rlw`24GbB1<Q(gMoa<ETJVUP2t4mIY1Y@b%Dn7diQWY60t!qYwL<m!b
zahzd4=Pp3)f}LB#=X*l#6U(wJbe$)sNLXgtrXeO%Q$|OKu*3{y;d<R;w89xpjxaBY
zTqJX*Bn?qUb_HnrR+RizH=(t`w!(avrwMB`!?=*Lbhg4GF~l)tR(&U2hnVE}yu|vN
zpBzMjI#$GVK6R@LWw1_t1k9s^T^d9B&ITxJ#yRF#+yIp=VRe-m$-y8jK}IJjVXgZI
zI*(~uaIVm_E!*oYt4&Y8Uf~;8CzR^a{g+MMTzC>nX;hYLFXaTJC#zzXle#!pi)dLW
zHS1!Q*Ww!_oNk6B1+G>aqb25CVu=f>EM$}GXyH2tp-cDmA}Jx|-_45jy^4Y1aKu^1
zsuNpY41t^yr+HyF40OIDo5+0n#6}x*$=rPLC5LgMd;W(0#h0|(Tf**<vYfbn^$L_@
zoJai4OSV7$Gj6{4DdF&dSq|jWiQ~t2Gz9Kme97_gBgq-Y;mG#YD~XewGW}g5jsw0?
zcmvDsGw(lq##lqYSy8;<)4NCBe)Toi+bx^z70sq)NuYYer}rPQTGOw4!Z0vQ0l$1i
zY0De~Dn+@_n+Anqo@Y$!na3FtIKYvJjiu|ZFip$xVaN81A5*4*-NUzZn=PhsXjSn2
z3Z*?I54`-*j~T}i!o+ac(RM3zx5hZh7*NKt-w%}b3j6E___oC?f#duUYYo2fe0=*g
z#Xn-UuW7oL&8i`_Z5@+n44K3Gfs_i}^VhVSJEr@$Fg{>XW(1615l@f!?Q^-&yyw{u
z{t$KdBZ|KQ*RkzhajoiwHSP&o(K*k@k01E**)?<T$uXkx3BMTB3~N>JCiCz8<Tc}I
zL!L*nv4mVurolBW)(TYQ*I&J5S|Z=PeaBbdeqauZy!ToZNEI)_oqtF*Q<ow^e9`Ac
z^=#vkiah3Aoe$Dn)T$7}g!c{Wt=Ka6505D8=vqg&oLH?a)^%L1Uvm2N4TpEXrJ#BI
z{E0NrTy3uCR~<`?U|LqM-cXi-u6#gc3A7oH_k_i8b^C(8@37{9_n$s-{pw4ezZQk`
z!{?7ECBm+}40NlGbO;!yxO(-RPxp`C=T`l_d*nEiD<v>j@KmmG&lJTpjFhwxnhED6
zf?R2ZDvp#MNnsI{xRwZQr3<dV!*#8k_Q#1OiE`c;IdvNZS8J>%#)YC2F)Og*5ifb6
zB!e;ys+z&1@D%%(b+)<)&Z90Rp;IM@0JaqAV5ImY(yA2Avu5F}^e;J+DUb>!Ptu>#
zQWTs+hg5ByONUrC!6cP&EgsGV@3{^Aua67lM^Y|KOJrLrL!e@Kturi>wAsmpIg0H~
z)vk!unkmPc3@8eIQ}Y#MlQvAlgt787io*J8Bb|U@AcjaThH09)+HNHyW}a}S7HG)*
zsT&I+N(fb!&WNfGB<DTDI9>3^IaS6%)nA+w3yUh3&sW?8^3VG0zDSuRi*Cp(CAj1&
zAF|%DRG^Y1k1q6m2gq1r30VU6x|Toui3r!vVL&NWq%w}26RC<rCEP{41TifaVxs63
z3L=_CGzQ3csP+U%s*rP#^xANSt5qi#3S~W7nHr@mx(!eywMcTIYbd&<s97w^DOGok
ztcho<=)dXl{UVI5rfD0}*}teP4~GX14?8y3R}|}Us}&(dZZ<u`?i1hs{a-LAMSFFN
zUu_8=Kag@H=D^`JP;|rnSHIx)`kHy12ycH$UIu0@RhqMkU@WQxj4sT}$Q&XMZ$C0G
zZ#jK_;MwbEmBP2d<cWUO^248e$v95<R=N<SXnyd=e}*oZ>9ofJoAot^X~FhuQE+OD
z>pFDXFz)a1TCr(b_Td02p}i%gOtapSoW&?7A7CjM6}Y~;VHqb}+fj;Uy}sk&>)+5C
z%WAt3CGisRtz$YZBqfBphp&GvqLtMx{SW>LPA%--{+zfhbay|*-+sY-_`vD@BSX&g
z>nmK>aQ{g*BvT^fBaNQWF))t{&Np}$8ILDAuesW+aX!)vd-TmSEZ3CB3FkDbzox{@
z<EQt86qw?I>aNiK2Ia3osf$PlG(JxxrztnDac#$sS1U|7Fq*=6IN@ADtA%VE=J7<D
zXTs?brity<mY^FJ8eaYJAJerfT4(t5^)HFj$e0XwS8Kjlw+yEf-+a2Kv6f$ccp#>P
zb(+WHB!CtrH=H%LE+&$0XN;DFuS&I}F03*`%QepcluG>Z7dM3YL^>TwC9z$tP&u(p
zdrlvI&FRyxvE7=vxg{2d>w0OQo=%h!dH3<e&9i4X<N5mAuUO?w?;XcaA0bD)wQPRy
z=RE7KdH2oVlViYJ&;GclU9EWd`sXy>kukjd(NEZ1ZTanQe@Wvz?DaFk=MUsXQVsu6
zKa>+-OjtN8{&QGJX{Hp#vdm~zn5RW7eOgF;T8jfj>b>eL$bc@SMf3)8N=?l30q-RG
z)oP0}UO+lj4>$tdvbN>ItVlB@Po$!dU`vZGg%svGvXzaz)Yh$JD#&SGA#MOwPh^xD
zSuA1z$yBImhd|xSGdW1BT>FNiB`B%nC!%JkzJA~`!#@jR1^-<xKQBkqGorUD@Iyp-
zPY8iRq47=K$OKa4y_4UcWW$}7c)@u$P8<kRNW@ea=8Q86Z!<>Ay%R<Wdb+*6Mk!64
z7t*p&fMs52+m?PMp=#C&{y8jBpl3}}QOJ`dd@2#v<(ib}8jnqxFilu%FRq1{<g;)_
z*Y1jz;g71g^inFAC)Msp)g{$XxU){E<b*bk)q2G+j96<4A#gekw2en8MdKYYNtbq-
zL+w(E;5dd!u*LdmB&q5@%a|mR(}GrV?P_yI9M?IIJ}14(|50N73eT<mSPwO|6;kr$
z_nZUXB!X2@Fl6O0rjsHIO<9hDUk(ve`A*X0zxnY!IRnB>9*zf5k*_v%-HMVFrV)~#
zikbDZ*QkDriiP9HM{cffL{XY&mOMbSqREE+H{Vife*1^7zhZrL%`PnD7|6RF(;P4w
zuAe`{IZN9-<No6}#M42>2!Lxe<2-`Vtb2zwo*YDh?%VESt?=Hl-|q!cIS#yh{tROb
zDbILQP()%1<S~IVth!#Rx{|qh^^*PW1S#RSSD-CcS=y_XCCrc#DkMxJ40>%fUE5HW
z2UeX%wYL;wvBsnPLLLK~yF1pe|A6uRSKNK^CHwuJ_WC8nk>PmHG|o&R5yFV>Tf#$R
zJ|6H+60l;<bn6X+_N@9I;~eX|7mT~l7%ib0Du4+Sr~Sy)-8JjZ@$lg{=-~tX=7!05
z7PDf#dPaPF2d<$rBIMCpGmU#pyCNxzN(;s{*c4IA0hUodYa<;X<9oXP3cng*8AT}5
zbi`Fhjsas@4(T4ZxusogC}@_CAK-d}u?=d8tXCa(&wfZQ;QgA1hkJhfXMfJ${N-Qr
zqvwTn*Rp9HQ;d`paZVFT;WwZ5IBQV3psitEB5yzMr8>?6wh>E}SQ=H4R_p)ZkKgb|
zZ*Iv;z;9l#KMkCACw32?g_M>(&aNr8#~HYP`x_3&6YI^Iu4!4_yd=cJhlfX0m>G-4
zo(@d+M;w`UbB&@f4M%)Qr2R*k6>o0&(I5T^FW&r!4{d`@6M1;#;kSRsC)e}&{(-AG
z@#58MzIgKmB^5qDJbs^B_4n@4tu`3v(M8}+k|wIEt3ZT>&N-ZMq^gVu%{-4#1Sczs
zeuZlqN-0dIL$&xtrez_<1&ScPvjG*Wn6Fwla|}4E<-}Uq`YA5NII}DgH08WU<;1dN
zsEMc4&0be0iqf^ntfzJnkDU`V%XhjOy2@QD2jL7s0^&TQi$Xarz$8F|&7!0pmMp|Q
zWf5PzetIpJ6<#UMSdTLSRBXP5(=gH2)^@5yI;{=9(WE4E@={<9iN5tXt1(5vBAYMD
zuvC^pt{YR%3$3$ku2!5!Bw?PbfK5*DzHyvIyHveIs)AqVrPGp;4u&yS;6!P`x9x@K
zXsu;lf~?(EiX|!hHQq_=bqYyF3)Wl;#h4N?Wr-BN6bPWS2r6@k=nU3*+Qu_Yb&L@c
z$2f_-@jM1VVyxvogeT!~)dZZy%F5`<fy#2iFPE*pMsycJ6Bpn+4wEOHhmyiUS2$Fz
z`Y|M9K^Z67Cv7nVRFuw@wjIXxC@mTrRiytClMr_$&C{aQzXi>Js$N>WXMO#QqBNF7
zt2Dk}VU=Ntfz!ihrfDW*=-0ROx6fH(<gj~W+0DH8@f(g0@951!43VW6wl~k|-u!@K
zx6G#<+Ijf+j;`I%wAZxjj()So_Z|E89jftl-jey|!$;gS5q6JE)?rp#!I$PNsZL7K
z`j*wp4a7xYFS&3y9&p|>ACBDJ-jJ=q^(&Sv2w@i!8jHSr#pWuY<~?CJU`(O(qGq;M
zRJf5uStjy4a~KAU)woVJY3~|xcY|K7Y34m~++j;1JbZ*WVYOk^uX*_N4UeC`;n|B9
zwAZ(k;Y9fKQ5Gqc=wG}d#!Q$-_Ky$Ty?(>GZ<)svz1hHYWZ8X2X-mJo25p$qg7%Kx
z!#%tE9d5lsJsx;`|BmT&V44c;W=q#t`kUwATa=lJX~xe3yx-%x4KYRh`ilARz?cHf
z`kHwjaK_@bmdtI_5OhP=2-Dvq8?f)!nEo37{53uYnHQ&-pc8pmSZ=S_y!?{Y^@gM!
z(=hQ*|LhNFuKr!}GNCEFe|TX3xTECAb>Bf0(bn3)pZ~`{WIUW0$BFB^Eiq(5NR&7u
zG_TCzIP=SQ_q@4X^OH9>#Chb^>*q{k<kQ0^{5sJ%hc^m+eT}gV<74Eoe`Is>lGXMG
z6=$ApH!_#?ElVu?=%;_g@x!;6GP8R2hHiDs{^47WpT1_jx#M^^FdXlB`#Iy=Eyg$a
z?JeW}k!8K6U9HK{fpLtd%**G`amMiY=^ej)|23NzugKc7JD$GJt@?ZSh+(1ep1!{(
zWf;c?N}ix-oR`y+bslRwoKhIqRb~Gp%X}hdNodP4QH*T3DM~(sDutK}Q;1mGump|4
zped+%!Inad3vn6A&eB=KIifd&nJ`Z*<B@SV$w^ihmQZT!c%92CE_iI~913;u)n5_}
z&yJQ{Dm<tP+_a+Pnuu2m0%c4U84Au<*!mDdx@cqMe=;RTQcXNG`d`BugA^a7>TBxQ
zYz{J}==z3vSuP@5!7a<rSZk#)tDE1s#ojiSf@TOAQzYdx#6(UJ?+nIhrg0=i*~nu`
zgb-we5(~)$$g)14vkOuZv~y{t!zBe_x%^(|ogjIgBZUF&JuzkGd8yqEC!Lt84KhY!
zs~}YXg*9KGh9BjU>&9Lx_(%lfIf!c|g@|iKn-Xg4e;6i=u~_S>UPLq^DFtF&(7Fne
z>v%{u?8?8eQcOuyu?Qcf;0lx+31LKOht>L`ZIaH0DF2Ny6}>E@(3k_M)O-!=Ft!zx
za)`{+i4Y@NH>jkThn?6&ts*K@1&;E3V9@&CLNV~ww;y?V_l(;YFKIU$Q975z5;DHO
z!nZA_5AR6xjLnIx4C}ksq<%wu+;bWhy6u|dGT~NhY_v>=Bgr|+^))#vrgy(Y5or4!
zx`y_uL+gcn960Php}obpj{RXz7zdhuMc4FLV>pzER~|Jbvd*OXMGzCFB$*vfCyH^X
zwkHlFO|xO13rn$>q8UGaVp<mZwk0b|clVN<Gkx1K`3+-!BrY>;Fqq*Ht)k53tR)-6
zw2WA->1=~FivEi~!ri_R0b87j_Y<jTVga3Ju%0}QXcW#^lyx}Y6LVr^EcB5)&rHjL
z_brBr$r!HR{Fv_g2DJ<{)?mFSEhDAf61(S2IuVYavBu)8#kiJK3Z^KEb(GR^d^oV)
zZt;z!X<Ag67{?<Sg<FH|8%l)rX3gpFNS=>4Q)qpIcM8IYnkQBso?kyB*@lk~J0hT*
zA(ld3PB@=vnhxK1bP19ks8{%+nG<3unoqRW<F&?G#W3txN`M^scYpdLrsWG+>@)<W
zIXyfw4HLsKuzqn(et*wc9y#ePxzo5!#+S&vANa+;`Z>W`T5ZTv<o;n`(|W#nf1<hW
z$=!<UXRm2iYj&SLkr`Os-lDB$vpVp1fAin*dEN8ohd+R3Mch9!e7<M>>^Z03{){}1
zBoq%H{ySbgd&Y<Y-BYx|H!Er3DZ|6#Cv?~GIL<7?p55R64azlKZ&!Tr<3Hv0hkwGy
zU;Zsg!RqFk!*K$wF)Wn%_<e5G-@C`@;R7ZwIMdRswj6eWFdnONI$~WTp)y55&NNyq
zc-DDhm{4f6_82FiD>w_vG^JRKX_)5&lmWBp(Z!NWVmi&(BD3jv7)iQRvaOa@uF~XG
zNLiDUblj9q{Onhsd3ZcwjFS^(Or#jF&XTnvv(zG?3j1=tY_g?1(dN!G{_|N;z+9>p
z6A9h8izx9-XFO*ONT5!+Y^tJ#cyi>NRrhBr?GyP=_#ehNrg5h4TCpb@C$YpaVZ6iD
z*B6s)%urF`s$fjUCMDV)#bx6H&1nqe9J$$cq@2WsWwjJ+6oQiOK&e7Q5$g$d`5d(z
zD{nw+jWU+T3frKRLf5wp!*t0pNEJgZudQY`P-wg*o@19w?MgUH%(-R<2$LYCjCZCI
zEFpx%G%a)#taFURNXkee?GjNc<GqA7c_*%cl0(%=Xs{K0l@i|tq-vwF#n-}5?0%Kk
zjw%&dT_nahmjcEL7Flc2PmsxJma&8`wG;L<64JFp0trnZEsvxWDRCy-2CW<ljq@vv
z)s>2QCbRyVAK!0GU>t?aG@S@x#<d;u63N++lA+LG{hIC7HE|jkcY8kn_UE{_hRGPh
zb_Jgv2*(qhw<u?DwjqarTW03dk<&6!Boqs`zF|}bXL}BxzvXcM9<0T8YtkHOtYN!d
zvr?Mz<9klWNLDjZEp)4%GzZ4rft(laUOeOa%?}xd17SK+Y>R1H5}sHL%Mw^wkLf+e
z!2a<w&bO2>;<^nddqrG5gUvNcS(fEQJAR^Z7H32iS_-snOXDpmNFrC;HdrIobO<N1
zmJIut0^>B|e23{e!gyl6-D27`Sc{DXlp!mK^F-Tsw3{o!XlM*H*S9inG>X(TIO`b>
zJHj$D&L0@aw5?^;cX%a>gS8HbkYMxdM_j-7F^z+(=hwu;iO=7@$2mt?A~r>dP;DH?
z$B*b-GuvG!oBVJ>qp(dY(bDCF))UKkqBJYI^$j^jp;Q`6j*&7QDJIdjYqYk^xzO}I
z%6bBkVm!Ou2Mo~Pz7o~=GLiMnYSp8y<9>fYWktVQadUSk0>rN25C61h**~%jBOiAI
zFJFGaxcf|D=Jk($N{pJ*JTe|0*=|~zlCfGkM!)&yYZ@nsg4%Y(@x=XykF5KSeSn|-
z(H}Br7#>e-)+?Iz6|41{)8iw#Y0yq{e~=E$&%b%g7k~Kgu<qB)+AuF4@jg<fz;GDY
z??ps+yJ`6ApZ|>Y?Q<+Ic>n%e?q0m6-&}G0^p>`9WTnAa`o1UpBTp?<O42BYVTUq`
z5@y1@0Fh-X6h=@M-)|r%+L-X(qfuz(s)AikdB!;AX@VlwIOiPe^$qi~uq?A|I%t;T
zkrDz9!)n!`wd1(|%p4cNc591s9<&7|SqBvPJ-dA%#R8RWkYcR10D%r^W9r;Elar7G
zg+(AI;ZpM1aL#us0<o6W=THSwnLufbbAtaaBK|th|1S*toMj_8E5)l=E}wM@^|yRy
zfWWdua$^M=R0_sfK%=e3m_#v&;>7wmEs>_Ncqc`cwTcFb=oNiLrYJdKhm^2daXbnQ
zHs*{Dno<_}#$k-DGk?Q8&2^srbP_&~O{}TVAEhx~=JhE>RL*pL%d8LrDyG^e1p<`o
z=uasL63V#}O{<vj>=Lp@lZs(pLd7&oF(w>@#KXgpzH2E!iiwb9Jas0Tw%*b;hG{tA
zY=Rug1ze4?Cnvd<je)XI>a}x8r>r7pYo*&D;X;Wi&Qe&K6{T3Q0G33Gvw($i6z+j?
zfFY`dT=wWi3y>*AERga@DGS~@lA0+cVT{LkPfUTZh%I><ez&gT_p=tJh-(_U)fLOK
zFpM+ZYD2qmw5uz~39AdoB{5G6>&-RYcEz+8TUIEZIV-xZ<B%0)%v^PbRRd)?p)g#(
ze1R>B`LyTVx8I_b!K}6@<Cxus<Hxr+b&qk5m)jL-_km`$LSL;}-8|<w9EszJI4xM`
z*!<)<;dDf+M4k`4xLI>N6!v4FXsMoFJijH*1C-2onwiH5<$7pWC}+rN!Q>s~_(-%3
z-D(3l(l>@M9MP_-f-CU}fM!K5na8)kz}m0atk%pap^Tj3=hMXLG~jzpnnyG;b91(n
z)?vxQ{r3iC9caTmO$?8ZZ2K0ooH!j$<O1z_LkN+1IMSKSt7ls%g^&su%{(62ANN@6
z*seEdW9iqT><<rbG24#rYRx<^45ty*dR*JFEGO2j12k3{;u5jl3Z)dwyr8uvqG{GQ
z#4xe{{9b@0O=5ibC_oltQMLhZDXPFc5U`LGUDp$Hrlc9ww9HGuZLT5ZO8Pb=tcWsi
zdW`kFezjp)M&^T}Xh#N`GSNBDymI{VU;QPiSjOjfT>a<=#QS@C-!sOM?&^(<7fNE~
zJo7wLQsAHb$sdrE#+kr3?>>;*j+fUrxOhaZC57_t<!erhWZT=u^Kk!w-@al$46M9m
zd%MQ2w!CgyR@W<b4<GT_GK@!l`PI)^H!IpzPbrFr#|IYG^TVJ15&h<x-R_YvPkjFM
zFG$0VqBPsv=de3)|FCD%_N=#e-{)5Sy?eOsiq*{>XiJ<=Xaq=;vqKXYjt9nRVyy~f
zv7}n7S*@;cu0xrs`#@o}#VGNqo3@jCPHXzrx~jG<(-etwAPf`hZVk#|+bs+ehuy^F
zX9}ZO>YR)ea+DknYdnATvk#Oa?QC7O0Otfftprb8vJ?X4oSBeg1sY|9W~hiURwqKm
zbEk70p{g@yz(|Ntj3R_oT9ISAAcj-j>{b2wi;K3p5YD*^=ORE=Iv-UGw=6T>H?nD3
zE$)R%1k{MTAY>V}xGZ2CU%h<AOSdE_LyDqdKnYfOm_u#DKVia8mBUYIh}F~Iwhf2f
zz)dHhu9QR<QR*pQDbYm46tUj3REu%9>IkzY#e}OxmDU=ASO%RF6!bjLQczffQ;uPr
zD78?`IWbHU%G9?`f{&D-ms83-?v7Yz2{Bz99L{RG*3tJZx3^b}#|MT%#yJ#4QITOB
zC^_OAj}3y%FS($t#~2S}a;YN*jTD+$kxOJ8b{Okvnib?gj3Xh8EYm<(PEZoYcGz}<
zbw$!UH4-JCWy3F}7DL8Lk(e{+0!mj3@S(b{3^Dv$NG;s{;15`>TGTSr)Cztaj*uhY
z{@vfAyr**w>1GSpcdVbkWcThXmi=d13az@~+uf06Kk|b={zKw8kxvKwvs<*zqP(1r
zq~QcH(>P5o3a294d7ABp-o8dP4e!4B$bPp+1;_gC6|2oHQz?X!D0!i81vz}$?Xhi(
zYaEA<-?IDdz~a`-%^GVBE(gZr9<3F2d&Bh)|A5os$m#Hiy}A|;{xstAObiooSXkTV
zq?kC2dz|+)ZcPog##l>?p~iWNuRMf-;}FHls2#>xtZ!HqB~}$90HqWI(}!<p`Zep@
zXTrfY2Gc+YGxK4`aX4`O{5kgzJLcg;N(<vW;QU(n?<JFigpTBSLRl*`uz99&9@olA
zJPkYM!vpi-!2GkH(QmE<*<2z_Gaw0IhZ1CQ2?=fdlWICAyfqjjCmpxCqQh}`_&^@^
z!~{hvY%c5`zM;R`qPrF722>^td*V0<aHeT!x7RdHLzqTT0TV}-6mk8UkRU6^oD^6?
zMl(kRI$@iJF=oPHAjQP9AN_>!cqGpg<8dU$#55dn-ZGB^8pSvaq_$)I`bTJEiI4BF
zIgqp!hJu#4=`u#zeueY5l(6t--E#Q+NEwebInneDaakCTd#+#qh|S9{@UB6}Juh!o
z{Os?3K|Bn6@x>cTR16`~ZC|l{{so_Y_4jnQckCV?`FQ_Gzq)2WO|sy}$T$qxys*7q
z5ldwM_~HB9s=s%SFwdmoSgmh4e*9X*e8!@@AvZnqa3IAK^X@&a+e&veIieF?f%YxN
z3Klq}C`o@>(>4thMIzF+9R(x)?Od?U3hO(xHl&cy+GG7SCJtoVGtVc=_(<0_tk)|{
zBL(!=p9engWC1nx@2VoRVQn<ZsA}Jeq)Hj2jtWwgmbPTgpHKH(?5$_!z&W5Nl`1f=
zAzntQit&z@w*SRPo~wJIqJ>MHgJ0Bc=Z7ksqn}UX%&L=Y6kBn@rGRmUdCJ&&56>0l
z+co|YQ>+v;PGhs8F`DcmB^N?Wj7!Z=$hqc`RSiZ;=mlUwvCy=RSn<p`29^cFys*SX
zE+W38YAM7LC`x#xKpD}KXceSeqR*oWaScesvJ^Ac$(%lya;cWi8Umvg^Ad<L(l{gN
z;w4gw6ip$>Y+hH#RO2+>TYT%eeRjvY`9MLTjIZ5QiJLel38tJ#%Y@1W6BnYc85K%w
zle&l|NY>ONDK3IBPa>9|hdnthRXllSP!vhMEmg2st%}wPZ$rq9k{TOx6#br7<TMe+
zffy#i6>HHx{4S60RSZ0P`I^?k`1q0K{sSdO;yhwE8!(2=%`Ml@?qog}Bk%wAFUiw@
z(||T?Z?<?1^J%7c7VMg&qbSw8YsO*1Bv8J=tu*ckC2KZUYsT@2*U)d*&~!M{uspoS
zZ#M+L#&p-D;l%zpFg@;w?SyYzVutbIGlLgP%McZh`v?4{u(BGz-E#ZG9}<^|-NT+#
zJ~ABk<S;VqA5g&R`U=;y^sd2%i17_U1@fR6PZPH3QAOMh$~jTVY9pS~qFI&%DuXuU
zoUlc#Z(Y|`s+T3DNV|gO>JGnsMplM!+!2R8B`=UfVf_5fmvsG_I2Tmgpq7acBFP!7
z5)sWb4tQfk<(kF%w7I^;mc+WV_`YMF6O&e`ZcVn1(_HY4!FN6LVOQ4|!!%Ch5NMUA
zU0cFDN->Z!`{9VGw%D#;<Jz9PmoG@mLdup@X5tb^VJ3_N+szG&hLi$2B|;ogO-E;X
z((y<}Jl>>5M44F@<MsBMd08lVA)N+vimYC|gkq|xDxq`WaQuWen&;1N(XPcf#o{eo
zUz0+jS@onCFgX*I=Cu2SDw#$B%A%d;I3KvVy}_<Jh=HUsOAIphI_wEUMsGG;KfAtI
z!FIp;NcjAY`}v5qJ@L~wq-95c`z3i}aTwg~hV2rV!wF?H&tKo7OvCATU}FvOut#k+
ztXFFuzTPv83&u7C*U_zdc-VcPTlM$u(KHQVJaCK=Qbbc2jt88xSl`ojE3)y-<49Nn
z##mwqe0cYfP%NvP8?5#8ZG$4qb(LdP)7y($yKybKXmE~9Ly0qrOql13JHT3x>bB6|
z5tcht(~!HKW~)&1%)j{O|7Vi+xYo&O(_8X9)%I}}?^RtuPKg{-^@S%X29n@xZJk@^
zl<O(~yeccsk+Vuw@tfA9MK(7rYQ<_J6!Bp$nNmdHr`6L0qLil%lJATQWG<hX7I9GY
zt5r2WRPT8dAd@v3r5z!S#3&eC>9}jGQJ^)gkwPUULs$Y5B4tQMIv=@6I$=(kIYyS4
zSYo7c4r3${(P%{qfr-pA%jRy(Q@~W1Cz*eviHmd`iZg;qjupOyjAmrxHk6buH1e}N
zW2t{vK4r-xkS>V~_d|#*b!}Oqpp}hMIOC)cs6s|lDTG&7H-za#NI}+g#2MjSQ}qjl
z5+|%u_`ahQSe7HI2yxY#4&@pVn#Y8(mae}-8^j`M3{BG#qroC<0An;U%6E`rAj}iS
zd8}XIoUfMHjBx@>GUl9}ky)mLj2V<Bg_*KUwR81vetfTD;P~*FPmepoGVtQnGgRYg
zdXM)F&tHE@(```B5T=>s@iSd(858)XV?IsX|K>L`6VZm_CvhdD(~+C)6)6Rb3Up1T
zcb<>$KcIld^`vRQq{MJK;hn)3=vqtn{3R)8PN#u+`Ic!svThtVZ@wgCC`vP)4rpU>
z&e8TA{q}}+f|3$8F0^XG6wAZm2x#Is;EZ8&{RJs6%prnpNFkzCLZyjuztFC(n3QFX
zkv2xWE)0hw#WuK`=V;q-8Yet~IL~mtVM@W|h_W8MWNw!%;0o&`Nv&ILS!M}gQ3}Se
zkaJ>lwZ%6L58r&ndV9yFyXNuJ2brVQS>x*2Yszwhc_6!%Ib@6}IFmSh{KQwg4-}<Y
zZ*TEk2aQK*OA3+Q{b#}w*xWtG8i!Argdu4O@tB4aAw7wij8deqpp9lc?(pMD6uj1<
zRvqdzLXNEambeTYzxf%*Y3AzY23<0??Qs37vH^S*VkM&Outs6WBd3Q?DA!-wp8fg?
z!i*h8l1k`qLq-vTY#@Y)N(t}9auluO>gG8fC0&~3MB|JIg{;GQhboGBTCm#kqo4c%
z#yZk6a+o9Ye8L-<Q<}Er_RBXo>)F5khU4P{rtjJO_@^Seiiubv?)DD5tk9b+Wj-;V
zPM8uHjt`8x9XL<7y+Ucl^>$0|JS7)?{x`qm%@?oPu2$@Jdvx2gz3WMFA-VSZ+^WBK
z59=EmX9&{?qoM28<WyLuft(7~c3iFBa6Ig}|L_*Y45e^72KHluPdnPybG7Q}TaT~m
za?$BDl$a?9_uY3ZmT8j3Qc232^A2*P#Dvb4DMVD;W4Z?A9NCJa;qAkbUw-wGrtQwn
zc5+E@Y27E$1yo;oj_DsjhR!2`T#E&zWqeqYgyX=|W_@;-<&vO^&YZQ3%Pbqhc|4(2
z#Vglj#-h(;0r_xMB@L*jd^&}6>{9ixr@CnyYXyB?#}`f8T$lziC8l|yQ?f2;ttyIL
z4l(B<g(4d{r4SdzlCq>8T1|*jEM>$dIZo0MF~$%=lrW@Na9UwBER(ErHJ1!+?8T+A
zED43HqXl6kIBgLmuR$3@3PIjy76;K1LUjXZVu~^r(B}i0eCR^-6H?(BRjw$k8i)55
zW6h;g;eE?yyJfe#C#6t@mRiOvB@vei6$h~|CnYC)5-CqrT)L9aDGn_22~{jjw?$i3
zM@fdP!DVrMC<RT|qjf`0k#YZkav9sM=+}2L(io4ZoMnVw^C7az#A=OfoCnvjOe6DY
zR~;<B!#(<aZs?!=%YVUP_sCCw^h1_MgW0TbC=MS#vwr@H7-q_8VIB@FhaDj<q$Tmq
zw_nq)R`jcm=I#ZD-GSfy<=^uB#WSA0*rFSUW`Sj42qU4f<T|rbx{yft^@`#hIgZR>
z!r+K=5)AA#%G}8b+3s|F#ChRocUqyFp0;tMWu};f_btmT5Tivmj1MDa3dCjNczk5F
zy`|gy2xXxeMvf2nV5FaSJPyoL;KSelip|xQS3mj^;>0p99Hs@g+0tCS!q|p>J+a(>
zP188$X(Wa~*LOr!uu3!lF(tg0$}y+J^!XiS9yuINy#K}D(p;@+JLzg9<p{-!q{+2}
zP*~l(z*<8XC-O4WtTxzH%j4T4IW6>?Yr^80haFejTMmyq-hFt_^8mJ6@%#rr!8Hx@
zI8f%9w(oJ>hLi$Ui+D5UfopFd1`e)8nT{}tsIg?rX&BKZ6Ln!aour*L9`H?r)p9}^
zCwcB}BV88pojXztIIT!Xm9Gk@97)Py)^~Ka6;W5tU>v&bB`&t_NYgC*@&et}Gm>e+
zJTQle&DC0bxfzyWVGc8$GmvI%njj@YH2Cc`4!F9$!8VTR;Q<>1sKnLX6(O$)!^C(x
z(AoxDPYt^Dm|iphDJ(SC&nUwIM?|$-Ox!XYAL!N{$!V7R2eei+SKBHkYDm`6Z&%o+
z!RE+M^T_6|!CQ}Sdz4C;@W5~gjE8^R*22>$sM-=G>U)X_-}IDvZCIx<jls7!<noCy
z?Q6PS;&>c5-9O@6L%zF(?G^KEXd4kqp)##&@XZSR3S5KLUefq{4W{Z>cq6TRoq#0}
zlWZbcLqyv1|J}d(+e`boAZdfShQOqlP)VqBC7)wr&!=li6cqDyshe?zR4sdTED&R=
z8o+wGE#g!ts#NZY=mLx}=#-!)cIH{Ab}1zRBN=^R&I_9NJgImZM}S!RVobykut21+
zEDIqdoT;1i<>aVIwR>Q!!8q9%rO42g*{{*2!Dop#A5+2<g>wz1WOT8_B%R@NmVl`_
z0y!$q>W0o+v{KCTLag5U#(S1D<E%v&S<mNr!Py!%rO!lBb%8&L1|tdmFqeuJE-WF4
zQzC{7tSOg5Ov3#$MiLN}7EVKB4bCbWZ-i-43xZS=4^7Q((5j$~#Lvf6P&uG9IOotL
zN(qE`B*&Syy~0^nL#IxpFyh>m96*dg6+^BtETau*Ck_DPac)J<GmH1q6=+sc+=37x
zD~=gu(9The!9*+LH|KH98f!WznxvNB=^p)F#lV+;{8QpIa=mI0U25U@xaZ^h&kRfC
zaku02xMS5eT-{u=eR0S3r+>`CZ)n>F-?sEO&q#K~$G`Y{e)0eKOI91tKl#%i^5=i{
zr<_6{q(mGOMmwhCKuH6-aS~f;Jcmz@tkyjxLz)KWA>jH3Z6L&%kO~RJs!|fVL{{|+
zw7Ms01-`+WhH!jj+3#^pN4s9rl@{X-!)fH<-EV0(8`^e74ihOm{M8lO540|^)|$ip
zBQG<Ir-7TBD|ELa=#CskyrhVPIm(yabzmL3Z3VbdlAu^~mUfqsQ}HrQ4AY5h4BIy^
zr8T7$DJ357@3GyQ{^q63%6Fd$i;O9X?OFfu5Ak;|(Q%@q<@oMbET<#hIb7p0ZOdx2
zW&PsSg`nn2z%B*j9kVkm!-2<dPAua{|NJ$pFaCtaKXCl!@95ixcJ0ZwArB+N;elm7
zl4E9?7S`*PxFcLcnntF1#Q2Vo3x2(!U2j>hdg44vmb@}@!KXw~8eIy}wxsrkIc7R#
zC~B11**K#;<QA-L$+^(1SCV{|0=h(c2kna&m@uQQW6T!kz~+hUW#*{e3Y#>_Wnz$p
z&-N?QJP=P0OsB(TcBpekXJ{HnP93pm<~d+|hg)6Krb3*K#Nj}iCzR0~hncJ<fldtv
z#%W-4wL$eObSiiXMj1lMjHij)FJ94KU32>C=WMq(tX_P9wHkXmVuyh=1v!OnzSph#
zd-wSK`JT>b8s9Jv18JPdz9EIgJSIvCWOD@Fu(`T{^#<o0&y>Txzh?@P+SaZ*x~pdx
z?;!=0g`_kAkG30>+pr)f@}i})rIhBB3PD+n_jupoQ5QR%Rti+%c-Zml-@cU&UR;Dz
z5f((_C(jpxunkJ8VmA97+;GZI&&N5CR*H)ANC2e>Q5*-Bb7G&C6LGGmR%0-x!hZ6Z
zOCVM@a=FCp!jp60>}H}ub|m;lqBue=W?bVL$Fa873r*9?)?AqbsfcdCn#9UG0p8U1
zx>Ay^sI4gbwN?TX(*gzZ-jK7xYZ#Y^qJ~dtdB2%LELEK7gxe6KPzmQCQOz;W%pu`j
zq_IYbqgL?70@F!!T~i~IGpdF=#gvIj7y#$OW17VZ=&Y#-XlcV}t*Q`H$!mAwGVoqT
z62=;|6ZOB+T2R2of=VcrYY}3|X~r0bvNxC#&?@1rCgcU{Ta@ka?G=~?Re~fP+8*Cr
z0aXj<TXfE3BjHA_7>ZID>&Yn)f^;0bU!kmY3R0Mu#ytUpb1kWW>sDC1hLUm4V@)F?
zBV*_`&nRa8o$k@^wZh!oJg0RQlOn@$M^uL9_Bs7=qH&I!n`^GOSFATz?8g(y$c*gv
z^(#Qolq!zAd~wHr_#glKeD#aJCq&_HkHZnQ+E7lJMk_u)-ZPI2HbNU4ra6$c#TYQk
zU@&wy&+s>zvW$2uL@Jdu+Db(m=ZrO$(=^d&gX?=(7Oc~_P0#*xq+mE6_v94VY_D-m
z$3{yk)o|Do=7qLjlM1Baz`UHW-t+A48DG44Lx_QFECoXd3&kZ$7?_rsFrAq1Kd|-|
z*Y^|!`_B(J1Fmk&MJr;QC`+N;uGsX~jE9k&6O+{x<FVE-9`_hs#M;uWIqpA^LM9`E
z7-us1@sk8y)PM`uUs0Bk5=J_eDJlY*u9DnJNsRjkX#*O^&Gjv1drLBo!(mTInR)jK
z6$hp%FfS252ZGVWREX2WJnV=Rro6^CFX^9Mqvnx$ngrck3f+1|({`j%2)@PMzF~g+
zOt^nbSVm006`Ell2~!|@OBfG?@iTcDu&%*$J#D|DZC3R}W-3Hv0)#Jaav{z$WjT=@
z*qi65WhA5-@7ACl)<7B}#lUiSL>(T{IigFX#DHr%snC~`h;Pz@QHIhsBoxG%@pM2b
zgKZiLnf0@m_&72w3wpJ|_iNNVvYRtx!`ENEWl4$Wuf8D1h|w~pn3j><`)^Q50#;%!
zl;aK;13DJQX<%L!uAe_+{ztAED7moP?P<MdIS%NU$X-slaf+y1Slu-ErWfXgY4Gie
zgr<4EuoTd`#?KljGv9fVlbdsdT$rMv<j606{%gq(scCgZOf`@bKmLP1;Mw!%^s5!B
z$oe1~+`|9u|N39Ef82?9jSM+aYMPyl2Pz)eSTfQn$W<SpRJoj7719Bw0N_9$zy3Gb
zr0a=RYf=`jfYMa}t~4Qp3p)ELQ$tX@N^1f7Xz}e+J^HEwY|h$&CoAIFSx}3SN~W8q
znR%QgZ=ovqOU@W;X<C>O6m(JbxhvB^DO1f6Qe+vU<TWTQF~eT80Xcyxf-knlRjZ(!
zd{dD`!nU#aY6A>WHt}4?3MEzRWhz7^*32cG!=?)JoQW|>ieRcqq^ahPq%2}WYXSML
z&UVBzdb(r@8FtR%tT;$CPqsu+=ZjmU)DKk^N&Sm)0qtn}8;Y`IJ%Q=4w!x6a^RGs%
z_E>*aKf6jj)eYWntF_XSQo*<$+iX!Ok<&nq3sD7(?J?Rjj2}p0!8JYB_hhBfWQ^$`
zZzyNrRjyqmAY|zjgg8SC6zyv;)V~Ecfk^^>m}5q@8yaKjx)xuGWJw#tw9J@h#VNz%
z$9sl-Bx+5ZCRUp@zHbTd-r`c=CqI16>h2BN!aN@-Az}M98CcyJ+;pNtHawicv^`qa
zxKLx+ef&&!eM<<Daeu)32ID(vpJgpJK4XX^j)x<SZD1G~#}RjGab6RXA<PlyS}<!)
zk8e>WalP808;@<T@YXXNcPQ)G|LT{7dFJND3)pP9{o=<AyL*oN&x~_Gd*R;~1!*}-
zVpi7+CNw7IVT4tS&Tu*nG;PO{7Ie-`_cQ(ainL79{>+7W*)v<m)%BeSjg-Zxg5O+o
zI_!mhWi)Y_Ih~H2P7mZ|r0sg*@c=P$I^A<j0b)RJZfKvqu4_|+$qAcg#xN15sX~f0
zx9g5y{L8<<x|W+4FX``IN!sA?p0wW)jF!*LI>upQh=HqOC~ePtIH8&~j5}GD3dAL$
zi{kkCS2XRn%wZwNi6tzkX_EG8jKok#<yS1jBP;`PnZSsaVtxC9?X%a|ZbQ4hgIHib
z5n~WfZqtJjzp%=Y>C;zaz$nnApu&mqG^6u^TU~K`b1e<s9E7i+g?FB{6QNEiz(JfP
z{hH2qq<J75AMpJOU80Dq!bl7=-Zro-q|f4>FfZOP4M&t!U>n-&Yr1}e)()!+5AVNb
z8V9;{M~RM2)ke5gOLmUD^yo1VauGV@KjMc`JuztJtWml{J8=D)ZhME{Oq4Xxu2v{x
z>H8Mk3hi1*k-qD2Sy8g#FpvD*-~E!`{`xC^^Q&Kp9k0|Rc_m2RaTw&6o@d+gM;NW>
z`<_>?U-6&)Xa5PW-@K&n8di<xx4-^1fAP=%TgXAgx+%(HUXR{6Nmt;k>eiPKp>j#c
zi&guhtI}U<BP2{9*VcfH@O4eFlZ`zui<}6J6{dXseN;N3Kz5WoucB@;8C}!z#MYQE
zw7*L$UH*hQBpwbER*Q~i34zV3m(fBUd!;1aeg41H{nyW~S(@I5{Z@R!8^5^su3fRa
zr+cUwk~18O6oYNa0rr6bAIJwj_5WtTfFo!FhDAssY{QgDLegkBGd-QEEA9BjH#})Y
zAFSv7s-Oe0>VQD^18h|7sxQ3hS?gZ+eO<px2%Ha0u0+vinM+YLcN$LC%9tWTNh;mU
zamK|lptGWu<#aAZaU5tZ>Bd0aTC6Ff5%1%Z=fBp9oDx$gB5WxIE}T+Ii;m(n_c*;i
z))c2zq5S2U+?NQkSOl#xPdb7!VwkfHbm;^bRk|Sp=J2w)aDeY5M2h+SG~}c)w#C|p
z7!pZYvMBi)N*io@4!I{zd$iKz90}8b7)FF%B6UUdBQX{Mrb-cF0UVJcek=}vq#{9T
zatLUpQBoJd_59eL3LaVLf?AXUG6sj%4Jd;f_JF{eW(FU9zk|dNDhAAQi_(gkO!One
z?t#Pcf#61>(wM47A@Gx<scZU&#4msQbAIu~TefHC{G<QL|A@<HA9J{QhZPQYJYtty
zy8Q$DclQV>S)QFyH#JH~<T&y8_{j3JpR#`WDMl&Y{py!QA2BIGN{q)Nx@oX=qN+45
zM0WcfXBQVG4KI0OPApevNNwrVL|rvhw!-@n86uT3q^d!*D^$0@Dv7KYpfsvlg3`F%
z1G>^EUGez-9X?F_=x_fSNhqQ#77^7fQ90s<15MLXuh!`884F_w!7&{=VouoQns&M1
zxIZF&S#+4`gE7D~jT{~yO5CuUSf5|8XsQBR89ecLAZtm{@T5}lYK2LW<OlrN(LX%!
zaC67S<rP6Mhz9)4do~NjlpuM>vaQKs;`Ys3>ct9bkBg49zr)AK`Q-)3ataf+1>2yk
zW$Xs1YNWQT*JrT1$G?4#K75V4ahUcT@4$qi&<2IT)HU7V0bya9ZJI;}ejrV`AWZ{6
zVj8=mEY2A>48+)Bj4jkMrCBM3A1qcI+|bQ))y%ZtAw)tdN!2dFCqlm`=72B`9}`-c
z*}5jsRa-dZF_IGVXdn{8Nbn9fjYtw9MOtg-T=yE2Dq_wUqp`~+QEQkSX&R`ThTK#{
z63SXudQ%o&T{H%MURdKe5yzgY)~KqY69QIS!tSB4(_Nr{Jg_>yB6?RC{lCXoy}Cl@
zh!h&zE=gGuj7FVbP{qi2+!0NM)CN?AsaHkdnKLpa^jtju-~NyPlrO&gnrU)HSeC^T
zKfloG(%kwkT}h)AM#*_ok?amf9{0cEH=qBOrm}qc$xF_+8{WTrhnvP?m79|YPlB^2
zv~Z@x(q@5=fiN@Tg*c5GzU%UvllJCTeTLE*MU**woYVFKBvJS+Ck;c+Wh|KU9O5%$
zkc}xc#GDd2CXz^VJnpQX|L>sYP8)bgnV2$GD^d<vt<buNVzsq1IO*F=03Q>HLI*7N
z#nX&K&Z5mysAyvlT9s3H%!pW89cjimYpW4L5~8QJ6022tJfS>KAqADyrRbKLa3U(6
z-p6_F5kj0b5hYcyZ2AIRIUR=(3T9b|B9@G$V30!M=gy6h20haP_1rO%s*Dx1E-tJR
zNj+cl&+;M8tcXNR14HMFt_g!S5ur1Fx+RVoQ!O#7o!|FFa(xlC>S}g>$l}tF0{jsd
zC!{P`Yd4K$TpR|3DV^q=71KDB;zgGN!$ntIYf_?g8S|v*XOVmb*^rDc$%Zo#)0p;q
z-J?IO7*HCWBViiw-4X8`?Zp-2-3{4078gsVD9b5G2o_z(kA5V$_~1FmeaBiPnrcb&
zVa?<1H%vEoy#D3CA*RHlwj`|y!^G=XuUM=XXj9>&Mw*JW??@@|?DCx5{vIhM^|EE?
zjzxT=+ZjwUv8*e+OH}JMHgweMhJH%yZf{UR(0flWBsqFwF0<#UcTBIoDD$J$lI$IM
z8gS#l=rf0hTk_ovXBQiq`couDd@~&%39ct=jmeoT6WT`<nM!MN3>*(5qHgfILbgks
z4-7XqkTY%5uz%d6ZH;L(Uw`uz*^gKyXqId2YDH@lx$h9mOGHx>W5D@hYtmX^je_Wz
zcDG1f1BNeu^GlQw@E}+{dxo+#_0@F={pgMaH`2SF%4*ghJSQiCA9{37^iy9#LTn3F
zg<PyjE|e3o5t#Kki<ch~yOHVsHOr=^d2z*}GAP#}v}QUU5GhbM6<SHEs%AK-GCOpx
zxIp#~NIZ43qFtVobLQdpUCBUCWr2;-ASC7c{6x3Er)?Xi@xbIuxKzE`(!0nw92k!~
zR8CyH`~<x`XH3OmAfqEhk6svD%4Pm41lnk#@J!tRV~<uDVQdj3g@~>!SC<#aszzyx
zF-x@4jNO6mxWje3;^Pe^$R)%g21>yrbVM#Uq|wnxkMHgfxkQ<Y<XKe$bixmhRH_Ci
zs2jtue<X*Yya$H|rm;ijf7{*pyL~mAORSPiZbB|rNKwR`yW@dsv4YhVUKS6(6bh+}
z1IS2;q0F7%-`w*0&D$rg{WQ5EZihm_6LMB{$GJ5uaZ+GZ*@OzcE*3Fc9;viux9>1Y
zU?g)Y=gHlZh~D9+i98n@LJDLOb9=UI_HoV&IPnBRI;G4BL<(g?{f_8umM5Tup1u6#
zcZN{h2yv$1rF7CCz-eTlmBh{6g&1Se%Lu{g5Ua$DFFyJ9XCa}Oe^5-R2w;uIJBMxy
z;A|X6@?5-VqZy}(G(Q&ko#fAAlMHkALPpNQz8pO{e5+VDr=+}G?3Pw*>dHR7FVZSP
zF1juy3;f5qFn_H<&m4ux7wc#dYl0G|Feu6BJQboG9Ot|R;7K_j<~DjMzzU79smo(X
zF_4m&1&6a1f_bw>JZTaTWxP~CQUa|aN@%KTjq@WSr`i7LiLNj3J7rPxSVP%Gu>)2s
zoa>7)Q4|Q!G))*4Avv52W!$4Qk&F=<Th&kehLX4`5F%1$F>`^K0$Ld)C2rceo>Nk5
z45c8D`jnjbZ@+#}G4T5HzlLc5GEKW6OHETPF&`|DAz)W)WGt<#hueEhRg<R!&n`E7
z^65)tOo-6azyAg`?AV-_isjf37^5LWcXPw^c+0X8Y+igyCUN)b3of3&WOH%B)Ezi$
z8fuyGDYJh02~&jq?Q6Qj5nHt^uRfr;`UsUH-EaP!<NXmf8)U8NcSoAV5?eK-Y^V|l
zZE2T^sqZj$fm4D?2_7GI48z33t|tpvwhdR;*DS1M|Nb5R{x!)M!q8JKx73>@-M*t<
zR7k0q`hjkDS4e%`an1slD7_@dvQg$TH018?mSJ+--Q05h?266V1>Lcu-|e|ep8bb`
z=YRN&G>(M*4t?<qbN;-9T!=^-4=|48p<~$JqmAJ5qYu$$zFCGN!`SgSjBMB2f*uth
zX7ZU(rGg(*!eyXpn0(;w^Ix%9F4-SDVoF?`E!Zqqh<;BVC&q3^yJ%^es(?180{>A7
zMn4swaX2sxBf*vMs=93%dxxnij3^(}6pBw=&kEt-JVGdBvqm^aoO*~OxgI(0?s3N=
z+RFm!#j*tb3?uvdThcIc+#fjq^d~guFF<Sh{X1~Oyuk>N0@okuA6&_Y$cZ#fO#555
z=Ub#PR7&FdJxU~Mt<hBjIpKyqMrJAj{q7E_6joO>ZH=D=Wkstp5Fba3)+EywTWR(f
zDakofFBiBNnT|W8PT1r~QnTC`vXE?LLmmT01-kM3d4+#hFS)otR~CXpDTP*&<P5<T
z@RTwOgw`l|T8x>~CFj#_&ivwUensDp^y5SfafaE<n`4N?8Pt=~w@HFhN`jPR0n=0{
zbupAea2y?NZCSS#{dB6Hp*6E(fSIL_0Bsa%p5rrT5~x%*`V`A4ObJoi$GM!wQ!ap^
zQ|F+ZLU~FSOcbb0(Iw0?|6IPmDBVvY#In&TDM(2?DF;huqKFm04KMl*jZl31wL%mq
zT8fEm%HQ31vOP<*HjFbqID|y9r85y{{BqgMXPw0C!%rzZbwjc^IVa{LcuI3uA`!x=
zpelL@HFq>hi(VjP5x=!fMUIIa1UdP+HC}?Ca?Zq<iP-~SOrC=vOZPy{YloagnJJXb
zx~?kB?1ZpZ<6ZgpqIFOupCadZEI4;mPT-<48)3P;Am@&p1=`ePikDwVLDQ^D5^udl
zo8o>7A<`^Y7-Naak)%M(S~q1XOq}NTx~8&ncB?qZk&J?@X3fcodr-uZNlnGCl8eqL
zrDClVO3u82@_LdBW;@1MKyu&X`a#8jcO$FCf~Khv3i@u~7ystJ=JLgJKKSr6y0K?`
zyl1~Va(8#&*~N;@<t3^qt@p)hjbK7H7C9<XPS~o!InUH}R8<8=Q*E~(G~=;jvAtxo
zHLNZ#sG0_$4Z%75ut&9p6Q9P3wz1e$F-{}>?HgpuIOqAzzxf4qrP=tR8aIm-!dOri
zB`n5RhT{WLWa@gs<UO|Cmbl)krxu#6&}1cf@!21+GK#kk4;+RLw|ii<Trv(0(|OjS
zCapifq#et)ns2l%(>T!YkGRRPTrH{BTZDJ`v8U?>v@vWq8|rbS)|$m?S>QMd+SQVW
zhXWVQhBS^$ySu{dfBzMW3kRwy`hqafv`hB8d-|c{*$2;9oNbWO&{h?V+3<LK%Z+pN
zZX{!{bwv`Q#CVEKRkyU8HC331s5vzbn!4rc!w*4ejL3*w=A5-o$g?fvj8VGq9ztgO
z{0bQcI2IJB(O^vBj4NF+IgfXa+E%!sV;VY)9p|V{fs&feb=0dRwyHo$q)en37~MpA
zxMduU#GDB}5%2ENGy2p`4lM=mUwy;l?H$Xr3v9dO?BWa~GQ^SCd4iuf+}sfDnzKLt
zGX#q6^>29hi@zZZ1MR95sv$;Pn5dd%p+%Ar<AKrpxfPu_K0dIlYNR(vQ+Ns9M`~r6
zLM92p@wmrUH9{HOG$3=vO&w!Q2-_m-y7;1}J-dfJ7cW0T+A^Y16V&GS+fC4DL0fBl
zD#1Wn39@VmZXo&+`eTGesiti1KHvd-m^V&}m!JIPE6=Hr|8l0ZSZARaFrOwfIYOEd
zY)J@$D<?tbgg)myg9KTX&>}T&1Sb@*txVAg%v8UabBX?)p@Z7!Z-r`Pv{Zzcp5Q(u
z>9yzqq?~caQj(mXT?T?Q7Y89m1SLf;=5i|6N)qQzL7s~nDNp<Y_0(DTb_1JtW+9$l
zkD`E=vw}a7B?#oKfhdK5wYH#_PsxFKPXC*=0MVD(Hl$eyS**}y6j4HSh=rMuQb8)0
z*Mz4G1*OVCF{N)`ms4<4&KYYBQV4=5oQ7pn&7{ISe_!Dl2vLfZQg5Vrj4+QAWT8XG
zTnaHQiojY*G4oMMUQ8y>V#D%UDLLB=l}0FykcyWdJVzRfZd%ZqI5_-tU>M$G?V^CO
zVwguj7OBgLej4_q;IPfI6agrj<`N+?DFyI?q|_W{B??m@m23h*oW-PbQI(5Eqd=1~
zQr6|Q7R;F85oY~}5*jJzsP9~WMKMnJVgJ4E(I2$J=qxdS)e0ZuQ_et@6MpK@(txro
zHXH16$+{LS*DD_H?s<Iwz<7UyU93sN#9?=jT{Zwzl_q)3_TmzZ!MVWZ{F?J;&xyGZ
zy3}0pD^p|Y73%D=bfTii5O6}GmMhw;B{$!E#ryyEzd@dzp<jN;a<gI@2KqzCN1uL*
zckwCUgIsQUKaNP9Ng)xY0o(x5Aw|Z@K+`N(rinO4d<y8M#S4XK){uJ+{m2*p`p>!e
z>@z-keo1ah6)cS5^8B3Zmx8<Z?|688But)ZDlK_2-;9$Rnf6B(AH1Y#TW&x98*JTB
zuh*QJhHAZLfB%3Vc2w&%{p}5p@4ja2I($E%E5q{qf_AZDdHtN_W{qtdtW?OMr@w!M
zENM4qjD8|XNs<aF6k2AeO}RN@pst#dbr3z5XIm;I7>5JWF3`=g90B4$@($wMR<{jV
zwAgxyxxB(XzDJD%#<b)dna1LiH<e|vzQAoRa1Zau!$@;>iLRTX)^-y<M5f?buC~;(
z6G6nlH=qBS;3kX~xT!D(!sM975!=)S^NhkOi&BENuCX#vtAxpprfu=w6Mc{Oj+h)i
zOo(Y>ynjpF-LTm#nB2s8KT$UoLRZwPA*29PM?xVmgyczn#QEY2P9if51Af298Z+C)
z0?Jh6;86NxS4#vR@ZMwUwpb9epzAsghdu40W^{@7Uw_T^`3FT9sWd+LA9Aby&RzY@
zFR<|%scTXYh^9u6kueaDJEWF~wjzgytlF|c&fqZb1JkkR;qLKCcvdz?<_^Go0?y?`
zoMu>0o=?}zE%RC6|5PYQA;~Jw9R`gOr=X#Xwg&PkWiCB=`pF8DEm`HcU0+n}Wi}q?
zLM6{9_9FW_@%OzCPxJp1=f5cFLzyzo<AupbLQGgu=Gh`8e4bm+QV`X*%XSGYDjkHA
zfH2SPdGg6_SZ47huCb`;6EL~L0$`3uJ{3~I2Ly!=Qi7u9gZ}*b3R$9d%eh$~wVo-A
z#TlTC=9HPBXKO|HPGX=BsaTz5NfL|{3t{o<XEczFn&&@dBqY8o@T1f`DScb~xKOkZ
zIVT1e3rHzvYHQG16`|++1WR!;>m+7)lF<6ewNQ8o@CQHsv<UnJXk8I<vHNKRQ3zrv
z7QvV%A`&%8j_#4&{T*7><Tw?6!s3i1BZQGW1EbtH5#xw5Ek>((+@nZ&&acrGQftE4
z6)U7F!pAA@(4qomK^crTCF@~461`)wzC!5*Qd9(4EQL9qgr)!e*AFWOq~z`}lGo=*
z6<9457C{&X-hTcYn(ej-KvHDcEU+YWa10N(WFI&V9Xvb|L!w!)sh68MHgdvLEj}g|
z>kSMOX3>)7d68CS1{<6MDQPM@D_V=yZkURwN63V28`_HtzWn@a*!7t6bCNRD3q>t5
zIy!{R1az_9jeW@ssFz4>sDeXgPm-2;b3wOzB&Wo1c;r>L<NWHHrl}Zu!{yb6X0@Qc
zPdxa6^UIcsfi&$Ij=9j!Vqky!o@HgxDNqSbNEzuf4-XHt+cRP}ar4#JWRW=k_+yMH
z6?upr9R_6Zba(HO+8|;=%$BCFzxbMmy9X+3QAX43kK_c4%S%3KTUwzR`<~n%QMy7I
z#cHvpP9@J<F3%x2s+>T9%7NfLQY(gs2V4OJGm*(L(8!wo-3`O&*shi+5=vP@f-!<v
zuBdgzaC^&md?bzzX>01LMq~(nVtBYA3?s|uA2FGlsB4UEkV+B83H->au`KHA5>Dkt
z7EO(<E7V6H<J`pIxTmUWtgRTVWqp0ka=B!?T##a-S}tH32}4Q3GgU<vW${l^meGs`
zwc6mjiQ{kn5_i0#Z7a6t8-@_^lPmmh+k%jc`yKkEAh!!*PK3!9p<=O1@OXR6a6GbD
zG;A*}k=Ek6168|33dK0~L{%Y7gRYlFhoCfR99b=wBq<RJ&R=|pPKn?WwzY^wg;C!R
z9P+z*9lJn|H!K!Qmg{p?s|CtP^7*<TSN+5^j$|TIMJl06YNj*{G0}&>U;p*r5PX<Z
z?`A@rE*o5kk-01=K{hibP%wjX=9zt(1)MokKJ!x1R%YHvN;ZJB7DOgFSAanC=pfE4
zft=F`#7CSr_mm5Q{7KP2d;Vdzt0_g!VjcxVT!=Fru*~v9Du<nr%5%vlvEr-?P_nev
zwag?}vKdZTWU~BrRPbFfq4+lWP|byaRAnP~vAljEcouB|2E~+~#uri)9Yf)@<fk^h
zP!cg)5>MKS@_c6-;+!VP{Pgn-N-~u#B1m${EfJKCz$wO9sse;kg}Wb4aHpJsa&S{Z
zB4&+2h!G!iX{paIw+o@T$5JW<$82qnGuJ_=QZP9eOOYi@$L1uoV=e|xG1MtVk`Lq)
za7tE~YK;(z;Co_nXuZG<4|Brmo@p8gQ$U-RFzhh4!kC&Vj0KrZv4cimx{Oi=0t%@u
zQY-wFuuW5V1cEuoK%s3-3=_)KNM+0WpAsQDl6NFu#tzeXz)(KlLP)Hxu<ho1-J?IK
z7<l{cJy*}46UT}5(y(f49`7DVs;Hxht_Uhb#(T$Me^0Fqew=2U?Lf|%#d?F!<-_p)
z^=rt1>z5x<FITvjX*U<xsv`Iaks-OnG!@ih_K`S7P<B>amX^^J10>1qU;hFnBWLF;
zc839R+#yyKXj!a(!^C2<#t)uh-w~81Ew3?B;QB}6*cSjtDmRx;f%WAjRaJv-=ynf?
z6w!FZv}f4g5{2aQqmNl!Jx7E@f824{J)-LxC4d;{#>n6uQ}E>5uaQELs~T1pRPDeE
z^%33vf%l*PiuJ`g=i3!2XX@pehv}YQ{OkXUq!dym*2@-6WWPVqJI@dk4Gl5~&OiKs
zMYG_`zxg%yn<ds-){B~YxtwQt3W+BB9n<|Aw3b+<@pELdUABZ6IQB<`DQz!p41|Ex
zmg8eb5SoJrOQup2`@1`iSd3|T9FGNrWDQDLjMjuKQSBOCS7<lVKisl=^Cg0Lo~O#9
zRa*-L5H(_UE`(`B%VP0bpPwP6VAHMX4tx4xWOMa`#rbpAt2KW29tkYATZX=4p9RJ?
z2&stC;iev|H6lkUtqHQCZZGM(N7QOf5ScJe7-^{13yi6-?Gookf*)y@YkD`3{6rKb
zhutcP5E<tca&<wydPcoHN3|^)N3%F9IO`Y~r+Z=$)XOay=np$|o=`brY|V1f%yYv8
z!XShyun_^d`F<jx-_>jV;zu+_6kF`}64kDds<f*OPy?0sMSv$}K7Gn0?-}}$w>S43
zx}nTg)3+xbC5v^f%9_#pIgcJHYf8f2ywIP7cBS=Nf@CCe77xnV<|gJDaw_fIoJe!<
zP@Mhj@H9>?!DK?0%^OO)Tq&&5#GGflo}630^TsSBJ|tY?33`-rE^YZ)XAn~+#Z2Ve
zja*7WZ8Yxmz0Mc!{QNqc1aD9Ofz$K-c5^Nd@l+%fQD$vti(bmakml!Eu*Ha(k&uf9
zqv#X`amMC;JCZ1WG9ZxzALew$qF%IhHR~iwf#&@win$bLC!${rrL*!h>H(_CQUT8U
znOSgZ7?gk{F%t^&JSK=S6(?FbQ2}`lj4Ca8rBtyK%ECyH>V$Uw&TE$wQW@6kB}k1j
zmKa7%wZ&Fzf*%pt5Th%_X-q}2FDil`P|_d@N_Ey;Ohxf8HDp<e2&+q1Md|sEDi%!y
zT2<uPxnZn@l<^@DyepcRxlmC`g0Ms&#u0Maw`2@BSMcf~IAm7e>mL0<#X!4S;HCq5
zf}_hEZvzO$6e3zyOuGXi1yU?vj{V&Y)jXfMxIAaEKBHZ45VMbTzds_SVY@k}UTw)z
zV0DR;bfHHXMRFtEVJv=etx;r-$9qzi$ZA2=E)Y^OMoF|ad9}gqZfQ1W97j);61QJ|
zg{ciy)u6>03t>4P9=W}LBrVpgHs|E-fOe0p*DF-j(mmYKAKtQFuUN0Qg?i^k4&y*=
zP4U=EjoiMVQ5M@&7;BI+(o<TKP2DmM$6|jpmY5RT<&xFeitV#2x+(GXmv5Q6Jyi|M
z?V98L9%U>-XjDkF%^LmjiiG0TtJi2HIdmoSI%~y;Kl+5#_N>4|q(sHUw0pp4$$E3f
z`PDTsMSQ;}I$s1tNTN@K;{)1gSZ=Y4RcQ(88j(lV*U#{NLL`B1R)nEvzdP{u=8>Qb
zA>1Q;&&#VV$%CvVUI?Cj@)^=-roka)Mn+GNlFN&8bhRL4Xl>2?H(!<3UZt>g85hYE
z$+9gshtZJIjA$*OspLH=M(V1<>N8dg&*Ud^f6vh!al2dm(6f8jV3#YjDV%UY#)yoM
z5hoO;Sr)~z3xt@djU|Rmm^y^c*h-^SB)X1a>d9JR>IPXi1YhEIW2SUubeZj`s^#Lt
zXQX<IU2YL>ARgbB7-}HP@-uX`BG)ZKN?JVrIDilgA+Tt-Gnr95si&nZaHG)uep?II
zn=O@9#Sd+(neCqEC|faO+GhbF8Dzx8NQjw~CBOR37wq=?*<Pk+BygEehfrF;T1td9
zbBIxT5_p}!KgA=T=Y_Y-<4<-sr6krUl$cNd!IQiv`tT$;JlP0^REU)4qGUc%&L;3O
zFPA0PK$ib5XCYV2MWf(+Vu~k)c<BrjL_vNM+?9PvKDh`45^XdgmEa-gXHnsNniujr
zl>w)jd=V!C^Qm4DXO}<`940b~X_}K4bBP`>#uR&BOr%u05~s%+yf32@^6cJ_WQr|M
z8&kxF;yXmf49<_FoM#NLEY?1)=T1!#i-s5r`A^BGZbq0le=|D*#x$oFo(dcxaI+)B
z$5fJHwW78fqh=tIR#cVc<ZL)GB2H|GlnTp1h*QB<zNW4$fBZ*3L)#ic!pjbU!f1<9
znG_UCLrOcUW(lfB>YBAwWn`jMNf=C?sk;Sfkj<)`>{B_M6vAT(v&x~wEEm|-M068M
z*#g#zfx0bXL_Z#JekcS(T|%a$uE;`DFU~NgE!w7B4&W(&J0AMCT|Z1fC^ADgVlOV~
zm7%}6CA&RJNbFJ<n@EW44-d$8#q#+}`rQMuAJBEn#m7HJ2+8ByFPQp0!cQ1$==weJ
zLDOt6X_|_0d}JCAES4LF!y}WMP^Ln;0#}%Z4nGyR(fRdrrfI_M?{VXRsT&q&S1g`=
z!s^9PW?gPT%Yf0EsT=t1FaCz1J21se><-ZFS=SZKqD9&YB?XPqB&A5>!13XZMN>AH
z?e%k#QZ((7{&=LS798&Hsoj8hyvJ0Q##UICq-n%Y(~QTQSgtoH0q*fVdRy^qx#64R
z4Sjb+>J5#lSzlctax7T^3FP5{x^8)PdCqoG&7A8<-xp>4#l;oN^E1Y7M9)DR$K4~t
z<0JdGUlIB}rdgs)gO-LAT?u^B8oND1R}Dg%65J9UtE$CR8;Ct&n5b7P?B)`ocI<Yq
zI6mB#gf(MvO4IEF*UOe|jzm8)^&QL2wupi1C0c-@1kxPde#P{-FYuks2Gbv?nNWTr
z_`*+b)+MtbyOHR|!bwMz<Sp+ZOt?{Csv40ZiWw_v4f^bouyAOl5vE4uL>$Xz>!u@a
z8d;rR7wZ#(``xaLQTDfl;Rqse{lRmjNhP>QNv04mdWkd@Q%ERXV;fOyU1X#+WK}Ff
zD#vn@A8t9^y)GogFaeTe8giHzZXc<(J=qw>?t!t}mu!N1L5Pun!dQ!+lOEGJQdc$I
z@codYzpIy=Pl8g_r3>W(Vh)f=Az+L~%1rd05R0ZFL{aiU`Xhh!S3jSBQaL$lU9#m*
zHowv_C@t<>zO0B1b0%l;ZJT^1_f^)QMS&Hi7}1q3dv_?zkQ97T#~WQX`A{~%oaX_$
z1X2+rYVoZvUMfxD926>C43WtPGKGgQOo8(Bi&#%0h$%jG8)j<m(+Rkse~b3OX#J!~
za7y7%BMhnF^t=U6!phSnm_?J!I*eS31+8@v6sE#p07?Lqm^Wm~7*>hmvCFd-BjxmT
z!cO5kW1Ev`k%IpXljCIZM3$eaq(rF9G)*((pg07)_f*yvyJpN!9iLOtA}B>s%n>71
ziKU+17($jP<x-Ccv@B~vWy*Uh1!Wu)iW8({9n8FhSq~z$E}Le_NiJg{H!uxHa`b3f
zikBQcVdkKu=*iL3)Eg2CsSQ~cZi%uBls35OP%;VXExKBeNr*H9q;lp&ODygHOQmZj
z*A+ceHbr0?Mp7J6ric@@tWZ+p4!58!y53;xExKA0lP|wR3bIgC^%9@IU#`gyDh8^$
zW_y0la<Ktp8FqX6hg*cmc#0Qwb@`mj>kml+!Z4sJ&Dr`A*X`(T-Vl7`;r(lN$DU=o
zK&HfcyTMNbuJ4E;FmyY@)RRIW1&6W?($u*ANN^5QFQ}FqjIGLPLM0>#878z?u)h9?
z#q&=%H@1ZOgn&#VY3!JekBt2z0>^pXpiM)yT;j)RCZAbk7F6|;tPK6`hKIX%ur3e>
zFEvhCVwU8?flaGmZBSW~r$R0Zu6zJ}e?Tk6>gqX(OpK1c@A3N`$DbcrZ#FarnyV}H
za*dD`wzAxP^Cb#MN|5|S5@o@%wxXUlvemk@G9{&zsucy6Bm`O;*5_APTM+@hI>&CF
zBX+Oo-`|j9WQ?G);H*{!M+(F#5T!+*y&%VLu&ij#uMzErm=b5_TYmhLk01v|G_McE
z3Rf>$TANVOLoBmjy}Td?$8g-CMP`(lFm&|Yo@3Xct0k+(5T;_=+;<aknrIpe)3M|Y
z2u&7}aU2QGp_or*0z~wvrUgnwY&;RW+6tvD$~J^?j~_iUN$T2SwWKl{BVp`27?0FV
z%P<J6StA!4ru8||Sd1A_)^J``BqYihRJ&$fwPp4f<80%Bm;zz<Na%Ou=<!JuM5z#H
zT^F5!)HvUR$PE1f7e{ixBZ)%Rj42WPKtFlv&@hf;(aHEihurP^A9Aby&Rsr~<iMB`
z(K#mX5K56k;m6B)0oGb0jVVm{km>u8U;N^ieErQgPh!DYz+UFjGbUI_o<zZUw#p$S
zoHn3&^HWL{EVd7X7^oDi*A+$t(l`;up@b8qJR^7u;ZuDlBqwJ`fYjd>AVR8nu0FT@
zXI_8{<>$nd7=2)L-wFrCtb6#@68P=R|MWy-D(s4;GBcJ~Qd`3~cNk#KEO<KEmY;Ka
zkxn5=C()dq`TbR8APdHEEc&aQuyaQs&0Um~A|WMonm5}N3mbt<bVbmbnFVUWKV>o?
zzAegfF6o8N2V^f;t`{gX+Z$8l#9q+zy>u#m(o<((7Jdq_RuhDH(p41K<SYslC1*g*
zo9yJv7^O059&r?*<rAbx3g#^fWYK8Jd9(tM{DkWsY1_KIhT4*YBe<d9<5O`42#UKz
zYK4>;r7eDry9R-f8dI+krY=QaDgkg=Dzq{MwJs#ykA+<z%KNXBK_Ce(k<-NS;SO6b
zNztQfTkyrh0o^PRVol1P5V}IP^qv?Tre2j;>FN8e9Y3fTSQyQEQ4<e4L{n2)gKHNE
zDKJe_+7h~KF2~0mp&zKNqHYu>c<#RX6<so1y!@EO%ZB0pJyqK>`iZ-DZ#eW{pqd&}
z8#dbw?ZV(=B%@KfMw$r`CYp9xD$$gPZlbCi!e&EqBh_+?YL^AGJM|<{GN^O#BsUVL
zp0OLSwqm*2ve=%pTrWA^-V)q|l9DM%*uKQ70=ublX(vCtzhMZ0R4u5_E-*PT^aCjf
z{OCx2q}eR7){sM>UY(<A1JlGbbPWBGEH$Fp(D{+rMb_&}Pzs?f!A(dbG1gL9jWm{%
zcqUQ;ESwK4+67W3vR)Q%{5Vh>%W&-Gg-emi*wq;k!D6{UR*GCN5xPPNk7yQ13H{*_
zi$JvJ*yVFrULc}CWQ&MHf!#=51T4{Wy(#3utZU9Uj+hhGVqF%+-FtL&SR%$qP=cfH
zNI6h73+!^i=KPXr%=F`gU0Q;i@Ip~94RWz!@(!63Rl6mb2BbpO6EcRUg({85qQsR7
zB@EGZCFU1_Hj>PuaKPP2U0cjzMH(HkKR}GQp(nbL^G!n(g8J$jS+#i6;DmuNAwxj4
z4b@^zLKcR)&?wuWq9VEhV=P%2w2%-p)5Cz87kaI2IZ38Wf7tW*@SZ$%EUunon+D|q
z<d&pcToTB!$1b;+x@H`D#_os<0bw+zUeKN`e#ou*J9kAlkZXgI>WR#UEZ+Tg(NZ-<
zM0yhZ<ryk;IDExF|L6bTC#&B{Tvy0z@!JAI$dW%mW+up#nX@S>@idFzoQwG25h=1T
zl0{{ybOt{V$C1#F1Rsla&lIHjNlBmLoSrF4xN)90-3;^0AAC#<E)a9!<qyt1VQ*tB
z8{B;OFNo}!lnDQwxxe_f@R*!qxoBq1LD|%dQfN6l7%~V^68i)o=CGc!PnON4BryuC
z)zp<nX<1-5qKF!eRwaJ?Nn>%+MhJvfhy*dD5~Uo|oaC6E#uGx6welpu6Np*hmS&(*
zE?S9UqSYE}3s**}(zOszBbAflU*LTiVWhd8A!l2nkGX_Bi5VXr%ICnyGE%Tc;XJcg
z^OWyUESy@YIo7yX5T%$4GnGj)KK*`@ExdnMSSiUTax4K%wml;TSN!{PK@)Q#lSo;S
zr2#3KrU5}f7y}~D-A53nLMn??24qQ%48aq^M2uqYc*&<R02q|ErMnUnAx>1Lf#n8t
z5f-}Xh>eC!89l@}kwZnTE&96!=6}!igNlLv_8x{F?<Sh<79Sx>Ge?aEbZyZ_^Z3nI
z<aWX4`Eyhp$!SEc)~q)ghLLKq=6HLLwwCt#qk;{tYTp0FU*P+mcHMG$b%_uqo4f#P
z29zwbq5in1QiWw76HMB$-|vYbBHf6*e~&xvFeHefeBCslZ3!S+Efyq^X%=hx`#W}T
zzQIo;wrUvnBj(~Gns!5cctaW{(30hP&Eo1A$K#&kalqJ$=xVw{PrW)L8_6KRJsjD@
zL}fMIz9;TxnkR|MSbR)RGdXQ6F=uLH(8e-FPnZTa+YJcybczYyp%KNWuFIzAeV{*d
zSX*J0eiBWg6mIe)A-H_;f}!uwu4h?WQmu)bEiq2il_3euFlt-^yF4dvpCOyJw7S{_
z{q8;e>#tc|MTW<F_HW*h#{;YDYwR-7Dv8yWq$OTzylV?RZ-0jtk@dw3HqW2o{6G{{
z>GVt<)vRbYYlMg_mlfTyV>})aO5rC*QW~*1N7M^664xE@hXd1ok8=Y$dzQ6<VL+*3
ztC5sQTt9l8OH_6O-<O%Dnwa{5EYuwIQZTL|I7H60Qs7)qfBPOcj?|`{ZgTS>i7lpG
zql7HE^2}{jV~Md?vqT0}*Hr6s#@!Cc<m5?lf-KN7vso?hwgex!hZ`C#S#Qrtre^e|
z{i<|@4-R1~TH6qhN2F4$S7&@bbI|YZC6z&I3sNBj^Sm~ZMTyi6(a*klg+gL%@#1&K
zBgbPIYekumQ4&HRrJ^A!2;DOG&pwv-ZMNts$(&4QNGahaM~;q#7A&fUw$?O_fmmj=
zDVB7-ljYcGQ#Q*G$uSXqfyvBXLXgE)M-rqQ3aCd4eEhbRZnb8Lx#(ZC2K+n*5VPp6
zEYwfS^~r@W&kd$2(A33NhM4Ev^T<FcN6KQxB4?7EQ}+-^F`umGpLL?~S!-tNVIs|G
zepzOWsY*A%&!1^I3n_Eyz$EYIuE1OzNJ$pq$+9`mWoF@Eaa>qibWdV-1c;moK9s{)
zJwtIqIaO;_?3Scc&O*W}g{(@kGCDVl8|Cc2C<47ak6D!9i6kh6pskAwLQq^DLV}S+
za4u&6Q5F=3nv%%bxsj!qJ4y+q4Z5le{3w+XkTRC6invmxh<9BP>bWCAFF**CY00X>
zhX*v7m_Y<bCXhlR#fhfb5HkedlXF2H=K{r;ZKyrbm4YG@vCUaQnk&$%LQ8>Y&rr51
zc?Uv}@<50Oh@i}h935HB0e#=&`a#9O?s#Adj;5+1!2bOM`-39@ZKFBA+MvS3YP;a=
zZ@(hFd&i3(eM%ORC?tf$`1Tdk+izHHFIj!^BkW><b0gd9XMFVOCk)+zeShHb@s<!K
z+IB&06yq@AqagXf{$WScG^n=5#XuMad;m8E`ZupR^hdOmEZdgd{XKr@IPM;~yt?AU
z&pxABH27Sij`zdJH1q@?Vaf!hX!I?8cZB{PODK_&$<gQrwWvr#hZdgLTKbzu_QOap
zHTC%!$6;dEPgHJVoE%aq8f!61;pWyFvVe7ni8%HoRTc%S=A3S22q`})Xn~o~nCB04
z#@3eO@yMY+uv|0*BG%ecR2+|t&e2we7<|bspNfsfRt9Zaa`5<=v9>C{_vlK3mn$QO
zbiBju?wKN(<rPON#Mu?1(?rvvl!f6DgTVO?BP!zY$f|BwG?ElMVt0e8mM9d#XD}`P
zaLe)b6?VI%URFc`%gP{}B?(E48q+S(wket1)}Y%Jd+~;^e)HEny#AU`Kl>QbT$IcN
z1<F>%8ioR$=Wvf$pztB!Lx<D`Q`JZ^e&|TS<A$yT+~h=-0^vGT@CY<h_H!0`fshab
zZkX_MEOyeSSj9q|1y$vJ>JRtG9MMKG`3Wfv))<!S^^<+ldxyxNXRo%lC2%DTdn^K~
z8lfwk4}`h(J&nKLnt{#570cy<7{5jI7JFY=m_-IRd19CfaL8qb-96In_~SqP5r6)d
zzhIimqGyeMI=RA$=2vFQrq<L|HQNbG<6W!SQW!PCTXKw8qo^tkDG;ZT<Q>r`#H>Te
zIgvsrn|3b#dZ~(Su|S0&Cqbsz28~kGrf}ykniAiuRROCx7wD!$=K@D6MjvJy-`oax
z>IA@3+hD$mz0lxeK%?<7%#fozbNVxZ2%_*6WIA;e^3%^QMHYa-`+zWNrUyE-5>$1u
zYCB)J`*H>v706G(d9ZU4_bq7XL{lss7a_`sQYkr)bLQ8KjFOgW?h=3?#eko{RCDY8
zJ0pnlx)i6yDO$OR2_u6~b8>4LL(G*2kda!=qnr~ks^l+_ONYr;WvdfHl<^WppW+6G
zQU&R*lq_RARW@Yj9L9Na_6TCJx_gg8!aN>QO5<Ed&Wg&e2yTzi8C5k%ElF-d%wbtm
zn8-O0{lLLFgtle$qZK&`LU1Udfka9ZhvNfXKT<U-F0ZbO#^yK@-2oQ@&H9=YT>&GR
zmZ=-aX#&+!FV6{+E5CdCcR(P#+fQ7tuepA9P4pdhp*g?0#7}|a?hdIkaokZ?D>hm(
zj7MJo_SZDE#Z)!!aG+VVEEWy*W=&R-<HI|Wn+Ri1cfaGCFTWv)j7b6!JZT(QEZ5{%
zD*HqtFB|OT3y%8-!W5atKva^xclho|<9d{m?2#Og9a>ukpJ*n};dm%2?Yco|iO>pD
zw+NNkcO$#IM=p}&{P}aLR#V3UlX?649i#JXug|IL8rL5nWRfye>zZx5z)vHp)i~!E
z_7lQtf)Ct29ysiF7^C^<(@zjuVog;P@?k0#qY%)u$R^HXiId|63Tr!tk{YSBMp8sj
zGAG<Puw1Rs?Fy+h>-G$D-0}GM04Z^pdd6i-v)UkZMV6|BoQRS~?!!dvcId^ru*j36
zd;JBEZ(ec!*&nif`7syD;NGtoAHKm)J=5`#>A1%)*05+n8@zSc%5Z!0hQs@JsA|QF
zpZtu~)dhJ>B|M`pmek-Ty2l5i8%mp38cfs7HXg^+?E#HeORNx7BGOuky*j6=?2}h~
z8b|7;F5Qe=PHRGxHfoyHLwXK53c+K@rTB9rNeS$1|J24{+lKCN<oNi2ZR#S}3KQA&
z#AzrI)zK5&gsmEk(x@DPa$=l(QBo&F0fGocnjE|LuV@xaHqV|DW560)?5i;nr;Z%l
zj5IFoz?=gy39Kl#Ht&WfRO;{RrL`%+F(H61@zXg?$e@UMBKZzCjf{s5BPEf{-TPY}
zZXfx{$1iDDXZ#oc<-fvv56m<=@6k$QEBn+TXevuJ|NXQv<P=E(q9~-ctSBdIN--XW
zp%is9Z@;o?#_h_dQNT%+nC8t2W><kK%5*YMzI!QU@#D|;b7c*yW%-_vN{Zmo!Tqjd
z-;IQM_dn$fo(9+DRG;QkA56~Ud{N9pVJxJSh-oe+V<gPK|1`eHqLAoL0z&T#{n5GU
zsbI3YpoYa<lt^91jLc34AHdc{q$ortr%*ZzbEwdXaBzCwN)#PIoJR>tmf?JDbh$rO
zaf1}EebG#ukh<C$w3K)kX7JQJW|%F9r8s65IhOb2)SXaL%z{BNzb9q1;9V+RV-0zJ
zZQ^M^i4iFZ^hX(uGNr{oy1-xj{B!=nKRic}3gtC}(UItf5}mCr(NFZp2NuicxUt8N
zM}!F&DokbZ2t)>>6h<2o;B&^9w&=)`BZN?rOtnP`1tdZ!39{SmWyzb7mXreHw4>6R
zW^sl#6~RrU5D7WsLLi~2Y=hP{IiwO=x4Zd%_vjC~p+EWCf5xYO_fM&`;`9IZUolQM
zEE>3Y`GWAlGs4*O?)@8@CXkIJrwX%J6PpUVTCv!ysAOgu9s4)0*njawY2gh${eDNY
zT=VRMYu??wBZ)}AJF>rj;NrPJC`0UfhzX%B591pi-n_=ez~zrVLu<pJ&1}DLj3MGe
zDydb<QCW*In&aa`fiWx-&PU?V7fRJ;QRecMA$2>#gJ!WXAPdHrA#wly6?$_9RXLqU
zkr36AEEEAvm^%D85<+AQj%rcxvkHZciMBQf1u;2{s?oN>JBbScH<XsOk)nipWUy6*
zv4y36${n}Xl2a-KK8lA~N`*F6IcZ8owOY|`wzzSm?;dfxdxEYB(jtU<N&zE9WX@Ez
zhH)g^enr5uXlj<*C8!V4l_K80V)6r1cO->aKJ+Q0E<Pe`E-}@z%;$wB=EP`PM!P@-
z$NR5-g&%jAc1!OjY*o|NEy-FG3#PH3^TUgcN(IMs*pXAlRu%p2f$r`Ncdx!=QP<QT
ze?oMQW7icIi#3$KN};tGTNNdEjwoZw;URhqMRYw5BjqGvFt*I~W6Bt9sp^(2B_ap3
zv5egTWh#8PpUGf`4D*Z{T)!iad%T;<ESmD(X>Gt4k)cZoM5aIPz*_2d@svatQ!31Y
zsbjHNV3kJ6h)|iVAytyLZb&Hl{`Z?eI87s3X2$&kZrHJ0SdNdc5g|}nLnh3im6F?l
zsp*G-7$>%yCH2__+8X}T|MZ_1AG?%PRaLU-jmB78EO7G-yfEHRJ$F%Tz*&%0C>(w%
zN*s09;axd7DOFqwV$QeEMb1C#8%ozsX<2d%gqV@Wr!0h-vo8u8zBu|6`7}xp#WyZ;
zKwHmprMOtvJnl!{JRUjr-*$dZc><?jlVifWB6v)xV0#gv*det7B>gx&&B$}kPicc`
zw%(PxqF4osYoWC5mC-X6mnWuyK+J0SIqPA54F#n$kU}KIl1g}Lsh?t-Q=k+uInRlJ
zbt%Yv5d$itQMP#Nv(WgNAev(F>n9(PsxTj-E90OL%e`WhWC&$MkY~$<oS&PNr3)lP
z;rweor#a5W%{V%;HKoAI88`W-u92+`$`+h(8G#fB$zT8CbN=3+{T+-c8+y|;q%hGR
z?g<G@vw-B0s2NbSC;Fmg3d4@6FJ)fUtjjv%CsGR7x<X0|Gq@~-NM#jTTSOLAb&Zx4
zsXa&Qs@SKKLn_Vo>?4L@htHZgC5{gr%37lLEY4q$b7bs}<n%}yJCsQ1YWcnH(H~R{
zJpcG3DqC@X^RC3jCc)b;zahlH*+-v3v!#*^?s(7Q{0wU>``rT!J#Af~tYPYp><<Sf
zpZW5)ugEE~tt_jyVtcVAOGUe^2{}_$hQ>;QG#qaCWH*9{bo&FUY7qif%a+xmCJjCN
zFfc~XG>pWYS#8g#6h!A{2uWgp|G?_}j7WY;UJPT;xz@b6x?q%shgV<FF4r{I*NA>c
zTeYaO(vJ2qU~Iz_2Hdb`yIx|n#ndZwE)}v!seJH^#a1nkw~w@yVY6A{Lz(CMp=TOK
zw5@U8@#@_Jl{H*!mKa+^F-a$MRTEr+O|;8J+06Zf+fN+#JB(3ORmFJRGj<)#ddbuu
zsOlxkS}LWm>rGLn#)L>^HY;sa@&~34tqjqYmS9c+X>!SC(1Ec(;QctKo|P)xUcIDg
zS4Ht^TBNBFwknaF<DRB&@Mo9AX(C&}!{I>fBD$)nj3#?ughNVUOihlVoM?)y7e5Ug
z4|@&|_a%>@>#=o%*<R4DHVpj{+<>jB(lSnAhEPQ`2_*P&0^)pffvQ~)T_A{zS+q<s
z;--n+{*jy#=OSZlg&)VFQQ7ZlmMcVx9Cv$C&NRzaIl<compo#&z{Ql%bxmE4a>yJR
zx{fFnX1yV%03nd2q^c{lC@0ORV4}=dL-1H@O9G-uNC6A22r1*q4E%l<1I=Q=a=Ahh
z86Ll+n?{6~czk@z^UEJ$8%-k9tQKX&BP%YiFUUD@dmq{RBTd!tPyW$Arm8E9)i^ie
zL*(;c|CYY%*zfmPQ#u7u{Qb-`Y$*!)O^JjglY_(eJ;N|DO=a__tOOrWX;!x9Q=47R
zJos|rd`j&TMV#iGFRcDkjbIEy=|YCfk&ptySZvcEaz)O?8l5t%7M6B(&DU=pcsz7Z
zOn*4lAf<h;lq7`mI8Q-XIVFTK2oZ3;6qqLzv6$Hhr=M2%2jzQ%FEjQc8Y~?QITsJc
znECTqlJrsrpE5D=6y|gqBb3KDTl*+o0%i15x}li)2N{K_^wT)x6qqEG##EMi(J&1Y
z(>OgjGMJqlMOc?hOm)#NKD{?uNR%j6Li-&{;i)rnnk5;nNTKMgLijckS*v2>)w%>0
zoxWboh2F1!`wf5f%P;uTzxy+!(m3xRMr^a7HZ9}WVd^z)wML|(eUVj*(JjFZNM*|F
zEDc40IiZz6WJOA4gs60jl)iL)W!W=?QYaG<sf<4=wIt`jGz^SxV(16lxX0QRLY8i-
zwk^RW9{Y)J-W<5v1YTY((Z-U)zXKfc=YREAymS%UROq^9bG|~~-}B{fzd=9yl#l<t
zpRqn1dHmw9amPJpAAEpnmOOm(Yu<hN4OOjq{^5rh4|n%(S#39LpI=jH!P#;JWTs<B
zKTTYoU$EFL@b@>|zkW^MJ66|MG@CVcdq%asAoWM~-+amIfAimPd49(E`HJJva=hQ;
zMMO0k5hdD|O?rRqP$~0lwFa#?9F9ER+~8scS<y_Mq3=m<q92Y_;|8lVqjMbZ@5#Zj
zs2g<3^wWSA1<`!Ge^0o*rvNlT%f4+F*t%gl?x{@8_T@`Rf-ishTSA=B3Yv9G-EMGG
z$+9-KDpkE7i8)~G7B@NEI5ABVb&P0Pl)OWCEHu4bJ|v6f3RI?Ev{ZFVnEaD%C%6tj
z9gD(PmTFhc_97vRT3eWg-~-Ai8mkE2v)}DV(o(ezCMlH6kP7~Izu%+l1y!>`Xp0+1
zvh$>LPj(}DIHE)m*Hy+No0|1zRZe?JkX+elnI>|WNJ>*LSLH^Go-~Wfs%D8WC2XXs
zm$aJ;QV6(lAXJ*{2f{EQLWa6xnml!_X>v!L_NDb-DMC^OgWT^+wt_V17|^n4DGrAN
z{lkvYdwfbP*IT+{&v+cEtits@<KvF|`yHbTRBeT=G@I25trXq|+~jB$Eo-BTy(mMR
zM*MgrC4qMnQi82(Y-Nb!aZXRGP)dL`C>s1U;j0>zA^Qm#B9dZhlh$IU?*|U~UA>N9
z{VV)pi)lAhCKIO)Wh%C3A8>!X<GTF`%Bms=b0dfSdxCR?N}2amLSa`G7ndI*rDk-+
z5uml^$De#e4xZ>750Ce}di$2IUcDpdNDOIC>WdgWDh!YVVf0MLfoY0Np<sZ0h-jUt
zYf}{DX|@(hL5`W6<^Y@!=hJne_z9uN6jgr~uvluW76`42dnH8fZ}&7y$JurRNODT0
zIL(4)Q}N;RbG!?T-jV0lH=M+Hf@zxRuDNV#C!W2OC7QXdGC8|Bx1pSz%h%7ddr|D0
zF-E2-(#*OFeM0umn_x-_AHtkfAioV(q7)`*O$q@hT?A#+ccz*7NnfGn5FjB-5td>h
zI+ntroX(39zoZN{%LZFprg6f%;vmpUqcuer7H8)JeCIjwl*^!1(P^X^R3_$dqVgJ+
zM;GI)Z;%pg6neH?mT}4n)m-eIA{A2N|M{Q(=VSqY@<%_0>`ISh24+cFoN8hYp)v$w
zoQTntjzGOaCSN}5T9t&s0Cm$6!UQUoV6NT~@&N6k6tX!{W*S4$Es46=LbXQg8f$89
zZr)KZYBrlok`U->$vAd6j+~vr<<%MOc7?55#P0EX-J?IK82H6s{w1caX*OppUw%Re
zp4fYG^@vW9{QeE$_J-lzD^}|zV%*VY$KFSN`|Gdx^rxR7w86A1KK)x)_?$3O5XTN|
z!|~=l{oOrhAHAe*7WB9G#KVEbdcoz#pR#`O0a+>@Z{Ct~WU<{~#HSqlj`#QP(Z;d4
zenxF-l5J6oC8^ut#v@^JY%aEt5>rTI*Ad-F)7G3{yr5pLi2WnlC}JqB;_3b^xoOa=
zEinXy_v{{beDS#B;`$lqFFzu;3pBQXY^s)r-41CMtj;#9)@z32k&9;+bo(7ch^*_?
ztke}~3FL^-lD4%4y&NNM#(ma{20_M6j%k|cjz<>jEzS09KE)1YbJ}hJNzbu-F9j;1
zw4F1u3pm7QNkAhbWH;cXFZ}+fv05^X6V6Yl{!lD2Ig>)kB^df1KMqK1n1+trA2~kU
z;N6J#4(}&ahQ)GCvs_cRZOPg88DlkB2qxzU!s5CKhDU6>MAuDO)I=Et>ALvtmrCKC
zqZ^92O9hWkj?^b)h{WU(w$LZ<-oD}D?2Kk9$x`5*D@9St#A#x8^N#2o*49OZD+Lir
z_wY!+J5X)6G@DD#KmIYz;Mw22!5{89Jlv3j2d!DHRxDNv1SOI-*_w8{z&Hof2%@Bm
zW#^fO9xVkZg=y@}rBCSOkrKR{7v2zvF+PdSqIY;*LSABWfGF+Dq-Jfx@9`xIbufn!
zkuqs2ihpgFEL4q97AXyJ8VO@ZB@?~xNKSHnQISQ%Uw;0c)uKUZgOZwBMw-STMW#}Q
z<HN++^@ji8@BNhf<HY~<|NhUof7llRtw`utqOha)co$|Gohrq42In)`dtAyib4$6X
z_!BYa*&bL3euZf;E6-(_V~3b#ro@1r;XNq@f)5OnBc#BhDj{0JSY#?}kDLXIMa{)o
z%iFtxTRwRVP6Ec$CSNRZs)UTqFKt!X`7;yoLEu7o+T==fa{0<U>kmpKZlUv?2!=5P
zl-7kXNYPZBcmrZq#7n6_%{ef$C%>#UMV}$1BG2MR@kG57GxI?}EaRAy2vN+NIdht$
zlCWqiQd1HXoh!wc4iP0KZthTIX1I_#xd&!K!|7CivJ+B9MyFImJ%g2$RG7*XZq%$j
z%BN&3QN*l5ByuVu<NwqD<)88X{h#tD|K2qhXXi+5=nn6(wjugV-C6{Oj3K8Hvcs5~
zBt)?+29L}FtqkDsQ(v?R7~<ic<N~T{Py&X-BX!fFR7J`qlG#=ojU>s0G&RX9uCK4>
z@rg&73XuawCN`^z?RG<3)d;20bGqdBxqeVFaC8HI@#lZR-~N-o%lV5dZr|T<fBQ(g
zUi0-Azvlg~{u(19uIotg3f->2_cS=xt-+5+hQl7&Y+0XQG4wn9%^UPI0+PXbUVZbL
z)w36fs}<wno~$&>c0rg%#=H0Q-LZtm)HP$>f-)@6wj3YsIov<8T&=NHjnfVFa)YWQ
z-9zNyJt;Wqrp0@QQDrVxSq*+9jgqnNkV3QH9eDNCYo1?R(423P0UquTq~O@^M|S&>
zYCYlKzvJa+KVkHl{&>KwG@@Se*d2NJ{A=!BzeN(zT5$E^1*TcB-ki}tz9XlI)S8Ez
z8-%vBwjwys!@KvS7};K&p#|jNp2TuR!J3e{5Z8k1nL5X~-$5=2=nx`yZq*^&Tm@^!
zm>7McYB$9mBVgD+5T^m5CC6Nfkb1GkR5ckzh^`cT&e8Ap_#CiO7us5IcsCWmhA0`|
zS_w>T%Sk^|Ks_QOMJ7r~k_IOm(xO5UF-=QGkY)#kHU^n8-c7`@Ba8!S@+B!SC9Kq>
z2)SNi*9&A!m^AUTAAiVZvjHL@Q&PO9ffyWa8VLxn4Tr8Hxe2Kxrd?p#nkFXN^@^+O
zE4J$;%cW%1Ry@4XWIv%qrfL_(^Q#ooG$Mrpr4e?Ctum-%dHhy04kMMV$x2{tjnb#Y
z!HftFY*n-kreX|{oafWpFkqrb8B5(XMGWR0d5mP|3FGg1G2nWfJ3>xOA&`@zk_ub5
zbBItTgg|gTN<m73rmbn}1sPA*_5A54HEmndO&;ASpe80s@DY<Iy13`r%a2&Dw@AI`
zfAt^zJ^tDM`@dp$7|;}-IGI8dJc;F`D7p42lVimD$lww&x`MJj1sKhuM(+xUr*H)X
z$c)eh-5X93yUJ5LPgU22UO4MRWT22!7cJ6MGjM6v2;{{1w&gfXJRXOcA_!TO*yI^`
zRIGrx5c`yrSYt3o6GFfk#fe>ilA9W>pXi7up`U<aTaINkcxv50ML^HG1Ar&+QHG>v
zh><F<iB!dc_}#&V(5jqlXPD55Re+G7OnDE;5a;Ar0f<;M2h3Py1WE{+O3~CU%hvF)
zAGkkwLMqOQ!ZFAwt5ERm_s=5C5^^MFq;jINX2ueyr?f*|SDbPeaw=MbloC=?0A~bp
z6nF>!$$$2*`4|5}@xS?B|A+k1AAUwA;gZLi8s|OafKibwWI@G;u@t!0Vr`3@i^p-;
zfnHIy8+>#qtI?|!+N@AHlKeo9#d5370%JTv%!~#%QEOW)mC0e1LCK7ki8$>@F;LkC
zRTZt2(1sj+L1TZ9>xYG;3(4dD!0WHRL`KJP>=~vbAAIr|<1ljn`8$k^tQIxx(jwx?
z-z``@zoa`J`Q}%jbMfpsHh31B6=Abv_j<>8f5+5KNTvDuZ+?!J8EZ6ymfY>`Nt37F
z-7!s$Bn9j9Gg6ks<B`@X&Q=R5GB!tYf8@(oU+~dSf5M_tg=er@VX|Xc8N86Riv>yu
zf-7wud^y=2_6MrX1uh7>lo6)l<~Lt+czr|LS`NnpRc(1ZJkYqF&9Z@*sgxr4k?cS_
z$Nq5S=9_z-Kfj<^wls?^&wl#1F-jAreuk1nj8W8WL*3Xy*;9te7t2o=j`N3HljoDi
zalbDyzQ&MZWE?$@!x39mG}|?%Eent_6~Vb#+%_U)@$iSK0GDi~u_DsPiOxmP1|=;C
zL2!-~%4r`Z%bc!SuvoT80mJ@|=p1p*Ll-$?Ys0c_Q3&ENpz0bU1%4O^q7d7XD>TKb
zZK=!!b!8Aa6-#7Yll)XvyFw6iS$M3L^UN~R`v57?)@xSFHRJtj((s5mTa%4Oh#Dz1
z!8@ch2s0I)z<fF=CpSUX3cXxXFBaI%h$It^5A<QeHVcFcG*%!#_z<fVVd`+*5y;e4
zRp$6Y(QYrxi6nWn$V|h8P?Bo3LTW{te4(t7=dprgIv%jLE{j<%VlX#Fq!vi2$tf@n
zeF3dNvDzhp!~CFI^>^(`F{4zO!8gs4VEjDaE-sm*Dd=l+Nq@X&yInK-9xWE^?p`Bm
zi5-tDmn$~s=fw)E6w~Obs+Qz>mhB77qM}*WjKd?>+Xer-|Ly;r_csUr<-h!|dHZUh
z8@`h`n9BCCS+&He;D<jtU!zI9+I5Uc6g+ZBM3H82PA24ti>ZiKP4VVWA(eY9D`eF8
zEYJyrPza^4MpN73Q&+m=*Qb<`N)%h2D8jvUTe0s((yadfE{8w}@$}CSB3WhXx|*$q
z#j2V_`K-m5(AE@J&B<Ai62_=G+o8B?)T|RY3F}fyMcg+R12Gj4k5mfgaS42jbrw>T
zLH?;C_)ZZ~3NAU%;58zrCpX3^4RYSPix*#f`yAyAK0*krlssG4Y-`P97kC^drciPr
zL;_SP1ad0&%2UDgl=i4fk#rIk8dF?eAdtq8L|Kcqn7a-!5oNKAYLg0GQEED$`QQH^
z{-1pM*$4cizyDJ{xV}PP)U?e4*G)`rq;8gk=~zzxj0hze$32kI#!xlqL~V<ELrCh?
z3uM)jlEaT3wpk&iLD~v;yk+d}$vIHfTk3j64#o9hwVcDSBHkUy*`st_3R*WYj6HU-
zCC5N__;=tYXszP<`W&AFzxd@ZSuPj&=(&0Qnx<}8T%M7KlD|xtIKF?&-OVG*vn3ZF
zykK$Y=^yUVv1j-03pUF?pgVMY@y$DumSmxL{^?8Na3CFS5sjs-EW5tPO^)}k?|C>L
z*_>^O`8^ls8_w2k@e9|M$$OLlwQM*$TeD~kb<?01=h(W&J-kKaNK<PjSE3<9N-Qf)
zQ)?dFNQ#Ne=P$Uteg@t#E>@V)^3ls5lR}`{zk}lgnZ)+&jCQdozFoV-Y%b^z2flp$
z4c?EOU0rhd;sr)Qr6ig}91qmXa~9{%$$nxw-lL4B?|UMG)!7-#l;|Jt@nc`8Ysryh
zg)}u%6l;)oBa@FzlV=!5l(ncVQ9zCXpNm}%GSfJsl%{DUM30*eNGha&s;UVoqiqGD
z`0dBwiY-p$GPkW;n)MmVDt7N)F?2`B1qs?L7bPG@SLpQ_OcQhm{Nacj`jQ`@BvM<d
zx<<*0s;R+`->RIwBj<!4hvMH><>Z~CCdCnQBBxMdZij*4aA5uT9%Ut_s=0aCQwNPU
zHAFYhDI+2W1d_#SML&!sET~=LV`7RD*98cXvFiw9&tO$~Jz}QbTypi{PcXLP;q@1E
z`#qHu^oJg!4b^&0ObX$8rmiCzg|T&sNF7ST=F}IZuQ7&{Gvhcg4g=m61+~;gRXt53
z!Ho!Mh~0otMN1-8X$NB}MEd<M29%cMI5PAD3vJN4fVTDzk8C$37Ix@5g6lXO9@stH
zP%l^1jl!qE{<tS}J!&&i+r-olwCfF(iA>`Ws|8hM34Wx1c*AP3VB9^RW#%9K=}-7a
zKi=@$zx*p6_apm<J;%ET4qcC0EVzDlf$yQ8CbkzF-hK6ofAA+iWy0|0*Pk;?{mlNC
zBxPvZmUoYj9L7k>3A<V&nk7H}TR-FTZ|)hL!woyk3`6UtG78E<5K<-yNhX!ZTAvGl
zL`l5&oNpF%z7)Tc_c-Ujv$;vstW_y?wCZV-D`|^yPVg(6YHiKzw&7b10L+C*nXMb6
zW~yRw11P2D0H}gk4lx$Zg%l;U2#T{}?g->L8SrV-hbJ3m85PLc;gIIXAW+&MD4|EE
zuICdMLCpON!FM+FIh#Sk)lzY}Trh^rWA8bPzNGriXy>APD4hml%JVDa!i0Ce0GSl%
zSx`<Y_TwCi!$2!V1cW@rWEVT7nC%@YN`CW;-|*{SeZl7ZjDPs||2F^rpZo#q)n<-E
zj^yk>L~@!Kjy+l%l5DWc=Qy#asuej5C{+`5Dv|YKwm`cHAqKJ&gzgcMB2r}hZ~$q@
zDdDFBIYrO~oaJ+59FJ%f84mYUt4pLP2MMJl$$h_&vmaCp{Nc|&17nHO^7U`NBu<{b
zJ232zH0_d~{^{SrAC629?+8AzxVqxm<ug{cgmkEs&}l)+g5l8d+n@iOhkeKW?!eFf
z=%@6D9gP8FTF%c}OugW4ccee`9PamwQ^J^roFd1Ydu*7vT)xD013pSN+bz$x3q;e9
zvSih0&aci1SVk8}(?rgm^Or9Wy91B!9|_JQoWN3AYF8h=AWjFyF)(#Kx{Lhm@BA4T
zAO8p`6vw-FJpB4EINX2DIF9V^Z)Q%pq3t@9R4iIUW6wCdxS(0I_z*ZA9y$K`{|}j9
zeR+nK8)^wgXLMy49w%@kX&lOYODU!-==%X_Dza55RWo!C<WO$N5X<~622Xd`qjJJF
z)!Yiqw96%?YFVw<Gs(~+bSZ#j&e(c^Hu;J3onra_(l)rmBVih;ms{%9IY^D~_Xw#_
z+AtgstX50v$})8m>&t6a*B>CJf!+IW5Gm0gI;N>-xmZw1FcxGS5WPc5Fs3exPY5JG
zko-tOBGP>G8-u7U-n)VWojmR^6hP2w&HDUP^7<TgxFflNyLYb`jtAQ1iWEK87?coH
zMqq83Lwd@I&ie#&;$4mcqZ+c-j6UL~kz6^J!C@Cmw%5;z;}PE<sHGtZgFbtP-d@4}
zhB0Qm1YI?S3o!Huk(vC2j*&Vgv@BLNp-mBC2|)}cQzHq1v=u%^qH`d?)E=b_*an{k
z-Sqnn9eTKZkCvf$qk|{r$T&rmsoD3AYF5SPR7QC*m4!D1N3+}@wIObwQ8z8q=+U;s
z`HC2bInfUzRaGopZa9F_#j~yi``sPl@W^`E(q2Ae{W#LE-WTut^)ot(^6&B8Eq1%%
z(zaZD@`8uko(~c1zW##6c8%U#AgYGjuV1mrFvL6l$v^&wj4>ms6;)+;@%MgArZ^q$
zA9^mXF8I~Yf6ncp<L>T(aU2LSpk~Y7IC+qTMes+TUb8v7;^x2ntMVyzlBThwn7KMX
zr|UXmOzaLvK+J(V1xR+WT^9l(nPGIas}*%!QEQD-lF?1Pd2>^AJ0U+g3DR6-Fxv`q
zF5?<8yD##r$}j1WqBsijEYzFfOl91n<{IGCsy}V;r2shn$%tYjR%P^;V}W*YYGvng
zx<Bb8q!eHbS}Agzc>_w&)}_0990K?Ifnh8i3?pYTq4+ks`9vL*s#pf0{5_{k^aax_
z)ZDqqr3gBWK(bKe6cF?CD5UGct~lJ@^Z)u!{ssTVzx+%77yq+=!hi7h{}^RNsV{6q
z3QLM5WKY*;g@2$xiHa$A2&Gvrukfxfatdk4aw<Vt2~@Q#dW!7l80~;_6TaV5wJizH
zFdh(Dp>@mtagQI5WLbeU5EFIPkn;C)BK@FZ;Nw5|F^5CX!|elNo~<-(!{hxUmDGsI
zkz%H9R$PAg0^QaOlVkVh7G2fos-b^);Pu=0ynVgrlTR;@QgU(DqTRs8YFZ;W-)u2j
z@aF5U*!L5Q?HOZI^lxsk#?Xxcp^J@WoE(SUk$#%^_~RFBmJQZcY(BbX%!1wS$iw@0
zw3{=W_td6%Z<SJL+oBpnQ&-qUgGia(`<oK9(=IV}&Gw^DxqkU6MpZ~Dsjr_SV}~0b
zxq17RL2Har4E@0BamV&z&BfK0!){+JbIuhQM2rMKB12>`TBtQ|-n`}b{yoiRO%{@-
zZV)lzy)X7Q1X*d^I1<NzX(mGOkR?@9k<)|{g5}xS3@|D)egSN@n|V|)k%cdUC7~F{
zj;d;lCp$(Ypi&@BJyshM7S|o<9v(4G!}8*acKaMqjC)&dl6*t<1JMoax;?_Qi0BcI
z@32a;K0ilPE1IsSyLm;9j!H+)mkr}EFa?LyO_?92P;PQ5kftJz9o~(FA}Ay|2b2UY
zVG;@17*xB(UVMVwJVVw?@PT1>hn50UH$W=lNo|SI(+?91W6`!^JoNO3o@6W3@{D9!
zQm%-mLTs)O1pNDN5YvR~4h59r#)2ZQE39b<)fTzAMk>YVJpoWD6WzR#ojlIILgo_X
z>N7}dS+pyZ(xm{?wj>lfSE6fUAj%3I9XSNDG9*<Igu(xw3kdT#(!RJNhvK^yx~4lG
z$T_p#UNQ9@Sz3gK*gc}FHAZXPv7_Cr$#sht5tL!_2Xx)g_j^`#UFN^l0%IEv_wQ+%
z7Nr%JAG~CIc+c2R7`-H|HPyv)#HB)CTw^PXkeOApps5#hpB_+JVX6wHGB1B5=x#q`
z_}Pzm_wJ78pZ$bqpM1*O-+a#NH?Mfv9r-(d|L<{ocgu?pKIGl2Z>SoJsw*Ck9ox-{
zYEiMh_#g1{#RvTJ|Lp(6vesN|D}M9kd%pPXH*B=xXFvS|KKSrCuixJC^I!ayoFgAT
zd&xKN-_X<*hhv8}nomFafOj_!SS@MWB|rP@Q!cO8Y?duk-(jnkK1B3l!NtWDO|5CI
zLFCB&{T;`?CnNavufF7Ax99!M17QvkN_iejJ%vGKLiEL&HUn+EHB?5H(bv3b=a{}D
z7Nq=pp64)}qK0SvfDjNtsMDyS5C@T^8<65DxT$0y<g$yG!XW3u0LXG~Ae2tTW?fU&
z74L5LI3J#h4yDyRst_}arJ$UP(_n^4X^ED$j3Z`X)Cuw=XGRIk0?Aq9P@bbGp3B)@
zyuXY5)BneR&cFGa-||2GkNz<~`SD9q7>aGXJwqbUQZe**SS1PBW9<@U8>aq|;c$bG
zN5~daHAtyY${>}-HY<k14T^y%bm3zl$kI|*Yx@3<syd^oH)Q8AT2s|E&IO|DNm^r?
z)%Utbe^`XEgQi*W^Ebc3jh>g!FEMS6??+5b?B2Xa8pCo`vuP?StNE|~>gW8$pZ}7J
z>ub)oEmJ@7<=Z`VRbh;vsVzBbMn6$C3v|^o^b>dcJ;yOJ`N&yQbMed~azwY~iB5O-
z_!Jo;<f<X=?x~_f24C7xG9#tt_3aIBUcF{LIs_3GHHU{s+&CeXAjCksI4kpO?^(Wh
zfw;Y8IQHzjfs4<|95<&xOc^a;wcg^Mf6Q?l(IO*BRI5x|w`l7*+qNtkOWzHI;OGuV
zrg1=$5h-%6&rwhi$cq)xv>2nQ7A;L(v)}E>(qQTZf+IpoY-LGG!D0<isB6n5Ic9X*
zATk^tZVMV!N{lj8)`Cz40~%9N0V~aFdr1zS>G%j?LZnb^M-;qj6p~R|?D7&+O`L{O
zxKAC^<1NCE7;Vs1g9@JDJ?~$A!9l_L;schS{*<cSpv{4_zvXayONx+cg*6q@7T`#V
z5iKR?l0>Ft0R<rn%`c>r6-m%6RtRM=S)<QCLboe&f27~v6Z%Ku)L}$XRwJNpi#RcQ
ziD(v>?G-6{>Y(uXz7!f#Gr5EhnZ~r}c1v<SSsG$=1b5(g+i`ffFAJp9B$3(e-g9_;
zkBgaB2CSHs*;CIrgtD2Kn%Xv`=m^tTNM|$CJ_b+po@&M%7nZ(c(k*$hBpx2HxdYwc
zlOSq^P}L8)Re$F$kDzKch!7YK4>YTDj4||M&p37{0+aKkY0q*0$l|<a=to*>xx2li
z(#0}cRa+`8Y3i2B)P(WK?*5+3t82#5W2y?N3Q$N1Sw1&epsR*jgVr@0+j8I5kOF$z
zgViV_IREeplEmtY$NP6Imn+&v<1W{H@^}6iT`!5s@CX0kPxxf`TLo7v6PF*oB!z&K
zp5P92-2)$7tw`4vp9PEc1&70tAAj}%VQ|Ro`0*znv$QR*zj@2%@|xw@8Qs@6=&I%7
z42(>C{QQic{`9B(>NlUWUMyHtHMjQ<SW{zc#l_`{tMe_yfT(LsWzcHD`Smr6rbfxk
zarcOGBehj57iTn!1s}h-rt1fC7JU8c4d1+a%P>p~!%$B8NQTjIf9P;NGL6%>X@;uY
zw|VX+o_GLH*Sulo)5-ia`$s(8%Q?m3yB9@^Q1;cxDXzGDZP7qvN?{U1Kv{#-X8wJ1
zCn05;%JAa(1vmFQ#$kHWR~e&`y08qCDJ}M=AR-}9#ypKq3J%#rD&1BYBT*EpVF-bo
zB1#vX7BN3P20?L2xD@!sU;Pcg`Sq9l_$Qz8KmHH@As@fkQcFpg26}%a`B3;IW-Jbl
zl<B*BvhT@K5Tm22mSiNRUNJtr!S@|nWyW!$UN(4t00eb=g$s_R{eZez62k#$Bspo^
zG=b1~ACNxKwA=4>kN%)y;Mc$UiV%w}M|ejYM_Q$L{%ngJ9pmwV7axDfcGIw0HE6B*
z=z|Y<{q-wWCNbRKatx94^9ARd1yxh=aGa2;V$-h3QgRqP-R==L4xGPy$#i?m{nzj4
zx*lCsY%VYH-ZQ2WB$0$-akhnSht8So9HOcbG1KoK&~D(__0`OhPoOIN<e9o7(=g$4
z=GCtbNK>)>^h4I0GioHSr?1c&ro$bFSHHw)3pulFYEll2yGN9t6|rPiSC=dnEBx*)
zvFk}v;HHsdHzITi<j6T872JRQhKKZqdb8pB(@!A_4*NZF#^$CB#@dqGAO)d2(9{(k
z5L#oanx<VAx}Fbs5@SD5)lI3sGxUQaWQo)oWh=(MBPVE9OM<KsRgFrJ`-h^?)rOEI
zV_z!JYOyAVz~kF*(7Mpxa3exQq|wxiC5z1^e(1S-_lm>u$bH}OI3zy1d{MBkV}~Fi
zwPDJEAwXpd1u>_>FHcZ{encilPjF)?oNQfes7lT8rkRXFs1~{25<=ka^Pe+4z6W5@
zHViqT$gDP7$i;SK+Xbqs(N&F~GO{u>s|6{+V|OHG#bVVU3FNUOO&uaRns$knmgDY`
zyZ5)O&o(G)7`l#EU%lo1!6TI6vyZOXZkHINQNiJciCSr@rez9q+ff5D6;HO5xbBFS
z3Ym(!U0Xwr5}75=jWEw?V~F_Zk!D$_c)!P&G@9%@IR*M-&u-sSEzfaoB8?9SKQM-X
zHW7p_(bD5c5`uQMBJ@2{3hJgo#w9~{q*-m4a>EB7e#$g;G|L5Rvq8j6!jtD51Sv$}
zZum^{_lRmmjGoC&EEL$ra)0+0^{CKFbAEY64g=jW&{&JL8e<jS4~%Xk%799lFgmg<
z<V#{=v0l;b152ss4hOV#thdh~MJ`rLwtxHweDV3O2vfu-!%u$p2_Js?0d?IFU5Al@
ztIH+-<d6S-9&g|A;^j4$7Z-f;Y{BLBjIX|W&Dpl)?*5S^G|NVDv8tK8qpepZidWSL
zB9-vCVJtZ9Fcp&BSZ2uHg`zD;nTySm_2N_RcRO~E2gq<ZlsWhFmoNG2zxg%8Fru}1
z5(G+ZNV!l^#hl<+TKZ)>IjQ<j@F4iMAUL5Fb2_Q`(@ad9se-u(l1~Cf$ed!1g(&M;
zDF{xyfzmywwdDErCHMC`hH)(KyDr55N};L><^Uu$YY0j)Cr^(GsSU(3%J3l+eFYGG
zfxc-|%^4TpiYkRF%m@)&$$J?({`$`!`Nd!Vf*=3n6aI();1Brt!;iUq_AxHqluqrO
zyeaS;A0NnR<k`y~({(+5a%eT%9#aAW9tac-Qt}R|E0P~EN;3941Ofbx$J;m5b%WFu
zE=aPpkRr!_7ZJucxA(M-;$pkvIQ7KtfUKaEk!ESp*7ETB6<hlfofEs=frs6Jk3ao@
zjfmXbz_>edeZC@#%>6@8z1dKoog-5ujU9bT=*=Z^;W*qq(j7+Lzq_SgF8I+O{t@;0
z756{?YvSbCZddfjfxE{e?*7QdqM}|bsg$DkfgwfeMNMrN-0qJEvqD<Sad(UKo-jG8
z%`@V&E&A*dnwD-l;`VzOCu9oTzxgdUU;UDPe_-r-Hk%a}&z}*d!0};6?@PY)=tkt#
zbE>+c-L8n<Gj@))Rp_chNyU2EvfQjV9uDMKIvB(L07}s|H9{%6;{oSJoDXDClCrW?
z_>fUnBWCeV&Vpg+a6^x>mJ|cV7B6=Q0j15X94Jw#DMo?|n5He*OKnP9E_$r7jFTg}
zfFB1`t<WiQ_vM$=R$<pmvXBJlX>3_Umeqo~DuO*z)yOH6JoJY>H^2TjxU-8A?HfF~
zYDroW%mRTXO+6{NGMBU!$`tQ-wic}_vh7Q&P=KOP$RuR}aqd98e^nNb{(;;-qT2;Y
zEvVX6v1~3@xRk*S2ok}Mgt6QZ)=12vDHJ)8$flYXAdmJVAq7-bA(X^>M~I~(<bA|R
z&oBjy5v*!Kh!bhr5#uHGqCr;`#u|*8i@#KYZ<3IhrYfFksZdoz4gsnbxmciViyu7N
z`NBx3+cHW~74`Wu>a!P&zvn?ki?e5VVW=(WN|R%vZ7r%&RLU|PBWD-qyt{eB*?Nmm
zip6Syo)_HC7t3sx29!aoi5y2(>lNO6n&rB9?57Uj9cZ=}2t0l`5K}@aLsd683Eks8
zB9C}CkvEqNhXW~img|P$u;a})JEC`Nx7Um=kcHy$cpwOcsh2$5eua{OeoU;+&dcf0
zdz!W(WY7BS0wFcSuwxts+LfX^K9IBF+4VWyZjVqE+E!#y1d?RUKR)|2)|)eOa6FuA
zl&KiHJyLk2PJH;IkN7x7VjOw-$xj##2Sl7`7SC{F$2j!(oEiN@@)1SmV!Py<Z|-><
z4{XmbaMQ>%`ho$T99roz@|A+F9~ehhMu$?;ACA1dzNYVcLX3=~BenIsIZKc{SrSh%
z(Wf>&GySkcG?&g7e2YEKsR;C<58pZw%Kt5!y!_6|dv+BFDa&kMsWR7>0%nmPLL#l?
zS#?F%j~ox(lZ)b%PeCpnfD$y7pXj9d<lIOFD3xYmBe}2#qH{>2sid0G>17;JU`4XP
zrB19AF-KE}iC_QnxBU87UvP18#((@D|1bIM#g^fI$K(U;vPC6}wF{t(XHAs|Q-Mx(
z`#VAiSW`lgaxQriCTmcd@o<mUOI-Jk<kmE`qFHRHtmV*k_z<builOf~91q{?9{oYZ
z!1ipxdRa5<_H3@txZCf!zv*cz!}*H>*crPHLgqKW{G7M@9cp!kdiDW#-+aU1J)eE}
z99tXqhaM{;#yiw;&)w|}Iz&GD*^klfmWMZA@o;m;<TI*Xv;OE)p8fRiV$Uv!F5zyz
zM48NVm`V_bv3&XVhWGm;*T;!~Ac!)15!tcWT+m#9O!5<BKd^t?gRm@Ken@q81wtco
z;J6zY?%z|_iu(K+uU^07_WdmYe)4z$YO$j49l!p~S7af1{`?$2jZ{s|2b<?)p_$zA
z|KHx9JX@A!d0ODN+tc3eb=sI(<8JQZ?$M=3Ktf~+LV^fYqmmlbxc;`{FMt^n5C$|N
zNG6a1nL%cRhj(VCrp;?_bJ~4&Q-gigq8OB5E=WQ+p=)URROFf7d(PQst>5}SB6-)o
zZCl#5rLdOm)eX(6<L>P@Ec48~^hjlx&TqgOo{_D0T5GH=&_ZUsrIbih5rkrzC)_j>
zQ_4urx<#4_l17XtmSsc$g{T;vXPzC)IAw9j>IRv>%@d*TP$E;Yr1VI@FBAPZkYwaE
zISL^%711wLtA@6)jNJii6(RTxB|09M=NXX#_wU{k=7G&>gHRHqbk;S8K#DH!XGcg&
z9-X8l2{9m2AO(jtis&<7C1Y6g3X)qK)BcXQzoXq=@Z#kKZexkU;G)O}3^L5e4;{<l
ziBM^*Eh#oTd<^&$uv!yEq%w-g!aR;7BqC-ZNsKT%k7mFqNl}%Ib6|aaP5WX)ySb%5
z9kU<ZSOS7+^3+9^4OwmS=i_GFJfXEEnjBtQLLdU#7DQ=s7|RDj^pvX&+tmf*{*kAL
zci7dA&GRp4HWv*2_e%u-T`g%!Qn6+}z9IPBy&3x>ON>ygkyTBzYSG7v+s6l1i{q0|
zzhLY;P7n8#b#{k%m&XdYfqCwE_x=snS2w7#%I*R0vD!e)p{W=%e5PsF4C9D0CGDz4
z*(>^|xA+9Ri%XQTxYHhY>R4Z1^3-=s-GOE4vE_=0$>{L@NV~oyMbEK&&p4bADbPPX
z5mThuY*~f@A0n&mC11aL&o%@OhXa=vJKVgdC<=68DN0L9hQUqPqUFMrEMcMBpV(|F
zl+>)+g0|Jv^$JW$_jF*pxkie}YE!ZJ4EoVpGS3I9W|w#RlViQvaN0j%WCW`jPDi#E
z7tDia96H?M*ljKlKB1FfRaabYHW`fNC3p8vZ0d^QCqLr1-@a!aPTzrsA|GHwWEMdP
zXX@Vv&Oq(}<jP)$95&1k*xV7wowIZX1d0Ej=2tqeOA0N|Zin-FV~XTZ;+ec45;1y6
z0!a@3t=fXJ&>T-g4nD?2@ByXtxxFvGJAlM^RxE!A8zOQ};k+f`m$N&-7-Gr~`UHp=
z&#<ISEfi<YK|23V2v7tNlBcHw|DXT(KjGj1voHB~|MW|m?FH@bIrmR@D4l4V6=9ii
z^FU!6;^NQ1wMelz>b9n+Thub-4p#v00x?avMPTZ@UK&DTb~C!%5~q>dZ*SRLUFNRH
z_o(0LFyQu2{PkA@ZBz67X2a(C3hxtz0acW!7-=p8_ip0rzxplBreRl>_@^1~B3I9@
z*<J2XT5<EqC(Pr_`>(&o?H{RSpstJjAe&Cau15%{3c<4<{4rO5{0}kf71k)q^@hi<
zX6jO6R}DL>xViozPxrU{`d8oZ&D}33>Y6Wp`~#}0&Mo+)X+MO~tVJk8*NwEpfGP(T
z?+MGq@a{EX>@lLmq=c4&sxA;nqA}Fl9YPz1aSuYGwIl^kRW{imI1C)UV;m<;vxaFT
zdIw0Blt&l)hkK@BKKs~#7&Cvos<35=AY0djE)Y^sw>67Xgs>39!ZZw-YIW`cOq0XZ
z4R%!_ZGi|AEM2CF35idFpbU%jjQtVu@D^=r-06|&@PO5d^)_D=Y3fOHPgxku^Ggng
z6MZ+ZOdlS!0aa(S*4Uj0B&M;)&m(?vSZisTma3`=%LqZC%PJ#GmC4d4lc74Pf544L
zgpcTYi;)5u7DS9WX`BQugiJC_2_f=#L`uo3fTCW*_JXk=`1&`$VeC7aw!sQdS{Bmz
z&0krQ0Zns;xs2y|U2vXZ7#W7~+(l7{&6aYtg>j-n(X49P*fIxAQLRzL3?xw&V+)GH
z=HQ<YXX2R7hmLbbK&D9a3$ydcdQH*>N{D{RCdy+EHiubr8!`*Dl&!JVDl_O^rV;)=
zm!C)K$})JLb+l7Qvs#}Iu!dnca60w8c<~8t+|u_w@87<qwi-wXnI%|Hx8Jf^wG2XG
zwBWK?BW+DsBEc;nG6!GEEUl7Kkcb2!X}6aMX_(W3(FMvDnRAc~ZM~*H9qA{Bk_MqH
zXhmUm!~nr9RAo)oG>DiO`y&_YH4jgZ%!@;~KvaUqr;g8_tr&)mei&)nhOs}<)Gcn@
zGaa7j=EU>QKE?GTwpydoLRtj2syXft$mB6rvxJcEBhrw3Bt{1^a(nldLKjq}K^K_~
zf{2hjg~{K+_2v@qro3Vn^7pZ+YWzH-B-~u@=uRD*b;H|t@2P9cdi4S=635ethrQ$G
zYJ&t;O~tyZh~pz61?~^0{4nQlxd16<z3|yz_+7Ukom>7NoC8A0^IN_+qmCmXW)h%0
zE1mOac0O#JiJ>1Jwt{$GHO?sGloH~D3Oa^-4+D&r>^3XLIbY*`$y@?;-e{0#oHOU&
z{eyuKA&}Z+^t3zUol|B6NG*x8CoDcYAt1kqlFn8|i9D<3^SWntS_l;&1^%0V{tJHf
z=81pw@BW-GMTLl-`)}XlH*F^128Rd&Fvtj3&%eOWBT6Qc5Tt4c!oo5$40j|P&p-JA
zpb!dVH=tC>_T`TlpC0KR12G!D@4e%9It+xc5Pan6=|CZ&EL3i(Sk04!dexwf$#0@D
zGTa_H9tO0C?AA4^S}_kJWmU4dY<ZYw9`7Ge68_-FpR<e;cXv;0F4q~^De?!j+cd0d
zgO-v0Y0u%?Zx~NKI+aAFN&SI~)>2dr5y|Wv?bhJuiThIz-VwS3w|5U{rD@xea=jte
zD~9frXEA=p&lAi3k=97GwP$_ooaaeOaP{(YWYKVZd>~96*Sng{ZcR}bV(@eicciJ~
z>e)3`m#C^HC0OQx6rrCT54Ud-TG4KH%(Kg1IPbGG!<dW;73B4+kh$vP+3eN#9YZ&=
zILCV3Vyl9AafA>k1S}x{_pGV{YYWPvU~(QePAozar6#6Cf4D<Skrl+w;iN(-ffnF~
zj_!Cwlm)u3!3W$lQ@0Inbo5V0YGc_xyU7O)H&L$EysFn6A081g&^A>@qz1t_I~M2A
zDrRDuE%F#BDWZfqMU*Wm%R0Z!eaH<sM~neuG-a7r5Uth}DqH1}uJEG73J@V+Wun?;
z46?PFWDU*DON#9cvEMU3y<=WtUI#tw5$A5l;sUnJvJayzwrbE?Gv4kIrv*PW1YO{R
z<IOMg%9r<%;~ZFRF4!sk-S?wA9vH%mk|KvU&nrvQFd~vhnuutMyk6Js5w0V|iMUA8
z?9mHMNuUr&n>#6P=m>p>JEy&WUu)_Qs5s<cD2}W34qX%!^$vleEEUr@vGhksiS2ev
zS+AIfp28NqdHbG=iwo-YhNLvgCc5s(re2>J`YF$`2`o$YV?<eE>JlMS_DDF#asQsd
zLtO~gyK7d3r8^z)LSfqrnk7)IuX(w>#xElXfl&o+o-wwdY&Ilv_$<yH)%F>>Xb~wf
z4SRIeWTBeL7QA&`e|L=s7cp%?FOL~4wYa?fA`_-8nCF2)2Ob~ZbLs|mSI;1208e+?
zqpV^cI;uj^RtlwKMp*kmTpTyoFPNOi&jWETa)(hzv?gQH6TEu)lCoLx)vtd^RcOL8
zBSfMoawoDdn(Lbl^0)7pW}k&$QgXA?y!iAv&#zkk@|R!p`u4<|eb3+{Ir^91{=<je
zeE5zl@F4{F;cZ?hjnD=WBkBX32ytE~40$6#2&C2DtviB1DxC#qANV9^pcmw^L%u>j
zz=|XSNnILJD%c-Rgy7Ly<;|cEhjn>YP|FN<6Y>M85m53BGs}&G=<@ygLsP;#M?Qp{
zB~P~e+XGL^*y_AN5fH*7l;-uD_xz{-*?+@}Xa9mf`SK-gA-UXKP_M5kwPG9xY_(=Q
zK4jlZ^k-ypz%L3B;PiN)ZMGTZUbMvVNQ@F?Yos!arvu5)JbQk@;c1Vs&G))Sf2YHM
zC^c8lYx@1^tW}hm$XpwSuBR+2v{3lLp@DuJP*`3(zu;oqkbI!uKhYl*V?Xe4|HL><
zyn1<w8+v~I)jh-L*j#Q{U9NfBcRb$hDZl<Dn-?#TS1%}xK?cYE@rhOSg6E%p!RgJ{
zynB0#tZK@(qF&W(*J}=sd)~gj<&)=^c&)fS9%<^l_kZ)rCtO}!BbDUcSHI!SuYQeK
z9P3Smu=QC4)Uq45G%-?dF1UX6L#oXcPj9~AR!hdQLnyFCnHku@qwAJL-dPQ90+BFP
z0ZJ1GMVhn3Lurc@g0gOL<49Vj+(s9Iw9HK7jFFO}X_%J4>2%_B>T;`7D)j8|GM%Yj
ziD?+n-cwpdm>jYwQPSY1k?wFHks(1cMTCr~!e#@SC`jHh_D9;fAqY*Va+tH<pYW5T
zEGo*Tr7SZVce84dKvEi8wP-(~gu=VCWJE}&sVB}eL28W5A)X~BP$u)z#}jTi<pYA!
z$h3TiUe-l{P>L{5RHdS6YLJ>RkCeuckPNb9&^2!8Smuegs<MPBX6$T85|Sn*m6`Lx
zvbwsY6%yOrvCKnOM=wyXHdt$r(G$ahS`MVf5PZTR(b|%dCxnS86{afLR@n(5y+bMp
zZlXIJuu9Ty)+k%#eQjmYh2(JR=<gpXw_8@bYt(ci%_GWa#M#~8<{=A_Y!167ABb*Z
z89V0b_uE?7ZZ>p>Th`l4rcqNjYkZO{%gi)P1fTJ`N^7QRq|lPOT@n035`uQ!BBTct
z4-dC2<G>Q3c8=clRBcPMZW+51%d}9)46ZR{g%1&l!;b^nYLe@TLG#;x^_PTkp{^A#
zKKlVl8|sToOj(nnr+a*hPm%TNGCzQ%qU%qr+YOubHRE6?>y~a@D64|Ht<PHH$dd>V
zJhiPkO%r9+(5HY@B|rS(Pv{=sX6!PN#z@Mx#Rt#WJrIK9<+B~D&5mISc<<R&TapM|
zUA9yT#$hJ-k!hNdvdU|i^9V|#bl}w8V{A$A6E`<o%Bp0$zQmdwg1x@HAjZfrI-ctr
zd|22&omgG8tX2j8;7>p0U;gD+C~LTW{t7vru&!h31}-bhPyX$nQdb4<k0ZZ*x94Ad
z^^Rc)^z)K~&F5TzDKg9~hyMdfs{DZGZ`IF9oP|MW@jyyBAEnOY0TFYCAZHNIaNU>*
zhAH?w|39xz=U}oxe1Pz#kPig0;nWRuV^$0og~1pys`)HAf)87f0lX*rtcfONzW*{#
z=h0<$IQV%%o&$t=3o7TD(s?yes0^BtDNC-RbBGv;Df0TY<2PTu<8oc`u7iJn`;6n$
zUsF~q+|VJUpx?ho3QfQCD0BvD^#e&Ns=DUXFGTT37!R0Y%@PvESX31FB#1Inl$unR
z-|HIvoel#RS2vVJ$+s@2Y`as>H0Ljyn;Z}OBfISit0bSTU*@!?_o(0*4o9plh)E)a
z;xx?6F=DGcb9MEK;#yG5GcgK`FqCz{<Ld?ebYT6PUvoK6G;PIpTVqXbKUZ~&zPMoS
zHOXjpmlybP=J2$~`@kok-_Umh-F{*7@`tQ;8`f>j#np9Y@C(7!&2wZ}m`?`^BQYY+
zvE0zJstU#k?Rtfj5qG#_db&m8*{o`$7MQBW2XNl8TCdPLOLCTRB*nn&M#{Fq7=t#e
z^T;Yvw<Y!ZIb%Q4J-$Z(wklX=K~b$BfpasuDA*cHQPvz!Cxpld(P^IGoG-{hmxVbj
zFheCJMk!M8SZi7Qf;w~A$6-V$gDEOf2n14YeM<#aQPdl(wZwVA^*ypE&`ry_ZSnmH
zLssr8W3gpRa+0t(s%A@lQ6Z#Y-ajzT1IkDUo)9B0I@H50Rez*^x+lyXg|$coV#(tc
zr4>R6Qi#lp<2Z~sA5pd<jVFwf*s>u^BTLGOUl(SiN~ofxTJ2D_#?K?muxFkQ0I1{<
z%ZyN(vZ<-s8XrX-K}v}tqJ@E&`RR)UMNQSTc|9qDG9w`+7VjyPL{3LsccNb9-QXBJ
zQ5Be?#8^Y{j)=rXK|f9meTN`oLOx*GvdL=P^UWxv{KmJ;6U#iZ%nN-=EMu3=n!m^O
zaDRuJ2g-KMYPDt_2Vx3DQDbdMxy~!I({v)Gz%m@UpB#upckEC`qil&%nzC$ALeglP
zg*wsa&o>DA(+M|wChsWQHLX|^eIm>gS}F=tuqtP6A0D{6UUPN*lJV3dv|^aYoPp_1
zSgR2dOiVa8QPeGgNK@6k`RW%GR#BHF(=u^)d(ZW=8!`(_W;<zFGW1Uj!yc(LMctrF
zLs?&8ikjo$K)q?1x+e;wF}k2?FDPt_u_X#my}qPTio^bi&31?Lj<#Mg9N*K}im8)G
zv9Q|I98Za5(Rd*-(xOexsXrl1c47SJCtpyNZN}4fC*0r(A#ruF<?-oA)wGmWQB{Vv
zskndlH9vZF!{zlgPs4&A6v|hOkB<z(a5pZzXxCidtf{J&pZ~!ZP-wn+cgJ7;&9@xK
znY(>Yj5)j<&T=AKl!TP=#X@G=V@l#oJUmnWAm6)E%JX~xL|#`sZ!944i%{mOh%>N9
zf|P`1L5SROMCMJTcBRnP@c6K2Sv(2;w(F46nS_}A6p%>gpOGRv5V}*3F`Bxnz?R4m
zkRf*^q|VCf^dXB7({~g^A<h{Cof#S*U`j#+tTjBHCjQI+>Ywq?{`vpHpZ)lT|L~vu
zV+tG^!{}y07=S?0)QnTlW_Q6H1wt%nRbWIxR2Cs5RaLXNg@uHWiF&i6Xlsu9?t5LM
zztdshZ~p3QuGbaI=y|-~6C+p^Nm5ZZYX;}2Hal!s&`L5MPbh6k%S3+|Nl9{dzeia^
zRo6L8B`vRSPkeFnIqORc!@&M{!atp8)*BY}o?rgvmT~fY`gF_oa*LLRtIu9wuU4G8
z1Agp?(?}RSDdzRHP}W?u6<>Vug4^3C-rNqn{KG%x^2IY`m?%^}jrtIYb;ZTaHR)o5
zjsX$Qp)w)!I%B(K_xu$?D~`AC^ZadhNfLr-npkbFD4K@T@s$6cqCmxfck`L~E$F*}
z!WNjaX6Oe-x1bcbE?^1+!O@+Ls8fb~G?&jP+m`A0#Ij5X1nsH;0pmF355I2DsoO}5
z3szZ-5r`$vy2f$H2~=&eltXZyJqENfC|k0roBZ?rF~1$_g264!-4U50Wz$fucN9%a
zX-y6mhKN+T{T1Al125jAj3#NB<tHj1l0xt~3m{;gXCA)(8#D=}1VvS5SNYtLOx_`e
zl*K%5&Z*yXdzYk{k0*i~*=#Qfs=!Z2P=?idO(AzAT_U7ri6Cr+9CD~pt9+^+y93G=
ztgddTwinEEpgTShmz>3jQeumO1&K6*)p`TxUFWgyS(XKD6iPs)1z|kV=tyG(r@p7K
z25Isp!IVM<f27<IDcd#8dQG7PS_zVrxizhAzHuP1#%8HaDk!QIMYGH4=YEfizvr0&
zAJ}Ym7^(5I<M!c^!X%o@Pw^q)CdV-J6jf$!oc2#>lQ2b*ud#AXSs0G{BgfMTZ4|rJ
zimIwHPO(fQ-RVeO7_P6cc{m=JrjfEV8A@aXN@hIns(wzhyCIIJ+{x0Cq&0EQD`d`1
zR8>o%1%5g}GBouId=Q8v*ll+NpB)-*@i_qOJWab{e|Jk+=I&PvfwHZs>PwPH9PXbe
zn@jrPKy!VA&<5oumM}7PPh3<r!dR5Dq>xp`!A*pikV4Wl8)oM)<%*C5ttfdM56r_6
zDc3ZcONz3<jRSUdO}u}{G#oKTQB@U>kB_XIlGEW4B@$XI$`Y=x*4VOO(-;=NuuK!$
zlnlonp|99%TbvTK%?<wIiu!nDf7r8MBD>9oqG*}B1M8~f=YR4O{^HO7gkg5v?GOCZ
zfBvtDA@Zv?_w=KK6!0#u6n^jn#Pe#QIs<i#QrX)eGHghF2<j=D9S2gvvg8dZpD<OK
zB}maD5a00tQj7>CbB;kus@ia4EQiApaTYMi?<j@oyRL*tF^>kM`Yuo4eI(&&Ru)~a
z@bgTZvw%#g-1+=4PY~)%I6RLpybqYVKxvyp&gmRl^b4k_Fu>vy|HCitx%ljf|M1`Y
zL+;<aCc2LP@Q9HGmA0(5J5qvLTgKroORT)f5G0Y9r<r-2P*zeJP1lbIouOynr+%ly
zz}IgcD7awuk<;jC%8J#hV1GDb>NWLh15zR*B=7N)1D)S`R@*JzsYe(EA`38t7L>a+
zYIY3w?+Nn+AtJR!*aBS{KKpdX{uKG*r(d$Wy2k1XttFx=@$-q(ySE&>6ZnA9lA#+Y
z>VoZVhmQ+S50A8k<%^&GnDwhyRD}gEz)dV}!Y>PM8i_Gclr=0fkf^Gfd0D7-m+W4B
ziK-j?ICGQ&QRp00k%B@tc`p`$Rf?ylJ#|yEc*i`C)a{yjca1iR>2M%SGpqHAvbn_d
zd*-J*{A@7BQj`_T`N7ofAI_Y8@KaApiK1#TS}+or#)+z3ad~}12>zUE2FO&eW%1uB
zSVQovFLs%c9rAwfvMiXop=>rRhew8SfS{PQN1BqNUV&6>wp*+%a{zG3QW=r*o~;`P
z{NaIwBx+4S;`S$$5J(wF%S2oj#^}FG{f5O8kY}Tt!u!ZL3^=V&#-N2_nHQ2wATwMk
z`GwI9IOkc$nQ6I06b<!y#l_79^>&Rw;Z6sRcegkfN$XD#yU#Fw!F7-LF^5F85cqk*
zPct$lWXiHAPyFr%HP16uS$wz76qikXbyQUCyY?%Hl9EG7mos#S(m9lLhtgewG=uPx
zL*qz0zyL#chjd9u8Fb4a9ny97`F-nr=j{LXv)1$Mwb$D5+}C{tj*O;b5qJcy2Gz%p
z_wJl)831e%bg2rdB^VOFYyMkH)E@!L3al^U6>{URhVdxMIM^V8*yOE%L2Q(3?S2|+
zndufvmgt@VOIJTTGzao!MBsKiv6j|_9Ibd^{q5a<JXP=++_>w#BiKsHvrrW#9$3q-
zP>hq-R_&bHk-xu7G25a_UBV&ayeXSfGm`almbwNYtFIPmr82e7el(M*WWBiO(LvDg
zRr<(U*Pm2??Z3q3CoI@Yfdk@MO$Fm8?j9>M`XFgGQKJcyYi#9B%?uLqZ?AyVkw+M8
zyQ#N(S;=`a^qIx6Hy10q)y`wrG#i@|49_rmnScW$Ln+~p>(n~1#OK%C?&z{|SJO(`
zGUb%uuj}^bcV-Psf;rC=rL3OI`30lu$ck7aQxx%v@37w+=%%xuitCtWAF>bTB@9wv
zb!m{$v(+f|{4SSy!WVlGL99&?RVqVDsTGgpgB?kL2d=MamhOjP4>4JkJ8nwF6a|bl
zNkOo`@8Z5^$vphQl+>*)Ro_Q>crc3Ra6sAJ%ue}2?~q>uus(ce`qugQ<cS=a;DcW&
zkimV2<;DCY$qWmfhWTe*!kdW=Iw1ILScy47>ftUzH+A2A6nKWYX>1Y1bS9pI0X%|5
z9TclMtb-BLoE}0HILj*6uq|<0c5eeH{>MJViL0-M0}=}&KqypHR^sM`rWrgmEdQ7c
zHIo~;YIwNqx~#q`$h}W4bKf2$AC0cke_Sct=V@J(FYKb7R@<(WnPqFAlbs@XeB`NR
zN1+DwO*f(&48<;wG{{ep7rY`3c-G{UxhAp#kiT5{_~t0G5G*VA=g$`<3o3h=T3}5D
zj%roAshx}^hFqzso&eIFl$0z>Gq7VF$E}|1`M#DlBC&4+&yL_H&j%~Nfbl=7$D`E`
za1ty{1HmMMAsdz7bh=RMBf7(j4#2gHS0*TB-^ZN|=`m=tBJ@6}cl{sgyG*%#>_}jD
z{+k4QV+ruN4MHBQkSh#(EX?PXlc`$!2fjs`BaPa**GhNZr#Z$;!D*SvMkmS9NY9s6
z1Rog#dp4A!+#DgXXr|Nz_w^26a@}4Y0{5@pXLea9+4fdP422a_Bl2XE+>ThL_$Y0S
z`hrq5*LLfjCusv~X_2kFI&qkzdl7X%#VXptZT^Hp)5BzFIB~@Hf?h1a0w@Ncl^48A
zSn3$drz6cMj)Vk6Ol^x7$P{+lP?l7T&}K?5|LgsTQP)!Jgy69%ewakB@B}puCvS`Y
zy@VMv9KW0(LIeyk_AuP5s-9HBSuWTL`_{hao{_{oz4<(>Qnjr+$H``+v@e?QW-2`A
z@=lhPoJ{W%ZmDOS%FytlACzXQt$TvkY$8JT6A@&y#<h0YmYf}{nQf9C+OzI30%Rg(
z1z-L}QYbOZP<#7+!934RBGA()ocZuR4A@Mn`e))E+=MCISN%Mz2Utv=Lw<#}rD}ca
z^li=p(2nOMav0fS-Ws6z-lI<Rqxg|kHz$w~wD?``CsC8Jj)|O_P+U?#wz>=2Yjgcw
zq?>7))7SxNK9psWm>7WtM_GwWGg;?3ina{>TI{-s+FHkf1j3y^65BNmj7o<({99mg
zSTn-U$WX7ZZdztPj&oJ1KRDbmR#I1F(a7BgM0xI%<&~N3)Ug)<#t_Fx{rvIqeX-fQ
zR1{}$FXVB(%@5&WVzfc4Ri*<J*(#IYXl*G-C7aHrug3!CnWnz>Li2L^<Qv;-Iq?KG
z=9fJ;q!M#&7W0XIdQR+z7@XeYH6BbH;G<AY)zCC^%_x4x&fexU**G5kPriViYCJ(3
zr+}uyg6OQ8jiSQYkj%#ACDMI6S!o<*-m*_uy&`#Xy<i4A@1yFt{tGI$Niv9Kw5WnY
z-UWE~Ojs^WzGp3VT5CQeMwCZ+95PUw54+vxOtODwwu2He=ZJl%Q|~bjimI@`an(rH
z9{X6pnJr|)RE_QIR+<I6JBEt8taY)nz*bp?7w2M`KQ9T<CO<9(;P&W^4@oK8^;Hu6
zibpA^NyVZ`Q<kC;M9);pcmH#Z91B8$-)=9&97s~X%?EUGD>BY4E!N~Ps4l1*wrkal
z!;f2vYe2x$lo4h9W3?T7ZLD-SwY1;<qY)@Jb}9jScE{!b6a{&;sW95it#03?Y5JGC
z8s?SUje&z6ovoPas5eIiGMB>T5(!GxF4KD%RK%mb!if}iVLD_M)fsKb+TBy^7JgSg
zKVhnLe;ZX8>gZAas{{8HAv(sEIX9Eg6NR#DBYiQV1sISwn;;UiJ$GPOGp4T;2H`R%
zscE%0r+nS);Wz>;tQ%IV=yd#5!TiWwmQ$$|6xP2dx#2|&@d|9Sd+?vQw@QMjjbyuI
zH7Psxt)FHA48k_@i$jK*NV|3mR=vE#AP&1pJIxiM%Q;ec+b<SEK>G&gXSRr`YIu@k
zB~##H#(;*>9LlDUQDtkcK=YfPzJveklgt>(_3)o_`t}k16XdBeP=?C9D*UvmwMcX_
zD1W7%?!d8OsI$eM(`XoHQFtTD<{LR-YFkw;TJd=t#4KFDq%MC^x?pknZTh3RC84&b
z&w2!+Vnkw5PNa3=)U73KFJJd>i_--Lk8lN|!8)5ibEZWnZVF|c6F#xa%R*x>!@wM%
z5o6SIJe#48E>@5dtbp-pVJ%o@<L0To7_0vBWXT9tWp+KrjJ1(nT!qED(#k{-0pUWh
zBc9k$c>CGT%4i#i1WcAfIfI@V*O8gx$6iQ#_%dma<k^jQ)H7Q*+8+eH9+H~n-iejf
zsXh@U{4*A$^;m2J;OPV^4w0W*CqTof7Z2Z+-hVEy1J^``=~$JCWTXlIzEROhbM+O|
z=tI_hGs*nplh=^ymG$Qaso7Xm_Dka?DDeiVRn=w75QO$(6Wo$$t<(;bM`sY<kH#l6
z6)AJ|NshDq#>M#7m{Ia(Q!>j~!7kDQkhOxVZP;iq&6A;1zWe0%8<ltqX8p~U6>or$
zmGF(eHX|l3dNb2pi||og|AoqHziNosL`i%e_j6t2jA+)XY)?H(Q<F^Go_xgc%UwwI
zqqdo_)hB>_haX|kH_Pay-mTV?FEeE>i8<L}cehL)wEZw4_%1j&33v6AoLQI#gZ295
zUs@Z#;OoB7?z`-+mkNz^?`QZE$al}PZ2VZRQ~LhJ%*E01rm<KVayge$w@ab9b(CUx
zgmu}1>A*%r%ZLAHnN(_<p8hGWZn?^SbG);ah2gUl7-I!eVUq*I-{}}goK*|O{r;g<
zC5}w2=nNbx6TV%!zg0i4$82IftQ!G=&9lD~zqNvL8^&{YQ`_@6{5y!Z>qh=T3Dx>g
zbHKsc%H$KP`6Sr;fdg^b1Y7sR!rv<Rc$qjFDCsV=7MG>MlEM00X*r+zHn#nPCHY%r
zbvDO($jF!@CCN0x%5<sqas)OA)S<K6CWF$|1q*@>biG)Rwbv@H76l_wWt^p1QNE48
zrzl+GI;3~>6hSyZXUIWhX-$@g52}1T`sRVC8bEa0%c<grUfI&&+NP=E5*!CfZvm}^
z?jPgq?d9rQofA&4#k4_W%oM+V!PyLh?|chi9YA&fYDxnYmBtWxQ`PqJ7bP5HBTz=x
zE+0&pt!9j%xcwDY4L9-zaLR-@8H(V|R&pXJ#Jfle<vm36ymtfZm)BB@9i1D(e;2CO
zqj-7h98D}b^JdN0@5a`Z0}qWo#EkXUsUo?#;ti=YL2VJT7L+8Cpj0dB%&k}wNxOD?
z2j@Yn?pG?Il6j-F8j%eF>Cyv{Bx!pc#`bBeDNT9v7H`WzsFJsMEd#X?-8nvqhXyV?
zKVJHghT9;cCEulBXfh^zULI0uOw}w~PJS0j&3WnP@Yf?5ZiNIa7TkIyHH~d*?0jFf
z1lY<I6x!n<^IuIHKC!}2vs?4O8D6`8q-R>e(|YySTeKRc*w`wcM}u7MS(8puy+#V(
z;v3-NE03`7Po5s|ug%k%y~+2%i<=;xU@{3*(TE+HzunTMfZ}oXYrIB;k8H^%md)=~
z$b9w;NncY_vgZe9_vJ<_M8l(Vl$&vS5yljPk5eN8=rYrBk_yZ4BhnBCVGU1)=6;c|
zE^7uzcZrjv*72T39#A2^n*5V5V)tbP;-A~Y#Ii-GEc3UY8c({$r1I@I3RMG&OKRh^
zF_Yn4WqiGA?Oh-{ZGABD&mZ~?*o1s9rHh8nYH`Zl4XNJjp68`h)v=FkA%3&?d>kB1
z<W5MTCP<_eF(9OcJxHrN5}{E_RNk2H*rZql)HdU1FB}&>@Xywb*W{Ht#J(*9Reo@y
zO^IaZ-JroQIF_!pILyEfFdxi(oi6F_#k5w_Ra6^Z=AJni7T~MF%+HIwXl||$bt!AJ
zs-Yb)q30!ssa`tem8l2+^|=YchkzUG3@8bch`&=|nVIH{unLyuyBSd!^ax{o>r>$<
zZb&&kD|YnIGa0{G-}ElO9_DzZ(eMdmh%(!h)R)nW<EElAvq7EukmJ<aUH)(U7g7*>
z-3&a^ru65o?FE0T1v%*C$ey^Jpp$qw(*&YfmnK=2tdTFOs-)%AG*{?r3<k&~)y!T;
zKOGF%az2f3)Td)uAx#T=?*iuPOA0+#fStdPDXM7NV?1P1LLPWHsmF;3V*iV<vSi_N
z=T$A+JkM^vd@(d0r+)u8F8%gn*QLapuFyPJT74Y=+8bE$47!QoR6CC*KU`US=RKt>
zHs`Z2kvL7OtJVRt@T@hf@9dDx3pm?LO`q`7Um&V>uL?L7dw{o<%mtsM9{YDY_uSv7
z0emLRLFdLoJ9f4v!Ls+m5wvkG1+{5Zy|NL(`Ym;TfL$HGqbt<b$mMl$BHIL(Ak~y9
zS&jo8eG^z)_nt@i&*GInvl&x$wcqVgoYthS=hX@M)(q=Xr$V~OJSPj81$CK$5-z{W
zth*nz8`I->L-7~d)9b8~Qk@R<A?IJJrdHfEv<+NEdC($&b`o8SutFO*w<LqrdJY7w
z8VXA|{{p=l)3x|~2=}~%?NLdY)x{N1?!(f44H>>-SYDaZBL8i#aD;^I9I5K|$9zgH
zA5~A>h%fmGwe$WbZg`TO`fx>PpZ)!UXJBAn9i<a~maj@}E6T}8!zSH|M0$0LhCPoz
zi-PChH%IF7ssvwuD{cvIt?F2vntd3S>pn*pEy#FvgpBeh>X*xg=&+`~-3d}u`>v^J
zMWJbD9L|wNyI{!_!BK?0;i(Am+VXlDo;22{SmMkyQ68Su62NB2qzBV><|$pE*KF?v
zf6TUJ=1NxL?CSH4va`yfk|&#^bRx)!Bb-T@DOGSxI0{?E4)pGO%e_R_+7&Xik?o;_
z6=Bd&P8re5NKAA6kz!{5RU)y6Toj?}pFms!PF^S^BRzxY-MF{nT)6l4rcU6vymjK$
zxFtEH0qtg17%rShx1?xJH*X9!x(rlDquDZ^UT@&hJ6q4`ioA+|9J_zGiXjYR7e5lx
zwY4bKjVhF6DU5gHAp7*zODH?r;k)Lzpyt>{zu-8u^kk>(TYR9kVs>g>1&NwHIci!X
znsr8;6g4pkwwW|gf~N?wTYI+TX=^5^<d^Z3iC#wgUH*xH;KS|zl~C|V%V+(@3JDWQ
zlUHvTv1%ltL}*c8-I<1j8gaEeO5xr=C+ix^+;ia3<y;KTgV4s$*MzqG-Ws6NuB1jw
zS69uavhy;L{SH*usmgGO{t?CW7)>_&GAW$0y#C4cSc5F?$T)kOB`GF$kOJ3^1h!ei
zkTFC`Dx#Jq{9Yc7ldq0qY-wpxR2>Hz(qH>FunYUew+`0s`mme8{{DBM+~5<5uavkm
zlcKL|$&3qdrDZzwGLH13DQGKk<{L6H{H7WkzbD+k#$uaL9zFiUqhz8>fioja60R}_
zM}J`|8a_TIiioW<*+6Foh<|>tIPELu<mnU4w@vds>x(79v3ZCoCD5=UH4Ssd+kl!=
zE($XhqUS8h<$D1Ss@SswEGOaU;uksD#&29>mSXdFO>}!4-rPU7a(xm6&>_eoDExy1
zwdy+y&8usaZGvKX)ON1fRU^jUWa=H*{{(&G`%Yz6^-1Y@TpW2)!RQ>RS&c~}(<<Nn
z=1@oIj?>RqusipOnYFIp4xvX+gHLW#ecMrURafWPejTBw-9G}yeJx1g_=UeQp_ka5
z0e5NDumE)ncObHGi|7%KVsv(W&UU>KpSaJPnF`bU)wqd_t;{w^z}X4=O7I!cL(jtv
z;H7k>^*}QtaC~yb{=;{`qcdbRsk0@Ymuxp(W|gWdaP;98kWTGAoV>x@yd!H#))vVB
zG+dChY9b%uG}sUudbn~W!IqRO1*4OYT4aXvsItA*B96c7cXq3>`?@LFRC2ZX5#)F1
zp0e>FIcy+F(Y}!LYHudj)*AC4w&G-3*8N=&FCSk^Tc>rs<?-b?5rqg=7*VR05fRM9
z(ZiAjfl3sz1^@*i?<26yZ0Bvn{s5>HU^OrXNWQ+kcDJSt<VDIj)K|t>o*pdGi$3<v
zPNR~!eN?IM&sOH{IWxvjSMTXan^Ghf7qx#zFmjf7eu87)2vnVK=#+G<-Jbcg=Z1-@
z;UOT&gBuFgNEKV-7_4wVjlLKj^}-Rd<ba%vEy67DLY5s;oB_FwCM@v0!|TOTwkhm(
zPH4jSDCf4Md9md*vcJln=O$;M8f9g@6Z8y7W??@So_dchj~5|8&5SEB>QV^J5%ct%
ziO{Ds0kJi9FJjzc0{Wx-$yhjgn@nn3B1v$RsWS5n+)8=hkmyIgQQlhc&{+#P$oeia
zevs<htfbhp3z)|JI=ghTr&G2UtDP6#^L;K)NOvA&joa`(LG(IjG6Br4qujH&76l+|
zIj5`GYFjdV_}LD1_=dY_UbFx#%&(<5^|3?YiY5Sk4Qi5)$@vIXBe7HE$Ab`1im`MQ
zB*#)yb(}i72{c*jz-u(dLbQ$;N-4$z(}`_$^XEaZk%DO5%4C7HL8Kimi9x~-99=MQ
z>tU$6uipjBPYk`)*cj`c_kdk3Sm9*!EGa_E!nUR~%<3sy6emB`PMFx+O2AnlgC$!}
zb+!Hh=G!sxTbRQW0B&l;u@F2{f%ocGR^h>9BO?3md8nxPT8$gywLIoRPJ~99e*5@a
zXT5I7=22ri!tPtP+QFC0laWeE9Wu<_+Nb|`N6E~q==ChxDNb%vhAVrFz(32;@D)N|
ze-~tDB9NdGo-|xGfM7L+qS;XdfV7eQ?ijC}4P`o2#mnDhYtpF%?b|vz86%#8+L_3A
zKThh_`vih2h0fa~3=9(%OMG_68x6A$iRAhDLAfLxRaRF~;ZE8ndmTLK<~ISv)JuMG
z*j`>+3QtN5pf;|qg3dE3qGY7<Bf2vxh7FZ%;^0m+$uRrC;B;b+4`!5YLkmI^R_X>W
zic`s{gY@nKcCox>(|cZbtl9fQ)~UuW+wH7WM!DoRh3%mV3$Bn|`q4?N@+gPw(F$p#
z12px=LFz)Y2cEmK^Vhb@SY~puQTh5M*Oa!+dPZS7MnG=R?vO>fyu9KiNo4MWd|wnm
zOZISb#=7?R`Tb4TJ&`-Z<*$|H(?A({!rRqT9uJ^+>4U9apU~P+Er$ffB|oVZhn=1y
z^)R#4m-V@!a!Ws3zUsL13K)_ztX*G6T@N^b%{&ABH_vv>?)PPGAuLMs<5?EZFGKHS
z?(rU$LNB_GJ~D>FzehP3Q3aolyZkj6xM^NFik4~VmkQ|})f_mD<GE3Yu8G?J{c;x(
zhra^ooL-%4(TCn9-G_(%O?qI)Q}rY$j=s3mQ(2N6%D`%$Ba>&j?&zhRrU|z-5>Eex
z5Ko31uGCyc7T;V1m~E}}i-!FO4}b$zll6S+nQz}2<_*7()WzXp!0G5p&0K|}pG%b?
z)vC&EpVWraTO{4g_UxCo{LX-Znfu|TYFHmEr{&9vP=0GJ4q)-@aak{O&#dF3jG1G6
z%~oWBQ?e6f>eBGDC14mbZzqQf9_?895^_eq<B-d&A%~wRP9KZW&eZ4pxYqEE)}+f8
z^g=?UDnSnJ!||?iYLIM#DU0H<nqo4k8@H(_U3x6+Vfg;I86a5%)^n6es?bHHh#g|@
zj;c3sbcLO(#AE4z#G9EuG;!CcF%KVore@)N95tY`yVD*TRB%SKxn_b6%M%`z)cYBQ
zs>p07zq#;uOPxuj)!P1;$cv6T5kHm*H=&uU%pq4#=7uonn2wcwV?!#sGkS^A!W_km
zWYv`OAy0`ZF-_}XcIDVDEh?-~FNBmJl$_%(9$5p1YwjHM3?~cw>oLVI(xe$F)+K8+
zSP5dSa<p0M7OMC)Rr3i&g?);V8~HY!3#=#w&IBF!(izc@^Q;fnIWHQ9rxQz_qriAy
z^Q>8TSWSx#J67en8Z3nvs20g$mg~E^F#eJ{m7=W%hgf$TJZ!o3+%w(l90<dHL<X94
zc!>_djGz5$%9f=Sk4KiNUb(4e7$r*+cR}7aLr*q|kEqFfEs8wS`>993Y!WP+K{j)s
zJx=T5W@(c36uscVd<SdC)6ij=ggOZ_If#>{#ee5;9yvYR8tlG5AS&y8dw2U3VW-q3
zliDlq!&((&-Pro|aW{$MZkk=D>aq~>JVRdSWyuJ<%)%yKihPSrezQtD=20mTx&rr4
zPGc?Hv{Il+(U1$3+s$X5CK9qD2Q{~pE?|7E@zG|<Jp=H4<b#duy@akVY50)awcC98
z0=Qf9`#lZ6_^zqYz!gFHmVy(SWj{V^ci+E9wtL2I3j8Vk5c`RBZq!;YZr7pCZ?|7g
zPm?a<3Xn?`-#;Gh&0e{OZ(ZZC@PhROc?i;}Ky?+yX{3jAEIqS+6kd<aSZ%*iHMVnU
znCR|TNl14TljM!U?S}RBD<9u1yh)YRuN;iVS#Xb071W|nN_sJmw%>jQ7ktV<j$h6-
z=#)ETU1SeL!2OzV%mTU7s`4gTN5E(6hI=-T`?LL4-ohncrK<@@b{17|@e?I^Z;o%!
zr?*EjEA##$8gc#8p>*AE^kzj!Q>vO5KQRcf9t!ZilL)>3=upV!@8`dMw>F&4_a@(%
z`Wjo{8^SK3_DQdX8hP;yt5PF+InCVsAz3T*FS|S&LUgR4;8jNt0{*c5aPMK|qZIm5
zCmlP^rKOGY^13-mvg6N{-~H`Sp%(!09wcHf9mlAW^Ht^M8>VXvJw?8$YzbOC-3UFx
zP=hktu=8K(q@`UC02uMVO@D`C4cgJUbj-4DBED0F;xneP8aQ?Hn*uXrCG+%y6m&_^
zhwp)XVRV?S$T!+8cCW69utWzVqsNCd7=h+M>;7@%{hve>Ki#tzi(ayheLFueuA$f7
zJ4Q0rx92X)(s$n1oiEt6-0BK3YvQN7)dya>M_cz$8(|7|^>_-H746NfdU1prsdYEC
z!jG!bDw%?|z*xxh!s!T!b%kEOXKunQa`+6*-0>U3l0KIKvL$0uHD^)cz8DdKf14hq
zu*DI?Z^US4p|mfZ6mfgnoOCAK9n0VWUj4~-q`InUk7vkgXTmk5U}Iw51%l#zX`p3^
zUP+yF&?sVs?9&tz2wm}Ra$l0mP)!eEaPBst_d2~8>Lb*5pJMf(*ezRNB2s?)!NeY#
zlK*PUfY(e<I#llJ<b{5=$UHb+KcY<8DYmgy)9DjSR7_|v(|iZPo5QM>&JI*f>Vdy_
zG?Ib^Qbv7r8f=+MTWh+3$sF?N8mYia@j^AZ&RJARNU0_klg~J5S(vJo>)5@%aWSiF
zD+9<rT7I%=AIH>XWBMoQY0@Jh5Lewy%po>x2X6zZ%Uyy0SpuJfaje6SsT(0OUi1}(
zNbU=fwgn^+&lH^;n~e$)MbNfkV@xNbF*>#vfuX2zFt>l1u8lwcQ<s|Hn9%1W?58NV
zw3ql<OikV>nV#D_UMu7~KhEFpnK6=(Fo{ve)bDd5I{N*swls;*i<=N?63jS$i0I2G
z_I6&%d}9kqRy-wb_!O7nuUytf>);NXV<dV>z)xlH&TQxGdJN;=P5B-@fR|vVjCy4|
z(Y+&q216=-w}hfBs%f@pdV7U^O3Ly}e^oSdZIw=3JS8$*@Sa{MQ0%=Ty-4o+mszQS
z)U_HpRV5Zd(mMNtx*f&1te2Iw>VUo@@0qx)VM0yR@0JZb-M$clZG*D`MHhxiShsm@
zAfG$#qCfeU#Z9zdOYJu~O!AOoaj_pY&z~ffipGZhD6b4$pIJ_8BJOCZ@dFioL6@k;
zT3phT`L`wo`eCKCZEH~4rz1Z=3z%Qr%cOnsR>FeuV|vlNW*x?!eUEGltDIjat*le>
zpCzJsX|6W{&KZxa3N1Y#BTo)F4pvq^HrV+{X%&+f_p$QH+&$VIeG>fk&&8!({^e0n
zCcJ^W-wxI3FZaV;LTdTcL8eO{(${sF8`|ULbGtV>VQk@M9C&#$<IoZG^5oabq7Gz~
zG2{wwq6x5^@fuGGoh7v%Exkb474oXjSK66B2JBlvMC$14A{-1|crv6!+6LkMf^#y7
znEJCA#1@QK_V)cL1n5y#g9()Cm&lB;44m&v48Sy{JQ=-5dp{VxF7}&aC@%j{@ddOj
zot8iR$vwHxMmmH}YBvzwY(T#(B~uQG2c?F)KOi)7rq&i&%g=?-Q`1Q=vVO3~`6UZl
zep#-{c>Q9;xo>Md6((+-4c|V!Yc&RZ9un>o-LZ@<ELIuby~W($KN5>?v9ewEbFHp|
z7`phF2r5UaVVQ`TyXWTr`nA`2nffk7iOi&jt56vY)#D7VR26dx&lX3%T+X$BHk1O$
zsc?47T~bN<TatC%zu++9>ZzO%{5XK`mos`@eNCVBjT3#6!lIBHvRo*O->SE3Gwd2>
zXP@!H{Fjig%Fl8TRXc$qK1EZW<Idu@+^5#VG83(F$sH<%@d-<A&9i0b=MIZ|;`CZf
zkCk+^KV=ekYOv~Zj`Xgk<Jf6(e1m+3#NZ)7B`9UnY?CF70Jgw-F=_c@K^(T&In0WZ
z5fR#rRCn6dPk4cc*S}C;Y$);m?>|t{_hFNBQ}kOYKmxOWA<0<JnpRK)8AUk=A|9W6
zEL}4u)r*JHpTnh;dTHXU$6uTwvB_e8VhPtT2aWs2#f06>++8VmqP6kOjCa~loD}YK
zYY3S*7*d>w{->5P7kGib%veX1y^7CU{$(H=yioXKw)p2n?u?F3s#Di-dPdStmGA6r
zkSRe%UK_#)l4JSQ2xU-IFIac@;Ykc3EoV6`A0Z3(UoF3Fl>BbEovG((d;^ZK%k;Wu
zG`ZG=${<N=pl=GMUtj@OVI)itBuo}t2Z9;bO&e?}z#(eWb3tJUs35I~a!UKl%J7U=
zMAOE5=lCKIFc0#6Fimuru6<4e&vVq|*i3u|2OOmgNC3YuUuLdiIVQcKOIl(Oql1pd
zwk@qjkt=jzsBz>!>1V6Uo0H%nv<YvTGp|+-Y8+A!jg&Gx#@W>pneqp>U@wM4#T>r?
z)}AE7VwKzJji|73QL?+STA)q9rlZ@~u7Pd<Y0%0fK-o)u#Uz(ImBDhxbXIu63t7`u
zG1B~SoYH;2RJPy3`&(qqW7HN*!@M64j>8ftww;yg+vQ9YSjmbSfj0wx^w{UAt^;`X
z8Y==F{PS%YrJ4R^>kIZz1`80TMkO(>NlGbVC`-LriD|;_MInEE%Of-^YF4+}myY<N
zVqQe)%f{a$A~lnyc84o~o~yX0s*ZB>@U&ritbzzHtHagNt>P8;AqS5SxGPxU;|fJi
zX^e(9g4E3Bmj+CNDQe$q*0#yck(IJUQW9>A$=s^Qyuee4d1eM;;$Eoy%_vMHb2(6L
zY`%V7W*5|$U&-Tl*lAVYzIN|5@%;;C=Nd#jB=vB~SM?`1utLCJChtd9^WkzmC(%#+
z*)ryUp|zt#1IJiL{m%-0Gix78#TMzcDn!#Y^pZ-sIO~CQD=*IVrQtFsn&_a#wde7X
zb!C&B?*f*R?=N3K#cG2BBB!!8W^B^53~&YwGvw>r0tbaVi|0eWIXw`ZNB?sCVwLsk
zdW}@<ren?E2h<I0)UGjaj+UH{CpQ(l<PRqCzxVig@shAmi0tGXHo>9Rd36ok&fh3N
z{~)_QNK?u^UO(Yz2}|TP+X}FLxwK~cv~Klej^H{ugYh8wRKq`TLWy;o@u~ZYpT1@v
zW3aDgj+Su>?#fuj2#AUgX)_P%k+`-SRO^{M+e-qvWFAg#UkBnQOn|MuiSXRyZ>qr!
z^5~n3nAMX7Q4?3@qPNGSjHj>?PW-{ALz7>|E2X<?{3o273nt-qzJ=`kXXMO}c(~(O
z(a^=i`Z2KAAdFU<Q-;DC#i&vt`u)8US$Of?qpmS<sjAw)g2MawB>med(@Zk~nr@bY
ze`-pSd#C7MG8`}fsmknj(wbTfQcvc~pN|a)^@{ULcf8Qliz3JrjN`KIeoubqv$=Ht
z=3O6rYbQ$FhYkoWhxoWQiLgA`f}swD;+-BYKcuisl<&X~juUn8qnvu9GaOO|nMlPC
zy`jU7y{UifA(ShTQpHoXgsh4~;u7aK<hbGoN&Tx%KSkX1-qB+BCXUVYd#%_#b|b^h
zCuLUG42EjxO{}lH9mS_4UbZq==9BhOkcNJSSj&M?f|#WR*_T|J3V$<&7%+6bHSjCX
z{u|i_eg>@^<T+bWX_HI1Wi7W<v1hNlXa&B48O`nFT8KES3en8A=S4GhK?=*Ra?jd2
zo}w&X06Hrm8SZt$E&X8|tRTy!n&YtLG^YD$fIRa0jYV7+-`Uzl$Q?jqZnG-p1{iDx
z%qj0j`BgY%v`b={&NI&rsIH@fTpQ8ZPX?XVuf{=YP8@GYth@C9HY@eV?ZFFNn-po1
ze;8T6(bL=n$CPJ*1R&whGX}%d3<%>!%iSPe0GfCtG^nq?|3pl)Mm+x0CHp(SP+}Ir
zSJ4--OK0hH*Uj8BiVK!0-{ZHRwUpYN?Ge~FlbmP1d!fs6j_q~Q(vl}?hNuJHMDaxM
z<9f4<C4|`)v`{z#4D>MzVVmo7FZ%6zHL(@MHiHZ&TWpz4!}%p13c01(gT?rBp2)wW
zUZ{JlG|jWIC&zbb#deixw*&qK0`jLxm=$AH!fL=Ko);}D*jidixC?^3vE%kCoTdz2
zcRalmxYZe$On+LPy3OMQJmwzNM-+d7Y-OH_8UeY#*c<#P1eOK=Gjc=qHkhrKMEwuH
z7g0=3zg}@*GBL6gc$Jh50*)pmI*2oTsdF}$tEmatyXCZ<DTA*RSQ&5#MD!OQ35%^@
z>9p`TEtgujUn>D8zONA<r1I?^6~hI+TPY+vwY^=6O?58NN9Ua+ofV<6p5QnV!FHiV
z%sr!F#XY2_l@D6xV&QtN6@(Q|B4y#%uadN6VbzF*BS!<~;<fCsHI~dCsXGi*W2M$g
zH1X?iu<m+@Y7;H|2eNf=yiJ~-E*5X2($!U53Pe|~cuPu>riUpi)t4pvo^`cRK0JJa
zu528fT%Wl}-~F9=2=R+dI;)lb`EYQ(mF?kww)5j*u50k3D(NB-7{7P_{_yk5MiBiI
z>-dg<)2JWLXP>dJZ9*!N&O*(L5aOR-cLZOZw9|b(R0!$t3;yq6_mGvKA>dQ=-zUJj
zL)XTPyAN$=H>dF9|2YNbhq9g>^B?{Hfjg<{r0B5!op2ZsBKdz-2%Xs7uv-1kW&blJ
x$dZv5n{oGl23Ah5{i!d3>-@igd*rVNs*Ag_Zi?mkihqDZRq3^2mHgZA{{om8IST*)

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_007.png b/0.16/_images/sphx_glr_plot_visualization_utils_007.png
new file mode 100644
index 0000000000000000000000000000000000000000..67adeb3d0ac642f784a2f57e6910353bd4a3dff9
GIT binary patch
literal 185983
zcmYIvV{j$F(r9ekwrzW3+qP}nww;_f+1PeAcCs-~oQ*bb@2#)y`!O|Bt(vNt?rBYo
zvZ6F193C792nga283|Po5YUMK3>c_?5Hxb2%)g1(T~ga!&B@Z;%f!_JM8U+}+1|<B
z-o})~)56uw#>tVLiJOU?fyCO~-Pw(gnc3lg1WZn@R?PDv;^+T_z&gw5xPgEmoBU^h
zRts0zfPjL4{E!e;_s+lmT56#9k;ni8S92Q3R@aRk=R+|2`K@#_0s_8MHO6H5u4x(B
zL)(RnLAqL2+s(#Q$@+n!89g2CwllY)UQ?cop~}dPF5FnOzHoK+{P^wzdiUJ}efMwi
z)f;Fi-1z<76^Iw{KkW1KGR^0I5&uASjYU$2|7H3QZNL`?-2dcUURA05`9BdwT1LVN
z!~ZA#tM|S8?`#wq*!RuA7x(AKz&HQw^?<MZ|2W?V{!Gh%P6T~I|2_MD3yS;@eue&<
zuJXs}AH!kLm(ln7_YBnQ|KFcq!}nLv_x}CBC-e7=@Yh!n8{GF_;5QP;pP;`=-{`a0
z{<!|QMC-%MMoI#O$b$r^A<@BIhs{JAFug(Fk?X|!cQ;AcFMZ9Khga)HO+@hP9z#q~
zc_RqHxU&eS9*u<fF3^UJ+%HJi{3!g!P}w>Gr+qH__^2T>2n%*50LTFXXYd`9nZw?n
zk4ZVCCLn02F%n{9yD_6|QH>9=V3G0RxR)PUN4oG?%zMzEM9H|1UUJ&IxFTykIVL8g
zA-;}K$)FS{IWU3!<8UPCs3Eicv<SgnBL!w0AY4a{uX-#vvx*YR1C6Zftuyy&uT~<u
zg6!xqfBqox7<Fsd-adS)MN^bP>6?MBvFY1kim{4IGI0t4?{ijP&DM2+Fry70vRo+7
zAIF~-5UOqu&e`~D&x5*&Cwq^EPLp1Dy33Qno1q_&14P=V&oKxRP_ED@u!EO<D9+5d
z;IffqTqn=tA74_I>G^jWb3@PWoqAs^A+rg6+_wQ{z;g|{H_rxWUjuO8BnUDSL$)-1
zc<zZ(`EKUPnk1M&Cf>84?+KGu=nx$x3Qzl9L+)Fz1a|b!%n_|AW8hey-L4(^O01=+
zTY&$n;p+*=iI;)a9PgPR8;Z+%w%42zgqI}PZKMZd{YizeaVL=55SZZ9$lc#(*tw5y
z$4it1KOn&L??eR};QMXseYtOTqepckJywam!n#(wRet+zk-tMn`V#k5^|;(z8h`?n
z`h;mlZpVFo*iUzL7@sobRPE%g{^bHe9o&oA(OG!i(&{$4Tlhj6Mce@*4&FUU&kTLP
ze82DO7Jau~Dd9%!DbJ62uY?Nm92yxJ$rKy1lwlMxl7wu_H8Bn;sD$=3L|pkC`s!0^
z-|5mYlkTG4-FTmUs0HuJ>9~f;Ero?b>RCYSgw!F3(W}tX7&)^;V9x;{ywu4d3*pvj
zaVAw=<EH0&3kxjOxkL(W@Mbiba;1Nwl~3lTXEw{PFyL_Wa9cs_U?ohzWaf;lOM*&W
zB&DOBSt`g{8;OqS0olG3;{CK`VeTN%<ec`#k9zfJbIui@=m<fgH`-OI39_<CGg{#B
zigjDx69+}{hiE7d#;M&3m(gu35Xged<4s)H(Qz2)NZSNwbd?%}8@kVQmGrM-u6zd%
zo)uNDJomX>p%EepxyiU@3D>)=QNTnu=O7yD*2*y?Kr*9TSm;pkcC}^XczQFS0A#P>
zq-<Bat>z<r$K?5N<fRGY?M(F=#v+Fy-6xS_I9{!(24(d>m@F`??<ZKmv<ss(j-*2U
z;y;4-(UWQ7RpGeFDe_x9S&TTN)+5dybw0AC)?L?$`oNkIiA2m3XhHRF@7kMon%Bk_
zz`SujLF=OqpMBSbo+CVbriN;PT{ilgk7>9puUv50ul+B@gLKuC?Q4C+=IOf=EPrsi
z<flmx8^@c-c&d6{&B{$wk}PKfOW$dIkT21>>ox3PDZuc%LIOZP6nECeB!)}(hL-i9
z{{Ht0^!G9xl>W+SH?+$rwgeHCG>4T?2)r)#AlLOga54CFWW0q@)NvG>SJCX&l7PRB
zi4VYKGSYUWWLXbg(xXui#zSXx0(T;9meMa?;$5ss?<Xvp8UJpV5z}Xai06V*OyLsp
zM(?_KJ*adoxu0D#0odt<;p<OkM%FmuW2lk}1HAx^1=d-6aUwwrZeTX-)KQzs-nq$+
zNtS<&J`C+#>2Qwfpisq3SApPw>{KqAw0SkKoV5iOH=g)M%dmdijb~Z_W?u$=&j|#S
zFOp&2+DwxW(ZGRAfXSPO{jq4hUzRI#K%tDiy^m)~XIcEc2jguVTX>VmX{3*{0m~I?
zjDqpvYvnUquQl&cFIV>Y8_~$#^j@#vbS}>9aXiMF_l{kul3LN@588zdQSk!ek1>WA
z<BKIo!BI6KU8pXq%;=318zU_rE*z(73xvUEX9j)!B)m6WHe+}@)VUtge7}XOJ_gw4
zb6g3ESuj$G7||<KP$h9?MR;?WVi+n|*<WBG*<G?eGY9!Rh|wthNqgvv{N&I`_4+H=
zu<&qGT;UD?bw5tW5bbsq2<4f<EM6q|KWq%pvtX}&<cwCmLBHQg%MMOEnVc|QyalEy
z-S8n}56X*2U|ebEY*iZV4)6UBe!uP(kP;ZQPf7hTw~iLmo5}+FIpc!b&J%p)Gju_T
z5gMandwSz|c1<(bCMxTKyVO$AivfBFZbJ=lqcWHp&gq+|)=I-?cX8}Em!{8O-7itL
z@kZrVNw4o6aWE0%Bi_!1@jty9p>E32qBRiA;__iQ6$ZiB$_Ly@Tda5AQ^Bc_O0DJ;
zih}-pKGZh`K4gkPl}s3VIs4Rq3X$0eq&D=sz(NPm_Lfe)Krf+vvAse!{4?p;nMNRJ
znUQ{0Ws?PYvGmCj-~710NH&}solARFjDi~@CIj8^p+E~J!<mhT4q0QxW4U>}jwljC
ztn&(S%s`%cq#A2ZgRe@5thbWxOm{!B8^;k7h&o@Y1W=YDjj#R7Mk<5~)gz#|KXs~$
zQb8id5FjjU0QjnVG=nTz%W9s3(+7s>uh0aMknqys89!&c;BQIElq8^DDph|<u%&nP
zx@+f}X|%QUX>gzI&GuWpJfS#<*+u*O6rV?I{n<A#T%-#KnPRDkKlWyfi(I>Tvaa?+
zy@R-gskexBpsdTeA^fVlnurGjH2;2pD>L5v*%Ynpz7ie=_S?~0fXG{@$O4B5+n@1o
zzp0%n7`1Gz{#e8Q(gQqcF6a!w!US0;i+&P(M?OB*q(eME{y`aqQlFMsux6$n66^yz
z7&!%#xf9x@5+NNaR&4enb_#sFBfV|P*$an0hW;)+eYxska_kfsSD_e4Otpe|1>;yf
z%g`puYrfIk&-B3>3|mKEWA<5b+5%z>yx88EJIRD%Wp>DYY|M0Zu{TveC-e8JAw_q2
zq9-)A=ws3o9iRb`3@sJ1wDt=h#R+#~983$s)dT!|Q=Aw-KN{$1n=6Q{F*)PbRonP3
z`l)WL^b8A#`t2sr9%JtW3j7s?dq6Kjk$xZa8o#X5oZvxUjw}c-4~C9|HZ=b(mDPWn
zMB)(DhoH*eSODxxqU$GYW+2#ry7ibV!XH$&ZRzVBz#4AO4KB=`%=|8>8yrZ48^%qk
z7@(Qq_5Gx@stq%+;LWzrCDB0UTM<O1gpr+Mngk>`inM^242HrN_@mWYgp&DZL$;uq
zF$SdD32R~??}<`q;Z83(+7aUBIcdTVsh&Ws!}=?_!|KnS@uIH4SsI4%ae_AGb8nka
z6FhY=CR6ek#5hC}lft{J;YS^OWs@o~tHE(8C=~hbv4Tm&(fRk00zh_;>GcfZ!qxE6
z@49;3!4*Z`bd!>qVk>Ka_0F@?!5%(8dM4<^a4PEWX+Fm8WCc|_8+Mo<XAO7C*XBeO
zRJE?S_*ZugZ~@A-^(gkvtvVR433o`OG;1}BdNsp{3!=eCyE2rjJY;yRZCTF60VnLC
zSezGhd0b6Dlxya3f5AYc6HSs+8|y`@<NIfz8aoa}bY}DPA36FgQ!%Yf7D*D7#Q$iJ
z#F#e6K-}(K!l}o2et>2&>Rba~_f6$!ro&o*rr75<HJhl5&)eNe$9JTh!7L~PGn0RP
zNU7>3w5mx{U6cBWP%+Zys>=usRsSTF%e-CPi`W#CO+U&F*P99iVL%~;oMB<G8ltD4
zvmvG|Rk3Z@Mh8X3e3gmhq+?=`z$gq3k?O8j-K|4knr$p5gYcp;pYEMQI{|??b)8N4
zVRB-23<>M(PqxP{Z3VUQyAyHHz%3aJ0l)z-&(%(j-W;+MqOJ0v#7r}5Y;u>cpEOwZ
zP%>ohJL%>iCx{BV?A3mRwvjVQqFq6--krWUX*P-CWEl?zkqRmjxxp*}vG%YP<>;)K
zWqxF2AZ(72e$fyEkX#R^8)^8XVYL=PGlv6iguu(lW^G%{55K$f%9J@`Fh08LY*69u
zbt}~-D`r@SFsxgrLdtU^CLBxudz5!vKYaO8I&`~n(S*u={Wq5#RfYKkC8NZ|9oV3<
z>93xtup4z=oCe|n`b~hOP^i8-8vI2tE9|YZKzn+=a*>!WW%3eNaH9{&Ovr0OXVXEy
z_v<|ggJ(`I)&s1=(;3tyL1e0M=|h8c99PBA+ALA@%UB0WtK?qJ6uN8oUw_lK*%XHo
zWdg6nP5YMA?P()HVkLnF>9OVl1RON7mp!)o=$#Cu4Q@6d*G5G-LdQ95C+Tux7lgrg
zA*Aw`_}<1fb{j!%%F<a`eMRt$oCsh7;H=|LBsZG|CB(b1w!fPo&qK@pciysZ;VL~t
znFzTx39cvFJcZopK{3$}w`iZ$1JA&0UJ`gom}u{&-ohlzjEvm5DewSRb9a%zOpz)@
zN*i}XWc`7Jl8XILznut@qf4K~UH1a^bci{`qi%T{VzXEB)vfud`;pP;%+G&9MS;O<
zMwUD$ToWWbndss{ybR1yLulM`Agp!2{5v&k46GW*uz3NyWjNL`!g9%7?*U**IwiMU
z&$>PmX*$df|6PAvWaX-xnr09-p&aY11i4|?n8F2x?PWqXf(gNIx=}z3Tpr?eB?X!D
z{)sCi#ZeCXo34}t-e|Z}_^<f*ypxe|mmUxZPw%ozF5~JZf=Shu?d0aTh|%M}4;lz`
ztnx7Q!e(EpC=z%(O`=j^2q`_L7wV2F8yme!tbNxg$FpyZ&Q#tL9jNk(xBN|9wqdb|
z8CwXLdn%<bNGkGIg_aRm*ji(WR-olfWONed@!ZQi)2J%H0~sfd=yXfhl<KoFU|=Gc
zW+tWkQM^Y`A_PhjA}Sf^B*@&go4_WU78Z}JK2!K?Q2Mm#RSR=KXW+~LCkm48ayvfQ
zoc#2k>W`gzRKfz$nHpn<Hk#}x7phSbZDz-~-n1|tNCi@*WxTXUf<dlNHc^u1BA_ED
zYp!WsDdnXd@@#mhV$Zx{KX+76VH1JsPb7bI6<0J4i=Ip+W3#!BRyQO#q1R6C;P-K@
z=PE$NeWcF&D0hF!Ek5#)Zri)YtersE3Ie%;iu91q@hckJ1xZ;JMY5$DzVC7#BEMIk
z#ib5)4e`Ul+$SQTcgc`Q*qVU}AN~I#(+|Rt2_PUqMz>VOL&A8!=TDay87)s37X)gJ
z2eX$G^e{qRXlnIh+Oi*t*@)$;B^7$xS~+rqYerHvEh1|2>G>f<Q3m>75Cr~+uJ#|X
z!h!9g1t_6<5xCTAHtL_)vj%ifxLRF*E48miIROUhju^idjuj}x`DK1l=P>9zA%V0w
z<IGRN7-v;wp^%m3PO2eT5eZkaC;ed)4~m$j-6@VAm7;L=YGKxW3!*DiB9}MdH)}3G
z`=M3%%{Zc4gqhKeX+{@`AN?iOxz^PiQCiq<XQ*J`1|IQRs5Vbk17go56l_vvb61I>
zK9KhNqo^FzNw#!Yha|XGnXYJxQ;<xQcwL|<?<{;q>?!wL9MsldQ$7C$1QS)HLB%$z
zJ_mmaA~#8bg%;i{JKAI!iT5Si{&o!8RvG*WN4D8bvHqCv11VQ6!Xw^w_M7#?Fw*Wa
z<9HJ=5k6Mo4T<+bn-1)EUAr+0e2!wn>B=!}dUSkrNekg|E;tltvL4s%I7Lde3OM^r
zi*LhVzgW=!XBO=fg0b;-G@=fs?j%c={Ip2raHC8^R!&PrNyi)}qz0RS3F`qz|6WWe
z-R4>vGv!8B9a_)G>^EJ1|1+|7Um=X#A0wi0-GN6F@0AK*2G*OoYq;_l#8M_uun~g2
zT@PWibMb{p;?Ct1vagH-+BlHq_+<562URui-qLEN+R<eKX<+K@&HPw|1mJz)?~m2_
z(6LZiu;J`K7_@LI>bG+5eS9Jj5IDJCrEafBk8EVYB7&XrS9ZF!B8oYHVliwWD$VP~
z@JiMt*{#ftqX5vWTK$l83o(qPYnAf>Kc_*i5oeW1GDi|Gz^nA9A<!Xr4fu3j!qR!F
zs;j_FMZ*#SNt1mOvZR+vZ8(m3<3pK9!Nh1zeg;Gs!NH%7yl5qWn_xs_J7fS&h_FOA
zeVM8j80HUsq}x#Kghj;o_{()@>SH-pOHOiNrQ57hV+)kY%nk7Dr*Bom<mI5c*Fhd+
zd7X34=w)3bVTgA*|4yA*^1BUNi{oM;K9GbrW*1-EPga?Ol74^qgBNMJK=3o8yS7H&
zMhdJW-)kv1S4I@`0U7u6gV-i%U?7mEDw($-34;5DsY7XYrjCsB<u<|=0^0SDmx4KN
zGXCoxK))cbtB|$rXY;mLxyS<4`sfMraPglVDzl2;#U#^Wl!$sISib6Le>li=v*UNc
z@;Smm-AiIBND;Jk@b|OB8)Si{t?hC!^kZOS_I82(te!LZBhrNj;=-&o1!13b6dk^U
zQwq$G8d7u0D?RW=Mgza^z1INbE<pSosJLMA!)7V#0eo@G2*vJ=Sy>aiU3R#sgat#k
z$31+Yu^`AvsS4Sm9!2Ye>fS39{2ESTTMe3_w5l89{VHYMm~IP9xQ2BlhWBEg&j_E9
zrWv&7k{RcKGHSt7UBE1)F#$d()D?+DLUpV`Gl1@t(Q+0fT~*ghUeR{{soe1tg)fAN
zAkAZp+4mNT^Hsn#A;0&piZKU)ewpeZ#{06i%{-4&N+P8)<YKFlvpF?~nx@|HF9x>c
z$cALg8*#Pv{*}EFZruwuS|hWgZ2m?!Z}2EEPq*V(Sq7_<i1K58i1Fq(RmyG@pQX8o
zJ;~>Yr^u&-f`h~4n3X+67`4^5Z4=J2l?>v?q$9V_*|)x5O1ux^LmXJ7m8&cCCP>Qb
zUolq`0+2!u(#~P)y%MF~YquHs5nB{1qxp_=Y~!Y7XcMAY-qIzcC>HjAwJc8PXuyIS
z?0*B@oCY;}<oz!@dybW?C~oUhU!HWeuEk19abneOgsdfqkYq&d8pHEskUesrw`dFx
zsqhy3w02(6G_VvY_>5aqvw3Ak=+aXlF;gs~^~Zp=JZk5{b9`qe;qdL$IJ2eFWTM^m
zx}rS2fhX{X890je^A;KaC|q;QjzzTisN(E*d#S~F1?`e+N8GE)UAs+K8M;@7Z2iaC
z%Rli9QaYx=P!bjtd{zpvjgo6KbY*?gqB}$7_wCBO)R88hT`$|6F7o{8xxup3BoLXy
z$RP2M$`Xh9kJI?PClmNLVs<G%I`9NxOPqbIKzF<j0$UwK(|fVlk<26z>yz961xeis
z@K99j39>AGw$3|KsVzzE$**Kb#kv#;!U)$q#@fGhEJsbo3F6bGzKqULHRN;8#YQE-
zL`9SV$i|t@^|cch4@v|y-E*kl$qZ<DuDVicV+J7W36Ku&n!g^ckt^L`F%g}sdEz@j
zXh^D>vLhvu$reF`$7hRd3`)M>Q`oP|h6zHEbNTUE68{3K$1Ro;4|0eiYfx#s-kIe%
zqDnB-73rZ22+1y-p&=8BB*FN(7zm?d^PrXvG_zyYG`<9<JL4mk_fN#U_+gGm$1v$i
z=>42rwz^pK5D6l6Yf%)(+cZ!-+OatQ_}{s0s+M9ANtsa>Hl<tQLiN-4WvCYS7R_|t
zwU!LmW&6V?<O0IqE;FI-WI1jPLHF)ZI>ivCSRLl`&RwL>FC_Rry>6(hvN(m;sZtJi
zWjo|*ek$Y4(rq@d<em1JpD+y9wF6D_D17-L^aMR^Y4*~sPh|gDW%vHEx88)n{?5;v
zh!7c6{U9<)g&)7_mVU#_cnR=h($(#A#|$kRCUZ^mI^{aOA)kFN>|{P19WAy~6Ee|=
z*JT*3YuxZ@$LHj1yza7v%lpcz5tj)Hyd_TEX76Rg5zc-?h=>c8DBz6Gv_t(;w!=Q^
zvr|OF2aTJ$r;+|Gb9^XhE2O@#NECNk_@OV~_KPz31G3dRG|KynWy1Du#dNDbld2-9
zK*bz(id_mhfb1J6pMi_9mYc<s(0J=(NqSZ`O>Rr1<XI%J_}6~^(|Z1juvUnMvH3f{
z?5<4eFtl-t<Q1_^cn%w_Q9D4|RB}wk(>RCpW&C3=n3-XpnC1~3a`KDoKyiGaAO_75
zhVB=ot|hq!-x>#t0ioY{@9YDMI)Ae66o#rOZiO!(Gh59IkGnY{8PUb0XGZ-lk2e*<
z@fZH1%xu+SE$HtuautC$zoR`MQ)#ZbehIaj#^dh|?YGa89a3?%^o8G#>pSW|JlrxI
ziOmwQ2%Ayecf(qaW(8|XYj*cy;KmB7d2CU2OR=LGV~?^NjXj)+@BOqMeDJu5!Pr`N
zGK|tOo!yP*C=9o8oB`EH&-3fWDtVXcQ5T$EqCdF?@{rdRMP+l43K}baW{ihrnR4;C
zyKe@#HAopwb>}vj$zVkMv5r-7chbaBNMoasi5;izo>v6_V?)0}&Qxf3?MRNtF^;b1
zbZ1gjU}$R!BxsXbjO+!cGyjGL(_*#Pi2%Gc4waUdVeTXC^b9Wgkj@U}(QBo@mHP~;
zRY2saB5xbe&F}$9=~*8R&jDq{zeR>XM=-+tM%b9mC`A@GiZl!LqVc3x4xLm4(@$}X
zK_h|94dLY;$r672K1bxbXomgU7)XBO$K)%u)K`gar6bkXMvO0{+|W1?IT7jL;7K;4
zD7&Tsc6`lAW?&{k?Yf$+Oy*lFX?koLSLpkWOYn`lZ6|9RoJWGs`aI5$60F|=3rQey
z2X3%JR$UBmMH1~$aevTuK$G3fSd@T(%-0yoxiO5zVxU5ghnqt8ns}9-Q`Zm7;+Ddz
z>@6$R#ImJ__*YBdrX3<yTl;Q|xzeVMZpo=iMir(0I}PD*lBW;pABAG&DPXwT>w1Cd
z=+2IH#4{_aGf|V?hf?xD85!#Wo~W<?2orf!3sm0zV6d39*2gcLHktKXuf*K_13q%c
zt|+eTz%I1&ymnZs=cXu(5z|@t>BX+sa+XUM@ld%aW(t5`IKytZ0_|K07?vRb0Wlly
zW<CYhUX*$?h+^1z+$pTyz+k!vg+dK2zki_jk526CaT9>Z8Dhp8$2ZmiC9dDsHi<7d
zu*QSqU^FFkHHK9V6p4Cu#jN8ERKAihPE4FDQiPve=^Yi0jz9Un2Qio7GO07(^%f?n
zDU$DnD{+gdqk(}i(-gqQfv1AUKRrW23^PH2U&$)bgp7?Mv?SuJ8sTtlF>r)Dzr32Q
zTQ;P$45ejAn!Qq`XF~~!Q%ZQdrczchmxCG0AXFFd97<J!CV9*~L$rVA+dL%a6lk$X
z08-stwoyTA*vmP-ox=fj-qH|@g0Zp&ZYpp{E2nJtV^>x`lt{_pG{X7nQWb_+bNTH}
z(z_gC$*LP(1kVS=esZ{;PG?cP^i$sxbKia#M9R1r`a6!A2U?9AGS!)-C=Am}cgjr-
zn81+^3Jb_bJpTR^!VHTa3Wr&8*NQkw9+}l?&__oMW6ajbbDgOryL+;={7K(<n2TLE
z$r&g2+@@1xrUWS-Mcql$%-I>U)UK<DR^h$iqvkFDSQknz*dcf>kP-gO$CX5c>FVYw
zJy*Y2-6+`*)Dof_j1i-0eZuOXIpy1U*wlkjy3`Xxl?X-Am%NV~n)c@I3)#I-O84nN
z+oD-KA;|7_eauS4>cN13RcgoIpy5}K@7S_Cu$9yn<xI2W=5Ts=)_qp3+ut#)$w)`S
z$}Gq1t|?mR2>YAFA4b1KbmZFPN21vhJ`dHg8_^9p`jDwgnW)7%j+(+(V*~LEM);@H
zra8UoS3UlLiqD%Q#qDxu-TTgpS2$QM<03P2-8>~4;eB?w02H>m7(U!I!euoj7sqS9
z6ezhK<(-{JYFAoe?~ueH;EFU{Gty|zZ;6JB@0Y11$auU|S;E3VQH(pg)dMaFraq<?
zVy>1`H6W7=jf5Tom`ImCg2;|Zil-Ngp}Ku-Fp$S7Avx{~X`afI53LILuIt?%3Y`{r
zk!CSClJL(}i@Z$^$Q0qMpr?FYs)`~rdfj%&SIn=F!>`jel;?3v+(~t4Sgi_bMM2pa
zl4xez0W_W1fbjK@q7w>%sRMuo;>xHF%?qBiK6r1c?<ZZ9S9-OW5*>%zKW#lLBcp7t
zG`su8ap<h82;9tF8KaIu!8ba3#=dxMwxZJPaLY1KR~;o=f;xmi;p&sYHr+S-xqXRi
zJES9f;_c?De)a@R_95Vi#+o;zbSwn2b8BO$j22znjh*J%G8a5eYW!IjVeWc)7j-t;
zZt`#A^L?ScVo$0F!WcW(SE2)!c(wNL<3NMuPnNY1XJ~1YbAJ^}Sy(UT?Btyh(0hW9
zfJ_Cs{L`(C`r3?4cc$w(EKs`SE!pc~V>=B+W8P2Q8`bqU$g@1f4t*2+j7Mqno<0Q=
z_BPDkF=erc>d*j@@Xy$J@9-3)^+wa__~`9PnDES9Ih%@XYV-kmLQ?~4YCR?@g`dU6
zO_WEHsP#BP5WS$cHU!D5SY*mE+&DI6z+se*WSCwvHM3PI4GX6&3cg2!o7iOy)VU7)
z8i7a>x?;<!Y4UTR3yoPBo2#}4l@(=MsC<3^uvT>s;m&+e%d`S5TBk&^ANZ8+N(BQD
z6eCne>oK=WW29KSXCtgs+Y)R!Hdb-LW?T82S8l$!Z<@pKvY$;L)A8G$Mr6cz=mmnY
zi$lVp4V^C?)*hWcO!Kj3<JEDcZjD;zU@zv?pxLFls`q3z$we`hiwtzTDS#YQzoXL;
zVWn%tEoBO_n@kcnb16JcXy)g9A|SrhDI$voD{ZTRpy=vHisWKH=KE@Ha0z{5Q`3f9
zZQy(%dZ|mqQK!>!fSo{7X*wBweLcase|C#^<g+67@j0aDpGw9eL83kA$Pu?;&_sR|
zZM!5^GOG=rN@{$JBy8@<@A7>3jY6mvP?`C2xi#dY(4)=o|5S3E_qj708|aJWn=?fS
zXR5O!syi3gG%Cdbvg!DZ#|3n-9x*xY>_q)jcge;M;F=Wh$>DpvI&H<bA{P5)j9e^*
zLlb28N#3*k6WJVnn045_`-oA$i^<kZs0V4tZ*t@j#(*$Ti9HP&3{z8P!I2U>vz*4m
zt~of8N@2?m0<+C)0zHcM=msjXl7^y0z*_7%sWUXiuyH4KmI~vFHlQE2hFg-U2LKT@
zy|ubw{#?7Wfvyz2l(vy%yCQQ4SM;Zluy{OipIP(#hHL&cZ9y_r>2E351XbxjxeDb8
z<NG}&z7Jl%4Tws&%p|b2@lHeswr$S>R&;0S>;hW!M&wbmPv4XCGJoJXTvvuPBAC}|
zOiv~lO8ALi8HzCHx!MV0uPtDs(WJBoavz0T;iVm=Zz7x90lFl7|LmyMa~Po?WSc*r
zJ}n9jvqI}SGaiYCVMW*Nqe5)F#di<A8R3R?*i7jE_?e$Ie6af#`ps?LZeMKdU~j4Q
zDzw~x^;Qq;WEXg0EhuA0tM<o1Fn*kye|y2EFh4{7`}S-mb%;3`w*!>90HcZ(RiDLa
z+)?<oW<CqQL?JiRI-+PR?=}8O2IghXIkmk6C#v=fp+fu(D|C%Dgfk>%aIdA)`p`a3
zKE$=~m*zR%B;ro444k4?=`>&7uq>R%lpW5k+sPvh1e)RU>4>yz&pPr$ItJDq>=-;J
z^hF*CBw}5LK5N}5vlrHs4mE0B)Tbk9qH-q<pLy`C<B7_?<HIpR=B@Yb1JWMQAmf&g
z=~1~OHz@1u9KJB-N`+&Zue}~EIL{>G{sKv2UATE5&g5rk;^5U+<j3u^jJ^58z8&Hg
z-$0OKhd5CMr!R+RIQ5(_5qsQU-nmtr+qh|y{6myood@cB0)Ja#!Te&6Th1J%+vw8R
zFep;uJCC7_Tk#t<-IOa^4Bf8!`8N)4Hq}}Bq5bg#5vLtX5xQ@x+<*NWK;_HuM9Qbu
zc~S6Ei5IJ3-D&m0dJMBKz-vp={T1i%qKn->$2#GG;wU))MTQoJa(JSg;2|6QQl828
zbb>YS-MTvMigf~UoQHWj3-h<v{DcXCjr%sq<Ed1zCa3(FB4N_Qzw}h5TJVmJNWQ5~
zfYwgj@Yp@q-g!ht6-rO}q&3RC+4T;B-^@FmjlK+qrAJ^3ev(%k7lX}+Sgqgc2qSvG
zEm>VuU{{V5<vs&Z*Kq!#aRAN>48B$<*zl{UTLTRqM(tB2xvW2a&NSWyU$~*8$0Yx6
z{UP+*(rn3fb{za;gRAI~S1F-KCITH5KrvXZJr>;ehp2*N*|Q>FoI~cHYi-#D5_Jr)
z16#|jX3>(4eaUTq0U~YW_E#3)z}AvVHR_a{)Q;q+nqI_BP1FSW1S?E@koIcB8%jxf
z=&4vJ8B9J9v5nxoIz(7uc1Yj)X)7gK6o^%APE@#zRYCP&r1O1vPc01(6BM8oCYZnj
zC?$e&buoeqpv1*tXMQWvXW;lToD9E!$RzzQ&?Fm5s;b*+>FS`H-bt&~kiWFfrgdGM
z@9_G&++iuoEyS(HrvrWF(FtS5QU0ax@aktl(b;_u8&ueNitiXn^b8kL>2Q(k77e-8
zA}pR6e#t$-sw@Fh4G$T~lWEW8mQ3<u7{zN1@1oeeo+cw&4E$H?H(8nT=YCI3^@k?6
zg2)-8BFjnB@7(%)*%M!BRiq!h{9SWqJvtEP)%~X|0&BxZy7IDAxAC`$x*7u^le>Zn
zwaBsL-~bLnvAQUToW*Qq;}Y@+d`s8%93&1Hv|`&0N@JQbc2>)<>7%&)&zK$C28^OA
zvOU}JtS|QVjsDN!H3VWBFRJVyZSdp(r&BkX8-`aegpk``0aslr;XJzBC`%BVs+Sh&
z>T5GP*0aq2iau_Qb@MX;56Km=Uxw{f+<L?m3nLX?G!+AyU>76rG!6^*QmjH{x*a2k
zr#(8hS?g0Fig%=iDsx|*d#MUv`fe=Ro?Jh%dPn#pU+!#wwTv*f3M_fxJ>=S+^d*|L
z+2FQrR(Sn>;ZXkUCy743RS)a~9-GAB&)4ZqjL;ZaBOB0sw85+D=Ej*+s#KmJhP9cV
zH77f2H1Cd!L>4>xv(*e*N%)EFNok#FikR6sow8u@^1(bv>4waooyi6~fe0BJprG$y
zq}l%Q@-E-M=~~fTBG|lVnzF^xQcN$_a;s#e!sF-whr_cg2CeH?1E<HuPtarIVyJqX
z|5xVZS*lme%eX`6XjLe$wRXwBUU?UfW;YH*bmjM+tN3(C#x*aH<Bii|5be)AaxrK>
z-@fYx?hR(1YhR4vcran0L#GjdCn$h{M}qhgheV9*UL{16A5OIGod578FDfDL3PTDz
zJ2vzMeb&dYqa@o>i#A}rJf$nbPIF7}tT7iFCK0=H!NE^=R^VaR!zE$u!ZoIGWu<|7
zZ2!T)Zm!+EVO-}*kYwqjK3VC@&#3>|vlSjj6j$5Ny~%q5%NtB$cOfx#Ua-;ZQOJae
zV8w?YD#+x-`wqi-Vc<-IU7~s3e|?s7ggDdl2Zr^>QM*WuPr-xr6H4NIg=tMjNOv%&
z37*#kd}AtT{~e3V^`i<`7vIfDAr&H`K<g1W%huVx?%~WD94Nun{P&e3xNpM)R6Tby
zi!mrX_OI6cFAgrM3TE<>kIHz1m$P=NvL|G6llkf}frBk<%t!`0ZVoHz4PFwRZ>5hA
z07bVW?An#X6CXdv=5n$EEjG+iT1hE44j;wJKJW2+{tGzV!Dq}`aUZIgYY%Apn(D!^
zH}Id!m#w1grPks?hTh@z%1-+dOjC)Io4yK`ID}$aR1vJ>Hadq0uXs8NGk#`i8BGzl
zq*}$S!4tvBz0)a1M)<}&KYJ(zz9NUXjSe$%3dVz($NcLf^XGxfd7t_(cuOnmh-qAq
zM6nXX(oDJTCI-Wq0%;4*95PcC^Ms{MEZy`;c1>Oq?k$#f<5?t$!on%qHq!L6#%y<7
z#L{8?uxl*%d8|u>y}CNPp<#BjQong;Wdq3+EDWx+d`hN97PSMZTV7|Fn0W<Q0GPcg
zwu-}yWL8^^<yVjX3hOIW$-Fq0GTVxkN}JYQ4Bm5wjRap_f0ho+GTJ#(Ic3odHrb3H
z0DJ?x;SBvF0QpRikV=w#VXbdVdE(DS>JK?mU957OfdpI8Hj+dFFcLi`e3@DK$JZx>
z9yr}xRu$JP2n`~QXL=8&vJK<Y4H&TKiE-;POA84hkGys*jjIml!o;->)`jQeSM$Z=
zxirzL&i;5aiQNgXJIsk*J4QsSGUWo`k|Fq+JllzL;Rk?)YW)|KdqImh&Co(F#?_hO
z`~@b})a}>%tLK++Utn^8dOPGCO-X5My(1P1qTBdVna*uz1H79l&VV-^@MPCkck;3y
z{PpS)B}GwQTuc|vbJ=2@cf9mxOoKKjczF@q-_2@Xz@web4?JA6b#;)6;6}^aPr~!m
z;XxRQQHDYUz0?kia;pxj^h>xn#n16X6@lC1IP`WeQr;yRshT7`UyQbsVl~dcSH4l@
zGqcA{m(^|H8H1AK-7X2w3?6OznvsXP2&*(bx6er3H;ET=7IpCg5ERL+Kg`{goFejY
zv<>)`DeFCuJ&+L34rv1fj*p3L7vf#j4Q1Js<${WJee9@NSZR4_FRr~NkNU$rmlpoR
zn&G=d8**Ow1InyHYI-?GW(BZ<Kkd_`PS0o))yg?~K1#IxdBAVu7~z%7M*}6wtCu5>
z4O3q%8mhvp4Ga-&3J3~}vTb_Red&J1_zIf|7C-eUi0;g=1B$}%u=~1rLrnyEyqlJM
zEva=;yW`kP*g()!HD7*}`Vg|d-<@9y55bIlc1qhxhm%;<`}|UX?4=BJ7?nI>EU<p-
z>i@oe7kydk2`95Xo}2J`&gPX@P0ZtFdc{y|Pu8Ve%a^2*^FVkF&>mJ@^mEF_#bSTJ
zwB00R(-vr-Vg{CG(3viUu<>9eH`~HlV7jX|A<Hp<o{u>E>E9xRrJVB#Sj)3x+w#k9
zD53OT<)*~Hz6_2=cnYD@Kvo62W)$EGdJwrGM6PRfToq|ZQoq&!-FFcaX^#1H|6V3J
zSfMV;>$y1Mc!?7XVrp!eXEUy7y2_SlWVMKhJ&3f>@=>1c)4!rSk-h}n6J0$_^Sh%Z
z@M7`DL+-Z=_EhRnafHB0S|eI~f_8nb$!f4j3%sjnJ_2>zXti*^xp=>>)cEMmJldEJ
z`6bq<=DiLa$=POHeJgjVaG%1yU{YkubHV&*7Rv-fhRqeaojq|l*^Jpt2z`%c&p>jO
zDuf&TLdk_sL#YmcXVO8!m5;N)jR}D(yEK2LU__q20uzwsO%YD16&3MRlAdM?5OU1@
zp9xFe`R-l~30q#o<D@DdoF^(ojap_X*|6StHud=^q3D1{_srhg!Fb|I|H_{eg%f-2
zAj6TrKmVm5<T<2!DBF=`YZ@h)LDWGc4%e_|bc>b1O)$@|HXg_lI3^FW$tkM9*h9~F
z_(wGDv8)s;<M7wa4t?u|?8gG0Lkn86K;A#s)FX)(gD|8}S~U58dww7^0i9x(69f{-
zaxb~rsD_9{=Oi?CVgf|h+aUo^YjA*FuaY{Sh@`GM=h)wgBa1eXYW~>5g14`{#Oz)N
z3kOh1F|X6K%I8e2dj{0&v@ts?{CX!g50n~%)jP3p2=r7}8}S2tUSYholSy~XHWZ)A
zrU2!xWj<ILvs$iwM0JWPe=WBn*$>UTfaRm2cg0dg!nn-0zqNCwnv+Tf=2vR{2!*R6
zY^i4{`VVyAxNiE4BDbN8+33{(__eQQf{hR2!tUn)9M}m1;rE<P6~7})=@iwkYJ6Tq
z@cxfFlxN;B0P&L!zajmZ)x^y<3yd`E#WsmKE~D_Cf|P%J_;`RfERVRwrZHe*`ZbMp
zLl4q~!v#WH22A9#>?N40Ve<8m$O_n8c}GM<IM%$^4jNBJ%WK9ShDS6_o)7UCCtcX>
zU&6s!l4ND@yGw)2scZeMxmj9+s*=hCx<l(s%z=c^cHbZrfrSuMMva-iL0L}XpwDRo
zU-HZiV`LJs17ZDW_{fKIx%%W#gt1xLBo}i;_}RyEqh&`ueG>=^R|N4>xv5=@AJQRn
zFtMLL{t$1<*9y0NgEU4Jhx+G^v`XBwTAOstRE`s_mbb3fN_N<i4K$zQ{$Yh?aTGOm
zjljO7l)7YKHE7_ds>-X&qb-<>PP9yKS$yCfd)$#Hj-X;HZWg(g4}1(8QHqCRM75#h
z;nNS}${}x-q<B@_r!u^PBOrLbIf71n*Ty0;W7&kC8&ca3D=Py0+#=qLn+c3dGpdLa
zo{z`%N?7w-Tf%#NNQ_w&mGQFNvHLy4$&7Df)PvBEFxg&LB)v1iySK^!mxi0JUqW^%
zAZzV*4L{qZ%ktXZ<QslG_ZoN3BX0_5%VNYTu5Hb#k>Pp@EV`*K(bZ?3ztG`^1eBW1
zOsjDH;>UF8V2z=1(x8a4e1i>#Gbe~N+IybZa`lIEqT}!+Im8|BMElx@=suYurNlXk
zMt(cjXB0g>W<nPHd1Xv{F(hn*9{7}80Lm-UET4%_q}BW*ITq6Hnv-_Uo^_D-Fogk4
zjgzhqqD$RJ-jKeA=h>>u^A`TE_T5!2niqJ(L_~YT2w##0CSG;y`mWT_2E8QfN>c1v
zSR%P-hd66~RK9;_d9rDK1Xh819v$rt6y}sBLI(n~?q-taR2d5jVQT6y{t<V1aNh}-
z-dI6bjwd!G6ym<|QKd$1`-GOuy+AO+h>Z$9pTqAdSyjCK#;p|lUNUFsoORFe)2%VA
z`35gNOAzFDRyP6rAx+aq?WLM2nz3vM1`QEQ1X92hr8TXca=H#!mb?u6sy>1(7w2Ru
z$|U~>bYuylEq#O=`WzHogNHVP*Y2?k>V<&?idfawgMTob3)HVlNom}Mh!YKj&Iak3
z1&l@9Zih^eG>n#X#Gtc&XMpq+Fz_f~A8-(*T1RA*er7z`z5_o$zzHgLP;B;01Ae_f
zw<KLiV%58*=FlqK=>_qyV;9)QJR=iQ!90yUEuMY(!1Kby8DZ*_Z6;qsu1jg|2m2!o
z1{bL~$cyY{V0q5-82oV&IrZN+ht{YUn^{~aN7RDXb5p%<W;Qs<Mb^?zIDEl1F|THf
z@-gQRjhkGNp0iCq3%DdS-t={Zo!5kq04)o9rM2KRiec@n!~5sKqqm3@oTo?H%kwgR
zjf3H&E!IAT6EwaW)HUkT;cdCQ*dr&C!wRG*IJx?EobSI9Ae5fJ4lssz_zMCPe^^)R
zVw+PI*wAOj55Gb9NIb==6RSq{-U9mhs?txJS>16IMt&^&yMEQ5R&5iFG#xeci+er`
z9^dW_xpkpmy8(+>zaNpgN0g&%TGk-9$9VLJ8A^S>31jqQqPwk5GPU6s5r<0GyKwP(
z@%|7rj}DYIy82VMmY-aR6Uh5Gxlb`m{Utpr$otsE_gw3SOG?M4aMvXje^?|>9PZMM
zn(h;z%gO|K7=B2&;b!1MQ~PkPKgaPS`?3>s2Cy;kw;*{wL~Il2{4MDaE7R;GB$eSf
z>B!*|DZ9Ug+kWR7Q<U-puW%@?9)+1~wpaZHIQQ}+#};?r-RrUN9?nX`#+`o2Cr$<5
zpbc}Fv6P&_f#cA*5If<NuZ$L{(U1xQbSwFTEDFT-x^2Y1H-UOi8`t*Rip0kEJed;h
z%>{Y>^T~^waE?>4(eB11*!>W0JRO`_($$mO_=S94^0Zbj`<$-(?lVP681J2lYoCm0
z?sT=W?TBeq=1M`04e?3D@0G$f7;SqkD&2599ry8(8Krt0tR<21Z=E~LmO=poiIm}i
z+2Po*9Lv_-rmoBGsy%xpN@#X-t~7N)!J28Y-ay_e<DajlVd?EilS#*I!<KN6w4bID
z#3!&go~1=$2s<o0=^1o1`CWP(r(-g8rXAft0@?jA)B8Nr9Y_d*-yTIb6l`&bZmQD!
zAy<2a8K(nPgHOQwX}%IV316Hu+=Cc83;y;$kM;9k#Bt0E3e(mTP5bV==#)0_IZ!?5
zO@pNj(dK-TW)9<tUh0a*^9L;2LDI2hYc(SOn!49~R~~KHF3&Jgz4hh>POCh4e91~;
z-EV4;^>T1L^V9_glnMfRCr~4Yyk0|kHTR~LT>O0W9>&o$8cTWnh~PXB+_{REaberN
zlFlBWi4*0)4W<n;58;!Ns62A<86RsFq4Nky6rOVw3n;?gNG&Gx39fXydRZ&fIfE|W
zS<8ye(=_U07S=#Y)##-12yNe-W6<WhH^oK)MU~b+bUb03v09YVa}8EOLT{Kq&B9*w
zMH@Cp(Vm4s4jlC7%Tq8?2&?)sOi#ahof02hAIt<R(m8(YX9Rd4MtN7Hu<1_dK(&^)
z#;`BldenrMK|(;@_KZ1=61mqDAZAXr;FWvfq|G`?rp7XU(6ECa<i9?E(P4N1gDnom
z$FCGl%*;fXjG9hvY}SbR5rWN4Fj~2StFxa+$Y}i=o2<fA9o6&h1h>BR>o*UH$Bkl*
zBAv}I{S9IBq1*;#ra+h6TqWxaBlf4JKRUC>aM7Uo)XSz^z`m8M)KBv(+aZ-5hMcay
zs$<_y*bQg@Vo)0mPWHjkqi>eOQ9915oB!k!0=zt(H`_k^F}x)9oSmz}_8{9r&zzOm
zeq+6f$@lr_L|$sMGeRw}pz0Iy1s7GqKUviREYn=bgW>@0&s?`cFx7|`>Af!4y}#{U
z82i-Sa&ocz<?zFShCgDPE$Y>fs6R(ojX3ytTXVmD3J3jD6>Uk_>%?>_JQ`%?GhA5e
z4Y|LPH1dF-VbGgS&W`@!9Y+Vdr@`kNi#KrQ_W^*|f^OE)vOp2^qD*X8<!9c0-zeV3
zKYt2nnRMyjH@%t1V8a$=HMXhZ`+QGQlT7g;mEJ)97<OAtV467WkO^lF+{(k!<?vzb
zHoq;*l^06shLvD$P3RW%c$$_G9D-&coGr-JvJc4~8}yYLvut~8?a}4vT#=E_hRo!J
z>T~P{4n4rp@a*v{U2v@OR92)`r>1?@t>2}9=Cyw~Z#v8Z^(BDOr4g6=1q-tIC1p7R
zBt05bx`qg?d#Mu-E!v5hup$MaNv#!#&cu}pyW(H=)}n7;bF7#=aDNQ7{Y*B~cCEbk
z#`ks;K<cGqMEtHDf^j0Vmly4$(W0O^?TP1DUuz5YtcZX<hHKK=2l-jzo@sDRz1hB*
zafplKWg@=>l7zLwmHk??jK1P6nYkaM?W#)!tRazo@~_;!guim$WavL$(reNbnIU7+
ztVs-(K5l`hj^&yvasTPl025!XzWK8^&>gs;=4_+}0ny|<!RJ}P--Fh%(D|(D7{^SS
zTPg<Pw@-YQU&-_6MMyx-W-vC=hqhfB75j~3Og(}v!@w0@8(zmSKv->2+5gjVR(eM{
znhP*dV_N1qFcm%Q&2L+><ba<VDaq8UYgVqmy({S2jZ26GItTLNO`zyx;r=rj;*6b>
z2u^s!;@!UQJsXu0egW|$e`faDLp7C))rvEJyka8XI^9AAbCaGB9zMyUQ=h0Xo)vFw
zB$z$F$}7)Vq=ezo9@zsVT16;H=N{c+T>}n2+q7XI>weA2w=h8Z8*@>4StK;WC*G0I
zSVXpLJXk84{7$UKRA+SnzS)T5Tqk2|1p=uo#rZrxlFyvhdB0KWi7)+p&uIs!wfcN*
ztoo1-zMB@C8pZX)5q!B@^rnZB`r%S9(1ODwlIQFMcwK!M{b+{rmz6VDWmi}}m2MM9
zRVt$Kjqn3{5FaXA1XROWksgi-V*Z3<u?Ysv|5{iE*-T(`gnZ6$@9$oYHI|Q5Bw|!N
z&q*78?elAr)uW;sai#<F*m-ToeGLb&sQWK;e5jXD34j2}u|FOZQJ-?m*6edPy9GlQ
z(;89DLYJ_NK534w<lRHC39Bw$t6-`JsIY>5-N<1*!!jwgW|;jo_&BG@s_2@VUWRg{
z@44fJ(EltYQ8tR2tXf>r?s=>EIapHPxr8CRRc~l0d-_<QG|CK)!a6CP*oWZUmo&G5
ziNxowh98@@N8^4ZtyH1;8z8t%VgY3beL2TRU<oxdm|>eoiu6XYkRE?+$n*X+WpyGi
z=gkN0Y|qgTnM@9zo8<~vSbz&9PFsXRPQGjk%mSb5V2ID#lNB9lOUVk1zzrNw!0To$
zQLG0?JB)Kf6J*XgcO05jJqBp8$PYy01GN$wZT<E(UVE;DY!!vX`2<HQC9Y3#sa7*Z
z<@=7aQ^Zr|>rHcmITcn|1pW%FDZie}ddW1GdW5P6eVwsj?JNA2p6$=-K?*FL;wiL)
zuz-fdBu~T3!>yz`DH}w-scElTGAn4$3|#WL{<;<ydg!z%6t4OOkH>N`k2^Ezlym?=
z>^Tq4l2au?RK@w%Y8_cs)dmF#yJcZ2;W9V3=lunQEnQ-^Z7?@Tg$kd87XS57lhZJ~
zF;-fbNaOgFE?ckk$ReU;ayrLzNGLDWOKm@YiWMf+nk(W#b`rq7^Mr+Z!qge9>XJ~3
zOZCA5=U>BO*C95R(wFRU<940hTDR`1KR&%PLoFcrx6cH^1Sv;LI4|E{JSi-)Z)UF*
zQa0_Ds0q=0QO%gsYalAUx%r?oFho|Wbj!<m+;5o>p6=NP%4yV`FTtlHiv^gGBU$ql
z^Y8Y*_!E~tHJk7z0%e%{uKP5ZlvA&TXrB^a9*{uvjVv2#+@Dcb(p)A-Y4_e;)6hsx
zVjE^us^XA9ofATj;R)N?Bvd3%t`K56xgF^#4H4miQ?Ibf#=(Vq10$|FQbS}aoaoQ9
zw$Ck1FOFMPL*v!$U_!Rxr<s4M|8O(Xos_QYJ?4|c^EjFUFfENHX!en%5qRxt&P0iB
zSdxV@<=_M?B+NU_QSR=2aIaJ@5np|0_eZb@h(GuT7Z;|b-uQgm&RCXWtgc)VLvn|=
za(*xD-Yql|H>mx=38FJc=>(EPJ8R#emE(QqckyRF#B&ykaYk2R+&)B191TmZ{pxPu
zOF!{VrlzY~pwo^wZ#&^G^-;I`9RNJ47HInYDglav40G<B@lIVcXrH!9&H<DkCm$-U
z)k!3IzyedIksVO`dlbMvZ(lvF-DpKGK}NMfv2*pg``a!K5C0r}IXt+wvYI{-dVK|-
z8q-dy-RMu?4Zf_7gTKYEY30)K&Ah7aezEup&O*hm`!reZTZB~y|B0u9JK6bD&1$)x
zh?g9bB0597nAHyDYVpRhCOq*Vyp$)fR1=)HeN~j5=t-_|Hh0aq&HMWM+eh*7?-XHX
z-q@W^AV^b)I-|5oYFd30GDo4Wa!AU*Pp_0Se6QtF=N;pDo<2<vU`Y@zRATmiiK~cX
zf8ruG1Y}#wAZ)iu-t+^2W+QIQ0gduff9F|pQ(Ha#)s1$u$pg^0MxrVO1;&<Jg~fSV
zR?^HC&-#MCE@C<$uR><o)&^!+7J#m!N}-0`m*vra3%EIqtK$(V>8+A#!_ILMaU0FO
zKr(I1|33d~`xo7%FVK@qxQdn4X_k}g+cEFUfY(lttxsw7sQP~ZuRu`0NWM!5G0jFp
zX5XX2SEC3M*~v&1qNvfy2TF9(6F*R{BiVyTMy@u8asqQ4q?z%I=`sqn5Q%DgWSF8J
zCE;a+Xc8oRR2s7UAR2YIIvoxQ!I6yw>rO1Ch-=ootGK5n#I&rCyq3})Z-qy8_B`NY
zQ=Z6L%IVA%@J(A$8nyZNNyKR!hh`g9Q#w^2+jvwonbjk(-u8sy+D|$M0-25M1M!i-
z72?XA=5alyMveHp-#KIRG72pU2S8C3xv`eu!`}2?IXgmge3#wpe|+Wi(o$=q4A)|>
zn%#%E7cja?(Y68Tx9iQX8oV|Fw3OZV#?T}UlPjYAXk;B(`N&BRILEyo-DYR_RH$|)
zyW5n1Z)DF_4|y2TbItj}9j+lrNbWV|;XZOk-M~<{ueDGwFZ+JJGW&kb8sfZmqB$}5
zKi8QC;dFXB>8{y0+<=em!4FlJnePmN017nT6NDUd1M9Wvw)*y6Gty&E&$%c61x6vW
zilEY}DP76_D7s_xt?(Sy;U4eUG=-xTW`JZQ#y`vWbnl;(VOd@51S_@+_Kj;zp_KT=
zBLahJyO_k7P_x4`!N`Dd(b^?$lp_UX^PNO{4tYFt7#YVV&|msKIRMWb=DW<BC6zgI
zDXUIaTfT5zqGDS2N{OsvIrlCsAcI*SBy~80EgxzQO-p5fl5VwmrPQq7@9czXKE___
zT^pMxYFO2L(8|Iuhd(Ynzn*yTbTu^uJbBtd<0!YcWB)ldES>NN$S%<wA8VMULEL9G
zBm>%FGRQQqM=N!0!0hp9Zu<Q{U%j9Eq3~(MP4+F0*t))GfUH}$=wQ1GRpr{QRu{2r
ze|fWR24o~~jNNduRZb<P?>MCV32wM%b;d;<dQ{%)#*Vw+s5*yg4y<!$0xvT++95M0
zg?XxIjtZHST%(M({nCDu62uS?^t*E{?3i&tUa%8j%wok6T1^?J4?Tg;R`*RY=d+G|
z-)r5CGY=!cdG2>O^oJ3VbVOw{vT@FxE2ld1b2z9ic%6x0TO8q;T9z@5vZ!Nd{`K`g
z*82z(S92<~wX!#4xjD7?#zk^&UWrK@Nsn5veJJQWip}V<x;X6%C>&FU17hfmC!@G&
z#lpcbMwVhk<rMCygVY1nrL*GuRIp5Lj}W5oUmIA-sY6(gb)z4JSyg<%g+>9|%p9}M
zSun5G0BKeETMY+S*HUPcr)Gk;Ud*p_?joYAhk?cUsGN*`U+-x%KJ!83nQR3jgV1k<
z4kK}0axLUl-lAHTm;wG+xSkiDj|-3MYBU_lW^`2gS&xR30oIZrqoe~*2Fz359A`{e
zE3o6(2;JCB^G&0QqlP5VMn_vti8;Nm+OMl3_j~lHkP86jiiA>GnyfMR@*fD@)c*)Q
z?x~F_SxI_JDoRqg#{BBA;vtq;&NI6+x_W(}hz73uBesU&<f^w>dEfiHD*I!6q`PJo
zJ-rkEIr#=Xr)h+YFtn27T%Ijp&Ja(fPO|T-R_Ujr3o6k%Bc0I$ReFDi%D>x=4bU8Q
zCi{+>uU;eN(>`R3GL_O(wMHTY=y%d%<=v0&hyZr?U(@~~GgF?ldl%M=<S^b?JwyGP
zjP(_6hBzquhSHyr3W)vd>&J2mWLwIr&B9cbs+b4W7hc*F(}WC|!i{>x5rCXff#{hs
zDhhch$-T3fIk$0g^IgS1{J8|qJ>luFFvLo%P!Udbe2pz$c^=4^?VLZxRU^|!wfVK^
zZMlxH(Z}Z{kj5Vz>pBY@fK|Ka%9$7t6KZKgxit6LRpLhUXA7J&S)wt}hJ#Yv+)ZP}
z_z`WS$F=*<Zh`k?C9$Z6rg>(SaV}&Oid+m6xlUVih1e?eChFsde%dvcVzwE_gH+hZ
zTHdrPn$mmBsD9mHv-9VWAwPKKIp41^T~RVgGhl%emAVP|G@M17k<9-4EV#6FcfzUl
zXk@2wUKNWam8UB<5y$u_ODzw4n$81hSp#rgPdwHIh5KS}?E76nAnF|}uk#(U?<F~m
zcT9hv&XsW;WYo1^ylX!+1ESH^yY_LghL+271xqYhF^yhqG$IeSY)N#OvOqoK_?^O#
z=P6*8^KNY%{sB<aJhuUxANSVA7?t3kVOdojh1uj<fTOZH@!4J@1d7VDOfRLrmcc=Q
zYIfA|SgQ-^tCj;VxiL?I=#t%><^EIG4;h8)n&O&?C!#037g4^`;WDSq0?EDmzQ4DQ
zjEvA`2OxzfbmCr<8<mK=f$G|5C+g%0U~4gqknp?OE_6jE>8t3cpchYZ*|PG3EENC6
zfe4Y}T2EdfIVTeynN5AV_L?UpgwI5(cQ$x(CEEiYY8;q{&=orBTurxzP_bZbW+@*P
z&glitP|o?Y6B@Nb2D;XusH#PmG+S(Ym7V*k>S^V2iZvD7aH_!lF7wmkbmrau4yWbu
z`G2x|#VsGhFtXIoiEQd3Bha;OcE}$Y9y)yIeAKMu%w2SwTi~P?MXx=*>_B@6?iZFn
z0PviU>S<bWXBM0J@&lAm3H6lJ`l*h<ec}21!gD?9bCHO(@y&qu>sxzYXN0ZaMSGhI
zk)ei)vscC|B1-SvgS9*Y+~00Z?-a@pF0f?WdR_DY88{pUa9BjM)<em@dDe=B@vtX#
zYDLQ}`X0wS;8gSnz%uR2B(UWNHRpL4X00m<q&=>SA?-RB=KUt~OXba=R+J~9yr%OU
z`D>08JsDUHln90rYT`$IMIEpl$SYyh`?;P*M&qBkez^D1I(YJhY9ugJB3;y}Fvnn#
zK5W9hz6ZwfK#SXjW@8){bOfeCC!>sYC`{V9`6#@u)2On?VD?=jFvHYcQ0XDT{y!tY
zv-c9i@do6{^sl8{Gs>OK!$9Eg9EBFlJ_A`^2UIgL$!h>@L~EaCdq!@Rn7r7FY835k
zo{v!`&nn32zupI|K<IDKq1ImASY;48EVT?P9I@+=nDgYjF<RamSoxnC;e>g*I1e_k
z=1>l;x<2OcQ%0-EZ?!j}E!bxztINJ_zoQG<G-Z8;gB4OwCC=NsIE?(^wR|ov&>dxS
z-UC8(Q26SC+XL<oqWz785kV@+Y@G{`Je>tnRmD#hn*qKqT-Ouc?X(l?a;*Kvn`3Q2
z(Q96{F2;j?GI{Pc2_)Ti=1f;<O$2&^xgh}QV+2xdIZ*fa9K$IVSFe`BPb|Jj0NI{R
zt$`GKn|j8})Hlq`aIJ;w;_t))!!c)q+DL@jkPuV~=xv2nTfNkt9(<s-8rFJaS><Lf
zuTb1&3%P;Z?=IjiV`jZ}?P*0+ao+Ye8a2-}!0uh}S%zV9J*vAfLmjk*DyA*(4b#ox
zPh3A_6c+0gAcN35OBV02tQ}h{^EPW=3TSx3qGAq8ZdW*x@iisIj742y8I85&?2RGM
zMBIU}InGkT4v1Rk6qr$(d5lHzz?x%m?}d=+?-R=4#-fEA#DqgvNa!^W+Bg^nmjprI
zI1F&^st~tT+a<P;9sexv(9k4H(PM}Ny$JLgN|j03l)w&(0sx&;appXvldOgG;H78%
z+l)m-Hdswxl)|$9-@LUxZ>Hqt?zsX+k?B2$!|zyD33(y814DEVqE92qd-MjR;_Z*v
z=byf%*6Bd4`0!U~7fwSqn;Gn5Bo7m5UAPvhwlP#t`S;_NQm+L4`7US_DbJ`7WTFrA
z;jx`_M;Zwm?nl1_V<xI;k*8C%+5C8}ybrZ0DNiiesIK-|kr39gc88qxXlm^;rb_+$
zBC<ra3@>jW5%-Pv{la~3Rs$z8tAU>iWBc9|{PyHyp5GdT9%N(!8;B-qS$iKy>vh%f
zopu=<3r2;!%<+6YnULRY^aO-zKBwNR#v>2bTIRzYz7lGrTLuhW<l65-V18!;v9T6C
z5bDD}bN#TVAm+w|evU0bK^<xz4TR7oC!vi%PfRUxY7$Zwc@{~XJKM!RV~aMUsDzGC
zVIE`sL|0(*EM1M94nvdj792Pa^&ZqJ@zn;a&kLY+r$rhHA=Zy1Vxn4G@4&vNbuR~n
zh>iOOK01AFVW`@G#VVRkuL3o|vA=ERU;3ps$al}gxayVYQt0XPpp?&RjD&=o&~TVN
zw519G^va%nr}%v(7S%)!J<g-h=y*{j&&_mbyMGupC3nB?RmUyaKkSej<kg&^HDmO^
zKJNW73-*b|gcs&#$D&y4+X*?E!R_LGUI;nh=8$jE-WQdHQJD+yq*c3%ec#oqjLar^
zmC<vK!*KFm7)9lV3V{R*Ql+q@(DZ)oX@WVr0uU2Q(j<Tdmy7M>(H(Wa5937K1Tk58
zrei+EK{GR~^}uyKDm<1i+_;fjy`Z>OUWimx)&N219>dw)d5ujqoeUu^V27?)7ckKf
zg5I)-2cVm_*_ny864$kmtRcxA*jex5?rf%+EvHGq7UNLW7%aOe6;Q&OdP+7$GD~Xu
zzO#%;PV4o5b^RdwC=0D-ZBn5@a1zoI?jZ^7g>E6TR9m1zZDfJ?T1fU*QU9b4Gh`Wq
z?X_UlgC(myH-Pi-jzUg=5_yH1$nww6T(FM*G$}MGflP?kQ%Ti{WoAE#&*J8+<3NcN
zJ5+_PmQIyUBmiH14s}T=+AC4WK!y}H&r$uyiKY4+S|qIh{t5I>i>WXXAlt{JdLG~D
zb+efxj>o#1{R~lI6xHFW$?MQxHOC=_4-JtLlU)FoR7@r+BB0(*CzybJ;5t<Dy9|O1
z5>j$Od9{ge*M_b%VXOI`tMQ`hHVSnu%E5#UN2%5)MjB5;cOaK$B+26{4_Ixo9l&}x
zuEz`Oy1GzpJT%96PJC*j_ezvX#CyNXV+vFrDN9C{6n)N|Pi>KqYt5naK%agG$)6Cj
z=B(nGP}QM~R(o+!tlM#|tB@<e5by4>&$Kbg{awyab8V{*|9L)t!DBtK?~NUSWvf{@
zXajyI%fo;HV6v8S=c2t=E@EfpDBPXlM2;a6YUiF3E(I+PC0<igdu7KiLrA$7Qh49@
zj3p=3XokvbX|fHKgf@=XMpsGd&I5v4{bp20WE%bdlJgXj^#nRs*4?n*w`RA)2%3<P
z5oly2BLZ}*TdVVv_fZKV%9akgP<&hc{l%53idLdHm<s$Fg)K9LIg`(7T*pE>_VNFq
z^bf3tkW(YNngQStG%4)NfQbuSC5Els#CHaW0Qi0iwPbHLdK-$rT0idCh(vT$h&rIC
zb6_26`X`GoVQQQ+@A}cI+#O;hscKnm3+!ViMmR$oRz{ZNK^vhXQF@F|hYINspi*3`
z+Hk;-*ia%~3HL}<m&-tr@+rhk^jx|wI|N`v(q#}1m%AtV5Cjz$wh`2ZH0sAZ(Dpo&
zIPsYB#?-ypC$@cz!zvsS4mdWhb>Z>&!kduMSzjUqdFFf2E24yNVDC3x=2+{22lGs#
zoZj~valdhPL2VJ^=o8>kJPuKMOnaC=&dgy1vfA3na8rUCR!(yF-=gk77CR8yBf+Op
zZYFFPnmyT6z!ui?SA2c_6_4v#pfLk!e_&^M>vpY{U-vqNF-b1zDt{%)qd=ECkfv<U
z0(csD!a59CFJw(T(0Sw9<JF#KGt4d!f!;qEQM!&BJT{M}rGYw6X&l8-YZO7C^pS7T
zuT5AH(0%ytYN>+*y!_|a4;lq9>01c0`tn?OC;Oz}R4RNMXIT@utlbAkv9&{)6+(%g
zh|ENob<9l4EHDVSkqT9anb<K@t`Pt>lDzfB{z=d0W4b*i^C3bJLhh|Jgf=I{)c>RV
zn~|8V4?ZXYygv;b9Ax5ru2G}D(IJH#JK;S0SM$|JA-a86JzA!-Lp63x69&&NGLxb*
zg!McA%z-itJ;xY@k5W9b(Z*JclBFU+dT=;CFBi6R1K`FAH77GXA7U<5s7$RIcFuDe
zRl&>Er{E{Gw3M_AwhUfCwu5<-qsmv_NgDuT-M-i{$67^cgQHc}JhSE04nwam>Nzb>
z#ql?6m<7yM7rI^R!t?nAkl63{+{yAV+VpmMJ}EN<3z1O=8m&c&;=+Azz<`~J*mt9E
zfv)oO{bQ`BN3C3;LRrIb1J8x7IQ|Jv&M;h~TY5@GdoAyHv0K6WexVyc!m8sj-*893
z*8^W)zv7o){(|TAC_`Y0Ki)6=?YIBla`NznU%&o}>$;FAWPazHEkI_!X95ewkh|<6
zpq0lyt+e&-0^BE9)jZwzjhB_Y>=nHxkLMH1oK@;?ypX`lX<X(Wx!QGLm`s{Yb2;*3
zARYJcKCI;cT$>~0`{!)Jj~NA+;d)e6t)b#!HvnH1Z_^<m?%o?QdlO`DgcEHe&q~7j
zPPOnQ@=J8cV0;4U+e6!>ya#y#g>FLV5#<hEm*Mfa@VFM{(C<@dK_nsJ)$J<*Sjgp*
zgKUt}oR$ef(QROen&eS^00OR9M;NI6Re1_-Kr}KYT5U~bE!^@vPB&T1o-@Rh{C*o_
zZi1o^dP`g$^Uy1xO_6M;OcY<CKrkmy3#n&uUZ;J>cs$u-oaStu=ugJTbJ|=O!1f*>
zu64F-IOJY)e|jYvNU!JjA*DmqQ!8t&5Koqyo*~#UvOVBMb1zig2A@j26v2(ef@QUc
zQ4(}|TR9u-C15&&beCbe#|TMbHyHr6uFG|bT6tZ*5?5Zp+NF$gh!`@IRc}NO*vcc{
zMh?CyG>RTy<t{40uN;>%Y_S_`eR?*M!`c>xi@nU0Un#U5!cn6b*TTBK@cjA(U(a9g
z@T)@1&tLG%FTcWkA@K$P{PN4M`0Lld!JJj%c4OJX+g^CR-bj$#0ouV}<IQy;Ebz{v
zahmBI><Jmg`99ir)J?HBZo9F-kll6l{=2R+*fP!|v)TK-6gd^W;6Nj~L?yA6>o@=g
zsDC86hL3#iKaC)Kcl}`N>?%@vJQg=x#bi^Sib?Ms8-s46SVl%=l=?TAcoyPmL#M`I
zqxeJ#RLw&(%o|bDd*R7mMwbf!w$(&m*Hu-6_~b0(K*5lWIXtrRVUl@}L%6wULbhs~
zP;E3O$T;4R*gaz`BgHz(-h|?BOEBi(c{PD+8`DCnt3x{7O2E2(L>m~5$z@~@BzRtS
z20)H@tWbD}tra!zw0#w)=EmW=^_kIH=Qe&)V29Cy-a&^h-MtuV@jMM*04acLF+`B5
zmFih?Jxhv3UIg&bzeWha3|L($&G+XxYf5uK4HT<A(E2RIcoUK{J+wWA@?^K)sGNoM
zx^@mlNL|$Ibg0s}Y}{0^9<N>0)Wy~!k~M~mve#B%fee*{9MK7S&Pjka)vPDZm*dKd
z9^&fBw0=elQUi<ZIjf(q>*_T$RIaRYffa7C!<d^bJRVOxuP2`Cf#r@!o+o`h@pwG&
z`u00~J@9z`f@^)z;|!QxFf2qGt}yJpTlXlx5IYLb@C6!$<8yiCxye{XkcPfSr&O0?
zCfEIH@aU$sB$HE{!NyotxD6i<bYnwF)chwF(wKUu@l@wz>Ayu?C?ox6t{?7_dJ13e
zVG?bjYj0JnGix9uyMw|Kq@oth1hFTFpngYUl)D={qk4A%tUumu5>}!(ybi`<OTnaN
zHJRf^Hn&n-Udg?Q=b{Z#A-e^gRF(9IlD%kJLFm6kXX_{}pSLFQa(#02TRC8a0GN$f
zd(PCKvisDWmB;{f%TZ)CEChKyyo2J{1<6}D%kUPZ4^rkf#cmpuki6zG<wZmmGi4tc
zHaTzAjimYJLhXvd8c>Nrd00f3%S3rDxw`L-`@W&hlg?SK?=IN7u&zgMBBSEa9C#Ng
zo?$1>VG6#~7u_$09Sw+G%qvL$an3Ri={}dk0EYSWY~Yxd509CD!r;D<c>{Q}8vFuC
zu6W;XysJwh>BgntBcc+<?U{@mB}6(Xqv`A7d!f83H{APmco?kfaD^h(r-&d0U`@)Z
zSUU1EoNJE9<8c^;ddHgvOGZUK&m!w&zM88ta*I{nkE?PXcHfA{7ZO%Te(y@8n8DYD
z$KzRjnpk6$akDSjh3oMw!_9N1jV(MLUIzuO{jPhgC*JFYjNh<d-*}8{>5#pr<#;Ju
z>jo&T&xF$P=>h7zZP><R8P?;%dR(1nQPjiD@OV6Mz2CTRn{E*f_##~%q{OB*(V%od
zg?=-$*i~i~X3Z%qt@CC8zZ@5h!hZ?-s1AE^#vXN!tF<y4jX(jFlFPsyhP-c>7s}a0
zm`L%lbBedJ5+!Br9FMEee+w)HGKA)BCe03`C3LG?gW{lC#o3$_1`tXL44TxK6g?P%
zU<v%vDR6ceg+~6)F0KLAN>d+2yZJ~E_0@QGZV291C@Nrdt)<L4S`Im@mB-PRcfv=X
zf*_Z&n#O5;L=nwW$T4dv-6pRt`TkPcGU6rdN~&-i4!Y+B9_}3#|4kYsA~J1mYok}w
zr{bx7-!=aZd@zjCh>}+;;X3bGD6a~qsyaC6;l->qz&0Eg3%)M0O~hhjNQXkvh#zCr
z^~2$u*N-;#fQ-Yyq7%dlscbk_8Hd;iH{{-U-!Hu0Z|n>_*249C;`w~m{@XV)7l4cC
zc4SJ_Zge<~CndJ0Q{kfSGo<9OK*#YpIBNk|qvE?xV{4EQ`v6AM0XZUIQHpi>MVUJo
z9g;&D3Mnxsk_a2~Z>+KC=Bn<UyXZp79B?KNt77j9c0F)C9_+?oc*MrzdE<V)@LIoP
z?~T{{W+gd8n>mGpz;Vi=jdAaAmcUmY^)e>qUT2`x{@Ls%z}gZA9*>2u=K~Rm7v*Lx
z$MtYr>|>@7Y8NPQ?tIC!1F#GhS@kSd$@=R-qd=;<FN6KLGj@JFd{i4&%9OuIO%q*W
z3{Sb@kkx<a&M0>$tHWt-g%=8eVb{-!2~dmtSdPcz;<M>-WFo-+C_0`|NXS$lFbQ7a
z*c3|jLV8;K9lH~JU@?eV7E|-LQSoK%RcL{vFeH?EZ-=*{ZYzB#TGdI6H&sTcJf;ov
zg|z^2`5WzHbdtV5U}fC%qz0eRRG~8|FO{d7RTxl3d7nsF25#6N6Y8rNbZ*YGm-}6z
z4zLSkVv|N3ZHO5&H*0XAJWOODw?Zds$Icht#<MfpC1`!CAXkUzETNu@eRe>~=+AmT
zu4h&MdeI#HDkT2Kt9k|Hji#hF_kO~ocK(kE+RRO3brcfHQ(_Y|DG+1q)opdyB=&@)
z+zi+Cz+*kSSW8O%jt!VY=S(G5wSX}PvXty{BZ{x{&{abu%BL-xP!EjtFop_cU3K5M
zVR#`jaQUN*-hGqyeS}CcM&5mO&CYTW56XBfhW!8*Z91BmFy~ejpy=p>+Pk=#_bsUP
zcY)Ja8w5o4M_cQ{<MF`b`79+1z@xm(unOnA?i=4;Z~XS#Z?zuVxGlH~5@Z*tolAvx
z0|G*lJN6jz>RYxRUj56ExZ_=22insp1GBD$$8%w0!LABpt*dgjWfW>(*_N)Q%q%H?
zv7{I(4mH)RqCXY0>bQa0=l}Zp;nvwoz-3ei&Ao-D(Mq`3BHV-Jhq0fFrW%dVkw=kI
zP?_=>>JI8_S|;p>YG#*(L><mXP?K0G!;np16y;Bw^xngbu01}Qc9eSEL+3!KL#sTc
z+{5>mqStMtL@PXlW-c_2rzY!4#0HZxHukZ>R09F9k`y3+eKz8Jcf2jV(}{2>j~~BF
zHSlUrDQA8-#-Ze8Cq(nEg1hnf0=HE{6QOkg^DFZ}<q*y@ug}E&obR1fnBJ_V>PrM}
zz>=F#XaRuKqt?7beUvxV7KU$zA|B-_*ur%^;OoM+8#<TcCcU$~S1S3VF;mjJYX6N>
z-`-pbxz_V_qMiX*4%mfKdZ{9_Nmk4IaI2X4T8~QVauM%+H~q2bj=-%mCTwJ=Jxt1~
zudC+9sXg3rU5{cZDTKqil*|O;#@ma}77++}ZqfxfMNUk#dA7Wj^7=0<XI~(CO|h$w
znO$U59gwGLmjXJp2UaI8);rf{^bCchBNFTJ#PjiFzM?`)NHy<vL>iutC%&G);Opxf
z{`mcO?0ffNPuh>lgN+xUcj6p{Rb3TBsOAR0*wF5BRBzBk;D+bal9&UI#Pu}%vKFpg
z3_JIVid?Rog0p@JBXc%ml$=Ghc>?<P%qwUgoJjINxqi$j+==_zjcoV5F>Q56!5RnN
zZrs5kuVmFI9Lp{MhC)bEsjZCV0Z1_@1i{5w);Jmh(-Fli*o=KoOl_=XE8MgeCb?u{
z$N+_O9t<-W5mzTmQ;E|;xGK);#(I$AKqsRd>hmKd=`-*wkBs|@h<e<A20Lr6fH5|S
zZsex5(C<%A0`%OHhPcnF&BS>s6qjR^H5J{N#E#8FU?`AbqV^JPnN7bdzMtjLJoUbl
z02{+~8nfhPY)UFd4t4}!cg>+Y1cowI2c!CkVpnx-OB|?KhM4ml3JzK^S){yHd+-Vq
z{)0V&QI1lSNqO2kvEOXo1^}0#4w*W_eQG(6G32V<xUQ>P$0w3TB1j(|l0ss<(MF~N
z1sXlx;qY};`wL%K*8}T%0LaSYF@LddM*z6Y@%rN%a=)f$)G05=96&52b6gA!nwN1i
zb5=8JyxbdF)65nDg>|aKsz5>&spZZo@}j|3vU^=u&6`b!WzA{qU7kc#b@lV{#N)c~
z`t}VG#iZ1kA8pWqt`C4!{y}G#FILcJ7(;cv*<fn?hYBHW^*~w`fO<R^d;=>sc8(`<
zEpJ&HV|UDNj+flH<iLTH()L9A&!4z{_&f#0F98aC;NbKN9Z|6+P*uSOv1XJDC?qKK
zqB)$+E`?}xht7>Q%n_S}CjuKlvQ@;EHxT8@&%R*$?4^#cXjI*C@}oq*&ouT42Yj-K
z>K%7j#Xq^Jm574RYBi0mdEE6~DPOsVRUM9YwZ++%VjNgD-Us-4Vzf-O%4;Efq@oVv
z8B{(=(d@D3Aw&!e&K$8NN5%W=bh<gp65J$KZZ53nx%#~^Oy$E!lCA5fa8tEW?IvhM
z|B}b7cykdJs-fGC;ps2C`t#XF_&a&#JG*@^12nF!7@`lk`G=I^6x^7dDdV+w&H4Sl
zalhZ#_hu1NaoWs}8jOx1kFoM4rF~)fV$$9FXSg>X`B@nEQS@nwjf?SQxUMI@e)$Dz
zr@m-(6xOR`<%PW-xbgveKd|q>8-HN)o@s^n1(?R3TenTWYyk@g_+eO?dkJ}XH2hg)
zbq-zEg-iL4Ox(bPn5b}*-V^{bn0QZmP!(n|=v;)Nhezv~WfjAJe?%Kh+IN7cmxA^%
zTtw*=)*jHRI<E303WQ3DXKb4D3bDIa_{Xd~4o^|~#}i2csiOUlp{%ve4rxa3K0fP^
zh0>LOC;tER`oZHk8-C*Z7_N}6VwERqD1pCw6NVNhXW+)qeV|7g01F@^aL3UMuVch|
z+nS(#2ljop3#;5h+HT8R7CaC(6@<OflW}q)%VDf&obx{&wj%0W=D2KzIy1Q%fl2Ap
z`#9b)b@++N9Ul7Wb*)aowhof9It>DOkLsAXNwF*VUIk!#1E~tqWM1GvkG)z{`e)2s
z<1aLHCV->b@n!-}V+ya0qq$NcnlS<Do#Jf8+4E}C7_93A+=}(W$4KrvPm)09?oeQ%
zYE`iY6$a(ijjlMUmm6>|s_QoEbB)XUy=xQ4#>eX0ti{k&2~5~hr75df`GFltgRt>_
zRpR_w3zi%18@@K~ePgk&m!+t<8@|LMntXy1t6~nR-UJzV-#5E;7UJLc#`ZEooA=xi
zSl0u;e*J>S^I2$jDIq<zB5rf2)y_JVEq37b`UCrZwWk)bD`&B5Zw1gyscT(BuawMl
zF%_cmWA7ONV-(?W71Ky^YF34Oe|-A`UteGNvIm41Gb)Go^?2ZN@8f}`gCxVEpx&_?
zq-{kMq?D=TaeOZk<w011QF%k=Xv%DD-(-}4(Gdf>EeU1Dap)4G=fiy|58Je+CVh3}
zNPWl(NU$*X2)bqIJ%lwuq;iY@*t`Ay`r(|yy+tXj@X@_@R|B`DGCM*T3E7hF{SA2v
zj7!SRKNNfl;e^91ppK=R%$P<1LXJXLqGje7o?1U=wK36fB;eCB133!ct<Zx6IYMB2
z5VcSs5^U6WF35y-3}}w92_?yBcBrR>D2R5Dbn(S#f>1Hgc-n9%CkTL%*)>KX0oX2q
zGNC(8SH-Qz7@3Tg-h=a2A3mt6DKBt}c}`Mw$+Bp)YF1a@iFPyR8JaK2m>ur~u4ALM
z^Q$?ExKHEi!nyfwIM1_&eW3YB(p{Z~6xlhW9XIx_+=6i)=G3~Hd{G2F5@DP4=iQAO
z0fm4Prx%@v4D3wczOn8%z;+$#_XS{NK>7g_CR*;eTW?S%O!HM?-S_?OFlj_Ep9cK$
zRmENkRg~d5!g?5bqXDpRgg@~0^$Yy*z@3SBzl!GR&UvaXphg<`;?c2ZnCYZt2aeD#
zLptsSu$*+&Yh5i=5gU)S@Or=S?SA2xc#<s!WK?ctalcgih<TPeAQaX2#(m$|Y`y<_
zzbdl8NiJeSg}$A}OU%cjtZVZn6}Z9Ep={7+h;!O>v@F;m7X;aJ<~)F-s8rjfwy0wn
z5Kx^;46rig1g&*eM+1UT<xjc_KV%g4-t8tOQAphJvB9OFGZ-$x{90iMo$EZ7EdjM2
zJO^Hyz1;7JY6lBU;-rgiDy$WzkkRK^043-l#{qX#OmAI8&$SLvF;`}Bg^^l+S($^X
z#<8B~)@suRH<tG%sGk@5B7$V%C<Em!1ct8YtMmf}2sU&(35VvZ(S#LZ5qjNOJ@(Bx
zhVW|F%=b_@C71{?wrZ&RTMMp1&ki5}gJZkH@@9QP&9Fj!63{y0u|jl`>~?7n;|eM<
zE<lGa7~v>Dpv~u;V^-TDy&ELP-s6?{gqXC4rH}y8Zw;1N3?sW?w<Go&i|nH~6swcA
zSqg#uruZztkj`)p7T)&@K;_!%Y=SbnZrQQzz*H+c!v%<B+X)gX_RX*XWBzM(y}^z^
z&>C2h9m9B`{y+Cd*c;)F_g#DC&HUe57rvffcw85~fl|N_&}mPH;)gPTD+~|6u&yV*
z?TvW9@VJ(IH%zlXR+~#6c~2d6o6s?@CBxUM;<vy;rt)3YN%XmX!7q<5T;|vr_>XUY
z!{2`Y4R`Dc2}QBn%ni@y1K-#Z#3S(S+i!UP<uAD6YG8HKFy8l#`+now`x{gU=*F5T
z$!l)bcc9wSh}!7}Td44;s>fG(&=tgBbYeGFWi+aC44{WVEJlDcWSIn}Rk<0)x2Oei
zv#O~u(Wi^C8NGqIM^O0@t?7SBokA2pt388JkU7IXTq!9-O~TAVDHAtx{;ga_DWQwp
z+Il7c_I}JvB6#~i%BM>zwnEb;fnJpqqJ{x^b73Z8UP_yT;)h*|h>G=0bobq5q<pY^
zuQX6@ECtO`imc<!2fdRpp#><_WKSVF-LrjBJs*Q`LTxx|-qe01J*U$mH~X0rIi%?r
z$7jiWdT^slBC_XrXTl@lG_uXk@R{F#cqNnanx7*qI0wR4H_bLJ?FL1=#Hu$?s|3JU
z3?w<y8Q@~|=<&%gLm;scF-1cID*VIqIgLWiWz~ojvR>G`=+X?r;Bqf6c^cv>w^PYS
zC+{ApFov6GmIh?CTzIfXp_DjavCD{M;=b>$3bk5tRMWMpD_Hl&HU})&TEX!3bxl8{
z^LSQ(iF<{4i<r9V2k<mp3Mp}eE9|FzB49MR!H_s%QyGf}-r*W1|Fsr7ZoJ>`YH9Ic
zo`ZY%kxzW>8_&n1=AGk}C!|y0N;U%T_p68^*VRVheZTST^$qXWH^lp0#E-R*8Sv|p
zyIC}MM1|E>F^QbV%oY`%`<=!aJW5wyK@(z5Wo6b)JGI?3LA||uQ`dU-K6h_0-TTT$
zPMzNq_}tvt3Q_+_N8v|3h2j5~xZBZ~?(?%s;K@oB%Fg~djc*lJ91~*(T+4YFhq%D#
z9^-0dtp{eDN~$JNlb}l>!Vbj%L}P)gQ6N7H16<#iM<Oo*wElXJQJ8YI6@V#3-1fZ{
zaS3o|&b#e#GaSzA14;njud4OrDuh-a6^p9T2uG3fOpu%uhKzVy>o0UBp^YZo$on5d
zRi{%8HMa=>g6Dd!eMi})Ks3uc0|*(uWPxM$8dYImKxxdRA`og<)v9Qf=k&=-KpV5!
z$QpQ~<Ct(_kOikWzSA&t*|f~10SSgUOHcuN4erH&fsEw>uJu5yV#ZmFoL0!^hOr7)
zaqV?2JRc7{h^DLe|4{=$Bc}>i*;Hg!#1G7OaPGwBn^W$3O2&Z$y%E$}JY5yy)ovvB
zZlxQ!>-}k9q}(@;$Arik-L_VqYqmmH!#M|z%6zUv7FvcmB5t@le*O9d&+97rG>0){
z`hLG)1ae9C;r)K&Ue&8D0KD%Tuh*+WM(;AHRSU8ZS;_Xt;{jyi{eI!zQDLmBTKu?*
zBP|AuL16e`vK@O8T9eZd^u%Y&(*#F9^@VhUx9kb5Sg*g1^W=~^DXrhvT;We$KV%f{
zdm}uNbhT>Xyi5NZDbT%7WWSs?ymNsaRfps_e$Uys9Jm;-O{+sFq%zfYwuRlxy{eE^
zHn+VkE=tItcxmgsSs}$|LFHxL8i5soWKq=jIg%rEbRAfD6mi3*?)RjGutF%2(S4vk
zECLmsuSLc8{jU|OnR(0aHyaXabzmQpA)$*#MC#Au^XvcQ%2qnS=fGt=N00-+`uw^H
zOlTa3r$GgrOor~eT8-#v)I!hvj`(@qCP4{iR)SJuqM$RzL$Eep<MoZfP`6dpmbfiE
z)|2<OZhx+rcz1+NR^J{5qpHR$5$v2`r6h#hh8;8~zV%)(8!9U4k3i*R#<OzxRg5o%
z(Z<l#^VJv1!QNk{oL?Z)KWk8|`v2eeUE!uo#PwjPZR6Yh#_RPB@xCiB5=lku?OAp*
zci({3^TZf3Rv@>MvQlh2HblKu$We)LJn4lKSX}F>BFYFPiy4(GP{npeVBc@t_ZzR5
zRep#d<G$aF2-JFN{(t%P3;+I?f2*#lk@)TRzv1=UyZ4+g9zm<M68vLiVKo}pzW1)1
z5-{ma(ECAq^JH~?+`aa%0c!OEC^FIlqvc~BRO6UE+R>Chas61L*=nPUa(X^c^g#hz
zQ=F8)lZz>&xljj8i(~Cpr+m2qk89y^EmjT!-|pggkL*e@qub^Js5q)ZTKpWRyZ)`U
zAdO7i=_>VKP>TbHA-&#6MKkRtn->D>IBtzgHOHD2E*)frO#MrQdT7dnh(idq`pkGU
zkGxGk+Ty4PMarrF9)J?sLX)cFXaRHet17x~PYuU)UL$MP`3^O}le}qo%c>Zwxcu32
z0GtE_409$YtJJ74qL-*oy;E-`?y0W-NWe0&QnIC3Kd5M|(XBe$wjn-@{`hB&7OL7A
zMh49T*W+1Lu&5R!YSUt(S|-CE0!hmGNbVQieo<T_KIoA_FGz^4&Ki!QE=z2)44e|b
z>V~=8TlPdGFd9Dh>PvL3i}QBS6jKz~jViEdbu;rFf!DVe?$;Y&tB2S94c~tM8{+-0
zHL&v1+}E&j2vD~2cs%zRR%kcr*#OlB!;OH8f82OFTPBO?C>!qZ>p~LUca6?*?S1o|
ziiLRJZ<qmltG<KzFE`v?q3Df-EyrJe{R_Sxb+&b^#Mko+zy1C<yyJ~awFofg8BhAH
z_K!jRp^%t*obM#9Nh2moX&i2W_CvCyXu_QWccwjgph@*Cn?(NXzy9BUg;D4!Xdm_1
z&^QWi+}J5pF+awqP0MbbOOP>fFF}8};p_2WPGOR>ddomo3>AaeoaoAW!j1_zV<>XH
zOi3fgp&u6pozW#dwH|;ODGnrt_AUS;+B=fyBM*)4bzCG4!+PR}wW{s+QhZY5Vj!}O
z4Lzf2nZOukBtV2aiA;I~c^E3_Ko+i{0qLXcRTZm>Y{c;Qwf5cf-VM*o*g|;}bRs{i
zC=Z9_aIX0CF|fcCs)?}%8YBzAm^7E6Zo_*Zt&mjarr0JvTCX6FnXf@u9lymjR3&aH
zEf9SMjZB3{R82$!!*ED3X0qp?K#W2=f!KE;@+=R-gS~=C8$}k?Qw5iMNA=lXv&L%g
zz9dfo+rz95p_u^9xnF^<F>nv$9d$UBm{bH47ph|VO?r1`;2j&k|Nb|;e)|nzqdw$8
zoy=q1XmoFEcw`^vnxJ>J=;^Rw-}1@_EvwF-fveet$<PQgI{(zh*1YJx`<5`balPNK
zI+pZE>PuFF1!`DBm#xHg<qd#jD-A>5mHX3K*+$n|DxkI;ySq;O?nH<9+z99hit?QL
zVNL1u=PE;M$BvcU^RoxjEy|Mjo{F4e;lHz<|H<`Zy^od)qOrJ-3OBO^Q=7oOSUu@j
zVa@DW03Ke6WFw+y9+|fdA_pcM&6vj+9(KkY(-)@6IP(*gq^|r05fv>^0MO}gD!pG<
zalo}Mn^n)U_5c&?D|?*9buYA8t-4olC`gI6Y8h2-G>2!{LQ1)GVHuXoYjd2)KF@_0
z%H{V|aof2LRv0hG^bKmGQ9?&)sd3T!n8tvpQ5(h!(=)`|r_1aBo|oh6an&Zh*aXZ-
zTg@Po$_N~kxF18HE7N*>_kFW4F1l{Syf$umrVD8cSQq?p0bb{0G~bP0fl*!uB$148
z5D`RAuAN?g#3x@0u?c)^wRhRI8f0Wm__J=|EKsa_295YUGni+nNQpJV-}^?wxkklE
zdcD8lu~sS4EbsXJewC-RnCD1TsA>glQ|i`ueqygt%!;nMr*-t#S227f!|9H_Z@gdM
z@OphKqj10Q?Qj2q`^Ac87&Z*Bh0C55%E_XZ!bB;r`G=u9;8wG}GL(^l<*250Y=xb%
z@rt|i6w3^k88QnrE;oEVp13-smw3Hic;D|ziYxTg##prn28C9tFI2K+3!70Mh+3T*
zM1%KT^V%VCfz38TJP0ae1rl!4?PwIazF+dL&=Ag@V1ROcWmg8u0nBY`yax~}XTy(g
zC~S&D{{Qv$W397SZ2(ayh3e`rL~CAIdV|AubY-|xFdwV9!~o-|TdkA_O~?z-%Le)U
zaiUem7y(&Zp#hGyles~6ety0VRK$n$!kDSY3l%1R5jzHrbBWv+9*>K$?D|v1BGpN<
zWJAak41dIt`xw+MNm2W4ovBbVW2E<<E?|no)bBQo4;-PYgVn(Ac7m#Zj_&7kQfxbR
zt+;2-Gr%QqCPJT&i;1`eTy-wM%>&ANY?0KFGDMp6V;V5yoDE6?%%*i~E%^0-J*v2A
zbz#?Mp<o-HJYE#^uvm0g#Db{@(0F<~9!sR3R$<vqfKas2B}DGt^w3Xb5j}6lJGIjD
ztI>UC;=Zf;ITE<uFMRv<trTF^yxmcuk@G!A;0Ux>4X=2A&ucTnCb18jdY?Hawv3GY
zkUIkV+Y9^kW|+3ZSzqhI^Z5&&*H-})CVk6jTJKNMQM+m7A<Noh&L(%&dHVWL==&}O
z_jq2oh~$m$bg_MUX4NOCEy{X4x=+$4b(@PEz#Ua(Uh;Y))&tJg8SnSIeiy;^7xOMB
z+puy{l$V2%89o$o(YnvLfJjb5%Aj1d+naNL2t1&@_2Y~!f{Lk%pJY7+?)1$6#r1>X
zBkS-`0?3_>fm^9MsyiQu%IU*6vhxTF;4zA&oRk<1cZ6BHJQ<1i4l0*v2O$jJo>@<4
zDS@tFwV@VHPJmuQQ%xDKq~*>G+%abav|Ei+!PJ_m|GW+fft3dYGJMn|YrYn!)&`%-
zt`w|<%$l-iY9$=icDfs#$*3xfv&sY6+oV;|=K@GE3XKw}8I1YYtRlK|nG<zDkXaY>
zK*oS*x)Yh4hiW=jURS!60)P~|7*YH>Wb7#Wm{8#NXHF=-9$9o`F^v3rU_G84VnIeT
zat50X+f5rOfeF9hQ$e=5jmjCo-Y{FRYgkO8J*pFLZ%CgUPSw@W0QUQgLMR1cP{r_d
zU8F>UZ9j-q-+|Yw6yEdk?D`yehGwWnayem^QogE$R*i^Fr7TGmiwSw#_cW$U<r26!
zsW6kn?lOdN?=C4aU-<g^3!aZJEW5ZD%Xn{J&<TF3s@<0G_Gn4<^i}7LpbW<KxNxEF
z^=6Lnop+HrvYJ^wE^n{|R1Wa#`NVY<`w3pXj~>q_9@m2{IJ`Z^KqBuOJMPgMiE2Ue
z`~{9mlK=L{Z+PGD4==FRT;Su-J?#Nb4>b#GnMip^(p?UA$2Y~F>UW1SJN4SBcVsj=
z!f@cqdY5S3aBk}FcKJWOe$XhCqSv7;qK|2Jr_E!HF>p@+o}CQKWZ1{7TKuxeq82>r
zP!XW5@EwIb;`>RFXL7!_{4z~2tXW0pISb`=7?#N(IX8Tg?SkHPI>g2pIt$WW-PJi&
z)u3w>wxbfWN3lpdm2^$_=23tWAdIpyt7YkA=s)oN%<5=VDKvors4%X+h@+`Q8%k3z
zp+f=pBTVWPZFoWUR|cxJMKfjqFn4i1SR>%oA7(S&-U&GgaAft%KNs%1HQY4HJA^uq
zwi_ynbMfVQzJMKTb{H2aIVp1h?Qxpv^j{2`<6H#8CH2%tbMq6s0RiA0!@n+e;sn9M
zX3|rJh~cAzj<~ds*2cXy_Boc=tgPM}d5v61QQh~=vYM)mc;D4qX<d)Lj~a~Gay+gF
zuA|C)x#PvNns-?x0tS)_P%R-|Wb)`ht@{qFuLr)Kzrd}OuFmOL>wzbXv3XRBZ|<2Q
za|CkV<yEZ*{PDo!@rAFiuL?0mVqb}!-|)I$02aV5sM{U)Ubr4tgF~PI!{hO&lgSKq
z;LchViuID0of|iqRr$UVC>EuI@)&{Fw?A;d>$uGK>(#)8M{%_GE(Kg*OUgOuMV5S3
zW6MN}g``d9wFj%*hH(eWE}T6LgLK@0!2yMeT*zgJ1U#%5nK^pngO&Lwt{-C`ZMl8u
zLA8f){QVf`aTPjx=g?3C(-9Fc+!1*3?9$6txuXhy<9H6hnt=1(XyuQO&!bT?s4!|R
zbpC459~0=LUY`ucT!eTB6_?zDzIFK+9W%B)0MR<bP&G=mFp!5(%Bv8pvD%r+V|fe0
zWwgTyo}$LYpy?F3aVd9H0h>l>G`Isj6c;^|RflWlV-v~~3}M)W%o}-MHSZ3wx!V28
zDwsk}Y_u1|^d9^>Wk&lhq7>1+H`}C7GF^Yv^-{86ZS1SAp+YUBCMqv+taApc7|Dj(
zJKjRA8Of4}tRpa&S69)?sty5Wl>_6UFJ@w_6%yBZ5E&CviI|>St#A$lsdsZX-0wHu
zZ^M(F1WE^0o+G^~%VDhL$WrV#<Nvh{7+K!M+*xpj;Lfh01%BW6UEZm$DtwBKJIafA
z?>FA>7t~Uu3;Pn~&8GIA!_s)(!G*`;qB2$?qLk)l$fs&c9&B@cU8~qhs8BMzT3e)h
z6|Wlby5Dd&eDU7M0a^y&;@K<LntSO=(DQXiu2p%f2-vE*0Z9P+T{3vN7eh&`FB6{_
z4W4S6=@WCScVhPg$ZCI;?1jbTZ$wu=_9*z-5A1a6U9PTjQT6ZWdks^Nq5h-x<6m7r
zT+v*yGiLIQ13COC#aZjKj^@J4ZDf+c4algL6PqCnm)Bs1quc-rPZ^XCJZHH+Z<_rw
z!!wzxT+^h|y59E<93Dr4+4`Jt5->eU<d(@3IIXvO%b1mC6x|2tAH!jHr=rSk6;Y9%
zQ|T@EwaQzSC!vsy3=oRPo}e+4(sb*j@%JOpGk0tx(KvC(F=rEBP8mRjb5bEWK%GTP
zu*H&)p~yDWr=M;8Upu@$?g@G?J1b;(vtFUPPwF$-@+4!Suuv784MVC8Y<#vo0X+(Z
z|72v(dr7v8-av(|9wj^bhEY*edkBRh55^JaKI6Og@-jPyK~DW%G;Z7<1zu~dHXFF9
z+o$IcCaRAQX_)&4r+;GGSpY{ZHFQRZ&M|S(qWNwP>sAn=@OJHsEH_x-#)I$M!XP_z
zwJtJ?kT9=`;ro6UbBlD@2MubSdbir^y)5y_3xE6lg}C2%u4g6Kvy#J4QgM-Nfpg!u
zH@nXYT@!RwJqeAKSD1H4;_?25zx~Jmf&BIdbbds4G>Ah+tvoqDREhRK7|w1lG~12P
z#+a~9j2<+zV$k`lc~Uo?dqg2uLox3<N&_X9XAgBGChB|&b)Lwyid=VFn*aRzvGCEx
zp)j&6xvV3EJWJ6gJ%Ty6xD>-?-F6v|&fJtRK?QB4TB@2gdxFXxR*0aeWPzF})e=-N
z$aSz&8w(Mof^^nWZu?{M8#VLu8wL-nYE)5)sgT^f6M6s|c`G4oR)sDbRjn&QsRa*M
z)?p=^ZI7=@EwTYy#>2MBn!psw)gu)L=L<Gc$u6N+|0OA$u4Kk|1cQ*#Nq6Cn{zLSv
z=FH<J)_1No1~K@&oxmMs99wYv@28n6gI-GI3}qd8h=^i5`NcjzQRQ$M!26B#cPo3n
z3m4?rgxZ{wTLREUi?yDm6ttv<QV#37@VLG@!QE}GSN8m>%V@ju%`)M8><zHvs+e=e
zBV0AkDs9L%q78tMRm^v9#403Ta#r_VW~k<*b$?6BckDjpEeSX-u37W!<*hMHRkTiP
zuP`IwiS{gRtyk85@I;*>ecv^IdptrFqDkRb`;qcU-tYGmyCtT@_F9$e$gKCus#5;j
zfBz5sv41NaaXs*`zu=G8V%uq8=f>}^SLFm=?*<HaY`pgy_I$#8;rY1mydHJd0_q)~
z*Auxe{PExaqZBXYG65j!tO|KXu`5(mXC!#9b%dHSqSu_zWdLe#x|Dd59cOUuO#v{<
zupc0?4(6&(uo=54wq`Tu<EhAdDxzZS1lq9w>+8qDM-hn4_67hf&JC_utX#!X&`TSx
zwS#?345&aArym4KgAxFPZ&Y3~qleCGLxuia<K7P)#GdjzYEztW33@9nO}6nIR4$k*
zKWYBpAb)x9evkgFVcc7j0i1<96R0&zHt9Q$(Bo-u9fDEFhocB0L3ugD5g#;C=pp<4
zn~?y}-Orxap0hj*!as329YUR-okIl=)Gt5HrZJl4cc%G69t_zr>j_VYR?OIKR&Sj0
z30S<ycUYa*5lS2xQP>6eoP(JgjfV!(t2h6}^P6?%fM9X9Oj@tK8Q6wd-ofq=QiqSW
zIOu5aFBr$~L!ROcm=f*AEw*frzlJ+TyjLnw9gNotZ#URlWbBk2Zh)1*Xi92#CGImL
zybprOc;LDoxc!aW=6Mkt?{D95Uymw0P3Dy7+1X(b9;+zJN~{afRt`d;xoN+VOrBrs
z!q@W`AU5v%jrYCr?ez_2UcH#s!p6pZf5YqBA9%lCK$J&v-*4RUg)4#Qs{TjMN4<+1
zaKCT-^85w={%`+|{r-m6Z@<@l&8tgpZbmGI^(f*=u^PuKY%n7bd#tfMD12<C1Hm7$
zcGZNy2yp57n>!Gy*<e1xjmd0l9SDABxmUeMMl(hKOB&+I9S`%_$S15k(KZhEiYTl$
zeG)kyN*RR=)CkY0r~!2HN+|uboVBT_+)NeIyc&U`t#n5ZtXK^DCVk9?KQHQ<aU44m
zc2`{Uql(g`ge%NZIfM!8D1R_<dA_ujkSU2eRlJLp%oo;!ZH!I1kfT~r=og0|Cc{Sz
zxkMlLxKPa8lc_N4<2Kc{?FWSn&*2pum&2Mtw<vs<tDIj+WwBZ6`|ahJwMU{9cx!G7
zQM=zcO0V)dZ2f$fK``&~G=K(5OsNUq%x0OGu;dQC78JA!AE1m&owcx7HzT9Rrg-Gw
zpCO*i`TM@_X6i`MS9=8&y$x+uhE&OK)lVdP3y_#_AIb8XnS{VbG?T6Vly~xezbb!s
zU4R4ccO};CaW#{zn$&I9UIZTNs<WzDg?!)dVly$EdB(G`H}>lr?&qTvxL3Fh+H$~Q
z2v#$3Dtuap4<XT%;vtTsz46EOKwRYoEx+*o<8MXLT^F`5fJ~wL{ld3zzvF(ts%huV
zPN=&;%0+|lxK>kIeO>sv9{6KD02tn}f!7~2&^!qag|-)&f$hQO!4wjf%0rxiZvRt3
z@!9`W9s@*mkDNcXe{i$_&p<H0pox;d+EA<?QbFGIZqq(9MIES{b)>+bxqhr6o`m0+
zgp=+jv`-uMhGr?OjDn|P)Y&h6&Y(NGZEY<?;GKz^MQDk{3PXA&L3{qFGds55N@o^i
z!a`#)L#!4VFoq0_p(79L&)u<5^n0)Hm7ZWxm7Qj67Uxvv<vu(3o?-P_c?n{6&?c&I
zxfc<d8pX}Yj>;p+2hrckP~3YLullYetJ(<YI+e9N0UhWKjdN^6UH<^!*l{pXGiS)2
z`AAIv{cRd?)pvzdH6Obf5>l|HZy$=Yav>;cquhKhpv94l1}4vA3deCyHa9Be)?goX
z3SB26qhgC~P{?))p=k_5YzOWv<-Pa2LW69%AmiWXPKzE&#XK@P>?nQjor%2P72dhu
zRpVneMu6zFFugpXQi@@3ybbW&Rg1wmyjkMn=2b#j=a###LiGo|_r0ofX!n0g&AcJ*
z8}HYv67YDG@-gM2Y95Q3#c+{OtFeB_O5v_vD|C^mI>?^Q0UwVi{`TA7Dvz?t3;Xu^
zhW)NH5(LOrCvoYqtP2k^lGfrm>y~&t9(dkQxD_RM=P&~I-bF~DhuED-N9fwy>zrxp
z6YD*i`Yv_lv^yHle-3HV{%09xd+_v5ghbd%MRRT3UrLNR=M|UEsWHTH?8|?4{h+6S
zPN>X=&5p$mg00^rlEzURf#>5U%GwbncTpRUu?rOI=-z?%!EdK3>VE^=z!aeIT1K-e
zmY%c-I=fkU47uXD4~1qd-10<(mx0Ji-mYaWq$)cIdLml7DL(UV<K4;Mlvi@hnbnpS
z<tzf#^MoVPNoW%zFQE}_vzmV~?o2PCJsDAk4>Q?Di*s5+WvwX9_)ZFqfSz7Vt%!`k
zX~@PE%R$caQ1lW0|0inEf{usK7)oPL8^sDchztrB)pLi2X;V93l_We5G8vGNz0uB1
zE%{BbA>yT`h+aoRs@6kmn5u%^U4gClsh?}z*y2U=XF3-eO<R;VdIG3!@FtmH7;Q&+
z5aq}iQv!*|jZCk61QK>H?CVCvg^1OZVO3|#<8IcOJbGiR3_sts-UCR)y`emu^FW*g
zSc6bTN?xRKZ`AE80erc4V%k|mhz#syN@TTCPSn}aXc?^gEgG>juRK=cO~dtg;PH5L
ztMF-AZU`IV#C2Wx%ddYy?xLdZ`)=^mz=$bOo&PZ{{3nN`y{A3Ug_@=wRkI&DRG0#<
zwU=}+=~tkPDB#CgGIfvWfz4PI9WXGoYrDMrKj|v`@EOfD1{Kd3nGkeo+qFCf(1s$a
zr@wGE-Q?v9kL9?oG79?Y&cx0t0@6lJA_=T44)hV<Y?KG!Lk(F8J0$p~>Ow7LwBeXL
z#2`Yia{_iGav3tKjjkh`aWn5c-bS|-IFUD`B&m@2ybenVtfxGhs?Ts>erKvLm;_jo
zS&%zs-VZnb9D-cgGBT9|uPEQaG=es-c+mVNL4gjQ5lYRdwMESD8}*nh&oP1mF305L
z{OQfEacPdqJKQ_rh(fKrX?L;h4Ko`@VMVv^Wu#DkrK}RpLrZ$DGAVm18mvvNa8(Aq
z;0!8NF`O!#1E@))!hM`qDHtd`HaQ)nktSWwTkGol!rmKk-w4XpW*x=*^{o=|@*>6j
z==T@Zx)RCvt^<V#)o`;iTd^A%srIfLkkLCitI)Z=YhCb5IhP6yF4k|@;Q_Sm*P0uS
z1hc%f`qkc>r{T>dQ0P|3UuQwsW8waK<GLOtA9o_I1Z?5?{KDgU4j@QC47Ht!H$Ae;
zFWBP=T-CekkKcc5Fjh?*<ripcjUG!Iuyqc$Yq&@vr~|#Y>NFT*kOfT8YbgUVc^-la
zfjG$(!^6&bu3lZ}#5W9;3pzSG{AbsX=M-|#rDl0B?eSR%<$y~GWT=}YEnAjoUS95=
zJPzkW)O_q#iwQH}uI;oe;VKfN;h73fCD72ylY@e|uB|Mo&F6DknW-Qa?bA?2Fvaqy
z7^>6g^wFSZO_lq+|1_@?SGy8>0kDtH<O_%(Y{ArLigRuN$^3(ggN!V~3<^YayYMy(
zIS9Q_yt;K7e<!7aKATsYGyzg0X&|w&w^u^b7hn}e3uM<8q~t^{T#l3|CTkp*yb4u>
z%I#K1(Lmyo0mUfO^-#RJgkC^Jm)Z}2zxzm>Cz(O+2ruN%?T6j6OI4r(zOsC61R^)y
z@g}>Y!bdU+No8*g8E9=>&v9>-S%Jd(O(5XQu&oXkN&s<x!+kBlDwpwozpK)J69rny
zWf{=bMPQLP3vTK}x%a(sT~ke8I?J%Acp?JZ)GRS^XPx_e3(@v5q1DXg`RlLMPQoiZ
zlc=`n54JwJNPUeEYY=rLs~H|^vF#0p5d~vozY7tsDtse^i!2$mRbIHQg~#*RA>>yA
z|M7qR2lnfR`GxB*f5k7q{1v}GzwqE1ZY6Oe@wzv@eR~1-M(*0v?pDW*z8@7{zHdZq
ztejA*xiin=O>}5?Ri1I8_p)sPJ7Xx}2piD@{%Bu*7P_I5$`vMJw7~^vq%UZo%z@=i
zVHXIxTz#|tT>R*VjY1>ygBDqalggN%W0a{k8K4cx<SuIg+{AQn()t2%PkipC5`xX?
zZXoqdMN7@!j)Z$uaSG>EXEAgR142V8h$=j1P$lrh?2$my@L*fj0ql*-40iFTMfdLK
z?`|LOs%L(fOL;x5B&J9#HFx^6)e0H(ilo%@^emy7E}IuL7(j$N(~dDX?_YSa&H-r3
zYU0R8Xnq;KU<G!lIPbpkS_^A!U{|;%IEFeV5dCf(b<c&0r}R2)tdjyC%Ad#+oVo&?
zanbn<c>tz)qLQ_K4S-4o$nHg?&;br4U*JIR{@D*Hd#KS2qrKmFzrWE?Y~-QDWN1GC
zB*R)d*y*!p7U~ceN<$*y;g|zs$}4a)yk0M!L7`(h9ZpW@W(Kz-5PCL&m@8yuVo{;N
zO2y@C9{1k3*^n>ps;V~byS<CZp!$u+qk0cH$r4|`e2w?U`Ljlo$;9-~Dtn#<@Hpmu
z-&EKmS2FU~^MPN!zEEw!4S&tRU;q98f&b6%zXOj8|Nd|Pj<4$po#CQzPeuXFZ?6~b
z{l+(ji;6;DXU&~Vw$&l71?7IKUdfsf)Vig&d5rur94D`%!)jLf9ff&SxY(}r;Z?O)
zPUg|}q)gYdi{4v$3_yqCKIa~P>iV&Ucm_V$M%YTv9~&}d0(%U--o5MVg_e7}j?MZ3
zyPBG#0tGppREg9QWhhz^I}_QYXxXYfhvKVz__)Iyqr-s3Tya(nM2Fyk=|D*2t__+|
zi0(3H%K#zRvP5DOX|c~F)b1oRh2QG`j5J71sTc;P#I_k}mIL)uXoMcs`8*GOHY*=O
zjb>oi)hOnUfA69nAr?SOHP01#1`7RE)$xYy8ejCE8na@h`f^Gj+ABlLglLglOdN$N
zWqhyoLD@A{eM8=eDVc{>G|k}BO49qd@B70$EQ;UEQMD~u?Ok-pFVys{62JHRRmwV}
zm7W}x4+DP~2$7i}P%?2c00vtN$bcz71VipD1sN5peZ5|Qdl>{KfxE#YUO>jMd8llz
ztI+(%*&mvF&i^}Kl@|e9XTM(9G)6_6W-lKkE?>A9uUAs}@p#~RJn-yeQI9pJ&>De<
zfA85HC}GnO-22A;`qq7dGO}1i&qp6ontLPG!q<A>FMs_j{@(^%zdZ4^9suh*QvH=U
zqAHNQ@$K6;{PD*h*gNq2`a&J2diQX;^>{0B4AG#)I$O4^_vOe|<%Fp)$t%`<%XtXg
zq*=E-YqQB<_C7C<p=BNiSU-$N9EEOGA4AsNpSXUkb@piY+B}*7WVgpg0+#9&hy$kR
zdsJEmBJ4s8vdmczss>U$Kb^{Hq~I(XvBRZv3ake`k89y#A~+q`sd)S(=ByJzWW+?`
ztb*pU_QGh>SIcb2V=W4IEzXW8B@Y<|Hpn^@-TQqrmY6I`K_gswQ4^Az*f1ra=Thh)
zTA8ZYZ&*qZDc`)G(?*2%J3tzAmPA^92$|N?jouUJYE!3~ZU7lT%FJw?_Jj;z*ICz#
z)F$GAI-&!%fuq5fF@Q1V8r1|Z%ffTKk{F$*JkN#<B2)!SRP|u;G9=F>%Q|%FzU%ZY
zTx|ca-^DN@yXfN}pJSP%EP7*Won6${33n1(HR{t`o!u_?{RVhoPW9GVo#T)(yVgaf
z(r&O>vb~zgqgn;!Aa;}&(8d7o_7dX03pGV4X?bTbRdWN1XJ3y;ANAP@)+6_)kXya-
zpjaIL?I2yyOZoN<uRs34V^vo|H^cMSU-6e;f303e^>Vmic=*Ee`D6~jvELi{sBnyo
zSwD>WN<LyWwyfrP^-fldGnB_9N~&sD3YdOY%4Z6|PI$-+m9O$O3`aPSWCcaSZNV)`
zV~**8gdBtEi8>Q2;aAB?9<qJ@ISs4iFjU0$C$1l6AIXZ@hks6YClZ!X2UYNein(IF
z0F^hJx&SwH(V8YUk4a5KZFm*R<&k*i0+iwmcmSStZi5+cu@%5FAYe#mD2f}k3PTnw
z0{5=5i9MwIk5NF=ye=9972S+*aoq9_VtDFOdMaS5el_6v0OM6a2e$}WX*?#QoCnID
z1(7i1G5#HiwBT72PVZ~?cbW^4n%ZoGiBz0<&?J!U*%V`J;dxzn62Xf`5IILes#n$-
z9srC**i`<zl<MBQjlp#8Otwj|LD*c6(ct>A5#5P#bQr_q9rV32Og($dVex&5z7<MK
zwhp+X!Z>l?RFsnC%Kx03r9D&a8M9C0b|gOsttVg!^%+vpn{oY<oxl#i;=|_aYIan9
zp56u;1f%zJ?|0WCs2|tPWV6P*#uMxs94VqpU$55-wibRRsM-;**j41pqd(gbYM`p0
zMX~$jQE=Z^yZL&D_ulyZ_umnZCmzoS*7YcgD--`sgy05!#|@9f`~9wN0!y_xh6Q61
zyujR>br->8a~(cZa<?82;4WsDD;ZADw9bGJ%|4c2=x|ZgaR7uzC4&qxcQk};Gvo!E
zCRI)H4iE#Cr6dT5qq`O)E9WaJQa~<XBYVQJ8@S&a@7I4tMRQJ!7X~V^ajO8EOnpu)
zP99M+RdUd!XpOpmM6xsL<_jBJ@Rs2b8$(E{ltRLSD0?}Lf_2<7xe0=w?H!WY@jZ_k
zbS~?D>#c6BWsZNu%&A19u^Ey=CFRj5VLH}&t2=5`F%ts7RX3MMSUQ0s(Fl_ZEe%8E
z_C4E1;Toip%e;?b*ojg^tIiG9{Pib4br=-xQ}WrF2Uw_@lCNvw>-pf(saG{PjKEfb
zml=_C;`AA`?(OO16mOLw2qM@KV(Lvu0A+wqr69#Q%?T$o(FU(+!IC+H;)G?;k~u1N
zpp^oHdR5&WN-;k7@>Nx-ZFKa!BgrqB<Y|A42RluO4<B;?cH`YV{Y+j!X5|zj5{O;r
z0w1d;?=(Y|_@$^XD*TegQY^(_;+e~FcQ2&cr=o4oK2vpq>Pag6@(297D$K;w*aASP
zw&tMAN^}GCXfyEu5N<q{Qu9`S-tQZmiaD7Fg5)?7#-wX4Jn+sN|DC78{qe^icq8Cf
z&2<%B)_R7C^?2d=c*57h{d(j5*6D5XB5Der&j)^ae!+k~zq%NB2EQJF)%n$y?21d)
zkJhCVzXe&^C;%I7Ui4w*N-*Kd){~ZMbXy)+Cf0SFea>|Ypqg<~*2OlmFvOkM@Bad$
zP#Z_1LpN(H0AjsKs$DjVh7eSmhIB)CeNpY3WgX5#UN%_*nMBg$_=8ry6+ydgC`t5G
z?M5ky<$bgsvZ%A=nh~vP(aa7y>}2lCHq`S(tLx(}wdrYSCR|Vh>N5}`+jT~OCsv`~
zyU?lk9Dzt&?^j|HJ+?UjU4?!VvsDPpjjWclQdC7_1*I`mT!H9fGxP?gk%IBiA+B>;
zo%#HrLHPQ5@GObyd6b(*HAPF6u^NVUy{2*Q!8k2X$~xF#XHnm<yFe;%8JeCW6B3_!
zoahWam}3l(lF)sj_bmZR_A1J}C}so!d#cLmzLnBByqg9sMAn{;Wu+X?Lam;Alvj7(
zyX#IoM~-3#6Z3iG4XUZ*S~vE3RKq~#I^OrY+s!Yl5NU-NZ|r4oySiSd63BPgSF};1
zhtQpG88T@iLzXPR-!H6hf8h201%G_|hR0P6=a5B_z7lwr37wo;>R8jHvL@Lf!LU)_
zwJXHxKy_O+hLa6VE~wr?bxPg0_l<9Vyl{W}hTxvQZ1ODTC`!DQ2b>q+ceQDGJf3)e
z`v!O5p))!xv0iT^fWQ9wSNuX+>mToL_~U(dvF%zFj=R(Fstr(vM}?wGpsuL2A}@64
znEOn{gNt<-buu5CfFMuXSFyX00iF_z^RylAurdfKpn2Ao{p;(;jDnONpHNEuKoj~E
zpDtE_!b3X--8NsJm+NOqw3)|^0D2Sjhm^@bCO=W3qY~^j$dUt6Sm;bT>YaKNn)m}@
zf~ag3W_~`g?ZY@7b5V{V)lJBIsA|)xLPZ%BoAG_{>|1?ixuxx$%^H+kB$drot}>JG
z>U63#)5|I(buO)9!<wCwCO2M%b^GPV<AKL|)W)z<5_3<haEbu5<?-CV7tfj5nLAkJ
zEUu%dsy!kKe76?S{Q0yz`95ZkoXw)K=({;MxZaoaQfH3Em|WfM`SATvV4-qga@!5^
zpeLyK_g-Did$cl?qHYgr>jX27Rbod&4gP-bH@fPEAxnkEjIrygPk2{%(XLDuP#{PE
zj?m{Y4JjP$$!j3@-q_#1A<c08{)9iSI_&28!2J$9GJwaE`NLA^uekAg*SQSWwd&|l
z$ulaXcfW7!_YFia)WehMywI^~4t7R`gy`|SUoRqQ;OqHpPwWFUaHia<&Gz*uK;!Ge
z*Viw7zV`EacAgW|O_p`a+w=Lve!sedg1O<kN>AMP&DQA7F0zSS@@q#<C2H0Nm_pgQ
zO$sGR3Zb3=x>iNi=>~-)(ExVvHX-GowFvkhp`?CP;$qwyx=H=w$zDKRMOav^5oivo
zlIEyz%+ciRD7CMLY8=s_C<~|K)TJKn!inWmn@BAlj_2u?GEnk7j(+zFHx+=Fq8k@C
zxoEl+awu<^QG3Q@RdGx}UKBW+IhxbEe3WydH={hHruh*$u4RaDL@an_dki0}8}w^S
zJi!+7HEwr$LUL^}%6ldXonKLCa53>H?<@Oh1PyxrY>1D9pY)@hb7!3-U0%71ip76t
zPz+_f+ojO3Z8pXszoKFG4443m#yBz(2-tojj0Xlq$zrfjO<ryrB?r9Ej+Z<hbI<uu
z4a+E6uH+Fq%H9k0EIK4j!IpQkVTJbhjhp*Pty>ZUVuVcZy|E&2^)sd(rS=qCws0SG
zBR1~E6G&LapRn&6uh%#H_S=7ld*c4(3;u24dOY#hzy3R(*Q32FtZJ+Ax?gy|-&o%a
z`}tK3^$1{qLe2M^RmC(;)h^=!rzaBfS~d$NNtf+EcFr>j5FCaX8Qp}>SmgHg^(!28
z0$XHatqYm&3i;B53}ml;?Zc6Swd#BbgR^3shZk+-#sEVR41)bWbQZaE0;tqafTYck
zAq5|86po+*<q|MIPG00b*1r8S*AG`TH|#Vkp(gqi2sQpigD#3c3s;Ws7mZF0)vEF_
z^zxFr4`e@sur%zvKQ>!rAb08T?Sb1juI<=f#%PDQ<r5?S{+gaX6&|VKCN<}5h(Hpd
z`D`FfJihZrhy*&h%%0|Ib5Ja%)zz8TwYry(SS>mcu00S71n7nyM&fkIYuvRMTC%e;
zS{1Y-jST7i17+Ib7(2o({<$_yB%on-hWQFz*IbNYQZ6Sy=B@vI*d!-&WL*os)bmP8
z>j*1-iU<zj_&fROb)C^4<A_<htV35a14gEgTAiyePrYmX_qr;#6x&+HOra4WM2^1!
zN+a-65>>MhAKsn7L!rh*n&h|=^F^W5ah|&kONHRaK+7NtLDzg2+HFrz9uC{rT<p@j
zVTdc>uc>4=FxI^T@p|F@W5Hr0-SG9xUl7k<v9CAQ^~B?HWT0raJAr%SP0#s$+_+Zt
zC5i~VU*GWiAAjKW`i8x40x09jc4Y{2O&_LeJ?@8rD`VA1?%E*^)LVw?Wp-Uvjcowg
zinR}4_Zw?D-Zv@?yDG;twCP=as2mL}d?l_cc)(J?Z)3&yw6ka#<ezA7?_gzFwnDG<
zs6=u%rIzlE@)~MiA3EwIr-2IT0tXoT&#xb5ADt)rP!n<DY3UV?F(Q7(&{d#Qzsf%7
zLv3E&*&&k?k_tC)5`=c7z#Oaqp1uK^u^MX>GQR=96z`4Rtcej0eM^dNNjclNWBV{9
zLUKC;xeudYl@wR+oq@rVM2yH{FVID-81ri7r}hn^@16OTnLq2aZLxN;52a}>qOjJ1
z3gdWJ8QqkpcL-63H>hltN0H}a9h#}NX<1KskmIsPmUopO4;MS8Z_x<&Y$VyJScY@0
zI)Q6X*rSu^2a=`w>^13pNQ=sm)Wqld^#`N8g*yRQz@2O$b~#mHkh~2y0uTeXReU~J
z+%%f$FgSeoB>lD#B0rn%NDngjSg(=^fJJxv@F+qDR1Ic}vJg!{EdwC0b5n{KLXr?*
zjO7{XlGmZu6<YJdfh5We<i6o{!6I?L-}tt^H2_3buOy@Qo=EKb1)%7*FH-^C8^8bY
zJMMjBWl`5fa@b@+DU3;9s12<E()+&gdcWqNrUtY84+$}KCP*E=b1^)%*249C!tPh)
zjIOGjMkelWH{Sa$Wo}XXND&5p7w!`V#6cNnG3w?TR9Az<zPW}+$Xa`ngFW`G)@m%B
za(AhGg`;z@{W%kO`PbJEc1g{q#i^i7?@;`8Q4M(rjm952t}qw$T8nLZS0{M^>k4FP
zg^$gihiglPpwqTKG*cw$r%~BY8w;<G;7|tR<}gvabUKyB5WxE`p6_9;J3PYq&#dPb
z(QK3&xB+xl>K3Ax3`n5UiX{aJgz5km(hd6t)G9;arl2vnN#`Wq=4K_EXJZr+=PoUJ
z!_PHJwL>cjqz8|H%wAiCy7Ydv@uI_AgZ@tjQFdi2yGPAhYCh#%aSTI6%{Ce`RsstO
zEh%wa?;w+dzHdT^xotJqaDYWZ70Usnm78x>`vQOzMc#u+W;JBPo>OMFH-B~nq=ioc
z)4_<^11!+gnZn({mhCMi`%T;3fjNHw{h9uwq%LKDwIS5|8@EfP;N8o!&+nz;M6Cx2
zL0yrxPiw4!+;#TBx7Qz7*H!05vj(WBt`+Y2<Mj>LHy%7sy3S$<yx(v9_WSR+?;Bs&
z1MB(hxsN=HUe!#{@OV7%oBiJJmn=|xy<d1g9(b&UNLKxZKvo%rU!PC7jL-7w`s$Sv
zD8^TL9>y0}-XnHkU4h5*iN~XkX@#%a!)y2HdiOq?Bwp>Q$%9a{M4-HaC9j+E;bd^B
z{^{6I?ePIdZ^D{x3@{YiNrlgK{b$z?vX9tL2>EgU77&TJ++-~f)aqACrC1S9B@pRF
zWLAOMJF6L<C~{!Y2LtuBDKN-c*O@c~fU|*DSI0{6b5on@KE2*V_fQNMsne|-VZ|({
zjYyUVD{Q+*e`@_E_Fi6zb-qMY%Y>7gt=a&_l}!vM&_k;CSZ)=5;@I^-wVtS$W1?#v
zY7rX!YA(!`g%LWZ=VJOiX>@uQBG(;!e^q1VC|+$cu-u-e4oLFr=`939F?1?hU5~1)
z1DrAw3&sM+vs(9DZ$P~p+H(O$22<~8H7+$EQUaQvuW^wdCBsd(-@tcWt5LDHjM;P+
z1?i`?=vd2fyAUkjlb}iRr*|k2V9cXdT}=(ls;=N&a&aHI3Wb{@2PL23sQxbq=0e6`
z686Ngm7J!>VBhJELb??etV3r;&XY=2M|(1qP)c;a@qT^7A5NpmrMg+1hz!L0h5dd3
z`);OOMj+mAVBdKC@f-4b0!Z8lcwDgcz`vpPk$LHmeJy-h;ClaxciAYcE-YQ8D>S@D
zs)<`<0*&2AS-)T3koPWQ^1AV|APB0%wd4uF42#P9aj6J0V7V&MPEX<fI7PD9hVB;<
z?_v{CP8A9pNp7M3Fv=&+FgP-+e!@J|To!JXjGso2{{=>Yb8uKF3~JNJd4-gbp-|~5
zxZvJ7+w1CwR2E`gc-8^jz8nZ{h&_FH#NKG+E?gMdCmOXil{~<Ki)Xyrcn%%H5Zbiy
zeCBs(zOxi=cc=V*M_Jz-fX13F9O+8nl~LSOgeUzHX0WWDPzlHO=yu#m*He9XbYj!z
ziVz96@*X86#CKk;P;Jn)mew%w)4mtxUX9N?^faEG3=@C@-PW8HTZRV(9Uk2o!os2G
z?f6;!@ANc!x|;~0HPkTM7|L)&DMsZ%HeLIg3jzxn>Z9%<nm&u~#%o<&uiaauo$R@p
zKO-{A*DNRKU=y~g_aI;ZTJctHhQG@VG52y$rEMp#F>)BH1Wy_Sc3CyEo_nF$%`=!I
zH><hp+-7heXrI(};2w_(@!jt?+`r*<UCsWhkZ2{R@AnI_?~mfUym$TCm9>Ap{y^M!
zwdnYI;Q818N3{ZXg`Eow{IVYS>+=`b^HIfg&#yM%{Vq{P)vdbGetXwZqW68{k|H)z
zXVrw0A+NNk6XP%^H`1+wp%_w+>j8tYqWr}@BYhOO?+y2NRbm5uaMBUi<eH#m5dg!r
zlq~^nNVieg1L!s_noQO-^f~08e#Sq&ez;EI&tIbOMJED#;SVd9q;*!4syYQ~@7=kG
zj6}G>V|^4NnIJipp27lf(LB*dS!(A>ft{1gQ8cQR@=ZWzgY~8|F0yyeBlVu2#aFDd
zVI@!s#q5*Wb6gGP^-kPqQouTxWM?hD3amQk5ZTD-x~lVJB@d|-V=8r%-Up76D3k!s
z%}Me6V2Bxx|IUK8y6r_iLP0>SC1290;QG7VfQtR@%%L)WjB!kUDRt$8jzJvnk*I=3
zv;$G&u|pQ8SSzsaipB1jO)q9qg%rgvALXYoNUW>-UnLEmirdT%`N7b$j}guMwrtj-
zWF&N2SvjXkFg}L8<(_j1%0}*uoq(@t%!URp%tdr-3mPu%0}(^AsNt2vn4&GMhrI3f
zyj$!O*ik6h3@=9&{4t?@-*0^T_6PR;YQjOX#YP#ZDz1CD;q@wyZRdtrz#f0#wcq#^
zm>PrxJb|yZuzr2Qmb2m+-2|{2Is!;XnB%o?{PFFD-+uocxC1FFGo-8wl*%ZUAz{C(
zn=~^TnCq(3>aJCE@D1Soe(^-RyX#lFz4|eS>g9FcH!_T53hQFVN;;PikFPv3-!ed@
zr*I_l*dLL@GWiqN|FBW`0KwlWx=gg7jUgDt>o!O$Gg;wSCGGFK41>;jmZU{iYg;||
z><Lo}A!$iRaj#=T1w$9*o>?QlvCtF4!#0?)qS~yDgF;U_hO*B5^`@{^I|=I!pvnyx
zH;*wy)gBiIU{*B-bwbr@RNBeR1>76h``ya3LoPXCv}>(ySBysEK<-1KkYV&eZ#9Mf
za44Z!(^gqhqIv<FoJ6#Q9$sJ8T1Y8r`GTlaOP4eRJEF&Fos{7{l}twEiAovDQ;;F!
zwrMf}Xo&p7$bGyULnCfwki-N5jBPHgH*bJ-Fo*KJ_YL0;D9j`{k;C(s0bzou6{mgg
zSssQ#M$XDL?Gv_Y?$m3GuNyj@Px%HN^Qv6ho&u&53muD_6c+6t?X65O{86niK<{Mo
zZB_?QeaCF6j<JD#1Fvsv@$FSfzweERH%@wYAlXKKJQ`v=9xMYO0`J!g-|lxG$J={1
zs`%2F_gq(<1-(eM2AG_Wjm(YLedF8fh2MVr8y<eauB-J!<#kvT_w5gOfJOs&y}kh%
zSeJ_8CYPn}iY_a;Zx0Uwdo}jQb>aQKJBgl<tUCJGRjp5NLV?Z+&uYIVV`N3*JR$;!
z3N6b6{}<PfQ&MYOfDo~vH{2wo6e{9k$Z92-EQ{2=<dK+U9z{$S(XIsk$hIFwfrdc^
zR*i(RtWL9m#6=Ernw79@H*Atlns{RpFM;YS5)<=oq<4gcK;F90Q-=ftT~EQ~gdO&%
z3`kMq9*?WqMJ$~xXT4c;+~(`PJE1JKrz&72-yZ8K!vb^?E0QHUVrvOQ7^=@YtcB~G
zn;J92UaHnT_WmTx_}`IyDnk(cWyhj(eV;6&r?V$XJwjIVFBq!kcOHt(`*>h{=0`n;
z)&OzDY5UeKb3h?QWU?V&PIE;+lIk*S?9Ee+<r44PCj5jfg_xg)djdF=heh(Mwh7n8
zIPhXku}d&RM>-DIO<_j>+&k6bSIPe1&nTz%S?AE35mLgVoYk)^dtZ?$bMHEPLiaV&
zq1{^Etwl+@#o$%?jEr}Mp!Tk_kb9k@!osGu9O#)+xMZ+_QC`M)s43ib;J5b+@y837
zIkIJDzyl0j*cW%T(D>uqAGpo1{_<~Cn0PB+u!|b|+uvZ3ea`gWyW8?eHp>{R*Hjb@
zhn3O*aI*`mokR`PI{E<=dX34JtfcxUc>MsAEzI&1QVz_7?-SLBjzIj+TtDV1v_M(^
zd(h<|%sMo}(8WNQaD-|Tmd~k7jivw!r(S@&5QUCXyZi`iobI=*ayOZiw_=!++BA7c
zo{2GsbR%aR!T)n2<3WCB&O(8_L0~7f(KNc#v!P*>7b2sz6e7_3Z*kCvU371GP!bL*
zWT2~zoyZO5z&0RxD2XKt_{O-aB6IRXIFv^HCvwndC4W^EQ}46Mj0h<fc@k7CLQLK9
z)P4Hk05k~d2{j0N#$%s}H@5JA=}ZN!2iH{vMwykchY#gbYjKP-GT=E?&gBG=aw|j6
z5J@JgHM=tukO|K<1#K}QzY3X7duUDZY|Gj_r?*{GpXh=%l%d^BRBoOG2AeZ48l@7U
zDg>95bMxxWbg|cx6ov8w)u3@Vt8F|K^gPKkYWnOsS*LE`7d84*mMT0{3><Y)7E3J*
zK-dwose^s?eYXcGFS?2)t!sbMN+sM8`2YUL|A9aL&;P)+t8t;&Oc8<bs2*wwsL1Q`
zg~#>4e!r@y?)3ubS>1a6E()!R`_{FvY#|X?*Tn?+tk>qDNb9=reEeG-lN*V@{q1jX
zKl7*w>{0TWY*l&RwQmVtgN#IhEEVHw9W})#)G8Tgm^J~0*Z-tV_|fweY}!9iqQxUW
zD-w&-T*WhDx#)86a1B3@3pur$+<y~yI#Ds<ekKhEUHAi-LZ|9v0vidBfJZSpT;p_o
zM#v%C-73plWYURw`?zhXSFQx05HO4_D-1BZ&?lM6lGA{x3#F2c!^fWr@iW1?QC#cy
zyS#(fdw1yL>hKPCgc0E_z%Qo}U|K1vYpq#U^RPXNuM@T=bgIP~BSD@Ql_)-vwl!{2
zHzlE&!y4n#>bsC&rZ$5CDXN}uwB&^fsqMiS(sh7|HSm<Eyw;Tc4H2GjDt{9Ih%u&G
zM}dUefXO)>;gu32DXiut%GjpfS%AZmB1#WQ?FnS?`J&}H=IkS%oJ7*Vm0?g&s^%;*
zv5epMTJH{13KfIrJ|9L+yJ%Q@WlHvuXkXo(a^|vgBJMCMZqf4-+CSiNy9cCIs7(Ja
zPaX$*%F2sRB#W*<!ohBksCyZ2`*=Kn0N$@}*a56?e0#s~-+%w_$k*>RSISLBkWLEZ
zS`Yy(v+^VluX8ivzNxEvfuRsWsiBd%@{S&OeEovQ^9zs16YF}g%WsWwSt)Zj!(V>=
zwb)Dy;oUqaDUs+?*ErvGDqSAHDXH4HcVhJ2Y*zwB9HH8r`j?dMJ0Qt_cKz6S3P+Mm
zORW~g9gb;<=R#^={Ej>eC0N9j-=XuE=VgEq@^+Qf9z)O3SXyaU+z{^)Twks{O?v>J
zXq5OY)!(Gtx)a8oQ`xa^y^NKOa(W|)3Ln)0?G*U-6sC;<s4%K;y~B4^f*QL^-b73H
z-hKW8z{C{uC5gnQw^6NO3PtOXb-B-HI3@9&N7K__9}VQB$WGpS-=y_0i)K>FYg#PS
zri*@O&EoOMHsZ|`g%D4E{QDg5=!p6E1cvHH?QKC6X;jFGn??8Uo-=+VpJ^PV7)mc>
z({{VPGv`aDr4=m;>-H9_r^fn^jBR58%=>tMR{uTTtQEhvH?!z@)w|6ao}mKGxltYr
zAnh0J!K&EJV10JLNX)#0na(-MBRAAmzJZI`07bE1Hd|EZF}F2$=)I<v71bJJKvJT-
z0lRPP_Z#+Wg;e)WL>OML-(h>h0-emg?>FA>ci%(cP`}%ISAV0&qoW-L!=0tLqm306
ztJ+^Du-1j=;{o?;Ji=?@_HI-RNn?90?3?-|iZ;5{N?(?w_ldR6B<2!CDx<4Da5HV3
zPaV-YjL6KyQpomW^ZoPde=N}~Ap`m6{L!FDJ4XBW-!1o=VFa^QB)Vta_xLx3s@z7W
zi<Z5|T6O14RAl1)9(UB3$7mX*k>#NY#N--u++%ywZG%WvLwO&)SXn1hUCG1bZ5FS$
zR?>6gMi51JXW$l(JgU^aS{e@z+OMmS2bhPrICoLck-*e=W+S_yPn=4Cd~CHX#>Zx!
zg`GJ}WQQ(vsLPnh>~BaZPVyi98TIa2Yt|4rP-A`<nzJzouup^V{q_H&?O%d)+qU#D
z=o@pc^=IZjk9+Q;ZmFm$3CXGwk|>O9L0}BRV2CgZAi@NWAY>B(bO;&{#&m!#K?CSO
z!#orwAlxD#puxj5+#oO@gJk(3TLMYQQdz1})vf!edmejd{%g%SXpHY0W3D`>?mfGp
z{u6mJcjo`EHP@Wu@r}2@fh+w4j!L6<sOeyC22imPN~N$ikQSUQ5@c$uqoGBGu~%AX
z@2U=U<V^u+R@xXfNC}RLd!)Z0Hd%ZSE#i%!sR#XdPwQK)*v)`VT${&b+-oEcD2Tl(
zb?tXxE2XY#SWMb>g&Zy~)yq_TBeKNW0-Egdu>?-w8A0)pI=j|D<ybWm6HA89a@LbB
zoR_y>;QsUhwueW+ED}^+p3%mn#Wpd{1Lt|ioF~RK_^1LzpHUz;+3QIKh}*Vd&Vl3k
zj+d8b+>blv7@6fXeZZz-00vl$BfZ^jc(^^l4ZH?H#GuRUyax$3&gfu0gBk-mO{4|-
zz9MuAqSftU3s6WP<+)d>=n<B1YroeisYx_mXoGVI#E|P-`3y*%0VkqSkcLHFX#p%7
zv$&+>@~O;~<vzS_5+Q?t04Q`@1_A>yz3QMzgw%V(wzWLCiZ=xF^lnT6J&6{r5E#*5
zR#@v<or*;QP$s<s?%+V25jotGj2j(hl>)o$gXF%~!`JxA2}jTUwbn3<#Xe9)XKZk`
zMRu>f<RPUs!=S$g0BzRgZCYefWM0jMHY>ZSh2_Qb%Nh?zL3_Qw4l<tKibgJk?mR!<
zTSqTNR(N4Gr6bs0$}2wprAB*J03fq^_?a)i$`aA@=;l`ot1N0gqA7@-t`SS+kcyBo
z3Vp5$?26Qd9@qORFdbmEYCk-m$xsvHu1Y-SAE<M|^;xz5b3HH7j4l62K0ra1S=<`J
zU5ZN|iO;#s3tgB25Mt*!-Cy3~i+7(xr#$Wlj`NO}FWyEQ4QrUki81amKid%9RO$0_
zpE%EZX7OznnM(V<<2amM6@ot;Vc~>T%68orv5mHEF?#g&0QIa=wv&$uv}mfQ0Y2)h
z)LKs-%brNJ6|cUNUtSf8+OofTU(&CuF=BjTS|pqNo>&Tj40wnld?4K^O$NfEu~u0V
zW|V|;k<E0+F``@Gkkvrprwj*K3rS238C<+rC5upT4>P`_DjLl!GHxz39SE-XhJEib
zRc-54Fr@tA3zx{J-OxI)Z9US|mVHm*nQ$j=+q4&KwB?{_e26LIdK-JYq>GEsr@b>E
zgX&Q*v<jh?{5fL~EVF<-EfAA}eI?j@A=prgwOx%B#_C;V%~H@Jk+6hrDNL`&NLq5p
zw_nx{xIdji@Kjnt>3ox>ky!)%te*Sx`iw+>Al+0xJ_Rs?!z;4YxKHB&wuP=DXy?j;
zPOzTJO}fFCG}{^VFK{Oy=n*gn-)_QMnaU7*kx3au<TF2!!Wa6S00T;B*EzC!SM?UU
zKC)2&pKVd9JWIO}aw#Z%U){Ya;TAqfHy#!R5AjX~=LIkUqZ@>KtP*cpZu8|C@4omH
z?gJa+0iVD90$+UgS@8R|0dzwh=SzIvBJdv;R}3z?&#(e+4|}w!*gDJ}@bu;l&UxZI
zP8{t8=luv}H4SbJecP}-JYsu%#O?7B`*w?3oLL2LgS8^l9mhFvj1eusnK9?UVu1);
zRLml`5-KxrXdsj=G?<F>75`${=P&mH_{p#LvlQrm1ytBYbDhC9&((>l%C#inu(g_>
z94JghA*_}hQ>u_cA#R;P9{F-52;4GIoC@fbyFsG_t%sWj%WPTstgBUXG{ft?V`5@7
z!#U`?&NU!2UtmD*(aR}E?k!C1SWp@dF@oouDcvKjO?~Raia=I1cPUT_Yd&g#FE1gq
zBKq6|?#CtO!@T?pS|^cVVQ`PEpp${_=3I|8Y>mZxnbB2aPV322jeY4%6mWBWd95HB
z!wMnGMJI;KtQe?>bFDpC0_d{hNHcM7Ec>jKQ%mKeRYzu#Gd*=7Yro!0-;$X}oVjBb
zsGtjOBWbkWvF&W@AaJE*0a^|#YKZ_a1JSE{jHulbahBCpzWrF^VDVYdT~9i;W)W|B
zRcBDz)Yy%-MLTg5`sd53db84<>B5O|7t3%;cQAFkTY@8H6e&Y3fsZMU7MArm2hRC|
z=k__~9O%|@9(R2D>1TL(eokc`!?l>$5ht3(hU1K;cJJ;lczArG`zSKBHuh;6<HY^l
zb8bf{*MWZ9aC;2J@#gUfj}MR7dXJP{k1?dvqxX}KsACbBaan~#Alt^_LCWlFOA8Xx
z#Dzcnl~q83&K^`KU?TdCR9NFtem(W@-a~e>t;CoA1hG_p_d*Ho9PUko8J!P1GmgA+
z##h>mvy5=2dW4hl5+U)&T9f71m3O5<i9Xa%No2d2@{LWYK%oH)n*2zFylo9_3zV;C
zf)0}=VN`+=sVz3Y1YX_~fnb}}!<X<z?TKFrBFsuD-6*PqGHYlR1ye^`gMpEze-Y6v
zg`Pq??ls1Y(WvKuhM4zdHg)E!OJHMBR8lZx?+T)ZU{h*vp0fx(K|uUZNeqQ}%ayef
zi>is6U^{AGY{|T7){w2gouEVa;D;n|<zDE%fCw(I+=WO>bF-M+F-9papOsD1t;4oZ
z2;074>oEcCR8ghSKB!(76H`{v@c8P}VK&ivDIPVO(*~*#Dq)IGw%KNj?m1!RXkE$E
zD-RXTl{qAJU};(9K9E7KH7=$Nz0jO!!DumQ_VM(A>`w(`8w>qi^Dw46j(hm&r%&8p
z?l|7Pz~^AEpHMPc^hQB4DcC0~li$6(#rAlkLbF_Rw}yS&aeKH`qd$+d_|`kxw&P)c
z#N+<J?4<^D;IJjo+#+>t%#(>L1ccVm`v%y=?Y2iI6`o^-^!*D=r~I_U*i>*-0tC-K
zWAv@)u(9~fp`lreDe*O4@7en(((u;a@QMCVChK~CEvH)yypPVPoSSN#p&-=P8k&PG
zE<B6gR6sSkspuq}qMy3o%1f{Qplp>qU?DY{V?t_y9y`-U8T>Q(cN?oEcl52%v;#OS
zHWUG~I07lCwB3;|pC<GUBFA80;V%WHn?+$?{%3^};jXB4qTS5YlA>r*j)C3=oK>kM
zBoe*#Qm~;AJSsp((jJ4QUv5>MDBtTqB%6J)fF$Rvv$y_C=*V@JGE+TE)eKa&hHj8y
z1?{YCX|+9Rsto5^98gS3gP(M+W{obQD+<eu2b8i<JSt-V%3N7c9>AVXGh7GmtQnvm
zdfz&_f|7Gy=4GS-vMOoqzn>5a(s)oHGHTW;f5oiU!8}D<>~%znYZjY1GGOk-M2o`C
zbv1(_g5k1u)A~mQq=O<99H-mJ6wfQGvnj}K;j_OAuXR7=l9Z72aPJ%k&f{Ql9pf!|
zHbA&YCn9TcV9XQeIB<V?iMi6p89b5ny;$je%ln(fyoI($8|{bN13Fs74m_;G<2>;8
z-50d*!{xVcF$cQ02DKES-LkYX={RT2hfb^{CCVg?3rL_v2FexjbdfGS2NIQwQvPYL
z_il){63jItq)bXxe42&H*bF&LCsF32og<Zwf`ByuE0au0-W(4pH%wtFtbWqGx7L%J
zR<?{U@7DXIf<hD5`Dpk|x2PK4vi(5>QJYI(Fx796h|kC(I!C#V3XdSK*;)&vPVieJ
zCPT?|nVj5E8%m9>+J&A9P4DIY5OE99%RSUN2P*#VsZcV5jEYIwB67v3;#Iehk}8`(
zn;ueljn9^Hv#98uB}k)fbAu^Zw-iH)%rfY2ikVRG^?qSq<8!rT-?H{#wOpE}%`L@g
zG|g(z$r{fPP*HsjTPaW&0->i@C7~2%DZ~(<J=fe?#C#ID<Fa8%qv8)+&et6*(!)a0
z6kt8F)GGB`$L%S^vYiI^H0w*2v#o*}S_qjb)lyLIdm!GXWD|}6l`SWlNjc$_M37&!
z+}76I(<j<w-I3^X0fWOe$xvV~?ht1x(ieqBKcnd<juC<B^L~VtaojN<pOVI;RFyIg
z(Dl8wEE=r+jQtzqjJk<&0=>h$fuW4dRGYHMw+$P19AiX?XxX215$hg2o@nDmCBd7o
z?j*USiW}=_({24VjfMaJHpFu*%i(e@knGmC9IKcIZz{B}hiwh$Xvu_$h-M(hL_<VI
zwxCm8V>J~n{x?{nJmsEa(Vz;N4Z46T3&GIPq2Vf>t)m0=hea{Gs*R<k&-;nveuiSu
z^N_)WfIC|T$3`KAo&xLI%@oE$)F6r<C6J}CgFu$&Ub1c0*GhEUo<t-@0Wt$*Eoej*
z$<;|OQ+K63e?_$EHsBepT{b2BHJY^dLXgD(o(s<=br;1U=DG&%F7#QKSP&scBq|Lu
zBw2!wtPDj3e!Kelx0VXZl_dkDQUl`i@}El)X-`()zdD}k%oEwk3IW&rYL7$zR%|E&
z`-ZKDAk#Xe7y~pFbwyBD;djU75Ep|1D_OBc!?sl=y)3f1qSti668t3yY1K)V*Ob*J
zjdY#kPEa$VdaE%Yt(M(#vs6aY0%F?w>@nu_IPREdwC8?)`2y#gC(Lbwd-Qoy1+#l3
zH1tLQ=gu_OLFE<yslI9;<jTg_%{*9jZ4KTVwuc8iJ-rE8)(oG%eYXH7onWnhC0w;}
zPs>ZPaHS^%P!MX1Oi1zB??FtV%;N=LlUSw2%WsG@Dz@IjT&4h&rV9bg>=vS>Ew-3U
zK%qSW*kstq6UmAROX%RL4N2Th0>V}Sf^?<0J1b*7)5p})pAtC<EcXJa9`e=zrmIOI
zs9#YOpv3@hc%?vg1{*t(uGW{ud)jZJv%d9IM(&Pw^eFZlrEJ82ZiV<me|qFh&$6ln
zEQ*;>D3vY8LxP6dmXQ5Qx=8wHXjT~*Dc)!z3pF-2Z$oxEMWXSJBJf7Ym$4R^C#h5z
zACMAVLH1G@ssBU5Pra%dr`~n>VHakazIGg6gp(9QOQj+p?WvR&yvFxb;Zx8CmLXUb
zIA1M7Ur>7qa9PBc!249c9&M9lVAf>OruEYD8K;*rP%KEo{Hi#(hDA}xs#FkPbS+TB
zW5;FQIFPk8)(mZnOr8QKBa<s4sJ!$HLsTFiN2G`LEwYz{)C+(h6?r8v<lS?>#~4u+
zecbWx<qJH&JfV45a`*ckFZUO`Jimj_869w2!)6hu@bkp;{T*9gMB}{Y;{`9r9S+CC
z<0C$J`T(tUJU_pMy}X6DhJC-`&6^ML^zg*EX2atLU%-C!+fY=;NV`(LEoq3!ExiE3
z0IEGnQx&uoO)0jx($?2_y|<;1#4Z8^m~~WmJy8OMTEkMQNk-M+4mC@212b_P?F@p?
zP@)*gFU>-TT(y%(1#w-in0487(ucA`0HirD)D?(F709}ro^!&zE0_TFY*Csn=u7rl
z70_97PFGaxuiODm0;WZ;6(EIO?;ZQT#jp~y%C9~<&a;P-*f#Wi55H$?nf;*=q(Zww
zk|Hj0cm{z(Nsc_vHAPj7CX-ZCuN(*yc-EQcmbr0QJ*^nhf)t-a%@Emi&(si*!Yw3A
ztcJZSG>-v>6n2Ua#RF6!m>xv>ThFk%vXnJeE#AX25ef$i^u%aMolVhN+G~sK1Q}98
zc)}$YRlVKW1aN6$Ct6O$EwIzbH)dEhmENx`BrH~knU}v_phBZG6ELKWuG7wi+w$A%
zb1B%aS_T0UepH&7$TVG<B4{3*GQSi=UO}ECkh#K*?n)`wY;P*8)&TZ|5^a;QGjYWk
zV2<Z;m_^b`rv)*`z<Iypet!;2;Ec?p=jSi*{QNFt!;(~C0DPW!d43y-GkwDtC*D23
z!*Skmdw9ab<5SE`u!e_+M?5?{VY}_PJv?CVJ#r+-j~lD0hxUw?Q;ACg(meAe9>!p{
zu&b=O0&fIe3tfMm*L!wJeJylB29vz>m6G5nRJHY}<XaJ37OjEl*IK*?n-_6Iw2N4L
zZ;1*Q^lz$?<}Nw4xJ;G^I#sEA^}q=<U{D3AM;w6yOB?GhAsPLT@;*QwoWx=9T?k|i
zk754ca(Kurr1aS%7=OB77^gE!BZIC>&>M*Kwtd@U&^l-xxkUpvBFhq_;iePuc<v9a
zD7NJtiYJ)->I^{7Yabua^OCjJvSos(W%{Hanu?G?K@qbQf&AkDF2!8VUN37OpKfx=
zsGcLWNfDq>Erkwc2E2K?Vf0nPPb)7gbfs(EL&DMF)&3&i5$l#Up}z_QX^LGmup24x
zq<oTybX!XnFPb5OjF^OCA}#&(tutF}3she574asZ+6j0xd2{%3p6fHGb1CTsL|7VM
z8tpc`-f6X;uCsMiz`ZXsNw9*jh}-1McYZ}`?y^2ZW{&C1(=M2rH!{(EM%IwG9JAXy
zTi!QA2jJs?<A8g^IqtY0&lu;y!rbJNYS}$b953%MEDF(Vp15rd54VSC#ZA7c8PJj~
zrGU*VH9ToeTkd?eCV?K0sGz>XV_pbwQ3|?Rf|(}J1K4Ux|Ag0jrljhQ>py+bOehKl
z@?dN3v0dIep~hGA36d<8zA0u(e`&eL@-`xl@Xs6M-qLGKDFc!<n~C^bEd^!;(Vd|K
zRVA`o`r5L7hqEV2DFhhfAuVDcQaVbdOff@~K9Fz~CPlR*EeQ#fC1}~KB~t2RgaY9F
z+I`z_W7oz^H`RZvjFO82LSfp&6?t-RGkuRPw$UBuoH)+tK-mW;*JWXGiPBOqma?Pn
zPvMv%ruH+*i|&KWa%wRM-OQHBbZv1*qFg5nXh{%bEb_CZcxp!TvDqR6E`ctHde@3Q
z@cOUmUwU?^3RjeAX-QO#NSz5BX`zlWqN67_DSYZJ!G&C9tI$(SCxD0L`{Z(KM8TcC
zup~&-`BgwmmG=Tp)1wg8WN~oCYs_Zdh}=sLKRssc{e`AP13dq*Tp}K!nhC%&D4y4d
z2ZRRnh=O_w#(1@Eu#q>8{iRz>)>AWs^@cgQPg~GfxV9U7o*3sH!-3;|hmR9sSs>@r
zWW3K}&*nG*#M&gd1F)#6UYvE~X+>N16TWR|-2e0b0w3e*Vub3zNW-trHfu&8dB^8W
z+OC9*0!!s$`)RNDu2YbUA%%Ne%s*~KXtfX#oez>HxR|YD#AGKxh^J7-C9fGpfCRE!
zLs=HBl@P`9{=5J}NhwC*%lAsBMD`wsHrQ1dWSA}0M;+Kj46S=rZ3SZ6=mH`2WR*7k
z=0ft3x)=pw%3?|3x4229AbJZla@#ohNodVg*i>hdg4q`3Y1h>|kcw<9y^Pc&XUvIX
zO!PLeHHoHDa%&(=$VT*$XRiuhPu!U(NAZ(J|EYqxgVn`3LxBO_Y)$>@DK0tj0f|{j
zPS+T7yL}qa%g?@AOp^*V-A-^dO25$8n&U6?zF*EQ=tp~#8=b9Hsid?Ay=o>*>C2ri
zSSm|VUp$k^)J}x}6Iw@t4CBv8eW$hne?Wl0;+B^2>%El46X!dMFM!IrLuF-U_s&;z
zl|q`&q!iZHIBIrzV28223h9?;<M60!*gBEE_*jK=qI*vlaK<p3RojqEp{2W;i4Bc=
z)*5PXkWcuOYZA%tYYKK))r0j`BU<X)1{g5UfpbKkrQ_wq+m~nTy#q4__`Ezn6ChZg
zPsl&Y0;gp}zRkeA&@qhrUqC~yL_nEn*MR;{d%b5&fg2(959c(r5}=7T-5JP>R?jl|
zCCDdqc$2<TK&^A@Zjx%Rp65!n5rkpTu`_!??v7v%rJ=H<Tw0%-p4aB@K%!SrGV2Z;
zm>PtNAQ0Cc@bl{PUKOTXtXG4tec$l#_<*OU2W<Neci<c&iCl`HQvgV~$cI?wGTloQ
zj?rwZdi6xR?oyJV63fClF9DFMjf!lrY~Nj}aN3s$YL4DwU^!EWC47@!6+59ojuz5l
zrwWiOpsSQaO`ucXc?mJ0y`A+fy@fy#n|`U$Ui}4YA0GjXsC*qM<`>km(3zvQWl>;4
zW38I}nYC=M?3|&zoddmZIBi^%dydz1?bQ{;2z6f3e@k#B^p~zUvr;;ZPbe+R3kuKa
z-$ucs{gAaAaiW#gJfFS3)~(9b)}Lh6EW$jk{l$8!4zG=Er%M>h%I=Av0bpq!sDSI{
z@G-D$4QI1R=vXZ5H9JRgOX4eHZ?RmtcRU}&GB6*)zOo_n3(jidu$0D}IL;T0@rb!M
zm^qGd$NhN0IPZAA-|_k98#)Fsqv-BDUP|b_^0-X6x&>mkN-1y4mWwv$bPU#cMDeQl
zrW#_tLQcQq>%HAa=}(NOAW;`8%MBxe3e|Pa4bJYUtn!UUc=Ww{`2lls5bv|T7Hd2O
zPnlLCaznaH^9%znt_-LYNeTTi9c~qG(0tXE_6t$+x^z2D3cty=yMq}=+Mm{d$9=;$
zKKKA1e((Vv9(Gu3(MAHmN%vT|MwFRR#3{kO=FG#LEfP=zJRIywi5M=TEDfNlNNI5>
z1SeS{7nTCb2Igq@CBO<#OUQX33f@WxhziZh^3ZURtI~}9+}P`W-y&0OXR4#xe8m1m
zF_3|2q<zb?cA=)UkXFH11e3IZRe>m0CW%|zv%Xi~mCoE;g;=uMl-e3Gg3jvYu}6l@
zwr!~xdT&{{MmT>-RgF^#fACsGKgZ0NLA=Q$Nn!b%9C3+9JiYwOPM`o+DitaQqJ8(8
zhNnWg%XB=sn0#G#xmfj_0O;&MDg!CTs;myijmqrFc-ExMV72d-Ta_IEoQOVpk9C}<
z3PcHhjLsWc>S3*wyJzbi%>rQbwh@#~051_kn=>Xh5^OD-id4P<(=pEj&tH6w=7#h3
zh-QxGmuJl5jy5O!e#grfU*IjmHGUokFwUg4Bv<VNJUlUoR9D4rx*F<Xk!^@dcPR+C
zr~6I9Yqe;9((89sr(i`BK@jcAIFMk&k}}9;1+~revk#$C1Tm{CXFgQf9%*7O0FHvl
za#M$wuT5)-4=SoDB;`rKVZ}IG4xnZr07;zD1yKxwBya^QDT32FSle4<UnMi#ZW|tM
zJ8r$l6f-jZu*tMogmuu=aIV!e8Vk6l`mo2CM9W^vem#RGmyYlepVzJk9Asvg!-mxv
zQ?U2evG<1C*08UpXuhcm0s*TmEYDSi0^@nZHGWgEBcxzchuX_*h^3%N)<~gb2Hp5<
zj&Oyj(o=4xNJG@Xb(hhC>T0K0g%n%SUb2!>X!&ee;Vzd1Q-)zc<IozlaZqoawCbz_
zZKm*FQM`D{(3mL`zVB`8IFoL523EmD$ch22gA)2e*davJ!zjt7+=-%tRfxFSwRi^W
zwa(Q~Z7cXl-ler;bdQ+?v;d~pv+&<AhIB%M2c0a|NAMl3U4o?sSrzr31dQ@dh&O$a
zC6Y|8wlsV`xtK{F*&TL~Nlu0CY75}NJP+KT-vaYQGsJ)&XIiTPbDWsx5&J!2vKzqS
zL<2N*Nzc!TB`~Xz#Q{2Etr~MH9uxqR{LEO41~Yt(*L%kl)}R>{qDi&I->c8j)>_U2
zh}Yx+u;<ZaGc}^_wXJc!76i4X(aBF5TGNt9MPrGW7UbvY#!yvjDJ@T{MQvvzJ7kUl
zM@OIKuXhqEZ44-W5Dm8m&T}NP0vK#+4coRQn&(py4OtZ_^`2$&a*?EvC5jlw$<qD_
z@5|rF;LQxP%@q1Gc0T*TxPxV%vOqkAgJ~~m74~Mh^^S)v+A}m}aIAEZgqezy)(luD
zK|!K@8q?VVM1qTbrM9KOF;9`oPO%prm?>SA){RsjfG8b-Qb>)8l!}!!6#n??<pgjb
zNoz2hS@)5tr&6#Me%h_8nOjSDR-<ss)*0#kk#JmEg)IG!v~an?1UTHH+2Y(b9Q}Zg
zrN59+FD-8g0u^Cy)-yvb=7W@LQ35}8zxHjhQjA#?PLwMCck&7-AO$*#A{JNiVu=?t
zk=Q7URP<&t15#-(UyG8<F{~F?K-?iR3GK3`H9qLmah_Ci%B&j`qs?^xX54E=t86$@
zR#~6G3`L~a&E&em1w7#6gg0Z1(u<ZH@EMiN=Q+~yia1m}C;N9sGrMVQ>(ekNdm4E!
z8JX?R+vPd6pUsi6)9Xq9&ad~cQ}87hTFkp>b<k98rXaJ3m<SC@N7Zl-WW{Xe)GugP
z-?HVev^j{TE@@U?4CB67FuaOU&`lvlp+;#_Bm+ccX}MTb05(U%Y%Lm4EodZEdladS
zbK*RrD`k$@6uLwHGX+dT<Yk+$Oa$5y1<Ra@SMdV10wLCcFIG~deKNa=e$nA!6%o&;
z0OHo^Ds@)wY^}_GPy2?Qsf4MK6;m<<gUzb$l<-hjRG?BabN)lBcm`<MG4o1u%2ooV
zDO1d@!deOyCD`IYrR<P7;JW{|M-Tn|wz0qw=%ZuXXLZ)4fSy8HLZ8YCQ>=#y0}B8W
zT5hY-P*#=t?6giSDBLPbg7K;zMM`05Y{2HZj}ne(mOXCbu@(5G^#jXVp=hfR!;(@V
z^Z!l6xVXw>QE2wHJ4%o(yw@m*<1@iKLaMUL(hVi|uold}XI-6iq!cQwIx!~zbD+(1
zJ(+bF#^t<D^2h71Xo@#5&xy9}1VIL;H8ciiDkN~zzD^?U9?HNcvOpu(#(iV5!AZpe
zSqq^16r9e-uF!E-a+OwUu1L70@?`CiwvN`(`cjO_2q1kL@A36s@2wP8g%Hdjzw)ve
z$W2iQDSfjuDwxj^XnM#`nL-Z{02Q=^7!#=^S~oUYE7a(iAht?$=m5M~G_I0`q0N;d
zlH!2SwirT+Y=tWP+*<U;)*FBaQL&h2jFA(qRsl9=jMv<^GyKbFX?`}Whin`swg^?C
z7ZpIM;7RQ|QXn=}u6p@ZQ;D_Zu?nRclgMY9&5N?NI``f?+7?P-hIB?)2r<$qYfu;E
zy!xTdkggqWe7;{+jW72V4hzeY1!M|PVnQy9?0lNYHE)la0qyx844D?$d-N`9n3<K*
zdo(-C0>ddGZWnr#vI&7&N($?b$uj^{R_Edq3Qt-<;Vz`hODfttEgGMlv^NvPNa)vW
zXo3oBs4_<03o{gsJ6e-s2<Bzmr*-2nr)npNl<DvHSw)0({l#d^Q*`982&X}lyjE{p
z&%XeZ+&`5eEVO@)k`rl3k`A<dI2O7qnpBo!qZOvvBk_!M&z_Nr-CHFPeC3N)S%IA3
z@#sd|Lk^5+Kq!yN>Sc)C*FDM#Y5>OiEC3=)>~eF+JuHT479c`#Or_qJKklc$-cKo{
z*}hl_K%&(!78bPI+Hx2Y9vbmDV#X>{&vdXfR+0Yc)gnpFW873PsBHjCYNHt#i@0_w
zgkuaG=fIeWyCEDdYap!-=l@&+G87RgVANf70vXVa!!8soq~YtnAWXvh{f>>6j{<Q~
zK-Y2*eIE3(mN`eHxSa#z2x3!v4$F3@VX>s;#otMOHxy#Xc#lOL2^3gsmgx1uyJn$8
zTZ47B>TW$+e5V_T%8L2mVP(ZSjDd43gq+ro9+m<@-8apQ!)}b}u$+R0R|Kxks3yUn
zd&q?)zb|vo8MNNA)}f*KT(X`-@1jA!#t1U*QC5Q<D1@Gv(8^MgawaMt3%`hl6j-hR
zH<d93*A+>Z3{M_g1Vr6$;Y$!T@@!f!{#)5xFU8YZm{{pv$$P*u^UqM}lsZnO@+xtI
zEv5AFphNPx0?;i1=VYoV0&wL<3OqGTjBHhoWpAqcEow0Utx?`bT?Gu8qMR;ZX2-?#
z>Yln*IN;>Rr)8Y(s7b#+lTLi(-micMMk_p1VB@^DZfJe50Cbh9$V3^r2(Nq~jV$R_
zXXxIS-~8m)|IZEaVzJky#e=k5{5fW1Y5;5@>xpz@S{U5}%ta3kScg!8L?$Ydu%Va$
zsiAi@=?att9(ZF*3i%HyJ2?V$(!s3MNMJ+5G-T0Kldvf*LZNF=&-Dxhd3~*b*sKm-
zT~2|ez%Lm%c2R|MSqaIa70epm5j~M8fEf?bKy@EcU^C7$_JC(#P9;JlnSOWekMOiB
zv4RK`#n5VK76m~;hI+nn4!BLAcc^I37*E-@a9xZx0h5CD@&g;H`zRH=n&jmKF>lyc
zV*4^x<{uKFTUkJcpp!kTYUyup6V^JK4Yby=^<|inv)E~?waLnmFg6H~aUW^rCE{&W
ze__^ODJ(#cldA?r5mD@1_(0F30ynlBF}Wb(<CTmM=eQaGCgjWP>Z~Fc03qZb7>Qy<
z#LgBC4rjk5zpPksKk=IWKIY&|3AL$^oS^dZitjXz#Ki>rQqc$Ca!a#2LRuG|0sj*-
zA>}2z%vj6R!VS0I_Q+BbUsIvT^(MC!rbk=(yca>R6A+3Y#J46eMCITrj@4Q!V^is+
zay9K+v>};Bi<QH@<c1gcO>0+YDxkc$s$VMx$)K<E`dui663~^V_Z4r^LI{|YLE!8!
zuOOr5waS+rR<s1Y9wvOP#u|ip>&yd9wA<L@Wm7IYTS8b27~g1m-v@>C3@I7z;@DVg
z$n;D#txCWBBGj+JM@wNm+{^l*6qN}qLA387j8->XQ}9t)C)~g0#XLbU+*^7?wxxhe
zh({bJ#!%p<TV<)}D}cYUZj@T<3mU2OJIBN}2f8zFs<jkaTBE~&HaX~?K3CHm0itVf
zXj_M!6C72Btet3)gWk6oq^_X;$+KL$$0CKi5hP6j(Z0a+>~xE1DxJoiDt*+9LsG0<
zYZ2^k=x}U1Gm6IK9F2|N)!>l>c(kW1gV&&XchJ)9q*uCidV^!qZD-l6u#`dkiye_l
zUW0yn2|UG!#4J<}b9lNiL(nM{i9CHNqp03rbM)(Z<d(bu7FqzaSD{$QyK*H%@*&-}
zrSKXo1`cuFcN|uMBB6PhY6&6;>u|QHkd-P|o^cQt(@wo*yh`JI8|g8-D{L*Q@|oqf
zQ6MXTojYOGAtAuw;^q>MiZ7dl@d5_PL-EBf4Yj!xr(uCgzICT1Iyn2J8PMcN_1L>c
zAWH=<_qfg?WwV7P<~*~n^ZH#V1)0V1^Q6y8X{4~lubZ~6G%PTEaK-@R2~0;NIL^J!
zjEvP}tKEd!0n5kRSC^JP*UD6H0NQl)(b1bj6RNUeGG9S7hQ@j`2E|%v3kj^&WzGjx
zhbw6$18A+PoiHhdopV~3A>M11uFT@~rO-(PeanB|JO;J2)&+BrtqZ^uHg!>*Ol{*1
z$`2i8yOe^Zxf}`(fUEsM3@8~R29bBvzBwv@9fur3)x2TijK6mk93dH5inU3p_lAAn
zapTv%Z|U~AK^2=8dF9OFNe|78B1mf#v?zfz7kA~E!TDJ4aF65;cv?!b44q0G28_m$
zt4krF0>YxnAf?6S!2L)oQ5NB=7}=HjN;sx#P4tv*ws>bEO)`&BNT|q7)_B$$tm&kA
zXD{V|t3VMQuY9MB9AAKsLzasHY^Eq-Cb!y<(l@(4FX_%zjAtvYSoZRYmefnFhw7fX
z`l*Sw>H73i2xVz5DXWeIjyZ5k+h7vrQ@gVOUrYMXxZpM6L-Whkv@R1(OlQ-Uz_5{K
zEi2Ux;LP`lZLry83m234M#`#UBv(3-9Z@u7uVN8D`SrVUAEnl5$hrb|M79byZ0*@2
z)CxFGcbtB0p-{*#{!xQO#60!yFZubzWUaEDV1fL7B>*o(Ei2%9_aaq?%xh=PcS1ba
z79$l?DKO0p!_el8!WT1EY0nh;y|);_$QIZ^9QwI3TI5ewjAb>&D@Y;B1+XO2vfd`M
z6ULmJeLWHZryw#ml!JW2Rpkv?JFYxz`I0@-=@>k-*#R~d^H<8%5N8u1QB?37baQOc
zk+SuMMnT&5j)&U=9&QgYt*Unn3-GXa<j?W!#GTT@C}cXl{toA?>2We+i(N^TT@<^E
zIVO(tOm|nK<*>1{XtStJ?ksjvAtqEG0A$agi<mF01od$eNXo)V{#(E(5g)EyRLdRt
zidI`$e=tMe9M%UE7fK}Q#Iee%O4TJLiac|@R4|q4)n4z<nc!Mik{8pj>0+C6qAfIf
z6(+dwxZFumwCEXdZ`NQPIfFqd+$vo4Zt#=l2pb~?_)(_Ub)PBEmpTRlBsHFFl_e?H
z)ll6LE330mv`ut$lnXV$gTv8g%V2fZ!0;Sau~#t>gH6wNJ2|>H0ci2B6wfLfu9h28
zcI11$#_RptT#z89N3QN_HR!MMQ477x#F-x13+JE<NytS{xxOq(bwM&ndJXX2l9O!t
z<J5ZQfLYd3Vl@TpX($9x>8w7ZBkQD~O9E&y7(+}Qe^c4CexHs%J{H|tnYyTT3L)`2
zaswrMS5Zs>U->CzreCX)lenm~2Q#6R+h2X{V{s99M!F(=j!3&bBVEsZiQx;q5-G~R
zZ-zwJb#}%9N)-$%DnT};lR%I=W#4yfw@4G+I<2up?u|kc6$e=g*Ibr?vnpGtYF1{5
zECpRJm8c3S&!|KPQE*s{b~(rDyF_z8sgyLj^oY3f;X>bH@DxT#4m9`(q1-dpc1f!&
zrU}{?!Uh5rE|yqZ=*p7*wJ^Kzoj{V#GTn%?ru^wSe90&(izdE0Eh^PUSfdrNr15HJ
zFNVt^C@1Ms04HcTtaan$I<f$jt!i?(Gs_+%>@x}ax!!7F$3-!)UZ!GeOMJj&WvB~+
zw$Uzh;;7jWMk_fwH`MtiedzF8Sur^rX%&45Ccv5GmS>rJwa}kg#X5ga+(#=05^=4@
zU}&#E2sc)6ZcAo{yVHEX3g=|2l(j)JP9J6YGYbtnKcM@UU<_g%p*8`~#pHlvrfE7O
zf|{ma=e^loXgMsdtA$GhjRh);^%A25oyfRyr=$er8Ziq7yN@#v_Cz%4N<F|la#x_5
zEw@oV@^d<SpM}6AKgLOjNteBI9*n6n<|>;~DS!cN2~nkSMs`HGn3N7`KnPC-i`^?F
zlXuRFZiIpu*^SdPU01HLzBO#Oo~{8|PE$Gk4Zvt<aBO|U)<S7WV6K)Y;4ylG@gOm@
zVhn3gKa^awj~M64R76i@Bf7;xe;FJV5-+tBUvu@fOJVr3HI6lF+?I2x`(D9-xLsCy
z3U|_GJdbDAmw?RLc0ECl?zBD^=rXgOmWiHSmX%Vk`G?>w3MRFjcdTwQ`t02>H_>e&
zX4orBCBaZBLhXt`sG$V%lwn{IYnZgOr-OyF9qqN1U?Jd-aVC#u_Lwbuw`8GJ(f}<G
z3gtRbZHaSV<&luPlMNDEPnRBG)y{`&Q(rcip=}+T>UccpCp`j93fYP-E?RC~8*968
z!7D%U^}EhKT9J90@<5{Em}A^4PacTO86e#evQ$@PgLF9=2?%Na!OG{Y8Q&3CT@Pl`
zpIz4m0CYn`j|`l1!p1~%t<{(!s@t*&7LbTmzGi?XB@Z(LJp8@VQo+x~o-jLUiXaE(
z{f-@bny+&bQF*1CsX&e4Q}~>?pB(Qwa&@_<`mQ3QbQdgUvj83zqE_(Brv&2}dFURY
zl&xzvwNXj5^shs-Ps+==8nKwPROX15+m5JFIgW@eVX!Nu1V~fh7fAy-BK41b8~t04
zxIMkW?co7iXNBi*9RJsU3+H)WEf&8#W`Sewx&AI86nXfA)`Mf)fb8xGt5ob&(5wte
zVsV`VU2sCE3KX`K?iU@6y3*Aa$a8g7<jzz)je<yS(z05xq|kQszVWPNM_=h$iXqx8
zgawvoP*@u&6zg~vAz42A><87OPfwpQSSVy(U1POW^kYsRTpBK_RgBgN&`MCDpD}pc
z8nf9J*&{GHXYqx5C3ZB$6t1UHin@Tn62w6&R>?D;ocSUvpw31U>U>O$Gc4KjoLASH
zT-1(Qn;VlB%G(hAqDJ1|zqDXachn?Sll$xIyxy};p<Oma%h0mS=%Y^N(Nf70GJ}!W
zH?Cy4^=Tqqx<K-GW_9P-cq#!(J9mdSKm|HQ&ot*c&}ZdB8{+4!w+gnxa>^4aIIkOX
zo8xe!N0QTzjBMg{&O%g*3xHxh5UfY~pjCler0LG+J4B&9JX%zbG1>nl2<U#ExS!!4
zSNtI@3YxQ-@~g^6GCt`f5rcwbF(q!M%jq~rxL^#SFp0oaZJZ4Vg>a;(;gQir7tEMt
zML43Y-`#PX1NZxhag4~eYytoVh{TxEJmv(x^??45{u1WTJ>uaz-@rFN`6eE2JIvzm
z;~04U<{P~H@BbRqj=@<-<StsUbcgg51_mr`K9nANoUtF<65rYS2K6jTz+)}iB)i;K
z^2eKqI`@Ck(x7jRr%E14Wi5c^Ia?%CgsHSZB|GEr0I<RtYLqyMf|)eYErp*Wl0!t>
zam%F_k@(mukRI-jj@Aa?5wLagO7X0TZPQhx_90`A1(;Y0$tbQXIUwu9J?eQH!ooCp
z77Uk~2(xOmnQ@0^OU7O+fh?;}|5?0G&%0U)`5=&MIhHsRCi+DVX7QRcJ=hl))grTE
z@1@vPzf<lYC6Xvj-ja+6hFfJ1HU~!rU#$f9vdG*c9ukQn6MbG2SbnG1?=mIT6SXhD
zO<~&?p|J*+Bpft~5lIl1h$E$<PzdU?wB=qyT^^VN!!f6(DgnJK7_*ql9cY_H3kZu;
zw{7dVZJh#=RcT6Lgd78gvBXELtNZB8!#<Bd@af8+6kbzW1Hpx`I*$_vT{FEIj$=?>
zSk5k3r%^>~-ZOedK}!jj9DrraA~F^6oC7D77sxnFPhYA;L#Ui*D1`GES&6;%${p8m
z6InN)MFI!ia~?5?ISO?%5a~V<V{skFIB_0Fv>9*<cOsTdDOoksLm~W^|DAaGz;MHk
zpZUhe*zp28?uZ<!f!pm7_MiA8uzkb*|M=H{mlNJPh7SzIMqUMY*TBAhPf}u5+`!pK
z7EM|&g|J!8%eJHd&su@iIw=}8_*9G|^59cpFRj;Mj_E@iuvpd3x&vjGp{zrJEAiJX
zL@r1#kd~;Ptf$NyeLmj<rR_?&P?1jR@bEh>0nsGD)HfzxPG)9OL&g>s*|0nVC^(xT
z_6$o}c!HA78l_fp0lX)h@1+%_D)Lp9;OaUO_{@Lv8Ct{Td~y!EPfWC|2N_Da=i{~>
z@jlZ8;ib6pbz@D(Lie;NGMI<5kYYh0w}x3M7#*kkLKExD;cn%@_$ja7MVpI~*m0gh
z)KO%y<zN)cLyt3C+yH4->A~XyVbTWjimZX_Au2`?8zsqMHE+UIK?}6T+-whh@048D
zLMd!BietJN>jygBI8j1BXO+CW1JfqXL!Am6+BCE#ms0dgInI%c8(9m&pa2>c;<F?a
zmjL2s4EAi%KIw_oQvBq10}GoF$b3R^e@}!xX}QdyG{Hdfy)|sA3MW%dKXHy3CT)Tt
z`9f>hO<G{QURpy@Pag<fH7Ms?tQ1cd$T*jVV=@gKr2s*<&Ck6-|1<v%oS*FIbK=_{
ze1vv?3*e5~I$FP>*#u6*zC8l}$REP|3%?Hk_0N&L`K?-kOY|r!O=uW=&ey?bDf?<>
z*R$eS*-7xaeiU&o{il+C7Y35y7>5dV<CH~fs5H(%$?-llS{!3QDa*1c;Iueq?Qu04
zv=k`Ogm0z}T$d#}7C<Y6Y6}=ZrFrtYQmoxHQ(f7Bi;vz-$$P0}N=b!m5HOJfTp(Y9
zG9|5u4lAz5miGyu$_3@ip7`|aZ{&{f+i#_uUMEYWAch1xQZYorg0lOf{dE+^HMW^=
ztZ=StNazbF%KAwwf|9DdB%&RYKA=K2*sScRlqbz#ik+pb^t2kjrcQWY_tEm*>X!<e
zl`nD0wpq}OD*Pt=G7b`#%U&&jRMJxnPE=zH!a^E#nLef(uGzw|k>}E)z%BN^5K-?@
z89dnuH9NaHY9Nd|I@q<82P>fIQ<cAd5dt#rJZ#@6irN-+(JF=jq6k?Dv0%&Ro2b6(
zI)YUI$r%ymw`xK3eoM54?wA_l&JU=#W=v0dDoalYMgcTllSM+B9})ijk^P^}F`|Fa
z7O8tu2xCYX0y5J>_Q_v9F>Szp;Y0Yp?E~O`$2Z=*!Q(bzF9T?MSa4=&wxdml9q(}K
zJH|iu%lP7d_!pBPD&kY<I<vRrLJCoKQGC-De>9m+17Dgoww*9A+LLY$`JOAk2I)D1
zm!Ye#s(dGUC0Y+xU|H6uz{Z>zSdNdB&;w~Ni}bGSl>(cLP{?utX0`|^tO|hifVRv%
z$Ko%y!s3dx<ii8CkIT=xtDeCZrI7G<c~>0STEyyki&Q9i+XA1ugtl4xlL-SCk>gOW
zu@vBBi0J{c`Z?hAQ7Of_aY#?i)rd)r5g%CGQVpz`sI{539FIDkie#}A@D*pV7`MU4
zInN-Ea@lK85#ejS-aGp!&nj$k$Wll@VYBpg8<0#FMJ;4N<43WiQC<kZDt(IdC9~PG
zk}HslRCsJaUB^M+-1H29cBL2@A{dP*YKur`wU<`F2Poq5lcL=K?@gcN8e53#Ea>59
zBy_0|&>}U>tqRr5q6dzTVD=;mA(YkRgcP1<tLNB?%7u=+@lYBOj2!G66Q^f-UN0^Y
zLzcgjt8TUgn#=kF=mH!kPB!NoXQZ>jBQEiBMlgB$M03ufIDzws_4wfZ7mIWHq{}I6
zZljX9wT_Q&5Agfv5znz5(;B*UOmFCZqG13yvHjaWLjQelaQ{1>#`TC8Gy6zKwjx0)
zEKukhV%Yl5LN%aLSX=%a=iidqLIQ{~%0e!nn5t<CAkmtcj&#LHdd)Z^_0bwrgJ(Qi
zEH6S=Rv{*R{-<X@ASsTLt0hfZ3zt^D8ae4w=~RnPr82~~NNd4R1!R>rCEG{$H^B5>
zOT?E;^c7eI#B-P8MD*7ZK!Ly(0Sfh$T7Km95|jn+jOjRzu)0F-RJttOvt@nVr;z;^
zVHq?t?6UY96Ifaxd%4<LZ@`$Ca+mcUnU7PA4S`I(k@eF$S|`YtER({=+Cid#O@Ibw
z3PLhPd#{5F8vdFYsqb$oWNMd!a7py#yR(JSu~z3=9Z+6N3pFVM%gmufwhIb!#3iHE
zM?87!GFMgDR`&_vRh!p%A8Bw2ZG{M<GTDpZNYIY5OpL1f;vxVEaN)ni0?f^(?HoDQ
zX7zV>3{1435Sr)=C~-ftaLEg`%l#w9OebcGO3<(tgc@tDVTu~IWG<u?Ck0`MRsvg(
zcG-Q?yy0S!30ed*{B1Gx9hk$j11>;U))~8UX&3y;L<iDv$iO!oc-VK?dBD&^$agB<
z;cy$#AJ_Isa(DpV{HOjNy!)sBC7k1k;I_KTx@XFC&W;uuOcG@VO;etIpa|vq%b%>Y
zL+P2c+P;cyBw!Z)aX9)v{0D%|G3P*QhKJ3tN9iigzxG2s|K;BVCc*;onuU;3R0uo!
z|A$hUbBWKO6oX<0>XIseR`~k*N^qlxRDj$UUetev=tl`#<N*Q1RFPZl$0D-oK2z@0
z$(X%@?OyzaQVN1iaSzgz?kk`}0csxeTU4u30@9L~g?KyrVTsO_bW(d(JQo&6J=mpD
zDF3hT*28_5?~CeRq#3EAJJZkf4vc{TU}s9JSxXl+PdJ0=AqIlR6!+s{{$91YSQJ8&
zfUzj_v86KC)=z}%WSZkZAU9vOYZVF^8{m#gSd|&hhnfg+Z3gSIG(3aK0z0V`9AqvX
zQMEZRIEFb!<3Nhu!oAf4-D}8&!GO(Dz?Ve?OqXj$Vm{W{Agjo-z9IY&EtO#KN1<=E
zsZe%H?xO@huhk6LnI^h(J)xktP-oLIdJkXgIP*-Dj>#=SH_LQPXUh@=rDMK!<a{q}
z(Ne(<j}DM<j|?QUsCbspaAXluD2s`C4q6Fdkq*?0Fta?GPvC<c{Xg~xFy_RqH~2g;
zt;5?6)ufz+A=9J7Wp{(w#O;rKf)D<_@8bJ^?r-Cm1LhT=RSaoWa9~h9_GAJEY<)v-
zJq4dA*#$vW+3E>%2<|GZsbquRchsC=2Rg9*>0g5X!@nO5Cm#C5({AWKBBQNuXn*SG
z@x`AV_|aef9$vD>BB+WDZ74{MV_#Rn6}Qou=4eu1+_t#Z3X!`ZS8%eShbwkS@$kAK
zx5{FWw^K^0w^)<AuCdNG;QQbO0$lCUXDYPm76oiGvhi}JfaHc!p_QFP{!VLy&mL*G
zjY$wE4t8h0x?qG4EU!N&(e-q?Hp;V+@hB`x(;U|0LV;svR??^9b-xI?66@E!0Tuf^
zm3|DI8#^!~k~7AX+qoskoB#vx6oty&Po?wIU+?EWdhLVby|RF%RnHGu46zrLSDlE=
z66-*G26qY(U~TU1D4!t^e+%DgyX0mI!Hf74!64i!@iPOTa9|D?fYX5HRq-1IXx!X%
zwJ=cJEd!N87%F(nEs<Wxj00O*9K@d*gcgu)qsm$Wi~-ETp0yWsCz|fe42eo#%WE8S
zqG901ah>7&Y}ucs08S~Tu#}J*TEV@{^<Gt<mrq;*Fb0Uv>+kvg99}YAUCv|wdZjNK
zf*H&v&Ts5+mW#SIc<XR5OAJQK%wTO}z^!4T0Z)bx{`k-0?Z5hW@#7!A!!gK|9^FsZ
zRtBc?UQBZUS&f?l$N;sC9>#RTmZhw5CSD3vG!yhO(HaW@f9YH3zw%vd|It5y?g#b_
z=rGtgQ$%1K7>|y}|H{7`-~6Zlb=?2rui}ODW21`o#+;ZKkxfT*jHxyq$TOc5jJ{Z8
zSK~m4a`E&_hB09vT`sH)A9KdMpB3eE4<dbgjc-t}GsrE4km>0R5-Wgv(wdA^=T`0{
z3b@j7uPvFS;E=A+wX^OV1Lrt!KdS##`RPgTi{)QKzmW4TRHvh48lQmb7i2lQdpbxa
z6{1GZW<Z~^s`bt^>_wSO#@;xrOW9=SIWf8e7D{cLV;Kgt9H?X;67%qOtGfT4U+*7i
zHf{ocme%a8GF4JIli1c)(rY?u8aJ;Ln0_bcjt)nEb!u)?af=r!(Xf**J-!eQ0CfXJ
zi;SrW$Khw9MR&)vNjDCdHkk_fOD1;>u_!HHimGLGBqUTl(ii*UehS}V4eAK<S#FEv
z03>tf#FzuqqqmG|E_!FLnHB|v`?e9-MkV>Sbv*1h+_oJz4wOG<_1GCpid!8z5!kBH
zL*hZz(?V%_^2#z}l{u{tAD$T@jlND5Iq9A;j{)#8uy;;+^9HmHj#h#QKvs2H!(a>9
zt#$N2^^5rSamW3C{x9LvL;at;e7CAk@L1Q#G+I@eBAF`~HBg`O@WX?<wr}R15H>q&
zG~i(vZp?1F{S&`}{@q73JMoB<iaC^CWarJ8>j1#*#KWKd{rKoF{|3JJ;$4EVaGOr{
zfg*2MS;tNRrol-WTM&J&Cz~#?1vc}2iZeybXtI98aWF$Ic{k6aQbDDNCfA#a!$=oG
ze-gra|I`p4RF>9)Le{i(!OsGfMW31hE_ak!WNt;;^~u7^ba<yy0CeUR+fYz=cQDDI
zXN`qffW%P3vbB^|C)9;1dNaG6LkMm1mVn8m3-*{ej~N~&I7Xj1`$XG%Dyvk!z$Ggz
zx~-Se6iEI$uit$|^DF;!R{>a4;7P<8kDmVa1@Q-3Pw|OW6#4Q$Ux|Znjfdr9CJ`69
zwI+tZ@*No?FblMfX7C1#ED?)9vupT36LX3AQH0_H)5Cn1551IExsJ>hbS*2ytHlGR
z0-_Z30K7%Zg7tpN&KV=}*asYEZ|DYi?{H4Zl5#PN&aC6C!Ws$Ly+<$=1`f1GqMYc0
z8g2>&Y3Fpb9%!6T_ocI%4hGlhd$$ZT)_b`Tq7xMjUQ%AQPh;S=?*y?891a?EW8sUP
zGuGa>04kAJ(9mqd{tx{D{LIV1{h$3;@ceRzqpVq-MW%4(#Js6UU%FQ&BJ0G?1KuZa
z0Gn^<z54je=`yh(^T8^Y)`5M){-6BA*uM3E%?3Ko*efCrcVkCc3wI;}VfLT9;mv>G
z=kfH<eU2Z?g)xK4!v(S+Wt63wmid$to{?FHLesLU>yyiYQi)Gi;GWZCb}&)DFK<o7
zK%VzjilO;}*eU$V8d(&AFCATHSe2XA8C_f}LM^1Vs`U1x<1wP%CSq8@i(|4=?m4pv
zY<&r=ytZ)a8B?4k^3MO*PDE};(Rw6n#Hi5X8>{<xWEA?Gm?OArGBx}>9pmKrH7eH;
zfOT?KhK#z5DHvTxjmib(eObf1Kk4<m?2=kAK`xr;bKX)2LI;mBsg#Fa7pm{6Ko;a2
zW*e2qO1xf5x7M_L=lbN|b+9^8$^oD?k#I0vQX~`W7ULJEqs3rj-8mG=q@d&tV=nWk
zg954=RM|R2_RSkG2ZdJzb92bON6>a6Y9YKiHWHW!B^1vLL1Gi<aiAHqRIFwCsfb#m
zL4A6}NKh4&a^FZGTb>oSS;A#9<<eLU+@qeOS-H66GI9^g<CIxTv?RtBI@M8U)|xfO
zNK~bb3=BI{{tTAIgwxCN^Eq+e?`VBP>xOd}&>Qx?<LQrl2jBkv-@>Q=#-}*OthG>x
z48-Wf^E`1LX9#k&RUwH^_wYyiz&Sgn13MfyIGQ&&Ee5zUfd<q+%HU{!<Y%z|$zQ<Z
zfBX+%?-QFv;QzFZ7S06bgdwio=-%AT;QN90VZ*~lA?6`8r-5nLYb}LX>S~b-RG2a7
zO1>=qiQr}`@CAhAl*Du3IOCb7OW2eUmcWlp7^aQWl_+I9MYq1tgBg$tVdl8ca6>xU
zG^V3dm`=QUuAsYf%-owng>;O;<cV;zRfb)`&{Y;CK#<!g<0M#=5NU_cA0U!5wD%4B
z?S`ElQuSGzhSN^gYfx|pVIgCYBY@6CSO~{G=y43Lby^1b^*P9^;GwXiM_`&e<?{8(
zuF)M|=k;DnL3@KFhA<g3Ki3xgXk|?Zea?j#(rjH}<SX5sjtjYCzT^{o?Qvh$X;KM*
z+7LLid#cSg%f?xq84F@R9rb5##sWB6fLw}vobK}fjcOX^*XP%oGAo1uvJSiSO3`$x
zL24Y<F(?e>M~<Q6Gn?|A!(oSGv`Cd3V}rG5@OF&ovoy{DRp`%cBIYSKWDwu+DmV&g
z8rNKez4k8RA&rQ<MH3GWmYHCR>WHkMt*w?qp@3A$t<ZZ`7z3D}-*Nswe;xJ@|1z0<
z0IfsmjEQ$?#YK;uKu5;e!_y4=AN&ZP{1^T>zV+YzA%62We!TdZtLfW$M8BtToS0*>
zrG=~<x<!28I2_GJ3JHUuw-6ADwGE}{vUN!Nd%uf^|KhLU;b%X_-X`{LnBFiP@Y%BR
zytHtK+X=Uh0k8;a;NiBVdo!)N>DiD_f>>L8trb{68Oqwr`=%=4+t<R)bm`1gUU{CL
zMne4{9-ofMU~?(SCGIe1=JS>{Pa!{MKHmTfAq`Y$b#qNQlrUFN+QF<H2Cid|4Ql5y
zhvJHoM_1vIN(*J3WUBcro@m?&Te`wV%~M#}-npL1UQDE0xlO4}d0*~{13sMbH-TvJ
zW@OS$^V}ny!76tzLNQ~3K*QHF72exYD57LA#UrC*1INbEEL$@+RAIRmMqN|utQ4Hr
zeaa%y6BY?)+UtTkK|$FOE}xu)WEz44z?GM5msT9{SvGKs;#_{tD<EvjSXq3iiN+a3
zrJZSs6@YY8%s@19899JqR5nz8#uT;qTcDc;WgA2`0_G%Pgl~Je<3wyiYbUd2EL~dX
zP+ucZA$IKDt7SGb!sKqynqMa?gb<!U>x+_z;OAHrWuQDTZ#%NNf{0DTQ^9dsQ)z`r
zn7upTfBi@B|A<hL!NDFnXhr+vtubc~zya9i*g6h3*e`#AU;J<Whk*aizknb8_UGYR
z;MmPMCSJ}vp3eg>_X9pA_TEuR%0ZKJI*xGwU|mscMofnrfVF7ZagGz`I6@)(%6EZ3
z^GC6NwBhk)*tg2FicFihZg0lMfe{ND*wB1pTF0;np960mc04r0I}hOoi;lTbx!<Cs
z(o^^Iz-Xm9GwVE<U`<E?RB@hM2NkEq?;3m{Td9X|otc234Wiu;|DVeK34SP^a@a<`
zu(VgiA+)@UMj|8=f##X{9->Q0Z)SCZZ5-qg3!;d0-Qn;?UO1J$8pPvPU??s2bYqFn
zOCgB6D+$D+^-15++lH965pLPmCc4ErMsmjnw}bL;rhCqfd(%`<sP&ZmQ}V~0hSPv^
zuwycx+1MMy9gDmI&|)ea5c9{s&g;D`g^ODC8rt1d1J46DtMdhg;ySg7Z!E<?^E(Bp
z2zGgEX82f&gcJ;-5e4opt|3EAXyO~w#YG~UUM5|$n3*uxh18Y|gVu#)sl{0`KS%ld
z<szxJ0m}nThudILo&5Z)lx_7OwtU8YU^>g1NF-|=h)uaWkpAP-CY8d(afUe#52kze
zXEb;0-NI+<j^j9Rzu$4c-`OP;sjIs!sdx*bOkn7BR&pc2qS#I8uc4cxd7SgCm6cVK
z$Sk+%BB)G$@Fass85RN^X9SKHVb!+|XU`|aMz&x2Hh$lK{!ilkpZqyIpC?W<4xo3u
z-0ygKIdI(1tcYKk(iv<$IpF8)18tm%zyP45$0*iw1U>w_zJvac{}H^oZTQBU9lL;p
zbqutOp_(<Er)G6CX+p(u##GRsdjo%r#)Pug)>aVVvXPKB`Q?xS)TOpSh+M5S!eU>L
z7}jU^v=pi>329*2qnJO^(&Z=ckOg{~i=P~uYtuCZNL0sivB-hEzpMjwn$6Fy0MZ?>
zvdkxw8_uC3%c}5q@@_LKA>Km;fw?~5OV3g?T!XZ-4L3jqaTanXXp*cNvoa3HIWg%f
zm2#4$s#my7$=kE)6ynlSbTz0eydM)+rT^1kzl)UAd_ur2(BK;6fU-$t4licm5>t4+
zNK1lTZs=d}7+gbpwb9un5(X^3HC?Y6fQ_+%Y^}gPNmiRHWzri*HCoKOh1FCh0BCdK
zWbsxpQtiuh$0llUA_EKL6IKDuX>-mqCIPq*c?sTBnh?LU(zqG69{r{SktT_4WeBx!
z=0b2c$MXijZ)k|_rRRBu07ChiWgSHggyJ5^n94{Lm5-VG7)-)B*>uI77*yOr7Y&ul
zTt0j$u=u*42cGtg8BR_;AILKO_~!v{F?hwNVUFA`xHZ6|=;!H={|tWiSAGG%@t6J%
z?&pB_K>sf<cf35`F^&=bVE6RbJH*5k2w;52PSRH}XP=QSGsb}b%Fm$vV}B1GKIr)N
zn<qT<Sn;rq**eBqZUsw)M;c~zaJK(&z%gMS!`$Wa$-Pd1Rj6C!vI3vWK`ozHftXHl
zRrXVUY<$%vN+@G^Dr;E*vJ_Iuh^Q7MY_<elr(zG{k_C7qu=G^^5qtAg7?#f&&}wZ@
z1NY2C(iAz*y0JCeh|is_;Mk+E?mYwWSsNifK>?jIS9mG{PM*h_LF}Q-q9w?^Zt;Sv
zGv?UYh_4D1SD(?cre;8IXy>@_=1+UQmr{t)i8>$#E3qmH5~A6adc3~sA&YqOO&7&*
zeaP3ZL~(8vjGhP|>trtxDo^2Wvrqu9&MbeZ2u0LD%o(ndjRh%Ni&Awi0M<Cyz{d4i
zzr5Na&dCZ$ccLj@vQWYWX8_T?7%?ebn+!wEA=uP_2}lKSVkUB{IF6(DP^jZLaPUx$
zW8nGuh%T)>gXr&G&BBZ*_M8I>3=Ve;AWQYF89X{IpmoD-?-a5V2Y|DOwNNn%){ZD!
ztX0R+;S+uzc=zKk@bQ~B>%k-0&?33vydN<h6nn-~V(_~IXlQ+h%>(Uc9`Vh8>feX+
zzxOZVxBk}maDH}R4#)jCBOiRQjX6OK5NES?t2GgihtfTH<YNw)Ps~5|E#S}m5<d9w
zhHpMSpxX%7TH7(PMOx>WQKUw+n(a=;fEpA^Ky(=VZ+`~={3Qe%6$w&+u`WUMEvq*{
zQ2{e46kCMIRnYUj7i(dX<cd2|F-cfssaKIqkZ^VR{^wHZd%eDX9+QpjYAL1A_$=ZF
zrK~QZ7pu#}EcQfhD|fbqKPF=)${>{T4Ze}mQPai?-O1g<=TcFi&Or)Bzgs_gv8rY%
zx0x>IdS5=&3N#btAU`Kdp$Pyi@<#5hwU9XHMgNr7dntw9g4umqnu)Iiat_b*M-f~V
zo?TP`g#6-F4uVyuymG?CXZ6R4K-!VAk^2@KJ*uDtfvqa1d|3mK88AuqtwY6b0I+d5
zcnt0+MjMv-N?I7oWpzzM)la%dc)qoc_)pEg3}B^fIvL8$r$R8+gS>7*3=knqgEfbB
zZDNEM_pRe$@7Q~2%`qqL=N->42VTwr^;ya8mr50-g*4`fZmcBA5S@YRXQtLhZ>HPc
zv9}Fl#_$`Z0am6Nq9t`sv@@O*t*U?Rd$|9~cQAhT+u{C`3ANCWgUaZf(R-;6v~3HD
zbnXQag0%+w@P=>wxBtWNKmV)v(f{S&#Ph%Kx59LnTfoCD6*UPV_)aQ7@VOc)dn9^G
zO|)P97~9|T3Hm?!hw=2(@yWvjx-xb821gH}H=;6^$%h$`4rLIN<eGIJxn(<<n0_`W
zqzg=?cy}z_NE;ecb}x7$-Md$Q5|6FfHYH5I8lDw;4`HLh_v`Q1F{0BQ-LKFK9{d||
zIZrF0zz_q(3+TgZu#|}%6oM@j>40NXapE9QI-_@yxG-pTK!a|=>KiO_V}V-`Y`j0?
zhwR^F%k#~3ul0lH8KgDji_0hXe^Ju>s4~HPkr}C!`OM*NoRKm=C)A71P^sMmvU0A9
z>L<P4+fwK~a`@W>+$%_14_R&wKuC3sqFb5FX(@cgYA*=>^~<qtVswbC>o`E6Xo*g`
zT3{^KH^|zNOQtcX3-}xnuqBeq=WJGiwB>;F#=dCvbcyi)<S+xGigCR9u+<7%H~3%~
zy@Y@jYH!`xp_K})%lAJ@Np$svpJ<4_IuExU4_ov`GB}=(Bl{QCaQI-xwH1VXloLc+
zD=H*v&pI+9_>I{Z>eWJ*xHdoIG>t=GvQnJcL<7LKZ5VSzpQhIH>`i+`q1C&=FmWDt
z*c{llEsA`I#xW;`IeOc%_l8gY{$IfOLm%VqAO0Bq|MsiEpZnV>T_b0HK!Lk+g!vr<
znL75q(WOST_U%Wsf8vh;zwj|Wylwb!1KPYZeYWQ{V05QVx@BevdIK>1L~9$YZE>bB
zRGv==jYN72x|MrE0%&DMp-lbO3D&OUS}0X&7I~F5pt`b-LGcR-L^vy^+0o!NA1Ge8
zW+b{4oO@at8NaaQ_o#ZARZ&TC<*J%<f?|Vl2o+{z#LYRGc%jCeUa6fz@r$J(S7D`f
zI<Il!xEs!g6HYfyxGfjg7H8L&vQPkfT!PooJweNb$|GCzkWW%E>zvmE<aI3RTz`07
zALO`LK0o30UP@ux4V<nQE}Cg+wk~>X2`7ZP6bHNd7WuFjqQ5?;rSx2WseiENinRzj
zs?N*@OX&w-oW%@kds&LJ87zt!7gQCvt+Qc;vOp4@@R>3=M}}28;k3n+OEHNIwoecA
z0<%bU&43_-gl+&^a)*FbplVb$Ndh9_3JT%oQ5@x5lL4~<lYVS>S`!BJt)t%>nud~3
zNB2fyE0p0xUCYY=XKuZz*;Xr|)kT2QfCe;Yd#6+=t&Y&8TdM+I!bkV<4!|Q%AOD-b
zikCm}`|$A1597glkG(R$&J)uDV)!^P&m)R-dWSte0@z^_=os+x1%@6(V>=$2<69p+
z;<x|A&*Jmn{~>OF`VV6KkN#o=y>GXODg5?J(9I0D&doX=y5ZJi{pXt-wtxJOqW$4-
z;`Y%4KDc$f*#L|q=)G-#S)}f^mIxdzO7M>RL4YLRdMKmT8)h@WPx#;X5sv@!-=QVZ
zBP$6k<ePV(ZRsCJBJ>~@bE5AaLeALND{IH){(w*}lO)mxBSki=YJSqbnJ>)?U75jI
z?^1=Hk;-RaEt4sCOl!74-RHnOITT6NR<J7CqKxrG(Ux9_>5gd=9ZL5M<*+rnu^3<;
z$HZ|T`P@>1>C4YlTw?tOw2mBVbqt5U4D@Zw)+z1|8xxyk-JHOnyLECm2>@JkL>9rW
znH-9zxFge1X{4llR7@}AE0wBySlPVfWKX2t%Gyws{`kF73iPS_#5}?L^cvfjQ;?R|
zgaV&54JmYmM3(JAwx3Yn9c_krK6Ev;1Jnaw0Y{bldy96>?v=fiL8=C{Y3yNSDWHTN
zE{aU1X}$!%R&?Q*V;}>$Li0<>dGeqo86xMsPRk;=5SY%uw}m;#^tma9XEuQ>>1cES
zqD{}jF}2`so2-TCvts&cZ1+us&(euJ3!%u|cOi2tVXvA{hhxKr^PFiVSSv+P?u6L4
zY=pPeDCA!0ym&9%2UB@ZY;8lgh5?Hvd##~KATf3EJPs%Z)3+^5em`MjQ1=eZ88tNz
z`;L!4c!MAO_zU<qZrJ}@{{S9tJNA8peg7H$|M)f7oM^0FxZQ4e*f-p^Jrw;P{si-n
ze}cz-$A?dkc-Ra!?zDI|3~UidGy#G@y!Us{ku^3>jPrms!?xc_s9F=D!|yM!<AAjd
z_v1+Lk^D1}F^>VdTIjA5Qk4a0J^_!SNV|ZvrOFxCGum3ijYWk+u2_$Zy*#M6Y&BfU
zL-Z7aVbQj{&zx*AlQ9LP)jvHFJjxO!Mb71#%+;4=(GaQDP1bP>fjh=PgK-4xk)3-l
zGGR^_YF<Uo{GbBTxX<a1<2-R39q0Xo?SU?%;BID|BD$Q;MEt>j?0Y6T<zy<NJ*BI|
zRnx_NlhD&Hyprp(4qQrr8);Zv)K+}@bzbl1KH7R{AdjN$jP~;-WU>B+B6Ruki<K))
zI@mb?0RR9=L_t*IhSe3;*YETot;Mh!`MUj7YMz5FJCRjqDZCI_67Dc>u#RkoHhgk$
zhNWT%<c3gIOA3KnMh{iNX5O<<um-X$R+QD3;s%VB<p<pZK`LIOAhaMux#0jXP`SBm
z3@QkVyyVG%x?B;B73XOD;`;{5Eqgf+-oL1j`R%Qcqvv}g{-N~Km={qsD#KS$fO0Ld
z5ni_A7GefrTe=!1?sEcjpyzfJoutB<l%>-#UhbHWM;I03K${cp9nR@#);pR5PY(~c
zb%T$14yE7spZfs#Pyce%5x8UDHazTmwB6o9IZdDF^T0=s8{X^yd`8QVIpIB0RL>cE
zv+Wz*VI4L%9ACVJk2`#huuA%d-gj79r~`i9F`u7t9tXCEj{9*eJQU|2?Vu;lR)vTP
z;D(QWz+L^YI?N`3eo+YG)byd#irZR@sh#_deeXDqBluyNgWv>dHE3sv?a37aR3)$e
znCrwclt~4JUUH%YK>{cBBGTv8kIJY}6?k=boaYI+;rV{x`Q?az_XZWQrP~K_cE>qz
z97hy<N_L~3M?!myaEA(*%(TctVaZ)0782^Om_kJo1Qt$aGUj{?z1J0IEvqJZb`hED
zl{NRHaTt(3^L1YDmy+6A^wjWZSk~wlS5TuALLto*3d<A7twKGCbQLYa$zPqwMb)%$
z4Q)FB%Rn^0d!A3vtQE{a2})A~&Gy7F+|(!%Yw$K{q9PC3tg;g#Yr%mz(OVRXOm_@+
zjT~$eR*MwsMf%f2(Zus*yXM-Cv>@P-&0}8*%3BeF>|Dk`q*&ArB784OKh=W^n#WZK
zuYOM8K7rGLnt9V!IxB5Xh0q$Z@GOauXRJZ-)sYsMK57cw3}x3_%h}JAW8so;zxJ%+
z&F}x?e+Aor;~&5|2W~#m+lIl+BG(XHW-mDx9Tmz4w!X*m`;K;dg7q7`Z=oEvj!!<h
z;g8)O5rN1vko<8Fm9wE63l29kbaVI^;ZAeM2Tu=p+8g|MCYU%vKsaFghP_Acru)k~
z+|O9+Jnr!04$K*CP<pR66$B)}fWLdj@n`?#h&#+^M5kU{##Zs_hvS+;4~q;Wue8NE
z1#$-5Lr{e4XZ4#4vsA^cZ8KbUw+#>bhG)7z6}dDl1)w~8V=I;L;)IfQ^|Ug(SeIed
z%uJsi=OFFa6(35z$i+73R$Z+1aPMB_gpZj;Uhkd{Jl~H<cWs`nX~vSKdN2#^I8@6Z
z;k%@(8ldXtIaB~DdBfk&m@logWAMG1i8V_gNp89;K}OdB(mW>*OP(fRTmi3?FRGuI
z0XYN4fv>F<-p^8?SsTN&gP@|lg?fu4(H$TZK9jjp)TLpt_{VGMxcvC)IgF=o-Lmh5
zhkPhmI6YHS=cvqxxE?J5gxklO(Facxx6!A|gN)^R9+2WCYXdHsYf^Y2qblnu;exkB
zHlQZ=%(t%DzG(H(jv#IkuX`xK-VB@CcW9d!1b;2nRbQ>x&Y01J>li2Q$H0pwbwP!9
ze3tw<EEY4%SWOcLiZPq#?<KHQG!x8FG_*BK_Yo{;Bs$Tmc*ifG!0#sp9Nh=Td7$kZ
z41s2SFaW*ZvU$V#GI87wY%ed-9-?)$eTQui0E4+`8{8s8>fz&Fh@G^e7T^FfBW0c^
z#yFYk8;}3y@rLdPrUB>u1?T-acCkBPJKQG5apLn&KgZzp%_p3lI-70LR-<*Ofn2cJ
z#Q2MU7wvFZ-|+7F$N?*03~WPKZMWXAZ;i!>M4lGWY-KF@9B|HujOKetw@v~&1NYSc
zG&pNveD5(rwljt}Ikif|V9c{!NvHcJ0`Uwm6L3WgT2?@;BW9WpHkF(yFw?zgQA;!+
zxJ}+eWWt5T)L=GKj!0+%#_YNgX~%LMF$-)Q@iqGlz{KW%Qlup-WDPF@Y}T{P8Jz~^
zHnTQkwvt`$5x<+G8egt)1u?o(=$+dty^w-m&n4N?#HF6kD{IPvaF>hXYro#pQs{uq
z3_8|g>YyQ)hp#?Of3*oILgTyDLz{d05}_+EE{xV9)23}5ecJ$==#7O@gV`|>vIfrD
zy=v3=neL_m40C!_0_|+xDaSom8-kAr*-@2}l=qi?!g`GLtlca|JaRpgu*2CP&)p-9
zbV@}!nE;7q!o?B{<<2y~{13UGE9x|37KSYuGamlY`x(v72Hj=FSc|Do#ZcyXwF1Y2
zY{9k38VEe8z2+`hrMsrf!9)XPA=vW#aK}6c=4LpLJ9@jp*#l{74fo@Yc^tqvBVg&#
zN9yI>GoGFvu|GYa?>EfW*`F&;$las1*xA7{OhrDMqt8J-Yw+HHediO={pOB2PWX95
zoW>`P=jU)aJphKVNY0l#-hTEu&ilY|4m>>Wc)0D^EO876`h+n7pcycaf$``54(4&f
z`VF6d@fNKyP!0esriG(5Jnjv*Ei&f@Qz=g%6hRC(D2R(_lVKKus;Tf1vJ{<X?_~j?
zdQ$-gXkN4`mKjWLX&rlH&$)3Ln_qnDpiuX;6e7)2^BQ{V5h&iG(O`6E9Y90&U~+e6
z%n8i(1f(n9&cXH;3IgkCsTfX<jkN|G=b7t>YImzZwk(2FXfu`U9229>EO-m6m*9q0
z+8Bdt@Js#BmtSEGK@vc1AqcQsyqG*Lo)^~$lr|=PkcY5zhxjS4_fiVSpyOcx7|u)-
ziogt5dPL$88*-e?`Bg?l9n0&jfVA=2q8ca@2ndAJ4dKGrA_$}mgV_V&%n?kU^Amh5
zrf69v68sTl3WD|-CdXp%<olWr*s#TY<Y8v=kj<-QgVce{-rLsk&>J48)Zj4$$A`d{
zkVuu+YsgDfC6CPDGLOk@^5Molg|Kpf!7~H}(fO|+?m#sKd$|w1yquYnoq<3D1La7!
zmo&~ggj~Y>lrKLmpi&C(kxF6sWal3W-yQ@nw5n`IAs!_T%-{GN$N%=PV*k(lQB2%%
zfA=Z2+Z#;u2$VO&^bs?oj{~h4_ALass}Rwdbz>c#`P_6{!7y6~`UA`i({7mKgl&#F
zPqgul;VIWc$8(ZC2G}X}_6FA9VIJ?I0-3I`cW<8q36H^6<^yx@7~_m7YjcEex^<X!
z_&DJI{NDur=8u5ZFwyaRM@*nYSVmzmdsMV<TL@5}7-t~UV+2n{vD65`J_koaB5718
zo-hE-oLoi=)GR)4VU3()hMUNvtwmU!<jFYJ%S1xC2rItkNn;^cXSx@U8GPHDp>GY_
zMp~PWaWa$0qZI}`_A?Zm+((u&qcIIy{iT%33a|R3Z~-y>cmT)qU~|N9|AERtO33LV
zb=NEl(yTLG7V+N5$cz}+JWsl*Mp>WfSx_K&AuHGZPy?hOS_D#4iaoGu2KR}r_ejSk
z?JX_5iiv;9>-~BkDJo7M=SH-)3coczpbf2(abKz$FIVQyy!s3GM&=U0?2O)GW^kwf
ze0mnDX&16>#&it75~SObT_ZtjOVIb^XwRY^z#T(qcB*4(`S{E)#|D91J_fo)<*~A*
z+<@*4CuU^{@IG~|%3w&;0j0DC5zIN^)-WwxW;L*5*;+`hS9R%mOx%wkSewYIXvyTQ
zsipQcgQ}&%u)J^<DpKn=RL9&Yav&X8y1c0H7v-{MALU-o$TPR&4EIUj;c&ct_YT|b
zDPlLh<M0V{N831bs5d-3-LT(o@mzK}o$JQH0NZ2Kry0<1Xnljt6XAz9L?bz6*Li+4
zsnravZ`dE6IPW5|p`wMx3z`}BEou<<+lGDLgJ%1N?cot_oq7HZ_Tv-ye}6v))o(Yv
zoQ|zUuzT|SqvZ;shZW~%M+zj+2ViXwWH@X%`gHo-nF34tZ0e80E_saswn`<9v+B*T
zZ=F2Iv!#%m%N~Tjl)#EFSwuh~U0LeGPaNaKInD+38@8=s+gkP>YZMVt#K$C%1e*@Q
z;Mbjr^JmI~74T$We$H^IF?mEi+GM?l%N9)f2;dd*u_=h22@Wn7mFapT5>n1^k<C%)
zUZm#}33i3*wR#hi5Rj-YNfwz}{wTGz_m0~ZGghXuwOlUd-wUOnr_clkt4d%ob?P;<
z6l6mcp+M%jhf*xFNK|T6o5YZSDwYFou-J(KT-3n}!8T(sEYrT?+BJ&M87yBv4ODr$
z=OLv(UG5zE?E+yGieAmt+yPrHk!av$+QYoKESWI#R0%FC-Q95dn!S@ZiFGMn<A!XM
z<z6YNug)&b*5R4yVW2|D5Fnu5O3Dt>kb4R5*W*bt2EgYJ7f5uk$u6`ra^_(T=bX67
zY)p5K_R4CH7^ri1_@DoEy!^f2#nZ3+Jhs+xoQ~sukMvQCImqX6z*_XR@4e&c=^>iq
z@!ZUMxIoosDoo6954Er&%sQrZ`oae=*~o%x)s(sXGS+!eK=y}6Y_yR5IPva_=TN}U
zFEL%N8P4;>e%sNv8}^4Mpx-dju=O2wpP2u%e;xkUe*ixRY=6RWzF_OX){SwAusruY
z(0dPmK~zjT{o@UYaDqo#?VRYCFz?ZxL3BhbMQ7(@l$&t|5}!WMI&i!7i2uloG-Ddu
zIR{QRx^APuP{n7-7@*kT#tnfJR*xSel)f7+g=SHZ*jiTQR@MBVBJ-$wXw_*q8wIYG
zF4=8OgaJyx$^yA0#*k_9m{A#2?qKKP5F6W=@Gfhj#%IoRL<;K|IM_*9Gacl1R_E1b
zQNzK;ahPLKtSiDZfr99J7VbS-R0tQ}-lepoLR+8yJ;^@u0rvm5@-1^LT?z^TNFnLW
z-(@NJg8spcm$r?^R<KhgF11l6zrPs~4~<?Hq<QyfmpfHh6`OC8(7NZS-7!Mr1UJk6
z%e5fH6+X$f_zEkIB1;MN@x1<;j;BXeuu}XP)=hOyg@Ju{a;3R9$a`7lG>0{%03z~t
z+C$@+44{0tRBYWEdPj{14H<)-X`^Zv9@`K~sG(4^Mkp#sm<l_nJ{eA_jG5KA>`Rgk
z_jA(ip^=K?hX=-A`VGAKrJup$hY$GRspI*KG{XCNpt+-Od#w9mkFTvmesojCqLz_<
z;VCLw(tihh9`N%WY`laU0AOG*R$9lx3Mco73*d(RO~?86Z8X^<;xxAPR@`QsC)(`+
zleF3Q4ef_7nE#8v1pLk4!hE@-wT_n~g1UR}xc7;ue2xn4ZHso?VR>3i4{V-O1Lfb&
z3f%-K3z3HO5QT*0OBz*6br|orH}q2t3`0O4<HY?81vMGNn3M>vKy*^?T9Xio9_M*7
z_dg2P<W1;36dx>HgF%8bVmKYx4A?sxNwOwsdK7Q2LO%-TLUG3ET@=~|Q>eL?pkuk6
zZOM+4dowKOot+(MW-xZs;p4>F44u~c9A^%28gjj_Rswe>CL%UGu-JB^P;eHVtuHBr
zn*af6vgW-v>^-c#F;_?6*MGg2QpjR1`jC}=7R}S*;FW7Cl#%UWr$>Pc6~ZOx;mS|f
zKfRC~kY<yt0a;UxNG<)(U@)|We?L=2N&Hq4IM)f(uaxU7lPd%>g<*--Y_|&rkp#L7
zLeoSRIifH8|1(iMFgyZJY84#qfXnKF8M4$KUV;^2L1<rMSj(VQrNA@S;<5fzP_Qi4
zQLG2U5yR|T1;M2hw$2t5BI07`LDp9aUNnqDK78$)roh!Mx>_R0QZUdD%dD?#3&39d
zv}(pb_y58DANeKVgEz6Z{f6@dj%GMt?tnFHw;lV#jg|K^&%p?9bMja!tjOD6<3N}b
zHV(|=1vXE{I+RxISzWtO4t>Kdtc<PU_V9?8`+>e~IOj_U8lQBFL@M=p9Ps{(+nW#3
zKD*=iXa9Bdzx6|$&!5Lm^*u7=3=v0Yp2~EM?My=zAeu$Yq*)YbHsG`wy@2S7gm_g}
z8OqXgRQh;CWa8C&LmQSYNn}N<FmBCo3+1DfkD@;J5L7_8fi&$dri=x`?gR*HolkEK
zI}<z%j&2iMYuF>w4xdr9IVSRK)Gj4PyheRXZ^09)vx};1XEs&1%P=RpIof8pwW!O2
z^=tP70wr!!bSpN-m>5I{-Rzjr{IEM~f*=d5YB-Gc)vd)fj%hgCjJQy3UvQHd_7?G_
z+t$?h84BcwYbm;~n&hI;=kJC4s0p8Q^Pp@LW|Bx~%B3Rzbrs3Tp0n)Dm7;c8Zxt}r
zhS<Uk*`t4k;!s8dXl7b7*wk8sFq^KEnm__-xo57S5WUtAZY`GJFKe^3B@|n+TyR!7
zgLM>R(#^o$?#@{U%GV@gHwj)ZVH*TUoXTxJN2GEFnu)#6QIeJ#IFL8@z+J^RtiE(1
z`e5tcWL7JlK#O2;6N1+{mtHZ-MzKXW#4~{T=aCv`MxS&8&yo)g(L*UfeGPv0JWhQ6
z&-_Jv{7?M{G0zkI@d>@((EB6iw!zPE^=N>S4>mue74<r<AORMFRZc#Tp$|-EAB{9G
zNGGz0*cv1*5f$RwcEgwhw+G<qIB_2DaGWRJJUyYg!)Nr@-?k05b!=P5ZQpVJtsi0j
z)$d`91Ac!-dw9YDi+T(LtCc-z!HS?uAvB9AV7sAD2P_ndQ*orvpEjq71AW<hD#t0B
z1u|cuVo6r+59ieqV^zdRzU;(dRKzJC2#7M<$B{8dfUz|L_SWFr2KR}*2NJ&ZsI-P5
zQpZQ64$I}!6nvId&<q<QINrA&x%?224dL4wLecCSg><-L+QJi(Luy~+h&booEL!DD
z7I`R~=Co>>VK#%jLJ2U)_>IrmskQdm@Qh1yG!~=*yEWX*@Ytdz0R*#(-B9hK@wp0U
z;%mL$FZ<|--0BXRknoI<WK2O!0}a#`h5*z}@fB~DDNc()a5sRb71Ad`>qBEv1hKJ=
zMLZPhW-)n*=sY@L+tnmX@Weqi?pS+?0Cn#eJxP_1Dk708kOwkdUsitAc}929F&&5e
zieQs7XT5~l8_+d#j>zmer<6+t$TG7E%D$1HidTa(Ga{6eJEBZtMv!?9$2?i9(K{xu
zar#6@gEAFE`0!boc@cFgv7#{seWuJAcVJp5r#T$=d(8Hotd46IbuBkqB_d{1OL7Sq
z84|J};PBJ&?(cq%AAIo^-~8|iZ$=3FL8;XrZm@Y`Gr%Tss!zD@SgsDKF3Vvz8!#rU
z)BYI;aJ+?&gOl+ZBI9vEzD`xY6o5rbS_X1k$K%rnc>C@J-}!w%gEwzJ#E*aY1Nhye
z=(07m+XHS-Z!rGG5ApVY@Gk;y?|=_{@bCr$4aXcfD4YZB1H+7MCc<hsdFT;KFc>>{
zJ$lEaCEr_vZA}ai>sy7>!jDn{`de(Ry0?}KG&2NY(R<8i@`KF@xh{*RIq!g|wRy~-
z=svJp2w>~LwizBb!$V`AB#XKZYjIw;&4`d&0+`bW8Z)Y_Bm_WZaND9g*M3G@m);v5
z_LzIU5oE$0x7N`2j=8C#JO17+g1SqvFXS1|0WFFeE&S`*pwKX3Hqclgo99IUA&>-s
zqdcKJwvZ+%gd)Ay7rx%OXI*D%`BuPzDN80@w^*+ClV9(p6i!a%vfM}#0gwtI2+J?|
z#4_jP_j2G8SGed1ui#ETBR^zxLq(uiQFXTI_Sj*$J0f*7nsH2wa2@eLjqgo&5u#v7
zmIz)EJ|7BD6R>5)U&@%F;)>dqK>MdESDm@ha*L#86>IXw>tXThrgB9J{>fbOHET)5
zK&Bd!7JlXc4<u@WnC#U;rG5IqK}%q_U{39ST1#Gm@)(33_d8FvSvcdY%|cN`OY@04
zea31ss2R3lEmYUcycaCbW<?KjP2GR+7C-)Tzlx`S=nvz~<5TYYnAK)ujuVVhN8y@P
z^FCux-F-whkpmN|;^niQM^>~ZB5*1gU|Cl&*(5?-{kPT|KK{nX7;T3+@bO0<;{HWg
zN`1cpZHrkJ|EIr>emZP3?E4KTmBHbT0VW%uW>@I@giCWKO<FVTM9A?Bj?H2alaVG@
zhBIDAw6<_vJUZXr5H6i=(N;wko)HKHDh@!=gqWAZGsy|r6VB8IYp|_je=wZ;#MTWD
zw+#>57IhIV;#ZxZXz#58(I$_B+_|BkTUxf9;m{ilTL2vx+Sc)~ZF#LsmO&5T190w$
zaVjELy;#YzkS;R~5SCCQ4>tsm8r=W5*DBqR-~w=V8IFc_=aj~v<uxM#MoK5n#R+CD
zGXTL)=uAL<P4xRUUhg-fna?u^1z1xdmet?+Z@Ci`s4AsEL7V<Glrl*dPCY=Pdgn)2
z8CVbK8B|u2xj^+D8^WJz&7i1ix)GK&sj{;%J48*joYxm2PX;E9%#!xAUh6gh9J}dJ
zy~?_R;~ddeV#LYH8w(GHvsoy4I^6=%5eYnQOje4@EQ~5yKr;?B$UV9=TNca39vYy^
zVP!#pt|%MqARCM%G{l~q<YbdOKC(~72y5UN@f@r)cMq4qd2&XySFXAgZtk(w_L&*m
zI4rRhSf&A=pWQM4#*gsnzxNs6_~0Y-=IGYZdbkW46?_)%A%ehj%v7eK_-W~yF{Rg7
z=;N3kQ`OpbO9~ja$X^MzRRNIB-4TItkJ$^`zOnWt3dOdETky^85&hv2{x5$IkALe6
z^oLu_0XPpl8*n&qwm9n(9{=_b7S<R*c8PLZ+LlMxT4Tkzxnr~FW2E^4Nu;HQk;)jL
zYlO)o6n4(|#b)YsN^nJi-{II02(~S7JwYI<u0t-ZW(~b>xIGxg9-yKj#;MvmPq-PT
zMLKL0J#xRnG(iT^6+?2KhXi?}+)>fxws-8=MK#iK8}Ea*q}>`avq!X2N-g$2V||8L
zPx9s>7P?`JmP-j5r38Y%WOc-jO~n4!eHO3?B>0hj(fUP=lX6lW9t>e^eO;~aeo7%$
z$?vbwfEHfhH+^mtUs+`bfLf1;HxapI27vtBiOMD-4`1XkX9^Y}IqTuWQ?4`i9?Kkz
zZOmL~`nID5Mks`~HMVmop_P}Pj<t4Of*;7G15$Vft}%+A&e71jXT|C1NP#E+f9CpI
zbh?yMU}SOG|Jb{!CVdT&&%WF^ncVAJoQZt&p26*8FAvvNpOL+y-uBH~oh`3RPdZdc
z6xP5<^w<yqWA%}d710@<Xc4trTo_+PN0H&8vmzTshRNUl6hHp8AK|;-{1!gkcT72i
z=D?VD^tNQ&*uc0Suv=$~Z%^yi1@YR(RMj9fGXQ-9Y?-CZ&5xlb%(MX5K4PRL1u&*T
z1lXFvM$mNBBQ&&a!~XPyZ2&KS{Ri03J32-dT@I_jHRg;K8URxz3m!NbO%^BRMhXtI
zmI!~dCDW@Q<>l#?C<DxLaBQr3MttH9lyK*p&5`ZvH=h4u<N+x(7sc8PeFwa4@QqI-
zgklB4v>pYZL-tO#^01`ox}6D^21s<zer8}{U9=Yo1U;v$jjMH<9gQuXOUPUsC-9n0
zD!umb+6h3<tT>F4HR=o`Ut-K?OZU*GbW;mjXb7wCrr*nA;J!(r+w}ON;)3Yq_riVT
z4opmD2IyA&0~as%!tlz+g@I|KL@?8%H;Z0YLMMduOrQJZO;j|EQZvIe`aE-fA%D0<
zuKmOWykoG?&AFbo^>k(Qtw+AU2ijF%cxkwas8CD;0RR}N)%q2P(=iw<kR2EV&b>k^
z1i4Z8gtT__#8SY83Md*8BS$S19-p_NMJ5Ubx_+nJ=0vheAJxUNA|RbLg^t!Il;jmW
zR5j|Dkx8{Rz|q)hA>56VmG2pIS~AESmiG}uTwW%IFrYb_H#DC(|H}7q{@u^;#V@_X
z<E_Ek#?LB)#B-gY_$Jyo@$UW^&^J6jeNaV|GbCCv`5YOrck9b}B7(#o?=jIoQxem?
z<h9pxgrC2KCWP&eRiZKn&hLK){EPo4aJ)o@Uf*$APq$|(BA!RKTPThUI^z%^k*iKK
z=Y%W@L9P|#bvy;rD#%4{S>Iz3yJ=2KE^kO$5h|r}kn#~|^W{FuvU6*cr9Ib?6h31v
zio0J5-bA43x>@u7Gq1yt)xi9$v06VBqY_QXX=#<F%Z86Fq@GH_DH5a%q?Du#qMb|v
zTFSvO2fLq2M#g823Q3O^mL2f4H*DSLnQ#I};MO|>(v~0q4!Eq~1^h_?EeC<G^Lo#5
zoFy25Dojcws-lfqX@ph4638X8S!OB_p$!SvO1q5zSr$&k=Pz1XgvRG>KH=V?q#s4)
z1fFP4-?G`#KZ%(Lt;JYHGno9DJbdKg8<Qt&q2VHk<Tn%=P!%#iwYZ~c3Mzjc`gfzN
z28i6>uQ&*2dN3v`U5;y^#+p7^O>7L}ikLka3QF0OWku^q8B7rbUTO~DtbC2nx{pN4
zJ_qylnYl9P?m0)g1=7_;m)&Fld~i5V-~_t}Sm#OWIHg#}kDl?PAOAK!x-~pLy+Lms
z^JHxU96&?s8~VOsPT-4AKf{kd`y3yA@(tYf9c|m7F1q5#kkw2h%sLDd<GV*j(>#}B
zQM0*u0XJ|%?;GIksM;Ft<A7PihM1*bKlmJe+%d-;FUJWycD(a)XJ)5grdy`~oa%fU
zC@C{`#A;ALpFVNe<-e6L9ObfINiCHE>AcNApzx~p4c8H>z|f1Q*9mX(2*}r+3QvL!
z`dz6({aRUTidU_mWhTI>)-Wk<M6>aOnjT~jU5v$s<FV%T;ZdxY>FJA7TIV~LmM%pQ
zzmE5e$t=8d?|NQ6;sVT$>MT67sOa|L!DGB`W9sYHtz@zB_EKKUzGue<yx;5nECqN8
zKWSZn`LHp>9H5);)m~|NX3qKsrED(dUeondnaqDvki_${P-XhE9v<+d_uVIbpth_W
z1&7GIdW*3IHCnq|FW%mg7#y0_i$tUGKBdXFg~S}J#mU^|i@+t~N}A%@s-Vq@DyWcV
zA$0J$fp7!WOkeH)64WykC-Aw`Wk)mL=)Q^qpT;DWC`@!J7z2IoRC<${H`4?C7{F@S
zxc5*notBd6`c{G8$pRZ=^Rf17_>0^Jj4D8)sx`p>mwyHR5B`4iZ9|)P_;Kc(=Fv2H
zYs;FNzHexK!xx`_9+l2`gUB9gFuMT=G&jeA#vJ=b1bUt)=A2-80oxHw_PL4dF-eed
zo@jl88ZrhwhtqT4HVnkjm#uf)|Ji>dEQI!e<95UIK@Mu<qB+w&V59QreHh>zv5Km)
z73InG+5I}_pb7;=MS4GpL+PSG0wF0xR}taqIHae7N~?Hr3}7}^O|O!}QZZl^W_iA8
zlcOvcuDP*T)Y~O%(4=N8w?1W0T#NTi6d$c?<QWh(>4D9#bUmnR<{3hk2^P(ii!MI9
z2x7~42iQs@JtynO0XEA79=4V}oMxYxb1NlDP_;LKK)iAhvC3NKwk-AXP(S7M{()v7
z6@&gQL?<(GHTeYo)<e&3n@y`O!c9co=re057t0cl7(AT3H<q>V@XAM+$G<R7ve2ul
z!RT3%8CI}FB2j2Ibn{%dl%qIIcTCVQ>;A0~^*Nww%tA#=CKY?UWSxhBdD~WJ>qx<-
zZi<e2ZYe703<PYFrXA}~Yg%yus<4pYvMwSY{NzxSF)A(7`3%PIN<+=_WuA97N3(^f
zxUkph!V+@L8ix8CnPFPYJ^1v~FYw6+4;bT)d5&l$@vvdqz&PLGix<ZaKYPKuFP_oN
zaC_Wvzu&R<Jp$5J3Mbq^`U1Q+jPu0v+jr3*&j%hK_h`f2dRQmSbTYaY!`cl2o>7L-
z43R<OC(f65X!{=7+_y(O9|zt&KVw+5vM`QuG}3b8myo$lps}A(x5&Q-ftoEQcz6|8
zJ3&N>MJgo|U%u<!B;agUqnX39ti}Y685w9w|JKl&m45_?;HYdu$TP4|;>yp<`}90n
zyvHZ==@>1nuagt!Qkju}1bjN(!lJoIMzW-_QkD8%J64yKxG#{B`}SJsE_}8YWu}~1
zHF=ykXN~mS5m^?&Mv<1=n?NC`FskrUI;s}DEbAiYe&^SFS_;(eENF6D%sf}PO9*g8
zHwmU<x)A1E7Dh?xj%?7C2b78>UP3t&9ziO|TF|p{6QwLr!ZEW<^b}NHV>*098lB?i
zPSj`9fZouoBinzgirK3WE0Kg1+v!?U2y1y<XAXDNJ47IKa-YJS*ZWGj0bq;+$q+&~
zA&40qwZ@~dOQx4(uh#Ts;t*rRadj51!IG|&(h%+*kAkCPjB2wXWFi#6BGGxVs1)2c
z&fr)xqhuK@=cw~?S{q=aLNeL&um2doxf}NFn|R!Sd4ECpUZ1etHvIUDclc|+`dfI1
zj=taU?z7MFhkoI^L8aq}>uwMAkaL6!a-1=V?=&3u7d*dxi!VO=3{Q_cy1j{lq<t%^
zvh@fE&oit|Ng_U%&y!%_K$klS6W%(GX?T0YFqqyN+)wn*Ym+Y~3?~r9ZzB+=4UZ0|
z4mPG-&e8;~<nAp&T!e^=;7H&kp}Y#$SuT%eQF;5Y8SVsq(>YgI18drd;!*97d!+yB
z=I+b>-~t51%PUW<)_`0ofZULPjfReaJ{@+9beklgU>!ll69U1*7EotW=p|E(ol5<>
z9kIRz+{m)mx?yVjW0^%1Nm5!@q^7DgE#*jddyRdHHBhQL7XB!JwQgLhGg~d8_kX?L
zJOv3te6=M~!7vKKCZv+<j`XtBUQ2Noz#QHdlrAY|(#{bqYOrkG5W~woiCp)8wWw?s
zm9`teS*@}0%4AC2uftF#4vAfFkOBj*);qhOHYpeOJ`sWCTIyLeN5_H46rd&4O)aM*
zomcLyHTwrd4l7Y;qLeLp%XM(E&dbc$q#y}fFt(!LH%-{7^hc`<Kv(aiP?EL~dn}zX
z1rA|PrNmn{V?!@mV=e!___c{pGiaRcMCe8t6XSpVSMl;+`}g2SpTEU-KluRTJTYO|
z_Z=sIFW$Z2<$mJL2OXb3e+EOx+jq|x=La~OVcTOgr~I{P-Hs#a<UCJ&{`p(n?*nh2
z4}AXa#3vtpfNy;G5WQ#mhG{KkZLrfOngW}^aUPiW7tB092M{ef+zp?<dxztk=+<F#
z1}*7rv<#*@wiyW)V@w<q(I<$0{_q~wd5gDFA&}BSo~e5(z6=i1@|GoAO2ZJng7((&
zv~Ms6?rq|9X32Q)$v937i!QENhl9kCsJ>X#`m!=#EyZa0NFh{455b)zsfSh5n@4<Z
zR6|Hfp4O_~K(3=H1V+UcJdv>RvYwG>eE}$oMICZmcwNdxTdF{FS^duUwBDha>yoQd
z0Q$Grg)&ZRnRTKBmH5D#U3Ag^D<Jq9ulG|5P=B7dt?pH3%Yvq{BIxCPqDZ?6R}rr?
zFwibmMKoKR=}cSwY*kEx)R<XWZ9#Xb7%;03j{~jEY>QA(Jp^y008WC)kxj+4-VoSD
ztSg$SpqEhF^(A64EPmq|Q{a?`ZD1g}SDd3VJr)Ppg$UQv;{V0l4E#@sz?rVdCR*DT
zV=}WC#yHf(qwyZBQdPtM9<7U;qgntsi{n;Bx+rZ!AfZlW(DKioib?@GSBno<jAY4;
z2eSXnXYZcz^mxO^PjB%2_8o3aQ+<58#j~}J?|l3*-o2f8_x2rL-o3*^Gx*kEeS>Yc
zSOw2P0q8jfj{AXgO#I+SpW!|n+rtBX?Kggapa1SR@N?h!7>`enXnhM7wT9^@C-#N)
zHqI0CJTd1P&D;QNPak4yTUe?d@9laf_jFI{-yIZcGb-8dh|!#bP2AjlN%-*ejmx6F
z2nf+oT5ng%T>rKT8M!tdw#XFPo3T|(0GCZlk%|ilo>C2#vA7Vey#9(+Btt=Q!)jcY
zYvk`UCK3byfkkWc`Yev@Xe$sRWUszM?ha6-wCZ+9_NUtvq<MJ8$9nE&DY<o{oizNV
zAoyjU<A3F0kY!P|3yYe_1FVD=t9xD_yJXJ-{2aJ=UVhT+y<-Xz+AY-s4->4Fuv(Zh
z6eJsy!IcKk3^D4KJ=vKIl9?~n0D}j~AtCwT7nPLALK;G;ViOT`%fc&uc8-}%>l(FG
zrqPmU4b<p!VxC?Nra*0vmYE)M3%RdunuW0zw4;M`7t#h@ZAAc0Eg&vgK!~~0Q|<wj
zl_PXF!^I;-sJRO{y=~A}ZNR>5c)V?RWLMFy_5n~qO;ppKLFRc5&P<d>fz;XIF^7_@
znFfw5x5yswD8K-qQt6hn6?19NtQi`5<ia<H^^X0v$JzA_cH8mc!xQ?`8(<84xKDii
z&2Qt^zW@Em+?Wj?f3V^C`5iv_=tJ!L1JHJ~zH!v06zs%2PP~2hj30mcIezo|Kf-Xs
z?ePh}{n_W(w++Acy&qv5&-le(_*pzWM2!UIfU2Iy7@5`e?*1G({N1qaH#~mu5xngc
zh<Em|0@HUhC#KN4Du?F;P9JzaPQ2U=``)qlezmZcAAQvdB&@Pu*=A*B@z;=@uht?s
zE@Zs5Rs-g_Z=E!%eqF{l2S0-yl4Zv%ZlhKC7_P#7MY*d`vkF2>zzd06T`1GU)4A)2
zI0Oe*0m{Auiapy!G0H{ib4K&K-qTRYebawa_h0xC@l0E5*m@L-3Yb;4T7k)#edV~f
zRD!NM$veW&&RKG1m2bk|i%9ueulKYR_#Zbr=k^-Y&CGydm3}oDX!Z#W0}q!{9hL*o
zaF9TT;YP{yMaT_9)Hn(eECySW`g%xfimNg>Lr9WQl<EggL|PA-THf;K825=qDa0GS
z5C;D(WPxOY5^TAb%g+-sb;JH~6mtYYw`4BpIa;}(BwPm-QV@+ZU|EFbV*+QUj<7F`
z`oI|WJz^+;;smi*=NK3gB_96pjn5lHkD|)eomIDnR0^IaR91m4ztb~$Qm-%D#dGd0
zT~r;OZ3OJ|1N*OjivG*r!fo4edwdg$u=No3#|ysu%}2a__k_3if!-b8`s5>g{P839
z$46j$1Md%ktbL*_G<1LWIo^Ku4&VE&ALEO=<LP0;<L>zGH=nTe9iM*sV|@1R9+n9v
zCi?@4rkYQjM>Obbh}gtz4W|A`E!=C(fhaDa8)b~lm4}6Z9V#?5pxFeb;W&Zgbj)*L
zY^-aLE3Qngg-i=XE#BuZ{apo(X*6SD3n^ubUOjVliMZ&&^4ji;^*lYmjK{fVmYtyl
zIb^+;#0{Wl3%^tfynOV7qj9I=SzzC<R+hyO?MhIqqMy%|Ez~XS6i=BG9s{4CWgG4r
zMY(G4kR1ie0+4ixjI~>D7>;mnfo>~T3_${2%3)R31_5tzjx3GZV6CUl13+d2Mo!Py
zQw#6!K3W2V3Z|^uE}srpKisqkP&cJzLCS$<w`v^VIc4c;4hG}IVcGTXUkz?}6IS3l
zYNMqTBrM&d!=!TceF+jfV6$@SL38QiSESYs1y`=4I<pFBC5+HxY4Jq(j>*Pz%!-;0
zV4zvJk%SO*8S($*we&n`L3o`*4slm~LW_R<x2<E}dJNlX79)$fUokb0!8`ggyRqGY
z0>-SujI1rkT|j~Rtd5=uD3Cy|Xd2fa4574Q&Ts6ASy)M85PWuE`?o&F<KO%9czk?B
z<5V_Bi>Z9a9Up%1h(Gv?-^Fi#`VR1kZ-46>_~3&lZ2KE{zoBjWqU>m$^%%@!4&1g4
z_HxI|{lsy9MyLPY%<!G>d<)%x<K-P3a+T1Z?g4V<7-Y&8Z9ndJ_<3T#0goR&;^jAg
zyR25C!bwPLW@Wj~5K_jqo^#+B4fkWB_W^Y)c1d$AmY04G0xbZCt6-Lpm2l=NzGQ_6
z`76fvRqm13uc$jABzT3X(%|Q7dQ{S{P*3PTi}xgBlIBauC!NWf;QO;bl@x>Vj67m-
zK%{pU(k?fv8J0dsi!#W)t;{U_-7M&-`^4Tm+y}-q9L;d>nCj_@Pu2V`H<W^1_;N15
z36ejke`}x4wN4D{1%LAEy|a&Ejh-5=X0I-SPQq3`)67_!a7$H^g54M7okE`)Gvf{p
zWXDf!<$@rDXv|t_XQ9s5SwOr|XmkhMII>ffx2dt?ec}7`6x@>b1S8C(DR+yQBpzci
zBV^8^inoyW0Mx)_a$?qO9x;#JbAKgl)zmLLT24=Nt{|lt+}z7GqOIh!$1`ZsRn#R6
z?ww7hfX1Q2nxT8VfA1ERm0c}+4bx&w>octSzeoe*nMoPQCF0Sz4~W(Ui~W~d$C=hZ
zgL+TF40yBQ;pe`A58r%%+wBo(EdV!Do~?EG{TUxUZg}(X5zHDM9v@)shS46-w+C43
z)ew$#Gkv?^@#zCJ>-hZLfluEZaJvC>!#D>%d}{c`pZ^(r^3fyqd0-v~+Y_(`g#+i?
zwqYCx&gXa7ZU8(K#5i7X^AjKc*Z%>0{$Ksy;pP5<=4>+p1A0%63!XMaW*Lh=F>y@C
z^Eq)lC${5^o<1A0wnwij>RRjrXRQYrYbt$PEaL(zLd9MZDdkaF?>gu#{#q7rNnaP(
zOB)1rrMbS}Xv~sk92jk-i>oDJR5UJbGu8}IA;<#Ke1S}<b&+PkQDBut<R#uNYD+i`
zPJcGk-V*&642;z`<AgM}9GPR{U2d{x0zy_e845hE$`c?ku!^<vy|uMocsC8JFJJ4Y
zz1}mXP(@M79j{`T(tD{CJpAQ)ET;gL-(w9VHI7>j3$KLRyxiW)bz``Y*6PY4oI>XA
zr-YAnkWwJh7ZM0G%zwX9a+!vzYrygVlSqUvmfIL9FH#rzybYO^<IB2|Hg$~&%w}jT
zNa@(v#<x;b)8uDiQ$a7Cm#5Np#Pe|sF91M8QcCGiU_-g+Gg?vezUnL)#bHa<m2*x+
zT9DG6wU64bEcBC0QGXOA36WCs_60!9CM(I%e(@7L{`>y`ZhOpBW*ScDZujVC1jjg^
zvESZgeL>rw&~Hyb-vU{W5p@PrirTQ<9$_$i=R3gm^ah{5dx636=3&RTK7NA_-)u0C
zS<LqE2K(a^YfQplYzEjiVcUtmZMeU@1ICH11x@%oqJ{JAhA%#Un<!QpDe!WCrd}|1
z6|l&NIfmmn9p^UDM-Y^H1g%0h=P;1w&5BIQcV6sKlD~w^rQ9oN_Ug3tI>zNl{Q>~%
z%~CB^_u{Wz*?<)OW{7s&KsKvPxkH(x2U>JgtdHPqPS$#8p%mtIoyGNK2n$tw4hEx(
zOA;_D@xrIl)K`Utv@i?+n58HX=s<NlB~Q!p)<R1>NUoGbOof~8w6uKIb{YxVzoue%
z&mlWyLN&uObR(AunOv35R>TAV81qulW}S!bmOY6;lUZFL^>BGz1C)Iu(h9Yl%RqYc
zR)w$#%ZrjgZK|%T_2pZ9AwVMj97F>w=Ho(??jba!uv#gQBpaDRvKH2}<2B}Fj=D#1
zwf7Fqx~4WFGdX>uo)wQx$b*l7W-xCUv`Wt5xSta@&QCThL2~-p8_Js4d(Sq&F{(30
zCb~O(I81hiwpiyZqNioiT8E}Z<wob8E30HtDEbS1^6_)SzU|oC9@pv}%^Svu%HRpZ
zw(S^)1Fgeaw5vwHVX|wj&53y&nB!o_%y2(h-_UP{kGCC9Zw}nggB@8NTkEhnFz117
ze?q%`5DKPQR3R&Xxf#$M{q_XgAK~W<Y{Y~(pA*(P9zJ|R|M91Id46H=9cZmA1uT2F
z_lBpvV|v7HT33fn4w7fDrPaM5>re+W9kdesBEZE%S0|8YvfLRJI2BKpBGA2(e{4NB
zfwlF?7G_xd-6hN-v&cC~5p<0;6GeFbSJs2BTnro`wu%v&0o^NZBGJ#XgyctFWl^-+
z#(IbvBE}^&^_!|?1c2m96C_Fz@O+t~-8s}p=b;#~FQTGcV{%;s8VxY%dC?;*Pz$fh
z%zww%`zZx!?cBFj+X5?LRi+K{!1G;m#yoUHle5(Y1ifAP>G`!>jw&_-XQadkTP$8s
zUpB$(dMVuX5_O;m`Jxc?mI6@B7Ks&Auym$HvEDmkC#{6De0{!Va@XFhN+FPk{GPz+
zu{!C$BuZKZGNBE0@4<>WWg!rSMW#^Xb)O?#Rc+Y|eLm(Ou)}1|MCjRm1V)oK8FM{*
zq>^q;t&a_9J@FX>|IGZHFH{}nD@G*dY2vrmaNBR#Zx2A*;1s4di4a>y+is!I_8W$E
zw048-H%9mTa;70LxPI?&w#yKckK+!T6NeuFMiiO#E$-9zSbGP|82Dv=JbYsB8*uKx
zcnQBY8n*8|;*<Z}pTPNN|2%&1gCAm^Cjd5^<eoVNn0Lq19s_h*i)4<Si2`LcQaMOS
zE!lQO$`a&;Yv56Cq6Ky8XT2X%04wcZ#Cy_S3a+H#c%F+Bcd<|t++3~}%`t5*0)9>~
zLX(+>D+rC^+5!$MFm&?B)}qL>A!Zk-OX~Xk!m^3{jX-s&i;`)-^0zwV%0l;r$IIKG
zT~h2%{902irL5wx{L*+cc&DXOi1jwid37DiGG}5S+#3%3^w;|Zn(dX}EB{cSRjXMO
zR;A?(%Z5&mUNuZrYgC!#7ex8-Q;0A3^m-3TIDRqPYUkzW^E#^unPj}c9Z0rOfO9>P
zdD6bC)#QXP#vuCI=U&v6!$=ZKFNePDZ7xeIqpz~DyR6&97&03RN;G&ND($t&N{|7S
z(wV3w9yBX0=i=Vu-J3-U;+6$%V4^|G^=h5DdI#JsnUv}v-A1W&U(pZTSIamB+1gDC
zP%#?-Yq<3d58IB`dLVxVWuVq24Xtf31Ew3i@37k&z&IVtR3YD^h4YwjKlvmx1z_7Y
zG(UminCBfB18v*T_eY@L;`1djF`3C8Eu?AnY=~x{XNS!*tch{LF!A*Cgb&`l!OOe1
zoTs5AhPZzSmKYSH)K7OYd`p+oisV7sx$um*C>8fApmltH{RN5_a(0T|XX`af29$+h
zsBcTwUoOY4|IRy#N9Pr=0mz1j?yQ4wCPReqpYCWB;A>wXK&j9I7o=EorqVmvX^Fq2
z+-OUuV9gd0sg*k`hxO!HR9lO6DLEpvS1`PCVztMmyi-i!n%p8e<y5w2N>=bpA{p+H
zj*ob@=V$!X*LzwDLIFhttJRiBfxncQH)X1Z)QKZ<hs{8`)+{m(gx*NU`b>(c6fz6p
zTZrEhraH*w;M|u(dBqG`IEqm8uKH;~U$8t39RT!be9l!t7P07VwHJON+PdpSUP@<i
z&n72=o$e94cM1_KCb|X2Qj<HZZUjVmJpn8<yryFPMnVs#q})$6ySdH}P*C2pidp@T
zaws_!Qx^GHpsLmHko#b~)_T5D<N^lVi{2XckDg!_m9uiEIRZ(C7fqI%;Y7!FdtmH=
z1BA|*d#$u;-OqjI*+xdu25dW2ykz}`wm$&Y6HzCmk+7s|jor(E8A0eajq$Sq`}=+d
zw}16Jc=H$DqWOvOas>8F8fb~ecQ>@HVN+@+_jjFZEFM_Fg`Pq53trzBUc6Y<>t_~0
zR`db7*<`_g`LikBlbNsTW?3pO2v|rQ`LO~GdcO}0OgNR}m?I1MF3-&WlqxLrUI{Ae
zisGgt+mNsBY3-?}QhB|>wN702RD0xLHvj-=5gSoXY9+TURu4+HNC;%TMV_Tr9knjv
zk<D6U>FqICX?iqaba(vlD^BY5>%A=n2{@rX`9UuS&c(iNpDSiG{#$OKAkr`uox)5H
zJk1beaClV_5dS@s9j~_vRUuzKJ1sJZ3rF@JPo%zlT?s3$6dgku?4pnimr86vwbziZ
zb6HFZHvTG8#q<lLX}j*V?ya6EQY6EJAuI)Td}dQC0iH>>#kaZc4`L8}<TS4ZeOL(I
zl3*1WR+dp-BvZqJ@l?q?)LPjjgzC-;+Pix4Tlqk9ztc((eN{V*IQoX|@c|$Fm;NOB
zwwLDNep~(+Zkk~RY>%-1fPqk$@Tk^S$J43Fe6DKJQczoKXxkoV**nbWs%jk^RhREK
zL(E}l{4yh5H<;}Veoy#3L#e;nvAub~<KrXz9C(@Fe1vj?x|9Am$;Ps0hL>`6yrL;7
z-O8C4@5ll%E4OTk#r0M1<-h7pM$7n>Vpu4+PTjFQ%c2}4^OozXlle7P8E^L2MOwx}
z!E>Ym36%R(sv3|5k8m%qAaFUmv82Vr(__igldg|n${LcZyi)Z2vR0jA&3%Zp{;h>m
z1Mr-U-~iV!3q+$pVp<_udu}56g;s-gX3;fZ_72cRe@##QonP-AXjU4k7OSRF&;qlT
zpfq6BP_F*|_@ANBZV@%(3>Jw>*EGAOC51NJb&z3-(cTDQJw1H5(>bZB)}b`WxGMD%
zJOjec11_SKLe#w4Fx_Eb_hh0GEE*i(o`BWc704~7K%(3g&&5wp;abpYGBp>hYVJ#o
zMJj6p8iQG8CHPbMS*>wJCtrH;^R2y#{UlLxO=^NTH92oN=Jlm?T{a}cQUR(H)!_2A
z4Ms<~o_gPKyWQ|`y8&&vhk-5x0M^2-G!5uGwuh$>9OoI$*-RDKfR8g!JI~r0fT0?X
z8T6Nmfyw|<2F5I9BDL15-bMT6A=v9dJ7XI0vm*lbec!R21IKH`tIsvS+*s^HFf_kv
zlP&jiH9J=^O9S#^m9E!u@?_DX)d!bz|I$Z!MprPPq(Q%A6Mp4uwbm1DUhktU#A%ve
zKC@}%6~&8pEk|EQOBn@ob!A&`U&;>2!Pkw!Qfpy)){=oK|9AuE$0QvhC!#@ft0OiL
zggiZp6I<ii8bJ(fX^Foq))&PC0LE5C3M4?WAhxRc8n5>rveVUi;nH<52!BemQdosM
z!;2wU*zwvuQa(#S8ZH9x7BPnfMk>Hkk@q^l(C+Jjvl&%%AYf^`W$z=Qfw=T*aa;l#
zL__I1;#$?ir`j)z9{qB)1_}nOJ=vjEFLxyqstwDsPpVP|uRoKijeDm0kE)vhJe4vr
z%#vBe-&wF~lFO>0v?g1KaRmfWy@i?vM*Dy<JqbFMofWz<W>eYHsZ#e9lj^-v49zSR
zgL1wxqNAy-&0szj=NTEg->^S^2x~jW92C$K!%xgHqu0}=9z!XEo_%>T?H2}X;eL`}
zkju#_VFP`(WOA@jurMe0vTg8jrh@KU$L+R9Ce+zEy-wHGQRv)!vF;Y+tMy;#7kOT<
zf2T`|{z|x%5MCAbRf8<@_wuqt9IqV(5{?Vh&h-?^H(dT;0A_tIpLf~&PzqEQo@BSI
z)60@j(hIX4Cxvaq6-u!B;Jyfzkcd}MnE`Bg4traXdnz!iz0wY6A)H0a5J+iFgn2Ik
z?iT%}Ao^mAu%S|HmmLJpDZDL<;JlpbPkz03_K`;WUB>%yL*tw!Aj6bS0qwwK1?1xW
zSu#;TcS!Z#F)*$;C1H>m9j#g9EL#y#TNM?zVEDusErx@f17k#{jb;R(uB`oBzSp*<
z=EveXvugVFiUm{Pt_o~2_gQSWDvpKjzLKc<15y|{P#IWD?;GI}@}06(%z>dml!a1<
z_IV*7v%C?q=^jlFTb@f>W*`9*t(}v-kGKY3(58a(X+e4^idTW8yjO!&T!TS9pN{cz
z;QsD8JUH8qEb}K#x&sFcZGS+&-EgpPmC|8f-oAvAuv-jpx6YwDa&Lh_OG{xOI#wT_
z5JswmHd+a55PscLq(E^qm@}?12k^_E0DtK>U@xeGE8r_!jB2|gg}|TfRR&<46z_kf
zCehGBR0LKMa;Wd6uV1}N%)yX7jnw061rJ|ogr=GNr3)w6)BAESWMSuq;Np%-#bqIn
znvk^1F-=}lFdxX)-%_&Ln;5?v@hG#X;D@!fD^(R}`b?#eyi5KBma`($TMPFQjPnzk
zp%a#g|MR^I0Dwip$z(RtWi9-$<uy#K6#Jn={Ti?LRtj>@=;n$6q=s~`vm=+<LYz*l
ztR9PNZ9%vV>{iOuE-yhj(}xPI`MFh1D?w+Ar4aX+J~5i(w293eXJc<3XOO$DT;L?p
zYcZNH1Q{71VaB&+z_e)H+gj8U@Ka$~)p{L<ISmJd0!tw@5*>}J3(}fIH9}iJ@o2Lc
z^4+NbIxJgI1j3F!{SCk|kOA}IhRzB)S#$=ER9i-{-7OWh#%adAOI3;x<&!;U6tS(6
z#4KVjTT_}V>a3C`;(BXkZCe!475{n^j?IC2p6RF8%7oVY*3q^d$2o8uCytjp_HDzq
zZ8(m5nAGQq{_q&N|Hr#1AZv)gU8zc_-lPDEq@~xt6B5qL?>(($@eT*HBf6_@f9e-7
z{||oy{(PWsk-2m&D)SU>q{c+4^`7<R<$B5Q{N$uo6Y4e<$dy7USK?J*7dw=51<U93
znPQ#A>L5$na-oaMQUJ|X2|@nPXD@}|KnYqZaD8U=s3jk_7Ts&-3}*4@PCTZnvI(}V
z0>V<_jRHQhxm&JR-jLiSxq}!Vt%_+ZdmMDrTW4JY#{nanOLllx`7Z~Odz+SuGhVo!
z?<;>`?nh>3;cLC#JJ8G@v_yW7e>Vj4??k3SAPwGMrFtO>pSIS}x0dEp2!P7qk$F5g
z!27k}H(Cp|zW7`>%f0NaV}oP2sD_=>F}9h#mM$-svTFFUQUIbCz19F$^cs8vT0}6>
zhf-1_!$1|ta*>P)Y@=P;QV0nvcX}0$V!t_M0@oQdc<dv;b{4?wT$h5gsyFC1aA#pv
zv~5-*Z*iCW;Q-k7L&yy&Bjl$w5UI|%-Z>@%WyRpCP3T<!fL0!~H*&YgD7TcisV-ij
zNi&0e`%MB8O=5#b*H7m&4>#DD`2346@cHL&ak39l2*QcJZFqWo#J=CS(?VAh$Ndh2
zN8>jbI%^<=gfsh!MR+X>IVbg#MN1b$)>VWBH}eak@i8#RS(E)FYqN6cit$zhf<#hs
zA4uI~{Gsdssz6o0(o}IJkbS*|BS`@l@-IG+SoY<O>)GT7ZTYOPa;-ps<epvrs0}9F
z&!xR~-NjlIbxscuGQ4JgU`))$SWq{tAa5EcDKc)5ofnoh=Y6Fmmd2cr6^yX9Hw({&
zT3qDJwAU7p2S3S$u2f=unXC3ldqqc+t!s1m;)1^x65>+eCYSHURQtV93XR%Gk83IL
zgm!_wHTd4KZ-FkQfi-w1%<rg-?X2M2me#+%M$Y&!lWCKgm6V5!rK&o#<T?{r!De)e
zYE$<1LFVOOIHF!gg_CGk@71wcS}dV}CT^SIVT&$^VnT<4;>EBL7?V<%R0{S=Ddej1
zr<rJ@2)rAuC0jG1Ub**B?jv(x_AqtNX&4hR9S;>8OACctS4jn{rGPAqinVbtxyLmf
z$os2UDCa-7TAOA=$7S0UWGspSq!|eqSi|<8{iE=m4d9H;(=1#=lUX%mOuT&l1%B<<
ze*@3Ycl3SFgXpbc+d7UT(D{dNp0M2>aK607c^vRDprJ`-EhmR*_>j&*EiPRQ&v;wh
zNKN2G4Gy?kY!4{bBcAFkUaIRatgxhNiuRL!Uke~9gGFfM?<w^cxKkfXHqcFV!xGx5
zP%iuetSzxMX0p`K_2sDa%GU`FB?n4Lt$z(8N6y?3(AheBM!L1zM4uCyd{9CvYJ|qr
zEca+@VhPORoD!USZ9#@11)gZ-I@&_3+9AvLCAxR<$;8xW<Q6dQ89f)(A8k?e2pX#z
z1<SQ-A7x{Sf6~gdT*}vay=RxyBCr5doEp>v-EUbs7e=y}KU@o`N}@OW)-ThLD$}67
zLJvg8Bt)O1R%1l<t<ZSoDpvq{l8DrjA`Mj8QwtB2!dp#4v5HVq+&%<(%PQo#U5x6u
zr`hE&JbM%&4_);iS9GT}`KvV;Dizf(@fXccFvrEWZA1%mddz<{$2J_tGM8aRkwO@@
z5U+%IuETY%5kokqS^~#aWKBgQvBuZ{tp%&f&Ac|~Vv(SPt=qQHLL9(!7Pc`ZG(+Jn
zSXDC}zW2S~!uNmc2e7_jyPfE}ntlWJGV$*1TbK_#-F_C$0wIp`j{Cc}XszMVK0x0Q
znPb{ZtwTu{ixerL=K287H@eMYp1?d{m`ssu03s!J94C(Z5pCLEqho=bMGw^eENGDX
z^I8U!5};5<(b(62iL$Q%^n!YFb4uN~*J-}C&o36St1+G=a&U2?6#YMME_aP+lyNM^
zdZJyoU6wu_I0ozI={jog?Dz|>+7)<N@flMTU`mQx;&-Iu*C5luzF=w94D>M~g`Dzk
z&KZM<TDenA0+&if>&k&jsEn#yuwAI4_B$E-r@!8>A>MMgh%Ho)9W(as;P(`oupCEc
zs5nCJojsJSToy@8R)^1Phc=usD1GrCCX(g*ods}nID5hjU?PQ5#xa&|dx?XVO{>Ex
zWHr}4m%=bZwF^j+K(fKYy+PYF0hb^_JUZFfy-*SlzfgA(N_=anxYn%T+>z8Y7R25g
z+T^*l8HH^v{+8V%6KRyPk*7e2KJF8lM)xxYzPvjQyqp6kH%YE5#f<b?E1j=`%W`kc
z*8R)BE*cwj9d79R15vy&wm`+jnZe)xt?zTK1AcIbeCx4>IdC5*`Z)0IZ+rs}Pmh7-
z#=wN(JO{Ruvwv+7U~(M=MMHgTrpt1XBUIGIzB}^-#vMM7uoleFEz(@)7#QQg7-zXN
zUNQKbqZGqJ58*mkkw^Sb^aj$G0{lt8uhDpJG53&5sM-d7wEnT`qO|aAefIdCJGMTP
z6y5UH)p}2)o1c@IlRv+~ZL-aWa{HxVCfo<2@9dIT>PR4ybprr4ixgLYak^er_)3n*
z2R88jiivpwPHr2Mz}U7s;?>zj*@2i^$k=Bj9{94yQoJSn*Vh_c2dGeB8F(`a1TS9_
zzxUUd`pVbu%2J4Suq}yBo8GLUZ82fUdUeZG>uNy$WsZDgyWw>XO``0yGK$dP6SnW=
zE|S?e!m*u%=z+{^POSOi-PFg2*M5Bn;sW{Y<g$>-mIy)If~!(knFaC@T3S7px1vB<
z0fl6^GgsN`dYPf6U!{^*6=KSQbU&<XigJx5Q-gakS$!;~O(o5Lr<){&6bYH8_c<qy
zG4XQ7_bARW)ap>Ctz(%tQ7S<{mc?TEw^gBN-SI<5yFDOaJkXT;?^^Q-STM}}{({*$
z=F3ZTc-?w1v;#g)oO=U&M6Ci0U>4EVckB-j037FeVB2qKZNr==oKxqVgJz}zi_KUE
zVM()*SARxu+z0C-CcvzzIVO(#f&0rH=W)+-Ggj<sA6f&mok~lu!oCy+l3u<PIitX+
zG*(?BvRM*r@~mw=dVnsBRT(&HBxRkCPO>~r<qj+NMlnFr3ZLPkR2GFCzm{v5bc{%-
z2pPoZ5G5HMlM(}qt0`xxP|!~bc{fDhm$d}&75px6z<^B6U!FnlS@%daLYp<X#Y_@|
zSG*}jN}ZRoq7ta_Z&~x<A_cEXJE#>pQl_Xc)7LCh_}#U+&?cx2ta`Ob*qR|-9PSal
zpUm4<Lp1r&^Y~0YG=rwP!Fm#Vpn;x99{}cz;2Z$>4DvRMU26cG&D%8N0EX)t`F>UP
zW(9T<U}<k*zaTU#3dnz7ev!;ort?Dm^bN|ju>2u@O4wg}vpQ2u>sjx~56BF^{LZ6%
zKW8T=l61g`R8;>VVFV;c3!8~#?Lu)R3^cSRf@(=G8g-cMfDOqYiDImXzqY}beYA_x
zGXu~!V1I<G5#72b0I-JE8$S8u8~E|3pW<|ItSWHaXIhT?*0F6JFb||YN3>b)8~V1x
z`i|o~alfB<vglBI951Y;*<nt%kU1@tWuHO|hGIDH7~=&#j>vE_gO369J?1Yw>``1a
zR+<V1SzSpfkVTiv9>S58`j-3rRkT`LEy14zC*_>2NTg$Q_KV_4YJ5e7u5;G8UKL3p
z*yYpuw=TcDzwS`3k^&VW8M_$<BDSSxcaK^a+Zwu8CZH3Xtk<r(H@TKrK@J2+oQvGB
z_#M+G_<qVMP=wpEJnrAtBngyk!ImkKs|!fF2K_zGb=mh70P-Cu>lDl8HR6XA1pnmM
zdntukIGz>*SSwK4Q90rgVvXsIMz=@`5|hB06l4elU<UIJc|FT*5xW<}sb2)bGb)(e
zKp-uRjZ?5fD4v*eQZcY^c_J2Vr<g`WQ-l{wutX^OT2os~9D7Blw08YzQ96E+cJ)oR
zu-rkGPevt%iJE*@OyKMFN$Ar<5y)euG(0i;#nEBqu9~CTU{4A%#cx_O*bJ+0?;TGM
z6I<US(C#3*iuZ38&AIlig5dco8MUB9rBp7Y&h^K2H{7;8CatlRgg}KX*WNcA$APCe
zPq;nYV8<wg=wZEXn_=5RahNrnV-P%qCEI&k`}y60=jV60?Khm5bOA*Tj55_C3$87x
zgi0hBuyMe~fq9<rd1h+2Z841tX1M*6e?Q*-_y6y}uMM<W?d3~&zuGXe5a^e7ILs}r
z%+*rNmhE)~9X|Jqv7^N`l1%MIyTPZ5Z8&L9G<n5E8RQl}t}F1%V_eqtx<xedyI<Z1
zvs$mImL=fG%)5D73g{g>4EIxGR~N-3@PWu&+?mPO4QO=rItYL>O`6J|G#Afb2`<y5
zF5;TE9`y)A`(jKKS<c2#e2n}#MM3PzavrZYFm*=q3YGZRR0evz_c%_`l9>~=8-+C%
zhBv5BkRBp_7v#}i5kx2xFkLLhMn+$gMd&FA<AF#`7cm2-X9Z|&ETRixYfH1K@ZDyr
zpQb1+6dIMRD@0kdtyW=HN<nS3F8~TD^5Yp6p;m{g*hzg(g5Q>VEc|dmR@rC}g`AAk
z61|!hliem_8d)fl-57Wu6Q?PQsVo>hn_KxyFG$J4g$RT-xAm5d@Git`<+AH7hRnwt
z2Fv|Tq2slOt8$ggLP53Z4!0Klhpd-tAi1ZC%0B+a$N1(a-@tGE)(^sBap0I>O$_jz
zZ+{CPee?ko!-@3P2v(alygc9W!|(qPW1jfv!#8La>7&*;(YbN-CyE>c01aRqnB#;W
z&*3r}C)@{m>*#&MY#V$Ua7_3&pV0roU%>Id`YUYfQtimqy9@rB27$tt@$7PctXUJ5
z`@?_x`uRe@l&}ftQc;^-?qN_Ke_g1h2k9CRPslH;<#Ovf>C``75Nuh}CBCroWIV4@
z#Myiki#%H7jU#G$_Qtsb2|mC+X5sqYH^Z$pH1kM7)%aSoY>N#l<P^YM-sGZlFch6x
z)+qzIZ&GBGaHR9#o0+OqXIKA@W4-cvofXOo2hzH0d31{)`3bN0QVO@N<F<EftSzYQ
z5ef_<Z~#%U`8zB^!YVX<^r6w=xT~Ss5}X2JZgC{N1AW3~xrH(~sLUe?Lu{##tQopW
z>L@^+Cn|SHY3RAE$IBH4G5+E&pF-IePK%#b2F4=nR>g4nE53y4>VqMq7W%s?i`N33
zDqI;DokGlP!|s)KcaDMM48gYdeo1FVTZ8Dtdld38RckkFeT$aH%KGqJG=+-t_$%v3
z&%e6fl9<0lF?v)NBB3(qi6U1d>Oi`{dINs;XTJlVCtjYP0SDf_95}~`?|k!{_?e&i
zW>5mpex}PTM9-(6eTv`w-uLkF#~%Pba2x~Mwqv{9!oAcQ%p#Z>nMDIW&q#<EcSNk>
z#GC_L@91rVBdXvx+tD5$@G=}P=Ue!n{CRBu+V}AC|N5<Zj(nyPU?mI|;gD99=wYRe
zFDsLLH@CC0<gD)c3jT`&vgL>2!sGE{Rp#nbCDfM_S`?A!K&?XjqCf)aX4`Ip1o5V-
z)VB_(=rTHywyYt@RC<pB*2iw>u1q?n16z0%wpI$JnbErT#RIYgz$bwLEiFlvd~R(g
zN{(s&D{w7^m=}N{T<$JuPJUX`PD{q=?-v-AiqU*g@xK>J;pt(=ZQsgdCi>HPB{^B{
zM4DH*Kg{P&RLHnO#1K4NFSO~=0)QLJ<|L8fF`~V2_>EaSq&dd!R1OLv0WdKD8J}pf
z%@BNRW<z8W)bMV@E5Rt@O<LfVUywRrCZ%{ih_6Ea<%2BY`1*SJH%qd(n40VLGE)iQ
zz<arT0M}3>SHP{|935^gGcI(E)_N+jHB+RJPD_M1=aOHq_xqv<i&V*MmMBPny(r4X
zY%mCoF*Ce1^sQ1$S0SeT=$ueD&KKPFj-UPRw{U;H<L%qG_{PVEzIA-?!5cikeG5Et
zzB3GJlyx2lKK<--eE<7D#O>jRZ+-g{z$R|{BewkkO)Y6b#OD%+74SrF*?et|5v@RE
zIdx{mv<>^i8$8`U0@@>vfARy&f95aX_|+f5fAcMwzw-L4&(ka9ZMDa6&J|Evvh*kx
z1nTQ&OCd;57bNpVmXKO6LSFR0Tx+Grg|~uE;=AakI`0NRTam7=uL*2CWWXb@$DIMx
zHeu#y>?|-?JlNTBAW9n?9TPiCfgh|xtuBCwN#s;Oqw9wj>4hW{%@e?pMqOv~O8MH-
z`n*(wCr_mr=Lj=)&TQ0~x8@$I@n5NDo|QJ=>+MSR0GR270DxcT^?sJZzTtM;AfIFU
zn6&|7YBI39vNL1~=?GY;BxS;{D$a0Z*GXQlEgC-ng$^+a!K`(3Zk!l2gx*N=9?RxU
zOu$?8yD<?bg>AMn(7L}&?DFA?Aj*R%zO5j(u8}ub(1Z4^nR+PUvR7jE>)jXSYef`m
zoMZLdPo`8x6TR}+gL|}XwWrqV%elU)Ca-6V=U7bG&f}CYhv>ZOMP5I%-c%V~*K3!g
zDdn*W!pQ3Jq#^nV^W=K(*l#x&JSv8d7}I&&@4)PMhU0d-0dvE=gPkcI&(F{J@sB^n
zyXO~d+lHV0?zi#bhfl1D>8xrFa1#Rf1jY$RjOx_6M3c6tO9|4QFhs^(>l@m>bB`t-
zA0DxN=K=SRzk#>^)JJH4`S0NPpZ<A_?|qi^zpDSGj3<Gj&g)`gTmVN2q}K$cj|c_V
z<?Sop#p|S1S~%y*Cd+}+$(^lBWR0e>ivMTnYSNrQ4HaRL7iExh(%wkf0h3B|))0kv
zpjo(&tTV$%14$C(EJKU%3Us-;GAL%j@lVaJdqst!C#u0ar95*Nvy%g|eZ0!K+5+YT
zmWCOIjm0}bIs-!B(fE-CpDadT{#u6oPOtY<3bzgWw#8zlV^hF6R>YyB)*!U6N|WT0
zQNq{Ip^YdFyDC#%S)`e%b)g=UL&SvHtDo6V5s2a%b-=PN0Pf%_fR-b-FK=sQ`xGJN
z`xAM6wNgptAVj=^jIX$t<^LC)xCp^i2*73Kw(M_y@!wj%OI8rcI(W9d{MlSMu&h1N
zKibkLPXTg+th9?rx7D}0Je$4#e(iWPnWN{LN3hV&Dk$ItA_nxfp*@6Ev)^ubczVL|
z@`7>Pv27i<+r|{>XhnY9@A%@2FK`?OKK$_iOWVK2+SYC9LD<(??_<oh_Wu8KPL(N_
z%jFALt_v>P1Tap5g&<snqbNXtAVm}b6p05Ec*#>F#6zBfl7|FEB0-5`adDzV2?|Nb
z1tJp=iedt(a=9)pLit)Z*X7*)z1N&$^xj(@zSdgrW9|Q(^KXkYtIogInrqH6`skO|
zzV@}-C;0T!w|MjB#(WDcKyiUC1vcDy4WmO6z-ULtwp&CV@wih+8J)-~6(;Nj5lL;e
z-cYt1_S+jg{_v-G|CP6R_uu<>VBdK0&RUYfb$#X!ti=QrQAdAH;?;PK{W$@`Jio|&
zeen@2d+l=1SY1hbWqv+2PV75e!Q^t%a-TS;nif)AUovQgvkF(5ArUF~Gc42kl(aCf
zYbCHkAHoXS1Q9XkF+>X(F@x7Nn-y0xLd+J>lV{M<N{MHp9I}~QdI|mZec{Ql!NjCB
zVdm_ZWPkNs5*0q8A?fF^7FIZ>uh#5ay*@G|)mt%YMBwx8<lQs;Cl{CrZcDw95qzVh
zzz)GSysUsfBTG|cuT{=~1j>acf%L}3$+%9R<XCu@bk+v~xyN(lWEn2^jtRB#nj2l8
zECjPHz-vxF!^dO^50rV&tDad9ge`{|`7h^65<N}Dx(I`?+TdHyrV^&m$7OBS%^p+^
zmSLsx@z1fKn_u?|LtXWQ4%c-*<cas3j}RCZf+_qqlmEOt`<$Su2K~UrT<D^Zb7HC4
zyj(<Kv`1&cTB(MoqImQ68P+-$v4Vb{FdNV!_~eskeEOwNQMc_<2g{_uOjzx9%=Ll^
zW5IfZ$UyH6tsSt@v#RhIoffbNObbjmjIu$zpMse?F$Jhw1wQ>U>_>kO<e&R{@$xVJ
z`><dB>O}Ew6|8tZ)*_}Ntk{seA{Y3z{50n?ud%$MJj3<nd*ixcuH#5Wv)nI_>0Im;
z&90ZbDxD?Fb{&JCJBaru7{hA_2~1<$Gn7*)HDP7WNj|q9j4Wg$RI0iT5Z=@)l>!#v
zVmZGQUuF6-Fwbqs%+*q&u=KncU)!);q<DIqF)ye<A!GUk0j>i(t4*bp(AO0ZU>L6#
z>RY@%E~7cmUL1(4mJNZBlj|F2QW}~u%2cxJj##38;MT<=+yfwGR1;#C^LnITX5FL!
zKoe8>a{0j$MEMJEvFBNVm(1Y!IR+KNZaF0O!F}d~zdm5$OjA`V$VE8knDeu9!kI)j
zH`baJoS(I2EV5*&#2m@>`i6N8nF%BL2&=Nl_m@Ly$lPWh`_&OY$L5J#ix68G(Y)G4
zhE0ni|Lze(0V1rO8100R56A(^qAXX3I}alLPk=!xOB(dvF#3t!8lusY7B+U55H_W<
zPqP+Cskt6J@FT~lGBD1&S8zRXzQ5z%I-Z}ONpBQD>W-p?{u(KOy4B%*8ls$j(Sql<
zpF)4%@51<}{uZ?V$$x<Hn|D|8-ZUx%fmetP!RDJJCD%Hq=4&55g{9^z&u^&UOP`-*
zt#Z$<=opt*p1`m?R;7}{B8K@&vB@HEm;24FVAsnb{40Eg+!-R=Gog<Xq!^;ix-C_L
z<L|R%h9oHCeVtY{OB(4S*k?IAzmA`XG3WL&J81!h%yfBShA6-%f~-#Lz57^Xabd%Q
zQllkN;%5GgaXVLNU$9K-M^{N*4?OY!PQeg=&BN<It0;!jMEB$C3(nt67Oy^*hg}4T
zVU~dnl-YYjAdXnDI1m%zOd+%i_=q$a0)*3P3gdk0%eXEV@q_kZ4aaXYhbb)miPlXU
zmO^}T*+*09u7gN1d^ucGC|<`oqZE15xmwG}@*wzzMJVxCTEi|gf<XMXD8q}8iaW^s
z4X>Xcu0()5l<@%*s&5#xj?aJPr%-<Ecfd-8S`@5X?wdoMhcJ@#hKCtOIcpK`rWU%<
z=-AZ4mYU3Pu8>lB*g6Z^VKi9O-5(9;q@^%e?5?+lZ+_z&xWBx|^V1Fc^Al=~9v}u%
z8}QI*SB%P@x?<$E_vdH4`2)WT$3J&N|G8hq`LF*J?EQlo)bK@JfkjwS@$T|FVC;eH
zYa5@*$RqzCb6Qr=@p(I%e7?Q@`v+FVA{|2UU#_=~eXjT1D*R!Q3LhM>$kRyB38!rY
ztdhDO-LRmCukoq`meA`DsFtW!Dnc6Rb2MTG5X|TcX1|YBmM#8;e2e?ihOO)%_oh4I
zMHCnYDf8Y2$~g6<Dmo)RBOx~f6KqK!{SL2>vlM(VQ)#d0)r;f#2Pe`Fjp!4jkvfiL
z16>7jaGMt{8w)lG<!v(w3*ye@biyFBnAe(Ep@#UAJDRWvQdd9f#m-WJEzfls#s?eJ
zHT(xo+4<|Rg;Mb^UCvU@qaabX<)08pE`LwUA?il1ie(CbI8>KwhY(&4+ePt%f;uxQ
zUp;X8L(&?#9{7-d%R4dXWgW;N$G!NGrQ|q#eMB+5j+MM`=^vU(jE5l#%0;lke2%$4
zwJ3UT98W1IvS)w+dt|r`sTHL#Q7&>n#(AJU-l0EUuvPlL+VO&MHr!uc(9WLfws%9R
z15ztQqenAfc0wAw4pW0^2QZ+rLr}7i+g5kfFZ~w0`)fC}@1bww_!EB``tG<0hg_M}
zQjnOFIsaM=VRbUF!<OyA4@bRPvukL0yb_t)6!PmeGC$_^n3nZi?Z$bi9IVK4Hm=ZP
z3f%m-TX$y6Wwk|dxKNa3{1L$u#^trOuq;Z_EfDb#Y#*Zq*t}RQo}pX{O0Bb!t}`ax
zTJFCXUr^}rJ|t~gu0dq8=W_;qwo-1E^N(DERrW3?OyFU--}&|N^pQC@!OG_bI``2m
zoAq_7{mMJ%hg@ka*ub4RQ&4AY^t@vt%&gw|N&t{bv)r_+hK>ZGxgWU#(GU7sxph@o
zz|6*;1L3t@kSV!`D2*MPWSz$NVMTTyPPX`c7EP32<WEeRUi=a9`A`PS0lNt7I0T`%
z^KeMqHV;!^uN6?PN_2+x&N=?zS(i7$=MVye+bK4PQ;ms`WVwz&q;gy;(BAUU5FuLb
zt<^ayfUkvJC^1z!i(!JI6)4EKcR25?rMg+pFl8AhaK7N{U;QfHzkgw(Qx%-|2ktKq
z?6m+kP`83lKK;bO1)%#5**A8H3q?Mf?+wyT(ilr`<q0U&h3`PA1<#*+8TWR={_~1A
zD!BhI{v!Y^RUT+JXFGt}I_-<Hn0RbXP{@!B<A%XZ)43YpAp#R**#i>{Sl(l`{$dR-
zmO#Ano`LQrN58K8dZn<ilN_{p|IC3x=!C_;1)l3Z=J?WT3qkE;(*!3$-*QZ2Fj_1U
zDf1K*g_PO1+u!xj(vqNg1+6RPB#^98ghkdxhP)xptL7Tab(MUM^op#Ml&^2M*PO=Z
zzdpJk-o+N1Pv<|+2Tp;y9#C5vcUy#;#$p1m#bcoMHM$3(E#Zs3P#xS;i?GV;VlMD2
zAv_|xIV2$#g%Fp=bz=dNoO8an2-k?-nE*+0Zt|VW_14SG_b!gaR8H4t$llE%uuYhv
zXW^V%oMkJ*$oD3<8FQ4XKxDa&B$7&blKVftb9uxZ@9RIjw#;K3>+^+z&G+J#4)J6d
z+8_G~$bb8fWa$7Hg(?P{*z`yuv+0(d;kC}yy3&q&X)QT7mWNtlqeEfn#{-{z^=tU-
z-Dfz*z{|tX?hkBLaN87bK6!&}qxDc~!Sma<M6#uz>^F#3n1`jNa<u``4s=#R<%2!}
zS`c0fEfr5+`Ce!#c>FDfCsiDO;?JPm-C=%-G#Uw7roi!i<E2-o;D5DdX3Lpa1V3U<
za_6OjS%JXCN3i@j6<0j7Uu(iz05YA!m#r~D)l^z*!reT-Gl1kOQ(|t+22_KB<~`?m
znu;i7OInw4c3@t0IYLvXLdo(aF8qh*R}h@~CuLj#qF02_Y{(SP8qa}s#*qRo&+VL}
zi(>qoJe2Zz=k9Y}AEgv*3}8eiU<`BW^BDAsO6Hf{k^DIsyVGcx8&E`O4vzVui?<$d
ztr<UGe+Z?Ns5BN><KVbvU7WI4G_$@WH~+=YF||$poSVY6LTr3PIK%lI-?BJWZoUQW
zT!u8cSu><(O&mF`-v?$LqszSDdU!+VPo?0`oB~v_3eshjKAfs`@XL7fPc{|LOmyR1
zWi`t<h^z99=R@KP$om6k4d^8!)yxaj4Oa-vTj4#1(?gPLWS*Tp&1#KgO!GAw185Jx
z243zDy!*!2@cy%RxIMqYE{4y({tmTPJipoS_PJueRTkOX3L}zA@v;^LTIftR^V|^y
z!&(tlI9&pC&WBW23@)UtRNOxO5=yB!{>U?)$_D#i{WR?8IY+s+vCsUnF~>6@(=L1S
zia#uc>Sa$X>&mrrOZe3Shk&&_e{&B65MHV0*njhThI#Y#(yyG%c`#l9)nyLm_2>K9
z-ZQ|a%=55<!Ao9FZjkscY2X;s%nfib<3Td;GCKi?HGwX?0hUfc2EgL_BNBWq=Y0aX
z6^p$R{;o81kq5H={R_N4QYm<uYRo~cWf|6+@>~gHiV>5}gt7Q+SOUOYlvD_L&Cj{^
zvdL`u;ClB}a9$8+bhh#@Pv?s{;F>~ljw{SCSvJdcyi;(_%@hZZ_0Ac#s}YLn>z8N?
znC9d9tjhyj5Psa(H$ghbV)I-9-rOMh{dvG*?D-0T5u!PeOj0^=;p?Vb*F*e!mob}N
zm0>7ZnV+9?9eYfc>#kQf*e`z*@*n<X;BWa|$tMqV9@fbIH15Mx%sQQj0-F8Td7adO
zjRBR8c7KP_PQ1K(4;v>w{hlx3`OPOdA9sxV9ow_u&6_9eTSYC*j25l9?Keo-yo7rw
z!02htvp@mSJ%N(uzgdIz2L+1Ue6J3ltAhRcEkuF-M}9Z(l_%Iw|2q2r^DoVL38Z;J
zi*U<m3d_0NuhIPy^x^w)Rmd*Y5k9?;%U6nO-9gK;`9;QhuCAl;ORR!?{%vOEtUF=Z
z8;Mu1&lr0)nFj+hEIt&Ke<Ce}MMm*^g0d*9=s~htpX(SR7ek=lde;`e$Mjdk=cr?&
zIZz<}0dE(cU@5&kd7fiG=Yd`@ndJM{zYFvcM;%V6HLm_D$guo<7J&m&;HNZlVYP0!
zd=_3aD;oL-0`cTqIa#lr!f}hsRp%s{(P}Xu>nbj2VVdFqsf@V@=`fm7)vLuy-Xjhm
zQL4Jx>ILK?d)l+T)vbz*iJz00&lhvbIhzT7zPnwh`ZzNwCeL`Hih2F$5j@Acu3fYS
zoWg%v93NN=s}h=1o%dg4!Ti~CR8k2qN*tfJxtD9?J>NByU->3vG_>fc>%tu2L$Jwz
zM@^e}Mm{q?r^=SFqc<=ce_*r|y*+T=-{Cw?s0iME@(HAF5McwS=MA`3!PCZY)w)%m
zBjvs^Tr3lDYk5VniOih}G=uvt;(U>lE+Nduz<%4&<q7={en0g0eINSM4*N5|f$=M!
z&3zf?ZjqOa)kg}#DTH4i;VFLBJ)QgBg?d~gEWp(g$aP9ZHE&QZqTCh0GM*LutN<pS
z<f15I@U}c?{&|ha@ISf&SgnZ|+7wmuSprZ5a=$2x#VWf@b`HdI4BTQ}=j__sJLY!w
zq>f0-buQI&4lY)@kLw~Im)VQQA0Na?e*WvDJFmrJg$br(z%n~=75Iz7FaTm1kuy?M
zyGl@G0lg6HDOkNyAwTYDI}z-RvtH2*FgN9OGX<hYZlu*;5h!66--=D*!f*}eUr^Gt
z5*+DA5+2}I0su@jk;!7-f-g{(NC*&I4sxQLi_+qIu^}QK1Ty(!#54J5u5?dX_D&!W
z-Z3Ln)^&uWrLgXUc^CrDOxyxolFZlD@tREW`8nQbA~S{5R2s`a`BDWS1J+K@1XWn~
zQ02IrYa8+WkTC|VdlMBeZ84Qd$G|{qEOm|EF#3RrGFM^SV0~bm@6leqfiet=jWkg$
zIVb#?mVu}{tEYC(y~ph`3nS#YFa!F7$G>d+u62eb0vL_Gub~ClfbAdpQQ+@A(f;kf
zfc_J|3jEsp+;eHA7-lG|;IfD2UYIB}#|Z+F*bZYYL38=m+(V1~&F{Se4HLP~eSj3O
z`6<81{PIQQ+Vq{Qe+dvkK%f(hyh8Fo$cY>!vKVHb{{@uh*cZS!k<6<HzTV^NyYy9F
z&M7R$JbexsKAUrJZGshZv&x0F!gB&;|G472Q2WTcrLaMXH?NuPeB|<P6k_A|23Sm1
z60!L_0J$k#5V$l*OTm>6!?&(|eF6VGknc^4fg6l-6n{f14_gIJB5j+W4Z)2?DF90%
z8E`317D_=~_?j)@k~8XxCBofY)F|F(K*UV13l95WgT@?Rk^Pv;G@eoBoJ>8v2y$7J
z%(8M3wQylm7v+a#-mbCS%QxcQE&$h>``rC5bY2|_fv5Q|O74m@uSZ`582|5IM*W+9
z0Q+x!hN+^P!1_SDzi@7AfsKK(Zx~jfXUExI&|1U3-+Vs`x&g;|V87h}0Ii*fkX94L
zz(Dmi>is}_dBHdj&urekIACFBs&Y>ht?p0`aVa5#!2*1<Pjk`2mbGKR(3qqIzy>R=
z2E{O}vpEk2*qT0Siw(%98|pv!L(o6`1CYP)8P5OpPq`AA$d=7m#{^J*_T~;)*L%J<
zEqae|@tQ%o*HRYwd5FvSI_A1vK!e{z;&L`O0)R}cS*@xJ`7xN4{M+harjiS~Zr!7K
zVNBv0`8|F$>JiDj?qV(EzWzYJ*y}fzv+{Y)V@sKG9mB#vSkP8;Zl-VV^9ycL=B1{*
zB<79FTzu=-$3-8hPR9s@C|A|J8oD6f#o_A0nZt7G39l6Z0Ao@}-Yeesf_IUvOifu>
z<W2+?6QwLWE+<$si;sU(&XkwsnAnE*TowHC(-Iyjf~==>*}KbT5sL!|U|G0OOjS#6
zj3|GxqW=`muPhhK<K|n<>7$Wlfa9D_1sMu=Zse&j(xhMJEbko_)a6Vqa}o+?{quFV
zr1Hu2kFGGwft;V=6K10rKbZW}JUka=Edu%K1Ns;!wZhEM&xZ3nphfYUzw%e0&8vmp
zHrQ71@^bguINqkHD#}*TM?>8!+CU<H8wTr4MB8e?(`|>!fE2*0pgirUr9g{93!jrw
zB-5niA@_U;sQU>5?G+b7iVN;G-<sl1N_31Kf}A0<y>+y9`ra^aCCWYx`n$e_@q?eD
zD$xFezXbb*uOU)v<E&j@>#i4E5y91J03?A}S_>=EUj*G0!SH0VIWN)JCsn=ip=ZSc
z%>9=C`uM~x*^9zhl!E_mDIDcG7|7agBL9%#-1)iXL2PCj&u4sf5&wRI&IODG+Lw4{
z3%TJr5MdGEl1h$ru3Pva7}*v8dqz>s3beAwoOupkE$g>=eUwtz>W;FNiGGrRi^Cd4
zoK|P@@>&q;YjnG6w-8z}DI;hWgmZ>UVb)=~^7{fzZ5({yY<N03_XRnpK$l$ZD-$Mh
zB!y0(RXi`>%Vz+93#xe)ckP=-JX~!m1#T31{Y+uu_@<WFP>VQOOJ__(Au$=(=Z*J6
zC@+X;MOnGGUxCS_szO+#;KzrrxPB_Ui>8?h%af@TYcCFoKjnQxAqU8q^@;fY(Hr|6
zym`jg{^GAdMbN}hYejGT+i8LW15ckmV=LmqofS}r0j(WS0dD(-y{OmQXHm31I&QZO
z*oq5yk5LbWjJY<U?5qQZjw&05u~f6Ib(Q=(=9?n$edbm8nNeZbsu&X#Mq^>XF$Vf!
zP_5YZ9i?n2f9vl+{V)6=+Q0n^7(e?P7(f3TzP=HC6brkaM9lr1GF7Oc=8`R2=JTAz
zuU>)J^BjdRbzpfBuCIxw>rR<;@46u`hi(ek<xIs5X20|)BIh}a8bXU)i_g3QM9VK9
zjPdh6<2oKGq}O3BmH@0^g<3NoBK(ufvo91L6>s=gUd>$s#1H56^IjjN6w0<^+X^l;
z!lm9b94Sd^L3uU1DL+D3FTyC4j0k%x$Ic5Ln0v|(p?V($FMqp+tGz}5$<1}ITA4$d
z0$oIM?6J_(iR4!eq_XVC*P(3J-`$xzEPn>DlzU!8TO5)p0}zKbexKay@)wbX#<^M(
zA@K7#alOZU*11MkQGA`p%Wpa4sc_QI^1=6}F+OiJe7wt?<+Uf$F<btS^D-|3GqnHB
z&tU)W{9(q01F%k?!P_r?8SmbH9WU=+ASP%>2Wr9CckItQzWm7(woT|W=*(?s$BQ@8
zDR}<m4c<Q8P=!uHRX|wE5~URGNfDH~(PADQK?O!XgK*l%9vsTzcY{14q;eLzML@zt
zsw0sJ1%ma)k}P0J_|Y3Pug&|8n4z5ywDZ8WZP0B;-8O80^shtz!@nK)wHKWK@=pSN
z07uJfUng|#%MUEfWg)nS*0r5yV7yw-73t4+{a&6eJaWHZ$FjbbU2(Z@zIWMsJSdm5
z;J}j#c&!gM1;;Et`D(m@>>;@)<LApr6I3pWZUM<3Jk|nc)=`IyEUCPgXAIDmz;{`X
zlv#^BiU`Zk3$gv2*GJYq%AqX+dKF#krn#2JT_;$?CpgWf>r#tb0Mp_ysLqt|`S%rN
zE(#`Jz6oSnCRfBOmuErNH-EKqev3=lm<2_rOinZ&T0egFi<o5(UHDq1@P9>R0VAyO
z<=^G+Mt+&&E{-q}Gx%!V)`+oJf^|7+DYzHG6f2Y8xtdScd%Zf-sZ^IqyeNSDJ#Hj3
zLSkJfTP&7kqyZZGFMfj&8)dHH<9<ikH+;|U{sBC{zvG)<|0e3T<M~UUVt;<ZwgYcA
z!PXmEA9(+Gpp=4_mj_gKeEQ`t;qCJcI+~-|5qXiQ+s;ho!TbPGuaP8@xe&lijAS7`
zctRbW;k|WeQCRC3(R$X~#|8i}^G5JdqWio9fcJeEu*QyVs){~_S25<CjM0E0Xw7{a
z-*ba}?;FU!@V5g1+W+Ry!v6f%VZZo|b>-LDz5H}7V{Xgd@O7})`)ZXFpL4zCwZ_C;
z{asToOu@H+82@!`7{qO~MEKzK-=;TV%IryLQXo?yUdKDf60}sx_}Auf0N&W9k<ARu
z>a~|ilJx=oM}iA1=OsoUfcJ@r2k_@CbZEb}qJolp4eY+v>#r*M$dwoen9sPN3|Z%C
zMigBX13#RaO=DMyc<uU#mJzpxG{1M<B(`YltJql0!kqlq!VNLaT)e9`b%hnBYF`}}
z^Xnv)7M9L(g>)o?2x|0^i!{b^y;Y#UQY?(nm<J{e$k?}G3CR_aTdY@5a~}v{EOQqh
zS)(21;peXr1;+oV%B3?^(ozNwuZ+z*@BdmX<n}eB!}Rr`k9nC2ctvZ*OyKTuH73|@
zHyodR6aD>(4%R}dPaA&g@BAH5Ehq*&RfFO{DI3P)z^8AXp|#@q`4b!ssQc~;{_t8z
z?4D-5LA9`cRnRK2594V>*!3;q!t-;5j$e&Aauq1<R~Q3QN}|OvSfgo-?pDY^YYhN%
z79w<!xo)&vnwJExwLt4eI%ITOE~P;C0<(hcAN?`lue`_jsb53;zx_GbyJOD!Lfy@M
zIFsR0e6BLa5<Xnk+C<zkmU&4?`V(^tLoPW_YrSJKtTPn!HSjaQ9q3aIe5e${k^G9r
zU7k|RSY{;7&vKtwll9({9$)+}>!tI;cojf$9sl*mg_+2yeC69SmqkLlYdtO{f;xb-
zCPIB&o$lZI^;c0PRda*Po6m<67i20+iEHu2Fw7Fz`GD83D;8e^{Vk3H?A85vU7nM;
z*!Y~ug)BJ1_yu#+$LERU5+E{7F(SjzW}Olj_^TIWnrE*kX?;U7!DU3EOMFS9Os|LO
zdR{<o+Bjqq2ap>>RW8qJ1B)qaxgHT-JNlx^7WyflYl^aE{;m$^_|%2Z71AI;M_LUs
zBa&iX#+Y%}VePC@OR-c2^W23zehD0=FXS>rpJC4Vc^(+D<CKcJzrov26_4Y<I1lV^
z-r%_?bTpPUvW~qJpfxlT)P45`?*fBjf7-nkeurv>pe1c-MW>@RtXc8U&{0`vO$1gd
z3>^h5w^3Cw3@D}YnN@+_S&v2pRRq1AV4PFQ(i(kq$mlqaJ75C}Lu);iy^VnaKnlZP
zOHn+|6UT9K&+P--(~Sz;K?#F20Q|PM(C>eX`g^_y$N%*I3mgq{j&+M5;4Z7wu4n8j
zAOYvU^Up4<U1nG+%@CX^PL~q;kukJToc^CF{9f%He@9;3d&=Wgv0<WUTgDNW4q%t(
z$mRw`gjWkDyYZS9MP}tEz<Vvw<GL5^q6rixE$8Rhu5ye=(B+}M%2!xkp=bevqQDFo
z-?EIt_4QFoVGKIiro#sakPHM6=Om;je9@k^2$+~`uEX3eA_^TcQSpNArx)K-!0k0J
z823be_tM7D57B%l1TO`v;flrsAw^C`2zTBj{2B0Qq<GHjVF|<+4ecX8qSk_lp-yu&
zH|!kS4AtZFTxz2GhPGvX;sRIx*DVc5&VAk&HZL$|&k@snMN0o%&%Ow2r#iHFuWd;S
zDVr-T_gI4wF3SgCscSOV160z}iTCQ62OYo=^(Hogvh8>(6}P7+^xn{WM{6BJ45bKq
zA5b5c=mvwKo5IQtDK{zvuTdm}1B|9Aa5^U_BaTnJZju9IEo_Ej2B3IPaavYXl(ag~
z4O%u>6^!%1d35OZlr{B-4ai^^>gcD3@zU8`O3k*jHU`etP)fm@rzaf8fye#f<D}n1
z3cJh^oNx_P75aO>2lapWcccBge-Y=u@n>Ntdmc}!Z=Qt&fU3}>LfqkRQ*F#L3og85
zuI1{>TK9bR1zLc>2QPW`<EyQEkxlCzF7)wKPM6Sb&{Bx~_!=%v74S+MuAnxqsR}Gn
zfuWcLGpjsr9f~u6Va&T=mIX@2Bu4O|oXIjVtAw+V%XV@9e!JJlmAUZ$U8s(zKAp0E
z6->)_UAxAn%1KYCnCdz~h@>!%yUz(5E~KZRng}B|l^?=IFD;073Sn9>jsX0-wu~Dv
zFU1`1QHg{jFscuJfUX?VkKrx+{JB&CHGx=)IT%<+KhVKzGhF8M%q_jV5%*rtQphlk
z5BnRJXIqqMY@m!fN-J(TWS9HLf%oz2Y!(k+-`=|`#87;6vaXR3Ys6Ph;p;JB?!Am&
z%EK7XN90b>r5^lB%P#{HN&!Yg9}E*^mDCdkg=s~pH(1%BV2=dzeaQm~=lAJL!@~dP
z!V2i<1|x^eghC%vc?PA{M}wUwR1V0~GkSZVzkk8@rc>y;qgr#m23j|4wK^5qQEJ8L
zo!tY)(A$6rn*(jP9qnj1?+@rR=b<$~v@&rI9>5_5s)G7={4U7b9sQ?&1O5N=ONg@C
z2APdqE<0Uv!(BzV&Bc=C<5&dTgy#rf*Hi{8EB;zJ*!AZ;2vb2_<CkKOt5*()LoBZc
z?a${llwJQE=V}pk^QR23u;s(AfX^Cg8|QA-8333`KgYNR^jzs9dp!_jiebpQS4x$b
z>5FNFkMt>U^G8uPj`?#YHwF^_b>ZT<s{Yn=fUH1OTmq&ExhRqNH<02;HHD;1)(s(u
z49*34#KE)>#i33XiNAb49mkRRim1*Q$0No$A7ZM*@cR3RWVZXG99zm-h+QcrpVvSD
zx$!S^>!@HA_C7yTQCK#R=ec&oves`}5AL!SO!S=yXKE(Q=eMQ4`~U=<B_o2a2&KN7
z>zCae2PhwCK9~O+8NXN*)@2U14O%yte~x1<kU~MIqj=Be2sbQfHULzwIMKZyHPJN0
zv*uL*i%7Q^SOcWzSd$JDTz8W#G3Bc0eV~Au*49t%;n6V8!^g&^Jfj_Wyco89hiO3{
z9kp)W@E+-_P%R#T)R|B>8dM7UXy{#0s-kWiYTeMzhTb}=&H|DnO1Ef9<p5EXKm7eD
zf5Z3V6oLKOuL7Sv@>z4<7i!=td}2>TLgxC#&r=?+zv`LYuM~~F`k1_WZ@%mIUqzBT
zjq~>v)hvSkwUAx+;peD<>-}Y+6xZ7>=bse<Tz+7l!+02h2U#paT|^fT+mv-5D2aL3
zIm+dnz(d@nl+XLp2d|GzUh`WUQH!lj8&Z{~NJE&LD>|PK?T08dW1`e>+{p#S=H#X3
zG&f)Ma(j+*2?4{hfFXdE_LkJ3>+ayVG$%d|x&bos?|C3>Rz95jjn^CgXb^bEtNHw@
ztezS7X2|WBmW?e}k`*-o*KACV?^|wci!jx7Tg7!(rM1oxvboB1yvr2`#LFAYvI)<G
z^_iI|nh1r;*|XAhD%5EyNXBw=D6@P>TzAa}NicfbQ1=_UR%q1bDw@|ZKeXbsmWiM@
zT5nYq>ZK@je@<m-h<zj&W9|jI5K5vwu9*+XCOv7hiWvaIXb*@8`gx$ae*j*Qbqqs0
z4n`7ftP0e7gu(J4_c5S#gNUFOI=wag4g!YUo(Jl-)4$=CQ7IK=+nqi;vF$qrgpY4@
z(qPOz>P&*$3vU19kHNm)as10a3H-&cFB@d?^_R_aIV&?b0<S-}E*>T@oTBe@l*fv~
zfyFPm00fzz5u26<b&|^xS<m1rshaP-?qn<<e)e;pYR-ybmh{OvS?`(FfCIDmTLhlC
z<$dX|kRaMF%J@o^rP#>p{MEcZvY!Gs`RahYK=*jLK*+JZ8epJE%?X*$5mig4|G|ZU
zy^b3QipT{;I>19{O|6+up%5-%z9n}jA8x#5^xP@{B0-4{fza3@OJqY_RwUcx2>{cC
z?hr4tn|mb2p~|@Am7CQm7^gKp=YMX5=`Rp0iY~7|!%~TEb(y>C`q8otf6uJV^g+SC
zoGXxqxi0hd13u6B99hB~2?o|a79miB_a?13SAV_)x>Z!S6eKc#y&rSrh|E@tAea#a
ztYVDQ3Gj+icg&hkp@;^i)#zt=3|OE#gIAw4h`QAz(}G}rOvCxO!&GP;v;!Ceb$`lo
zFcD5_3aC^7jKT{(AQY5%;XX!PjqKtV)MF_Hs1;gwu8Rr)r7xj1^s@oQP?VY6HU_L4
ztnMJZX|Q&p$qD`PGq!*H$8i3!KMnh(&vGr-bDu)pVr@czO@Ocp%-mqG+>0}`+y&nU
z;n}g&D<H5rmdp8-ODT+%hrf(twR$gjZP{NbLPGYeW0K29=D}b8^7=cU_ntpb@>NT*
z%g;MZuYlhSfsfI@qK6UkX#v>Z;q}oWsUdXCXzrHcg(gDt`@mjD15alug?=C)i+KF`
zl=B~N5172>b@i}1uF@g7+zXJ)ri=TeGugSh)V5iqO_q6fD_~0Sy!RY))L;o^6ZeR>
zFITt<8JW*fOkjwT&x^tmT$+@`a|LQx*Cy6twQ|<IFx|(_wZBFj%Q||qUY{CCT4;VS
z=G>5VFJGC_>$Z+{nrjt%BA!vRLDBl180Twn5sCwz;a&s~W9Vn>QGPudnj0{9=m<V4
z1kVT`QXjLFv1V2|6$2xGT%p*ynhKMddlFR>tv9E11pT<f+JU2Y)WRA`Llp4Jy87x&
zh$p}C@8p@)LsrM2!&?RDi?s>^YONRsY^9Pa(CQT4KoMXR!+AXT^BBM|Dh2_}dtXC}
zTYUrl{!P^Hdq(|d{wCO8{wCTV`zegi9`hVV*zvNOx&8s-Z24Qf7SIzi!T^sTigUEc
z<cWq}=gI<b+A{WgDu}rLDmj)jb&;-qGnE+;{aS~8pxj>X|M~ZM<!e>akU{f^tcnu|
zDn7t#P~5P{Z*sqc&AUv3=C#_IzuT)m-rwo<QA&YD83i^j%y<sMF47%5%%Q;+0S>^0
ziyykRl;(1QSm*#96I$jcQc$ch!zSw@En~~^X*yq5zXTA1L_QQg@eDrGSEb;K>4#ud
z0p?bOF4uv7HxuwA=7K_47Sv$~VG2aO7{nM~za+-L{(b$Nd7@7|UmPNHM>~+H$lz3t
zA)>g}d09q>EX%Rp$(7+6ei{~m2-2~iARxw)^dT-Qqxq>t*~Zhw{Dk6swKj27dAsRL
z;d8k~kxP>+hd6Yu=d%Fhm5`$oy$D!U6(f5@pU0h<&*!;-8$fGK1)_d$S@s4nt&FrP
zjkyP{ooJoqE@GRH(XnmS=SI<oq53?OQcwUoqSa8h8!Xy@0%*On`s)G?WEn$056JTk
z`XgUK`N#fx*q`_r*w-IZ5MMO-`t$t7{JLq~E?{Ivmq{w}*T5o0L*C?iUXuo1<jo>2
z`RPSw_%@%wEB+KU&91k<=z~{(dUd*9{jyF_2=aWGsXQ+<H_;e!uV0>erR0|RU4Y!Y
z<CIBjfbJK0eYB%lMM;IkfZ|rMiI>B*elLG=L6D-ri_NW?+%k1Iz8w<a19yF1Yqp&9
zO%`8S1n!)y_1XZ^{R?iQDXT7fF!^mPfA_`Lr8lar3!Vaa(e94Y!(koDFU-6s446na
z%NF~=c~w(s9S}dWZ1zQZW)ZPO>M;-VAy@p9$~Fb<^&r1SE1$dNf=0||T>kwW#b%+n
zUd?|9s|#%q10EjX<Wx*?v>uII(t3-Jk8uf$O(p(0N-u|DQ~<E%6o<FLv<|d}0Yf|3
zp>2%r0`&nI#Y=hv>NyIc1y#ih42vhN3B$d`{aK6CJe<m{H|8(&j@Fv{y%=#4HJwT=
ztSao9Gm{#%_%%<47l%ayW7!L^l=z%kfzYG?`UAi~{r%qy`EULhj(_PVq3v>FbB<;L
z<y;=*+WFoJ<?nlJ?nj%#^p#>*UotnvRj`LtO=}^pL9jIe$Ojjk%HwlyD2c>BZ^vFP
zv%H&k3!u2}$Mp)A-{sX67BL^-;XBCoPy{hzPKSU=G~fC4(LI`TQJ10T-`5S1*@==L
zd_C=wC}~ch!9FPBc+Cd9=15$Ac#RTY3%y*F`JIbZup-yxmgJ*-Bg?BrQ3~y;i*1p^
z^T=F$EKw6`8Gl#<z@k_z{<hrD{2s3au`P#S0u)0;UR)L3su1EwS*|nh6@Sl?;6Mtj
z%g3^|eT-p2gRd?(U#|}7WoqT}9I3GL5#(~+nOKKZ9#_3Bxjt(?N8BrcUHoNH%rIl7
z<Y;slTZ35l4Cj+=Ki#ho;ifF~4*FUgOlWbc2Hv*OtqfWMz87E3VMseV$>UnwlQsrM
zr+*-ntciO51c68QtY~Lqp>+d{NpVGruPZP*m0c;C$%Yh$d~Vo)X-E0tFXQQ-{u^=r
z@t=Zy)~<Ur_GYdTUOms8<aK{cu`UoD0Ltav^|g`_*U<9ioEQ*g@N+bR<$e~sB9%#q
z;LG;^Krsoj#J_LWDg5ITg4f^r)+b<sN6twUQFIxSAb6P}`S2>&D=tcQnd1)z`fXkx
zX(>!gV49g15j=lzWSNsBM1Zs5$)p0-@A8^s-UGBI7&13$mk*xM1zY;4mceB7Q<|^y
z^Ve%kAthKhT}-qsTWLYatJ#W4nV8R)87_(}GO0t*tx6$;_naGfwOd!t^3^sr<Q(O0
z_di@LS-H;r=S%@F@XFi|ls}uoEH7^*I2DKFyj^r!(s;A#=4&eKA}l0${pDIQ9>jck
zwOk?k9*f0#d6{9QN5-(M{2r(6su-NdjWQGcfc3*2#SNpMUb32V70M{6AB_+|U(^_@
zc@ewN+v*8z;ZKQY34}U&i%pa~ew17f!30@j``rMOKy1GnDrFYs!{UD^8!!yRlij2g
z(&_!c80zb8DAjwQK%6$CMJB)Z4YnWuQH-Dc8rr}Ai-89yh?h2+aYjVe=(wpAK0xjl
zL70MO`QxkQjQ^%}E<jS5uSNV-$bYU+PCk$bIfhAh$Gz>kuAe)?<s92w&wRR9wb9ad
zk-8$P9?$Vz7!Tw&!K^%dRR~vc`0domeEn6FxmcXHDP7G`!ZcN`v;&f3UIjr6R{R4o
z;VSC=6L5*rA=J^&RSHuh+FUS%AP<53E%bN@<9J5IW3AbdPBP>gUY*C6tA?_$P#%_>
zgNiE<%G{J|zCxg{7|8TxEQbqIStZSs^Acz#j*Y#ZEd=6R-wUCia~(~GqJ`x;VF>DC
zZq%SGuFtawbBocKSGv{wOeo>?dy%{t&zf_%SUTbgCZfTnFui7E+cGY@jM37Hi^`_4
zHa46l8Xf4(>kqL8Q5@J(3gRF}aXN^j_%Rm0k*306pqznGQR)d~3brSaoR?TbgZD-s
zj5aC}f#hMgHkjZxB}$-8f6h?Q=Fv)C<+Yy&&hv!Uin3LRb@YBhMbUeQ%7$UeP*}jN
z9sqs7e)v<!k9;r6r#sre_2(QEI0d>umiW&s{}B6UU6&7b=VJL9E}MS++wzR@iBt%2
zw&I8xazC&4zE~{Z3QXiZ*Ku5x*1GrCXG{V9s?q`fX3KZtS>^KgNUUA=?JBS?W$3+6
z%c52Ms=T*-!Ib8YwiI9s8uY;l!b~$nNR_}hqeLkOA7p$#zMIM;G`_#THno6dsOk)B
zT%6sxFo6%^_kb{1P{1OfmYDn1w9PN&FsE*w#t~7QS5iiXyJ9u@Rsp?OO}K3FoSW;}
za78PUh++!m822lM;(wb7b@|q;J0+6N&G_2~%E|JUm%pz`dx1n!e|b%yxo@w2qC_ss
znPPb56c(4Rd+>3@w)P80C~crl6F`6aCx9xtUu$|hmP2XY<qj0Us&$Sj_af$A8r>bn
z9Xd`}&_<xpHrb%^<a|G^2~nbHLT5$|6`?6&K$cw5GM}3zK(0j+8R->CK~sSs!dBDE
zD;Qu?AOle94roP(7b`RXGGJadMpc-1*Xi@HZ9_W`^mf2T^FHJ1;jIJXJRw>z&f!UY
zVEC#G2pS+K^uP4mP}R_X;@2_$;%D$;3(G#1c^XpqUi8A;Bfjxy4E4_y#45;!TNf$2
zB!aH_FA>iaM$-KJRT&`9(^MS%jtldPm3bZ0`p^2a#rdT^&U;MVcvTjYGR}o}8SfM3
zGDa@rU2eK6C7V((&SjMF_)f2nj6Tvl$cX@BfU*j$X&sP4nKg)JmfzY308;=0DcVDb
zYc^YVue=``v*9sn6YVko0W+ge<uchB4lF<>N6jJlwI)sRO1k(5ug0kj1C)XwHd5jQ
zi*2x?ny3x3(pndlH5HqAttLnI8L2Y$*0PT*=Vy+a2QB?<i$aa@O{FQgOvb8f{D8&0
zEGj7x8YEg#`uC%eQd}ET+04#-%Wc;4CD*ZBo++Q&AfhP0=X)qbU+sB|`IH!2SW=Q!
z)r^(r;Kt}M3@>ZK%v=)zN`-lHk(J%oP+_7_47SMZ9k2$m230QvD2B+ji2Eh7u&^L!
z-*7f~muw@;J>oT*JPFT0Wm0Ozrc?+?JqbfzxfP5#azVekP&GKkriGc)Mk|EpE5aTP
z?K~lc+0!lPs1n81sQgp}<?s9fl)vo<(f_rdNB_xR&y3`p|CD+Y_)JS6exIMo#K+uw
zb4AB+OA`S~rcB7B)kJf^s~SA>?=ixw{#ridyE{Mfts?$D=e_m5DYN4JS70eVmgp*0
z-Os!GsA3++h}4y<43fOYRaGQ)j_ad-yVpn7K3dHE&|H}f?5I4%Swd<_Aw;A+YmI)M
z09w}t3!+f3GMVBfQt9)dDO@uGVx<)PZ?RW6JWVt)5xiaQIjNph7jrUv_*t5E0N#OR
zK_!dOwYiz6nd%fpZ0eaZei^rkxiXdfKA#~rF0VcDgDi(3aPy$DQ%RwYWew6&iVbb+
z!F9y798Lo=IWH+YOi&Y(9X&Im`eoc~*tl30Ih5Qh5tcW%{|rj9i*+&nU%v?bhkn%6
zMBK;O@Ku>CeD$<O=7L+4O|~hB%?zf+owe@Q@@(N@QLxUlgFyu>k$|KC5x#g1g1DTR
zs56?(fIA1`g$DhJsnfmEY$;BTtSA@oR-~BLvyq-N1dyUq>h1IiT&0IQz3CNCkOBqF
z#2x4iMei(}gjzQYoBj?N4bTnx5C0xUf&J94=c2woGcZZTw!pAg66-bHxxS`#Cn6{|
zeI~C3I&udA33~DtHi1VfBp*f`AOF+mC4Q}R_4RfAJ;KZ{e_tu1oQwIH>1yEq5FeOb
zA1(Js-si#?Tq)bn8{4;feRPlJ;BV*8F*&Xk6J-Me_4bCg$SBX$#3FRUe2+TwA?O21
z!8=?@NsC~eniQ@f_%TZwo1>$oihRO(9&~E(ob^Q@10UQg%U1aBW`@Wy<-9DyW)YT`
z4IA|@;%o+rT@a(2Sx5?ORz;t?BM?Avhxz+Rg~w4YbCb&Pl}461Km5oP@_^-YWmqV%
ztkJqj43~{~K`Ax`mp2KClHuZOCDCvuG<cS}z;b}%(j0pqfb{OY9dc!Swid+~@3=y^
z93;a<8RS|4x#yv_%)67O!@%&=0YwLdmYs?~tWFBT@26TIg^2IkYj;Nc(CK^)9~7sJ
z#1(2OOMyS}Xq^Z}MIJSX2wFQ)wak!6crzAX^Xa_Q%>0H{l){`B8!R1ON<r()T-UPu
zX9vu!BYC@_{u4ip_OEo<FTY3smG^$U<$`3_MZAcq^=!^%Th@F1p!j;>y5Hi1GWWr1
z8Q7|SmNiI)m{xRLYP_;q0zkyM`B2DD1r+0(K{6pT;wk2RrlL+uXL+nuajwVz8bmP9
z8t0JXOCb4y0{i^e$F<|sI-Qw8L*#)I7&L3ZC9F$xkfRO9+wyqRbb`!{x^MyW;V<*#
ze~X<Yxft2CZK>Fq8*Re)xxf;!@vhY+7G)~9sivo}Td5r^1;%2OHpWCnIW^)9Lx7B~
zf_U*Zk4mx;?2`9)h&f*7!3?wfAAg_J*!3W|`1&05wOU#6eDRV+;0wZ!GdI_?On~5d
zSBGNkTK+vFd|XIcF6_XtC@|{6I#=0IQZB-RJadl^1_*hcLOgL$lW{mGOqV_M`e!q&
zv{71+g1MJc_!y$~XtG=a#G;8Sze!I{!D1d|q)SQwsZfPkpO%AQ0-Xd1?msY=><-Z~
zT-n4E(W*B0a}+3m(NE?x^o~@Wi-4TUzR-VQlU5Tcp4%X(-t?#20WF);!tP{;0aCF2
z<3ED&%Xiqn`m?ZK{LF>!+*=7WaNTnkU+k+bc~uCOiDoHyqXA2PWP&gXR9U`12Qv55
zdKO%kygD#fg%Eoogt6sX#ksi9O%Z*BNsmpyvVKo5>lhJmvlmTx*_L^ebshphtowVm
zgZ(0}k8?CHd~_a^I3$=fOO90H!MvY9c)xSf=R$;)FqVQ;7av1t*K{aRai|JlW#Ka7
z+JU^MAc?HpS#x}9>}=?hRqdvU>Vv=<H_XCkdeQ8(o=T!RIPh>??)%4r;$BdQD=N2K
zG%X3s_LN-rm_IMgjGz*zY;8e*Q-K-BFeI*mu;{M18;NrAc|&Q)tNSmyF_ce)NLna@
zRajmQ`x45$gfNEAxG3Tr^71?v54b<dqv819{v6~V`Tcp{ENmBjJ(7}n{m7OIemH;2
z2qb=AB}{m)FVuT37x4h;A`?2WG2Hi(NtsiL2AX>v31t>PyIOQhI+E}BUVPWjfa;=_
z=i0K=6a5YCXlUmHw|ytV7qBmb@4f6v?hEcR$E(dwg^E%k#vTwNp6NVJ?#;3#c(B$W
z(lN9HzimhTZ~cCp|KfiP``QCBX)$7pYlkp=E&W${`+DU9WG?Ul9vL+SqXVXJ*iWf)
zImd7x#WXoI|D6i+!+o=k<*E>tF=3vMI7_P%hOO;0ZC=jLM?_!-c*0Tubk5tP3@<~9
zQ5bVj0IvNPd3_{(B#=zeUk8?=Ay2np3gkeMl!ioP*})BfM2}&oIr3p$D619sC)hlY
z5CFvsx0yS>T>NJCrnnHI+6u&wd@JN4h|G*-ZeUqeQ~YxjU@o8~w7!oc1VG3x@hsF7
zsFDIeH5%3c*}E5-i)|h<7@I9aBomMf?*TCF1Q9P)tpG%)v1UG}03`ebxtdeK<c~lq
zri(H_h_~xHUt-N|CU30*KF&1fF=k9=S?aPL+9@((Js`HAe#t_TAxOi)KYs1K-yhs#
zxfXLSvWW>6ft^o2QSoGmJYmm(F&cX*_|46hT18qZO8N;1dH6OOYdICCJpC+$LpVB;
zF^rY#&NU`j8m%LJA}*D*5O8Qwh95?Ar-e2X7@PNaT)m8j(F{ct?QEzX4ojYTc&E%j
zmf`@Sd&4ALtOXRe6qS}xZzl}m<Ej7#%er{$Q29#5_Rs!7wEx*p!+!N8H*w0n*c0Ya
zSj#JTbohGAmnHSMJTU-;bv6fD!{*_Aw37l%i?^rvPL&hF*J+oz`(ytYro0G?coF8y
zZdwG@E1Y#zFS6`=`t7(kXSg(j05Cr>&)jmAoEwX0a;N<0zP(wi?u)!W&QdTR?3_jZ
zFPqBvLmL1!??~nytBS%9rm=rMCM}Q?oPvU3^;=MvxZ`o5r!MtctfbJk_%7gM2&W-8
zp{{5(1%4o-e3He)TcoL^2+1#)EzYkHf8LANe4|aGXkE-HATy!OTtN8sC4{ekcX)&o
z5e8Cmh6J3M#~F;5XqeB)=Sg6+I$u3^fR4&3D08IC&-1@AjjrUE?DO?mZQU<151z@p
zqNf~}3oYv|y7T5!K_IUN;8)*c{Mp|Ge)xNRiy4Aiob%*~S)p)~&Jqee)Hzpib&DoJ
z(U4}Y2W8w~l#UILKBB2V%rMX)X4p!dmy5k6;sVIWhPW;=;)LPfY=9FbJQxnENV`0I
zXStub_gd>HrNA0XV5qWH=jd$;Jolzf{{lQ259sC!ER>^+%h^yZkj=BKnP9=73=6Aq
zFcbRA1@*^&2<N~0bHK0M5e|AToCB2wK_j1H?)gxnlLA?3-DSN3%=j;J-vn^+J(WCb
zpdm|T#X0xtx{J4YEPya85-gX9qJ}~4tPd1q!0akGpOpH3iHcj#v*pEs1SA87$(E#&
zjVYBWf<vg8g@tfcNZ;!9(N$6xcZrW=(lOIiS{KJsnTP<<=sgxSb%=X|)h9F*0{5gs
zQRXICbUY`AmIvv#^=}Fk_ZQIP?N$X?Ltnx~lz1Zk=2S=?yyY;<B@597`685QW%#p=
zK!|<~8;QPnAIX2yDzW&_d<7b0oPTVBo*Q9+7LJrP+*IaMS@cn!s|X1p>{*j8a4u&2
zauC7^cjYTX*;s}l2Ev)=VmTnbc4iq@cU6-9>?{9n#tvn;I920ve0msw-+Y1nrLO}&
z{5}5NDGw}G#-<YCFqwytRtG-3{}{|B7D#@U%b^%;kfS$7;0U68xa6KU14~Mjk^vW`
zxN^eu)gTbA?>ZA+dwAIgBGd3YMX(2F&`i*MPsja7pmh;%iOR7R1$y*&Muh2l?`Ef{
z2HZz6yq`nRygo1L0VJZ8EjX=X^uas{i*QvhqKE<f;4|t!^<!xN<DUS&!G+0vzHD`Q
z{ga=a*z*yoH<iJ%r4fY+r^1eFg@xU7{4)2SpY5Qy0|<%6)>^`oLm>!jM@?Xf*`GSB
zHfI)6A$S%$M-*r;p4F9lewT-Vw7K&9LKfj3iexIls|=d}DuK4|*)Q_?NS{J1R=g3>
zJ9atUW8rbBWl6`VmpINfxA^>+J%9tz`9o?(IQc`T<Qy0lj%9>qnbe0<4->sBtia1b
zFLip>XPR1p6wI>d@re9ou4f$lMI9nc;xu7H&`dxx=gY@y7R}h{d$G)dc2hX-hfG5A
z6@d8|xUs|B4`t-Sd>wQk6f@YELS8S?Q*_|2C@h{Yzq_FBfGP81D~I|y5o5!d)a5%f
zUjPs#O_BmFavsV$Cw^ER#M)o_I`IC0yxDzvY_2n5V}gj$8%8_P8)>1}uXwi6BiT(e
zn7WigFa*k$!pZYwXCJQ+<(nCLccEKI^YuPZs~5-f{Y}N>R-h%>nTk}jA|kXBfH<fM
z6uquJjFgv}H*fMrL8Bw-$@Bq%l*}-JX4xTig*>kxuZ{uH>VCYbKslJgJd4rL%~0H0
zmhY*k|Los@_CNm_$S=JB;+fpbe$0K~Kr+q-r(t+uLAR0&NV?f3r3B36z}RY>n~vT(
zg2K}bNnhqX_Oh7MPH<Tf5osEuTa-kM)2E`i)nI}ora~S?j)e#UP|hcIz$f7cGe1x9
z{L@OJ&YADWxlWJ?U@bHniHsuOh2*u>Hd#IE=6~a2EfP`$5H>KE{5>XPeoj;{y2R4}
z;?XQ-*KpF9z%`y)qRc@2!}rG+P+0m8QeZ{)jyYllkj_g-a^{g1G0x=DE$$y^UEKd8
zUI6WiHpyv@$(Q=#(VBSe#73IpsZ?aOzuY<O&+W@i>wun>Y@Nv=Bs`BH<<CSeNJTt6
z2~bo~;GB92pVg^%?M0hnVpX14HgLiu<bAKoVWPk=gUv&c(s?g%D#+K|qj%<d`96Wd
z{?so){;uB*ys3Z-<Pd(4D9JHkt)aCCdOKa=#9neC0d_jOeBV>oDcEZUMkqsiNSi1D
zQ4fc7;q^$zDZ;6o^@u_PCUO*yUX$q{hbOYRfL&>`5O|zd$ukp%*f91_*tUAz7fX~*
zL|PU1eS`ot1Io5xjDcX^FG+}_fytWUO4HDf2XxyZg$fMTvvn^Fkni15{?Q*s`{O?Y
z{FR3bO-KL3a52Kyyc`h7rC%u4amgh~E5QL}0GC+DumVy@>s5xOe2TxvEL&LqI@cya
z*$m5%Mq4SC6KuxyyfSTE+pwIX=+*~U)RY4fS@&2JQ?9`>S*hS-A0aUDTyyXoeUaBk
zDup#7Lj7OonqyNBxImny%49Sa!LiigAYxh(?C@T;1EK|`l%*+;EoP+pwu!LKh~{yV
zc0_m}KVtz$YznU60CT&Ca>G@S+G|=QG<Cc;z_y4nAAG13W@|J`L_zrzW4YYryGiGG
zQdK%}eKBSa1YT#RNG!4TIl=KLD}O7?2fR`yz&sLJ6tS55x?VX0qkd7+XTGi>couvx
z@8h5XVR0azFFqZ;E^L_;vHb9YZk-|l_gjI%zWE^PjWI5k$gmD;EMzwtD}O3Lr|kt;
zfO~3mCak$2kfM-UA#exs7>waJuwQ}ahro;#REME%`v)ECLJ=1&2{=LtstB6FQM4mI
zSrssrnc^G04^~Qz`HfE)_gZtG0ugneQAZZxtUB+rpwZT%<>#{Q9(9<|SiJwV=PdN|
z0X2@tWQJhIV(Sk)VgIN90FM9BPe8uvQ87-7qz{k3a1Fw$vBe64r3INteQ>Vg%nV<X
z@GBKnHXd4N!q|TkOj)jzC8OZ*FOv_Q7KB%KSVGX(GA%^@uF%3!Jdq_r;iO`SX}1V+
zF^thLglFXgfQ1FrrceshWlP8|FpTru`**?791%G*k<TUOy5s|fFug9~6nr!RXQ<w~
zx%2zOSmS+ka=N3yJXwc-_XAVvoD3|eQ#4T<yi=8-V~uEnKZixsl7J-(JfxIo`_%#l
za?Yi64QwHZ<56N`#eV>FOfBybAmI}U9|i`6#TblNXOf=_-9$P*SFUgdgyDjTcH<z%
z_#@H;z?#t<^8j+hp>1R^5i1l<n6H}t7nk%h4+W2}KeG<!-Zc-2g(S0CP%ig_3(|>Z
zA}X!BUg5zMU9T~%sYFTWP$@Y7*q=uE@Bi%>FpT7VDGW#N813NUFu#{LBv|qT1Oio8
zEUt`f(88#tx-n}Q{RG+z?0kW>6H$aum1VMf?|v^*rt0;B64{T2QZ_yZp*=(erBodL
z7N@Xy=6on>O`RXqfO%sdfm?navO36(@ErN~dq4fY5e(7M+7z5)0DTyW^WwxMT(gmY
z5Ft?^7^J1d_hLKHbe8s*wN1Ex<SQGtfBFxg{qdj0_{I0RuLRQ%0}Xm?3O-5e-uJB*
zM1D|2Td9Mn01IY4yXiR;NeVHJ>KfpX?qUTHJ?KS!FNE+G&xvt`9v~)Yy$~7TyjkzL
zK5>RGR!V|I_o4W8)TuX8b4ZITWR}_dY_3Q5R<ZE6#oGd#%|7wp;q`H4E-dN_3C3`q
z4&GE;*(5xYgf4Nj3+fF1HcdbHtR{MK0bv7xCvA+5KE^}>2xRJ;B=^mgyAD1x;MGV)
zGzDm6NsmYn8Rj8(6EP6oD@z?Lf-;|P3>0sD7+j-yiQ}Y@!i*MyjYU{FDlmY^bPNpZ
zQ1c95^_^<u&;t#u(Fen>HDU7a;*-xi#g<EJBQ~?ZEx=HYA#4+&MSo_6z%91CKd#-?
z^1og}Iw8<{UH6-~lUvo3-j>yy1f8T14L{2oThIUJ+=v+p6l%j20stYOZ!Vy%_Z){f
zT@(J7EV>9ED;yCf@HVvbpu&~{YX|>#-hti_^pJUtpx52%#+X!1Y@w{@<)|GZ(8?sy
z-W#BX3LttU*qorOX(W#3vx=z=j5ay(K<iVn6A-}_Mq1x~@mhEEK}UDN^Z|0(ToI(x
zl3yA&pdghwTmVH^Lt+R`^es|g-}eOk*!RHx@@LK}=-i<)UB(DYd6gB2T55AItiA`t
z^MF=w>?=dikW(IIo)0KS<{p%pv*c%f_&Ejy7Nt&H?rB!bU1Wo=NpG}FGbvJ|*k|}p
zy$np29e3rjL{6rUPk;;w2d1KmG1~$jmtzv=?u)!WPAT+`5pnj;#kTW7p`Zma0*%KY
zkdZmT)sB8`@UCvgLM;WDT6X~+PsD$%8!(g&+C6RG;0M4Kq3EBVi5+VhWfwjo%SPgD
zGA?^Dmt%8T0mMpL8zIm`xWogAcxGvlq<#|pi=dV|Es0Q^A&f_xRWoH^z+hcx%()OH
zM&!woXXl0J-0E278{wcs&@6NALqRWCV~)rCyB<ypD;fAhG#{I^dxB6LJfE}C(X&de
zBOzD5=JS!}zd4|Yb&fsl^WyK$;USa+oL)dGdwqWZf8wX1KmG@J$jxA*q4&mc%n*K(
z)6T1f(pvUOl$;JaGa`7@=u;?eqYR%{8{rq}>`_pzNnin%7%s{VYA~RdlFA|WMBR2Y
zGe8R~q3RTZ0ldP8CoXJ9?^!G@EZ3kb$3U7M7DsIXZ*+(^RNcLEoI<Kjf7m5>B(8m(
zvth%Xtkb6#YbPiuTF@O#iYUt8^*b<r=^gCPd=)_j=i28!oB}>*nHX;>DS*DET*C?a
zkY~XCB|dP~!e?COKw|G=%>6dSPhPjRhO;pzrEWKD`xA=oyn^!j;&WjAoc}YBjWC;*
zspou5Rpg4rkieNya#4l@0P12Pj!Ez3CZ0eHi%eXB`nP+1lu~HzKtxnTrzxm<2dWP9
zI_(h#2ZT8|oryC-Liu1}a&=<}C*TP*F3dydhK`Vq;2xu6Ri<&fc@~GKD3Y2<bP%;b
z03Lc1Pk<_kIz}w0O%C4%U^+au-Az#&jzlNwt#zYtisUt~b#EpaNp%f>^bI(cMZ07v
zT4o|rAQ++?a%_ayR1#y;=>TLXp0|#~Vum7`f%`o@A|vdeL~6z?^WF>yVWI26j4HOy
z|E_IGoz8I61rFSEebhr;;=C<xsEZO`@mJymM^qN{ns=Bh0OX`|G#%-^GumizDyt(j
z8^e8tVOb&fKn`6CYB-0D&2!k1wjpX$g}<tnLP}5InZJW!{KcQu(V90=!?1yN9w>Fg
z<e{A!QAOQ$gqy|}FQ3;iE}6#W9CTQ2eKZd#_0Rj+3xQ#MhFWs0J?CIxKpo&j!Nu@J
zTzm{b(nLYzO<ZRFjK(tDb!k8~2K1?5|0jP8=O6zWjGqd{;LmBX6M-Pk)M*@otksgR
zxpp52`JVd?sZ!j1ZQZxSEJGq02{{!}Qn;>2!z#7WF+gTGN{U47C|Xo>7^}~T#TgvT
zHBeQzu12!Vw(Pw*7fYT5qB`_qsfqg&p@{zUVQqynOyABIczq;&G|&goPVVS^zd`DT
zaUN_-7@3Az5c#$lJ2N_>mqK*B8pGi(FYZ>vLfx1lGCV_xL=zM71>zkv7dSLy^x&Ow
z^)|;N3)&6jWD8-D9G_x(>@74y&PRYIgc!{^cN4LXMD4X~>r75ldf9-XG{RyrThNgs
z1<GH!O2@26B|e`~8;3$LDwXoAjf|;YHy!-Z#2a-`NYXS=8;NoU24n>P=@wH53=Lu%
zv`obWcythe;9x`)Y8-P!koXZrAlp^w$9(2cu^^$`UGw{Z!G7VZke~TA;75Lo%L#&o
zDBm9IwL0*<_5%`OiDc;xT4LK290KU*DW6H>2!=Z!YdGISRt=VxfC%#%)ax9T39!r#
z&iycdA30wkSk?Chosl8PXvNgXeK4RSd@vEJ>~rLEz4}iU?F$=*-a6_Ql<F#MQ+*TZ
z+Y&rso+o6a$W+3(jS~f!u<jB>oN8sihEgl`|HhA?>HiP?Pya>&r9A(!KSeRJp4y0m
zm)27DAGf$Kmc9!>L^EYikWe10wH{k-McZ@Z<&}yQ=r;G9j0S7na~C2q%`N`1P^BrK
zkW@9#KtxPAc#0vxT!Dmj>z^CIkjO!?JgY&4g`b6hGc1X*JkPg(eRT4gN<pbS_*yDV
zH|Q=vrwL3)Sttcb$4~F=14m0ND2$Fmk-722DJhLYa}vTly!+14E}5R{yDyyIGj3OM
zgG|Z>e!W<2|E#0n2z#8Kf8>Uh(f#=-)JvhE+`Dg9QJ5psFz?oIK1QYtM7@^D$uAtY
z-&0e0c}vEaH>S;EMp4Bx2t?bwh6OqVQgJz*JkhcN-W-P6-cC_D{~1<NA|Suj6w3k|
z4F(<zEv~pDcHbD*XS&RDMvNinB<9#RKnk{5>j!zTET3yR)FDj1_8#`@pTU0UD`|<U
zdLg$y9kQHGoj?QX5zg*sak^+MTwT!u74}U)KY(^XFi=DwwfML-*M#>Sq+h%cBEk)}
zD0k~{XY!!e#@At$RU$K|M+7WqVg9ew{NNY^I%YEBR45btgo7VK?z#)3kXmb;?;5Q{
z;nwz~`vOdi9-ZUFOAL7siZDEzEkWsvQa@Y0Gt;}20vlj!(_2CL@gGJ1AN)rczx<h>
zA@RTq2R2^tkHwf>&<S7>m6o|p*FBn1j_TH-D~;j&e(W78cg<T*8eGzH5cr&y1rpgn
zx0ZWP#g#TdIiLh$MK<5Jv~m!Xwv@EQqT<7h0E?ePLAmntGWZE(U73VFOy8@ow|GgJ
zY~1u6ULRK_wQSh78%hdPPPU<B=$JUQ6QPP3{-&%=0(2zreklf3lHbe3E1&v6Bx5wM
zX(Rq!coIiPlm#}Am>7Lcd^2~1#Q`yxwP})v#+TU!G?UIS>kkbu%uro8XFR@}YGQ`o
zPS&osLB(L0^-K@|(tsWufA}~e18+dm8j;Lmjpr8$MQgbRxn)+Q9YQmZRZwujsjI-$
zxp0ZQISTaRZiX3B8|Z3U7-3O`MOa)wotPnpm~#XmN)^NJGbt*2WzqTE`>+xHBqZj2
zN!u6ap~K7%l=!p13HckoA9${Dxg5V?CM!klEG8TE%tuJylq+F@pl+4>wKrfij7AID
zFaX^k0fLtKiKz=mtc~Gk1|FqlDQHI<D5ZL6unU<|&^<A1cnOLqz24i2TF6oN-Z)lq
z%bKSl5o69NNPlIt5d>0z@bfUHzoCrz91(^Nn^YKMAIzQGitdP&6&V^#>7Y)b6;6B&
zKb2+#&u9%r1@&+GUOfJfUnAm9Co7u{VGNIa;{J;uhM>eO1@R=E1W4JDO2wZs&OWoQ
z&G!yc5WD2LS`#_H6VP&JR%ZlaTRMdu5m-7>*jvnjAIr6LN;C{|;g~+tS+$q+kRc)`
zc~)i~hL6#Rz=kU!5dvn>?}nq<r<zVpBJX}aQ0fM)TcVcX<#`Q;A;n{uQ?3ArD4D4V
zB{q*xKxWt*FgLm>dDHO7Ltr*JZgIgpoP;$;Z2&Sn5^fgT@>G*+o0P~}%Y@OSnu|un
z;0W1;m_`U-@38bwEdCT1dclxOSRzif4BsVdN<GKznXryt`1Ts5fJ8j25p-F#IQ+5s
z7Ue5>C{{E=!4Tzov-+y94XuQM$nXT5KX+$mZx)yGXAH`M<}TcNjIqXHq9QK&e&{zC
z56_~J=KA~nSD_#R$WQ$m<iGNJU{7TgC4krP5t}V9CruIen=NOF<OpO)eRiV$0RQf+
z%)A2{1k1Yk4lx6)dl_pACT!KqON8Q|f>xI2QVAY;*1zEz7<vzX8c;SWX7z00NRmq%
z)#(NsAq+WxF?RUan8n?rh)b?vMD7qhyRvbg+$^h(DKITUW(R<h`^{i!b(k>^srLu;
z@BDt$)-nF?zlg!|1N@#VqzS;{H34H-La;er__@t5yrPQYx{M4ALG7Hs5V;3nyV6c$
zZcd3UX&Egfe!k<3S_o$dAZ|@!?&YcEVm(}bVG4eP^$xdOU>OL(i*PSc=1eOWFrb87
z#TR;glu{U=IhSn#&7CMgk@yU$7)CoX4AUZaK9O}qL{PW~#Yy2A2~m-3!;%CP_yesS
zJ_(NBMBURfl?yB!#Zw>=Nngb!a?$2~0o+6vNhcknOId-D`%)lwO-LHUk-<VoIA-<o
z9y6m=hR+j-WX#7FiX;#|BSOx_43|`_GonaR01M#7(iTC-D+c6t!^xFEgeWd3CmPO&
zc`NHqN(mi`j+V^~PYqgTqz?qKDv>;t*Ln4tXDr_!APz;<FY|Kw5;(8J;K2iYum4Lw
z0sZIyR(>Q}REB`_-Uk=@EpN^_4J(NLDd0FBZlKW#E87!b9o89v#}=3BXJ7#}p2ReI
zcjrFy0Mv6627p@Wq#tg*%=_E2KIYROJT+j}p_RwjTp>U$Y1dHbbj&)n>v=w7QOt3_
zQs{6Vy(b7Um|OY0N$-&i7yAK@hC>AcQ6)LTaG`@f7oX(OhojdW0m7sR%HQ*Y&`r_)
z?O(-ciUH;lMOlXo?M{gtfJ{6eVEP&$m;o;tMdf)k{wx`d<Z~3^=@0=_yaj2L81YT!
z(LfQ^lrqHkya^b}3y)*E_!Q()O_|5ABxp>^H_j2@N{Rj|GdNRGGIu9@4pG2xR%#d?
z1;(lRE>%hOx-RYzPOYr&R(3>VJ(`|Vn`K>aM8paC_a5H)C7MOFK#kM}6?spk5RF=b
zqh+DIIM@S04`h~YD3y%5K$`0Q%X}||rhti$lAP=egX_srPA6L+)VPikb*Zo}eJ3J{
zjobTho|N|w%B=UuG=-bt9!X$S#G`pPt_&-5rJUShF4BbF^PS6qjkO5#ccOY9Q%YaH
zr8<9JmFqOak@LM=ek!NIVq^*zbYja8XBZ+D;T)(li!HiPoRKfD!V`TjT8}ja%KH=J
zPyZV5pZl#OWBi^f3R91SQ-IXhRHEi++QhO*TyG};w!UQf0T7JT26%XFD37qlG?U|a
zFV<OV7X_x?&?YR#*z??DaG^dOtFaK%r{diOm6^p_X%UG|#ce>=8Dd*e-w$rJy{>H(
zIVo{BuEiSm7<(Y*e*oPn)OGzG1cY;(R+@*NM!<ueP!y=_1afl1{*J$fRNMdgFZtTb
z^uKui&xkQFamur&=M)Vu3lkt<`O^h3aiCMd+ImjhGNGTZ1dewZ5LQ!g1>$GN!Ef(w
z6(b{nV(m0Ct3y5om_V$<$hl1CvjoR5rYwuG#@P(;VDn6v8B4lDfbm^W3g_cM4XR>T
zI??PrVUIhsxPup=p$%rF4N_1e*47iQdM8i0(md86-JA3Db8@L}W@~ZBRrC>XZJB&|
zDgeD$3|wT5T3cSrUl1l8EClk51mJDMwD-~mvH2t7qEldmV4(t|`76SR6Ds$J3W|db
zhO@?8EsASonmQdzFmkoiDk9fv$<aGstN@aFlXWt#LVMc1$dKQ~pJJE_ig%pHXuV2o
zN)L+V?*cHC35dEyd|m?`+H-D0i3M_lMaDbkp^L&ot`xY$x>#A5>pry<^wEGn^Xrhm
z@%w=>&~qIA37rJV1TJ%`#1&TbCN7n{v_`)U5hw;k2Kw;;5TTO1LAP1SHZ*St0W<bI
zmL<fOs8Lf`P%b;qJi2GGbAFQ+B?@wY6w^BQSY(K6DHyqTxo-l=N8+WQqKQgn74#8h
zG*NLjGFs5PVJewn110=)15Pnb<wE7#2^@od3XAh^`yS}o(EqQ0dDe7_{haG3^i|4q
zq7Hz49{NaSnFgu@DJsj5M=qQ*wAklN1tRVz^MDrr1#ri6o|oKl?yGX^B%JT7rlge$
ztQ$A;U&6}IwVOdTJO{%;@`<Dcm&il8gw6tx;<9oA&oA=&IQr;puzds3dd~q<<9A2G
zeJ+xPLvMIpm;xv~fhFAh>O!;A1cy0g%7;@J(Ty*#D7jruh`WOvjm@3}zW9UVf}^t8
zg)7W#ILe5oI}=I70>Kv~qZ>ky>ZFfo5V$am1JN3ypF(d9hcWz4OZBKK*Xj|01Oc?r
z`KMGqJvU1zsTHBQvW+Mq@rZ<&o7}0>0usS=8s<ypM(Xg)ZdY0ee}mt5UNkrCI!9)P
z$Sx025)~==t*KbD)NrmB4-*59mUwtGG!gEfun2wav)P`nOZ>Pz8Ok>vkZ(LN-ZGGZ
zvF_&nfcat7*}2XHC=%Q*CLP%6B&HyA<xqBrHIz!qROS|;uv;=~rB<AM`U`Szxw3$B
z#9BA>xE3YC6z1ka%PprsvakR_7WihOQGbTKs}Hb<=9vh~BhFI%mjWJn->86W05BTD
zFCjo${OPD)P&dz`k4Z?&AhMi8`4j<~Ky-uI0sNic4;7&OTYt`j4od3%IK%47D|vXQ
zKf_n!*xtwEfPx5<-$mqHiAOT}-ipXC8_Y<?nTjuH(m3n$nm#ch8>FQ2hAU+Fm_vqQ
zgmvjk&8a)&XJvu`2Po&`okvd<lx;&P8zf4fI9s8my4(cwoW_8b_1u5!*GE@L?Y%pS
z@N{<HoEDnOO8~&rzafIw4vgNQLbTQf4`r<zq|%9_VzX#hXkjy`R<I~urBXpe>=>j>
zbRmYUEh8f;JvP_g9!}<;pKb0aL`$GEBW*B1flMLdo7l??a5t7*DDuArkmd{5(ULOa
zrqfbz9tY4GsU=}otEiDPdILOeT_WDNxUpyky)4s$%!C_;7L>7WkxQjDBo#WTmbfOJ
zqmJw_OFbablmL<JhPWQGdTb(<MOkq}N7!*#huh^a*_fe_A)O@98$b&MlFEO0c?u>@
z`{(1HK*+_|p#IO~^T*Hq2K48C9r}m9-#^H?Vn8vUmkiOWviR@x3q+zHAY(v#_t4b>
z(+Wd{-~>DvfxaG11<7imgJ)vfVVxF4Bzj`d-%z*BdtxxkDSCjrJ%U(_RN1*^m_UHC
zkNe#MEKSQV6&e=4pBf|)>=+EOg~_b(Q_7^ZLSbXMz8UK!c>-gMYo(cT2QSupWah$a
zo1qB%NpH{S#{=8n{R04i$N%9MP}K6C!+k2Dq~f_I1<k${QmfNHUZAg^A#dlNx$H&b
zdI%=%BMillC~6CkD2@@9`)A=i2&7cZKFb6}P)((weu0oXy`AX&#Cg&Z*lI=HDkARy
z;<+9O!C#<SU7mzrB03)iZ5TAv5>fXZULU0tYAw+11}Pf_Kx+)o^T`!QxiniUqmq2!
z=kZ|Om}1acp;AyQOB?ha;c5&;8~ub0K@mYM#V5DqL$-cm^o~-Czwg2+64=aO*2&!`
z64n_%K8y`F!W17fv17G30G9qWZ+23OZ!mx%a>ghU)^B&g=yPFic9vRL#=!p`LeXFt
zHVg0}oO<Dq5>U{`!44#2(?=ta1(g8IHV-H9<cTFC-DakC9#{zncsRL`n>sdoB>trI
zP2n;9C3Gw@*>6Vkm<yhT&XULp@Y%M7CW`};S-3<~ZA4w8;)2u_$Epz=5hn+$)d4|6
z`#=2x%3uF|khhf?&+yo(F)oRMi`;d@B{Q$VGm=N|XaH)^kfM;fQ$QRB7(+w*0NBrf
zin_TXs9Qy;-d;9AkU3r9Knhf*SsKAVqn`9Nt)nT>C56xA(6bg&zG!*`1vfGxW_~EF
zg+AExA(Eq_^IVaLXhKq5VKw*G<{r;tZkT8}24jA2iN;CG@QOmELd>vlJI>Zn{^sw4
z{$G9pcAS8C;+sd{gmEGg^H3OUvsD5`oz+_5`#X1^xsM~kg70Yyt0EKvD|2BODM*@7
zx#yGD4>L12kq51e35HVEL=b_m*8t_!SQr8W>w?Dgg(5BWiGe5XSROVh?*Pu33l&Ps
z<;fR$ePoYjqWf}t#uy#P%X^G|%n6YKQPN9%AW}qH3hI8#5Yf8rM54mUS5Tlc8**-N
zm+ycLoW9|X<G{<y3j~Iz=Qp_Bo&dL6I{6r1`#{uDsE=ce@W1X67~x3F&{*eSLy@SP
zpTX(Q9XL(ZGz~5OtjIo4W1yCTy>5sig0SZFkg_zh0D|FNtR&;oRl_*vM$YcbexLA-
z#CuUP<I<zR$jPg>Ae_{NO@L%EXFwjA5P1GPz~b2TrBjdXPM+Hn8;6IcB2DXQQOvG%
zul+xc)S?=$jRYB`Y)%&}3SNS5<@ZJu{_~T>&U9{?f9(PM{BNNAq2I=6($2LGkU``a
zR6qnJ1EVoYCHgF1IN53r0u)+qfD-78(Q@u_yukV$3TE*#6B!tT<uH5$5+&Ci&<)*)
z$XS0#AUPVg$b`&H>s5<QL>_dG)1nroNfhFF1{~g$7O_UaTo)TGTQh+IUtwM3-NcLW
z`FSST=}?;?t|hG6Fdq;I-048^fQzLBj(BaTtyHw%GVt_Y|54ok^`G&`Ar`O9uA+hR
zW%@3uq)xYP;{I{2!|K-wYJv)C$FyWH>DhF&CwOw7Rz6o`OS`f`_*KJtp<Eee&|!?$
z$_?w5sVZcv8H^)dU^MfjY=8x+8L4H)u71N&pJGE0;eGUsI+85<`CafS^u~&Ra7UYZ
zK6~%pzjhR81+*yo=xD8@cTDci28vW@B@Z^nz<JzB4-AH$HLp)|91k1^dAw3)+V<Hx
zj>jXTdB!ku^;$5USln1;+o5%L6hCQ;q7wPmV3YI{{d^><4X1QgTy&H$#>trMK3K^T
z6U}m8@K91IGuD4tiNI0H0Kt^zzThml<}`^fYeWDI4`*X`@ahb8CtvR&0&={`l`qdR
zL;0p+FnC;gC=rjxHW6}=?%)+f!jfXhbZVz_(Uqg*Jsfp`p^*tcAqbrtiD>@mNp%g-
zq2`LFs6w~RA7{>QAiq?ZPVbGuFy7uyh<R@a80_EtdF+4ax6xmqlC%ShIr5s(t}&ED
zWaC<95p%6c1x>nXI6@osIkSz6aNmzZVEy6Ekeu)KR;op41*2|2xlt($pg$fM<HWXa
zq%D#z^k^odPewIpwm^}{`2k48^$?t~E?z?3+~IGavp4&dBtYVMS7mK8EmJTR4F{}Y
z@oq!0Mlt!^7%Ko!u|^YlI56mGF+rCQa)B!d_NT7s=)d<X(BJtbjKA=8gm1ux?;occ
zV-H2b<!CMEC~cLM0uIVc#Qk`)uS4!P2OjEmjWE2lL{v-&%frue(%G>V`JQ<$(ljR0
zkK?{UxaG0!t7UHU9D4|z){;uq4^PS#`t@E_P*|CPFYx;4kW{n>&Oz$NJfcH{ygbM5
zkv6{iqjjA90TVvVz(8wfPAJ@gsLDk?+#+bb;j{PeahxYUdGm~?rzZ}$I!8?~jousi
ze_yl`RD}wPRMhPWF!FVVmIq9=`p)D=pkiZTF>#9_oa7O6FG4d=3vjC&vriq>4Qf9`
zM3lmjW>n5hK}|%|ymn5ctEW&cvrGrTD3J(}<Qn*8#Mm2#rwf*+%`m)sT3B$I?Hz~F
ztqK<M>$BjRDuxFMq{#EK1td?qZ_JBLiU`GxoCQ^dRxkAJn=^%$TPs<s2JoBD$FnrW
zEw7(S#TC{}@`IPw*t2-tcpyU;W1RNY2ipJc&qDvcA41tHX}$<{fOEBrV4d0U$of@S
zqKPD)E%*zk1R`n4Ah3<X2c4xbSRk!M@o{vi(7FSqLZre*$8p|)<3agvbd!?8@v!dz
z?g=n0$2k%G43&=Lt<in1K&uE{T}Y3l-2(A)NsR01>m4<lW<s39C67V`W)OocgAsTl
zm4l(k*NN*eT(yG*9+zku8}MonPz>mAdxreZPk_Jlb=KYzq*WLDE2!R;rY-7O=*!%M
z(>>4a=op;=FXDd4uvo%>I0GxxBZor?F~m5#=>Zwlme!ZwUNx+;u-syfBWu~M1yz_+
zSfRuim@)Dks8T+aBB%?UDk2y`>lFgm-cHb6Gy9u=yVpl4g>BnVtKw`uwKS`Oa*~D_
zTJI<j6fJ1|Ksyig)-Yh$Y=>yU{pB5A-oK;2Ai}%+DiJ!TWy2?5`Z6E|TUA$xEX{kn
zJ)`x8$Ni4`abVkaynS<Xr?k4MZ;LSDYb>>il$kS#EY0FR5OE3$0Hm`zXmg>xl{kE3
zF20#3xKdb&&WIg=7Efy@Yv}vedmlK@hTw@k0*-~$9EJ7PI47~$hvmjJci5`?ED%b?
z5g|)Sgn7)LBa$cN?-*W)4FjbVl)8bDV2+9r)`*XxvF9$4nNFpHhzb*!aRZl90o82o
z5tbHTL{(7^A}<35E0j`m_OcnyCCrw>AbmH3wP(~EHJ1Xd6`M#FWsAPa?f>x?P`W|?
z!5>ZKYm3#Bf-Z)chmi-mkChq`WElg55RI}KE_};IOJKZ#z_WNg1QTZb*P~oW88PIr
z_XAaHT0_ISM+)`4Z_GsYeg-^?In=p3lpX0Z4<>m2K(0~G%IUs9m%(=f9=aOl9p<xo
zgwdK~w4Sr{fjQL_5k4`#pl2+=O6oQrw^Kd69aP4;;pxYJA6|a)ufX2Tj%(TIDe4Jj
zRHJw=1vokn<g-K}j*hB<Kio2Qz{EA=SrVBMKuME%a6se$m^V@i_8^o(0zL&^M=%1O
zE8~7ZSkTej?_x$)#-px8meXR9LsOC2{y<u?QTxpXnVZ~iU*Pp|K85XuA%Y<twQLYk
zwBtd+qLZ&L+s#7;2hdKM=Z*@-*;o>xcVAS+XofEIC)84b0vA%1(Hh1MY^^mY4BPfZ
zCm{rn$K4m9pj1|E1pu56^pmmU05DoZ>+FtIbc1YIHanf)SqRQeY8$jd&UQMgR_xmb
z8y%1Pd-8Uw7*bGgJHHbsR~UPJG#17yjsQxDT08^2vuv^mux}fB-C*rZ#n_!nQ>`dl
z^%jr@8$uxhgSEzkG`tI5nRG*`8&p|arjHY?H$?dZ6{7jxc{r;wAfQRl-T|zS@BFLM
z@sIFq3=h>aW0q`MLfrVFEW;!ZXe&Wc1-UKxr%8uc_+A1DhVnL_7jxvrKm#2e)=nJO
zQMWzOX&*t{Q6PvJ`k(v@uv(!1zz=g@1cLR*q0xH+3;01a#q3(;<K)>dzF%nlWUZNa
zu8?%<EWcI&D{^%xY>YkctwFV7D9fG<;ul1F(NY*rt7h35Z`kmLOhA^C007UEiL8mR
z<U+Vlo*Sei#yV~ZE@D!4X^n&e4dpZQB|^{-7jC*BQ}dC8Tc(YXeWAlL=K9iF8(=h{
zwUg0#tpmH@_Fw+%@c7^Vr=+EY=sBt^6k$E32r$Wt3g!*8!sq4za<rOU0SbdYNM5X#
zg)8PI>L);1`YS*`;H182QEE0*A}BHpZx+UA$^>_5O#|&bkp(m*tk3CZV<)-dfYj&S
zXnB*_aCV|B4S^zd?YluA$@UD0LDf*U8{Zr3%PQ^|(^|0aJ30i~I%=ud>ISJBTI;ZO
zKuW=F-yJ<WQtF(H5FuJB_NOP5x+98zoyU=jQ_6;Y>xqz##{;F_uxWJ;t)sUG#yGIm
zois?Kd8;{v1xK=he#L29CfsxzXuT)0vfd#ZBdU7uXouH*QgXjh3K>o-*mlGI#3+=Y
zk)%}Iwj0LQ(a#gdafkKh2YO7tJ_Qqn`&KuU$|fcKY=F1cG|Ze2LpTBrGKz0H&BHdv
zblC58Lv_E<7#+uPqHb`7HKT?m#Y4fOy19SlK<h-ctpQrLq|O+rAr5|`m%~&FaAarm
z<ndX%0NoP3rhxU*!rhhH8WLFR4iv#TI{I-q=dI8|Kg}?PeyT?c0f7DFFGK$B-vN2M
z(>d(I+_DO2l+5NKALhXZAQB|kT{7y1?*~{D7xY-npotD;`9s{3ip~YiWlortYYuWF
zQY-u$$)U$>Q$0-~o_QVNQ*h_*q>FgJT5DEFMbJAVk}w_7`)4(uM>04BK?e&^YQcFn
zREg@e)1Ma&b^uJredyb;5W<&ZhFU5{^J19C3(6n)t*CwA@o)Smz(MJ*I!ma8V_Y~;
z=rZHE4ibr(9D_(CeM;_gMnUOf;PMK?t1yT#WRwcQa-Ttjg%bVDYjNRimde7$fQbiE
z6e{1|9VB`!tLOy-GoZJ$&NIlsq7S*HoJc#s&v5SzK^3dcZhznD^>N8-=9*QEzEV|~
zR)}_B9CVfr0P+G;MQ;yabV#k7FbG<^W1Mu<4d>8`D!S2;sw${;bCk(VuL5YTsH!;6
z6K7+uU9*AB(UbKP=XpojHX^>(C;Bm#x8C(R6#yTI6b*%7L?RK@(DBO>zk98YFuK=k
z5mYH2$<m-uoQEqM37-W8=+pCa9{Mn;!?YC@3=ABPj<YvL-hd^ZOI4^=pC4KV=h3|A
z6*o-8x%V+V1g(>*vjIID%6TFk=YS&=j=4vp`OK~9F5G#|>gBIBov_0jA#o!@XkGD;
zO$#9X5t%4RBsN_5K|J~bE<;2Zk>e)2y0Qp|ZG_#iP?=kB&Nuh&G3_E$tfO;ptwJj9
zB=L{Ie)XGZf9xkve*6zWe#_hZd-y;A__K&hW2YrRlbxl(oyT41NeL<>jT8C;%tDz2
zS|f7o-U*J%p&^{UeRLGzJ`5nU<|GXBa8aiNv!u1KP8VeQ6ap*e-ypdrVaS*vOD)iM
z;53qE16W7Wj-V59hM!0hg$HF7KEw2FKs;1+vBGlZLhAtogJqOdW{H@(f!sT-<N9~}
z0B%3~8#sRI*Zn#J^r+AEGjh7jH}HKJxi185I>D6LRI~t)(K?>M=#9BT09(*>3iQG6
zXkku4gyCk6im!Ws8G;?MEupTW=K@5r?K`cvqTWY)c)-d?KoWBv#Xm_Gspl$81sL=r
z27iu>{_phqD5dbYf5yh;A^<wJZAVqX3B&z(pdANF*_eDG@OtkCJqxrk+F_h02Y7PP
zAx<N?snq*G(cv{=2HJ7Y&8J#W+?;P?ASx2N6^z~<IF4y*u`ZCXamZe<?N2a>3(y7|
z@Z=EjUI+#mhVCJw(PvkK3K$;V>KjTeszY`^D?I3=U4Xhjlg>~@)#8fL6YB>3C)>U|
zmDCuY?dY>Q4Q6II`w5ugJRa!1W8ZG5+Xfq6i>DMu1-U;@Bb*aP(_7#T=hfw(Ao^Zo
zFlkWBq+I;mus$UvnwzK{15uhmQQX<i*@;aHqM|@AUJOz4zy|W|E@;$q?_4@L`ph!n
zZp|3LzRM8$d72B)!hDya>Jed6*~It4f(jum5R9MuO^pBLpGN&>{t$HcB%=}n(uJ@q
zvNg8@ZtBmt@Hl+PLr;n&=3%00TNF(81fZGB40!A5{0`wWx)JDDI8XJ8_;e<riAUh6
zu<Kt5>cc6+Da=7j)2F;>l#C~62`A<ZeMKT{es?A)#vWyF1@SX5e6Iu9M9B!WYFIdk
zKFe+udXdObIxK)FkWyiAOTRxL=s1rTY=6)1MEm)#Lc3%chvPia2LYfs>L+-O)~J5J
zP6v8~Xlf+RL6|(|fiQ7qT^+b(V~*@GSpg41I3EDtKp?*-_qvUwKf}Hny}NL?JQHj2
zzVs7GEUUyaA?m>mw_=cgGxwD&&+2Ck?$@Bz=WG7Obi&6eh2#A@3`hOrRP=IpvxEnn
z`2ro|2y@57M$K3o#taw?6Km}OK%sS@?l%e-GaSbq_xlS4jwAO7`7EX6hQ%1@XJjXG
zqbbO9*UEmW!wg$tA*+5im^At-M#ph~frIueO>Mog!J<<Su~DL}V+aG)lHH>aysK|;
zX_C%$*c}2o;~}JxI0r$J(IJ1^Zy2^Cl21k;DFurI8lCoHMg%!doOdF~N2yuGQT<uQ
z2<xZgJXt%cZX5RfDP!(E;2EP&C$&M?dlo`vR?9SD-D=}^9TB^U8{?iyPM*}~jx8cC
zxMMbig>&nfyWs*ddTqqZ=T>WQsGfUL%>|k}eCzIl+|a$x0W6$r9>L{89m;J#{*$la
z`0xHn)c@iSqWpb70;yq|+uV09j1dSQUUoX1hM0i`VF^UhH1h~3K66~FU#>^1w&8rd
za9`L!ty5?)I)}-YJX1anS3oS^;zJ1ssZh`|9b-B$&1RHL&`%;Pc}~IYJ(lD8US+^h
zf$vFy*$irCc&0EzKmAHc)@0R#5TL#Hv_^xX3nf5B<+RpN@2#qkdh>$3p5xO`;CH`)
z{x#1SzjB9^f_6OcI1Yv!Z_l)P8qZr5?$fYx#i>N@2ZfeN2|tbp?)N*Mo}LL<YzBBR
zNJZSSf9>wrdt)yZ*KLf{btf{yA@5~ugmW0zwA?=l^OFLn<;Cc_OGKQb{;GwRzXvhE
z`&b|?G)w*c#bS>>)~5htCg&Jv=ZWqURBDCRC#dd_al$<2U#n*}mk<Cfz7>w<J~$Dj
zP!KYPThRKz7zesbZ51ZM9mkQ;6e`RTjV9wn@D(s#`)mi=afjS)*f&;SJRdL2mhK%g
z8ioqaW<UeBy5$&%d^!wXdcap|s~eznW~mgYXAox(WmW9^4Mi(jIo+4g$?Ln)P}k&9
zUmuQ26+rXmDt(Y&tW|K^$$PW3aCL>%&`$QP7yz_xj`pc!7@8Lm3A~oTU|LXa6(|(|
zYp0Cfa2^e3^XL<y^|9?Ylx;_3_4ao1(0ZiU<fyrkSu<&(Kojvig+R~Uy}C#x!W|Xe
z(7IDIJDuJ|llFis>QKO?R_vt!<~bhj)GXT#!wiq}!29=~p`RzVjXoSL9)j%j2`!fa
zygShUoj*^c`d|B@JfI9LpirNsh-dgE0t0%*%mmaT92Z8?4&f`XK!+A=FhNgmjXti?
zycno@Zw5<hg9?++5I!PO4@e;phrhR$kwzH_=Lmc_rP`Bnia(D8Hx%SEK`@g9mwTk<
z4h~2p!Aw{sK@&w?A`a8%VDZOH5a1UN1#Apy*SbNs9XdM3`GQgjnCg%JKD_@I{zDj;
zkFh_$!M5L+96LIIy9b*fdtN|EyY@^_oAYCXWu~8QH$2^*&{{(~5BH03FH{G`(X%45
z!|5<qs5|qMLRGf}T#gao#&b(P?sViMfk{{=A_bcil&T)IGnMTu`pM8*Th3>VB7R=-
zWYTrt{`FBxLA3xSx<}F3Hf*3kus^*)k&1D?fR9CmF<eM<fe+LI+_nvMzoF{p#m@?T
z0<F<cUMkukw^$0z;vxg^zNG3FNUfD7wGGr#QMVmh3y$LeVAyZ8K+3kGil8+Yr2V9$
zuvSE^^U>+&Kq#FmE=)slX*wmP6qK_2AsK0vnNtg;iu1G6B#e#0qIerdJ8?el82v<+
zhVUex>Kk(OgPS|TCDa|JW-z0vJPeS+4c|{g*C;hWdaBln&C%a+9*8o)W&>KN+)53Z
z;p1>BEi!te_jauXDw`)xsYlaL_%yE+8-gJCWpnFD{9vMYZ--18N>x$o2D9o8!;NTI
z1lBugy+-8^AP+)9Vhr@rFr31ZLZ25zm<V~$Sz1NYngYV_Bi>tmVEoBn0MH?S|L=8E
ze&n+S6>0tqnal?BggqB*NhkOlt|ZqC^ALQ~(MzSHv44V^j6^la3{r@L=t_8X;zns~
zQDLchEi**bX1se?2Wwn?UQ|Qb2lwvxS;B!uy!azl1Z|7J5lm_-A8?T&n)5o_jk?K>
ziM-=8KqG}#D$4HtD{xTg{lF)9`Xj#^{ZIW06CAZ*-=93P$II~8@K9y?E};d`!kSv=
z3^obWzTbA#x)ZdBV4Ms}0cOBLiDcX{^E`qIN(Jug8uiF>J);p$T97hxQik`K5IW9>
zM<|*xx2rp2q7c$XBH5urYk_s2Ggr9XBCFXcV*O&`{^OKF;e={oMslqc_m|IboDC=s
z)ZI5jt=Q_#`Em-Q^LU^+FKJSsWykjX7TpBL{RP8_lB00l@Xf@pxYN}){kA`4^iw+<
z+T%SdRANA-veu9dY_($dj&QD*iK3;TY&*7XM>`KhBOmK@wr%?k5if;dti{yM6R1M;
zi~?AJh-L5T?t)XaWRefCVT^|JxC5f7b%%`t6hKNvy*;6y0`pkxT39G=-#0wncAQ5c
z6$J|G$Q&PiF!R;mo42?R2A&l?@cjISe5iU+ITvI{@3`GI)Kb&hP*Ez_P$o*Aww*6H
zPMYNVwnGYOxj=lJQ-MLTZO;Ih9e$h-7f>b9(ddm<h-bz^+={Zz51}U%_0yeaJyBRj
zuOJYWrFE7L)iH3M4-~7Yb;GBheg$uz-;$#T_Z>rVB^W6>Gl#;2Z8+^u{sjg*{8xW3
zWUtvEg(Aw8CaAoF?qY~29MOANZCDF|PRE$Y(H5FzoH*_;c)WiHv5xAcxr4gOFwKST
z;h)Vz9C=2fT(hduN8~gf!43y+qfJMvpQDNVE$7nwy!jgdrUek;qor1F!ZQVblmMR~
zpNerUw<As+l1lZt4eB?1FCn1As&aKRgx{~*&FL<{cEk1qZyCaSG6J#p#%M=%D|UM#
zeT@j5C>|Ov7)3MrRK$HVs8D0Ls|qM!mGn|^YV-cMqqWW!q9qdds03onrBZ<)Bg{Gw
zQEj;noC8*D2q=9~uPAZ-(HV6pf&oR@n3zTyO981E)~Emq^R0S2(IRN!QWW#szdkY~
z^$GiJM|9>I{RA^~8_t(^uzt7zIQ%fu5vVZKQc&wB%$haBaWq(Is2dwvt8Uob+@Yy=
zptnu|%pEn+jzl*!O>UQw5gq6K1q0;TiWgSvtxYEngcd~I$<Y@jk1mFGo)A?$?N9V+
zIM+GEaE^iFyhD0J-Jc*EdEY)reY6(E@eHWdtr>vL%8B;3x&t=Q9`C7`sw)Nn{p4Y2
zt>fKipE)W4TBFj5;)3VN(A;jrBOaK^TRZ@uod<w{{q~d@*YL<7pty5;BWfDJ{W#Dq
z3a1fej=@qO=Xv1%cuC<SQcw%2k=`2YctGopTB)FIbhL4xwFkDkW2@CSC%nkl47jvG
zORTrkd1-g_c~n{WZZcGu-<k!`5Hih`k0;-mi1WaVj2ok6`^yy}hcJs!zZh5l_x>`k
z_YeI@QXuqIP_U0-uy)`U$_Np!?|>-8u19T&h0+=5-uPu0!Nl?M4iEp|ZQr2V#&LLs
z%i!E+jQSLkToNjsh-(i&Qdl9=0_DFI!U#7FR4y|gfvfMRHw3*~Wof5`U&1Q@+whQP
z_ivb_LW=JXr>zV?m9&$fp=mE%6w+Vld=7=UiI;yiuZ*yv9}lo)sP8Mc%G&vW!3Z?8
zGBCw0xT|GE<t9S`iPGcnJ$Mp`d5}RV3YefTz9JDSPjd^PXvsYc82zE6W3<!v8}oj$
z$}o^VCkw%D0<xr%o!ivMWYVDux2k8@HmReMiIN0fEDyqHS8{CvV95RR1^cVl$7M9%
z-k@$kKOSh!O*mDQoz7b|?~qnDnm?m^Q7Z0`+w(K9KV$TU_VR+`JTb<JejeDL-q7-R
z|DH4oj3!pJ!&8>y9%G=N2l`3mgVE8BJI48fT1aWA63LvyFe*`Mb5pWzjL!#v)*g;H
ziROk|400T}`C_;8q#5c`h|jHnhXe-FF2x<6HCk`Nw|_z1cK36fC^k?8fDE*=CoOR_
z=LA9jNmr$xV;GLd12_(7*>D_G3T8B^Yjy55I(o%{ax{268eZ=2QR;LwkIwFSbcjDv
zV3$e--Fw4m6iWMk<EGNe5L_3;%*sAs_d9B-Fr}XY5lIC>_jLWyJpl<wAuGC(wsOJ7
z5Mv^CPNYH+vDphsj)P5`8+|mKj|cF8SEO~{5<>X<I=FC+T0sG9;8vIuo~#b^@BT$t
z5y;>FL&=Y~!6d<m9D`>jr+|YWUM&1M;_$kK8p6QLW7{L-(XA;Xii~6<ahgE_Svyh>
zb0UO82)qcJC4wK0*k_OgtrnJoF%S@?v7iD$_yT;;P7ej29+iqY_cK_6!2XhfkrWWe
zCX3>AC5Lf|h6__LR4f4ZyV*j=bU-U3lH{P0Hd2S81-GYXDy+Iu28d#`Gjl1ju-s^5
zA|(11coLtaUBZ7A8K5xzAB^S+N-Z*|OQDiM0tk<8GOiDN-x1k+lAO=4mWrNdmyw5c
z+o+s8Bst5t4<_u1Q%5F-Ah1hi@JnwE%{=iH0x7%iyYGUfaNOUcKbSoV81`+$wmo5#
zfzeJ5))f2G8*I;SVZGtFzoTLhn%V`--qrAukB)vm(2f^~JYk68ejX6(S-37j-WbIX
zw8J}{DKdPK)CoB>acg9ZhR5k$+l*#E`#QLga~ePapcFiniXL@ll<4(5IO$atb$<g<
zK1c5jwN#XCL+|G$+>3{Pq#)Y2XEbF-aqEpa8EB4(8L3djaEvGk%uP3pp?!6uGB7l5
zzUOB@EDVTl=P`wfAI5Kd;~S6?4U7h}m3>~j3|MP0>!^iU+QS6Bd6p>*5U1k2Hjg>7
zGA*1lV{UFg51i-S(GIP%T{qY;-cO@E15y}^e~vi{1LxUsJPw@gn6#9!tV$^nvEdC<
z2BTEepY7%Gf_`!yZu^GYenYJ{Xx-=o2}{kvik1zmch#IWoSCVjhNx@be(tYe``dmy
z?8`Sl*FlYp%-F34mer6c94MsY0C!Rys?HS)TeJ>y3O16c4C8ztip^z|6_ZNRp*hS@
zD4&x;iKwBl1|+2!94z#-hK4I^^9r3bA~mV8K+1ds7mr4YXAG*UP$Z%S)-6L<sZTue
z%6WXig{}+#yoZF9pPV2<9mIVjOf%bp29h^g1$BSII1ZG*>vv-N>93;wCtvpnB>>&B
z@F7E5qi&O9^Z_V3#1qKI;C!lEC-)c1th#Lwt*BbLe^g1`_U6_D=ePHc1EDl&x$Zkz
z0WwgYz3h7^dzym{$GXju<0^`F9%Kl-)>g?RQ{Srsh!8M#Gn87e-JYRd@}?aRoCoR6
zZUeo&&=-`Mc;Ei@aZ2Iwf+{S9%j6lr+Uz+T&eV*twt@RWzI>|%qa8#P;WH5RkS(B}
zFBqd^e|n3$-SBw4z<Q$wZ3oQQcOf$_MPTqEc$F1b9o~{s(Sqg;U#xfZ^KfdUXTno!
z%@4d=BQTWWre<dhKg)M)jEEsNf%WD-l8*g$gPE`{kP*>g477f-Uqs=8?&XH+VW;La
zN%vA5ad2xXXx-5IK(mJPJW0Eiiu#sxOdQHFx;tV!&em}n$6CDi#BjQyR`Rg7ofSNL
zJ5e#nsYaDa+QL+@Hd3%LzPm!|#w>)w=1MjMrUog%t^(`IXtapLDDz+!7=|Ll=_WgH
zqDL!IgHuhwnng^fo@Rar42hJrRH%k^L+(~=sqbSWsYe@)5jl0B$_ycm*D%XDU@3C}
z{`}W5{<WV*{l|Yl^wS%<HzJBKM;CvnalRKPZjxR8bGDPTfzcQh7CkXay<xu%9PX2-
zi+}_?{L`Y0eAFuvuP<aX2@9lwBGQNFV^$uGkjr5M;-x$WSW{`zQ*Md5wUCKs7NM;0
z8b(uTWFCtKe;x#p%K}+KQZ442|4N`}AQfs*b>?6xKtiB(`a&4=0q5px0i-G(kAo#?
z%ux0xJiqybXTnVYGtUS%(mLz~V|bhmpS|1xDcF=$bkTxZ9YhES29^k<3<!YCdGqmS
z006qvLkfhSF|LEy>*7?fuzCX+=rP7gaOmK91m!7=v|@E%06o^Pdu>nm`;@|@H6x90
z0B-yKh2;FlIhv*5wrx-m>`zYs3WjyGKGO80COx6Y3m(TE_m}tBwu-G*jMmZHI}B0O
z{RTkNd}euD#^Kj(gTed&dbpJ}^z%TNS6Xk5wuoF0Gn@~`<;(pA+tai2d<9Ylj?vMZ
z08sA-3!H;lxwoe`5Ganv3oRZG>m*(?M6>zkAz#f~ZPv}pI*bC~<a^a);QN7o9#HEZ
zM_*YgSqmM+S|No+?SikZj*4qVBJ3i@u-<LwV*3Faw&6Sv@`&0l8PMAnEmOGxo1<Tb
z%<fR5VvFd9ZVaC+?(`Fn{TDEqhqaRob*Jy5+fvol(V4)!ITDet3d<z4PV_ZIvPn^K
z9@`wj@7o5I4c#-c&8aVS^WG&CuyE`EmHOsZ%rGiR%>y)WIO1yOLF=Y>tWAGL?>J66
zCvW==x1BzWEzJ5q_a0lT=pMax9ES_0=tb@V%~9+O)pgYD=shd#vhxnVFlb%dpz}~l
zg?Cg!W&`{Gg-^vCbVLM|QA#6teiNBt!A=L&N*Y(xI~xW?3Kk9w|4Ni_pF-GBFsq^h
zGsAm+j$4<Z$5g6bH9e`Mq@&z_F_pH{Q>xR-5-&>O95|K8@ZYC@L{U=6_TB-FdE|T@
zk2}O84XY~37Q$V<ePzWt43GPrO_K~b&IT()^-oW4aNBQzLz4D_&qvT|>ZPltFylJT
z+vuIXjd4Oy@aD~@j^aH)WU$u<t#bFBl_-Vb)?3ng#dC`WhA^0*I?%}M{??$l<I%fY
zjzKLlf1<Sq41#U{F3?A}ok}69S8DNAjxcv{GA@0b2hPp|`{|oEM3CJP!{}Zg>E<<X
z;171<xPOMvKKm-__Kf}c6YTp9=ShbgLyT#L7gcPv(82w<^Zi;-?S?W=SZ{c|`zEAp
zEKTf=X|21{F{Br??%5x@RKa%J(N0Dr9FGH~2==YQIwNX|*YzJZpd;Fdf|<5bVNn(w
z@NyCz=8;0o;ANtjiI?N8H&m_6nK&P5;tpZWmEJ)5pm%gT9q9mjtqkeipK0}!;uUC}
zqg58htMJ-1tz(Q60J&FKff>0>DHYNg95B2dRM9~7?mRR`T)hA0*RVf5&rD`v@iA4w
zCR;{*6}MDWfKqp)+tj&mI^-AuLdO`lVN`_-LDG#S0g4hikKS;e^zUpNLxTI+pxchE
zZZIz-$hK!5aW$yaqa<<oTtc|-`xDFtwryhs*BIG@0V2??LiU@77!U6h7Je=uoZ^n_
zvBr-iuihgYcEHeayu7eL<4MP#s$}ebAGS1_r=>t4WWM(%${a34GO;Ye9?b=*M_XB^
zb;^)T&64g>a5=J>$-__UIs`YAiC^Kr02ZgMcLIzAP(dVydA+FBFW~}Tb>_3ED27hd
zoD5d@5wP6X^9VGqs|p<r_D}pM^q>EaaVu<1xz{~?br4~=r+2YC+KK4;R`)jm6wh15
zS2jj0jo#AHj>W3X+#fg%=5xzjB(K8jb;WvXxW9acvv+L!4FiHVPdDEG^z$11x@$!#
zJ5+$P(<!Z@D7PnUb*HlHr(0pn<6?nO@4q76U!gbj$0OHFm4PfnT!tCY`**?7T$H?Z
z>n9G^KKu4GH#Lbv55w$C^r)K`j&hT;bF$CY$c1e?Ac|fN^z)8!9F91<7mwRffb78#
zoFzlFPNHR1r#UJ^+nj&xMyFG$8|?P%eh+TWb}($RpC|5jnrgM~xb4rl-EQdHf%9>v
z(g738&UR)ZOB(~{apLLe309dLHActL4n|v5HqOT2czZKw<-x6mR20i?cSb;X6jRi$
zVZ64N@{s7{2Au80dE8MRcZTRyQY<Vx&336nywrFfu>PR6p#|Lpy;C&N_r(S_S~pk|
z&n+|0);1UBMAc7QW%&$i*vp376Qdx`vqQmJP^~w#gG!dNs>8HquMLSJc)4-C^thL8
z2E5J<`35@7qJfhPoR0(F{OlWOXT$TGXS{iO#-@rkI+~+v2_fc(NWuJ*nyp(k^DZLm
zK3Y5WsI?<59De^-U^nyf6O1s?dShZ=WtHG4u@K|W{|e*J4e*R)FVNQ;j>ikyaYt{5
z_bIQAph>M1VLhv2lY$DzI)$x2r_WQ8VF}TLf}+bbilK>H3R3;34uc8%D)-iis9l&w
z+4HDcN@L9_AtstQh>3(x?rj=j!dg_WBB_+8WfAmJ?l7mOg6ab3q>Y@M>?Iw4gLo<m
z6FSSb;n5FZFvnmk1!k=8HImXMARP#0M{0&nX(f{}29EAIPyTy54~#|{sAjV-(oqDt
z1~^3);nrqMxO@ACz_xn(RU3xt1vSkGhGeUm&@1YZeck;~Dmc%E^W}wBM}(exu0_?T
zgG8CM?fVQ~fWn;emZ+S&b`JbDua8m+=Q$voXO5oaX%8GpnOJhZ3Q-Ti@}gf5&1AKg
zmpkO_czPzPEn09K<kq*kV~8;OxSh`J?I^crcOVZBTlA(jjt<R>*Fh_^fTpj|0aH9~
zSu7%#9yH}zgP>z8>OD_6$;ZL#hi^oLiCevQyu5n>L~*-4LBQHf`?jGJCcN~jUeJt1
z^dcOQ5kCVx!V}Gt!d7a*IusQ2q7~gn7TgOz!g1Wu&J$Yctld;`yD^J1qH7`zAJQ;R
zV2l$2CjaQR<Nmng@%Rjn<AGWp*!QTITJxv|=M4{6z3mlSjZ9#JwTG(=QVFNQv(yKM
zFg(wkD(j7rK2Z%;L@0m+<c|%myxG!{m0W>ZZ>ZJK&jZ#PiYlJ=9S<{>Bsm{Y&$=F+
z%0^1@65kZADO4>`wmCJmpyV<9Jvi;uXp;Y~HyC<C+sWib-SG5=VZ`B}rIZ@pC`VFw
zQ8;KE3jWRL3t+9Q6Qd1uu=4A+ZC;DYtqY4@4jOTZd|$%3>v89tL>Q>F=!mCK4dq@+
zL2oV)G=zlvAo^^WHCpprB=IP%S%DYsqe{mwkcvHoby`MPg-=*IYqFdR)rDxvFk1D*
z!B>GBWG2*;UDJ`nsJG`g(0}Ad@c38%EXr2#I2+H6Mz3iXisUwVmWP*@u#WQ}&}!Z&
zYbzxK8DNG|)#ui!NHAXs2&$65SEp<CQn5dO;=Ms=c@`Jk!lgy`5ccAZ=F{_oiYm_Y
zf%os<JFUv+M%fd@KsTp&BWqn;8P8xG1H3f3V};(+AM~AGALmo38_c6dwoMn45$xyc
zna+t_L>MCKPU$vCFiENCz2n`xZ{U7^k2i1LdX-DCQK7IjGsX$i!h8r9*vueRh`01K
z-7xMics%Y90p7fM1C$MYa3h{>O_<Rs$!#dEZCM?43^3!dR7eq2t!T~Bel5iG0&Ind
zst@3{RZ>0ugq<(w=jk!jkszgL{mkx4jF4i&4~*2;aHkl36g-6DluaoWo3dEkc^qgb
zm1`{p`~Hmm`3)ez@$w!=YwkEC=d|rNh$n=V?dEwB<gLrT<Ebc$P_Uw(kY+GP=&A~I
z)|U|=oohweGq<^KA}G0_2mJV=6h@9iSTkPBM#10d)Q@^m<D4<u5y1lxm2ZXfw|Q@b
zLp<6>xF%1x(|Pg&*`KjLJ*No|N3<AWTLrqF(L<r=6N&cj5lrxB=6B0>Lw);YV820y
zROhC`a>p)E!T}nB&8NZVg&?W*#sfJ9#^XdkPPEoouDlnNN+sHQ!{+m9!e9sHpA`1)
zNa5O@s109$-(P8Qa*ui9<ru!VH6w|fxglD2z(RjgC#{H@)>DvS!v<O#vmmB}n&@f4
zBrr=V%|(3*`()-_sR-cxK^Qp}8aDh38X=tS$4E!3`%$!bbe0xK-LQY(Cpd)uF$190
z`;WAf`#*vxgwar{6@}&lhVlGt-sbW=?q24`QYdG6a|o|pHw*xyd)aGasCNrXBRV@~
zw!JsY@rW~4*;Wrh_ApoXm2IU!F;IkbYT0i&S0m054E^L7qx?=}k|Wl2On=rGok_G`
z;PsKwM|bR7b-|}7)k|4;EN}08xVr6G)o(zt-=0|0sGZK=R~+ZTIqOX3x~X7$+F??>
z4h$=qvW5~EEdZo&3uo|oo_L%G|JG0R<Asyk4UcvQ#26k1Z&2bj&*2%4rEK(%m5SC*
zjQbtWZ{E6JpaP>2;f;>o45e&DAgxgm0%-5vGe%hzI0W1639@gf(2O?8r~@02QlUkF
zTCkUb-nR_DYu;4EPICHH)h&?{>|AR9QY!TM4chS_LO&mA5lJ6_7pY^yAkf+Yd!fUa
zl^2bhwv)1}#T&dhA0NoshG8@$FAoUImwR;6zHgwTo}8$-#V7-iZAYm)Mk5k$ClidG
zx0?&9pl;~U7N^TpEAHo!&OQQ8(n1sz%o;7FVB0pBqqKe=6ehK}Gq)u|4XT>SV;urz
zjs=8(WON2aY#YXJe}nP|e-J`*cH!jRk+)pX*6>QVwZc&t9~`~G+5_#l<MDF$x?P5$
z%!qES@%Kt}8}yEpnd<--xD%PH`<Wu^bu4pKQV6zAOC(BdOQQ5y*^~@bLDoEi_ZoP8
zKLnQ~IYy|YZ^I-sj^ud7*Dky5iHBEOWNK^9-LM%}Za$kUKUSO??CwN1uMMQX{Y%*X
zz+Z#@v)=$}<koPH53nO-YmMiqH%RsTnbv0M3IT=~b0JhcpNPqA>=&e55BDo!MA$Hu
zRv!*B1<cd~<x26O32|Bwko3uA9|+Hzxxd^Ubd#QK9k6DkNjqDGZiOHrf^=HzjKuQ{
zc+Qk7Wb$NmO0VhV-{$pkj^^jLs7mJwC3<ubDwnddqeTX(LW@A)&PN#9aR4gVRas=s
zyN?ZFiQqSHK7j#vwC+wfWdwnGIp3s<8qU^m9tYk$KamR39R|VuabnwcJU>@fU440v
zS~kzdW@S|s0Y*pFiqqr3iwKVMf#c4J>^4}_-9sI_0J1TP>iy$R1Xe4yeZvWNUbX>y
z{{nk?fz;}m+y&-!n#Sl*8|+=+Nmj!QkN4E#TXkO6sg}o`zKw0)QMVgJfPn$C6K#0H
zmOQzSgb_sjJUv$1a}e5z<9J|Knz`rU!uVk1O*qY7jz`iMTir;<DCsbWVB2pnu*E0<
zwo?_bI;2*d=L3{yf(=yU?L4vHZV-=I*P_^D#{l<9#K4zYTyRpc)HKhT*Ah`JDw1Ko
zMFr>w?w_5BxY(?{&?*vAUFUgD03hDG+qq;w&_{)A2I(vb02uHaC-mpP4*Bh0ffWTx
zK{Iwa(^9kAY&S+c>8cdWNjQrvWctMVamTyQzL|}V;3%5cVu8oa1#jd~MB_UEd8T-e
zSq>szqzsf)n*YvZMRu#x!Upy%F9E<PEP~SynupGKBrOH691t`dRO3<kpC#I5YQl&T
zn_<N9nnm{V4MZq%_ri>ZHJjJFlj&57S#4GXHX7u6pHY73d(eOGH!)6-bKfSKFJ)t%
zkpSnxWYFe`ZQFiB7dFe0>dCK&P~Hu?-5?^U+rV+$F|45-FU)r+imgx~-)=i#1ib2&
zt@VyR7(5cRC9}b~p5{$;%pJ#c9xH*h8mt|7JWf3B?{S_7wp#Eg4u-sa@NkQ#onF&T
z#1kyNQt=@5IqK6FlJg&>6igK5cB9z^z`RM#(YtR%YWLnwC@5Ik!Hi|`$cGM3K1x%s
zqihwo+YM+9<`3Rw$G+X5QXM9F`Ry@qJ{~ySS$Nwva>xQV&kBxq$MJX%J;kGS3PJq+
zeMT;Q?Ki)MVZiO_8Fk-DVXETk=?MZs_j>-KlF3BlaiBFi(8Yj$5^XkNo&dCabOfLf
zY+xAL<Aup(5$<Wmh=|q?oaf01DJ@P%uy2Aox7p7J&>P-=GZX{)#40y$!T~@l>N6eP
z<Ge%M((1kAagw^*i{QM!;PL)5qW4IQ>H~F)c8)Bw0SI2+zW_$ZcJW5G(aZ<F>AC^x
z2wO`87F6#)?X`)HJMPDOI><cK@P2<_yWJSQVJu0|50)W8?>HNsgxkiF;G+W%uaL_E
zdY+_n(t2db>|n@jWn+@_@gAoIy`iKTipSFT4r@l5s}xAx84WiK$ICnP^FT4+@h5%~
z^*FHo(H~BMEb5fQr0883x@D#;x9nh_^x(-}<;UZZiEn+7zpgHj&vv>6NGj-Tq{6ns
zNVVQQtk;F=!pI<6qKby%D5*3A1R1Fa1dja8sjcqw;@mwv;Vl%77ab%$Rop*9UrACG
z`8of`BC^cGh(RZLD2oh%bt@o!G4X8EH1h9r%+U!=5KdW}mXQ=}fAsg@@uz+f?azM`
z5XE`-3c3KZ`PHve6gb)c;Qix)V|<3~>6t+<!|?W#Pw?iM+0o+knAt$<4PX80Z{mJ`
zhgt=nzWD^tdgI#l4z>pcq`g4U(y<S%q^f%din9l9L55;K9*6gWAT`woOJfkYT7sFP
z1@B&75Lw#$?S|*4Cp_*CjB#S$ZwW+#nqy?7;QjmW0(~Ub*?x2M+&Jc_Y_Fb{eIBfA
zx!-{OhH<n+1Vsv}GR}Y7o=`jz<UCH?#~n|5#nDd`VWGP(z4;Q%7lchnj6I31<Evl&
zO>DK|OJDvn9udxI0|f)Ot>En4YyWU>8EBIAcEXI&29F~`FzMyL9}nF29RkD4{f=$h
zAQp*DfK-nXQbD=xncxC5jN^s$jfigo`XWZBbs+Ba+_oK3E861@7~5b<W2DvS9Y^yJ
z$1(6Y9^@ML4bL~$7dp?A$t};MSgHZFun%*^z}Y)qUS6=*Xb?iG<QT?yYA>Mo*}I1;
zOhs#~K@~l7&vWKn@T@zX(HLDdQKR7;9i>pXFwB)EdT_EOGg9eryT5-=>P-w!PtWLQ
z!^_JZXJeUiHBT&bXRHXw&kDf@d0sco{ZRGZ@VLL=+1_Fk!Ev5AUtZk0*`dlZ0w#<g
zQDY8DAM|U~t)m?W#u#|q-vOiJ`IAooD9(TL=Ye0jQ}O=8KMH)Zp|^+cdlrX;vPc;~
zDkxvWlk+?$A(U5C8e-XRgASnesCvqXpW|TiV>i%x@Q5gY=s&A&4o0Y{Q#7>NmhN0a
zhdRS2JEB37je*J9%#7y@FbEr=5CB9bx&}Sw*B<V?r2y>d2gqLB86D+SqRfk4dPJ1F
zLlDh)q6i=}fHC9s$!~e2qti|+9b&ZfODV98b&nW2-+<B3`{6y+g#jnRTr0gj;c?Qz
zUbhOghD{8c66~C3qvZ`lKThnmpuTyFtrS*XK6s9{T1a^tE2G}u-Jzx8`ORCl5Kym`
z=83P}o9+leAUZlK1VsUrXx8N=>%~xn0RpuYeDdZE>$SOsa6BGRAv16u51wB?7chfV
z7WK2R@V>+A;~dR4z{-wyjNX_ndwa%ch9Tl*6b7AE#~p3(Ah6VOft3O_n(z__o}^K0
z^#WOnveBV;yu9GJy`V=3r?QOp`_JBUB5ya?V56PL;5&{3N9)+1-vWBWyR!jMJXOW{
zctI`P$j8}WqImo1r*x8e(o=P=PR;P<nXNnT?_bb6Lm<cCbF|h`>W)u8{W23_?kC2<
zNG23P8(hTu{efZhX#^L0o+sM<g^6iSEzz;w7{%qKH@0oZ{r(=u*|0r5v#fY)Z0y|)
zrUxS$<iMjhJYHU~Z^a$g%xwU?9tx;rx6MP}M#r05#pChtrYAeg0@O+xP7JLxWLS#g
zc(8%Up3b$#Fg+u^cN_<{ePfy5<H&rAqKd<vU#&IV?=P@nc)H!_c=aeE6~q1IGvGz=
z>Dy;EUmxa;>;<xMY-7v}QS!5_P0!AHPq;n5!2q1zV#(u|`wQs+1GcRKPdA`<_NcDp
zk*jo1a099s?(N`VFBS4I^gs1WkWz8}%&%em&?h+l*6+jf+fPux_e;Q6!dVD{1Gj|J
zq25`>Mb-O%^pV}|!g^|B;5Zs?Pfv{Cdff52KTto}FuGwBbE=BDF9q!FA&{)IH6npf
zW^8>ppR5~(CwA2$PWyPGlsI>vf;}t%cZ5m_bCB=3<2N!hQt=e}31Ixp+RTSTnaOA7
z))rjYXRm`P;gfnYS8fJ^s#5p7RwPe|!7C#ew(olj`=t{o8*XntApmKtlN3s`_lDcs
zCxV4G@W%U3zzoMY(O<racjp}@4*I;ehmMAu3Qn+83Fn=~HfyO6PoiWGb<WSb_b)KK
zVBc>HZTF^EB0%dL%j4xeAcDHz@OInq^6r~>|NaH_c4L{B;ueyGvm8EuoLiRP_dXFF
zw0c25N70HdEDN%i+ZU4aAEgw!fhB0wOCtA<_uu$B%C>tvZsh{EMl?J+x(W9E#u(tU
zVVrl=t)h>{Dun>HeZ%Msam8RL*q6WjDF#>;;QjGH)ST~w-PE4nyut0ZlOF{09t=RK
z8<WLUnfzfLkNcg;9VeqK7y`)b+~(-3wdPaI%vHmKkw+@HZBNOWmb&51enTIIci()+
z7LZRj)GvL-9mNmafAiOIe)bJC6@2o^CyxH;U=NBUqDEM@fL4u-ftUC186xQBx{QJ2
z<?hZw`VdfsWh#!lbMmamGB%>uy>2uWRk6Dz5!u%!^n?inPft4-GDj|WjDgljbzzLm
zF4kJnVYrWuecvI&pwjWUzYl*R-o1OrLS9<%&2M}Y&(BY8busa67_jd<o}S;bS(181
z)=p*j=C(srymr0dcH5app)3<TT4TU~=REY*aXd~)5pU=tc>i)|5@Ibd75aM46X)4b
z!YcCe%J=(w^2C)ji24|4y)z{6aiSeOxclvfCn!T^+kx}n|5c2C_dmsW`z5r${<ot1
z%|F0=6J>a$efbIU2fq(Kp%|S-&7!#5?db{q{R^}fDp8E2MbsU~qjfwU4?I1;!3HAZ
z^E{G^9>dbgVI|<DlNp>&^15J|IO)X&!$B*e8H&sHnRK=iU4|eJ4qr(jA}9$s&=X&O
z<_(zVDP*Y|oAtj!fKlM{8D{j9rT-%OfWRZeR2SMv6o%r_KuiWy@%Fkl2r&P#--pLf
z{(Bfl$2cGKQMlhkOGPA}^|PakCEW`Ta~HQPWGEBMYN2J}g(wFdTev?S*lJ;rK<^lX
z35L-`N2H>4gA~EdLD1Jf`zHEHis$z9jN9`Q0AbmGF+fU!vo~*^0WV@1TAvNLI?T;p
zmgyM|iUE~^ZUZm(J06b*YTco?irfAK)q?l;&;GMl3T1z0xrSkk3^IWkwX8=M69MfT
zqQFU(KK88AAqpoyVcra+w*&Bl&C&=1k9VXQMmN-F7JD)msOJ1J^JFMIM`v_YF~jy;
zp~jL4M?YcbiDFE;6E8sxu>R0_GI8zf9na6tI36d4IG2u2Uca5*lv`l^fYyqm8&E24
zPfv6xX+fzsD!MOswC$VjDRchQ7>QMC0ld?kYJr?AnSr9{4{p3#-oQo&{6N?^Vea@Y
z?yPF(!NOWrP=wjG1gC~lJvzz^YCE=igVuTIDGwU?OQVAC{q%4-W$VXM(GC^`tNRYw
z3JVdMx$lFPM=6T)WbnW+;C?v$;}q1U?BWKe>EIclrW-zc_YUXdj_0QvHYs$p?mN+=
zE5+6u#^Zrqs|ynLofw@?Y8!&DfBhTKTJZel8Cz4d<3ZoiwgKHdF^QqS(K})|pZ)&*
z3$~)DTXCw-=+hVtIt<6l9it80o}Xd00KKCW*7UcZ{|(rG`VGJa`bmmOzw!zCU;8CQ
z35w%5(C+VW++QH)0X!?}Kk(NvTJP+T!bW)8UKzV#EDh3JNok4Vj?8=xlnJnWiHWj&
zihxu}`URZ#2hh#nZXw?%baIEoINFDH>8V^SGLV6cYKgYA)*FSkXeP!1NM~>ef0i5}
zt%LA!lqu|4;EfcUx^E!;B)z2roKD~3);4KMh=3jL!1L7=hSQTl2a0Z40FIRg8j43P
zsRG><pE~8jGSe)^2?1`^P)0*LJc;r+*}O{?&u^Y_+oB)1dB1e?419xufeZptV8cj7
zRi|f%2nLuSXyf!;58&ngf|r+ffC%2ceG3SyGIuMU!_pxA%rgDaXGA>EL7(ZP+G@qS
z&)(5WXboBp^kz8Tzx&T#DWEsp&jSV{3asSZi+Z95d=q#b{fJ6p)R1CW!P6(VH2;VU
zJ5iw&9GxMkP{mfC0D0g%KEsm=-oAN*^KpmlbSw|oh$zhLvrgguc>jW}?zlhR;W!Vx
zd42{e!%b`5$gj5(rF1rsF~jrQPq1%0_NQkmy6%PU#=zt8z<%2^S*fU^wVqk6FYiCY
zc)S2w7&$ct>?F_F!Gm4*w`|6dXpXJ&iU9hwLxg<|?)Udz{A~ax-qzDl%MI4)u(WQl
zTF`|Dx3vSMR6Kw33AWqD`cj^dq@vii>UEsl8URp?6+Ry?5AyGi2j0AWivhuLoJ3C|
zkZr^LaYq@$H-4C|Jjlh9<cd~&>FI{EHR$bzK5ib}!+Jgpu_UtVoiy3e8ji<-y>9Ls
z0ORs+1Q8jr;?WxCU?kXiKd^Z}?<kG^=FRgAGc_YEFdC#ayxi}Z2T<w;tIRK{0`y0N
zwSmXk(3)a@e#Y(T8KrD^JlIn9-TOPf_Dg@ov$74Je)%na%lCah+WkGYr#CR)7o!jK
z$2*kX(0=V}thCw=I>aCM><O+#QGeoBQR{~HU;R4T4}S&rM}HUq(6)*$Kq~@#X=d|M
z$ucU%#_(iO@%=dSg4VL)q6+hJ7Kp$Q_f8KpFTU7apm>$9pnGf<)-&h*qeUkbj*-fc
zo=D$a$N;<`TvVn7z+8Z!c-(#L^h!`pCh!Wyj4Xo9(jVlccqU{B46_V7B;3(JIK5jx
z(ax4iVe<lw=W)XP`S<O~0bwU-aEJSIdBe)3M+fx40MLe4#oa24Ri2H$sZwzqCyukx
zzod#%+1~_X;N82=ux;#la-Ii+E4Cf`?HRQcJRWzJX_4r$K^CMP2k!6Q;r8ZJ6!(c0
z6ST)2_xBI<<Gn{XXolr)wX)e6=$}z<*mShh`>r#>>5IHRN-2EpH+~Io-+l>izVsyk
zhWq2an{nJc?L2Y2-SUv%kAuj}V{*4$mnx1?f>&u~88cQ@DD-5jY&O3U)t`?$HTLLE
zO_@_g2DtNi+bg!Mpgmr)Q_^ufn9QX_PFgG5ch(JMzj=YI%|l&j`f<^nI=bBeWfjzR
z9E=u$;N~H<_s>3q^b=YXpFPN#m&#ao?dSiWwtv~RY|GZhps%%l8Dq}1BKFRm=lnkw
z#z+<dWXZx36_BVRQR5Aa#53>?JOvLxi4r9wz(Rl}BY^}JOJHnl%jQ3M*%`6soMZG$
zYZSiL$BIKGGo{Qe<vDrTJ7UG0WAxtIcZH0B_Iz9~)f^L=nB8OW&t@tW)ErfrW*CEO
zE~^DeR$Jh=#HJ@kXHJXd!096wNL6uvyI&bDBD{d7P577?K2Z}f$Sq~$x+7QO?vfN3
zgC2&g$XkUv(2oP#zGL);=kv))eoI&0yI>-hjI<S$RG_IMr3xD_mKDtz?q^VG)jN_3
z+VRBeAT+dZJ45ht)Sg0+!O4CAz%+;vCZOw9WB33M!jS3pvq80Bx}yp&!x{3}?85#J
z2U0C~|A*h=_3<5}HBO?!(7+Ph)^`7b;Tey|6X){-eHf1OK<^WG5fu51>qr9l==k=>
zUvVB!s!>z}o+D}L2#jA9=`<Yw{(lbtzx)%X8EUQQqoMrcSJ;35Ut)aug7Nho>A(3~
z*l#7|!AuA6YffD**<})wE4xaRFpQ>^^Up)q=EOiagng-vxHU{$u_B*}plKmDjemAw
znQk;?AubZg@>I}cbo~{d&!UVyh-HWb7xLqwPH1WUZ0m1W7nVr`;t6@1wb6Kx=?D%I
zVccqGL-W`W!;=Jao+nGSB?1X(cTT5prqoDhj`FRT$hVEo;@&#E52Ui;>)SgdB@U^<
zl|G&y`1tsKiAEAUYGaKpa@!7>?B#Y*)cuBXyVJQZ2{6iD6AeNbed6)>h#{Pt#+;Ms
z%!%kjjy`$~4sUJ7y!e53w!i(QQ0fi){mz#;4#*KCNX`kG`BJ(W6D1wUHC?$ba7R+d
zadukp0RWEgA0HU4<Nf`OE0@#$ttnp<NNb$B98_w&jsyNW0a5()<(&;~;^^@Q*R3GQ
z2F%a$2>@v4iFR<h&b1(KB%C<;3UuGmoWx_z8+z-Qoz;kY-6)A21K)oA6>CM-s2yb~
zhGUM2{{4ab9#UKREH?2uumxFWW|vlf<?i!`7Z&gsp&AuME*W{NIF5sLfdIs*npz8c
z$t6hZxTKvE0`2xZJW5W4>5`&KrZGOz!oPC&&&PxCSz<D#PM{wr19+WKiUwqrN+H&t
z6&M{gWlS&?77EyiAUqMAXOEnj4I-2A$hp8|#)BO#XrVthkK|ZLY+inCqSFnJ=L7q;
zT}m9Nz*B}+VHY$eq-1O*qn`&o9KIaOygcWiZ{>D>N7*;@F(C-GhZy!+P`>^aKmE&p
ziE$k8^NFN_;Uwv$H`p|s&+jk?ZudK0uLs)eNhL|jxZUrlwPK`%);gZA2QQW(fI7Z>
zxuI-#JRUF1K9K(KMEhU;hiLEb7{B=$$It%>>}$bU30aJv?y!IN-=at%BF9kh$wLV-
zcUDhQmb$tQ<T`kb+k0#aYtxTAlovuky(_(R8O|8z96W%n=nKCt7+^(w^RCeALP9{7
zJjfTJ{h68?{~jI=u=T%FCaucsGa#T>tn{Syj1eLgtRaBuuqogin!!f==;3@0$?w@3
zaw#bHowd1hp!bv9XQAq<^?~C!al746YYD~OfX$AofX;!@2YO2IF|ZdRCS*bY$qB7Z
zR<Itd3fQ&{!wf~asroeh>32`~IPqTInOfwIQWCK3<cSs^1VK&YacZr&-`?QHvoa;B
z^B5LdIONFRgHy-}kH>fT_lf(B1V~OP+?d3JkK=i%6Si%qLdYjt@2I6>FB@hdzD^DQ
z6E!E?wjH@-^z%hx=k>r`Nq-33P)KGm+l15aecxCc`8e_M?R(tnil2Y^EpG2$s9TR-
z12e;z4)=r2VXqgyl#0=rv`V#9M^ehTW@kE~xJPi_5iVbZ?91`x{SDifpO94Xe0*>l
zNtC<#Xe_CT6?aaqs>ey#CI$2b7bd69fm(@I-|m&JVJP4eZ>o@K$dZ60(3(+cvm~Jp
z@&vPYBp}o_g{lbvqmc`7RoFRU=fvC;sn(0!93GR==z#HJettZ-(rdUZjZqT|8l2N`
zj*jF7uV24m)(s{Z+nZxrN4Eh1-J!ELm}Z>ggZk@Iu<twCc~VZ8)0IS-l~UQ!8cYfO
zB~^G-R;D;?RTVxB!zQxs=o85G4hN9kQ8K5llrm;-jHwVxXj9I>_#`27*|6O*L+g$M
z<`XpmetaOM3Wy`W-x)0C0f`wqFT51K+_m5ze)|h-yyNll9c>u)+YS5e4roT}9dBP7
z9v>fgwTWC5KmGiMtcst$e#P->k$nTyaNW-T@&)}@^!Z3q%=8WO|M5>i$*{znV*HQ(
zO~`-aD^|s({2T8yE&&%-zud~Oh?o!8FnrZ`>NTnVSlLNg^N68NubYE@1<vUfa#UY`
zk8|r-eA;y5vwmGmar;328BAvDeT&QH#~?g5xcTMzSr-D4i5N~p!agUZxhZcjXMal{
zC&Yn%_6Vsn_}L*cFx@c5#O?}f6X)ysnGm3Y2}drON<#ODQ3n9W^ZN^b>f@5A&(R=e
zxZiioVd$Opo)AUd-r)Ew1VhTG1vuM@M{Ag^UlA4m#qWNPC8fXJZ`bB;{lwe*7u@e}
zmpH@B@bT?i<YggXT`B{Kq-+aC{M%j%DwC?05e~dzSOpF#C)nsgT9OQ&+At=`ow{wX
zIq`VBV7V~6NCnz^WPC(sPRckwz5|irfB*UuC5m%IrQ{8@0Co$-0UW)Nd^sl|^$tlu
z7hrcocS<<3ZuBfbAd+yN4Jj$k#{>KAMptCC!1Xbxp%*Hqj`O7JR^@`3EDiv7zd9I_
zDEc^AFPatUwnGyuOJ`{8MFd({SGvCZnlm0R)`>Fhlo8vrQE$BOJNo(HWRo}KAZ+L8
z=;JdY8gM$!6_>h%F%pK914D*jxWiSMvgm-!f%bf%w}!G+zSKo1ts4f<3AgvR_}PhT
z_uiubM;Qj_4*haRDI28TS${fM`#I-;B#D4JM=cqn(L`pdAxRb<t|?rt=9sPVXWw>)
z>#0C$y>Rj^XP5wv2?N{ZTq{g7G*`%q(C7nMGg9G%I2@2@luH;<q(rD}8lb6Q+Z8q`
zmCl&x$2T5;nbot>PhPSE@Jyk=w$V6OYmOw+JKj=3*=~?hk;@L<fwFg${f3(Zx7(dt
zpAF1@VA~4jOYmi<i3vI5P%!==5{KP@G|5XzOycu@{U0GIp&k=Q{vZC=F;)aKCkrtE
z0M*n<L_t({9JtJR8WHgfc*#ir_CLb^^*cbckjJcl8paI3if>;75q$kU0>=y(X#F0`
z4ebH(#IK3CBm|#4M{Ea(i^#*2u-=nJHX{yv@b$%VKQFT#L;e6*Y}kqZlsN=P8@S!S
z;C|b2oG%&+w|N-|YpHPIaBFU8A?MbVkkZDBFfpGhg@bs(wgM*iCe6tAor6-4=S6`1
zw&Qg+o;A53iK8TbMmAO|1RzTPfzJpwtQE3VZ2QD{oS4o(8xJr?RfC-md}}YbB)q+U
z!JOn$O0-(Oo(J5?iQaA-h62|A4)BrKgyxE}F(Y4e1%2=fwmD&A(#bqKp63xwU>eF1
z5?0n4YRZ5KwvD~Pt{HG+pK-)alvIGSp(e%Xj8-W7M(G=mcg8jAiCT92vb`Z~JJS6P
zngs3jL<J?Xt{Lq(a6Z5Dz%DnaCd}C}J0l|4%;u>3h6zEcJF*G#w6G})YR>REkP8p9
ztron#f5*2!eG4^CqJ(k+TQR^AnzPVj;i6+q)O`=*kK%P6@!;_wpMAolqh`Vz?Slka
z?*mC25Erbuqt6kCd-N4|$F^;d8Xkk=WMO;ei_;Gp>_{9>v^Jnl0=U$Kx)o^780U#T
zY4XV;@K9+nkBH>2WvX6s%}jF58$1`d4?qm%en-_5k<<WYY%)opwT9HOe3FZ7Sb}KB
zY$w0xT%orwB+fNq^nuY|^cvI*YV}EI_txM#;nJabbto^H@<1v(C6B@c$rQxx=r5ed
z0WCYqe&?7oCh&6j?C9373-x$>z|2th9eo(YUMSlHBte}KC@Us<yWJsWhvY3Fy8<?b
zRqXYSO*NiD17^pj3GMg})r3+L&fbwV;eOjubH;2Z)JDX7kbi^F@U$SF)pR!gH~(+=
zZ~z9_#5o3Bgq)l^vLw95fc~HUKgd7P&272Dzhu<^>A!<<OsLY}mP$m+BpRZuQ@XoI
zTNWpMX7h~J=m7x_K=q8-T9_OO-{jcbJ@^9vp@fxi5@&J~j_dLmqelp5TnhXhFuDgk
zG}%31)aUC~FqMPq`SCz&#Qdij>h^}@4WoCIZ9^`EX42b6BkA$P@$C=vz=UG)^?Y8J
zD;(FKClx|jrBnbZB(Dw6##Dsc{f3-!7~DF({qc8UPBm=zcf7rQL8%32?~r$1EMrW#
zDz<lSrG$+YK0C%R^x3d&TO=1VCZ_Hid@%7Ll>|u}q-3=7g?-;cb@?~G6k0oR9F&i4
zJ4ua2_>Jfqkd$z?lTO{FFk}2RoZAHZniZHmh!9}g-l&6r9T;O^zrR7Xz?FwrKTg<~
z$Vs4cLTU#5)8To4yE9nN40V4)NgKo_9{xh!c6dMO4RA+InbOr%kxN2vCtl-;oD?nU
zFA)i>>@hg8zuhT2b;tR9VVQ&G=on!%`}*@2u4}zV47Wy`mf&m$9uahPzui!`#E`pk
z!+ADri6t0g4m_SOHb!M;y62LCBuHfgjwdG!6*!{MCe<Jx8O)|h=v18s2<rl+Fs#oy
zyRvE6<K}h$`bI+T8Pk?fNt0ADbH$gppI|mIjwf5>4BQ%2Le|N&Mk~O+-_VW*pA%)@
zP_`{5{xdFd=a%BQ2z<>Y{_Ny7#2t===gGrkD=bNxV*rWjuiNd0ejbDxA`7dx6X$V4
zmGuD5Krz3eGk2=sR5*Dp1&w}|yci}V(kXLNwDW+*7Mu4sXezMLFx!ExOyrbNqogI}
zg3$+Tv`clOi3B-O5rw7-<ee7-gsx-=T=^`iOu#2ty2bA_@9)sEVf2P`Jh(x@kTPJp
z#d9|xy(0xg(=XMKp_X*HVEbI{ZlM&@$UbsG6GJK+%6A1mIuG4>V2%#^pZ#yiL)b+6
z@BAb5zxS_UMgh$H*WaKws=pQlmqTqhB8(7D8L)%N0Jqx>JuU@_ZK8_==U6ZY1qf^+
zGt3F(;6cXZ|CS5;4J@J{WNkt5e^+V%xA+c7fg#9$n-p0@rjVFHaAHf*%d8zT4QdW)
zzyK4BYpxg|xj&B+;`EEG{_q<A>Z4;$!|U;c?l+WrN6H226Ycc?=D^twe4GvYzCp`|
z)+Y8Z<Y={4^f_^!4Ra1Wo==!L?(gqVH{>LE-#4UE@$0XDMBeV$Z+FVhZPKfiGKwac
zI2DJn$=tUMz3^H6`1m_;3UeBD`I-YdiPco)GKQS(!0YuwlEkUnsm4jA!U6P<c$TC{
z(5ofwIH|iA&6M9IIvsJgONVQ2cpe9fcXGz_@g1|B*!K+rVe`;540%Gtq1M;|5s}v}
zj#LZ0x1g&STE)jIj5H$X&lA&LNSbkc|Bn0pM#&uLQeE;hpCrFiNum0ovO*&y&?WpC
zx#H_@wzwQg9*^iASgSf8c%W=shz&y)3aW8ZPROO=oP+vxQD7P+YzA@|W+=CvtG4qY
zPo*+%MF1!jSu4&0Zc)=I@p2BVTp2o&fgHxee$YTyLw(^%80^=%8A>Ht?KaW+fYyra
z!13`N?RZ6dnqWZje!oNZJA(!>G0p?lJ5njg<%R)6+25dG603~{%!%P=Owk-h&ZDuo
zCou7xnXzM-$O%YF!gKU^hBPr^%3|sip|lW=v~3{%()rxNp@nKNOo2>Ck_lvi3!{va
zDn_WfY))PX8mXF^AX<P_ae=u)CD>>D1Dga_XT=l+a;5%X6kk#TW<v@&W;-d_mN`*(
zHor0Km#=X<7%`KS1BQ3Xi?@x2PN>4CLsP*|zx;+_Z$M0thV%J>*JGd`4RcbT@BfGY
z82<15zhI%j@qgzZ;a?d+uQj3kkN;a}sgtk{vl`rCwQ{SN6Q{h;TfY)*;~3{>AarG?
zNE29ux3B8UwT*-tiU<4vTrUq4f)kSj^#A>T959Sw&J&vo+UO|ZU}vSMpt!-$7b||x
z1TB@Ys2@PC+)B+TO;^Qi4Sk+)Bec}o$#%)DGUqH8oaYm1B={*%$xt6SUQZ-pP4OB$
zrEJ)LdZR~ToOu56_x$|8uJYb{_^=%3>wzBGQ+eBABcR9qj+_%7AK!4CC)6D?dSi~)
zBZLyb9+;m7?DjXm6lyKe{~P%v#@vmHjgCxJl(J#;5icN3T5U}1h4^HyJMQmaP)bE_
z2Oi%~yk1X?KJotb>$Oc35l$^OA(`^cTq+ELes=U%$7pBZozd%?6m3kbbVBio^KlT9
zZ|e(HY^5N}1SYU=H856Y)U@*qC6nO$#|OTD`v%9re!HRW8&=VUCWRuil?yDzW+_Kt
zo(OW>Lx~Y)nHdh=-Wl-r{{AK8uL&-LFGe^p%^m8F^Jwsc-iH+bB2Pdcg4PTah4q2u
z57C^-TY$<B1#R@qT$$kn<zb%-Nn8skh8M(yA18bq5nTd63nmOQ2T}y*NUZkr9BkOj
z8T<VU3Aw?JiMsEhZYtRJH|TbUrVU9Zj_-fO>-(R0F}95M7Sg&}(4JKB%mo7_W-*Ns
z#ZLPTM{BUji-{?8hEzA8l*=HsHduEip+bP<?5=PlcUAVvEvt8&kB0sI3lI4@h_$;i
zm9}h1+l@qOKS|<2AU@z7a4kq%0ifh3;`4V<mZ(c%=H#~E1~L9DjC#;3gi(>8O1Hnk
zp|^|-axGao$>&@!Q-RsU{{GI&($Qc>hzJrtvnotdWKk)`*->^y_lZ0O36AUzA00LU
ztsC<GjxidZ-@oDGJ2&L{@BK%pwemU43CI8SUr_^O6I&?&9P&T;w;=!U9ujYdoNTW1
zcYc4X6lodd=zCZSq>#+^&rNR;yS!H^Q+1y}ir5w*tR<(x{)7Jl7!$4ufa3Rm_!X&S
z+}`3m8546%>^Y%`;QRMiD69_Tt-`BAx}kTY0<T2GwXy7}ZW}#Py(4qIg1TUg6K8KY
zUr*fbZ&c(3Io}3*+P7^-9|K>1{spD(nC-;x{>%R_O3B!_3IWfr`};fgw|8vY4Xt-p
z3zScx#GZiXw?8oPxYIFT%7%RZ88v6z-Wukl5w`Vy!Hi#i`pw_|Qkb!>rW9mLOlZd|
z)@~*7n{%P&T@yUi9mmOJOvxFp39rVu<$QZXo{rmFj!Mh0xr2>IE{f@boHq79pB-%;
zFcI8tJ4Wv~Unl11*!LTulYNJ0z*;9R3dh+S6)xMx_zK~~B%;7zw=&!)9gE5C{eH)-
zZdftB&*Ox+p_US^T5<w5Cc4r6yNr8?dEXsz0k-SGrfMtWe%p}~qpABe##>jKxwyMT
z66vz&n*qWk(issE{dD>*a)BVxs{lu;OtgeE(+NS|Nfx?Y0oHU%WW5K#(>tRh1cKXE
z0@pQ8;mR%z{*^AgUc`|G6(sw9LwWncjjf*%NT+DMp%1R>Y7;oW;f$=Tve7g++>zZr
z(^#E-4Q{Mg)u69>@1g!s)LbbhfS}wW+eH**+kvQ16>-Q6#m5q}#Iw6dMR7-sa`(r_
z1CMWiM5^1>=Y39xiNdwQUD(~x8_a=JqP%HNR-p?2JxeGAk^}ac18_=4A%bKBK3m8=
z+2^g9c}UX?K3Z^h{AY^F)s7~rbjAaWDG%u=QpSJ+sWRgI`9KM2DSM?Avk!P17(SsT
zvB)Zd_yED}zCp?cEjz?_Y^C5H@gxw({eH*q|LI@h`8v`6pLEXVq>Poff$~595Aak8
zxvNLvjiCIy|2E`am(77G2$9`1Advs|FUWUxpR5F2NetHOtwoo@M9PIftB;W0D*WTb
z{9pcu(Et2<$X^Xxy`lV9{;POBpOBPsz8ZR`cE3-DiXnwU(IDtiv}L1XIL(J=V=!jU
z8L1XX%DCTdNO#BU^}_3H*i>l{ONza|W2;m&3h3P4BA`!G#yC#=`iI|Ps}zis#4Y66
zI$Y?B*;2u_?Z~C#%g?{S`aLc~K2zg(;d%6cMFo$K2YTx$=L>tp4V79@wgjmeM`v^3
z_s8>Ze<`HI!*EWz@Y>+Zm<kKvfru2y6X($|!U>xb@nW**Oz~t7Lx#l!n-g<|`iwTK
zl<@rcfI_gpe}#zSe0}E!>3*3_-0A8}n?h@b2?=aWaMIcsu|N++MZLZA3z8pNqymyS
zpAQHGWxruLKDLlAK92*VHx!}iXPyVf<XX4#_4((mut>j@P&=iRfJ(y~@%e+3n>qSe
zSt;})P$`qJ?Ug!cA@0r;Rpt>vS=U+8pn4%!2Ci8LvckovcekkK%qd)=oa!|Z#xfWs
zs=)QXi#yu$1)qa%%3NV8fO0xZNSKm=Z3i?V+W`83kvftH)EJ6qVarEo&~wT#>zJ)^
zWsifQ)B;7A_{IQwhd`MsIyo%@fgL+}fgd+I=E|oiIEhNb%wA}l2C;$B*gsryA%O`>
zj8|`f*+hH2&|fDNNTO6~I29WLMqi9+n8$&<le6;KA*lkn24}#kM%K`Feip9|l#{tM
za{wtt54B=8hAnFFPojiSTr+teqatX|5h<w*V#`XE6X2|x^yqESP_k?r<<y#RFB!F#
z>youLnExMGJ4S+4MY`XBc47zBVJ^&xGC1!0hP@V~T38M>jXl%TFhJtAx6b|(0YoPH
z<ThGsMLz~gNjUx&|2OPA6+Iqe63r=|M<D%o|0U-87SR(2W^i)kxhC@e;J*g_H@?88
zqyO*z6QqB7@qKWEJ|^t{{!h&2ik^sWCz75ha3mAlY60fp#p?n%6@20e9gh3$CzS0S
zK3;s5m3<^DF9}f0>3AFsx3528d;5a(`!{$$u-$j${f@f5Vf2pPUudr<hoFF#M|8xi
zPaMyWFa-kJcH>4HLY1S@Q^TRnf$&&Bgo6kYQdR7?xGW}pVcG@fd0^y>vF$*pEbF!d
zP~2~CfBQ?pW`mhV-uFbRJEwF1j0o<7Zr~LSv28n!*8%IHzHx_0gR9~M-MLi@CNgy2
zpfW>Bb)cOm?r&T}=~tjZ&L%qp`gxMjsg<zD+dHQ+XPLXt6X*?oKClU(HZY$bJYX@P
z_ji<SLrn!^0B<>?ohRL)aWjv}Dc<5D5@*AlFu95`*|j2R#`*0V#5**F%QflMQJpmr
zUNOO_6~hcriS7X=shT7V8xVCWTT+fdF!@}?BPU5G6))H}$aZ6tZ?!le2KvDRXO4~(
zTLLDbHufVc`zZjzWuVUi;P2;!dTfjaU6S&Yn5gQ2wH8wx{Zl14rQr@~M_fRx6XgX!
zM|mGiFm(sMeEG`5Pf3zO*uOole?IW$TmVhi0l<wbNW)yHo_ako+KXG2O_G28B)^dg
zO8nf+zKOVzfbGN`t+OLM11MYN;Vy)Tp5H%6oVwuI9?+CfYlY^Fx@VxBbZHxN{avE0
zC31Mo1~jiE(auIXC5}IxkfVZmVm<~h+oZ^Q<ppXSn6zX%X~PiJj{|NaCPt4YJ!87(
z{hiEIK$SUhaBv;#b_I=K#>&c<qi8}YJH~84PDuL(=^dCSL-b5}Vb6h|zkI<~*{9xm
zCp2?8e<jjkS(Mes#F&oT+YZeI$Mc2bMGmo6!&d3^_y6O6hMGw>+lW{XIL06Z=>O;c
z6mny6)%?Q)*&=qPQ65~k9dinX8II!vnj7|Qha|!4IIz`<x3@Q53g96<j|cksWIj>5
z0Zw9AAlPbw^a+fKk-+y+5->WXy-@ZYUIfFV<HH6<d!p7mJvg=E`TPcPLA`we0Em7^
zZw=di<6xt~ouo{4oGL_Fm^1r`=a-wOK#=c0v(NtZf~JfTim)T#=vV6)XTx0-sZ^Zp
zz_bo~_%#ImO)rIJj`QP*ZQr0!?6-|_K(I$R<s8jV#!xk|nsPdDo-B)*b5NHJN?7;0
zq0}AuerMXGopkE9iJU9azVXrZRid(FgoB2yn3iUyi!k?zalBwYL$j;sK#;Wsx*E3(
zkP#tQ0;~owH%A+dQ`MZ&m%u}<gjwY@MmlFG>})8S7@#K18cEJZHt!RR`2&E#)W+Vh
zA{kaGkWU_Bx$Zz^oJNg);@N49Cl?UaqGMoAL*6!;aBPHU0})b~;j&>|a^XVmBLdUr
z7%VB8jPD-fGk<0D7ABmC^$yic-ET;hIZ_%Lb9ILY32c!)m9m6siuQhYzy=9O6^RU?
zFhE6I!@|((MlN7!*vGXM(HoKk7?asXsTOG67%(^|%C=KU66C%E&{UCigQtR-Hz*9&
zUYA5R1?Qt8*y<j}AF5(j)N;xdWn->W$r*3^SMmcPCdewWMU246jU;0td}r?5QiF49
zC%>!FC$ivrMk&h-oxe?PDk|7FlAbO?e}xcE9CPq9&`$#y|15NOOQeQsQe<7-<c=i7
z;S+dxIw)ry9_3C5p#*qQCuN;yZyl1Nqht;g1vHw??l~hVD{L(caqOocmvEdX+Igbv
zmD^`G<Vnb7dd8~(k}8h&!h78S7-TD$ZQyvk@OZpXZaa=&AE*Lsg^=R5Wo)$qp73~_
zxJ!kpqii>9=@!1G%FLoMa2^LE?#IM-e+RO{a>tk38-D%u8-D*!zr(g=C<I@A{)q!$
z5wzC}>W1U_LfNPSdObcc&l9tErUbNs)(?KR?-SDv&({HQVBZU1gv&(~ITvooq1qho
zjNA{&?dyEuzE^zv_ATOk5|UEgS8sQ0Z|`_}Z}|AruXsHV<U}Yv<qa|nw^FF?gJUbv
zTVboS{BM0J^oiTH@qtdjc{XJK{EZxuYUqSc&hx~PPvly8+ecpiio8vU8%95#cs{;I
zwPeB&hi(bzj`n!KApG*dOdtqLZth@(UV>t@0mHz@>xJ14J~X!*UDZw%4%0p>r!51_
zTJ#A-a;pd?J|m1*6WaZLM={5Iy#hzq$eQ8BF`_@Zl!TnPs<+n}R)2n}gn**-W=yIB
zR8dmCGMDo0E9`h-bcdyik~4;fGA9ndRlZ#c4P*Uy#3jRci75q^Do}S=l!wopf%b%0
z1FJfMN62z1IF>i01_$7CkZ{bI1@`+Jl`)!-BsRV|=p0Q7^A&Yqmr0P1M#yAA6`F?p
zHbo5f%EDSEArAr-&UgX@pR(c;auU>fgVg<UdutB4B39A8P<QVG<|KSar_yP~mSZ`p
z%p&qxuhiYyTLhqSL)qTGAjd_r?KC8cIehT@cZ=TWn4+=P{y5ATNzM(LB|p)P9Hv7-
zBn%_?4gnMz^^EfhRl}H^>Jz1!sZebKBdlmJlt}!5O-MZ3qEUup?D2&%bBMutgmGuo
zzb@YhY+|0=BsG<5x()=Ao`M{M(aKc1ltR21l4ckt-rnEv<?C1acL3}+a!&1h0lA{=
zJK7ldYR-(H*62ci%W(aU=Kwj9W@tX4p14^k``fb?q%7#e@pZf7c{Ye9+}?KNTBrg{
z1sMr$o!+GOy1Xi_ow#p&-;M}XoGlcPp_sBcaNC*IoVNm4be|-_D?B|2!`<uoK(0G(
zzx)GS1V=kDK3?!i^IolS><S{36WDX*b~QTgrNGV;fBf}3v{Zck`h`+v50#<=ITgHp
zc}IT<j^_hO6|dtJoF+M31)<TXvitj}kI)UC7SRQ2<L7cTXDTwL;cOixSG;|BW44WX
zI0zZs24UpEZ0)L@D=EQy!#NFe%kWZ|9n-^w8yWmm*9ff7$iN?v8zQljMyj;`L?H@{
zEoq4ztFNOWnS_eP(T$q|D<?B*-N+NnNy1l!m~AAQB7?1GXA312F;=sI8m7C)<AX`2
za{}Vnwt`YB>iwOB<$0ja_;ZJBY$dRsk3XRuC(U}V4`5B}wIZj4-Ws|QZaCWs9pT1Z
zJ_itrSp`NZ2+386q9Q0MU#6}ZeaPWoz4|Sd>zdK!hFT0hPU52)k-=_+6X!qASK_7c
z$+b|SNRDIW=5X>Qxh)d<^uWkH4mgicwj?~%@_vW+GYV4-ik0vw;nzsyimsUF=;v^r
zU?X6oRRE+ZG4yOj5hFs_03<EQV1#uk)RmJT8K~P1&HTQ`anc-EBqY=;xs;cVID5z4
zCff1DOic3~HlZma<q8(3I3(oB{M_g#u*t)D7+{@SHa#i4{BjiSm`~LG4&D#O-87)2
zF((|W(219bF=L%MK`K{1k{Lq)UAU#qh-t~OWnFR2iMduj33Y9`VAw>DO34!So}zJy
zs$S|u0b!V=f_=;I-qDXI>~&JLQ+AkCjQ#;hihu22`~&{QKmHbiXa#io0R(i8!)Z<}
z%1u|?pg!^K*I(h|#J1g_LG)%-D)}&R8nEvf(1OR&v27Wt3ci+|L6?sw_EG?!*lL02
z#5kGO@$rXWvE6odw3iGShFU7N+m6@A2POnoZ;<IIA`mUO{fpl)PwDZEsjbeA@&x6j
za(;`AKp=(jH!6<j$2Yv)_lObU)>dzCaAmpFc`}-&)QvLQh<w>^_h>-mWtDTn@p}Co
zE`@l35h;D!wgE<*ws#i5=Unhs8S*40A#H`tTzkbhPv-3>!Ds`RhW+ggmSO^#6Z3UI
z>kTG~K3^=d(2R-bf*uno9FZwWz!hAV%^YRhV2QV-_fC9tFxP&Lfoi}^Gh}W99x*oU
zb)p|9I|rN+yW3t-6!6Y`eoq2(!M5)t3L`&V4b0sKVXpoCPKD5UhQZExc#7b*Z<x&>
zV=|o0*9|U6S@HIT^3dK`u}5lS;&t^r8xIr}l60YYmw3@-4oEu!n@pE~M6tT1=T)Aw
zQfN4(^8mlv%_P)DD)2eb277Nyu2)?1v>7l953+S=jVNC)Q87B>GoW;Ea#P6>e{B=y
zae*6Qyww1=BAMylo(uvMUjAYe0GrI*aZR|4a+ny6M0HLX+g6a2KAGu3{0p`GaHH;h
z1s0lP-zwHB!RD<Dw*l)Nrh=4rBu&tL<F6U+a!f}m+ZBZ}-B4Aa(FMVQhI)7dB1$2M
zZ(ACF#OJXl^w>NHgZFaI$aO>S6a9E$6}3~E6@@#hRN!Qg40nf0f`hT|9Cj$Xo=(^H
zTKL@w{iad?-1r`ekVli~E-}D*kAN+1>xzW8+Z~Lqb6yO=9nJW~Jq}ojOL_Rj?|=74
zLb)lU&yMY<FC=(RU>+wGgaBM8pupSh4R8B~9!e-TL%zS!GcyR`eJOAF@_xUbL%{HQ
ze8cfNF>K<e-~5csKrqyjA@_~4_Bm)qz1?ujU*Lm6jMEGy!W6gL8w3fD<AvM(1~opP
zxBUj|>_h%?yCc;MsNr>dU@IHsc+y+d8vgL@S8{;L=kV>zJNlUT`1p?28-DrCFMv31
zZ+HCi5C6!h{!$pBJ_h9VghWf@3`z8uM!SGc{QCPpAZbQ@d!t%#l?&y+?708yQd0Mg
zh7BzdbDWuQsR=0vCy|LPf*j^5I~tPHSvigq`ToXyc8&hrh|EB5k+sjnP%}=KV;XXy
zLP;W@e@sr2xsdp6$4TPUfGP^uU_u}SLln&$#3#1G9@REFN~zet`~-K$>*G(5(OEYN
zAWIa_w1(Cw4K1~zlnuUV_$B~5Pb3GB1U03|F^LyPfVAD9lauZ2Cmq3_KaC_|^Is+8
zlu2TFnQ#q!B0$Ly<3nVi3=*0``bc)A1R*@;1SC`GAgdaZ@!M`aR61;xVy;X~_d+4<
z%9L5Q__+#Ke4?EX<PvW0>u0AH-+W>Ef-ww0i$4INA6emb!{SOps8tkZpDkb9p18kg
zXz<(fJ9_K5-|yIWhI`H4(9e?_=>5i@BYYwbCah{Am$a~OLNMeiEMTgBw$RUOJToNr
z>9d(gQFelP;`n$$VC(^oO>Ze_Ae4!E$`mE67CTA&&#{r&bqNK5{{t+O!rSxNjS$GD
z3C8)tF%0+j_fQWR9O2}hfDMIfLMa=KYN^1R0n_8K4(Te^cLrX}Am@hknT17?p8^C0
zB9~}+3iyK?1X5V#x~gi`Exa~^FlizLC?bpud4=}BZiG^9Z#Z5D`q7X&L-6YTh9ntM
z1X=(%y*0=40e8V(_i(-gIoB|7@|kwCOC6bVLMj!dR-~kOdwauxV7tGuYyS8M-ovrg
z9pgNa^A1TBA`^~i%u|J=g4g#a<|fc>$F1xD3_~5;`x{IJ`|S<8aDYfjAvqzHg7fi!
zovheQNl~+2GIPy|GGjA*x!s}f@1a67?6-Gpw;Rv~e*e3F!q=a^;&#8o+KKNU-#7;Z
zEL*mnVu;r9^UuEl9C&>H#tZiT9?tW>LmlVW-~JZo>)?y*fVCGiZP1OlT@m7>V-8qv
zjHieXwPg08&VE8m1-2W~z9Z#|^W#^bcV@7Ufhq!~?*r#~;Cwvr_Vpe2=u|#FUR<Zs
z&i)1`)HK)RJNlL2>7cShQVRY7D2azYLMMMKC1jT*2Ol3V#%&~L_DjmlLE5+>xCn86
z0k<6B!o(rb5gb!hN|_f{Khci^-3F#BOf#k=x{M*026>*1!FXW!o1~0V*?Dk09~g~=
z;Hd;5=Z+vq2+hagZRg1Y5jP&{HgWcL)pD|4F{D9Is$=Bya$-~h`?CpZ@Cf7m5_mnR
zidaOc+0yJQ1mun+XUq}*Z2aG81M)n1$!{B^=F2;vE7wXjjDDR{TuAov{f5>j#@QeW
z7!(F7Y#hvFQX!0l;7@#TOY_Ogk#6pJxmKY}am}%9NGV9N^D~_T_WA%uhnZkb;Q8@E
z<&)|^Hz>}@12sM~vk8xWUq&$a1&dwja<b0;n5F~9T9FPanTJcN$X<YUglwCeaY`98
zps+ay+QD2X2k2%;f=7E>c5YaGHcVlar^Vq4kT}+)sNpd%<<+h6nVGX84LS3lJzF@W
zBP24>wJwt3nD9tPE|rGY34!LFr8O;b#%?ztC6t{ZeR(SY1Z_6RG>ntMZl$vFS5rdk
z%<r0!qC0bj?03}t9RM#+wT{tGh$g%adSqbS*oBh_6rhwdXR57aC&_<2aJy}AO?W<@
zAp>r>Wk3>J?D`z=Nm+BLJ951PTA*O^uq4MkU#Q8E_JZI2?mMyqs*d{gD||-AnXy5y
z-haaV%U2wK`U9SS{2g@LAtLzxr$2?{UE}<0FgKLj8+vazUk%kAfN==0fh>UKjJ(~^
zUQe|Cf((U?zXQ?Sk4CIY6{8=RZ6J+y)oG5^c_6s@h+FTd`;NNbVp1WgDOrJb;(Wa@
zj~89KbJB#d?bz?{xW5CBZ!~ie6Iblpj^u!i9;Np*)MaHEjJrc~r8`*Tz;ch?Wk;@`
zFhgr6O3v8!ir&NZDvJH)czyo{6yD@z+mY&q?o`gOdXE!y?>x9lDVV~SVnIR-*h`p7
zVyw4=stgyTZHJbPlYW~hiM5q!b?J_QexBHGRJd4g5%(}c+S%cAlDJg83}-e$Viu7}
z72Jipb9(Veu8ad|uFNWD&#Rd;(?~yMwd%yQVeCM#sKAMe7asNtZyv(m2XWAgAW4ZQ
zhzCgKHN;6Yh|jofuAfU}b@N6oWiI4!l;6qdjFpg_ks}hp*b>A;a8736j{zHwYM-rQ
z9%Lxoc~OX5wFFz;sDcXTF~7Hy{sCe-fHGR-x>8f&hTGefePkniA#j+5GHY!?(6}V*
zns~2b`-qBArOM9c;z1N;t2897SPetKI}!7dGCDe5&nNEh??9N<7M5>wz`6t8I7h^v
znXxT|3CHluytx#3OE6&ozbm15KoTJ`a@fJ|Es;Q_a*!X!x^>8Bt>@QPhl4rZZg&8P
zh!)1P4{IR;VfQctNj!{<R8Jv3YdsIZ3CUSM<8!1Rr__p+D_)NWQm*Ke(KjU*6nJ!h
z@XsY+l*aS<LUMr)KqR37*zWH%;sti^Q!rjHU=5#>zAaH24Ub>{fLsN-ZIs4#=C7p*
zAXm&3-+h@39c%|`%@~mzmP^66?;kkY6SH^ZyJED4*YQLy6}SDyjmsuVs^~VOpC#e-
zI`H#vf1w!UbwYFD1^l<Z6d;J8E6UbTHy&1VB5R7CQ<k8ZDpD!fG-JQtk)z(Ss7B%>
zaK4_*#SvlKnke@DMpIWz+bIc&Td?D|cPfTRbUJ*nzd&+9e;t_TiJYUX1(8Q|B{333
zUk6AnOlEbXbd>3ziu-*>O?=5ECkzL?4{Y~4i8`C;t>J79K1cvD`j^)KBobY?6KzhY
zX4HtwF58YV9nUDsx!rD*4uS_^y|b#%%<%a5Ah{e01)t0ok_f!BHZW>N)&g}->pA!d
zQ7TU~F1ikfb>qWqT+>q_?kfmg^dLD*!j1>p?1Qr7q!@8<uStHXhFBybmm*6dt&GI7
zhB9aM4AYUV+Z|d8R{A>=RC)199F875NzMuGf)cG#n&}gm9@$kPuO&0VX*T7=Y^_ix
z2|gR!6asg-9L{?v)NGPmwG`bBlAy{0DR}%k;T&cpxm+qH5kc0(N>UZ1QZa18#1WBM
z7#4g=U476IpS)&u2gaB%r-6~~ewy2IDpCDuSlb=VY8mj3-UpEUs+OFiqb63@(x}Hn
zUgEoDdbNevAt1x~L-9>q_@e286x~AR>V!$RZN;XnNbHD&jsb)mhkjo2u@<Eymzx~I
zc$#Uvd}`eQP29v)fm+FF4CZ~+%2*o0CM7Ps1>+5j(I7UF3R4wEa8qMCd)V)HNJ^-+
zK!OV$ts`d+cv!rh3Z}Dne9VE{?fsu|UANl}Cmdc1QHR+v+r%)(mNys}1Ub(MJWiCi
zufUB-&1noiQwn6KAErOQvr`!p``e8^DChT43nN^tM;K$R`0~>)93JOj>`!D*<$8nX
zjN2_@Yb2q*?Ku0u5z58C>7|fzqT!>DONX3mLaiI~^=pMks-GklsHlc!J62|&bd$QV
z`Dxo37-r53#|95vpYtM6iS+{gc%keYTBA%;_8JE~NhY{K)lni7h#5nKlBll$O))&N
zsLtpeK52c23tu{wOMVDvV+4D0c_DEF7}!fe-ZvnU6agMObJCQgApEj&phOaqZpaQs
zM+2ytKPToy!C;E(Xcuu0i-_Ldc)&-nm%BmIAla-ripe7i0uw6E@|r|qvX+9QQ$p$y
zs+LN=E=_UAoCb3tA?4fvMmMMpUe3mR9uY+n<~eb*APJ(Hc-UBk3&sVq2;9}n4Uq^x
zOjy0c5cVk;*iq{q1ubL73q+UlJRAC$$Xi7zJ8X9J<Ge^laY8=R6{(~vt<u`b^C>c(
z`163j?{b=thSm+%8Er5P!AB5YlC&T<Tr|C#<N5l|EhCCs1UwYg!W@>U>VbLuKJ0&h
zFGsa~hPwsRJe|)hLkoF;u3B3MDAiR;ctHvv4m0*T*ktCuor<HKG&<!1NO%)eu3$%r
zgB7uL^7}H<NSs#qT#Xg2@`?*1!^7a2G_j7ga`;mNBtj_Jut!56)vf^t#xs&~MixPz
zp@WM<*&Ni~r-W_eMq!~e0|?j7h|yVm6hxz+SlPAQ8x(^3?G3l>#$2y>hi7YX!2l+P
z9B~zFD`OnbzTfzq&WYpkfV!de&IIRNQQp4<)HjhuA;Pn$E0EsFT|g3a+hH<MWG6&i
zGPEST{`4JiLwmij?>7t^aPGbk75wzmPk23_s41at8_wgv^O*R<@jKL#k;S3rIG!(h
zOe)oHrI1tH_ZuFMCp>S+_3ywbe0=*8YOZMKiO0tWwo-6=zr%BcMrEcj+B9G`<oy;e
z5~oH9lB(t$QM?LsUNjygmc{C7q3IoywTcRw;sBQ2q{`KpFcq=VniY~NR5<-w#LAje
zMN(@hP$Co!BB61gaPUEn;y<i9!fBK~j%Y;^p#*SAH|^5HQ*vh`4o;AxAhu2Ag_*;r
zKytypZ6W&`++Gs1?&s@Zgl@nAqzpL0XxWhWdl>BoToXvyB$SMp+?TY|uiC*D?nm}0
zuvA2X#dhQ6F$eYjEECzTu(31-L(*CYCbtw$)bVQ&6O8YlKf;_#=7>w}A~B&u!fbWC
zV&lj-UoV`G55NZUje701DVRG-V%&z$j@AzJv*Vr;B$o@*F8j`mCjpg5i`-81j*R#W
z*I+IwoUzemO9LPe?&pKi#jSO`y}x0<-8r8tJ1-UtwF*^cm;h~iL)mv8Y;!{TV0liY
zf0mTk>gUk|;6J2)1}Bws=2qf7AQ#jpSTbjD8_AhN*c=d}$xMXrh|M+W64IH$pCCed
z)8a!!@Xy-KB+6{$GxgYfaaDYtshpY-L`M!O$9t8+fp3E_lB)3Xi+ym!`wK^?m<-+r
z+Bh+jMpGbpA>@qnY=Aj5D^{9tiv9@?W1TsbW|9)U2$lXFH^Y_^tP+pUxljThPW+?-
zSEVtrGm^(_LL&@w8V!o#j;!=DhyuNh00`;J$f;txSHLJ`_c_=hKL(0!m~#;38UqN0
zlTX9*+jnwyHKU&sxfXgBROodwyP(J!+dbL|Ns62%{_v+?@yjp2;P!q8ykeN+^?cC(
z(oXnnIOc?>4D^n9S_nS=#+QP(6CaN!i`!N3cHdFlfzg4KFb9dfc{UgX_3fPzBW6gV
zp?mLebBuOIDkP^ihC)49yO=Xd<#v%P_5bJNh4wm-a>o5_=gS$&7sMqnT94H@Ogj*2
zqMgg|u+U&A;}f6>GFjcnl*WL+Kp13<WexYs*ktYz^-Gyy>y1uY2rNS4vWBcM>iMX9
zb*ORropE519Ja`rSGZ;>Hk7VP$~h<X;_>sAn8+z_%owc)JTO9kuYfCBe_e9T*KuNu
ziEZC;dw*lW&Rn>(K~rJ|lk<T@st}><G?xU6ix8@GZ;$@jyq41^`VoIV0jJ-kyI%>h
zyd2|EcSAcIl2`(Aygry0G`CO<DP_Rop?)IdqnfboDMHcs&Cswzs8Af3GLh9HeU?At
z4EHEQv529wGea;s>prt4*qCrAdhgI=P-T7$M6th77$GXaVE^wNj5$t;${Sr~Hc&E~
zlEIy3u_TTn3WYJoB=Y^&rfVJ>h$sMZDWeZgeIkr*4_}W>2hQj-Un#e7IgCDO7EOiI
zY^iVvpR1I9kmRmj=9u8)7}9{nxGlu=E^LBxWri5xniX?R933e$jzI;UGE5kpsm||7
zltW=knf{3}VU^F=oC8yp7mO%Pk(p&x?P#z$P(y+ZJ`I==4bI#zgXS1c1lOdU@_t=Z
z7OJEr?aU=%Okl5fLRrQtS~t7=NozA>hFHcH(bf`5CrYZrzY}>%;)BnekXy^MV@?v<
zvfqKxAj6=8gJM#08d~`8Z93!bohG%qZP%Y!RUxzCAAf#BtwR3{1PaA1f-wyPVjbcW
z-pLQzFns*+kAM41K~ut)`yE;mYT*YoLD&b27&FJ~bz*;ihxdVT9E5N>TesE&z3&?j
zY6r^ghHby$^}FA*m_-=ZkaI=eZ?HM>?N1;0{(PZo!TBU1Q*Rabx1G@sk^<uCm_i9`
zts7Li#aJYrUIZfmnM|N`Cp5y~Dz4D8^&q|np@FT&17h%LI2$WE-GL5zHkQnE_1bRY
zP^AOWO!*lJ)le3}e<=d~&E}Q&l5&QdA|a!=p?6*+t(~xMf5g6rVq@Epa$>JFoQIDv
zSWR?9RDz=1?&!}4K0XL@{QS#rVd9wW;H0<eF8z}$5EvW5A{3SNo&pof5VLbjkeo00
z!IsKqsa5PVbZ}{F;7;Y3O?YpSt}Cc-J1|R-t_X;U2gw)%(+^(Sl2NyU6tNL3PUj_r
zRCs;o9g+%kf8z_`6KVg7;V)Q!L5<UzWqKd7_=O{ved8hDJ8JyYE1E-73fbd`=CB#i
zqQQH|@p_^sMUM`U+xt(@{SBfSC3kpx5`r<RvP!N@4W0(=lf#R3m~~Ff2KZ>`A#vov
z9|x=?KErN66jv>Yum-OHjljX+s4fCTQ#iPlo25k;L_D{=duS1a1~lsY`*}{zi-2a7
z3k4)z1KJ^K#z|Q7#akr7^)tr#4c*@3ePFk^!lM|7E`Pd?Bcawwu5;6k3zX9u#6s%Z
z1zI;q77k;x({C`@r)^-cXVT!mA`T>{0#EFsmP|Ee&hZ2|&xxc3$Lkp!B>UVkC+4e?
zvoOQ9ansdYkjsX$RmiTe;|ZSwdu8A9>-D1ael9NsP+~|)Ny3<GXy&M2e&V}i1LyZY
z!6o6#Z+?cBit{|7h+*6qN{PYuGN%mHg2(p<Y9-wIx4#rhMZMkO%49~(2`Q2=B{b>~
z$JrW=-~S$uj}JH^VR99IR2Jk6o2Xl*nkiBOZ*T9=!b9yc*aQUd{`1e6)A8}^pZHRp
z0{L$Ehu?g~Km76=h_PTqa-#Jf^Xf((jYZ`tDa(Q2F%fe8wZQ<nTrwvWBM*^NLas4=
zL{z|Xt0v_%yCNb~LID#wMS@uSgs#t&5>^lGO37TY2(!1#gaY9~>|xAFS)nMv)R<1g
zT-OI6=O>IccnBo2dQ?q;4ptXH<qab~doyw^;-5iwR17Ir`Hs0q^a1#vM~C-@BCOFA
z{Zt)^*{2DQBRj6%fKXB`C$R(2l7OjHIT_7T0&*Fu>AFDN!iYA*xTec<P$<7V&)`Bv
z==&G1lc2Q#h1pVy47Y?m7xbrMv?E@2i9sMS=uCnfPl~Er?0<of%>nNR+&NKl+lT?e
z6|QUtW9AVd$*KAYA-Z8qhc#y9zyS{)1m+NpZu80aWyLPLGZw|np*4CumO*V!jA;<#
zMnl;(hmcxXw&W5{Z_MZz33c4)(Gwh}cN|O}oDo#wpDlbA-5^GO$fg5J@s;@rrAU9(
zI0MY=bHYYk^zIN$K1;LnKn~Mn=Dq=gvIpBi(l^pwi9%~Am_m<eBc9a-<}99y0}Cmg
zaZU7d(u?DL#Q#XHaQ2DmjfZIbvvaZ%Amt2Qq-`mve&d!jN0>J$oxY`nyYkXI8+n1(
z>%@6Bh;^t>3^SZ3Rg+1nz9Y}bU&BE{eY#yq#{E1{a)!8Jv_=RRz<C@f%7D;4Hi<D$
z@^5261OC(``02|#j^jY<e+QyDr%aX6exoWSL=cn$DteG%xA!-+=PNQ>3R}j;z!teW
z=D>6Vlv1-SboMUEltfemzy~Acf-hgbgkG4fS;HcmUlaP-Q1*@4D(=w(%=PX#4$K~F
zt{A-Y1!D1uWAt(To=KIHI|Q|`jVy!GK}<R)R*2f1M(SDwITKFeldSPlH~_;OhFn(E
z1Rs1KhP;uK>jxh&i*PuD;W(agwkgC2C#V8x%K<HbllH;#4y>0iF(TV1i+T)p9j0~L
zu-|SF_@&^gY4sy}Twp*-9)&M}`GgG;x855M|2g3jFPBrQ$}J_P#F(%o<aW!R?{mOG
z63`&1C10H!>I5zbx-3a6Fg?5yOQ^8mmQZ&19FSDF?eRnl)r&(q^KX(fX4;n!Amvc(
z@K7*|@|$?xEdD(qu+QQbfR~%kFpGf-qf*FKv4cDYh)5<Cjqk%jS*kO$XwHdT;}A3+
z7Au&KoIp5koQ`92g-FHJ%*?CUNL`57C;2?oQ%)!=&&UQGlL^Ed2~(d81;p%Fb+)p+
zCF{Zfw|OT<#Mq2Mh|+w<R?V#u%C2%rIXXRJP>{H6ek87Alk^dgOTe_`a@ZklP-v;_
zWEmZX6MzJisIfhSYS4oL6os`;=d?tPXkvf(GN-8~NC^`i1nvz^A~0rIZi*}r3qgWW
zxM5?2=?~Z7%W78+k;MkY3H3Q>^3}3oH#+vm30R+~(W=+ZM(KU65$R5|=eEVb>dqED
zpV(E<BDSA`7!O&vvv8?5`j3=nv$^BdPn4}-`{^sjn3yfx_>pDxXV3Y+=<kZZ%D%DA
z(mW<Gz$Rr@S-B3Ul#o)vZQq%PByt6Nt<|>Xia-7FkFeu_wuaf+L!ishb7Aix5ewYA
zCP(^m!<-8z<(Dae85;uTh=*r-6nvBt_~?ZE@WR_*DN4ET5Mes!ygCvjdZsh_oS%lh
zn50$dpyqSJq|BweE+wJVf)Uf$9D{~EUtw#+F4O8Ah-rvM0gtLkB_<xBZ|t58C04wA
z5R!%|4j*%pj2`^}bIdDCeJNW?W!&`W(eK(v^xryGNs`&Pl0=zE+u&n_<k){8C&8Al
zF_B9osxGNx<N)#I_mE)5=a@OsF6SwJ(C52@#W(>+ejb(#_L>3nk@UqC2|<!mFLld#
zp2%ZRTwqkPEiF_MyWqk(TTH(J4T6R^9`NBr2&L5ePE>)>d8lfrfRb~WlcdN<7|qZ>
zzSEHn%4)MJW-c@ag323fjjBE@7c;ph!^GJQ10+yW+MwGVQZi#HW<=2kyt86JQvA$w
z4yvjkoXkNXfX&ZIeU8q<ZL-CTP$oBjn{gSX42NN!FNm>-%ID%`C{>=NlDxJJKtPN!
z(XY5>3F)^3JlwTj7si|eZt?eMi!7)~&LWx(O(=0y@V{eJcjTcZc5Gvk*auR~X1p+5
z?Ya!;iHW$83=RNhkohSf0O%}dcQy@UG%BXpI7TEWCt&bkF45~@(=P+uRx30sif%ZC
z@68;8{U38+)!;I|ohj6sG7ESnFZ5N+RZ50uJ~Il&wamsQImrpn=PSG$%u|}pFoc6h
zzTFtzQkP%oZ+I!-?0m?a4{#q7QVWSoP&!!>wc514lSD1~2VB_^F2EXfnof5r63%{J
zT4*MVa{3<UBjkO4753z^!NZyCafmsX5rUQU2uDd#Trx+*S1_wZwgM}Fq(ltgX)Kx3
zshgey=W$}Y-Klu60dFi^0f24Wp$e>`p9Li#re;c1DO~J*z?6CC2q~V~#OzegXpmYX
zNXn8q%}S~;DAOe4&@45fZ1vKILo;>V#&lB2g&nhJ!|~+7QJhJwb2?J4$n^$I6XSTm
zB0hZ12{8qvT`pZ7(vfczW}g*(VKa`{X~82Px-tYk@~Kj+z{_mppwdZ4Le9(=T0Id6
zlGS6Y%(bAFEk09TKz!2?GRn$!uE{tGP3a=eQ({h4n+@aDV1xa@Dd$MQWQdrFFm-l$
zFT&`jWkcF;7Z$&&QDx!sbNMxBl9NRcyF;Pz0w;X`{tW^pPmy=RNc#;kH-6vkVD03j
zpTUKn@w_}W#>TiYF!Y4Y0xvum=A^GC)eY88@>40r<`;%M*T{i70e@VDMPo^Y<L8DM
z+!|(ZJ{~XXG>3Eq-__$83_c8@P-+ewFO+SEZ;bDdl#z-<k%Puy?ddY;trS)2=)qwK
z0Tstkg2oGkOORa9lly?11j$UH0x#kcc};;_@TH8HmqkwV|E`o=6~3>W{0%(>1-X0<
zUA~4kaC>0zL5e{^um677Qb!43pi$KZ;m%x8%EqlFCtw2aU%mjMm=Kh@VZYrm&4YtO
z^omdwn4^ej490J(2-5NbDSuv64(v7>tn)MC8$-D;%sx$of99{hPJP5Tmk+E6By+M!
zi8tVECmam|Ty9h#9Kd}<1TC|RN-hvLoX3IVe4+jRSA4vlP*L2rirf8;FTece>U=gQ
zM<9uWPriSFq=eB9O3jXwkkn4Qekui{vk5CFO6h!bbO*HVNc&DDh9vme@c8}#w@FE1
z;*_fZV;m9@Pt`<##|U-+%Ri7K4tHh4ME~*VP@_9D6?!l{9@et9fLH~giaaK>)kPtt
zFc{DnXFSIszYv8pB8ub=L{x^w6rC%b(&;#1_KmCOXvn!R%qvCS5k8YaTO-`2DU5{y
z1<_fXkT;FXszCW8V|GrIXB<E!<*Ny;wSE<_You3NT&4m=5~X1gK#P8CHxkN(r>LUR
zjSu`10`M=3wLcU{zAmL8S^0DZ#DL52XJOnr&jX`%)LfBk#+A8b9wjk~l6R_0u*`Tu
zhz1602hQVwW94xr7|5yQDHvm-1AsfWaP+FuC^sO`x(7c6q&+y5^MogXWO`iEcH?J1
z2FBwFYcGg8v@0g~Gp3Z_$e7|0R}m9ABrbKsG!_RzX;Pa)$P{LL4tyC9nV)4gF=q@p
ziDfs_7@-0cxGAR3Ya29Yke-U7<HQQYRG}-_B_{Pnr~;SLN#^85k*Ff1Q=C7)+T?e<
zpgU)1pCEF$g80u7M?V~Zq)XZ>8o!Uq4}J*#QcLnKz-Np<n2^xU17*Kq1ymx2702@N
zOm4}l!JRSVEZ6Na?I|3lKuK?b=n19@lUn-##tKL@NWi*$hU7|tp!Uo$8vS57C-!G(
zK(6<lvenjs3Aj1-+m8OCe@j)7>V`4+^DZS|{(A)C)i1wLqBtXf4T|Z^0P#-w+KiW*
zYp4J}R9lj~Lbzq+s-sg-W(Ox$d;j_~e)EsN#kTKZgi0|n7~01-jN^rs)6?5Y%+Cf=
zqFgM)uQ&*Xvo*})!0a7U6Wxhvzfwj`6=jjB#4orWmQ4BO94v7u6$(R5oQ7KK7ZD2Z
zLl3$RqCnj;6l`na`5iEU8Phf+j={r32T($5?ij73C>zEQYi14-ZK<Bzfl{c7NhM?S
zhVgnpju)&Q(H6#2+$AC3?o_Fa4)6TjF0!xg5E$}gz+Aj%S0s9>gipldfDP3QL}IIp
zQknI}arVY?!n!18+e6SN4OyT68i5tF@ZJ#H061X^qVFL*Fq9&neCa~23gw)cGIMx5
zB!(p2xgoUU#PN#UoD|Z_?G7ni-~IKV0Z_qyWRZyRPSJA>SSmoKXFzj}=a<v=v>tKf
zoF4ObyCT9pzBhrhGXjA7hAADM9oX1ekY?}%;VUS$vV~3MIvBld(8@S(X|Ib<G5d_M
zFJ+IU*fTCq!k#b^va_AA8J*z)xzZrG1O&!cc+z)+r8a|Ku3m^-*gf1ZW4JJY5SS18
zl;ZbxceWJ<Tu1{Me|=Sn?v4sbfjKw4AGf?2!;B&zNTEScVi0qH6TCA{8;hHa>@URU
zv><ItwjG|BzHp0<bru!z`PH<c$St7dAhtoIgAI5Rn`1oZ%WSI>hw%iaz{eoBS8^y6
zqbDQ;0K*K!7q4Zkft88(uWt}t!IhM)I|^Di%-*r(ggyu6<AHg+=%~KmF(wsUX`vVR
z%((iSUJ8N<F(9pJ*|@%s-Y|Mc**0h`80QHYJh*HQb{?Bzn9y(~B*#)3KsLOUjQ!2=
zHfS8mGGVVLh6~Dm$F}ps>*ovf?+DJ<1IKY9=Zt;dFgvHmTq|r0yp9*#CgrNCkh;Om
zQA!O*@C42gAzy+$Ck#Z@qbllFNo0?X-X~NH;(S<C^)i&?QbQ$S@We=97z<TW;gQkX
z!0~EO(BeOWz?@@VQ48jb=}mzz`WWc1ClF*LA*Fbz7YE^vapL-W=8saolCi(tIB~&(
zhv3U;q3T)1Fk%j!gT^!krMjRLbHP4z)^?u3G2NMQeW$LR{R%6(I^_h_0(UxiQ<4ih
zm}B78PfnmQ;ii<LCbR%HQB;r+@KoXoKGh&4F2s1DVI5FmW>jo3YR)p6F_G$qtQqHV
z;_-OGCN<=(aobDV9&y}MmuT=s0zzaSkkqhA)1)fWUaz}oP99Pkgt0B&$qXuC_A5IA
zYs3WAE&>k67caEk=p-K#{`w%{t-!uj`hPTVa~Lm!N%Ed?5L7}j)?TAucpc)vBGgxT
zk-MY!j^!K{3E8cZm$5m~`wU{4&gMX40jKFJfG57=F(J>{F-IJ#Ac4%f4W1Vg7s&j<
z4*;JSotIsy?7G%0Ax2R^3lmVy;od)q{{?ZyNvDXPLluaGU2_8lzdM8<XmJ_~e%1Jc
zX-JMGE`JH1gT_|E&mfdn=2sU<ERclO+{1LnBB1Lc#OJT$$}Ng8a$I2Mk@8I5l_?H9
zBgJMwY7C=Hn&4tcF=VY_CndeSMD09~OTl>@sI@TqM*i}1{;%t3?kC2O>k@EdhIdv)
zkLh)b@HlHHW@}U+Z9B;~BQ{OVJ0XhpdZ4!#iYm7Eikt;%j($9`?K{jGa2`w~>?hnP
z=@LmuDbz}WP_;zBn>nnVVS{hTw>!#C6Wp+gejLz3tT2WAj$Z;;P@Yn7AFKq;Wo3i{
zJ|>Riq_HGZf+$+y%Iy233bX?<8|E=N6^bHds*dKcYob~ITK}0cwa}2K#Pao!r)f&m
z#Dhzh8I4S1vWK`wTUadxB0$|LEHj2+w$8(#w}2Hrq<WLi-pub~_8AJKh|HcW^l>v*
zl$ue!<4n_}GyF5#W-d}ak$i;$*12}~F`$Fn(0QCV+lf+Gr${J7&|2d~WfKx{Dq44x
zL?4PMSSuR1H9(S6L}1E5Xp6Qy>ohc_SgCqzB%iZh8;b`a+aDy2OJ*~gjR6d{qm_`-
z0{k4z*=K<uzs)Avqyp>G^v9IhTF0~#l26o90mlWcB~|3DAc0AzDY5&a4-%|rJHp`{
z(pOm+u>-S_<1$vFGTuA#zMMM$;l^e-o4j{JB}6aL4AWtRc}p1P;^6Q=Pc8wV8<NH5
z2Z82|a@ik-lNpn+&B02}8Qeo~M52na?;I{dfdz1A?CrezrA0UlXxaeH*Pq9oydby2
zfTaWCaexXgv&%eZv3(1iCN7(B*DE3;=Oy9@mD!bYfa{s&ckVEZP$tr=v(%KlOyt@u
zXhKafX!6&Q;AqT<L@<2PFSt~NmvCC`gc8|oVDxqm#akFukPva{O5$BavF%JpO=ZXE
zghnk4c$%ne`t%C@`M>_EOX2bH9Z4enbF$THWwczl>=<D8aDi4~>5T~bppgjxdha-o
zgF55LC7P3m+?V$+NJ&|l>CQ})Pyi%V+ye7=bFN&2P*yHMqV<7(9+;!Sr(nD9$T11s
z-)?w(Ja9Z8sM|!n-LUUF+zh>aav?c3A&J9BOU0ay-j0~K43xe(Obq4r6{&1IsODfr
zW@ovNsUqzgQqE|`gKoh<OIsdq2Q<9V%*m&xK~%%>sMpUl0i3O2gw(c=@J6VF+GJug
zFf^_;^bwH~u=u`ky~qWEa^rVoCk;dr%}|SQUTO;s5|i8!Fgs&EQppf0<W7QcPbtW1
zjnbUR%IbaMJOG(MX1`vSmmy~c<c%@N9d$+lGzGM5xYs>!>A}v8lSH`gH%Ou44I*(|
zA*7AG)UzZ&9PNBT#c;cA(4295f5$~p2=kI)JkVaxC*Xp8-<c?ja3iM_f&)qq=}`$1
zR3*F?xUlAwIME%C2edV4q5izJPA*3kutsvwC+tN&$XW~nG3}?M0ijBGYZhF_Z*)bB
zBPLQHhUgv#eRxh%KoIT&D=u5G9Pl2}VW(?)DZYMy|5n4Vi$bFDZ49O&hYA>~Ja*Re
zV)1j4A28VK8JIku`4QdLqn!-nJ5RLUP;U)se`8jU$aS!h)aUP!1XS&Wp^k3v6-CX9
zn|&`lO#68PZDOmH(rql`Xp}_>4RE>+cl}f&xp8X(1W5%oC$u@j+(`+##EF}dP_O}n
za*mhm=#0i++d@u!?ip)uA&?Nj4Sj}@R+xz30D6>!K^57_bsCuVs#)O9@Aqwc2aM`C
z2<v}Wm1?Nj|Lk9X-KG%w-&!l+WcsD#E2h|N;Q8@AGIXe^c2NdIod=^Z=EQN(XtuCY
zE#<SrSk?rd8AIUC2y4AT#=I^gQAG^`x%VF0ZoU=<FdkzJ2$X?ob4JlbMNXiqDEbOS
z9mzjK;l>~gqM2l-gfU8!z-K@U0gZ_OzOww`80U$PKYhbIpGcy}+s3NVkS==S`s>18
zh6PnHR|c@zW0GUtqFu2NE1WGwE)?M@H$ZJ%mi&5STUIJmDV$@384B=jNQDPCCl%_v
zOPxTjkg^B-dqRb!@@95j5H<&3gdRp<_9Y?~QmKK(kBH}Rh%ne~Ju}k`wUT5vw<tg1
z<R*eJ>LpBJ1Vh#6bVh7B*TiV_AqXIa>XIerr0t*a>FSR5df+%HV5nO`D!ibH10#$?
zf~<nx+LgoQzy&YxdqVto5$+eTnmHlK6q`dyDPBN+KV}pC>~K}ovZ3DJuBkjZNqq?}
zW}N)okrUXmL6}71q-PWfp}?F<kfn^Rp9{8#XKIxzt?Z)Z-<iGBxt*xETLwV@0Vgh%
zAaQyXg~2G|h=aw$yFgEa!y;YS!9;=U_g*m={;5{uZzRDJkyApzFw+e7q7Xn|C+fam
znO%uGWVrDefJKO((_b_DfpK)q36%GDq}v_z`O;%!{z7}5aPxqa0R)6J&93dDR)MF0
zez1lO65X^`t%w;7zfWutA`nlP?;+NC-qoteNrJ>T2vVpe@i`pE_V3O}Aou0u4=z9w
za!p9JP#WJEV_~hs6;TsgFlTIS!Ywcf#W@!2*Wdh7SQBm!hiWa1uJyoHpU)>+9}rFW
z&Cfp}?;EE@81sPo0h@H`v0!7rIuX)JFC6nKJYj@EgPKuY)H_4H!q`ovgYt!65Q{l@
z?$ShrR8Vqebhjqt>`aSH8Ng?F7~?QVX0Os17l8zE33!lUiDec1reJg8_|qSV3+H-q
z1!fA*RIPA2B(x@}Tm%^Dfu+SKtO0R&<Fvep*D(f63^m1#k5zud;JvLDCO5*d?>iL~
zW|-$et~D!QKx+d<Bmdr<TIaK4j&G2>L+<a$TSXodIYR7UOM(XgiS>S}$e|8d^4kQb
z`3bm)$~hg>YYM~87OB1_EFi+@;`0@SY!%RnadzZwgGeIIzM|B(8r$FtshueGjA`Cu
z;-4fpQ>NA(2oKE4M~adkVJKR-EBHN3L(PR7=wuv4%^9VZE7%btu;?%*7M~*MFCi8b
zeg<6mQ0s<$-y$a}W2LidT9f;_%+_+xoc?JRmw&j&mpM+UO3V8VaBdD_5;D@yzm_SC
z5e_+&Epadh#O9OG6C3Cdof=C%&VN|M!k7n?no+hb@OH*n11olYl3@0gXSQlsDFE<*
zAt@Ay+^*;VS6$E`7*{bT{CMFw4{igRaC>_Tk4?DmB?hPn#~baWF*O(Hwn6qA5Pb-<
zQ^+vJgrdcgMYz_k!#{}O7{EMWM?*-FEx5S&zSaS5HleN|vM^wJY{9Dqj&P@H3YA!#
z8*b<al|?H-mcJ($-WZ(88bm|_9&bGeZbw#j*z`W&W1_x&i5#%FIOa-{Am93zQq6xQ
z(JWl^+^}yIl2}6G;-ECL;C|mDA0?DEickd&NZyc=<Gr#RdyhVYTq{YtPP5nJb)Xjd
z6eJZWJR-0=YN}U)s=0*xjNE!Il`nXhxazhemx>xjx`pi_ags7y3wOx)ASh*B4zUwE
z|GXLpa5oadIg!AgqvQDaj`ny$^#t|`z3-4&vBhD~&XWyKY2bFhK~iBpjyr>|7TG)_
z9TXCz<hWjT2ILLEsTrSha+QyR8R7j%Hp#(&rn6}?6d{I``GRRF>|+>o8Mi+1`u;2G
z+3^1L3kkN&a-Hjt0jR)A2~`*{*4lyNb@Ig=6GJ$;YD_;GFo<iYhsMBYgOknsPspkm
z#}o5>kzg-{8g~<Hw}iaWpx93um0UxG0}_?f4d=-$n%N^$NEls`GK0jl<dEXAyj&LF
zAz+p29GteyFb&)=mU2zW4=e};!cE454223(NUv$&jTmm{WbN_dt`Iiwoyt5>xR>x?
z1g<^VR7N7%LrSXfWk6Fv%MMQoX}@23_Ziy+_3LrSIlCHsS`^5zQzMrRNF?`toamFA
zbgmVyvE698T);Vn%YCjC;2eVrI2hFOxy77}T+vbjQjxd-KD`sZ+}f_7SS>C^qBM*e
z$%ve%birFGWv0twL_c`+B-nDLJ9&UKnF!k_N*}OsV!R$y;ehZVVgh#vJnO~j2@*6F
zZtza8l&=d-NO(J<I4R_*gp+c>wtn#kJ|_h8f*6-7Y#C<}eb=Rw>$Ac7jC_S~0vbua
zz(NPfyY=%R!cZ$r3k-(dI);D4?f%B|BmRD>8H2C<A(--ScPS`PbHRBy9*-A$i*Gk}
z<&KWGD2PF5-CvIf$=8&r`HyIUTq<f!ky@yzwP5y+_Ikl=qUJ)!Yikf0*E*UrWqYgF
zSTv(7UxqPNQdY5ygA;jp8+a32kZS8eAW=0Vlq%)G`tdoqEfU!IuEJd);>cTp<N}?H
zyzQ@(#C3LPEkV{vl-fA<N{B&$s17CJoO2vn77bl<MRLS-wY2vFUI@0xzQA=O6JEfT
z@--Phl1hO)BO;m^APJ=ezmYN(OwVVOn@~bL+7stXpc@}Ho{jc-D7qnqwp|6(ha1a^
za@}J3WCyZq*zu>pr=cB7-w(Wf{et_pLsjuO4<3@j7b_VHkEm~nC=!|1W=1zPWy-VD
zV0bBML_Uo~lg)KG!p+jYAt&Z0;SytT@|$s)Xd+ZplrCg5CdVRWAjfw)W|*dyZ^dIU
zV(hZJ!vhkr7%KXljNfp+yrLO<VvoiQIVHdDkc^r@BG?d;=;l?0ZhVe4t>})=sdm}~
zvP8TKl~Bm!sOC7Km;ucexNt}yk!#T=rWa{yA-xm<vIr#Nq>1qi%<!?O5IS@v;cwv_
zUkL^Pxx5+Vwj|+}4nZynd)}hYAhMHIM1{%~H?u?+d{sU8h?wAmn^2`U0AjFnV7xvs
zd}5qW%*H&Rt!B&|O?*Z~V$M%d#fs+kIoW?5ZuP}JU(lh4_yV|oKE}jHq{WKE7X&<-
z*-{sut6iTR`7Su0L5p|b`kVn8QB%N(0pl|KGx!}{Q7aX0+{(tF`Y@*m#$5M+`L((I
zxxfCpOJPn-Cvh@0alKC&J|b)`E0a(swc5+Gu%d-WV=I!F(>-a1Tnol=(5(l*Zq5-g
z3p^}&nCD1C%_YopLL8XJjNmv#QA<UW5k&Fo(2cwm(cnLpufi@u)Qs*wi-!b1{QnW~
zLW;j{oF~-Q>dT9#pB>&K<V+RQXq?Np9a)sqn;UGLOQFcAL=%t%;))3k{$Z7$fcE_#
zRzGIGC;n`6Ks{ZX#uy~sOK9OUgs~Cg5f!4K@y@U$<YnJbZnrQWg;qUGWGm=SMDcn)
z(S#QbUtlU6=A4+XCu+{{F>xLzkd&K+CiL?_I~!K2<(z{(x=QokI^f)t#31<lx5ooo
z`o%TLlH-PQs+}iZ&lma_@e&BLVK9*`p&Kym1xX2S!Re%ppY0f(*sF&OaCKKWBV4iy
z=IlU35Ac6CKxRAx%xg+(qXcnv8#9z(4i8!AB61aS@nry~z7`>8+qNMu^0*{IJ2rWs
zDmkf?A}?u7jDChGZ$hRq74(N;C&%wGN8@|j2kh+FZf`(vMeAT&(3=aG4;HGbBsAVh
ziJ-!D7!5}fL5X;J)Ph{{HNlh1>x|AwtY5W4U4M!pC@02#tTH8yE{54U9p|#<fPis9
zFvV>Cz?pmwbPjf@l#Q_PJfE<BA|(>`OUk-p#8K>XNJNzCgos8$COsk_&4qbPJO{`z
zET=O6^62F-v>tV&DxtOvBPD<p^AYd{AzX`$C^5J)+bpGJj*I^vWH&$}e05^4%|k?x
z@<y39K!a~u206XH<G&0%`YSJm<2+$z*lu^!TG@m&2DTt@Dc_31@lufH4@|mp5?k{h
zz{9~$V)&z-vHlCj9HA~E2E8PE786mH%olI4I6IfZ%=u+7a-h`U9F`epIV52rj}I(7
zU?UG)n=*pyiClXdR1?N%;h?2l^n4!3s!(-om8KMcF$#<}h{Q%9>hc&3l2rJ!8gF*S
z)lg#QIbxbsLrt-iKh*fsk41+m<BJ_B=a@th7fQx3`U+G6Ymd0$c|3XGg9l)$l^jIK
zaF<|!4s|-hQ(hAp(<#l2(c4a^EHm>BeXz8~1ju=VmK!8FicoV8n>>tL$Lrgl=zxYJ
z6_UxK)a<iwqvLpd13)RG%}$AE<^|z141Y0M4V&aYR@a0H^1h*#eBu7j?~mx@NY|=f
zv$7*0UQ`9**JO*9gU}Uovw-ER0C2%ooZcB#veZfRdPS;fW<SQt-n!yH%m7W1tV(yR
zfN+BfFk3_G1E%>hH7=;IcU~rv2x%qwHMOtYCKvj1)CyBQ$G~~Mu4Ln^E9L=Mskf9e
zvUv{+4vE8jNkA7L^T|D|En8{$>;qjR%4Kpg&b0tqP`CX`{q)JHot#a4CduK_FJXoT
zDR0jA7B--guPqYraK}d&V;D~7>@-+{5F2Hne5F+d2zRfB!thdY8R4i}3Wi5u;DUJY
z8Pq8u0Y;!0(8NPM#de=F;qRyw5Li5ihz)QN(StCq@S>Bn^5Q0}HzY-HW#pGUt<RoY
z>f%1&G$1l2Xwl=bCZx>EDK@J=_t#%XH19hjB?gOfJW)%VGKsbo7zUeh6NY@#oVQry
z)TCUfJTOK<)KbF<%NKk-{Hh8q%vRErBkzTW{&_Zx-e}G!g}MHLV>*c8E;pY=R?TAM
zudF0Tbm%VhkWNpUc+#hIFCL_%34KuRHU;R%gPp4-T^UAWP>moF$1O4Ovj$YKm5N+<
zASt|0*x4992EZd`nbY6^Mx%>WXnK-(X;=KljOgVYxOqxBE*HV^eDKe#)MoeA)<7Fv
zSc=B78|wYe)pdH5w=ap~r`lqfv_2J42$hjtGQM0{F-RFXCu^UEtb($$TUQdN!7*5(
z!nUm(WHaTrQdp%Zt>gK4;oIXul3xTlF<VHLnNnkZ`aV)hLFiG<Gv`FzcWkwy9cMUk
zIX%(Dn6EM>09R@#!SUF{Ab@p@gG6_Znns95^aT$}P5JtDBrn5LJomy)&?hHWbwXvs
z2v-dk^fsXGP;q2sMJ3FxY<|rd-ufqRQtPFQU*3r8;L3>sb)lRitjF!FA1_|^acN2T
zFjl_Pk_y}4ahjsq5<;DRN(nP2>R~@9iq?hY%qP>WUQsXZ;02_Sbp(eS#R8)@VD8K<
zii1+(f|YgjTU=ZstVlp~J0rH%l!EA9o*xb9Bwc5~=&X2E2z-8qI&$m4m9$_n<x7g3
zw6%#CIfV7k&^Wm+CT;_=XXK6VvE*_=W#SWJ1EO~Qd;tOQzf+|<K5vAKf^gO_#o~JV
zBS$Rd3mIVGL1dqSkdww{B7tJ?&j$V>|B`e5FZ#RUuhA!O&^h30l$iDt-4vtIK%#*G
z9?utYsYpT+N+Tc1gh6m{$2fSf%TB)o0Nn03=G~Yf<r*0*{1P=84}x(*HQ~#bFP}N?
zxx|Mg$eMZKgi}&OuIau~M8j#Q3JdqH08B8Q6OJC*UPe7%2($y!1XK9Xl(#p`Lm|f0
z$g*v5NRp&IBaki?ZaM<NOjPA?5wNJ<WbB2M4LO5{Mn8eski-eUMF+!qo|M83Zd3RB
zEi#E{Mw*j{i-k1Qf>%+Ai8RVLaw_a#h{I&Uqc<+(flDi&q6=cq3v&6?a06Qj1z5@0
z@B5_yJ6p%`JYlWD)~1jmhGOvLT#y8zJbnvVGbAU}8wtm{?|2<Aasg&+PZJW?)16yl
zEy$%3!m!W8RRu7Q1CQ5n5$)T~LZ?#LH8JLdP4;xtRF*yuIcO}GoN6KLQ}Yzj8M;0(
ze8yO)B5!Q_nuDqkUcea+y&$3U>^ROAM5bIWXrBXbTZV5`&0-n*CX06Vx33ID>;rSe
z<{^w9hmqhe;q2#L7g%)?Xg0XeFG6_Bp$^D;HUd&zn~z7$>L>0V=1ce3q~tUH+Q>VR
zo8f?{IWn4@ZfBc@*_|%u(c^QUIG*3*U1Q9I=psska1GAI;V?l!nmqWqy$Uor1nCN2
z)ZiZG<Rq@jZHPk@IT@M{31$C?=WBfCs)?JLPx@!#e<Sz8p9}6m(kr8@B?GBIa*b`@
zfEl0pr1)SVCO|+xB|~zdpa9%nQsj({enOn^?F_QMBnbyLRZOCa<h&>ranFm25OVAk
z@vHxhF9pd7-ALe-oogV%Ol4!@d>wIf2cEAQc{zsFxqCcc{DWskZ0CJvH*pkb!<pTp
z`-W0?xD7fqEu>un^f@9{nxR(H!%QX!w<!rLH=69W{l)`14gibH9`R47DBpcr?tLkH
zQszG!@<|hijRx;0rI}92sg#0LceFlHZ;_;^2|5etiVm_YB%8<c0ddAopRLiVFWfG?
z6gDu0b2eqYFi7|^F>y)}>imL0;-5ofT1<(nYbu%1$U-?BC9#<vh29{AHL5Vf1Q?T$
z6U56UX<VDzj6~37oJzSwWQap%L$ZRDmAI}MUe?jiCx5*p$j(grloPUKq(t1_L;Ifu
z=+==3FDS_c_Zv{s!pj|_^O7E;Lz80LchpiTR}A0E>?5A9NKTD^xf!gTXf|P`atikF
z9l!}o#IIrflazC1D(gBlBwoJtxh!Re!>9ssT>Z^U_<*RwL~n#w+KF+}W1x|lwdBd)
z`@{=|k`lJ0fDMduLh8ntXJ^jViq4j0qKK#kRi$4f4!#vV??5CAE4e6f&^w#1!`FSm
zjSO*HpS#Ah_G7ReumlluJ&%Hjf>>Ee&aGfUx0p;0MXAxN%%}?iuo8c{wQCT|60LF*
z^W=t;a=r$=)i2_bF(*}NGCu_ybFf&ZL`bG8xt1lK5FsyN&R{!6uyYW-ycH{n0UV$#
zE<P$2DFRp?kC>H^;J|2@{lx4{;<bnw!D!qb7auhZa~yC<mv;)mpU|80L#SWCa@X%I
zt1^IJgHMB53-e(P$k{Kr_%HeEuT&p#U4<sJF@WA_q6s30jbfkn?H)5Y4@qL2Wv_v8
zQ394yLOTxP;l@C{x^QmIG(EQiy&Wu<7a%7Az)Pl&4xbv*yUascNG_Rkj+>(gv=k|v
z@!@(Lcvv|3oIiYUeqCffEzDX2u+EYY({LnbXuYG{?#TBQ78sLMTuPeOM8ya3cz)pd
zeDLK?2^oQlPr@5B(e<oZq0D8*in2~|h$K}8p7oAy4$G5Mwh9mvwyl9{Rg$)j<aAE=
z(P4cc3(y9uKvU*Kn?uI9R@6CKI4%p)N>bHmv2&Il`8Xgj)VX83?MOb5EMf+fVRt;3
zxK&WT$HRm15j-@!HCUSAKao(U)gT$^%oS0L!ACbGhSim_TICEwco63r9T}sCmYzwz
zy&Z&#B58H$#KS!Z_{1=h`^*(0*1M+t1#56yqjZ#IKb3h+V-VAJ`WPg(Sss`Zl8jE`
z^=tTqrvy_0o5eD^)Avz>52+!VINO2gEVJ?m*fT6BgqXF1=0Cwedpt0X6SZzgd*uaI
z3T6jncJNgi@XnvjViU}z#6d`Mf8{3e2wqTC==L4_`z*rqz{z=~eo`@-_)JId#d@Aa
zIN>k2XLY)(2wm`fMLKv0HTe6dwD`ghf{c#G^W<gG8xLOpjMH!j_RAYXuMV?~`1u;Z
zVMlnO6Ix41MS&Di<LlClGwo6dL6{dNGY)1F!*jHFcSe}oEVgo?j6c+CHaa~NrNonU
za4-$N0RSwefano&dXI*-$g3JracfM2hB;E|SmFaHV}}0bmx39KJ8HyEmu<gDlj(rt
zU{rxo4T3S4=Q1ZDkh)*%8PiK7DjzBF%$j)6pD$v=+0b4u9FGr-(HZfaGj1C*RD2rt
zy3zGFCM0XD)HB?Y42T>3Q`H0^`M08WuZtpxoD|#y<8eU~8wBbgArv^O#LJL!LE3K6
zvhiS8kEtjm3@#E<LpUea-ABmRS_A{;;4U_PHi1PG!(9X)E|O{jP%&tBOk+*v7z5Zg
z6v-GT4K7$pDlm%%eE?$E-W2Ix(2`$S>uSsGxbpH5lG1*u7*Z+0o%prlkJ(5#o8$R>
zq4kE=&g-DK-`OXv(RfB;oLMdHb<k`k(QqZs7Bs94$YVH{7$A6gYrqB<(d#RUAYR(_
z&ww!45-{X|y3!bYc}1aRgH~>mG7k7Sm{H>n_vo88fjCelBPYIODKzyqKf}tfvlfwl
z-vo7H%PRuKjAb@zhJD|IcMHk2D%xbx%9{MvbC^UTJdf^em}828of`!!^T$F)$|1sl
z2(%kU^PkKPNzC?_aNHWW6*Eerqa9~(ROz}5B%IHDAy$L~GmOl*J+6W*KKs$8XN+N3
z*;Z@2Lg?^CR;q1mCYihfp}Y<XPh$g7i4<aP4(sz-guV9$g`wn~ViydW;euDuQZRCX
zr$iq?ic2XaWLeSZ>obqbD|jM}?P)2P)<(5H2b+_hRs7D71dB2tb~h~JW0+#yxhZIZ
z*@)+xe^2ix-wC(Nd?_&qNm`eX!+p?{XG|<!f|Nx*lZ&Ac#f<OV{LAiyzp`Y<ja7>Q
z3nA7RxU)7eLa;=r1`!A(MYf1Jwrx3y!GmlD87mIiO2j0v<>}*f;<M}p*!DZ}2FyV$
zSwV~~9ux8w#@m`72^AWuf?7hAggD6YnZJUy&4etDRg1WYD4KQ@rKVVw{VE62Tp(q;
zx(BXGKPUdV^5m&S4#GXA;_#>P@a5Bpx@{!foa!3@VYHG<2{M6DIRvgh9Bbh=u)DQ2
z!V-+qpef9qP(cbclDflNhlg2+ZlcinM{$l%z62{KIQ?Km8gpD1I{T-|HQ6wXc?K7;
z^zFp87ZI9+S6Ir4#X|{}vL$`GOGU2(g@1337f9k!@{8}loc#P_<D|+e{<%e})`F=$
z^x_3+Be};d*w8zRP59ww7~&ZmhKUB5Bup)a3enZKo53fGu@5jlmnkNADuU9*qtuch
zX^+W|7xy^~ucKkOW2*(VuwWi@zy=GRRH-CE;<E=;q<)}J22cucDS-qu9OjJ?^s^0&
z*5MP__ZpW{V9-gCw))QwNQy)P*Uvew>*&W2KuSXLd>!gc_>AW|B(4(Qf8ZB<{=xsA
zIDXWhE^Y}3X*vfolJ-mp)+sR<1<bd&j5(17xNjRID~4nY&>WaDkqJ*3zC0h<kgUgd
zyj<MT`w-6{*1t4G1WsH~kPEYqSuN^s7vF{J+*;?#dX`1vXLfE0%=>C=G*p^(6TrBA
zTe4#hWs`>Rjgsf3tR&$-FX4<x&~xcCFi=Lw-v1u((WaTHli-ImTZdY(=K^ed#q2?9
zkl6{nw8sH!*96fA%d=e)au(`|g)tKDKx+(7b5U&fH{|VxEQ0oW08WT$juF}X1CNjI
zd@xgq?~=IS7NJ_}fpy2VrKm2=FV|9y$w8NDYP}rsC1OnxLA2(G%Mgk6*=cj<SFsj(
z5Rx!L6HG0ovo|GW_D-S6rPNAdJLG!o1SU^HxvyR%mMD<UkPxoS8h5lYxjC5=0?CC5
zoeFp)JW8KHD)8x;-Z6XVzf(fpH{^f~=A_ag)yfAd{tT<q(Za&-aKauTbMTn@QYv9u
z;w3iPfFOD+CZmj@!px%|)L!e-<ZH#x+~NRO5#0PsIC;BXy6v<G#()k!qmS9)0F~>2
zmVn6M17h+0rVPy$?)-a~4`o4#V|M7|pA)%UvEG&|H^cb3MEG1`hVgHu0|zGT0o`!g
zr2=R#871I38u4%`mlR;fgBd#nl>QX=6uR`p-@`p(umk9cXBtqXRH22ziJB9{1Q<*P
zm?XAPN~L)dpA&u35|UZI)2l(ygJ;kf4BUkgB0*RS;uVPLno2+JuInq5^?hgxxC(F^
zknkx<v=G8#IyQs&U`t*+1yqzVABovPARiL^z+4-ab6dwGnxRP{S{7Brch(ty&KA8`
zyZF0M2^OM_I0gA<NEx0p+z6XX{DNi{7rEj`Jby&w`FWhMamFynEmJiCvr{k<p9wTG
zN_=#N50*-WlPHop(T-3@jo9)vlz1ZalW9s96!qu+`l~L5IeM%$kuKN=4*`?Q)FwgR
zt(6giE9{Qpbq};b*;(Co$Sx#r=KvCu86{QZjjmObkk^%hejJ!{aB>avh{7SQLEC}X
z;{~?{$?Qn3m9j%l{e(YkS{T*>VgY5Ya~f19vA;-Mj3*jy3Rm(Cy|XYr2~AtcjWX1j
zgjT+c2kB%!<AILV-Z;(D)fiUxQlTOk?Su@)XWTWh)zLd8nuW3Ck}GWVD}r~aF;Yqt
z00^hKv$ac>wjS~^97u+o1?7H+xW_~i88^eQIIuvJG5Yu<v}y8)OEJePy!fWY(Zrz`
zL}-dbiygs(iV9HEzKn6)v_Qzf_+qZNRTHzDTzoA}a71&KeRe6Z8$f~NNUqnl!h?hV
zmZ_}ea@isdXI&>lE^%ma6;Ejy+ng_4kimo`cK-dT#4r=-w#Gl#JUeDT82l*9_(guh
z+Q9=O<#<03hbOQ_4Xtb9gd3kRCi8@5J0ZOx^@dzGXeHS$IeM)jfz3M$XeNCvS~8vP
zSklWJCOt<F99qNc0E_QixkYN?wz#S^-9}8mE9WRK^ymoZ#f3OvxfRSAG0^Up6xe6L
z9b!<6+SV{9YSMqk#ks}8F%3{|^OTvhC<~GB$^_L66GCD$-VvXK!Z`H{q*fXQ-9lKR
zNcbf1H6Y}bBc)4l5(tUL24tWsJYy$}7n}C-^uPlibqYnuEe&#`&a?;E#5_+b8+~3y
zUX0kt95_O>B1&Z>B9{uHOq6wpb;R$d$RTdXb>s8+_kfSaoV4(BG{zVm*$YZ#?Vy=s
zbmEUgSHbi|E*1N>6IMEo*j6R5WT7+KX*!Wqu-)#=T#pFsoD|X=cDx{FkqJV7g1NJ;
zDJfLHvjaGV(#RwK-uUl%+6e^AW}iVguVRZi2Yj$>M&qV8$EDsd$K`|IVlZLtz!bW5
zRk?b5e1R4R=z>k^z6D;X*lKt)Vv3k!;CMbR0&npW#o1Jblu<J$202g6^MTpe-3$<O
zPMO`+lJY0HZUJct^IJH*4&Xc~ft${QrzVW^ggsx#wV)JAXh$C;*#ibyFS8)<+)$h#
z<ZfKAB@9@BL9as|#D>KQ|7c9$*Ixq0hpR%A7v9>+H29{aK=T#0Hal!dLk*a|+~0C3
zLM-s_g<Gdm;K4VkIYKPTZA!y4A`zz%FYG$ppvugxlwy<6&jUEN6bN&eYz)er4e*T{
z@;Fb-lg3NTiB#>gv%&ZosS-jXJkmb}5wf5S&`qq2kj@dY?xy4#06;Q46Q8%iP1p_0
zmqLU)N`mAZ+dg39|1ZgOK|d*TBMw8E1Wv&Vss|+1X)YZNC1+^KLChLuuS-n?{{b=z
zXk}eAeu1fg@LdDg!7*(mxhAZRWo8P+KrsWNb#5Ov389KRtaXge=!u+D#E~=J#mxBb
zq!fY-y`ajgan3(~-ZivbA1H4^K5xMrvAbbyAN+gA<w(}6J7f8Oh)Upx;gK9#F6olW
zntv*;R^Aq_&qU)rqn!O3S~9nwbp|Zf@X!7AS9UZ9apy~wkuSWk9}Qzf0;VG}!}$O+
zo`6$Fxowc%@%r{1Nm*2W`|=L0nTic8h*L<koCjXV3prN7x-;ZXONN0uWC0_b_kD}p
zpM>Z034oLRoElOydI(Y&br=U6>%I9UUDT9et2<VJ;f`eM#UyAq#sutyh7!k}9mbv#
zM(^moMKBi2>d8BN<#q&4N{H<WprLHWG?`L{O?E}NM*P4Wn4P+R%mE)AW-NwbY_Y=;
ziH&n0!I2X$6h=9apw2mC90w092MmgQ+fhn^kBK=M)R&wj^WqoQi>gAB)g63M{gV?}
zP%9snm_{WWy87t{aq3!sz?%ij4gaIT=E?*L(pr{dGhjb8PPlUribGm1PnOt-WFYDe
ztq&T;k}`5Q;LEGjM>ug2%NF9+g1FH5vy(pK4|Jtu;!_$-0+L20pDm_4ZqzALb<#!I
zznv8rjDqk`9Epcy6CLWtf;~|P6dDs~PCV48z~grg>x4em)EtH<D>ZOzy3@yW<c%9e
z${CPI=uHVFrDz019LPL?mnoF6NDBQws7}ijLk^5Na6Av3&j%sA_cur_Jh<JjkN^pI
zj^|ul1aawK2Ey11G~NZKpdvO#LS+Je&WJ(dpKY<h1FlYQkIkRYMAYw&Am!CB?0-pH
z$=z|NyWlTAfY*D#JHz68kIN(e43tDOx<!NcTdj9463=eHaH`?0i6Fd4d|mfqye0I%
z04PDqyY<ihhIoevO#)MY2>J{-mXhGl9`8T**I%hVVwv<b!|k3?azQSfYBUK(8+5XY
zUI$981tqhh(hV1OmSS}uQK*tz4jjAV@jNcMpT|@K4CK0@-rr#<!lf)CHR9qS1jEUN
zM6kg61d?Cy0GFKTP>sodP1}>i0;3khuSqb)#7z90c6hd;Ial(Cff#O>9bvHRS3c0u
z8lj!D@zBrtnh4hBFb6vn>@o_)Qz4Ab9)smEDHkLKa!nV1A;M6$qj#vbt4o0}Pe^hD
zqg4T>t|FWi5FG#;*s{WP<KY>~29L=Vp!%v0pqX7>j~wyoYkP#RgWRse68<Sc<afC+
zZw3Ye!0L^#xs)l07nKAs0;<5;0~V~*`C^-KThVygJe=KpcOq?cbeyjzwd?`!(PT4W
zk)G<djFRCj$0R8qUk4$!B{iPVf=i$8gLwBmUTCim9IpeKGV1LO_4WqO1)Dj%AMk$i
z|AY8%Qejitq_mI`@eu?kxN`N^2HW}%eb6Z*l?~hd4O%J;pZ&wU$teR?iFc|~a!X|q
zF9iS>skyej#l`V|iwzm_nKTOrN`cpcgVz#DiAf#-9mWx}ewOv!A>Iv$#KFC;1;oJ~
ztdi(OoUeq06i_8RpkP}HGsL&eKeNbZLhTZK_?*m}N)%A6O>`;Lt|7wK))c=s?79>e
zD7mUUC5~_2AFBw-FSQcZCg_~jN2A1h-*4R9WMzzTAmqhF1>%$?-R7^gB(aO9Sv&|l
zxLl7Q#|2xZOxf{;Sp50F{z^{4(1%lIx7`_eV07!Htk`bsnuR9Jv*9>i7&b6Gp;(Z;
z@$hUVqAma&?Sw>{A!8L3)7U-BtQ6wYxiAxc^cho_T{rdU9sM}b&c^-%cjU8S-#3)4
zu(?P&N<?P3Bk;_{0nnLP>V~}Sp$y<;t})3DpU_0oYIPCw%I1eXJ6Bz^;2A>Ik_yRN
zXgoOUK~s%DHMwqa1;78KEZsTJhStWFVN`O0r~`wF3l;WTxUtpijApXBRg{gobjYWR
zdm8hKJV>mR)75jZ$W#HFn^b=+Am@S2{$55v*A+n$n+jqzUc~O@^Wdov*m1;lgi^=?
zTH_%t(MjMxOpXTF=hpB^1M!GTDI+V@W@ZK{v6?3mBML#P3E3IzZN2gG2W8CN7)w$L
zxdy}qw|=grnE3$@2uQ#fK{EOkpCSNYvC)O2JSBE{Xf3e~ILtGl%A7t)dszH4<I?oI
zmg{gc4H<Bv{BZQajnAkI$|i&kwJ?SkN@u%qLrNv1)_h%R%iT}EPp~hdJU{}jTfN`n
zfGeQM<KhPoLw8_y_~^(<k#0LQFL;OWA6Bn1Xt<-NDGrncQC<Hr;J6@+*EGEjLn3J8
z7T|UreoN|_(`R)l&4>{(6H;_uxXfZoD^lGl-IT!M2>nFOC~$pFB8xCyofICA2hWfd
zt>(cQx~EWI1>9+T{uu_zoVO03EVwm1Qdg?;dS3!k!cs6@o0W!?B`#05a6=PP<X=mZ
zxFE5=!=>O5+`s;W`u>jF+Z$t<k0&1Aegy`Dw1_8eu)q~Xc!&1hAal^VZ^jPUc&R04
zu8}F$%YdGeAUq9%IVQ&IMb|GRZb!bbE7t}hxZiHjl<+!^YvKYZ@tkv_=Z@`m54B6=
zyMq%TxdnDa%(LNqJaHbc@K{t5-XfSfKeKKK_y+&5wC$mC8vT-SuI+#?G}AOCC71}3
zbJB;grA#8z2GY^TiBu92CdNr;WKIc5R;p!Wo1_e=qmP3Rb;?Mad}7H6@=+Cn$jh0I
zhTw_9oMZtHwstZqLGy)cYRE&y=r~Sg$wSv6O$$E{6ah@(KI4VqNkDkV0s`P=BC%yG
z*XH{5_`|RNO)fnBdQl+PKP$E(IT$~&QbN~eCMhA$f>~;mwol^y!y~qPRXeVjb#7qm
zrwOJzwus<D=76_Xj0ARx2RJR?kSgOxk_fyv0F4Jz?}Vic!9BFdFH*MOl?|SlglnJ>
zH0Aiy#P{u*FrygC+%e7rhd6e>F-ON0W~;w}aBk89kqN1lhltKk&dKFE#CSn?OqUc&
zpb8{{=n&D`qn~(vd`E95gCXyCEbN1=etfq`(*Ah(<A=?0`5vzLHmq<$`C))?=Qjk2
zgWEn~B|>|_cgS)I#orrPe-d&q0O7$>D?~!kbCKtKKYu*$fc5Or2`&Jo&ep>Nz@KFr
zU|>q`N;qadl3kqBLN?ZR(R&B<47u-ur#OHD0avdH-}pVIH9W<$xy-GUj8Srq#L}>}
zSxS7q_<NYw^0wj2FTY@W`?9`SGUJkPwkPK6gnXR1-`|kmDsHzMvF9<++rW7qNO|H`
zE6pU+7|!M_FQ3N1G!;4t7Z%tfa=@nH@%@8UjU@^`Le1cADEo%G6-cJc(ZVZ1^~@rZ
zbiLxyj~a)SJ1_=bXN!Z>u3EcNE9%=DQcmb+<BJkFqM3avDE!B@cc~BNHF5i#jIj;~
zC8tCsh4nB2IhJ!&awLMj-??FohH)^iyKFmh)`%!)KLxZPOU4$VcBR(Xi1^SL=qz2?
zJ_^BpyHORA2CO$`#E1mRiHK;3=;F{hJAfKfdrA_AQV{xzOE2wl5m-FjGbW`NL{W1D
zuf-*_FyzJ0@Fu~9%dYrgxWUCS$EUJmc|cZ1kBxEVaC!WEIU-rEoUEAeJ^q<%!$C}6
zzV;O!2y5FBCq#vi3HuyaCM)<PuaPLbVGfi5uV*-HV{$k9;r9}%ml9Ygl?yx+NxHP<
zbHK+sw0Ic0CP>6NOFT2ZpRn(pnPHg_sDm8_3PEyUB6yRCX_k0L)`{%vUGUh|%&4FX
z+?qNCDxl8S6LWC;CeK%5z+zks;`>}a1i<;)L<Z4v6!W9CaPivpv*3qeBCV~#odabI
z3LdaYic4n(^rIrsBH_J>@6ibhPM>jEenMpK^AoyUhdO|32NZHu45I9pI&rE0d~KB?
z@E=KCU#h7fe^YGL5lu5D-Og*{UI)0Xmk8j`7O)xbI=P@SSwEi0CFxb?R=LC-zP1kh
z9WI4ZYqSM1H@!Db3f?=i2#^$`8y?@k<Lmx2Tr-C80nWEKOil3F0m;x>kz*ro?PS-2
zFz;nOG&U{DFb1Wn!WOF5MwHe&_LTU*<b)CpZYd?)ZVA2h$U`Clnb#qw33Gxa@PV0|
z8)m0m@Pd6ZFOj4XS8E2ff#i-e)I9>`;sZRmN@3mRDjd1|5E^npW3;X2&pO6wi`QT+
zq$Wr96wnA|xkCiAbMr`bgOm*E<Z46`v~J9k5lUO98Acx{r2-|PpUmG!LNt&Cex5T*
zO&C9{r~oeM;yM_UDEefnO-{<@y=#3A+*6maZQ-+=&Mcs<Y5WdygyxB?m&vY|UsQQf
z#(|I)!MHv@{$2bv&y(M`F4B4h0UD+Vo556IZNmGw;*0^1WENc91{@cJw+^t2gmI0<
zG@PbuqFV>4$Ik?b!<7GnxP%P&gn8ss#S1(`!_I?Z!>XO+FCFHy^~OhRHvD7;kvL4l
zQN0c%0ss5hp404DMN}7@6B`<d-#GFX)*hejGQBM&kS%HIWRo8!Y7+J4hP-W5|K$Q*
zt!xm)5)+sDk{mB(6esuvFRp3Uz{}x+rzF%%vQh{ncr=NvG+l$o>Q<O@eEK}%>k|tX
zaR^B$5@ED<%9*k3Q_xOc+=Q3P1?Q{{nLHWa(*=($+4aSfIENW3M&t9N1NIuoB-!Ui
zBcH*38t)J9H@UXr_{ie>;Cr*SsP)%+Z$CGK*v1ieY@`PKC4c>uoC5LBJ|gc$spOb8
zu<aY}rQkdpo*y50etV+af%@eu>V5+Nl-4j_4`6n{fm|!)r{BI&UTMp&AFWjYO=AwU
z7vUgrDrhvb5JsH`P2D%tZR0;Rf+$Lgk~6mZjcSwleb2`ez0JUH1(wovK-7IFF9O0p
z7z6Wsp*yPplL~G%!y_;b23R830R=R3s+P&Mcge?k=TwhrC?#XtH<F<aR*eqA8FP|w
z#!Zm<hk=PQ(|?c<<u(x|;YinJBZCDYY-^LtAF!ql4=90VuZH7zK(s^`#<(PJeW|dN
zap1YIZ{YPj(2tYKk{VTlahPjVUNZWC(a$Msgxt}RKc+!Tjwd_?$sbTq9B!`5I0$Y5
z=8SlVOP1RKE}kRB!NJ2Z5w~8^3Gp`Z-194ohW~X;s_TFUuE`9*4^YJVHPW>`Ak3OH
zEJiPgIm{#=BLyU33LEiK1ir29V8*nmr6Q-w)Xq>lt&P(?co#uzvukjOIF^+@ItCMa
zOA2pLOyk#~!S93<e8b8qTO;P0qIu34png41B;6%AzK~c8i=)!EBZGx7t1KpYsOqL)
zha)qy!0#l6-vxiD)6}`#`5YjY%1^P#RDnb0_#|XKV%!tq#u+@Ln?(W9=fIX@qj&Qw
z|B3-~krOp0rS|4bEFR&H$SmtoWwzG(JlDB8!^NzLo~ZQ0^Rl4S|6hA=*CaWTBnPUQ
zJAjC+?io_<jsO2=-RtcG?|599>dFkjKkOb<%@IvG?sPqjoSd+*TyD>FR%HYLw+}T{
zwHZ{}(=NSY8h5orL+#?SV3*}c7Loa02249&%5nrqFDJ8_4GaP^4Q2n7P>Zop$-Urz
z_4=u#@W=P>7=55VpZNIwcjC0$K|xBTauCH~Tg4wgKJnY}z{j>@s~a(M<Cp8%ChDL)
zo)2^jygoa=K402NVEEgA{yUvGXVAo|(GO*NR2<JIwyk20fkTSht8*{yNVX10osA*Q
zg#KMSPg($*Vc#D(v~_H#h4k&gm;HL7ef~f@xgpCf@a@}g5&=~h9HUQ^p?$3BL^))*
zl@*c>CLgAPN-1dCy%1VHGY6NSTX1Vr^bUX(e`HH2`wr;ORcFO|T_>*dMZDUW5H>f8
zrQU{~rS7cgMRDZv8bK(@C<*eew|z%_KB(9oT8COF+IziT@Ip#mQN@*ujA&9vcwz*H
z@g;G11daD<k)xvqo_WYrDRCzP;kWCWTBIU|h552rc5%m-w^53{WVb?L-CIfj){(Hf
zPp_nVKpFBMVr6LZoRP2+Se1D(!Bt}l`cxH}e`mrZs>btYV|5$1)rkdnQ_Mk4K@LbM
zbmGj%A{i=J?^4ye$5IOK$%+d`LDc9R^qL_$mmzU~Gy`9$Rrm1Em1K4eMQ~e<{O@`H
zGfA?LW3e$E{QpFm4zNQWg2WSAEOkS%oBBCJ1*S{n7%H_vGC=Pa->c-~>p61VA*{aB
zX?ZkxnVBK33nQpYF)QdXFoW=PLcij#N@K->*QZ<!f|~^qG=emk=8!UQe1A^rR87w%
zyAo4ZF{O)HGqCr2zB=XL%rA-p{{0?BxDbW+=$`NET`bsodZeh5X>Gua1v&k*>!*^!
z*Vh+DANaji%+}Gbi!{^c2VaUY5hY|orGaP!aGk(Krp7p5usO&As>I3*fN8`H|M>dE
zbzS)FZ{MU7V0Ew^2lmH<nbMh^k+d=AUW73w=5=wg%eJ9xm2gx5vq>eapk$@u`1rv7
z{7^7A`0Th|FPM?Z5C>I)Rw{k>`-NA#P$w1H3`{{;f6?N3C(9wL@+slhrlDh~ip~V(
zVnDbXd@HbeP^DU2fdJU{O`<u7ryZE>(sw5$GDFUy>I~-S_&iUv#$B_eZYcYKvLAe*
ztD}gTTaG7(@SVaNw~D%PhP2p4bD(Sm$8j*w%s?2u`@K{i5a13hrr@aH>xIf=x4Syc
zc|uZuUAw$t6f2En$IPH^q*kQ&QMkdU$biOYv27eO&oS35BM)4nybCVMs!q^PhUghs
ziJ6#@SV*hlehZ$?AjusG><UL~m@W^X45SJm0s4TE&ifV=JAZpR(k6YAdk<vRh=gNw
zWW}thMIAvvN?(0HXox+y)eFXrXg-?!{|kdp1eN0Jlq+cYJ?zC>uo;8iV-VbkIVt=L
zkaivrJ)CZ%(;Q`^ZG3yZ(65Gl-{E^DMsLzGOzzl7XV%52nVoj_bBIb^mAu715?T{;
zV5sBfed0Ue+s-rY6(}eAb!t$o<iSPc`g;iHZTpTQrE?K|dokzAhPu0wlB}>{>YDFZ
zIVZirJd;3nWeXxqXeZXay|oHhU3jeb4+{oNSdY95dxQS&4`9U@yKukXe=#TgtaUaA
zBcGo?Xv(x7`1bqXQ1+@rd}+NTF~D}6xW0a%8dnRB*6{lLfo+rRu~C$RIS{SERAFqz
zsos^h&J3E*)op`J2ryfNwFa9RSZINJwT9UzK0cnPj|VIa*Y(2n^#j&B9?u6o%dKF0
zKF9z9U*zZ=ah-^E;c66|<xbtr9oOp%U*Esut6$h32cF*;Fzd&Y^g&hOM!#?3-me!c
zI&q?D80a{!Yw^S>z@wLvZ5y`Z3Ht1G1vQiVEa4Yhd}G5H74`AJltEy-C;*&e;_LGZ
zV@`bg?VA`lvM^9<&V*N}ShXsCyKJ8k*dl;FMVD+`_;YM1*pSNWQg>F3s;pAtAA`eZ
z+gQDdxem-gKWT}aMq#zq)rG`N3v-eTa<&j4CR;Wxi(kv(gMrXuAciThjs-6zCVR6q
zd{s=oaA2~}M)Za8!U~ri4F)APL7XXgt|JnAsP~TaSfahR4*i0=V*N8$MRpC(#V&e_
zCOO-RV8ECudV(!zaQ*ZOEUQ0Ad<L2(Gf5f+a;Q0Q)UX^1WEw)|aCDhP&bzxcgE~7x
zE!k03h(4XcO*c0gYm)s5&H&5<n*r6S+RY(!s9K*rpFIO&Ke_1<;GT5HxVx;kq?JY1
z=mVxb(EIVg{_QvH`+@KZFB`_TVRWuoZRblw0l;g)zHcy)RaZ;7*+NN=EiAxFNoQBa
zAC>diw68pR+KlHw-Bqwv&8V;{e9oxKdX{;=33(0aZ|*8T;Zx{L<toty!{#GMQ$}-k
zhQuTPQk?&}LGxB|UL9Yr6W7)7_p-s06g#<L?R=fs3Vnvg9Qg15?Z2TSmRS{-dMiM9
zL3upn7sN&K`(Dw<M7vJxN4gi<3LL#_2~ImH$cm7(n{dO`I@;@Gm3ut6BXw|jMHn|{
zZB}qzjr{FGOun~EI)ea8!E9Y}eizk@*d}*6bI>+M!&df1ZRL7ZK2R?eb6!env^OXc
zs<=^%RO_}Ll;<F+UFfPRW@l0;vJ@U;;Oq4Q=ES$h6We2_558;1#l<13*MbR0pFqMM
z=XD`k!xM(%@j#z}b~Ow%ls1-AXekwEyI9qojSQ%m3!=%z;>8by0d+g%&-7Xep=z(e
zb#ZAtW3iTm1cztW#pNfim9FId-@NSHft$=10HWr}vwo&dh4}oqL?{R%0%<uhg^SY$
zAeJfO`rT#=`yLT%mZ0OQe;=wYNp4s>I3g*D>8YecZ?+gGSXD0%<?>FV@zsJ6i20Aj
zXI~Y%c<zAcj`GjXx-Sr#ExiRv#0tn$zC!`oRm=^B!J)g@EDR)sK0Vnw0n7~6qu-*a
zAtiCnN0fvzNt#10zA}ysXeKhK8T$6XR!In1UQsYXCAof*)it--h^B7p;#${yDQ&{p
z9eb^)#}1Ss7>dy5R-_ZaM6HhM2YPYzAd5<5)TM3lx2isLmo3yS;C7YewmJY9Yw)XR
zOeP*ot4yK!0gJLMR`^Im=Yh#SwK}et#P<)YqYb&Cl0j@zyQlTZt+G#21T1~9{;I(R
zub)Z^kH-VA>knM#1vCEVkJjL0pj{WvuTQjg;c@)Fe9}Mv@g41I*tUW`JE|Kj44d!N
zA{2^byf|Xgr8BOEHab54_<?O7DEk4_i7@)bnCONU+>q59V4GvkR1<expI_LI1NC?9
z0S+=!#vGXK#Ow7!?N#leiDkinWBaZ_j@^N14tPv##ZixgkX;aepI5`@`H6ZpJik3w
zg_l;^y_0Hc(>b(Na@lvTRa}ZiK*KVw8#M%JyOYDJvTZ0c@H$_3d4uhhYUJ}mpA+-E
zn2^2T0^4FQj;merago;)NE%p6T1G?elc=R|PEiW3^TgK=njr4SCT5kxoqMLs&W&fp
zbI-f&1=fpSJoo<=xJ(62!!Q7%CGs$3_p#bs%WjzvN4rk+)`;_hK_!2-j{z*!Q(7$h
zJyS?XME5ulh(2||)v9Zml55Bw#a4wI2#nz8rvOmn#IXC%BjDkSlb%)yGk>RM6{e2I
z1Q7wDS(u))RI{P?Jl9LI!E<##;`!$plE)F}DEzI-=U<dkK+n&YNOV=)Vt5&tRo`7b
zLrIf=+y%8LV9hrZ$Ou>%%D$r>ADkKX3-jv_c7&UTvbBm1uJ5C6(997wmbLqYEZ)@}
zR7(;Bkh%(&*sJWj7<ql^8GuPdA>LU;>Rk$p0{FI6&pCTqGeJn>5-I><!i<m<yWy_a
zl-2v_<4)+#xmPX?%0YE@LB#)U!DRPPzpPFmgT3dFIp+hR?$B6BnBC87L6zznCfgu+
zb^W#E@N+g7gPi7C*zTO?3*Z0Se_L7A3t(ETy*Y<7kLLpyh?!R}sP1@va7S*;f%f$S
z?QF0ZTn6F{y1g~FgkLZ0V*tHj)D6#%57@pTqM*NCQXV&8Vr^|Zwu;d^{_p?uzvA!h
zh5d1eVQ?Rmk~Y1;A$HeV&&T7TwmYGSE{}5#=3<EUm<$MWbWq6C=fJ+}{zL#?xjbl_
zuowWSNFIGGW>CzD&G*%+mW9Fl!ol0wFwC$W4?OCQTry)DEpe~2;m5!K!A)wl!alj3
zYsSE_Z9sKI=ZwdVRw6K9+i9&kjS6QPnl;Rm{zn<yORd<Bhm6FyQ)4@xsLv0SQqb{&
z*B#|8K+d;nv=o`4$xpt@!)hev9U;@U<eMAuW+fTw90roL@@}0y#H<(!#BOP6?u$;G
zhp?2Y#FI>>H~*Z(h3Dj8HU>h1I0jBCWJe5nX64^6wAQ$TJTO*}oAFjbnj@TIO|P1Z
zj7A%l45-MtgSw@aWErfgl?AIzbsN;bumDKQmo#2rH65aYN?hlg#8{bef#o)u&n^X1
ztLshn3kRoDog?0SpfZxp5rB7WPG$Y*C+28yOmtZ?_%;Ene@aa38VqS#Nmde26{1oc
zYNamA3jTPJn<%^U3ap|d70mN8&Ho7R#hh?3R<QFDig8^sFjUc5?x8jdcVA1B08z=2
zHaxg3a}3n@)ZF8eu#YJKus~108Qhf8i5$Rl+gp@aNrbMdz93YdA>gJf<L%zZpP<CK
ze5^!tP=fg@*H0ydt2Y$qLg;-z5E1zC^+mx`E!g*+eD?D~glTh_qilH?r8O9n_~sZA
zwVe{1RBQ+B0D7&&p2xse9Tpw^Y?zN7Tir2d;5<*fUMC04#wF<6wrPn=qxK;t&EtmD
z!j%bQ3dU%y<8_`G!2vMPP$#f&JUm-1bfPR=mso;y*?IN#VtOe!9tYtt(bqykDcWKK
zvq>jlbO0r0c?UAKskwCgUe3YY2oqK}9FGsU8TPF*VC);ZJE8^l=QotUF$lNof=7qV
zt^pHSD*L`;KOPct<qM<Ln>+96>yPhf<5PCsgIMcvVBdGnBzjwPObc}xG7;R%Cj4Mv
zbn(Bz%Gl;W80RA>v_tnvuc)L6A||G`RE;@Rjj}>c-g$AtZ+*<g6K8<up)~1SoNIM%
zygk^has2dF%?ql^;74d?E*R@=&bR*FP#{eO{1ONXYfl&_SB|n$EnIQwg<W(^^lJ()
zU%e6f=t#<Gc0F0iN2wBPO6WKO5gkpCBm*`BqG?|Q5_&3COgF3`vVzIXWkl%@)RiZd
zpuv9*C6c+ST?Mw7>+_FJ%lGJQ1zW0oId|;mi82ANK#2mN31i1fQb5jO*xl7iqQ{t1
zR8e*_(l_!u_ldO7;JRF~&r}>2*_qRN!{Bw7#^NDu49OGYI#=*b3aE>E8^ni6I0&+?
zIINa`l=&)Jk`YC{)mXE{WTd68U{38PnG@;pHK(un4DT=B#Y`ThFkwJ!F1itCcLbyh
zK*HGJENdhF6*Gk*mqADX<&M*Y%?0rJ^+h^qsc6?lIAYtd9|tT#`w#{`KN+m6_AKNQ
zmdD3~RgOEh;PUs7KMyOohU@hO%z@rIzW?JpUe|@^w-4?PZe5kzM5&Hi4s2D6;>C~e
zUEV!c!|V(uR~ri_K7)(ZadhnahL4Y@7QEB7v{bIgbYLx{-~1qcZ;pN6xMu(cEBm7E
z&BpsN#y}|)R*`B-vDXx4&FZtlK$_#6111V*;+0c@2CZ}7MBVWC-+ssQw{NS;y3QBc
z*AK+`g*m9+EcM--j0eh%6^w(ecyTR8iizDTs*-F`bxUT`K!1Iy`DKd^^VJ!c73{TO
zbOl>4Xzdd0s6eBt7U}acxjWZnw5PXP*tHP?Swpt629t!CvU(<8dyWYUvTPRKoeA6v
z-(4dk3Q=xYX=MP`wOK8@RfK2dl`JOFVaHr<yJZs#EAMX?e}A!e_<6XHT{L82%%q>n
ze*M_6$((Nn1%_pR@ZIa347*7~#yKWN!j(FgOeEyc6XrDJic=2n1*gc~Vg=#srk7QE
zBJS!a^!bwptjWrHr($r{7QJJZhHa;?&5CpHM5hTR)bYGtFHCJmbh)SA^va}3^S!KY
z+<vF~g@QGo%JsEy0y4R#GHK$O*p_jgo_9%wyHn6wC$&o>Hy6%)$AWt)D)lD;U~QP2
z6KxD^?kLp>YxTzaAUHFP{LCjj%9>@Rkt4exb_#rVG%%K0`@Qqbt|jTI5u&leZ_5xQ
zgjn9X#x#xm=hsgqh2Q`FJBNh5p>F#UMtn)qzNtNDTh5Ac-w!-KzELY+#278O<;yA;
zuGeqhm?WeG9+NYsS`9G_y%U=^7=C>Jj(!|yW8hRUdOn^g)p6PvxN?%XoVzKY@x?gh
zt4??zfjcatF51{?DUO&!doegRj6r6Pt3x{W9gn@jwu%|H9xmoU@v1>{?y)6xZ?qE;
zfo*>ZzN+ZWsfgA?>$8)mua|k3S~LLk>vdz(W1^2Q?8gUw``drU_U$+B%vSi>_a|D}
z*l}H-WXX(<>uiFpG$0hTF`NGT(cm7kXD%`jlu6+T46p(|Uw`21>pKvEVg>v2pla9U
z`#uL3sBar=21S6^$!gAohXxj=&cL6ov#)tTLN8F_L_$h)E_1VcCj*0H?-y+#)T#1d
z>3PS*G!B%D-w9zRj;`4Q`vq3axfdb=+|?Z!2>Dkg%fwyJykAR%SxZ6PAM~a|!VG-J
zN~yGW=^Zf}%s}rjC7a$Fyl&X)vG!z`SM=;abbsbWMpG$FXn;CS1L>wQfi7zZ%}{&|
zN1?N-d5$p%E)kuQi3w7T>2RNjpzdT$vX<tU%T6ZDFwHT%a4}SUAZmft0bDO+cQUS%
zo@nl<l?j<5N|Fprsi(-5HM{9#<Zz~=iW`87)G=@mc9TgJN+4qdFei!+c<(6vBFVxG
zb=wiggSTXKYLN=*wBZH5KVZj`^w!aUN!DPh_d_!aQh?KPog%hCz(T@PK(y?9z17v)
z?H-)rJM_>vUm`0G`rc^<$XV9Z?*)HG?|ri009(She|G&;Qh0uR@DQ9FgpM(Bic8;H
z!|~V`E<3bwtZGFd2V4cy3y%F@<?hCTn-^A_eZay|tK*6Qro)}aQD0vJ*Q*iREfr&M
zhXV8BgpI(is*{WZBilesnDg~QYc$iVwW6D$_e)j~+nZ^Xpb>Kb#E)A`Va&joCrXv_
zxt2vw^MeZKiNNR+y%V2=7tH_+O_+JAQ2_;i;eKAYNiJPXODVLBSOa10Zm0zT3fK1M
zH$0x-h;zmbL~Ppze}2GhBBEn_ePUb}lb00S0VWssLZ{2hfL^z{u;ukTSKy_XYxF59
zEq~{17gd{ui<ahj(%+?W4@U1p-9eGy=~gJ;J}<09Ib+}=bWH6f#<=f==$~GwCeFYe
z$>CSAl79Iq7E6Dlz>US0(EU#Qe#V-e(e!YgPIV~^wzChtdwIYmSdyhRF>w|obw9W!
zHtDaOGa`q$`VlA#J{v#_2@V5$<yJoO_&M~i#eto$fPq#PoDs}5h;#$0R@`><w}e_m
zN6n5V#8$b7?t}oT5x9*EvAR%1$_Nv((u_e##!96x)Aj)Bf!H($vw<iidW2Yr$-0Ti
zn%0p!nGxw_G*D~Y+6jn$I`s+6ke2-c1Bw_;qF;v>xd(IrP2CW?i`hG-8P<O3ya&Zs
zB7_yhcEIWZuk@RWlql!~OzAdK6EyW6uf#1z5&(6yw*H+7f%(&{K<6Q?z0KaNU=-|c
z%keCF1~Ck>EC9Kqu6fmh=l<jCr;>svMrlqM=A`qenJkQ}l!C7d{knh|+{|>I2mt5n
zL_r|7f=#^n8G-BT6aDIFUngb+e*631xM-$zTvx~E*9)x$USAF0zU_P+ZQ^mLGHY!~
z<r4<lwjvC*<NP}D`h2k|D+Sx*iCUi+@rQish23l5T+I;2j^o(b5zUD%DL-b%u|4U8
zX@Jcs$50;79)SVJ=Xv4!{Df)uGQeuI6wXv)hS)#!Oqv1Ix-rpsuIqHn^The}#bJ7c
z^bV@{>VfU~K_jxn6AcSVx9TX{j=CLaSHr{!?y8~)YL2F1nK1Sk*y@h0Y)nYn>B?p6
z3<IN6bsZox1FM}c{A|VX`s|$bTo+t$ninVYBze_H;=;{rA>W$8bpg#XoCBN=tXAGC
zSS<V=>I!J`Mg=KU)kg$L4qKYcu|pZDE>^(;BSUUkUKja4O(r8=Icw;e|G1Khs%(Uc
zi$JLqJN;=n3;;&~LWwemlM9XrG(khO|CZ%<zZB%1g+|@^cUyti!DTG%r6KVEoZs2a
zVBr|j`z%iUoET6Sv_K-my=z33=<G1aej15eD>w-ngy0RUd<VVmguAu}O4+c6wVKBb
z>2JIi?B&)t-0yI*WNvCQk}|5#F+1t8p=Y1FH(;WZ2H#nrz-v;|8!!T+z2JrCp7(w%
zt8=8=f)10H3RuC&9$W>@xSD2@b7>FKBoCSkCPAQ>v6y+=jgX0u_t97Iht2~F{+w|!
z&gTyE2q2sSPJSldPb3`c1z}o;yizw1%M0T_zJ4kxygpyp9y^|o51i+PfBfSQ^gb~9
z#N*?@wrw~LV#;4#+pGpL=K!mnvat%OMwU)%4d4Ixfq#7ef!Eh1ADBWldmrc{@cC+p
zAjNB|q-t_qB-qvi;+XRx61t)GAo{@7Iz?N-8}6mRw~EKd1H}!kbsQT-DX-5LqFoe6
zO-EDJh*DvVZl;b2jLtUFh;I^(nPRBWJnTF#m~|wAAA=jnxZ7Gd-`hUdm=nclKMs&R
zVF>KsKCq2}{v!O;8%2WkapUD7MoiK)v+}501&$|&Sw;)-ZL6roab1Q|c9gP#MHrn<
zYjqUoO3tB*IAD6#ePQnuO{S_G+$J{pJ*`lKVtPq$f1BJkmF8`sKWwXGnHDJJ?O>^T
zO`j(iapoMxV}^D^NY^qtnk>7EggtKe*qLlOU%M+p2QDjT!WmlkI#Wz0Vgd5!vO?6c
zD@A%C!vH(IS|32!`Nz`KF#<6=nMc=aSs@r8jf4Zv?l*!fZ)c}|X0J>J<=Eld0WTY0
zKQo{f%x)B5PO6w^_oZ-72hR1mrG|9{qQDCEgqBbc7EFv0YX>ts%&hDQnVAxm%>c>{
z+YXpqUUU241jU-OpP2Jrnd%u(gKR1j9TziLz7zmfV09-1H#-aiB@AJl$-tQynC!3?
z@0jLogkLp35ff(y_a@BQhyZAJwwbdV;ul@4z&<a8ofy!H+u%mSOp?@gP*fZ0HnLM;
z(V@Gn&-P9sHG4n9gyKk%y^|^bm(OPa(k9h4D=<?Ro)#Xl$hX-`AXZ`XA74L}6xtYY
zEvC7I&!-U;=0sUl0Qvqj%j)>?$A!no6OZSPeXCl|LV*cT@$KUSpPyg6xd#0B{trE8
z3U0Q2M{5Jm$ARNffOTL75DGNAZio(iAB3A`@34JC-4C&H0<(2i&b`Ct#If&GH&;Hu
z@kPN)>m6Z+=f^izguTI<RLHHts-bR)KMNNh9oKnbT(1Q^tb8q)0#Nlm23n_kCJ){P
zpB;8>(BzlZ?Y5Jt1w-$&An&8YddD0L8iqFk%&I^bvC<rGD;s>b6)0=1u(Ht=7J=EB
zL<5}Fglrau1Xn`R*TAk5YlDM^g|6;r^iEBM)ZqMXor&8R0u#<A48(T5l!_jcP>B)d
z(@~;;aN(_?^{h8%KNPWU9f2*H<q}$T76&d{6IRXq4NJKS&KpozZ2<=JiNwJ%Z@&IA
zEsJ1<X~38e^d|?-2$}FPxjBeeljBajFOAW>(lVrkA;yJvU8_U1IWZjYI`FIyShAqj
z9m=l~i&o1;Pg?c4lB?h)BrIX8OJR~06G`7S>90BPk9=m%-^f2sL8rL^b%*T_5iF)w
z$ex7BbdO!%yK6YkqJp2TDxgkisSR9vwMp8T9W2E1IqHxF2s!o~CY#+h)l>GG*`Zi|
zb`~*XG&wal?N2WGmm<I|gr8q8T!Yr*K;XG=cs@22DhW5Imy?PVGv+F%0GGBNq=s^Q
zWt;=4OIR+-*)d6^_nxqslI2L?+Odv6wcPwB%}WqIwPiUUdcXhq^;1dV`#*l*OAt?&
zjWcRsT--~$*^X*Jd2E!l%OK4nNbPjP=ojkq$hR3V!~S^S`Cv8skl)ap6K89Ppq+Mg
z!{cxrq77FYc%2=`gPI4JxG*zt4N_T8t>t_?xX1^MFU9rs!qsv|ucJPzs*|_RbnJZ9
z5^pS=4HOytB?S-w(&HY%;b4q`Z;u20a9TvW7HxYfBe8;+#ayHsGsAga_-Zd$Ehx_e
zN8L~dE8QYRH_UJxJGB)3LVE>qQN0+v-wDqRT;~_g&o3_ZKwuhVa98E($^eHsRB29b
zp;8&0cG#{1Gdit_A$Y^xG26h8^ADuRZrerq^CHcO-7pPky?#sh35#SR=*hh%(65(W
zw$CpA8{x2FVqqXfVnbbYzS?S^gIWgu#wnOVCTFX9!!niwTw);yz<B_yA&1seomnRI
z80qovg19F7MXL3Pw0<c|8NUQgMmC?Nw^P}`YH-4Qz+9e4GZ1~ihM`<1-DW4N^tqA@
zVZlIfc5F;+TqwmLlol1xAQ^h^R&dS^HsWomyu<6Q{!Xj%tVIKaGbU{9T!s-IY_Bg|
zKmNf%sLobySWE(jmxAi*ETAN%8BoGO(6h}%Y6j#kZ2)K^#ixdSPb~(x2V2T;B(#R$
zD&eaA51sVh?Apu5XTXF$a9t<P>q4Iftid(sa>c}Cb5`X(Np>L1O+^tD@U^2NBus_9
z1$&d`W=`7l!`6Mwj$_I_H)lFi?hdB8Sia{oa@bAi%ghwpGU)!t*H0ydQ=asr(_iNK
zIIz`%S8phou+a!tgkjT=i`d}<b54By_<?;ZT&km?$`})$KR)4Zcs_T0J1QKGmsHJt
z1Zn|}2c_|^v!PS~?l^w?jSFV7r4K?bbE=X^<sl6H<cy<T4cA34n^GK)$AboRbwj@z
zuFn@nZ#ecHrO;34IxkE!DzlAlt1fOkJ2VbYdmobh6K^kNN9&F5twFYlmm+gL?w}x!
z%!60zhEWW0IBehHbpuM3dn>KAW3ZAYYsYZCJ^^reIpz6-pfG1U4tOnejg=ZWpx|gm
z7lGHrJ|_1rn34XtDF}kag_pvOPbuoF6~)tfL@QD+nwF8VwDb404%<MzgJ2_bm+f#Y
z+&a1Sxq{J3A+2%7VxlYtQN&7CQxMx68n7A#RtSlH5*pDnaeLp)g*WI#biWBwPXQ$X
zR-#K|oz)jOe_5QTZ10VgUl_mpXqUDJI#3)wr4LcD<oUbjlo_C}1xJx4fnErsB=ix1
z_VokB0%hAVl2R(DG29U#iDSCD1hH}~?XQ(6?Y&~w|7P&?3`B49o}4^nMSAZ=w)-B8
zfo>P)EWIPf1&qcdn$MPhA!*hz*L!DT?8T%dI+Hy+02?Xtn+uu=OLaU4xlwPY)nrXM
zut*0}&&aC=*^t&jP)nwHQaagL_LjzWq{q(4<vPqazHM|KR<gGZq@AimeW#L|{@!fy
zcuzxlj^0Q)pCU@48A(F0A`;mk%P~gMiz^21K|P$aMBfgw0b#)G+2sDy>*o#edc$#S
z*mo`viAhFH9}|CkUa-Q$JR|Tp4ptKu7#ElF9NPyE)oAF~#OK$E)(33cQ8(J##@W!W
zhUyN(<j}YlwAL~EgkiupV@6FGcpt#vE@R&}l<k2C!yFgJ`9->%InJx2cZ#yC7Ce9d
z8|w2nggeIk2b;4};MG?LFvmon1D~HSwzrr#wu%FWL%u9+1S=(B!V{DGYl8_SwhF6U
zLXuVKl=h*I!2V#{y>A<^?X)abQk^}6J~`lS#H{-ut0>gD&5`!Y48r@d<9Hmbh_Wgb
z!~<8Q0pJ{+@KcdkuTtnV=Rkng2|1)){`-><&VEo^Qa7<W=0Je+99FRdGe{j}Rn`ed
zsWs*xm~ql0@oX|mW(Y1eitKVtX~J;)yZT~CE#<8z#Gyl}q?9zk4DxHx7eZBx^O`V2
z@tuLmjO~7&0SS<b`BIJ2{BI~Y%JN|{DTI+_RNc_qbyJriza%H|ptu$aLGO+*c74zr
zD$=t)zkZ<hOADat`Ys5lV@#@~f}X^99IWaGHJE4e^Tk{WFktzn{oN^618Lr_?}-dT
z3+<qxB6=asGlGc0UJ2Ef!a~Q(+YnXkCw*p$!7>4;lL4e!_7z<D|MEKq9dZ}+a^v>l
zTwRMu|1K(u(<<A5HSEs<Z8TDv!wbs6MOR+?a$*KJ#D}_1pOK>*Qef|DfiVR7uQV6P
zpik6bE>y?;-hYstoympRykKCmThyPmx+AwW=DrK=Ksit!AImbFzn8G&U%7rNDg5^P
z2WF?S)$GFMW8(4ciGDSleQ?{91}kNw)$!=JjjV$k)n2a?uj@h^lio%-Aige~jTYVe
z^T5M>iO2G#v<R3OA{fl(?L2`o@$v16?WiaklFrt&C>uQeL3hTw@7SJCY>#i)o*z6E
z!iLdL;=yxZHVWlNOs=>LT1>azaQ2R)5BmR1+7m2hPM9c>^nxk-la)ZetYf0J6Jrow
zDs|`L^Rmh43Tk(S-(MHz^@0P~_Jh7rec<c!3)gim95wiMY4Ao&wvE7bp7`<O6Rk5q
zKOUmi#Y739&xvEp6`qEs;0?u(K&q{sTi&DquiM57Zxmdw7g}$tnx8X8fj8BYhurd*
zD8bo!(cQ@~8mju&lARn7>b>^J#X8x+$lGSwPEf@OQ2}GJyUjr9DJq<VNpj&2op~F)
z4q6VVn~5>FH8AZqvLcNDX8Otm#^fr{G&vhmtit5qA6TxH)M0S%F>!4}jk|C_9(XzA
znqVu(1TJ-1wZiHRKf<uOaDd-ubvnBHnH)R<%WB7Ncmpxts$1k3Vm&YZ7s;*X-#syO
zk0q2GLv-8%)CXO0Vat^`zncjor}~>z)7dU=a-95KZYl<fpe^lD=2EEp87=?VIoJI&
zGrBM5y$F#Gf<ITC@KpLSjnRQg+j!p&xEE~g!e}SolgKVhkda|A_|n76G|zG~sgU{I
zvI|;TgoM=eyoa7kzF#JMY0`c{T{Q>FMTMh9Fut`HLfuB)cWnDXF{$}Fzwxg~3W$kf
zz~_%oT;~OcV>=4UE=+$W-sjkgAy-;nt>YS<)(!x^|M7|Me|(a0vK7AMHuRxQdvqPo
z=TkSD?O~~b<KRvO4v9Nv4^lRJL$QK!HSCWAvz=rDbTSIacp%5y_6PRA{hea3{h{u_
zu^k86@xawi4zYD2eB;8JW8V-lFs=(<U!V9oFI*!~Uzgfp4#i#CgejP(8i%M>94Lm-
zsFPSNYYs_uc+)MD>=;%qbI8Hf4CV@Ij@1Dydkw)Xb>FZZ2kN%dAh30u7l*xLP6_)S
zgctN@2{Qm!qXR5qtpa!oVECWifS9SC49y}2VKxnkGw6nXYL1D~Ny()!j{~=yC*-gV
zcEcuPa5zWXkPJ?t?t=rwga>2Nu#O64{ctG&Hkdj#%(<$2RKP&3%~Yon!gQlGcx}Ru
zDX7X$ZFY7*`*AFL61e6svX*uPY%B21*X(4(TtO)mSZMbG3FBmWvF0fBp&Fg)`ymcI
zfIr<Mw=~sLb#Fzit@T%_+abox3V(pR547ND<^Ys)KLjltEb(lmaacl5qCe`N1@ZtY
z93p^%$@jiim~V)UiDpjfj@orH;R|Xd+o}lW7-(~%;I@rH3`CzRA@CmgO15-H=X=)D
zLK^^Zp=s7JRLR!@8SVwRtxw$_0yZ30Xy`|s$|UTkoyzZe0s{o?29|CNQl+a)<)>1I
zpn{OD&it%5omS6Dvx?NTU@{4YOUq<_?~14u(Q^>Z80KY@I$Vj5TXx{EHP8EJ*Uzz!
ze!O5d@wx`CHn1NDwt#4hz<%s>_3Ry?WiMm~ZFtNlYAuLb@OgH8HQGoN2s><@idk7!
zAF@}-3aAz>wl?SB=ktqNhyb3C9qteEfenOh2Izs(25Q~VdS8^)?f2hs{QW<1!MWf8
zD4VlcVCA^!#xvUv)LKyMhPiK;2(G8>vh6U#x4(Tu1=j?cff^0%$}M$+M?t27;v6jY
z)^N23d%dvN4W>1>5tGW*(J@~ya{ec)tziwm5!M1Af!-h`O@qla>FT3%^`imDanA~h
z^K~t<7KRk2kbo_D=|vp-*;TD7AkP^n^R7tSi!+NH4)RMtmCv03(g+Ypa5XF3x1lZA
z1{bTCLtTaN>M3^9jfYE39%7;u^50V-npa}HQ^6}#DFUg#aVb8h77o3ZO-76%|156y
zD!LzO2NAMINCuJ@DoRrc-Oh_iQ3iYq{;*qoMB~76-?m7qp{>G*-I{C9$F}S`FiCq}
zOv=IJS+$=zmC%ueN1d?VFvqzlu}~r><ksO(pibC{w_AZuYUmrWf<gXZvY*ZsmL35^
z1uAEWtu>5xv9kv>m=@JPLb=@F`-4HwSF}a$JqH+Id!w-<k*tux;*{V`yC;I^K$HSM
z4iw)oCSXhzs>&%yEc$0Eh}GQ<76M~Zy**;0RtEhgFw6bXAe;?o!O*x@m@=VLsZM(6
zB#h>?xre@|>FnnY?;G|rnC7(kevH=TQFglkBzXcs2AYQfCA;T;L8s7L$Fa)+6rh!{
z16HfBvHPKT6vRxSlAJQtKK6OiTrGg}Y_hT7QXU}iaU5_M`TXLyZ`+1#Cuh2yfp%T^
z`uxOJE57~q4bR7c&`|g~FYL#Tq>7C>n8^z(g2zGIZYVG!0$mG%`uU=0tPh|P*Y5p-
zkAYqcwLF!nH-w!s>H_NK3S7+UO{)Q84s>wYO%4jp#Ng0CD14BL`8u^mGS3?axa0K&
z7yE|Q0J{*XURlA+8QgYP35|u3Ju}WzN@ljo+fI=o-51%ej>dURH{f_YWNc`{nVE<a
zLK!J&3sRKZb*^Ahilgon`;E?uFFTA(2=5}DTpyT*Up!O{YC{zvnKxrueWz4^s}-dd
zdEwAbg!k2eN$;Ym&sv>H32rEO^TNqNw_(9ka6)`6bWSkllHHRt5R5<{va<$OQiwjt
z8kz+K+D`$XLofnBBF2>d;37ey)=G%6R3rs7$r9Ufpw?YOPwpPxD*c<va&YB+xQD_J
zOr@RNZysaIVbl$A(oijf6R~R`CX*_Mu=g?~;4N?`{aMd*nqited{T1~w7YQF3i^;x
zuM`p$NC9Ob!WjYQhrWNTt%Rmo#h85s$=*9!OIj}vOhPq=yX#DN(UnL`tr+b>KQH=z
zjfPoBK_24qN2AjLwEb}Ej5j$R90#x;H}#vHBD;p1m1T@f>T-4wVYbxzz87xzbLl+P
z2`;*F<cw|t18CA-q?BTU^AH2p4u;zUOhf3Ikd{;A_7|?76R(Yqy>6&;L!T6Ko!5mv
zy0#86z`Ah|*RG2XL$(R+I#I^NI2$~I0laZf@wQb|2ad-k1{3+gJ{j;l$P5@`;Ch|%
zE{B{gfjJu5)$zw4F9tIco@oQo20Xa^ElQw$eWC0Jv2hcVVNzr_&lk?~g{f^?tFj#(
zudh$hAWg6Up_1oE#c>=G)){_$|3q&ckE7z)cU-;WAOHUE0E)T|=gPlQIoHq&F?&bv
z4gI9YQC1BW<UOYeq1A$_%KAJ{R*?Bww+)YPzrhMucwS#$IOq2@00ua~tXpM(odNfo
z+dU`eC0s;~8`csxoi8MxGxsPo4se}DdM1;#WDOzPWd3TT@f)WB49Y`YNzN4jveI(5
zbgt%{(o1+^=AM5otJkTdOe+jqcsNkjhAPj_LPPmJx;k10G3b6fz(Zt$$I@Xkp&bRF
z8zx|KYrO+~z!QFi$!{mP^kw67Ng6EX{p@1RLMGs<wiSH))g+uvu@h$i1VHY2LcY)0
zrOng@NZ=K4_*=zY3KIhFO~As|LLI&<Ihak=HASaY`=GP>bTit!g8?$kF^#j9ZibR<
znK_lv^sN1cWOReIc<CUGTm`{m)*0>Iy%_1ADwZVqRhLO@y~xhRld6pgn;mKANBs<g
zxC7Ls7>C`hU6@8#hFyUQpKT+1N$RA#T@p!Q^rcNG8RWC#&!5(#qE!dGC9jn#bwG2?
zhLFH+CAcvFk!wb8=sL5tr12e#2_HkgVv~B0UJ0MIe?d~n7h^jrha%%`K?+X=m(~~r
z7chu3=<j5o1*^PMLuxKz0Uk%euAR!YkoiIvQC2KpUtegwOO$rW*v)Ap4WL~OpI;3>
zUKjSQB0y}v?xgg+ULE}kcxgDl{=oJ)=svm~NU!}`E55)*D;C_!_IQ54ilcox+K=y4
zDJv`2Z38NIJ{Jk!&hx~N&o9g&Rj&MtkYXVP32TFUqSMp8RKyrK&oA`z#Q9=k+P4j5
z6Eg$B`AbX=+3SYm;Lhc`?OMdX<MH@L?-|gL&4asW`{2NzLoZqR79hegUAwC*nGPy}
z)kf>(_nc?&^Q;9nc_~pPWw02TS-o9aJBZch^Ry7An+n37Ah>Yi^hL@bIuklDZTuR!
zr&jSdOl>(uNrWHxe$LrZG`k>tM0|R6K5uQk8Dy`GW!wlgJ!CRw1t%DV$TOmDtBq&C
zX`&69-jT}WS`2(Vd_MbD>3J22F&B1fb3gzp+<>8kyEy5ZJ47T6UC_Y_BDw03GaR~b
z3}EDn%UJi0Pcij23Uo>|?qv3?d!sYUr8w$5Li>!xA{roJU|OZ!L@+UCB3(vpV_+g#
zP=QEexj9j0!YAySs&BCE0T->=bJjMc5vf#Od)kz*3SCuq44`*Z17a_TsRZVnr?_)_
zBg|m6sJT>!3$o+tw$pJy9V!uzd$1jIQX3-c^5tA=DU39PmAmMf7ZdxoqaM^Z*yyx(
z5nVT_!qieyAD#sgp8?j;IszB~2AH7eH^z6e|0@jf+-Yx3Uaq5GjTNhBn{ODs<8f?&
zQ9#r?7s~JZwk%_-(|~T@cjDKh<Jh;gva`2&W3KL~IkZ(<JrO|RUI#Bc0K=UoW3`}c
zW%0eYx*;lq*O(I>z!rg+9q6Q~hC4h0`^G)3Uq8Oom8M^g!S(|-IxEmZRkoGFN|~h_
z&esdQb=2w@K5;d=SM~`ihc|xTh>5yy`u@BpZa|3RKOYYOfpJ~3d+;024IcJsbZ9)b
ziV@(zz9>0il5r7}jG5rjZ7YSlDYBY$F|0a0+P7LI%(E4|lLE&f9#>DgsFYR;rl^dJ
zA8M5gcT$(Kv-&pY-e=2LQ#5DRrY$xKtD-W`3YNj4yCcqv9{kQ~E-U^JY_xV5n6KYM
zhgw23Mc8^Ngfp^iV}cXLo7MFUTs+E%kTP3Kh!BJjI55;8QJmINcUWaI0-#;Mvd1Wn
z+*@38b|j&lD+xx%;j+VmMF0hU(sEc?0zx{LPV^ppzUrLJSgdi@5h#P3<VqRBnLGOH
z3tW5y>LD`O0ScZH3TF4gTZLS{Gb7Luf};${jN!x8`=uerL<hWJLWw?uxV~(y+@JF2
zt#|<X$)uWRl@dQVs4rSvepFatm*~Y%Hw|RfB|y3i3sWa3!Chd96H6_cEAd{~Mg@8N
z4@F8bTDxG5wP>ptJ=TWKN3!KibA;TV3<y5TWIiV?{2M#%gfmNFf#3i;IL)QfsvY|5
z=C&xNql*-RK=@prE!X)%bEs8}{gdnG*hgk?-<Hu9@mNAdV~CPD2H?9yO~4=j_=A<~
zzO#ZDq;T%XCW;^sgV=17I4ki|77`&W%ocmZolpuQ(At329sW4*C_Bo2p#Xpz*)YLM
zp;poy`<U>v0pI_?{%SZL57fP)Zmb3<$TG5PH20u!-Q@br5WX7Ffo34QWJyrB4If(9
zd7h0>*95kG!}0t^<Gh&iiQ08rIjD-Dx^sV!a@u#gz|x93G#JBbd;jG4IVQYvNH$WL
z+&kLWCkM@$d|a)x`>Lcn2z8{qKZj$v`f<JuV3QVEO1=5mf_rQ!g+(KDR!mcN8Boif
zo4?XNp-P=SI@hg26?jo4>avY!odGsA2#Cp@Gx<CWi{Zfp!GoyARi9MBEm(LYG%%BH
zukCXA(`HFP*AP)%jYYy9KzdTgoofgkxZj>ng<618IU7R+W~0S++y`vE3+`2PUCvUr
z%}}>v(T#~+cMd8i7UODZ6LcSFlL3Sl9+4id&3bN#xRuJW@C~eFfW)Z<-*JyQ>B!tz
z%cK+%&!ruB1f<CT@p1*%q#TDC4gKoPmWdz?JX=Gw7X}nuvm3$0XfMPlv~e(7CY+h*
z6IX{wBaXOfv<YnmVY`(OWLB7+5hB1>En&eaSUh0CB5L$Y{zSY7b_axY3teP`sk4zR
zw>fWtq$l$)pru%${WAAG%OsYG5ik`;j7<|Vtc-ko$#7GbS-6iS8H4}q`l+Na`o#C|
zKd=|)aB>1II?jt+=%j&Z!~!!yEK^m%ejIH3#7FPJYPXc4_Oqb%vFx0CrzQXha&py$
z1{h3Q?>O6qZ}q@&Y+7=`EDsnyzfROX@bP@0?mMhpXxBjd<Atx+g~wL${Px6tY+R1B
z?U)f%0vCtx^{$Szb9qW6LbxK=M~WU;0>^RSx8J|P9r)TB{s;~wkB<+0{Qf&ty<*c8
z;qIh@N|X};RC3q72(MMpTX4%e+W^G+x8s5Bc#x(Rfq7la)}d^>SV0h6Gs&~hnMxlN
zfi3ea(PZ-lOKv_?2sBY<y}+G&0>}pG=)zcP8`yR&A_^S}G1I0u(4K&Rn3+ke=WKfu
z^ESQIZUB<vJ6FImqs@9Af(+WrHbYQmYFt=2a)``<b@I&ti@GZHq^9PqM8E9bzYxJL
zGl!}Q<^gcn+{Q}s0Hk#0K~MN$jIIQ7rv(0R4k=UXq3;GIAZK+S!q0Q4t4AfcL&N|j
z>wdks+eT6I{d=W-c|tAtQy@}M6jT!WAB2Jq9}feQW^CMx06|$`S1^WbSmaq{q?4Lp
z)9iu+NxO{YZVPvP7D0%Uo9fKDwzt%cRBx0ugSsIwk!5qn<q~oY{;s+aGG5l{`V%Pt
zeNV7SHT9qtr2yEf%U=nIF(^K)b#<)>CX^|?%?tu|Bv#n+Sz6l`O$DH3N0EkwdY8mi
z@l3J^Oi@mkR!7dcbCNA*0j?0OMFrPCzkY5VXB+tM|NT22)v;C4_yTynPVD<3KOMfE
zeQ-tN-~Rre@VYL{2prpiQpDwUE|6(gqcCi9*tV?~b&ScSIK7LtLaLz|SucRej+pK2
zxEf}^Fh@uAvL5spbG3=vUa*Y{L&5WT;Q6sr32BBtfoL?A&dN0~(5@G33cHm<C_6eW
zssT`T!+t!l9jqqC=qSEnuLU2^C#G-Mzx@sS^Qi?l^u<qW4uAve$(|eOh?wo8<<QUW
ziYBcIwr#_)AJ`uf$3dPmV<NoKwgS-BGyu_opnp)pBDDh7d2#^}0=YkdYXr3(vTD6!
z-I3OKdGL~0$<?C;Gk7>(Kfc2&ts7Xa6W65&;E;@2ylXSf9aM2oK~-X}xkC)LJ3yo!
z<K8$Gx9v8>1qY~DHDDT995ao^Le*L=Yik;9z_9`)63;Ga|6#<zV@h35#uIUA29e(6
zjBDJ+we56^jU3Pq4o*|q&R?OPCLjxJexFrUy5O>=XCEofV%O5O>zaD1+l!0LsH}9M
zE-v}%x<@mI=vcJx;0@JhyaT<Lr28s_hJ2DDmPmEZ36lsj@3VWQb-Ev`yUITst*KiR
zWjfco3K^J5%S@iX2oDB!5~Wli9MgxMiPYf1nPw(CpZDfAu0<6r_?DPsXd|8w3(sM{
z8H%MQM6)cxrdjL+ThCP>lTw^DOe)4h_c5sIR=1uVIEx86B)fyVB@B*&X|YHIv<A<=
zu8M$ipq7pAN-voGA}pxc#{3`ZkN?^AbCgtNI8zJT2iX?sud;2G!5x9~yZ|5Akz8Y=
zfJfpKfWzcDI%?fycN^ID4Xtri=B5>ziER#L?}ax9M0*?<uTR1zgE(d>skY@C<Ua9u
zY~;C5!|~j)KOd@4i7Dno8@*vhBbGTrRutL6q*8IBUxc+PCieZ{JsJ~sT^y9bux%Bw
z9ZEF5UQmmyOJaHVjIMpHgG0=s3Xpz5TzQ)CPrh)a7BQLV(?j+MpI6FS0$6w=s~r=3
z(GZ0Ph%~<oMS(0ppO+ClWFAZkwQ^n*V^C$?I|t>pVB4QCs~F<~`h{_w$o&?H|5gQB
zt~ul<>`nKp_<HG=hgGTY8^Hk(D<MD%+#D*-$sGX@GsiMOE?9>*CgG<(6hO>xM^<SA
z=3u8Xl~DYhGB|JM6~ZyiF-@}<K}4+fG=Gr?*MNerD*B;HeANbk&cwTj&5$>MdpJNU
zWkQL@P7ad{nq2FeomQ+)19X~YwnCBuGVws4EtS}rpkmVF$y1<b2Gay0cZl<M^54xl
zaa|pHS>d&^3*UEa2ZMcp6?bnJqV-#h=sTiptD<*IdSD4wTuEfI5b?snaY8ZBy9&eY
zYo+Y*&N^e)r@7DSBJJ((rq*aB1VGk@dUyuGp8&LRjts!EhU!yAU+R2JJYmA$yjNZv
z)ho<G^Ol4XCzduR=3574>%LPFAAsracb-A$_n65BXwrg(-S5AkQxIe(+hPNvp^t&B
zZd^{I9lga0T5o9M!gZZ^emt<1r>p|Vy}$aP?Y0CmeUL}*dlfxz;3x;GvhWzhF(KAN
zsg*5W1fmbL$$|Eg+1G+?FM0`CiEnko^9Y<5Td{t14$VC)qbaIf#T-Dt0_XX{C2H-M
z6T7_G&-2u70SM8{fH5$;<3hvp@xb={2J9c=w5wH5zI!rk1~K=v-NgzJDH7sq8Nw7z
zpVi>2iXm*)>D}BrW;=oLW=)(?$B1*gX<*&Y)tO-irqQe3boh&eIULvy4A**7*#j_%
zH53%sXfPYF<Z3&U6ym1i<>4@&K&WE11h6F<im}>xQVzAFd&(<mW;a$<B#`ye5r{%a
zEeECPt8ZC_fuFlzLk7ZBU|VrQ3v<9r!Y2$ADFBQJ^z#eX*C&(DM!WABCL8iydw&_&
zI4}lju6El-Or%j74}TwlJM@&L&%vth9T8%(m7DQ{U;xRgnX{i*fv;9mQecJ3u7U5v
z1Yt5q<{&bgD9X#?BNLe+*Nx>m&^b*+1G0DE32~VMF=v1K<3QaB^(07xIe=M!cHvTI
zXE_T1KZjWX1PLb#E{EWlnr{R|)$~<3B-EBF-*u#1e<!lglEt=*`7{T{XscT?@Qg5j
zm}soo!SnCza{_MI3bjG5pX+Iw=~bGIu>yRGXp4k?YZdIeT(b(sqB|MMj}7!<GcoCG
zOn7jI21<7Cvi6AVM!U8J{`2eS*hdB&mG%X#(c`5GLmM-2T^FivD31;M?wDVO&p%Gg
z0a!Tv$rrWFz~|=|AD#oBqXS{+1R=2<yB@qMJ2w<7I35SAIAM-H<W)5J&pZw=GtCF;
zau*CF=4{IT>~LCdXuV@Uc2q$eW7}}3gP3rP^Fj+UO^WE3i-Lw!1luxYiy%EV0`QF}
zeN`nsdLF=ge=93;;h1t$&DOlnF3Kl^r2$eW>dJ~J0;8Si*NLWNb+rp)j3p4t83dQP
z>+cXx9*iMkb!Zo|s-dz}t{gPSbz<}s2q=;0fWCfw!i!_yW$Qh7fTux;Fm8y(-oA(0
z;aDZT#82zrC4?Z(xd$+^*i9IM!LC>a2Mr*LUh*JWz?5x59QP27x)=jRAl9A;Wi`%|
z^d0F%bRsltPvbY24~Iz<8QQ`&XNUK>U?VdzsDk=}F{hFWJGN3QY=0_X2vU<MY8VsS
zw6sD9cnoTJN)?<CC{BN?T2w%Q_hwEYt+P4jyOT6&!R(r>xS*H5bR`q9I#B8(7!X`6
zGyuA{nI!mr<uIOeu{YDeq@?N@03oIv2ePpQa0+r}EBIQ@Aari5h!qq`Wbn)muqk#K
zvZzg{5jj5LyQkulOf@JWPC<=DEJueGTRu%#6k>H<De_aCS<bQQE{5LWTIzySdmq|x
zcz2DAd9InG&J1{l+<<WB?ixXMCfkwHI_%g{H_j@<4KpN`)6f2u>!*@Jl^i?;KTW1|
zWCHOE=D_Nxj`BQYgWd7{-@jvyiQm3G0dT)^1Tc-e763zN?`qMYuj_w{`wC2$8mtA-
zuZ#NyW{PacMyMOT(oriQPPC~_9KxJw|Io>ypIp9TzG1IB1`28^D7#^MYDfmYctzRy
z5Dv&TmjSeDuLD_C4&NS&lRU=>L|`j2PYVva@&!Qo^X_6+M^gctFY>ZG2noLWz-*V!
z-#|Z^7{=&a7dRxu(&0^}1q;RC9$8gMh5#o0<C|;~YOUZJY_+pvaGs;e8b;;!AO#40
zba+=Y?k<51bt<LSrR*d`QP#T>3Jg<031BkfP;LtHdxyVVbYx<(IRNM9<O!NFkW8Mm
zim;%CL(JsB#`-gy6Y=1WKGd?4;v8t)2QbMWFAmgWgV&vESmV4P4|goFTj-f8Spnk0
zOKTu%u7mHdZ#!wN`v(I;3|Qw1$%w&CUo$XN{Ft;r6zXXMS_BFLdv$7B)LkaQU}i!N
z&Zs!c2xQ_&Gq<#GfSIs!e}hm~?nY8jg^ObC8Md?|rzzz$z-PXDe5VqJH+oVtA&i^$
z8tKZb&t<ZnU#;}|e1(gz{`4%xDwDv3^~l?IQW216XY`8}CTVk_Gm>tmu+&lj$zBTS
zE6m=E@AT)wQt{2(o#M;2T@N~Ia?bT6f(ZI6sO$(q%5&uRg|J3uE!Jh~_=a6@A1KDP
zxbL0h%|`kQasDs;RQ|B=^ioAB>$e~SNb!mx1uEY@3O+s#{QJLu$B)koe|-PMu}NiE
z3}<6?UG`lSr7io1exCF}%7+qxesye*4cmkG{fO^?{7+`cb4V42IbZB9TU_aWkZNAf
zGE`!)If1$t_`ZXKHp<EOgPuw;_yX6hvdB$(=4gmPRd$gdx(h-m$CEFNs%*ju^G$83
zULlAxMy;#!0L0W8xkTQv$$oltz?=u?JWrI~^+NR(Bs_ijb0FzJvQO55V!fe5qH0!7
zmjQagOx;IB4!{Gs<OT}d2C;qWhP4}a0wau+QR8saHhQ~<K5w?g5h44jo35Jv{TRRq
zZuv6<qVWEEt&6(6++h`Qk|`ssksNXZWJ)2u3wirlEkcPSG57#sg0hLh)Emadh4Wig
z$5~v)h7{Ay>q2xx;=0J7ay;p~v#5r&-L`tsnmQ|VuL^K<uC<r|2A@p@91g_5t2h2W
z1EM-2RmoOYHR;8G$u+hXb3qizjB!ztO&w7>mF8^(H2fA9-qlrPrG4*k9t)DoWRSKG
z212CVJU7>jEHxR)F7mn|cb0oh&BioXes@asJ1rSNk1H<{GD?4ddj@HOAPw1VvarY^
zEMeE&g2+T(^?e0trPd(fu}GbY=oP_)Ftp?-wNx(EcNH8rd)LA^_FZ$Rawnfy?{B_)
zDjK|!&YSP;7%WWYTo&(Sp0PWnX<%P4YIf0VRI&v93vvE0{u}_p3(t=yj*sVBXIU$q
zAV<rVN0~98#TX5b2fy6w<QAj-*jcp>QcY=7%^+8W2{R>QSzy2%;RVOXZ>W#ws$M=P
z(9V}CZ?c}Oaj{Js6JEImOco#jKuYZlpt?-1_z->DF)a`Ep9fFLU3c(h>baURi5YqT
zVWfJNZKq5>0Gm=W$`sAafZQ52<9^Kv$RcoWS~F)AwN^`l!D`z$=;8}WSYieMXP0Bu
zPy~6*8QMcT<^M#h05~yACdIeICKyE>J2=E8&Iy0B?RA6q0dG>WM!fmcq<4}kSR-Ao
z6z-O_7>kXR?zlN%OlT=Bhf_(L6@OCl>YI2<8F1Ma@UIX&r)M=h_c+@c)^TNMQDwye
z@*JI5d30z^9lOd-(>vN`D<I538yDJn;rx8vmAjLsyzjhMb$_64+%uf|k8#oOE3_su
za<%I0fIEttj0?F##4u59z(-?+SPX$(6}8J-e+nLSnD2<$*Fc!Tjzv~kqz7LK+QbM-
z$Rovo{DWej<*;2gA(6xSX?mYrYF=Ne;Dk={Iiy)#cKqxxz#I<#4jXT(_Q4T-s*6+X
zWrM#ni<jJZ$Mc$2CIeYv-}QE^vy|VL3jA&D<N%oXizxMTPR^T@Tuk?wHZ4QM2`q-0
z3m%LyNTR_z4H%e_o#LPAznpDRzd`l3VAR2R;T&|Z?V6o>$ul9$3n6$NtDt7rO)c49
zxqd1sY+5f@c2<PuJXDW^Vw~POYB6qU^bHmh=c{XIGlb?8eR=L5roHoq^nuxT{`0o&
zs9DW8Rmb&sV*mICuiV169}U;`iGC59>B7QmWnjs{N-yP4N>)RlcmPOqKR%#T!?H>=
z#Gnvtwu@C@3}AA*T)!H|Md+q(6&-+ka4$k7wBV&OdrRmvOHlzS#*tuvC9bhM5vBFH
z!zq!s_v$Rv4J0k+e#!a$h)oB;j_W+pMW^*rsrp<CRs_)mt2fs$E+9TT?qEznd_yTz
zp|UOu-TU?yh;Qm&V#X1C)zXzxKE@0Z8HDCo5EJHoV1dwTLv~{fJgS6r7!z~$U<agM
zMraA4#Xb`@Zv8<KBoM)%H6~5x`k*K*EN3nX1kU{+Q4BqtI-NQ2dcDwJCq}1Btcm$l
zi;U<d7ro~U3^uu{whu(3<%EeVHs=Hy_0**lY|icgL1tC@09jfj*Bxo`or79~l0LB!
z)h)IXmKy{s-u^?JxsdhMm&cTzkvrf8)61G4M8a>74q=dDTWS?!tgJH?P}ibe1zGUX
zLu;%;!1T-rU)|Qz0E;q;*`WE2*pLX?)hC8O@2+6zbKSyC{*H83fSDk$)X88~*BtT+
zB9+tdkog?D`$-FIy>~2R1%F!5&{IyUwvY>G30GakUi7{!(|Nn))1`vY+~p1|C@AP5
z#Ub{6K39T=3~`p;1#33*?i&0bT|e0=6dKQce$rE?kcm>aivLtA&aY2G17Scd2-|RM
zba?dm!t>Zs_0Xd@Hgy&yx&KF%Qg$j6W~*%|+kxZpA+ZN^Py>Ai+T`FnX5iR2xTvTF
zK%arp8oW6NzC)h=+m2c)>X^8$29s^Y;)+Lz$~9CKby(z!$?QAK){9_uN0c4@_&})I
z;7djwz!+umtOH!A!MPz1UXi~Jg7Gpf6+B^<NtP2AnY8tgOWkP7W(IA3WpK}dS}B;d
zU|xN}HGMR4u#LR_ab0NFDbe2zKNSPEO5C~isTcas70;LhO*Yb5&0AI{F}W)k9KIWX
zS&N_r$#e99IozH=-C!=;;Lx0*6p#s3q?pZ4Qi(^ZS(rMx)q;mWvV!#82{3SQ@3sc-
z=CoOdp$`sWEhej91xv3LTyt5_)fGj%&|eMx>!kLkrV(33IW}zj0W)sMiyz-n$|ng8
zW^~<red21`K~b_?=l|)w<Lm1a>8VrKJH_{fYFd6CEU0k64fn=41bd9>2rDq-Y$7IL
zo0dNnbqp~OGE8&BO6o57Vp>5sJ9Y*eCe^9nH~L&ookKL<vT=JE>82Yy!XSLIy2MGU
zb=o<^+a6-tMe{pGZ~8NW3dO=<Tym6pAQ$e?bs+Ub6vwfM$VFYE+)l!DP~i|dGr5-{
z4I-!ZmK#)<DOsg9#VuhtvCD#Ur8$S3XPH>nX1G#;vhhDJT}@W;aob!>D#c@T&SeBq
z#T%T)Za~Q)^(jI+h+cp?9Uij36z6|#rcmh()LX~vyzn@7JiqN6G;iEZaK6443<B^&
zxj8<bA22uc>mslHcrZc4L{&StiE=eZ7oHgKYN(F~w&w@xzAa;}Il11mkBO@TKfXli
zgyS1H6lUNnIL`~O>qQPeCSV&A!sCe=fbU$WVjDTq<m}J2xO~W1N~edcipo`#FvE5{
zP_~2B+qNN@31rH#8m*P#iN!N3zr?|lE(jzQFW2<3^|Z)ZFRN!PZh7uaHbCqXt@vzB
z(ssH=a<I=AnnqkNc;&Etw07I|s*<-ET-FXH!9ozj(tF1Q8|tNAyOcATIZdj`*Ji!m
zn+&>~`i;@~taGDPEes&3rcMJqQVeq*a?eGQcTPN>hk;Bp(pt?z&bpHW&j>`w<|ASi
zD1<vFAZ2=WEUtfef$s;r?phhi0dVw2^EJDetZHf$C~_;C7;s=AbzT?w_9?~^oRZ%E
zbtBYNYpRoDg$Y))w)QP7xCJ2sMZ#as?qhb$ZhVK-A?5l^i?|2J87wt(fH5Fe=L~CY
z8XH{8n#f*A-EXQ!r~a4EK`;|+F_&OjtguQ(Ocgk{O4@Z60wHt4d+^V19QQK`+I7qb
zl$e^qWuYKA2u%8;n9ic;+Br0K=qxl=^Z<%&eAmRDnrjXbNLH1r(>GDF8Q*NZ_Xwq`
zQf2nZ<ywSN-v|ylkBP*ev7Y;!)L+1{x*e!<k0jb?p0M`4=-J-03ArnWw#nIjp8jc^
z|4ToWKkO1j)Pi<#cflc#_W%b!`{T)jFJAxW=O>~MJdRzGaAyB{Ja{kq1s@kN(s_HV
z%udy4Jq~P-2O+CDxf?)m$9bM;ZJ>?75ZB`K)$sZC!pCEy|5Dvl<#z2pAg6f;G4F6%
z&k~=L1#JkA>`lYoRAIYz;tJn<d}w}fP)%&4kj;@VNYd@FDhhY#V8k0dqPIeGgXATp
zDx3e-NWB9oqM5MnASoCrDEgUUpmI4GIAV>phJH4T`Qo!MEj3xci-R;not<Yi99*}@
zM=%6IqGd2qgCLCPMww9(H98^4DAA%NdWjOFmq+x7E`{iQkQs*Pq9q=k(aY%FU|jFI
z>#q0vJ>SmSd!7Atum6AV-$5L;tt8&V#_AY59~dX^8z$>8X}B(5mDJ4Sa}GSZ=>e2Y
z$B7Ba63WI0dySs$j&SF7b(nfOn-~sK{WSKM5qLy&RDI0*`THmVb57o_RDNa-#P{UD
zupGS-SBlHCP*)0}URkk=hlOemTw%TW7F0i7YBmN%hC@qWOXTRBH_{$I(iw<-Oo>Qw
zDlP@<&J&Zg)r;G27-6?th9V$AQ6LwYHrLW61jLIF-}x(e*dcBf%HZFgxh}|R&q;`Y
zfXZ$+`g(VCQU_HQUxY(5b@Y0*11YPiC13XCOJXm6+w8c~_l<`p#V6nL)`*-GIpS3*
zh9R37fX+U_pcHPH79r&@*A?nlVP~51a{Av#b#?SjYx>#c0#cM#R!Oh<{%TXXsHu(}
zry-NCr0i=dzI1~WMp|4LCGFfx{nUGeJ~T%Is3mKvrI^TK%<V=u>gFGia>)pFx?f8S
zUVqK&<4|mnoM2BO!Qs7Qv&{E`6!jay;~Ljlve7xMimk(yT?_9X3p}%l7_ccbGF#xg
zupK}=s|N3FhuwY#)y=*>OT;%rLRS(P^J{#)I&M%^O2I+aLW$1g<*!f`U9B&fipJZ<
z++8Xt@ux04JliDEH$tK*AL*knuG3XYm;%LInuAjFOaZCmVOw%@vW><jFCNBUpe;Al
z@|7M+R&Q^O1hKdfdUE%0_xt_C>#g?wg4m=E=_tB<ik6tSlA|G|MFb+>fxg~fcp&<@
z&2&`nQ(Hu4WQIbcX_)aO;#j9F4~)eaZyQz}w-1ViIwc?TQ#CuIKlip9{s8JQ84rw+
z^Lxgnnwz{cd>jHD>O9NeR$k!dO?i~xoVffg>XwS1T54>KlKu%rA2HOOf)oRi2R4JI
z?pts!iO}WTY2bj(o?k^t)nXB^dEy*yVN`rXFKOO;p>4O$+H-`{n%@oqcMAM`prwm!
z;3Bb}ERa7D%|K+G`8WXUIOQ=dv_Ar~l};9Vhe(I_NN}30%r{Xe$oUrYp5wp-iORhM
zl?z$-G1<6;OUBy>K?ED=^)DUJ&|qo|WS(S3JlR8)Zkaavt6l5!<;V_$#+JU_u(@@V
zd1C^jH<?WpH{HN%&2E(Ev?1yuYd1n?P<GUT>IKH}yjD8W{nVlRhj&#_qXd3ARHh|+
zI$A6u3x6eT_%S<2EB3KaJ9T+;sqHohzHTN)<kNWA*nR`iTeGKPi;63!v#_zI@z}O|
zMB2Rs^Ul()e}foET;7eMhZ*)Ou^G3Eh4-x%90%aS_Ht%bug<_L`w3dmZ`L?y#axq<
zW>~U};EPXe9>+05rYhjDyEC#ulECQpQJ_|VO<C0_6;1)cYe7n`sJr@nR=<HiJVMn#
zmQ#y1i1ugC!w_3PUXzOO6kuSPjGeZ9<8kF}>VdC{Wde+-ZQ5KXNqY63zF%LNf7;&A
zuS`n+)|VtMU(>`)FkABO_p}wpMdl-C4$@BJ-O=$Yf5~4*8ISHu<v|$*Byc)BSQ-~C
z>mPxw7%e+r$?*j<OM}q|QALefIrR7c#>Q2_Tu#y4%WE$5O-VV1-l<D5?=FG!CxLH3
zU-+vSVK;6+?(xS}EvEOJ1oC~AuMAxBO}zvx%mFtCy%s6v$!t*VceBZ?Pu#$esZWlr
z(lY|YobA@S(BOI#gQ{7vHn(%;*W~6?`S{#Ge;$_9z#SV1jRbInrYnn@*H)yw$iaib
z%5fF9Pzum|3R*~^?Z)*D%5!}CUNB2S{gm3IWx-g_66k=Ax?9?>+xqp9;Cd#5m2c`5
z4w{svz!^|K?@_$tzF*T!AK<m1*ys#DyChZ1ffYkpZ4E5UsA7H024qJFNfW}8NX(3$
zirXzbkuvh#d5+G{hf_b)l-ze}(|1v?7dxkHTZxCQg@z~z;h^A)NCB3?QWD(r@Da8f
z$P|3!g<kI@l%-{hg2Eil@BxHgkZVQu+K#MfeqC{!_1)r_ryP<G3FpuyQ?HJ1cA^vU
zzd-A%($7mGPvi_qu@R<W<72~9ZMO4HeGgHNT^5DhQ<C(gsu*M!m6UW;lk5uN2e-}<
zs<ANrbLzc|r<74>C{O6)tKoj}@7drLGnGyIH2wKIwY*!4TCP;X|0pd~H2sJM8MRDn
z8gTuQbyjXLuiEv*I$g?B5ePs1-QGp;`R~}iH$9!6m9iEno~6@wG#w2#RG0+=PF>?=
zi+x@hC)xv|Uyk7=fIP0^IgqvHwF)9scBb>rieG4_fBwHZyOyuCLGR74W;ZdQG;;8E
z;UD`RY7a0PZeo|e_?biofbf~{0;Cky;+17k6(-o|0Z3N$B+XfH51%3fc;|y!`hNOT
zNAcMY^SCyR6W!Q6_OiJJBtE&G>2={JE>>^un4y2LAHpL;2U+?J@EvPr@$@g@h;KD4
zH*U{&&L=gEk(@2JAs}a`M7;Xmp^vh8k+qG@VWCKeX$D|MWwr%y_hnE0udb?F0t+ZM
zWFKhUD_jSt&s-q?#&EZ)|B#9=Y@TL210&?CLN#;C)M3=K$lhtX6#0(uyTCtcBqB^(
zg1`3s4kZlZDOA;O=MEH>S<^6p(>5@0<)a2ALi*=Fi#-#q`Lljke|I@D1e-5QP#qa1
z7zp{?pUN!86<h|OtBidpio6H9Tuq&*Rs)6i2Lb!l$GgAv@q;Goeu$7@$4~#1tH=jt
z9L<Y4P|WBBjT+c`1+g>nSbX$fY~~xhdnN~!2R86q>h}|0VDqGdFp>>F?7Lv`k?RRj
zc2L{}>GA%ZY3x-sU%+y}a64+vB_u$dov@x=8rKuTvA>oPaukPkSE|~jNUyaaE0wBL
z$GZ^fd5;fQt)nu<cTiFBSScAbpH;ZrETg4sjiHupZN17AgIj@2;$umn`vVgD25ObD
za^o+*t*E+93cjq!{?dgQ)Ii<Vh>%3U>3lB+LPWW-TkfafTg<|hTtI>Th?Atr&-}=R
z%<wc>Uq+YWO$kyvvPs9i5lB%HS<9a-<i8Ih@yH>pj-z+DnXr0i%7-^$8d>Zv+L)B)
z#t~|A2i&2?>r2M;9HZBig9z^zu;5SSb{qinRulTljo+2rp#zL7_r>D}tz9vAo@Ge<
zqX+Wz4@OJ=O1Ayp@ocWXu_Cg(Xf%~I=mv8ptwQ$(IVs++ia`QZ;@=n#G{tu@0?x;H
zuohsRXZMuDF4Jg{TUBDcPZ%uIX79)njpwneZH`B8N+lCcX`&xxm?`J#Q`wsbl}=~A
zUzB6pXa=Uf?`SAFwCZHpc|Q+-+>YGGw7at`&EfjKYL#zuQ!(a{z$bQNteIj=lKL}T
zKAmVlP{X9ybjVJDRm?)(e|DJ7@$92s5Aw(3=zK0tk}#ZAW3jOOR*?a*xAJ+#-G|Ix
zF}He1=uAlim>3EeUAif%CzIvzFpGl;rB?!Cdf9_W2X}2fD-ZUAk~}YCVW5QBxyh2j
z=c~6j|1RmjTQ7%g?TYftAf309BUBMev@YFYHyYanhi(&sa0LM@ADkS;?F&=*5uaGZ
zh5TcT{f>WAOvtjuH9H=4t~~i5FJVMoKkk<3B<p3fCUKFl#FysHXTdZRQthF|yPbA3
zy-G$maGX5Xz*;tXCeGG=?z`-)`~I;pY=dm$&2y|~WQ=gitx$3}U{iC(dJLuFSHnuX
z{~J1G6V`M@zGI*~<S)83*$>WK0}PK{Eidp`)C->FkJC;*hNIy!C-2&DHBH02jx4t+
zw~@Ae=@X%QX+dVZGpgj$9bU{7yniM({hnTelr4tSFQ$<8p+095Ic2}A<4M1Z+8g|6
zVQ{Vn%*W)4=T`%}Z!?{YEZoUlu>$utM#l3{egoa~9GP_j`kCmB^wnbKb%6EVr)t2z
zNC~@F9llsMv8>fw_zQb1HG@JHJdJdbc7V;fuI!URP{|@U2)^@WiO<Y#v$cpd>-q1Z
zlBFl=aWz=GIhGNKtIaRYC66WXiIc6>{gWKONSeqBx04-bay`D86!k$i3%hee;FGAi
zITA8vd3J2VM@F|rVqkdq#b1sQFg$(Q_MquH0x$Y7f;}`CgYh8SJM*JY_ebl^?CnpG
zfFVy5?hA}kVK<)xqiQ?>UyhDT!iucx!Zt`k=>Keu*9$&N;W9g7_)I#)zir>DsLOgd
z`D-7NNg)WBrfyiAcs3HokF4Lm%w~w3WF-|+{mdAtc!Nw}odhKaQ449W2AJRHl@lNn
z(s7~)XoZNF5M+NY9{v;?=mZCMXp+*nX)v*iBZ}fgOw;2wRr7B)yR;vUv(cLKoj!^c
zXzk}~S<N+^yu0{Ph@RZ70&{Q1XngU<jv+~}h-qTUC!?dXT0;#kOCObbH@i;W?;%uA
zYuVXr4L6r^bpbIebKv1@9T`8zp59KmRW%_7wu0ns+-RCBX*@KdyR^vCQbtCo?}lV(
z)+Pp9M24SU{gf-bLo+=b*j7*aGf%W`h-1A{)Q==_Ej+KoZ})puKo(cc!Imk{s&JYj
z3$0Ud5v=`OTc8OK8`8)yf`>JEeDgEuxBX4V+;0!0Qwog^Q#5?x=WFT2DP8j?d*&)s
z<7=P>J}Ez!LcwHg`m<48Z`_b4BKvilvS&n4t`WgzSg=+2K3SX%xb4Hk4JRMp&EBk*
zs8Kl;|J1JE1dRs0Tv6gH<8e1g#uK^LQqNB;d=8puOJOUEImCk`t5oK59#oB~2j6(W
z@RIOiuy1dGAU^1HM0V}1Q)!JJ+t!o<pZ$?^oiaHChX8eXyq5bKGyL9jv!S%Lc8`}C
znZY8wwCq@gySw|BHyEF<U-qspd;x9O#E{EZ*Sku4?$^hamy)d&uf{9!<HvZy4Yiqg
z1tk(sTd#Gm(Va?KY>2Tl_v;h)v#ZnjUHHQPW_e#V6GJ@J$S@(-iPs0Oi2ZqW%GXQB
z8OiPcjgO;<St_Idy=S0;A4dFN8SiquslnOJGr^Yssonhl3lTZG<>dSj-kC;#o1>xj
LT(v?O@!@{}{HVe~

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_008.png b/0.16/_images/sphx_glr_plot_visualization_utils_008.png
new file mode 100644
index 0000000000000000000000000000000000000000..f87cca7b83587bb6e481e675f8108b295c0a0a48
GIT binary patch
literal 153833
zcmYg%WmFtNx9z|T?yiAhaCaDdaCdjN;I1KqySqDq0D%w^90my<JVApy!68@(H{ZMO
z-FJVqt~&d4^;%uM_dZqeS{jNtn3R|R002i>Nlpg<KnDIhzePv=xBr&<JNVy0#8=+P
zSJ%VAH^AE49-waR>*?y@>*{O+_P6);arSWM=Mv`P=KwqT`g;0^a&x==PXjIwZ%6Ja
znC#&{5DZTxV;=wj&-y<bQmxcGX8;lape!e?7g+S`dAiAMHG>KD4T0@s#}CQ7bv65J
z-Ws!b^Oy?T37I%tQXIKhRKk`r=f83cl<weyFZePy_!4>pn=Nzb<N6QlU3b5Feo)g#
zD1;K&He6f$=o*3OpQkxGtOrajhdk%`B#{$X%7qfpot@o>z3J<<+-N#qygPZi4fs9x
z_Pg-K^W*=zYk7M63On?FH~RC4i~9d#o*%`+-k<^hAH=`d=U08bk>UT1_<x}OrdV$D
zBmR@%|02H~X*##s{GZE>roNrU0G|JycjL1DrGPA-;_q?OUredz8zQ+~kEfFV&PtZs
zwG+_y^y+WF)IFNkvp>(DfS;HD$y?tO)_8WWZ?)O<FYF@s?L+R1R9DA;KZxb|<GYue
z(rwzmgKzJf9)-egZ_NY#)2+9Zi?s8v`Y^Ho+{|kN;BQYi?iaoKO51;GhCbe)@(&6!
z=@Czl5cT~@+}L`cZhtF!u~YgIcYPXDgOdVdUTnVzHM<XdyZz8@H4jF<&w~SFg{U1B
z`vPDn`?w_GDnR^~5+Q3+DG&7-6QCRbs)thiVnE>Jf=`e~F^O7n?)_(VOKZ!N4i0ok
z5n;AM{hm@;I`aVzUBf|!eMUkC;DRMbes1EnV{IY}BdkMl${_gK(Cxo(D#8xj@dJuh
zb{yo8+&l^5Fo(iLn1_uvxB<Tk$Zdz7^ugxkJtNnhvT~kX#n8w2u%f`O7x=nHJJpAv
zt;@d@KDh0g<k$e7cY}1$R%xm-98WVZQ^FAFhj16{Ns0}mfV}V*cSinw-0(_gMGr4a
zT*Q9VAl=H0N#u}?%#lE+SPYIht)@6Fa306eLrZ_sncPPPfDS^9%m+da8+T%9-^zKs
z@*KIrxJJ^COzh{}tyGQ_ihe-YngvRh<dK0t%>0;rF!!`li|p9xW<PUOl%gyU63~Z%
ze1;DSF0TtKXeCONsMD50Vh6={L4X>HeQI)B7;Kp*5;#-iq>cogWEw^hrvXmig};ov
zoJuu)x^FZK>U%ltd-9Mr8D)~&Oo6^znMozW?e~+o2(@V6M+gAdw*E9t5dz$BdIaNG
zD5ooRh1EWp9L$JaO1a~@qu;dewEwgQFl|ph6={{?r=EWg6qbWCMWDi3x1hMehPeE(
z-=I$)``~e9g7O|&^aSX{jefO4OXPtTc<t)wx7j)bpBGHiMDYMD^n@<fs6M+tk?^6Q
zeMD<KFY`Y6<$-{ikB!ENAljsjxgO{}=2q!=2nkXFTT11QIBs04QAk_bw3fGL;m6^W
zd-GFmaVa_5#(t{3pPj(S7Ac7^;cv5Yo~<!I#I!`0x0-`Bd;5g@mo;xHw+>&a`<>zi
z?l=3o%T?-hEJVcg24G3B<7g#6-(b)JuRE>W^Y+PowmVDmd}QRnE9L6t6+8iXRKw)W
zBC@zOI|Uyd0A*~@uI<Bxpo(tBtuRVgTylW^SFd%?8=zY*9gau@OVkGE9N2j%ilz^9
zT3);>&T1<B<@#mo<zCTmqzM^Fu`V_ZJea-o@B8<rl0>XO{=F<kpIW~dd?y_GxJBfC
zM+J`HW!f)l7EygjbXw~Z3-)Ng<DoqLK_`*2O%BA5_TeVM)b1}Lnt!Ffvnh=LzMHA>
zn{JHuN|2?DsAS{8^qKeLK8rX>0PaAX8h4-?P96a2{qpugHnSqB#A)7%OjZ8ssYuFk
zIo7N*-0+KMfK}dLS;;^H+#iaROkfwsq9?0WAe2WT$0<2MT~Ti4X0%P&x<0!azZx{+
z6*BmIswm)=C&WWTR>hN5=}i9H!HWl1qD;N)+m;sK!JMDC__L1IehM%3;ye#pb{mp<
z?p%1(fH#fg_n7r<7=h-++ijRIi}a#gCJH5XikYZ$M5SMQY`L32MRI8!E{wQQgpZFD
zlYMu}3||z%E5^ee_JIq(hO2oiZ;%&Hafh@EY^R9_?JqEHC|3av(QW&Y%lhN&BVpU5
z5}V^8qIKr!+*qBZ;n@0#{AttA$<Zs<VRwGF_b32qDb1o03e13^4l*Sc6LNdKoY2~+
zwc3~#Vx0UP#{|qx%BFYEnwv-gaV+j_M5p4s{yWFGmucK|X(aJPjp*hfG|cXI!f;3H
zm7y#Og&#obDh~n$`l*wkBIuOZB(^LGB<0{1gzspgA7sEVMKYK7)dc>wQ>@G+b-%R;
zw_gCp)E_QJ*{`f>tM(?tr6NmR|1WFA52~qa4tgwrJ5p8XD~Nk6K0e4(qeIY#{1_E@
zuPXj2<b%2_mRUG%EGAB{ZtI6ocTv>{zV;MofZ~Z02B`I`u^f~bdFwBV6Y+N=X5`#p
zROo`NhYA+`3FZ>W;W5Nhl-Zj7!wF`Z$|=ob-pNnO{w)MKQ4E&C@~@uqyJq!Hilb(f
zj{&ToK9V9anVx$&T-{c9(y4LweUxr>D<w-@)5F@N6Ii7E{-05Qmjo5kJ9I=MJHTFK
zi|pxScyo>LqCV^aqVsr4pnY(Z1IUp;GI7745lT#7RLEJSEkOx_cgqn}cmq6xtr%*t
z?<>dvaX112aaeHW)QxdKJV+-}oht<xcNm=@Q>0`u7%3i%m-L!Aj392PKVVvlb#n#6
zw>gYHEcYh*og}@I79wUaR-Qo`W&9SCd&Dc5fJVK-p3XuD>wFGpkckJ|My4O_(Q8v&
zqv%=Ah1o<5Qeef8Km;QGP+)4}tk|mawGbfau}xglXRa~A#Z{VjEJk?R@sdP%6xT|J
zWCmS_CBek*G%;prdP65BQdsk^z&w@w!-KUd;pc9;zmYTssoF&G*{Y)l_|>byUU`Jy
z9@X#`!0)xlYDGQ18`{OHMBs|+UHygU&vHeDGD2~Ttr_AsRjxfnC0ar6;`ry}tR-!^
zdtVhGHat1yvu499X%8E%aT8gvdo)#>nv*aqkvkC?9_fNRE<&_v>77fN;+1T$b2Pox
z-(qY5l+djQUr`aROLfRmhY+9~O^;aD;f=M57x{F1>AKj|H7Y^0@^R~D23A9>U5--v
z-po!&(0aVUrqa?T)uu05?X0zSbb&PEcCJe7*u#2vPB<y`$XLi8rP0~gM8yXs0DW4c
z;7#R-Xev@^jHow$RUW%m!eo5Qz-;`R(;mg4*v@{&7Z4oq7eC6K312Is9wcBkj;VrP
z>*1sW)6%v^MsuaSin%+$%ugIu24u&K>`)LG`{Yp)_e{NiZ_8J)C{Q8B=BpFb%Rl}e
z8yD1^>B)i-JULj$Zl13Vil`SQw9)ExnV1<FEHo9-pkRHnT{H|2$9~Y{7XTge@Qarc
z{H2}uq)r~TnBYm_y$rQb4V3c?idu@|F{D6JvM;u76JJ*+W<z*JDxtS^SWimJHh5OB
z^CwO-k>evhHtrl2!-lP|2Z|K*Rv|;U(bKTUH+V<w`t7X5=`sDKdVDK0tMN6jF5;Y^
zf|-HTeJcoJgBbD=dnk5VB~iN?Hn4<VX1fFV`LGK;b_TIY8BwiXW*$_F7xNG`Ajmrm
z#2swKny10%r6jLIS7lSY!}5ukeTV)~olj_7{wt{^4{otC<fE}9&{~t%%Ss5}`hH_E
zpgT2Ba8R^Pi|5xn{4=OIDu(p(R|Xo#*%S}HgA9At35dEvrTX>gMSMcpT^!~u6ZvA)
zo=>9ecX|>(0RiMPJ$DRj8(kq3G#VqkG-hlP(#>>Y<SaG19)$P|$=^{41TyZ%VK(mj
z*~eO^G1)40NJDYoo!R6~c-^mT1tYzQvUe1u`o5=hBR&Hk>H(<D?-(Ej{1^>5H@oSQ
zw3%9GTJ8~iH*RWV`^D8jqF|djt4r$Nz@E}eaR<+s6#k5~S>SgA?YLY))q}`Qug1MO
zc@;Fkb~%u$n2JBoZo#>Dle=$-19lx$(*ASmzMJp<2<6=a16VUMz8M3}FL}aDCB=ol
z6Nv+BFuK!mT&PmKhdmhU37^iN@<9E!ZBx!7dk1BU1?iUWF0W;-)x4q{_V!2aB}-GK
zW>}d5I$ihMadxKpFNNYN{OBpjf_0w|Aspn)CXT^KeWwQcZJ+6#h-0a4tL=hwe)H_H
zmJR~5YDdYYer$xkvA2UAVc@dJKFnh;UYk(EuL0s$hI}Zt*y6O-frWe?-P>l(6zjLG
z*k11dbof%a+}N=0I>8zEuFSh!$t}`S+L-m3s18a3fqmo2NTN+iq@i+soI|Kdq!xx9
z3pGBXJBnw(Z5NdsyfV9r^9*e#!doIyfVS!0Qy7#~TU(^p``WdyU363DQaA*Q6@C;5
zsQrd-@h)IZEI9ST<h(ed+DsMrXW*9lqEdE<_kbwkqn!naC)|;eeG%7dhJ8Vuq4a5g
ztxJTS&u*SK_Pb+F#X5-B`?Ag{3X{>wn?hDN7{9hQZvH|wv8XS*0iY$lND7<7)`;fd
z%vN$LiozOie*L;K4q3TGPc7eXdId=$ufn-j4@dq(g=|o}?dANEF4MdDC1xGZ{f9$A
zXSVKz`KbdU1u6Iscv$GF6<Ff)ZPumbg(mhQHMx)~9ac8)i@b=L_3ag$gFcJ$PhJ=R
z3=RliCEAbKfksZ3=O6fC2dK4)mUAIrW>b+V;}euTv|>>0wgpPeTx|wveD*6kcy$Uz
z(&BO(7>*7yhik^M1WRrA)!<jTH|O$awhV8QPv8CSDJfwZ{LWThSTImy<q(%&-~0Cq
z6nA7+HULU0Tn0t{BV$RO(yd2jVp){DX>XUnm}Fj0><r~9=dP2jGZern<ufnOKQrEN
zOzIxWUv39jRKQh~7pyZ(5TVA%9yGIQgLAJ=K40qW9Yo_oAH$bXxq@PTC1F^>y0Uz-
zj+{E*&iGQ3KbSasvvvgiJP~KaLSHb?epd=ln_aI%QPaJ1s2YU#6_3Jsl0<$Yr!UZ9
zXs+RC4Gk;_h(8a2XTLd+&wy2p$PH=6?|%}O!8YsEaMl6X(ub=`+&5qUVr^Ox`Ak+z
zkzZjIPp|zhIvg#+@#$i7>RI4f8y3a4kBh(1!LT0~1{$^cZ~FI^-_j3U_`n=xbt_}Z
zBeu@cb@EXWR=NGm;8?<6S|Xz0M)~#bb>T=_FgpGg>i5y|+jUQBi7NxZK4`dINruo$
zG!!)#sRfZeubhICbS|9u5^*q-{;(>3kxi3Npg=u;5lZ<%5uYAdJ)r|Id%+I+#^4bw
z>r(3!vp`1zx#}KK%8v`a1Z7<0#Tc)Q%QGvhV5il9Je6H<9%ah_^Xk@3%4?ZAB7g`F
z@cS7i`$K)xQ4h*1Zwqo{14MDq+#hkaUe05(yN8Ii9&y`UbM%=4J<KiBo{U2VJ-rk#
z@xepiu$={T@X*<cz)HhrItPse;WMY3d}zR~X;z&M#7Fd%wm}sg$lb|P;6sy++{NA0
z@O6l5F$E<RQN=`De3Jx0Jbz`+PFRSn7%p67N2F<!d=mWUeU$huja<(<WX_2`zI_S1
zJ?mp4i!|jS3Z{-?cBryK&p@Uhz}5k7e5_4BPIjd06RVcN9psL(0KAwRdB|3$3RW;7
zV$t%z1QnOR7+6Jtu^(0vfImtz(m^<cj73L~%BhyGJ4&18tT>7n>9uWww`h2)SZVKB
zUd<)%TOQRi7?n*j*)#g8D!6D3_3s0^WrNeCKMNj9$08COwc5-Hkisl8zS(zirarTM
zKqmybrkVK1CN@p=d6aSfnx^gR#l*f@_exH0LzdpX<lj9Yto2O7I0~;A$+O&DJUZni
zE;yf#q?O}vJe)Z^(hT&e{d&S@Ni9tGb?A#{DMfJR=u4vBuJo`GT$Y-_aUj&&wlsWm
zzv`XW41zrgn;%V9)`#^&gTGQLM7uSChd4#1WB3q84Gm6a=%^Z5)x!#$%pU;3k4{C<
z_KR*%A%=dw(XSoXIj$F#mu?`G3b<lF7<_iK-$uns-+{G+l!e$?T(Mtqb#MY*zq&2q
zgj9J<J(-GEs+&zSV+3gb^lHUzoHiX-$g`ySr})c5<6x$NR1fCKB=q@(r9P1BYh+lk
zCA#UMviIpbijwKtBxq`iY^4liyrqsN`th!~(-*+>8ZJv8lL%<P4xWjZ7MpXG=Awzm
zoq`%)lVk<rH=0>(lD&I7o|}Nrf|3~eNWyXNobJvyf*Br3WAV$vMMusIK#;$jWyf=b
zmIB->$B4-0<h)v5476Dp5@f1DFN)#ld5?NUIOF-{A43PNS?xrGkbCDT+&?EayBoYO
zy3=$zl`HDFX<_c*2LmOh(_FntuNp9rEhH_Rp1LjCD%1;kIqkSnG)!E~<}m`k=H*{u
zdvrVNwJfy*DD2e?vil!@`9iXsze0ByG;3sYU6x0kxW5sy8R|vI=#9|z-}s-0rfT)6
z^TKCmeEDPyIUvcpBS8^j3_<R6p7HpwHyyUth$PqcG!G`66;V0E&nCTGuW0hnMz0qt
zN#QWk(SX1s#RCXKGyd55&kE_1^}lkTEH)WWO&>62g5nBU!-E8$CO+N^E*cV{7oK2Z
zIG5epd95bRmAIJAEFUdb=O&M;7TRL=d}CPVA#*v)LkGDFTwo=OwqmS-gvS2tF^$NS
zXN6TLevk5{YuG0O8ltxh53A)+f8*sDL5W4F@A9kt4Z!0xy7$4x1XcJl_h{kJ+e`B9
zlRQ)r3r0lBc|a1;DGEog@|`_tz_|jhz{u{H9|O$^imrCJgZ8Lm!JZ=2S^~8+Gm&)a
zRPDnYLDpU<$zd6AxJYQZJ)Y-}RSJ+6a`=&L2n~Xu2%qOAzi(C51hCoNx7FQ~uBHTj
z2<5BJM6f26tTv>*ceMD<4@p-E<1NdyF5Jr*_Z#)Y?P$It4Fy(y^kOI6rr}$Dgb186
z>VCd#c@vY_h5wdyaO(AlQDpHwvkOM@tu;Z}%Iw>9OG=b#w3}pp^y|rqgrt65sRsO3
zUNgp)q_;%n*2WeZ!TPXQUZJ6(u(I}z4M^0_3O)>c@l$OAp-)hkO=kNQoi#6kug_4g
zz8-1)NLO~l?0DWObCJ`pjF<<~=)>U*8_^stJ-7r+Z6u<o^A$+1$-;4auMi=}jdmw{
z-W=MTM2c_d;Vw`)^jJC5z~2>HYTwwgt;t(F6|z!vz5R<~l*zjHJHsEzIYKgnJvx$m
zsbOe+nRxWZHJP+W7TK~Tk0JFSo=<U)Ugt#QUf=0Z137MJzeGEWUxz+H<V69_IE@=d
z@72k^GRXnK0zEwXiK*K{^y(^(OAukiDBwXh1zQG&m~)ADte;!f>o2JuVQBPL_kL0B
z7-JjzE8)c=*8!iT?fN%P?SA{`abf4$)5sMLblITfh1U1iFz?mQ%q~Sqq-TW-E-7(4
zTepGp(K*ruQ8;>l)n8Yk&M4@5RILKM(lOD*RN@J^95!po<McX&+LF-tg$(iCzb1sl
zbO;u)jt^)!<`vQz>Va{EY=f7>`0dsy7}+uYzU}B{1_H1LuyRAbIQD)btVPRF)<rSv
zJ6-qgiAXuf#=3zVb@2VylFVB2IFsY`a&PeRi1m;mEo$3n>0os`vi(}ZWiexE2VS~A
zG!=+E(q6k(lU+MpUXHd&;~K2I-?!k%PQnJ$Tp<(;0bt=QD@SN#B@R!^D1vJB;(Gx-
zSndbmm5M*z_YZW^Ibmh#CfaQFAIf-17(jyB%v0wMf@<OD42uyWDjP226r8n(8WF3T
zouLzk=MAJ+@eE#PJ<t|BSqgEM88{bC53RnjwGb(KATM0o`_IEQHxGYk@Dl-=a+~o=
zsJWphhSMY;N#1ucPc%wZq!6sBIrpfK*fz*kj^oYgd^SFOGOs60HO>n}RdCWsU3iCy
z4!q=EfXV22;Hhw}5zfnt;2(eZ7t5MMDM^pFe)$mSqo|gyO|vv%oAX&IU0HhT8^qB9
zUO~$biZ|qYQW(bV<sf_hsaT-6B@saS>l~6s9|bRd$=049$z`0n!2RZiC1sh4d8ygB
zO>}YUsnf99cJy&uFSn<Vh0D9z<SG4{Q++pR3Xu?(nU6%shv-#=T$LS?caqS^j%uM<
z3f5em@$Q`nv#$mr#*PKw9=$f7{!JB#KP21Wf?Ec!@L5!GyB+~^8}~Ki%RAqZ)`_Mr
zwrB$pl_YW?`zFuZWd1>>fuZpf?iBDN;80!M4rMc*v1r*n@5y8fbMPZi>!pc)^y5|S
zonxGyOe4Jt276w;PLkKwWne5;g$+`8vp#)N!<Qs8mrej;@`Z0vGYq9_z`;14=R^1j
zrd}Q2Y<%E#>n+VtiQJL{DUSRwpr#V5Feh$6y&@cq{uM>Sdj<N=gDw8u8o4@Cwt?^;
zkwzp|YwY7ZZVH$zCDaZnBmeBg`yex_EsHM2AchqpbJE(AMl}2avZ4?c0KIyIniuP<
zlbd$3YNn_`q&&G5`*0-1Mb~kHS@_u8u-%>_Dt}JLU0Q18ol^_L$&lztnzas$St4~Z
zM2y55Rhh$h=GQYm!NlSM?3e23x*Q;T(u66_k8Sz}{Ut5(n$x_87Ose?f%17odcPGL
z#)%dm5+t_aNiA$;alx*OVG*-DZ&Fja4WvyC2n`OAbrFWf4-wc?G2S^tw+YWC<mQRm
zR+Ama{bQ4byMxjIJvw0u)S?lNY~0oYBxCW&Gp$L1x9a$H=rV_KNj$DOlILF_KoIgD
zepuZ{VZ;?ve~7*{@t50LjS35G<&ucuWo`M))>vX2<%f54_LR8YWQ_@A+?o<+**RG$
z^1tr)<27UBt&g#vPdoa0gU$?2@I~HO)23T}ADRAw2?0{)w3$E6kG>es7V4WEUxH`C
z07*RipZ1><P>+1p*Jn7DV@n{QwzyHze~xKJ^N`kOw#uAcEIF~3*LG9-5fTS*6R7|A
zT~IX*2b?4#GW{+Wmj=N(f&Q&&m+B1(c{>e#8r4a}=dacj4YEXTX*FIm%;~-MhfwiH
z7Y*?yil`jXv~()rt?l&Dy+{e*To1^AH6iA*aQY%+hcW+CML-P_{(uqx+ro$*)@9<k
z$s@(!fbECCkqbnfBNm<ZZ`#p<AKNlZpI<4+GGrRENTzpY!h$4_(XsjNaB$rWzOkCP
zwEsMF>XFBQDCgynU{ggH3RT3_kZ);sX`)jrIu2s!(}c11Pj|^c%xE00!<~)$kMsQH
zX(kLvB6XcUQ7HKdWa{}|c8O%tNqF55<uB^RN7U_xs6h+lj$TJ#^5AF<7tZ2`r%!fn
z`UjI$hl;292Jg$D5UOOhz_rqnud0$T_1r(;JC7#d02i!x(+%PREoG7IJS|Z;JC~K&
z98FIYn)l)XN@r1GLq?pkXG-hmZ6Wd=8QrAq*BRBV-4x4W(_$%4RK&=Il@_^ti7W4t
z@Im-p8~?iGVYfKxXsBk@m3jJ<0xVq*i_6OPV6;o&17j^T^6OQnc#48PP3fW%viu%R
zsJhv$I1kV98dqZNj^;>j-4_T$iTPMZVi~isCPbgTIY+^3@9e;Ldte4r(_$P>4))=G
zqf27fb4jB!Eof9HU4EAmx(Pjz>pKapo<b?EbVBUkM1wj!C$}OKQn8%HlnA5Q%R4^Q
zreY?@Q9g#9V=6nwb?OSxBhQ|ET??mpW4uYqm<7UYq&+-9&u36{J{r8|z5z@19lDti
zT$xn{Y6y++$BumvLtFXv8GO`Tu7Vqlt184Y5o6!w{&E|<bAmwR!I2h8_PmS(ifOdg
z6@sfEI>SbSqwL|9+u<QD&{R|s>nEe;Nj0$gjP#G!wz_Dujr2sDgk1Jta8ZQ9)tWHt
zVC~ARb|tT+YY~-{{Y>Xba<#q3R@cUI#F=y&pYkh={f5TsMlW~?zw&Y74@-0x#XzHE
zxL*!&4o?3G@7WaK7+gpZsdAk%CKYo`q$2Be5=JXKx+A5ly<0Na$wgYlnpe{uINCk^
zS#a_pYLPuRV|-ZGIDK~6Qw6=Zd5W2*c)je@x`a?)8sf)MZGM-QO@$u)m9u0J`4;O*
z)~O}}U{^hgYQXm@3rC6%rI~V)Lz7w+GJeC9%cv{quCRS%qWP;-cF#fJF<ml=IG34x
zOvCS{mAZ=bq9;v@i3uh&(jwBkU;B5=93{(gyeP+-fb#a<`34I*O;j?aQ1dg5Qs7Pf
z+I--7(3)pgPrpP(rUv-bVYXjy-(hA^^6tmQ&O1$LN~K8#^OIC>MmCcJhjfxyc7q6~
zu{nZHjuXks|Fnvx9@V`uiI-FMjY9Au5?oe=be5xE%uq+`pcW3$YHA{;h|(5Z{qap%
z$1O=<9P`O1d6MFRpZm+3^^y0XMi$UC5+0pCqaY7LK-)P|vrgp#hIliNQCgC7yH7Ge
zFS2v<*!74SOSRHTX^)`0Qr?IgT<MzlLFbOlc^#vaU{AYvcC=1jGtN~tY|JFdY&uB+
zRl601)wovJCEf1SnljicuE9yxpbQrEZ8m2x`Lw(Hq5!k%kqN8s?2(P<BG8NgGk@05
zwD=Pcyd|&fd#tNz-!W>TYEluNrs|8^P0Or@LTfmMf}Z6gCRRm@@A!_@xRky?b;BmJ
zR=pY{|C#Pa8Z+5efI>1cCuI$~ZXhUO&D?jh?M>98YLFduzm8Ts=XgY7<QBzChUZhp
zD?Apqm9mLMO(}t2mf$akwR;U6i1TQo_2JuK79dq6P%HfP=iuv4sJmle7pH{k47|n7
ziD4z4#Xh{DabGf8P~F@Slb+IzAt7a&cAB|Aw}<5u1jeX0xvDXfHZ3hc9V9_HWu{1J
zlSc~mh6a<3jz#8BBeRjmHFY{lhlh(tmAgQw*(#(}?CXw>^9y4+MJT0ZP{xvKEbL^P
zRd=c!Nnwc-wwJlsmvq_Q<IZ9F<HbNG)35)e#G=Ptf(W*Peeu@OZGcQrjNTwifB4uj
z5v`#>kc#}sw4V%ewvzmsyD%W)z%QLqDw5oF4E8o#1dO;}Jx!^koke>~8JbeN^6~PK
z=1lRh99u-!V4jDAAznNrsJcPOzc&l{oy8}UmUD%ir8B$!(5uIMNM~vUN0<THJyV~)
z1RSagSA5(}3|R?(OK}epGCKgdg`|ki6L(NiyfG@KS?o+y^>Q&0WhId|wjaWhgCDhU
zaQ{p?kJ%rX#bV|@XQf&?6~sY$A0i=3VNy15FI34KWr%`4Q1?;Vk*!}}{6K)!c)TZN
zR3;*!NLa=^yBllkd0!;$a}$YhVT>}V{E1tlgX>Y&qu9AiW?K^y$Pxv0Sa2XDz%kK?
zqkM~uw9|DV0+LpP{T8caaVT3DvqAh53)yN~;~Ud+D@3XCu)N-wiO<-y<4)2y<+2te
zS;vzZk(2!EFZ{Q>+(H<0F?JOSoNvooJat%|7>FG{xUQ@XY+G>5ifc^_CV3irB*+pw
zPUmDIH|PG&LoGF&bOMi}<5jzr@TK>uQ&R004UJW~|1#|XOrGVw19NMd8K85%DpXI?
z(E<vv_#klRshOvYRw;g>>pnYwk$NL?0n7Pa-FD%v9nJ*BNKxcTH6@&6D*QSs$B)#>
zW&i1o)(00<in-<(>y*cuG;>KG1-gg74B=CcnV(qj-+cT2s$Q|HtjMC=QD(=2pu%M2
z*}ATV<M;PJcs=f3QoY|@#7ntL(wk)<>}DA(?>XP6Ss`u%5$QNlNd$G+*Tal@H6lR3
zS^EdYw4x?v+21h83uwWZCh#U@HE3|z1d8HtaWW7pX(efUugB&M0F#Y<#8WnP<UsO}
z*%W^|c|nH+zUXgdSbHkM6OzT>C+M>!GM)sn>bqeJW0hGGc2{>g$xeJZNJUBG%qmj*
zi*5V$LCM$*++f3(^>)r_KzlS!5%VKwh9~khD%hdZm0cH3K3meWPIRRcng9Z#iV|38
z!<@13wK~w=#y*1k@%e7kL^<D1t|Lh*E{Ujv2ME=Y$XZTvt%TZM9T9`O5DvCwr215B
zyF8p1dBJmS)Nc*g@AaG%$B3c{j44fF@W&5Dkh6`Zj(?g>yh*zwusS@JekoZ_Xqklj
z3><7g>lo!znz|>b$d$4uD1iIyI<xLs#JgDb`S2ExNTWl=XJ4LLyM>89X~U`3aK7wy
z4qa_>z?dhYgci^hKH8m*-+`;KZ>Ljk$HlA62;MHS3+Jk8n{h*CBG6+fbCEqyzN_Kc
zv+H$1=zESLWjk@N2>{aL2WxBsvySou(X-@7rD6mzmKOT3bu#oOm31n-7Kp>;stYSP
zd9(_IOX}V1b7K`v?X2^iyQh!A{h5vgZ)2o?W@-L)Hn@GRv->PDCgdQJ9)(t-F|Dd^
z`%Synw9)UaAIN8#tSm|ao#KoE4YQJ|7duKZ7{biR6o)N|X+%Tb(hA|}j?bb?B^^<{
z<C4t~k2&5c)2leEj_ciHk8!cD9v-*8+mUloqxyX)_0o+x+|2_eXBw-UP4E-Og!$8}
zWi%TLh;*T<hSxQl+6HC^A0NI7mSd1609k#am_IJjyzj;z)-0bhl>7o<dKX(`_{;g?
zhYRV&uy@fYUai)XLP}yHfx;LYJ7+1kl#BYVYDdCdb%SA`g-JH=N>L&3Gfef5m^ucV
z%$D7f0TPpk*HP=|)MX&3EIHfG(X~bl=gT`Em<73%RO<M8udtuC__bu$prP;ZtBktk
zKR-fgj-zv7r3}rGIM-zDpIQuKyJV_U{iufCr9|jeJ9=WdVAgv3TC95TIGm|^kP?&w
z@G}K^(%QT+%o>^;{9&--BLcR-YGHM;8vE7M-ut1IN0pMnsP1kRsZz1nydr};4VWD4
z;xw10x(_hUoJ~Hv4BZx|GQs2f=Cw0ZG0A8ebvcs9Q+lXAyS&RRqhY}hI;LVdh<@J=
zHRd9Z;xYaa1`+*{4-<uQjQMD8apDD{&yn}5p+Y!uL8Xv)e3;}+D&bqhwJHJ{I*q<E
z8ovwU`1+aa*;202(=y9PYr=)zA!as{w>(nTK4Y*9E*3ZP_8;Ua5M}|L#$MOxWXj$D
z^|s1Ub7N#zD$ZU5Bu>Hepi`t}Zj{6AD$a`Uw_)dPaZHm~gIsJfVda&*TE2pW$BMC%
zj4nvX*oegR;`tMJgKs>2J04nvIof2|+_AAYwLeEZ^5sjp4x^|%(_x~rYdOd}qbj8A
zQTe2Fy$3uF{S?JclRACKuGWR0!(Bb<$K%-8%8b5ZH1V%GS7rkEU`*awv&G7o*T*Gv
zcO^fZKns_kpPQ(xm}<s@MBo=IDBcnQYIBlfDp^gWyo!X2?u1)^<Uz16cs^h?cf=e+
z>3s;DN63?>!fnPPdFizUr3XlD-g-^^*;}p>dP8xQMOMgPf!~$euDHzCmwf8F-^V9|
z?}Xt%WJsYNw&dVT7Fgb3GqGpLV%?>={@Z455sl6fdXI|iKVU_jcpCKZPK$=kzV2jQ
zCDpD8gQplKB`<IJ+vLmi<P&j@+mAb>t(je=;CR9bHMawbY0vIYeR3C@CTh{HF?6PO
za7%I@|NSo>GUhTZHJICr#u=Rm&NpbgT0#WlMC_ayL;2}H)x7j{`S04NJ4>sb|1RD7
zsmHeS_6QbnjE#@!z*w~MV;s_-2;FY+Gb>-+UBwH?JZr6mwaXqcCOQyR)Jl8{Lzz#q
z;2xa7MHh&B`aYbJU~Dw4PB+He&RVry4h}UTcp=mtQLwwt|Bm}(OTb^d4-cHTm*fc_
zYQC`xCwAbpc2*B(C3k3_^g2n;$a4ssM5-hsLX8$v9(+G%_08)XZC_1NwwvVuL|r@^
z!&D}39LHas{KI0I=%<ADGsTC=IOVms8E!e+pD231UpxJt3MoXk;%ktUkwiSMT~F1e
zQ%X>v*T`k$kvz?<gkJ{^w#KiQ_9LYmGZwL8rPJ@DYs<VagoqfhzLxL^;?_xksIW2D
zurENU8<vMx_$AV)0j4U+>k+wz&2y|KnEk?XpZD75@p*<=BM=4!PoH{v<238``+g=G
z)!I1~NSk0AIrmh4wwP^vExD&al5aqd)Qgg@D!6`7541irRzRJP#@=zso=)NIaa%pv
z#BR(GsV?_Y2tCwk#dVYZMN&%DZmMY#gKebn<5N}E=bfCGobzP}1_s<);p^W|KRwT5
zI`vTI2vj2628))1W)Mu1+_dR;#iesk1R)S)gi%eNM_J0M9c70qK@O}$_<-L?L0i~(
z^sZ59gl*M*q08BUJ-H*OJ$aMkwz7^R1^=9d-V;|ZSeThH#_hEq9>lawfxs*k52~vD
zlUkR+K8&!lV;-ScfBi8@q*x<d=BFyDIlwOwQrVcVn~vgx*MdC!syTigPszp8JEZ|K
z%?U?l7XE;$>C*GM|84d0eW(9WvMp1{YIn;=p?oyIl_Vrmt1&X{X%`%R6^9P=nP7bS
z2!^Er7WH6NftC`w!kVxC=Gya1-E!V_dGqz7I4h|QP!}a-9!j4UY6g_0^3h188J$|m
zdL)_4RtCUhOPj4iEuqE_XP>eH7zjTwC44{SozM-v%85-)`?l3<d1Cn*j5LNrr$g_^
zo!ai;m<EZy)Y*uv$ORFs@Jd{3JzcFJz6Ik03t<t$I<4QWNNRO;PU<0WBoYKkvc%LU
zWrY+Z4dN!jV-ZOcK@AhV6zU{1D4J4ZzZ`xrcG8A@Udmi-^v4d1rL;@JITUn1im>bW
zttoMulWgZ$NJ8^CQgn&#xT%R~I)I6YAFL1YAz;vithXS3;aXh_^)Lus=jWgN$!X#y
zv8>vw(P()oo>wo`hNxz^$tTTSB0cHU4)K(&)9c*jRkvgLW~DRU$LilOJq1LiTWr$z
zx#m0I^H7zcrL0ncafCi;<0+{*j4tuN>A1v(U#0QirM^CnJT3gOLVQspd~Ph7^N3sM
z_}6<3)7m4n97V0_w@p*`I5!#Li_Y8b9NgVvPFRtYwy}N4!xQ9*X)*RLw=jR&!h(yQ
zF6x(Gu%$nJ^18!e>(};urI(efker~JPr!EjE*rhq)_rYCA^}EHJRCO>Gh<lCOP#qY
z>aJkM1?O?#ExPh|B&7fr;yb4!VLK30vT==#j&O+FcQ^DuP%Ee>^|C5_Y6lk;zWDix
zesM88{Cbnzif-18R5gI$n~|X?bIH8b4?r@Pm93iIJ27iRiSLI<KTZ4=Cos%rms79|
zlX+S}dh(T)Zz~Q?Etq^}*v9w_Ie2CFtlWqejwY+q`O|{(^!V1gNP2G#FRC#X(tr1-
zcTc}=91H0hMwHF?79O?%+pggyTgz=FoSB(e6S8!awtQ_ol;X)$rG~9xInQ4fgZybU
zep2EzhSKn5L!Ju0J`vU_l0jirAJ@`O;nkPDHmMnU&MBJ`qp#eT`RTk{mo<zi<W+be
ztJoK2)9(#N7oy3jO?GLkD;{)iba{7)$AW1vz!@a8Dd@Ldlb=6~ku)=0?n!osJ$g@k
z@$OJ(WW1}G8Z-EkuT8Mfd9aQwBBI$FTI=y^-0Y0+pQ-sh;Sne1IN9ROk0&y!R|4WI
z7yd6%ZJJUxdD4>!0whU(6%E|x+VS{(U&uFsA5~cGj$R=<x}!4h!b}E(o<OILUKs{;
zc9dk{N|b*DhJzL_*{<Va8fLl{ymx82VZ44il7;9X7xUx3jI!!j_HAoZJ*)}mac=d>
zUq*f>mx<heC-15Zr|vLC?7GrExMvieEq~XNm%l32iSkg_Dz)O2iu`V+)U<g0;Yh39
z<L<l(9#5Oj3RiwMhoj)8m$Xa%kd-;1(At#3_=Xkum{`Pk=xzEjGeN0O+~Y&$qGaSo
z=5KD_9O=BNB2w1ot@PJo00mT;5oUt7Po$yWZnGm3C?A*eiPH1vcs{+cA&k<&mj&;G
z#*D2NMKVwdw>3QJ-shbqW*L?cCl_;l&#!u{!B!7>)CwPZhtu*$1XP7Uuu0XYacj3(
zm;39})ukS?{Vpl%9|x}W^cJTv(~yvyeKD#eK~7xwIxZ$gV{3)D^cO06pkO0m93?2&
zQ_hU$v;-4G6O)&bTYsFNFH_<@sl@hWUD;zDI;k#U?L~x-iVks=iE;(+hUkk*8+?ej
zd1PridH7>>>M!6^Ko3CSdyTj)l6(!$7LnFuMG^Yd&!%ruRV6TwvEKYknouFS=>l3J
zc$GQwMTLr>rI1g1SgDHXd%g&LMHlTrhNkAjPweogV{Z6;Pt-m=K%VtyFli0tRsKI-
zOwyCfQL!63rCk!l@~dn_)AT8qw9|PIPwG#Urqmx$F6uphFB~&;Ix{%O?Xt+prM>4F
z;6|VhLl(EJ`a)jzYFxO(^nJqWLrhrF=wAjDngGpIEAHP0i%CJtJyG`U*9sHLV_e0f
z<sJ{>7|56jy<w=OQhu2CAsb4~*$#ocYFJJUwMe%+)V%ZXe>I?>)so!mN&S#M2>s}-
z^3TCk<Bz;`A(KA(L;b$4nC(33)^RF~)kG|bYk9#uQiDa=u%-UNV+@@r&*MxH445f-
zE;SR`qd^5);@tj6J_wAoQoHv{8yi*gGsePOv0C<z>KBE5hDZ(yp+Ic4)I&SMg*RpG
zbj!<psB)Zu&Z>Xa_>`s{Y$w^|PwF}wJpR_({Fhy?knjaaEMjaX|4mGm5?!AHJ||J^
zaz(61BX9HizOx@StSoZnDJl)3U6wVlN)D`Fd!}UW5CUIu351T#1dQxpR(tlbuguy+
z@yJ@z?Ha^R)Q})?Q$g&F6&r{K@I9=+E*XwSlD--viHvm8%JV!T6Pkvm=Br906yM{L
z(I%EaJl1XXf#t5F#=oBsSGa6wz4GLn8Wo4g)cHQ9M;S(mO7UvwOa&T<Y)-4nrfwwx
z9EwFEA!O>w{@+vLb|UtKl!FPwp`=s)=mn0@`bUN=#{4)iKS>gDVo%XYOLRgyD}aXF
z(&}L1b)55Z%(ANCm#Y$;qsBD*(Ox}J9dUW3LfOD8!P%m0I_LY`g9u3dm;wKDV`MlO
zQ>Vr+C}*k`rtumm9ns81<M0`mErM3^IZm@QEXv7uzm>B<D~55}ujPOMwrS@mBaM3n
z5ftpvi%;X9tf$rhL8pkG<5bX$<Ua538vXwEq%sKjGODpGqdqgiPAhqO@mn?MTk=S1
zYr{94Yi@3%GDWAeI_5d;2?3iNPSQB0a%kV|zi&{eH{&so&Y9PbyIw@)UDVXao0hgG
zb-fZ;=lBN4^e(a(N<)x^0cf}q=Oe0Sy>nV+BunGTbKQiVNSPXv8~&+%+ZuMA7C+ms
zvt4w#>n;vvcg0lFugG>CU^oc{X>{{AOC|a8yhAz2GUOR;2(WsMWs|B$$U?r~ZB>ys
ziyYaqizu4$k{>aV5cwuZIVF;<qi^vi;pD`D2wyo7b3ZQS%_f4xE6uFd=z;0)R=s*n
z5Ip?a>H&PcI!8RO@ZB*|Nh0JN&)69iJk#lAlBFc@i$L2-BafV-mLkl8qW0$RmPyB9
zg8_a#olM1YWr*NPhr=XM?VGB$jlv)0@ro(EZ`7$QoP{Q?Q-E;&gN=~1trP%^mD&9!
z!bgPSH*mtfg*6`S7w({`87B8TeTLzBCf>|2o)Ge(aDt7iv<$ec5b%h^9{)oh?4LKV
z)LlmSX?9+8IA0G@u{#k8i6!{1C$dkW|H*XQtPEr^)W6haOcR!gaz(hrz{4`*IA)?r
zYi`p%ChA{(D2}&;O8hzwAIDFDIt~aC&Wk3g(`zn%`Fwi=0U(6Fwn+xOQ4HlS_%Zt0
z%J%!)O2dfRDbX8&;jL^pet|h#h&5Kho^Hfb(BP?ms@4U0pd{`6ec%ZB^qe5Zz1cD>
z>FOU@g%2@JMz-;@SoVC>>|=R_dBM{t>f;k>IUVPo!hmpJgU?+}IoXExWwpgz0VYOI
zuVTKb9=NcpU^dkLKy}S+>x0SXF)b_?x_&h~0#W_|bv&VQ*mnO_Ez`13Vbj|CL*Ba<
zu=R6o7o^JwG_yrt(oU;7;fslQ51iCd-+;vcr_Q+!Vzh|?E=&MYBgJ%-BR#s?QI(yq
zPNy>zgF4)lGXj(wpEUrqR^@Pt9?*gHhUTHFp4F$wM1l6-yux(6tmS0iXOL?|*cqiG
zjMP?pTDN9UTO&`JCEIQ${;6_IY&Cv?0D*DenjuW-Piyt|%ZV)uGq!Rb8Az4R>i9pg
zd7Eoo@F2ADd<0Cn5xT~EIOYoMt3$8-8djW*%xq{J&W2+KS}uXw?}x9c6-T^oFyF}k
z`;qN-eh!};Bpv1^>6~90eq$HN#VlBpxJ*oKNSzqChPSqH*8VE}(|CN;1(?sFU!i41
zmZ#sK=)o!GcrF_=@*5v<y>7urajtq)Z?~LHuFB(Is5z15eC}*^-d&zW0!H(k*va$c
zip#kD(LRlB0y4oz?_Z60C630cf~|$_$L1zpi|dZ(!Wcbq^1n_6->3`c6Jb}(Q936{
zOwigBl(PMD%=az(m2aKGVAH6Ad0=Avdu23~%hBW4pIqr_;r;%DV5N?bav=SA9Zhnw
zxBDN%R`Qq3Zg0B}U&tpjqPVy`DcbnbA#@>3`XVz=6tm@C)@FF(l9CtgB|M+NJJm!P
zCFCUUU;i>Z>!Tw?%84<kR5f?OT;izd<3|>mv|1$D4Ap>&%_1GKRzh45D-E^5_<Dn>
zofdy&iP5wDrV)HJbUsv@n<u1bvY4Q9tc47iTj6YyV`-A7XmJZNUtwhijqBf@ykk{8
zECvRUQ1I$9v#o^9F=U-ytX`oYr_)f!hp2mx)ZN05#;J)efWX){XHTOhBR~<+A4i;H
zF>z?4pqB13u}sl|bRM==HZu0KLdje$Rvuy~hv|LhXjd2^)z&y~m#a+D@F7acD!hdg
zq7&Mkk=P1MX*c}Fmf!wq^L+zmeE0cAZktr-8}ouPjGhIp3iF*Z(iVMAPq-Nk4I+Pw
zg82^q^K!VT9r}S+Vq=)g<aNyQfRp7SOOr>8y&08iwPh_IIW3AR8Kvo#0o^enxd{=q
zV^s0ATCO#C(!hY1<As}QSI$MXu+n6>Km=8_7=Z6Obp3@3>aK_(>yqG4qg_rXU7<GY
zyZ`$)<(!{VwTyN&6XTgnqa=nj3_E3eTdkB>>5Ds8bw(8}yS45r>ZH;r%;S~1=28-C
zrt#mS@ouq-BE66BuR&c~u4rr6(<l8gJg8QmmnON#EN~MWsjM6|QaDiBz<B5+Jwh7d
z4Sor0{Ax}!{Gk)GlQgoeYT_!qbn^SEJjIAvoNNoiOD7uSx<QI<GCmgk98OWIaq0$J
z3#6e|)&};FoJg}&5E+MVHlU?Pkzy@z^%BQIefhIaMW5241f`pD-;!lq=Vg~w+uiz>
zr`a?bcFMtZ_(^AiB44rFjPP{Oo$I$sl`$Bc8`*KldCqf}0!_N5rNTK0I^rxRZ&RZb
z?IJA&6t6eU;>6LNr*;789lyu6_~K@Hvd8^4L>kGo6@H>%OQ5(^q3$2WJR>lk?qi_#
z=AD>{)a*17I~xK`IT@mJqMf6ZVfyimJrc<}>C0^(hD6|-RG``(*8)N57b5w#zI(p4
zqbhKYh(IRr*dzsgPTM#3K|guuwdQ0*70XE|T1?_yB4)u&9u7Uu7$*-FyhIgJ@-5S%
zl%JX>vGpBQ!8?lzeur}IXvUMxPY1?Ie^IWIh~?Ics4ZO@iirOVSjF;~Keq{dZa#oo
zp%o>&Zu1T_46`go&-56j<+qs#z8q~W=LJq-yycqaSfN1M*km=t(4Saq`Ly7I+5u{~
zoyOOo<K$t8%DSv}ViO*2Hg;iWkaVq3hiLdS{%Vx&0zIfzqmuud1?WTP#nDpU_30ra
zir`fH%Ul6sy0S(@283z<u);)~zdrh`zJMKdkY7u2^z53AHc%JYyVMfStemIQ$f6ov
zMGR}0UN;`C$Pex)Z=>Q$T)yy;%WZV}75-R+JrWLTkp0A8y(95J3$j;el1kcQ7&Lt?
zB|AM^uN;~2{8Xnf$@iR*4$ermI=U}|5spD~aj1LBgLdMM5LrI5dsA_mLc-}FN0LQT
z@U}RvuLO_p=JRtTGrdKLBiPi|k;={uJR9Z6_@sw!)gwJsO!vplf<(u!-poH(`sh`)
z*j@t+30cnZm=zLh_(y~}W2VqBIP`*J{`BxUDXuD3ElA<eBv6ahjN+M~A&kv<%AQKw
zb!>Hx7dNYDpM$9FdqF7L=dS*PRTdy;g;Ew?_)D06X@ej3W;(y=z@eQlCd0HmXuL9@
zcg}k*?eqHVd0XBBTCC9*Zl?B)r1APY)#o$z4K91I{n7FqV^i1>I{?Hta9Q1va6LZL
zafa#{sjPo?Vsz*#0!IaD2mPy7<GjhSYqFP*ktIYQ68B-Ibw?H&EV&esMRAg2(Bp9o
zzDKE!c-f)o@l~+XeiT-fvP9x=6K!)5X~Kf#tko(^{QY4W#8iS{|NTaO^6%b%)`3L2
z+1&lng-kI7R#jj(z2YEai0{9$URGzV$8xyw)Ua~IpbmXivQBd9e`QR@JN!+AfGdBK
z-La_nP6i;W#otF;Zkqa|kBSVj2IoBih=CL*Jl%w35+t<CD5tw~&MwlNZnTtOU;(}N
z$iDBIPY7-LC#a91Z24IhB<RZ~OT|9ut>w;52P*O;;e>yWy0+{B#YE8R0zcx;XVg3W
z0a<;lufB9l5V0NAL{8UJz+jTm&iKY5Qf7PWtI5sSYlW15@j=LHgykDG(Oh&Wrfq6M
zN@elKfI>)1KIWXNwg~ZoS0wCW;>po0HDF1g=qzo==o_r%Ay$diOuv3hS8{?ut{{SM
zp{i0DFcM^>Z;syZV})}(%FY|_8(WjO@;C=tkVe>I+ncXJSQQ()_+_qypL=Tm@)T;h
zLC-z3^ZA(eZ-??K0e>eBuHVit3Fvn}Ul74j#c9b0RnAUScE<6P9QO{<3c!@zVDXL~
zB(3gHp9pw$;HMz<oUt151lJNF={`9-m5?d#^6)J2Hmrt71)?Lc78Cn>N25)nYLlUQ
z2b@ksmfy`kvd~va>-ddM>40&u+82kfb3l^y*V4f+Ac5Sm5mEp`A*Fz8P5b53-AXM$
z4msJ(^l!+B^n(s_o?3A)Fp^tDFr^jAcj}%x_V<izXZPiw45Kasf(WHWT!bSw)H`)J
z&{*XZ`MY%S=Y_nS&>10DOTro!K{V;Fy&Cp^>=W^MIFm_&61)Ka7JoGt?9DPwY*)IK
z{#9&hV#2Q(&;*^fc+*hS;Rm0mo4S+RaPa!XA=%=JbJrEcI<VvvrbWJ;qjyP;P-6X-
z(f$kEk|~&xl(pWqTchFVE)u-a6^q=K-f@ZFdYD#=zh|9IPF!Is;N<vHfQDHr_a=Y!
zTTW+VjDKkSR8^X_8T$*hwZp;CdqMq~Vn)uYtPeuP44T1#{eo2B<kP>!knbU2O5fHi
z6TaYPETIhpsp-0y3W@GS+p^C<L~Xwk2?1c7C@e+Mleu;9(?`XrG^J@OOr@q9`44_{
zXP|O{A2WfL=ZdY_ev1B0kx(<@NQr&f4(Mah$m%mQ0)Ak-`Ez}1Fr5C*N3fF(oKC=t
zVPPr9@&&s*a_bZpGmfr&v*OXr%kx1xlIWyjyP5&A;C%bM)af2vwzF}~()ZbA662P`
zbi$xN#e7m{3h<jS@Cm7H)l@#A=)Cri`1My3$VkTO>N8H;6JGl;e99@~t6>x~XFWz9
ze8E3;*ibus_ZRqJFS(Mq+q0fu24C)==QYbz*bYVI&8!Tjfgh~patFxYl6UgauJ5ck
zSQaX7UmRbJW~*(FB{EO%A6<fT3N7o&rE*z4^h+US2WKjn3JlKvsxfg5W_{K0o+lky
z-xAY}@^HcDlF)d1tcQ62#gGB)TbAFiPa&f05lqpOgm0j?G|$9yT$z%x8@LznUjU9k
zaliXQV0k~5O~wYJc?05$OsXw<?YK7R9E-=a(9P>}bC1$tG)rBLdTmBjWJ5u@9@#uq
z5mRoxMaRzG4`e&;OrPY;o|<o=Yd!wh^;KX1a1N=7klh5E0I{?VO5_NzH-JJ(7+fBa
zG2GXXhbvj8jsTghX@G_iq2l8>aJxmvy=wT89x(4Ki4^7*6YUI@(jfTd`SZ*w)Rh9A
zxm%GW2-52=vRlHS+^td#0Xh(%Lbb!j2mrxIK<1pXeqH(bAxN2bPrwyb0n=9(J`s<O
zSTj02Mu}F3d-UASSqR0<)*wJ-^NT{6yTU>ohTS{Qs7J1>2%GxkQd)#)8$FbxAGm+|
zgy+Xc<)+N{j&8C4b6yzpz|l_J&N~E(>+zXh9&@oJ3vmpT0MpsGzO{*_F`Y6EKoHFX
zoOA$F6`00ok>k=4`BTLHocAB`<Bxv<2=Fln=5K!k1#t8m?zcNb&`+3+Xmia2sOS(e
zKu2+3oSCwmUmyZJAJ59O9>)oR;F1UMcvQL4UL8`=f-ECUz!0`>k+)tQpKtEkOR}od
zpO-z#e}Wba=NZO4WP$5VKRj@+!)-;m%DhPA$@_7ZOx0DTas=WpO<J^4Oxf=s!H0uA
z;M0huH=CbM1F^J27zR~^c{)jo3a#aH<}*`yryi>lxZS*@>qyxWmXBpcP<2r>+eC#>
zNyu-~7Wb*WIs=X?hC2D2Xi31O5jF!<vtLGhe}M#~p~h+=fh14I>vr~1J!`H|8sgNw
zn^P&ZPr0`NQKGMmp*OM%M9cUq7io`lW_ZFu()5hUu1yq{ky?q??cp>EgiQ^mT~Rtu
z&{G_=0tB3?L13k0L^=jhDUq?OT+2NB?v8=z8xZT*sF{2XRH{wK!WQmmU7_uSoVT#E
zbpEEr3Q5y#u#~fOcJB@GFTcJnO>lJF?*|+X8;UtrULtNRYQY^VPj98@Gx!5!n_38m
z&{Rl7aP+7&n5+SAD!APa97hT;KR0u_SYvL5i=fDjjy(DV4~cn{ivX<srm>>Y7JBfS
zN@6>^>Cu&E$JdB}`MJ!O^ShyJz=JZhb2!T<4F?+=B~UR3J1!>ModYqLM>mgdX~Wp&
zxNV*LHV>itn!7_d5fYMow*>-LIb=p<GVJCR&I2ng3dJJ?>>5u<8xU_~`?o6hX}!b5
zF`p4G)a)Z591wGyMJlz~fs2t4o0ztZJk}U99D`q?fBj7&pQrTF=Uk<VOMxnGw=<@Z
zA*xt}9=>bCr?(&R!`lyVIDY)$C;WdH&G_+$KgaF3Vax%=2R{5Ac11%QF~QLryeVwX
zO5xN<Y76fQA{b-f`HZ&R=W%1>t2n6X$z`-@MVbNx!!>;yWC}>>k!8v0H+!XLv_sAM
zpG50`i=ee5LfAa)H5klS_%k|3hT^WYB2AEp;F=Tmd_+H#84a$gMYk%vn$W~ONE)au
z@|d!fB*u6Q>X9_Gn<30H-_UodC!$(Kfs*TtNI;|_B~jTD;C+RM*PhqUQYUe3Jj*j}
zak3W6CSI!=4psJC8Ri)7xUMH2&nL^O5yQV+=qxQyw)BckWCSZX3T)0~<SZm~dJ*gd
zJhjx62iae;K+amu&0?%*OhpVN+`90bWt9(NMNq}rPry?IEe2t^M`JlO=pj))C<?G*
zDEBZ@d;wVYX^QLl0FfLOuhC1#SG23ZWP&SS9d#|~E|htA!p;?;sI)E@u!%yB6X*x3
z+JGFKRT{mYxm{lW`1Ms_pmjy-jk&+oGr-}<3aQJ&^EO)~zX;KDzJNCL%)bK&UMJ8s
zMk1f*f!+jzL!XrS`ZXkm9;7P9bwx?GgrQPo=$s>snU6PM4YVn&pBqOnj3~RcoSK&;
zNKRR!_njn=#ZPEZwM9tcF3`^dka_XQLCz5eDnN%LORS=SPm&osOQ~87`?<!%^}OJ2
z=+e-S1KJK8{UlnA?=zcKe)B@Mkp^u?2(8VS;I<vqOcBJh(Tib7FTe&T_St~@1;_(k
zPmDS792X`Wy`SjEQTJgsOBskZ9Q~ljW*8RL0x^HRLpgprJM*>PFsBni2jS^Fj_?l5
zFpQN28lK~vSIAb6$V1b{DidH*0iYkaop+qQN7Usw(Ri?q{)XFm$4v$8xS_Wbzu52a
z{{D^uLvK`q76xZUGIG_3`hBAL#O-)PKO%yGDLBp}ko}P8pxGm06uL$^w%MT`EuD=u
z?!776!7|iu_~J=eHmV{(ti(u=3I|Bkyv7qYC$8rcHkq5%kd4-lFz|C=$OVk$v6>kk
zj|b$simVu=!=Y+p$W3~cT4x!~MpH9+9&xjc1GXR$!EqLgB@sFz%N&CZ@{U;6*CZ$1
zEr&;2uQLTYHj*h1ddz!Tapxidc{JH-c#awb1TPy~7%S5iJ#}G5C+EO*U7Yj?&TEO#
za>ig11c*7n9wDow(#zv@j7qCzWHDb?HzNzfKr)7q=AM7c1Emq2g{L$13Lc{d3>6u0
zD72rImWc*wF>DUl8%sGWl`g<7&Qw-1=}O%o!`O^0)N6p!gPLg?S<hR}T|L_Z0&xZw
zh$zUAWvKbR`Ix>y$0895T0f!*5<)kcuQkv2N-nl65b|rTuk1ibAxL!QsD=;GqsCy@
zD7JSTDo3_XY8n#{Z)X^#co+(>W?RdJrgBjh8HAp$b5yQpt~muQBuz!5DGfa_6$O-|
zlZ9m+iN?s_pVOhO$Gn8BJEwgV&wKLK6p1qtvNdy*B~B_7t2NqDsZE1loT?@kKu8_K
zEzmj<T<=nf#yvKDUK2KG#Cw!M>Xw#iEe%vO)-iLyq8zgyH=O4k$8o|PQJyugs0aax
zJ?=|csZf9xn6Z(E0EoHVG4k^oPh8_-D(B7yE{<UXGEzA@AT2yT)|sm+rel%w8bfdJ
zQ1(sneA08G9AqOYk$7)ENW9vkImoDuAc=^|5!%vwq+MDA%uxNtA!z4`+ovDUo8oqR
z!`t}=f#Bou#5E`SAh02*OYc-Ayt7O-wI*msgtDg%h(tK?cI)``=>*iFO=1p(3MTsz
zK%(MABt|ORw3x-w6nbP|gKzutRFtv>p6PXE$X}@3=D0%k$0Ug>I3UfHrG^3G#t0Cu
z5;F#39)?FFBA|?91BqYs$fMEBBJNdFv@v2#tmpJXfr>#gl1Izb+kt+BQcfauD&SP=
z#hi~PjYwryQc3jaNM|6UmF&Dh4|y@(7p_n5(kt4zfdBHg0TiFBI8q+(2X*sg?~XLe
z>1^$~fS*(j%ih2Z1n)VeH5&s}>EpKdq1R;r!p*x*<CuUefWZO;^<8|!T4t|vo@*Ls
zek?DCkrA+8#gavGB(i5p8MHNJv2KrFm>jQ~L|Ct=TdX^bc`#D{hOD&g%Uc|HX)Vu{
zi<ZP*Bz1noazGfl%IC~zc%iQKQ7=y6zwG)dFyQp`#~2k#n<P19wUo_J-jSnWxNJIL
z7C_*LsfUJOO;3-gKx?tsE}Z(8hYP@UUE#NuC%<wq=h<)^9lbSr-5VM+3MG5dBg1&`
zA`QlaG1MelWFCQ(@v8%FX*5xXHHACTVIsUL^>MC&1)F+|331@M4CWW0G5s$QeUnyA
zf(odRk*?4pVkT;zHd*4xhzC;(GNT<bTyP#Y9PNg)-&X^j86{fPJ7GukHJg_uQf3%W
z!<+-1oDqdqfxW7tYlOvn?~tZ=To;-|UTauX8`O++1g|2?>00kN&U-|2dgQpS>jNf&
z@$rEB{DM%l_!j+WIQofJO{+NLLYI;pHW`|a=tPEu9bt^m`+@Vg;dY+E?w0tzx3?#5
zx1aIL?|zF14ShtsA`p~eUkG5rsLm0dyflMh0CvGKp$G8xV@JD1TIoCz>_z_qr|zh8
zY+0S*Y|!*IC-qP7+C958ai#QuVBdQZH)n-tv-wmw22_5I^5|8udVQE9Ji$EIUf-Kl
zF`})ETqBsC97XKqY)}dFJI5So2sw=~zM>yNTFoB#(`Hn+?D?OK&?Nw?iq`LgtVS~@
zY38U~WH-7%Jve3J_kZro4$#=j_L<9YlVFXAJgL(V=9yToeNA>`ky|e0YOjq1J*$*D
zNtE^5P{8Zffs*T&RETWAaKj1fTmJCn9>zW$)%3<uGNurAq;**ULM0*4#*vueaAlk7
z9BL$BR6Cs|S<QrG`?>7TQstIGMpsvs3$8R^aG;%g<0~4#b%_cPjmS(!9c%_*MMlVl
z-EuQBy{*<4>-fj6uL1*OM2XZKwiu`YviVQsE;w|^EDQw@4JKxw>&6BNHl0m+91`7X
zV$va%dFE!*Q_cC>V}^1!5@%EtCnu4CUYoh$8f+EH$UxXP;~e$O{|U0?Y%qd#H3#AS
zR}XbWr9m{)p&=AwlOUo+N@tcI62XK)x(3EGSmmMInb_#jg19kKk%Cgu3nUs0BN~Ff
zh&*FNABJhmV?ugg8br)Wu}JJi;Ijy|x0;8Id{FW{;*ihjxUPxE#}hUt&O@L&VYdY!
zfCyjp<HT_s@MjS6=Jn+I^R;GM&32?iZs!fhaYX<3*2&6NjPnN9hTncX2r#{rzX9m|
zz|lH8UiQ~@;u;whS3hoXu#t#54{&slR9QsP&axq!0kUq#fzRb3Wiz1(gTf@bZni_<
z65dxZu05hBpsd2keL2y3q<vZw63o05L$R<eJ5a`+3rhQbvsWt20G_F|k{wU(%bWp_
zLs+)|s0<O{U_2k84|-=*nPKA>PaHcYh;tZ>ILc&(bxF#dtRuIT@0U{4hG9>b2Z8l=
z2KNQXG!`4GVeyC7cUwd~G<$&X@s=C8rYzPyrx#i907<5ce{<dV?4TUgv*t&Yr;JrX
zXcO=WL4YAM=d{-BuMp9T^Sp76g9l*QL(R3Gu}^iTbn%$N;hGh>C}UP~%r(yE<@XBK
zUNOEJ|K8SX*MZq#F@fLCoe*+!;LcLd*lU4AWTzi5dMPKV!XrXe`lR%5q@X*~xX$Qe
zmYs{N_17dlsv}Fdi|B~-Egen{+s+;=P@yqjVJ1m*4q!ZK@?%A!rT<a(QRukVH+_ZL
z_`1@<=@`QL{@f4(f$Z=$*??!#Yitx~q3}!#R_7Qr%pz;UEoKB{e*lxnANL-L0>y#{
z;9K1`N={bvz(S#mNPJGyg<wrrDGkfmdT(UzXNQXg@YAp)2x)MPnW1#Hgxofw!ewN(
zCpqvMv36baP)He~?9ae+7CfDX<yDAaVA+Jo<t^jdT1P(wQ;g+?5`;yc!1avPc@jX(
zXN0zaN~a{EuOTcv<Tf4Snz$YVHYQXWuIs|{dP1a=NuRR@RudF$4K@`b9^4KnuIoZ?
zJ;2(Wc>DAtKHcA<BV}(@KH7VO`vIup^XH%Oc>f%VSTr(#jeLq0!$-^s+}wdIt+bGl
z+i{CQSa3YXh0l+7e0sYFQy;*sA0(J9V4FM$GsBZnshYKJxlfC6-c-@L1V<*lB}^)T
zortVBI%^n}<g!g}bA-2(Z<99Bri0GPH+z*%eF=ZWTAjx&y0zg6or4C*N+E~G8|TFl
zrOE!wFF0<u2-VxfAJ%$O=Vv!EuE_%sjLlmvYx`xlAEv5&IBuVAxZmF(!uwz}A$l7E
zmW=W*8iu9AkfdXx?QL;(edEhhX7s4~0BoQzd+Gza2-I6FIlaXismyiAF}c}XDb2hh
zq5-$_j@#{CDX@6{e7~a@QQJL<NHFU)x@DTrcwIUIMLfh>=YlpHf+5Ip6S#A<G{PCH
z$OHloX$vO5h*Tv-u4|09kEb4<WI&Lk#AZD#C->=a8?bS~=U{Co$$WxJjWngR$c@VN
z6FPSj3*10b?)blHA5bsa0FJP?!IA+f5sdfdBuRgM0CT{g5VrUhrr5iZwZZ{@#U1#S
z*H?jo>$-3;)a}EzF_X+0@!@6jh)Z~#fM7Z<k2FB`ofip&ng~6?K$eybbKp4OZFLB1
zs?mNFKsKNWCnCN#CZlc2fQLmO6@v5`<e2d2H!mDOm74It3AqB3q&OWD-fBC;VB|pu
z>pPw4222b|_U8)bL<gy0qsF^Zn8A=9OPv_%$}m#)2$+0cNVJBo2ou<8kY{mRPs}+m
z#seZbk;_6MwlLzYHNa#3mb+A4yhuY=L)V6(IU68yx}uRfh)RKuesrcNSVucogrxe(
zt1tzm_;i26{XEfoh~a7u5_^2p1O2$;{rO<yCjrDNvO<?7(oHP6RHjK_E%sR(-rhc;
zi{h`}f5yeb?yyz0Vur`_17p0wRPekmJRT1`9v_%<1QCpUmezvnp<SX`Ob4V1I$CgJ
z!0C#USrA!3(er%hO3lQ;Cr}YM+>}!>!!v>b6Ty!EmM;;h%DMvd5cK1~arT;U+Eg&p
z+buwlQXT*6j_4n80hQaQ0*{Cr*Sb7n-h{??pXY(2H%ti`%`_Yw$xkm&b9*cLal`%g
z3B9#wel$b=Gb)g9#m!g+lMZqGUjq2OJ?sK1T$|M1j;L}ftFW&DYW-BBG>yIl4RHKd
z<=Q|Ih7$+`_ab;s2}VT(=W(JRCz<(HpS8|h0vj;;#qNo%NACi}<WbDr1x5hy%1MAD
zhEXBzIm|>5m(#oUHSUu-nvR}GQ~<CFzyX##hd#*ee$g#9&R77k`1?x-Tt))sxBxa5
zN+3|P2c)rNTq^~#q%2Y{`x)o0#(iouCkhc+5r#PPVWUt+o9USCbM`g#IRUwfc&`$=
z1{rlwMUw@**4uwwc;G9+zzm_Y%!7qOP6I5&jc(DfqiY_nVDR>q2q!8lE5Bw4!iAA2
zmtmf!Q7><XIR;dEY_clQLx7THoq(i3gt8vlk;l7!m-m>_0c7|xLx;JFJk1s2OygUj
zL1qi`9$=uxDGxvqVZt)h8TN5sA+CMDL<$H1o7b2Pdgv}jkNs$p68ZLCVf17>tAt{8
zk6f5LBPB%6jSYSxk4hqzCgCnSm&ID2XRaAPcTO}Um}|J--*7ulhyc$y+0HjZXMxP;
zWaP&Ux7&fv+UG*P?zn6MD#3mx(s!GHHkMN=qZ=0+h(zS!8qw9RpC`;?=+T(7q7*8M
z^LC;gN6GVG?hR%!0q^mS)*9X)54^vB#&|r?TElUkIPdq^ho|Atj^h@jN9bNk75`IS
zC>dc(rceMiuPh}EPcmdFQ$#%(>{Mj`|8S)-5@Al7(IlfXZ6Y{SBW(g;lqZPPu%-9}
zp~R(H<$x){<)p~Wvkd@(^LEF1dkafTnWF=?m%XR&`vTn0^9<r^@@VFTnN6*7gNQw<
zky+h4YTwgAsTWt@2_O+ug2ml~et`m#I)yoWTV6A&@*$1Z`iKMs>CGk}0MBx3L5fAC
z^FpV#MyG|~B#50+B_G7^M%kw6&iB^ZWaKn%EFCgYg&_#%YIB_fdgQQGxv-RYkIE=U
zb&Z5b;}D=0?tB^cc<$YsXKxn*fk<bpJbb2JqWY?p1Js%z$1_<r!E&#4&ArWStcnTQ
z`5()kl%!Q;;PsHFTozU+G4Q&F9Hs3&?4P#vIWOSrS6p8S1_t{}1axtsAcEx!60GN@
zf`%4gz(dxX3hqaYFMW)-$pBUk^B}Ti^+DA|=ES^IvJ&;%BCGjOZ!F7lU?M{DiagXp
zL%ni!QI0}h0<so<wGGY<3p5<+pvq<}Hn+9LVL1|wj%EahlUxGier$Q_GVYN>Pp>h<
zU%mm6tY5F50VHik6DBX?sv>Ads|k>SQ(4v|Tb{5$MAmu>`f&!KkVtZxMedztG#S8d
z+JOd)MgojSswbmD@W_)NBpandSC#M-94zs2bh7z7V*W(*PGBgIGvJ@kC!W`ZLnE#5
z{{C~Y_dC50a#DgT5i1%11~J_Zp{!LSd}cHtQNP~NrBO)<t})mrVBmUu;Jrr)*)%J{
z5CHu2)1Tp|Pd{Rs;rTC~aKx~o)~Qq^m29H|R(dQ}-pn!<1fL|xW-HMIf6FD<lpucH
zVg8f8JX*PC-F@pFtu^+Wh>}$?LxV?(Nq{l&0u+smt=Cc94)<B-Pe{LZN2JJ74@mG@
z+Hn*h9;`e7-^)+}l0=(^bXBpSx*=Y3;JO~LF{0#JvQhzzJOSkcCP4567((z-9zrfs
z{<grwmpOJVub&P|hTPTg;t<ks?q^>mxC;U3IlDjwlRrz~X=6Z|Rs)^{4Oveg>&0>J
zmZ_Z8cf)74@3oTquzf29o1h>8OLqPQv>MsP1B1_%4pfJ<15Sf(7JVq9b#(B#+`)24
zuQh))i1G7C=dyRg_WO9=4rv|A?xqRrWQ%qSP%QpEO`N;GLAQEM>Yn^BiOdBEa=!?=
zl4nqy?){TJK-u+tqN@PC1IS@pai3q~M0{NbLizoYIcOq7{u2?ZXkeoo2f6@nN5^^e
z@ctz7G{!+U4ze}32QA`;0MO{C-On@f$>y<e8|qvO6#q=}p{KHTl$U5`n6mnt_Z}4q
zS<_E<Hfq$FthE^@t+6$2(^NRV%6sTuEMOo2J>nS08#U6mqoH?txw*Dqpto$eG7&;M
z$7<oKCQvr~SpzSUZ@}<A-<P3x@cE;cLT6v=2A|Q;#oVKoo{F@!94f@+FdlF@!dQ}J
zZ3c908@V#_cTEJ+I*e5$=8=|Ys?~^U+Jp~>Pl>V7qBykS^T#_LbKv9oj?cgOYxpqW
z#tH-%xLs^ljcNlM<U>^Ng{T@3SdIXec>dlZHS>H%DH@nFem)+!-GX0`<8Y<Ny&eP+
znVWk&KhPb;7|!#cvY6Smt^5CM$bR*L=cLR{xqH@YSDw}J68ifVF9O6ylG*AGnqJ+2
zx)Fp6W-8bj<XCZoYBXYkNP*?-)|h7wC>4NgmkvRI)fT#A%u$_Cr&|FAY`R2b?Qf?4
zY)oAC5jt#y461}~kQTr>0rBcfz;hYY@c@BsA2SOYImCB+yw+O+ovc|;gR{T@__IhS
zQCIphMI}Qt3PMoqDxpIpqbxK?1lch*U5hm@bojzyPVzO-S`cN^c?%EwV$^brX61VD
z9|5KtvfFCf#vB4v&*xyCpPQ<PQfh6;O17-ff(N*P#2U|9_qm))?!yGr&;UTLM?BN&
zCF0FFRPJCTD{jWwqh*(q9H94QbSUKN6@=s&Uk#ng+$^BPeXOMs;g7f=t^gfh%!zSb
zcoL)qaESY;V!6MSJ_B7NY{SoXjj*(KL5rCl=puM)!GavqFj#Y+#EV3`Zc$rL`G<?(
zWTgR@nWctoAVdadqHhC`<X~7Lz07+Z=q(0njO-bkP{Y&9&lrCg_+H`a4CBWk%F2Mr
zkW!QEa>X9phG=Wf&K)`PsJI5c%?eP#*&2ShM{^z=4Z~RDOz0<Mzovk=1^Sz8SQO1u
z&S1lycQoerO7K0E2#HaX?KBgKB8e?p;9?B6-sAyDGNH)E&h-QoYwDzXg#96k&+ngc
z^f30%=d;jolAMGg{tWMe?A~jR7`h>6yta414A<iwA3y&F*T)By?TOp*28UoioYTaZ
zwrDKxqnFfs=Ma{-iKY!=j@xnKM$i#0wMQw>@LF4i5bOEoMGrD^%hz==-NAiUt!Fg^
z4<$T)Uj!v5p=_@YOy$gbUMyorg$0ovJvSinjb70}saMFg41rsb1tAet0)Xy0+S;J$
zX>$#toEc29`qH7<s0R7(Xk^x^mOS_~*SIPLVf0u<B)oeCWM=tUruyQGW57~I?nqjc
zt+ls1ph6%m<#F;vvTQLx24jmAHH!P)7A!Gk4m;Tl>nhO=c~McYsdoW^`lyU}I4Xh_
z=bO=z*23vQ5dMNy?=ZBQq$rReLd%2{;TT1>Hc1^ZNo4dg{^eZBdI&`{*xIWAvLRGP
zapI*~Yf>E}iz5(VZd_MzY*Y{~DBYsX;MM_k2a;QcjK<bo@{3l~Amw&X^DU=sIZEmH
z)oXp7*Myh(y4SYc$-Usqx0d9&T=~8|9gj^X6k(p1F*KgXLm?IY`eWBufq`tPn#3ju
zfF@`?jJQunM+|cj7huNNR#nC24xhmyzc{`&j9$XNr~{2s4XHI(GB{PjFb;&U4o((q
z9s%uK!QL8<^N6;$ZtyuUnHzRTOh{zuA|Xc^eOTP@#3F=<YL#qE#{x76Wh0vTEE<Bu
zf-EUXgO%e&TWe8HcXmujv?ume`5?Yh4Tdzv*&AR+J^y(_KX2&AZEN&F0r1Fyr4hGj
zm6sL7RJ=JasD9u)4gwK+smktdt%u^X3)yxQ$nK0C$8n(bg9CjoJRcwU`1p*s^A=Hn
z91T4~;g303b`%60Y08vA5cr0X#KovWMk21qGnySi@%Hv3&VGkWcy<}}DS;2y9Liqd
zw_B7(y<bl}uO~Vb#_m*t00{TnEoA2GxZUnJ+5xCUF8I1gE;3x#6VLO)$KwOf#{)6q
zT!0)nA^;@!$mlh(g|KJzN;d*s9&j%i%=&Z9K_dxTD}wOr4)mYu)!E~r+V48s?=tEk
zNC3RT=pLvFStjd(Itw&~%kU!&w3L@GqATL(j0lVwsWHu{L0cLd)Hkw-WvP@SxU<5|
zRl-W6Q!zeIQXdsFujK?eOrqqhJV1HPBk#=fGLCrtcclo{FMj`8IAHy{)|&uL0t}>K
zSspsJd!TV0SFBM#0)RW{B^$|mw!o8YPVVL(XG+)I*gGU{Fw+p(i$SRGT^8Vqs1WyZ
zjN*Lx2FmuRNfI8_qcY7%!u|3z)!%7!^K8aU17{;BTO2d^c7CPKZ-FdV6w_q~U|3sT
zXB-aoee;rC;OpNB2tg-SxnY{W*A+=Y%IJn3f2Ak!bxnsR$f{(4iSjsZ-mn9BT2Oi-
zSRi;#L%#;>=(yNo7TCR|JeW(!gPgUzh$uz807yg}rlBd2JcYQy2qQ`|M!}FYBLN~p
z51XV5l6>~OVebpGH`NmQ>k6>5kNT`Mgs&kKK$7#gr561e2Gb1t4C)$-TA@9b3|7c6
z*6JDq&ud`JiFQW$V<&1i5AQQW-ier<Vf%1aw1}6E!Cd{5EpfAH58Da%gCruruYt=Z
zn?SLOLbGfN7505TA5lK&4)IX@0C0_Gk@IeEZ?%`l*`tb}HJDF)T+sljt741`*BtnG
ze!!~dI?yzj!_oxjdB?qdf@rJGadTb)0?s?$-ag^%2UHXv<B8{d!nNaiUO11AvoE95
zjs~`Q3R#f&Zz@TytwG%Ic>j#+y4b=rz53WQ26G~cGf86A(5jX>RU}UYFXwZ^1jU4}
zPM@2Xl4+`M^{NrzgC)HHy)4HXgPR5wtTB&JXS8b88m>d^g>3Q`h8HW$3yJpEN5k!Q
zM~1)C2*r!Gy!5j60BD#*7!qD&4)@nT3x;^+vsVpz?s*O%B5tTkpwyeCvI$<=7HL9(
z#_dlmbCIA4Sj5Q{S=x+MA)2+mU_?PgAg5x^=#;708tfpM8H=G|mR(4b^cvkgy{!mn
zme=ye*IjFr!bKWv>Fm7I;wI!lQF{exEvgl!4X8O(XGNMano#SO1Ze7rIR^D3ZCI}D
z7yFwfi*hi{NWBO03u144TT9>4DH+vR6)^c7O{0O3(X%i8kqz$xQXng&YVY%V3VV80
zhDdU6`kL30CCUPziAK8VCFJxsUtbLdT88Oa0us*C*2Kdc7Zu=~5z@=Tx$<_9V~Jz&
zjhih*>s3-iU@#_XXpDo+WZ*E3p(f^5WE`hJAu9<~%4niIXve&wpLS=yC;FY|cuTL1
zjm9t$7FHEljzV>8hKMC`Dpft9L~4`a9YZKK)o5RA5UeU%)c37tXE>51yFua-0<Q);
ze>}OSbwlMjWuB}`X(}`bhWZa2$DQeym=mB0enLiLf5)Uz4wiojXesi*{c}X0^RvZh
z)CeI?jPZoe!HA3jGI7dQv+DrxjBxxQ!{4_&o&m5>=F?;3ax_IbYZk(^f$NIyh+P}b
z;|!0m8*X;tH^2E8_&g2LdhE9Y!v-GD2Z5Mj+Q0;Go+o}de*_@-`Ik3bW)POWDfcjK
zftwdFE`CNfi*g{_#U^iqL#5;W;srp$Npyl)0uAo}923_yF`nNx7>LMTDpRWeG=^o<
zBwa~@s4!;PGz>s_*G`3)-em>I=^eo>$f~`w(F~)5^*!sqk~M-abe3(Uk^dshnn8hO
zl+&1L&AJeJ$!VB(PqLKy7ao6w%FByP#JGR?woJ19G7b0^VMxPLo^I+tl^3>q^a#m)
z^5ADMQnLac@tG^t7hJKxR8sde0QgwlHUgOC{1MJ%5X+kn9lJfC%4#bPuuXY_mP|c(
z==9zLh-B66T2EF+^q6!jCSlCA#djlGe`^>F)UY!_Od#wBq_J|%v+`k<;dkoN7tE9U
zm%0xDctn7t$24`o6*<x=?wK4KX@JkLW9IdlB|lh|8p`|Zw>#Q-R$3}c?9(9)!~5%i
z0PwX<hsMOrHj4<9JISPPvUmXo)G!SJ&{e>+ZV6(bK$J05U7^zT`SIr%ky>~O(WyYB
z1+#tTf&t2gHh!UUv|->@oqt9L+=1Kui28Vv2qUkv@=cRof0h#gUfreQCcPpGp#(rB
zCu?X3avd5DcGz>|tOe9F#q-CBO*xBs+6<*h9Gfv>{)1g{z#CcG84^^)FqBl7iGoyF
zFRvOC^%|%9N#k=Q2%eZD%BJR=7=x)6jgV@jCPwf<+@KQOu_k9BIJ{Dwu*eCIYvA$t
zz=2z=CDS6)lLU03NsrkE#|_Wp#<8$rL>`|%<IxXvKlv<?3b>yqAi#4B)~d%|^cHOA
zv&C?t@qFOp`M~GT0S2Z|ynp@~*T)B1Yj|82-b8Re@A!1TNAyJmkLLsb;-CG$0SB&m
z1(=m+#MGKczjJPS?o;(2&$UMyH>+ux8I4LHYLwmXQQiem!LZ(YZXxGy^-6^cfPgtX
zxb(I%W#Igs`FdYOqnD?hSAmITYwN7E))+B}{Yh^YjQ*r|U&c9J#Y&IOE5k<MlI<t_
z2-$w`<o{?koe^2D-S3(AlVVc;rT*Iujc?D0xUU2Pl271Ehh&o+XWtMh#{khg1vF9s
zAmme!M-!#vQDL)tr8^MVpCr6&L6_Z$p{}HE%<iP&Y5R-k8$6w*^MMTS$7f|_K{(i}
zm2#K>A?Z7v%#4I+81+nLti{{bC`0S)ivZS7wFYS&&fYiT7fFaQc#2S>QY-sDMK1=0
zCD_?}j0i_sGeQlgTcEAdxG(|i!di^&uoPkq0MXAA{k%bY&fp0SUi6$;&8mnXe%19=
zU?4)#yrVo+z6d;~hDN3*oS}IDM~{b{Bn=vGvXdFC2$rgD(+sn3fFoMxUSqCutbn&w
zhS-megJ@pNpmRhUTC`yOYt|GC!9Xb2EP2cjW}^KPF8A`tACD)-HPKr`CyLJ3MA&`}
zyCCMsor6HwXF+L<*X}0&AYQ$QKWhqDct{EW6sG6(4CXbq{Gzil2rC_0i=OMLATv#%
zgsG}ADX{e;8ZViVYcO!JLC|<!usJct#2AKtv{FBlQwNP8$;UIuS^Zc9D=vhaV_r`@
z9-m>D&?C|)C&AWpTrhXMozZ*X(uVhJF-<S;dVJt{y+y|{=8v-qBvtAY<7Gb~u?}ZH
z(Oq%PiO2JSTR(AKPkfM^*f}R%Tf^;k;yj`x@^;*6<nqUOuMjuQbcLqFQ?!y-^+-oK
zwtR2OMHxIs*639^Jqe^Y8d?=mF~o*T-|(fH#vcgJg`Si%rIOJVi}Z}rTCMylK;~Cf
z3AnEywu;g#LUox7vQ|#@BT#Iuh@~c*mmcq|hEO$uFC7ppW4z)Ef<nFKBpBwfwdE;R
zWFVj4w+ewe!<($+m3mY{D}hG5jDNjuy=4&}W&0T_RjTAGtmUK*6EEY+9YWT-Ds_iw
zBX7gjx)Xe8aRD6N9k?!-Yn0~o6WQrin5KkBTljPWGr#|8wBrnmuXAJb;`(K%oX^_`
z0F!%=(K!zeRT3MWUy&ufo}i4CP7IwO+v`VDsf=i~Wgp;MX=p}WrS2)$r?zKv`pP;~
zYMIZOHfWcFdS8O1OiM5Z%@5BGjM;rQJ|dmXUxx<14h$IkBMZH4O~cJdfZ3u&lOcTe
zj^l{7vDXA1V`93adm6`BgdFh+=A$i1EhM6WXtNY045JCIY3K%+1zCu_ivodmsqTL6
zgFOv`v>1drNubU4$yQ}=S^73dL=na_j31gJM?$9Ood+C{3dNPzAX!SnFs~r)L%cd>
z6&OfVrKo$z{8D~i)YZ<IJFZFt4dmpmsJsY3ATAZr(DvHDJd7!gug_>FP42;DT0oeN
zFm`|utO$<4;q>l-s0AO_6HnDJ@XwEE*ILRe$eh<DXwL`EoI1w_MRAZDfl7D`?x+%A
zH!I?yi>x-BjQ(WLi+=Qoz9?Ii3PWZL$N8L*>SBt4feW<b#;J>Ia^~=8AN%8{pD?Z`
z-sdl+g!h_Y%LI0AyE9@Y96ij+lgj(c0)c8+Wpv7cUnE`%0Z-#ve9KqYeGGHRY%8Nh
z5Dmd_AcRc|G`*&!K#&!E=T{GVY+kEW=S`I@jX;C9OzZC4C1|5#lTkH-*@#3&__u(=
z7oMTbiR*flaqHS6%9H!MIeHmSsNDM&CD=dnB^vs2`Svoe|Eg_vDIvgJVVKCa=_)ZT
z)e$<#Y`{&Sdc=X_>O_b>0CmQ?13XNi{K5uhg@HCj#{u)`n%NLe-V3+NXQU1JJ>H_h
znK0*_QHykRc*3ZMMtrxe9Sg%A^i5?GvyuaHsN(rZ7%y+xbAex?1@Vl=_cA@|U|JyL
zGWX9D5MbT!3OISaW+!vk#6vnvHU|9Iq90%E@ro`GXcPpD`*W<uOA3z$a=R>hM$`WI
z^>t|i^L2nW?Ns%Wi9)Ew4-p)lq!BkbMwDkwgK@*ov)3X-ii6$9JR%4YS(Gxk$*_h4
zJ^@t<IgTKg$x-2~@I2V)Vaqj6vY7)EB$pwya-lgI<`<C0&xnh;CgwJi5L3joDtC*4
zKx0l^W8iv@Q0_E@=Icw`5sY}CgmxbdibTVRa9-<S6f^WsfLUH~k{#!q3;~iCTb|a6
z2t;||6JtE1HD?|k0a5r;T-qoF0!$YiVD#qs4wVS|Ut?0~QgKU!Uxff`%!xj88e<4R
zqz)qFe^n!qIEd%{Ev6zKk<ZHsgJCSfQ#^XZ*%V!q+pQQgK%Jxq)H;sa9k<r-7cl(u
zfA-IT*6``aKgZj7i;0pmhH<?=o|w-IO%%ud4RZ|24hp;mrC}m_(>o|<X<`D#!vT&f
zPLC$`(;yIaSaat51{G9xgA1U1W;p5_y$+4%2&Jt{o{$7QW5my4GiGc6fLMUCjJgR>
zO;|((CK|@UBO*|o88_H?#x8KdvxA->cp4dw&c0?!!*4Js6?R1A(pv?9)rBt4!1B<e
zE1Uq)6i369WjE4g@Vy*{0D)^g^2)!jXra`1tQUFx{FQ%@;DhPWdXE=>3q1D_VULg*
ze3jnm?1o;_-pimT7b#OY@v{LRR>bXmrq((#`c=fjF`~bI{@d8roCf`+HFugt_h@Tw
z<Ps5(tTc4!LZkAl7W9%jJV>C$&XWAhD2d+C;0r)VzE6Vx^zthGSx&jv``6A`$#mL`
zbjr%MF)}^J6XS{|Z>2+)F?~S?!O05%>|?@Ytf(1*;7bIGLiTGNiLWay#3BktT2!j(
z&hmok%9*2|4V``O%>*tJ!-30I_pfQrGAu{y4mGXLap6%DrHV$5)~(w4CYdnu+KO{V
zzknm#+|r}W4rWPUoGkNOjX{{3Pp_{8DHb_%yn{xyd8>^wMig`-lw__O;P=lE{A)}+
zt_#|OL^v&dns^q6qq#1xbLPB;_%8=+G<F<}o&d}1sOk)Qw$ozJj+s|IF?^N5M7RqW
zHt`$}JU%{Sj%V!i79mMxlIXc<bOBUt(H5GO8Q0?h*u-^>FxbZ(z0)(^+P3!AqKOX|
zUpz{M-nrk8$2-RJBfQ6^FmE`{8=EB^xZm&5R7h6jMX~xH<N*(`Hu3TJh*XFJxBCrm
zZ*NgPHh;jUpMJu$iBHG9*y(V`a}K<Je8&6x2PYebyfhAfX{|@}L{bhGhdQG`f@n(B
z5EBti9M^S0+U5z;yLHo$O9^Gska<vpYlghP-77=T<q5Z(;^(0o3@jt-F<8uEeJ)=^
zh;|1g4IUsdHo6~iwy%N9uZ_%^-lkI4;>`ghqfw9(v}5K7bT7r6-?rEGm;Im@Q$G0b
zdH=kqN{V&9fAa7Bd%f)sf`EF#dRKbTFK^~sa|~}Ng)C>`N2&ubbVl>ilTIE${+Hp;
zF9Amaq8M}T<)O!KQhtFw1?iRZ7l=gn**VD_C~%+OK<_*!O8Pnjsh>T)bxDVjGjfVl
zy`pTZLXXTQUqCI_5vA`|H48i)(mjm7O2Eq*mVH0d-q?%W?m9Icwd=aL4?rW=B%?44
zGw@yn&Jt#u6Roc@PViji%GI{X?~BFXbbVz9!X=3A9<3<dp-E7oW;Am&iO}G|4sk9l
z*^>4CE4$X207KAbfQR$!=v^a!DD<B+!iv>icI)M=02Gd+Db6EEdM@E<4uHaHT=-y~
z+XB@j1_GG2g=6BGcJg3^lp_SJt%b)oC_Hm*I3$oV=q0KEy(x~S=voCN1UtEVOHksw
zSkgx?Iz#;doRoOWq9~53t_%?ZUI;Xv)o~stJ=Qh<eT)mD1JCC>=Jkk)UG6MVVuK7G
z%<R6WRH8LMf%g&Zoe`gFhMuz$5am{n!Ewf_!4)yKShi^EI%lx^$9Tf70lTgMoXl4y
z@uv4yhhC(|I+z!^(%oxrxp{!0pMU;2x(J>pe*E-9q!>Ip3%1@dA2GT#OC=+R8p2_2
z(0)dX<<<e#kSCy;HYk%pM6wxxrnO{H3q)E(zFi?u*f59^<RXlqjGQ`0Kp-R>64YJe
zTOU-EG79E3Qwb|ngAs1i2pB++0Icb6prQ^XnK@rkMLX}%-Y}mdhL6cbBK_jvz&!XW
zh<%Lxj40Fa=qMN*gc#Gl8{gCg!tVoU1~<dMI|ECF(_qOWC+3@841NB12NL!7rR?dO
z#5(RTLK>m<-RZFwfRQb|M|;}<sB`T3=2Xb~N@T~!7stST-6Yq^*pI{|^@9^&@O@>>
zJ)f$cHH~68Xz}yN`DeS}Y#D5&!;(*xush#Z-4NX>W1WUH)2}aUM!=Mn99b!#v@ffs
zWnDENhu`;->M20DqN!D##?)=TCiLb79nrPddXVZf`hhJ%kl<<EgtMaJD589k2e<!r
zx%y+**OeBUv&1Bsuma_SE+ZAgW^x0BHPe&jaU#M_hEA{DVbjB~&P66Uw84~N6=mHd
z;`8T*9;;DN632iFPZ>SUF-A2Tnqy!NrWp`9+F;w}IAulVw?fEP{>=q7j3a;Ex>8mX
zJcKfF*LCPfF#L2o@xwiu;ao|kgfE6E=qb9f%iSn~Ud(=|m{lFnM58SmcC^eP&85%`
zwJy;fn03iMF|Vi`fO`!MT4|O1sz?pQbivi7G1fJ}(c}4u^ap5&=jw?TLyFkJk@F`e
z2Q59GA9#FxMy6Z>kPz`ZSi3**{{9)Co<CxYHwdKal}$p1re}^nW|Rk@3QzpHE+~LQ
z@1Y>f;2K`%`}^nUkw9{1lOUX)9XO8z5Q#{~aYTiNHvDjZLvM=rkIxVY?#CT-4m@X+
zNJcwm&gqbtcMvt)3Y*0H6?Jkbqh+ubDVP=pWNvck<5$M|{zR`lqm{y|yfy;Zry(^1
zOrm5>Ui?`8uP?b)iAdRTAkOMsUKPE&Lds!v;3j~^?26pan7kQ|N8C^RDRl$O^M1)M
zW3TznXQ9Cq*@$mm@2x-g*zK3r2J8Fl$N3rCOJ6-c`Te!>RlXQeQ0JsT0i;zKWFP_E
zJWFS}c7Dzt`N|Yn80+OW)fTS5`SNXPY1cibPDOIWLSMuZY4Jd$0EHD{yjuAJ5Xtm}
zED+^hYa`TO3Gj0;Rg_Wp5)D-%RS24g(>f8Uy#zlZD5og+1rb3q5(z6zUr~ew6wFZM
z?bsKuh_vX1-UNb!&yYY43yg|H$KS>|67EHo>^<iZ7{sF$d~1t`CtF12Uv_<6WT2#0
z=&|hOGhj|oV*$zh&)pvu{i0(&Kp-9$5=$~aL&feQkQQF0&RFQookiG-jfS?<P(8Hi
zStAi}<+&JgA1iF*h|PrtAu@w#lX=I+sD=ROEy~$?4m8TpxhPyS4`ih@EPVqW;<Gn=
zy5I12zXkE~n%Rh<Q2&l3_eg>G0u(Rhn@E%@D{4Rw#U=g>{zC4a!C}!8pj`lZR0S|H
z(wahDa1Fp^g}q}knN%1s%MGt9K-Rnl%s8czhv>?3L2Xf;z>&{w8lKl9+W#7&t#Gk+
zgDhHOl9_X}Y53*yFF1bwE!x>}zu#dBe3CnW09GH*M}=S;BMFbk12zY4$H5ZHkeTT*
z8{_@`9rL=F23Q@$R0Q5*N?kh+oT9iLCx%U&W8ik)&|8n@SMtEo8mvV<`?%Qb$pAp1
z5-pA=xd6$F08o2u#ztNy=tp=DXucvNc=c$$#cQ9VQPVoPU;BDYtg_TpAK^u%5g>mi
zLF#c-<S3#+^W}WU04{yeQ*gwd_h<*6FBXK$t)d*Ayz<P&aS+ZRP(~<HS^H{~<YmOu
z6h<gpSymj5Qv813)!k6O8cozXMWmoiEM8xh&zF`E^}7AO!VofI#k&5Aw4~l4?_z-&
z0F>1Qwzg+TBbZSHpsJ>XUQtFA6C~vQW?BZ8I^HVEOeMZWCXzoDK%n(A`W=)5LBqYt
z5U>D2p7F+NrC6fms{|-QPM)O;Wp3oCgu|PIvXG83yk-Kdo(NI^E(ns2#z@uPzrx}1
z-1YqlC9)zRo+45NK#n$#GWm&|EF0NT&Xd6F3?0F*(;Z(223$6FI1sIKB7D)ZS;S_@
zQs^DwxfuZEbiH(sqB9f<`DGYB8~H^4e|p_qJdtFa3q5j%UFbz)j#|+0^`mRyyasH{
z>T{6MjLx3=MfBJdwg|}1kNWf`iDXx9C)yGR^IRQkWBH6C;FLQKO99}BdsOxuNEDf#
zLq%RhY6{?aBs-ZR(q5Hoz9$7Y*Cp~|o)8~~3-kF7w;p{Ird4D?r~(IDk0z?G3Je;$
z;8=vhNH0H2#{^(X=7Fgi5f1kX+YQc8FpG3bG=>?aR9b5=5iC}4@DYCg-EYv;@%eV*
z&;R`I;cN#Kf@u@)AMY_p=62yc0%5<uzhgX~5fwNN9K9D<EV=RwLwo7hCSeqgBidSj
zI&Lg~W$n6SBYuVgX|3TrZg3OmxN5Yh98{17T-UYDE?7$55STC;VQZw#oFR3RBe$j%
z-{iH+=rV9$z9{k9=H(?|B5x;Wg*461YrR<6bFhUp;(W8_2WzEc-H+pj^EiWih!wt$
zUL7q$xjX4BfdY7M6iZDdO&@8sl-kl02&3EkQRR2(_Q~!k$4@H{y#s;-AzxmhYEuqZ
zctL15);ks;SP=@A8MDe}HTt2!4Q@IC01dcgjXdR_-<_$1aB9*KS~{rW15!>em%-Uc
zEaY26Sbc*9KzREZ=`C`u1ZS%C0P@l18f&jb0z~9GBv?{q89S#;rqf7rVnp%yIsj@8
zYlFa?Yn3;~)CDpEAgOO_=W{FLnfZ&}wuqlEpt-6Mc8N_?v!We<z~_KHKY-iY3;!l<
zZsS^#zx=DNuPZHdLo|3_^`|MMVh)TXRK;lF-WrbHFpPQJ>M>3A>nlf|IX<&!Q1g`4
z;V*l9r=bykZtpSq>bf{t&~uU=YwlCv5%&~5146KK>~2iEYZZRV@Nxi-#SysX<hV)!
zSZ<eV+=Rn3JP!e1rzqjc@TiLrbcC?CCh*D9Mx>yVY(%dx!B5uJ=On`%z*6B%N5DcF
zIUoWwaaPIH<r+Wj389g$G>`TmC8lS74gOsis>{P@$h3k3Fjf~RG9|z$OluMH&L$*-
zhYtB?=J~<>m5R~`V&D)MJ%R%m6Z7#7h2i<}fyeoQCT}rvb;b+<n-?A}N~C`I{27nO
z2TlnBUy_M>YnX_&5=f*H#zjIW$`kCZ;eNYE86<$?c|zJdj@t=^;QD1?i~-Ro=hS}S
z4FMb<7f5^xBtPfGxF+T>jy_I>w<!V0iee$aUdcsOv3$GN4g>(~_G|qcKr-3WY?R!2
zlya$d9Jd>^AMv&sQMMXotO9i{k0{Pic=5+^!~N4INb6+LFMdEObOnllws#+_Tjkl^
z^YJgwLxUgSL)qe+UE+{@6ncR#x#llEuK<30yX2SbUU^0)@{6}fLY$k`9Y^-ph|o!3
zjekX;^1B&b#V(((mm?jaR#D#MSeaRMmW9x6CV;?Z><d_g9$hV>!S&kZhUQLmbqS=$
zpW&iN2XM7s_Jzh;8~@WX=HgLbT+dFO?TqHc0Aph{R&H~8=PRAE0Ku1Qwq!XX+o4+q
zSi-7l*|44$+r8se24kNl#`7KRgM6l=)q582EhFZC3z2~cZ)`*+S>crF4orsN&G2>}
zxUuYQ*kH4vRr8%u6qiNPP$1Bv86MA3VX8ES!ULo<qGxX~XY-dr<6cNnAQfVtY*IAF
zj0%PlOe$J>q^bf_DZ-w+;~EoVvbVe&`3{VZNai=!A*@Y6jL>zU%Kt=!rBZ8qW$R$4
znYffC90rzVC7RvBT`&39^itvu0+~l@vqE>%CdM`D5CbR@O7Ai4Q5^8Y@O-S4MXq~<
zJXw1ThxQ;vK0h9e%6K((DkYx$5D3PYIM$=X5lrI(#9SB4e}igOlB5;Uq>uMsaGW=&
zHnvQ*;2zk-(W7b5`{#Fje*cV5_fJ4q91_au=meAZdwGvG4c8nnJJ@PFK-1fCqN@_*
z3^XXNk0*KyaPfRR;SRig`h?r<4(~JM28N5399<P~i8Y~&sA$Z4U>Zobo)_+j0dVDI
zvg6sed~Ny-NO@dmD0V#4dlf6Io*`9Q*c#6J8(NQMP(B9RUB_tIOh-fuXnvtFkJ=BM
z=W2kH)kRKY5g5ct$`!0$^egO}<8gaNI4x42`BiCfot|WGoj?FoH>=l4qd^W10?O;I
zLr_L$Z`hlQN&Ut1&PUjg^>S8{f0OTg`6NYEASHwVz>x0d1mvD%7$je>tz6G6>%2H3
z+E`_!!fF`HFh0wi1?w|CNVCSFRAnGGVB{h|Yb;~jxlt&BqAYV+c(OnXqact(Cl)2t
zj3&Vxj^Iy0JIdp4x_|=Z$(f(u7ghLNO%#(OoNEsx1KbNJ<optmRZ6d(dT&oP?N{;?
z*phxeF~<W<Z)@Fh?Y6c3HP_b>2unu4Kr^16E)A7$we~#F!-H@pVo(oOdga+OR4d!(
zjZw-X>#YS_-e*HULL}HkQ8OgWfwnfIz=vn`K&Nqx+S{5*?TV?<tuC7m4c5jcV;!<K
zu#o!8y)2w3MBN!qEL%-QkRHU^Fk#5cYQT}bJhN<{pPtO}-t*g2CZf$H*Hg8|AxCP5
z&3!-;CqFYitLS=1k4)Rj_3{~dYs^tOrV5_dL`L5QU<8Dd;%0blIju5te{%+tKfH&y
zhl|9?>w01`pO@bqB*W-P`2NdpnR*aBA0K#ryvK8;@)}Rrc;e^ZKJk3K<Hx`F`!Jwp
zHT-n{1i2l!JE}kWV2bD(6Q4gmFs=)Fp5gIuZu%G(+!W&)Y#ap~{lxveMO)@^L6A*o
zF6^U>Us>5GW4OQ^f6ntH0(vfAb5j5Gt`{4Z#xlWP8ml@^83{n08FxTi!*P2<KW+$W
z_=Jrs>ep><UKPrVgJrX-BYIRwV>&VRI_ut(U?+8UJkEe5lPLR>#J$=~uQ$MadZN32
z3LwC$B9G^+*0v&$-ZN~sBADXb>)$s5brlxyI`&`M&i&#ku1~3cFx0OZs@~s`(T1FA
zn&oU>x{mpAUA+v#4$SJG;^gK`pf)lqc9IBJa{-(Nu{>H-R3aVD)lg?e0~{H>;dH@N
zrkmm~H;vH?sYt@s;vR_+=880IQRRG1@)uCRqkgk_;$QCD`axl}9bi&60E$3$zbsgW
z{C&uGY|(ZI@>@Okjv!jXS*>V2*TMIL_eZY(dIUmQ3s4W}Yg$s)sGCQb*>em$pTXi6
z)=B0-9I)=x<Beg|HBt=7E`2lxNF+a_)G*ssx<rnoYvaT`0VI0M2ZJ;YU$)u=Z(jv5
z?|hqi4vUh<!7;ppJci?Fc)Oh;P}4aEc`b6yK_HS8ey<~iHonvHtn*p1c|Ia*-u0X;
zz1$)YLa#^JydpUQMcUy>&oH~t+nHto0>m)Jgh+VD(Gx${b>ewFU%ZL*H1gguNRVnp
zBuJyE@;~-n$dDerv&m`#1u$)_Phg}ejDxMTjD=S{=2c~x(qg_vzR$*m=ljnQRZxu-
zfn5;=vQdLr8X|q<<HE<sXIx1ZVoq3F!vVoT0_o&HrH|)@-~8^kXa=%@!)twfd_dhX
zu4ovdN#3l`qF`8nxahPQ1Bs3nWwX!clWQYvi7la=6685Jtc2uRLHzz(PZRi3gs)+a
z*SvYYczL0GP3DI=b0h1|=NQq&4Jhxr%FU?IkMp)M=c9}2FbmL`z)UPg3p1735M!>x
ztM^LzltaXnl4>iXyjh?iE$%6{jgS}mz@nV8#nj9_)!>;}N1ALe4{|B??FHh@1}ww#
z<*FtisPi4}ba=ema2x>ToOyZA8430JT@nw0mCl_kfqt!}p9SzjFE{phh;(xSKBn+y
z>dQ=@1ndZKjV+7rK#qEYqkw?&Ve@qY0J;r7B1oz9K!UeiCqeM4`~s{+^QbbwEU(Hn
z*upKj{VP(-RVGl$Du(=7$+o|~u7DwB#J~6sA$LG!rB4%t1Zhx;%d%}Ke$DlDk%8JA
zcB12h=dhx@vVlZ|J|ZkXXE4R%0IjjXg~Adc(VLF(x`M&BnE>lDGx9JUmxf_;_hN=d
z$UpmRv+LOea==LvD!K}eBgUMn2I*yx_5OIqNL|Q^c-)Qyw=+u3#ufa5Xz>}FnGG|l
zISCD^%~j%OM9woqlia5uEgHRuSd;8A5TP~-b6dQFL({|60u=dtd;ZaY822m<(Zs`J
zY%iNZ3D<qvgj`R+tvuP%H=bczF9#0b!BJu3A|rru61#*BZggOb#z&06@bH<<>Zt(F
z4qDOpWU>RHN~B>{Dg;<;YQr2Ka0jj`Nz4)10YdVSsb&_HgeU=i`*_Fr{0{&02{nbx
z2?)l(xMG}g?+P0+f${y9pCjztqiu0N8q)&NNNc*`{`Lv(j*rJP026;NnrWwj!E{ih
zo@D727Xk!Sp4DT6zUHX@NnVv2zzX4h@qz-iI}m_INL?j5ncA=zn#Q)-&hzf2n^?>w
zf_9uZ-af_0GZIjPbkYFk<T^N}I6A}0o!$39k6s0pR*J_k067qb(S!u-WSj?R2m@5P
z<;~-qbFd+p)4TVQ!vuoqNG!XMW+m4RC}tl1aDM;wBOH=myBS2b3KvE=3e4x(-+)Yd
z>KO^i5$ZL$6Pw(o_qyqlP|;o>DbK$H)od!)NiugR;o!xhkT9310I2BeOF3prgl2p4
zB%n@~lV>U8D|pIvH=ji3Y^LOy(B5Pddk+c9oHnS8XvmilL4;f<aN2ne(>xr!7r?)R
z;eGEjN<ujDQeBZ;xrBtd{(o2=zP!G!1EDVVdl-{Il}*FT$uLG&!O<hk?8a_`wz^vt
zS}qX4k_(Gk@v9m75hz`oR7bZ|j$vmvG)(1T7Df~R+4o_*)8oz&z*&~)SoO2hfsf}C
zACD)lX*gKQ7e<QU<cP8poenfldjyef(2)wVAAm8pv$8)qsc?BdizV$GWt&8tErjLH
zYM1z&mi4ea9BwF}<!a?+R!?k(<1*mMv(Ag^!6w|4p-}n}R7i%xYvo-O3DnlwPwr(W
zx{Xj^G!se3pfg0D3MQK?sUW+zmSPr$U?YIyNGC>sYJCdl^QHHqafr<v&*#Nf#8OcV
zG2s2<1AyQ+KmP{f{TGb$#25q5>xnu*W;mo_cyKHd*ku($M;Q6@yy5ogM;zzLUJ8Pv
zb>Qs{ACLD?__+tnZw85#QL724IyfPk9tanU$}1<pecKlu0AK9pcuFezx)^dVA{)!V
zz^<etbETOh_LUOg8BA{p0ul{7%T}*oVQ;zXJaHDA9^6|C9?j8XLZ;1uJTfb<pLn15
zYk*6%xLurt*EY(P_7Q})%SW_1F*6mG3j>x|u3QSTUT#UkibSmF#4acD`d8lo0S!^u
zY1=+39f|nNmC8+nC-s_)dL~z3Z*!h_J<Dv%!*z!w$oH)^JlE0GnHGzE$!HFd0)Qy{
zjOZ(%Xse2eMtzABO95%jW$W&j0wUXc<Ms$9LzPfwC&ABJ%C)kxo=qf}{c{V|Gv!kN
z*HMw&+6{NkEpW`aRX5c)NHl_jMDsl;FuAsz+mrjbo@0TQUt5;^RbU`;L{g0$Uf?QH
z9v3@ZR52Z8BvO{LvC#Bc3L<ZPdUTBw1)42GpxjWa8PO~)<k!z>c*+Ig)U&g9cy<Ex
z-8d#1n|O{19Rp}el3+$IqVIo1Fk(Zqg)9KHBPOJ&Y32V)7}6uBpieVSXXN4YG+eBU
z&WJ_h;1&KpYi_sWUiq^SV%a>Xm6l0Cufb8a>ZPlchKE<ws*0j@Mh@IDCqdk}sJyqZ
z7b6u}si7*+8Tr;}UQJk(Wsl?=u4m^?$NjB#%La$E^|YC_oHVAZV_h!wiWE#%FoZA^
zW0E}-g5MIP)CZoAcU(uud=A+ABM6Vz6VK-d&f}<v)${oP6gbZl$8q5C_`t{GqZ-uY
z^JqVC9yh$*-r$JPvrNOZ5$l7H_v;$0R)`JCE@N)Oazh8&+S4>NRo(KU<iFi30H&0j
zrXk>Ok-#yPwn~8V>9X3=^33EpDEmx9lt8of8&o@aV!?~a9wPY(rFN-+M1Zax$8knP
zXFpe1dheyA&n?2qses8rNU)zeG5=Fpq~8QGW)@MQL^1jKDGRTvmiLmc{o;4|_v(DO
z{mj=NNh;1CEM`8+$*N~sABqgtr>y4Zyg)k1-Yc2DT%qg9^{?FhW*;I-^TqQ@3Jwi=
zlX6TOwvfFqW0)ybk+!8f*8<4)zb}|+!&R?w(EWa>ne)CjppZ78MkMD4<`)%U5mqB)
z?}2DC7fzO<CnvRJY?BG092#xRSfH~&z$UYQE0KX+My5q04U|F3waCSgdfi}%QwYJT
zoW>Wz+*yw?G8kwx*y^GJY4J-H!FecdXOzHQPa=I0j2TL{bw%qbBCG>-ukyw0FEHUT
zMe%t>si7*4GYEAl4B~0nLy<=bUc==y*fUUdb|eFz$0kecEJk%!cw8dAknVN@8$ptu
z`2IX7;P3hQt4IANa9BMUkRU+Vv<Fzth>&|XAyXi_!gssG<lOK@?o0rp)#pGckU2X$
zM^_i~8+iFMy@wd~GcYY`!n0R-fRM4)F#(SGYUYCF@kK_>pi~~yW`R`|DUV|^L^*6i
zuP1)@+kb{2o(=r`o4<~5^SJPQe#Z0Y(IL?dkLMG^44>|AxZUm;b3{90K@cLFLVH8M
z-Esf)3Fi^f1$4pm3y>h8emZZJ{#l(S_i(1b*;fFD)p?EbX9P^YX9>QsYpGyg|E{p8
zmp&Ds+Pw-H*|DF#>)SF&>sn#$1azV?n%V}_PFD=zsuU$880<6@0^n#3xAV3*AgPx;
zzccTCDR~!(*%un8ca9O@V0S~y5n^XJjq|iXb%2R1Yp&Cp5yx#6YfTf6Y}U*-CH-RX
zXdr%HiuU&|{k059rV5sE44W$35_o}k6~Wv<On$|3Mo@-p-Q;?@+*Mx>y+3h$m!QBv
z$V>pR^v2`6lC!ZdjI9rryyLt_S-0W{Cju;&E+e~m48MoL_-vB;kSX1WLak^;d^{^D
zBplB<?L|k%@VwNz+cvqBW0&h_GYIaHGV8&$*zRnDgMak;%67z7Gr$2y_GL~)B8X7z
z$(%GqaE)2Rau$;3VfR?59lBO@#SDH8s5cyEE8!B>+n;9?sEsSy+iGWrz8)=IkKR|v
zBn=CVqc%l5qK178k`W;_zBa?o>3|4Hi|$^mUu}Zn9JHj{;`8u~Iq`gkXWcY<8k{*Z
z!R<xP)3DYC`!c9G;3abK%RmOhpSfsRLYl-%2MAIG0_Mz|;ZSp^O+-b-I(4adjlBwz
zK|kljDF=>QNAI0H3RaYG!<52b4TT86cH$8bDWB`QsO*<=G<pwt*p;BevAhr7;6mRJ
z9voCOsvjIy^o#{?crloL4E*w&zvf8lC;s}M|5q3`iP8tm2QUTmy72M%z!2cie*80>
z=YiJ#9FNBbe)rdZjp31ZKJRb%@uxq-4?p|>L2wjo8oMu!SSL0rIgSIF5ARLt-pu{A
zTUsjK7l%TA?Rox_y_^CW>LM)-|3`qF4WFPP%fN(bw5oJdi9R9-MVC-<DP>mBNJGr#
zxEOvN@OiP>I8dpuyf*=k)^WStq6wCJ^e>>{DuCu2DCO-t(lN@<Bp~6r4q~G&@GSX!
zTKuZtyIw2-kP(j^P-RMS8L{*V<@b72B@B33`ud7w(`+x`mD%6j^ITp$UUZd6;E*8m
z30&8>-z0J8p8NI;MTRSo#XP^H(M}znM}d2y`!Z2?r0cT1FYbGJ;khph-tB-*AP82!
zZ;nW<dLh==1<voY+}n!&D8SUm9!;<Bomzl2?;~X<rn$x_?=tHZrOmb9I0y;iUT*Ez
zD~_)Q1IXHSgmFn<HrJ`E;_T7w4YCFcMF`3&L=6>4YenREQG7c}%g>JfpHLpG#yZnH
zT)j1%M@PRMtlN%Gao051m1MkCH<|_lng(a(saK!!Slp}~t+G<4Kn$7dk9=_CbQog4
zg1O&?RSvCj0-x7L*5S(MP~mq-NuF>ZMU+=YPJc2Qq44mGA&9oOD+QCOl_>8Ff+ZML
zh|;R%^^_;364MYJHGBISF($K}RFp@Lxe3hkuM(LBLd=smiSp2^T{1!r=DZmCXNrPc
z4>0^ii9@wo?v3*&S}Cm6jVu6HW1}lo!6ZRq9m-heF$RA5?Qik;_`rCEq0k<c3Nce6
z#yEex|AN^gN~DPOIgc~?W;oD#!})f{r`soV#O#6R^~AWI)oe=%W|VUej(*_daYagF
zLjbMzI}ZFLNUnhFuZ{ElDPNmLb%150sT5_marNSD1?tB+_{j;xb8t->^rlg+=N>XT
zEZdRCu$$%Grw5v#JN8FKqrB7|m@zhY8N&2Umm%B#41-M$0>3M|fxeEPP9V^<Rrk2O
zpVzy?_PH;2y?!nP8vJ(}z~w`}zBAXq7F6<Y3mh%S1_g%FNwG}bI=MjgOo+;j|LH%(
zPk;Gmz;yWY!Y}{H{}Ye@;^(N2$?B@yVL|OMEnWPadillRN^%HzPbAq?(<DKqBxPIC
zHR_o6{nH8gWuQTj-73@Oti{iJVuL7Kdz0%E(SeL?^Ie{$shd-yLNe>N&(3N(0RUt4
z)9Uph8xkZ>HlB})(=VlfoqcUqbZgz|O6`7u3;)LJs|OWH-|*D#u_4O)&fY^2wpQg7
zt%W!Lydvj3P2iSocu@-nd<klN+tww;sYcS<RBh<3<LJFWf|HEs<`|C)k1Ns%g7miA
zY8(PoS_6=niy&Sd%F;8hzSF#?X@=>BE>X@{-YNnR%o)$VLRtwt_&h?-#!(<ddh`Ow
zaFaM<9)jkSy!g(h3hi1gge!Db#JV`D0U{z*fZ-*;I}qs_gMW_m2oN=A<@{r8bF^&k
zY$D|lG=KnTA?8LNhae5LRmo8A8kHL@+y~F7Q0}O#P;G+1Y_;4pB0=Zvj3`2;FQ^F2
z0T*A5opjAnh{<3(vR9Z7j+Ne`Dhh~9mUKR%y|E}B*9U%ld#hdubH?2F^JrZ2*vl~n
z9?uV$8E)*Co(-(}{S78sof=b?tJKX65VnN%Bs|KJ_mr1!_)3090D)J-lx1OYHfUfs
zeIQio0z{dZ`|4%k$=C<9b@bx~2QbG4_baA$l5^2i;M5Ph(;;9fAUL;TL?ar%t|)yL
zsR3ZVspORVWy<;UFSp9q?-y@irTH=)A$eb$`~DiaqwLuFX^;$OEtu!pxdYo|*ToOm
zjb81}_P@Cgc$L3&9Ar0)z85i|{g3`l-2TIVj5nG1=l|?q;J2SI=>Pg3;rJ)N!2aX^
zJMh_&k&J?RFJNQ?R$nY%u1!>xEV6Cr<?XquY}M;Q1c2N-^gsN^@PFse@&14Mr})kP
z{+~d9^MRlK>?iai9LN2k6I6f>Pp-ii5O)2Kr*8EHTJp9iv-y0=CT6Rg5Nzi!6(x?f
zSrH>gc5YpA1O&iJE=_>lR!bp#oyqU7=^whjdW7-I&4`fr5)EvQqqBd09GpQk8O9nI
zQ=%+r8X%KtoLOCJYFGp%I-AWv!tlDq*v;utgP%Y|wY`>=hz2>4HRI36g~u}pTv?BK
z9KA~5=DtxaLb?%YR6N=INOa|(+^m|-)IF#1DnSU~gAMub*ivZdeGD4@wgC=MzHAA6
z%ZLS)m?{TRWQ_o-WG`-#-ib=2PYBLRPvk)nt*FInu(A$jL|(q^fVN~Hk5BGXROn?o
z!>Ww0WeSJ)EGla@cjhHGV7XfZ0eK*~20)Ol*iS(Xa8<OUp&t=BsP@l!pB|kgHOn#0
zUe>C<Qr0l1%M3sN^4EAf4gijikI%4q#W-qF__*-?{0y9Td^{g`jtigOe!vf(euV!5
zpm%)w^do-y^k;aU7y9j=L-0!+bb_`>Z%FYv#>_q(Z8MPHHV^=?Y-Q<aEAW^5DH&(x
zYg0MKIN2~G4rh*jeP~4Vvb&sCXE?KLXtRR7Me0(mMpIFj9XbO;SQTOrBe#d$<`#iB
zB9F=bf5|<6u~)0icPaWRKdwjE+;v|Db1hw-+cfN8=y;vM`1`)bOd@0y?PVi2q~$<G
zsUW2vc#8i31f@&EGgY5<@96)Nzl6>U&-dTq`FP;BpFf}wod5QL<A3>Yq5W_EGwA2&
z36UK)zr6Aj_s>qF+$0A@g>%{4=A>1$aIH-Wn7noL|LEU^{d<2OJ_a8D!M~2*{hNyR
zpZpil|I1(Fr$76tN{bg*iT`my&75#L9^1yPHLIWPk=%3+QTr|XGx1$jvg%78LONfs
z<@7F~lg?k98#4~Pt0>nO-}URTz}Gh&+GTwgFp-#3-+OQc%t;<Zus<6Va#3*2U<2i$
zf{T^4lO>TN421*6ZhX?i(@oZWauJ-y*F3!|kTjU%fWzyPhVcB`s(6U7UU<<*4-Zrj
zX^`XW5Gc;u4M*?MUw~zj@bLU2Jic;)?t?Y0&XBlc8fsXRv=~O?+)?rPJZ(pzP;->$
z3W@G(s8YxD_FDn3<!Q#>tJF@I=1|4Jt5Bv;X?fBPc$8nJ{4@fA=qwh%=pxB3bDcto
z^8_kH??%)chB=}G{cLOIfM;4p;bQB)B&Ews0jFv*f{ho2s_`Ai(Q6pd<MDusGlF3-
z=J8TFd%80YR((ViD3ncrkLP1xJl^rReukUlmtTGZvk7T&|FDVoUw(`Ec;Fh*qWII>
zCmgNge*1*;yx~00%A@uky)Y6u2~$-fV5BNS)LGS%r&!jae6!ac*-Otp!rZAA<1A#|
zx_dOGiG>Q&)+9=R-Ch~wj@;8#k~_Gco{mKUfshrbBIAw@;rs~ZZs&k{j_RR7H98xV
zA?PN5z>6MBP@A%rWU_j%0DEXZ!SfDI9m{LxlNO4%YzCC3D<c5q3@z{!Z~OgFh2+?<
zP1hz+NZnB1i8Lz98GJ<vj`r{WH=+Of8}#uruEzr^z}pc`qHIol{I^aV|BL@NZvXLr
z4*xtW%CMeoUrJ|a-3u^>4kHj~Yl5rlYSEr{sODIR07parkN!R2zyELG`T1wej|V<J
z9`HXuG5+WO2*3TS|0_Oy9ys)bf7#|!!WfrDbwC-xbYH~I#diRwK`OdOQ@1hg<=JYP
zk)N~UKsIig4MQe}HhJmu{d1J|HmC9$Q(F~LMq0%H4Zr64%Bh7Yo#fj?rBUh33k!|)
zu*n`u6kuc-g+r`r&f>rj?d8X2ZYmbdIQ(XYAqBoDOIf#P>}8;cI(!4WW3g+WOV+?=
zDkBho*N6&4Zo1K9?zo>fs4DK~ExcKGWSrCC6_#H)q_~m6vcfSHkBeoD?4(x)5P=G3
z9kO_q3`Ax`LTU7~DbJz7k;u`z?oO9Hkzp9u^3J=4auu?DW7YXAOlxF5a8s^Pb+H4C
zX!<n9#b@rN6thAk0TD#PqsubQCTNGE_lDM2%EUZs?V~i&s{T0n2YEna*wKJ^a7Q?0
zv{tJ3hVzKYaBdUIAxa{s{uZi=egs(19_5e_RAeIgBuPZ<77;BQkKjJoRfp5^vq#L`
z036y5MjaR`1n|6`u(xQCGaipxBSY-7QijT46o-XMz{wB!wy{9)RkE|%i$@WFzzN)W
z3d$#4URo@GI{UJ*WDg>2oW)kS>}isXY~nmh@nXQw7g>XG3|ogmv+9a!^?s(cB*MVG
z_X4T1gB;nr$8AvnM#HjN=!I3!JA$t&h<xt+jP%rDiwKus*`44r?}L#7S^u+bLMqCL
z%I#M<2L$jLwW-&Mlq~M-#9W8=@BV901&#(h5oh^!J8?Ttj4|=c=XX5+@{Z$w_1{7N
z-~MN?U#|MbS73pXf935<Us2G;ZlB2&TTR}=a}E%#5JbNe{qOxfw15A<foooPJf8U7
zZ-2&f4D_Spk^%n@{sHWN_g@0Xu|Y!=V56ejZ!G?jI-fF}-7lk@I&{?0C$6!9QY6r!
z{m9SCa}ET#SqatX{NZKKmqQm$)*=xGSji5!s5Z@JzYYz2B^dB(gdtc83-Cg0)m_e`
z6)_~pc!tbTZwHJYn`FMu(5|^-%G^9_R81z}1q8y#Pb%j^Cnd?z$9`_*ZjPpk``Znl
z-tKT!HYsxS<3Q`J^5Q_2h%_9n$5>o_jS9f&<(HQcn=}}W$POk!#Jkw-Fg%0|+iR5!
zPDVSdkZsd2p2vBx<6T4{&ej)?fN28vHNzlsQt|Ui<yb<_$_M8jsb@&vp&Sy@*~Kcp
zw@0~V+ghe_@7<GJw@7}NUg1z71w6|IotnLV4}uYAD9&F_UA$*gkf{{+E{{$!YU_f2
z1c)2iNNR?+cPPP0i&TODurcs#igC>^z>HSIq}I|4iF>TPa@Gg)pgM%*k^|5AfF3t|
zJl}CWKJa*c;1I|4d}2Hw08pNcnPEa=q-iC^gG4&%6-Smuel@!Plf9_$UTWJTq=ak=
zvj9;39;I|sO(&Z=jiexy;!F?8sT1lWgurn{g@Div{m^2K3)p*r`)Y8sw6VUzxX~e#
zo~JHD7HwknMK<H!+5!YMHg_q1`a9bi<p(aO9lWpTO)@K@0OURFpa2E1_Wb*6aDaq5
z*CK3GDgs+6V>vmiG<@HiG~})d`O80pxZ_Yne;Evxb_SQ{JWq&DJpU5F|KRVT|EK>Y
zoT2btbuR}qFh`QNi;uLO)m<NDbW%K{hzj}PME{@tJD4iCetyR<KfmLb&mTC?6XzK{
zMeN6piG5Jo2A&dR@RssB_Gn42rU(|`WYnOKj@~ZJ;k92$p-f;Eso0SF>0z(NV9M5W
z0%N{^-yzDEMEmnOW;ryixG2=o?F+Q^H(Xy02Jj+cTr&u|hSE&I53hK#zrIy1WkwNR
z&U=P7_~5JA4~aq&#<y!k1;#WOu4p1M=kksv8(YwOW7okLFcU!Q$4__s#ZN!se!pRw
z;e!XK_l|aS<ZSObM~t1F6Gv<GEE^ZQA@iB7w<6UEA$0;Jl^~iOElZ(TjbSthmZ82t
z($k{Ns?$T(-f_EioM$hxV`DhHw+7(|*YIwol+kOcQijYZ+uORQP#2?`Em(VNm}B5L
z4xDEo=A3AlfCGrn3(kbHnxo;#?qU!`ddS(%)z|l>@+qXtR888$L0(A?MiHdq&k)$E
z<#j|wMG`EvX^cwddJbd7i+U-Pri#%7M{juQC>}$NgXSu!G>9}rr~2TmOm?9vFV5rV
zPQdrO-~3m&Zg)IB-l0D6c>loXuJ9!JNbU&%V-pNz%V^dmhnI_L%A^_!_glX9?DOvF
zXFYh}s#g!F3~j#NSE;SQVM>y!CpaMJ{nBX3;uB?#y^*o)63SR4&S;XYiY%5!U^7-`
z6Y~-@>B~T&k&h&+g)Y{2x7Gq!BQ@xH#!8`l88M-EpJykcCyP@M<)^#nzIl$jf`wpB
zaLbR6pNZ%1q-3+!4NE^4fX{nMO;W%<jaIbkc7kTr=&^F%PPA^gu7S`0!@t1WKly7M
z|F4S%yv#q+*u}c%xqJoc<*{Sg>qXwqh_j&okN*<#kN-YCe)$FCy70>{pOL*%ax^{u
z?VsR(aYCN^@)R7Zg($;MXLwN__`jNs)><uL9|QVm`2TYE@4uEMS$-e(SrM`KIrrYo
z%Bt$F?&_ZIo~K4LqGmKAIfN;hB1K5DCF=!96d=HWVGFh-_zC#m&`<u!4@vL~hGfGA
z4A>S439@Jdq$HX&$q~un3^{Mk^gOETm3i+udq>3b4{NQ6eX_b|Dg_KM<<qFn%zMu{
zd+&%9Ykk-2yHX04*>Iv_-4ZcMM^RGux)^tIUA2za(e4H=2|(7G4TK8|9b>XZAgA9C
zFfi_9Vs*tl`z068FT1krHukqVJ5mhwiOmaxZ|DZV7Yh@r87Q3jpWv6Tt)bP(8CvV)
z)afnLU7^NsqUHmVs-vco63bew5C!XlEpF&7|8DJ&R-xI=o2wOF7N@$!+kxUK86m1U
zB@HJVG*EM17Vb_9tJgZj;Sb(}K{U~I0;uZIifo2I!ah-q7Ymt+DbbV?+1G7ECtCPA
z23X2soC2vNWQ$PJJmh8ZqRQlT2woGAb0&L!Dy3pSmmp>e4t7|V!oDnY2VUadwKkjI
z&>|gEjLUz7$_C2yDVKgwf&w*=`cO*oKZ*RUF}F>n8USM+hN0LTyo9|;%qm_#k=}jK
zsk?iC-M~rPO(wK=RPyNR5HCdRBh0tak|5fAXo&yJCpYF~btZdr(4!VTKLX}R-vmV|
z;4Qdc;8*OA8>cYvmOJ;zQ2<0e>?5;hK9pZ3BT3N$y&7neoGb#_4H)|+CjrsnJF<h`
zU`QR`Ez$EX!}uD=_-uWE_C{+$k}@2bFj1qr-88-fk37!G1QxfT$DbCCV5W0emjn4!
zfXagd5)T=zbluO%tVWp6FdXn8D3Sy4dCly|;U|VOW?eyiSR_H~!?8<}xa^JID<vj4
ztAZ#kCqDh7@A8%Z$8RF3(7O*X8_N+?^l79=lUS#awNR(gLLGHLNa(+WCg9yW%Afd_
zIir^gFF$=n?VZ!<<UqxuDw%9`K;al2o;lK`u~_PlJoOJlE`k^?xph~ynyyk3+csr)
zMtVXZt-ac2u!MH=$U%&uKkkU*w&9rj<^&&1Yoi%)$l<Xo;dgtzF&Gf9lr)1T9?B38
z@%t@ztK8meqtz$~8kXKe>tbX+sv~p_!D#1jU~q+pw_qXJ!Gqmr5r$0>Y%~G&KI2v!
z`?i}wccj_9M}BqXork9(9x|T=8P$j7h#xNT&5EjJqagK#&oNo652F-$_jR#3>Y8oC
z)hUz5d5<A9s>3q`AtvI*X$)S2?$oscKFu!(O(0mQM53DjPN#)+EkI{k3w5g_KP@Tw
zP^dJncdVL@Tzf=J5;E8ZP01PUGYZR6XnQfXzBni{!;V=p2@8Qqc2CZWK#Pwg?Al4q
z!p5z2_UeT+%x(!is3Y^#W=BHYv!{c-Ei%Ss8<0FIWaD<2>mDAa_HGIs@z2>vX;VZT
zLfSg^%{cQDT_<;b5$H8iRnB|hAjU9bc!kr;8xzOeW6+LK7JtECH$`hOjFcy>Zaw;g
zXDgwFF)v>IjR_^ESz)bW0Kbo#o@7+j(aRWMB2s&5N_U<)zf77Qmpg(J>~ru&9aTm#
zZ#u}OIZ26>3pp<y{R$8?>zrO&ptwmKnYtYc*W>jM9SZthTQ;-vCObFZt2ZCN@qy>b
zI(YtW?4y%OU;Dg};4AmY^9AMVYm>;iP%k^J+c@y#j+bJN*Zw0fxc=dthrj-0gQl<$
zxof^d1a<;}!73=pz3Si?I2us1FXhktIJ|qu^`lSd+l7x`y`s7E*dn?j!s)ak-+YGt
z+yyC%AH}e3g7oS14!{+Fz!xD}pu<7IBa|f<bCQ$snW(SjAZ-$_s~XNopIarubTs{1
z4yJ&yX1$v}Yt9L?)yCqWY@`ali>Q6K*BiDL1_GL=c7*s-6oi@a-kgQaItQBz)xpE}
z{|wcO=k+X)x0V7d>~-hqx?2udp><ORX()(5!7_|OYmCW;MvfESl`T$h%__}3T%5B|
z9OXqtuWG05&LG%3$x*rXMn_C&wYxVS-hkzocQ6knK#GEzwjeLra+Iz5!PHSB5iO4?
z*@5U?t&cK^vno0pTWjpCvaq;7$cVHgY#jr9iD2Ck>q0IIx%bh5kex>$A_8S8!&@@p
z?_)hi!KEwO2n%LEsH$RfAbf9AOyrYe*mWMD6umUbh>|o-pxAq(IcU(P^x6k8OSC>V
zjc8LJBZmW&SyxMPRGvv)$tBr9k2F&#k?xSg!Hd6JB*kW0nB&#wV4Qe=n#XM9!qwKT
zRvW(@N;Q0roD2vpOQ9;Ht&Z;suovF5c0H^cpHcc>wAZnKZ^TIlpuc70oCfijl9W+^
z7@lyDjU|`)X+zcNfHQlmNZm)LNO-NNM;}Mm&%91}cfn0CjHsIT84P<Ovr8eoeghc}
z>Qb2zID2qLIx!y6*CX7W<Ie|J^opo36k|PxWK95aT*m+efv7$!LKE=7!2gHO-&+3L
z>&FT~djA<bKf&X5)Imi}M;k?tjdfMlrLZglDYN~N_c;ITFOj!w5rT+{ffK{!n@$WH
z0OfyAHg~|*>wp|H<bU!9NdMq_Tx;d&<tMOJDv2dWv2~&=lmw@BA^p(X)W7ko<YgV4
z6#??c#H9d$@!W^SIA&nxIrX%L2+4~JW33}C6cG)3-b`}ZVwC`uGIdjI&`6^RniNT1
zY~36CzOikab@VCpWZ!qJq50ijZ#K2iS7hWp4nY-FE7%GnVql<oj#Y=j8pz6vx+HaC
z=uj?^4=)ztNKWiiXVFd<B^U9w3!!m1<k9XU7Co>+KJ|^aeBQdQqZ&B7(Fj7Ia09Gp
z#T0TBHKOX%%M4J^j9oCRF#(oj(UB$%#V7cMN7*`yDoe_cgs$#=_fGSoqGerJ)?y<?
zuloQBxnyrqv-$w*vba69!FeC4N>%k<X*CXkR2*Gruc*%yr^mJHF04y&Vios{!!RKO
zJmj3MQa;TAO$ksD+CfQ)%IJtl*}%cJ#r*gn7?e?4Fr&gIKXBko=t^^zuJF=e9y%Qn
zo``OMs^x>rl2NrT7ZsbNxfCN9na0VBLOt?R+=E$G8*o#KcgP%6t;(!qK6(k00A*52
zAN#zMr=v>AEDFow$}IWbB?ogbA{n8}KR3MpgZgsgJiT0d4R$!<_=Ue^pE(+8^Sbb`
zE~7O#_}t@-89|m4DjBKOrWUq}1|T?u)5DnvXlJ^lAlZ~`3FEFhY8nD0g#O6MB4(+S
z0X8GV9L~>tz<8F!8FuGtKn8#2S$goAm5#OJuvSd3T@LSv)=j_Y03Lp!gI67IISAa)
z@#+Px^Sh?ai?p5BakTDEka^d(=H^7ibXHrFjJ|W`@kj4D|K(3f>*C+h$3ENfn)LzE
zeFN&(@FH!Z_)(GXdyD)}e2ZP-(@#ESzg~EHGJU$P3o51yTHS5GbE5p{m+0qTBdM}`
z*V%Y>kKl#vk8}RI10inHa&iaHYnE`CGT=E2&xh9V&qp*^vFWlFFm(NKlogL6sP^$y
zNfK2%+rF`G7p~bzjdHSU&v`b#`|Aw{6{&kHb8PJJ2JOnI!qzNY5qVz$^ZW+@m^j(I
zPH|G2WCarmNQsQ)@srfS^&PL2U6Etoo0d#omdTvAGkb|WTwh9IT^A?Wv4Wmsul{V?
z@mugvm8Athn}M5s->tCA1vZN@W0joPcSfO^sB<mUoz@nbu7hn&aATAc4M@=~;6s$M
z1A)8qnftpttJL?RimqxQG&lS@LPvJk_kCyEtJPMx@on8GnZ=F!X<1N|Wmzbxm=|+k
zZ#=d?O&)czJfs7$NC8ZOp=q%cMjNOCY?Uz%H9}^6=KBJ~0E4V3WXz&I$4sRYN0m%#
zd72hS+oAAt&S){9&;{!l$=UWcXC&2OXGejkrx*;3&cedvw9eKRJKMl=$h2DiamZF`
zW8LE3)DW?CY-Dv#4(9MDaK%}NvYVjiyBmZr%*(^OlK{vaiO{lo$ZQM4gSIi)It{6+
zpGOC98Xl?}Y&SORdP0L=qh)H{tmV`dx0e9f%nCP<-3@&yd9<>n=onC|6NtfDGc`KH
z*D301WXiJX?f?iA*Ui+2|5Rk0BYX3(&O`1eIS@DRiBwaIhzSP(aP#&)Y+QJ@$Kqnk
z)(*4E!Iec0dv?5h$65fjcb|76)aoOjMM$0+ijJiylnrlt|D*50U;Pf}mRtug`sG_!
ztVIwfS!}jnS!t;a;~cHLoD%E5^uz4hxxBpa_{pcV*7@l1LS9ef?z_vA;!{yyoO!I3
z2kBh4=0^MgwfroO+6344BNrba!RgQ?XFzQLny1Tqb?%uPr)GNnB39>Sw!`$FMX)kb
zVr)I*UYZkXO|T$r`_9wjqebRRv6>~ZZiCPBdgGb_YjI&5pqvtGnSJmI)ZKEuVH~2(
z#aheExR7KVh{&^u2n`E)h&F=o;pi>oupR0Gn*=FF$fMbeYf9;s#~=k5Fy(8qAmFTJ
z5$<Yn=+%n(RHFFC1VTH5mmd!iTCs~o^l1&PmFCpPC60pP)$tXTYzEQqt?Q^V@ZNKK
z<(ybf3+MBhby<-l>?xB|M_Q+(G7x3ttJOhPt<`1<bZ^=7^vS;MMl~$Kb6}t%<?J-h
zQMzfMPLH0c{};zwQ!I-PE{CY)2-FrF+ltPxpSN0B8pXORM5n3=uVUZp2$c^A?8?u$
z{~gGcY|lA50sIwcoG_|>KTC0jKd+BSOoJccf3<px97RjX>6n+|a^k4p4Y_LM`E_(A
zL^Q?rqm8tYf{`TUS<B!UVcY{fq6$;}MuFlN^`*#77&xMgXdB+@@H&ZqSSZ-$N(`%H
za^Fo_O$F!xlMy8|V4cVJiZ0%_%>WKyf8+pn4yNOL#9qdGI}AV*PEc^pM_9K9IEiqV
z7l)djbI*HN<ngo~Qn$Gy1M*5lo4n(~56O%cdjYcgsO~Nf9@N<t9n?#kfN5MY(z>RP
zgZ#=2)BHVzm(L(l>t57ogU_r@H(#zi6Qx!)?hv)e%Dy|RiCFE;UK^VI-1JU=vBFQj
z&FQavN=L?o&eRf;l`fqm&22@r&dogf05~Z5C%%vEMM591TpzFWy7TGdl}qc~d8K}|
z^@b7Mt1;DH-d;F;P`NsFz16U|L~7K%`GSNq=e8@1yCfTsSE3EQ?(ExcwR7$SeOH)v
zaE4lI>~*JAyZ&ie2=116-rl|2>HvTdp*Q8}>0-n8$}*ghqm%Z#z1}cQ5SUTMLgnm(
ziCp=OO6Vw<a-$Ii7K~zxbO`xcKXq+Nim<;x&PrY^*WOyEWMR!wsAy9ILvg4p{|KIR
zobfNvvM*+Bji+rxlX&I4eMjxh!`KQPF8faHO7UVb^RB&c5j?av6_+7$8(&)Y-Fj||
za4N;R7~I=1p_<-Nv=r9X=%RMry-`%G^Px<>eLZFL?|;GhhrYza^XG>8e*F_J|M{<t
zsfvW!RGkJ+ox!nIwtXi{wmgzYbfP%$Mj;=H>hSNo&08Nm1vg$^c-dPY`6}CEy{S*I
zbVC@(!+mlX9zrN4=OIX;d>O?^r4$xbv>wPuhP%M1{!ht>Wg=>Tf7ep1OJd2Hl6;+g
z1zT5^HOQ$Pl+K9dkm-2#QN&mx^lU|C0<H{nJcmWctjj~XlA=o`$!FGi_T_)EBLmyM
zQ-ZT*<o6X0-s%MY_N?kW@7=-Kz<PeJd3~KiR)IZ+40)H%xPV6(=W2%RMhd{>CK5w4
z-)f=R7_5efa}QXG!fU^Q=Fc;uqI0%A0|4mZ{v(>vdeo-4jHg5{g;KnL(W4r1%2kcn
zt$L@*K<@s&D0vL;Kb~X!!>@TXu#DB8?+efnsVnQ)n7VNJ<~{ZL$lI%O?aJ1az1~I&
zQ%X))*8HC0Y(o2^U#9-d$H+BV=U4B;xQ=$l0SDzEJ2RCjcMI!JzfZnYQtGtFmq^`t
z`E*6QSTUtPwh@RaWQ#s%68OOv^xyiB>viL@)wxGQMvlZmOhDi=8<C+cD`o4B85_M-
zPo0Eg<xdb|K0v<?N3&L|g-Vj>g{jvAkVRy5?9=?ct?y>h6;BU;-k{<e&nB>EnEACM
zzkr1#oGi#6cofVao+`LGhHe1-Ky@W|AMR6Jpo`TKl*I}_PK)*G?p|Eh01=}bCPD6Y
zXxiS`_vV$wo{M%*y0yl>@6=L|-STR+_krMgZ#-?gDF|SzmE<VgSk64J<bv4UYi-p$
z+HJQm+3JO8W$~PQib;=RGtXPIKYKT--wv0{mAX}ODI7haz9!_){21jQc$Y7|`!@F{
z3-kAy$j|TC{`@bE+14>FFBrS*AQ}kSwrWT_Stzxn!kVm3qpSLOWefKiA&`z1yBH%u
z#8TY5@2*T+Z`8dFPbiFPC=Fv^OzyF%f7y|ZQAQ3nFU7p~W(PLXHi0mrpb`0O?pFQ8
z8u61Cea3pEgs2sfb!Opfj2#_HAiT1hE(-<`&8P_=+7X)rQf`5QA94$MGNJ&%U?`bE
z+LR(9!$JAFr}stgEWBPeYH9?5ib$*v<4<!)1tE<vwj>8q5xNbp7>0tjpT!UTRCa5v
z>+#te{cAGYFcFLzplIXk$-u`ko$;A8MgUlmanY2mJ+AqAux_42Hp}J){P<mmnuX9!
z-8{Y|iDg+?5hUjU7$lxi?UuJ{ZLqC6j|H)i$G_@$jCiT%<>Ot)BDZ+|1Tm9dzC}TH
zL~}&Y_fM3k4N9VR<+4|H2jbI+MMs&=6mv3elkKgA{tthN^cO!E>xY6wdvQpK!f3pA
z0B(_4VfkZUr@#Nqx|8-Rxn8+kc3QK#AITO$Y);x#Jk8v?6(}ENs`a3-n}B0>!?^`C
zBzQ;_o2^qd(mA>noHMvHEGiVKg#(PW@6)Ign({p(HHQry-!&Z<!NwrRyr0@Txp$Iy
z_uc2z9d8yHIC}W_aH2>N9B7D}+OQW}w{cpDVi5JPi{;HH`cwsJu87sz^dvz!B}58~
zBu=a4-P2|UeQ%Ab6)=G_ran(qc(^`!*~wta^XkjmDu_^9WpjcX^D2xi32ZaE^PIQd
z*?nGp^vRc;IJw){T4mcCwYq>7&mlWWk&axr<&au!R0j?=Wl={zJ4MGB;ockkb3csy
z;V<#-`OK+3(XUs;+m;^w`5#9A)(6zze0ea$zGultrK&|j(CJkqgp$1dMKWeK&AS<j
zPQfO_sdqO;-H81?dAg_fPTi`55euCfNa{X(C#v_&xT;!ztr`j#ViJ|v0iJTUV#Mrq
z6rGt85M=b{hWsJkD%h0Zun)4Qh=N3Dv3V;~_s|<XjJMs_Qmh(0QaYoHpxbxaep_0m
z_RhL2b{>kQf+qO&TyQe(hlabzmRj`A7Ee2cl1*Y^-_aKL#YO{9PfrH=yf}5F34oy*
z-P;O<8$UE-VuiV5kM*;$61n+{3_wLc%FN{V7+|h<>o$oy$r*L0B7PqePmx1fDRiMm
z^nrs?pRKAl(GGO*kc``PnIEYwFB0xf6;58r7+YvuB#M)FZ+VST;ZY6GO*8;!FC38T
zP`~AE$sRnJ!^LEz5CeeKY9ws^5FJS=A=NrIE*JRpsdCxPD4o3^y0$(XCdKQi1ZNAh
zd)a^b%bb4Z6XfIFQ>)>b>kS}?_U>4}u8Q2RtpCIhQc5DxVSgg^&Ske*Dx|3ojW@*|
zm0H=hW_3kr))|RZ2j%Y2h6$5{W6ROQ7e>UwZQ!UsJM42B0L;N(q_+mx2t$7;P>e!n
zm)#qJCjzJ`DRmoX?*ogpcjKAKueF2zF0VH%ARNyEPQ;N@8Z^WlLRiNLz&P@KhW)e|
zYF2_wNWvMlLX&RjJ{LhsvdVqL3hmU}%>=Zm<;TT_%Scyen8LkMLuB@}2`QOWvR7cL
zk%y>qm}_R<fhGS8^~Kmzl&j}-wAphe$`ITrkTKNjRJJ~5SXCW#SQ<fVXWNYwNihbx
zeQ)&IZv@@K?v?cR%6(b6Pm1<Tm4!qh*G9S%mVfN)T>p=MYYb@Vt~g^hxHiv8r)-@S
zkrS0<{r`j9Lg!GQ1E^57{p%Te_PnHny+PFDDS}8BH4pi<Y6{EUB3otO^YNpjqmoSE
zjX=ut{RhuIyrXGD<OXO4yI71*?j37i^yrl_MXT3T927WqkZWxoX*#5|BD^nF=u>NB
ztJMuWqXr=+GX`OkMYghex8rb!m*aAyVMz4wSk7yqVjjh3Oy3vl_3H8I(01o*>CbeW
zTXJK;U_?rSaA&q!&F-0*yPs<~JXil_%MRTZydJqCqGlj~>Sy$)EA0R$Q9E_qme{)p
z<@Po`^=N~Ohnm1ehBF|uy4ZE2MJ{S<UJR|{8I9ATjqV~DKubI|njDOT!8v%N4o+t3
z;rMyq2@NpsfP0|zv8euCCY#v@30(Kiwl|iNS#qMa##7%$H<)@oL5ZSScL26$iD&<<
ze-Qq6KZ|^PJroho8j1lv4i;K^N51i#{EcU<NszA8da--)_0K6=2SHcXoLI7O-8Qy;
z_hM9GbH{Nh1=t8Xe>8e>`2W$7LPWWyY{j`|M14HX7^nh94kz^AtbMS{4gKLM+@mg*
zoB#LtfOv#89d4P<DDL_kuQ!VfI7+c$3>L~7g(rT!j#}1CAZUVZeH6MlVuIe(iu_d2
zER?0-`$B}5sh!sCvq=(7pd~d#ne91msHEMa474<Dk67_gHZUNiSS}Is4#D%*Ui&Uh
z{Qk2K1oA%jBu?(IQ#A_lWK{2l@M>2|F}{SQECh*?5!X~NCb25h+Pd))RHg3KK+-Ul
zYTDpF!;6)ZhqGEr=B+DKGDQ>TKlVN^|Kq>Q<(EIDwjEVvU6#QoK%YXqE~}5&Oan!#
z*FXrzYUD$=NCNrV1IxeqW32aZx4@Es4|hKKcm66*FCWc`b;GM2TD2d<DCx-ahQ~Aw
z!}JURkDz#J!YhA=G1Tc52G;<XLK#M&2Mu-iwt((l76VzU;-3{_!ec~XuKRun>x=Kb
zr&}Utr*^}poa}xj%gFn;@PKmw0P*6=0j~^r?3E_MzBIB}n6<+sT2gU_#V>YbV5`-=
zE_WQl>tzN%qNjeo<Z!-~AY}=I&~bhVDL2abIE!R>V``Dc0DdF7HLm1;I-^|_j@tJ-
zNt8#z1}Rh=7&olOw-G?k3kMI;fss-;0G=qGj{j~8VLc)H8u|@lv^mDube|k#<k<RV
z;rcYLZM-mZX2Q6yh@l=l)SHQ>IRH_gk=+v9DILh<B-GZ8{{cv)cwN*!jh>{Tn{!I^
zl(=p?{NOWA|J{F>>%a5Y(C<7BXKYS;bnp-cnGVm-EdRofvX;WBB=%PucT2&Jz76t;
zxhi9IG{|~{lr6QbQQ<o!Vf;1pMS#`N3HCr9&Cwns97ta98t)$?;15OZAT$CX=@2EW
zX)>gcypCzeM*yI49(<2TX!7a40dSf6>-Px^goiW<KZ3mv>_)1_z?={qUwjy0t~p-U
zh8S5(TTB4dab6;~BqB8F^aNKwcnw9fMGa7=+0Amp^cB<0!pjtAz(RuA&j$Il-n@kn
zLl=RViHdj+wfj9?_e#d{-m6pi;=pIJTXBeZ_>{Us(T&v~xCbFXSEImfoyD(Nd}>vH
zNbrt<ik=IQJ#=bDrm7&%?|A#a_Cvh-cYlYcOGVUMq+$}}sPxyHfho<I7Bj{lfCD$_
z21K$rN5fBlo%O%*!>nIjSVWN+b^K1@o&VlH#P#3$Px<JhPXTA_>q=@)*yL5TW8}2<
ziQ?R-#)&fnuWZ-Gb+fdKr6~+R+7op9yLw%K+_o%sF9xl(jW(Z-BS-h@IOOJf>sEhK
zE4@{&+jdB69Q)Gk9-;!<ioCMri2YoI@@~Bv^Ia$nj|$j?K4^Pqzc%)()o;|Lp-H_M
zY`UD11@U@~FY+s#f=E9}$}{{c$;WtGpY`Y-^6;yDw%PlJzQB~P8{=?3Y=2|14y~Q|
z{vC>9?HYJXuBWHGOT{<?@&K1nTpC?*0X9NLj^CMbqStwzB65Ly!8c(CnUu3p2a~3q
zAyNm&bsW)*05lTc@6$Oo3}G6cYj4e?wcY}mqa`x#$2?kFU-wcYN+D)?$wCnjF)C(&
z{LQzKZ@q=?FG;Dec(JXj)fkj)g^+789ohGtlwr+@(^|M&Gf6vj7uxla^@qO#|JqOS
z^l$$a^mkqYeg-4W6ZfRhQ>Of*-{SrceU<yW1-2`_?UbcZue%iqr^I<FuKSC(0vm`-
z4o<gy=lS`>x-2}_I=bR~RnYLu5Bqjo?czRX*tfAQQA_8=r5+g*Kt-{^njXcBEjT=@
zwc1bJ^SVPYK%-3X2WDH6gQhrZpXc@Fbi|<`bBxhFPU4yd&Y}MBpun8m(W4y)%ITYI
zTAxjy^J)zD5CN<ejFi}D2v#hp$-=YhBT8d1kv3?fkj;cS<)Ev;+rkp`PF1TNXwkam
z3L}Q(?0aI*0(lzIr4*`Ga)P_{#QAh0i||BLl^+hQdM~txnN?;;x(!j1Va(6~T1ID?
zd~@k)vlG%n$B=kTBTT7KyRiO=AK)t=J@M*){g+VHQO$1Yi|WOFQ?=}^@pQcau;@Z5
zF%sG98oEvNE8q7P=YRHxIem2{O9K*(Lhn{snZLgB;tzj?-}<%R=1DE(QM<CNeH7@;
zLPR5iVj$-!9|~x*^uo3`3pv*AG^RS%6OrL{MXol01ekGewb3f%h(d((6YK4t?Hw;R
zt+cvRn}LvEg>(x)<mM>f*{PMg8>wio?QT4dOl)X`jyqk1x_7oqW7|x&>)w&<fXAJb
zoDc0pU+~C)Vjf=3Mt(KcML@_y=HeKJCpi5H&za(mf-BD1t~fomq0IGYtf6ZRiW)cM
zpkGG{E$3nbg>r;v)rLg5=O#fn7uOw1H`026P%L#NiCaU4sh*+RCEgopJQ?YfQ05T9
zfm;J~8(s5AtW37TNX7LI(;gf;h`cLH!^=$0b1~KjB2X}c7gp;dUcoL|ket{bD|<E4
zFFb0|G@;cB_STYthLAa}3#WBq$%&kx$ilu~>HAK-yh4BYZI*xYkI???&(Z(-Z;^Ii
z!$UzM3hPgQjnlvU<2--sfz&I%{pyNV<+KzUom>iQ&YaIHcjr4U*PY#G<)mz2w0dX3
z>D_0f<t1F28BY(}&sIc6?P9FGrK+ah2?z3Mq|KRhFs0pOQ0L-?O)|MiWYUQEgk9(%
zr(rX7UL*A4h+L@Hb-L*jg+nCkcYpnt;u$~<^)AH-g9ai|My}THbc8I%2|g(8&x9|`
zVKYJpSJ)hKV&W+Yt4UDH$2RZt2CpBmHXG}yw@_9PHdq0mMFNUgmbtaY)-7+}vT+;K
zYLgDL!oF=h(b=<A+()RtI<uLfKCgryp2<YUO`XNi=Q-@s{Gg~2GwD&mK(Hz1lvuL0
zy|t%`yxQTN5S@d8t=&lu5;bQY{^a-b-e37ue&s*^C1UDb)ylSQ?AMLtNdQ$NhuV9i
zYUhDF>#fh3)~pavwNw7|4{-M{{vqz(UlD=YSFiR}=$S4F3gj=JI5{y=S8wa=)~i22
zp>oKj9u%F-v_bPhP20Y+Ie6)wb_nWNBzm4LcOUZw!jiNAl|X90ZiP;**@$2<_iUWH
zt_P~>rZlxywAQN6q=-qFA(LUq)xArP?u<x}QkKGLT{*8O%jGturb@=-V(Zp5f)6L!
zw#v5I<jo}6z#TdwI0ua6-r*PfRkG1t=TfFMj369B??Srqf$cSmHyQe5@R!&o44#XD
z_r$~9<188LBwe66=s2`^B{2{n!MTy0jq|mTtew#g(or3KEAw773R3Nc)5wblP;tjZ
zlBW|Q#LRfc03{KIw`UPTg@$K2^}EjhL=J0k2vg?_G1ES-ktc~wth0d+9S)H_*91Kp
z#)u#)N{8A8C3&mwvRG;$Le+~G$l9N#GnbOMKd<cDhUCPuEWEnz>@QzYdLg}kPySbb
zobso>h5Wf+LH^FiE-be0^6nj{|N4*c?5zh*S@`tQxb9FS^U1Zb@5WbX915A%8o3l|
z+sL`FdUUy6EOdYR>2J{gU%zuhsy60Ps9Ne%Zw<ft`h{H@lT(SM*5T<V0ac?|S8ZcI
z<M*@c&g7!^=JjGg?_T2-k+m@VQ}&|$eL@Sj)8Fkdevj9GIR-+pn_9e;D<m^2?&lDp
zKtOTF3<VoMmczrTh#M=fG<VcI<_LHtdbIy3V!4tL4)h|v-3r#ahq>cgy;~;$YAuP$
zSG4<Ny-T&ugVvQ^tqrQ}ovk-E+5C_S*S-2AyUJ2B=VbvsjA|Z{k-2hlFhycO%B-qQ
zN$A5P^6K&&^%#Qw)+7DbK1IL&E>;>HMRb#oJv@N+?8KM<hd<5r->Q7_bHBrNs|c0H
z%Y~=Q#(7=b7Z3wNz8)fQ))Pewz3tRqk$?P~-2DrGANOB5lO?gM<+Mk>QW6=ZQIMqE
zuLTgB$*ny)2Z^fO*aPeIiqjB^UEMqMdV}3tIaP69#9@4CcJDUwxs!anvyQ_N?MvNQ
z-7P%g=>fa{rZc6})reEtZe1j8@1(`O%Xm+VsfGUj129RFSeKPj78^=KqU1)&3n^`O
zPtCuhC@OtSBn(vFJpEApdQo8%AW}Z(7}7teuat|;$0%heukc<ir&{U6K2CQuIoR?h
zHB*{ue+Jn~)O#>I*YLG<c$69Z{8=4R0MrQKmjEdd`Ali(In!I+v8JFzvF)v@^+-La
zLC`VeOQQJC4uqs^UZy(h+|fC5)H6c7Hy)t8c2yl~8Q)?0r47)Ui|4CHt1E8Ur##&{
zj&k^cJ8yoF4GYcGZO)nejTfd@TVtyYk<7AWwjH+J>Z6t<+^q{a7tYQ!>sK;E*t@q$
zHqy4E?Gq%yxg_*!E9GDNhe$7ts4e;MFzz$yofA1_wteSv*^LU=|5u8TyuZJ8wW)|A
ztfv)`L@J%E%H!o~yeA@D_DXY^P>s0<FsuDc4$yE0bgWhE^^q50fHuKXMjg@Qt*+`V
z&gRL6*)~<>fUH@&cJJ;ON}eH_vHw$tgwuVbrPsOZzvt@>BLj8-$DHXyPDC+531c|=
zG6#IQ{pS$&DRWkYl4s?l<<Nrf-4;o@1BN7M>Q*a&uq@V>zi*AMb++0Jw59HSna8Wz
zM;Anj>xZZ36||8rU;GAays)xu=5<z-jIt*4{x6p&_Pr8P6gfL9D0AbZ>Zt-#^jW-E
ztSc4!thKm!?hS4{#>9b$#S33@anvb7mqdPX$JhVNPtyNqe}Si$SDJQSzI<Z8R@O2Z
z5W7b?_LR7!D~ht{Lf1zBli%doKl4q#{MJ1s3015pClrk*G)Epx-(ANcKPsv4aAO8}
zuvbDvM1s3uDz#~JbqIBM^9_qi-H;+3lv(s@R|ga7v)YXZp?k!_!?|H}RA@Cu7gy@m
zD7lyc_h^=_RkUxJEE(vFx`JR=Ox_t`u9SFDWpyH!r8ctUxi)^L_STT@k+lPm4C>Y@
zU+kiRlFPv(HQ961Zf_V7fTO89ctI&Kqo)$3Mf0<v0FU1pzlNJ`ki#dZH=Yw5c<@0-
zM~zj7&(B32(CznFZHFYEz$ddiDRRf+-!t;S!gPG%4D@as25}vZ4|=#32cR&VtN@bn
z1nwP<yz_kjjb6SvKes?33=Tb_kr0t#kwpFnzT}@zeDvuP&+hLy6?2-pD*L5URd~Ga
z+@DX}Erq-DLe$nIan@Nia<FaRknPI5_Y2SRVrh@lnfm-5UGE$~hyy&UQL9y1*24XA
z$C4DfMOB`*GnZPa&2(?cnO3X)eJNJ3-8;)tya2eMTnYU>Ja>t7pITjun&X>vu6A@}
zn7)feooRp}IU_Y?!OASj(<%{X(xEG(Mb%^x7V-fg4o}NJGq*JCXTZqMi3Q#a3=}Wc
zo7I#dE)xulD&4MgveO}QaeUcx=qM}`aj(s}21LANOUxi5^Khx?@Ss!5-V&m~&h=@Z
z7m&FnHZm8Ib~9=Mnv=I8x{{BZa}T}bD1HI<&gSI79i=2CCuP50*i?D>>cUnnFByY<
zM4X6e_rXbi(1H=~?o@o}lCp^zOG%TUr(uJ7+TZ*w_MiS1ryM8WkZ9en7T!-r-HKSH
z@ad=C<NN-ne}Z58Km9d6`kCM0dfCt#VR%c6>^`0ukAz(HN<9hrkAIE3fA)uX_u+&j
zXtEfINK+p>J@cZ3;C=fQ_~DvBz&b55j(|g(3`BhhlLo@|!oXt~)NU~dS-hmAMfHCl
zecRfj&Cy#thhlSO*8Tg#h;vJA+FPgYmZn&uGsKzFS~Ef=6GH<7t@h!O#&?cuDxTV~
zbu+RStgFFq>8N-uDq=rBq$W)v-BhrDF<%DyLqF))m(D>h;x)h`!^0N!J_DDX4x?a8
zJVH5ugd4?}uG+Peha%<B;&@cz0ZQh5#>|g-TXwJEoQ>!NvpqWL1jNFjU^2XM|9<?Q
zXG59i*_+7>Ug#!18&$f@l<c^N*J=s|@}BExI`7z_phH*tKV8=zWmgaHmvJ`jU*ZLq
zanF8Udt=)Zq(aWkm_fDBOB*2ky6-%HIFn>@e3YEHTsO96WJOCB-g&s=t%pL^PFWH;
zRW9|3Cn@xD8X&tX>}r(ir&Ho#Ig^>>OXvH0F17KrH6EWXmOkEh8@t|>TC4HItPAz2
zxo$h|xge59^h#0hgs`(d)7YN{u>lAy=dY&!{5w+~y~Wlk#35Q0dUbr~?2L=zbD5j%
zX*R2WyXjg{w^9Qs{@$-Q0t3t9>`C4GaXHxcKrzD`45%_ON*ME-S9pHjBF5H`q8@f}
zAw39w<N!5{yZRW@oGjcQ*mzvp7ZmGe5WJtaMca;@?3@#5u+@nd6PZy?#c_I9@D7HY
z;k;xX&MQkwJY6qbx7`Ug`WSMOvQ6wVE(1(Rk`5xB(`rS6HAO_kid*uMy+Yl-Lx1TL
z>eI%lECfSOcuw?6y*?qVq@|$UD%0<O>V3ZU!*B8U=YJ0QKmQH%Z-3yG_WOubETwQN
zHlLw?;~D8+`EefpzIS=CDrlwYilz8%{cGD$54okXn&5TNmo_GL81*<P-Njou^co-a
zm4ksk_bt@h@y{4nJahju`V_gn?rn_C^ZweE?mPvO5-~}8M6Qw%Lm@Wlu5Oj>TAdTY
z4Y{YJn%A0Gn<nggW4rDa2@)tvg#P<Yn__Drk#k+f0Hy;@zUin_EUnRx3g&ad`#-=h
zdn|j<;PHr8bpDKbBV5d#@-d9P<0U`V9zi@7mmNHz=^5$h<DYmzZxr~^(^4YE6kZy)
zK<4m~=EILvEJYR$z_IOb0J($rX7<Ft?-1o0I?3-Zx$ct2WIap<vx96#HH~#m2%7Ri
z*1C799U!J2&K>Gemg#YtO7!AMoi{xA8r%UGsWLT!q1u#>K7GZ*{hfD1S+TWl+wg{s
zUOiJU7f>ZFh0C_{c-;^YUOe3K-dlH+<Z+smsLczCFN&P*=#pq^Qh~s;Q#NYhYo{b3
zVWZKL2)PtaNg-t(U%jH}imEvaBEtRoM9GELHiKOet&aG9Y}^5uD!6med|3|>tby`l
z!(~J(Lk^M`VS?!ri;lRKi1i*Vt};b>h<02l?3ewSMCd7zLdUpWl^f;qy<cwx2G)}+
z-2)eV$hozXnb8{n_^^<6-*FP-;R01zc%}QO;zTy)xej}`Lcx@UEM_=G53M{NA@Hvy
zIO1)7c@*UNatIDAjGUqQS6JOcLb4i}kSnT=wLW%7XFC|@{*ow?C`%!q7Hi{Z*(g3a
zpllf4h0&6#Q`TCVBfHlJjG+_JNszpaQ$44Ye*IJS-~5>Na3Ynng(Tfjw(AqE?$Do+
zIMMsXxF&Mu`Tfd^Km2t*{K|s->L)Dg3jf6i@WCrGiHGx<`*o#%<sSK`e}ETnKX6|Z
zX&X%zRGqh>L))(C?$h6vB{mhV8||<Ewt<F-8g%#Mn@2;OvbVKPE;=SGPOng{r4gk_
zW+{oQ9BiZ)5JtgSfWeeJ&EiJB2Tiqu0X?vfnX$n=`XGj~{H-Xot#Q5hkPy_;CH@Sx
zdYZ;r=dOFBTzAgbja>39SgcVz>-QAt1mCORX5U+%59g?V+@o69Fd_w`nDz_&vRV(X
zq#oYY!Mi<pwPVz)cwwQ16<z}mGE|G>g5i0a0S^8X*HJ+QtGB32HR#a&Jshg}NBu^0
z3OFliv_N-bvFjp~h%k&OKwLtOfB03W{=1py(T#Dn=ioeX;Ogy`svQwCbee=h0-c0t
z_L%0uj;Wt+7*ew5AI3$y8+$V<lTFc{FM8u`2$DU3ZBd$l%GX-ErDanRzVqoTi+G+-
zBvICqSo7HfBHhnGW#5bo<+@ise062(%8T>D{rN=g){(T9Os~o(Tjf%nD!nQ_2~BZ2
z@kFa-@Nz6UdwE2s<l-!_Vl0k#5A^Fs-*=WJanfQsxhvK8<LI4>9Cz#J(3;-PPV6+!
z=F$IU!p7)`vy=l<*2mYHE{nq9nL>;R>9lV{T{88>e1|6pLjJnarOiUf-{bX0U?8u>
zX4R`vAIdn2c55~mD0IC4{&2_xb;m1^VqUX_Quk^_72ZY&-UHq(>|RSlR7lwrN%O;5
zv5<akoqgMBdmYF+9|}7>MrcYV*tQf>?<4F7I*gYfRLv9z1uY3ZxgaDVEg4OjhO+B{
zQ+qzFKcw}H644>m!P2Lwx5Al{T8oY5HIK->6fEUZf9(^t@Bb3_g14J6jG)%WetBY9
zGrd(foxn#Ezx(0^musbe>4gD|pZKaH)x>#S$Yr4b_iN&<Md@uf1)0L=dcL}D^zF%_
zEX!gFsyEu-`+)lL71w<qg<b%4Z&df*qGfQlEXLWJf}ccKOX2>saDQ5Pdb-^7*NZq`
zM2bUeN`G{5bSxC;`F{-qffw~O3k_OoK}Y*vN<!T$*RFi}>d6#*?R4jS-}jyCb)$Kj
z0Ae(?(Mf~V97K^JFp5Z<s*jcx|D5gBj2Pg`JN}LzW;^;hs>zR*@C*3rK7dEnIO$`q
z{S0&4`RoD4)r$YLIT#+r@G_k)W<Pyk1@nYwb@woGU^?Njj<-s+69ho2YP<)2z*K!m
zS!*K%3kF)+zT;kpw@pU2U&i?k;H$$6n%@z5;aaVhDRJyVIFjugYOEqkibYVRhjZqg
z3I|z=w3~KC%0V3;(*^-L1!s1hX({R6t~hjP8sDA1e8>66-lNNvezI<(t`}ZCU0Ihh
z3JoO|TZCR4VCVDY<;t}uE_<h>#QoiwWm#-a%nmQB?ZL}y=czW3PI1HPw$9qBlhM^n
zu@1Mz_y5XHazKz$g1o|NY31H`<00vleY23^{g?0H<Hy+bA!Mr4K|B~TM?zDEA}DZt
z*++EBLP%hw4I+&1HYbJ<g?$plYLI5=VeW08oE-Z4RRdH@L*@5@XFyVhW``yp1XXp8
zbyZ)aQ8_DdC>Vz!mPB$QFPT)5^<Vk}_9i1wYK0Y@UK_0{H7nVvUjxx=vtq#3l(x6A
zNWv^|G_ZKcW2bBt@68H!#OI@TH(ILQLQk6Mf_eR_kgXz2#qxc-dkl|TW2?3$X>6Ju
z0Lhu+20lX1rC3#|V1p%UH^sl?NnhNx8M~Ap=K8lkp#Jf18a;7Z?Vd`ZtS5Ti*`6-k
z-JeOPGjh5&;m<4YynW*B=g(cxoN-HyTR-M(D+UH+f5rav(u~t7)3x#GN1s~fgLZ2X
zI^WHZY;Wj)@=NGr<FZv(#L7xy%|h|EzlP2C?ro*F>R~lUwNB_#QX*@nFPZb%*b{5_
zL*WsCr6gw*gRBP3OxN08?zOUS7V1x8L$4Guk4t?RmG^>2%c-hG6JA|+o;G7jt9xhP
z8g;8S1LL|6!l5ywsDoCeC+9U7U?7bML~yw)-agr*?Q))5Q*16PE;}9TqOeGqe$hu0
zcqE{sJ<Vv@Axs}BrXHOEzQ$qbqfk>VjN&>*#&?HP7Ez|~>~4%+cu@y$%YNmK!uU~m
zfrseDYxf%|4a>=kgOEHb$&Xjikkg|x6o79;f;{&g|1--D4tA0QjZv>96HFZ=AQ3f-
z!nep9chqYv<e;DR=5Y`Hum4!a_aJx?W}KOhEScqDrEeQgwe#%xJ@@OAO?q@%^154B
zk$RZC_eM&t^Hj%7nDw-h%fiLmiJz{GCYi22W5Bx(>UMGP-$${fq(oOC<%Pa3RtJ{E
zQf(pz7INWqe-D!AL=k#o-x`<e#_8_F>Bqms^%uXxE)jX2K=F9|jzmDgs5{`pLAJ!#
zS+v7ozl7jvFoxAWucclTt}~s+LH4He?cma{KX3I=GvIe*X}8MfdA$)BXo`oY{pCns
zVf1ZJPcPjf&<GETheyS$>DNMD6DrBWyRJ|L%bB}~b~UfAO7$s*26IKD(mag<uz_?B
zIl?5c=FDBOoZj9nEm7S|j5eiq82XW&88JEl;+cB~q<c$^g}k1wMp)zfcT@0{atz6d
zp*OZUL2+xFZy_OLbrcVWo52(e476YVi2BJBn-rGS*0#1r-}jj--*@hxJwPghdCKY}
z5|kB5h15H#@6fNHt9JxI-7fT0=(V$5FO+;{yEa~a^a=Z3Ih_{n?$0Ctl_bHheuw^7
zzfIG`zITgih;X0bv>GGXwpI2XQ@kpvXAApl8cNbY8b&0QY(P<2GP|c4s`oADELO-0
zWDUXZMD=oSb?}UzO6GJbG&Q=DSf730rx)e&)tUo^Iz9V+t8CYu>vd<}s>!AXu~sK$
zGKhB?L>fdKyw+O1hUBw`XiA3+M<XRDG69NB1kJP9>kIr+MtYUV*Gm|fp$pwv@W?>a
z;*AKB==giQgsFp~@N|6t!2<{ASnFE=h2eSOdMYaY51=5%rH2uXAhNMxiq)bFqu!5~
zo8Ux(fzGah%X?5(kN0~TAXpG10uU29r_|_>8DMnYNY~+o4wNY}Y8|Xj!&43(4VA-M
zgn@~nNr0teQmjSt6oPnbx3_EWJIiX`_w&<=iV^>u*34sqZBx@zzDKpX<LJ?oTo%f5
zqLaDQ#&vHrHNKe!-uYzh-!@aGBHbroVmv0TRoZ?<`yH$&uN`Pa>ZnTwt(@*3j8v*C
zMG~j;LQ&=Je&yjG{W>rI?9UHRT?z4tHbHTl6VVz=Gr2MwGAHiId;gS2no<b)Fw#+M
zZTe5=zIwaqC^(e^bPHhTaxmmAoHOS~NFt2ci?4?(zQ^m0z(CzQ(mM;sLKgpBRH?gB
zn_(6m;T7@!kHR=9nY`Lef}SmJ71X?aZ)3)SNSfZ*IH=Xq385TXr&n)n^vxXf4{C>T
z%@IN}%G{(@R7iLh9<s9$mZrY!-wh4^fw}s0Rkpfwxop&}Qc`AF?4FDB1cVMq!-I{&
zMJOp#M7Rng$K6#&SSWg&tu?OEe&bX2U;hN&df@5lkyK8Dc~NrWwBB*LJG*QM1dWR-
zDhpcf&~ifhj&2R<V(e=8eu+RTEO!q!4Xt%vzI>#6Th6j9NLk3s8PWnf@aKM!^wemY
zI4v1TN-2fYl315QxU^3mu}Q5{Jk11(^xHP7-WrN|_2;#))kfVKdv7$ap9uc{6sZ&e
ztH^C!>&8}X(8#i6)-_Wx559GuWZ9a9SjCG|iAr_*KM!~r@T=Z#xL2pHjy*C7*=rOP
zs<p0GOHpmx#<o>z)nUw?2F2S$LyG2KY_OH3phT>mVaZ}ZiqChT|De9C7%Gi@W`v^s
zJOt<tz2Qo5)+Gf{u>Wv-_1V_a!{7ZgH^57x_}QZ^^4f<E?;T-!1WC>laF`P@yui3!
zzfHT3X;5T*-yu?};q4`$9lU!hoVJ-Fb~5IuWS1e=A+rO@Jy^_9nU11I2b?Q^db?pH
z{Jxk`mV?zfNSr@ovJNRwG6}x)fTCQVE<9Tkb-VKQ_Z~c5+gUU0DY4U7&g+;-F{T(I
zknBVtiQT(>Y8Njw6r|&XP@SZBbi*PZf;mHpMr+la&Fdr3AQ#MMaN{Wr**k0t%l$nm
z6^c|Agu7=ecMm5@GiL+vwbvV&x~bL=uyi8>GUh#GuL<J-aAV|hM0%rWdR8g;{YHVa
zO#3<=Z$-kJzJ-S_V|sJ1_L<*@0zx`t7JM3M0vwsmkZ8_wK<sf8X5W2RC58&cBgMqG
z;E4~pHHF<ne~bgCeLkX9aDz6+=-PA;aK#LAcH`POY_LKRYhm15ApDu@s=cFowQ!S;
zP7)OJ5NkJ&5*5y+7<*z8%f&BwL?$Lm51Qm)xKbYh>=6zk#5WJ8cJnm*b?4Q8@N+!-
z;rBS-o!FWc?no)D=auvQiKQeDi5P(kUe!)sAlc7kzmoQg35hs!fr6w0(vbCz<sLqL
z`N(!PkGN#>1|%0W7t${1zyCMs|DRt2wZ8GSBwBB*OXmK(kc)@;>rT&jYhN=8WA&a~
zV&=E_NO=pDr3~vSldlH4F8j{b7MeCI+?l7xQcNAH&lR|C)jaDWoYssqWv@HE?MTj?
zv(TlOm(u5YCy_ZOS!@oydPhKQ>iiFaW@l8(R#)A7LtCTON)?NeZ9Ww(xC%n~Eje*Y
z!ZnE#E14{%1nZKxJDU<SkI?R6t6zLN;t{&^Q1}dw&su^pT<oX><E4kzLc*Kk#?gom
zr8KK7LJy4;j2qoNJZ7(p8JBU8Fp84(=1+ha2Ry?1$FmyXTd{Me*!-0AIs~+9k378C
zomhFXdUP-lkAJy8OU$>J#cMOY6gqc;dk}1h(J*2My*!ob1UVzRJM?=v%NBvN?g?S}
zqhDjaTPgP^PFZO6%H0BTf5+of<$B%u<nao!a+iwfgMFWMWj@JK#F!sPCs%fEjcOeo
z-MRZy5>6$P#iAey*mgC&?Q>^T8+E(F0y;QiI_n~oq@1LZwX&=Wx-aA;+$~T}newXh
z<rmLUedN_;BLJ$1M@x`VgAzsJB0-t20}LdO?%JhIH^nGzjXgT7^|52aoRJA4kF^4=
zcTyLE&RYhV6@D)+&jV<30D!;u>&+MlqoTX`6t_^ueO8LwLAdGRu^SW*qxl0S&+8<G
z)NQP+h%>IW$aUB&j!qcVsCtcIN0BG%L4~ZbG4_6&qhQg6Yi;b#nAMzRkdn;%_VA1u
zeA|%PyLo1V3ez^n4q+cW`r4IKv7s0#S<7I~b0hnqU%f?<nWNU`Y;+#N@I1JT2u3g)
zw!PB+>ThuQ`QPIHC%(zu=}gU;Cc=6ul*KEKQ+A;U6xxvDu&Y)2{setDCDd`0?aE3L
z<b`r4oc2cN$;L(tq?BoOX9e1y{X6iV{Q_Dynq)}bz=kyIht7D>B~YwYvbRnZGi=2V
zPk2e*ac=o^vB6WNOvTW86nu6suF-K2TCbo+9lc#P_UnPXNrY1dEtw+9xmb~AZ;4Aa
zU?%t|?qt!U2Hx7#EgY$xHNok;a4uGqmc?s31XP4Ii}6>~2CX5zu>hM)=0s6gMYvzA
z;`y@AuyYpSt|Xo<)=86szrS@JLwkOKU+IwkFdgAwJ>B-VkC!-mUvm0c5%UTY)~;#;
zKZbD+JrIgJr2yLp-iw(VVC3WhRGigtMiEBnJ?Q|;vn_NItA#!j59JP!<2)58NcXYA
z;be&1L>P~rxp<ZTEYsr_5XfPO&R94578R^Wc@z-F=V21!9PX8P;1RF61z}=q6XAj&
z#L797p0BJw`4twAX7zpJlwiN?NEg<1p>0-({NdB5eD$5T$(|<Ns_9c3@ReCkD`i=1
zT|Fdc=@N?s<m~&Wo%51dFlQoZLrdXW8(WJGl+4z;5zkd2mBhISUpiYY(_Sy!<vT=_
zUJA8kWP!B^s*kKiSeMMxo<@{AOxaA6q!E?sJ!;zy9{B)>4l{c=JYioBQJS%TRuiRa
zPIl{k?8Xu3MQuvRUHp9h-E3Dl(jm*A=k-QlpjRu@3!#qaKoG+CJq%<rgwX+k+wS>*
zV%-AjmDMRFf;*`1_7Jm**s)hAm4o*XEgLOR{Q)TpQL(FRd$ppK+O4h7#}JzWvB*Fh
z{ai*wRGTAlFZxJnG8R@<_Ptq=Ol>p~TJeftS2nQ@M$>E*otvjqnph;`vQ@NMj(Ca5
zWs$o)RE{Y^f2@4)mwtuwkG#u$PDm=O_h$+XiGn6%v@n_%VB=Lq6|Eb(Khe|%<U|ju
z25fac@02ok56^h<u99To@uTmstc7xSkN)isx%@{zN4Z`}rEsZ@oz70Dg*>=dqAKL2
zaCc{6^jtEMLy0rIDaG_o_q`2cFycz;T_36#-K%dCUKrIor8O3md#@?DE5a(ElE{m4
zDw*?Qbqu?7wif8y4&r&uBnp~@`?YY&26)dUQSMipzLt|kzFaA*TDO47MJIWo<w-i4
zg)IqB2d^t<VM(yhDpMk%wTN(6d}OwJ1iR;$zvy6KR@J(ZmGGithYg5|7%&y_$!mVj
zBXu9yQv>XlW0&9=@b4Lh&n`C%y=oNM`Rg-=P>)^*k!Bbnb`4L?-{VT$T!h3RT02Gz
zhBJWc4bvkMzPX=+(1-Mjv@5C$=rWuSNpjOw61sIpXXDEZBTc5y@Id{M{i5kc{&a#Q
zzi&ySXeqD&y{nT2DeL!t8J^qy*=X-hIW3eWv0WSWVgUWwTKMgkPyEJ*AM)Odw?KVR
z7s7^+a-l3MF*GR=jz{vrJV9@7kw;OBQZ6LcWNfM>3v%7b68Dn0s&MTBT`9S6UOXLH
zJ8X}X<vD#h(fZEQwi*5W<Av6$r}B=}AGh{o9A^<<>uLN$noQPAFBt&vHtXI;n|tW4
zIN3=||9A9I^LMbLVpp)JkH!xMko=5_a5?<En&Zt12wQi9Z0ILK*{ypao(qw<rZz=8
ziyua=ESVAPtvdtL6zsehN4QEtJ^zXW1@RDkDtQdr2!m2vjHDT~B(YZPU3sdNqKUoP
zBs{yK4d#)iWQ-QGsd?k(abhigl02N-$;veHU1`;M<(qP)j#kD<ha?ZVS+QZ~vQ@5E
zn|QdZ6--{P6~#h>V{GN&V9x1x7D4&sFZ>!0|ICka|L~TD69J)Fmx1?iHwAp`GD%Re
zQK7wV%mE_<PMPdr(T=FpO*!%6ofp)-Lwg|Qg<MY5KmWJ6Q_wZ@*gKmDn<^X1u13>b
zv9P)#l;X3~dt(tJBZ(`z)Dra_md>%^CN&9*J8;r1ny{c0oeEw2TnLFGnX;}tdtqvG
zSr*$rHGYVklZ{}-4{7u*`ZZRM9^{-ktz{6Oq$Q(E_K3hF!<xpyBVKB)uNkEzq32|I
zOFyO2iO>-0$nENQhKw1EY~~PEpI@QVXntW|;k~4T7aYdxz}+uVHQ2l7ZMl_Fne;iX
zk4F^c_#VWGe@2mBAW!S$kU)Xck`BW<3@5I9^#~I1_r#5a&a>{KS+6wp<mkN7=?#s`
zhj$<P5{0&-?TSdGtn$!(kq7SBx>;p<p9bB3u1?EqwA+3BE;^}jz;yFDM<&PIxTKK7
z=IxgDb$soC^|Vm5@wjbVubWMlXhvXGubb%{vD8=tF5AY#-HFHRl}i)rCR-PB_L&rt
z$i)nAkwk6P)=zC5yZ4=%qn&UPGO@ZHuyOJNNtAmMTTeVyd;YyPmYi5KuwU683!Jhs
z$6j{s5}bA=?>nUwqA+=y+9`<x5a4@v1Y;qaz0;$x5V`r5;WZzA6j3efi~w1L*45TM
z-Wb?Y51&7v6ZPLs4xRuezYmcC4S0Jv6cC|oTa=WNx6qmJM>THi5ehe9lHO@MCbSqp
zNGTgv|6-`LiWL%y*l>$d63ORcljE>?ZZ2ls>*1gYbg`ner)&427>3>@XHqI4#^)bI
zM&i4yT@4I$W`Udkyo(*G?#en^9J`Zcm@)1YrqK`q^<46GtE|No<MnDZw^lb1hTn~h
z@3=(%Fxj-V_Q4|`{>*Rim9Kn>Wr+>1mM>F9_RVTLR?mAQ7Q*k{t<K4w8IWF~?|y&P
z51bd*X*Xk-GV8OqtP1z<d`S9*-vZsaiWzJWni%M+8VaU6Vm4!!#R`FJa#>7bGo?0$
zU5SZTkwi&0Oe&94XPJlvSY&#Eibre|mUZD=mKlo1Lz{yu9IFB7R-McsZj#}76;CPU
z<n2A-C1GFD{JZAt@_tBZ9rDNF!5+fN_I|s!CBc2k16<!Ni8UoFlIjqJK*p|eICEdr
z*TKMq(lWU|jMDW-v~aP1E)KNI)Ci*+3b2;DoToChi}{~@Do*!=2e}j2hDtxQ10Ixl
zg!JQk9V8gbo4-%-sD#U|N8-ll^r%jaFn@~YI6NEO|I<N|MgW3Tte{%b06@4`C~DEl
ziGJcsC6;?WjmGof`p=@fhg!(jwrWE43B+VR$C@cW@g9pPTpzh!u3WDc-K;HhJ}vYP
zA6+iIvH*e2ZrL|Fooj70${>Vmo~lbJQI?E!(>o<6_TBgw_N|iC2G4B?9<LiqO6=VV
zoozcC=L~N@oDuC@)$B-D*qe14S+`T;^64kkt&z^pXf5&K<;q*%e#Ko+JZ<~vwm2?(
ztaXTsj=i3Hn#O+M@)~RE{3x%TWgTnF0ACqEjk$KpVqE4}bU4B6_~{{CF~#Bc0SttI
zSyiYGcB~dOtO%0Di$SVY>VZ#zvNn+!J#fTnq|9E~ZYM^(Eit8pW_#4UWR`UyWuu}@
zC1^jDzFQI5$PN19C%;Cd2@oA!5II@FVuYl8)#I!YH3&@-vT(=|8DF>&e?<|LaZg7T
z_y|=VLZhZ6Y&3B1jjq|MZud%WY943KAeK_FUfmgiu+Wg<Ah|X!f9tn+^~XQuofj{x
z`=afLWb#tj_l*xf_>gzM{2uvyHw4N&h3dJ~WCek}L*4Bjd!zR{He2jn))q-2)6u{2
z>*%Xbxhh=J$;2!QFS`yVHG20Pn8{3*L=mAliB59bKjEP7U2L6K1ZirdGq#DuM8lDW
zVSXqL;e>z*wpYzUbi*jmnFa>p(zpjgK0R)GpaPihQ3IH^rO{`8XgxxFq_`=m+I4&`
z3KZ9pI4z00B_^1Odo7cadbWq@;3WUTzizyDTjMz(2ro>V3UYubfzd_~=i7;X#yNMe
z9qVK{dM`-}<EB8gCw85D2d+?~y8!|$M4iFmeQxx-^Z{_<V2X%EH2fa)weQhk2;zR!
zBUF9p)s(}5urq`xnxgJuEaAAHnTimO4#YG$uN^M0-Rg(Xwx7b`fzDb8n9;-d3PrMT
zdhdZHDYb4$Z*)~2w>m)FRy!YWomabs#!sH--kaJ4%MQ!3vM%c|uB|r~bqB`v$&9jC
z3aQr0UM&p2)ehO(q93nU9=FDqA6Bk=<N4jf%U4&HT-aK-2ph^}+qpZf(5~#S8ab~>
z$;e}8`;UJG*_7HD_9>p3!>Z0;z>EY9xkkp2sv$>WQILd1|Fm(N!Xdz2q{GA=g4owS
z^f53Tk+^1bLg_p#L>k|^Ab%ggfIrvt>bzCLn!zSMiO`*Do8hf4qtz~#1U{@GzyMfI
zN_ULdD3oG=>y$=2*R15Uhwnsdo~dtBy5WV$$k`}ZD2nLamRX_fhD?tKUtPiEWLz8K
z2+s|E2yqC3h!}B>7cZ$A8B^*)qd8@*dEU`C+iVn|RiNG!rYMs7K%jOg%*##!`nP_Y
zr_09k(<}PEBPCImm8ZwfM=!5jp01qEpRi=Hbb)pQ49)0M>%McnJh5aruZ!>QEOKEe
zK?6FADC`^kXFs6H%FFGFrrjbL-E-wmHX^OFAgmJE)X4kJ9@yb!xN;xitx%B7dj6}|
zqO{)GYBL72r88#~;NUT!v)%JZ7Nr;-6WzJuA8kmgbPHv8t6>b7N=HXve6<`DW{`!%
zkfeAiVQd0$IEeG#XMs~mEG2WSpqC`xMWE9IP)|YJIul>mSE#<x(=S$^f*VQ;2~Q7c
z<kgvgEY2Vjy1>6R9D!l@Lz8<Em}@8X?v17qRdS>1z!g7#<ImS$a{K4-?DVzw9Q6n|
zFtwjYHGX`()kb7X52bF=G=Bz7{`<IpZ=)G)X0D5(K=}w8)u)<<Nxy+QuW=$6%5ifS
zzvv9_bWRY3^y6Pb-dd~;xHpd;ijl%~;Zot+)Z3HI2Gmwr^U6{RwRcKcSx@UY1HBs~
zYO9U17(YrWHixF}&B0cJoZ0s}q6*jAc=dE;Ysy`+SqO3Ndh1rB<K$%`aJ^m;l)JUS
zUb$SK$mi$mzwj}y{?-TR>1<~`Wu#A=!ePHczD7sDVQoXTP1q|$>tq(=W@u`4a?yp5
zLY5C-mTtah^3Ws^Pd^`I@DOd70K#l$QfT3o{~oV50|RkdoY=yMS1hk#q4#9f@x71G
zZ|KS(m$4hR{B5+wv_8Fhm>#hNPlH$lLH3Sh@iwhChJv(eiyIzVkonm4bw&oj^Eg2A
zZcT6B!bxOA#0ZUIW59<!1aXjzo!975bW~s<C~|Thk%+WGZ(R2}{@uGjYBdEVnK30X
z>Hx-I6fXuEZASvzcb@pozw>SG{(-NvRx9M|4gU5o{|=uvMXU1SooBTDLh;d`XrtH7
zQ2eV`eDs}<IG+~Ye({{U(*lyMXScTz8gKg!>{cm{&`7-8I;|=tH;Ts^RiQXP1Y0LJ
zWnJt5kAtGGc`OdX0&NenjUF>9;G7fpi{%`9?bNN(az#hYNOIsgmA?i`^WTHz90R2z
z3CWE`g-hFq@-Nvs3Q|ggCp<XXhaAgn12x4SkunZms89(i;#o@reH`eLQurCfw9JvG
zjiUyp{J)z~`a-^@R}~Sth#Gcn7TqI$FH^aq^FLq9?#E7BM|B_T6zdq9cFfvLLm2nt
zl6my-<Hzbf{Q*7>KqK_lkYSlJJzEZML15^&;dMvRqj=YZm@>|(&ss~Dl{xbuesQ=~
z*C9i#1<%U?5Y3R+SPP4a%m{{iTN9Lz@0a<f9G+px=EWv4w!e}oUCG6M{<Il=@BkJP
zAl&MdQpzl+lOKQey|JNTS;FD84x6%^xabwt#I|kro^>6R)}KCZMnCQ~Ssy)K(Np67
zw4h*A#9gVXEbC$g+*>scfZo|Jk1VHWoPN8aQt0Y-!2xN^WSDM%13ZZ!q}O$c3<Ey)
zr-QxAyI<7jpoq0T_h$N_r$b{4+}O$E_r+Ay1I6uYa(8r0{mOBOKF8~gz<>qxrf>xy
zu(Ki4oA!egciZuU4<lULqqV?8_u(PU9#<tg2YOewt@@A~chNDPYmr2E!k+M&12rNt
z081=#go$z5OUc#)oWzP0gB##Djya0=#8*Llqk}pUu(q_4S-mw3F+!T!4S>{UjC9hK
zMGP1j<6WCNMJ%^g#M$_wozBDV+9Nv9NuO5wM^`@h^pW?UoxSx@tqXw0-h@v-`GmLb
z7Ub?N<KqTws7UAW@rmF1;1e#jb1on7Bj0$Jciwq1yqZ=kplp{5y<P$3^z2LQ&hKB;
zn3<D@Vi$$dg}oYmXgrtr_l05zJv<3Crs9gwS=4TvO5yFhl~YdCt}Mm)4EBAes2O(y
z0(L-6DSFx|3~P9qu2csYP!5Hertor36psqD03@moY(ti9hPw7n@%>8TIqQQL6_vx`
z<=dG|?!|crLPSS~(r18QTN4hR-WMDYs6Ci@PPeJT?$p@TWloSmIc){F?S_syC9yH*
zs?IUthD`)(VPSaJesLUj5xMV#ct?`mON=14Th&qgr8=Vz08<q7CK53WZRi>T#LVZP
zsVZyaO_AzQyzuQ1csTqeI8zR|W&(=?9j=|*=J@6SB40BG3Z_rSB?M9*Q3rcWE5?qq
zF;WHs`6u49BHCWjx{t_!V&`;I<z;PLc1*XX#Qo`Ht;s5!?o40Ba4+pj@knu5vXO#G
z0Pjl4fC_1~39BU;+0d)2$xHY3*lOiFj~#U3`CX%R$YtS_Gj-p|CGo%mFJHdm(yVas
z<85b=PkAq8&Zm{9Z6EB9iA47T;E<aeNFWY290&|&CKma)nXwh4<u&Z5bK-c4co+i$
zTug^r9h@FtGR52V?3BlYfAqOtZwLmW-Al&d3&k0|lX3`AgfSMc3=ZCURi*BAc%S2;
zm~9YqR<so;;<;Hn=o5$}SFT$9_bHHC8e@8gS8y;8ikcFUD8w-Z6dd5H9c?<$Umu6r
zLN2{``fdYl1h`ulp5HB;Pph>#s#0s?>V5HOCu!%LtT?UJ&K6`zsDmP_gcS!1>Vh%7
z3Cled3ORSm%Su@n)`u6AwSWrmeHniKSAUx&*_`gO-XY7q3%8=Z@#^D`__bgEfJ^Nx
z#SZW<{=MJi?vz>9GKi6SZM3#iuUF1@XV&#ZtKC}sqW9g|&AM|;^xk>AHm=v5^SV=V
z847e1kRCum{CkAU5AMpc6dr)nQpQlKl*B1e8KK=i<ZgbeV8F;)(Zd5|9BS~PLFxxI
zc^!pLrP_LTuT943>S{zqNxO$LM<#J7dvW>ABAk}Y)|JK_VLA$rQXE_%oVcLe4G__z
z=HLtdnzeMX)}9OXs8=7;9lan_LODV(wT+b?ol6CJ^>ihKg2qz-x4j4uO<DoUl9S$W
z@V>!1gT~7}?@|Aw5w1Ucw~JRqctSgiT8d2%DTyAILM%rd&u^wSQUuWw7J7DEdVqCR
zCYJ$IN0?so`v`qEK*8-ggvC--B=k@UL^_v1ILM#~`Tcu}d)5FcCyH7()^)3Va;;o!
zdK}stITdnBETyoVRx~X<ZM*4C!ReR70kY={&K8>;<F{Ib#iK%|9}@R>aBZDh%~9wI
zTZ4};ZLrOrPb*iCG%fchT322@J@QoH)pg^1x$^k<$chg_b6~E)ArNa%0De2K!DZo?
z&PbDb%omuBBtupvOvIm6eHl}lauRzbxRb!-rBE2pA2ExFfZs^F#+6==D1GnOn}Gqb
z(2e-;3`LDVPStWCYTkSlE{w`}FDgsfit21`gBhC<qPvw>TW~K>j1_C{txFc0p4gkS
zS%oHL5Y3pZ&ZqzZWYSc~wjmZW>msx!s5DC(;Ask9SW{SS{2)e;8dm|Fm&|$1tc%Y`
zXx%5Z8L?4GaLUFtpAfFPd;hsRs|1%d<Igz6Nenr5e>TmY4*7*odGXJ^#d5ksJyiAf
zTX+1}55A9f<;!1shkSkuSx$poC*@*=RSMhQX|=L0!k54NF4}hXrwvHtVhnvWd+&bT
zkX||ez&pJ9+25eIM#d&N0;51vQ1;q*+8XC;Mb7DPpxo$3LF70faICsNhIUg<)`2k6
zN=Jk5``8a`-A89gJ-0U9IbtAAYfd;MeyDRMqDI=<V41TN*o+_qc~FV;5)6uBn}+Bz
zm<<Tj*17D$RWZ*?Cm0%&&@>LYOX|T(?dsDozwobcdk@M`t*AEo$fIgH!u;m-c=}E|
zS<C>|zLU!84o3lTt`gU-Gu5K=l1#}l8S`rdPCV#KGXzsd2s!{pXp=M)u-trZ$fWa<
z2oYhlQ$1!K3}fn}gnOe)wJ1coi75=3xgoM~01!$)Ng8_D3V5e%jkRPxlYy?=Joqpg
z!)AE|#fh6lg!Jwm+^;-sorLw|=k-jYa@`wG4fZaiR4lih7E;bEOX2y$iL?|Ng@#hw
zh9U;;l68!sox9V*TjvGojfT|?h&FnI$6a}OX>7excja-r(pn?0XQVZ>?p%83vUg7R
zXPc*!66@(it2@hj;&~~&+AG(3MN{JX_$j~g%fH66Z~r!Lz5Rk^U2jq?;UK(@%FK3{
z@PwCNofTGnPYm~ls~f2==jch(fti=kxd$Q3hkXhxId5my<W0=yZLYl8>&?0eh(6VM
zgX)S^gMCo2@S8<i%#+hVOvB|NCN?3nnty~29sS!d5FH^9+*6{<A#CWyc6}bw=ytH8
znuAvz9^mjq?M50lJH9n5gj>{DwsLq18bfxDh(C%v{D5jyaEea|Yh*fi33AG;*|__&
zw?Waou3~Io)Ems7M(!F7D%@<eo31p^f9i*LxO;{yC$!t4745urmnr9a^64H~A0TC@
zJ4W-n@chNwe4{7MA3u>q`N}&Fy!EhHkNbKadyso4uP5r|6-g`fjfbClpAY`r-!vNJ
zl#m{~U>;Y=nM>_FZk_wBT0!0+0yJh!sCvG5dRg)JDa7%+sf2|HHFH=!zk^WN8bPFP
z4CaE5zP9>;X=nnyd+T7gD3`cg*n4hKwMEA)Jz$=m(+#^fkewATN#azXmE@q#{>#a{
zkDP)KibKdZfWV9jH1|Zgy^Sy2OMSk_A&@!FH13I6s>EcwwdS38&UvBTm?`mmeAHry
zO$rpRsIuc|z;}4U2VE5ris_@-{|9gwAdQ0oolkKSeL8j@j5kfIJHSAS;zT^%7<;TX
z)C&Jo{4+Cob^wn<Cg;zlSU8P#PH0a<54$}dYj6vP0vL=p5M4C*J&$Oc7$g9Y_n)C}
zt?ad-OQx)MT<e8bd*zc$=cx&rGU|hgQZk)ze_D803q55%dbGh`h|rtq(j4_8t$90f
z8>!wb!qd~0%eC^s<Ibz=4hU&EQ=Y#d?H?HV(CK6hrrM|qPus>??Ow~en%)PN^~8&}
z-{GTI8zRcxd*7tK{SmKT{W8y<Klj$wG<ZR7K;iJT1;l^vk)-ZvkH~-<_CGkgV!9?v
zMss$D1fLxdJ0mVT9vhOFcOXV>&LQ2TW-}yo)WNZce~b==&+&RAFpv*n_QSQOLJ!5=
z8f^!pSQyAdVnOjas61U5`#5|~G%|Fj8NyMuEv6*tX&40u;r@9U^#aFgPhf^8{X<8j
zY@2C2n4Ajvv@Y%)dzEgpynV6~1MlC5l!e-bYwcWXC#5K4v|jR(ta5xU#m5*MEpL*2
zoXr4DrDRgF{|r=xT5wOp{}(M$r{@o>%Lyr~x8d#NWi`(~pONLBo-?U;`o5V_l1yIj
zc<;*#Z@+ax+h$}Ubtm7yAf4`QCA(Dt=hHKmy9d}_p^{jC;BB6N^&LL=mEWZGKITiL
zWMJpTU18a~(R(s^7eZn}UJS%dDVh`C4&VR|3|%tzXzpjEd?3QNQzpIWRK<qOh);}6
zIyfa5_hMh))lynXj196o@{jK~xta3Syc#$vj&n{{5pF0VbomH7=ERt~n6sxiMhM)-
zJtt>p98r{mTz`>Yz3yH?uk@5isSvdT$vyc<=?GZl5FJD_BcJLrVBp-V%t!ACLxF}R
z9AR?t)SEKEl{jz%`+D_G4%hScnGrVj62<_D!x8ZP9R^>Af}RnsgLgM%0@rOZ6pcf)
z;x;nyIymmpa_7$&G8{%DsSPIzrdU5t1hw$%T@ptxH~>TpM{0*_se=6d@1nhPJ|)iQ
zOubaDweeJySIwq_n)8)SFXYTp63<SByHn=UY>!IGRt(%aQX=(dEQ2X!ax%*9wl|)(
zD^Gjp)2;FHva#(~-hb!I=)-gC3b}jY)#EEtD&&XfY_DGNoyQCNY2kEkyyf?lmj%>}
z<6938y!-Cky!?$1=%4%!XBurWN3GVKHLb&opE}||05_y<+A&4Qxp;eE@){5mCSyru
z)KHXMXp5q)60>SF6j!XFh}UBzQtvc5=r@j?FzKeH^<a$08T&l1H;)Xs5J!)={VaLT
z#DCW@WVktU3C~|JT0b}L;hotmcg)C3qy>DDdw2ro`B_cCED&>xVVlgM-&{cJ6y&&9
z!M&}+0SYh@(S#&Il{W6r!u*D^GDcl%m6~=B9Xm9&Dt&Q-A&Et-Jtig?M&U?oe55#%
zhxe={aVmuAMfZ9CKBPlZ=59Tc%Zf@tQlm9Uc{Sy@+#%%z>LH`NK;L}tAeRhUX!VI+
zck=lG%8EUq5itc9cg`6p3(_{ymxLET@>RAEUs9j0Bdn{c$hwe3cxc^t1xEYmF<g64
zpmER+K=LSTi8S_>Q1W9P+=~%R-u!!Y{J+kx?0~p&QL6*6k;9QjI)F~X_=2SH%%P5`
z5D25+YaBpJPx-u#9@EJfC_g$}qMa_ht~k&sF$ruC@5nI#ab7Yi%H{GVr}YokD=Mee
z$*44SQeMf?6)LVykF&xLMl00dEb{Br{epM_DICwax95y6cDWgBkP~D_PBnlhc)_T9
zrJ?-on(EA*I=u3b<0Q!lf`-nLQ0eYrx_%@;f(fShgwF8ZBxM^0mWrjP=4v3w@CwJz
zTz>k{XUTXl7>_*+yNcl_D{S@m=S-w9W-4OVRmJ*rRrHieKk~ML3}*st&1Su9T?Yvp
zBnMVHW^R;hRP0GIYe{_i@;b(SN9{&y-N?(fohPh7w{_v;OXcO(cy+0i#U{0;`U?H%
zCDrDWbNAwH^SG0x8m=#Qo@(Qhr$_D{SMEEMbs?uj>yS$3{V%`AtINjqawTVY_TpXk
z-GRV?{W2ZaB;_=8bVQp1qy;7R1d0)<GEh4P?xloaL+vzWTC3c!n#h#-a1<BUrN_Nj
z#3Mk5Nc~aw9c1V8yxy>8z`P=B`Jaf%yq<vu?Bd-keNzhi;>LP_Zz4PxnX>=^_A9T@
z9FUX)h;UEh<|!?GQId_0tRAbeFr4S(H9*B=jT)G4#MA6OZLb!d^zSPPtMV7?91u0~
znYe|)!WdlZBIJb68aU)Iz1Cpgn+dYa=do?<poh`J{`0KX$&uBn3)XMG!|4aV49-mu
z3Pyn@Czj=oE=Fo29^L5O4snFQ%d(Pq021UCq|Cg0%1B;Z<DC%6BvH`D`44}Mw|@3F
zdDXjDFWcTpk(slE@yt_*Bg8)ytgU_=1jgD$|1=@w!GAD>C-m7&4a7C&5P~1@LA^+=
z+WdrC9g!cS7VdSKfs!G6UArqetNON4#J+ESkt9)|IRTt@<73J>S?Vc;x7CM|5HTY)
zykkGmqDEqM&Y5y2JXHAYkG}WW{Dc0AXvzqHi@cUkod$pI+rhhWEh{PBBjEkoGbWCW
zN4G#TH$Wi1GjPA3uSh3^KJp*}eqkcIFm)D;39`)hg%{`A+vOm1ieFcCc<i`LM+_kT
zGiF6+S&iN;2drUyW)7R7!w!#T&lBe%&XS~YhO7h33tFQmcE@@E0RR9=L_t*D?X~@i
zGaLdcs^qUdqkQu@B^%)ChYN<@=sgCB;jQ0p8?e^Fr>*m}?L2L}ajWEvNMhS<ta$1T
zmNVC_^0IY4dEB@*Woy-JLn$nGZ*%QCiH;Tn=Ve(~*E5m}ulB3eMoH)KTKV|ZlaZBe
zl~f9+b>&<Z)>GjJzyJIBd*A*Y_CN8A{x@Ede*GhBRrfP`Y%d&g=x22t{##MF1p=`z
zP8|)PIU^~Pdj_#|i8|f0D~9>Ciq9&sf3{-;z~x2#nzQ@kC^NU9`geQ15f~_028y+b
z(mDhz3TiqG>P*o?eh)@iH&!iqSky4G8Pe<!xoM|YrB$Uxjle-+PIMgyT79gnP_iju
z(ZfM=<viLN0|nni4*cG_4U_1-0m51m%MvrHt<OKcx4B269KyPB_$5Nw2cjZ3pnG*N
zF!6(V6s;rDz(GiK=?8mN@qF{$TMv|nWo)==M<mhHLOH)>{j&S+3RuBgfRJR=&Dx<9
z(sIVe9jEckBI&?i9!epZl2RZgmeYy#bmFR)Q4n|x-S<NmN3?KpShu0v{Xm7G*Te5(
zV{WUN=V624h>h3O<8feZkE2RieQzT&fE)BMMqQ!m{HE^z834^dGQXbif}tUK4sbMz
zi2}65cG`I`)<HRd0VP7)ai7s^ocy6e`79+Ffm8W>p}s$;uXR~Qp&-KG7#OWnK&+_A
zse>&6kvxp%@g+m{4`){+%`)X=Vg>Q4?YNeo8>15){#5kFn~J&hvi!m$a;YNZh$_ht
zH_$_U3;`<q^B!4E7&uoo%4~IT6fp%3%3O64yB%K&1AyMtw~UBdc&~HP;(Q3kVU!q-
zh0N4gteIfO^}qTJa#B|75vMgHOIw%Bd9hld06tN>Wfk|vJ+p0<S5G^y9yhD4!F0BE
zYr|FT^pbe>cx7vi$Lns>2r*fCIGtFQ0$IuBgwm}DdfRAQrCoPU=QDNN41`FUby%fX
z7m)M2NbTfBxI3+U-}k-GZ{2@F{?~ts{J;4bo__g5>jpY>&du7U11LOzq0t5O+Uw?L
za)`c2Fa#ckB27auc6ScEOB4UScA=?HnE>D*AM*Oa`5vz~0|P1fpcZer5FLsui=_ja
z&K9Z^MSC%>7=etn?uZvsIpZ7dbq5%z+S&KUwwV__z0HZz><2SKk&{(%M>m0Ok}bR`
z;!yN+u>@uxuP}y<Sbh*!+IBp%KWFi4)vi=AVIN^kxec`iU251i?MknWy~c#Z31o_x
zVM>bS9GQmH`jCyRMiBaSV}E)goj}v#OOX(-A!vIg=giaP$y$0+LYj4Q1nM;-nnoBf
ziUO09Iavb47YQ}c&>MX78TnkuFJn}4kOjHo@~)2vGx|39pMHuWe>oyF-$Uv0y70%k
z`hJZd&-5aPGMAZu*HKKWz6N?c%iw$8u<w*)-)ZgarW~8Mo7#O@cL%D@ztMeW#gwm*
zqm+enG0&(RU<5GQ=~9GNgK5j3#focnFu{FMU(v;)DM(n-I2+vob$C-2B?;Z)VAju9
zOgt21g20hWy-gNH7-@h)|B-kbX*dC4IFEHCBV{xQ%;GEOd%IV`k5*BvkhHhzbtHX!
z2k{i+QNm*jdC(a>VE)d$J;4Sz1qhz9paD7}^siQk__~KU05=w?LIaSckwzHALj#1z
zHI)99J5GP(eVY%1MQB^=)MjjecT3^rwOM<%;B`+1l-6QkYuhVzZ`_|2E*17}-AuU{
zMRz9ATjNqUp7us<-S10TlknE6yz|To61OHiH8F<0YVML~W4}B?xBFO2=K1}JySqD<
zWg$zVBh+ruiLP*3I?Gyk|4T3UXwB^Z`cF~+?>~?H%!da<9_Nwa)VSac5HJlj1jvui
z=pl_67-Hxv+*wR%MAkz$3I+tMn+&?9;mkzDGDd+9Z1TE&`<y@k-n4))%>xu8Mzg@0
zuqlK14XYW;WtrnxJ;jQ3AK20)Zy8HIBS4sx#a)$_GBsCf$sVSS73VW;g6g^87%{5?
zHo?*uuZj;gs@+=Og5fTJO&Gnd9`g51FUeZOPD|pP6H87ubSDYBSaDP9)glq1p69hg
z8(?U7sV)q)bv7T{Y+iU`$>NcR<hf@L4f(nSDb3jWhhi;o;q9u9@ZyKx`H;)yO7@=o
z`-gk-{T;PdQm<CAo3p*&#ZJ-51%VnBXOQ&ZSw%6$*1!IMEZN&ckKW6i9q)Q*KJ;Mn
z3Wa+xZUF|y`;T7PZKz9NBv>^-fLMLO5m1b)DbD&BhtSeH%_Bq|%X4NNNQkg^YYPp^
zVZ^?>di{!~<WU8?PDx_Z5dFOoevW8I5@A^umL*Y8vPNf09AR5W5`*<_tczFIPl}i?
z>MIna8|W}V1k|3P_EAC2!N|ezoI>IHheuaVoH8@I;UQ^U(j5**<^EsXLsX8(wT{t&
z8MV?G5$LnNDCQifcO6;!$>k#)gOse}DS!a;zqmc(d3$s3+;`o9KtwW!0~BXXy_===
z@$aJ-Ie?_v{m1|TQ5+fH?+YsKSitzM^he*PuU2alb=cME!OdtZ_os|vwM*Xg#d)=q
z&0ZVZZoDJ8B%3stg)W&zlv5EhjhbMuW;dR88){@a*ma~>OZm1Nv0AI$#{Om~DO>&1
zX$5WEomL+1?>L{&AWm4;I(t{BiLKhKgS9Oz>&iRN@995w$N3xY@brKEdA9%X7e7N{
z7DrZH5Oql0@%7y-uzhIw{KK4uEc(DMPhlHA?OxbAzg3ce*E-qkoMUluoC%rf#_w@A
z-|h8=vk4|rx5`y3Zpy5N@YVacF(9zyWEJ!T>b3%kK`^5S19UL#f=vd~lwiro`QUYq
z=Z%mw^UxCoKg#`Dczz_`@VtHRKGw4<lS1<7PeehZ9hBl@X!D4Q1j$6=svBLKg|ed0
zSr`f@fW{nzWY^#OtZ7JK>y&4qp;gjO)i%5)#$%HH*w@ThyVrpvT30^)=tF+_SAU&U
z7Nc5DiMv-X`Nr4YXUz-y<rP}5$lX0EUd*9$$du`)s9UhTOmw90(06-As~qwOz$A;A
zvMpYWHNCN@)Sh12YX@j}W|0y)LhVqZuBgU$zg}H?sQew@8EFpRXWxU4<u;p&wZ>{5
zQKV~E)ENYmLQW0>G_GOkp(y^$91P3EtMm@F+kss@4S=%}<^!OKmqloT2`-9t-vHYG
z0e)$(4DZyv=`!{t_7~T9Q+AGRfy7&eGG?S1!j4TPtyW<zk-9Srrs2)Gavo>%yQp<l
ztM7@zuw#c(47vhvG7Bsef0%C)_vW%;9VCiP^YjUoC}{WM;aG<b(uYjr`Z3Zc>V|e8
zZU+VymGJJJLki}$B0j*@A?+jw4^k$O9^S|~lmF1W2%VCIrI^mBHj2br_fq0+QCiFF
zDX}*lY2t0$NFuCDVejzjX=87&ESZN@c<Wqvc9+?=#M2e>vrI?%_D4@9CkYD1eEN7X
zUANVSighYvH+*YJXf7Tp$ed0qy>%|v-R+V>UQlcleBJi(`<$P6c4s^Y&!63M_pkf}
zpX9{TfAI74CSzTuTX&QdPj~9=Zw70v@7XbRBZ5VJRY*}(*Aq&0oi#zJ+<dPd=+LXz
z@Y&et8LfX!yM9g;^5)$HDMjt68LrbpUv0=vceI>SqLe}jE`Pt99;*-r$mx6|YJklA
zQEe6u35wVOsF{MBQnI3pSSOhWFc!m)<pI^XrxCRp=ukF6+!G<_r~(%xA~!d}kV_|b
zD7hWHipU*~$_a3O>e@R``_AK5*_uZM#76d}Fq9#3$7sSKaxs3I5@k7|l95yp652L?
z^V`44%U6%2vQU<SC_H}hF)z+5@4Wjqd)?VCSJHAqnzcr%PeVyb$PNrNkcTR&JH1|@
zUd`xrW#2Zo?Mg%x0P6*Jb~vx$8%o>OWjy5E@1Zzxf|FSL<6h6gE*V%?$M+vYcjBJf
zar`GRh>w5~fk~oP=;|B+;UTrITzrm11{OEu0oH<J0JTc+NFNEfLZ5)aSTLQ)$i2tI
z#xf!ZOfM|vB9ItE!9G8}|AYCe`#v^RG><NTX_)=ghWGDGrDIgI8%>G7`)lT4nP7kb
zj1F>fCuMZ=FuDce^A@2yL<dABVOgZ}#zmuWOKygnMOwg|5;;Wo4g#&ysMxvoXlt5t
zE>wgbdMFhnmmA|9T^GaiI~sFldM;9*$1@^HJ|jvN(Hid>`(S{!Bkw(9dGFrpS2CQI
zOv;(d1@^0j+fz!cOXk#+{kj|PLATnd9pSt#UMmC7P7BY@_8ia73(wAlBueU?WnEZL
zD<8jn;$h7;<qxHJ0qW&?b(*FG+1B~A7$_(?aowRMH;BEntFqTl&e_@-_eMof@0NPn
zjDIC3q3)F=HVS<`pSb@Qev~Z2<DdJxHgn}5b2FZEMDtTVX5x6t>(EnTKp_cEnrn_{
zqa)oNyitsPtN{pp+z*Ve1#;+~G}7F0pKtPd;}OQ$`{4}?Kq_T~^`h;icr8E_2by3R
z!qupL$5R3*z3o<&?v?bpb&~k(fFuJ9*aI@cjdl>`q(AvChi{6wZ;2jqg+ufs6wOd9
z#{nU+us5DTv(e3}&N1%aXJbgy5Qr!qQcTJ3UAZ>r^Y<u&MD=<g{rShZeyEVRZ7D09
z9-PxU`&q7(rEqzA9Ka@vwWX*UwXlk?obE`=%H{IJ-TgCat2S_`+k~qm&igL{TG4t%
z>&1A;%L)j)cG|XM{rhP|bnJE`M4B9ADME6OXFd(gYFHf)=0A_B@nfsmVGRcMk}33a
zrC^5eLL4mY++7)4ai*npA`MZB(0sh8ai)7Xu3H))flmVKWFo9>l(Pf$yX-VK=xY)R
z2C`D?BN8<Ls7@(_z^dxrGXq@BvfMA`OM4^Q{Zwt{gZ4Hei9!1tbr-?iAjVE+91x<D
zsOxYR<G4#Ygu0p1GSs#!@feFO&XN~7eF<aixa2lw;4<r8cNpAAlbPC)aUani9?r=P
zpu4dlk*eqpmZxq*668F-M~+5FoT$h_5*|fJ-Ze15hD5i`F@?ssXAbu#^n>2<x{556
zKlv?mwYKl#JQ%H2>R#Dvr8Om&NZn>c3|w3HSQ^on?8bP>%DUK$f&04!)ClI5WwDlL
zv10YLKx?+<*@uW(B+?wlR(FzEv?61X+aeaBPEP6*1~{d})~r^mi<r(#o(^n{9oV*w
zwPe!eihS~t<+O16r@q16kG;#&zx&td-+7{UxdEfGo)e4>@#F{@j`9KFY(;|1!GmC_
z+h)~ErW<q=Eyu+acXpz-z;#LwZVN-dejhqWw0^JG8>b1Thr}4V5Mhj5WOy|a3?Fer
zUJV(>bL7)8{^}%+9&LFkaVo_MRH{wl%4POuk5!376AK-E`r~fHF!X8M#;*^CRT=GQ
z1I@W-5ng>1gNZX!HT6pKuA_!v^c;H_6c6WE@t=8>>r!}ezjC)`JWWi_nbV=Zf>1QX
zp}l=p6xMQrvYLklw)v8H@!~C(@@@9LQoGVbc=7x>=es*<wdcttQTNVAAAM|tdHPPz
zg;a`@-uN@O9jzPePxSp_ATSAfpDF+1--rIsKN^HCabu=@EL9_TaaE9TJNU8v^2OIE
z0nkGzcj`;+-3;r^VGHAb01HRw`8DS!B0x7FUM$yNd*s!uCweV~BHKtsjJlXn<kTI`
zCu1Y@gV(UAK!j(L2&dvhtJJ_vNs$K+u8Y@heJBfibycK`7c+jbUx)2>2dC5fZ76&C
zr5c-W(SRW4c`f`NyBrkJLICS&i08ZFx_aP4@hBGF<~Z)dnG)mP2H2DXDcbCu2}Boi
zG#n8P7vxyYK)mZl&S%8S*JL;}G98QF{2m$LQynbLQM>cm#htC=g`ze`A`%*r6{f=@
z;1%qUzxjgllV2f=I^}vX+Feg~Y1|NTKLV2Iy@JN%PGzr^Wy##%oiJTygW>kw>aKFK
zL2oHr#AwOF!`%rH`<|c?N1XJly0I))$F(eFv`m)bh0PAAErn|1(D&NVl>A&KTC-ts
zNrc)Py(#Nj$iVh=Mec<4hu-4u-}u8k{jdJN=<htuCB2EJ96;*v&jbE5KsOIVOl%CX
z7*4D2eaIrJgHIrNr&pi3gCVzR6e7M>Ve~pu`5vz~0s|3_$=So_-nKDNx*NC1uNvAl
zZ=kkD>vb4q%?V2J4g__iS0ip>UUN=l&6cB9ENm*r>fN~zGg25hOXmUe!xM*LgcW0<
zB*O4>TaTb%Di_l<;@-frWT#s$oU6@Au+&Htv(0v+l*EKAC+^lVq6%@r!Mc_F5c@qH
zBR#52R>D-kD|W+EK!m&dGvD|1_xSibA939)sbs$IYhN+W2Hzu9`1ZGdi_7Jax8J&>
zEDL#ANo6r_s8_2Gs2BRa(d*_TcM~a}Xz4`z$*+@t@~dor<=c$b(lMics_L$gbP{+3
ziaox@#6B<7n^7@!VyDmpa`0U2{f9Su@ndY>Bw{&it<xt9n$1`v-C8BRl@pZv3@jF)
zOc)SYd`5x<u<?a-$S@4uG;E!eVIkjK-;{vKZlC&NeBUqzM^Em^X7>esMLot~g2aq?
zq~&6c37COXi7p;L)i<8C%vR<@)@2&gSr|9YdpOh@y4b?8x3V}5ZwOtycW&P_-=ps%
zh5i{@W9SZR9gm=K+^c<loM)ZiJ3lu8hA)acT1V6is@A;_DO$rN0U~sGkyFOTuMS{(
zIzjOs^?4upANxKyXO<$Y&dSQZ(e|CErzc|=4a4R}C`jB&vU)2)Syv<S>AiYgl&8`<
zXl<PCPc~-U!9xJ>Jtu1GL+|G->|13ig#s+a82?(W15_zWZ9cc=0L*IBA#11}x;|`w
zYO_?*x~!b;?yNAl*j$DDW6#O|#vkL=zx5ZWPgh_3V;Y9f0)dcVu{DsRK8mG2M7W0X
zib!Cj$}k;(FjDYQHxT?b#}bgRUi!LM^n1MC2n;;jomiLRY<ZFEaxb#$7={zbKNK4h
zsK^%IiUkw3SCC9Di*d#$BPxpT3I)+z(6Z!-cEi#f9yY>&p-jUN8(IxfnMm#k3g&Nb
zy^n#G50MA^uaHL^(Ke=aCFktDp_U>U)6T%xEt8k*k%Pm%6Jhr6!T@rR6?L>22zq!X
zsv|G<rFUO&%EGHxkDO174ItVsq?FmOS3ddpQ=TqYzWV-`$R)F^rx6)JEnO4fD3wH#
z0*V2KT#(Z<o_+Ie(*O7w{6GID+rRhM=}-HNyc_^bJc?=*i;hhhNZ%KHd_sh|=I#a!
zqmU*RKj_|JAc7WFRgI}J5Vph~C~0_gjnu5&C<};oGM)P@oKA%pgIlstvJ=_(O!2{6
zJ`V3e&cSJ#Wj?!1BOhLGIfRjG6l8L^e>}`8LhCbU`9*x?k{s-b{~4Z+7v<(W&r0uT
z@{0T@SXECm!Q4|gc47|l6<17zcV%)jvgod?WxO{)w}hM>K&OB65zS4dJ<cqnKsmP8
zx(;whiJ{~5^{+rU*3yj0o?es~vH&<h&>Oirb`yAw^`WrZ-J<CD38!a7r6vHGVX6s8
z(l?(|e)6lFvv9Xqjg9ukUU#%N2L^pON{S3xbpet#PQ{#`lAyQ7dRjQ27J9eIl(pGt
zVaw-NB1M*SCaD4S-i(_i+BgRdx){}a@z(X)?P>Pb4--iz=m|o!a4LFhT(3Ups1)n?
zIo)x3c+TCkXRIX~C|L>_;r=HdSpWG?_|E_3uTi%>{L{m^9&l>dpYa1`ZA<``Ap&~)
zbCG1^S^ZXEQphGkF9vw5$kk}qZ{q>Jw{K8@{6=8ld|Fvcu+sS-;{KC>h__>PEFxg1
zfoPwz(5j9w7yfz6#fs&>Jx{Bx(zSrjx$nJO5!z6wzS5&sz+bNykA;zq{MqmlLV5U&
z7&xB?WWa{-Ug9&c)~|)*B9$<xNmH4qZ+N*fa`KUseBEc5>lXJu-x10|d-Fp*!-I&i
z3f@0FqvXQ*?hb90(^|-Mv~D~-zCx1l{@1?D-Q5YvR-h!{Qy6!WT0T-+MYP#9R4p1&
zPC#c_3NOC(EzbYSm-+a`b1wh(Uqqj_0dV3#jI|{~!V*ssJ2TcUHgV8H<LdU%;%K6v
zV=aaprdu%3d%qb+8j8?D%I^JS5OF3o^$ru1C5w+goibW-aE;G&SJw#Xr6Ds`r@<I<
z6-qY>Y1IL9>Sh3gnJ=MK_lB*}B)B)C(C7^y{vX7b@68a@qY*u%*ievU6uBnj;?C+M
zVzIM3qh}CD#<YLPngk#pdxTxr0b~^U9-}q^AZ~Ex@iF2(0MV(y;p<(NEOB&HQzi>D
z3O2rNpoHmF%5*~9fVhda=RA0U1DFQ*I6O-%NpxGu&66C?n9fu|{7MDVpZXShGTDB*
zT+mi1Sw>o-wKjBb@3UBQ$(dd&YfhYt<^DwycjuLLDYV`wInxv_*NwN%3%wZ=pQuH9
z_S$&7b{?NLYPVu_ZQbX(2<x&?mc*JZwYgVitCh#ArANCO4`nINn41$>qu#^sxkXKm
zpJ@{1Y?0E4q^by8waD~xTDbo+KS}$0ulVpU{JcdL0`Ps8m6h=iHTl&Tpry=0DKD&!
z%^Z4e>Pn5Mg0FQJb50`i*?U6{LLoPI0eZ9V-~07ubi{{)3lV<P(MlKc6bQ&plu?EX
zmNQk!)3~UG(OUP=jTrK^c1s1VH^@CIC}#_o5<(xrf*EG^#YmwH!fTT;=w~e65g3RJ
zn%WuNu&>j|#W>;~z`*Zm7EIct!hy<}%#0m77zsJ>O6&ASZqgq?z~cW1a3cawSCeKb
zF!sX=c#uq4SDr0aoYPvRZWpw7?(R>dys)k(Yd=iZYBNx+Dq1hpdS$;nA!@ZCX<3lG
z(DRC<)j17>)58OA|I<IsCn($h{x8!$dK&Fjao<xEr(vGPO>X{(DE#ou0!SR*!{L5G
zIij>tlo!f2_Fbk(#u{3kM?%#?_u_k+bEGR~7&~f0Qj%dIx+~h7K_Fs9o46rU_xhq5
zyd#5PC?}VhhytYG>hMd1+pDh=Hh(c+Gkl)|&W$rUUF!H_m|3RvoQjg+034NPoc-t;
zh}2&E-p*S(eC{awGm0c*e<OrH<as=c4Dc7vF(W|Op2b;y%~*_U;o!&DVkmcY{Wpd*
z9ms%=rskj<C-^?xW$3Mgm->2&PL&%&V_-wb|HwB;Klz?@e0AaO?nFsu$ZSAl44CTd
z+fGWElnbCN-ePmN6jHYI-Qrv_t<?e4UVi#GA{haAv?~>5tBt3vvo)q}S93y6YvH^U
z?>K?Sr;Vqr@mM>xNBSv|v+ZxllQ{{djV(%Gd(>JZ<w9CkpVeT*M0+zMn{&1rpRHQ4
z;yVvKd;Wq?KmHK3Ps)&^*fLd=M(vAphTM<}4axSpfH?cUzYz%*qU%m>hndB=^@w~a
z&Q~z^{xib&Jzj6vP2dM8LLzZaO``mU6J7e@VIvF`6PCx;3|dxgEx@s%PJf7W@Fe1(
zG#`066tl!32n9NE(Ol}+p5rq=lbip7#~VLH0WKoM6rsnXg^HieK}pVCnBYj3GLil?
zQqyzBi(UfB!*Obw;Es3i=wJLb^iTgdDHmp_TCLbh@YXG#3@8aH3pVZ1>Jw5~Nac*A
zOy76<_R3iF_6zOu$i8j#eP`b{-g@yCcgxv~W!r51x{XtY#OdzLi+}2;c=f~Yu>CuK
ziTbyGa~SopHUpLWI;0tjjp(rHGL66Zrw>J<@sse#RT1*R)3S}1(U3j2NXHk5%c>I%
zY5YD73FtA$;UH(Dy9B0U)v3Tmgj|Y+PQB83RK#;wi)6$s(!r+_At)(3Evf1_&=KD3
zy_5Be-Aw>-EId~~Rz=3%%_tho=ITIpVcg>*hfNRQ$DvJ=TLXW4xj2spp#8eW93cGm
zS8}+fs2p_8RPB+1_^d8^{Sq?dDVXF!Lz+Y7g@I}613f->tj+6Qy&h8(0rGF(cJqPp
zkjWYt{?Tue|D_*C)=Uv)Ekf@V$*Xzc9;QDPp;oAVORCDLWDDajnVgK6sN_WIN^-|&
zJ)L;EZah9+dHdl+PF6c$+zOeO*UGDF=d!m^SF)5s_ZH|&?OeAjRjqBXcICRa2~biD
zhZ3}P5@~qx7M)dUZB|#%JGC~XD|y{nO17f)Ay}Pyxfr2SZ`4*<e)NvzKdz))sM|Iy
z_y~j^#n@+2i2vO7AkZ63xN<1SB+^d;$qD&n7RiE8Qbw}HpCZ~FM1bn>p$-D;&xZ!y
zu$#bcbPC#}9`<E0#WwOC0TSGx<A~Q6!qnga>E5t*F%TpQl8>DPamZ(w?<N``?up4P
z7U6I$2?aC|h=1sCWNjnT%{8a;nFpOv+#@;?;cy?95*|!!^wIkmN`sJ`uvLdOm~3H4
zL<kiwmGrp5+b26j?Lc^<u1M5SdDra4amx_0R95R@*2eYI5BcDCew&P<`v$$S<i!2M
z9S;u=l)DqMo=_BNSHD>wfVAABBAlK*hadR?^1qqc{}=xm_2<4b*YP0a@ys`cGybcQ
z|BQAsLa7h9Y_G1z<su?gGi3K!FcIPr3O;xpMxkCDSiy@s4@O@O5r#oPgGF&XT9i`q
z+858AW~E1{IFvJZCyn(Hhxl2pHKI!0=&|FOdbi@z&kOJW0Kei1tfkmVtpfOsqvw$K
zux4YAC5U%q7;7D(;U3^I%H|q+o;9W()`Kt%;Bb4tH-(1-kWAw`EPzZ<pd+GqIK%$s
z0rCgT8Dbj1Z`ueq2L~W*Yoz1aV`UEO>qe<}I+bxIxcLw_$8H!d`=0a@U#0xZKMw0+
zdZ4&tBvwD9YST1JN>l-H8<2Az-2m6CfrVOkAIqNTYRrVC6rP?gT(7%#C%EU`m3>$C
zzVq^F=jCS4(tMGsR`zOq04Zg*T4^2y2wL~SzvfzUZ+6CN-6%OzJT2FQ($|Z%qhPaj
z<?-o>pvcy}QMU`%%Z1h|r**YG_|B))U;0HJ?(aCAPB-fqw&)lc9fEqGy|#UnA)n&P
zUB@%y-B&KzuXWT!o3rTPsYFCZVd0S0?+O6F`|HiXfc9Ae4TJJCjGqagtKxx3CnO)m
zJ<LL#*^1+aQ|E%3Jx=2w1tG{lP2*M${l|{{#Y4ceP9PRirYq|24=pf`sW802V=wdg
zh#bh24y;mfuQGr~w0q1W{kp~pfti3N){$|I_)}ymJ^l78`mcNkp5Gg`?2TSGTHRqp
zH9W+f3y>L=w1`#3*9)&+z2uV*z5_Y&$;YoauZefwdf@K7l9s~hd~ZgsEPyMtUhO`5
zvqxI&9+vxOq#yVu*Z;vv{a^ki_`4s19@gI#+IUut0w{m#00=jR!Ii6ye=@_avw%w#
z5A`#Rfh*s5!!R7LACQn^|E?pu+->;x00Nnnyw(FRqZ`Cp*Ye?T#-Oe=49rMHc{#kg
zJSXxnbYQtoL2R1a7x^_X{}C|fah`9ZhH~hQmU*TF47<Jvxw?^=*Suuo8G6lN#^3a{
zw;yzw*bg8~@@IwS26BG(LFd_y&&rU)K0vDlt^1){BF_3uaoDxw@cqZu$~4|<qUBE*
zauqsrI4^TuhqDz0o$ua}|LGqlofbr3DVe*~ywIF*U}{5udfg2GXQA|D)&1GnJlB%h
zw@N96l!R^DS(aqQ$evPrxHk`cT^9CQxo$hCaNVoXweJQV_8!DtHWcaZwA$p+ZmBzM
zZNPo&XaJ#)=beDtQWOb`Tns>n9tu~Lw(V5yT&_=CA78OOKGIsXh~N1R1llK`aQXNn
z?$!lemyy{W#AdJI_5loh@1yWI<Sh+*;`b_vCXpDFX=dE{kf9%Rqz(qKP1e07G7}`P
zWyQbe>y5xbuZ^CWM&)ooOldkLZwMMfA4C>1jQer4=S1`~%=WtB4WUjt6-F$Q_9!5n
z9;fB+^N0>;C|z`bh%h`ctOF(hnM&f&5|@N@1Q7u^2t<DH8quN5QpzLLKNZ`wZ3Oy{
zgUi8gj*tQ4^T+6h{DohJAAOH5iK@->M|0<k_t^&PnBW52JAHrT>D8yyedDcnUqFYo
zZ#;iEvz|_*oLJVg<pIU693!B6Z$`S4GFqQ(kJ1X-Da(ocjjvMwNB<S{KmT>w|LyOh
zPy20D2X6j-6CMVy3XeVM3_kd{e^;b(SdZiP3^^EAn?~R!f;8O3gStHcmD#=;-|K}(
z4lKNB(DJG<qGVAWA54f-CKFbG7=^YvMv|JGAH0N{eFv#ua4-Pa;!dL*xnl7a(XLKb
zBQgVv9F^VNr`X2n_#O<NjQ<;Zc{GLtC*A;o@we9u{B#g*TzZ|#Hg0qP&?E3Uy5U21
z=PkN9n&(==z)!NFc~<72rU4wpJms7rrI<`;BuyN2!8CqMK$g<fPhYwt|7-sceCsV^
z^Gia~&b~FI2zfcoOF_L`#tQqEMTUMTIr$iHWAbcGDY;n9K(e_4tu^nLH{DdX&g0g&
zT$QJ7=Vk3&HshV?l8}<gCmY6g+1pT_+Gkgft%xb^sPH!$XJeLBZ8nF*>j3sz2LrAo
z;~S`T<JHSgD5cQ0E4}WP?!ydmcR>F1=YNqTiLx#u3IL4B)`GkiF?Z0XYBOIBQYIqI
zz9W0IJi6@~tvAH~qgCjA*frpgdJyo`YZ4uVe$thiKOY)+<1|5pRLnpIA;K)mIr8v_
zphn@HdC3eVdmLdpy#2VwQ0%X>c&R`4tPZy8#{F5rO$tFDp*tPdR1qKQVU?uD7^sI6
zs^h?bcxcRx{Q(e+;-vW#%;QjxLwl%i#IFu10S=pZR5Y;~?ors<MI_~RFXWReQm-S=
z+_rtJkEBH3O({hGeJF{2ztZ<BtyWIwJBq;d=@ZT=S(R=n-s+WHVU%(I=qz+o#e{o%
z1;siL==SIO_C5U{|9+M~^i}GA@~iZ}@@;tBMw;k`miJa)+H(^*;8~1#E7B7{=vW#t
zhD!jiofwPzn&AkqfRL+$$Hd`%@t%;;C@g&ay`OOhhze_lF&a?;ohh9oD1F^v-P}|9
zf`b8XD;;M6OjLcpMuDUTaEnOTe7nr89P2iyn|W;qgt>;kkGI8CuL1io$3v_l>gng)
z9U4`JV`29cH~HPJ@mjVv28x-+KIl8<J;wPBg&)Vx!D;jJC6aH}u#^d?T5Hg5i%yRK
z{HSx01Ydq2{dax>zV%{YjGUl|U_+8D$DXrLNW(Z<VPK(cjU?)Mecx{1b5REla{Ij7
zzOk0V^}68$9(md-A3kkt%|L{Tfso=;RBNsDTD^7)8eW8*vKKIBkHWc07Hb>str0`<
zHm4u9g-GV?wC3B>sQZ=P<2tacXHKURty}#<`J{6C*_XWY{`=&zjI$j&Dm?w+y{pyz
zMA~+qX)8P)-6g*JB2L_w1ZiG0@58I4heOr~>hy?=X^>%wrQes)27i>t-~IJw-2^#P
z%3{V`7==L5KOQugN3<j&)*_6$0awIBq$!)h4r!gnH#~V)2oZ${Phltt-}C?#NAJ^-
zZ*X|4&f-6jgytFE4FZ$!(82_auyN!7-ALQB2D%Q<^LXHHH{e)LaJ2l9&KN#!Xrp+)
z1FGru%(wu6{NktN51z<hz30_sr@eZ~<0qezOF@@~eZLxTD;aTm=yg4zx?5pQN-QNI
z`-QBPyk@77&K_9_fi{2@|Ec&;pJ*pa9qN+}PSTFH%}^|gynkZ-SAPusliy<hKl}~)
zFMVvnee5QX*VYOSYZdb+4kRzU_WqEA8!&lWe{YI!D(ev<RNz3v`rTd;i8ulal-Vlx
znl+kvx?3P}3=@z1Zh&jy=KeqP94&wE8~h@?%zGHRB0S8S|DX8I|8xL9_I9qzFr4u&
zImT1)W8~{+Ej;$#<30}=BP=d4oNL7V6jbyO0**b2NK%^ns2s?GjtT$=fx*Rs7l}Ia
z{?Hk5ZEvH-^h~4`dh4UO6=?&!q=ODjntc_LfscNOiuiLzN(==3(jDo){S)k8d*CGS
za6VgIE8X+!c7M5;w^jF@-i$biW#Q@iL=qzi!l$+>isVGE#`q^Gqk9`{f2h$K+nlty
z&fV9xwT6<&c`*auJH1xwcD43r?5s2qL^6SkaNXR3!dsPIt6xIc_Byy=P#b(?W83F2
zz23~&rkr{H?sH06xW4=p-R&NhUwXx}7hmFTnNwq>tAi37F+U66eQ%xUEP2!??BpCB
z%~<D%3`X=_`?MQFZUBaUb4#Tb6&jsQa+trOpHFwZ5g1t3JJ$915g8uMoc<*aSf&RR
z2P8RqY27G>Q=S_h-q6si4hSNrdNkz6#kS|3AY$BEv{(3onGu+UX(Y%T(+l>#0S>}z
z5h3H=9JhJn01p$$;K=Qe?->yXZoQ(R0$)#$93v(E_VIP|y#W@|M;r3B2Ws8neB#qz
z{SB^<7gG*Pp)U)kx6Uk@SyJJ0y^^H!?C!*xg~hpCmS^{fB=S;jQG9QxR&N`^fJH<t
zhLZ8;?7M-Q-q5x~Yt}K6GRymC*8j;*)BlU#q5Vg{ME|9a4@&R=faYQy)+DT{ABop3
z_c&CBz)_&T4jts?pT}G0rmCEIKQIrdh9Qx_C^vgHzU%O_={3G?aQvC|_M99!G%VHB
ziDX0+BYsg|A<y3p@YwIcEVz1F=)j-hznZltQv_y!^#CH`%xPTjC<nfeh7H3i-^kSL
z<~h_d2;M#71J;PxmEFu>yHuE|Prr?^&pz)0MJ0PUJArh(mdxZ$6_Jkgj!GMhr>!+(
zYV9%E(sfr#Wb#4Gd`3asdqg^4+Y=Q?fBO5;Z#<(2RsficQ$uTG-;HCWw}z<G#wTl=
z?ya)>V4p0|_6h>ax>8CqVj|4Q)mjlzWU&ZF$;D*bpEq*&BFf#mP|C`+_A&EBwX<I?
z^jgQdYcj7sj=dSqZv6bwqTBkIQLt@SiqFw$wGGe_pNr_JjV=eN(5tn!Ta6TG?d(7E
z>+s1-9=`M?5HE^PLJ*YYoJNYEdnb@N!X_6{vlSS@?3~%i);hU(^vs`Y!qr<KJa$tf
z$s|MOhaUAD!VMVwU0!bl26A!!`e~9};ib=bzmYdrRl0N}MYt!C9qA@Xh<l)ca{USY
zU^FMf>BCqUV_6=GHH@vbQ*gux-1K<B0cCSg_Vao(J`nGBMdihaL-hnO(c|~OW}wFR
zP6cKkO*i+bc-4PAkAMFRFx>R$9WEfyq5b7wC;!;nSQu^Rt#5plk3aZ`To%qRp7Zcf
zxKEv@%ayKh-8wH`yx^Q6_03qmmUVbJk_)I6<xCsx+zV8)WLKuyJsS^eX}5t!y&>9Z
z`##pyY8e{kbSD4Qm&yO&SLna~lJ=keI{hzy8=mGNzj?l+Hz#BDZ{~Z)-cKMYhozF?
zKN`<RDA3sWL!{+!=}@A9&~<o-y!NsXpzwOW`uG&-@H6oW;d!X1-7+}|UWx?HKkl2~
zUX!mc+{?jO81ow-e*j$)K!^h{ny-Pa?Tv>N5e~F&Q*p7i)6>^+w#e#6b9}+^ZVmHE
zcw~OR9@Q}?Md<CAHgX^k=LW2O)Gyc%d1nQmOc--a7<M}X{K{^#15^eWh&03WB*S3N
zZkYL;{@%F%6h=Z|$>e|XhspoQH^?dR{QgALeO<3R8MT^$eM7ZV^1`|-2uf|0dfgms
z!M<D3^17Y|4@dI)fZe_Cb+PUP6*JP)$R%pEh{yeTW$DV(4vosSHnc(Sl~#A(4|4`n
zE*5cc!`Esx-RiXr_<bf7v_sAXRU1YWat3%5;xJDpgbV`%IU7Ck)8ixcS3aTrwcq0I
zt+z=@-BFw^($HBPP-%y~H3xudw9|>PKE2zXM<>K6kZ+9<=uj&x6~*Ku(gZiMdORSW
zRvynMw{QJDUT?-TkOEEHf1x19aq`~{AAjgaxyg-0o1DpQ4=3ryXvP<cBSQ&g5|yuE
zC=SJC(>o01GQX)e%a?bbb8sB3eAMHLdKfcJ1J&j5+gT`f9H@5#Bm5A~780K#=Hxh$
zS{Xvu8f_>?=Kjuw8DJ<3UTj0&H~QX4>xs1OD3$vc&v^dM+bp^8ur##qK4hq{_l2Aj
zS+B+kQTqgRDMrhRg95+zG{K&Q=}aTrxZYxiyYKYvN^d(+^e2wqTdkH?^a*S%ne;<1
z$UpQp{ZIb@_22(l`Y(Kl^NG$yh2a^D^*_qUfdFW%$?Hh?sF;W6Kfgb&arnpl^=r|d
z00pCW#@{Hf$$CUgj$@O@%WuX2f!xTbQFTYC(0HA|>cH~<Kd=tr-tj<lGWN%Cl6!Sx
ztzBVgb-s0Y{l?WEp80ptIf57chU*gFr5<J%k_x?>?8kXM%85u2ssU;tcMT#Szn5@&
zx+7Ne%6)G!S&`63@zo(utvc<j{X4)60KHRnCyjJzq+h`}@pZJZDh3n|9t6;*GluEH
ztR6*p|1FmP=1-D-=xy?1biwQ5-z8$x{r27NueZUHxqFCTw3A9<zh0^CWiKUD7Nc9=
zwku1?w6?L;%94|{4ephkZ5Wu~>DR4$*MRV9+l`g5s*SL3(}Ndr22f8m<!sYZLobAZ
z7p!%)w`O_=wYoL6bDGn1#s>MF7^ueqX>U;wY+zwik5d1?e~mjWJios)T7Er(dfWeX
z$xa{MsjXVUWVMLOlFA_Hv#57-eg)GTCT3RmA<+@0>UCHL$6!F|!;g+YR^#1=_kE7n
zn{^X(&~#9ifOHO1!$F%4r8z^OR!QBP+$aF`4H*-+#-_?B5FG4Hgrdciz>%HJTfx+g
z>^KwA*9oKeM%8*<i!%g{c+J6>w?2pMOgCc!hrt6d!oCN=^7}O7aaaR66g^4^rFq<~
z)?twH0i^hnj4gfL&<R4{cSEOn<@D@1%iSHdR$CWs#$whgBGz7*&-Wy4R^QkYrJRg*
z*a3kRU*-u$dL5Lz+B>~%P=%a3+tWr~&vu)=QTGc``w$*WuNzA)G*K>(kDP&WIwN?Y
z=hvUH{`da~?JxcY_0RqSe0m`+ZHH44J#?;w0gm;Ll}O__F?J!W%waiSf5q+|zxx=W
znLj@adpUe^e%3!1En#!>7~xuCT@IvNGVh!4TK%8mvYr(B_|7lt%RzeNb3*~hl=o2P
zDXD|WV2nHLkyE(_`y6nyCAMT9xf{DXJ+JVnbu6?vG1&2EI$J%3xnFuP+Mxpu`o)}Y
znZ<zuG|z<-o;*`;XhWOl@Q2KWVY0su%_p;{7MVQthAq?`9WNL@E|qrDnvr+zN%v=#
zKly_!fBgH=dn3mx$rv+h9qx#w1+cc(uJ-rpK;U#*Ns?*%PAY}m90-)`hTB5wk`vpu
zfhw&xYHe1iU5zMcDVdebwpE^L=hfD^cxs@l8DYSS)u&BqZD+~C+s~hob0(LSk6vwD
zoSl}|*kdM%MoO-`6XWgB8j%YFcqv84nge|?2jIvoT7K~r{cnEAvxm2N{%{^3Ga|YX
z@hLe9JN>>r8WicxS&SMgLx*`8(dMuvYV~!;`X_Oqi9g#+tIGUd4B*I+ozDje-UtjF
z+q-V_5TVF)2!7Y;wMY1=AB6XqGwg1NZE>w)ze_L>+G54TuHnU}c-VvY5DLe8IW5BO
z$hoU~0hY@Ne(I*(jOM|}q}Zvk_#V0)mSf)21PVIzTTqnt)(7gdTkGSyD&oV2ELTJn
z`V2dCs2Lba%XtjspytW~xsb%C?Bzl#A^4rGUnpwrY%C{oIgzxZnh{pphqT8o&V5x6
zH^t2NR_U$M>V{r7+PY9L7c=ZJbKTd)_PG08`bO=IvYsI&+FnVyf<M0`g*B7^<k!i6
z=sos-?`P3p_y}n_fQ9XMD8o2FVMvIbcX@IKMu`yO+_LGXAB>R9!o(Y;?8`m@z?@ts
zGZr>+#*hUJ5D-bHkae9#E2cOGcmCmb^HfbhlCW^(7x^{Bc)Sj%2VD7iSCTfebVy1I
zU4joeQgc4+Z0q>$adr+eF}^lHSq!i7eT_nBQJv8~%#-ikJBYF%ac=to@E`PF$fI_p
zxw1}CC^=-7ur_We;HYClNHy{v{w&9J3$HRBOozi^a<AHuchAUOLDoY4N4|;NoyqUM
zMfv;RLvltI>$C6s2EAJPsb|(DkxQX<FZ@jc4fn_ml;^Uro=%`P_hY+&<W3O_(`U&j
z1{gm6^bvr&(?ZwIWvk;%T=vS|I$O0Gh^MXd>e|?SPJkezxV!c4wB4ERGXlCoNy7Ji
z<!#pW#5*sa_~7FUA3bhdTeX2oajsRV`##nh4ZfP0mnU?*+biol$Rmo=e)a>}|MqXN
z=)#M)pL1FY$7r`|HxFO?ps;C9$`<3dr-qtsq$z;m;Wy)GgpSjY?~plL#Tb9wiNgGY
z({q%cH&7dI+*%lvyzw#st3zqV@eCn8BK7f(!wm{n^%%d8LmBNaDP^Z}K2&^b95}y@
zCu!6R0+ad5v6iR`l5V3EIp?8{jt6th0v8Sf6c@z{6MaJ^(4Gvc-85uQJ%4o|n<0d*
z&mG2UF$}Fv&V%S$h>CTQX!VmD6j<(`q036`Ajyb<g5;+6#{ZAIfBTg!%hLO>-<X%R
zB4Y1zGPA0>s=8TyCn=F^HD!<@DAFb+>+&E$1`NR#3|OWuy%~P+AL75jupey6mSDpV
zwrtt31;GylC7F;ZnQV$`Qq3lteW|WXW}dTmM65M0!w+N3x#Cn)s!Rh7bUM1T&MkIa
z)|zvUF~0GA->4NME5$%s&bCfMoE55|NcW*q-6j@3&JI`Y)cXzTm2$tD{jqek6eLC0
z6#IwqG0fUKt?r(XHS&5!VkSrPr0=~UO&qx~Xq*t>>%#Is{$<*~_nWl;``@*;fq&6~
z+ssfs56TY@M|*+{`|rn}_>nbxyz;RY`Cua*e>oD(nZo~`*Bn1552v>rKWU$T8jt%u
zjd8ZYpie~MV0({r;;a9d$=1x;#H(OD(bfmPgrk9CjPvU1D*fO<_g5X8m=R}nUh8QB
zCgFp7fmmwSQD<Q5Jgdf;VKP8NWCDFA0+0FsxHsC{L<WvEIglUqzyWBW4>h`u_p~>b
zFzz=Dg~B+TW)OJvF;ONNk&jox&wic!=e|Su`L7W^dNSEtUTpZ=*bMy`0;j#R#>i<A
z&TA%yNU4=tt9y~y-t4WioHFOL4e$Z6DT1v-2;|ese!tJbK!QmC%}^*<;nfWZO)=QW
zVn17oMNpcr3p=w$(%|Q|SV9tgYUmiPe7#j9X3lTkkk^ISh4ZrV{g*4ZZ8xN3+$u*W
zv>$SE8-jTd#+!Dsw>i4Y&ReadzxxjUjb9<%JD+^xYdk%#h>C-#o(#0(*qANN{)lKF
z@r?OrFRG^`ew;?fju-!TQHwc*fcOxmhXW3yB{?Z^#7XIh7r!6+<NTLD{(qWH(42kA
z`~pMUIW|K$>{UNV4<35H4~5d96p#=8&Gyug@*ST2_A}uFLYK^?Xg_G1^f*R!zM(_P
zLx&k`(e>$_@YtewNL>8zd8jq#!7(+#dbChRrWwK@JU(8@BYB$1+&Hwn4?Yr%Lwfme
zjb@+{VqPFdAADqxV{~6j3Fba1VoCus;qVd5Q^w@7K@_A<gbG!9DbNe${%Ur(B9jLo
zLtcm=<{rY3caAL9df@h+X<4x;jvZ~7RcY^poNeK1w8Ka5#DDfDkw5$q^?&m>;d`$~
zK{fBi<9$8;N5^$SOqZ)pp7N7sEC+$ZyhT2&@^w7_;csNV$3sH<unz2SAL|>3<;FTc
zlq%yCkg;YS6Gv5sZON!yAHLzSZ@$Wpk?4B>Jiek`uK~x~urj+I5`wi!j?8|p&zYNH
zKc-_@<8bVJkB1vCgR>*kO8T*F@tEhSj(H?UYr$?qcqlZ-p4a)l#<+i+X(29~vmrkp
zw*CC9k>7idkK)JHPIyGa-ekfb{U-5GeVgzzpOC-(HEY@Ve(S0p)Q;dkS?<GtE!YG+
zYgM+H2$WLU%T7R9b6{OpD|WO~ig6Q+g|D=-8yad;fs7XrWR4Z?O+8)+#ajp|dBE+;
zUJJE!kDac?f2ed-tsp#~E|gMe#a^RT<>iN;vn)0a`SRwO)0@QGHy`oC&p+n}pMTE1
zR*p9G<DTZ>{IhHDv}QiOEQA#JjqiSDE#<%cL;An*pCPY>r{_1keL6wwXka1-lb50W
zfmrXn3P#&^$h3{jh+;_aJX?;oIy$&3GUcL!Kqz$gQE5{EGm?O@)<<$}YVbBa7{18k
zOHVC~@d!?SRT%~thrEsnfsQj%9R?izx$)s=raoPj9;4>i@G_nA9*)`h+>zbQgJI-j
zVI%?ia43jFU&BBs*p2Lu8}uP589Op?)MyU#`Sza&%!DJy$&?NQiVxtRaZUJei5??f
zAOCXn3qJ@wM|JDfeQ;*&Tu_Wr(w6Nq5s%}VX;-vYgA3K<fZ7|qY}OKJl|fM=m?UUk
zTX0CtK2eCtepb6#mHNtzDM3sqt}!Q0gA5fM26TdqjK#eZe(_BH*Z&0j|Ke{T-@hGI
zm+Ark&--JC@Km%g6aXG%#;qU!;v0@}wjQ_J5=lLh*N1xT;a)u^0*|Zp;l|G2&+O`h
zgu*V6qd*Kb!l;Oj_p|jC4#?>%`j~qZ=BA%*zk%{Rta9-KY{tWxK+A+z+pvrt!QJ;f
z3>V1X%Qb-q^f873fITFF<2N>LUA5c9I(Lj5RVotbYGynm;S6%1jxjt~kt;m5yf{8J
zqvbayU><#7Io9boy<#A|SxA5Or^)}+&m%wa3~3>Rj0Ty?3Tt=SyNrQZak#Adl)Hch
z!NH<$eHiaJd9OoCHf_+(7%R47h_qIxJTIm=fA)U8xmUjp?P4#KTKx5F7cDu_YNNMK
zo#u1qB@!HU4G$`%fkdB{6<Xu&Md;otuh)(E-B0=2ANn?@r)T!p!ugUopU+%U=DY7c
z<=yq__f?OHei8^N69b$gJf9Zco)<1lBt_wRJ@f8|H}3!bZ=mn)y!q%YU;p~Y2%S<3
zaakS;Ghc7+&KP7Uugu1IiAt4YKX*r5W=pa^w4WXlTsyDB;igD<sAi6&;@BU)h8T&b
zAp=1UspF4(eEDP`92vKGsrIAL^?-pMH~%9~V8W!2pYbsWKLi_wv`+Yd(D6YK6`8@#
zUh4PScpk5JWaAb)qHX%{@Yy~<FUHS~bH#(B2fInUd-zNbZ#kLxgZT>>iDkb3LkndF
zfw9O2Z{VTtI5I&$WJcgR5{@JIJ!-6F{*PD1+MpG*-q3PG+fJ{$kBL4kbX7)gKh1mm
zVVg)MiN?q|%@=*M1kTX}oRO>&5J`lrw6a-?p*LE;&x27!(bAEhTv-40KSBMsf0h25
zuj9l#u8Te>IFKU&Wgs4pL1yNl$3{F({)Zbk?nR$T*azw92Z8$Gy+26R9$sezxQDek
z5;2bwAH21FzHuLIZOOyh{;ED^8*nO9Kzvl0{eGw@EZ7GR)PoeK5BEI|w>5p78X0Cn
zU>3A??No7}6RVa!KujN#eH~;o|Ge{mypmQ(DYKl;<aH&4K&y@WedBh!QS^cQ%-ECm
z9^><?_M<8viQjPK#P#-ld>4)+@_9kN{f6|<{}J+^{AuLP1&zr*YFDZ%R9E?6*GCX~
ztMt}vr1n%m7(%ed0NSXvld~y|LJG9jsjU$tvIKQ<pt0b~woyuD$&T=Kn-rLrh3m`p
zNR)b~?7R63VEzS7?M&#c(OUKUC*H1AVlyPu3K~Cm)X0prHL?ijH*aXIvhM{A@c!Od
zV&&y~KcsDil+P@uGcjk2Iqq4pl|qW%zU+-9FMRaPZ?W_nC*7ven#tTd-}%NzeE#qJ
z2Di_I_%lDr>3o^%pbw)VMk3X%xDg&^02p!}V`M=PmxK;GcPql|v(ZgeGAzzVL1Jgu
zIIz48tsfSPrZ%*7-A4OqgxrCEjo14Ek1v-D%+GnqxEKL#Wb(dACnV^H_So3tb9EgS
ztcL^PfJi3vHvZv*>R}KVIX3e@>`Wh7_v62f)c+yi+i#CxD*kVSTon93IqZ({o1{~_
z{+Pr(04DR=z(@+b=Qk3S`R_9dFb=bESd3&~tQ27+MdNj&n{&tq*G(TzH61nUn3UHW
zy%nTxP&c&hXfJk<#b^f-mc<4cLr^>+87yuYm5+}~_Je0kF;nmH4H&+6GXR<E!O@v=
zzY|k(*vl)om<dtPwj<w)#Q(`J!T;qqXn*;)_+U`hLvnajfFB0s<A8ri2(|OL!j22g
zxaxBqj;UskIU@YX`W$n^=35`h-0_jdfqW#XqrEoBk&pYvps0tn`oJjuDn7*RV6AO6
zzU#<o)e}2U?t=Y~atwnSm(Y@dW6w1YNaOB_AHoOH;!N-1RWhC_GT*m1xBPxMV`k<U
zoWPkA=gWo5(_8Xtn*3fHIb9J|ZhEElWA`yXyH~~Y{^{J`^Qq&rOYqxn(mDC5H}FS3
zA^uZ8&-ydpK$n%OP-=%>oQGiWeSqmUXw~6x`+HpX1L5d2lqCl@k+bh2$?R({r%aG+
zp7<E_zc)+nbBdHw81>(>E^K?T<SoT%L}Z9xM57WzqPB*lL@Ax#h8jdEr64hbBPStP
zI}-`)rA`OJRtu+e;?3oWdn(jgXd--iyV13CeZBFuZ~Y|8Sy`3^&S%QgGqp52=GA|{
z-`Mt@lp>$Jd8XYPODV(@rlr2bOnATX^xJoqx6f1EFxpfW-W<`804W#%q(EE0|H)A#
z?PKVaB{xBXjY|Ju9e8oz_oKCDh=DCk#xz1-6~K_S8kfQ5F<36qe2|m+`k9<J|LMuA
z`vQ+I%^8S(qtqAJ2Qr?a`+h(GW^LtxFN_VLefBfPfTK}Gi65i`GP6DT5Og2U%^Bp6
ztX{R-rkcD4nS0xx0C<Kr0<xdxI$5SO#X1s-AfDYF*Fy#$;GsgX*U_1kjYMPU&O;nm
z;UNGz2f~6w^dLU+po4$$K|vb=>E}MCJouC!7sRi(I|q)S*m|QkYYj;62m$gLkta*c
zT1DD!Zi2lt4VT2fo4Sp&+94af)iQ&XSS4h<f<Q>g1@lg9^y*AJqmDT{bm;-NHS5d8
zHpz-mi2vdrBK+ZxsQ>$4L*H)??uYyy6lg-L4<7RKp%3fPKi5HSl_Tgp9@aClJp#yB
z`~Q$Ebgo-_DLDeX^8{wwK}TEXAx~AW8^@aR1W(&nJQ=_a`jotC9T2~$dj=fORB%_i
zK@=2E04(UoXa;66D0?~Xn~vn^@vP|HGQ;}yF-BsxJ<)onR)fgLQ8E%0#|aQIxHspO
zWj(X3D|uOLPp0N1-QrbtJ)j4%o_Y|W2PNQUIqsz}z2(Pu7XN-Bz<1sf{^_5i|MIt}
zOE&F4Q0^OupJg3$;geuR)`K#o_vY27B?Ha9L8jB5sEzZDFc#nWv>Yv9?bNbUTW8xV
z5&|hjOWILR>q2YQ359-AE7W2(#m-wJ1rSM~%H_16y&=6*ig9_82%FlR0z{19*d5_}
zNPZRNhu2pw=L>Jn7gVpP8frnW{P1??vMgM-jps?kRMykV)8#}ypNR;!Yvy*}Xxdql
z@NU~!WZ2>xa$9PJzxz4*>CGG;84_^2e%4xY7=Ff%NbA)G`PFIzj{2dI3!da1>W{$@
z7;?~&Bs>Ogb%LnfkKT_l^{p$_6M^nAOvl=ha2z+jsHZ#Tk9mAKHGu?!i$~=;;5%VD
zK~0~3IfD84hGCNQlqs+Bj0!dTO@sWCZKekp_>qxs<gW)ZpB~lkSPJ7H9g7J@b}XJu
zjA?P>5~v3xn`(6wd86VslG~^_x;6{&qe_<}F*4Jg$z>h&&K#=Z$IMK6W`GrO64Zku
zp9FKS@_;;&gc0oe-+zbxxg-Cw>woyai^%*pvlecXP$T(->N*c5tDtr}G@`o8mQKgV
z5wL@c@dkWqnQI%Nlri8ari?PKkCO{=LXjWP63j`Ap#)IVHV4-R$DGj+jN9<XFQmWp
zM=1a1e~x~(do}qGNFD<1{EHv??PO%nq~Nhy94qq3tHzk_agPTv^muJgEasp6o<EKp
zeyq;s&-l+hK;Ur9CLa>B*+2f#q~<I35HXXgIEn%%pm|z;+{Z2lkqrC(DTIgig8ezA
zjELC+PcH%yLW-oEW?<{BQ>zsmQH%^1J&4s>DW%Nd`2nZK6Uq?6e1F?=!^YOmmk=Y%
zX>~zUp9#G>vSHmSuR8xB$$+6-*1V4YwQmvs;?Gh(S$U0tvK97HseLfS#btg+8?-t(
zP+}tG<fJ}F`Q{mI!6b;Vq`;aJOEiO=oPxI%#+YqEB$J{gAl(y)oXz^zu4i=*0;QHl
z*{WHikGRk~DMYG5igq<pvfg!~6Ex8Yb`C_4qHNXK|6LL4O#BcbhD2$N>-CjyzI{VD
zpZHWex3cqexuAD=*><jXa~B*4^s<@?;{4_fsLfXhd8cV73A}lFqP=`dYn?SGG#Cl;
zUX4)MYB%Y*uf(CUP;VEs)(rh{)kQ0Xx)*ab?4#=O;#oP)V_+0^BNmJV)=DdDDQIg`
z@iJQ!z0Z-(eH04bMwl~v9*RXD4f%!k=$AtVVlw8sc;$K+!x{ALFv!fbJOi@IetdEL
z;86VFi1SL@Dr5Ts!@ltm1`fjp@u7m7@UwR7-GpNZ(mY%qykKNh;c?8JZyvxOk_lhf
z<7WojoB6-?ya92m&L@tKYM<X_d~=T2BdjrgZd9lK9jvz(EC9j7bs0$ws5Z)Nr`P_#
znXpCM<--7EzTM6HeD*vEi2*^|ckYOEL>d7IA$b_G_Zu@L{M0_6$A=0wrg&@s2IO}n
zR3m8^zm6(c5knDTCVEyZ0kav6*^rZ7k)K^y{-?i0`9J<T{XhO)EHUHAh8_?D9{y^E
z>i&3L`tbw-ceI+xo4>H#j|cfHBiS7Y1S0{|xwsyaln+?q6Rr*uX|d1V{5U=m(1&w^
zuX>vxN2~59#BW(8FpPzaL~1Ur=29pkq$s4AoRvM!xG~=MFpt6E)wHa{yn6L-_oX!t
z>=MXJ_CYnKXc;}=!{zeGEysBvm={PXmG=5djBzRjLQLemnD0GycG&so$@{#g;b4c)
z^&!YpCjQB9B7fmKY+qk_-3ssT8~5v-z1F!rx=T-uhc%||S?gse71YtX?q{GjyQbQ$
zIEgu%T`p>4ij(>F_f|P&tNc?+HraIVW;2XT2^|<QFz3b80mX=fF$6-2v|_h(tke{(
zXjqp_PQfbYD1=j{H08Eeq6jH~ra;*&_|eq*{SV}4fOp%S%a-`$^2EA)#P>h@oLVa>
zt)N0)vYkxL&XyR2bzO|qC=jEN*R%KN3+v^=zP}?npD6`_oTw>scSNOl&`r@MxyHw&
zuok*B<5`rAw(VxE>}{vs%s?y5flQ8+Kh&yrZ@oD5duqqoP<HhX+D*;JF58Os7$W1`
z;&`0^#0V)J#KRx+_>u&|bU>@Vz{j8u@c=ZOqY#->_rNRPAP51}t$xP^n!$2Na~#^%
zfh2?p3oB}aJFwo{7%GIrdyM`+HupdfRK{n*t4EizAr4_XD}G4q)W|MV;t>r(3Ot|+
zdSn(liETUtmJb!;$9EkGz|iuKk3L|@K|gvVTG{(dB*7CaQKA3J_tAG7d~}+hU-8Pu
zgP=aV-|Sb8`_kS0P@Ki92JiRQkT4Pln2FCs8peHV^X~_lGQ`AW!4v4M6H}n>yUjaq
zmUK+0eN^NQbtz(IO4^ao>2-sc?Y(=2vk?E~pM%!uf9?B>m@MH?Q}_a!fp??>qZ+j!
z-#m7YN_}MVW13t)eznYQCIlZ;`7;^Pxqf7>HRk7;>IS<(h<Go%4|(Tte?NGBAG035
zx{u3gbxUkpf5GRJ2uxp#-a8r^ty_Q#5{c2=?M7=s)fFmY=*u_&34xRnF|VX$ooy<U
zeT}2j*#3rUDk-Iz>>u1C&%pHI{&zF3srNhEw%Ol{kW=DxK9f?OT!^9U5l7SHu__w{
zh$lkvPyZz8pZPiVpE&dR>nopr|5L8_!aiVqs{-shidA1Krjj`J^-yBrfxS6*=fiyv
zqL6Z=)=FLiOHP!1Cj{ZNE~GflHnku$6Cooc1~Hj(bTZ+#ZH8J5avsW(Bl}j&D!8i=
z5o2(!NMPUYhA51}#8Ea?Yp0x+#9ljP5bVH;I{zGdZG8Xbl{cpoPs>Vse&XHrO3}{c
z>4|(=t#s)=J-0(nd8$g5^+e8%UIO3!-lsgheap4Nim;v&5`=^h{`RLv5cRbXgN<s=
zIZw%G@7?Ac_`0n7?x=tsYiZ*3p5Sx1<qkA!5C=yw?>EfV0_q8X`8sq*1RhT02e(VR
z=7dfsOn1h(Ed1yM*Z;D|mqZ5QtY$25$vkg|Zabx6WjhO-4=N$?to(>;;+xY=AsijB
zL%q8G-v^5LW>fL%U#+s7!4G`%2r_$j>ac{)_Z+PPiOCQQA67AffgsS^0rfgDW4Qc{
z1DfO9V2(bWW=2ky9FoaLA9@kcF;K(*$}4cejg}aUV|CI`oypQ5OsRlujrJ?wL;uO|
zcu+Zl+kndgYRObgvx3lGeI#mQa35MXzJT})?cQwMXMB)g=|k`g_yo$q5ePo$D45Yn
zAC-*SK^4TLPC*`E^l9J}2o%uxpbD@=Foeap1mc8K?a0zd|DB(s_D26terQ#-9?;tZ
zKES-~Gl1(nNDriA{F8z-zgOd09aW>kv>p}j+4{~uAi#l^z_AhinkjK6&j$j955NiY
z8b5v<o8tG?o?3W0pNWHgjWP3Qvi^_^^wu1~vY;AbBIb<XL8R3~z0`*hSA-C4h+52+
z6vT7{H`TeHfO$^^L%T*`1y7}YUCfyRQHf+$RiqRreA#_HtkBKKD2SmOBk^GNiO1ei
z<ar_e%YT&iC%?_R*7(8i{D9Znol+{b9q8R-q8y^bz%$=`M5maZ0^>(|H@5#6v=oT+
zS~)F|=d-CH)^)M!-9KXxJ9kSd)YgbG5u`g`YZ#u{u%GRIpUDK+Na9*`@;Mr6JOfVz
z%)?+5f!MUvWU}sSX*7itB56%r_r_k`8MY`A`<#`s*Umrq>>b~Fe&#1W`GmJ`-|+hK
z!d5HUl?-^&*n20RPDhg0`;@)k_rho2{Vr(cv-cZsKlz9pl}_aC{zUx7H`#vYyEGL~
zgq+lgBU#SJ7=wrgN-|L@KM#hHoC1s<zY_r``e(P8dB!;M@F;;E>(+j?)M4`7yz0|n
zNpW0OqH|kPOlYfLXpermWMBqNX4P-%4GYO*o^p&wWk0+DW}i$Bm%W2H$ia{wiAD^B
zo(%4u<ig|7J}S`%cD5cDM`mz1vg1RB#8E{*Kx{Kwhwu=b2W%sL@on=QJ$%1`&0bTY
zSyieG7@Qd-#-XRjkInaf=p-?QocO-?YY(mLM!Jld`KGGquYMo?%uhm!Uciq9sc7A8
z@>oBz!|9hTmUM&=ZIkqdcyPxLmgWeFwSBy&I7pKsP`$?+LB#w7kT@&0<~##hH#uV;
zx@^267*2=-8N#cAVL;>8RIoYfc||pVR>Db$|H?1eRsK&t^`Co4x`4s9p3@&kh1G}W
zx6Gov_ii5S82fmr6=sz*gI2ROqUyHUVXmK{jxqLeH&;iIxxR--!kBB%P=TmUWMizE
zkzn9*%&+oe1c~9hA5)x#!ElpO?erk%n7<wZAuhzU%;dJ!9lVcC5+8n*j_Ua#0YGuj
z{(;&s&xBSh=)>sRksR8SJlVH37$$KsvPv)#Za2$s)RZ(dr8%*dVAj6ITd=m1_D287
zC*&7CA^gRk<@${a?|%3h?_TfhwM`yM2twB|)m9@Z&=Ji1I}ZO65tHAQF;4({UTe*L
z5_GQlB`-XmGwU)2k_B2V-VO<ry%@XS5wN{h@R=ORTZG&7%D!!el$1(Jfn`}xEP)Q9
z^yc=z+C3OG+FqP7Z*ATs1=bYMzH^Vl)|A!JixdK%-wJ!{Gx3T(j<|~QS}MQu{yi%2
z&9@(M`Q&5v+E~x$i30W7&}uCcYPAHRRDu|%<oV4Te(Uf2UDnHqpZ?BI5zbG9*oh(0
z*OlA#Moih=6~>y-<JW;4jRM3AQ+rCEEd}F?STfLPzFzul<vdE$?Okjiv<EByNlNwC
z9^<Ypj2Sm;w9fMR%yK%Bmesf}K-<6A9{tj@2|lQ>odg8ffi*t7qX!~lJVCF%S}V}R
zv-X34D8n8%bZDL43MR<#19`Yp(V3FIhb(_21RpT7)e{2Gp!>LbUd08!xHgxdG3_2v
zH{m&26hTH`?qOKXEHVc;tX{RaW;F!HgtN&v6Xr%62h~9yv*0cM`Z_i;m?A)MTgvX4
zhX-#}<Zpc+y>969X{JmfHYBL*H#@BSdWk@DyG;p7NDColza|~J9VoRLEFDvxIhcK>
zaT$dmW=D*OS3?hp&wNKvKd8nSQf6u=bZOJgvbA=M-<*U>o$E?OX#1Uz&Df-?!WxPH
z-CuzIcc_2uhaV<@JdkR1%xf4@GrzylzI*G3qQNe<e~FRSIC&5!Vj)hd3=y5&3TBc?
z7pItblGp7xKn~5hj^)@tv?7KBA|6@)(TV+vK74Nu?~0>uj_)D6$5VdlXj(`KNw)q&
zGc=&%vMsMxhf2rpp$*p3dPrJ(@AExcYm?g%&`GdTeULDY>pj*{jH9pwj&+W9dhK*M
zl85ncy+$-GgdE|gp3y)59m*g72DcQrmCpXX_Z0W$|A1xQ^`XKX%y9KVKtA$T2IJno
z$1IH0f+s%h4^MPN;B;PDmq7G2Reh>o3Wi|RViWe3WdY!R-@)63AsDiEzu#se-AVz)
zoC{?qIY*;4LmfvO1~sB)E2gv%4rN%YjZ+G|St6+_IR-9k;-2gpzHF82wtLH=5PBuD
z&_m;@@LQjK&dc?MH>VSq(~0%zG84NHgeE4Lq$|Btv{fs7MEU5=TQ297`|FJ#{?2!K
z3g6`P^^ZwQ=ChYqw%g5~%L7r`lgi$VU)J?7y*jEkafIk_pBx9_@n=O__cQZol^#kK
zQ+z1SX&Z{W+R#!QIRXihWm!q<ip0petSqOKkz|wGH@8M<zYhdL0Bw~vGHUO;kKVre
zLHd|g`(>EUblMvl2lv1)r5~Up$f$~uUuNa)!M?k(hG#*uihBHf{=XBEK&I8FfRj-y
z`<Bohy0PjlF}xBEXm@Bw$KQ?2O+1LpU>b{A8`C|TMrJ@8pFu)8pi$J|n_#b_U?xD5
zg+4i~^IFOXP-7;#XQtuhM*HjErT>}lSerr{wG>*}oQRhu6k%+HcJy1vf!8Z-dqMYC
zhK+FV9eIGnMWItoO`s2f78sVy`miy~Si>r?L4@qhz0W5>(gLyRizB@$A$0rx9f-mp
zZW^?!^!+}sBLcDr>0kLpbgQ(#@#(yGV@0T%l=Vo6j;cQh#zW}+@J$#=<Xlh(r=MWN
z$k}EXf$gE^KefQM9>~CeYv=oV(liwreu$2DQvhoX*$MZ>puRt-k5-4#mJpL8Qoc|Q
z1%>*d7>`4A+(+-B3I`N!0b8s{z)_fH*Q}{0=>5Z%(5T9+LKRZ7R)l8`^YeZ{5Y;gW
zwbMqXI|_nvK6DSj;s^-9PhANA>@Twa@o(@Bxb2O4t5)?51%&s6LA})~-g49VIY23U
zK@ro$KeUZfvNLNW72?;Zk5(Ucy?gJJ(qT<g)wJYH0QSu$g)Z)_hblQ|df(YMv%bC8
z<`sXVlsZvZ5mRssnXy?G7b~D>+)J58b%tWwK4{Gx2Xl&?y6w+paehcJBxxu)?xnNW
zVu5YnQSGF>@>-SGyYjvFS3W*3eDujD^kUOhwVIhvE8a?0MQcS8XpgMxiC_H3{xN>z
z*M6PXcklR}-~F6rIkBEMwwL$wj2`i4pz-5un@J{fF(P6Jib+V188-t_58iTZB2+aV
zz$`dDrXj|O3(<<|kQOAm$rE==Y8;7>X!kHLi?sy&*PIhUu=A%?wEjM{3Gl<Jw?<=p
zaUTlgUl;-(I_z1NgLa2myV>4)6)iENz0Ls^<C<Hw+2>(Avf<u6fR9OH>N?Sw{RTQA
zYcl*DX6uF48<Fg8c+GU)-KO!0n7!%X8jOH3At2=-kZAyC!w0WyX8&Ku0q^569mYJM
zzUFhdnSUOEVvGRoy;+aS$&Ea=Fh-0M=JyS)!e9R#_$Pjfu0k)n4GL;DI&;oW7y${f
zz}p+G70cpLbQ%zKh{=b8*ynD&nRkQ3<Pxkhk112enB%Dek#Qj`_+#Uv4;00|QupT=
z^xjE}RfXMyzM__l3<j_2w51-`dniU^6~aIBQ?$ST9(i##`nYx@;c-O9g1mi_1G==E
zr@i{29OprNv@S;FXa|33joMyFd1YBI27BXKf8D72-H;YPAPxf~nd{2-L#4|xhhmId
zohOg@;r<nV^xg<vAiCskeE;znf!7b-5BL|3)ES`^qQ==~8U)ty?<4se@U8v)Q5b|E
zCSPO(TIQiWQJ7JsA#;5gi`+?IxDdk6g5Jiu7Lo|@YZv0b^hapF{B7RF#C2=bb_9PJ
z$OZ-_ceP`<d;eLz8XfDjd&O_-M+`dG?gYru&K%dn*Pp9#!1hRr!n)>JoyHK^_kD7B
zat>p4gDTs$aldZ>)Mk!?aUck-vm_fzmhw=21(V`+`!sv0?4=Q-5U^mBL|K=Gt?V4&
zyIuR!5(rVbtR_4B;me(`J)JnE!Vg|Iv<G%o8!%NXMIkQ>D#F$}_a01=-Uj!~(6?UH
z{TL2eAiW{I@%H(`AN^y0l<$A<2YmGQ4d<s5ZNCxz)VIlh{~dML!yNB1VMQN~6?HkL
zowFf0zk@@SG*Pe7BKNgxybdR$=Ak?rgmsHSamf&p71F42q&x|+A_tS&Mr#S+azx|T
z6nd%D^2PcDUn&{c_xqI7*kM>_U(CKV3>jYt!@2IEKkA3C%+01uB|`7?UO~MN=!{e&
zw~dPP=%Mzmj9J^3kgG{4)S!Q3u;*#;!+>-=AU+;4(<o?WdrsnO<6w#oZ@M+CjuDb|
zzV0I<|2WoMuPXHeiKIw?m_1%vg+8pCM|OBT9-w&-5Th^=24h)?zxe_$ciQtxYeO<+
zyZ{T{qq6r87lKK!aF%u3=-8NPv*rw*e6@Kmy4r$mwU`e9X>u4OM0|9aUrVpjT0Igs
z|9%HQLtV8H=DN{`O=mveji5Shjpu+Nuk>3je$4{m=ijjW5C2ih|LRxa-RN76ATqoW
z4kF$-Ksm0eCrpCiW+0x#FeF>P{y@xau{U(TQMDVSEn!xu!MQGG7!>S&`5|$%wY;Kp
zf}G6TWA8Hvc3*V@AuuIs#(I88t~`K`pk_l~>|q6gBQq{F8)rvr1L1NfM|%@XB;^Q@
zgNrfI8%GT+pwFe~_Svo=as)@*aZ?70paU52&)+-EpBFEg_@}>3`wKtM`>#E5+dJjH
zn_gTk*&j)VB5uNTpmvYwK&_Qp3~mn!B<9&P_QseuFc6Af{d%-Iv8<E=&zCdjC6k?1
z-@8&ru^_^-E*O6T=$7QBXx;*?R(iEo?p_-E-dIxdbFq_BHi=@XZ1?S<cr$C`4rpuS
z7|1zLYNJ-Q1XtD0ZvUKy(wGA2Yacy>DoYafZtFibGulDSqoD3Pd0B`#x*{U@cRvsz
zR1LYPevYZ}MI;pDe&gxw8=ijdr$H=Xy}kdO^iTf`&p-V!@Bg>Ig8cUNApy_><qI;e
zaR}b73LlQHZvF6gFljhQjLs)9a(wfowe^lPGc^L{`3qQq;zOFiitrSZ5oYD^C6HFB
zwNShbIa(Rp7gQpB$u@z(s4<@m5^lR!>_aAGz2@2D1;%*R?wipr%FqZO6|!Yk10r(m
zzu3>}SYXVZhy<|;VFr(e^g-@(HGv}wy|u}j$Ke^Y9obw=bHg9ZecxbRLA<h%I$^Z2
zI4VrW?&$M}aXo_08=~DAx#|H<3>wmhKxlg?5_Ejd;WsV}BOiF|6?(O(9sT|b`ggvM
z{_?jNMUFwDl|kARf-g3)z#S4HL?R7JfnF!{A6$80m(iKi;-?b6$eRaQF@A@A6emu4
zB5+jS_@T_43()=7;ZM<u4FMULM{^6*E}m#HNM819h({thldYdSlmF#kWdC3O3bV!2
zn^g@#=64y1*~qX*U<!`L^j_`2Q-kq3ekkL*_7F)ilVd`)(Y4sD-4Lg3cZl(W&tMht
z=u!Lka3b1VZyZ%ASkI6ybFX|=A9k<&D(gd1=KE<#%?u_D^ufX&3Yl=Ug+?$}ec0Cz
zaPFu!En(wfzZ|y3_Wxd;`TyA(8_yW$%J>F$Mx_|>m1v|u5hMlVk9`aMm;NZXubp{!
zue`qO)Kcad*xR8_l9)+3vu}4(KBVkLk@|jem|U{K`?WaP3`;Ei+!|23z~n`A=R^!2
z1jN|?=NxEb8l+zjbAyC^+sqUQ+aD<ggZcx+B^j(;Hw)wn(i*1}S(Z!)!oKg+QvKWr
zn2W336XQjg-oE(UgA}689B{ky5G)qtC6QNy#rqg;q_D;t&x;Wrs|IeRA)>D7uQo#1
z<BLj|*65>&TxScz-p`_M<9pxxl-u_|=X|*!BD~)2q`vXd3j8CVkp2gMg6se6uM+Er
zXq7Try^lPA`IRlPA1F`n5Cg-hHKjyf@~rMdu=5WYs5VBNbz*UK^-A5XxH+#YOE$$!
z2tGF=1aAqs{{cGB@rtm2!9wCoB?Boh<ecq8OC+x6Ne&xTae`5*W7Dz3Dt6liN6^*5
zS|7oIImM}Y3t{X;)rZPWkCwy)7er{i5~TXJwZE&q`h~L0d{hl+H&|HAcUoLpDBidB
z)4&ga$12?u7Y?G?fO$rCuOr#us5-|XG@cf^(8^AG_)Z^;d>-wR@tWg(yT@Js^YO`i
zLH?EBgg^FmlbFoUYY2rwnGO1=prYNwu+VEoYw_ySAVdWS5%EAKIPAvpoj@GMj~3`%
z*27E&niF6ig0LUF@^Eczh>X~Mua6mVqlYjDVp*@#rDz^dEwQxECJyEDM^A)5_YK-#
z`MxVa)RRS-n;4Pfdi%AW*Soi=kg*<Yvi6Q`e+HVxJXnoPss6JBM^<`A^_V89mX1x6
z7<8a$&TE<1NSx#ev0L)>;VS&<JxrEn-@9u=2}Hk-h6Wt-<`E(E05SR6@cTYEKEP-}
zjfr|AOZae>i&yz0I~Kg^Kh}X2!y_5j`8-Qne60a=O9-O`B#NIyAkPcSfBP4yzw}Ms
zH+X$5Y}-b!1zM*L*Uk_q#j5d;3bahU->3r1(;GrwZU1>v69yux)*@=&Ix-N<_e1Yf
z?mS%adOswQrR+#psI9S=!a3V0;uw>)B2%JlcSASof#?D(OJ>nTYmL3@?EdwxG=b6}
zl!lb%pV67}Zjh66ruCh!oxB*C?Y4VyFc4ie<R`_%;SmCLFD5BXLcJEAaw1DG)rAC3
zF>o!Fn|4|&(|}5wD+u~fLpfPh?V9!0uv*Bz;}74x=d<h85-LASzVXQ$IAyEUfAR_W
z=YE#<@BTWAxB7Z_kD5HrXkTdq_E&Etd0Yy?+9{Jmpph68Ia>^BMJ5S*w#J+oJB<7U
zCC5l!5~pP$FWC&>80&hpa-GwXqK&o(LMw!@eW5-2rIG=2>GO&^=w#pbS`H};L(khC
zhQh13<mv<KWr$k8+5(v~u=RimM=+d-@&mjE<2Aifwk3v2J)l6ZyhDtZy{lDfwrIn2
zKbw6^!KP`!Aqmpvfq%T_A^pM+l`#7Mqy69&Ed$QBD`&6Qj993SP-t&60BDF#iaQ>T
z2{7P8=tw)DbViTR7e2rGL*#FNO8iH^Zm@k+)Y<FZukTPV3`lU_H-nne3F-#Dc#!of
zO9Xl|w=Fe}03ni=^be`Hyux=Cg~T`)`TR!y{`2#Xsy{>v!p$J!K^E=Jvs|@1?fVA7
zXC0_ad7`#T3<;J<{BQj%goyr?ADFaAiyLvN8MpY%cXhtVV=U-?U({Ee4(6~3At!S@
z!wG*LkY-|t+C;fFzJnabAfw;oiNJwk4RfqfeGC~S^j^_!)~H|Q$2hPLRLyKvMXjGJ
zqeY~~_t2mWpL!Ui%r-J(KYT~>e<b4r@i`>Bp6Kzg2B(dU7a@;pbbiL65AmeH5lZ71
z`FZsVACdn1f1L96%xBla%Y8>{A<)TK(Cpf29ZM$5PT6i&eXr)7(VA8MPj5f5ck_0b
z$u7eTD!=Cp^7W~%deGVqSUd!JD?UcqoB}y7#31CfFce_>wplyX<<bh%W>_7Sw4}r?
zOmpdN>q04=x0=l!*{>El)~T@XWfBcT5Ucbh%t9|2!wtkI7pnTNeG(4$ZHLx4EfG;A
zhfLE>i;)l@2TPhmjKl}^$4G_@ez*2Oua!7j##kb?-CnuBzWQYqGcH;(k`_)u$hp)0
z!cX)0fACxA4-4xOA5;QNM8J@(54k}4Fe_*H6O7~`#)vF7DYB)=kVG4LFs`N_0C^Df
z#S?-&y=<CWBCB)gN;4PJq;9P&wqAc<W)rl!Q-(xKcMy|bjB2^}VE_{BgVySc(+|RK
z*?jNDC^tw|oWZR#FbZx2G~$L6Gpr&zw3}l#wH?G(xru}iAV>*;C_*g-)!^Ayhvu?E
zyhrW}LVKlEJrIOo8)e9u>~+K3lw%?3s3`i7ig>2y&lxvl+}LrmB}B#~G5_-Rea6pv
z08zzdnjDp}3~8rXS4tPsfAm}QKlXL|ydr3D$G34f>M-O`RCi*^NM3<Vq|&!ndp*z6
z$7g7DpMAKrtoAX3-KU^fpIpa15D<4^^Wd$)lf-@y5XT`u=UIf|Ksb^#|IhdNfYyUV
z2;CA>nq3#a_RU)`nmY0?e2dT;?Js|qVbn1$vPzuQL@jkzOf$jq1Nf*`!wgLGb>^XS
zRG>D6%mT*P7hrot+fgwb=#<kyd!yBz(JmWn-2knK@Wr6MKd_Hr$#WDe$oo1s#EDr5
z;Olh+=m(68X-xJbIEfBk!9V~QXSB|J=8Mq(p&!ZFc=C}Oc;Y<%UOw0hy{k#w{>axz
zf9X$fKPP_hZsT<;2%Q*(HD$Ck_Pz3y0$ZtU+eVI!r*onMyZ64yiTAfF*AU2g^`aoq
znpKd7HuZTu#=Zz6(Z&+S=5n{Gvg*p|ydoj6@5KnZF}ek_t?3vKHBU~ag6L|dLrb<v
zf~}Nk?;8YC3e;*r=U%F-gJ6j^f5FBU3)Nw3OX50X$Xjq_Mpv=vq+<NwnBs9JV2Sjy
zEPgk;Sq39qw_<dmQi;;dGeLw_cUr4NvDRxFg(yy_Y?T~OGY(Z%uJ;?YRxVFZ5QL{U
zXU^GT$=-a_^ViPgf9hx0{&)X5F<JuB1)?5nX2t=JXE4@^J4u2eroQF|Qgjr&2f=VA
zENVX~22&kHRTwhZPOD~81@@VmCqF6$-R;BG-OQoB$m7c;17-BO@kw9S7YQLkO!jwm
zhmfdE%?!jf?yWbZ6+~>vh-v4&TJX}}f`{q#d#uygXpn#m+8yLUn}8>wTS79(eHhH$
znvrknZZJ~oomy_>Wi{tP5on$)#AFM<6+%ceupq%MzE{U!S*`{CgB+<-WPePw8i<B~
zs|W18Mqf_HT<j1Dd0`OAW&+fWfiH}qE}c^Mqc0UQQ`x8D#uS45+GoTcUg;k#p0qF*
zryXjZz50rqLt0{lbSBh7Z@ZCF#DcX1A*Ka&7qJmAW3-KTRC~Tq22vn{IoprQOmL};
zw_x0a23##LxY}OrGdGnzY;34J6xlla%E37m#6XaQE{gt(f5;Nkzx-W>l~jysmD&f$
zO0dMqUpU}W$8jHhGj^AM9tn&<A4!6E%Oo8rl=waXh6Qu2)=!SX%xR<tYAw|LN=TWM
z-CEkOaR|}<^}pJWQQ?k5-`2VFCPpw83kr`YSr|l0)Y~bJ){Vm$yt4I*ZNzvd{~bwt
z^Q4eD2V$(<v3LhUG};8>mGN*wRaN+fx1@jVm)TB<*Xy0uD=C7gkqj59oG%GoI;S;5
zys+O3wN-jkwtXk0%x5n*x*8&Rz29uFr^UzV8r&R0n&d-+sGVGjpGEzVTf|fl4cTN@
zBb=@4YQt@sO<Bw-S<p{LL{mDY+MI$Y;?}x#qL)FGr_;g`1NXhsnkfhN+6mfVN%Y#-
zig7psO3s52X!qXMa+vqbJY4jHk<)2ishb6_ykzq`>9FKv1BPm(Yhz7;CFGgdZMAUU
zZk*T4JlhbnjBd5E<b06%M7%u_cseg!PKl?d3s2`n6jujegy<9k=YQnulv89c9iotf
z8rs~C!r%jESS+QnV|@6Dc!A*jsKHSg#p)RJWLyvCD~v&D)lr$QWs<;KZ-%bT;=x+2
z`@X?#AS7Ote}TuBn@zAR#Pv+wZq6K!gp`qx$jgdqXS?6*#;=R@o`*iVo1`#>vG{~i
z%;!5Vs~vuxNj0z7N2NGfnFG2WX}flMsdTCCQWnfo*Q@&00kK|R&I=;$L`A1Nk#P{@
z29tR#TvISuws$~~{UF1QtaI2>j)mX>_js`|52+9M1s-U%H;%zfjxKm$J*-xZC0pul
zgFRAWSc49^lZcp1^QEEx!EclPg<tT`tdoWO2!&2kU~AN}vxGAmS7do6lox}coDAB#
zl5#c_!IL2~S6ci#R7OvHCIWtu0$$BHV)IBWox@=!m&4m8g9-kiJQ+XN{rgO4pUCJd
zcM0Z%JkTBiIS2Hg`x)9_{{i~xK8>D&IPC9!>*MekZIqFThvaC%YzF}7w9F;AO|@^z
zvB@W!+1JECi%ibhsI4Nhm^qf%VUpKpCo38U!IDik(wXo5{~wRRuvYh<P;H|CnpKi-
zG6q^^mAfN@vx>BPKMr7c_z|!3Cl|y@G*c@)K<^Vl7^sc+kO$glZ7?GNhd=x=@qhHo
z+}FVS*Bke{neSu~N-31m`S_!ikKe3Bd9uSkC3?rDdcwBdiP84c>-El>6Q5u2eExbR
zpY6JY6dmP!K%x5ZwYPymgrUOn)JDYT5<t$rr$p$$^>zn3IYx5~R3)VhYBEAIAc}-I
z&SkYACAJ5j&a1T>dZ*=tNMqkO#t7bW+o;93Suq5w<fU<1k|$El&tOaL%5Kvb=kO+&
zYNc8m>*ZvG;us(ZtvDP$?j3f9B%d`QS-4+cxGWb$A$l7yB(q5mIwZo~t<5{%{Mr*S
z1x}}v$uxmd_9-!6@<NJm`h}0UefvWB^><VPkwA#zL~*}8BEWI>+1WITg+9@m(SjUn
zEr=Q54fCwvmgK|Pcg!UywVJ}DwRxU0)@mFHwzWU^eRqULO`bkJ<d1oLIW>V-Dm@08
z&sa$zLV&X0XRj-!g}j~}#;D+0{(awgt<BxEl-u2;ice3^lkgx|<(QTgU*9CevUIu;
zULN{0VCD*~D=~M^zN){LR{`#9mqPFiyD=;bTCWe4tR)!Y1G$Ch5HXgh_(3%o;!_@p
z!Og7>>sf>4#l6*4++NgowaMvx_=@5^Y|mI-Lob7z=0v$5k!3mYLH~0E;Fzhc(~4bB
z`76Ic`tv_U{P;ZE6V^xcEL9E7>29kk-eTzzk>v@}YOr9rquV<|GQy|9lgS(a^M%r(
zZg;Y~T{&c7{j;I%wGTf}=urVY4C|d$S@)L5;qTzd0)E~7o|;6<+|gnbx@YK$Izdt_
zbC-yu#PSmv{qui_`~T_h%;#tvS^V$|Z|>_Lh7S2*5?w|+kc4sHENC|;ocebX|2`81
z8SBlH8ti_il*xHzpi8QXsYKnJOOMH%qwoEQj(jyA9()I+BZn=yC3uWdW)|!7I*zpi
zmcY#f(IHbA$?kZs2b2nC@;Ltu+;e`MBmRa1uowsJ!z081f8-O=U;1VCHE?^qdys=T
zq>Vm|3(cIKo(wUy4M&tn$QFE0mlL%%J`xK8W!qU&;<kOtwr`x4lYNkZ;yH=;P(|5w
z8!n-9%y!pVuo!YDMjZAvr9jHZbVcpTwr}*-ShC60Znx{4%vZZf0++ln{TxIb88g&F
z1X_jKY_B+#%o2p!Q7SP6mSjZ77)*tL2zA@J?bZ%h0-V>Cw(ksqQrXSCNL;!oeui(g
zl2d><6h=ahiG6#a^-nC(%69ICZ0K~mZ!=l!y->8;IBmsvN!iTW1{pGl4RwmqhB}>|
zPgHql?Zcdj*(8;N<1@(c1NBfMknu9Zydi3ZeWR&yi|qTvnG8+-P%wU3sfGKt85h9K
zvU(fwp}YTzsn$}6hk=#|W#6DHF>nkI{4tL&H=Cdp%Khco99ShD{lb;9BQcUh=y!+J
z+%mVdN~w+ewsRI6lyN#;Xti?RHWq}O7w^?Z(sH5KLTmfHNMpj)2xtbmnq()q+-p6p
z)ZJL75+X5xmO^QDGWJWc)`hhdOe;H#ZQ2Jq^!_lmw;vU}M4-*j(rWcScmyKa?C`g)
zS)=nHf099pf}Ga}xU$b+IZ1<bnAM991M+IAH}g2?5?GcK8kN5L!SJq7|E>Rw_?Lc}
zc(GXtt+#o;nZ2bSdLBf@HL|Fh2ZCT$z4>JG3R<ziC4n9@s>;5-L(4{r=fj+1V6@s%
z4npaG;th}(_N%y3;rO1tk5<70l5t=SUcnuKvQ3LoPZS=>i)ze$(MtNW-y;6ye}I0r
zSwIOs_@)$3jz*yJmVpxz+aMFF^+Lz6E!(`wod$i45+<TBP=Vk>bVRV0nGQQ+@l73P
z!X{!SG{Gxi{JKa?gs*s;Ah?N^j`b$N;7vSn9I2rWq`>#wK+9(T%oF^swr+r#Da<m>
zj$=!->}Pu)2e7pl9+3g&k2>$bUu&QXs*0Qw`M>u|>~AvVW~Mn&EQv#f7!6H`r<HY0
zgmkii)T`e{{~q1E25KoRIdVQN)KZ`}&ZmX@%VukOT&aExaMH;MVDeIiQH`NhhxM}8
zn5KqNGknXAmX&JIaS-z}h$6JwxnEz&tB(OzJ0rWo{dOY&wKn#BXUUNm68l~V>q3bY
zc03ueg0a*}Q(-HeAl6Ea!ANlSJ?tLr1U|p+eEhUfdox{#2yHKD@7(LoIbB?xV@XSE
zepWKb>e!h`W8Ypmt#5#cq=)lEJ<)7d$lW|B2ZZbO%C_yC&L^Ui^eljPM?i)3&;K<0
zuYL+4kk%9H`Q)4q=fRcTCC7f{gH`E>p;w_+8is@q{2<2YDkMjOQ$q4$E%#b!dS$=c
zbjWf2hAe-acO4k~qLeb?iodq+<`~)gOd5aO<I5!jIooXc5UBgzO%@U<uhjYqILme4
zDeR`Htftq_X(8p6k3Ko0G|Iji%FvYS^#zg0a(+f)qTFBE@9(i`P@9j=w5&R2YfA`F
z_KjMN1aew3t?cXuEfDi*!YK{7yM!dA?613<1?IZW@7G!*<xI+p6Eqs#C6GacC_)Ge
zF(ul*vll}tVu%E>nb$F8_G$|wG6-O1_*3`Ig1(Pe6&V$9V`Pek77X_>=E%=c4I8t(
z(Xcu1^u~65A?1uXbS;4X%J*ph)~^u$wO^i<Skt3-Bo2k5wkd;q4p@@aCn_On1(s3!
z&LCOKhWbG{GgcYoFzaQGIpGEcy0$(rVQ=+?Ja@CJQyz(Wa}2f&nAqh@Uez74Fd;Ze
z(fakikwc)}cX$?-fBt8<{U81dq;|Skg*9w}fkVN-5q89rx89G$$pVZ2*@>>kEwENc
z_k;lLHoQiLiI(+z%Wf`@62#WBSl}0Jme{DGcohgtL$0szBTP=up#P0)WdA-ZeXsP6
ze;BXuP}%n(Wy6VyVeTD!jXtY9hZC%_?j4&iZwcDb(wb1jTnl4}%vgTuAO9A7>q5U7
z5~jt7r?r_;+IdZ!mX%&B=hHG1J&C~$mP{#Wqo~a=sQ1qE)0y5HpOse|K`-{4yc#3D
zwdz`W^IT~rIkD?0<|XpDXD1;|J2{#hZ(O-n8@qODv$Z!&e@ZLt+F6!t+$1Lqw$|Bu
z=i;M*_kDafJEQFRc!FR&tn36gli?|^+eYic`@03CqiC=JJFyq8Tcy41h7zB%weqU%
z&AK=AY^X>Z<ZKSkQt-9aQBn|UyHaZ;rzhLL&4w1)x~a5QxRs5U?e61ftjN0FH(GC;
zPbZL}Y_OORV<fFkc1+e1Ppc&kHs=Hy#i4&&HzC*<;X(Kl1nHeFUSQ6($~<esdM2(b
zgkUB~2x;GFrTAw(rXUW5l1gx%QMQCM1SBnJiu7vn_lrKhoU|~W$T=et(OPZJJb~Ub
zqRP6QAZE(#g`ABuAW5zF9Bt-zUNRvpEIG69JK8FzWkGA9)j|$|UU#&1Qd)>%j2ZU6
zsRUB6V|jSmFF6r{W#{IOH);l3G|PG>Bn$AifP}Q)x!+z)nt>6Pf;&5jSD)r39eqPK
z>b4V`5@TjSk+n7mmMoN74v=`Z6qHfGC^*dLJ*{>?qRkKcAji1i-W92?r53@gPYw03
zeo`_>t`w7Nj?I>Qs^Q=G9r%wvCj8l-L1i2kVIl$+sBkZITeza8Oc<H;BV4L#e*4B=
zHZ)`ys#pcZk+i7NykBdR3?B}8u|JRGVqPa5hD|cx_mG?&_wz6sIwYw&PyiV%o=%j6
zs;LB8*^oc`HPY{1x&J@@9$`odeKQUe!eBZdG&Q^uoS10wUF}+8y~hw_29}(&TM(O*
zm^R!DL_v;$NH`ZJ#cV-~W`nuB1vQ}HF~4Kj(SD^L4^m3-v`XttGV%7BjCC~Y^2gU0
zsNhiEh$nQz`(ac~(OVBrX7gltRL&-O<-=Bhjt0W`{g^MJaEi!Z{EoE%nhnjVrOrw-
z1~+}F)v9JO<Sqh80<qyi4uydz>2xA><$il52*&0Q=K8nQow9HAl>BRoAO6M_8rIYU
z1wfrdJ9;CkBtH|EoY?Lg`&PI)LX~o43CX3@1F`F7;!|xB;W+<tN<KNz3Dn|7In8Wv
z8_M(LV!eIs&>E$8Hd76fV>BsZtz5Ut?$aFs-aMThCWmWjyxw=#JRJNwwOQia%1Uoa
zt2V!2BpL(xX{}A7VvIs-SKI%3vMGeMP+MUycV4$UFWWZ1o2QVRPbUkeGDw&<W_z@F
zB`E8fEg8cQzsLDFV@jzoT7v>qA1r(LuY{OP{e*lV_m-GUww&_}_}UiQ`Qq-hVbTPw
zO=*E&Wf;mGsFncT6Nf|c{^K8CDjC>b-$Pqj@&al-f4CE_FLs~?GqTaPA!;P4m=;1X
z*h54p_bVoAiA0p9)nKfkl<k$nu)!|$dU~R@O55+GoUtKBjq1TO<;CE9R|SZZX;cj!
zX|0g*$y5Y+0ql@S%fiaVV^f;)!cq%$w~9aK1<}PwKvO5(*|)3pd-6g;rd%?{#9nq3
z3>5+IHJgeeq-=vH9OCNkCdPxzq`Sd~kw0`of*}URjDxPt5`eywb0#h;b81=$G;Mz7
z2>NgTEy54)#J}*f=-HUbXrD}B+u&icY*j~K?lbUCo<yrwiPz#JQaj9BtHfwV9l-<r
zFwqG?a8%D;^TAjL3^f@U)S<->$DfCE(&64mDb5EYp~E4`p?6e_tQSLs<!qkabtC?{
zZ;<|9-z8it`+cKoC#U50)rrCQZLW8Z*=(a<uP>zNu4{zpNzuD`stcWRe<k2O+vX^l
z9wZ|0s%U&x0U;u|vc^`0_kbUQfHDueuj-@qMpQ8asgV#k@sCl}kKaEeX#hh>fXD}C
zRRj6aht`^1)8NuO{JAopuZ}=5-@~3W@>G3zzup|4{iUCvem&dR-fD@ONhBLd;LRno
z<aiLAJvjHe6MX#gXp0S=h!9YzKHJ2+{&Nf#pw<O^;$LfZR{kLdf)29S_|QyJtf(c=
z!<8_lXw|gulhzwKXAAI7d``)VGW_f@LfvZ2^;*&3w&3b)_hfS7<cZ39O6>RDIZ2hK
zjYwe05$QJ5;kH+LHU-g?!U5OY#+#>$Ck&O+;pMh-UJ?s|*Vh}0c?r~7jgyrVF^IPt
z#pWgWd=*jSj}e4Mr<7OrYF5A9)h2s0HokOa-#6o@VUocV7uF^FtQ$v;f*}|++w7Y-
zd1uwQMe541fmCtG-yix^^W2*%7`zeaEu7$OL-n~6qH6ObE+>{HAM1M{Kwzqql+0U2
z27;)r0BLhwe9_038(~}v<<_WopY?otL$7yw-E0)DU!1g9WvZCrLhH)j8l_gE_s$Gf
zwt!!&$*giRQrKP{b|SEESGz!c4Dhk6-=r<wNF4Z}qLj!UoDs2`Sa#pkZ63_xHz&n3
zhklgJv-%L|A<<jm_VPmQ-8}rAv1wVk!+s$cStSa++E`Ep^PbLUL;_WJKU{pF3|BG5
z8K~+me<PuB?n1Q*VKJ|yurg##BazC>3JyA?<xD9XpMUyYM3wV;cD}>{1=_#!x9R`I
zXYjB73G^u+(7Nih&Q%YhbD_a}j#Y7E=DHD=_WO+(EMYPxe{*)Y69K)?VH`le_8bg*
zVFY)kC&EAt4wDr<%sU<q>zR?y&xf;x$KiOW5$ru8;q*kmb>!<x_%q+8{`>y`QsTUx
z2|?JduO6ceHXQQRBv#EwZ!ryvSwvWt)gc^%prspPAdUtFPt(<5f7=u?gVfl(c{%yy
zJe_N$mffJ?Zb;bZsee@;#^s1J==y>)codm@p79vhc>LWk_hP{JL1$m~SgXg+TJZO4
zF92Z@4=oxG_09<Y^Mm<ocnjkkiGT5D09aBoEq*tnwb&c$l3CWodU7OGa2sm#Wl`;B
z-yQ;V@qe>*+nP<{TbG5hj~;$wZ-pG!nTU^>7=wL2W@e0pcz6h)Ry_xl^^{3DQCgph
zDN!R8rWAcXfCb9ET0wToiOa>c_@Z2%E?zho0lR3S4Fb4VqGPBUeh<5UrX?q$DxsNV
zQGG5z_xqGY$l`;K)PnK7RJ6N{x(Khgo%51i?O?MPN;%en8I+l2Zx(AKUyHdH2A2wN
z#l}QiTf&B~En^Od_w!4wc3<ZGcf`_{X>X14{%(@cm`SZtx6M$}y0hPJ(7F|eDIaal
zFgz1Xq7F>NYMg1>yw&5AWbH3(HRceA%YvqUND{ldEc%*ADH<8sIZwTtbW|{u2a!4N
z;Kx0_9A_Y9;<6wTXk}+To#-hLS|;U{Qg^odPC!`CPlWY}Qg%waqt0Fz6>`pmgbfnP
z)@N(`oh4i4n)AZCEQAoOQb$?V%Vg6cke1Z~z?A2i)NQ!Y^=4#_lvjF;<{TCxDW8!P
z%n8gdTFxg<r%2uJo=AWas7=jIG+~uqtz`Fw*526mo%Nh7#}g->IDFOXPN{|HH_L*U
zF<k8&th6#Fk#(=cF={SCUKXn!+)CI2NAc=^RLllu7C$&o=O=pEEn{>fBZ$zt(0}E3
z(f@TL{HwnN7q{EQHs`Q)9h(f#I7gDA+AO$<C0b>>9nkAwH+QvZ!&tU!lEUHQ=ZCd#
zv>;)A#?kLL;$CDRvy7h=nHA8e8tu>B_2dp)GAb|kmveZ`1|^wuVp`8=*;)VV-$DP%
z@1ob6Lwkuh)GMw4>BTH%cTZl{)0q$=)h8CFl#Ey67N@Q2gKJwrMnz{Y>Q#nS0|$y0
zP&NLHvN0Spg9st5b{%7+w#v|=e03ilFopuaTU3*jIg-DLNCC+73{X5F@V|e2$b5ze
z_W7KFKpY9Y<pTCu4pd^i$H+KGvL6Y5>ZeFQ{qgj>A46$UvNq7=yl^_7NYfw(M7$Dg
zt_+EJwuX8cb5ATeHUtpP=MztR;|H(11>mw71T4eC+DU|uRi=N}>^}9a7J66KC9$r?
zZCVy*x3mH+aLUh=Qmj~tk=s@f5iaKw=bVrLttrcr>Dq`AT;A3w!yAJGos$p|?QB~`
zdnLta$U-+xMa0??h#?HOyI1C}r!S?jkLwg7wKldIsI@wP)4A>&OSJY&t){K7U1_Bf
zqOjFY!rHIZE8-X<>$0%d=2f>LnuBbd#@J=c-at*3*;}-N)juP+FU(*4ZBj}$pP>{>
zh}<Yly+}RmqnjnUAsJeo#nGKMhfj^OF$4jhu`;+X!R3-MMzotzRO$O6OB`(j-#Z}&
zmq!B{ZBpbcqU0c{|Cq;@N(Pqmg)n6lm2&+Y=x7jP5F*O?{Dh>1d#SA1$Uf_8%79+m
zJP0MGdGLbJwpUuQLofM6zwLgvEc0}+K+zkmH%cv>o=+?((aKI~mD}x=6JhrFTdhd+
zn<2sY1*-J2fmp_cc51t`-ER!1!CI;lO70G;t+N8`t@8BrOl^(Y?BB8*2EE>{K7-q4
zhikcWD;vvlCQ3#{+3q)DT9BA&b)(h7`SgS$cHnz7Gk@>h5fmHBkz$?<d2frO3br*Z
z_+3s<^t>8V_(9qigCH8vzx65lZ~jf<U;4+;CCpYqD;pxCDhG%b?88{_pfqZ&++W^t
zegBTrVmfK%XeAiD<r(Vi*)k8;(Rv_6gWo+s3i<&zAPi!S_cIv|mIFFRj|q(aoWM~<
zA~Mkmyt?nTIC3F`P_V#b6*wUhJ@}5-kPt|DahFFwi061|8*DX5ecfT{xSs51wbS>z
zj~%sSrQ<;;1hjQTl2!Y`Y+lo79rSMCL0X97Q^`c|4)j;{5#&dLvv!ckVDwM7HophG
zkNa&uFCz4iCWNnJ&De!yCId6FAIazAf4x79oJAgx8aaRh|6{l;<p1U`&_&(0+8E%a
z6m#B-LO!j;yjZ|jou*WxPFPIl#SwJ)mV{}yU0a)pN-c&GEIG0lH_!^vQBgxdL-6vz
z34md!f`W}T0LDN$zb?+eR}5i1ozLw1`xy+E6wL=FB_AARoT!w99EH3lBNW!%#uLXt
z*^Qrs2<y5Uiil5Nd}y_lVhV`okDqM@Lsjm(6>m8SNg}1G2VtFRW3cs~qYbI+U3uAd
z?t3Rk;kuhq)Dp$j2yT}wR`mxjra~O;5n+kuYtTX_W+A7*X<gX1&3F;o39iX+y%SvO
z*S#GPW3VAkBJNipP={k<IEonmFS;Dik~lYuOW6qJLwzKm{g~a;9pMp_kPI=(IlE%a
z`2n7YwBh68ZROyS#^CXDpk2}T*icy*xqm^@@55|@vR#REa$Z<YXUnuCu|B<}w?f&j
zgcQi@nV8$$95Dup58GJW+#vw1?9|=l7Om`bI%zpk%Fe!ff4Nupsx~Hl3<2_r_KFax
zt=NK&5$(k`aj>apQHCK8tm}nR3&mv?x?K@9h^V?eEyifchBHWq2lc+)%~BKL{reXq
z-nqnwwxO+<%t!~j-Qdrsr%xbeG-cnUmHYLD$%K_a+3zz*wgw^DNX}X|ZnvvT?_gOL
zH;`bG=eB*L?3J7Yr}a$Ar|I(@qVXWK?zF0t2v`2o-$DMvkCFezPy3<SP1>Z8vN?K<
zhxTsbE)h@K0?YYA-EXvBNPdt%!uBaTnPwcm%^^;7j!f;n5n6!w5cs{~mw8Bx{+ct;
z6~-81V=IqjXBZbf%!!yUH{fS=Bn3iVXdgrT#W(2REY@O@;MYoh>YDZOl2`m)mA2n)
zEoV#SthHhgtnoKGH6<h@vcrTSWrre%O2p+#>U}*&P*QNSm@#}L2x&b7O1a<AR@uMc
z7}7tm4{sX{ncN86L!B{{95C8e0}A&amw;${Xt{~c_%Ml`_!>Ce1CQkL;GzuJw=yK8
zqg@b2+hD9dRpgr&_}YThYF-*H70a^D0wXPp-Q#AK*4i4{3nEGsMO!hVby^vcO7&SE
z^nqm9{wXeV1-JDL2kCd%)Q<B-+feM-c{=<l%>57EJUt;|2xE*w-y8SWE9r6~w9a}y
z(R<~5%GCR2gzuEO-EJ)FLJYyVKt5%*8}X-<-A3ppL$P8??A3zJx+c0d_Pz7N&t7R=
zIW39PlG*MXF$hb9ki`%PcPl)l@jAw#811#Kc2tz0!rt6p!n3{T`#Lg$V&-x-bT6%E
z205AD+4t%#l1L9qbQjx{5~tHjh=OWx@}9*64d!sDDDv9KRp==qeax;o$jn1(tgTVD
z&Dt(`WszV6_;~nMs2=B1nEhEpMUFok5z#n9R^Y~H2#g}0l*EHGgYB0NZton)ixpfk
zWsVd5zxwgz)C8rno>oH4#%}h^P@7pnwpv+5*k3<KQuYhbDchCH<t<yu2DjA>_8TE2
zBk9GJh*EAAD3|pN5~$ma{l3$BrK!QHDKBU>ql~(5L~#j+nn^?RDowQ*jO|WLIVDc(
z3S!7X*{<}ukz*uCGAN}t&gYAf@?82;N@3qORHC5(xKma%l|#ULp?;IqE88)or+cri
zBM>u7?$rHGt7b`R{`%cny3s1|vJ)imbUrieEQ6C6$3rBN(t=*ZAV5HSrFJOW-8O1o
zp!?$Pot&}_g8O%Wo%BoJCVlh5O_lbzT09!h3e1CiZwL7=uNU_1m0=)c96zfbhAw|X
z<jRy8;yqidWUSUz51f{1H}fnX>{v_^CLT0K6)<|f=4lXkBo&TU!t9g!4eOX(%7%m_
z(%V4Q#`X4s-U8>#voU?U!E?J0m3iHE_NF|YS1#+xcrjSwm&O{Gj`TVQylnMOltf-m
z2xcQ11iHwqUab8B;>>M{_ML{!cHi%w#5i-jeldUXKhO_h@=oklITyuYQ*iyiGpENP
zXKSfL)7ddXe55g+Sc$ndelT^^+eTBK3JmnY%)MNZBGKo53_XEEtUvz@u|sXPSG|3<
zu4hDCRz|08H(J?A%L&x|Kt}bf9a^!$F=H54-I;zV`2I+d^Z7*FtaUZfG=|BR{K(g`
z86CknMq?<PsPc4K4JH;LucqAEc9YJ@Uf^WGY(1SQ`_8g1<Yi&o3Ikf{y_0fw10FHA
z!`fMwWkT$hpqP8>X-$Z9>R!0r3xvpJHDYD28|%7IZ<`luKB*6{W<3F~-Ozv(h3A;r
ziv68~_2cfjIT(&!L~l9RI!Mczr}Yi5z4A5*+x>;x^@UsOtZTNUFA6p;+ScHbGwa#!
zsVajfQA!~Z(BFJV`P~=yHyBcHJsMGtO!Z!Ab*JuyUJaoUlw1=rdeJk!&ma=dx~Ws^
zZf9?ZL>8j~c#CGPhwcNuq&spPyogIk4Ccb>u_E|>s`ZPZieD-jSl2WmwiKgTL~3Pw
zeNT$f354eGw5%t@y#==G=iFXD=k)Y3EGO=_7bI8~n1WEt-TFAo$*Q3HD<PjDCerC*
z<V3OFX{VcxvUke9lLC~oAu&2Uo&c-RYO7{oqs{tDJrnW?LL#1?P1;p=@)GD64+2%T
zy%D97mW7xX&co6&j^3PQ_KHHxNU-NwpLzGJ)u6!r{z~hO_55VM0KmE|ZW>}N^f4Pj
zy+V<AO!69}L`G2Vjr;9JUKUGaQihc2T`g(yBux<aq_%+|l1*Y4V}ulp<as(9dBzWw
zDaf4Kf2*`=YxG*UzrLsLw;9lyvxTj@VJ(K%)mgG$iHBk4b9|>cp-|N&*lLrwG+>t0
zUQK&%Y|L!1mx=CGFd*R%QGwyxKK^ZJ%>@u`fJuVFpZNyquYQly>B$oZyUx8;TCw*J
z`NAhpA0sjO5EN@AnWW1<k72$p&8n1r<95GtIzRDdK!L$6hEw*wXd$G`npRr5)7w6g
zFB=+@X!~_C*F#JnB!XYX$C#kk1wuSp27YHeC^O+tL@ZGle3p?}j(+LTSz*@D^o9oY
zN_x0NT4L)WLBIo%M5H^XWguZQN%7DV6n_4#Rq(Bma$@+6*oUzMuepxjhFn>4kzz8Q
z4TDr`(nE2s%;0~FL0G*tmL+pOEArXZkh-Ww3Y<v{Om|Yob+Xqn)s!U=rbQQOsXQ$U
z`~J!>*oq<2N+B(ae@3fv&!>|~yIbRQUhRH}RiIllr@JJ{<=0lxmfaZ^WB>YoyKmM`
z`PxTsoCvrRVw}?0$JV8dRR={3jTIq@AldHi;yrq6IgZJ#mI97xik3VJTkCA4@%4D&
zshNMr8UpcjL3`zX-zcS;TySKG*qLPaFS4wwNeP>oD-9dq(2ujXxja2OK}rR}<TORX
zaw18lVI=VTwNX<t=SMjP?8Rtp@1tO9lM^Dr&H|gPYQ8c<`Gq5_hA63@xdn|8^?0Bm
z*_;)zq9e@y>3{X(%OL}51dV!qA*ak3K-1br&}iQH63k@c^_9zWwCc(0%(~x5Y2k8y
zwkom}S|5JP78GO7Kp>r-h$)%P?{1m5ZA>9AXF^Oi+@<V@DEHgd4%F6I&PvWFS}&kx
z3#qkIYbOO89y7<(cGC%myqL0}LDR-<yAtk!93!XGiI`TxdPWdxdj(PA@<eHkT5n{G
zZ5$uqP(itzGQ9<2=!i6$bi0ne$qY6Z>lOEH639xc){9Kpg02Kw+32NE?l<aoL(UgM
zv&|Bdc|9b-9Ph%IYBtDqeZsN&vv1J1LhQ<NdLpbRJ6O88IGPML5R%X}K`qo;%$H%`
zNGUnvo}mJqns}?>WLonMi(Qq+LthY=PMI~c*wCm+!f_qHx@i;U#XdJmQxr>LhD>k3
zhojA4h8zbWZD`R|ji?iT=IgYtFNE)_B_L||OI?YAABL?}dMU<4UJ_;7JO){|9wbw-
zwu~ibiPQ62Vhqzw@Nu|KSDW*2zu$e*oV|AtTZ3^xp{gtiCtF3*SDZj7BOyZ#qW3i-
z<3ESX30~nF7KDa~>VQvi<j1Oc`!il)9VEdqYrul$G301y24b+JL`N`EpH1Q%1`$D%
z5P$k3>u(ci#UyLJXS9@Q2xb3`85SPwMtj2`Jtjd?4ao{=v9l~@h&$4ityaoz(*gxE
zEi>iHK)aYI9pE<Hrt7b+i?;&A*86Gb(*=gR=xEc`ZR7s>%89c%wwA*EdgXFCTfrj6
zQZL(%VC2SLj0=&M#T5ls7*_Wxh$))%vMZP8Gt0Vieg86v%(XUBG%MjX+kAxkzLR6(
zz7?9n(<u{xoFeyP%CluPM5Yv*Wzd|+oT9Y}YVXwE*h`~#<tM+f^8Drv=U^spWxvy^
zSso9>(grvw`(DV&?U6^DGbhwrBPMBV)X#3_Cxgv}v1GvAA4kzLk{pGs-uhCXBq?L%
z$?W})_OC72K?r4@En1c41QzLPj+E|rX=;v<Py!e-X>{_AN^$Ly@0s@d;6T{Bl5?6G
z)2{X<V>GErT2`7u2n}hK>+3sCIok#aiB|4ty%VC(AqatxPe@d@+bf}LoG))|6yyDB
ztlgLm`is##1cGImIYl3y<L+WMmB{3Y+m(Ia3CWxYEf}d){%$L51RpbdXby{~Wi@BL
z+Buyrq|;>%2x|R>)GBE?Bgxt%L4@^uHoic)(Ry$Os6&Cn4NyC!6>9IK?4%M=pY@c?
z5UFhR+E8siWM@buA&F1>iiyC=o43Xs4(o~2(_3%77}0P{L+l=C6$^y57KSre{Pr8-
z^T$@55?t=pkK=MelJ@&vEA_Uq-#5y>k(W-&N*A2eXw$u#K8&j#WJEv6Y{5G;=Iy00
z^!KU;wM&{tDvBW;lO*NY_auzX%FxMRGL!~~gwaq5nXe~^+bU<`M^B_*{Fw6nA3`$6
zxVGP&Kx92gZPoR@k=`qHcbGpUOG0CWxX`*GBBhwu#2PcAW%3+GYh)-#dNaac5Ji*q
zYVC804>0mcn%05vz4p~M9qPTKVJKE~t}&cn<JK*R9)@9}UU8}%J^(s2_d|Z_Xptn~
zsPyf5_7+}CPK3y=t2*c4D4W8AssRta`&Nh&s9WKDJ{u7ge;rj<3s~|U#?R82(zI^s
zfhcxH_JKr%h@?O~DawYHtF_qFwEn&6{H*pJWmbAt9d>Y=-!YBPo@f2^^V5@&s4*^(
z#6a8aJc-(=`_ASF-?rU3t&4vT`@8+VQHr6Q;*fY+vQ_!jP}8Z0_oA(o!n&S_B9v{j
zGcs5p7c6;Mm+VT6&XO}v>tfGWIOoJx%$@U`ZT;tDh{U!B?t7tkD5bF^<1Os9vb9F7
zod#AQoHIE_;%PzYyxiVRPydu6d9f;8cB{GvCu6vT8W$pCl4!X9N574Tuq>Gv?HY_E
zX9nE)>sV7}%%q)%))s>zz$lyWH7*>efh{dVv~_S^=4%W)^dq_OjMmn*x{UcDAwE!r
z|LVt=n_8GNhc~iJt3&ILg~<?Yt&k9x6&9A9SeBLhW`wjDf-hpF-*1#!S#knNHk`r_
zUy(pdm4NsJD`SKAT4<y9=b5QWpw-IWsx3|_$W{saPRh$P1Tlv~om1`hT4$?5Yo%^C
z(y|)-+YM6d!K(7Q->I%+ZDmLAuY|myz0!(}N-kLSDrzt%VpVM!`t;uDxOSIDy;loH
z-b2!^)UmEgp_P3SMTYB@Nk=T9DX*XVp@z><XtdTRr-1>xro|RTZ-h@9yxY)^P4^vx
zW#98qk>EDthMer%jn|jY**B8~Wdfx%LRwu?r&g$`8r08G2^tz;1E0(giRtn<svXNY
zhW`3T9QYnKf=vk=d%_Pc&${Ov1FsxS-s1^}9)z4JmDIrU&-@(auYI3>-@T$!@^Z0?
zW8b)McOwjHAZCO3Q(lm~LfW}q-$Q*RNAsY6S`wuef{rPU)&>e52())<G0Ca<Vg$;*
zvE6Rq^BR_%2~nu~-M#$n;2254SM||bapyHXhIXkQ4rv0jCDe~u9x_f(|5crvVr0`H
z5gduEBUE}E(DSt?uy9n`28MYJHBV`i+R8ujHQG}`Z#$L77)CQBgH{PX7;{ByB?@L%
zHIiu2jzn)%_+-_2kOxbeYai_IF%#o3s0ks=kzROUm}XYKF^ps5ef|{r<l{HAUJbcJ
z3CRM8S?dOR?QDIcb>(*3NI8*$&6g21o)joc%FY$2J_BSS<VadP4pfuRio({$Y!)@U
zZB^DK`?N&6Whq5YIkD|^4rhxoa5=AbuD8xgws_#@br3UOBimU}_Db*8J`K@Y3B9Wk
z)3vbOU+J$m!qW*9L)6r*=2IZWz>-XgmviQJ-zj^co)%(jNEB+dGdzaK)@=e{LZ-^V
z5P{JeVXm9;-;pF(!Sn%2FP(?mIUhqcG+SSN(6Ano48&`oBM~;D?ZZ=iuT0}O5r*#S
z$34DOG7w`zN9H1dm;z<H(=h1swrwV}z#P+LS!mrV_TH>&S}t!uJNx}g+c#d`zvt=s
z32=g35Na#)%eH-^126AiI9)E2aox)1J!iZ4R;eWmk};YVUOO7iv8q&KC&z4`Y2P=x
zBos9-?t-y3MXANI+U5M@^2Z1)l<Rw|I%A^OPVE&f)~8PavdiV_ZlAw3%got#C#ERH
zgW{+dTPgJYZl8ItbYNL+Txe|&#ncobI8Vd<1qSqFmHI-+E8F|4Rg@tiq4=C^rHT-j
zvrUe?Us-?UchLX&e@yz*KSTUeKS_Hw$TFlnA$B`l%u2d!cXl7`siHLJ5iAP(W`-SE
zf)Q?_8BOW%0thmD{j=)Q{s5cGoWCFs->rLwE@3dmnbq20-ukvVUT+`;4245Gx`Qt3
zn|8R`iGSz~;o}AUosHTGX8dtv;wpic`^LJ$qMh?<t)!rL_WO<7>kDg6ppA?YdZ(5v
zWrt<GSQ6G6+ui&thUy|Y5>fXZZ9bMdF6_0iZ+GwKc8Uj|S_@4-I8wfHkKQVEOyWeQ
z2@7Hzf_aaC5yWLwKI1j|kr-Ig>3u50$1Sn=k;`O7*m3&yv3ES_HA;)2ga+$+Wxoa=
zuWawREGw3**umY^O<i!i>+06N)+ED7!QmJvhIk9meCbU^p>s@atv++)L1K&-7Up(n
z>)a+<TA*~AgL~v#%l3561nJamLxZ(V*L7umy0G7FsM^^V^Mcey%!z<-yKjJb9*py*
zi5cLu=8moBGm5y&rX@rY0wKXk&XX)XW>6sZ*=n<dcnz>DiLF#ZFcjA~6jrhCj*J8e
z6IF=WP|{%uFJjish_ytD_Bqx!hRD0?3tB6@Y=p~-RKE|if*x%9OjIk>uCG_Nt4(}O
z+3RVsbE}s|+iVb?kv4~FUk8B1kqGd65(F!d`v)zaaYl{FwD#BDHq(PDIe;t64_r{I
z0=>;;k0S{9_ZvobBe}KL`+Z0Tx_fd5U|RT!*to^}4FpCfZNOSN9+S%Gjk4Y7m~CV!
zn-9CN3VJVvxL)XP=pyxQq%T!2Pc|F8c~)%w?Z8%xw=Ob-L|RTr%7`Rt+qu8KvTud;
zWaEEp**(h*h#KLrmP*bGd)bj-21TtlR17w5WuG^B-_3Xk9CDYA`R?oNf%a|&M;Zd#
zwwW?PmA2dK42cnHOxE5zd%aG1BdyW)JNM0;7&UrjeOj#o3U1NaOoN+3q!gQWwr!i4
zhO8EFq9YTnY*dHxdaLBWUeRBBPx-(8E$SbB!}%9}68Xpf5Try;5&gu-oXv+Tjm?h)
zO0&$hRwL$Ip5M~Cx(u(Gq{i($hnCj1gr&Pp`{X#7Kf$Q%N1!+aWRF`b#^~*kqq;Ms
z&8so|^_+n}Z<;g4LtZ^Fj0B>Q|M)l1?|qk;7P`Zty;YXeiRFCZZEa|;l>009_r_L+
z4twF`HpY7`{@&)O214yOV%pj7cgE0?z1bMm(^KSrz0oDI_*xo4s;g4#?v}AGH^U|u
zew8002BW)JmEmA#)xNc}P$u^T@B{EbNW;YV?Z#`jN=dzaFcS?Kyb}vuNI0NcJxpPy
zHEbUvqumHd9{47$6K6FXC+bwU%`B`JZ+VXQ)NZop`Q6-80UHsn!Get)VQRzBoTzrc
zZ?sy?hTOmPSj&b=_#q#~Q358SH(CEGT(37y%Zd&KQZfyDjFDQ*w;*RF&<F};tL)Xx
zv-~AV$w`MXvaT!3$vg&Z<S)>=eby9N&L`?_vh5%;NuMJ`+S$-JugUlZUeI;mw(X=C
zh*8Nod%|ZVTBTFmq-R|cIoqVfy}E?blZGCItyfx8ZucACzrW{OAAKEpTH$_U|FoFb
zir94=NMwqE5aHSR6#KT>>n1ysxBJfZ*FL5Fogb1s>35MZv(G<+zo&0Hf9L=CbNhoL
z;34UqD1lmbcUWs%$}pbVbs$q9!O_jxie;|%(Th<3c^*FBk9&Mc2f~%uBfadNodk%{
zSdX5ywPxcnE%|rXErq&Ons&;|3(MtVfuMEP(`pvBD(rh<NflMkkjsv-?R{)#*-_8L
z_hLQp<c1qI!ZA@*Np$k%nT&vw6A>}lh4=TcW5TR;_c9M`Gd^;Xi5rM$R~w<}$w=s(
zWf?y923O&qFQmm7^=h;3^Z9H9H0_YAGT!&$!Qe@W$lR=@DevCDBS&G)arUPs<JJj2
z-97MMuQ&GA$YS+IM39L23dBI^o!8Ib+3=ANDW&oHawW<;&d)F8A13ned`|gmpV9u$
zzeXz?T+Zkp{u-?nw(BeV>*wTE;g`S8ZQn^h`w3w&V&i=;oO+)X&a_KyotzVa=4Kff
z#H&uEYh!wsAq9wqh=B;iWNm|C7G!K~=Jmqo71-W#|AGF{CsSpZ3=LGk$~rf0BA_(n
zC(rb{)0%=-8g8ECa-Qaaey^1*0V#L3QrNbgysiv0NISC?iHj>woGI=BD#c8482IJ=
zdt$P2+dX6>MqO{zeVZN)>tgcO+x1GyEAcDdCWv`4a+a5~vv)r37o&YMenRHIh#Ui(
z2An?B7uG{RJOb37IIFSYboYods5Y-$o%H(1VeoU@3}*scZ>-7s>FRy}Z7()S4l#M`
z==*v!Xf9_CgK@5kmSVh@MNmfv@Bkk?I4<kLvx0`o`>j!0HA1kr3lK|WbRZ;t{nWh&
zI^qe%%Kd)h>CGD-T$Uy=a8#gsDa4>GOJu7Z5oO<Z_N{O}uZU!JDsRp!X<4Q>Q`tAR
z>lH*;)`j(SnymO?O4Mtll%3n{ZsU>_7B|YNwYl@*o%?P<e>nctRynP!A!*%aF$^c&
zaW143x%;3Yr5gb-M^Y57TjjPD{@!o>0|a<0mFw+uY>t4it_$nQ#;?~}3{ee%uvi<p
z)I!;JLjnA}FfLSgVsCffoS9iNFU$ZQMq>K#l)0iu5i|brk<0WW`^5;h<KL`xHcZY`
zo#^eeY8WejCdBg%2L|z@+wmXz_;Q1aUS4_n=q<&~fXcQR%Q__|k3~|B#0+Byi86XD
z*o^NGeY9~wbEGs2nnRb{?l-DKVMfNY%u<3P%ZYVe>9rBv8dNbiCuv5C2`Tgb-4A`N
zu=5zy5)mZOiW*tOk&9l7&qS!iWhF|W@4H!EHZwyyoorfTSLJj%F*p;}uW8mZ^F=PW
z!9`9sKGXx`{dTwS-IP<##3dtXC8d?tcB>{FU1+7yYIVdT&eli>(LUGR5QdlnPnWYv
ze#GFTQtPb7V~8XX^c1*XUwC?RCa=NR!@KhR&#urmzV)r2psBK|*j(=Gop-<X_i1%Q
zfLb=&jQ`KSN59^A{mD1P^THLS|H+@A{z9Ss^e42p9(0#c>5KQhWHP?R;Wrz|6CaZ3
zK8>AJ2L$TUJipg4bm(4j$sklQ)eG7m2CH<n0~D<HtZJ>G*&nuz+-qUq%}&)?5D>z!
zlC{2l5*wRc_stH{XrqRESBQzcE;f53W!`=I10!**3o!;;v)$KcUP$?bmW{gIiHqIu
z?e+@2QPqu(Fr%sUe6b$s7o17{2lWx$Pa-(kIL@lbDw?4nncR=Z;4L%EGwxfEK%>%1
zb^b^?dZzXa@2`f3TRprpoU|I42-B#<6=v1BwQ4&3b)5;hDUM8Hww^DnXETS=UJqld
z5RFH)yH%}yAd~_Uk-Sc1L~OB<*OSR>EsbDng?+bSZz&q5N5OeTe$57d;vtkMSb|=X
zD|IZ9-u8XIeplso-^ihpE-TGjoTc_zwU@mj>w>CHg4Mm5N!z{=Qe<6EET`4lS-b6<
z{IpQFjg%wzy>Q<fm-Cr-uQyjD_&}u5*298sOvcXN_D(H#WA3YxR9UjkgV^`Nw)-4|
zoLoXZOvvDzBWsStV2RH6zyF(Tw@=-R*;*gVl33Q01>g{TXp)^DDZ3MH8^mQJaklTi
zC+EnLllSkfMK#X1S%f^acI;+*bwANblaCL6lX}Z%tV<Z=!BN@!S%@c-t`uObo6+jB
zPjIv$28ZP_N%(?~FGnDZHeNUgG%Vnm=5%vMJ=_;MnBR2o?5}r|%V@KSeZlnntvB<b
zz8k^sa>_`r)Y>?`{fH3Fnz2+H7<5`P%lXMahoVV53vZLLd3$-GY2*B4&)c`1+95^r
zNKmzkxwc9jI(0dMtXAt&oz5q=y|4zQI0r#OAatv?Wx$DGXhVuz&TkyLQTA4;)eeWU
z7s8TR&lfv?Y|Z(ks)&l+W|KK88X|dJX~ibyO~$mE_kR?jMk5u~Zqp1UCqg=*t&vX<
z>CSq#1(6b*FVDo~G&k8tPba?iQy*CypqUG!+^_7nS3=5^?Mk^{$?J*B(=&UieE!+z
z>|5peht@}bd4Hw->JKO%{TBV>3-X-NoY?;KPqBRCiEv&BfBX|OMp?|o@X^Mzxog`n
zSy6@!tCM4TC%E_dV2L+97_ws?_%MhXFkSZpe;hj6)u6R*wxw=U$I!I*<}iPVG{vEL
z97O`eW^I)FjZzH}tKQlQLC{|4rK81s29l3w?$s0m_uF-vxZsl!Z`T)MtHfGJ(GEO~
z$>*u#^Dop;FIDPSoHH;Ius&2JVm*0NiO3`$2T5kxtX6O+W~<tJoz;7v6m3IyY@KxP
z$NK@;N9WZJP`qL@<xx3!Cz2d_I1Y0`=)d|Ye0pd3`kAyWE;B2ZguDKKIiJXj5p(Ny
zn_l-L`PFU-O&i8p%?bwtC9<&#oI{kB1%hz)Pw;;0$le}4x6UA-BA!F}88mz^P`V~w
zt{ZQjPSmn<x;*p#-8(Zy%86E!Nd`q|Dr|f4Nrq-BbU-6^54PJCmNfYfF-F$QWhw~v
z-3GBO=d%?ttsw+nt{dm)x17%l_ub5l&X)_<>y_5jq{4f1(lc}J!CtG|vG=J`vsns|
zLS)-3A%$scJcc)IrBa)z6_yx{%Of4Sp%R9~Eu@^;w!0xrF}Ot9W(L&SczwMgA$pra
zX#bz@u{s<dLzptnaki?fd~9jjnuN#n%+|Na(fT>SuLIIY4wOMX9(=%q5}63uoQjD0
z(fJQS1z@&Mf1fx58*O)*qgV2^HeckC(z}ncO^{ZkRo0Wa^_BZ8_s>5y6QOK;gM4}-
z<`uAU$nW1>2}>r|jzkNLi1nVf``uB6NZoHpx2*S83bmMqwN?v`mfZ>(8!6ernq3Er
zz?euUVhr0gFXRx|wn7Y<x@|tb*e+tNg<IK);+DO@kOtPKKp;h>UT?Hoh&iHNxxd~x
zJwLfaqE!J3ULhOo*lw=||AX}@yJi3D`9kt(VA>465)tBJRf@Du6suG;WO`f(>uT+Y
zdM8Kgdr6E$RdTB)l~ic^-DXT&KQ+S)?Z(ZhV$#Q)_{P^iVY{u!a$-3@^At0ekG{dy
zy1{{c=i|3m_SY9AXZCxczIP?Y|Lxad$@CQAt&NS2fASmDi#bKA`b+>I{_ID{cb-gP
z@|+3h!}s57AzSR#4LTU)GM%O)7C0WD)5%l7oCjU29)cxasM81}8mTl!ihC|}wb896
zMPd?|idzmx33iR#_n^1#JQU+WXjh(}EqMYvh+3<Z>dtAsvz48^p3S@AUdS1wH==ZU
z5tj4CeCD;A@rmo&zsiqNH-aqQ(&^aI)6s`}Y=LofI|Jb6LlPNe!DHfILqbBr2=HxI
zUAEaAzGqaAI^Sb3t>=1?Fpg|PtVGbQBTbzk=jfb3^>whWtBq7gZ8EA82)jeS!Lyse
znKWB&J2QfqDViPf289>##K&#Qdo`mkgM-I-WD6ca#B%{X%zVNK4yO0H-|sv>ov5|2
z-EUmZXKGHyzSs94a4!vB?|kzUYhkTfxfQtIE43GvC89N=C3>!;JUu@d>ABf}tr!x^
zlM%abpMFZpk@qhfwJYEH#y8mSS0GSo<D<9FmQ<B$GI6n_yE`XI#7KoHikS`dM(cK-
zySEn-#N@vokV~mYpj4Ar=WIxW4OvoRgf)-pn0C$YoB5@PR7>d8?+<W$y|b<>r*$QV
zL|T$r7008S4<5Ar&%-2bwxb5&k3;n_D|qz571N*Z<ZN+cHsfenF=Xo`5yEH*;mN{y
z?Qt?bbSGwFFi^K2|M*hLK=te`MH?GDG~u;1_FBy+TROeo$qwf(muKsJot{YNXX3Kb
z_Z@m+Is2I3R*CDG<;};em$ydP>jkZqa<h2|r}HyqzoFF@!*V_&F_W8;Rz+JSL}&W$
z8@JmX5wYMMg!^sd`tr)MEY7~qrt=Sh`}O8b(qx+Y)(PEyzI8h|`U6zmy7|cWW|@1y
zYLpPgnBvh4JC>(6Xe;d7W|FiR5wWb;8vB*=n>XZI4I(bP5$mEI7Nh5Rf2oi%;q+wH
zYVVYGH@7hj+^?@_*a*vokRnTpsJP>mlkw`9iEz3Q^Ge&UP>n&{6eMIDMZKJDL50Bj
z^cFciqhW!NSki@M9H4cliw*L)-)=mg7MdqPqLt7EwhG_hK$P<DeP(c@k3%&#(m*;b
z#MOdv_(#7+eX|mO?k)WE6Fq}|v^b<2oIqG+mfn2*NnmQBN7ZKY6e19i-b^CZ+MMpy
zyGfQ>-H0)eC3)*>_xB2<ZjNPUn|-F<Un$oY6tN*%y|LtFnwTJo_-x)3QE%4vQ8ndD
z7xEi3b&;@fyWd$Z&z${wZrjZhoPf4!Yq9J|faP-Le*a>_M*hG)T0@gnYARw}k^$F)
z@r<<7V<WTypIKtu>c$*FF3c8DN~Rua`)v+EJ2)kSu|2AEd#~O*P9iYwfR2hw=FvV5
z(!6Arb#aEaA)|?GZOmk7Lf`Lbb%L?c!J#X4^YO)DaA<rG6H4!|Ha8_^#63g;X3g8Y
z{Wx6eY(9NA8-<y?h_yz%1vwI(kqDL6x!%m$cE4|&7i(87OX7}MvujuGZTE?GhC1dD
zcx{DJ%^2-|xA_Xa6*E+ek(ZY%ITeT4m5;yn2{A-o-o2yjrh2(vuYCO*-{7<^)Y~gL
zM>EY)L)+GrVK2mRhptGJ>636ipZWZ~yC&93_s<fOoxL`+$%1Y7UVd$zPA6hK4-v7E
z$IU!tQVOg~c9aj6)5?9@t(dEY>;2A~H_r&dniJb*ZieZ4XZ_EA$e46EIAS;oFug1$
zsHARd;_3lgOf%?@KY}X<{8v*wS)4f}#Og(nsUaS;Q%7qQM+ri-gxRgVM<U`&#Yvez
z{_*A71VLsG)!r$vm5s9P<i%zxhm>thY);hu&h_&j*v(v4GuOG_W`<sB1&M4g?^&0{
zCUjljLpNowL4*dyTQ%(j#opPjH(I+=w~f=~L{qhifNFZu=Qq!mWW?k>cj4)yPb@Ij
zozQJma%tWC6A*)aV<JWK6$k`!43uIJvt$d<F)i+?ALa!VI?KuU48b9$oK9@7?}>T9
zfC|m}(5>`2&Ukx$u}U_rtWTd1=<L^bXlumNGjV;Ql^bQdQPt3ikRsddW}ge6RW8O+
zXsyp`qV1I!>|hH)iFHTTGr|I5#2TWp?DzYPuAQa=aY5n&!6r5a8Bnee)fl|GTu3qT
zCMBfrRB=maQG|xNL5ZrFFi|VZvd|^ciJU+B$R>M*!2R_dFMstnX=-T4c6&un0sWab
zq&1QL{C6PRXNW)ZmT-zz$LaWPCZkMCo&%>s9p)i!`r_kFgUes3jq?BgEn`2|N>?*N
zL6t>!>g|PkHz={ilUoWqa^b=2TCc396F~y6*B6qiO?$iB0IC=Q`S}ypxNv*@9QF6#
zOChDq?e+87k3PS7OU?@~ukVN<A<GGM_&LU}_SC{*EM$jiKz*oBr(pxW9Qml79*X}+
zK>9H9v0sMy5Ll~d!su!(xfo3k9&jnChUrx>D)s@lI!uVVxy(@NpwLCg!8r!4nqtal
zhQxRz+1=YXgVfhY#TKl(uca|(aCSS++rF8t?Rr7uf(C=CC5Tz;)@lU@2VvHR5ZQY_
z_J(JoGB{9O*~?BipV_w^MLC~O)(Wj=fOI;oenwc2{CY35(tcX3C3@W&r77>PjWq?%
zOC`pNgv{sfZ-k(H`&-{cLSVn&DEB*EEcv)RJ@NeZ4TzE#J5TTTjnW!tcRB3uuOCPw
zTWf?ESk5PwQrWLp<7qWRKlWP5(R%lN+h<z|v3V#qanypV4XyJodu^P~Pc!E9UI}U8
z?Z+SU^8P*9l>_fSf6wJ&`}llX*emo~rS6R-J3`wHtr&LMzWC<vt!^BKQ@+k*B%a4U
z7>oHp3VbcX@kHTa-5x(N-_yUlN-%`Xe@I(Fom>vy+W3N7^h+iK2CJ58P+3Uk$fS!{
zrL0g@$)~d!Y$(+GHb(#lZ8WVAG7+J?UeRD=zT4{yd09<s+B@pr&?2y$Pp-2+algN!
zrJyMxV&w4FE9wx_=n*6M_~Ju%zVDW;YptBmCns_=4_=;$HB0ld_qPq*I3ruK%41m=
z7N<ksI~*cH3^c47Ng+btNJ%W92AhsY=xFWa(@MU)q269NJw37Sg}9y_A=pfF+->*;
zDbcr1YlY$A-c^Ztp;ns&DIt(gXX<{p{#37)W%fqhD|uZ>DX`yOka{QOg({IQiJVXL
z{R$zvrdg3>#32dA$M_o2GiZli%)Cg#)U`^;R^_(_eau8ibl1#_L`vTX*+#yq3jKDc
zuI9}^?>0>8{A?A3M7C`s|3>ArAAFbcm;WBgQIGVUH-se+q7XkhBmd-gOgaQ;4Akd^
z_=}%V+D=dAb<hMPiX&Qfn3TWz+wh-#A9^I@Go_d%Y}qzXE&?H#WHkufuP>Bh($W+o
z>2xvUmTgD0QB?@Fu<eC_5*KTMwOV|9w)HAm0&m`a#QEtBw-~%dv$MT?#yuocDm3MM
zzVLK;LKW`YR~s3KDLY4Ikf$7uZ+2)6L#;591^~oi@W+ZDpDc;VINgh(wf4YAaJXLw
zjA~@9F<ZcR9`?Cx{z04We`A)0reJG4#F=o6e;d>K@YggKyTic9rp=APW99_f+u=5c
zLb=^&U5KmFWQ^j)h;rUy>a8(+wvv;H)sX@(FlW!)_yUd;bf{hV{QWDJ%R&&FQ__7J
z>*;)^*2Z3TRE67KdGmBKcwL3Qna7Ft#8!+4)i%Wp%{HFa%*UU+1v;%3w%d)`Dy>@b
zyPg(==E$*0ftx$TE~`xz>?pM<Pp8DXE?loS&X)^a&FYwC;db8;0&BMO!6cE1^SaR0
z%#ETr5~`;5+C52Mk{g@h?V=cM7Q(g@yONd@=ktl%^@@nFm&*R`)kxl5&_4Jn7J$7m
zGiWQmc$g^UkT%=X^*9%YIS_q5|Mxx)DL!%}d!C*i+yQl2<KR4n<cYw8@zz8F)EzbD
zzvMlC@8io!3&#}*!Swu|y$mh89ke#avDI#!!F(d0PPDFO^s_h4Z!VPWMw0HMHv`Lh
zCS9J;u+XZx?s2~nV<twEC5?)Bk5^~Lf_CNgb~BHDHY4<9bNaHZ&9b@}BbUn)(i^4h
zs2XV{`0$Q(wQ4`4<nDWlxHFiXE|5F)-C^qPa}#`QsUJSA85~%u{rxb_i6NqCg`A0P
zM`gFU<lEgOni9w<v)!)jw=1XfnYgZ2t={i$G@>MDmiJcK_l?u#iJT&>6>`YX%)kf<
zBu~Janjb^iH}-8O<^+3TyIqa&)9s*6F%Z|Yw=&!oxmCCqpXR58F^oywyh#n(jAEuz
zJ=koD7;HM_a2X6S+pwwAg*bpoZ8j&qHu{>$=V#KgPbbN*ef);FeBIR=%5C2%+h+Z=
z-@al~&J>Z)2!Hc4?oM7>)(kn4zIJiRlW)+ESK^=iCi0u_kzfDZp#0a)5>`qp#AJtk
z7O{jc7&oQt*56Ng<)n@C;t-%ixNqKmM7doN8u@e~rW0kqQtq$hyi!V`mW^`1+jR&w
z%%qi_w%>_4qur!(ttw|n8MgbCm+LF}d`2aH<wqD>-_j>-0<~^VzH*0590>?!f+N#?
zaijY@cadXg)p*T;4s<YSYY6n_6+S>~1aE1ow~JyJBdBdgi3Mx-Pzj?wk!_7SAx;3C
z)U_-YU~MkSXhSpxOMDPNM}MCfz~_&&=2~)h2<=^oiw!^PY9?M2_QwK$s|tJc)|M-9
z`n(9t>39B>B?$M;WYN!WE?jRnqRYQi%Err(z_MgYu_RU<(Nct`u%HC`z0*>lw8p+y
zTCJQG<88gZT&ZPeT^9CI*!P{B7T-%d^pjl&cd4zdv+ac^4|02{)HYfNfpuLhQCSws
z{+_U|Y^8F$ZKx*Brv(IW4PyQVYUffEtSD8bcUOJ%=H38ih}XPjQ13h33wbH5my3~s
zw>t>De}5$<bIXm%(8lh`X4m%6=RXL_<8e5qV)t5PN}a7zADotlJu!cFBm-ftq1Hyi
z=Y+)}hwt4CnA9b=3a#xR-XizpVRB3`^5PeHe5qvM<#ywII&(Umi9UG`Rg)DayP5af
z)&22}@RQdIJ!Nz&HmqXFW+j=j`H=6`Tn!hOvAn(mx@Ctckg}5q%bk6{8=)td`9yDp
zn6ml%OJH~FS7{13BpieFrM$Z6KC4<{L~Ap`V^bS_tU|rrZ6UcXH?C*jD3KPQsA^^-
zsS8rAAKqI#dR5J)nME<S^6BysVZGS+TGxu2mYwNEfIAi1L1bPCU6q^zL_5R=^e;48
z7uL%g(z-&kfga+UUE_+TlNqR%&EUfj2&ALE69hC-da*g?(GwRR!KosI1R(|J)|RNn
zcpgFUAss-iHqk9urmZ1bFAh7{yF;uYs2zUMB&3$4WukRWW`Yw#u=(U7)VkAl>!W@B
zqc=W26!09q>^rSCYOUn+g5EaTwo_$68?^5i0-d&3KKt|?EeP@-{(Y7xoYqW^0S$q-
zA6?LxNvDg+LGO1u%_^86<W%_V(@)7cae01%n9zMUlIiJ0dHLK;R)m+^r<_kG>eyN$
z&NmQbF%Ni6COPe8Xa8)6V1rVg-h9mZ{4uxppP?vuT`6Tupln}w*vKE$M|DTgAd!$3
zw3%kxnDX68o}gIr7raLsLYxOF5~e#PFmPhPp4(feXhW7nARszyzh$;|x{T^h{Mx$I
z)umwF&kFg&AJgAvJ1@H0fUwQR{k9lrYO>ql7U`{zVMjt&pE9RrN6REH>ij$*nn6{}
z3$^T?K&bmu*k_C>(%h?}*Gh=#gV*Zn1lq3i8X#66-L^_f3$L$N-acQ5K}cyOro>(v
z`Ltpw)IQx#!QPXu<S67QY~6y2@rvvoeEzc0+s-$?@rK$e@7`Tasn8*4XAeT%S9&Y7
z+SzV*H{I$qSN)XQrngH-nR8yPu<FY5^Aj&Gubfw#%xY$7m7Jm%7iMI4UKffh6^t(;
ztR7#MWg(@&ecvgqaat0!?kI-7yj*X*QDt2gUf;c%QQNYzZG{jzr!|pMqYn=gW9?1Y
z-`1Z_bOsOf!CPQ>5-<*3n<zb60iMu?0}-(GJqWH|{Uh_e1-t&!Tn!s+H-_y=u%TXN
zRFoJ4-=?R*7kGTRHi1iUTC4P}`)Ep&`Dhm#p%^mhVpVM|JNK8*3D-aqp)ZMg-{_?h
z<H%qPO$fcwsxj+rVM9nO6(LA6soQol_$_6|w83S`B!qguQcDM{4>_e-w++?KdOopU
zo~hN62qYpgVe_Ee{x+G#a_F#ccUyGb5|=8Op3ab0WHEcpSOY1T+;0Thm?ANVLwQ0>
z*1J!Bh^atK$$In6Hn_w@S2MwoXs~%0<60cjj42XxrZ*u9gyrN5Gx>FoW?Iu5ZM#v$
z#tWyEiDK@6lWDC&K0ULX-<a{pX{DFKcD)gj9Y#SrLPTPq?)OPfk&w*$K`nui=%b}e
zC5RhN#bhdx-rYIR{Qz1)MM=>rvxbj;4bc(@3GDj~0n&Q$9$!PH88K4=TiNaNr^LFh
zHiIEW-hQ0uyA6A~?}hXEiMnq<<MQS$z1t9=T6UHkiCDmiA@ctEV)sZ-c0IMS-EQQR
zC>`E@^mZC&<!qkjDJSZ-+W;*QwraA|cdu8T-hN_-qXb&JnWLP`mwK`Fz2?Yg?>;B3
zX96c~x2p}u%9(OMsx;h8tMe7-4ETXv>rN6lhKpI%rPg3GMTRQL+in9URXrHk;?8pn
zf>j9QywY2DxZmEd7M~i&C@KPFABHvtuNpbb30~rA7l4n>=CWCuk<|8La|U{+?Yp%J
z+*oQ%R1IQAF*r9uea{{CN!G5c&OZ=U>WkW__Qg~;tu^Nz1jdk~kuYhiZmEs)c!v4X
zNR<1CPoQwS7v7#1mTW4QoHHuXq<PtfW7&+D#_LwNoHM6IEE&kcg-BJXLD=bN6Z2lF
zjU@&?`}|7pZs)#t=r`V+S8~p5*H`kA%rq-Ts!v$3){r4CImcPZ<kecirR+%WJUu@V
zz=zA_If!muv+;qv_=;Xk#AHcXZH<_Q?RO&e&h5T)a!1zQ%uC_EHNN}1pYit16AF}4
z*{ezMjLEN@*32HP#S((fAU@~54AQ2Et>1pAUGRG}k`zK_s6m>WM2RNf7QSX$`v3*A
z8*TAXQBz8`F_%J7tq6{WG1^dOs@MJOM2B(OkAHlrWMGh=EGTtKHP|U;`@7K-D@|#w
zlk-Y^O0>N=)DjI^j*+641uoH~B-$&euwKrH+CZV(^$N&@{aeVC`zwJB34zO!(fwu&
z{G7?_nNm%K(7W}D*SF8~Qml{O3g<ObikhXR-2zVM)i#={O~0zu3;}%jh`J0W+;)0!
zQkp6$iaX?mS!Ku=9v#hQc7J_`1RMF>ooo_9pjB%pMC4!^w@&owYi4NEXEu%)8@}#4
zONxYEsKW)#NieNek|>w6@in>v!I;{+4^mpPElN8K8hyJWOCm4XJkpEJ8VE7cdYz*}
z!$_Eh1EP;ZRV-;!t=#W7Ph^y~-Rz(ZV%Jd&DUe>RQP7XE#H!6nL><ZNG+#1}tkA0x
zR=XFMY%PdZMtcUL8miD!;0K@n5G9b$C-R#o^uF;N6SY^;a&hRclLnHaov-D$HkUkf
zUSDo#N}Sh~&?{uqQ?Hk^c~iJ!+<n`K%Y{-aDcY#jkQW=<y(CK})s!;tUOp$sYzAGx
zwr_M2Pj5dW=Y?(GNv(2!d56{w>#d%wowM1P*R-r(W~=^zdxW%*R!d+_HfqE%RfF({
zMjgEFX^jvdrDX=}-iM>01NSmm{kA_+&QNX5)Mi$~y;gF{KqSNk36*_sgkXD4oJ2by
zd2<v^gn#0vtX){#uuM$_Bq0za8t<hyTCvt4ZX#m=_83-FoKG-Jms&e`1?H!3ZG>Jc
zwKN-q7L3=@+_zv_E2|C<of%jsK@FW%=$M`;2<Owv{eI_kI#ZjHmSE)6usPCVlL7bE
zD6O-o1(!8jo8<Xq&ZkKtwJ2Rp>RDUk<yuU-mxQ((v;XtHSwhy&<bC0O-*|eu@Wb!F
z<N3`qXs6c7winyG(ONKOdQ@0WCvLA-6nK97hV6c(?S-(~cM2hpFs|0RSb)D>?;z1k
zg=FF7<%&(5+-a@x?q#!<-en~O+^@xC@xS}MPg$0QtrmnpiitG`o-YeeEhDFe{r!s#
zFJi#|_8p8r66~Cqq&{08A5{73sEyjatZlb76q2o*-fXZ`YxG{c^<$G$hjG^AB8)_V
zfkuRxOxQlKBC|P?C|FYYMIK)&8Cag)T87<>JZQF#mH#hw|MDwanx^+*zwf$MMC^S|
zR%TUKceBZ6H^rtXinK_|vPH-+G|<R`0RtLXhCQ$aPyEMt<goz*8hBvC0xWqV3NRs=
z7OjC`&?F`CCiach-IZDAvUkK<-_?WXU2CUCT{$SAGr(q5R%V{FPeiQmd*A2kX5s$v
zXW#?Td&aWh@<O^?D32FL-S87aT6is$Q5v@|-pLNsDq&p=Zt-B3`8iSPk0Yy1PDFo&
z5O}<9lu}@LF3a_-Paad*_t9z|Td@M8;Hlf4^3v!&GJ=}7{L}-DbA+5(o}SfWRSPa>
zZMVJQN2iU>XdBV3%sH)nPMQGy0XT<;IR+T4H%2?{I^vcDf2Lqus!fbz4$9+|5EE%#
zZMbj*q_*zXc$e3eeHToAaQlj;)I~gx0G_UPzCPZl3~v|nY(*PQt@Mu{i79i4SLRR%
z&lsI*p)CsdIEsdJ_?(3=2Tini>i%H1qIlk@Pp|E-#GHkI9MQ6^SVgnP1+Qkw5Kfi8
z8_l9rWdWT|jfdEtrA0vxW1EO>nv?bE`Lr<hs$ZkGY6F-qq1HMxOw#p9LrS?z;D={f
zq=gAy?QS7Pd|sHNb4`J|6)25hx(W5p7nbW2>2lM?GAoxQ^6A4TMs3_~w^P&=^M!o5
zv3>eTToy_x=6TM>PlFp%UhFwYIWZYRA->12VeoiNEA{cJ&B$pp_dzL;ZB%*%AF{<p
zIEEKu;7nk6Y*NEeg6I!h;9E$I)9uIz@q9Lq5EthcoY%15dt>&U7=og+gFYko8$ZDR
z>K9DorrX}|bIAFU0zP?B9M;`2$D4q#*G8+lSX>CF*sr5Nv&p14If+J>tyj?sYFi9W
zO?+TT15Ok)oI*)|>J1ctr7ctM^I3Om=gg;<SKi$&gxI()>-oH<C9%DBOF8vC_R4E%
zQq=ok0%5o=fz~|N6jXyg2Bmgh9!JK95RhQ!Ua5_T1K9EGTyI%)*OcXZf586O*qhQe
z4~yne`P^Hh>~@hQ&$2F5r@|%GRvU`bDy>zcKO%Wa8dQ!nz%e-jsbX;4w?ZkE>mql<
zeXByVacrd-imF)S%Nn_?iQF?|T_kj$M+lkDz|*bg%sOh?Cg&X8)i-niyVrvKdqeLb
zR**I46g6v+&W9io4n(L~+0Y>x^thlkR_VImx<d}ctu-5J!<#Vi=XiamVPL&nRQFpp
z#x9HC=nbC~mzkDDS=AHz96UaJBhJ`(m7FvF8I@Id4@oIx%z=+m#=8kFSiNP{c9?Q9
z7*3tqSUcg^ANZUYA>zZrY?V?K13GNGoxms?%@W7jgncjg1j%av$WKqq(b@OH<UuQy
zIU22s7mQiL#b|2!lO6$j?RamKJHXTwcWljYI21T9TW1%4wEX=;7!_h7<Q1P*ZN0U~
zlII48{CU0Jh%qsRu}dU3hPg5<JH5Si9t`{a(z4*0q-a*TG3gD_5?(c`Qb~KuT?bH>
zI71%Zn3A#;^i8E$w>uy|euymB3$vTe=Mdl67c`~i{PH>*lAIUJKP=CkdBAb=d>>O@
zPC-^5DhZF53vjq^_3zYu!+EF3XhS8&p+_0>q-WfebxbsZ1W41g6?Ih64=KRR97B=N
zh{9;#QewI4_s2o~{Ge|Sb9*Gl_R8#qxZapp*vXQ}W1RWo>vaLA12#FY?D~`hvAz_V
z37)18xEPt+_j?%dmJv~pz%n%O1g{dN5F%;0ih{HNnlYvdo$$nE!8;w;7;HYwZmg5l
zgk?V1lju)7>0tq6&NpfxW@cO&V*yKl^cm7+lKv-uh4AhjVNPO-EbBtQziOa34qwx`
zwAw6iXq?Dhvw)#wG-y95k#JHNg7yXZ$0Xrx8?r_^qZvc+i~#|sgjC~>C2Zpy&IKt?
z`|}xUfa=*_w?R&U=lAaf)OYdAt!mgvNfhJWx&{4U8X`|3;(XwGQ9ep<m3`avevZfz
zRW8@(wC2Xny;io;*h=NPEX3&8Mp2MZ*?Fb2-FJ1D9Q_?lR^rww6HnO`e?HeJ4M06E
z*MK)$XNb{+bozcB%kj4Dr<#5u8u`MZr*Le0fiWlpwyGHA;5<vr<lwnn7NUv1!wlrS
zur8|=Y?(6vpg>>0U``C^u`|m8%B=OZ_sXG-Qf5pzKT~z+=yRWFU<e^F&I|j{o!Ix}
zppy@A?O}^H4aekIF!jmzE|f}x!7%#b7Z5sB_l>Po#(1;lg&@J(dVLaHZxC`X(gk$m
zSVryaFQ3hR<EYzI>}YE&S)7(*NQ~wKkH-T*K}KD<{YswFBpri?%jHBTMt`OCN(!Fk
zdNamd<r4=qA7UWng*cI@a1H`e3-47zl-Zhk5iU35tfj_3em+_C6fN8F^r<b;qfn;l
z1AbX?P72{*hXfo0!c-=|_o#;T;EXUh)Sv2C1x`R4x*Sv1_PoEo5c5imnX*Y#sWJ=5
z>qW3oZ@Ngv#Cc7Tb$jr#KLAHu76~uI8}C@GASA6T(RqSJXm+Pa;AXuChs{dcirf&b
z?u4c^u&Jp{seyA?$VsEUex&wJ%8T~<pd>%3Xw3#Qg`oHdu2m_0{oB_xq@?u8rjB`M
z58x5bp-I|=YR)mdC#Q=ZWAhoSnQ)l>OTV^+pPr}ZXH%p%de?&(ybW<7vfNZnT<@>c
zeaFQ}e0rzBd@4WOFAMaDiwo0N<>xbKb*GhTZX6BJIcK8RhVbdbCvLYZIYq_1hDcsz
z&7A%_J*t1=UWavXbT#_Lm^egz2P~S>4Oa`><E}%NNvng>Cu~K*IZ@HgKD^}*gb-{K
z)WH}+d}}7O=sF*~ef}B<TA#S+@nJc|R{-uezaaj#U*odIGaq2g&J3PWH548Vbgd|P
zaCR$J8_GbL1BKvNx@8QJHbn#2j<5gU_~qW4>4pY5XI{5gVpO85E*)>N>)ncS!heS%
z(52*TtHe+Op=U}M*3da7#O$d1PAwJZ9oOqc!waay?D62)U4cmmfo#1jy;Zu$#Nk#W
zJOX9+Y_(DPz~?}0)w(sN`Vgk_6H<(nZBk05)<F)A6djNI!!#Ig2xBigv-ii2pV)$<
zXC(QoEj=ZCj5JpWUF!RNuk59mC_0?+(b0RhU3O=ICZ5EkhrqDz9M9z4C9|%Xm;y)4
zu=kOaGo_kXZ`!D3`l~TyrFIUE_Equu_U|9-%4k+;?+GDsRJ$J(u=CP0;51l8r$w=q
zUFUa*#I#y|Mde6m4PFIwvX+Xv4Kbd`SFh!~U;en)cRFD{ef>41?!0??q8DdjJy0JT
zX}y>LV$fS*ScK{vll%Qe_O%oU%StVkm#;stURKsimM8{wMz_%z{&);ICn%qe%<@K`
z7WC1Tei#mT)h6cF>d7-k2Uug==pDT5IsJ&Abe^_1=tEiCb=O6kvP5oOR!Qe1aBEGM
z#K(`HNGWrDeuB~IQ=FR*nUJnJ+-9fmuY}<3un_D#FdN9|r(Nh!fA;;(9GzM!_fH?r
z5FJ?<B`TW+o6Z3#A77BB4L)TYk@EVPIR`Oi#t8IQNNJ^(oyX(BI>Cj(<@Rhyhu(u2
zBR(AQoDy%&*;uSYpgOB2?p>78bSI)S++bR4us`$PMcuS6=SEE7EVnw8;@)AEu_u<o
zz~k~NUhQ-sc$|q3iPBmed|JpCrc)|nA6+%m&f|g6<Rg#X!IW54^o_&kg#hHdh;rC>
zOLm*2H4|r&?aR_Pk1lzF$61=;90R|s&}11UW?YbcF2>08j$!qvA!qaddWIKAIlP92
zce9gLb>91sa17koX?6VEBA`F1uhz`AXBS`_Q=e^^H0uO$!Qn#&HnSHi0koDz8@rVD
z>jF9Bh7Q7mPY*s2f-)iS0p~0g^1!a5vLa5xy$`bxOVq1p(vkkgpT<R_`j{&2YN|bN
zRXUxctVn#YBPI!L$2=mVC0d-lGyGd{5LoYpeTNRq?x>a7r-p`Mym$zB*C2Rui1deC
zOucnN9QYW8l3`24u~Clxt>wj0n`(d+3<s6Twr#3)kAm1sdDzeG>0S1{lq715*<3Lp
zFh@A^4#pUaQmpI6%D84Hj6~^^tyW_6?6na?V2nzu7UG2CwkCp8fm`)n3+B4A)dx#T
zre(-H?hkUd2zXmS4jY~i?i~8+UM+{<ErWnL*_&<u-qYu#R^1gT2A-^W)_b_*$eJA2
zRi$GgN9JRgg-@T?Ar5uU8@g+#fN4S;hviVV&JdCYU_(JwgqF_GqO)JX3q5qsgoR#0
z{5iV~v3BnHx4<HOorHinue=vGbq9FwO>bZnnfv*o%)iqxaC`rb>u3_N8CFqCWqT-#
zx|BjI>ZeZ2LcT1lxA$~D+cY1D%gXeX$JZbE^oQRNljjFtJn@4szIe0I*ded>swPGI
z=-|ZJ#uV|kRrEf@J2{x9;}}3V`Rz$ULOkc9!XSLWB~>x^QcaPb2vL`AYn}c6LQGjA
zx6$a%<8#7!^^GSqVKa5mElX1Qh)r~}RRkUCu<|i-bOpcxYaMWw!cwpaxOIgnEw{^?
zXk||IZHGj(JOH@|%yQ%+>FGUwy(s6NK)rufSAy3j<w79kM7mrLgd2|{Fr)W$FF|B~
z{lu^?gLr$7?}hsMff#~%7rYL-R?2A)>s<(d_{_lQ+L)wlH33;l+6YHGzo<;(9-XtU
zj==DXLX~jmdo|lo<7@ydsyQSjJ^DFtbI93o(VyQrFT1Emp*}}LolobbG!*bIbYV2w
zzAJa#!I5Y<r$kQ7z?ce8i6KfOF~xPyrg&E8p*BHtG5Yg=&&kB76TCMu<`h4=T%Q#H
z_4<jN7vkk6fn2NjrsO;4JTv9*@~Hladu`iJ^!^mj&fb|`-uclc1k0*-fs~ZV;k{?Q
zc>H5O71$v~LSFR!nZ~{1G(7qQJ_!zYq96|qK^yj&`alT5s)spbSBi&yPQ-ut$5v&&
z(d)+81&aqC7~WI&LMesO<z_l+nGd+z8EU4NJM}es*d%%_#d<5UCq~Zmr;n-biy&JY
zj?;1By`#0k>{CgI^T0mmN!1-OwI^ooBqH8xqqWMqE_{A@In(Jb*Q@qRuL^ZH8~44F
zQd3i7&`_n@&U#Ml9E-uTm!04p>k?Vl$motsfR|0S-VpSjm0D@TQOqH-=16dkS}LV=
zdM|kADE+~sRMr%^<Rsg0O1$1TQ9!Nn6s{6hj-esJM4H8{iZKS-nB2EQ>rIJ;<EVYM
zOM9Br%E_JJnXGFfFG)QFIcurfmEmwe@W(bT7+-XYbBl&{XZ7t2{0!g%A!ZJTiqJ1D
zTK2WDlBQ{8T|yD2(=srpRTevUI<5|P)o6;5_Fjvcj|mr(#Lb2->*M!oNOive0pYT)
z;`)#6-^qs`YbSX1Tj!h%R#EW=4uv}y$Kgija#{J&Fa40$Paou8bxOO_=qe6(D+(^^
z0m4vR<us1Xbo-=b)rB#XA~=t2&uel#B0g*H1g9ujab}K?)QC877D6-(hL7U<oTpm5
z;=u*YR@D^fJ?pw~yI#+W?64;)V5FLwxzhIs$-<1L!%zk}Dxw|mV|0A)gdFi}rdNsK
z`Z2gTCr1Rwmnc!orYKdz3B8+4%VEF{q!8s8n4>9Wz=`>a?f_vY&9Vrl3oOe5d12U}
zeNNPBIGWkECZVe`TFPnr9sqqt{Bj|#%08G~?PK06JuL)<IgJqT%avX>M!8Ej5VJi1
zjlr;}+RAAAIXL(b&rAseg!OpZc69Q&y)%1-kc2E)gs6*2kDJqA;T?>+s|RBqzFiHL
z&Z!^3;f!*a;*|lWIwTwtff6QRP57z7l$hi;8J*g7h`qkv>0@%a-gId=@yNY5+~d_^
zM{Ovog<y%6?WoFKSNiu{KnPGu5h~XzeoUs<mNiU&lNTwh)h}Te&vf`z0cOt937u0$
z1kr?5h7Psha4F#b(vO(l#J7eX92x%OAN~wJTXzN$DSPK2I^w_m=kQ;ycqi0w_6AK!
zsEXB|_~=O^(velP_htfTiB6Bf9l$v~hX=HL1QfMlL*|$SvmX|#81~Me(ZCL8!@dNL
z`u&6w`Cvgb)3#O{4(1pqMQWh<d5zs)@7!NE(jq!HCA-*D=6Z!Yn?)Fh(>f<Ru}2*W
zgBb(fCCfJngy_gIlcJ|L<qYKDNY}*{tAS_GhG(CRr%Pr@LP|=NPN5$f5xdZPA3`+x
z;IWI=n@q;0@S~^eijVy)YIELG`yfn3lOJJYZA>!mw0KUVU6+(tbHGgq$d-7pE?Sz7
z=2UZe&B>9Q6XL?TZM$f_0Wimo(rOgW#(Ba{Lmw79x1<Bn(v*QCIBPIyDzS1xa?j}W
z5N)WP(v%$85{3>O)WopQd(3XSfw#oUpXc@65(tM>i+xTANg**oo^9v#<x`1I!VTqA
zx2nrhu(G!Aalhkd=lSVIT(5%TO=aor;!Uf=H(g>AC!eN8Odbbn*><@I#=zMF;=RIk
zv>iCRWGuH_*sVm3E$Sc$j?t&MW0$DquF3I|2zY$R@&>npbAkFvx`$=GK!)Y|Oe;IB
zY_u_Ohnf{kEe}2_7MU>wiAH@9Yink&5*O9?Zm@pm=q&HGf_LP+>JT)@D&|Zp^%Q50
zF{tf9tGYl#jEq|C_uQ#dvB@qig!Kwop81@(d01P-vj-cNK7IW)&Igvu1;1QPpRr5W
z=qx$zut|<V#zTnunL~nEPFQwyjV<-%lLmt!C&cIjnw5VHak>-etw3+O$ennxw5;Uy
zDk>(gz_5SD#8|<a@TT-2bAL4vA~u0neOG43r>x;;a&pRw%0={o6{C1-j8yb?5cFG%
z21`^#<Ax|8?>%u0#;){1f4R|WCA+Ms=bSAsL7(CNc%`>yEti30se2&?X0U;z$xV@#
z^}9T(f8t&a#k^``iE=@uL`(}li+7DNg7ZWd%#NzfDMne^z%iBp|MHJ;|H2Q5f9Xfm
zb|g`nnAA<e7Z-~xKC%jQaHimyp9}o{3-fRN4)dRVghx|#`LDmj|J5JjW`$Ov6((K!
zmoBt&-%Z$N8(23HV+fvzqjpgX;kbB!Ug{BKD`Wsc?yc*YnXS|Jorx#MWJ-Ex9J#jZ
zV3`x%J43T%oAeegBTia$3+{-pJ%X4g_r384-`q(amK@pl$`9T@3z`nW&NnM!^PWek
zEL-7rRSW9u7;VVu>X6RCbA5hN`rlA%BXe@|N>=nv&vHtEPp_5R<-(eE9&e=?$`-g@
z6hif=O&KcA^Ej5tF<5gvMWto0+=0AC(*<a7aQ1oBD&Bm}ivo{!@%h1ff_G*|9Hi(i
z#$8!32a<v_I>PBMoSg@1J&k91l<b?w`{MC&1sby#6LxAD)IuteHg@LmErX~4c8R{F
zKbYY9Ilwxn`{5LZIui_ri3E-A6bpJ6Ab6dfh(4Nr!u$o~4EVH|Cl|r0WoMLmipad#
z`4oE@*vsVOhYx)Cd}mo!e*CMy!g{$;dEng;51TTz6zY8kCr3hvL2Y8aOAzw>o_u}A
zJIDRgM=8UniFj(o$B2)De#Vr)bXnJPVxCNi*1V^gz#@1Bm$)IX?+gJK6TXSBav`1^
zHO@0i!9`8u?V<9Rck7RIfgI%&L6FmERli40c$bO(2q$R<Z!2E8m3k^)Ie6ruA)@RO
ze)KWO%R;n`wYR~n^{`g!3u$~|aQa+E7v#NfJL|eKTw;0{<w1_l3N;2P3Xa<Ecg1gp
zq41xwGp53bob&pA8dT*%px4h31tF%iK-n3O2d=!(L&wL3F~AKA%5e@KgNb1r{Y+@<
zrw4WpYfr=;4BK?hjm&y6ob#mB<GM;?M%Ur#aEwxvW#FA{=abQkLANGmbzTX0lXGx9
zoC}sUhU`+qy4gkJXwa0F#hYL!c!v*Cf-|}zjzchun~CCFCfiowQZ$}XpOgN1(Kj6j
zDKpD1I-w6*sifs1VOszBpuRt;ub3_dr&^iWgc&?;T?hkO-5K7jTZd-Cd)e-oq6M=j
z{BQqz^k09+-Wqj6OH#t)6kyeQXS_bn+VGqLi+#_R6mhF({Khl>&-_I)4*u~sxQ~VK
z%NP9WVKk__nh2SOf{KfE6GoF&u=PqU3I*b<)D=Z(5E3CcaD#@<_?)y9#1z5k$k805
zc=F|<i>-IYP(H<+Y8X6Xme(co7`sr?pl8UPsf3(XY`x|=2v^fqtx4f#Yfj{>@SdDj
z#vDq84JuSChQcbsc;LN79v}R)Eh}kV4ZRW7BS_sdYNNHr<sx*T)WLNPTykJKDE0XL
z9P4WK?GUI&5r+_H^9|bHYg03<X>R)H)ar4*;&7Cr(yuX;*x9U3I(W}s6b-+mfU(KS
z?J&wSMxicqV!=<}683q_HmILt&St@O03Pp;rEg<)qz7>E<nsrHE?O?YbUFvT14gv~
zvNy*7b28inf6&I5`{)f}8nYW9@D{nP{0&1Yk8|x?JsEGb)gSYWeuwAz4CkP$YkQ1A
zTrSl8fg6LAR(Zi)Fe_T)`JH3EUa7T`+?0syDY9>`^ljty!#Afob(%13e|;rp<DRjK
z#wu?5x-OKW!6Bw-VnN$B988-wyNOA{Y=j(XeM)(c)cMVs^$l4<MjO=E7k!Cxk{x4q
zYe}3+KinQHYbJ(g4k1gg^2<vEf+q8!i7z-Qm}{+43kx%6jKTKUfXNagb#LP1g5%}$
zXVD&Bb;v$BqBoaAT5<76m(nGsHbDn;dfBv1;raaOBi_TdL%LjWA(Aeuz8Fn?<ZWnL
z&e!Xy?DkBRj*Nlx(S{_8kF8y9j=T_WH)fSUGQ~i+uEZmNV|b>)?;#r6U>BYi@os3%
zF6b*JLe2{0+3u$>N}Jwc!@)fL66N(?zr`bG+K~0BRf+hP+Y>&n^r?$F9RVOQpWPFH
zJy1hR?_SUC=cwm)-m@+%E?7Q9N*0%_-)q!jKd82%qg_s3VyUC)gWv|U?iyMqxH-rk
zrk{+a!W9Pi=;@`<W|Jkg?siaomnZd4-YYE2871t@LBi<j2Z<T-#UkBpBRldqj*Wjz
z+z+qxUwfvbJb=gjAsq)+EAFI?X=mtIAbRudJ6IN__^AMFvgX9)&;5Wk2@O*bj8RWx
z(8^AEJXA!4zAK_2p><t0bu=qAgN`m;p<7NO+UNz}I&VU0Es4=5=2@zwq(e~;or%t`
zQ)q~xfiavLe`}5DJV}S2iP)_2*kJ7=*hqOLR{0&}WZAZz^sKIv*ZV6V6eat>{oZs&
zcA*76iklBdHiHE$g_so%)`YGc-_e{id9g&r+E{bucJ-$lDfmPij%&)qpxlpbue7Fr
z_So1TMGk;Dy3xtXs~`j(kHW-pyWNP%5XEViH%b)5c~X`uBnHpZ?IKr2YxLppU9K>1
zodU;ui*fTuw9g?fan8>5kabC$xKK{6X3nT^E@>{P$j^tPt0kTE_2{Cunf?6Po#jw?
zv~)2|@5Dhf9cY3HG3~55DX`u+go)G8f`j7ud0yXX7`T7@pvf6U>SAISe|^S{LC6`V
zLUQ)a;(R+?E}5t2t03voXxl5k6_!Pt*`X$$pl*P&5#vg}J~NqYub)}3H}bO3n*@wO
z8ZgrB*)Hq5B5{2nU9QB-lk9MxzGiF<mTb|w)fDsAbcLSIkR+b-A<>=6clKTFfqN0$
zA7kL@dSQA`c{~{VD@RpnOj4lM(uh%Odv8KdoP*({LY5^c;oOig1byT7ZKL(UwryN5
zEBPq7DV2Q=mZ!<(y|%<r3VyxdmuDTCOz1^EqAok0_pm<QIC83KmDVeJ>nwTER%>7M
zBTaD3LC3Zm8>qHVAIT|EEx>18qEZZxV4qfTF<8jclwM#cI<_B{n{mKk$03sFvnc8C
zKsO8t@Xo}2-44T;p=;|NvoU*Aq^u1z`YsB+>Ah3-joH*bxIDj8tg&;{`miC(W99Jn
zYnXMEX#>wjUh)YQj;{2|!%4uDm)C=(32q$WIUxp#faZjt4W-*)$3b|_VVaGJPl-OI
zSjSI6vd)pRXJRdXW8$5l(C_oh#Z{2~0joL~-K?NadFqGn{v3RC2I}ye6ZbEDMGcOZ
z*9WCFOZO8RSZiZiP|xVylKvdkdrH)Me}5d@dj;oty>C2USHkm3Oh;IeIaj(lW4d_e
z2@;rDJgu|*h+s*By@|udd1jPgv5zk39*=Z*j%1k^W2%c_Fkw1MYx-;B=?@+~d~2y2
zmOaqBJKvtOP0_Sm2a<KZ?>0~z+;vQuX7)hmye!;PsLHw~Uhfao{-+B#dQY#-Ao-(?
z+!XpT37NCOdOA@c(R)=*4+$Q79~L~M#K<`fns@WOM_e7O%fh?c^~@<5y)x$1uV+!q
ztxtO0Mf;3w$f~2(s`sV|m3Cet)!;q1>p}{Fm^|xxrEEJP29|X(CrEg+4bDS<Wd9v~
z2UA*sYIH#`hF%bTbQz`!^@KKR7z^H>Va#FW4mU(#S^_3h*$E0u>h@l0Fx3ZW=Z-rD
z)q^-c&<g86FrC~HI^_talaOxV=XrgnVPMR+`c|B#(+}VLl=XV$c6+8aC32Ng*|(|`
zr^~9gHU}w5FyVq{b{(uCM1~g>7kmH<0&?g}nDR=_-XN(URi%l<oaOb;)@5Irf_D3M
z7hl&0pT7A>2#!3JX6Iv)$j|8_^4?qfm{5y8$LT%xL5lZmAIA0hJ;^&B+kXC8<-wH}
zAQPDR%AA!P4`I}|!$~?RC&5u8>VIz%$mQ$^T|&&1l5u1b-|c&4iAfW3+4RMIybzbc
zJUFE(&b5|})*F}Gg|uWqHl^sHY>vm!--n<F6V8a#5aOF=!F{W5Aw<^O&Ax1yfTa`s
zRNJFLl&v=UoaD<Yn@b(EF0ZurLIjSjTWai}2i`<L#{}z~<u?T5>21=*KHs@-wxMXl
zo7>K}!(E(%$LlL)+lb3bh)F?0Hi^?T`PawB2#MtfKf>Xt_fOnEeqgjFC>L!SVI(dV
zC+*O{bsW|p8&e7%H-h~+q>(u6Omli|z`>}V(WDuf&Ga|n^qIS^2;%j2({sYWbp^qE
zpI^Z{MO#mI!q*`NT4|<_X_iG0&d(L#?0T`koCw+L-E=k1Ho`90{HTsStbYRy`8Y>+
zj=Bp*#h}y2zVi|$;hKpq5at^-;Js(PE*vR>&Ow`lT6K{-=g3h59bgwgw2b?1D1i9>
z7{xbx?{U$yZ|48@qT%|O)Fzm_v_c!ArgDt5X<97r<P7qn2co5kYI{D)bPVx2H^ak*
zCmOQU#<cRGeN#bFh=FxosZALM%d*f9?M~Nv@RqM|<a5X|bU^byM&hvm8Rc-id{EkA
z%sSsSe=Cvl$Yq#wvMyRu>JWnT#nZ}GBrwD>`ir47DF@kx`#4d6T17n!JBQjFtX_$O
z(OZOi3|ybCCKy*_ytAL*1yP7TCJhKb`K;|cB)l^_>8P~!AcV*f>ZbSGpLa&Lq{a}E
zK?|N3P4hE{X#(8a#gAtOquZIMKH^MU<Zbwqkkj-tqai#>j)ZSFA2L<cH>ObhpXc@6
z1cX!J3`)Qf#~WiLpx_MeJl%Ox&V-c6K|IX9ZIpedhs_$xI<@SCq@g01PcpA7cqNbJ
z<-&5gGMI0QP2-LU3GdeP0rWNthm?tHWP5$#@zbB;mIZ&iS%P7smCDoeGo#OQem=xS
z+K!Ds9!+t#mPHVpJiy6>I|G9{yOT<NYD@Y2M_;o(Kl5~5)rW3klx4ZGI>-LFv+uGc
z9TmSwE}(pdQc1TPAx8F39~3l^Gx#WN#iq8fkTPX^u-{+l27CG7Ny|!1iL_oxS)m->
zIb3fRXX}V*QFb`8IXP3<yC4SzbC~j9p-1boSdMqK+SLS3O^ZhqsSAPOBQr$}ewS70
zkgdy$Z`(8o%8Yrz=Y_F9pm%923?*^B+XXClM>s5>bMo@(qu@IqPLxFzuU#r<?~OC_
zxQKW`^!a(Z2^R-Po$aJEMu6Ux?@%^!mVdkN;x6%lV4ii~t0{J^O4o8Cq;f|*ZyAO@
zINUT2++t((fa;>tmM88KBqpu35M!W^$>Zgf<$6Wp5Px!j-`7`vd?2hBeRmEG3YNxM
zl_@~<%o&+3YH6~V)nT9U7)@=ZW7uUcN^9cG*5%H_^yJL_aYoodIgVCAonv!)ss-8v
z@72T!q|2f(nW=t*VH;LRlVw8c^9|Jmao%kTULRIyV}j5agoNxx^qY5H%LEu@QK=@<
z8`JS9QrRzu<*#?9)iB54XfYuU+IV#D@yUXgu<uvbCBZ-z%Q1D@PX%2KRC64Hy(bo_
z3{XnLxk;^q5S`Jmc1a<zwF|QMN>CW(26J-mq?|Zn*$<Ad6mxEs##RQ)bya~;t<<Sl
z<Ymdq_i&CeWyRf!XdP!y-T<A~L*;7+e6*KhLc`8$X|%4?*4w(Uq-c~X99n`gH0UTs
zc@mbx$FHs4Pw4;X1JiN|EUQ`?Cx2YT#@QdJW_Q$H)Jx(cN4cAiiy<Z5K>oMk)rI($
zU2@PmP8L{a^sr+}uf?Q~D(j-13*H<yy;WQr#Ju8sBoJ`tnehui2E23ZwL@CL4RT5n
zUG>J}^C!^+0CSK`Oy@_Z?;CSWYHwh|I5`WlO^qtR8FLczVim{Hwn5IajsrRl$);fw
z&S-MfKOABtt}CBEz2Zv4FALLB{Q}}(bZ01-VOpf+5z8A!*XHXzA+5Bsn{Z_3we75z
ziy$-KSyLip{kr??tZO9Dn4{3EfjM(fw+AWcLJRJ#bYkD-hL~lry}UA+tj|w+Pp2Y_
zo&7v*R|$~dD?ypBFE8xdt_RwCg4KDZW#uWpBRa2#)XJ0`o)jIyAaNHLn8R#`;4H<^
zc}L%Oj@p=xwv6ZkrET;+$VptBcghjy6G(~R0{yY6mes^SAt*6mj7~2PMsK9+6T{+m
z<>YYIrO-(skggX&|D%x*7ySJ6q!6fDm0RHh%@5LgA*DpNhdvsLGMhNGZ*zjOyH0Qc
zm$J}{&)+cWPK*J+NPjS_GTB)%Q!hnzv;aPuut?!A8dR0w*lQ)GNM3|4^`@F_2WF3*
z;59TIjfb;$E7Q@}c)i|~aWDj{Ixp|>_xV-YU_Kt!__&a>HmkX2j>zh6mzj5>ZnOaj
zT^Y&4U;Qz)2uV5+pw$9tYHc(Veu;uG;VBY{d5a_Oy$h`w9p@t{1fDOMH7DjAMDOsI
z1>9t88>8+9#}BjhbsH#^Pv2}fJ%+n}2x57Im_sy3eM0Y(HalDG_Vc0^!=ZDm_DN|P
zR=Obtq6-{rO4Kmv(^CE{@lwN(XYj3rL+|kO4gS`k>?}*gv)Lu*TO+3nV~WxkW1Iy^
zrB*IUXT({R8-wL?jI*l#h&(@Sh$eP(pvAAredU8^nXs4uU<?>xuvoo0)AU}>;j6cJ
zWOG#w@%~y&2KV0C+k+`D8w1!%Rnur6?6r|?!KlZ87epm)ebe1_h&hciIl|4g6u$kL
zk3S<W*fOM$Bo+5XMw2yt_G+#yEv=*Z6Yc6IGCy9Y-EYXME#a(&ejlSy1$aXVEK|T)
zhQV=%oORiwGeESa{T|LiFfs1W^ZIU@fyaZ(<;v)tC3$iT?AvY)bGo#SpJWV6^Yan~
zsxZcy-cwe6Esn?Q9hh8hxATFCc`*ye;USkW$J?7V%}R3cDuR`@WxxLH1I|sBHBn20
z=5SfT8Lf9*bHuE;-&U(q{o+N?ZY!2*XHcf^3?_M5ZA*;QvWxq&%;)~PF>tK6C(`8t
z16nPL`wS)~nHq%JP!itUUIgX%fJ+NLcCPM5t%b*@kECT~S){lQDN*{)hd=zO*$Ssr
z$tc>hjX|3bgA&}9+l?3_pFezJ&kutdgDQnNxTk5`@WTXA)?t7rnBu)QO~Apt&(oQc
zqNsJS4$J>>o$}aen6<B&ohEoZrofLvP(oPoqf_^da(|H+$A?od5`tNL?7h~Y*iZB(
zFDkTJs<!CXI(fYkmn%6ZMo~aa%$eRgWnzefW6o~T!v1s@jG0(2Q5_U1$pU8IvtR;N
zZz|<EWQ}bGPfvT=f(f|B@Um^Di$ZZyls+IWddTOb_g#ExQmbSuMWIhqeHRL_F(H;a
zY(r9D&+mDKalXF8z%s(J;w-w{0R(-Q!Wd`P{IKnMPEl~>DNvoX0pE63MDIx{a9I+M
zZRfESJZkIfN6^e64m)7fsl-grBuCE@?X2;ti}qtMda>cPQcG2%ropyjqCM3+qR-UU
z48oODX-SFL55;^iUR+Q<A+VRm)-?p+1eN!xKXdKVT3;v5iEB<V66fHuWcJe7J3O}C
zG!D||98t8#W_eICC*o)P_HpTqseXdnsy5ALRA+4+htLCD5PCX>Md0fFYuj#=+qZ;m
z=i;VdCT}P!&T+Y3l|^F1L)j&M#02AtkJgzp!nZC=;Cj7)gMHggf1!)N)=u@{?mJ7C
z4}UiuN%WqOlRZoG6t05Y8Gsecsi(!e^ZK+!O9=4%RJ4Hkfa?txGA@Wlb<=AYJBO{P
zXLYo<U(ac=I^AaW{h5Xo3UPRgV%Oh~Hu&4!=5;0>zt{0bPn6-}rcu3SKO9q84-!Cn
zb07SC6ytXq24aegQE?sWn6yn_hCU~bAWyADbWZ1}+k<Aj^s+9byjVM)V4hkk&Z$g7
zmz5x1P2TN%BWu?~A;tbQ#jO*>5*1I|CrlrhIWzZ-m=;<aggMzhzu==X2)qkK7pS$c
z?S&Z=DKAhfeiXs$KC+jM_D~dN&KG?sL)M=V%*Nq7<f}r8Z2RC8?+jvGl<8fq<`~E+
zk(SJIS*d;U`f`U>@j-zZb)Uw!IVA(ejDJo{=Xia+<D4^CQ{3I{`H5tH`Ge2z`;PZ2
z_*pL(LH@>Lod?>O%(^jxC&uNBmUb@SW+z`RX05|Se1d52n=%IGW5Q_MF8LJ9P2lqU
z4(FU5p62&Q6wqPi3ts&S-2_qIpmK%p#8V6d@@y+#!wysPUVFEYqqZ3OPTeY_Y%0!r
z1boW2Sq?kI-k`-J8yr)W2Oo@gmYYB}LZgdHGhviLqeDaF;4MblAp~K)t&FfJy=-(w
zeK6WMEkHg5X2{GCaX6xzjJj!19XhCVNa`dxE*S5>|1?2LnP`!{I^aBq`vML=)SqHK
zKd0#}U`|K>&G(e6$8{yowc1aipb`QdF$6B_DoV0zV|Q2)%&{FiM{wfrbM&mH4T&MJ
z<VbLxu}P%cYoYCxW}NgPFAtVyQCruffuh1)2zG8s^d*gi2GZIZeLA*nXDgkD`Qe@O
zGz;{a6RN5B^|l0i?=<``InjoShqhhs>l}kVE$aDvC#}5bcp0+lz15WWVeO`R|L+gW
z-AAdM%d!xHmouV^Z$Cbpy6rGKjd^ap0~I=@Ww8ZIL-jEF<*`c$m@iuLQjGR|hO=zQ
zx6a~|TxBK_sg|3tq=cW7*RdHbci80|d#iXa0cNe0B}H!Qq6H{8mK?cUS5cK!R_<s;
zcs_Uw_KEi0@TXqIwa)klUn_^f&I!j6M?RS&i5{n#HXxXl*v_frXPkrHHVqXa%Nb{%
zhfcLaxN4sp^BhK`ud*2XlyOe6&V4j#4s4OL+D}4QbjD6O;f9GWPyXkRd3`qlA;>DS
ztaR^bkBvD5ON8XXQUGh&AtmDVMk@uHB6HWLcYvcjUIbrHaW<`0UEq1OV3$U1m6S5A
z?us1t;z0UP0@pC>+j6<+e=EB#f)I(zg?N1;T}=nj#esRJl&jAlKG0gB516$;E#wdg
z$r`4dBRX$Mc#XR6YUT6t15f7&A*n($MC!H^hQdR9@I)V&F^ht*K}NaRx^PE%;IrqN
zGOu-`)k<74F(&eQ0j3I2Ce*qULgccp<QPuQxi=j;>$;MbRe~p@3lxG>sE-$BQ(%ts
z5s!H5Rwa(|%VL|ew=f>jH9lmRO<ChfO^T!LjB=;#n;v$rrmc@<a!H9iC7SB3T9Tu~
zg?!p`Myu5FnXy;Gl8HGhkHI=A4lX~Yqy(@-5H`*KKc>WgLty9`<!QgZ4<5g)nusl=
z$W3djbJ8q~-bu^lycD*w(QAXcsoTL#F4v2=xhfYxAH&{1arw>%#or&3doX=~d<2S#
zts7;#!%<g?QxVP_wC#nkta32aT^al#oHW@qH05Y|JgAo1zTd;ZVRv$fR>B6;-Vf!q
zJK^|wd>^J0Rpck`fB37|WPnup5r#y-b7-q<eJc5L0@E3Ns3SI5iklCf5IpOeN#2Q4
zXpMADM2|8y+AhWVocJY?LeM}k^=q9IiyNo)uJwZ-meACPw_vgRF1~!PotLeijd{Tb
zyi-`v5U_dXP&m6EG%yFzXF0Kq!KF|36xm9D8^nk9MQ2n9?now-p2C}B<`_EgY9SRA
z-&wK-59PBQ>4OJ6u5cou4{|!hD0scsVsQGKhUXA3mWPVZTrllSpZMfS5$^Xp%c?5)
z)&^zYxm*_A2Nw05lBKVjur9?ws~w;8y!4@g(FYGr&tveO6g)9_E~{dXgYoeVp%XdV
zIXni=Xb~`RxwGf_ph1qB_yb*;!zidX3L(bBCTcDUKh4Dz7+v+mK4ekR2+2`>@<Aqb
zLv)nD*+(_n;Wz^uvuKD`bhG98lugJ&$oQ}*zhg{Mk+$?{5dH-S2ty3SRc(1Lc-mfQ
zyDr*c5Oi8EEbHy`pZefPd1bUh+wP1p*&nZz+DI|t#w0K5a#z5N`r$`cuf4uZ3cNXb
z>U~TE=TGnV0r`o;b)?IUba`UAS)J)r>fCIN(RNzffiAAfM?vGg>kEE(fuqWeTG`n5
z$H~(V=k<R$D3yG5kd1FL(aIR+1Z@@P9LsvKgRaYKTMD&RYpjEKxu^!3fRCR2<;5T*
zCj~E<z$u)t^c?w(moxFt0lRwd$;(Rj2J@wbkgrf)X%AITE8xeguGId_vri@gw_KU?
z1&?|Vh#BT2FOBDSH(=25+>5MmAx5GX3c&32*70eDq3B-kpgRo&trc9WxMK?#O|W0n
zrWY;FnQ6fr=y214zzKP|DG}#k<vUOhHD_9cXD@P{Ol^n3;%U9()KbJ4Y26WW!lg{l
zi|H6NY%M|GA?H0VW(AfEea>g^IFqsOQtkSK!=08NZkM0)XCh^9f}m$FZx_{sIV5}v
z<6D02K4_z#dG_De*BB~p8BFo^#`#NVcZ_rr3(x2=EMCvcU-*Ldr*1|wehUgWq)t!*
zrcBqsaK!b_X<ZWDn_eN(=cMeF+A5dm`0A?*tE=>vkJS4M!@|Ji8dMr(3P%jBU9e2N
zX$bn5b~$Og?sF;}49CL~SX&?5_sZ*D=>uJ2u3JyVfhL-R$EcENVL+uer2&ROPJ!FZ
z+#kE@coFx0lwTRd!V!TS%|#H*ZRMH7=(%1N=5*X&@5%t_gLPfx@L`%|da$=n^g&ij
z)8V)|6{~&>Fs(7mZWcg&jz^N{z|ptLqZGV<<%_T0X%Xm#{+e}8)Tg<3f@h3CZQFUK
zIS1>KDNbTb?9B5{@0kyt#U+B*`}x=!-c8mVO?#np(R)uVh3P#p3bj)XfVEsYeE_3&
zM%!WNTs4)qZR+-&v-4M>KY*JO8-E+cGbab;lF_-ri5wguMZ;p#&9`z=2YpQCarE8J
ziOD<~ID5qtA!27m6M<mEnGZ>4{rClR6C~%ATGw`%Q6HKGdiWSm^{?}uQVO#*gYQ%(
zv%P$xP0w<BBHaRQf0&)Gv)x~*?K3V)ES8sSel!d4aONEyR?&C}D2#KOIoqIM){qH)
zv;%lih>)ee9X7_4vnlW+&FXs<3geV(?{v_IDY4MLZF=BCB4!m-w5l&gw}&O>Xt47n
zq{S{b4LZ#%PQ*+bomnS2ulnD-7c?~8nRa*ZMaTT;ykb7pgg5Qq>8C2T*LreK6Ywbu
zj!sct){tytwgJo8WDoLzr;Eg)QyZK!d(^aa$vFp?9B^KI;+nP5gk*_w<$)gqAFS_t
zN_9I?Tcb}&LrO<2hEW*}14pnDqkbrs<0AOWd0g)VyL<;gujeM>@QhKxf?k|=oCs-G
zVbI}VnN34R@2VbMvx=*NTX7-T@MKW-n7HT_tioX-awg_GBCNw{QDZ7j`VfE0YNUgu
zm&*QlK(EC0YPKX%BGW5>pI!x>T7zUW^b79^J`$JjcZIPwoN!V$2NmKS)?#S|E4bsa
zXY@{ax-$Nqzrg0T_0?0^uS;mmAsE?vW^gpRX--5h-IkT$oo&Q}{qaz>?&!Q*JYQT^
zE^#pT7ux-m`uUZzt5P^D0UwW;^<lJxQse|ah?_rzfn!egHh9@9ZAvux^4Q2J<CwHA
z=zGkgQ$dIRe0!7L-Z1-nW58qE_~QMOAx8>6$}tdwM0UMN@SKx0HPHnNVKS<TDQl~D
ztk;$6C6iO&u~nm}RHk*fh>kO)PlQyZi1W~@T{gD3^>%RIhDZ#dCr32>=z~@pwF=?e
zYh!<TkbjU_*921wrVomRpEd|cOXS)2$bC~=BXg*WU_fb;t!j|yIC@b+<PscfimWNI
zBb^76(<Um(hAjl%$<Z~SBSe(0<yAbkK8A6=k9LtyFL%I^31IKn;pLx38G-%!=(BO4
zG*r+Td|0WSJM{p^5aDdG_BMn}_on+e5Q!jO*>no3ww?#Q?5sg^hSQ1n^Sr*(3DfrN
z%!?0s(eQ2)%oq|uIq%tDUm!;E<$|AsIUB@G%Chc+n5f$>o7wfI1=xEY-~5cR%j3PS
zD-KO<z014moF>O!tHFbU1iL?qbq0)1*&lREqC4QO*{vqk`%}3cQ>JYj<t{kihbaDS
zebObcKNRHSoVER3sI{1%H86+DLqf_pFBod<m7MgY6AExpHg9ps(d$d>j=v6v;#9+m
z#5?{J1Ra?JdVi%9J{~kdjo1TZUoI24t=F3@fuoUUU`EHx`XV28&Sq7vJ}u0A0UPdS
zh_wAdeSIZg)T>=z@629Fmy75F2ibA}hWCsRX|)jQhI50w+{o99{#*#wmy010&EuPr
z!c1C~eq>l<V@wOqTU{}gQ7gTuILb4WGv7MSTcOO<b~gL4dR*n*;~dlJ-~!7Ux6N|c
zh98}yk4oDf8ZPaAkKx8l`5dYaRN{<Bo&!If=~pDu?8OR&=9_xhIYrS9@6-daKa`BO
zEKIN3TBg#=#xx;YUa3`qJSpXG#mIa=Un*;P%g{bj)IbOBtcSKjgp2b|BCa*#e{f-G
zoe-qgDaFEQa<sxPX)tD~#`|h3k3s*pE}7mt`(tCYop;&s?t0_hE%SWK&>qys!;<MH
z()A6NxIpmALTH^<syyvB1Wxy@Oq5nBbufZwubmF05Li=U>*9b@GFA6{)3F@kHFLg2
z!#J?No;<cfyjn&EaMTqqOD1jVO6XOV#qOgjhP_uSpq-iezC+HD%O#(p%-)7^;Zr#r
zF;KT%NQ!f|Wt+36>}GSrh64}oSUxzHv)~2m&pEiPi4+_UfTllZ*&DB)URZA{_@wh9
zMo|E*N!V;T5^IVC9Bm99kIkAv18sECL3RnZK1d!Rg-)3k&(~E8Aj+~i2B<NF62_>k
zo4P%OX!b6MKj;kg!{V^L#K*^XShS#Lc7%xcqijp<x4Op_q%_)&9sQUmOG5452~PPe
zlXjl*w({wpGs|p0;)vs7Znrm@z{h-|UqabBb5<cmJ`2GJ?zC<Haj)+*4D9Cl3_;->
z<xwa_ec&-TmQ{yc%8|$C7v|%^_4!!`G7@a{$19J=tHOkyo(VD1duO>^d3vt2R;jg8
z%Ep+T7zO<_<Iv}rlzr2t8G|LbHS~E->ite=N+L@ks$8U3S}CmS%I*1``nc__FTKC{
zh>we?h{vbxQhQtCRCM5C5dFbqL{EE2t1x^dEQ^%Sy%WsGigVUYekfWo_|cBT{uc0~
zCG(i<4E8jE)lrP0KnZ=JTivOx+Mj2hp(ef7Q=G?8{i*B4ATfvciy$wf9>mXfn_d+X
zG~3Q7JAERHB>|Qd*F4kP7S{@%MjuW6?X8hRAVw#it39l*!HYrhk<mBmcE>MK1wB)O
z9=agL>DBP)eKgTtz{gB+uRK2ejQ#$?^Yb%AwSYMeuk*QXzFjnSKsDPU4x!@wR+5@t
zwxx|~QPPe+hiyszzXK7DE{fvFH0Z`-A4hqsJI*Szj{+s{y}B`+f=|X!&i(NnNM0M}
z>_sh{Uik)xw|&Byw)lO0<#i>cbgHUF-(eysEU{APT322(z3goN^aI;};A#KbFT=C4
z?$<Thb71eyn6xQQ`RSUtUN5Am?}P@&K<|~=Gj%VFE?e;R8f6E2C>p($X5utu5y-o(
zbIUnxu>-7AV3^ob6|y-6u8SuJ$1mTW&I@h~UFe@4ollQSV&z_>dLJBc{+ssYcpcy5
z1LEFmCB?*E4KbRNB}IB4n4?$$GsB(nw8!S^gQbu<`SD#g6s|W?^xUp0W81b3rcwk$
z5=vKUWh;$~4G5#vlctF&37s9o&K0M#s2x$gLM3954xpH5a<s~Rhh-HC>JEKOYu3$R
z&lO%FZBum7UOU0fw~PUQj87WQqdz$Pj<?~*&B^xISl4W5oC1(Kfx7SPFB>rimPHoS
z$t2`#b;4#_x4fC7IOz}sV-(lXj}m;6z}`)qz30>DFi-{sb1-IOlyNS6#}Is!vgw)c
z)VfofT65P`j<?<eV-!Q4SN&{W;`m<gqMTeL#Dyae%Ke<H#CID8YR4}ZE|&{)RD5)-
z>s2XGWmmvQE5wk<Uit0!&mRQ^`M{`!5R<a(b7neE-5-KETBGiT*Uzs^ceFVT=uHj;
zn@}f{nId7mQj1N-O;cj;DnF7c+);2;gNXW=)IO-KXeze}&^f1xa!g9?lu}uzW4T@k
z(Gg~#^~U3VS833?TI{7Y@_9K4a#a4bkAf6i8^-xhwz8@D&U<dpPx?I%kNHqI$gl$k
zv=tn%G^dzIzT@Mt5}ay^?A}Pp=n>~h*9Bj8%5IY5-smY9f+5?=w8K<WeP`B!k3~OQ
z?X=g=?DtoecVCciPb|(ewnX1w8NI2Kp%zBULKD1ac%`Oo+nsWM;6vj2{tMEoi+j$&
zj1mTo-f8>(2KPH}o_#BIYQ-KfXVJe?9&mgA5BjLLZmnZR2icqG5BT*EPfgi#{Nxl9
z;>>rA@rfoBt3`38!Q3paH)$JAO-MH<fj`lwF^>q@$(#-)g^9)~`>VvEmkUR@jkB<m
zIW(B4Xv%nQG#q$mltw>dO25A^?+*mg{v7tbqigdTLx`iAIQ3a>kBxoXdHi2Kz~B2L
z^8fm;kiL2nz10V0FNBm>*A;*8>DLR_%PMURqJu{7w2HJWBxd)lITGl!`zOlFE2CDF
z#+L|7k`OqCx1q(XTgunhA3$y<6UT|y(C}>z8D}nr%O&&G7y7=|p<3RbZH}M(^dsMV
ze5KlFt258&E_WaV{e0gZ+s;;s;=rBU1-*A#m$$xKK0&a?Mm=+KPRXG(*^!-c+csi6
zn*8Fa$8aVrJZyn=4E6>Wb2P|5Um2|{t<}Hr>~aQqJmJ)U^)AgpbdI!Ur5DCTYn@hh
z4h@Kex`AZ*9Hm&6!Wb4br2(Y&Nt@VnqW8<&86lT|XbPPX8n|2QhO$LW;7@Ix?f$UZ
zWHl1ToQO*ng&Gp!U->2a-}*g(bCEmk?Q;mp8`8MNB=^<q(hZ5$J}!3UK&XmY!|Pi~
znshElr1_C`&}}eSvZ!4JolW{E1n-H5Ghyg69}N1mt{roE=x(U0yZ<>}-)R_l`tpZ7
zfAvFxcYON#59xgpBV2A*mg|){8jr^-F-%=HN6AeLsw>TcmV4`z`z!VHXFD((trTKP
zTyCrKvKh2}Q0C5hHSWYFPMl}&clNs`q2>8m7k;tUtZtRS<0y{lW`pq(zBni7cjzLG
z-kZ<dDFU_?#_ovbknk}=n{1C)T$+kAmT|)rqS|znENGbuqZ4922M%u=u%Q3(EnALl
zp|<&Xhds>tPbIclEw2e+Fg{f}6kR;-{37XLmRi5dHBmN(3vVjagPXJiE|_|MY%hR<
zMbeVM-*CY(w^v2|HXWF<15Moe#LTNzRyTre<lRMjkcQ;(@{!Rlc+*LQ5Qrfgj}}k+
z;v0NywatcF^sQ32JKO6EF$Ti=B;i<7BUm^veWwRi@O?M#^2oY3<%m0`c+PN?13<ug
zS;m|LbjY<@abCi}gX8_SxM#Ft(WCuTX)DylDQ#>jmE5813A`tybkaW(`zYd6ZWTd~
z&cXCBrvzZ&zW3b(y-QEg>&|F}m^^L2iwid7ErfLTHaMptd`wvh`#<?e|M&ks<KO$s
z%%o>|y<S*yA}^U`S*1;Kitg1jgU!<p=k5KCl@L48Lw|gxeEP(`>H8L=BWA^Pi<9Jy
z$G6Qo#pT?xG3U|f3M7&m9X+4k;Sik=vQ+Ahi5r!)7T$ju`BPtA`Gc=t`KN#Qkq<AA
zQxJS4(s?IA<>-Um@($V<tn1385WQnb3GjTrZ&X8^=CD*e<Hb3&HN7+JY#CGkY)naE
zUrY@$DVdf?+A3v`SamvR2(W1Nxa8BKD{0h#snkZu`3gV)09ZarL_t&xLB8ISEsM&3
zh>NV%J|?D*?2p3Z^_9LAM(uPjZ9@!RK~i0r0<8~n47`+1>&Lk+I@STfIU#l~=;uyR
z=UWK)7`fj!7+Lk;W3W9WfXvCWUbEuvb5apmh~N|BFMb98+3%Y$Txq*=bk)R9@NsIO
zJ!qa`&H+0crY#CSs<i1CA{0w*day34CImZIkORQEK#Yso3%lqf=g7;G1jy+7S$D;F
z&oPMW)t1HzbC-*b`HP_mu1`;d5ZE6NMv<lP@%fcG2FuekGiCyTUU$-(2|lP{uvubU
z5VuyF6t$l}sV!^q5+>$lk>$m*2}021I-I9&#gz4vR;tCB%7dFzV75*f%{=u=V@f$&
zw|GA{g|b&opOdrfruN3-9k+F%xxoJVS-$AY1zJ%Cc-AyJkT12*#%P6{lP;l<aoN#}
z?Wq39s*aTHmFNRGM<qt#aIMjK%j$o9>;$h#P>Y6anm#y2)8-JP1?U*3)-1*~gg~!_
z-ZTjGyQSV4cX)8#2m0Wg(?Bzmtf9{1{IraSK_5rzojmn>|7=u`Q5NfsXefm9R&6}Y
zp4f|Gc9&J0uNTA_%@yLk#q&B7$(ZGC81$$?N1;Ns>>56r_&A?e1y;2R_L&N16y#>|
z9sIYrwgYWZsfs4&6s0KdT@~3dS)3Xjwb69y(AMct_rSD+6X)K9R;Ko!@P_oviO0lN
z&f&AUK8B@dhQJ{i8=%VQDg6t<h8G(Ml0O1&j*PYM?<>Y=86%T@|4i#0+@S1tF6%o&
zwBRXAk;Ex8z|F~HAJowqfBz2&|Hr>eUUjJVF?d}9?cew_yj~;WSFfx~X1&Osdt}2a
zidI8obc4WP*24bj6EB}1+#ijU0@pR*_Cg%Vs2SS^58H9@mVI)#S3IK*wrw|5PN8eb
zLY2n8$rrv?d|p|fp731hkB^l5oj&2q_bczdc*lSJ-~IuA__I&?ZrFGBKmqq%P<Abq
zl$CXn%tq#&$Ggh58x>~)S?8sXIjDzj7S%bmX-8+7CnngYOX#*Pi>TNlVey!-_d)OG
zp+`e{Ty#b|FwvbZf913U3n?159m8@>`ofzxKBS0G3(IBY_U^{x_0H?-<IFp7Q$tZ}
zJEcwbCK2M%nJ{gLjLwmQXNmgHBX5FvLaG`jYh~Xnl&0{asm9p6uB5yOl{8DG?0TN~
zFMdh>AN(5SKmT2hw!uT2<U+J_DQiogdXHU*I2R4^)aCB&Y+}BNo`v{$p%p<Fy*Fg=
zI2RAAs5eWZ%HGV8)|DG^<Qq5(ZYr(wdatZ`a$BvCum1vQf=@sD8MiUC{RGdt+;DN{
z<?}24`pnZ0-jhnBy!?z&3wgcbVq$;!$m8XO5ImRLO`%2GgS=$&<wEeV#J7pGnaw5T
z#IOwh)&}Xa5_87qg^(9!tJK#QUcdR8by--KMD3AEv5Dw#x$9M`Y%!Ot?X2vbvTX**
z2ST@$#L=m(;(I<LOzXZgXP3AJWuklJH?RA~+$@VA270XoAIP^GOvlTI&-(Da!Z||9
zXBUEOJ^iE$e8?Qx>!WwVoM)O>0L4H$zq4(4GEX=UJ3xbp=MKnn_IfJb4{OZHCEBG+
zuozWuMIOT@c;^W*GN(hwq5?{4Evn_TL8EuUqDP+tO>KTYc>K^8H0N~we$y`6K8{iu
zZOUo`9H_)bP%Y!AmyU{THacVmvm>@{;u#&<A$r1WOs7Ji7~~28QCGd$&@+_{Kc*<9
zR`wG?N_kPjT~lGyXpOBD_Fc;4L;0L7(CF9*V~S)u<Q$>~)aItRT4&<V5WHYSXF@QC
zE*MO=aZp90V}3;Yj(B|ce4k(S<&PLC%kAS~j)o(Umn(bSxWso7XW^KmshQF)Fwj=%
z0smk9l!3P`SkFQK+rNXmMAl#b0rfZD<Nx}v5`OC|{KZpySMgV`EVB^m3yea&@7zDW
zaNinRRUB|>owN;ZH`(ucw_X%4_sAHAjG5M?4NoZz?;`_4hdjLhBC)27&ly^yY?a^p
z{jYg?|KP{J^h0pc611&QhU59GD}Vma{xbWca4-9L2^@pWFy7mFJs)MS3L<jaDAzUd
z+*js(qjka1(_nW6;0QVnPGw?`(RsSAr05NK;_OW5y;Dqc<ZRfEKG3IQvCF=fikp*o
zAc!ZeYdQsIHSk6&Ok}6U*;cH+RD;wVJ_c^@pUBsh?e)QK!pM{ov4d-s(hAdgx}6aY
z<p%_>hS?O&-8QT@#zKYkW9eBs>ZQH+EXzU+*np(t(Am9qoOgcBEdR^D!kClxpa0G&
z2tLlU!&-bGg#lak4!Axhg|2ml&Pm{oeJ)*j1TNz0VA2>d$;Nqd!=_tOfV^2Fh460M
zgAfCGSrKKGM1NAr3#{uxTnyznM9qK9>pKktAHMlU2W?(Rw`W-*+F*uV+0>6abK7Y5
zSCv)tLY#v-CLce&@cjN++iJ=@|6rAh-61u2h)OKncJkAW5F%~c8Kn|)Vtrak*BdxT
zdEA*yh1>8<?Tu|y^`?f4!1NK960;WBF8Uy@+0d}boE3FY2bT3}?T5uCYKboTc4t~z
z)*RaIN_p_I?<|)q%k5^RNhUH2Ndg@YX-VXiX|=Lm7g{OwG08FWZSSm_%N#K(FM%PU
ziNk855kJ~{hgwNVgHZIrLYBM`l?jdZ=5ci>*3!6h(E~707!SSE-9b;twllgeYS$`Y
zdE@qww^q+M-)=2fb$^&YcAT_DN|}(d1fKkNx3WGu!v#j|5+6p11gF=}hu6U3i7G4U
z@{!LY5NQr(+i_+S8zySn9(NTgMIE5y5cwFLQnrVxXw7+n3E1ks(@G=cOvo$C(>p>N
z)a}kFJGDHdNOq32E~KP#9Z-l+%nQB`L#)iwX`xv9eT+M{L)_V6qCs7B(ZPit|L*YP
z<f!$Qc$h!G2<T7j%bt<LqdohC0L~|3Qr^1sB36+r=uQ>rO8^)7=F@`|4_lVE0x=7F
zYK{8$f1mpIe~<Z}{WJ1k_%Y*u@fYDoS7n{JL2MiC@rl0e?7OPAgA1gfMAfZL=C-r<
z!R?YI0MtYW)|Mv~zApZb(U((>oZ1F$4qSL4t=3J^b$P$uH%hO(zpljO30W4~e&3lm
z-oLx?tH1o7|MvI4p_*U(WTvL9xOA!Yn-yFgNYv#bacZ?V-bbmYSk`-I%48vb!T9v(
zJvk+gfTva~F(t~bE`}v1wF_FON<*rAH>LP@fD7&%Y_ZE~4lNHyBPw|5Yn&6!iv-`!
zy!qaH(aAoLJmhy5#?ysb8{4L#WG|{v-kbCS(}@;HArPD>3Nw427xa-#ImV=z`z#s#
z6GC8J7v+tNNu2XUdyYUU=Y|C2*^~d>zr=VKssHIe#=nSmY4suA-<D&nILOTo^iva{
zfo@o<pwY{{ainmLx=EmF{VlET#3*NxX-jQk(2#|bGveYdvYJkX?`g@I<P^wRmCZ+z
z>mfq_<6hs*S~wcteEl_Fe)Sd0b)}iWC+5WK=Z|b3e}?DaEH2{EMk6ptQI|-o1;RpF
zF5*76JI?s>-UqL*chdEWTNAzROz%ihp%8t4(5rq8?*lC=4?5*c+jr_-bQy&}cY-Bd
zaI}3=hLW;kRCC?2ZAXKxX^w7IuCf>E=T|Q4LR>{D?0YeYtx?M$<j#0JxIHKN=10XR
zkBgCgtCUZ#?61;3$lrU#Tl8_?HOzQCWqW`RN-Z4nLC-l=baQNpx{>Yw2@01TrP14<
z*P<uHHr{ZgXgQ0#Gz()NQ<Q;u<9peey^EJ)+J>F%;g+^wOxdDZsSJh{E!kx}zNK}r
z!^sDK3cyYeGLz}$m+b_IQWyBc#&`$_524@aa%l8s5SQRSAsa~Clci+#v-8Efo9viO
zqj{~>&blB5qIha+r(n^?sM^&zi8*n(Tm+jYM^1tL)k;I$WQmER!{KO4)Jc{I0fkE$
zEi{^Fl4;wRvLU1>cIHlSHsSX+?U2^CIgi4rD85tguejWUD*E2OMq}){45Y8{1e`_t
zTEwdbrx>~`5od_tCaqR}_#B8KQcb<iQBu>@71CzmdVM1LMEM6lf&aq?)_?D>;(qgo
z#N?R0$XDJhoXEDv<p`3hjcsqd><W2VQsk0#UPMd(EPJE0MjIA%9i%a+eW)O(HC!lM
zww>$kCSmPRaFnl(?Xhu7Pq=)+)lSY99XvZf{^1iJKHvGH51-G9>?7mAJ97yPJ^Onp
z1Rq$JOwJ3nG}c9VBF#`WUFw<El!9t|vIP$*&tl)JEi;%)z!v;*1ez(KuL3=(LfJ*L
z)yBvX`OHE4rFcU02NEgoeZ)m?`kQXKA6j}iEqvA_X=1ylYvk$nj`!~h>rY?!$%ma<
z8i(9cJb6?Bop-kN>x?)mu8;Diye?eU1sIGbH%7j!C&wPflun{|Vm6221gax0E6e}v
zFX8|7Ut#=T|10Ky{WIh3l}spcp)TXe;p9++wF^oWG(1-biBF^`1lW5IDd4@bC?qHb
z9HLciDNxg?#M09!hv0%~OVp+~O{d|);XIRC()}Wa0Z^0G{pAHW2er#qcD+8+TVwnD
zS<^`j#H1;ywZ?RiFDtcHUO#?jxhO;?=EO{iQVVV0PqqBZ&pw=0$<2B8qTq>ElsS%9
z?mjqXEkx`6_ETT<*(xuupSit%r=EGQ)<VxvK04|b7msd!aJwXGsf1;vPbbx?3p_r)
zQ14RT)+!gpUUx!o<P;^AarSEs@Iz4B<8|kHSqLfW%lQ6FymLId;z@_uLSi%s*?7o)
z3@c5#jOmUZ_+hPW<9v8VAMCq&$YahpQ`UzNEagj)ny0O;w-d7Vs6nkso~tR+nI+D|
zkT^rEL$r)^O|P?CafW@*g7<IlX=@6Pawa&kgK65Y7h=Tw)ga3GcEI#*;zm8tf&#S@
z&pFR(*pxFVX8Qih>;=Ch6C|1~(-|K(G!)IrY3G}EaQT=-Q8bGtt-Az{F~S&x0PF3h
z_qjK^m0Af_oDAoPc{Rb9f|7={%mt&NrU_XOp{G3I)h<8``jbG&`6DdK27_bxo9aFv
ze7|XegPmeR43XjA1|bc+QJ+U%Xmw{<luzM^l>IJoV6UW{$X+dbSgo(m0Y)dqD?SD3
z3EHIp&PUq+{_imUjo)DWkA8)DYjji!G^a%&JD49T<MT#zliPpwGq&IT2><FYQ{J!m
zFBazeMNXw5PPX>W)&|3Q5}rOAkJ4}<QAWh~$h+am9dW`jvg9nZVOfazYQwIa2eWr(
zjs)lV>fM#EKfWHd1p2PQsePH~yHrXe#>wq^JqPj>qY3RIwRiQw#K=hF)IcmtVo6a7
z`%)Cmyr@h`QRfz~JBFPFs-_Q?Av4<>&2=~?Xh)szfq<DNX3F+^XqOm+Y2{0YNLaMz
zVMY8_V4AZ4F$Zf1{FxtK`QcOIk3JTD`tsmjWEXZFTr@|*wD7lX&LJluqBd<u>rEQ7
z>!M0!=ZQJcEl}xr(?=7E7-AqUo!JV(N5WT;{=fN4)PMLh`rrJ=@cVb>C+V|Bv+%c=
z42f<yy8XU_>AiQd{2kl6bA)`gI%p*d28%>?J}~M=Z9B|C2$>MFtc~_-O+z$VuW(r4
zhir+XEB|JWL9LIU>lyuSam2xKSyl#v4<A385Nq(bzY;@~h0LHJ#-KhPY}>9(?dz3T
z8*Sh5vr-<F9N$xF<MZo7?gQtzKHoGw)(wJEwo22bx7~O4TFFahPRFujm5X=}N9hC;
z7bAJe5|qS9jEg+N7AqXH|JLL!%UXz@VOihH<th<UAM}*)J@9mUCqBD8po-U9mc=5B
z9o{E=UTL-R`Spb{^f0bBg(@i+XxfLg5|<;HaMu9P8jVI;7SdG6&#($!B|8SeK}bU&
zcz&vV@mr-VM}W!b-K=l^+-A>>vscjt?kI8Lw4wlKh3Pbi%LnoH&O6HE^`r^Loack)
ze88tgD*9R(rBS>z8v6U*Y<#xujb;}q+Sc2&@6H=iFx=$ZNnf&&%GQ|5!@SKw!7RhX
zHx}r^nIs`?MQfVW{fNqNm<awg4tVxPGco3z5c10U{0{FU`~8y+PTNqu3law6plML>
zUI!*a7BBDI*&lHr9<o|BU0}_kw4K!N+K7|75)5Gx4D7U~%<uOwAaSU~TQPa#;^poU
zsv|yoj7qILW#6pkT){WB6~W876TED;F=^lweT8wxjb84g6xCPq*cpHOw|V@h|B(3u
z$9zickDa=0?4{yTX8i6e<wK!&<^RvmulT?FfsW_%(<FSg=&L%RJ0X>2m#|H~^o1D{
zUwriyA3i;pf9@+@{*_K&GT~QmeBP@{aU9gHNb`wOH~U_A>>C8fzHQ`BJM%yKC)D5m
z8S|Ge_`m!k{10!quP)NX#7L=?x>tg~;Jmo-7$d$kmYjixYie%1E{iT->+Ep$-+Sxk
zug{i;(rplkvJP7hM{P}_<e~JxsGG>Re{(AwP_;W^xr`nR(P$Kl$nCnoqY0YZE^WXe
zmWF8qy;B+5)Y21z(9y-ipZ+59r@na4YZ?5}=bfM3i-x+<Exg<@oc=vki6@1<_0GQS
zc+@MCmqmBL5RKAHqxYgr3Oe3*#t=N<5Bk9?yC(kDGx5*=Rr=@2_}jlv|J%O{ADdB=
z1L%fPY8h7RYaqlMok|SKVR0@poCL=QlB(+T23?MXgIXM2=MGFpFGep%(|S2Q18&m#
zPDras4h+ToMZ^(*@%|YX0u#^6=NE%v)XLXl;{8`&$>+a67-JIGmCJe|d0iZyiJy|B
zwOaZ7;b*GkES2{^c&{`n4=G0$6F+RF(rTyf`p~^a{nl;AkItHI7E7wy%}Y+&GI6Bn
zSuQJcDraDJeMrk?fl?_C1(4v~nOwEruFPJ|x7+Y-@Wod@VtINevEX)R|MU%Yf2FlX
z**A>XA(==mMNjmZWnDFKn-}%-*FTY+WnCnsI);Sc>7^?L>$D7dgO!^2qaX{R3RJAc
zadJZTt(j=jAovavG=@mZ*|y_Dbf;}Kguw9OY>uOwlGe+eFnTAZtVT&Ei0^D#t5t22
zA<$bVtrybe2E9?|%lR2A*d!9Br3;S^O~{8Qh<7{!pN1LnPSvuko?}QMO&<!Fat@au
zEzgh^T-}(Xv)?tSL<_g^CYtodP0ndSFUBPvLZNA*9%p$JqjzTLq%jE5kuO)3PPIaC
ziRl7<xd`el57p1wKs9Y(an9^?(-R2iaAp2Mi?TKS0UD<IoNwsKuq*<vf}QqrDuVx{
zzV_`QtDGgKb$uamK2qDj`ypCu4jWo67JLkKp*Y9PSF5`fyxPw#RAoO$2Yhc;E*cB_
zs{00Yr+?U)AyZ4EJs!FkE}q81`fg!=Z1|j6ePn&QvOgN`(}VK+Kb3HKS%^OJc>P4J
zmB+sE%Rl}l`rN>G#_#-d-o*gF{mS;A{8K*cmG^)Cm-$>@S^vj>mdC|&|LYfGx|7$;
z!14Lz!6Z`t;FbH|{I7ZafBl60AN(GN_nyD`&u~9j2|s?v`0KyS{!34c-+YJ9k#O<!
zy{bL$AN&;eT7-_4!c$tfTqh|!l#4P3{P{2FUc%}nE=GNr;(zk3v&vdl$B}>Hscq1w
zCs{0RALDdtDbPz-$*{8p4&h!$)@P%2x)L^B47eVk^)o-lX7>|e9it0I&+3%G1c1&@
z-aQ50z5jv_FO{Es^O={|jo?M`rJ%s6VBtU$Wgw)up7}C9NZ-)5s)*twL{NDL5uXh8
zaFO0C9@Q2*8qRe^^S?{<fBQFSf8$Rv|K{(~{@=fg8(P+!pE!aE$2}nm$yVL7HhAY9
zDXk2n1{|JY&%m5Va-f8mM=8=_@eUUM-$qr0y_ef$v<mKM1?B9T_<3I6X&88Z|DIYa
z`&LZJE6dQn@0vQzdJ$q`eY)afP|?PHJM9>A7G7T;Jnn_3=N0E1%M$TJ_P*#L=S=X9
z*OwP+>*Cp`qun3);AvJ1>w>3uLFcV^ZckTo4g?p-PlEbODeR92F>4YGW(_)me?my&
zCZp1-_HC0=JtlkzEVpMa*JmpNafHhirl!bxztg=!ZCZQK%g(Z9mUS^Zn)vo%!4h+H
zsY_dc;JLrvX^#gXE05r?#2g#&KwI+YT>`i$7`=5N8Pg!6Bi&8~75eCN&UQIT@sEkF
zdass^&k6BM@u~-JGF2$F?;0f5RV7C_dfBzjqsw>nW*nFh2g+S=;Hs$6&>C%fg*otk
zvc^d3&6LXa?+i2yNOG^Cz&40sB0J~AcXlRNI+|a^yAo7y+8&>D;f81epl~X9d+WGt
zfj)H?B<KUq%j-<+R)4($gYkI9bqN83_b@DmH3WQ$%#a~pIEq9DNNhQ*gTY$>l9wp-
zP=q`0iB_L1&RWG&N`Rak*#nC3KK$)(eF@*|SMS8@#njZ9;DYLIndId{uNGNt!DT*}
zlc%Z$a}sr^dF$YafmVw|VMAM}_mO>n=vnrj^?If68?BiQD|^Cnfz{!bi;x^Qwe7{k
z#kxJDf0<mL9c|la@1J=*Hm>j9b9s8^{^=9<`zzO4`SO>3$i8h{Z%;g4UxnyIPpOTZ
z68@*JJkQDXKmJ4h`5*iNum9mcWn2^aU;8ya*9YJHfBq@gPmUHdAAa{6K79DhoPz&f
z?{{Jhln*=O@e%j?cV4yvUu4p{(B8dczh-g>Y`_0ODWFptm({|#l9y<$mL0xaN$)QB
zzxHGL`$YVUKV*i;NN=G!t=m98q0~ZJRu%v3ow7PESy~!5Eb*|9Gvo+6Q{A!Y8|;rQ
z5QhqXV&!td3ak8)-EWb-V=Cd1Bk5B^yEBafOl`^D@xw2#Ty7Ws+3)_8eQ&nGdx=tw
z4wCiG5uBa#UJ>fbQ(5gS>}o2sgv^kS3UoD4&Zg*Lu%%|&ncN!r{R02aZxR36uh9RW
zzf1pbzNT**vp35aFzrX%Oh+;B!J1H8q1Vlt1(hVIs$>sd#cQo>I5QYalnpp7oV9I)
zun@w+>=U2NYTo52I~sO>p4WG)Fdh(M=6?S~Z-d)q#mB&?O$mPYyA=H?s1n$FK7I2u
zzWMqi>*Yeuk=_O`k7CieN9VUqnx}tL{_obcK`b$_TF{S=+6?->pE0B}L}uAlW-|sZ
z_)|4r?;H2mJE?0YBnE0JrZ&}fcGyzrgM7VG9~-@?U}Jr@tn;b=^gzxF<MJe{pTYks
zC5kvdSe7VjQtOP_sa1(x&iLiUuO=3m5Hr(9<C7vGMhz0EEzLEDEIH;_kX%NGPSRHG
zqEzm6SEZ%L8Eoz8<6F_B2ysYx)g@VXeW(^bB#Vu{cHIShTA5B4w7vjUwV76o<jk5l
z$Kdh$nd%^|SMt*rW+4;ZQ@2gOb}#{wcN~dY-Z=<nL7o#oB{Upkn!8TZeXvrFfXA&j
zT*%DU1O+z*Y#f12pe;x7p-uzCvg)#)jxqNWf#^fxoZy2#i{4C#sWh>=RUM@E00NY<
zQC@e(JbdfCX$!iALj}<~bDCJ{eE;8k>cJmLaR83)0%Qqi{j9RAO`}cx_nRgdIE8Qx
zeeYsgY_M+zrD`jiy-;f*ErB*C(Fe9|JKrl87rdA5AUHW{%D%I#7X|+W(O<JmXcm$R
zal8${-Um{M6x*zPFUoiDu-sO#EvsyIS=-`;v1XpX`clh`^SuA1uXt`>oCDi>yO}6L
z19EGH>zbGs&zR0EtCi==h56HsjXT?Kf97%D=&yI?`)9ZY+R)FrE*gHHt_$zqzvttp
z&*T(|KC*2)^P8O!JE3R%W717DEJeO&xm-y0KGd?yqUjy(V<o;$=KuX41Cx2nv>#t+
zKVEU~^fQ*PUjb-s5dPI)CH=(<_xnQuU^d9L{cf<ep3O6a33PFT2?|ZfY>BDPC5Yah
zF-*G?MVHoQ8m>1^JjN^pbSP3h2H7<a+oj;n$<rnCn}7CK`MrPsH7_rB=7idZhOwMj
zt{396oc%QV@f>kkIBbElRc0&Jz0)i<9HMu$Rv4`jbC#2&9TYP-DzpN*5`XK4|8su{
z{|`Ul{?_l({=pw;XzHDI)R|8$W)pB>N`E!p=wMZOOViZ%`{}Idy%Un^jd6|;q??K9
zMhJ;HibbtxL6-1z8nU+hT+irt%Nh9i*+eFS_~PtD=eS-MS=H*!_36fvqq?ZQ=jnFS
zW$Y(?->73s#9}2QdlQV5V`4fbt<@rkdbwS3+s6L-pjHXr^15ns#;h!Od0mfUqA?&$
z+?ces<45CiS(!Frs#Kz@_@=9M@8x5~`M~mgBc%mD9rxE41@Y94djELZ<5Cdp(u-N#
z^aPOB3o#`|xu4v3@3Q(hvmYJM(Drg?H?HyW{KUjjOEKTSLTHYllWt$s7_5%iAUTQk
zOgWEI9m7$SF%`i%rz!fq)jv;&IWe3Pu6)EUQhh<kRJ`v*6Df00?yq>SZ3&y)o%I{U
zyfAuWyWfQvwMJ?azg`tk(yM6(<Ys8Y6!z|b5G{HY4ODU{NQiR?kNkO<U<XQsv?xXF
z)6W>?!NkN<y5dRsn8s0<h*_786vrcOb#MANqD8IF-k>*_-JoetD?9sRBW2Zg_f~m*
zx$~&falE^&Cok!S^6X=fE}~Clp|f%RI5bCXXSRYaD82V?9?fslgrtfYbtxTE@bB&G
z$Y}=^Y?aaoX))AmhtaIoR|i=Lf`xGkdm2`f6GFhnL~E58GZ47mp6R2h55Q*=irT@8
z!+97(I+s>=oKNZ=7=z?cZTR-U?NY!mmkV<$X-*Nl3jgV&39ag)C(ejkJ6lr9R3^zS
zXCPQ!E-)uc%H%IIpFe$&zzxs4FP^wPU!@T58i>|4@ylQSh_XGnUN5Zc!qYXe=FH2>
zon`rfoK1LAbgW5XL($6DrV~vVoDeaIGdqqTt&l0x9ggvFr~T1q#@`+6wMtYsrX~Lk
zF8}3EdHkDSv;3QX7Lv2U)LKACI|a`ZPVpR=F!rFqoP+7k3;d)3=a4q64=Y3Zwrfgm
z2XArC-$oZg&bAn!MM(YRT(jfXfBjeZ$xpuK<@K&*C}$0WhHwsRmJQyk{H_+WU|J=0
z&dlzaRfw_!I*ztK@ZOOY2|uT4fkF(H88C2VV^)iE|BZ$C8~+0SpZp5-KmKpvCmZ22
z^tqWhdmd^go%^$KxxH7I&>2>Q4OYX#$sAbgs_qjlmx{$7FbZY6>$&lRK5Q79Q2Q?+
zXW+3FeF1Z%_9|s*fPkzwE_n9WSF*q1nUqr5OXcbLN@lWcg|=6&i$2V4Z^R`N@}jN1
zsru9B6*oF{E412q+&4msy!+w@gyq8Pho3R~kQcT|b1?SGn!K!W&M9GT&eNW>l?EY!
z_td)KhSK9gzNndQSs_MRuZ&u7y{qKn{+Y+iM_Q@$)<`+AUN4M3se4f-K=8!TxvW>^
zFf7TqV`cFlx$HT~?%1qh@rb;|DRogf+Eu*J7$}KOaQP9IaX@cULyMC?1Zu|@Ti(;d
z=WD`6gR#euhn2(o9*loIRJdLBt!_@XmrwE~=fuR>HY=zvWjnCvP-&7~daYI-AAd$)
zmh-|Rs494bBs9kK0T`c;w-z)n6)%UHo<pqvW-si$GWS;j9er0eL^La3RELF&Ne9<x
zhQQc%ZaM@NB*ApnKsV9jc4Ur+VBDo0R1;IZ51KhSa&RYlFsC;tGLV8dyU`(v(x1Wy
z`#hWmJtJu?AVq~UuIp!TA=0OY<gi?b%T=jq-{V(YE^=cyPQEq<Pxs#Hw^h9Jc-&c%
z#H!KD(us4_p-u-gr{nFC%UOjwfy@|6uN<uq@@mlYQLH0@=WtU!DoDsV(<-OP7^JKg
zz59zIrt@;K%3`^w#<8(&FE|Em7)1wipcvYamPM#jD<*XGwEAG;xn34(5wCkZuYC?2
zo?l*mq(OEJ_C=bJb|(43)AN&tklDCBy%)DXy99og12QHS>Lp#yk`u4@o!%>X(eTy#
zIGX|ek-7)eVzs|$lz#m454iuw#{AVA>%aJ8?)L}d-OVT&OCWS_O{|Bt@9_F#fBzi%
zO;2#RpyUv<t{AIzo+gGmG8CNSOfB>=;$qS>6YR_wI;ZiTxCH*pZ~PjcKECqd!za$F
zYg;gwTF^!nbsNn>eQfHZt8TpS#kP1kUD~!0a<Z}}&+xpRwO+Ik?oFc*=D>wP`wQ=h
zzxCgz{^U;ozx^ZT-`l8Lkx(;5+%z$@MiC7^T~u-_o8HghI1Y|sXtp$F8{Q{IyF<^?
zJw(~+YgMbKkBb%m{Q`!8WeMaIsil%H3tK5{k0$%ZHAyhm)M@?c<0tN=;?jb<T-jb<
zX=CvGbR}5sba1laxmMV=4L>Hg_wVp&VZVPCe?P#5M7}+9dH)rDSp^|KK7)r|yM;w4
zpZifNmuf<>gR5c?^Fq8lodU1gs;2AfO@n~*xH+l4)Ak1;I+l3h{`z2hsOVvIO=&9H
z7<~Hl!bHf(k(PGNS0<V~C)yfPN~DyON_bi6EwkNUgiH)mxjRSmhZ^^IsJ!)i=MzeD
z)@Lvc%9{2?aZ#`|2uYAS-QlfFLl2(`Rno;a1u6gKDu_5YY~b#+E>}qHol*5??_&^C
zA_Xthr8uKdly%T;^V=VjIjVZx-4U~C1w@Cx?{E>=(W8q~l*3`0akkaXp|yC7NiQ3-
zJV?tzzAp46rS6z$T1GH@Z<SegW>n}&%!?ce%rlxgddFnYyOp#|L9R^bZs2LAW)=sR
z)(dOQr2No82X6^$eOUK|x8bUvhmdz($dqkm$M*uXZDAtNkmMQf`_8`K@o6DnpGkQk
z)bIB&fDg=&=;h8BlbkcHRp|g`#)Uvk30ESIeIw1m?ItR&ZVy6C%;8L@A_S-Q#{F?;
zT`#6?*8mqRX%f)I0;H6z20fCJ!-W;Mze=-^FA|TI!YGZjE|gX!q8*ByJrC<zNNHs<
zD36CZYo<!KnAQo>EYWi1<?&#diuYZXh2D)KwNf9Z#|;P}v3Lpk>R#2j=N!a9N=}=<
z(<XVzE6z<)4jieS-fNqUaj|XOkczSkF%8mM<YkfVvdN+xmqp0+RJ8k=6^T8D<FW6A
zm&*J<{XNS6w6gz=KV_~cV?~b3bed?^GW`!Rl5+=~<;+4m>>nR$4jn>E`>;#S&cD-L
zFij_rl4-;Ohg#nq-b*H97U&L7A_mWwUtYLgGvEB|6QvGX7tN581;qJ@8=h%mMa=In
z3$rs(0MI*SFO<@_<*W<#pj^k$5)<OLl5VGIs`Q`Hbvfex^gG7??yvItfByf%-~1=U
zkVwmwt!y}_@8%p`j(`xEL&H+*lVHugK1TddvgsJdIno*3odHjCTG&!taJ{j;Zsc{*
z`}T9ZzSA(!_riylJ26JCmn_I$-?LLV5|_!=8!sOoq#V_EzjsE5^>SfZC3d^qq(XnZ
zywLZZh@c_YtG2%?s6BYNeDNjgmp@X8gm=W0dHvjpLH3*_I4)~u-yVGa^up^mAMr8r
z{Qj9_9K6FDWNjO?d*drpv<l<VsrOe^6vu!!luMR4$Mkx5@SfI-xE&934&_sW+84)A
zT(N~E&7tk0_f9K?vTd~14Ka|Cw)a6iw*_xP7#(mi1xEkY#oG>uP6t6Vh|Z^DVjkl6
zEwD#IM9;trzTEFPALXqt8^L+fqCD`~tAe7u$6r@UskD0hT6&O`WgxHc1c@EZHywN+
z#zYM1bO-q8&O_OU1YR4It2?6?3BY20LlW$Po6aEF^LUvjf|9%;0ePjh#>=Nq^wx+e
z;u+>?*2W-5Lbc&ZPK#lpq}G}Qh0av{c_E}sY@HbuADwcdPoYj4AF$0%gT|OcRw&*!
zrK9Hc2#D%q;9`b&#GRUGur?tg!9M$GFs_e=a`~S-Tl`P*Yg)cVGb>r^LfLTIjCw27
z+PGeCjFzcQO`%6z?vcSzwpY8@1>LV}QZD=17vRk1rXU-AceI(NELV)$_5$9&(=&x9
zg$ziDd?t!I`NHQE#Swy<?U>M7VNHpx>>7eC_IKYa>7xD&XHJ^d3Na=|-)UuMcJ;iB
zTGVMU2XhE%D|^9NP?@N;t~%WTGSyGE#~trwyFM<W<M{}t$H{3q(?}1iaSZW{LBFih
z2;>yG-yg*2dCpfnP-|h^2bWc6+Wg$8|Nj3Ee&+-2-%W&Hz1gqxmW?r5Gd++pAI4Pn
z%7IvpVVwhp^tj-y>p@zr!N3m*n7w6r9Pz*|d^_L2!SGDiB;vDlKTYrr%nrCD&oBMx
z2fW-XpFh7)txM<N<YS5_f^h^Z1s9m!5thtoT5P)&j-^HR^tM%csX~MTw6Yt;C`2)2
zYuK&3te{Ea)EN`c|MS00{#XAr<v;mt>fie#m_yBpOEOxv8^tAo<$8TOL%6^(?QURn
zTHAoZ_4bUoVW3u=k1W@B3gO#cn2LA(xnAFC7&v(EvKN9weo5QPcf?2gU{RfO(jJvk
z#PuyX;AW;*MF8h@WpDC+IyZU$?wQd$+hZrMS^5*F?J(y=ikev4zESQk^xE(QrXR3Z
zhEW8E4QtH`^VrIrqoVim*zisjx*=z#agXD*nL55R%0}{U1v-af9|xD`CkRP`xK_C=
zfxKi344TyK!R(dQB9UD@3QxKn@|}IVgVVv(#}L$P@Y+EqxZ}O<Q)0G0=tGy5yx7X^
z_VHUvbTFIMMz)JZG{2yKAtka9)m;<ms8x9d&e{77qgC-(F_^gLEq{MBB@xCD$XS`#
zV{}5!++w2cyKN-ltPpJm2YPEw@jl@4diI5nB0sa>vH>%g&v~31PqM(1;8jF|16;Hw
zMnqLOh)Xgxzo2Ayx#9B)tth>6NQiUtfrglPga5n_ma67xj~(h@v}EsF=l=0$20Kn_
z*8#C4+kTvMAML0kwnO^JkPvr0AzHKID9m0YFdRUSxZ&t?*B{R`4oYmR*6;rPef3&t
z#?uc5YtqINV#eVaQ974lkmh3CaIclc2VU<F<$9#Wk^^;VEp}nyjLN|RXS@(2H4tL<
zRK<z+q0E%Ltfb(nwP@E&3vo<*x^T@GL*q~rUhjIQVg|5FtP+;1T07^YTg3M4-MT8g
zato|$VvLDTfl3i_IJ$Z9MV%-wVX{A@k;>~uOQ!dxb#SIhkzT;LX&0Fmnag^$?{pOs
z#UyySzN&rapKez|%)EU5B=PpyouTu`d&hctg3osaEp^eAt8*-W=lA&Zk3Qi4@|T3)
z`jYVHen9x)6Y=Sy&HiW%H0{ojZJ^nHnh4g(sg*OK9~~TeO|$iSv;T!9tGKF<Q(5n#
zJ=^mb{>FJu>(f8<2}0?fuM5m8fB4gHMDr#I+m$6C<W+y}WCNAuufzyZ-6|fPjcwcL
z+a|}0hq7%am4H|9-5l89qVF@#F<Zxb(V6SN_$BLK_!S=i%^&dcAO1bsfAQNa(-ZSb
zaLNmaA=Bz68|Hi~W;$FRbuXlt^*;H;>`I{ZUgdG3p*xrBqT%uvfeg6dSudgq#^|(Z
z<snf$-@*FmN0aKlw1&g;bY0ZTJcc~JRRJUWw$obYc9Rh8)8~g0wAM^sGW%ZH_d<C2
zNWNZi>(#7YgMHsg@yhk-nfm&|<Mn}yk&q(S6v;VLHrW_&*OloVk5bKBn^<o*o)%df
z@1H+&|MUsQAmwOYdxh2H+=*!t^TK+2N6ahx{e_1YzwZwI$fH9js?u0?7>nwSmdoCp
zhI{YNRIV7K34IhNIgH~Lyex-78<ScqwN{#?3A(PRT*tJbxZ$TZgTb&GOj!x5HIOM{
zbb_HH)9V4o@*+}<OguA6(^2+n)}cU%N!8S2kdu-Ian@fxJ1!gvjMD}(WtfYqiKbD&
z9#Wu{Gq=Edk28wk)IK<Qn2!W8oQAvLY~XW&=HuBs8RHq*ro%y3Dj<Os&Z}PlhffcM
zKaEKr>Qm6qppYZiJ3e}IJ~~VGvcb&>mdnwTmuzi>qC9FXc;_TGJYi9<WV%vmkDah&
z(sIG|3SB`Z&IJudHUu5Yd8Tp)Mw8g}`}~SaqLz)ESL4yMv>0R1n=~|nioIpG+eS4y
z46K+(NSExtvfjqt8AI6;;{5G9Yuk{E(S*A~?Yy_|=JdS$>>EZK#OSy_zf&UJB8L=n
zvK#cD(yE^gN2`^Tvi&)90n|1L$;%7xL^tibzK1CUO4r#q+Q7NUyXP;c``u#jC(%jN
zI>uyg8*>b<mxYuUn)Ptxuz(2+j~e*noNz>=rVh$+Qrjjs%II9yg%AQuUhzKAT4i0b
zAvA-FUoC36Sf`5TtFLZ+^YshPdoI@-t{41hl#h4z|KtOgzw;APj)dR*lJ@7ng1`Jj
z!VjPDU#fWS=sd`mi;xea{iY8J_QKnupfpd_Sa;<Pj7hrYGs#o30igy*v!IenRW>#U
zgk{k=W0aqF5?nJoPuI-&!54h<;S(<}ucQ#UTr$fN4ONY2mjg%`I*ZPyXx7!<8s+s>
zcFm;4r#^N;^hUREP_oVuQdW(!gD{1Rjcv#Gfm=G)zxX5Eum9`#zxQXzzqiAG@l&?{
z_IC&&GrSdH^#=}5w`SSdk3?VH8?_Cl+s%rbq}^~{`itq9R%Au|1ymU4btQ(ttBHSV
zQKH)AlIeZ0?}fZ1yo0Cbo2)lu$Omp&+P2oI@M9_fV@?<`5iSsn2lvhsQeb}_xV?}+
zePZdIlmg2V30|QLF+|+5Fq(uQ3);lXzBA|K_I#n$MlGFueMed{IYriW#e4nnSNp>D
zT4cd-2JN;+iVA~?IpLf*>%PnPxg<*;j8S~ev9a4>K3YS;5VD1QdjgdmF%Z*5yy*Tg
z7(EyrWih!nDKFj=LZnuyq7PpB&=<g=$aV5`J7?k(apc`Hu8%+0VSi)H^XE~uv9=!{
zJh3;WzQv$a$CRWoKsnhydf8gKUEFbzD9%COC0cSm$T#0jD78O`tx(Ix>}q#RAWC3Z
zeuUF!FvbvfI}WdJwE-#V;p%T2h;!gVB&0=_zR?-2NNo%11sI*_k5o#vG9EI;BiRwN
zkd6+7Xtl*&Dfqq{GGJW;#w*7hPptyMoqX^aC2UlV#G=X(2qBvC*I-l2PfH<u%i1LU
z!swOnJ5x{|8+|ll9HcB5UGRNW)N9ZH+M1$OM=NYyDU!A75_dkBjz$n~G<^oB$chQA
z774*-C&Wxb5!SN{ZVx_^!eHAsmP=+`uV?C`1b73gP$=h6)w$7d6D4lmUp`8R8U=|z
z9uF>;3+7!g>tmqSMr}pUT8!l2330`TNGUrZCpi|pM5iGqg5@PR1~IQB=O|^DmSm{9
zxRnPkM}7Ag<RwzdWEc-FjZUPt3SJiO_qQiecXOJyOGCyQ7xf&(K=dk1OEI!NUU>I(
zVO=k(4{i=OBW>SlE^sfMn~(H=^pW}xz9#%f|A_Th-@&gv5&pGb!Tn$*{n;OwR>>~0
zIax1P+O*LsL<UV_=rJv1%P|g-?KlV?ZQ{5fOEH*ptW)cb$Aiy;(S4xpJ8pE+^(qa+
z9K8QBlGnsP|K~rY?8WrE=@becsQ2hicR^8H^3gr;@n@gNOXB(IiJ5P>&~Q!PPoFcy
zqnOB9c#+Os=gl^qRzhAyU-p4>job5;^w<9^?XOhE-}p7s-@MZP<KM<x=1b|FIi=l7
zNuOuV3t9^_Sb7)e{-CsEyJspgI;Mr0B6X`j*E9N^hJlYCU$`dyhOgVMH`MFniZM~U
zgnEYvtQAYL>YZ8|Oo>vw7o<PtWIE65R`Hi7^5U7TNx8dMiE;dcPoE#kH7F03CF5LB
zsL`6Jty&gzm%CxgzBYs2z^CUckH^ma(YXEKE0*g8H@h|mrvrS%$Z}a>&4SwJ&_#|j
zi(h8B+`v2P;~^#LvI1tQ$txiySt93TPHhK@1lF?JY7-0{qH$3JS70Pr_1u)D(pfBU
zT(86wX{89R#93<|HUt=KW`dQY9Z+%q`hvG^Ru%7w?%1|%O7{Wpy@s4IG+lbDD;?v^
z4i8BeY)G=twf2@HuFX{g&zSmi;02@O-?p+~(|zxq?d7AcJ#Pt!Lkiz%Xbik<6^DX<
zjKRRui<&0$a^xQ<Up!~i8#rxd11E70<&0$>y_?-DsM#_v^gb;^V3;1mRR2SRMM~Mk
zSVF$uh86jbXR9Qy|6W^@*wc$L>a#2O1BnDf%t~{#>=k>Q@h+UIY3~o^bx;<0v|^Q7
z=-+RafH5c8!m-pOxf49Z^@%Z?@t~CuG)&jBE2T0BdiF-gl~O5HHQSdpS!$m8FzPM{
zJ+2FT6%-$a=ql+qHF)KCAuX%EKYdt^gG0Ga()NVlt;Tu>JTYCF!x`ndFifo4&0g2W
zB*jeGl=yepXG8R-&wJYnF-=-4r1c^Ma+uJx71g-gz=NSChr-?=tyJ$FACoOif&HO@
z)VZt{*BFF=q!soDy%vMRBip`{Q=pX1W>80nuzEqxOH`!$VeRdu;rt{oiQye7UKq<t
zEJa~%!{{8Jr1|)xJLyA#zyFbO3G6?*z+e3#E(P3geF5(-?Bx}bD6>*3haX;nU8*RK
zA!;b=I@gW?$tQ)G`QVv#m!qprmZvMkMOJ1UAumFX`j8gT1>A(E=Y>C=SHAw~H@fg!
z=O{LsvJ9Q}SBK63m7S@eZZLWOgO%momz2ky{UKXlOiA|NBUd4cikK*3y|m8ybk)yQ
zH)~*>%vM+~H}dM>Y4G&F`XS|)euevg@OOE9{KOopSa&WEy|+Fdh(6Go`Ud9kr}klV
z#Y{U(UzDwVz9r25F|Y4547|MT#2fTEb@7KlN`X=;J|?0i={Zx34uiUJj<jT&l}o7E
zNn)S4WKj<LL!0feLooz=^xSR>rB0r|cu!te6aFa9a`ZyEziL~a^0v0x2yu{?%$S3H
z+eyLe-@o0&<qjlr8A_2lRQAfu7mpQUlzM(ySZ?pc|F=$Y3ih!(LKmHhNWJ%zZ98K=
zdvBzWiOXWcQDJuVtH-p`+b)!$`}2WSR`;Oo8%&`Cf;EqVpEn*p#>8^Da@6MPHBF&=
z)cDG)nwQnYJ$k^#P=HfNnNfCY&9jmsA4GRJMLZu<x-<2!F>`08|8%BDKwGTlEaTF~
zpcNAz9!ZQ2g4HXJsd{8@)QLlv#UXm}L9|dCsxzH!(8qRe?dxSBF3J_?wNM|@29R#J
z<w+LOx-)09XjS`Oc8PU7(}o3;V43CWpdB36-gxvogh+H6-kN>zZx5jmp<1f4;=NMy
zJdT*NvLcQqMf;pR;dCU7ZVijUIMujtw0_clzI^dNsV`mfN0z+_!YtWwjH&j{Iq{bu
zo9k2qUI>w#4Z0l-&tS2r*jkG=pr{RLj?KiHotzh0{S0O4t&-v*BulF$rR+vub(|OU
zzn9H~hnB=yimbJx@pwFlAu{JAuUAIh-~9A*GG-Sni=Mw2BEhS}qSnIbO+79-;nT%5
z4CYPuo_*hN&iLb?c-cPc85101R05_&^_KmP^O@FmrtA2qkf9jl!tqW(VJ#|Iix%h>
zyps5WKN>Vea}9*9gb-M7U&3@|ot+G?jE!jn%G0|iX(HAOwG_(7oz@3&iB#w4pLVvN
zd_(&${~6AE)-Rt3mxcSskGS7>Cj81X%+7jUY5&rfG!Nv!`hzReM<Ex+-A^y7ZF;2#
z_PXI(q1Vb@8jA}`QlC@hR!-lK{pB+uJ+zUB1YUyRU;6Qn_~xfyQ}>;9UF9$ek=Av#
zzoDhDZ7Ru&S&II7y;5rD<EJ~{d@e92Pgl=nfFl_KEZR>ff#>t*&y4uQyUPXFJ+!V%
zJ}6)<tWV6|>7$YUtzYNzkG^65JNMIOo{r(qOW%?lM4u&sucCvd=M;x-`-A9}Rj@1<
zdf$JpXY@M_12MQ0PE(!hz#+x^%csvQ>q7R=0K9&FIk!>|TrLYC1=bXFVW*^T>kz!&
zwgRR$jXBHS0j$fyob?n!q$Ef?FDrF-e0ur7oRb&=y$wpOMC0n2ld=_V&rhsRSKtuy
zS-oY|R6Q*`q)j#3F{hA&InO|m_IRaj8)ds=pNt94^rp96jf(Ts3TTp-lDFcCcEH#L
z`u}h5+<GKCv$Xttr+tXX%&M~8(=(=}Vbm8$!vz9y#qZ`9AOR8}fiyyM17ZfI-ENQV
zuByz4efUmXylY1`7pTjKrez#UvihR!$T;l1zqQ`=9&*09<<T2@3rBL7b(Ob+mm56&
z3LRHSgE~+L&WS<ow7|x7gS`h4Dr1mDr&^3}ZOHYjHTLB~%Nds<g@%t#tZD2)SBve&
z_1d)-3U+9XXEj7l(V!b$s}pl;Y|eN9PRNG#hsqIe$6mZi_v@RdVxIm^(0-%qyYa!4
z5=WlBh^50U*p=3B)l&|O<C>}VkR?QV^&IzS$=R|53<`G63{C{}Dq7jynul7l$gECs
zOp((BbJ80XIV3jwwQBFWe!Vl!MHB(E9!-EVwtKiCJ2+@ZqFxrCAxz3@?@sXLkGZD#
zL<r7IqP!ZTUhyGWG;e`sal{sSRZ5oHB_=``q}#9A4zffT0IJ<57}+)}*CZx2`DQ&1
zol;hUQ`2Aa1~fvNsoRBhc_EG`kx|27vDcCAG^MuINW+MCo_$;O+NPWtG5F6&!Y(MP
zycHoj$$Mf<r0@zQ*L}a`x=izl7?YZOd-yxAwZY($p0ft;%lrr<{SCO}sbMfr0x1T3
zIMfdT&mc3Vab_H5g+Q&#O>Wz2Rz3kv>xnR^p?6xtdW)WK&f{yKW32h8ynqnBBz>Q~
zW#3kE5u<&pjiod$myMUp!n^Cr`TW50;RE&WzGJY^oji|1cx$1Z66-(ug17(pPl>;L
zBz%4*el9C?LbSCK<~~}%Y4x!0I@D|J;v}HwNbia3W#RGk1RZ*>V7!orH}81=-FG^B
zs)&B#KshM7W)x}Mgy5wFE>L~s``>=xyYHVRL*Fa-NUoidEAu=MQy^I*TU+D#d8LIH
z9?p|c-6p5pPN%L5wc2_)C~4LO&p6KXyfF+DaXibGIB$B*QVc0cw7gqDpTD8C!*HxW
zxHU(gQa`8}NI~SA(mG>Gg8qRidE$BI`DNvO`M{eu50adCx#&>^wO~_MumBMs2Ro3n
zrt#nz$B8(OG?S==-f=FotP3TtOykIKdLqnc;+z<Yw3pEbTyG3x!ZrIFL3XYTGo{IM
z{$QqK2Zbbe#z-w<g-U;J83Fq)2Is!8uPfWO<9%S9C&BhM-Gc40a%R~S4LS_Us4gW_
z%{efaS9{tt41%BOH>mpv*a5MEy$`o(Cfj<!oBLHvvLWux;(2@HNXhebi_w(9u=kLn
zmqgKvXxYn7+gDlYo+e66P>)GhU2iY)`<H#R6cWyPT!_SRCJZya6<k><b!BgIzY3w-
zp(@?+7{H?@Y~N*yA%Ai;N$y}9w{E|U!#lYw`mPD*XaWSqSMdh6<>uU2Ycc4x(aS1n
zmD!EPaG)JVLOOCA5O1YbH4#Tcz~MTVcj~@rBJ}A0bCwzI{LRndXw>8qsYAXQ8#s7J
z%YwJDpq|^#(hwDF=iTjIDS5;7Oc>K`r}<;97y==BO3{6fPq->+YBYsOE&I*Iu4GNw
zwN`R*3{jHLy07HDFwCQjc1or9N^q~HEo#YL3B1l7xg3>V8odh%Y^@PuVsK~H>-DC3
zsb$A|#R1O`?^w4@Uh=(*IJR%E+VYMmjx7DM3km9ZC$ATon)z3Qu3C0lsUpsqh4z7x
z9@nAv`n!4Ci<BKqqS%@Sl|2{YG&4=-8?4`3gC_nFU7wo(p&vHd7rY-iKYq@7eJ?_+
z>DX(nq^Owr-U|EHnCJ8DI}9lilZ=#(=wf-GbUH;6x;6k!3K1{&+2@ZRInNWP(*vaz
z=Ep~B$&6-~j01;AihJI<{@Y*k{2#tyo)hsOz9anQ&x!y1Piem#@Ieh1E<l=aArZI6
zwr7U2vzN+ndeEWXJM|QJ{^o^K(>nCHHgYM9X(Ii4BQF=$bt68#F?1Jds<_YfoQnju
zoXJ_J_Q&UiufP75Wh>0H)}UK%l9RW@)-w<1$atQ$KE<(>&Qvqo<%1yq>4b{|tw>hA
ze*YajJrPe&gzhOzC8d$g2Zlis%#NDmqsdf?s+2V!pFeVbcx0ST;*&ISo8)YIkf}f6
z9{nLV^fXV5(<Df`cWC8?7EF@^241d3Z!~#o>7;3-?wJq+`MTh`%rXd(`Qa_yb)IXJ
zTn|GDmX+=!anRcv2a(cRxBZ?W2fb+z**5LAC6{whm3iPoB26=4m?WWVh1?B7^g;U;
zlcIUL41ij<z@SFJQA)9izsb|UN3)I8j$cZGsDEU8XFF(2Z!0&zI7~W>!frAnv1S#z
z48$acb&-Cy+9ddj6f|S&%bs=!7N2ZLMy(mw6|q{(gF6I00KGr-$lai59gwXz3wzO8
zrng33H}-Wk5?o|HKQKH<MFCzO1^cSMTME5=#03acLg*AS%U-D|;-~4RR?w=!(Yllk
zvg{3(xgdy7lQ@F7yf;62N6vDUJKVQ!VIyGD)vUFLm(6mpIP2iVAlL5INiOLeKAHD`
z#szN5kN#>g8_dY#2pGbv2?;TY6xj6dI(y$oQ`cK(NX`=0G)3Jkyq&28Y|?PUNveaA
zDceHXcY`bkV-FX$Mf&{12mZ%erEEabgdN2uZ$;=;DVtVXwHWcz_U7J`1|`0A?`Wkm
zOp@|}$g9WqR{<XfqEiFmf;D@_CFu>uG*V5fC#=9sr<&GM+rF58MY6bQWuw&ESe8Wr
zYEe6DhXr84?Y+mSJcVLA`QSV$1j@FWO2LR;A>cd~(d>;|!fg$uVac;M@u7Sgjbmjt
z-7fGjYf{eJRnUG)LKli>IGu$e90pA^umd8E13rz!GRbI3N1)bvHn7y{iC$604t=(1
zI^pn?vftJg&N=eFQEFv6&kWA7uUBL^CfVMbw@<wN>@6XT?3?!7`?|;@!3$}`N%3=<
zCvcJF<;vw!nE&P@<F}RmufHb$gEQ$ot10_uKc_t=LQ3q<7oI<Sq&-YZmVEPsdl<>z
zDmiui`tSMhcH#WyiQXLjt7n$~^ee6B{>L-r{r5~E;?E~a-h~Jbjs9+6Su@YqjUnmt
zuIt7&P7JxRyj)pAr&mXtXCD8>pY!}?fKz1tr(d$4#3QL&V@M)3zPxxLWQXG_y2o>A
zgi__=_|-@H-@MnV=5(Uwjl5?)3oQW61y30C?62KI>5QD_y%zx;5%afgAq~?f-J?IK
z83-YghRBpmb{9RNS3;8gtB;OxGWWa~@F_5)MDt3|a!ssX7Zc;d6T?}%l8{E$WhIwN
z7`QQB8xRcco~DVtR4_;kB%v#Lr*2p3uJ^QWs~Q16im)}zH@i-5H*US{_+$4nLOH%b
zXA_?7=@mHR06~0d7Ey3tF4R(FP%~+QE%rB}Y1L!&wC=%2d^()LWLR=Qmz_hoP-&sv
zc2&pjGa$0fk?kGR`0AtuBJ+7q=%&<gnANDgmY3&1ZJn}7o`|_Ew$}0EO!0wSWoEKn
zFY4I+KuWeM*u}mDM_m^7?Fv{9c@e7POkZDHmn&CNq9C>6X*f5~9Mqm~tE1o@DTxHx
zOOeHOQ}aYfwD?Nkqh_;TjSvC!5)17TF^u4Yy!^A!o7UyUel$?td8$*hQES1KZUo25
zFwd6VuEwh#oPa~k;6(!U-s8e}LnJUAZ?Oi^fzXI3l9CISvCthZMtL~I0T(@eTWB^}
zX?n7h%2EWAoASduXNSKb0!`EAkGXoSmJ&DG{k58$)6wcq3<;lP95$aGAx?^p4M_*P
zAtPbX-oILmvFi>WkKJ;hl_LYe-Q;3n7)83PMJR+f8{2)=f!;mN1(M})sKF?VCOv+5
zcw*f)v}zC%w60PCte0ob=SM)A?;+~h<h>L&1R<OU*;N)832#O@&U<3C;2L){>7BA&
zCFu=;VkBMXBPoqQ7hj-D<xs2n{M&#bB+OH6oe+bV=)J4)N>Rgs41glNRO5I_ny4T%
zZr9*2${u>(6nd1#fu5T^qdgB#Z}gpoEaaeT<+W|Q_z#2DnV|^%i1CC56H^3vPqG6y
zjsuge&{{97r4n(J|N2{o@uKJKUwy@Olkv~T>&~x#{WU-T><#C6VmObCk0(BS^Pc6h
zQfg%zA}s_U@cFw(E|)9Uz4FDow>O0x4)KeC`f-PkJD2w_avywrkOXmCIiDRb&oBJ)
zXJ7EE-~2swhxyduU;Y=i%gXub%=U}Fg{OCffAS^2ZO=S^88|&oq%h)ABi2T88i*;G
zt#-I9jE{x-Z~jimaci_v$?FPX)a<owii=OO)~-$Ihv)OadR=f;jbH0k>l_z&czAO=
zmwsRUFlk|Mlw~1KBhxf7q{Mc;=tJQM9nLAzTO-bAU6j2-(mvlAj6I&tjMJnnf_-Dl
zm5+<M*zt5?93=a?UROfzj6>q|_((FQ_~GnV%C5knQgODsALkPxjil*>PXicSc);AY
zaOa4~(KOj<u1E*mn}vDE=`WZ~YCni#rtyj~QCnf(cXE|RzU7_6Wy^Ih=+ZlykAx`W
zl4Ii6q31PC6s30Uy^~_lF3RLi&gtN6bdCUv<5&=3#6#5dO)WU|zlWc`4;J)NEdtXS
zoEQh<`6MgfRu$XZuj(>mB#jesROrj)<pb69+trYgG(^U6pf~O8j`T<Af)8!59iUkc
zi5uskb%X1j$xVg$$XTv%vrfjNXwhS8@22JUHx!|}K=%W#*PFFyaHy%PmUUnVgS-IB
zD|vMo)o9@HNs5Dh_+@mV3Wr6gtXKVwY^oL1VWeWe$)VH;&QPFWMo`WWh<0>(T5;@)
zJSI2}uP7p}IUOF?39Z_M_v5Vt%El?SH5gi^CaAG*SH|hYl$6#soZk^rv<F}&jV7sT
zMaWl(&|lRcA!@L2KA1v73KTHj58(V@CO-<FiD@*XD4FH7f_fBRUT#cx9~{#-%16S;
zY$=U+AF2HUD3IuNjTLT46*S$SDD=wPq11t#7kZPuu@54_VllfNAssl6Ytq%HQ2|PI
zU#MBO&avy-7}7-dGDIqQGj&S1kpx2sl2Eov-4;Fjy2zVlTWmcx5Cx4u_q2V{H5n4r
zPAwa0oJ9Z)p0e+Z^CO2d<d8&qZ6({_I9cPTRYMvlmh07iCwWKo<Gv2&+u;HakCD7>
zEUPTguNSR`#@ra9XDglWUp9Jv;hXm#X|3@7vU1rgynp8LVdVO9W!(z%G?Qy%xh!J)
zx61Q%C6~tF9bbI@KupR834@X}n}d0tiKhpA2<+>|<z?Y#@7@BPm=a^^{KcRD8Nd0>
zZ+ZLlfb))d8o9h&;8&H)uU<I)?KeDs`++t<KaK2v`E%Sa-Vh$o?1nPd^$M3OAJ?7f
z&)&no{~EfuMtNIkITO>Y-*Ontb{+E)Xx4Bh;xu7Vb|mkNJiizj_tWmtA7(n_9ldlG
zGg}zOf$8y)k_+pya{cfux4>~=OoL=I+alMzqL7#AJQL?L+m=~Ae#c%a%eqmWXRVI+
zA71$4vqv#Xn>@X{b0pibMw{H^$S1QB9F{H?7$4pc=ZD*Fma=oWQ|a*W5E4g1o|3h=
zLC}XIR{R~mfa4!&@lBTJB5f~}eP!RWzK<c<?jB01T$a@o5Q67x&LZ^%X;&TU;JqE7
z2Q#_#Mrl%O93@o-A{!ntyR8z#ARE)1*_R8HYAOgNHTF`-_WIp9x=)1D6LEgTkF!FZ
zww1kP#&N`l2{)c>2Pus!Au58|JHjA2REV-34aqYc-?c&RwN)_RZCV90@bKQ#tKyuS
zgHkRyI^!_h<T$=Xaw)Ve)AEZR^v)5S!_Qjz$kgc7Hn#f+J<uq|D6Sm73C?KMei#(~
zkp<;%>ngHoIt+rk?OMfksFomhATYHQLn%Za?Z~tV>@%Wk$lG=3jGIzxrEDvQD_bx$
zpquUXu{dpq+%t_hGePp+>GxbOz>m8b5I4kDUe0UpX~Z@9s}o$Iv#nQLR4zk^k-RP9
zh_ph>8)=wrouCHvh|yLGougKaw3*cGb4<C$?dWn%^&whHyrB#(>3xPo?^+r95KK<n
z7^j(%Gr3mAG*D~j;qk4S2n_uILh__>qHHTQi$@@JilyxZhi{G)2TGO(-!)8n33*$H
zQF7jr)kIT^?7zDrcI~lce5=M~{ip%VRtdFs#xa^zbGfa*swLkMqEHp@w8991I8GuL
z8i{r<W*O{YUp}g@8Rf$f!$9wHMBLU)Oi_6=VNg?Tgv45fD!Sk)Rj!IDCUaoaD#kgj
z5o;Cy>$+U<M$UB3QP-=uXFjo+tbH6KrFW(w;ru}E0oOXuFDtbc&ZB4F6*oRlk9_u4
zP0V%QSaRm+bjBO9+`AHAWj0lT%AQw37>HwFoE~|2del18xF>nP;)CNCzxV?0<o^;v
zq1D7R1zw&vwrk-}zIdV+DZiHg{97*n@i#0~=4lp5HgBA!L|u34=NWH`u9n3+i>EU#
zDqyedMR^lds;n5iES!5K>0rTyL29<viE+gHz`p+hf{F-^*0kSQcR7HaripIa)DSf-
z)hcP6bisvcegnh#f#L1v#Pbv7a$(uNl`%%|cF1?31TQPiIWFt2^@8nPL$B1VeqRLa
ze%q6TG!W+p!gy9E>*Q<Sjp6J40krC{D5aQtoetI*rK9iFc8*>o6b>R`5g9E66m<&>
zG2skuK0K_CoAhAJSGzG<`5l8G+YPqX=(UPjU2T1FFwH|S38R{Tqq~>f=M`4&u4;PP
zH%-#XGng#!=*Ua9?C}tZ^NDbJOB@Cx={dSh^wLNFPE)dOMx@geH%${E$aeRbt{slp
z`%TARYbDs<-YzdV9h;H~g~T`xWT&{)TIKK+yvIk6@0yNNkW{R!3)pm5o4EQVXG#<A
z!6Q=-4C3r{Cs36fXb`k070p&YkvJafm1>Qb!bjQ>Ts0jT>Vk`dna8wR48_oLr?iT>
zjS`dC@vh4muJ%p}5gP0{YdsdC6g1X^a%)ii?LqBKXO7duID|;|GEZ_pay5Z-hgnmz
zM!eyX6?PhD-LKA3yJMIh#ExHAd3q#Gn{~VDfOk;yiZ>G-@H!ydPVb&L%y=Iah_-J^
z3bnOPEm}Rqlo-ZWb&a#@Tu_k(cC|t&mJVkq+c?jpK`Nm*M!ZjyvJlMZs_q*{hKNJq
zRoxi@c(3Maml@_0;BnqQgP~nGHL_*9QmgsrcO63ezTll~n1lDEG$^XPWJ2r&*J))Z
zjZb<O*cw6&tv7^BmfJb+a?VQ=<7vJ<zk>I;UZ_>~OQ%taEQw=?<n1CEUl@gAmYpM5
ztoK411~b2@tk)Nr0JY4Pck;e*dU(?FsY_=s)Yiy)@mQ%8;&dXUfxNAJ{>6lIp6|c=
zz+M_}pB@;dfmYzn!-?y*^Zv3BT;Sc)nRy7jdlPx{SZG9Y?t~yNSR6g2JKjD`!~;?<
zhFBVEOf%G27XhVZ$@f~*D$04%{J`b;y^Na15#73ul+Jk9iC&H5{P2$Ta>ez=yT>Qi
zW#i*_->_UVr#GJwQ)J&OaUAHaQfwH_+iotMmD*N@^hh=L$TW<%_bdD2M9rBH2gY&K
zZxSOkHNfEy!5Qcs(lk*wNwaRYprumx%rH%|kPMO14bqO0X_)ZC%=qx6$?n*9I|VYt
z;IwO-6TLQaE?oD@o(m73J<6ctcl5l691SyqiNnNre#3Zvk`1KSPP`O@j*s1I$7Dh}
z;7b)eU-M2%NjpMgZH5>mYpa?NYSRs&FG5qW)DO}#XPe0Lz5@;)y&ePxHy?+sj4(Vm
zr#+kgz2`NS(;i8|WR0y>qC>k)WN1S}VM?~wkHaXbP?wK)-HXYmV2GMhn-Leo?fbCk
z;(gQmOA5>ImX}bBP}xng=;-+F%$TU(P?bY_9MV7<G(B&-*!(FCa*On`6+Rf^IIS^q
zEyR)uHl>6#YEPU3twauoFzq)zr4{n7{cG@nU>*Q%Uqxt*0iWWHaM$hk?gvpS_}ll|
z4=JIM5+mkmurE4Hyre#o$VN!A)NXrbUw2I=^GHggywY{0(HgjBzp;<>W-f@%ftPEh
zjFr?)I85u^Wnv>A0|2dA{HRFpA9D#UD@5~c4<XWR%A!`0s$#&$8G5x9j~XSd19xb<
z8Si9D-e`TrxhSKvekf8LWxHw}kVZrIbkDWY%(*leXGV77YAC9RS6*HtC9j&OOJNv*
zR@sYkEFch*cqHDMw_XZ(M;K?e<??!8)Yqn=w??sDt_wj;V)Kl{C^XAkP*pGN>&7rU
ztt3W~{UY29jbiZZdDUPsj08pwQl$?<I9knslSAmCG_omElIdpg&z@GdSJbiY5R?Ov
z_eH6CabVvrcs%2H(y-GDt`~YkzA3fe*y}NR98vM~S=aeEj&|==xuUja#5k|mX8_mn
zG3hxTBF@RQtklZSe*QUN<X~I93`5|v6nXsckyc@v662IeF|aHz48zR2?bNa`O=q@!
z<4^x&B5w*~>rK)2*XyDoA_Pkgtn7OxHj!0b6!c#A4Ih=*JET#h*BFT2Oy>**PAO5#
zO4}|--K@_t&58cmTU@P-r%B!c+s3k9CDZguK}^$3tqWm%r0wt7_6y!8YN_P1-6Xij
z01=YE4ZeyxmP6q!`{WbqlbV4emwj0_hRJbxzVP@kD|`q9A>)bjnRI%@nN+3bg-WKo
zz%-x8U11Nd75aK*jGeq?)+G~C;>~%c*Nk^!_75R3&a?9AOC=Z6pW9@a*9*}H#?u37
zKHuP8TA}Qlc>**$VAVu8(~lpv63z!otK_TQIJVbL<0$`$Vv@+BJv<R3Xmt8+L=TWr
zk2z3v`_6Kyc;84te)`+Kn-S9yY3oH=0^=}v?ThRtm48CSH3bB%b-j@vV~kKEzI5U=
z6ULcV3uRgK*_w<(oI$TH7fNlEy>cRHjZg~v%SUL1VHA8_OU3PC!f)GhbAbyXQr1Ok
z1&e5P=Cz;(x|su?I0@dl+h^<@YH6%_1>~o`nJiV-pcaiB1Fm-PnOd5X#zNHI-HC^w
zpV8!+t!X0b;(7?tfNjqfQ;hyTW{PCogJW_#k_8XxA|Bel3sG{i(9ZiN<Y*Z5K(Cp-
zb=)v88~ID*ElqbgZ|K0nEj3TeZbhwvK}Xns<4j*}PM)Kh*;X6A7E7#ql?NU7;*X_{
z_j|qWN-=d_Q%?}VveZs#ncxb2SA*Ppr7WAG!i{G$&pHrGv(+7eJ!iZx>{a);YY^;j
zwzdm?KQaywqE>4rIQdc-VT65KjG!yGz-nG3I=4E@sx?n-mDV@L>4`00L`01f-8Cp1
ztrXJys0Q17QviA{;;{6FORt^)rO0%tRw)|f+oB7xBc^_Lg=s!3qoNk%FdQh&Q=7-3
zKqftU%=jx8Fm8+wa%jXQoABkXN)sd>hQOXTy7}3g;{eWi)^)`J>$c&1qSVefO!N?0
zFV7s#ezkVmUhpomF8Xhj;_1oDZpse5a&JZcCU&osQV0Zc(dTo$k@HSzkOrsMrW%T-
zrq0a4i79D?<2_Gr-zqGs?l{x;<Dwx=C-zldF5M}i(v5-MWs27c4AY6=p_E413NgqI
ze_cgR4oRzwea}#I5SOX}b6w>z6TN3I*({75ok6zv4tlp{frhtGBAnh2J~GTFa%~JD
z*z+q|^PK5DvF%q<lw<C(@~X8nO%tVN^1hKmVqF$;UbR+3eh1E{|K(o%-_<8I1JBRT
z0Q~iT`_=FHH~;&myxb3~-+cRj|22Q|_1AywH*l~2!+L&x=FOWof9$vM$#pzC{?)(x
zt3UEPx!1jZ+V%UM9j<@EsL@*EtFOM|@$r$v2l-z2y4U}^>b>*){LGhMetBcZ{}}52
z?6}vx{<!t2&yG)O2JUsQd)@0^_qx~rzxCr8G46G*d)@0^_qx{~r0&hYz3z3dd)@0^
zKk>RZ1NXYuz3z3dd;P@g-VEI9UiZ4!z3%lBuX{6auY29=UiZ4!PrUBUz`gEuuY29=
qUO(}=Hv{*&*S+p_uY3K(>wf|Zatb6G|8Z0R0000<MNUMnLSTZ3zZd5K

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_009.png b/0.16/_images/sphx_glr_plot_visualization_utils_009.png
new file mode 100644
index 0000000000000000000000000000000000000000..ec940440858b21cdc651d138f39c24301e9222ee
GIT binary patch
literal 242568
zcmbSyWl$VZvo?_6?h+&fU)+L2a9G?nu(-?Oi$fq-a0nXQ-Q5XU+!rTUfFO$mhv4Dz
zR(<bx|KED1YpQ0Z&Y$Vi-OuT#qcqhOaj~edkdTmYm6hbQk&uwX|6>?vFHeXVPL*D6
zA|CSk9y%^I9^U5e)<|mR9<Giq9**`FZ@sMDq4qA${9Hm@{2XuXJUm>XqTJk0|NjFn
z7k69ksSmRIFN<KhDj7hLknqj_W60%FW%fwONJz?Z(z-s`f1Xs5twxh-B0f1J^sF=R
zRjK#B-P}bj%a*x~xkEpz;wom7L1|wv12C+$WmLG=&x=>0w|~N$QT!p@sRY!swpQ?Q
z6q*~JZAk;my_PclUwpWedcb`Qc=|7%;YYUGw|Dpd4}Lv?%;9-;|Hab({bkh?Q_#rz
z9U|zT?f0PP_>8BTjXC(ze-U)X{_gg;ZR1ws`9Ml*>^V^CjGQSv=pQ0zFX)!;U&M30
z(9Pxbe-Siw)cg4Dxi9D+{P`~E=J{D(dF$ck6zOr}dH5N<M`)^P20r%iFe>%$iTm)a
z1A>a&Qr)4Zdwqja^6v8aqW7Zi`Sf|L_pvT$?{YZw^4DEg(7pdjZ}wVm&m24y#9l%I
z6m)RXubL3?HPMii%6q&s4MO;-(&mwG`)Xm$YiC$qoNtdJPU$8fZdIwKz3UwtqEC*l
zy8AhPUJjh|+6N`m;F2_;KCIsW!uxIeMDBGu^h#X}pwClOU#i2gU*p!5)18CfwynT6
zg!Kbw!hqcsg~OF>=-mWznf@Sf7o-mQVHDHCe2e70XB3OU9J3f#MA4AViv~d@{1UFk
z<6uuctSs>+<@4`bDhf?VSsQVTOv4w_emA9KznbtL27%@_>7?&Qq+D)88klLZ^*@-E
zXjgxD{aQG<t7*}yCiS@%fjEh6z`&mZm7`7-x%Y1DKYiA|*w^dnU7X2+R=?)WIpXce
z4L#3}<vmAjc@I1qFq%M^#f?&hijWhk_s}%ADM-9&dk{moO(!V0;Es=#YhplwHPtN?
z+l8%fdU~O~K}9XZF>e>c>qf7$2i&76p(b)1@t0G+YnWKHu5mFP%n8FTkN>uX7O9&-
z2Q&`BvvFYT9?$s2@{)bZ8rbFe=Jxrfp9uk;P5Q1xeYF23t=wny6fk|rnS902MXD`>
zUCWUw$)~y%hrqniY-0Q$`T9lr1%f$oVD=il=z`17mB55Sa16rp93l0*d9xwqE6(*U
z`@z>em!;-CaN*NaKtTOoAGN-E`!=EYa5(za<K>f=c0BMXr)~93jTiWUgny`$Vu=N6
zES{%Pdo=2a3)+8b1GR6stih8XR^QkRArkQa?cThY>GMR;)6&M>?fItH`|mwHmp8rL
z=i_x7cYps)xd{M~y(z{kY6$R_jHuhjp0eDs!!N6wA1A_Mg$FPAvxxh6sz|B?AfFE&
z7i1usfaYI6%%%n}<dFy!IKx-23{2pommNOwEa{S29cY$DvAh&et?C_HiwQz;DV9rj
zj@~7=_xI*L!u{mi?vg>OxxMI8f#N)X-t3LHW_Q1X?t}hm)cO1iVr6q2DQ}tbJz6`>
zxpeIA-SB<>H&qvO!)|^S;8Dt5O%m^?nWhCqMK_eLiff`<n@V1N8d`08b8{mUA2{t`
zxr?snuq%RgJ)_UgQ<Z)Y%GRP4WByvRycFqxDHJVE`^~X1>&=%O8g_-#mp`x&yXrRU
zV2w7h>xxT8%)kRgTT_?9PLUC^<SZj-))sNeS)pnnci_zwtUja^DfnTp@N;0v>K;jE
zqxXHhJKy3vX3db@;P5%W5>lp06ZUX3lVmVnz!Z%TTgi+=8J*CIK?54enV`pKm6EnM
zg0b~jUy@K%r1L3gu}zI52#xUy%_ypbQ=%2=%kKnZB++H0Lw|gOs-<{<G=Z8Ed>@Rf
zTZF+ydhItuHF@(Z`JhDr)=Z0~{Ha#dYDR0iY)8>G&<TAIyF?EQ)n%%A2oP$3{hf?3
zI46Aejo_<AUs_ROy;IS?1}sI1{T}wjZ+*vq*<ISC+<Jq`(iCn75N5S8@C1!TK_330
z+a2Fevn#M;ul``(oZu#lEiIp}myIDK77J!6;b4kzNCB7En4>{OJw3Wr!6an3C=(<t
zen<BlM#ywnas}5z?w@*Rv5Z2?u~AMCIS<FUMXs&2{9Ec-p;~3RS{3t)6$X*M?a?Kw
zps@$hekg7lw|K}e{Hgv7%PsSH&_PdcuBE=F0s{(=KU!7}O^X}oIG9fcK22Hd*x0ze
z{wAfRGT4XVJqAd@LQf4O%bgQ0Bb>nuHFU`yR+bXIc~GtiU+pqKo$e*QZyR$7;u3(Y
z-NSsv_7zKLQoX#uHFB>#Wg5(UG*RwnL|oz?Cfj>ys1-}z50dw@%7@Q+;+CSbqr}Mc
zpFJ8Fx?F)OCR1E-hp1HZ(%Hz;Eaof4+}Fw4Abh3Ka12CdDnxh4?p7C>EMR@+ZZEb%
zrlzIO*9*)oC!EZ38G`UXdn8*6>f(*CBnDRKaz8WYKGOlQ>ZI1NhP7T|@9?r7?@JKw
zS=WcSP0Fj5k@TyU#A(<{vK)bdUzk`)e{G!Jcin`p9^+~<qv=Bk4eZ0vzZ?Wl3}#~U
zu?-jzOle^wj(2{E4QuVUP2G9T{=&Bl?<lRbht^kqMr%hOPi^VnzTQfkZ4BjckY4W5
zb$!BV2(c!<QQ*vJK_k;!Fw?u1=S}O6?zYNs)ej#oiFfGi#AwnyH`0ZLcLBiOiy^KI
zeJ0XHtFW2Vt8HCAz@{qBFt7+*pv_iN*h|)IVHO^FSGo6!h0N#mRw)r6x-;kTdGo_2
zngXkV-Y4UPD6NGYIkSRIT@@IZDH^YrlXY2nGF_9V?G-!LYsrw=n2;-N`5o*ublNnU
zw4|U06v>Lx#?Peb?)A0mTE8QAs}Ew&=Y9E#D6$ocdIW*?^D81JE&&gJ@dJV$#GHry
zb9+aPG-}VyCPv%200pa!A@<}NLQlONx7}zbu0coCPjTDj&3Gm1An|ffi$PYK`Aqog
zOJ<pmT9CI#Hcvqzt5jc)>mWM;e(j+FSnws*ZS70NS%ibdN}6#P{r(b)O34Lt6n|Kk
zcEUByZ4*<`6Ed@Y%eQD03*j67sJQRT9dlp=7omj52$DlND@i3>-bNqTqDTJd?-&|f
zb-VBi+Ws@VWH4MoD98RfJdk#+u)ez?z#P<069;<JLJODz%$uaQXm@r_Ejp*z&D+%>
zi6e>^pKxL$`&<!JF}!4-Kzlet9?rGT@YfK)4tT~t8Wh!0zfM{={mDSRTE0gJe02Yi
zziO_R`iYa|m5`@7Dk-cbC!O>u*4j)Ub|@Bo%ePd;T3++0kPcM_b<UTz8T5m??7}g$
z=#641m`W-zh$m0FK^#3uIs3UPNv3Y?l%;KT)b}XhCCd7WRkR%DTi0{Rx9ospM93f<
z`kL(2irY`eLv5!Uw|~clR@|n+<gtZwxeGIWORv4b6%-*N;y=FW#1Ibdx48epb{QVc
zFdkUF@s7WF)g@Ryl#ZETma_8tusU_YZGYNE<utf6$6Aw`eQ1WtMnQksHrlntG7-7?
zi_T$iM#wvg@tBZ~H?+vFK%hS%;Ov#yB2tN_X*eB--U9;*1`RwkkemJ!cA3AJh4O*7
zywa<$ARnvLO52r0ZX#BbIHt+fwB3-In&!k+TIYKT-H1xvS^mgTshOF)arkC+;XnHM
zF*#ODA6^SCAc+HoClxl25QrDZeT!0=^y@lXw{8LzQh@@!*%=Y6e*Z}$B_IG*(sRmz
z{E?Jjzx7An{f#TCHcnMISX8bAT$C1C6twO2Fy-sN{l{BxL{fi=v9u|GXEE7&%7W*3
zmfwh+KP}1HEk5w<xP8uH5a^n%AjNKp{I{5_5iHR6ZSync6YE;xT*434LB%k$5kb<T
zr!h(HkX^1@d(2GwSP-DZ>xb*o>12|$4{CqTE|-2Z2i}O4VN4SG=64fsp0lu8lE*&J
z(}~x-GCLhwFoY$cB@8Twi#m-}qx&dqch?f2HpIW*`#84hhS4aqUOL9iq#=uHCnl;s
z_X+n?k9Mc%Ns(-ZYU~?QAbqu#h`HAzUrv;!t>CAF?hHZ*FJkjN+vr<#6Kit)kQ+|U
zwKks~6}Gh0Bvm3!S_HoZfLAz$HLR-WvcO`ZvceUsha4&9$2-<2nRYR)3tMjKikQY}
z_$z%#igv?p3YA<<I%SWbpOl@R=F$}l8Ezc`lm~8Rlq$jni?t+2dJa}DSOlKNf#QAv
z+H{<<Gj=0@k{s53BFx`Cf>i|Li!NU(l-u2<)3uG0?w%gEY;PWlrNO}x**NR)A`bDr
z9jZ{CdT=_F2pcH85Mo&Oc(590A_j`PG5uBF^jnX#t8SBuAyjIj0ppWZ`a5P2dpmIh
z#R(62i_D%MN7(VAwq47tgB~A;K}A}(dh5Bqaiw5(M+lheTJxbZz{0XIU>ZPOVpbYs
z5Dmy}i5+Lr|2+f+G_+_6X^PpPRyo`uLd+WI1=;0E#p+dq{k$!LJKr%40oyTM@K96F
zrh{Ig`hWUv!*sqkL2ea6D!qK#L((z{-IqvUqW}Q3P$>-{cP%O6F?vXnbmS6#x%B4s
zKd>!_V>?O#zNQ@Bdmek#W~rbY^JM)P6R{%oPiO@zZP!yz$7=enBIVb`H$qHdm+CII
zUR+bLgO=iy{|1~0Ic&tE_!_9fC|gkDP{^ozW&znf+L>DZ(;9ZKq1Hlnoww<Po^H8^
zNbw+2N0oy4LN0f%q3I>x0QE&2(vLqT?&5L_Y><E?FtbXZu|h{q2^N&eS-<UlN~=bF
z%?Rqc0T>tUnrXOBn2iaQKZ*JJXAPFp=pqWh`##y5KCBkK_?1YEA<tbVH*l3)b#@%9
zWD2XkKqY$5dx;I%OFDS%u&r6?rKzEL1FOwiJeKZS)*j_YOS$?l(YY<+VhTM$>xihX
z=r6a_VaD^a^0kVtSEaAJLO|HVZe}E4t&d%*#%tM^sUv;n@jq}pUnLGAiH7=p7Xu~9
zD>0j(Bt#7}>ZdHum)jPAKGgr#YKWC>Nl`5(wa~^|PUMRJrGN8AlVWKHD)GFX+~@ng
zIz+JO2FPBD-}KowReKy+xAxpBP;#sDvvP&&FR_i@mp$Hl{_o$=Qy}7x5E5-V(U0rR
zJCSxtvP;Zuu+fsPUTe03R+S((N63Tk+0kmw%|o0_aMM-~Ad5`;-B!qlzBt~9q+t4h
z1->JY#QxpwPDgeJ7dU(E>VP@G!nYw5G&4+{c<YaW@a!Lh1ppG`8tiJ^u9CIcTc~W*
zOE%fjHCh~mmpnFOXa+niOn6WXar0`ejPSu#eM9ijhhsHPU*Gy@E>r2$SER<BBTag-
zM`Lnogtc^iqD~?5gOhA6n{)JlWMxk2<ps6l6rM@h!r3Zw2p>g?Go@9^4~q}J%<e*=
zv)Qwqc>XitMRrQ>>9Bjrq!uUU$f&?yM5j5z48B1p!{MVBgIA+a&f4*c+PklM+gcI7
zWT$Xfv^o~v+yrhPKTi2Rq|U0pP^RN)qbiDOo=L1B^rcJFjLXo?DA9cM9<&l}pb#u}
z{Gi9IRoJkEs+l%Zae3CY+2I2ntMpw6soE>$P%nYW<Vuw5E7?&;wBDy9`PTve(yK?1
zPE!SwyEXHhr>O65L?lLEm2SDC`FXC!9EOxtB6odOQL-Z@#7Sz@Z>(b7%<L7BE}QPw
z>?t>-Plxg>TEs<dQ`-(#5^1_D8+UusS`v^{<Lm_h==B*U(j_S%7+)l)T10Bd9VF^`
z7EzA=9zHf=?v>|2Fjs0O?b^m1)HJUX!jxm0(qHEekO@O<!}plSSuXj8EP3K#yI5zu
zp;W2A!?$97;Z@pecZ-*|K*hR72xwyq-i?M)@qUt-dI!<5wfVEGWj!%y^o5iuQG~{{
ze7ios4a>y6bw!}RLSFXbk6g%)nCRR<>WKznmtZ*dDGYqrw^IFluDm6jg`%y6It`a4
zP0?sj<8aMJiRJY*3GL8L1+z78=hLA-c}*wOnk`cy8|kF)4kJS8dV|g$eTYO{KXuXF
zf&->(TN|Z0aKX-qxVAQL=VAIk)4C<0D1s*;qDl5vCKtVNMIb+N2D_TKixXs@_#{Hj
zk=tiF(G(A@-W~DlkE03|IIx2>IZcEeA*m+ep}EU?lRS2~c}(m|Fhycii|L>?uj#b8
zAu~#g*m(su5u8)>8bO%MU<T?U$wYA|i-&zD%PSdUwf2zOaIGV07jL`k!TQKuoLnEA
zT+0J~K|&;5tt3suNQ2!1TWisg2Hk~tgnSo4m`?r+mE!#@kc7V%a^#+Cc@|*eaC-jd
zP&E|BtW=jyG_qnWBm7Y~K_rrtN@1xG+0o0v+%;Fox$X4v{)zm1Zx57K2&^~lK<Qdr
zHnqw}ar9C*Tu@7;9}(+ohK<wqXlm#PX};N#*EA#nk0Pcyl1IiOa>%{U(aZ7j_gs5i
zspA&IgHPW|l=7p+V#2qo@!?R>bW6M2Wh^KcP)n%2t>{k&p;d%g2-o~FXBzei)O(Q*
z?tvlNcvf9V4#n|&kmnHRuqDc{b95jp8}r|Nqv^*+)@=3JJiHapb6}yq@yv&BsnCFk
z3TKkBNgt5bG{|96AHsWyxFQ$jm4#+4ims#flkVaU>4(@hVKx$Il78R}aTI9R8@7?Z
zflqfZkg$E@bW~6MQ)m0MaQwx0iaQKg{)Q3=tf~5dX(GETC@*8cUQbj$LZY?9DnHjf
zE#5*_zN`l3^*Rk^d32JuQ!FGa0aau6#q8>;T)<4v?vF&n9}39ELJnZ*p?ivw_SV-B
zoLb1mLxM)&^Eqt$T=#u!xFGJ^!MBHgUlU~qTYSX10w4d4ZeLcILmG;gDpq{LSU9$R
z`bn5lrT!A-&o*md*1UQGwXTFP%>N|~E%F5OgLTA&mS(NXHDzRfBPh6UcbfAJOXj<c
zFr|F?I08_&fsbogkjm8W>n`(`n}2eO=+Y@&j&sjeLV3AhOm-s<i^BIg?jSv;mQCoU
zr6y$8D|;;=d3z*`Ao!tA=EJwtq*uR3WUIzo(?VbrA&3bF=3SwwZJKfi_^{bM0d(wb
zC7z=+^q#bnqwj%OF*@wqMd<*M_0>J~S(pi6m&buSy}uic7<Z=b;E^JESn&0cyqA#P
z1Xh*8I0Z4KU7y*l)vZNlQ}j03l|x^X;WS=RM&Q({@7Z{pOMw5gf(yD?wV?c6?4VI0
zQmwuj><r!VY3K?+T5U{==XJ$(Zfv=UU5)IIY_VxmRs6%*KkTlRUsUx;CIl4uh$qM}
z((HPwE>8{_`^%<^xcywNUip1Y^^=hD2{fq|9Q<0;=uq#Dch$njZt^w7Ec=VKc3u|9
z0QC((e|PhrV5HtGa}*X8_-N`IRXbffH}Aki${$>D^sY=TRU4AY`${J4dd4c>2BKHw
z=ABU)x}L*n_~uva29_l8CW2KX84MN(EFimx;lUx!IWRli6D<KQO@vgzO@R+m<ta<c
zZ+o{mQ!|S<v#3G}e5*?@CXt7UZ66PaRI^PY5+kqFk7fda$#|`&$OT1{u;`HEa~a;t
zTXF}Pi5D!hQAY6i&)hJ)Ua$zx(9bMzkahJUWq0h4Y3`l)@w*O)b+amc#L13}fEqPd
z$Z|Rw!mcleli_r90srx>o0vrXZCYmef!vD9+1z9|wHJqcR2^aBJx^WMr2KWZbt}lb
zq?P%N|H~Ko(9ue&y;>G(ulfL;b|Uzb=#9nwXVyrFW;B#qbe-WG9NiSqn5GrYVyz_|
zM2PtY6^g@^mtIskykK43Vb^l*)iw7(fA5F~sL1V~Ejj#!)swS!>Y{9_H`rAhL!m=z
zt=Gn^?qP9>mNr3e&!8&(wxY~Zz0W*$(@!19)|a{(Wl+{q8@-IxWMWjBPdS<kaIPEU
z7T27qxFJ(#rhg(do$4ezViUF#&si<Flb&f#%}{NDpyoj@+L@Qq>ph9MSQ$s&i*;P4
z(GcWbQjuwe1ip9Rm&RM)?}4ttt%Ralpt*4?0^bNuBnGs40};zGSCBX+8T=^J?TPX?
zo0{ru+&SaU7xVL7LQ8`!d!N(Tj)*Z*aR1+{u_PX?_pKaMwVOB%rGJ<3$YVd-k}8oB
z@3Tz?BR7OlhPI5c+sT@9>o~S(H-`QjBNO2V)WXU#SYdX5zsTbSk@{ps9pb{jaWPc`
zrfUkrKE@?Q@u_}|VwhCBmT}dtth|U@N@z)UQ??+sup)h;M>R8TN1XjEIALfHwdR}Z
z8#@Yc<^$GE-%FRA?;n1(bz9Udj4AxEQLUNTiH${l8tzr8Ie$uCc>pEskY<&uejC6%
zQ^+`1{--}bivX{cPyY=R*CM{VPK19Fy0xWDzNiMsH_Hvfbsz#c@}rQx(POPRio{TA
zjL%S#C_!sMMHoi#DER8(JWno}rjpznTz;--?9C-l_*JfE;2#S!n%brkr_-o~v*hLB
zr(-eJIABQ)?{snjtJ2@VheJ04A2+v$S>WsipNmUsrt_WEkhdpsy|s30m1u?RgvtBH
zc*mv7LlQH<0VQ94h)3?W-v%2*Sq*+Qda%KIYg1T4W|$Gh(1j&f$Nrkmc&Tr(wl5Tn
zjc{`ni=U@vtc46^O3M|bs6$XT7^;ph+udkc^m)oY!8H8yo8iW^k=od#Osyj@Ge}Mm
zP30A0`s)&V1D2$ovV6sSU{RY;{C3VSjc8H@-)J;ZG#qK{7&B8`<L{ZUj{3#aV3}|8
zOJ?F|mqe=t9M3|?_6_ZezlXyR=|<nvv0bBH#pvnDrtH*~#rKy{2rXH~vF(A_Dv*VR
z<A>ZZ7R6oA5l^2{Ajky;*6}~xN<+Fda`VV-@cp>WlFwl6l4hT@RW?;r^FC^N>-H?l
zI5w(dl%0{hRfyEr?5war1#&9m*cBR{jxqjNXLaUymDbq_c@0Nc7{NM2&P9AB(FmpM
zjA2{yxA*Vf@4()E<+9Q7OvS*a9c!g65CLr++3GkG=l~i=^ck}>y#wJi*#yLKZ<EMY
zTxc#)8IrmSnp)j!b^MDy!UiW`Sa2919bLEVTWzty4~}JZz3I}BJR1{2<vCEtho;I0
zC+UjRCc)?^7`fuM#ks0Ns^0pt_JVRw+10BW*#ph25c?yn?@U)BM6>L^UR%M<A~t_R
z*m+4Q%%j-~VeHGbOpKP;vjTPJs(1!U`lC&S2oAqqCI}!8J6Bn(@dMDIT^V~WYs;LP
zX3-!;vp=%&U!m~xN2c)A8ve2?l#?kzv?r=&h^D`_dJf;srk~OMwXfFXK8@dL-+~qa
zHyqn^3V{esNI*}Bn2$mFWFIMru$TzfuqWrZ+GLLLRZHtCFKkVS(5jIc2lir-`AVWg
zxEur>=H&C(%j-bP;R*M}A7P+}_ywy#BJIPe0>HEy4t@jmyf&D`Mh(xKPx6KF3+^B1
zr{MUot<84L2fMVm{c^r)iuTIVOLmDbFV(`DSl&g#Ok|^I0Ur#rnLuZqub%*N!FKI$
ze!_gaFK>cwHqM{ENfDY?i|{^t@cWK!JF`cxE^38bJivgVys#wr1<#3a_smb+AOY4%
zCZQ>2<PkMQ_2?sh=M-H=$NW8jEfqwZiPZv^B$rJvzN{p@h4R))`)F1v)2{*$Y5;$x
z2!-R+S*-EU4y=<Hh~$DblhFaRF*k;{*yJ=KzsgO59<4^VlzxJIhF)(Fy+N*n6L56(
zm=uC5zmXT8L4s&0{>~6vpW5`s#E}IbSMVD*EkoA^V&|^FbgAsMM~&c2ep7U{yejvN
znNB>iiB>qr=3l5>y-btpMX9q+jRD?uRWwc?E3avY930PS3U(Er_1%_2pPkxkdZL0~
zg`D5vW75*uf(bIWU7?yfY9*;UzskEW`~AL$j2}3B|BjE*w{NMS78<De7%Q8GXGyD0
z@;db64PxMbj5xj?B;t^twy7y3ym`s?c9Et^zZE7=7?)+=U@>k$539Wz+rk1sQzO8`
zCNH7ZAZE{{<P{u2#|$gJjRox{CT&!02qCI?ah#?%Rr-N;Yo%2;Su<J>=E7Tn(|Z+@
zAR49?-EKL+l`a$afKNV#WsP-66md5aP}!t*`nXBYFA-^Hjt(aw#ZL@A%w!%3yPTCJ
zG&wbbcGKQxT9k)TUW%Lz6CWx&J@o(V{$yJl1&`8TO2oV+^a&N^9yAE!%tW<4>I3>S
z*}7RIwAWwk!l-MvEI^==W^%O1vi(ZIJ^-{r?_W2qS=>v{$=UvmmW&KXn0zyHTAZFD
z-@vG#q`|)*bu3y`OPSQgxH6=8Z=5yN`0<m+GQ%EJ>?>`eEvYLm(^IrN*1yjkMRjNq
zL&`$<!$_B=X4Gp{`uP&%N9mW9I*ZT&BuRoKkS*eXa>1$yv#gylf|j$`VU=&%5MQ<)
zL#Hr&N0At_*x-%<dBv&_|6IWPXc3yCl#-Y)uJ?;#2w!dg2#bMT;2xB>4qspW&TD+k
zK}&xC5qcN4oUi*49g^xARjDs8!x!>Ppip3DMvyfR){c^=Pz~hFF*$8KtFJUK-L>Wl
z5x1xzQL<_FnzY(eqWVi35Xd1Jm>2L!F+r${C~-MRdo^r9EvT(X@A`6VC8ZHRYkIQ_
ztUSMxKh7c5?Ap!I`?%(ihzeBt^Dz;tgAP|^75yDWDOa5o;BTp|)<k`+MRm_-LDZSd
zQ2RtA0gpDyX4VN!#78zwLo#M#W6O%G*@{Kd*URbM`&+s7OwM860a5ilpUg7|oDs^$
zz==tmYMpAWC=y-3=zJQFBWJUNTzO=(pLQ09umvhfQ8Si_pdzr94uj_BS@A+?opGGH
z<%?f#62A^+s@BI5b$9=1l^*u_r(gcOl9NM2$oDVz!4(gx`#6BDLq>-RY8r*;qknxO
zm!{ZsIhr*7V}1FpxiU3MszwA_b)#MHiT^Yh89Cz9!eiWzcm^20e_<_4fDdDnuSxNE
z;(r#@z8Qd5e^{_$8I<YU&Ag^|jT$7TjfI@~;+CpB5@VCSVqs0u74ZYqgbH4|Ycios
zk_An&<=FnNr^PM>>j}-QRVs5Zr+rSxp37+!Sj%$aTP$9{Vhjs!EFtah+~PG2WP_R%
z{#7RrwKJLT*dHEUeahMP(gq`zbB#23$QL;l-j8M{)fUgoR~+pYH6aLtMdS2Uka<P`
zR)mn5>JJcYCvw!L0KnHE@A%R5(8*l0nqoi_N0NF)&cPH#<&Gb*iBV^zy$af*pscFE
zN0z18_huQwhfD44wJZvoCI@Pj`}ZEzq6l*$p;X!V6|UB#y0C{v2s>GG;i5nN(DaEm
znE5{A>vX0;WSDGZIGJAM4iCc@6*(a?g%Ul~6_EVr`pi_iiUCAbA435W5fyzzv*|>B
ziB!?{KY(~OvqYy-K`(vU_RDe8Q<e=7(y+IAc?qKqZ9rMfuO8fDQM5nY8Nm||H`)2z
z&8XRMO<Y3O*Qmof1SU0A!=yLRZ84N#@pr9hTI8>{Pzp6m9p?%tAHUvJF3TwouuJiq
z%a$fo=zO~>u#&3ZEjkT}B6?^?laiEV(-Mw1lxZ~+6kzx9&1W;fMj=v7{)ATFqL{WG
z9>Sd!Y>Sz9XXQJ64Ec+ic<4K4o0Qi+?}SGhz3@$)@eP2Uu-pg9Q(#M7Iq^+8g%##U
zGOE=a`U~XL<`{ap%VXnkcyu4NN@rQ`=Jm%uMWM{`&BMEC={3ck9odSKX&|CIXJR{w
zCdbF5S(|ccFaeJLhnW)=?GxqxuIw=#Psb&01<;UtpIK91xur+iVo-wARy4=YtBh)#
z5p^Avadtf}tKq0d@m{1WASn&9`34(^l{BCFHF{tmHgri7BP^6m=fXjr=&y|)G}iHS
zh=L@kb%ccwHj`;qOyq-?P|QTpu2K8wpOAnrDHVjc^yRfYd|Colpbv`HWT!C4dv8vh
zYw{VtQDWuxnr&x>t7vAIj2w%rgO4&%8BN8Kk`BK!s$vbRKGS3!X3D&iFk?Ebc_MkI
z#fB2Cw`^c;-YDKc0!B`wL20qsq!HKo_Pg$Dp;_U{PhRr(VndskI@FI3Phqq5zl|jH
zy$2>LLXP~xs}PPwO2yX!Ce@9v8g#x@4$g*8g4wEh&l3MK`ZCvUvFV2J5u5&%NhqD_
z$Fh|WIa036M3+x(i(?_S&Z9H^2AB+Ue!G=AaMC~c;X*%Z$zFXr;a`Y{2f-T<pf}(-
zRqCl<N((M#sW!m^A7)_(Orc+PYJaM`+|bYf8B7^kL~@qok>!O7M%XC$=SSt&K8_j~
zNUY{IF^ehSR27r9g&~rtXEGl<|C-%W6{4j+{@QwoKSusEHQR_qp2ZSL;wus_Xc;>=
z3_bHXyo(bQNl24ai>BvvVwZzA<J|hiXQIMXM(cv(MyRyrd!{jnC))bljAF&HCpaTU
zM4%W~jz7JP2~s(fUt?z+OLY-V$&hi5{_&fXhsH}MN&SR|d=)WSSBO$xcUmQO<fx2V
zBMLH5zdcHld8^-b&1_k|W%R8)iIYf%*R9Yn)6q?Jk1u-=igWKdLrd;OyIiO!gU2=@
z9Z}LUHiN&Iu8KApy8yl@W{5k(QmhhHmBy1N|M>bJj!kj!NA(?($Hxon?P+j9Vcn4>
zZ=L+PW5i^=K%)EWk9Xh^0q%zHM81gwE4LQHQ6)b-(nOfZrUqYIl(JhdZM40c>e0}%
zgpNsEc1jvBu$Ni+P`AmFYDzRl&&nHCL5=fbc}XU;t{c-X?|fb~Cd1R+rSqkf3g6Ua
z7dCb}X~pS!ZJBbRpIh7x#dm$pB}8sPA&XV{nFM%_wPVob!j!Stw?p$E4^|66@Hv91
z&OI}u&|zKAF13}Z(xBEt`n$tEi=~hwUg$0BkQSs<S7F$0-n}SQ0ixgn8<`qkBJrO{
za!!xZq2$yjdtFHBpJ^%o`@n|hi)baegh&!sJC#>9_3)dv)vQil6pcvZrq2SU#MU_3
z-$IPInp@c^X4?S~Go5cxCtnSeky~=U6P|k0`|sfNMKKV%C=zq~*h535i2RLm8|ci8
zB`GN{@#{vfsTCH93E6AP9Q9t1{~A**fYVCV9LAM}_C)k}vh%gX^$%XH%}ZDmG~Xa(
zH-Eq0v+K@RjsunvJHIfV_1A<YuC+ye)0<Fj5|e$k4`R1HZPw{Oar%hNW8BJM2dcg`
zvv{u%cv4cTh8wi6>$-LYsqUBzWBRh5CWwhyDGiv<RC)#caERx1o;*6Be9B7`j5dn4
zdO0wCGiFeZ*pVMRTGX(fL7@;#&geD5_}B9v2029WdxBY4VDI-`JKw{2<ailgU4RJH
zFRUs`jXd$aO{1imn?(cm^l$pA&=E3Z74sM~FY`yw-z^P}e?oxAL7dd~LWG=W+Vav&
zRt#d{LgYHwZL3ffE#0H=OVD{aPIAXG6*hOD6s}6UGX6#-t6d3bQrC!s&C~Ote+fDD
ztaPSYyGWB7bzYE(ui=k6hR18cvT{8T9(2kIO!34jObS82CfL#QhmYsK<!ra-8)6zO
zJT}i3FZK|Mo5kw+D4spM9eOF;fVFhHdH!-enoS|-bT>OA73$N{&8=2zeSy1nIXo3I
zf&)`NWUy_&-~3_Qe-)Z!ir+J$FrA1{$Xcr^ce>XoxZDRAW{z^oiT@PX>mVu5g~PTl
ze+zf+hnQ!Os5^*NAXk;Y^YHexSla18tFZ%>B9{d+j|XR~5;cWf8mmRGyy03gCAf$M
z5O}C}xit&<eNy3XwfcEG=B%|Lw}b4CvG2nRCR~l?2^d?l&!X=fj%m?ut*=T^{W|6L
zi&fV`MDeBVTu;z-64+(_UE9nqDjYjEed%j#73N?42K~zn*J{yzqIXu*CGNV(WJMT!
zEA<qkrEI0~zVPUI@ltq_{qzYtAFmRNxxIXx{4WN7-du4p=iPA+mPik6CE9<n#o8%D
zZq(Tk800QDFGZyL5sQT2TdLX0xlyosu&7ttz&IEsSt#8QLt(6%!pbt@B&Un@Vrx3u
zvSb#;ZrRcqtMZh)_MJhl9lHqfJ^%MKEHRD9N|fA1s&hTo@;3%`h0JS%k=VKeBocGK
zH1u@@)}$u<{`Dt?EW~*y2Fm>IcH#V>YHL*4%J3RLrD(2ip_BT0m;axSII*IULwqB6
zHDZ_tNYO)L!<&0j4dJN^)}R>!i7$*tYmSEwXN#|&Xa337PGZVu+&;PQtQlP$VT@%f
z#VgMw0==DBTBj@^yJNM#{|GP?<*vf{IH+?4%PjhYbDi&G3oou21l#XsBcn4_k{Wp8
zoQ;f4aIkYTWSWccQwH*;#V)(Hj4oY<id^}(SbqzZcA`p|j9O*61gxn?mDC=E4mZh?
za^7B^!>$8Hn(Q2w`l*wT38BWa)vU?P%fR2+CDHEzNblWz4jA?XeM++q-LeHteK6N5
z&5B%e;lbwF2)5od)fGjj%+q2U^BPq@%W^E0hIl7Q$6PgB%aZ$(W5%x*-??z#E_p~w
z<yO17(dKOPe5fvVa<v0tm{uxH*4x+zw3-BJyw~fo2>+$AGQ;}x7C7h1jb8m-H=iSq
zPXHUy$-J~}VNy6uyDqoxvFsKUXc7=0{=%%6bSi&GM-A!%EoL!milcr}kRzUIr<Tb}
zDzZ9qa>k4?r?KEK;;6)&4WhwCEb|!f*I6bx#rO=oVHE;%HRhQUGdW79iY&GYtoV;*
zHMQF7cVOl5SXBD;K_K+UaWzLwK+#^d^21xkv*x}kF@VC#LX5taVTme2STyK<4yJ`a
z)pQvsB|N1dA%MR;zrNx#iN3(3kLL?Hv^r$BuTItbl!*K;{<o};ObBJQaiNK{`jj+N
zCEIdmF9u&pG0di6NmKtEX-Lpld{`xtg0qDnON68p5n%(iCNJh<E`F;te#7cho2ITy
zdicDhyd}bgDbYycsMTIOIE)w^!+Fn*>t7mb2*VSJpk3VeHolKEi^V=)_v~ELiC!1%
zPNU2PFYz)uuhLEHUl_ZT20c?G6NMuiL3k%xUIa@Zo4<lh$K&{{N8hb}dsOhMM1Hc2
zLNw;n#3J4rQ6DZwr!?89Qb`wF$*|+b$S~RUxE23U6}F#Mu4<o$yB&nG*Xc7?1<RPs
zSAS5xfm?GNedtGQUY^_bWrM`{ndz2fktbR%3tU$6l$$iGVZ*Pa@a(if8~6rs#L4Ma
zNb{F*1!F!&|0<Id1M#*J0m|cB1vY3JJ55qU7A92Al3cvs^$Ut!x#B$|VGZBdc7N2D
zJqOgqtq3wY%df9cG}3j{o)?M?dx%o)=NVJD)zK94+MF4|I`XTL#vnGuO+vHZKfKzp
zw@7Cb#{6}rA1nAX^OfxUNMzJ7fk_tPJ%OB8TJlXHD%VsYA<wmPJpHZ*j{)SYn@f^b
z!(#MzJBfizm>6Bc=qX(?96s7@^9vXA3T*%T@#EItlS9kl?4JP_=eJgbYrjso_3`Aa
zuLHaHya``PeEq^LU2FbsUU|^efHyokQbU!UUgC2H1w_mGN*|0?hMQ$|o^y%*F1(c2
zbjai9*awmQnhJ*IcVu9l`h;Co(5xlu^%$4R{7n)8u&Iw<H<2#l>a}gvnhVP@hdWz&
zZYRdFUe)KbuzvTwZ-I&mbDA`FOj-grT4im5z>o$*(vATmE{scU1b)4$@h==-lZJOr
zgL-yy4%Mz5?X|m<S+fYt133#WkQ=O?gu@*-*y%28w8^nTX#t+New_QX_1Fe-{r8+r
zA{iLqdX<bAP99E(2*l8_e!r&;X$@ho8bFm}OIeg7?Oy-E;k$v%F?2afp(BO_e!gH;
z_BlMK2oV3GTc;EM1v;9wTN%zv<8So4m|R+7QK+_J{DdBqz*-Wb4OFK3xyb)0o$|Aj
zP>XUjQh#}{+CKaGYX*(7Jb%>iSXKDKMdG$_SYA>vnTE?<$V2f?A#tN0Z(KNW`Pd=-
zIgXnAEa6-Z0)Y;R%)K$vA&5h73v?HIQxn;%0S?oju^{Tm?>xIH+D;32qM=5N*fzVn
zsWm1xUL<^Y&QscWqjGBjbKfI~y$o=${AJ0(RxTSo9i_7G;enN$DP2$ZqIsNpST%IM
zB=UD)CTShAs6IYA<-RQdZ9B#ldHRVeP(^z}aU`XrP`iq4ENf##xC1yT`&uq8l4le%
z(7Ba5?HF~bxCs;OGN^ucK-v`5-me&uTbfvufAxU9Gw$2fsJ&Cgj)ApvEw?;w1<JD)
zh}Xujdpo}lY2zUNq@rqcV~1~C!cJM+B)XWq`>INB5uYR=FHfMM+Ge=|pXtk$x+osy
zsCgQV2VkaZs<ULOlw=Z5g|+-Hq<k5dIKzkN@MnqXnRQt&rJQ?EQE7?F-+Zk2=y@#k
z^Ogx_2k=*x*hX84lDFk#<FG~>8*6Wh+D+#_y<KcOQC7u%Mn2!obc9$r^@ayc)N$UK
zl}ECUs=HriYaB>$qs`26Ksm86Rk+t>*Po&WZTRmXUb|cmbnhfe7ioFk*c>e|fv2*@
z1RO;;D@Pat3<_PYA6tLiCa7WcN*P3SB6D^p+QPHpS9Dsbo+Xqf@?O}vKye{rPG_!G
z(+#DzafW3D2Qln9^rh^8_|b&n!C@WR8%q2ia-LNkcWm_jX_(+cMeQk|Yt4^5u1~1v
zjX{xiR9qRaIfiFoShg;e9<+ramG5Gy^AGchW@Yq=x64yAUvYe5Z2`r1qF0`;<hMjF
zk<}*YuAJTW6WHN<=D6A;4++{k0_c8b8bMtrVU`CO%_|Lo-9&njoY^ar`k@ArGl8nF
zsXhirOhcK_-%D>x_k$rgKVP-)9?GDIyGv~58nlOjAi_(Cmxe>~@a5<>{WC2bTVJb<
ztQPj@=5aq`u+`nud+8N~czVhSEWX>iW>i=yOt(I$h}JXyMA!mv?v{fGv73e>E-m7v
zF3$s$v`lbJLk07wd#1<ZBy>|!<!8B)_Y_~(Tn!B9`6iO>a4V>$diwH8zvze>B-!{e
zjx>uu)Ba<F*LqsJ@#<l8lXN*S-uy7g1^Y@vNAOIWUh}Jj$~Kg(nNP$Zd<b5utPuUR
zlMS|rHU0i%R7>sl?;{N=hlZ>Z6e$2WD=OiEy#r)rNi(<5?OSG8^qNBC1gT}80E^7L
z-E!tWe-)d_4n8vAU53$@6WwTGgswEB=k1q?fe%Hs)qlRNNnIYLE0Tc#WZ<f+!ob1>
znF^6LQ=pRS1<oZEk?a&fsb3gtZhah~d=t(An!u0se$;V(o!Qbs)%d<`k%j`XV#)@>
zg*gRTH~Oz){n2&HscGVzkx8_U=b4xmjjPLk9Q4>Bb(v23V5wEImBi4?>oz5<zV1Z`
z{kUK@f(d<8a@%^R_!H{Glg)T?;A3!9U0}`YNh9_*T_r2+vkuz^)L)_ub2Gf!TSape
zxBsNxtj2+34@}2w+VTg+)R79mtg&}N!39gnMdVYmhq^{U*-G$7_HfYKBsFtA=}xLT
zc{U0*oTh98imX;8w36Lqz3=rWBc*dcMbRhU{|#+)3~#}5YDoJj3@`I)?4(>ITVzr&
zwB*^7Z=bs#Jr-}zqbk2{r4d>{tLmGD5&ac6L1qyj_+XG*d0wo>rGbIFOb;KShtdkS
zqRLlREdBPw%U!a<8U7ady-8(Pt@UofYBk^&m5Yn_1mJX|*X{Xl&ZRj0rkim+S0Tuv
z{8m%GOXPG^2HU*11(ealB;aFAE?H`phA22TO3O26Uz>Ur^R=Frt)Rt4A)yDQbT{4v
zH>)>o>ej}1Vd$4&uyXodpc^ha-d})@vbe-DuY9UAQU|rXE#0C(W!bx)Pt2;EuX@0%
z5RIXg=EX(>HU?WWKlK*SfpS1d8_vXGI<SVvBYtqRDk$CUAXiz4=rgGAq=)9KvWS6Q
z7*W7@6VIi182*ISCoLdcX3+YlQl@%ll7e7)&)y%7irqiP1i9WS-47oYnMw2(vnX#0
zgPfz8(=sEEw^Q8q<)PL$cb|K(f-$sq+9Fa{XHnRZ3=6?7HA)TC7l#t7Djuy>Voc4)
z0zBWE$+ct1MOV7EFgwB#YtqNTdm;2*3mAhEqgjktt<61}dgOrD5VgJ8X`@9Vg{c$y
zrpV{b;<I+Ntx!*U;oHXm0(sLuwS`~~zXh)(m{OOsE~=3wQHOkWTHi9&ld!HEV}dXG
zOb!hMz`2e!iM4Se^u0GHyQU)Z16TEU3w;!(rFctkKFO2`>*9IEr`L)zXBWcwD+Och
z)9Wxf_?&hc(5Vk2CW^2pK<iS*FwKXsbWt6leWjYLKXuWc3svqIcgCKJ?{rMe(EhTb
zq?Zg;T@*ZLk6dAFY^>&CTYJb+iQKCK{?fN3qB2>S{w2X<EIVYQEracCVv1>Q-S}RQ
zZ0ik{rf0{h$!~lWqa<FpI5tPIfns}`wbk$WD_NhOy~;MI+E!t0Z7)+ah=8D?b!V(Y
zjRU43kEow}s@!)nkRe|$i|Eqs@YuBYBB&beiz_>$heY5_#!NYM(`1%iH4kHeMKuox
zM*ZIQd?V_M#t~Hp1~MU=Z;XmJ6*g(AhFa0t-)WnMG-b+4%}k>l5tJ}<>JKV8#RcjX
z`hlWh*LRwvnK&Fy+{DAp;2G*ayGfX0G7hZSk6tgD$^L+>75(awjo|@j_f<#Mb+KjM
zGR(I-`5>52_^$|6tZV-{at?h+{o+_2r(Mh3EOjlYTrbxcwBuP)qQ|r_cEq%zgQHaS
z54^9a=c<lj=e-cpu`)gaCumuM)*Az<mbZKrL-a1{5TBMqCOfRk8Z8<vHgycAt3P+V
zPI%GMjH#FH(yU^lYn(d;_L8!HtORN~nX6W%l%N=>)DYFi)+_-p)#vUmv)vk6K^8GT
zeWbff+HKlJ;L{mD6{0M_P7*`CH}^+qlkic`IP(7e^Se!9br=J^Kz$~PTOHNypMs7A
zCXJ^yiDq`Nd(Y~r?w!``xUH#7E3mU6-GGhkL=o}uAO=+n3?kXv-j<gdB;60`bHmzB
zxcsbVUosa1zUUc{G*@*Jmo~NRM7h#SC2-G;7Tm#Dwld$e`MJVbPk;xx_)b5+m%!9n
z{x(vkX<eSFG&`j&4&UU3JspzlL6zAPL%WSh!=>%x;*&H4^(N|2|M(`7){|S}4P|oo
zx3W6ugm*=?b}9f8Z$6D9gv%(za${7h1DDYf=p*Ka1%39s+Cgzm0?s`bfjd)XRn0n!
zN5j7ET^k)6&eRcU_~Byx5;VFj@Ru1ja*l6ZY!8%NdRZcWKz%^JT_c?LOa6o{vo=LS
zMLFmt4iSp$SnJwQ((>Bzn8OUx4z3-q&MWf8YQ0?dS&jpnE(Chot5M^M6S#%Wo1G#C
z^wnL9@#bH=qORzo>3h(g_kj=Jr1)v!R{ErqfTSXqlCSI-P@c}X1%kliG>#fv|9~92
z3&c7WIP#BC$u;twS2(#>_DsNAe-m;}ePm6krjGm=k%2I95mA|WTPEZ~!F=VXVT1M&
ztT1v8Vqf&d)~`P?xs$8*oc#Y51!YQe(%@r~8ajz=q;N(XENK_7owd0uhM+h!UTGJ%
z3Sx}2^s}(ra82<=qnof>ci@Pr0j!sG@&~A3XQM{Rx33HeQodwu>S9yOv8t$+rM?sA
zYanMpFp5xItA$O%FYw``O`oCc-Fw{D{neu>{4lj6?s3c@W9{}_9Wk{$KE^DF`I63Y
zrDO()D=M$V)6HD(W4Qo?8@N8h-n0A8_e$1R@~fACGj%NQ?ke%LU7>d=SAi7)M~Umy
zsFo;H90MAlyI3HtxlzS$7__|EI?IG)flqGSip+V;qW`nfNLF@-Sv<<C>gqvS%D+(;
zYmaKp>FbE9ikE}du7G7pvcrjLdW<AmOWW%dpO(TmI^Tz^B+-txs^N(^d1yTcC|~~Y
zzZJgc^k%n{>azjKVIaH)K4d_O6SuV8tNK1N2<WE)eZKF9t{{<S;3!41pI;BK^;^gL
z7;BrO)7#;1a|}KG>CR~@9z!0cpM|-(U%e@g+3jYYG4dx$OM(g3h9!oydF^Z8F)d~)
zw^KmSKTP^Fs&iHI>}h(gIgBnx+hgn3H2<30P|91Oz2N*v+J5&;#G@>oZ~!^U_{b)r
zF;Yt(-Sd-%$<0poBu%@fRJ1K+jZ?E?f+PBM)61ZV2)k`B1svwqb^e!z_OEn*Gnqi|
zuU4CSo;12qX6+x2^1&bHx3ZH-oRG3=Gl-UH9;&dd>W0kl1>`uCvMvaYCOFeYT4^DC
z{q}}yLc3|>Sc^>iWQOyS2f4`N2Sbd1dp;Thq<3)e%H;24u%ktBM3NTeYWaBc<E_0L
z6}yxv9KdxD#x8%#4B`v;A?+aWapNX>DM_=1gF_tdLsG<5Cirpfd`;S9@P~WbQeO>0
zxUvNp-k~9U!j!r(1ST)>kcfCI$QlK*GOtXaPmv?0o%;|3Px9KTewIoKD2y-d`Hd8%
z{Tv?GXDVX%&xfasKX<8WFk9u@vP|=Y_IHk4-bhL?&9c}{o7gq$)i3!9h786QO}v(=
zwk(u2>YT{F;RKjsfQ@y*qN6Wf352vv5<vpzb^FfcrY88ujXGu4p79f2O_mXqiGhA~
zS7Lz6U-{lDobf8WZwB8PG3w@E5I>>Pwk!L)n35ul5DVWF($<%TEyrk7nW!Rb!SA3N
zTSjQKWMP~vLbiayPMm4k;O?69kIz`oax^vHpU%}2mB1U3Ne)rAM4~ToJ*~Fj1IA>y
zeie<Pq3P>&{b|wHx9sE8r^qjn?C>K|_OD^pTIIQymtU#`W{2Th@Fa+g77n`);F~PB
z?f2_%Rex(}a^A0J4w<5tqR24d+#jO&C}aKPDZMw#R~k4H@Tc9r-neTD&-$aWL(E?S
zBb9kQnA1K>p=J>QrqjtlgOJApGlf>#l&N#!Ar7hTKzswaE=Dd)3$L{&CwsW!6mjNb
z@-+=3c93e;Jq2v92h-LpcFP9KTJiE>rr;E(>6NL{GYFmkF;f9$`G3Ac<(9(}Q1N0c
zHGZo`#bBD5L?Sd~bS3@@dGs+U`Lpkb#M3Dwg#<O!g{3+s$kqH#@wX_^%;Mn1C(7OR
z#;|E6VXpVh7bqPbEG^@o=Uw&xWp{t6OOua#hI!l9mdLe%?(2N}lnh+MwoKSOqZfWQ
z7gr1<jkw}h6Kex<2Y*-TjOR(6ZvD;jrc=(+)2cyZri^H{PXu;D4o(&M+WxY%deSp_
zdnbte6nhVZ@+SQw;}75X_sAgSGkOL&?qf78%2Rv1EskF(X$3TLri{%9!@B}v)5<+S
zfZ^kz^A}bwiYVAx%a#?G70;g|PdHNMrBM8MW)uP*qBa|H^C~2be}Jt%jQUa+_R*HT
zX5exbVrO%)ha3&sG2hIYHB@e<^!zFD#_erUY@1uH_yLkb?#`uJ`{y>@%;xzWcZi6*
zre!+5@7MS=u=rjGe_o&uThd&7_pP!71;1JJqK#fC-*RA;g~?vWAm>*7i_VPfd#=Xb
zS4=ZCDXNNY(%WH=Mg2yu4F1q>Uy{23ZH%}dz`nwmyBJup^R&&{?bky!R5#l{0J`4Q
zg8S&dnx!5t>4lFnllN2SPCQl@^tS?g=L{KYD>fwxZeA*<N2=;~0L!0!HT8~T1FraE
z{Ra_?Mt@C`4ypa?3N{M{MwWoj**lByX_Qo|zIp2ppndBjKE8tax*wF3`luWGuhy{*
z#Jx7xLsyz(F)(}Y-M06=!LxjpB5nS*elVRltiOCNu|Iz{C?nI-+rroAr~S$njeVSE
z(mB|9KFX+GcEy@=3y-L(m-YyDOtuc0jNcVq56oTDE)i8Zf5)q9?0MgeHhB3AAE9}7
z4$K5a1Dwbge76xh4in#g&ACG;8kgR4W8MX5sYDg(n`gcx!=HA{(cORP?51;Op%x;=
zaZ#+-?u6;J#z*XA&dc&$SiI%>9F-scO8yYU)dx_u5`FRpN2*gZkqZu%?v*BPkbTdf
zHqNL`?aqZZa{Lbf&OkB0jBNY3<pwvAsM3BtyOLHLWtp~5`bhZ6*c<*LhO}P)(MjR+
z@qDQoEkSTI<90iEf4`~Hp01U)Inc5kXzh8P{h^wc^P-fj=W%j7G@X|gVdmoT#We;q
z7pq+9<@4j?lU9wnE-oL`Ubx>*N-v!6CvV^GTCaDUT6WErfTIs5#6<Iz%z<6OS>(=!
znsbh&Zewc?dVARoc!82a5;<myY31&!$aXcF!YFI1fYu)VMD92+9U%!-0vkpQBqAGM
zl-RU9dA0t#JzXwe6Sqx?OSXkC3IDbRfxt>*+2rZqWAe<cXKTz?_fc#^A!2`!{QvE|
zYgx^fto9;@gfo-P9&*azG<Rj!LiEnC&BY&FOD%gV2gj!TVjQh;J1TxmM$^VG?L}V9
z^~n=;s?{pi!MbT>W<!a>)h&sIC3a0IXnmuF)?iz_g572|)!G!mLljra4)W64*jJga
zES^>ZWU#BYZ72lK4Nr@ZZHb?s`wqkrDy(|DQ%cL=7__r+Mqz$Fnd?ER(C;$$8!^+r
zDZfamR1d642wF;`A9vY%w43biE!kXqc4o=m36LHpUO$71n(|^CX}(hF$H94O^P*J+
zon2X<tLN4EBko<;mtN<tB&Pr6p1L$aRo9X;7z+)~OWr4a9sP|t)4&fEnk%b&3HWjr
z5Gs6<-GUU-`CUVBoB5_kZ;KdDM`L{*!P!X>O?9#_IS!Pfy~m)rW3fNXjFwijk~SdP
zlU#EqVt<%l<fy>?wRne2^5nUQ<q%CXm0X*;5;77cdGiGXZ&;Okt#5H)$cea=L$Pk^
zXZ%ZEe<dX~E5*#-6}7~@8Q<Pc&f2&hgXiy$G>`Kf7U|+#*Px~~hL!Z3X^mDpXKA8c
zrT<H{FN)fn>E}}lxa<}_K0jzB=trS7(b~#cs7K-3FYmm6e-}M786}n#jl?sWi(Hom
zL8S!u=p08H)kUR*V@tNk7IIGQRi=Nw85g1TgtevberPVRau_(`tIgHU03H0X@^D#Y
zAzvh!jA<7E6UtC~t^u+oJ+>0XSgwjjyLX-Tf{-k!gWHb5c|0FH9uMY{od^Z_v=kGC
z^m*~T2IPhNTI1=j&rhHE#(h(tP+To^;=`*p+~(_&MdIgsclHWnO?dj`$GCW2^8dGL
zoVCj<tF)JAPAP`<+{^0@Yf{=`td5sX^mr8&{dQ1#qtftcR0~+L!>(1Di1QFZp`^K*
zhjc4r4QXt`h|t-rr6z?s?HAI5p>A&((k>w@<c-lQ@a$>Ai?hQQXZ-8;*xD7(IgLyF
zB*yY2Y~t~x9q{e@4VzU7U<?g2xwL|{$O;6j9B!1_x!>P8k2m`9U1o*#li?Ga4|O-O
z-&eeLZ>&Ok8GS(>rB+%yuz-nk+-~&ap!R1L9ZCWvNhQK7RrK~+v0kZ%=+Q0bfZO(h
z2q6XGNlmSJX2?4>lYwJdtZ0|VHi0D1mo3?!PoG>#G43sG2`ZWeZeS^A555F%$(|>P
z9dm1v)G2x<jJchI(|;;~3+8lUD(O_y&mT*{d>2)j2lAOJ8}IX@;Gh2<q?$RQvycP8
zrk_UAPxG{C@a*|U5KLW`*q3`fuumoIx+cA;Yn*>sPWbDW?5OokOtThlN9Ft5o%?yD
zyQAk1CLZ~I)7Kb8Gx}Q8HHjHw67^0ijiaBKdBRHKJWqFdlh=y-^QB~UM|+=R(T>XT
z?auM`Mm-DGjJNkY_xoFxW0sF{YXKCr#+W=VK~J>?y-TfJT1|@$;fPYRTjKM3hOn}g
z&(^&a){^<$S};A5V&>wPr*W8vT2^X*Rr;1~o+W0^bZn~fulci=ti-oc^orcW+*Oro
z$gDwgDLKnX-=gO=a_gKY6co@3s|1h7lgIPodJc-`7CQID<Y%s-BqjbS{;^bNLkgR=
zw~euC$<D-C4RBePIHpBI>&b7=CqJ$yVVzQwa8N5Pt(Gl=t5&kg0$YWFbGFWVxnZZ%
zr=Cl%DXwY_`RN4nVO&pn@f>GMucwYBmC*Okum_Fy1()#(@Ows-3t~#V1VidLH}sKp
zZ+2${SuA8hw59cAxA1kfabL_e8Dq#3EW0l)Z_kB3ZwDNW@gV(qf;PL@b%t~*P5ZT@
zLTM-G`Oe$@JGc8c4V+sOn`V46*Awj<-`KjMQ;qV1(kRAJ3&(Nte*30i{`ouS`I$mC
zF}fn4w=TPr0<pB^tITcM*l4F9zt$?0EzZj`HifNMOX@Dt1cXI)kOd1tm&@a&s#dNL
zODvj(!#;y?U5r>Px;lm>I{Ko>5+!T;l{24h-ExhQU8A#XrXEFez*u74h38)ABD$|<
zw}m10kSkc1WLA~+MH9A!E;mfKVOSv+D(F0^tHd_&5K$*0JKNpm=I5mu=GUI)>ACnT
zO$7s;L77IX&;JXxYkyr*XqFu9FpjRF@cVgk)Z7zX<RFTTgaU-wT|QYBl%u8are%`R
zs=U@(wr>~LyqIJ1@$t!zAD=uvA97w@$+vd&kRs^SxW9G2{rz{|fB98wHglK~OEj!v
zmF3C2SrUoh$)xen&@!w1T{U@D+5pwm*1Md91xdF(08*maU(W-z)X61ZzBmsJSuOtH
z7ayvmi_0#7#P+p+I2EsJ{F}Bj++}T%D!QIdt+g4)X|zlbmaVcD$PdOCJRgFH3}KI}
ziP0rzTHl4M0{%{DQv<j1hnYgHm1SAy6b;QAomh*R#n^LPJm=!9f~RI1QoXzE8=muG
zMu;lh8u!~N`}2l`LR)QwhL5FX)z##gRd09NxfrX_kDj`ggewq3Xq<&}o|Iy|y&v4}
zvh^sfrg}Hn-2PffwQpJi?3(Tm^i3_9Q9>Z$jo00b4DhkA$Uv39|H5>QfAf8veNjms
z`;~*p4z8fr%oNZ{;MOpj&V~oQ9SN=GCbiODMTiBIdT_gc<955}PH8xfH+uh_Iv!kc
z0c6#!AYx`|Vd7ZPoa1(U<92+*tni$R`|XXR-)QB7!4yK1FpY$Hat$6s=fkSed*$d=
z48DZXh{6gbKK6_`X}`4!UaFe%=j<TahNjPYcM8F-ww35th)U|yxt8WalS*5SzO-zr
zW_3No8Lppqr8UP~Bk8D07#nYn6$z_Ki227|H7fSMj6G`&ErjyQa-(JE^9SbdYnJuu
zGO_GjB?>McgwH*97mU}9Rm76?gr_`G?3PDUovbr0t62o~lF;QGe}kDqZALE#t&|tW
z-gkR=G;m{-3BHnrlvR_<^>eMs^ZBH=#_fK4nd`}<Z07#n2am@mKYst<`Ottgk~+D4
z{LQrCsh^GWesbP!x!k7_RIW9oXa#CZUjA6jf%dv)1>CX%tX-Zj+dkQs7$=GW86<M>
zqj`azM~XUS+7+sTHafW5Mumc`rrGf>q{?(HyV3SC*?5fW(hOyJsy>^P+M(@ldq;JY
zy`<w02h^p_@W>?~OIG05xOhIaUa%CwV6AFN5gycd!M{psQ-*Eh8Wbe-s>RYeBVnp^
zvn@><xF+zJgOBHhjmc-N{Af?HqO3CcvhY58<-ApHXXpN=pj|pp79}aQFU-DbNLO5}
zrK1*ZZwE7^nr$szXelq(-Sr;0i;;C6XU>A^YkL@w%ot(+dWQO(YYy4<i|o=Y{f2nW
zi`Zv-=>kSUqD73&tKX18wN%|EJ|CgQQWUAB$Gi~La&Otv0I{_$kzGCT-grJHpf!-A
zP2tzfm(a`8Z%CNCVJ{fgIQq$P-YL)DRWB#Atfm^fSYb8j)i}>LZs#{yfrQ>~v&!wb
z@pilOd+khQiSe>q$k~}LmI)<DtM0Eqr$0(0g*bMwm4{F(M%slb0Gm36!Y+1{4)p9*
zZ3~fC1zoxdt=)wgg>gAv66Q^^hmePr-Im_lcE?>I+x@w=*Oz9DHHidyX};b`s=y<Q
zzR*a9$;^6P3@{9JJ;~;Jc^yR%^ULo*NQh$07g<O@OSC<=(G=njx=WEraxh&>g%;Eo
zdr7eN-(<<o(K_eRX|-hASe5--g{;WeOvUodA!QDei1hc#<lt*@p|ze?bU|kbX1c~w
z+hbabSxU`HYGOF8I7ctELjiNli6z#$oAZ|m)b<?8p^gs{hVpcs>>Nkob~{t$3T5hn
z7Am#qqb+s&od?qvWk}BuuF6j<BzaxPuu7?FwH2uCg+~^?W)FG7Uy#;x;i4YFUh9X+
zA4N>Aq^9|xh4x8r+lt;BB`;c*r|bpaW60UFmX_YMVhmXhsV#@e3ss+EI?qeh($PE}
zQA2E>oPij;F4)KQ;K$?1=i|cW<VUIWVvOn9cinUu-j0)cbn30in&aL%Zpo-BmUiaq
z=p-iNg>kpmux4~CP!IM!t@TSNxDq$kv&%J*wS9@wk{uQ%?<(oLcEYwK2&o9sCndJe
ztK;8iY<CHvG9VmM-uiC^fbg0{Yz6jACs7b3a>Rphw6?TV4MvO3CZ)1MRdPS1Q>o9H
z<PB~4yN33R=x2ZO91ot?MGK?!LwAp~iW%8q%$0hIZ4)B>PDIfAAxNsUOc+6HJ%7K`
zOXWdcYXWj8ZK18nnp@;->eV6bF={@ymGhOP8*g`roXoJTRA3M11*^FiwYB9BpM!x_
zXyokB6W~moKYfx}?$oHX9QZF^uiYrym{MCB;{_GtA<rcO(m%<}vKWX=a?tFLcIJ96
zGXDF5fd#v3Zls?@#&*tTW{D;%wi{wvnWYonhHM4Dm6A>6?y3P-f0f;qoPj+%N#Jj@
zWaqePDSB&7=4=IVq)Xv3CgX9jmJTe144454D`_oWhd>oEdcYn|?J{k;uEFzh@%S9f
zXR2&ddS6Y#mwRs<hr)J0ZrtC$Q(6_SIOV-w>t;Nq(seB6j9uitV6ajIsnoRPP=_zi
zl%4~;2z!Z)_Mx8Hy=)fC8o~<Y$|w^+RS)I$Yj0&LmT*y;dhMa31XM}}QS_l&%>%o(
zUP6gBmY7aoJmyUV$d)t{akvg=crHRu@t>JU9a6jU8ph4Mnzb6G<S;br7|I`#ZN!TS
z;jY&AnoIopxwMJPjOTRv**V@0j@zB%?WCUv<u0uGJQ}TQ7L*`<Zs$@^&=3!zicnxR
z{b(xe?Pc5hepkO_<ymV(UQ_8fM?M$g1sOy_1fDRFRKwZbmI_{9l_2(Iz_vw6uYgxG
znIwti)lFrk7?D0V7SyQP&w%*mbGmdbOYXC^e2JWDZ7+DQ)J821))GUNwekuH)5cUm
zB^vg@Vc$jEh{+gFuIr(doMVbox=947x)vmnT`^+YM2Tb}RqRqFzq_Xus)YCbCTDSO
z&R$HoS9lWfN~~fACj^w4;?+g7^XX<O>G`^x8^JE{4J#R<dpji*C@Do$)5uDDGlxP$
z;fP=@oo`o_r;a*u(N*LYz$sX_T2}XAY%d}9eIS#V@o$(`LO=A!LS#&xr%<vB1RDna
z>hw30C?aPH>BppRr=sJPK*)VHIn*zkz_^n`{_u)5vIZ&vW7|voJFmYgDfHgxw?oiT
zml;?;8g7*_7VBF02<_s{0os}Ytza%MnU>ZTdk_~$Jk}#G?h*}iJjAbmj_b9!qE+Kw
z55C{ucsuU&-YLC9(-yMFeDHW&V6NS)#bhiu7VgOh$$()6s&t~c)yUE&L+k5$E!1+*
zV~V$K@=wwZ2suiw^j!*Hsl$+`zm@iCVS8&YIAYryY+=+j6ntJTr4eaat-~>VvYx_t
zt*Tly=S!LChS7+~LAT4Ha<ihuQ24UE*p+o6Td7y5o67-V^w^HBPV0@boZ7aPLoO96
z^<^n+CfI5(Q=cvj_Q$wb<{Un_*~#~N<+fGD&7cXYsdev0www?AF)38pJ4DiL>(X=R
zBu#H6D@s4#i!o|)jjg@EFy~<J<CXouwlsJJp8b5vmsm>womF32jPFaj{cW&#?ZzMj
ziy^^WOR?V;Ox2JmQbHbG)P)!fD~5K5=i-;}MLz~d3-C!XXPR6yF{9Rw5nScTbv+3d
z*LV=|q&_O=$3ZV9=H4~<`23BJ-~P-<Vu2Vv>V&p~XtTo1O7yud`g#gpYmF6y>C(Db
z2P;*Mqe}uWC(l7*^wSruC0u9Z_lw*IO$IC=9S#(%MJzEaI{{%adVKmGD{b|Pe+`ap
z&3x7(H8hf>5)jtp4jTGiS`4TzG4iZm>BW|Kk)BE6HJIJ-)qaortdLA0vg>y|jq5ar
zTz@Q}Rt}vl&G92%Ma;J0*4I<<;wGqEvn-v_Qc}Qave*#n5U+zD?xM*T*{hV?b%86U
z?v<|p-s`U$G`G&#4$=FL&X(U9!rs<g_?TSJr{v*Fdy?xe2l4@Jdfs*}oO5xsLg{E*
zRS_+5!!;33Eyk@i{_g#q+pVRu>Oq9@`T4|6%Ttz5ZGXyY*j$QJ3j>$Hq*%$&u1n@F
zXIF8sBWDe!YRO4&BU7MC3=*&SGBl^nh3eaaV-G($lA0#^+&26BTu&XQ)JweJhGIm4
zu@+a_N66B<aFj#A`RG_}+J{>bM;}A%2vRWRX%v^wviHK8U;7VaE1bz7zA6n>wNkBG
z32ogfPOFWcwh$r*d{u|)29*GPEg>k)C?>(wb2wg&K2P3PFqXWWs<x%PW-q;UB_T_W
zJQa(oWbR*B>Tb{V@8v~X-U~}CVIy99nk_qmRDfF%Upo`uE>FMq3CrEHCd<X~Ts|ab
zoG|WuFOcwx0%_n`<>dv>i(JaC1cmc^2AFMHMhPi_r*Q7M#D+OXFuOri3RBO^llDj|
ztuv{6K#3wH{un&3AN>BCO!LMVe0==i<KuVcc)V;!+|yG?c!%O<CdKf{wH~qo_rh~N
znRC2YOB6C;i~hK3c`2QhfWHFSrdh5(D7&}XS@U~6J@ce><}=p^EiaQT2&`o%vn+NZ
zVZW^EE7A{9<+LDtK0CCwJIeJ^<9m2^{rbKY6JtyLP5zAg=SlbFnm%o>&vADK@w#_8
zxnkvDpzPTTv8~SXTnT>GS+?AL9qFX7_dg(vEMrZWvxyQAQcgEP%P-++Ccj;@YK86f
zm&N(N=uhRV*9^uf^QTCN6dsIgaD86*LzRf9q1;^QC%)vzABoky@Xe6n9J%l!Qn9z!
zrdni@I6CiV=ll0NXP2|;{q2o<$XzpRah?Zu_N>}PpRGq1e^)nHa{{XTS(dY|q=+p&
z4GYpHAlug{;=NO}PBCpVtW@X+wqUw!kqeQF$SYPwF??a!eBp|<R>l}Fc8U&H+TI#G
z=H&AsEyZzmZudsNb&mT<>j$3I$kCO&v?5f`m=Hu3(-YPbw9r(RDZuBZvYJCvOWM+I
zW$Q?Lgdw-fnoHe<6ePSMQ+_6BUSdXV#_g=6+xRQ+s5KWP7G|t`R*{UEZhGc5*Ay!G
zG}(ycwlv#N<jQScHE3Nu&Mq+AU7DDLf!wo#M{^i_ubRoYmgkIwGxl(^<Y(p^uztY>
zS-mDkZ!efYmHi?pNm;c0XEVX#!Cs#7HTm+~6(JXZHJ7@D>ym(Tz|~dzR=d(V*|;7T
zEWxUU!qGHqAmklo5sb%^kLRG2lJ2n&*1Yf)^iqf_Gso6NW!9j$Q4+>;7IQwhNX7UP
z1lfD%I8ToA$(k$IZx;5d&4p~<+g|WfYO1Ft&djD<(1?+OPZdU%&(b}xC66vFtv+5(
z0XtAuWKAv@Ip(GNovLvSn_wC*=9py>yIDIk=W%-zqlHoMEC|UR8I`>!Tg8RF&ZxRc
zL5f>x9j>lOlB8gZDc#^pJ^1?c*^EEMfPL1o>R*nR+~Hh|SXn5uk}A!Y8v<FTWW%%1
zUm#nU;Hv=o_g;ThQrNhIl486d+Id}kem+Hcn(-RK7^(;g(5v95R!U;opaI=xa%icB
zhrfoS+PXEdSlrH&Z|`s1&XcnhZtr*Q-`=V3Z`fIhva|lc$6$Mn8<kq1m-2#lYAN{T
zd_FEM?pdk|+KTKjlyxOgGUimGYPD-STx~kksho_<vx3$Dz8y!>{VYBAH$0;%MSBdK
zX)^d?`<X<f^~2)hbMSmFtQ`zD?paCKCIG9~!r5j0-Me%T;fuLoj(9m(&TB|yQzZ;+
zhoyD2l9N$#n9U!mokDo!YzNVVk!sVG{tlUhN^7cea!%4pf!-(y_etMiP=%qlv(j!2
z>$<QD-wRFU&^i6(FZ8uFO4nq0E*@E$DVU?6P)lAc&*8m>zu_q;G(!P>Z?{`78+stU
zb>yB~YfI|mPSpB)UsPf#Zauq9_A{S)MvENmWd}^lsxk5FO&fh}*GnkLvtyKfcA>5+
zYx?I0@W68I<xwxJbk-%hrEErmUer-}Ntk58^?WcxNnEY~&#YRt<*cOMVE=Omx?YbQ
zgho<Sfl^byqFLSTesWzW*YlF4y)PotfgoFs+8Vu|2?b^?pEeXKAkr1pLzR72-Lbd9
z*-EH)ciQ^AN{D{y39*<Zjvjm!nEBr?VnJ2`uUb80m%VU*%@B(fazz7$O5~hT^~9bZ
z$%aPWL;y;5n{pm*38(2o&GiLC?HNK~X+55MXVBk|6#wmW=j7Ih*mqMNX#uhb3C#;O
zxuPFQXxKl*el{2jn7dN@cV2&0QW$gb92bj93++1f>0Hmr^}O)0u({F#K@Yu|gfM4s
z+P0M#dyVpHZEM*+EcTLVZwyyee9+Iztsk7d)0+f2dr;Vlc+YcA8A17cVZ${qbZh$l
zEp4+`b*`y(qH}2o<bUwBk68FLal=>6J?^s2u(I_Q!h0<pTelcbfm1V*y(w!SK9V+8
zYN4XVAr!}(76eUK`*@ATm}+Qe1`~=2#=&GyeaijSvKk!M;5k-0bM9fQv#J)**J|2l
z+_k>)4{X9rSYnJ7k5`ggiGA*!#=fLkXn9$~{G|;^;DA=`#l)!n2KUN&?@C~|N<GAO
z*=A=EX*;lh><p<ht20%R!-qbnl3TWNgL<spZDofP3tQ@RY`-nrQ-!^c!d^kaP>0fM
z&)}GtZt$w;#nu^&;S_7M99+k){(Q0dOJSLqHcd1_K_7cD;e5{?($<66kVJz}g1nxC
zF|?M`r-qNSPr1v6^fN{ZL__do%^eUPT9~v@#qBY2W~C~))hchKu1$j^b4GXMmZpu`
zC!mitVM#WPl$??2SqYAc{Z{gPwy(YAIdzUARXa<=Ye9ZXwSFx*m)PD&2~D9+HWvB)
z+bd{R+fWja2~o}A{yoE)FDnY2P@OSR(+xu`IiXl0W6YU}J}Kk+cCnk%RB|IS&AvOK
zGh?}(?~AsqPTICKCA*M0{bggXtgv2CTipdG4phE8qf*2dB}H|USQYP8u-aZYwOL8K
zyGb_Tn12JE!t%v)4!lGfoym}^WGiTq?v}g7cEJ|6G&1YG$ZDYOfmzjHbS<tcplC_^
z^Ks=4+l5c3y3<SHJUX|#Ov9`hk2$dGlluuCs)qJ*l1kbihZqyJcUAfsEF+K)5lfJx
zZB2nHjO8X}^;)@8nq&Zc3D;i3>4ZJDpL*LDbTb#w2W-`Fbim8LP@zLyYjWf*Fk5Ju
zh*N8Do+r2c>vjSeM4E!(@WC|(k4uJlt#)4iGt1SPM0OYL;C7tJsa{D~$bo-W$U4=s
zpk7PE<CU0xWyyFhy9;*(xPq&hZ$%`8qUycBRgPcY`Q`6_;rsVDZnp#Lm4&nS_%4Sr
z1TB>ERr&2zlXE&_T`y0UnpNkTC5MLz?d*FTQ$zjiY+s!kIcI6=OpQvV(My+F-@aH}
zD=o#%1bb>D8LzI&xAq_aa(^(Ygkl9=LX4OyI7RG$F9oV4Tp+8GO3fKaM8VeJ?49eH
zthpGEfl+8)bfzrT!5Kiy3;IgPWoaQ)<ahLKTaC0lyLc(JDd}6~IBvY%-#CsNR*dOS
zu1mwt<tn&*ZG$$=rSO{GQ8NQ-ejiIv$F}fHR9BY2(1kTtTHR!~ozK$UU+w|=nxSs6
zkVQ%EPWJ3{i|Ag#hDl2nJh?o}clEx5HA%`aSoWF$Z4Hv=P<Tzvc|<w3S4SQDT<$=h
zh3Vc+RrU^V;XFDED`Cr0z7ohK%lNV8t0UftwzgcRle<2?I)y|VmOfokXM-LPy};0)
z+g+G#oHWGM`r4iGAD9%_K{KmFfL4H3iK9!lzEZKA*lBG@(U?K705#J4J0YfP4(628
z=brG-sRiG@7Dp-Ejw4-EJB~AA@tnbWOgKa_8&j0l<$`m16F<FLrOpKmjw<~0<9hO#
zPi8D=q##Tim#&eR-c&;HC7S4%i#8pr+sMi?Af%Wia|Jm}-&nmT>;Sup5(cY9)fJmc
zxUFg7c<+VVt<qa6$L|OC`z>87t6I}qq%L;R=`iEC-9&#SH<67&xN($<S*Mk%B{%(~
z7?Sr#igY&9qR6getFs@^CnK#7#AoN#YFp8+g7Z8W=ZW7gVix*Qa^Pk3?>F9lz4OcW
zH@>~yIeH@uo~b4cH|FK6vZxh$$pLV-x+`ssSLnUgWTRMY9=R^*%WfiZ+wCw}k_N1t
zkO6h32Er-@F`cxI)L(x7GbU>dKx;(B_&KDiO^kMvTK+8?X^YuZ6jc{&Su)v*@stDJ
zlJ2{$r*ggucenMTNvk4ssu|$}pB*gIaC-gJ^G-F=dI34sQt16u5#x(-jSNo7%FVNA
zE?D<=yK|mzZ1+#`cq;BA>8zr9tE<s{TgC9PxUP!`r*~Oi?4T{D()J2!LX`^?-Jfl&
z=?ndwiF+4ox%VL{`b5%9MShTj@@Cj5&<&GUQfq~|B!sK0vP=nKrZqY_2l8Zb=}F~X
zw!&CL))$Yox36-tReh1AMesUr%hyY3juncx9A;|(?^8ZY741zoiqL=&jLUgEr|7Ba
zL)DJTany89XjI~5aZyYKrDt~^VHqQxx+-G|-s`<-#c358rRAMDH*N9XdHq#Mq4dpP
zw+wc!agPN@nrX2Wrb$hdDpewwz$5`D*fKxia9NF_EIM<jh3i2a$Dsk&anNexu`aH)
zcr52R;TiIYS?)Y@a8%7Cx;i=~tKFkI=b_5^<9TsiBSViAk#2D1>O2d180~DV6pidQ
zNRzDP+Fk5U=}ql_Z5l#nuy7|SaoY;Rzi>p$foq9ZXV8u!m&~_DETeT*4Usgza3vdG
zlEJT~9loVD&XQITf1t4`U@D3C+BwcUy&r0@OJQ=!AvF_=yBI>(x)^K7c4RHJ`<r85
z#O}iMy;?VJ?<Wx#>w=>-Da2c)z2CUs?wnOSqiZ|#`I=h$<OpKEhrz(nvg&j;;k+bk
zY}<a9J)qqd!uA^dz96aBF5N#&*mSWgkxS9He~^a8nk`W#b6vP=P)p6hrt`9~7QEza
z4Tqld62;iQ2GP<g1RgX0-j~-^skv3Ko%D7`-c3-h-+9<1Xt3n#V=Z1{Buo|b%6HHb
zk0;dOV8+|GU-)*uv)AB`=O>>ZzcaQ&z*>Y=b<$*~kptmOTLku^mT!UNhP#T98VKnl
zKtIb~UAx6tX3{TM*=#>m<E04e_DGs#cA|SXNZEhC6wTcJu=98e@S;_$>V956XhjQn
zyqLB;mLRH<oTuG?K(ct!*dzsoCG{z|W&|teMABQOElo`pk*Pz_-G<?5dWf;6Y_N+Z
zGcOO>aL;cXtTn#284g)$ES9{yrn6>n9EBc}HT6uam6uLvTi1K)WhOyfi_(l%ovSZy
zN8>m(GYTvjA%6o(>iw=o8#Ot`8(Z^K#a2vf;nIpXdm;(3@^VqNVA^V7jrr9cVT@}q
z(q_R&E^o<Jb?+i5tx;Lb$HniDi-!f}a~}>EX}5j!#_e|F7!J0?hUhITGqHcx7_YYp
zJ0W>$2_hHdZ0fA=%kbTU`YVVva$8qbLX`2Hs=44xd$7O9t~68;FJet%j`T<mnV}s9
zu0`RSB~eBbQ>T_({NaqL1KEmbyEUn=#YDx8P-R<6;p``^pHNbe#m4OYrLI)xBlV}4
zA;F7^n${1kX<u&dg<1nR#Eq^F>&~rrjw1*9y>j$}R<y`|%wT0#=!-9ngr%CZ*cW_|
z0oOv?BR~$HmXJ-PVl{cK3TC!v&qZEDk^4MID9_Y2h4b2@@OlrVZXhi=JRv{9fb~n^
zT6R*{={=GKBDlq~BQ`G*fQ98^m4p&aV!_r9ik%$ojobMh=#1yW?ZWwFT!M~bNUH)R
ztuS<GO~y?u229I2*P<VHe*M!w@TXJ0MX`c89{fu&e*5!(WezgYXGs+P!uKtQ&S1n8
z+2eZdoh8t$ra@d}aMVoBBJkD(F%|xhEJPF4H`;LI6m%4Vq19E^SHi&!?S0a7+$w*;
zHwL*+BJw_tDRax3bF8Ci@Ac6I6W8)p6cqN*KM4s*Ri<PdGeK;#i1@O`f+iuSSU{-2
z4Y`r75W%44d0oqSf4g(Py}^R#$Adrr`FG}=)F^Dfsog%Uxp;nFOtf^#yJi!m9^{mT
zQiF!c@UUt&HMi+ye$Tb3@x|yz!Eb^+>sgq#XTN`gq;RxCKMq>jPDBJYC2ENjWC6?w
zRULV4v`WxeS?Sf@i7~jswXPD0dNgERuVwOUE(-1CUsFnbVT0kpd|o`=@mxLvduetM
zr>0J$PWJl3mfgxeB&DczRtHik$arpgE}prxfc?dCi&QZ$1yXQY;ib0_Cb`n4JM9rj
zm?Pw&61(Fd?X5!>Z(Mt^j~FGb6$<U89uK6$2!u5*cO}r}RO_@7+|Glee<`nlteXEY
zWXnrD$7IeYk=y0gTFKSt^`69FT1nHC?R4#xb`*Ly`cXNoaGZtnYOJ@D5?TkjEgR<c
zs;XU3QHu2D6b5XisRA?8U=pMrjs4y=7ulkUZL#sH*8lK}fYK_xoO0O2L^Dj7!K$d6
zvL)7Enscy+(3^8U<6_%rhn1uxLPSYRjD(NEU!Fc+NglL(Ue&ZJN5ONECW{)~RIMkq
zJH39R*KfERmp>RZ<|>Tg#B^F1NA55e3=>&*>7Q#1+%+8bV)WzA+x^#^Q$SRwwaWSU
zjo(71fS!5>im1jjgm3q%nc42TAJ@eg<7JNM^^XA-HiOU|X)Tb&eW~&@voNgo7vx-P
zR~2TmE7!auwJAmEW5s&S4kE~Hg1lxATW_*wdnF90JRdT?Y(=|O;(k!3L$5EWY73*a
znJl*qQmJmzRSnIGTCKTrAt3}3bBg-H>MJ=TaQRVf%I^F5&b{3!t#B(h=F?f@2V**C
z)8AW(I?PRpb!U1Ojff2uT|Zwfn!D9pD_inRuBB(A9q_?ilTzTgs`Objd@Z3UBlrvb
z@cvqzLg${VDCgROp-32IGOtSmBn(SWoFI3ZCdRC6IvI22atlE=*GS5tZAODC^<JeE
zuGt37rM(Pmq+M-d&IYdwS}=S#&-MI*8%pKMfCs7`Zg;I{{NwWSh~OHMxv!yLtr{FX
zhfxZ0+d_QX8HDAYgB46OwdWcdyn;#%dsUo9q-MjJmouJ&ab3g|Wl|LFY`b^o@*J{4
zI3$X>MnXrY?791q#)R3$?Y+Z$X0Vd$pU4Yo4$M}e5@O>lPqs-;z)CaPA$>~k>fAIG
z-c_F99`6zf7B66%O_QrDcV?}XST2`WNJVv8B<C=%P*S9+pVNI(l3)JfTiafJxisj{
zz<<<}TRVl{^F<H{6P?jwQ$~`mn1z6++hl8<;>OB(QG^mxB>qg47mFwhOa%zre@6ek
zL+73+NrWjuq)xd0y=I3QVd_lX8ME+ws!PD3MZqR;s;CiVJGX)*qex*(cTy^2m%y;J
zoUbfOI|TbJf4#So$<g<Au^goDIe?`m$Mz(Zw3{h~6&ohi|C(cBG<j2HcQj)r=|jw#
z^EfH>jX?VgHc5pS+*Y)?%{SVShjkZIZmblP)pBxn?bcX7e(<j!|H|i&C#{zJOr86A
zykJKV6U_7O?wQaGf7!c86QT=PZD~Q3PiS(=ZKT0pLX>HqmCnY_t>0uC%7L%NSr7j7
z{a5C-xYl^h?Dirr<Bz6wN~~)IK1D(aoI%iLaGd&na`MnH{K_R+?vZp4&1XssSe5w<
zYEuy~KmP{oqqDV_xtFg&_#f1H&Py=W>i)7iV6_%;?7J$_nM*MoOV_}Jn6Si6#qv=*
zC~WS!CXeIdu*PubO64>_x5{e~RS_Mol)^%Z<62IHsIO)zweOWU_TVMb3r2bYPs2XX
z7A-KAca{r|sEH|l@rlbT2+f&mNXgmA(&1ZUvy9r;fXs{h|F6aMSUf(TStVsN;SO=}
z!3TYwl-4!-xDq=?E~UMSnbmX4m=2gOzFFEM=b(7g608+sTFxmViy`6nYq!x|l-ZJd
z2}EW5QvSY{%cQr8;GI-_n|2qpP`4us13_Qzp0rMTi2-vGtZF=GIcB@pEX;Lj?{Wrj
z%b{6>a~zGMc3M3+`a!K-@YtFuf^^}r!pia^7ZfEFJoA;w!b6bT7qd#RLJBsm;+|d3
z>y+zfE=i$X*~yfxKu&R#P;_$0EMKeIN`6lXrCPgDyW{aM%%Orpdm!4^d)PCokmry4
z_E%agCeQKU^YKAHD`zg5nsai^U{WZxt71mU-&|yR%llYNd!X;f8|%7Qmp0vbPH1=M
zy|ge`klBRix=_wG%G5%tRl#~b@7lN>H;&t>`Q3G~avl-+H3=1c?!g#he~}QTvvOa8
zHrCzF?|i#|(=gX7KmPT>dj4QBWCs(4BO#e6Us@4g!Msu@V20H%7lmHY?}g(vh$&RS
zq7bWOlb_I;r0HqR6Y0KOXw(<G%iTGTgSWRkCpU~>KAy>F%iYOi;y$yRR1v!>?5yR1
ziXUrvuT{>CH5r!{1-Z+{g%IOxJ+)WlaV+{du~nFVi*=l9P0G`>0B0`YfJIaL)*9Y)
znQF<UEO`O67S~1ZYr=E417>TTb_6Yo+)>BH%)J8c!Sc!Tdh&Scpbl@$wYZiivq+d;
z3uSzt<UmFOr!htfTLN-hSScEYmYmLux|VdexfG1T3D-_rwe6~;A~X;eHku#KxI`n{
zrDH@w9qE}f708J~Z=%P(n((E)t}_0EIRv!VCJcS~!seuu4kccLKqI|_%qT4Fj;>;D
zhwXZGhgwPHJyIByL2&uvniJM~1&PgxE~VsR8U>uz8l~;oNfrFDgQ6uYM$$hsDJnup
z4U4J5-D_2sXN8ysVLms9-`Lg=t*0$PY<O|KK2NLKS6!8lsYOYl(7=`a$@j6G#JacI
zSQVH$kMz2EX(p6}HpB;!Ngz9g6v}P2t!E<DZ8Wy=ExQ6X8Q<+h3JQKo2vT1=l~zu{
zXC=73b>7~8<M#Gn8DnaeQyZ-dl3VkV4^iH+5~{i{zZeX>6n;GZ%&i^7G>&%4P_8tN
z^Uk^+8oI8)O=3IC(^69?XT><`Nv}N<WI~|i;4=rSFUoZ?#<WnrX?R@H_94$<ZwI&g
z8?7HqcW8}e7am%P8bMOBrB=S?<K=Kr^I1nM%5iYJz4Lc}`X|1fzwr2YFs{L^o|MqM
z&|6TgCB&$+v)5mW?WEP2PH5|372EG@Dq?(I)KWOwsrR<4_^(5+XXtr$v$`k6Wm<kD
zit%GSl?Z`YlNyRQ>&K_=NuXkQ1yR9fp|04yla9?o49z89rsY{6g=}4xYbdP*mb;^U
z=JGE|2iCO#<p1{TuSyDYO<me8((0Axj}Lsz6g<^r1QuF_X-?@y)mqtNsv^Qk8dx<e
zG<R$)v1+i~(Y%<~#Lc;$gLw|dAx`>;NF-uA30h|N96uGPRrs<MW6VixwY1A}`*uIL
z-48AQ$O~K&<|46La{f?r;=fh4AO@A^Fm1ZyrO4tUY6cX0iNkzlU1{?Anai=-UV}+u
zvpRC%R!OHy4ddicp`>b-yMSv=yOH|!9<96v-C3;a@V@>%-5Epk7f&e7^HRI1n`p<u
zyp%k404t?()UQ3oj!eZ??)aN~SZS83()1t-nFcC}l!Ud&077!$FP@k9+O0Is+l`|h
zO89=MYs(qO7=sZ(UlQO!#iA2l*+j~{2RU4Nf-xOiQZu&=Pr6{K5KZYTNh3RwMNOq^
zOSn*47(s<Oz}N{Ui;g^%FIDY^scNpFEG52_u$C3SOZ?Sp<)|l<pxKlA{m%Da{*K>0
zh%xAg*mvQ<2wwZkja{`WYMF`M@_78<Uu)+!7vFEPCb`|d5mY`ue&hQ5o$&dBb=-BX
zikbX~UYLCalkU6~(s>q+-U)Y(*6F>m##2RBI<(dzm@ujo1XBMOtaNHUHB+!fvz~L8
zCy?9bN@K3ckH>=*L30&4zLQrwx&Qi4{OM2sSH8df!t?u+PamxMzwlfOzb57VuAzME
z^z#j?#`E#SSCUvlokX~-dTI?i1xhDmgrkIEo-8SsU^2i%oqjkkn|WV457$^cpHFVN
z1H(1iq_)msjmeYCB*=4j?Tnd%z&%vxHOrIbE02(N(yb+J#+oVg^^!AG(`?R`gn{m~
zVyT!9{7Qwj{|%Bt0KHY=@Ljua@6Gt#o%JEJC0Q|~SjAnI23E1s)Yd1xE2<@?Y}8V-
zYESjAqRxvl_?$D>&8cOcE@wkAbrc|Tl0Yr8uRaeAw4QSik5oxw9B1d-_d9?3<r}3J
ztm*KMOGB#c|C3>nD&F$7xW<;|317>;U4eE;fm@nUq9$w*?5$~IZb};Tqh{;AUP&ea
zN(EuFbU`e%dQhz=%ogeE)LzQeqn{f7CQML@@cEUMvXwo!vOg#tOLD0XF;ily<u(mt
zYppahn3pQUOyVUw0bJGASdNE=(KAu_qg9U9Q_vT$&Pmo2g4dKtl$eOgHB#Mah5NU+
z7vEjg_+q*xbTk=0#ZF41gLZ)nhb>#)&$hRoujTTaffT6^SZgp>Ccc`3Xwy*I3M^Xg
z7)aZTuza2p)48v-Q--Z#_1K?jykh=qxs0Km14V7VyR+uSbv=Lwtqpp;5lc>+@8_NI
zr#}(lJb(L5^iP${l-^*ed!3Qi@|&rW+J{<H#Fev(h1a$6wjVc<J;E7}AJU3Myr`#V
zlZsFPfF;By_kmIb0mYKix0Ej>|LugM-#B{O>zDePXRWm;Avo3(N)tV~F=H_n>X6Pu
z=8Zr8{8E{>+Bi*!z|47HyM5=^KmGUo<3If;e!YL=^2IXefBWq>{^$SK|E!%Lb8#Fu
z{`<fGPu$+$aGW3Y2S4HmkMY6uOK|rR>#euWVFzcuX?<x7`I^~e&GE7XS7o)BsBj%)
za9xUhX2!?IC;!U7Ql`_dn>5L_@yjp2aMaFtUOdj<2(mCV&0emjpw4Y;?@I=V+uDB5
z&-RAF2+{uClWuFWBGFh#2uL+KAZ&_yl)nM{2s7^AZoJ)3iUsXxJj}Vy%Hs(XZF=fm
z>osd26Ey6F6VItb_kwgHRmG_&6pOr2qGBd5m62j9U3sY(SAK1h2}M>By=y1$c{T~I
zrYd|1nb7*q?|0t6-@w$BTq6(b_RksPg~^wy04f8mO020tWJpwGp{-PA3cTGU?{`ZN
z^(6tz8u@|N9fXWcQB-B`FStRSf-Ec~cu_}C_ELLH`6a7=BR8|Hm2k75u$`E!h@RSZ
zrJs~MD_SyA`W`?l;OwELmGxpn$Sy?KYB6!k-Njg$Q+kG83+;BWF#}n9Z%j0p6RvB?
z<I0y<YzT7H&(=74ryy#v6$Pv3xcHpf)e(u?*0L?hJ7!%td9r2%|E&~JrEjO@ViebM
znYHxZl}HmB`NG>|-kA0eMtOZUv9Wm1q9>ECelNv<nf$5HDA08A9VK-!owT3%aec73
z(T+)}f`KeZ<3dq;XP3F7wc}+MKVvc;L)~<E3d*D(sU?F#X`JVs+x<OvgUE*de!COn
zsX^Dc*xplXC7fuL$ZciCp!d$R<f>aUYESMzGNwFlC%5Cq*>04Qb~3Hv%Tr~qLSSnw
z_Dwd^46&BR@pj|w?XH~;4$sGf;~3yh!_rkZDDB4i{&)P{-~A(h|LZ^S?YJ|1^7;Hn
z{^1|~J6b<@WcT;?|M*Y*U;h68M(vfc24{KaD8KP%`!m1&`Oi!S>-p5w>ZsIJ=+SuV
zZw$Ne>;TLJBb8ELdr#(^tfZuic%WnN%ek(LkH?cE5G-{+Qb~Stw1d`-&&NYMr)zz|
zy{+V~^;*+}^L4@1&F;dEA~7CPj?bjWd&qvKCOO5b)bd+Q`Zux8uDx)+b-umdi5U3a
zCs4@6FuP7mk-b6Gf?Hv(VB{K1u_km$E##?m4ON^h$6S2qn$?B|r8Oy@qN6R7vNtj-
z_NoERc{Faf&T&>P_8GFv?p0Ou`}>XKs02&$dw6m5YcA?pHeSI&k5NvaY$wbwVMC(z
z+xl4?=Sq_TweGKAT6msSSNPuQwT<4cgeG*bTJN&owp=D+mDc+UE(#JyH(j7PC7enn
z*6Yx2huW6AZ8z^*OGBh<SgncyvYf5LR>zvdQ~$!kg|G7(I<7gz&KLuaNjVOhL-llx
zEGsn{Vl>U{2*%W9vSvAJ%Jr`O(7+UOCL{L8^Er4tKlOJ!IFFOJ<3{Tz)^6NxZ`{s%
z;`i&75Ud2|MZ3k5lDHGa3V$U$x$IgN7)?~=oaYe9oXJZ$eWm(cGcWCuNIaddOZTEE
zmu=dXG|<<qCD*HR5%%Em{DE;0(O&O!%*iz;USxq?kAvg9QC6eRi|2geLo=S%RWJbT
z#Y@Ne#_jgbaolJ{?6cP9t9y}Ok~;gEO5~XpYNTbo+&J|N=1hN|qC0J+(a(eXdZSz8
zs3+TAMRTV0>J)cC1DWT!8?khsZ?_xw`#bmZj#<!Zml_z&Ps)@}R5zII<o*3uzQ6s-
zFXuPz{owM!+wB|QzW>7eFW>o>|Ly<7aeL>NKm8-$Zht4+jy9?Fj@iLnvXX!L2iK1@
z^1E`!b6Lbe`ZgsznjPU*L}zHzXHqH?6)v$0v_m?OxGrwDJN>+I_In~Bp=aJ3{p?T^
zmMqdBNFW!qy~PiG%UV>h9la}d6anYH){7uxRF2ei^)3qe`PMnkB#o@UK~iY7X!%Sp
zSahz`6<lNR9FnE?mbS|*X7D;BBj9s5t&s^rHWmSzpbuCRVmi!4!8wW%wei-q+})aR
zz6H2du26<PkbQ7(#{GWd``gKJRICJ_k4T#UZR@(<I=y%N<%beEFjQq-N~QH1kH>@O
zc(Rwx9A=a;Gij@v(Iz;tiO#Nl)w1<1d(uHIA#0Arj(sk?)|XVhY{hMSc_-DiJ1kwO
z^N<?3=Abw7@038+41^px)8bd=fGZP?5jBTjn?_m0AS<Qj=Ow!X%b9`{zWUzF_c9O7
z9*lJLT#NOTi|u#}A|@97%fqSc;n|n3%-(=^xnoOhHARWCZ%gm4Jsrz~$Mxj%<H3)Q
zAFMe9HPYzTIqHpCPD;P=cKb%{4M)R81s^5Gz9v5BOGENfrze)?XnkX=m2%wWaBGIo
zL5;<>!QQ@55g}Gv3J>@ESb?}>xg2KZrmgUVV5(4R4~a78blje-Ua9K?;?KmYwEBk4
z$>(_T+s6-ne|~}o^*lJwoA!~9%Dk&E)8JcVj_to%OQp1f)^8l=o#VKR;7~hZ#*fFJ
z|G%{TS(7ALmYs`zXHr!&_lU^Ms&1fx1_YP?|CdmNd?BAGycd@MdZ_BEjKST^RL`{e
zu(p~fghD~LNL8eyWJ+cX?q;Uv>|w2MNpfvQOof~ma#@upl79o$^U6M@$hxhBb}GY~
z?_?x4d3kwZ%L_|dfJkedy;TPGtZ;LdDX#VEQelmgm-WJ07Ou<2%jJc0**!iTOak@T
znU4dU5mYX(e0uqb>-OS>L}IkYR#q<C3)kys+Az{`WnDfI(#qv>AuokN=W@T1mls~Q
zciz|Uj5^uBe`DVpc_d1i97iK2<y(XaB^EKF%*8Wa+|A(`8WD&-Ww6dZ4nl*M^2+*h
z<@)-<vaWJ;J4rU>#HZIQ-`~IUxJ6<QgaWAqmNjzOGQ=c^ok=ZNa{5pN7Ae;HkZZfU
z6`<tA;%dRPM3zh9vSmtHp8LPQ=%ui3D>)Z>Yvi2CWy9>9;qzrFUvRziDPdFOWIJGn
z-CH@~p^w_!ZP~Flp$@ETj-RvU={aRSUpFqB+|(^P8);6usYA)Wcugfe-OqaF%@?!A
z@C9vhd7lb@Nlr&U)6VSK#k4skWyNtF(?~%HO?f7l6*^O-ESi71G6$m1Q6nWSBxI^n
zV8`&8;c#S#V+dOHWz%GigHSzXEM`(nT%)>?r<B7l67#gGIDFY_=44atJ)X%;t}%~(
zkx6MJ6u^wb(_MA=_1@_zK|iyF%{fcD92C<j!<)6itb^V<Gj?GGLQ)lzjM^%LK<$%9
z?Ho;qQ6RA_nxQVauq<+$2QvtXX<ahpeP_Qv2(z=KED5KbNCxqU^0N8^Yb;BCx;|rQ
zSf2DyX|1|Pp*;~51SreGc6nhb7cSQqaw!5dJ|4``BuyM$%krmD?ZembL{_*bAV&N%
z<FD5<w^v=(Y}LD`!L*3YhS4aw(%T_vVQt**Z~VBwarDOZa^dCkPkjFLGbLw^`e4k8
z(V5oiRcff=UCRu~<*h|s!n&~dXKfiWmo<}zs1(ii&)q=CQXUOQH{`OhuA%@Ohq9li
zED+2(o;!?^JP$kQ@UNp0Vq{%bauDRrHNko&qpIv2A|)ox^s^S}Jr#mwV#uVp5Tgqx
z2v{y$){U(!N`hQ6#|$5GVuYPgmc~MPdExr$6W8m75aFdYUcUUy{^L7eFMs0L4}RS4
z)H<kf(z@bN#XZxrHQJzcfiu;O1y1k895S6^_GAr(s4h5fh2<hvTWw7TbC7o<m%yh_
zFZ{fJ_JdvLRk>ugmxb%LKn(OA1>5ti4;6LBvKDz%2E;W{duN#%9hDroysV@pke0~0
zB$g%8>tDms?77FM%T;_3J)GxX!%Z*~an?ucoN~ARXw0M0O)q|{4VsUPOqar}oirvT
zMV6AdZY!@Zs|VMJcV4G4LS~>jY|KW;kc!k#)tWp2V{kNeEQVA+wGLt)E_0mJ-W6v!
zL{?=$ZR&o;FeMr#DQeCK;57dP&QK4Hky7-_#usWSWC&6abkBSxWu?*0`D{_csXjsn
z@KFaACNVFpTb7h6BwuWvKrbKnAX8RCMkH0l>bAtW56nf2D7CwiHQIz69rYZ#ORo*w
zAU0ist2sw-owsfVo2tI8d$(}NeO`}A&4Uyidl2C0lggmZK_?0WJ|>%A-Z?9@<%Gr^
zT^^A(2K7*sWE$v#8UaxRRYhqJ;jaNvQr@J(Ktg~nd1}r$fe340^MEl(De-#!%<Icf
zlycFiWxG(@f$@Xm{!VLm#%Q_>XLbg@Jul0XZxsDd3NbLoK|<<0sl0}pG37_;V<?0r
z`ZBjiN?a)+OJ#L5Ws`Wo(d+ig>-Nf4Hpz5TpfkAl9rD5wd9s3HA6U!;3QJsBQf5o}
zsSHd}imi!2lqn_Fc<$X2r7WZ+Gi_z^;;-3*z+m)7y1o)Hj>nx&pwGssr`IkdK*^N8
zGOQ9)QW%u)s#026@<vL9X@R3P@*<Gi#4q64fIxd3lAwADV+es+G(s!uDg=8B!in>P
z$<&Ct2kz6@mRDXrf9BJt&s;AXm{Drwl2&TI5-joK{m!p%@BH&W{G%oShsG{)r7I6%
zZ&Iq|oJc|A-Hy?kateY)ax4r}rN3Ph+o#O>l9;h`Z#PQn^sxD^O4ZfnweY#FR9lJL
zpv}Q{*?4)qa9tM&@@mvW-X*3JA+|)yA_uG_H#}xSRRn4-kz5d}eicMi3+I2`Isb1{
zANhbQ1j(Wv`$4U0&GF@Cn~H~P!~G0C>KMH%=pjg=HHK2bVn1D#1DXyS3_?n5OX20Z
zux-l|?ND3ec65#jmNkFOCHgll3`2-YplW@1)i70cCp`@NemExLTo%pe18yiesds7V
z*C-6eW!2%8Q>N9*tlhgIJrKXmVKvu-CW7xmgc%s8q}3~$+Pf4@Q5XrUQY%G*XA-M|
zk{M8!MhZfG%;|wK!4*Mf{%}sh7@Uwd(3mQ@l7T7no)+ACU(dO*9P0jFo`L3GHECQG
z2wnou-Whun=Dc;7(CXlk97q7ZKtaD)BO^jLn8|1MuC*%O0p>Hs&6N{<N_N<qb_t_*
zqQR1~m{xt#lYdQP5{e3x)|(ddoQJS3gb*SzYoF6ocm5FKRs~p=3!h$p;<COH)53N8
z%(`8O@$m%5jZvvx!h;gEJDB0j50D#NhB6gCx-345q&ZGbjlQpAAJqLI1Y=utnUj^R
z5lLQk!pZhes9U_QFRbg8vJ}=ZAVzMFJCFTfE143N+-J2ilMpFVuLTJkLh$!Fv924W
zL|PUog^(96*H_l<!gjqt&~v22crx3A_nhdV$Z&RlK$k=_quk@ufh;-s^pRO{7S}@A
z)J?x$UMP9POqbHp26eAI_Ja_z`+)4}t{!c$*D9r5ULYl|msc)j^=x%@?76ITgW5W+
zO}5uBeER%_%Xa0mUAS%=HWg!<N#q&$^7$uz{`nVv{`xEb^3VUwzCS!5%1BA?Z*{q-
z@>!z$bfP-1tP8|KXB>?%G}g&mV%ZYwC5d5XgKiIO4s3QhJ86}K`1&RB%eryd6LoY_
zNxW>C>v}2_voQYkH1&Fz62e~u-xZENi5GPW@-q!j6ZB3jCBk20(Oi$l{@A6QnSuSW
ztD0**g6TR?VhVIiG@BpXn508dfiZI9SsTP5S7|W)`q>o-6Ibb-*KK846;eLB%RqZ)
znp6=>N($X^e}PGu<5IEC(HgT&#Yqxqeb+%40=Y<IJ;QkNUB;L+XNqV6UD!(Dx-P7j
zD=9_}0&Vi4R<bFGpFzsusRT;+*E1%hpa*JBvgvD{Zj?GRou{i9lqoI5cu6wTq;oVC
z1J;G*B}kTPOomtU!}<2ZO*CTK>orcGafp97q<|SuW6$%ToAY_eoiU7Y8Z&sBnPbiC
zUfCa_tVT}xrbc5SC&=-`pW-_6G|#2GW}?I8zxtt?{|rNyVk|OzFV*Luk2AeY^5mm8
zI36kvmQ3hI7`XPoxPiz%s(Kbu)Oa&zE$#I7sB1|n`zWFiTt#;DrVb(N0woB~@j)U!
zF9gYtm4xLyplP(h6UY2`uX=_EaLGg)<m9ZMVKOtiV>)Fa=yTRS;c~gMZX43h!)1G8
zKPvfLlQ3EGDx|d+YG=V<?wv4<vaIC1&|9OdSJJYQ)|Kt$mCNOd0ox|tSZ|$uU+Dp^
zX`%l4&$RZPc~r?MqgZD#m_xTUY7DWkl*p0`mu1mtx2$ZJR}zW6Hs+8N)TUmazG(ba
z9ZNsVo^te<3f1FgOI~CEI~qL+qZ5-%i#e@ef$ef7<%KmB)}p@y!jT*y5+`iS%4Oa7
z^64|L*B5?#{goCQy=KA^T{f&U0W(6J;zmVva484(Ru%20#-yy-v&I92NlZGNRg3IW
zo6W|$B_flTyizvUj)OWnAxSKAS<BPQ)rUAzz3Z|aQ=jP=a9*1+n?RCqDxuOxp)|b^
z8upjJ6qu1#9h^pBbo0OUAOsCrJ)EXH4m)6JLX<g5ZT(dB)dp!wK@|C5D4p2@Pco{{
zh$rq&oo5C>B8-VuSIH+WJckK^F{eflk?d(?9hl^$kk{2wzkw0OD;n-s7$YfU13mmD
z7uIE?WQFvM)S2m{^6yiJWa@)BRK3hOSR;^x<h39fWS^6gr6QW{wUAVIlDz6FNj{I9
zv+pZ(F-W;2WOr~m?{sBN41@VZ3v2XcAy85VudFSo${lm)b$sqn{PzpGD29DVfHfY`
zg&R;pTHPykZ|wJ-$NhnIC`sd~8~}=kgh+Fpf|GU4u;6MA5;g*(2`6wu>n1oFTk0xx
z8|qGt`BVW-lVU2RNo@zMS5i;Z<{27M^XOnFW679QE!=%4GY#tK5E>~iei623l&Sfs
zOOVDvii7MK3_?Nc$C-+SBco3~()3?n+ho5t9=DzMA9w2GLCFbvA%ugxG<ut)G!)wI
zO0Y4dFe&RsUgQQ|)|ENB=Iwc9-LBYU=eLi00AbZz6?wQ-p)N*y>gq#CTwlKM%kTe2
zbN^CUu2<G|BPYp9mt2TjqF!FPj2F`7g~y|E`_sQrdw)takKWie@5F-xF=Jva7cR@n
zRu(8)wnsNGo??a=BhweRIcFCnG+M9pIoU2Rf<n6Y=|-Tos!J~_+kqhYHUzHgMWe*|
zQH+$B$T2_nYjaM8cujro*Ic-+8}Fx-H>m*AUF#u3L(0R<q=&>MOZbtOz?KRY5%BI(
zo=BgVCW2OhxNS{NWa|<)6fK2Qii8wdrsp9nkzD-sG$V~cSv|wup!(i5B9rY@3bG0@
z{bBkPSX7~rV9v<DvQhXq#&NdJI#xGAsqIf8KUwTJSEP<Eb``-7@<E@{NcX91-Zpf&
z<<k^|J}68jR+$6$dLP7=n0>If#$L6g1_*OvLxwWl5`p@t-1i4<G|k^fklnja-K;Jz
z9HG&sd8D9btubm8CZxC24zrtMX}!|sOfd?c|DcO+_DP+MRP_cY0!KVNr7Zd^dSceW
zO=Gi(#A-vlsNf&kk`=rn{rCC|k_X(mYEMr7IjXWKHB}v*eKaxnT^3t<BObmBN{MMA
zD@sFm4$bswn&+i5<#~|25T-T86dK?-I{Uq{@11&hM+L6k&JKaozU8F6h7`2`oy|E_
z$uCdcK4&u2nVc7pH{#@9=>mm_=J)cn2st=gb+^6be#^45Y#SjI%-p<lN{&r1#XdW|
z?&SRrtN!2KZnT*>+Jn{}w6-%wbJw{!R->po<BZVM?rMxdtzFV!H_yp|J*ai^xF6i!
zbtxS82bX0bSSFQ~ydCUy$3mu6BTOys>wb{ei^i7A_A$O3({~t!oL6R25YUXB<9?%$
z$+j%moE(pxW7$YK(OTtb2YYL5*H`}bm*4a0^Jj95Y@6ov#ow=puoUaI5;6AqLRmNJ
z(fRT1PdsjK`u%&QwZl7T(F!`SuKN70i!K>UCj;yQWGTo0kc>M5Q&WO4rP6!3yzu$+
z7o_Hs3U0ss;Ll%w<##{-j$eNMq)?wenYFQ`$WNa?aV;zD*s-Qh+63<wH5#tHb8iQ2
zD6D7Scb!?Lq9d20ug5zrMU!fLIb(q=WHw1M{TYqe-OCduox!9CZ5}7nD!sc(EXgHq
z;arYO)tO>e$CaZe`AQK7uoTJQPaae&`ka|b+C99eQbMFF83$O`^I8KGj4{qK<}Y|D
z^gh^k+0jFYJnlP>UDb$~8|aXUI>d)#vHNuP_!Jd&nD7wOK~5oQ<~96MNH!T=%eLMI
zF?Pur9sXnla!xW!MGyA22_yNu*_hoi4&$US22a&8naQHdH922M6LKTE>om=$>b&>q
zfjiSxB=h;VnK2H(q|n?vOj+Y&(g!k^;MHAFcu(tEe2)42dQ)&j2oM}Uz2r>(4azwB
zpK=~bC-?rm=xk_D(pq=iupZJ}^jgHK#oxrucjk-P=)oN)^9g%a8w#_Bk&w<g84n9?
z;t4VlK9>{c@>FwI8L(C<f<mMl^yuVL3$9{u=2Y-~jw6~=jFPwzRHaRIJCX*v$6yR%
z^X~A>AYK4lK+p6uEM-}Q!ERmZAtJO^xj!BfAdJcklN1+WH;y~|ey6ux2$7QI*v*p6
zTHqOLIeaNz+rw$UP4nX?9586SGkFFxX)Y}J=IwFk?e>P@FuvB<_k(GEFnAHBkEssF
zlC#JHhenEX4uRcLpgM-VwZ`rK{>%+OTIE)ErUicYyWjJ>U;dW06t;C?T^919&iSNt
zEQ?pNOhV6mT3-3(m%rh^y?)|8JN^BQZk>|lCujsnkyr}bwrUp`l1r%la1CHSmh7ff
zvQidm-HHDFTC1)iEXpNt`F87#*KOnbk8k|>_JiNIpIB1X@7Eh^&MYYrZBqA#(C|Jk
ztz8b~$A0ksxO02lIclZ#&iA(;Jof!J9ZMm`=~7zljFw#V|FfciHA-?$S&Wofpo2yE
zt<KNsn9WH5Cy>&GC8vjJLP$c<XFsE+=QYQu9CaxuHHP1Erdgr&Cgr57K0|Q%xQ9XY
zW77R1tzF>1@z)q1?e&3089H>s9<;lhe{p)n>7#~|Uo?aJ0nP{A<$Th;d!;=ghXK<=
zQ+%n36HDt6c=X1@XM3wJjbqBLVla-q6M`1Etu`5|o+_avc2Eq=0DChSk#N!1rsU-F
zjf>1YDQM>XOxB(1Ud9~HV3+B#%-~BrjqeQeEQxFZMJTFO&IBxqvRS)5lhZ~USclNm
zriN{07XQ7%B!o=)&1l842Xz=v$}BNR`|qlpb{tw5W@Q*eAS`0a%!y2eQ{y_zGTWVq
z5rN+%dP+ccrj(694^WC0kfR<x`*&qi@Qi~1OIf5U%AQDv{0bpP!WgO&r`~hvr!{WM
zAy9IbXC<ik!2AI%iI`yqsiT;|_dlIlD=B44RN`Cmkz|UKluq?o2i(OxI$=(Z$9GyS
zq(`P5Z>-zu&R%sCwH@S`DS4G8J{Qs%)>Ml77&>8Hh+$z2(LZ}@q`VPG<S;0qvt-Tu
zTR5H<N?D3f&!bb12bgvz^|%vlvaA<U*(mN)NC}P>se6K$31cu^>=As)Z8qu0gO&=d
zWNtUfu;1%J?M*Pb&p+|`<;oHZ%eGLKBGuVcQD81<cDj^|P$BXuUD#f~@a?Fyx9_}M
zR^NdQh|{w%@X>}57{oDArgUW=I=uE{*Iv<#tG%-3!myyvrA_8&JS_0jw(>XEpV?BT
z&dUG(>p$|3@Bhr+cgl=}VbocPCGz>xC%*o8BLPcq{QC8udH)Z8$E9!T>_=yBjmLg)
zyWP1xZhU+DN*|Ie=P0#fPI_9$+{pz}2A1F%B9XLa!eXRN`@k3`X@$KuTJz^|8lz68
zrDvgJVWO~}(;@Hs7gPRB&Jv9bbCPqI&)}|j%DX_QlNgw@D8rjhirUj^r{Lhike=SJ
zF{cC}f9XqMv`M3DrZ;@Ksu_xu7E>DS*@rJc9Fq`aBqR9f#WY`Dg4(H-KpaCASx#Jw
z<fsZ~R~1+NLwcIULX^74rk2;i<qxeZm!S79zJeJc3`nwl$Kb&yLzQ-3lh2!VF)X>j
z=;WyDpk%euqcvvCPs-vs_5fG!<RmFuaFv#;7({54t6A@7qzBiNcfC=gX((vZo-)((
zvJ}$6$0)uGKL6}+?h~B<r1@iacpsQ6h~`l8-Smp#G~t-<W=0>X@WR>a;7SXZ+7d-s
zq!YSV^57SfzR%h_xyd;hqM(B*`*K``@>A_}b~qZ^p7>_Xm!C^HAIXFeJYc@NQOqT=
zmQ`q((B{j!vaT1d*Q?MVLAVgfLZ?PzG*0BnQypZ3R(BeKVF#^ix6mA*sIM!NVv#4J
z$S+eIL1J!v3k>PS+rb|0Op~4{1}$2-1elS2bm&9D_XuaPm4z^kc09;&0iSCxF>%?h
zeE$4}FJHd!`tqtvu2;(aK_7$L{SK`WgMx!*AI#P`W~aoRb-57JfyKuCctA{CuNN-s
z#+nxv=S;;YV_NH-8KmB;OxA7T5;C;OLgJD#TiG}r=V!v-lC#EiF%Y7;EW5j@PH+CP
zKe#>aGPl)(F}vWJd6Bc<I@Vx69=zS&VY{%sXwR|6L=2gq=1=^5Jox<bN{WeAtE%;!
zp4kT>1Q?w;8b7{&XDbVHQ&s)wmACsl-@bk2{l{1K_jhuLtm{J0k#e~}T7)pkMHkFg
z$XlYUfxOC%9J8WxPiic{7PR{VXJoah&p<ib16;-A%q*=wG<xR@jnbJRHQ))!R?1vX
zIumCY{^tbA$;WioS@Qj&;z%okhc@?N{Z%j*$1%ls>3ZRxrmT664g8G27n|nNw5UUB
zG$n<N-YH>HVq}d;Uxbq%a)?})%({f<!`wok`^bYhNXqcj3+Qa0a3&=6Al-cXaPQ^p
zVdFuUf+TcPOX#F3$^FBG7$^mQ0o9J@9NZA64Sye}V;HBDYA%?wa*TB!L`U@bn>}<X
zrYIbU4nzGj_<qH|0{^PLa(6`WsdO^`HD<ma3HrKJZSY)L3l;4UI|Cas-weZ@tdcxh
zFa}D&!YP@$a5ZZ4=o=tuf1^I)iR{1{VD#bH3_$P;*vFgR5$f<iFC0;3)jrP-$>Q^a
zo4o+&eNbzo9hK2l4TU+Gy+I2Y*=ezb?Q-Ge^%Kjooi86T1yXkP7elW-$u9;y$fwgh
zM2(sbm-)7)(O^zR7eXlVo%m+SOFfOsa?!42g3Sn~C3ndeN?w_5Q1XN=Fs_q!9MpO^
zF5M}!qcLkEteL#%poYNZa^cHosbf~hsh6Y->=4vV=9F2Y9%U739|-h0wCo<9QM|5&
zH5c#N%>jVX?`L~1;*T+C{U9v`mW#Wv8;@hB9Xs^S?9F#B3%OiamLlcdoE&x6xFoOY
z3WFDkbIK9jn>c7)l-w}~(+#1so|zOXHtZ51XKFo2sqpW9`5m9HSFk}$nJ@=OJ$Qe6
zb5>Jkw9c_VxZmEnZL4IV(|F%^zW(^ekGJpaKi=W!EFrO7Ho6(x>npF<S8NR4Z*R2Y
z2YFezUJ~myQC7LW37wFA=N9n}>qNz*XU@#I%Wxx*E7x?+n422U=YC=zKY!}y;Ah8)
zU)QocJkN(`9?j^_9haVYV>ZRhrsO-g^Jw{B{P;Ir3XjK5JtAW?Cr1u!W-W%#(Wz8a
z^eGR(bY;wQZl4k*$Vt0}$m;D=Oc_wSddbGM1lA=nW8pF8S$KO(Dy34N?qYRri?zwD
z6B;m{g}mf=***@upvA1-C(RJoq7aiYrXuo&`0lwRNn@AwH`rQ}!pfJ|YBxQ@A?U!w
zGj${C9A-S_JT9LdlQ<f73|ctls2+9`J?U&&Bv*~$`LOB(<+uyqTbzgK=<<2Aruq63
zSIv|0eaG}RX1ZK5eFznL3^KEwA4LBj4&mVWi8Ak0fYa<IAnR18mx3>*2h+OLKOfr7
z=VekITRrCadg}XFL!eF}h4;r!^||ltO~FY1z0Tf<7$pXB$!zPY1@p2((8FM1(yh5@
zKq+SyB4PN_-gH>bIiHedF?aC5beXxX+b7n15t-FTWPMHu=Fy{>nbBA*W&(*h#UDy(
zA<c<pgP~9zTr%37Z`6Aumc_Y35)qto+;v@f*)Eb^yT2t_dM+VLN-S|<iP-^mnkkRo
z;Tq)FUN<bj%Eo%VD&T%-ncZ{e8A^c)tkx^H$DLc<sbf-K8o5s%kAts|AN=_7EA@Vp
zSEbA8tqy8c*pfpM_j(u2H)eIIHYqtWhGf#GU}f55Z<Uw|+qON!ppK(*wEFDwEl0)y
zFWZ%DP+Mj3+2Bzt?~faOs5=?$V87pZ+~4_@Bai(~pT=G*_uIQ9;Y*@FBCR)K&iwwD
z-|=7m!++xC^1|qYfBNG;@Z%r9%k-BM%bF=$_M|+GF)api3fNT9sb@o%LbUAEQ^|~{
zD~8S}QC%ANLnu`z=g!4M;T5JE|L+1dK66#3KiA>J>fquNK0HUku(RK=zXIl>qc?UA
z_wf4{F<k`{lMW9H<}aL4XD3@=O-gN>Q*-Y%1-2Y1rsZ;pA_Fcl@aekn>3SjOOeZoQ
zjmIux;)$BIpa-jtiRGzW<l~S@&MVeuWjYN*Z;k~#S}zlG(cj5cAjvaLV_}-+b2&$r
zWpO8PkOOy0nt75R=LlLp{ie!~DEBsm?B=oPn2n6ZL?2!1fYcP<i7JUwGTXMWT~^k$
zNUacj^b);7z_AG?H5aI3lIGtpx}?#-k_#pJA%0RngA^E}9eycjJZ0G<)(z!D1dV!z
z1*|u{wp}juQx=<oI+nErwk5D;UHIxi%zF&HxH$uK;?tdtiIIfoWF-H3hHJ#@-l#_>
z3?oFNO*iPxfY~6MC#E?fI3<@py5x6^sxe0JQR8sA<7hnNJ<&|}1i5Jkq^#r1_J!A%
zpR@}YgToqQR_ak1alm@<C7^e2j%EmSf<+I4%C5|7m<ic^x}E#bdy#CqLJL}L)Y=G3
ze2OG+nRm+SHgilQf;<40o|z3P2>>_@N?CY){mk|HBB}Cp&3sTb->ft14x_5O3sSeW
z+F(c-_q}~*e|v}CI7a9D`**(n;eXKH->|N3b6c;JvMGby)GhUK$Wcxd=+LvWbJXRw
zl>((umJ4gXkb;~v!7@u++18C|#<pDuXdjeQV!d3Mwd%Y`Q9D6@uFW)}8vOw49QSt~
z_cvlNj$;8|<Zd|v8CwhQDR9ievabAxfA@F%^7@687Dk!;{L|0;!ygOBexqCVF^*VA
z7Lu>^+)*F3Gqn*ME$+zin9emThL41!&Z~3$JIy1EafOtt-KGbL#dAs#rpDeLz@sPu
zQ!gKrlTy3F#^%SLd5U7jUt@ekoLZavL%<9S4R#@F$(6D+=$S~#f!?L{mx@I%V8U1o
zsfRqbwv@yY120SB=hqi5Yr*EAnXv?ysZCStU271M-lC}rFo!@46VKuEsJ(MU;s>g9
zX6hD7W?7~EzPxNQxA{mXrTF9&#WQHc*+9>#e142V%*O~R7S2#~xv1mwkPSgfpOVD}
zVhpOCA2W>sVt{4MTrLainkkO<a3i@dHFaPL@jd9ym{CT+@MNY$iBZ0a6j^*RJ<pNV
zZ*<}kd`XR?u0|crIJ*WP3Y!yAhj%m~K#7Tv^>w~pGuxI~N+K^=DyN_itr}!?Dbriy
zXMw~=Qk9u8)Cx(Ra;lD|>1%F7P{Kv2d11=Gpzr@w`kYx8x-`b9crWhXuXhK}QI>x#
zz84G5qKYB0<c(Z5a#|#dH7SeQsI=oC#)bj$?25bSy4!Zq;o&*u=LHsmE~S0HtLqGj
zIES39t&(z9W$cUdl8Z>G^<efZmgIqXUd(f_luV-N&xk{Xz`A0=tJ|e;VrHU`T#weI
zZY!Cicg0&S`ne^XB+kHNe{g%h%WpIrw|?-?fA}AK|EGU~y{S7NCLuwL!lBG_sVNC|
z;6blLNr#Tp$SIMwix)vxsk#=K^R}`wVy4k4DUluCm~x_yMw>$t&XmZjy7&F<F29UP
z2$Dhqo7~>tx!-P-MGO7Y;fqt7W6rQ7W|H(g$4DF!Nwm>fR~4qMOL$NRgn^dJoO?2Z
zjz16TL^~es=2&<g&gVkS|J_WVN&f%}Eu)J~UA9veg1L|D%ndh}BkQJVt#Qg0yX#Pu
z8Y#u%u#ZuqxBq|HOJP~FW|K*glzP;?%9WFZ5y5i83dE6Fds3AU0!x6;C34A$*qdze
zGtpdfiA}axSYx2X0I5*-15EHZhTwX&b}Te4SE3HfLSV^};!C=~qz@dJrnyTG;h9*r
zl!dactXC&b`Uppnc9k=oZ1{mlh>WRdyU`KHzIT?AxU4H{DlDgo#Sz;^AebtMFwSaJ
z(%bW3H=p;K0iG$ats7fjq>{<W<$Jg^Ril)Es-*5?BURRJ3apbHm5!CBT3cLCKf4y(
zbn|L|kRt>ijrHl#lrb@5U`C^xxldry`;N^{ih*7NONjimE&TlHLRkyxvXHk0OMyPY
z=D|Sp%<ho2XVCm~^tz9#hUwRgR(U)g9FLuTba^>?CoHg7f;I`IF!D;=UdS)6P*!FP
zjPRkH8KV)c`^>j{D-Cm2y3gKq$WB$t770_?1u>|qUDD#?%0`YxteyJU+4r3g8lyS!
z(?@#KU<PB^R!3|n=4_ZbI8FC2y$$j*6b2vsurk>9cVJOv+&guD@a@MpUe=Agyi)t*
z=!4t)57HcL>nfklfV8ZP*+@$!(1}H_dkTWl1utS^(C*<@A7V_6#_yJ2*g|H_x<uAA
z*!K&!Z{PUw<12j}yx+g^^}qi&Zr{H%j>@n=B@*MtC4I(nqK?Wje{ez}XOl1>gv6Ft
z0#QLpMNHRjl{LR|eR<{e@`*JUdfNHXAH<y5mX$R`wvd>Xd0j57Ir8)CXRbv{`S)mi
zd;8A&?TssmZzXZmMjb=5_!!Blc=xFBX-*4U5si6{>O0Vnok#D=k8Vx4oV^axF&UxD
zPmn%R4pT&eiqWO^IYlV>6f3x@(FH9wf-fBA(;IO*`IX=|o~NprN`7TGdk|DQ;|B(I
z8hfXb{V=B^a7F+O)wI76#{5Mug;)Zq6h=slrln>g`cgJYr(OcfA_`5dU9`^ug~*Zv
zpVq<_gJKe^TH03iLa&SDix#5aiqgY3AJw$hsi%87iw*KLSj8x6qQowK@Mx|bL-Bc?
zboZqcPxne%$c*vK5lAsGLV6NQQ;g!%gii49RR{rCmW5BR7q)Go6zROj98bg38Q*u-
z3DfN9j-B1J2K+MA3{dEWbZVt@b`*x#QH?29k{Ka%rEUdN7v#unn_}Q+r((vC&M)D)
zfP~+49fn%ovr-=lFbZjCbae)PxS3E?NG-Bs5K{Tm>nmSAU0Ao3a#;!6Mhk&QZPcm5
zs1DI5QxviGnJwWT)Tf$jewY|-ez1*3>vG4gSFV=}>E%j!edYDj7uIFf>z|8$FLUeU
zx&QD73q1X;oia|D?6MO=qYrP@-8rIVI{(le^$H7W?E6Dri#{pkLReNn>bcN0vyZ+Q
zZoRp#e88N7qD@~A7pbP!byMLo29JB=e!DY<aryC`Z{L39S~i%2m?Ha8c|2~6E)!E2
zMwmJX2!Sq_6zj4`5R#>sX?{SaA_Jg`)afjLDZg@(iYYj`G!aN<^wv1OedqZ8gZZe8
zW*khG^@T5=f6tfePt2G&`a!LghwWf8oy{P*Wb6ky7A$z$pGj9QBt#&s3n3SEIdv6&
zBr8$y*p)f*^5x3Qwn*NL4#ewr;d<3F{5W>chK^ibUwOH{vgFlag+8CRK$vh0<5BOt
z-`|)-Zu^6}G)mYQ0lvSzal5^X8r&nJC1w_3HpZ!p?O>Yq4;)$Vgj}2C+?DB)BEurn
z+z5%rwI|J44aM0}s}mq1wV@X|GdilqJt^38cJUOj1n;m%pR{(0eNMF<f1OJq_@Qd2
zlEjQ@Zn)58M@~q`8_dl#k<g9gOT;zdn)@si!_l-zoba*~dW<yC5^Eaod*i-W(N4z)
zqvu@Wj`Ps<xxAUMyfJGYD<IxrPE}+|vs#`!AxvR0Gk6w3A3SQOxjS{qiHuOtF$|a8
zYGGKi@{G6Z3zuyprX;y;KVWw1QT6+qJP98QToE%{)nO;Pq)^8ZrH<uRT`CK$4IZ^I
z<2k0%`_Vivq2$a`R7o5|JCvOD5Dbs-i*B9?zll5@W?IPiGZr)urbbM2cEa?<Sqy}6
zUWS6doeQ!YHFnU0xsfglmZhIROu;TV5Kh528Z!nny{hs2T6|&tn~{-?NpE^D`j}i&
z;pO_ucD=A%Uszv%B9-;&CQdGkbgcsh9to)`{anKJrUE12Au|ydI_PWAAuu4`$az&j
zR#a$A8<jSyM|nzZma>N+6~#%$BXCshh~^v)#Oq8OM9dwmACBS6<x1Hut|%MSIB4~t
z_s;tEjh7#v`RVnGlVuZoJ+$|Tfn%@4kv#rVRsW+Nl*^`+wiC9PA}NXRxn>0#g=nN>
zXYUmS8fkeyrG<vsHGV9a<Iy=DciKZRyfFe{UHR$fzva{GFKqeZs<!B{oDw<UopXp%
z*$wM1(;lQHQObq56gPP;T&^#?yb7H8UJq((Jo<x;i~;+8@OoMG%=m<%gv3vuf8u(*
zXq=rR+v|l-U%v3^^2)Ysx{N%rag0WL931rq=kRtKx7$1Q%kL<39><-3{^LJ!zu%<z
znYv7vdWM*q5CuPRPa<(|QYLDQD_lh>8l4SmI?RIya0NG^cD0Y^`Un^0{8lX#>|<|{
z{F#}<6?}HexP|diZb<(P*1`$F_E*>`Scp=-WK=$lE|o@*ylMvXgUhL$L>-1<l1=8V
zVSxsk-KLDij**y5?zQvdaY$yDRYl%f=iWNa7c)5t;T)2yZ@iM$JkeNAIVAQ>MzW|1
z?$m~l-Wkr?>5h#yj8;4Q(KxCkq9U^ztCKK?4Mx=gms4V0JqKV}Shfo}MOB9_61qTv
z0CaQC{>fcAsK>#mmFY`cPyk2ojTEk!xA^!%{WPwe691eM`MMEedeS0O0P9WFN>H`Z
z&t+7y=dwGRCu5mCTh9}R4qDtAljMF9x_-{=#=bZ9U2gLb-Hl!rwwEj0^&%oEMs;O`
z_&7NC$vY049uDu}HNW}MO_+_=I!EgqRRK=Jbn&NB{J>dRw~Y`pM?Gh-a>#4>-<h*Z
zN;$g7wo{ivtBqF2Q$`#kUb#uSH_d3hV)t({JyKbgm5`v%N{mwVK%y)w>$-BeUb$Q(
zhs{|YjpMlKYw%gOyb?ns^q9rIDT{;*DQV6fJj^NzEmDs=`{T}04^IIzX6t&-L*RbD
zlTG`hF&po<AMD4@_DMQ^0TTsg%uC_r`oeWt*_MSAU|Sb{?02S3j?rX1^T4Oxl>vS4
zJ3sE<xj$|~&9}z=sH~S)e*XF2QSwC#_B9h`=CZv=+PxNLfX8u&GMh5%Qp5n-cPItB
zypUf%@%s4}e){=mE^%SD!G8VBV}9lP`*%Xz7<pAjb6(kN<>mE->58&8Crc{4zP_?-
z0?A#kFKp|^WxcSh8>tkIe(*RB?vFeDc4N#+j*)^2Mq1<Rw?Fgt_MOM$!TZ~H6qKbM
z<Lr`86CGyhUHPd7nkET-Jvy~@iBx9MMUc_BR;m+?o(38XjR8?2iEv&H=W*_rgv_D(
zb;6%Z{WM1Kt)!OT$x{wv)J{*rps7p-`|a=i|NqCo&AfKVBzB^KQ_g_QLor224rm%4
zQ#o~l**OXdf%g%~X0FbGV?rZf7P;3(c{~UaE@8z8>>WpAXvA>lkH+NAf)J%~38!Y<
zLqdix0fVbiT*5cUWP~#=PtlMuDh=tVgQIq8Q{vQWkyuU&;vlEQNTRu#dkA6-1V`w`
zB+o(#lV*dQ7kobByO%)kjl(;H>8Nm36_A1w+WHuzX~abVC}lCSzXo_{P);(6mAt4T
zcC+4bH0G?X3e&}GX2dDah{gXI({XlRx*#N3ixG7}ITLW%=v38y^ucjfLI`F?I+MPf
z_tGUuuo$76<4(Md^`c=0Jy7PIgH2=j=wo+1{A1s_?*~V1GQve&x+N#plAqFIg`4Oy
zXsy!f;jj5Qe(ECmHXYEdcjaf~AWQ-$kS}=$q>-MM)|I0-Vks<XVOg()xDccEI_q+w
zEUV`dXWuF=5=qpD&!wx73iY7&$_SI3T)ym_8=kLEWL*}PlDWPA;Qjk|zWn^qLvNFa
zLW$<GP;zDlV?XZH3UhRBZ{L``QGyOGGo#i<GFYz{wsqxY+gOvj(wMY|xji1#UdaS<
ziX5$RJ041ad%W}G{Rgd9y?#T{s%4m#@G}rf(r9_YdU<8LT)DnpNoAq>d!Wm@vF|%9
zFKm}9FE5{YeSPK2=byM-E)>)Ayp}>tk=iRQ4_Xi+<LmnyDZwwVFaAsi<}qkhm-Xdk
zBbUf!Sttnaavp^<L}p5`6l`qV-X83|36L2Ij7Hz@9DUMiXUmx_XGeKt&lK1GhM<w%
zi3J&UI>Aq$kYVmA2tjj=?#PWfP^VxSi-x(!3zwh9bKYadSu>pyR0CHb@*I7hR#NCb
zE>+>$JrDf}E&XdGuk~sKq;APEkzGPJ=lmc~y1lzGh*3#Q3IU&3S71d7lN^oF;L)ep
z9DV2oLQl$YzRoaX#`u&!E=i#`7A0>DLFfoTMAGPdku&ki$0%CbAUWPp_d#g>-;`g_
z=1DB=STM{~1#3Y$-K-|ug5ZqmoEAp|$CSv!MD?tCz;XF(IE@@gb~<Z&?n%s-yXQV2
z#=x>JESC#22>g~UJT?8x$t!X;{xIVpaKx%)vXO=Pz({@}pC7EzhqV7q&hOTntE1w_
z&vY+Gj#^TmCNrDr#+oT>A}4XENKs4aAPVMmGtZP{%2h()BeVyeGCMhGr5XxR4$jsa
zd#gO^!QLvhch;Cxk@;U5cT|ma{Oh3&#;lC#q}}OZPOZ@$JSdd8ik4h7Ml>{5Nl8Gy
zv(1hOR8|%tpD)`BV;GP9PE3`Y#Y2kDX_AC6J`bT7f)e#wbsp8}-SI==H<y9_8AuV1
zQ(+UJb#;|{^G+cbfA-{_U6z&m<4&y)tapybO@H>Yb23SCbC#zeC6-j!@}hNQpDZbJ
z4?$f)y^}&@f9yQ=`_pJPY9q{{Ebpdyer?Lrn6ExYi5lFMb)}>%)c1N(w^|m)Y~1hf
zly#RO?Ddt`bYWc=US3{&FO&!uxaP^LJ@|(|e5c*ssFzRFIr%<~WP?AiE9vI|!#G;!
z@woGN>_VXgMb(Br=~LZVNZQ>jpI`VXM(XIizJKMv{~w+4>z`RmRwzvnSIXM>5Um8M
zjdF@k3-Fy-IGIjE`<<X7#9Y7-+_^bCQegNl&gLK{xr|Fu^{;2E`@Zd*2sl0mBPCI(
zLz-F~o5qvrR4{pAGpFK2ACF4i1p>7hm`#_&U;I*-$A_y<ZLV<)A8+XV!$M*Xmyrcn
zibvFWKtSuw4}4{-+SE*&Iml7g{v`y$>|z-CEI9flRI<iGlR%F~!wA{js-L_fO{BGW
zwWq^#Mn`Ygh9NX|9XrpwRj<&^d;}JtGB`@0EJ^B)uFGkR!AzNqVNsc0HYW`&)cTx+
z9NCscwZTNNIw?GLlc(9}`-A;{Xu*AfvMoB)*F;$pIR{F)=;rs)M{rP92zYfH-07@~
zH-$j)4-}?{L(4D`7XA3~g|ZfMF~}vWhe=0T@|M*!%w(b}+g$2Pgf(TZTOnnss+v37
z>kK^Fq)&HShtrT1+;<UFTm&t$73ik1i1I9CW@}xEZnHI53&GGucGc3=-hVJle7Z?<
z`iQ<{_^%!23*)HnCMkHvOb`FE6xa8E<llx8i6kh2b-VI<{e_q9m2KNpzz|tC4?2pc
z!??H{&r}5AJFsrF{UG$6*$zfi4gp!IX!!Cz1qJd1*PCaI;upqf2m9keyF7SZ7I_(5
zqP>O85jF0beJ5HYlBj#59g?<M1S9hr73sP}_hW=eNLhQ2<IZEhgC+m0sGCg$UY3ow
zyl@YR7&CoNtPe79yZzw((`RLxG{-U+gp^sg3#Dv)xx8|{uEZR9yS?*%yHWC@%6mqZ
zMQHlh%S9R8!zd|ojLO@OAN-F${#X9|hkw?N@A^Vb|B+w6{-D@GPJ#QV{L7DT+<v@q
z{CF4X*E)OMc|UICFFz6U%9qzGpI<()U9Lp*J$}CY4C@Pj{BQq<5+Il4J2Ro~li^|z
zXoq2G5|g7%3V;HgG=g=&<*AuK%qLMWJYDkKmzgo5RAI@wlMwW)&gF4IJKPXq<OFf(
zxfo5!ls*+v$?9ZO*Hnx?Y1L>AdL7I*==HC_K5DHIL!=~C3L^sX!<nZ*kxJF~Bs0gH
z9=_r_@_4#@eJo-2v~DjckrQI^J9asBU&J8)Ku$iIvT(u|e9oo~d~_@X_}3P^(nSa8
z-en9L!yI*dc53F1;eaItXw3X!+wbF~41VA{hHK34^^j33dCNXcRc0wt1w|cr=Otw(
z#6~?tn2j+I!@{y8)~!e{?xe|e-F{Q?oxn0FnTAbCXwQxcjf1L>gknsbh>UZVd8&*8
z_;^T{RZ02T3NsyQ=5Et-F@0{zqp6c9DY2D>%Tfp_J=4{OUt;bA&|Be`Z#)m~$Ww~v
zTRgz1P0WAAQHn=Zl8Cq=$i#HiUDR4&H%Vnv3KY-nRdMC`bwgcdOypJ2j^W)vaDI<k
zOSS4TI{zvJw(ZJwdzCpWujH~|(I6zos3_SiVK&4;GO1r`ebC!Zs|T%VVO(qD*bfnf
z&#O78%Ir-D=i1as^roe>Fb9LtE2HglT8C512B%#znpRFhJ5!lD_nq7O8*9#_Z6R(#
zX}8vRn6a*4Q3p)-9ZqeH=p(8WGkJ>=JG4){yu9*w`=J6Lswf<L<^9*M+^(O<Hd*W<
zX|p3%QYl<tuDo2Yysj&GEo|$`^?GGa?RLV%5!;1QG_oX2mXRqT^Riy~-A}*6>dyYS
z@nhfl_3NL=AyI&BT{&js_Wqsk-~Pn=_itElJZ|sY?r%KW?|7Y)bzOM9zVPMsh3oYb
zIR+lLJAd=#Xa2YU{9oALzViJKe<bHZNrj_5INI@~J0FL->Ln`|nu0_&9@{yQDzCSJ
z&FBWr)BAFo8};19pre0GDKl-Fca1o(a_GG=$Z2SFY8o@8IR#KOMH&^NjGOVpC0??W
zXZ<)lPeUDHZ-0eF^Bhv@1Ycf)FN~uqN1MU~Xeq2qVUGD6HKmk9>I;E5joKS@bplHY
zT(-=TJzK#8?qiO`uJkRK?g~~OZo<eW<w(wMXgUwDvj+>#-I<dy{SAun*c->AI`-Lq
zBg>N{&qAb)#9oEG?q#y(kC)4mjC*x-ZjYTl26=Il9xiR28Zo75j*pPkY|g0#YVEYq
zNfWTax)rY13+r{ItoaGq6E;Cz)|u=zgQUl;H*Xm$`+jiW58Ba4L6!Wuu$PHNRgZ$`
zrBGV(;*wikv~x^{@ws8})Qr0OGRH^>LN%;8E7>Y7&R{w7GvvKEFa7iP2>uz6Sa|py
zWBwj!usWQNbH<656d@FHgx)2^oMDj4!3<GVEAr5Y#iR>HlW7?(V}P=JEM04rxkuyS
zGwh{2M@h@FLPar>%Sws`ND#CfCKf`=lKqCDC3e@T-1@<>-*q`0yLLyTa~ze&W9K+3
z!Ew(7PYRfW(G)e?$I#+2oZ^H@tvknY=a`M=F{5xkZxb7iH>uj2%nohexxIg<1S8iA
zl*s%0#u6jxdWEF!Zm(7Qn?p#Em-U60?Nyz3fZOdxJ*ww9Y=orF7$>nlzJKNR`7=31
zDCsG~6$wQ&fNfj2x*Dn!J*%}19^rsFSXb>(&WXZw{KyhBuh&=V<qI>fT$U?8{`3!g
z`}U2O?UmQEVRP`l-?{(z#;?Er3%B=otXBHIv+oaXN8@KRF{EB!dAVG9*;XyS*DF6?
zKJ)A46++}zcliD%63Ih&hHy5+SRE^Vjw#h8`lAuem2Soy;W>U(hkD9F1!yz^K3~<d
z=EW@2^HBy2Pf+V~GLn%@V=WR4i}V>NQv_DC#Da*%r#~&ArFF_dP|FbLKEC>kUkdGL
z0>Z^fJgd9lYM%rHMd(S=CcC$dDRz8HSmtTe3f?vcuRNC=xULIJ);OUJ<E?dW&8Smz
z<PZWZnx|US<)LL+B>Q4C<*d1}Ne@4#2_0+0qbQ;I1Gpd6A5L{EvZae8nrft1$wv2W
zaR!YmczjE4p%ac>gwhaL1FYwz<r2pbJlSkU4-SIbH0xI6r)0c22BE66wDNNreRkqd
z)i%tXqQ7xd%k5evquqDj-*4P+J2s4UDXb^LAq1&brdaQ|oUf$9wum`3y94=Lt3OZ5
zA>%2BIA^Ij1)@c{XO}`*JT5efNa}x_-zC1!z?D}`CbFeu2+G=MZPI&Xw(&InM50qY
z;<K4(z4NGtG~0b(p8TaznmxO`6Y>^QB;`zw^3If!{UMfUA!bid3m&YJlP>9`B(E3}
zAudlkYAGu*7iw$%;6;tL2r@a18;|>q<9_2f?$pPP-XyE6hg2!;n3T<D!Qc$5k90RP
zsikJC?5%Pe6I*Vy{l@)vWA$9?+Pm@sX2ZI>dg6)B=+wu9oFcbi<WaFD%GvAVwab*_
zdheC{@!)>!%sD7!<z>67a}1h`|5$$zV<xYwI%Hp1<`C$Q2e-FxESC!}pFV3`7x2zC
zkaA(WY+Npj!gTsMKR4y6TV!1_B`cjTCoN=S78r0V3u{@J7TA`JPoF>Y5C0&k=)PVV
zz4P_^S00ZW_xp_yB8w%Ks0)6^Kq@QO%azxc7p|+ilF=(I20nlK#OM7dQd*hQIK}~<
z_pt9fV?fR;Oz*7230QP9Y79@bjvV->4?$zn);r@d$i-okL8EEBNE@yUl4-7g?94{@
z9j1gi=U=f1QBP4Zt6!(-iG_)^4r7plu`P)(3pOVE9JK#g==Ybo6#6iXPDd?y@a{i(
z_JD=p6wV35m%G6jVL;B5sTX4UOmfyv4uK^nu4`gjHJ=^^`!H_Rct4u&2~<f=*W?D%
zO&x+RgyQU*kb<`UQ=^MPca((wTWd`--X?ippIux25r`qtf`Di#B~sTCYwgg6x}Igp
zI$YAkM%0A{FeJMTK_e2Om9uQ<von(^y=k7^PLcU(oK?E!+L#tfZMs5hPNzuv%yO!V
zA5$60N3G00sP}{8cF^~xGdyR7#t(H|(@qz)4sB%uNXq(~U`e0*gG(mB+?AbA$)-Ah
z)B6x}{E*T4Y=89W<|0||=UF-Vh{${|L%{3wQ&x<^&t07cLEQvmI<+d;A5q?y7)8rX
z(WTE*mwV3ndC<$y7%4@vO?><^9O5`^ctVzs348a;FUrXs0?W2?*{*Dx@)67s;S(u>
z%vCkIpc`XUY*Yd=tJPLT)tw)&XFQ9CqD)_=o|P~RMx+|gQREn%(HqC@jk~KQnl<+0
z&R!q1Q6ab*Fgf0%S8ObdzB3Oa-V2Y*h2wIewx%4Cr^?F%ee%Lm79D7PU=f#IuQF2N
zSd{7H-#Mv|o!$qQBg^Fizz9ZJ6d@YUfP2T*i|iX{v(s#FSqkgAvM$-9DTQQ-%fKG$
z_KPRC4I5mX*lJ^X?s(*~Toqv(va0MJupXHtUS2=*`ud4&)4rl)Vf<5$)S!a1j>+rw
zmB0JjzhfK+?>~Of+fCEn)5~LMHxu2!*<hyd49l6gU_@^*IALH;nSN)?63N(yG*nQ+
z_LQ>@FS^bmArP_jk=GKV45cUk%J9H4cWL`hT5+Nwx;e5yJV?p3)A(2X_&53#)Y8oE
zhg%G!5+$i3VyDW-AAIvsg0#33%W7@RCe}MP2Dz-fTozt0nRQ)=7<CxCO^zY?+tF2x
zjUm+!rV14jG2yX|=Tg((KtEFmO3lKZxB+Fh4hMA%2mF~q2=lpQifIy>4vgC2=tgge
zm&=L~h2wT~?mQo0gEL_%<>W@F@tgshxue-W_5xDAxRUEB6*a}0Hnp#js7u$|P#uzv
z$8dB-tBrm%&l2{%%5Ub+dN~}yOVEeb8d#G0!5{c%rIYj^<6YfZl{D9>TguM+5p8o8
zE$QV@uM!auMYued;pb;#&ZiMC*xB~F+uB{m$$ca<D!?fx8HQ@@Sd5e`fE_l;MJBM*
zm3($B=ME!hje3%gkCJy7XJ+;^V;#%W)gDpM&CwcdG~q{zf_sMVXT~_i6`lQfFk5?O
z@XjH4<GE84Ysv8FKbO5awHFmPs%&-H#~|?Hx?X5em(;QEwEa%=e2TqS-tIqm|Nbkp
zR~h}raORLbvqJ)sEAPk7e%yIH?&PdQv~H8f@sNqpXCXNzMSGg#+)YJ92$9x?n?<z)
z6f#SAZ$*YKm<Q)9OJQ3#PxjM34+CO&CLj`{yae0ITC{5nG17rYJvc_?nSW^VPed{$
zZ*1$vx+n=R<(2#Kd#<masdW+$qt=1tg)d)z;?wIV7MBZ~c@E4R9CPqq4~|;7EGzN#
z3-$sp>xF;$!yjmSV=}cvL*23VQ|0_5OX(K$H9jZT{=&<7tX%*ddWf>znvZnH;5e$Q
z4kM(<;2fW68lx^^K*)ro1ii&Gw&u{@ZU&vNOG=LVmy>-C80#Qp#3Vaant#blVVHe<
zqf=;$I2_Rs{5thz<8=2PzU+%B(F2%Q%V8SHAO+)+GixcNr4VegL?hHr8;#m4kJbn=
zNgf=8FTTb|Nr>J#weim-3!sIT!gU(MP#&?*%&q&5AqX}V&2$)=F?!{rI;~nfrxPqS
z3ATtulpmBWa4oANC6b)aN%(-I;hz?OL)OPL`#FU~>y_PQp~1mfGZp&qtOLZ6FV50A
z-^z2eV$R`@Nu#vc4_&q?!q6Oj&LQT^lT$jMx^=)^%l^p4`CL%Wt_mf_MUXh(WDYZq
zRyn#BqO%Y7x=6aXXvAVE(dJ-pT`pelPGXQuaE=c%zd9zb98F_HbK_cf{I?BN?lI~D
zSmH?sg}@{QV|`h;UN_CElLCIu!=GrRq<@-s6~R8@Ms1*rG*8xLAv$<0#K<v>5u(uF
zX2P;FgevHr(He(U+Bmo!H)RX72gh;4X7kOjsHG)GDyr0C41zau5_koUl*dL(_1!AT
z0wHB5*&6%%R~|1HLTr=>?{DuqIP30Tb7Iu)LXaZ=MIZFXgRm_G+o-Mcet&SizS3Oc
z8qHW;%@<~*x5|-}C^zO{AB`U#T}e0kekdR-M-Cf&{&#=Ndi~7Hdgb-9F+=9)gU_#@
z`Sf~aTNWh@b|Jdw41yk}#HV%P^RlqTKnc+eqJdzEC0(fP2e&G2R5Zm)K6)jTLM$tB
zS$X~Pg%KeY_&$yQ`v3l~Jih%CF<*H34}Z_^fB&EPo6o;+HF;)s8(2#8WAgoe@awm4
z*!{rnRY>xZDW4*@+fB~#`$Haxlv&HFabpS`Rg&c0FZ^%<8U1Xupw4EZqJ{}^CMeny
zX}ktwL|W}W_0T1pi;6ZZ__GoiF*28^z3wzZZ*&)QMBH4LGi%JGGyKR~_?NO0*8tbQ
z%A#44rW_;oj9`o;&Rk>Ea{HXiJ<U(Hi4H19RdkJxN)j$&T{qTkWp-6()92^=80@uc
zi5G`-uR%BDQi`<P6FQO92dQ;J$;_a9<l2>&YJDgi2o#(Wy3<=-Jm#)xJ1;&jY=T9o
zx*$@F#G{G|`6Ar+8WUR<g>fk#Dgbkr@F^vIzJ_iho6|7ja|CvF4rd22nwtVc@@mq3
z1lktskhKqn(20v^qbuhj8q1PZ1=>fYD!jH{#}oqDyD2-Nd<Hpxcot*?gwZ<tekdz@
zOxl>d-R}HYwQpg~#H`&%>rJK^Gxsp)5LSqY{=6I|fi-W>!!WFW_(WRo&%Mjy-(U76
zWhu!GV=~M=UmG_y>h<91{tod&6=fJlSAMawnLA=F>h+oKRUxG3c#5FvcBsaoj?OW5
zjzgEg?d>bGH*9u#e-JtMSJDW6P;I0RNyXE<9Booc@yk;qpt5N!0&;lwKIq56?Z;PU
zeUMV(`?qhjVdQ0H%NI&1v{Jz$kH;OM8}KgPl}c09c8P@&3x+vaQ$b37bncIbp4yl=
z>dx(P<L$>+j>nx|&%El09k4D6HrlRNUS2+FKelFGK7HYrUw-DYt==WXrxNf~ndK0<
z<VByg6#QlOSILYVq^){9Zdg6YQ*!T7y-3K3*Uz8$-7mi|=AeYg-;c)M|Cj%b|HmKy
zK=}N^fBF0W!q1l%N{ED#1dj_wZ<WVU`S$jM`{T~HAK$q@9;_i!@=D1XY2_puX;|b(
zDP{4_uM2>fGm%7VRh%M|#CxLYL!0`%>GK(xqmzcY&q=<f9({yW4^r0NYx;%Y&Aw{u
zIhZ|SOVTJ+Xgx)6&fw#~uCpTO&s<A!h1yj1{BO>ozu2XaU14;#rP0yIC-*x=wcR$R
z&*6r!uTh!46P}ksG!tQTSrp~xAD)ns8#FRAW-5*9^GhjB0`uXs{?au@ii0xc!i_1C
z#w0CSJB~RppI48mcttaHDmg`*;7=OS4I(bV^n(DO$M@mpqZ%2>XPLQhT{f01MtriU
z>dYX9s9EsP2x!hh8#1f4W2n8IrVmHf2Q1{^>Li0coozIqxdy62Tx;xxmpO&T7=9V_
zE+}3p>Tc#V5A8<E&)_u1;o!Jg^=A4GMy}!KZuz+b5{unL1oo4rjmiBuxE}}k$b`{p
za|(WEKuFTzPc83HB~l0xO5{xW3DW&CnmUA9g$ZarrpzHw*35R<*gO+BduMq%)lZ{O
z3{r!f<F6Rgvx+m9@z&9}J??%14K@!umvBj4Rmq7^E~MyWS`C;P2FIxE?ZIPz<M#ed
zhxMo`er!+@>OjYINTEF;f&tw0c=|T<J)ECyB9v8G&+TX={}9gFdAz@2zRURj;|C!x
zyyT4(S4v(eY0>w9F-LvoZpRo{mxWK)E3cPVmXJAd+FBCGCp2p1eGKiX0^A<&+~0q2
z|Nf0xcaj;~x=<F4y4Gdma(!i8HVD!48$R*!`ogD|t4FJzBv4N~AB{MTwMe<98M)u!
zF$k%Gmb`%}1pv!HG{4dJZ!G8ZtxrCG`pnCh&wTpn3!h$I7>}K{Zv4Ey@*n>GKk<M6
zzx`ji{PGjOyL`qTm3G;&>*@hkU~O`Y&ij7j*Kfb_r$7IhdNh9e`6o)*yxY@pT8cVI
zqB2-Qn#}63NdF;3j7GCDo_nAvFrh)n(LevbZ-|*(9LJuF79_)-9+ITpo<C+$U8P*g
zfibd&6b0qN#7v*+_<{^6J`or+C-3%J=?#wB{u-Bp%wgxyKBg!sUKfNaOi05BeG<l5
z6&oQ3UxN79X~><tW~pB;YhhhhVhD_s2w5)XhS8%GCDY>(tu?So?}6GTsW#;%?Al3`
zoJdRNa#`6f3zri0@F(<pj*5*AK@pwPWb;o6R?ZPuaKo0A6lX`IAzi1Wh09vFt}Cyb
zvKVB58*b@$2k)GUlN_DdrFc4O<92&+?3HHDO9!84`-sKUzl`}@{64E*Uldyi&+$sD
zmBV%BwW%9w!x0{RD&rzLGm4-E*Bp{YPgNG$IH<*qQJKv-cm}P_r-3I$V@XmdEjf|1
ztFdfyCcDj{JOiTsY>liw;x=Op^rPzmGJA&iwBw*2F4c7?-57&nlFyb!3L&EAuDv%x
zku<m@oQI{INeUkBlHBp^RZ{nz%~2bV#{nU-lvQxCIz0x`$)Z__VfWij?$iiJt3391
zj{Ti>+~{>zmtf<WJ_yMxN|)OP$H4d=<xJE|IXkCI2S#@?Yf8R1@_j`2P^)9V6HOg-
zYn_r;i9UkQ_haI+UTDXid)t*Rh&$x7#vg0UzT6a0vCkpN>wc%!s%5ghQ`^Dqaii@I
zY=2<&Aec!dltc+w))n*DG^NbT>nGM_p(MR-$;`7g^gWKD_ceq_Nr^K87xdYFW(gp1
z#r}Bk_3KxD{OMnKT{q%V$SLu=zx^Bj?r;8<&##|U0hi1cGOx?V=bwHe^*~I8{r!Qx
zOi8boLiRo1h#$g^!I0XiEUYh|*p>?ffj(1N$rjl6gLTnGTJj18BMgq-sI4)l{0C!-
z{%Y<i8m3}7=fn~>6gv?}wh%^<;;$J?SZJ|vw1Y}j*1JEEF-7vC*j$cD44p#4;=meG
z93hD1i8eBP-!KmC&bmx&d;KenkH%1?_f&=uML&#U9x*1gsf)p+1ksBUeE}S(9p>yG
zoFM<eN}6XLkTU5Ya%sTZMZZ+Wpia3;yLF<;cMyZfipxl>rU0K%GG)>7CHmr}55vMF
zW@CxTy#;}=AJia|Tfj|kXznLEAuK7dmBe)|T(^aFNi0kDrSkNSVm@c)vo&hhn8Kz}
ztJFfrp}$9;P_nuo6Eg*;rdmv#Nnm4W9Fn6JlTI0Q&dJ9r*-VBm?NuVRb%>S1;`;DJ
z?ZZuF9~Z&&>hjFRwU~%Kk`XIuOcx4F&5{#Z9==`|etNm^<+`!13l@?Ntx<Wpu)996
zC-~0%(>r3sT*lX`$2T5TSc=|hZOG9)2OAn`#VmE%*~gq6O!oIEM#(a@SoT08cYN9$
z&qel`cioQuu>^NTS3hH3J)nLoE5RncJ-FYc*16xl(jM<vt3+QYLm{Ly2`gw%HBP3~
zxrp@fkfYnviTXYru0t4--)3YSwemQQr&&-QepQ3I{1gWp{9>1oLn^zN9J>*ORmNxg
zb$<}o!C-JdZlZ%8ckSm)Ce>PZ`mw{5N^sQ9tW!`pf)_2u_aEQMmx~~nInwGuPKlNh
z%XRad2{&&I{Tg|j#2<?tSHZaqd&omFfulF>`<*}j@ej0@&wTp%GwbyeFWZGnUMQ%G
zv+S|7CGvLM$$jwl{)4-9Y>kw4k;)M*a;+3zw+rvvMvjr%DzzQd+Q>lyitY6iZ;w0m
zXykR{a@lmTq~sJ*@2o>%EdiT@FjbT(3P6W-C@NC7wCJ#pk(i1+G_w&+fk43$jDV72
z{hST{x_D8jf~Qjmkc=6WyPk^N&fO(aW*XC5)R{LA<M!To`&R$*mqNF*^3g)B4fB>=
zAG!!?cZ250qKGbDkx#mZ$`A7xM_)94TpC7eoySqxkH#=z3Cbd^EmJT%%=s}k>6+Di
ze0@t`3Gx#xNwI$x@Cv>WqcM_KC8HB_VOg`3BSc+f4xSQq$evwA!6->zZw$EZzd|T7
zZT&`7)cRnwMs3wAM!A@ez0#^iBsL&;0FdN<(N`u$2&d}9V?-VMtRN6A&Xeo$b4pJv
zwnilOG_sgMNLh$zfF;9RnO(e=I9;J)6S^;+J$2mBWiYy;NlVFUQ%mAyUHR$Lg<n3s
z@cFt?){SPs=<H2d+eVZmau}rYRJkUH*v;WN8C~u3X~Y|2$Z31-Is9@y^Bi<BoT+wa
zc#K!!Iq1N3>S^%N#2iXZv!TSY=n9}YPKhWk-PM?{*rFLf(fLY))(&bFt|H8#OEYCy
zHs$r6A_~(XF-`2KZ~`QgD%N$|*w#h+nrYbRv^fMzJRB!-$hWZ9gZJAzAw;g*l{r=6
z_TD+_!8%{CFbF~2Ngv&_8#--hBo$9Jm^<CuH1dqp*?7Ob@%_iI^y5K5a_8CPlc06L
z_f6;G)BLiV#_@RY_18bMUSIhB;~Rfl|5hDKP}<+LRnp4V2vew-Iib^Obz-COYU;FY
zx<qr>Glc4Xzw>xJgxqlF_m;DYraB12JL(ddCGa+O{@Wk_5o+aM{`4>0W@CAIVY_T(
zgwRnr8KFpJ^ySlM)@9)_;qCohMT2({-Jk<sUO(}-KmUwkJr5cBdLyUA=a)|yljHH=
z7{S$<JhKcenD26MGi6q^F9DM1u60&=8(>l=O)tI&8V}D9i2C33J_&uWU02QXgA8x;
z?1pX9tv?HIW+3!LyARs?UtxUoZ1?7LNS$^E7<9suJrkp<%p9QjLYF9W%4m}?wA>0I
z5=&-8BSejrdN+<f*k^mP(yhRZ1M9(6D&fOHjVir3Wen?6K)MkVam|#i=mVXg%*@`E
zqZ(51eoNKV$-o&y*$=`txJ;~fa){(8NZgXL(_jOAfJIf`)24s8H6Oq4g5IT|CB5}Y
zuVO#>beUrewh1w2LPh~0#g{fHkR=C4BJ@FRGPtCYDXB2+Jjfv=qb#C1mNiqhg=JmH
z1UN@#LHP^Qj9|(A3ob`R65Np0ef6Nx$FM{j{^w*NCthAxe);l=-+g)Ea=D1@Ka6$+
zVwo-h?wFm<VSk^3QGynm=0>*|gA%}+E(z_@wY!QLN>=oz41Ov)v?-ErLLCA`M_t4i
zh}j6<uE&t{=i8)FeTnC4v8Yi-FqZ?KMMpd6b!V(EstSV~>iwvUCekGsF-NXt6Uw3m
zL_Jb=)@X(`$TP8qObCswD6%xB>;^?iaF05@6FDT+9za)4L~C8{bTdb0t01~RZj{TF
zy-l`Nxe-TyaJ%2AO&9ADBc%;`t<?4)Etw@=BrQH3)ccKkzth_8Qd%QSVdTS{8ga!4
zIWp6vjYjB$WPx5A?{DAv_1Aylmmk09*RNl>ZdaWFeZgww#Y7#LOQfS`yAO?CrcDZ&
zHFrkx%=OmTAC2$#8~gqsn`u9hBO~CWUODyKY~)nvN8{J`JKz8Gjr*U!(Pw9gnU}PZ
zqok;@Kre$7BC92a8ReE)%7xd@UwHfaja=kwSW2cFtl!@l7Wj{U_xJqn@`aK!U*Es-
z=l44$1YR#IrDQ$pBf>Fx%tHtMl>B@#2qdsXVRh7fe(pN8n1=m$FvetM)x@NC;u7e?
znC9~EV8k`D_W*SA5(wER2QDLKY80Jfeuiqz(KsDJfALGf%v@=w11-h$)Z7m<m~vL0
zD|6us6bl(r5|TQps+;CXR;L;v2D&fg_SR@a8OP@0-w-r2j#+{1HdW%<kUArU0AY|u
z^bv*mSOky4b5n-{+NMqADT9`*K?k$IYQa;L`~dPL??*b9AHuFvpIu`Z^NUHGA`|e-
zWoENZM(y<4X^)C|I6MeCr;sA&rJu$oS8l~1LTVdouazpN(On7}Aw-NVXYxh9gLNrd
z;GYs<?+8vwYD!wJhNNoaBzPv5e}3e;>T)_!6)`5(Qn+3&yu7@KTfZ$BLCTQ!q1=k+
z?!(7YqGKeVG5^m$f=h8v(_0SmIFuZ?EQO`W45(~leBo_RKSDZ_!+dm<vi?29=<1r_
z<Xm5F`}dNZsRRm1QO_px*RkJOav{=%g|<#@2V-;%YSIaQ)cbMfMSvEoDMecN81F@w
z%ZhBN*I<q=MN`#$x>p6>^g8KBWj`Jm=5l8owJqLFtDWQVU|-*ec9EZ<Y7BDJMh=1G
zhR6H<Al}~yrBId)qIZCXWX&ToyT(hWVNPOZ^JiJw`R;*CAqF1D&i-{Ua^&TYKe2qe
z@bgb!c;OTG(D+1$*U#h_88I*;e6J6_y}vP9<LcuZk~@GA@bG?nu-|W3?_8Iaxn0>V
z7eWz}DlcM^9j!uC658$d#*gpcn5_{Fe){r7`6APIOesnX;sT8&2VP${US6(Ph?MQZ
z&%gYhm+cBR`Q-bN6hC=EFvynJE>}K({>*WE)8&*3DQMI?hH<+;=s}L+HYcOWI}%I<
zbMnu?ne7n_W>Zd!mK9T<A0NjXio>N3H(Lf`=nx@WB=kv|iM)skVJE}u`I-hL?mZo*
z|LY(BMoH=<1z{vGqmNFAiq1SKf#(d<n6kAeXUc?tp`TBLadb%|r|%z-)T`N~dL(2l
zh2nE>^Miwl6h6ISZIBFRQupB48#h@^pG|9~*{4iiy40M>1HK^jgWN|nL@kI;iCj!t
zl*SaEqa4ki2Wbv(nceh;;D1X_SXJlHd#CM{`l$515{8j|Z*k@#=%0nMEaJPLFq5fq
zQ15!ZLO^^ax%kLAvo4uhGi_O9c@Ma5Kf3bDKY;iw+5$0Tat>4kUdt4bV7=S(+0_TK
z!VkPN{YkRDg|aL}PvblB24K%rJH0`^Ob_9yHVHxSJr{Pk@la0962;|-Zq<L;7H1bl
zEIRHm0#U}qkdsVuN&hXRGwM+7xCIpg8Yww`gz$XL#Te?o+mWeq%tkk8DG=zyzA(+$
z_lKT?=sNHyN4=V6-wn0yyBhOO&YrumBnzwze7@TN6T8USM^mRWZRmoCn0I)(_)?Da
zkqAhtT90?kDmLKQci&}zrRcA(RDqA<;L@9Z|EbTyx?Tuj@-{oOR)~o{1&WM0`|G7e
ze>%IAb8&tQ_IrbGzw(EF`fubic-()d8T{)P{?mW_FMQr!sHkJ6B#w6Q{r!#G+Yi>1
zxPSVIPcN^Otegm&#^bSbe|zV+J@kx+AW_Q8C(7kTOf_Vbw7Szqr#=q!X;%S|J!nhO
zt~jklY_o7ueoI!acv<=K(=Yt;%ipjVJl=1NF}Yl>-0wGz+Gug2!Pk(QGKgRrV}hV!
zW7|3}*H6SyDCsoZ8ugG#@)Gp^FS)=JrTKQdaXfZ?eKB~<t)2m!r{KgR3K02}IA9Ny
znTctVCrnG=QuLH%@(V774V(`8f8DL{Z@UysZ7c9}K-wJcMes^=x{<>)CTJ64mt$`d
z{18?fyrjfsDO}ctk`k><>`)lbOc6ZPLYi~`h9+22i{4vz6PlC_s)miD6!zm_-wUf}
zIUlWZG#SQ5x92SS)aHYh-o!vDftSmw<#itdsQJJn1Zl^Qdhj@Qkq-H&bW%LSICu{A
zXIJ}v9IE1i;Y8QKWfd%JSyq;HVcRwxR9>C5+NiaAFpro!_?$h3q9uD0a-w@aYfb^1
zN&XimeLV3&df3sSYZd|7=!t<i<8v{1&gg?HD+vDQcv_kC<<T#W_&fx=<0De?-^Vc*
z&-m7<4}My*!vYmtq?9}<y)ycGQiL@py++rSWm^~+hpA<b*|RNt40IYNKe`;h3HdPX
zVG=y(!1j(IofW*mVFAC4d!rsV#vG(v+yT5}2J5ntl6DtzkVY@#14lR*C-^ZYjpV%R
z49^Q;5NCc$mRl3-u;i?fL7YD2^8HwG<K5`g>ES7@VZ*5VBEf<GKDc0mE^$Z*=b@nY
zFQ5VsgPaq+XEDl5&uI+q3t2Lk?IMH%fwm2DewW0!58{yYmZMXD{NVBDmG6CrocZbZ
ze@Ff4Z+N`Fb9u?EDd}uDI;nQ-sJyr8&&AGlyRa@Rqc?th|HdEx^hdsZ|4Mz_NhVzW
z>&ulHqf!gkjSw&D&if>n%w@Y0On#*M?XChx<QOUM%s<0Zbm48?c)4EX#-4E5HeQN$
z9A6)AEZfHW?Z&dK<W=c-NA1t9uBHqjOBH8<vJ9@5Ppo;PwMr?O+v6rkXsyg~RhD-5
z-Ry*RbRLh&adcnEOH|=3Xcaf!9X~blQ6yUGpp>Y@OFHKJNs#=0h@HY{+5zYMuRHGl
z&5mYlJk>H@da8Pjft+;^;j+WSrIo`!e8YUxgt09}H+4*0N+B<a>WdkR0Sie7ar}6&
z@c&QHOt81^2nJQ;x;WGx_BPq~MhRNb9Iblx?sP48gz1FL5Qt$Cv&Ig~l6k#cNJ%g5
z{ZTnQpyDG(xGS>JFNGLobg^LIXz{b!2A*LrtqppUxh3Vyl7*jMa#l2BDL{A{xN2?G
z(cRQ$K8F{}C-~2+Rm=JNap=Gbg)jv2U~0Lq%Rq->2!RsC2O9o9N<ud%p?D>vxA!EQ
zE~TKO!JbRpvtl|Ep1T=sXulG(E*WB=Tk{dyhmET{3{jsQRWT_>xxg`qp7C80Ws%%?
z-BvARTW9!(Jg20KA)nnrc*=Z%r@H7IXHH+h43`Z$22kHm5XL|x*+Cmp%!JrTMOEzH
z4pPi4DHFru%i~px{Op;?F*0kT<(*g(p?bIC%k8sEHUA8snbc3OPLSfOcP09zsc@@c
zBHT7JTGiq;2F?Fd%1``;(&ZX$8YN{$Yb>rB6h8aVP7kkJrqpbwatU}MLQ+mx*Xy%x
zE89g`<lWR;soOS>ZLQe;pllaH?>ugAeE<H9B`xF}ur}$pozw?81dcg)+<$;g9`|=e
zYxc(N_QoIn{6`-58w-(_Tzu~XZLipMkV0ZxR&N-iqA91U_z(iO$BpmrKe+EZbu{{1
zpJ%F}OD0EshIAoqT#A0qcDeHD^B4Z~{+&<~>&u1DUw(oyd9M%d`-9`C%EL$ta}3TD
z!e!kECDUqTSZA30O=qZ8ikkf&hez`EPCc5E>=rMA-MDw|9a4;hWaOazB~EQ%ijq5R
zYL68P!70=FwdXmf0&gjQg-fA!adl3{`0x)}Bv4B7z!rQ{P-Hsm21wD-dnb%Rw7_-S
z__SS|TDbc1ta1!Vt4xSYa#qA+jHk|Bv)?o`F=a;R?heK!L-Wa-IgW$KfFz1#Z)bS>
zM<{lTo~9Hv(=Iu(mdv(hmXh5`yYS<-Q;+5+Wq78BohHp`8n1-tz&o#WXNtZAvZuN1
z=nb30Q_)nJnoW|Ozt?=`801n&CDU`J?gzD1Ek}zS#I1E6`@!w8b2Je?Cz7VxIN3yc
zJtD5sAz~gBGJa!o{H-%OMa3&d1Iw8u7Y3?UbCLPxsaX$^>I$NMGN#VotDl$YbQvY0
zzxs)!CdFs$L0*^VxX>5Pu|&1SNvIQ|vJgT#*+$6~Sf{&N?<wJ=Qfg6b1gW`xi&)1m
zD{n03wC9701ybl*4mv2W_eNNQ&-JrsG|TN8j~#DC*q8g$KVZ+x;PiWFWNM>(Vy$*S
zacI`Om6fm*){;bYJq}k&oxCmwJGSN?ksxIc#+kW2J@BQEL9LC_+7pU6ZMp&mYOng9
zg5Y<(cYY&sety;gLWs{f^&l`(^vvR<g>-<~8Y4y?w>Q3i`!mb-!uIl_k>N1velTjM
z9tVH^{wvg-*XxUqVj{IQ-rs-lPyhS}9`^@7zkFu>^aTR?jCH*}+ol3)ag6@!=U4ve
zkAHCIzTU(8<Hq;5@4Uah^17{TqmYM$F6LadoHKv()6bMS*;1jDg}1lw<dQgA<+5Ew
zA-><;t0d0Y{dQ;6#(HhMe)+`f^_5F{VJ$1k2CW`?6@zJ_(7Vn*S8kp=n&uMLF|~A0
zE~PdmM^jF^g(d}C*8a*2H4MOPQgx;fm|;l7GBmc-xJ-2B@UP5ghJWk$sGAbZ{Llj5
zX$NDNs=O197XrRx&;r?-lCT(qCBl{>pVmcq*OFznn1MRsHab<^HzSDS5u!<6Xd3A>
z<D*jX2RsFVtC)jegQMXL@iUiq5M!W}NJ=<rK_nfE3Y^K_a;6kmvkvndBPB(e^I39?
zQrGlOUKUEs9IY`pm%aj=#VnFz=fGwVvb#=>rpsj|`1qh}Twq3T0<e`_d?Bg@`bd#h
z60HddoN{m$k7E$JMl*6(%HI)ONg~6b6G9MYy}QO)OKaJA1Ck%coU|LklmCMD9&-*7
zi8Zgpm>EQH2WGb*ZVg!n_VcivloS|4ZF6s(?6dAObNjpmW|!2sn$gVILZYOVwQQtV
z{0ddo@8`U7!IL`CGu=%WjOP5^{^w3Egh@yVa)9O%OFulLxf6AC_Wi~jMoNpQw{yW9
zh?jyr(c;UZuP+i97MUT!s(_?>iE81!bG8z(40#c^3*@|R1Z)V57)UHxuijh|p;z`H
z)!Kb@LXS*S%;y}#skNOk#oIFfTs{p{Q*pPscV2v^uCa<=B(2fLP9KNHM=2797b`(-
z-QIoIHxKG;PjQHuo}ncd`GxvqM(<1??|lFIXVS8=ZdVy0yYbjvemMs9?K}JXcWi%Q
zU4@qzebC?D2*=L1KmCER-wD>)X_WRxSYP<hQxxH}Mn0#)aas6<-?9C>H-2=37}=JE
zPvydz7VU<xCuul3T=HHI_WOfnh;U$8)Z94-Z{NQ$9y_n=M*n{2$9tEuFAVaW^y3hw
zIz-C#!b^x;FB>e{50zzA)GaE=?)SUGZcKuf7&0ksj44(WNF_J>B%UFDr_}!}ZcYJ=
zCkNW#xkIs@1iQ5Ul*Z=}nX^+;`50&ar7s2ZDhO|}F>lXZNfCk{Vn83x4KQ<-$5eR8
zm<%*eFQt&nqITQNDXYp}K8}hR^in9ZkSr6Ys^I3SWy6=RXR=su#tNpHX~Z~YQ<du5
z{^U@#Sjsc-WvWE<t$li`lt|GL;f}9O`t>DcQb^1`uzZqZrvumgW}NDvPW9CDYnbyH
z`#H)&PDRY1DJIXF6WJHgQWiymMS4iA!Kf{>*F$QdHaYgnQRRo|qv-?Whcoz4AVy+x
z?Xz=`f+c?|!&8-TA`u9}=$Usw;<u=c%^_*Y?eFXK&j3N+t2!th!eHc?=mWwO!S#&(
zG#^#;p^Ku{PVHUJ?A~2j<XCLXjeM%Iq6Zf#S2_|ya)+!haZR69oA5Ls_~-WQY?D{L
zGNyG}JxEzI{xcOXh>4(!Hb#Z+<V+-Z2c~_%d8x&eiBlQmrDzm++PtPq;bQt=Dy8IU
zb*|EjDSDcnc5gEnwPach>|F*jNqe2rI*pVQbIj-SAdE`&kQkHy#(qocRdt$?a!wsb
zRrin-E(03f^*-3q4yBFCUU%-hd;<ZQ=|a%hID``;Vp7c<js5+N_vO$0Sg#Z&ITq?@
z?Dfu%AAjb4`%XI^l%zcB1hmINdpsEXgE%@re*F{u{m%8}%Jrwu{QUR-7b=RoU6Z<n
z5CgwlUinWy{hsv83+b}*H=lnNbTeq*IfsI+Vvs6uubp3S@7x}D?)#l<UiGwuMDLUR
zsOq|haqI^pcCO*Vw652BI9qBMITl_nuY9?_5)5kH*>B%{EUoceZ;gG|VmbQrRqIac
z^*me&-lg^yDT{V6gGmnM*#*axd`IF=W|uSz@)@5--2lW>T!Sz|<%!n(3tkF%+j9Q9
z#xQ4f=6`eUZRP5$qhZl85d<H%*i)`LgE9?zn>;)@Y~L&Gs88iiZ;jjvsU!@Zv6#_E
zIn!I;8031jc+SuS{lZ)Jv+dG6JUi78S7J=ZMnG9sy4&AD`u)jp^@tqU*RoP_A_pC=
z>D=;Ll;CEH;bLBWEh#GpIe8)*Johp-&jVXAhA~UgvLtEq!!TML97EX<-5>1HJNxbj
z>FDGXc)4y|w}q6lj8kq#vS^GrNuE=TOI=Ug_<5bDoE`S1#jm+U&X}B;kM2+8#2||)
z6LcY`K=5ql^R-nPm`pq&$LE1&{+=9Fo{rnz*-weFOJlX5MU}1QUnKONM+doKs<cUg
z$j1RS%}AUsXr6OqNiAh48hy}d>>-eg2(a~NjOqE}Nz&G3DI$@&f$n7C*W;jVP1SOY
zn(21Q0!xmhEMs7IXj}1OMAI$jYx%Av1|w&M7zKlAIvh&T(!TcTmyo2e(HF4EFTE2Z
zGKPc=ZBC9_s9n|LbXu?3U=DQ~tsQ=0O%I7t2vQ0`9Y__}@Yr|mk2|#;{tRjE9dcqR
zLdgd`ED$EMR(g2ncuRa+5@YP7xG+ZNet+ldw}0Vr`@vc=SZ6dL8QO6WMx!u!Su#I<
z+!=GS&q@f5^!kNw#~XhW{*EP!!4xKxG5Kk^u>Jlg^1ASn3mM20L4aFl_s>O6lOK<r
zKYjhiKmGAvuzO_|jb&$mN8JUv1NME_GaNF<epkfj?WR2E6v!!a-L72M7cS-EMG?He
zedTt)E77i1@3?dU55EXfR5D#oiTzL>{Td4p=|f>$DXI`9M#$>wLx|7#)o8*0j&4@d
zvzN?~-{vFS#2M-Kzx?rU%v1Qd6V8YA<A++BlL;vVu2R*e@x%<oDRs=8r#4>LirPB2
z{ow8P;QlyhP0O2L=JZmSedr)E9dyk3j0M$9MTcw*>IRl=qeP8JhG`);`XIXXJb0iz
zC=5gS$K8Qt`npea_C%_xoDxfntU0sfg%A@x49WM1Y$dX0p`z>LX~Ifb;TWgP^qdJw
z(>__3%<IdA?Ye4buoOvpgN#_sRT_`uV6WY~4(KMgb^x}^#+TO@E^Cn~#w(c~H2NE1
zFk>LPPC5i%B!|SwSt>b`N^~u~V=&B~D293JeqHS|{r^jKq~Q6Gq?2Z-Xy~IqLv278
zee{JoKC)>p$vlJ43*p%|6Lpy8T%-;f3NtyEia98qF2$@1qMw8Oh-KkcLQEQS_TeKl
zzl?{e0BD^tjr}+{jz$Q2V3(4}sj#ftsaW$J)|~Xw*!R0Kd$&y;fJs_ChlrH{jXOsl
zl%jFKm|}xjvXeuo&!QJ0IVXYz`U#Trj-q#ay8J^7t}fG3IVBkqha13pYjRAt<~h->
zdV4;$Q}YKW0SZ((blAt}K}iY*s;#*+b0=V|B|*u|G)X0~u91@bH4R=uj7F%F;|4!y
z)Z?LyYh=`_$2%!bsfA{z_vn6)sgIVfJ`%8cCqZl6ethHG<6SOaqS!xEXq1jvUpYpW
zi?X2CB~!e!?f(3m89#1!zJLG0KmOzY;PY}JUtYP+&O9bIJ4?xo-uU|U&*X0OoJl$I
z<NH_4qjN)0p3K=j_Ay|JVO9GOy?H=w65N0;0_CB@%VpzzDU_VqgO5{lA*M|2iVC(6
z{cJK49~sZj!`JXZFCquj6V7L%6&OT17S$2|YkvG2$4BSaq6bxlx*M!~Y;bnNHhj$F
ztxcHIZ)t78V^nP5qgH0t3)=eRXj+VorkQQ*lBZ5HPBNUKxt;YPWbH6!90WUEviTWi
zy{?(OWNH_&HDulJB<B?caoVd_fEJJeD>g|hOQv^KXE3DykLfu+lMZ@l374Rx;L0HA
zEeLnv$nNuyE3TnlOJXgN>$>pi^}_4xN-kOcrw|-NE&1YoKlt%@aJwJ0F{E-rTJ13>
zB^O?|l}}f3T8M#qG+OOg$t2U#YNY9Jq87KYWI_<<J|GUwnPq(bIIEm<{G$LRRlsd@
z=u#aG|L^C>M%;MOR243SQ%%Ej1UZJe%ulH7Q)92LQf9?e#4t)JQ~Wgv0r@17-j5Uo
zEbC2Qhel4*XS0*?BysAZ7eQB8l?d2Ulwh|v?L8LxYnD=^5-WvaFlQIW!X~X%EE?$^
z8Ka7e!(<Gl;ho88=jd>{h+vFCOnUt^<g}-BHu*&zh%#n6cB8f~fr6oqGr9*MXYqo<
zpw0e-wDqBkW1*DA!^$Z>UEvdT4(1E+Gq?LxG0pHX_G-0Kj|c78<+hFvsf<!b6~|w@
z_|JC^&h(Fl&Bm<n^p?0cNxA3LB6R3O!SEfSnS%jABRdy+sDoCv`=cKaWTc$|EEqMx
zKfisY{rNkWm|+ohSZ7UqiWx>5)cxSNJ=ovx{QB#kxu?YPcYn`bAD(snV2;Y$_pki%
zpZ*7_D>UeG-Kg~-=isk>A!Ox^_o>}Ww@#-M%EHT+FIa%%Pk-idbW(QeZ_4r#<)I=h
zYb@U9<QR={OlogTQ}%|Q^mEz&QHci&iUcKSpX-Paf)}_$<kW1EyZ%)&n$P3J=QbKQ
z5XmJyL%DU>s{-ksjch}K?>P!@ec3kFMGsqcB}kuQeHf>Pce?8t(v;k54vzp0C&bJv
z!g!8<Oo8hp!NrCoT`^@nAVCp}W@kSmc~~2gv&ll?<kTsC(IoosC7w?T(hSWu>9P5u
zRh6FRm8U|(fmT7_ukMTPFb5?i=4@JE=fb*V)*`dkoPuphlGxez%A+3KA3{`j$3UFb
zlyCvCtqYfRC6~mqCSvwVe99<iQCjIRd{1D(6+0isv(vADSCb*0yAbhotWO=#pqckj
zRZ4*7vdW1n%rP{2>a*)m>-xS*&V=F~7X)8Zb$;mQoO#3eZlg$uAj4E3QCJsU>PlNn
zet_!G9bMr*;BlQyLX-nE1uf#IE9d$YcPSCDd77P!7?g8xI;Bg=lqD0#q;_8dTZaL$
zLv1@_G{GN>K#wU!H*d;BWa#iZMY%-BDWAhu`{=%>@5rWi$w%!xYGbb+-53JvWqW!L
zNP!uQW`koiUC=|0{S+gYWz%OPL{1=J?TvkZa2yX>-JfAbGY)s?;v?eu9G6X})@liD
z*z7o|(qHG`mqYkyCZnE>-gfy}Mkk2dE@yZM^cZNpavT~-?MLJNapUoL(0ilK$t5SQ
z+eTbZdhMhhhc1f&_tE%&{-6IZ{`|lHS6*u6R(C>*>OgCy=lmp`&M_z<aM>0<zg)S$
zz0umv_aEP|+8J$Uzu!6TH`-p^|5F*`V7qw6vkkWEh2^r5Q<fo@L7R<klkIxtr%$gu
z?hk6~{KG%~134$wwUCRlM8a|SK2?QR_2>TdRzz2G#uQv89X=N2j>6YH>_$pchhR3D
zU5Jce`bi17q5tBSLdw6D^Ql@&F|uA(N|IX@gD^xC4Hjtrfj5K&&cG!lvV-MvjIwas
z<d}^*npgkg(1%iV5%|M1ef(l$x<v03AGtrlj3Z2P9h7ad1Sr0=%xjWcb@&Y%<jS7m
z`KS_2FMOUlC~Tgs&-B|?UWb{?DWsBM#%QE?`4~M76x|mTvw76&9fNSVVV+EMjRxja
z)`HO?W~z1G?hoE?H*SxEAMZQej1nVTmJIKl$Amb2Bo@h+OuA%ZDpDWSP86OJ&wSAX
z>r&1XG+IlZrK=C*@X=C`DJBez>E^}rAk#rG6o;o6&mc)eaNWDcN~h|>51FZjT5^Nm
zw8@$Ke1Z-&;+XUxIkDd3;7e_Bs$$eyd5wu^;#sMnn9q(+OY}jQ5uEPX`<d>hmVP4i
z^BA3k#+xa4rxAU5;)}gu1OJHAXD4+PIkOL9K3B>xqmS#lb_r)dPjdom2AI=D1R8;6
zM|2P(V+3H(hd_jVs^jWYmqwq1i+dVM&g8Ag3^&5lCt*aYO<r@PEN6VFE`Mb=C`qnc
zSL|r3<dTTCJU=MsNNG%Uc_AfY?96DSB`}&u!o$5qGa1ojn2IG~$?@}kdi6QvwK3l_
z%?U9Pdj%imjsbOV9Ea3Ur(z~18Srv6KD~T`WAfwMSK1LMz3}VXJO7_Q{&(irf1xvJ
z_nrNy^xD`TJ0V8aMF%Rrzw3MBrEHKY_5N^PTk*%;6#CSLe*`6-ID?R`%Z1mUe&O|}
z-|^}7C)UeVh><pY8jy&?SW@EO|NFn?$M;{EqhSR0Rw*e7UDF$3h_oQW@`R}Zh&ky&
zon6X~AvrtAQ=DL^>qbs!5fJpbQ?jEmllD@^U!hYdi$GrBqw(n~nRG7m&UPKwsYi{C
ze32^WuhNCqZKbRQVqgHZ&nNurROf_qzv4?|;;AEdYNMNy&h55ENl@pMMmS}oY*|^#
zrcR{$Ts&AHnYjx!F-*4dY}4Z&{iX4}MKGSKk2*S5O+i92(z|4rdC5%AW)Lbm%c<&f
z=HB~rQHf3ZtnA0m{a9J{$}EYP3ey_5+IZU^-1dX_`-6KuFal}OY<*c0a~iSXy!c4I
zD7!kjLLtQ<7CY(;QqXIyq`c^wXVdaFC7)Gm2LoqEF;#hXjmU5$g~+aYXvcKU%joVS
zk=mm*g$!v{K70(oM|(T9+z(IGiTe!B4>>%?k})YPCr)`vPP3Vvj%$^*&*ZupfIi44
zcNl%0v)AXJJGwh_i`?oyw2{~4RL<F;H=YFAIn=?%oQM&o4MZ1hf+KbnP5b_!)*Vb2
z;Q95Pf{ZZ+t?ti`C>JTHiaWvKW2d9{Pk}%&dj@zCBD^e%Mveo`<<EQufB&MZ^3G?e
zxUXbr{^zgPr1`8p2UX(E5mAvByx>cjnfmj9??Z*aigrhaxd%c(ww#4)PwpY9ZRe=E
z7>18K0!Hh8J^OUwsJqltT^$9*p>eGeZa4P#AH;ueJRS$N8mS%p`1VKAUWpw+Cu?Qj
z-?+WMsX*|KJ%vE`9m(VIP(aYD0>5%u$RSc2JdT|{8)lu>CnaAfF+aTu%d+wL(-+p|
zLJo-~FXW`>9Sf{6fk`%8vPMsjo4hWgccxQrANx+5oimm0<c+ylj@br13NLihs4WD7
zN9+z?D1(a6K3x{m^}fzA*g^nK`Tbw`Qg|*me2)P9rpgXpo#`Rf()RqJZ&YJb&#BFF
zO9r3McYOcqUI0@!|AdHz=(uIPdi|IjNj-!g(q2)v-g$rDAp{<KBQFI@Lfe|l@75sD
zTpv<(BzJI@l32GQ<A+UW=ct9-tCR!Fny?r=C)K#Fg=#Anrs$krEyr-WJ$+0ybQs2<
z8~qrKd+(Ii8OK2=lcP`G9}jvTgp?`k#>M4rTgj}O;v9uG_C=vDdM)C?8-bLKuw}Xf
z0Mk;)6GlBzgg)OtM;QLzs`$`2?oliIQK_x5loi$6XQyC1^=9V6fgy9D`PVdt@`|;X
z6CUF<a%t|Vq`qisfd}I$P>4}`m~=*L;z9<m1?HS|`?w&~b+{<uWDDWr#i>x2KAIMC
zE}fJ_IglK;ekQb8m%1omSb=OG7>P+>%O03%nwje}7--s1upc|MN_7&@9^mv+JP{mI
z0)&{94V?V&E+>!4G-}lSCcJhT;W<}+j{7)^G=CrChdX)np#(rrJe+Qtv?fVyt6fXc
zHZ`85N&g=FBn>CRCiwg5`;FRYb5ep1Y+`V=*#Ga5VSafEpFe0-yCknEp5le<4|AWQ
z57<1FH+pW{=!|~w=YReOzW@9G#LLgWXNJh}xbx%Je_=Tq!PE5W<Iem0H;#HI9T`%f
zdYtgAjrX@VYOjP8xoj6M%gW_?rPZCc`x}4$`X|1>eW&(=C1#gTSMn%W&<Z@JmFwk|
zpT7LUmoGnI5w6#Zp4(>Fv12y)e*b|1>nfw;c3BAnzQ4Ws`=in7Ie`(M?R+L9(UQhR
zjgP5GGS>)9|BnV!%&jpbKIm%W>BA`S0rUJ{{P;Is3f|2<Ph+p#hY^!TFM4}@wxwV3
zEan4#X*jegM78{FOpc+{GO09uu6=&M&T274h^oA1H0AoDpJz{3vs&cQjUD~_-Uek|
zNQ)MM63zsyBz-Xb1zWQg(=V5W%eDc=ap<<UZq!yG2=m+?T~+mx<zFZoi5Z@EFx5r}
zQzd-PGKWodS6Kl+q)0c|`(PZ6dfRD3q|p@QnJCu_*TunPemQ<f(lkz!+^y5+U?G#T
z7PDaq>=CLX!L7Te!u92(<lKEn9YiI_O{Z)2uH>dZ1~CpoES#~I5$$-A1tuqf7GFe9
z&GC~mxGgvV)+wQ)lMIgx)nSfl1{%<>J6S#YpbrPTogYG9+^5W(O1(PGNsg5f(>@r{
zm3R{!EPeFnL#Qsx)4>7|edOUOVbSHy<JrCHq1E@O+4{L8ws|gZ{k`_tK6V-))AO*_
z61BK8N|#YgS_X^!YQCe;?#&P?McQ~379pN9WP4tMV@TLAhQi=$Z`9hnOYQVtnPJd+
zWprV>nWyG`(mikZgx{S7-Sp0&PY=PeDe6G{O)>aXi{Znm!LTTSgYU7-lnOYS3NzyA
zq}LwBM(0c|Qz%l9D$Ry=Q7Q2D>p$}k|NMV~$sxRdd*^ol&aeOQkF@(vPKkPZW8Z)9
ze*0BO`T!{v%qG27-tW2;rcEws<@)l%%j*lbo1%V)X^D?194p6!tw|%9vJlple7SOY
z{mkbtUnr%}dt+TzN;XDl7?ZVM*lS~NjhAI%DVdTNNT=$qf|+xVf-BbI^SqS&G(DHf
z5U_D7JoP#$iP<j^FJ7y=q$81I49X%>^}puFzm0t~=Mx_gPYLIo9M+g)dfc5xI0B)4
zAe&PVXJp$HnR&Lw1L2@=`|gWW<>8)+CVVtivO-#B?=mk1M_hQ|hiYSUBO4HPyPmiL
zp)th2PvsmtIEFf^^I4b7r<aY(rYKK=teUs_?Mbveue%f?OU#rNUapHOc)#eF_-tC$
zXgG_*DTYbOS-Y5}FakVk=NN;dcVZF-<Wh=9VkTxz+Nb!1EA>VohfKGDbw_hM+E@!%
z{nh*6$Oofo89AmnPZl*dPb4><d4=t8wA$MLm$rY4wXNII!=SJA+Zc1MwfBEHw>nkj
zvRy7a5wUQrSb&6(5&<>|LW%+k0wE+MB!rNV-~kB>9^fSsFF`^AB}jP)4?z$jgKPqJ
z?8G6s?1~i|;%k+wT;;l)%YWOKwdNS3Us~hgYkkc1pE`D(AJv&hyXu_1*JaKz`sls2
zuYDKPTG3r?F2Y=cr4+0=W62pIM-Jku+f{_X&IL}N3lTA~Vxc&Whd#-_#LJt};BXQd
zNZTV!b{<l5!}EyC;ltPnz`CG%7V>F)2O$P7vwHSQY0#r-z>VVkfR@1bgm{M<QfsV|
zeU>z5I5BWGWl0c2I`uNc)J2=zB*z#h&35#`v^i74K9>b~aY)Knl{+PViD8oRjG1MH
zI^1ppUhtR<0Qu~w84F7Ve@YB8VtCKy^w~oDdF6bIJb3<L(?J?4hxF_M#Ti@hN>&X{
z(+27gq%ls2>(rHmvkW9X62!Dgn;pJMQ-@J<A_`zsLyJT-Q_nFGV(8{!Kl_gQ#i#iG
zZ~h8&RIG7D+i&=d-}(*w#;^V=>QNEa1vUm+z2W`s3&av#LzSVuq1Fvg+YRsPfs_(n
zfAomQ$1ASaD?*I;=C{6$+wqM3cEi&rUtrsw(Gw5U5DV7p6%QYO4G&-YIzIfy*YWD%
zLyY4<Yn6UtLKc`F2bvB%Z5y8VidXB!-HW4}EJmnt#seQL&)}@!5G$M|CmrY!h$?QT
zNzIHYV3F&a1R14!EI3nZqa$bUe$Qv)ckxirHauKjNY8WrI+Pn{>fXk?-S7PO86aL#
zq7mM@Fn~QLs0waJ#ckVh+cwndRUC-p2B$EPIthpoUO>lLnWz}NTsd*+iL0ACjrO64
zEn>MY%<hFt-lq;W@M15e;CfkbSr_`!=Z4#MV5pLwD8%NM3r~F$U|k|E%W|qK<pq?+
zEMd+`SjCF_DJ7J3MOqiYSrRh1ASQJ%$A~p1((Z=4pn3;lgM&G3>K#)jJEZB^`2>%y
z48{HTt$DRKkjx{Kd<JkTQacBXrljsY-!|;G9SQz^a>gYWT$X}$DF`v3hJd|uk{n#S
z;QORkK6d(7K?v<R!3=)RascmeNDO4OP3qpns?{m1WAG4!2$Emp=r-Xzcp`_@ehrdW
z;3j15KwP0+eAr&%X`E+Zmofal#ULolLfWoZuj0KgKcn~zO^}cP?G6X9+iJC=9}W9{
zxZn>cmyFBf%7VZ<G7CUR?jJS_3$RkmDN!hyqF0Cb()Cc!MmVc}^ZX0r=1%*RitR?N
zKv{|xU3BEkDPpKofNfHz;Tx5>#wMNx6suIe!=NnS#7QW;81p{3>#olx$hYS;3&{_J
zlCi85aE(4V*!7}OIncnekjr0k#(vvy>tEoL^lPZ~z?v3pPw(-IfA5#^#V4P#s8WV&
z1sjMMpt(QAo_#-1w;P`KXMF$d=lJMr-$Yp#tm}%)Wx@5~5h*7e^@cZ}d`kGs%*Bco
z%i|TVzy39R^V{FXx4!vJeEiYZ@UUEP+#+gkfX@zZw`cVIz+N{d;lY^F7ZOrT5Hmbp
zu81+9Rtj}<q3;o^{o%A}JxLS6i*nA=^n@YlX~p|J{O_aF-f6|nITB}EzsrXL2E**e
z9Eivrzf`A`BHl6IF?c~E1h9c63LZ+vnlp0Bh_R4$Gdk)Z|9qZhhDl~|8-&bOI;6;z
z<zlM<b^-AAROS#yxXA;61q0ymh{K&x-PjAp%_`*N4`jre5=w}$NKP@UBSA33nljds
zak(ximj!A@oIfw8FsY9LQJ{C6Zk9S4A5<5>P0Q!#-H;00iIfA@F$YFIC*|hp{!(VA
z;EeMLS6WMi`Gchf1h`jI;_rh@n8kH|Mp|JXgT7KBoYl4sVBb5oz2ew*TysG*uuE7o
zR(GyM5Qq++2awu3b27R!lDOC3X9I=_>5vs=+~qJi^(M4g2>D+9&S=h74Q(`j@BCQ;
zn9X~hoRYx}EIKbg1g^^*{2<5y!~FfHNc#zAQmGLxtM2St*7Zx3ut8`R=;R6xViR(V
z=n}E-v{dJy&u|<Kdv93c1@T(2zP{q}u;Q{_(c8e@Syb`8#<NqH8#-W!qDT>FT~H@$
zr3<7KbPwmL`;OtXRTc#aVG1BYLD{iyfN=OZv_pjRF}38sYZhAmAS!&b=l?J<POP1F
zol?GMXZZVP7w%5}79itJ%Lb^0pv_G}Sqhd?kW$2fgZIg!XqTL^<baFM_fqs6=rQpA
z?Ypz1d-KH?c>3fER_4CXgc7hs@zlZr6GnQre&E=j@%;22-~0T#xPJUGBozeV%qf^)
z2??*RkNDuD4-s<mxslI8FAJ8(SNPfoU&pr|Kf*(14z;Wc-aiZ4a|4XJhG>G?ZivDY
zu`3`kV_mP%R`J1W#pA;RsMm1HFW{!54uh&8Ms{n=fmV=d$^R}Ov<jgJb<(KKlM)%`
zG!KIx9Qgb$913G76pbP<XEnTJs;EnB=fF~YKoy15tn0GialJBlds(SaZ8+PSk~5YL
z3~Q&zM*xIO|8NT(5)vc>GNx-J*dZABfcs_u$8Na^38urN!UZ0I{Bv4qP{|oHJ^qt>
zGIcX=)dCgLrJxjAI6qvjD3_IqOyCAAh-lp~yp7bID(#~-x9n!T=1c_(k<_!4GbAyd
zF?f|QI+ed8CMLl}+B-0CjM%P@sY0Iocv4KKBgGuhOAN3WP{C9?Q$s>NH?A@631$G5
z(zOoMW1v@%!drqL<gDOhX2oLy1o5=B6Km)C6TpRb_l+flfLH|DK}c?LWc?li>Xycl
z$#Ka$22hN*A6y$DO|~i_I`{y0T8oFk4k72{p+11QpejXX2#dIvRU+d?VnEoA^U-1M
zft&}zG?>dJb3mK~AtkPtQg}GG==5Ac8;u#u$Hs#<2VAaKJU%|+dR=i@FNCRT$F^^G
z4=+D>WM&_z-G}rCD?w96zIG=YXzMf-R3F-=fMw0dg(A-2#)V@v^utd35`pn>Ukca8
zU<Ve$-OFg2U(Q#^5m_@}hGi)z3$-@uWns!%>kj)VYO9Ee;xr4OHAEm4y*M)|mFNzw
za;_US<xDN`dYY+sn!Z7r;r4XH^V2h)-ali%Rn((nG<Jb}CV&tPA|0v~eH^H5$F@J=
z-P<ql^!6RL=V#P*aP}nW^fpT&VO=hCEKojB%cWp>T=Dq&H9max5wawVqjN8^&V%NA
zzlj(E*1TX@GB7Uq@bxRKIb*a6o3N=9GfFN<rC_TYi*6BA4?xjLR*4A`JkTd_zB65#
zXRf(<@g?H4dMD6O>%x$qs6P&y3-jCpAC67qrCHri3@muH6@)NHN&#z1xRi`_DJZ#+
zVG#qY8q8P`=A2LlZJ&2Hoe~4J;%WGR)map{|6=C23vHlfLcepc#R`-%@mY17GgDk0
z4N^`4n){j1qIvT2gVIW2Nf~P?cvu+3kpgL)X=2=}7{VNCQ-j?)-ro+~j=^jRHQWPi
zW`bFwRc@lSfOsp;cAcF^O8I=K4dCywj>COxCSxilF6g`B_5R01&~%^<GG!K5eg{7&
z$?oCCtWb9}HMGWIGt8p|6Aj&>TTCciw5InN5rjGjn}b%&;$TKm3i1*eION?~O6nv7
zpyY_PBtPh8;X&rl(VBDCS>30Skrrk^s8@4j@rY+E#Pj<}u_S*jCem*yAY>Oh#-J8a
z?i5&NakCBYV>oAvQL#%YSRbxf*2OJ049#%&N{W$=tF1$X`jlnm8e&KfVUiFPIcJ|+
z=>34jRD_t{<BSw552O&tc<KWd9ZvG^NJ+3P3F{@}vKG3pIvrR71(HGKOAsqy2*Dh#
zut{@)M1yok1tMoKAqk2ndo63mt5*+r^_s9t1iAD>jZ8A3k5-t<Z4e{RPC5w5G4vRa
z3jNI>yjx8>(g4Cn_u>?2nKI23+tJXrhA=3tIK^DEI~0RNg@ggK4ZZKE?S|*$J>EUN
z#rEzA@4xs0@1DLu`})_}>37mlQ{o({RwlZYBn-pKnFnKlQ<zz#ZTk)H-oL@E-mvE4
z&?gV}%W?q(*hfbT0=Zmpd3eCP<3JVCn+e}~&WFv)-;HpLL0y^vAfJLr_4wmCi-0rl
z%oX9Ppc{kuye|A0It4!5z8jlwy#6<WIY?DZe<E%iKuHM?>xzeUMRt}B^$CXNr=~i_
znNxJ#Lc{3%WTSOPE(X%yw0EErLzl@~k^5##OqOqxv5vC>5xSCkd)9`Fy42~0oh|O(
zJ5A^MI9UkU=yhHa^MbQOKiR9;qgB*OHpCcY`0VeW@%H^Qo^Bg<X9G+=GUF99)^$Z*
z7Ml1KImJ%N{gZO3;Vl3G002ouK~z{Z5&M-C$RlrLQQh_(1B##qlgvjh2Ith{9C+&R
zfxn3(;rDPM3VA(WLd>lRm<1umxyxvF+T;+DN)Vapa&AAZ>bi!R8HLlrH#p8eGY_Ie
z8!0h&BE^UvIYf>y<irAj5Su@G%AZF!J04DD9vpsvhA}^%!zX^<eZs*3xY~I*$B<BR
zV7O6CsMSxUzynMsZPo|f0HaWsQm__AkrLu4fJA6-=R=Ybac~G7Rg)Xr<xD6fxQ+E(
z6Uk#Qb=Zq}l)3IDI`<zgIOGA&2VbFqVN2@xe#&&RUDk;8lHJ90GNnkdrYVkh9D}Si
zwE@~MJ6!2dl6a9t6E6TFFfrz5UrDjdInor*;I0AOuu=}^MmokI6cSypv?P?0-0V=i
zFwCb5Z|@xz<C*K;`#`U>s|kTxqu^plGd+m|bL%^a1QFnFGoaE@`;MpgZ*jbT#uuM`
z51)PU9o&B6TWFUnOb52(8N=wD%i%ImGSlO-DYR~={cx*o^#hE;+d2kfGCbyjC9mk#
z&;eLXuqEUAwU6<QZ~Y{4NYI$@%isGBnt4&`qzJMs!!#Gbv?3ALD$HVH^o}<AIV)hu
z8Q19yYckA;+(0(fkMR0Rhe8Me2=Np(nN5oaA5_W|W_j3-DUt{-rQqR0R!=Sk;vcH1
z;phX+A%s#gEEx1?tkh(1kkUJPqa76eWCqX&2jAu_mpDC`Fi8*oe|R{VW7;n<+uG_`
z1(raE#M(Rd))2ZsXdn(4eRy*oJ_er5!C*dM>(0g+%`jTWZ9njQ+wr_Jam`dcEi7R1
zh{@G6nZ?Cpb#!!{VH{rZmW1VU!LlrfA)@z&w5-_cIN3oaV24qXC)Z8mz=3u&jDwlh
z<o|QipvFWgwu|7VVE~#yWxvJ;PWWKkxK|^mK^^9`bsGtp*FLa9ih^mNXY){X51Qm1
zA=Sy|u#V>#^ybwU?^SaD_~cZ&5Q670dn#DYJXoTf^u{>?jxo4^@J^Qm;C%g#AfB?f
z6k4u?cpA`!;N6~gyYq16S{LN(IC#8o&|>g>35PCm|N0cTp`;K=!yS5l%^Vhz-YQ?>
zl26!%v|K)OF$!oSB1j>#8xztd)njitM{?h<9KCa4Y-E_)pg`XK(1z~fyL(PK59i>;
zh2cI>Hy{2=3_xVDLQ^=-uDU994ldx#Ig{O`iW~&P*=I2)q_tqZTo@E2WYUn4lo3mz
zqFd(L1F57>H%R0jR8nE*-x{p<yPm=w56m;gc}LV?PS*$39Z<#lcVFPUpZ)`U^2Ili
z6evr^yLWH#=FR8${EP1cqoO2sg%2^JNI>15@yVO-VZ9VADWlhc&)>blo9Fi!V)*Fe
zui>Y@`O{c(!Ka^R#LvD1%?Ynwe~6#_*6+c$KKd4-0U<B=+E4x_<Ujr;1Tinh43a$g
z(1-pZ29y-At_!=;7v5yO(*gN(74{{_To4vDlaif&9E0Z5Jj)j`X=U-;f2X~&nukUh
zD}tHFEH8$tZ75>}J2Pho1hEv9s5sO!orOkvssc<(sLb>s4sQfFL)-bD?wnmS;=)Es
z1wsI#a2k)HjK*|3jkGSLj>0GSB0d}=?8DC(-C1+Mz$(y8QJX)!#%hKeaYb+H!8!xA
zlWJp52QB`<hRh8vltf$SC}>K0-f7#s6g<4T;<8+jQbet^1-L%1xNQa64a@b4hsOu3
z>y?y8ar*~VY7<&_*HQ++XU|9ut`a;qQNwAtcQRaTpbw^Ajqc8(gH`7cfr>zKL}&rQ
zNIz71D0_i9nLUi@Od;|hb2ds>J)K0&VITz8e&j^@tZ~38MpMP9c+Z(|BnO;xA<7AF
z!Mp$y5Oa5>=QK%6k&KTR0-Aph-FZoO4lisr1XR~2?{<>IbO3SQc_jHMLL6KK7iZ2v
z6~|F&pR82HSC3t7o_;pFn}9hq<b6V#N><<$4;rD8yPp$j${|c<QA9{IXe^f%ODW`@
z&VpguS4eQ!f%Yc;;9%;7%ppMt89^cXj4nnG8?uv?Xg-Xdl+zG{i~M}I3LHVxqpM(D
z7Pk%|gNgz?!I~rVAPrSDp}7ZBPK!g-(F;eQw$4O8fA-ZC+DE5oZ|FGhUl0y%OJ>ke
zI}U8`-{Sq-PjHPHaf{gZ4WE7XTloCb?^C^;IV*wBLVCO5i%-9U5(BQw72Ceylh3}7
z_wV1}gRgxPU;FyE@cPwj#1QeJehnXf?HjnfevQ{3eu#(Vib&5BgkgAme1%*Rs?4H*
z1(x|<G<7{E@*|wLg#gr+Asby(x$-9^_`U`OIMvYc$1!N0OCEBEXU6D`S^Hvgc!EzO
zImnk>aJj4~mxV(Ui7=%#a2*5H6wPp2F9aV76Nlu^VII69A@G7ZJxhEL90Evlb6aRr
zICYj-E1v~L5Jj?p-f5*xgSi1MWNG9Wxv>dn2~bxUx`i!$MH*tKx69m!qJ;CJ3j`hJ
z))oUtr6L%E14wlbW9l#5#yciPPa?twy^vB)9BdcTKbJKlMYiH8MJy%ba=id1xISF*
z@bE~}Fmd=}fDqiAq4*|314Ds@$ba+U?1cMt&_biNfi`GSK@K@^R7LBC>M3a&01bf*
z9tnuSfI!L}86Dg-XB=pZgyFR7X$?HUE=vHYfEt28udR?tb!7goAc3~!AplFHy+BR{
z0NmNrA1trl;VaUV(WL5(AeDh^`NV~36)*r%e}I_#JK0b3378Ni0--uJ1=KUN&lZ(t
zhymi%;=n@5=qB4nrx3CQ9KCTiWa?r<H{0;}m;`qXa92`;AjXWCIP8u=&^n_*CCpA#
z5F=1ls^7~sW4X|ZU8XqC3`h~eFbp1GBhbLo2^#e6GT)H&P6}WNOM0ppx;4~M5hZ&;
z<=NMsz$Xs-THx~%Qo!Z1;<~O_+}lh9=&iz>3Y=glftC`zHFjJnGqEnfAu=Kt5)&je
z?^szl3_oaVZ&-9-DTVac8d0)fT@#R4U@Qvs{XM$912khh-r>!=?_+;{NBVQ%XXz3!
z1V|9z*zok(Cpgjq*g)(XK6ohj_`{F!usk3P1r#VT<Dop@QdTU{<5f@Q&_L!n$uoBr
z1x8ptL_)^wMp>Yg!a0#b5f74EhC4Eh$&T~GMZ7S`Ia6%<qY0Y7YRJw!eSEUuBBj$g
z)Zp^$DVs9n7_ekN92O=B!MlvsJN90g`7DDiq7HOzfKc@1o?s`Hl)smG3^H3wXB#Ze
zJ7)-w3#2;Z7$cpd+~5nIVM~`x0j(&KfNit83;J4c-zFcRx@vcrk}VJ%Vh~UB${Ec%
z;vpD~7-($+wXyQb67K`h80fc>GlmFF*4)#^)K$+ZF&I8Y#F&wDhMA#cYAHT^^%{bJ
zR}UAwzCNHVg<==jsjh{a`l;WW5D6a;HulQcH=ucq`xrPHZJ0&b@!XDvceewtj)t-<
z#HKSBata<Z<jHKRu5gaTvQ-tsHx~h8GMT}^-=}lHD(o0zDNM)9^!91{K!}3X?IKEw
zSXU1IVxZ!9!dL)Roi;0Ocd;9T%CxQnXahY&^bw&W5I)1ycM#g5{X6*c914T4ZWw65
zD2h;hKnqWa1N9Lh$8#=1)i6#LQ)`?*9QzGhI~?XH5F|*-xU(n#mye%^I&I;n2iYCE
zlfbSZ6(0sKkq5&iVOf|rA(BqGYFIq2FKAQ3LtnuI*`I2<)4uF5?6kR~H`M5;y}22l
zz3`)QXqn?c*g=V6DG`^41=q`hWhwrZ>OM|%_ciddGG4TK2DAYISk?t)&EDl0a^^jo
zFrgZ0%W4Cr9We@WVPMs9M5M&Lg|q}*t{JZ$3v_HaZf|gy;OXsKjN_TsCqiA44*Cj(
zz<?ng4ehuA$HB_o7`<C!V!lDgSqLPdlUh9v`eChCmP&2l`OO=Q{Xh_+DKQ3g;T*t(
zg|9^>9&(RM$SI>~_ql=R*!ba1@##1XDnkqifyK_yA4i^o(gR&)KW=tzZtBu|@V;Sc
z2udP03{0;Zg$jqMM{B5q^w5|C4b?FC?>Z|J{yg>H56oJOvjRRO{5?66#%R%P4wRgK
z^MMPQOAKKdU1<i{34=iCY@a|pl?o>2-S%Dq3m6$tB+Oi5b7Q>;vx~R%{QsoXw%*+G
zoA|p0fklCoM7Bjq#1K_arEZE5dCf>A5gu6<q*$D(#D^j0f*cFhJWyiB>xV~NmlZK*
z$IT7Bn>)@bQ_tKV327>ifypv~FD^5nwR<e44i58MA2@2m?qNtv$q0EtG>Yqj(1tq)
z3e|{ji#wV|r*{rl*!H_ml7DXM4JedaF9l^?kV-!D3?x)fy~{qA$W)U%aUe*X+bDk*
zob?m}mC?q7CuMpbwBC_IMeh+-cu%R^^$N`&urUaiO;Mmt^D&V22XGc;5Eeu(Ok7IA
zvaUWaI>^{cBqsz+R@_|U#DAqZt3`u%3F5O8W`d*;=wCH4Wy1R|KI0}47aqzI6&lnx
zWOm*ux$wykg=n6xHq0{?_#UluGn<}YaWaL7fx(P=ln_ZfRij0EfC4Hmu=3}E!WtU2
z*?>et%mK?ntxwL2Q=<nYL`X^=C^S5+PfllMGegQ5<x;R*7T%9BFj~g*CJ=GBQ+<H%
z_aZpM4UrMRr4*D)#&RhL(U42PB?pwq13M@15RHJj7;!i?j^LNM^YDNPuSM+#c2yi>
z$B+S}0ue*oZrJx{^kc`T`widw{QG#dW=QS$<cm-7#iyUp(jIU}fC#~SR$Kr$Y1WZ@
z9Zr@di%~T?toFt(uX`5`7$hjll96*f9fyC{4+XOhJJqSPD&q!uPNkIICou=CIbvNh
zVUoahm#JoS8?EYTU=q=_q06M<IkxXmf|F!`mz?SOP&ln{NfArVfPmpHjTFCOz<~Sh
zyDMk!9Eb(;1TucWbD%iPy?+MdKp_Er1auovV-+Ry;Z`05syuAkQE8l}<|$zOBdSPZ
zc=hUnuYdhReB&D*<NE3m-GJLsp$>QDoN&2bP?qAI!b(s7>;q}{0YVVMOEICugmod4
zi4R#vJ%r3Hz#$|a=1f=1fO{4h?1Xt#2U;Jf_3#V`<HL>tfG(&<J#f^7wKE%GC}F*l
zGR+JEAL<kX6vZI{ip@k=LA9g8x}F_i3JF0BYe~3X7L=9kesaH{5SUo?u&yXMq2z+4
ztb|h}A|*N34BKcj4uZ2<U<PUTK{=~iz;pr56-RuWaLE*CQk_26qA7E(dvA{YbC4bb
ztygjojF@0UalNcq*A-jbkz-^yeW5Q^5H5?;FKF7vxHd*WxFtnawzRuQK%Z<Hbt65W
zgXkR2lgmHZIP?Ac6!7#;6OmH|i_`8TAenCoj=>K<0Os|Xn++F>oK$9*Og8P=xy*vm
z#7;-mm?DFdQa~;RIVV~;jDCv!dRJ1o8wnsWa-eJykaIxH(XV4b7z6#-aal7;PG~zb
zre_9*`DYCw048U3Eka#X3dGu%!oa2hQkH`Rl)NCujMiyuKMP-qGY13%Dv{4xKd`lm
zr`zY)>eo3(3Ip%AFYxy1Gi<l_`0U+V{MxTQqF4o5!<#p6aC>?~R{;Po1_zj#A1pzD
zm|fdoiW1<(xd293yS0wZ0GOawr9$0l<jgqXtmQ{}eKq*V2e1-e^7{Dz&4m_x;}*Ny
z3oJRFmIyhKi@ddpHW<KijPA*6OyD{ivG_Yx()r~aXmHBzq-1ds+7crJz$3pV2p|%n
z)-k%B)#%B^AIgkduMDFI7vs^m1?n6S(;6bqP~WMxO@v2=8@NsQY<Lc9uNBAcG(i|5
z21*I|`0<LL`^lfe@BgWv!iOKdM#u@<(NIjVHR6IPm9x@}F(D<UfQ68#@a~<Jfrp>u
zvl7fGxj=1T=u9H(oJ08VDaMGrEGQ*|StmT)Qy?TVe3MdoJwU*OI6ia%DUzp8JB0z&
z2qbQToJ+7GH)mqSAgGNU6Jqi%hUS0fUndf9DGQ<*9<K`?A1`>k6kOJV<$SP7ZG605
zd`&IL%Z2ldAm^M*f}R_yshzNjjTdYv@dJI#4G-S^dY5<p_xZIvEX})(;RJ*b8C{wq
znLTS-ky1cw6{>@{`NI{*V?~UDhleYcWpyfRK^xtN^Mvy_3keLv5zcCIj3@2&vS##J
zk>t+AGAJZRWW;+$g?2~~+QjQ5hfKzk+eZNGkR!Yb4PHzU?hqmM+gzdEI}4rG8ik~M
z%HYvw4k0Go2i=omN~@BfB%Vkll$>Hf5M$RPK=2}p#*Pg~9q6?o<l<J_ESAj47$Flf
z)-|!xzpl8fJN6qpCQ_P{i!uqFQ2@ukBUC6@*9%fwkn;s4U7>Lxr=0|d!R}bA;d&`l
zWIHrVmITQZq{uh)?HQl`)~};j#P-40kfP$V&pttYeuq3d`rEho;x~T--xWm?V0*d&
z+w;Bhf4R?kKoykpE-~h}9Bp9DxRXHCotI%ayz6e&nRCIM;Wq~gKMtD<ugv8KW(!e*
z!`5Ee3r)k8n0eCAipreA>Gr#m{&}m77^SiW>V7KEO2=m}^X&1-;barQ2VQ`RppF>>
z7*8eyMA_;YTQ2iTD*7-jN7^9wKCoHCG3x2lavWTwB86OTH-`(L@QHNq*&cPa)?LvK
znxQp+fRYGGiumBeM|}Om*LYkmSVI92pwVzd;@c$z#7wH&k_%EU#HgK4Hk4KNvC*1&
zza2=|J6&*YaxuuMnwEgm|0HGF3&aosH5X}38)$Ze0RmAT3<6+D5wDhv*O!b}Yeva}
zkPRw66A(8!GtvshX~U)lb9!nYtiW0)Q)aIn)q_@|0I39IiMT#wTrR~s1y+NtH5djN
zD>*X(Z&F)RKG|+$QyE=F-8gPWY7P$$GINEs3%NuX&&o}V^j4b>sCN~UxkAP%*=GV?
z-z686van;z3!zd0@^Od$mJ9U^m$G1q87XAsTznz>P|(xPW`@FKH)R)0UYCsRx?nSP
zHVQjM3xUv*2d^+Dl+TL}W|~VP5kOM(&7_~I;A!&8=Y?-+<|mlXLDY=TWu<=*&lmA$
z!IRkFRA=0;S%}dg-iQ=w8!|H%D1798F#VP|=qFOp=I~UV4QC7|;&7b~136{%P!S<W
zrdU!$a?#g>%!=Ewh{Js+Q%hN8!{I>w9!8Hkq6h7Eva>kI1x64Qhg_L4LY4T`Ox5i@
z-hTESpl(RPaNKSPqd7bUv|~fNy@$krer)Kq61t8NF&aYPbJP12jT)K^glOnWwpwlN
zMP5ns>mNL=ahxjWI6c19nO)__FjMeZ>$y44jbwJ-Co>@mN=Ue*j7uqgQy9bz5X&FE
zqju6lkJf3!JAmGmNQHME(>o`+o7gZ4sQ_VmTL2uaq8#0=B6Lzz;SrI-4X+RNoM3JY
zDe{kibj3qEv{t8iQFU7n%G1{t(EEzf21XjlfpoDL?6f6rd&S}Ugt}L>PLJ~x0+yU`
zeYhZ%=sE{ih`Jxo3b&D13TFm6BbNnccQZCqh1QPa`I!ept5CfF%E71FV8Dq(CX>?G
zz-Y&%$axRRZb$$k#K`W7!2^w4w1ogHIpNj1;=_jryt=Mfa={W`<O#Uamt&^8g|ixm
z`Moo9A_@<O-W&GXvF!)?=ny9rl$@|!3erNQD^GmG0J{=|1T}UR6#Eg1Fy$sH(N(wx
zKnph#x8>-q;W#27hGrcD%mT<I5u=V72F0>O1h+JwVyfs>G>aX2%akQi;kqmyMOz4;
zMKb1&>>)Fr(XlL<n|v%tAtR)PuDudYza}#m2Szt9Ga$u;ODTA~u6SMwMs3`{jdQB#
zeqd&YhZ6yZk~Mab1t~_XW%20JjNX{VJRe=PDO7_WMhPZ^3^CQyuA;T>qQB$?oOe^=
z?5%lFIjlMhGdRo<<{!xE7|a<BJ;~rtD)E0N&`DtS^PD%qg<{cRHHH-_%`tUA%-2y$
z&MJ#1DPVLO)v?>@*l!K>P%MFVG41HsQ+GOfhgMRgNA1WVd-Uu;?Hx%w(0HIQUs>IC
zm)*~38?<I1B}5EFp<!dr8KR0_4<LYXsEsC<B_)(XC~of^t-}L};6vk#))+#jMd4c=
zyY4|`hQu1qKwxC?sT#CV`93omey7)0W*<$QRRY-+F-KY)1UNsD6wg%L71ga`-CGgS
zBJ*UG^8ce5WBrVH_2_yoIC%9N1RTnyL~sAfDsWmAP(Wo4G092m3UR^&+~CEjfx|03
zB<>)Te=E9mNC(<zZaG0eqG5ug4b;7&?-jjdv>4oS*xOXM9B4<!{(Ru}wBdH!(OT#4
z3NM6T#!ZIQXNGf^=qgtiIfWn$#+bf7!(jktL2R{R+YdrCAY00;!;G}U7!!xqB_qU)
zs)oZ9NfHK#gvA(WFbFWRQjN>o^%QVwVonK{%Yw&;D?Yehv91dwyW)+s0+=wCu@7bz
zG-n(2p>E&cvzl?I3idwZ83{Ekg^7IAO_hfXSyRmh3KU%w8h{jOgHH5|>u?Gag9xJG
z-27l15GN0q@1vrRj(mN<<EuwJzIsFs0f>h6SaCfX`qojqAr;y>_tEM373G8`NbU9T
zoG6$9^V}Qn6*C|Ba<XfZTcjWr<h(-Byqg80WaLnMu-kcc&CqJY^?IRww6by!k&Kj(
zBKqLw4I>j}s3Ibu1VT$0E}n}9i3>%RT_NJEtq_q#kX=n|pa3ky2n9t_-De)P@i|Q~
zc*pKkZjkwN4rGO$!4y`Ft^k^oFMOE&va@h~1rJ})hb$2mRvG6|TS0wOH$`uR%2F&4
z*<Cds(0j+W?{}!r4>yJcH5P0~MSgrlbY_ZH*A*R|xO}&H#Q7|&d%EDv02ybz=VWEk
z3z9?lToSXLdxPn~8Uu<L`f*?YIQEKGJFMGjQDbTlyT4b30X?Gkz=&TJ4^3k6lS0IZ
z^sov>_+E5wp9ZZ1nmSA~Vc*~NLqWLMLzHvagYZImDK5G&hL>l*sb_MQ33Lp!VYnR!
zp0@+X!OY{~C5x$@&=7}R!Oh1wZDtxMX9Day4G6%n;dmhzc1j5fgXRFUgzh|EsvYU#
zseNGh#$!Uds+;)0z?@||ue#3GzjySG-U1Hc<}^l!jDc}<w4>qJ59~X&3B3=b5D^mv
zB<-kpdV0dE2X3w@0Hy;$6<KH&7K01_3_w?yHMa{_Xl+hyBgRXn)Bv<o8{jZVh!N5$
zev6R|AW50DM8Ccgd?O8pU{sxs+MKISPWpKtgk1Qtu2`-MEJySJAV(OYZ!)Y-=2{3K
z^xFDMiw`)JdD>#k4Fu==o^vV*aHwyfX-CzL4#koZBu83&2xkT{1&?iIg&G`hR2!Z_
z(7~R=MP3nDE(O=?1?yT6A?Sm7AM0hozV7Hp^ZboOI}jJt#T?y#XyV)K^yECuCUapz
zF@$GkS12yM7DS|&>5K0(vlJ6z$jFh5s4jH8%qb(6jAbe4)yWSwKzG=?#eDNoX$TP|
zEoX#e@;{cS9U0RQZ?GT=Q<?4WTaC{`jTO^WvbTZb=s0#3O~pg1rY?rVbMHb<1c4pE
zq)pBUP@6~g+`ydnI&;FQI}$7l4~05d9CTIGdcdHFIT6wY&gys;iuYK{yX`}Cgh1UW
zDTj!M^@5TZQ!_iFS_ii6fGO4M$)<l6PIzs?+NvJh<T=+GfLasw;|8E(KQ<i40ocIe
zX?>&9v4F#Q4`2!r2?3UXoTrF!iV9~j>Dn8=a0Yc~m_I8;!ik+9l2En)qXq6Gv)KL}
zUSIK0pmI1SJ{Td>w^N7Do4^M=hRES)PA*pCf>p=vj@Gd475jdm)dtQ}h(imw2@G_=
z`Q9;m_H5vx6Jvl3;(3x?%^Vaa9gu*sTyR}i7!(gy&<A~j-fsta*bbi37|*jIbdKTZ
z69hHb=$J;V;%eI7J4Wjm?pHp7t0jepOz+er)Nbg5gS%+{{UWHf;_chFxGanN(-bm;
zY~ZqFWUs2`L&`KS0o2fHMca4m+m8LVVc&PGeh<W45M!L5(7uCNu<(3~ETEOJvZ_e4
z+ll1;o6?97iHnh;VRoOz!~l%}Rx(0il`8Hgo4eB#C7}0)SgAO*qdWWtgsUrDg~C0|
z_J0_ynvLr7kOI=;blnh674Ft5o{t@p0xl;}!_GaJ2a9_v`9Widgw86Qw@wjUi5yyA
zmIX^${C#(%Qk>Z`^(5S=g0R3TOTn@hdi<C9Cn2Wn%56TZaK;CBjybbP2$2VKk_450
zV&AH>V<gQsg@|Efg{44tSv&{00>6AGEIW-4>F6mWQb{}ogp*dP3MQ+?01XBW2^#d=
zL}v#}%B!Kah7=Q$L|9k!W1Q4Sb~XVy6l53Ojp0JcAZInL3iLYAyH1Ac+)g0y*<7TF
zV0QZL98fQXV!VEkd2_lm$9Qh?t-5)pYlt9R(<w8UE4W5U1dQ5U3SuS%n3F?3y#pbk
zwTfCR!@1NQdolG;Q~fpXZ}4KMvG5rziAL|JZHHRJaqOt$;ILgN-kU5ii6nj$XO=A`
zQ)P{TWl7j-^UvZc=E&}8z|-%PMUD>?iIf1CS`oI|g5j<VgmZqE4+R!M^TKm}5JSL-
zhS0^UM04j$MxpmUupJG}4B41%Tid{PR2;2wsN6{z?Y$vJ9tzW2-W;lu6y5iQbXJMZ
zFiHUkPIZfdr7U=OyfU?H;*PDNB)~8*l$o^J2PkOZCMs$ug_ayi=|7?Z97>T<j&5rJ
z`1hELM2uW?qbj%de*U2y2BQe)12i$_ZUg&q0H$>5j1<@8NtpwDvz*3)0f-S50a_cj
zZ6~CVIje{h8tc`sU%;px_1IDO4O%Ob0gDI{c!~=OvSYBhWG-w|Gz<(+tLuYNo^8xS
zj#y@j38@sMr7&L62CnM`LkIRv&|`(=1vM+CE%(FqipymIM3MK75UOie3~-x`5Ccjn
zxL&VV*9#Zv=*T5vw2q_gSUO`BU25*D^DLn-%^;>q+mb?L_f4uJ24GzZ9xoSMSB5T0
zNGPo#rNT-*Wt2s!m&kO^T`mi*j~867{IgzHEX#6s2m(F>-Nu<8pnbR?k#_94@c*Y&
z*xkl}FkG?f)*Is8X_N$UD-rlX8fRgY+&Tl?;K!hIYcy(T>gYJ8FP1|(5DxukP)kSw
znkeL)qB?+(rs_Zwil!DfeZ(Ne+8Q_LdXSmtqrVpi++=#qcvQtv57gs8Udl-mmHC;3
z6?D!d7nCLAa#<ja8+QtH2$<esWm&LZSA-OB+;$ku^#=ojiWxvN#<UXmLQ;TI=&tbi
z!2_BaBvKtsS}vUD-w4YID1J+e8|wu~Fpq{+Xw-9NG4nz+uv-BPy&o8(0vL$mOt26<
z<C$GgiL{+rbHc-Q;ruT&ENemD?)2J}k{7=Q4g72-vlyUVNgJGBP63~v59a{qM;s}9
z#Y15p!#?4+p%^A-5IUKB;q>o-aU<!2#%w_bw%+ly9oUXaMnmtkcXckm!cI6KhMD>7
zzcvz|9I6nb0oHWd42c}(^-{20$T?mTEg=pu6m(q2N_ynhIn?W{XoiuBiDZZg2$?u8
zBtSHw>)nz$xyOG<a#~We;$_D4(Fa;r<^?Ek$dVWkvaAV@k5_Jv>w2n}b0I?_r0l^c
zaUz+n5*J9!fEsFK<wAbHU`_n23{p`#b?N{jnI<*{v=QFA1fYb3=-6WtD$YwHzup3c
zIYSjq4Am)bZ77b`akPrPR+O=zxY=2X2~g4)M{>WV)-g(Eo4#?-eZ5?8ePsM-)r_((
z2>E~-Ela4K;Get{tk(;!4-Y3?S5ihd+ESnBjNU_rX8a{Y=)?&Mv&>k%(_d)dsO;>P
zoRL#Oh~*q?hD0@OP7!IzSk3V8Y9(~g49k@UZkNjimz9e0lnW`&zVFVQUGt(zheBfH
zJR#CJu9S$J0=jXX%OIB=4y8@F$D*Oz3`e4E0^FY}IAv0bv%~_wZy4EkZSUCk1GRD%
zKsKFMj}VHaBzVWcXIR7qmhi6GK;fg|&g${QLEAuI2HJ3MFq`wA*%iC)VnVW{@gRsW
z(lr*O$RxuM1Vd>DGMQ-WB?A#2Ul~w9*k>&fmrPBPq=+pggp|+@J&hBg0j;r6u>eRC
z<dUFWUGKx6f7wZS&N<?8U17nUUcJtss+e$$Va{DR50VhhO*X`cIM~67Ot_2?!*J$Z
zcNjI5;1pp1Fg|lBBCkxMlprVtxLga$&KX0>0U@$L?4m$iD0I)9sSWQEm`cd3ayN~E
z8)mTcpRmX8=Am$&HhwWU41qIDTooKhN(-*W6u*FzIkdL}Py3GN%J7+58-`1?9dULD
z$`xz=59T*XohQla6O{F*;5S<kOCa~!wFDGZ0ZkOS4qQsXkvpullPong7xQzKH6tQH
zJ7DO1FnlOm)&<LTCF~=h<EsR8RMIa+FccWWATc6By~_}+>w;I054hxoR5{-OCu=1o
z_q3UQ6Y3V%CJ2ETY?e8zNp5v<rS}iZ9CFJ8XD)3Fw$`l!V_=Sj14xmbf0mq(oq{-{
zQ@ai{-%R>orMDeBYCF*Sh3R{KD4L6s3?vCi6b@)d^gLwHCV44dnYr-Dvyo#U3oBm=
z9v>g@`qd*oc=Z~O4-Xtt4?{PQ8XN>UQq3JDdeOkotzL)>VKL!P3)xP`fqg%)ZR{9=
zZ{%(8!4>g#T-Dhvg0L1mUN8JSmEyByS$wS(<mle^uBN8ax1aXyz{Iqjp6RwFDF#Ra
z!kBba->5MyLntL=*u9<sRwNtDHS1Ns+$|_V;G8X&4A{=+t2K(R`hd9h2j}l2V@14@
zRs*fO{XKJBHQY%cV8&8vBTHv=!%$x95GT!f_#8pb!`IBPw}#j}-O#|p#mFRr(`i$r
zhgmALMGqoamjxm~RYg?b>d#)u5t0eH#FQ{%bM%0jC=AkqTa6Ho8tCr4%j=tBU=*lZ
znk(d-70m=iIy1Y?iEn*W$-7yh2=?PZHx?vo$%GcO>uaW$6}VTB4(<v<uxki8AuI(k
zXI?WzTrVrG*BJwxkc($eBsvew&JpmvMoHKhen<~<+Y19=(FL}%tNus7zH**|xzbm<
zLXx1>P8)ze{euFlPrnf%J0=YJ-RxuF);o^Q^e~uHTjymGo}4qUHU^v+BIF+469Z-F
zF**0%tFM$f{2swDqVeI@$cH$!bHD>5A4;6>eGZbGw=Bu2cX7~gD<z;Lgv)|-S&`Q4
zZM+W-gCMnrlsiJ)Pfz@^P?=jw#^vF{MV$+BUYvHiFfKDjns%j3EFP?UnJ&lHnY25{
zZw!v1hEw3ipqYWu8oC0<QE|I%*lUH3x#6=4j8RZRM3IQaMTXV9vKz_@vnwlq<NfWe
zqaBAsQY-Sh08!BWl7<jrqGVddh>#L06y3}-d=qg`3CP!i@=QzTB&;mg>xz#)_y8Y$
z^btOI^&!?JQ$3vyyjl#09N6oD1P?feKswo3lkm_S-wHlgP#rjq9oxR)b{yD`iepq9
zM{_#|R^h#OJUwsdt-FyO=Ly&4a{B+T>jl?kMUD$%D4Y?T#y&WUnIJD7K36KUgW(1x
zb8WlXUWfrv`F^M7rn{1OD767&@Q;~Mg&xGaRO%GOI9uSHY9N&jlR7$8xHDX8*x<i;
zMNc9H-|KJ)B8UgsxV;IBi$SOl6qAl-7<IV3KpZU~%{J$V>q<p;@6*%E=Kyjqz!)tX
zU}3t1l0sZc0WgqK222rS!nzQyOgSMIrWJ;g5lci|GHg32Jpd0r?jOmA_YoS#IVg_N
z4GSYH|0sgsbB(fO#Ow<BLF2mavBBNwAUsq30D6|KRxrT8t0gf+iP_lw^ylK*0A_xR
z8J)7a@2MKtEd}MhoB>mltP{Q9HaIIVa%QBXI{ZiF_%P0K%tMC8%!mK6D4OB-)65me
zaQY(779s?3+XX)rJ<!{XZOyGVJnaXb_R45Q7fYFM%E6|1h!rS6Ln=B`iqoxlIHcfV
z;r_ki!|Vp<@3sSJG@jJLO_W`m8)qpYD-PmVy$G120Eslt3vOdj60Ag65@_Tfi4n0B
zsvJ`Ug9#i`0$a%#X%oD-Z@&R-?Ko-T0TdEa<|ZS-2lKeAEC%55qsFdmkYOScy))<G
zU|v0{@?kxWisPunKqG}cFgBF|Kn{WgBaAjWwC>R3K-~_sot{n6zcvXFEReBP8)~gc
zZHFYzOhzBD931MSn6X3Aj2vziTBeif#RtL6kbI_5)`Z6o9`NebE4+Gmh3k4jE}2C@
zq)=(Jjxc~E2|3U)c8qcN^HDolC@HZDQ}>@bTE*dqf<!ki3jyj*w!IJR+m7R?WNLIp
zEYmkfTr8E`cc~zy1u+(qA7JM`ce=fp|95rYE$;~COpf$Rca{|Z(K-B)DtQ<MfnEDa
zZqNaz@J?`e&KmeLB3Ln|%&Iba(9evQ3=*8p<=4qO1i*!-XUFWD6quw}JLg}$^Lw}f
zfis557#!;3IYU_zF6)YQS)2tp+=An-&X;h{yEtDb1m_W;swV}MF1a3;yU-iU#93!0
zLTm6LJTQFIPOkS5DlA!{3(#F;%KIi?8eXTAv0fHjE*GRsSZ@s4T-4UUC``JKmO{#U
zY!r)Pisn2@R!vcI=A6VWHFPMBgPZf{<}n1~QKf-2e*?YTGGU+Kg0PoL@fZ}jCj<m%
zY{F-p&g2A6rcxh+a9aB@z(?Z7Ou(dd-DzwRNu$gu;kp#OS}%B57p#RtJ9yRdbnK3i
z5`U%W#oK)WyNbX{XcE`Xhe!fqlJh|`@hTmFA_9_N7C0j?#oikN=y4)O2q~6`<x<ew
ze!^Bc2$lfkK%6}yvBEPBKrOL*Necxj8VD1`AT*Z)5H->VNYq_lfpl8dx^P8k0sX%2
zz@9N8&aP+?fy4v@6`WlaM<=b-K$lnR9eW?B3MNvixPZmaWcB1OgDcDl-5@Y`dOj4x
zjq!p|_#RDRN=VImM?Wgnnno`Ak^+_#kptY6FCqm%^lni+MuR7-z(BLG)~oZ$ol3|k
zJyaX0b)Yo`vkwI8Wx@5;1?94!Eaeo0wNB9*mt93d!nIs+DOb`nJDDd(qpEZ^`5dcI
zfYo}ZeZuH~G4Za%h+u#YZn(8J?AwOA?;snK!asmRJvvTCr@&5wIERh=sP|hy_3N04
zW_K#CdiOWfPA@M)MmT}&0Ww}x?KE<nwjn(5)$PCi5DLP~gpiPoR`e7la=jLZyp~eV
z4it0aRdro}A5M|<?<w+9V^nP)9ko_JaL*DS3=ncpGj~5uk(z~){Xi-SDF+HwLBZrq
z6};~&_k&WLwbVQPg_tPF3Rrr_%+xM9pAbn1K*<4XVYDq&7<rm00*#R*m1zUqTEp|U
z5%$bZn8=Ab6d|m+tl0<90ez4>FiidMn3?HL$M+j)g9Q+5!Su6Y!eYWw*!{FhI1duA
zSI$#8(_+UF906D^3ogC+EJ-0j{a|H+AKW>BEXlbazJLXdLuoV6n%)fkN1QZ&<xb&T
z5OAl>A<n$ux-N+@$aN`HGdqqN;Da%jkM~B)UEZsE!2C%yb8i%hj@^Yoi|!!Mz>6Ue
z^dU~SGE`IO=m;7i-lfL*pyh~i$rxQxw}w3j6r<htwIp1f`o;vbh&p^2B)B`a`X(^E
z%AFOmL=XEj+BnosH&)ny2p8%MLutL?{kEZlm2~!pr?&1UbuRF7f!baVbTuT&(BAOw
z`5EuF4Y%5Sco%%oZD(uRl;V~F7l%*WRwBEvFfByr<C2^)qOD?o-f-M@XyXFSG2psn
zJgy5anV+RB1($V2DT|AG%oDfdj(LNHL#-9+f}yP*P&3@N9q*oYJU#8OqhU!2WzFsk
zTh7c{afqawp&G5cMFeYEu;vBpvY`7wJ>`U0gB!1L$Tqi~iKVK62V_icJ?)OAs<774
zYsbFVvvauM%644)837t|L~<{k5Tb7i-j<tt3ytCTnna2m?~mD>iw!*;h<sDzVXBkE
zcZNaO3E2+k!}sdSTT@n70*dDpK%X5ui;`t2xU3hf>k35X0f---bMq~Q&sa28pumK@
zX2(vrb^eDP4r$KG=VifRnJjZ|5>ZOQx~|ZnSeJprH3i}aYlyyw01noRSGJ6q6u*wJ
z`}LsWUQP&fW~J*)Cd&!vhD#|}QgX)uT5a^vDSk7$wuV@U0_23W6sWn$91JFUM;844
zgb*1f6{4%M3Ac_xYo0zj1a3`k)IHJQFr3$99~<_)A*6_HJJ8x>r;)xNBD?I62)6|d
zu3@3-IY%;)dgJ_S&YaE-7}H8y9Y!Pk-p#xy`<-83Esir~herT_K{inUb&N3u+WRh%
z{9i}sIc70M@7xsnFql&022W`+Q$*#Wn;f^0a8k)g$(*4(^zyO6DkcP{s*aAN0m&$~
zWgaXLBr#LD2y|6E`|v4Bz>)-)CE{A5w}6owc4HcxN+5$L(&MC#h%f*ff@ox4APU;i
zu^q!te^%MU?SfgalbYA+fTaOx6}e^{t<fB<SHvnnj2JNNW8nFy*pEtnHY$2+c=LS2
zyXPIX9#~3&M8+y^tpOI?np&~f18s1@4CC;3V%yz`EZP*37;d7yVKl{lG+K1@juZ^n
zEV#I6>Uvo{XPDhWiM)SZsO%o9!03=sVZEbGuN>pzecm<<RXpEzynpwM{b@%$40;~=
z!C76iYZA0xF)&cuj`#H$-+J|R{NA7X8GQV~D}-QZtzoMTw_1_Pg02+G>7bt$dPC|o
zMI&|AA1<HcNMH*-lpAtu1JBPpdh3izjRD;hm&+9?6fAMYB`vsQLL*HXKdYu3<O1xN
zl$9ckymcJ|dR};{?1y)?gqHxu*t+h-%@exA0qRb($IkgsuMG_fbn0B@;l;U1!w_=v
z;}RexAS{V690t0<?LwkA=VV<Cm>XFpSX7jlP(0WrT^6M09j1!feGdq3aRJWO0w_+U
z+d)EERxD}3fdiKKTqkglo>D>%3E7bc6FY4fR1L?`QQ-4S1azgPO-TtQ@F1|1Jo7pN
z7S1lV+can7!5t&-G4((!-YgKN5$IWr3OJ(-QJ@$QQ6vmVAaXF!{km}jcI!pXq#8xJ
zIl}Ybsm*c0KUB=u7Y#qPPL`rN6`poGG0aS5{vHcBVX5GOIT68GZsBx)m^9$wP68T;
zs5l-+EBqe^e57FL#O!&=oS?-*6hZbdosvDI+y_f32B*YPO-k#c0O3fSLpYgLgPgwf
zLGCBFy6DIZce6Zapba%x8|cjtd%!XrS6u{>$q7~?Gb1el<r0xMfyIcD10Ek1yuPf6
z5I5;lgfTE0&}=>|5l3OCGIZeRjr3G7$*R?X=O?PPYiq1T%>15kz%_?#IR@Bhcz%Au
zk|Ng3G^nR1KFrMo$LJjJxl1ce<SO3n8xChqMF846Zrer!y2uz$QzNU(6x(rNtBvnP
z2N#ui9)NF1s)l+D+-?m|&pY1VcC<c_QpD9$-;!H*mqHu#>|#C&!W03DF&aiYFkJkz
z?+0fM>JY?!pehfCr>AGM?Lg@PX2sQ{l>s<T^KHGMX-69sZ=c`cb$Nw<>L2?#ypk^v
zp5Ed&zxM_|@e?25gNLspme;_3p!ePL6jmI4$7qakWhF*9Ap81EDRHpxo$9Rcj=iDQ
z1KY-MqYxtEHDQSh)>Lpw3zq0JAQ_0JZfopAaLhb6apHmjM(XI8MrWLG1Kkic_)wEZ
zkj)MtGWWweDY>I|96M)39H6_|Blw>CAQ3^8nv!9Y0VwPeL-a0OU>Y#-+5qC{+*s6y
zuq@<CBIGz(axoFsO;Jvx#VGs}WRf?WFyhL5n81vXaBxqwQtr0s(bdulZer?Mh}J4Y
zwnl}4?z&N?5hCft>L`T7c+z5-TVk|XQ9Zr9!n`NWBm#hvBi53cw;>z?=Th7SfC}`S
z14Jjug-C>s#(}c|wKwd?j@r6+RAkPil40t0;ts9NnSiR^3uDN6n8Q1SFx_h@pE&2C
z!H3N$28>CAZ|!tyR)-kJycSMGFh96j|1oq5GK+XOs*)HWesRPv@}(qPiu=3d<d`MD
zv3GL$yLNBSfa!`j$>%h93P_r0Z7~;%DPXR0h69NLF2OmIFhGW&b-<wP7P8k|(s_Vo
zpj;C!y}&xKieOnY9<D1&2xzq<?~1k?>e0^XvhE$XrvqaQ96gh3?gBa0;hXA?V{h(W
z$U}?|Fb}y@h*L6Cf?=b*9ys=mG)0NnN5$=^*m`#>0uE+d@7P<#)+*|t61I4CqoboW
zMsIfS4r=eHLvgDWwKlf9V<2|Jnj<w7@vb;;X1E;<&s#(5Aod;^hO&CtO?DMCnOZ_y
zWg|n%Oo85^M?*gdan!w%3M=$->1t?Qv27K4klJWSchQi$rjmI@%5rZFqcyDSia+wx
zzaRO{cVKvnKldN~68;Z=`seY_{?mUHvb=`oSNKDJ@aOO|Kl9TFC14*NS%97zLOAXa
z4CtLC&IHPSSS+r&2!X@H)|#`D22x_tQewsu3)aL83lYMC-4sz74By3%g%>O&olFxI
zSO<fhLI8S&4RZ?+V*7OMC4<g{*&o))`kI6A>~O3&<B62X0dq*9bsFN)snl674u7Sd
zd|2vLrrn=U)_UCV&4POhaiF@Cgpz2(@Y&B+*l!sg+e|!Ug|2EyV8tGj>utic!Js`m
z>Eq0ZnEy7MATheqY_d^tKZp0~N@kK)7&;VVV&Uj?-0F7w(;I0xTS&(s*6*F#Xq;!M
z0eM}qTo#Y^yw7o_9~v`DNMe}7UIyxPgk$4eVBhhyZ-4>oy5O?X4Y!m``=0J|HcyQU
z9L8Y=IGH#D2?HI2T|i*wwIrOZwh&ycA9)RXqT;;`J@?}%&Rp^kF%~1tF);;t`!UQE
zrf>a8b`JntwjGRZVfLwf@lF9ohf=~u{Z8u&>LX%K-a#h}mnFFP%^#{siPU*w!r`XK
z!$bs_E|X&L!5|R2APvPpaG5y;ALW{$wjgLgR%3SSx?u4IYo=IILkt~uur<^+Hx|R<
zIKJ6{Cf=w%fZ7zjy)g9ifzL3fX_kn%I_1r|$`U=RS_y;IKCrijqczNUK?5<tri!MD
zqmwoXuSb)RSz$`ylB!`l4m=+Rj-#P>K2%K3Yk+3dOXp##f`NfP4An(yq_dmn3yZh5
z_hB_zJ`E~ve%Bk0dSLX1aU7^e$G&&mZaex>Q7+5rfplygbsw<q12S=rmH9?oJ4UOh
zArL>`4@fus+_!!TUtb#1vEeWNwO_&i=KuKb;OTbYfB%2_3x1`6Z-4tI@F)ML|3kd~
z+Bdnb94a`~oi8dTgAq@T5peZzj^(V5Lx}g(F1LxuAz=v_C1w<dQZZ(;OQF-e7i9Y+
z=k}AnKgkP|u|rW4**$<ljGcJMbF&`~MGWkIxVP5Gx&v`_xOSkA0rL#%+7!KxbNz@H
zTqJ5lB=P4?opKp`ek?d$Nhj1u7E^SGfq-B8*$v<M=7#Vs;Pu~J>HZm`BM82Eg~TF6
zjRR4@>6q?YDsrFmIK074lAGD+F68sU@LeNdcl<rbSrHGTk)5{g_72pX5O(wL^$?(;
zK&u0-cEkxSh5I#{hM5YZ-=oY<pN<7bYuJvR9W8*ZR&;Ghv79D&%S^{J`Zoo)Op!$7
zB9(;N4yPBB&|m@R=n(Z-*90IcjW)dyj~-5h$;K4#Mni!6kDsK5^MIe%+B~oQM|geZ
zPC+rb>kd~5=Ls{XH>ZTfb99N(C$H$)URhwj8Ov<te$c-I?kB>`W|^B%>rfTMz?0Du
zfIvSO1e&Nx<j=7n7*Nz<K<%(3Sb`vzfE)m1K`aqT5|RQj7sx^bz1kI$1<1rX=P<dq
zjv<}4;1}i~aIpa*31$PWxgCVFPeO33?+}2JApXa4M97Shlo)XU(Ap3p(9AG`oPwU{
zjEL?Fb$HbiV}d{t6T@fLOxuLo2I>oFsrTkyHGT-e=(wdG)u|oB;EWUc+Kf7=cn*nn
z-wVUSv-@2^3~h7_ZP?n5{Wx&k2peH2>fZ3?w&VTtj@}2dW8S?hw&$IS@`EBLL_ODb
zA02Jqp{)TiB8(0_Dqfche)hwUVDJ9{GzR?U7YE+m4qxo9Aco<$zVluD$^Y5Egg^N|
z{~zPyZ+snf9H>fRQ5_9U$JtVg!vjdcHw&fbQewdUL%HB#xu9fPi|15mMdG@UD9P6v
zAR2rFQ$&f#5-30uxpM)$(mahAt)Z!UHK_=qh%%61ux7Nw9&{tEO=%U=IP+LDHvm&A
z)@!9&8;w~2Z5!D4>Vr=2HfNzCFE-^QcRwF9M7Ul@PP;l+?iw%*zy1?9{MOfZz<~B(
z7;f}e679h0g6y%b{`<&J41S*iL=vK8P(AG%wV9Joc$oELrkn}?et7EGU~v`Qc0<rP
zBNzkH4PlUlb%>HKJJr3$?{BI&YRA*<z_zjLm&hFBm@~pql(JAv7VaIE4Z?o=e&Ff0
zBSpbOpw&o95r8wCGIdgAVdSS-Lr;oasNj!FL|PM;BcnHiq0`|lF+s1Cj8q8g!e^Dk
zhBJ`jNgFmBbRRYz7*%OMUqW%)6e_u=^eT8U-+w$hg{d^YpML7r#{y3I^K{<y6aEbR
z27uscQ0DG_L0}ev!G$^0DT<K?fpt;=O$}NZ)n~yFbHtDUw1J_5RAxZ>T_HMiX!|f6
zm4cHxNIRuyiDBRBLzyPrqndEBk3gtlGL6&>(d~AoiZr@P)i9}{(@wz8I1}1b=Ymi&
zEJetY5HIW5zMJuLAH5-|dXSa41@N4;M6|uTnHDRkm@_cj>B#myYH<6DfXh-iPuMo}
z<AC69AQlA3B3L40MTgQmUdKIhke1I8*LA_e^};;elD++9byi2mejK>f!-JC=q8g6c
zal0L;^SZZ&Aqq3W_Pn7Vou+&zBoNMxOVl94FkEmK)bQb}R|xwPL>C-3(7zNQHMcLW
zp?B||@W1|-{}uc<|2zLReDf#2j(xkKR{CrW_s+Sl3z<_sW02tfR1Qt15U{KlTnZ~;
z7&jY%oT_K_0VpQ%1D>_u*!rXvnSqgyj>AvItU)2efUZCvC`s_aWyPz7ZnjkoZ`(6&
zT@i*LS~?2?8*`RG=gVGc8H3)rVYiM}U9fc0l<mB)F-FJLSsk;u@ecyRNfPCx(ArR+
zjohy`0DST7ilFhFqKMsDNbF)=9p5{190p+#nmxL%2J@LAyNWKQo|YI5BPe<!#aoAK
zY$AVSP}duoq1OQ!>NAc2s26iO&~zY-fTPh)`*z#t3FQZ5h<B@v7=tU<3z=`~U2OnV
zsRs%%BRD)xEZ;k_7!WeG2CQ+_Nfz9TVvfvGaYxvgBAP>fyzm}eCI)d|FEzuq?+6M!
zE(NtUgd~Jj-9lT<kbQQNVuU7c6bUXDkf|;CF$|i6u*wfJ+{9_4YKD>+EI&8rDZmQ8
zj7Birtm80x$It36c%Z#PAf}#pDbB(7-jl>ZUc7OV#9+!0a7}~lzT0by2y6r>8ioPV
zx%guTx-f>5H;p8~BM#K4h-OgdKWk=Hp9)+2U<O$;gI&s<ffHi@W0>jide1X<%L3s>
zU^&2I07612g%wyMHBg+<Rej(j29(6$m-)(x)(S`mB#~48xU6{08P~!pV+^He%d4kj
zSv*cq>7@@)pyM9-0|rD2xW?$-x(|pBHz*UVIpC5DvZQ-BiU55y9OFQ3><&iYAg$Mq
z=CH`{<f=Xt+Jq0h&l@N%^;}^wBp8b8@}qXx(O@b-Yk0jb=<Nx}D}MdEpX1;D3x5Rw
zu~_(4#Wlg#cYo`9_>Eus6<l9G;^{aLlwm|rP;;>q<Pwnz$qnYFb<|_gefN5~V99ja
zonFr7LtXR%OC5m2qw0S4!;nnCY;{j)GsChxAf@7NodXgKf{3nyU=3ee8gzSyfB9ei
z(<nLOfB8T77x9hPukrdJVSnN*XecSt0g)d5L7*d`H$mOqUc~JVf{V#UX9X^1kim#Y
z_)vI?9PeAad*n<`jnllxFtld)Bs}3CeE1fQSU~{1{dR@@rb9A=3&VzD)CwyLH)dzU
z(awg9Jafc>Fcn;J-7u8?TJ4}U_ec@F8|pA@y&?4rVIar&#o;L^p(Jr4OKaU-S=r$*
zovc7>MQb;_-y8PM!L>*D07AWtIHO1cGb(x`g+3NOqw0`b4hc)lsEzO*;lqJgB31ql
z32(;@M?WY*6*#?gn(rx8sOeGb&=E1*&c@on*c<AtVH}1E!*(#OFpP*k1V<yJ90Kh&
zh8RWyfXLy$0?`0fcCj;N_IG-HMeFQva%nN!POyc{F*zm7Gg(DW4)PdOW$rpoPVI~x
zr1bt?$(VtWieR-er|P}COxps8f_sIQ1TLv2e>w`xI!2&Ud3Fw?5AuIy2BRCARm7l3
z3S^_FxkW*<fga4&Y!OHu0If9Wt>h}wM7Ru{Rb!Y)sblgYR652qW%-1;2KXLQ;mwR8
zpby1922+-d`3C$?@X^Ygui!KsGsEd{nsa1)XN)du2cC|G95%eVKJxRnj$CLLu%-f0
z9%I4Hy_y=Tl0eI_n&4HGvsXfZ`T?fK*1Wq%%<Pof=%}8(-3Gbzp>dOJdq+JwMpeX4
zeME>n(AK4(B!)5dT46e{AN_<8HDqXBv1`JopTF~MY+jZ7wejK?|K`8>|G>Wh!%zRg
zpThften-~pX?<OCIbCm?*~umu${AQGWEYXO=Igi*^jcBF&W$784~yVM&D@k}V@~*Q
zK|=|oY==OrkKg>QU&p&oeiMK97ylmq{9pXbm<nb9_^<zu{#Srk*+22^Z{olF&;8T5
z<V=5_gSIk9YuJy5W(mvHH5d9KDm2mxy*q?B+-gFDxdL2XT){cR^J2@)VG6<~$I<YY
zzVY{QV1_06VPa$gj^UlAvV&9v0LVZ$zaQ$d80?;~SbOJM@BREffSvmk%x(Kl2jIP-
z1?~mC`!L=Fus}#Dj?=zl41Nw%oKIovgh5AZ=*NK?Gmia+=j|Eqo^KQ<hN)q2DzMwe
z_?W}@9R6+_bPz}>Gc!9Clo&Cjk*PEdJe?LgGq0P{*|!d+H=ZcD7-_pHxDh8=RA$CB
z%B|w|TycA<XqDkox255@bXWjtBfIZ7x(~=_`Hg$mB0LLB&1VchpdkN;yuPwaD%G6C
zWJMSimlA!4NJk|e$BRpRu@5ROrv%&7TTJuiWpiKWPG>@#F(^TSF%;Uy`QZjQpiDt1
zn9}G6ipy)##4%@%w0I(tA_y0%xP?ZD5!E`L#)cBoDMG8#pwS?O-brRa$j%55QZW^g
zn_(k<sY6c~hVPZ;YE*SnqHhzfQNlTKD#T6EbfAt(Jk+mu4%wsj{AI!_&^l^7xB>SD
z5wC)CMiW{`T*`vS>jOqJXs=ju#C5&kqt~zS;q?(8JU$>s!OfM&tu@p>m;rHaw{i}G
zSq#o$mY$X#8Uw6T$*mA{(7>!apl(LiO>@jx-9smb%#I{wDsF{z<kl(x!L}c6x4i<I
zf$#uJ^FWw=!IxI(c`&FN{%?Qk&*8uDC;l_|_8<6txZQS$4dhH;n&hEJRDrwAbnn<}
zL+g5aZHY76TJN}RH$sTu%ql6(*pfdB#6fufM{gLt0j4NPu;zr479>Rc=CA%L{<VMo
z-@tdj^PL~^{^>U~A*tW^jql)3|CxUm|LH&WpJatUjA>niFkKXS`Nt4X9G#obbTnm$
zFrh~q9#6~a!tXEmk8|!}fDJv_I>+8H)@iT7?$2h?X~LiB_rE)_WGcE_6VBrXnOq|%
zX0o5@a0)401m^HjW0zx{&l)_dn0yW2TYL~rt%V0-4O&J>2&hLR$>ON!wc>~jyLsLY
zyu0mKOGXe+Idnz$nav?n@6*qj4;<#oF=0&wtCJKC54I2&;+eW=EfFH5o+F%;(h%Y~
z;0H*#?+SOa<;-tXQpC6Cj@wgX(wzafrD3}?<V%FQ#z6<^t(dbBhcJT)q!Z?Bj=-ew
z!=9Pn{`%_4Yv4n{P5?k6HFC`f>%v1}26LNN4S@JClhx|zTv4hz{TSG~$~5@`D<_8^
zXu<2-fN@xN`pPK$!5$dhO~+#5!Ju@ORP7wGsyJ(abk!W0Z>!O;ql5AA^l8xoAd`WR
za)h0DDcjxF2A~zyI{L2IJd=4eR-7}lHsyTkG`bl^U_~yYBbb7zWdpTURHH`^Q{n)T
zNS`HTOWZnLPwRd_g5U^-=i3w3Waz!X67?c82`YwwB^T6Qag2eIBOcckkL!gB<@JIP
zfa7@P^)|Nqtx>SU1Kn8h_d!CzdCc-+&@RloIWwpPsBm^<KCCmJK|qz!s0#OlN^Xge
zu$F=~d&rRS;NR<k*N?Bra0?kDCJbh%-Z!rafJk_`L;=vO@!$Egzkq+{zqDX^eSw<$
z@5JZ_hx^*}j(Qx_0c;0h7T|94V1`>g5N=OgKY`|si>@e|VVUAP@oJye8I=Qb9Y9_n
zX@RzeH*Y`3|L$M?*YNolZ+_rdlNWgXRMHxtir@H+@8a#-C;XFt?2kA`Pbl>`2BLKD
zpr=lX{yJ3H4&U5ZiAyDYnr;Nn+MD9B^Pa-3vDFO_JO29PKS1}g#9kf<{(bN+U;Xo&
zxNp-Kq&s^8Veg8O^mHZ`pS?|gr&dYl?%mK1g(k)oS$BnSH@E3$x6FGeXk(yO8XTGe
zqYK)0;C9<lj|1(X4fjziwpww!9av&^O7G<Hv5=T^nBE6^RsK%JN#C9RXfxq&x<H3a
z2IKW|!BPr89GHbta>gbJFsh;<evpoVW&;h1wm0-d@=DtWZnuuxtzp0MV6q?>m7Zoj
z(4er_;fFle4C7CkdWiw(;Bz?UJ&Ru&#{ENI|9{wA_>FfDg$NGOE;(T>>2yMzgJv(r
z&>Q*Flx7=dvB_0NYXGE4^@LYl#KpLwu=+Wx8mFRid$99>P-WOe5|`tLNbC``p0MsG
zyHgEC0wfub+!8_putdZGq{A>WSPePm7>!kN+ZsC9%J*$xd%vAAh47*KH01)gQ6H^g
zaaCcniY5U?$oKDKcy9HMW{MUYMju$qLf@vNW4j$Vjsv4MSW4(e#qH@Ci_nd=#sjs}
z36^%r4YgJ5$AQrXatLmhJ-Ooy%UM#PZ8st-I%P`R3@T}z;;a<J#XSO1Qj22@5DRos
z?M#H)d&Ab6XF)r5>@JNWRG3q}8P038q=f6T;F=f25P7h<|GjEMN(<HHFKrRNv@8{X
z_&juOw}Mt1-hTHJd^3L?b=V0<(*C)l4lv2?dBg4fhQ2jqNj{9_uW6%Wv;&7W+&Tqz
zuWjI!4Kyt%&OQmM*!zK_9gJoio%B!3c=zTl{;mJ(pTXb##b4s4Cf@BBUK|Cp1G@js
zq@BKd|Ae<+yn$$B?5Pm%x8_zWiAtvl$qqj#4BUW59|)l{Zvg?A@tK%uW`e-LYf@ah
z()OcPybqf*U+hcQ-N3?3eHPUn6(H4_(35Gy?#leu8ZIdjl4+!1#}I(dfGw`oj^m*6
zzh*@SSx4&H7#;UcW$18UD=P7!giVibIMDI5?G9yDi2CfTk<B#=n^|z0PG%qB&{j#)
z_VEk}F;j2T8XcaMaIfUR1L5HjuV25$!)0|6K)~n%jDTC%NZXf)63Cu1Q)pusf~v5+
zp_^fQZg~H8!}HUQde9muqyUTo>q_=oCjmem@@(ClWds+ykq(P4(8Hr|e>C*yt1Ft%
zVdI!nb-)r8q9w&Mw1Y(mvG3{m!=QG@62sly5AI@UCSU4C_`B?6#CT~#PA=R+V=@R{
z@NsE}9S}l7GKPK+oi+o4l&Q+4mLef#fko2$cvf|{o#lZChka<s>J+tUZ_pLTcANuL
za;y#0o|p3g&P`e;-@Ffj3`G}#ri1`<b>e}&?bx+LAUNWIH7`_6JBzB<hF%-tuwdA4
zhWDa~M??+@`+gv%j9LfY?K__LitTyh!h*Q5Sp(30xWz$ab}W_un^h|VLWX<B5N?Qp
z>2wgFr;HpKexq@qhXl2X?I1SXEFmueFM?$!`VY7`iMHX%t;1=cXkzF`M^{pBwGUXR
zK<YSaojFuFH&MCoRG=tG06-Og<*)pI@ZbL@|Eu_WpS`Cso>!!;H<%6FZco^^8@8us
zj5<)(^|Zj|EK^Z?ML#;8_l;CWcTepOi$;%+-21_Up&p!#48_m?{4e0mn>YBSU-^}@
z8aI0Z+xUg~GS2_<AGvA$;xGOx{?O0<0Y+X5(7HhaJA#yE74O&C)x>PN*)2#M9Gn~7
z1auytVZyjPtw!TIYpvru@&(?+&7V{J!280vK~L*uwE-J1Ozprp)>gZF{d=*Y14*O9
zEquO*4rfkrefO@YXsAO`BfHkKOLfe@`y8$+<7j(RCv-6d@#qalI{*N!9%x5JJKR~<
zuZ{U0WPU)5;oej}TN#4`Vf6!}x;NQGH*#k9(z-4vIb+EM0-#?EqXAkKOJ0$~GgOq8
z*Jd!^<V6h94OU4Z?#DnoI_=7RUI_(8Z|L<v7ci&0k8=<HQjhH7JNPURK><;ud?&8_
zqh4R>P~b)6h3Y!Us!>A@z#1b~_a;C6ivl02?t&|-nfU?KZFiw?RT+noL0R2Q8XbI-
zI^lr;sPM$4X9?~!xIoN1SVl5d#fS#gh!Hxa^iGGRZ38#B5)wiLy3kr;9D>md79x5a
zkU~f!rh=F_y|D>BRK;<0>_@}VScy@$;p$nr`>nJ-u<ex=0llFCIC$nD1}b{d8jZjZ
zaKwtyl!4wBu^kQdR$-kqyA~2U1~zo3vPLXvL(T<;fZKLp+beDd4b=8+M~ne$S@3GT
zAP)6hY5D~9;W@L${MiZDsm~H%2C)W60ju(em@=ey9AkR-3u+_%H$+C24(qP8_Cwt|
zQd~|?`lXN{P!exuP=I=&<WZ)I(G+SG+UofToN&wBU1#g)?|ZsN!W4i0&;0`a8~?%|
z$KUzlGu)0H>%JgKM6DHXw)gn-=?P!FeZn{z9xesZ?F7YuFgOQ#zY%7OGb9~~>*azG
z6iE}Z32xgnwt9p13T=x2@c;X_@VEc=FWv?CFEJy#9|12jlFzR@4-OBnGJ|S>4GJl@
zI#6Oj(}0T)yE_5~f?LrGL<PMmVg|yvU#l3ym#^R8tPSQKN)GkC%R4rygm8Yyp3u_(
z#OUE~)aXd1Ek}<X3d2$P{n}{#!lLPJG4C!<7^EeUI;jfn0o|fIe6or+H8K=+kOiiO
z*69SQoeUoe1UsyC=y71&3P!HjYsa<&-OVJ;*Kjb-G}H&zoF#DfH3n^YruN9Z;*KGD
zjHsY27d)(2Ect>Q7o>cF8IYr*q>2&?X|8!ek93!AK3f|{#SlSn1F}SiyVVao1Jsd|
zD1b%OF4&UcsSa#yYB%P*H<6wiX%{X4N(OoaIvUFIqiy$Ju~PsqW;aZg-|GG`DZ~?R
z6Cq}7^1`U;l(09?gq^Lko*4;d^lvx+?@;edz<-~L>KQ)q@-?l6gJ%l_H<3~$FDvB*
z!6}3(Ppe-zwBcE&MyJ$}BCJ+4Gf3jblmeBCIVV8WZ4g+^wn_ofXf$IBbI|A5sln*c
z-v$Ocsx_d~A1R4@rzb*gq8@GzBP)YkB^Z(i?l>Pl7{mso11SnL;>3?ftr!;Ze5-ig
z8urGAtb)}^GR3PkVU2)|f#;_uynXW)Z{NPhz8@!SB7)Nff}MiR{R7?HBHC$-JinbK
z5+l}irFuLunNAF-GTf|;i<n|Zc9}8UH}vT&IVs6QUEpSuAaE~ESVFw0tevh2n}bol
zKKb+wN=*2@U;7ZBJU`<(6x6*zg5ird@9_B-@9_TJ4GEquHawU{iP1NQ;Q(MeHrzJP
zV{qZ+vJ`Aa(OA88yss7i{xAGh{EffyiyTCQ7xWGp-RryaUCKP_W>yzw84j4?i#Koa
zTi^RWe(K|oPt5`6KCWJlfn9(rvPt)5XKOED5;aO&=kp}$??PEQ57%Lw(8^Ri&ui@O
zyoZ7R_@};u-@kv1&^asXO7@d>+MjfD@&#Lj^O-x(Z|^XuX6~nIJ1V=45#(ewjiIQm
zqxH_>@Ps$r62^F7u=1y6O|1>R5?9{0iv3p6_l5v!ta{^cyChdsH$KnDG3{#RwJ>NS
z8&12;oemNy9xZvnQWnqJD1ZSuHJ>|p*QP)m0t^ygYBlT!LywNVqLtt_KT5ARop2&A
zdv^;H9Ez<r3fOShI&of<D6|YDz+iNUkkR<VkM-aF`bvjF^l%IjgK30VbWf9brWf&l
zhf`ESlPiTzkM`lfzlHOGAf=BSY4sgx%H+R1NF+=YPLU9N;0%nWFwhf-i#sqX(38|s
z29Pp>7(!s1+XrAxAj0982yxXA1Qr2JWDm7C5G0~pRxBj}V#uvP<iJsfe@<^X2eIw(
z!c7rIr_9iJO#~g*oBJaHCdso_>9ZfiKbLQ8YJ|)L9_A4ON)}uS@K7=y7W%;00BWaQ
zMXhW>>*1X3L1lty!xwSj(-n~M0V06?w&VHvhPpSH(eeR+i(!&mYjd$Sr5h-jQp4jI
z`Tk6R?iG6}5zCUWW`@rPQ`}_7;dq0XXE{W112RH%-ME1@>*z4NyFKGuuOASUJaBoj
zuw`B+7%S`vKfauVdd2<u+xPf|zxFrr-})E-IR47t`DMJl-7w0Er~N%XfAbb^K6{7T
z`vZ^5;&}-4kO9OkDOf=UpHpl*b3wY&$#m^2)H?RthP@5^zyH!-!{7W{zjPk<ILDc@
z<Nkq%>jjo{&KrIZ7!cso&%Tduy?)IhxUsvVYWcp&vQrIyC<vy=O<hkx8)!L;FVnk?
zH|)iH&VzFhe&FzW@s<soB}wSa^fnGx&Eb)uyl&GrLtI#<-dP6Nxwfh^oq8Jx7XA75
z_X5Y#ofr_@YpJ=#hqg|gk;{Y(f_jW|Q*V{S+`2Uk&+-_~%7Xf=;)J&i?da%@`xK^L
zY5>~qv8Y}rfn3Mnwi%Ionfo=7DJ3LXs5wQHzoRz8QK^+`eKIq>*kR=~orDN04eB@)
z6yhBY$&n5lPFM{fJwGOaWJ692X!`Ao?uWm=(xFfaVHSg-`zCto8O_c)!gS`luO4R&
zboaP7QZV6*$^eMHQ1-oWGW^o(gadrB422E)&kPu0o#x!)L1NSVDFIAnN>PZXIz^8^
zGl1G&=cC&|&I?9wkX%qMgv74P6{R>vu7;2@Zbw6ihjXa;*-m{qI4D$8GN8I}91_T}
zw%!=n6NS-;0kl;x;HVzzVd6<j{+`V}O~5wsvSeHexz;d2Z3FN3hMUXdhjIAZl}wc-
zFDL=np5Eb2peO#*(}qtzdyluz8>w+-XB_1;Hk24JwBuF}pwnz@AHET&bNNF!GaO2e
z_~2o|t98K`6-j`7$v6V@3P`W=Lm@M#)>1&V0Yk?=8Z;Tc_vt72*>C>@^0_n4A^|62
z^b&fR>*s`FOwO(N&;G?<z(4T^e=q*<f8yux*MI5Pu<aGC4!nDN!~6F;`rdI#wAik#
z@~{`+Xalu#@E;LL32qI|H@~5Z?bwk<$D8->@bCVGzl^{7^MB1dgnMPpO_<Na3zQ4+
zje2?&eQC+R#EK30@byPXDPcbjQWAv<-sFGyzawH8rwQU&>8b}-P2)0isAX<&!Ue_v
zc=bWC;)lP?Ka=MH&(QcF)EK!%MeDRq={jhOE>oN25G=Z<2`Xw;>~&1p(K)B^t-`;D
z8~6}cck8^zjP8XOVlfJ`2(+<agwl`e`RPEfj>)%K#JPWz2Q{_Uv2W}!1ZRbY(K}~d
z&78J8xK}yFR67Xw=<J^L?DACMdJG;)y$;mYQT^+tKb8*eYoa_zbao4VJ(wHKjz%5M
z3eJqY-xm&@BLa{TBpg8Jd;U>gU+GZDh4f4{MR3QxoM|YfaG!D`KUv(5NIlAKbOM^e
z;U(*n!|Mxj!b9AtG2tY6`B#{WW4M<@#25j=tsNxTNyk$dWJJf}3@NI+ld^4261@WF
z@0}Q-KwvqcK@lQPju|V+_AtV<Lpud3A}=0xw{D;iD7s@Kpcqi2a+_*aUb`XWj1P;5
zHjrb1q=dxt6V`pWFTp*g2oZ!3Jo%Hd_L<=-LNA^;$Ywwev=A|R(iDk!ycT@?xZ-h1
z2#wCGZ{EMdr*Gck&GSL`RN(;lvMf$9PAItm0vzo?mr8b%O>TPU+(3P9WQrw6ynbBp
zdb!}Jfa$;`Cqxlc<>#^h#F7vfD#a~EC|r{f17a!IpPx}j!{Udgojb-{4Gy_@5iy-f
zzr3~bwH+1z(*O3~z<=+*{m&zngv+|%)2(9PDuZjBJ74R8=k10NiP3JY<E9-Lilt-}
z2}mliMv>mA6o&okFaCY}+OPdOe&Of;8ad&9P`!iz?8Sj_{v~F0_yccPUexE_{-+q7
zlFjNqgn(!yWkh4wzQl|y{cdsL7S7XRjJ1!P2kU$vgKe4%{aBgb%^iREYucg{%MYih
zQ5}lWJ_ocNi_BaDoE?hV^xaY%4#ToD5Hz15bOnY?UcMI;a2UHOf>N&#q96ys8WO50
zt|=lYNeiQQ>|4clI{+88st&Y%p!bRt0>)@)z2ew+Y}-aSMtHvx1lkzTRd>=sW8f^H
zcQ?%(clxRs)#Wy(s=J?u?cC4ITMaowZ{<9sr=XO@GoS6?$2kvc!pM0tnL5K$k?FgY
zB8d_M^!qXedi^ojTr6uuS=?GUu}TlYH3{m2Js0~-%wP3had6&1Kc7?1&D+b(4}jw-
zq&0w7Z{o4=uay^nKgB-5FK`P1*bi7^JSI0$fmndOpy{48ya`7IVuKM8ngAVS{)|Bv
z(iCY=8=`A2M)%g;X@I^_y>$MRIah(Vzee0vBIv;AsY>wgpJ77?5xFGlHrjx>-@Z*M
ztt8|^2q{Kdi=>iZCJ-Y_N0QKz!wfm{D!*PbzV_iGKDuN)7DlAjr-s@(wxiM^RY@aM
zn+aNphh~hNrC38(7byX!4^13XkHFZ<^^);$S#e1T0SUbkB64`uT^|$Dl984Kq{u2a
z20#U6EhqvcJYYL2R*6^&vpf9u{{ZB2UvF%?<@~#|ruDD?SN`=sg|C170Y3Wp1N>uu
z<R1sXia$8bSqJHn!8q)@)rt;yzQhHO%LO@e&O++%z~A|MzlJ~e|NQq+x19us+3CJu
z)Sd5h;@)^!xf!cY`QZoJgvA6--!}LR&7AW+IJ*G_fD;B9-d@f^<mJyk=LR-)Klk@@
zLLXo1RetadvFdM?nH3s*e&pBNt35)_I<tlE8YaT0)C4fGkn;i^>YV`ZTsK-vk3OJn
z&Hx?q8UFbLNhjB}AcTb2E7p`?-B7oR)+)B!j@x$7gvQ}5q74IPodfpKF<QfDop5ao
z2xGddcHEwCST70NZG&1y>m7M%)P;CgG@`SJy3hE0zS25>t~yQIY`{i?ny{<yt_~lx
z07ndt9r>boq1Ot)3n+xwKy^Te&r$s_nQM>~;U9V<f7|OT9SZAXLRol0dri<5ks)LN
zIj>tyPd^~2(Z75QR5SXf4?B&e)M$?|L+1@_cVR*oOJ86B&Wc6&W9df{c`=i+W!2tM
ztD+qPQW(FP61#)e*`Bo#=y?-qL6Aa129PD7NQNMyX-7&7_E1qE7;+3qsqkh>K!@Sz
zoF{M`%!QDqq7ZzrMgXdjMJvKNg&<Jm<qw2ufJi}=;3~^ZXk}D-1qx>)svL-?AP5qn
zf$R^Bg%8dcsKKB)Vid=NA&4b2$M(7)zh0283sO$d-mxfMAuSYQ|DZdg=ed)%7z`Sj
z6&*AnML=R;@|Q4zRYhQjK(e3=pcEi_#%;(MiW_D{9F**xrDUw7c&Zd_O;i+C*g8MP
z09zKkt?%)11~4)^rXRLV**Loec1%;sH8s0B%Lo9#r=PvS_dfj`|KPiy;%9#D=b+tC
zgqa8!idHKEl$IyGA%=tx;)0LYh>w;4i3y)Q-|)%jZ}4yZU;b?z+YTLr*X<mr&K>^6
z`z1~@_xLa4ymJ3{|Kjr>41$key~0m@@PK9lMS!#llq>DpI}4_vcH-Va742v^YDKwp
zm=z2NicR|Vgf{e?{iTQ(B#--%a6jjM;Myyol_<f8f|3haIUz!#aic{?5TmL)u$ot$
z?>PV!7`KYLRcu>DZ}h^VTAuXMAz)!Q6amRK0O0IQnU?~Oz7;4Q*Mx_XkOmL=eXH1>
z7=$&c?q;3`qQiZnxbgSlnx$k&CE##W<Y^xmZ+GO!2Rv;D3<F&au{K->2hr`QT+bn4
zi~wjjbs23VBTbw7X77q@<9)i-1XaPlE9#-ndgFR9_p)l;kR!VswFa{eScF-Cc0n7Y
zDpL(E7~b8t(NUe6`#Zh9(xGs9xMEq^_O)PWFchKH@N^BGR=ji2Jm7vfohf3oGa&TG
z@$xw<09^L&Rmx;zOeTo|D#nU9fUc9p6{>@6s5s;lBWQ{zY^|Z{YF!{L0Wp*DlOV8=
zP7H6D(QdkvpHH;H(=M$039o)?k9+!&$4<@=bn@?g0OX=8jkL=Xfky}uZtF&`oTP#%
zFHVr;S+M=iorcoZx<)}&L4(jLTLsk&Tkm*k9aRS`fi};<Lt-uq(y}10JQV67uoy5P
zo*isF1p*b^I__Ro@CqN|LYfEwfdg(RL>4?p(rrsgh#msM+Q99PU6wB}<RB<9prk2y
zWJf~#5T7&!V%F6Lo{tU4SB9&68TarF%l}Tz?G<^AnNCI(?MuA`KyRnFpTEU__#gab
zeC?axz|a1Hp9WGwj|nvdWbn{dRjeiAdRcH;3YO)9cef3H=Kuch;&1+~U%I>Ih8G>e
z*%3@Z9bYEo!}CGV^Zm&5rSWdtzxqYT!Gdd6nQ*&@)(4JSag>ClgY?aQzy@7r=ODJV
z4vC#bN9IPTI?k%{G`vti27Q}Kn4}T!9PGFAS0?Tsl)4^Kg9Tpmq3$`>NvX{-o(XL_
zNG}=MUBBcriOvd}*7fiM)PMJIhXS8LG`FzOMnXs_1-TSNG3FSAh<!hhB;e_`;mzCk
z*tP@btWTM0*IS36(k-AhI{Y37#j|deK9u{IJ`X%S-SGVW8Nn7HMAX_)YdyoYdSh`g
zVIG0FpqHP!564C1-i5W^89oKY-Y9e>F@Vecv<-q^C!=rK7(^E&(Dd@|a_sT8E>tuT
zPyha%UtbMAx-KY-Ic?GqOc6R61kgDzAk+~AWYmCOIKy#5^`XoR@1D(l^wN~o(;X`M
z(!AM$R}mnVh_V*sg>!(ucM8g!5*j0`3Jc*}Si|iwOart>gcWEqP^}}NV)%|d#y}sW
zr0HPm9bB7WAdh?B5A4qy_UC7eqft)*)^rRW37)1F*u_!#<^k~Z-D#s^%w8}MpJhrM
zalmPG#V(MakEvOLy&IAQJZPc_P6Q1BEkv|vXmKz)(N)YjN8~j_at2byh!IU14&%Y%
zaI$0aelO@T0)TTE3j!n{E)g*iR)a7TBN`aU6N!E19AV&qyAI~j_CeE=OUYQ*jLRjX
ztQjGPGlfrLfT~+J_=cn~ylV~8symH-d2vrz=u3aCdO{heFzbYr<b>2?AWr+)_r8yQ
z?D)B#`#pH|>VoLs6QC$juqHtXz~~i!_TT$U_`ARS>-hV>_8a~)s1E0Q@%J}7Uw>co
zxtO6hv@yQVLZ%nji%&&}!w(J#^x5hFX`T+pj>dGoYvti;23YSvA2<%iclJ&%CN=+^
z2xo;nyM<YR1tZ+E=8Ul}_$TU5;P2^o@muotw_W%1Z7Q#Ucq(6T>hdg_rd@ZA%w_i5
z0~2$)Z^LnP97m@GxEM^7x&=4E3;@gbL8Z09i<m(N*ciMRiOf3=ksBmoo`BEZJmJ$X
z-eJ4#kR_kF4Wn}QA=Ys0JDzU`p58xW+p1Udz{D%YU=H(98|u+eqd;;-ZVc^dm8o$W
z3=-j<Q!KP*22GvC9f0TstN8q6s}0O_8LsQ`0>J-&S7q)x<h0DjTvKqyqY5H`5&}|)
zD3LI<D#d<3%Im9HXXgY<^zDh6ElMyT;hs=6vuTIZ#p)RRkWLew9BP{t$ZQmXoD{D~
z!O|Bef&^08Ab_NZaw)hxE~G*><y68SA{ow;?g1&$^T=Q)2G{!_BVib{2{Z|M6{Osd
zd|}r%d^p6eKu3dSe&8_%j@vvG4vJ>{=7i|xfF6DmBd>`lJDii2=N-}v9#X)wbs4te
zs4O8l2yuL8gsAXf%aI2{6CkGuiw5ls0AgQnm^s)?fFhiAfCtV~SG;YCr*5e3GsVhC
z?gRd*Iu8NJ0eJ!9LeD4{R6sRCJb=8Eh~7Db6EU=2u^&5b`$5S*1lOGLPzsi%fSKrq
zt{tH_l%)VrDr1#=a23ODiqa1%JiRG6SBJn0i95tNJ?zg~OoUzNS&8#hNGafY;n(z|
z;urt+-^bto<zL5t<xl))QH`^%(FdNlhUf2pjz9fpf8H~*x#dTnOPM>)oreNA8#-%#
z9rrPr(?&7h--}N9J}2?7@#;K<fIsv{ejg4ANMA&#DsrN`t{wwg2GpOxWB~~gy;U^N
z$?m-&dw7vbLeLHyOyA>YU>6fbu*3wDfOx6+>HQUcpM8Yy$~%W}zI-%D5Tr!mVO$bB
zcamO6>kWbuBb|aBNgzBv(7~A5Ry&TZ;r4u>w&A9Ceo%(lsB)gCW8m0#1c6gs9rp6i
zF%LI^VH^$ZFx+a#_db7*&)+_y^?@8C`rffWHzomY9ZedZpPupjwBz|{!`M5Se}kQT
zvw=PcZ|a~BuN@u1U0X*TCp6ST<6zf~{YF=6CK~s`I`q8f;WW_`9F3cQo9T@w`Z!G~
zIioNMr$<xp4umRuAOxlYmXHt}R_9}^__5?Eq+B4OpMw+n{xhO51?OYGz<d>krUQp|
znx%2;b|vGinz#|E`)5qX^p|LcFIw^#10+XATb4-qxv?u@S?L3;E*KT)(U4;HS%kqX
zz%YPrc&ZI44zyyZWgsR&%7R=nqY|4_M`duR?##=f<S~y%OI;i6lHeMS7`%!GZ!Gh3
zflw6DGF9giP*R48pt`(Y5GX|RYh&Kdw`MrDids1c6B)1+aajVoIz%IYIwU0n9;-OK
zE(H3l+Km?7dk5ZaJHB{6@b1>JHMeW<hmkOtBl2djSqb{}NjZ}(rd$h}Y0F&-4Y^7t
zMRJIu4K8#TkcFAPC37>Od0K-agOEe>D)tVoSP)bgT0gMw^}ZqbH4;B;nWyW5D~B79
zJi2lo2Iqm`TTL)zkInR-3hjzN^=JNH=lVOX1l<28I7xxOGC+)COnWic;JUr*9L5;u
za`px81p@8|7w%th9&!HfJsN)bznjrsCp`Wg!x)0{9~d5gMe)I!@M_6;NWq0*0aXC)
z4Wk|~Z5Zu<jf7zdy_3y0JV!zT?P)CV@L4j`5;Y0x67k26e;ogj{3eBz=Nt1+BL)eC
zYjP&RAc>(Y5v?1N^1MzFNWmZ&Pz&gTI*8-gQICq_=-4(2*O-kWBvOQkHUv%eWGc2!
zg2&Mt#!y7@Oz3G{0VgLyu-5}`pLcxm{)Xqh@gE1%0r%&M<GG?<J3;{3-f;MAqtyY?
zfbMfoGjZse51Vz@EClWo$KKGcgu);Zk_<M4nyC)=Tyn$9(Y;%kCZ8ZICM2<pPUcd8
zaHcm&6y{8B!QF<3I~y61UhIj&%|HR`<1`?|kJc!BMM`Q)&>-m2y~ZKK0m7=3nW>z5
zQ?x=xLHB{If>par`&+m%Ta2D^Mrh<d9&|qy_4I_`X=h6zlyzNzsEjYvfL`6(Uk645
zy0>w%h*u{zB?e?o=s*L|_KsT&$Qj6M!u4`Fji#X7TD!42NQtbI!#VE5dBvFb&f$Tx
z9rq0=20@7#*SsPpuI{GJ*lBwDm;mA?a^MlFR6e%mo=3`%mMHK8i&}&LXFa*^iMYM@
z7y~8=iiQ#bQ{J2|_-@<q{`tVu%`G~#v(_{~F#Y9cxx%U>C3?zdj}wI|2I@I;;5;Xy
zB+^ei;QmQ`uwsl@Qo>RSa*B*92o7HeZL9%qyc0e&7!S@-vEQE0>%=|N!<ZZfIfuN)
z%4&)c;V|MkAeq~)|NQeM^QZdL>To6hE#h7$-z_L6Mh!S!dUjUMz?)2)slq1X?Ik<v
zr79ce-@SMX_@jUH_u%LMnF0%fxBn^refcj&{E`1f#!rlh53UO`fa9nbYS_Ynwu;d!
z&>Ju!4{$TI<}<FL2yV>|c(<cuZiT>1hLR!*5~TUswmWN(=bmBo3PVClz>*VURC->;
zfYdlBj)|8<wF4a(jtou9_T#A7whgTfTynv>WIS9FVoca7U4kVEY9ljjsN&X0jgBcH
zhJ4O5%}BBc0@R~oKN|MRfErSQ1v&(EXLoX2H%Kr%KizO_JL>H~uMMIRb<onDYna_<
zcYBx-00{%FHXPdw2O1Czi1pK8ueC-8-lL&9yJ}`q19QW+Q|mNo%M<oAqamRBY*nEM
zL6BoOpAT51Wrwg>Dv@NBN}P$aq920({jk?pIuzdT4T=FA22>yqZ^^;Rql;l!M|NF&
zABv~v9k=Zsttet_nQ*6J5+$3?sY*NA962M9$u{7F*rOn1LA)e{YeI;M)CIbCEK7t~
zMyuUnA_H~~x@h4{C`AI*kk%>7pP@1%M~nuDRbqlpZ5+cXWuqfXWM1oY#eSp5OEAs{
z=!p|fI3YN{IE9E;>w=F;#789{FA2kd=L7WeH_*w|H^ouCYImM|-v{b(oE1hGLO-TJ
z{GYVK4zoERrg%>nWDS@i7-^SPci2&J>>Y33ZTS4nGoJSfQ$@}ho>k8IKw#TxPFsu)
zx1{XhJCSP5-d(KZ)m%Zq>f!37I|}MB9HXO%p@a-UM#>2Y{G2rMFig&BGr|Kni2{aU
z0oHfWhkpje%w5#GgV}aZx{Mj7Z`S$x^8V|8CojJ4g>^M(82H=1`GittM>xOV2k-Yc
zc3NA{4T+tHoClad-1l0}_kF@n0APZ@`aiic+zIpRfc|?De&$;R-`Fp>URFeH*lvO`
z1n+mM$Vacxc3{zhbnFOAM(EQ<M39G|1g_VJfD#M|!~bI>BJ7Hf>J_(G@ukDm2=gBz
zxWL$HC!mxCDe&MaONJ$Oh{2z!<ORT?4caSe-EliMJnsh}z>*A?OT_gO5sLsl;<%A^
zu0GFG1-9crYX>CItv8nF6}S*0Y8|z^%Wrj|ni*2$J~O(a9~E`qAP9K>`GKc*8@8>Z
zb<PBu8ER!#wrc0xW;E2JA|Ns8uv3T8!EV67jt#A%-=5L-9oxR+7>cT#6->ml3Eabt
z&{`Y=Q3CozT%2$t-lCZRh!he_a3n(MhN#+by{veASa~RSgB<=Wgdg!lz*jsJdLL+7
zp+`lmiWunz7$9zgoJd9U#XOwaS8F}F%o1jW_wqF>vfzWnv%BVIZd^<V!Whh@V7aVV
z)&(?HFi3Kbc*hf=06K+K+5<+!X=NdimBdg2T0<GW6M`Xh!Ljd%nKx+wENNhFz3lOV
zhX>s|n`U-B6A}YB42oum04^!xS~Aw05JSWe(#KM5$j0=qF$ND5_2IVNRf|lRa{J;*
z7gb`#tpNyt24fWFaE0Vq(#`hM=RBRQ^Sh^K?Awkolu2x}9Tme`A{KX(jcE!zIM8D-
ziLkl>;lIgV5(siiSlubIp)xtG5jzeRxZxWIiXjjZoWV@<XGZw&yOXnSYS<a+2jEUt
z;BzB$9)5?EM4TG<0|&yF*4%vkKnF2dAu?G)Kk%(T>`>t^r)YDw>3$GQrT$F5n);ZN
zffBw<vz=N7{`Y|2`;*_mcVDk~ColBTzRo^+{fMuZ3oXV6JDE}qX{&^Lx+2tp+!SSD
zM;Vj@e>EU#zjp_SNFl*Sh4lup4q(8v<Bz@jNqpzyw+`2xz#g8;fzPfRUdu}MQ6w|)
z-nH^xPX2-lGfVp59BFR{-XAwS9~=7M=3bVF%QfSA<?L#7AP-|I;-H_==mWLx(B7GH
z7w;Ax6Db*HCx!WBFbWW&&o`V2HNBbYaiH!ybqX#vqZ!}ue2@r<8H#|d8z3EPVz^Q8
zwW>#BVHN_$QL*0+>LYFkj=iHF>Wm>SBr%T$7*t@3@pI^eEFFC?@gj9v&6kW63~^e}
z3t{o4T=4o*utpBA;jRYiKl4X<eKkWol6M_LB|ev8KmzG_jJf1u5ut{ldOmAwomIlw
zQX6Kh$lWs7A7<eOZ#HSIUX~;VU*{z`HIeu}%pjR|3>G?6<dlm;1sW7O9KQ|XahbD4
z_dipvu8fZ~knv$KPp_h!U*OpR)urq%`VoWc6L_jmAs4-#1~rTRJG2`RNihpXXWPW-
zI+6i7a)`@VL#Mg*;aRXa?GtnmnrkW2Y)}+iSM1GUG{0{KQcn9&9Bn}B!1G>l90NfE
z9#R1=>HG{51WOdGOUANhWT%*hsiqDdygn;N9cSmqwl<yiz&QqFArqw521vrKR(v1}
z#elBs@MFt;V*K89^1Ibd*50Ed3xk~YdWQn$y0H_d7fgSj`(Dcp{|9X(zC<AfzkXK#
zWU$T$$<BX2|L6Im=O#Xl73U1%u3%Nwb0aqeEc#qxhOR(PHTL=QjF8<Q{F!g#_He;_
zxzAIA0A&Px<AX<h7z-pt><!pQL&z1HfL2KX4o64s4aG7d6mh)x00Y!!KmZHqov_z%
zWq0of#;C{<NJ4GO%RcHkSA0^Q@e}O<P5XHms5l-Tc(9oO77PkzZ#Px+QE_V<o{t-z
zwheowTW`q;OBSq`h!7KwsuYbXU5twgFb1M2mdKdg968JCjT=8Ag_Y`q|0t7L7f+-j
zN@VVeO+z{5Ah>&Al$!TTYyz{BNX!)5en1;%V%J)sy+eCP@71AOL*I8i-)?xiu?wol
z;J(of!}pq@0}4ToK*_>61On;^XtVm8^@=+p#4Nb15tq!lOOEtif+%vzcy)=0=H`J~
zF`R_c&uZl#{rXCWLOaxRlM`Y9a!4p4B8eN+!On+fdWf{v-8+W)tQO+grJKU3cvH=L
zKOKWpRAne23X=#)TmbYhnw}pCNPsK`T?=sdCOW$F?)d}7K<rup3F;Q!&ddoy$l09?
z#RS^L+cVBVhI77yaZ~TkG?<l2m{{@43nrxMEis`)k1>^i?x8fzW$`dpQ#OE{fHje#
zIUuN&j1U6_Y_8%!416@V*3fJKN5DX26|Ovxe2R%_%hX_ia4@zVjcxHTEFt2$W)vUf
zTM(24JYE(oOF=BjnMQYO;ppzEHU=@%*3br-5g`VzkjZedp|m6zfJM+ABN8f#QM`h7
zTpK)OM{DTbVUEg}P=(+Re)}i!&;N=49RAOL=HK%Hpq=gbIk=U3ve-YYsOJKFiH^^&
zobZU@bX$CR0{laYE%V<GP|P`v*ybEzW{gav$;?iSm(AQfw7{avVUXYdli$LB{-658
z_#5(TU;c$6_~-uo*YNFjLAXX}4A{}2Q8BW?G6@O|9j#YrZ`?>#ktY+}pFwnI!?@3u
z1|E#6bOP3(_*S^$58u9yzkdDxd5QQ08NP@&JlBfaE83_C2`rw%fClDPOMqLb8!&_f
zlzQN{-SBj~;rV&TzBiZ|QWjj+jMrCoA5|NcU<lgK>IM_T8i2==q00(w48Cv50tJW=
z35Y>CzlsS<Vmf34H4|$}D3OPPjgG!o)Y@=#A|g=)mnHji*bsqp#`3ThT$TtE!nUIw
z&|cAw9earw#=Ym*Hr%!wp0^EqJx-C}7&_UmAh{&s)td3qnh|v1*0~?{jfc~Hm2lR!
zmVn2H1s_}^)*^_a`2W%N@6oqyTY3=m^*+X&>$mql=f0|vs!HXZdV~NW<Uvdtn!<p=
z5G05K20;XjAVNdZfToBx5fud#ur+KTiXae_M=+vM5{iUSfrRjspwI{@DwV2Kr5>sK
z*!#EE9HV#t_`W{o{Oxm2-FyDH>$Y?EYyH-mbIdXN<Lh7lx_gINV%s)+@YIpV!1Li4
z%CfK7=U01uDyJ~V1ct$niJs7mVZ){=W2Bn%n~#II+g*+!CdJogn1<1xU7eN*OKE>V
zT>NRvVVD_Z8#Y22%{ipJ1xRmI6zmT8-q1%^=d@^Bp!Bk(J~J526}OS7gsT8}8T{cd
zBTFwMbcic`Oyn4AZf}!ZH}rGc&GG8hj#p1R+#8OWcy~}U=y@D?HwNC_IiM%z1n2o&
zhS)^4!<T&T1gAgA`C7w$2F2sdk=8NOFd<PgGr)op8?6!loD=uQfiJ(k<L$c#@JQUY
zhTFd3!+n#~HH8B|cy+_JJL@Z;MUSj3gzFk3*qC?>sSR*`HzPx)ypKcTV*!LuMvD%h
zk-}Teb#AO99WylST~NXO!8x_=`0`i2yaqYBHhLr-NwJGofn2$up)!#R{F2u*a{jI|
zT`a?#u%FDcRPEv#KN}?%mcU((3+#ghPgP;J=AAE4MwwUd4ZrG_{(N*zB&_c$7sj;)
zW(y2A3<pMQi0-i6(PqQ!hIcmboD;9JqI@|dr7byZ)%HpuA>43eU>Jbr*q%DJA~@Fd
zgAVVHeE!S$6X#@)O$>~5ST-_ua7v{|3_LVI>M;i1y?u+fUw(_P{KPxF{qh~-NNmQ;
zY446*wDs5B@u6fzdJgP%;Dc?$n<r+h$AfUr^AVWCkznIiIHlXW1G_8O3d*6e|1!{W
zst|v`Vj#yIn*~1qwBa+i9k<Qlc2nO`Nxg4q8)aF-fHzyio7S)+5s!n=RRrRG2aX5Y
z7}x{Y4R~O%2TF>p0YlG;9UDHob!>2KE%5d@@a1RK=@RS{1=(RAbjRm!j?ca7c(p5#
zXq)D;G~7CHKLQpL^B5dcIxiA2U*+}bD5=J|y)j7zi=OZs2c>MSEvjivODo-v8S3nF
zK?BGoe#g|B1!ybyNt?B*?dJVt9yLjI!(a!STtbm$<YdvNJ6dmSgc0Apg!oQYkTA#O
z3{&*HAPFvto-~~RH19x5Hm3w5Ute&AB(Gf%lCbX>0{|y0%^Z092)sK6o`W-gpU1#s
z49o`;#am-P!W2v7t|+_&wgTn27}33bJg|X>Xo8B9P<J`JYusXuPnG8W<gD7q<DfL~
z3`E!5Y6G^7eI&bZ%Wb33ShS{MSOc6m=tDwdWP}js&BjRS5#p4O#Iv;Nk%?xF>F#py
zVxq~=OzHJ-HbV+KW)tDNc;H8W^ha?#p5d0thFn9}ugbcYP_|z0r429t_58Pd@qfx<
z%Yc63%K$qKEz8KPk5#`-Yi&VMYabTOrE5PO1-Z|E>0j`l$DjOd--^e(p<0TZhEbQl
zPeXWOnzJU=9sTtUw+TEw67V}(ON`BMG;UBrY86zVsCaS;6RBFx3|8x=C5af#@bkX(
zIs8HUG5kn-{>Tct>Up8YI#^0^H&88v!9pGYU>pY?&(9e52kvj5aesS94xrsgQt?#n
z!wj$Y4WE7egq#y^UTygB)sD~JIzD{0!OZdK4toa1cmSHmUPJ7+-mo>`X>&|$*b~@Z
z-EixU-3=`RJ|`>#djmeWIqbtH++g_pe#4t5$F{MnxmS&<QH^#a1Gn3@;e*!jWDd-L
z9RofCZBFc&_@FtSw~p8w?$55ZC|xQJmD=6#!REMmgSEhy9tVDuF`Z&p26~$tKJ<pq
z^@cC(JDzrvj@XXvMulZhz@7&K_c)<c|4H}o`TA5&;VWNy%i-#9+?t^yVZ-2ilO_)h
z_mmF(BRRT`vf6>iOl;^>#8gEgwJ)i+5{m;~Osd{kxbvBUs6<d1iRW<#jhVsV*nLCa
z20C^cPBR?OgHTo^_RfO%j8i{8r7+ps&hNEAqOWe=g#$xcKAB)&X$odyBnl%nGA&@%
ziE(>hVt~)@?g#^(XW&bv=bnZ!4`e*hd&3b8>4^mAR=hh5Z-?RC<KS_1tSERKSOxg(
zP9C?>Da}dR+U8^j&0%z&9T@kCw_iE%cupM81IN1qeFi?;8jg)}9~%-++YM<c7>@>P
zHbYBwfr3-@p6?HQ`Ry~l{LDGf7K~aS-ZAb2KmMf$zVwv`9(P{w=Iqb-@M)tANZ-&l
zI`RaYgN!~E9l&7?)^?1X$afF?vFCUAmY?y>_?bWboA8Ie?*|EwiG{Srz?Ok8C*c>1
z|K7Qfna0b8Vr2^JUthi(*T4Ok;$6|Hy^#&G?VRB}jswV=vjO(CG+&Te)o7Y9?^hqb
z##fMd2K#*K%vb{7@gKdy&;JiTz`Li8Y$On%H@Mw!i@+Cf!*OH(!K>Q_f3*R-YaU8S
zJeV>x>9(M+0J8+rumfR^wrz;<z#HGNvDX_{$7@LRK_kS%+ySJcrPL4Hk@G;yARHWl
z{g`+f2cE_qZyfmCzM;P|ym{sL#@8L6>%fx%y>006K<^3LH+=4O!)IR=X@lBY78Ctp
zYr0)MOEcis9G`pr3M~?EAJ0fLY)?=4h7Wdp{&mNjTcZ}!G-k>l+%TRDw{63lZNsaL
z6zbN%j1I18fFJ}mCwz0<TA<BmiSaoze+)E#ptlENOAH6T>;_=RjAwOX6IpomYG>Dc
zPCPoBPuFlM5OVII^h$TbYd3sh@3`H%oV{{YC!yrgj6D)F5rdu1@ijIod@85#_AAer
zd*H*Z!<w<O9h|^cP!d2I<NctH$c!oYhi&+-1U4$lR?w2l!c#1qda+<d9rZq0IGZ!z
zuE8CV@Wu^l&CzW`>l@$=N8XWU7|#X|Hiu<!Q#Ka`%&BYSGPv$4sLDBAC%h#XaTm*p
z^$0A+%#u(5Bk7ER=Wx6|I(ASr?)m<}yCd-Yd|)OYe&$4*6I<ITCT5V{3Y%4q$Tfrm
z4S2Ytb;23}NS-$l78o)3_dS5##j4`a@eIOx_jiNMHU|fryp6;cozfd0w1n@cUQa*=
zJyY~tju}0IPVL?Ez*{+*IXSoXet&Q-_Hp1#Um<2ZXJC`o(W@IL*lilFQ<gj})lseb
zi{OKT_rp1o^bh~=kKmX5N#Bm|`~DwhmRB#?sk5v2e4RGmE`M|WdHs8_RK7;0#wWe1
z@ZEby5lnk;sHReTj}_y}U`={YKmR*_4nFvX&*7Z`?_#1lSL&()Y)FI!*;i2U)D5rg
z6<TY^+l0l$>&BW<cR+JanVCb5I7JHAHL<S2e>a<Gp4g2+!He4c<<&B<<$)VEbT~|z
zDGX!>MW7|%!WA<^BXu{}n;YJI1MvLe#M9pJh0ndh=U>0ZzHhMJu^VucmH*(j<7sb%
zml~x%ECXXa7DSCEvH=2pYk0y2$1B_({EXYH8$S2u32$zl!EI~sTcYEJxi@46_MLUW
zy>XVi%f7JILRfl_z&;1%WEx$n#YC^U6b5YF(YJ=@z2j+fbkj4)6d5G|6`jMV2>Uz&
zZ;udCA^3Mr%)@}_@MLovGSOjZ-qD@1`C;HX1RXL83RlqhwY)x+Q>cvkxDU+UVEu+J
zV^X`d3K}djPej_vBD|9!z<`23Kx6Fd5Y6O56l}xhD*!I-oH20Rk4tx6l3_F_KYvr5
zqdA_;VC}%x8-kO+*EPD48{D^sF$Y;QPdnP)u-&$m(gRor;W0VEXu(`LVK}pnY8_-M
zo;V_b$3VP$#%=C!GkoRkGrl}POp%uk&qND$4fg<i@b4&l;YhGwfzl6#2*)ugLe{;(
z`v!N?P?zj33}%ypW)%qr*vo*9sn3||GkAL+=<N>QnC&GrS31~t;+eplu-=hsio1`*
zm*3v;6JKN#oEv*HJRH*oj^_v7zB};t-GLc_Tj8F|=&MUqmsikOjgmA$k_j^3?qlLB
zk@(WPXZ*Zx`~p7n;T!z%pZGF9iv8$`EKHw&zesKu=dA~K+2H4|)sptIwd|`qi|c@>
zRg+RyXGM`+bXmJdIcQ>6Yxo)8_N~}H_W>TyZ}Aptp3KvE&7@Eu87WO7f@HrvWyiJy
zX$M+5UVF!z8zbumtw$u-WuT-521s54U}o9fVd3ZwWFthU3G8Q{pyj?H@fwLYX6UtA
zc4kdYb*=_N#nT%$?|8bs!e>9*@cjD5tjZgno^I&3eW7~c9YgDO+ZuLDGz1U>8puFx
zr+M(HIbg~-OBAk!W0xfTn>SB*^R(l2ceI?GKVrbEtsytUnXS_SZmpqp)*xFm-9auP
z!8w^RGd&rwIW-kBvFX|N&GAmsH}0BypeH$1^BgnqNFcqjmNU6uIKj~s_y#;F8?rgj
zlJHUsU;|}z!YGm5M#2Z7!RC&+YF}Ud^{Jggs&a1GU=^mNG~Jrl_M)*7rkF-dvTt2C
z2aJQPlx8g8P$3hrIq{fznLA6bpEKac#9?Hy%sv5epoei2Zwq(eEPz+9`RD0|M>4BP
zurA^5*mg&+;?C`c{^}+<2h9%v7$F;b0;t8NX-fd%rL4Xq6*v7XW>O>mz`J8$-v*jD
zJf0tTW<{2T^Hu{wT;hQZM>ck7XCPUTh2=r2YqAY~w1l^W9XcS=xdfn$O`1<)k|3FB
zlv?SbQGO}MG$L>xq~y)XV)A1g=&?~(rgxO^R!74yW=-$zcYN{fJAC==Gd2et4KA4N
zD_^<eOJBO<-MfRrf*l-$vNc%0u+UOKqG4h2t#M7PHK0+0e&$@)`Mp2zL--B9{@36i
z_(%WQwLQy(uVNJxTNib;oWc8NV_{AwB=S+m@!mmr@gaHr$&a`gU4r%pjjLY*p;e+*
z!TOuO^{3!he%CK0WHuUp@LLYNdwqT%IqMo~g@A>adI-3JrOl1ADiU~V4X@gU&0VEp
z;`Xey0ma&C+P`6jo<R4&>`pwpms9xoRp|-r0G`rtW0xAwj#HQtP0F<aM!D|T&5->y
z;<Y5KF+~V;q}8ZVJ+h6kRZD}_e1gU%)EEbG^r%N!Gg6Fl5xrwGz|*mB9Z$C#HcvL8
z9W17kvDRRX5+BnU=@ZUPv<O7n5&)e*V!vdJnaG?=^snrm8zI3}$d^N8Nr)LnPD$=)
zA9+!L2b<%%15eFZyzZ{G=RLgMS#-ZyLz@YofE%-d1R=B>6E=v5n1XN4=KeasM;Q}y
zfQ%6XsJU@A=QQ&*c;qoA86%N^sYtvvw5^k(DEjHrKe`k`s>V=ig4`wmB&};TO=yr>
z>dnx5!yFn&Zpqg7?$~#B4`T)%63?1uu<q!uc04`p*tU+_s~aW^#|Y%DW7{_v*i3?)
zcsw4+d(BaW<boN&wgwHN5gpUGVUUf3LtwZ|)s)kwa`FA!RzhJ0SfXugB7^CkMU(MV
za2LU8N{3ON`bZ=K-hhS<sHQOP1K9(J26RGfi5hgy=fftiDfZRI0J22rLe9T6ir*R4
zPcbR7=SjuT`~8{J!p9x2Hpf!{k0THx@s)QEJl?Sv!raidhQ4!?n=-}_1Ucro2*&Y1
znF^!U9N9UlbejW*8{UnHANb=xj+$QeF{XG$)k`j54nPdCQ<&nizsq{(<@1U>0YWMt
z4YsaOQ#q86{+2$*C2MRaOl6XWNxs`!!*~7aU&i%Lb{c>0cfP}yKRbjv$iWq+HUq%1
z?lY59!W!5NwY%e$H>qY)ITM0W!&uOrSyJOrQyr%QPGb{Rf=ysYUMbg?v<1*8N-<;<
zc%h1<%@9&KsJYxN4cT@?FHkJknq&kX#{<VfW65%3d3mnYRCGN>Xo*><jFHKh6pBM5
zq}jbMk^={3vG06JjiFV2rxGba!cdn!b8qU3rS#RZ{)=F-7mjoaa;8WSkg<P$3_L$3
z9)}1Lq9iv|hRE;o1l}~mH+`_<gS}zSj^+vYhTaUHd)n~f)^T%+A40a00H8DZ3pTsP
zoFvl#%0@q_k@My2Qw!p42I;{k)eoHydxkjd5o8t|qFq&=fRl03w~mPeKB*8|bU|C*
z8h~^|5*$D1?QEWcM5)HXG%RPpr(=$RHX8aq0U5Z8fM9@khwmNnnB>Pl?bzSE#?#Y=
z+l`V7&-a10JKCF9*tT7*@B@ejR|n>tY^tk`THWKVoPtETIKj*?AAu4cIg+$f2o5s?
z_I;<g8~|?|Z`#aYa-4=(V*tDuDwEJt4dvN9)g&|-8BAc#0|-Y05Yo91X;X#ddt*#?
zTzBHCU9%vXWX=25v2D~)iJ)wP5$^)##2gQd#~sJ<jF<;<+kqs{`|)6--2LqxImtTe
zt)Xuld~58+o&(2lJPvNw-qaiz9O*iRr3V7&i8g?43B=Ugi!b53e%W{6cmDqWNI_W{
zZ5qzS{RfA=OzGs9Z7(RGF9sgGa4IMC#Q#5y*NgXNzWC`vtm;z@Nf=CUmYtt|F@y<t
z!-t>w06+KVek)aVM<B;M?<-G3oRRF2nUtfRNKv~J-B>)mwT@eBxWP4nTud#nxY-1A
znm$)%28?c{RGkch`jM~^$n9ecLjvFZEkA@`@b()ht#H%45RiGqnhA})M~$qdc;wtt
z!?aG~aO7O}U>wI%9;g7+OyhqQ;1H_j?;{P?A>5yPU{20Dfpch2Xe~ldr-INf_zhqW
zR6X07WMw75mdy-OJ>`@$lSOsc+0(j)@!yYwdSmy4($^fE$ImZ_t7>f@Y>wU?Z(iLv
zB6i++?@l`GtBw5=H!=5WU(O7Pa4KMkywLI$;C(&VM>RyIyNUbj*o+ZWjET33K1arC
z9cOdV`CodV@7!|7JYX<*>qv3d-2k7CEjmK*Qrp{NIGF)4nT;f-jF`z$XpBSv-jh&{
z0TAyS?y%12tTjV(&OWf-$nNM(vzl2BHv@d@e2613{lN%(CNNa@Fw#}G-pQshAdT#*
zcN*gtkFhL8mu!TmLtR4LaGlw~5$b-;#5f3}m4H@EO>$LHJZE4$4#96?CxLl%$wrtv
zzYC_8c#Rn}9^8h+=1oQ5hFfpg1fOxz9B>~4TQndY%^D&GMm{j-K|Q8PIgdOZcr>!X
z9!KE*pq4*ULr|<?Z#(wx)QOptUwGUPRMFFn0*E81C&O(%VL3%np4d9^^VWB~{>*21
z&lTa95x3LG(2kn2sDS%Kded>T3){2|{>yZrioqs~8*)JFXI}g+QN@cFj$wHVzRS!}
zs;hu4!LwiYYyWfH&-Viu2W*6jzL)ROpZA>jAejMVa=t>t*(H99M0a3oHPMg1mX1!;
zKB)VNDVTwUOb<$g&oO}Gz<eIaJp`%H*_JB7vrRni56pusqozVN>K*E_F%7>V!*D3@
zHx<|!Xhjl+Y%a1yCQQSEQp69=Cs2XjNc)ChP)`OSb50K2aS~j5HwrOsVh{nCF@;hz
z*0<!uECtIbV|NqoKNRIpWqKiJ?6H!k0oAX}k>em;{-GJo2|{U;k$Y%%%B$x1aCdy+
zgBwwgIdD7@V?I$y!6b3fH@w~wd*|9)P6m(w1_s9gi{Zr#;;f10*Moh8tR4e;okb~)
zAeULGW=Pd<y5`19_CYL9bZ=<R*$kxqv1_-RA;3naY-ApJGY*g6I%-DbutE8YsP8tg
zAHd)<@rZ$Z3|RLU2RPNS>7Fv0eaY1(uoxcmG2nBuA<tQe9%JJ9-7}8+fj9=E9$}YT
zO<427i~*z~RFD!zs7AW=(1w_0=t*xK!UZD}sh0UsG7qM-yP_&Zi}M(m+d%Kzy3a=X
zlJuxp8=;~t*T3u;IrL}EzLYVWcQnm>D8c6$bcEA@M;v&2f8g!?f&1gZiF(F}>o_JH
zz;RC;cQwtJX5G3P4_na(>A;}1kDTm)jx!rIaBS+Wbv%;T@zxuTnE2yg{0V&9=f43z
z_4A*>AN`RZhk*N&)4%E{aq16wF@AK{m*D9fkW>G}9Ud!d%af6IDIlXVh);Ui`AZ^K
zqg~BGj$}mlo!{{@@zcKPGw@@;(BR7G{CbvUAb#*?9QeVXJ@73*;&=t(;LVi@D5LBS
zJxi8{U41UFs^4h{OkVy|KvD(|QwYE@@gLm&IDTZizqr%7{X}Bi59DKD-UocLxz%8#
zDq9WGaxsPg);1VOPSqzv@M_ke6?2%nkh`gJCY9xcv-|k|z*`)6JKl1Bb#S8L%mAD;
z=&p|DZKF(&D-+=v=GryqnW%kGVMT4HC`YhvJ_Ij=Yt<BfWOcC<79wGTvt=sNft(Nf
zbZC8=1Gl~7_3cK3{+doM)VGvi!{Xkc0bcfE+ojZ%+04o8Gv^ta<WIIoFRxEsQwW2K
ze}-t>d{GBD=@1^rf#-3sP{z<1=|&N>BVigOk4CMJNE*e8+G=_e8nJM5ZZO}ZWn$#<
z-yfn|DLRV`U=+nJ(Pjf&p56@+wnX{nOiJmxVM_xV<jE(pzHg@>*YTK?$$ktvgb|{l
z(m=GT59pxt5kpb*T(uF0W9Hzv!$>@ulZ6t=;w-p<J=q|59D|CW0D9LTBl88JR|Gr8
zfInEsH`P~a{GVyl1Ewaw!MeZ&1xYab=#46`x7M+dW49d192Bsd0o=y}@1F1Y^8Jpx
zy46DwW`LM=9Etgynn{s>W-#~8RQc9B@m+;68G$i)(2&BAv;G|L2s8xtZpd&vGVx<y
z`U?KAF!4GRbjE~7#+-Qfe5`;a#>CTYuM>315V#746MllCtc;R5rdT1#iN3Oi1kT@-
zQ+h9|KUcXNq5i4@B^i-#`0NLGZHfIT1`E*B@C{#d><`0pQ`i!TFMk+#_u<6;!@z44
z34lh^)pW_pUX83X!)j84rQk%>BrcAv;D!?N%Q=w;pYM<LfoBg`U<EhzUnKB&K5#rg
zaO@9cn~czPAEl<evNtuW#+35`6j1Wd6}_Vogeo+gFbIpnXd5CDZNK9k-r-Bfgdcaz
z42%e}of;e1cIQx{t<y0bF_&zAcM}a=6X^t(QYu{#+5|#!H-<n*>fE#LQ_!8S-X7Ax
zM-ZkuK&k#Ir_h`o*E8pGOy-=!vNf`uCiwh;#@b1+Grq2y8e~|{jgB%&O@(-owI#nc
z<Osm0oxGN+=F<VMsHNsHBgk4A<G=t%VfL+qS|q|RjS+EER&bspF|Au8lOP$wCP+6$
z2D)#IT+QJ|%xoH?(FBdpF^3Ku7sX~O!%*t3&3u0xx)wSIZ|C*QWORz#MBms^dw=&}
z%BLtNllWpc!>v=KtxF=JA_I=plNzQR=1`cF38dXIGZ`6f8@vJ0JG#*UXD~A8PhI^I
zk{3|q>*>oD9mgQF*L>Ar7E~d}I3hXE0=9H_ER(X;4cNM2-(39zrhb!&VPaT3fWs4y
zF>t@%@%%iH$AAykGTd#cMIO%s_jeO<U}bAe_bJQ$YU_BCltI&V<d}LZBh(zi-{wTj
z!S2%>usFz2y6yPU#{*w}zMnR^AfRBD5Q@gg5aAe14qDQBUgWa*|CcPH_JgKeh1esl
z=c4)+hyoo>X49v~_-N0AqDgVyXT&-KHUl?vJf)EW9DyE*U-_@y@Na#m<NLo^QEZ`5
z6Tl#D-z4{9D5III|Eveiyf3CAXh9^1Tz)K)h1+t3C9RUd@A#RI*ZAzyhCl8PI+be*
z6p7>cfp=edhg;uZw;RI6k9WY^juwP4dB2qzoFhRmNgC~n8$8)-W6HoOQptKG8v8y*
z1h!k?>1m?Bn;1uG5*Z`)eQS7X9or^tFOce)+H=_)El9ekC&W08vGtJj3WXX6gOiOl
zvX%*`h+gyI-6`uH0Tsdr?qlE>8X25y%Ed-9K~1EiG<anbI_)|)sm^c)d%3Q&j6T0J
zYX>if;6ffg&FfP+g_eok+M-yQqMSMsRq?J=9nLxd@nRNCr!cq_{X}ceEFpQ=+TMxp
z9WsJpoG<zqz~+ju$o2IqI%$w~1dBp$e3*Wz4=~*d(lRAQ-5(EZdq>}yeNd0TAsx8S
zXB_t6G^FPPaU7H)R*Kyki$axRI*h|ZQjmqhs7y%{>+poJNM?gEgbSmz=J1WbT1*w5
zcl6t)g1#c^IpD)VbBZNL9#9s`wp2!hiuPv&g~O(dWzZ<>?s&6xe0baO!K)pw_6@z0
zf|y7gW8faZb4$S4w04XG_s28t_XkosSX(1Y!jiHUk7qKrstJL`dG3|1xnu?$n=uMZ
z8F3UWHYlMHGZvN343u-81D?R%$T*w&+2+)Cs)%WIxEdj+(6q0ZA<}|f{=;dJ))~qL
zCmb4rWS+p|z`l3fB(Gjpbt$!8ts73qu;{U^)0tc%^e=q)2EXDLd^>C;fM*~(bCOKl
z1alPSwhE)a`&YchFaEv<KAapeNn^$=*#HfNfjme(2k!lA(Yp3e$<vfHM@>H@6y^BN
z@mc&=|5)Px{8Rr^4CDROZEj%gY8-+4{eT(Ay3S-q(BXjQ6w8YM2#j&yqVc|hOe>hk
zO?@_W4Bq;gEl5q+W?K|rJvG%@#l~t4PdCTb$Z+I%;9N!g)(o4~P^l!OH#puhQ*!fy
zbS(p~9LR9h_L_pkb4?>-%`}<EA$HVUlIn!M>6n}99a4_N3<_^{W%|t66x=OF-es><
zKmn6H4$u0Ca|&S9b-&hZ;Zwp##J@Lpb>8|)!_UXeP0Q&-H35kjYDQut6*djII)|xf
z8XP#TxyHUVbYMx^k7MA7fX|NRWE51BRvk7sr_-L5W(UIAvsRiV3G(zKS>W8ZO&*ol
z-iV1gA3VfUSZ2RbdG$UJ$JA!xgJ{;&oy+T<L3-Snq~o4G-BzI-el{`98Ne~6_zD(b
z^rX8}e9L!^JFP;c9|O@i6a^qf7*-K2aEwIcP!EJ5CZ}5+_XEfML1P<%Zor4HpYRQz
z{Q%$i`48~Hn<u>5J8o($JAxD6-sQwQP@4M~6JxTVm~*15f$r75Au}+JfwyB|97BB-
zf)xeNxPwjy5JsG<@M%HD21iPUV8E%hV~T&<Z!kEv!ASM(m!D<xOa?^ILUL$N@4)>y
z1Qk)bJt8=!b9L}Ka4~T8`=UE?aEhtEM11`57(fJWw|yDaOR?ta6PAjrfkwpZS;iT$
zKW!b~^qJR$5@*74tjG}>saQ!6v!cWg{gea$@i)H3H~qj9I&8sR1&a{_=cv%jttpaZ
zhEjj71ANlJ&mVAbR?4f}4d3zZv-nNl|Lu5pKX88xDdW<(J?%G`0guO><9a*kzuuft
zb4&O*7N#CC6zFiZTt=@o!wkY;rdbcE-=}N|_02W~XY6w*W#sBpFxAM)VbnCgF(&TM
z4?3hwY^~$yY}QK^rc-hP?29?@m=wQ!-5TuGE?Tv*hM1tOlN~FuO+roq1T(qYZNux4
zxF;ROm}(G&K0`tCV@{13;ODf?b1r#I+Khx816EI0*+LEl18O9}*ELgMw96Z8tpjSh
z;uI%gTH&0ba7;V~nMf9sk*;-kR}|3cnR9RmJr{Ojroq+J)putLdn5<ii~|;dDPu<_
zj0Cp~XLE*u>6SM|PJzQk?F8G{b0l1|4h8}?kHu3yi))N{!Uv9VV7?o`!2#XmfQlDv
zXSOtF;s|ypnl^ex%jF2%Yz?kqnirZ;(J+x{8R(4-W8E7Igsro~H+f9WCNly+YkSU2
zuw!Bjs>Vj<Ig{b}fp|WU4{p?_-SN3M8@}+thA+Hn_~6ZkZQCg&907nu_IJk}&#Kdm
zIN+Fgg+K?;_YHnC+`1!=fwu$rN`eKhgU_ULMTe0M!oeImsKUI#XOVgKE41E_6KH$M
zKxm*y?_#qUzV{FPAin&SXKby(tevTWQtxd`%y4F?28|PS!4dioKow_KvQX3Z4$`Wb
zSR}U2y*CCKQk!OFEDUCSKxGC>LV|&cH(L>#LWLFC=A4LkY#4I|50A$K_s7J~|2@F>
z{rUB78W2z5xBl9nz%Ts=J|L?|7`L!p>fY#EYF>Z|Ix$?XBP7B&6NmoI#M~p>Ie|2I
zZ`fZw;d9OK!A@A@I1UU^A4ycOWvnnyy6sfc=LDNvqk1mP)^!m=n*cU3qg=HM7x4jr
zW;4uG))$G#{lHKU0N`XlIpC6ahcz5Q*%<?WE+<uZ8OS<X0*A5Rz%f?Adt^0qCVD3N
z1~f|%G?<3h{f$Z1Wd@DC!>AvH#La=v-5g_fJRbvhOVut6ZR3NO$);b{unwpIge$$e
z9Ql|*Hl^|UX+RyER}rOiNc)?~+QirB8$Oj&Fk=H&Z{$I5qCI+fy9fZ|EIMkgflYOI
zxiMOwAoIvsJnqivN9^>Q7-7=?;X(Hdz$P$waHghLyCOW^tgkekK_NGF;pt=mg^Vzx
zvj`eF>Mga^3y}fMgW$Z?>60v=DNeTn1a}t=wV_6EP8E|Dlq@6F+*em{{ar>ykg7Qh
zN_s?dOd}3{92DQ%jMT%`Io;ZVM}T4}1l#>%D)XUpjAM|6^Wa3SoChrD(uV)cn;SlS
zwd3_|1Eg{~4~{E5jt9nZXOYWP>~+JdZg|xjZd*rh9h(Eg9X0|_tr}|#Rd*^0i>BXq
z{nzSkk(@fEnGx>LoDOpA13YN6;5rb$eD&%{QK7FQ{~1LQ;(NO7JVeN)w?qeWg&fu<
z=d)RVq#+IkE%kmgVBbjEU(ILASoj7ob*|7}uNrOYT8C2k%4AV~b6Xnv%?*bd0sn-5
z&GFlR{fTcCb#g}FC+^R1+pymn3(JEH!5Tmn#-`7?DB=v@c!@<8NzGEP;Y=cZC=iT=
z`=1ziliF?6v_Al5AEiBV;MUaC`I3p?zK^vY3~V9-7?Z<qJeRhSl@>}Fyf{$IrN?n#
zjOx%QT-BUSm%VQo8Q7ArYm{X59K5b^a6Nkil5j2S?{n3VGASU){odA4D8fJ~k^~1~
z%}Mqh$y3z6y6uS%a-&e)0eore_~Lycb1tWyNk_%TUS?K>tyiW~0YRO00Px;cz+|2<
zqr~`Q2A+?h!C(wnUNEPAlGmrBq=s6ylbv_X7|pg)dR)(K5~3JPeIe>2TsrfO9)0y4
zB<Cj=<YRN{U|#8QT%<WCC8DipWNmG{=8g_vhDlbrQnEnA<AruiPNw8cxhdS&2K9D4
z1wn-xX22%4oXRc=CvBA9-IXmJ8bI?n4i%Jb@zbrjCXF@0Us9+{uu~jWlc2{R6PR<L
z9|zhKyRC0l%o6dCRX~r5^fZH$xb!#+RAM~1n>66b<@QW;P)YJ>Z+O}pBNeD<Hpi0x
zbtg=<2XO0#H+#ni`-Z36hE2kIkM@8)CVtEtTGcI9+RhipZvxGj9hnM|&A`ARu!^Z>
zhRqVAnL2kjjCqRuO=zyP#0_79riE!%PfFAGUuy~Kjd*XHSD*s`ZE^+(i^s&oTLW&l
zjVuwU@eGEd$gCzImmv*hmNLKumqW11$@V$!4`wj?4FF*u3g{KYMMucECJ|})@y`Z+
z%YX5U_=|trj<gNEolsSR46J#aovpPeTC3md^GrcZvM#fng)=Kf2*Xyu#{fu!tRRM<
zRj0&BTX)Ttz)a~b<sxga;gaWOh9oYG&vGc!KqLtm(q-b$oeVQ1zoXXBv^V;;p_S9F
z0a-DXHIvonnUumv>V~-ju_cnF$GU@o>#M!tY2#oXWU3p#U`9C*MfQ=LKDYIfqQ7B-
z;qg52Lz#f+#*mI$H)l3&>>nZxmZygPn1RO>TokfFTY`E=NVw`>Gu7Cv;M3e*unoW3
z>r?hAkRjB}*1(CT!x&;TL{bC)co4fL<!!3KY6u6=Ck*?dVr4f#3O?)*ETgQuP{AmL
z^QUSFvtu1;6D`yWS57TDSXj)+wI~^}hIqpuypoM|@FB|Gi4#YNKR+}njmfPZpqj_8
zeB$7S4pVcSL0Gp78*j*zu1iIeYOtH>oK}UzI2?Wi`mE8PiQB$YuAp_eL(!ic*fD`)
zNbiJ>!^M9#bK>=$2-krM;GyB>kp=L8=yVP$KIXt1XYspt$7|p5YHxUbYj}0raoaXF
z+eG3yfUmR%B@E<nSq$VoHiz`>9ldRE*Y-=mnxm<CC{!e?hMfjd6B%(1ttYSaY2Oz!
z>!niw$}}SK0v8!zP5;&V^ZZYHgfId-%&mzSVB8nI*GSTG1~t>6KI{6idDah*=T%lM
zPOwu%;1~n@?Z!ed7#+`$;ITfR5U8z>jszYVczb`qJh62)r(I<{NP5WC4^eZ_kyw=O
zl3ECH-Rms#+|+d|^GK|FPt}VOP&5y$BBQBr3)Khg@_aI30p{+J>RL9Kk+J6Pm@Q<`
z2=AB~`py{&UD7M5hR$+uRZm^TSGl(_18s&hk~l*llJ(Kh`R$t&Z3S65CM?;gyBTL0
z)VfyRz+*5FjB&7j>gQNuH>wnPWA!g?8=iJne~O?}G}gUo{`}Ul_YT;(_NnxD%~cVD
z(O1t5V4660%)z>!D96kS-|MaUDz8sDWXG_3$F`b~+WN9<$7V&YEY>L6cM+wIfnf%8
z(&qY3quHGeYfw`c>FL!K9gM8S7^>6JW=dxWf_13>?q)g|IGdofS)#N`;xqzmu1gAw
znXNg+3t9*;GO<9j1SId=#er@{LpIctP)+-Z2^f0@`qt4l!AD60QMQcEfo=I|jGPmI
zIrZlcH4nv{@K=HD=?QLa&6X_<mN*9H@qi%(!EiS1=4y8HL`d%A?IdLL_7V8v^8=r~
zAF#Hm@K5{|aiAksSM04jKD_OCwRg^MzHRVrgSSpe?fXPqI+}cO<1-4|`^LdNy<yvK
ztPM0Wwu(Prf{CmN*;1r+uwKwwW0M=0ns3|w0&^{cv%cRd>At+Dj27zc9}S~SnL+Q)
z2u!e58Jko(Zsic_b8&ta%SmLGaCD>20?Mdjou3(j=i`7orB5>HqSF%A3{Q`aKmDIH
z{N6v$@a~BtGhx8(t-;>`_s68(P>R)`tn<11lB$@K#<w>?mo8Qq&PensJDySbg3AyF
zGESx82u7u6W>mCDW$rRY6p5oFPt|t9l<jDpTuQGIgnEIXJiI%+g$$tqjiZ1ACN?8S
zah?vRygK`0pfiF&c(xn6>u>I8F)^S5HyWD+ol!S0gmC%p1TYJ(i`r%ZHg0NgkUM*X
zHgoJ=a1Px|7~!IcA(W9B(B2KbCALkO$lkH<9f_&hLJ<ZO7X;<FTjL<HCQUO2=Z$kU
zqOuEab9Ym~{dItkHqp$QLmkRzfr+ccNPcFB3W{|Q*Vc-}@i>rglbGP#gpJ!WCT?7a
zn8fHp-Fc{k0;~s=Ln*L33=J*)R5%Sr2BW(BQVgf1?wZ0a%0`fmwuZHw=49cjWjmC>
zs*%J+sSHV113(qFvDgqz%{=3-SrB>JCPp6O!_TuW;_isyD<W!1rAU9;F-<t>B%j}t
zdMsv%-V9q4cJI1=sQIrq3@0S^d`wP+yH9M}z}AfY5oSv~Y}*>1HbG2%!)@DfyX`D8
z_f38DY*u0yr_0zd0_CAM>^rBqb-_KH;^u%etC*r0!zI`TGLi^KG)316k2A6FJD#?i
z3ixMmLbE0#l4O<9IS7VXy$O8g3n{psnM+i4-E+uOZSsYaS?yJJ`5-&@Rb!y?f_@1)
zxh*+AW4rAzK@~Nb4Pc}ibK+0?H+FpczkJ8{{Dr{V&l5iVu77;PPZ>A(YeS5{JrkZ}
zY?VV`BjCwb&&pmD$f;@~xHKdMWh+}?#DA_|?M=dL*BXKe#x0o8Nth7J5MYsedDn{r
zK}xG<J(569+KiM_GPt_ud3nt*@SBR-BPO~=>axC=>J3P0N+Yn{*vHe+@H{Az2qzf<
zlAcecV$DjJl4Ed>5vOoxOk@m7?etD<uiiNE3_OP!l(<>;3S18cGGgQzV9l{@44B%Q
zaW1FgK!)j_1wk*u1mWIH$8pe{)n2VLy!S+lM)~Zo;q~eI6sqN1k$vI)g;^03Bz{g+
zLSVY_hBD=C!KRvHBJ98%iEK;<tu+KRPuY{YFEfyn19A+uAf<FhE^}a6)>c(JgN#$E
zq#r;sqQ(?kN$djGya#qrQ7WscKC!5-&M3=Si-Lz?PZrv}iOta+HXJigjJTn3($vNx
zs%M~sk+B;%p-u;*sVOH`7Q@@hK`LlVL-vXJn0T~9#r+Ay0CJ*(QrcaIy<1{$oK4$A
zw@o(;H<$&cB_7^RX4e58EM+MvjjPRpr>$X^y6N6F_9R%B8YP4g3)j`dnbL?!B-}P0
z7zg#HL{qG&jt-YFA8i65J4@;IfQ%0gj8OLJ9iM%AjW_#EQT@e@TYa<~#oEE9;GtHx
z-UDDcvGcVMM798(;a67ErmydP@q-r%=t~hYKkE$4HJC*?)qFqm;TyC^qP19bVhXw?
zEH-$)LAQ=y{m(rR0ql*#=WjdM%$Wup1vh~;o2jKprDHP4=A@ICYHL9i0M_5Ft1c&%
z4UQ(<n%TeUUcEB0fS6{by`{|#ImE0)*_VS4tp*_7r3I<g+yS{z56t7CV6wxu4RdlV
z?D1e-VkAdvM$vCjHBAx~x&jB6?pLU=Uv$qS$aZ90H<5WDv!nM7(P)4-=e=;980ggq
z01^m5rG^+%P?*b5TY}<{-f(kHbnDv=_fCqwJ1k9oM5XOyD5{|gBKLB(je*{r)IRFY
zi4Xhk_~7-1&(FZq6Mr_w#E<+jeAU;d%M`XI++Nxn;Aps}=rVK&cypwu5lYf7Mapqd
z56Cbv0*bWBDQDA>X+gNzE0%FE1=SNny0aDqqP3jwFp&rzKFP=;E7Pc`qz;WD|Coeh
z+@*=aC<~b!V+pb|uF-pMOcOUXycDyeXmp-)dG_778C?Z;lgt3}<QJPI>6yk_L<MW6
z>vUrjcibPOuTFMhHzcPyd3D)BB!H>m=WW$UqJfeHz+y{{i+Uq#h7#9H7W2CUn{x^k
zXJoTUYQ})2vlf$yc}yISfp^aX&-XzH3MqyWC)`zC#k;7)y`iwu!m>k=T?QdAN)snA
zq!AVL-+SNjW`70u{jx)I1?pLUuKzDXC<9JIc{<<o6Y~75l@$^OVzgYDhs(h%&+*aM
zGOSkzr0A)>U;FLHzW5dVzCZ8>@n`+Q@4!T?!a1XHo`DSw-8Q`1sSnlk72d_m4`!RS
zrYyZx<J-yAFZqdrZbE^~2_Nz0t*IVkOJ2p5Ww~abM>t+|fG_i|b0;Uj(i!0)8c0D?
z$ur%=21~KTVqpiS;`UTUb&^5k?53ZQI3CjA(u@`<$O}exHHxyZ?2R_1wYK0qULAv!
z=f?vx7`Zo`QFr7dbYo7L81TQs`z;L~>^Dl#v9b|xz{ACK<gg@fj;C9v1GQZBsX0g~
z_jW}jDzPoJ;T+E7NVTsP<(I|$SIsf@j(o$xL37;EoE_Hx_J8~_`}Donr*jH!=$0yU
zGEMB-XfUy9kPCj<)Oruq;A)!MdeDOf=o-hyW-q4NUB(vjioAER8Jrs_j4XvI{xfX?
z4|5D|VDE{Z9cdF`rp7j5`e&{s$qGor1dK-YnXV`x1~ZH9hFf#oHp8yLG0)@dux=*3
z6ij%S#t~8@Ba?jt8E4u|I$GUhEV^R#HDt<B&cvJt?!l<2>Kb9-aPnq)E|d}3&2TdS
zrdZdA#ZEb^v;L!Tp3;V*!6EC!&9HaQtgS9fsAj{{!wnn9f&24;clQV0J~$KG3@A=e
z6$3*?KIZ{U8rhko;#xUK6~{JhP&Z(2jtvG*7VUWb!3W4}_#;2`BQN&oN)mAXCJXGV
z8qs_Ex(wJ)_Oi+{1gETh{2~*mldQ;iWe|Br>DQ}NHnZ?~Graon4IXcwv3M*-`4w*+
zb0bZwPn|0e_jYGcHKdcJO@kU6Ymf+`oslLl(~yoDz;;D+R-xb2oMn1{HR|z9C8iBT
z%@4?O34q!XS<A?j3<nUHi-d3u)XYg~FBa5-SadCw)v()|?7}H{?D_dwb+A~*+5w^i
zV1^W`ryS+jOf{?C<aC0B3uxZKOu(H3c;XzUML7z<Rc9=!^xi}nmsH4$&TvTPdon=o
zog|GGEHYPTd;qu~6Ny8Rk(gx4)Lc-OdfI4r#YAk$X4eLT?_7hYW*DzFAgbZ_C*R}s
z`gDcny7{Ka-3-l2qcr(Y=b&6NBm)tNz+8({?fNOhi&P4h4VV+agn3^)dtzoHx@j@m
z*3mX9J&s5mN@pz%y=@e!(y^K|5Pl#Y6C-Irr#`6F1T~|v_X~z+GT;T}P?DI&aB~)2
zpEk#<?m#yjlN+*Lt)iU_U=HK~L=Yw!BUU6)hmBiquE>B<j!vU9%YdTtF<Fd<*iHM!
zX=mn!E~mwMLSk!j0F(DU5g0N6G))pd6FY|1q5jV<*f6t%itJp?IW#DN`51US4m{r{
z?jtax;G9eJonXV!7_dhfSyJ)Jz7k3zSV*TphqgPNLWAS>^a`K(?C0=<KmHc~`tSdD
z>w*>izudD~5R!~ixd@<`D*4eL`pAc!5YvKRF8ReZE5Y?&uD`Loy!q-dZ2i@Kd%~+X
zACNFGPb}1uhCe#s;pkqh6c(r{^Lu8Ogl4LjL!WZZl55YcOi9K*xHikQM>!Td*;U12
ziC75_fb}zs8I+wE3)u$38^qNsli-?M*OBYm*11<av;1e`><h>l7-nn~eH_&28mujt
z7=5X|(^|uD(!5QOam_s;KQG02Pq>pERIINo$ElTK*1)Bse-uon42tKPgcvsl17+$)
z0^#gXe%d>!;t1TtW{U_s*yN3^VQC<>y51y%(>>7}>vpI<4?Q^;A=kEdJas{XxJlmQ
zTL1av*Qas{x6N_8?by1(J=P3?G~$kjyjy6lLU#umJ4%^1N<*{GgrFEo>@q_DjP$H#
zfVp&n&N8$c2Wf9xgEb0c&DxTYoX#aVi-6&N95^0{`?KNxJTV?lP4uV@tVqS6c+%E^
zN+e-}%rz_3jK!SZjPT5+)_CLnp*pyAD7qWv9Ew2|fjLN}t4_|d3w3Hu1mRo^P0U0U
zGwB%AnBopZ$7=^#c63WTwZ?%pF&RNzEQ4jTwxR>iQJhl(hI}{<zf8wV{#@b8i*83l
z!p5b6<Nm<oe&G2&@OV&4#E^unm~OGJ1k8AR2894qwC1K^CrT@uQERJ#0(zS{ZszFL
zu=|cZ9d>AX9B?tVXRKQF1iMIt)cP%$M27OZ)@QhU{uM5td3`P4@#3%T{EO>nTrh!>
z_JCE!0sM~N`|sgT`MKYUpZ$f;;_-M$u7C!HFbw-TTp0NjRe9wt3NDJqNj%+_!!_+~
zUc%W)ZEo;sYsP><6mw;IW#EP1Y7KEQ*7Z}2!YgQ~C>JQElIW~f>v(2NSttQa(+gj&
za;op8YHm5n0vbsw?;MhUfa)9iIWrO%W|rzLVX&rPWPm8cNcKXcC`dUI)qYZ1Om2;E
zcYTX!?J)w`CuJ#Q=Q4qC20^L*7P<h!f%|6gZ0HSmwQaCwI3|a1krh_=;yn8{H5re6
zA2Vcg@G{;Y6R9nDpMXG7hbO+4*QaD38J-$P`mwW_lsy<R_a^G2B^$4s8`WpCV<hDQ
zsw)?ubLfr6-MYp#I)|`qI;<l!&smRot3fcEVcRz;lbUeJT#T8_K$_t`I_}4YcW(#2
z@|6S6&jZiTCzX>Mr|?@Xx}OJ(36N$rX%9Ci3yYLg$B?Y@NsF}QXj&r!z$J&906Yhq
zP2AT(Eq+Ba*M_ZVCRg#AJ3tu+K)PaTtJUv=*-4XT%$BKv;lMo-!wq+LxP=ChOTqy)
zoP>6StgI3wYfGxv#Rn0j4oky$BytQqj=;Mk@H`HLn(wEC)0zb~XQNy*HW{|0N~9+<
zu)G+HURjD6ZpN7vH}9+$jDfd5_T%_nzvtg#_H?nqsNV{bVS}VE=FRzUl^IEos`Qs|
z8v3&JFKm8=!@vvt{e*}3Z}0hz%ay&%D_FGVug~|#s*#mwVPsXz2)zLbn_}u{JZ*ZM
zwAeEAhP5Bpe!f@_Y9gIP2RaA>{>8xuz~c2vkm~yRU#R$gElQ_>!JQgY>z|6kn*q&A
zPz_G+tNg)>{nuO#S3!dxV+B&>k!Er6qfiJrTNSsfE;u4^WXLfjFbzEzfE|&9b#MkH
z6M=hjCVWiC;fCl$8LGn@sxdS$_?pQ^&}LFtu5=>#3}*xrI%B5rdLx~G-#T9Ph9?Y|
zP0f!l>aLhYMa%alc$k#puKHzErmzC-%H;KYnrfoQ*8x7-n%F0X1+fHpaaBOGO!%DG
zEfsk>M&sf~Qc00fFbmiF*4Yi&L2Z@ZDd_-+-)(q2+R9Y7=IEW6<Lzlf@0~`~MW1w1
zy8w-Bj1Bku#9M#HYXhD)!&kbkzK3OpVCQJf-d4nuW~jN!sg&Nm;nm)-Z;oBGu-go5
zy`lGssS?$`bv*42w~d%-q_PxHsWL;wdJt!Xo7fRW!{kwEX&8{GSe|hx#R&~hVi|R3
zclGPu@YEYN19l7CIx)p}mT(c=ALEWk^8WXYa=fl<DShKKN@%y%1tYO&T~TBrR^21U
z!2JlkJ0>3UP-NNxWkRpI;i;=h$LRcaZ<Gk}Q!k$r$#{R8WG)0MtcVSn$m5Rx$M61q
zsK%<YT&JNHBw|G+7F$FSZ2eqTP+(XR6ldyjbqA+G|H)tanQJCeuNNOtv3k9Y@%}&m
zyM90ZJO8EcV)0kS++xmgOr@y?#J(1QsB5ZZ5+Z0cbBZe(6Ij$_umr3zG7mYVTARuY
zRqIvXkdmoo;Hwcdka@C$R)0c+I`=R4U9}=W!36i<IdF#Wv2tMLsC4ZksP2n`o>-8R
zG&+z$7%2p)4U_<*GL`0t-f;{zcBMN^HNayWg0$GUHq9^+7+KlPWUoLN9uP}V>%+R8
zzo3C~eGQS5!iLT)VU$X$GP74ZM{2%$b;HwU=rORx#LW^O!Rd{asV^B7H57^#+nQH>
zd}UUZA@NBuoS@n;o!{5;`gDSLPUeNTCX5nTg;6ldyfsQ7oCQ-2nTWoMn(m3)zTwrr
z<JHrK+pVu?rdt&#Nj;n+@Mw(2YgA_Ij&1AcTf??*4CsZsIt({Az?_rJ+J-mQ@u4?7
z`i|o<aesAS91~UBwOV{Vw7z%t6tuz#Q-}1>!n0}Z)!ta$_r$g{E4n?I#N^13*)~Vl
z2tpd;z?beF$73K4M(0%nXdRl9T-cTxkCOc85yU{N>vC(rZEtve>v;3D<AYnrn~hS~
zPg^H30Et^?gVkH#;PV}j!1Ep)o0^2SGC1Cp*_{Jm@w;irQ5h|7gqI2)$YDSP-aRJX
z-VZz<?4~RTp*Lf*+tZHQ&e_PlH+1XRtZN20ol>Q~X@1J6RP+N^{{$G3B&|k8{vUm%
z@N8(_bDV|?74X{<6I|T<f-|mAPkp3akDUITx!!b%mvO>JmxK7iKb-HmN*w=(-}^@>
z5;=-{PgY5hFj!i;z+acYTg#wyk8@5W|FEn^!WUG7<tS43Tm@kMyMlvYQ)o4Omh5xs
z0-bbT{%?ub>7H`N%0AZc9AKT>wW7+l-8tukdQgXm87E!XB#M_w9iV#*+~>q&u#k-V
zfnyukddI$N#BK{%>$nHu&PR|+evE@mIGYO|9+^1CgF&ECvDu6?>&T(rEHVmnYq%MO
z@7&l&M8++h0X4vIGWCvx_r$Glc-knxe(R2whHU~CW336N1OnOmg0o{%(%V}Tv49yz
z1Ys)bMWvoUQI8oM9yIZB=j5ZWPgiJ`2Sl>U&iSsS9j3Du@aC!E8$P_@v#&PMrznB&
z7*xNUAVcPZ*EifY$J4&y>FFk2l+foZO;!`m<KUp1?rk|57wpvcCfqWSMjiBqiDr)J
zP6>-<h^--S8x8!JoIqAR4^bXjn{E|WW%w#O!r*2E3qyPF*fvMRz-ON}eD2MSMb@$h
zCK{&~PRNMiaU6&v5RbWFw$0@fyqEI_b1y~;BN9jO9?crO6$7f~D;qaiZ_Jh&6*Id5
zdnfGDCP!oj7}4K{<E~==jDs!lm7XSG-B&u>dRMB?9Hlvp^vHo@1RmLN5669q;OD0A
z54_$t+#0X%)z(E;=<G;t9V@6L@QSJ*wLE`c5m)N>tIBzT<nO&qhKq9@IPt2U=LxS=
zBUd?x<q%%LBKF=FTn&BYDJQg|lHc`8&ig|u5`OuPt8)v@|6a6G5Sy-ra?!#obFFOc
z`ZWbRz?jiP@B+X3G!K~&g}2T^EU2x#g#~MfltPcy2T-^-gE&!Mp>FBc7F_z#YYS8l
z#!Hd56M{|6WuY7mnTXXTUD6ygtFK^S%+x-lYQixNWZwAwDg=ic?f~xhfz6!5a|UH$
zwg&7iVH!YNgND*c{5(WLIR*S^bi*2~oB9uo6FDmjLasKcF@a;yfsLSSi+8|GQXXRh
zL5!hoWcgj!7~tBw)D$R*A9&<IM`ItAGZ6B%mBgg73K+f)g=W`7A@-igaZs)ziFIM`
z$X7dj-WYK=!UJPYJf=LL3dOHp-EeCx)b5(0ki{&ZYh4~Q(1?jeAjYB3Q7^>4Ni&_;
zH`~q$3CN7&n29uY0X8F(C}Xm(L1~!+PZ=R%om+1@JnR6HWS`z^i;)xKKqPP^PE1(u
zSyCW65i<sIJSZzXS+ruuF%(&x#xIQrs0j<6nZio>y_=JH)J+?qs}=z@(lqr<C>0$?
z;MJ>v(E?BR1KTqNF`8ctnlXxX@u8bty1c!nxt;0%eK1S-%8__CC>ah><0y;L@M_yM
zX0T!3cC^+t22t>l3|BFrihgG(db0|o7174T|Mk0nueAJMGJ#IKpLg^)4m^&7pRrh}
zrT{(`M&0&ZkQm{KRJPYNtKy|``^h4%kI!1xMd2ENRF;t|T8|9;Gr#-az_0)1zi`13
z<aBpT9Lmr&1~MGYS-q+2<r=Z)#49s$YK*}^7HdgdR4|0g{au`86~HwLQEMt#rfO(1
zf>&tIBEg!$dG~&a?s5J3TA_>!;|h}0sOCOHm3M($Oeq*Co!z+(ZajcvaFi_ve)Y5P
zoQ7ipU%peK{g~`)=tw-ZK=cOWRB;`g6z3(|U4PH)tFdHEc43s#gOk3B>cQq~X5cXg
z?lJH@C+;z^-=Fccy&{wr6XSRw$AhZNnr+>v63Gm%oRQP+gz6(PtFt=M1gkGPFla-I
zN^I?A@bF2mPc4YY6fZdu%)$BsSX}8(9k>1puQwWz-Z%?j921Xm00!*d@bt7}-!|-f
zqrzsB<bl#+(!)#zWCx@g7-L}G-=Vb5-AqF&blBYyu_1!PUJmI2*}=gXTQl5x!)@;@
z%t})vP?6MGtZBtN4mQJCCPr|Bu0x^da)u^p8H-g#PHpDI=Ow(<3_wiGdBD@vkYzB}
zoMAIqkhznYNHvOW>HzkNmYQgA&9Rqkx25VIU{j1BG@U?X1bjcR?K^r-t^WW<z_KQs
zO_&8~c}@vyLlmI`Xs(EzO*Z%Y#24S*@nc_kiy!{dm+)h6pYeP!qTRe>PvF+2hBO6h
zkEvne=f(vjgW!h1R7RpgWv%rUpE1LK@<)F}5&x=2;2N((MYiKOXef<QSG~hf+<Fdr
z-!uc8`&_0{7;cdOKGqe`reCfB$QSPvRkN~#OkEq!`^ghVib(vSAN&!qwc6@u$xPVL
zK1pT^ttUJ?ncdZFsc$TywN*$~J%Z&EUwdYXWmf2!TS0Rm(@YMiind5Bnkp*NXY`^b
zx1vwVxz>F_^TjK%vpR3Cb!NssYJ!>WqnuOq$AkxLCb~6jNsY9lPjnk<lw|YcF&-H6
zz-CYEuvXTYX}I4f-W?A-2GN4e5_bchI!O<`0k8LtTX)<V@YD!dn=%-L=6ZP3B+9_<
z&ORrQaMQXn2|e2|;2xTKXn}FR1N#PeXbNSn#xz4^cm-c`&1!$v04jc4;4uaquI^zy
z2b>-_rlh#PMsxou1@S0Tu(gJ#eMfJA<%An+_Pg)sPaSXejZ7k8heJw&20{q?PTAqU
zZ)m-*Jt(9E(+pt-Ya232`RvV+G0{GJgVwsDs0qx;X<|~AWJsSlAg%s6$bd4AZ@g`t
zyne@W3U#K<n<7#T^;U;4AX#qp74YwE9hxHx!XlQN#z?gLY51&CxadH1b)eQcQ;O%L
z!i`A`m7Q8_i!ZfRcdRJhoDqdX=Q5f0qEs<?S1?fIZX83@TGj^kt>nH-mV->BzBNho
zkQ@*&m~Z6zo2fx3ai0Tkj{{%4ANcXN2fp<Dz!-tvHceHdW111GPhe^yAG1P|5oTq6
zrsgnZBGexA#VQF)ToqAInF;7L6qLg>dNE#-GSp`#R5b*w=zT@-z=DygE`o9(a!w}>
zUM?o<3iO%hC7s_<|8T7boEcC>f~g{YW7x;*PBvq8O8^5UonGDj6hl+DaGn8<L-Wng
z0wC8+JpO8s%t$&2dQT3*QccSZaF>IToiS(dQ{C+Dt_IcWXE>d6Ou;c0OWVI1)!^k+
zL;-j0IaKCU){r#S>WvEL3<p<+&<xl;a5oy{1;b^{1aFwb(2{7w$f*f}6n>mDNEem3
z<?cz6hHqTDIhTHzJJ;JSv9&~VHm}u;1;g3ohT0n$m^sl9c-0*{Soic8$UFG}y=dQp
zdm;uBP2##L$QN>wytkQRp+!<fI1)uet|C4s-$}`Myv%n$>GdgV3cUHPb5z`I+kiAQ
z&&AvC(yed32G6ilcq$0nw+-7y%4u&*Yi3H12!?0|_m1ct@Qw@~;J$H4_TC%zomk=V
zc#H9P$Dn3Ze!&7=Gsi79?E8jeuwV`2kd^L^)&SUq!B9hidTpq}ei76XTHJ42DV%GD
zgjop%wgO-j*^!P)ToBqW$r?u$YTHTAUHEDT+9Z1^W#nsYX$6-oAhlEYuRdK-A1gA_
znF8Uv`|}<5$ASBO;xT!plu%}qjyKLl*jq>6Hg$CdwU?4`SOdgD5s$p`Y;Q=XK-cRh
z5UNW$m~8|42JCxR7Di~mE*M#wVNA)IIM_p>^sg1|#Z2-axeQibnucD!$^cT1`1$#c
zVihg;LjUh}{;4um+W!l_EexGdl5q9Y_N<5|%gE#9-}9p@-MT~}3qFFfxq2^Q$ksZL
zhH)H{v``VYeoU6vva%}COu_fX9CWIFT@#0Q?#pKE|AnVI${E?aMut|rUV#V=HP^?@
zn4O@8%%88SNtBFy!534|u&)fvup~!e8b8V`$C@v8Rp|*q84!X6=Cza-*d_U5p?w(~
zq~mj<Z;spE@bqZdx|+b|<SY+83yvTmEMy%R2N2C*TZ4t6O;M?pv1d;7Q0I^(j)RCp
zcfxiqJh)Jd?uOS}hhZWtv3JAfgY`erST|HaBbXUH7swF(yPz~W@0jeKhR8e)Ico^i
zkvz&sW`N^kQ{9igK9y784Ah1U9`LO>_N_CLa1bN*zB(h9hEQw5_o(IGkcz;kR6#?j
zvm>3|xy+I&@>O^{q^cRU1<e(3u#G=f7cDn;+cvy;x*-C1e!k=W?iq6&=+<Dy(UWeF
z66>k<YSvD=s;a&*L%loF@A0<M<%kmtJu%uuj9k(l72(!r6g;tFNJV%R!9k8CP+j1a
zPGiY`ry_7OM&OG&3j?gHSVh&^k!sXQPO8daBVliC4S0{TH16o;7;#4d661JCvUjI*
z=!E{BZW}&)b;CD(_Kw&42YAdHp%K^`Sz@nx#|KXv-aPfC0%>)~mmkg-u%OT+79~S>
z-wbn3wEdOt`?|}|-=`cxMX|-NzuICtbQ;ON_4RXEo*p=z(F`$EoEVs3piGxdw-p^P
zBflI%VigRsFkUP;q^{?BAAk6d{W$)`fBXN7U;8V5F(XfHLMmk{8h06V0B)TTwz=l@
z#)=e846yT>L>z#k^E^jk&J1Nor|bgq+`9=jl2&g34c;{>uYiaX2z5jk#I$y?9K{-D
zqqAVBV1Pf<ByU6HO5_vJI-04!1!U}P4cOgr_js|KMTF`5hInnbJwMpPtW4Du<kWXR
z^@b1jh7WEHPaSA+F!M&udTuB!DTi}z%1p2|EP+m2WX8g?XX42l?9~%)J@C-FGzHO5
zt+8$g)=zQ<hv?v}6;x3f04!D5cGHY@E|`I~f-N9~q8%XIb_s^Qme;48(afS}cc7~)
z@3uGg8F=CF9R??j41^@i4cXcf&+?|imu{(vN_seOpMl5Yj>KSTM=9*q8?tx!ZG(Yq
z7;CIIWCDluPu%@<K&@lzJ2${i*3mc~h+`uC)cr5Rv7`gkBm`5_M>#iC1j!30)y*-F
zxR{(u8_y=7!o;h<XqV4fK*Z&6t!%(fjiTlK>u3qrU*$JN!p&5J@xCG=Mh%VfzFWgC
zzJBiwtq*MdfhQI0O&#VX*$pW-!h~b1XxY*66h(gnUf-Gqxl?41;(!J;$JR-`dv)9K
z>h^@Ey#}9i9lgl`hU!TY1f&Bz3uSY%n|01o0=<Uom)9j2Z>?kNoyF!e$XYIB+enM8
z-DsBU02v7D9LK@T&s~wN9^x`m7nfx9^szF7>NPmChRP`He0GU-audM+@B4lbzwGD#
z1is<3pF!U`VhqA~);P&bFk3-3l(-IbRY0;X(ZIfe3g>Ho)t*a7=^^zDWw7ioS<Q1^
z`<H8Of}~24nms-BJOIpkG`v4fGb?%~-@`zTkTAeM1GYxlju62C5asg1b^_EwDTD8h
z-4joHqu^Wuy_HrQ=Yce97&h?Q4WI1|!@*3`NzvV$Oui3q4IkV#eCF1$Spwrg##mW_
zEr-S{2{l~?QkrLCSUMSkRCrEwCw)E87uv8jThwbL`=&-znVQbMlmkloJQk&xaBE7n
zFJ)_{f-7@FOEv&*>zI5U*hioL;1xdn;0dpvIB2D$Vw>QU)-@3frWWsstE9lfPv;~a
z;Vb~dJJ7lg`~#2UkZNbbtf5;XdLyJXcI52_ZyQEvHtL+X9|s=yhtjIK7Aa-$y*E6)
zy5Sk@K;(h1=|z0c9088r6iv(s-}K%T-Rtiw`b@poY_UI<jWS~1E44NdDhV=Xs#sil
z_2qP^)G9@ub%3>g4J#UBDh`yPS<PkUu-?cxiUc(W6Svq9n6L)CetJ?CmU#R8z!%@$
zvF|(HyxK`+HACwSxphW>tNu!j_|~Ky-x_YW4R2lmoj_v0y~2FF#^az;rI?UFCWiW2
zTI*<iBRtg_p)>WSK!q`LTZ|dJzlV)d#Jy@Mp9CA#UPC!sVBa^I5Up8Fu&PN{3&wdb
zoJs(5OgtXPavUWU57?rD*1pyou<8_xPFP;%5-mdxzF?U-QDoKx@CX0M590@a^pE2k
zKKmKKAmvm8&?sRMf=jg*ZJro%^=DMJZbk^QV9m8>R;I!3d^ohOf<`l7>Fah3KzagG
zgp9=kLZt-En)76k;Sz8eB%c9B|K@~E&g)LJaLkzSUQLGN<<r)5L`<Rs4pOuM+`1$7
z1dKHJdm6ktg`2Z)DJHf!@Wv*-@u^`SL<wLVS9|MCcCF#&R2S_I<RC@31pNr_Oi4cp
zt7?4(cv0KdRP3b((>jxMk_h$RI3ODoVPrzNYLe5?cAflWfqbe%KefiShnDPls(xk$
zmxh#sjh`@ya()f3PvsPT%IDtT!w+At;pf}daBG_e-si$$iS<&2v;?A%fF~o92u76d
ziF8NX<Q>J??VSho@px!A8W4tvh9mD7_qUvzyLI&a3D!2WZO0Z^E$=bLGMp<Cr+5fv
z1ldsD(XDF%9cU9a*=&|;gL7yM^+!~Y8<U%9+txv=iE!Tes2q$!GEZtA4DjBU$X-fM
z2}SvoN1>A;TcoRqh;klH4_i_C8di?;Y<5NS>8#H*<6PvebM05N%M6UM<16<Y(mS1k
ze%9TvZJScksTxY%4`G>o>#(ik^(}FqWPQb$3u>WAA+MYDQbx7K%3ACbu7=iR>Q@?@
z3P1xv3NV1a@04|3ca0aHu@+m7+l+Do)XteerT49^Mx={XUk7X^*-KmtI*(<97sOQG
zbxBW_li+Wv@$bwg{Bq?6D-7Q_L9eok|KtDjcj2%3^ZqRS+;92D1zD206r$qR@EWk-
z6Wmj!v9&c48obvmb3&^umQU#;*=h(}#+g`zpsO<*WVFw5u{q9+)-EvA6|Ms`m!xE>
zQQgvk)=kc+>NL#IW+YnJTFz!2)iCSIY$)~Z*upWIV{8r94s4@i-zJ_XGvmEWVuq0@
z_89oECGgbnh4#eHb=Ej`H$2@o?0aYS0l;!1+J@Yj5zd&Hkr>esjR9@)B4xh{7EiIU
z0B;@MN^!Y}5dllG>{3v&dnzy;$N~7Kzzk48YB-KTc+i}+R_-np{bMnmT+FDF(K#OP
zFeCByG4Qp#K9y7W+-Kh4!w+7;O^Te2kOaGt#ci)HVB3%p7>|b<Ik+XQInvr<rtpC3
z0Gq>G7lUTP=CRtjH}9l;DVlfRu=gGNwqwo--=DC1$2<-^l05@CCyvK6_WcHjQ^_=$
z&d&@?70Spc(0EIm3o@$c(FC28<EL>n8I*c(6-kyMOEsbKhgMXXinbuuhN&N*cZ?_(
z21Wh0c+7=c7c<bsd@^X<pf*o+eY>s)4G0w*y8{_*u|EiB-SFy|QVQnesMa{}D7z!N
zbXFAY8)#^=Hn^^%sJd<E5Of&rREvtB%4in-v|ZqgYES}r9Lq2(OVa*eW6rroFP*^o
zBD<()maGF<ams;JjE!IyE2FMesbU>Q&drP(TWjde3r8d@>Y5eUg%*2xo2}~=XsP`U
z)ICy1ELum*134y+Iq}<m@4t!v?w|i>Vrv^&Cc5R?TWQ5QVvV5$1&P$ya=xe<OoR%`
zml@TIg_<klB|+)p%9LTKDkk+%I<uNe$n4<Eoa_o&1w-ksJ>NU~V@!rCQOE+Xd&Zz~
z)5yunB-n~TdsCTan;aI^n`%f0Mf|%><T#*CCu}Y7x&v)*Wc)PMCz8#sTQ_zXZ<LVC
zF__tC%N=24*Ad)-S>pk7TTDOdbrEVeV{cB=8IQD6UId*1p3?z_CRV1iCb9J{+v>}9
z9IfI0c*Ynh<yJ?)F|77=qFX=51dFjXq{C!+O}#()^{Je~XFhz54_?1o15;opjm(u!
zR#9|<(gn7uDjtD3k0nwD6Q{nmvawK&1O^iMIAE>AppkgBf71lvP{P4e*kYps;T%{$
z6LVnR6C;T&TZDApcnTr_jKJ`Ls>YlLp;CVRa)pER^A&l`aTe<`$>fMSn^tU&q7CL^
z3AAz)U{RF_1y2<Zm^^@8ugMl(8_9tkb+}z(fKAwc(Mo|__nm^7hy&9p-sW67GIR`^
zXu&}4eHo)=imN$EP=kR(bgJ;VaCtLuCIhICRrdz6Ju;UWLkXD{QN}OE4dB5UgN%_8
z5BxQM`ES7A@VEb6IL61N#JQdiyLsspy$+`u*2tce)2>V<RENo6H@b{hIVNcB*Y}|A
z<NRJ0y0O?qFP)cIN)=&Pfk`|Uln^n5PsBLzeLwW$_+9_re}rH4i@p<^b1+MNYywMn
zjSKH*5Thn2%FOVrF+=3nlE9&JWfkb?JTCb1yoXRGpto>uw&mg!2r*+?#dJS?W$j@T
z+}zaI3PGafL|f9))u65g?`p3C4x526(xnZYv%K{U)H`LMGN+2x;{jv?oF;T@N%mFm
z*n9Ox^(E%HHCq`e;j{$8zyXSZu`_^*Ou#$L`+_B5AXI7)rB!ob6xcKnrSN1!YwyaI
z(=ZR#9x8b5t&wr5MmE+@9FC4VhOz?9+Azmq%ud)a%sz>(dSs*n&brCh^7?d?)V^WU
zp!V9-<_#G`JYrVNT@h95jDBWoY9gxUH^L*KcKeKp`FLK~TsHuF$0qq`MvVc4Dp+YO
z7z5Z7@U6443ITIYE^>gH{t<}Bf%GXh003)jW&&2D+qFr?-7v?b!74@)u$)4jaSq49
zDeP5;(~ZJ^mEx;-%fq6k6Ky0OK#B8dW3HN=4hPUAfOeVGhFsR6hVqc&M}s#DN=ps>
z2z7{Rla>G;3(KyiJFJKuW{csoTK<!CSK0wyhqM@BeM2dAdZvtl!EK<FN~?80B00n;
zr<#h?gJFsQVGQt<t<l*A)ngmK<EMV^LzJ$}ipZ}oQ&y(m)WPz0X0xX?O%z58!y1ZS
z`(Z`Y=eu=yPh)o4eSX$*JnuP>^WqFx?2#A$!tAm0J;xaMEx-HU!Y}wa--@694PQWt
zMqPSG0TbIqgbAtJ7Tidp27{=80+ux`jgk7~hn-FmEKriUmLd-5{g-THO-?pN@p1~K
zo|<u4;&t1C4DI}L0_S@jK!~os7DQdbuFSY<CJAKFvE8<fMO%0Djnmdf1QHXTWUw_$
zLNW%ymMNH44#+fLTD8L>k|+>o29v-L(SZegNm4*!b*T9tY?gK5_;XU`y#l|kIm>2X
zGiM`-pt-T{OU<*}#%zIX$ON`c!~&fEH4sWmjp%ffdNN=8O}caM)yzKW_31jAE0e59
zK&7ghr%;EGxMBrX<TJO{5PicMsO<*tO5LRf$qq;!;LQ`MV$&RB9VseDULI<)H=#GI
z&0EGiXQjZCu+im!T5>dA!KKJtkx7=ui#886o1h^)^%18*p(fZB*5I{yt0)x9xLB|#
zk1_c7FS3QxATKQ$GZx}eMf<tf9Q8ohz05<*56g9lg^#-#vY<MptDFO&o(C^h6F;-7
z_aJehlE=w#ScR?C?$7I4|7-l!tDr0cYI2MkwkKJDlj~j@M5h$_xyIHw_d8}_Jf89E
ze$AhWfA(MaZFoT^Doe7W=R|d?DH&XEW1T}AK>tw7`hFMnGGr?@u2;VITh;>%7m95w
zBVB<-eYW<*$9Xj~c;oSmBk(W%{y%`f^3VO%K$3qykEyRqSl9LXSz{=QFAbNgn4cIp
zqyB&q`&J;Nw3b=pnx~<KQzNQrZ`RoZD$-vIg_geH!X+6}brd(TnYiy`zNC<rV{ue5
zJ5PAhOt4;_nT_^IGe~ocb&Tl-KXlo7I&nB*FUFaHY_{N?Vgzx9I5o?b@MALjn!#Bh
zQ#P!~9jLvfW#~q#DTe5#&W6z1s)TlUHZmE^nx2P?{nuPF8-jt5EE~5vm#M*7Q+*$O
zTQ$clWmzZ-j9fI~ulD*>PN6&}P+B;~FA8=9GR>361=N|SE<@>c%;Q*!hwhEkvLwb|
z8=0@OJGlyP0Xm3Tk@=jsKfhZII+lTPOk^+uon!L4GKnjD(e?NQ=a8WdYNa>R?A9`2
z^?oFBkdFH@b6}*v`KVle0IV4(YA8{#_cwk`ekwSnAffvAnqGxExFCN#gNAi6x{5;{
z<=_}0Rpgk!3j@0xwIVx%1L=U*yoiFQ8p0?KW@(G74pgRtk`6;!q9IT@chx#p!BL0B
z(pN!^1$r5Xyue11*r0(si<S|AiAH@GMZLfD7ygO(XaDcN4JW+D*HuCE#lpiDwDt0@
zP0zeOr3_r%L;WocC+&NzU;^4BxM(J=CZsVCQBa6qFY{#>{URI5rE2QlfY$Nv{U<+w
z`*eKf^%LgtjOha{)GMLlJl=B(XNI7smI=quHJr?;ZnpXfid7d{)0$PXu1<#61*;^m
zm~(4A^4#}OMgx6^nG7Kqc^AXX0A+BwYCX&ttyY)zA#1Jlc>pXL(gIDRHQg^g44V|4
zwbU~v<{(@OXnbwW3h9)pXqp*QnS9Y-O^_LUiY#L1-Z7FHZnqoSw(+El2j=~bImVjm
z$TdD;%uItxUkii%N2PFUtrL|IJFl6;ThtQMF@<J28*tmVmGRCvgSklBq<x;vu0Pr9
zQ#pl`S}JzVsNpZMn!nslQ^DkAjWZFd?ViIB5;g})RL3$185RDSbSN_~5Q|dW6fMGy
z!{g_yCLK0r#Za*dPWc8a=7)Yr#l2ZYVqxywaB+5F?>ilYqFhEBxs0za7v`6P4NgUB
z>Nd6&vFCyk5?GQS3z<mr@#$CrPh$VM46O094G5ESo99gE<xDT@M&^=?uaz;YFq!M%
zy`h?M%F&lIF>8w@<*gy1!&M?`sk|UXCxsekE~NlB{_G|EMh$)+HW$QHaibaF+f5mU
zvKK*ik%7%OjJuOv)pvZu=U(Aw|D2zV@BQBI)%t5OKe}=2Hy?T1f&(w`-<3TB6olmG
z&t(YyjME5<CRo|Q%5f6Pj;i*f_v0+E^Sk*BR;GjVT3X+5pA&!2Klxkn*ZoC*4u0Af
z-e4XNK`xwe4V>fyey<uE-PB>M9uSER&OzhiR?HW0W_BjF4Tv!D=GjP9gQ%cv-jQNJ
zX;7A(_Zx8$8gxhYzV<4qu_r7{rzW?h^>Hr@qHK;WG|abs#b0S)*S$^j&NDV+BEigv
za4DZf)Ab+6fpH9ko_7uTA)%!pBtnceIIL?v3p<UILYi|B28F3Xuwi?8!qe+lXuY!`
zR#`>>RC_f&7j@$+kQb!c0Gn(`Ya#n_#>DJxX@bdh=4-f@be4LfOuUp*w58c2VE`hx
zkDiT>d3|d5C=wtpo|L&3bs+`osG_&Ldgf5-Xl#B{!^%nsQ0XLNHAd=*G1{pG4?#M>
z;@>6`IYY%&sdvuF8KzKbCxyW-`R6Lia<eqRh}ENn(nR%R%GQOo%P=}niiMR00Hxa_
zbwU-uTif8hA&-NmMNvV)7sbi18Uj<=inLZ&0&3t)Q9-YICKWjyK~}*TeKX1|XHx8t
z(@JU8i#23rc<cTAbBI!W+E2k*ibRU0xH`N=Ej*)perKy*jJhkWL34T#8WWCVh1ROI
zrr}9vhSK<DX#QG<*zvR8d<MV#7ylA`@Av+v>*mhjyPi;e=}YFt@&VL$U}SDD-ht({
z>wfA!GB7R~t6Q$ld>8xyBPLDVEq;cX*G4)Y#PucI9%X>*O5EVq@qK^nNAd6fp&!I|
z{)|5r_xqO+*6^6gx>}<n&%9(U=Ta<5m5gA=`Q>aaR&ZP<nMvBC$0R926yT6Egj8)w
zPrf*X(q+nGBdPyDSvL1Z={(8c1#L;<TqnN_@L8Ta6F^&MvYEqsplyK$$yexltA=FK
zF{=QuwT7b~C};2bYz88#JDH2|sA7Hdb-g$SV4SnSW=oy<eWTKD>&hJ1PXq6;7BC|P
zmFfu<rw9fL&RTDKC*xwyOGd@sj#&5GBp)IQyv_w10cj86434doTWhn=7aRN&U!QWw
z4tdKfU9C-AbiODIt1#pct|d0c3``6OX@$f?86h!cR3lhLq3(cq_>eJ$lUp80^v$W;
zt{v&Uql-e;>}(2h2heN{&Y6l_8MV)4NGp9VRnm$esxS*}yd;xW3<(*0lKYv87Wdi>
zP(~6l6c0;lzU+)~#ULnWF);vPx|eLjcsLuHx*7?Qv1R~87K*^+_?Q|uWefUTh~?nt
zAtVGxN~<8~6_Js>b2ci@3_W_2kvp630IcT$$jICEa9>{M7s#Qg*rv|vaxQCr0I)Qs
zJb{@HNTj?LQommdxH7@(%hs>wTBDZd#WmXHS7nqDgfJyVfCzQpR<!AYJakV>uGm&^
z^nNz*ve2=-_Igj{EZ#T#)Bn=%#!vsIFW{H|$={CSamSN2z!PR<1oGKa*|!1*vzneK
z$YNpzT(M}-rbHlRxT;@7bv|pd%Yt={k!s*%(Q!EfrH*=C_D#W7edg*Xiem0j1hWiY
z!OGX3kO<P9L-oG4;J8{LocdA0+Jh@HH!*R>oY=>{Aga?*;ruKGBdu72YirR~nGFUK
zG3)JO?NK+(!CDFXF^a}L*NkcoPU3o(ynSo^oIf**iNq+p0%GpB<T_Qq3@Y1k_{Oy@
z(h40P(3Gjo_H}Ryc??89^oROC8Y@!-9TA+n+x)WE8AZ)8R$+B*qpYDBwc$=9)mleY
z6LgD=a3{@cYmKzYJm(!3w9zVcn<1EooMG>sd|^c%qm5NCGuYxzuglZtWhM)MD-?~j
zsVKQ2gVy^AOC_6ja-K|&2%s!@70V1b3)NN!L>Yo27rhBky1OW++R!zj%=GU|o}<ao
z>vI=ITUdCiM?;*@)@q)E-eU<jai&d9uL_K<F=7%7vDRz7?2>&H8>;+>Xv&JV>p%dd
zS0sFVMcMMA^;}X=B8F^2zSx&FZ}?*)a$Oq$*n9Pet3!j*^vW{g;t;$wG^9G(0oQP%
zTH}+AR`Aou-NXCeU-wWM16E2?ckz4-{Jp>Vx8g7P@~^=!`L>^l@A%elLTqKM3X>lz
zg0E*;velvNpp17!Vbr-UYROv&Upk>9U85<Zm_yG~g-{UYtvL-<tY)ndmuxnc-dI6v
zYeZ+}x;8zKwKgo4W-Z(zwB%)hte(HLz)WjcQ_M(>pR4n_YDpD<x!0bpz);L8U4?H%
zL}mmh`c0B#`qr`cos$!t67U%g>l;!*AtE^)O-w+5Jut4Cf@PvbV9w4C_kiYd+X`3`
zU<PmrVrpGoYrE<Y%W+69qkSD}3iB8ki!H-+s}wwIo-zOJ{7>5m8H4mMy@PO$XqrLh
z5M<=ci}D6Yjei+NLQLmDTQo3rY}P`R=+}CnCJu4m7VgVtkA$59B6MbUI+#>CT`~dg
zT@T?J$#5>TD-+>iG*L(=h7B6jt-+>R_+=DJ#Lf${>KD2%AfYiaH7+obB{}U&dZOra
z?JQz1NJe-zpOu&+s1($}uIIC{g0R3u1(D^Td*i6ge#tNZQy)n=x%y5$BXta4okH1{
z>L`U_u|HN8Q-)NBv?fULhiiXK{OUjL7vW$1p5KQ*@<)HTa;nqeT{mz2;UmG4mEHa5
zD+~Lcz+!JCdglnyj9`=woabPj0lOH~)8M|m2Gd_vV5NUF&@q_5%z(5m&yNHD@IU=q
z@!dcDoAL91_P641{*8Z$o?o$o)K!kN09?kdrj}I}wPa`#x>sD*5pv4v6A&v8;7p7u
zT-|hU#e69U3OFgbxdI-Z$vGz?bIso7dup;t`Qdv1c|9ixcwquU0p+#cm3czH*K9+X
z6lN~*z*?kGZNu!6@#|+o@DT$dI*T#J>b+u?cy?#QSPQbg0U+VPBB(@Auv;_i`;OL$
z^`{(7%&8!nOv9<-LsPRGXJeEO*+m#q_Mz69cEqUsYKHXV<97K+UZ1X}Fb8s86h5Z|
zJ?QCLTbqO1swo1HVxmFDTzI(Y&}4bJGxb|gLMKid9^5cVPu^MsLJq;Jz$V#)f=NOV
zX{rlSj>CIrH}9lIyoY#FDJ;}!ROzsws1R^#t09PuV`MpHGWF}*#s)d_<*7-ZoA9<F
zVq)u(kM1;TF$PA#KWZ9d*Kol=AsJpesVV7;Cbd}ea;mY4y2L=McwE?j>irdw0gF47
zWKgZ?aB>QBj)e>JId*vG8dDa%^O;x;gQ>_Isu9tEUE|;y$POq1YARCKUO-V4>oA{i
zi%AnXvwVV~!|?6j^bPpzn^*XL!5Zru%GiEVq9Es!+OiAQxm;sq2^pu9i0k=R07Qo6
z8Ld^Ze(goA>HDwua)TB9p1%iHr<vJ{wF6i)`%k|A2l0o#{|7K4@t6MkKO5ivvwjwF
zYXbEs!%0+EP9<Vx>8CCkAqswdJuy4i)7#nu85463i0{1$%264EoPPx{tSOu$K9`Zt
z3oFkg@6G5@>&k1WI@y`o^15C)H$BU>*Du|DG@#e#k;~4^n2Rk|_Z#Zvsm!WK9BI&$
zvsle^71UL5SoMi>&)4?^#&Mv}>RI5k>Akaot?D&nB4TiTC5t_#WP%LKQJKATbXV49
z&B(a!vwHu5$4B<z$Gtvfp_$SBO2HJB8LEA~(rv2wb@8V&ja73Ql^XVtP>}{K)Xp&!
z8K1cFF#{M6q_%SJ6k=0OfP%KK8bSara@_K=;I*_?MPxCPl7-rm=53ALxh7tA!7ygj
zL@IX2oQFDlNuz9?ZR@>l@ERPn17jR=c%cGzLkim)W5ACC7An?BP;jwdbbY6SWEIV-
z8O%=%R#}cI?XALK9j4OAKkZYop=>#jdT-T83J$6FwAQ)C#jm!K6Si|+a7D_BSc)B$
ziVEwyvsEGaRADu(3Gwc$VNlS#2Itg%tiy3aJnQaW7}4uLT#TJQWBnml1!wJJ<QY9z
zj{zU<M8;U!PIbTap)X;mi_<w*|HXQJ%!T4~YX07z`Kq=3?Z4;Wz;FLOzaM|&U-298
zU;A~x23>1bllls(;eIV=UpN2K4Mf#P%D}@A<|KH)`P|%qEOLO(n~Z5Ln{fHPRh{sB
zKfm8G)hAP#H-b&3K(4O#tA1GI*&`8)0$h%*vKFzdPBse)dai&L2{y1szSt`-B*2kN
zHoeF*(B~|96FEU~1qDM@AXAO0BqpGIy&B6}ZwuCfkcBD70AT5wmGh|#rVG}hGgf^@
z&gv}w7u!wy6Na1JKVh%Ee0?gXV2VQfz9W(B1RmquQo1=a2e!ry1gM9AH!>&CcC7tf
z4L39bO`VmtqV4YV^mFpptubO<#XMo63p=l3R)mI70Hyols9uViyIT!aY<M%HjIBu?
zfcwi}(l=?&S_Oj<wbTQe2*z0w+rF>+nVIOFpVJ_?%{1)WP2JSTIYp}bit<wo54XaK
z1;3z-6Ht|z83V{z8O3SPX>gg&`YPV~a^O-_yoA@8K{=4l5vXkB;|A}&sd(TrgcVKU
zB2Wh`ILIjYRmw1#^}JI2_jc(n=G}BF`0CVu0?GhDhZD;X&lwtgqWvU@A&cUn3p95T
zjdB%PsVlX<o<_yX=;Vv6;cC$nGDhop*!hkZ(Y~!et}l7d2dv(J7rBF7-uu48t0?$%
zV3ig9@BWE@25`e)_~-uFxZPd>^BpqC*mE(;EI_qf{ZF8guAjX9A~l#hY57)E;)Xy-
z*R19nsBR_a$t?Psbmi&~szoxKdwS~^JuTG|R@Peew>WESA;bVU3+81UP3vM8ShiKR
zYQUmdCF#Idn344rQo2pos1S)_m<d#!rW*b7vUcmaL`rvy?<tN@D0_2|O~jwm8L9IT
zp-e0kjGRgYQ_hdF_RQ|DaP8U?f;r=&n(Knw$Je8#&^l>XWPmu}196)&($rwDh_Y`R
zyH*v+(&#o$ILef6dalSf!&aI}M?yXFp21qiIFMSDzVGPU?JP7U7L$glwwywQ<~k_i
zgz6Ekv(W{W%T}6|iyvLp`w%>0)%9EHv7jaZFFJ!wUO5x+3}njDnX)1SyxmsBG8*N8
zIZ3S`2v@N%R;>Z3`Uz0YEhU{Xqhrzh0`SFTIQiBE(UkrVnJ5<;V40YZoJQ!tD@aWf
ztq&JSsW4O!mUzKBDu?8*L#VX90TrodIVg8E7zKP9rp*Nvmh-a|o2xQGXhdo^{ICAz
zzaD??Z~PlIoP<S218^M2IX_uFDj{*f%*2GrzbRvM(V6Y6ZJad{tgi2ZB3429YSZya
z(f`MIgmdS8!h5fd;{A8L@3ero&v*QN|HwZD!0;FUx?iKQwL#n?*$2?Ni#&Lx1e_TT
z>Y3=KPdKd0!IRC#$ViZdm%Mf^81VY3oHZaKL9x}WUA|VeJ7MXGv-b#0O1K*CXcnti
zh@H=8Ey1B{$N-Y;Bi`gSi{sx#$*C+KE*-U)FIGE)4VP3)&Qln%`cAId-}QM|)r-z^
zlX6_lYfQWm+J}NDGf$`zDTpi2Iz+3XJXVr1b6vtvk5;>`iO?N0$DSXz|33QqR8Hab
zho8ag*RQzQtho25kuYzknx>$dWq7p!=_QUgWnfqio5x&>dW95XP8{RFI1Uyqn`7Ib
z&~H!Jwr%O$SMeB%ZW!prat^69(10bqP!V%mMy?_vUf0y7Dm@X1=1NWXSdN1_DMh!-
zRWM9XMaCfx`^8{rzq-bcmH{k@WLIQD!_1mVQ{;|hHqbkrLJ@(-I530jT#ZZC8oVGh
zZRDn2k_JOI)hA@M*$v(n=l&|>SI?#>erD$(hl?d~JtT6(0zQ~At}N<=5U3vmAZWBq
z)4lBS^Y9eIi4nhpnPq0+_0toIp0}QXGQx4cV*wYJ0HPqFs#{#)pUkT>s9uRW|Gb-;
z6;MC($uQ96VP74C49Z7;>ZHlO__ym{meIg@omF5xj==xsAO6Sj#V>vlzwA%>Mfl~v
z=ueWSm^l2CjiRw?Fy&lMXcV>IgQk|mkGA#?w|9yzDM=ZYd@dORmI1V+l)5z|`=<Ie
z%#ec)%c`zm^a3@jnxA7@bDhamBQeQ;C(LCW*YrgNG+N}^ca^;~+fXEcio)xB)?Nov
zYNVurvN`jFsH-t|J-N#>UQL7m&xuici8iC1|EB(e#kg7pda%Y>@;}O+SyR+|B4Bx%
z@b1OvQr{KoEGSs@uY+{<>kmG}n>TMr&2(y@5I)kPi09r~!?bE2SITn1T~8`pF5`VP
z4J3C?>e~9oO%ybCV;n$&jcEM|zTH%)>2w}rF2SdEc~CTTVhJ1rF~<tR3L9tfx~q`W
z*JcTEotqlXQpUJ&W~R!aP3Jl%wZR<B)y3!b<1F|UUDOIXyReE1&kRQGbVS_Pr7^P%
z8~}LDTmWD$c>x#A-2pw1(;U&cDfGpv#i&!<9Eh|v{6ss8_mag{2@(XnOt!SGE)8}t
z6U4yCd{I^l&kaVdX|Y&on&0w*>t4~EK~Rt9+24{XGF}>;R%VbCKbuEeb){SZxWT(G
zsw67PUH8ZLt*+|Y-)j{M9;o%a7>4)G%=@*Cl43G_$uh7X;~thDcn>OC8O#OZDg%t<
zdHLV}Q@;iOx8L!v;Ai}_Z^qyExBZQ{-QHl|ZfM;EV^D(r<ntH`j;ul}8uKMWYQ=8i
zGdBa^XOYz!R-Z&vv|i3uL1Fzl1(#Wp&W$<MiHr-`<SdrfvzuUYk`->)8FvR_)k?U<
zPQ%Ub#DZHQXh0$1zST`;tXagh@o!yAeZ~o=0;W36l~|zw(=LWzg?a{Q+DUbr=0JOy
z8y38YNYYoe&sV>e&1FByTZQa{X#ZJqJPfWN4J~V52li1t&_J%PTn>WZqL5e4$g4V1
zPiJuHbcLfesJk-6-KR=98A5`ooS)sciEWl@sZ022Tlg9PWFz(Ke42`GgKFwF4bu#L
z3d3_l;f4ld$7kQpX1G*+!PH7_z+<wBZAH)qVCJH(WvJMs?=PiEK0F6vV8mGGu#9yU
z%LIhq{E{w0Lrz?qnN(7&;*vEU)+n6*YRplnNjb6<_8uopZTV7XU5qDv+1iYmbr`t=
z(=;gE>K;xUzEb;&S^!++M)<id-J=B7*-SIcdFa_Gm|lCx*8wae?7FXUe_oG%9ey=W
z^rsy)AA#AR!&HNCgPNo?pBRb~4NDJ)&#`hHEg7n&Qx5u?9=&t|m(znYGPiTy19Z_?
zsnhv>MV@8saaqf&6T>o&WeEBHAO4Xa#Sj1JkKxbw4SzL$-LL*t_;db@Uxi=t3x9#A
zvK`fI4L~(6te;<zrWj@p_M}iYOj*5XqgV#AR_R;=*9S7et@B@B%4?p$>bYsM8>0+0
z%IrhVry=vC#Y6kk*J>t*6=}bo=aDcLoVrd^SvcXtwj6Cls`n{m>?`69U2Cp@#vLzo
zZnN$+0@G4N7Re>GM5e>0An%n0fmdm{U{nv#$#sWP6*k0G0FWsSxQWJ?jGxoML1ERO
zp<rjmIQQPy@cNV;&3=xHOBQ#Ulunt}n+ke{?nSGdfHh`|&Y7mQF}ydWNDhyQ2p-yf
z>(a6B06I6Jv`v~e7*i}oojMq2`o-f~+C8Qg`!uKYFs;~%wlfVmjS$GSajh~MrPvXQ
zjLoao$_;=ric)C3Y>+9O6iGG1$8jKrbc<Fhnkde{I}5NWXwEYbVQ5V)>|*~EGs4|8
zV<4ECpYf#*Bd@CriuzPf*dcAD%~toh@Zxd^C4X=#n!+5>J6d$W3`gYZ$~>Eo3}fDb
zq+wN%t^>%!8Va4gqVyOm@>(wwR0XV*ePPPRXzI`T&;D}!mVfzsxKB-bK=U-H=7#;Y
z11M&W_5fHv=#}EPGEibfU7}BZR55EVvHjIBFDrW*bttPiD>wCXm*1?2+TQ<1eP=zd
z3-$N<^OZSORIcwc5NrPD|Hbddzwq7v3jPOw?O%z%;Me}y!t4D-&)?GV(jKVZ1Vn&?
ziulG*p?w)gjR#F#U!6yX>Ogg0T6ig4T|$f#mRwZo$*%0vJ*jp9qaj=wld$%J{wi}&
zM(fs=16|$zd_d=A)#}EOtjDDYT~M8+j!<fbyI^izp9x0gW6Wx$!P3pE`V(?xPYEFn
zT4vW8nuBYaV1Q;P-IezSWh8wSy$RtZTC202tM-}mto?i~uTSX|sM=TQwZSl=3<nRh
zhf+^r!fr-fEG9+X?4%I`EDBE51IQTJh~EK78{mqat?E-BU<NXa@)X}%pPH4{p1k~w
zi98;N<3Q$=X2{v;O9fI#j3#yxdl|@LU{69RD{_1ySrc<!8gAsMQk%%l5rcHJaWF-u
zL`TTMQn!i4Vn)?7I5iY91mP#Gt@WnnwlnFEF)&9#Kb&naC%Nd}93xr@9phvz@io;G
zFp+05JXteoR!vB({md$ARn5g4+MLLI#+o44oP$#UK@-4y9Hib>yqAgING#9}IJ;P*
zst^Uk+3Kx0d+P-4RVoechd=F4_<8u}zWaN4Se%1KqSVA%qu?SG+h1(&ia^(nEC{bw
z>5RT#ID-1u^+tQ|U$4K<`sNpZzvAHfer1OBS!Iw+25$Yt``JiMA}iQVG)1$x1Rit_
zfA>H1k0H(QU-@<a1z2y0a3rKp)e=1co06xtf{iA^xKpeglp5q}=K~y9Sx{Gz7@9Mq
zpBZE&Wku(_$L<Z*`ZCxftC4e;fxiSTW-4|=&I#+>+m#I%d0FjZlTAs(C<9LJQ&)Wr
zppCFOkO=KbWn<dAAMFdcLRgTJDfduXAr940nrHh&q<W&XdaiSwRW}O-eR7?P3Vosm
zP;0=oFX8<i7}`U7zDDH4fv*Gm=y5!7e_T6m&#VE4v-sw~F;vvmF;8QPKC6*MFhk+c
z><V<W{=kq_#A>!GCv=S>B35PI-#aul<j0$%SI<DwF=0#~r;3%VhnQ29G!E&hRF`0$
zy$*8@jDxDIk$|-gy)jCzahT-{WOwASco>i4Si0wEhF4Fo;N5U6D5zixXz0n=thYLk
zXTXjVYC1c9u<ir8*M5muo1HE1Oaf%Et9fS`nS@m;TikYj6GoC8K&Dt7To&G&bmq5h
zQ;o;enHd<51IO_|!5<0I5Zy0ZuOe!tyJhsrNUS2iipKzoErg^*N5sTYwW4!1IL|^E
z$+ZDW5eM%#%lPT*1$FY@@ZvKIt~l5Hk`-M4yu40D@YjF4+;(0fVMUm%z&oxpF{)Hw
zS<1Se%9!+84P(kQ8GtDA)X%+rzT@xs`~M-ldw#~B`V0R={3(CZFJKmC8zR+I*x6~G
zIdpwDM4Xysseq~re_}(bp<SB%n$_SMv36W_syfq>Rk4NZ7o;YqEXEz)`+t=4L&C3*
z0(m-)6B~uWq&GBBvs}uF0XWZvB5RMa0+nm#b=ELyFI}JGTmc9f+YQ!seTM>q9B6{M
zW|XRQL0Fp6orK6Uh-S1-yHtG>SL*l+DG=?Gnh?0oxw0X5AS3Yjx=3g1;a6{gS!o=d
zhpjpRF;C32n%t@gl~J^^GtFb+Qtz)}^pjlw7y2y8qSBZXW6-NRY^C`nV&z`qQWHh1
zvx2h}^*5c@)##C88C2J5WiK%)2Ob<Oo~fc}@cNJOV05$=ig9rye`d^y#|R9)Ve1?E
z4zzU5t|eYO<{=|}>7p&8-UJ%efE_cqD`gk`tRh?kaF}*VVimMud~ZeJl;Myw)$^$W
zcz2+*eijwhpnCJ!)meQd(>PeX_b#d|jnw1$j^lW!nTVf1w}6`*vb`*@U!6iC)i~8v
zYrzRa;1o}b0OH!wezD1`sH>l;zK1irgQ7t!1Gi!>EAOtGzx?tse~|G2YEaSn_ZR1J
zalm?b{EAZI6g{~-#d|pc&x#y#J@b?^<#!8|DJYjw03&SE_od<Q`3L_o{PdspP52eR
z<WI$CKJyv;roZ~H0<6Ke8%MF`M6+t7Y@F^^Ml!2WQOq?HIROR4T(Yrp;0h5-632_|
z0A}JYT`yN1F%;N+Br_`q`C@epnlz+uSu4EO!*VB_yznx)nu<A-`^1V8dRh3Qnb&8&
ze?<{&Ov{-3Jad?L)?u`dXm@C<JgIb0I%EhF@*$h!n)=uJd8W-)oKJnZmyT&=GFo$(
zU5q>TOTS(FT|W8s={lOz$RmzKwzf22>grA&{dwT3i6kyrLJ6V8rDKo5M=^3#3<Or{
zTynmdAQMP%MuL%ECX0p@ZB{@qfTd<y1@Ay&CLMz*;y7t$&`7@8G;$Cy1VX}deCWxW
zhHs6qD*(i~p-YxIvH_|IFfvf_+xuCh)_smeNmHXxa+7reLdya3IGu(ng&)>f!P1`b
z1V)Va_|B601{j?!`=`U;i4lax4h^tB9j&7%c65XjA#MIg9B?-_@HuIP$DF*U31uw}
z$fizoK93jOzygdN8hqqk-0-#xO1-s;nB5xs)9Z7eRwi(1B6<<rmmuObvR}#f1rn%P
z2MhVud*6p|aQ<_C<S#2yEJ)*$v7qWb>rX&M_mwRyN2pJ{WajmS%NfY&<XT%>?=QQ&
z<fAh2hkxLY;`@K#|Ap2Xe*5qIR{=Bpz5nz7F+Tr=pMuYP=JVM14+)3LLX!0r>zo&T
zy38}XY~!SC>-ntC^tHy<0RKR)X2?y`{w_g6y;n|<<OZloF|^8#T-R|1m;B+Nw%LN_
z^5uG&3b+cQ%J`q~Q@I&IG`2o=Z5hvf)!=|CXgkc5?Kps#qUHh?B?KuW6}ktoAiOC4
zeOnNjhN<!zyqw)-r<nn9K|YH53&Q@9f%fj}(9zt~ojc}02N2CMDr>)_R)7j5V*kwK
zA|#SSYuxGS^fLuXG?ea;1nH8EER1>pWMQv-m|N@mjKwQN=@ymJDqM^Kqk)Gw{O*#N
zO(4~fQrS&}A(|`Q1z@SD{E)G&Mm2Q{%E&Pa$J$s*7qEMBN>*)lB=-G=eSbpld|+b?
z*l3qxA68_r(rIN2Rz*D{R94K5ET1S#QLtU82t6->YIyO^^V+e$j0xw^HgVQ-J(tz9
z0ezTh@?L19p!!v+wgIttG7`<G@6^4)w{pUw1q<TxUMHHH+NB4DqV)M?ZMk{it>N_t
zKZ0v#FrkPOPO}#$NJ&&y6sW)U_VP{ZXD(KzjP#3(xa1ul@fTNT!btzr+qqs(6&VZ4
z$`>;pa;0keD>?DSD3S4Bbp>S_b$^wqzgP!3t1$=u(?9Y9FaZDMzw9>xX83RX#eX6G
zroZm5N86uZRw~7;6_g?}%3x}*K^@{GcKAx_xXx#zDHid8bZrApTwkm-*G*OybZsKM
z#C|9!n?_p9xH#8vjY18|?bX`B*fXV7P}YTY4$c*F1wj7(Kb%2FU0C%yh%Dl3)lyT?
z%8PJ~ASa5oZ&`5;BLvvCUB-TLm{5+DDZVdF!%JX&GOeiQc4iMWTVK=b(@|30)od~8
z>7r;`=WC^_5sFF_DJVS-RpG1Sc5!dP16owp2tq_zosSW#IB76N_bz&!q@VR*{XN4$
zbKI1UX=r&%&aBIm+L(EEGX^v?285<Y0>0gd{mbz%pKzFMOM}FV0A&U!rCt$wHGnm$
zs@9>b0&ABR%8cNU3<K3yrG-!3bRNi(1VA~@(u4_ad~Yf+M<RO%dp|i9k1Pf07F~v~
z)ZWQRw9Z=1RC7`HzB0or1+!~Yj^mEAD5ihhcOFDZWi)T-x)<9j>tU^6t3Hul@C}_y
zYdc!+h+ZOPwaEqJNvLpt^%~cV;LH&6#r4wAR1{sFLGj}SCU|e>zxUfq+Odq|`T6|i
z>Ns+B%9eqK*6K_quG}PB#_sy{_TmrB>RuL|^I{Ax(h^X1bv3T*i?M>NwLX#fKmF$a
z3tDUVoBw-%1JB(uD2~M0XgcR{I?3t45{gvB>s3=M2Ti-dJsXG-LIn&3B_SM;Xntx5
zS5(i3MKd0OT!v5mK5&HR$}Ge$ec{-+0Cmr{GGt~TmuHz=XD-pn>N(fX*59!UyD}HB
z)4C!zbzWbn0=T|b-NH;k>)HK%?PocmKt}&(7Fb1Z!nw>AW5Jm2TJNvNjAl~?W*{DY
ze>ZS@dc$I<HUl>%vAI)+-BweRYb0Fn8+t$6xR-WKRnpxwGeBHpOTf;i5<-wFTv;(x
zxB(X58e}MS*VxJlM)s}4JkgDYygC|f>yiWSiacy}V0+(@1~jpan!b<wYUg;*49uCp
zoHSIf=}f|L3!AKZOltw_EWjZh`_v5CIgy*Bz)9@kEQpGFGLgxC3u~?jv{C>%<JDm;
zXijGdGZ^*cD*A~n^nyjj;&+T-CXj*NH*EWczF;Sy3XF-wj@M`L>tZMUVzgMCt|3IK
zM@Y`QMjH0MuX{<e1^q3F;Bxo{Nf@)gzHKr-c`?WzyNfSq%xMf?7~B_G0&s~gfEVw(
zuGmNaZ9S8sII6ZmUuCB;EramBL%L=$ALp#Duoy&@P2e1`az>Mjq*2b_u!6Q1POk1J
z6aVOM{-+Qb_-lXTe+#c)y@tEv7!R~8sStk1c_vLYFYQsguC?ByY)S1P;Os`%#KeH-
zSO`PV8!Ren_2&RseI_+yp(X~V88Epw=?AAYlV}$T0$oMzbx}pxwDn|HqopO_wZ3K6
zQ%{9wl8mgYj*%hRfkE9jm%v{d=7cD}xjLko2r|fR&Vfk_f{Vn6133rbrwq(_ya>Rq
z6k0ktK6AqLrAzj8prp3$=$n-NaC**{DPHk<U0Fg?#>57Z;v##e9!SgcA=IHMQK`kN
z)kb0jVdZCD7+F{33)o@~0FK@P;KVuIjRmwu+9@Qrp||s4<@xy)7bIA)Erru+%jW^}
zWrSM~WOgid&`6ZB<BI;wHCm72e$^ZoliieaOk9;AO<6+I{aquLwf^-d=hP}+hvIa&
z3Bp)>Hp+A)`4C>;Cjx*0=?xn;xEbY3q(>E3h84sw_gs6Ro@HiYyF_GV_^X$t`0NF7
zm^UfuUhc=(<=ifNhpc2h89gWW${?!dAmet)65fx}PDk*5CUE_?t3x=~{N>-}<+%X$
zKx)X#3)Yi<v;4tJJg5G3Iq~-$R2iz5KXu85%BC0EVfMoAS0m$#MOo{0xh7?a|Hwb_
zn{hnu_~pO!m*Ttr^j`_{j(}s1;Qp`AwK^q0%>=JcJjo@bsk&x=zyLR@Bfy%(3V{@)
zlsFyy>P@*gEfRi8z5@%kVUw+9AqSh)#MEc~(rnS>VYVb?BF_xx#q%h8dNI5JWO&cD
zT)RV#gkUefIYR%J3HHZ?lA(~eooW$L-%S)kE4ZFzLAU2#u56*sL!LjcJ_?dB;yrlu
zlV6{%Pr*p>3_B^9xl+@L98@K5wYe`Tc>*2@xU1R35KR&1EC-XAO1+jBR9bABp^7Xa
z?GuV3m6Tz18ACuUhYUnCd1u3F*O$f+P!ujTX|0W+=s+1k>nFpZQhzdAeBI~Si3}hn
zy9{NxS3G9hikeR?A#Q3)5YH1<GF#5BNm`_uYY0nCeSIyK!>e013zx3q8dd{b{Z|sa
zWxNtf!<F_^uGd$i4)kXO$8uT`1FW3W%jkK{0WdMsR6l^J7IKYJW$a&O1bj=(v)CJ`
z(yuIMP;Iu1<!L+4sD1s_i=h6a&fw*5-!uNHy?AjfwI8i|WAv=~UApG1qTuob=b!5{
zUpTLdN@`ioeP71%Wb~YdYO!Rl_qEplqt54=A5>(2{h4wm|M)-s&*5MAfBz2r@?ZLk
z@z?%Ue+7Q-&-pfNd*^dmP#7S~{1Ky?Go*V&C*q|Z04O@X6ssp-EoDV3f-VLQuh!~u
zL;zVxgDnOT0QE!6Ggz&f)H=V0lD@Bv!5<C0s8s7aQ(9oBRalR2tz_B3>+?W0>IL@*
z>nW2?Nk7T<SM3Nfk}U0n&b8+#sA;Q@>*XS!&)LlBxCOy7)5;}x@zq|RE>nP^D_X9b
zE}zZ~>!(NS7gkAXEmHnahsE2aNY(1z5-g~kH%a;1mnrZvyaqah>h>)>RVz*itHFGp
zw7N(&jtD-nQ&!n36|I{#Y75ag_XT%c4;djW8Y(Zj2P9b!TApeY0}QS2Yjak#-o#cQ
zR$2z1_vpTM9}kD9uvboYYAP1(sfvnREt-_zy+NJb#ULuI)`h`lRy1nB`hM4#QL-@D
zi{Pp!C~KTk&lV`I{BkT>r-GU&lX4czO!oyD4X(|Jw<oH>E>O(*Vj(3e%DWbCUpf@5
z?|=FFxTx^`_b#Va2IG<u*CxC2)~ocp0`M1ELVcILxVBHahWg%O|6n-+Em*#HM=!9N
z>t~+pVT*Zn{>b^+FCEQdASCb;Kk-HUi{Jgb@caMm{}28T|J(mHzU7;KDn9?YFJLAO
zPE%)noaJ1`MVWBvW_jBPDS;4fApv24s&D%GOaD#cWi+0uy`;usX2LP~^L9E65;lw<
z6;*4z^rPfKdjUn-)odE)GmoV!MB`2f8*7he$(2-nryTg@xm5Qy?@Pt?JOc~ErPk<l
zv=)PN<b6dKuyRFq?w1$!B<%+l#2tnP^u}|pXZUe}!N<Km9VHc|imIq{Mcw?;HCoSr
z5Er@CoLCuSB-u#T!~!Z$81*!H@n^A$QR<lLn07K!P8GsDjf8XQt`po+hBHzWSdax%
z(~ii?uP7O4{fs24Q@Yqt0Tz-g`Y)&FE*?FgO`@VN=5kOTQL)hCAJa#k8<tRx#J!Ld
zFQZjn(j4{g4y?wY(~%e|V=4Nhx#(;Lz<FR(3?W>iN^fNlY(-wHGrSS@sqCU_{fYTH
zSw{Xq%yLYoSqPQF0^0r*2aG`0P;}HjGIe>L%!#_UK*D;{=PKH()Oc1wQzH%6|9`w=
zxOk|q{<=V6%L$|$!^J7Q|5?|70_*4P<-dOH_uGr#u0Oqa??(**fBW8nSLFZVGjui$
zB3oRu2pOMiZ^n!Iv;OkOe(Xo_=l|FLa(vgH{wwjn_#gfE@r7^vshHuMY-k4LnJH9{
zO|Hxq>)DcaPS~XO2M<aIt^$YGb;fzgNLxKl$WZ@~`U9NTaW&xl*<8O?fQpOsL#i0P
z-nXO|lF3XtkYtY`G!3x6D76<}JZ(7?)<pwx33y2qaEPzH{=UZ6W<2z0Y#|((co6`d
z4SPUZTg)TcR$r{mQ>5^#zdkklXfcLl6kAhqE9`8!Hk0Im##+LbfEz%alH#Eo*ol{O
zoZR081+jp|Mj9$$kpb1=G6H9~<seQMtFWsTMiwiewN-#Q@2T*7es}9_O$bXf)O?4k
z@~1HPf~h@bm8k#$002ouK~&DeTh>Q9BvfX@b+&Xx(BaaBDP5(tZn-{tMFb0;F-d}q
z^=*KowNgT?4Wf~iS!|0yfU*^U>>6Awhx5wopq@sH(O5l87-1`BS1eRFqktLb$tI$4
z{mXe<IW!r9CTgt3gl#u6nY6E9MtWjvmyKTXA55psSkoxSVFp;!Jm71G^v4+96WV(D
zQap(l3T66%atNnkdhZA82VMmJG6|Vwkv<Zky^PMU|91ZQB^Rmcc6`qZs7bKNC%#@l
zN!JUv6=YP#NVZPhOJU;YXIw{u{>%UB@4;{S+x~X^gMaVeg|@%O8Oy%NL&)B<M);`w
zP2v6u6ki0fX3N>jM6$-1#ntyqrf`y5V7A^rleWO3&Z;TUymkOzrZQf7g|1o2M^HFh
z4`HoW)k&-z!(~17w3cDNcFO}o1LnMZr<_A=-<i5}M}s3(_gftb=c4id`+h!XtuJOQ
zW@JrA{A#aH*}OJkIhYceqqPk)IFx>h8B`i94aP$1ruO*}z(pn*t_N3r1!5Xa_2qlt
zVOb*_C-$u)%AvFk-fu7~M<aThHUI;tYIsYP5EwO4K#uFSAT9vDU9fJAy)?G<3VY;s
zY=s;pi(8q1Z6IxQWdMbGQ}&uebgEl3R|-!B$^?S={HgJR6?H6DgS~7TA~e&G2$-MW
z8)l-8H{#qOr=rF|Yn?NrRcBzLmU2SApsG;jQ8=i%qx%Ll@u=aa4wfg{euG!PLPa((
z;-z(8zF|vkVmEl-neE8Crca<%LA@h0(X!}t{M^=0g977d(;j!K_5P&U@%!QY$MShj
zXHe^+!&|BU83@>dLay%xOwt*4#w+p~S7TY%@cwlbq;+1!%j^NmS)3gCB1ORL<@<3)
zz3V#8+ef-zWKLeVApP*AY47^odjIeFy}u9t?Z5sv;eYwJ|E>7^PyHs0ys7qdhRfkn
z(M?DeC08a2<NhGz5^}(XV+K5r%e7YxlSAmo7&N3-QYgkVP(%!4y^zZ;6X!42*-~>w
zwAZVSVAucGIm}4pxYBGd#d@Vvxi$v>T!h*tMiHsqa+9vRF%IoP5`HQ@FWNH&p#hLi
z>p*m6*kUu{%&4iqE!(OE*Y1VQ3D1DdufsfrzHMlI=VDp|yu+evZ_~`J2+k<j6bGZf
z)*CQQk#;q=2*QEzbEl*>y*d}FG<M9%u{UejdPmzDVh)VdRH>Smo4I-(+QzxE6Q$@V
zn9gwyxxW;~inSyFi5}XoBLI1y#d)nR;5o<I_%)P<$A<>3pqMI_nzVe7iZrI;RX-yP
zK)i)Cn5i3DnZOFU45v}mVKdEXKBrU3sW~Uep>Kq*6g6I(f{GPHI^bltww;lA;w;qh
z!0~etEN-Kp&y-Z(3Cn6sQ!!EXpsjaGq(IS}D}w<R#W(>(5g62!TF$1jRkxbCeHjP(
zaj(m|p62oW_siVApA{I^o}dG*?C<g!mN+H0RXwqnAM(NntjqbRW4LAv*Jo5e_Od2$
zMvG^Wxi0QyCXy@0%=KATMaUOf1b_Q7=EsYyp|0zDe&6rM|L}kMcj7<u%YGUDo4?^N
zLt1BGVs%YW<4#_X0G^rUbO98V*)owcC^3JD%CFxUsi8gIDWL?_C|XxwDyNVT+laNW
zGTx4hQIC`6&a7UMeyw4Im_!%oWd#nlDD*F$mG(2>$e^evHIZ!fTFFDya-0F50aZ`y
z-ZAE2Pzl<MW6YCc??!|J26u{P)_T?XjtH1~WCHK+vyXp$I_)E+lK^}`gTHBf0BD^v
zbpfm~j_S;;DNSrZs@V#GF~pX^C1p)U|Lf%n<+Io~Tp+6m7|E9I`Y&}L=Rv37orUMN
z(Rsy0+FTFX6KkIH66gZXh}3GPb46jP{V^9Tv!-z+dkdnRbv6~MX&6{&qoWv;te%WS
z^F)M({=m`OridS>x+jF0B7tPlNOdzVQ!%$9Fq4@Fb%GkuCf^W_iAgNepe}88Tk_g!
zjmzPXo@&z&ra>fPGlY>@#)h;&wdW%fLmA+jOeF{jlNsBH0|sACX3Vi<!7H1iW9n*9
zx)|R|)h{~{isIN0bDF!4-}f)U82<A9`|}fi)z4U6)60nbqz7LQ`*|qOpVw6^8@<wq
z^Jfz;;EPYbpYt<$D<9|FPMD&MbFE<+qH<KH@SjY7+QnC0BI1jYE*=z+_?_SLuj6}u
z@4t!P@jJf<f7xI17vWcb*LRU;4+qYfR$z2&SlG4#nHrFzBag((MzibP1tfu^jM8V<
z=NOin=miE{&2tXG+PNyFJ=7YTnJTi@Bj!P?X5zW7`Ndk-o<Q9rqyLk@pgQJIu$PHc
zy@U+c6T(V;RMC0ql-bOidC9Q6dK36rb1kK>n)p?Moba?VkE>(*NnW3heYD?zb^}_6
zZ#&Q$4Uke4&J~uoEG*fbauTY^cx%`?9YDO-VOI|@BH;4?yoP6RID2m#6IrQ$J)Vfk
znuNsfS~FPR0N<z^)HXdZ6>}qK9NUSjM-Zw|tF$%<F}-5<dU8A-jVO0;#6Zu6G<EwH
z)lOaA(!X7s*3Bg$QPCOG>TT=Dw!@8sW^$SyIyudRbJ*&`ja%K|bEz&eQlQR}+hs&3
z)T6N)ZB*E<j7A%&uzbWhevlbl4fD}@#hC@ME8N!SK+lJdE%l7${4J4<PJIkM`xyfU
zIcami2~Sor+@yU(SdZUZuT_Z7#N+Y!D!ca*mVM$YKjEL>b0nYev)2fmzt+#a{G{u<
z&OaA{<Kr@iPq?1S4nC6cT+M0W+?UI|{wB|#PdSq^Iy`6Er_0FHKb2$PUmWAWzxnU}
zJNS40z`u{b^KbiG@SWfJo%mTl^JftPY%F@_<+!h+tl9OOY53OH*}3LrSJpkd{thY#
zSCGPBW(7E49n#YgIWZIKb1(wU6m`eoAf`qYXj4sa&Z#}j^{Sd`QP!*HU=?y>DtkBw
zZ$%k%7$9Lhp5${uV{N>uol&x)`W9G!q_YnBy^t6rmq&Ih6Xp!$TqF;7XM^rfy7$l5
zr*aD0_6ojj+$^ob+NOwKts#C!Iw3k{!3ZoY0DX5AE{zPosG%;RDXIwTg|{a=7IRbu
z%+?|Xbf0D_UMYHLoXm#41H#h{;zuj8UfM_~4-7`F0OuN8MH;>~<!b)OByDU?G&qZ#
zRRHfhgNz8oc!=es%`9VSlix>%6&<T@JJ2`gK?cBCcwe_yhbK;k1Z&RxdrL@?gB_HW
ztt6z;Zl|q=3dSgB1E$AbP8nytf&a07WUWiXYG{*+aR%yPq2!U(WzADbXK?nB;L>Li
zh`~u~);7{|qaiKCh~3cO2tVbs&*nI=NGDd=F$m#(#6G_Ndf%{L97Vo}?$5P-@pJDT
zte0hhyuAP76yCpt_pX!A<Re!5<#*L@-Cjs~FH>^;s;l!`&s;y0`GQq+0wFJN>iqNd
z22WO7oxAJjWCrGO;BWdL{~h>_Z~J-ppZ%?W3%>K)zm0L$r5W>DR|wUx*4bDQd%n1)
zlfA?sF-K6#tAsLRU{TSRMi#ASCRKcuF08xF82pU-{Tw_$#=$`hlB)Q8DIP1rH&Xy4
zXOIyr<{ATB7lg&f%i2PMO@U>DIAT|hN9;Z&DPnF0a*k8!cTSe0#k7RkQYBsic>w8k
zjfOd>zRa`s)nA{kd980~y(_bqCQltGXEwl&Q)6p)8R6V;JUC}Lg%OjkS}dcK1b3`i
z%&x2gN_#@B^{q7yBFCjkCOfK~$t8PWZCkY0l*SP|a%%)<!7!IYu{2q@u{fhwGoWV>
zG6Rq~k%LEsT0s0x?>h_RPD*VJ3`v{?s+fUn>N3vBdPH&}U29zpDhY%GbHYkuS`Hz>
zj3{A9(KD;~Iwv!Pl8s13G1Qh(tt81p2+%mM(R~0i)Lh1;SdgGs>Kg@1(0LGo<;LvR
z*K7=|rZ%i5w1ovH*P*#ah${0S>TAn+r5jX7BcV@l)Lhf=WIzrcs42Sh8P$Kx2405p
zJ&THp9<Q%Uw5`Z_IgR%$@rfeXX@CtluO{F7%$MK2Fr4)Rc3$iImxIfDKJuzUup;?_
z_^im#xvoE7eAatccv;5leXa;aXOaz8*AF-Ws{imm{)6}f|G^)?cl?~6Ngc5~4cBrr
ztrfY5=iQ2CnsQz>N$^_73Zx*(vj7dmxS!CgqGT8YHO2TyoODgDcPt&ZMOZ=EuO|DY
zrqZ>hFKax-s4S{-1(}7Axiu|WASX^G1`%tHG2oJt;qhv^hrwVdRd<ZB>~5Abgl1fW
zpMM?v6H%FGV!V3?066Yn7pKtBp5C1F-k1m-+KwhXjM25X8_bxpJ;sA5VRM*GBnIFe
zG03;?eM1=5@EnHq9dI?c<f|I9rjey}7;o4?h~hZzTv#($-_f=k4OUE`6%;eT5q29U
zY$>+Z;`yA)?&gZT-Aw5kX|vTqJ;gvsGf6Wz(vvn*b(CdL%OOOdZIE+JI5sLFwhge(
z=y(oMH67$$HF2a>JD3A8?#LOe=eVQ!2H!W;$eOyv2{&a77!uYKImP+6q%)`ptD-X)
za-(C4Iq0Ai9lADqV@s-@k1ght&pRwRdekyF-`cCWQN}=ry;(yv=7XGI$7f0;gT=YF
z8W01mi>4?iL+8VLr8=?UbD#YjzU7<089(?(f9R`k<exN){<ut{e&pkRu&$mF-+EU?
zx<4u8^hwwENf-E@qbei)(GR%%j%&^i-uP(!;k~b77ExKnN57W8+7jbi1zEc+l&=4O
z@^}CJ*jwV~f9Eg2w}0DrKvtVh;KD>g2Ii3%QZir$O8RN`4X&n6mU4KGV^*!M0*;Ot
z2bwYCtp?RN?!54RWA@-1yfs8T1mV`_YJKJaykWYlcas-Hr2|5tUd%sBLaV7DDoe^A
z0Gz7JR?w>)R-8@`5CV`SR0JqoS(II~z!3*9hXN-BjF*~0!O1DuZ_FB4mdt)0cYbF~
zJf7cv!oGU>`jnJZ4V!m)NOO8yh9<l*wVpRjAP3TdV<&GrQZz$Y19BnS|9{&4JXX4O
z&+EfJ&$Onh+Izp_Ip^MPZsVQ|1c5;$n2;zzh{%zx1d<RNVFV~(5^T94;HcXW85Gi(
zShf&>1Z8kyMNUHOKqO485S(CQ9H1el&1C4N@9jIBbKY_9n$}v+m_L5cs;a%;_uPBV
zA9r2d_w1@#wWeC@H-Eq1>aoSE7F8osC6i9d8i`1vq-A>H9x7UhqYC64(WaICl>)|L
zT_X{Aij*7>00+{&w;F94N$J#LjB0JXn3G;`PFc!W3l^hhK{dCjML|U_GH$AksSk&M
zX>>(?mz0o6V-eQ599NXGlaZ2_Qj*T5unbF3v^_Dh0y1S4G{hjnBt{~dMIYPQMxvpd
z#L?i$DL`#<>x#@2+=!C-@j7d5n+EF|83`mvmQiZMSTEqUC5&8n@@sR=Imi~Jsw88z
z74`x_o9exyvKGIh>%Fw}IgAn^qN@3ubEK$1(uH!uA)F_amWW7e#Si_!5Aa|7<^MSU
z&Y$@+r+Zayl#BcK^13gdEseql#`3gUjsDJmKiFvPYp_JO_qMT=JQk<1WTo$q@@0Z)
zY4X->o{sBb=A!PO+k<VM>fX!om*pHEUVzK9tyw@_?X${Xvg%UXKc7=sGQApN<ed1?
zAOEBL@b~{*e*DLNH$VJ?KTI)}m=r04Hj<U+Tv)V2#6Bwua=+zzU5wU{ll0BXG8myG
z5IkkOrIt74__PxDQjjrWsRa01+Lj<?aY~CUFeM3)Qz>CUs2O@nizt-P(^$Pk=whUe
zcAy-K+%DzJI-Dz;D<xiFB@|0TykKjfWuUVD3O!GPyo|us5>N@0Y$?mMT@q>$DMJae
zT&nSc8B$6Sb5%a%yMR;3b-b{s1<_Y+DFEh7*L?cla&fds#c3;i9`%OTTFcZ-0c+(Z
zP{lz3?Ho!QnRqe6s9I}00$9oA3<lAlj3Rpx0YWP>N|4XSV3LL$$x#?cV=g!&1W{!%
zbk#{wq%oE&i{dF&TQZok%&};+5+rj}rq&CMgbbD6C)IOYdL}uE4CvBeU+6KEu56%)
z!^suYCcQF6QbsTk+MsiysEQMiD@@FwV?j5yIlDB`mu#WbRJ+v3N=a2mTP@Vb+Et2f
zCwj&PD3Kh4I5J~MwWp%0)T7+iOC7b4s*w;j(_9hiocCXln=*;ml8kEA>QXKH2)rZJ
zQPvDmBt2(8fwj1%L)(^RR3k5$j?ARccWvd5x|pwj?^4eHxiIASP3GAVEN<`|z%RAr
zCE{DgQMtE;)9XryaN49g2iBL%a=)E3x5WpmbK{lE|JJ`B_`1`Ja<Nab@(<4A%uWED
z`hn*R<kDH3K6v*q%Cdd6Z1vvB{bxV@GyKjU{e%2lfAoLKhu`yk6jkvpIiLGx)Jdq?
zr^v{?T<UM6PyMHbN`INmtA|ozC4Q_W>!pNZtvpXL65|XMDI;5ba(tzT(U&pVgf5HJ
zr;z03=w(_?3pOWdL$VgF?5P?OYuQFJXOS1kC0_gng_7oK4^0(fS9J{Bk(E&Uvy_|2
z<YH^ky*ST+&T4sdP$+VVXAIOu#xJ)^?NL-gFnt2Ief#B$A)1xYEhw$X6(}P3u9J(W
zlyd6W+d3~FQ<N#7#$cNEd`lOwwT}>DqAXxG>rPpO(TlTD)y*hGV>wAgtBp`o6oqaY
zlojpfoHNSgTJtZctjOAmXU(Z*Bl&cX$&I*_SA={{CUQ(I3@nW=b*dLG1Y2?wsoG!3
zki=N$6fw#{&cv!%C;g_{#Hv|ON{L(oMrm>q=arRAO;Gy~R*7z|TKHF_V2`caN;Tq2
zNwi;0r4SJNXoX~~9=}tj_7svd;Sx|3t+7~VD>h(xAXANUPPZY_@U7Q5W?*5z6l+(I
zOfpx#r$U`5p~!R>g3U;=)=f|L^*oBGjX^gW9RuFaloTtCim4BxH2X`AkTN=Je#5W-
z_57v(_&?^WU;EnmroC5+IUB1}-TEREc*mSCSp=M-;P*TfwF!8!fs>WJ+|I+e`#sw`
z)$1>g;rxhy_VVz+Pv3m-Vfl<zm-SEC1T3Xd;2x6sypj9Izc|i^S=93K;}<Xa&A;pS
z@o)TrALWOB=pRLEOAHJBvpzY4lh{z@R9bpbAtqAvC}pvxHej=8gQS!wah6i0Qn;qA
zlXs=whY247^Ei@A7Mzn28h8<K5^gfNUsJt@r7ToZluSEk3RZ?7if}F&U21he$~L8d
zX^Mb6l_e=D+fb^$bxf2J(K-WCrqx9xXN%IPl0YZHUgZi(tpVloJW+doi(Yn4NkE7?
zo`^F=oJyXPQ~)c_veDlK@Y>>aRar>t6el6?rNodWZNV5RZ4`q}%M81zHDvjoL?+To
zin*q^#-7JAV?o)j-g3(nPpLmt5+I0@@h=5rT|H&W#1Mguam^{k(^kmsa+I_yr6^E%
zEiOTje-L9qnI&>5XExg`|18E%=gQNrj6~sD9MB?~uC*eED8ls%H<QxDl1_b?(<78K
zq0AB;=Ts+<q@u4?h8#qAr#6SRTr70sFxrx`;2?5MlO;1aM^mRdraEUoA&JIy%!xP$
zq49RLSCDI@mMU{>&NYQyQtp}&s%oega<g)F2$C7q>`O_*?iBeh1~(@%<+R;WvW)fC
z@nKyG#zYyp%?TBQfYnqc6J>NcB^-^=bf05v-W#>oYjJ82KD3mI)4ARM)`eL=l}_$|
zdf^qSJ-jrWXR}!k;<T;%i#n}+|H#kNFW-5&IQFHC^jT{z0saNM=flIgJi`a+0ZWto
z)ML1p(Om8TzWK?g{NMllpXZ0Z_XB9wLMdP3#oCt+=Ta0ak#c~_aYT{v_Y}O8c5=d0
z<El(kqiyLudGDn<5USRAHKwQWYtBL8Ngt(1U?-D25t9Y`q--ONU=+$2dG2*(Jl3T$
z9BWWkDE5^C9U)8kyJ#zaqaiQ6WeKG{TBTYwNs_&%fOWm}Jo8cl7iY4duTKF@g#yGR
zpJ`JBA5wqIoFk{Q&gWRZSWiKVOd};>jPwUi;x-i>ER&}qQ(23tb;MMD=$z#y670e<
zJy2@7L!$GP0@HLPd5^U%P1j>xgVsv4l|!T;WhJ8=8dJxXgG|(vOiG?mq`V<C8kSPi
zN~MzM76kb(N)VDVxdRI-w;+9mQpx>JhHIHrsMH>%Btj`frLc?WHnm5xWG#9@`eqTg
zS2bm(GBwscE={k~G?O+^f*9t7{|RZbiME5>*jg_#;f0(7G1iCsRQ@PSoKMbyC3Tfw
zAW;egKT?3E5xj&^y7rF>OYL@v{#2dp5N;|1bt<jIM2=C)gu24~%NgOlfbvwm=1a6A
zIu_M|=w*VU)|s_MCr?S4qBA0I7b($8Z-tCCj)ua&^t*n9-}xiI`#kr(luYWKc3JW{
znY)KZ^PWGs?;O<P5H9O3sMC3IvG&O^Jlyuhnh&Gva~t5ZWBSXD-QP~RI8e@!`b8$Q
z+zpqu``zQL_v5`oxNwv8%c_3)<nC&J<#c`ST_aik)Svqc{O7*^EBvBg^ozN9@}UG>
zDV@fYm+M=KLFbHd4l31&Q3~5g*(an(h@lc>DzYzVYq5<x4S9G!OMFxcjk9!J%c&H1
z(T9jJpMq4uAxL0ZQWV?)s!kOeP%?gAw8OYo%9#Mw?kt)P6ooPgl?^0KDa$~|1rDTU
zK`gX7^|)L~D3yufNLJ-EfwVXS(VZyO*jMy?mO6?yi`@!T+$kZBK*9UN=eVc7s8dJ@
z;~G+m_;8@4fYlD$G^i9wN>}VaCVGW6GO;8RBdT6(FNPvZiRpZER{n30Io>csnDKsu
zcqF%-aQ4*Zu>w?+$r`k3WFo5eCzQ5WD<hgYM#(W^CL3Gp_|nu#L{gfioTL=lIGk(h
zO;wf0DB|5!Ay?pv__Z@S)#M@^P(Vzfa;g+kaZ=u>5eHf@35qD1OiIf=22c_uWMcBf
zFk`Hvb%IXLNTW$rxO>qR;vOSusmG_tso_r{r*cAK>I&?kER-QeCdstM7L79wft3_=
zoaFelr4&(<Qwof6l|Ckmq^hwP+o4@URyFHUkei4OGdb4TYAvIH6eTs)I6<6iW>sX`
zO4S=&Oa4U8ki2AFBp_j&lLt3X6t8P`A?FuM20h_)%Lk9I>&<#LmzNYJmv!j(*MsUO
zYWd!KqTZ8Xm#j>k`ZjgH;@Rc%r6f~eQTw{s;JsrxZ~ip@!)4j(e$;<<C*^Wmr*l2;
z;&dGM_N(rBrjyCPY{p+i`R`jUuTHtQgL3{*-9O*Mapd>>tACuI^QDjYBR}>d{NNA#
z;OUYr<=T=9+R0phEU{J}q<54n)<Bhlu`P=%rCv!T!D<cBi!xbBln{iT$2Gztij0w$
zI<qf4Z8cWN1zRd+-vG)`l&~U1wCF;pBzPvPSr}mj5@FD#((o54MUw@S5a*dRPm~l`
zl;BPRV$=-U7&Hr*R-OZ=3@Han2r}Z2mUt-7QlVKUuvDGk%VJ+Cp-x6A&>EMek*?!+
zgAgw#j4BwTm}6uPGtmh8*xH7s>#(&sl(R(hMM2S;QylAOl9GD+iBV2#FpYdyAq}6D
zP(_HGq8Ct%wu_jxsgZXk=RmIGx48)Elu?omFVjD&I*XhlC8_FprCHdzu140<ypHL9
z<`9!S@WvQy)3S`wom)DP2((RquVj+TS;W4?K#CEi6WUrbjs+M)2&g2+t$0sNo)l-u
zjTm2-q?DEG!=W^)h(T8@J)B4kK^ogiqfBKfRm+VhM+rjVu|y?Gm3pldaRC99^JuV4
zT?j(cx(4kmnzsI~6}4FuRxK3h)N)tFB9y9wPz^>x#^iEN<lxCUqIAX@i`Ayif{XDu
z*Q4vX)UtxAy&j<!ET;!8r;H`jmg&B!uSW864DkM<E-l;nasdy`VL5FFDuWrEc(KZb
z^SQ_<9&X^_hVS{E!da%XIwf^pXW8DlyTvIoFBj)}-`QO5>e7Xj^P$uUvD0Ck8pfBo
z$i)FKYB`sCIvry@j5@sW?)PQYuYL0qe&4_L$N8WA{@=@&zx1Wk(1*}p%XQXpnyQh@
zE~OVmCDBA^>p`7@uUt}X4$HJ++bkLm`P84tQSKp@=h|qg7?F{wr=`27PYI<ekr<=&
z9#9lFp}j@Bmabc2U0VYnp}p5xp3CG%=EIIK9mq)>g0+@KURJb8ipcotD&$jdu~g*V
zCTF?NvjnCTdoFd=exzJ?BkPyiJ6L*rqW@Vbb-wfRg`7h46VXA8p6ETK$T%Lb#!{LF
z6-TsjQddh+N)=u@?Ghm?qW4H?MWP<NFdI@N)hY#>JqutBmI;YW_8yfY#aeXJp^GI)
z&m5*ob+LF(xv6ZGc1#&kE@u>=uGvdo$}LnLAVrFqQpC%Yl1VYtyWL_-C!||SOj53f
zTF%9QS)}y}%IZ@@64Q+LGuB$VI%Orlr!?11jW$v|lA<TaNX(HrN30nsW+53=LK&{@
zSr4K<I70aiXqg~VRpX%~nIclfqbpC4F_o7o_|wBJ)YOvh=M-hKDGE2!(mOJx!Jg9N
zC_uxcuyrD%iUyKGpaf6MUXHNpX6T%WDG_5N_>mAi)@rhKbWMMvG?9|7^sW|Td#VPV
z>Kdw+ty)U<lIEAh<CWuo^gDhtKmNb?H!loUsae9gWF-FP{iU6E-?%XNEcWlR4s_n;
zJyWUPZ4ocV`!d45KPp<cRcGxVRCV6FESbRBk+N9ki|u~ack91P8K`Co)i2%mS(o3s
zWcyR^dAt5jr_1xswJ7|9pZQsS(~tbi{Gwm*^Y~r=+;8E-4?n^!6!1AqAXMajk|{V#
ztfS>J`YFyJmx8XniCAihveXD%QPI58kBC`HrPg*hW6>oEgDGH&A|+2Lv5u9N3dmPd
z8bT@p({W@s9>_{DtRHc8{fxHXh+{LA4IpMxm`U@2JReBoPO1;ep&EyBuKw-`qg@?X
z;Y6b$rC14$OYx-OCG*z`O)t)*6r%UlMgeuQb5TxQJesM~N8c6d?Bi~WDV`J~!Fx&u
z<qTn-$n!{cj;3kR#u1`$(WpA<van+YpHK>%v+^TX_Rm!5Ff<Blgj+wwcrw;YlqLjI
z7K6;>mm)CkoYM&uMokg5YAc6Gt%a950$MMZHCEhF0jC$f5heA?5LjrwwUVSJr-0SA
zM*HG4iV`Z{+0STW==y<DplMv?I1){WoN9Ehsy+cS_7+oO(S{Y;25N&%&Nt;qic<0r
z(fcH$y+O92w2Z4dr6`pJH7{s$Z74b7i$ax%O3V1B48Nuvh)X@w*g9)1eW6m)1!-5n
z9}}1nXPc_8y^PL6wiczV7-;Qe04pU*@Q8ke7T}`T2^50f7KeW}TFZ4=W|jmQDH+>R
z^pOavj<PB(lN&~(bi^pdfad9=Yq|GMk3~8C_3pAnzW3j{AJ$8Ta(Z8#WA=y1$9>k)
zC3wFq`@EYjERE08cb3nUy2$n}_HkNg$xJRBN;z-)y^Y^@PXFw{E>GsNa(21O@_M0u
zcz1iBwV!<9gwNmG?soj8|L<SrFa4FD;J5v={|vwFH~e?%eNxc1*{he5ixDto$Wb!y
zBH&Gt0j(Qc(8)@IzAW-TETT|S_c!7Uk|5!8id8EF2!RwmsZ`LXGIDS01%o7!xXyu+
z5^)E!0{nz-8nhFhCsHDZ85KN^SchN=r36ZdloHU+VNHX|q0-GLQc%Pw$7PgUhmwQP
z$WYK&>Dg6BD_ln!s}yKgG1=<Gs*tmsXDNkLz6(wv=fFI};c!GL%jWtKZrCvI?}+oB
z5)%=Ru|=Tm)MjN7gDrEw3i2|850d3c*FAVoNvTHos*Z7ru`tGBt;0E6=_{(|t3R4R
z!Q}xHVQ^FD>Sd%eXAzn?i?ditbv!9cqbB7f&cp~F00l-Z%`<5{6_Z$+wvu{FgGXka
zB~sIsgfC@jgjv_rlFgE~3f3a0P;V|t>#S+OsS~<8S2~cQ5;^&kV^w+q`iQ|T#z`$!
zX;exSn<!Z*j7v_K98VKb3-yIe0!edhnY74CVQfV}C<9JXc@>4SH>CtbOj!`u@^I&r
zDP=-68QV$&S~*81qT)g(R#+rD(F*GvIY*hjM#^oZQb+y@P`RlV!nU#m{PlWkYALMp
z^{YDdh6Ej2)lup9F11*4*^E9{e%_lfJgnuP@`Y*~&r>^}Wq_Bcc`^JK&ZV3Uc)9oe
zdhUx0xG2$_pm0lB?A~rKvyOMX-De%%Ig2@EE)Qb+i&{~sC8Xu}-&<@ut$%h>r&sFY
zu&YD*Q-Au;^4ov?fAdu8iKW!B(*(v5V>Na$RP?e!V@NgnSIR(<$+wzw3RZyH6Vt*F
zAbpE7t8+~~mV!|=-ctUR%tv~1xjOF7HME;6xxZ3|a3qz999p#QgdU?LNHQUHDwQvC
z07{{f+{Y<L0i!8VsYFY(FN$@c@+3*WP)cQ1dKFcNA|)AOQP$N!cwv4v;t>1{_&Ox=
zZh-N=<qIO33)VEWy+Ub6(+!k%gVPr0dvcnw#?p2j5mSw$fgJJEQNSZb5Q&Riqtvcz
ziKA*Q6r^t<bka?A3PMS3tZX;e3_%;I(G{8+&1H<Wv`sDbPz04+#IWZYMJ(kKq3cj&
zxf^7vY5^;Y@2<wCNOL>|RFHABl4U%!YaAKJ>;o!g>G=DJ<R{5uVvv1NnN7n?m}kgd
zXg+E~sGvo!!dR(^Qn1!R3`DP_SzVh{ISDgts-CqdWK^z#Rg5vG$%e|3AtQMy6UowK
zu9;G*v)DRCOp}E+qc-IgPNp58(p1z2m!)zNiQ$fMq69;!$FH=E6iR=hREErSs!M3S
zWJ6MemCUR<-%|!;P`U_VplvEIP6aL&>bey-QJHednP2&j{XG7uU-c{bGyl*3@MNDZ
zZsM~sT@3s3-eSD&l{yxqFRxApbD=k{Mp<!T&@ZCycYVZVrm$4<%Q>rfr@8yCAun8l
z<*4`gxq5%v#Hly%F4p0^n{z(3d>|#QcXs>kv6Q;yvuC(WrcR4oj?02sb!-Z`pO7iM
z?QX~K{sVuQ-~TWC79QQ)kYgYQp=mHK*B-c(HA{>nB@6?kOwtx>tFx+DhoVX?e~QDf
zs)|uX!2%d94eOk<*c;oB($Y;Z=X<-BJhgRL>niDGuD!{~97l2q=&Z4-kus2?Rul?R
zD@xNK!d&t6yycuwwm}K3U&^m}QDdv_sZ!OTOvUDiP0BH!el?>M7%M%yBF;lwhfa}X
zqz|l8#=6gE?!S;z==v+x>&Nue@h7V&CW3X8<_apCpVEfLI?$r$YqZP{n=nBR7~9A*
zl~Oe{38h+z6mydKevC{kRWKWHLL|0YLzGc1trI0CG?}bKGk_EbQ?9@iK`iGG>J(H#
zr$}b3wAAGkscY@<lNfMgWSkTP>_UN9JSh}8?uzV<If4-qE0r_TVN26ELXPC=K+KSp
zJv}8x*Qngj7_u;rA`p31A~~zm11@Q61Fk_~WVB2#B02_TC7l*yscNbz%{EoUISIde
z65K(qyi2)IjKL^NR%h^gDTU-mFrJ)*0XHoO;|g3nfz)b>YK@}+nL!%U;~o_z7OtJL
zd|wK2l1ZURajty>soQE*r;(O*vy|Qp3S}y#358fR1<z^3wPp%X1%9-U^I3bT*<6<g
z_K(vV4@x`b9rJhU=|D~Cxo`>;$<pqB<6;kImvd>#Pka&=|1LHC`_5x|{oufAsJ)0i
zE$hFd$?(qUUL1*dvwO#(PRI7}i=VN<WnFc4b&C^NPUl>Dx}OOxOvqGF*kAgqf1Q8-
zFaJXR+rRq13dnrEDTWd}frKd%uvueBB@@%4O^_-dmgq%O#8?{RgdfZ{6vT$)9EeGy
zNjRfVqoP7GCeOE48B<5_#Yj^T%@Wbe7-%AsF?JBd(Pa+<VP!C#m8wIE<g7_r0?}+l
zgT;r4%|iJi)JQceCSpp0B$WhHsW*xeplD1ormicC!W={Ow3;SGXj<Wj$`PY1u2Gog
zJJHL_@`ap2D)NgsV+Bl-6Z7ttcD14H*H8*^I+A8Fc)0*6iYzAyr<VIi(xa@X&5xz7
ztp(s=3|cj(3acQD3#LSpKjOn&QN<cvOEe0wEHyN_S(Mf%jc1iby)4B<%vIX*bfZ@F
zGcD5(g>oukrF<5mFO>ydDO=eIYuQ7I^UEa@W&z8XI<gj1qUa*$RthDHY@W6bt;$KP
zC>7D2AOuR0i~y}=#I^0Do?3(-Nv1E(!XI)i^@%dzWK)#TJ*bMaPz8+D#7Z-*Os(xF
zeX47hB6Hh09V^mUFeUOpEa>uDn=Pt3bcNLl9}_-|j3sgIb0xH~QZ7o0g@K~fkueQ5
zZI<X-X~&WSlsX`QDo`^?eNpEr^}z&KMAjF<^8fBX^=tTxfAz2P;?3=HMe1^CnlJ1A
zm&=1F|6X~dW=&kax70#TH`K+xIPc?Lnd(e2aas4jHwE;W%h?5-{6x85x_WrV54_3!
zOrY-n<XkSEjr&!pQ@V6-d#9uMc7EeM-($&OYR0&1```Vuf1Y3YOaBXe@ca>|7F~^F
zUiiG~LFuafy!5Qdi>Aog87;KZr_4}pDk+<i$w9WOwZSR@o0cdOl1W>#GD4SAC+eh5
zFXOKA5Y-93qDZL_vJ7t+D*|L77=^ruOzBGJPZ5fs3IsM<`eP|Cj1ckE6QE`S+Bizl
z#8l}ztga}3DJf5L5GPR<(#?veY6j7G{>;+m{pE`}1ri~8nMp4)4(bp?maH&{igC}(
z<CYkcM3RlI2C=eenuJk?7(FE?tWm^R(@N{mx(Cxro!`%-Fkw}K<e82;iF}O_?jdhZ
zH<f0I{AJ1|XDO3Nsfv0GwPc`c$*A6@%XDl}iWoe`7+lj*l&KCzVU*BZYo*DJs}hs4
z#HhhooO8IQslYG=1gvXtZ9`51hEqAp6t1c~ODIyP?5L{h87LXeJ^B$7J&m)cne3Dc
z#E2ot4Wng>Ldr)ni@k_-D<w{&)LNse4^t_HqO6pkw3SqvB9lSE3nT0)D_E}G*>Myg
zYlBXaQnz1-C=#mnNpf1~ehnoym_}yBb-7>6$P`g%qbl}+j0YE?4?$(JpBE~ZI(oe1
z9&#+P(xa=2F8#`f&-wX3`2D>2D}ViDAx;IGOKbmaf@)bV4DCfj_lz)>9?`=r;p89c
zKdSEQU5f6LX}s9gl383t+xKlLmj`}d4Ghkspch8?RI_>a3@>w~Q>*);fpM=FAQ$3&
zb+HG@&E<=_t#WpR=WS}vCok9NYhV8c|J;xMkNKbf;r|I=dis={XLJ!)9G5A=k|}jG
z-MW^H!pDgeB-6K6X7@v=a+tMDSz;u4Pp<sEsT9Zwt27}?nJC66gmVsD-{QIzS_w|3
zq;TqKl;l~kBPG|NkyQJLZ6oCzAjK%Shyn!xgtCBsi*U4|b!86_V=n6JQKATCF{Z(o
zR?6Smf~|TBl-ieo)-|MbChsdL5W<X8kb=S=_wW8@?^(W>9A``fKa#u{;gXX8s**e=
z);MzVq*?0eIVOq=_;en_bhf1wC0r!)M2P|GDkdN%jBBw;Q*0dx3p0K?5LLof-k+Fc
zKHXMcsS~;{N$JU$+7!rz7(~ZmsT^A^tfIncrp|z}g*QYShQ1esu$A#o8Q>J~m5h;!
zF{gTVTW@bu8$M0bNTW46P!ni_a}CKU=E)O-7Z8eqjQoi6>`z+5GCOS~Wuy?|Ovn*u
zmuV`AR&t8<=1;^dzk#+;xFE+8z6?xJ#9IGe3|d;sPW5?|K8q2}jA|If8Oo%PU_KD^
zDRmO8ZbP{7^VvYwbUWuvjGmM;4FzRnZd@;!fjT#7WkNzA`FY`cF`PK*DTy3pjJF`a
zTPiXjAkA*+7eyuK^rQcS-$Klp|LCv%|1N_0vy)ql-hK0So+!F(>@M3rp>+XjTk^E=
zW&KMdeqrzy9pZDGug@#%isa>Cworo@-tWII4LWtrvP=ejM~8QDh?jqyrm^mCed$e<
zdU4b}&v*I9r6aBHd`6Heq?u>jfMkD}7cXD(V}Ixm^ZWnBe}SL(eP5E&U=jkjrDPgY
z1Z(8B)4FOyi<6Eqo}PJQ%<1~&g?_$<Y*M=yWwJbbQpuHQQB^&6i?uD<bO;4Soji<n
zVkk$+!ZFxNU|Dh`Peqt2RUO1gQL=HZF-;@aF3ZS!$PrbE9!td}ER#~Est@X=?383+
z2F1m6VEv`E5Q|DhUMeOz7t**FM&mGkuKVE&Ifec9j^4_zn^VMD#~dQ*@QTa}nx+wS
zv(kcNC<(0`N+p4b8%I%6e-9}WQX={p?PY{hp@i}|PUxy^=BEQ8Nc}lGi?R;aI)S{K
zMw&TU#;tM?dRb*_jHM$f1X9jo#EaC@7o%jITyS(#RgR=qX|$H|kCF)+V;ewGQb&w|
zT+&Iz!<0gpj`bOm)YOzhgBH-9(c=6ln5IV=Ln^5n97UQVDGIf1Hny^H*IH&RTNpAW
z7JN(u1%*`0H?U~7#wb(~uI_hnaEp~Hi~hrc5L-rckpGs7%BIv2U@Hw~InRX>T}mf1
zWF_}Os*J8K*D+P(n3Z-?0m}(h0uqo3wTW6#&I*Z`sO6#}eD<m`KBipISy)w-aZn;L
z$+n~}Tk#%sdPLRwbjd<gcHMXUs$c%g_$z<oZ=U2o;r-G2dHF2!3rh{3il|urmL;Eq
z@=K3Uqx#DnqyBL}Ybm^2#=qR}WyF6v&I_M&+UCPkSv>w{ZJT#LJEaYm(6e%X_q=-@
z?`0AXyzMEX{>)4N;Da3bsekwn`LQ4WBmA#^{KvT24CGu{jq5IqF`}+$9M(2irHCan
zANPbP8G>sXw5inGxf*;B&LUGwz*)G|Ad~Cpt1|L=$))0aG>}m-Kn$ehrFW+lC08&g
zO3K$Ul9UX1C|fHTTFSlx(yJrZHAoBIL#EQ+*T=Gm<bXCZN-OuZC#RV(?W&-bl(7XP
zWhhy=<xuLRE8~pAszQpM`MV*qGwlzgb&oO@?|oIzk}`+RGdX)EZz<a1tfLfBR?9V`
zutw6HIV3_1(nwVe^E?Wxs<Dt0(T~+bDx6NZR>a(`YtgQ&V6U2XA-xGhk*OD`?47m4
zU!~}*72LpL51lpDI4O#8<Sew?p(tX^I3rpON+C^KWrXtAlv7V(A@{S!Fo!@vS2DlE
zJWm*F$T>5PNBnL_JFIbShi*EI0;?1$cyvx!1-k8UuEE-d7=#^Ep>S0*L5y=k)p1`{
zRP9L^M7*hLzbb@kqZ(aQ&0<g_E#z4zlcEHpfG(6do#+8%=95ZSSe33+72Q!4Wo*oe
zrjV4$sR+Sfk><53qS2siBa!6^{VUD{B|$(g0xHT;`cB$NKd8h=$b}FT#(H!SYy}vp
zr>Z(3rF5NAx^K7?g<tb4ei<)bzT)5fcm8k8;Q^xaWHgthmNZUNJKttGS>q*J5To(#
zaxMjlqSu0&IFB@6hS9QFiFV)J%ESFXc=0~NfA1V08u9bo)p<u3$8*nNd}gSB`a|NJ
z&!|zov;X%v-&3VYA!FN_Z+!eQf9*f{Nq)tD{+D1)i$O_{prCOLrfn$73KMgk67nGs
zLy%{_)`cxdrb*Qmu};cPrH}%3Oum9ZE2me^k`YXY9ZKy_CDxLn7+Rx=acM%!c^3)S
zWqQvxSla@b5SM_}2~)JvweGLw5w)iF^E_dmAV*s3gtfTR0;!;wrX$<8-y+6<u`S)O
zrtQ|4(g>_a6_l~Kw!<~jTu<|aG7YZ%F5nc5hS|@k#*mdIRITAmq3Ft~W1Mi3SVz+j
zC}Y^~@5u83v?r&`=mk(U4C@M*GB}fPO<TQorK^nv=UPnD(KG|jwSrsFatoTakslpN
z^=vyx7LXFbPh@{Yp~=Bhl*P7ArfE`;9;}wp*2Xxw8#9C`O`*w87{pO2t#NM2bS%oY
zM5WOwVN@1o#wZBvoILY>mI*A^k;%BW<uKimk9!)|faxhZqn*RGE8=`4r3u^gXzkG2
z(li={Xb)S})xb;Dlu(+xN*UpEvIeCZbnue-YvDLb3q`oKr0TG8sTvBE-aIRfYrERp
zsM9|wO5?iLVYL>~kAfy@Qys4uVN<7}a>|70Wz<%{C2}vAt~SWhsn4Jkl&XWR0G%Vc
z2HG;IWpJ)T;mLl&PY0AzG`0D!j8K~wkTKR;dJ!`!y&aT7XHYrt>;LP&nt%7-{|mf&
zdwXHHE)3Cw7w^>2>r#XMI?<&rwzHVmv!PpjM!EQ+;+(N9k>%yq&!wrQ4DxPt{;s|K
zOi%FcDi0iBd2a@xE*!(+6&5FPk+Iy{$3+LDoKNA~9^wNBe3|)31|^_VRWkp}fBR4K
z%YMl(##F}QoH7}iHZq|~)7Ic3r-F4Ir9j&{%&-y}VdKa__=grFekz-yBx6;eOVwT|
znG{O}cgpmV@NSmVaEZ1Zj=YeU$@G(JB~Vfp(+pVGkW#>8z!a1fu&GuSkW4413V@yV
zOvgRD{f?3|oArvm?`t(f`i=86BO2@u+jQ7=MKK-ObQGls6^~`L1ZF7}G%|sC(wqMd
z%NKGA{c1(CevInY#NefI5sq?`YlF5e#&wd;wJW+|gKJxy8`#~x<*>iQ>V)kFjBV(L
zHCln68mNqjIwit%Bt=836<yQg+8*21W|H@e(@5|WZQo-Xd0$vez1+7sle4ec2+|kO
z7UL{AX=0YrmC6ZiG%Ci^7^qNQJJE@Vvy`7wB3e(=4w$Curlv@UGcl*iDk?ldDHTL)
zr!WOiKXkNxM?rcfIRtcuR1zV{xT!V<tsAs1kb*cZp;0|G$Q2501)~s?swEVA5)wkm
z1Xsy{k^-?vtv(kjPNCM{Qy|5FszA1)&sYNK)lgYy>%*Fh(0V9$9=p_XjboT8N=akA
zf(W$^LKTuTN+)6n6}_zFy3`D2(GZpzE`nTkpj?A-jWo}VBE*?v6h~Fsj;3t|Rjss4
zU8sd+cA3#e?!~4@D@8aSQAP24e#dX+U-{SnIFt8hlT_}<$?BcIs!_hIpHr$M(&zQm
z0_004J9?gY;Z!P;EsHa`SD#<@skn&J@0VXr5%gy)3q{H2Ya@J{wW^`KuiSPz^9T0*
zoS`Vr>k5uyIsQ@|>HClUo$IgEN!(9BjfIzQ-tuq#sXxtc{<Z%EZMP;RLHK5uaaq18
z_D*A+!?rE`YK5lMduf?wD<?oGrO0!j<@qVKZWF^K4m%}M$&z&&k-;n};uqEgW3k%G
zSa+rK7J(g^9$GF<jDZk@1wv~o51YnXnWW1}FcL9{Jqgj{!vshVaL(Y2Xqu;Luln^9
zVS}{|u|UcOtSGiwW6)&6G?MjP%aU?1`rC~vUo2lNQ!t8ty`jB+Mht=56p7Qe2H6_4
z!!`qG0d36DQ<TBA9jluglGd2U(XX#ab#kD@fNg;kh%w3qu(hc~g(sE3em@dRV!OTL
z<)@zlg=Zgpz@tY`QQ8r*WHF^wzLZ=gF|DKRTHzEiBJ@*i!|Vfvf^`m+MfVkQO&c?y
zg~v&0N}9<bkn}7@t%B@Q@TBBXbqYZQ#WmU)QdWQn2w7v;^lMUz5=p0#7(8>ZWH*R$
zBVdzcJ+l{z4eLb7F6MyI_G~1wz-w}?N$QG}Oi7T-0@9P!r5T*C+DY1+V{MwA3CBp0
zP@7z;P9`a;z>!rF)u_rlv!u#KVM}ExtsFhZwpc4uWx-2Um@*hA&2u;xSW*;;RMEl)
zkVwe54r47b33;G(4Q;oo+&2bgEofJ#j7*IZ6`|D3CIca$mEojxt>Kscf*<C$|1-ai
zKlZ=>KV3xc=jDMLkbjrb_)=S6jIO#wB~uOHVjwS!Xho09_8*$la~=9r(_}$#Uq<AY
z%ljGt56|L#k$io9q5k&mzx#Ri_-^0e(rJlfc=!F&=ci=kecR#mo<<R4<p2Ed{RMK)
z{FYz$Pc9Wv!3L;{YpFE05n?){6yx!L$&+w(DJc&ttx>uKD}ju*mfks(7LDgLZb>1K
zldqg&)$k{wy%7C|igwnDqOH(sAa+ackWhm_8U4<wptO|%ktpR@NQm661a3uvb&g@R
z#x)(Fux*bv25aTnCCoH!N6rS_h>e}5iKGf`)--liotJ3V7o<OEoJ=kGIDgiUe{cCh
zPQi~yro)zE+e$g1p|SWm5N3~R60K3V#xlo=<6&Y}J6zL{N+y(m)&p&SMbobcA+W!F
z%{)$+<fZiDCsaw?-QIF{x5XGk*EOQwn<B??;_aKaxM~n$$aKR>PQ3DiD5Wsg(XFm=
ztBph)DM;hfXq461UP>ZX7s9w_+8ziYppC^;08EM^&zEze#2}f6_vj|jbiI^LiXqIi
zJo=T}$S5Teee*=fEtUl*kg-;yauyJau{dLhDdN3np2iAK7i2X>5n_<uh^d;zHG<RH
ziDU4dMbo#WfDmk%i*Z@$uOyvobi2$_2`D%Ci4ZdpbwWZLt%Y*f%2b82qGsi^B~oz?
zC8Jz}D&qKJ@Te^0erQKFLJ+QMmR3}~HdQWD>0Bp2Yz5ubenTu3Feg+Tt?O{wil)Ws
zIu;h~>jX8GZ&B6~LSh<^_}wd1DYVwf`IgKt{&_#h_kZPk_^Gdb<KBJ!S!Px(#_681
zEDy>smj<nNSQh_~m-^9yY$z9zw>p1!nVO+CV#|fLz{%NM8r)?I=SclD<#Tbo56fBS
z*ALFAaB^msjo@>0_>73YJo@kCq)zMJAH}-Z{@MPYbE!)Qx)^;ab^WP7_a9)j<~RK}
zeho!IC^I@0Y-LWSP9&97gvEIpX)Cj`Sx6a;&_QI|l2RswM5`N|bHq^f#VbdU;Ab4E
zNI%nAjWSNE3{J|tr=PZFi_&|O(x~94G(}5;J4Glb{Fb2cHA4TQl|$jM*3$PaC;j*e
za4k6zv%)AtGxX9oEeUO<bgQ)l?K$OBaAzv-TS%f%m(zD4Qy>Qz?>P7w*Y}j<@hO5)
zG^Rx>K@4ikNt)7+BX_qm7>(^aVu|dITQ*uiP~?Cqf-Fo04)c+eB8Ta~*FX7~yX~Iu
z`O=3xd-?>0qU+W?er!YpBf~fz8E21a8?3Q3P8!@gYuf99Zv7Nh3iEMGN~4TT)_Q+I
zSL$V$=9%aN+6X$@7)yv1L=pp(geH<hz-C3;uV}iS=sm&D%-*A&#TbcFl(EEd7P{R!
z4NxjGY?>pJpV3A#i=2h`i4t**Fd#NfD?Njl$$2Jaj|#QktrXTItdp{xOa?`&$^c23
zvyrG=DkUeDdU~o7snJXSB&0+tV$6+gq#TwDiA7^pp-kmZszytdLW$A<7p^JcIx1S>
zY=h~Ih;(WzxQ;A17ONNgQw_XAM)MH4(hQY~M1X&bgU?EscC}U#(fOHFVx?wxxM9Vx
zxhAB7FR#x!8W$y=@BQ*e{K4Pz&+~i#(7(<%KY4LBN*BvLb9vElJ-vU=4nBzg&lc{q
zPqAv1CgsBTF2xvCAO4f+Tex_n=5p?DJlytjVBCNIy#=JDe5JVm>4gVSMZNEv_e(+#
zPoK~A#=YZu*xWv?eZJ`RVlR#B<@kgK;xGNgPx4Ry6aN+Zz7xq>1=<)=N`w^1(^w~f
zf-w5pRpt0vupn};Dj<XzKaG?yF;5OZdrTV9MU1QrakLa&%Q8ku#$F&RQ>kV~N-h}d
zPEOPqTkmxV%t9=nG`4BPHrb|5L#9gcBX}*N!MO%)MK^|=P)f)q14?vY6jFY24pO91
zhQ4y2>5@@K6R|Q<Uu}pV_h>5<o}Xv=LQY}WY*=*-yK$yOQCb`N;dJwdl$hp;q7;qo
zn8pLg{f=?klgO;n4aT*+d9mZwCm%DkExH79^dx0~NYnKA;Aw{q-}{gJ5MMD3>%Nf)
zGX|__c=Y%=<8<WhtJh4oTh>=seCflFaBYvVj+6_2j#4hs=JY&<uw48|i3*G(M^R^L
zyA|C~dD^p(-q{d19*<0EA_mW@70g180hE-UQp)%uO`RAbwrF%CWuQ_rcXwOH!-1x4
zq(QC~(-_$w_M}{BT!%3lYoKcy8Y>4o9!CjCa~1=hia_j@Fv!lc$GJ$`wNk<f9#aw0
zT4f>0GgkD$liYZDQI112Y5g21#nZMe-Ec+GmMH{c^f(Qs(w*m=@zaRUGW%_<#TqT>
z>!@+gVXUbTenZm@sQNpU1xHW?iwdIm;PP3QS9p~K)oJ>5O{WX<Bu-ozEm|9>W$Kq$
z6beHMo;e0w*62ot5^UeIHo_dKlW6-cfV?NySN!20`#t=Df8$T^Q$O?7d-Z<c!U$g&
z+6T4u#gxe!?=QR5jhxFu%T4%>Ct70ea$1EGYwCG_(%3#%^-txl^03a&z3<)o>#}ZE
z9<F&mPA|N>+7eWp_W4<Tf+fp&=TOT1B<r;P{Y_o=E~GRAi$2BIKY79b;otjz@jHLx
zuS0=MBf4I6KVl$-C}W_;(6o+}5k4W#LP%LvIo&uhGzx8GEVXFi^-3iYV<Bme)zb6J
zT9aZXVNL-{oF~$_rzl<3(&YK19&0g{l%=nxjRZFW@~ptSB7irM(~RU22tHE$B)zgy
zg@lYompV~+0_Y~3u{h%dlcA(%Aen!pX&Q{Sgj{fqgHmd5?>jGF$SK&>W4wlF6hpU0
znTFB7hMZ^`0X<<F`t>7>wS<&7P6we)&b8KcIO3F%*?lb<%HAgufyNs8u9yE08=jbs
zZhb?}Ghse*Jnk_HuC5-_%!#5bZ{NI<rlfA^hDYdrBW0rP8>Yi8sy6@pbYvR0q#Ws*
z72N>Z*-D$6X!`+e8ll(#QkaP8fcG=U<DS_^a%{M|+K}VS+ZUe@<{9H!qG{+?8>vNR
zgsMXsl5uPgBhkO5X`M($b^?qO493iHdy6%S&3c6?8%im-t|P?2G>uRSt{Lc715MYG
zf~ZO5IOAg=kJ2+pA>y2+)uLQxtfd<^7-NX@$aF|d(~b}$#yJ|}FulfmQLE})i)%Vs
z=Q&nvNJ!vR6$G&wr&|<G24a(E9{0qM=-Q5C1U#gajL_MX2uV2Q<p~$3khR5GE2TNA
zQdT7#6grHCQZjL#QK{-b)Tk`yMD&3vL=xhRY}?@UOdNOIz4WX$R~SV5V~zrJA{RlC
zXCYsEc5}^-{I=i3zxF5o*L?Ns-#SOHpJm&5@X=+aa1p^TeTR#WDP<|#s7m;HHkM^E
z#kDuI7~4vJb7A;7O+1~p`?;Rj^ZNHEhwgvA3^Sb9zwZRqeT&UPP4U_LtmpGSr}5tH
zN;WD^=ko8RoF>->4JgfD`N_Y_ulnh4^20yp0~DS)8_BL~Mj4|AV?=9R$z&2)dXqUz
z8HXgGSMBNmg?2PrP{`gVP&I%mLfARa5vX+ZBvP12VMbNXE$>A_(;9P9hRkgX2p5@7
zg^*_EAnKKEH$W2h=te6HnwSdv{jP#irN^eULpzyD%z36{F92WX&{;<Eslq?=oQ2yd
z3Z6%$L{WxRv-E`m<U1{2$SLUcQ;IIQCedzggjOL${NXLiPdICFO~=*MH6DYRj&yCu
zYFJ~sHOFZt9q%x%;qlcqt`#s>&Y3jtWo}ofq_9oT&C_Re>uVq~jR)F&$2^YYV(7b;
zw7N!B)bjSt8~Wi9tLqgp&yWK1;Wf$MvKg+yd;IYr%^}w@9rpriDKgULY=g2=htn8h
z6x8)NO{6HGs~Bc9YnrZQnmhpDKp(${?JaG;!n!riv~;GECRC;R&YK&qpJu$Ds&ppk
zX5jJl4Obh_c-XUj`;y>~f>0)iA)SRAsV30${fhPaimvaOePBEu$;Du-OlKrNi`*Pq
z`XVKxeM4?KiBV%FhJ=ri-TolsZmShT(_pQ}`<d<Tj@259M#@<skaR}tRv-&m)%oo;
zq(qoK<1{giTMkFC?FwUCjCB|i(KvDvDZ)iovlPLLu&`04gs+AnEi~w}OpN6iMdDO}
z6k|2sk9a@JHMR~+VHmEVwd~%`%y+jm$<s6qP4G<9gfF5w?W_~>zmg%1eBTE*{44*%
z-@%Xm_#flr7cWnd?E9)T?}D_{yJMFRmL(`(iY1q&kbBw5Ql6<syOi?*yib1f{GHD-
z%ojU(=iQ6a&Anr}Fu>fi+wV>deVcQ>TtzBnH3fKoMd>|R**PORJFN3$nE({wYv1@L
zKl%6m9?x&2G-RE__C1S+I(3v<X^qq7td)R~0SV<qXN!!VXS7K;2L)ZL%MI2^nJEf=
zP#t?Nq%a9>w=SZcf?*C)`s}(6WeoG-mK;WM&XPrIDN_?7x@5VxL%`L(Y{^lc>p3z7
z=>aw<j1`hll}C)5D`3~w{zN$aSPSzrEM`C);XF@OFCx=eD=GHpSiV@JxyTqKR+F`e
zS>t-NJA$8?4tsn6%nN85{QeGt#~Mr5b`%Wb{uTZ>(>0Ey6s}v*Gz~slLYi@|V;I)L
zE}9}GMVwQtS1bCqVZYsRI8Gc7TXOcS21l3@+3(nIU(&5NB!sG)AR$4Co~~`s&2v&6
zTO$`rIHJ+Sl!+u<qftqb$xO3n4uP)ixLWm6>y0zkNQtQ*I#Mdc`G{4H-F8ADBR^VG
zqu4c1KKKyt1Ktavo3p~)>a56H#_YMf^&F-V({!k;i1SR>Ta4{#8%Nt(QkdAz2fPoI
zT59Rq9&HWt+=?ompYY>IOcC20>GlVVwpHAxU|dJrL6|4}JY#ag7%>Jh%p8u}6J}xP
z8Z;W))JP}>sek(V@KSKyKvm5V0x)Z_*2t(`oS_t<H7^-og_%OgcoR`*axNHaP(?VG
zQua72+%G9dlt#35b3XCMlvI`J{Kz!!2-99fndXTcC&D3+M;U!psXi<UX|m_4S6w2m
zM4nvr{DFVzpXZPK@&64!^$)*xVMZ?u(gTOEZ0}S)I`ths50M>bZ*ft|IXQ&7!lgLQ
zyXfn8bA<cO<MJHtS@*Lux%d9X;aq<E-uCX7vK}t)&7{i1*NZcHpLgJ1ujSGy%8ysa
zRH;1|G=K8{_-FVp|B_$G&^1^hOh<Y47AKM<D-24Jirk;t3SU~$7Gn(|dWzDbNuWd{
zywvDC7099T2<dd9<u8@r$Z1V(U`|3Y)Aj@1N+fxc(WF2Hy~+hh6~s#xNzZ^jMG7fI
zX>uFKYI6kzn%2>G8*=i5I0{uyQMH_;>L|8=;!;XNccWXBQy3@vEK(;10!bAD-v#j6
z+ZW$t=vqmk(#+=Sn%)`qrSRtVmc#8W?XY3H+f!5~hY^($#&)>l3`#Q}wy09@&M;3S
z>NtYY7$xAhd75clgKK-pf#dcy3c0v#)6f`AZj`Va5`?kS3H#=Be8cT+i#81@Myzd6
z1?I`quX{Gvk4efB=9xGjC>d76ilPkD9K<V=@o^+aU%mWFBmzsOBuCr!Sl8g^07}xl
znCHsBl$qy|7-x*ptgfChY_2OwUQ5$70E947QpTE=_00{5$aFkl?SOVI7R@lMP)d<g
zWPjX44(ztKOykJa)gxBxE3#KKc8zIUf>wB_biYZah{%NbNK%Tf>jh=rf>oN7JU-48
zpJlShYC`b%5GYz<Rl(KqT&<JPj>JS{lroQPY5O%zJ4oMwP(g0$#Ji$UXx9M(ax2=^
z9)-+^n?*HFgEi7{SK1Q%g!f0Zt|cOCQ7WTUL(G9R&m5<bY4&ujV_2=|nikg&LUfoj
z+8DvT$lb$K*)T<UPEpVhGao$K@XP<PU&v4Y%+J1K?(Q4phgR<Lzsvd;>sEVq>1fJ(
z90K<;g#5s|p2GNxivN3ecrnzp;2&x!YKg|Ve=P49_lvhKt4HtnhVt%rOWn`ONtTP_
zyl5C)nEMB3a&AMPzay{3F>5NNsC+kPvP%44{vUsl|Nd|Ibyxx=6mEBS_!McH7ORAJ
zOBvxX6Ou|x%$ejpDQZIWLjPV-_X^Gauw#EXV2q}3I{Lno%r$4cpP`OnrwFzbs3W7~
ziT&+Cf{rkg!z@mqW+@>EQaPrGaUF_63_@>Xw4rhC-t}I<q(w_F0;MHT)<U+45_=<5
z+Ev0<RUw{=@;3k`v5?4@@cD|4U&txke(P(*)taX5i2FOH+gCKMmF~H+6zilGJs)w-
zRy|`&%7uCRmYgFxjyPi&_qSwaY5EQQ^<&z81u3$<ea-9FFX{Rft&<tx5IqIx+7@dn
zT}6uc!-1SK&WZYxH3>f+>DCUE%n7fD0Yzp$9w@G-9X2FCv3v7U#3vLqZI3n*^~aK!
zLtyfPN><7cqCn`47I7WrI$GOH<eVanHc|#FU`E4Xmzc*RCB4EZ=vEt)vcxdSL_rej
zZ-t_BE#~?$e%MrT8%NBdEM}}UZgWyHu9`BfR+Ra`ej4SWX?sjNV2r~VO=DY>lIB+Q
zBZtEdA3Vcq&DG5lnx-->Dx@}=JTXWXlTjSTnS!EUU7^~Zq8%k=(Yi=dU&Ryx3XJev
zS!*$-J=JeRn20&kSVKXRN+cx_46?TBFXT$0k)wd%l82JXRik!UbQSAyPGgeLCTrtp
zRvS{vbk3k$LpyBHTAFV2VM`7>sJ~e$i5OBf5)Ri&hLhun3KPHfSN{rr_8Z^gKm4nI
z<D&Gw#O?Lp2f_7e`F79QJxl(;8C-6^EIJa4X<Q=Z%ddZ%Nc*%URcrb6opRpS!z}2X
zGShoXJnx@)`kc$9XSwIR?w9D!&ZGWQ9pI(f=fW>kP7W!Pv*IWI*5Bjx-Ja)<uQ={+
zd9&S;wBq@CgSM7%oG38~wTw2Wn!2eOK|vG-n_+`>J@Y(qyS?M(%NIO*_MFDr6UUo2
zhA9QX4Y(GAIc3JiTBh+xIPNO@F?G@~qia3CF$&!@^mZlUQ$?neN(>_XS-}^&W}-1g
zLYgQUjA?MqfU^x)8R1STqDscOPRgqDgdtSSsWihC1<O)SgXVMolwS<I26Wbtqvvqg
zany>Y>1c<6X8nZNT%r5{j&F%E;F=EGu4&s2LxDKcxfNE!JbUIK<%7O!Bn=CJpcLcm
z@lm?#Mu`T1_cPnuw<4N>T+rh*5&VQPmcAQkbVtsLd7RiEwjAzWgD%``u84Ew?#&B^
z=|G%D@JEu>9FK*=VaIA%kxItTUM6$KiL4Z>)#^mxw(2*i#*&S}8i%11^zbxGNyj>_
zuAa~~1LI*&@CR}X<lspvk`c}&iV{R|4wNL3t235yI*_&F=p%lf8G6Ix>l<3@h{n>k
z15sP%Y2>)y6a0)XN^a_=!C6Dsb;MK%^CaV&CE+XfW7T%F-GFL3G6|zw!Zfkp?eXI%
zNMK#)`W362$8@VJv~kSS#C+VLltyQ99->TTQASZTCjv4`rf5UaiAcdBgnv2Cs48zK
zOt2jbo2B$^W}G#&s}?tO*set(ERmE%2$f(_EBc}5s_jullVf5|5o>z7)eW%}-fr*s
z^c!DeIv&^z12<QXXqq0B5#+13!gCV?DtWX~{K#+pr%5UD|M`i(dA3of@{ZW2_din?
z=2l(y1m3OPU#@$Oj4M8)oZ8OsDFrPDaVCR%zmU#*ngZ`#=Q0)ejICVwj(fSmXJiZa
zsYA{d|8h6-t&9DCW+~}>F7mPpOF<n<kWaq)jgR?5f8u|~@A>cjMpmuj=J}UcKYGkC
zY^2PnUNLXq*1RwgtI}E5cZ@z(Qn|X5lF7+qbmr;fr#yN5jOln_zuV%h(C0&uthDV#
zUw+YZUU~%DVn~outt?aoDY`N@E6Fx=1=v;`O3p}WORGvaDOo@+!k%nRVzp{%TZ=Ld
zRZCIE7>shH7-Z5fjO65m(nf3g#z>xAL`BYizC_^*I)%0c=g2XkT#s`Op8~BiTs{3V
zo$d*{PswjYEZexASTytOAqV2PB{iDs&6?rrhE;P#HVw);^6|j#cpwx_ZhEE|h%wPO
z4bC+<V<|?)!;ErdG(*={9;3vn=@~XRxPD-}yG51EqeoBZ`i^#e#c*Xf9uEXR;^%{a
zQHFsOh1D?39_1{>2y(a`43n4i(rCe9IM)(Y#xyNme^tdg{Fx`SN{lvLk1`HFxl=>Z
zD&f89h7HyY#Oc8P%?rYGMC%IJiZiZndHm=p$K!!GZApH@TIr1_C#;7#7gE|YPJw{J
zwuzJ?K1@u0gp!CM)3}CVH3<1%$?Udo1;}Gti8LB5WuQSu<5P}O6ZH{dCQdV{%=jq6
zA>Ghp%}S8sMoN4IP0WdDJP39niIGdC5K|(A1UaEg!4%O&kc9Urr6b!gtgc8oN~x$>
zp|v0}qo0{)X&6U8)_#pNyk(+9MhWLM;nPSe8P|6xTgcXOoJX1DFb%zOSgR%FFES;9
zwWOqFWKd^*^{@P8{6|0Wlbl<_oQ?TKO@6Uk@6tY8uJ)c1&E<Nha?<@RTvA}Xv*CBD
zR`1`(J&KBZn_1o~_rI#{t#LoAC=d2<VX(j5F?`nY@VxFF!aabVI_;_;Bf3hRKrV7<
z3jXHb`v-jT`i@`xL*E0g=lbd?P1lps%sd`M<jXm9)8ZTga&#@}I58czGD4a}fQu4&
zbal<JzCr62tu*uDmU%jm3$(Qa+d4-EqMs>Q0>T&*x@5v43%g{0W){pq#V%y;YlZ^0
z?L=S1C{bPXLV01drEMEx(#(@5B+oeQn8HL+GHNYgR7HbD7ezDlWUmRaf=Z2MjsdF+
zrj~+j(|@kt=nFZ8rtMH|Pt!TNVME*XyngX@c83G%_6@`J6SOVFzUOLq%=+q@STei)
z9e1~{nMPzzS37KX&HDNGGH1iv*PjyOL@KG0sI>ytND-sK+J>fSX;BzwSzSG$T|Yvh
z823B&Z$Bj+?@%hz7#P}y6gt+cfvfcfktEB)g%dMbxAg04t~S@~cXyN!nCAl}dlAIB
zy2d!iJd3!8ANTC{2Ur`L^)*I!sM4WLVLFb&i5?;m1pMU$0H*YcFdx}%U!#?#8*a!h
zp<0D*d%Bxv=)EJ2Z=po`RydMwt~NY-bi-lWvEROBN|Bt4+#o?n=jN2~S?F=IZfTnp
zrWE{qKoxlQ>;t;4mxh|3n1g4IiS0D=`0*91X3b&0BhDUc97EsNalI&0WGZvmzG8d0
zC9A^K)eXA6hGJ1g(3Gh}P%W8;)n<hv;l~}j?OPNE>w2^mO^i4TNU&)enZIU|-iuPK
zR@Z1P95TTl@KLZ5DT%022oY7GXhtbi%H(OT>Ulz%mP2HJxTDRURlnl<zwd_$&%aEL
zUhIf*(%Uf>RUZ`XG(H;HkAh$LgWtzL`K$hM{_KD7msx5s_h#k+N~s9l`wn6;WQ%P(
zN5u7;4{hf%b(B-3ZN49EUp9~5*GOLWe(F=yMO}WsezY8mTAbU%l^<;N0Yu{B&kNGN
z&*KLk`lQSKo-g5hmqnE5GBY_53NF}%I;Q#u|Ja}8-~LyBl;ixCdD^pDUkN}>S@imm
zkWMHETG7}ZZ5!J5NNk-<*+oxEniM^zThXL~(*<i3$;pJCYg-!Il1kzAo3|XtBaJhx
zx<(vCW%-REpp=##p2?5}Xc>c)qf-LgNZE=S@we7twI)%eXbUIqsJ3%xrSWmbPe;kt
zRYSY((aNGQn8u-0&&(0iK-&#yt(lH{=HO|Z=qv~p>^m=C%qeW1arNkm7-n+v#3`Y4
zCi;>6i?36rJ3<`s^Oh1Kl$oY$ab#9)&+3B>%C-0uxw|_O`Xj??g(?Nxths6g@MTM4
zP97fvlsf;et<>VQCYy?u#fs-w8mkj3O>AFJ#8MbsPv<%`nsGWxojXRfGF(4;Ot-p*
zoaj5pZrY=DW`EeD8%KZrjJ8|Rw1V0;m5==P?loo_88%N*R}MwShl%KC#&II1Ow$c=
zVXdnrCCfC8D6Lry*AP9&w;z*rLo=)y+!IEX@jGe89;T6RedDXNO^Y>(aU6*;(hqA~
z)3Vzi*xhaMA>#UhetpGieNES|QA%+<G>|=y9zUb&SIm9{ZE-8%3~Jh*)zyZsapYKd
zv;7uR_9%S=N;8imQEO6T+3#+7_2z{rNTrNX4MW$EN@O07#O+(!{)%C^0UBbQ8Rs2`
z!!4WDHJf2TY0W%pa+LDV5*??YnC8GV?wEs+Au5IR19HR~Q=81v0MxnwiN<J376)I7
z3|F*Gi|f}ED|#JCN3?O|rkBicnn)!JNYI+frWsL){<f)uAT5et{)>MBfBCQfE#BO1
zPZbUkgFhpBA2x5_XWLF%%kR|psih)L*o#tGZ)FCacgjDPfU)l+sd&%t@L-=6Bk|rO
z;beRtY~-PXShAqY@>O}?7SH+AgOmBT$N$cGER$Ge5f7@{C<~`l!TUfgmY7F&`z=X9
zKU`zoKx=Blxq?2I?_e~|FrY)An2s}$Tr(aIjAcuTLc!fMB7^CxDqBoJ$md*3s+uM3
zAX+li46wBZ6(_>D6Hr}}?`Bnbm~tT@Qkz9vl(Xn$2_ce;rqwMZa7LVTjGo5I07|JG
zX(a?g8p)}_bZcxkU<-6zkF}b{h&@F~89p4}e6DN%g`C3X$#Yh{X55ZUhdVHq-WuBL
zHO0@^bcB*I-5S?*%wb|052PRszU${7ps$`2rUS=Uuh{K&#A(mCyXDc-j~G@fUVZW{
zf}dz~qUcQ1Hnd%baxKa_Vhl{XJEp@PV_JxT<K0W*{w00eU~GqN211zG?;^oBct0{8
z50n(?*K3AWMBw*_w-i58vH(83pYh`n-M{6=2shGxcSoq4L{JzHx1vn8sswj_#92$*
zG&mGp*9pgaDUdbBG_2P*%+tjFxTj5t)?~1OaT~CWfM>RcnUW_`fyoD&reVEW5mVyr
z+qb;Ay`$|`+->jJ-rZ7Crt1d!)tXWg^L&Jy34X2-p)|j@`y-#c_?W)G=KAI-u91<{
zGEW@0cZ`R(%weXKOh2qqx-iZM=1J1u>+8o9>oCm<pEWL*3X}=#wr|Ph6`dO>O$#bh
z(o7gfX04f=G}PwtC_+S<%4uXUc0)HHP2e0Dk2_KfqT*K)G_Hyufvb3oQltT=pmIMY
zrhMf1`Yl-%`k|%k*I3(AN)=zq!f|humS(tXu|=kz7Nxk77e-9WFa7yH#FLvV-rQ~9
znc`f~-~zcWwcdU<d=CmiXVdn;E-sGcAtl8^Ve(ng^F8NtD(l>L0}oP{2j5)2v5eC$
z`@P@TeWzY@4^QwNjJo3N)RyaZ`smwbDWAhx)MHdik(cDF6nqH$Z~w1<j^Fvu{<pBs
z(DrLgvm$B7JnrybCJ7s(1@#+75sETan!TzKw)eViRRTt3aBv8TQd}Ka5Z*9nMdL(k
zq1V=0)3hDkFi^A>X<Z5_6ml)K7G&xVB>+tgBqbs`rW^3NF#8d$9Ni$n-hA9M9gh?W
z&Nu<@MWNeKMfjDHPNWi01-fBHh=JqrAU(X2aZ01w>q?9N`5f{W8%cfn2}fqqv_p}&
zT5s5_pAlM5oc8DvF<3T_K4kO!d&uF)?aOa4k2?y5NTgAUs|s8z3RE1K_HQ}dy~OB3
zpzvz9L+6RkYMg>tL@KXU23z2`+sch$6*0_&FhVN$X~eb#Z5u)=9DU-=b`n1E6e(dO
z%m*4{(ak{eB4%Ng#<e|9yERjY7+q)#7^m57McBnPj>pXhD4U5T6Q&(qx1w2JGmTrh
zS&Kv=IVHkL8nof+>KR9O;5h9V!+~iw6h>0?IIXY+T9de&L#5^gg(Jj(YdeP3irc##
zMNPDA!?O=QgcKxV$dTjzju5=S-HibZ|KO*83fDBuekQ~~(*}<FTNFp;eI$(sa`db^
z$Mt4IEQV?Hw4KyB3kln-=vUXMW<^qlc36>ZOUV(NHA9h_>+$f4xHmMep>-|mZcX-y
z{q9bvK;kSSI4GggwhB8mb+kD$?k8?<ckB;$B4LP3;%ruHy0#N8B_}MC-g}hNIM*^x
zBQM^3!h956=$o5?$Jb97R*z`<4PD>ktY)s=MBA=#{Th--jR$gw62Z;Feygn^#>Dr2
z@Qi=()vv$5LQ>AAcrmQ+)c-GR*m+SWBdKbnZIlvS01)T2Z1Z8={US42Z0@}&p7%AD
z%lW%!Y}jH|Yf<b$HR-h9OE%Sq=X-D6hb5kOP8!s?H}LQ>-J|CI%rerl+2w=;8Lrhb
zr6vS+U;p@9(k}{yY%|Jc!5wu1FtgTTozO&-5U|;!RjSMjnK?#Wk>}~<Ctqit_6(~l
zOgqrDJ(Nu2EMbn!^GN5c;9^3+&l9N>Y}=8P!P<_i%?4d{&P@$yi&Df?NJ_9jN-5@$
zPW`YHJdL*GLt>i7+Aqztjl)^dK?!~Yt7+O5Ar?Z6v{xHekDueZfv!Xjw{O|sd2}hv
zX=XZpiqQ)HP-N;emM=DvdiyCundz*e8#aWj$j6zOJvj<PrqPB}=g?yeyuQ0dmB4Cf
z@rNxc8FJeYP$CYKB1&bZ{R?6kxq9+thO0-sdHaGm9hegM{f?3g^XR0>T$_^6U~G@>
zi)a=2y<iVESFBveaTa>)<Y(sXE#6DS8vK#QSPs)j%3`>iwnYPNtH6jN8Y$8mgIjOt
z+7^eU5IBx|cH3Kyhnee}#}o`c1wf%gONl}iF$@EZwm4(CzP@I)*$CPY%Y3|J+P`Jq
zy}^f>rfKNg7INk|d-lg8Ww&S5H4NQ~_4<a@YM^bbD3t|InnuRivp-JkL&7wMwqN0z
zmY5<b&kRFDzk0%>M^A|%F^^l6i43hKC@qsuTFc065k)+qrneZ^iiSkD#-lM!M=^#J
zHOUT`u5e`yxobHd_jn&f1f^zKMjM>U7z;+rh;mFt^dRDqIYwfT(Ykd%i0Ycvl~^(3
zoTYIMUB!2lQZS}Lp=jCx=LWjr3Wb9-W6F%vrs`UFjA~K35W_^64%jdYf!|W>Rzg%5
za~58iD!=%fe(isa|My?}Pb!7_Vx7+Q{c5G&S>7qhEbm{|nI5dXsMrZ7|7Bm{zJ9gr
z`*JU}Q}7-lyK`mdB1Tu2&g3GCsk#(3eR;>$pHKhuWC)l0JUh#B|Df(I54oJ)-R?VA
zQ!eOIo;xL1g=yS#I35T&(Oo^^`ImnVtu_=_X|7Qi?TBS2j9c=2M3u~&{gJPK`WlR5
z(^#y^w63FHt!cWo>?Vgwo1y9Z70x+=_c%nMO;=tpr39cFW5gob3`MnM15INwDH5Wn
z!MREwGv|PFnsvVsAg2%b`A8<ywGGNBLQI1CH-?a*RR(1<DcLGP;K@M{Lv3S7T_>uO
z+R~{bF~ilPn{WRc*5wO11xL%xdP8Fr*H1n`*#_??#wnp~2U(_g!nDQT!JE6cy!zx5
zHk%cjZXnJh{%|B2M|=H9XlbLTXpM0Wr!t*W=wVGRnd#UIQB@it&5XM{jMDVO6>YP^
z_A>6ObKp&h9Nvr^c3ZT2$>zy(8rNc)9^HGkeq=rZ))JM>{vUkCbv@TlKf)Nt+c%#Q
zk6X0u2r<!UMaoHtq+G)k4&)pek1{#auRFA}=+shBG}h7&J(NP6xA?;?O`0%mLt|D#
z+iN?K{A>nn7fB(4-cxd7)im_ZG9*y8p@f-6%Q5!r8!?hvX7@vat=Y2f0=60HhYs7c
zSl6&wtx(3`L%_$x><iQWfNdO`t7{qpX*yDx0Sfq%C@Rx90e%^0Wm>{$=5dcd9Pp_y
zY&JCATC}1=6#eClOmAdqGDod3D#=9IoSDW*E)i`sUE5(?Co-lvRGRd{dbPn_Z3tn*
zaXe5E)Oh1sjFItOw1rYIm6a7`VLY;G2QrSBJb6B1l&;{p#CEr39FH7!ca-R9CLs&6
zMi?#qJhR>1GLJ`Yt{>4~$#luEj`Us2IQe%0Y^t0L(Pu9YYUvM*`C@>V|DC@hrhcjX
zm-iUwcVEAgkKB880$jt=AgR5QrA&44dfDM;Z}9$daY~DmJ)i4CXH|{=MRM@I*YAAe
z5}$FA>6}D}PFYl`TFi;t+gI4}fcKuk$xw{#2b68ttTyO&;209GU%nyE6Jsvi9**ET
zTx0q2vyXUm^%zQ#{u6*Ql%jD>FA|hmGJ=vard?A?B*h@e{mFCOy(NVmS~q<3;Xgt@
zbd>1vS+ZRXbam3vH4UzHkcv?1Scl2N7N(uZY8o4e(PC}GpoCICMo)=}cs!DL&F<BR
zLUd7L5{>6+nm8U_lhJHGPl=8%=oFq@uXw!f1&K_-8iVaSjLIDLcj%I^Wah(`G9TIW
zJ>U1GkMJQ9<{;gCKVq7eM>kKvcEl*-lwsPjeff2M_GkVs{q<8GJ^u<<H;>sL_PqV{
zo0M4O;wfkv!+O<{;>h9kYpk<0T?;pl=%$7|AGrP2SJCYTGhCyzuxh&DhPG?zuO8$2
zE1G`8=qF;HaGhaubHliA31Q;y_Kv=>T(1Y>d|-RHrCV)42^*@(5@A~nK{r@8kaFaB
z++h_oP6=|_2Z)*!CY0^5Z6vCmq%3N9MDIFmIO6xWw9bmItg`q5^HHI~Nc0o8+Z`!s
zoQCUF&+6)i;rc0IK4P1WH9{y?icD*{uA?iEV&Zt5@$(*4CbYYzRBk7?lIE26BjLD}
z5+Kl6VRTI;qD#SLc=PHd-}>s7>qpP%*H<*{096^Sv4+iREt422(KL=26V7Iu?utUG
zS`a`%s4cXmYo#vQcLNH^4zv?qq*O(doYoZtCus4*;Q$IYYoP;}cSqd%n)PZ$ZX0|M
zrb&Vbm3uEuXX~7Z_k=(Q38O8x>A1RnOx|3x*=%U~9u*?bpE`cu@BA(N%m4bHJVnR~
z2@piO5f;qAJ4WoCcWU`%F_z`59Z??WFte<p&fk9T=b!O>-sd;H>`j#O=QaBf=W}rm
zg$qwt&e_7f_a7X{!<VP}*5}A7E?=tiA4{L$a+B}fky^5(0y$%?;nCwqDARE@T+=j`
z2vtWtk)|V=Be&zOvX(ZiZazdQ%Ll6s|LA&!b`7&XP{Pv076X}!h(*N^P*lbM>u}Bx
zrXx8;{5&yDM`D&~#TOrco#Q;S-rSIq<%4HW$hmO0-{abb^=i#-oX~ww-(6v}<~VPK
zqQ*1~H&3w@Xp7%dvM44NwCpd;%=3s+;D(ko?HTXh)TX*{$C<9jYPjC4NjWngZ@>NT
zQkO5}6goG{%&b3f_sJWIvRrMRiWWkUQcX#b;3p3Ak$zawt=6np9T_bg;#z^R><>rU
zm1Ef4ptWTlSIoOt`0<V~?#N-rG(C^j*EAW%@qp2FmcAtX?3s@{4!bSJIqZV$b+)Rw
zC4ABJ{hD^YA?HjS59q29Fm7L?ipKAEym`GN7U=sM9zA-_qsNb!)&sjYFWKI{!l}YI
z9@!suSlbX|#``V%-Hyjko>GR#n077vMEg4oiS>G*?HxG;LJGtw;yP$LOW!#xJxM7_
zXfdIscux!yg@kGwta3!_MKVx1axo;pABE$oxqX4L9jngKwGA0X)Acm#M?`I<UhQY1
zzooSaTN>tKn7tzAgwYz;IqZ5(Q&!~0GVQ>Wf@@n+E|?-(xd471FePB}$aQnY&671L
zWe)ops~x@W7{?>W@m8n=a$(Me!LDeW!!_<i3P_4%9L_pak><Ei+iGmT!6{SAg(Js#
zX7&NABD?J^U;XMo<gnlK!ABqQ(X;2Q8p{|a#ydg3D<c4(5=AY~lt|+=UDx88o~$fB
zcyb2YIJ$lyra;P>ah$M5)7VBR0xGNO(gRQ^Ax5i$L8p58dlz*Avt3qJN)gJ0QWi0k
zy3PA0k0ob#PciDW{dWxW(qwqYnyP!>d6)I{B11XRRD71}d9c}c4xvs7eVc6H+g-JX
zd;P4E-14Sq>j_6u2%f&*uzuW=!jU;l#G+Ak0%@LR0&@McF4(4}>DO4-vKm%|6u2Ed
zeK*h#E0UiX4?E^@L}^9eI2r9s3beu)C@PXv!dhq>OQ#guFwzVg9)0O6Y_4wT`kq)4
zDHewHTAr^`XonuVUemQ}j5Zi&$R#n(UNA|{GVMn8uRbNsdyH+cZ4a(Txt5{n1dM4M
zF(o!vH-wl(2~itx9a=j~>1imaDn<L9mM`QKw!4X|t2OQF5n*>^97hUoS@nY;@p2aG
z;~Y>$6LVyoNAQti9K(7;)Q(R-eSzIgJo`xGGMl#J?)DZx2R5r4k}~XWU*L~B8rL$c
zZs@xex^ej6@y9Kt%oO9GX=PMwo@m<+>v~MvW0a*a9jiyr=~f$zv*;L^_P5OYx2WWq
zrUMkus%@}si*+5=!0p|wup#C`3IT0k7<$HO=Eb+Zj#7@tkDs7QWV(CB)%u#g+n`oG
ztKKquPh)#>$!IlG;*6hW@R4Z}N@$-mbEpD7aVC4wm2kkYYA{VN>E>~)fHg%7N4DEn
z?Dsp`zN2*wW7F{X@gw^6HO4x0JrAQa(HP?HmN4znx~IQ-hS^;4`ppYozxgJwZ{M;S
zUUR*<VrZ=N4~ifuQ;K4gQb1|Lqs^MDp~aBVCX<eDNhl0~lrsCEnC6K&B%-n?+lb_0
zf;o9g^u*~%&QZwx)>k+!5~4X5T)(0n)?gin!$`_73|FYWFdeq!<7--Dd2)4y&5=AE
zIo`d+U0tCXONcXM$tIgYG<UVtWG&eUMG;a2C3+C1^25XwC{Z}9tkI+t+3vP{<%7rk
zmww4F;4l8fe@fBjWH7ZdCjs>PhEu)!=Vg?AkMg0Mqv2A7g>#{Ic;AC}Y3IK~6ux96
zml6IwVVle2DT|OzbqnQG<heM9a{g>_I+rJTQNp=@0q$R$&pC<u8Q))eA<OZr^YJJE
zO^kuveoH$HyxkuN(*f5QW<N3Xov?>yxtE8&WvbdLWY2!PXTN>PIL*>m=zF@>vGXF>
z({&A9+eomx-BOgIUta-=kPD$GhV?ZaF#D0~uIKv0uh4WWN(uChr7;b`N6~IjFpU#A
z2UJXq`A~akhQ?X;yElCL@h4oZS2SAVng;8xaL!^(Luu9+=R|2OC(%901#+ZQRmNBe
z+l0*-Qv?Va=G)J8^}iTSVRcP0j-oZ|D~D^P=8IA&<%oegtD7)Nz~$NvxNb!(g<~j`
z;;{XiX`cAmul+PHKK+<4f9cDld7`9(&x&FSW&f7qwnWzw{6Ro%S|d~k7BgJYw`+=9
zv3>iByEiYfy70k=UqZDT=A_W3Ww+n6-`|pQq3>6y0vStjJ)}gMCt@t<)?(WZ?F@r$
zn2rUvO0=sMqlChF)j6i)ju13An>CG7(h%Bx%<*t1dennRY--~eN14ddBw{$AipQEl
z$pg_J@W&k~MULYr?1vP54(JR+)3bSGu~xIcyW_BXi^`Fq>$tjk#?`Ygp^Rqx;v3u@
zjwsg<azT$rR@aX(!<9&xZf=<TNKqF0miFq3XRV{LiZt%Q?@{9|nGIktO#}W&9Cv~$
z4pC6rjUmkw$KwI*glRHlgv`)Xjk5wxE5aVyG`Mb!?N*c+D$q<(vd|06`&1=K!8SeC
zD2mGHG_h_C+B9r7*Qlb{X*kB2t}m>H4acWjycaDCbMu_*Cm#sQs=rXdN*-MjAd~^2
zAQ;A!gse;}AsvfC333P$MJc67VaDmgW<8*c5R3umvCUJNqnfab^2hz<!nmD$!csaa
zr%a)6lHoZWgLr}q%Xc}hdNRTf9npJ2_qy|Y*}!tFr!vjM=D_`S@jb`z@cfou!u!to
zVyh31_Wtoc@P^A49&YKKqgxuzO8AV7(ac0r2G^}{T~A`mGzra0iUDIR$y%IS(U}A}
z;DeO3&Crr_Vt@MzKTlGYGT<7APl-8rvR1fmkkSqsKWC;n5^}>l&rDNd=zBUhpfHqh
z#HW#<I*e)2jlmj4ms;UEvmHq#3<YZ(WW}cKXmzA93T37L>bi~;JboVKy5zt(2Vzca
z)>lF$pM@$z0Vza$C4tm25W>WM`}#fi#=FZGn?8E;lABF0bhS>?tqiv9g(@P$JY}pi
ztcDv*F~kfB!{*UbB8J29K#n6%Rz2JO4y6j-dk$|u<&jmao?Me{!+1QR<|8>w0@^^~
zW2((EjqL`E>+w0U+wZyE-l3Et%sZy-r!>Zp=fd6i5_Dm6b%WLpPo`;GxtFa&X_;x(
ziON@W<nGlgKK$r=p=8>w2Y*BrO$Za?G0?Re+D0)SkCZwIwEfm6H2r{b4t1wkKYog7
z*9^TQ96d@y$r(Q%AbWIs6vT7i;@T@#K47gwn}+dtAj}iTy$AyJ&T)OUrtezFhz>y!
zg{Q9TX+HQehr2g~;5qDXMch7l{NyRNLu&=5A!Hc$TT0FhSJznE(YOflK$y0q`9RkV
zglL2<H3h7y^yN7dCW*pLBc-b2IN@VpShZ}fHdxhg_jXH;1(m_12^$o4eM2*>@Yay5
zVcKQ(+c$i0^_Z@6gy_iyrg2X?9%QW9G}xx&)$W#@6(tuIrm>V_1u{m-mfDup)rNv*
z9uG|0EsBiodQ_O1r(*^CXhDf)2&u3eM~t=%?LbkM7=<Brv*~HOE2bFH?NA*;jh{~j
z@O_c-`$OSn?WOizauyxkQtmkw={=W|lQ;$T7kfK_=^h@&MepEjpU<Xs$p+pz!pk1Q
zgQCu5ZU4bezU^|4BIM!M-f;vEO#i)h$SG%ea2%J1e8y@kt!OQxp0>UL+oGMrm53+d
zedajHNOzePws(x<jxZi@*3oE3R!V5y-GDI$*9hRV*jApS>`+aTd&_i`n9xlp&rr(D
zzTiC=Qy5x<Qo;h!I*ri|V;r;h6s2f;xwmZ7fP&yXDhFP_evL7TVZEWG#5iXnk+n07
zbHD;a*D}qs1QkVbJdC)ep>>Uby_Im28Lf#b1BMhbNqrZnvmZbI5?!l_^N6HuiM|2X
zcI0_NC|o09j-;5;*2(mbinP|?NgVfEv^Cr~!+1QvJj<NDN%X5LKK#=6l1kz2G~<25
zwJp7EXd5dHo~9+nM4S#Bjt35RTRwdL6h9u&=8A&h)!QTM%@xm|t+=~;$(tA7qVLz-
zJi5Wpf^o1)f%o*TrxfYrH%4>2zvb;qi)tD%tka0e0c&B_3bP)Vr-^UfzQQ(+ei#_8
z*X&Ee8ihVa;>Rz@pT5AZ)^u&ls&9zBkmt>QqIE5Y!x7i^Xx)%<;?<{b$R)EGI*iT4
z`AEr$H!t6Cc>9XY)gy-MYm`L{Y1(rPg?{K6)HBArH>e_z`RTAj`$#s9_2!yzJg~ic
z&BhwKt82Dz9ic?x?Hjpn&f$z<clVafdd(C(>(xdkD;ix%yfo{Z8*)y(y}hLr82TQo
z3-fVLaRb}!TS9_d6dP4Y)1KpD$Mwx)w6Zt@scqO?-w<<VcbrgK5ptyOdctuc9(Saa
zSzTRGGQ=zt8&w|#tU?ctHffyIFl~ABwf_vP0}6gT;#>=PpwVDSY!7>~vh>56q6%#V
z8admdm0=7XwYtJiGp$uX!kU)f`Wt@(f8+1`6yN;#(^|rjsQJ`<{;a@!F<KWUvQ(;-
zlonaYvjJbI4eo!J^RUh3N2m3c50)nK^7iGn-j}(&bC4JBs|WADXZfr;(EHMXi=^b;
z<G48PW!(3@rKk&cb?@*VbYGN4lZn#~?`3MI+pLI1k$k3WuUK7uh&Gzt-5cWej>BO~
zh>`Vmhf<n3WoghlscdKsO@D=5J;GSc`YLdD_Yxmx4tFmh7HJwcir^!U9$%Al&u)8%
zD}oaU^F$5?V-wc-N|W7+UPFf96Db8z3iLDMv`5>9ZgY)ug><}S_j=FUm!GiT-Lf8f
zT-(xh9jny}qs1<k=sAu%k@wYxk`pl&tkIO%uyR*~;k$rS$flvcz9FZC)`j3_l2VNG
zOqfNZH0Q#6Jm89Fwb?Lk_xQuD@ReK3oF{bBSY`3q6H1_44?KSIguFkHc3ZCcj#=OE
zX#GggsX3u@#^2r2gvjpp4R^s)P~5Cn^hRiBMW(Lf?f8nfZ$74V24@skeT!`rcdtIC
zbCy*<@b~`q-ys#v&68&g*Egh6=vv3qr%y0SGle5sE1>|-fXam1TgJmk*RRkCru~d6
zmNL&Aa{@DC&@{-8(b1t2yncC0^s~@=Tgwy`x)jFkmi=~*Ti<|bQF$h%z~QjNxt4X`
z;P<yM9r@PV*Q{=CSgki`t7+_G+OmdFuuaGC{7Z!B@w-=;afgnc={OO0Z-}vQ)f)D%
zKjrqt*D=FJG8=!yPo9-+Nm<w$5620m6feH@0!n6o*wOY4Xu}jd(NCZZ(I;~9B&#@V
zXPS1RUkyC!SL}8NQu071?_cri_7kD2)<O#Q`t@rzH;=JaZs=ik#oKujgIP;d-jA$S
z8?3d2@s8W=R+_-hGW4yC1|LSz=9ot2;2Ey2>G};NDRyse>m*c%Z5`MS-C9&m#A&4X
zjOPfh!LF`YT|a>sx%>26O#7XnV#AEeQi8f4N58-N|GsiZ^$71w8=bNQEk><8G<5Yl
z8hx?8IFkDuN%sxwdmF#^dII%*1=Q)C#knozr8D5|+qQY%d(NkKAK3BBt0K+lZ=ZSG
z+w(hhr*j5$dRHqNXE`4BjA29>%kEVWQaIrH`YFz>K^t7tP<9jB+btizdCNZjEUU-Q
zSZ%J*A@k+${W*N;qYtGF>_i78rO0?Z@OS?9-{R%#Z!xr*pZn4G^2|NqFmB1C$C-@L
zh5hj*N*&Rz1s_pu4@xt~Npz!=CnwNo;rt@AeftI<0<Ifqx;0@=<T61R8Cn5;c6y+3
zJs-dLq=G;#H`g~jd-{yGZ{85Xf$OX56P-tjGlhiF137Cxe)(OXxiE1;Cr=2XA8f5A
zH9fW&Sgo#@_IFJCEwc}B^#nGL@Z})2<N}>;ST&jBcx1i4rcss@BHQhbS0BIR@y&|6
zaZ9TT<L=0ZU;2KE?a0aF#{-l|oF-J9`MZDnZ?Rqt{J;<VAa4#^thMMAuwCHMy5TTq
zx_(9QS!j0^tT$_>!=9WX>-Czqw_B`%t}*QPd*1Bda&^7I=EBR@uUKt1=$L4&k-BBq
zQ2b1BhT-Oh!L_25kRs{!C4T=FZ4|g3=ukNj=1BCylvtEzwYkC$YxZwnLGf(wcIatB
zwF6<E2}c?63@P$v`<lio%FQ*;o<2qw#qREo7!s*ioayP?!m8^j(@0kmhk4@Vi%+@g
zJG!R9keK|;&PQ~!LFquzJ>U4|8yQy{EXVP{uo6d+V&t&jp|qhJR&)yK<CH?4Ja@Yt
z+FFb<lu4LbyS8PVCQ4B>?SZ!K@bg3|na(!k0yzg-XL$MPr%d|;S;3PhPnnJfUc7k4
zZa*=sH`u(Tu?BKv9uF9;=(-k^WxJay<+p(DtTuwwoQ}MH`GV(P`X2H#$KBlvwy!@$
zDZ}Ra2GjJ+Q7FEL&S7K5DVSp<DMf!Za=6<Pg6Hk4Z&AX?hu`}Z3id&iTWj#SM!Ba>
z=VHho7_hP!{Bm{-4+<<x1?3WXT;Rmlb1v-EZ!SnU^}dOva|!Cc;Zzq!egU#wGQ6r&
zui|8A3m1ESxSTD1Sx)?cjei&HRxVlr?}MzJr=K2TFW%kngG>YJD7B{TT2$Ag+`v3f
zsC=NzBhz%p94C?6EeUTGH_x8aK7PiV-4WOI46A{1Jg{+vm4P|T><<T~!wyvv<8H@w
z*D!qeC01R_;4IVK8=T3kS1l?`l<aYF<kdGm#dT|*fA~GLU5mCIO0{@Dvb%eOpGSh9
z&=j6Ld4`E2oicQd#@d3PkNDz+eOM<NV=%g5=vuz#OCPa6>_zuJOOR|;A#1~Sd&gn7
z!x+sp9?3aj2a9g5$W_SaJ1t+xDcBOo^T^v*F9`F2b>A^O`hdwNg4SY8QV?PG?F){6
zN1To_v+i1Q*Mn+lZ6hOCRjaq}2O0&+TH19>j&S$lCA-I;Fxy7Z#0J{d@#)vTiUOWI
zc|?wZ<6%qFG<5BXySrQN#)+%v-$Q90^UduDSe|SghuasljmB!r-Tp{chVS{xS7=(n
zU0AJXn}#q;bLPWm&&f*j>eEknGamThqYnr^VqL@WIMEDO{M_&R0h)fr{?(3ndqC+C
zTFo&gdSiI~`YkCXuC8w=N|SSBe>iY>^@0*+x~`!g8Zd<zkF16c0+15x%@wa-9l70)
zte!sOFph-7k>hSpoFmzG^x83N&6{cD&C5@?S`Fk9IUGl(X+ojt`VN%hus^U~t!SL4
z#3LVF54?H%ig{Et?G0H&+jr#gh*F8;JkqQ>OxJMt=9c+*;A&XmOCcOb!Z@LFqU{{p
zyF1obYrK!dH+Q0ZWi?G>nUZk3AC3p!yn0EBk*ALy5xg{llOJ(JUVZwr@WS$aUw%qf
z;7eva9&pa0xRQQHm`+V!tqs#WQM6?oJ^js7Zk~TgQkw1VEg=1-7cXC8+n$(Uy<VZ8
zJ|g>>>tV&vbtI)2lxN)UQ6)3j$eX(@hutl%>G{3C{WtJ0{=q+bE(hf-fJSY=T(V9s
z*8TD|EK@%6XE|jKi)~wMyclzRav(Wp;n=!oyg50BQ<Z1|q}@M?#eJO}%EB~ynPn{h
zy(o#jQ@(kxS9p*?JeWECb|a_v4)Vc<zY`Gew6AwQl`Kq^$a>XtV2y6Cc=FLTemoNP
zcNj9(IAIQ*_N>12oae4%ItD&Cj<nW73EZX~(~F<so7<Np-O;Z$nAT!c!p<`{t>Z8k
zn1#+}_IqL&8QK*_J3^Q^?vEm(R}|m+_+w)EgzM`^IHef(dvOL9#&O3u9eMrf$0%bN
zt{$=3Z20sWKf@;<e?tH030>FHH<p+SN{Mc{Q3@3$8_-%YPY0~E5{&to>+2gz$%I%~
zH64>StgfDOGd%wGze`=dkW-ivv}lZL(HeZ1xc!-bK)2d({pcw}-=lIvoJLgBF?!FK
z3^zBA2_bN|JK*;(=r=1Sf21D<K(o1C5vCD;9Kl=WdFBUx@aK|<%*Vjnx39@L)0E5{
z605%F!;e1V(ajaxyIWA2-Qgfn-)h6;3#D<a*NzxHZ(hBGdC$1>v`!PeXS2DX9abEt
z5pv-;9??o>(#P97BH;QCs|}ByJSC)z#;_jNTz~Y0(I>v~wQth)EA-PF`XBuHthM8~
zf5WT)^lz~H+D|b$!&iRJ&tVRc5CeTb5Mu;C<7`7pMrb9}B=hny@bZ(7>H3bWHQIIP
z5V(8s8mmF8LN~Oq9%wsDxP6JTmJ}oNG!j<>w(Yp;SA6{OYm_xmG^@ri3_Z$eqE4us
zu-fwE`86dcoYis@B>~@s;2{S5aZkeFh5=9JZSZWau5l?6jpNPBPel7bYrgf(PZ;_Y
zUDNTkpZ*!1KYzx}^#-fp)!R4tl)1XO=Gg}yad&%%H3mP8_`{xg8u{qS71~;k#{<U^
z9zD7dx?*SO265b1H`iz<lAeB?xV_t9oFywuQkG9&zNYI8tG*>!#jv?zzdLZ)AF-xz
z)jRfY-r)Vn%NO6m8cW{~^j*UgJjXH6G?wj~E&IC(qoEnr{K6mnK7R7=|I9^%UAg^F
zHt8HGm;3*)fYnZ>OzV5Em*D?0?p?BrlcNCgg>xtiq8BcZ$#whcp>Zwu{wu7R`d$xW
zA^JPBv0i$D3sm$&hjCGpx<|=zd2*lOAT9@})PpzGL&JYQ*qR~T_f(6o&AE_@C&dHP
z<e7K3%*PSJBv_elMal{79OGeT))8GatuB=Dj+91pID$W*twpa^41JF-o@u*dzuglR
zU^K7azG2gMSTf@@6Qagw0hN_prw@bhr#*i17(dT^?Q36U-0iu3^oYk#pY!;`=ad3r
zyhE!Dt|bB8s^`NmJxBG|T-`iDg-A((LSmecWH0iY<8&nDR2$-oL5qgM%a^Zs{q`-x
zaD{!;^Z3bAjB|YY>DS)-TRdF87<lda8E&<riFK5-7;-+)Sc_GKqzgrB+-8l6!cp%U
zCrp6ZV4H?uy&!LMB1|Jq(=Z=B$|!tMgpiqL&zo<4lleFj=S2VPInB_^O#3vGVr00!
z!fGW%X09a$*uHv$kD1Mf9VRQ%II%nKNc$aaN%-l2Sr0hZQM9Sxw>x5i<K%hv{5f}b
zTfBzD@j#nB>#k!O_pH_fp(ysV$MkFNri5;L)=!@ZTcU30){p3h9`nN^hfjJiRwe;l
zOUa6e$Vld9Aml{bG{nOZV>GI3@o(?g-tAa*1JO??D^izvJdmAXj?%EVT65gqp`1cx
z;d!4&Pl|#uXgkY19!WXVkk|~5u|rSW_RQlE-4whZ>D+)V4x?cn_t>^WDa#xqP1kYQ
z9vRkaR##V;b<g2;iz<;ACQ_PNUk!9^!}jhDoWU13-tMr*a5!wyjbYVxBB-RHbq1p7
z)as_=XTR}Hc5h$t!PCbyZO3u;Y#v=>P%>huEywvtczerLI}k!97f;`H%wfh@O^SxU
z|M!22cIf%wqvxzz%W?abrfIlZtx$7h-FNtugacn2{{OW7$C6~-nx=<6OK;^WWM=;Q
z)TvXt1%!38(J;1wG}4ScBStjigagia6}$ruIpdVL01n6j0g?dP46VDms!r7rze*L#
z)z(XKup@s_RWR!^oZ6Gj@bK`6aQ9eiec$)K&s$UlDlhWCcW>X*l#ZfyOq0h)%W)V9
zbB{?0%_INiKl~$x_RKF{zlZM@>hDkNd^ax;@ExdaS+~-MmQl{nq4M9)g7Y)#Q}7(a
z_}u$nHUW@7XeoX8!XM>3{7Ae1e5b#>*b4ET^!10U=y#jY5BwAP2b<FOcl~R!9DcZX
z|CM$Bhu?=E^vC|X73I0bmH+b0=7-+{zLTE*=rMnM36_*Wf=5M(b|r&fAcKiO*Y&s<
zndSkl1ykE003$W+agULZMC9RgKnTTU33l%2Pm!@7=nhAYr!&$Tu5WMIt!kuz9H6XM
z3s)=#LP#JMZjwroB*8zs^UG(+YY+;b1TMuTx>y@flDqqNynpu^DHX3i{urSmLw9CO
zhH**=-y!GrJ0Fkt9wjqcE;CH`LQz;pU01yR_FD>5GLHjee<b&b!~55N-HY&7uD{Wy
zAg6(c_DBSFn+>vABU8j_%iIrS1R*C}Ss<jsTE{d^OZ&)Le2iRv{4qHvQV3{m=*EGI
z-4-J)QH+Ge?7(VsiO@$LzW$0!t9bR%E83@f^mdKY5+4Ka9v+acpfHv&b>x(JdUr?D
zY<Sqe=5##r;^hlORWgo?!+IDe=5vqjM^5K6F(qz4{)F{v$NK69pCoo3+3qe0Vdk)Z
zVzb>KoWo^sMl<w1tMw(WZg~IZYv$t}QkT4V_35I4KkqS0GBPs^6F1uxWl=8RDiP5J
zjL;NHu)SO(gk*m@@Ob}@!Wzf{ktBs((Zxu4wIxlFbUcw_Ak~&xFG)>p-%{AX?Ps5{
z+O3#|5kE~-Wr0?Tx~d2<V&xL*;mMRCP=(CX{R5}A<NUPe)yE$ZQ5;V#nvBs3(^wv!
z9x44yxmqN0A}_Az*?S7N<W&@|;=0*z>L={Qj@j=h%N53!NR=49Cxr<J4C9iaacl<;
zha;Q1K?{kJk~q%Tokl5%QHxYhXo)hG^Bj?R;(R)?DvRex!c5G8S1)hzIWbHFB3rC2
zn8$$*$^GL4u3Q*Zu5L(Wg)tH<B*U>K42h@qK+!uk*Ed{TU*lZO)6*049C-cik@K(K
z@^>$H{8sJw@4tS_?&6ZK-`$gPf)A*O%#ySv^LOo_4=N*a`fkW91qit~`-5MMd&@(7
z-@f@_i~hr!n{s*{I{)Bw4&N`9KR(=le|=b7`My{G!#Tx|um7=L{)av3uRryF=md!$
zeee9^|NoUQ;76^rAAPSM-1L2?=?7Iv{%d~nXCF+;k)iMC_9J6l076Xw5<%_0xx7a0
z=}gLjpc9+*73cO$V4`kH=5a!6!;7nr7^lQ=9ze~^oEW`da?YEI$$R>?qjUyWR2)tn
z$8(1+9LhLcy``!eL`uB*=GRoKn(Nhu?sQl((yJ|-iyMlf<nVOI+;wQF@F_7(fu<<v
zPe&fU{g&O!&vEM&+Sa%T!%^{Y_r$P&hct@(oL7H7pQxLPW4qvoVhZS@APP9PM^60-
zK_t%8|LE)c!|QLfDg4R5{ujLb_zPbB<hRHo@%Z?DVe{t%5zglm7nhf4BvK0YhXcwO
zs;WXDSgkks9EejwDND)%r5(c*8N-0J8m%K%mB?yE)vQ@p79ln5<2}>R(#-?XE<h_+
zu9)UPN&_O#)Vq>qRnk4(VXURtlpK$btk*lTTF}d;EKxR6*EQwt5?8OVWsP&?qNYMI
zwgW}sP)f1gUM?g<AL)+=N^8kkvNevWKlAYISBSaA8OzsC??@q0R3&xYVCRu(7&yKD
z1|Ju~=P(R7Cu!E}1vkMn_)G~_wL_}{?*m;Qkj7CI&4SDoJ<%w#koeK#r<v~5FPQGa
zb9r&W-1Z15kV<j;(JP`42mwlFq*2U%CXN$b7+4jKvM5ltp;=#2ZyLswk<O3@kBS*t
zYMP5RI%K;02e#LjjHe^--o8T!$#!cPhYkRhJ=LnD+^rDB3hg!&#g?gikByNykC^<t
z2$}+^TPmYC+`Z%O;fd{f!`0>Gq8Tp=mPBhor8Q;UFaT+q5%Y{nAeF`y3aJ&P6=bPU
zWl3E%v`+`tn+;P+to5=2jcvzx>JU*<u2&?WSa103?_cu%>#yh!2UaMAvTU|%R!u<&
znX6Zy)2E8BzkNq~_Z9z#-~TDU|A)WNfA#<R|4}p>{`nvOoOkyR@SzgOc?nZViSInh
z-&GU&`_5ASurR0ZK8KW~oc>a4itkO#A5P7GrN91z>v?7VaOVH6{qw^;Am?YE+z(gh
z5C8i?+vxe<9~J{YrnY{lWW--Rz5npTzh>tAgI{R5=|_9ik9UO+;X2~GkAFN*faic5
zp%F4smW!i34kM@Ip3~E~fDVfqr6nn3j!%#H6tVW%P*#<M1j7_iMag=-!uKa~3Uu8W
zDKbjIw%($(rahlHbv;G1<0l_|f^v?#hX;@f<qF<>`Afd~`b(5nWFK%MGIl3?@OUj~
z)CFR05n&`r!{+LS=K3XZ7%7w^&mGh8jubkCw)EY=w_ksaViCqfGW|F(rHm>Z!WC58
z3pT6ub2wMe&@NSol%OR`MgJdg{f#z-=HiN*7cX&IaXKHVouVr&AtuVYWLI9XUabk!
z%;~t8h?=rMsb@5BiZ~7D?u;`PJePdG{`yP$L&v9|e}uvlVx-vl=L}#V1<uDseUyZx
zn`Z9r9*OCIkc!{_;xnqIM##kZ=}1OkFV>Vw!^{#!lL9#>&W8iX!yQFYQmj`ryDPLQ
z2;Nf`CHnf3GXw4U%w}~#dp<LsN33;RY%U;rx~}7wfA%Lp#;AnbY)L6Gb{*5`x$6eb
zU%uh?s^a3}f-D49X+ny`IkRb+?~<5?egvhMQ&<Lz2&5A9-H0m##1X9wbyL$^Zn?XA
z#Kerr8C6Nzu1CrxI&9T6ATmBiQb=^;1R_${f+!NmOj9~ymQXlW(h+5aa+Z1SsNBN*
zIlTRrwmnm7xl{}(^Zwz9kw^}a-OV*dTc&BoHETB84HO23mQ7JJ35%&Jvd)AyEhk^8
z**`v_tz%v95OyRCVrexF66qY$Srmd80!c0koZ1#RA<$AV=Y-FRqI66t@a@}gxxCt8
z)#9*5DYC0ED@F1XbC^(T$F`~IpB^z<GN;5GmrxQbEYr}EQ|9K?C#<e6IrSrwgi{Ku
zW-fMXKK;qx;iv!TAJISDlb_yVjpCpFFa8Oe-H!kKAN>O!AND-H{g#k2SGTwP`Okk&
ze|lPow#LzH*8Erh&;O2;pR1b?i7Aqw@fJ!+5RhVeo*m>L9inr}&l|<_2J>AT>4%Hs
zhgJ1Q+rjgv{u+|ecU_(D+ebgT{`z_D^TYjkxc#B+wcP!O^~m#W`FY;(-A45N#`E3(
zz8l7W_alFQzYo!<&$s<$&)fHx#yW)$6Vuo-oCl-=r%I~2AwACyMOqkbDTBEv$U<^B
zohVl;);BLG>lJAnXwQ2lKVs_=BLt_m<No1^o10rU>oxs2l44Js&r}xl)G>5Doh^t%
z%hkF@2#Lu7E0!7M@jS7Ac+39o9kZXgyx7wDnM!LmH@BoI@E8B`HL-i*<K}Z5YRRB)
zD#Z0Q^X!?WVVFI2DcKYTWdz&Jh8H&<6Mf)(I)cfVMxvF%HY?Umjd=adUw*>=+UswI
zxhP7$dHrkhmoqPJugEb{xq^P^h+(2xt+~5<gGdRT6Bidt3uc-|PVJf1YR&b{CF3eF
zqe&rfynkR*m)u@lQZ*Il!#z5AruW}6n*vkSpg?L($_qjCeCnC!gi?vy>l@nhk-Kl6
z`1aRd(VttSQvA*z{atQ8{Ru_2LR1B^u*}oUI3+^5XB=nFckfxPwshSS)@Y`w#Sbk`
zgQ(1sE*B#@C+72+hlhK#T;AZ_xkV`W{F9gLKK~hoHN5-gYsd-hO7^ESxj#}>tHn(!
zGSf7os+xy~J;(Ejq9`a`fe?_CA<h%q?T#@7u2ws`eo11BA@Gs0s9~m(7Fj6FwqmOa
zA}~%flLS8xOy?7p<>b_?R!jPlg#K{El_k+Srl&L0(}2FX!tNX~Cj2nbpU>o!XxkPM
zB1MS2{`NImD3Xz^nuZt>Ras$+VzKp46Nj#)hz;+4{tKq{1+y%<{3J4k6YVe{mBxxC
zPg}XbvHyw~J*%eS>gtN)(*w>JT;T}wi~vgG7C!#@h#xzOqQ+Q5wW^6^oVMJ&e2KA^
zG|lwWOtZP>JVCvCNd&5tXsYGZRoIHTJ+ZbLAr$>MQC2l#>`_`#6cuiL!M3Of)5z)J
zjyVo&Tq50ni!0V364j@#V4BFuLmc`1cE?XYy5;=I72O<o`ROP8!B0QIcSpYY>Pyc3
z#7}?k_xRuZ(|?LEj`O+axIggbn{T)~&QvSSpMUu^@7~{$LVyfI-|_2j-z`0B^7kil
z@%?G}*|5M5rKvw^EB)yD%YCK%-6H+r(|_5=@ZtU+*4`fm{e1t~=P%FS&o>f3c$^=v
z>3?Nz@Z$>QL$UFfHisWpAs>G4=kn<XlM#R=4p%!;Y8m?$s~x7=;Iqa9ZU2N!nf7p|
zEQ<w=ayTK*JyY~_{m50+%*TQK{afbtfYxfMvPGt@8`kR`DMpUR6GAM0tvA2E=X~lY
zY(ZVt^nHu)Bd>0*=*OAU`OGLXLK;?=udpdG&GSNFbrtKiV1GLD_Hd#<b%^MhW9I++
zul_X&MZKzd@#4jT)y|6#?Bh>9<>g18Aho0)T7vg{{i|Orq+uaYV)68x&u8*5K@7z7
zx8P`Ya-y)3iA0(w#$kp$k;zN>)weh;n7V~qKeUE%oapC??e>B&O?>n1dt$Lh)D49-
zbmtRJf^d#WXoRhJ_~zHp_e?(X;<tWEvXVI`%4)@Cd(Czu2scZ(k08^X&h*pF(*SDo
z0_6m~5TxX3!<jTiYHjfWmJ}w({^^7(4NtGX=IZ7Jaq3ZVqHmw*yPoUoTPEo_w|gF*
z_6x8|DQsCRX1zJmY${^Pg#Jtj4yQHU>A-Y8uqhqI?e(IbagKQ!NI9XDz^ogLGfNmu
zh^*@^-@boK;~Ysyin7AElE=ddsU0`h*Yx{uNePO2M`1L1?il+s{WvWQlo${xFQugx
zNUh0!BE?8dfugeX(})s^F(;zYRMsJ5MouHAH}7drM^?Kng|&!MGIaxVo={3|KmQaZ
z6}8iZ;OWn2j9EZog>&eb`1sRT)axB;^OB+O$=yVrC&sx0qbRf@5h<Gu-RX=UM!eEg
zRgF=KqHx5r#`it=$Z$N7lBc-5BDs=D3Dm}t=YR-_T~iShL^0y?j4~y{uJNYE)N7Ov
zWCFw7Q`ri)lB9WN%o$m4p(wDzV3ff)O|{)I1<*osac!9U8Y`g?6Yb-#$T1V!njB{2
z=?Pbrq#+Zk4Z<2OH!H}V&89+>Kza$bq2AqKS6j;UCCv?N+K$c5D_&-}zJ9^~JWEs@
zn9g^oJo5gRf6jmNKm0pPQ6rtD8z%1W9{J~g{Bxe&ZqLi*4`2}D2bDwqaFG6SD`-(I
zEg1nHT3A0?f%#rR^kIJRgLV4*>qmZuA1~}byy=HCh96ZAf9-?*^%IT!qv5&WuT(UA
zXgy`35A<;fXIgF6jHBf7)N=FkCv2~;*dN}aQewWG2tF`PBkRqYoHIq~h%s;+NBlga
zOGjaMkQ25na795;R%oT^x}F>YA_dyx5vNP)mlxQwWcD682R7>!S2tJOJsx>Ho+zq{
zLS`DLh%piSo-8EH0%b~SYhm&PGUa-U+-%uhj66Oam|`La3Eo4FY_-SlU-PTu13?--
z`TQqLQ(~Ap>bk+0lKb~}gy`AtpSZocU^I!T`&(!q$uMEd4b{a8az+PFnxFG?Q^XGo
zSG=hj+Og;T{*l5uoLO+w5kW|vY2LFfN|-X+%Nv|?L|Z}hl+6-}8Rwa%S|L_jHie}>
zJ^`8L>XM`l3XoYaOe2$R*i;jLwD|~kaZT~^CDtk0ySJ=F!gm8vR~#N6*uVXfZKbg{
zw^-vC`;O`Sh!Tlm=m;UQxw>Q&l5%rJHA|`~kkvxnT$dH~JW#I;WmU3U*W7>gCBOWO
zU(i%FMNv>R1z{SI$s?@8pCV5WkHi>}QZkGq+sjL);6ZBcAMcmg$x@SpXFQ*o^_gv@
z8HSd_*ULcs+;$Z1#BRGGOfw-w&eKR~9Xfi(X{0+x%*7U2R!HaA-`!D3(4s(PN%!=G
zK(gAd2=mDC*rHQpm<Fm<$!gbN>WXf^ryV_BN=hY3=OYop!|_bc0%sgWRf2(fU896T
z*B7L!A*LRrXLb9MwOzE+N@VgpFt#ULAu(b_mIha?P*Re@gtSZcHA*ud4`gM~^#*<Y
zIj*iS#?p5^{pmf-1DjQi%nIcy8d-2Y97$#(>K6B0p-7P_ikc}56ir2ZHm((}#A!{a
zW`@%}`^S4+eZ}hLB~yyzVIs#s99z!wnQ<DBNZx*XU<?zx%PaccBhT+y?<Noq6-z=l
zqV*y<-Ce&#tS-^T1zsqyE3~o{W{ut4a%l`o7x>A;bf&VJ)7{rx?JoG={;&QCkEfP$
zy}}g@ZQJsn{o(H{5wFG~s}fz+U<{?yNFn&Y{9pfXbbbHb!nv%y-<#(?h>b#uq?mu8
ze)|6UAwK&90()6=KS1?9R1H5WTYfy9cvdXs?@EUcKf_;c4}F+R{MFh<JR4qrw6T2m
zGc1+&him!u65jWCBF<O}SE0%^lRje=?7#hn{r&;@%n;gKeT<7R9y_Yb3!2ph!F$}t
z7f4q#KfdAVmw(J~?kTDU>m2uY?=VV{d?crYvyQg!C|4zdNK<cUg+p4)eA@H&I3Seb
z=JuMps%Z*`^c}zY_y2<AJ&$M4?&W8E{K;q3>lKH#MSN!QKGL@X(HM50&Pzn-!#$=j
z#IdF8Cfeh~W_QW1Zcs#AS+Ks?lE#_#@JLyg$YO`8YmSFIuFNIY7XLx({NE6Wx99Hu
z384+U-Hz>MP1X`^G+o~<gJK+y#{;L+8Ep;o9C4;VgvA^+v?sDMi1ExY^k^yAZa-os
z^6+#|-}hX;_!#3VlGaqsith1|d6=lJ#>^wy)nqhb=nxrhKYB@+Jnb~Iz4?Td+ajll
z*XfS(@|t{lf@#2p$ktXI9`3oaTgp|*>F~t+yN;@`oK7dovZSh3xS~WTix+VH>Ln<^
z2Oi(QL*_&#Gqo*4iqxwWzxA`9V2$B)I$;XCplxjdTJv-`@y)wCv{Jmhy=Aj2Sgjgv
z>J?9i<HC;0&<;J{zJ5=bXUejmsw=9xhC)&_Te!a?j4fEbY_NJUkImkbg(78xE-gt*
zm;;O+Efun<z!>IpOCV#d#g79;+3@lwpOB*GSj#0pK?=eYD3!t2Yd-(@XH4@%|M(gk
zCsfSbeDX2E8bS_`BJ+75GzFJeH=G_0glS@Zc|m=#L&U`WX^$YINJ~FG!uDbZ+F+}i
z5C!8nBZ~qRBSDnB*}q0)!NqEWP@1W45w51tn%mnOPKQTk0@7;I>^UDE*gqaIwnUXH
z=DNjI6;dmT%^D%|;x#cAr6oRiz%P}XR+NS1cz8rrD`eGBUA;oP2Hzi<kN0d>Gj&<8
z?mA)&6!nVsbS4aEl(tm$4&xj?Cz1qagSBf|T_fsC6bi2`J_)jclmc2#kY|j}q$G*C
zq`LkTwcgUVC&cg#PD)DQah0I?qo1NO3|)_s=Gl|8WOZvP`9J?3|95Pgl5sebMWXed
zxjU1+$2o_v1rf<V`?tT~=WiboVu7X!DM=*0{PG(fo(|td?<}(no-2!`I!HfgBjx9w
z(+Akz_e1CpGmyoEC*}7t-S0=>&maHjb3d9ye1EUZ@^~L|ti`f1{P6mK5s7|SB7UVN
zr9dGvVIG*mOj2ga94`VYKC&$WhKSTB<T#^!LZ2i<?ulK`x?ZD&W(uD9@d=p{?db?P
zvuajs)-|h5O*gbmgXiwe8}!93tId|Os?lNObpI`R>Ulao^5x;k&E*cWy5Mjcx%&Ku
ztCz2^>n;5>v)>;W$B}Q}yuruFt4}^fr$A{XpI%>3G%No6=YL9d@e$WIw@kx8^fRa9
z9&AbDN}`XfsuG_fLIiFvceG8z&~|+N)!%|7HH3-n?iytaR+}{`WkT=>qcCMj2$84b
zk%zbMaN6O@ikK3vs;Jf*V$PfoPi)p3>eAw5AdEfxH^1io=|~oGQ5LlwDu76dd^mtu
z1}1ZydG~nFMYF}#4R62wlHvHs_2nfp3+AUM((LJ!L(7Z`N6zON?Ht3<6LrLHS4+l#
zNTlRZB7<Dygw|QIkhJd}nY_o@lKWqM#mg7hpkc~0goG|jjMc>9gh-L1u0Vp(nk<)i
zJy(=OsSsMD>lHu!!#|=hj)n3zGE6hmyeGzlbAoyHL=UUo<w9i4EbJAjF;z)X*VOZh
z{nRp!<AU8CXR6XMM9+SIA_Y%#dqv0z;b*LO6wL-v7(9@T#VLcc#ghJ&JmYX)6k^Sa
zsqYa|VYFHBz6RW8hni>Fr#pm#&E*>79N7v)Obq9dRaql96@D5Cv*&ysFsn$J;Nhz;
zIXt|hG>Rw`L(*(tzCvEUMCTck2IlsG`FzHvHRbAtbLXkAKEg_mj}vo$glS?(d$e(^
zw;O~~JUu=ld?d$6@H1s?5GoVINMSNgEwHhnO&}%4I;;ShS#T#xP&$JqvE6PErGvR;
zU6y1rwl2_eLJkT!M;sneN}}GN%@Q8epA?f{oc=LKg4PJ>5V|3nHNtF`(=rM5oCU3v
zAVxvPU{ywhnXF;XhNvw`Dhg%!!dV!PSQ*Jta@_YsKO^Un&1Hpf8I(Y)29YECha=u+
zs&YXbi2h6^GTW;yQ%GcO8Ro=4`3L_o|NB4u1B|P2NQ|>&0{@@?`oG~{{i}b&FwRUN
z@&Ep3{~8~{k1LJzq3`tExcV^ncrF9JhXsDGs!F8qstYN9@Z~>ndw)QB{CH#eu;4#e
z7T*oHzi%mh7<#7^@qQ#uBf?~|o*BjiJ1sbMXI4!^Rg@T`ASPDP6A$-{=Oe-v=w?lr
z13h<8?9irQm;&qF4vb|w9XRiM;yh85j^lApJ9MmeTeMKD>NQVKuPMud)1$(e5*>OT
z-@nI*Nc5h%(45*9*<5h_+rNiZmUH{ayZ7HLl31XsE4uDPwcVmcX8-<MI%YDGdUws#
zwbWNP+}ymPEDGLy^D7>o9#Ane>jr5IPmgafN+6}+`eKQvRVt&*Urh`ApS&ql&6-uy
zAk2q;c;Ni>#A>@Evh=an&6cU|Ft%LGP%(1u2kd%_CsS=MdGX0lSg+Td4^O<_KXL!|
zEp0#Y*%!Y>QI`y3i!+M3J<*2=QLgYLn)L>c<otNze0QeZ?~!4~kBggGm5$>uBV{B_
zBht?tzP;!2>YDbrC-@O-4VRmSN=)QQaCNa^=(}Zs>?cl7Pc+SnEG1ul^$lPC;uj>o
z;QIE8_0=WHShQL)Rn4kI#~DOsx7jeyGv}O8LLxILy(A*}5Xe$80@F0$`<C{!@JT-V
z^ou0~L|d|!4Aa0ok62@$6)K6t{h9s46H;WH0w)C#k82D~UE$UxqBe|Oj}!`LEVii@
z5EqgZ5=u#$%>~mq;^zr1C2{t&506w$gR_Q72uxX0tyhb2@L`W>m%)E!0M#%O5MxUb
z%_4-!iL3?FQ%_Y_h^plB)u(JWTjFpcD#hmN3O@}v*RZ?%7`uCcsxBxQdHC`*?|=Ol
zOP0jx$mt_Xxw~Y$UUPc?4g0${OirVoQRYer`tH1_-Q+cohcjUu7OJbC$uUrEFG$TA
ztrgx+SZh&s$<t+8$gC~}R8GV&5`3h&++YetI-NPZK5+Tz=gcJ1*i$)4!Z1cKs~y$$
z1tgeHkI<f%=OGTH^FVeDZrhNQWeyoF<PxAB6DBYHZV=B38jDB)6(S-=LJVXS(rQH3
z*z8ccVV)+Uh!ihMbV|6>k<u8VpQ+25sce|$iL1{(A&fn3e`5BDLKH|XD4Pw1aRhBC
zr3EbG{*1X)oTI0l6lIG1gP(rNr5HHR8MV6PAODm8g7$pm-Iu@O|Nj5@ukk*RRmO*a
zVA%jtO3zDpMhf{|D`!!}<sTjY|0<zEe81xVrBT6mEvn}$<@5u;!UvGw_r^CXVd^^+
zi>o>38Idjj=HLGU*Oa`_ilOV4IaWV#S(dDf<f|{=fwTzyK)qYTHX%Z!Jx`pvmSWR@
zln5Uf+cUfEhABqwPA&C%4a0~@lD>Q7a6a*?_iwrV=o8k>nmqTUxO6b2v23@Ow0*~K
z{q7&I+TJi6pZJ%5`rolym0VnGQ5a6c#EVZp=knuExY*usI-TkFZ`rLYhW1Ef1Xm8;
z{KX$rt~Q*{Cx&j|YIR9<`!hCG&CSar)7;V?PaIEYj1HhApa1L+e)F4xPcZu#rzL%N
zCdJ4+P24{`V5=2MDKwd)Ziq3XjAM=?##$a8_pB~2&}Bn51$V~|qf6?mk2#NttB*=t
zSu*=YJ1la<){ZC)bJkR4iOCtiKXDo#nIzo2xMe(@INUwa_h+uJUyuxlt4mTTnD;$O
zM-q|!>4-Iwa<{`&1yV(<TcNb38z%a`gAj1!g)5>GpS--~cC*3HORn$x$0tMxY*saO
z;g}~#(%}1%qd(C!Ys#uY2(=JFS)_LB)r!;cNY{7E+c`p6mb|>Vp)Bn(3z$5|^Aqb;
zgF%sVAmzm91IB1{Q4zgIC@89$ei%qoPd7!}ZilVb2pqm0$TCx!0$~g}W(wy3P+A}>
zg*OI4##o1z3atzhg+S7tPD@H1aI;$z!UW8er6HLbS1l=ULOLXZvYe1Q<2sL@XD+U9
zS$*^q;yg2+_cWUg)@V-0BkeF_NSHXGpMJ^X>3~du=JJ}5C1c2x(KCmUd74SdQ<ViV
z%=nzpKxQf7<KeI%*sB%Cm<d9&zP?yak;<~&U6QAcbALq33@mYsMo6*}i0BD^W}ZfL
z;Sf2oat&+^Z@&JT*T4K@+~$gEy+x^p(-4TtQmP#}DwxiU-5H<3G;6fXq+!M@gHaMc
zj|kr*TLtBg1o*K>M2plGp)^J4iQ@@~1V0ick11*}n%R3o*WyH2T3%`iZ;Io9b~DOo
zgv*2s4Fap|f@UZxI$;pSmbo8!^Y$B7%7HTIqQYv+)OEx|OP@w$U4nJ^IWhZ$kct;C
zFA(*PPghqE1N)C(@ppdqyBy9(LXQ0R|Mt)L=BsZw9FM%)pP$=D&nNxo%>jNePktT@
z{ACRk&xnp?+>pP!X<7TfHwKF5*+dr0W+6Ycmh$qb#%TUe|K&f#Pb0Qid_GEKwwoRG
zfA(MS{_R^-QS<cfNZ&4;pwk>M-eW)g31zur7K+wsoU2HEOBZ{(vE}LXM6+(F%8F{;
zpvsz;H!rAO1>U`X&;8Sxja9TmPj;5e^(EWQmf`$>j1i+XWxYk25>wRlkB|K6zx!8g
zw-?k}(G-T^c%(g^NkVe<(P!*7D=sgts8(A(|J^@iZeO8@h<PCUp6>CE_m3ysYKxGP
zH{ZUdJGC^Acl6sGbyZSt8k(&`t_z;JGq!Mi_3gL6`AwmyYRal)7<!Bn9J`LB1g@!3
zMG0i;vcy%6r^h3spQ)RM!j&jj;BzL$k@xT4p-h1*N@gV0W=G=`wya^C7nXTZ^7hML
z<1k!({slrN_V2%<(1ON9HdR9mftV%6mTU`yURQWoktRu=!N>x4A;?N#D$A<gu(|yd
zX)BBjRJL3~<h5h9IdgdXEgH${>V~8xLMe7PhGCp|_wYb_JhIYY3eC8`r=J2jsU;iP
zIr=bDtv1X)Fr|bF%NyGj1#6*}1gCMLdwQg_3hzC+>+vY0UQ{^axu>#@i_0C|G?1jk
z)+?l~@Uv&`JCb!6Q(!_xnlq}}ptYj&XDBUM`DH*hM|4r38b_RFh!Lc~77kKoo+gqt
z1SyE)gh&|~5~nxsh;vV|-k|F>K?<@CkVkATm~2JfT!Xcgp@-uGgG&@LVf{op9*JQI
z@R2E!gQO@b=8%}rd#3XNp_Y^0dcDHZpkg8qBf0GnSbTrR_veLNNxk&xW5S0>j)B9|
zp1P?itYckQOg=G(h}Md#s&K|I4*@j~^ruIPs-bDt2xA~;`nhKg)6x?UGpBd&=-xk4
zZnsRz;Lm4BkA&`k6p6m;=>|{vct?G4g@_|@Y>~D>R@+4sBu*qGbDRici<-kS1F9D6
z#?bdr6qss7mYV7Go@p4+&MpzMrd+;ORv6<5K2kZD$AKs{URv;h($xzKCnu_^KqgO)
z0ki<+sCSnO;LkW-H3jSK8b6L?9|+yRsu0YTAbF1*9FwuwmsiNckxLQTG!+O<dwL|M
z#H*VNt}Zqds~THq{@4G@|D1pJ-~KcH_)q?vKl%DC^EfdLBhF}?UJxMv!!KSVSb%LQ
z<?mXe;sXUW!}ApShug$+;qXCa^aIde{(k74NY9V8R1p98cW#-64nK~F^#+M%_5)pb
z!rCKi5piDda=YTKAFySKh)a0WJPhbG(WT6*-~Mf??I+~^jzUX}RjgMVVvHzj5NL)e
z^7`Sx%@;r8cR%`^r(ge)*gkUc@|x=E1?%l6jD1IYddKnZ4f;7}#ahWYv`8{K4LrU5
zns?6_j?3LkX76!Li75&m508{u<7CFjh^T5pj36Xqe<BD=vA*Eq_QjGe7(B=Q5mN}B
z4)^qRO}D!MU2uFn(Og^umX|l5{N^`>u{$#t22<8dVW!w#(OkY<OgCZ481DL>vT&?-
z8?<p;TwL>Te@}b9|1K9n$rDj)o`hr=W;V8A$|dUR0;5M{^z8Q!oDN5BKmK&7PRf#y
z9j-3urw7Dvrr+-=ss^*#V9S#7qfe3bB{?VtKN9ByqTLgp-m+5y<wwT1-=K?vDUX<!
zpW&(<tIG{zC#mX^<Kv#JC8S99Gu<@NPcw5$=(<Lwv?R=B3ARF-lDyqfnu6EA`V}Ic
zF;-)oMH@q!2C@&7+EEu3C`BD-M(+_ub3C==oLH|n%*rr_9-|ZOFrkWi$qtr^5N2YI
zVD)o*A(Ep(TwWk|FIc&fdb=o|lJ6G>_|!9WEd?2^6n+{AQ%~76C}|ka2S8BOH3FFD
z9w{VJOQhWpQ$h_Vl*m{F7>6xZ<Rq9*2aGdl4S`HkRKye+Gt8&^=gbX@Hp?Hp6B<&a
z?M}p;$PvEy$xl(n;iDpl8Ro#;9_SDE&wWsdDhx&4fG!yO9%A5dI0K5Z*&wPlMf8lP
zH}v}nZ44$%bi<kcxaZvUD7&I6b_*|NoUvBnb3zpb+B%4d<LQ78Go_HMHXA?@gy4KS
z@cye`Q`ZeKO5Pm5#kys#;i{6&WkEbWVYEf0h?0UN!ITY34yZ5_Qil-|;S_OhQ5J-D
zNEsN@NKP4pCHR)m4+ve*I8AxAVK^VjS>lQY-PClaGf^pQPIQm=#L!c%*C>_HDUhQi
zNlSD4F(Z(NfwnzTxedlD`k^O_2qF@EKoiLQiRmuVXvb=ONjtQN+EMIw_`ZYjKxH&4
zNW^3bW6$sY<WnxTYyM~dSaR;p^j*t*J}*I2Lh+COla~I}5qzLMpOHvje0<Ht^@jU5
zckJ(%*5Z0?*=;rq-AIm!b>&EDCQLI@DI$q}m?(_JPcvvqWgYX}Q`L^b8jSDRzk6hT
zamnN*{T#S^_!?KQxwyDO8O7bhJr}DrpMLZax~j-oF?MH$r`NRmxA-9G54WW78Zq?<
zB)jz**$(JoWU>OP3M(vfI3rK*sjsgIjbTWZsw@!m$Z&kiX7`e+{s|Y`4eww70+|)-
zo1Y*^U^M5$o{P&Jho>WzGZ<qKuE4A=Nyec}q-{^s?S@l3V2mKlJxXZmm%qcY*fZb1
zM;A5gLNiSx4~HX0Y95b!-aQ_9fuULn9uJQUb4&06AO03XcG{3ReDjvei%XQ0TwUE@
zivoe5Jsv2^l4Q*i=a@4|D5{G~s+^gg1Zf)ZeUA|WkrRFzx!PUQY_2G8UlOHQ@(+aK
zbQmb>4aMb#89|6M#dd?OG%?Kd=Z=xgloNI1n4Kf^J!L!5KfEWziF#FIwdI@=rY>=1
zg{f9c$cC55{><rkKqwGe5$6CtaXcQ$F%YJKx>+rO7(!DkiHw26JYbBZ5{B{6a%C!d
z-(m%v+ZIz=oJi32=o}C^lj)h;JyUxGWstT&DT6Lc{5a7LJ=NwCv4;M1pjmGgqsqA@
z#JG4Pf@izgLDw^mGxhe8vMlj)PYx3jNlX%-YNCpC;kmDyYt%d;LMFvT-wk+w!cPN*
zD;WC4z!g*E+?_DSQkMl~>5x(pq-A|`i(OwM$_kSrbxfEP$^DTyj?81X3}>}LYYAE{
zp*h3CjZsQrl*So{)JuB}rAR^|wZaI?cDp4vD?~~V1ARY$R!CzB^N3S0&5P5$ytu|}
zFDS|q=7H7j3N03-HrkK@TdyhIj%Zg*%?^`h!s(827*R^#lq30!(VC0R6(KI!8#<*$
zHDfJB(@^a$=(?V|+3?ZzB{@e5V<}ugWi)NqGm3tBt`HcR(bgfHA>@cDD(dwHHBOuk
z11WmOaio%hRkK1{gRySO9SNQsN9J~>u#(%WEy_5e)C@5rWyV@fKaSY-3THGaK};D*
zK%o&Lqf$WSj43tf6-$bCq^L`jlsKUn=RnMYEDb}9oV!O10&NTh5|bi@k#s|#?a#<z
zT*M<&PpuWNRx3izq^iQc{1`1E7)?UZ9*+x#K1BBWJ;$d5&DDn0<r3Z66lYF-OHtND
z9OcCYkNZ0=uda|HB86Z&4cvYEo;R-_S>N7pckbBg%(ki)uZYyt#ZpE7>dl^y^crsx
z&F+@V@)lRstXCVVrePR*4)^!WhZfn5T)g^(j3o9gpa0G){_M|w$-n#O|D5e*jY0Co
z@B9o=m9SQv@9r^aV*lojdV7J}thj&u4SxT`*q*7&ig+A2v=0k;_tTLv1>U{CN94ry
z<r)#r%kWtk>Uu@9*`Q28-yi97Kp00mc9;~o+rQ`K)rMdG@)vY-q^b*?fvc+*tgl{i
z`SMfd?!;Gr{wLhOeWY}X+ZWfkdW#-BFY1PiiycB_cB>`EHwBOU>x>}(Q#XavdEokD
zOW*Yrw&1WoQ02HRJW`UgTq5Q|Ld@{=^vJrpq+V|r?TjuPCf6<uA}J`#lHg}f_iuRn
z?XPe}No@;yk#H9!bZ~t0c1cLvZV)lh4*{VhZnNcfr<r}kWzBFJm>Iagd&hKqB8G{J
zn;W(lH^c;PwV_<E(an14-+2h(@LayTCC*W5CNfGHa$5SeF(h8Te8tK~#=G|@BXOl6
zDM1LH@Ju00QqnXfRaxQ3nbX~#Ie4^KG{mNE(AE)23;#$<q}Hr9JA5@0kyO`TFsldp
zE)k0@MO8A!2d3$fqIQTOaXQ{ZhGxA*lIR{@FWetnGYiY48e*{~Z!QVCL`sD(E2iUH
z^l(72<U4C^uu5R8LnyUypJHUaUN3nVb5EQzqHHLeE#>7)D7GZ$P$HpnLQXx?++mC%
z2TgjGAkOoI5NhGkNQH3)<U~=H*rFh%jM?l!D}0DcRgn|U8IrLiKSPu_mf&(zlz1N?
z&y-HF+T5bnm&_PQGuoASKNE98+Xa4?d|(be5zXLXXh(FK$m4{w`Wa>iIRsM9C~Z+v
zASCqdi4Z)cR%BIj@zGCcrU9u0yUQz#Q^**J!%Q3|VwgZ{Bnm(DNSQDqB8(#_MfNjt
z=qO!5bOx0ctU>V8;!^!!4afo|fs`>uGR-3o_XAZ^Q#ETuKa%>6DMor9DVrK$EzE(b
z@9`m08_-gc!-N<{RB4cUL5Jj+7OlFH_?ST(l(LwjW*Ry~oG9z{!f*-`^WmP1Kvxa6
ztQMZovrIM0ge(=MFnAx3N;Btx4+*U;ISUHs@Y3_@(@*%~r$5EkC9{B=7r#XuXHNS&
zQoUw%dCjToNikq@WICM?zQ=uB5%r4qV<r|G{Bh<mP3XchjU%D=m|D`<6|Pd8`^Xo+
z^NRNm2ZSxS`RJNz(-5UW6*UD0GxyBXnc;9pSrk~UI3D)AxZKdxC6D8b2!Z+7A+1Ii
zj%HQkd$_u~!IX|_yFylp>?d*|x!*t0&Ymm<zVEny`wivhiWh(M9}%SD?(j(e)qB$9
zkutO0*2}0x!ts12CAhA(OzptbcD!6Q%ui4J$<Kd{tv7t}(=P~FF606484}b*&BvdA
z!TH|u`1Tu|QM|i<z*QweYaXAT&~j-Ji!2uwm;4V~?*C?SoWn@ydtB*IRxE6rX(R)#
zs>m6}Ij}#RS=Spr{o*5R)i8(190KFyx!hioWx}6M^rs_J2!t4Dj3%}x9>)j9i;BEn
z;Z(*B<C3F$JYcm%Dn%hQ^DyDVnPD1G*EhJLK$8jEn&SG3*+&wIWF)2AA*A5_!;!)Z
z%Az68nR9#QbUM<u9j>a_>@Mi1nKaGpRt?%K%G#nelv0u_$U~reIua06n-!{%_%O0s
zubHNq*WY|Yo+DwJ5HeA18&>NLtJ{w;^_u<DBdc-G93xRkgc8WC5eW8ge#znCk+suw
zci)i9r7X>x4OLNa=D^uc2r`Gm1Gew5N@9wVn4r&rU=>O!gf<kVLns5!k&jX;s-nhd
z1sSGc$wOX@N&;IHC@D}{Va=Lh=n2z=Z9KU@ph%cdE%=2vGmU4$cm@%1MM)+RQdmg0
zB4J&HFqV`&L)R{gwsBZ(7gmv!Xf1GhiPy{-QkWO)xA#aTamt{Tq`A7mpAY2zh|C_t
zj8dSLBE^h#%Y4Wfx2U#4U>-VBl%#lO7-x);7-z7`VO_yyyT%Va{V)(kLK#CQQ8YD*
z#5g6Y?FFvhLKq?Ttaq2_#-eg&cs#HV4-}0;I*p0~iNsk$PM|WZRt;&MIX%2%a+Wx@
zq<O-WHLhHf#B$0{lV7+;LZDrN0s@NQCq#hQ5A;L-Y<M&juA(YdNU50O$maSLT0!uE
zAOtA{MDUEqo|HYNa8%_Qtu#J)M9NrYkVd18LT8IlCBbL1f??>Hx|XtDQ573Bk(d*H
znvqT*Rbrk;w6<JaT`^1(?P1UC1GO>>w%r(ZR~O4+C?t%th@8=@nnG!=E^1PH=6wH#
zr}t0Xy!wRA>I$zmbnVw@XZg?n;osvQ{HK2pDJ&^wPI2b`%^MVv)w*6#?tRY}S`tD+
z#TlJOe)*?AC$>*qUfyDw4a!!O+VJ#rAf|~vfWEB2u6gs=vfVaRE6Hws!PvGuzIn^8
zX^;uFn-#e%>HWlbJo5D)|0$Q(SLo8<ng-bi)~gK{*H^s%=4<*a@VgrrBWop*uHoh%
z`~i9DIKBQgD<=^lG4}&bDDL-r&WAszC`+7G$hx4i7Aq6)zWEg&y}Y8ECnSPy^z2@I
zj8Bqr7&+|s#FSXCFEMsxoIL$BadrC&Yu%EKA{A9x@OXdso8J_afU>X%EeT|Hn=K;2
zIF6jpJxUv@X3fXzEnV>Bx}?~xIGv8HF4uI21CRR?sxXYFo;(GvKYE3?ikLHEn3?V$
zu-hBFG<4H|C<=_L7_;CwP0amBSqTcKDGG%v6<TLB0Y3*;7ds|fv(lDuJ~0dj@Du0#
zfo>Ss&Y7{8AxDnKfj9|bkeu$HXdm`CYZ-GU%>%a=8)_|y?H+3k!74&fTwL9N(u`?d
zLOVj_@cy0%_!PPP<OM}tF+M&Kr;d~Z^VlJ^;pS?~eD-wx#1t|iCZt-Zjzy{Y<g?Fk
zR&zc+VdRpS<%fay_j~*_6XTL~U~P#mi-iLp6EZEB3wat4hkK$iNaN71B&UpHrmjm=
znwa{YB$k0*ObLOYD9Yso=x2l!Bx%4DC}W6I&;H%7kjl`M8wytviOll|WR#W|?U<&4
zFnfk^0HINC;RyL@Am@OTl4%?$N{Ln)QX+W|IT2zY#*DTSR~C#Z6LTWw%rsBLaYm(u
zG^+$CKZ2i_Cr^xtqHa)1l5=F97miaHd&;8ZV!OfF9VrXOv<%v%l1qO*c&2GUXoJp)
z6e1TF+a;S>1gfe*x{BZje0w5jL0Mko3Wwfp2`Mv;(;_F$iDFe?3(Gi9%wc9-NpxM}
z!{TPI-G&sFwv8!DY*l|(T`92)KYfBEB)Vv@Wkd86`}3LTC(62_Zg#Z&!0G-DTR4Q0
zM59s4lC?o<N0MRD=!`8g3XF1#Ph^-E6<%4Ag~G>37|ze;FF_0*9}=!!xLHyfveJy>
zgjRZ)0nHHGGchD+G}9b0%JK2-E9R7d%&Ihy5-~5arwEDk6PKF?=_JNl`lR@`|Ki`1
zQbxFn(|lt4;xmkMY#(8KIuK7S8bLAx*+@Qm@rsYHceKYN$V~K}=EW;yjEfmoNxt~>
zBR>E16+=Iel>{c<fAgBhyFHzs_}L%)9;L|8AK7eb<n0bWkHo2?JI<_bZjsGT2)!q1
z&E5xARx!lP#qGx^QS$okJvT3JPy*&*=Ec<urokg*%}-w3P^=p;fxG)}*==^Hhqt)8
z;`QmkpZwxaX>37ZON4}vzxa%H82ReUFFBo${Lvr%A&|Lw^@>0I`~NB3II?OMrQ|sF
zobTTu%L4q!*!7SF{Tx^|4N}ki*<buSjIH_jqmMDlAj*oABC2eD^P56_b;W5ISm`xa
zyBmzD2=k0G1<J|Am@$r2WkoG5Of%Et0UJHOZyD|%sb73V&>62K&1QqyZiqy3d&c{o
zvQ*4zCit1*+|y2;Y9**Q7f4$pWuO~J;_PYG6(U8>4+pd{Og^)g4lzHXg`mueK|4;}
z5z@p(?a*=N_|&o5thxNjPl+*d_w_GPokXRGAX6!e0jATD(|J!yk=1rfv$-S|8&tE#
zBZ!ZWi;BSswyr|Ngi;3KEJ0+7%Nz1=L=y4sBQB5VRFPPOB6Es_IiOuZi~(Y14w>K=
zlaG)Pb7GbXsv7A^5SbtZu3n)`LG%;M10n}%8Hg&;Pcu>#K*VWD^a0|G4<qx`;X}f>
z;yL%i;6uQBzvw_D%PateaTS8Z6qe^D_?g4>hNf99?Ski4K@5_d0wF|%1SJ*SFcOa^
z)bqd|0!ka=G}Dd&o#f)XkP@XdIf|vnD>PbLgiRPBF=e%+_PzcU=hK;&FJE$bb-7S!
z+Y_h9N9GW@x_v=Jl5&0yaS9Zr!?*%lJ?Bpk9SkF3SO$K6nh+^+I_$|pprl4gMIi-Y
zn$d+Lgpoq7nEE4P*mK|BFR_Tt8mSFRWn!8MlSdefaE{>WMOr!>nZt-MC90?pLLo%P
zhn}nrktF~}3Q0<f%(5s8d`iTa7Mz2U$S6=!aC*3>pJrZs@)=Pqed09FXepUSn58Cb
zN3}LcW5{kzNC7z|=ClYrXQhzBpwbc;pJF1DmXyJwK&DJTEO-WOR|qniL=ubH(ff#$
z;E{L~YPBK?fe(pBI~0L=?3w0)X&A6YLzWt)6=`lcKb&xNgV^j)GP8d15%+KJ>F-b2
z(~J^^&8wFbvPM)ITPJe#L_hHG_FH0<Tx~CqDHDOZS}jmcAF-w)hM8&V&{E>65?im>
zY_=56^2x`a&?DKt`Uo|9#)oeR(*QAXvALkzAE@eUbg@B*k(7F53Sbq|t$FzJHPcSw
zw%1(OmyiR4NnCvL5#CBJUwp*X7bQ826jq_ClA>;SeDf{gVb9xl_mqWURXCh6*s{b_
z1@q9Orob03UZOwy45cmO?76!AnBDD5IG*^||MFk(TR;14ZeG1091r~JSAT)kKuWl(
zWVOAZt{cMP%<uf}A0SfX^{;<{HJbHDpW%v%s^0$QH-(@6-M`QM`#TB&GAH8k9^<x{
z-HspywOZKW{nQcKBd3QaVuEZ9tD<7=Cx&*9Z8mUuOL=*XYgRn{`p-$-i9GhmabP1Q
zBZ~|7Y5&Oa+i!XGlh1kelh4tn;PLAReA_WN!NuhTcD2D8n6o72MR75X1JfMHWOU5n
z9J62MA9Z0dB13mX2G4C%F&Tk$4z%T)ukRR!9^oQcXG~dPH&-+lS0rsnLgCM6tQ26y
zvtc#hoCV;xe?%yaau!=O7_FJl2inJbPHzW%Gz3@ENsF+KlBEaC)H56p)OLyGQ=1*a
zRWzHLdVRseySH@52lO-1Gp8AqJ;F-}9#SUfh{yxu(;n*_R@^X$MDi1d<DP0&B8<Wn
z4luyd!qUbqDQ_u}Q)28}gvyjvO(KIxs2oW_;A0?pPe==)RcVc}OW;RQtq^irums5?
zih`K{NfGlRA)MwB`DTxF1!8xFDhq@Zs1(uCpd|=vmx$56#ZM#A&*)V}{o<BWJJKBw
zXeseR<Avwk9m!cN!#pF=#v!u6WI-BQP>SF^IR~<j_^`}V`XL~+!D>NO)JSE}%^I_@
z3&MLI@DE>-$1^?(rZMsM>#wO+Ym~M~V;1Nj5!s!QV-M9pU@<C&(&FQyW=SD229J~q
zp%ghS)srBCd~V}C&sOGV$f4C0!fbHmnu-<q_&_&~q~w`>AjSr)WxLrh4I{^gd(t?P
zg}|*g2w4)HCFew$K+OZ)bYOm-^@I>fLXZ*IdW9`CMr-^uVys39(8-f>#HX-e%8Qyp
zJLac7aa!g$Axs1-iAaPhNZAp?%=vht-mZ~I#-uRyC4Gp5(}4G$_4O^n7K~$0Oe_X6
z1jq4&oF@9G2U7H`HXGKPORO^-&j*xLOkKxtIxv(q=ISNu%a>#!8TWT6-_oC+Fs|g?
z(~<KSkn@NiJ9Ht*%3v}a4<~e8BT7f1u3#RR9?rbDz93r1+t*JBAF1q_>q?+pqBNOq
z=m}HD94DI14b|oaLa(sG;OZ^XW#;``HYQPAulN^V|C-tel$N~w=@+CaGMxuz9~q~_
z`lFxn;`86);phL3w}0}laTgb8t@!ny|Jfptah7SAm``Vd53F8%%=vWUaD2^%K)#P$
zp4LR)QR+x?4XcYS=lAb8p6<|{#*cpsl+?rHp6i<z6jsxp9%xSkMR#QEkDwL9;fTnI
zBr~>H2%^VR%j)Wa&Gj`RLr{`22DH)S)1Ifhw@eT3aSG;X;(R>QY_GWa=o4Ok_9?G_
zwL)h_6rh#i`r;ZFAq;2AvP6)Xyr-x(=$KJ4p{os1StdQ8twH6?VZVQFd99F25_95s
z*t4oiwpTmOon6w3l;L`a+<m!+y28|s)#Vj#v!jg~Aq_D_rT|ljoZ22gbgY`1No$gL
z9<)nE8au2jvDPv6J?H%+{qca$8n?N`Y<5c!Nc1#C!D_RkkOnDW9y|6=k6gWYNo6Fy
zI}!UMOf4ySPTiR_O_YVf8jUs!6fI7mU5!SejAiso{9lX-<J_`1`9ur>sWoHQVzk0K
zyCmrONFXAtL|cn03_61EI<g;0G8053!Ls3JkV=tK!cWt3(sm97&+<$~X*7aN$`Lac
z)PThxT@B75lt$!vks>~~v!qy<L?HymaX?oqs`VP}Din^s8wok1s+u|%7^4Z#D2HSW
z-g~T2ocjUacWgp@HdSR}ii?9hr3KY3^`fFdu-k5yENJUUNidBA-SNOM&7>&6lwd4_
z7igB#w<~LOsp-!LLe9jL@WTu-A&Cs(L`Gtt5#mY;w3c|kEVemA4soe|VnEIUqYS~%
zgi)ZZ#cE4^v1Yw3kYP#J^FD(xB%ScjysY!_#NE4hy!!N0a#1bDL}QRZm<HyUAhrvh
zMhGIy48s^r8VCA5B63>flPTioi4bN4kpLv)kkX<Y=yG869wQdfML%VfpUEWLZiBnJ
zAV={X=A$qb7=_A_wriQE5s@S!Mn1lILCP5)L8e4il?>kFoki%9Aw{gvqzpbMj8F{0
z1K@`d^LS56V69<ppODd`v?Zm)YO}#rYu<nJ8e6Y9O&(<hO|xP}M%U$H{F4$@Ntj07
zzIlgB3e#t#Dmcc>?&T-AszzzcFnQ!W(7wM%jA5y^=9W_w2<uom#p9p;8_JEw?6zE8
zT%pe6;z*w*vZ|O|O~`^!Pn5MI9v<0lDu(MVtBVVYdd>OEueg8rj_KS|y)q2P1MlwN
zQZ-wyuWyjk78ys@#fHvLjNOPfmUr*&_`SdTcaWQb@$NvbDmJ$-fAgDykdnHoAt$sg
zDc9i7C&DyQW=EC~{7hX}L}PGG&DNA$zkErx+Yv&fk&1Di5hAm{d&BAeJyr@-RbiCj
z>iU)t{2~=aQaVMLBBV^A73brDX`09(62^|06UtTSqTzUV&v1C6DjiBee0DoV?{S4g
zNQKA=s}0^KwpB?=p5#YV4&>l5%JA~#HQQ1%o%hJ9pbHSsJ-e4Lp3xPa_IPH$U*K;_
z$i-BZ1@?JZG)^;N9vR08Q<OBT6>C>eUtBRviMRI;=q$Lsy+H<n5ulW0eS1SyIfD0m
z^W`_3Pbb=T83>1IAo&TS1bGUCQ_GZU<fZ{M);cOxAcRC4i*^oMlq9KX50N=|#(trz
zN@=kxgAfr#W;k_B?Fn0!_!yBfAyZ`VJ+3UlS;n|vBt*(6V?aQB2A*jlh)R*-qSX&#
zX+v4F47wLDOUBp*g6A$@VyYDZx!@6|p0PdR+an@JQsh}N6R3&;pBMjt76MZ^{AR%o
z2pM4Zq<LCeYdIj4!zxE14W_D=1(38DeY8PKMTW)UtOS>rmxP#zQ_tMDM5j?=LW%;b
zCAqYaO6J)kwM8jK@Jov=B|%OCsWgo(7fG_!6sv;4k9gAJ9SE6Wp2$*QR!dB+_tSUi
z>zp$=J%_;0<D%S|J-!&06v474OG6?qcnLH~iRJJWT2AMV93pj9;-`_ZYuOYHWu@uc
z4j-2`j7|&DIHXKWo+K1WKnlxzN2>);7eBCy^pde2Qy|VB5i^>I)Q;`$l6jt&d5B!*
z5uxu$IZ+t1D6+JMuw)4cr4VEaTM*`%c^p9tq_d1+UMh#(1=1Miw!<0?o23hp#t9=0
z>)i(|O~RE8#u%hjC<$rm=-LUDJVo6Ql4R@$l8^LLqMJOc-6gtoY&Q*Jn&{e={&=RW
z6aBd(`H`FgT1lpqm{O)6J8nPw4A+$Sejw<`_3br>90~Obqa;i%(_yBc1Ka9?Kx8=W
znVv>?^F;INx8SDc^mvc$dz#7-HB>B7oaZ;crYKg7Y2fOk8)RDMV^YFqRg?XU@6HT!
zpxA8q{CECfnbFN1-BZVD95Fsn?`{Z1$?dJdqtL=}`_UCoZOg;y;Wxi26qRLY&&+*K
ze>f7Rkz!Y}sVdrYOR-&(LWy$*?P@MAud%km6a`r>+VwmKj$KO(Gu9fm7h6P*gp?@C
z20u--hZBDEkQNH3)EbqceSAlKamnHIh(DjWzT9G;88Bl%K$<Z+(bSqvT`~0oZF^!G
zM}$zA41VaA)=`Xz;2F**@;EF^nk<oGCTRsSvAMj43_eG)RE*t;aU7Yao;VL&u2<C8
z7i4LfyeDS?N;1zgW0(NdKq|kmLQ>VmqPTGmTNG%sqlyBv-T;DiT@m^g=8;;AY^s`L
zccSeEl(7)OInCIe(XJqu4Q08gb!1VX%W?@R?|O3dSS5)$phe{Lbifuh^=8X@wc?>4
z=%;}c0&*I0t3`>Ff=8spIM2*kBFhSqA)@H}851&Pp_gIs>`}@r3a11q1VYY;0Opz3
zlwv|@LyF-UingSvrL<5=rBp-|&Nz%Tq%b4JVoC~SNq0Jsb3$v)90NsJQW=L;is`WD
zJWMp3E$e2>Jhrsc6Mf$!ltjo(9D8DxxYa_IjD94X_b4S1Rzq4g@t8mgNntdBnSFa;
ze|o^5?x|M|V-OT&O;I?US{5&)TBM^&AW|g7z?=K`q?Ea+OXe^mMMAj(9VRk?YzwBP
zwqo)Va-5jcyciI3`VOhCwI-(}zkK?DpN^iPo$$jzRa6KiNkSluML$<XqFbUOQ}S4&
z*=^U%<HY><ny}YA91hrGNgop|5?rYb+Bu|gNMjLFpo<ct7v)fno|K*)#ZuywK#V=}
z(9u5L5z#1H68yw4jJU#4RTXA?LH0AoCm33i$Rq{2S|j9wGM|T@6awbCU0D|erBU=T
zA~T2>7gPw~s+zK_*yfD(OKP4gN<@wnrDGhPIbm6ll|mbhk;{8`@QLZ%V`{<e?JKS>
zuW42*20!7)fnk55>jn@Rqh=~&ICm}D>LoeQIxenm(WXF*6HjlxMaRJDd?IN{*=$&S
z`Z*yiGtfd9gc9^)N2dhUC!ZpNCkunVyu=GnqcY>+L{XKjn}T=0_zIb3PWSIISJ(9O
zNTkGJ5JvISpZzv>Z{M*hN_N{FPY;j0dmQ=W|Mp+9TGv!6qP*wv?R#_v;}hkzVO3Tr
znYp^T<=Z!J$i)Tq_To3cDSY!6KWEcyI6m%q{nb}&H!E(#i)RMDL>7)TX2LiV#z1vb
zAgozVK`|myLJC;bCDRZ|+8}WB=M$v-43Asd2kE(*7^jgrXN=T@ae1RUEhyJD(k!P=
zpAtC^tSXBswn&)~1VR{Tj|b)$sp^`|X1gd*#u1?eSuQVrh%+g8v=EFrqpU%zrSCh<
zOPSpsk4)zlDJ4EkNU`X_eTewf6H-P>L-c_VJqm%)7Hup-WU`+KF%o8lC|6uwUNH7O
zQ{SR;z*x;NkIZ-XJUt$n=83}93uSVS<awf4Ur=AZKw3qZ2a+w9wvZ`Mr9?ze$^r61
z4wX5R{mk+14(m#!lvL{tndNmLEl|VL{=nEj%OtCUqP{@55~UR7(Br!^)47FlL@Ahu
zp0PVGMlrE0tVv5;2_eCxMx<KoQo}HzwL(dSbsC{G<V<{chspwB3-)(+?B9Gvo_eaA
z9ex@~aZy^u7-?LAt=EL9=d^$1v_GL-fvOwg9MG<$xipl8LkK~4J`?8&kt5nU>P<r*
z2l6~E83E4=VC_n#p`+^u!ZeW6a$<h-`Yj<u4#yK8eewx6w=eJ+ge`GZgOrLKBR+US
zSi)KgTQF~5(Oli2$S_T)t|j`O!7oz5P35pUBeO?H#S|h^F0HqeGsarVs)V{C=JXwK
zPQ{2wnJ|wC0sSywstUW_5XJ!+0#+%sHu!nQ&okC)rWBZlo-vIW?GO%PloZ7lrOh(L
zRvINWLQ0T^IWCo-wIwkvFvplPAuBKiY4Z3yq12MYUawZvn+++6C4@#x{Lmrh8Lbo{
z&C5JWWX5?~=)4(Fj+_LcJJR<plQq;$gU%A83q)P=@Nfd3DM|%7Aw;Gyj(G}8_>xih
zFf#OKuobFos8$<fP9z`MIY(HR80V;(hUs{wKOQipLn=YEiuG>GsXH^y9zRc*RSilq
z1y5ldGG$6*I3D&$1?R3?l#)n7$`nn*X1hg1kLU-I?~z4GH;go9@l%ZR3|)^Hm-ks3
zXWsq#E3$-oU2@)!RO^Pr`AF89rdTuk1^45UU>qm14;)VibfKw=%)k5jpYw-*^oP9o
z<Rzpn!yI`3?k&n_(s555k39H^U`xcQW%2<T=imIM5Kc$Fdh>?2@7|-7;?>J*>Xb;`
z8H$2&5Q|fCe<TXY_R~+uS`g-dlWG|pjx8|+`r{rSJVFWxiQ#zU&9A?qY}QL)#I^yY
zD7QPb_dI?3p40m`xcY$?A6-#Yc6kB>2qjS2Gak=ODdEF}D=k8REJ}(Lh$O~&AWMU=
z3gZk?7pS>lnp(8X<d}#lpjg@<gCCjaCEZPlOq7y2XSC6{vcb=u<N1UU0fbsk6bMSa
z43v^_R8@^go}2<I29y+xenyf>VO~mlXOLPGOv%`f6t>2onEk*!jg03LRbh~(ppg=7
zmfUOW3WjsXa6aSb0c#}IS;&^d{yl#7#5iz&zo+j9c2`%t`0Nv^W=$CT=K-$AxPq>m
z$q_2oK(R(BM^1^daZ4$k7hHE?mtdl{KjVzW6~%J8%Pg%UWdN3phZIR!qSVr^NimV4
z2OkhQaCk~cTXEdqasTdX($M1OGljEQyV(ChJgc8HF$t6@xVS2C)r!eO&K6hIXaz<q
zbcV4VFw#>N1yXBl;TV)=^rz*GY)Z7PD2fuPBtv(iY&4~Hgkd0M&AIK+)?<z1aepLB
z!_CLn6!i|N1mifMY$Ej?WwTy@jVVyf13t{CI3wtBHj=d>crYTdate*09VTXBDeH#d
z16|)OI(uyvl~(i&(@fX3tg0HVBy$Sn5Qt$y3<=`AEKF66b4%Q;(Td*B^zDps4X#*I
zR$H<FXDUonEWk-67wIq=trx^x$csE!WJ!*Y{E{vvU^#^sHC8!Py<XZi!IM}#F2Y)H
z6=@n!N|BlpuM|VuVUi$W5IUiRq9_-BS{^5|i0E~}H1<qR)9ls+p_%6nTUYoob9y>4
zPd%HeSa2aCqH<(BpBKQQmJEKv)GMS?$Skq2RCCQ0%4q!1Gmkw+DU3*TexO{HFiEa9
z4St@<lV|F?CC{WP(8eI8Ad8GC9J8<3)g|Y)N9%H--_8ReuQ9fI)(VV>wBWTzZBa-n
z9~q}Jx-KcM*L3X(-Os3TBKnB4B|<CoDpJ}SKLkW!2ocJvVza$Kl93X$(TM55fAaeu
zv)O909{`CFj+^xk*VKeOBEtzej0m5|Mlg31zy6bd|C`?wY~lFq#SOM8SvMuyx?%?E
zvRqUcKH%qx`}-%ZKKX=r>glIPx^v5LZW+6d945v&aC3D@A~EDlef5If<t=yr^51f^
z*^!JOL_t|?C@*jD)5u@^{42hCbH~ayJib{5GMn2CFFv|l27_*cQii_W<B6Pxmd$#%
zoI12xdbo4IC`eL}5ag*}5CTCF(}a~7t?e?PJwK49nbWZ&Ww_XG2xg&|=B(JC_IMwO
zDboVG-3}oYInGN8nTV{k#2AlHndmKOOOgpqWFD8kxwR!CdBV^Vx(=x<wpwN~Wv)qa
zrY$VfIB`5aGXHAg>iQa7L9@Fi6^ZlX9g0Xu5>&lpE$E)m9zYxF+gsYVZ~5lyuZTkM
zdw=xzNHQ=To+wrg>+5ThaZDj2w1$|GIpC*>Y$PIq78deEe9ng`^J3IeRf)6)nI$2o
zrF!x+IcMg1rgAk>NFtdmGs1i3Zjmfzdt8(sR<V6`O`11L6_FB2DU#9%6n#4~hDTD0
zWK*$TttnR<rfFbG19|KhJh`hFghSLf=%T?~zd~qDNE5}?PZ{!C{P`Z^7X#Tm)ffb*
zO!70vC`@@twb|j?j!pgu$Q0HhrNOuoYYL+8X{=%Lo+$?knee9vj(4v)zA^O26GCKe
zKmQr#;$x((sLC_x{Epmq%rPRXg1XwE%91z+bUR^lCeAHBw8$*c0djYyn|o|waLN*V
zCJKYnj+8Sx%rHV$OO=|n`Ck62q1ml5ZXt(?9G6AkEvR;lLZ(OxBUwO_YPpZk1SFot
zt1^WEN#n36Eih@q`*AsRv;zu3(vrNHaCT9Br8%QyrnDtlF1a91KnMY;C1qVAq+qlA
z6ip%y3ns>-uy_XM35miL7UvUQ8DgB^*}I}tMhStki-J!2gr5dz74>F|as@ef7&>C_
z$<qj-#fMDk))Yz-!;A_uLVzhMvd#E$z*Y@eN{V_zVyPabu~=90@c!LW_2(n8Ka-~k
zefOGfW{Dl%E$~5CuPIhr*1HR`Hp~xikwedBv*P;YE%$ft8M0@d{8FW5Xv)=szF!xN
z-V<lfc}{G8@-u?>l)h*G<uAzE@!9YHKBn1lc-mu-SR~_o!U>5O6SAnNcO}}adHv<L
zY_~gB+snU&O<}!a`{D)J7Rcyvr9=7waztiDmXa(p`_qB@e{<mVH(wKSrtf?DbBiK#
zvDsj&ihiD0Uv9a&y~XWr@KfOWlTR^qg|r4^G*#6sy=CY4{C9t!X-a(c^S|I-n2}E!
z_3N4c_z!-{fAn{L54>O+CUjAtMMBMj!f57crr)=;ZHp}}&88uy%+OB+0hL(c`;PO&
z9jm6G*enC_d7LQ=N3~W6sZd%HvR_WGDPdiKaSEX{K1^t3FxoJuB^Fb%aC*|Xw2`cH
zgqR5g%4)Tw&q>WZ^i1a?^VlzhNST)q9FZ8~Oj%kC4Ov?9I3cuSjG5SW%O5@xr6lt(
zfDtGJS!+yH(8v|a8X74uO3-iCl-0sY>bn*~Vr|Q3N3ccOrA&`u0vQ;`9u)*KF9}|;
zKd@5Ur8le%IWIbQ=_5WYEu$zC#w-};44_~dd#tf&D;H&2Q^33!7SkN?c}6_z%C#;?
z&+&~hEG7LkPUsBUIEr$GEo-J};Pm*g_$a0UYb}KEd}=Mxn+DmGDC1C~Kvs^+y5juK
zG9Dk77C@FLxy(YewMZ$5F=Dh~tCs~qB9_#)aUh%?h;bmryu7B*!~8J@vU3!=#8`*2
z4U-=klp$M(wFO3c+A$K(9Zgx15~Qv}D+MtiHH_mx5|OU!AxEsygyb2!nb{}wT4M@_
zkrr(YVVdYhkMJI63zRFEd?2RGW_Q6HBv~%cDS0H#36aGTjXSmYu_dPjQY^Ul3?z>>
znxqs_I<Pe<WMY`n&vLUkM{-seXCbL2&!8+ZRfCcesUv<4BmvefOs5zVq=b->(6@xS
zLq?CtC0YrLSkz$wLn>E@LM&>xX<FpoT7c0AQ!jFF5fQ#$7-b0*o{%NV+ND|*5K;oa
zq?4uK8K-%vdd?k$B?m=G$<&R^@km)zivdzfrg>((f4{WEt_@KrOub@rwY0+G+%wOY
z2%(v#o|i?jw3bugY2V{T$?5LM{_%wL9VQM8A#&KCsj332HP$s`8FAJy3_YM2#)-*$
zY*8!#$)`tNR5dCD#?uL<B!!n8@9t@|!`6<=iyb+_I6hL91@t{8!0~v*6q?`u{1YCZ
z9yuJIe)F3`X=^G|QC^pfZHp0tq$JTtl906HOj(p%ee@~EckgKGnqpNiPjEl5U9GUz
zGAF?lCWKi|Tcuc>z?<7^WCm%}64D_uB9549#TQ?Ej)<PxX4<}|pJz<Y7;QN153DZ~
z<)&teiTQk{v=zk2!_y<fX+R)|GLX&zV=ZBhM3H$qcjTN=R%5KAunw}st~ONbhDs>r
zIlw%j(?nS;%Ajcq2(d_13%kJ4AS1jci%1B8oFa4UDOVchBq8{Pi9gJg+nQ>#;gmCJ
z81O=alB5*4zkj=&UaVQT`O09OSlS`VkP-N4<}}|UjUyv4u0Yv><R|<*EqS%sqZ1hI
z*j((mzPUz{X!|3>(6gyl2&IWJGENgngUS(^!!kTe1AYj^sVBBaRPfBh1S*i_;)eEJ
zhfJBAW?WH#P{?NW%)-j}Frg5HX+UO;sT@JVX1%6wJ91py%swZQ5O@U66-bfkyMT~_
zvT$T65K>dFHiR4*4|~EiEaCRbkrs|JMQMP9Y)?pCV5=HZSXv<nWKp5|j+8u*iGIR~
zw1DMU%y$cVB;);ZDiv}`Yn!J<mMerJASs#^tEyqWS}!r9Ns^Uf9C~7V!Z=3?Gchg|
zKu(GCG&2k%<#vtE1JO^kVJ5|lkqSRfgkeO;gwKH~MtrmgwPm&y^=e@gEdxwX6apic
zg)ECD^0sybMNu$K12GQ_qQLHMQKley&wO}K9y`LcKnzvALP~?m%c2(jVnmEW0tF#K
z?pvnVA>vG$C$!cCsX!^JW`(j9wp^h^CdEKjOY1NjL&}MiX0k|VrHElB`iK^S9OI&3
z^CL+}qO+)?psH4c;Y`R+h_rZ05Sl3DG9Sx<nC2x18p(7xF{g}B3-_vU7AYajOV+kE
z8X+5`(jWvedP0mCWzaGc<BZh;s*;%Uvgw9^R)Tq0(B(o1q+S$^A!Vd1s8%~xzC}e(
z+jqnyXs#}>%?{$SZgl(igmz-xtUzcE`vb?*nbmenQ8uiEq#tHnxgwJADRO+;Gy9p{
zh35QlPj@~N<{l#!$$u1)QbdOH5!xe_sVJNy<ive@pxLds{Mk?G`-$mrW<H&n4~O6U
zrm(u&VORCy!gS{OkG<gW;gPCbv8puA6%@sq+v|%Z*j-A-X=0kDMHx_(Z0ij_|Fb{A
zKRlwJj*P8eW(iu8<Klo-#xRXDIm{UEG1jtjncw>CmT4TRRvR+QDOT5x!5Tu!Xr;+=
zdGmyjmJ>|f5d92sK?%qRX<mS4UMcDqS11Mi|7rV=Wy!KD%N87)AEK(4nVGu-$cT7J
zUa>5ZlKTIzc&$o#6_F7DoSI#t<dDs)vDA^(%oIWN93pUYHPv(WUS!TOOC#=?t~K5Z
z*4hj+%eLVaSk)y1;a*LqB4#Y7EX%gC?>km04r5??II^!-;<mD_8*Y9Ow8G^=EtPe>
za=RX>Da%1ap_)KzjqP^BJI5GCY<JK$V%*84icxHuB-T(Vro#il2U<?V9LYITW8~wv
zuk8EA>G?B#a_D;$-|^uIGi7YpAnn|i8`I%H%?)KMno3J6+dZMI)yR4~Q@53%6=Q1{
zBhFuE47X(?Z#&~5$jjE+eMI0eT9a!+p#*o4G=l{*I*~H&J3@WB?;5l~^e)>C>#Uf^
zTE%Hg$~$yLiV17LI)|Aia#K{L8N!Hj7NwxXg^%}F#^uK2;}a$A0uw`_RK=?*8OcD7
zS2~Rx6|H8ff{dji7k*bt<diO8t!D^<;5xVKjg&+o78uYCaG-((9azta%b))h7rs$i
zW10h#HLUZ2cv~1Tv{uO1Gs6(rd!^Kl>Y!^T=fvfj*!zGUP7L80`}`gHIMKEn<@`!(
zpnebJ=?T{rVRUz{)y^0^r7F_8P`bg6P+}x4H(I=rwkx?+^qx?DAcTQdcT_Fp)XDV&
z6W(dP5I=r}ctJM_C(?Gn_<-_(czj@(A2{YjQlgI1enLArSbOj2mZ&+3*P&aQKVf{J
z^g>N5Ep2ELtz6KhVREA7jb0;7E2?&M?QwNS)tzm<QPwloTI@J8j{&v~l{Zk8Ze&sK
zoS}krw}|Shc7>lF@PIZNT{1fEXr)mUP@Uid!Fzfwq}v(o9M%O=+=N)XZ;azWU*Cw!
zl{t7$<H-A7=w&Cr{e{QJNBqGt+ZW!noK4%ZNQ7=&(N;4*9;KVDaETS;2c8}tnRa1T
zUCw6|o#Sz$Hb*NNhh}dOKd!WW$BZMte)|Qh3|onWZi(l}s&CX*7{C9H&o7Ut-+ul3
zUkd$v5wk__)Y629a-0|sM{0+`d*;z|+e98%O2LP~VY;_v03A{;98QmH=U1S!ynSGl
z=k59-!5^&?rU5t4&>Gw8Uof?xdu0d#HB3yWXYc`Qq2-OZ?{qrjA<$<5h0!bX5Qr(F
zHOzLvY0aJ*dOT5bNBO`oPqY?swvu8&4}m>yynlP;`Lo6OftXj8_p>NVP&_?9W33{$
zN<&Z$hlB89YRjaYX()`+T(38Z|G;TWFLL4K9Eq2X4*@q0Xsg-wom9l<Vf}<29!VH3
z@4wQkgv<}~%>3}go+3GAx)bSKN?D9Q;}JDJ;NCv?`pZ{pf#XYM7)JKB5vC*a<9E0*
zvVQxKyxll|oY}S=?JVdE)_HoXC}%`2N<(|cX+Gcw=}byYXazaR=qERYsp4f>*PYtI
z`yfFkwMd0nE9ciY{5UWUCsZ%wdsvL>9n~x?@9gWsczodW@Q5{DlD)KO^OaRNYf=7@
z?RJB>Np?k$-?^m35F90FF8h`3*DGng5`tugo8H;aAEF*In%l=WXc+_I3RebmdSdwW
zJ=O<W$r$Hk30w-gH;4&cGg>)H+9i4wjZy{Y9LIU0O#}b^zyAwbTaM39jPoa~8=0QI
z!yYomShnj|mX9mr9!aXwjOQ>O*+W2gi%||0J8@fBudh68GkvzSUZr4ans9m{onPr4
zf)8k|X+?w&IT=cdEax{=`zzx((iChZqgAK0d$eVVv>e&icgbL`O=!ot$`~eZr1gq!
zMI@zhr}hhem}sqYe*KN((<jRFg3=D{C&HoQ>Gv$~B1F-WDXpQ6ldb@TH4YWTgi0@W
zoXSEu@3K3vcOOfYbD`CZTI8bBT@dBkdB%BWm}fwdN(H(Y(v(H(_P1ovrsQ4KO{0yL
z&!;LSXKvRs$_0j5_F(izt+Gt+y`i0?>e<25jlw#~0@TK^=ZaBH#zG}?`#7`4oi@#c
z>A>;$L@J4)H`E^a$AA1Mu%6g8hsP&gzWW0)RmNe&8rZfg>-#HdTbZT-J9yG96T*bU
zGP;2d6z7>94)E~6^l&0BXUeiti=lV0Eiu}Om@1_9_rDZs>nN*a5$O#EN2g1^@4B#G
zcFIj*ni6_tDQNHMNtTQAG*ePziwno;fN7c6uU~n8`<ajb<G=B5zy6FjhR=^reD}v6
z_>cefzjAtfWV<btv`Zj`AF#(G^Y{N7W*k^Qz7o$D)^Bg5QfXfV=H=%D+j6CM%{+|c
zQYg!f*c$fn6XT~(R6i0-puV5^>Hqr|YT21Tf5Or*baaQZN5VK#vnKdKJQXF=T9pHE
zxnje_!^0zVC?#RECgs9yS#YH?&j*J2Kw5W!e_7ArFkqF;-K%bB9Mcx*wPCHp9gess
zp}o4>fxNBw(PM@KxFBFS-{{7J38bcF6!-WEJ3r!I-^t})IKO^n-98w{1E0SCBR0%5
z9Vk7~nmokk>&h@qIPcivPHhDr20?6l&-naIO^L&Fz})v1@ur|wYRaG$LpU)WPXr&h
zUM~0ft;RY{Dus#ytuamEm6JuEg4-rzm6{vdvdHDGwd72{d}5prv<|gokutVI%~dM&
zX{OYK(uUG1Mq99;(1b9ds-nh)o_2Ccs9HI{y|JAy9FHfa<B?nuG)igY%ZkyC(`?CR
zq9UfPQ5%BSC=%)N#@n}7YJ>6V1?xPO!d@$5)ofd2>7Z)k`O_mba845Uwr+8cFX&(>
zwKF_5NSXcpjW)=I7gNMPea32oTUXYfekPv3a&#Sg^X%Nv#xe{81x3;@3?n6HQQM*J
z>ZQtF&g8s-ub>Ulh$lkzjIn`YoOmTlqUT7>JA16us_=uOQYkr!Jlsn66icF3goe8<
zP&>w0uBdx@-~nqaHE$HFamG?}B5s?Ey|faDW85fZMSF)a7Muq?V3k4@kzn>JWB_WV
z?l+OOmX0zGr47?KV~3Hg0cOBxO>I>$DR-1e&9VpVy<<A0)-ej8v9CM57o6&(?Z(IZ
zH;ymg%bw?6i1gkiIj#%%Q(KF5K>&Y#AQ{E{@Iu}esEt82iZh)4@CS@mbOp&8{BcGv
z8wm)0BISay1INcl>6m~p&ywkZqSVHIxpDgta>KrD1V6IfR@(Ou45M(4T4~hS`26_`
z*RQXX+YM_ihr<)wz6(V+%cyLcr@#ND@bn-5E5kHk_l-{Dpbe?4l<h`a3e-Z%jdfjV
zbwf2xSD$It(5*o^Loq<@@?}M_$XnGYHKLj!WBJ2>`M>ag`+xtxc>eASq>OO`%4&>C
zT>kveY~Oy9*<o$sL)aE-+8Cyp<#waBN=pm0jzzIvu591Fv0ttv=lSl-KN9>W!Zcuw
zF#G3{Nb5~FHaCg(JIo{deqs4_<M8weH$P%kW(W=wCeG&zd5>&+LeZF}2}1+b&=wvZ
zXRID@dVr=;VPK`+Nt6aVSZa#2n7LiA&<%DN#o>6paBv4u8s!Y5Ju(K#>Mgm@P}I^n
z935^9gfy|=&g{Sb6>WZ@ethFFjkM~(0Cr&d`~%fm+IFQ~-bl+0+Y34b%)<+|X2!U)
zmq=3`qs2+vx?(sz5sXFGM5+lhjkJ_OYldmSO_LlPs?&|Qe0!J0ZOxfxJj25yy(^Yf
z*w>xB?Zh>)FDqSX;Y2A%>Ee7=y@>8xE8?Chv6558Iz#OlZ52H>l+ogv))IYMQiQnE
zY6k;mny|fdeSPQaFRvJ<2(@FoMm_z2nWcm1zx;@A1>+Qb{h(^gzAcafeHhRvil6C2
zrbU?n>!xrGR4K%?u`RcIKu=(P`HrfrcoER#+R&=gt)f~@X%9qmp-&AS4}@vL#d7bu
zrOP@!oS401beguG(KI-GLY*G5Wu<)lLTiGgC~>29k*J=YKeL=Kl-QW2gUEVYMU_Hr
zhVA-J-Y?=;QIaLn-Hi8|)*HPR>XxZdycn%E#FkRY+d?UoVgzZT?j7)Ux!}CK1GZ+$
zzF?deY(^9LtO*09HF_!Jv=YymS`*e8+&BrQ+#JaJnY6B?O(e?0JQKAi=MAsHx|ysr
zEs4|{rKnm)$GTMVx}v<5E+bW1iIQ7U8qH|(zLKM;<eIaD!-4OPN8t=Pk5Y!-J1)F%
zd@)cPX<upgUB+}e(Tx%1Rn8)3Pf5I3PP`vAM`|lTqO2?44-7|<s2k<wUGIsS5?e~7
zrcqs7`DwYJi2TEU`XkyqlnJEjsD37^iF`|J*Bg&>VBJ^Z&wnL;`;9dhp1%JBYr0X=
z&g?7-F;V`{k&pgYF&8yrG8k2;S!RpuJNxy_dcAOXe!@K*xV^ulmxihWKRca15p7_t
zisAxIgLf7;ZuDCC<2>@|4~>7=3P1k%E0apJZ@;iB!~F6CkDq38-?_ejWseQrH6OqH
z%-7dfrfKB!%QO3}fpr{CkECto>$hJiIq~@PD7=-^6FpZBrw3`HYeK!Plq$=l&(p|C
z<>T^3x1PuEA0=OPyRxi1R%a9zr7XreQp(g)C7CTpN{rNSzz!qijd^ekvjHrZ^Tzvc
z?-+$B4YG6^Y2S%y$7&^RVbf85Ks80Xtz3TlNz@`Wa-1d^anwRjir$L2kUu)>S;ho$
zy^>SB!_X#P{`kGf@><1>2XycE*wjKwGA?S`pfcEXMR{mWVRpIHg7tW9B-N{Cnd7z&
zUcy;24L!_MXNh|x=R~W8v`5Lea5JS<)^%eT6|GgQ@m!W0*V~z%Bh)PXpkZV@9trvY
zIWajy=$*DF%y`7lM|vpKUT9rWwj^@Ox}dbayR4ObN3Bw~8*z(_rzf1Bc>ngnkAMA{
z92deo^8Wsn(lX=oGfpGR%;N#uWCWy@BHgRxSTj8*f-~~metxDkKpRrraax!^v1BoN
zwMI{gsv^cK^7Mdmo}MeQYtk*EP`Dr(e|I=BynKfpCfd3afBQ3uig%jl#|Kid4AwCO
z2NWn*OjHuDx~_QZ1+5VR7|&od&S<paj^l1pNEBJR*49|AS7<xVikM*t9%Tod_oNuP
zUC-pa%lJg@R(UR)rdp47N-jxc!LPcg!18t_cp2T+)`>u`lD=odh*q6g8ta{E&}(A5
zT`>9{9HlIF(v+r%&453!mx51?R<5*O<(btLy5v9i+V8m<O-4EvjMCy7-wIpap=O+y
zq(9XSDedHvF-i;Vz=_aAhk&|wwbodi^%!gEy|e00jw|F!Tm?6-6aoP&l603#CTDOy
z&`P1zOjicyJ=%D>*5oasYUbz+rdCpvg?C9i);UzmSlvbNkRwJ_^19H9V!UI~jdp~m
z1Ey6z-rgCkA=QX7mh;;;=HrpKZ(pgWBZD)5bdi7eOCjGr*fzy-{b0S_=%sL&W=7|@
z)=sJL^!$nN@Iv}{BVIl@91k3xJ~80gQ$q!Z(GF^1bb<3%#q$1@IT)Tle`XvlwpK2`
z{mA3XCrZ4~-@noK3-$I+%Qwbxz-Y@bd`F|B4EQOKQEclKmn-RVCJc_ao|(UV$Nc<6
zYME`{xxIbG%{zThjK+c07^fHve114Fgb{ip#hv5)z<hj?5kuFM?aJ%x8@a(W&BErl
zmgZ*c^uXv0?eY~J&&0PxH3qc9jh41W(pqRr=ACGfl#0SdF?f&l1NJUq-LDr~DR?*F
zd>}Tl|4W!krkBF9MJ~6U%k@IZk<;;jF_zQg1H(MhV<cX#G9SHtP)(!v26+`vg&mm2
z0Hv_qcF>I>c$!kgJwlVw&G~i~HHDur(~R?heUB2UsTIa(_H||3R=l$W?@>-uS5Y=C
z+lF0NK=bzP8}H{c$9dxE;Yb(<DvFQu2g+D<T2bqjw5-_UGsoe9ZU+<_rWCri_|r_&
zvV-WxP>llJDK*ksqoCMhV(d-u>gNyM-rkrd$8?gM>8=#pwsBn+;=1xM1?JI{wv1YL
zj5W;8+{X{X1}ZgU#))PWIqsCSQ}z|rE6NA<oMp723?J7kDZgWeH%<=^^wPLomis8c
zGd-Q~O7=a&;IYHRvL$*>ykFk%gJ+&bG37Zeq3^2F<5lMRMcmGnh&3K<l<XU}h1RMl
znNp-8Bg<i&XeE(z!cPOL88iYkRN5fw1yQCd)udRVlS{!jc~-kN*!e*a75gSx=B*Le
z4d)v<ZRE{Sw;Oris4<bY73Cbq=NE>0U{8g(ZE~ihmF43d-5M$-wAKuC%=pA`cU9N*
zMs1OnHc<e%g6<v=>2g!Bmqg74XLhvHl$=?YtNgh|mew^m!Zd<%O!ELOpqh9mT9jg{
zYea>oL~T@Srj&$tj^GSEEB3gv-4>MUIA<_QiU=P9+BkBS5f)z4o2RrcU_o^eEPQM9
zT3K%k#u#dfghA0%C0L6Mo}>$Q99g!Ny6^OyQKjRDLhC!)2GAaBEKkn|YT5{=XO_##
z!{J2EnZq1#gJxe6+8T;7fB#G2uYdj*2CK;XMq1@ZjWuGBmQ&YCSvHIf*h%D9hvz4X
zHY{sm8fLm`EX%^%*RQCu$<j_6(w;G6;UE9$pBTL3`f;K4#%vVVmRWCSg0p;ndBFrr
zE0r}RwtYu+#W+ulTCrWO+}>YF%LgqPUdGRi27)(y`QbZ$`pb{J{`@1y=VzWif5-F7
zCtAte&L1*mF_x((Vw4L*4Ksc|vFFP5;~TjoYRM=i>X=+R$IpMlAD=K`LdBKaw;zd@
zGir+zZ}HF1glXhj3fp#L8V1nffVN@y+kxYhg|6=AuSzu?V;$B_U@f^8SyFd+{W!BP
zI}OFMB(@ka&XBudy=^>~&eQRct$lD?zhYEJk%?UvqrEobwlQ_X@w*qQ)|B@v4uuJd
z)*IdinN4pSDT-vTmBfA)lFxo!Fh&#9D6yThCWL^u1|I^oHncU2<HY%Lp{1SWw(|b<
z6|BZN$21=p#(`~HS+6VUy3*pthXsGy_~G~oXDz+1wCkDl@s6^A@cBFJ^hlV#11%C#
zs~bJ<R2^ttG21{_o%8!|SQYv5-7|Kc`1~LKfl@@Xue3(>0;4C#jh}w{3Dq<&FP|h~
zjz;MX?*q;oQcmpGMf?Z~y0d^HRawGftVD>0iJ*4Az5ay%=$K_AmCX5iqj6Ws<uk*>
zkz7~GFO~V>g><>nTE=Kau$IAln$cKeNc&Fh74BG(-+4((+{D-ifT8vsxMHlO)<h}|
z?}SEM<3_C+P$;Dc-qV_7W$4~X#Yksq3`s$&omwhMRk=v5M~xF&Th{H2Y6Wj43*2;3
zPi=9d>>GL8P)&%u)I|O*i7L%`Kthi8Wnnpg2*%;<ox|nE!<P%g=RYx>o&b#rLi0`g
zhHjPKJA+afCmG__C?W-wH_}pBbHR@@R!wAAFiNqU-&rmnJUl*P22bk{oE*?GCQ%{j
zhBFE`NcUunrMF5>J2`KtJ12`_pzbSak2K}6&Y-$<W}|bmOQ78y{u-qfTTHNQC@bT{
zVHg=FxgaTxQZpe~dd)0b#A$^I1A{rzd*}LbW=|{Lb!IKZ+e)Ml9zPlC5I{Np`1pb<
zo$a=Oswf(6o<$5}{~bDok|N!BP7f#g;RtlxVIuE4));cy`S~yZiW?@vVd8jrpee()
ztz6&V=-!i3;_KHB;+~jB$K*8Q@qioWJ8wrZPa}R98HOXecQ}j;_B+(RQ}&%+!C1#B
z41x%_-YEMji!6lDIXykHuPaJvF5lj`Z9BtsV46>iqvHAVcZ?5@*g^aZzDn+Jj)`^s
zptOeb0ueLAi0TpNE#2RZG4~l`H;Q^Y)3rkP%65L|U;oelllb-<L2KrhXAW+ldy8`e
zW$g4jQo1RPcY@<kPTJsYyOHX~I8IE{LFlW9UOr4Sei$ix!q~uYH00RuxudN{S;OO(
zFKFu+=NV@;uj|TfThMgMwJ;thf_1c9xh)$s&GF$8rW5+`>Auwcz#15j2Wm_r3w4&1
zHhQjHFITobLa7`MGlTbx#}m#d^0waPrx{csRT(9nP~@7iC(FSP<X$-*kM~@Mj@F8C
z@Yvqj6|6Dh+=K~}K;v#7Lg1{KRtx6#hBX~?5K?KC^m@oEPzlokS2E|{e&XxbU$ENp
z^n4;59~h>Q6bq_KsLc=zqg5=IGu!P-71tWMWO~=A)`W+pqzWwccV<zcw8s4S%wS}c
zbo2vV8H{rbhXZ@g+}<x#>p4EX@c4s8IVT5l-JvI}b(G#%KVI=hqo#qe2bqU=S)BI{
z*JY>Gf}ajdha=D^S@t@y6v`HH#^Rg+3tKJX8R$?;<rX8`?M!fnc^s+5P)a86k={Gj
zIU2vK1l#?<wWOYirJ(P?L0To3E;L;lx{D&LYsGTA3gV??N|G^IX^j#K`c8vYN|;Og
zz7cmpj2o>;Ir9G7Z!8~Y{MR=gUOwT%h_x1HHA;JGxr%(bWt8w1=^FOyEAjQl7CZBQ
z_y@x0AFwK5`bxWf5P$uJ%WrSgTsS?Q1h11KIYv^>C~L?ia?%dxM{2E--K{#NS87_Q
zwM%i}41;$p*E2D$R40XyZo44twFYA#Cu;9#<C&Z#?!qKP8)8ayB?eUMJX$GANz_ts
z)<LZ-%R=g+1jBfyhk@bYh-sa2yU2d;dLypz`0|1o52RSw)(x*UMrj6X&^}O<LHB=$
zPGJ}WMr&@%hIgLhamM>V(6H^1pML#?>$0Lu;Q#S||KFKDy-<3SR505pw{PUwxV?QN
z=gPzInRy&gw&Na-#66N0aT<?@nN}83{J_S@{`yY7U13m^=CR5%91eJ6D8?{{fiR5h
zxe%H-74^77m^n-nhv#RS^UR04oqQBeKu%d^hkg>mrgm6o8NH_&gARdIHneIur?KNe
zq21a1kmJU;|MUM~m`@CcCvt1VEs~3&jfS8tALlD5&BN0R({MmJjnS5N52^Y6L6{y+
z47yRz8*Sfl;Q%PMZK0^lAxu({G{rp56f$~JJU^be+^$qQ&tE=seExzfneF|Rf|AOp
zG_JRmZY|Y9NtN6TujdQf{?C}{6Xu71V0t<+t&!;-2$G|`2w5wt>I|c$ThHOH4k<Mg
za}lXsmmr;vqLxN38C1tw!|`}v-G$J)#3Ugzz_zZWw4nfF4de3(d|-ROF+INEtfz00
zdcII~$330cZVR_>zcHRaFvoW~hFr2#k$Fc4>F{)p^xl}}6D~|xt+1w((}ppYU=7}M
zNefGb95bi+#Fi7wzEN`GI8B@$PPlt9fH9&BSr@bh?+2_hXcN%m-KWu^b?B|~^z_K*
zPfvL7n8t}(W#<t5K-Z4m<Ya55NdZKr1WVe5yQQ_kSWhjLt_<A=+{1SaFP~903fwR`
zSvGeihh&G=DpZmBj^j+257KGUp+@OMlk5dkO7}R<h94$S0_nq8ls44T<hMV}93P(~
zPr*7K9*^i=xm_>B80C;?8Cwm#M&iEEQFNsRg`OoU)mrg0?fXXA3)VTh6}Y0Roth%?
z+b?{3f0aeG3w-ydKXEucU~IvAhoY0>O4)belwDqFOU8^R+|x7m^LMnOFuS6xD-FxG
zMK0=$(F$#~MC{f=M`4sDmnz>I5>{19q4b8X8L}YfYZ1z9ZdLYxdB^oi@0QYb>W<*q
z-9Rgu9201ZbtAzCQtN=ZOJF~k#sNP)(zFp6Xzy&>j%pHY)p{qFj@E(PE8EAJrVXxl
zW}{fo7p~VUWs5xh$>Rs<;(iAxE|)9**uVyw`ky=Q|CcTWi-uIurSS6ni1x5=qT$5^
zbeQ?>^a)SHXiJJa-FAjBvD|j_6!1gkA5Ndy)*GK525~7rJkYczUT)-d;jiC*<9Lw#
zS{r6;>nyh$`?hc#9;h+VO~Yn`bxKmJRA)LJv8|J87B>6yGvncr$4_4flT@537jRB6
z4QZFvivr^uttsZi0i`;|N$W^&n4D3qfYvOz(2Qj~JTN)Ow%qs@|CRHPzv99fo*%G3
zd`3S$@bvkG!CKDmA6!1(F~iE2@4v%@nf2|A>cD8lS*bN8?h?z{A(uky3mr}G3T?y{
z99tqU7g9`AZBV__tLzt!ha*FZ==X1A3d{0AOgladWSt4~fU*I1IMH*4WkF*(-y*;L
z_x}&4w-5gDKmQZPSu7S;L2E792s9p^o-twINo(@H^YQkLZ{J=?G19igrkd<@j3T%|
z?;Wo@%^CKb=<CjMxzU~B^pF3A(OR}|zmQ&k!+DQ)kC-qrRLv(Bn4cb~DE5yV>H0=<
zp2<17f^}br%LlJ-Khbh#yDg&r(2C)3U^tvOK0IO%2j*!8Z5gM44Gwc(B)bl3-$<7m
zH70TeKOf2aE@*P=Kx<m<VjMIIs~lb_>@=ah#%U|dSegJFT_Bf2Y=znzqqR6=@LJK9
ztK<f23u6`YmvZ!mRO@wzp6Nw0o0W5P>q*K;u*~5z)epFr?_eD0NfVn!RmnH8Cdhoe
zMR7C7gdZnxj%E$h;Sn`H&}*VCAMERf(w6z@NHZ4W1HliZ7}2#d4Kq!>;p2v?V9mf7
zW||iH>3lj$u66C)-aaJ1<lzWT;)hbISSvXSy$Z{x#7JwEVH_p@yEj?ZcJMAh==i}i
znIKiP_YC7eFb&%`d>HAyvu!K2BxthS?b^|8B-fpE+d1AA>g|>N>rXOoC-LR;GqlRK
zt~9L~N7=)at{Lz+f23LYJ+<nzu4rf!4(x$Wq&G2C#(f7g!Ov)?D5^^#P#Z07D0aEL
zRHx)9xRkpPqUDTMn&aUp`PwC;Yop}^z03F*aAA~v6S7brl#;ZyufNFR`g9`u1BJ+P
zxv>B1Kl9}ef1)Wvubrk9cAUvwB;JmHhfYCT#l!r-`|CS@{p+tB9#0%Uf5&<|bGzO6
z;}1VDhd|pmbglgM*T3M#1Ad%wh!F(cm<|)C$AKDeSnqjwII&%?oWK4?O^M??lVT#q
zjMg`t5$3-vb|+G)80RoyW*Q(Tp>nQ!!gwbm7NA=zZU65hpVB}nj4{yr-?s57B|s6T
zNsKMrD?PMac*;^+rdG+))mkw>9?{bQ_i*HLJ@eZyZ;&<#{&+ah=Mj`<+Y>csULGF_
zTH}DhDWb}h4%2vG97c5Qlz5Sgp@aI}8EP!vic^=C#CfY7cJw%H*)AWf%Zf(Q(Z0oh
z!RXF#Jkc7AL*P;hs%j3$!1(-#pGL+QQK=BqN-P__-B{1R^6{(2Jv|W2194Y(e6<AH
zw_2H6L$`)iu(XECmDAyY;q)XPf|#iLLN4MK%2`vl$j7&DoWFgeX~hr!@gMo_<rBde
ze*Ac)YeQFtLdTSZ>4io^t3p+U6<K)G73vOGbI!BvJ0Gua+^!dTk&D$j%a$v4U%9@D
z)4m$b_~nJkAMjyDnfrpZW>TpVo@2Chm)df02Xen3%uUh40X5BZrGSX-8`>M(;4!@u
z*Ntj4Za&azrNtzf)vc0SV~LUA479@adPDn(moSs}jhwFXccKeTw<KD%7&Bm;ryEZ;
zmNq!d(}2|uKOOEKkqTW)08@$#v%}8=ElU!bo`fPC*AINpD5DsrBj|~}7y4yH-2>gd
zb=Z3cn>G<4SOue1=;H+1qO_%@m0T;{OTL9ShQkm@4-b@>Xr-|=S->k0tW7SYyknf^
zFdk7_F&_p#O;3^of+2(;Bd8o1yp;<AgRu@DJY6e#%NTbOU}EiPH$v}t7l>9AN%6d*
ze)%&~-C19MVZXjJPJ!o#iG9>ux0RF=>9$Em_!s~rV?sfViJ}yBU8I9ihORW7dj?0A
zz@{b#BFY+!H|SoeDY4u(;<{k8;`I0+87khPjUjCdrRF<gJfL)=x5{2JO<TH=VD>J!
z0`zo3`2nN2?`0~-<D+~g4c%yl!vlxIks6~!4Xir_%{WY07cg3ipPmabuKx~~!n(5;
ziPYQ}n4UhNhZ(ko5IpnXNVf}RUnI(~6t;C`dXO(?yMCZk#HL2<4Qk|gI<b8F%C=l_
zRJ4Lt65HMELTy}grEAUna6(%z3MFIET5itPD)W34>720~P7eZ{vkntHwKrNyl09G@
z7DX-v)jKgIYL!c-WYIE%qSOi~dC7Zci4pdQ_m&bDZXd6>ENEpD9A@xb-fq<Ef>8#%
z<1kppao~9jjMh@30KC%rf%8hj>PtoYL2@0lBa^|)Wu}dj2O@2V|Feq=bUvc2r{x0c
zf)4{$Ly9}gCYR^Yd8rGO=C<4fv=;)$(L>L)mdLKrDjZ)PIDPjS<2=(e5uBl9!BW&#
zQOcl=;kRFY=Bi-~lIL7&VIC*qwv&GQnbSPc_Q>1USCk((zKFSNd3zPRzLJh7#g+K&
zSBkT&mv?&00$t0NY$XSx?upBfKVc6G<_@dVy`fTsw9|6r^8UszKmEqKE{ww<*y&Vh
z%g&TXE_G##zhT3`AHBsGkAD^<_qyFk%Yti36i>q}nI5fUoyR+a-EXv9@q+{z)tW`=
z=8ULPb_sxjCNue9C`)8He@J$QQD79kHZ%>ZVH!vLFr$>__WsKI#~E!MPmhmGUQ|Ff
z43wNHIg(2u*&txO4wgjTHn!`#U?cK|Dg`rWP|)dU45Jr!Z(Jqoc)zT~w=2)z|B-$;
zkaKX!j_CjZve8LIK~$kh`6K__pj7$uYl_sgVXPK4%;AW2K@LZyXzRjYHHRPyo#Wxa
zJdN}Y+rCOyzbeKs5Qc%)5M94A4D-w|j2y<m=!NWRyvL3M-dN)0%znAhw5a`*HqsF*
zXedgG_~5BVq_%sEVlq<}JDfPG0eWG7|An%=5xf@FP^&0ym=6z3O4E;(+OlMcYlk0Z
zy0+wfL6wA2itToz_D0tZJ4-0mmc`W`#sgtGLhq<v*|&vtOI&U{)Q<O>B~{d58HNcI
zw4CTD@q}nYDUFz-e6Jd?1KM~(kPH@F03+EQtz@W8L>z-B<$TZgR%m4yhk=?4wSpf8
z{5X?yMq5KiXaX(uzyGE1?c;+K6VKoM!1KciohxZud6-8I&qqpalwRmL6OIRJRNS_N
z)yP)gTRD)F9JFzlL()0R?fgzo>8@(nC^^emq7=e7;loHOk`6W+%k?6rwNet}eBf}L
zF~QUB;TOk;z%(DwR$8W-Wy#YO)Y7S0Moi9GjMkJAX{w=2rz5Fi{k|B}+TdJZUoW(h
zFvb&a8}C2AGFprEoyUg*!kMV-mv>H|o@EDPM47Rq9bG%OZ6j2J2~weK>yECH8rAhc
z9X;B5tPu=%jte>N7~Qejl6yyIQIF&lNxc%Lks){{PBiVYeqzmuxLvWTqgzH9acRe|
zKS?F4I;Bdq-{JH3*kPczjzg2uMqF27{19{6IHBlxV^CVK6}dMVk!V<#4fF9cdLP)X
zI~C1vc*Gt*V?t-v9cQ8E$hO|t&KG{k|Bdl@#Hq^gFw<H?In6L1>7{ZzFZ}wKf1zp1
z>G_46t{7{?@~;&wMb_&Dr9~#WrUWqyM9>(9qM2SEc{m>MK481y*MIww^KV}nr-_^s
zw8Ci$gz1Db4s8vl%PhSai?$i^jdc5<*A1h^q^1ycNi9W`Txk3-V2ovn7jEl92Xt^)
z;~1tHqd-+r75$F?h`FL{z>X8RfvP}j#a=e7>g=~GAJ>)YM^u;@TBVmvFPYozj4}!v
zJO<BLE5SLctCEMH9o1R1Gq^I-x0QX{_;~wD$Q!*>M(^m|NQdO~U1U(0#tFPZwR=jN
zatzZ%>xzyziU+GvR*T!$YC;&%#xRwcX&hNDo}8uI_uey3C$yV6K7ZmcA23Zi%Bmvs
z!wEaccT!TO-d3=8b)NMS`&z|(mD7eZ;FQeub+35iX*jykJRS!2eI;&J)I8H#XYUGK
zL`emb<J`u8>j*s{NbYvrS16TMG>7AnWnE-Bu9Y0hTFWJ(6k1!X)qneJsth!jQA5Xi
zj~^%eIHRW#<sG^gMmJEKlB7Fp>A6swBCZQvcdVZAcA_am-gokHCRl~42};IVi^db2
z#b`ivS}CZm=uM-v>_}AC)F$x&MJWgqT1j}(-~Cb;pI^}DvlusO!G?gD0>h|52dwtY
zQ=r!fT9Z?jYPM8*5;cMw1hJc9B<8}hEJB%V3bd0r%eYJR?gQiTf%!1whKZQ+-6h+3
z|2VUrKX`n4U^<>K<0vtYInz<{V>=JM5#ufhZl%#kq-5S@Orw=(hUpDsVLvZ2pLGMd
zb%E~PL*%;;++xCx6G_80UO9jL%5a*5VKjNxn=Ik=wsQOW72^%gI^worl;-hpWIDb;
z2-Ia|T`rVZ1SIAH)&?wE7HG={HD1xGQjMb;NG0KvqNj~@S+H7jI35@Vsld9iC^WCX
z{7kP6KUjt#a6MlHNuVHCiHmf@1IL#yoWB2_YBVZU+P2G~)|*@scb-m;JJu+sVW!kf
zqm#Fd7&As|;a(kP$*3-!?j~Ao^ffYC!{OzTJ+7F+u$^yg%MCY99F7lIG9_o4vIIM$
z4~{uCK0a>z`s1(s%U}PD^H%=*K5#rvc&oYI7Ph@eD{l?nOMr}bjsnD(c>FZ*`0&W(
z^@DX?$TeZzz<hjUnnuFlP)dldF)rdzMS+u8N|c6tV_V))MlnqXnjKKa<FuBMmg*u$
zR)RG!!K17ZI%v!k6R^Px9PaL<F1fO0O)&$dYS#0D=@n07_60wVP!c)qxbc9SC))W!
zDjB0;Je_bMpsge2EPjzH7nT_ggyRWwI7rt~D|O%4KE87K`YZkX4w(BA+hD!LIzexs
zG~O8UUIFQnnv%VMw~i13(HK%(*`oLg=*W1nNH;r<BVH?V&RlO-YEz8fqP@Xu(7jUP
z%JTLb>-9{JpBN5Di8?KEu3$uRTYII~4m8P%jxt41S_{r(81de-?~&TIkX?<%xB+cP
zycPA>dOM4tLw9m&;_J``Z$;ADRJ+FlSBwo9KM>HYDa!Mo6HO_swIVs~jS^*-)Ro2!
zBPTa8JVAFOb~-YR2kCHYM3+s{**aj|T|T^{vmid)c*0CinDHolt=h14r=^I#OO!jE
z+!}V6a3N4qq_mE6fszXu#1%fya&8%sEEgpjiGLUKT3RE=(~)gos5)bZK)k-QB*S*;
zY^mUUz&X#iUw&ggJkTs$-`?@Tv#%>zMfQt;uhws8*6SJlsj=6L((w7aFB}dBa?Xsi
zJosWzv)Ei%ZVT)3j<tb*`j7ufoU`+cn-0XdgEkC)B<0L<y<!04G-yRHQn|F8B^TFu
za;!p6RIpz+a*8xAxbZ+&7C(%1qfvtTAvMijH97>2fB1nIG%7{33(QZ??D1EEG1#U!
z|MDB9RmQ`J4<l7+lB?{g;an%B!pHkNdoC1fFvn-6S<9t$eMfC)ysb<R57Me?$H&JT
zx623ny3$MI`yamNFwIyuLO0}P#nMR2Rg6`3#u$&)0p~~5JYj~Bc{oy+h|-!`3prPu
z(m+QmP0AUoHNgjb7$|AycD)gAtAx2*>82i!0Q*1$zek+0_`}G{_g}<#Q#x<&SHAxC
zO8GdGbj1Zl_nI&T@*3$~aL8S2_6)sAXwT?I^fYq%B4(h6FQ543x37%DOe^B5#%Q)&
z*lTAx9XOnh93CDRriq$m5vYyEb;b7j#`^Uexk=XZ%cn0qefq+DJYk$=+c(O-P@^Q#
zg*%E|wM@fG0$Iw`xS_0OQwBE}{4ipCz;p!Ltqmntv~|qS&sgV(>q=dBY#8u&ZGVZ8
zoC~jS7rwrosoKz#m2B%6X=&#VPXpx%Vgltn&1x>&%HRW^|M&-vk0-WeWnUvkYerqM
ztw1is)|i54emvoDR2<e0^k8vgqS`ZQ-$hPpIw*@bR`Qxlk%L`nTo|ZD#y4T`fb5sj
zRs>F`1#fS(0V=&zsaiFHP<IYvEg{^=zw3?K5|PG!g|^>l`;{7ZzW(%YXk~HZjGJam
zuz$;T&`P`{S}A&iq+}9cw87r9DwG1>Dn@%sZ8#+fgslnFD8~(J6gtR7I3EvCh0<GU
zVY?w{c4O#h>@d=_mVAnJqvnmcZgj12A&8k!E4-G^R1H`g7^e}h13eru^CQj}$@qv@
zO4;uUt&UYleuYAmSo4gYjudUlT46D`!vXJMzrCYs#aP3hDzzq#gAoURuwXsbI<$wo
zT|~@>S`)m8GUjdsb6?Bf|59M|T;s~}`p)t5C+6{h)}E*oX<2#y<&9|y*kG~hZsIWp
z-HK%Hmdx9?UrDXwrio#km|i}y<V4(7R0I8~<K~I!^B*}re4?h6yx-VjM4P)&B`plY
zfYF_Oy`oD7?c_%rMwGVHJyPRBO)C%2U(m|3#|wF1@6=R_sUk=1Ab5hZOTKSRU>(X0
zJb(E?0%Lk7QHjfietE-80U8A3P$$oL`b59&JT`~b3TFd$K9I}KJbApA#aV5NZ$JHr
z?ww`5^7j5ttblcxb)z1R9L5>l3T_-3CJ)AnVX0Ngj&2=Qgm<G_V%t^%7Tpu3%A9l<
zXUaGe4l^pQbn6ApJe*K|W?eS=?HzL%u#2XrMAQo9M5SZed{1DLfRs`)IYx4CXk+Ng
zieYe)SVFyD3FSzbJ*UG=PnoZA;r99q;ovdNGSq<94nK`dv%(LC7;jQ#riwS6B`*wH
zXFy}88UML6``<?um30{bSXTiAKEJ&1a5^!a9x%bvOJln(gkc~Ifs`V7-z6J2Wy<x9
zbKWV7?77<ATqR-lBiaXAt)z7k03V(1EXF@bRjehOSeBiYUJHYB*p5_fe#D|lQKA;@
zAo5GxgmWYCL0ng^=XchR56KB1M$n3s#1r!8TjHC_%wgbha;UhmTwhTS&!~Jt_s%rU
z)CSfXw3Uvymx^f>?>%)}X{CsxIcHpu%dYu=QwGvbP#LEZrh62_eMVhSRp!<1@F4SV
zI;LeYprt6?ku~>RXMN9L&#3nMK4_zMC4XM)w75z|)jB<I_;E%Z9>{(`4GC2fs)KR^
z+68p)f-P^tx$4GJa%4Zhp->XP>Ktg<b<v^giIf`adPT(>E$+}eX?djs$A=?fK4AR_
zO^eJnCzO^zt<+$Cc*GBr<gL_7&O*=4wPC%N=|?Vt&L0EHIGT0xnXQqFOUoE8&|qKK
zZ|}tGE2>0tN%UH1*3xyL87u!g7t->L<_E$-()j2iDUIt5S`n0d>lhbES-^Z1BOE8G
zNre9t6E&?^qoGM6X36^<?)dLONp-=Y^vDtdmvyC|uQIpShUy#sctEMfvfKzJkoiDq
zBknxs=+-ekKXM#K2m@g};Qhem^{YszN@vfF)8hk&PhU`OqTe80zHvK$V_$Z>Rg7MY
zL$xLub)+3LPbfd(@2r1o4XPSS6x+WsTAbJ^-btkf36d&zHq4#f)1j#@X>zU5N}{#Q
zx^DEmqT=mtJGWx2aRcV`M-E^9#QX<Ojyt_Il(DF^lhaD@8lz=tUQ*`UPk-i@pMOMA
zna9AW3@FXM-zY5-<H{5~esluZ)B0|tQFv>3m?uVOSuR)N?M5w?JVs_eih^TsoIZUb
zqo{NN3K~zmoEh4VZ<gzNB`+(ZhV!;l)(t-fQ6>!sx(oQA$!TZZF65F3<Ba#7Ey)Ev
zA5U~T>%OtvUNI_jiZ8r;{=(QdT)v`}&>+J&NtC5>gyW2Em0By@vw%yjpcU<MCdVCb
zJz8smHQ2`oau;W9@Ds)u)^($`#xzb0Vc;}B$cQBstg^T<a5x=A-LWjJw;O3)*gihE
zoxhRef(`?2JmJR^svA^KP?BUwBnb~1y%&<F57=R*DT`5>Rx5E^>1C%oNn2C?MDS3z
zRrX^I{4k>2fSV7{8YS<TQn22kS|{!odfVwCkc~l26S$dc=_m^)6A0Gus2#?8K0aPa
zWoH;C*$nqURvxP@)?0M%_l0jl_m%UHKe2v&<KgKM|MW!g26ucU-p;64@Ve9UPQQvQ
zHf}4<S;qU0PLUw))EuE!98$TrlIihYefEy&jb1YGUX&Q;vEwM({)%)<cEAk-)_Vc$
zX^plXO~o3?Aa6?3a>QxFcKu+xC7kzkDrH}(D4Nnd4znc0wTe=Tb-VJ{U*9ms(U*7B
zdW9my15~H=&a)Fy=5jkz3s|es-eU(v?FyU$G%-u3sqH|RJ<5+Lgi|Fu7D+PvJw{V?
zP&uP&MSFz$d)bNWmF;$>rB%jQCDUsXrj}`R>(NS*=+<?|&L?auEay&5JH1G*bO)Rt
z#s8oi&IlEF%T;K`&av+sDXp|rDC-Sf>%EH-Gp}-<J^#<0^Z!eiLX10uGn`&taKp&M
z(+l_sWnk;Dw$8e(q_m?3pc}$CaN8r(>5<{|Lbp=Q7sU6FwvAE?$CneM3$&WKe*21Z
zfiM=XUw`J?kAG%cZoHpw4CBOipPo@AQ={S2;}d(4nW<5Zw60_=%hG+lP;;W+&KPYZ
z@Lb(v1YN)=L3ZRAg`002VVKbN5o;YNjSWG#<i3;ll~zQOW@|&$PIx>qJb#ZnJz@PM
zYN3u8e{!bRh@lEkzKTx!rEb(3*_R8>YKCEws?WS&he1p<DG|IUjIwy5BS;D7#Q<X*
zsZ?TG*e(%uTQN^B=xJbmyU>gS>q%YQ({-=t>xR~ehd&0w_eVZLV{`+gOtzAb(3@o4
zc7-IT?aO_+?(l~rei#X@-mA*avRnj>oio4u{3{>V3zzc;G_Yj8|KWQG0XI!pJCaMs
z9}cv*aQ^xgl_E-aj4@;yt`|(Vv=RlAp~NdOj`QDgpM#f-;hYoJXpw)G9i23-HjD;q
z6h<pz63MC3hH-wt8qcuZL=9pB?(l?v_#z6c^&9o}jl5lPRg%MEDzwtE1x{{c@S2Vy
z3<INqvUhSSl-e<;7nJJcUeGPzCWjf142LIlm}R-`2VBm4`R)_SYD(U)MpLchoZy@~
zMaekRT8(y&U@ga?$?w=2s#bC=C_kaX0aX%tPZAuYfIDEZ7m;7Cd8fR+Vb%p3Cgum>
z8ugsXJ<_zn2E}c?a(;Ui)cZ8hQ>2$2owxg3U!kqM4<#p3j*_R*I?ftscg(cuw3;z@
zL4^vQuH>_q+UPN%l)4*C12s#)k{u-(t?UbNzfpUkBY{Z98KzlOTU|?lle40#$thBA
z7i=rc$43g8+%{B)c{*Z-0dqL8k4IYYR2#8(1dAC02csF4M3O3{F<$U9K1?Vrco|hQ
ztv2x*82dMqo9&gd?y#*?GPNhV3Djbt><r$Jx*W6G2qwI?jxqt|2Xb7|#$vtXhFq^7
z^c*ogqsuN1a5qRc#N7$6dX<sZO_KY}JX7LEjyt*w+bp$8Nt>wPY9)*ZV*hv8DX3C7
z9FLfj%(2KB<0`#^y1RqZwxbPfDWS2r9)!!2GY`*4^f1wYay^sN!sWMLF}9=U3GXy@
z%f$Cz`SZ7LC==v?F!1>DJ>!!m?vYp;4o6Ilv1v?i+}_@CQ)724O?7I^+^+8gXK@D)
z2%17^h0_-81_^f2;!$wk;e*6URsb!GlF}O11`el*W`zs0o!@8${BS@Yo(Mz0Dur$p
zYc&R`Oq$Xd>rqV}(pCz!X3_n(J4X5LB-O@pe0*X$oy4Jez1}t0o^B+Ou2o3QP}~6~
zgVlz*bOr^jYciF$*Kb@u&U9m`#?g$&&PPhCIF<3bVd{=;mU_NY4;}A4b{<jAfcL~L
zp|v6xS+q{a2fDEYH(-p!itce`+ZJkTgfS4y1M7BUUouO4lTe%Kh{DpGqbp5H1*J6e
zIFNd#wMK6O;PXC^dc|6yeU?&4rHVElsmxpJ_%MoqKxxKlqU%msch+TLi#yg@DD7S-
zBwFvN-U&)xD(6Qw<0!2%Oeg&58GSflok!`!a(yFh5!Gd3Sn%@S-V;%uF>VEI$X@a$
zj16cvW5*e-99DHwHBdCVcY<~}7w!vCkMkpCyK$OFOn4EOwN`X2?@Qs=*E4I&JRVOx
zo(_^mm~N!o`5w_a3T`D#n8P#OdHnjy?Q-UJ`M}T_#tCCBiq7%nh3Vymo*Uc8%6hp`
z(mS7DL?W7V1urqFgHaR>!N{@Da)O)zjT;84vZ9Qsm25!!z%Vws7>v+^d#@sNEs>%X
z-dRvq)O`05a#ykh?P#zKs}0&3YAMuGu)Rxm_TVt1$J+63-c&d=#tj(nsdwUUNs&pz
zA6}kOLtwWfi#iIKV0xs7nd(QH2~_84C1Uf!J(-9QCXBX}Qm9mvmDD|D1KN6gtLSzQ
zgSvN0lF_Q@s?%L$Et&0h6J!mR(mN=Fbyl+4=~zF=-?1083oOgZe*M53MROiIO?Wz9
zP1v%tor^f#T@b%YZK$%dD8)D*a86Uw#x3u1F;;~V?E{RSrnRX0cERJ5{~g9h4?Ym?
z{gNt?bXlN8nsYeQd3b&#1xM{w7LhJ641;X;t)cXStC?If%d$WzU<1Qw@m^z%VweKG
z!25Y)SrWxKUcUPS^UL>)N^|-8BdS)84^O0h;Tkh}U-|35{dc-D*kMLljUNM@ihq0p
zBL{+R5;9V{!gY<dvQW!+NSLwGqH|YLXjQ3MQC&m%4q7T7>pj|PhVg(JCSkeIQ6-^D
zMFJ*eaZaza93e+2k)Ag8>%w}yF$T*xjSPnqew;a+PK0r|<FPmP^+wAgRTH_dMNLQi
zbdt;hrN|{QT7|;UTjR1N_S&c=vE~&s9(g!Ef>zX`Soe%s5^8YNSlQdcrY_VNaiiz-
z=@Ub6XgX;xxM5~^c({{QGd;<CK5q*pM!MFNuEk{2kXBtQ?0CeF2Xac%;)cL>S?FZa
zdZDF^4-VyIk?H&(!SF~mS!$(qFxp~VxMv4fY_M{fbkS327qH$+0-Dx>kJgQti^QWk
zhqZ=Y#D%J?F!z<V80T=NLETVFq{lkUxbZ}*5$z;(Pa8w;l~NMRcEh3Rv7&|(b{KKf
zkusdnxd=I^YqSpHOG#ObM<$?+M_EVCiM-zkrlYaY4aRC-P9ullSYi~Gj@48fq>AoP
z*DEb2sTl2!YrBjUgJT>5I!OWIoj~MhLO>n-0J9?u57^^|brwGlsH9P=B2|dHNE?sG
zM~qUGE;Dt7!n+y#NYPd<He={1(VOg4T&TD(2+}<6)GBCb(~%CtfYFYoE0m0)({m>`
zFwG$5q<{%V6kKUVC4s`ZX+XP?UK*^1Sax&;Q>E|%=p~`Nkt_%zL4!FxpxlV{15p{8
z5}`>;8>Ee#R&fP;EAOdMpd770tB6Dt?r80PpYaQ#!<ax<GIHw5V7!KLqNf$Bg#z6X
zc9!lM?*p|J8BNpJwuOAX^8Wftxt)1<e!?H0(7j{4r^ki3tyn*xg9krR8tmIbFdcUg
zYOqlbZ3}tdi77$Xq`XrBH;h;xFm_;knsCEFx&1qs*V49OYQYS5spoY;nM%%v)Swwq
zcMNkFN4c`JIE9rTs5uhPZ=BbN9!5@2FW67^PBdNUL6Mur_!Hyvfx!$U69}g#^x??u
z+c)CAayXq}oEfGAx)tJb!M}au*B^f&zkc8*nJ1qf9~h?@W38lBO*1LwdlJ!=Vl}#p
zXCmc-i#q{@=>^IztvWJxYI&p8#`W?pfj)X5Wzi3B+d^+Z%fu=NLstf)1#wMF^pep=
z(esY8n$Ztft8igrI-De)@$x}j4ZTFR%N2^CvX{%1y>!gsh2ivqnU19Gimr{C1Wx9x
z#)pB!%O~b9-w7-JazVR64v|u@SoXc4w#+m+PCxt+TMNtegRBcBH`=zNyvJCL)r#6X
z>$1^mB5$jV!Fs*RQ%6c`G_7#PqP3X*l+su;P-??iM;Hct7*M5=<H~mZz$(je7I(2R
z(v_LsCB!7(liG{~!@CgZSxh?I$6CF2wta;PL0g&ITa8f;TxGELQHwEv9CBU3G*3K?
zl6;j@M8$={Xsp&?G;NwuRBG9TZ4?A&UV0~PJ2gd0tyn+cPoH2spvMDd9H159{f`sc
zX;NNzy}WU-j!|2|S*$nK>y>nV#r1-nM82p71xn+r;@ONqr+H5a6SXyQu;$3Vt>lk)
zl)2#7GxNhU_V~i-ctVYiR@2=bE@~l@D>^1{8avNS#{;!h5}n{?M(<EDbGxmmlCjQ%
zwUpY3`--w2HI4M)NYVH4M;D5E??o!?y3=E&DMd<|o+7puv{ld&#zRq!v~QHOqFTjy
ze|Jb%YOf@nv87Ablr~!4F@wh-qtg-v|IwAmLrd%T3=L68HE(GxNmr<0Fat#y*7eGs
z8@YCB+%Pp_TBdfekB^wcKx>I^t#m$qlo(KBBy7jrpXuIYxebD?uny-2TGtdRcAA;!
zl(>^(#2AC_osxG{Z#dN`t>c}>s?HWyCTHo>h(*qY?ffRhg_<}H4mV5`?NELsr9vr@
zaS)P2*N)bOG1jePIzG~Sr`({`iq($MP53Y{Jv?HJA^lIL?EEiX3fs2O7IA5!HOuYB
z)G~*MM<|lLyxuliX{?LPH&cmBrxTx^KBJA{{p%|RL0KE!gvR;wgzl3pgoh)iFMq(q
z#K+q=;&ze5vDPTxenY212s6WYWI7)IX2AMKFJIRA?E~ixd)g_b2zfASdMRk@??{0}
zO&e*iSm!a$qN^pY3vF942As7Lq*EKYR9x^PNsXD_G+S@fxKYmE$hQku6yC~Wv=woq
z)+9`#UNT+5>G=~j1iW`VJU&u;10SfZQ?`ZIpZ_B5H9|&I#xfOHQiU*c8fQ`k^Zv&E
z_7nO1i&S3DU=4)HGkt!-3?p${uuhX>!5c;Q1Kq(I7y5dmD@DPurbhK6br@K_y|VxG
zmBZuA;V`06Y|EK-PoSmZaf3(E?-Ic(->U*uL3hx~pp_x#h^|Vq2#O#*pvwi_Kr8W*
zl-A&G5;R(22pH`#rqG(ah*~S0_aY;<292WDU2=x^m3p7Y*4o+kB+h4LL?MG*%z2mG
z1v}1+#{+%2iEFoZMk-ciuuyHID8M+J?zrBBqtZZinW;AeIShh0u#TcYmm-XNDoQVu
zoXE?<cKsmdPRp5;GRwys`^Q<}jwv#f!ohl+HK?wr_k+W`k=6<-U9n1b1gI_%np$xf
z9NH?p?#x;!u7ynMNwUH9NOgv;R`RydT9n3ozSE7Z!Ri4!j@VjgMuLl!So%>_W0e(E
zpC9SNj5Z$U0;UP)DQzqJ^~U;fp~jU`6Z>}MFpoSO4>YZ*HQ|aJ;36??vXe6!ZJhX1
z5IU<W5_qNPvKL92bi2{j%DxM2K)Vs|95fkQX&e?qOFO+38eOsj;#H99wNWlt*4rvF
z(kT$UM1<xVNhRaEBKQ&QCJE8%mEJ1lj+UTRj5Zh}3OI>|t#UKAj;ax@G<ulnS@N!H
z0<Al}$xbBK{k|9U_+j8Vj3{l0F;YvqlN|2*6Kx2?kzpR_TGNfidxr~>B<GBw=7=g;
z{8i4OorBb|Rbl8j?+Nn}qXl`Q==Xg2zw@O~l%yz?+Qh6C0tyQiZf{qXC9!Ww+A6P^
zf<qw@c<d`U%lzdLeYx@LKmQAZ@f;5azW>MnfPI|Bg{bbFcW2NkQeVIF+t**2g0zAU
zpFcA^h|;9=#uml>c{rYU`TT;@ieV1?^z+X^<8pgv+Z@v<!7#>J24_i4G~KE<dJ>Iy
z+JsE$hk+6R?`RhGw9(3)on#!;PF#2PRIxc>ouih_ahj-MU|DW=Y3qvWo$LEMVHo&F
z-8sE{0xi(5+8R|UI*?jrSBlbM3?n6Fwk^?Ip%|#~M!3FF&S%&++I}XdmE#!jgQKe|
zX>2t^RJdM6r4@Hf%^0VNsxwX_%iAkyi{u7LS*FjQndg~$+lc2grhu<P6Wvn+bGN-{
za0)^DHi)@Flh5A`vOK(V`hR~x(Vdz#w1!biKx3+-&7D`T8wxFBy<F&B{f~p`Z~F8u
ztRiiwrJ|g{4?(K^^+wzlfsc)W+A6&!y!UuLqG;3@iOZGJI>X@6HA=`!PFQOo?`R`?
z9j(ARLvSNTL$5`4QTN4f>k@2Vio{4N)g()}6`Z#8vZ2Zf`x#vltyJQ5A<mEN&2Uoz
zHJ=EBrEe?vk>)0}o3J(0>xv`eS|KLr$T*^HJ33}&AMm4R9%Q7|dxUzWcF8T)%2GF(
z+t=fx?B(jpc0RMm!u0rro*iZmK%@1B^PbZ%vX&~cWh1%T%6rT>qpZaniG-A>N_96}
z9nsda?F-+2`8VEw`VnKm|HGfqrz3R9bvIq1l}2fM4~de0?p@Q|-E*Mt9YHO$l5xKq
z3im|oqPcWFU_!tsO>MH^R;tiiA#EG2i4I?7L4wOEtkJ#DOTr9;81uGBX&LQhuO-i!
zAUk@Ky<f|jD)ckC0Q8+YfJn4^gI*)8%KyJ>gVBmw3fm%qR6ba=(UhDhCCd)QOXn-g
zWyf@!X+<ayze|$)@6yrUu}Y}C7DWhwloNHo$zQD_Dk)=7#zAQKY{<2VkfO=wBp14K
z_wVHId@0n}IbRlHEqwWI;y4|d58=)jUs%%(f-pqVw$gjy@nPaH1xO2}z;Jv*8O4|)
zy{@cj<ITc+p0T}<^OYZe{zh6?(bx8h_lBNeetPEghd)uAC#Q{+1U833UP?6U<;Hrw
zL2B5^P;L>1O16?%rj^WDowbx&M0wOi<q~P&48!qJ6d+E5U_#TZw-v0TV%Yagy4-{U
zy6^beqfx9iv)|6#ZVSuphPRelD&Ni*h8Q`1d{7mHX`)oI&08;S*SPP5!vTB8D72#J
zM$xooU3a!+0UZeQ5$|AfBkYp!q-pG1<a)j0{lGX+7%PG2nnv0$=qwetwT9g}taD7m
zfN=vw18L)N_Eh&sRgG=gC@E2U6W4F&ZpK>WpxYBD#o^&3azbmVfbOD7l8DAmzo$%f
zEg1yX8d=uafKrMa#UoHlrS?Wn@?y(pm7=1zj$|$f>%TX75xjRSUGOC(SN5I>-l2`9
z8(D5kHiD(QKw59S|N0YgxuUd|hhNJKyX+1VMoBZX^5PmT30_8Jj8e4D9g4P6QpOBB
zz6)wWw@TXzx-8_^U%CA97ry=a73(La&;NnbcYowG9C^xxzPwVeue94YYFF4{U~q<B
zHsZEO=*ys}%G0f)Daf@_ddC}0Es68D3u#@b4OEyI4$nM&{|EeZB(swCN-x6r#5?>n
zP}EMXJN3G<L_;eH)hi(cKv8o=IgtV@Ys8P>1jbl%LRBpne{hlz*c+-R%6j0pUh%3k
z&ojY0pks{1S||97Dx;sK6kUr$x)foWH6!k1qd;{@D$Fg*1*|P$7}3U&iVzayGYKGX
zPx+pCp8m$xQkuWz8YrhR#$c`Br1QRWJ)hCRa(H;Wqa&mf(`d=MPCF$h%<l;89ND@!
zovrh-Gg4r+xY<SVBe@>^9*{+^l%m9^<pa(*6iPDVKi<ft@afC<*lD60O|BYkWGB|T
zcwllW)Y9%8woW&ORtg4<G7=-&m2^$iA_<On^<c@eQwl!dw7yR){vFI~Y0uoYikfD|
zr|+5OnUpT17zwlE>ANS!moIcP5Z^y|ef`Q>qNKA`gHaaSIt_R@9f&cKQM{c$__zqo
z(1%g@N9oGs0}oFx9A?ow5A#f^g*EQzTBun9I$|wsw+q*A?_6Ime4Nje*0{W%dHM9r
z$;{k7&P2N~&j-RV%2F)dK~owk(5&Y8^vLx1g>79}u5Xl&Go}kue#^Ur#-k`r=7*h5
zS7Kim^?@d(%(`u~X%?C5@sYuMFiv2fO+LHvgP4!BCY3BB6eWs*Fa&B-ynkFcpBJ3z
z=qA1fZGk;8TaB@X+p<tn=6bpD_V#DAbDUltIX#~6-lLmTKCRvXfLbug1V<PHHb!AT
zwL&LD>nzs=?;Ofna!P`aj)@c_%k|2#uAGiX4i871w^Sv#bms=FwX`e&FSTdEM$}Gi
zMOgMgNvn8AN(JPwvD%`8G`afTL6uU-J!7>&X@#+tk_)O-v{DSii0bXG>MAHb;Km6X
z2DFxTdpHe@pCB$5^6gA2vQ6Amq{WTFJLbd0Fm<q=ZX7vhiWV4IEm=BBWyHbS3N0tp
zzM%IT#_ZC`06K5P^BZ|v$T1RziRt+ne>|Yw$lRLXo%fA0jHI|x*9*6nDcefkH<Sxh
z7ifN_xqvpB92&}57)Ns6S*|O~w^v$enBl<utQn7w=pcC%>MobeDUw?yoMs*#50qSK
z`zjY{O7|e1N<)^xA>4lpcXwZJa`80ZumO>AHiIIeYenrh+IHnIdWJ8b@uUCSs88gR
z;*RgVG0)TAjG(O}nIz_a;GB(7SmSBdP`ed_sl8`9v_`L;-U_`)2bWX2=R8+ZuZ&^D
zd5<$fH#SCNt;hO;GM2b)5`7sw<MF^WO%hX^Bg*b#N-R6Qh$o|~F6oLXqUdr6Ya?*V
zR%Fl9#mSxO{R}A*wA1vRc51}e(tCwmN&Ck3{z1-(E1J{iFZ6prQ>$_*2PZD@_Ivl-
zvSe%bhP6&cS>4==q(-OEoAAej>6AN@iB?ekeu4HX;Y5G;OQ9>i|IhzBkN@GH`0kH?
zWby;&*B|-upZ^<ejU48QX_^=gPo#OmA1AZ{L*;tCV94y{9a3UznWn*q2|Z18Z_!Sp
zrwMB;htU$0Y|#^nm@>=P-!SV+(@J<nO-2LPw|BO6K^x1L?>}J&&u?G9vZoF06?BQG
z``9;b<IK~S@5C!nGG!NSZ)qK4J*GffH{#ow_n&^^?dvxQzd4=o$487&L~ZfIBtJHd
zo<$2^TEEjKMM0BqH@p*e&^XJCIK@qH8(YE-m8_taCQ{xU@jeJX!405pT-TNHFc4a1
zeftVIGR-4F6{@OWf-qTzk-k>0>sM;sVVDRHmRdT!fJTc{*Lsv5*ir=3xb~g0?&zGL
zSApG?%K7}xd_3aZOx04ESnJ4WOlb_-Fdk=Yuqdr0e*lg3qSz^>Kq=HL;(=NU9WbiX
zsvPLSdupp_?fzEH)pl2u^e*l1-5?}lfjbYUl)q7IwXvu!Ew}cDVVcp}F@^z!rK(Pt
zC+ski4+r9WWRH=OBk^{j^-49Ctc2ZVtuS%g?`bt%$>^+AP+bVPLAp#87wX4Xn(pK-
zfH`e6^I^cc?+B+8)&>Tx>9;eje~?ooUe8c6rQP=(>qQPu>(Hmq^!br89_iXsl^pJV
zkQA_Pt&E2JL~=scZMAUw_=a;G+85k-!gfzgSJrI<I@TFRH%ln|@d4EeIR$b_;_7d|
zqaN;>dVpT2d!*%pc`#@*V#xF@Ln|!TcfS4gU%9@&GJ1#q!}p?6>z#eu0Wo$7SIp26
zH@ZON8rB#|9;DOm{~lUkb99B#;u6j|(Mo2x(?n~QIlK^jHJFYgh=R@N`=6!I+KJgP
zIFZ>J$)0vWjF;Ld;(A4C=_*o=<S1q_l!OEQ-tib?an_>|YH>{qy%!k`#YC)%+`;|c
zrC3MrT}C&*dqQ$0mk2a;uPoOy%6t4Y$zR=wkfJt`tk>4=d%Hx-5v3H)iHlq7j@DJu
z`3xund#!i_1cMDDtw}c{JGd(Qk-zt)Fh9TWAOGwBlZQY36G~f=PP!3g0_Tq#zy11#
z{_`u}|KX1ue|YB0ci&T5CoT)O^TzwPuY_sm%hLmHKC$hISUavX{5%u$itPqv28M&d
zsX{A}b=jaPa?Jer-~W{@CLt}-Q7NHHq4dJ>(+dyJN6cUelO-3Cg)X-hilC#lvs9%3
zMasf@v2;x9)U@%-+XuC2T2AD3#ob3FPk;I&VH~->e{d~=p4LX7e?3-O{r|fVx)@`m
z^_01_cSeHV*PBQrwP6|tfd*EoDorcMg<mnoa-2s@_yf&(+ID9B_KK~U<M@ej4%GcZ
zZ=LD%fbuh&oq39twr?Cx5BM<3!D<w_H8K@H1Th<-`Sj@vSsUW8P_Kdjad*~D?FbS)
zc&SXBuw-iIPL@3pyoGz0Q)(6@d6dwlR`2<~$eei3Rc50iNaHk%6SEbW(W>60y7gWz
zkes1+oRiU12*S8&tzxt$?P8qjt;@W&6}%mwK}s7`K{7X>l+2J1XyfqHi7*MF*IPp!
zgTzVtaIa$3-#ihhCI^%C@`AUD>XH=3Ad2diHd>bsOE*XN19tGF+OWQfLhb9%^!0oX
zdD;YNYz#E$%D^yVM};1axbZ*@6K*;R7NZu3f{o}kQgWe|PBRXDco1}Q-eh-mej{#I
z@n4K5s%iADIi4P{O5;0#5xVM~@vpNLM7>f%1x5~W5jiXQZKJ4dtg)2!#=b1X%RBGC
z{X|?B=6Pm5e!?A&C<P_%EH^O-T4x1MqxC%ys-Ug?+gQgt@es9|MQi`Na6xrcEq4WA
zXE07s6ScCpoqB&ZYHMPWz2h!ol;^@)jk6Z+-L&q==qE^iL5w?lieT^lijqh%(UDkE
z?PTnw(2|?bHLdQ{ypy+umR9JUS}V0wi7{=Rsx`)HoKccqS7fZydP9eR3s!bPo9s!}
zZKY_#d^q5plP<azfl1!w|FvXtjs)-NN}_9xxsR^&{USA*>F!N&MPaQb*G@+~TM`)d
z?{Fy`PABH$5vvv3vT!@U^Xp&#jka8voS`$Yt(l+y<zJbeN7^v4rABoF)0aPy>_FMC
zbUWf7PN*XotN8eMW4oR)Dx-_E%i0-IER=O;+a#b~Ye&X#KCfIa8;ZvBFftp1n*ztv
z39VrqJ>&41>*XqP$vI%COu=JDN4EoU@9ZB}Are|ctBLituq}(2(!9mZfx%b?=kdm(
zOTwm&p%uO@SFEw<t~r04K^eRsnU6=57d>@P6|1!zqR!ECq9kbr-Qg(n)YhnZgI?$g
zVv2;`u){zI6XRhdnZRx#U2k|d;2)nc^NhX)R9eMgb9$jo4=82W&Tqt$+18cA;e>L5
z%f~x8?F6Gl%{7Wa%uFMPQ(&v2l4`xu6by$0tv9J)P<Z3G?h7ApZ%p2Cd_D<X@}5)N
zTE{s>%ZV)}lyUbQ28b!~{;?2qVHyJS;Mr3VC>ZLl(2xKm?VY5Dby?t=A>fCAHBOeu
zT?}wJC(a-5Kqmx|CK}a9+d@i-q%_0vi1C(m-Kpz}8)oc0qk4zDi-^v8=>&QgUA|I+
z##TsHf;IH!1V!9-bQkTpG8Vc-Z}tv!O|BB)uI{mz1_`ZMmv?I2?$sw~V~IIq=z`h^
zPv|gUU66f?325WcTJR@IDRQl_MszQD)d*prqsX;UxgQcnaMC?znqDN6!;Kj0pj8RE
zQevu8U4b%pRgjp*8d9yQj&u&KQF3D6MR2fQ&$PQ<Kkpm46vl@qhSLLmIAFspVOCw^
zCWY4bOWk>?SZhLeMA4)a_5PWx))*uUqSo|#dx0`qxIf0>f~4T}T2Z}Y06zrDf4)B}
zDMj{uVcRyG6N$7}o!%_1RLUMn`_8tm)LL=Fz%Wge?M6j$J)fzi;>UquK8OcIi!?PU
zBM9p%il5#(t<=BCo}INgBq6PLK!eJSoF&K!WAXDSin$@-tP%IOwvyj*AA7Zua0cpq
z(cYRk++~N-D5YXdy{oj8bm?6L1FC~pN>oy{pqsic%aH=d=)eD^koKLjEWB=!ZY<hc
zbdGFs=jqcEKO9fIzkTr2zx{<D|K%qRKfu#J{U;tjeF5Nfxp4dCFQ~XeZA=f3Jbn3|
z_%Hv=_VFt%7wWcCN~6%^R<r@_I<H^95o1QR&M*zQ!80F5Olu4w@bvu1>G_4Kpl_RG
zFAsrf8hQNwJJNQ8b-^_F`qv*>E|D5{FbX?(lo1ZfX&e|HztHKleW9&4((A9>%EAx=
zld;&xiPBEA3S>wjky`~1b}~2{AVO)8U}PS8yI#eW921Arfybw3+Fi}C?mKbW*l$;2
ziVR_(H^VqVS36Z3((Qw`u9Q(zQ=$)(K>G+1TwfAq-_R|it;H%$@B_mgo1t121(*fv
z4P{$-{q1M8o0$(MN)qM`#&X*?>K?`1HIGckN31gZ`qMWquOH~s0rhYYBUlq<jvGe0
zGHfxCb7mL?mEerxS{CB%#+EX^SGv(?HBgfLp0yTwse%l*;#3vSL!-6|v^!9-O83_!
zvar@laE8_@xkehDaT23WN*R=pFSYGrhFdSh{VKZ~qh;oNC!Dsf<Yn(&ki*VmjFHYw
z8<ZVrwE{+}OWn~<4$ru6)B?E~!g!QCXFU*?O^BE^%3|_9n%Vb_eO(#m2mEkkI2_So
zL@6b*#4g1`i5t#{=}PrNZwb|#DCoi<IoNr_O#{<3FplG0GATwp)uC%cN}}ISTf!Pe
z?o~!3%AlRa+z*^u3p7LT{T|p-<@+<1loBx)8HEl5{_zv@lfw_5k_)k9hI<z&hkJnx
zHFb<v=qZq^7BgFy3*0p^bQ<sPtf5Z60~pbA33w%h0G&3JR+Q3FE#XaP4B{M4xw52<
zmLqB1S+5KFzPz4tMj279Ete~4+o&~Ry<EU$k1V@9TbK8<Fo=RDjH8%eonr`MU^Gg0
zezZ=Ek(4uT2pk@cSS1+pR;oCbvE;h5tyfwF;{*A25qwM*5T^+ij3anM%aKw#P79v9
z8%^z<R5ihiVX$P`=_#e}BQ5nmV1%mEnsj5TH^G7jOSwl4Na6B7ch3JWT?+s9um8rn
zCFW0G`2L^%2{U-+<C8@1%>%nq^x&|Ek-z-)7yj|-Gk-ihFpUTD7C}k0;I^;wvipJ3
zI!~WJGnvM>pZ~(IKYt^x8_sL~^Z)g~FdR;dQ)S%>tzBhI^gQwW!*|T{NKQM?rw1sV
z^SaVfVqaI5O@L{^dRol5u1GPlZY#^jJNuGo*75l1GmoFYpo1snM9vX+`b2P+bbjX+
zBg^%j^W_7h8-M!Kf56Q%!#to=AgzfIH@pex(MvvB&QQP@(XqFZ*i#m!xsq{&3xQ>m
z?1d5|+qRI>MsBj4TXLc9|3BLPtjCt@I<p1G>aE$0yPqZ^NKs0q^4_d^tN;HK3MkYC
zE-Hb%KvGnS3`WFh+;?YYtFMP?=j2O0WTMCifq<l79QU)g*;;dc-xw>siGR3NAth!i
zxBCTUEv+O5r)diAudm#`evuY5ig6s6ouRG*gDY{vhmkO!QN~ic2-{2QyYNTv<(vj}
zp`)3fo;iGYPdFSXYi2$i@y@fY8=pRZAq2;Cp5$aT%@P+ag<%*Orx9x%sYJShVULSK
zW1I%u5GW~8av{Y??;EjXd?{EBet)>67|A)~tt{zF2X75w7L<dvmewoEYPo_Bk|3&4
z#(Pa6*a9mkmfoceZL9+n1yP?GYxmPz+mldGvn0WE2xz4#(THzBS(NE`D`ln}6I+fL
zRrgb5!)T3hl5N+<Ndk-gVY`pOx@xE$b<+0HN!*s*$!(;7*gH8bP%@@ul+g@f0Oa8@
zh5;YMa}dT6^(}GL6)GpO+{K2*P?tNlXSPy6TkJ3~&NJOuKwKN*OBMjXF_vLIO2fPN
zJ^3YtL~T8-Nt-xt3tDxO1(JnrrEd>1(5<Hx$ZIEVk=9_CEZS5Vhm{SLowRSgV+fAg
z3>7bm%XMQ}u97Zv4s8OoCJl4d$hFC=e%aXW3-Pg#w@6cp3T(A8%}10OSo=om31>u=
zTT7+vOXZvjx9g4j67l0e%!%3x<2cZJBLqiNT4KAY$O0J9&WOd)2!dXru*TA2q{Que
zf~l-U*jGmUEOMd{;x2m(S1D-5$!Ate`OVPK)=B3hXE|8zH~5a)_YQK)*IL*}x;NPm
zX*skANmj5rfAg2ZufP7vvTXcI7%9ub?Rp^>7(e_3MhW&HH1s^92YEOQt)VU#`t^po
z-$<{Y(K&KF9&n?lFBe8SRVlo-{QBEhu9qu+_~(B>drw!2)EbAU7tZfKieMuQ%%>yd
zD8FDUsM{C|>Fq*Gg~RE9pC;xI(4~<dE8F#sQjK96Fk$57cmKfg{YUIL(NN?hW-8rb
z7zToRLf3*FMwS>!k2k(v?+m4(d|;R-tPkj%>3+bR4ru47+eSqqtYwR6Z8#o}c<(7m
z9IY`G)^!n^R~IC<o<<JS6XVkhAvk*4(5-PeAE-%Mm?{itRY~_Z3<}&tm}g3thr}F$
zK(N-8U%!6k{#dZqF-!-BmuH6K3%NJIGIT??j^X(OF-5M+N>2r6C#I8Petu>;9<kn1
zV`P{;Pw!rc+d{65mJ72DxG;TB3|Ub^?bGQdwSBEH-VnMhl(mzQNNtT&3s{N%x*$9u
z6r>c1QDUzpSBBuR<1DG7;Kf;C1SjB(W|#)%;{oe}q@;LIs+8+`Z+lWhp^T(A&_+sw
ztzo*r^I8{)dj?cfsFHS-raWX~&X`_h?4*z{Z)>Ey5z|IcjrD$$IeG&n+goY@YNOGK
z+eVI3#A-EBQj{$rP~yt^xQZ-COFOuvxPu(QG$rz#J-2V1lk%lfwAv{#3tL}ldRL?z
ziQ7Wc(lOW`3pU<q#$fy)RMOT^y32t@3pzy{*86x#Yp~)i@4&YEAX=@EGpc5M7e%Ei
zD>Z>>4XeAne2mA<BMQy^^JgBH8|U*APv;TU<>xB9wB9LjK^D(yFNfv0vfQuKlw?E`
zCajh;PmU53Ze8tNlsoahf<_KN<BUB#ad`hDF3b?V^0<7JR(@Jp*F~@wwNYD_Ci4)O
z&qv02pw)u+4)2_dX!hS*ID2|+k`((kVpT?V03CW28g5({IT&ie;238yPHshL=rs8}
z?cbGJJH0j;TUjSur<P@}0XXL*t<{@E+V===+0${x?EMBMV<puL1}(`l`1`yRj4>?R
zgTMarXD;_E$I}zP|Cj$nI1j8bv&L7JC1I^BRF>-(uD|@5!6?@2EBSHf^Ur_ba1dTl
z&>ovNNGtVvL$3?%eq(ly(>#bX**Hq=xM^lO9+|bY3iJJje%xs}kxE6ofKjkM9;7w$
z(@!6mKYYMe#r5)qw_konZ!7OVyyM;Rgti064<FFOK&7K<qplD7<&Ao~ala3^AmR)=
zOdOo&`uP`r`O}|BzkKE8{X52C##+nE?|z`T18ROkSqt%oN;h(jTt0o3CD8GJ9|qF?
zl6bwm(Mn?Jjcr>o3g&sn9*)c}A856*#8r4r4c43VGQuHaN}+(Hvf{+}UJTA#E<E0@
zQX0ycU;p|WTaNt0KmMaAAw!_K;d{wR)rty%HqKNxl69agch<OITjf|gZB67Yl5cmM
z?o3YyMx=1lS|yAptR1N}QH$Wmjn<SHNn6BD6BdQsK@_T19&cCTW2IziEl=~rJk26h
zN`;&yo>!_giF@nxCdr^-mY8i^?vxlMc85VVO^+LKTiMn`O7h^UMfg44O0vUP5s#=Y
zt#zklwp(H$RS~FLH-PmRy-~IY_xp{~pq7a31=BKGLyeGgW?L4VbCRCQ0`*f$)97MF
zE7EMuIk7Ac;<liTVjKqAo;J~YtGm|3dY8Y$j?AP{a+~S3VyY8=v6fe_>i;7pBbC37
zLf%@+dbD*^GgI9R<6da$1#M(JQ(H$hnbU*JuC;e4>tzS=ZPccqq)3aK+z1imc~B6X
zpsfu;J7_@H7K}toyY^@W-hi@=gYC@T5WF;QwGrT9E}62f;uEeC^?Y34h|2>SEL+0f
zqgW8T;H{Oh(qY)5h=<k<JI@UB#OdW7&%gfz$EWu+Ghl{^!|6b}-$;)u_pgnbB3(87
zG~$m(w6=_gkuU@)DCWHzVI0O!_;Df(qn!G+LFEie(nf>#a{FF4C{<$AMi$-HdfYIe
zot1sW-i$USp|nREOJm=c7z@ZqxYQ~y9W4f(n)8lfmSa~4lG^L7(OTa1!5Y&vJ<BaM
z*G!YB+u#1Ba6Hd^e!KD8r?0rTJOBHCl1O5g92NK1H(oz|Wgb1l@xYQc{y+cc|3PS#
z$s1}bY|A1MDJ<Q1o<F=Jm55b^(JM~pnI$G}*Edf0gETaoWtxtR^O<G2(YBRxeIvvb
z+bSkFYWL*UdA;3ve0gJjeqnlgk2)OaTOgdzRBMSSW*fL(@08x~yBqd#yGg036;v3>
zk0y~<Q^_d2(P-5P)=~S$wmwMf!ZaPoC6j9;ObV?4YjI|zq$nkwT6kP8f-O*DtLnWm
z%`+_}Ii(qm>XmrENWR`E%JQJsf*%i{HCv2){`{Gq8o&GDBj<NV;;mBrNV9?KRr(5c
z2&8OTGteEmDZakF(QY?>K=btO150UK-|i%C_(#GFv+!KTnb-S`<<ln~mpjMcWuDy&
z+9;f}I5*<MKt*vlJ!7<Gz5VuGCO!=lIcF|!7wocuQs2=8s>{RV@^%5l(V?dx%g!V?
zh*}E9IIImq;M{KywPt#bV5C`JYetpGvLu!(cps!4{f#Wy5xlmQ@krlCLox_a-;*(P
zNTssnDqdh?P)(wI+8BtN=ut~T(I~CZX*|AV<r~byNEj!u!fUFf@5vbUrFoH`EyjHm
z=)rhUGQ$UPdbA=-Os$PHfVD?yX#^Wx34@Hy^#0HF-qB7@@<#AYbk$`6sWh!i;j(Cp
z8YU<QbdR`k#C%KsbQu+OrM{2kB$3;8m-#|z@_?*qlMst>GTy5dWsQv2vZU^s8&Knj
zw*g%nx-@#ve0cxN@%JB@=Yu#ITKXT#+g34Igg_!F*iXG$WBe#>^V%toJ9!g)jM7Ld
zDtO#HF}D+HXH|vZndXr(1e}JZDu*dBxfizJxzmY{I|fO<xnaZ(1J20H)>ym8l0hqr
z_d%S;rJ_m~9qTxOj%d}ya$^h~X}de63F9cYNi;(kvB8r|-nFSagReJP3TwC9?4-F}
ztJo%0S8_n;B7@qt4Xfqz(VaZLowNIAo4sjY=n=XS4pvq2cgx@7rttFqncv>-e15xP
zmEyzucl`c#!K{wucIEBUU%A~cyuAAl4AToKC4Twif2Uus9H)V;tc=c}i{ao-Orv8M
zM^Z@`@3BMX^!}N0dvISDe);!5b3AYO<1^E6r0P3t0_F40V!9^X9>i<MoE)WeZugb#
zzH<5NZ=9c=ndgc5Fr#Z`Gn!Zu`Fi2=ub-G6Bf&X|i{>o4Jgv~@XA#qALrF=p#yR3j
z;xKro_s=LN;5=0_`FiEAU;oN@x$*J$AFxVJg{oC40hI_Tp`&3wKS67F>o^^s*dBNJ
zkhCU|y^=`xFZkfmy|KhZ%vWk}42KgrcRqdnjnCnVIgKLb{(6yk@8=7L(J}ZzYX7z4
zz2ow@kZXnp&KZL0#8vX=P)W;;z9qD=%+tW^EQEo#&#(OYxS@OD>G{avG@*;f=)iP*
z$MF2jd^l6qNM654AxRtBjsmb+6P&Q(OK*aFE>-3bz0&CNlJN%BDy=3tnJV_-wy=PD
z?U-6gZ?A0k8^burL#qjVuyu%AVq5b*8q(t9wwi8@;0d&m(=k#=>b1zV+ejY1)=EuT
zD4%J?x<K{f;~lJ`>Bd%~u=R_fsZK#j9$p#QLG*@Gj+QDVNsQ2F$m~g#BA`DLCj4=h
ze{YVoz7lhmeNb(97o_D~MLuJ+5xlvT7$;c}&)SF^#QQ*vk-SB!(UOZ-5=aonkD?ok
zRZWsWPEM^>%R|f<gB^xFiM7ALYLjTHQu}zPKo`M_){2TCq<d+UeOE+lnAY&a8T;}h
z$`7>r8|m|(QN1vpCJv{OFwLaeAcA#{(xe1ttf!j@HDg<)n1m~piU8)Uwi4?uRetYi
z8KI2>;dDYDChUF1D(TgrD)n(C*Me~#XDp?4G!5@GZtwyMG~#B^`@yfIEJZS<@xuUJ
zSY)MSv>CC7vk+|81;q{1GqlFGMd2h_EeqsnqG=^9WMlUTYo->tO=_XP&`Tk=E)iiT
zsU!Ic7>pHvL(S6Mw!3pv?YF~{H+r?wLs05Fj=q#a-$!fBjhHZq!SQeXQb;jjP)y#V
zo#D&t7yjG-`~Q#!p9XJ1TZnDrXb%MMDIZ_B_JZ>UN@EP3AAb12;e4c)mCG-G;r@0(
zmyI<?y7$b_p7n7ToW%Nxr!At^J87PI|Ksn_$0O}HuzYE>`yy$BqDZN6KA(9x%&g0U
z|MUO#|Kva2zH<Kf%(iU;$IFS93Vv`>2C9`cWgg2)Y?+sL@Ae{{A#aKFxD%AZDvzHh
zYO2I#qvQ?WVX}s$XO`PnmJ~UkEP52H7Jgt3Gvo1ytsC^l6asNej6oB^K&u&RpNP9Q
zG3;<R>jK8h`1t;MCuz;$`8{qrVZ0|ei#a{h{lN9>oxl8F|4zEyr8WA?7h;aowlN>y
zp+`?!3e(`Rs!0L|#qm6#Y@lc>sh`$}%Y*p(RS?jsQ+s2aMvgDfguzi$Wt@&2UVdc!
z@FT-8uzdZ38_$eQBBrfZG7&$GxM1<#qH9HKPb-xn1fD*8kmzWXKtPHstya7h`&F%x
zS|t)#S`pBl(HJ8L4LLq3QtgxmG0OtbSj%*nQO;t5XK;?zq-{NfftoTZOOv;xL@f=h
zJVd`0<8qY3P|XJC9jXfh$67<~g{>A)8si<#+C9r|h}%Yq(w%5kzL?Eux|Ng(PT~E8
zpC(-J^tjK!_voQuK`OZvS}D-M1}!%QI-?ELT>)B}aR0T_)Fe6oQUpTSm8L62PYJCm
zx+P4nf+%+i>%=ik>#*;$tPxaqZ_=X8MQ*jqXsjP-+DHmYWr@7@vg6Aeu}Hd!-YGS4
za0`dW4V@O^{VQ#|QmL%h4P7HHRhk(HPMiglw%a;87&%y|k}$qA&NE7DQZCe9G2<k+
zl)W*nhaip?ZE@Z+pfGJ;?naReB_$#&<AUHqa>Dk``RNH8Cb=2a$acT5-LBM>vEF?*
z*|k0P+l@z0NBrSQc0y6&v(6g$Mu^>iZ#sC~G{~QQSxGs`ZBH4rRePFAZXVmVQENdf
zOAF#$&}d4lcrOvRjx5)WQE2V=5fkKGC{(#gAbxdvBNyokp$&T2KY#7-5J|n?ZUk=_
z=D_LvM5podr%za87^Z>K`9yG@+uN1iG{HF$zaM65NrFjMa{AV_P_~WBW##S58^#*K
zVPJkfar7hGWn{UCnsFK}sxI7K?>xPnIgE}{8?T?=`10kAH74ryO6`i{;m9-w(mJtR
z->@xFA2(9m2;;!u2G+hpv((hM-S2$*@|9F8!#vWtGZi?Uj(DRv&I3*>rZG^BW+;u^
z8YMo&zN9;D7<qaB4qRXy1I8#Y8Z(@kj%Tcaa{Eflo1D~&sPalFq?GYSV^DND&S;#s
zcx%{7WvfLZn&W}zk3ZtH;e4Dht)tIp7VEg&8|NQ>$M7F#toPh6ueiFP%Z5Qt?9(vv
zbUt%_erB3yy4EPIF?OWe0e5<$mcnoU_9yOZ;{Lc2QzA@(<MR{8rzdQ1w8x#G6^4eP
zQMLy?uNXgb7>?o|D48%$<S6`h<2^Mc8Ko%DeqcBp@ujfdm6U->rj;zbADWb=kQ}|X
zR^p>(2PJCQVv^|cjswU={+XaP)>^DKSfSa#UYaT?ibc;knpUDktX=4?#t^&)>*&4g
z0+&qF0!-^Pd3Ye{C1sq@b(<0+msT<p2dtKOVx^H<!<3A3L8SCUAPj+A3)?NSEq7$V
zAb<bMvZ1_T7=skPkfpK;Bh8PHBD98XjTUA6l8wR|8R0ess3z_PAFvMS1~&|N?`ip4
zu|gi`GM31w*hMnm3N=QkO>S69cvEOi+ZWiS03AW%zGUI-sQpjf>i@u*KkoFLXv-bz
z46P@uZdf<4<;e2&m5{+t&t%`B%S-W_Y)Y2Q<@@Q}SkE{q*dkj@_-UpG;c4a8sAyv8
zs3l>Xp_NYEHcA#$eRcA)rrmJ%4ejm^XQtx|;q+dXy4#&x8|m$pUJ4<2YA&o>#43Xu
zJk|{;<;bx>&GPTrR^hV^BT758m3=|UnY>1f)^fXvNn)g4uoF^_lxTHM8#!ee3H2`Y
zS8W7Dsi5w})@|LOWx^mMX_^${tohb*mN8>%y93Brj2mcn5JXG=J4klsX~bGX>WcXw
zOOuKsE(@;0FTZ?ZT^HOCxW8S|+AtlCJiUKGm&)Vo7uIcM`O5{>GV^r6hJYSN+~9ft
z!#lDOLzOvBP%~|CglUji=k3Ae%Wse~kK4i@|J$GVdU@lA-~SFjdLH)&&MCU843lRb
zo|z}dID6*f1lqE#E05cQ$L&UbRC0#vEs|=bX-6(^Y_|pVymC6veE5eS7=}P;h3jJ@
z%o9avobHVCfDVR(ed3Wj$~ex)1H(K47B!yaFQpdp{esqxxJE*79L`4`_f=##TOn_&
zaN2`ob`uW88dq|P=-{!VM<L=9;lMPw-^oT?*YE!y|HOws{DZLNQ|9*e8;@WALVo>3
z*&ZlsaC?-o?hF~UrmKRYU;*zvqw@^oNb8x?>A>MIGfq<Salun|VA$i!FX)|6neTTn
z4(~^tcNjIHr$Ab+bR)^4`}M}#<$@A!Hmyi$X&eT0s}wq0&csa)4_z4)ny$sNR(2dY
z6$n92^jWUVwO08emV#EAxNXon<@-ngYOSn*4bolF#t7Ei+FiY>Xtl_*5?RjGoN+sT
zL;E17d1GY;@2o@*Qz2Hto%_R#YM{F;3~Mg(qS{kZa^;t?PL3=0`<=4Cm~zUPJ)=IJ
zPt0MUrA)33GaeXU-s6Xfx;!Yi3pESBsis8il^oMv+#Ces+!QtpU^=#er9q9f8d0r*
zau{n+&7!FCLa3FRrNOL}77kZ0lzoh(wV~N>aJjlK#w)6`->_uK`dAk7`r!0*CLGRa
zA5hbbijlrP=q6A~<#GE&ykCS$8%EiK<P2TB-byu`krCuP$%8MYO4F5YEv9yAm4iW#
z35tv%(|x6;Eao;pqBK;+{u#<>>rf4X_4skdx<K=uDQD91Al<Lx)+mWoWYMov!%l&2
z0~Lks4Na$}OiUZf2K*FIT~i--%DPZv5?0sFXi~;vtH<w@fL1FpWzvpqF@UNC-8<BZ
zs(MFW2yd%&!Csgy2ZFxiFp76L?{GfQkOzO)0)=eaQjy<N&VL6W-ar4xf1<X++kIg=
zovCPA>il-Qp?c#u4Adf7f883bG@f4G@&2dZQDebdPgyRQk_3D=jvPKdG5r1yn3_3w
z#g>xPvki=+qs7j;CY+I~`CtC{Cm!#vJl4eJen%O{$B!QgqvKU|ys8XFaNf>Y=2;3{
zT|3^cHy-zuk}5q{8Cz=@4>PmX91kP8Y)}MvcKNb$oGj+)NQ#Nu+YNhuf!cU?e&YE&
zW9O0F6eSgrnAerT&7zmAk+jJ&Qsu;U`$}((<MWUBK{Cb5&GL4AL)Atpok!WQ3c)&{
zIi8<5O-CAy$K~@b8$r?*>&p4%cl_hO{NFhK;Xg1xy<ql$+tc$0?!$?7I&#1Kg<K=s
zvT^&mk_KrrqxH^P@s!6E7Y6Ef#o5N_EyFNKly)32#*tHD>ju>;+owPC>DPaQ+6cyR
z{`ij4S4cr})p?~a7u0w_(P-O)?d_`+e4NBXONX4JgsGL;Gv^tj?XDn{dA%|o>fUF_
z4Wkvg6k^T<7o@@u&?`ButaoU`fb|aNtaJg2T)`EYqQSaA2tlslO3A3FbuiKh^$Jp3
zpmm%!;N*5{I(s6d3A(v=oH8P1@gpTA(tV+%m0Ak(Ffg9pW5Og~)Yd6+rKgS`B*_x@
z#EtHq(=<ucHYV2V4LhDV9$pycXIz+2ZW5PAt%8HlhH|}eFUvm0Fw&-08n2<LCIGkP
zjcyc9cdWKpEeH>hiW>pr^(wJNCB$0Qq${FoLK}xQj)FlshiNkUs5xPPnk)ISL4j^8
zNn2Joq7EMjhj-L4qRWQ8enGV&hoS2$kLwj<E#q9I<(xLWw-_xC-AE;SC%9=8?5$pb
z>dA?_)lTkR3SL0$8W%?N-slD?Z{?CnsX}gamvE0e!+6C_2b|VSha>gf2UevhX<-O6
z#szvebT^<*@3BumV8V!6uc)+Ow54h*%3NzHRk|6f<oy<#Ni9*Sn7((IkY|&kOP|F$
z8S`k1@gAk5FC(K^g|dcT#F3zs{NGCvm}lFuI<4)`8KqIy)4lXaN)tJeR+<?94(X#G
zfB$<b75nuK=Lh1J_>xyNo#k=o`JDjcwp59w(Mx4?4&Ay4Nn8Lo@bvB-$KyzM7JoQm
ze4yNJv~|Veu!AGliqZ*d3-|k--!2<(Uv3yvapTOE3diGt)+?qJOl&-b0IhQHQcWKR
z$Kf!MQ|8O3uiS1ADn?Gi%JBB-jrG28JRGPRc_<Yne0(djyp%G_V_|)4tdEE;UkQ`v
zcz(w?I70B0+;A@^O7G;}NNeJ;7UK1?D**$}86LIaot#ix698BkMoAtR4THtf=tk4s
zfFA~G%;;P&bjl{D4Mt0&Gg-^|$KUbt(?2jA&Zvgu{ON@Gi2LD3w9b_NApZ6<%iD!k
zD`^OOJzmfb#yHw~XPhUjG0caF-kYGsQ^a{kRnX}4ofEX?MBKI=ThPe2E3^Wku;oa6
zeWh;;W|E(OSyx&~cstNi65nr?g?!oF3tdaZu+~mqSE|bBRB&NL^~}0$WG$zCZ7kj)
zfWam)JU5Oz*v-&#rCTHU@=`Ebb~wg3`MFa<>w<RDg=tNwowZg{O=zX?!^pR=r>YVJ
zP)G~<8**Jx8Ad>9+e*CON%tGHg7$$hPK@Vg%y2{-Piu|5-igNG)kG_i);pm@RIB8=
z(eIUQS%|4}n$K9HvH4pZwt_eG^Fe0Z#-ZIrHwM)++E{26XB1PAJwaR~9oLF9Yqi!p
z%Slo&y~x80Eh!{r9F$B>4_UY(L8;!PJJ4GJI$DFKQ(6|nuIY58rAu;rLe1~#(=%;4
z<J^QRks1@_{)M(aBr)YYa(fkxtx^_a#L3)iC8dp;6T@ykQ%0e4V|%QmTxBex6fKMx
z<!QRn!a>HAK@wig1(I=ex1zXFlXNXgO5}AFy{NJ1*04sQcY~jC4&@E#K%Jg2hiB~J
zwEtZTDY*k(X-Y2WlrY}mrvp_hTB~S{e6DN8DuFxJ9ZaWG2cxlS5R8sC-_vlVHVI5v
zc{b>5#kGJ!_e$xztAf6V19qQ^mCv&fnJsxgz-WbpP5#y|1?K}sca*-fJXUURH$hJ*
zO_&1XG=SEWdf{=sLMf!{jrGeHYFE5|`O2UgZ62AX1J)aIExi8vh3&d<`}LJv8~^yf
z{R_h|5w8ok%fhdpK5@TwKK*v(AO7KqN`qc_cb>>45tl34c|kGvic+0<@(dx6W8~{^
zU-<OvXDZNL;PBx`rk^IZ+R3}DJJm{g6dsS2$#mYgGf&S?OvfXs6xPznDMMa)Iey@9
zkTi&{m0%3vFba35byggo-`=?W^{<$5W;&iw3R10v;P!Nb+@!QJ(!#BsaXvDQ1Kt^i
zAz-beetG5VZ@;o!uk<ePI~$~IH5{KfeEbPFpD<do%@R?US)>^P;dEwvJo9+FQ1vUZ
zR5YT}T-Sw|5`(ilXe}}xj~s?GVGL~RA{dIC=r~HR#C0P{Af7NzQVy#XHF&0{6V^Ju
z{`TO@=P#_6E8#pdP6wQK9F9kZal$$;59THhoK!@WsjSCPM4&=PRblE^RZv>|3C2j$
zM}T4Mvs6&tGkPy+omxo`!C6S(q2ZlpJ{++wK$ECyO7cRo&S6|YYsXr|(iXzVFwQ_1
zH+9Q<4?u}YFRd7@cT6%n3R;kIwKa08gu$YRgG3!e0A*!)Ooh6!-euOTiloo9lDO`D
zjr3Nq!J{;6j|=7T#uQ{%f)BXqh#ig??Qw(njMMFvQWC9-eQt0A-pR-!?x$F+1NE&J
zATLWXv2;3yt}MnWSsJH>Rw5cLx8N>4gx+^fmC`up#X;aLmO<j2)0yG?f#LL?FwByu
zZ_?y0y|Bd<YQj(A?v}KTyx2vrCch_%YI{%}<tOwIaHT>wq(_pNurA>J?nM?)gYn{u
zD9sSH$IS1s^91XaoHOfvK@W9zOh+h|oMbH8wnf|#8kDj)AHWVfGFw4Pl(dm<ceZ7b
z-;0)`L;prVwR#8YNdO@jCv9|~i^mkD)t(OO5(T&7u>k45Xa&j-0_OujllIO+Yn|#m
z-5UW(rWJZZ8HMr#xPhtzO&L@-LK;SRN`Ld0!mq!6=J<5x<A<MkeY^0lfBbhIkBzt{
z-akEYKF&P9d%;-GugfOa)0jxN`}enHm_#l<TFpEL?vDq5{P#ce-~RP4eEIFdG@r!k
z<dBN{+W70|2emqedEzh+XjM3!23gKEk-NlRusUFe19q4wMe}-prK}s5>q0l4-a4u5
zDU5mI<4+$5&hU6Vcsj_)=gY63sP`LNT#4Cnk5OE*VL)-C?A-`rM0W*EqZ~WSzu}B$
zEt&25#_Q_^bz3P-KJ+cEINk8~-ND;LDr0srJUQc!Gl#<@*W_(MtH$wnKQX^NvAn*~
zOJjL^W7#6}JGcK11T3bsoW6DI^1{-FX`0dJ59GKJRcF25Im`po5HJ@@7>|tMfU_EH
zGtOxalV|Xj90O@b6Ep>B5rjysMHHF)ZonvL{yl#5^i(NZ!i2!^?gbMDln?0OC13A6
z<2+HTj6k|F@({3=;5=oyL2Zmd&=U<LN~ax4jeDjanV~w-IkMmE1byMC>niD`)&z%c
z<nP5Gi4W~G9Z&6YTdG}_wBNvd$~wlwiD^Diaw66n?b``?-xp}DMH$&UwJ8D1X-(CJ
z4MTI5d47Sg8_41&qk!IN=|NrZ#LJf*g&<ZsW8F^d>}U#|&i1&GmOG|Z@;u<iGsF86
zc7BH)W^`8sC7y-)cq68T79TV$b#<g7Cvoovj1|#5N>RGn8|Y?VIYZs8cSdO%4M#->
zDfg{g7GRzneFb^06fO)xJ+&D-G+H=ej_<I;0pkKKC3M_K{X#7p_3=hd3#NaQE@>eQ
z_lCAs3XDq8Q>8a0AWY{mhch~ikRu(>*fh;pQYkyQP|S0w!nQ*vwKe$hgdPu640MCi
zLH2HSqqdctEE|#Sac6zp1&!YJf47#gOj>Si<x1ACq%I61*D|)P-<R!1%RZwl8+|Y2
zY2(D$=YpiHXfpE2iMBuB#m6G7s9FX2ol}w#rVFSlpV3yU{3W4@Kyni*Oi(Jl$g~9Q
zzy!Loq|$e!wnbwEbXornJB9VJ5vBvRWiD?w?zffovC&e5obbI6w=7jp(;2)c=arxT
z{3qW1@B<${JaL)_o}Z3PlV|Xb-ba4_yPx>^FQ4|7V&~ug^<QzO2}(9h98M>0_mz+D
zo_T&AQ7TbtXImpQh4GH@IG~)&h#Qt=i(J3l$jgF0otVdo+uIwqR?=<Z`gS8fHcZMq
zpHCb=e!z|a=PlbOht-P5W1(ry$De*jI1TW)vDFRT3h{Qs95i7LdkICd#oM-W|N2IJ
z{VKn--gj%4L;`99kcVlGk;h{JC9;{6FW9`HE~DhEgJV1#I3AwpF>*7G^>JrSk!4%3
z+d|&%<n@Z_BicFf2=|7rQW2y{`e<+i$LWYyMt<+DFby;6WyViOabT~n)HPDooo=lB
z8B=77nNlU5`t5_Aa^jxT4tmpEe*KI~ot`V(V<8-l9OjweM{2FaxDso_D@n1)>{{(w
zLWTANVH{XXrsg7JtZAfcMeUFg66K!Y7S>M?%(kjf=t8?~J2X!ltg(B0+tIsbSyp_j
zIOl0yiQ5+S{h#Trh)pj9lpAQu%b(Fov$vH-&w}0%u93*obLsN(G@30py7dghOqh<8
z+Su+_kp?LvG|;rt;)+HP6W)8eHMn8I4-->Ur1i$*_QtZ@X}O_WX1OZ-ocMV7z`S=O
zbpMtj+GDz{QcEJn%Cank5saGvjougP2<1fP*?Zze?aR$?{^6b@sw8U5<hCb`_9T)q
zLiSWjVYIv;x-yc)nxC+zcW5sGi}dz|bp5j9pcASjtP<GeeciBKp|zkjl+rkdK>Ltl
zns)&=&iMI^ae<<4q$IZkH;l9-o&iCKtgK1?E}hZXAfNM=Gqoqwwn0e}$HmC)x^%X9
z=l1r>a=YS(0UsoSOOw*rdSCeYw=Y~{BjM<+^Wpixk1sREO8iw@gYJ!#CA4C_Cyas&
z*H&)ErR06wWyB?|cRzv^pF_!+R*H~nfnGLh({%MMg{@KDptM05LA8`F`xR>qRUycZ
znn=B%oX7Y<1_(Qr<Zu2`_-}vwGvjZs<dRt*3xjhocv4z0dEscD7*!{{KN5_><Rbi+
z+&Ijh;5Bi(F%Iu!W~gNz`~3WlpFVyd=ar97&#3)s`}KaK)WUHZn1bTNOArv5KXJXU
ze0qEE^6m$Qd6u+5lZNGMypdz%?Q*A;%7+gFY8>EZ(G}i){mS+J5Qvm<9L7l;3ssgo
zN?XoPM`=}RBXd|EFn;8C?9Ahg$c>~OTXLc`pcU@d8#S){?&Awj@2re#yd{@{_M*XT
zty9Y$zpI5(Wmd1X;_KHpetmpG4T0&07oMjhQ*3w)<2doS-_S<!@_fcRO@6%b_*BSR
zJ}eUT?#$ze4USwAm(RcQ`rB{ZZ&!M6WG$zWX_|yWX&pgpC@XDUsWG93{mb6iqj^bU
zjUf=XO=6$c?CxSixkdi+^-t8=7)}RHPd_j{JrVrC<NhF~h;bS_40NT~*6sVk&=`$#
zMu2qFEGL>)cAU2PE?`m0;P+8a@B_wKIgnJDtrj}12?xgeK~hc<)~0soMFIx3mpzW_
z1jF45IcIu<Fa*Ld$!V)=(S{E4fFcRQrqoJq1*f5RN$Av^s6}uFlMVFRST0|2TC=uI
z7STb*IjTtvR9ftYq7=#nVLMF+@m{x#-2&P+N{z%=x!)5u!uaw9H%~B3Xl*ImLQSin
zqQfM66bD&(oEb2Fpf-7dv{v{3Y1FFGYR~(tZ+w0k&B=>AQgTKai<>5Sfi%hGTO)g|
zZ=|cDyNprx892IAwvF4DU%CAGkC<NM!KSs0NQWTw)>?4hGaZgNH$Yd^EQLBBf^e9e
zMX5$BQDjQ0i*hp<tP{B1V@+)9O4pY2S#G}8DEjsQ8!+0^OQe<!MP;6znWiHeMNWmB
zD%+Yt8^-y7(h^_Y;>z3Q!sT*d$qM`kt<h@0Xc%1BKT8(ft9Z6k#5&J7PFOGW;F^;3
zI?V2^*ZU34?6)_q$-6JZ@7`sj<Ti@67~?2a4wVvcsFd|aixJZrZEruz;<;0eluWxF
zFv>_wRs9{xYj4*Fb{I)95!aOuFE1RYfiKIAp;t`0GaY7(>THAAvA>0PPwz=N^ZCzz
z=5z?SE(sy6E!S=3=bwMUS<UZ0KJnA@L{N_3UT<8tg&Ygx0PjA&(CK`ByD^M%rF{9r
zKQSFnLICUq-7;z0NLk+8_HdR4D2DazM$etw<<8}Lr?iF-fsY^FGdw-xh6Bs>#_KO%
ziH{r4PbY$PY^88{eWRecz1=yOMoL8pnMw%;+BsU59C7RwtN_}E;mqwJ_O#YJE{wPk
zSmQ=s?hHmyS1lKm_W1F{7BuC>F`UlW`3PZRYl)OM()C8$A}SYZj-V9R*I($L8+E<!
z;+YJsaz33ozC7W^f$g#Iczxw@yVHPJ#f_Wj_?$SUianmBbTiKQ!(so@D_ZV&->6+<
ztVcUdNC|HZ^EfgcCQ2);mj^X!4&%UdI-*o(yFbteO-iCDJRXl2YuREX=1kcIB}J}J
zyBgJqXlJXWT=7a4aAg;T)Y{NSl&-y2^0rW7LVv?cq_`Kdyred|+v$~BPzgKAT+#Ov
zV;7uB5mAzL`)$Nnx|LA|O43p^NeH=CDSeejv)Uh`t%>Dr444l0$Ax&i5Dd7{V}c{s
zk!@!fwNeGYP%9}G+%RFZCAdMp*iKZGHVjlOrD>FM^wzk2{u$>Q<9udCcv3wr^00EA
z-X`#d=4P@PK|6u!^(+e8{nm=o^t~yJu_zs|x>IV#SdTMaBvz`UT#%(|m!@p3Wsd+W
z`5rXUt=70w%!Q@}&z{#SJ>Drr+{)SqDvhoU{%}MHdImGi=waR!n-OoV{5({0+R)pA
z@PG=&D7-Z^#CuGaj>tTTXWNH?){77YP+~)}n%d>x>!uN9lprY@6dF}2IaDCC_u4z_
zwy{1I(t5}0hPRqhHyJZV#c~ICJTM;5i1(rqjKb}4Yh@(8mCHAVl)O|MqL$T0n(@6W
zjJDVyH#%As1EBX3AP52F2T?hcj>&>|scA*mhH)cNHB9fI49Yr&!xL_P20vnzp{IrP
zKUY)uUj*Vg&AQxpes|<}vUn6Nc6@N$x5U5w{D$=lu~u^He0X_+-qEU&t_!!fH=YhZ
zp?f254<2uC{QTQjZp(%~J~JE#wi<c7-gsOejMk%-;&d7q2FLBO64#9GD`8fI*`Q4o
zG4+zE>w?n_dpt0lKR`3&^^NWIH<tSf%J6z!S(Z({s8itmKmQB!4}YLLM=XD&@$Zr&
z9we*YYp1M*dy4$!ufK7yU|Zq&H1Vt~W|%SBv22x$rduap^j2ZJ(W~4>>@Z=+gUq2*
zqI5(yziljA<@j>Op5HSbPjEajOb7X3wL)2Lr29=uUfU}2b`@B*jqTT;#UFjYleX=<
zN7;MJ`T4}l%X@+0X+b2ERH><=Y$vY^b?x-HVOqil%{&FhdB$7O6F%+@Z4C1vNYKC;
ztTMEmnWh6zFYl=+#=}C*k`z&0SW-$EQjC;Z83sWfr*$PoA%C_ix$)9FR3%0*G+G(*
ze!52~oY7KZa{*<&%!ErP$AxX%$Z5a11X->%qtMo2oTry2ETlbOuXn?kVh8Q@E~zae
zyP4&BL4-*s24nW}kfut7OFYs((5)jyiLo}aM0`!BSBYr0v=SfE9It9-u$FkeQA!nC
zpMrTF3C2=##`}ub3QMIR-UMqcsz-WlGVezcK#$u6r5gG8hIfJB1{ojiOGss`^gQ;a
zZ7)@V4K>LFvP<mLIxiy><LSnu%+A?S9pfEJ2egyIRH+q>Km?5zOmy%4dlMWoIj-b*
zr?0P=@rW*!(RSQ%ByStad)#~kAFzJJdWUL(+&WEJn$pA+Y1;~_)3V)%g$Rk)T3M`i
zIYHZ<-b#mQ9?;&(7z#l`^j7G8#H&oNnMOfbi|O)mS?_lY@_kgbin2Ck*6VFw<OYJD
zm`%XVD=8MXm{3>_AwciMQgNkWtd@R=^?L(e(aYY{SG&J?Z@RbEu+Hw1ATjCDmEbX&
zp=8KiUQWgZLYPr5kb0KIzp*&u(BX(~g2pLn5x;{QvGWu2yC3m~GunC?DU|~o=Ksf2
z{r}IG!XN(mAIR&%`*%;IZDV<RkXz=%?|)#qT=}xB*s5_p;2nf%!ugr)v9N6mRvVnM
z<gM`e^Ur*_T)A$E;rs%_hzk?x?Ts(jU)h$Gmk;l7-Z4yGKvW=&*4|J#G31Qi;Bmk6
zczuJI`01w~I2^@y5SK{I4<4^?<lN}m3LP^Xn4X?FetggA!|$2j|Bmh~s&-V|==WE~
zc_OEkQX==qO3ab-(-WpfZrjFfz47(2@O0leou34#R*G~tTE{ogzSGdmr)S*LM_d?b
zMGy-$W$w2N+g4H56AovFA3os6BTdUyP_;_SnUpi@^^J7-f>WLO>516~ZjX&$fBBWm
z<w`9Ty|+#=mS#Owc`}9)D{o&f{Q1v6vuqn747_{y4BrYptuy7s)9FbHQcbh08!08)
z&K9CIiAFll`RwSe5{7|sI%2}Wlpb8ZeB!<?g8!{W3OC>K)}@d(ng1%KL^DapSP`3R
z>oC^ht^5Aq$yucHs>{h8Yb5H|WlWUo&a^jn*F`$8mWGx_uXc1@G0IT$O4}BUmU4=7
zo>m(vi<&X-E#4eAw7sL}jXKD(S3q!)x-O)=fmXOMG7N*TnM!4F(!`GG!ICqzCbTnb
zQEXpbYf_X}ciJ_RRdvqu1ii9t4{n#Ql#)O-YDsc~Ns;ZbkW!ZJPZgZ9s9wd_Z6-P4
zDy`8<G3?Ts-WoY28Bg_2?^+6r`|MiPhVIZ?qib;j7-V!+YC-kRj@m&k1WMGFTKXdr
z7Lr7ZQjL;U;=0n6CLANfFfg4a7(bGhh1xaR50tKGR9cdom(dpXpC>PO((Q_}nqGF`
zoAGGt(bmdBHcMK{8renEUg)LZc8FfDf*hwcflIPT#5xJ#)f)l+#he+O@P&qHBIPV2
zs0{PV3-0)Yvw_3$&UX7k+~V#Amjjj+dLA)Gd`A?cyNJgdku=2=skI0+bN3+_ZNZ5z
zqbc^=ggh5|75b}j9);{R@{T-_gkZwjfpI?Jrz5&H9&ew?Inj+wBgW$s)9D32&9Ybf
zJ~mW;`<DVW6SU&5fBu!r>jkGBhiA`tk`u5m2l<6L!!cM?a18Un<@VsRu1w=V?v;Q2
z??3ZzfBu<nHGU4bb3&^`EgCTR`9zolAAbCR(-vg}NifYbFE1a6+e%s%Ox!5<J8z$V
z<MFsN8qa#W^YrnFdjCwmE%f_UD4zL&A~OcZbbQatfBerJ-oM8j-^*3jI}T50Zcp#H
z-d|aF@p#h~KN(J8=J|Z0=9TSs;nOdF<+ryRm(&=)T)`OpG;ugTVSNy;f6c@knXQ2T
zj34OT5braKx=@-XcZJz!tz8K|IY&W9n-0BY;{A=d-6i>AB(Y(<r&O@U$oy1`jj2#*
z+AyDvoL*j-PA8W8%H_WBc3Vlcuw^-^Ow&7D7;wWVDFtmQ4KBAkX_Hf}vH{)dZsn@`
zwO1SgT`8KXP$G4WT))1NQp8y+(a$R9YXzhj+13>wBr1&B#V&S_m0BmlcCBzml11BQ
zXeFYI$jP-4Y<f&dB;>V9xlJ2dZTtVe%jw7(X|sBRjt^RmB2Q0}@$Pg??`%m*M!7>P
zQjBW%J9fUI{Xj?4;!4>T^5afgu2^e1JsE5eCQ-NY0PZ<bO2qWWbQlQpz~gZzmrC51
z^KxPr)I!ObhR|ADDXhzl-uI??Ypm;o?eXAoxl(f3X|G_6$2kGv8EwgFlf8fwkFjz4
ziLlFUNLfjDDJ7c5-Z$urps}|i<~i-bdc3u$mZ@nK<C*h-$Y8!Pn0l)yEsXxw>n<)T
zDAo6fYbW%~q$M$E+)Rm?oExQ9uGbsu?TwN$#wfxx$<CpATHd64vKB%U8B?$hRWo(V
zq__f-g6cldy6l5=m9R%{h4n5<P6ePYC=AUwqLmRST4Sq#8G};{qx{ZN&)_|Fd}4a~
z5jRihT5vVP7EnmM^V=>;DM%#|v$VQRR~V(Fmt?I(clX;{-%%?q7i=)})}&14gCy5f
zqw7ky`<_M{de@XxX=S1B20UjiCIrTL!jB_$O>_)RX^fxo<3T(U)(O6)cZnGrBYU>L
z^-JOU^RHM{`Fve@yKg)_y>M=t^|7K9JUyM+w#2q=bgxCoZXNUa#QRbi#(_Eqe*OC3
zx9h_3bd)$<sg%cq7z^8V!Pdg_ab_F?t!uPVoR0^b_h@aoy}k1FufOuR+)y1pzB}W7
zdMA4dpe=W*?$kAt9*J>0a9i$pYp}z_=|B7vFaP=fj-O5#tLaJ-1ie=pigk^YH8ahV
zK;5iEmyLOHOyhz0e4qk<{p%;LkI2tIU(r_a5C8lROizt@bZ7;S+nsf}W2_~NGkr*C
z8oCvNH<)Q)eO$TTZn!U>d74k?BYSzJkoOsF{QQ}8dBf&JN{ZFqaaD<grgx@sBxk`*
z2W!bG3QdvDwl2i=fiv*@bdV1;jo=gpfzw$9q>c6Rik>HoH8SGB5QYgGAUz(GlF-qT
zt0XB}F05M9d*k~0mHX|^Fpel~_T%MV*VhKE5RIe{B5T)La_Bn|RT;OpamCZ%oMRkE
zN-{J=*V%d`C$W<mrNkMa%#OT}k&FoeKg<|qu{sDwdgsGnjm(wzFDg4!O@MEXrX#%<
zmb9VnUuf1!dvsf=+g(a$j|amLNb5pLB6(^#5w{iMLNf?i(|OUYI?dzlM%*GXMc5NV
z%1G<j><<^`#nVgc^c~BQibQ|qb}Z>HXD#C}Fdq&~<A^a@(o7)p>3vjDTW4}3W{^~q
zb_31GEsC5V-l)qfC2s<6G_tgH-qMO9Aj`~FD=kO)JsLyZUjjxOiP!Fb&ql#G1bhe>
zt(XTvIhfWlN}S-<H6D*E+x^Dv%Vz*wuvo+1^~hM=87GGgXCb*7Ew>&GDQ~nU58#v%
z%d*H#wst5@0E?-xuA5*Zj1k*kH&h=8&+pKFX00&ijjo#zubs^Jy)oeVj=1Q_?1zCX
zIhTDJkQpZ<Cbm`uZ|)@Jh8CoXjFr+(%9Z|r*W$u%JM=K6h;kEJdy0|t8M}<9mQ1S|
zQ{^z$_C2KuqXZs+QYz)Xu-zZT80BVct(04p>~eE$U^Qxxo1OgO{>Cqb<+1U6^7JM~
zsq5{+F(?kx02FF0<dj8pQFbxBGc>d$36w^MfqS~3y{A@q{d{L#GTvFPZx=L`A6}ku
zN)zuZDmZl091lkh$0MzGZeKgs>y_Vr`Gt8J`NKc{j*lN-=sENH`NG@lm8X}pa6+sj
z=7_Nh=PlM5-oHFCjRxhwDT(ntUN2mJ`z!00PdMW!j}<j(O6xqXZ}=3s2h9|IAOykx
zHuZ&VTS=ue4PLIO+eWsQ?XeJVcY4h{zI<X$<-3;A7#N(Rd&hV>QGH@51^wkSd0Rxh
zO{cZOdb^Q6|AuW@Mqh))IZw`sby-N;iqcw^SqkR)!1Q#cX-Ddc%h#`L*Eeb|%wBUi
z4U#%?MhaQIkRB`d^x&ZtynEsF{6w&psvIVGw3fnMi3@FcpiuM%k1cY4`zkcmRtaO^
za5@wBrf5nDnqZEN(S*V8sU`W~8ZC?Vx(i3Ts)A+CnYJgBa*o6pg-n}@jI?^^c$f%7
zAhk>>;)O<Q2BT>;)3yie`T!K87i_dLqTK9gT8CT<H5ICcrlbp@eSjQ^+lsn=!6+p!
zfNiB$Vd3<(3bfF9P|#DN#D%hM#8$95<L4RUJzl}mA+Dk&9jB3AI_t8Kb3xTEe&y#U
z#&M$7LfXYj#(9qK2Ik|LZP|ESE@Iak1JgJ#3?Aqx<%wBdjHw|vPYL~@td`&NFksw-
zwH{R}td^}Lc}W<D@dGZ57^ekofuhwY_=?>-3?k_<R)4?YIb+3Xtz-;Xdqq2gbpyRA
zin6F~aYoaTorbEJVX%zjNEipgFksC$pMWBa1NQtAZkD8vi3>G8NVQUPrPj*hvGBOx
zfsQto-YVOg_FI_6IE&JjijnbC2u$-#m}ZvSmF4vfrz?83q;Wv|f$bYU<9=azJkY_>
zQliAjRy&Va$W`(53{OweOHfLHkG(^T8?`2SlRt;<lFYN(%kNYfx~imYqbZF)1d$|Z
z3oghlrgu6Ey%ea0S~tOA7kLpXH(*^O=0@pCKzmBzhCpr7eQG&V(h41n(J16_-t+pm
ze<^(07M?!NO!I*?-N|X?_3gsCuFP{lv(opOT`-RO<%$~wK-O#J@^)q13i0)V?QnU!
za=YE}K5#k=j1HdP9hpxftu)r_1M3~7R%`=@V%t`3pT80x8@+diF-R(4d2qWu`0s!I
zMAshU0`qAQ@%=SoduO@b2^ns`{laj5!JaGK$Sj*u$oCuR^$TyWU&xO;-g|~o*hwmG
zeEs!Tsl;2u{r=dK5SEWW4!pdanGXX7#b5vJXC9Y@ch56V&nMoNjX(a^|4PR(4<qBj
zb2^>yVPt;)lPE)tBX1AZ<;MN%E94{(%pwnNKX|4wV1vg_6HW`k^6mAF+8WL|hGAeh
zo_P2Bf5e6ZDP=C7e&MxLK}#S}KJ8#QjHD>c8{;e~Wvn&Clz1!;ybBo1cSaHW_0t$A
zrR{*N$hzFIRx<x?5G;VI2IDM24-zll^U)##DO9a#JL<QUOiYQ6!dQWYfpA)ubs@%$
zlp<Rc;99G4TDDe8yLOk#bOdLyZi-YJ#t5%oD^LvRVG=jC5!XZ7<Ylw2E2TBG8zkY=
z<U#Gw^t8z7J8!hQ?TuccpKfuX^(HBqUVuh!mHT}`Z4tAqyBtWdZ2~Zy4iod?K#sCt
zZ#!x^_`nb*I+CE$WRz(PUfc&7g}vVKTC|bc2q3LiAzwPaNt%!dwRQXNI#G%j7@jrY
zJq?RRb{a}`yboe6bRMTWx;1pK<RUM0<E;E$<-WV84Wl1qm$BdWlu}roCCt!e&r`G`
z8&5Nis<oiV;|irPPb0(NabEP2#_lf!rBPUkfX~mEd8Vh0_}IuxB*%m@6UKmRT3icz
z0;*X{&y9w{xqu%h+z>E;(?Z2H6m(h$RI(D9>AGc!7dIulJmV<a#{K?A(g1=-$-A;M
zOb7fpiMdeA_>9(Qy|Lb|^zDwf8tp8Db30*IZhn$^<9h@+t}AQRG`HfWi7=kf3VPa5
z+k;XzT8$X1QQDAGL3e{um6Y$SwW5_kCENbOE=ABZ3Q?3M)p4s#JL2+pcqx=p=u{5R
z6R{OqvN)&3%tfQ5%HSN%bd-?~UD^`Cd5(t@tyE~8!!+~ta;KDpcZOi${nN<#I5Ll(
z$vbk7tlNsMk<vTA{g${t7S_1&`uUBw%N_3>r+Grtxm|Aj_0xmDe!6lxPPBT)S<iHK
zbS$60+*xjgZM))r`wPeSFU(WGgaK_FVGt$cr=D5Xh_0O>1m1uA0plH4`!lz{{E4q`
zSL`6xvFSK5O&vc4!X$`=*Dn|T@|R!8sWJqO9|P-_pj(W#7-dLn;h|S_7<oRQapQy^
z21-2f_W3i9uWx+$^o6uMIG+xThd}EBzEZV7%M9a4?}qc!nbHN35vBvf;mo^_zr)T?
z^xm1r0hc2+uB>q(#uZwj<b<&vYrLG)ot*UDG^6d#MydtUCI%=P%4)P#*dPz8s^uiA
zwWL<G@~Hj|4(Bm?Mk`n!H+0YP<!z0$OV6oEh@rX4X{R=@vR6n&7VM=oQj}A0*~dLi
zby97pl+jj6BpF5Tnwl$0H9%h2)@phyl(^uHL1`~o<|?GpZC!y5QLx%|FE`n4=tbeY
z#}31uLK4h2HOud>H?%p>(1QFPEm}+Jsn#s%u9%=3hGD>u6Ih2;iX#VlYs|yQaURjV
z()TvCwIf|?T8X$WC={m^(HeT~QmjH=tlM%Y?|$KKJ=)5?Mmu&Mkwe)5ltcH9svBy%
zl9sES!nMN4R2mTLo80cC4U3j?qOn%=k#xeK@m3OET^VQvTe7s))j$W<Zs<0U?L_RF
zUas`gsl5`M6EkF~^qMF+OY{<i*Q)|8XF@K!#HWFA4D$&yzu<jf*jB>fMA-zpUg`Mp
zKui_u2gbtz>nvt>Mo8}^q1Kf_LpTOn5%64h4y@c_OP3O)odzkd8Ntxl*71{OI-HoN
z33lm?^Hx$$IihMI$AVFEcr$Lmda>KpCd+u^N0bS~R9Ws1((8q)U^ty<PZ1vm)b@bw
zPTft78VA~9y+nvB!VIgWqOF$~b`*%A>75Lu<sy3FUFc)26ng*WF9mB2?|=BnIG@1C
zOn<TtV<E=ORx4?}^X}yc8m4h#T~@TVoKHub55n8Gnx^6D=}2&f6jxqON4(QeD%<@b
z3?Hqf8EBp0Wm_YaE)Pei7)FN+mTnBuSyrVm!@zJD2t%M&#XKMA8LTs`Su)ngmL)#<
zxYLcJ+KBCq;~Y4h5B&D?uh2T$1g80f^#=X?jEM`z8?+yIo({O+*_K7zw6*i)?F*0F
zgK_dGo47n4tSPdVhH?jn;|c35wPvQ%ne+J>>l{{ToH8gMI2@0>ynl~wh2?r>2o~oI
z;?Ago7)Xh%R}6}m^MPEOxR6IVfoUqF2fY;Pa)TPNsxt@AakSFDbPjDtQp&`wVT0pv
zn8lTBw9q_LrfnP6dsMHal&HDm=ZWA)7zV1f7;E-xsu7ojHVmhUVK`HpM!V0X<&HIS
z^AL!ehT@1uQyhZ8rVG5zY03LLN6i`Q9QAU+I!Ew+zXfPg-jNzw$?REMj5gm3SScn-
zYQ)|ctwoO$WAG@grGc*0-jeQ;`stlkGck+Mqyy7DGfXqi%dMwWL5j$HN0#VaX$dP#
zXVlz8KHnSZaU(8^{Qo*fGm`$9=O^q*QcaGIHJVZ)4J{ZBKjXBf()Kb^-T`S+`tm`F
zd$mN?wa`nHnYLFro#|GDE<F|Mw&AvjcN>&U+8!)le<N=X407uVzV9$PF~qehr*Eyb
z#8eF;nQ4`pHqw&h`w_s{0JYFc7V5C|C}S~2V2in|g39iim=k%sqjI8TP|h+A2aF%F
z&e1#QDkZo6_`;TMf}BVzbYXTmtq6M~KY2l)s4nsxKWe6K(AMKaKpRC%5y@0pT282)
zjn-5r_KtA_VH}0wWDL5?&!kh@ccONe{A7$W1fx(TlGi&WM~oUpghS)LTj4x=Q$NUO
zxR<=6$2F}fYOln+al7Am-0v7?1b0yzWf-t@Tvr%3>>&(E*BLDY<l4a)BZX5sX}yca
zlunI>xGdBbN$VmsXe&j)l<VLArJ%IK`<XEecw@NNN^TlsJSncE+(}#Jy|IkrOlt6N
zfBo;AOXYYz(UfM58^_a$)ANbAZfM<UF<?<Vu2))4jKe?(f!<(?1(h3JX;Mj)T(RCU
zO(V3*^|qm=Blg`3AHoyU>4Mh+nYz4PDYavCXFLp~Pl{!$_^)^LAOFPp)0O8B?-|b%
z(>&r$W1c3a!BVY);AkCc&gh;wJ{`%mN@l%P3>s%F4$bv)<$k*{9}m2|e?e*Z`gZ5-
z(-&gV98O1Ge*8VdJfM2zbUbiAOJU3yNzl|LP2BUlXU1Sz?-%GzY-FuqdfQK|jTm>f
zRO_UaX*T0$jk208t?+n|9}BnNej%+7td#_Ub4Cb~R&vtDIl5M|>~tQqBA2?y4;{5s
z(kfrd7&qxnluVckZI8USF6koE72RsAQ+Rv8`heG-lJ;1pl9Q-024k(DC5&S_p3&A3
zmm6$R@D0WgyrWs+#D`&IT~}Id7;ACP3Cu9c6}~n=%R{~OhQgq%M(MsQfBW}QN3EH%
zv*o=PYU<9r+>=06elWc+z_s3^w1S0DfHD<p<XR7MW2!qrqIYE95adOomBtLR=neDC
z>2wwp`1L}~;;`;rig3nA!o}HuGx|FxM{CP;2oR*uRCA)&hPERagy^}8lT;V&W3P?+
zcyQwv+ImCvPTCf>^@{2hYYn~0LfBd(w~bnXE{$6yH^5d3B}p+fKJKWV33I}X4Q(rw
zjkew~3WjmQ4<oHWPLW(!umZ|d-<#`!Hag5O3${AUsGUugw>#x=p+3G)x_nNpHb9}*
zM!J5bt#@Lp0wf$qDYG^4T!>?!ip0KwqhWWOA3Ghk)l6&RwhrScn0mc0VwF&m=}@(R
z8gt1|i=_T)BrlP4ze^<6b+i^Ih*m;qXnT0VS%))*R3r346>Wvs-<s?~{2<g?-8(8R
zC|%Khz=TOgg2G3a&uEn<J{W@cJJ_(0y1@Nf7s5jAyNi4W?^z#k;|TumAlZpq;!l74
z3&+8;+!uOo-1Exo+Z!FlwkC|;+onpEM&o=SH)z^$e15@i8!kAkHEfRscBwTr6T-l4
zUByP`29`B(eTcnlnr9KA><=M~=f2!|d%V#OFT8tqU_3Z<t#U0z@#}A&8Sg7U{`i6E
zbVm7Cu5VZF|MrRBetYBPH1osnKJxT(;`nsraFB=o;r)}a_|GF-&6wSeWQ(K;Qpr2f
zvN0H8*jw-M!@%>)6aV;+KViM$@c9P0vSiKkPru`*fBI*}(~NBedSP%*9vFK{LTS+<
zj?;uu9UlbAtb51wMz5Kigzl-0MLCaB9qT3z&mTCwyvI$_h>h3V&i(HUha+Ja<pZ7)
zxfTXDF`QmNYs|K>tt(Y&au>(4ZNh9~pG_MrC!LZS`5}-$Z#>qILdIPRkH<r1(iWye
z;&AGOX+X7vp;EVvWnFOAVf`T2Zf&vJ5~c&zXfjtyj>0t33Wdfw3u>U&OiJbZtlntD
zJRiS*A;ftrUmT~sJb?CAz229nHJ9&u3XlR+LrH8CxjlUIDNA153Bp8M!r<kGRWp5C
z1&3_3Btx_#(<wR1zDpZg%_th(2&YGfk?C+?nh(U~A-~U#U_AElZm&)7)LqcgCGYe9
z{l%@Uq8Uxgl}1H5jdRm?_K~xi!P`Bx;)p4+ynP{EJ_~ZDHMCPZeN(~=wI)hYDAbOq
zkU97_ynD@AX5yV4xh*#swSO6Fu9O%lDUx$$ZXJvlrEAWVZ53UnZj9rI$Iz5V4-<CB
zk<?bH=pEqJXidTfs!PnaRvCHbbQ2SubsU~w2=f6GJRQ%vEW~vMBt8xDf~Yis4SMOO
zl-6j<FdYWQSz`3HcX{Bp#+D)}Mj4ALXeFW11f#L7QuDSSAV+*~bQ)XU@K)p82wF^m
zeMi%$n8_*Ax~6nZt^x@0<A@(ej8+7zF}lk*-Wc`{K2}?lma<)^V-+~(WVzlOxdFyX
z%+-2)5D;OG56bqyqR_)cm`2vWhfm>p+xRd4<zIO>I}Xzb#t}>7^Osl7=VwyPOd~*L
z*&=?J`IrCgKlAl^A=SpyyBCIWBIk%TiqqlD{c<C=#C!<f`70$C));xcT)96MNh1W$
zIE-l3xvxoT<-<r%nUX8<a$|d3sN2HiN2%FY*m5D(j&Y9D)5H=hN_l?x;e`)BJaIhD
zXsy{|qZ290F`S=pK~u>rmrqbCak<kU3#Ar}mXlz~m7+T1DR6#y&;0a+Gq7$O&wl1S
z4gC1Kk0f*8=@0*rr;k4p21m;as>;b<*z`LowN+?!XU{Z@A23>>vhaM25nQp}p;}na
z;dtQP`Gw=l2kbC|f*hmdo9z`nk9Z|`jnX=q1zjrEkA!i;4FkiWasB$ra(~ciBEso~
z99QDDVy(dqN4cqtN5PXL4+P<`NW5}7JmIXT<dw8;e7XJ!H#vzZ(uLGmGFyuDHA2ZK
z>#<%I=ep%RfuhNIp;xdbNUq!{X+;CIRC0>cTA5ELtefP^+oS-+epBeksjQ|%Q{Z+a
zvNWo*@aQ|(A?&ufMz7K?{ibv5U80y_oC&jFI{LQCe{T~)tZ{~BJ;8|nb6Hkt7Hi8e
z%}nPPY#4?0R3Wb$d6RugZ;jsd_mLQ_k?vpie7o4P8gdhe<qkJ_AhsgjWoyL+5m)j2
zIt6;a!qwU+aigS2Yz?b4#%jC~UJpuBYbT|+|D56?*GihiG(nG)QuY^;#ts43^>^-n
z>zz`?k-a`v)^#K1f*B^dazesw9p}uh>wVfi!sa^}G(WBqUhrpmnS>)s2XQ)-D!+p)
zbkW`#Rx6AVRJqog5GKj=$Bjr429eeDrbuz66iwIx)VUN&ZCL9C>F$D<4EF?eP8-Ye
zAjiZo?IuCzF-qZ`m1TL)NcBU@;dV=GkA<SN^kQhN%SuUEMsO&U@@Q)?wV^7QpmD=M
z@K#2jK&c6(#M6-DW5>B`ob_OZT&s+w?~Ux*B!Q&;B;TX?ObAD@W3E@q!(n?v1=(%n
z{y+De|6h73JW{1Qv0`B?TQ0oa7OWf4+HyF~oR0>rJ6~Vl&~D}2W$B19vN(*3lx@5x
zrIolY9OeUCN>B^)>B!@HCv9172gb;i7)4AGr4=DKbnt}Q1fAR>%jdt6W2ELx>I*{{
znGZ+w;6=Hp3LoCj;yTTSmyc(jetOS19&mQRYA;S%D%;~uDK}bPA!W9E;=Zll>77n%
z;#OF0vdy0<oK8=GX5BW-<w`FRkKr&+)WbXG=O+PsAt!F#<RsWyrBpeU#$~0Xjr`3e
zY`x&+QKapG>-{@WuN1kij^n_5JTVND%&ePm{OFz1<aE(Hq!ekY@26)iu8YS_5>(bf
zNo@Bk&RK>rKyTuYEs5F<wpFxsa_caPmNu5R3u#?(NKtD(eMH-lxL#4)h4@w6*S!ff
zvNJL(vpZ}n-N^&(aeJed#NaK;IXMAsQs5)Vl~<`Ir|u?8w$^r6vV!?=ka1K_tP4A{
zem`0FP8B?kj1F4E8W}&8ElO<J7{+;)kwGrPTFNqe&MDEp6_VVJ+b)H?Mbi49l}I&?
zq*t^LXytIj0d1V%r^kumaNbc6;)&3$<FsaLJM*cD?h+I^MM}KNOGO!)l0*`8aw_zk
zuu5(q&2&*yrbsVoM>cHKyzK~zM#};gR9#V3-!T?iIw5?6`AK_O5G)zvob)HEcqa-?
zz8m|EMPtx(a!qV$k66}*GKQQpx7&lhWiXyFjhxOWrt^uz(}`{bVQqS6yS<WbD><&z
zQo%pb{Q+fyB<{N4<IC~}CCSOV$#>o<g^P)97TW$&rDb$)bS*fJUK-5`&DC0gQpQbg
z0LDp#Jm-Yc27@NIEH}gsLkLn1bQTmlex=GjQ?sAuJ3S>HkB6|$yrn8d|7M4)jj+2k
zw3eG_qhh^0L!9@FgHV=J&g4zW=&2S|$<p#}1=Foe2C%YI*p<AMj<O!@26Py(!zj!y
zCD;~i6{bNckxrwhzsI6kHLQVQ81ccdv@2Uq9Hxn7UB%&SJVAS|>z(`Sh0FEA#~(g0
zyY~<`O3FMg4=&4B9`}V_GqlFCCMIuiXN?;JMR(qvXO1zkZX4T{3B!akU3mB7MBXB$
ziKzRYBF=cWJqj46z&s8p<EX9To#8kiDbJm+mko+yxjY!gBi&k>Qq)%P!QxI6c~dOc
z&g<)EE^k+CaHL!~osKj#ua}!t;)|H4j0FE^DUlb$_4AF7AK!EM_!B(+Aj?{WsH$3#
znQz$z-*F>n;Q{@RS)iP1?6(uyj!HIryR@~8#-W}4p3_!&5wZKSQEp#pDazuqBhCS(
zP}X6B!zd*S>wQkY#w3nj8U#75D{XP%0LDl;t~QFF@OEFySwm?nDi#6*cDtiZq~(oV
zGtDbZ`-WJMsl9wTw6OiVMt5jh?`l*{%bOI3cB@sc1&xeRN|hI=>^U~VAd9#VBv#ye
z#d(KT4yy#kTuZ^(fNIj*WtYeVrRheZn{IbGDDSAb()Xrq&5=x{l}>s*@Q9pdef`Ya
zFF&Jtq0R%3S}ACZKM)RQ+<26OQ?013DP3WDlkd;o`c1u3T9uY-YvirbyOH*-8Zh1o
zY*8ukK|4ceflVigmez`)LQSZB*kBCwCZ#2<H7<Cj5ac@sXe=>hN!2u{$e2v+>?vc-
zuEdnZb?fYL;)W_3T9O!N+nv7^ijh$U#)DO8JJ75_85w<y<G|_tGxPaH3Y@#IK;?`I
zBkgu0w!L-Riu^mYp(2Y_r3`&{5Ok9S2&c6GDDPL{`fLfTMJvH_fH14N)i@<1quM2w
zYxmoxv5u58N_CtU%zRCe81LBNFj}NHHD_U~=}PX6iV|Y7mI7q1iEa$@X`=5j@!C3T
zOp*fgBi04EnRYp(IV-x*aU3y9l(a(gmD7vT4($b!=l!sw!NoTrSene1Bg#5db?7bQ
zTBmD|9vVtjblY}MgY4L9T8Ycw;ice==IJm(EnIIaOUg{sv@<>um&=vO9|-RcJRdzd
zMt=FXFXTmGt0DYk8Jy={8bAO1Gg?`+g~z8~=~c7rt;XZ?SvdDq7CqW%hSBr<az+Qw
z@>s}AMAt@J6M5G{=DcFG;&?i7I2<4Zln;ALR>+V(1gtCEmxb%&#^bSYdO9-)!S)Wv
z1Lq&!GfoEzmhJ16YpeKb2v&+cb&qtaUh#tzqT;rR2%g6Ew$WS19?#gWnXHqBVJR3T
z=!#l};@G+pH$*IyWuY}WsnUhKX|&oi$`PkJ-C6lYBYQC!yA-zTSC;En9&caS9t)S-
zl^7%AG%*Z8mb)%sf+WBI@x`_#zhkXr>00F!oRutlhad*HRs}JT-agY@kW+ZgkQ=FF
zw(E^gfBpsM9p{~$Wm8*SxUl<_;!wN<r^Q_YC3s7L;_mLntw^A_OQ5(lP$am!6qi78
zcMtCFl#}<J^9#=Yv}e}Ln*C|*b=!3@*+k7Hg_`JYGw9gnXj?cs;$*j@?CZ(8B4co~
z8!?qXV6c3AxQd;tdP#j3BbSoLJ3`?CQ4mw1E(Z(USbOgw$qaVEDaVejl-gRCesm)_
z5FAxeCVNI?_=7tWs^zV0Z?MNwy=t`2PXNaRQgO$eGSDA$3`$Iqc++j;HXg+f0E)@G
zaO>a!W^AE4gvI=IH0KD2@C);Ab&fn*Ln4l^xv_2zEZ(2UoJ&l8JuR36iZF^s)yZ1T
zk-T=|D0cZ}oO}=ROf{*;k{cAMW8OvA(bGm%7W2M@{LaEpr(B7vo<-R!dHqVQMAM9v
zth=f$Q}SI%j?e%gPlhJH@dL%2*3B2wz!K}F7$d@F&lZ>UpzIHEYIJZ{km5z>TK;u4
zaoNYaXT~YddFV%B>Gs=GS>CqOI7t(hWybRSBb`K;yGHq&T^qT#&Vt__V(~r;L)I8X
zG~0Vt00#KRUyG4rN#qb}oBDY9o_n-+lX`oGK*%P#Hr1@VsbOkc$18UMEOI-H(Z5+`
zji;SoN=XKrMN<rQ=TN0y3z>SIUHO=a;aD%(D8{o*MzbW>iT%6b3N^ts;-3IDz#SAJ
zc~;QaD`ccal@ANK5FrcuIo7mu@7;FU|8l4dOS|i#@e9!s%v9P!)8vPH#SzWa;LDp2
zCui`oRNAu;1oi8>B&q#zo|CumGSEp4rx$@p+4XTl6vERtkXG7stMYX+z0Y~`xt~@W
zu9XY~Qu<L7-<4eMId_gS5yrXzFvr3WHSxC(4^{meZceS9Xqu(#Ubm!k8_sXab>|{@
z-+Fp0*9USip0hQ=McyP;*D>-scqGFrNYru*RyJB1eTK4#4-7@-tQ9$8(y%jQz#W5>
zyi&7`V<Humy_S}LRwGq+MpteA!;ZaBvo=)eZ}yFj&{8-pg325YiaPKKU5e!4QO#ga
zD@3#1UXzq_IAUa%?+MPyoPKb3k&fG()l1P7!FtUPbzU=oGvr|Yk^BCYb65IE(*wi0
zjH*g6UY@{1B|)&+{oO7(zf3jvn+2)lm4<xl@oZYSh~4&gQ%IMawaxE%Pm@MfSq0Vc
z0WGqIenva7LuadOw-~!nYHe@5G<$PD&+f6WH8iS=iP-jsU&iqm1|9RF_}ZM-V>$tr
zU8UO|79xw0RGgrnzmzu@#a?13ez+}gAL7=#%Pl%{{Pin!ur{wMAgF^`LnyWxv?xAn
zDkb)*GyM9=*r&h}Z*U_eidzcprTX#p_M+{npnfHh{ooJPh)(oAb*(vNhT0!&oeznI
z$9TF715O%=zf_E5he(`=OWrgqYO|0#9!jXI=_Og{<vK*_Ngk<t*rep-lzsjs%Wfru
zRe5aAUHfDGJKKlW38ik#a{0y*Y^*%GLKlY++DAoqT{U`6flUObr6gOpFd0AjIhdl6
z&TQ*d;w?&a^~LKb9@U#*dArwZ#I;laTozf=iI&f_?&XP_-H03MGQ}40ANa63#-_Bc
z$>ecdA==vfw{3ZPi4+<O(GZ5jIc%e^tztZ2uC*mjNR?CGLC3xP20pCfr2juBN{{&P
zz}gE`icAI_MK&^05oK|lrz-@}%XY`VNbwh+?X{Q3N)aj)xQx^K&Aj9iSSTl%QNL%M
zeb$@Wcy0a!uk1VMRJFX?fQE973e@JG<L;&2&`(v5G9tZ;DvmF*2#`Cp|Ao(xdEmyO
zy@e$I&#a_L=G04Ya*O_)za)~wb9=qWIyy#k4j-Syl0&xJ>@iJ+9mg~++Z;CDT*|1U
z31>P_#c$lxEt0E^z}yqVwWu%=kV}bc_VR4oKG4rBM%7ST)HeJ%bQZ?ChPy8A(EmzJ
z52+9qrZ+bap;eqn5kv(5b(W&_wvLf_dX-f=I9L94K3m45n*)RU1c{vdjaXrfgYWo%
zr#N2fTkbC^Gpu=_)A62whQM6D7B;9y(>XH>0Zu+^gY7(_hTR#XNuexU%Ui0)Z|0?-
z>*%ULRmrqIwTOLuVZLF&RsT~D_n^BFB+!%`^9M{>TH`4Cp^nLe-H<}g34}Weu#Pcc
z&|0!xSP%;0_$X~L)TT)`0G2;P(fX`^YX7tH>bgtB_K7=J0zWWnGAS)0m2KNEHT9PI
z6z6^C(2u7npeaZO+$rQ*QFQ*=rtdb4_br0)Z$T<M-eZP}Z%>*8K!_Se^zz<?E=9Ty
zpD`Q@WiduaRMie{CtDFuGk!!JZa=*m>S~L#Dg^PI<eK1J<*kqMPHtanwV=<7r_XSI
z`)w;;)nmYT<%8Zg-CXi6P@PkBd1hwa%83fwDmqKicFo#r;rW0}z80TDe`59+YRlhq
zb&xu!%WJPCmd1=&8kA97*zBHg;LR^H4GNA6vQM6oE~yDNRAufuPM(Jb`8x_q6wUNd
zer669nwyZ{kbcR#FQn*SPEG2h1bW^HG$_8xuuQem-9?JN{*v1+M+^lW3=CCc^MJ}r
z>3`KRtJ{@N&sxi2%Po&cgwO2!_zHtzu$&{qw8KdhCH)Lk!_p*;HC7G`mPQTJEbGM=
z_J=+H<9u$^q43{+Tlf8V%M}8HHN2yN$L)kp+I6=$N*iS#-f`P}DIR{j1*XMwRkEcX
zh8FG;nUYr(<L^XrkJQjs#CTm~hsJyfe-D0p8R8mihD>JP{)C*0<+jD+$sR7Z`vi*G
zdaQlC^4(~VbmTD)i@d+bk%qq`M+P2NrQw0@bCNCZ@!v9)yt~xX4%1*bwKLuQ7o7fj
zR*zaRK|N@ae%8kGE(IH>kaXDxt^~Y_J;6QX%m1vPn#SYE^zj{Y)wi`<*Qf&!0yD9@
zNs^1C(SJYN+}Gq|ikh&?(tLyF&Trg03P4W;^>2Whso_b&sGe2>hiPkSX%$)Lo5(jG
zC%AGY37B)q(#M4I@IM8-I&Q_r=?5ag-lNE=fM-RO*KO5(iek%}7`1R*W~x27X?U<j
z_evNifnv^%gR&c08J;jk_9@t&CtxN4mG&FX?BD&h3={;nS$#^M!RenWE0TglHERw;
zGxQ^wC1?nwB*Dz#T7n%c8R>u@tFo%4w#+;jQH&$S6?CaXjm^psIK&@Q5;c$Ac-(~;
zO0Mj93geYYoYjR=oXu|Z(@ou@AkKAp_ZE?`jM|j`<I=NI+x!{y?eoqpRH*q`#yWWl
z<U<~_0{yZ_UXs#ftdLM$GHHyhW|W281SY^AR;{ubTz6Umbq9Yt|FI}*hayeuE4}~R
zMQ?~&7#cabzgctk-}37CN~n_HU)MS7U{~qZ?my8>OUQPG9{rSnPh+k`<D+IzwEk;Y
zN%8t&K&=}iy@3ISzZU}6Bz6#d#rm8QPC>ujb7^&Gmuw6LY7LisdXvM&@qz1I98b)v
zvlfZXIzxa>4<Dl6r6cgD334y9AE(afkT5A{^CcO=AGV0aBc<BO%1~3lxPlQ5>uB($
zWWjZIG1Z2hmp}Z=UHF5%wT(|GG97J*e#8l?jGz!Db!|~i4R!Yl_z0?WF*3Rs9L%xy
zWk;HayuIRu1kBv&?mBuDO{u?b4E$F5lUfI)W*nZ1H^N>rH^o9b<!2l&O7MDohy$f8
z0=~uEQlo-*k+sKRWv<~%)sI`caJe=~r9xoh_1%R0+Xca$LQkmwlc<%SMDghTeb>^J
zc1mR>AH8uIu|lmRVWVXiKes6i?`<XDbXblYkO%C|!AmpqUp~8@O-6#3je1nM<i)dG
zbGw`!aD@!-|9R;={+J3w!gZ-<Ic0{tg*NfN`2e=%mOr$Sb;=J{YRY;YNil!J0b!TN
zMOAH-rrZ@H&E*X8DRw!@5*$IQ9;3%_rhB!SHraxDO8}GY;oT7167nkkD0fUEfTPe)
z<bg52qN=j_L2%DkB=3q={AYYc<jNo9A&Z56<Se=1VoZrm=eP1v*0M;+DH4EPF4ve#
zVr?1n+-C;OyM@@+XvayJ6tK!#R{hjl6tM8D9KH;>G+Hd2jXsi&9)KMqgv0X%uvd=q
z@XoR3W*5u~Zi8l#XFr?}?S5=gh=;TxNw>n24COw4mbHy|ItyHFRAw|IoN|(F?Z|R*
zJw;0W8yPSN*i1GCsiX89Z84~@RuJ2BnDI_c=Bpk`x35GDk%SKCsdE-RCa(%3*3;X*
z)H4gS2>oP;PsJ6W<TmY8U`oQfAY>;nL@LA}9@!*@3^jGd9xNlA90$0C^;HFB?{K_=
zlRl~~TYPkaP{x(z{Rupf#@M8&u%34a=Qu%!l~}K=qsIV0lxi<xx^b=8c;#!W7*72v
z*uwJ7OKm_<zT$w5X}~1slZb2dRx>dlNz_?%^z*ObJ|ft=MNwXrzBGdHOu*62Y5?&?
z>+PQoeOVWE{U<m4zLSdUnZ<#?zfb;4!nH=tu@=Jqg2%;Sz&K(C-I|S;mtPP&ot2jI
zz#c)(`rA>SC;Y$PpVzS)106fz<*)AO;|c$k3_~L7+qY#8r57uXMDZl`G^-<OBaX#L
zh*oP7(A{1#Tzj=jqiuFsM&cA+iJLwLl*TkwY+WbMM%CnX2mJndcOOB8#vdn*TbXOc
zOu^6e{Kq9z%%0fBv8>;OXQcIB#2x^6{$!&$Qt`4U|9%+7vcu(huA_Pw#V<?ODj#oj
zs5ccUz@DX;#Bs=~y6HshE3`O3E`-AiBl}oi*4_s8fPinqiW$m6&-VxTl;fxz4BbCl
zatP`cOc9N(J%`XtADP-R6n1Nc$P-jFCa19e#6$XbFoi)@Ol7jkbbju`qh@NBl;}JC
zbm8X%gg|4?%6SACa0KKQl1+YqVN-al?|E(2Q0iw=z@2y$RhR28c!|RM`~1`&7L!3i
z0Yw#fZzjX22IXwkAL>ouQT<|7<_@W)?wzE)-Y$!mu0mtdBI>1g_dJVHbKBw@L4P*W
zSM-}qX9QC?zZ>LMEr=AYSV5m`a$p<cvnktduR%I8ueX5(MesFGx4w0beoI==koy=P
zkhW_xXpTK}F+!C9&LE8o*bo25=N+Q(E(goN*ohTQ$I18iq>vA;9uQXK;37<Ujs~e4
z{ZVpf)1BVQ9Wy*7f$>nKvRlk4v+ronMb|JvI<EOu2`5eHdZ=IJ*LrYYQzy5QT%aLP
zlikjio~McMeB{8tM<R`M&?oyoa<8=|l&f-Ctj+cL5}Tzwgn?U0h?<f{nUaR)+{0?c
zaoYGr__EMgS323L(7@~YQoxAQSe9V}T}o2F&?s~1c#x*UxR_<%+Jh25kDbXp7;Q?K
z$ZDLX9!qgrfB>2c?c=KfnYn8SnCpGk>rUzB=1|xLB0!rE6FFF!Kz4jN+IU?Hq9+0`
z>*#CLOG6_wfVImO%Sa)zPU;DyibiXxkj=CmGsubc6?X9cKtVOqCbF((lRTqmy|pER
zO=rTHQD7J(p3o#Q;kd$Qt!dNz%9Ju4?`J^1_m>h?^;yXKyJk<xcSN?HC&G$OL&@Xh
zK@;f7bMd9uN-p=KEE>+pZ#V60Wm$(DX8yr^F!WNt3rv)g7jBm9uG>SDBR=&CTZoty
z&^1`tLn7|2jHV59?UK9pVEHEki|q-Hq^<^^A<{)HWV=CXC_Z>14`E$A8;XN#tVv}U
zq<{Bm{8=bdBa!)W?>`s&43{VSk(WPd(o;h(?*rG_x%{6hxJ2_je2ei7)*tcK5NJKu
zhp6J5?8>L0`2f50bP4C!VE!b7t5+wTEaGF(wT|7p4Z5NIBF9ON_(}NnHQQ2_*FuP~
z|I)N7OV;my(WYn8rp`kHg<O?BzP_d&v8q_#zFTCyOF@FLrJU!GMIORfWLt_H2aLb=
zR2WGfYgd<p>&<YaWnd0<@9$H=g<_RY(?^wUm&BoKfIXg@G?`tO=+SijskvITijR6E
znb-PxoI1Nzwp0v-W#Q$_*~4V~0m<#3*c9`4Giwh9_$4?(-B)zxe(gM_K=879_NPb8
zGcJ<l>4qs>3}xrE<(AhZ+!UPeHqR2R)`bqqq~*KJ;h`=WovgRAUz9^k&!~dSvEtrC
zG*mPHv?!22woKZZodqIX7#WihCy&zQ$u3!Z!3$XMgFTFR$5#&xKgO4_d#!gQ7pE!k
z(R#f{q?m;*EH8x#)Z}C!TLn=SXiaMA@4_#I<YXA=HUZrMWPE!?4PltTe;cw(Uc%$f
z^C@x15|Ge!v>g|-ER@-oUw<8hsB~h#vOmN<JX~XURI(-12bVNk&w52AC8VWBhW7*I
z`=zG7otV8U`@kE9Y4>@L(5Toc54imwD$KsFHU6V~a&Dvx!-Q>jLsWbeHkF=ry<K*m
ze0!fW^$!h;Wo_H?q$YLUc&j3pFN3kgs9a=ISj|>7bz-G7$>H=`+cO$x2MJCeX?<Vj
zolB$>qf({uS4ip>nPmf!5Zh78=_0L!6K;7H+u1|psQ>9@z&>67Y#)ALNd640;sAPP
zD)|&}8+9tP6q80*jxESu;YpBzo~U-)>yyV4{rZ!GvlE)wxANQy)dc%@U2oCh>zIfT
zimxT^O?vq4sJsj?denV)0U;`Boy3~tj499(F<o|8YZ&K6jLTy4D0Z;6M*;yA-ooN^
zU!}hFW{Y`wPLhB3oi(|Ly7RovRtlD}xc~M!BGHT-zgy({zcVU^(n4DJK&AgPh5HKl
z<w>&WzpiyS$`Aq_S6jD(a5EN`YgH_&u3k_k-285*pDFfKoLdZ9%uws=r@H)^&F9}$
zdg|fm;p)5|t?AwMP%_QG8}v1QPp!To@K#G{&Q8Ozz2VxAUUkDyWr#D!$<${NyHb6A
z2DS=;$@X~j54*kiB8YWNBcic_@LPpK&yY(ZQ@saEW|?JfMbdPsjP~&#(ynL!(S0^<
zQd~HDwL*}09wzz;PClh86vSUb>uN=NH(vYmI(S~MkN??ATBF0rwnv8c4Z35uU{8Bx
z#@4mB<=Tr1ab&QKGq@DN^m$^koX;a<$em6ZS6IpRgWJOk^Q4eJ6A(8;-tZ9*@?Jsf
z`^OT$$jm9{wD(}2aW3xJnv-~kwPbh8qb2GtIscUju1T$8T@B3Hc>MhE!;p57=P;gu
zYv~oEzr~*>mm_s39K}o^n(3x5->F*jW<HfqE`VaSa2-MTa8!S=8PZ|0b0-7L`GRIV
z05^4H+W~i;%K$_j$T6o-{4%Om&Dv0)54d6fsq-El!*5EU*&4svJ<3aTpobH%yl>u@
z7JE9~H>st!g9C#+YL28&AZ|sv1}uivfkW&O5R9I?=#qa`<D5aFN`+&&qH<cx%g4uL
zpAKe=#$T>0Z;eYt`Pe{Uo^M|%vurKZ_C~9vXQ60HWcX{rB@p3?<d2zy7@&aG^dx#a
zbc_<1omKXH?(auJVR?mP!`ULY*?(L)JhPl89QO}M-`}x^I-~k*jKOlEYl2W5-XTG0
z;1@5$h4z1u{9!%!>0N0jm{Cy}sbTv+&<}(DwMcEG3o?|sgT#wKUvNJBcRtu-+pH2T
zw2{wGg}D1qpa?Pp=C?pQ<>rV}w_I{G{FhN=)04aANoReaHzg8>Hl5+ZT;5m@&v4g0
zk?4#8I*P4#7yTfU9K)+`nrgsK@F16BDJ?m=>o>wG><8Ky>LNysu$nPCUd|(z8h{WK
z@uxG<@Jkdf<Mwei#(X(Q<XFe^U+-ofP9lT|ued;V?dO{qaI=az(&%S9`Q`m~)Nj%-
zgmCx1=wHz6nDPD9+P~1$%1X)ghEsz%s_JeqG@`K=IW4%GYV5zujmz$_<^5xOd%L?u
zioZE2)0w%|k)WZipB!!GRbIxrUJZHbX4o{G?<XJ~!|27ejOHvHScoU?0{$0)lQ}f~
zhMu<$M;Fi6CT21cI{6?!y*I4Szm$8-T;g5Lj5?9)Z9&<j^{*UwMlGQ9#_lZt>Yd)V
zRqt_4i8IJEezg*i2;YH6ik#!?%}{6()}^V1mtQY>41hR~5LG+WSNSQPcS^+7&l|1?
zWVFnz-sx>O;=vf*|1Gb8dU!o)Er@{bZuro`LkLQ_N=gdK%{ZFrd&&8qWTv>JQ-A8&
zT?acVYBLtY()6zSBBZ~|jYQ>n7}3cgMl*O*$9}UMhLuJgH1zbkPF8e!Msj`VbzrEJ
z7>Tjs7#W*3HTAxW-Pk}e@3<NP0YTA~@iSx<JghzbXK6iw_e~}=Sv;4)Ru-ph{KpZf
z0#w{1w5VaVmPqVr3`%+~+JF|nwjgK`pBuN^b)8tlMpGzS)>iHOOF195)mo$gL!{7Q
z(Gu<~8=8#jh%Y=M<E4ZbY-R`WU3UeQDXj7?<ki5?s6tK3kb~U_IN5hY-AaMM2KA}f
z#@r2|y)@f2tL3J3JoGS^NtG;7)gmy$y5VZ<Q9!1#CDv}rXNNI;nYVloWl-1fLb#BT
z4;q_WCZO_B(T{=;MZyvqpH?e$f`<;j58;b3a^R^oYR8|WxZE!ier+=^swxk2>!H)~
zeOqh~%f;`(zY4O3Pcjo@^38ejzs|P|wIt6gVb7CzCvf*~*LroYC_)<V=uyeQ6UPv=
zF+<O)II%UTH+k|_W5*7f0Q)}6CkC0@K6mR3Aq!Kr8S*7|u$jJ=C|iv2c5;E6uZZWU
z15$Oje=Uv9`SypW)2H0IT4uebS?&0i@jCpzlfmyj9QPFE(sV*!S@QnwQK_X3A0DsA
zcG#kRJM7*38h<Bvhfloy+x>Y5Z~Kcf7%|XSd&$+hTw$O2tZK&q=&`;IBQXcJk^H;a
zP4;q<4CI<io|_|K>fREy0!+V}+)rz+bv7`5cNo{Qp@LiGhSSCU)@i8i9Eqm83u#h)
zZwB9hT02L%dFis~l(v#V_FJ>D8sZb$UB5>;o&PAOhg|EBo|$+4A(&!K@Mv(DQH6Ye
zDCsiE?9%sj?+482aj}M97V+uGos>7p>5ZI65SgAGj63eta55pI25f%;r5Oc2)pzp6
zao9=8&%cTbHKFnUHuyt0nNk>gujLfUPKmWwqN+|+Nz+y*p?O;GaiT<17+b)NSe=QS
zbXTKnZVKv>8it9)M9bg#;|;EyAry^sm1SI9uplc;h;15W0jhwtM+oqsrhw#zAHf5t
z);O$EH}U&7aCh>p`&_nXte?R8O<rCQx6=GN0Uh~PUV3iR+Z<wodAj2Vtga95nfcwt
z;-3dhjle!2a|7hq;b41tfSX`-crH3!+t9GfT$|!WuY;6#LJnYHNv_3Ebx*UqaTtIm
zrP~W{RiKOo_N*@)!P}2}laf6(*yb!vai|5}bhcSKQy?(S$6sZcYM?A_m!SKmo~Gz&
z(ryxysj8c?C2L^Y4f9br!#DQj4Fk2v%dKEXb*&pRIOV*6<FIn=h_Gupbdm`d*aIuO
zJG2vQXckPqp26yup-NX&na&tS?0K;*Er<MFS$O-5JJ#%(E1ok9qPlWcRj|3}e>TC%
zd;QAp3nbRbt`r7C=*LIZ;(iO|d82tWQ$knlt3EB6Mn?LxPt|qD!*Jo2YN^DE13j98
zQS)s|E&>+InW|TW*3G?9rirHUw{8|=Dv+D21g-XM^+7(yWk|hJap#9Ehdz3$Q!VqN
zo!{duK5a(alXrKQ$^&=FTwDUZpl(oAh4#OPA!ELPTudIvvL$%S3K;BdL5}X|Vvd92
z(suE1k4K&Ln*iqrUVGLKN<HS~eSA2LpV`Bi=;iGT*~`^`OexjGiyNP=OLBe}&^7Ak
zCR4~x2h>wqmALqLt;}Q7v-o&7yEaR&ibDl6#~@QUO5ZZ*R;<-M)ZDxSZWoN4W7(z6
z&s0)3e7r6E9tN(?sD(%8*UvHF)&g-Qg!IJ^c6(*YicAk^PD#r$f`oZSSYB%bHo<kl
z*daEif}CIhkI<j2jNSMECQa_+U$nY-F?zpZo8+Xcwy@MRYqaJ6V?}^$KCIpk^C}{H
zIU3R;WP60v4)T7hRV#I2YKB-+p8I7dq0vJKY^PoNMQ~*JOB8t5%X<KtG2qi;o(WsV
zkbjT(aY|Wkvph;fc8e|y)+VhpO!mup5A^W!7>xbrpd^JA0lAf=uqYvsHces)4$MR%
zWQIqXrbwEZZ>J}I9c{B=Qf~GQ*PHOsat3hk#Q7*SjaAq_!6DPJ0dY#~!R+gFSgrFn
z&B4V4Z~BW^`Tuyb8jRxRxJqcmDLZlN%6>-Mq^GB#s$*{)Ex?Pv>`Erjw|(c3tv`jM
zkEB@Vs<ivLIJxBaDh#P6o<3~9#B_)&(ha$o0=~MkBC|RgAL*s!Y>xhjfzW{F6R$oX
zqQbhs?gGLz&iU~HgrXp+a+t6AR~kZoQnMDq=qTnW0KEd3vdg+Ej!+YuzhWAk(}$fB
zMDw+q7ttS3Um`-Zhmbi=NX84zSb2E8q?ebOTV(xyF3N4o)0tXW8SGR=5_Uhp9i#Gr
z|D>+qe!OS8KX*1UAvSNvrTfw8i7hx;T9)FGvDH2~4fq}uTNrunGd|`$CE^edZ*d_1
z+=>G@2LV-La@?4ISG(ngetsH{@v0~kDshU`vYtHvl=*x4V1QLiJ_EIGJ?q`8tY$9r
ze_<&CoTnrQu#?L0NFHygZeO%hC5({X+sJ9wDVGfkk-2i`w3+AIB^rNMF-q=to0d$5
zPQxF$+0o&bx`B>}SAG_$(lQw&__64W6rka6CBHD0PX&FS0`Hxr%YufuC5*IRzhy)5
zGA2kR^W8o_`dfXrGNn-V-(RwvGw-%DOS`8Vub#$%eFRGXCDP~)p&_MOh+x&ZDl2(o
zH_&{5{_T&-tTx0wq$mFTUKCjk0EzqeDl&ERvGlJ3pgeVyvK(qtQFZ?^q^>W+B!DWt
zs?kMi|4pds&WHpUu%2JPz81)${_W)h*7!?LqW;V1Q;&oa=4q0@dbC+kM~+B_O-8i}
zE=m>>JtxgSAlL(N<2m?}y?{%^*k{u#^JlrJe$S`9F%Pl+IMf!u8G|JIIk3-4$Zi#>
z2#UXL`b?UgUtF;w6dI6RlD2yeVqPE7f!i>qc5gxl>4Oh;7_?nyjn%(6{&}1wx0T&$
zJrFD4djhG2V+zm!;OUD|CuRjm;Kzqvz0MynvNy_$gf4ArQO+Q0GW(the2VGurb?OC
z_{%|nNIk~F?FqXTzfPOq`z^^E>hzzWDPmdEnA~K$+=+qJk|;*T68bh`Xr2Smo${(t
zi?xDc=Rn|wSa=j<Q0f%<-P~A5Xz5m5kWQ<~$heCf?qWgxt#xAgkQ=odRB|*OjFRh=
zM9zOf@V-I8HklbUQkc=<O<@`aU1(QU@yZ#owDKZ5Ly(=JN`R}k^Jk!@+!yksjD<$X
zZ=BJuM}YrUzE3+8+LIo5bY<j{!F^9opxRYgsO|CK{C$d%;y>Cz2*qqVb*RvX`Ux2K
z)%QW2A#MmaxwO*oN0VNYp+WDqs?&HI!Of?DuJ_ed1r+7dKF-so^ju$|)pboaH#ZsJ
z`-?&Mu?V2TRC<r;4v1z=?3W4ib5KrX2I$<LOp?+OWYQZqtIdwf6!o@xMfB%XZOANf
zy<x7>`-g@;p_O|<V4QlsjFpfg`GPG@eNkhtUS9O^z9cZKr9(x2%5qZtL2bU`mk>xm
zez?0(xe{Ol_4DcP>4{8X?eD?)@N{z@3dK4V6R8#P@4vemb^{%DPok<t&A)qZ=u#V&
zmE^K{_wUD3kGf}c(3Ye;mad)gKMPyo#3B3r!Ci6Al0KTRS-)unsjbu&=bcEcN`=E6
zmUQel-Neq2wypkCF<x;|!XUkS`N51&mXED8eca2OTw6of?>3&UdN=OVZ^fU65r2Lv
zd~GW$!_z?o6bMK7(jBugd~?UYz4x)pY?$nFu8*?4Q^nlnQ;fJh;&Wa0diD-wbZQ=I
znI4gutd5R_xxHn{d=w7}RMUbTKs}rr5Y~JJT*+^$;|(W0G_MHNHfRq1OXKB9r-tXH
zlNQJ;cet}%)~VR_MO2+W>Tf3Q7pv*`R)?NQIvL7Z>Yq|qWlk1HZy6BJYIQtAxw$P0
zm6XkkcoJ`V8MQatCu&NH%dT}MBK~Gz&m2O%{8X-W*apdSQP#?Mz^58OWE~gBPuE&>
zW_XgBr`FiB;T;&7RsAtwqPe|yz^^UtHreetsIVvr2!`FY2XM@1A0S1dH|~~a1(!c?
zg6nlob8JOy9W?jTR-RxY)W_XdCex!`dZpl>PUJrh-73}S_)#-J1*RHG$+c8m$5r*J
z4KgcQwtst<q+2U`siW;UdL6bss*_A<6@;RQQ8-TcL>Px7Fdkc64F$|t@3JkF&E4LZ
z`HXXaDD<9yzV6#^XF3&tPL+jaJVwAsptWEDZe^5;DNRjdQ)?+rZL_i++aC{`^J<#~
z_g}RLC=b}wG*|TXpvNZ%1#cZs(bR^c6e=|{7iz`!Tt|-%W1zqpfI-?hGTkh`Cnitn
zGA0><Y^dJf14xhU`>ca?&XkIz2-s%53S7EPbUdG(@_qBPR)o_=S9lB1A23Aag{QQw
zAGis-$d4fp<GM4}`D&m0ojgpc-p)<CRAlOO@rCmXh1HwW&P((3D~bF=KOu%42EJT3
zDvP|^Sh=NPuD|F%{hbz6=mYMglC~YP&pcdZ_0|t3@bsn?c@EH3Y8vILEbFcRTn*ci
zZA=#W7=%q6=<PLgcBSSbh`5NgucVZ|yE~V`2sQRcrlo5(Zh8@ykzLAQ9w;@_sJcls
zQFAfu{Cr=QsJ#HC4=~t32xW+l)9VxN163Zvx4pL?9^{%00lm(hhyoN3k$4xG$;{x-
zB-9Lthu{DBShlr(V#8{6^#-z8eO(ZpEe^^_k^{QEahnQeWBwzW3UB%9_=~emZ7~^x
zt-I#yj*S^spaG%fT+s+65GlmzGsNZUk8-!|5|}NaaMhog{EeL-**@C+2_rZcCdGwK
zN1Kmteqf7?g3#bBCf_IWz5gCS*~v%t+N=z$mET-$42W>)ly=su=_$vm!%hXrkEpR(
zNDRaD!@-w~3-<L1YVYAi;op?r?53q)?xN3X*lc5U0mVhW`*`<*MNGbi%SJXaa?X+D
zY-Yw<@GjOwyJ(OW6@Q8XPen9ecsHwZMsCGY-pGL{oiQF$l@0zqh$E?cncUr%o~Jr*
ziDt39xI}wCY;r}>t<5H1KSTjDuFIz<(@jF23)Iw#Lv?Bx2i$v-E~(Wu%xq&?*46)N
z*g(3}%5oA|RPz(yHYsj4anWKQ<wj&yeq=T7&{9l&$1^h<BRLhRk2<-OV^Bar@Vbf&
zhhTA3q?MZ;$(6<|HB?tnq;=wZbAD3A3%4PGC}FFy_8Pc(>S)i?F&vXOF1VH1!aUr<
z%m8}Rr>Yeh^@eu)ECDHu(x|UmEY!BGY>3C>t%%$SqD$fo&^d%Cu6f?EdlYQ}{vMX(
zjhJpWu$72c2sm<8$`B266YRu$y)o|ma8zBhBaMyXM}9VY^NCW%g~1_a*_`v*RipBY
zL)CV2S}pnH1A`TWoLIkhI;EN<qloYJ*M&S+fCPbRP6WV=Rj3O;khRGa7BN^V9|>7C
zhs?#UEKmo)NNejPlp^)fH+_Z2UfOss{2v4YI-a(<o}Q5I+p%Yt${#C-Mz;_CO@8;=
zq<#<kh5om{Ukp<v?4nx4GL@X@;{X48%iN1N!v_OSQopD3e(Hy-j>?WhwWhhY6Zgs=
y&;3OmPE@}CyAhu1t^9`3Z3}YkTI>J#`a%@K9$w2TBlP7J@{yHNlB|$03Hl$Z>?9HZ

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_010.png b/0.16/_images/sphx_glr_plot_visualization_utils_010.png
new file mode 100644
index 0000000000000000000000000000000000000000..50b04906e41a09ee96527a45e74dd0410a8064ba
GIT binary patch
literal 189893
zcmYgXWmFtNm&IKM3xm7s;1*=?32p%bgS!R|?ixIhK!6|v!QFxm4#6$B6I_Ek%YM6O
z&;ICs-KVQwb)D*db?e<*F&b)$xLD*^2nYzc%1Uxt2na~Xuh&-?Xs_qd#B1MIBkCcq
z@1gBt?cr_aW`&??=Hcq-;^Am#PU~gm=5FWW%+D>%&Cf||>*3+*F2=*-^nV4oUEFMV
z#-wHUUY%gNDjB#VAmE$*_kma}Rb+>Nh=8ChC!_Nz>*z`Oi`7EPF7}ho`??#+fswWt
zzb)(c%5UXN&v#t9!__;J=(N$|?nM02nEgo<XwQ;-iPKi%U4`r1uc6s|6Ag)qUpvb%
z!c&KmE=e{Cn?8Dr+Xy?9Jd2a?^LJF89zNW6{ErcCy@X$*-~R92e+>NO9*W~{vNm^f
z6%g=h;Qy}Lf&Z^W(AJ?J=<Gj_{!`+mf9}S7{r^06JjYj^_I3Vm<yZ4WVY2o)<o~zX
z(Lsv*f9`LtT>Se?E;9Rpj{`5aG(LfW3UbedQq%wAdx4K1YGQI%2EXUf36c_gylLNh
zZRz>2(xlb<vDAEB>UrR05dPr(a`NIrbMl{;cv+hby2XD9f!}4q@2Z|afmin(-v8Yn
zc*2#*qIjS#=<M+kr!(jV{%CS`cflWYyT$T+StO-pxb@WW1^jnd-Q>0U?#qvx9O&KS
z9L?R*%fd@Ie$eA!ZRMpTFyJh(_20C5)#J3(<H)z1HP%AH;11GmUT(XMd$X<cJ0#Hw
zUzSLvSTY$_C)ZSPSv(-kQ<c*JEZgarBy0KMy<7x<{@s`@#*7~8apy>9_$+z0T5K+R
zFqqWdj&gvCBAdz4h=(TTUAV{|EWb154WP>-SRhi@&>@aHj*d5VwM(IAFB(GznN9Ay
zr+vxgA{mUqJWg}ctgQg|kMta>GWO3{1$ot9G!a{Yw9Ey+0KxZxyU$**n^5kMv-*^8
zxR@cc=;X*bk`fn>m)+NoF;dTG!bWrW_0V5bCU-?ax9T$w*|0UB4$ej~75~Qf_ilPq
zLclGBXa$YDtNx{>v85&P?Y&|&phAVlwb}$uK2YJ1s!RWCyr5Gr0BB9w!0QOGFgKo4
zEF{-Ku$VW6uif75z7Kr9i`T-Y`N30=-h7IF^N>J~LUOQ|It$R{UP^;MAHug=oaT8U
zt=a?G>0-<gLgb>1T0@);KkLP%);i9obKtIR`1LHa+uhqO1lrx?%ZU)Gu*4kx9LXvP
zsR~X|vb%!(!n?u-HJ&g-is6)Iqo+GrG}eOLbR7G#dXRLS7Dv<5Sl}7~%Aa1_w2)*s
zYhyzCuy)GjHFT*LpWULMPunlILAO$GSNqn6#iUMO?t?DiV7J^w1kDyq=8!DXA3r2{
zRqoF3BX~@Lb^_Me8YWxeEEYIH-ehtFvrKiB*dV#DZRR3KA5*hoLj8NRk+?95OY2KX
zfHask=n;7?F(@}$qkkQWw;0HB6()iuCN_!2s}Z4#{28o*t*)+KP|sUF34t}L9fk?T
zP0FA)I++bquql8siWS^ilBU3-CvAaGF^gVc45OxaAPVLXAp?O@zMQTarM#{tK|k<v
zasR@PEH)1ZbH<sn;k;vQ;YPe^DI2D<dJ^%WWH}+s5TBPWQb{R<7)exqsy0arEU)kO
zcmFOXagGH%q)!t!yH{au!K+Zz6nYhTh1PBXoAw(R0bnYrM6-@Oc&)Zh#`|#`-GKA>
zzfbxy*sa?c7&PK)%7ljC^cQ=qvy#+x@ouFs@TjXxynXwz*+^g8Qj-ojImFbGARo;<
zP6-Ey2m+v?MPpbil*Q*^*~ysyypef}cMxb1<X3dvJ;{Fa&~<qDKEXm!E@oK4P){p%
zW|pnNwRNqeaKaZx)&F#6LlDNL2<6Tg{%-I!4Zy%2icsMV;Yh8W=G7W%lL-2IcNdct
zz>|{O3wT_NjQgoHh#8M9Qzu13XU?vKVXL7v(wiF&5J|{>d3w1Hx-nV)?<nBf?KiGY
z!Nc=jGU$2m>JB1;<%xq7>VZLT#7z<Ok2L5n&+~zz<nso)T-wspSPc!OqKq66jhsjf
z=P=tz_3_MOqs6hDhG&oogyQTCzh&;0juy=WRiC2wiJ@E}n~)Ibe^VJEki(#>{u>9_
z>xs7SO^v3TTI>QwFrmshI`<Dzcqh0aC_pQXD#tm$kxZ+Vc;{oh8%xJQ%NFul1Oh{&
z>5Ziy>*Df?@pK}JAvnn*I3$HpA%mUenkU)w=w)mnBT5usNjcH%dFt(pQOs4Iv%kg%
z>M)|9Sa~qyUh=WRcu~Zvjk%2)NhRAJLyFX;&IhEPNE}+KX@swHpqM@rPDzf^0&-@@
zh-qm>{!Cnu)!YwgWvr00h%5&6acp`B2BaiDK)B;sXW=w^GGHqnl$0Ojtg}ylxhfOR
zRN?pyTa-mE;EP4vT|F7hTkG5p{x?mKD;k492ECKEpx13Q9oPPl)4_ZxpEQcSz$}{a
zE6LPm=G5LZ+%9}dW=<_S`)QQb<$_v4<_j$%m^2~4wUxB;tu}57yDaT?uRF8`??IDQ
zk)c;Q4tr4ABUdH{6fCWjHp?C#Qj#2_wA_g#77o-CP&AJV1QgdaLdHnwK#@;%-v4fD
z$HqC>Bx9G%LUW^8)yH`xqyp9(X_PS1F`{V_cmcSB081RX@Ki#OqJaE)Z8G@MS~eJn
z0mLyCAP>dqD}WJpijgz$@X-8f^IsQ2gHacIf)i1W!|}w(#l+|^N)?Ec5MI4Lp7rDa
z2n@wvKdYw>ScB{#RO_=D26<`{{X6LG*?ID}A{vp`Ddmc!X6*3cY?mHQ3#*rRjWnDV
zqQ_KB10)ZnF}2_|w*lf1sie|%(`twfel#_|T>ae)LMvb3)s;i$vT;m@w`wZWeX1cu
zl3_Ts+SnSvfBiT=j9^SS6lH_})4MQmA|_F^X<3Dk*jRLsD{v2hro4t0aUl@e@VjSu
z45ZJv<=ELWGo9o_hu?oY&?{)bV?557?L8vNMCE@p#yhnpX`7E?f55SfD-VqYMCLL7
z%}P$?#e^1LCM<k<gwRdkdKl0u_(nw&IObZ=)=pHo!*$&BR-tPm5Gx>t0!BzZXZcQq
zN?E)Gyk-smyWj|AzRBE3t>t;jpy){0%Bg!(a4I4LeGfK_lHLKTKj--dIn<U`?uFa<
zkpRE6fN8vrE~FbEBd4!L-=~BVYw%)oy7|&{RTdjE-~y#*Lv>34W=o1OcYhQ}SfZ-y
zs^yXh`m}TZ*z4Jf(0$x^adjBqR3*FzusHi{*KM1<yUUSdBAF|*uPH-vNCl|c3rjL*
z-0cNc&kc2)4o(`lnHs!73dQQo^$Mcnm1)tkr`sK|Y6gH!Ev*EYqQ~roKsqs6F>+`S
zwTvvHx#OXv4bld0#_0DY?fj9GL#8-%d*B2P9IHzOE^*JJy*A9cXayC5V_U#byS581
zmqKEihjFsJozQ$@|ASM}A0c5BroWx08h)_)Zb^0*D@8@7zrh8A2$kyN{+|Ljw(pZY
zpE_IK!Y~v)8P~mDSnT~!ofz#3unQ=GvSt_tbXEYnVho|w)%aCZ+?e4bsD<0*j#t^G
z*H#DIJr%;$k<1sjie^C>?CrwFl!8(n62D<Q28>1on$}AKM}cUXsjLK+6MjVvSk?uU
z`fr$cu#u(dd+-*6ixFYTZuB-xdkDm#Q;P@=d;ZX`?~q8q#0QeRWWabJk|;+Zx%sn%
zCz_u)5T}Lgoqi{6xROf9s`vx+V`l*%6kd)|zgpa2ws{&)sH9nw7fj<O_E}qJ0;mtp
z6{O>eKYK{Vd<-&*G-saY{f-OR*mtuUCB63B2$W)Fnc_q1Y18k+p$MY|r)MXGcN&N_
zuG+2OV<F1PM3Pdrd-exUiUmj$>+`7p^}Kozjawuiky5-to&`wVFsDqb&QA@s7}xyj
zERIGqSBfUi?K!jYHob(_tZMKrZ4V7kD+(|Jsc=Ikce~iZG}&gW(xww2%TFJPRNA%C
zc3zH^ej@HX6vT!q0AN>2e>%VO%~CdF&=-?wk}!dKKeHq9!V(j3n0h-eF3^9NNTz^M
zC@u|)E5xSw@P#PcPY<bf0t~oGxz*p>J1nny`Brq;lmK*|PBLG-DMCvCn$eI6C6#0*
zmd4@!My_I9OAj0i{Ts0Q%#AV0Y<-eDb5SoLgA5%SQ%Pt@DIJ<Dojq8GMT{=NHY++}
z-A_mK@d2oRI?pSBq7z<_kt%%NQe93F^n5YCeEJ|)B+kZcJ`&BWaDYNlucfa|%O<DW
z$W+G}l9~!Y(PZe8o(T4cmMxJnlRiw7B#$Pn6AtFX{%h8QEraQCXzSf8rxC=h(=|d+
zpoeMhF#}(!(ubD-=V#LpDI2QiLltZ*&aZ|kq+TBT19d9Z39{2-Y3pa3SMWLleR{sr
zl!_`;t8t+l;)<cf<Mat1F78=OppQ^@Z<Ny>!SW`f3G7|kj=kl~VleqnA@3??7BMXp
zRRJM~E&u9`mD$X^Hbhq03Qf`2oH^&>0rVA=mxj%i6o)PMqd<izX?|Hb4a_*Cfyq5&
zE&cg(?8EbLTnqS0uK3@I&SDkTX`pp6lGDUV+LI?sW3Z%O^ZSG_KTyX>#1ho((_Em>
zVl}1<B4xYgo<52(EjPGBMB--SF6LVisUn?kVMEx`F?M!Xdwb1X&V<G;B15@o(zkH3
znJK%ccqoc)G+U@VB_Bzu+*pj2a&K|Ho~@?-tQrpb!&9~|@mHq1&sLEc?+ZW*QS-X{
zqcop%WRO<;`AN_Sv+VcU5urwX1--1cfVEylD<yg^JW~8HA!j*?VoP(p+-#v77%>Ao
zrmbn<CtDpoV|^`C$B>)3Y`wUP_SngJAuGQj&JT>+(lhI)l-cdag8*Ku^w7?93VoV1
z7;XV&bTR3ebuLu_oeTz{w2iQYumFLSmbFP7noK^y^l2-3wOJkPYD~0WZU~zh%oef5
zdemn}Cf9`)%U2$j0efo(UZNu~@9od^TBB8s1M<IXn;KZ3$5!-#H4N>AnL!ryD#vNa
z&;~l((G#>D4U$~-P@aj$%ys3G)?MMnQ@#wHC<S!k(imZYxT(+HW^{z1=N<tY`{$tm
zvWBU5d2`@XTz&>y+Mg<rJDD$cnVV70H!KJ!I!Jt)QfFwgf+7-}tGS$PykEoTO)tTs
zsj{wys+Re=>jB3xUvag@+S=OA_O}$ZRG#bnio8#c#>cIM9)ld9I4HDL*mJuqF15Y`
z<%rPUK0wTch9E`UEY+mSZ+zb?W)_}gPlI0)Tt;7&-c;)$!tiBgu&w#?%xrIGuZA8S
zOAk`F<V;X`(-v4&>`ezbIUkLZ?{bztPE_2w3skkUnu0MhtXfkP6nnsjfw(7rGCDDA
z3?L$yxfyLcy~mH)IMagS;+*Y^mD!k5G`J+2DbA4Qj^K4DAEP3sEJoZ0am>}2Denvr
zqfk$PGW-{<HPNVRsePQ96EZ~zXtByyF^GaM3AtzesqsrwM8e0^p~G+QJDx8g3$G9p
zkw577>P_}dllSWnwH?S)&Nexa+mIlTgjFbDPnEh@<DCAfA^difMo2^Ow0i)d?60GQ
zo#~}CMzCN%VCbVMlHVz%9}|ulW=zrHEB1dk(wkILkun__kI;E_uL6l##cyoWXeE%R
z8p1?0n<ym^OmX`;(+!N!3tvs$_BCtu-@Lmw>yQYUVsAf!j%MbT>I>VKK#vpppS>nd
z{(SZ)ShV#Ikor^+zbVdAs)nu~HHkKgcL1%HR9e^%K2FhZ8o90)lPe*Z0qX5Py-vB_
z{qq<#AK~zPuN?e#fPnevxL;QTULnqi>2D4(x;`Lb7-F4t9ruj$QQxmMa!=ue4}EBB
z)a&SL4;4V5J8Uv<a4(vjkiM7@d2m@pO%gZ6`irvx5(jqS8%B{UTDodZGGs_q{YP30
z5jh~Sh5#d=Z7UHXEFVQULid<T$bhB4<8@)b#H&SkuI)8sNis#x$(GMNf7We>lbN*=
zer^@)As5i&=t)~zeD9ZsMtCXXRKjin%!h`XLQ!xp4u8~DSH3QNGogQiJRMxUnmUjv
zz%*euPp`obdgFbveImcHEb6J{{8K^Wv{Wm$Abn?N)vjW4M7(+UD#OVSaosD(250;D
zkkp={?G+;Ww&&An`xW8Me}LgB(#oR{eVqTKA^=8@bQy|tJrW<KEmf&p{)b6&I?^ro
z<E#J1eh-cQ4=50$g2Z>Fq>19;fuo&DPPKfAXz99l`fHb1#nEc>qFilfReFXA2r+Y|
ziiW-WP919q7+Z+ZKCP9%GZ@I4S4t{@IO4I2?<uDiet0#s?e;H$=HF(?dtZm>y8vEy
zr{MT8(F@7#-j?PQn@f<UimW3blu%dSf}RzMfT)|rT-+`iK#&TgBDs)4cDR&v)6c>9
zCx<<{)&OW9MA&Kght!e@3zZ2y^qNUd8(?`OZR02UI9>{r4PVUOq$FxUl3~iDV<7)c
zXTFs7;I^R~*srkr>y$RRFSV__05~{@{Vo9oZHSf?=a2~GlxU->i+P}$xGtq-rUeM3
z&Z_PpbxP>YjLt#LpLfMI#)x82Cdb%}pypc>p@@6t(Znf`6W+2-dn(@hE1RgrT6l*@
zlFG$oSmICS(gr_mYhSfhv=Zh)8PX(kqV18FEYvCY-ZkoLUb2wH7|^WAqJ-jyYGR80
z3J<ng<3xd_bq8LUclHW9M0+59k#>rII3C;HT>g5n$>Xmv``Nog0oK&RaSQs%+Cz7q
z3sJ%3q?Oa*n+Kn!S^4Z-L`l^c|Cmn{zj;y^!Y;YP{-YS>{Je5Aris^YJ9E=r90^6X
zp!PLcb0H2(BArShIP~DZ4k)@&lU`Umjnh-c{90ht6j8K`J=~}j^^y43%BqqK0Mk`q
zkvsgTNK=|pvXrPryC-xa#4y|o2;!7mh!*2;)ok%H{D(X?l2^P0y}NhczADPrG+vKD
z6Tq+#2$~5?x&UKLo?h`z0WFENl(?@klXIa6eFPJGN$<(xwJ1Ik2(W!?t6h)si<|7r
zXInQ8pMs$NQyc9QOs0GjcabYpfcS<?cj2&0lAMm&4Km-ytNB6=1jSyIy_4?C8wP$M
zwP%H8kV>-ySZJs%ojV}KXgk>m=Ha#%<wP&1UBXstdFW0Zmlo3C7^HVz(f@ARLeX@P
zB13@|9;+uVGeRPe?Wc|hL^H^#&3<m3Qmhg}n$ir${NI28o>JHsE4uBp!z<+6%C0>H
z%J}q`&a1n+3pngN&|x|1<IU{qJ{YB^L{>8+e<psoHl<f;^y5(hl#Gs6nzoY%?;w@_
zTM@qh^CiR5G{3FoWmjO^^kKFI9AwB;ghA)J!0qq<+-Vha2`xq%2$T&05cd%vGDB(e
z_xu%An5}4OeW{Ns4FG~)3+e^kQxzyeiAnYFw0RnwMk*!T4fK4EiR}cI^8P62fMYTH
z_E9JzRa9jS^(w!`XTfN6oJ<2I253vGRWYQ{=ibDy!91_A<m2cdQe{|~Yg~IXo@~lE
zkrN~?)GGT<fHT?Yjei%N0_FHB#<ZpMY6BSaa)`YD9GGLHP0Hx5SsiC68?F>O1<B|*
zyz*0f>{#j&igTq)BA%YIL65sp!$NkK5yhbp#lD*d3n3u4DNcf9OtXl)fNB1iHS5Hv
zHNh5=3iAC-QO1yLkD2w%Dh7j3Pt0g!y?812LF(NrxQl20<=D33j6bg+|9g@yGmC!9
zb}x7-Q1K>>?_}9IVpgmNtpTq;+*IY!bcn4MWj9Hez5>61Vw-h2@gOuc@U^FXzfl#f
zRG)+C`0$Nv159D9zgVp-Z;@xezqh743QsR8F1u6s61R9-4{ct7x6n5eLj~Mcp!zx7
zM@-uIyB_{8gT;FN9(aG-?#&jS{1P@IRA!ZucP3_zgzPI)Ve7}Y2K&}x5Ibm=BPgE@
zxjXRFl})}Wr0663bjE^3KJP|lD7&agDS>p87;=SINJ+%!U@HjLo=V}RZx=8Gk>d?`
zX^QoWNJnCmeA`>j>RCg!#S_gildUOvC@Y*aL{4T4yci<IF8?+%j330fqauz6g}1Ob
z)r+sta+SdmzR-9@PSW{ff)}t0^|Z9XdS59gf=Lm@uIM$D+2~mxI>bY#$V<Dw3!OUl
zn)C^Mt~1Mq2?jp<7ExyJ9!7j4wK6~gv3yOp30vo3JczXg;<(EgMjsp;OxhGx4bA{i
zbUnn_jigDe8zeUtFyY_*#dZ4A16GebWH9JzIIAn+l`SuGEpoJIYbZ4`hw)lig`Nq9
zG&8Hb*_mx(<y>72RHDf&I{2$x4XeBTxC5HaYu=J=KoFWoE1;j{$1I0kyX)oH(gVjp
z;I5ODh|WF;^b<j>G=C=t6p{7~U+Qds8<is7_XScW?0Y&`fIk-hUQQ+gi0oi=dv?d8
zKZ{=(=`xzP;yW>n_$vk8h{jxeD_Mxdyvo)OVE*g0C(E@?42vd>&*#&8=gA%L%+2%W
z@pPZQ7cf8Dm&}{%X~yGp06-!)Mws)1Ew2FwvQ3{y`ho3B*&1yz(+!o$n2ru!Esh81
zz7=tLn*3VP<!m(taytK{g6vVv*Y}>ER#y7t_<Ob{uKi9&^X4SOG{s=}I$1&j6a{-2
ztZ1-d4+9KPiLhfAHVtXw58tGJBEr&Hkw}XF7SG%<nF=DR-x6GGUTQkT-K`$T#g)WE
z@e)A1hs6S+T)dgpX2=RK9jO7wDS*Z8vjAG$>f__V8cI~fn}h%xV`0bqlFCzH+=d$g
zeQrYTfgJgGeh85~pSy8mHq-da?Y(NBG?cJ}$<X~1xBThBS1U7P{%uyuhO_%X21#Se
zSoX06%<k`mFz?vX_X9PAp{@px5ykE99>vFMCE3n_^Ls}qQ?(Nm^)Ss=9~M!cy#Bv8
zI+#SUv6|UZ4)MMMcqe1-+u_!<IKKS7;ZI&$fkB1w3i_#5OG<VOw#+>Jn$DElc+B;Z
zzt+4H42#@lV%Vnoy8>SzqF$tfpg(0Wi-6$fYG>u-5}S_#NWZA!laWLfe9wB!r0|^m
zx*}_wk_8od<BTJ!{I-WReM-XG%GEOV`WuaYE3;bdvU;zsd3XQC`nJ*Jr|<1&aWmuw
zzW`)lBo}8K-iHkPhM3}zU3z(IrU~xlvYI0r{sRwKzo%hfQ-G7w``>OT{s_7yP>nG>
zp$%!`k<xUyg0y<trm<E^m9$SBhBQQu*uxUl4grj>fX&LV(Zf)kl=MN)jL~9(M#9P=
zzJMD{^|U-_OqGDuKvvkY$l5>XrCh+24EEQc68bY^Eg}&AYqdMvIzG}wUGPfQ2=3_#
zy5FG}6V6dY+|CC7$bK{ZUC2=yV16m9E|`ZTGwtG|`|^him%AP;-cE_Xck#I0*r4Fh
z@ex%u5NK-}6Zk7Sng%7kCub*H>$Icmc}waExip04<&3o8S8X|I_cp4IC=~vh97zRN
zkh~*1!>x-;5X8XQOu6m3K!3j7$L~$|fj_CNm~0T&Z=7mXjEeK%(uM%&=Z4dA&9G+3
z3#sGus`ZrVr_YZ(G>;WT@eH|&vv0&wE<z1=%j<R3I@at+_}$;?VL+o@4|u7AP+S_Q
z0`ZU>bSz^iQ4`!I3>kPh4m}wP23ga#=Jz*^17c&+3E$zMq_H)+cdWRk&P2p<&rEMD
zhDrQ6mU&GVYq`EFP<`L|n`B{G_GWehh%vmGv$jr0PD;AZXQZ-RI*kKsjRzGB0%Tiw
zBgMq%E7>r#)n$_IR__RtV^p%`Q>Xnc;(R7rT=}`Nho{lcx5ZW*6DY&$tN+Fe+FZj3
z3$zq1W@LE!3DCp{NP}Le$oAS*HIE9sN2mz9g9k2Lv?Gw(b*~??+kQFY65PxEiBj!#
z1d2W1(VhBLnhbsk%>qZ5Gz4mcCPwcvpIFNkWtcSHF|1SBm=HyPP`NbT6@G=3d3Hm=
z2{DB9O`zOX5$;l#e3D;H3W!a4pMFObC%wkgG+z{Z3TY3k^Lmnq{i-7mum?@cX4-R#
zUJ6P<rn?)8{o$&NaL{i*u41euPc1|O6|_PsN~a=9@e0!PCeH1>s7UsP6UqUKLQ?2j
z;Loq(?Sh6SakHY=J<a9>ZX?Aq6?iY-7DY}>l+EZ&gcWpJ@E}*+3~2JmL|)Us-;x+!
zfp<j68$<|cEoUE9_<}aBx7M!?)wJ}jyO<_hu_*$yApL<JhKK1gaT&wY5x}Zw>o_&8
zO3+?~_BEs|jM7-d`zZSf$`T~0M5k6f-YVUM0#=L3?T^}Jp?_WXQB(-~a;4u3Qo;Cq
zy9^bhk_EL2Z-26kwLWY}ak!&%?Pm{M@n%MZ{*uPdoyEdkKHseelb&{0e#2J(u)cMX
zRy3?I!ne_Vhqj<YkAFI-;+W4!<=UHf9qYN4uEVg9FOC|#N?w`_`fK*Dq|hBsI@c>;
z6?4uipc3gGvywJV&s{m%^7!@rmB~trqdJte;fWD}d1T)>(QjdCCRMse-}PX}*M~kz
z$qBvT&HJXUo30%+hue$ioagBmX^u4b<7fExoiFP8lb6k!H*^i^;}>M3B+XYh=<eW9
zf=fna%t!O%N3%2O9KFh+1I=m#ZoeD?)z37xGhC1bYr|2X>14@3Nc@Q!qv*z=&l!5&
z&Ct^|#NU^`4}tA&e}zNe`%mzljOET^2q<UReM_pf?<1B#)LSh5aS?+2%y8Ru7zbkD
z?QQbaUB)cAV$h8dz@417w6^yw_;%E_C<2RoYo;-40Ur#Y{N3Y;BvaI&<!Siair!R(
z1+zXTWeC`7xI!)FZ$eDQO-y-GZo8t1STrF<XiUUwH~^L&v18PXnk#Ewl9ihv7L~ox
zAyNI!qd@91QSe01AVVwX(Ms==k<@W<!wwmG%}<afJW#kiYz&%`@(_N=(-j%9QvhfM
z6SLHctN=Q!Lj*q?%0z>ZRxRhW*apS|k8C|ghj@zvRET2pe-8S;G0OZCu$Dfqq~+oq
zYOTNlKcISgopLI&q3G>69UZHp-!*Ugq2gcpHuk%Hm3x7dl~mB*@+59UV~8|q7CiNf
z+2bjl9YU}877-7PO<v)R;YF#2V=iKl46%SZo*yy9lY3XzXi#(IgTJq>>cTl#ZB;ks
z`+Ljwv2V*X)cjF5OkhK#S^kHoG^v-%7jxFO(*)LXHozA#cor!NQ{SMKf$f!AMVekI
z18HG%T^@157A;_qjYz;<e|w(d)a7;PPwJ;)YTen=lHRgfU`RNTa)0|Q8zs$__V6_J
z6c9BvAL^!@LCjat1x0tD>JU|{F|Tln{hLoyiu`^vQ4IsHqABoK9a7Nx`Iq;{PiGE|
z-@+E1Q)@4tI~a&YP8<GhAHLGdGF2;Y&)Ht-K}oUG8}Cc5j|4oD=CO{43NG1Cv%m_?
zUhlWKw7+u4a8i8dC{<#}jSuONwsd5%M~!+kGEH0@UE8_C*9E{1Vb-niiAi^)ws6;_
z>8tO*%{K1tz=kLuiwo&O#dP{xc?SD1f=y!byGiL-Jj_P>nW?O?%shiGlzenRhR@9h
z`y0|$Od0lBF)HfuJAjoEi?Xz$YxM2h*P9o}TuZy@&iAZbKtHkUXuKu@nUK5@?Q9M8
z?d`tej~q$*2XUD1`h2JEG_&)|m%GPw+-ummy42I#pt~X(gE^YwLNT4zwO?!5Yw=wX
zp&fO;$_u#mT%2_0`#mgLLvi~;=LutZw^6keb8!GPqD!;O1bx`@DQkdMK174`*FpHd
z0QVcXVrbH=9C4wrcd9K;oZeF>6&2Z9<rbR|_JNvzX4W|{Y*D<unR-~wPf!ItHA!dw
z&G^7CXw$ys{@>s=q8P__mdeS3Snr7CC{-DP32j3!F5FFOl%2x&oOzBo6Lu!)p)n}v
zYn|+&L5+)IGcvIe3~L73n|l*86C1OA{L4y~oqP3@8t;^Z<sx97zBkN_ebMrYe2Ry~
zf6yW8dQe3Poi~<P<PCa7_;2ar>Lv2*KI-J-?;(@)D6sVQUSEbdIJAu>v3P5YJeZ*S
zwLg3zKw@b5z*-70M{v(8R7p0)`0;Hrd%Eu^{jzJlV=a0c0<HQnZN8&Z>FFVm;^PIT
zi5D@ZPViHCi%3ft0}|nS=9qP1M_cV!rgVnCKe>S{8ovz`UzQ~Ks$#^~KrwS-q~yRU
zN))Q)u;nRrd*0!`L@|1X%{8P+Q4v-Sfg-=jT|o=m$fJ2ao?zvqOLAXl&0bPr|1zmL
zBkY68s9Zp7@O31GzM)ljz*0pqQw?v(y`=Xsue71A*sS5~stCTThfTF{cOJ)BBxgQ!
zxq~i$n!QMI{FJT7aeRhbHn%xp^VY<=s>P3{zWp`T9zQonanl}pmzOjPSzT`V)|57$
z6J+{9_}Z;{&AHxUBGxA){F8{hn^Gp?sf$bM;%^Y2uMC-lv7VxmdAdZs1DDQr-nmr!
z$Z7+yu-$zoSh>HTW{#raBvlJItK9WEAUKdrGg^KR#=b$`-5rO+wo|ETMOfm{)%_;K
zWRk7=W4j_CZ(as!%#ufBtbr>RKems=ht`fID_%OrwWwp<v}oH-5OxtP^vLVnK!1x(
zV2Sq2BZIR1q3Rn`iY)CwXHtTvdMdGYIoXt%$6g`-<}xMBL{?PP`D%C2apHQ1@_@I+
zddk%aa_aO<?(CCkE1&)qjEmPgdy>Z%AV<nUUw3=^wkp^lTcTA0kF`tu6ohQnOW+*J
z%0m;hes__hlz+LkGdJC)kFha=lpZzhD4~-&p7T8D=5y`QxyTyDvV%$+JF(sc`+k>4
z7<T3#u*vg$`K>pdY^Av2BW0`X7F{>b6o6{-I+n38u3xUHt2oUtA%ktbcy*!+Fy5@h
zFX1>ZExUg2ab<0VbNYj2G^QmK{$8EW>J!43vBO=O)fZ2<1q%<usu>?8&}Yt(k{Upo
z#7}@Pu;M80M3h-GE$b(XB$N7L)?zimPmOW*pQ~LD8w|9Sm>9ldRj=$JHibGxcJu@!
z$1qXI-BeA#O*5fCuu4BXR|wxUh?@9cXRdbA@>>k$$f<vS=aN&M!#HF4fScdB%BX<7
zywr9{d<3792+4_%Wy~4NNd;Wei&>DVBi!1;bidZSwM)Q2HAa<2KTL{mDD?0Yi_8{5
zt=jW+wVW{kK0+G=<LCsnd196*7kR|~Wl{+nm}il5f}1IwSm+NzP13nh5&U%@^13r^
z(FRiA*=B9imw5I&nh&pZ(5HY_e)2Z>u?H=68n2U!y*38ZkfyjI+;31(t$=b+)Z%6%
zk5U`wqrrR0w@!a^kJ>fFf9U*7uD&H*Js<i?=n!ie?>vn4V@!}SKhIZVxcrMJ^F?;s
zgy{sc$JBr{!{zU$ViowAE#B1{{^H6+DuzQ2Rcx}h^)NOYvzA+;y92I;#*mr|_s_$^
zsdpBaLJ@QUdi#~Urxn_<3=B<6GG$@95lm=u$a8}3?UnwIw=qTENN0C#d?}U<^YCm(
zU2a>fuLC>QlSBRm!Y^8A!kNCnTz@AYS}{0hNb)053Tcficbc_;#F&%5{>A)A4yZU9
zcyCir629STA)v1KA*)J}e`$5$R4t5iTFSA<hu`qyoSZ1Ifv(7YRZVuLz<b6i+;E8)
z|LzJHy(@UB#w(jya|5P9#RaQz*nMm@Rgs|A)5fBu7GLr8p4Va^Z~tPyl8i1j*iM^u
zkiGEl_sr%HF|_%%;RaGg;<TezypNo)r}Bq_A88I%w*fbWywRFZjpXtqGml-#%@UTG
zZ1i3@#)JW_@cXWFH0wjn%YV?DeNSfCWYh<ze<nTsP0l40e9J!7TpTn|NmHR;r>KxZ
zm5{s`(Mz1$7!h&X5PY7T+l`CFOS-l-Z-tWO*L>t+Rzg<99mA%6c1N>gM$k(U{8c#<
z@z5cAdUWG$C@u@%TsAPiwL<$Dl2rS_qEFnj1_Ynu)By_Y7p%)?la9oswy%$hN(njH
zsEt{6@0~SadUoINJsP6~(!IIZ-jtcUjGLsk{k$s(3|Mp*#h>o+q!2>w5@q|wR9EpC
zE(=m|3LcYS3SxG`xN;oJT5@}uP<~jS`5b*JTvXX5zUs0ms*Z5kK3MTRA>TY{3J36A
zUi~=CRvW!jPBv22XlznUAQ-RQFL$mQja>QGbH9D^pbH}-{fm_FMSfJ4GK%9Hk9H4V
zt{+kwF42Nuc6~^AUq$LWuaLj3D5O-2O*W<Kp!{jUFhNDImg2o7R{ZW2(G_JVZyjf7
zq_Q$u`w`_$ruXgPfe)#jjQe$@r4U9Se9IrtaxvXTuXm>9@%vR~Y=~5lPf?o80)AA4
zA#F)Y+aoIIeE@eDD+IFfk4dxI;2#%A^UucB`1p81@#FN;`>J(^qQ)W5_A~z#jnAAd
zw>@OUp|TF^L*v9=hL`bzl?0f-+v_(O(|&dWV5tgaD_?pxFFVjrIX>7OHHDq)CFKFy
zU1{aMuImOt?9Ro$Sa6~%a`CH<p$b$G+y6rksNc}(syE6HS&a|iX*(TYuz3(w<Us>l
z`gELMK8{4~+Dz5*-LDr%EPZrc!oQhNh^o*MX${+W^9{7;1eoi+`u=HEvYK&RK8f~n
z-$-dvx3S^<?#?2@KY^}vi4|2fo;p>{EX-)aZ#bKK(FdqYLbNap7b*A{yh&9!gjN+}
zr74!tY=>*mf?v{|QjJclOwxuHC3~X$bFY0_!m~$JN21|Atj;(hNrD8VuFqGD-f=sg
zuJ^;VmvK!<*1`>jO`j}Xpr;-&Bk&HvU1{D)K5f=mN=q@qzLz-0?$f-I%Ao5^Q9Z;d
zZ4oaBjltG^#x(9ki=F4dMpHnybEhwP>A<@3IfEa>@GWTk(92J;P+hKHkcvIgqW9Qi
zv(3Zs4fk-Y$JX^Wwx^VBlQ2Yl414#@rqAu5&X`G~jby+*4t-lk$K;TVI9|eyw|Zrk
z6ho=jN6Qe9zV@bij--3L#77QPwzR+?8>y3-6SjOYQQNH#UBG<ruR_hfx)1V-LVXU2
z+^hH=B<jM(Ze9<Zc#0of>fLgY?0zR6`8LV;A+bmw3QhAr83<W*@K9G}TKsg#u9IO#
zg+VYdDH0&@di(q{ibWHg+mBq&`kvMnpU{u5qKa_!)Y4ks_C{D_v6J8I4eUkh#oF12
z^!IGj)&W#Nw+75&pV2a!>7B4t`z|D0b@<aidyV@cfE}KH1dG30o(U2|%R&MyhnCA2
zwFNG>=kl%9X*MZ;$Mh62G)l}qf9|)nvPQ*-gS=gxq@^|(`J^F9mFdjx=>ed&E2+>u
zNYS7B+ssX8_T?)mgLtk7l`oSsOACwsQmijLpK}g%GQ2pdDE=f=?@LT6sG^YXr&{-Z
zyyP=#dBp4bfOHu*h|WAq2>5cH+uh>mcnBFIsEK4`DvueqqT^0(n#+Fg=iN%`u8oVB
zrc@!!2rQ-eFjjt;#+X4%P&PKH;F)LoCHA;SyqD>tPe-+JBj$>FTAA-(M~e8x!}8ho
z9_$NLD(lq6<~6MCmF6b^#ZT(cddK7}%J2Jii{up_Q^hNEH)mLmE;<))R;W+{CO!mF
zehQU3YPj!Ad<Xp$8-oAAYeNyE+7djnoY?J9ve&}9lVb-<24CY=dym_{->aaNuVFpf
zHl37te$5jtq_u;BhkC9DiK+Yc24jjj8a>_KiM2np`+Zy5p(76U^On^JWo|si^2(zO
zd){a8u-)zq<(1_%=-u)S5DN3VQCB~?Gq?vPB6me{H0ZV;zR^9{(|4B=J(#RLeigqa
zYZ&klBn!xjdDvq~hpMFO#Sl#;(oIl2ySEO2(NuZ!o;M6~f=RKR6k4bow05LZgvi@H
zz-7G$lgn@QdP8mSJ{5u@pp{(3?Y{Ngb-b9|%(aTS*d3Uom3Y#TBEH+YWEHb(=cUhO
zA5LAW{vPq3`zZA%>DQSY`sy^2oVS2ffKIQ-l?Ny8;i(^(mdgKqD~#XU;X8v8NI_Df
zy%YGVIbDdl^nU^`-IVmdk8Gj!j>bo-&3)DV<w*JeW#UPlExA(j^Rs%{><ORxzTkGA
zK|-!nxrnUm2DSdkw?d0!$^6590zJ#Gdn#4}?Cw7}m!7H@wF+!+ambCxC_<oXSuJT2
z>iYzh%ifgWk#Ky&cv5Nbl|;9R_w%^pxm<fjkp=%>tn}3V4F5g9RBb!%Wl4UZoRL&-
z!N;+&S;vXCz(Q~8RORoon}Ny$>16ESgs$ZLlfQQ9QF%c~iRFPt!WRHKtEA1gpB7mA
zp>!(@g#jimi96j5KK$!A?s9D84Qluu&1Y&l#;(p2rSOZU8wv|JT@QMPorm$OPS_WI
zC(2s~>@M*ZtN6(uUOwe7Vj%T$ta;Cd=et)>9PWA``bfS%^QM<t#2BGn90E#^TaNR9
za$N15Pp5o(<0WI9MLxEnu)g`-q$V)adi{_hHbslxglLJ02u+DBaY(YSv(f1n$-8C=
z<6n|4fBzz7n^j`1BqEo)OJND#cE%q2FiBbckTP0m#6GMQ-JD3}J61$Pq44ljVzy!`
zfg`+wf+(nUW(L~yp9nlDQhEesJ3ey4dRW1&6x+}k&7@5bT&7o$F+k%(mw+_ML00@a
z5}j18!?C-mY6jMfiO%duc-M)>3ORjBgabsvsf>Y&ZAxo1Q0JuTG>-Vw6ln@8kNW#Y
z;+k1IG1d>Jr|2lKsL0ZR#8{lKf+;C2p4>4}nAzPFx{I#2oG47O4Ah%6^BaXF(*0)%
z{+Ej1U_t;)&O68$sSe-VZUkiGEoMgC`ZX^|smt_tLA?)#_R!Hl7cfm3&8^?1aS?;9
zcIW8Xkl?%}cQTu%VN`X<8tJ{TA;l%;op)xv`pX+EzGjX&UbE+BjIkH*tgT}z>`1`A
zXGf|`%?^Em7Kb87{iK~vX){wTZr%;=X@D%I{za;i=J?x*b)E2qCjYSXBSe1eE`DL&
z1OE=cQ(%w0b;I4nPe9wnEmf5pF<9S-xG3Qu28~;tV%U+Fdr3j~_WpjTqQgDt1ya%{
ztD7&3sGm-!KQ<>Gz?85aqcPm|aNPg!AU+UiBT$6y2E}~WAGxb*#6$54r_^br?&K!7
zOsj+w=Bj4qcrG6(22CfFT9U)mO9gblrem@ZX)FTqf^LR7<v)pWc@F+Axd<MG$Dogo
zgbLlE@Wig*eXu$5r2Tmto!_wWoit}~w`aLOz2>(HuK*TijxH_xmnn~TR);nNB3VT{
zIUgqcD?<mG{n&rDw05Z_=c~$Xen=LJ<dxggM*bW1$#kUK$V!NQnjYuJ$#>aYm4i3@
z5<(-Hc$nHKEayr;;^<PwLR7yPo+mf$>;#K`kY>53zg-2pWKC14QgNMBsI-T%pilW9
z6i=2kxKfHQich`}l7fh=&us&KDXXFM@k9i;Gp4Vdb`>a&ppJ`SOHd$FY(5B}brrFf
z-pVBKMSP(O4G}%o`HtW4<wOi7a9pEO*pHmNnQ5!@C#*hj^xnxibjl#VhxD7^MU=Yw
z%!k7I)?GElEB0bR^Ye*Vr-k_`q(vH<dsg&Fb-#!G;rEhloQ~=tuUgpqe=&(PCWF84
zju+!##LhQiTG(^%HVOI65y_}I_L2HTM8%>G)rfO1!Gfe~y?@C=L7^Y`Bh4!;(3J>0
zMoKx@<7HfEovLHiHTel3Wy>zY*AG;H0$I?;@srv@qT<zy#(AlvfA>d}uf)l~fD5VQ
z<W!N7U#DqRetBj_7eofgK8u%@qnE48hQDLER{BzxAki^iKplqN&e~=*+0cjeL_^@{
zvq7;k2DMo!(ZONeMCgEMd=~tYf@_O<y?0OA*7hJ5eVm757RisTZ4F1k0y9HUamvlA
zHPC$N2VKX<vQs4@TN@EHelef`+IcOl%t9sU=DC&i`0#S47#*u9<N4-+X(pAb3l#~C
z+GghL8J+CUnb~DMrfJMcG{3jDLi!g-){JcwWEL6;YHXv}ZL_v|Hmfo)Z}+l%9|WvA
zjoKaRU8eC2q&k3v@DvEGVT8DML6sgC`{bSLS?6TRWCtcLKB8Z5!;|EksIc4_y?Zo)
z7JX3!AkN8P&}ZZ7YV>pKiHBLXW~sI0Zm;15rW2p|kbCHt&ZT1oitMGp-D32@JI6#x
zG+D1uPxDu182Zcq6_bt!SuP1B^q{s3Jj5?8ad=TOCL8#V`4NOBo~tEWt@sl{>&RNR
zDuI3Yr~Eh2nd3qg$QE%(0?4-EK16KU<>ZXJuIDDVhWHD|v$|r~0>iZANN(fD*q)Iq
zv?7t|29X#h#RNAILv|8t>fwdr+wn9PDRiE{R#*x0p8bG_jjAtO)I^~|nkqK61?=Q7
z8~8luR1`yIKtGlg7T}+lZ0Oj25+A>u5h6{f`xAS@PB~jGBdIk0Pm&qeds=xzCw`I{
zj7>lX@6QX{+MNxRvrM9wDTR@;&l0kQ@tFh0MZLBI1%z!$b}+)3=$0f4OS0@nO8(mz
zMLJqOPmLna56gxNKcOn})x2j5>bs7pe)>*SsuEGSf<+O6`Jm~3*SDWb6`2MGTYSk#
zz-MI+yN_8eqs7lHhU-lv$ZQyF>~n$_ZUzQg1rD-1VOkKG#7;vFjPhY{-0z<7>+GAQ
zvA|DCqqfu3GveQNdyC+Y42PbG5$OTz3HW3Lz7!|zoal%2w)2l?2Vh__G{D+NgL;!h
zGt(Vk4B5hXHfoU~v+E7olWm+<o?k7Uwz7T2oys1{@Bz$LHEv7f6a2WX!xyjf`$}N#
z8n(cT_=lUBHg{<ZI$us}SHBGm@zV-$*`uogD1O;td-oxpzPrWA@g4n0mxK~ae`hWC
zrBm(__8j5PL{9YUUJM1%^}B;cx={k{_jHHh)>dv#yVglx+>=i<f0%<Qohh}CNWq^x
z+={3USiR-8>K|}_G)hmW)%rC`b46&OvvK2urM?D($7H8=fW>A9y!P}o%Y9Kj08FYs
z%OWZ9k_@AkFopxRq+~AZYpj&x4p7P1jkq(i-Y$%z=2LG|oNYR(NG`A-`vL(r1&0Rn
z4PmhNZn%!`n7x~Lu0bM(R|Z>OewK`~{D|KnXkM`feD?X;N{ip#V_@4whPCyAgI3WR
zWO&b6PpF?>eOv&XAxT)RsCT*SJN3jZ>Zc;(Wvw6_Q93kT`Jq2c;y}{V_pQjiZ~D~w
zS3>Zmf<vz89md5KlJ$|*4ZUe3*3}(nPPUc{vUm~h&-KO?tDd}nJLI`fxpO<od*Lx&
zViEk`v7NRWlbT4!34yJj&1tfm5syjDW&i0P`gG<P5+?Vi?-z5oP^>M30<6*3`N3nM
z<|F2Bhh}6k{W<0`#@?X!7>(>Mh6CTrF$`!<xJqAy2KzXzd$8c3aXWwDQ7>PLpzQmc
zht=N9tkSD#N&#Kx>uEL0B%<9ViiK_2fI0d+k+n5I|AeJPP*Yl}x%)-%Ev_(wHjgbP
z*vP$&sK`|2QfN9v(H1FFQMP&!+oXjP*}NBBM(AZFV!ZH%QgY?OtPE%{(_Mhaffo~X
zLdBo>e}$_IWiUhDva|6FKJ5sEfa-v8w#hU6aoyEB-NVpitmqsUA?wvn4+h>SQ3INp
z$>6Zz`_P{w6YFC2H4hy=vEAQB2;0rF?tWZ|-@Eg9r<q(OJU5DLI&HU<JAR93v(7Oa
zeag#qs+e@x62Y4tm8G!1n#ikiF}XXxO7l(4nApCqZew4vvM}HmwcbsMRs4Ou$lB_&
z69DLLtakdm+w7ewD?lkm?!=oPB0{NK^i(j8UbV#&i$!fY`911+8UOkzw^6c;BaaH#
zR(qcBqO@W>p-=a=X-VxZyWh~?^dw1Y&6b<3^{3-VgkfTA<x5J}_@PyDleN<V(5)y7
zx06P1w5CDh$n<dk#V}-0L8*=&f9!&VZ)sXlO#KVRd2(?>^ORUon&&7tS{Bb2Y;GIQ
zO475+qO*e-ew)0T9s2P-URb|7XMO0o#?P-U$RYY)w26EeqW%PQyb(#b*EoNmTQ?3J
zXlK@fBGMcX_7G<<TPb6IyNbxgX6u~Z&Jg5UhIsy!jz}W5vq7zda*mQJkCg54%13Vy
z7cDdgsdar3t!Mq+Iy_6EXOs+$m3?F}486UPgcQ|Bddfx-(wvnfRyAz+el=655!1#h
za?_8Hbn*;{Mct=0rk<XRrWK=BidPlakCT+@pn<y}ZY)@3i(wJvw7j_vQ_p$n1%@p>
zWY1NZ`1tv~a%05Vb$h#Y5|N3a37U5s7bdmP%4KJl<-{kmW4E$RpLN7fy>$(n7a}A4
z68XLxCc`pZR*T!8SZj?JdFwV6!+xD#{!@&`wt(NwkyqklT^OH>xY7JM&TMo|7&(5`
zR&}wQHoyIYjYN9T%zMsBjZ&nieyATi{oSkdz=p-gD|L`~UKNE-PD&!+cKUADy7Nzo
zLB=@iPBRAx&D;>f4l%<bph%;$!QAG2Ypc+WOB$<JSyYUE{id@I?(~NLJeP5UQ{iVY
z2-VGtuFi%!%cGD~&V~b7O%?^4Iz<3ARXI)buxS}n{Bs>q{QhO2+V@N4!~hTVYkty0
zEJ$32XM|Tt+Z?1SVhY`OIgPQ7HpDI_n;g-!h1(iLBN5tGvKfh+Q%%OwKDwOPf9BDj
zJ5A^$d1IBXE+yE}w<PQ^qg8x*I<spdL}3(GtAZtjdAl5OHqLqw6@3M-e&yuPF8u>=
zijsiM$A31+wg_ze7N3U#H4-2dsPxOu;!QVUP)fI0cB8L@EtZ=tgy-F>_p+&Teit6?
zV#v1E)qN>^<`7<nfoNW8(US@<G+h3meehtxrK;6Qcjg(uI-RUgmq!|q!!Oz6sI;TI
z*fSu?^)^6dw=-*jt78%+B3PmrtEpOa%|W|;(Kk%;*Prgg<Ag>Too%;N<wS`Z2^U9o
z?4>u8M+bp@QEkyBe!AM=FWA1+88bWPuv|Q@F=5k=Ul#qvv5MFqWK}8cElQXD23-AS
zb6cteZv}fsOxR*P4pZH{8JRgzNX~1=Dn)!)=q2^-dC(i)Lxk0nwQVb^*Vfn`WXiO|
ziC`@JKYxd@_c9Uw?8(c7Pm%AY&lpVL>_l76eI>Lp`{es)ocnK9nMVKLeN;IyHZ0o$
zlBnAiJOSGtlHR~i;&UdM>QvKq1?F$wNH(A7B&b(bkHZJ3MiSuz5fL5U=$S$cVcKlY
z?!mFs%|sjNJ#iu%XKVXfS78HGB-*1t1B2^l{XK48_o2475<=f{IsdY?(c%~z?oYql
zqIpT7dAV#ls;HhjAGCR4ma{m)7IBLikSMx>K!prxG(N*H2UO&VPmlJxdoXx&3W0ny
zQZG!y+YAG7`lq9l0y|y4)V-3V7S_HPtEkUN_f#@9+HowQ*!~Dq#R0P0a#sB_V*v_3
z<@L70tBNeFr!>+79_q9m^%KMuKhiDj3H4cObAD!N#?tP@nZLWR>C7BOzRA(X;(H;-
zPnaU68qGFLa2Keh>Es{F%!qP5XSOlpl`5O2QENhVA01c0IeU78=%A#MSUJanKZ04O
z?dsSwjrB0%1pMM#=d<phvq+iX-LHT8XMpsbQOyPsHF|fF^gs}t^s*>-F&neO?|jW?
z=0^mh%-+?qADqNCpe2(!x_CrqtxkjQr>dCByP8Qu8AG#T*O{)TK-p2$KjVB8^+|3l
zeQIMF#s5CS#~dCU=3Ereen5X<O#X>`XxH$@Zx(^bo11PgVJyvNe!!dy_pmK=r$pu7
zYO#JkbK7olwjHEgnZH4<8+(&Q;{qf!ZX@VT8-%IbvY~}?&v4gmj?i{Xay(M*0FuZS
z3BxWlqh(VoZrD!FCI>xBuP<tAoUvtA4#zYSTaLFkD0C<Ho3CQI`&{#_c-Q%671vX+
zfA?~%7N@RKvuU?`2e!1k7ZFI6OQ=6{D!lG4YO!wOAuz63MI{>Fp0$Lt-Y!>?SOm(w
zPnfiI*Qw1G2#Lk^AjZMhC-3pO0-=#-LqR*%{)VRM%Kp*n_&um;H?2{7oWl_XMY_qN
z`@HUP;uUWUUHk){(2FD7j{NWTX+%%F*F1KLM`?d_{ucn-KqSAhw;QQ>k=I7`&lFq}
zaE6u4yuLXZ@wBUXP)$&u;AYiR{~TfVT_I>ASRkM%UZ@3Cnhm;m=O*EddG()H=7t{_
zBo)qq=)HGB@X3S^-?MlDMfaE7rhU>W3~0U0m(vSOk7^KP2-3VO^a5{oPh!O6MZ-~l
z-vYuoq2O-qaj3a&C2;Jw*_k_>9AjkQxg3a12;C3{DXsj?-~4<2>TmuVhA!B9do;(s
z$$sm$&q7V-eM{?wPe1+@Y2A3cy(6_yW;j?>ff%6%Wb_Are!5UTefkrZr;pxE!wjYx
zk8q!<2=#8!>a-&2qo9_FQ4Tvh>=}8jZZAE9zd(p~9>OymXBG=?O`Jss=a032_HWlE
zcmY}_T~~@*V+|Z>X&!D96eIyN*8{wGZne<sBueI<m!fygr(oA`aa_uoW%Y@qGR1;s
z=2Hx(VDLPUx_RD#iwt#&yuC<_GeX?wkubUb`F=Og6L9PAAjUMqnI<(nseu$G3a2xX
zbU0Ed8W4?M9EvA;))^4jd*^=av@wvSfG*#hLuL>Hn$x6Rbq*csu1PtL2l{#DKD6Ni
z`99{ngE>EA##=nTubz66`R9gxi#?7((e9{IYZa+Gr8n+%u$SVc^j+v*1c`|dqDRha
zK(mLmdHr=hXTL76>^BtyOUe!f2ZyEa?|15Lr`-?QFdvBH_hxpGg%rSsm)ItTHOMLO
z!)4<SAD+2>d}g~mVL`ZDFPQm58l#Z1ZyHe|{pb$omqCiYxtKCxInOW97=e|e`W)$;
z0%_eir^YEh@S*lTEk1!3lrkDWy&a4(_~SLPK3|A=^~qh)VXHPewR)min*k;gly%K)
zG5LTBPw4S+vlGrdef=T&XE^YUep>@NqZQj2boHdKq2FGZ7?N)$8fi_iKI2-8L*hC6
z=;hvrBL@$Y-qZ>h6Ox30zi=_d+5L~lpA7~sXcG@Dk%1)lE{C9g&`pG4Vg!%0#`}He
z{k;<OP6|6>iMm(l#v1cX)N9;cztV2+?!>siyU;&pLQ0O>gowsK?+5q&%M7N0^>Sgo
z@5B)N`VUtJFEJ6sMWbU3Li8Sd%$;g<=+!4vo(X>g+!5kZL5DY|pI4&IPTT`Zj@H<V
zzwF>IY6{LqwmyI66eNl`f+Tp#n&(b*emR%k$Hz{#)CB9DZrv?}U7-2ggOrpJlwg2)
zXV|52PFMTAt|euLf|2|_t7EF|DD-m7hGK`{;blQ(j;oy=tPFAt{MBFmJO1KNe@lo9
zrBurOow9q5fZzz?FwX;Nc82c=OWt_-_yg;9@fh2()5?zF0yQrhPKCC?6fTMiPcI+2
zK0VQUgVt%sPTB92ay+yFh^yRPh})4mk&sj5_@67f=VmQ@6MBi@R98G<afkQAq)OCy
z2+dBh7<@ZmWPXmxIC${YUO9G;CHKzM^#r}+6JEo>e%aBFL9K;aAxfMDh{q&{5a{MY
zUJYT=5%IC7aK<6`fsPphr)RS^i?Ue=NOZ&`*i2oVLhHx1X$_Na6A_!4hdgy7#zCI>
zF$5(gc?cVUDUgnCg_MA#4b5}lPj6T)^jc@b(WmC8(eCdUh1!i;4q6+mmka6o<fxG~
z%FOw%rTBEw7~$|D_dHul^|8d?ApbF09M^?>yJr9sUD4>q4Kc#-gjNs4tCag2H6->X
zl&Q5t19I%V-`>0+OUkARR0&I(LeTL*(|*nRO~pWIm7^AdDf?cz-w&*KR%<X=Lgcb~
z{I5UjV^)h}#?yLNoJDxKUb$XYwrwLXnXpCna(AUM?loUxU|Tk#M(*L@cH14gA1g5@
zYBOr-s7AKy6$ueJcK75*@#x^3J>FKQ|FpF^!q{VGcY&id-tX13?7=N=i!1rJZE?im
z*eQp13^qJ6Hz)P)J7)*L$@K-NGbV@FCxjQHr>6yRF=m{0sjzLVPYdhQLfqWvk2z25
zgLhm_-G?V0TI*<+`7)pWc<4<y=HBE26ITXD0y(b4I6o6!O!PKwfRbEA-&}y##V|X0
zKF8Zz?+$g_0CTaAN}?T=FJIp&$5)oHvgRv_f(?&_4$-@mW1j=4Dt%OPTFA?WM!!Ef
zujF;5*@?h-U7K@anTcVg93JDUt}jrD1W|_hWJkB;bbOq&M|z65AksHlrP?4!pWW4X
z*q@MZ`?Z;P9PtRa0n40MoFOMZM9V27n}2qkgrPYNHX2j$=H=5~>Mul@N|+4jDP}$G
zhP}ZU2^&UY4p|W&$n|?(ajMKOmn)i*5C5vWAIi<?M0>m=b0dr{Y8#HV^7+#r`26`t
zAaK0D^X1FW{Nq3U9ryRIQ^X7iEYZcC=iwGLaJ^nxmc?VDd*i)T4jX7nT-I&Qu81Ce
z9)RLqX^fGOBBd32DU9CO-`_a)8{ktqABXw;uD$!aRizM2NIET~W{(ci4+wi4<USf#
zXM(B=^lbD8SFWl=`R9ehe_jlP5Qrl>J!N1qf|Q3QsEs~dDn6JjI=lLyDq=XP1E&e#
z?AU|P2?@~;<lu_?bFRdRn7V$#eY+<cAhQs7Oq7k$(YIk}bz7y+5t$>B@jYwIe3NtH
zYpacJ1+5Ls326aS@q<s5F`pTkp89cS6~u((1x>4?5oM<oA0TxK?tG}4i+cC6(`Je+
zDWaEWbh!{QtlsE-pL-yWnB`OB^DV0STlyfJnzJdup417+-5Vl2u8*fZI<**PW9d_3
zi#I(@z9nzAhBW1VyOZKOFUx|hmx%>T^vZrA^!zpN=x=$UkJea+M?)s@K{6&_;ueYP
zl3B73x@f}(xYSap&Dcxx0Y#;AO^NGeW4o+WI`{i4mw4fJ?9^IGF*<jW-6SqSDAuSl
z4?kRB;UPYQnEZeiRp<Nd95OR)9y2+RF{jprfXM)vs?`WF8ZipjrwiBXg=JlQ9=5pf
zqmMz#X_C+#)rBP{uIt4GMics133MD5z8{5VE+$%L_>W9S#=0bOQbJTNe|X`?A3yWq
z`RcI>V^Aw#gG48UKB28fZ3pTzFQ_F>5gRkzs?2yxoybB!#XIOIVW!%3>L_5o(WvH`
z=H??dM;~;9u1=%$)(M)Xy$?hJDJP;VjHa~Q_{xpreuvOWC?=6U2YX0xWwEIb4idnp
zF-De{7!pX!#`E(hLP#9#jpKOpm$LalshlJC``s<8G2&xgJvs0c84UAL&>DT7z`z*Y
zpOxUb(u_9khdN(08FM<D^NpVUG9E9Un2X4IcUQ`BKd+g4z-<iw88&C3GozlH4{@^Y
zg2~>y&+|AA5c7EMB}}oDg1YWu40B6sAM_q*y^$A()8%*WCRmota=Eaq8>KbI-Wk16
z=4R>LtUVkp8pJy!ltCut$kX=ZnG^3{xxatmpZ?(=`SaJGXswdf4{ZcEd3L_V|EffK
zZS?j|#W=*+1%eo<Cz1*|C+u+Xouq_ycM;voK^<-fE61Hy?`~_H2Lbp#!<Z9Wr*6S|
zrHf)I!3^(tEc-=qVD{j_U2qDBTN?G8QZx^qUyANj^^reuri+TYYTm`KUaW-BU{Zpm
zw?{1WsiV@V)<2yBVd^tf-S;2G=X&^$o{5jqi-FTJSdWU;%9sMlA&!)tJi1R3g(SgU
zCA~k=1O2N{k?0)CWfSpZctrJ7B{GInyjBF3NK|MO3K{Cuh7F^(ikXmu2cOmEQRkjn
zGR4MxE5@QktTh}h_5Jqp@e}F!!)%m_qjyZ(UNBOc5yC+|N*`tcG=sJ9`x+)<X=894
z#RvBVVOiot7G(wrc`<XQ5%ykPpe%}r&!1`C$250^(1RzlPBgT&PBcecbka{_H28J3
zz;7!CVs`bm*&r!gVxZ?dg&Xc0HucWZ(QxrWA-(x@sp4wgZCSW%D;fj0a`3err2YO-
z+vXJ6mX*C!_O+2y;`MfSt~f-Bbtng|^l2Ay`-fTcSpdz&LlPz={)+)4<Akr~VIb03
zmcY}8gpxhc(#%I~_Q9wFkw{)PVwf+#bQqB}Z>;MDM(4QS>Foeqn76ls+i`cHOH_0o
z{OcNd`Se7N#<nE#npt0-xUMTXtsLEX{x&AmJe+ngTv6)N_NKUIjt#a6KgkJsJB|bv
zJ$VbE^ixP!DQ$PYHZC9zR-Q`MHkxy{>QTg)(5QfG4e}b;@|7jO&<^8>4eLrT1IwpC
z%5@mv+L0lH4<A4CvOP0e=j)e$WWV2Clo=odFIaqE4U}5A-QEElic`0L?Z?hl9p-fb
zb(l*2shCzly7_2q5dt2;Y9@5+wAv}fqqMtqnsrJ{<mkB=h>t~O;1#&}#7)7&scP|J
zG6gTRP)BR(u!mLE)H&YTpp=HGxcKns9B+lAS5l5_m&Ce8wD?Sg;NlE-E*zEHO^8Xj
zToO;$<rGE#Zdb~g%k>GF!B;w6I;S|!rVzHbHaT$T{Z#{bNvxNZK8)9&|4g;c*ROx(
z?du!8SB4ErA4D6|fA6BdfV&a=5UBKh&_!t&Ed*@I7{XpFU4@hvPv#sm8dj9nI=6Ah
z>_%^1tOT(+GFKmm`It#}IYg&J?3crXI$3?*fQcV~;TOM0WQJ6oT{=7z=fehp3FALH
zY?*1Dj!?=>a}6UR)tyj`LBv=V-&0G~nRe+70cI@M3(L#POh{!GD6U`^rXDKH%=@W}
zL7k%Q=Aq2P#TC;!=|A@@k~DmqwP`?*fCc+!sEAB-)YBwqq%JseJ9Q#<1~N?tfYpu-
zVJN;cb;Hu2&BDfu0Y^fLy`v3GeQru04Qth(6`j%?J+&AUb%}{>UC7sGhI){iO$V9|
zaYQAYsjcVy1t+wWfA0Aoy%UB!+!d`gN(;1>*w#6s9(jZ~iOjUunF{IWRcA;;nA1@W
zq))^-1doyq!Nqi(;YY4l0pa+Um|A$<7IIF=FnZm;!D8kIN>^os;1#-gs8BcKD2=^T
zTJcF&>$-Y?j3~$b;LCo{zc_5$dt=*H_FSiSVA>=mWSo;jz<V2%z0%rXNlp?EFDEDW
zGy{(6{!qx;tT6p0Y}8t4<zQGRFM*)mS?tF_pT7E|c7kRu%R)+vI|qW#&rZoFUe(rU
z!zmE=sQUyx!>h-bd@|m;C6-mli(sFh=iGzn-c(&UioZ~1!w*Oes0CbmB+xzC&?{tz
zt8KzYAw?1yMQ0dG0usF$r4|(DPdo1A7E28A2Uyf5{O%i-QyX)@#qg<M;Nrt&xw6Dh
zT$Yc#*PS-LaNHYO4b720h9dwX)K>hE%#lx@{={><IK(y@$9<ougKx0SJd?aPPp3SN
z;t7#s@?K-kdWc?7mQo2!I}?hT(wpa*hv*$ot;LH&%>88J(JwN3=e}27>z(VmaJhJd
z_9?#cAx;g8VTvq0hPl<sc)Yky=g4r1Ma=m4;e|`Sa*KEV?(hF$&aMz>o<&i5rPX<|
zn|DQN@n;+H0^+DX=2+CI`-%>Iz@NX@b;8@NHCijwQv8~UbH?YYwb5Ip<HLwjT3DAW
zqRRa^IO-ew;q*cu_0gHz6o8)mY*uC2u56bV_qw;8vhNHH5a%z)$o(K2#xXx<)Id^?
zpOs-Vw5KsnetU{S!*q~*JJgvYynT?_aU2uDtfX}2&+DY#{BsZUz?r=G{FOFg{Ij^X
zUuZ={<)PSD5z?Y0oF4EAot}?xE_^}6v-bB?&``d)0#wPDmFo|GAZ@E3!nM+i$DDUj
z7t{U%;&-hAm;^#hRP*#tHJ`{RqVs2unfQ86l{_;}rtOlGdc+&0UnaFqI_PWw{2wO|
zhPc*7`pn?4!KlrL&^3=vCw!aO=VC=XTKHD6xhKWs6G}BWWqXSF&{NA0STvz|b)>2m
zN<W|;v{K0)6G&9UhR;NaVItV-fq~tI7mP8|PxPqvZ?go(43CNl70*%d9Tvcyc3qV!
z(@i3V>Kq6qOoVq3Vq)E#{?h0Lh&R2iRXRt$r<8uxJNny-fptYgpbd}v?S}_^1ocC}
zf}+W-bhWw9|L9(|+YleUYoH;zXzpl~y)yPfL#BdVX?t_8znzXM<FaLzoczJ)-RXzj
zqhYrt&Rl8*e01nwaMay{h5Dqxx=d)y)!fIuaeKYfkLHjOZVS5Kk7+j?m_(K(u&jv?
zeGKzMq~;=JniysNejU!Q8$L30*)q$Lh-s?xgNF&x&0**%`Z$F(K?4!bE_dC#QvuFV
zSK6r4rsmajN|~HiwtRWyZJUVyom152ymUpUO|jdUG>QvX?6hu8<#JaK43Tq#?Vh0E
zFD=BhFop}Y_7W-gPPB*;IMIOCr<D${=7LjOl8byo_8=x>a*ch)(>lzq5@(k(2vfXd
zGdlRV7l(XHqfv<=vgQlV>!%3~wuh#x_u|1V;xVe7fm!6Ted1XH`~5%h(@%d-TdQ|k
z&sPXaZ$>Lq5!`)>Uev%6I(>JcP}70nRFibU4o7I#u(5`fe2E;-u2%1*_;aU$QM}8j
zaS&$LQ;RV+r!J1#x$Q1u!eAUbeg76<^m|=42K#-d^~UXf^I~wcUvSnZ2kxwADF-R7
zeE#82gpk>9H*WhY_v7Ze8N7%+9RrA?L+3QNC9eGVH~$_a^ZNDAwAV_v;`X}p+<6?X
z0jyJ7^9+KgCl(dzegK71J;9I3OAir-x<4SC2b#>;9Wk;l>m%>L#+b0b+bg3giX&%Y
zgA_A)TZxNLFXWt(ICDU3rW2lGJddmOXxwEwo@@+|=5fWn6Lfa$_(9+eNpJ+J2J*J?
z{KF5d&mUZv*DK6nTnhO1<e#w*BgN>k&mj@w;t7Mt&agoe7dg9YW>D_q(fN;a)`<DF
z9@9<G$2WS!3lx169e(inPBIWooT?q86KrtKbT|ijSsNbv-TPxSFyiwn+VHcdbkD4a
zA##Z>R(8Sh%m^`Ws?L;JGf!0<z0umi=!L%DV04ztN?cczj+hYh?H2{2KUh(HUeA~`
zN$dSVu?+JNt$}~<7-jD7{@u?^ji589WWrn_2pJ_IjG*LY;rjG}b$cSDJO!-AsEry9
zau7t~)HMA%Tk>xz23oCi46Wa|sDVC*@SGXpA$ZmReGqaYEgl+k9InpPX54RgzP|bJ
zkJ39icy2(mF+KI3|6hBbQIy|CYT6j|+UGQ;Ixld0c+PA0Kc{$_S5ObPDMx1%VM!}%
z&OTUY@6>yzm4QU}?jFaEN$}})>eB&p6mt5eRy|d_WB$Buo>s>=gNGn5i^m0@;Y>3N
zVg3ca5jjaV#cby=p$U_9qcjiqNhuJMZ+6G2x>grO<`h}dMj*|X#-Y|M=Goda?vz>G
zn*{o3UZwYrE8(Z=cM1j{Ui}GK%7pAR7$GTWMB{~RyK;THvu;lycVb?MDN@Sa4|t5y
zeU$H)_dk=<!kRZMWYTh>7q|BKbqgd-Sh}#pl{YWh2~nqI&_7>q(?{GomfQ_+y*!h*
z%_md#gW3u{{3Eg#zc1c+3@#sk;4gmspJLVc^3#pKKmMLd$DRZ;SL@c&=+!*qLOSab
zSksDy!V<k;v?jFqEc7<$b#P8QOffRV-7!mFsJA<}_ZuMo?4ratsA`<Dv%`B~kEw1-
ztFCY8`(RY(8Gp}f-*@DMNvAdl`g9na!aCph%pj$eKmPbveEj?)`|f%1eIj}x%)pm<
z$jB5Kexm{5^AA7r`44}^>+ME5Zph7xEo}0+s&3JpV{}ihIw#+=f*F>52+kSH%WN6O
z<Cll)EIG3++e5@B>VuHBZKJm~RrE|+WDdoeLP8BXwGP5*0rMjH8-4O{8aSuoB0Pym
zhz~KKQxfCx{Xu57gAX8!QOPNhKP=>@XVP|cyIt#)GU&C^TO-QAqHC{uA2U|iP>rPI
zGZ_YkRv!m_W;u*8h<TwNg<ek`gMW_GPV2K2X6V;M2hQn;k9-K3qt=m!D6jWMj7}9Q
zGbGGh=&SmRvB2tXADP-D|5-wvL2)s5R=YlO1pH?@l5{#hVi0N_l;#75_T!+{9UYw*
z9XW`L2dwo}$oD%jeL>iNv1cmhumR7FXj2E&=bW3yX*aa_oyTyLMO}N%tdAdsau6e_
za+!_avTbbJlNYkH5UaHjMdy&bN^NuCoBRvV1U8Jz(?Uvtw^A?_SIujnF(Aa58x9hc
zNFq(Oe>dE(U)<MOs;AKPli!Gb17q?jghx}x`;Ln6{Cwrpr)M+;+TI`DQznFR&XC3i
zb@T~~8xk{gZ%;?}x!9_WY-vJSJXE15J}){3G<X<LOp#?-SYjk-oN(C*;qePkzM~zr
z8K{!BNR>O$Af`wPt`4uYavX&^Iw{P!(2y}2Hk4jHm`F^h^`NxQ)AgAY7gPpEch38M
z97HujPFN4Lt}Zwlo!T3<H>_7i@MvVW-sv~wMLhXWXX$bVd#Ij@SLVSs6CP*8CSvsF
zSpiG*+1t;TD?j|jU-J5o|1hnN4$qzbe4`UIa+G)8j~h>y7e0OXBPxlnfBrjK-6x;c
zQL`zi^GSqJ{XV3e$$23!nKf-(wu@5;=7KvDC4`mrVw9SZT38pKjP`cF&FOzawZVr!
z{4Ia`xBo2y{P}+4_`m*NNJDwLe&mPGe`G)Id?`Qsyp00glo(f1zPJ!I38OWvIpQ=J
z9LGU99GUw0&p#7q^zrrS0}<s_e}X^%L^&Eq>4fA5yBA|F4s4BwHlbCYey3wvx{NcI
z=y$u0)(GO3tTCGV$$jho=W+-MNEp1l{E;92_}BRO-kDm;KqaCvxmdT?IoxTUp5li~
z&YA7`qd(J}|DQrghz)<nVnWUw5}OuggS_N9>CSToe7dHRmSt*M+&=xViHRa15f>K{
z%R~<p2r0ot1CoYGT)LL#VVZO-ol!kXw@pjxw}wu?9SZy+d??OY0@I6LOPyH_v!R$s
z%UKZQ><D67E|52l>H%+W?>F9WZ}j6Jt8qzzK8)Ud!e5KkkvTG^GKakRA?xq=OfsFh
z7z3e;BNJ)P0THvAWg<X8#6`TOV`c#OcXErN>mcGRj%+kyNDRhfM0yB|<G7mq-Jebi
zhatR>@gYU#^9YJN*?RAk<6w!AI%$X$vri@r(F>r~U>qFxJGb|PS`TFSgkA}A%IidP
zdde*KC(?Ey-0qIJ;^UTwoXMmv%JzY09{glS;bk^8#Oyi^`$hHyII62o3lEGiI&n$F
z%LnxN6PmIM2WgbDQzQ{Jvdn_IkKy)5;9n$X;OXPW51*be>$H2MzCpk}rvqb%t13Ah
z_8;A-Q@j^Uhy)s)Iv*w^P-xhU&Wr(`wuSAI*yk+jl$4J@JaK*6I7(xTGGi`_Lul#`
z2yJrAeRwA+O43BsM8Q?A)`$CB)zzoHdro(VGA9}?1WjH6hrqXa^PX-9{_lsscFy4z
z5n|4B{=7e&3)OiY5wMPRysJNUS}(j@mMIGIy{a|1?~PKMr$`Z4^1_mmi$+@IzVF<R
zJ9J}7i4<32T+xu|)*0ru6=4doy7QF?bH>5+Qk&}|g7_q_-rN`4d-uX%RG*QJE8cB5
z*C!aI?7ZE+5Yrz}+lZRT%R*RuaK^H2R55N}?hF~E9Nn@N65IC7)6)l9^#L^X{+0V^
zL<ychP+h+frtNZcM{I&+w#x;Pz#6v)2X@plr~GxGgSl4X<I_jhC1XR__X^Dmjg}Vv
z^rye&umAYhm<|5sum2m*p@{JB|N4K<$LCMn$~#H^fq!WKKzaM({9X-QuNyrl%DoUh
z{&DiO)oGB@=&HQm?xgsY%k_y*PoK$YaaH#6cUbTjGe++~RK&Z7;NsF@f?7xM4#FW`
zC9RRP{Js+i$LP>y!uT^-iL>3sDG>uxC9PMk&mXD7BZS}Yckai5TMxq=E1Nu&s1-`_
zZnTdx#3t~*@AT<3xZQVJ8>HwoqsHLc0r_^&p6I4^qx9mMnbCaQxvGn8nY3AJ{((_K
zNRL1-l_~IQ1DUhxQ%d9z=IB`?#zYK0-dB~!0o6luZ+SmyuBjOin@!IdHYDc-b@xf}
z;=CNi5v%#h9@kZbm=e$7g_KuB0=GK2^_{oZSN{C-7jFBG3Oucu9KOv$XftK6_fG2-
zEsFIzmHw`|>Q+2|p!ZoM*khlWPzI+3@%(Hi->&(YJ#gCe&TEPK5F*4a;X}mtBb(wy
zQ98i~Mv0<nBc|*+iQ$py(i>L49k9JOzP^41>-2RYYzt{!h?gfcW=1_c*%4^1deUki
zkJ$*P3rW90#AwQnB&CE-gl66VFD^!>54Q%p9tIF^9LJnZ;HafNgp%j=I}0QmhV|hn
z@bwwJJRu>_Yr#gPR-*+bW%g~IX_$U?{g>BoOCVe`m&?NKcH?M`G6ri1bUI_+_*-ct
zl)Y7I=^T5Z9<BgQB52NBmW%|YAyf%8Q!Xj7ZHbRB7q%^O+wX`Oc?m2__5`hA->Th^
zct37fYjhRL?y;0<PWrp7>pZLjU4$hoVdg^Hn6bMvu*h#+q&Fk4i6yTyLCkdybyVm=
zKP@Zbo04x%o@1XEA1A4tH~4%39L}_1)V*-LA6TzMT}aUw(&yOF&V6t0j#wfo1-5M#
z0|IYvJ72%PV(nl{7q01vxIXjr^wBHUV3=*BdeDk*W_>=`2v|tO?a6JMwE)B8lf|eL
z!t}^jd_3w@3cIerJD6Sy_X@qgBJxjc`4ji!#?NoB9KiK*<@)sOo`vpyUQvIMHLYB(
zAGlm!7`<^TH)?kgOkTDr#+=i^g0OBY$Kj)W+c4H`<N5l;5?4xX+>RX?jwp>$Im(R?
zR<_GCpI<)vO!wj1kn;BCwxOqIet7u<FPAH|S6-ezvTRqPD%Yn!^7-SBe2&6;{ejE!
ziT|Ph_x#=8{T=uB#${=s1M7wmy-;B$-ih<nt#$Txu=m0bPgl>rK;clHKU`V1NVq9+
z5w>d}rAX@@T#}as#BH5=K&+i5M7B$0x#l^s?f1LRE_U!u!=XCEYQ}jjBOw@sM6HFl
zw=a~V@cQ*<>VESgKL~kUSuf9wVWhJ2{`Shjj!u}o^~TRX{~39Eqqjmm8ZmfMsF=r?
zTa0v34|~zcP>;@0eEKM@PV@qnWh2BHYi@By;|}uT(d?*O*O+ubYaO%ZgT^EgMOl(t
zL^UMVW%EH=tv*72tOpm#oyB1gu+yz@(g!XYR&`Z;DF^%gPThA#AKwn#+4UMgP`1^J
z079hJj`c!2?!3Kz<#v0c4X0B}DO81#ea1%b-IF+5=P;lBa0Vhd?A<(?*=A8w>Wpo7
z{vDYP2%XLl@wj%eX<a-IJ-jGB=Xf}RIYXA@3<r}5sRuejH<^fmG9;0<7nXHpjK-Lu
zKOvm&%r_5ztCjb+SF9J76xh@YONkL}%~9l5VKj%6MIMN;2(<3A7La9rKc^-h6wxsH
zi5hWIH)kQ_^np!S-;0_PP4c48!@9lL@P@QueSmsVvR<A^*Jp-x_R;)Y?BDV$zG*LH
z(hpO=^()qIn_4($%OjEDR4-k8jAd;DE0yCFh{C;A%DrH<!%?9$mK3-y3!h$|NI7xq
zonwGSLz9x0j7GP)xo7{}NSycI%Ry}w8Nyy0_tpqeiIZpVV~}RhO$v!9k=6%M5;+E}
z7etLVx~Dx2-@t2c+>Y*1#4GMg)){s(1A~n7#b^+i+`5utnq9Vg`LV$<lgUP(lWxr;
zf7?-M@70sRQl)9275^Uh-nkX;Zl5m7x~}Au7^73h;C{PP-wt&0cS4+E#0&!KmR(pm
z|3B@O!VaVJ^z?z34}aw8=_A7iZ(o0+?yr<yy!)79C!KS{Lk#p$C*0_8V7*sbt5ej{
zU~81|&fEJ}-uD~NmnS}d{*i56L5;oMh#`R)#~1`%*p>_1dSMKA4wNxSQ&pSOLfRHO
z3+r`dSt4J)d~uKdenZr4TWeZSjU0WDCUu}8OpmpPKjoB(q8?#AloS?+@760%*JtvQ
zkm}(-eF#MY+w&)W_=~?}T^2rl{*fO){t2<bKmPQE*Zt0VUr}{I;jw!be#$;dxAitV
z2O)3D`kat#+}k@bt_<yb{Cwpv|N0ZM7%>Mvetafp<=7jod54rX7lds+QcEL?vTl*(
znuz-QMh5JJ>+Kvm<Qg3B43#z=9yYuny5C+I!#M7D%CW;}m_~voa$dMRJ)<JT`+}H|
zLgF|Mj<Qo*<M#g5JKWwK;aeAST)xdP2$7f;(&F~o(ma)}+;7;vquupLLH+rRITK>^
zq(c?1+Y^_|6CRQ7FujGTy*Wc>x?$lQDJ$fdNXzVCRXn?(d8Bv14+afs`un}m^xO~M
z%suK?C(#VORcth;O?!I?#f0$itVa=Iju=Ag(=F5*b>BJeH>c-hj%uC}xMK{6?izq^
z;Z3dkBvuutC~P=oTWjMe#fzX}$h0F4j9}lIBfsy04*4b=^$+Y6%tQqIVD;cM>C@3<
zPO#|YzXxEwJHiqYDKDO4s*`G+NG9TW|7Konop}bm`B>$YGciQY0aG?Qz4~NbJ*k;f
z06MLEwHA6?JyQaC9O!4~CE~6oL}mft*DUCyfaa@wY6Z}#X%c%}j}QXdeMU;383g}%
zQl(9csNi<bpbrAeuhAHfBIwtw-&72YK4`V`{&w?R`##uP!$dj8^yimC#VEaFNAtj!
zQlV6?+v1)7HB-mnZ9gd8*ft+U8<R2=-j9QOsTj~rczt`HmW&zq+412Ub+BzK&({mv
zx>4>sOax(cql@2QwK%=OLNh~VyzVzovMWe^&`v_<KeZM={M5WkYz=F!83^KoiDpu&
z--6)c4I9H@@Y<&p256PzedqRmM+UsU9{%z~rA5I7`@UhsIgF&YLAB0a3ZpiXM9)^J
zV_IiiLlARzp^XhfN-QDJYK35ES3bP_z|;1T4^Mxjna3*M?r+@pJ8AN9;KCI*Kzjg>
zic(9bl*apO<+$IuZc2_qBxBXDa}1Gfx$tuR$mM!PbK>W>SCTI15JGDNNq+DLrJItM
z4N0yqNtY{6pMGFjSMs`GHh6peN18_7|LGqIlWXi9dhks%rG*?b+tbR^^@SyE+>h!R
z568|(*;W7Dt%^A&Qd+pKSN8(P=rmR7eE8urA6`E5@#Q01PAnmlP5JpR{+iF9{vZ5@
z|NH-q6oe%QN^O)jSb{@`%NmhnH1R2z>r-NT3WVHgyi@8sP2O3a20#AImF1JsF`lmr
zOOCWIE^ak>q*G$hi)UdVuI5g^QW5;Gbt#jcn**4J&BQ%*0rDa~cf*>GD(}5}LS}I!
z%`;q@&*wP9#3cHN@x}e^tv1di!1F*pA=ptXJqU5RxP{DI?-WyJZ%#M)gw=-hF~eJ+
zcaP6DjAh%Vm@RU>K0gu-TlcK`R?8eKu8%^*TBpa3W-nr5oRj@vc8}f>T9~v6!vnQ+
zisG24Wj~?bITQ(*nEN*ehKJqggL^(F9Qqh<AK~2QBtC4AVj%b6gLr0h^O%?!lx@3^
zf*>{%1<ftC#>^a;NrinDB)vCBTIN(hW1<yPYhg3YNRSD)p8`3VhxQbS`%Rm${5gu6
zdBoYnC1taU2pUL1+(|TThvy-wI+GQL&T-@2i^z!xozA7vI`ufH`<+^RMox^5kf<0T
zFo!A4si^AIxX3Kra94?mTL4X*2GTePy%WM91{Wmjq%vwB<e`YTy_!j}ocBEh56tQw
zoZ}9Pvcn#y7<CY1V9m*OW%K&j`OZBooBsmE!2NcBDaXB14<Av>@J_o7qu0(*<rss#
zR;b-m;9G}M(aXxR_;}rY=4G21!4lFtrR>yd?E9VLsANrODWnv+Y#SN_d+*%t2c<Z?
zo@3(qa%IUY(#>z^XpGy<M|heK&Kc@i<$EoJaPh>dnN}FpMM;;uu*?95(YqHGQ-?Cl
zg<`e2oyNMyFqg8s*rtEWqORwJu$k)FdguPO^Zw<H``ewYkv#Lp!(}1H$fCh3T@t-P
zuMWw9_~>7CO2y+Sr+BVa-fypH2^_^o17Ef)8oZJZBTdM@dz`KC@Z{Fv+52RfFJbhp
zXiODv_xpO3#_Q{ypa1?V<$kBWC@CJ4k8?oGa8K*9UU+)>z|-}aj&Zquq>NXr^#>Q*
zYVoRi4gpH}%ID9YczSx~<>e!nbz@ssqDJ0dKGD+3*MIyo^{6yD5(3Kt&0yP}dETz1
zHL>Q+Hwg1VK{j`fHXOCOynOWPAMs+%e8i|qWKA0}_!#OnWiIQApd9;74v`N}FZ@46
z=X>hmAxABq6sX1}XVz<E=t>=fyv~V|(P8F&->K$;#!t&L%Ox@}meohL`e2hmZLTs_
z%rywfH$8@TdjB~e_<LJ%cI4g8sA}J&L;^7^ENO*ur`|j>o?c*D-o*Oc)Lhi|{`F_p
z?ZR@oPU~r)oV@fr1kOX|A;=n?(JGdF&kA8?{tu(o>YZ&l7`+isrTauOdM&*7O07F<
z-uUpt54=3T_`HtphhS?5A|6bYa`GZ%h7OJ1-AXuJI-Cmq)8{^|gYpYS{kf6bbRV3w
zqsmDDcyLiv#bRU%_nFD$kM1(Wh;^glymK!*)((Qek`kqMdUt(R{(n>VZ$FZx>6s_?
zd%u&Jxkp51R#sKBx`%AeVF#;`Agu0Ir@dJ1mHt&-ETB7G3EG7)vy0i4mV+72lEWFY
zo9wP*9wNfs%)X;8-q$>m1eQ}JfB@Tp?5?iPjErzMv+q4T&+l=L=oAh9@ai?E<&5+O
zZBvpR=6D!#3uzh_uiJXZ$3%=1A*#NjlugC9eOL-8x1G9}l=28j@;(sKbh|EUE#Mm7
z4tM1(h_L}|g@RY%r&n<%O$E*+7t@ReQXIP43%X{>MQwYh<-}<|qnBFIK(%e9?HgiL
zKj*YjJXRUrdFGTjEek0IO0A^mnWm@(fz#)wm<78DgyyLhXYBn;iea;<bVu_al;Yd}
z+5{yAPYE4L)5dVUB9{-0p|QtJU|ANH)1BTA-di)FpgiE^`u{Qa=uf$!_pKpP*tT*z
z>{^A~lqrzA>Ik-`)(lE#aY50Kwb4tXWF50zx6GESlKFJ$MBjzyZPhV*AF6r}dIM#b
zRXlW5%a%Rf47|e_#$#4HQ0@hnRe({w$EQedRY&5r(sQMkp}0!)O}#svh)y2`;}{cJ
zwN`uQns-WRBx!`JXffj)wQIHKokBv??{2&IroyIeEnKdZeJjXhdEW?=Pt0l2!bOif
zUBR>ge$d9Spb&4nxtH(9)y7^g)K;116X$z}mzkJ!m%e8W)_8h;RL^VFXz1(uky>_~
zs4&V~NKo+JMy(f1R&4BhV}06r`fwqy8^O;!K0kAJ`G|K5`?hN!(=?N&S;LTgB*mE?
zHgKMGSDyFv`I)Erk=h%VZRe|>|31I>#h09yh2^vmoYzR_W#+SfhlI#)fBP$b`0h89
zvN6TP3P?(v?-r)CAWqZM^pI~B=9C!)XqDHm-f&)Y$M){}#Bbk!$5%i5oX`dycyo8h
zb=@oyRuAOszVWzT$ffZ3^vDo2Ydv_puMFvwQSr7D^(l~Wj0$yV#BwJ?*9XsVx<ZI9
z63z;l!dq3Z>r!g#3M8tPnmfT;AzGkroqhlDH}&7gS8(n&833m{MmxA)J-mjHXqUm3
zS1pK|+L#^IEphbCo%Q<6x4-!-?%%xOe7=Xa!)T^je;GgDiUT`*+<=xdW13m!1@WG(
zsO>rDtl9<V)Q&rjP@ryN<IX-BpS^y|=bwLxbB=Al(ptS?V@KDikE=(QE4E@(%Wf_e
z@8a$24zs*E^Z;-ow+sWY!#r9?Xeq0rKbZcE&Z!RKMOJv*)8mv(2TlW<x}(^$B$|FA
z1yA&jQuTWXBAn)h*RQ_d&FjyIZqVz-*b8lF5S@&{a0fED*?rUWpuD>yrIQv|EhAOi
z`J3ZIj#bcj`MHtX|0gEUli?YolJ`Q8j-y*pmlYH^8jiTk#CcXXTh11eq}M?VQP+mf
zGg_w%#n*cun5V?)bVrPVRx>oA=8d{usQaeCrlHRV{DD#yV$dYq;FK$&>;=n&utZ!D
zY&xUUu){W~Ah!C~CO&55zO76?Ee3DXliTjU(e?}5)Az7#RPp#_VObV@NThk@6g0i?
z<zleH^3#9ux80*Z<%WKIdgkf*!g}4Q*Mc(!cyFD(R4S+s_Bh-v*WF{quZ{A`7+gxd
z?TfvSl-wwzux}fA*J8js&+x*?O(T3KwGzhQX;UCH9XYnDf3~%8iw6$D(>+$1cUr-K
z8bZlCJ$GuZP!Ddj?yI~3Bb@W#Y~Sbo4OX(|9XxECyX~<$z)*krp_2AKXhf(uD%u&f
zVil|2N`x4hmxY)T1BZ7uhB!u~PCd8yKs8mZb4K~<g<MpvDwYecHlm?Iw5}mN(&6&_
zo^5^RdVP=cPMHs7XPy>9nwjQ>DNRThf-5i_>-w$!Om1S1Rrvt98s{Ch>oXxV*8NHf
zk?C~Day~Psi0l&sC|T1B#~5taXP%!v;JxQs3lbxL@a4~Vd;gl{v?y(_RjPH3A$UHU
z?>IfY<=YSMxSs-Z^dObslVXC0PH!%|7LaS+nbOQ0Ba}%$=Q#1RFTP?4f&A^C^OUcQ
z+9@>7r-gmjyzuvr-*MSKCzs5J=O^C1f5#6WzGvH4J+!A8Cxe(A6M=Ygq*EfC9AWbK
z#o=;=A=KQdUhSLS_ZzIK2{=5#aLO_)$6bWzHJGTi&h9e7D+{6p*imiC@AAu;p2mwp
zobaj{IGxWtJe*l`)fiUq*u9F~njK==G3oAe*pzPDp7D<Z*KIW18Zy+1Dqy@0EvAp;
zLI58;_YbeQJ1uOYiJE)UgT$Hp=V0Nh=~3@h?>u*RulfA#S1f5^Goj(OU8yze1K4RQ
zs+2-9#IM#RI;m@2!A21~!1a-0QxDruAH&}9^6VL+9i<yKPPR$GOHPN=oS31Yo5Nz~
zLsIStYUOlJpF=SQrViU)@LuiPqjiFhJiL0%{r!VFs;JPN7PB#Gy5q6m@7WG4#W<hl
z6Y+EhsLCD7EO27`7LI9xX%a+#RvAtOfnD7K7UQHv(b4qqqQTG#TX4&3{B)1=nw!${
zt}CT6H1^%%$wOCHPOB9Ugy@-;g<vbnns-L8<n6-PuSZgtid93@_cy3sWQxhA5WvtW
z9$hii!LH|SCW>JngSTP=sn)bwb_~Le7U?3M!-Sw0#gM^hItkG|v?eKff1!2QiA3*I
zKq#t#i7_&HCEy;f1H=DgUO#P3@T@SGUYli6J34mgx2hwbIvBN)#$b+~2_3i1Q{Xh~
zE_<(yIy&jdHE5mIVU$LG)-eyFpgJd>mMiOJ!}~=eg}p;&6<?(2IB>PKG}|XS(&RPN
zXdEa^kf!NrN|tL%gR6RRsUXlM4};sd@lc7nuA}pVDmH4PIn5AoUd1W2Mi?pr^Imhw
zJwngbNM%6p?4?k9$0v`sl(u%joGik)6nzhpeV{pOj<$@Zw6L-<q!Xu^hgY95&u0|?
zCb5y;5wR;OEG(Hmc6D(SC7(sV5d4hy3GXMwdqR(Rml5;^h%Uj&Q(r0mvo}qF%i9K1
zP`X)|`SOb|dGqQOL15li{r-o7sYdJMn#p@+TQ_pAyt#W|j)^HowI+3aPoDP;g6Gah
zdR8-kEjuGZnv|wCdSlx!$fVun_435i^}@Ps+&?@}?mxqYz-3?cXP75q2)us%8T0*}
zddLS{w=2*YHS_r49Y6f=9ch~Q?!yPZ`S!P5-+kaTP2|_Fc=-HtdTs2FmAmQ0{p(0v
z5)#@CZR)(hP!Z#(+CdG2aEasC>MHeZfNOeC48v}V(}cTbs~qQo?viU|{~Z?)+H4r_
zJSha67tYH<S{5suHM)HAj&lr5V<Kl=DYR~9rOm&+bHq5|7;Mi^cpqq;+X3lsR<dSc
zQl_mF$E(lY^5(NIlz>^A<<Y-b_cYu{^s^Z+qQ@8_uU@^uhrm`cYu>qTSJw4{JACs3
z()F`#wc%ng7ljTOS{sb0UB0s-fpOq^xhcXq{&^g4n^>)(2gtDka87RV+`|Oj!-@gT
zdeFb)Sj#(<m>9h$dQFH-A+pKN5YZxGig-~=XX`~>I34AD%#(G!9k*_pqvAX<&G<Me
zU!ZF^*r?5Hh~1z-FKz*KPi5NBxMoDyOCjd0a5+(!n1l!wM$h)#Rc(YzXVP+yOOx5>
ziXx-;-41WHyC3kQqc0Y;VT&O!Mx)oEZnUz~swNpa(S|U@M3~PiGL0(IPBGl3zcpK|
zd$*#;)b0wyvg-zOhgkN3bLvPF(Mf@RmiFJys|oWH{B*ah1~aMHq+0(@agmaXx=FIy
zO0yH93%AL{k&W>JqWarkKdBgazC0@c1*f}AKN!;KG|Es=hqvhA$vN(vD*e5{DLUTX
zP56}fVXJISBO^y?rYd*TZKq}XkQnMR5C;a#@h+-BO`N7=^~y9S<}{n2CDYn?x$q6?
z2wtl)uRx172G<OQRTZznGo2<KtBF=<&ik8;w6#tv>eqDQnQgZvU2pze3qv&NZ-j8`
zLa;q%Z;ftVcTdyZTo8~!9TktNeVseVUzJ?zEl{N7!t?WmZQXIk63BTcddK}VaXOvU
zRwIKtGGmmRs=1GD>RQ_wR%MBwmRGz!zoiPaUfD}#PA665N<>D&;hED3nc<S27&gw{
zbH+_GNjh}pecvr7lJj^6DFmYT_|q9_1OM#lwX-ZI?#_42be3tR)yA54vJ6D@>wb9p
z$otDP_aX9hdFGq1e??t4#xgTaCxTDpT6w-c>O&D7)kWK#1{cVsGEG7Zk=z@@3Bx<m
zyl@VQ+BGBN`Ff$&OwSvSAAaELum2Js6I;$iQ*?I*zx>00#6SG%4|Hwt#~&DdXPN`k
z64fRodaWfF=pEefcpo(s<z^O=+XKltXSS_MAD-qNnz=a4KC8d;g(%l;!}`1X8l#g!
zAf-U^f#q~!Sv0}0<c(f7TrErjahfOss|a3%+$zO@S|Xlc_*D*3an1B01gH8ReGuw>
zhbY2VM3_!z=JUOdmHW_S%vN4@^mf1b<M6VT!(sh&P89Hxcb?Y^PfzdIw`U6(OX^;c
zanq%|<O>`%h1mpeqhe<Xht>^_1E;M2tr6KTdI<p&G^=>kn^N45ZbUz`F^e+od#^Wz
zyjb@-#i-#qPQ$iR45Sq4MZ3>lGIhI9x;MQ*B?X-(q<K;BmVon#f)J`w0WtDjEjx9;
zQu9?;J*I>gv5cAFtiW)$)f^&})>y3xbWx|F`kp8fUe_=k!UXdJT>?jV(DiyHZ#yy?
zDd^AVn(|<pCC=V5EIKBp9F)D&>(1y(DaCBB(Rt!D<CimWxo4RsMD#rr`@Kh)QxG9|
zv2Y<3D7MjYL#2E-8>Q1{M3=+%r|5R^ilgrbHP_Zh$Mx-;5WS;`>IB5@C3SyP_I2g?
z<0DIq_+`d18NPkMXu8Vme+Qg_?m9QVi^ama<nYU+iGIN3BFVwbAc^35XYr0N?@zqB
zTgbihwA&AIt;+cxgXl!R8H48>0X`!*TntR-h4aG$%W2Vrwd-CfdZl?CD%)Z7yx<bv
z8?W3dejhw3O_b7D#k`#YcjuY=yM;8Vr(f(t=~W{qL4$D^jyX<DAu&a#q^>a-rRan}
zgyEZxfW2rqMkyA0)Ai7rCq8?axI0Hu@-*~=9aw;g=ta~=r&TR%%C7OxIag#fLawAK
z5naH?WI!Q(Ahfa~C`C|%f)tQ48n`LYYiEil&ZpN*X`z)X>vmCr%B*Q%ZFH99tfCzo
zceR$;^TzPXkO-}E-BxNDoZ`%85yJ#shfKjE%R<{L^E^{}VOdT}tTao~nl~zqu{rL~
zXM#A+VdCrWzUBRX<@=A{@%5knTl!WI@x-ve@YGt^%0_Q+N;AQEPSeb3nn}w8cc**q
zmU|qIJjEM?%7@6DCZ=WP>G7Fu+ZauQaX!3%$D7xmaTZ~Aq7acKQe|+zoH_X!FM*gQ
z9-n^TR0Y@dp{!v_;9X?2W3Q`sZkK_Kj&$1xW5G=JxpZ@`h|wPVQLuGQ;T#Rk$@q~s
z^xxN)2*ebbrbzOU6cep=My-^*QuY<!GLr+JBE{Lx+y{IqxT@);X_^@bxfV*Tnzq@8
zipjiJh*!5vcpvcA6`hxbG|#lwxo*lydEQpGR_KRq%Xx%)>%tUuiw4_>P_~ul+Q_x>
zv_6yTN^ruScLZUc6%j5$h;pO}V$dFg_tZl0(@o$gL*utU)j9Om8E)K)K_`}~=bTOg
z+{-UNx_R5Jh&+l3`=u&%?J$FL^q^74(QCxK_o_SS=mD}`FGMO*HP(3w-liBjx{p%=
zw8KsFf-8kyn+g|eq2&v$tU4_T$0*$4f~H5(1vduMG#QfmWoJHY-_G&#HN8XB?(Z03
zAaQ1dS;2K><GQY_*Q-H!bj?t!PHfv4)Ovu)Y(3kXT2&h)Q3ryPp^gHZ`W~HqnikT0
z)?5eaFe+`=h-P<NAGw$O2TKPWV;sgtbxWG&1#`Ew)+kj)fvss0szVemHH=DWSuwZC
zf^#;-Qk!~jP1gn6#^v(FoFWcAgQ87~g4Sp@b^edPei~<B_Vgmy6k2U)lOvuT^CpY}
zvlD)AiNw~pZk1~pOmrq2-Slee+FaO5ryJt3dsSryyZE{|eDri5n5LO|*4?`E9y(-H
zBhW>w=h0Px@15uV?vC?mW;kJMJGChQRI5={5BJt?okN;tmNetNC$~niu8ppE<JKB`
z&WzR+9)h4Lc}lpF0?vHCxfgoZ-J}a1dR3cWYlP@{eLnFIK7Y;f>WrLDTx?WXn$~@q
zE3OLKJ`_#6t-7Dg>xK)Wmb&?#)2p|n^F0z1jX|qBtyVNbrBm`w$tq;iPDi0DnttUC
zz1i3<7j<Y%jcGnxp1a3Opf)9qjoOf=EP^>{Cscdoa^1j3zW(;NJe<y?6nXsc;f7`v
z?{U%Nri2elcXV_l20EQfUWrp=_OD1j!03EiGtck7<@;a#DbL^iz`hp3bkCdn-=m4J
zwclz`N^KgooRsxYM@8IVIp1-*e^8i`6Xp<cwe#lgfj4(&CLg%QpftzcX$;~#sW>P^
zuoYJCh3odrnpa9!LgLf&Guu|Uu6n}x7;aTRz0t}*q!3~?ilRP<LeK{}UdN2aVbQ9+
zAF-zvAFDQT6@68E0Qlc+0pZ~YN-@GvEU|YEX$5Lg_Pb5;bX1@8+6e()BFhlO>%p<-
ztRXT(yYM3>`$)$-1cSkQLO3xewG6j0xITU4`T9f#uDw%4kPrwyFomQXk4*z{q6DIo
zn+SAYuWaK==}JI_2FGgA)c)ce^Ry_F-V#W=Xn|dOWv{xbR4bb47(^52#Cu&O*%U#I
z|L&%^ztK*;zp?xWj^RDE>iAxrG}c<!#Om;`4<|U$ZZmjCaKfA>TJ3C->Dc7McOso>
zN-8dPj+P7My5fB@EU!)(<akEr%&41^VChKkH+{iyPHopi!*f(*-1M67IydvSa~A_a
ziF-jRdwXw(4<J#So?6(pjpyfQ%AScKKzxY~9!6Bvk-AM?n-<u$4E%6(uT)3@4ycV-
zF}ClSW$0?ad0QP@n1I-O45q_y{+7+-oKu~Y@eIU-m;U%3YBiC&u1Q*h5HLLf3O&MB
z0!|Ni?;V3m?`nY@(&(emy?+Vkvg@YQ3aAdLbuF%5Y^i_y>n9ZhAtkzZ?5mD6tA{B$
zrc=a6$JiV5K)Lna!PeMv#izivb^4xpy6!x#yB>bMYizJy)q3%IVfH~Yn@zzPW5hc>
ze7D}oxpBGd^xl}0-9TMc+^1>g?%{!l`+I7yuwS`eS9;M7Jm*5*cgngUh7ZQZO;rYa
zZfw0Pamo#Ra?CyvykqUE>Q~l*AvDpQeBBKKg6{3QF%eWW*(pUQ7xS|4=FMyF9?tB}
zu|KcuT?vHFIRc^qL(WsjFxyfi?}h8SbACAU@bHGa*Kb+w-)L}8S1K7+G1jtEYo(Mz
z-irB~^`@@1vgeP)v@q`OsePxkOzRa1j(M4x=b3%?Z2N_h>+QieC(kr5+}+(XEoXA6
zJeEg3zW<)@zW;_VUw_UrO_Y5nm(19;%X;3fxTv3_ivc$Wr4`OId)c|x%wBdz)xya+
zp|t>}Kv}<?%ZCq?HPa;V<(I$Y{`{6MgX?<LJ@IH%gz6l1bVi7T^BwWD@cPYXO!J9v
zFCY2%{K(lQa?oqaBP?;^vo~)T5l-oz&)<B(dAUb|aNReyrzc8Xp%tE=ANly|k$3Oj
z@$tiZT!K-BQ4PN0m1$9HSEo(a2WgrDeN?^1Q(y#LAnSh7$c3m2CR)Hv_ZCmxJ2@+m
z=J3q_F2Cj}aeunAXx~l6c}{e%+^Q<Cbp{IPistATy;n-y5DPHNS=V%jgUJ55<{i+E
zUUfT>)9K7SDUj>Bt*q;X?x1>44?&~8Z6u3W9*H~A@{a5Jus_xjI(|26svJBr<-qsb
z=i?MX4H|>4{f4fEMn`HT*P;-qq0AG->Z`uTJAYH4|IGlOW1ZoR=YMlc$Pv18Xb=o^
z({(E9#euVx(@=7t6D>++(bX2>RIz_(+pLhVz;C2g%G%XgbGpAp+0z=*HNstbK}6RP
z;*PUT=y+X%iIC=r0|g(Vkvzw+scq7rsHotjZeqS6%5X_(0jDC{>-BmQiw{(<bAZ2`
zu?m8yh%$~rb~|O?b?1JFD2%JnUru1)h40aZ#?r^(7*cPbBG5I9*e>6TSd_6gyv{&%
z-7K+3=ECu3dh7J6g2>Vaz76~s7E^uzi)@y0Ocivds0e+rwX(4(oKGrv$F9fL4rQtZ
zoE>ZHzy0--ih))J;vFFh-8DsO4xs67e|Q*Ocq|%&xgzA@Xlv&C=PR!`@svAz(;Fv6
z?VkG~jBy?bp3y7O2b>Y!N-1pXPA(eHdtD1#E*jJ55^3^Q6?dbGMU4d=okCXx=g`Q>
zJ@4%6Mp<`Sse}}5jHGDcB?Np|6>N&SRu}`GoulSz;~B;Dq!9GbZ)zv(rgREEFnfhr
z)Y93u%w<>qw1h}Hor%+;33X1p$8PH7j*_kf6(sGo@N~WM@p9#|?=0?2Oo=%qLQukD
zYnpj5nvU$YZKvej1eHRZb)w?ESMf_NT(^&;C6Q9%e11(zjnnC#^ZCq_CW80Wy7Qbj
z)@^00J46W4SvI@=oJ$J4|M-E+^GCHl#l)P{uK9HNz^H|%k3aC{i{AsFl)~NpiO*iW
z;%=JBHS_)DBiDU{ka$`zJndJ$Hzj}0m0Dp+cX$^Gv**>TFZg2qIX)%!+GsU%-7ZYa
zi8rr5=gXgc#lz{I-~!)nk9^oa(!z<S$7fzYoOpdY>)up^<uoy)`Sm_{suQa7?7@@A
zV4fyi%e?#cYo4B;nS#;42+G<YTE%uWNpasYqbu*A`)<>#rsLoy;8DA*<F-r0iN$MM
zml?;454mUdy^^>5Q*PD2cdt1pmUqjAVFXZBZa1P@3~WqlKo-*=MdwH<AR)5mqV_n@
zqRJlJZE&NB>L}LZ)5Q6_@Y&nXaeZ*PJeioV>N88<cn%!g7~@82OL5XfKqDW*%h-LG
z)lfteO2x9}+vsYqo6m%hR0Kz-jDG7pd+!XV5$CNRfFw1c9ZZ1aH8aG?O9Ak5*z2fU
z*9IEXK8^yxtBtNZ{TbAfhQ%%0WON*Kyh9YJmSXR8{nDk9=y<21S1(R?(_L&*rF1@B
ziFws@&5K%NYqqSHMyaYl7#Nm#D6%)CnPoXKYF91;=zFQ(nxfLJ7!VQq5eO#`W74oN
z7uYskNygjn9lC}%K2GU)pO|98$EbuyC%79y))Asv)Q85o>y%zoPQr3WI6vT=r<9G+
zfL7?$p5bob4~<|~jY+T~&U=Szy7Kj?j)0*E>=LzMz#3ge;R|3vdAjXYoq=?n2n=+c
z<iu7c1LwNpszYF2&djGXAqGITBBfOYBE_hG{zr<Be_AoH59Y}cy^y<7@M>0TU-U}O
z>vTRAwOy25)0!r<SZJ-YSM}yk!IOd}TxoVg(9yBen*w80x7Mk3@bvtsmAOSyHU@KY
z%(El6f*Yd6h<6&>=*7^OLX2KfoaXy>qqAREN@=&O@A)({pU;3kz?}(fLX6~ExrpMJ
z^D%N68ctJ6VcRp!h*h2=BuSw{y0>c7-c*Q_0^J2Fo_%y0?l!t}&g+Oc=*=@)A-BeL
z*RYl??*P=U-Fm)!q>n}}!U27#rC9{8j*eoC%=4r-y&Yr{Jo9|V!`<h)eu-DrG{k7Y
z6diMo(G3TyRIYB!bbTONOsPN^jl5pjwkvNICA_7W7(KIIKG3!+y(qF(eCAc~EHN_M
zjVz;c*`ByI(fwbo+A$3nrPIsca=DN{t9Tq?nt6D5&H3(u)A`I^Hl8jQmU&?bGp`<A
z@o+w|EHm%xg+KX|Kjxb~!#s2Hp3sELzVhz<dq%4a2itxn3p~7f&Cefxi8$d`fB7?>
zOTq7zC_)&T#MVVacUoz<sD)^0N<VDXB62addr^LP@0~Px1!Lu|`4*x9MT&E8jBbS%
z{+3y>UU#0K3YVwf?P9=tM=6_ijsZ8SdikB`C>(vDGltsXYQdTRJcdBCE^IW5FmGl9
z^m?G~Rc(uXXcRcLa({QnX*p4IwvoKe?QRTwJuH=bZ}z{TfSMQr)L=Vzn|B|bx0u4;
zx^uBZA;iS8+^KVD6zxvk%fA1Z?ua!mhciG!j@%6?>)c;YDm*|m`p+-_{+NOszsGy8
z`W5SrwTo5F!LVY)IeorJ8#kM03KP?^;M^e4@x7^Tz}e{9cHrV1F=`Xyog+*WBTi<W
zJMg_ad1yy*B?A{D^K>Fjixn)I+aN=Q+DEz}HVF=&#Jwvn*y&``Co~VnS$0cejLM!5
z<-QU3oVU;7j0@iI!IfTK1dSo;3TX7I0@PX&7YNHlIzJHRlm1-2E08F!D7Dezvdsc|
z*gw2GKEI%2e~zgQq@Yc5HEZSJBpCz5MdCCQJ-C{Y?ijhF*jgjqigFD!2&xkV;<B)O
z_Br$YE9PZEL>V%B(}d3h$^T-i{g1zXQZbOP8P|@qH5GSkrCFC;HA&CwNT_!xmY}yJ
z&*VjSC*C2o5rZQ|g+~YjpKOeL+~B6t9t<wq&gG(Mg{rW0NDMehY^CCNcN;V2Ub*ZW
z<8tiZg=tE-;b^&19Trb{W%R+@&+a+jEzGBx`@54m6>6nex4zd(DH?&f=gb&`bhreD
zk^a0R#hD-~1{=)|gkx8aexRqAxLX#S_f#NP4dEFWfh!)nvX`vuy=282_hH`U!DYM9
zABFu<FOGRp>zqTY+#o`nXPVi12DSr@fq9+@WAJeQ8OwZ6S{Cp#qg(YXP|C)(J+ZA9
z^1icPF09uTi3t&*maHJ9Qt7?m-5~E5_Po;ULFc;qqDL)^+PFSG^ZtkLxO@ACy?n;H
zt$63zTVY+Vtk(;ZUpUPtxNkf@e8#)^xB9@|ciw&TTYfm7IrmqHseNC+dBfXR4~Ta>
zU9W6gCOFS|o>``eh~w3K$J>XutdH;c?&EvDjuR7(wd{QR^{;t;_XBO3d4KuHm*RN)
z<}F{n{eme3P7-<d_>MpRAN~dO&hz$s>*}u08`BhNNfqRsW+5;yRxYs-v!Gxkk%9A2
zTU8Z1LQH|w6LoZQY%D!8Pw{4Rn^NQ=m9qbCy9vaKBYj`G(jr9&7>@&mMzz}LG6=nC
z(S?bn;3oDwl=a7Z8=O#05PXRLVvJ1FOsm5*bxQ1uF)*B?7~lpDi<n-s11%=JKhV6Q
z5>MggLV<Ck%pFufC)#{yc7a1Ib2?bYz!B3c$LY~KVi?<7w{K%%XKN3A-~PJzOoMy+
z%R^p{LPfjzK(Hx+vsII#pY35v?MDX-VRUU0Vq&;J2*TtXqh>}~kzUc3Re#<Pllt>9
zeXIyRIFg@{aNvVA*)i`s7)EzG-Lu?1;PCqSNi{K~H>|i<U4Z>Kx-x+ziJ0PTh0;o)
z)ruP(FN5eEa|&9FTVi7$>LQxv+4wEmh-!eG+yFr#MYRMQrF4*WQk^Eke4;xQDofw!
zWvAAxtQa&ty4lY^KKY!~w&*Ecy&cB{fS_w4)k+AXiB7W^h@bFjQFobhkaco_i2=o?
zIePs#MymwxIX%4QbpM8UT5w)h=(SbcsKnqEo@djn<C6Q2y?#<Lu&r5r`3xjDh8Id3
zv_9y$(n4j5Nq6OD#|r}YCv_WawQ;SD+QId%qaW`HXayLPiiVn~g1Tp|Fmr8Ot}7*1
z_0|TDWEEGn-r4R>oaU$>W+{Yqqls`@CZ-sa>K9QuUGHqUl1n29Op{|;65>Ub{Z2pV
zQWYV(?K|7L^ZxxuuFosWG~s6Rw-!~!rx01DXk$n<lj$Z)-2o2uPcJEPnx@<2LwAk3
zZ#+5^t*$jHxfHb|N*@I0iReN6@#!O5Ep#!9kqn5D5CSPiCPQS_s_)-$10Ov;24y?6
zL5K@6F3j^CF6cwHmz|O~F6$GQ$M;;HK9cv1%VlG|WKQ?Vx~NVmyG9Y$oxSYz%OlUv
zA9+08QM>9EddVj2tgO$EeE8vO%E(NoiP<}9SL)u!=VwZ5{Or{OOPYC3GhcuGm!x^t
zQCcfpe)yipFTY?qpOL5*ozrRI)%}@+Pd8BY^E^d@7n&26Y39}G6$^on`^x+G@A+(b
z;PLs9cRzeb+X~K2?5z>zna^Lp;q~1K>4VSiU-5AFn!Pr@zkE;Lc8#YE9w(7f)V1L~
zH=s<nhw`2?9gp)MhHtGv+xrX4y?0{mG_h-@H+9o=E0L>;>duR2U4FMY12GEG>nL`N
zaYNsFR3{|p&cBR7?FtVfK<gSkD;=a&eQpFt>t6HY?fyw2XlRe<lrg4AE15imnlrr>
z)nx>oL=JHjE4o|n1{xE*Xm?$DyM^`$I$=CI_CES;UyK+hzjpTZ$}}aKh)$9|q0b%8
z1P8Z+UWJl2TDG0CLkj^F5Sms2cWaQup}YNIhwl-CYomXiW`>(h8FXjeYoVe!3c-6~
zFi4U4)`M66!xTK-4La307&;O4G2&ySBf1i`Y>5~nE;wSERO>SIxjA@7S|cuZ%;$T2
zm{ciWiy?|Fxitn2f(kYz(f_y|+Rk_hiha)He1)pPPQfc(Hin?$WFHlPHqEBzGG|Uz
zUV@l?Hv~tDUiAzET!=WYP$BUFoTm(^b*F9@dd}M1Ndi%BO6^DXgNiJ?59WTc+=AY3
zt9rfW^`AvvcKFrKFRWPfA<(P41wGjb0!~-Q-V3oiqIa5cVL0tUBbNfLG6lH_8b{ZN
z^%p(M-~Re(IO41fZ!bca0|ZaS#ut51b7e$>5tV{(o%_>?hqFd@KCP7x*UVlUOApLb
zV{&S3nSxm}Y*bhoS`?H{-?h8UrO*x9Fc3m2T-L&6AH2Gsc(^~|5|jbCb<W~QDM3`c
zZ}mE|74fvDz#<i@3`*^|QprVye{~G<Uf8Z1>t*Hoc+p(wm{imy(0esV1;#m0rMTEd
za-?c?%PTlKkzokWnv42@t#oTcg^9XmGL>l6s9~K93N<&@>rU$qpCYHzV##VopqeI@
zNxj3h7Pguxt&ql~W5Zz{Y$u$~XO?NEkBwF}m2lg1++XvG8&!c8rKvR$B?HD7@N_zr
zCLVE`=5~F2&!Z3Qz4HF!cN!RkB4ni%T3NX~f8^uid+zV<aXxU}S8~l-%?B$~Y(1le
zrX2{m;A3Jwoz)?5IunwvcvO6<Khyc{%-y11+r1V=T6+%?n8M6fccKeCoL{k=9>`g!
zL0Ily^VyrXsu6Hn7+!lrB=F0h{{h#>N51{-Z(!_5bGWMQf3J<^8|s772V|?2wiOzI
zDS8HuS{m2qd=vH@$(+45#OVXdMQ<yGqD%_sIG-k_n0|D1|M&HEHwUJ098eyQFomE(
z!J%l|avZR=(tV)!qSwxV<3jKbe0Kz|%JEvO-g5}i2fZ1c!<lm+*yy}<b!kNJaX5{N
z{$zl9++V;=Xy@(3uT@2$-g%vRiLKk<*6EMY$=k;B(+6S<MDd)aMLSTZtHa=xrg%&-
zkHaLGb@3&4-~}Pu%~I+NAol5n@B))LI24`WgDT*iqYs^ocJC=RLGY!JLS%?$ifmcE
z>a7%}c&~a2n|uTp@Ij}E&U^hE#!zp5wS%oEl3R2&m=g3mN`tJb8bX{A7oay=Z*G)w
zrB{tnj}~99pr8XzBrnfUyEbasXmw+>m0rwwHFVM(hZZRzPE2Xm1XB+BY^LTbt?q_3
z=9Z4B{03t#1YLgsw9ynsv|p%sWi*J+gE<qV3wZhD{nzWn+4Sf5Y>vr+PL|bTt_47&
zH5hp(Z&zBGz(=routADBtPA4g#+4ZE==KXOuO=?m!X*gIKI!#^Inqqr`D0!`sTe3W
z+8iSIW~vl1C{H)>1De51t=u`s!!om+Px^sgGi%N~=fYm~uYUCap*99W61_?G4!I3#
zhh7IxR2;(4+<{v44HzqgX}1uNfu9_&!kOp;Q?&HIp{PxU?rTQZpqbtq&TDF1ZJjM^
zr|l0JY}3P7BYTHZg1f5wkB_T&r7`y2=t--;UNxs6O&U{LTjjcEuG`KfTzPuBs^3|J
zb>FGIfzhb?;W_mVu?to&psIdUP+Jdz7nmo{`Od<hLR3X<(BFfzuBUajZO8kd7c`Ds
z4<^IV)U~p6S)aLFKCnK2ByW$jwu1~C;i|pIJJnxI$tgoY1YgC3j+Lj6-!b{dwyr#0
zF8bb!MlDC@X?aJc$hxkq>&lWAS}8>F&<1<fyo9Igh3mR%*FF>>`tZdUJlwzLe7?h{
znYXWBF@>mOJ{-*ncc+<$DYC?1<K4hkG@n66r}j=-PTarwjCr2;XQS}1{=a|3NQtk0
z`A_-TtIzNqE?Z&UGvB^@Pkwr4auauNf5vvXu&x_p3?{P_HadIX)hq7wP*7gIvG;pd
z8rpz;S39fsN}6P#YU0?bbEkz$t2@0zkia}gf)l3bxnHKAa;yHmd)-CPWTST<ZIm56
z5qE=fC8sKMpEU!*iAI&TJsXO+8$%zuJ$SwI<l;=p9cg`#v+_0e%Olke77TqZPVb*M
zglN#8U_!OfhvujBhVMluH-N-o>j1U+IXilf)xB<mlpa^M4H*+I48~}T5o|i3s|GLn
z=e_aigMDAY2P+8lYu*kO6CjRZnu9Jk!QJ?zFmQ)E!Ty}{Y7KVI(Y(JkB6A-g1$GrY
zC+{fx&bF^a4{|=^q$93_fx2Rv%;#QdSvz}uHh79`^tBS>Y&Jv?7l_kh6HlQwO-U@<
zm3_Nv;Kt~by0UMN1P>``Qmg(f+yt7fcgr2v7(L^R{XUGZp!orTGUTQQ=tCQ*Q7WVC
zNX-N)hg^@|DEnpwmb<;j`Y~$wR4+TDHhsU<{C~^7z+5Cl(BIRB1;}BxaJ4j+!}eM=
zlV>l5(lj`(RNeh=mq)I1Brp+0(Z<6(jQY9T8e$BB6Gb=IO1o}|zml6;4q@C@A1_hI
z`n>;<wZKm+1{_Lfa-#G-FP=8^My6XjU~dXusg2Y9nbVvUFi|Vrtz0`i@0B&{IBBny
zd721&W7;cmj<_kA*S{N_MZKIUMa>@=$9P3YR_nHrR?n+EagGyr)5H=dwHp;xkk+bk
zwmDb!y#dN|#ycI?@0C_n{pm!<QUg)%cx%+!S+|{(JabhZbRVLl;!=sF+6O3vX|~Ox
zG7W03DuP@$-aS1LmPq$bA9OfygCY8T_!Kq6yLNi6?0aF|cJ^GEr@(xgIh_@tG8rD!
zd8a7i1J&%+*SyuNBly}#(?V}TAC8z#x`P`!0a^2v4^Q9o;fLR{T_2IrETp7r7)e(f
zSV+2xh)F-6kP>ooCRl7-KKy{xji<|siwke=Kj-cgkS=W7v$Es8<2)^#!i>l3P1ZW=
zy7Eo^mNFXe-+jmP`yX(0(i}*2QsnY-=6t&2voAj9{(RC0r}vD~iQ<@JBqqgBx27Vr
znls=1@I9e*MjPCJ{)V6Z-p^Q^C+EWN{ox;yOW~`#w@B-B4~)j+^~yIte9JfAe#beT
z?4hkL4iV3kW@<BG+EACoFnCFB3R7_d=$&a+P*su6)Af;%>~k_7aVz@%<>0{UH9dmE
z=c+N>A)0^qclQ+>%qbv;cq{3_>;|;zXvI;R0|zlwLe8pRsnwi+ho9PFV?}ThsnUt_
z#C$r_O6U4~q19?D8@-3Ms}kMQ9~vL$sS=pdOm~i23uD{O9aM3narEJW;-a;amjiKT
zibnPHA0vOU|Nr3jW4eMPDKpsk8WRE@U-PCj5i!-j{j(h1rnwoc0JwF~sQ6Lqd=(QO
zqCbD@v|;oBJ=*UV^{m@Hp@r>W^g=6}7HcM;j?V0IP9a-z9K<GlmoD^D*m5Prq*!*n
zB~)K?6NK)R{h3~iiQ)v2PHj8eej(1fJ{1|X1|CQ2nOZmMu2TV8;Ye(37I;)!2d_K?
z0exm2yZqjuH7yKG_&bh{x$1M>hoD8CNdgB#8@jG@#s`prA%V4I5U>0cvFY!jWjnGY
zj%R|#aJQS~9Jsm9dhhh+$otCW@{FS}N0|Lkw+G$aEl0+K*jhw&DP`XXr%sHXVJ?d3
zy}3}1T!bU0;_qNh;EpI)r|_OoJFaznQDIQ@I-V=F5e3du<ZL6R(L2p~vQx!>X`SaS
z>keznNc7Ch#4;ydz0xA0h)~_2jY5o$A;CIxPa5jA#0gPU40AU{UY!>1=UK&f`_76c
zm2Ekbx6Je9%JcJ;+J<Fr3%OKA-w;=HH{;Y>JJ5Z1&Y8Uwa_i)(*i6NfddoTJ>bJG~
z5IJa63b61ZoTkX#99fd8Sj*6^y)bm5BF;iZgxor1FYMQi^}4g}nLTIX9GDi3fF5*e
zZ%i`qL2zCpNn;G$D0j8$&ssaL?_TQ~r^7VQ&^p)4NA?u?`1C#d<pXtJ>Am7*(7UL$
zZK%rEhd`Vrd`uvY)(fTS!>QK7w(k`Cd)Hh!#l^;>fn`}(uOGD#34tZf^itW@D{Hy(
z^!%QEyYjv^uGeSs^+LHk<2$$za8X?;-g}nQqTN7kgn80~xK>jj4-GBp8Vg)o<?;E6
zzxefE@M=DBdi|O|`05wjEhl_$+)XpfqRET76jF*zA<%|qy$>{f^VQvh;+4|^X~qp@
zwTJFm%SNjMhNtDCR>;<vro^(G2r=L#L4nW%enj$Ga6uoaX%cXq^||0Wbl({c{OB}+
zGL+SEx$HmXR{eYTI^2UHcvAAD>FCm-HBWH`b`YXIfN9h^@Xpxm;~0DD#7tK|x{IIQ
zd_kHP^3#>JS2lOyh!c12K!O+Q-taw=<{2M7F`e<t9oJg8w5D2u-c)<g4Fi0%>geGb
zLeLfAPzR94>zMlhOvI)K*1;rDS5aq(9w1H*S)l%MpIGQPU^OCtZR~u!jtPb5Wu%U&
z+mFJaV=yKQ$r?k&dKfCb^<nD(2yx<kcSjdT2x5MCOn)GRsGn1_reL%~+I-V3YMi>+
ztA`Ktp^eR6Gu<>Fy=?T`VSfCuhyGDx>jg0e#~|2*c~p&Dre?EFf8x+;+RTV()`bjR
z`HoTTIR?%Gn!FQ;2E99qF)uGgkAalbqV8$V)=h5Qh?d98c~3}DKWjk3xLHFFAt)&E
zaX5ue={hJ-{V>;uuAkQRVgW-MRMu_%p(}`{1sB{Vvl?(^fqgMDpYJu-#7MT$LGr@n
zg&1r~Q2!3Bg|(@$r<rJqE`+AV!LlS`3e*nIRdcBIVp5CUb<bRL#(B>)Pa1kt3#AWo
z(aiE9N^}d(ah?)oG;-M~MXRMb2Kq3?eM1j(QQ!6{I+jT*%RU;NLG6Q53hVW%u#zov
zy>4u4fsS?Ks_GXhD3sHy{-Ia9ah;BnoUTVybfV&s;9y=7MbJ*(4k1r7r-9x;QwrS8
zk*_{`#V<a8&3roXR2tW&YmZW^jUQhuL3_?zw~cIB>)A91a_9~?_EKuwnkIe!rztUj
zoHdqLx(X-V(D7NRl|_`J-&$sUeqz5oBCRUBz>7w*mQJleQ+S7)gXTgwPp*x<RGzOJ
zBkTA2{^JLxkaU05yGFuVZbi<Sebceqyk~kT<g!zx^Zw)a{P6B;dfBy#-Zm9Z)vASk
z?UX(9%{Ra0tB=2Ay<Tub@v^lPiY0gD-dSrQdSL|5UK)F8JUu^gd46I-IQzH6A&gcj
zUU+Id-+uQE&mW)Zb6{R(lK0>OuTE!P-`yk5@qD@P>i(XzoatfW_rLfVrxf|-`>%20
zdHd=u`?~Tme`MVXrBo(A5eX!}5d1{zl}w~&B@#9ZUFl=sJJi~7wc{6uM4?a(4Xsm#
zvi@#Q2BP<>a*p8^7UR4=q`aB72jDA6>9nTfS+Tk7VAw&eP$n+}+_1cX2M~wzyFQ4=
z0bUs(!Iap;nEVb}Bl^H;InhES+rMwfUsH%HHO~H>b$<RxT8&2+u0rJCogRE$3l=d5
zkP1bslKbp{(QeZYRCF}WIQr5TuE4<-xV?@wfV?;h;Fu`95dRqc_I@4DFlCG=BrB!}
z-l-VZ1;hoM^Co08=SR(!Y2b8%m_f@WbV)MAl3cZII>d-dQMA8*K$otD1TRSUH;u&s
z1<<*Q5-Am)CWPQ9`ZwdcPAA8$%kIWj(d!_D2?<FN#x|_cPTr=x-E5ua?i!*K03RdM
zbTV6~x=b{8LFh3Monvz=`?}GOqRlB9KTZ=Nsm-yW4Xtw;8z1azIB|5*bHazfs1Sn{
zPPO8^UXLa3tk)~c6gV%cFLA>ic8!}GZ3Iv35UYk@jnPRd64T82l(d)`>Ly5DSfb{D
zgy8Xi7n%Vvo5JxhE43*u(Rn<YQrL0iy+AHDtI+V3HCL{!YE|VN1Wzyih6V1WgR9zY
zI!{Qt?~lpDrWchk2+<SUQ1oN)nns9CB}8L|_fqsh+Xk&PTG2qHx@GcJ72$5^s850_
zjAIN0^uowR`2o{B;e%Gnb5h&n=`=GX#c+b|{9KHb(WpJEBf&eWp+RfYo$>v7=4W4g
z#^;~EL7Zp5tjM}SRpDVDsu)#EkUq$c)HOGRC3;SC)Re!VYF97dgJPSL6$4s)39SXY
zfOW@j$O^_N<m(gLbOt%GmrdhwMVO|E*aDXox)jDxu}l<COpzFZzQ23n`MUG;d|~W`
z?fFW{g;UV1?=}?j@_c<}PKoF16PI<PO^IdS`S|>SbUNX4r)-;!8L!GBXhma^htqi3
z(K;n(e)HSE<kheK9m}f|pS}H@l7z4P=ZH9-O6L0yAM6@XJ6_F&ckka*){WP9_nNpC
zJmMXD&D1jZ=9_O>AD<MHD?zL8sA}A{L>?YqG2h+u%11uCd&PC%d0el&et64inw3HD
z{DJc_QNlu9D_7abt<Z9Z9!PE?gvgc^L%OZ|EolzdDQ%Fm5K?E9!Q2CW(or#<TgcAu
zI)N|*wa)tJZ`L>6ZK};}484&v?WSTsz&VJGoNaA5hFX`Mi_CZT_^8RIy%p+#<QBCW
z=31E~+P+@b612f-UWm)WTCKD;!KMoyFNxUI3Mz0bxOHl%an_26?F5(XI8%|L>*#$W
zcHnmvafzs(+&Q(;>4Uf1-+hdrUm!!r>-gVZjtPR<7UfW@YY`=Gps`(IcH(jCvJWe!
zGroWHksvePN5pA_ybCXgV@B7#bS?B&_2;scX(@7ZWeLJ?&3<}?#vR%Pz=}Q3a9R*_
z5k@VB09Tph7}j{9D_!S=-gS+G*-hyh+NQZ3)w+Kn#F=<{Af}T|Xf>)?*Fgqu(h?Kb
zl^p22rXfz#i8RePl$jEoSe{J8OT4XoMjzyDSBFFJ#3|uo0_QdNLWG8DtCF2<!F#o?
zHY>(Z8?88hyOuR~Lf@*<wU{`XP8lk4J#4ECEvCB{LKwK(=_TXp#uNiFM3T2)K7{CC
zPKi0G)zOW=gJMAKs!H5)PUyATD0qO2cpuaXCYU`6qSM{<-WpAWuuN9}4o2<d>q_6V
zMM*YFscc(j+mqsEOV$0ojoOatF$Tvp1+9|0y9r$OvcHVkFlIE4S}NPqm3-X^L&ufV
zJQ3%_c{ved)QH-=YgdatfW3FjDY3*z_gw>cY!%W?OyYf{2IVNwj@_Yy)^8tZiaIt<
zX;!v2suGPs^UB*{Rq=uw?#Aj4;+VWA1<&jA!aPT#wSiAT+4x>BB(>t`WJ7moI!e@k
ze+1tYa(Uw7J*Ax4YoU~ElZ8pu@HnphS#R>gW9*zdA6jQfC2!Wrx58dFoEuExg?+wl
zJJ+Z8<Pce}SFPr)kX-Xd>nb9erio=<DC<Q@lupH$O{}8U<r~lM-}9?~|0kr_xO?-O
zfBc{PZ+ShPNkdp$=G*VTrK~GYpMB1&I|aXd_uaQ#K7JqqukIeWJDqiWS&HJ4TSa;&
zShqDgI2=>ZjD-7#JI?p7_}Rl7rZGq{saBv5O6!Jwb%;dfQ{w)i(5jI96Kmcm*&s<S
zFxuex<1=k<NEkN)8se!9N*=V4sbi3eQ@UY-7<Cj*|J}A0dT*3G`*p<j9><<G6`x{w
zt$CDV&sr({uw@Qi>74Hf(?TwVvR>HMXZDgoG^V$jW<$^svUh_J1NRSac({K}^?_^N
z8Lh$?>UVbu-&GhZbPG?z_R2>Rs!kfb*!|nJQ^qM`x*Ne%yL?9o6U*sdLzbEWWyI?U
zTV9wV$LnvB2M;&G4V*NN5F%`PVst`{S;XC#!oV>SM@P*#XRxNI7+W7T4mw!exltOG
zi7$*El-f5Mt-M(Ms)@dpNE;FF!FifEP3si!K{X1;eqUS#=WbT&I=W6Thg!N7fTRaz
z8^`xOzyj!6>(o+d10+oN>CC*`;k{<4$Y?rQH^F9a)xc2P4&@Y*Qfg;iW6EfZp;76>
zf%B2!JT4^SfH<j*aEyUZi4c;0CJ0!ZybO&#j#dmDamMcWT)abWG1oRRZdKc}Zmtp^
zZafA3%(Q`1nA33$>igkD82gp|@q6yxe2JTr7J)<8YQYOBI4u;sqx!$TYWiP#{WR!M
z*F?Ty-&72NFb36uD~)l@8Ya>jr=&u@`(@!;E6qVOUidK{9EM#jMQ?CMqvp!<Wn&6L
z?Uilc=~e$yHw5Pf%j`Ix7S5+c&XuRDD&KR>T(^}{ie6YQ5`#iAdLL}t&b}5+xJr?`
z)52+9Sf*J6d0J&Jja+p9KTS!6gL~#_TQ!laI>;(0JVJkxk97Ji7yU>Ub3whU9nq0Z
z3=GkoU9J_JcCjs2YA*_1vC+Tov6R8!9mM20-A$bDPV8tgl9EC>d{EfOLA)9!+F_{E
z!<mQGi!hohTOaFt{N<`Qk@w7#3O%RkgpUaU^7B=h2zplDF5=eofi$H=lR=&WjEX}e
zOt)OgYh#->rdb~xd(H@I#d8kr-pp5R6j*)lf@hv4syNPfXT&lZMjx>4?C-xL_R7=S
z@A%=>7tFy^FIW6tc=!B~ykGb*pXjaeS6}~z_doo=gG=f>$Q3tWo@TvSZIbZx{KV7s
znZ0J(IBOT+9H(XBi?^TiW<KM9)5AUPUE=P|E24Owe{<z}x#E0co+qZ|%;#@C(-e63
z@x5kX<iZ$`*PYc?RLdz5qG~d{2q9{i%)rq~r;bh)!H>u&jX|U4-*M=WI19KjQC(LN
z?vc)>x~`W_T=A<~ZTGGj4XrA*Fs3u-H=h%i6MYDMTN&F%tGb#BUXz}jkqz}AR946R
z{Ttr?-p>j1!oFo{uR48`Vg7rk#ef4w$KT?JKVbm0W-C2W9IeqtWvKW~_qW5w&xzCN
zo_StwYfW)#H$9#eAq1rYy8q?Lz`;16A1jli6A>#s?0<6<2bzYc15BL@f*nvIApW-J
z4#BA)?pB0gU&|>s!&VJrSb;OdqLS6=<<hK$fVsUkv!J&z@Nq)Ss@tkIc4BKE4eA-H
zk&~inmhL4+z8k$!cOsUoC|XSV`AnFX7tFJ1J*4Y&;Ygj-m;jw%yO?lMbx$rtq&JOO
zZoB3nG)qei`g~AV#B8hKMmTaT2J}%GB}1#YF$kiR$LRg7U>q)}z*L6>I@%Xc?}646
zN-iC@jGLF_+G9O%_*PWEc5o^X=q0m0e#`T_SET3&=S8nA5u@N@0!Y2~f6VKraR#a?
z9LG_e*e{P`s~x1O?W8omcvyISUPx);xil`VQTiZw^UMaP0yQ6Wd<4ii4y&E_A5<XM
zY9$6seRBb<@|h>k!)c~YgVNN??!|6kG5bYu%p>7qV0VF1I@dk3*UB;?jHWkmtvK(P
zr<o}R_7un^6H}n|&huqs+Z!Pc$bz&%%bJwvy{E--!}NMlHSytR)`Lxx{(R61T193i
zq(M6t#vqqU2-Pk`Q~eGoV{qpc7!<q`uACbhm72UNoe!Igi4mOy#0N_Y)6o<_&Y9<@
z3*IF{niwK1rxVV5mgSx$&B|LppLk4qgDVYEKRfA~Ns*=mGKi<dDLNj1c&`(cT9u`r
zE3G}4Hl)&OBa1pe?wd}JB#1dZwwIBxUHWw#iInIr*D0oSl*)d+@a;E$#rgg-rZiEm
zJ7vqf`|fK-SvjBXpbx(L`fL8`>u>n^7eB+DCQc4wjNG3l?(b&);`7hQLqD&&?>ucA
zZ)>Aer@P5q_~LxeoDy}<<aK3nf#)ASFi(k*Gaui*r}e>?Uwy@^H*XmxMC_yS`1nYx
zL&dMdvF#c=>QLZLX^LI;LwC-{Ub7cy+n|hUzV=AZKd!p}`}!JAjLc_(zOHGb$7p<5
zhc-r|k7__JsKx-P-Ga^1%&RZIVmdu2W_G=3f@R*2Zkm88BE90e+W2UN7#z###KXfI
zqEC#W5BO4xMZ79W(m5eITX&V+#`D&(jZUcr=*puIolHu&>C81m$O;?h1Iuz^o)<7O
zX*Z!=@aoLbBu_2qbtG(<+T(p2<LL7_fIJLCf)9O$3{*dLxD@o&l>j;pO@gj1ML;<4
z!9x}2-of$rH|zGC3UWv9H$i1LXBDE5ohE}YO^78zmRi_L=6PFL%TBF@R5yZ;3L!h_
zkxuX~=yPSL<AA14kLtjqP&T@1ECj#_!+V$lJ|se1wAm^8HEZ4(t?2Y)=x6C2)S=`-
zoI;aAOggUb4O%vFVo||(YxMOIQir(cd*r<a^mT0zr7KkxDpHjPVq9&!3q*rAtv;wD
zW!RJ;1O@5YG$?paiV=(Cp>>P7_J<qAG$8HMGEYBOx~{~^?%3D&<ja{kex{-X7jSqR
z<998v$B%0ye&U+I##2Lg?T4bR8B2V$roQUdH6C-CbbmfzFP-akMO<cDX2KLTjW6%?
zrehXw)oE{?wQjmE8W3I3;383RopYMlq&K13YIMANBxxz1K%A&A)pl--L93lT7X}Ki
z*=r>UOfhI~d5`Q<;(5KQh;Q9lHyvxwZZbuu;KES;zY)BKU9`8V)Cv<NN9l;v;t|v%
z?LFSot$eA_#_fTkIM2d$%UrU0Q-k*}Br_ECt330d_8*jcAat$P5#wb$U4L9RdW@{s
zjeXw;LA4D&PB6yHbVh^_A}K{u@;V;004N`Vvh?ktvdoFeL!z=jdOkc}^ugV*DL`ed
zjeXtNw~anJ**FcgDot|VS4!D$SqKBTY*$LDJYF8PDDx`Z@y@a;a^?Al@Azi^Gm;D#
zo(~_s<Kqus6J!uaXVgx6exhDJ@~e*@`1tNye*b^{pHXD6Od7=WhhKb!fBV4a1tvVw
z2kXA$9Xzi)zxnQ4u9s&f_nO=)+qNoNdEIDxylEw9aCds(%U5sN_bc1wBQmV;dVVHG
zuVX<Y%C*wy^#~o&&eEGoU7aan!Ydt@yH=t8ZWjaH9!x`qMwSn=JRJvMH?Q{?24!h#
ze;XRb+B^7Z=HcznINiSiJb7Cw*NeJON@k!`I);sV`_KczCr+m`r}G0LOw`iYOJUu2
z_I;-rLijjqo$X#jRCkFeK!SbS7_Cx=0+Jj$L1O5d4i1#DPA2B%#FQpLZMwbd<ZN$>
z&34O)9uqP-9Y2e8{N1t(ZZG@n?KD4Zi=%4**66BOzjgK2J|0t%m!m{Ohm_gkZ)7~3
z0qCKrtHrK|)Zo_9mz*i(NbN(P1wIgBx`~;aF<h?ureQg~(A%!h0&R^6-ZRIEl#-&S
z>!2G?pWZ1>*R8|pip3py4WbRCH}8Dwy7ugJeD69D7-j`T3#IYtRK=1k`xr>8Fe=h)
zRcP<46QP$(Z;xQ1YpP$hi3mgW3$-Dm5??MD<d#tAjZLF!4*B%+`{a{Q;-tFjhKUMm
zt&j{w?t)$mcHSQFr9+D{Y-K*|eQ-2lh{UK-%3>>`(dFj)_>aAQQZaD5G)>`n_+MK`
zKggkwmB>)rT&t`l^YQu0`^O9WWmkb*sf0;KE~Dz0Mg73q5l?C2G)IOE_Py#JJqE=<
zwnA%_kQ`yt5%!Ql@{x0zIL8Iheb?eWDL9HncBUAQXU}bfGM^IjVk2@N7_IUA^u%RX
zyV;g2d#&8xEqw9%K>;$QQ?5JBx7)7IMIH5opkh1kgldS{F|@<Qi6UuLXn;p6;$l!C
z?^WA?2-{wHzV7U$8Y)#GEg?lhL3IyS1$qg(z8ZSKhos}q;B=%_b7j45WLLOeH$q69
zPczdzE2_C`QTFuwks~9d_oB9%p`tAnYbpptuK`WYN|UEokBge1SaJogRex=I@$RKi
zwv9n&&l)4@ZsXsq-C!?xn-uK1@bvV|b-fUS<2*;g6io=J50-L$<ioeWV$>B0k>}@+
zl>G@GC(3@sb-~k_94hQwpZMYXueml|Asw-~DMrpQaGDZ9U|%zNE4+Jr<WK+opOLyS
z#Yo8;(0TX6_r%({IJjP)G}&^B%yC9gNX0c*)|?fW-5NoJ)4VV*6CXc5D)QPG6#$_r
zYu#&dj{3Q)qwT7uAAQh&$JYdh=ubz)X2Z3c=MYJm6kgNg3<W&YyG2MR(sCxwC)KKy
zOv@X2y^`0f9^_sre;O=k$$}wqe*K2|bcaV-^m)H>&Dz1<VsJ$d$rJ*?3CTI;Ff*mu
z)|WfIIBEy!onDGk?^1%^?9U%NUx;beC}VMq+LUV7$}P{nw#p{3OcP^Fwue@s^fA&u
z-nh{%%@jN6KIs+pkN{|Qh**Tq5&olJ<YoWNfjU+mOm9UIxVFdtYm<aS+-<*xDf{L4
zSzWvl{c_3d`_7iLPIOHq*LuETof!_91L7l{BWYGwMzvNICyx%jk&-n&z0;3_+@3xI
zbUib&?)<v~d4di&RBM9RR7LP1Y6Lc@{;Xx2+p8ABxfj?|aE{Untvy4X7|yg;h%r~}
z&x~=iogQbQ=D$Q$@S9_#cReF&AI4nhw_i8g=>aF|LvvSRu+Q8QHmw+Of;G)@%g#_>
zl4eBA^Yn>`@nAye`u)eeeo`^etqRgN*s!$T(~C0jTh}Bk&860{$X+sAscd_p?pehx
ztt+|73GK+N9-Z0-eW;RiT9U#qVx`&mtG2@4)K3}qii8ax0<90?<khELb&M69V~Q4r
z)UkjQT_q$h%*pd`o_Te@5TaI^ANMPl^~(FlXD*v2tnEc@f3NS(yncPB2;HZZID{(M
zINX}8(0bQ(hGnt$(aCw&(Gu##U2RR#Y!B4?V6Tm>=;R?<cd@P;>!vZplhN8-Qfy}u
zhl{XG3(*RJTnaE$ZR-7PH`s=19_El}6k6@f%gj7Y%*#ZsipDH?<#KV1QHU|9_{OMO
zeY`xNWhi-zVfanws0eE=T(+H>JM%f>W1!j3eBC!nsiZmS1jab+rDRIpbTTm1?$ty;
zt8Kj~Si?aI4j%=zmJ7pmtB?!(<p=5*T-HsI(_?V$N`YL<M(YagP*rWCVTyD&`1e2j
zmg{f7q5t9+_~1E0J@#!UU#_grPdsfK=Q&Z=mFv?-zW%FUA*FCS&4d7Pnuv>PMz*@K
zwaVL{eMM*R-Mf#7z?2fB>x!cG#`AUMYSWMqR8V!ZJq=yExi0i<p6b$}mY;H~{yluj
z%LC6uQHN)K9F;*IFB61rJa=*Upu6M{XH9`D&+M0#eOonRR=SEbbq211GEMUYmb=$H
zeDRgqg142tUsS_Vbdt#c=}<*!qOMj>E|I3Bj)Sx?dZyP*5cg8E*kMZd^iSLm-l?E2
zm{v*0*ov5zR&B%{ZS63~`AR9XW`=n8*VqKNe%_`g%0+3nqUq2FfvqL1RN&Av*sfS^
zpMZXiK6GNzdn4CE^&1B>Uo{;bMCi59O3`Unt12E=H69|zT2h;tecv@ytt;54kACA4
z^j>w`ExPV02ZFsBsyFNR(g)<*=DaE~QW^(O0tslp9nBdsZXLO^LNFXoouR+e*?K^H
zpp<I51?Bi3VmyC9pUgcVHd=3GqdN_88beg*C|W2T>zdouZN;Pyfyt5<HF!&>7g`MH
z8Fowoj!0}xn$2234a~}Tum<#NkNz^fP&Pu)Yok^x7BmwfhDa3EksPSv7i{w1{`yJ9
zfKTxT5L%XrDR~%bh1k0y7(qCNz{$7+qjlDnmGEUy3bCC=FG_+sMufEvu500G&Mfzd
zlZ%8UYRJS;yIboW51NP2j?s<U2y<<)t#lv-&#U`02MV;c&c1Cov4@C?V@k<X23s3^
z|M3GqynEz%E1Xl{JZtJzNRH`lV!BJ<5+PT@80;;phQJ~{#X(38rR9jK!+XIG0|AZp
za>ESu&X%SLiM4h{96FloP;ybH1OT03(Z)C-xya-klaHGJUI#5#8ajRz?{8!3V&uhE
zH8jd&uoRQ1IvF%KEXFgEcVRSD)3>SxOC1^>(*{PzQsuRXqc%$^Yn834I(W8%FfNh4
z?}}v}db2IFvJA94D(q$Fx~+JxJOddAq3dv04Df1;>`pr;CtfEpP5~o1SL+swVu$NS
zEuE*Q3wh5doXs&uk8i@a-+#{-jz9U;pK<;6w_IPpVqLF1JwEEq*OVN2jx*PFXS+PB
z#c|KnJtIwd`E!cgzkb8puYRBVhX=O2GZ5~V6aMor`TmP<kkK_Gp;l_nj5b(X!Bs69
zv#W$Gl+Z~&7>8dwfz}7PEA;B{rT@;pM5D(U%>;hh!D`H{4PmV#l_9nl*L7N&g3<<C
zXKXt;@9g`gD?^4_QB}65`{AKQs|3g0o43T}tZU5iOy5=w+G*WdNUx9Hrc=DqE3LqM
zcgOu%6EXMI1cN4+<Iv#Q16PK+Mf%twM4X#VnP16e)0HyKbUoU=I?-|RV36~sX|KUj
z47;L?`|UpE)<u6h^Bop}d_*h{$E{ER&4{_-Y>xwS46{`$Bfr<mVTDB09pFUAgU9$<
zNB%bSwosiz89%zM5o)<H|3^P!os|dBnu)8?2zZxwqHBy=*fxa`xzX`BDue6*(aC`p
zCyM;lT2Bkem?F*^qacc?4%XodC>T$4a^pk1S+~1Ed_)fG>aj{Vu2XfI9qZ=Nn(7(;
z<`iHIaJ&?4;vk^w%;-NE)gK@vZj&yxh?<tbnygwXWv@zdbP+!^4H&T%jzgdA!IMT}
zP*roc?et<ShH>EIb=oqliEi!3Jv)9<F`z;<MGlMCrI`|`u#4ZYo6a%Mi4X((&_nvz
zM>``vdGE<Bn<w-pjUbo8hsP&e6qeE%Ueo=O*IjiTS{bFNqi|Kt>%sT&dchS(9gT<c
z!k1DBX(Dx}Ym8DURRd+Vy7O^e@z*PoB9H6Fwig;gjE;GUq$y~}?ZGF91P9WYPmw$t
zF-LkWw+~%xH>DqdO#yP@G-<ZNpxs7<t!n5}De59flP3hHyC75DdgLZZ)X|VdlMaIu
z)n<7M$|NYgDZW|sW^9($<;EbmK@fc(S}Ek5@gY!~lDy_Q*glL~zW2Ib5EKQimLg36
z3rW+#K847Dk{mTWM3d7F4alJu@BOHDl^HQgCGV9zHyj<`{Y^>h?e5IAvDHEhN7rIE
zvWZZK=;~qHE4{*?s=cc<w*5-21)maIExiBmj`#1r=J~v^K3>V+{{XqMKR@yD<F`EA
zohc@~`S{^I;v5eTuQ{LZdH4Q1zW?w&Ij=+)U^II3%*#Em??2<q&%UJ6xm+&Dp2_>B
zBSRTPA9yYoEf&V1p4X~x6i45>@;K5k^|N-yt!p?}DebnN`Q3fVt*h>EQ=~IgL!f)!
zj(X2uR*~L$mL;-xPg^(3rq?N$p!bB@#|~9_H_%UEIlm?(jWMpdvRhhdu{6)_v{P>O
zo?{2)l1|*8ADE^i+)KMfH@NMCZR@A^M5~C`#wUo=uG1;ps@9@{FfT7Ip`c<|A5;q^
zh&D@ROLk~iuOvEEsUNoPzo{HKQuFk%)z?dO?Wh0F{$to@@yT`8tr&1#JAES;*i`Uk
zqzy+0thx;CWK}D`-#Bju*3`6a>mkLak711tMW-vno?liJ(hVi9l_E}tb>O-i58mOT
z;G-k?K$3xHAT|LM({y=f&!8~r&(amZb`xk?Atkr!!MbVt{`=?(w;RfNQDOG+$sf#s
z!=-V&W1^Gu15D*TdZMTm6)*slV|1@m7kgiGDO}c-khCP4y*6%owwOS597&18p2w9v
z=UZVs?ETdAQ~ka=y2kl2ub;Li=nA_S19Gm21Q=o?9xVzqK~L~O=mKkQT=sk`5>&M2
zEltePnxl_)8;`+Y-v;k~cx2uxVTpv~@N>L<{rkJb>sMz&@H{_VsJTGs$nNmXF)fj4
zR#IW_6+t5}hq)o9IT7Z_HU{sXSN8Tqih*-XoL^5WXmna#g&}OEv0GTmVx#Wf8*MMh
zR&c@LQcyIeQvqPfnQbfNQaR5sMf1>lM;S+=l?o?A^!T7rp*P}}u@&6t3JEdKeJO=L
zbhjQt!WDHNv}TCW(M+7D_Qf$!QdLy?SYxp7J29&Mp!cEcfT8JmB3*agmJ(S@)s7ui
z5T|8QaZ%7!(UDa!r9|sPxe&eI(zTAASgn;@GTU}#-><COGdXX!(2_O^+deEgtb^z@
zqF9k*pirv>tg{&nx&oW_TnJtTjv)l{wo_VTuR5A_+e*K@=Rf@E{|DMeyKemYPyQ|Y
zb!Wdk^Zw)4JlvnjCDTgg{o@C&weZ!OpYhq-w_LUhab7@#eci}&Ae<MbH*b0S)h~E;
ze&D)Z0NDEg?|9lazIyYTd70tkM@p$LpdELV@Sa|kD<BQ3i73U$Zn@~>!~IBm(BIdW
zj@H4BQFW=<VdM^?9{O`@9IR3CTnFd%8hCkbt5@EMeNnOC9^=6`-QoK<YZUqZ%)VX8
zPmfS`g}#VR5}Jq>215_u^XbIf*Pk)RSp{Xsq-Ge^PSjebn~es;>3+RdFy_FvUWhp;
z`L0)nb2q|Y^o|sq70VVo9wNaK5nI*o-FZyxCw3p%oA}9U;7Ik;!^S2E<6sadSnXz0
z9R2onj*Nq0g~3bGo^`vQob0_<?f9*O)i2c<&eohw4<IosXBeiNu+ck1#odSs#6gww
zYQ^lgU3+gYx`<&@G3l7F^QD;>uIsZWIva_D8=8&dZS3#tS>@tk1+=}nu}wtGA)-Hx
z)kmtB)QEPFp_u7`ia%9|?r<(xQ+O0h`W`yx=?tA##AxAh0q0dP-Fv_N{I2UCY`$BA
zq)??|_j;W@cq?keY2zU_k+A>PsZFDfjrF>cH%*>x#o3SCZ%yit1(yA^VxZdoXADZw
zZ^k=^n3B|E32Uc?d6^mFcwSdNK3ynwXgTjyNuPq^6KxNr9e%g5zGebkpI1_E+=mmR
z>xUD9<Frh?ez*hRdRb|$D;Ow6=F?1?Jr8#$qlD>2)T(BA_W@~&EO#fm6V|mLmH~a9
z5>7O3wzjU(k98om({>a2*?2rmo(rAa8i{~(O-^hLdaJDK&g0WF(gsUP+%KA$(4D9b
zK}$Vu$$(BfoN8Iw%_c71JL0671AR1V)x@jbD=ro)O;N_q6O1Pij5~k>i4GqlE(Q&6
z5gmcYoLR>ClZT1zJWY*V_pG3%btkVo4{z_edvnL>yeK0<u=heACexHiDR3<o(Rql7
zs#dJE8QQcKo}V5q48;+GV@i(REdWb~IxvF1A014O(rxl0w+Dz9r^w!>V5Dh|obMM>
z2t1(&a;K9f&oo8)b>lDp-5-;$&j8^-9=}{3ANimE)&Io0@3<(8dZo21`zCCU&wTg(
zdsYYk_z(UuukP=d{?UKJyZ7Jo=YRZfxjwJ7R(SQr7yRKr`p5j@7r!9jSo2D4h3ECk
zWnI~8LqZ~jiTm@NCGS}<j2Fc=_l{IJ4UF<O)OKxfFR_)D@Vi|M=(AMUitRwJnOYjB
z`jg?r_OTxCA~AG)%~Z=62wrHxF@`s6x6<pzu~Y3`HC<6{wWW2QFHe+Q+4l==yAU}>
zwptihn$hrtyVIG^Kl_YjK9Sp?@B52?{phlXUXMB$wWLuSK_64nF?9VW;%+(gS^yp#
z{AgI?K$?_LscxV~w;ebfs|ZX}qFRejYr@ewU7Jxo0POu7cCU0dJ}Cx<U4L@J2U|(;
zQVfhw_yx8mcoDa~sNRQ&hM}RkAcv;GIfrWsN)s0dso_W*>VKU)=ySf0ffqZNebCRp
zTRi=6_>hcoqJ@OPyEWsXYmktac(Wu6x>`C4Hq~wEx<yfc)d~S_QOQnjZ-Uugt@F0E
zRcmVP9S#eoLnk(3Z#w$S90LOw!Eobd8x2uy--tRH#lF8W+>Pc7SdyVx_MJ_DkI6tc
z2ZX)H;#4~YihXarkZU$FXTKGU$7iTN+}}YlP;*8?FrtzIZFH1#+|z?oioChI<9s^f
zLeQ~1dK2VY=|x%mq=3W(r6Cy;rWy1>bpL*UbjGlbu?_R&M&@bK0>K5mcurH|_5GRi
z>wAPj^hwtfwNNfQ*LAzu+2RxlQ4a`0;E0{Ay_2FdF`1HrWoQ7<<UFJ5#Trt;r%3La
zx)^%Y$V;d2j#>xTOW|_KoSb8hiPJnYr$n}KSToLg>rI7Z-YHtxIab;1kk*G}j9#Hd
z-m~Y-x@B;K#2}`iqbz5gdyKk18J!pd^PKekn39eghu*+@&bJ%fMV;38=$V!onvnNQ
zE`{goRg0~MGoQVE&G~$0&4qPaIk@d&D})#w4fPx6eTOdORWlM|3Z%*W+r6@_&%~JU
zGB__2*Ih5*;lymAcH>1)m?HBsk*2_Wn#>EW_|k3{`!q%7DKRfIF*vSQ?HXeYO7Ro5
zwXK&6zy1$@#=Gy{@%;Xg(mQeTOaVkxV{y5@=ezHI!#*o5GI-Atgx3$RNb}6|c+Yg6
zc=g2>{QR@uV<9ly;OY61%kv}iG;z1gx7AjdCgyqJd^+>P4^OyKk){t{=RHJsvkM6*
z2uhBJlC2E>=6v{f>@`fBrq3%wZMSZnPPoCR4tv`n8gtG;?I_}Y+|mPE@8qIdn>K7%
z>x7g}xHwsK_(so#T06b&NH4d2?Xb`_-K?r(<o<NVd(XCQFdDLNs&nec<WeWr&DNT3
zsP*$WtiUW^$AOF>O^X4*#?r^CAZj6yCW98~SXYk07@u}_`hCiw2e1`DANn(UZ&TO9
z1%NQRqxpkcc;Kvc@2}_3K&vorYrI2!J~-A4w`sE_w`vF%B1R(B6$|JY(EDI=oiEP`
z$qFa?t`8oa!?mGgLad9w74?dURu+IBU*3Df2TdeBiZGi51xpnjH8VoM*;EB5(A00B
z(>E=SoUJ*eH#~YD`IJb*Q|oS^9&ClJ2xMI|XmLTbrlbwCh7T~^k}?opP~ontxcazn
z57^s4=+5bcWT?X+rKANAYP)Y}w7CzS6oZ~az39ZtLiM2C8Vb<oM{b2uvsx`}1*?;a
z<FoG^85aMs*H2y(cwR<Nqbp+?+iTCk^Xhcs?ld!{iA<-9(@uPuC>I?UQwpQ_7lk?p
z#7HZjK2X|VKgKqysy+lT9idr=AvmTuakrefJDr(hLPYE#(P%ldzQ6GPab?f-rudzb
z9?-*Sr5vKFGsS7(gg9Z_HcBr{LG2>($U_$e#Oy{bJQj%Jj)+sC=0@H+r`Z#H(CXiL
z^}lz0fQg|wTN_N1C;C80f$KHg9vC9RX`V<a;$2|Rg}oN$Ig;!b1A^I2LbNKo>76`H
zf#CJ$?V<{6KRlzqRQb_+mSy4Y;ei+@YR;r7;e%(*nGigu^TKI4=^hm^M66w`_VcF{
zNXNsvHmVd_*OklpoVb6Oi76PewK9fP?=-}<8^%@r=R?;o(Xsf6SFi7Q{rZ6vJ*`%1
zSJ+GPUMDGC3xcB>kCR%7N~<aywQER-lx?GJ>Tz~xFp7@@=Q8_#<$3$S_uqZP?j!FW
zzvcJ8_>x*T#0~a+BL(69yx^VfY)9w1UisnU_taW=eR@r@NY)xXr_aBn_&`e$Yz?Hq
zDqVi`uGdiTfzq>{8`=_rjZE8*1c&^6eI1?uFm6C=R^$NC(JXu8wacLhSCOG(7`j&j
zWjX_D#}Aar(3?4mPMjXz=)?E=k+!W|wkPsl@D8Tv=ppGY-7CZ24YiX^DVguP^0@Cv
z&9uuCV=s6g^m-DZnGZ;FO<yKza-5v~80(K4Me(-7M|Z}~dFveGZ4!5^A%-}_55_P!
zldV&n?T!b7(JYaXp%KFUKw}U49YUw!hG$p@tIy3t3t)S1F}t7WpHX3)6-eXd`_@tR
z%dTB@FZMGIIn*FHaU_9P`xF2AfBe_{+1KBYPItWd`Oj%@ULi>>bNZR8`dtZw`n<of
z_)S=;YiXO33|$RP-rp>QUOFLYqv^(gpn4w}{Z?FI?f`E^kc`0(-~Pb2zxswLdj8=*
z`6bKUOtaa8aJXGmOc_LVB0cCcrgjYQWOhxboqiuf*Q`!(()HP3g@zB|rfu^A&Lv9O
zReLAa@HziUFqW<(^f~TqfS?n3vB`}$nrADGvTX`=YSq4;yZs!&d5ZsWild(vRAh&M
z^aCrZq(7{}yIW>n-<?U*#1Pm@CHF>3f%7tP*)lCx%9iPNQH@-nXvcnJdmmzrW0WS&
zkxQXjZhfwWr+uT>&TyWX6e1H$d?P+kYNl+NvQ|EPc;fxzmAzDwcPw+_{_ezbI^k0w
zw@OES-eI1&zq{Aanv{6DUUV#&bkq*^0jve7+BG__g^as_v+Q=P4MsBghvTrKh|`cI
zanM)o4r7Q+DcGJ#ZqtK^kdoIBBCl#(37!z6h7(1vAdX=bwS7_yU5$j`I4=|PG*c`%
zWAqA#*wcGX%gi!OOhIj()#j68O!VY~lH6?CbUf(X&<R8z2U1h%klINNO3CbdMw~D&
zk>!*~QzT6hpFG{E1|qjYHRk*fYqdnzv3M2COiN;(RTEH0xg|CFAr!A1g!#O1IxURR
zh%u0+$n*26qRZNNzV3vnlYBIam_8PJJ5a3ykxC<Ty*%+;Du3};|CTQwUeSBw`yak0
zZ_k{j$b%2G>l5ER{+8}O=iSHmd|03P`OkjAKls`2^XuOToX>pu4}Qr%{KY@!@y(a~
z^FR6{+`s?ziyOl3o0ML5jQwWA)aziFBD?>NuL+LtzT4r_>>3fEVgW4Tjoq$EjovW0
z!3~DzwlWgK<WjW5)vOe+K|m&O9tforYAyQp1`sC$0yzkgfM-VNP786h8&=$=WI9)O
zq;_Z(-!`2XI?q6EVrp@n!_ap?UOq(%dz)@J`>YiK>~DDC<30FP@OP{VZaM)JzI5E%
zt?8cGgF%+8*m${h^gXwU2FF@JUhc00Sv;(O0sS7kz~LP}Aai_j?H$e8VNpE3ze8ak
zK7Qcw+rQw$yLbHS|L_~Ot?-M#_$z+?PyT?<|L~W@Infvdd0{f>-*Ig4wHVP<Pg|2A
z1QAluW@Ac`(=zEZD+Ya!0lEsDd$WVu#`apT3w5v5b>%mI`se)XKl>H0PYW?P{?Y&H
zA0i<_FDe`qJJ@aH9?%9=oT$U#_{2Y%T3}>ekxoj^nlO)Z&9oo-{Tv0tr|TQ<#T+!9
zS~IQ9h(w$mkR+XAjsEgpD>hm1&8tgk&q!-GnyJW9U}~}PV_rXvK*%ssl=`?2Vq6~$
z?*r#Fv79s=t~4zM*4B^|Np@Hb2BWBF+Brd*jWxU@ME41nwu2i4*9i>rRS%ZC^Q=_8
zrq;K1r2cgsPrDu&MWal8(|F{4D_qvhmMd=Zgvk@9K$;UidBO;kR!H;2-Q69pU%kSK
zb_wT)JG=zmeD)dV`9v6jS~EQtMj!g2Cr?96AQTw2k*n^>Yp?8kCzsAvJ5?MmM)lNt
zK}0X))*8Xhwg(@Seo*>6!$F)AE(N4jP1tk7oHXP|A1E&<)B~5?Y6ttCrr^0aO>WaX
z6YB&-A8aA&x}?@}tB@UoC2Gpro-^yd(imL!mDbyBg5tbD5V{X$JMyag=z2Bkn!_`n
zBB%SAd07yzqh8ewL>1A9{_kP?(LqPNb{t(ARVv_VqEXuFIB*Ovnl*8M=I*@Ed*#Es
z56YBq3SAnutJPvlX^>d>Ox`nF-YLD_Gz`lUc-&Xk_uujU<6qEP=P!T#CzQMsmop=H
zq;7okmw(3k{vDU=Bj-8t)fYeKemV2`XJ7K_^&8&YKM?vL3^*;bR_odx;DVw{eGt-=
zI4_B7-np*IbyxhL!EogNFFx4co0lq`gZ8B3(}P9}={n3i-G+7F+7Vgfh<BdJIdZM!
z*6m@hNbruX=X!7Kd87BvRyG|=8>{{ZK^mf~Mu|yXIikgH8I`u%?si{^qY_8Q+nBiy
zJ?xwtnp+=&vOf@{>w{cy%E8bUTE*1fGo?gIkzk@oZG3{Ae1>>8hKD}-`i-7<#3vsI
zzB-4@D^EjuP_2dcxNgEWr>l^DTV0qIK;L|APL$oQhuEj1@?o*U$4LBG8K50!?Vs$T
zhpm4`$GHgYN?Jei{O$*~T=?PRO0J!c*KhglyGQ=%`z!zOfA^mf&$CTC4sF8AGzIio
zs5UtjQYzv%vINf@Jxhw5Qecjr7?bMI#!x7bgHea+3XUcW_5mM1d|><G9p8Wdj#3-n
zU#|SgzxxY*{)b<2eo&I)Q2@v=@gjl?I?dC~iUvf{U$*jtc3a_uC^#2c%EY`RuDd$s
zz?wjj+tfl4&-U2@qgC>@vF#Jr;8?=3R?vVrOy33io-AQv)S@hdZO1iT7sVJ%NU6`z
zkpB)Q14o58de=FPRjwjJaKbW0<~iX+VK<k&Q)JMjQAa^UZ5pjN28Cf&ZSa8<J*Op^
zbq-2xYR4OuR)n%QaxTac)lSm7?o!OVsa9J(%)|@P1yTqU5lYsAAlJdXcw7p^DG{fM
zGOC8-Buz0~mW64W2tIIH7T&&o%@iiyzIu(1K_3h)+ZbMV+6Rc~h#2idv9?>zY<uMW
z^TvnEPMjl>j_yGBq`_Nv{Yi>dy@5N19vI6^nkOAA=FB;ImTBTNF9a7}EJs~e55WhD
z^Qz)+g;EkD1%qN7V?EnN>E4Il2f4J{1KRr|7CGJ?L`rMBa9y8m*QlMCc(Hh3x$)LT
zj0Tqm(-Mf2V>$<xb7DCq99s3(?g?2ylwuaVDvN_I2er%dsN0}y_-=3@96Asg3~4Oq
ziB}J2rfH&-tUU1n9_Y2;oe)!Exm)P1F-<|&5uhIF(FnnjLS&vI%frm6Zk#44+YhYU
zPA|{59b_9_DU<7kt!kK1T<)zH)CYSQgU9EOWHS4H;rjkPLDX871qzMmJu!NgIjVQs
zYraIc!=bf?cnINlTQo3Qg{FkhPw8tq#1a$vHa=a8Yq9DOnqg+4IkoQfR`qOi4rgC)
z-#6BM(*~gIxAD1*VRmLq-Ltji6e5JkJ_c>is?N?CsRd7`ntQ5u6-ztca6YmGRg^m?
zs?E|c0-OuP7;itDcg{Xv%^x^Y>5h~^wR|a6@P*Q6+x$Zn_(}KNOasMew@!4RP@d70
zMIsou==k%;`+rRKtWX)PnJ!1{@9g${nbzUxUJs{-xf)bnpr6IDB4Mbf-_f|6gkP^4
z|IHu&1+5Q~Ju|=ic;(&y_;2{`>B2An(?8_R&)#x(e{bL7C)c+9{H->fbPV0QPci5&
zJt-<Z98RVX%wBn<MS2F_^laF(4a)V2AKrcBFaGS;JgzItY3AvA<+82h%a!xJ3epdx
zv!G&k{U7c&r8&g%`Uri=h;WhHbA8kZUASB`m&>Mh@2UksXHarB%jPGmnAY^%7@ZV@
zvT!EN&!N^^u?T}B;A)DIQuVo<bH3><(iC+~!ONNWV_rXPGEf^Xx|fT&Yq$g=%+V`0
z*1!+7HLfLtSMfzbgD(03k{IJ<JR5~21<q3<MpXfCS>ZgrRm!e80MZ(+>j#2`80CG_
z4|@cLO!|R#9X7SBOzoPhK2#Y@De&db-tv3D_XUK+WxFy2&i7~L)1tv5(+oJgk1W&7
z-MsK{zQ@N%E<0t<R#XqOK1TX*H)ZLzXSQuOkcp~2pD!2g?<bAWbb%$!<gVlFAu7&P
znpdHf3Ik#YoKF*X4`=3O){8$z9_E?TyjUEk?lNnwv{Cf|6eFb+M1+0YsdWO;uGBmA
z#*U$Bf_S}ga;-$`(vy#Pf%}Jh=4DpE&aAl%$CJ8Mg(mqJ2*HzPEd*t6+C6#WG6bER
z;PmDerwy<7V%hSJ+J)Gop*8Knb?ZvHKKq7_^L1qCZ&Qg<3obg+95|g9{n=|ZyJ2O|
zopVg!9AOTe?k2nlWxev@UEun><IIkD>~dq$RmL<$mMOBNNGq8X)f2w&2CeB8>4l18
z%Nu*$c--ET9^UZo@oOGGe9yZN-*NVh=MNvaJifCBj@pIxTy??|g{BHQ5G7`=Yr!Fu
zG7OsZyIl;>^j_9N;84sT>7F`zFRkhN>J##si0WYwp^I0_rwGld9oCzUU@L{*E2S59
zgFk`o^PMx6{t=YaX|=E{CnO|#)xv^dk-Du>t=MZ_CtQ7~u+v9KEzxU3rX#>ez^PC$
zdYz={LtCkn(Fa0E%+sV?0Z|~4+7!7R2%~qs#;rKlNd`6|R=7{EyLRFVHK%)Qa%OgA
zD{zOcQ}kHtw|{rzTpY>)l@A;>0d|DwuG<TJ*va_mXd8153}r7I8V^4zlW+V_fBBBb
zHR~aYHYnbEdhh({pZqz0@$27mez@cR@L&Gt{Ospnnm)vSj-w#5v5=BvBPnPc?@<IC
zD`ij>gu2kg!t*qMsn<@!v0Yc5zWavX{`wpK!>_*PdCxp;ndj?D|M};vk01H`t1sz>
za28vGjDs74PKA!*lYVFmoJ|Nc%fTmYg!JDz&gYf+>6vx6RZ`dRv4*b6FpWTPqPa<C
zqh7C9CLf?yVz8x{X)X9fuc7%;Lg|%l+mxx|9Vtx&FE{N(sr5(QqCagiaE!7<>sg~|
zBM5T{EGZC!KB%uHb1j*tb)(e6wq@(~Uv?#kfknJCw6R*tVsPZD#YgLC#oJ#-*4It>
z*QIpUeb-&<=++f$xQ>jjJL`30+cGjbOAvnfd!O^a`Q;z-_U&g}YvG5kq_5?)aJsuw
z@4gQjuCwo~`_A385Mm@+YT&kRHxSRf%qq^qtZsH-^)`r3cy*q*n?1pGjTDtc@3zOO
zrv4qBT#Ei$cWSM85gtw_-n=>U>eY$UDKSKgnTPX<WnKuUvpy`GQ~Yl(H=PC8SP&&2
zYEl^Bj=ht+J%BW?98%KB$Z1;Gw!q!p9rv#vh~hXeCzg4pwx${dTL%QEBfRr8^KifL
zl(ViYyy!ZnD-XVRvn~nX#L_R7N8p04Y7SOK&Y6-k`?gVwt|KsVSMteL8`>o_nofaI
z#CZr&Cm^jV&1>BoQ93?2mgtE|JM(!8+?^&K{iY6#6iu+HK`0C%g^A#2oSSe`=uOB)
zC2ytb=S;7(>}dPW^LpVw{JTHm@%0-%eE%I&&3yR&JAQkPw0&o6n~o}^(~E{t$q-Ud
zoO241T8u2#g;qOxSKv}ezuN-BTgz{qjdtjQvq@5K8d$?&JNz4>yW;>m(%uk9O`STl
z!0^t|yDI;0d;sC5dkL26>Am{FlN~U4hwnxy700metG7lwj??477c?Jb-`Txq_ChfC
zkwd2dbMkmBG)yp1nC7i4%Skc1Vh6O&;y)<{UPi`X9)6Dh-(Y05S{FamRygzky}hU7
zZ)gxqv^lJc{nT?lY~8D8pnX0D-nn(hR^&J<4s@h{G{*YxkF|myoj?1_Z~51M^&T(~
z(y*xJplgiYJKJ?-^uho8|M}nXfBG-~bAI;o-(yIBv6|XQUc{4P(&>tK#{GX`P#Cw;
zv>3+X2O5=uP83>&$M+w3{NV%t!yo-IfBwyPniwlkO66azSN_=_{(@ip(|@GG#X63U
z+H_O?wl+8*S)gu+BU8f(xac%DCQhb%5a#PuwLpHSG^Y()KUfcXZAJ?w>b=uVh*(P{
z?>k-&9Zk1%M6Jq>&sc044X~~&`?jk-0TrnprC^Fb?iT%N0>V$$0-xS+QwTgvGgGk0
z(4iy!>sGmJim9x-a=>+82F?d+)oLf$xHq?M1UuzeAEWZ#^kz@d;pnVe=J{IaUD&rm
zih;b%)V-)MD0o`Y$l0flS7i~5!6|qi9?lv@LLdb=Pm@~ek_PvPL~7Ug+RMi$F4qg6
zpC4`%r<nBMs3H{erHXUJkT}*T4q?<r44%*L&wTazfrpcI*8PRseOPrg4NTJ%FYC^A
z-GRZfMD8AD;-ur_T&wP^9E{+IOQL~_EJdv?B3hzt+cRYph79J@#QAQaQ;DaDf7HWM
zoOYZcIL_yVyYs>{EzD_Vj7mz{?aw|k$AuIZ#1$===y*}mmJH$HeCEx=J#Ec;aK}iP
zJU#D}ti?|7LDw+zM6OwFOO|jb&J$B&SW&m<LazgcR?WxW_=uDBKI$)5a9wwHs)E*i
zoEG$s8=v#&cAOFJn(pzZ5R~#b^s}kGu<tw9OQz;dE1mU{*)Ex~6<TQw@7Zc0*NWQ<
zF>!On?3Z_x2-0?q-j>0Bd1PC6_FPE8F(=2<Hdxm}ZG#jYd9RdMZC~2x)l+iO>CJ#x
ze%#*g@9XP0(2g;=52~D3k9lHr6?8HVUD8VzD>kZClBU<VU8mi89vi)M7I8!;L=m!6
zcg8>r{VgvChv>n4KA&+hvJdSny=hP+23k>vh;mfCsk_}vCp|j7?CiTExRW|>L_ceH
zd@cK)#XHSvnUbyw5NiNz#c=FIZbCer?~Y?)(9K~#D)e&0QXdD_0T}vJkT$@qkR2bR
z*>8{agpH-`uZDxN=Wgo`u~C7;7#E@>)EBLhu5z7y31?cR_wOG0um78Wqd%Kq7+!r3
zZJ~GDRBYQ$TsQv3|Lb4!U;J19XWo4ATE&~TvcT&>IZqQQCW24IkO+QKSd;00qKPqY
z)=Ce;!7UIzeE7hB|1bU}|LuSGNBsJ`cLbZv9FvFI8vpnI_FwWOj{ocb<$sU2wG|?a
zp(_Xd%@#Iix)3i;n^zOQ$Eaa#G3r#gwZXhhoR*{zz6#BIT?ce^#38z3^?hivsku9x
z6FL<{1>O4I+jVm=M@+w>E0V#!?d)r2%Uzu{n7d7stMFr9KW#D~;G>R&+Zd+&f6)uX
z7!U@%6`r@1r_0X1Hd<56t=L5zl2@;Ov9XwkAsPaad@wPac9XejxP(KonS(|#C{Sxc
z>&iQ*+s?jjhAK{slG!dBj~}18TvpluKSk~Kw~hN&aMD@4C?3=~d>eFWP*lkAe7#ck
z9oH3|>Acd>Vww;a=%y^bu?Ix0QqBCn-U-t*asP0~!>fCG3iKH1wNr~8UPB<~Vqrjy
z(mpu=f*z1MCmz&M*?Xh7&PW1IQMYMQuWb;^Xy5nS@AuL1=CfDazq+GK<KfjE>&Hzk
ze$nBRXR;iLI7Q+VZ>VXo-0f*1ri6D9@T9n?SJp?Suz;)38t1zc+w+z6y6FhZYqzxz
z0X(x$Oavwq8n)IbRKVFys{8vbY_~E0^|}*-!fKKWwg)>bXWFThtn7+InXOhqM+h(_
z&vHtX(I`~|pbnT32qF5Lhr0J}w#Z5?g(Xe|Jhe2|HM3m`(FF@3Inw2n6;Wy?mz_OF
zO3UndW%R=0L}`e*GU{ORy88@Xk+{-ycSuJ=yHMtYbfJsd3P&Hb?RPz>2%M<rTI~7S
zO(}jP!Kqm8<scNqG(=WpJDqqr2huyG2c{Yjns%qr(PN-=C2ux7TgmLTVDbc<IOa4F
zgAzkqRr1?G*ApiY1nC}Wx=uSlFxUYc%)c-FIQ)e<MWPJTIB0VlVj|8b=IKl{r;Zi_
z+zxr*uT4SlFNgXO+SEq_l*0Kk*4JyW8MN)7+x<0#zjqpBW5Gki#`e~B>*@t?9fsH>
z!M^@6k-!xFM>Ma}8c|`ZIJ`jXDgymCfBqXDuiI@R(~nG<UTtz=V|~%7LD>tt-}yKH
z>VM|H_^<wtboVkDaXygJaquUck2s(3KH|)(9c^lGgfboLMCSrB9M6}P|J#54|KMNz
z^MA?5%au><n#cDM48i=*|NB4U_kaE~{<DAfPfYJ8v?eb?d7HB6s?!LD&e_V-AvbZT
zf|6pKI87(c^NBe=l8fq$#M@`k)eR86zv(-~nz~~}<DA0<U4LjOlNQuRK^Mg1i#E!8
z*?GENdAe@g@8t$<L+_X$b$|Y(V!%03I8E|U8%?mPNIxuPQ7hN^<3&62noV#wDo9rp
zaSHh4k>GVy6&=KZ#LoFValb4eXP&NCzJ345r9i71V3EA~%LQ;lQ}IHW814T@-M=;2
zl5O{W*l)~BX0BCLd+$EojdKpb3kU)vz>8(cv}r~tWJUNvFa7}ja`vkoj&S%AiV#Dx
zB_bq|BE^eDP#^&U1kS;^b)W9udso$3D|6n4AI8Y64bmdI9~gm7#76h&uC7{@nRCuD
z{^Nfc^itV^P@6d$uB)LFr5QOUNTh6quim~rjC@XWWQlGF(XIL$wXt1pJUu>HT3JwJ
zG)CVnb!ywtR+&Q}4K@913M}))m6dKjDMkiVXW7p)%iRJ~W~|~y7K1ru-nL;m-}B5Z
zuP8Rp7Zl0Daz2rlnK(MNtURw*9$r15E`c$nR3MQEA=_+L&BFHX{*Kr8cbw1nlv<gW
z+1vu>iI5}bhckB%XYTG#oX(56eyQ30j$yvqIc4&^82ZsGIVOe(bDEgq#6Xy$Wfi0x
z$jKIG3{EWEQ6%S8#285fa!AZ+He!^Tmb$jWZCaT^qHm3ItH@%y<S9l(B3m~%$XYAw
zY905EqQwwRWpSEW&NHzMPUlSD)D$a|S(LWch|w0aIQu#TD<n>*h4cMF7?BWcR%|o5
zb`&|$3xc7;J?I;|C;sL3$aQ_TS)aO2Sd@8+%pvn|f9BeSFCQ<S{uv=c^HfDcA3f&R
zyS`q(-D3iY$zx>255f=T;KQ{8$0m(KwAK;DJy$zruOd334W^X1J1xw)`>c65x-%GY
zLc_%KdL!JP=(W32#3Y<3-lPoobsyfXKu(h(QOt%$?j*b6`2?~f38qrWAuuQN3<z$N
zqbj7lkmuQmXb$h%IVx^g1LicTh+EDQcB|vzPJu5z^M;@Sh_9*rK)2I-9DD%3=feWN
zY$Wp`0O6gl!`ws6ui!{I+*^KLij+gnrn}_MPc0y%;GBx_g61GT1<~VsMGSq35qnOh
zGIHaK&%UHLtceTedyuAyd9pOeoGol9hKz(v2oo_+#5C;&h4xv^5Eu;P<L~{$-{Qah
zd;iGB&B#H5#K$SYkwU~28vp$Fe!yS(%YV^e{^9@6xn{@<*D0#d{p7SFQ|z7$B@8ow
zGf8YBhd_>QV&rJm=vD~Yh{lTJJP8rE9e2ftL|10TV1=N<@C*hkXgVn;athQ^d3t)!
z`}glTohKXnX1mv={5v2J;ve3KF8YC_&7^&6>1ePx%J-Km>$*{j&$xn;vq_~iU>3Oo
zs)3vXd5SDc=Kj^0`}yRM9c-<#T`RSiJ#Y!;7IvB=8Us-c=ALpk!<}T%T&ay~wGf=%
z;gp4Wj=XvO!0B{C#J*3SCUTlROWvyKlrsx3YG=taQ_7%?db=VaQnwpp+qetD1SFIt
zN6u4XTcK)a8-vjX*R62fRvyOSeqIPEQhPNGd97^Af~3T&S9jdMzT?&^k}fcc!DOcs
zZ{ED&)$0cUN^7<#_cu}1Hq;o=Ff8PwH`cW<hH|%@nDW^(y=`Hg9!}if-O+XM>h(RZ
zUcKVs;emNMF^~_G4aRn!mzl+tB2+8hv<wn~B<2@@n1h@Uh(jpanZ%06lmqKy;d)!C
zefZ`lKHdn7X#Whg6w5G_EWwOOw$3(irGp&)7V2^=Z8pA{LSl*&DMfUcI_2(up(&iE
zi5>%Y=b4A|LRWb2L_tw8`)My!mucd3nt1j4gcdlRXY%r*JQ&{5#%P_=hpA^dnWWMb
zCpI5!&iWoGY050K6<zlO?j9CL+X6kR+304odF|aZA{3NC*7%K<>YutR%}72$@75T2
zSFA1*+z%^pO}@IZ7fC72JJys5Luq9t1e1Wqlx)m4#tYv-0`cxU^e_|G3)AgH+g3&|
z<e138Qd4&Zf51i%1ADIXJRxIY%8}EQOcvS3k%}1^%&vEzr-><NLkyP_X_~F`$2M}a
z)=m!f`tR(BhaJck)Fy(DUiaL4m;d|08Z?-vf%@DSUOYMcPd}Xf9P!7-Y<6!wzd?5+
zBzpwTeKaGb0XyA@1D!GSYshf>e%pKcK0}XzD3PrVe)xkwa^3?+5wU|12QR`H2?t?u
zTURs$e)NMM^J{<kmjPq(?_|#y(lHnBYWWd6*P^vX)7|I637A7+Y-}y?U;I~pm);7s
z8V4c8=rPdFUxJsB_Za-(M?d3>$7iN_c4b0GrMml{=N<%{#OMZdBL0lcL}{J}DP>x#
z7DBgkEk>u6YN`*v2T`IEs+~lu5!DLp(Zi-D%F&$;MNMjs6~qI31~De8R-P{JdH?PU
z<~dVqVM!(%|94Uhm;_D@mMxuLJ9Dy)H%%Eykv0a8>y7ufmCa}2_6rLko0V}$(H@@~
z+jg1*rzNqRGGsxseeJ@T<#{8=t4S06|3J)MHwUVS>GtnWoL=7(XCs(g<cg$=X^CUt
zG*8^$op|&573cZHayl`)+0HUs<Y?^;#vl$g=cyQ3=7pTiIH;9Et)0u$Gp$q-?z|WR
zrx-csY@Yain5RL9Z7W=_SJvB&GCcIhJJOIN<VazV&$G!0-BoZ*iTOP9-Cy{W?|%9V
zoX(2}rUY8=OldNBuMKK1#zpXW>8Iys9xu<Vx0TU4y(?R(TyHB;%+c_pSFbGH@&2CE
zd}5jwQZlV=Ll1~)j6sOe91wFPswGZpG{_6g0Ocj48~5?H>ORRZ)%d&H5qz{G0XB>E
zt^=F9E0_^cY4mC?fs8?`W&tXg6WKD&+?`I`%1SHdBxoCnBM;HWA5$0R5Gg^pdpI+n
zX1YQ?O-Q~SBZHVsN|{o$u%jr1IUiD{Y0eO!dFN@iteB;>1WJo6apvx{uq-p%ZT8Tv
z=$oU#{v=x~r+IKb&mcmcqs{fi#xi^CYi*r2hI=-Ala>0X?~<5MiR=!dJCqZ`0SN5W
z4ufW85Mm~$iBX+M=ZHy~W+SDlxex}4HdgGN-tB#T{pKS^oSCjO>%4HgKGEuWK>U4}
zf!P$yB&=g<geZY!URb8gGEGd$Z@wz6Y>ZY2qcbJbyD!<)K{+O-=|r4Prb2L+z!=5Q
z0_24_EOInbXgEf4rY5ox;hyp;AtD}`v;Wx@2>YR{?r1sMIN0gQ54hcyUS;?w(g|$A
z3kbEeQ0v%tj)4C+ikT?+GcguH$6X$b?A8WR0(GlAy?ZpILiZuu3)*m;;p2~^7p-+d
zDZKmYExr9YljXY0pVrQW4eb!91X)?8iFUH=7i+c^=^fE-$%P(L^*ev?89(~rkF80q
z&7Z-)b~$J<)TE4GzJ26>`rRM$i+|>O?)k7c9)>{g_UkbkN*O(IQXMszrp#%X$@7eg
zvR*DcU2a@&7P;N(IQG8iBZ26uh@gSiq1NJm8aZaO>;~q%a~+i0yQ$i`?GGa64e|c*
ziTu-_(A(g2UW{-b5|@7$(n1lOdG3J|z?32n%fk6IlP1&HUP|TZw$ZBDEe>_!R<vqA
zrg*?<NFc;)uvy4d;IXcxwh=K6e{_;WK$(Mm=-%z6WJ<*IjGiX?6v>Op?P9Bt5<_gG
zpXZ6w*;41Ksu`7pOrB1Jkmx!n!}Ol?!B!g@B3%W=6aU&^-3r%rvyD4v27?@6R*1DQ
zZdbPF8?AKCA(F+6m3pmg>q-|jV;^@dtlBAV>s(Rx$F@wFSKoQy<4-^4&8s)0oOsF?
z=2VHjaL#9PPW0Z{YGrF1!?pR>+l}YTjnN@XqPNO=3p_nOQEug6R|jl<I!!00yqG(o
zSFP$6zk9c~!?rObQV{n8HE6qzX8zc<IaG*A`R2$R5=}d4HZS%$n}<VnX#J20vCqgm
zwRbMptC96otq2-6Ck8mpGxv9AmSy6!EsSWANI5zsj)l|5pn6A=rp)Q?%vLMYk~}tA
zm~&#z6H&|^uB*k|p7M*F)6^ylL)7ngK_bjkVkjOzt>z?|GpBiCS!M#y+^&Vv21A7r
z1WkcyPMlAZANDr8oU$D%+SNEQ-sxDO`t@Q+|Fm9%K<ggBr90`&WUi<`XYasvg^Yv*
zX|{hqhK){=5H-+iHH%`*xC@{EH))6$gJH=GU5r(&!)Vn$|E;6h^H5=k@kvGq$T05*
z<HKa;6w#;$PlXumYl0YgEk>M)i|N||Yy{Ycr5dU>c1P79eo%UH-##NBs;pbFE>5jJ
zh(QhpeaO)`7PdTHg4yY{89$|4A+VFilE{$?xCeagh`^32423cVF(#IroFTuH(rmP+
zh=-0TwGB5@GhS5fg9eC1p5DEK-9=Pi*5Gja+#Mb0!$vHIFk2zhYUKyN_XqsOZ~VFu
zafg!TNSY=}sNBNpQ*Rr|E;*8;9Ah*aMOglTh|t2q-~Rvn*KF&`Ry%Dd*==z|j;RZX
z%SN57nx?@Y{P3rYG^5%MZwu}HNLt+Y7ju;~LNgeTAmzmQ?#%teJuwN}?V0EI&pcf>
z%Ytby_Rasgv*VuEv2G?qUvI@w*VhjwsdRVQ(Hf;yw$hGdTN`}{+t&E`=U;NYHR9=x
z6em)e+$ic#zP_=o*@uytl9bat^XBf3`(<ICC!5_3p;m+7Y%XiAdlI9Ob5cC!fCBU$
zs6|-AFi-6SIS9)v%=Z(^{ld0Y&N=dG$sY9^x%R=>XE9M|IxLS~2VG!mjoNJnn{q@1
zF5AlI@83a=Ow+`&czkZHJl!soHqbu!;_1S*HqO&*aBs6X*?j>Ro(99upQK70jdHzm
zeR?L0&Uu-6bGLAwC$rC-CY$FsQ$mOiO4%&SdRu8S`0D*L&rd7&!#mmYLh^gpu;1x^
zKJ)Rbk4P!7)yiePQAcAp-cU8el$`SM61BB(yWDtudLpNZQY*wjnkHxKX7lHE6Q|Y&
zrBs`zVuxz$<Csf^07IN%D@Jx|wXv<{!!VL9_>=+?l$Zmj^NIWWJLY9UH*<;`V_0@W
zPDB;9ZMCkzXK^t^#!%xUL`#9oIdfVT?(Xl{MrYh^*3s|8E-@I=7GO2l&VNpJ7tAR|
z?v{yVGFjxDXD=u$Vs&ekhr2{xW|D`B#1z~J3oBBS1t9g-X|<X!#tuD)z%pk9<#JhB
zo3V&fGJZguBFkxV&I>-vj^;?`1(og_avzRCeS5FR=L@x2K|e;f6gC~~8SFb!VCkLv
zjC&BHy~}$nw{7Nb2-JtkRNb4P_fA(cHhaEZp;_=$t<4)$aoF4w;QV`y-ssIq!fJ_~
zT_RN!E#sJv_u7!rP1PWguFBS&%&d?#GFSmMb0rV;NjVu8hrOd8hh_IpP7A@{+}pYt
z8$3nM^TO!_OGr$?_fHiUbT?{Olh57?Wm}=Uhl_i?BtN(Y{Y91GZLBzFU~5)9EiqZX
ziB}uq9uSHc(NazARg0yjn!BncK86r^_xVo^x>h@2`-cQcKQE4STE>w@AYwA$cW>YE
z)1Q3C{lhCW346xQi4ZKk5B0{(37WlOB+CY_G9nm)+1h)4^hZBrbPKaOK$U$=vR4B>
zmN_^FI{5uR_yMQ;*ETxQe*E*^yZhqoN(3J<MCXpgOqv&#OXvJph)L;frQL4St(t4(
z*h!aiSV!;oq>sT?tkJx!h1+doPSep8?1iHtu(p@jD=kNSw+`WUTTzABJ29O}x&NDp
z$p6gs%?N}sLcnJ!LL_pY%$UWZIfYxTw9y%Ad~X$Z5mP~;1p<LXeq#^`wAvUwP+Mnd
zgL#R>lR?PSIhk|aI`J?kK7Mt!#lAJ#tx^zDNL;TgL}l()h2JiP$BU7^#G^qGGp^az
zmG_U2yt+H_d@)zEORZe4H{&Rj!sqWFY1QOTkCzMAR_MKR+X}Z*XkL*?2u2pv!Fp{x
zKdrobe<cXKInUgk&z#Q-%RHH#>gXhevR>)y%I&tACVpMHY%5=U`Ht)R3#ap$y-Si9
zDXp;98+|l><IjJe?<~sD3J<S&d%I(+E45c{w`XEZ3}BiP^OVVPqITzUyuXsuO3sm2
z4<B)t?szrbampvE4$5e>S`dvetiu^~(5o^k*7*#V;&nwsrS(Rwm2I<vqqmM?CL`-^
z%?#vIA}pCaPju~$9GN`Pk|`$}+vLQ2cVb>5(;C@oCyUMG5iz;evXEm2xAa}6nK@4+
zoblWTWvyr}R0%ZH(kJsoNQo3PA-HnIXOlS^H>4Y$%bX)|%4APYl&DM*w2SN8J54P)
zY+VZ@xCLyqO+v*s#M@Tbs%>a7LQcVEoe?vu$@ZL*gbxsbVK5-rocP-f4DjU_@0g}5
z2`6RkOlQ##U}Z-PLbU7>gpfB=50wHww>B~#heTH#QQDOz13}7{)F2l3UF$Jk3aWH%
z2DJ_->W*;N73l_}8^YDmD~?<^dC=R+Brv8#-QDlJ-OiIS)@y6DY6eItXX7Y@<b*)=
zI9eNjS^Te|jJ<c(Qmk-InOFC(co;&MPH1;Ru3AUU2;7SDcxzlQ8|(E-uY1UqeeU3U
z?{Jk$z{!RHF&*=Bn_ig*!=d#@%>yDN2xddP)yBG2YW2ibRa1ya<;VZ>k0@I;BH<X0
zD!z2TJ9%=yb{_^2p*IVZ`@=u@Lw@7e{|a^ZVO<-0<gs9CGGYRE=-d@RDOp(EGzU7J
zuYURyp5MQ>16+6cxA<7V$(TN>I82!Qqt;rvt{YR%*4R3CYLAq+qQ;9V8HCXpakrs1
z_t0A@P#UAIP>Tmk$&2CIPTmwywZOJM40W+WOWnr>z?C8*hvZhocni_@*ntpYgJ@&<
zwN~a*kTB7EJW@9Qr0W|76=8nd$>vd04BVe)PK#yP+rijbTV+UhiJE=*F$M`sZ5w@1
zdNY_MIXMIv4YqZFD0$8_X@m^tHxnT!_Y(Q&^#hXx(z}i4=O><T8_#aAwAIRTnu#%R
zdD?h>Ua7S*xjZDz$!79XWc0%2b^#Hd%f|cL?U-3VuPaa6=Kju5%SP#i(mGG;%HwsV
zRkMh^n=>E33Lu%SSKd9XW^@w*_os=|GO^4PQ7j+3)<P>Q^>$;utX!Y27PiwmYwO&e
zR_a#J^NCW56H!)fw;R1R{`#N&CC-mO;s5l1{Y`%K#UuBxKjK$@?Jx5m{m1_yAsTT+
zfG%o?#gv)nnNUp?U>OdLS{q3schkcCbYf1k6)F-KA(KP0pp1bT;8bN)HD+!#+t?VD
zkj0t7h0F7W>uqI@5mDohwAxv>Ms1xKgR|!oy_-Gj7{>B9d6We><(xR3CgvQOr%2Vo
zJef(%02>2|k!n&1OgS>m>6m%8K4@FF87g{6dg7Fsrpb^QjL$KKnw+z@!~eT%*5RD*
z@5~`Gm>E`x!I{%a*UqinkY~8uR@Pd0_3FVAJLky^g`$~}7+blbK`7cNQW*)tWU{U}
znZz-LWax_sAxC1fY>NXL{uW=KfBu>AA@=MG3C<08*`UKd67)jA*z(C>{jxE%n}38$
zno+0Dxua>=xd=|UDr3-^$@5Y&NvHs=7nc)SSJrw**W&%B(&vfRD;ll)uhm>ZMeJwI
zIni1s$LOX$N+|_{*45F7dLTrWPPr5KF!MGS`kO+)&@M|cwJuabq7I=+LY6yDArt$H
z!lbv(x)z>2cV%1Mqkgx0w%cSsL)8#Tj~xa<QbvM}#bS))5J0OrwziH2bF!PJnGlWd
zaJ}8QuA3t*iUuQF2d#YP6!}9jqHnWYW6O|np2ERz;NRp8SjueU2fzOZeDd+9eDeM8
zyL+oiDi89ZSbl|gfY<)t#bE3DG-u-2_`~1%=b(mU?w9U&TP&YNF*LIeHA1@+FW<a=
zO`fuGQDsMx#I^cDkmO_XVI%Iak5SE!0w)mnQO&W}&3N`STl*Se#2ulq5z!IcrT+PX
zji6F8a_auQB?zI*i+NuNbQxw%?2>yOhDw#PG4b!H7;ryCn~{l-f|F?G>|L9aO2iI{
z6kqbZMAa-jL!j4AYo;aD$quX_0m7gM*xDd$jj#lkITFu_QwYpN^8JGJPF@lclv+Bs
zwOE+Xc_K}=X}1Pju_`Mj=pwLW!rjbtT1YW*8%i&Qty_l|oG{YV0*(ZGNlw$mPNA)3
z<8pbXIYTugR{8Z#DUI7^%7+{ScT48(Jd@`{*Un|V@U+?(p|2b7pKiQ+_sjs+t@8Y|
z(ypeLC)g%cZ!2xR(Qh}tcYek9=E3J*t^DRc`B$_)2z<sr{q29rwXFQD|LAYAww2N?
z;x~c?s^o|^VahXenoJqse?N!JoMxt&5p0e-OCpElIqUu+4W*Yx>7A{1B9*Kd%i7)k
zuGd1{D)VAq>DyWuwX$9tw`*aK@dQ&DgcKbrb!gw;FG*&cG^I>Rm3}KsF)^h~9Rtze
zb5(P}jFQ{~!2(5k?Uc1qx6UM(S^z_~QcP&AHX|;>tw4u&<TlfOzHVHvn{gzv{Wl>Q
z$uGn}D~;=I<Nfp179+#MuOiE7K8&fR(?U8Mk#Vbo7Q`jBm}PcW+Scg9x&*W978)jd
z3Mmp=caq$<yBhe>Pk%<v**7Kuf^Dpr@<L>Ir}z@GqiSQ4fug-33}<sEIL0RG8>WQ@
z4NrK~@gj5E+50I*6enhSF|TS_AVzey2yh*in-K=i<`1qtHzd=>hGw0->-*K=<36xr
zyEV#M?DP*p$`c1OZ<x`Kg2T+oI{Ea0<6X*h70OVqwLxiQH}g?9D%&=Dk42)VHCpl0
z9gaS_U>`U2fk_vGt(+Wv^bjrRskcVQ;^9XZq`9-6m?4e^_v&|oWB>?C7=aue@*jJO
zq`~3C&6_^tzaB8VzX=EqqYwV+|M*Y&s|er!rC)?@D4#L!vC&*kM9*I^vmqns=AgXJ
z!WW;t<+IN}GY*OGT^2Zm%L$FJ*ZK>pw39bg;g|p1pP?fbw~aNC!804IyEoEcA|6GG
zc|I5!v-eIA`zyoFln@ihAntrDmnb$>Mt6Qzw~+)GzfMwOniujkTh@#7xkLgn=s~*O
zgZ<2uFS$y>Fd$VkFhjnc66qU`2?BAljgdCU*=03eRT0dUuV}ZCKuT75b%%_a1<s77
zw5=1ONo5AGObgSJ&B#bATkCKsn17Ci8I=HQgosclp(oR53Ifq)bJJuUXYa#h9JU<d
zJi#&%P7C3DHaJC<wXR0|%84o1%?#^m4p=#{%oC@1Ld5Jww_137y3kuSkA+~)X96}i
zE)8uKg!B6T%>DUn_kqE>ZCq{_W7Thk=gY>rby`sB^OfiKSL)Wt5=^>+BF$#vx)r|j
zaECVdz0ckol#JP4j;8$Q|JnbS^Xr8--~X7)dO?SI49ut57Q91KIVIlQJuv6lkOEak
zYm8=^qNtE15W$oc(OkY_!ngso6W1&6zx)~RTjPKA*ME(lwT@JEOIdrTdmC)mg0@Cm
zD|It<fV!r8m!o=v_p(u7hj%RlZC32*5Cj`rh}uRVZr8i5dlaz|qUc6Qx~;2u6r@0v
zM%7)0V;SnBE3I^U|LUNXPHFaB-akE=8YEZ&5quVz@<gd-deyrDuX&mXrBHh#Oy<Ec
zbWqyHkWQQ;F$ZD_l+jS!8AUC1^LE)NQn@~C%#v73qpNMgL4`b$#k7CR<phtH8}mFo
z|6u}2bd3m>4hrfA`K}tW&TQ;Zwps#V6x&aNMA~R5%IPGeXav%scGwT^v~n`1%`uds
z)tTYLiHt80@m^|$5RHJCCsLjavD#5G?KUzB9udo~x)G9OJKT-$iedq2sEzqTNdC2I
zdzQYK>qtltA`(A{Fy|%oY(Lb!y*Aj|<C$^C&X^K?bVLj$jh?e10;3h!y&6I!XFFh`
zlVc^A<Kj;EQ*CDI7579_gUlD)%b*lmt)4lMpjUd?02~E&V<@p$c#X<vK5ntEI?%42
z%V5Dm#=ZF;{^395%b))_{?cFlD~9L=Ph|CCJct-s%aH=r!NdK+M^oUx`R(5^^@F?a
z?b5*miEy-JuSj+Sq@Zdlt2b}>^0Ms|79bkoaD-#55rWAA6>A<8Lt9<->4)`REJacf
zB7$~nFx~Ck<e9F3+0`W}L^PRlV4f#pP6sEVt8q~F;smVFNx+ottZjYJ+KcQzIMViY
zu5VThq+}mXJVJ4rnWuc{ty^ncdZ8U|VlVK}=t>tm{Cl-(F}O#xM4QFMlnBA(Fd>LZ
zF{;UPN-y+Dct>DLiCbvs90^MzjzKz^qn^1SMz>BjEpZev$Gth*=HforL4;e|*xJSt
zCz}~#Oi_Y$r%TFAQ(~DGg4hi5?difP&Wzr9eSgQCXCx&ffz)P_rx=)$9V7<jKp8fx
z>}tV4x7G|+3yL<gy&Y>~%9&-#oTm(<vt6FK-YU~auc`CG@BYzS55e+|QmxYcFaNXO
z<Ujo1|F6*{LvP%lPt5Z~z`CF2FYdgYOplMStt+?dg`tCxjBF@v5V{@QOBBz9i2V4E
ze!y@2<A21z_?_S3`RSR_JOB6p=R(MtzxKcQE4KLzvl70oFx?fKue$uG*G?~$;fZ86
zQgp$nJD?3GDLuZs@%H_V$LAZy(l$x!UO%^D5vAK^y4>dK1W_C7T&`Dcm(`NLB+!kb
zZhr4=crb|i_h{6$Qm$9Fb>++V@2PE&LOfz<jUf(Y8<tBTkvUD=<(XUS>|rI{SOeVZ
zN*U#t-%2u5DDfg_yH=jQy7KgHMN;JXVWsDZ>usad#_$d~Oj3sOy8r<I07*naRFTm1
zk3pW_#%p)3%OQGb&_o|5RYMXwgu$>Jjf0$(7!nThmIOP~%(hwx9q;a<nY^TA8}u+o
z!d+%+p8XyK(>i0c12x#)<WN(jET=oB<wQQ8teYQ$UK=snI?Bmsx1K2ycT=rliKD1V
zYTFpZWCcRnQ%c<_PkfxA1aqu&i`;3R&3iwXI${{@$~wXTV28$zdKgsggus{~bV6!m
z5wq@12^Gs;p$H*?h&iaHX(q-*Opg9(hbG309)_k^^qn21G4I`}?T^K|5;<n7G(&ym
z#asv75kKo*qdOLAAAVq(f@W6*jNTco5wltD{_qcfK&g#i|7(BMoE&qu(b)%C0isGA
z@a8V_{f|!k;9vYMKm5thEctW)JYqj*CqwQMWIOB)2_Ahg<;bg75B$QXpYU$`%zY;~
zkkK8|Kahl-3&HN`=+8s3XCzjTq!7p%vL-~Mzpv(0TZeH|N_ojF*pa4~0?Ry`j9YP%
z?5_3lKSv|w4v}JH$=*AXOnG7C{poA1(Qn2X2xhDxZm_b?#G}gu+fYiiIq^>N&}LQg
z5Q9TP_kZtjUXQ_F1e<f@*%;r};WcWrXx%&;to5j?b#_S^vl5qx4C`EbZ#FV0us(0p
zt=hcM{lQh_VC08ry7skhJeMnLY$#?tRBNZL8@E1qI!`<-0lKox6Db5<-Q98N7p7(A
z`yYLWPv3lpoCBY$6$7O^BflF7%u<4_((fYFp;X19Q+<w9Yoiq_gmSWo;=5%gBc@`g
zjcGZd`Hm4MCzN<kEe7qq{pykb>TmyD{>Fd&H_)LhvlTGCb({A+-)OGc?!fbQ<7aQb
zK!MlY%$RaYj9#f_rEjav`D5mXzyJIEKmK3;Zyw*j$D>eCRo*>c2`GQ>zx{^*=|C30
z{MldmWky$SH@Ni9eQ7qLP^+N3!?Sg(KGWPKS$m+0Ax7(E<Nf2xS5Jla*T!ivVpFe!
zifty<iK^RHt&r%$>|=piirELRgLx<+3`S6NSV4e9S{rOzqjhDytX!Th+%C`D)|Iaw
zFZ8Bn3Szr4s%DQ|>tL%^h|D1|#Y9PoD6lMxrEb=G1ggZCK!wpPk!)Klrzl*X3h&--
zJiXt@Q)Ydtq(!M)r!|X2ou*_m+R><AbME)2^&09Hv-8Q^<6<TZ5Q$7o_RiIf(JH&W
zuI$oY1s!`txV3DynVaOecRLi#9ySrP5}l>LcgU@qKZJt&!aIB%;(<KoX<=IKn9paD
zI|oLTn9R^9h>%1XqxwwS3Ka3ip;Ze<Ln5jsUEbw^A;u%zCImOPQ`oJ7MT|ta$MO!B
z(Z&$HnRN%ceSWSl??VU{u#{p%gL$|JN=m_rn2_W}TbFYtO_OWg%@!OrC&9e)HFgNy
z4<T_skd#c${NmbZ$$rn*E6q6#R_Ivqw;L^`Xk4D%z0>?20znM=?%j-edaICTRN==z
z{2?EI{4u}stG{XpM&9S=fZ8EH1Hb%BAJebz`A>iI?{lk_IYq{v8zV1Chpmm{{_TXq
zJx)8N#QohFf;sB8ZV1(0G!A`1G4_H;+;^bsOP4M#leP{XbGpq!CNngmt(XK7_lw{~
z!nyAkyUQS0m{N@e>5+gdm2@vC<tRdiKihre(OM(==Stz<!I;1@wT(zK&yoA{#GLm$
zX*>MZVr)~Rl^W~!AYXQ6H0KWN3ld%`<u3H@DR#l&pHey!136e!Fxh5OGDBJd^PCvN
zW;ktaJm0R=YvcL-Gj(gk5Sfz63U9UYyls?PC=y60w_14IE({UYc_op!TsPLsjYWmW
zckjqFo~LI{%bDxec)nb@U2mkj6Olm_m_lTZ`H<yxXWy1SxJzbQ(-@?*v1*|VwM54;
zNWtdCL2dSvQ)JGG`{l&lG;`~Ac*-zJxxJ)v$W07NfsSx}y71)>e#H5EpVC^RZiTB}
zczXAqr?*ceqDO_o`}N9GSy@Nr7wJ6o$}-Pv^~!a<Qn!`M<Awj~zx+G=kN@<yoS5Wr
zJb>EfY?;}GAO7G+eD~8|G?Sxlp6_!qlHl+zytPi*jBFJ`vRQEm;6-6s&8hJH<;vq_
zrIb#Z+*z?Tla7jbH9fdqZw8T7<1QdVsT<eZ%I(%zhRq0d${w6z$%AXDT(&~plv`Jx
z)+_6Fp{#}VT8Yt!YYdYb*4n70(wkD7k)k%26Lyr--5qZ}dc()Bzr$KK%BEc3-N^F9
z6cSnoqc*rTsw$6fuROlHP}fcrp>2i2pcf-x#qMnD0Fom8+g%HAT23sdGpF+%NHlMT
zC^jc2(5rQPx?9GAg|a9FHCM*n&)>*a!F>Dspq55S)nIkhMmr8YO?hIPW@;Y>D}+d?
z8!w?r*7YvS9VyMEdFFIl$n%1PJ$P*#&U*A}VLEaDoO|a8D#cK`W(2@Axi?43hElM1
z#O8!IFxNmc7<xHu^E6qMvd_pv2pq{kzz%P}AQCTyPus^Ad!|4%spjZ=;;b3A<tZPB
zodRPIj|yh7ZU$oJyE0<5@thx|d!dkWMpB~OjB~RmCk98q_Ss9^=i2h&jN4?#qjv)K
zIcw{5sYEqoR{P+e|JHA@tt<cD|K@L!mopNR{casZI_>E_zyCX5@^}90zsr~Jo(KfC
zZsvG)pX|P>s|wWbD~?=`ranE1{KjAUOZ1Q-CsI()!q}1R7wc@F5m^!E4vZlNZvY}u
zeN3XAloInCNdh{I8`4VqfWHvkW!w28J_f3FbMv(gZ8J-Xq$ob3l9w}6KlpromzLx2
z#Rv1RuW@}t2SR&ELKKp~Ic2hcI6|QI&TXr--s#OBi1>V{`%Ay?Y7Wta5FL&3|BaE?
zBq$?95S!D@^Tg~aRizb5EsPM*NlBWR<_WNk;c{8ITsLa#%wjCwp-Qcdr{^nSG;XTQ
z$>d3G+b9a{T1k>vw?VxXq71g@%DcAlHYDb0rgh<~#~at1$>?64PW-~F2exK&v~Am*
zC}$P5Rq4Y$sFN8S4WW0q)o!}^DVp6b#sHpj;<QZ6Q!<FUm?fqaVjDX8-kTrKM*OLF
zg&+OlPx(LmwO{A)cIEx!g}{ZMe(?ogeDRi#Up;t*8Q4nU{q-4bX0TC5<*uJ8>xJj*
z6JLJ$760XL{(X8a?r-jdo`X1JKXXVR3=4PKwni9(TQ_r(J@1@g^3qy5b+U0$io^ZR
zclnMx6c%NDE^MV!+Tf%H)Ai9!cC;yFvjCd9H6}+F)W6Q8j%M^!v3FBTGl<gR$kGOz
zyMgsSD7TgEvT<2gu1^~e56&p=mOoHSbtGX>nxR`=yKzZU;??~t?v@9Z<(?2G?xvOV
ze8(JTWC(3DLRmXEhCtmKwRS5|I;_RUIM>n`42)&y8ug*1^z9BRn&uPd^8?fQ<dD4`
zq%j#m)4BKzoO}h2Kn-v|i;i^`+H8(KoQ!Dje=}=b=Y-hkX^?WVXjz}#$21X3IH2}D
z@_RX*2qA-yhNf)s#!2iD9L?Sv)qrX;Lh4~UVH<MH0cnbKMWma%YO-i-%Vx1II8C!L
z%2g2WvUlv)cm-xewewYi)ublDRD*w;>?T5b6a=Vo9@MRB&8gJof#Qh5KCgC3BYr4>
z9{YHR0h)~ttXm&20$s6;3{R1r=b4ZuZtY9HdcLw&n`y^rseRpj2ZngK)9$wCOn5s_
zeGhSO0wd#WUOWhG?fm}l{w`m=d&gh+_1|Dw?zlZZ@`Hc%d!U`$(-S&6+g6V6XR~tg
zU?2Q_vz>UC>4~Ekwe^FX`@B}d5T=yqj$Ggcj8#mAZt56x@5sBpL?|0y&1Kj6La^L9
z9ivxbOLQ@ghLKtYkwjNhD+o$!1?|eVZESU;mW{rcmAWJwJ6oe7hkB&@XOR%xr$ybv
z1KJ;>Z~o6;-!PlNBqlLLPAQuuqu+?GX1-F^VwI0OPpM*7xmI91wB*%cjAp<9M}<Ex
z=Jr46<auJsk?0Y}tqdbEN@h-ho*MZMB8mI+31sm6xYEA7QnqRjF6}Vepl^i+Xy1&z
zI@`=FsxgOqD-7+7VY8k&!Z`&VvayTT+m+|r;OYIwb#2_Fvq<ChWZ48i|M|~(eqL=d
zC1Nr*Hvk%~yBn6t{;Dc@nwTTZQ$TCys)Ilz$H+X}jIIGqO*>k8Cr5*r_Kg5lH^Q-(
zNfF+?f9Ce?iC_BV@A3Ip@7ems4}bI-PhVYlbvhZzF$9b1z1?`unJ?abMbpkh+bFjy
zKl<sH{N~^N2lR5Y4aObIY(S>R&$8~qxEZx|zWVAbK6&%TO^UiBSGKVU(Mg7$z!i3r
zBm2hPi^%3y#I1F9Q;y;q?9DaUy);TK^yY<;s>6PrQaiWnMk|dGlU2&Kv6X5aezS<}
z;j%MDSZ{^x*0?;cTpw4?%h@?AMn)@J<96BT%?)n0Mz4l;gb<nY%raS+)96MzLzUoY
z*mdnZUn=)^#fs1|2>V7g2G?~%0<5L_rf-|&JY^8&n!nwqL-TwhWpkkfp={N-6yEOc
zR<B{-==5be8dD@_ra9|u?>s*&NR0?+ovg%YZ>N*R0;bGZVnjkSpA<jD#bNV#o`AgT
z@x0Rx{)P+4IJVRfj6`VP!ai#VfrxeXX0{e>rX5UG6@op-J+lCZsUsY}vtT1Vi$jhc
zezT(zzSh}!4o)N!IgkMz{ewc?I!x3^z=*hGov)RhgQDi2aDaWPZgR6DkliH3(a=aE
zyA9%$os(dP-rC{$wlO=Q+?Hc7DY6&oIoW+nX}4G2k3awW;D=l9oN{1_Cd*|s-v9jP
z{FATV9kcS?!d$TZA;w?|D6yX<2-P+A-S3ILH#RCTGqZ6#XH`W~a#J-3Zn_jhw1V8n
zvb~v>zl}ldF0tI-(O?ORCJ{6>hslsl5o1V#5mck|RfY(~4h4VTl`zc5r~8?xW0+L^
zY%Z}9EilTGCI#&rXX?I&UO<6AK2rL(UEin}SQaDb1cJ>W{N>!?vfga=AF!%)?ER~g
zVfBLo#TKz00HH=&8c4Lcd5Hs!HglQh=q_ZYGmVmn5*Sfwxf)_HEj*k~7#ZoAuReeB
zLkl<7P?ww)Bq%`}qZWEpVhYUjL@$l%vlp9h3ONVvPBX9WESuonW#j$Tpz2O%ii0H%
z9>4mWciQ>kkH6&cvf90xEyN_I$fd0GQMlI1)+_48K!}0we&@`R1gV|J5Ew-X5%MV`
zQ(|MVQF*RcuBGs1K4T^vb`M84%3AyYx>AJy;y3>R|HqI26Hcmp@&1{we*Q?mHMX+x
ze7VqpOIe|Hq;<B-jkmq?{P@J@KmIBI`QQFK)a%VdW1^GoUb>(I@!4T0-+1R~GO1o0
zBoVSiCV?#0nGgRHb`FIvPIH3mv|?sN!52F&RqDM{YNM%`SAXf0tx}6IbhQunaj2|Y
z<9aKMfoq-zqjn>%Iw@6$k#2GlatOo@^;Wq&tz4fg^^Gx`hkHY8>&EqYwL^L+ZEcjR
zWhfAoR*jT6I;Nf(_H$Zkrc@}M(lIO7eSD^R5R+PKUN9?Bt#d&=B@JU#f19tg^NL!-
zq<ZEjGDfpz&<~*f+7J00&0jqRgPDi#1E40)gd;R6%HA<~7>twsYPE#X;AAq~;Aj6i
zIp@QImk|8fH9Mp~U}6tveMgt)S+i*9P>jg}wx;QT?~T0X1UA{xEoYjGArMwic$eu1
zq4t0V&$^!`3s0*SuoV1!jNXCA5F912f{G}F;qQvtny@&8@KP|1!O(hej)L<oj*&z6
zVulye-{{>683_z|IbinAbh}<$x)=ypW&{K)_|*{7J_aJ`DBkza+-j!}rFI*+E>mXC
z0YR<P77MAXuD|zNX77O7GmI&b0|Y<(1W1}aN5t6v`|l^a&w$$ZPtlsa(v{h&Ak@_y
zF`MclSx1V5MDPp*dZhJsAly5bMZJ>Ofq71}X2=MFNF?V8JHhu%5<>&~K8g}o0C>^3
zJ3996&9x&Ii1AoNO$8SEK!TCiEfCBanU}flpRh*1=2`~e;f=9=wVCy+N!z?j9cEvv
zuC0{45b&7IqjIq&=<ljQ>8*DtCW*PqjkPx79GQ}kqp@IPawsMymb}oyU^FA@oKG`p
z$rhKqr#ZQSgBZ?3!{A<%y9hl9P0jW=#9&EPT4~*??KowoS;%u{T8w-c`hcYq`(Rpx
zPwr=a@sn4)o)bU)>MOo_eCDzhnwlaY=R}$kZ8X-}sk`f3g34fC5+A+3V-n^0=}HKJ
zE=pb|rt^$W5sgAzGAGX5%Es%1T(WmWn3EZmK!RYM`s34u|MUOlKjqU;KH}APKc&~<
z_Qxwv*GEcMZoTvRbmpVwM3BJSFW&Jl{>AU}&;EyhMjYMQs{8!Uk%!*w0NZC=$U$5Y
z>|h<<JxEiedYFt_D9Lbgr8#!la~fU<;4}I_IH2BLsZvX2y%jWPFD!?507{M~i|jtv
zzpaJqb)&nbvMR_N7-g8ebsL!cHt_1s@&*<Hx5pcgZ!cVL1)K-bTA|g#w%#bExMeF`
zFDq}~KJm%Zg^&X0wHWljRYNsOr<O`tt2+vIx4kuzZExc!+>9@y$9((+mhN8Pef{eG
zPwS;V3NlyEKy~hs>9I_A-+Q<BYwVHZ;z|zl?T{F0Vi^w_lo)OQ$T^c!ay|K8v>M7L
zo_#Pz^D*5HOTWyxU;g(n;<K1VE{e^|{orV|8Gj%e)^27)o5A0%TA1e9)^TeHMqV~3
z*hqr}OgZ7*G8p^2Ra3*jKvRO;iNhh^9cfJFwP9##qG_TvBlK0Z;wYs^kjV<@W-?m=
zM+f$zL_KxZt&rg$zwI4#iisfOP+p+MagZJTz|EL^Uj1}4X#chqrfhP~fF1Uu@0P#*
zbHrr902(cLtvdP8?7{O9LWGDLG1*}#F`FD&U(k#>+2J^b6^wWz5gbK|Vg;nVdoQ-g
z0_13!4C+Ouy&k)KPhJXC@8pwY&;+e9iW)*BxCt6CUQCkAr^WabTd8EL%1k~g0UXUO
z&5{=nb0r`5L=Qt!^>L;bReK|Cg~3jo-A7&?Ml}p2tEC|1ueC<MQ8Do9qZx)!Z^GoJ
zG`rNIwZU2&TWi=PK)jL?UpUUd`~cQMZI&OS5ryit(CT0w1JOv0%KJK)PLXMjOeuI$
zV6b_JxW%i>Svn@!tF6*U^W;I(cy|eC6neyFG{fK<Q6X!vjUqdTpjm}Fh?D@8K<S+e
zgOY`KpUILrzq(`j&VyC|w+mgO&+TlMr~(>0DrROvu$kQ;6Q{r;iID96nID6>I^Hd_
zr35B3`;Y{uyTr9$Ddl<?lh|tuqgO^9Fqze^s8y?@RX+RK&-qtB{3%Qezx?OE#}HV%
z1q)?Kd~%-oXqovZzx^-yZ~nV~%yzwz-TcR#<mF`nbx#G;;VED)&vNE-PKmpmxb<LC
z$+c1@b8ggDDVte6&%GJq;0J`|S(sca0_Qw&UNSier8!bqO->lo<opt)^+pQ8;9s9-
zwzX5QmD{b->oC;`LhQn5gHe@jwQ)hpk!7C9Qy?nb9&dd4^Ji|?4b&jPvK7i!%uuN}
zQVvjrw@)kQub%nz-6Qj9;eIVfK3uC=<$AZ<N~1S}!v*1X+YnKnuf=BHJ{I3+=zVxn
zVUN@{po0*o<J;{=+*(IUHIIOl%|2N68M3W^>rOrOuA9k?P)+V81BrU1@CD2_LsB^A
z(Ou2BC<vsOUvg1gz7|6`{N{n<P^h)}oWBu5CTa36&ydUgucerYP{;`h9c`W4^>Ucs
zgzUytBQQpA38L7%UHUO$^P1NWicY%~D;&#4%1V?4%?VX|XFC>o-n9{Bvi&kdAp6gF
z8QMj4hu2m?&4RqOeca+Rb4`|IB6<|nDcJEz@YK(g&Aqc5eSp3b5}j0s{c3`;UZ2^D
z$!Kq-9z@9fo~ON&-O+^c!VBqruofF7EXl~z!<&SC+_3dwmhTLttCKIW-&h14FG-8h
zJs@^2M2I$0Aw(n#v<7|HJE5j#@T186O9rL{;Xpg#1*r<ryetBNNhViTjHA)jeP32P
zOQHz1DnfN4yhHgr;uMAzUD^*d#!hZkEPM(LAN<Z7je#{&4!6rbW;>pa6_?vamVXD*
z!qclWp<7mc>Q;B}?oM5mZFKkPw~q4Aobtkc|6?7Fz<f7zdRXYCbKN$A4swk!S>@Mj
zGi~}g+t2JP^EBu>*tQKF5M_`{X4^I{?;ly8uZKg`k|R^fhj#$c7@u2fbYBdub;LYv
z7(pn#Lu*7CJdDUWM5@4=5^ZXvDKe+Xbhn_>%&LQ@A*{pRhA2Ld8pDWlQ(!tLrsS!B
zy)#dNkt`6jmd0(Z1QlMLXYTH2;$$(d8kMFU4RF)OV_A_8@cCnd8OB!EV4G<;IstR6
zdVc?&AO1mMnF6mL&gf9444?n(Gk){$`~!aQM?VGaR%LrUtV(!+y}+toS2hsi2Xoqr
zjLxzwbg{3O0hjB_C7YUn+PGe?lx?F{>##=`w0G}xloTRwP7B|EbH^`!@`}$t|H^nw
z{^Ly|nlKA>Lnvk^Y3(YGZ7W=!H`cWn49x(I7WZ5lx9dtP722>Y`YDrA<obN$%ksnz
ze)2inZKK`_+jC*NtUO;YeD&2M>t$ni7FkQ>`Lc3-yn+VS>qfa%>f!+^73gAq<K5lT
z2&}c!t{b<_GZ3&%ZmYgIy6asZMi!exwE6MZG!FUGdeu^F=9)TLg*v*0Xo>f5zL~W;
zXr&^Gbva#uK-!VfV@t8v*PX$Qcet%L_ZARC=2G+~M33<rx{qj_@1p*FgBoNkgB}`q
ze>oiWEPl|1;e^JmuwI{y)RhwRvXHVl)CJ_QnccYyLp^F(!?CH4-l?^4yWZ%%0Fh~)
z4H=2aitd9<=+ZTpGHOnw+H8~*&GB!F!N#@X`*7WhQS<P}QMVX`fu%HdFDPO#v$MS;
z4GLg~o%swz%Lc%)NfDBE{`(*NfH9OD0#`LT+%bp8ZsTsUWGJSc1EqDE3T>F<W=@F^
zqTe^A3>)*-rY@~@5B~k|#zq{&xf3G72$tIDZ8|c<x^|Ic{}<Wa`G&Lb)s2_zXZ*27
z77nRi_yAc!cP!NRVgFrU4r+B|!``jYJKY^DC8%)-hMEamZ)j_1?`DhcY7X!Q8hgHS
zoTUNPZX-7zgYWXh-M0l1v;GbnF)d$fjeawqfi$7R<{wMUjtm%CVH=HWt+atjf@Dv(
z3kM<P1qN|e?L09*EaW?L_u3Etwr;eHlW~Mm24&q?^Wb*sOiN&zGo^U6X{{{NMA;ha
z^F}ZB;9?L%6Ozwho072E;xx_Nx-r)KXoP6wn(fwEubVB*5h*%T1x8BN?sc2f_aId1
zJoirBHh%WF@_4g@z+ge~w>6MXGxPmyR;SI1o6~t>v`!M?u~u$tV~UZ7)66_YI*m3e
zGBBtPr3QE|S8BN(WD&$5%033IRnnBq)huNY3mG{r$r2M4e)NYw<MSVXX`>xA%i!9q
zQ_ndgz0zv2b&+g=l=bH6Ub{_h-d7v;3fgXZe?F6^Oo)TmWL{5+`y6StvnkZ;l~z{z
zcBMu@trCtoBMNtU<}^C>Vd0aHADEZStqoMnQhLgP`@4zLk_a&{H%k(X!R}GpI@jxk
zr>B+Mt&oys8w?FdG6L1*a%Ei$L97e0j=;PIe)9H-&)+^7s_|5Kdbe_USb2WF@buPX
zfVB^r+Mm&twsuNxY?n$|8|7AQuG*BcRxWFyRD)Wh2))^%(Caw5nW0d}aK}SC?S?i)
zXEFilx*LvtYp>B79ox{lkGc-swjY{1*Q8c+8|>Z2ZKEknei*vx&T;eL;2+rLZmNQv
zwQe(KHK?`QELL29zHjH=;`-b-92>X<`|ozW`ri29(D<m8(Hdx{wZb%}7w?aQPqh=)
z_D%Zb5b1l4z}Uwloq)N`#h4HueN5BD;)X;!x@s20-r+ibAh6r(;#&S>To$)m?qe8q
z?e{&RjQx=0<&Zmme*z-W+#Qpzdq*XQBPLH@e97mZ{me*ib7m^dvj~isxc~j|f@`RH
zA|P&&Y(+wh2?>^Gq6*bfGxM+6M{4Gl8oUWn)x(_2YUqF-eh&M=Y$QFCX_}mJ_>+e~
zS5tGuAUvEhrzAXMq-{AkmNtR}@9N_VadWuef;E0lY81;b2>~J<(P~9dSDFeV80M<o
z_$)?(9_XmFY6QhnD|`Mz5GP<euSJI(xf%y~RI%U6`726s?n|M_?Q5;kZ&VD_5uiaD
zHb3dO5B0!e7uVu5D0{(n>qf7YkdWx3p%#KoTr$h+6Q>93?&`I&y1%_d=+g)03%zr@
z6s}KKmQ&{L;pD+lN*RMSt(2{EecsTfFx=75XU{nXUQH99o=(IRdD=G0)!^dYKm-xi
zt0f5b(OL3xgwm+QqYH1{nBb)qo;K5q|Ln_Wu4}<1^A5WKqU0$t&*q}l!_J<@yINz|
zmu9TiWtz!Tba+cajgVQ~7-SoklWp=sLjf!mYaDZ;oEI<1_7Yj$Nh|`v6Z8fQtKj22
zJGxLE4KPiA2+^XO*Ofj7DH^lZ^vb?rxVg)eGJodNkEz4#s=q%8AHO;AFelzWUb&P)
zD_1BtVx6HVV^~sPjQ)&d&?9!gro`R-#B-^X*16S6m&P&&&gTh5nVKzp#WvH@8tZ!F
z@%fojD&IMu`RL6|A+TyN^5ix0bh#3vFoo=rEwyZj>&ES7GuBZDm-km5@1J=0?vdwr
zSJqpl)NWbg)hhFHE5s>tyR9Y#S_@Hxa=meRzH;4)No~hqh{4;UK3~)q+3UdOz2`X`
z9Gqp2oaT(SmznFg^&0B)T+Cc%mt*ZS`#l$8M|gC+NWScl_rqi;y(t{Xvq=ms5T_qR
zJrUC>$k@H&pi>~uh(u)9>mi46kPbNxSNk{4eAft=9c~{k87tMa`7v7bcXMC*4-Zyw
zYhFdfcvfmkNh=0?Be&6qg;|BTu-u)PmI+*1IL|ZFG!Z{Qhit@w&GJ7aAI9K;TfWao
zq^ZePr8cBnXqRO@SZAF=vXXx10*RRs8O(37^N2Y=nqgF=7kl4->!1BLwN#G}H`i6`
zBLXp+V{3PbuKEUQ&+4VR71!G<h7$C?7Z1B`N)C^S-7<M+rSDHl$6z-@!$Xhl<Z<+E
z2<RZ?$>q!#(UC|~CFjWfEJQIUNn_)W1BMUsK|FeRu_gxGSjh@0b2~IYfFweop=}@`
zA||I}4>B6XNR5Wq82`zlxSM$f=!;9{K58F2%$9h+C(eztpd=%H4mE_Uh2ZO>eXTY6
zjf#Qi%SPB54`gFFr|5c4EJ5pXrPcOA(%BiKB5s&rvzvfIjO6pge19S>M()Zv8Z)*I
z%4MT%)fRmSX3n9Fe!zqthpbj%-4M#HvEGW!a0bWX8N;%@-z*cabAp_xtuqbGQ@)$T
zr>o@)?DH=pku}z`p(EKL(6AW}LhYTmmuJ5G@{#M~m2;Z7zn=(ExIfvUE<uP<Ea6NX
z)-sHZyF4?eKnh|JxXn}(A-N~`{?ZQ6(y60z74va7Z}4<f%p!r54L04okE0M!HCD8h
z_lS1)$+nxh7i;P*K}?fZol12AUUdDjxQDupAe^8j!k_=ePx;<Q_dIWKzB}>JtBKbS
z3pV*u;&Wlt&a@WJy^(vh1)zf{gNTKB^s+HVA!qpHqZ5zUjqBseQ!Q*VkYq+ReW=}E
z31}E}t!!=M{pFE&&re8z`}@S}H#4gS>TR%=jpsh7w;PFJmdt8Ur1^x~Cc7U}GN7)7
z=f^9T=PO$&Y_+p>Gr}4wJYF}#l%RInR;aftr4OFoJ@W4Th1=Cgg9<cNreMNZQ)$>n
zXv;!4oicZ)$eaw-I-gFwIwzhUSKk1>y;n@>rL*<!aJPxwyPHEV@2++)T|6)I=C1TM
ztU-<L7vPnWvTw{Bh{N8g%dt*O29~T}u=husJ%?k0X`_h{l6l`dccx={d-cNH(1L{M
zcu~K6fPRh-!e;~ztVw2-+lD*j`7wYKnPaq($h>456NGelCxD?Uf)Uedb+3vYc^%Hj
z!Vj|$E#zkBJ&bP5b8w{3Q3|oa#v$qa5XH5dXIz4mn5PLri8*t9zVP{HKclzCl*|X^
z;XL_aXU0N3*x?<Mla1N^5;(s10c-o2+IiuZIk<|d@bZAv4}1ST^hG7mTg3@#)=cfs
zQ^iI_BM|fC#ektHd5Xp#%EFYb2{Uho@xfSIjntQ-b2eT~n5-~yI6p-4l$ldvioj^*
zu~W>Qa;Op6&2iU80y>O);e&BEH`2r55dS?5-RXowUF5>tU}N|lY1?@?LCpInjF-ae
z-*$a7)1ewkD&&}$Q#4`-w%}H0aktu-CjYj&lhEvaN;wgA(1RjVWL_-SAYg$n%Xy~V
zD&@BM%vA|dJo8+Upq69MElzhQRkhlTZMv!l+;lQd91%cl6V4$p2=r=%E$NUuEXg8d
zueI}j82dA);DngMcviY3Zrv=G5uuA2#c1n1T{dI{KDj${iWbZK@vAeZyV)JqoL!1g
zT4SpfL_Cw)kd?l5O6zPiBSr0y<bYNh^|n#gjpw2KrN8)R`3L{_cPOn}l^m=v+vlTJ
z4SWdw4R*PT&FwmjhH+DaFHDyaNrY&2xLYZf?&b4XRYMr|%~}NGKjh5c`fGoQS}Jv{
zyndMZ-Y54g%fz)+^eXg8ji<40<X&0EizXPg&2`;umR$SBw7{E>7UnMl?Oo@~=O-Sw
zd$ztYlNdoED#)OYjamxV^~#q|?|AqALLY;>bLKm*PCS;*(|V=cE`(*~lT)O{6KxD`
zWjFr_%qeini58S8M&=OEVg$mqbfN?vmdO2L`53RxnZ%%OE3K?->y>K~%2s&${ymS+
zS9<AYfHdqNx}OtoCR566quKj}5+l5RnE38T3#Z9o-+7uir@;H?W>D$3_zLlbJdAQM
z?sqNr=#4(Q%inhb*M3krRO?QbL7k~SY(KF5CAfqRBfkj;BVBzT+L`P|IL!3!yj8Ws
zMp61`$ViBI9)%7s1oY@^dus>8FA@&v;GWECMaZByPeRm!(Dvh(5Y3WSEWpUZifo?<
z46+PPDG;WNNIIZ)fHurSK$WA529vgF3^s0Yq(lTH1@g%3DuMlQIV6)ixeoi?#EY2d
z=<beO;i|7Z8aYNZ1m3@UOW8IvM-$NANkNgE4H0POyN^VYYw*YDey`;i_3Q-+vfKaK
zxiWMwa`fd42_`ud_Y&wX2gjwJ?4u$~>bDnB(XAb#4n(_W09bNnim)sbf8iHD=F{*0
zdB*f5Kl<W5W3&Sq*yX1o1P_4n&n~+H!NwvvnuFvrWs(d>sZCZ{yQu=ko)hCEuU+{T
z)S4p^Lz9ePyYpR?trc>Mq$y#t#EgpF2le7o)lwv*STP`@G3wVdEc<45#KqF5(kU>f
zOpd!Jcjb25*tY625P6CAbXl5W4@prAJeo68j-;;SCTO$yS6-6IY^u%T_E}xnUx?tE
zPG^_0zeukA9$Q7mAo>d^A&_(6yiBw{`0BFqTq;l7%6eN#49+<p<fAdHJCAN^aw(1H
z)@YJI(jj%~eV`iIZYz&ZE2A_%zB}{D-I<&tAHBL`Ua~0&@L1E)xUDPeR;(KprHszH
zRcbMV5J?bo^lrfXylZW2Yhiu9v9`wfEaV77qq#6Pn-PZS3wkdEe1Y%87=r-@+AHo<
z!3UNf;I%0s5K<<MVa17aK!O+?hkxJw`~2_zxBouh|L#Zp@TYGXaV5=>I0agUA_8|a
z)BQp(g-fV(9p*{ysKqFQE`zSHp;S@Ql332>9=8{p9cZIcdNG0&ok(CSH^vw|zklTM
z%lACJzabjQQ8+~*%!#&S?$Y4nlknzlrk2E4myK2{Q3A`9Ii<j>B~gNKN`b5$DI0NA
z9<uW0lY2fsPs~$bPX5`V<xRZFk=z>ndSh%GschV%ybnt49yP1*DhGb)erAfoSC{pe
zfi5xf-6iqoU!Rz#K$<3UjI^zqq_rJw>$mcXu09zb=$}Od-1gaK=l(nTCOGBf2fKwv
z4egE&cY?TGY%qi%xXd+0A*D=?7BD4halb0oDXSlnwm+%6<-K%yI~Ae3K~IdzkR94J
zBP^TiHgq&c&Ek>53GN8jY!4g5h<n{P{=H-C>^=v9T}ByIkyeN_lC#2vJwn$}D<8WK
zzh`|vOe8=Y2hyP!6=Gk{(Tl3RS+MV;N3kX9Mp$kE4Y;3!?XTU7o1uecUPwXs@ehA!
zg^rltLJaC7if;D<JKS%zaK9{vxzT}M*d2HwN`BCV?+3Xz`3FHS1#wv2_pT6J8D?hz
zxH5}jqk$kyK{!VNQHD0MWJ48VWDF}j=bX9E@aCL&xSL6ji5*?vM+19q#7+|3&82!z
zhv8H7k%z@Z4<BJ#lsm3GsKfRDUNqv()|3K08;SDPs*MxH5e>g@wN<7(nYFdc`_)aC
zb`HVe!gz3Uz8<Rhjf#PGtDJL0hQTP=JJD^JL*7o7u{qhx#6}%PYX|hf@p6u>n;c6A
zQ%cDFV#JcC!f2Mjwa)-i@$7GmIHET13n4)6Y;4AWRUb<@iLt6uTjlX`;nF(qZ<ZM_
z`ewNWL77M5c~frHa{xprRng6K#Z3kwDU(>}he3Ufr}rzjOQl>^ZXvQRnKXF>adL*M
zC|fPmp*%icc)F~NR#`=8Yvqe~7g}kY^ThpmX3CkAcTe%g4sdi0)NSyS^*w*~mp<Y5
zfAASXAmn5~rtYo}`!T<d3gq}FjE)W)HS92)<EsbZ#x?**d15{(LZq%YV>IV%e35Pr
za?3LFXMgq2@F;_4>D&;i4%TRktmnw`>W=RXbM0G}8JQE}+=@QTNoxr1{V?dxo}O~@
zv@?M&Oud=^L95pFskwE9KA?6^Tj8O1zPDs#NqqN|c!-g^2&adMH+M6i-k-ReCoZAz
zc$0(wFr~mJug)atXpDUC^}_2@WQp+cdE)f_H=I)9^)j(cmO^LU*<kIBJPFg-h_#b?
zMN*|Lfu|h#T+O&e1kQ8fJ`0aE9GmeNfsFDh2EMz@%!?@;RN&G&(rxDYZN7px2SYr<
zde7Eyu8zg{nyM#BB#NOWF`7Cj2HULS=!BtEKTu-uId)3N2AEPl4ms4<PB%;NsCMw#
z_|d#F%$ctHJlEYZ$1rXBFiy(wL9&?wS*NW2y)6kL`rO(NE^*0ZFnxaQgA~kU$;gDm
zK{ju8qK^vNjkvvIw0j?wn0!)_Md6B*Fo)WG+Z6_fH$yPqgO1^jgUUhN^L;UdU@`Pw
z_`E2eT$dlCC73#DJH~)!UI1+vxi|{^{O3Q%I2Yn5ZnOZnK3u^7IR-AJ@wDDJ&->WW
z?#bTK?}R)7)Rhu1yp5M4&ANQIYF0}mRP3ywsUt(e2tr3J*;TE>PA)HPL!q0@Z;u6T
zz4QM0%5VMl@AJ{8-(!|Q7TCklO_3E2%l!S49i_mYbHMym#8Hib5gxU5TCdb?Guv^7
z0Yoy~ILn3}ISO%E>@#lrXaqk;hZ84VdMht`_5PXE$1m(8TYuNbQDptwuWwWgthdVD
zfRO#AaoNLGDy!#PXP>F~1p0uxyH_n%4l1!mBMKt!B4sega(6~!LgPxmnnPY-n1m=r
zgS}1*dCB`I3s7#ig6!4qK7SGDD9^R>^UIA>pLxD*Jg$xA^O<?FE}`gP>q6~;+B!*K
z?e5zxiP~Va!kCmfTe?(lgKdL#?Uc3imK*aF$x~!tEZF38%Qh;PweiKfD__04vLrjm
z*2~71?=GNbX}rwlFR&y1yLY-c!Dj8s<7MUh-}{sw{p2gwx+jH=m!iq%^uFWmudY|3
zJ0c)^4Z!in?U#2+L4<i(oUk+brj<dAkO|7K{N-OJee#-jUwp;8F?b%G8=X}fF-{DG
zB~1LnN3W1ld7U$-DSGi@IpU($<;4JLz#JO)De@`?9wJ7>X`R#PJkUAA;XW4>vM<=X
zn7RAt6@NC{VesnVjx;5PLQ9EdnK&hz+0zQGt_CMZtE^w&&zw_dUM9Zt`pjK~DJt=P
z;X8LTIS1w(nLZ@G_0HW9BvnXi8TO|s(3Z%Y0Dl^0VLI3(aBaIysdWZXm05&5loOFj
z105rlN_$~8e_O932Oyf?#y;P)5m$Oq6sW}p)9ayC7xyv^*nGf6F2+TP(Nr+u!%Tub
zEN4SoaWpHk`wFOs<Gid5|9oO|`>_)X<%2H#1E}6Uj3AgaGE50c@rCoSznfkR>@l>5
z)XwL-Q3SMFajxBie3Vh1=<7R+c@7IncN^b?#Ar}E95d|D;>%}rqHu43n;EIQ(1|Kp
zBhz#`h-rJ_q94NE(G*HCM(fze05N3FTqMnG>HXWcL@|k=G2)fvMmolqc$s0_``?&6
zJ+M3O|H6{z<(~QHI6{FOg5O;^t^4Igq2T{}_ZUe2ka`sA;oo5~>QTg`)3ri0o<eDZ
zB=Ecr*2{&{>47ML#s}{yFGb!?K-~Aq{gwFeZ|uGoq}jU`5OW<Jwl_=M|BIJ*C4@+v
zvXMQLu+_m<3W&Loif3C4Z~6{$;o$fj>V)nC@sa&m{K?lh;|#P0DcQz!5MazfskDxU
zz%)<hx>qZBPBdaEbE`*42En3gWu=u)aJVkd6B32xJfk9PDBE=-gv5NxoX!)|X@Q(z
z@|RUDMCNwuezPs&F$6)PFl4Z;m0N2}Q!(jc2&84Qyz(g2kP!`Dr$oZAbh@%BX;R{3
z#1QQkqua&u>02L^GE6!K)UMpNMr)m|RT|2iqIE$cv|4!k{)xAb7haztw7}c9H!hF%
z!1El)lR=+dcgIBcz0Nih9AfH;i^3-#zvjn3{lecVW7iH<q<2vFu-@NL*#`o86c)$J
zxWh=!c7=Tn`e;Cc*`oT)Ow_ulG*A5df9tQql-YvC7T@}yVa)F6*#v1QOAIXMg@>3p
zXS0-Mm_LPzITsRy3FVX%rxDpiI4M>zbR{+AjPN=nhY4}kxg8MquQDGmuNc43j#*r(
z#R^d$X=&M5Deg6(4$)fvMHaZ5?l>)(B@3-p&_PHEjnTaE9YKlRHoxB5OYE*xH_9<B
zI;ynReWYY}ymo^KAF}ZMyNNzbPyXGznb$cHJA|glXcm3FHS;ijX-K}6SMV-2&e3Ff
zI_$lk<|90&>p(Fp;4#`Uug(H82BMWP8j%huLdxWMCe4dmIrp9aW?j2F3N}DuvNhNn
zXk-l84d95dbIgv$&o<TY!Cc$EZ;TPm_RgZFu!$*~gg8aB#jdV}mk$q<mle0i(3&jG
zMr}cd5siaU3fZ0+=}e7AQcV6Cp_`?;ykrw#4I}%PC{}=2il^O-F-(<YPJWdz?7cq<
zm=CU(R6o4I3P=>BIg%*e#R{}uokOwbBN$mw71q`$<K_9c-DpS+9r#dM%2C9Lk2-hv
z!f_M;_}S5c+Dr~OhB@=u9afUSG9?6)@xhCX(Z);G#MH7eq7u@?P+@J2pM3eutxtUR
z_(W^<fZ5F#fj?Fh=&<g;Rx@@I{|r`e>}st}biPW;nLG`%4DR;3xT5vA55dXiAtgpL
zQ#Hk!=ntJTRqbboUFON~{qE(@&cS$jEB@r`8x;dF2mv#fz{xUwD64C!5g|^AykzTC
zhtKCyH0Z-6C^4sF#!)uYV8-aWTb)RACPqlJ<&Z1N^l;+Un|ty+v9&?GJDEX9Z&bZf
z`<2m!(TpvdlFbaqunxUc<>n-?l+7V)IiD<g(nlXuLoQ_4VOF+AOoJuZ=aZ(?Q4J0f
z?`E3mRI?1G<R0lE8ZTn9&=O;Fc3Qcyt`|yKY17QLDNoNUZL@T-X_+BHX_a-`oOBfq
zxKG2d&mIQvFE{?&FTUm{KmFV{zYe^34y@NFmh+jE^2__6J3Dsg6!<F9okiuI?y4W`
ziUX5kIea%Nj7C%V)nENpKK=NO$86e+v~HE&D$*LcTgMWCB_z%%aSoXrETWnKBPwZ!
zFi|F;Do9hpHmF4nmhGKdI$8%=EE_|{z9}nN0y2S4O5o%RzH+;X@ikgIw7Io4wyjWW
zqxXrB?O^EY5Vfl%s`sQh*hmUBMsE$4EwSr(>9)~oh^PnJj8?g>m0NN0AT}DCQ{wfK
z`1o$(yZ4#i3dG3$oO!)u55U<|=)@t$eq^S9o3Gt2I3$b2m6({O31lbsWqL3^hy)Am
z3A=HbHdHIPiBAe*(0MWf9y;98cHg8sVN{#LhhXoTv%XD=R~gblUkG4nhuqZ;RI{hG
z15<eET7hF7g&3V<lYCBY1~TCGv0j<TFs-`4%iMo-H=U`Nt#)9VT%pygp&B$x9~@pl
zX*^i|7LROW4h?4Uyd#2p;s3{Nto;xh-Sh6<mRDYu{Emd!_{J4sAB+$C7)TtH^=f;A
zAJT)J%xGWQhq87v3$mH^{@+^xF8Y!w;llufxrxpd*pV1LDlri;2T9+qMl&Qt!JKEi
za>|k32XnS0*n_8_``o>oTy%6sUs2!Nqzk|MgP-v8%ivm#2hx3<x__=0m599$t?#3l
z;dv2$kLm1<zanNDHO(XyW`$8U``(Tg885<>cR?UVq6}nf-q{c54eTRej(5NS-w*tI
zkb~ryJkq#+%@N}_GabTacqs@Bb9?LPk?G$3rb{qS|2B*i7;_}`VX>eB^Jyl`2_54g
z9YR1iGhC3h+C$!DLt+L*#9*|WWQY6+10+&axZMh)xam($$Gq9*e%*@|bGbsS%Qq6A
z!7<VsTPZf92Xj|cla|d>!e(QTf)kVMwu3@Yt7=n>EZJlwDJC<}=w_Z$#>Qp4v2Nyj
zzpWc(V1D~!FijJ4GEeX3fjgxdYqhmzFL6+FX!Jw$bedUm;Q4ZMVx>@Pv$<4j#Im^y
zq|LUB<YIv%X7ExCnvWm2+kF6i?Xakb!TG4oUAvfK;(z(qeuc0WQmy3LiDiSW5^sfv
z0dK@2J)>A5F$aT*lNd=xhc6<ms~ck=y(6vB*1~oz+%|*V*V=fxZj|ea-WMd#wm}S&
z{NO&|Iyyu2ASuO^j&8>I0VXSJY^2+T*191NZ7`+4JVmA{v(F;K-W=-QnOH$ws>^@6
zedc`(GcZ0&ucjI)t-FtnKrrRX$M*{l=h?`i&PtCda>~ve?%mumk6@>7zt4cpJ$?M-
z*0;gkKoQ%(o#tcRRfq&Sdo6Ihg8}0WiFNwyo8&M9m=HZcNG(xrxKBs$4r|b81B$w&
z4|m>+k#11Li@sY2j2AV-PR#nS>oTQ{{4t)ciX&_a;>5|`I-Q0L&shjIjtD40lxbR6
z#3Z#TI73{`aBEi@43y9TF?r+gdH4>g*IEzBAV6zIbQD*zm<q#-{@!TLN3!C_EYs~!
z_wTLNScsa{^D$--<&S>w1M0TfL5LCh_Kv@H_|O!`$aYo`?2Py5-uwGK-0$z+1=}sQ
zeRcF<zBK??nqPFaNgqbA?sgQog|QypyToATN`4QT=Y3=&EYW&J(gZ=c*3Re88xj*~
zn$YSQ6~li1PWF2#l3eZ3_26}-!?()Cj~b9PAq=Kq=T{kxwpB+xocxD1O<=8B4#soX
zf9u+s@$6#sG~OKovr*+fn%c?UB1%dY!_HQ|)*Ah0YJ%MeB-lf4s<hF$wMq*RrbNSF
z^JqgHx|#wk#7IoR+~lG#d$E9+6l<u`tD516H0#(MJ&+)*TO(WxA)EYY$i``z2@=^Z
zh3;@_5MfHu$cP@T+`KasFwgU$T%&b<AxGvIQHQ%?N;ZS^p_F1yaw37WByx@itTS5o
zfEn8f%t$D3nkFQei`H#380lJXynVV-w_*hXKoz#<&Zq;Kg`6ViIkTi_a9C?cH_EUo
zJn3MHurm1JmtWdAp%sJv1AOuRVm6*bnU~4TMdJRL^?*`qsjTbG3YdAaYFVrRQuN>w
zjL{ML5!E?Jl&(CSCVt^ObFLk_QipJEgP%Qb=wo3n%7eIc)Qf?jVo{yyMowd3mX)@!
zx@Wt0C*4GZPfnTd-Dfrxwp(L;ys@5cjL8;th!&;VvFdqWVCr2(2$7rvdD@-G4Dt{2
z#=92yVtc0bPRfC0&YUL8B`=<zkYdCIv{qom$W|>0jow&dAccgCVH<l0K;){h_D*rj
z(3lcPH^eK#k`tP7^@=@{-3s=l@^`lI(V2a_uMazz7_6&LDKS*2)k2|m$U3G-M0}3!
zFgmu6q-nMU$f52_p$ru>wt+)7CxUOLPN?!ABAck0&qH*Z-`(!l_Impay_@1;KeRsN
z5FEeL7+{ek9@^7CaO-?2s@2WGdhZM|C^#l7mZvE(3oJpI<8WD-t+(jSM{;%1-X-q}
zjNZC^-_|MXW@cggA?rT;@PTO#ZcemYGlLT;^YM&&xAnfq{Td<L4q+Sd$O~)T;cZd#
zOF=9tw7L3ge|2{_d>>gJYBPm>gc6-xj@NjHgY5^OD|q3yqZq{#4-Gn3*481~UJ832
zg)2E?Kzg8!-ThM;trC(M!d<Rc{?!kD#Gm;Ke*uY+poR(=QBe+bB4B$-43Rl?^M#3#
z9A1Kwtl;W?>ueM*A&}gFsP{qJnz>x!#!uqoDK9`R`r2;DaDc>;3wO1^i|o_P=kzF6
zV~oztwX^i=T;He|$SNPqu52SKwbHc{+$0Aq>sh^CJRp=Qk;G>1quF1@wfrGQ&y9bP
zMrc>UyjMk?_m39~xj9Wvq7d3;BZSCy-O$=CwJYrDUWp+>ZN}oa8|XKaj$SruSHhfO
z%0wBYIU8!>q>E9<0lSO=Q3FCU(n=q+;wgNpoKob?-OTIzg=Ml?&hu^KQU^^s*KOr`
zxe{yVJ_MpFkIx(1wYXV`vFq=q$T>waVoX#uc(K>&#HT?Vl^kFd;b|=f<NFE{3FKhp
zw6bmH2a#XA?RD73RBPpOEwrJeaN|3#7M9au8-y<`*ehe>M5O6_JfmOx#b00<LS73|
z2M+`8s_=LzeDVIqcG=*K*`E%Nv$oF6y5#P#=f1%9#lAC}BY}@jCw~1G-te6{bFH0+
zB~jMK<>^9SY^I#2ncS@+R_%xBh}m+w%UulM<TZ8Eq;25*j2COq)0{akwzHRR#vnQR
z<`cZxtAn*!;dlygPMH*KEQXuRwB9JC9-Z1w#*8r$2{vlz{fPAi+#|xf8t<T+8f$*v
z{J+iD(Ba6;9)yz)x3tzfxAn^Hb}>jf1Wr>X1|_PE*@7Cj5Oya&KXiN9pcHnA9vHeN
zQJr+T7q*UUu@m}UBw)KC((b3f7p43Bb?;_(pMv9e8skXS+~48W-Fn&`<&4{6qoV5k
zk&ckC@HHWr-$@j*2r;O!%;iP5o<cZCXuBBrkGT%+MON#S>&*_^5RZ}H;h2XVvZLF<
zf!~)9&1KMHnUnd9xc`H>B+6lI<G(pS`SFk0E2|fv#Pb+X%Fx4GL&OeeTchSTqyDuz
zcC^Sw&->xMch;sv!yA<sPltaU{#LAMx($UKn3D6;Tvln%QIN?9?`<fh4c6X?B<}Cd
zh$#Q$AO9mh`K4bnSu5tYs%_YlOpzCS!q9~!&BhB0xQS3Z*1Q5cw5aL|f|OkCVu(kr
zok=luhmRyoC}J{F8+BJ~8&H?3haJx0gB!pepmZoPT#5Ce+vC6O`bNb-638C<v!^%p
zDy(j-($p+$g+ZJh@~+CLm4!^65-HkZyIgNCta-&<$>ivSdQ&9W3E65#Zv`zK>Good
zKDd-h3ejA+uIAu3c~O#LWYx~v_DX$c_gWV)Fc-fm21YdZz9moOc_w<^`6!JztXQd|
zTlDO(`CcuR?Q)~7l_>DNkM8*Mzwp2(AH5<4p^U-xa^rp9DB93on0n(pW#$;zt{d-5
zWi(}v;!VK_FiFH*2O0EXBLV`16v$KN{{GA|H@^7tnOZyNX|@HeaCe3;-(NgS-Da74
z)o&fAndCe^UCmD+*rI&*c;Qn&ICqK4?wdY(KhzwpSl95$-HEtW%UTenZx!j5-Tkw-
zD_^~R;^Wu%u$;^%oIQ_QT?hO!ju>uevX7Yd=;??1GruIlckUOqVh8J#Bc)VZ+%eJz
z)My>U=sQ+xgrz)9N4#ht98xAiL<f{&W=AQ7BZ#FB>webT##*b%Je@eHs%DE_8)<az
z?q=I0#R`gA8>2Tr@H<lseCKZFsSMVw+To9Nw3{35Sg{#haWZ^E-aEfechv!3b3y-U
zy>_;GjFFrshlDFsa}=!WMlT)k(PEs8h~-_ZHJEA}W3PmmnDfk53#}B}tb>@jj0R)R
zJ50OJRRx1jLkL77IZp0V5D#}m^#gFfh7+R?9sGDfH^BE}Q6%7Nd54IVKqE8gZiTxe
zKDL-&5C&5_3_`4kc6T(4J_dQggogjDT{+NNbw3s-BjTf&Ht5>SHPBMMP=>#&u+KEq
z&B(?WNNWs9NV0p~I|m~j3~^Ij0w{y$r^f?<6WN20ENE(LgX)EVelW_2homtaraQ1R
z>E#$%7<4YqpLs#~lA{}n9dtv5oK2;}kmCUE3R_c`A{sxedm>=SR)E%8sl(Q@HuIv$
zIq~l8SN!xxKjK$@?JpWpS_0iYA5cRfL<5S+L8m;CqsyeRbzRG5YZ+dOMT3;U3p(Q$
z8mFXnQ)7@48Z1PKV8~bsyU&CAzGx#DA7tD24{|7A5R*6#fByyJw|$-Kn-v2g+T1so
z`Ho72wa}Vnx3^x2aYDj`I+-vA=v$?>MhX@yDnrfLtr(;e<fZ1_AGWf0#I{(pb!r<t
zw@#X}S;dx0Rh!4D&w^5nZv89N(rC^`l^{l3O!h19?9|aZ&E+^-DfARrJRl@^C8&N_
z)vb8v?4}}Hfm)aXe0;y~>FX1(?q+H!tXrjQE3Iur>D(oFGf&(t6G?=(L-|~8xSOw4
zfk8+?$v&4>wG4zQMABrAk4sGSJ`)7CqE4VPxgP?{ocQF;1E8Ew3n9*K?6Ru_tf~}W
z@MabwobOJkc1CkakdwXkxzupGwiqL|8u_l?HUOMJW4|<6-g_y7A=VMSzZE`vd*hQg
z&&2N}vN@=jq-a=5WdES!GP#odvz<g^A#-X`&QX|`Y$GII9M}YlX%@A>64T)BvokRk
ze(=$hFUC<DT07fTD8-z)Qi|khvO{`n^ja9LGJ0dvL5q<X(gEqqQ0^vAp;XvfcWYRi
z&rVU9ro_khGq<ht<x}A+E!OJ{-&lwG{L**{Mszg;{^%ncLZA#~^JnsHz7CF!^`|zf
zY8}+nH`?5%J?N1QzZb8q6?8OmP)4t;{h4@rO`K-bl?bCW)NQs2yCu!=0w8$j<cA`9
zI3n(I=?^=8@l$Cudmf|5gCnr3LXK3v@bXzDSP?qLK*Au&Akpc~2xuZEgB<&qaWrIf
zf_Aj7^tKXHgB;m=V+S;hQz7OhU@m^fDcNO`Ihi*_Ov!xjhZ4MlG$8uLxM)9IO<MSZ
zI_<N1ycn_jV&js*OEdN3AN>)Jk5`iO0YbDfodxkg$0ctcK>aa9rYV_10E;Q^V}Gx0
zp9#L_<YboIsG`G3bw28;!=!*+Ew7?F5)*un(_wMPZI}ey_6rL_8lw|qVqOB(t$kGj
zt!})3Us#rzd0F`H|L%XoC*S>)`;%wWjGdq1f^0F4TT`K@#0XY=mg1&z>*`q}iOD@L
zb}8%bS&(C5+ey$q2IylDJyXIY5F8SI9d?ff&c#sJlNK$uUSH4xc|ngpfNlQ7>zfq=
z3R#3Xn`6|TMZcAez8TBByQzho7UKx)W*l`mi3TG^H8TQ2#2lS`uDY|~Yi%Ym@_7%<
zV?$fFSqmH$P^%SbMnK9@m_tB@XN6ab^9;$1H1<4$S_gFVv@b*b0543X(>t6NbBP0J
zqcciHYO{qxCyOw`FuiEygLzh<F_2a`X=Bd9d761WXWrZ|oR-;$Ew@hQ5!Dx#@{}W!
z&!|x=Z1&1Mhrkp~cfAL)ynifAdGf>FI}%lvEQB-}F(ysy<hC6$oaTjB4?<cSBFf!j
zu6ihDjq39_)BiVmGdC^N5CRZNsk9GJ7X@NYoac!zuZ_>1D&K#1W11%xckEKHf?M}a
zXea&ohPTfsoB#gDtkq?f<|1dE-cIf_i4}WG;gz<<FAr%nAjCl$O7ytm-A8_Sx0a$6
zf4wQCRirnhRT8*s+y`(zmV|lA$&IuIrFM_#HIfhl^Bjp`=60(@_8IM9PJuZm=A79z
z!FwipA4;5TB+-Wzznd!w_BhpV>(yJqf?t9kOO8P814S_+YYbKxmu<BfbT4Qvpp6hr
zwWF$tHlhT2nn<I2mubA6pv@0Id*6dN_o5Fp?EMlxvT(~?=Lgt`^2JRKN44N(ZoU7v
zM<@{-GkDv3)els+7fyn9ljwHcc?!6LXGKP}&n(8zQHdA>#Sn!(Cq&eVVyft92<q|Q
zFr`SFXKT!c(zKadpCe7jfqHcFK04hyZAS%nE8yKw&JGku7aZolJU$zt&5_4Zn)zom
zrS2W-4wH*@t~mv!X!l^xQ`q5t`!Yuc#tR2a<mI2YYKft@+9_^mL`;NvL2lQhAQ)J*
zdB<eDaTwWL;!Y#AJ#Jo3+-gS#<nt@2g<3bJd1Ae-{EL6~Tl{<f{(s;~o?TH8NNP$0
zlSUe`Zwy1X%VyE(Z44wuyI;al`~@(Xe46R=iEZ1CbjGpcNe=ys*q)n0FxSJ_sl`<)
zZFrE;u-NB5pfz^Fh28fTNGXP$n)q+OzELr-q)1BP#Uw&}1Ml72Oj`tFEv(NMBrJp&
zm}0cKcr!c2<CpQurQsDgc%TKw;2dhR!Dg;w8nEtecyLZIf?-erHRr7q133oTXqNY^
zwwPnI?k@zP*LpzL!=-@$QUaxPM(w26_sr)LT^plRv=p=zrYKC&bhfHOS`ycyyt_8u
zUJ7|CPJU~=O0Z0sSLc~mrxT~U6KS5<N+X45aF1>g$SHES%*@kF$}=O1`5A~J80*&#
zKJ(NbI=EZz+}GSR`)Eg^WxErSk>zBss%?WqjKzIE2j;0cd}-r=m=`4aNWm8)X(IFr
ztpWxIPSZrwL1~paS|OLF#CPsye)ZFP?k|OhQ=%)}R^vLDTcmL>42D%=bU0S^;9kH!
zYuwqp#y=VEyoR%74Q@70z~0>)-Wm?8^<q}OCT+F2Xi~DnVf3S0bac&_{iQL<;G7~`
z9}F?gv?c19tT6{87Hv&Ih2AzqVDb^rBnouf@E!UI0dj<8j+{>u^O9(-89eBVe+m}O
zO|as!bxTkS0y%xVi-B@^rpI7>ifJaN#Yn|dCa2`xn-K=Z59iuC+BQOO1nD@|YIeTc
zjWlmY^ul*zwR!uI?&oo~IErKdfi)Krtq25tA-1U>y7&ev)^(XNj@|#A9d8AUjVyPA
zp246MV~0omASg$@aOfQCaqqzC6?R9uV6%SHkJ~!ugU(%bFnsUSeGR3TJuv#`AZMbO
z8e<<j**dY#u{*cLikIkwL$}xUB75hs)H)8Px{krR7XIbG_~&l>EJx^49VQ7jM8nsw
zBfU|C$qC`jp>7-Ny%?q0`c>c8Uvg!P8)0LI(%s;wckWMTG*9+hq(*7R8%cSlmBOtw
zMv5dKdx)Az)BbzUQ|5G9D9<-^yV{u%96=hLKlqow$6x%lU*oi#oCs)4{*<%L@f~^U
zIxOc#K?zLDnJL)U?tBGv8XPE<m@-qE(B;I|8@;U5R*pT(*1O%0=*?w{gA^NGY_Hf;
zA5Ajrfq~7r7TtLmc){wKHDBlYW<CSr`ra{+phTZ5rI<)jsA~UCdt+Ro3?)s@&d0gY
z&0yMyNnPwfuBcb87TXKbhOuE=vqE4DWC%n|Y2Z>i2B<|)kr3t(-9==V!!#o1co#9a
zc&Hx9V=)+^%ijX)&_d)ISL=*17;jFtD^Ec=r9c8k4uleDL0D_&dh4`@VX$4yJS5@n
zlv(D)k|uJ<1k5je45c;mjgB%IDv+H`F1A?O*geth^JNLdV2H+?X1;Ux+Fx*k!c}c?
z+nb-kkp#6N8@3zmhoAkd(afOd$SU`faetY_zjhB-K?n0Zk*0}<(?Uu%^BzNqZyv}o
z^5@pZ5)-E-8GN=@|F*u<tHqlh$#TBo9a7SLtP?Y+`Hp*(%F)&D#K0gHE30FNK>=*;
zrux#=0Vp>Zos=@`ZS!Cy^B+K6nx@91$T5L5%20}qgL~3dK_@iZ#vwkN%P}xdi42Th
zT`FmdJIBN_O(u7Z**ZgP6k@&<MnsFzk*fgfX5m^(4t)BsKp1@X**96Of4Z(#3%v}6
zc1GRMsgh3*M4cRd9)vhdrVHQ&>h7g!@1%9_L4-OQ>(kGXJTtlg!w82G2-y6UkSxmE
z-=Bz5`U|mdc=x}D+3Z%t5`yhVF&v7AgJr*u0d~1w2)6&FOx1zT4XK^byQ5()e;a0x
zt3C_10=ys?88|s&5z|bV&S*L@B;oEJCM#6it6G1#CwrmU+?6mz%PR>{b}fn<BSqu2
zbc<f!O^BxH1eobs?}mbjnDud2Bi&7s?ci$4SD$^x<J+&8yyrOLqv2S``~I<y36d8W
zeIF|X=*2_UY^SmNuMl+5J5<cCe<*BqQ2L;@P6#gZ?iK09yLL$PVk2Egnp-g<`&Jt{
zC2}(Emw3||Lg3+iM{Sl&D-xLIlSQ>lq--0%|2x0SzxOx(#-YZFF*48D_h>ng%$@Ws
z(**Eh&gO^!qa$r|Vqu3o5!IyDy{xS3#X^e2I2?$4z_AgQ92rXnHFGD#t(?8^GQLY>
zh>_!Whr(SB{dL(!-%L#q15*m52vfAfcCTpX7&&YaRnRtQw_?(kZYfMjgehQ|;a)9E
zgr-6TP{9qQZIMZ!O9bh-jsoYqdNHx<Fipj<q{KM~=47x^lwfh4m@RM|-8zHb%*TJI
zRsY@{?9v_Wjob4xy%p1*C*hn7>YBY62m|_IS!oiNQY{=qBoc!`!gupT&VDF&3$CcQ
zji=|0>t;z9(SNqVjphZ=n{R{)=pX@kbY_jtetpGHKmX!bNWxLTqvA#^$N|G0WFMc;
z$v!v5USRg3!oSuApt~f+XMpTbtnf!af6J@C`PUh>`k^kwQy|4m_4~V9VeaW%Dpoic
zK`b6~+C83kFF>O6NOtl~QsWs650BZW9=%%;IELk3=)U>HgofSw-^FEnz|6HEA#lB}
zY{d+(Y`Yah7OHuQlcLX5Z4+8cW0z5NZKwu<M)ODjauBl1iF#`c?X;$BV=&K&SKVZ%
zBCr)V(;0&hT<0EK`G8gS5oL~vkMA;byyMTjiu`x~{F|)SKV4S}hOpqoDl7KF(Q#X+
z=$$(ph*gwuSk{_$+LPH3MmJxOvBTX-=maFqZr$#4S`($4Oi`PSW5+;7BS<2O`AX~s
zvq8XyfbP=6J(T7{QXOz?_#p^cgE7f8SuwiH>tz4ET*);29qR?rW(px!4}=(zyfA_h
zT8E>0GREQJ*P87cyX5pkLZkS8tj;XollEfR6AevC)LSzYH9Iom#rpoeN;5L2NHQgk
zM7H&r-}{~4J`fM@E5HjY_0sJi2D-n=KH-YN?16W)9`O-^9juNDs#(}KH_X{;Hy(mF
zL_r|V3vEX<y?6}H>mdZ{=*NClduJ`?5Gq~$y-Q>t9lv?~ny2S0!$+2TV3VryhyUvL
z`PJX}4enpPa%AD<ejXf?7hL~%kN09=zibSX2*W&H!W4*UVwq-in;-;6-H>6VPlE1X
zKLVVV>ELi}r5yKD#r$`4*u8D)NQKQ@52ESL7yA>}=-2uxzs{2t0%k=yUb^}*nB1$s
z>24lk)pH!dU_O*7Cq@kBC6ADEBG1w0J8B))lnuHowX<E7^(NHfkX8gC5Ph@U9rKbH
z1ihrd-IBPQ18cLFD|&?)duNo*;HGBIUr_=g{eb7JG81TxR-u*3EeH?iiHG~dIh)t~
z63if?Fj&>Te^b|s?;S99zUFKOL?O5xZfk7Um8Z+bbsJVV`5{K@$W|@FI9g_P7C|I2
zp-e$I#l(_kzVp#1{NMieKftz`9hwuANJz6Emi>64I2h8_McLdic-8w-iS7*J-C)Hl
z2Onox(S-<gz^1}^S6&P&5^}bVy*A8{2j8fJSm7mB{Cv3i>_IWYF4s!Vk`1;v+}UjJ
z@$FeTDY08ou#YOk{@1V{G)~%cRm7;~k730?gCQ<kvjeIqTnDBx1{xN6BFXp;`Iz0C
z$%+ow9d|2gMthMHwyspnx8N0#*^!^MR>bGYJ0#lOi7|$q32zJh;WY8EB(^qvlcoBn
z?t1;|-uWUGo7PO^DIR0N!&UEquX%x4fh2nPJs5#YtYB~_Hwl*BNyr8-s}tx#aILqw
z-lZgTM-$EIvD3S6?mJ3l*u0MocH*sk@F%c`7wzw_x(|XGui<QeKlr^U*$t6Y8w~aM
za!Wt#+DTE77?H5w6O#{0=pK12$DwbR^Ke?n`Otng`LXQpVefVbHIH5fjzf6QB1qG8
z6gwIMP4x(qOEEA{ncw^E|B26j`XkfQ`@J&7f_49G?4^OvtB?D&kMRbznVjy}BmFrJ
zb*?}(C&e~)Qn43l0+a8fBI=gvrXU*b;y8k#elNKQA3V2Rvgl5tGL#(5gzn+hJ(t@?
z?|LW?6xh~{-~Zj;<v;kFfAc7C{{KmQ5C4h1(EZ@w?bwG8N37Y4Hmc1r(?g`8R1xUK
zU+a#D^H!pz@5W?fDDc8hUOr0@@7u)22WAvyWWycZ`8wA(Dh5t@vYB}f<S5Xwid6?o
z3IsBx+l<JxrGc0OX^PBKvV8a<<_3tcOo`JeGtHBcOrkKSh>Eh-%G2}2yZ0+k&%q@$
z9DdQ!*}yu)>`SrEYsrBreh}Fy=BvF-_E+l&j0_U|WlSDmA!60l=!4;%VhA835Hv6m
zSR`-~Wu79}2DcbV(KjbC8%hgm+BM9kZcq#nX<MhRg{NC%t=^@GL&7nV227nJ1PW~S
zIQdNZoHCKX=Rf>0KmF-vKF{)YS;Q(3clQetsIA%@Pz;U1D|ZnND384(7`xlq2Zg|g
zueAs}9{7B{@q^F4<k!FdvFowzu$69{i(YLWX!>Daka+MD15tOF9K$VAhi{?|N(jzK
z9xp_*y@*JAb&kiEdW)ji=0B7@_ye(o#peF$41-x)b7)TtmJb5!7;HYf3{KG5nt7ae
z0&A<b@uYx?P{fom38l8eBzmN5BGv&76{b1!={a$CnmCJj%d3wZ%rIq;Q>LqJP{9i@
zOW-q~_0HGJB>&TTEvJ)_v3felURLn>ChbU26ekj5ma$O>LSRpLYoi*XWeW=EKB{>k
z$EnM(>`eE=NQ^Te!_^wW)}#gIeIR*NSGB0gm{paWGx|Z-tYaKTW;rEdh^91>L9H8Q
zy)j$?5k!b-B1Y@r1pJ-1a3UiJHghcu!07fT?A={OLqJA83SbdeLpk!-`|vE77il8;
z!pj<?6M`wL1em7D3M@BRv*!?B;O$_hXHIWmjLuhIe8oTd2Y>%y)QgV@)C<=f1GSr~
zXXgs2f9+wv+>INQVt`@#_q~aTPNLQ^sA@6t2daq`CUb<-WCwfqu}e{P4Zi)X-ijA!
z)-=r76db`_W9`sZsvVW2U>uM8^USu{_@wWR5%34U_j~;6um5FU-M{)bQ2{?=UQikN
zH;YIAKEpj4O!u#c_CH7@gk)rI5zw7%91t<1q!|7DnG&r=mth;aGTgZ{_``a!adwn7
zyu@YyN!K?j24>gr?<Ach${gV#2@f&RNtB)$Y6tVf-NI?f#1zOuNZ8A;Zs{l%<~g&R
z76?ZEn^Q)`=RNbnWuCa)6-wR1bPVQB*=KpU*@K#i%j@&Rk^?;`H4JLcOjG7G8&T#~
z22a<OUI&v1ISbP+sR=^s%GL%Qt45b8@X^D<$9FRypECDz;uP#&JjE61m0N*wz?#aG
zGbAPF7no7}aO)VU?r}A<rWi9zjy#+aa}ws5IK^yef?#1oLz#AB4aWF=_wFrUzI~4*
zo2#oI)<R&t-3*4q4yNq$YO7prqcDr&@W;zo`Jk@V@$&b3q0h?>Aqao(xBex6=@&k7
z7c{e79>dtKYSadH@AKru#?h@R*YHw|4DSrg5DK@ERm3~DV|J|j7$LkAF#0mP{SYqQ
zOQ}ALw0fAc2vi2!=xlvZ=#<(yogm~yD~Vg(*w*59u7eN~4dvFFZ9W>=)S_I21nyGg
z9BmPblaL~AquPy6@aq1Khx>_z#@q{`HP}g#A$rFqgvjBKZzOKpEbH$3{;j?A==N>S
zZHH*sQKt|{L@x%!<iQ%zS`TNzVP+fdqUj?vA1evqjVXIrmxSg8q7?!J^TL-HO@Tu=
znu=Bjt-o~6sy3r1exTtWFD`<;Hkv!Kc4>?tj8SN1qm}|{u5(`18qwFCJgulTYjM?X
zovy)m^i~;lW28tYn`Cy3;~lrby2+(2vcnFAT2@Bc7_A?l)4V(ey&IX@rhi5No{|ya
zLNFWR-AmzxaA*X=(Hm<i{MUcy@33tfyH~@G;;F;YV16CSp2uL%Vz)Zqi<Tg}E`A)+
z#vtmPiNO#|aS{w!+l`A-kfT7+;R+nZO{&b}z|4iZy*s!opq~@_GqqqNVCWHG7S;CY
zWzMK(qN~^jaJydl;<L|r^YF$IN%7xj9B9`5z5n&=!%@Q6ztZ=p9X&z?37HVQ=+#PK
z_k9>v*g1JR219YQS^i7e3OO2~8J8aJKX>mBgK&%-{-o<06$8jHXp=z}C1&B1`-z!X
zq+n68TdN>}`@4mQC7aeAbwZc1SW6v<M@xnn$@4@^PL$~b35fypoZ)VPJ{Ywfh97&^
zG-a2;jm~iG=j7f7-z_J8rM;$Ouvbd9=xeLAt)OKv`P?%50vJAesH&(K3-df>mK<1e
zWKM}WC2|z=T+k<KyYaXd9yew52BJo)!;}~1v}T`^0i|v{Rk&1zU`5iaC380=?&gU(
zPt0kya2$s#4?AC8=1^ZfUD%K4eO@&bw)MvQ_fNLr?cHUq#QS@z5=*g;D~Dq=VDF0T
zXx#n}Ud9*RX;>%o0+&5sSDv45WJBH#&O(e9^0CX5)Hkgi7Po`k4!^yURqqlUQ*>W=
zg|Wl2A3*<O7-aK71q>Xs-Tl=c-8Mpo9nuk{sp-DkFt+;I2kW|Ve}6LGMhLvEoqlVS
zR!ORyQn!t2bb>@W!3rK#QiPLO#I!Nj^(cgDRj#$N#Xv|9gF-WZ`YG>Fu(84Idj)ED
zXcS27wyVlZ*7UdYD&+!r=ub=x?c~Xkfic+KMgD)({b{UiTXx@v{l;w8+WVYyo7Ytp
zi!2sNiK0k}YDkF|MTwFvIU+3Cjshrfgv5y@Hwg^H&X)jAg2YAw1n>uc2olFJBF6z7
zi?JQbmJC^<8%dNzN~YL|O_9~H8dkBYUVEE!&faUyX5_;dbFF<#3Hj~^0aY8QdhgwP
z&TiITbIvjT<KK*fa&$Q%2KU3TuC@(PfePI;CN4R37B;2}7il(=Z`C^MkRjNNIk+0c
z?1&SCb97l;4tU*C>8y*pj3{oUQp$Q^CZpEHx3axgLO|6z?GdQdTcwxoDyxh{`wWK!
zQ)q;spalzNvaAJ3laUmIjWt^rT6af4C-tHxk1p#%Z+(+#k2QR`hK8l@Wv44)b2hku
zl?5d7tDpN7-hS&%#{7SHER1PN8zVwMgT(@GQ%QzTf$g<jm9nMR26D8<YDz+HmTM3u
zyBluLhK2<n{~6!GIuk<;#KZG&lt(NqtT#~GzinLS(o?|-aD2ZSW1iLsQcR{715-As
zc4?K`luvx@V|?_Z-)i6CP^!qjcMVdjH8*w*aRkYJbX?CNw;ZTJAQCL*UIW?+T9wA?
zih`l0@YGVZAg)>(s}taJ3Pzrku#sU^9Ch8spYLytzEm+_5~pZ|0G-2>m|uIy8;_3U
zm`w{>3x_;$?WjP?#t<-@N{p2rqE$ELOjiqK2w=%^wXv*)`(>eOU_MM-9g;;sc1N|n
zO3Wz{2Gq747J*TRHO!?dgy`l*K&#Li(i|Q{H&c%utjdbX<0MlwJ5aGv1pRVms2rBU
z)!I287S^WFE6_|UEu$J0BqUHctqX5GyXV)Q-tl~G#FRKn<d8DSfT6AP4`N-C(yU4!
z^DjUAf98{)w^-I-WrD4c#_jDbYbns0>q~XBB|aW!ySZw03F5d>?GW~{{yui_8yIz1
zvyXYB#n67~o%i_CyHEI*55116kXoRYLX1i#SpzdRSez6iyOhiBi|(*uAV#-G8A=fh
z!W%_{sx7*JYzx`m_h9Y<SjuB5W=lLmXF_oEp$1co%qfyXA+*L?&Sv-#Bcj52?Y#5+
z#99h7of}8C9?aG~TzPf`y^%xUASO3Wq9nUP%?_H*xmA{`wB?LOi_|q8x@@yy@oO^7
zL@boV{S;zyq~!g{gT7WD=iBENE`$)LnV5#pLf>Zhc9h$->0@TpyWiR!Fgxy{>DLa;
zEw}-jqmJ9wRZT(6dD6^?3=L3GBncWlRn!l~X__`0;uzyLV;_V?Yu;_MPchP3H0MO$
ztW^smpCZLz7QC>HtAdDaaxs!LnQ|p&J77~cC%YKXXk3RDB5^(t4p*d{366veZ+zIL
zdp&GD0bRRsrQ9FG){~D$(@xskkTf;%t{{j;FSc#(I-eJQ@#lUX)MSS?1~rLut+rq9
zh@B7}0T?`iVL_ha$bB&K`8auuplu_9MmJezik#O<9)wzU5~9WW8k#gb48|BVxH3!`
z$|Bp_d#jt@#764G-t$O=4Pq!_z`F04qK%GI<lVR5=I4I;r}#&H;13w`(fa1E;x9CA
zzs(59k_Kg;?JKsg50fnWHfRB+Xk(0W2Bp$9d67594mMgGO<8N^Ya*(&tAnHLiD}B_
z&vVhRy^M|d%c%*1aGVm?hlwbic?wL&Yt#3Ku*LZv=9$BknC8idU0vuB=_#X<ktie)
z4pT<yl=B_u`;%>yy;IU^mc6BulWo9JY;)*3OceTdFxY6>s$mH}s|tZ(3Od-ec7@XR
zd*g-}6~8JjSKDJCgKyZ~=Y+l6fweA`^SRKvGG#kORU%Ui%=2V$RBv=KKYX;aw#IF#
zJULr9?oFOtU18nyn5oe{&nLLbpaa2zO4M1$)@u43aS#B+vH)r=C`Nu0al)RC+;SnL
zjD%#2;*@P}C%Uno-T%<hj8O<wBer<JNbAm=SApq(*6xX6W-b+@F?<Ifz$)G9q#h7A
zGD*=~!YZB2&^#+K+0H(0=GJuUVumx@CbQWjXLheeMh=1_D?wbrf)^WsDZ(6-dt^f(
zj>m~58S=2K#m5AVCV@kQ1DTLBQ)^5qkdqLlF-ah5bnUNoD7_k-`f!R|0kan3+DHLH
z3UoDIg<YGS@=&Tdxse5F%Ca_M6!d*sihiv=S}AByo7KCNtCYq39>9_swHoph#kza7
zG*MbHt^m!(lv~H-s8dj4RMKPuLfepoDUKX10AuBM8g=PNhma2xg_zW!{GfiISTT}w
zz9?p0ZR7`Vk80mJMoty6xwai<LmwVJ2Z7BL31FyiEPi{u#}4i=sMvj^F{Allg#d;M
zyx?=7714@l*$(RzjV}}>?pC!J|A5gP`+El2_CH#+^iDB8SBw+d;qKWpo<Du+9k8vX
zF(PRGzpVqs$$jkh(VM&uw()=kfZ3l<X}90rtb4%`2tQ|iyb--z#1Hh|yUV{LAvr=I
zM&jHEp?1Ny&)K@V&9fkF<Rh3#(J+pa3bNqerI8jt{mD=8fp7Q*zV+L_)ndx~-Y&%u
zHtF*~OH@0}|41-dzWC@xL>RVt>N9^rC!~lnLoIY!{dTcy+jZ~|uzR>wB!;coiDCFm
z1Zo|83?FILe+X&<)k+M?)gd!a!DoAU^Gpuen3(oOI;`xFX$xYNyvv1hHXB_DP}j5T
zTI~=@dE%ApYnG*Oe|L-4#&pPpEM9dAssVHP@(MXdW5qh*%EmE+7}qxVZ<7X%8`KZ}
zF>~IswEggp(Z7swkVezgm)<GB+N^>-Ot$btOy+an8oJsXJ%pJM67xLq=)sZS{0)zI
z<MEMNyN9w&yn1us`k1{Fwnf)_w`k}f7TBU{4r;0}d;zvOzb(J{FcY5d8dwoH9A=9r
z?Yc3mdpCIe>dJ0hbt%;)q-BVJF?F?7m&?dj=_<nC{rG43)^GksJNP}3E5xxe*tN2G
zrOn>WUx0nl`$C`mI~1`E_7bY?rEO9jBaOQ0hK%>={}dzLI6&YV%s?6XzA!N~>X;&T
zt97IiBf&npzT%jk61vGbV=zMMqhsRHRbrL`r6X!t4ap0jwRUsgoXqGW2j!Kk+0qG7
zBezCFz<O<zYQ-XIp5`HlcR5a^QuAC7g1afc%u4;*dkEGU8UNHF__feu-_f7};d%i{
zD}+`Ey%Dr-BN(SnZ#8gp$*BqiwfS<)aZ76(X6LP0{zF5cRRr^xND4Wd*;eqmaPttH
zjgZLbe$6wbda*E+1Xgq#%45rNAQF8;-?pWNb%tF>nPQ*a4hma0f(I#y5pdPHFATY*
zuTOeGi&<wIC>88Ij)-8<4YF5XE9&Ifu^;Y9nMOMFp&KCvKJiPx#JVgSu`|WN&AB*W
z{DW=fjs%11$Mw4$CHVdfHr6qZ4|f}M-w^e$cyg<Ad_Nx@I3m8M4@6++3|a9Xw)Zpk
zdeuGC!O#*Tst*z~Vs4_AMPuu<xea*CLM?@#_^}@&0N?m6A7M=52hsbhxdh|2#y%&(
z(*%6)OA;v6<?VhA@b6Dz#A1LA9X8Fr-NPwF=A2pA5m2>%hiOEDQy8+-%VO*|dc0II
z@XF1Bn`5@|z!XV2m>q7{EpAuboJL!t69{pF6u3K`S#Iy?2ssG-Fu~Q4kS0p)oY%tn
z{@yC35V+2nG-VF$KwDRa6fWny2Rk_Id}-7(;5@VJjeaA@K38wu<U2lF*{W8rB)1#Y
zS<Zv?X+b@SAdylVwKX2xT#={5>#g(p-I=OzHD{(2=`qmh{=UqeE;(xO!N)h8&n9I7
zMWbEEd77B=WOKqe($z*WV!K`I&GOts{0E0Y4i`OqaLxIArZ$_=KX`D>zFch<b$549
ztt<0vH;@c1H*9c}IvRkRtsvsAhH<O?Kl#P4@YyfD#c%q?51NmF2U(5H)1*;pE40EE
z%jum19;g%yHt&j?MfpYoM3)9`3tn7;2M+tGe@9rC;DuCe<2!bWfg>PD;A%=dzMha!
z={a$oghK?;%7-R_n+fP7l!wQehsVT&90<KaQ-}I1RM4R4c<;W6MllQNn?tte2{KIb
zDBkV_AB*$>U3O(CIW}YioM`v{;l8iMM+id5=B(EwZuiCrJ)5S95C<u(89L(pB+-uO
zUg$;V!?dmzNgmkNvr6=XFnQ?K#ds|kjEzE%8F3}RC_Kfhi;<kVz4sK{&%}=aH8o63
zzPUSJB;Z^a(fvSZZCB82qbyrHcE3RCHZ()5WNoceX=t{hLcAbpwKB5meX>WZE9G?0
zTB}P8)eCVy0K0XOBYa6lt3A4H4VYb<&$zoCJWVU7bK%#%@I_Gj{^D?AbRJLLQaQb2
z*S&xOJM0I%pT?ehQG^)5(U7>DlPiz#v5X;X%{h~Hs9TfK$9<#XUMXzz`O&ZM+#%oj
zl#S@P7Y;CX4|@*i!J3kk62m`YHyLXD_>cV<-`9UXANiJV+4rDPtgw$p2BA9yvFL4a
zLo0tXgUGGj?h*4bu=gMm4f!2>5I@vYh{z<=E2UI(Os%z1O52)ZLyChrBnqfRRQ@5T
z2|n_nS9x%KWXi@)Hx*mLY&#=$U0t&sER!Dd?(bq>T3F6#DPFNJoXZJ~6Vvqr@?=E8
z^?Y`6S0Sftvomchs}&m4lnEMa#?vg#&lp6uSVHpIQ8$TJ@cG+V?8#?3BC;KT<4X)7
zxLJ<hFyQ7FHp|LBOUiD%beQ}`YxaK6RXNNvIa+9a^YMw_CQLbVm?qY1o85P3h>P<n
zgg^{wGh#6b9v2m_g?xk%Nh$A@uRlT%BnrplYz80eYBDNqHaC+%YnA1^P+Q~r`skZ+
z+BU*b-;d`N5!9tzA(*efriq*mNIFx`C;sB!{we;<Kk<V+y1FvSpa?^*HEf}@8%HAL
z=!@QEMZ)eU;O`-fnXxY@MF`FaHg5?3ykfS@#<qr(qeZu>p%Gq@Y;J|3o!TS0My6!*
zyLoP`-3o!&3I__`a1>q%N1`ZEl;f1RI?kqssRsMj)>xb84vM+H*?D7d<(#uK<?Z@M
zA6<2KyOe6N-XWNu!RdUa7TX+~K&ivoFp*>6w5~6+RKMmPhr<NP*zqakt)odPlJmrL
zn28n!ZzQ`mIRUVs>jK>%an%uhZccuEJKUT|7$XuAGHklRC3BJB-hXTBG%U*cQYtp!
z_vq-~|Ic}=$kw$FBI_2FEkf@?R|`(-%|m6RZ7li#8(fbB=eq=7FK(`d7;i^Go6?PZ
z<EX2T&=`b9i_NX;LUFHw=otVbn!DBJ&5^5c+?2qq&HXSN9qopQ1m1q@U7kIAN($yJ
zGYrdm>)!YUD~z|bE8E@)T)27e9EBK+SZ#-%!P#bt<l_o;gf9eld33H=-w~D(O)O)*
zjpD)&_vRyC|5A7GPzHjEe@5d^jYcy#vYYbcz6-W@rJ*)MsIBt%fBL6+{R1E1cy;t!
zB5Z~GXka#_L06Zw`gcIzQ6nEeG4^)}R*a-b=o#&mUhBpeOCga0bPX&AMVIA*KhV92
z(rw=xO~wTz_3J)fsu+0vm51D1UE4g~^Onbkr>RrK4j@$_<YAtoq+t2vvPN2OBzFuB
zT_HeL^N$a6wosB*X{UR_bRe`w^11oCRz%_kJ#90a80>=k4OboB8DSVAWBwmD7-w@6
z^9^V~bc_q?8$<G9CdFuaTrmk#3P85bV!&}jSDsR)#X!%|D^@3m1@9tKV(>u#>wd)j
z!PRuV0UIF%*}WA6CpvcJ`@iSA`KeEQ#%{oi?<j_dOw-1BK)c`6igsf_Bf`zim3dZs
zB{^>7!B}Nf`s!vN7+kHvOrc^3#Onv7>A=&|Er0XpKF%Ni{@;ZY>Mf9}+o!`JV2Ozl
zz%raFMSQ~8eO~I2;$U#A213kRAdA%t8}5m8E9wSeEBGAEphnjl$KeQO%;8bdUw4Sn
zB!=`AR!f<CFb8h(<X#napC7YXZU@QGEfsC;?p7Clwk)=B9HTjHvLS#%&W2dD)xO&l
zU^>jCWF)ps=Nu{>p*HV$La@R!Kxyxj68ts#kZ6Qllk5f~_C~j0PTpAuBekYHp(&wd
z{4CJK4}}H<Qu1U}-v!3OCm|w|2pBrhz34|Q2`~_F%8@7C$&R{+b<4xpru%(3P%#F|
z5UfjMle@a*fj|F{q;9BQ>qc7aVp3rlOD-T9k<jUw4X<|G<HRJjad3THUEtziLp`ji
zwxG3U91288*%VYAwqig~BwEegMYi_@Y^a;@A`H5(z4OVBf1I(=R!;ym+QjbXz!<xZ
zgNI$!5rXZLgJ)pMDj&xN+huC0%sDy&J9s^|$CYa4bK@Fsgt$@QZ(}tx0IS{nL)!jz
z`o2E=O{~qav5ZMGHy6k8-bWUNzn=K`0GuG1GRt}4>C<N%4u{?PcXMUj#{&^@o|~hr
zy4s!+gB8PSa!SF9Jng|lko=z^(1O{mhu+P6C<G2UbA3FxbE{I;vLP^n_s0XJ>9U<O
zztQ96vI%0|bH6dBw|WGwguEI3wAzhX-#gmed;>U65skBXw|i%(s+2X7CWBK&I#U!Q
zBkI{@b$ebs$`(Fi<z?KYw*&O){_op|bXnoh{9l?x1czsWqZfz;$kbZ7%-g0by%b+y
zeS@;>kZr_lvO}LP5@R-7i)Nno6|Z`A7)_YM!5qM>%WZ9l^Q}8J{q2|~dQ}@DNtJJU
z;~}p;e#E=)J++l`X*}e*=4ojAy@P;G&I3lZe{P{u+r&?OJYZERetn41I*AdY;)iq0
znMoe<^Phi{f8@8_@sZbGaXFUyO)hMKQo%SE3g+I2EzYQFY+1zf80<2DjU`gJD0jpP
z1oaV)Ius|-q>`;)cn?G&`eqcuAW2)-6hp9?qy%OaiihEZ2-6|co5^9*#tgS&!{)2j
zy_9C0j_6&eI!rf(fV9;$*yFIUu9R*{i`eY-=h?<yfthK_CP{R}A^Lo^xAyhV8Q9ie
zLgEbUdx*%86o;*g+Z+i8qSB)iOvI6-$jGa=upPUnY~QxTI6^k$@&#^<xq8~TY>-iH
zx7_Szqh)8<fxZ)n#}eFOZtH0AgU=Y`nUuWyc0K&gcd=_>${FjN#m8sXb^3kPv9asA
zkPZb&$*)!1#~J>9Ez;N#fTn~f46tMyjrQGz&Xh`cev7@oXl!5aQq7;_GoSi2U;Ne2
zV`>lCf{Q>mmsJtl{|28*f~}3{ooU|qK|Wg-7=~vOg!5wIXZE`GepG!_;sr_H(8_@<
zgk28qjYsW`($v<?#eFjh#j$&M18(ay>}m@0JaCSNe^=e+3bYM%P}bI*{1{o+Mu-`D
zq4u4pFp5X}r@im44f;x?+n(<Tk++FvMU53_UQ)1GVmk_zTrrV?6$*-=y|OHeou|##
zg{W@swd#dRzt}#1y~j%y13N3zQr`k0V2*wR5=t4>Vjn_4rEdp+r*D;1*WH$KSr@y}
zw9`vNo4K<kZ4i^yPELr^=FyyDW*LY<`Oz!0=o?D!n0EH7=IWQTz^&r<?%00TWPO5#
zssynOylJPXv(}^EIQAqE1YsPOjxb2HC}R=CzYVMU6$y6pZG^F4Jw=KE2}<vR^k%^;
z9&9s6K2-(k${WYWe9Jd|Bk#WV)R8b>85f^TFX1tV2xZ&M98Q&BE{3i6!mehi>zzq!
z%?OB|@Mi94cvo-F8B?O^%-{al=lIkA_=mZkl80FC^Fgp~#)wm$P>s+SF!aT~GbM--
zV>Wg!hMd@jds*B#q0*6$7#tCF#3&_GaEw`RM?HDWu6I=17^BA>A4o<3LoC3k8F?>7
z_lt<p=bpO`TUy^xs1PH^>nrBN0d19Xe{1qaW6o@zATo~*Y=lE48-0XGRw%8T-F6(^
zX*UTM?{h_ejXri1FS(k<qRNBu;{YR8s<_0<4_>PYz1Zt<bIKDLmedB@<Ar@wmt^6E
znoF4(o)_W;q008V<6yX)>08s_1%t!V+j!Cc*S(81?+gnE5)#@B)$11rggyUQkM?;(
z2t<d6Ey2_flg?oH8!-^|nX<^%^@7_Q1RsOir?Y*wDLpFCHnoQK4sn;Kj<vGQxcxJh
zQu)khKF`no%+H#;oroVSj+RP5#n!b!`?6DW{S2@*=*;&%BEZK%Y<A<NTD<(6BO^|E
z9Qtcj*6J>Vqs#Vv9dXxDZ?dfTw=;n8{rTZ-_Y#bZ5p*<6rcP4--Jwe8o^rXNlD@N!
z@rC+~Kc>j9eenzYmhbouZl5pqd>45U<6c+XjaMxIEd(op#WnKU>}=WP@s4@=JrPnS
z2f=cEta!9ByhXx?!;!Q4nF3oO(7PcBr8YZr-)A-Oa%uu#*mcsC>IZZ6O;r8ho1&R*
zsOwE72!ueXN~dj2XcU8oEce@)r3fL}MM_CL8Wd;o@7;AwHaCa8;@(`zf>p8-11V2-
zfQfi&77kgD(_?H5!T)EE7_IuU8>KU5V)$I&&=I>?Y$k2p{wNe|(+PV~G)5aSIt;ZN
zery{}unkQ;IZS|<GCAkX5o|)}{eXtRAN;=G%TNEpFR`(Nb-zBl;e^dKHch)P2HBX(
zN?Xs|-`!g!>sjt0K%O#Fo{$u`<SsKF!gpzfkO3j(iC_A)xA}$7eT6^tJHD0Tr(f`~
zf)#;mpSN{mE?cLit~zO!8rj;`$@Sh)+6x<UcXu_xb<e@cNRBFuXzWo2Y>U%3r0fe`
zV2e&x?VBa_y1MhAJKGIp#_qY~vtcEy1&L-UopNTnx;9(kT4<Ab1&FFg&f12Tav;ZS
zvCd*aSE_-Svu%8#c{-&P$F)|HL?*tzeFlvGYK5~E@ixK|Z44(&MZIvk2v|{dS)H%v
zXxtUtU}fY|*#6@Mey~3mC&Z=bJRC=J6`!~3rP4&fi;7WXSU~`CaqY)uZ{7DW;c}TW
zf!3>=9__@lvCeJ(+Batjqsz8pU=YijiX{))r}FOF6jTyTM{my&P}zyZqv<kfu`N>L
z0(1ur#{umhMh1i|^bX6i@;86%$N9>azv6v`x-40@QJ?y*7lbkJvx9$YY;43~hicnC
zyS+EN7Q??GMIonM3DCMV3TtVsZNG1D6tMe@U3@Pe2+}y~9DnhvW@7*mD>8@T!2Zq;
zAh|d@Ff^c)%7B2hTXJr7UP1C~2@&D*pZXM^|L8|}{Mu_<AqL}na>T{wM@AkT+$JM|
z_i^rY&(=YZWaADx)DQ?E6XQe_MOtz4vK0=K_()@(T*7YNHfw8Ci|fJ`)juRP0kxtv
zYzsiNqYa3b_|#Q5mcJ<OKX28#d5=eGXwXM)wLg_Zg?YRwm%fEyk{|U&S!;1dxVVI3
ze2ri;ciq4AXmdpgZu#mKrhr4r{_BEsAjDG`Z8M1}6V%!8Sj9Uk&Y051to7RlArRyK
z=i`7LqZOMgMVq_H_O`9Fz92D<O=G}<A-JWnI0;T1V(Jh>xtI;Pj=GPH&A%fCDzL+w
z<GN|D)Ut9uogl!nRB{v!bKZ!4F=a<FF0QxX?N;nzP`(v$nZNiGKTA3s_ygbdo4I*#
zwBJ*uTZ)VyuEDtx78~6Cl-~Rvunlr7+TQARZN|^3|J;xc4o&Sy7c&>Mk=s5Ns4}2w
z{QE&MnUX`Pf^1`=@%>n##)i<0Vk=FRT$_y;LS&w1ax&{;nGd979S}ZlF^_bUqBXbK
zZPnQ4T}(lea@s~KsO2663FPwiO&}bQtTIepCXypfLU2g4d45LNI;}A#7`X*;tXG_R
zjo&{`%o;`h7}v?gcwi*bjbuIJt~g@g*!m!@hA{ROJ7mPO494EpTHW|Dz;-BFf#Zeh
zb`V||Uv=-+)hvU%bP9#wd*nFWLz+z@=#2zoUIc@=ZZ*Fl5hmHqUI<$e)oR(0rBO&5
z3NRol3ug-{L(H7cD?k3XtPn8d%GObGv$o*H+Gy@t-;0ajdzg*lO!{c3s(t>}`nFCt
zNpZ?nEX3f-A5$@m(LnQm4icOd%E5A8yrUnAgct6y?c4U^Q@3M!8wYHmLky&5uL#U~
z+6c(4th7=#k|W-HD)8*dyPWRtc<|^U)$i@>_W0~~oLzG)#7b2yIn@i%rizGVCqy@1
z3ckJeQOOMdi!r7yD?uy;H`2*Lh}*eh=SWRTij63ZHY7~%<MC3(Kslc&YF3)r4+(tk
zW%}>our{I|9WHJ#n6r7B387jS5Ug|V*bOFr^90A!MAV$hu$yB*zk)jS=ZCe#t#dZ%
zN?@K3ZWv+S2eq#Ldqx@@9^T12?{PC3iOn~b5XV<)^J15q_?t5{K*Y%}yAnWqr*$*$
z$Z6WFLB}X!Y!LWD!e3=*U<afaLd2&>8+&%#*b?Bkf8;}a=+(!3>8*Eco6r}(`0cs9
z7#KGxDB23On04vlZ~(u#rX06U%2WoM_Zs5>Rm`--PzRg);fMZOIzRkZe~iEJlRv}1
z@TdMb-~C&@-E2sEWt2{WY(>l9jfn2cTEUTsK<@>u^~FuC@9u-@h{`A=wt230e?8g0
zTXpngdnI*^vH$r7|8_G}dNC^2xdC1@_O2Yf>Mmsk)s0UOQewWkqLe~*Bd!q4O;aS&
zYN3`wE!9T}j%3N01A74?hGHR*mgQxZ>et*uk?mgU%^2~<wHOMJPN|C-=EUg5fE5N3
z)MJR1fde4c%{K3*!MQDpZ;Y2S;kFU1dT*Q80YpD6WP{9iz}?r9fCfYP1_0c<S*wrz
zWUNX<GAw;_++vr~Wh+t@RVySAgV@a}6%0-^Sy_lP(PhT!q~>AtW4DB(hZ4zYMh{9(
zCL`><(!MHZVHC@QJdIfKp64T_b$<MB{{&zD@>i^ZNmc-eqj2KYz}Q9<A6fW6HTYNM
z&6oobTj#!$GKZYd5e4svYqxoSt)0?3Ywy&ybBc!cUT|K8Kg0=~4q_O}!sT_h_Z@tF
zTdijM!e*A_Ma7V`mUXdf?hS&WjAQ?>d(ZMt0%@KMvG5T^t>(oNz45hiRLAgeOY977
zHGYTR+bIWktaWs$ccx^8=Xq_^Wu>(RS?^&vBef94&l4=(eVhrJtSB=Nm4r@8l)4eX
zf1}4s6$7?Mb?Vv(4vng+Icz42l$=7c5rHqt6lNsYO_)+9DwJwn&XmpTxidNdgRXlU
z9WEQuZ5))SF)wXIV6PK~T*bRc36Yd1raU7l8ce&%auOPBF5GJ~so><8gt8qh!KEfP
zGi)RgLLR0LPW%hW4-oId#oWUTdiK0-3HDySYZ=hIjL+;+xs7=|=0!fAZhjDsV!)RG
zAUuv16h*TGMZK!E@5<sb$6^$R2F;nueIR;w*js7q>LxorlMTb-SA3VnyOSMVP{rq8
zD`@ilz?Sfr-g=jR^UwWx{>b-#AAjag{nH$e*97)W$%=!odZ<D2>YNLaNNuZkkCz?N
zILuw&J4ON$&=G#qn{{+vL5;3-`~2M&si<10%f)vBdndcA8b;J|-+U0Z4d-H$8HJ+a
zkfSrwhrAKRx`S5ft!?(VDJ2VYVrcF6!cIhxuk&Icq)1Gev^5bpauzpa0K_}?k#?zG
zlw+GZ_B&&Q(3+1d2dt~k3xVxB=<box1{%<}39IAEm&LdvP9kXYLn|mD<gGzyy<0LK
z##sqIPQ>@7-rBYXHhCFr&~bAY(~u&PBC64icRH}63dr#OaA?1EN-3;$MX1C$G2s0e
zGgn9R@+fs-Sx&6$uzKE|JIB2|h=*oo75VIEKf{;5^cBCS{rf_;;%YQwm!@QwXh-)3
zZH!e2x^WzA!VtX6-LG@+N{qt}*j}f$&be0Fn07h}F~Y63!+ER=*v2f@M22l|7zGPB
zN6CM#U1;|Kmq*s^=fp=98}b;<UuL_Ou3Gx&w|*OsUwzF#C!0}PjL|IZw}P~J#B#ql
z8@zc;k({H;&z+CpR_`f9doOFD*Fsnep;pUL&`xX~J@1BkJ?{-t4noK?IV8@pzF%PA
z<%)qc%^a_<DYbZqZHIZ9vdM#5^_$KOK`<?AbRtOd?DDWCw0AS&ndj{W8;tAjrb0%*
z2*jM+OW!KW=p7V(Xbc<1ElEmn$x=pCEN8wqdoI@@H}CvXiY|8<{sM7_;0Daq>$Wi^
z9}#rl2;z9pc(F8?`vXSv&qdfHbuWv75R+|SJ|8tTp7ROR2yBCd19ly4VcCZEz2Eg)
z_`;XJVm7vArJPr*<YRProozHBK5|f3HprMw_i7{;m+Oqhx7FE~yTn+A1A_C7So-Cx
ztwHacPL;p>*Z&q8gn#Kz|0$*z5v;H>xxE!N%D7ezBL>?Hk!ac^pYMz85}^I9gpCP2
zNRo7Oew54i>;F(Sf~swd&IcJmCddW)vj2bYjp0RZk}CFtEXWr4wCU2fBEewe?S2|C
zajT9<^o=7iAX&3V9fw`t31G|DcZ9LH!o>=LXrWEo(JnSSXP<2k$=Z;R4p%;C-e{O+
zyAPy5F>6+Xt1QFaytOT;im|T!J2iH(t7+`J8nbX^6x4Reg{J2EIYsaMJFOM0pWj>2
zk;>P*lU6TGqEz-`pw!h53A6kapce(P0$sX0bF@w(8liBEl-8B96iPdTL}SBr^WI1<
zSDwOz8Ijw4^P-wzN4^4qQW~H6?8o`ppZY0kYor*Ml837K;WoydgETuvbY>46%oZ0L
z*IlkodmSH3h+1*w2Q8`_>C%4K(F-ooJ0m#<amT~n_jSYA3v7+4Sn+5F@>qv%x-)ze
z@aOHiA=-?wi3C?O>4n-MFio?Kw%27FzoeAN$0Nt9EAl+kN+pVHQt(4c7DDFtMXkEY
zpQAP;CweUirgF392U;~reA?pK15?iQb=iuV&;y}6irR~9G2QQV|88P-504-O@w2ME
zKW*Vl6$1}neS?P&ZZ_C7NF?RF&B_7SznVD%Y74tn3yR3evIv&tOs|zZ&CJIerg^f(
z()%_d=)D=My;^}WtSVC)F!tU}jKW~^IhPAk9uE3#-KcCg_oh{KKYX*y^&lm^kRl2;
zLsIbqAeo^NJ4r}=9yV@T@fp`3s%=o04EF3U5t0o;wLPSZ9UMU+NT$~6=Sv{^a}9^`
zZ~MqMI-#vWj4aDSa}taM3q6T*-qXFB-Rx6}UR2piiBq=Ui6cD6mynM6oUxYWqEHx1
ze_UgRQIPg+<1AJA&wuo<@x@>JGJooi{}8|Xw|xgGW~9yRNtRq3>Vu?{kdQX#uI+kl
zhlRb=_?$9|!`_Ug%RmFhdmM~nyZ*)j!FVlz#$Dpo$6+CZ3>63$u%br)-Yz<l?IE^_
z+2J1vaWC8u*-Vm@U3#{mU}A^=7@f3IHVCwEewn5EHTN*PT{h1Q>(Cn7{n*;R=l;T6
zJ!~>!bEagB()xa14pq>G9<n1wVdIDRSjxRX3_iATz(DKlcTKM!q;Gn8`x$kZ+Tvri
zjo4zl$LxFb4=cewj~^;H+}>KHmg@V~%%sMJ&=66HG_}J{v>57G7Mg%GrPs=NSy)SB
zDV5syB5BSOfD>T7n~jMO+6A&WN@E+JJbCXue(_@;=dHKib~JMAqmFp>-AQlkv%^$q
zj6(ZGk=}jW7Jc01++dgJ`OihS503k|cc4L|cFrjf94*nY$EXqcc0SZtfBQeT(dz|r
zxpM^ef_oqgV=N>aykG_3HiGO{G&b9Cw)?oZN?RAc_1nIUDMV@-gk>XL9}bz>VgA~>
z@!b6S#t=v$8Bwr9RVxB;W!<u@9FEcV;t>etBLu}T;o2^mC)fbGKt;b`l-B6Zhme>Y
zg$YElIPmxLc&TDwJ|3BlN0PIoZI+#ooVF}%W5LTdODU~FtfZ7U9FK$$Io1kl!9r;|
z5S^*MSxRm*u9do+L7U6>+@xfhZ4vpL2QfKL4wE0q$`0Myf!0)!9;mGtfpIGUhV*A=
z6bDhizINe5N{A;5!YE8^lN#YZW5XJY1GNHV=;5V7&_s;7WhNlIT44(-=|BiH10$$T
zd>kVMz;Xn<OIEKk?GS4m9^P=iKU?v2n8`ViMJd%fp4MQVXR^R%Fmxf4yezU{^vC<6
z9-rHHh4FRSPhP~S_P)<fd#n8P&;J5H|M5@qJAUhT@TdOx5Ai+U{o9CP+Rn)VyAQUn
zxTk?>bO+352Y4SHqj%-IMAQ!8-Yq7Vak%)`V`zg~0iYehP?GM-jqU|QcL*3i(5-6`
zmxXR9rIT49gk5rG9f{pP!(w@KF;?t&QRbUY^zU}u%dIwlZ3wY$hntl5XDRwM`;g$#
z!`;^ffOzVrZHN{O1FbZAHNO($s_vt)-bQh4NFhEZ7*O-rD8Y-yXocX|P)CQZP0i8G
z(WFZv8h%J5KOjSZpyPC_W{TE|LHx~{7(1xkTVJ{>{sxS0_hBuCT8&@Q1WJmukkD44
zHhT5_s5L~-GzrWHvtriPSkEh^H9y>Jt{y_(4!aoey~|OhvH!GYp=TlR?z?aE%fI|7
zo<F}e(jHcznZ(bpkFQ<vb&R_F(Z4H`FShMm?oJvXjgd8q!}r0~UkqW0y4Vy+n3swn
z|4NRL5X{g^89`L`c}C$e4*Scd1a^J@UMhGq;Fj-J@ZlUYwx7E6oEi7e_WylA|AAU6
z-9LLwiN~)z;4m5I&CH$@r1Ri<B1dz;y*r(0%|?O}l$Zlk6r=^#QmI8*TS2whUSAu_
zd0__Ir%ECAj?`|p+s*feOQhC|upov=)l%GYe=?Hy*j(So<K@N#BhW(Fq0yiux^<iy
zA>oz2K(w)3L`ccC;1ZBzELp2;jPRltRlj{WG1-we)S)#bpc>HtJ;V)i?Vgt%Lh^3h
zFBrQ~$RKN#vY@J@Y+6n585-mIR$|J;=yt#A5~Wd{IixSZpc8MTKmU$~dcg=ko-`&V
z6C4mwN_i_3>bm%Ix+SF!G@~<MX1uZ-J_c4PD!=WUKg7+|k#kiGM*{L;Vmch0OroH>
z#HSJc+SZH0$(LDhCeUUo<VWQ8@8jof;}g1nC;mQLb=8J1ChwX}HPE3He&%Dp#K%7M
zY5we={!{$PANr$&kR9bP4eelaTY=^JTZeijjG3P;`QcID2Oq=Hn2@Fo?hiV;4sod2
zHvrxG4u>x79yj8UGAyO63u+|KsoxG~E|}W>yumv&wIMe#*>j9yNyXr55nMOl3F<r<
z_03z{cYPFleof~0>m2em`xvNM+1=*E*!<C5`LbJ_YZTUVhb~5hQ@n69`DuWy`S26$
zg$-|P;9+w&3>)Io)co9WxmInFAjx&>-IW7=;E0lao=YoMoRqVfyg2uuYp2vkDTaO~
zSN@o&VHBOX9r%diRZC)A?k$GA6<fP~kmi6|DYv(jwICSD(i~2~k`4p&WIQcZLtE98
z9!D&6tIa>Rja5GVsn76hU-}BQ4&=+;V@|_72{Z_;dJ(V%yp6hRG;4!Dv8$Ix;bU^o
z?h2m^v=?&*41rxe)w&f7sx%c!GjeC$R1teIur*o!o`YPB3D9jXA3wMAeJ)AivMIp0
zxMggg_omJZb9*k?ixPjHm=fRcp%3z{-|`U)AsYo(H>=@CH%C$oETwWj-&58z=hK-v
z3ODIUjtL1$shv_*R<|7=a`RRiwV&+V>y22dS@?JLV%pyL$jlj!kr819YXGs9sb8X&
zzro|>)C6@9p2o#-sI8f<QAFsAQU<BPD_9WAT3*gug@#w5)eUas;tP*4kaG6!+A8ru
z2GbB|XPdPf?ANHZu$F2YOPUw~Lwi?iFwA)|XS~DJk?VZ$s=3;s*8Ol02%($e!Ec^T
z<BsQ=3tPa+I1t)s<#ArpC<dG?Bd}M_Au!D|f{h=R`;*VOm7J&T#s(`!21#nnsK-GO
z5LzvEAcsJjW`g!@(b}Qcx5aD8gc-q7_n@1EY9nFU;u_CCcBZk^tg-Xx(_q&=#N>;~
z$0P6p^xC};>1xEe1gvY}&;9uy;R|2>3jg#E{o{P-gCFK_ys-|1HeS>gn09CJ+xutR
zd<L0m!0{ux)*<5HhyTS~aulH3Z2tmM#JffvNDMpttlGBkbBrr?wz}VAj*8fOF>;rV
zVrK9Y9Jv^rilK-Z#HT?98r{w2B@fgzz06YmntN<AD<7dD!Z7RF)=h8=?xYYqF)4+i
zF(2K76G2<24{hH?A+R4xVKb^R;-Fc1Z+n<FC#VHtbW=3%E;~XGw)Ru#ekdCMKzgx_
zT4A{J4YZ-I#hfa00yQFQT@CuL-HVN}H}_7j3+l!~;(J7Q(;jUgG17_^=Pj6ZGEOdz
z+Bfo|MaNgxy3$&8M?0X|8h!Ts8K3#wuk!Xg@7eFC>`2Ox(>glJI7s)oycNxCka==K
zx4IjjIXyagko}MvBR15|idtRr+sr7)8XFH~>dM+I2@*$vHgxy0S6BP<5Ddu(&igo*
z<{UM=c^F;UxGQ|rF(T1@4R1|c*fk`3Nud`Af^5Ds$HRf&^*eqmuRppWB3?vw%DNyZ
z*N55PhIwf8vQo>Ls1eb5-&@D$anBTVWaB7W8es&)Rr3}QZ$L(>?rtRHT&q!Fb%o#~
zG;D7foP*!!@lwTr`F}&62I=nNL;3u(s@?eQf+=k`L9xS3b(=R0o<LKV&w*RMCQ@z0
z<WZ7c{es{caY$)&tlc|kn@iflt6sGVN}y_EEuFOqDv_v>CZ_pry0ES*IMdi1=fqy`
zf`x1G&Hj1)XS)|X0~!=R#8R|FYb{3l+RQk%8L=XFC#Gql_s()!%(W~=<28se&x4qp
zM}rX%NtqDhMhH@Y81kmVAV9=)sbldBqYiV%F=Hfgjkb&>Fk*kl*Xbv}?Z19dsbb_i
zW5!AhCixrk!%=aM1+~o?ZE=Q_5b>K|tdsfAfAp{MqksL!`7OWcTlwwZ`JMcuf9MbK
z(Qp11=4oPd--F$4#h)L}m%al7HXKM*EB^D7LD1vRRX1^1--@H52H6WI+5QPOe!>gY
zz9B#6<vPkE?BBhJFqcCwUeuVs!}w*ON5fPpro59!eFGeu-*zqiyPdwyvk3<JhUzYL
zQbDCLro+-Lhe32<4lA0JD4n%+JD@!sZ{t){-$Z5C^wUStU=lZrsU7>1=g|8>QEP>D
zCC0Xm4SP4>t9F>MPG5C31<G1HT+F*o@5(oHvR2k+D1hcn-V`K5!rBn&J1XS|@<1Ur
z4*<T(7et1gdfM)D{B_1?tF{%b3ydh^m^iJKU;FZ#eDRB4;=C3l&A{MRnBR-Z5yUO!
z!`@9}*KkuyiG7C_d}gR$tJUq4jieU5IEyK;IN>#=h@!MQOoDLBaf@bGHCf?$`Q5M=
z?1RuZB*pfJHb;MOb0CJqvKH=_x)lY62Jd{a7n&pX0^UC2*zCqXoa-TBdmj7KK%OT4
z;2-?`{Lml$LpH9C(QK=0r9``?IVxR+sD)4q(yhrGdyF`@O~7oi2d72wd%_x1b%#W|
zzL)`;MnitsrPQjp<UMR31ONL$$bWyOz{{Er36xqlfBA8NjnU|a9_ma}_t74RL2_O|
zr~8<}E_m|)Nie6rctKKY2q9-e%pTm~UtaHYwS{VxSw%wwvDB#y#NOP|$}6Nv$&OOY
z^Ta$`;#AX4R3rGr5H^XIjA{~NgANvK^h&sOM>twTJg>no5qPH{nAPmaJqHNAxgiMF
z8Pb@quRzU{U|1}NV0O#N{qsYdNbUx>OK1X1ZT9DKZ}2|s!9uE@-e`w)h%TYLn3D=@
zhtVJa4ju3~@OmwFsMq3~AjS{y#VlLUi)@I1dN(|D&EnrNxYWlRx@9@@3%~qve&OSv
z;6MJ0e~CZ&C;mzPr9b=6Gfk8IfYklwlOZG@Q10$1X7IkvQzd4$F`n*HBxG!a_H-(J
z>%6zI)mXSA)kTeCAWoiwG3WlOg%JdC=egkH78QZ+85E2z7AImxLZ0nfRnPBmAB%DQ
zM<I0a=DuzpqYK-1UDRjN0M%y2r7~$FXeIPw4D1Nh&a&F<T-0Q4!+{P*8N@rQuIAbp
zVkRYprbakc7+qBGJ*IX`t&8}<>+iFp^wL-rN=zggy{=Y-HZVs|*}Oe!F@qZ8DA?!&
zx|{AmNN#ydaIQf|`W~mP5(&}aU)a_$z7LLjJ*DKutRlV9O5NQ25V${`IWLW`eEBQX
z+PFV2+~40b(6hnk86(*kgcPmY55X<pos%(&W7SR$(af;MeXT>r76?f;gQ)T7ThgnH
zddJp<ueE{Mj)Qy1ddJ=a0|{U+inh+z5S>ywNF>dZh1}^vj*0uVP+diJdFHFjH!n-=
z@fecu=+f^|l=n9N%wYWQ`~L6akNwaeB4{*Xo|7fJ3>3lDD<WuFkzU*=%a9QJOm+`O
z5sNiXR+#QN0=jvwBng%YLj>O+1|LK&H?8QNIDJp3?4E0bhE2uy8$DjC7_jPIS=VCD
zfNI^4g)G>nW$(>k+nguH%xP$}d-u8RkUPjoFLR4uX9=T2-fq&x*gL&?v5~DqR<XEO
zVkW6y7dQ7A^Quve*{qpuvyRCMkDLzF8hwsvRaQ>Pql5bnRVmOL(JwmaZg>(bfMuU?
zStZ<BBge!%XaC|HiswbWY{0tQl(K1B{aUA#$Z7Is!K7Isnk}o{3lU1)oC&vX)$dH(
zApOtB_amcL9HsC$^x0xl!Otz{a<p1DtIuIeYFz`2?g8(xw;DpmJ}q|m`S&tb>Btrs
z1U<oE#UJ^L{~4;CKl9K2k2oHW*6H{ZV?!+3?oKx<{jqsyaC?}xLTOI=vo6kN&+7GL
z-`*Rt+pJnVYDA)8Rigi(c@e%?FWJlE(hcOxwDNX+1}l5aIW77-`+M%2_E|dH_q83I
z{(A4%U;JzJVfV0oiDUK)c+st!H-~D&vOCONC&Wb6$fCCX)vP5A_V>|ZFczmZRc&mj
zf`;gFvoWFxK5oQNziN`vIMObA_vS*})4jQp$6s3w(pH%q#`X~&YQ-^1ikT?~Gw({l
zNpTh@+`6fPOzvn0{22X>{mQoo!%A!-XWXd>^aMe?;3>TI=9|3v=G(ma_B+Hhv6RYM
z>I)+*MVw@&+cnod(2X%aS0`)Qz1|(!6)ZO-rnof^{o)`J(Jh93xH^rI=oTy7_k4l<
z`wH88sI_nT0%I-j@Vs5xp+xJwQP%EGl1e$94JlT&qJY3W&778n`}28=^xX@Li+zuA
z|JlBkW8~n;qEBzn{+{>y_q)FHxAF(S|Mx;IGy;diZ02>PZDX3jyO7?IR!O4FIRdj4
zTtsRGnxeH6>7H!3=PXdYp+N}AlTwqSr4w7F_lSmS`%}}s!E)rRwMx}N7GFs-<mvdn
z*62$W1Jjf_%#)uV0vbSf^22~C0v29lFtcoMx*(po1QPch+z#io&A{x2a80%vsZLU)
z(Hs@n=UaHiX;rgqx?vgm)QwORw`w+qyjcn(eRJzlKL|!F=3uezH%@os6Wa)qs@sP;
z3Cu?a+Le^{1AQ3zbic8K>@XOr8y_Nig*WQg5CRNna1<qW$^73(8sRt?<*WXlty>qy
z20ZR?*((1*IvV_f-u7Ad_N*5j?|6)bJBpXlb?Bxf&~4PR7Y$ae!bPyh*r?P{gTL^X
z{t`d;3%|gh{3rhefAkOkqokO&?m=l6ISJ|nscz&<PhV59N_vbp?2QgFrg2=1&~=ei
zCa#zmdUWq3c7zD_i(!bDUThZj*8?f>`@X5cgu{OJqIeSR5Rw-O>Z7Ldgo8BY*DEAQ
zFLqyFZ;xTCy^**su3d;OnYFeX#AA$s?s|XYAKA5wArS+tt+MYQKDr53m`zh6&)LYO
z4E~C4a^fLnZG9BX10Dv{1FFObi}dQQk&P5H>2M%~L}|+1>5Pr_15*r4Q)Zqgrff|`
za)eYd>ESl_mEqFra#}YCYu&OK@Zz{{yFsB&&`afWU-&gX^VwhJbUM@2yXNJ*FaJ>;
z>&`<71Nm@rwJ?ez@7A{ibBtjI|3dbkXCJ9;MDJjSP!MC>o3zl;m)2crv7<uKeN)uG
z5Ae=^paTL!X05s_hf<8xp}bL3hpnX{cPEYyZb*5em2T>(oX8<?yPQ~x?~miT{lCP`
z$zp&xC8iW?tP^0J6Kn08P5MW_<s<x&Kk@^_(73(5<?twN=LEsewU``fiKZA4;W*7a
zIw<Gp<6?wXEDg5TX2kSr1VQy^`tH6KF_|1ZN+e<<2$L2L^un)?D=vEIlx`<f(?;vf
zLa=hOvGx0Syi_reC0H!(26^}xz?r+~uywSVY;<@cdM7ySF)up<9XDAVmXl%YD0>&E
z9t0DD`!_p`ZFi+DrWgq^djzrATwTKUW#Ttf^Kh6Dvo8iS{?N3~l}4b+<qbdPP9s@S
zxPaTl>_c~*tG}fGd&-#*(zdxNn14by@9sE=Z|rQ{D7*XLSXg7hHh=D6M>!f3wuLHi
z(cy3ZyeI?)VTInI7efFXX6K@k9vk*1Yn1)9E+3nea(gq*qTY6#(e3)Lifkk#L;e-*
z`=afw^T|(riht|>{BQGfzxXl!Cx7~%;Tyi;o4oVhiJ@Rsc*MKDnEk*FpwUy>)WH&;
z2dgpK8U6<NJvFuo89b4?kw&-AJ7)TuY{dS`LaFq{_cdgf)Lmvd=CL99cU_%C$M(<5
zo<${EF&Kya=-2OKz^!S-K>NKN#l}qm8yA}L2s3nPom#4QkAnd(TpcFcV-`z|4A>Yl
zq(l=VAJ5s%#k}jK?fXUuvaJE*!GMbIU8C6ZI;?lo_-j+Hu5Wn!+6Q>>%In0GSk4P?
zzx57p>YF_4l^D!TE(L>uqhM%)Tc5YuHr@GXRz?U;>z!q-w63I(iD`8AZXP6wx8Ht`
zU;O!xaerQ|Y5FR4%y{f0kG{#{_8ZHJo)JK#O>O@kDy6Py@cmo%v8{ls6S_m*1KG5Z
zaB0PBF)!KO9a3O}$kn+M8x!95c^j&DK@2bKRYL|GMM%!EYOSMU$%aun>-`CxI}(yN
zm6OT00ra#f8pfU;L*Q_IL(oE|ZTp)wF3Q8}14T_0k*CZb{iA<?2M-RGX1W*&*2V;x
z(1o5GQ^-DUHAC1uO&n(<YNur}<k_3z-Wt7Bq%|moc0N&C-$J+4QSu;ag|phYTw7&V
zPzY6p(oK@M+Fgrc1X$a4?b?R$4`WO)h$b;?n_+MWIJ)I-a}xxsIDJ9%;qg2mW~FwU
zb%!)_d2+Q{x9-|sz&j-GoW0wO+n4K5Q152^gBKSe64}PSfRK_CvV3G=BRf0bw<=vC
zL**de$*ZbeC>g{%BLmS*;wU2vTy~<=c>lpW7Q0E!Orq6j3W_j1yX|`R*6AYVNj{)!
z?;!e^GZF~N&5#sxNc(E}$P1^^#UHC~sKk&z%9yopov46xM!m2f=>B9fN}9b`8F2TQ
zMH&(iUU+>wZe{x;_22KcUKpbIGrRn*mW99aU;Zec{>*3iH~!cE8Xx-bH?gIZJ0cX4
zzy9DsIMnLP!0(UdP^o~sQI7jXb=aZ$?+@0o7q!MB-eriRxYYfFy(sJ1{$cat?p>fg
zpI;j{4jFQ%m?ElS^GvYuMcRr?+qA#F9SCEJOjG7~JnS&B2syaJBAq}rqpf6f^iHGH
zvia0YfD~cMLJFp?II|+7821BXo@a|&C(w)e&1)O+<0H^4Y-7Y_IWt)0ama;xgD6Z{
z$ONu#u6W}cKFo(d@~u30{3;<v*3*fH$7@n=gjP)6E0|J5mAbC1wNhK#oDiXV@YpCw
z?9)r|P_S;K;Au{%1S-PQ=eK<P<DcZsH{Wr-#3j?6t#cp6w#WdsY^{pFhTU^s1mT^3
z9Q#%Gg4CR14FR>ot-A!!rfP0*hP_Tq(VMJ2jdbiWThzCR6Pay%yV(pd#&u)=(yi$b
zKTCF*ppRVhL=7&NL^#)qo|UpzjyDg82||kGq;xMH#)z)^cM(J2Fi-TV1WzZm_v!mB
z%A68C!)vd<#)m%iK^tQ=(4d|>15?UqOhzKq-6StM@h_&tGzF@H1j{1XTom1_p_R&V
zwqik5E!9@UPy_7*%<Xod39VL2?No)eE9cr-n__u^fjq@cz8w6_e7}#E^BD+Oq^-lY
z%@6%8Gdw0@u!XdBwSdMB9YRQC1PPr`3n@>eX*Nf%Zr!q?n^}TtaGwH$iNbLE^CBPw
zv+J}?R;_F7O6`^s731)f9$z+l2dmX?w!Zsq_pX~C_R!o9ce&m>q=;y+7)b=lft+Us
zzy_lVQ?xcSf(X&Ds4y&Zl`!oxvC)l0Nj|;~q&(XJ;&QYYCzspoUgN4(3FoQ!-238-
z8ZD-b4t0m0H^|P58uf!`3s7=N42}|99{86B|3FU|hn2tZfDy%uI9uGKOXi}&u+hM7
zax~teqb`GRsLCfk@p1ll|NH*~|I>fr&+@x|$M3aBWibt{j*jATB-9P&H`~t(vmQr7
zRBZeb)3lNKM!r3Q^CE`*^B8aJXw(QY+R&Q~fAryvc7N}e9S2r?j3U?WmW^=KMTlZ!
ziVjFn^Gx@RRM)TXVjzUToFi9<2?;`V_|{5xyBEeNuVMy0y_+6eiOiFycOo37iL1kb
zIZZs1dm2Ium;;+wvl632Hqz-@v3!xpVVaq5j)+9oQn<anV;!;GVq-#6CpCHx1j$@Y
zGY=j<;^C{W^YE2ddHCqDaT%_zSWDrx`{!tPvR<_NT|zRC`I<b=Rvp3~WW40-!n^{y
zHzFbt4T&BjUwQYAkALhF{K_YOnY-KD3*=%fBOiQSt|??Az^Sj(z9E@=T_5~kMv<G+
zNMQ}WOd6(cvpr~YD8%y^T-tc~Xx)r&+yHcwVJkM1M>|!wG8(~M_y#v?8yR{|#Kq2f
zT37N_W=duzR@g(!PNkA#BG<}v9NnK4HB&akD2Pzj)leFg7@)RF9M|!3pAUORNNb&S
zJzGIkDoG|RZ@~B;W_b_i^F8&vQ10(p?r&MvYJ0ef6%BY(T%ACvCP!b(%Gw4h>(?xs
z#Jbl?sTF#qu7$QPv~}fN8t1AkrLojTlYkCxhksT@85XlmFZRG+=kaoCf;bW910hXz
z!`EUnHDmg06^VwCAZTA@5C{q)cJbo|6$XMfS}$lV(27BA`3S)yF2!bmTk;=x(O?Kb
zKpjc2pADj0Fp<sk*(%)T`co0;tr%>Xk08NhI=v#&h>{rAn<2`kOO7OlAxSN(aT%f;
z31J6-5heZNdoDR**21kD8y_7x2v+R$)~OU9wZOWbwnN`^>4s=*B%L8s>y=IGuG{&~
zcBpyxJw}LQ1h5^xA?>VGh7GS50Rc~=d~wsUDasfbTpVnJ;5L3<MnSb_fDf2<cZTGm
zSQs747%}VxAHXL*^=baK|LwoQ_x$ef;a~li|7X1R`Wp^q%I4@e*tp|;4I`&fuyll&
zCSn-!KKH!$*Eb&v%ckh!ayRcv451Nsb{l^8dq-pUSl`g3x>iRg43*GP%-L-L?s}NE
z`^5{1R!jjh?7hQqn|#q^<m>GrqNJeA$-46%?OGaY*fvy+&3Vu_M~Pfbk(|u-T!6zo
zbA5G1O2wGvwNeo@M$&QScyq;JHW}Tz7LLaQS}Rk^yz=TR9IkIbgwy?r=g*&0*0UMq
z$gme@<9SRmb2TM$o(UQ#>&kLESx8&&O!LI!*Ip%O+Y|8bRwWxD%+b7@#bwh*h&Gha
z8XOIItJiC3eEd_N;qU+aC-~{V_ftH7`povXaUB_lxNHZ#Y45|v+ZiTj&XC#Gy>Yfo
zK@fv%vBJhD@N(YQZtqx1gRl9~VGb`2wvJsL4c_0I6;{EVJe_ndJ_-u3ANXC7pwy;R
z-=sQ<7`f_&&)QTerEon@2ALP7v}XSO71q`$^&-#=<D%r8NYm`^NeIz-Xb5x8^z}^d
zN>2D^a-`3Zyql{7_s{M?AcaUTb}+BYiFG|P<(XP5>vH1s{Fc+xTh8YbwKY2vV}h7A
zls5*k0;n55q&ws#L2Yz{+iSH(?F*<WUe>ixmW5UeOKmL8VxOzpSRlqk$kyE8qgTlj
z%+nUV{2M)9ZcH#=A^AwmvvuMjFo_#)NN}6aPM#-n9&$k4BsS4JD(PAnKJepV*TlND
zI2|BngIb1H{>5Bf`p9qI=jo;~ws~+;g_!Mm6q>sZMn51`)ph)VG#?QW%dV~kd;}1>
zIvHb!hqaasR+Dj9;Pa^NWF238BRf5&8DetRvPx9X`VIo>GrHb4__B4kI&GDdCW1s-
zSdlmrrx}EefJ1AXPN!`I^VQ(Mc45c+*GppBIEXI}&&z*jXR6whya@Z*+U>kh5TJgr
zj_-CL7G4-`uRj*7mvrOJx-GK!HQZr;`>vPe%-{LDe}{khQ=jC2{xALW{J`(~eb!0)
zXTuAFK@{BH*Vqh2&~94f-i)lKes7H|d8_9Ao^sMot1j#D(ad<RHV`85QA?|U7cjn=
zNpR_v4i!VSLLlK?Xmf3Nql#=Ug`?oGECu<x9~0cnk;4?2^BBh^Cy<Gc2F;?pVZ<=S
z3gN3m;(ETa@E0!@Q_iG3xqY$yzDgwK%+;gE%nx5-zP=&^Qwc4rSpZKlar5vY^L((5
z^8Su>IdAEaW7aN+b4x6pRkU$lR-Qk<<y<QBVJ1&T%$g5ZTs?fq@%oxUHce?Fq=_zo
z25%ZXP%W5p!8k6ivO&1Jz2&d|mH&#r_TT(X-hAg>?(a{G+ylM1e>a~2e+;vq@pysj
zq%KkQ0$HN_F|cP5Xg7*tjItFJcHrwCa3fob!Je^VV@-85K;y2TA3q#iJrrF6ZpFRj
zqO`FWsKeP~UjwU_T|%^t4S=N<aukl!M2rXS&vuY+h4VlvbPy!1xQi)nYtiI+#>I$v
zg66?)jxgF>+YpPc$}6ut;`!|@*M|e?kPR8{N-c%^yE~S+P|hoN&!2Gr<SD1Sg|ZZ?
z8uc+x6ZtTM^D??YAhd3=0@X(_IQ=X}`>?H5*1pneV_mCrXDa8iQf)$Qr;q&-T>^C=
zkzlWPI1sLmyW-<Fc)T2eaK0ku!`2CJNpaDPZ)z>9%fjfgf|$m-YomMDFDXJ#Tjz@e
zrZmyoY8`X1VnDEsJ4BPzbvwO?s7Zo)Getnk2FtpU4!)U6Or+^(v!@vS#u^HN=GAFL
z^5mZ-5Q7`z^lk-2Gp%sUv+G62-xnhuc8s;I-3V|BJq~!=4LL>!93q0DCsC_yVBOQU
z27_Ju*8Qjqq{T=rh*yqg(r+6@x6SeT>Wa7CeeYtUxY!*0|9xB8;IK`Ww;A1xd9^J*
z0}o!5*KBAc?A694A~80r7dQ+9QP2(P*$M)NLSSq*<M;hJbil`rjf0?_x8Ht?|M&mr
zNBHmuUgcXp`kVIGh{MNV=bYG_ID6+}G9Y`JPH&4BJ;_Hc(eC9q+t0S*oQg`*hPs_J
zo;|Q%<1wD_B)#rlB5v(EQV4M=ZX4Z>RAU>pN(dQsL#WN3xcl?A%IWs@WtQsK+~dQq
zzslh-ktc)5^K`I6-{ny1jQz5lS<2)IZZq?A;Ogo~5@9Wc)*6T7mAmt;9Ip?C?xf7&
z`iA3!S2#X+Og>%{qY*W$TMb7EOw;6tVB`AghFS~AC=S68rO{Z}+U~V-dwa{<Z@<I4
z@4jae!5Fx?Ir7FEALQ!#0qHP%7+5Fe12N6+8{v)1aG=DK5`!;g0>AjNPx8Zm{y*X;
zfAVKo*Xj+3ALthgaO-&GGOAa%b$@AYwf(;Rq8*IRm5R}3$$c2PrMMOPn<UX+YaFVV
zWM&a7z?!QCf+KG!1nO`m^zK?5eUKou;v4|cos6qv9}rRu-Ys9OFHuL>tdqW93v<fk
z7&%T8=ThiuhprLUbt?uGKMQ<MNs_ka>-O%%beJjK4#AY%6T(rEt%#DyJk4BRU-RJl
zhB*ad%#^iIpFXkkWhvY}ea7AM=d6p{T{|CRG$|51Ut%rP#hfC0w;To|r;nD)OqtZV
zA=QJambFkyvyp;Y(b_|F!>HtAsYVeXMW#G6T|H!e_~?SK^6Nidjx&&Fi(qv7VUuB5
zb*E0uI9?qG;IwV_F(eE8Oqn>%o)qVV6as-rYpdCFN+it(>$>XM2p<GuHb*(d_1G5%
z8g{|m@osFYF*zKoNVYkk3@F%d>Np>XaiSF`5KWm7vd#C)f-EPm787ZjNO3k#1<Yhb
z)Ew$I5`7TTe4qm{xHdl|8=E*RECMRCDH@z?6MT+uu2C+lQiIl7H}Y6t?x?+5H7hQo
z6u3Sf`M%%td-&`ZzVPB<{leJ5m$i)U>SDe=W_V6=qq{TNsN6$vzWs)_Zyl8B^WBD7
z!d?grj=-3wgO3Jq$bH{zE;lQK!bfj5-nQ6=y8XO${U7_~kMpnoul{xZm4ES{;~)FM
zKTIRK>~ttY#ysCvSri+`Y>30y^t!tXI`6=Up=TR&Ik~cyiUjKtLCs(Wn-`{(>56xM
zvsF>*p6v?SErf(%?DkGVVm7Y`?d~*I(Y`{O4EY&Jop`s@D!r{QvsAz49v}L^8_ZL(
zf~hOVn`_c^0Bl4@Z<KYVtY>OnY1IzZe7xd#eM6AQS}J8Z?GoHp2|1Fc1CnO);fj2`
z=6L;p!{M-@H?{ekSnN)&tx=l!2OQ>!rb3UyrOxk>*8M&;JMF9M8y?qAh>7!hrqlvz
zfitNTS}#l?5TtKD0VxO}V8kz1GX&H(er0KazyI?e=l}M9`FD8p&A0c!BQJu-I<T!J
zY;5SgeCGoE>KCg=cL|{{B=OEWM<M6PNDq{qVZ9X<eX~vu`~EQgeq2-6)^D=FAQ$O}
zcAyRg$u=64@p)|YkVGITrH*k#AVi@w-H-zdG}xP%)|IE{GdI)B_3^+w&#dc0>-~aM
z?)z6a65qzF*kqMLZH?vj4m5CHE3Y07m~(63a6A|U-|x7-y5iA;ha6{<)km=r<N3aT
z6OmhOJb!k}`Ls}0b3lzLAu*D3wvkd0w3s1MUCn~OjXlDztBTmzp!JRHXwpIKtm$sc
z?Vra!vndF)V`-{6lO8|j`1lpBA3f%HyrP}&-q#v^Ne9A<tb0`iBXhJ?2?D(}+aTTD
zFLoyasS?tGkP?LKMmkFGg<gx1#U`7__hvVBoXpgr)wYE}_?tjcQc4?))SaHHUCqZp
z+;X>@`3tTbsI@wLCB7Nk4cDFRD>2d#a?A#$dDYv*$Y2T(laV2l_ysgK{dHYU(l}oe
zL!hk-qGk;`B0IG;hcwyXI&+Nbjv#b;iKtiKz3%W}L|N~i5T_&QaATjht%TP2-rxD%
z{AYjpFSD$E@N8R!vd6)WUBht#9mguZ$qbc5XJ^m5X^4CkqOie+KCkaXMi~6C-r)mB
zf;Z%YAq`Zv=i6Y+pfFSb?C)26rf)Ten+j{^?*5*C>)-jmBft;-fgf~olI&jGYMZjf
zPZuY#iZgSI<<f^lPDb=8t#ZD9hN?16lig!jrh<qP9MUgkK~kc(N~wjI%<&Kjq%@&j
z=zT?ML(9ofp=89l9*h@4Sq%kHF{zO@f+R;l9Qu@C_j63d^!|)PzE&TH!-1>W<WWsK
z$K%1qUMgr$pq(~l%IeoZ1yUwoJ!CpQgb>MVVYz>X)|H&IZMO4FnvUj2;L4lUswMLI
znlJvE!a#%?%juq8R}PbfTGiG!e|zmV64bUL475({J*Q}-uXV8{);jfkqEJ~{u>{hX
zn6IwM^TCTup)MzuwHVn^B2R7`Kk;{ehX3lX{wU9$KK-hqcpR8s7>oTkeMbStJ#O>;
zA-DAo;DvFV?*wX{_mqO^&aEq9lM^Fud^5a7M$9zkl<4HaT2?D=dfzOFB@GF#KkX>|
z)iVgxlM{X6kCAZ<5o#lf(9jDWgK{orPux|J)_g2&I{MN&x66VGT+cICha*k99lHJ+
z0|6LZ9R^b2XV^5)yz}mRq?~#9@P?2l#EXs)6J@z4gvj;bz|HXrq0>t<6tJ2Kt5?g-
z>1vIJB3t5ONRgaP>X?#|f>jl5omy6Eb9Gn<*4+7C)x~6)rFKeh-t~v=du3V8><%Nl
zrfFijx-s|KdFJ68uk-MY4{-hPF~@l#tFqj_I9mGu_jsvdK)Vk*aJV`p>ufL>NP4$<
zQk+at5CpPraQpduN07)gUs*0dTa9s>6CxQRXe*TSN}dkZl}<BSR$5u<#nA*+n#e}3
zW8({0xKPIqkmNUF48g*aF5LJcR?*gawneXkgk*%qdUk7Bh|@$2*>8UHCYKG0^Dj_A
zwQ^n;LN*9A#E91FU0w0cI+%N6Owc>UAueMSdsU4hedOA^0n2=Z5QweXCRJ8KP<lNP
zoAOO>yvm0^^g%xJxnJED<yHu^k)J*u2GiTO7F@>nqK)zBNMVrVE}=@V1xA(McSugQ
z-`?K^=x&ePhlvl{Pi`=$y89mr>>a!nEPZ<m>P4Hq7v*$5^Z)u!euVG%<`41e2S4H?
z4(9^kp+P}t-3XtxHp+6g!UQ1=qS<07)YvS7@1&cFkSJPLsH<g?#~F>uy7_!`PP|&N
zs6dRUOWN#xd+>-<3(0A#p($8VCn0a&8QQ%#%@FNSYyImR8K5;(qboIRtQcLf;X+I6
zBoDKb+RQaFAFqhVD>Mi>2B;HdY_`dm`C}ZgmNT{7)0RNH%U*M>7KWyFc&*D~n>me0
zr_2+CNbTm4KMoPJAzA<-PX|lYs~uIJxm))`Io;CEP}_o53sj3Ku$=F>zItG0I9<8F
zU-`<D6TkM}!pA@LD}3%(KEtnk;+KsKJq|FigL80#;$Bu?rdVD+E)_lFzlYz)HZ~MP
z15y~A126I&nH*`Lrop%G#2i`2USVshR3pjSp=S5B8Rf8z==Rr#VdiAcvoER$@FpUO
zFh`+uAqFf#QSGPJ+9@8D+_fRi+&fi;(v|1uGo`Fd$q1$?=e+=u%SOe~!i!?+FwIOk
zQP#rw_Kry)T^*1ZDeForM)rR9y(jqG!~^vjby?|4afzWfFe?7Kfh3S)GVcrVds)mK
zH+o(`T~`|u3(i9ua&hnehndlQMDj$hm2^5&_12K=u2#rAdd$_Uuab@j9zA%4S6+XE
zN3XuZ@#<jY`t!o`_5OV=(U(I8RH@5}S{Cv&6NUt?b*su_pcBxt+AO^{>*7M9G+Q`*
zU8wVcG|g5-NI+w<i-EC+J2taVA*0c{z1};u*e0IRM4k?2J=z-*BVA0M#h6F6LQ#As
zAIxaR+5Z}$Xd|egtx*?)g%9&=#Y9;c;XADsR9BCe^?7uw&Q!PgQOcPV0~wI@glh8u
znzp4~A%QR*kbH32TB9zfP3oq-qsy)N)YAZqSq+j*H6Tg|3MxoDndY??l+NSp1HbvB
zAK_O%`?+mnxI_d*MnumG$2wb!zFo7`h-xoD>7(NJ&Ri6$+VI)FZ2!KE5-#qS(WM}l
zh=gx_BGwt}2%QQq{Jp4}3~=ji{O{a7aBMA~`PDD-ul(!(Cja99@W0O=`2O$r+zT5m
zbg{9AY)TF#q=^uWn-OEM#Z^~w%!K){r6Z2pN!0Gn9vb9{F3}YxP7Di)kcK=FA1yT2
zyL+@Q8(AyFK#J}L7p;qFbumbqMyscdeGUPwCpuQdN<F{SQvF(coX_`&RwPH};}QHG
zH)P0C*@h%Dr3vt{lM{q-8)A`8jL!2?a|hJ6+IVZdr!Dt}xT$WzV71nvlx0~K_tNj=
zB<S(LG+*0auKR3tL#)JQwjFvYl#ZFYjH00%x1`lGt*nHYiSa;RD=nP4yFK&C&%eok
z`wO4uv!DGuZ@={p>vDF8c+BT7&@%C+L99F99ryb7zAg>UwgcDK+Ex?{M9+T6SpTS_
zVD>Z7Ggz7<rqLtnLA;r9;~)|0FeSs)1C9aOJbWJmjB%$w3)M#njP>qg1@+A$L?Njm
zk;5HQ6vRgvQS9LFBMEmn#l}c)o|Ut}Ar}sZ!=@}6AxxctfI70`me?^RUirWWc>3N`
z)@9-L?rbiCxzScD0#W$YFMbJCQxb@TE$gI|Vtf+}Wd`G691b(nlsx0cP#BZnC+2CQ
zPKk3e)zGpOF9s6xA(A|MAV?s)^fX;RCLXToWhKkothaZb>~8z|qnigjc>I{_hmW{=
z^pJ;-9&vMhB&9$--J#1JdU}6+qnASl%9*aPo===lw_F{L<mmvt(+ZYgN1~V27A;~S
zJ~2b9#9ENP($AG;xiezk;hI)XwDo?Q^#<2;w+>pX-E3ms{k5z{GAVXwwOYv2Og`SA
z5=>KE$JoLbSqRj1rPUMj!C-%_D?tic&#cQGQ3PqlNIxi~i4e0vXPz`jnv9Ja8(J2m
zS7K1x<i<gz(2E7n^je71foZxzLU!`n0x@sC0hF^93IeSamh&C8oSCMHdAf*M4!m&E
zTJ&n7)rC|~{NC^SHva3s`L}J>HVOe(8I1Y4El{i|8sma)8R=nkrM~DGI)C4z#yoJ>
z0((syf<dqj;a`>;s2z+0p1t^E|ChmEzzUy>?qdHr-5VdPh5YwM(xO3>?>>3T|LNcR
z5BS)pKgXZ=WB(mK{Go3^2Fl{m$~aOqiUPX1+o@WIf>wI5ZY0Ht<O&Z%gB!h9pWm4b
z_3nJ;Zdo}#c!kHWd;?)JiI}bn<@pm%Pv4<8Wtt~*E1R+vdPBb0j6hh|lcD1C6{2b>
znv8wjijuM{FSAs?<{o8T$tieeH$3CpR;2d6A7n5*LJU#q>zS_R*p{M@LS|&0w6d7B
ztE!=SI5OnYuwI+}uN&N-Q}n`vO;(qZn;?0@p4A|Ghb$<QZ`8fGLMmW#-wwrjAw5J6
z9%1~>JJ0wFKk>KtrC<31@4WkzbtA6s=o>Pkaf8?9d-1)m0UsI+#n2cF=<pl4EboVU
zz%}|_Tm*;ZQ``;Qg3ta<0U(>%(kSX%t%eSGBR0&0EJMO2gJF0N?8TG%%Uae#bCRSz
zg4lsyosVIk(|$KaK}1Q>8kbfohba+5L)S)CVUXfY0b#|9ib3<$HX~oxeOxsw3VaV7
z#TUMZ%{lYh>#uRTzh|Cj(v&Rql$8(#b&rueO}J6Mok?n8R)gr7B|4cs5MyAT9c|S?
zq-=;I7ugA{y){<|tmbPV!<cRm<{_nt`N0E1K2q*&4?dTbwH5+lvmCxoiR*b{LSZ>m
zv{BkhIo(li@3?#N?0v1#mnsIHK6#t_5Xkcp)yD19cbMmcZ*Eq7nTK~bf9RZTBP)fW
z*{-#^gvGkckjy{4)<Rir!<*)r$<2riLNQBU)7!gpp*7R`qr_?>iZ~yr4N6~F*23}n
znw)0S;x>B^7V3lZ6;^}BT;E@7p{-}L5>3u!kl>x9F_tAw-VIcwub>s_n88hm<mrf$
zkx^|d_s?0Edk=Q<sy}9AdH{4<IU{`}`w1}{Tr9S_5iw75bGA16q9qY3$OJi%wDavB
z{Rk-twd{_9<9UX}$h!UgFgEj$3thEuJwa>-u|s%2m^+swz>zNLTKzD>e)d&9NQ$sO
z&#riotr!s}&TKjN{`)pQ7`6Qc3NfBv05`Y-xVyjSFaFj4iogC7e~*9jfAhcOn?Ll1
z9TKe~DI;|1x@?6|a2QdwQ(el0SR$E7q1466ZJk~gYFms58Z3G^v_dNPw5NspwvZ0j
zmdR08Sni2^u@3TFV4etJ1`TwHl<Ii}&Rqy8)0TBBxN^!Cx+Kx0rv3eqfWB5AA`nt=
z$2x;ocjQTp_v7QCw#5(!+PWg`<a6dk&b}6<vz|}Xx+2(G4PxeGT?C>RxiOM*r1t@V
zyAnsc74ks{Vs4ZvIn*1((?ZSWxT_L!Af^M;tTbQXV+^;3kwj~U&5!-vkMTEt;%9mD
z?I%V69*6vf#Cc&ZgM+oJZ?^T;FFgGn%D)^Di0p=FR!QitTDkE22ojhaxeR_X5AxZT
z7n2M{3kJ7rT!0WAu~b7T%rdu=Lf8-ZFbIRXx#;;zTXXK8DfoNo-Mk4#fj>A=gU1!z
zx%Mzm2u3pO-ZY5x4N*{U(xS%*C*MCAO~*hb@oVSG25`xDGsgckPfXK8#`ttyY(Gh2
z+=OS(Z+Z8<Cw#*PUN;p<m(6}Z_}-P~Y-5-d$=OJQf>27OtSclNHHDN&^Tc#@U_Ktq
zX1i6IkBpaPCFYDKNYhN1?EGxy%zAgr>Glro%5gq$eYheY4`dOdc0w!2xpTU|<$2%T
z5PPq@b^rW*t<je%21F7e&&-crhY)F}XBMh39f{F4+vWZlajyHqYj$Av-i%;39SA8C
zqnSPoNGHaGW=jQ}=c9jt2LE;6q>WhAu~^3B+2`GihU#Ko^sSMCpvnGSWu>j>?Xd2x
z(a!f!S0nMoi8x<b*hYxZ3T-`EzuBtI(u#S3%lV!dlN~@<)<Ri2Q!>e6E34gT`Ct*M
zOeia@S9(>uMhbPgCrKjtwP@?=91C+F3?ZPcu&ifLGsg)jn_MX*YOhX~v;(xa!UrE;
z^PS)J&HT(Se+v9a9_AUyfZ;EsdwAmz#L0_(ZDl}zVeix~+j5t^nd77GKkwVaPF!;0
zCi@1r9rWI^jIKa^lkB~1nG3+y4S73`zf&s|><RaR#gRyDg=bHn^6&ro|Cs;9pZy<j
z$bnuAp$Rdfr5L9~v6(+3Ps1kH({!+IZ&?hXaz=91OcwhLrBozM9FI3#AFj~4vesMV
z+>khvT&^4sGujMJ)KY2Ej0fO_E!wDcv1`!Hl&W+kM2o&gqKz$TG_u{-@gq+W3;e9m
zm9;E(ulP8pl(Q9?!A3mE<GfofrqW3V$YH<FTBo&zTF%4}iEi9uQgQ1T1A-<tl3Z}x
z;tk%W8^rG%9YX|MLg)9@hG==5?Q}SpVyRbJS*X?QXmiTkF3NxNxBeb~<-h)0oR)H7
zJ~j%Up>na2Q5%iXxaWO&48yJ<yUOS?mfUW`us@%_#wZ9F<+IHiXsvqKm!N%kf4Hd`
zI00PHZfh;bPzC_rsh3_i^P{Gnh(q9F?E0Z)mmtu^_zr#WA4YM|=}k9OXQUYdFvx?R
zeG}b~XN-iVC=7Sb>h`<nAPEB5u7t`c)cQr=;1J1p7}UxA-HCF4CMP>|*JAP6b5h$2
zQO@Uu^Rl|9#Y7Y1ti=$~-rYdYWr@SCU!vzwRBA0Goz3DoM7X}@`r$*$YGdn^3@HgA
z+9<>(Hi&cWPPex#=Y_jx&k3r0@Qv4aczwh5@rpFhTwPt0b3$sR74sojMNM)aB4_P9
zukVj<^isvZ_03~$Zmwxs(HJ?r`ayI(+szlExpZNgdy!-ZaEi!$upTu|e&_~zwW?k_
zgx0Lu(n?=D=kpyg<?RqLwzWaSc0ibtV6(>z0bNeS)2$y?#<aXY-H~&quP3USOQHyb
zUPuC(vQ?_<O2=Z2CC#>QB_L8Awz9-Q+rLecBtVLRQo7%?322^xc<b-mi8L{8(Wx;8
zl+1GXg!N7t?ERFo0a0!RY;)$g!)~n+Lf@b{^@3oCjON3E-}PPJ!O#5CCtvu^41w1T
zHtaHFfwozO7#&jGi-DbZG=ib}4r7ibX6ZbmqVPle0tHb2u7-o&C@998;vYQF)%g>)
zFAhfZKB^dn#&~&~1@7I>MviJL|LxEGEI<C!zmp&QeZQL&jr^xQ7(D8y#y}=i_vmW{
zYB^gZsaU5PIiTxm2$N(Znhs|*x_Q`#qkD*-iStaFCqhx{?xUHkbQLlt7wxq|H;zWk
zGq^n}7?ix6iro)oHJ^@`S*Txwk1z)$LMaxhqFQJ<6T)O{@@DR8ty$5Ya%R$_cSaWK
z)kEBz>RKzT>piWmNIbX_LtUmd4n!y3h2b!1hkcCMxJl|kS*6g1oYUvq;}BCP_$j9U
z4zZJ`fVN04cF)JCJb7OD(I5L+{>#7iH(8g}M{fK26k$_EX$Xc8)oSNvo^9l?72_}p
z_fbTAweH`pq5r#%HTNRNr#CCMhD)GXAP<zGZ^{e%eU~Qs+K;|xY40@M&4h+gR9CYq
z9&X=*Cos^S5F(0QyWwF_d_Np&1N(U|z%K5U-dxfwP)cWtzHWVtoCHcY2fnU-8%2yD
zq|0k%o5R?ie2;K){%Kh_mx5gEiC2di+>&`*&tLn>n|$YQ`OU->Xeqj#ygDkvhNe_^
z<ZYWYGm^8#D$m&h$E41z$0H9PJs=0+JO^qqW3>>?PFp>gtc5@;m7tBAdE(~5HE+D~
zfSX4TndaF>L}EmG3%Xi4cSVjz!p%cMfOx(oJ%9SX*62$W18Z3<WT&o#-XKi~jdK6o
zbf-CAK*eU!n4*0NVZKJfY~Azeme9{Eet4%i*+#E6+ilhUIo!3xVQWILECMxyp)}19
zl7->+LJcQ_*%k8!$VODGA&@$xd8VudX^{|3bx^gLJ9tpm^$uMFQ#NvAuZ<`TiFPDO
zOipAms7TzX2rvR=N<gP9i@6mJ2R{L=N^O-U3mrl0N~^0$D;4@=QY%~MJ}Fcu|EZCo
zv{pjzK-<PLHqWzy=pXxi-@{-3yFblmzVIdRpd68lIplcm(H%L|X244_bifr1TL<QN
z-`Etz4V=P|yD+rO+dyyInAkr5u9v>lEsw%Nyj$9hO7^@09UJ7<{OkA)zt`B24!>TA
zQtQgU`yc)=aX#_~zvs6To28ej%ahV{Wt!h|Z?U6oC{^%|Z#93y-i-rLtpLssjvTKa
z+jq9!(_Esqk)`OwexPCYw4y*x&WBL*QmPeIL(B}uwqjY-B&W^0FjOh)iCWIA%h$J>
zASX)|4IS71D@Z3wMMJcEj}G0?I6TlNCX-#YTBs607H6<m%6djodk*o!*<IL1N@{@3
zZFH+~*$>gtK@T$>{Lruin610-jg3r~QbK#Dw!;10Er0))KgWOZ*ZvNl{nalS3msF%
zsIZynn0nWx#n@O-{qI`qO!L83gQHJdg5UOz_wH78C=cL;1XW+>WBfJh3X=<t$lka;
z(}-=q#OusPejp4R@V6D*+Lb2Cxhu0?UQ@eYyQ;gd8u`%n@4y)M85$I}24F94c9d|)
zCYwj8$7r8Fio1;lknY#WNr5u<l8b2!jM1j}&-ohkA855u*L5qtvb$PZys;3pqowlk
zPk)Zz{ky-HS05b>kxq#yYMdgm&)i!hl;~rSzPU33%yZr>pJRaeFd=%wlrzUGGuR3e
zO~NX{&pD+_bm_7A30xlzJb3tkcrbU%^Lb%8uk^aI_Qvz&OeJzW-jJpvBxuXY*VAwO
zc&TFGeEXEx&V&%C<tgFaw=7n&HIv)bOvp1S9yqTn=ev7E8!1KRrpQ_->%GC`*+vFY
z3>t4f=RiPGrmf5FIv9;<DJesk5DDh<of5=MkYLGVG$bT5h4GFxO%owS?r-nxOOjw-
z2F*tcAraz?;`7;ZW^EM_g;to7nFH0X#2hvV-N=f5&-s6gU;#xJ2j+mQs^1v37ot01
zA(5B@r4~vt{(`uVgNXTY+brL!;tT|6ZAB!bF;hzeH5=sDAKvgk{_p(>{*V9H{|D!F
z*%sH(E)Q5=w}op5uQ+p6qM{y#xe<_jn;69auvw^X5WYX(_5;4K$1HS=`Zw<ab-A7w
zG-I!Ofjxe?qW5>PcM!5K{IMZf(c>c#?c6<o!oU6J{yiSff0pn5j&B1E#wbn!jgtjH
zEobiT?g%k)I2_FjqBm1`#EjO$+AAcp$`<kG5J0<=3SC(uP|M1yYH`XV-3_xHZtF_T
zZhP7*NwA`lMyo5!-4li-x5VtvA;f$jew`Ns#Nc!IiBjV>Ck8v%$H5$Oc0-)B8REnk
z2|;PSQ0rm`v;;GwQ5!w9YWJILajiyBv;ed~OTQ4DXuPIwB(^pVeRajoC<L&cqiWlV
z3Q0n%7T(ocqpXJFr5HHf-Sg*u_&?`of8o=dSK~p1U<WQfU(d-h7;39z4=VzU=o>7q
zd@Y3{vh5p4Krf7THW$0ga`*zpFoet2--afj<Yb4-#=!R?X%sq_>qeYRxk(O#bE}%o
z_{Sr-5gMpz>b>ZQz8&;t#rtMuJl?SwL2#`9p(Go{#>jlxen)pJdeE&vD7AZxyYZ%i
zzp%OiQ1I*2$3D<^#m3+|43x1HE4p*aAYv+jXi{dAOq*-tlPB--g)hIwJbi$hEZzj_
zR>0dHFQhy<l4;3~trp)OtT~BdqX0`6v}D?t44DnVMom-BHs12WDh`yw?+;1I2$So|
z>2%NOyaGZMVeOq$S*dr?#^k+l`}`g5o<4bBOZ4T6fzzEOL8ZjHE(9vA&BS~l%~vdC
zA(UsVsc^q`>%Pu+M1*BM6P8;-j2y0R$n!NJB(uegfrOCqOxI31-_h2S@ASqM2+3@Z
zl|-qPb*-dYNwHYUU!2V0FxcjKUQTpv+&+6oN|~iBy!Y%W4<5ZvJY17wAe`@zoH<O_
zpq*v8H8(Rci&R2%t#?G%)ksFpV3lZ^)>eu|(gr2viPFwct&oXc<s)%p|4$*(%Su^S
z^VLoh)Ab{R{oC*zu$#NAD}i9ij;(S&KQY!Vkmf)P)@^yq4kCQdcYZs6{15+r{=$#`
zb@N8|$5;g0e~m_d9Oe=rBoAn5FKjHrg{On9O%#U$4Q}v0ec!0;W#r;D{B;KkI{wo!
zSjR3lrNZSK){D&uVMiq#kudN;_IWT~uyy6`_8C9&U;HTF_RW8WIT@ROiV8josP0Rj
zrU`XqM^<C<N}OzlFD8KsF;Y%<)MX*3VBC?I45^aQ^&AQLK%NfNVx4QAkMs^97Fz3^
z@1Jq|{29l2a%<hn>2%BO(=Rg}Za6-C-N=TuF&&Of)6BVkoev7cNW<n2Z<!m22ZWV{
zfj7r65pyC=(+*jOWL==6MZGI_ZhE|{Dy3P_SG*WOh0(Qbk-L3M1GQWYhn79n(AM3~
zw}I|Sse!f{>SJ7m6JL4rj(_L>`6K+w=fCJPUqFpaX9<N7(s1F+)aq&kM@A$D&P(0U
zvf|kjAw&r3#mI{V!5{)}846yDUmn(%YK0LN7kuyOjv$P~Ub}8Cfw21<2sn?%Ov3Qj
zol>NAL0aDqtL~l#M3a~s!Zx~Kw^H}_BSE-SKa9^Y%#{=)GxloF8DeAvR>f%DviaG#
zL6PicN3Az%Eq1W8?=it;mBai<VZ?_a0kQ;gGLq$#ExuUTjrA<;aN%^m=lOksS-8n&
zI%ld4H)#sl3V7K$P|dBZI}IuXvGBcC>PBdcIgvsjs*sbZU2KncNqRTb;am!*`+L@Q
zOU1V6rCQ>lBnu;wm{@L~^X%!ny#3{0wa;5kq4s_rFI5a29=yu@>IXQ#_Z1_1<%x87
zK$@?4@aPq4t-Sa4SC9~xA3bC~9BJokgUISiJKu3N9hi?-GzmO^_JrCB4<Ek@l4!w5
zH>DOjk(4qqXU^-&TAMK^!>Ck^fKyH<V!UU*dWgixaz0xWY_BZUX1rD5=D|ZF@YU6b
zM|mbXk+D`whZ<V?813-x9YUTVMv@q*uh&LXNZBF2)ntUYZoD={G$6s}^C9~|YfgAE
z8ovSw^uExW5cS{;bvM)L%^1De!j?js&>-C1J!gu6tK$_RCb!Hr>(3AZ@o<BNl^^;?
ze~@4N)Mxm_XFk6>U)k#Oh0O*@4hP2Q1RauvcYvz~e9?yxZNnQVg9szXz&!WI7R=sq
z;ImQ`b8s9XNxSPH$hLWm3ZBsejqYVDAcSo$y1|p9;26H=;xBLCfvEC}AO95p<=^-*
z{@5S<{WjBUotP?ukPk=Z<26D@k`vZsWI_b;;mWSjdNK=Y5rT?!`Lz(zM2@p<uqjzs
zQAn0K-Uhj~89_EhN?kZF_neoLb-^km>=1bL%14kXQ+gzai5Q`-XZm_Jg~HeKL&pr<
z33MqFlbMrMfnFCQ$~p3u(u|Boz?~RdE6&YvyH^pPhno_{?lhOa5w^o{`#V*7uha%n
z6KRb2v~BmDGCDt3-iT+yCpUu;xVvBY;s5MM`IXOo(NrPc4|Z+zs^mFwm?I*_DmMur
z#<ph=_5!9@sl*UiO66P%Bp+LNQM;=UM&Y6x>NueLOGIuIOmWQNy&x*J0?|1WmW5)C
zgmG2$0u9)F1jf1($T@kpuc#V{xK-z}sFCjm&my=IiEgCJvHw`qJuYjsfF5g@hC0Cx
z-=>|IlXv<7^}HCboHjH;^}?kw7%G?3<Uiw{CD9N&*)QUTW?mmBLa?#V2)t{)=ZPcI
zf)Y_a^DCd?18=-WlFS^EIXUKOM<ygPl1LTMszfF<256Pmz42I$8>3#JOfeE2ftW%f
zhGaI@3Tv&DQaLRvx2H3A_xIeNPBb+#EHOZg#(_zuHcN5l>FEjgr+Wg*G3N`_$*=!-
zIn$xuI2>l$oH?D9wRW!O0~(d5Pv5a@=Xv6A^C}Nsd4tpa9p!YQFZa~-#MAfQ;S2X?
zZf=g0R(b2~w@D%LP2coUj@J*|EC%v?&Ee5&2<Fv)yt+Zv0)B3vyiM~3G{Snya?es5
zQ|+YWCNgS7!w>@He9sgD$DHZ4SU$Dq3Y_ll5zi41QRsCgC6o7sIMIv6E1P)?XpCs9
zRPRK@^P|dgX1zZl(ICXwJ9$b>^EFKpwRd7RXtJye!4(*5?ev(*DOtQ_wQ!t}3}zO|
zv<5j&JbwI!F{>q7;nJF=R!s*^>&p3^x0odG%JqSN>W};&|Mq|IpKyCR0p6{TqQL)q
z*Lx4<t!zb!2uQ1FTQ=BTVe>R6WSSaT?~)s_4+91iyZ^g~Mff{#M1uVv<Nb|7U=$Ji
zbE|_I>S2px4mf^Yk+46<fLv>D{D*((N68`bgWvZ%Xd=cecX?}RmK8qFGl-C;Bbp{^
zQ|9B5UMpG`XpNX=%-zx?v63db26MLz5!^7y0+J?M_|4KNgD4?nuC5;P;KA$Q^U2m4
zae|m8)^7LK<krl?C9b|tM)23_Lx!BVi&?3PnG2a0M5Use^QO4~+VAXa6T4?ecvlF~
zJICFy$NWeD9kJ80b;4V*=;N`jR=^v-pjmhhiUu3q;XPva@l^~NO#v;+PyFQ1^Rqwy
z3F}k=w;}T4#71^$$_(-=-fiY+MM2U;*Umu%SuOM@#);Za)K(pR+;91T3S6QC-OK>l
z*U&{#uyF@)NPXuG#S0$KkmunDK@b~Rj_!8M<714wToJ5sQf%zwN`m3~dU;Scwc>2}
zIoNxW4gT-7QJQXwfe_P0U=$a{p%|qED>Q6|F0SYF?>+`vvu4udzxEl2tWYn)dhGk}
z_U&AJP8*&2^$R%Z(hlm+f9`X9_jms`u8yx0uO4uCbi>uv(Z?nx;ltf{wJVVeQ$1Cd
z<xIJ|qxAPoF6-C)kY`dD21}N_NwN^GyZbZmy!V9L`x8s4tYvjMJ%mt*DbYhF=EiY4
zaCJ4CLM=v~KY7o`+`+7~-^b(SB8=B(+`sjCRFphVAd&NW&paPF-#+L0?K8q*rp3tG
zDoGoCJt3{KobS0kpLp}hGd`FSQ;NL)`ZsVs-ShT4@9;?B=HX+L|Ahlrk3UE`-LjmY
z5JMnM2ikhFJbAlQXsvNNFSPYUDl6CN!1aR<a9;}N^NE;YPH?`z<#sufCnGpMe}0F`
z9VtelRE}3SmXY1e@MUq+mmmxC;mXtzTItEGi6u?Y7JDzfb2^=AWg$*84WV~o4$wqc
zIn!X%6UP`mUpjMmaK&L5@V07_v@t7G<0+WLA<#-E%?F$BcV#`_qkW~y#KSjUr_YVm
z=V-kYzUQ}pC;!|(^C$UpKl~${P7D3Q!EAqsIcjk^gEaWn?hXsrVdOh7*xGI*TUTgp
z>%jIyACT=x_RV@b@wZnWf4-AWt>_qkHwu`G!GizdfDC-Y7{({q=I=ScPAWIJx2(!v
z`I~=--}k$|iwE-+IcG|(obPVYR+y)3EHMeLuxRw<p0%E>t12g#fn^^hLCQx;>#TRr
z+&o5|gE9OI)aLy;+orLe=zX<>v@s74(a;?+p23_lIUR^DH8O4T6>)ukB%rU=$CNYE
zbU@YE=^;*AMoCK(Nevn5y4yFmZr!U%WCzr5<APc$!AB>eUWg6e&gO0ih{5v3P0B_%
zPmJze;$37+e*XqLe9Q>ULb_TvWJgls?f0JWbHDgWBSnLqxUe1j!JX$sGUB@Z?S3_w
z4hJrYLCDeM&h$=7k-B=qqCcN<Axl)<*2EyzUVw1^QlrSeoVD5ihR?ue5A6RQuu(&4
z>vm8NevT>@)CC(UbBq>j+!2j~;kLmrcm;+M)Nb<J6m5hOh+APcxJ>8LIp*C*!FmrX
z;9B#-DhT4T#L%rN7$nf~Iqm0HLNrCdB~ZC<Kk8$dx)pP!6l!Z^M+iooy!kxX{ZLEg
z$$L-v#&7y&u3vqF#~*xy!*Qnf#(KVEy?;($7gHKZB&2COn0vRl>_sBAS8L|FGR+6F
ztJk`A!tsXi;4wNKSWovX<=1%H-sQ>NJ-t+>6b+5QWQ>NI2oe)9OkCYO<oe+wZXP~h
zdHRHUx<}i}-Tgm|prWhC$b7?`XYSkEOgWIJBgcoYaeuP;Xs?y>>?wt3TwPtWoX;fb
zTwg!raQzCu>9@U(_QLaL?{J)sEKB7pU;0&II`H_7j}i_?Zr}ML&))hRS}M<<zsr2M
z;&5feq51lTln)4S|Li^9e)~;M_xHT^_#x-pCp-~>G;w(Nnk}-FSqQxQr7x0F;_CW_
zRvUNE-U2PW^7s{7Dssp9eB$c*5!VkNP?o}SK2cA%^jgR<GKI)AA2=MZI6Zs9^Cxf9
zyE3N<34!aw71QCGXZO!Mz-BT#%rp@d&d=Yqq_z2gL?elr%xp!a(St!on`0i9VA!gq
zU4>x8%G3EwQQ>&FM%Od%efe`--#j4oLaU9SN>b$qe%E(XYvbSlkN#86%U%#{3uS{K
zWrJrgy9oQy1gQUtcTatv!;bUT;K|)0cpFD}XEl%k>-Kl~4>EMpT+9S6rAGd{JKO1n
z-}RyCKDzM5YZ6qKHP(tE{Mwsu@nb*ri~P_J{2nx{#FVUCb9?1lRyJ0<4iqFX=giup
zO`3g-CF1e8>lqZ5(uvy9wwNY41n33>OPWlf(hAa*=TD!~%4+arIucU`UC};~^8smQ
zj#SHH*P%4l<@8cZ^=s{sW84t8sA`1U5C{Y6jgc58RNH1&GftNvBT>@tz0yowp#mw8
zawY^DWyP52&3xaxwr!l%%0exLG-c9!AVudHsk(27j~MWvyfFqFoDm6>`-Lz5>KFOc
zXFqS<xuXuE5T|G|<lc=>BI-r;#Yii#b-osW26-|r!Ktl|F2EM7!<gOcRwR$YVuSF#
zh#E!5SaV==^;Y+_Jm&jfZTi!E{UnF<yGwYDDA+Bq%Ta`sJYrqdM*w)SWGGG)U*rBh
zdb82UTKnbO-$osx-7_ETFQyc?QHO|Zg_61f(?8J6cu}C3H$?B1CdQ{4L3+b`VaT0J
zZ>A2CecV~xFlmI7Z3ac!iC&uh+|U0!zw>*)m+Mzv<?7W}IUHtFj;XTV-5EhOSiw0>
z6M3G^RL?(8Xe&hu(_uz5K*~09X%&r${MrYZ9>0d<iFSWa(!k-_6Arg`w2+vx1rw!g
zV*!&NM)K7|t{z@7P3FH7kB_KVGxh$Ca)0OQiQo9~QpG^OdWCfJ(Bg#eZz-qufJC~g
zNN>FQ_%)bsxP9vj+&p;1oF_yZ%l$3!aA2OUc<|^UIR_p+xZ>H9=hV`9{gubm({q-)
z=Uh*j_3kPCzH)VCyn~cy>T++h`H;!ek=~WlvM@~(Qxc{Wn5N8nx+Tpw6b0yndgjrq
zZ;-DZP)gx2&*Tuex_Xzlzw&v`>xn8to{!||$kp*k@XC9dkK}nah-W!*|KvSROXYC9
zc7=g(y1z&BN{WHo+h-iq4$UZbL$%fk(kWf22&JA$L77CU=We#=y|b<-7;vO-kfk=_
z-HEQn=5TGMtSjsJmeM=4E2bcn=g(Nr&k+$y>l}_Z*1WefKkz%h9hHBIfB%R7Q_kn*
zVy5n0guhQCD#>>Cjv49((T;n-NkyY_H%N9YUWe-apIdkD&#(<CBrp!#(Zvjf!)}gr
zdFPFyWp4yj_c`E;_Qh-)=Z16y8|}2R@S{KWxA~sm`fWVEzBfk`*yg1mW|2M3<eXU6
z6LmQg;{;(M2BFtZ?*&+C>%x>HH6KX93eEHLcj;vz=E;Ig+Unc^n<uVIgD{Z}nXBUi
zT3_hRM^j=^-gS7~yKCiIE6Y+|W~qM7JyaXL6^En64QS2;#WOt2$`y=P&@Q^ff$&&}
zOWW^AcY=$_!oOlj#1M$#K<%AU%qcU)Oplei6l$r&*?-@QU;^&5vEQSXH{YNsT`N%|
zx9>dR@BZX3a5|lLw)_RetliT{$AKQOd)?8g?H(3vGuKEh_Ps`j(oOqs7RnpqB{<}K
zNp9Tld${ly@c)kPsgm5>sMY4uOzW<rgiZQ3DuW@58+iw#Ao^<U{vaF<N7N~2_RIm}
zR!B36auge+wq}R^u=7=5p3)}yEL|zJd6<%aCg+26-HD=6c13`{jZKaWxFfMA<u;Ed
z_t3Z2E2UNz&o+q{vvgBlj7wqYVsKJ63L?sr_uk_(pZ+x8`t84&oM+NBn=(zIxb1b0
z;z-m&qr>7Usbxj9laE*Ee1)n|Rn7ibGjYBmKe*=R!4+wmNmHaPC)N*sh%CXzn%Ro2
z7){15DM2*#Q4~X;gMS86KysqT$nF>U>pxzq7<lUopQXL}3J)JX25o%#&9@NVA%E#t
zp?5;egnXdgzlW3)r~5nR>qq4I!1;8~RgB!e^=q_lhw>!K((VXtCCAS5Ctsw=kQqf<
z>pXt#!<^4&mS=BseSJgi(9dViet2HrJVdlIPnkSj(Q5GpvvPmB<8*(^&BIr@dGtEr
z;UmCE9M_K?(e)e9ro`#WsyKoB`&(|GzC~}1o0~__9+J|5kSBKrtUP(=E5sPNx_O8Y
zsOuf&eBy98^6dE&nkrY<*Vajvm2$o#q)a|uGhID!xJ)U_VqArMAYh$&Yu()I=w`x%
zays9Ur<vpNh7gSKcz<_GS)U*e9&&j6nwhWkLW-HSuJ$}JA{zL<@A}RB%y)h}KlyVX
zV|c(1`4MArk1>Jnp(mF*{!6IRVqnV->HA{b9K|jnxv_A4lj34iQSan^wQWDMs}U}4
zK4mW^M#aLuA<D(?Z<}tn6irJKv<T@hzxgIV`-xxSpZI~_!+Y<(!>LssT-{LCmBZB)
zF(ztl)CNrhSJw|{UHRe{KhI&FdG+BVqKF0W&{;K;XJa+rzxzc(uNLrvM`PDU$nJUG
zm7Eik+l-0`Q76*`M`O=xui%>YkZnwpQ07DA>-n*+C#Ed)9vq&B-b{5f47P^*AOt^H
z+^QBf>6~e=O$MBDCg#bx8Vn^XQ4%s8tY}1t$v6fnO;j~L4k4Jcsn5c@cbBRzP3z6%
zzRd#H+I`{g{p>IC_kRAD0iVY^pJVbwzyV#4PTgEZqZhBjz&kc`EGI47;T_xuz{0PF
zM@PT#5Aa{!4&!l%U;g!F0Uko|u2i?@iejkMuqz)v`S$nfgu1v%4B5s7{$02afw)XB
zI_G5@(KS2FqmvUwgj)Od-`yd8RVbw~O=05!G=;S`3-}ROOR>GDk3FZ`bqd3PC=i0l
zAIC*dHzw*7x~{bDcE<kq)eY1T%T6#v)Xm;Q+@4>Ck=j5^5o&9E@rz%i*Oj`SnQ|~Y
z@%2ny&PeYBNlxxggb+=p+$^UCRrhHDB$7l}n?S9^6cO#Tb+H_u-Z&g0uYB+WJbd)f
zN5_`WA-%bSs*sP@%=0w_A9qyq-)MDl1H6V_?xVlX<K>Wn^BvFMTDX7vO^!D=9N+jB
z>fL+P`}fGlhvfMR=_?N(-ym2<^*m3UOXb<=%=7zu=9CDcyuGf3km<d1fA^euns{)X
zS<h#l-ah4(2PXG>{?3<Z>&m*^QOcQlK2TfZd|o-%%GL2ekW4z<5RX^P5@}~E6b@Gp
zxXn)~>&e*Q%`SRSvktyK962v%)?Ru3<eo=2Gjq($`N&f5`O4=%&$%|{n}<yEL`;dY
zTE;mjb4nZzH!wxY`NaMCL<i>MHHQZuAVfns+$Yi8>ryZmwbp4ETa~)dYhliroF@Rz
zchAtaaJYGun{4i6T^mg+^YOrkA3cUH6jiPseUN9*pYX|#{W!5VUVrt2HcH8l2mz+P
z^3VQv{}|7n-||bJ{w$j(yu*r@hv<OuOxDsj(;?sC9J*{<(Pgu8+%8po1NH-Y>-_g)
z{DqzC#pB)o<;5|<R<i6M1)JXbqB9y#y}2FA{xfF(I?>feO@HC9{&l|fo4%0`z3~bk
zc<qBsIr8q8zF-}ZHm2#wT05;+Tyn2xrWCope!$)Bb3_aZeDLrwK_c2#>beq2v2?`V
zj7=OOIhxB`AMmcr&zeGQg;JCUk6t6qSI{dSuFc);6U$Txi9;6hC4uYf@KMSF%%FDA
znq;ind^OLcoM~S0iJ1Q0&Q#-*gydfQ+G#10dLpJw&IcPScvScJ{p1$2!z2uBhCc;C
zLPA3Z#0!YvdhYH@CHEfConW_|SN_h=e%$y5DjVxOMU#TXls1x_Aq9>$pok}A4yD2<
zk_}<A=jhEe^1U~+ybX3AT-^5Cv`oDK7)C!M=??bdIHtJG*T*<P7_NE7YqEkwJwXvt
z6mqsBLA~QQ4};+y(A+@j!c0wfM*jt@=wky#x@SRjgLuZ0wX6bhq|p$B+N>j2ReBTF
z(#SEI3d4<{hN8hfy6O8{?;A0)c=Yg47PXBZ;VKppWB5BN3Ra*5QMjnLFZVD*I}F|U
zl}~+|_5PN-C+|>B#ua$}_E%tCxw?7C)%8O{nry8nLo`~-^vU-1)+@`h8sRbs+Qh7h
zIn&F6))`eBOgzYm!=s0^>!TG*b)l~-^}KirZ{&D&<oe;`O?|Ph3#a?%+)Ah3pO6?x
z^Yp&f=t~s?@o>fA!6Q=2oNu4;__a4^$BDRHbM@dc&+ne|<eewvL*kV;zLDwGZ{qg(
zd)%I$(x(G$Sx5?p>xZP9z~SnK>G~BTOFVsx2iG@d&U5pKM~@ycO_|%<Tb6DJ!MdC+
z_BTeZAHT*_jo7E<Oo*B3@R%GEORKD_#Sq^-dWChFEtxPMxW4%y5;M2$P1<rtpz+}P
z5wE}YI(JXsqLvdKbE&&uE2p&)uL5mt^!s}rJb1|Qc%YV*_ntlDwfUOpo6CcTS4OZ2
zk$V4(XLrvzTphW7@R+WU4oAAo27fJgoX-}UNh?-82GDi_4Lo@G3T7o*Da8m-=le6-
zi_IiUF(M3&iXb`i`UgHtxqAi@EJZ6==EJfEZsx>4{sZ63CqMIfYTFHj#!PLO>)D|$
zwj+B%uon*6O*=c_!W|WGF4YTN%HE}4ydXn6J(#w@Hox26)7S<@wxU8W#|JMoOD~i(
zzKM!=D{3?05MBD$c=G%i|MgG&H2>^B^+%9i36eNGdexAMJdx)s>a+K#r{_F<+r}ZU
zzwtp*nz^qlT?>c9ky<Lta?2d8qp9l(y))%VUrx}QIcA1PNZH6`BG8nSW-o>po<Dns
z{`5Ud-<&3<n3--KQ)(s8N0!qq^~Jf#*WJU$B!&uf@0KLkK^bEr#tgmF`?_^^LE=`p
z4s$GVSJF#iN)Q+lx*%b*w-pi%rIZRj2RE2CS#j>;sd0$gz35rT5}2kVIn6Am6aU5E
z_<Q{7m*3j%zd@Q!d5kG8ln;JL8vLQV5t6SLe=++bfBz{Za_&emBAzLQF7IPc;v2Jb
zM$e8*0(Uq+dUS4xfw%z(4Z1<h%^~~{gi=fzne)gRu=i-T>2`<?KY`Ya_!mZkIs2Ne
z)y9V*2vaZ~MC;wMW9$psX7l*(1;G)49R=#$(rQ~*hrCTS6@$qwO$lJE|6Z+ts*Wlc
z&@-gaHg@XnENX@2FeSpw&AiYYpTR~uGT;?QMeQpcypdKL_s^g4;O2_9u9W+y^z~%P
zhAzJ>?Tp6Al&+bNM?+d@w6b!#zoRt^?CV}k>XgliG$ku8Crjl#946}F!2Rtr?(c4?
z>tf2LAXYHMY|+T`%yfN41bSOJ9THhU+~&B3_lGLJR59@Q10N=7F{6-n=I%?MM5U5a
zV4eaq$_GF2jfAU5ynT1aD{G}x<&{@I$a=b?bt7BE=|Gx~g!w?6ub7CO-~A%X?b~!R
z$HO(!;k~!tqV>k(S3g8d2YNlTwuNWU?|9?ks~jIZX1ROH-RX`u-}(ZtNoJl8EayAU
z_b1YH;Ba+KNE7Sr6JqPEx9>qr93DL4)C%j<x4C`$O=8YSuiW3=lIJVF{PsJXmzA4G
zuW|F}HG%}z<(3bB<3~BK3wNgz$D3EUdiWS=g?HZk67xKf^GsO^_4&fv&)(y~D<5Fa
z*Qhq$dG{+Ej*rQQ2b}KTW4XWOgKvBzwFK^$dv_@el(n(8r>31fod^MrS6604qmd*w
zJFBgun#m5Ye&~&FqCB`~Ia^9%idIa_ITIs%?{|JX-}#$9$}fEKGuv!zXs;9VwqcDm
zAP8zyV@Cv(QWomExNT?HD(Em~ZL&|OUg+E}Py$<v1KJH=^)vf>P<2<d@v9dKh3$7P
z-tG1n=!7b$cVok~4%?tU{+ZA5?Cyl7_aJ?c>FTlhTet&cm@>zk2Q-;zrLvyS6z(X+
z6eeAvtVZgqA(M^=ZeD%EioRA@pS?vn-_bRYVxnQ@Ig%%1XQv|(W!A{DEF9;_K5HxH
zQhEIfakkjm90k(xWtQsK+#^LxY!oo)y%i)(TU2rN;G)`!cj#_5;{{_JhEP6^8bMgC
zp8T-wi0rVtjwo)1an5+ok+|9Y9yB?kx${|kIs=`^Mv(i<f8}rSw}0{%P<4xR9U(jJ
z;U7l|tB=@*@<y!?7#6&6QSknMsr%1g%a-)a6Z?f-X1in4UQ>=_k<IDpo^EM00%RcQ
zZxsXq!hDb*45S$Wn&|<iyVVk#WV2YluFIR5$8NV=IDLp%`<%-UV%-k}surqlW*)Qm
z+H0+dc;k5=!))?18!6@URZ<dW<DOG{?imV}csAyqJU?c?kXcP2ihkbv=|t<zghGGX
zIp=~)k|pa;{sVJH!Q4Wa|E=_sj{f<PKXU%e)>egNn|IPNrNW1p&3|U^ho}9)iL+Bs
zg~`V(Ih2Aa<ib#if+mF7w?Lfip%Wm&Ty*3;Op1}DGa591wj9|!yDvUBRhx^2R9>_6
z*~M%NElxXmrB(Ti5s@Zh2$-tMi+Y&o$B}XJ$aVp?#mWj<TZBv~@A5nv0&*I&>#mGw
z*-+IDtJMl!H6_Y8%MG=jn5LevbJ^-zDQsOKOBRrA+VcFBVov84v=TG9e~)^2`2EJ{
z@AXsV*xi4mVPv`5aB*=(7!D}aa`E;jxM9cc{ykM|SY5oJT20w!d^jNN8OL#CoF>+7
zLj(+hal578KH$T|blBo?wChU_$1M-r+pO=OdMd56u6-J?+Ok+${M6Gl6<vRzsv8<v
zliUH6z!-zq4r_GLcT3LCUvubo?E8K8I)C_p44z|m<S_IYQ_<87<2Z2`GUL-Ifs&Ek
z!-ot%Da~)c_yQMvW{E~YnjT5R0gH#AnUW_aj}(SCpZ*2OE{OHISW|Y~zyAUsB9*PN
zTC;n&fs}VjQ}CE}O|`1%4qI+Nyd$^?No2iRQ&$a$gm(_3D<1au?6w{2rRM7VCC%!J
zs;Y@GV4GzoUUpmDn054xw*2^$SN!cSenYWWA(h@XJahJ&2RUUi6{7TIV^&D`C&#uK
zMM$MhAm`p8PrIYjhW|8pBXd^xyojGBnfYfaliagK{&{;KcN664edm7dKRlkmhBCpH
zV5avUKXCW3<JI*GjIyYzBF2QVhL{{8Of*(8xPZXZKRz(^d+OyGvS~3=F!>QJ1=C@V
zbB?}lk)~l`4G4{CRyjeB>^D9oK|eZVj2JPJVn7L~jV4S3sj2~mu!{El1%k--_C0Al
zvfX#zXQ=+c9W%V`+=Lr?1QDTZHm{j-PKPMF>KWb-E=M{eKu1H21BhbJN|x$+Vvz|J
zqbL$7g+dye&2Hpr*OyXcn!4QTkvdyGOI6%W#R)-}o%il<9{3;s?O){`rJ7mmx!;@O
z>`*wXSIX}dG%f!#K$K5HoAMq{{djTe>Bf_JQX%R-FU;rG!`vExVka(6(E9U5x-uE6
zR?p7QJC_NYGNDB-uC$g!KT9rU@jOsZ3#Al@_~b+y=3TlJ1+B?h8Zd{jWf#I>$k`rR
z=2XAneO^q9U9S*D)qp5{dRce?oWAd831LbEF^~UwuA1EyOW~p<Mr#t;ub|8nrPcFY
z`E)ieEG?4~RSlw`JB6na3o=#amE49X$(7cgim#ls*yLUL*l~P#V0-8h>od+(O;s(h
zb(ThEg?5H8jImgekq(j4Q!YX*2#%KJa~N!8sp=~GzL4Kv=sL>`b(Y5l?~y391Zyj_
zsj%&ms$CL)KhL4>R}8r0p3BP%v|TZbLzbOIK^P9)-Q6&B$L!_&^)HZ%P4Sr@=^o$n
z>h+I!IIQu*k;Czk^mv0c6^r!+?PA50BKwbD6Jun3{*oj${lf>g$4ADoXTlSeWzn9I
zx_jI-(C;3pb@7ysj>BP(5C&V-__U`InyC=IR&C4G`6Wr$^h3|#{vEF0v9LD#gCmL4
zz{{7fSgzL8Ml(6jVZY;Xw*}F%SgjFKF!_k778qkN3}7umS&~q=6fr`fY>f~K=Y5v4
zM2W1|tO<168>ViHSuDXx1RkR$Qa6w$hW!q!4R5Yr=BQU8u*R0C<QW>Y)Qc6bUTfmG
zhXmgBc$e5c++eyL&GL-oJ^QXFhQOv>^RNE%zu>?5H~${z!*|jQPsj7pe}?Cuw-NxQ
z4Z+WI)pK}0ozRAP!pak3`p&`+&!m5<{N}v<pOd)e^Eq)9p1;Oi$mB`#)DD<GfS5ZT
zr3Ep6pcCs@AczD%@b2Aji0{AR#jDS#>V~Lt%Glws<#7LjcMm%r$H1r8m%O-mfwgrm
zp<<$LH+e#wCX^m>@X<K((A`t%hSla0l^nw`piv|#u~y|`bn^HZsoRziTp`e@Y^&=9
z)9z!wZr+iSWa@X{XQ=)G9(niYi@Rku4$3jaF@&PfNEtm5I>*W;NGZGHWlswaS?!m#
z<WeJ2$gd#;O4p#P^8AR56d;Haloq9b4Z(+e%^6Y3B6Q|5B(lGKhT|j2d;a@>^Y7Vi
z52xoQ>(I0JM}Ed*4j+mke?84cSm|kGl&p)B(buy@Z-)Mb$ab`2$QcJuK>VpemVe{<
zaGlYE@*4ADjbgR|mRg~d%B{g-ZJQB=ETan{7X-2p0m~WZ-#tO;A-6bEEK$5O`B9t-
zxZGlx)B6HR2#nGN3CQ_0lgswYMSO9B;gy`tpi(jU5!s9@7Pnc+GyV~UOvlO0XeJM5
zS4EB06)wf>_2MU-%hvC@ARAAJN<IgrvLx^6o^nE`W-p1k1*5gbJI6F+gi#9$*)!m$
zo}t^*ZFjid5k-Yap7GdHRW@&ef=5EO+b+ciA){r(@i8k>OpRS_um+-cxN&5T*bd$?
z_5r0mIa)^{;u=zeuojfTScA0&6D`(g#(w*IjnMB^3|yS8bCQ@8Eb5B>(3AYg{Wrg%
zsVYp>Ku9=uWPR}x-K=R>7d(9POFn${3$8!?bJ$#R_x?*1ppB%pg6?pGQi8?V1!3&i
ze*6lvegU>+arG9vI>SoKap>@3LN^<NpV&P<&>F$+@PMvrhOWcfhD3|hE&b4;r3XJS
z>^pR|q*`7gOigq7DQ&$Vc3ak)iV4GXd?2RC{cfO=k>z?ry*}gBVu3OiWgCpKOv6Zj
zJP-+pY6W)9VzFU5ZV`ujDqE3+AcmgZ?HwP!7`c4;HWw{K&em(D;fRz0g`=vMKxTak
zWoRmyAD)z8%<{4ne32jy+}wO*z22~_DondZG)pEQn5Kbo=yL*=ko3nRQfh3wMkz3L
z%Rl>*KjF`Q_&I;~n=eWaZtlf@7iORT$OvmHJGJpt;X@hFVqSAa*{wXYJEn}dd4VqE
zsZ%oAJYlB%z2`+l{?*fqKRYRbNKfsFdCoj3A~?O>(>a{bgLHb`6a$Mz&CBbzD3J)E
zCyo&}_KZMG3bT5{Pv8DCOj9uoJ#`5#a^7bKvk1i(!r|OV*Bv?R9{F(do~svc_*_Z6
zPbiyx;mu-+H5J{kM=u&y%X7xvEn(P$%%M|m>RGfc!^200aX=`IwGGyMKTD85w4*W_
zrAnW@DxXCZEJ=!71jvls7*l6*Sug}Ql*P9Azzcy^1|h4A@+432L$T1F$%9#8kdY$+
zWMZL`0!?-r^`7iM5q(Y+O~r7k7`Z^=_U?iI;otqu3C5lw<$R>4RsfV1RAJng{DYZ$
zkcv9u1bckvxy}W&QYw3q4K5=DQkJC4a-q`Pc1!bDzSHZTTM4p^g%}f7WeaC959C=+
znC6J#yu&xe?KJZZO3ZQjp7*}M<&ZO0@>~)9)7U+o5QRGcG<5}OPDOo^G0x>pwjEB)
zdy%>AGM~Q^Jh{x0Wt-2jk0%4;d|hHl_z-C-jft6`poBo^yo>j~SiGxQNm1sDqKcTa
zV`g5*oah*eC304nB!2p5f0}LOb^Z)Kfle6}O1i;pu4vVQA;8e>xqUy<w=M0WL1~4O
z8dH2RNSSLTVkEc;H}*Mc(^R>LF&6J8#-Ri6%7qGepVbReCX^C%?tt$tRlUI0Ezvi4
zS7%wL_oz8%`X4#I*9_xRpmzgz@3$-$OQQEk0aL%lg$h$I5xQm^BVT;|6|X=25sSqd
zP#6;3;|JPy!|Ruyfl>^^fys3Y!-x-_s&1$kOM<X0F5XZtSH!WSf4pHhJTe~ljDu#i
zeu=JILVut;Y|$d|{=++j%qeu6>&SY2%{U$MV6{uaaK!Z<X&p&M+%ythz)yingI%o{
z#vVT<MnCZJ;akj&Wm#8TUR}~|F3{&MkV4YE{~9SG)#fexY2dj1$O>53HMKm46!6|7
zQ{wf-3O8wLGl7(h!s8QkqX%s=+0c_S)u%nCsxVq{9E&WoT{5{yB{k#z0e9SUy!$|Z
z{}J`_C2pMXAyBmoR7%t`=P4*#lY}}khePxPH{r*Dzx?4_{`MEYLMW5-8fL^HJva3z
zKL0WmVpq=Vj^1a&(sR-v-}MDZ&;BNa_|71nq5r3eEIqB<&-$eKn@=;%{Q5JjdV>9*
zo$tI$c|stb*vmpBif3HjSxD&n1FBw-+z2TkiTI%-Nr_OBN*nxm<am563XwDRk9Q~u
zB09WtD7$7PpzDUEt<X)&#V0?estvWVc`}m%H)S_L7ZPo|=JxiMVHgl@fao#GU>0kd
z^@ce_#w;T~<utSw^L>WuAKW2jLdirB5kzKx$D-a+B?}^_vZ?G$m=a1@Bv}qOn_M9U
zQrdzHNWc({A<a_P6m!U!2t^{6TL`n1PDrpyBV`7hV@w2cOF^kRTO><Kl%D_ei?7)2
zjzDJl=iRZ&JN@+ZOy|Y2?A&K~eFm|fwXbHZOo%*o^7nJ<pO14!hypQld7dqx;#rHV
z{CiFxoPGOGPkl<o;OS{>Fv|&3Rt0!Bp)qK!2r(;XOexqPFOoA+SQ`T&oQjqhpOPMD
z>EnrCkigjdy3^=RZJtO*8oUqL3qmRUm|=DovNwU%S$3y{$oB1mVhv5v^NSNoBA=lN
zF>!J{T(ufoS&X1?+kCmVnbVQVxuyg@<lGA}s|zSaQYqNxdyonOq?F{vi)$8(B~_I_
zW6oO8AoI1TqhRTMo-4ydauZr<v{aC?=fO1k%yl8RK(fPa{=1ey90^BHkma00z@&&&
z5hEp{X-GnsLOLJ_2saS%_%wps6Dad}g4FMSuQB@lQVU(+@~oj<%n*1sYN+bm>#SE7
z3^Aefh#We;{pPn^u2)zs38H1%-{FsURP_o(b`DdkD;NXY4_{&S_q=@b111mS?R(7j
zmbzUbO^v8(lyfN4FirytV`-O5Y@<kiz)ce=DOQceJ-%amctqDV_4*1IJ&4HEJtA>P
zTY($1rD<I=#>_%)&d%xg516Cp>hcwft2cPpaXdaUxi49a9rgMpx><4@2c%Y9Y+hkS
zgu@M!J78@?bR%Q8B|4w0v*6kF2YT;mmkm}KqzL$=sp=NJ*x(P3Ox>0k26WRPl)||V
z;d^Rb=P2c>L36<me^fE_2h8RL*Ps4~sqdJE9widd4VhgV!Hokk1&p$E$2~qotdYEa
zeT}enuAZMT^>UY=*?5v&*JlMma9Ou1Rpw%6Pn80Vis$+3oOT!A#R(F#82qFRC};q{
zY&ARgMl+;Xtd8f&kbm$d<oC&cm8bWbIS2Fmrt)24%vQxxp_Cwqh-}Z$1cu!mU3Z|W
zYHYj4cLzv`rmFC+=dgRgSc5Q@Zofk}HKJKFI>F#P$F8F?OD^gqIvp~3N?TAO8>|?U
z!?lKqwo>_i*$vHNi64$gDd_tiZ7jw%IiM$IlN}%bxU&g@pNOF>q=dX%CIJpak+7u%
zA!g|<WIc5$9HmrQ;Tj`8d5qF!cPLS+Lg<E+^5mbO*p)}L(xvxXOl4*}dlgAC8w-h;
zKqj=(CEc$$_4ypelb!Ig6JL^n6y>}&KeYu!2^T73X-3U5(iIRwpPuXd;UuUzy};*!
z``PY(M*W}F*r!&(Toim4EtuN~#Vf#=%+Vn)sNRW_Y?74Q1)(5mVb-}nGad`xpIQp1
z@iY4$#2nL}NDw7brqjZyl*%3qrFak`%P^r7DXO#?$^vb4u{KUw+Bk<K>5@w!i^9Rn
zPzck4IdxIVT*&wspVZGnAav#+2%XPya^uN`P#3}{2t+8w4Ej{u%sdGxWND4g*I4jB
z``7=H|IPpEzoKau7^6?-O!>-3lE}tYA}Uf+geeg`bFN|GtQ6Cz6tbsC3Mi3@qgrM8
zIx-PmCg}rNJ&?hntV%L<A*_n5SPCPgY%--%$YUsk)BR6=_`SyH_bLXO<qBy`CZI@#
z^Bn?#N*<vES|@CPVSk{Ng8hEWe%NxhT5+~H=k{^J3q@;cybnw+Fx|eV>pRZYOJKo%
zcf{HX=Q^S^2zS6p$+*u8Z4!d1+cTzs51wi207p=oz8l!y_AD1QE*{W*kJ;_1mn*Ev
zaAe=@FkykNYGg9(4+rW>G3@T~!;$Xp1MTJlWhI0?i}g99HjMiR#N#d9evhse47&&X
z-8;s`8s`S4{T6F8q%U=zs8VovDX5#4;c(28pYQR(F^waUk-_)SuhGTBJ~&VB%fsRh
zjJrp)F=$g^Y>hDzTWPdv&}m@z%`ZtZH&TVINp8T64pT3|7?2gl)TDNu12TNiPygbd
z^Xlq<;{EMi>6^(@CH#D6J;U}sINUVg!;>s1hCFCy6C)u`oBMJ+6#=smOYZG+N=p-P
zB3f}m5N4U<6T$E_ftTtz|F|s3!2E7c%xaiPjN()u%pWVoRAQ17MnUQxnM}wNrVp&n
zUlU@a+dgn}_mOIKj*6OUv81vscDcqZ&k^+_A3l7EIDXBdSrGh@H=7m5;})HA3~Q7T
zsV0ofTKF(|y2F;k@qn!tv`xifd(XqoM?}m9Q>)b)BNg4_Ew#1z3W|t&%31lNJf`tL
zZHjqJip3$b5a3i+tt2!AkMjdc=2q8<_mL9BK=8h3@f1>J=erOBQRIb36;vUK`1B0S
zyL1Ffs={RgbR?xI?_9%d)f@1x!+VDtCuXy)xx~yn^U~m&7w=TG_j40^9uH*!%Fkpd
zB-|N6An($7?zZ@}`_KM<Gm^$>IGvEQ(^$>loZn9>K~q&H)F8jVJhcvF>3;{8c|gwj
z;+iOkDqpV<GM`8mgPd98FUVxOd4Y~8B2y$4*TlI^7qZPDN=dx)^BLwMMP(J#Ja(lJ
z7^5(VQcUD<E>*rSN=rehe0@WRq*NB_qN<8zF-<AvU@+&QwAB<(=1p@dqblUbe118?
zT*XYF6e5-T_JmT#n9xS@=YRHN)~g0>6qI?ZB+HOeKoGNBRwX2nNW{xnih+!dh~l7m
zB7CMi|7s~unI$P@%Vg5*A(4%^l8{J6aqG-v$vInMMp(1)Q-T>O6OhvH=U)1~iUC`-
zIk`(DytLSb;rMV%A0x~4lAF6bs;WlGL{(SFv}T$d3BfcTF?Gw?>mQPWV|Vi{!}g9Z
zfA<BSe)<U{O=S#d=!YFr8ia_9hev!$eEHiixq9^mRaF_{Jv`u3M2CP<&>eOt(_rf*
z*oxudJt_r)Z7>>+$B6TRX1QXVCU)C<oNVxbANMF8kzpdqNVmUd8Yb2+-y&2^Wi4{G
zX7}yy&?#W5mT8#i#*wt|>4$;VfY2F|?Dl)4Gz>$Bk^;8dT-AoaKKY_eu8>O6ACHXt
zJCvz8?hm*G>+=hC`<}r?Dlt%*9PCurHIw(GlrSnAQyh1XEH)eL@|^pdZ>g#lSuGG<
zN3B4H$U!Qa^OyLkV|)9Kv&$E>l{$6Za_p`=5p(8AX0G<5_hk^~Jwi<RAyj4HOCbtJ
zBIKZ#nbpnHBK-uH&J$Uw9(aP}W`IJR+5pccSmtx$G>Me2oWA%}h|C1Tbn@jsaSBS6
zEmFc7g~E}Nqwji%1Hp;Yb&0Ct>zjMdE(Gh~d3k=uG!Ddg%kJg_AHMz_7ps;<99fHm
zb&+BBj{P>VzIcVU7RHgsyIZEQ#~O_bk)~N^x69olVcbzEMcb^{9S`j8J`&m%?;OYE
z@zcc69~n-hvp>#b8hbk7ivdhdEJVZ%(H2srl6is?Ww8^IQ|}fsWBRmM29GcfDKtS8
z|9&CM)?bw8s4N60s5cR!$PuwZfy#_~876$12yOrZ#0fu*eDU>%({RWEKH_P7l(8{C
za&b2PiK#sCb3rgK%1;F>r8s^{4Sli#&Q6O@*KL;Go(hr^_n>?q$^w#qN$X4=jKSw?
z5mW9}TZ8x6m?@)s1@13x0q?_UsLcztn9+yg<eBB2xxh}d>O$legnCw_WPZU@yCJ6n
zf=Zc{pNoN_2q@!J%IqgHn{&;D+$pNLTpPiYf?#eF%-2i`!RWHr2hJn3F1zCyHOW^@
zh*?INC8f?~n`$9MwxXY}Z{l=sUcbEJU;gu-qVTAEQ8KhIQfWbEX{9{PI|(T>+7Y8G
z_qMn`QjjGPNg@(MA|+J{UR4T&Smv{=tjN!0$UHTn^1YE{G%Sw!Gnf03pL18vPZ0kp
ziUBtq>9&WgYd0;%*7<=86T7>Q2q74TeU?VZ?4P_)#c&}d#5AzK|A<eCsXx-~A2~l;
z5&b~Y4UhYg=HeC8qQ(1w>~4`^K*YeSPd-H%!#IxlfpHFDHQU31<*K1>mMGoQoL!-s
zB~rHx{UdkZzT@$6&*jw>LMjf29aGWa%g|HXhT|cd3GI(tFqSY3j9pK>yQS-n4Ex<|
zW6$IEz<RyTKGQ~`M50mIQ7wvu(1K~|kv5Qw=I-GUYi&+g+&-dI@z|D{7!#ZI5;u9a
zhi`MvLfw$2Y`=VS|G?R5L1-!t!E<$anX5~!8K*I)5k^6+EW5)4)6{c!f6wFso6Q+o
zB#e^OZOe4HryIxo`TL&#>Mwu77w<k~X0>zZsxIuE(q}INJB6nSc!vICECoSf4jXHc
zW#EJWO67?;50a-z>m-Xx^F;78>wQ-inCZiL+DpuNt}_dIo=|}T!vgcdpJ(71lPLcy
zu;6UIQ5F{w1I?ml5H0oM9Jha@vX*J=u-DgYUcSNiJ#p&T-@W7Ri-BhIBEy0E9aoD5
zVQ_4_1Eh$SmhN~2q3GR5s^-A%_7-gvRa3DWI;4QB*A=(#J`$B>xm-gEgy=C+;$omX
z?r56@;}|&{j)2BA%kMK(|KJX#GP;!=;uI-m=%1Jf)5{@KB>({B8c9S!Q~)v{6S+th
zPvpFr8<I8nh43XZLp{2&?Cy)sK9-3RNr;(Hm^T%=JVF8&JR!QwN6;2xmLi&B3?q>!
z1w#t__N#A8Q8LTU@^h35%sxvCp{&ihfqz1{=HeD&cJ3A;=ibjd#Tj)d#kKe0#2T07
zMNhWj&sqX=>ny)<DJ)Kko$SIla}R`+h?F7qoCz_%P99sLxcX&%y*(jlGs9oXY+$Bm
z8{KpoJ||~CV{}Py&21fKMNv6oO2w^lj^uu7=lPO)O3MBElLKTy4?`^XL`sw-yA;NJ
zFK4tPiEJ|*BY5vQx{+nwBCH1Qvs>h8fj?Dd`8rB4rEL*{XKop!1l|Xn&mIv!`0O=}
zO(>D`G*Zlb3MJ<ak9f+^n2W~fA$f={htUOJ<}OwCK!wK=^YiRe%;sptDY3jbkY%AE
zOeM?45wtC7g?XM0X)O0NOCaaqI3@L|82ZPK?-ya*Zm6uq9Uf6rpPiDb3TZ3cG$N#6
z>@%Z%*83}~aZzyG9+}+4?u)P4ync<29wx`-#W`LHd{XST56HzPyC{uE;(m)t4qH|D
z2;=UNAS89&vfb}8VZ&&KDW}LSS1reoSmIf|`izbI9Ht2$CX&)bsWDZBvJDb}BGDa=
z93FRBic!~$ha*0C>b7DW4|vxTTqG5TsmVpg$+6iikjfC<fVL7M+5N9wF?o-QAWX&T
z<u&fmAw?p#29ZM6jE8_9yG+`VlJ$B?T~+MI$Zvl88&;L#a@C+3i!p|%HT!;`ZrV(~
znMOtzdH3!k+kT`q`SVmlp({z424rREoa5au-qExMY|Vb((|_@vCJFwtSLghvFTW+q
zr>Z`=5roQy%Q5C2Z1L3hA!JFS;6xzwA)uLeH@Q-N4o^~8)pG>F2|`Uzlh>)g%2TUh
zPO6)Ce6!{0)8_vl77}w26Q6w@NaDNhhNoSXAQDN_5g#P`=@IEW`lHNM_0U0tsXG9k
zZP&9D1}X1&*dKU&+_F5|WFwa-m;#VI!k%TxX>EyqWQvX?YFfLZ)fFx{zWDODG<C!I
zi*uY=au_}L@85ISJ?6;{&-rFWZ7d(Y{g&nO9P`JWU`*AX(XQI!R5wq&5y}oIB?xgM
z6~i4V%g#GRq%<TUNXbKzD3haMy&s9rlSGbF&)ye_SS*j{nBS~aD21TUl2T+si)_=H
zQpo0788Sj3vz>2Yo=d5pR0b)_=Tew&h<Nt==7KqfSaxpWw2(9Va;UjyF}E3>o=v3`
z)&L?hhC(1Li_MID{lln&r?x_VpIlIAg7+9}pLX4a*DxP32BgTF5+8Em5<@6b!(31$
zBHHNeA5ex(Zlg?qC<+LD%8gJ;mvhW0q4WNfh&Us8vpb;k`Mso+r%@eK=4E8{NIJ=e
z^F7Mkj4V&}%pjwZqVPg85z+a`_OL?;s6@p1vFzGKj;z<&K0Q09suM~ebg{{vk*N8e
zl+20$@?ZQEjl+*SWL1(}NyTP)p4xJ&<Kz)BZ>mb+lcS$yf&B}*QtoBOL1yR7f@I9|
zT1;6%F!gzUR32iDsT!m(g-|_vS8)2>+|HS~ApiLBy^4XmufO55AODD*n_%eJ-9J#b
z4NcR~`#@FKwCgq1dQEb%M3EY_l28k(vumumWzp8S;ec9J9Hx<EYNT*T4c*Oqh7iz7
zQPnlx!RP~0D9q-Pi}Nd{uA?5N%!DQ2$y)Djzr(bV-~9S-Xsks`jZp&Od&cNM8HRnN
zZfqu~2!ZQ&w5??tdfN2{u~xXpEmKHz<3zvjxxBh!v034L;OhFam~Ldx`7jNrlz2Q&
zRJBE>4iOX0vSxSekj|l%;O6$0McXosfve?;YO_YRXSD4()3_y!9U?_q4Z~s2&>vYY
zRxFwpRpkHxLI1wqv9>j%4;=3vF;dVpnor(dBdQ83HF4@WggzTkD5&ikEhMXJ!_7Be
z@%D!=Sf5?d9S+=VJ)eG3vubKyefoyK{>?kS`uM=mXStOd$BYVyTm?gMM-!3YixXV*
zGk}nHHz}Ri!>2`eF2qj{I8TH78Pq?wAWqdl`K?rTFDVtmnV2=j@w8)n)-X7|*LU0x
zf$aG41TO-_ln2Sw9oTLU1lMDAh1LS29U&!jWeLg9ss%Ddj*k!Q4ts1}LrS>GBkY3Z
z#frl)($p=#`n$iOvYMA?E1IfebG{)s$8q<7U7oR6UJ-Wp^!K-zO=8>~xPSi*DU9?!
zF@(UF;9|4k;?*m>3v|CHBlr*X(Apw|!q|$GBxOp5H07OS^dxdj_k2TR^azoV6q6U?
zX~(R!!B0K?a75}FW!oZo%}9X~8Wfq}ig+TTrc?}x#%yFWFO+#Bm7|1&$ctRcBsZn?
z>A9O722)BXV+wRmS=eVYtg_J0B)W3W6h#73Gs|j4`H#}Nw8iod`xFQv6~e1}#+4}$
zdMUW)qb!*ua-fb55`Ai)$)d1$g5{mdCN<LN9FU}B@#xRvHpGY$8M2Nco{H>QDyX#1
zgvu1jflOJeAA>IocEtIFF^0*xQZ$6q*qpNiW~W0T1<{35EHwFcoKh}B$SRRUF_D@V
z*)pEB%8R`e89nHZ``q>^b5KTivgPlLCB~wzfMU2-?r%<9d}>3StyWxJZjdULVy_q*
zi5Zdd#V~4q4N6O>EW|>hje0^_L~b2;U*wchA%#Y${0y8jGD<;_Qug5pF{5fy2!zg|
zN^`p^gmChS$oUglK@(CqrCk1_WAyj>sd89b;f{N(ObjAYFB`O!bh}5IszOVF_l~+U
z>^hHgj`v@G$!4`e!~xT8P)0KxcZ_3C*NxoWJ+fSGczwA>xt`%LaQWju!PaY3Q_~&x
zeE9lHE;bjOzx)Ie=-Lxo1tBsV#|&GG#E08282rS`w{HN!<KvdjMHZFd?7StVh@X0n
z!+~SStnwrU#u_FcId&u3Xs%ygbN_f?V=P8mbY*k(nIft(<m7^4eX2Fa*1UXqMF=B8
zM0Qh0-yezIbJ%UMn<ba8US}dooAQnH4kIg~intgHk;Q|lSzMem9W?cFnYFKgOp#h7
zYNIffVIt7G9$OjYa*1g!2$N$aBNd*|?F%Co>bB+j?WZ)GYhv`YDe$LnuldXW?Vm92
z9uVV!|Kh(G8H3>F-Pf3Q$-A$<;P~*7YO&_q{YW=B?r!hc?mJxa><&k!(a{eXo(eIZ
zCf1lZNt6=uWD`11So79qPB5Dx{!_mAv*rE7GnoJTtamyi5T~u>cYbm<NqW{!5UIeX
z`DY>$V=LTvhe#f2>TG_sJs?xU)HTvZqLf^|_=Nr4M~*{}Fp7R0vKf@t#BjhL9dUnT
zpCqjcsFVo(5ve41AHKzhz+$mLG!05se0=v6cIc^(2b!j(Z7PCT@Z#bUp`mRWqJwG3
z83*I<H=h4skC+^>^hJ{}IjWlgA)75_XjY(fgHW0n9LWttA<)%=G#*H)xUEHx^Nu7m
zDT*ALrL!w%-l+#tNY9>;l4KVn!HxLwfHqk|sAQc(tVCW2BO#lRIiI0xt%_@9*{K&)
zOqK;)X;l!kkQYqxw5yBxgP&S#sl*uPPQi>clxJRM3tR(`JSrrZ(T9|Z_rd%$KdZ9m
z@27Nn)F&6Hj3$PFHkN6eo;3DRW+Jf2%!ceJn8&D#@d@HC=jOdDp+lmeBr=CUod|n*
zfmd3e`r@fvzgZ^ji>0tJ2A9fpl@h{i&_oGpl7u;fA*GyXn#!VG(83%C{@tQJD;A!s
z88T6kvbmT)pDt*RpYsvO3un$=$nw%@%0Wl+8Itnp+n2n!KEqm@Tf;?ll4Yz@-l)u+
zBq0&$so;6yVM!<(n3M8>{ND1^hD``Hw~Sm)KP-00N@<d*a(hjg!cU1M3OszLrS`;?
z$@a_f8Cv@P@AzKDKp5^AwnH9haY8D=&<%uPK!%3><f!VF(YK`48M5y<yG)q2<#_Xs
z`(OTy_Ubie)u5`H)#Zk2d5#c~_cyov?pNPpHcJ-0L)8IkYLaxsDB0fLQrn8Qt{Hb9
zk$ynwijRjqhhtA$S-P&n)Rsk6gHZUYr)^-dXj!%m$Ne6pU<wK6BeK$r!@%Wog;^Nx
zw@1#5rh9l`wb&rTh(yqJM?M^OR1(_NGFQ?>9D;subW}7h(tr5K*d3@>6@K*GefyD%
zPv6qkZSF5xl|@y~A#6pIhW^7lx?#YGm_gGBMq8HGuV|_qwHmr3hwVK|O6>X~TUSnw
zMoNZ$AlU`JKEqsIP)A4V?x^cUW(wPyCI(j1k;Ua}e1K_tj}asB{m-yTfS-_IV!6Cv
z^W)E{7Yly+gBRR?e2-aQvwHC<vRQNc@rK9Iu~@8d`&+{9TmI_b{|Al%hLHHhufAY^
z?6aCe=k`b6_qk7;Qke|%&SUnTp9i&^clk0~fX=cQaheR{^OE3cw>1|5^G@M3M?Mt^
zLI^UNR|XKpTGe^H^VGHFI4RDrKf%QjcRb=oj}(cfZ4p|sJM8h!VU?y?E~%R~b0)@-
zZoj1;ChB&Dcb?Bb{fx&$M+7!)#WZ&8j~&J~5Im2MTUM7ZI35msbF*i)2H_?ec_bVa
zo9ovc_D5oH487xUJm99^PagD#dK{0BXthCULlVA-JbZ?5BV<{iLx}7lAcX*#5K0n-
z0b$B2nu}&11lrcfD&bwvG!1!&B})=$iKxx+bj&{RA(FzBJqeU3Z3R%$W}bq`u7xs5
zKHh%h*iWa)Rj8*HKuYP93Yc42PvkI4&pXWsR@sT4w1UOlNT?^W?%dDLYer$Q$GqcC
zBD3D%2{xX|vCmyWpCfJ{3ztWq7M5(yUgC?#EK_wMLGrXs&c)DD;$k)qiYcoBQWPZh
zL<qK(%|%gJgoKbe*F{SFFd(I5be@=C+2%<3Jl-R^s!v48Tq!+C4rgdy7I-_9>p7#1
z^PV8LDk8ILD3;Kr6Y^O22D3Bayur>8ef|j{GPhtRQ7R=*w+CQy{?r1|TBD7oT`sVd
zd1^h*(@g?Ug*aWFm6)HsEU65q9F`dKPCYAEQ|7Xea%(4zh!D`WB8StB5ORSlg+$35
zbDm>~6Cq7R6-#R&e&;&Qi0eE@&UVrN*zvuJfo^}mXv3n(vOMpyTh?)$*mfO_ZHUS<
z9&cGmg-nt2x1X}_2b$|Qv@brT-dy0ekEH3y*{fF|4E@++&tJ0n!Jl*f_Q$yW18Uk6
zk4FwS@2PNHzW9{G;|;+m`tHEy)hl$hWI06I%Yg4YY$dP?rtKXcKHj0UqFpUWR&o3O
zk#}Ex&DmzdYO|(V)l30ItGNB}5iKR{vc+{B-Qcj(4MJ*0?-0Q=4g<QX5D3O`MAeqD
zA0dDN3;|zhnr4lvY7P%uHgA7G8alQQ-_oo$sLCKk!nO;9w4~Va^79|i_dU3QZoj2&
z7RY9aGC5;FD9vuyk?J-2>Jw~TqbA2;yg_S)tt>b1z9!j6)aEr+Q?puBgt{(m5{V+R
z|IiUaqPc!ey}D$$e+T^oDnxof*#-Xih}~RgBMxi1`s62w%g;%64W?n!yy8rh2dlrt
z-7opyuQ%M>d}Om(ar~eEbNuK)D(rF%3BLa7YxcWGoF5s6o)jEXTkiHfzj}9vQdzHi
zdw<U`jCt^e*%o$YSC@%m_Lyg0=bse@<<Hq2@p*-OY8VtaGk-1N-N4X4VC{zGCd)E+
z5040KXlg?<9a$_i+BB?}FFD?R!~Xr3BqWb_w}f$Iy;{>Q8Yb^S)hu4UAx#}k`iO{u
zm;(Lb9ycl07cXer2CW|WaC^(yt53OpojDi}cQ*(TP+AeEj%Lvi_6}tP>sQy@-amey
zq521RL_ZRXOJFP{x>(|#NrIv4NnyfO1IFeEav=pG8A5m@xQP@(HVZ0=ho-&6G<DwX
z_XAUinWv<(yPlefXWHWZM3Mm?1!!F)b`vQmgsKol43>A<fp0!MFpM)159W*xDP?iX
zE6-5~Co)-Hl~RG$!)YgY+PTl@lrPV$%)R;Cf_UOi<V~Y^;>{G^LViQaJI&@@Pvc<z
zeoP7H11&k2NNaUkWPRZfkTOCLOVLsy4l>FSb68OdnY@=$JngQhaVm>wF4pU&$rkP7
zm?fSm;U-7ZwkWL#e##`ba&B2klC6*D&o}cSAfb(B9CIectgc8>oC={tf#J!m!u)fy
z1n&u1jD_2ga|J{}H%gaAX_B9bn*rw?Rb?}Zpd>ydLQEy`QSi_H;>V{jpb%UsK1_CH
z4W(rxQdVS<@L?+Flo1Ch6PYwi<`|IYse%a*M9OyD!I9joh0cnQKz31-LJ>ti=inV6
z2x7{~g{C6Onk2?7Q7+cgh}?z<?w_C-P+C&e6=Bvd3eCg*f&Igl)!CY)EoQO61<(1a
zrQ6=}?Jxd{F$n7Q1$ME)-FzfOkB^T1u_MWf`>%e@#n~BSKj7Z|hB$SMGT(Tk1x{&1
zawwS?r=HPy9^TzBPVYG0J#h8<s<`m2P;sDL)O_}%PwD#sTW1s^CCMj$`sc_L8IC(N
zuv(t6A0~9W%nVMcF-?Q4EBf6%Rti>C&Ccy11f<QZ*=o5V7>yDpU%-@5!f|$a&eRW>
zx*;XWa(T(a*I!U8&0@0xnMjkz)`sCY5kz*Vd-(cy$P^Ka3))Zq5+f$I?|y}!CYsC7
zuorI`_a8al-_XTKyIf<Miu>EVAgSab9;oDi^p3HgFt)-PPk%VlSjl>|#8^+&?U3hJ
zXwFEF1I9>1dq(n+$J-CM;28XbXx2!3j<9DS1V%};6oACWfnX%*{57U-_`za<a(fmk
z(eIBKqmUIuRnh$5B_^IR9v|_;z-qn0$%+Zfi=X@{P2Ex(#ee({|Bk==*)RC!q2o7S
ze#5ulzGvvif51avo&ryk$J|4lCku%x326aDNv2D)>sXN|m1M<f>=?Tv!f3+rKuCej
zVu6T(P1te#_&bibze6`m#_9r}45n#GDKQ<7q~v*jbL9N$3ZpDvfBO}S7-_BHaC-yM
zW3^`W$zO2QobmpfUl6^+YRm4hqh2g{`1)5=)(}y={QO6()*HV3@;6krL7!dW?miNy
zoND+-eF&2Yr*2|q{6{wt!bnPjX&R9tF%DCX9F`)(q|%VvMLhMIBOnuDn#v9}!`0)s
z$C@1ds-;FKi%<$d(SVapx3sQL{e8kniUT1D0(qg3DG~kfL|rKyfly3%va{jSf+U{t
zT~4Z-vap}zhw~yIb4sX)5ur;+Qdw~4KKg9f0~EVlaauHIG~(%+JhcwyYve<~>MSLM
z9FIINuB7~L7hP@@#KcsLm4Ivl6cSU&?EEoBv=k(j5gZ{TVH!asmdgc*jIhP>Ucm>9
zHF<HRK=dL@7ek80idQmDQvpVbn_vWFX}!dVpK>8Kw`WpuJyeDH|BP02LP#P45fj-^
za3)q3d+Lm(6u$vQU01p2P>Setu#z>I2cWek#B_S^x38{pfe{?u4QOMFtS>!v08*B|
zhUkf5LW=zCN@bbb0lC*<PUn)JnasQ51nv96BM{7vn)CalStU@&s0l?B2Bn&unSmrt
z2T}}q42gs_PaK?o?D(EO1KBgVJM5X9$1YYh&4LsJs>)7&!bg@@uMzDA;S-0ON6ydA
zi3oHYnPLakf+{L@-+W6d`RyOS{)*-LjIEo9-VwLkT$xKayS~C0O<i4afAbCgxX*0_
zz*Lfc>_JJEtwAeGba@A8>*ge_u-35K?Fh~h`i_^^S2@-)j#!ltczkfgK7ddxnl@K;
zDRTbmCA;krLcng;AT?TO>UGWJMhF4n0;<X$^ioTjx*|;vEE<WZFGxzGl||Tz*gIC2
zmn>iZkp9D0T)zF3-EK#H_7Z$gcerQhM+O&gu1B^t?!hx2_E?h>uzZX(%N3opEZP=j
z49m+G^xHd>%%K^=fpP=KUC;XRjAdhZ`1l2~`$)SwXHb?&Em$tD@Q?4oR5VK9T<#4H
z{SH+x2tp&mgfbO6c*L}asn6G1WdumsmZn`}7d=dU9=KJ5Uo=GTP^xAh?lG$i>cxtL
z#NXX8o~<yt;<L*OmjC&mQC)pTOz?}p{TUx_Kl0Uw8~)SZ{*3qcTR@!Hy|GLP**Bo{
z>&Wn3D5R~~rj;@g=AD5c5qW#P=CYMU=Mc$ZS1;HdJC55U+lM=>(Wq!KdP&kPM#25N
zFW4WBELTgKs%CNi66X^iA082Iq)!rk+%w!CFhuIr8G^*n9k3}7b~hjuX?4NpKlw9W
zy!o7ub%hQCv3um<%b#;=TJG*2SdW1huU_-^&0A7P+&w&gpP~8(cQlI?%2;H|5Oz)z
z6-8B$iG{`(j54KfoD0u!>>&lTF}dfBz*H3}C5GcZo6H2yFim(rLCO~2QGxJf2VbO{
z(L0QmIn~tVcL;vM`;lqNK~u(BL=3q8G4Jf9J}vSmZwoQYRP%0s<`T>;tNCYU(qW#^
z=VvkhTJRo`1VWL~g7VrS#1dsrDPD7TL?KVv9?#z=BMc>A=ZS(tN{MlF7?YjpX0HVy
zC8q4A=afJn5|t4sC6MK%L&{J=cKGATl`(p(R<w1UQHYqO$(cn@Sxob_Iaxr9`S)`K
zFwBKgDnUvTA3eroCrT-bU9tB-?*GpVzAENCDP=jKD4c*axBkj`<evXbE-ntKMMjFu
z#&E{yB6ZEJks(Y)LGr|#$d@A`aL-zi;#6$pG)qY~S&VsypH54CiRez*nm950w?BKW
zVlXAmlg)c9{4w%#CX~tNFEt8>@Y%RX>IP+6gv=Zpk@ETakR5k(nx{sJe*(@xhBhKt
z%ku1sBqW+ZEhM&GpoL=cj^o1}!!Y6^w95<Hi)$wD*>4}2wj(co@P@;~ds;Q&oo7ml
z&G|LWn@>o3!+6{xjfI={H1&$AKBHbXw5uhyZP?ZC5tS|dmdH0BZ!p7v+deXAgIa8g
zW;LE7Dr;M_c)0>UWwV!<*dO*7YZ><kE-%iBT4CA+Q_L2^#s?%CdHIUf+9AdruG^zj
zBDI;#thK_&h$oPyfjExryAGu#rpb%6H5IWrM=q9B;|||%(J9~_Zom&{saUj29&W$k
z@%9_8Uc8__yC8K(-0e-Cq(q>7@e1!FZs^%PK5+T^4a>G>?2oi+1;c@9dxwykc5?<&
z<p<mvwh#Ag@3)xc0`a)#@xyzD<DOwi)T<3mrD@Nv!PW>h;=_m?JNU4}G%I}Y*yWmW
z|3Duc)$)vS>@Z4WrOcVzs=~>NrpliFmOKI3c8O}wu-C7#(dB+|7;!QXyNT&y&FbZ+
zEY4O0X&AbJ|MFk_Db@O46Q_ay_rLr9^2Ns+wh#A6?-`xvn~x8S!-xdVuGVz>Bje;*
zt{PHESS?Ax7sA;S`&vthv64S~wW8ZTvX_>_aEo0ADr2xpAkWTFx}tyd?01i>&R??F
zoFn|m)n=6^X<IY-#OFW$6AmB0MWvC&)f<|{1-plDIoy8Da&^gJzo$Fg^X@KU>kZ0Q
z*v$p~?vZJ^q+TqE-heXn$H>d;Ym8Dn-oEF{4_~u-@ro!d+r#nu4Anol!@G&fTH5xE
z5TWlMNO6E9F*fg>w6SR2V1z)KrZD>-nZ{$D;I$(9K+5<n!RLGkk#gMa;5=GaOirRu
zND^Y4&?({lgr9n%(KK`NUW_<5;>LlgKhSr_lFFC(?Qg$kW|PlF<lL)%vfPcQMgKd6
zKuRaQeD2$3=y&e(=PwnvxdPk!VtpGzm><<rUu0E6EbsAT!c*Rg@3zS(=jwbeKx}1B
zhB{Ba{JBU`N@jn6f^5wCdLuIaHu^*qGRI7Z%<?z7v<s#Q6(w4gYy%&15td>$cFL>$
zNf{7h4id6umm9ugzbuO#a+V2(5Oa$mMx4uXQLQzia0p0wu~#R!1(Yo0)bzC2W-k*E
z^2uN+7S_ECxxf@c6w7HCN0(n$&Gy9kUZxoEenQH~9Gf14D+_k<ztA?LDpEkyB(eg+
z2ba}Zu183Nmhv<g#d5Et%*m2UsdBCvIS;AiYiNVgHCigP^Nhn0=RHzZh#(ldEm9=3
z7I>iw!BG{XEQOT+1jWEt-`?=@@`{TWFR53n%ydqH$wh3t#@02*kM9ZNh>n3M6wAxk
zgm#60+;QwCn$42KG+-7>wAT##1Hl;5*%{FbhWEcg;HcXcnu>beAmv0nbnJJxNUKq%
zVYlDm`i`ny&{Qp2DfY=@g+&a3C?lcV2z)@Lkb8&2k)#ZwZt-1DRjnAu#FR8DN&1^x
zhH0X$YogFJmoJGi($+PDUD3x!{4`QKh3X!W!sDhXv&xlXn0mC7G`d17LHqVkFc)u#
zLLmK!zaI!m;uFX?g0{rI&rD*a5!zsEi;sbY($IS1I5JEV*48Kn2BkQA^8=c*bHvn9
zX@$0yF!iK*gKw@GWWXKn&`M*pMwyzB5-JIjHl);Wc-XVtY_OH4s%nHA==%c^fmwoX
zE0TxhYRPf`NE{Csm8flnH3D3Z7zY*>uFtLry5i<xiwA^J_>_p#5n}^Y)nF@&Ob(GE
zLN2gLG6rxiGLA=TZLw0KwV>;^Ovyp?{EMG_&QCvkTN-}ggyi_}$j}XRUB~jp8TTLV
zd3<=ptTsd`F*;(BXWVuC>_7bqr!~U$oYjVl^98Hg^2NJ7YV`wRv*7INHPvFt_V$)Y
z$Kv7wsU@q$o}d5hZ@F8xy!q@Cs22F$Bi-FS%d2aSU;dKV_XGqFAODsYS6B21kS$3n
zY_-S`o=`kI-XfcZ$9`n$c5MItX9!bqwpsA@2S4TFvp?hJH-AqMusA<se;Bg68xv_f
ze4nBE2X`DFKBD4;Q8m?K!+z^<{h<)tJla%*2r&tWkxEL0GH7G*Za^TB(xOacnlb?{
z71%l@jZzikxQEoCRxL7VVhRk$5nb$S$F3*Hlnum$%uue7#E=M45<^DZ5P@I(<|ZRc
zqAXe=dv3_0X+Q1U#B5s|pV}A+W`@~`uTp^CIV>oDB_fPE+4r8vUuBCwGu`u!JQgFK
zfAEUWix8d)2p=-fz)cQoO_rO9taHyha*Zx$H@l0@i>yc~dk_c+0#f>})%P&v$mV5R
z;bLMKr%X01i<}UGsPoTcjeN*li}+Mr<-#iOgoTnAop~gQl#RcpsYnH9OW;sE2{;u|
zT4&qud^zS_x+0W7B`LGaGNy<Y`qU=zA<F?R+1p_nCt?Jpb#Y=$Io4SSHtPkKXG>Ca
zgb*{wqvR>15ICtp$u{dc*-W}DPLITVZ$%PFvZ5dbpYN;ENR=gmsw7>;Fcxl8Do9u^
zMuZSZtw=K|5l5IFv)s~;L|q|;K_D@9fzny}eR4beJ&y0!CQyb%WEhS(H)86Baq`5d
z2tkpk(00jYeTE-8`t6RpFMox3rx=vNZ5D9*fIA#%tVS4x(G}4-%rr6{4;-g~6v>gh
z%X506FuG-b_bvPT51=)+SrSG^r4^gal7*1;Hy=12JW-5zG0`kq!szI?dt#ipd~wCu
zo6qRFJ#ILVbd9QN0v4YX)8trai)t-O!S3-6V{4M@G0lpjUgH;6usTBs%`_dU`&%lb
zF)Fh{Q-Zp#sjMbAhZ2&yuCm#Mk~nuH%A8^EgQFjMjBQX&gX@nhHyc#F1g+6Nflx#N
zejKRVhI+Ne^@>VCb#_6LhNJ|csxaEp?YFq8ryoDk5vgm<qHVAuF*#q7%sd2GU472k
ziyu%KSYMyv_ebu&{g9!$$)kfu8HK0}hsPUaoQT1rOp_P+?ubO7t2#$Zrz27hO#P8m
zEoheKMDMdprP4(2NyCAp0(HGaO3fG(Rof!8#o<ZHvfX|}0qw=B%o?8tq8wQ)TBJ7I
z@Ail&Xcr64FE24wjczJ_^3#@S_sG=s+-y5uzI;o+y(2O5`tu*-z2rFd^!rCPtD36x
zLTg*|#W!D5nVcvnb%XDZ+<&}b**5GP{P<6P!Wf}@JhECYsFrIM%O%IhM`BeWjpTlx
z5tv_n^NzPa`DZNKC8Jb~(+$>o(%|U!9lKp7(Os-6{_a=5WO;Fo<^>-<e9PsF*R-oO
zhue45l_3fVO4GJ2?vFgRPzVJn(sf&e5+pa`#tHCDLn6_G(AahbF;M$}wHAp$3X>D@
zLiWp7$}kQCBycXn%gg0C<1{f%W0sjAnGOe%a~KqhMT3-v!|smJPuaFxYK*ZUHHe&b
zKne__W1MoII*Qz0%7j3QV_#XABzX}Jvvc5Vqsvpz|LF?k#U7CZl|y)Dznk4!Nu~cC
zpA{9%{ItZWXTZELlcDUBqCk8HDeyj!j65aw$zrK2BuJH!14S+#CLhpRp-U~Il*oAj
zJ`x3l(t?YHV5p>I7&CMpQo@HsN}kH7oW$y;@@H-bMN!I^)8d|Y)U$Q*tlQ5F{FK`@
z6x9)&5EZS}lbS$0IV#SkO4+C?w<%_{!1?T3Gb0O93POP6v8S?`+Yn>Km-m=UIIR_{
zMMG_K96ZsNRK^@-PYn54RtnNYN}*T_=ev*#plqb1i&Z@-F!+(+9a=3&NoNUQ3Iso9
zW1;8~N`o+f!i$NRcF0NMraTX8AjF=OCag9DF_Kb1DT}hU<jG`N^3eT$sN(mN7KW@H
zZ5C^$X`<^#n#GEGWvCZxhyg7Vhbb@)6U)^Z&1S){%d#w&EMpWjO~XD4QXg42ibVx!
zIwDY<y?lXAg7LWH{kPvDg+eVh2xS<}io?x2jJQV`!^_Qru>DB2SRmI+7U$O-x&yvH
z;)e-s467eq<BkWU2*h#Ei}NLiePY);QZdeXaeaXwJ4liKI5PGF!n9B?5JnNhgo<0z
z;SsMZn#BsDr>zvOJ0Oil2z7FN0-#zDQ{?g8uhIG|R*NNL2ndyT@#Asec<ivPB#b?h
zY%*eWlY3(+W@8Ym5ZZuLjKe_p_{gTMQPYv*{zyzvFIPC{8M_0GioCkoKuTFUq$Fd1
zV0YM|wPCYdA(f_Gu83ogdw7T1G&Jia<2cbBdqiz9b;C3rS=I)SXrb_vM>h+ED6C>B
z@kpxWIo=KI?mi%-!c>9&;RaJ#P#U2tu$m+ij2#XQLC`cUK1Na)5mn7NIn3$|LM%?-
zih$0f%<~qdEw5j#n5Ld_FG$i5fhr9&#xgDpzxhvpO-P#l`jYkMpWyHBX>CjIdz$qd
z@PUv53u76_krW(%`DZ^Q2#HaSZ{B|-)(tPu&N2RgSmqSU%a^Y?P98!;SC)tS2khn*
zyRKtlHJghida>bkRkJu-uzk3}Morfr_{~?pVp%Py7cD6X9v&Z<M9XJC{uA2O8Qbk6
zZXCG#-LDDVmLw#bi)+~KxqsNPs#}(u%kMK(|KJYOtXZ62g4Fo&h$K=~6)_qlj_$Ch
z9|o2pL6l5GkJggKVvR8kLJ5M;Tmmh1HpC&K>L!mlDQOqW!rImhgU62!*9|P2B?yDA
zH_#32w*!M6N%Sb0gK{LPBy&!D^ZvkxyF7u0*-elcqR+cRsT8x}PBzm?CrJ6^Zy_>N
zEXsl>;1o)f`_V<tSXvgbK=E@*;#@qOuyCFh`usBAZBfidMupuCGLr?pcW7<OLa0vX
zGryjavgq^+Ax=oNET2bZ?Y~48-a%0w#Kh=k9>No;Z=AABu1+R<sCbVxhJGBemCY@b
z5Khj7Ld-3T+!jz;<0l6sY-RB!Jy9reYS)NjU#yfSr@uR-D6IpPH+FN&CC11&1yX{l
z@F!%sY%wN|-B36Sv5*-vQQ8`XF*++Brirf0ZI$e3>goE96g{@GXqPSXQ%p#0QOW|5
zm^>jkyd)QzQ4o`-@3tsysj3A;hxa|s^-M#Dn~snIN;l|wg*MqXT}U8gxuQ}ww|Ytu
zm1034WG*bT3-CU(ZluDye}dG)%_l!(v8WN#NL4EMzB@qleEajiM_5Z^D#G~&&MsMA
zzhwLVE2iBYHHliB^X>M)wCnlgPk(~%2jX#$y}Cfkn3<QufzTbnM=B$UNnnIW8cVfV
z(dyTTs^a}OH|(}MgflEJU$IzSFr<V_ffOg&I$I7Ow>xxQV=BY`=3BO3c1&i;SS`_7
zVWOiy?6N7v=A5(7e#miu;JAN8ug|guf9SDs#Jd6CO)T{_!3XyJ4r48qSyF<n3)JL_
z9cb2-JJ)gOe3mjwgEAUzD;7y)sfx-5O2IpZkKa(WOO_W`D53IBr~>Ck#{HH<x99Bo
zn)`<>V|OHkiGJ)bcA5R(Q_Q3|AdUk<YbNKJ#*xZcOr3Y)q1!U<9~k$0#$W!DcD2ry
z&B?<s0s%J-?2|(vNiHB&PK2Ytc2?&Y;20BTv8KVWfA~o3cK85E3Uo|tAHJboZ;;Ib
zWCbGPx*fjnve`;iQE$$ustPysARNl~Ou=K?B`!b|hA{}xiil+N0%Sl}6}^wReMj(t
ztIvN#e>f1wf&S3rePHMg7^@lk4vC=eIzrvBeEm66Y5e1RbaVu%IX}NZNvJE$)O+f7
zfw31PH}Pg!v%h;J^#`gLsM-oYP4tHyXRkkJ_3|y&R7k($<@thN{`}YYeaG9kZ%AIy
zIZv~B#pe1gx4-=b&BZ0#$474NA88k7?1mvjwBFNq9Xd{I&KCIO**$*zK11~n?r`Hs
zNQTAog2T<%narm&!YX3bGIo1{KQeAVV46)vSA)Tu%qx(#MroZj=)n_W-VN4Om5FY6
z>bfDJvVJ=zbhSX+1}PQ61*EiSdxr8IQMZiak<>rZ)D_F+0#)S#`rG?1Kl7!>9-}WQ
ze&j`|EZ%u17*4SM9HujS;m>{gr}aNC!14JGH09RAyf{md7w00x|1KRN#_V4p=P;%x
zC%+RvVLn7=>W`Dl21GJqls;7xN=f{bo(zY4U<v_~@(w-AC$pMMMMYAweC9|98A(X7
zDBKgL{(4ARIysWn1;(foQ85Ki@LA(Oa}{RiznM^%GXe_qud)MVCPn5q5#@7bgrej)
zNO{UKnCGYY?<pme$e%BUfL2+NF!^xe7x)skCQjx=xox3}bX%It0T9X-bQSU`<|M%o
z17RAnlEF`SnQgR_$d!{!nG=-f)I*V5ikvmz9b>m6hN&nWblFg*oO+uIVX+V>wa$JU
zLW(l?f{#A)_oN_(0oQkUH)OkFsgS>)8}aum2CiPdrq<9u-Z0&NBzcb?d-Q4rO0hb>
z;Oy!$SB>8D;qU*NICK~ZNX6!SgOxBIM_QvnFA36Trh0Qm-wmh$!d94t#2g?+$!fi%
z?+;iB?Pdv8gHaXJ!w2kUg|kakb4KWn><%5n;}%~J*t*6?=pXLrY=u&a&I=xQ57<>=
zp(S>+;o`H;@YBHdVMj<ebo(9A^>n)j1h6<;W2&0gRA}c>w!#TV>;(OBKvyjyWp#rv
zCTmos%9_$7nWijrLn>kn=p+fw(KIzN1eDT*;Hei-tuL{gE22=i!xrD|NO6K<4SfCP
zQ<`>(9}}Xk5YvEjo?sMOWajG7cUYw|%Qfbm&Fbs|odU~7W7~#t3=C2rnk7*ij$^`B
z3fr`d`)!$k6hl7{oud|#dZ}?^U#k1at~)RreKqY8Q@32cd_kBdLezvb;wMLNBd+h*
ztj?Jvgy4`N;Czp$8XDCS4u=e%rT}5eq|BI!LCeiqmLkQ8aO{x5vv~0mk}5l=fQ*j)
z;TEYR*Vh+FQ=^PvvKr3L2rf`9vI&Zc5icai?JXijDgg+MG?v4-=ltRvy=cHYf{0As
z<+j&;hwCD8wc_mR?9?yb{`LlU_n!OlfYvSk_8Y>qrM>u+xY8ID=3>QWa*XbXP?GD{
z7YJ2xJnUI%h2QTGs};+|l83K%bp3?V70xs?ixwWX-)E@)!5vjq;rcxX?;&_3iSDq^
zyD!^PHw&V&jD3%rT;BCL$H(_KxTIN}pQE*<tt$jsdKO}~*p)(`*xHq;v(2<IL@JUW
z5k%ZLLiEUzm?)YJw3oQ)5>Zverlr~lb{}u~w?F%jS=(LE03B5hnwi=96xrG=dkYan
z`p)j2LNuK1Wlz2R`H$y|o05P;gm8MMm<e#tiG<JebfTErpG-WTAqaD;AjE75tfau!
zRsL?w3sgoR3P(VQCuE@P^yOTT2#GP}Q^!E?0Y@MuS#mk@u3Jb+3dKz|<o}zwZ|_4P
zTm~Eo=fa7JKer5mpYpXNdn4pUU7jM1$(!C&dtmk)c}AhgO#1S>`5I)4;>43PWbmQ*
z2!M~-G)b1%Q-YHxgD4z}%2r%nUK5hfb9%Y1##9IaN+yyY&_ZC_hLm8M4v3V|4Xql4
zsWQ>r2b9({?V7B3&c;VoRpY(N91c;UpM6GXf_J!aKpBg+3yiTv5*ty*<W`T;g<lYv
zhJ9``2uW}wX&OlWpTK9}@NmcN;}$n{ym)nms4P`$v9{v+^`}(L3Sks(7?~dLXlhL#
zz*Y_8ap3-oFLF;rDh{_<l@N{x&Npj9a46wusz_@rHy=JCfXcLlp+kj0cRXT^!X{{H
zO>_N{5F^K9$2fh<&>vV<hVwU{;vys=>5qG)(imf?+lF>?&N4s>0qrMhF`$y>VSfN5
zzV9$fu{wK8h!dmppeus&ND&ZWpx;f@i#3DLjNVgwkClmTe<0}!bAFA~6~}(S;_%~`
z$vH}8XRqiH+JddhAS?P`lath%#fE9jAuK{b?<QgltkxTBRq^o6S1dP|teP_(Z$Dy9
zCQGH5SX{j(O-C3zqNy2OM5%;Pf&I-bUu{2<gk-t7z%~t379lmkdA9d=xXH1)yhba7
z4UvE%NP!QYp*!M2Mg^2oC?N<g%W?X{7TX_DN)fd|EE>eIhv-?hHGb+ieDg~V!^ryl
z9GN1zZZPd4Ym1f513d=3ZnD>Sf8h9Vi!g1T$b_I>uECAyt|y3qY*tzL<MIN6^N0|z
zHcOnmHmuLDv51V)Oh+oCm?mGyng)>s<1nG6=JOx@5Ty-a>e-K;aXeyGCLgJ~=Hk;g
z7;V|T`-a2g1FCIV{ouz4kx)M1lgC_KqNfFNwIPj1#^Vu{9Npo8e!B%@X*O$w6r61~
zw8oNR;#WWW4R7AQX0uqZ-R_We%jU8r_=#ZZ?=w{Y;0|poDx+}25hbB%mc)>lhK?8#
zS~skhZ#e9C+<$zBV1$%79vr(q!0nd0wyYN|ZEdmEWV%^`x~{T`4B7YHHVdX<$djeW
zDR0Iah@RvFGHM3r5p|1dDugjaojpJPoB!8ear1D<&qX=+<ckNu>@-+5pD_edf%E6~
zz`W>Zvz!tVBxEiU!gt~RSv4?Q(ax@aQz6{t8~~7r++Ikj1nd-KfI_)@2Kn>sZz&Z*
zrVx1{Cl}YD?2>chpm`=6Oc~Y}dG{+yVxf{*QIK<>(n$e9N(7%tmm%hsWGE8GNDeU*
zQ#>u+^ZSMXN{CV@xWd6GXD-$0+`J1<g;1RRMN}aX7OCYkD_-vdK9sphmKKeq_#8Y7
z@j;*jq@tb><vLcSEg)qk`ay0*tTr3=`v*dB`Ew$(v@`|$G$4G3kb)3I-q{lfamWNo
zv4E8DzGE7Xh@`2S4N{j@j!;AiCRU8ogb+|QEmBs*;OTb{2ouruf_8a{HX7F-5HaQ!
zx;6QFi-`9FN+~L1!8A-m&v@LP_$&YD82!C|svN)kkN<)F_K_cd{u$GwLai1U1pAvi
zme;TFZX_Ki#%|BF-{Sm4m;&E?`!)4qLAz+EE?=<U?)l=c|DNj?S6sc=AS;7pf@z|2
zJ+9J4N)r(>5eV4jf@BTR_l#~pVen(fHn2mF5abIL6UW0N##Sh$X{10_Ep=rG(@0VQ
zYip)4n;1>9qJQW~gTqe)hr=U_%>~WsbA*Pf>p48!gGwL`heO9WI6nU3x2)D1UVZ*4
z_<?Dh*bftCwV_(SLg|WjIWXOSO;s7jp~pK%(>8dK&?0jHd<a;ZE5{fD!`*w**t6ds
z`S9z%r&=$mn~Z7%VQ@+3xE536Tw-zl0<9IUABfXPwOFASHIMHO#4ypW&Tx}u?6#~o
z7wjLmy#M%tYX`bnaQ(@TFjd9accgKoZd**VA_Rw)nY`&^$C*{&9eYzFRD&BbsWC;(
zvFnj3;$>nu9y6TP@3B>dF0rHikYBr5Wt1iB&JDpkj1&a2O{Yj8q9+K0T3*uV8dC24
zp$xKaa~fRR5QY&UH7P-}z9Og^)B~d%Sgn^?mm6W~CPp{X7zJTOhXF#sd4=7aVSx47
zIl3|o4-aVXKm^v8Yg}C7x`F<<r`8o(W>-5|TU47l0KrXEXIG?dkHI7A4a#rm4i7ZT
zhF~Pq{R2`8s`aLji7J9Nw3`LGs?gE%<JhyhtgzZ*>lPsb%01BS9sT}~TUvPRhpZR(
zIqG}x0b8}C@@!}$QI*2h=frf28+J%3nOxv-=s4a#VrxabynxM`aa7b*CQ2b9wW+Yx
z0_*}zg_f2nB!n#{J?#RkvQj|?U~*-L7>N?5nD{sU<G<yv{`&8L8Jf>TKS~i0V$80R
zaaIL95ydhBkjZazuRq0XqkWPeic}cs6d1o`*oQz8v$StcZ4@VyA0Z^k`@+VL$aKQx
zpRxS&B9~HpR%zs2>@5A0Iu{k*dwhs!A&^QC+>9lpLViyLde1zl{K`)|a#IR{81i+I
zNuaGNN&!WP=@d=uOEO_9#Y8I125YiB(>Y(92xD<3EUu4*0O%(F5A2xBwJX<D#Oz-X
zN>QMcDuqE_0G*44XF%Z*6bB0_uo-GV3eZaC4T#RX15qH)SPGvE$w#C}NTrKZG9jWv
zNV1s{fs`E1ABcXWuGbiCa53QBk>GlaS?5AiDMV7lVuq!p0%@`WKv|4g5aY;XZAJyE
zMJ{g12#ppc?NKPCF(jo>Ug!3kv6yO!Rt+Rc5Yz89M!#1v@adoa7(eu!Eh;o1NgN(`
z+<dsBn>>%(Eyu?#i@IWce#Yj-C7U1rDHC5%*A=#|Y0s|+dcn<K{eoZrPk+r~W%<jW
ze#T$?`A<1IhYJDU2b44nhmI6FWM!DVLn_Pu_L0T1B}E8B$JjYcTOk!VKjK0n0F}<n
zV3|CNas#z&2~vQq(5k{69+`GKOw~{?mo%wHSw(m3d3gUN^=d_3FNkg+M1x(gk+!2Y
zfu)q}?;m*?p+9z<pRbY40w)`y&*UY6$BjPU?4|*2kabN2LJHZ6HRd?fr1GwK8V0)I
zNK}f=o0l1ylKEkOyuU{`OWO07xmUiu!%ewOk#x)Qv!7ruUn2cLL(Sp+Z<&q<tTC9%
zqUxH(YQyryt326|Fv+2(gtCUwD5h@D<2Og9zNfu@&EoB!QP~F$-~61muBexms4HUE
z)9oLa#skquhGAm4Tx4~GsR%>QF!m_h;9|lqSJcZ5%SDSH`<x{&RDR$?_F$JO;dM=@
z&l!EB5sD<Tg|Y8Pq=i_6*4g8{S}t-AH8^DQv<B)IFHmkoYC|72#(<6k(WbmhFbi~$
z2ov!x@8ER1AdDUU_`q=7XNiy$$QY3ks>%>UgHMuibST?k7H8BU;fDjh+Y`nCp(KZH
zB#I%MPIY_wp<}gPA=(8pBrFLf6fQ;j<G{t+SG4Ohj$i$X&E}lNi??Vk(Z>V2>j;C(
zp--#tHB|rLj=Q^i8kJ$!vFix^K(rOX1;##*f+MN}$coka9F{ALF<c3Q`EbwRa%x+>
zXlT|~C~F}&goYp_4vW+)gjq3><7t!3$d(Y2V@NomQPyH@gGH2G9*`3M?SK2<@qhn+
z{142@eJT3vQxSY2u1SQ5L?2HJ&%9{o?16Y<yw7plrLE8W&wL1c$>mTgYs@LZIN3)(
zFSuD7p7}UJ&R+c4geZh8J>&Vyo;dL0Io~iBtEF|JfpMBBDo3={rEe}y@9#oHg-E6K
zQ-rW6e-_yxg-;O6g0IOaTuKomVRD)NE6QT)X9vfWzov`Ba|j_94#jmZ#Doh0ts+KA
zqR4Ve=W<%1j1g-MsWA4lRkSDtPdZ(@m=eyplj`9)VlyKe@>vl8sT2X3dmyzs@i>Ic
zye&jPh%AScQW3+5QU;;VQOO}iz-oz$6WZ1Y-C*lAs0xu>PCC>rwps&)3ukL&j6{{g
zjZ9J`AyC>9gTuLuI9R(t=!_HuH!$=&913G<LITq)(0U0eVvI$rD&HfeXjWGwHU3^>
z^n1-P=jYedMx%nKJ8bbnQC(cq9tJ97I6pt*Y_n#$TC?kqgi_I6yna<KxI*}zmoF~)
z*Z+%u$ydMr1<pr=$aw=|u_7HKl@#1P-ZS<S+C%LthS3qFMk!E2p-?pESJ-n&ntH5;
z-~u9KnU2y5KSs3D9EX8QDooqLG@*^etXg)*0|~|9up<V?YIBCE8dfr=Qg!<sZk(vw
zB_ToR_KedJZ7o-qSG;}m2In17X%Y(OCXxvx*D*{ZZa6aDe`IMjrfo?Ac6SdL1*R;_
zNea9lNK>NTELgQ?^!uI|1A~?%WzkyEA9g62vb3dHa@gGxTqL5<QeadhKHlb_ixO~Q
z+BIqFNv@|6kt94IX^N9uN`Zd&fL4mA4CiMTq|F6E8TR`fE=0!dEy8yUgJYaLc67Mn
z-|dHivD@NF3~`CAUeaEjA;z9@7_xP9N;Jy_RoxI$!r2;q@rLp74tM_!H}$A?lidts
zj~g7(YFxj^^>@UnLz@cKwAAf_x>^+2D>KNW8vxHZPN<mh<4Bs0L<9Q#8Zq^_Fk;OT
zq(LhPohK=n_K%4DBQknq@+9vtb(1Uf>6pnkVL}N-sw)Bl{78S?BZNX%6^Y35>Lu3q
zbkl@ftg!78G4^c72vPC%SMQiY;QG~DqR+j@Y4QwH&-TMNxmQd!&Pj(Y#yg}>^h3uu
zO`Kg{G5(Qr22xCHw>xTUnGPM&2civ`z&UtCOf1eTY}IDZ3sqt31p!HQJuxMaGVhFp
zGC4Oi_)H?x8kZnQN$$LAoOeairB53DDR{)_sD#E=6{4i@{XhS^|DONr|K<OdGY`ZQ
zFMy&JIECv_`sqkgDABR;>378QB7k`ZI_vMDVE)7>&SCx;A<6=ucj0p}A;`{s0(Fw9
z%}#$x%-I{I+IU8o<JsGl{u+tQy7$f(i6O-sAh#xrHMq$a)1%q$`}9gDbVJN$JVo`8
zTMdOs95N!JBt^B5iy3?(CP7HW4!Y3!Lr72F11ayrXVPWPM@SInLMY*UMpmrVh**M%
zieJHrOsK?(Kp0a(!4uIkrJ`Jrxd@o=!E9$P#3|rtegq(+G2Txg4Rw1?5}GInPz_pF
zC}P(9iykd4+OD6TpOlcY!rDzS8PbH1P^Lv!8$<}i&=LKF7Y?Ocl(cmHM}nI$Rg1PQ
zQAi{ar5cDUlJ}*pnw1z5a4{nV&X3?7N!r54_$N>ksF2f~MjsLNib`pkrpBg}ovD;!
zm_}5!;27cY=ALfn@g>n}v07r=8u$JkCOCfd*=rV;Z;%?s@j!9`-7bm1;!<ITBMo`M
z!zfhUB4x;4*h;g#xuZF|z&TI9+hc8ovQ6%RMVV!MO5q6{_6I6mLD$pwJ?2<rti%U}
z8$HI<pq3mT-yu@qY_URC7G1Bg+R`1i2yNN@_BXh3<ov}8Sgp8t`vbb|J%`;L{pgT3
zW;;~@VLBir*rv@|TOSy^9u_q+!tvNq)eTdakTEjckF?7*VHz_08585QW7LNA*=4r2
z6dENGc6G*azsv4dO5&%H<MF`p_&}U`>ZZjX_TW9o<2{Gqz&qsXocii@8QT>q26Pzd
z-GCp4Vp1fzST_9nKmHZk)SSO~NqhN<6chd7p0L~El+4#m8~Sdb^Nw{=q`GC?9}(3O
z`Ym%Ls-FTf2@ZF^rK-PWbQ94JOm0F91ENT~yi0`iTc++2rVc+1pfZ<YdGUhH)oXOK
zqTXDBPcR<w-hnVswIBj6L`3uqw_gzfB|xTxa7X&%h>R0vvF76ZEKhvVWq$=Jb8@6K
znW&Qzu+Pepc1dF!!r0*s57>5rOrBavgzNEc#OeyBiEx+I6Y9kqhVFpS0(3=vc1F{#
zkkX)K;^D*B41GtlY)Ia)D%=ROsENi9rxw{eTuj*=@{jmXqQyJO7$rhBNCT!_(rhlV
ztAP|c>cs+~G;LdhSs+af!P7PkCJItiOhhK1@IGN|Mb}T<b`dQkas(eoY?ZJ|k;Xt<
zM<AjDv{jxQC+7;mEb{YT{f7Vg|NLL$oy&dVlu4{2hN6ql=Re2w&guYJ828MCJHaV8
z{!~m9_&R1R|MZ>jWY&jVq$eRzh!c}O2X&=l7BnYvl*Mai%;%MiSwH{O8-5}hCK8j6
z#FXWCT4p=wrnY6bUr>*jihEv$_MM9uBeGR8PXx$$(M+W<GL=2S{IdLvF>{1(K8u`W
z7_dg?R#8etO^`pIkr`F+MTRNK2t~Q_MO~4P5CU$>IXO~3vsD(^zhO=>)FqK{zMiwB
zP!+!cCFk}|DcH(;WuW9)@sqMuaEcmZTdZveJ`$BBiJY&Yw86F)kUGM+Ln}>60YB~u
zen99YQda~w5Q59^9|HuB5D{X)4OsyYRlrM4@*b@;O6p9wE>Dy!a#3V<7u2GlBYr~b
z29#mycYwf}rWhanejADJRScNr8A>Z^62T4h+XwcC2mCY;l*Uvw3V|CZn!2WYi2UNW
zKjY`Wd(YYV1^>(c-T#`am!Gk}`G6G@)8T+!o^jkgu={Y2kdo#3IdxN`ghUP_kB^Tm
zfB2`Y-~Irj6d!*13xW%n7$HW6!vWni*g8^Gn#l*YyDjIJmuM{r(GgN)xjIK`%TbNg
zRYPSfoEwlnP$@&K8brH79T!+7k@W(UMpa8tnrZuht~5$lJbwIu^CLg{&;AurWF2bs
z4%IACDKPbWnx>^*t<h)aER4bXiE)3-)$wvoyIgSC9gxoFNvOCF1`p#nuz!5Wal_Nd
z`r?vBQ)RQ2-VqLalFsbc*}q(Bg^7XadfafNdwAgC=8ns&YrI|%47i(*Y!-?!LUa?$
zwkG+J+i%}fFIG@HrZ5qAceoI^xVi*s@DYS<!8Rys8IC<vHBwvF>vP!N<KBNn@4sTY
znK11IZUW=`>^3L_rmi{e9}pJCBGV*H;CkYidyF|;D>4qp?5vd%Q{NN95o2ubLn_Tm
zDO_){+A#G;kP;~q<8F&6&|=jt!9{$xBP5S76D~xwGDW&3&{a!JqL{`|S_1*^2fUk@
z#sNvh$3SaM)}xCW6DvYW7^AVvB|&Q#C*s&sHw~$&2t<^%tn`M}BFi~d+u{n*EDR%I
z=%{Lqsw$2`!0HUYeLzOglc#&!v%0t@xG8(|{}CVc>KdH_QfO?uBqoVB8g+3=6#~Oy
zOE48u8&DOdUS)!3O2`<|5uEq9abj|ghlgD*PMzSs%b_=Q)6fqCK02(e0m0d_;;gMm
zZscOqWCSK;6Poe3<Nx@7_&?#MYy_2;+&no`%x`AW-bn>eS^+*~y?lbw+ke&;D4$sr
z8`wzxx+?v5DFuk4+mD&dtc*D!5Gje1{BTb869SZ0B$u8LfSH>xzu&XJ1E&Q%ro`mI
zl-Opi)X6$n$zqsxvRalIp(=g(?>ZcoMKYB_rP%h)DuonFp^%Dou~E!MLROcIfpT84
z%nMqVL_^ef4TR$S88h!;MqXwTc22#@#Y2oUFCn8OarQ8nNrX@c&-vb{!ts!$Adm&Y
zo}-5IN6nulrdaHpBO&(mN0)g`7_<opop9qVVMv&2iBWC2zmYg~nbfMQ(gMowtEB*U
zU<zY4K=Phx9CEwZ_Xv}b=9CoU(C2op&IQI4rmR|%5~XVr*%~>M<SR&qXk1P<EQp9P
z?e8^4zh5z+G&%+R*yD}|+%(Z%UNhX?kle)La>*EE-USK4;&|joKN4KNdBtIOWGy01
zwWRr^=JEDx#+y6d{o-#3F|w#FQEPlZ^6r~&SgaRlQ!z=6G!=1oBu3B6s|&Wfd!&@q
z%a;Cl$mAp47Gsi;WnJN>NVQ&L{gHaz(2bGp?F~u@I_KzwBn3xEx%WJF6XQ3(Ly)km
zCCN>sv1b~32A9|XJVC?0Kirb;ZaBZ((9}OblBGlA;SoP|B&{(ikz_=>fFe<8O$wgF
zen8X>PFKiw$>co!%?+eP+cfMRcW7H<8qHT<e@Su!RtcKr8oOH28b#`kh~*WcsR_X|
zx$K0dbtdBw9OL#DscXRS#jk&XG6Eh1tCufPwx+&*fz^t1Jm9B+ZtAG4X8q<BDGFTQ
zp;M$AJI2Z70-&mpi#2ibc}HvnW_`ip?I(oez<B?TWmD6<zGhJw)O18>&3M=&qNi>u
zw31X+O}|$}5`LPp3S{?y#8EdZ+T{f)B_3`+;HD8`flsADNb>Kwk>l>3wrv=PJ)_Iv
zQuXSLZVC+jp5d@XrO4&m&(X^ZhL{xwGEDg3&<n#93o%Rxw9y3N7>_*+9a<%Xv3Tc+
zKA<bh_0=V^s!`fv%o43M!*S2?uw^=KGZWh9gfSm787O*(&;hyJ5Qm9II^6LNk#dx&
zh>leyKu28vNTq6+1a)KRcaJ3B=X<b!U>uI80GU75quE?xm1LYo<YI*snasI8?5P$j
zSY6|!L@0q2sszaxjFcF%PCShh<6&SJ1Kk+-@cssu1XGf9<A^Z^Z46!C;haaAiW?(2
zTP{H)F8d{J>KLZTFaG+k`R&jCmgDv@N8F01UVn-iwvHjAkeSI&fyK#ia?C9L6jLUf
z6~tin+Xo;$?dp>NRnjsMAk$Mz0mx)L(n($y)6<T2{y8NCUKH|P3}ulMC&FMh<Ke{k
ze_|5KvS?<D+Z@yDsZ@^bb>5wbh*By-bj0#mavLLi?g}Z3Y66J4Cyk)QI6pn0N~u#I
zQ4A5SR9?(tB$9XHAqBKmCj%qrGf7w%dtO9V2NcNV>C==2-}~%LXpJS;4rOnjG9e-s
zVxcNo4pNG&x+?wzG6`aOY75LssHcR;m?*zmT15h_0!nDAYR%+^(q<B*<Ooxj&F_>&
zm9~bmi!2AMR!mcu!?r{=0UE~<qXNW<Y4Z90DNP_@gvM5No&#lxw?-gDDN;*bgoqxk
z3=%<1fobY;k)ssB=Yl}$IXUqkKfYHn@a{K%2V)N;nsz~wnx<M}-Yk&bV^?crP#^^l
zxA&N;CXIVuUTygN2XB!fB78^p@oQASWpj}$mP6NLOt#}Z-rO)g-m+{2o7X=e5xM*3
zcU-=D%jWWu@wn%_X{cqw#l-sU=Zpb%x9>Ra57??@dHsgw`cqU29Dn^+9PSUW*&u68
zw>{7-me{HxCPNiLXiK|PjNK7q7feb}DZ%5zmcAc(*mfiV%eLYA#S0eJvitY}cmIxP
z48HHEmS@zPCCA;7dQl;zV(fa3+q>-4cRVn314>JjUXnsyD3i<&_ubts{Wx)VbIXgD
z*KE!&IUbI5+Z}h&vHQgH>ZdP>!+_sy(U&hV7q4>2g$Tr955qv}kMz5Hv=Ll=`U%?9
zD5Z)0fqpph*bkho&!{Sm6(B;u`<Pn|!7;?hlz^&Xbe_B4{F2Rb$?k9@#K`6Og3WS;
z=ys%GWH@eV7cEVbd$4KD=07UJ;6{`ZOny(_5BMpESJiFH&`p@C!iZuAT6EmmI3w%G
zg;2<5jhH6F*g+Ub^}u0!&veM7y(FMsEOXFLKd`&MCH4b{-JXjd{DkJ>HE2z@`v6m4
z7MP3~PThg-VaoXsDH6w#ad*qv#Tn8VDkYh^9ZE!Mt<hBjDKhmtj7(Gly6qiODXgw&
z+8U?xWRkfmJ>hY{XiYS2?qMc}k(qg3FBeQ9Fdnu@9kFpDO3iX(NJ4TZ8`9u0%5xlk
zKdtZ|>LC}G=*ohhP)ebdBu)lDW#cJj76`3#DrM1+i!_7d6nR!)e;oPZn~xk09os|4
z-N%o(5I7DaF$C`3eaSTR86GXfIHfa5Pb-O10_Qxt`+Gw0R7TTSoflQ+-S(^&D8BMV
z)1TvbO9G$-kVsK>qlJ`4f+U}8azjc5YA^d=rI@YDV+tqI->eik)e$q9P|Ws2h2<{<
z;aNj}PV1Zd>fd1xGV>V9m!&AW{ZzUb0_Pp1;@&3}lXGP^A4y3SC_WS)0bf)ExsaAc
za`+G5Po7rMl>G~4bfpwrsh}f*qM#uACTOjxtjRwcih*2u@*_y(*PGiRSvinD%Iw@Y
zM;!~1iW*H|i(D`p3TeC-r@znA&ujoyYO_*M%<qZJar7x2NlBnhO(JXgg%mW+Iw$eg
zXK0hvPu>Tb<qBggAx=b@WswSnG8HC_1>vcwY?hG*KS2s46eI;AA*IRL1u_#FqZ-Lk
z#+hdnV<t)pRRWfZ<uobVZ09!BKY`DHn+8^k1x-^U6m-X)pa0GOhpX4Gc=O2*ISw7e
z<2}3WfxEjsFE3YYuC7o`hTa#eHG&b@SmcmlDqA&7lVdy{sj3Q$raC(Vp&1TG7H3y%
z&J3%oORA<pXoH_7T)#uL*(W{@BW+``v0@kpy4!D&F)>Y!U;oX|sVmLKW!AV^tPsY6
zvM6CO#?l`ikRnmn3r6R#?OBfNtvYI<ITM<sB(Hz?C#;O({lf$M{>Zd_V6|K_^b_Xj
zSc{steuIfymTko}jF8S4hn{YCU>YZut0ndN4B;l+&~ZHWXk$3rY^aBUT5A@oWj4oA
z(5{v|JnXq_HpF3I+}>r+{*Pa>xST*$nP1?0ns&)<dr#LNdHLogi}MXq8rrI&F&iFl
zZ@HN!x@jO`uysWgf?>${<W=3$Zq`(OB%n%a95i*y^(UWz(io8tDG;ZLT1Vvh8Ki_!
zI{Q8N#M!HBr0?Mn=mv+;U<|pHqASM9;iie&R!seou|Hz$P@*~oN@|YNk$SbnRu!ns
zu8zSoOe68(mZ3jnnUsr!yL+_OOp{|8C$toN{N`&OZ|_*1Ut-%O=a=Ugksu6&qr<zA
z{ml)*t~vk9e}zDCeD`ZU{QU3neNVf}g{lvM$&XacGP^~R5W}9qxkBKI93CH7RyEQY
zq{)5>&IM{^OY)oGaM)q18leo+*dtS78jlPiB5aGSYkcs?amV&y$K~7qA9errYulQp
z_hG*;jxgg2JDprr&90`KWQ!(fNfu;SmL=baFa%IAbY=L~fN$-;OEO@af@wk)DUeMI
z-Re}`)zwue?zqBCBYn{q-<WgnB4PTR3j$rEaBxoTz19q4&N1HiecmU0inL`#qb8`$
zZ?>AC(So+t_*8;{v=U_55ZplY4uQZ3i&704h(6!}d<k_alD9<6FgVZPCjR1&|9~(5
z;%9|TerEX3&v#DCgAWQKVm7Xk=aHKb0#1NH6;y(mh&Sunwb8$}i!ZqX?^Ob4*!lu?
zt68OR*|C>tQ_hM3W;2`M<K^|0g7$uzEs1`G7^kHCAj9B;DOQRL@l8;Wv-eEJ@>)wl
zD-Fe284zP4p9y>+6#m2PoO*`aOYH6$WSCbab0cuxg9vdp1DfCG93qrvm%{gOMe#{U
zlpvq^#g@X&xyZi{9xA_8Cf(X<!c3Tyq7>)z%HRy`=Uj}QpbQa0ic15MGDSXl;dCS<
zx+oHIzPJ{i&nf3~I+GvE_e7%W1=4up;PBItVR(zRi(-rwi_$@Bi_~SJpN0b|IBc^l
zO#q6fxkZQ)0gV?VrApI-D4qhTWD^Ktu~$}Qr9hT5G9%~aphnJejAAo9LQY5&N@%2<
zqrP)7EQ)c$4~Oq{jsBoJjLs4RSgr6eUUCMiobXeJmIjn%vDsjkOV+huxn8lme_;3a
zk@4XjcCjW66UY4ncG&<>Rhr~A+nZZ32Im5s>!(~ldrr(H6I#u?AZ2Pyy+U2xmM>29
z7y?d6)N)1pWXZd4zvk`#_y@?VEA+>ou-vSehJpUr@#z;|;9Z=*ETwytLg2>{sWT}g
z!Zd&zAUdSTSQ%)V1<N!M$B0h>-L!b25X~BL&#@o*@?ZWTH$VCjpFY1OH}eHz47b<U
zJpEX3|Mm@!yB%TjOjCj6i!q^OawF5>#NwlmY1)>%U;a6^Zm8F5u1rI<UUPVO#1DI_
z^_u?f9lJN*Fm@fjAJCOyd3{5>Sh0NioaJVXZ5ymq$f2iy*g=-Gn=3{?k)$L^g%qmz
zkk_UVQ(~a5nv!)8J-1g|DkT_)BhoI=&62{797x_loMCm_kVT8FmzdipxZPXSIABam
z&XH*>I(buB7V8__<_7ophCGZkSGVZ8fegw`_z;<bXSv!^&q@Rl1K<Ag7X&w9w7^Zp
zV<1e9X&kXlU2L9FSY=U4(AG6pCTf*1+0nEu-g~0&@!k=W!-okmO^gq(iTihKHVY;<
zF+NPxO@+`EwQ2|{z|@gY2n->4k{@xt=z^2T48wpw9I(a|Yg1LA>Kq(O&n|YM>>$1O
zn7S<z#4PB#j^pt_yQmpm;_Wxzuzmg!2m`9H<A0Nn{@q_-<5Q%rNkJf*x=e;K5Knug
zmWZ|@hlZ@$0#nb%VcrL(ZeVws*zbFS_dL9L$2Y(H%Q^FYF8G&X_#QHTp8{B>Pm#}B
zd~qpA#V&Wg>CR1r)CzG<7fc`S&LujZ;pp<aQb3kx!<~6maDGk9@VzLFrJP?=38$G0
zVUENsF7UpOWdhc-_dww-05jn(<zG*t6H=hEx{wabdq49W^1R2-%&9;#I#Eb^awf@L
z%#xVng_+k1ye9(Y88ta)L7YjFMQWEW&m-sc!LM-|W+Gn<i4Zf|XoQq=21A+|{$;fw
zMM5)2{l@pu|NL&5PgU`8D1Sr<aaIUqTrA{FK}Klxo`6`!{<(?ggFvYAn$MrtSqYNo
z^H@-eija%kPgBsD7>YzNriq9|O_Gz_v46NntC}1qQi@cIE0T;5MshCx0B)R!aYUIG
zqtDHUBIP;1MpsC!31c@~VV4N!Delpt0%buNj5fp=@Z*W-9gFo7lwKf3MUX{OnB)7#
z^WSv*ux3C??vEpReT`Is)ly**gmK{YFaL^WyDcUZDY9%9SQ0uohQ~Xy51fV$9(Tl$
zXx3}$<pyIc&P|xA#mB^Ay@6rEELxH%u3uV}9c*w8q@<Y{tfd2lVTzCiLMCk6(B9ne
z)i1w+eUG`mCMiR`P}Cx$qeI9{Ko{xV*q6+JdWqDADmY~JBx$KPH+1_QIVFbUj_<lX
z*H50(G!;W{xP7vrSuLm^5|4i1`nIKFAWa8`Qz7A`7&zR$Wm#Et3RFT9QbzjB<KrXk
z_KMg|y!-kavPfKi_8CT$w5||6It<9*>F(bmwL!#$n4P@7`SKec?;okGMHx+VIFS=9
zZg2U#ZE1yO?0a&5Lg@-+6syIWI)&n>xV(nosB!`YDhGo1NUaziA8}<4Vxmk2jjTD`
zzhf93+tm_9LMcl~Fh&r|6}7Gy?(P^*JL2e&wx+IXM26rehR1h=VPyIIQzlarb&atN
zQYpeXfgf2lmPP%vgj2bZMN?zz3iat{I5%-T9jK}rYbyq8SwDTsa=B!?T##a-S}tH3
z2}4Q3GgVbgU4$e_SynR|)M|t8CQg6#C%Dr+ZCkOu-Y|rSpIo`g+7^UlJnYd@7AH8n
zAm&7vd?6~5Avkt-cMPW!i$%ls<`!u!t~*k-OQcYYV^35S!Zhf5S$GIalg5$Na!HaB
zq2T((C+L(2E@4}XSX3DG{ftBYx*n%4kmEZRizUnTHLKMEWhD7}T`X7q#59g%B2q;v
zp-O6|Gz>A}05=AXy8~_<`SLG+h99R{CtfDfc;5MzKcU?C&(Xc_rD*WJQIJwl88wsJ
zX8pM?H{KirDb5gUE|cXsK&VX4Y4-3h4MuSd{8!xGX12PJ@4fR2%f0xa0GGie=c2Vg
z`yZTbje*(hMxkbyJbhRo^M_A2D**CLXq^8i=K^>}6T)o1lSn9g&QM4*A-PQSF(q{I
z&E>p8IZw}pbeKQjC%oV1Un(i{<eoB0o0$X&dIk>iyt<H5A(Ff_J2{soAzt_w=d{6D
zIgudb;w_NRg`E;Ab&g67DWT^k;DRQ|@;*Z-Mn;8@Sdf>qN`iAEFt4f3{EQp}LROe+
zjS!09dt!2Ey}%8RbHeI@X&MMqK%15@>@l{&n3^e!#WH(ta-uINqm+SwLTZcD3O^-m
z(-c1f!JK2D(6%Oq31w=evgN&}M2L>$9m$uq!!#Z-^BPJBiPaUh-F&ZW^anKquiw1o
z$@AyLabmqRtlFC0{Ub>gb~MqI6D~5|JC273YHjf2G~3w@<eXWoH~0)W3*Nqb2|4ie
z<4>uVD_qR9n;UFZ5&TpjGMAX9VzrokB#sf3r7$iFOLvL^lH~5s{sJW<SJx}{#{qHL
zBUTk?S)_l%#A36?51!%B5tJq^pJJrI^*iF&qqQcc0v-7jSl`}KRW;~_ZvTi#5sgPo
z2ZqBPQAln-{fxzv=ZKK#PkWC09lEYj0*HZbj11l}1y8>F1}PM|s$q3Q)egK+pVA#3
zdHc)1WPNkZ^>#(dnR>bAaeCk{{^kEbQVOXO>t%~3ayT66oo9%Nh6Wh~*PndEqFM0O
zpZ|h~%@S)Z>qSkyT+XnULZZoj&-CyLttD1z{2ZBVmn|U%PW=gC3hbqgfe?_|a@utS
zp*ec6WGXdrxWDIw#h8}ecq&Fn)}WNdXidlx)vnQXg?1zT;~o1~Um+0GRW-AmGqtrq
z08t}m<wBT7v@BB1_4O4}3O3!E?s%XdMmA4gu(*ECdbP&y-y(s<cFWLr9J0XJ1|bzO
zI^5J_wMOJfr8PlT)a@;Ow?nPg1d$2jgpr17y}+0X+b(f#B>0hbxu$mm$xlQf5lUlK
z*-tvJkgFT&)idhtHL7jVIGV*(v9peWae5#ILA~6Pf&REh=LwY~#?~ws&Ac~EAPhpN
zVm2ZmH{Xv0^w;%RzxXkY5k(fey+yUF68u9!8*8BQz8KU9NpNKsn!IP|M|Q`KyM4#0
z9~j4;<8B9Ykt-~Hg%6?l-~&?YUt_|bYt)4yKd%&&5GZ+`jOSf;e*ZH$&UphD=v(l<
z{E2y|p3a+lgp}x+sh?)KUV-RKdm<<>wm(Dv%nia^YvbG)q?C)sjzT=l3XD>h%#IY!
ztwed<d4}&l7@+5Wy<aVy&qJPxk})QXE{#eFAevvZV0Hl%m)Do)EQ7LYi0^mnA0QOW
z=QoA;{yd*oE$7F{XSrV~zQtZy&2axYJm@{K(pUo&zDP=E?nP-7h#-8pyw;E+Vm6I3
zGfHr72rs5X^B3XV9G=%3`TTD4`F)*H2U!N5=rGk5TdfIxL}WvZj^G11roz}46~PZE
zX$ogS6vQKjsWASfhAc}HVRcy_DLwzA%6bNYRuyIORHB@jZK8vB<^1L}Tf$2amI%Z+
zLN0Yn#(;ChUOfbd%pa&E{td?uY6jZX0yiDe6P#S;bQeG<rV!DxV%i@GDUf2xVL#lz
zqpB-1nVZ{d7V9h8^#&mYE+h_z6H*$sn``RTmMjHU>yn@qdX!NlH_{!)q8Ha1MdoyR
zAZ3ZH7F6v5Aths!L|c<r8{GbmW^=`9^i(Nv_tn>!+E7&uT3oRZmeX;^-NTNwShLz(
zle;6@?O3l@sH&xVyr)0DX1!jqUT=%*of|oh1GP0pV=pyw`+`PUY*S&ZLCQ#vfShRR
zmT@>0`J=JKl-MqptgcpUpFN?Q65o9FnyEWb)v(;IIXxUu#v+79g+$w|(VwkID8Bpd
zC0a?2UCF%8TJgz`KWDYQBBxY{CNVMXA2C|8-du70<S8*le19N1Uv`5?qECd=Bid+K
zZn2A1$&S}GB9E+}KEwM7kp#L~5r&?_{>bZhJAyKV@PP0=A3xcWJjhz&h2YueKSCPK
zG&rQp$mj`Da(i=)t`>w0t*v?Z_G?@SB^;?P>mr#VS+?ckFd9;tqfLv^RPx2gysj#&
zzGAiTOnxHw51iZyx4**=J^MEecDX80WJzMjh~jSMgu*n-l5g$;A!cf0i6IlF4xux)
z(r6Wlu49;bvR0V7LDmhym$==SDTs_NyFFF4+<fwkRBy4%Ey4}N)7xUV3S?OxLsu(u
z-6Eu<#p6!{2*D5ni*`FZGnRlFHM6^QRS~-1Y-!<ov!$}C=%H;jd%KsLToz)sX`cy%
zWRN9AFT_mBl4C#d*qxACuvxD#Mi+*BE@ZRfC|5{fF@%}0HZ#P<3^~vLUs1?pN*1}>
zd8e;+F?Pu@!;A`~7%wtE;Ic30yyF+=;`JVmKku96T+pOYm&jfpB0iSM73SiZ&PW6A
z8-X+a&8!%pF!YnpC?!fNC<W?H;%v+_ANjnK=Yt77U)=dbI+G;lwM2;XgwH%&m9U<a
z5-As@MR_TL7$P%6VV<VTxsXyJlqRLQNtsawa<L;%Iirj!&D#fi`7<ISQbOsHcT!$=
zv7v{Io)I&b3&#O5<3ECO9%sU1ilz8kt!5vALYOS%^s`cf_X>h}g>>eSlyi2rlP>S?
z)$2Qc_0=mr`*@2$VYEf5ObQC6A*DT4vjkNmb<J9;(x@p_5(blJ>h3@qWV0%heJaCA
zIR~>mGx#}lDnXivZbB)Go5m7i*A_COA5S<x6bC~s5z-)aMHZTRafLB$;Wp(mfTx^F
zD(?3<ei(sJWQK0U-rUkFL;voM><%a)u}h7TB^By$d_=Y@md`(?-#-%j0bRG;eD)KB
zknCQ6$<!YZe!^Ho*B^+Fnr3@T(^QP(j%hrySZ)}OJ0>@wOoejATwxkI{8Y@1uAe?<
znkL-g0XGhqx?yqkgvGPZS-luaAL$0P3>dAMx`AK)#h){DN2ZvG-4VJ2>$;*@v`AZ_
zq@XdHq!ei!I6dC8Xi9O}K7CG7il$xCpH5WOg5$#jwHpw-2TWyYY=vb>nnwIIky2)y
zCYI|BN`TwFMQ<yfEjN67dPm=#ka|O7YSvGlAaX2O0SV;sk-Bbqc6-fsQO!Qrk-jg?
z_?w$2EU&K^yAfR(tks<MJBHnk!|SgJ{Q=V~QKmslLyE2hK532JUZJZ7Ax#NxiH=p(
zVyX?qo-j<*s}**0i%@&^`|mhC-j)5Ru{fpa4uPl3mTitiKQi?l%gq*R4YFRM1t>}&
z&GGfuOuIud-`Q+1{gIjp<tKuVCA(o=G7GXBiEbpPj6#&;E$<;rxKUuL8j&K3a#Od~
zps#KT3x`%3VQNH9#IY1xH=S_P$m;qj#uz~Gu-}(8%HfVMoIoU=e)Jq^QVA|nk|_j?
zULsA!6cS3;*rvqB3Nq3fvMQ1xm1DWck9QpJUlu3CFaeTe8giHz?sin$o@@+b_sG~C
zN;W~gAjC*OVXVc^Nsnn9sjHf9_<ok6e_f9v+mlFD7Zl0`ge-}kDFlqsNSTS=6Jp_1
zgeX87hG{5HrMZw-PrDt5{o|YpC}1|8IYaOl06m}A0r|pk7o1t;#kEaR_V>~hWx&~j
zIL&)jp%hZ&OLAVR(euPS7Z9oBWy+Vcan-p2Nh#rHSHF-F&KDCRpXPl%T#6JF`8=0B
z{W(gw5LK0wB&9H!V&pcbC5j8?UXIMKHxp;VA-qTYg%pT6e%$AXR^{(YMDBd<&JDv{
z^v@<tIh~`2i$pi)Ojc)pKzV|kizDoXw=pX%&LmCm{Y=g*uz!pJt@UM!KS%Ulq<$1N
zM<EHGvmo-t!ukAOrI?#&Auvi^5TWyDmQqRiJD-dk6G>(SvhW~2c&wZf<2Z8KAJCuP
zmgkI~F#DjS=*iL3)Eg2CsSQ~c-x6gPC~a`lv1Ag|TXeM`lMqQ<6cU9L!K@CjRJvwz
zUEwojQwX+UB*hVB3OP~B3MDn}cn8{|>kYQvqN_DA`SKi6kcFbEm-zhsa!r0vGf>qv
z+v{tViwzjdus_g0-XTQBQ=E`ix6iqK`Uy!u7zR|Oxmw@ix;@>yR|Fq<eEX99sb|?P
zkSVd=Zt&B9>pNly4BehE^`sC;!J%w}G&Qb25uC%+3##P?W2-VvsDvaT!-N(K)=xiW
z@%(eHjV+-*At2L88at-bj<Me%a9r07+B8(lC4QV{=QE4Uf~sDUm7(9i<MI9ttc!_*
zml~%mF-!9C$fi}WHmEGgQ;BB{u6!wde?%+A>dA8wnHU{?-{TK^PCq-b-fU<LG*6zO
zmurNqu$ATh+pkbSQi9|sk|>&IYb)xxkge9Kgj!IbRIMmxNkX8tVSW7sYbzq4SJ&9h
zbHx5T`nT`MF)~I_S#VV=l-5*0oB~l=^wkS;{1(fK=K3k3-4Ija>Uzshe(O`nff3Ei
zW0Atui<Z_VRP+$b?pH5w$iXq3_GpnAWhM+AeRts0b?9ozsxgGA$Tkn%M4Tp?#=>+e
zc>_X|g=8E@f^(?(nic{?^r)r<=KbQ?v8%06+M;Yj7!UZ-Ba@`AEmlh^qcIZ3zJu{Z
z-LwpYz?wC3v0++Y6OF}~5oHb6RYgLgj6t<))>T_}e=%NcU1JJ_{f^M@$<gDJC>Etc
zpmkj!O{sCd2ay^2BQB2Qeoqp`Su>_Y@B{tisYAm!j)f=Vi#z20(EpH2^{?IGLrD&d
zDG{Ax@(!UCDJ0As8%r+oGHFbS5DuBX9~p*$<MF`I4Il-68ZR-o=>vzk3yD9^#cF;`
zS)%KTG;l7~F&0=F2~BIzM3M_6@7@<UW?zAMuYD;J^Lr8(lc4wOhx2q@_WlANO94%@
z@4gSIyw3NV_RBf?0Drhl$7v24vNPdMsq$0K7u1BAy8s{9=$;?o+yu>k_aRTh=yL1#
zvy?F9;@Wqfj?XAZob~oIW3R+DQ`|x?JxD4_pNmg}m<f;bJBl$CUx3L|w-%|+O~BdG
zxTGnb;d&`bvmMVQ!1?plN)loAKsdkW@^+;FG4lWl2{T=O=RHa^^G4>s6Xymos~N=*
zh`AZNI4(*(-%~9)ID#9BeS9j)fMQ}5N}hqtC~fi6cv+DMsWJ5$Vd@h1oJs&(mI|#*
zv04`r@5ka@aF($ugFq5oBBzPd<2|-slA=e|w%8XBM|875h&3sDLg>o(#d~6Kn0iHu
ziD~+N_l_Ub3@nUhy{L)DJ))_ptiiR5VoTXH1(wjIxSV!-LO)PjMcpV&@Z5j>mvqT+
z^YLdaK5iHu-cq$Kqo25c^NM5tC90`0wPCZ}&@Rl}Rcn;4k!C`KiKbnaT~kU#H&N9M
zVY4B*k!ra`waa3&JM|<{n6o)}k{gLr&)5xETd~}1S!}Oau9uwd?g(x|Ny!u>Y(K`T
z0*F<)z{!tq-!X(hsut8&H<%n4`hgS#esm;1(rlJRx)K8Q>KauWm?oy7W9Uz0sS(YF
z&X2?{vR>bUQV4AcZbBM~v6jkeq_LdIGm#Qt;e24xE|4;j^|EN=$AQ{dhErGcrm|?6
z?dpn%V6j{vD@CrC2wkCsM>Gqhg#Nh0A`tC0cKIBZH;5<@*&^al%x<JE1eWM|x+%_s
zS=U@|95E-V#k!C=_ixeBVTl+cK?zR2BjrHVEU?Q3o9kPqG1HF|c4-N6!V5*cG|0t@
z$vb3DRPB~z8juQAPskW9VJc}n6(+7!C}D`MD>1(aw2@>M<@V@C>e^x!E7ItQ{Sjis
z4L#9~TyGkp5Y$heBC8f}8k{f?CS(YRwxL?ANt7up35~K1Dk`EIFvgOVK??~nGd&Kd
zc^9m;y^yT><AL4dTk_Pgc=8<EG$<D!w<O)-l0c3<cDcpWHRI4Tb|+j22%|Ceg7#|h
zLoU_7c1Lsrxi%=NW`{OFf;NhF(NZ;qL@Fp8fSfYZ<dAa5o40qQl&Gwt>j$RM&G3C8
zyXl#YE#>>YJTu&Ev_AXRT*zZFM#za;Nou1oGDGwv?}(Exu605nwJZcRQQ-JIqYvlD
zJ|E|OcrnE*_P?2ukmo{}-&YYT#PsV;zy;1fPxBw_{e_$}1wx>%>}C2FLZFqvhclFa
zF7l;ek-VSc&#y4Wvx}iDSpsr4a$@%IpQVpy?tu_zHOG0cUL<~q`8=e2DYDFriDFVT
z7v6cIPjfP4=@}By1yrrj`m7w7ITh!}nX9{yq9A2H7MbORsL7m)IA;=2jE%?{Ek#Kc
zRkGMT&%BF^2La4v()0UEXTSo%d#A_p!pgZUHq?%3?B>uag>A2h!4>^}iU=W#1wDzB
z6<HdPl4%+c1cWgl@{B%$Fcnf+q%t7STo+FW6ETX?ln8lQ1Aswk3sRC}BE*TxG_c%&
zF2q7Nov_i6DXWJVCvvE$wMGBhxAp(`@q?Oy{_X*W9`7ca?G_&)N<)bn4d~jUjb``l
z*W`A==J|6}9LZ@!uGXwK8itW-vF3F5fVP(Q>8HguxT<;kM}LIxd)jr&?UP%CDB0x2
zux2P<;F#!72P##%)0kk=hQr}dVmjT3e0YmH?J=aX`wM|M4QN{eh*paQNo1PEn*QOQ
z{i|>B(}=Aa#>0rY`IM&J5FcNWh6%J}xn8q)@{H5z!09w#Y(;c6-La=$U6GAs5a1q9
zY+|Can(ojO_jA`uqB0gAQ;EBjvhWCUrZxs`EK~G^X<)P6fKX*$o-@HaG@|I#bt#J8
z2l``&wG~#C{b^+lD21CmNeFIVykO`%wCh>cmQ-uvW=o6{b!A9GGmILSz%H-J+h@q8
zEl^jxpx?ixfB6loCy`<Iz~R*^@_1zR^eJ|kXqCijOVScAHQu$wJ@0Uj7LoPM3pUT6
z;ru`pRY7_tk7`!5n>9j2mdlFn)G?k;2&M3oBPoqoTqEiQ8j0(U_~Vi3aKO0%ojuFi
zz%Zaxk<~~_B(5Jl&Lt{4f$s}BN=;1tKo+wEDkh{<gy0Z4(@KGJJ^kHV+&EI3vIEY|
zCnUC*c8wCU5X~sDt;Sel3>O`^s%xtCHRFB{WODMPI6)R@nc1uscw2&x+~YeMEm?1`
zNv3A>W!h4@!Uu=26|HTErxQ{s)~hSNpE&5R?;({zYYVbKxAJVqD>BX%cgP~|QYa+G
zN|N)W8S%N^Y^a-wrZF6Lj}+@%h14II+=K)&Ul4OS?~V0rv=c)KHmVd<#!?xDwF*)o
z1%Hv|ohN)XqYF7ENX2X?#dMMAAw+TEgNsN0d7{m^gy!h1@H1o~)me@Qe8BeqplbhX
z@PPM`YUW3riEmI6`;;t%yR+UuiDIgBE}9pv!~DBSDU4Q^sXSk}3B@E!$ifMb$V-9F
zaCz_fYzr(US>)M=<V;P>@3{&EE+{H6kxGWhjFiNX@RLJZ1ER<=<&4^-Sx<Uy9JG{3
z11Xh)c@8CtbK`YhM<6nnpk+~3KWgR)T=)z#5|HN=Ns(-(_pJW2%4GIn5OdRzh2+2g
zmw(Lf{_anZN~4S=D-A-TQpCHi5cS*%p%)+o%Cuxv;lm@EOw1sHBNIp=k>W(tY={|x
z@5wn5LLlc@GIa!a-xFPFC^8Y-Top@m1zJ^TDG==y%2p-sKnPME2=NFJlv$CZBda-}
z?|U3Ss2SLwj!ePPR5b)RynW<wbOfMnG}liys4%hGE_nT`uSsv-@Z!f`kcA`)2_Z4Q
z{*LMOx2(3etUmuScCo;@k?qrGeEP-b4Be4Ke`L42BZP^zT~HgvI868`NIr0Q+|x7-
zs;zM`5QYICz)gYv)k}{32`wedwq^hDfFC+e`yIDWp76<!enhiq@VQ8P4#UVa^u+{Z
z$^@lo^c{V7g8l(ZD3OxM(dY)Xs7OPH7M|Ez`gc1H!$>eS_4O5}VPfA;RBmFN98xJ7
zYcWdU+^iLs#aM@!h+|Ju<)&cOOgvJiOc>r@q)KNJObPYK*xGVBojCSKmWzf!#9CXL
ziqnbFIoiq)gU2X=ngmK|Y-P}<B?pg>8EdPedyfuDM!K?cNT++;{(&iiSw7)Jg}8cx
z=rqx^C}m;TVGuaqVMIkdomkZki$;=SN9^9AswE0V@EJ^tKi+Y=`wqKZQZFkafn{Y7
z&XR;AMvZBgXxo%bZfnr(ii3E?*MIeAJih#fFMjkHqPZ!V2?~_0ip??#be_XKBIRle
z_|PG>!BjPpj2}8u@VKEXgza36ScL0P!6VR2*%N|CNR3d%f58nCevZXX+LRk{C{yNq
zYotFuAag_;#pEZXG+1L;uGbg&r1uVyDY7jx+NuIu4+ks)sv4myoDanVKWC=#H=8rC
zxp~5JxggF6Ptn%{vpufJ;3iKDQ!yNJnPLCH!`%Z9Zy&io3}lKo{ne(yn)hmff?yX&
z+$f1P`VvATm?dQ?LROcsA+5^fMUF+96+(fJ&yYI#QvAx0U1V?Y-cR7XuNSgt*^L$`
zAz0e7u2XaIDZUwrDMZFx(9SbLAkJ?9ehn3vpY}4T$9XMvjsu3GqDnKAe=+cxWr3G{
z30!1_=f@FBT@(aTiVGPt%&bJIio!soT*${!VB{GgnHz%n=ic)&#9U;>Y)xF?{YXa6
zpVe71nP(_lX;aQ=hLj5F^TPR<*ECY9g7~C_i$(LF=WF8JLgjRR&E<XPT<oOP{Q80q
zGZOLMrLZ73v&1?`w4NoCSzdf0&b*1A|KdxIhXbo+OZFZ`$xcWqqm;(Ej+_;hT@l;?
zp);y#kXn-5gqZmlQ<%s(5dFZ>IfS-FeWDdP2|{ovp@Bq76UWmdT|ZJaD{h}WEgYNE
zNOVVB2sG=br09^wqD{-x4dgU|YN?mkgvpiXp8hR35Z)gqp01zr^x0FQ@30Ha^^;rt
z6gchgkt!3%J$1EWqb0+5;^nV?K~q~yRpX9FnnlZE(NJ&JWF<L0z9G4ZF!pp0d%peZ
zTaw6_BoM)q#(~9hO^yjSITDGyY_PX4I2|4dQ)C(gQAzsV;ky%!>rqN_Kyo^DXl)sM
zqM1C$)3Fdt>IR`DLMu$&B2?ngjqLAt+$72M^XF8prjEr-=Jm@rjLx%tdQDx|xc*q8
zIhCPW*KFGbei~7&#yQ7um=IPIeBf?(<hb8sjONoXK1XPYHC18AhpAj7A)sfNF3xL-
z4+n-20x^Wy_gf>8RMsGK!i@vV)e7CNkV>;|uP~=QyWJzC#Bu5wmo3d|gU}UOmX2Kt
zKf{F+vD>2;>*AD_9No(=*}eLX>mU6N+mAovMj71O731T#_^D?)?U+sn{9+A@7PP@z
zhph~E?_P0y`vz65c=205W%cBSJf;$!Q5Q*Sa1-6`k?4lm7Sv#x24!r4<+=l)(Q1hm
zf=WbME3r?msVd7^>phJlbyF8KBa5Pgp2<ULwjI`U$WaI$LoUsq8%auFXZfc#2HQ4t
z#}lXBBetoFjc}OAt|v}IiKvdA;3jO<V3bDX2o(F#$%jHVLli?0p-7Ws|MokY#gfgl
z=foJW#unCljKrxU2RB<97dSBIKuiKFA~6KK8|DPX-`qoMgA@`U0_b8|lhcF@ikK&o
z?{L$|c<eAz63N`Zz2oj~$6+@i%o3#pi*`}C138x|R!Ov;oh#=?;A}Gs%+P*;xO2>)
z%Z<FK2r}6@qH}W?QAS8rrsSCzIM;m6lka=WVDa9uZssPO|GB{4vtf`nl4kz+QU)I*
zV-TGBkzw*5pgtdb#%IpmTxb);l`+OROX>1uAD*QuFm+0$_<q1&5X=q0jDH78(-cCu
zG)x6KdC$E#dm<p1n+-6t>H;C=H9<j9E_n^`-W8EEnKR9=znll7X4d|U)T9_Mrb6d{
zqGBB_W|<@Z%HzCOUYr#j^VSBKR~<7_akhpQqO28CN=2eMpOX~k^xKpXGJ}wFZiAko
z+)Jy$%UAFC=9`!Njh}pm(UIr}Vv0y@iGHF#J+fFn$BjLHJRwZLP+=;Ir|=MrQW$MW
zfX^9Y+QK7CZvHF^a;8LST0$rZvfJ%#$(xaulmg?lr_!2cafLP&!A+zP3AwDK5{k+;
zXkC*-Dxr1zci-<C{k|77z~_Ja@A1X&{aq@p_~rlj`;5~&77g5d{DSb&Gs4*O=Itw*
zCXkIJrwX%J6PpUVTCv!ysAOgu9fw!narp8}tTCjar$6jzmTR7U^prR6-jGD3-=8=<
zJaY3~Ae15YeaW}hmdEiGkFQ?hV&L}2KSFE6pv^45aEu}1LSga><*2O17|m(-h!7IH
zEX>|G^wid%H;b~DuMDZ%6CO2-g#lS?j2RLSZ@)utuAnN@c@zmzEy+R=(1fYOk0T*O
z#^9(H#eP<yurblr2B9D(hfy`!R(L0IA>f7rSsN*+D@)8^s|sW1X(lC<)>v!FDHR7k
z$`@Elg*H{0G^L_it!Ot}+&I#AJKX+(plgD(2%!jJ)(NNLh*s4a#*uLMH383}sabB9
zpguuYig@=OlOLG66Dh><rB4}k^C@9-i>a1nKQA;fCq~mU+65{&-hTa;_;HVExAbnp
zRyA$ilB`9sU>f_9%r@_4Rd7tlJvn7;Rngxa>F!@~|J_$C>YDnq&xy`)>bjz0nRUn+
ziPjops{;Avk_0COvrA=J1CGN8<bo8~ve%C(W3;8JTe6gh9MHxxc1M({@ZDi{1~X(z
zgs2qY`aO9(;N66C5l{K-v^Fpk(OfEw@BXw0YpL7CJkb?mSn?Bb>R2omSfx=iB2*@8
zNR_0m8xo4X|IKzFoTd>iGvnbAH|$w1ET`Q|L<q%?o-iA&l-vePO+O68II-O<8C8Y%
znce=t<Kv!`3tDy##W};==K}g`azJMPd8v^^gv{l8;T>TbiD7mflm*%sGwJR;v1fQs
zBt*Su$DavvXA<FgUBP*aE<lth^~)|iMWmM0jbzmrPJ?6DO^nW8RtJ~jTAs2L`4EXI
z&IUf^wk?I2O_v18XA)&1{t!_}XCm9VLCE4V@d7h?ab8`NW=O!Lxezmj?~<;N3Iji<
zR9qV8$CWc0adErL%y|IwT4H_#sb%>+sQ^g~5%1jP9+%>bFo<GFoHM@`cO2rJ_b?ZF
zIZILH2Pds421>JV_7ZSDVCJlaoHKE@gH}?Y%{*W!31=JNOI|_YLj2Qz{>S{@-})&j
zO!UVGLITq)AbBLJI7<2h(HAaL81_tku_~^bb-B;@iIf7iu8`8gY+M#Xq_PUFEg}o5
zx<<>2)Lx@?RpisjA(du(^(n)!$7fBP5~s%wWi8Qr7S}JxIWl%9a@vu`4kZ%0T7Iu<
z^anKq&p-Q=%2qtQdsE_Kli>AN-x6Zr>eDZv*;2^{cY0uPeT6la!~PM5p0=(~)-d%a
z4#y*t&wTZ(@5m{#tt_jyVtcbCOGUe^2{}_$hQ>;QG#u{^WH*9{bcZ9VY7qif%a+xm
zCJjA@Ffc~XG>pWYS#7VV6h!B!tRdyZ;o*_h^%aqPNnQ+N&$ZUPcyhxi4UgY_NxNLr
zJbjAj_q0`ux+-wAj{##FrZC`!1Kafyqb;Uhq4Rv{q(rs`Wh}O8+1>4EE5l~9#D}ua
z_e0M#jA&cqyyLq!k5txhvsq$n>0Fae=&B~TVm8q(7p0i{33r${9rhTbsH%$bbYSc{
zn)Q;YKT_38l(ke!Vb`0&OpOVV%5GNLs^kw$9a<TpP52ndDIiTQ*$g@`_D8%QVfOqM
zLSUbKOw+Ck<JGiCQzL9uB00wcP2J$HZi&-Gwt~mwk=jLcRZ|&F_P*S7mBN^s90MUI
zWMTW@r-9S)!13{+<Pmf|wr()n8`{-|p+A8euvJx9)+x*;R1r-A34WY_IJ=fX)h>uG
z5Jbi-TBaCr)5QL;Bd5f*$QWDU$FXo!4hNd$3XvkG{ehG-&2m*Hc-!ETN6Zqqm=e0K
zspoyJ%#oq%h*DwJ8)6C&0$EC`x<b!)?5JR(Os*k#tSt#_h+=WO(29^Up3J~+b~Df{
z7A%)5B#~kF72Px<#Kdm*n&-Da#x|Nnrdch@ibq!5KD{C5#N9*W;7|Ba5S*KvE0&8A
zR~%C$`;2!bCthn+yfgCqNZRuROD;}+qEr~78*wfW!VH(|vdWydHy<#{%anhH!pp9G
zHp@wcDKBOeKnRek<jtoT$TI_9D2=wZSn&CY;5|7*QyHqd;(piB4bx@Hy&OVZ9zVs|
zXePh6DHhC=WE69*(yy}+F7xf2bx=%yVv*p<!h<Q#S$<OKLOw1e%koOj9EdY1uw*RA
z%S)6WkpHAo<s8N0?{VpD%<N=1H&9ZRCP^8MsSMr)yz{&#QkHXBnxI^e(lbZk{Fzc@
zigJD*GZ1)@Ju<JNN;p$7ON#N#u_%A8bZIC{BPTD&&N;GBin5CN*<XImyN5?MZG~+X
z)TU(|J50T%t=5QCxG%D5F}fwV0jUg1fRP47f;gd-Kx9QqWre78i<ExmfRvh1&dY3|
zq!pD~l5=1h21YkA^aE}@VC@Pa3!18JOK^!@Kk@CWBTqJgkDn~j#*)MLV;udUX5g29
z`lo#CBDSf}b<O5_g?@P8t6zPKe)a{Q{ZD?%`gmgZ<)7hB2d+N)2-Pfk{Pq{T`RZG$
zTJ!vqPcR<tA6~QCZrDD5N~Hx?%N3BBP96O;aeIBkVza<MyyM~JOZwiidisQBv&L?(
zsMa^6{>0(iuXy>#f57ea71!4*PD9J-;eZzr)o4VNXj_W(;nbm0=GkfuT5&v{*u8s)
ziy34^GkJ!-C%K7!I8luotkR6mae8<l2gjmr&?(bT16mY|=F`Jl!rcRHyTH~B)9FBE
zYPKJL3`y|S&woXT6IwyDZmHW1ZYo*U##Z#>MDims2aMg~CWjj*rfH&%5iN_S_t2e+
zn_ezolErcbDpN07s=6gi9%b~L`q$y7Q-M&WLP%XmIyuczq#0ovf)6O8XsjZ5&tbnO
zNlVo>n50lLLn`*ihr<C~FQ}RoLR;K8lAR}|2eKQ<!wDq{xvnxE+0?8ztI}mBL2{+g
zGEL+#k(8!ht_tiKJ!vMDRm~D%Vu_com$aK3QV6(lAXJ*{2f{EQLWa6xnml!_X>v!L
z4h8zJ6d|c%gWMlVwt_V17|^nCDUQb@{o|g|dwfbP*IT+%&v+WCtits@<8IHx!=BLv
zs<y&bn$2p3RtoO}ZgMn>mbKAEUX&qDBYr%QlEAwODZ$n?wlc)=RN!n|6*C!YP&D{y
z!dEpaL-rGLCUPk$k=A0S?`ItH*Y!Ak{rmXE7SnF1OeRhp%2aHxKH}kY&(rp|P*#x<
zL^pCgyd^kSTq*N`N+@hwI4@GxTwPz$ee*3s2qN#pYsz#POW~Sf<@2Ab9FVCb!$lu)
zqsRMH?7<Vn6j3r`j6$kXToEAi92OLd!BLKBp1xz@^ax2wk}E%7H43dLW-l=jQ{;5)
zu~lHXXwUdRDI{W!*rw%ryCj5I41hjZ50Z<mdGO&qn?6R=+~^>|2uaGC>;pNy&qk0R
zs2=2*;hz%TC+tjOmUB^0IbQbZF`Yx7&Sbh-dML|UHO-BJloF|Do8!wf&6vo%$~dna
z&Spdx)P<RUFgI9sCZBunXI@7k`6^Ychg0?!p2@k{IivZdv3alEFCIegrIu$Cp|WBq
zO-nI2nn{K8EB}x*Ddx2WQX*31&AUhbtN+`7%m40w`@h6mJvU@UsV0<0N`)a1<3x<E
zAOiIYnSA-KYgG~k1Jq4R2otDSg1LG}$OE(sY-MK>goZp6-jb+EIia;i>l$ln-o1N6
zy{Or2Zb?F*t0m*u;W%-11-DPGXtyhD-6Hn8?{$s-pl0AN{^U<EbxpImV)^msgy4z2
zCs#XkisZMi2zT!o-h9Vuy+n+A+Uz*^$gh6!HDCPp=Ll^u?TRn{<`aBQ7%7Nj2e#q#
z?k)ZO16QAZOx-N#?;eQ9Ba8Kd+t0pW{o*6CRP5fpCg;dvyTOPrIQ1QGAKs#kWApSG
zwW&$AMJ<-3ZjT#JgvqhF*+NQ8A(352bR$h$bA9uIdbuL@JG4>6P@v-J;WfEw(5o#m
z1cdkOcYD6v?YVjSjO&j-CASMSwiww|Esy&>(kxhAZCI_>45t$}&u-`rdxj8M*DHi9
zx@s+f91&X5w)T>t?wmsis22@_jGG+OG|`<-EY@3^?bSTR4y8D4w}7PQRDO5<Vk>Ru
zjO=2>;j<*5krA>RaMBn5{;089GK~|?PpJM_Br!RYLNQ_)`W`<HNNbpej@+L(J-)-c
z5$_$|PpAxw<(g)>rf%C}=;AZRYO)Ya&Jl#gbrTFbY`a9)O_?Y}Sq163=<b(F;hm!!
zMv`}^;L*vE`h*OTm^{K3_sRR$ueiCoqFG9^6nN)KQ<O4sn%KX4Lv#*n>%zj7f{3Jh
z+|lokRNF1h=9cTvenK;N4)0#!j}IIl-;sj{ty!&BELMxM4;6xBYufDs;~Y$5**OYH
zcAjbI(Ncg?n8qSi(J|}fkrKR{_n))OP0A9|61~IE#w#&7K%lHeCN*;teuEELTnBR)
zi#~rUvz@j}7OF-li<E{qjfAnIl8N4TBqw=#Q;|i(pZxMIt3`v-hTr|2pYrOv?=FL<
zgyKEGS#w@8-3v!iBgv=_$bxZ}qe&@IxoH1g$Ur2de39QVqkS=z2;LO)rKn_>t$k;c
zoq6}oRAQQ$i&F{-?*lO<YNN~0sS2W%X1Kbp4C_V3?l=;E?XxaWbx}E_IYL<?<$Lr%
zn>pS#C88(}ikG#-h3$XlAG}vtC{;*=MR{<}RtWPsQ}0oLGum)IA42d!@FSNb#fy@F
zi^&l)OfQ9ixgofu9SSbYeMFh;tICk=JkW&Mxp0;d#`J!@A%sAkWrBzg`lnLN)2vcB
zH&5b>M!<!%CphyJ-jgp2d2r@`q;P2}VvPJh{_#JDocJ&P^Z$tGc39gGeWq?L0z<}-
z(+JsNOidC3p)x6WWEN;;0EeG?lqr(IczhtafT|jlfZ@2KZd#P8NV#|-*h-_3B$<$=
zCV9ovr%wqn5tB!m3XuawCN`^z?RG<3)dh``>SCbv|L^!g&A`bG{Lvr&5r6yN`8}>*
zJmK!`J09+KwCgqBeEAFB{^g%xM8tI+DLz5BEATxHj&*DB<B8#TKsH;}*H0MwJ^tM*
z^fUsJ!Fj&>_9d%lFAz^wjK>GE(k$BrVHz3l-_m!d5*kz2jCBjju)NxGdVJvcuw%Je
zW2+jc8|vi-RY|(X$kBUJaMVqU_YR{($uqYa{74!lW8Wc#=CD8V-PbR9ese=}y+sCi
zd^nPV<8T<+A4aP6gn#>nkAL)Aj6Tz!j+m83)Jt~ViN{}l!~M(GNCH|5p1gR0X%?(E
zSM<9#<P?!w^Z4!^LR(r}5uE4o&0A88Y;Ug60&;K{vK%uJM$S-Qa6MD!7!P~M#R58n
zcwti`T>0V~%@`A-PgL!u$YTTyhezTxAhhI^OA}Hr)|jd$qX^NJhR-?r!vUWIR_em0
z4i4|8Vz41f#<x}iQ`>SQ%oL*@k&z-3r6fs%lMQK6p@^8KB_l|)f<haEOd0Pc;@A<!
zfi(G&6qpiLYEpz;udwR{GA2x#_~}nRVYAr)k&p{Z*iQp7INUT65MUdQT}N^gQb|m^
zz_c|@OtkA2Po6$uyI!(fN>**f<10<}6G~*Nc2P9HN-<3%QYcUwVVBq{gF2PhZ#CmE
zQrViU1lHCleNG(Ah~U6hh1*~%#!#eTF-E3gz(kKSmbz(3F%rBZkCE&=Vf+nm23(JG
zC&-B@1aeYTQeo?Mwudbq0KxSr1t|%dwx+2UWISEh^LIb5Y1^7^@_h09NA&yO;rIXG
zFJ`zrlX67c(tM>9$+HGr<WfBIT#Rx$OVY~oDy@)nF^D1JVj=NrB}>@Nysyq_w$@E&
z;~)wrK+c&0v%C-)(IFN-%f%_L*Z_y%v6U@x-7!;ohJq9<+KRz>`Z2!W;|noIB+syS
z*<X|l22t#nl`4{1t>hVW;JhqQGg<G<dpK|Z%lk?AS!-W763m5AT#9wcXh;-wr7CE^
ztW+p*zp|`d&b5C=1j<W)Kgkz}WsdF5XGO%>|Np%;@R^Gtgut4Dpw(WZd&g-E1;vF+
zAy-OX5P}caBNvP3*@waCv$w%}L!rXGIQvJ;+>D%yv{KA(BZgu;G#h07H~-{+0HOHL
z{?oq$T4PO(^B!`*s7MyFSjC62G`QAcZHvsAlm+9k2fd<dH~8pKR-;!dv{}u~>QIm;
zk<n&>F&-f@J|vu*sI>)=$jM=qLCK7ki8$>^F;LkCRTZw3(1sj6@3D`6^YOz%(uHJq
zIP&uAuaMDk8heK6#7Ce1h;bNs_~jdnjI0(l?b0IRS>G*KJ-?+po%r@If62|W=h)y`
zY*vKLlKsm)<HJ2uHzAegn?L^<T4t=#07^l%z6@G&zkeW2o_>GNG&zzKtgo*~SrSht
zTC2EPEvU%Y9LfEOufF?|Pk;NjSX8R`8LU>A>{wO?FC^_^ff9n?3ao=Klil%fq}trz
zf}l$oVJhDJ)i)epzN2j|$J3Fjw(O3NG_Gf}Y#?STr3ii`JJ8N?IG%X-?E}xB-_R^u
zn#Go9zx}r{N)x7@n2V`~F^alvs2f{c_LO1r4y^=XIL%OB$rleXaXK7IjIS}I7#T;;
zZa879ie|gUv^7YHF%`kNncOxaWYO@4sTeNVN@GQ&j}x7Xpbbh|5`y3yDU@j+RXTF2
z1&d{i6fhj_iOvzdFPS<yV{5~*ZBYo~FrexhBL#jK2%?z5Bv;%NtG1;wH`J9u<WwY)
zbxraUA^O55&ZU#LTH<HuviAW}qOI4gmTSg`m!x5bxmuHrMu-|IHNiWiGzc>lKY>(k
zo*^b;7G$l^%O&+<f!&NqGU4<{A0}+GK)66-1@fa$uv!tO4%eN4OkGuFk1rJM_O{%f
zlShlpG)xF3sa7kbR;0<7f<&Iz3XbV?#M=6@f3?=&ra0^6H8};wp}(-ZW0;6Z;4nYv
zQvK_8q?n6)UWTSw5{xfJG8UCg(iH1!b4!1EV7py2`W`J7?C)P9YKa|BESD=b*Vp*y
z_y_;VzsLRko_-j)yWf@1T%N;&^1T0@C)iY)fs{%-Z<&T?a@}lmt0bt|B~b|Gc-nF!
zJ);3J6#`o^(ve72iin@5*MNj1sSuP@-?D$!v&LU3jZk);$j>Q#2}F_iwUuF*{&jxH
zrO?e)y?8W8%tgvK+x6y*lI6M89N(K~aR9kk@t#ecT=Yc(snjg@%(K<-x#>8g2_F~(
z&C<m)67Xx$(KCtg{A$yg!1sZ&B8&6+Oz(?-exLMsf$4dVE@&axw1(Cyx+&1RfIo*F
z&F7;$Z6PXNf~e-_KKoGQoQh&VXhMu+Q4}M@`SpCTzDS}pI#QQ4=^0hXDe{m0@xLJW
zf&bzk{4Yq_(>4oSH!-=9x>*vYQ<?r55lS+S2Oy)3p=z#)+Rh3ONxgc3tXfiX__4z_
zE2K0?Tj5T3jNJn{2da8YU9ZR?A~INA{1#%2cy}ackJ2?cM!cIC#vZ%al4GDd{##HJ
zv{v!-=`}tF{^I99XSrP9qvzermo#<5;`WL>l>BAF#Odv8?%(ZLUM;!#=mm>oPyhIU
zjy?M~U$R;L4Z35;m*2i2X-O7}=U;qG9FC;p9ip+cm1W=exXJPM<pYnW6Pv3oF~8;J
zdc)ParK&X4mdSgR0JUtmx>~bn40Y3>7uVRj#y!49<VaI%CRd^%LrN?wO;c+g-AIaw
z+vhL1efkW%V_d8-qvg|&e?kg@>hK0mk7N?tt1H^Y0$~)&E-{-M`s0zWUVe-BBUevu
zxqa~hqo7g}O(Kp*>g6?y>*r)YF`XVzM$`8_5y9%}ie*aly9fN(BZVL&N0JrN)JTzw
z`<NS<d}NwD!#JX>MP-Qsat!!f$X1Y<#u247O(P+C+;l`z&9GceNEvM_2t{u{23KTp
zBA2~w-O{YDP*$;j^BqHXf?V?1o8@BOC06M56-*O!NBr@G8~TzTpd?aTs=7wWimIu>
zj~56tdq>U*KMsWirOM=;qb9|%OcE)SnA>4sI38K=-lD9;R5kA&57a@UO%2h_d&-E&
z0fA((TG0<<2@7hM_?VcY#B~8eWb8V^*fUsF&PUADn_HfI@>>{N@%Zvfy2F7=3i@M@
z(S~ZhCMJb&JyX{ajl$TPlp<~@37b=2n7+msQqGLyz&H$e>zKw7sY_znG>rr|BBUX9
z140!piByUt6_^T<ezThar6oCz4E?}D8?-J)TZhLT+f9ju9lDO-I*!Lj_K)wVmn-T<
z;ZxvnIuN=ZwV9}GV(JGzdj5p}%m48ovOgX8=G!-fAo%Bh@aG)6k-NteyWOD_nw%~=
zc4l|Jl#89QP>QT1#euE_W2LiiKB6?ev$lSohs&-wF&l!-4uewBv`tAyRSF+6nc|!`
z`Qpl#Mfn{y7qK%bFoaMV;1G+6)vTz<WGF7V#Z*U%Ie{|~W5k#Okf-y0TNI<8Qmqq4
zYs~EQcXla+x!9#S*i4;~i*s`#Ij0mdb9kgG`um(`kAcfRUd-pTc>7-z2jy7`A(;1G
z1o@mgs8m@Y##oxP4~T&uk`6z#4Hh}01Z-->x>mRdy$cL37DPv!eFx?UXDE(@XT`%z
z^fYLgQG)WCD4$JCgp2$zgUrYoaYBHIAG~iN`G5Vle}L2#|K&gYPw@Q_L22~Jk(?cf
zNKO;OsYh!=k_~qG948J`wIYWBrD}psC9+=35@<Ie#6Wg}(Cv!)L}dJM1Zhg+IUNhi
zr5oB><8x#jPiPexjt^9;Tcjw1gi?~^zF)}M4{8Q}=chjcV~Nu8&0l>*oIHJZWH_8?
z+9kjJcmEFlcw%~dL-2{klP5g8ea6ZPTui8x&}l)+g5lWltDpUh$3w@%{>V>%_qXYf
zdm01Av|L}cn0mqe{zQN5IX)a1r-U&LIYmzI9<X8JcKI=`8}L!G*=~8hT_Bo<lqIW1
zbN%F+fMs-nG)?5}x&HVCVt-`!c1Lg?;RKdqUHasc7sTnvI0mM!M|Y8*{+++a&1XMG
z3dQOE4UfP0BaRQ>FpeXKhj%1EYD3#~D5+SqhQ?lTb#p_rXz?L%I_^0A;XgwrSl?cu
z<%U{<(HUJChTQ~iB#i?CfmVts3;KRQnu=@{O4SVABRK%FB$uTaiNVty52&25O?43i
zwaX=@YFVw<^Nq_RbZLNO&e(c^HklA(nS^5bdubcoaYvX&>gAStbq!ME`$I8U(T3r4
zWVKpSSC*-pSl>Qn_4FgeG_ZgBEg~iQW5+c0EEfwZ3C4nq1EO~*3C7eVmO2EIA4q;A
zAxpM78Dk8hvUu-EK9MJnI}XJlXtid2{RMe_jXK_w+`#>t?-));+U1HAJ=PeM5L8BB
ztt|vC%0%aVf)oiMlcT_>hO9NCkGN?hSB_<H*u|3V)91wTgzt~k(vXBfUp+%_pTOZA
zW6XF7x@w9qz|bQ^X7Up{M(UK%^Zrt316q?rabuGtk`PE+6*C&=K!B+|N*S;XJ`1|(
zH`{dR@$M~JhN6uQo|q%!6j7$;&^sz)5F(IMS>?qfNKq1kqgif{+7P$TsGF8)^k}Pb
zAt7?2stiB*{FbU(^SgiZGn{jDPV(-}9e@1izvQd0U-9bo9U<l7u^|Ong0v+EA(GKr
z6tb4mXerUw6qP}SIOixxrQjT|EF>`%LSV`i-D*lSRcYE%&J3fYnq|Vu+9IQiCCRf`
zh6|jJWKp#I!Fxh35w#b*e})6j;W}w9<mX*`%IPAjI~Q_E83c?ezWQgR<f7S^a$YIS
zd<WkDKJ%Kb6uoni7W10my?(z8_49uqF6x4R-BB@Jq?qEIDvSgPBF_ntAE4N|yvc&x
z%)hH<-hoJYRz(PEE2yfP#zjtp!}%~9Jux>_yl;NUvs|$_2NpY$@;*{XANV6M6Hd?m
z5wqm^oWv+*4~cno@=yNpKgSBmKm5=CBihBX+*xEI36M4-g(bxjvZw2-@>wWQqGHM&
zLTQ%ECwSKr1Etu@i4;63fvT3pu`_#O2*en0Zo>Bms<tKJ8O9?bE3|Go><;+xM3xmu
z12Iuo4Jm&=CDIRS20r^6KjC=ndAxf>q=eMav<<t59hKCG$&q5FZdTlW@&et~43lI3
z>JDAi=&GTAeB|Zpx4eFN;PWqTkWzAU)uP?N#%fw4xZZ3rTJY+duQ~J+i|rL-QuMFh
zVU3|1142ub(TtPhxIfWP6Q6zdg3YqQ+KSDmPZ_gdzd!N#_6_ak3g<nw5wk=`p>2z5
z3{72O7Y!n1_HW;npq+M!scW{Me!<g^zrd&pDJAvO=g8RM#vSiozh=-HV-!O_u-fg}
z-mJNKvgNox6v>=(#S9`wf*+9~vKTGYnpdx0b9(!hX0s*>NmDn781dd0c^iVPG;SP;
zW56^MA$Z7=s;S6nDvoDYSF?dp+3^ctv)#<Af{83V!dRqGjAKVtwM9}COZK};fi(44
zZAe&LccgpVVVZ{J%@f+~b3id3Y;pR^@5p{2x`BOnK$sR0J!1C;t0e2|Yeco8>3X_%
z-;txE(vj<B!#E5~!69`+jsa3A!7x%FO+_3#yc>&)ppfJoP_lT%Od=s0gKD?fo6nJ(
zXUKX9J}~U>(NbXQ21tcGsVy;j`e9;WEZSC#$DaP!lWc`rUXe^o$`#R6h|LoO0sr<}
z#5Cc$Bel`s#$rWWS6I^!sx5N!6sZ)W_XI$tOmuT0n|G7=3Ykl^s?Q*;Wznuss=$;&
z+mcY|93kc6h$bs^bmS1o%8*n=5C;DnP9V(VNc-XmITYQs&^6uZNY0t{_6bwpk)=gw
zh}{ldtub2TP95!LO|DzKh@cFUKcee~zCW<4YseX0Eiks>`0$pdX?eO`^YJGi^E*F&
zf$KYVyN=)gqhIi6U%ugS-*Y;i$_o>DcDa-2Sd~ALWx_QTq+m&*B7}gp6{f22<Ae{M
z7$eQ1C4^A?O}v|P7m7<^oOk$DQ)7(bbUYDL!aGlt1=SlzhqYCS=#4Or6Ht6FeDFwF
z40tldvC@Z7I1C?p7l`?uHk&nBqtVI~--Js+42WDf2Iu|wd8aS(IW{R@)>D}>nag>C
z4lxq7F2vZm8@N2qTu95gyX++9KPU*o+0{|x^j->(imfqc&feVQoK1)dNbo`PA&V?;
ze&DiJX)MMXcKd$Llqk=6Nmb0u=1Mb3S;~vjL`Z?udR}j&8Ldh0C6eNU_jLA>NTrc1
zY7oyq`fvUzGH3qx|K)#%YL}oYq=`)9p~$w|D<lFf6+?fIRg#cB)-F-DVd{4b$9MR6
zf^0EWGpi^JQfX|nVmQ7-F%U&L{}Dl!mbzNg_xDuQ6-~V%JCD(ts;+S^5M58w8q=)4
z*ERaXB8(k0&5EDB`b*sC`S|%QrmgY)h>3~)tCvV)SgvX|O+{rj|HGgDj6eFrpL6r{
zDOcN;sh{}j^?|yoFh<bSmK-&spQxGzx@sBviTlHW(-@h2<f^H;d1etgqTBLDr~3zd
ziVP8Q)e!d&RM8<rvAGm7Bc<l$-8)`=_mcJK5JXti93OYMaY8CVh=F!-Rrc53vwZOa
zad*dX>N#`+H$N(S+?)b2Wwe0RdW(Dh8K-eXi;N^utuk%hqOIp@+p=gZeK!z-qdT6M
z#-RXdDRQl^QBW+97b~J^F-B7@TAI4%us@Kc!PE-`Cxn#P%950V#To!xS>$p_jv3uH
zhz!TaJET$=DKW}WSqnmysM9#7aw*MfdrJ<U>9m6|AyVn8%XYh1M~~8Cm$#s5;xv?N
zd+L~WcL+aXv_V%5DtLnTy#4M=jtbT{AF=%6x2f6<+8jxTJC1jEqzI{2SW_XbCJRl9
zu@JPhAS6(-7zH7Un_rl3G$3deD}=I`tkKt>qT3a@KhYl^2>p&Ybr=CvWkEpQ7II?r
z645L$+b5*xse{7jhtg<B&Eyh3WE#_=+bzlUWNC=e5!{i}UB~h9q1+&)CW*{`|CZy+
z2VBgwGGK)$p6^r7I0Q^pp-fF}8&WLBHfdH36gb!uy{DROjtetgGU=AQSQ3wq*xZ3`
z@JSH0La6G8T&jQV4v(N}Hi!@yj*m2}Ym71UW6wBtC<2r7r0Kxvuw!vuGxQ^^wcOv`
zQ|ZDmuc|GTmNa!sWop8BV*l{K?USdBqsLSgQVHVJqlBbctZ3Ucw;wnB^!NTY&O^lT
zPyXq@z`2P(`;$NCZhs=C%;7W?C|i^upv`(!cKd0Ly%nmUfKrr+b}pbZ!GIWjxlhcK
zwUCO0KxsotiS@F<Pes~zdwa{KZQ1P(T-{u;TD5%j&1(+3Bb|38H=$B|`pGB!{I7mR
zcPfbCq3a9NerC(d83I4=^%b%Zap#66##yE}Pv&LCkxAm*L=+3)qLe5`STM^dg+L+-
zq9{v9n3SrRG|lRY3;7S`y*!td$|YL(+_ao!d~@N=0)#4#kU7WqXvqA#g}ZQGF_ZyL
z%qopbvyn4aE3U6r98P_)e4kMTDHTG^=R>Q@XM6t4iW$`GmypxNFe#;3W;r7WC2cUx
z#MA<joZoVaiU0dQ`hRn{zvCbLga4S%e*9xl7E^DTQVC89(?IV}Bp-@Di5V|8(tY<p
z_B~k&Vsup1l8nUEE5^rH_`XA{%s5We%Lea{fS_)l;DV!RKca4y#Bf9!NlqF!O&~Ph
z2c!=)?e=?Jqd%w__{Cp-O^6=v0>V4mIMOP`^JiP+=on9ry!h-Bwws34szGbbryqU9
z%WuA8WfH@~9j6evzFu&>Sx_|<kEaQ#DmLwkEG5Um)9rV-ap3ynkD2c7c=+ZGUDu<l
zip}jU-h0NB%0(s=i>obkdvq@LqN1u1G1DI&(Qe?`(<if+d;(SBC(qQKn1%_TGvEE?
z5os#6Uwp!Pb488h<@7aL!*smo_}$Mjx}=6Jo0^mZ<9>(oku;N5R!?qOELQmaYhu@v
zq`*xhr*1@OJx8t<Lz9PZUh$Y-QExUp{o)JAg5%+UoNaSc24ii>ZIFV{9ck(c4+yQX
zRZY__XBV}ACo%Q|RoxV+Q-*$Uge;L-qin_4cjN@kYDthaqN-5|a{ow3g+LgxWb8`^
zS}oS(5ZJx`7Of>!Re>83A|j2ZUMyK`Zt+9U{hRMNo=!aU9lIg%quUoq6nX3rB&0S>
zIWPpMY;i#>%<egSHG>l6W*XdB8Yf#98LCqA9yF6t2-PCjTS5rj|MF){ySD%=+J+$~
z6q(g#3%P78ZM#5KHM*+tQ$|*XX0;$C*mWmjRxDNxl0Y6i($pc6qiL6DX*un8+`qkJ
zeYHVZ!_akn_w{Su9z8-Ce)Q>6w%cWK!wU{KOw>wKH7(ODjS?D=31y18iR(^isgNl^
z25k*FN@SKeH$sdkt!G%-Bh9khG=GB+X*AiDV%eX1_J^Kod5v=uX?#TZfiVQMi6C@|
zmL5lv5VWfmq3@AWP&W-SE*ZKL&1%Dx8$SBv3#O@~SuR+c4I*X|o;>FuNFj>vhR-DL
z5Y>wR@ZbF%CO7dP|IdC4y<GA3&1?SR=iiWGW_`7#HG&^~{t30U2n4ona1o{?FfyQ0
z#!a3qOR*O*p|qqsK9YjxbULD~WVv|?+Rk|y@YSz=QR3NYFt%o#Jayd=T}P{77_Ry0
z-};;E?%wd?<EPx--0;oUpK-m}@YR<uIZTfG`yGegvu-U<uD67gsAp5PBn>f3n#vTl
zPY6uYKvS2ji>hhL{acjWgud_j=H**ny?KWZ5$6J3Kaf%(yiMaYHyLMTK$)6TelHI!
z1-{r0UmAh==g&p+atKknfHM*>qX6f;fXnw~p5o6$z?o543b;DQ9nT96Auw97-K;pC
z`tttdJl(6}6(F^lRU~QlBPi<@Dc(P;m{m|vnkb;ih(iei63WcP&pE^5EK^R2|L*ty
zlz;JuKj-iM&CmJWzx7-E^l$z)AAkNS-ToEEEU?WDF1;&AZH)7&^PF~%<TUc^<DbxV
zeF<?=2FaP!81O)#aFCLBNL`Wqh*6TU-y;a%_w4RoQP&MpSGXX_(n5-y{%u4Uzq@;&
zZ4@`#6{o2uc1L6dt&B8Fi?)`>m*26qAER?(zd!Q0Kl0fZAF&aUckf`_pLlw`B8$w!
zV^6)=P+wgmQzVTYeM;!fEpp*F-tXv+BX8f_Q7;$#_;-Gn`uYhEKl?M{<k)Uk^rwOQ
z-3j+_;$~4%FP2nF(fh!VBK4xCwhQhKCxlrct>v`8LwZk`9M$F-@!1xAbqh^PH=S^Y
z1B??g1s-1gig#cCoc?fR?0Pnv6*teH5vIWDaZm3{zVzru<df%Abw#^f5xr;Z9Br%6
zRfUp@^|EETS#dfZ$*~|9!{G=@(Ka<gDZ0}U=SG|lWRXkuw=9GtW%Y$bl(S$MI^58s
ztR=;Ou|>-rLO^LVGY2TLmKX^xV4AkrUTRYaI?-c|Wt<$*1^hUah|v_e|LQAhtFY@O
zdES*aR+Cd^SuLomLfA7^jhr&cLw`K*?iYWIySgdSzQL2LmZT-YED&hY)RTfMUAe7L
zrfA2rwP;n5ZC@@r0gAHAl`~n%xg+u6yK-UokK}%bZWkoAplVk|vbk8{QU*65NCZC;
z#&SVeBQc9+o{kdPRAo@e9_>d$3aF|=D2exu5DOyYeZ)!6Fa?YetZG4s6KUEL<1O`~
zL01*V8jPBozmyQUNGDBG5pzaMg{m5I2vD`i#R6qp{NT~f7mtLxEvpn&QC~l!zIrjk
z$G^G9;_4Y*7-|c;(&U(ETZ^g`m9k8y$koj?Z{EG)YQ04$#bUKU&o}hW7s+gv29!ao
zi5y2(>lNO6n&lcF1a9i^-H~Q{gTUj5BQYhEGE{YglhExR5P8JAiFR{K97k#+`49j7
zzs+#m6LTVZN2MX-NkXwZJP?G!)Jt}6zedSGKPFaJ*N7m%dz!XE>df}~DYmK@h9dWB
z+nT=Hk+b2GkDk!&4+vGEZABJ^AW7Ez2UmZO_2!Bk9FNx;Wh#d5fRX`Y1z-I37swci
zapdFAe~aOGM8t_^vBZrX<Iv-?W%N+jx}b(u@$I)?lTu`ReM8KWY4rHXF-;Er^opN+
z_JYULiEqDs!{OM`bpvni4(v}I<V*|+p-WO~u1qQV7c^j==Ffb984Q?tckkB&yiaV*
z=XJz6pWxRx2J=o}ZUi_t0+-?|FK&VI10_N!){^aZJ(H@ZSwdI{mU;fUoKsPV<g)+I
zynin_7elEtu@Q<SG=*3YhI0-@Ix8JS$)_mt3Vh1^{-1uyzx>lLX_|k^Pk!<l|MS29
zyL|HLr+o6GnlvT4hdq-Iw96KiEY>c7vYs_nB211<raRmdLcp3caw2eU!ek9fGaet%
zdWq}akldQ4Ry2zZm9-qZ4j&>_TQT$<$J6n9U86s!8Q5MeSTAda{ejKXEA9^m9^Umd
zmErnDG1wWq4npRye*Q~dANHu#73$eX+<*HmgZKRCljqpla5(l@5i#DOP6zJq-l0R}
z)1UqX-EMh&^)-+0?wNc>)oa$De!;Wf{ypr~4bdgsyRT3tGaV;lC_C}5Ucck*;l$I^
zL_iQk*}aJDSZr=+o_<F16JtMc*d0Jv79W2?b@il3dvf5k9~d6qQrC+5`WfH7e8b(_
zI{^IT@dDIhMc+Gq@mF7yh2;73Yy31)H8me?o|A=Uawif+YrkABSuU4U*0Q~Q%3^im
z-K%f#)5PREq%w^21S}|!nKlZoHP%*WA&b4Glt|MMg<_mW+&Ily<O1C;k*0yH5yKJh
z2Lw=whQS4<$>E2Q7!zvs6q&(IBe6T7L~+HEG9ZEAN4lXW%fxYXR6-P2#NcUG3zn5-
zIPI}k6_b(}7!G@;X+q@4-RoDxsb{m=Ae6)?UE-Q!BqvAC<$}t-tWI*4#1s)ZlB2^K
zMGE<1j1oeb0GxM>yLY7BJC@sPUVL<o+gOq?xFkx0K{4|Q{RzK+BsLmrYpTr^Aw@!t
zSglDS(HKR-GYvzzDWpV*g}gJlfTqVNN!8R0Q)K<*3CkB7mYcV9$3t;tGnR;890P4t
z!iKCi<@0e9ZW__rl8go^@iCGBZ7Y&AkYx!?Nr8H`VY|9!*gf!g{~EixV)Oh97MpAO
z?)x!<|GFO1)Z}W-ba+V&1-%)%13o3F*2t!1v09=J4R0UrS$W6DpMJs6ojBg#QMV;K
zB)GCx!1YX1$LlvQd2;&{RX4~i$RS|0fmBFSY4+D&EZ6kIfHF19Rg1E>bdRqH8LqBx
zP{!hpJKXWa`sRkm?!<W7<9&y%S0qfa4(|>u*Ei%8IGo-v^hZRFbdQguoLFqO_`WB^
z#A<uP*RS8OjgkF+&&~A}ZrV{*6}qxiwIydm??!C3<l5Bu=sE3<Y&H!_YF5jN<x<nO
zD=;;u$35H46QoG2HVr-$_*-kqH0@~?R}`k6W4+mM+&yAs0;}nd2e#MOOnqSJPdM+m
z+FT<-#voWP8g93nVvH3e-@JUyyZasAy}e`dff$+f0<#(LoH$ri4`pp2&R)(Jt^$0x
zD!32;&WHdrLcqU*BIL6-LK%L<OqiVW8UK}^jq~#qE9AN{7^8}_Yn=TfVw#DCzfPJg
zlm3Tph4TxUdzW)Klu>HtSG<247qsHiu)P1PesuiFpZyho{Ig%OS}pkx{=I*PfA@Dj
z=4x59SYNZedd}VBJCx2W7b~Kla8plZ7bG8M!?i@^9qqEEYL_VA7lbQ9aFLWpoEMn3
zDjpXxGr0*}Z%O08+wb18xxGPE-*3hEgPH-id*m;^>RB#Yo<H5Nd2&k#naY5wYE()r
zZX$PX<m*5GlEr4hRb3MvCxT1dK6}E|%@s;3o__o>(=hSo>#uRU2U;0v+o}|@@kly#
z2mwtcc=n^e#qHnv`<V3#YZUc*!^2k-Z7s2@1y@$_^vRETynD+p|LPmQdG~Xww&jbz
z@gth1Ef9RxEYD;#TTDMrr-5bPqv{?X0@083ufHV@9Y)lcoY7Lyv=stLGKO}0h0un6
z*nyB}Ey*#^)QgfJ*!LWQV;DxvVh!U!3J#F?Tvivm`#Z*dD&ZIf55jqmY8q@^Bd~?^
zqAP?Hw96Ln6ft^I^o)I9T&-plU>qH$U0_!Y(pHE#f<F<wFYM<mh|1t)VCW8r`&Ve&
z;*JlD`+KZbthYr5mB)@eb<~x?JilSTKhm9g{CF<35!Dty^x<?Qk{O4NFb#y!VXbAc
zSkg2t(GL(6x^74zl!>lgfGMUrxx2>=2SiBdc8ie$89gGUlJAfOF2*85${8WbcSK6b
zs)DLr!}gk?>-qXuzhdZ4ES3wb2xRZc^W?9rDF#i`Ku(F_tQU8Iz8~oOVZOyE#AZvq
z+QKl>pjfP0mTAcpHC3}lkx)?h3$`k3@fe6g%#LHaG#xWXK;}dWp2-DdyC!P`HKgE6
z66K)-Tga^1R3)m_N0e=`%?hnF**SjgZu1*>glV8{EPV)s7#YSBi`9B=U=4l0=XmUR
z@#16Lu%+ud-n@E6Yl|_AkR=4`@$GkPR!e%JFj{c4SR-vq^oi&^2!*LMQkKxFvUg5I
zAz5y35YjMZkI@y%R>kKa8<y>w?s%XZ9ZDL6vY-`}xgte~&ePN_&0>K_nV~yyy<T(w
z_`u{H!bOr2JUpKG?AeOGKhgCA%jJThJF;k(xM9b*f8;b}o`3c!t{bq;8kIfS3vAPJ
z*zJ)yV64W+SneazkV7IR2Qu;Y-76|x(bNWA6>ktkf*hz!`3|l(Hv~5piCHY)$3@c;
zrU@nC>60r?#}k|Nf>*EK(6*NK>IGV44#y++JI7}qJ>`p!p71;U=ltT!Z+ZFVH7ON>
zGN~YUXSbh}GjcXiIuG}yStzo{v&qnzP*@rVQC>D>DnFwPXB*{W2P`gAcp)GbCO@+z
z@I7~ci*g}T?1>jmQ{tm1H(OMn8xAoOJ?GYDMh*nC6XI-+l)}vC$rLXJB}l$lv!4l^
zDTd2qh<S~{EX|a%pcSX%fq(S>{Ac{L|GwdW^S}D{`A`4XKjPH)h!nW{?hRqHEDpHQ
zA)){bGQsWhF9_3sl9?<7xw!&i!B6!4J2H;vAO8qY2nF)gqg2iIqu*k9eBksDNy+ei
z?;Ss=8HmvnL*nsqPbHzQiY7N(&7*{NwLlwFCef7W-yS&Zd$dSgty@&HV(JI#re<@q
z<bIrZxVuM5`0c;(Ier*<_wJF+%^IZyLKKV4tIdK{YtS;$J?_|l_YK3bL+6^LG`ZVz
zy|h%#f`nvpj^)-6rjfg22f-0fd)~ggM=Q;8SyQh!q;^GrI+k5bm<ZE|-#xH25^e1q
zA3GKDs}kIP^f|IxaCo>Ujwha6wQR1|RFxsc!0G-Sc|39Z><L!asHP=n@KaAt&`pl}
zx33UdvD{oSO(i}xg-}dMj49e*L6Kj@l2kZMMVH=nC;HQXcaHUXiESz-?}#x{i4u(p
zcdVMKICoVQqYJoU#0yE1nv^r${vBG15<%=7PAZfVXaTN2aXK6jb%kzQ2oX0<w95r<
zaCDCcT4UKhdpgUkM(Wj?Puewy`v*jdEEf$`Wf4&@Ob+iHT9wY++N!L9vLZ>Cazfdf
zx^BxfcjgT^M@kW6G<97hh*oPVB@s%Jbwdy}R)B~RD>Kbiu|c+0ldWO#^dqY6Q&P8M
zczn&|Q;~!2cZeD7@GfHO682%V#WoAH)(mfVh@&U;3!<(F(ed&Ze@Y?~LgFw*R-0?C
zl)ijFPKP~xoKR8}@}?{|=CL0TStCtCEUF@}JM9qeL`);eOY#)Z9>!euv(gqM#q}p*
zcfx&O%<`LfjNKlU`a%rFal5`kR~1!zg+Ni)ig6h6-2rlDyWLW^E2h4qvK231z2W-$
zns&V*D~+<5)9JvbU6-4%EDnU3i1#HQBgq=m)(DwPj)Zd@c5mndw3T3e^@LSrIUV-|
zp|Hzq7Cus~pYYN43BeC#AFV3fG+}H-z1WZuC}T*|iDvr@T`dtgGxj@lvnYXT#v-+5
z-L{u&Ji3HgR&?@EjHSFQ>KB<Yb;UIGR66o-|AynK=j!$u%yPcdafh;ssXx(Fie;lv
zx~wyNC^pMapS)mnp>PtWsvwLyp*6)OJ;Ns-eMG%j@zpPXK~rg>pAaHbRRxKxjONMH
z4f4y^j8njm6F19>&wuX=y4{*Te|gVWZyqiLLopi`$ytA&(s?<_g|K&C4wV0%C;ho-
zN(GsCStI1k`;Z`YF0M1fU#i(IcrI?acpMb4VO|5A$%>p^40S<Y`r+cOP^6aU$D9MG
z#H>=t<@e8FOnFW}L_i9m3^j8<lS1UZ)B&WP-5BQ|NGTWjWef<JI36Sa)j#_0`1^nV
z@AKl>lJ0Tjv_EpQxu#t|q1K9F=&{Y3;c#E_UQ(FX9FgD^BEs=u&vLOXR_@i3G#p4t
zqHK#)hT*s;hlyv;uh~ED5O(psuF)UV42W8D`@E&w9p_j@Ssa;LLx1Y1s|GC;p?7GY
z8+sI$7tgP`-Y&=?((NAU4vL}cxxag4=tn;J=myt!{PL?i`oXce*|54<^SC?l@NP%_
z^)J}Gc!9iqL1hdwI(83_teO`*|MUxvFTdvXtGCFerCv6)tCsD0&HiD>t8d@(@$(yk
zR=hnNShPj=|McUJxw(FVRFc<U{fd`=^-F|ztTzq9wsR2BlB;3MA|=|*HBUbIG0o<d
z$CuymR!fHAgiv6snv^mj26Ve5qY*g~-3TIMnhKOA^@=<d8%(7wRtV~Li5mv89}A2w
z1ldoF!-SENYO!GSk>l~m@pvjwr&Q?45oD3-q?8%^9vuR;Rm9ODs~RN@ZX7u6_hgDm
zkjx1o6RNVsctcd=;263C%XUE&n%ERFXSX{NMn_#W)Qcr`U9538t7W-#mBuzpbQn=W
z5!{5IHQ3|Wk*0|#HAZSs3ZF74Q*Io?5!W9}gP=4r`-}CmuBzfKKaDiCVzFpJYT`6d
z8$(9Y%bH%dxc-EnMwU%ebjc}^V<2V;S(CHE`Aim;)$I*SA+d{h_^B_^(H`2>25Sv6
z1yb}Vzb7vYF=QMPtt~kRVjM|QVVat4Q!*iBa7YEwjhyy-tdcA@Ym_Z=G@&fIlI)Ks
zy1NJJ?UvQm6V!MlPXo$m#GKvWroIFs*_o&mBFT;T;lwolW~UY|Hycj-x2(4}jDx0K
ztO;4-{lwUh#NbOKtTp2}P-#iKT$OwQAy}@L2pIsy{ry|~(Bl)d&e6G!X1QdsUNW4H
z_|a3z!Ui{WLx>59BMd#-YO?D{QS;@W{RwgKw5{UBXFnoqLwkLLsatXioE}~ga$>!@
zxxm$@?#Oz%VY7b1&>QM@$!YM^O+~wGORx=zM-d?gTHA6QN9txlmm^Zu{P@Q|<@E5X
z*e;V;7)iZc5@KLD-4mnZqi0vFHdpk%m=awzTe67U-YjVp4E;n51LHU$Wm9CB(*R1N
zbmVxt!`PY_MxH+1Qa3f*^$pe(66}+kYf?(|gX6h=N{F7_<B`?%lGUo>_x{eO{PCZB
zg|dbx&p$zqN9@yv|LKpO@c;SwH~iV_BfD<oG`QKrAR*)jj)QaNfik)%nJ!YpQjCf7
zE<fdqBB9jwVt%Bw{vi7KeGozxVvcFfKzc9vQ+h@v66b)W!fs4dMv<GEahw?4gcP!@
zVU_$fmCsoX0ht_%o%Kbcm=eMHITcZr)n16Rt#iJ_Ij5ZG?STSSUW1s|C^-fEFz{dh
z*Z)`k&;O_Y1M%5M#1L3q-=eFAx4-;D>Sl%OPY5aKc5jeE)A<gCCis!Q>&a5lv@OTZ
zlf(mY*kh_SK4y%us3eMCg-lenCb#wXx<-FcGjM(Tl)9?<&IKB4IUPI3seIYo=(yh<
zxZ19;O7hwIBT@*65KuAD?+;j8k+MVz#j&55Qo=T6=jz%O)e}KAO{63+!ceyr58rz9
z<DT`ee#y-=vRF23w=LEb__=A9=<92yPLqx1>gJj-Oza<bgb?}o^QUyDp3}~=`RK>2
zt~RWfE!VeCiU+?C+&+DdjGpPZr!o>F3X#P19jm5dNU&V5kTT)+?-(E7A_;6(Em8|i
z(-I=M;8?9!Xk9{b{4kJHWO4)ba)B`hZB{c%mS~qX?fN-GH*k7*g8*z(;U_`WtRRDP
z6S}I{8cS8T9F9kX5ExVT(#-q<Qh`aQ=wX6J%ChH*0c(q9S<NOJLq8yt!Bh=7Mk2XD
z-%^29RP6?9Eotg;U5BhH^kT_+xg>N)3?*_`8H=r#WG9K<(JZ#K*9}4nrrkZm)T4}q
zSVCG{a;W>aG~I#j@s2p1sH{aI5WcKilvW5;9Kj~<IP?Q9B$RE)!x5t-wq6j&0iO#*
z=i-D^8C5N5R#zz75~cy)@0i9t04fKBpAbq@FIt*qOGxt#NJ<n5Ees@692_MmTAIZY
zsWm2nG6OMZd<axZBF6*nbfjGsZE%W#q$*5RW2_+tM?&I~pc_W|?t~y=Vrj5!y+CSJ
z9NDBK<}&g5X~a(he)4oV<A>A5ll?dKxW9Xcn|kWyn$>E}H1wn#NutHtntEL%v*UOq
z=ZNnQ+>P%4+uob~SaxOUeZM`P@lG*jCNoJE$(p*Vx~nz85?}~63_sa^@L%dT8wUJf
zTR>YaTadb2R!yub7Rel9ywe%?<cEEe-Fh&@f*`BQ9p6(TBanF`?mg$Mz1FvWABe<p
ze?%FLvL#As%CbQTNuw=BYhv{I{SAWdaA2N1qn|0;C9PNzeIiUFS}F=tuqY>P?(ey{
zTyk;wg5l61w4$Ge?14ERuvQ}^n3$O7k)m!1M4GDR^%uXOu!_1Y8QsY3%^jE5S7a8L
z%;}_MN#8xv_gkdW6m^3x4P||SDQfn+9rdbZJU&twjnM^Fdq!bfj4e@k>g73&QtY;m
ztkxT5KhxF=`u#f^TQMFbQaBc?n*AZ+oW=`@krr)g4&4D^au&vC-~X7ZY%`wrcwp{5
zAtWx&);v7!shXD3DyquRHWhbof5&GpuQ<QF<gs`7UZH%&^zgvT53cy^yC0)%$=ipX
zfAN3+nnOSGaOg70F#NTD;Jt%C#{7`i#t_r#Vf{DK9=};dl<&I-Cy3FB{&=DY<xM2`
zCaWQY@YjDY1*9CdBu<4$61f9WRuZKp$3uS#srqY3(|hD{is|(K<yFR%G7cfd#5hlA
zojaRoa~fryd<{zG+=Xv^2>B#0rRFrx$i)f*i2RFx^>cpl%U|--pZ*Dprlva{afcno
zRLnu6nb8=Ac_M@!2oz1tFm|jqXG~EbghQ(WBMPFj2qCGe8aF!}2_X~pYD3Z1?6=2n
zb&dXB#lSCr_B$?@6>ji6+-->wtcoP5D4QkyJX5VU*x=AgGVBj1ZAosV+YO{7xxL$>
ztf8vwY^IWyuWk-}eDx8_a|?aXc7MP>9%z;;oO;KvesjYxdR{%=us&a-rQzbk=h%w{
zhvN=E9EsyV7;=PXVJ)GoIcqCE{`fgJH;=r&>3Q)-f6V#wYh)NHRBjRZ5Q%lg+0`ZK
zY=w>i5l*HuA@e$8yJqw3B|<CqH}CTNZF5c%f^i&KtS%^;hQt1lf2Jr<G2rLvg!>e9
z$DYC#n6jqtdWPAd6wJqfDGY>}<6)0FSW-$f=hu{N%ea5Uxe<Y&T{IwI7{>hO*X?PZ
z7KzbemCcA!HxpxI82ao$)dpP_x$^aq7z5fElr34+P5%CFpHIiSpr0MnagR)qvS}!n
z8;Yi-v?d!1Lqsa^O(uMF9_M|&__V|Y4=NW)A^7YC5HL*>_rLoknuJn<qAKw|F;7R5
zF-V>9$!T`@dBTTCI<<F6nrVL^%mb_SIYAZpaSzI{SS~5#hNMe`)VK)3R>(e^8nwz*
z^>EywY{BB<ifVnvGzE_PN21GKj3_0xC~!!m5iFLg(?t$L*WsK)n;d;wX+ancG&<54
z!J+FYtU;Q*Nie3orWI1|h?MP;X1S!$d51konOoD^5@Ys7SYwGP5>i1?Ehw4|DF%GE
zWsZN~o&g_Ntu`2`@zc!B{R4$bH0Q7IAu*3Lecw@3nYnS;KB7&+6h*$q$|YrC*l+jj
z4+peSY!(Zus>aL;Hx3*Rd+Ne)d3nkGe#bZtl%;``&_<kMe<h3h8O`R3I2>{(OG}d0
z#A(jsqj{vNS_&=j;|`Lcsh{J6KqSFty&?FVp)ot34ZyRfX;*A-Zz#*${fZ$_wl!6K
zP7;aT-6LglPS@{fF0T;Upym-52FBwfXH|_b7NsmHWcPlUM?y?UA!(Ww(>!C!1tAGq
zQSvbCnEE|ZE@@Wh6lK9Y^w`BE@$M~SzsDFwRaHDZJg{s^4!Z}GNNBAnOSrsPV#|V6
zV{qQ##u05w`hADc7p&GTvl6t;75?mkdcSA8+hVomPd@sDpZxI;dAz$Nroiim1Mlwl
z{M9di$FU!nruiE)U}kLi(<kEnCc%5-e)`65@E(!;9&D6(3Npv~3U%T<<ZYc0bDl(=
zH{^dWXME(2sZgiR10uud)`q_Cb3$x-4^aMV$Pw?o4JTj0`(h!6h$B&yc^fD=M>>rO
zazy+onpyDgm>r3k-JpspZ-xcW>;wP9|M<`N?#C}Ud;SUQtNiouozIv<V*B`-cVB)%
zoR4(72aGJJv}Lj0kP_6|GW543Aqd_elmL;K#))YdQC3nK&9NI0Iz!LCjef6U;CHX@
zDL7;Dk;5?4log9b!FIRD)Jy8cD(`egNZ#Ye8FW7NEY@p|hYn#Bh@2fRw4mH9QPWI+
z_l_`)5F%1rge}mO;lo!Owui{aKlqf*#U)l(Xe|*{iJuM}-o9afJb(`vE$NRvMP0Do
zZ1B<Xc>h3KSU&#2cUiuCNmW?z0_G7nk9g;p$Glftlr^|a0IRB+$vLXcIh&WCqUr`e
zOzfpV6gr#jq@a)uBBgvYSBl5SEp=1k{LC~B)a{abbBQ*JaknFk6N}}7vN>n&woH$=
z_{m_5r6?=h>A`f|-k&)8;Kz=X5=GTwv|u1G4kJ~&;QaE65d6v23=lHq%z2cO2r}7j
zdA7-n?9=Lua}HBCl+6mad!QeB2#QI2q$w%t1xUqey~f%y&&pg5l@Teo$L68O@9s%R
zqSgc?^Y(xe0x1K@jYQ`dq6d+P(PwuqlA<g@D!h*jeb1~F${4gzxXF=ZURSb(C3(j%
z_sp}$4HKigMHCJ7a>3cv8TERJKrtV7>~C+F=SW(9fY^M9@s9cUfFH6+R11NhM*KJ-
zQ$nU37Ujus8SgyPG*OjBzKKXlbPhj`Xr(C1ib_aS3Jg<3D4Dk$#u2FtQjF|(d+Mga
z*al+43QfIQQ<s@GIJzkpAb!GFgE9pP!8A>HK$I=ET~pL6gda#^=Gc$F&!zgePF2-l
zjlzu`(Mycfq<KUj@pIs?yCnsWHkN6g2}w{_HAPvoTm`(JiM~e*$9hqq3d`a6h>(e<
zt_h}K?;^>2#%W@;KBuZGg3Dc&u{*G6HyO~962b_!yE{mN;AaYBSuM}l?(TDzMhf`x
zz_8t6R}I1S+}ymQsVaO(tS+A6x*bO3O_&f9v!5wT1q5avD2oPTmxKh}Vb3&<+`apn
zdQoGR*9@a)zuTR%I|L68cSvQZmgkg3$>DIo%@b8!<)qcBp=uP#m$-Ri+r2^CHT_{n
z)FZYL9QTh@R-$yIFp?<Yu)U|QO6sy?8b(TM^S#6vw9%MC;N1vuh6J`KIDh$L@RB$m
z@F7PFkKKXbJTZDc{?13B6wAwNj@v!k?E|Z-;<N94M%0S`{LNdw{_Pk1`Cop{ewes<
z9CDV*F!0x=zg&>S^d5YaG3<!by!~$&{c_?|Zs*6~;RFKmA)wwH4&fV$AyN|N`LBHk
zLcuA-s%aXAzR%Dzt-paO#ZyMa`+^`L@ZRS!gw8>B7?WN5lV_N6@nMZXD=_3P!F#w;
z9uNqo*<)m&P$vKTG4f~s=fB~9`k()I{@eeqrn>xy=q85#NMUQv&YrWqzvDO@aPPZP
z5wz0u{egvs(kww)hT{#&)Fd-8&l6H6?(grYj6w;?=KO-)@E>8duqrfCNJc*(PsTPa
zGrZ=@&ws=E{G3&xNh$L6Z@*-oCbU%;1eezvil*eUF6p|SYO%=X44>g0!@~n4P)4yh
zTO-R7BwU<r2;s=?;efIQ%ND?}IltuasQA_Ientp^rYe~R$9{jLnrEbt#NgTAZF%w0
zN1T87Jg3vm1Mze-9=eY4xFxt5fWdnN5#a-c1kx#nGzNy-ujuaXc)WW@3?ZMkwLw=U
zl0^S_&os_?UOnd!o~m9#Ot_&#Q6ogc^&MgC2!{i<F!`YuJYIlE2`wej6hv~On+SOZ
z7E)q%j@jk8x*NxQ(ku+#gOoK|*=&Gm9GZZZf+!T@G}BZiwp}nDdz_n@ontz7DCgO|
z`wDm5BNq)roH320T5ga?f)9k$p@PeXni!bIK8G8XOYG_eS}T+YOrv8=9$_>(PU!b8
z@E9;!vnXq1@^}yP9Pr_^|LYwDNeY1}!Z=MB1u8n`c=|qDA&LsC3xqW&DOj!+Y%cS>
z4ewD?&(Y60yJ6bnq(K@BwkGKUbAqA_ha)~@*pC+Ed7nI8_(Ec;5+N0{^XM3vx+79X
zv@Y<|j4>K71gdOEA)953<tispBA8tQU4j%iKT(zyb<^bH(|c0P025<OPE8ZJmFS~m
zjEQvUv2&mh0c#RoCc^yfz#)HIb$R^)V-)lLEh$;5(%}4zu38KdKlg~Bv!PQL*k(gN
z9O?JB6xEWl*)R@U=CPx!8*H(lvVxQ*oR5eELTkoh;Ou;XZCczo<m?flNFvbn_ml#P
zqM_g4BVwRvR?H7yargF1nzQGm=qajIF3N0+mtFqMhM9J`!79VFd&r?$(o!``&M>6t
zkufs4301bVix$`KaY0a51y$Wrmu;SDYss*Cr0)lY{=n7cbBcPwFirSrX3;dr_eraA
zg_|Zwo~b*awPJGFsjkZfswol08fgsOFi{s3x=^^uQ(B8&tf=aS#n~0>^%ck6J=^<R
zN@Hm@XSm^r^Nu1(lBsEn71MY`Nr`S6bZv7HAT*{diP52}64NxaZG){eMl@7E{vrE^
zcO3g8vZx7ClZ@s0vkTl;@AxnO{-=yl@$-NCId8WIra2c(@ctB?#|X2F6e>rS3n{+I
zWI(<LkDdOU;t6&H!3T`e#M6d>IE7IqB=hVkn&O*Ne<CA@ZJE*NF{P8y?|t5bI3+V?
zuE~2Cl9V|BDhkOwPV~b>IB^Ny<ELYY7^6utM=Nt0Lrk+LiTN8Z(0jw-|Mgd&^I!bq
zx6HGnX&0EX%v~ATusOTJ9Ue&Qa~7Kqu+4&DzooDh!{LZA4aK6tA9sWkr$VVbmRYUx
zsv<El%mY$=yO6Wrs~Bjkq#FmuDX?C)9Ct@jOk6B1wYugvU*7Sv{ug}qy;n#jNXDX<
zHEEieoCoI^hk>FjFiN891>><}ICl8~To~5p8=A9oys;!dQuu(-M{aN4(eLlM+N@cA
z@Cnt`ORB4ebvH4OM;dDo(X-wxz)e&|fhjBIAYgUL;2ejW*ED6$m={uku|#R{$wPGb
zxkoBRsTG0zCJy^8q{QLzhV8>0+xso1C@HTlY1$g04bJBerb>{G2kze9V++fAy+o!!
z*L94_pv#)>*yE>}5F*R<l3^GKDX?5E(Ml40%(>gqGY>hKsVG}?jC8}mG<meiev0Y@
z&q+QL(L!`+ZLrp0gv3uHl88tit)Q+7+%V&XrjQk(+v0Y&G}>dIUow^pq=R{!nEQS9
z6ZDe0uCb;>8;dDhf>fxop<G{6uCK7JM;`AO=6iNwW(W~$5=GM@WCbyDR!MXmiE(DP
z>+mVB=$WEKk&Tc_TY_`nFvG1Z5#p2;FEdS1N`*3(I1hy3$h5sB6-7SL4?WUYl%Me3
zJ<~iPRNm02$^xyMQ_i<WYD<`B5)o?+&SfZ5j!>3R=3V9KxToLVQ&ly&2dGFY=N#a2
zB(xL~?|t6vK%kUC%al*EA!4;=>N*blE!Z5?6M>>$<ZYPKj8ddTm?viE=?@R+c)*@5
z5h)PE{QF$0f9q5h1-39SC``2k(~`z~|0v3m)#i$^-%{4uD>EK;lvXl~6J6J_s@7EX
zl5u;-e%tZE<)=(Tmv86{c}{LClpc{O=EC1iCj@V1^-tqOQZpfOdhj^Tn-(EIV0!z8
zX_~MyFduiAqNML<o_+jB+}=H4>;j=8`|Xzf_TiLjnBDWyc}Sjd?ua2^k@#-U;r1)a
z)isq7lx0H-j;Y@>A4ldf2U=N~2w_Gkk>jk#yaqTQc7(Y{mK6v?Sv3sZp4kPg6zsYK
z&Dk@~FJ3TDE(b0-4{l)p@D35cD}%NInLNbA{_!h}-SFy@KjH56&rVRSnYw8hhmmD_
z$^LjC`hnfO=lXdz7fX>aiv_h*496{|EKx;?Ei_UjeC#O~4c>16PgyNcTJ!4FORj$T
zIWMm+dG`Dz{WS1LuU;_rdxB5&ljC3g`fK*X#M}El-UsH{;e9*>Damif3+X+knA7^A
zGzg)PCNl)|DZoh@3#aD4Fp_zmF$tu~84H5Up3v5!z0ba$5Ku<vWJqyZbCkItkjPAh
zjX9!vnmxy3kMj{@GQ}~(baMPl0vV2_Z60|7gwvrX%p<R!UGkGZ{sI5wr+>uvK3Jk+
z;{LDxHR;n&Xr6yc*(`8L&@R?In1XeY*GGMdsG?x8+%Q|sakrzEXC$Ne)vtfa%MZW9
z=HfZS{uUW~0-0qJf+K#*$NcYA3|w5Dv$;5D>bl&fnR}{2B81}dQjx@h-@O?)?za?m
z!8m%_vops10b5rb-@ZWwSXLF7n#+%WNF?AU$Km#t066ElyS*crl4enpfX$*pNx|$L
zK`QiW1v<B`SIZ4^@)S*tR^SerG%4!RB8!&A`E!cZIo?mWd7$$%%k>ItO1x9ZvchPE
zkvT$_5U9%nf~P{#_XFcJ5`w2}muyVU(D(RxrrYn(Weui6+Pq7jqB}9lA-M?`Jkl0a
zn=@>+Kok|Y!0cvR2uw+5`(CxAUMzDLA<RrupGPhy+-dZgPUpSD6b;Mu8BJXy$*%MH
zg#Hau(N7~*E37S;#t~}_w#?ZM^EBYlD5-J5(;xS=l|h&S=N$W+2X=S&n9^cx$<?cm
zSiSs=YIT7?(?7n2-CZ7&bRDVx990Mk2FfDP)A4pUGPgChzCh^`Tn>35NL)7)Q=+mq
zpInLpDM5vp$%#@>S%Xj-RkcW^m<|W#?wH|9MP@K4o6VoQ-4PXc*^O=sgekEoWC$b)
z(=aeOheuLX1wu>8sz6%9e9DIvG2-Vj&*$9?LQ>Xc?!b75GT93tw5F^V*#+-27(~SE
z!RY#q7$d<4R*MBpnIfr*0v`j;WzIqfkt7tRUZ9rGP_|?qj*O3Q8Ma#-30o{NMS&j&
zw(kZ^VRMGVAFzhQ4v{KmH`AU~L}{3ZK%pdK|44A2Y3wo9kWwaON^Mvy4MnpiDvcYR
zefc+3fBgTukTqLEXvUrqB7`hsH>5C@vhP9+B1|+QjeQx$F6*T1`!3s9vXdywpsWdF
ziLniz`}uy)_aFG)A78I~&i&z@d+zJ`ScT2fz<Hh&-F8C{@L_$799<^#7FU+#dFdbd
z;oWjd_f)EF5b0NuR!0j7IZCxzYTbw+4M*VLexFf!QbroxxuiryAfA<_n!M8Rl8rX@
zYtj!)+jTTBO#_5N2%-osZ#-4Yd>lG&oJ;HBmSGCQ=wA11&HGH(2N|14F=2DoH8<tF
z^;4N=Z<1vWvA-@cOyf9FvxNW46eSp+;3e(S=2^U%YCm3<+X1+urE2XrAHvyik}UjX
zV^4qjYfM@@>C1-+GR&?#23wjm@ns3Dw`)ZA$i!J$1p~QmJ56}An-YQ81|EMB!(>nI
zX68qeEzkn1)mV)zHsx0<^gJAVu|kcV%G}0_<GwBvu6w#wYwsHr-l6NQ>}b>A9=V6Q
zpbEJF3NVwB{DIl2;ZyK`21e{O8FQy7|Ial}jl||R0Ykkd`6YKt{C#GJGjwSe+P~P>
zd$`Hny#VZiK~`R7TZ{>xi`3|Qx4lBNEE}=w&tv<QE$D(noOTou#D^CN9jDm}D}KYv
z)f#h2zgK<z37shUq&8gHUG^a=YiNxdLumt!V8t8^l#lxYW(<;z^}#3TF(dS3fP_56
zgY7*+|1urtKVMXgUe+(XuJW2WvpY4r<M{2L-A<n%%{19rQGQly&Jryx8cD5Zm2#Y(
zyp*wK!?miqF$ufJyYW+}r<T@*9gFFCCC|z&^k2BnqD>tkc+F~$42=R2s;q<XiKK#(
z@tCAThb|9>$MzEku?!$nSr3&{>_`mL!Rb#2Jh|1f#Rq3qPO~r>rxTpSCgt#P<@b3u
zS(2|r2tMP{$^Q);0Av<(HYT-r$}24xo5#pdJ#Ni^W<6d<-}eiq{o>l1n>wvXtgzN2
zAUjG3TbH=D^_3?=X;Rcfg(}j6(xcKMeop5|&TAtg)Scz?n7bxxtaSt|s-+kbTTsj(
z;aX5!y?smF6+SEVF9`dv`6(?WIO7y!IHkJ7PX*5S6#6Vw|3t009*2-?{)p4}|AOWT
z6@#C&dyI;#46r19b;C5+<p_|Q*1hHV>$J1Y9RF=bFFcEvco}ez8u7!nkdmg<l|!87
z{UWhuMi|nU)wo4me9Lg_=gTc#d6Qw<7s)XSY86Pdh-Lb7zkQJ}J(zR{m6;d(@2BfL
zvszL3P828#q0LRddeyeF_Jf>Y`s)_*mD}zPGjLh%aLAr7(<JEDRf*xHgUv(}u8$15
zNKM$SIekIV(5yRvZ1?-)IBuOH`<}QP@RGKclA9K8kpR*)SfehIk<6UpGQWK*0>k_U
zt$piVg(y503t_<z^RSf^?}?byYfoJZxnzfeB(>Bg$|Ub>#Qja^apM)L>VvyRnE|Z^
zVHzjln8k`IQlqK;4$Hu|im^S!pcC{31}<fwm|4*E`R3dicSTMJ!&De*Yk#vu+-D*b
zh8>dCYPQRCDSJT4Cz;YYjbvLX8<iTrQ$(e1STQ<r7$zSlQ=}^LZkHnH)tC%J<Y%KA
zT=9+j3gHq_?dw}xJVLVIFQ?eEh#**${Z|6M^(@H4$oq1NBY3w<N6?a2Ut`WAQP(bx
z|M-0OiG5Fk4sP64e7vYvlNOR|_*WRZ6#nh2alz{DF}gqu5*_BVMcY;Ap*5L<fjeF9
za&^OI#4`G$ilN!4N*2IS;`WCSnWRux`j;tOgtLWO-`b8YW}MFC5$Cl6-(kmu#7vhr
zx-_(Rg>;2ev!rnoTOe@8I%ahyI#6jGtY{^AygNx0h#j%3M8!M_OT5aIKy(G*iXp0L
zSC2c@TDsO9yitiKca6I@!@%TSX&2(Nd+g^Mt+`G$&v_JV8^4=AZd|NYv3&Oy#n2YC
z<(kTkBKr?r9(7zuUrZRZ_fwCJn&(vhw94K)sYLa+{vZYKcS%nVMH`xIokba_AOhvL
zPEP+ShY<T0mpAW3x0EnD4|sd=c71nqnjOF1vE09S84f*N(dcvgz4I$4?Vio<Du?ZU
zVaP7&#P;GOS9<5hy_^rm<s9d0e}hoxA713xP6M;La=6Xz(~yhj=jXPULQjX#Q=74T
zF!TW5t1DWMl>C(#uh=YBorqPs-obpihXT5o7x|-Qbp;s{NDPFp0N3R_AP>B-gDz@%
z#Nq;P1c;Aflu1(<Cwj6wMUt*Qu-qVtUYmtaehKP7Il0A=#!;3^4_atL7mOlCc#-w4
zyQ32}UsOf3Zr|xCaN-g!S~BXZLhW7oTT2fQ$eV-iR)J3rx4-@j$W{*Oa|gT!Yuj^{
z9y9<mIR-spEH?=hEJ5tg7cH)L=KH9oUK8h<P8)-yanByCr}qNM77HV6T2b{gODVzE
zW<`Vyi&VK1n4;Xc61!NaD526v6b)h+_baSL_D<O@yhaVO2(B3-NFn|fNG5=qhB3Ae
z7COb{d2_)`A<&sKqfe^wowe#*SV_`9^4FiQw%~cikBWR&0=HQ7cw@V=lWMV}p3S73
zOe7CxkiRqx$D?n(fKI;_3~OSRVwoP{?=R90ouKMMYStG$40&RBbK$zaoNZxu>%467
zi_8&)ZI`&nu*&vhg@bB*!VL|m<cL6wDUg=S-_wa`6+)|8{@5@^Kh(q~i4@vByicNb
zfl7~of>j634-2Z0!licA9=G~(+_GsVVQjYt#zdfe1|}WdW;B>Lth%%XVzk11&Y@WV
z$q8}d*N(_lwy=yX_ts;X|N8wEB@v@HAtu+%n@l$)vFiv@#=3{bMj<!69TPsgmq{e5
zOT`ID37gEUG~i5UHn10ch#I4xE!u2C4>kD}uBI&QkNSMghZzr|OR!3?C<)PM7d2D4
znb+7ST@O(X8<$ne6bCZ-^;FjP{ez;r;o=gs?K+hGh%<C;Ek`zxI?ylYp>oXUJooaa
zGLT&}JH;i$dYpq&bljv<%TrWJO6oVO+&_vRIAP~|SRi@pN4fYlOo!Qn431s-Z(=~n
z(dSh9!5RmiBT6vFGzLOlml!KiK;JuaqCmJ{DHSZR7?&88Z0KpOWcgP05DMMXkb#?9
z(WEc21eVlXE#6?XEUV@;Ph#a^l%s4cG`F#f5z1pJ$@^TwV`kP!W6k?U!f$Rwwkv@l
zBH&+t<MGU|<>BunF%AwvPiFp!+aM!#JN@{tX4;N1(G!O-sFX_E$a`8z#s0ltYdyZt
zjHkGaWPp<N0J-!U%`|bh{b2$?5y?vFmdX0{D?jG|vpB8<k}@02ji<S-LVC=u|0wl!
zAsXaR4><Uaj%3M&Bh}ppYRZ91Q@J7aTAOlA%L<+-vjyI1YW?EKbK7eNF#~=*8lU6*
z5xI3E<YbEC*Dm)(F97i=9iGos5|7;#RnNnS!F2#L>(#t?WR+v<P`14NLypFc8L=kk
zwIk2--0a%LzfWM}M%e>-<BCx!vm2nL{r%|ElL!$@@1~HY(n`cJb;v&o&FH#Pp{xhH
zi<fejSm;@82p%BBfd0TA9M6MKR-cM22i9KFt2C*suN|Qaie1Od<YmZ`zVpP42$ngC
z^uC<e6Z>lf3UE9IU@q_P0dg|N5P7)Vm}F6B#dLULq)KHK8k?)wE#W$P50arKYVqd%
zL41E}0Mc>|2saL$3UiB{pI5P)O<H~3yQ%}81S{?=5cUA&u6@15I3m?bY_yZ<_)HNE
zWJ-_5;)mB(ZyruP0yPF9pqJO@>{r$hLy>C;*7L2Ww)>x;7vY~}@mn)HjDjD{DFG<S
z;zix`uyL++-6$U=)_DQQDmYg;JdB_*PVCm_A@w8|>x-rgcw4M<niFAsP8KOjTGr8i
z{4d2P`dAHkGubHhG!t1o#Vj6d5nzpP?9O&A2V3l#Em%!+1dhm^e*;2SPZTVY4e|(}
zLRAy0gY4=YPjkBz^~7I_QkrYb8wCCGQZE!q`)}JCt1@DXR&Ti|s^$Tk^N62P)f*Dm
zY3KlQa74j1Qp2F*DU4FhwawblIz;mgneS-pmbM-`LebxvI!m@iUgv|dMrN3Nf+l}~
zyAL_GDx>U^m9BXj%-9eT^PQbY(bKjgJ#xb>AlIM~Aq)cfSWB?|{=VFIMW+C)jg9n}
zrq?%yi@dvj9SD0t1vs@3_<zTma22v;4c#p{6uQzY8Rv@8P%J=176=Nyi;h~kR;RbP
z5=LJ>{(7J)U|53r%k!WB00+*qD0^9oV)SE@OAJ1$iB(2`j4W7;9eAjD1AGn9A|5t6
z$x$*1u2-%`DU2LD8Rf<Ho?$Nb*lAs<^q#>ml03VP@t@e$e+R3kvrZxmfevub2L_Lo
zmq&giUR2f6>KpNo!`$xkeu6SvX}jN$UH$VqE28u0NVY`9@nlO8>#WsCNDdPir8wBb
z^lpmY_)jCR+%h{0wjj;ZkghI0Y?yM>w6%vb>-L{J-?Fprs58JkC7rckcPisdzq?0O
z3)NUKlW7)wY^MDPEb70{&oe8alJU|*{=M=+HgC=;icV-~&y`r#F*SrLtK11MVn)1T
z80q_CjXL^pb~`pvSf<N>_wad?+Y_<0byo~<u7x*n4>s7#^Xwvoi*qgl1)BxzK}(vG
z+^rPy{eKrxPkc{!SZ)tuffQzx8Q&1y`n7W<0al=4i2q@gth!5xQEPK2R%mK1iA+|c
zyV6qT{idkEUFhe+dGt!YyR3@5M=QVZfD8y3C_nx{_Dujzr1qCoepQPDXCqR~%>A_b
z`77k(NwM_)P?>e3SVI;|_L+1go<Q=^BYh*d>~%QgG(ryhimK?BeRLbDX*GCqy#MP-
zjar7Rju%D+sDRkv?2k`-PIV@}6~JU=i|PRbatSP@S3pAB-1f=8-bZZl?Q<;4n&#7-
zFL?EFGB4$o3{4(wz8aWWz0-EOWgHF@@k~c&Ebqt`nVdhpJY4MGh~DwelL(rh_x3T`
zZCzCHzHC_rlfR6?-W8T`BS^{tXUeS|$MiY2Hihi#v1hl(a%|P?`2v5>FJ}wzFkwG_
zL|w2>ea{mopqwTo{`6%q=QVKsm4Q5|GvhZ7Ei*$7uUs{4?Me?Lqe$fy695hhGn2N&
zudoKp3GfM3+>@YLkk+SYs7QBo%26ULi)-2c+-4j;1La?jNds43bQu)eqWX3ZiGc^l
z_x`lRW0mR$_>@U!QMMsI{bE#=#3LDP4Wr)<xZQZ4js2J>v?=^OZ>7*B)^KYS?A6{?
z;|H*hw8lLV;Q`XK8~PP=IWr1qd%Mkq<6@o9!^%~M0B>&a!r_`H1;6e*4=4IczKmA>
z?($paFz3mFib~g#=z)=*L_9w(sfn^M=I4aHaRhjOWL_jRab{$5Jwr0?yYqYQ1BlVB
z>u~!)QEtMMT8ED^TIS9h(=sjr$mkwq!J^8z7nqjQRVKy-1C~UDvHeLnYa4WaG5CB8
zVjG<6Ua!gh9usXiB_dw^pb=6ISCA(E%5goA2>#N%w0|Dftrver82E={)-~#Vn_Dz_
z`u*(`7xb}GFj>pxW@MYI<523zNfN%$DMVfp)bRb|$2u-km&T3;p_mx1Li!dzdZ+c>
zm8<H4r^`8A3VpaLUFVz*rsHY>21~tP&)6RNvgu0<x8<t6g}li0O?I>6%Q_3DS*p#r
ztEbW0&ij!k8W$7$YNrJx7tdBSOzY4s-4wN)ZAdpe4d<laqx9Pkoqiyn`}6igP4A7G
z9+T3oN9ld@AM}vct~m#%Neb^L1dDyAHH}z4vHl=p4#56Wyic-jo?<sRtwvkC)vM)O
zg=w`6qr2|1r-AwE9Rizrkp`29^!On3Hjv=D-?QG-&n^pD{klAJy$6}*dyRceQcKrf
z{A^ob`=u`QF*A>2h^FcF$6tW{JjpogcBpIBY)%siB*a<^PAAqPC2^4t!VY_7w_j?}
zU}_)7Rnxaz-2PZhi*lG(VLcCLVq{VND2prUrQT?`(GG<_&@mg!2_1(KT-Mua<cU#y
zm973HuZ)O?+#fVRH*SL#9qMp%#Omp94R_HfLGK<jmKUAd4Ki6{pK*LeZy&JpsoQXi
zy}*h}n&PV$lhHXZeY#F(ght1DmPBH;(V=ci1w-ujczio3ZUNQEc>IrHM*T1O#g9%A
z39?C?6e<kgfr)KqX&uuV5FqGzYc0MXABx`z<*01O@3&n6dlT4+JiTW>oJEs-HRa&B
z%AeeS@PKxXCv$D`u)pqI06JtQY3TVs4R(&ZR@AxeEo<A$Ik_RoM<2zYr~fD_eh5L&
z5hDRm1y+Zl`~$2rPU6vQy|Z^1F@zL$Ev*<eAT=f(mI(Q*QW))fM|Au;iSvUT#7Ts<
z#Lnk1x-#TUdQg3$>_^Xg)s#$dJ)Mw1J(V$Mh3!S;TxEwgu9Hax)kPjj9RnoW>x8u1
zKNeqz%%ZnZc{Y_WgiE@TZkNwldb0fTuoAS&73#zzM)1xv&t5)ffGkh_Jq+9P&I8Ov
zqoaRLl^cp9RTgt&K9bFTY!EXZ{}&S@y>{mDv$?hP`406O%5D~9ZfgY>OWmk(ilj@S
ztU<`z+7_mzd8k}qUn1356~dmv;`A{^CdU@Z6{+D!;Ir7@k*uC$YCGTS7rQ6nTjA%2
z!&-WEQFjLlB6aGU?a2GmqtIivxr&KCUO0}ZU69L51^@cN0<7Ue-xP)7%1F44_W^|E
zOCG*xSZFsp+v5*&C&mU0Cua(;SdvJQ*y`%6LoUlHo`@)9rot8zFQt}|VM+$mPTv*C
zO=tVv(fqQ39)*w6R8pZeS53deUKj-LD7o+gQUkjb)uOk8QYYsYep7a{?0iBauQh6I
zq^T2R&~70(#d(S(Q)}M3ulD+ri66?l1m4@xNnccH-Bqi8me?J;cVmHi{yU$G8kR0f
zT|41P>MWZ+1TPaFnbDTUKjolSYa`%bcNQ4Xg1HC5cC)9pRD7oERbue|O4O2z<2e<W
zPIz`#WbS&&Gs6c$zE52u{Qkk%blv)pE8ahsC6RCLt<B7~g{U4g8l`O%X%1Og^>Nbh
z{eCY<8z-vYmyAAFRCO9RG@lBzXYQI%UMPs6E2p2|NdH9ZM(>&`C?`#1R9XO#o|WIH
zVq;N`PXrhs+=C|LFu_5y;|1ro=3%X9h~75{=_&-n85_!}wX<0#+K=PM^f~<C`22?{
zD2rEnV4Q@#zl`&oVSewzb*D!j&L;bf9lL$A%OG6g?NOlIf%p7<l46pgVdLGk3)Mx>
z8J$o=p{dPmX6vF+TCI*Z$r+d1*MjJhYNme!5l1JP$&i)N1lCV`Pn?c?7hpg?@~<nz
zhimQ6>w%K>YU0%vF<q?5ws11Uld7;UfdZ}zxJ{YlZ27RfO&Lp9M_P*N=nI$_y7}tK
z!h*UZnf|Xvw}fLQm{-J-fymptknyQ77hgaztu3hc!stHyW-9p?J8smlYV{V)V<f?@
zgx%8QxL80arLe(R)jbBmo+ti#6Qfx)V^p8nJeSlRo)yGNigbAV<!99*q1rW#?GTM8
zr-?Y5dsX;$DKSO<#8?%$Rs#4ijkFP6$soU`ZnL~&0FRp6=!3g-#;WNeu`WR1`dw^8
zT~_Ul_mU2OGaY=;H9~MLI`NH&sF34?5It5j^K%c=w^e07QlV~K?FTVqKEA*A3!L9n
z7nF-P0-}TZ=HEN^LxI8ly{nIQNR}4^A^y6gpZ0NGXQ1-bHso2nmRnU(x7qc9u`Pcu
z^YGM2?Wq|Vr5*qD*S*^poer-?<^9uJ3hqB@A*Y&aohwzjTA(lDktCr$b#qhKn+hEs
zFzAKKOVCVPQ>W|5#U5XTfhkp=S-N^{Q<DqNf{~}n?<z9Y9O0?!7l-V4KQhI^$^rVR
zsX%?cmB)Rz$v9-)EuA{VZhi=O>4Z4dD0*$7^40vjr@7WZT1U`GfQ_|Ja|ttS&0byf
zkuRBVBqm|pc}ONMHp0>cNonqlB4<ko7M)qH$gqGGeWqzzB66|<bV`bT1%Yi?KT((o
zcBkEl0Te<9&(r<Y(w{$+v;UX}_4gnJF*bU+dJfDSBs{YTi4@Ds9$&qUzrYX(ikj&M
zZro4G+(Hjnzz3+$qIBr9jEiI8#?v-Da?GJNf1FNkM395C6_fauJv&QU#Ibe6UOVc>
z(d1kBNj~!pf}~x2e#PA+W+g;Gpj<vq7;=3OKv`dYey^kVPXCi^{ebf~dn3NyVV{7N
zi=$AVEhG4v`|E!jeB(A&_M;WF8e`s<@iOtZB%B6$tXL=xd~n>PHo}F_Q!!VGibyhs
zz>%BWjhSlE?&<s<=Vy(U@&xZl!B>!`);_^a&WT|to;|)bbDR{6vX(R>=p@tSoDPfb
zm2(;MN%tZD=i|cpykU-l2GAe^_U=?!Pve)ISGyVYm%pNdhG}L6nbif<g_BMaIr|yc
zZ5@Z`vZ<ktDT2a3Z=YGgJ|uot6Dhd=i7=Na@YtZbOV2GhyM`7oT4ZhG6U`1-?eWX0
zIGL%}y5scOf91m|{OcWzIA;f!&IA^y!$&2KebO_C2uW=Lt$6PD844g(N^D|J`r~|9
zujK1&Cf;}~pOed}j8q(j9$nH<{NE0w_(lE(`7e2sAW-Uz*EVNNumm`?zp1F`*!E9`
z0%{%@y`;9n)(*Vcqj<+L$#|vK7FEc|2E95%plE!Aa-U#4UNO8H2BvwNfI%)u`!z?F
zW>QH(y62@Cl~YPK2*Gq;9&LdaS<b_yfHB1Zq~k5!r`HxUzd#6$)IW$NMUAX}+G>$&
z^1*rDgSo5o1Pm`@dJp^cYXv+=k$(Gc{FHtE4Nv5oy};J|!}mxsh)_;$rLRdf4r5?$
zG)7g>EnPzA%ptX=-2DV)a`hOgEH5MZy(?|E(`9naF?JV7svE!YEw)-*LL>=sOPFc)
z*IDSwo%ukVmDv`_Qazkm2t?Ig;MBE^XTPqbqQdJjw8UCEg*K@J4v3cF5M~>x$2ayQ
z^PC3o+XCv{VpN9X%3GKQrLI282@(VOEu7A;YdqnZxPm_gVl<~5@TM{)`8V5VD^IdT
z^5>^$aKB5XV$m|cH?E;mlRAbvCQmk)L{#YYF?X88>Q=V|$20RK>!uXBr9LxaJ<o2<
zE3ot33;RZg9^I=_{+60q18M$3P*1{^>u?C25kTm-vO>!%y1Fk(feGAq=p1Lj59!Kx
zZCK2KqE^<gbIkqAAh5E(Z|D>q8>Mqz86dd~TR@tnI0tQs#M#LY>z~5(=^N(m6~EA+
zOLW))F{${VP`#2oe;NX0UJa2)6pvbU3jXSv$i&dOaa@kpA72V%+ItDBrIqc!r%In*
z+6TP9qx?trvg>qz=pOQ9e>G$?ZLj>$j~MtZ`NDr=F@Sh%)~gze`B7<G`%LT!1a9ye
z*yu3Pi^BivjN!P~zK1#I+xWAvhr2`mG}ZVeAh^_FN%wzm%J~N@T|Nzw*NOe#C;Z<f
wu$Dn@=l|IPmO}auO&FB^-&TOt|KHXn)7mttlzDam^dE5PYCX}cP`3;FA3?8~YXATM

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_011.png b/0.16/_images/sphx_glr_plot_visualization_utils_011.png
new file mode 100644
index 0000000000000000000000000000000000000000..c5c7acd5a5d51b46559f1f0074fe886082854d72
GIT binary patch
literal 144896
zcmZsCWl$VV*zMx(E+G(Hf-V}|-3cu2upziE?(Q1gB?J%d65J(_;1XDTahJ>ceYftf
z`=e{BX1aR1XP$oKoYPV2s&ZK9<mdnZ082q$S`z?(V}5I4sK{@33gwVU-Y$Y}GJ0-W
zj@E9TW-e9$6*D)dua0hC!RFK+RxYk!M+Y8ues&%<YFjrqCs!d34*UPtf!)!?hGR-x
zYX5BzG$(nGD*%9F_Fsdm5Gw}*-~a#xX$ftwoa1MMgs<c245)tAe6V?u#pM>QG{}Xd
zj&4l2Y+$0e%>3#CjvGea9lgBp8G_dq92&2rr105@+9nbWQ#(aFx{kXmv$U3WGAI)S
zZMbchZhh=j<JmEbHM4eoxYA>7IEo}(_E0^J+{JTnm}egtx*H<hP+#{+$X}U0f!?@2
z+AUwb)}oxbKDIrt`a9G%<%cc&-v?{{#F}C)|MjbNe+0?@_gJi@XYs^_=K6nnzv#F8
z|2vO&?<M58-u<ri54*g_EM0(kEC25}uhZGLR7*!hJ+F^FIMpUgY2AwQ&xg7_{qJA4
z-oIdwt^V(u+Y}ySdhRfu`rbeF{Rp_!kv#MjMF6~<y#9Tqbe{(RQZ=0xw_Uo-DN(#|
zNCE4ga;9x^R2iA7w^GA3(d8a85Ib0Rx#D<?jg2`uFj6l7!WV4AkGC5Im3Es2D6@w*
z5|@4(TNf@t?!<v1!IC;HcYE26rpm}(I2Dh#C&PPP!@Dkh*AKz&Mun*nn&jp4WU!4M
zx%6Ft_C8c2Kyh|s)q%v2gM{q}-R@&mlWjJ#8f;yH6$vU*f&u$32Bpk#ty?rVBJS+$
z+}s&XxSMngh!K097aKHr`kQZ-Y0XrfM1<s~LCeAVPba3KDrr}+agKAj&laf~m^ycM
zc82b>XqG25@f_%U%yolMRv*zqtj@CPBBQaH@OqbU*7AC@(GPolf$coPdft`=)A-HX
zIJ%iLaP2WLKF(8R1YokYKYMxR_qx48w4Yd`WG>~)f;?OT<GmpML!6j^zzN$0_c+fD
z|ClmxA>m|_tHYylo?SK3$Di<<L8l%A(CK_GWrC+KaIo10`n>fXsn3LG>u|<!ohI{s
zi8V9;FR0UzkYTN^05p*-M0ooA>gC-O)RR;8{V7r@5P$;}Nqo{@y!M6!p%gSpYNUF-
zzW52qHN3Z?eC*1PhvbnMvU9W=v7bFj$Y^Y@b$AMh<re<-@CCBqcs525aZa-4iTTNT
zw)H}AtO%pohWX#-d3ldm4Ww*_ee(97eq7Oy*?4+bx_ke0`|2an+`0H>L!*cbDk1GP
zg^;<2ZWO>j2%H7xCgoUBFaL&bND<RPA4vS2O6Akn>Ole38QH?A%DT}hC{JGAx|AIj
zk~>}b>2d>`(-iHvD`7jLxS!wS$!*0^7GK2is&3XT)s8^FPlWJv5>yb7&`@=D(y~zm
zR?Swq%<9!rG^3OgfF_Y3r7qSjei-NWeLQIyxfH(I^2=d_8s0fn51M0^IZJ|4nHnhD
z_|z4pz=J5p1Eg3Mn2n<ajX{gZnTs>Bv^zZ^x`Ab6Gl~N>f{mGG%CfH2gu>1JQg(4{
z<B3qgu?A@!or)uJoq2J)IOnX*E-6qtz=f$)V+pJK`Q{U$Og#uj+ehd6J+;5RhBO>*
z-ACIBt{N-4=vT{zf5?1?C+862mK*0^0HB(%$`#n`9O{m(Xr~Q<`|+`|;TQi#r`&HK
zU0n*shj&X!>IF?DerB-}=~j)p8(mOnKqpSKPuSv3`-Ia}5u1{&g=uVGhbOFGW*=`F
z%P-@+qQeiGGXPXMn<`{dVk%9vB3tkBi75AkC0e-^j*xv;l!!T#p9+1N(vLCo3OCSQ
z&_OB(SYW=*+dx=E#QugIJM^(z5PEiTfgiuI)GKE~Uoyv#TXUqCvL({E=rq+QUI^A-
z<OIy-3!m;PJ>c#Lgi~ngYoyv{X%VWM%;vc6w`0!(!>|f@XMk9hmB}c*88U=uQ!6Lo
zki3BpmZ0!vdaT`j!MT9)>{Y>tZWz}zPmZhJ5)mMdb-ZD6hzN`3g#Sqy-SiILGJbxt
zer-gNIdN6%jOE4tuJ%Ix(CcSOR(uojQ9E)UvkHRKQFQ&$W;#;aqLU)cG{)Nk3YpLW
z8aoQ-5yjJ~ZHMjHi=S*u3wT2I7K*@XX5DiLJutcR9B?;gdyUl6Ezkk>W<y!?XgWG6
zy*n|Rvo(LdFuVRC(YLd$)<~yUE$2CG58u7th=zbDP?AEUktPH%Ta9Y#F7Y4ivXG&d
znR!Hy7*4EJb6i4DPD~Mp7K(H$v<#(rpzR9Z#F~dPUBOe|&v8h*vAhLPvClSa{vfox
zoYC^{%CCr9DyfYfkFNfe>mM7py7-#wKFafvVDbU$v-RQ%#Z(;`>-8J`<<(Sf)4<R*
zfq1<oB*h@d3V4<I0;YZ0n|Rf7wg+R}k)Cp}`3a-LHXEyxQ0@TfmdN|%lYgbpv*tMa
zk5p4hk-xKmA<ZI5i~F#RfD+~rS;_~vGwgLiCV<LX!S~|2l>}%K+JQ(W#&>Y5$@|ix
zy#qKV3;8TteGCK{6yz$M^Q3LHR5eBAY3X_OD^DswW3W5%4yQS3gj13=D83A=p|d-1
zy==#H;RhmiGm5avGiwZAMAtFeaTBFwjs8TFDSS68Jx9}FWOgFSz!q3pGt>N0+1iUr
zrZ}T}lu}Ev&(gYl>-t-G(9n?e`u$UA5gRz3QY_`RsxiWdqU`gf-DT&dcb7RiqZbls
zwFV*5n}>)lYVD}#FHJd)bF2M09=viLZIw6hTPKY^FE1}RoN*?`lYJsHa?(x9u9Q=i
zpk|3R-71K$#GBte?zU>O%%bm<?Q|F`oh)M<Blr68xA&iQ7WA#I;Rd#<_)p#UBHSs3
z4*K7OLv&X@F%z*r8aR?7E2r3MSXL6s1XZ%!J3f~s6f;m%z#g4K043?s)-4)8pU*$C
z?_@*C{5A&kDH?wjG0irY&tyD3`%^wPvtc|KptRJSE%sFtXY$$p`HI}^+9VdBA@0TI
zF(@$oZ4a$w1?z1p>F!(iQJRKeKDegvMY7=2-2;vRqZ+~jB0@|60n8s}8-JxLHA_Bb
zS&{@s!bt#U&D@G~wzwtP7O>eho<xZ~`Ww#;q&r+$898^XabhEE!eNCqFgc6nz8P`c
zBkEDFM?h$dTi)!NgB}#g-v|KJ(BYjbCkEy9oxF&SbBk6|jm=wa<Hy;PB3~Kdz}Vc<
z`$$8yd?eR^oI_{y``fmBP;sPG31%7iU5{c{X^#GXWUA!^efS-I_54Y<@ctnAO{@Wc
z7fv-WT6p+g*|j`X=y)38e2$?5e~5JAasu8B6E->yUw)}|tTxrR7X5X)eY@$Tonp~F
zXc2wP*Y9dRJ2dr{cnWm&j5QCu5C`x5XqVN|F*tZIJ-Yyy??LiNVQ2*+x%&SgeUG!l
zgQ9g7xvuJp^>x{N6g_N8z}QSZJD5X}9|1)t%qO-}U@qXzJy)`zR?A7Uzsq;U_FF>1
zlyW#M1Mge!#L+UW7r!(v(j7lgEuDr#p*Bx<*&Cbl5eZSfZeOjB>%|b2IkVjJUO6gX
z>c6L$M_CG0O!h4OHqR_GeE<}USJv>C4cd`NvV;zs^`3apvW?m@S*W*s%xW%u*(IBQ
zNfk2GwfpGcYnw}}zG_L9braf91A8&Hb74RX`BM)WDhwsCBVx59QnX9e3A^z{w^p0+
zXzZur(xQq*lDT|+Gj}6h+v<DCIO|Nmz_xcru~UMWehH@@7$iSFqo3Rp?oE|0ACWx#
zT-^5V+V<!Ov*8}vbQ3BBQ@~%~$a&TcMKV0iB%D#rEQ|N_Y5Qt%mfuocQmU|McD-we
z<ty{zrGl)L$cQ`iHn2H!e%;8hLJ~)!4B;CW5%^eEMrn#C4M?>5i+rh|GAU=88(377
zG-{`#v-bDOM@RU%<w{INOMs59<g>O85h65$SpUIo>Mh>7`9&OE2fq(^`sja|y<wVF
zeg=t#bL2cFbncWM&^&K4#b+QL-V>CUw9iwZK}Y0@vf}W<VJe+nC_$^pJ-bU@Y-7V{
zXEhUXVLB}9J8-iuM*;*STprT^0+lnJY(XJ9ig?(+n`f54-VAilLVqGi&hILdTmozp
zGj$0Ca#fk4U4GW?Lm>JCx`d1C1jwxLit7gHR-BiVQt-7#6$he}!+0$#AFs+AWq$*0
zmx&q1Pj50e{5Ed?x*P4|P|w$b$+_Eb9()!o#n9ch?;Fmje202!x)f5fGRQ8uQJx}8
zKESzJY2cE6`}!b95yV-IHV?F?s$fFDpkjIVP}bwAKDLV1Bf^y1y<W%2ZC?Z@vYb;C
zJT8GpP+-3jh<B`#+E}`i(w**(W_oYzgnV?^CrenZ66zGtv9){2_;l0K(^uU?!^+*O
zMH+;P=Y`yV-KDO&6uS7u2aU2?>tO;c=GHecqewGP=GaN@zP9SOZv6OndZWFj)o65L
z{%IAI_vUYlEh8c>^5lFw^Z)h32ZFPlI}GcUc{=GS>uOkmCuk4}@`o+#v(cBZ;Nb9|
zmch}VX{KGisM)R%k8xUO2^#=QEQ59u8aOnMwGp+yL`8xhvxw0-6Q`{+J3qhqT{X40
z?c{~r1@Gqa>Llm+jD}4=#fTKvUWvNI5<UFowFW80htj#taD?00Z8X(AZkvc>zq?C8
z9?K8KUfH{^!4zDOT9G<TZ6fp+K3#57PKvnz{N*1+w%s|{!OkAA>gW;ZKo4>X6zyO_
z_1XxDor?!3;CX;e0P0J;c2RpV5h}*ClQzA;I46%VS5{%lv341aw9vowj!I#l+|_QX
zFX>kQ3+V288AL~&;;dmX|2ryuR^O+y%sO^haa;1YLB)Bytk>%MjH&sUP>$IrJF4-|
z1To~nrrN;)K$W-#raG^f$}tUIUS<ED(i^<N@7DBL;!A6&M_kB5OpWeg%{6Afwh9x;
zEj(Ro?icf*WG$!B{>NrrgNM3rq2b0?Rwq?GPsIp^mY}qlg7~ibyTwP>JnO3#Y>!Nu
zKgQ`K5(T|&J6xmwR-J~drcJC5d5wg2AAzXf9G8ht4nhg}m-Dk*r=_Kka+>x#;u{Gg
zk>#yHIy#7qk$*?_{O&MeTex9`dUQ+U^r6IX;>Q09duWx_Uk>h%hx5@G{UMOT7s``q
zHjbG*HBOUAZnc$_&;<cxLu6bI)sQEmN0-{u7-HUO*9&n(+qr&666x;uXk>8Xr5u31
zW*i21DvXM&CPFq639WSc<z}pU>VyZyd=2r5ZdiQ{Mvl8vy18jqGgjWTg}rKD8>#E|
zyGflT_bO(6+v2n5TQPAPvbVfP)X!^=iyB@%yKMj)R94LH33%s(*19&;p~#Lr9bv#{
zoM%%}i9dw^DSIo8z#<(|O(<UcYCp7lz0<7<1=y?K{_~WY&A)p-VOnA6TGe?<eWquZ
zas4vUBA%!gKkrz(wPOiJ*-bZVt^m5Qun}$x_NDMS3>e+;>$?cew5z~S0MnlfQ^SkZ
z^#OgQY45hg7G*oqL|G3Zd*|w_u9atU4u9w$S4dZX!;<EWe+vBlqu}y2v>xkstGm&+
z%fOUeRW3didWnW?QLrpMvXaxsi^NY<CgGR!WGeHfd5pgHJ}HU%dt_J}4pXeyhd*dc
zkNsy8fba2;;Yo#{r~wsrmY^wyzvquyGx`pjB6{rlsiBme;V}v(!3hc`7mv;MCho{3
z4B?$NRyb*>s-VX9%C|B?Of4Ge9cH{UCSHG$w@qf>y_@8GPMuIe2bO}b5;_BnFNsn?
zrWFrt;1VM=gf1+_L_y?8@aD#bPF!R>JGLX<cjJ5`7GyF{teqH~^(}?B16jMRVeQDT
z8<k=;6p8x?6|)t8Z-7ay2bAb2#o|6^O6nw)&&msDDRZ2LrC!xGVUvH|z}aeIV?A>d
zMIoh;?%<(flw|*A2l!D%sKaiZ+V~bEgz`tCw)uHp>qalfcj@6dt~`b8+4q+oG!!Kg
zM<uqK@N=r{|EF>Yw%OSoSRb=oIC!{0OT^)1Ti_F@@AYwGPgu?lbe$w=3YSPCs$fa~
za~$+Z|JqHf)U*w{f&{3nl-1B?E4((7^NCa521{S!J?%4zyNKm!AHm`FUI3%pYL2Ln
zZ!p||WsMx?dxFqS0rZZVqk_z{uacN-koB1ac(#z;ZL1Pn82X0)qe%DTDN3qLJuhAw
zDcHP{-n1rG1j)IJaef$*Xfw0cr^&on)HEc3d3P}Uo-2XiutBnnmqnBmOgS}cxK<k?
zwc)wnPLcCdotzUCK7TMF8KxP36D<E>dzZ)3{ud%WX^`-DQ2LIWP%)Dth?BJXz5VaF
z8~(`h*%}GzMI9N_tJ9fi1T^Xtxub8Yi0ilJ4RS$p8i%|F3_gXB#(>=>cO&O6V@W{c
zEn$JTP%e`y*Rc{^m&4@pS?<T|@!eoSwBVY`=ywQ%2r6C_g89u!&J_#&%w$W+F%Nj`
zz6<Lm3Q3BJW;ZgR2-|JW<J`TUTtwm&QSki5jl^}UP7X`Pc7Jb^w&gw?|GO4K3z6ro
zX*>xsoISo_DE<aHuvzDCQ^9>WY-g9YRytH#1I~Pyz%f__W(AE}<%Ni6gn;&BIFs@?
zIPiL+2mXF@Uu7;Waeke8+-%hneu<Q^-^fNcv->FOdxvJf@}Kd^<v~X-C{<R9*jW4N
z*3K{AfpvE{5qUn>r8ZQKG;S`$(em<Oy`r|3yY6>jBV->1e{8ZR*hFi)5{WSRY8F;)
z&2Yp-WsEgg$i?^%L<OHVV8e{{BhdZ#2=@R<i|{9i1V@I_f;>pcQGc0)y^DJ8a^NT<
zUahG2u^Xo@=y<8wo*899qx&%TAm7$5&j<lr5$;K)NQ*whvDq%S_t;*8n2Lj76s$<I
z<5it_U4L|>bfS`_t+iA!mK^-?hy|4xfP?MqYqS7$H!80D&db6VqKO6FhnquFFnXN3
z5XNb~qn%_tQF7?*hDlYW1L4MRmsiY8gu@neBOp{}C-6{P1b&r}Dy9%!mUX8U<KWM`
z?~b`G$i1uPWBb*+4Rv(9ZYO<!<sKBM!@<$G^cV8XyM<eD)Acg=O~t~Uft$ux9~A`{
z3|aE<;Pqsb5EtEtgyRH$(|14bi9pn{iB(+7N=^_t{?_UWLE(NLF}WD9&aqu|{)*zo
z803GlJ0i(K*>jc-=rGS^Wa{bdNb7{C#4fpf7c4zZ9XjJ@EtkTQLh(}C3mKisS*NKb
zSN*swn(~u)4C1AL*z&<tqzux3?a|knn91n#;wvhZfy9Pm`>6bBi+TL`-B6*Nr^V0M
zVWTB?!c&S9**RBpQ3*qXUWlp|-cMy#TRg9Ff<`_kD(C65$6Q+t->tr3;hs|usSWol
zVaiu`(fklvrM-ubl7}i;3vbXw5(!ZOrZRkb53(|bYLdAq*xD~w+PEC1=9qiM+om8L
zgKyS?GTyV$$yn!08gA<04GS^KN8cPKI1B-NE5jSWb-Zvl3dtmS#IHU(SFzL`KM;|+
z@O8UfbnntQPni6><{fA>9o;1sNioHY1GAkEi@PlVSD#;9g>F3VzTaGN>>0d&l4ke&
zywStJujZgXI(a!IWkU|iLsBuVJt%{T%q^fDwV9C$cueE7D+s3VI<VuzWueYf>%m>5
zyT8k!Ppb7g-AzOrVuF?a@(%iB=|IoyJ9y&V$9ZMec`y4zeap3N;Dm=(hND=4wKW};
ziEEtik`|sW^T(CVZjc!r7RTgV#V<CCAV%`?jB^MMmJAsRTSXx0I3sy+V;PtadXPl0
z+jPUA!oq=Wo8l}w@4?f3Q<yB9vMr~76sK@q%7AfJI&#5ros`94w1hN)B5&S3W<Ldb
zcwYB?GaHGewQoUvNaf<>fCqXk4gZ!JjBJR4DHTe}n4w5J;Z4*ZXDRjVN@C?N#W%2S
z*btMT{u^!?Ve$mO;rA*67FKzK<BDvYJPVJZM$14IOov*U!seQzMD^_W$63Ru0;j1h
zj}9(A1%cE>WP4V)(&1EQaKK~zwMdw^?V_s8kwSf{n@Fy!0y(MHksbz2{>Q#UY#TGt
zM8$j*0aRoscxAMG{z6(Q2}lm^V_Si{Kj?J4G5SEAZq;`kww<8~M-K)%i=Ew0iZw@v
zE|Sj!w{~}}iiJ~?`CCK6)|c1mN+`9}Kk!qX>hkrA{?SC#*3qFs*>4inyrj9t9D~vQ
z;T~_$McVCt4dP0ot@Gt@v*J0;)=Z9N`_92<t4XmJwkD9;PNLf2++KqLzvlD)47u;K
z=GK$^ei%of79H+yKRoUD-|iYK&*$VLUNQ(H<M=}3g+DRzDQ<ZyJ8Wz4N8}d+t*fB1
z4Q%}zB8&LNDUuGSE7b!HRmT-)6-VI$@r}qxa2v3wVx@@2RE)A9{64sXk(ju2mrfL>
zUkg4Muy;VpO}6lwf!%wap+c0Lu+OL&DuHxfSp9?b1nPaA=0`MGOjux+(cPzI(1rM>
zFFuYNx~nCK_(4gf<LoE=Urb@y{8-Q;@9&GBYy$)+sghkazkI)UB_Rb#y%?FNP^}gm
z&_s`pXb=%`7=F=w*VL#6Vv4S%ew1OA5di6kOq73u-(GT;*|^rrNaNL`u;|9Or{@*i
zZL0oT$T@9+%i^Za7A-7gz)u<ZZR`>e^DowKgBUPn(Qq<B=#O@EDkZpoVQ9T5X`+8q
z(F5n6c6X5gS^OAAJZ{gdUOoMvL@niUqC-A2cvJ%IHIZw!W#dT=+DUHiQoTF|W3tGw
zQpg%lG65#`{g6@xt-b{T?(2P1L#|0$%7m^8MIfUhd_>VV9<-@gc%=!NOfqbn0~EzF
z`VQpGLhAAM<-PmT>Bd~Q^ydPveyCc$Q^%T@J?2;?n!ci*D=1<iZhB;UXf!eO^9ZGG
z)J-OLJQJN)(dE}er8>^sjc(Xzi3&F-{2BWe+*T%%XY`|T9p`;^&C$$;5cUV_!7XV>
zh~fcf`MMm<YwxhjpBMBn(&m=cu6qI!ZN!+s-G*7|m<N1+n&(eyWG6cUy@D8-Rl?)5
zPUJ&U3nLplq=X>RFfNu-^(7rsV|m9vSC7V<NtsJXm-G54SOwLBy}uz$YV}N4L~Zpq
z`St8P4=Y%gK%^iW7yaWRHhv0c9;m470}H?y;$@5b0}cZh3D+@IUKxAeQa(*6vEHdA
zI|RXN@XT)J?zyn#A_+OwTQn+lx}eG~XG)EeA;G~n{qV4W1BAf++j0LfzEKysb!8sS
zN(W(c+>Dm#HY9r6hg|l`D8k)%;Bl<7|3KP}{pzKna9ylmqC41~YC!zl=Sn$nmh7++
zJJeNzR?8qp`Yq=-XiNomvy8hoP+dHEZTP&;<M0jFRLA}{jMj3*>-7^1BdAFmMS8a<
zBueF2sG=n<pX#v~UKL(h>Gb7oQZdfp$EK&s3eGQ%_?j&0KU-BWCyis0`KB+C#qhsM
z^<HSRdR3$G^of$Sfhh<lKv(jas|hL<GC4m3FO+}le=YLwDajgx(f550Mc=Z}ktewD
z*{=LEP*$s*o8kX!!Y8yzggdGqmQsiOkl~~*`1$!AKws@0kd{7r&O>y=>F5N9J7^a}
zi0e{3UpHL_h!S6B&F}s_odK==QrL0AF6&3JU_!8B)Jea7*0(?3ad9{O-kYby(_eMY
z-#m5?r=_^YanYdEi!^yk#6ITFyW)pAx5O-$1+FiQ)*aX^wUl%-0e%>rxj?W5s&a<Q
z%gg0)er{-e(gG3*Fk|m_hG>aT321lk3@mdD8*g?ku8#zZANwY5_Jl+zN@3GYO24Rz
z&uQ=a(c$`Xb4ecH?>M2R5_G$J%e*^G1QRxfN+{1GGU32JHhxyF(*;DfN=3&}qM317
zATgzSl{F_#-euLTI2`*Cf0Ehcr(rc-$fyM^J?4Ds^+AAPZt^}K6KM>3qwK!3;M7Rl
za4d+^J>KW>WN?0HZz7i3usHtITf>RY#2rH5!&=-dRg{td7y9GtMtR9B>X3q=iA-eU
z`i{`S1fnax2J7kLiz;el?@GXpoh!gJq)iaXt6F_g8k)?AS&A=}aa!GFN|KJo^}2hV
zg$B@~{89@-Aq{-gE)edi>2NtaewMjAqhv4{1#IPD^!>chkBq21;vDw;fm^aa0PeJN
z<hkt(;fmH|!GFAvS3j9o1|ViU&HC`YPhndf>+c0o@Kr0<s5&x`pxwngu)>%zgU`6?
zhLIxj`sAqZ*JIPN4f!<YfjMi{AOUzp(D3W5>GmJQUzFv{UUQX|{T-ewD<?@agp|zK
zgc21|p6bjf)$bZ_)T4LG^ons$nAhTbw5~9sPmPpREXL*-UFl1QMN02Y%{E8>ZBh}$
zerNF$IELh*#IZd+pI1oca}SZ~P_+#mwp+&@yHA@wRJkZbm22!7|H0Bs7Qm}I@F^_~
z7L^Wit@{ysa&Un5mz|gswKBI3US3^McwzSf9$HZl>6O0=GUDct4^Yk1GA4ne*ZX-J
zDwVsOjv-f}BAf|4*%9h2?Lrf1Ngp|`_O7ad;9_jd;GF#k(FI1~e{Ew#eM6;!AI#+Y
zuLt+&?Ul=0M&M>iao2u>$+7SfG}u3{<RwAzUFwKrSI(skEvOijrErFYC#Txfzfl7w
zFwyUdqqhBXnn*jSEH-`uG&n)=H{WGkqm2uk&agvkJ*0&z^ya)Ax6*b~N$O<#brd(v
z3QT(;aZ^%)YMD=C65zcP_<~e3xU#KInX*7-X429RyTh|8qpu;Cb>qRO-h-SsXZmom
zw6!b?<vfEc5<`@{|4ifQ|Kzgu{=sQotI=DcH%$H~W1krQi&Efx;3Wy+11VW11UL)<
z6$<8?M&|$vAJbM)1)Y03-1~r6`;NN3AW|eIVt5tTU`+-+xT3)k@1#9tw*0J^FY>Jt
zV<UKT#$((MQXcv{>kA05TnM>Sm!dFXA1<#J6I7<}o7Y0g?Q{po1}4=kzFE5+)3iH-
z?o}hcLtBGH=iT6g@!{H<My52d5x?a~gz2NEkJ%J6SlXJzF?GpsU~b6y*Ee}PKu$uF
zisoBh1#d9Hh!s4)%n4oCnR$Jb?Ta)D7-*O{Q!X`+mC(_l3!H}vCHF<4@u*l*p{^7Y
zBLlcDyE(-?S>l&{5#Y&#D!59K3qPM354fIw^VC>Lx&A8(56L@IH%OXB>@0C~dv`)s
zQ`x}JaawK=Hl0w6vKIwsQlm&qBDErFjr&I}?fBE|$PJgAy-#~g5=QZIrcM!95c;7*
z0$#SIN(a%A3&qzVSB5HjXQC5D&@kmst#*oVsepDR9+j}8_t0^>OP$^OP`8;gd|@Y8
z+?t*C1sTCk&;V!X(%Uu>WNycMjCEDZryPFb)vs*iM`O03uOxNjAtjT}nGkg58JO=4
z%@psX;Kr-vg&UvG5$bE417>ktBLv<@H#c|19w_yzX8B@sam1cYp$M54601Tbmy1Q0
z+$A;$5CZqSy{t60ef=ekTG$Lt)QDW<{9G*yKXk;|+KGUbH38z4L$*j&s@o=dK4x_r
zQ7MSC|56eNI(N<liTdCy6^-iIcv2b4l)eXlC}cG0(4b%YU?sQ))jb_PUnWNlNV8J;
z#$J_vbpL(kb=K$BhdvGue^rOcpKjaekkZ_y8(meFoG4RbM_|*V1JPUoU`x(dyui1;
z_Gqc%X>H*xh+`y|XZ>*txjYeiD{r7&FLK~$mC_7*<V2o5yZK7PJtbgr)p=&@c2eu$
z*k4lBqz)K<UG>?0OIK%boPHn)SJ2^E;4=TT62d$9RU%ey>2dK1k8Q47R=iX&<l<Hi
z3^Wnq1eT%ssp0*?2X(6@zGTS?pBp(9V(77x1_5A?(>(*FRGF*TXY0tp{9K-mnpO20
zSl!ALmEJ|lt+Rue3O)Po@02SeYP`sqvKN<n2P|V{K|@vEkUS3F!2rfQwezAc)hXgb
z)e%QZJ|E%)lc5Ac8N$<iC{MNlWz+2I!e|Z3f3ftV@Zb}-TUKn;kiHK9v04p{Nn^`w
zk`K8WbbF*DQl{29--yh}3?p`;*)!u}B|&dLBi&U}970UB#Vt1>B(v|yFatOlF)wSV
z;nMCY4^>l3rq7Qs&%6Wn#SR(Da5&nGwxSr2^KQmRo<6(D$C2Vx^z@~X<9O}g`!!U2
zfPCo=7+7!arKHq?sNoTkw<XwM2#za@TIiSCSr((Ht6sBKR>lHn@)*4ERg8Ante`$3
z+J>W4h*tzHf&i77XOY*AL?T{@I$<2sS+$z2JxAL3j8|jiLNKm)zXByb*$5U|E}#WR
z&@NB&<*A6Sui0l!NWLvgs!O9ttDGHf;H{N}=Mg9NENvPT^3M%Dp<TpD?%8|#52lY*
zQ!?coo#0X%b%pff>bKY^g-_E2H5shN>>LufN<jZZLh7&Maw!JvADVt|L!%zDp@LPq
zI_Yo94Tqm5QtwA|%AZH_vWM{a_pC_ZTDr7hxqt&Z;8&dL{p57tp<1xxT*>DSbN6jD
zxEuqxp)xv<XA@tY=pLKizZ+i_@zF{5(D!JZ8Clx3V6!<?$~ZGz@hY^I>F+-FDG0F%
zD%xRx^dBBs5?{zr#TJTJn7az;yT@!58#uU&TB=v}YGM~=Tet1DdE7d6iBMeO1OIiq
zf*HI7_@0G@zg{@>RR%N8X|1~A$}0BRJuJW9T%ZF8yxMP8T(E_3Ast2oESbrD`1vCz
z^ubDCY3Y~`$`@!5<9?p+0p7EeMwkt?;r`+Qv{~jjfaQWO`h+?%z5X8}Z%G&YKQ2Lr
zPA-BWlm(EJ@rI!{s#cf-;2~6k!@L|p*qN64!RB3=fA5J);fF#*6X&eY@1V=WQ`(dh
z;Z{1#M2E0AL)_m%l-%X>1VrMfE)I718%A@ua6%DvI41|eN*Hn@k3hSlIXiAcFEs`C
zsZUo@3FG6tP7ZItVdz2w8;v7&WF{>6g1Vmd`?#A4rWp^~C|xN5TZ!|9zM^$Yz(#Sp
zOZv~pV2h$L^~Q^)?zyEw9sv%FS%?=#8ilsKruJ6BNm#jU_F6SY9tlC^Wiw%6wyBt$
z-rpZ(#0Fm^oa)77`ue=a=I4#;lJ1D_X-?7`naPBc_3lFw{@KuCwGIfY{7OPBoM}^w
zzj;)3qZF@Hx%I>8kmacbFMpc4U%HW6s2Ij&W6k5E@mN8CfmbYI-~Nkx_Ud#_aWDPN
zk<!O-S@`+J>vRZAgRYRPkCv3-)ERR_bQ*`6_1_zFBbW=Q-qOLm=$xPMW4mEhMQ&pP
zc<1x&)oe1=5xK1kF7E&~eQ|-~4Vx})riyiY98E1?&7-WqX;cKI%gcXFsT&8j-18Cm
zL8-`riW7<7An=k@nU)8#X)t_(-B4KsanPP^mN!q?l7rH{h>jnP>wtcH*m$M4?ViY&
zem~J^S5BO<22ATb)QX<U0`xq78A0#p0dnA_jKW&y2#O+#OgTI=+WqGOfO;vQcw2Tw
z{)e<w31JjPnR}o4N6Ns;?;4952<b39j$+X{gyPg?ylI(2{B3Xi{Y|BeP-%y9Z3%q9
z!$@{UGhzOPs7nkAO{O`|aquw74Ln!}eUZI&ok>~hXn#Rr+(2Q|#;t_%HUr&abV^-_
zTX1!&4CUQPpX2!qG7FW3lTbdBY0A<`g)B9AI2tjHQG)jAZh%vG1XI8vfSzWzD`J+R
zGo71YBt+5jJ}HC#Id5A|bC1POtzcbbpYep&@1w2gC1xoTWRY}o)AoMwwmhFk;=6-U
zVHLvXm`WyurAND9_7g@GXZ6_Pe~|baY3sbLXn_)nm;!IV;`hAthuSV$KTOq}gX{4v
zaD!;JoW>(JZ<B>w3n0<VyqI1PNT0BpSpuQedE%PJ&$`f!(-3X((J-rO4C9i{Nb<jS
z44I80?zOi`f;)EbG4sun*rtoB)w9i$Mt$)<JMJ68=PGMV!o7myR60VRWHds>w@6Fb
zTUWan=o4eUSv4qINq||%pvlo5Qi<oEN6Md>WqKd4y~B(?%g_Ck?u@x%4(NalAXL@Q
z-lj>hnjbFdst5+Qz(XAd+b7PZX3MUrq{O}^69xHjcND9%w>)aW3-x!{S}2b6G1%@E
zl+KBLoDs)gL7C@~4Ss){^u_Y1-JBA+)jjAZ@l=hve&KftLdgAD*Q0;x?YCjLwQKa2
znvvLWK6D)wYP)os$R^mGU*zdl$BX4@9Zz<vMr&EPMwTLqGZAqgCSbZPqeSLBThHs<
zhB|mZ@4N<UC|H}oI8)8eTrQBVnrhewE;b+XURhgECmyUi)E(l<nzOS|+AN_#MDg(=
z3BsU)upfc<e2qMv2d)ijV4rO?AdZ(~&}R|O0Pm?sQuS9RX3#HRl0iVfsMjs&sjg8f
zq&GqG7dLJ8v&}Z7*kxhmPKk-l2snBu1KNV<|G=J8Z_K&F_K(I<CvTx)CF8vz`q2TD
zbTo13xl3E!ocMGa$*;Vuxi96gPwh`&70`IMp%mveNyP@=$>eYIbV#_-C}s9g)Y`tr
zS;^@$t&jh!)5V(*iUCPffrve`Qb70#ftTOj1w$@aBtlq?bwoY4(ak=7i1}_77*}_s
zC^`0N!9AptsV!lt7NHMs(W;E?(Tbs32EhvCG!}}uAd|F%^K1HL5tAV@SeopYzx<qd
zec~KCTOHvn`XyF5V$%h2B4xL!>KmBSFH;;uEl^xg;uK9a&P_Y*N^0>&Ed8hWY8F`&
z2?0Wl04Y{^ohU=jl1oMh+*icw2C42b?bU^o7L)F;S&KrbX^<?15JdP62@UCLXj1xA
z52m%h=}Br`QY>)mP3yp{pU@c$-YB|e8^@oZSNv)n8_TPFWYl~6t7?&aW*xVAWb0LX
z8b*H8GDm&_j!A-?4h&0#`Olv#2v!D4L97W?whQYzQs*(;#S(6rMSBL&hl&c7=R+G<
zxG>)hIg4v(J5h>c;}82)ldHJ3?J%f+oaSM7O{1?_<*>lfgjY0XB+~iA2Me3m&|+zz
zk0fKwTFaC3>*>2~>0JHbw;`1SV3?OD!!FTO`m%k}R3UPDGlDR9fwtMoP%0eD-Pien
zK*=k8WOcr(g*o4pDmqx?Hj9Ih8a>U`BSh}YUL!Gq$oRxw=8au*K%P6OqM*Nb0^heU
zXIjk{G&<4gpa_{J(t*b_0A9j(XI@?Gb7SGn<wasC806e@?`64BqMtXZj9DdRk|-@e
zAs!VFmQvH?hBKFuDAt(h{<0<vW=G)O{mDV@?MEW@i%Z&IktnZ>+tF2#Ly|wA&*2h8
z#<{h7exMH#1`NP`iMQ3<Lh|~}*&+2ro?ELt);*y^PK&=VrpssLjeu$(Dda4fn=I7!
zUZpBl!HKUBp_#2B_#o<0Ah|K&%Eo1M!*mT_e6h@|zd`+ke=vy(!_!9O_^@t*b;wL|
zqtr6kw?n~EahOg`Mx$FlXj0@|6L0>uE?QLKRaA9gFfzg!m4gu%wt<|SVhbzEnO?T+
zMW*4@lZ#A1hnYL^q6}p{8>?_xm!QL23314U&K1aitoCYUHvCid8qRS^@4fo>I1H**
z5;r-!Z{8atJo{UeGM!8AF=g>lr(4g_FDYM`t^`?Ax%JYC-sDc%R*l12n|>Sd?3$(q
zG2-a)V;c%{*UjitU1dG5e@~p4s<C2c1@hHsOm6JnQh?N3s_}*f(2${YR&4l$IuhBr
zyl`~vnplo>difJHSVo2;{u_8D9X-+yQ1LUp)mL>zylW}&bb&`6(I=h-_Y=vO6}Z~W
zFHrsTF?q#UV_kbA#hSXz>5N53CpN7x{&Y{sZ>Vwg;qsEl|M|#dr##=^QcL029kHoP
zUL9ktNREr1NEkI!rjnYz<iF+hGO1%|cqp4+`cLc~$#Y|K*^zd8oYNgtFwulagmW}6
za*XisSm0F@0xbe&Z$EeL?ej~rDd^k@q6TmoZ*!Y!YYps>WoTq$A4Jw<K{Y{@b@l7L
zF^ejbyCY6?>Etk*M5uO3*O2xoqMzS&k|sIfRAF^!!N|+chqEP3DIAIE^<kHEelP78
znjd|gIF;SRSHf*SR@LLeTrPI&moUDI|1W5ki}wr2{CGR#zx=5k|8G}oSWHo%1#dgs
zn0OOQh_cQ|u>yWZXDF{cfZ9yoGOtu`vRgrnnKw3B<_+T4L20@e`K<c-u6NdII}&_R
zr3W>=fY<Q}9*dR-xx3-<P)cv0vAu7{UQaq&{6?wcD!C>q0cJiWgPU9c;-?ue3&ih7
zQ03?6pRc?R2r+p+<fvyS04h^o>gqe+bcbEAZn{(%BlKsX;+(vT{lLSaMKwd58Nc;L
z_<WEk3&GWh2JTI;_Cl(HW@c!{v0^_*QBU7+-*z+e@i+BxtB`iUElyFhO%$?W04LFg
zq!alJ6>G0PR^<-NAYC@GjaXKVW$}I@&gr<0`D3%n>Z;b;q13sQtpDK}w?{SxKJwka
z{{)5QPJhNbd*2jpZ;qZP;a`l3n_kQ4>}B-o%i48t1{8{mRMRN22K<y)vlLb~@x$n|
z+Y-f|8J)M2G`VS%7fLh3#q8>+XJ27x5efjEL1>}@ew=&LZiOOj0p1=^Dty>3nNGws
z1Q!@SdhzM<boO7QsR9OYq@*lq7yMqXw;9-%P(|PG^PPOP@%C$E!Vs;bPUt(G_cXVT
z^9x^BE;q)Z8=2fm^rn;%o7j9qh-IPnXPef9fAU2$8JWf32;{SiC$gRw=qu&gwe0RG
z0a8k`9)y+=?`hG;t2Ppw!8aPG5Cxq}Uk|}eFIcd7Z~%Zv^|6|_!LV{PRU$?)|L9zT
z#`jHT7Q6E%6av{@e|-UFAGUpGIJe~(IaynC5DzW~DkPSsf`twl;pc?}T#J<Qfm>N^
zb;cOmBHj)#N|_)}zk=CWHpD}Q&&_DP%}lNi+0x2cr|Iz^(P9>*MmVUi2%Olv{G35`
z;x(Ud$KA&3FI{qCtTHro92}`3-_mQ9+M3qq*7w#77@2m>xWkIr)YGw2x%ApnXGBYP
zW<%iWU7Yl_#fv;l5I)>Kmwe4`Kx^WvSRsahgc}P>OC_C1AE=KjTFZASy22L|9{<GR
z^auY5li|iEa37Yv_s(Zd{OkU1&-vHP8fP5se~jw!;n~4G<Z??DkpKlwwji4?#@f;-
zUD0+L;Tck7?8Rp^TD1gs<p+VF-6yP5kmkA?8vcs!wtN{0ry9mvi^I;|y(8P+TXvD|
z;@iqfn*~$OZU{vmZ9xg;RAbcxs%r(TJ8B3W5>tg!q6K~KOkZOkZEn{sC}lDEFN9(?
zzh{%{Xd&nixadvOe&pgEAA+EFLNCbaH+W!e3BSC-pa)u3lt`=D!J=9HV*K^X+@%XD
zij{b^);e$1QM|1@{A0gKLB*A~-3(+WeIQdgG1J9KwM6-pnE0T1x$cO%f0MJ@&-$Dm
zktp1{aLPW;-L80UwT0j35`F*S(|S19pi{I6d)e)~zK&G5g@_^qLjF|txeUjYR7npI
z@r1`vqw3vWhofogkeZm^>XRR)(Vszl$Sk4OEsup{q$GNBXVyfFZwAG!OIt5#h{@A7
z3GOrjS}@_KXbSHhWN^NVK!rM!p!ry(qz-S+hbwou>@z|t*A7h@45T?j7S-jcF;r?4
z6{Oq8;4xjMs>|S)MYDZx7i^k_z9Gwd3jz+mkEj*Szs9eFlLSWyBQXqbKAZmiwPQr%
zT3^HJT_*KG7@tO<sd6lC>3){Hq$gj1&^$-hR$g&R<WlnCr5J<H9Ub+I9VU{iO0)fJ
z#}4)L(fV}{oFpwLo9{FQefA)r4LyR1`Mzcej{sFD@|dWR1N|q<QnRUTEsp!(Dgwa(
zn)jWsihwZ@x~kq!(Y%F~d|<0TR~V?_T5}u1YOUA&6G!la%W@iogowl6Ipi*Xz9-0(
z#K>?cpHxAq?caFn+K_s5Tc=BhN9_x+Je5h|0O5HVb8ts!{+dQ1NcT~(!3O9c-yaO9
z88oyC%dG)(v}5AXp2$@#{$i<}TVNc2LCCfOrBfL0R}2akkDR>e4UoW0h~)}8zX>O&
zv(j#F^?SQj<_Y_;0Gv?YT<vWgnraIUKs?Y7E#{nBQ-1P3OGe|X@Y1^A)ORHiyVuQM
zMbh%9ZzSpT(m@$Iag`2C90^Sosm20LpG%laD@vZO6}PyaTC@W=&;U9HGyXj`^R<2l
z)r}=jeB`h8C-x<t@$lnA;9m{NKE@@EmGP(V8J@dq>`(<0d(u=Avmhq46^LTNw}zub
z1?I@=g~a}Dc4aMB!qcR~ct>B*fQ{H74bFSRUkS|=pFUFK@=J-iy~Tq(6*}g2y(L~k
zxsG!Fy;<SWI6+JiIsTP`zkZi521Cx~gE(y(ME}AOA;+O(!l8WXVE3a6+z?&#MLbm2
zFcT}&l{%b$vNwIHNpYo}@9Wy_$R)ha?@c#<Nf5t;X*Bc%6_y0ULpX?h8P+;*h<b-m
z$87Xs`@H+N94x&-5S9BRXhjMfQ6|3R$I1VA11*e%yYkrKzAedBYJ~$6vrO#TSy#n=
zyr`%>TTjC17SS-0EoJtRTE}Q=(<7qZ#oG!hNSd6NkH<<YWMhq1pLX>Y!(i-N{I!}w
zZ2+DUyh>*OSE2rWk^K6*%$f@xx_5!k>x?QM9wm#_@zwe;`cSrhh>3;N*OhGG_nIT7
zi)BGJ!;0jv6QjC6cz+nOwKPBfGo!bnH(I4FO!iMG{cirxrC#Za|AOxFToL;EBW@>|
zMUgSdx2siAK{g=6;CCiol&3Is?$)LK_{M2)7GGLyS9ya9Y`tqo25Rdxh5b1?Cg6+A
z!gu{XFrrZ6*={7eg)`NnkDi>#)MJTZs~59E5etBWgX8Y%TnqTI)%Lqr+YOy_f_|%W
zt?tA;DYgzsmy17v|9A0St2S!5B(ZYhevTCAa}f6~ugcfg%R{!B77B;wxXR}*L@vlj
zWrY<kUpG80hZG4WKa`?ZetS|22$Pb_q9a2Yvw#)i2X{3Zm&7U|YCojK>*T;Aw#{^d
z-;Pfplm~8z{J5eysv9osyoKvgVhw&Y3NTq&pO|1aIco1+9&unox!xbxHKY3{enzXo
zq4i~pLuP19KvQ}A?+Bp1cETB$t{tdA_&oNVWBq4DeXO+KflrH=kC)nn@*?}bP`@-2
ze^lt+{ldD<G7M2$hykt~>3vFwZXJaYBH>!1P{qG7+jz4g$&x(&u&=oEfK)F)p{Yon
z{`7JxW89b=7rlAEPXZ)My#?!K5B5qm1=}ceFwnL~`045*%<W@|**D9jDCim?%r79)
zH1Y#~kzh1-ChGVzPHMR<ky=i%Kw(EEeYuu=N21a{4W;3gTaosy@ZU8w_T_KF{QRjv
zzB*w?llHqnc?JAvW`P724kyvvg3II0zwyHikBu;&sqy0ID53G09DVq89KbQ5+?$Rr
z7)f;i6#+@0filaL4Yz#|2s4XX$WhMJN(j8T>H`1w^(;?IBLYCgRUR^h75H3h#0FOE
zY0coeQsMn#;hv?1{ZydpP!W+&DEZhIH3X>y?3m6x%>gg92%wr`FHGf6F@wHz>n<U@
z)#-t>pbb|)8CT+2pKJn`$xb-A+=Llgu$E#}=ihD8Hx{g-L8*CO?w_Kw;d*u`eB?sB
z%7<D3qi{!)YQn6tq4%$|@3$Ud0jO84VEgu^J*HgO{`l}oD%-FJTF}w3A5ERH;$IK6
zh{ORJ!`(OJTz1I6vjk-LZ8UjZcj-67>gI@$z=e}S$@OQitj#sQE4RbmHb#<zT^F;D
zCUBJS3r-;ZJ{ql{MB|!+qwU<%IfLZn6i0Y*EC2q>cgH&SxzZIVUr0+Od-rI)Pg_1{
zITC1M5`SU3``yP*?BeO80P=H6iM~&+%waa@&){p%$+H5D9yK&j2S<!zES#V*Xuxjw
z7Wh06&(@1zj(X{M{UB_Z`l5}c3`8smPAlEHVf+u=4$y30sP#9p&}P|}nP$d)wqk)&
zK)AOib=`G$q^Er@<5dgm;-*!}Q&YPGlR+L*5`!IL+i1a0Oa^{VNVqUhl~qt9mOw8n
zDIn|eeb=p172Mq9@i!lR6wbwz652wWKv?%H^2Y0#k^0tn13(gw3yA|?wTzWFMA%GK
z3Kg#)S$oO==pNGiJ&7WY4p$WQ;*N~_)@@Ugy3Tb*upUlZC+_9|GsT80*+hn>+Np;F
zuD6;s%mee?AO{>%3^{CiBlKGrEx$h{vBghn4)6vJCEyFqf`I#cfrFYjK2B3TncgfG
zDF=!5w{o<#6QoPRT%!M6ZhP%i1HZ^F)HM2&B4xT0LbU7nY|e7;D7zyTxf*;Q0QGaC
zo-BSH8nr+aouuE!dbEs!kA*&F9r6|+e$2y^FEs-Gi^-(kL`JGTkKe{@oD9mi1*tg_
z*)%*Y@$N4p6Q9D;rZ*|yTkxi2woM_PMS7v_V%8s3v5%*IUV_AzA8^J^gzOSUS6SkW
z=FR(r+<o0sh+%f?FtZ$nPeamD_RNVm1WKEIt!Kv&q~0{GlHEJ#51cAc!LGKbv3m0I
z#>JTo9s!JrQ;d=0hGQrYmqoh7y`7H>=96peeN0TLsHQX$NER06p0ci%(MT1@&r|<L
zKUY;AXc9V5?_R9*0piweTNLI%K`4{v#U>TuwGg?s86JN*zDXAGMNJvY?`i$YeG*<r
z0;$v9h}f1GS7tKjSfLee@{chA0-*_|IwgYF5$6pdRo%nLk0uTr&(IAtV>*&SAZDxa
zTXvV%=u;h)TE~)dDBvO{J-K>PMOhzPb4-Q_=3tm5CeU)<=pHguUCHKTF6M1}<R!_z
zVx=S)!;^JNV1&T<_sDSRPig1s7yMNM7){%3w%PCW*jYbTIwicbH~G0(*mJE1C0d8~
zi2If<h_CKx@u6Uu_DiIB3`YGcU($6w8G8&nL;tlR`>GYlv>FTVZe@il{V1K>Z4vF1
z+E@=qp;DE#kBRG`V5`&XR-PxbjM?|p8e2ZgzIY|2Es}G9Sa0n#2;0ar>})&c{<0MH
zpM6@^H^l68*(J&!I_b3$b(AMRqmJ%B9?xYIk(SS+G4O09-8?$~<Okgm1o{pF<0s&(
z{-%1H@BQ{d6C<~f_w}HXT0)67htg669ak^u3mI4QmI=<4z5z&T`J(kG^R6REKOIk}
z)@MQZD0=$za>^$Q!v1bG8_RD}S%u9^l<&nqn(91cthsG*1k_2zg%W`chRQX5WEC7(
zt3C<8HFMkDb9W{O-N=}4@BKAzza_h3&kNMqjKWWpYNxxS()sz?j+7uVeRJH}&SS<9
z2QAuW^EAZHOZiqR$mCe61$<JW)lH<q4KT{UJ<G5H%}CD2siv7)N@FreNH7mrRsQ>{
zg*6<Zj>K_puUcU)uY`vs(`^F^cdySj?6<ejA?I!`$*L<?q)qOp@j`P>YvJ5WyCdVP
zx5~?jxr;UQR@8VWc&MZpH0!0wc5JJXVI?t~v;t5u(MUt2&aeu8cN*RlAOW7M_{5@R
zox5<UB~h%paqFav#=wh3J1jNWn9*)6*;%W64bRjWL8*d=ktzfsD)&i}0ru)gyh+fV
z!L?UBnl>3;$n)q!nhS4I9juW4E#I#Cx1$?YpeaP7WVL_6Gn)GLTL@p@>emp(K;5iH
zrjsTLC(80Vu4+$p^@_H}`wU5}BII&THJ<RGPP+9^dL7gpq3#k)z6sfZzt5;Vnhg6{
z%NQbA*vrZ-mi+axjfpv&%xE=^BH-d(k>f@sbK55;_XJ|aox*d&hZpglkBg8Qvs7`l
zpHQOfTXTclT1H0*?}es6KUBRjARZm&ImqIWIreB_YZ7rx8R8PmJxQ5$y0+?!us<Ij
zyP@~)K@!xQ90Kp}YUq6XVMBDT{vB&hI|4MNgbl(<fPYK3J0AY4dz)jAS!FbLrWxqT
z-2t^m%uH;oY3s?2B(R{$N|?ak1jN4Q_Iq65kqCXeUFrGS+lg*|taSIFCf3+d0Y=wd
z>OF4TJ3Q`4!a@9UsK}tysPUQ!X2L{Q%YR&YHI@vQ0X!lvO2ViLdszgggZ#iP<l^R+
zvm8!?NgyOouqVdWfO2IIKOi+2azL{j{C@ywL6^Rr6q(%hH;l^~0#lXNsOuJ4MrDYO
zTEInuvf!~y6S}^GgGXj)8k?pezVbW^8p9r}_1*{0a1lCFKM<g77dU63;Lx_GV8&X@
z&*K5ZKycjT0KAa-#lKT73L??yNb4eyo`7-^y2?|Kse=!JeV&;tt+mLq3@SeSP9p$=
z^ii8q%$^`cP=}#sO^nvBsauqF4I<42TQ?Fk7?4|dfFrwA&<4h?C;OB&JOXJGGmp8m
z6wgP{!qnDnPU%KhSi3O&yF$~P0`tPx2x1XF%e;g-it@gqZqS^c!d{5Ek^GFF#<bc0
z-osgo{(3{#^^iPBsVG!c)dn8|wnAw?KBXc>@v^Rf;!4>~gZ50k=UHBa3=n!A6Dssn
zRkexdvUWw+UD5Yl_-qOJNQo|A6Hc!?$Bx^ZFeys1wWJw&4l;8#<S~VV$v5aQ3>#)k
zv0Kw%v7;csG$~T-U+G7wSV1P`Et1dP3xrf?2?MC-<Y8+qin2mcY-6jA?C{i_k_x@T
z(x_DA1lpH?KU@?Q7c3MCT_T%kRT0_!eIDce!lzi1QVLKWUO9N>;6Ut_LCA)`RiTca
zQxx<m1bO1!6EkshGbf)JKl^^JMROU3k<&|}b`h9c^}X|8|8%6ykd{ZR5*Ipbh_3Dv
zb3|<}roDN#`y5GoNxAt2hdSc#KALlBj@b=11!x&N+E@!m-R<!%Ah2CXgt`O-hhXP}
zG3>Q5&jNp)W^|WptmUU-^t?bMQn5(JYV8c?EQWC)d~}ZawOxXd0WqI3RLR*t-tJtd
z`z9^i@}>4BP3n5RKx;6#N?MbU!iB&>L^Dq!%Vl6fE1MQ*IC>a)jj)<2+)yIig!1I5
zpGTW(o?{@X_bzbo)J+5PplD^PF$g{Zfb01UOl?9*w$15;s*%)}E%vZ)A_&rGY3GT#
zz+@Skq^b?QKWa@!0+c5~I;R^`b4m3OlAz^Hp}I6UXp#ay@3q-<?H&r;qh~{`(mYKQ
z&ACe|zU29a8S`~BT}bTYQXtH$sN!KulWO^$cYS-tBGz?@+5q!3nIa3ZT4Lfn+Eemx
z(W3gkM?Z9MZY6U{XK35V1s*gD=k7dk8lV&`-p9i@ZZ1CnAxSz2Vox4G9IQopIl7J3
zLvvtEq)*<AgCQ7W492nJU|AFD3syB<mJ$yo44`rjzZP4TMlo(*U%~ioEC}<6mlsOU
z5dLH-#gj1)vxM+d%{v!fD<ttZ@k+dAM+4i?1rbPD7P-$K+4*UcjUwqd&RKYCIY2SG
ztX1;2kD5{%<2dkbVqvljMjNzEL)Awzs`}#nm-~`2iMDCo0uPU-J|_&IG4@Oj>?<A0
zD9LrNG@uM`=4c|)ZA_^#EcFs@^Vc?6yan$lZj-qdKwb*7%B3KlseL)~^k7M!drc(-
z%JP#G1(BywERG8IM>L!6&oI@vv(YnX9Spn8Pp31QwheQA^lbOjog~!GOE7(#rpBp_
z2RFr<?q+lcX-MvyhY!!$CK(Ir6#*q<3c<yFk89TveybErrUOrT`uC~DlIJ<JZiAOs
zDKh_`pL@RGS2q2LKTK$Aj^->C(COy^l+6E@3dU~|Qe{yA-lMzS&~;r11}m!>IpY?V
zlV3%h^WKL9;xx7+eVF|N%A}=Qlwdd|bDtlXb!p$}<V@3qX&f0(qq5dTQKG48j!W_I
z$syeLCw(n;#jWp%{EvllUQ&tlXqw#vqLBQ;@H>ZjT{xI@g7<U_7X_J&Sy8a(E#fmQ
zX0~RPjFWFz5jsA6ckosL##4!OM6)7f1bYAuP%N_~dR_T=_yubM_@MAV<zUL+iSQbR
z5w{!7lV~|r)1Yo!fI~MFN4zVshP-%()BWKv4jsDN70y!J5e&ulYys(eFPTfUDZvE=
zf{+7glp@}w^Ept%?Du85$LMvMkKhpC^d9f2<WqBsJZ55c1+y*=kDiVNh~<)C<JiM<
zkgzfu&92%@>it79*8P~LX~Ok(#Vi_`VvQ^zfjqik2$C^DtD5H->oVi=`5Dj8&&Z1$
z=kp15(>@H?+|SqkJO<#MJ*r>R-Xx^X=O=*8E)g1zg9?2`_41}^#59coJMx~;e*xBx
z_?FY3MJVhu{o#pfwW;cd+Sq<Vq#x48Rsw{1A?ms&LN7;~LNJ*}Hz+Y2^yr5^m{Vk$
zQL=4}uiAo|-aBZBkYNDQJ9Ir&+OdwjD_R+}4ST5s{u1p#fy}X;85s@%IDbBX+gwE3
zI}q2VkHtMTX*z+bHgwB0CN%H_zfMd(dAdZa^Gq{L0*Vt?Zx<}f9L6YBnzLLyl_0M5
zAVd@gHUh~p#L6K8R0@nfyA+P7B<y70B6h}W=RNfm3dS-~Tuzz_(ac1A?PzbM`-`Yy
zI&|HQ937sREZb9yQn<jA!N|%xY=xFnWCBvmQJU<lQ8y>l%+Zx>sJmy~k5`1iG(|R1
z;nm+??)$lu`4?zLFjM6n&L^Tj%JU)wZ0_r79$63RXGp26ebV)p+M9$FKzvOm%K)X3
z7kRL^;VI#B96ENs(54q}o|M*k9e<!Ed08Q{+<Rz6c5_jdgrn?)+rBRmh$$Vg=N!Gd
zZbBd_;?di*{a7X5%dQrAX1IKHy<-wR05f@urW(hTcw8g_wq~YFrPPMGK>8vL=-`=I
zvZUNn@iVttRVosU)}(Fl<L@!4{fyjUQURwDL8V}=<1%MGIVW5+tMfbq;DT9pYyibZ
z5zRLY{X`KGN+Zt-oKED;1LWraGEOGWmsosOg`xm;-2@!*HBVW@dK|?tU9JQhc$$l*
z-d3U!?Jbp<o7YmU=Xwd%lG4<YBMzqHF^c2J1{{k1OeJP^g`<mnYih^-0K##HzQ^iZ
zVAjaX0wu>dere89;Uxm<a*5CSwe8J9%7fI@X<6aCXYZcFGB23M3F|x)*;|&wdt%8<
z?bb(vDt+LTZBrW26_-$o_>YeINe{I#CPYz70<m+B(kGf05_B4b0*>ux(Z+_p$2^QM
ztYu_Hf%bGFOv?+uklu_rHcr&l2Gca+%I<3C2*9Tsed{5K6^)ax6nDYy5CDN{MFdq8
zETU17fJ*zP>^^rVq{Oy3TC3E{1Rw~CfY+@Hyrlrq_SB+D0BOj$4rVDewMPyB4Vm>y
z^9|lWMCjM#e9A>7ps(zW)QMx0r0;Yr={(C^QVi2=bAbi`SwMWH?q(A#xD*t;VaLg`
zu;XM3pDAG#<%3dLj~dkmOz={Zfxztsb=?Nm)Yq&SesU<KS$N(s%_~fn5xZ<t<K6pr
zhcS-({`Rj8!w>{iifevNdmHH;=M-57Ov@aqsd!IbxMiNgZJ`x;meR=bM-P&=wB{X<
zmRqmblk??BVxc<`ZOpTIq7-HG?<3U3Zr?Zg{Wy;3IDpU?L(Ga?!)+^A4i;T0Hnk#6
zN#Iepk)pV&VpZgOLC2<0VuBsdW5h0$R+~@`IyG#LdJi}iVxbQEO0=s(@80I$?18&p
zG0zjHz5qo@tc;m4BnHEjc!=PJisP2Gm?SWA*+w+CZA&bh9wA6Jfu8R6g>x>H3zzca
zlDgD15-t+wBq})fpf8YcS22k1okw2eG3b^T$<xCmtCwsq?o)X1Ls@i0mww%lcK3+3
zDD9qgX3X=1%jE^@x&(P!*A<?gp1v5NzWp7Qt&0_m=^`w$45y|=Rg{6bHeItv!cS{W
zWPL9$p+K%u3QfX#@ul{*1Y^$M(#m6<W+J>V%8>6xwrl}wJ#u4ERn3O3A$_Q$wvSK;
z5H@HoW?u65dzCdkG}AP4?L5owVZug=x`NwbA8NaeLZ2SMGE>=aZ&%JyR49uIr}in7
zTT0(^H)r0%@wmo$qA{%t6qgXM8{z~a%}!=H0@y)70Te2rB2N@g0{xYWhHFHO5QH0D
z9Q?W_49i;z$km?e@j<>GtP`a}ZF!!I;&Mor8>V3l8kE+Ws6?_1&dXCt-zr%HL8NHx
z3nwD4(;N)KqNupKZ%Kg!-ox;CpzZ~VFmqTtqr2WPb{#}R$nyg4-n~avH(_1UCyd|q
z=5_BK`u@fNpq5Ly<rHqvwD8(QCqTLeJm+C+c)|n6en5A<!7Xd}#HMZ0oX^m&@Vjr!
zMV_>k=YO3RN{0f`C-3P}_Un4ozhe-;>#htDC@7;))D_X`C<SjBS5BY>#}|FtH-OD0
zA)haCSzp`UXzqECCqrRfW~}QPk_+=ZM_HCwt;M`p7@Z?eR90z^ye8iJK);dRhH0MP
z9-}$xhcp(8aU3?_EtMmZev&tCFXQBO!rJ<-k3@C`$P5*&tm<NeW4?M*0az9)6flk>
z;61Yd6lmHOSzf@&VtX*WnrdZfUp@$(@b7Tg<$yBj3JY$n#r%u?$-GED-Xjw%{7$4D
zZ_acrhCcE-dG3#L^J+jLiIToIp=5r57;X@W5jC`y0tkddZE(5tzsa7U<ZrGNe7h2@
zFf&3B)l9oHH;&gX9ixOBtm^_OVzDX9oGS?>te4Sh+Vrfn;)_CI9=Ql-5=<I{feT9D
zPv-Z2LAYl{wfs`<DtLcb6gOSV{dr8+l7#0KZe1an1H>+SA}h*T&NPXE*h(pQmIG40
zP1f^vyKO4b2zS*{b>&CxO+sp(Cge<^x~^*|LvD;gQ4~yErZA2Rp&$pPsU98XG_pLE
z#zN!k4;F0?0*Z<+F;Vn40H9IV4VPT{%7<kWw(mRQf6fe)0-Cl)%S3k8K32z5Q<Bm{
zdh`Qra$@tOB9nDhv3l;q&C5NI334Eq<r%&RKJpc=A!c9nE_ryulGK;`QEF@QrCRGe
zN4LHl{IaZYYRRqd{Qd2^$0t*>I+%I<doim6rlsg9ygVPG79E93>}#H<aO<b>eP09I
zDgbbFd3M(u=5d7NIp#%$)44?+ST;E|N!Dn08H?$*x=4s8P0YNqh<SNws@J3Dy$?8)
zSTafpE9U9Y%zO0QN=XIdE9QxWr`8H(RpE4cLRnS;Q?xD?J_!XWv_={~Cjpx;FVA5O
z1jIj|&##)Y`}QUwH4Z)Ox{!ocRWS2}LDD*hEGwXl<v5tFMsU<*(j;Xiy_F<%;V3~9
zIQDc%Q3atgan5t$%00K!>rFPMTO-3kkrz0%Y1Nn40nK>$2Hb9Rn|Myjt{X`Y;ZpL@
z(cF44#QhxHKMRK$uyQ3UKq(Pm^_+;+oYHVE9)cs6aGc*X^+vutmSx7!-7qgBIAIV)
zN!3Ous(d>SM;f)fwkdYWf0KHK(|#Nh^EN5dC4DZ-g6s8yX&zZX5xD*I^t4IX-ag|&
zQwo;%!N66CJUqMMPpHZUfPSEg-*v?%V~4)K1>8<B$-1tgjV5#c6-We#fxRO~sITiB
zgnFzqxXpa{lIN1<?25i0C{#izWE{F6jxQ0K5<4%vV*m+dl?bVE9Psk;%;88R%u<g2
zR7iiRy^*SgqD|*{!m`Y`HZ|VAe~&E7nDSwT*362?2j$*t6q`((faPfdC=vcAiO2#G
z1~UhD>;eLor-7ux6ZLgP3F-IFc?^AzX&i$Ylo^BebRy0o7|za*@7X_-G$K{&Fi(^d
z<-O%+J0Z{VsA)K#2p6AljUzmFG|?>6<m?5D%tgeaQk;%{rz49bYgVLDsqmRfZJ%SS
zL9x-rn$I{*!8-~7byXuT3Iq-)Y)yLUFZ>Jwk%zv=z+7PfF-o+j6UwR#cb)U%9prIA
zfk1f08jK3_-E~*MIcN=-EWb;F4KsFlfXO`ePAUn}fiflo6wow84MTycG3ZqU#o;dV
z9@9ACdWlrK-V?h)%br0Q&qjdmu?b$EWf87;;b~LIB_68CgJyo%SMt2Fp9qJ#!FqCf
zQ4Am@%qtZcJDj86D}W1lxagqaxty7?^K!X>vJJA3Gk6rx{iXIMA$2-E!FwMl9fqz)
z-*v&ITXJzp4Fx8rJ-lT$c!c*!xphjRARE2d@;kXG?npMogDe7Qnnuj?7=UsD>52`7
z#QyMY$3+^#Gz}Ps9)Js4RMpg|+ZHM_d|2c;j^|1#6&NJPe&j$>5-LsGqU6wpSb9ss
zAOswZkwo4*EOy4ooKv5lpYhXAKLs#Sup*ic2M?*B1Nx6vP)6dUEa3+ZFxtZ?3G~##
z+@zF|?L29%V;OcuL4Lz9V09~v+<8<b5ziT&1#n;1YCqp$Jv?zI4P6hvu5<vjM%A=5
z-^PI1mhU%`0=y^21L>P6iZ5{M%vk}%x{&hLr(@X&ECDD%f_RxG)_TO0QYgw2b=yE^
zSxn0IwC7)IcD3lPS1ePSNZzp4S+d{tK!AF=jAhQKs!i8*7`l$K4HN<ma9-uXYbzSC
z^d(8g@XQaT6`Z%M>0hy~3o_30Br`jSP={Pr2+D`}_?)xIK_=fkO?ZBO#_f8+y3PPJ
z$}+>#(-YBYCp6w)+ujD8YlX6`kd+0|qh>M0o=m<qVhqkgAxkDKPJCDZwu4jdb6{)c
z#loXoO)70^KcYohYlFG4PeeXyqC!iUfKc~Baq3GODO)8B;;1{(-T)2&?K;MxC>gt|
zE1Wn0bW0_>{SL6E!}B}#18&zVo}Zub;m4ow(~m!H$i-CHkPMu&U`fORG}D-9x&1*l
zlT*p(KbT(`n-txkIpVoqJ(UxyiGYyB?M@5WHo>swS(0j}aKp>`X9t+(ey)R_=p`_c
zQZ7V5l|>mizEX$G{@G}TsA;WXmz8N|BD(?612uf<ZRcyZ%Yj7TIOhV@uJo1GWAwYf
z$6a6X83~7?v?IrKJDD-4YY{5h=I{5<mOi(K-|&=+0IRApu%;fF#rvRQx{z1Q^NQ!^
z7kv8k3DZ0QTA?Tm2T#Ym-M0uR;I9zONfMZ|rZs`%d0sHrlurijptU<Z$Z{)5$bo?6
z-p4$@z3cRmfRaNqO%rJVAhRXc25nWN$d7~Kj=dZ)kkkXvMnf5mx@o9%5O$`^11v(c
z&93VhA0yn(u*=%y1@Rjx!74~VhEx~@c;!Pv=`yeA`w`d64WB-I#IJw#E6mdbqYbQ!
zjGZ#3m~KK}MpNY&Ch}ywF#g?fQ=ulF%#T(Y&H`OGP?Zvh*=<l&F*noFfWoE^gGqZ=
z3f3;@J1)ct3wR=)OApAam=S_FfVGbE->5E~2auTz=kpVL2JbJ(bS%^tWlmUY(cP}N
zTwX#Tutb7yBezolwcDG?-ikq%LU$Zv<0P0=ynpGKqL8xys7S&zjf|B#KvC8}T~%n?
z6S8beS+kkCL}1GK6C7$}SOWmEJlm4%-7nppIqRADlFDw~;F<4D8I7l>Cp<m93#`7P
zm49hn)AlAlDLAfD!Z~o(dCbeov@KK;YC|3RKw=8|FPZZ48W1DYfh|dy$b@vy3*I51
zyKr+m=QjQ9F_A}O3LNUZDYem*fXZ|fdw1<4p$U-7dKd;+7qO$JZcvp6!VZWy!ER%%
z>x%x?VH!rP%YtbfVYI>b-+hm!xetgV42?`6I*dMOS)yrqP*|sIKnYj4_j9Q5OtPk3
z*WvPV;X)BBobxEl3T2s;q(a0)6b|J$0II6e^&K%lcn59tU0&Fi8^tgU15x*`tO09f
z(6%QwG0eu~O!2++OIjKoEka6yNuw~?nP$5$@be4ZvEWbuLWH^$FO@#`rk*STt(_z_
z-#A%b&I<{sq9}GBzMrS9P+r#+x9+wPo>`XRe4=VVDc-x8ZSOZ915euiFbud{FD&o~
zzdoPe;hS&1;}kR%G{GYdx!2m8tHto05AOgC#}UR9FzGB6*fcsL?*dX>C2Cn>p2bqu
z^!tKod@~LJ3@&C*nnk#s!CHV;0ob&buo3GWv@s~lDx?DJ)wv@ritwWj%QO)#q7<jl
zG$@J^*m(x`9;*m86Q_57yJDUOcpC#rCC|@oc$V2?;)IrkE5dk6IU1&s%C>u|%2O6O
ze)-GahjR}9{4f7`G9M8G@@bljQcy@rkE!snP&S&aTG0g*=9WD?BH$!)cv)6FJ)O}s
zM6p|Bl=PivS-6vENbwH7H-*6NZX(3fFn|Jzl5jr)kqMuSxJndvK@;M9M9KPJ))`bn
ziZI0H6oW}=T%Mmxgr`o21dkYeI80-RA*;#)?fJx|%mM7g?y<5c1!J$|;Q|#x&59E3
z(-U->_JX*lJ!{ec5PSl>KfC^ho~f^uQZPEh)9Hk|suB3Yf>!A71Xb80p27nt?kmQl
zHBF1=bS4~-Zka>uPF|PSmSv3<jJ1L>84C8MN`X*kMKIAbKIY56Z*Pn^m%KTe3`sj!
z*9GgE!C6a^OcB^8Id(vLs7XT_$B`q;<3Sg0<?vzheJ6Fj_e4GGNZ3(+-kRKcO~`CN
ze2$&KlpAjxCn!+zsnP~bU8AWRzKI1BNCp?4EgsByrW;DAVW_&Y38@pYkVTXh)aSkV
zF)uTE2B)>5pv-T+d5?GRfzO|xDd#N6f~^3JW0FejQJ9Iw6wJ)K>zB+l(S`u>z<&7f
zfp6{xc3n{xIhwkrJhSle;&G!EJcba*h3E7VieUmTZ=M%uPiJCE*vy9f%kQ=U-Qa?c
zk(Z0irh-`)@}lHq->4#POupSIUhNXGBc6sV%ZMelER)HL>%Lt%gx4jIk-XLx3LZtC
z1#FWv+Vs-hOg%p#ODqKpBVm%(E|hY{UhAfAV6-{RzTC9rm(#?Isq-A3hb~d7VC<7;
zJ8J0#?VQ7{?{K}|Xs>5EdsELFY8Y*F;D09!dwKi;yY?n?Z<(j4Hp^?v5<Xnwr9Ao7
zxo88mtrj(g06<awqyQ!*=^n%t@6>Xoh!#Zh{G^s!W-?+$)UpNdRcU#@G~!Ud!LBQ1
z!~-bHA_$iK=STd94|!tmk0W_JWgZk1;}0p7?8c_sjYoyK90wBQnbOFZUgq2HzQgHs
zl0OZ?Xlo@>fd%b63|MPW8(rjX4vv~=lmPJI!v}o$@PSqN9L8j*>X>IHhuY)bd`jm$
zG7@w_6Ed0MiZ3xEAmT=k=gRw7ew;vL6H-;aSoSHI6!LOW<h+<aV)D1`38&LJP|?IW
zPM<LqfIft<$ez-S=QXtiS!P?j4yo|OW{(3gFm++?C$MG3{3_Flhb}l2#VN7Jl35qZ
z#JO+;n5Rv=bdtQit}C?BA=>@SF=;ZF5yNxI*VnZ-X;NKx!)jNQWre)R18AKJ3DFeH
z18^*jWs*&6Odbkgib@FZVu>b|0D<zkLarwg-57(mu8|u9yA!6;j?BEVAF$2~#SbYy
zG`q?ihxR$53Jt`7#Ce%Xi*gP{UWUULxwN(hLaNyT2pZ}<O)yHs8w#8|ozGb8ifLI6
zmRSY-g3=Q-GNX`b1!%grL(LBlI2FyfM@=pfp-cX~-}^n}MFwq%zPD-{s4NSVFQR$L
z`QJ~OXsmtTQ!x)(1e7%q;zi1o-E+xoD?k#Hy?2-oN-I?*F(qi7<_hwAcLNGJh~8tE
zMqF>#km@CsXh=Yr;L*L?1FRjA8A-3r(}dgQ3cC=8nNkX6U88Q#Fj?fl7o|K3UDk(0
zWAo9J3f?z}m$5~`@la92d)yQb(vGx&=4f|`A=^LOk+722r)k2=%L~SFM5c+sp=oNo
zd-o320RwC5-P`bFeWATcNJ(+faTqX7BZ{&_o@aq}M5srcwNM#t{xn^}ImhpvgKZ3u
znI~0G<slw5d0<L?8CheYe#&ACmhWSu(x1b4#t~iL1(g&EVbN%FlFhysZqy$_zf?)%
zM&+`?O>8!KN?{(-=TrMWU5G()%)#xHLRFUVTH$)@Hd;&Cj|fbyGzH{o5>iToUn9eZ
zf-hCjlpfud!W=ozGv2*>!Vf=uk9SW`6y(=5D9Y;2E)bBHX9nrv5#qUNn#k1h9z{uX
zy8`^C7@z2VK?|3(YcCPfTr*FBalZ=Z^J&Lj76Vsht>QRoYzvQTU#3YS)TO0+Ds3qQ
zKtjFav@Fg)Eeo7a=S`Jubs)u4>%n^pa_u?DRbH2}TTRnM{AoMQo3Y9oIPcMS9Se1)
z<5sbl#oZq8*qEc+XZD!qB_s|@{%&2@czSx;jBP92RD4kXx=>Bqn}k$*I>UQ%uY2n-
z-+C;W#<i~T?%h*R?N+;@%rcmOBhZ2@^Mal|UJ3vlP!&Y>cC-QHR(5;~YEK#2(>!6A
z26(#=HLr6hvK(#G0w$v%vj=k^{Zk#cDWnw`KG;Q3;?$gwnMen`!7-AK6)5GzdxxnX
zFm|LNIBVe@a5^<8tK#mtwgwDq9vpvQSBrTZiB?#VK*{q0Ro%kDqwB}*nD{@EkFLB8
z!I{Qn(ApB0w`OJl=irHEf!4+eUgZNdFbQm~zN~<vDDaEF@dJMN#rHU!pKxkUDD#5W
z!^c2B@OMS5gd4smyIPEWhiMp~7!EMT;MAN@6*)8jn;ui23QrMM9J=Bx#(u;wjO?1>
zA`j;VWtp=^7Q+ON1WO77h;|`mr<5`{H)k|uv%%#anXqtl2{^Z6xOEf=tO&a`TBB`i
z)J=6K2;{6P$eCsHmKUZU2K3z(-Y%5jt~KhaM%A8}!(NPc68Ogg1@N9ibo%asW$B?5
z0poR3JtWuDAZ>dq|6I6`uh(aE-33Z{7^9Gv6;4m@HvyaR^JT9_m_$NnzNWpg?jFU8
z6h(n$nX%3@R=eVQyWrMc0RVYXGDo-LTbt-*op%_RnIM9M3RPX9$a6rc!_DB6&xWTk
z{Bazx&ND3xz;RnOxpYM{`NFf17Da^fvH>IkvaYt}x>L1xFKm?3Nm(jkU7(7{k8qLT
z@jM*Uh;bNMJ7j>gZBSJeh67&cRB&oVpnq-@@=PJmHOfrEh}q{otn;wmV*-`33SiZN
z@+!Gl@7}$`58wX~QcKFRL}m<IP=2#^@?c7d#z-c+)VZUaHM$Vm<`gtm!Z9B|oAgg+
zK1E1dyF$VZnx;kDoI+Cb{RY@muh19;Q?YD*I_HSTTL|%X4vC{C?s^F*rZ60eYigP{
z@L=7cFe&oBn|~>NKF?JC9iI8vPftX9O_Ops9vQ0$>1A1Py<9MiJtVWBND6x1y;q<E
z#0w||%Db*(3Se>y38r;KQG9r&{o8=^H-;eigg1&zy;Z<wet>zNu`CO_?PEZo#0_yr
z3{ICQ>pCRa-f2;(s>rtbI1VAJS}0BOyx<Z-J5wrY`wbRDAjrliV7iGeQ}2$>?PhE-
zbfM>Zem<03m&3f*5M!7Ck1P3*X_|vM1Wm!6by)`w4sb6l_lKBdj)PGv3caGvG*$S}
zq;V`(V_rRs^H@2B#&hMEL(^E1hbzl6N+=Cx5>t;r-wRtvY!(2hP)=;-SRti6ilRzz
zNW6Nhq|ds}6T0h_l7~4uRm_Ym%gO1Y1jgKw;3c~u?17^gCNUYKqvH!>Q&br{R(o+I
zRH_Pf)v#9ndOzG56iDmrZyn|_UP>tr^$2o4IzQt5?E4<x(WPp#45#x`!0tACN=2<Y
z>OWz4A?|CQ3A-a4>a@|w?ln0UaQuB`Q~hh(n}n1cOxdJLYZOHWr2wsRiackpg_q2J
zc&saxHg|4iwV122HVv5-8xO$t;nK~u+6BYV!>%j9`FSJ|Y8XRa!ABcm4@?VqFF~3@
zFi-h{_wPNLho~V?_C6&LQ*lqZiH`!;!~zzK)O&<<9PxU3??O59yvX4#2ekTaiFFx|
zG_4eJ1Jt?4`wDnh8RP{WW(R{|_UJ|nw^}UT!n+kpLPwJ4lW|JVQD|+2qEG(nk)uSg
zI_E632C}?l?xZq)><$AtY6}-;m;^h5N1hdETgrb6qdgP|KKw#1+dU4IGj(rKd|kA=
zypNbHFSy-q5YBqpPqF^q7rUZM-8%GL7m}@&GOXc0p{kg5;VYY5pyIJqN&&{8VNa?!
zt~Spp5C>B61crGy?PwJ0l?+W%IYtEDe&45OTju@c<>jlN<IS4X(ZeXEFfS`SQszhs
z5S!gf+~?~GqbZx(7>%l~XaSS{kFf)4WC{(WCVGx6XMvFA!X|J(-eBsBiN!>ALr(#c
zqLCG4iK49TZmA;yC2d~jP+l9L7_z+JGV&5cs90&<bmPOB5zK??!F6Gi8$&D*VKZ@z
zcvx_GKabBT3K!H0$dp2rDV!>Ur^cWui3@A>4ma)Lt;aMoVVRF!KL-KTeKJhZCR7lN
za@++AIi4SK#gC&r7gZBdis(eDx=u=XKd?8GKILNRhYpr6J8cZkCklf=qDAj>;Yi@b
zxp$6O7y9nT8l($MqE+4Sv&v4m#}hY>6YtUWJucT9RyG;6QJk+9Luwv^q(}t5^O%Au
z8^P_0=Sw`OsfM_n^Q~;2rwQHdMxjHBFuj^X7qU<g_klUPIqIGZV_ea99hB0@G76ZM
zfX#GFj}#8aR5i~p&v<_M{BS<@?ajG+@`34-IM2)oiALl+nLX>UVi>tlC?Be<$WT?~
zog_a3sO7a~nJ|r14OdK)EYDF?Qay$=w=}{XV7uB&?7PM?&FF78*kysrG@P?+b~YTH
zd~ZV^HL}fbn5H3kH5C+^=8XJ6*hzSR2V&(ZZWOw3gSDjDc~~@cy&*1>i2L-pv}gRl
zfvJUoOe^GCA=4SOR`ABdYX_wP<v@V0c2Gt`=Q;Ay!>ta*>X9ugrfCeOi3-A%Jb{F`
z^K&TrOejibPBsV85i5m$8bi>gg5!!mbrj;FO1>hXWIqWr=A4wbYN{G-bAmEDVM>U?
z<H)5B#AtWBENmB2Qf=Kt)&;q2kH!{KjsT1Ug)YqNLd+4KQfSWSC$8lu13MlIfS`HI
z^AsvI%imNr)$+@VJdS2>>LZ~o;whb%8Qt}Qc^Jqk!8qx*lzcL2yRo_CGpq!rc0IhM
zD%)9B1TW*G#NoXrHS|5Mmlu5g^bwyweY*SHN9}E-cjFj{uEN{{fo|jW=>=nU!+7d3
zoLW?MjmxbAyeIdoqgzf2g$izP^nlVtkd5QSAfm^*kl9ccb%?hh6AZ_Fc|hq{)_k7F
za7z_M$z_thj&oUVjA0l!tsq9a$9R)tl{^ZVs5!`(rfCd0a#@yhVyrNUZf~b>TLmo#
zP@W>FH9U;=P#{wd7F5p58!SG9H#Mp@$G2^b+7}qc3Ek}i{PY3CG@%bSB5&f&K|dDY
zd(%j%D0rs;_G@Grkp`B<fe?Q%UrH%9^RAf3Az)pSW|SBCofz2*Ckb~EGcUv)@n=L3
zd7Gf?+);vj42h5wnsFRp3<;>JtnhSxcUP%7Emj)G6xVfKRx<O3A&g(Bb|d@7E;25~
zY?C!v*A-nK<DsAxGLxZg&#c|i{!7na+RuIt#-YdcdcnLbFh&Aq+c38D^Y8Ykk1<uZ
z+YO&TKjZV~XMFnfJJO_vp%0pe$#R&?Kr0RJIo>GKD2f~^SIA2P@If&;3ohoIy@LQY
zlfF@@)pjwdlJP>IsIy#wxJF)-h^48H1zidTQtHbXey5cVVEKL%9FNBXoUhpTH%v2W
zc9!v@s%>C`8Xe8EeLr#wi-tfOlYUQ}-X|_Yl(7I*9BGkY1dhxp2M5ERBr@GBa%GWg
zk4zh6ri3Y5<VAxruTi4J;thsX!7LsQFX-8oPh;DSQ>D+Z5}_p~p%12@0PWU$bmM@b
z9{`T^(ppo3HFK38t<#ZkvDRYfJB)n~I16P|APsDqW;2HUx}|fTg2+U_;^3VR!HiO}
z{oef9?<o>CX<u|fC}@+ReLAD$afu)j=8!w5j-u;|0*Wb9K`aZQaw&>3j9vP?vpnDd
zP9*mLhM~jlc8NS#P+`7gOj^gwXy4Lxn5GGrz_4jxvJ6#2j*67ZWiy^#r!rTUmkU09
z_=L|NKjZVq-;t2Y^9+)l2oSeaQ54V_X;-Gq@bt|)4$iE?d*<1KD*ooEnI}j#<&pg=
zjrfM<z<usG7cS2-G<6NF%stj~Z~zp5kv(V5S!hEGN4z<a;$lChd*D<kLAt<}Ae&0O
zdGY-+p5)!TKIPeH6Gyz}d_E(~GS~<9z4S#ub9ez-bE)=s-{9%H?{WVAJDl1Yc~t-^
zgM-1cI@rYqGfO+isWV&cOwOky7#m}Vx`qXj_s4=o1n-^-SXcx$rO}hO9)Z-MFRo&2
zG$^OCNxZO$v*dMc+y;};vZFvNg|=xzT|m;dq6R_lz0=d+z2~aKH@*aE9@-P7y(pyv
zOF?*+1G6zTWxK8;Zg&oc5pH)%)PJls?fVObXO{((NKI5jr_&jERsbLYMFQ+G2u}_H
zPNC^~x@e5xQ16oBy4x}BH7N;{{rveeK7RO!=g*(<@xuqq(|GrpkJ=lVIVopso+hlT
z#XOIg=LsgG`tjZo=@zAcSY%m_%;a!x;TWru?Q~H$w-2^=e+yW;EDM}n0cW9sz&GWT
z5BNy{#*@B0-*C=aOnr}ao~d$*(WL&enRj#>q&INX5)4y<3|N;H){#fAt{b#<MJX)2
z(c=l+nxQf#@w`ma7&duPFuP(|;6TP9&5I*U2*mNNfZXJ0+Hdjn!!Pml!}mCS^9`y~
z3zeI|o4t}`fmm_5T4A(>Q5KnUM2hEG03m8dp-Qj_Ek^%zGCrCkk9D=^x&hNTF(6Oc
zbKRbh6(wwZ!)BfVD!mR69-=hSRNpQyn5T)ws6tsbXix8;Oa|)_1DDe^1efpfq;V>x
zu}lkw+ZC2;Lpmswbw%t3S&kU|iQRq>dSDj>**q-)J5t)idxNTJ(4L+m^M;P=zZb^Q
zT&F1f;l_AmMvkYduF;&%RABdkyE?rf6aeiQI^c>#;!(`Y63zLELxSY_07@^yW^bk~
zO`!o0<Ds6OaB8>Jb{=U{68HD~{EW|^KjZV~Px$odBd)jWCa?EP?QJ0lnz+oftbkot
z%nNz#>`L}@SC@FdykjyigR-blloj&Q!da^NYm6{bMnUz+)hj$vfmdqB6BSUxIKqwD
zM_~d!iP54|6{QX8CVI|Z1IcB4N)e_t_5<9qLIW`D6+E>KvOAJKAQe<J9+Idce8ue0
zMzY$8OG#-RQfuza7Zx;W?{p|Y-ZG`(Vc<M5h&u0}Tof|!lysacz-yo|3Y7th3@~iX
zk!*^f?9o$&X%7NT)cPCx{szH3Os3GZ4a%wtp3MF1@I$Z=rcT#+MSr_t=A1R96f)*L
z&x<mW{xhe~J`PGgM=J$q7xL&117!nfjl3wioLGLS_3nlinqXSQdrsrcivp+f(=Ht+
zAb7jJ61;~mDr;9ZzZ49QD@~y}YH4K_y7ai9Cu{f$3VD{J76L&YicHgy6_c8$!l^|`
z`zEC0Jh7I)f1dq076Em+T=4QjY@HuIe8kJ;MghfNFQeWjq^t)r&%&xO$crqnAk6cO
zVH_|`6V)|S8OpLj)l^tli)pb?D#k>m516)8Af#rVXP0bMRY~S|szlS#5fzr~ZVkIG
zAqrSPe<7?-xzCTrlWJ4WI*eUUWsUe=&yAsij&(yf-cHuuyNc=cg@YR<nN=w=o!f>?
zPYQb5;vwGqSg3)&66YJ=%v3PlIx<UXGq4V#6w7oEqX44;V-%K5VaYTqqhUe>I%!J|
zJ_K*3Sje_{p7ZE|b6j~?aHzI9ZRYt=+JtyOVLirSK!4k^S)NWOv<dI_Yb}~>D_07>
z5eB6-JeyN7wue6aai8=)iTR?)c_xZ8L=?9-e;$u(8)SLA-Jm$WF5`;d;>3AmPVXTJ
zDVZjr3$j*i+u|f1Vm6DvFl}TqP9t7kp0TcE0cA-ks$8gCaBteBB$YzYM;g>;eEReW
z&(|x)Wu|C=<#!ZuuB1=`tJz9X5M-7W{M`Uv!&wS&Tb2pgnq#%m)<4{w(tl#^3Exz@
zAXQz5;taSuF!8;zcq?moQ*s-Ny`QpwJ_Otm$2?9WZo>_$Tj9Ke^9~g=Qc?mk64Ud(
z5BeU1HUnj=V$Pz|IRuh=AD<%?a%lzUbj0L-H!xWRttod+X+oNuBknB`Hb#+l1E-+9
zLa8-!1LQ`5OFD)NAYGSKU~eZ%HYH(x%d$W#(u(rDMBSdC&32nbyeHicB5`T1g+pCF
zyJm(|iLLsQkkX1s7sqiz*IyI<eI#X{q^~1@eA}ds-1T}zT9|<V>F9WQ6!x(1D=>D>
zGc;c!WhvS{3H_r>K|VJXV3Z_(#j-9G4x&7&s%4Tz6Zg!~vDuAF!XWyd3al8dHkDrX
z&vk_5${rU1_2I(@eE$47C<|nH{jMU;1sM>DtfF0;WhA7uHstlw(e10W=5o?bMDV8H
zbi%#&Sk{Hgk#n6h?;Wz7%Hqh)C%8iF1f&5ZFD=WAVHg9Ai&B)TARAN|4(>Iccp&B0
z8^;05x)MdQqmt`IQ%B}Us{f5pZV(4TnK$P&u{NeOw2qz^tqtKuz#Z@^kLVQ_PDs9c
z?@^XDo}Ru%dwNIJXzChSZs0u=jVUtVv?ddXO05H#;ny^oqj9|P;O;lJQVCt0iY3WL
zF0bT0#jrJP8_c}aG(9darl7pTx~#a~Zs^#YG9bPq(R}UheWD@80f~{Nuq<<6Byfx^
z6<nAhXk)x@;E$kf`3$LG+z}4N=5STj8)CPx-s!q0{c2{@*cLyaIK#TCHcq8Ibg`NH
zL@*FAedC&y^t-C8@bvBpd9h8$iR+lIar~XN4g)&`#Of$mKt=i+ex?UnlnC>PLD5&=
z_i&c5<+1B9+%6Ah`@X%2^v+rZkld#)PvD$&fb)R0%#h$>VNPc)VIAI*7ADga77=VK
z%MD>M3>_9rm44m2Kq;VVDwsT@$ZDmqdK!s*Dfpo20Ycq+!@SO9+h{5UPn(^}Nu{RY
zelw)NAf*=cSmv2?6IY5*mk<ajc+CNhI5<Rb6yG(^7`vW|Ich*@kGiUHYR}MF7Lr*V
zXUNKd-uJh0L3_X`g>mdKOjmg2u{e#@dnl7p-k4T!io7~TDP&3`*9LiJpferns6j<0
z1<Pxr*n}o6lxSa^0!6?p0N!C42lUq~nR(8`D=JCVoKDz#kRD-xq=^vk`PdJ{rLR0p
z#!25brG)G?Oz#|p0C<X=CsUY8HTB&ML*IpDcVmbUugFXCR(T5k74`!OO5I;l%MyjK
ztV@;~t{jsCI}ad@jmN?;fL6p9;W%=eFjCs9S&`%X?g=HEjP&te))oaH>p6SE12B1x
z_WTZcS@Wq7C?M@mppy7+Z!N~|ireKvVI$DU^Ac@)Mst3m5?(u4Bt16{4%U&=<mKfB
z&(9z5@zZbd@#9BaFGK?Atj97==&qL}e(uZdjlrB(XiZ6>bQ{`;iLhyrlGh8zlxA@*
z>hPo6fj)y+0G!adcnWGOnY)OzGs&HcW&=+(C=J1dT7F{UnKT|I(R*Jr9h6c5K%_zs
z^lVi{6_K^tVk3U{-t<-CsHTlBZvB`vS_ChS2wBPS+>dWRR>F-mln1)Y70Wb&IL!eC
z@?yd|z*}UBI9j~7Sf?37|ANb>pYZ(C2YmYRC$99txA<-@;>HY52tZ+B1_*L_)4NJG
z&8TeGW3xs2RZ0ad^Lo88en_tOrf$BvxNEAB54WDniAj7BnR#VZam<nfxjtkZ+x3gj
zOQ{@PN8IA@9=XZT)TdyY(R(+mPwxfeh{^}xT5zZopF5pSA;<4)no@L;^?irFBidf2
zDDuB;PmHximHr5GWu$@km&4ficzJmV>nP>m1c%zs@BO(^TvF_UPoF;X*E6n{TaX9S
zM8#42T?xA{wYLH1cEvdKFeXEm<y5JbGtr5WQwJgCybDa7N)d&KeFf4F-ZS>*dFNr5
zl>+zbCaB7!2Fp#b_YeVa4(l{y8b-i6WSqHQmIYN~<;{;|;f?@GKw><un5KaQqIJl%
z!PDspMOGvR(QAza>oAHpkIZEvzL%g)6Jny|vyX)55qPZ>X#$3+Y;IkTep=D29;<c8
zu%a*u^Dtn!zQ7L8n8z1P>jl#=;^lLP?q$Y&Ti{kpGpZ#1D%S6l2z?bkm-oXsU|AO!
zLmE@4qMNV??9J%?_yJ&^mw;&rzEd~Er~YL|f-S~Hu~kwjm>dCRS)*mQK4Hj5E_Bgk
z8C&um^EhJYZgA^DQ3WbB^Rnb!X^YP-sYMR#F}q#hEqR9(<D;qh`DH=nLlpS!a-qT@
zT!FgcG@+Q6w-roxj0ZfSp|sUB(S@!J%4&;4r3%K~Md=)tc}91;;^p}>o<DuY=Z`P=
z{Co)k&+{~+zuwUGJ*>Suxv#c238|{6;o(BjgL#>e7df&#3llKUGnRGbi0H^+uC#_<
z?Xd(^G1X*Q7Mci-RIck9MK1Id;$=DpBsX3`p#=euXE{#oiEQ)``ath_bdbnnLN}-g
zTYy<R>z3FYzG|*x4L-}oG?mh*s|Ib;hO$O!>{!sI9`%@P6irzV@Xj%#gvIUoGmITz
zuO5260@DkW9Rb_J&o?N{Sg`=B2Nn-_4FyLgdC&<wn<^3K3nrl=t$7}V)s%um$th87
zLOb@XrYUwK-z7CjLNWjZ9I7mL=}HH4p03@xvYF4pg3kLOIJ2VQ!VLHCrJsS|VFU%+
zUN4t`Ap%6@(zaBNDr5lYqfNj9iXn`dy<O|N!0=fqIdJziF7|t86R|<a9?uL_9L|dZ
z=kq%bU5H##$NfcRVYQZk@}Vb_SX0d7)A?zm!S21_>CbeT_jrDO#`C9-`1s)`eER$u
z!$|yU>#`8%dq-(x5FGI<+na<`-L%kJVOeJ)y<1m|qlKO|@|=o<T1SGyGD`uKyK>3o
z>~e-!?Hs3xDt=fC?}!qZm<fu&5Gjj9-<cIsY(gk_=P^tZ#%Y509>!?ONN*}s4W&JZ
zHnyMaz4vOK7L5JCMEaz+6;*}ye1@X(D;%>V`c%A`NQjb|;he+JcbLY3(x0`Mc}+Ng
zaYG&TbFX3x79+Kq<iFE!!^=kpcl!xrrP1dzGT)&<hpN!Xw1amJnV|$xr>TG|l!FA6
zs^oJ>I(blG%2?D<0kDfWH^>7AMKNO)SJz3ODSd_wo(2-G^EhJYJ7R3rG)KZn5R->#
zW2ZQZn4F5WNC}V~h7QhKz*8V<(=;f`BA9C=IGFJ@X`Zzn&BAWzJ1#a#aeapK--@c@
z1Z00;7-KU+4PwU_`T^rGglEX}9Ou(HDK-k;pRpfW|E*`-dyAp#Fb)IZnoxv~of}lu
zR_oaT5F3+SaZ%E~yW(~uk7aH$D$aU7p=qUp_`UU$Qn5&_ftMEo%0GVkjOXVU4BY^4
z9h|l76}=E!2rpZn84PzlAGbG_wDQm<L!pWAZdq5%(@Jqt(}ZQ2u&xD$VZizPgjgco
zLF<fd=`CltQVOdj?Z>W5IDA?Y**|5aOO$jlHuzQtpeQ;>d5vXV(GNXi8e|q0WsbTP
zck_0!fUR(n*R8b_x-gD3(T>;&np2A`r_%9qSoD6A1`io&MAJB6o<^$nEbO0W;ZSFc
z4HV;abo<Exx7G%8xVzo(`O{}ANyN~@vP`g-1<UY)DF?<<W2rov(xJ*rSf^1IK&~|$
z44E4oAuk)w`9vrtb=<_$!!SDBYPkmAj1WeV=dmP}m&LgkvZ*;kn)k$}*$l5FpRz{P
z)X*jiz;Ch%5rlcbBI!T|ri68dM8!8Xr@feXnFP&1ZpcmFAP~9l(cf;MiZ+VVaZaeJ
zhQm8(ZjKZJ-ceL9&pGt{g#Oltq6`s-iuLmL8fH2udd)ym(uRIu2}s!!WnH7JN<gd4
z`HCq$dqGN?@X+_TTwkb6mgGd9PH0c35S)AzHY9kn;7pgx1)n~D#HUZ6@bU9!+-@B#
z*WjFn0o_fiR1?W!nQ1(=B`(*4$vbFo=93Z?UMYnl*T_r>=Paga;9GIQ_2q*1Pw!CV
zIR|5UWcht%5}AN$9I(tIv0-umVpGQ&d=m3=rz5gQ!-H<*q3^IxGnWuD%o=z`QB)D&
zd;r*`o22i1u7)dh!h|0vR$}EV@!qq8DD;+Pp%4h`h#gaDG!0R}>MUn>`WA9M0?DNu
z{q=gq^Ydr?_Se6`haZ0ofF#dm*gTYs?66p1Sv@qUzM6x_nki%+D3HNhhiP?KoP(8l
z^kltIsDfcuPQ9t%JPbqd&M2*rWd=>X)eMiYmA%$V0gXMTY2tzt2#K}jCgNJ8N7`pJ
z8N4U{`~cwdTohGB2QJH!b(8npYpNo|#&HO7T3SWC=X8F8dJsr@FIecckvW}+!InIw
zbzPIDazJsi7vl69X;c)$6mdK6DG0c(TVz?WwNSP*k5o`8MdjeGmn)Weg=mRY(}Yq~
z`*Ec|3ydS76clbi%m!3?YoTj*C6dCvCt(joR6<X6iEmCd{^DSyN9{irpOjaWbZD{+
zCO5Flf@P+NQi?*KP?QDMbw!pHF(EXWtZAAs4x}1-?=f48Jj;W(Bbs_gSQdB%73O*5
z#AgSksjgYo)Wi`a#A6S%sPt#SkCtT)+D%auXqpzr<Qt(N0Ddx<&>yM66IJNd8MI=m
z)tb^ekl<=*PTuD6SSSdG`pffYeE98e@t1%3=R`NdxEMRq`59==Ru8ORVRjms1*}ql
z({M_|8U<g?fCHA*Vs*gkRNy2hOmM{thb*J=VICIFQ#mqWlE|_Q8VXfeqpn+gDPcAJ
zITX)i%ulFjh2`-Gi&1g~q{2!$EGucgX$qCe3zTh(tSBLQYSIs6+a$OOHAg|IJ;fct
zdpJ;>R(n1pFUt_ooxU#}tNc!b0$lssg*>iW!Fj5Ne0ujp)sT1ZIc_9ODGK}rUDp%j
z??L;!mU8XU%1F(+D4-I{WnR!<Z<xmk$^$wxK?tX`Fsa3}$D;&nmuT`&A3x#w`Gqbo
z=cugL^@iK+60umXaB3=i_w<BQ`*jiez+XW~i3Ar7!g21DHVS!BlEBIgQEC)foFJY<
z=Nz)k1QTYI$|&A*(jnmyc~0pe(Owtrev1Ps7-Gomh_O*<OY)*X%^cb?>5oh)X))<{
zf(P@tgH-b(FXIlExnR#7R3e}la~Q{o84e;-L9EcMLg{bE=J2B?S>(eoka~W(z<ZCU
zrzfi3JTBqTD`+@vuy|m09+M^JiG|~hibA2t90~(erhs)G{R}u|;XG?)1WM6n3|=!f
z&88Xa@)DlIT8pBni5)eggkN!N?B{v^?}_s%SG$H9jcFwKJ$I4>5=O>M1rKM*Yj(Ze
zLIrK56z66(8<6(y%<eQP>(&r$Eq|JG1>>@+X-p3^M2Ux@p8>nB=&m;m{Sa_IqL^tz
z*2Lba5C@6mVpf5km_kIzs>rj9D22-^XpZ;cJ@*&8vzD}*TL*6`NIJ_Y<DqF!R8-e*
zHLny41)^sE^70ad)aTF7%=7OEC?5t~uNO}JQh@hpstn(}tMTpo2Khk&h)3;BLdw|%
z@Hqev&xH5dD8`-u4=On5y#a-%5THrr##iJ<!L1Z@sg%Ju4HzaO#dGWptg410iv<+D
z8-;`r5Y?n}7V|j3t~0sH4V!tbq`pMe`;jM4)aha9F!ndVEf9L*wylv@B|XZ1K72gH
z$*Y*B8MoU7!$48MPHSXkfvRbtL+%sV=d^(n`V!c%>78|0#tHK<!mq@mmFGEr@r%EK
zWm)m#k3RxnZ>aOo%ENkK@j0ATm{x_WGEfuXZ-9p;Ar6O(=Vcl_*41NK73Nj2J{^Pt
z2^|Bw0-ry>koT^s2%pR|H0Lw2q9k9|E^bYlO4d;UT$tgCW$FW_7f9^x$}2_8CVW?(
z3b}e<Vs0;Mmw;PVZHuNp1u7pWrnwA&cEx6QIp;8s6D5#4Dx{#bLDjaXm>=qq8A}Tq
z%}4SK59qp{FEPTDPN%0sjWY_EC>rQB5_ZF_$8Z|}>!H*NV<;%Ott+Y`z@I^IMw-H8
zcuh!9D-U$Hj&l6m3S|JDQz6{b>4eO%0+M(Kx1Kx~BFZn9XFNZ%fco?qLqEV-54Uo3
z{`Cd()IoV-CoOW|eOus%_YI!f0xJ6*390L?!?N1IBjrHCC6T%0%~P7prClb!*0LoP
z=4HXq5A4mO=ya|0CZ~!fCT{wzNko%4&r@&_=XuUW3Y!2-i>4><m+xs7x%c2=5mfY4
zJd$~lN3AYsM5(E9uQgcqW(>m+(iDskn3d)B{hJnfs3^G6W1c1q1C@~%ppwYODt!9%
zc_VBUSCAFcJ1a71O$2mGDdsN+R=zZ>({Srb;a^^HL~txGE<m~j$A13&8SoCj|I5D_
zHh$0y#VmiVc@pN7`A?DHg2&}qewR-vrt48B-aB|jG$*&~mAz)P7b)#2)m8*tKQ{BF
z&GU@w^%g3}D5XMQL@yNg&R%07=YgypFfEY|TFkWo3_<h0I`BiKW#kVg2`Y@0fax)K
z#>l%W{7|P@-ag$$B7`^Rhx(r85!dSrmUV*C3MR`!z-COgPe!v5P@d<Bfbx$YKXSn4
zCtPngSVtAir;#+MzV9HoUuG2Eooamd{*3oe4XUz$H?Olir0q>YYMKauB`&E21q&-N
zr#xtLt#-w-ELheBc_!ewB{R=xI5<ob(Y&Os;oPM(0X3bV;5T9SYN+~RZMcz*&d}5?
zdnfkqzc)R|?2Jjj@KK1hO@pc|p#Z+I_&VCC!eTp+dwUv2BCBU@$QTpMm{c&vTPl4f
zdPdKf-Z=JHmKn~iaBfAOXLx#g!sT*}>=67s#t`36SygD-7N@5(nx;mc8Dx10C`|#Q
zZULNyTWP~}eUHoSim~f4^=ZJV3eJYx?S|X!iXVRXo^U8;qmY2cy=J;Uz>*H-T#UH3
zt3{S)P@qhJI#i67`BfWJIrhiZNLCx@ZiE9X<3m@LTi`c7&+!Ez8T(V<VWPzO)9JWy
z=DtO(OrokUA*mezioC-4^t6fGPRF&GGeJjG>}#H<5dKkB6;4lQWMvVe`NhPfet`X@
zEYEnoe#SKSP%7?AqRy^)zOi{<`kjFCiJdpkoH+awE|&||Wr4F6>pbChyP&%f9SfkK
zmBXni@csK1-@R{fs!No4iSFx35Z)%F{{4UdA8?h30!3M(u9}c)K;LszRSlylX6n0d
zzvBm_TSg+6yM7EF9e{54q9{-|5lp+^P-NGVhXeu7k{c#7vBXYUQEuK|J-;{eBtm)O
z;J6jwr0AxuiTx7_-bz)OC}5}$z3~i9^NhZ~!Py0QPTFAGHh6k^3Y-1@AbTEbG##Cz
z8+4kcNC}dWfSTtS<2VIxoAkFXD=zZ|?&TR^$WSSctJWIO8I;bDX&^Hi6(?&;Xv1nP
zR%=O9L9~vRWx<CJA2>+!4^R{(C+W8A1&TuZ$SENQh#UYD`kt%37}C~?A`c{nHaJzb
zH+tGg3vuX&J_Oo|nb$N8<-e)z+8+r|d0kSd2y~AZ3#FITfiUnWJYf58A1=FLqzc9&
z>;f12o$$WTcA$UzBno_BGZM^w{|;qQ9?Z2roHM1E+rPV^?>i2I$xs$0&ZnnP(dLmL
zPI0K)txE{(KI8fM8PhZoEzrDhn$DFdUYN<DDGk1VSL6G0fpb$3y|~UW_rI&+prDNR
zbzN|~bpSqx*F}J6MNA0aeESXlEC1Tx#UKCCAK=~l_jJn_CA@Q(rZG@wI8RzYEzzws
zwFj7AdZ=j;r)h)&sD!v8N{LGHslbaTplNCvaZd_#2#3C-#Ar~YJC&v4!j=4o&ikzg
zJdB0=_7t#3H)yO9+j3r(j6sB4cqC5>K9Ykp&okErTqxeeSyWX4=L!yskofzZV1VKx
zq6iz?dq6t}?-#7Ah4sL!fa}mFiM7((z1B&T%@*H%_Z{B9e~+dm>fSW6c^?O*Hm{Qq
zTh}Epk$O*@-+58s>D?0r><SU$omcPvVv*{;Vw!o7nOC<Bv=Q*ohf76*O|b=iKqa?E
z+^$#Hb%lZe_zb!qeExopz#d0yEyiJB-YUm|yF_HzG^em9j`~P-b6MM=@5v<Q@rUZ#
zbwvxt<8fj(`gF}D4h$3)#2$2|i8VLR3Y^YoG|b>|6tJ0&Pk?d>H~94F6Fz?WgpsQ^
zIcw4N9bR6ZF>&Euc!e_4`0mu=hjWYfCuXnB3XBV|zJc4|?I8$C^TZ`D0m&hA%;R(k
z=)PT7xYfd2Vl{Bf4CTPCZG*BZkQG9eDi}i2H~(8cGk66#Z;7-t3R}pEl1eR!s)~cC
za@xIc-eTx(6awMNQ>36!l@0Qu5?FgX5E+lBttCQl93~814|w6N%1~AfMNKOcumSIV
z5Dwl)f$12Fkc|6zo?$Ivby-H*#xPFs{QS<XOy?rFnBsH`hY-ZaGTO0Lm4Wj7U@eUI
zP}%@e{+rAG#=}U-$Vz!UJ)QBx58vVZbVk**D5@Gt8|(uS1<ML+ry_UMIF97)WEx(j
z4XV10(c3(q*l8E303_nMCGv6cfy~Mb%_(I^^$0f#A66CmVw@C?@&&I~ST2(VQ1zhm
z`Dw#G32bxM<k^JUMgdr-8Qt{;w{ZTW)*;6)5bHgFBjYZkb2aomx?2Y?{K!f%M?41r
zLj@<NlE?Mfe1KSe^E9Eqy}++Cbe2OKqB^eI_5qtL!hFkfQ{wlPW1*h$(}zzO`Y}{W
z8ipR1>oW#Uo(B}NT;tul3g3To!n=13s`qDPd4+ZH7#53hecgmk+g~Y8>fY6vJ{*++
zk_VfX8H-&ouM5VR9*j0BZSXA9cMO@k52J4oO(br3@e(@ciK@7$!tD>---6M0CqXif
zQ>bL@Jb4Nmu0H&F>`HiA0G?``F;FVlf#jY^vgcMk$DO9Al;42vq_Gla*EHnu9G8^}
zfNBk;hzLq+5>C!ptZs&K76uMlduZd4YjRiTnL(x%N@I|l93GB5dRAdtJo?3B@`xTU
z7OGWQ;)fr8U@eMRB=Rha^CQyt?is~t0gU4m0xh*6q1p(Eoq+T%jga21opUajrjk$T
zSX+^j_7uIXg7&ywCJ_Pgzs^Q8Pg;=|o2>u4)jrst5_fdHUN`h3MIn8??#0^g`)QtV
zxn40ReN|l0v*6srgUEvWF?u1NyJ8a1ow39;Say#;$<Oj{A?N$_=@UMF{D{w=KjX@@
z!rpr<ES$$aN%=7v?@tx}=KB`kJym$u)Tr_ts|ChYVXz9r@-^p++S|*lzxo$Z!?M=l
zcI&ZP@=Pu3jCsz8HIl*9tS~5wDuCrF0bSZ>`uRj;)^mVp1`maztWdR7*fITE;9z61
zIU-4#luzV|F+m$jp<%7=^=DP#01ZRWC4U$wEef=4!)D$VE1+Ba{v;{yu+9^fafDme
z5aLmkCGx6dap+_DQ%`}1ei=}lTD<?}JDi_R$g=`@R-&IrSUVxp9#h}bEqJ}ac@1k7
zyd?*QS9I$erMbYQXbGANBIpo=%P1Uaf-F?LiqDk_E)U9^n`fzY$SkYQqiU)^78fe@
zKH5>~0iGsKxv*h%>Y4&cBaX#6b)ra#hT#Fy45`nd?+8EiRMa)g463$9QCCDJxWkIl
z@$6&kZXLq~JXWcf7g%Hw6xG8E6JDsUR}8m~LQXigD9a1fr!$JGhUGnn!%Hrn8?7~#
zWy0n1g1)=aj3|Xd6&o*V&|V)jIU~dmMOf6!%S&+B5c9!}sAss~1-wID=6HW@a5^<;
zPG{sz3x^ETx?)}|)`juUuWWxMP3lpL6*UM9zFsdF`VQ`!_mnj3E$XU-JvCHE&x0t3
zfdB$W%w})!2`@AW76Fb;@aSxdG17Rp8%lmo3n57Ai&Df1U4|l|^bD_QR0O&cc0dIc
zs0gEYzK&eF^!<KoEGvQTgx3+6ZSo9tT~j6KbtQ5?zXI!wQ&ZtT{h#_z;+KE$OMLU~
zchFf5D2@N?fAAmRr%x~VgTM7J<NWRknewR047qlA{`gz`<Ny0V#&3W76J}>&1Po>t
z1MfvxBr#>e2_`{)d%P%#`Js8#K?tuaRUDa;@<mymv1V9C>|_TY?AJ>v6{7B?q`ikn
zS(HqXn*{+N2QQd-JmKQ#@Xm7i@f36iYG!Obil^I;MSe$IQv~2UK6FA)o${xLJ`>?5
zW}4#b(5)7I*N0MR9$*}b3)dzZrP@iUbew^U{dPrny96_}C<uqD>pG!qITpe|q`7{6
ze&!+yL;)<$96K)<I%4s3t6Un2*ceU?PECPRo1?4@Os27{4wEw&tit4h)vb4*{iyv5
z6;iS-WKsIQN8b&ACqUhp97S1?r%8=iR(q$Sheecm4^RaG?&CNF{v>U5h-FC+R@yvk
z0|7$<FwP^op(plFP(_uR%{;>afZ+mu0GyK7WJZ6xVqKOH!;+U38cwv`pbeXiPfiAk
z4O6hr;(F`w@^THK8QvpQdX9Jqkd+m_{k>n}-7kKLKmD`6z^YsP_%gvO3lELc4}XN0
z@t@%`EHIUW2JkY?$h^g=`Ud~?f8*c8pa1GF@E`uee~9PLFHq3X$Qb)TydU)TASh)Z
zz&1dCPEfaF(@;E}p4cG(Oyht-JaCHUunr!$RJh?zNTnBya}NFOhKa#`1P`24Fy3m-
zN_h^nRpE0i3)<CMT(4KG^9(TTP?a^>_RLuhuV~@2#aRdofN$->bD+WnMOdF=CGI{y
zQ#2yMj;3kC?MizIig>)5w!!&)LRzK!k;w_xyqzPQwYw33Z#b#_e15{|bcQi8G-rQi
zHaOHU;&!><g@x4T&z~`m6TBsg=b@)u(s>%`Ie0Idtib3p{N`rxaW*JES{O6HISsE|
z;OSp}SD2K5kB{)E<nBVKYRszeG%qNv!)ghCfCrm?2h}<l8wr~Z3d>3^PUkE#O{Pha
z=bPf5kD3p0R~{=fWKQ!$Ic|=K@dWn?Op*@{jeI8xmY!l**o4fI5Qq(kxx1M?*joq>
z>pg~X#I5VG%(U4l(!Is0ZD5Fx1IjB5{e(aN%irSC5BT-3e!}VeExv2MhcT2wG))VN
zbB%xLKl2@#Zy<cGa0|4va9Huv#}|C_?mPUq|E>QH{?R}Bhxn5}{Zr(59m~aonLe?L
zAm=qvo9Cc)l)8ML@e-|wv9s@bSh;1D;ux+IAC$etdp}?DGo=Ix4KQ7TZPRjmkH&uK
zgT|*KzM{0o>MU;E4dd98iN$MH*A3^tZ7-w!x`&afSZrXK#LV+3i#kNjJGEm-*zc$0
z=Xs_)w`FF+KkreN6&Ie+EEBg6x%KMC`(+#lstGu+P|83ngSI&lQDa`PMkd1i7>gQ)
z0hgB-eEjecKmPVdeE9HzG$u<<uz8#U1+rj?+GG@-Qnf(Qz!x=;S6HC2Sb63c>R|6(
z<qzAxP)#a5#8Q3uc70)bi5^v1V=fjz`4FKjrr~RSiicrh=0L4AOh#b~RaGXzpIgl*
zy@`e3ejEl2Ll=nqy!WVUsXnZs6yp{LV@Yp!0q?!{tR>RT-Xv9e?>DzRXB;C%I$t>i
zg2q}F6uOWVIW#m@>+m1_KmQ5-AOF!mg;fSc`-C6<;9tTo{>Cq{>JqDWu<HV44A#}d
z!C-Y%x+*gTlqsN%hjTOK+)X1sJzw!({JZ}e%Jzi+_n-VJGOdAm4(2ZuLhj@tHUqh|
z?iMW6las(us6_MG#fznLyI(he6TX>6fEuMvg+S6|KJV9wH9JMbb3;$~iE|YEtGOgu
zUDxoUtsJcT{=ubi<}^?eGQ1RcICH+BjuW(*n)_#7twne1FpNV`W<%kLykIZhv4Be9
z1EUm&2Hh}ER4l?6gR(3`mV!FK{=^|6{YiY*^UE`ypP%va^2}s?D_|X)-B(r^D3_7K
zNh)MzjjXJZS2gmy3cO-zaJHDbujmwa?O(9D7h4FFV+}#PRaB{!X+J{5wB9VB-+|5Z
zJYyUPn3iPJs;ba(PL+IK$WuaUHhL}u;V7w<*#YI&AUE*|T4y1JW{<V(TUz>Ev(_cK
zXCe$!56Pn@@h55azSf#t{WTHk&y&U9|NH+4|Kv~q3_dGye)m1vci*9`Tg=YGtv$S3
zDK<xe=##Cdc`}*;;WGtO6wp4yI`z<H1>-D!{hJ^02Y>j-`13#i=OGLu%qJ2?PT5?W
zLl!YZnrk0^o`L|6&(yZ30B~cszG+JZScn9lrm$DavgA}16Z_d|1OlQfG0osHPXqdn
zbNmz`K_u)>P+1h5kJc<=trQQ}UvCKxWw@|w!{Bk$GGA-nd8!gT^voFQp$z5HHMQ*R
zW9(=<wwN3}3=~MJb&jGa@$~eLOK#}^8_2Vy@loWIT4K`h$4~h2#~<<O(`U@{j6f00
zH4N>#AbRL?WJQUrs-cStSy987JmeM1*(8P2e`#9o+FwCPz5Wp!lgK*UE*%AZSqsY*
zV3dk3?1)Ki)mOo`7S5<%)&<&-fXee6O&xRmjwmI<I2f^5mxXKajqsjw^CYr(Uy*gM
z@yO5=6r6WKu!#AaOLqGMx9+|_Ij9lSV0vQj>1MB}it#*K{Jp>T@8X~R#cz<cXPm$J
z4$b)+WJO7#9rFaZ6@cfwy_|x}?FwZ*Gz^Noz%-5c>9;RfhXLKt;g|o3KSr(zf9Y=>
z{tN%P|2+P|Klu9)T_p-wwOOZ7w1DR%HS^@}=4ryMV_qjs@1V4xcE|L%-AAbqlmj#j
z12aib#2O`z6Q!{8Qc9h;^^cloz(;`OF#+(2ey4%SsigfQ%`W}7Xc|3Vbh7>pbB!63
zMJx=slP&wbISxJgtIQQ~$k%Oy=G4OQeA}H6j=*~|Ukp76fNzXOo)HvsK0lE#QOC5~
z@?4g&y3e0K<ENi~!f$^4Yy9}*k0hWJLY4B{?aEc9DZxI=a^z)&yr@uCHL|LLHiQ#K
z!B51FVxoM_1)jDy3aPjN6v+0v=gWJKYd>NbCoJp2ft?mb!R3QIh>CXye0l(uh0DO{
zAtt!V4eGi@RW+OUr4vGnCn_cZ;p5n`4Q;{lfD>eMHf?y|AzIySurjR-6l)1%*Hb`b
zMzk<hQ=@7dK*!|JM@=cp%3qcRUEg7zC(?xMjG`#fv^BD#0KCHg@&EAe;7|VipX2HK
zzloxKkGv>wyS$*kUEr1p3LZ}l<-HldqA4hPcv=?><AA)#p^XQI7yQ5e;on1D<@j6w
z%D;-XX)q6zT666tc)#NJfBF0PlRx?Y(0nS;vn1SK-ZL<(R6^M-&qSuux~}M$Dp|2D
zX!0Bl6VJV-tq6du*cPBNSr@>rw6~Q)+199<hN*kO2l!q<rJo}nxM3UtM{#wTA)1q-
zZoYi0b3a-G>pY{oy`Ucl;_`AHby1-{onVR*mYFRmC1fx3*hA_(w8|I?xnfx+_Od$E
zWsRmG%ubjaXkHQl1whMrceLA<X~yl=!z~s@b3xC#Li==v%1jKF-3tgM44Vqwt;6&4
zCw%<)0YClpQxH%}DQNF8P817Bp*EEAqK!gUmM}#PQ&h<F5*g#8QWugU?+r4qVIeS2
z#5=#H{l_r(js?_y19nWnE*Q|dt|SDQmpcCa_9m5EOK=UtF7Udpf*aK(#%-#>$&Hbk
z4U$GdL7SdqnG&->Mxl_5f&Q%+yk3_zlIa^Mk6xgaQr5|%?*Ylw9CglN=z6BNSOaF(
zwk^t%Z2AB2|MGvs=WCC@^LPFZe*E|ezy7m-f_3TvM<zzi=B3G77~^0RP#9odSIpfF
zFP}f5ZfYpEplwe0FaI0=Ccb%hLUDSB-g->ygymuZXYu*NZ}If*6ijsQspPk43K2_C
zJRd2@4Ty<MjHZ296QhM?v)HWd-A+=HFGzs*vcxi30Hsce*Njr9zI+P{nBDh1CA?b8
zdyFesCncmFj!7;*_)s{r@5IAM+G}%a15unDyks}s^{14=x@?0#opS^iaKL7I>F&qG
z9}vPqmU+RIwM3=J^sK5HPfzc*aaz0mC~N0EFb+LFKYzlfk00>i!$<t|(@z+u5gFrj
z6raXscQ|hbSy3P_Yvg5xvTBg!C1oPeEf08MNpXDUla#c>;uXj3jY3Lp2zlVgiz`iV
zh1C+3gP1mkOJPN`V%tYxI9OL?+CZr^%RQP^M;lDAB?=Q*Lls+qa3G~kNNU}a){|h`
z9zv}ZldJVq-U?u<3xl@dI2JK14>rZtv?XmURF83#CQ~;J8pbrV*7(yu{WE;~-4D3D
zyx{if1D5Lxnab=|w?+Y|slbBqXzLQ%FK~9mt$W724wx@Jl(Eq6f<OMFKg7TMxBnLY
z@-Kda)+$VSgJ-Q!>lOd%AO9XckIN>HOEz$r2jHP_@IF?=an6ycq7(;O)(_WeBP4hY
ztd;c|0maGHhSoj*3J#+5P|D@BS_;pA0*b0ceWKFuubCX-FoG_rG}2vi{F;W#q7xVx
z!1~y4dV$9E@B15up@Rp7;7EaorfxO~%`%RAERf>LB%*v6IzWTC)bboB#x(b5F`b*J
z2oU$#<>eWdmkTb>&lvlGO--uwa=XP^fpVefJVRbID2f(EQ6no#D4ntW6Y)4Pwluch
z&flTA_jPUWs+W?|%R(j}aefoIe@Ku_n)8(SNmX7$uc)f(kbimq+4P-oD)Pb+k(mgj
zinXzO?1(-`Gy^$15~D>NIUXj{Fh&O~LNSf+Q6#xE&(A4AiOV!o%*nc<EDE$tPwb#D
z4kL!X2mD8T`uG9!c*8o6oRdgin{`o`#~#D=8Lq6L?2LXGh&^-cF?Sajtubc`fA_!n
z-^Jhk*Z*tyvw!xdnE&!e+%7-i^FRG3_;wyp=><2O@h5-zA(RsfCFprR1uwZ<C~i`c
zO3VX*90{jWn`oH#Ih4Bzg6G<F9Wy16W}rEGpJJ=T<oQygz~5WsVpob&Ri2RNkzVL<
zEW0*MW8_aKSx32arxSFZK`H}9kK%okGcoi-@Rlh}p%HCMg*l+Mm)UWP`*RouE?VjX
zpV;|)M$?><fX(9zLU5?tt;5UnC%im=!lzFkae29rlL9_?o;#)n5{){`bL2&hqHK@_
z0Yw=M!ET3w4-PSP$36Qs^DgaQh>&_sU3Sjldc9#9r;ydZ+7($&w7%k%lFjEmgLHj|
zb*0F1=M~DLrmXEOBkvz#mV5Zj<g=kjP3`*|(-S+ac12MNUwIY;M=1XuKuQqM$Qs_d
zP>j}^RO!v~992`paIs16{T5fbn~xajK8_<CYm{1(H>W5$w#t)`n#Ufes=zY!Smznm
zQyoHQSF-rrf=PRP_7+)2MOs(8LcyU`8vmt#<G%qnJN(1H|9{3{NBmF!$NxjL-825M
zHTdJU#4rA{e+%z^ctV*O{4f63{|<ijm%oP5l&0bxu@unuQc*rBDvBXyy7!K&p>z~`
z#F(Yd4C<yvmgk8Eakp!d{X^d;*2W|_)_aGts#(KSA>Qpvcd&C7!_YAeF#WYQ47yi!
zz<IVB^IprN@08wk3%PXOBak5yJ+==!Z;Ork2v)OxvCI<z&m6D`07Xf$V_8n#<x~fY
z?)q(Hz5^ATy>?eDmNa0K=ct<o?dcpZnY}qH^vB-8JBy+3@yu0TK71e&yJ?zWonuyq
z5x2_?%RC1PU~MRIJ}(;-RgJ8ys0f5;jR1rwK#Eu>-ZS!K-v16YsV_;ty=qDmtI+o&
z#&IV0K1U3YQW+Q40^YHmuY;#zVh@5+9y-&gng&@>GBW{!fJ$%Y;DJ(#t1S;y#m2k9
zVN^G5NFXJ*ETu+hrLplldG@Z@71J<ao~R^`(T1v{G!1Qdw!OELe!uzCo6japn^gLp
z09?ySv-32<E;BCApE30VR=dEu1(r*?XiWrm!#u;f6<Jn7E3PqE7Wg0i+y4`szxxIL
z-GA@@g#X?D_J4za?|=2b#NYWF@A0qv{tx)0U;ZAx|Na-KbOvKG)MbwU)_?V1$N&Bx
z|5FOE^W-g6o^V5LghrTrXhJZjIjMOZ29B9wvVE;kw=JqhlC-zw%aYxa>=)0PPlsuq
zf@dht%l|KLe;TV>vflS$zp6FA!=BGRWB2K~`!?@DUh;~gNJ*9@nTjl1v0z1sfW+_w
zBoX2d`H&!h5g-VVkG5e54lKczk_eHNSh57kqAZCNNs$!q<(-F~`*fc^!=B!8&8w>N
zp{mxq-hKAz+lTZ`0o~`Ez4yEJTD7X4dWQe=|C5a-1YyL}iZoX(c3U;IcqpIi%u*Ll
z_v`ZN&~tN5dw+RuF+s-lE(SqFHXakFKIPzxN!|<E4i#Ot+BQc!V`tpYByq}kvO^fh
zmXPo=+`u^V=DhUY8*e*S>zul2kwIt~CE^h~(_M?LGWNOsBlh>><hi=GwT;8`8O!;C
z`E12(v8JlKz9`W)4f$$C=M;$1AOlGdCIpfFFN|Y?C`2O&jXhwsKkeOq9L|KB=|RNu
z*~TBplNyF~S0(!V&(w8g7md;`FJ;9;Hziclm#2MM-X?EDM+cnsN$$%o`Zm7GR?dWW
zRk)&bbSDa>v}M|ieUj5JF!xzw+`Mh)0_wW9A`i+|PNWP-(rh4txowDs8(^4p_PM#n
zy>zX0pS%yl(B8Lo0DFDCrmCpw%2EJh7pip32U;uIR%7%=%Mdgd=MDd_zw`IF|JHka
z<E#IIFTZraj~|V>dHsmPqiY=QA2Uujbg_|9y!8Bw#KCivT@IY#6U-Z*Wg91D95G<q
zb(B@*+Fk;r6r^dwcr<lQ;|#9BC1d5zTp@Y^jH)Ow%AWQ%>F%3!hb!vRM!vKLd~U31
z+P-l)O`lW>zG}S=jpp!B(b^rVE)TU)_VvG?rw)NVeDXZ+!M_NcK<s3risaGsmu#*P
zzJ*F}<)M5-yOfe7Ni4R`HMD!@U^wUgX0_HmTHa%#W)~Os-kh0iS{LT!c~1v~5Xdkl
zN)m!FA&L{r@Zb)2pBj0Z0s1Vq;%Vc5w4BuC%{e@&`aDe6(zTX+&Eu4;_hS?CTDPp%
zON<la1WV908chhpVN!RK_`B+1Z7IBP@>-g%)9yNBS0de3ju9@`<FC!Yu#{y<*S79n
zEm528^NQ_tE-i+u?#X)zWm#^*Dd8mKT?Iy??7|X)Ah3pAN=fHjZyz#9f#rqAdV+}1
zwLn<I5=0??`@j0D1Wm<{-96@KUb@cFXw0>nx7a&8WICRrTSXUwRvl&2@~dC}dFJ{5
z8X>+*82km`0k+wA(><`&rZg}t%Ck`z<8nT`L{ajD=kuh;v&mWd_9>$~0%ySdL9h|J
z^zUBu@9uIq<5@P#vc$Jrc^$vY)V)`&n_=T^+ggZ#YFdlJ3ItJ-kd4Pk84m8ldn|*4
z$*(K-$W2w-79q!c>fgO@Xx=(_uZV&?{n4hOd2PL2A^RreEF1NE!rJCdTrf0L{;=Gr
zyv$g+9GO;*B6zvzRgG;tArWCnkR(KLW(`zqg?l@tK}zfWUOGVf6o{XfLC#Hn3RwE`
z_>m#bVY6XvV%0(bl(wFYF{Ve4!=}}sX&Ni5gGHD7)5b3zf88r%T4TMavMenKO@ivM
zDBkgyI8Fw6t4)|}^I>kWxgQwjkK92*5Qf&(PhvbOez<urKezvTzX=D1%%8}9Y))`n
zjxMW`D2)2>!RPJ#z1h5*p>1oTFaTlOQB>D5k{x?TDPR1tmzj(*_Kpq^Qt<7!Gv4}m
z!eRRBB<c~@Z{DWnkMW=U&7Z}X2xGoX$A3x?{98QXv54*Sv8pO+KQ{=%FmeJ2p0RU#
zKCXCGHgxG(NIRtaqDSvPvEX~p<#I~lvgc#^190M`%&$5?ukxg97;f`Wb<^~BV6jyb
z;`r(Y-6!Az?)hrR{Mii!9<mJ?-sRs<A0qW%wdoKcD9eI8U!jd6=yeKrnM@{@@DS6l
zmnY`kY$V%y56V%mw2o$`>pIq}6)UGG?XMAr5mA~DB`IN&5~m|%7@?ie0>)Co42{r}
zk?Yf2tbW+|1CieSu;iD&DXk%27c^B(*S40aQ!An<vOP!I62}!~VJV2E1R0QKDWh?A
zNj+x!_u7+Va{@L^O_k^9u0v?M>C-eL8D$uu(Oajl*;sp<tne45ttzS_$8^g6ek?%F
zSM!Ya0oc>#z*qBXiq(p?u2IH%MRAgljYkk{uwKUCP9rS72{FL-^bc#<L;Xe*2CX~Z
zc=L6>^y6RQzxu0xjZ8G`j$^*`W3MntBhqY-$|&CZaF74=@BX`-EN<}X@lW#0Z~PzG
z9WS``@-9gPjia14qC^WxWDKTf;x!m;y%>$5uI;;0M%xBxAWaj}G}+oDm)DD*7cf?s
zOc~1Dt_!rqL&Z_(WWM7|*AZKvzntsxSRUmPcgAQ&qe<_rToEN+>D?Kldc2o07L*tU
z0b_@UdRnM?In+lO+OB1luc@3zkFesPwukXav53KqZQT>tJlkruqU$uF3<+ewI2*Ay
zof61kYaMzoZ^NB$jc;C-%oa;7oEJ4;tY}(Ir$Kc#k-lE8=<24QV;KYlapJ0Y89_Wk
zMzLdab>6Vii2l9!+qd~he;WOM4HWzc8jOcyGx37;+ETu%zK72)CMhIJHI$Ww8$_Wc
z#tWjPchf)3AP!F>-wW4ut;<&$64(CV`O`Ox>5?Liz2QF0D_oVypU6p^^lCO6kFNh@
zf34w((YaOvE0d*BTDd+iD~`DJyWOd5Fl3_<$JeiO^7t`9AW_(Aq6gDyr7%K*78oMl
zdh1;tpDd6VcCv)$Z(nB|NU{K;&`_)^{=dKRr#M+$2jHEPpW)HV|5N_hbKgZZXY5To
zR)r0Rq`rqVnkJBhu6W?tDm*b;-84N}oxqm%J@3_3*1#VKTk~y<RUIx%wC8nll)GLy
za(g`qe1G7^5Zdn#o3#Dsl4Jwc6$p&jHU|FB;rj78QE%*gsGE5_l-PuLPd?vf4d4p_
zMNUzcy@3kDu&3X7+PG?+Zy+8m(&-2!amvI|)eI@hHy_`G#nzj5N-v99uQqJk%`BL>
zsVr4YRg@b&Nht{9lsL_Z(-BcRwqc$O@a<;(@6y}%ZSzy2H&B0&@!12z{+ECDU+6ch
zl+re3=EWAM6CemARbA4y4M7+X1YxhR;ROtajgVJ`R{p=LD&0c0_bA0Sw(5zJ?4@Vw
zu`Fdt?H0WMx(AV7_98F;-9NK2zBkY5<~hcNG#w!3wcfgad-hn)qZeLy5v?uKSq8R}
zh7d>raTp+tLbVOLQ{<}^AHM%Ss;NmL$vBkU+}Yu|!vl`?cR1Lc@a5O;5lB0ady|4M
zf3D$RXO~AOV?KWDReA)5pTz_q2;6_0x#?6#krz}hzoNC>$Z4APhsj1MdT5wzp2R0r
zjj{4*-WYfzHkAIl>R>X4e$M>x`94nrEQBD6E=gN`%K9A6zXyLdgLgk{9QSi{)pdqz
z%L+Vf`Z8hDglwNLeBwG+MF{rK4MC&lEDNI8BHc$Ll(EI4t3L1l^UMVcmn5If=j6^n
zwQWmXmX_^cwE~xeLIwm;Y7ynBg&V|4>YzJ-5zbQ+U}Vou;{V$U{rtR`4Q%L#MPbA7
z`-UJq)y=&${MWm#W3^uOYBO!qI%8$kETohcS&r>a_b1n7zVfHtkI!xRMenO;u`?{N
zF?LV$_`G4UJ$@GK;F%Uj<I&)LuEPCnU$1FAuXodLtR#thfym)~`1$kNeqGn`(kn0X
z+yCOX3B#zTU=9L_X)8iuNJ6_1*7<cbFM<9lGF-F2x6jLWZgG7&<>vk&<DDs4ddy3A
zKj8Y&KjSyQ^*Vp)PyQ~y^o5Msqu>ep4ohYy2$gY1m@g$OihM=e)Wk_l90)`p$g;Gr
zP;GK?+u@2&qHArC(pcpf--PTocpS%!Cu1kquK6tIr$12osu>%`g-ud=J3n~U@LsPk
zZus{+tiFC)VPv=mgz)q`LqX4t&}bOOgKF&6n|1iRilXSz-hor+o=m2UCp%83P+yt<
z@z?N)a!dbv#{9x2%Imsug0Pl7!HTA|ND*LwAdE<y2NkClv?)C`jsXTOE}MCLVrI9V
zXZZcgnYT|H&n9Vg#S8j01LbebxTM`;zM!sanx>^~8!I5%xGw&{BGBV7wyH1TcH6?!
zUZ!7Ag~L{9rO-+tr65TY!eH3gN>77(dExfqSz$dNi9P*nsd&R(WIsDB&u<$&gTZLy
zB#8<O&~tgS(b(oSE<cywsKRW>@NV9|&Ee4z>(#0^YFcX!4|e!hfBp{<2O%dH*ZI5u
z_%E|AUqXt%Lm>V)Hh~>S?4~JQSy9Q57|G7=gn#E>{T~1RAAbiCgh&%JN>ie^A&VQz
z#%kzIlO<g%(2?__+_U+r+5YaHgD-f>-6uEk<@MqGT7<9)*Ll9CD2skGS#|?QJO@{9
zx+mA0*?40P4GmT}V$=(b`X=#DKR}J~5P=+*A|#Hil=+4R`0BN|eQ>kkbaE8eVBrQf
z>F&ck*SvihZ@{oPV7<*e&wD<9UlklrhHxC|azfk2%^1_CKNgDx^Vy8qY|g@Ue`u}g
z+LnC1X0ce%wN2ju9Yzsxk`cuzagy4Ec@%+X*dM+dKL^8z!u`)9@xkHhjGb%Yyv@%#
zo;lommLUZF&2(C^$_tvNwfP_?JP=63D2s`sfFzDcl7x6DO@!tNJX3#tyNNsorR_SJ
zuAxyaQEI_v9{p<h$|YVj)L+a0eOSH*5P?L5f^3ozCNV|?Zo1@^i?la<f*$Q`lm>^1
z%Cf8vU#*yd={KVRgs}w%fkYZf7N@-N`8O%b5@CeR<+UBV;|Nu)DOM}K_T7J*b@`&b
zfAeQ))t{wR0n?p*$|~n#F=M@)b9(leM-LuzaemHXIS17uM8~SsynO9F{-Zzh-xJ9-
zFWz{EpL_E^K?FFF20#khw&U@mQ;NJq1j2&)qKGsbJ#m;^y%6Z%VOds`WsWuqgvCU~
zSxS;QAx@0j1e<X2iHiZ+@h7!Sjlecar*Xt|JRy`WB<YRp)2?OTpKYC5W0V4I2&H74
zjTjH)$o%g%=IVsIU#Gca{`70h_vNMOsL!Pd!4(F|uO$uWzNc=;=kJwXqR_VD*oZ-G
zJ=$<RSCtLp=JPohvjwxoilVAfokF!1>^z@eP}d%^VS&@Oao9Eu$4N>UMhGLEi4snr
z_fjt&5`*my@dXz93ViX*?gP&YX0CYtr;R^Qb??)>D8D#-0?ioH=MMZ0YprdlASKFC
zScGA0t*%>GK0*4lJd*G6vwdf_dX^+fpZ9@Hg8S3*X|{#N=30Wlf;K(l!chAC$>Dqa
z?DMum!2pMG@{7xF=q=76pY<`@3#@KiUVh~jzW%kZVN{2%ZS8M0tErYtf=F?F|1I7U
zKkW?g3?bg(@p1`Kg03pMPO~Tq;x?klYu4eK>1YC~BOZ-7JAKT9hYz_He1pIA3;!9v
z_u&Wp*7qOM8A+f(gzm<$7^uUe1NQdzol@{dM(T$RzgC8xft0qpAWf4#!EMY|jp~W_
z?mZ}DY^znz{D~Fxu<b>$iz}Y)dHdl*v-Nt_3ylgPh?B&&y@a8qgBc{Chi=5?T6PWi
z!?GKAc3EawM4z^fhhs>(xL&WRs;UnclO$m}oeo0lZLiW?UO#osa=Gk<6fQ0<C<-Uv
z=HNAp#oXocgi{6zE#*p*Sx+iS2`));ZSsnn2xR!a`aK6OZ?$>e0GmSa)IOhe{77(8
zPkRxzNw=`SI1KxkUm>XK8m+CjqlIVk+op24JhZmx+oq`ii@)-eCqw?C;eA=}%b$+c
z%B!xmRgJDA5Rw2Zk8MGuk#k$0<Otd%hP|tAt9f4D$mbD8qmd)|-*l1t;b}iK1}!Y=
zSQ$lG7f31i+#9d+?QeaHD3p|);{N%Jdk;sv_WS|Q@BEki(VtHE?%Qv2WBQ-*(dn=A
z_$1_XmUHc3hlq~6Tq8}uXfn1!k!3*?g-G48cYH`qOI)q^@{Npt{H@=mP!>#TJN<2V
z90Y=c!+rMm_nA(2$kL3sR~^~R!*E<SUOwkz=f-+g{{4G4imTSpRR^K<d^U}>AY7}8
z6&k&KKUam<7|>lyzFN|h6+&pV5yYtz(uiY}*?2xf1NCH%OY}|1LjyG$jhIX(pUIF9
z$1pd>0E1>Z)Fe+9k~aCFO`_a}aTgaC%;)nyjiHpHb2+osYGu1UEFi>rP$S}WOp=WW
zoskmKx%+1G{FmR$;0}7j**nF)fQElxlK(&PNqv^Fou{blhNTl0^d(*)EE&FNI(&&)
zEB%B$80K!khi9eJqE~&wJWbO+tg@eb%AJ`_D^6LIRHYR+l&+h6G_u@IFpQrsN5{k5
zmPJmNd4U<+&t!Ck5TpCFVDp&TeN`Jx-B?*{-L|~;%B$S}@B{Yt_xR2azR$X=`TF}0
zDI3L0H+FdQ_Fv;mFUH*e_#Tbk;oS#we)V^LmzV$SA14k%Mp|JEBykK;jP4Xwrx0~R
zgCW`9V@DoP%a}j}7!*Rc7!l$bRFgDecXzj!qKe|!$+_EBoTm-HkS$Pm!+9=*?Ul6L
z@*4?y|C!-C8X6p9Hi&6U0c{VpEK3GlQ%`^K+qp=a8!L;V&z)Ff;PnzO-^a!yzdV%M
zut4}{7#iTG<?F5*-lO;ORW@9l^vtI2j{FSw&uXxH!dE-z1C^KMaz)p6Xl3YH#d5W>
zxjcs`#~5T3Ik<uKlEWmktfkV*fcd<{Cdarz=o{m~&>P&K?Im>_=fk|Kzw>9X+s|yI
zh8yjN|DMZL5W#9?UE-oF$VM4yGeSnv83MO}Z0s|*#*-DJ)~N=!ZDTe4v;|2HOW6Kr
z`x6keb>qI8S2)H_Aq6rJq*+3irEXDd+-WZ^IUIg5*`BdAP1~TA>&<hZO&J8PH*er3
zyJs@)wDF);r7V}UQ7xg+jCZH}^jChI-}}99Fb)hNmej4{JNHg_yqIxof62k#4zY~6
zzJJ6=kCxm&J>yZHbN6#E^Y{mEk;WNXwN@2ITdysz3m%-F63d8_4<BH%5oK*%e<`Ky
z856?da^wa~>(e`)lY98Cw{k9nwc;zUTG=9r5SBkH%`zA2Z%deOSA#acLu*Bm+a_Im
zE*o3em`q(_caR^vv=WDhoPF<24-pV@pqJ<F$r!%|hwI2M9?#Y(-C?<=DvLgBw~2C(
zbk{u1Hi^M{7s9sUtX4Mk^mx3{DC^aWH<87*CueP^=@I42<&wqhg8A%%d}WQN(k_u+
zl`K~)>ZS&xZIw<$B*};*v4>C)#+EjyQwU+%My-L8Zf;yu;iSX7krKjXfW?Nr!N3Ni
z=sehG9ltM6>WBTFt3oWWF}D?qw6ra4*U>3W9J(;iHBtJC;M3xNt*zX)Cy3X|5Qu<r
zw!yP(otjt!C7eajW&pPIvQ&T+ferKe>xlk7EM2nZvf&j%qLlaQ8r$|`G+`9=GFg{T
zcVX=PNJsqV%g9nnx~fE&j_D}2)w6WO(QZsr=5(*#;-y=+`Sy2zz&js(!fcWAt&h(T
zA3m^Zz+p^!a6p#DeEaPWc<yMQw$&stXrnMvV**KDmpq=&`RM%*dH%+8+&aF=jo02_
z^iO|-^L2&BHZX@+uq}Ep|J#3DnvY?A!j4}0sckT5Bv#ZjO-GxO_NPKRHhGpjUt5pY
zisNla5aYDZe|X;Ytt(u*xcA09e-_5{oaa#(LT{k9$5shlY%q02Zi^%y@H$F;zV1l|
zi0#nRKA(^qa>k~qdt;g;mK%QPjrq@A4q<%{<OXhV!D=z5t}3)L7ExZ7%;$5;GIxv!
z&f^Iq;&jB`!68xzI%DYC8X+t@D?OLJrt5sg&ME*I0WuKom0T7s6}{17U|{WB{3v_(
zHWp;?rB6=?4xs`*A*edlE67BN1^X#w@iCuP?lp!U+-8&S_TjV9h%`wCa#9Y5GQ1CP
zhRQP}7FJL}SZ^T+L#yYOrFgoK%lF_uzIA`Tjl$!he5v@dBCdZ1;V$G);~)q?3RK&G
z>DZZMNHkFpBD6&r|J0X%oN*ShJbTC&-*}!cfBAE~|L%QGFG@c7<UY%yU{RE`+Hg@9
z<i$R3zxM&xqnJ2sh$N)R7_AhYF-*rPzx1;|4JJU#kaf2vO+(1-GX^3Y<QXpAyPwOa
zEDt}l*5s=-by<18(ZLPIgN9#^J2QQ_f7ROY{g2w`e8h(N^D2V3t1x}mTd@YJEX#fa
zMPbZjJnbO@PkMCQ#@F-c*N*pFwaTqGhw&S2w{l%xLz^4hT+gy~+RE}K%o|_1hKA4R
z4b%`1W%E!A^1PyLt*$^*)hy<7iuIbn!3eYw7!fhq+2`o`v1Rm>hR2UjsB3|i5i*F7
zj)}p(f9pRBDZn*38<)Q{{(S+y;aK{)-+Hs#Vf+spzwhearyPAx{Os^$dny($%wTg1
z%GthcPvxo?+qN4yDUZlbV=HqcgXoe6_UY9;UupD4B1#gH(FkO)B|<QGnyqU~Df(hS
z>o}r3U$t)&x#}7E7wKGdu#Kp#Qy>YXJ;9<tVw%c^H6ctVW7ZcThu5xg>+TDz^M;@J
zLpOQ%t+#pZ_HDlYFTcj4vl(%k5(kP0XD7(LW0H7`%o~KrDYc<P5hc*5il*);+Lqm;
zYy8+NFYx}^w_PcD8`=FNbnj`ywxAfisApkw8p03=L6VJ#lR-%3uYTp_?P9hPVwPLp
zSFY16?W=-*4*goYd?@%oeK^#EKaJ_ZkzUy2$ttdRB1K@maLd?H*Oe1{-ZVzXQH;6-
z9DZ662MlHD4>{Xv<cmK(&o%@!U%rMh8==zqe9rm#IkVY}yeR3E?$ZaW)sppU2};}i
zrwp0w?6P}s%+2TSu(!7Zy5ndZu{%!q)_31#T`RN*kj~2wLJKAhql6%c5JuT*oiqqE
zI74k5CQiEW&o+L#x%7_KXS5&i`%avjC(=Y$c~e(y&|#ldubOMROQ+Qp-iPzV^1RrD
zdD=LrKt>ow^f{j4z4VJ@V_t@pCjWb*D0VW`UVrb>6K~~7>}UL<_tiWvsNk!4!_#hf
z9qEj`eet2oCp1)Ti$tP2TNRvNoHGgq<D(<$<&1pxn6}8d{oD)WdCiU6H>j(cS6+LK
zy~%`M`1zkk7(-LneDv@Exf1+~w?F1f&%q0aW12k2L?KFOWML=+!>#Mrxs{I4TJfj;
z$S?7$-+GUt?zl`rGk6(S&dp^&Yct6C_yoB1d`5i&*n7sq4cn)_j1Ti%g4<I`yD)D&
z@Ww8`qwTMG=YK;<-E|#-lzsBP*Im4nq?gib=xtZ8Jf_aK;^3;m<MG7FnmJ$M(gD4x
z6mEV;+t$5MiLIpC^*ou3w>-qBMJqmKF=RHd3=WnOscY?8U(V-ND-X$Jvd8}Q>l_>%
zb8zjLWRx<^5~OMfb;TEd>~mbqE_iUdWWFvjNGGJyS>_AYfFTGZxaw~drG!C@3?rlr
zU0<gq$}@Oly=gT5EJXxAp4muww|(=k!S_DHFIUQt=M}26h*#aUbV@rT17cuwx2rLx
zr%qAIpfwoNk)~tfG{Z=ncd_@dnULO4U=w%RQMuH$t?C3pyw&(Q{LEEjXn1@}h8LGG
z#`Gb`RVS#Q8|6s*l~(B1Ru@<6HFOn8YW3UbN}7625{`*vfJ`#7STmhYsOpx}lMAxR
zE{n4{w_m+MUFNJ7bK17y?(=tv!;t^_Z~p^6c(meTQSq<*_~*Gkl_Ulh^EpA9Qn!Y+
zG92yfGD^oRDoqfj(5!6H&KrDi?J}T@LJJVi>lMOd;e3ukSa#Li^0NEx_$am^RoENv
zO-^Dn)x$i6z%vq5Wr^_f7zoBYlV0uRi9>0y4yD)-)7iefHBFrpO8dDeOUcF~j1=fC
z6E*OF`Vh|_mXe}avtH-!Yqlvd9XX~AED(0+!QrnoVDYq`T{~zT4*EQl=PUCE^YHnG
ztzBK$QkFTZ)zV_2W;2T@cMag3>R3AutEy{+4A|K_<i%Is;ONF}#*+zY93mwoaX>&q
zwQA_Ph97(V6^iNyEQ^}XHIw(bNg-YMXa6(?+PbARj^4%D1sPbQ8-x~j7zB}(hVy*r
zpK<)YdiS<`o=-7Mmp8hfbSvj$t6-jNON8$G#vWbT=&cU*VY1UpV{KYi<fY%Y@y51a
zu(<@<w6S#Js)A1o&-}HA2FHdVd9Q}-L7T%I$<qv-$LRo6Wo0Wm8bT4EjHXo@x{kxc
zL+YYnlEu9J{tviy^E!8)d!AqY)nDiN7hgpg$#b{wpp<4Z9#c+7bX|;+3T0}3=}T{*
z>u>S?!%z5c>w+)dy2;)BF^`vzk)vHiW4KuD^6`F!9F6(L58h$Do?+0mL1Ojb+<e+3
z>*d7wCa<O)!-FxJ(&2oJwzSZ`a+V|sYA_F*sov&_@_+B&RZj|Mg*ZKk)b{0Vzx%Cx
zjl0iIDc1R#s<IFW!^Rs+QpbOF#hSg!i`fXec6jCOEE_Qzj|YZr>*3rroa&I?<Y#?U
z;JphATbOJ?Af6VXD5ZLA)_gu=adFOiF{gH(qBURzO9$8#QnGt+jk_<u!gF_DWW2La
zlEj44V#S4SX&YPkx%csXUby=LyL*pG7bWYS3vGxvx0Sx$E5<mvIE%bD+R%2&9dvG9
zr8ic=r)6TE7{8C+y(hnZRpsK-#(JIihN5-+QYN-Z@77af3>M$8)7{75V9~sB9Bt*Y
zHutwNRDL6MUFQh#YF7pF6`~EQ{HhRbcq-ca=_t!>^t?Bi!w}?YMoK%rY^9V{cxf6i
z3S(N*Y!}m3&^D+rCW&LJsvwev8`rOKF}vXS)(u{J<28Qo>)$|^6-_M>la&4aeFTCJ
zzVjBkE_vgPmpDA!<5zz5t9<LD2mIsJnwxvOyn5}BJtH}-<|q!Bid}x^pMRBi-+vEf
zGNdqMrbYx&?>%5#o>%xe-cEkng=nbj%7#J4^u22%FKs30bA4C0Itjd<T+=u?s>bnk
zBS@087aH)n$0w89b}RJ-4q>ROg0jdl+9tSV5Hg-jtpW`j&LwYjo_t>#Ukx6r;uymg
z9?H|#T-C_F@x<)hm1RNc`WJ&B7%*&Zyzr~oKm2<3fX(@Q&c*o|^Z5luo?B)NZK%td
z^<qJuuQ1AReElZ3Uw)nA>(|-a-zQ2^;wVCQ4XUeYtAa9LQI{2~<q8=-B6dF`26;Il
zH)!!8BgJ+u(BU7$jVJ4KfJ(I})!KHR&p7@-y?amFgg@*vRaN$_2wi856a&xW@^yV>
zrxzDBh9pUP#?B!+ee<9FY#(B|W=kvZ;HY*3>5#nHme+PU`8phyI-{T{HiD-<%=45m
zeq5#%*}jesU(3Fy>G1H7S6_aa<7?N@Mp4%-`v(W4*_b!K@I?*|4nPQg;>%xVU93rx
zkfP2xJUZa{J9m&;^8VMqOEX`xo-cXttKUSbhClqXZ}O*q;&a>y;PIo!ymdC`d$XFy
zg=GKvukgk%|NHzq|F{2|U;4BEGe#LGQ&782C4CM7<5r8=`fs~(W)s}n<$P@Tl8`y<
znyuar=l3$bh&2}07}yPCv^|vr*MvNTh77%~Ew5)V$9}D;*0O90X{m{`z6p6~Y@dF}
zTKT7Se^A;yROA?6<4^URH_r!Pi5YDkYQS0Ls=-4Z_REL=Fi}3tL(S())<s3zDLUg-
zcdUmch2i-6HE!Ly#jQJc*xB7BO=Hqn(lrHjxni}rV6m99SgtsqEm-9hAKbgoY`(-8
zdPU~TUMsNG;%00IvCZQd9M6?2HVFwrx_nhgAU?Z)^gp_S@zcg|G1^lg?{)aBks63}
z3ZJMHv`s_dIGMbgYb!Ei!XU&0EPbBivbX22=|hLEYprCv6r@SQWIRF&+x=~Z*>!u@
zThHsOlV$1e$9~6G^M*ZszJfQbgxZRp{j+I1in82<G9qSOD?WOBMpsuT0$#jxhaz90
z5SUo<&b#+`@aTl9l}z{cxPIdnNfHqTA;0!d{ux=6QgscZz%pglqM>~Jkh*Jm@wt6o
zxOvFw`HCMrc+AP_oX6*9y!FAmB+ox!_vT&x^}qDL=g<8w|F=~6iu>RH27mjn{}sBn
zavebykZSFeW4`ttr*L_wvgrA|47Q&x%SOGl6<hCRxF-CeVqqg*chESpIO$Y}Ju9jQ
za+)U_6WqMpDpJ>W<m**0M<syRag~LxKqAD(vmJ(rLyuet-_n)$UX^ruc|-Bj%Q>X0
zia)Ir+qC_Q0YM<0M&We8S;Z5%D8Cl1NNve-v0!#FV?LX+TrR2Vnoeojwqw0ovsla-
zk23c6rrbEb!NL9k)A5*WG{WePuIX4UE-2RPzKB%1j)|`8XgfvQX_~eJ;RPc6p|E{8
z8Do$}`WHa&Zud*ScWVb*>PO0P?ta<RzWYOx=RSP%FRD`wow87bFtT(Z7`q*`wt@;}
zRiU-AC1@eYvZN<8^AEdORDvOh&Ndx)&al+Dt~8P;W;_}ZJKa3DN_&H1*enu%f^0yW
z9T1{**XLWpFvbcg^mkz|*`M&v^);Y6O_`UpZHG0ohB%ZK{t!gSNU=XY=Gx7h%uglj
ze8uAv%ac9b->0lgRu2oFd;SHYrsIS6-s9f)-r|j~e1T-Lk66tKV%VFEiE4vL01iR%
zzGJ%9QvL4l?XZ9II^m12^4I_Rf6H5MeUotdBDdZs_~_k`mp}J8j;_DJ&;DCK$(z6U
zukfwk_y^qo?ccJeq!bt-F@y*s?0*2-ie6TEg;56W@H1AQE$uNg_yr4i!ZaWL7-Pe{
zx~eGi9HXo_r~#5bTy!V0+2qGODrb9<c}kcSb!C(HT1Y~NPEV2;gF@iKSbS5g=?}}{
z61E`8mCHlfcj=qYFJrVWM+3vX;H%1Y-B@oT2pl?n=gRgV`*)96--uSMmMfO?1&i67
z)nZB8v}kQ?%Tr!Z=WC93r(8Qa<ZyS&?qtGv6eDzl6eF6dpe;-Cyl`-Zg0i#;@}_BN
zS_`yRN;xQn1#RLSfMNL8`=J9?10(D_`ofF*DjowVTzTQMjvsOF-jl<Z>OG)Nb<-YW
z<P&kmCC$Blno?aKI(bHc%L~bmCljBx^z1>KK@ftJ#G{O4G)7B-8A>k=jLi0Zdxk{o
zjd=!2-{Utdp$|Pd|L-ALUR76?PszpQRu3>EPApGTAT5XVhaY@MSLHNq!=p#1RE^>J
z7e<T|=!%ZT#hQ;FJmO(BX7|f~iq+XSIXt?<y2%-(DX~%{6Uq4SnECk<)wSrZVKmy|
zciwtP-5JWW6J{r$kllEWqk|(BvzfhDE!a7DfnWG@{~pz!`qz2)SN<X8{1c22bSRJ!
ziBJe5DC-TZ#2ACZ^sOA;yLfUP48t{_Gi+I{tn8ORh0`qUuVsrWy00Dy;oC3lHGERq
zm;u1nDYOKE?D?UF@8!~6JrT0`Jk*e&&xd)RMpJXSy6Yhj8}RT359QUguS(i)2m>v%
z!OAggo?l#Wu~@P$EhV>69d%LA)iuY52i!V7VrMdDG)hS0h%l72^@?@ZQdAXXRk2vC
zDf8U^+j(ND>v{v#ZwT>Z-fvj!xa1^TZN|ZK^q=#t@0D}^^zrPyd&8S~)(`v733>oq
z=khvT*AcrUqW`z8p0&=f8R9r)GT9-DlBZIq3_Up^1WnWR42gj!)=N^d(F7qvloK|1
z+Ozlx_9bk4KemFABq`(ZWDw?EWi_{-yRIAZa&5&*h5dmv%@~cRoAX=>t8!bdDHj*W
zbVQWx^YQ5d9v0kvewR2K@!;Zuuf98@6=Od4m7nK8{TQ=v{(WN85E_BhAtH(pUBM`V
zW!tjKbHsSe>!14)@4o#W&4YUgIzIoUpXA~FkLkLOv(qyxf|+D!GNR-TU;gudiI2bj
zJ1jo@28n4vhIB@vyO!m8O;OuqxiN+~i5QPZMA4^h;_X6=R|~3}5|4lIn>bBUJHIa9
zB>HW!9STX8GxO~sZEJI5U`W#qI(;*?@2w9_a00AK+3=@sdi9quuu|s3{Nv?kyeixq
zHV*rBI2u_Rm&-qY)t{S0`Hb1j!VU6u4oWde0_YSbG#nluu(!9v-gLq!OGx4XDWGd=
zR<m=?8bQ$+l!hX&tblIY+Hg^8%YimnxIV}CM84R2QpRkRP<py)=RI#3HWvj-ANqa9
z@od$-XS(#$hY*6ID0}31&w1{uaBbUC*A=a23p5Z#Vb6R1S*wUjDXOaK=`*C1*1aB&
zNut;qb|Kv9{A9x3pZ3EZKkX#BeA|NOd=}2&T%{i4uidsA5UEGwMsZ9w8W9FD#?j3v
zgA=2e(yj~EmE`e4angj$9u)+;#{^Nx`JyA7++;c)Q_kl!FMpZqUx}D~<G*8^DRd{H
z)hHdH!+=#a<6^lWi*~v7;^+AM&-^04`49dZ&OiA9H?JRY{pM|&(l%T_eDsJU4mpZm
zK!#9P3A-=;9D<bklix#V8$zuehZ5MP<1h@FOvgk~*fU4^cfLKh*2wtqzbBsSX`4ll
zUF#S7ARHf_<eP&@^qaY^YnM4NBx$n6=hf#vpIDdv-i%|%*t9#@?{A<KPxIvc=e^ej
z8~9Aub-jVIbiz;Pb04n3uC-O^<#NGncF~J@q7_*rxHe5lgNWmuj37*jvy3E(NfTSW
zYpRO%V$N)~=A<x`Mxr%zovp@Nr@&S|jj*JQ?uBmVeei5oS%KkK#o&LJJxbq%{38W(
zVm6)4SG}bl_RIatZJUnuD(_hWZIi7xQXO?!+nk565PP4T{<NHz!EO|v&~2*PHntkC
z{2N$#E3eieaN&p;tpCd`8iJ~-Da)z{Z31Lj>hfZn-ZOW}trNuHt1_CR$|>ujcY6b;
z2bjb$!a(OfFHv+ItV~3={}RR8kRHdpekbGM#~;A>Hl!n-yEWw{3?`bQqmV3&7{B%l
z6v32})9)~nHD<LY7z?@}MNRK=_~srBDY*{$nLqp|`2N>^i$@>7&pY4xZNBugzl@$7
za61(oT)R#f#R#JabVS)0idDnt&QH)(D}wbs>b#&TYYXlahQJs`<1wRb!Vs)^<+?Do
zO~x35Y8#66imqt^8}=tj#%Mf2$gp2G0A|a+Y_^3pEEfD(7M9Hv3^I%u4MNZ1d)?&3
zhWEJX6<p^zZP(a5rVL23)Zw9Qe({H4>g+Xj-_m0d<&rRrwrE@~hk3&gSt;9Cy<9Gs
zU7RzYU9ea!>AH@|C`08lq>>z6I{*<-H3|{LU^J>OSTC29MZqdBxyT#VrJ+!wXX@C5
zb?$kL@y!O#Kt1_<#%BVzG9=E&^#`^wY?go%cj)txKSCh}V+=9~z?IDXY4I$-(2q_j
z)~lS_Y4Les1}UL&(pUcBZIx)ld;YX?uXiHt!tZDt=ecJ$jN|x;?sV@hZ9k{|(^Ie%
zMM>k6liDl3Y`~vaVXEw%Z~k37!Z~e}Zz?BA5|T7^vQ;hGM1)#k6cA<v(>sLG0lO1}
z(IrJ$v3LD>biBihuYZxQQ;f$Gl4QhevE=u@@q2vZw||FJwfy2A{R)q}oP1J~s+c4U
z*ne(^$t0nv8>&h%o?ho?{^-BP-}{??mEZm6zrq)P=1-6pit%Wdot<4&Yo*G+@f&dO
zEy+*)v5@3;$lWje0`Gl&O>zDfQfVS-4gF*?C7X`$&9;8iUo|9PT<N;+{&(?!^_nMy
zp=a3~oW#Aix(ONl^m}nojNL>*5HK2#$+8V~@8vP{#(abGhPk2bN%f*6W-^`jIHAw7
z<b7_;w$GT}i<(R(eO2&jU`_wEvMN~@xm6rKzhJdk5r-iMd;65@rA+{7MIZu3;}KF;
ztk$LD*v~D4X{%_pWK|o=x~1;iGhdN|lky4(#HD*0dRLo}-GD2m*hAW|RlyU!Hs#)+
z`Vl40hdozMGfqFuX!YMShP<#89-V5@s-<f!NKUH;gF*%%4a7;Zl>^vR-L7yuovU8B
za=6AyLirxQ%l%$Y8=ETRx+ktTE}ZF^HnqOOz&+WsrMj+%QTU(kldnO54--RUKy`>f
zlI-0iyY&?!G-+AV)deb;Q8yj)bxjaw)XkFL{LO#D*MI#tdHCK3%r73Jrc)-@?~r}}
zI)C_QevacPLWd=xY)OLvZ6wiHqSPs6+wsOv{0zVKPyZ2*-uo^ee()}@yzwQpgu2O5
zUB@^7#UB5~|5;Gx@Voy!;otp_TcW!o?tJ+d`QF7TI=@d42+}wu8;yuU@ie=1d4e&n
zTGcd8-!Kr4iDKkvr9B16lL_VFzl|}yUS8MPT%QzjpcEYR325&z_u={00b`T*Lz-(L
z?7Om*DNonSyXu~X45B_KIv$U<B7&<8=LWD@<T;Dk+=h60PMRc4M<b#zK(+yKx<_PM
znzE#+D^^9pV!39%$SLZY&IB|H8l@>aN!z(~>DV?eAF!KfVuMBV-sM#lOZUnCo%(+n
z=z+U^q>e#iX@I)6;j@lEP<8M3YpDG9bC1z*c&Vl-34?$z5Om6hdT|nwCf578$|wx4
z3G@p#&kO7ENt<vTjgl=Yj;rqRa&pk0f<sW|<sQGQ#^thenHRKm-JdI=5A%|wSAvx$
zv|Z)GP>`^@f17Z+LmEc}T}N52(9%$sCF|mXZ+-1|x&QuKeEi@Z<8j1GZ@j?8@{pox
zDC>&veDmup7IS{~=YNKqH(nqN1x-04i#6j>M4KFQa<U*EWgI{M5+8l%tGxBqU*pAB
zzW|-0TCY%A@%R76h%&cr8t;4~<ojRE$)*L<X~dU*=}+@7f91b{Vnvpw#A)oK->0sN
zevTZyh-K?^*drlHlMz`qX2XyD!yDJF=i~BF?%Hwc#axH7=o=4cW2lDL)mtnHUPEp)
znvka16YJ)3zGZlSzPmvwD@7&)NtDDkT=YbLS3HmZyKi1!&F9R{PFP-?u```A9%t06
zHKOadez=EfN}6KL<C8P;vZSaRin=9l42=<JA*nk<(<(ZnQGO1d(nN2)|IPclYK0FC
z)^Hx|MTYUXa0bf#>E>9g4y}GfMymH<{QV4G*3&lr@aHa%s;Vgpi@5H(7Nt6pC~`QK
zz~+iVi%=ah6FfEC6Pv_o*R=#fT8W+{+)7gR?kU4RMZYj%TbSXpZk}&_xSV&ne1FDR
zHh{7msKN@@>*aWYg!Pi0+`7id5yvn59HU!rKqv`Af#?E)@PK4AWir`gXZMscU-H78
zTO1zjA_7RW5ym96b<Jvd!J~@>4^C$M{r~Qt@XDL7^Jo5zUnV#>B0>QbdhrO;R4m$-
z!&`T`_ua4Z?ce$*{PZvXFNlkXe7PV>Vt(=$b3XWfjOqlpUufAm`h?XI?!UF*_Djcn
z@fZIj-~Y9LM3My{3_?fr^`1cm=<+63o!U_DxOWo9v17)x3c1?oD>liKo~v&>_alrf
zIz3I(&0aLy4xY~)`gdYImzvVulTQ(NAqAU*eUj1Qs^JOx>bkak@j~<<Qp<WVSb09}
z;BL@pjdGRF#bUwP>0{;>r|eEf>`XF@(a1O;3LuL^F4hf;<(k>-f^}Jw*9xTrx-db>
zm_~szpoD9DmXZxH5R6e<Mn{NXV7>dDKA7<BSI}4U1Y0>i;qZzUzOd;8(W;|sDpc2A
zId@l$XEIVz7}C_8lX(B(FYBr?T&#6dv(77w(S%V*6eYyjh$uA#VMAH9R_4ejwXblG
zhZF2oFl;VCVuT<HlODG4<oymU{$>H%Q!J-#Yg=toiZBcYZ3`Q~=rE_XmG>|P-F8$(
zLDMz;i%ncLF9-!FLr6?bi!=>7+$B5yGQq(ss5l@F?6VUFgb|1~;^6ouqwyYXom1ww
z>6(gGb<~Z|?KSKk9dPq*#}}HK2lwyu{da!A-~F3w{=#4U3+(J2vMM^5YGgJ?G@7HM
zo2(W&ITw`62ZZBm#FGi3fUo@V1x;7*!S@6H^q;R8A4`;iNOb(*TW@pgPyZY*{?sop
zd*`c29ncAbG#XEiHXPnqDWzEDYt}`M5SAl6O%t-wgdmEXGbS*DWNZJr#3mO7R<<e6
zSA8#-)i<0Hr<rAx)b96Q0(}~AED6>?wr#_DxumTd0%1RwqzRMpWT1fZ^gN1v@Amse
zv7)YWOTA1$6oyPEJH%1qe&or$V{mOH8tSqnU(C69_<-fv8Atp3WN}1Omn2C@JdUYW
zbC&B1<`*+gFP1FU6|1VI)~*5AKoDAaE~Epf>JX(3T5PPOCxJw465^XJY7^;~)u()>
zQ4F_{a5X{cjI9~200G)yv_h#Gqg^Z9XB^LDq^$mz;<F_kt{y}0s~Zs5TWSfbXdH$S
z2!#k#-w(9C#Gbq!`0%X6#?aCl76)6{AcIXGUR-q@pX@BlqK8IEDSCqbr@FcLLvdrI
ze6OJ2Q@)xPB_WNm6c)OxEK_2%OLXlgF-NZ;(~!jYC8Lp{v^bqufDuIDn66D3O?Ig2
zmQHmx3Eg!TP25&=ZA05sjK_Q2zIBWHAAiE%_%HtpzVO8_@Y?6z<oMR}?7Wh3GJC|j
zUi0}s@+W!o$Nwm!=}nSMk|aCG01%o#{XcJ6FBhDA@(@)|XfYH`!>lZL@BRg!fBlQB
z&Q9R;o)u{56Rw+_gdZLcwOUyOw=uRkJIh8+ZQ6$&SHGWOS)!^egM@E5H^%g({SBi<
zf3#l#(6t7t&qLK!PdyTbR?stw;-`kMR}D`mjJ8z9G7PMeP2$BXHaVw0x2vrYHO3~&
z=kpnl9^7ZOIA?EnY@q~gO(YbeQxti{a(=;bxn@;1oX!`ltB#^+X*)v@#K=$)Mlo@k
zI*Js57RGu`fWaMrCg_dMB`1k~HJk06)Kz2iew=9<Mu0|Pl)|V6)z+x4#XMCa?#c1Y
ziE|sLnmz-0*(bSbTx}5Si(0K#6h%qCE{NmQ0yPb^$`<5(IrC}P^OI+V5c0lt#u!B$
zr#-{Lc9K$V<s^)u=f<|l&W(Zc3*Q0(gTd#9_jEZFh00a)jL|lcE2U)vND=}Q*i%^<
zFfoV-ar`B+gI5Sf8Azy=AQXlm5Pb)MF~)&UB_aq2r6fyYFt(?wYdX@=7-6hXg;E_&
z-O!YK>>eJYx{l>y$ya~pciF%80h265i-?;y-eB>ChUxJu$S6aEiYN++#}OFFCWhUE
zDaW_3Q5PK-r!&qUKSUbI;?bP)#pk*4+E4Jw;sj+Hq~10r_C4Xayb9Lqm3vQ{@HflG
zSX7JdKX+AF?++L+#L+Z$|6Y6X%)U7CWP#`MzqQ_=)BQ%3fn+qX5?R`Lg`YNjn!t-Y
zXk#n~m<(()b2jd)W?RP2m|hT7XhT=mEavAty8j{NV$S{~BZ@-0wkE)ejpVBp7qbf%
z%QZ#SvMgKXdCR)By>=jwszoS8WIB|#`yh%FLTLqWwXt0tN}EAyO1gUK_VFORF=J?4
zw^zMSmz&F5Lin&>gYH1J7IdoglL^638P8;-!Z1Y2w$IUg76Wy83_~*^torbJl~e2%
zq@yuOYWte35~}lfwC*SBs_S_w7rS8XXYA>`spdT`N$#J(8h{Pc=8%HLp3J@>RWRH@
zmy@VcdakK8N;$zvHWEi+7@|!}U@{PjIyCIv`Dv1yZ(_725CW|wjnWt`A$0g_Z)62@
zszWP5+bWc92!yn`gg_ES8NxuEW&NVjT%&DUs`Z-bddZEZVLTq8gNViKA>aJ^*T}LF
zj~=|wbI;u%40e&BKpW+nu|dcHEmCAx(bP4`Xw3TYIc=p`oEF?VzD96xhxY7kBoS<#
z&vq89Vaa;~1(t4SI^DTKY~X61a<JiHQRIE)%$N;7lxHNcpK-JQlnt?|z!eTJWZXS%
z^V46IZ@paU@~{P;&ype_%|=WnQ)2gQPgL`aOC;-#t}2<GopJK;W2(iRgPkc^8c`NG
zs;MZOhPG}vIXU5MzG78WRGnd6wI~xH!-P0CfJApT?_-(<)oHX+7~K&iiJNCZr>ys;
z(3U{bLkPsSN#4TzT{5t?V&}cBONU7xQE0SkQEi28JGxF=)tKN%)L<NjAu^CC4G4ko
z)c`)bq1vv{c~$}2Y}>giAJ}w)(%Sbr^jHnJykJ#-4WAc~(sD$3S*qdZhTY}cWmTKs
z@CaO^?FJoX8L6-4@$>`0pQcJ#`Rb;wU4q>Z1|eCN5{0odbRk++bVP{BD@4b4DT|sg
zmULRsHWguDs0gXhgkfM?OdL*2NJx^9C<@Rjv`wK+hi=__LA!9*>gILYDzr?FN9dy?
zy0&G$TGCY|JG)2x+%Mjx$V<Na^<U@ijga?Ga%2#a1_3e@U^HE4kOIa@#Co*=0qeS?
zR1K@yg6}+Dus69yc=7`fE_>5^oEy}>=b94ICX<IPCYzj|TTN~k3{Py&o$is~LRhZ(
z$;5~Cn-kr(OK!lumyK1|wJ!JA_P;MpGscq%LAc?$zp9$IZBVuI>bxX61WCeZXJRYL
z#-=PThmArTD+F7vSYMp;`2NRK#hTr5$|MbGnwqM#bEdT6!J`u%ou09*I?Af0>I_;0
zgi%Bg#>gndU{JC{0$ta*E(uG~-KmZ=%?P82Fc6pkq!MTaN`t~jIKzc*LvH~#*VX$?
zxIC8qQ>(tLyZM~42GVj?H7I2blnf(8@FSvX2}4N`3Yykjy4j7+7C)rY=ZF2iEkjpT
zN!PV>UE`XGI|2l)@oF&GB;OSl!uHhVuJy%MCz2#(H1aJgc9RV6$BPvX{nMtY2fcy{
zv>_X1jM9uiNZZ^hc*3|14W92(_pk*iEeCWm&PcKtL;$A4(4j?2EsuHUof)rY4=@Io
z&#~Uc{r3`N0MEUm3FDZ2t$Gr&VqH;{4M`FsLP4Cxt~uG>gI1bOS=I=R)u!tjOV1(0
z5E&&L9$k0dV@Frlh#&!x^3lhSxc2-T)P-QZsxT`}mc^u5jM0LsG2Fg;o8S1=e~#1!
z(>2IebAED4dg}%Rnax4Cjp+*vO;u4978W9~&pLKMOFvAX%<y5p-6F==elG(wbwyR?
z4#92H4{17LH1aZJ#sLb&pg6T{Y>aVjFJ;m5dD)h!nDJ;r6vkUZ99NaxhaOf{Sza*z
zbJnXHSwU7|H@U`uwu~G`8v+?pcMVmsW_J3R^9T3QO-&j}0^Oh*O<Pv9ZA($LJb3t+
z$LBMaWl3JPbQ+K$VH_cY7#Sosq!b{D&`5$Hq7@aoYiX()qb)WqNfVMZv1EHvB0@>q
zX=-JY_cn_0^JWDWe7d!thv8l{8w@1|_p}Dxb?CN1cMV$GDy|Hj0?m)gNEu@YWU%F(
z{6U7zEwnwl*Y^-9rI8|VW&4f1`Ichpz$nztYXJ|EoI2`@P4%X~U*VamHWO~chk~Yd
z^61(S#Sz(fL>NaXv+aF7Ep+s_pU!p68!U6<D9Z>$fWUf*osf{+;n#obZC-fe6^hl2
zx-1Yf<nZVwU;7`%{H?#TM-XcM;(s*bOFv%|h5=PRCm3fO?2kFSC}=uG90ab<FGL20
zC<ut7sOJ(#*i+JkAO|kG?ricjkTEh2i6YI*ul*R`fBO@D>7_3q#}@ewnx<7Wwe5e5
zW65-9hgV;Dl?T7|Rnj!!;_MTYkW^L2Xnd1y{h_U{YJm^|bzN<RLPCfhXc+{OPhCo8
zV{25`QI{oJDPosrjN_P*%R~7zL?D7IFE#XpiXyk@boX`759Ny^Ps8L5Ju)FS>PP<H
zL13%EBWHX)XO+>m{YN#G-Mm3aXB20je8lX*$Ec<vj0C3E%!;B<Gt_OzM-LwJ(I*dC
zS1omCo7-g&6GRCzjP1b_I#yAEg&Bm2Z7C7~ohVUVO;cAG9a>M@Xp$@?3~f?9kdi<d
zx;CKgG@YYM5+VfgB!uL0=)39BbYIk+J&?4*bjq%UAVdTaG6;AYeCMk1Y@U=eP<TV}
zEFX4bJC9>~lh*6JSC8=|crvoY^x7r0hZEy*N?w-Ahgw3|DxA$HZ1emKZlM2~H+p4J
z(3BOp3R&Pf)FV$1!sRfP{u=&kn<nI@4}tq?UKoKkXsJ<zi0p{3e&-Wr>z1E7KBUfT
zWR$XdaKw5&;s5<tc4=z?b->^H&-Z!rC(j5*AVtFY*@CvIxOMv)i*-v|XjI$T->C%3
z^CUUBc;k#xASI!zBq_&1t&GMPI|p@Ha{boxJb3t+d^sbX>{HbmfhI`;f;2#NhORTD
zS;lLxzrpw3evc%SoSi*FWGVAS!Q|*BdikNvW4S}9EIb*WowFoKm`<mTJ>t{7Ziy|a
zYo)2Hs;7hTh9w($srG?q`;>L(f4?oV6n*2l^~Ta334WPy?rHDcAF!4$-wP=?IWrMj
zo_>t2ytWO^dPP^|sJf(X8nhJT=a10U$|l9Dmac7Ct_zZE3{l3t`;YkG{zFz}O{*+s
zI}BoiDDFKHAw%mq8XFrMfi(!y;glqZPNJKJ+L1>Z$3>o|8EG1m#FC_gR!SOas9Qy&
z4cf>b8WnB!Q-JW^dDoy@Tg}s2feevhLKMfn=*nju&t#;=<1tFxWMd!4KjYYr<*mW-
zz_+&Tdhnz_m2{^!mBfK3;}?kMo7P%8F-2ENF}@snsprV&(cEWjlH2R}>3$RWhE#9N
zu1dW1C1hi^45r^e7PgQeg+$1JmJX3jD9izGfAAfC_7{E@(;AGCOn3GfO?JpvB?s4Q
z9)DzeyKX$!fq)>45ZaKX3G<63Mb&cc<_)s_n6gmRWrxr>k1BNPCoVS?2n$~^1}z{6
z1JVRKE$OsnxQwCz)oO0vzQO$u@A3NQ4^Sq+Aj$I%V>-gXvPd8d<H?BEUweu5a!#WS
zvMy=jeRMQI1__wXHT<qA%e-%x?YZE4%H5}X<pLY#c>`4yC9MO0N+}s<8F8F&nQ39j
zXS7*tHgs&;*1;Cq{@qz-%ncsuv#en+Yh?}HrZ=<)4ZE<`uCbb?%&8U^G^-h=Z79l|
zx>dAIL0K21Ny=irqG(!1y9caWc=+CZ-oN*NMPAdoDq|$A;&2!yh%mOfs35fOdh@7g
zp`8a5Sg9!~Q9`1dnpP>gszxb;YAphMI!YKtAu5C}64Y%#-WtkQV|dat-3ouL`vykX
zB7o|wuBNJOIzWbqFd~ese7gl_|A-(6K_m!5tFGF0iXoZ)Rbv>tUw$zEcwmv|RaIJ&
zb{F;u41tg|rt438AE?>R@S;~S_I*)lnmP`n?GoqZ3H3{{?K)QLrL79~)x4O|XhIlz
zrUcJkv@v+YQ`wXA)upA*@S1x5WDcAZkd_)@0*p*~cvkZhKlu}s)soZq9&_#5ZANKI
zvtE)WCI9wcUi0_<+RG#<{Dr@Kj+6q`O2Q<;kP<`{tNEO!EV+K`7JG*iiWrKuqO3Z?
zSd(T6LAZfhhyd4aWC#K<$`D3^z<>~vG#ycw9mmHv`1ZHI!SgQ_#A$}^Kneqat)jK9
zvHf`xc6W9-fBXr$GiW8KlY&YgkR&_k>Ws2(sPZM9gCs~PNk=1+%r|*IS#=%GuQs--
zTQ@a})xsVE&KsOeQbyUBD2T3DhhjU2A_SmOonz~)D?FdN)mxlwQTO`$7_L#fj<xgZ
zZBukr75)0M%m!17^^#a_rnW`b6>YJmS}j?gKc-xsBTQ$B4-17%B7)HlpFBFDtXgC`
zW?35^o?h_y@d<fRAf#QhQb>DPMNZ!?3f+3sc1^oPUY{SffC>Z>83Y&vp$IUY?UOC*
znzrj`we6i1;}O#=VH^dRP&0`ns|Mz^#@4d}jI?Uw^n2PGCxOA}4&61FwnerbqO+`?
zHYBwMRRjSzW|Pl4o*B#;hM~=m)Z3xka9mlkyy9p4_v_j+4OlnPsmJsQ>cKUI)z&Mk
zlA_3ab&et5^yT0EL>b#`DurcP(o{9NQ~kB90!=(90MMVpm!EAD&Uk?bYs><t;}=Jf
z+bELGG=wr{ov)B4q^;L<guM2-FR^!Y%<S<)*2@Ls$(a4)l>gQL>3xzY!sgXd5V%|y
z7QdrA$~@<z_wP}c#~j}}B1!`CbwgWsR5`9{W(ebuPTS;;msJV_K@ivu3Cq`KNV1SL
z+2_vP=Xw9#_jvt{H|TT-GG=!NRM%1$in`WlDLFjY=R0{$Rdpoe9p}4+gy|mZ#RJN+
zLU#&dECweT04@9E_~2b#UN4@uXPxJk4_?_enNgPYg^{bqrsll4zBeqsSJ0S^kgJ_1
zdjM`9Lc`qqCb#KDMtiS02*^A`D*?J~p|&uTrdrXhX0(eL<>_PAMNZSTgz1RM{yxeG
z-uv(#dEJst_NW@g`E16clQYVybVk5dLuH@SiIG8IHSByI${BoDpX-U`d=*ZZ03l+8
zkZ36p2)fo%cdJslemaFfakOU{J;$nJuY<i(@UUoEG=@OiW^FBuBNPNrajbRc*k4<;
zw&E&2Q62=5ld}5Zw&C)4Mlfg48!EY79J}&F{A@$eb;=1d*bu?~x5+d0*K%Q=r*rk=
zvYgbX;cYyeqbeP-e%H}CirF|ziL=z6*4{wbPd)Km{=Ukxq^dW4V?u~6gQ<-{V+aID
zqd>QmZOM4=fY)CC5>iS&y!S3`Suhz-h{BK{j9BLdD(cwT+oS0^5{adVZ7R}HhA|_W
zrlBbpeDYDrdcEe_tz&lf$5fRfUnn|dt23rk$UxFIhA6W4)H%{KA+QRxp+HDO*&3dI
z`8n>reV?k($gDz{h_lBq9Z8}{kfxHlQlwc*SrrJGl1+99;}jzT<oJm7I;SctE72_k
zaTN75z)xEzmk$7?4C{44o>yos07(>zp6Wy((aw|Dy5^voJ3I(&!>@0jvD{)F12^>Q
z`UCQbHDvnqfcN12!DjCV=o(B@(dH{w=ciPQbM)eZdOc&2FVWG6(cyKX@f0N_AH8>v
zyfTE70~Sq7mDgO%maZ4kS12ub*C)!uAhIch!F+9{U9qr;;STi0q96#I`io6LXjfIQ
zJQb1Bw4LV0;h5blL`E&+1hOpQ@lr7>46XLnR(sG$jIG{vEnQpFwH9Y6gMc7T3B%ZW
zfI>Pv*0UAeeji{?3quf6*nH6+U~Co=o8-A_9SlJcx@4qrUV!hZE6W^v5=*BXGwF%!
zei(g<V92!Dx&)%D=4DngX?PO*aA|lgKboeoAxQr_hv;;#zh(`E7RJIF(lK`qUL_t+
zsmg-c`6E=hU_9O@OEWs%(Uohfgd+ol2r#ClZCj!sAc|vFs}-Y>1qv=rMYS$Ceei(w
za?Sqn5jUQ@#^JSu#k`?yG(yW>XV4g%M7NkLyT9wok<A5$G>svMxPJSX2M_P@;tMaL
zOJxNTM1(O4BdqF*F>w9nZO+cuh|uyaX{~629qPKZI65PULdP$jBm*OFo(T6Yd*)SD
z_Mlkl1Sxl?6UHMifZ#uPDVH=jNUXQpuwJkF<ULp}FAJ`9N;cck-NrEWw_^{;;!*>0
z9LG$vl)xxVS3y@{$`#A=6V5-n&*Je3qN{CKw|C5R{~GD;0khSb2cJA*Q7fY99%a)~
zl@0UdlJ#n7JrSn^C<7ldMaU>d%Fsp?PL8W5sbgbUg-@irM=)+)gb)a68`qVzp;w5Y
zZ7b@w11)L0lDah<@5Nj{7?CELRBHC3l!sZ%$+DqrKoL4wFk2Yt+PbILmVrdZu{BVU
z%bU6V=Tj@6bv$#p7ex_a7+RvXPWhbnQ^sfUuI$DsT*&0L_^f)2s{**w(kiyDuH=b^
zt99-!pU%c;>awJ*Ejqn33K?0hsW|a{TK+WhpV=5HzmdAGqsVg_uj7Xyb}R|LZ>-;R
zjsQ%^h@x&eSw7;{t$=m0X0=|DjmE_B*g_ouDGeG;l%#YnxvbijFo<ZojvxrhS1YH$
zdxX?;isc2Xi^r@MGv=pfy!i4Rj`sIh=Yp!$be$vrL)iS70D;6dy$jk-(Uc8!o|B}K
z@ovgwy+@rFy!c$qhxZpec+_&|<_(lFtd=Vd_9onY@kNT}BjO~rxxFMN7^Qe6mOu(e
zJeOHm0>eOz-nbiw%kkk!+gQWWwRNAr7>`FxCXSle*h<$`>&#xney!^itCcOxc<)%s
zz;S}bgS9LAQ-14^bK(9r++`!kusfYFibCqDfUcy@=alm^){h=C|L8t>Q8Br3jmh;}
zjCPI?NycKGb9yqPEL%(%v#Kk)x}{hyS<Ws{Z9^cXqmd4fK}ZmUu3{JV`8LF0eGliL
zH@KIoJq{{hu-6T23D)))N<rJ!RBc706^&9fZAsg;+&P#q8byq>;`vap6OVYXRGjCH
zD|U70wx;VEj8n3fVMq|gh`_R#`1|Vb^NIZQ)5kLzsitn+!jnW{NZYhqzwc^ewV_7w
zz{>4qYHz%(O6sz<Nmy;@R1Gq7KcMN_j@8QMp+aF>HKIXP=CZVqy{~O#xe$V?uBqz+
zqdTP0ghG;~BSy)HkN|DA@6LX{Crou`m9~w%Rf%-cJwX)q0-@V^Fbgm=0s>_bxVuLO
zly$|$`3ckMF5_$m<JxHg2|=6?1`$d%MBx~PqN+;5ETe83_NEgqs*<9v7-tb(xJwkp
zT+GgBs)9$Ke8}SBf?IcA;M&b=jCV4sN~3D+@G24=3qlEjw4rF41ekF~-E=fn%ks1&
z5^(?F$6VjPLxg0#nDN1K!7HD8jdX8^vT6u|h~N74UqxmqaTv0{f5ad9@=qYrDQb1b
zBn=3sJN6_Mz7I~i5KRx}$syu|ZF4G%HBD985Jd!pactQvlFTi03z57$`lP!%sq4C=
zX)0$}0w)OTo5i6gz!I*KIDkZ425gpP>AWjV5{K;WOgKsdlqxYgXFWe-eSXUP^bw~I
zAF-@z4(`6h{_W>TclQx8X0cdua(2Nguc%u^-8L?Tv0^sAusU}##0ZOBl0ifexhkHt
ziE=G%w}<EryDbig!9?~DAtyU#ob~T#OxRp$6h<iLU1=$i${vVKU9wOit%AHU+&-MJ
zn+8ZB7$tCRENQ!dvsGamhMj>DGH|kNDS=D~gK)#*>kt1yuKBZ!XEIV+Tm5-UrMuzw
z-hkWe&ux1k-f#>Y(XCBvUDLX%$Q^rS*LA1?6J}M_RHaR<he1f-PO-_(&J{-C=_G01
z`)S+G8B}4ZQ_>6>2A0!l^P#~>Wwx@Yo+C<WWvhi!k_^H;7@l7WV+4VYL6l6UM?~W>
z_wL;z$;QO#h%k!jy4uoe2#Ha$ml7}XH9I?d)OE$;^o;$311@GWCgU;7*@Ac)qLd&V
z?QkobuwKupo0^kHC%pB&Z}Z^(i0ij*bLaWnB;zqfWs}Bn7oa3`#t=ug)g}lf**GE$
z14>bWk$m{hJr0fz8Bg}Pa}@C4>^@)p+VAo5D=$-)C9C-f`_mC39dUGcNSut&T5xdl
zWiE;n(%tK%VT`!sLp{y;_ND#xdfmUbFtiFSJvTku1(B<YBvQ&g!EOJ&!7)uH(<f2_
z-ZdElHNEH6H2yjgP>j=rI7rY{jZ78PE2`OJ^7B(>C+FP%=n-9%a_!Yu*}wf9lifpv
z5S*Q#b9R2ta=Ee<;ktH;G&zg;+{s<Je3Gygxgd%N0$aMrQ~Mgz7l7<4cK>){?K@*2
zw)e%*082MlLfT@H?ks?_YijDYMdzBX({wFlNrVhT5R$4>tZVB<p;|iCVi3qc61d{p
zP>tE%jxAugH~o|q@Z@-AoRpLSN^9Cy(KKpXpNl6w7-Kdr`BeyB|MM2qX=yjQwnZ!L
zDsBNy)6ljJp|D<n2yFA_Re7kZ7N3+h4^>qat&_+y#xC4AP6*@3#_hh9<%;1kJ#E`M
z+T*s*mj$5>7iAEj21(c6EK7+{1}!BdQ%)W|vI+L-E~C+euC)!J7V~1`_fXnC$xd}>
zqnJ#loIQF-Q<n&7TYJ(ZA<tKYNs3Z1&QeC~FqV?t-2*Pp9<!XC@&0$e#|J-nhdVF6
z#PQ8LOm?OgBhy*KyI59eI=~{<(==gcKchQH`RSi}lW%<OYy85W{Npr}jDv2UDla&H
z^pN#>&Dq5RuHQH&+q=g8&MrpSFy-LRo0!FW#7T-!#&!hRiG1q%@|817<1*lfasb^8
zY2o%ezG_|d=zmu!2H{8x$Np1WYs9akVOK*C1azIH@YP-8<|ZbUf_gRQVt#>{OERfA
z`Q#JMW^)$v8ZqAI`rQ{9U%O7WcYr{0aeB(dY|ebKpsH$jP#T)LX1!QZIW|xP7%5!t
zH$u2RH{tR}40xKv*4;Z}J=h0*Udn}8PwbJc>ZL}+5*<XyphE@`s;%g{hP<w6jj?+r
z$!xQ|F&49?T2$*ItRS=)t|&stzyXPUz0n8MW{@QNY~$HT@1l^ll~(a0iU`7pPHE>c
z`6`Rya@gf3=Bk{O@e?RvwOS3@V=Vey2BMc(4?;_S5ymlD->b#e@4CuJ8DmM!e2<{%
z>^)k=7u!Cgg>|2`@x)(p@GP4)oSq}fEr>9V7&&m_<pkZvcMu3!qvZ}4WkB1O92^{x
zWK%TInTEi%Fc>K@F6o{)2tm`-j3+xx_I7#n@F90!xXW_AW;_~W3K;M1qlHA(HAym}
zX<J6qF{AMg>%BvovShVh^3ezH@!mUcaqIT;yzt6vWTOcNNt%r;msitJtqY1(&S;Xc
zb1>o7E4NuJSKNF1Jzo35YY5Sj&SsoEKI7J{n<&xo!F!J>$k;!)#&|Lz4nstIz<B>I
zqI!f8R=vb;uFK)LAD02T;OJMzlm1lxG_Toz&$C&0PsOg?gf#Ves3C~->fIv5Mv%d%
zjx-dEl7LVeW+x9RW@nr{e8fD^WLd<;vLx?9c5c4Fcz>VC{sB9?2ee9aadE-LV#dXM
z!749kn})7aG;PCbv1Gkk*#=!l7bIn153DdD2oi@Xml)$cJ+*b8eoj3c0hikFI5Bg@
zy!$_eaa1g>S{MbO14|ki1R-rxq1%e4ZqXW21UhiVCEXy6LK%Y$BBD4Yh%7A(uIr&^
zOflYEdBLn7A)l1TNd?kZj0Pdmr<lj5LlQPsl%lApDqA7YN}<};S$j<sTJ>QeL!u<z
zf*V}r`uib}bZtwX`)+cD?rhIjoFvFF!1sCg@7P|*oj&#|#%P+_cBOkXu@IuC^^k58
zdHL_2@eqMDw!&A-L)KMCHrXdmrvzbWJtJ-7^dO7~<Je}sw5IM_k|ahc#YZ2!14a`{
zIDh<@@$Mc9h_VSqUa`A>fQb@J*O8<n@}gq8vqO-KXsR{i{R57!U#F}~9zA-@NB{I!
zc;UsDc=6R&2x7}A9)>aTG$LOwSe#ujdw9X*c+AURc%2X5`jGqY-s8nLUm_Vr{NUS$
z2d9tu{1-pZZ~phc&NsjJE{{%5`Kh1!NnX5rlh%aD-P`E;F@f$dNVKc$NNIXHWxsx`
zVJWGqx@XLclbF$XLX^Z@W%~VRhq)t5s#maDtsRub;-RuE9k4Ap$teG>2G{bvb^=3F
z=9s1=pkow8$fjg9pYibh_jvs1W7@hUlp*8&1Ez;J**m&Hwtqk}&dAaXql_(NI6=?#
zdgTn16|7yZRxDRb%Tnp`wnB!44*V#6pPLi|f8>f(h7i4~&9FK-n53;j&}JqFB~7>9
z4IqP%NJz8_F%DPNH4R<UQZ$V{i!pXR2GUkH<Af-R3F8zU#8j=lRw##T61J*mT&n1^
zj%Tj!Nona&qtI0c29q^>*<01EA2R$RtLm20wcK>A1%`!zBugT?&Te8U0;>o2l<>~~
zJ|DJO#h2Wbj;-+%G9*b;!Z_*~5iSJ~gBQ5TXO;OH)wR8WisRU#%7b7~slG%3BS099
zkr|boG9C%S?2ssp=vtR-G{SmN?k>EFbKNxd9#uyu6d!%~A-C?_VZNx@Kf1;`FG;f$
z)wNW`is|k?T_Dk|VrOTEvZ=|&DM@-jTQ_udL#jFsuiYk}o$%nnC;ZFbe8io*FYw&m
z7ePcQZCo`eLaUmyPtI7Kueg5wI%kj1`1aSo$(Mfoi~RVPzsR@0^F1CuctjGyYIefA
z?-=r8#dCk|-=GN3nPkVb*>jlUW3=uNB0>i(t#&X7V^3Mn*166Ld}TTcLq_8XNt$i8
z($;z!2GpB5_uTaM$1JjWG#YJ5kZmer&hZrzf%mR7LKxa&O_MK(OidDn=ylHg<S`eI
z9&!JJ4_K51J4eSH+`h~1wOi~Q9x>Y8CCV(DXH}IfR|{r~B^MWSmW!pM0%}lgN1hif
zmJ4d<H5v>^IRI=%$S`pIcfLyOO!QS->XL_f)wt>~?}=Ef6~nM;1(QM$Nr4LN`qW*G
zZX3|ba-RjEg<wolrsIqxwk<DJ2TL@0gU#ETpsy?jHVMXMWBY02nS~fO5QM<SpgkVw
z@)`G;FWP6bs;cTq(LA@fZQ01En%03yLrZ4=Dfc)GH><j4<yaPsQ%TOo*{2GPZllYc
z0d--XU*tn+t52J5n?cZ!9-&D{!k8e9Y^zJx^dba75caXWKiMR<@fSx~vv;u1dc9^c
z*(1+$!YIagfarQ<jplTU3ImMNjMEs^Dbh(wl18+3W;aQvxOV$R#ybbBS92ad_=NM*
zQ(k@j4bp5v5GAD37?ZCF+LrbFoY}=GH*ekH-dpeT4fRdl_}rIx<IOkt;Jtgi`o+)l
zFTegxP>NA9qFm33lYl~x7$3e)EhJ`r4<TxV36Md6b~Ls?&m_=x9YWX*5|1)Q<M9)4
zj!nq2DaQCsRo6AE)yg#?+c%kIBPJ6YE)KCd!~6|^5fWn*p)|C0L4I+9stQI~NV}e~
zJUiv&<ByqL%n8DX+b_Jr;d3vtb995r-ag~Wj_m|M(9{j9)zZqLU(8r6mK238I9cJ_
zlEq?ftA)_>G|4cwIH<s;QUb>W3ubFyJz<2z(+$&8*X6*G7=G=Ed6tsE=^vUvqGe!(
z6zI@WpbXSXlW9XH1ERnl4AP2MQC6B3P|EHluQ~Z6L!2w6`-EZ`MLkx@>yvGJn4isy
za(CRj;6?6r$^>B;5XTO!8nia{727EEt6q-(ng?y#HXGXumd42?4s4%Kqd%#@hi&#{
zSz*sjpP-k?PO}VLqE%k{-CKW`(1iOKkr9Ro8bjOER$tIq{+>t%1Yu+g>%uM$)mTx<
z>2yq0u1J!Y)6+AKu3e`na&Fw-XSK=^A|Q?emWw5$$rvF7X&Mm*AVd4yGLXnX5XK30
z)zT?JvU5Z<nsRjgCigygkMDo`Tiku|C5~_0Mu!nWI3bD*K`2<RR=oS}drYzk-~Hy>
z+`0V%y6d=c{T68~xPE-dX<m_KF(%L$k<v)eB4m8+i;%=rvv&!*7(q*GtVio<pRM$~
zUA$?MFd0vYqIkp0#Z?P?^ZvSS!<X+XWKk43*3XRu*%r-^^Zuo)-i9cea?Qn~2XyNt
znUq|d&AB*zM4jiX%bIYy&-FXcGre(}WcPsS{t-!**>wr%wqczwSuW=+77OO{Ir%!L
zapJd4+c?T>Ct@nB2Q32o{(~@ahkvkn#;34e4xu$$$@8b|hbwbZKcq@3uuLRITM8p@
z1B}tAR#Pa2QXO@xDH_dWl#xUsLc%zfNCNU&_h<mE&DQVwEaMq<_e{_IjWbq0G|}26
zMxW85gPRn$h*oQxJP*TwD2{AJA&Su=pseb?LUcKtylkv|DZ464t5jh$K^T%wwkc+%
zm-xA4WW6(NTXxo#KE`d9BpDbw`^&fPVN)@&xlkjZ3nnC^9a0fdo0_s}X`32^B#K5v
zu|;DWA&>~-IHE0cL=bR(c19qjl@Aw!qA18FI}}Ah7>_B}3&I$}C?L;scK3HMs->zM
z+QJ}|VO<nlyKxPJB#tGrQXm5w1gllf_2cWLQONn@2h>%;jXQVP*`Lx)uQA%$W4A21
zIDO2?C-*ozIU%4TMzO3V`^VS$%1?cTzw;0O5pfg}$_!MAu49hx-bE%M-N8Ouq_ned
zAv=K}e3Grx*twmKcSz%8i>!Z`ck0U#uDIa~8ouG&Lsuq~Y5%T=|Mk^i1A%}5g)j<T
z7hF8JPd>jO)`m|WJ*Jq?$kz+9-Cd4ee2wwZF_VL1(%n7M$&@&Z98b8RYZ~%m#cDO@
z;^Km{von^<rA=Wpt&>|@v0ksR-1p9K1_42wAcM#vv>p7!n5XZht(3xGj{KjWc6j^G
z^m!`8Q=<W6j1BLU>&sIux^sPbLOO<uhE`cDW79NDM<dcWB#@FMvP>GFXqDw^coqW%
z&syC}B7~HbRfRSRqgw=)bLes}pC3OgKh@)B+yct;g1jhMt=E=3K1xZ1pzB(!{<udi
zUu~c^;Jd2l`Y~D~3}o4uQ8s$=^l<U`R$PxkD@9!tsJ7|v({pJ*O$pi61Q`atjmU_y
zDOob2QZ>4?>KRgo7$iYtTTyH`lyRDY5X2E(TQfU5Az!aJzJ5$mRpfO^GTFgs5V{}=
zAz!bF;+QgDa_^mQb9(ZKv(v|1JG#c*=U<_0EmzUb?jdnxyU{yk2&0(sXiD3bWIMZb
zN>i*>eDwC)>|Q(K)^oSm+s$Z35n-ATB`H66>pdRaKjHA`h?A3#Iq3Ge{n{%efB#pI
zfh3G0vT;ULwk$7ZjCZE2x`_Dtmq>$#;_*8ep{ds;&AP-mzHJdYy}T^8+>XOTV~~s5
zctomy*X#AVujC2gVGAj;2ap=HkQm`oBDy1z0U?UWw3Mq0E*^ct`gFm1wMOWMa=m0e
zJ7?$S9j?9fD&uQ6813van(UJF3Pg78cU?=>)GSsxi}{Mxe8qaPq^U~FtJHNArB}&u
zWR3=h&W>H_KCwx2=j}bMb>ymitTC7V;tpO9L+DqetNt06f06?OW$@K)4+OP3d8%zu
zoqY~r1VNNxg1~hn8E`*iH8d@>Wjx9lrwL&ckYGRr)QzU@I@-&K=^rwl$w;MXib;kX
z($>>>8n@JC1NBrxC5$^Y3}so_Do<Te*9}eE5GCoe+-Bbd>Y)otsX+&Rcy;`KIX1^w
z4x`G`deE~icm?CDhA(e>&ZACs=wz3+QG|MhmJwAxr>-l)Fd|7tgmDTY!065{8Er#b
zYlRG9AbI@oA$$9KD6J`qlAWE_Vygm)(H(WZWO-h2c6!QtAG}8p2$Cq^^h6T{A$tc$
zeEi{iy!c{36(l4{LenWLjh@<;ps0$FT9C_tx+*w*^oaFp$&Fh#xOV-R-Dyge+~M^%
z-{irA$4sU>Xw~relaD#r-(ws~V#kBOE-D~kRaWRILra((PDyuviq-OhX7&+fv8Glv
z$bgW5WHe&rggCwS-te9D?`AOfK3p8G!O_UGBt#p~F?L@Bx+4uWX(EuSM(PGr7p#|$
zIsNcm9zB|~%uA+wW5%NiN3Xxh(ep2`bNx1x-F=cYBaq>i5iv&f)w;!E!TH4nv)RlR
zYueWJ{FN4B({&rp^}vRCG7Jf1L?G?KcU61WHf&>)UfKNErM=}m#mnp+*m2z}00961
zNkl<Z{nuPB6mdo9Wj6fe)_bbxIvd4E5g?<4FpixOG`4|VRJ3hFSyi^^rL={@j5C5T
zCX@ypf<}U}R5qV=JX>{7*y=-?Bvf&-MGp7F8hib}uNuRNZkl?FPFQuSKRNBS200|2
zgx4Fh;jZ>2PwV=Rw-iaP^1g<nscrLc=gPyjnwOBJBcyQ2Tzd=K)j?N{k<N*(o0eej
z2u+Bjqsm)UQ(8J<Vf(TA0)!Fhu0vvIx(2Njp_Jt76+r-Rzx@{bN7tE-rmPlo;y|Kw
zM^)r>b-{y=KIHM~DeF8ZP9l=sU4($e>>RC?Wj?s`9IC2l%7An{rpTA1qp{T}+}R<W
zOquNOvCbE)&u7$i#r=DquwLZs9~}{86JC4$4Swr4f1NOl8IMLhyf`P-nit-D9ibJY
z$qtw?otEq#jx4S%f}rbIR2p43)V`liSYw=}BgS5i%2=wAE0n;u-k;W*^?L1)+V+r(
zqL}eGBMbuvCy=&qVFVE!<FI2t3mC^x7X@9pX0^INtxh?6^nm%SChrV|(0t*=SGo4$
z8)TDR#yfk~>v1{Q;hJt5n`fHMX3S<Y&d$!6FP1cQgVKtkbZfe<Y>o$q@^*|DVU!Sf
zy5Fk`UzaMz+e>vjl--Q&pEX`#cbI=Q#!);E_mq7f#%PS{FkOSzEeJuBrljeFI34x4
zz_xCvi;_-4XFzogRol?6J4Sj$s|{(Qh~tPL42Z4IjQQ-^pwG^6?jOh-4PWWoAkUv|
z^wkSDo0iaS?X0mvxsK_rL?TzcELY9zxuz_W0$7QnBuSqzM8nVd;TK_1ZYXbsa5CAW
zN#8YZC8XWG_n#KMM<}JFt!t{PBMU}k;}P}p)D{LhMVKUnv2*Vooi8qI4uZ%k=co>C
z1a(~!g#mj<$4vJQF$72f%jG!_Ke^8+OL_e80r&2GK&LuJql|GjaS3HjFbS!v5*a02
zoSm_Mbe*oLIe$E3xn6PO)*T|J4;O?9<J}#mhZEXs*ICV$T%4WuCG?}~w>Ui7=hfHV
zfFNe?;2K&PKD_q<KmA93fd`*F;^Ofh|JDEXfUbrA@c+_r`=yfi-+!O3ZkSBQR87FV
z5uhYtVCB`b$(T4vEF_~p*sf0O+I3=O%*wI=bl-G78l}$Dl^A6Rga$eyVc1Ot`-vb6
z8l>*fx}eNwEavCrb;H5!+sIwR!+Fk2Z@$6pmtSYRvrn2$`d&Jvx7WE+st-3;s}<+x
z=PZ{?>c(<4HBCca<djwE;4ij%D;+vpIvNw{xR$&mTW^dn8LX{bl=p_*M)El(V=M#f
zFn?!^O?mimPxRiU3$@t*A3@;00GW+8E(c|sh5<(7DbvG4_70EO+utV}Sslu*ZRx6(
zqAFRhmaG>u*2_6%z9ugU>ZT=2Q<5yP`DGarNr4GJd%X9|5CkEFMVvcOr)S0V6$TF{
z@IU{oNiv_8@{9<|A<yGD*!r2RU|_47SGwLp%Sn|cDcLA<V7I3|mvrv4aoz4#+NziC
z7PdDpPLhpfbSL*#e!zGKPtqx6jjc|QVM4yPFHi)6Fiwe*)V9Bb0Rn?E9bMBBNoX4z
z5=bdoub1RS&UidVD@{{3B<YAS5QKptFdgr`^8-Ho@I%s3#@_BOhldADcXlyIlx?MF
zcmDu{Brgi4Z3|iv2Ub^Pe)gDyqid|niqUAwde#s|Fg-lt=*AIy2YZzHiq&$-#n~A~
zYwo`I5_MHF*|opt?EH+TZ5XHf{NMlM3FnV&k>Nl8KLq^w|J}EE{`n~~8#6t;!L{dZ
z^L+j$@BQXKrYp~wjK)M+;-=OWQ8Z3Xc)OD2|J^jMT}4^i&mfRYCN}4&jfHPW41sP*
zBF!|0-7F-dLDiPhc(I(bT;|kTV3Hlu!!bL%A<zB99S)9fkY-cjSUNnFq0v^GPdJC9
zZ97)06&Dw$JbwI`Pd@pC(~~n-Tr%iRQ<W8!6OWJ%lqntFPzDJ>;Mgi9cwVgGp4r$H
zvGre%+w9BH8)UcE+OnErdLZT20X+x<J%_(BcJWF;>8>ZiGg=`-izJs}1VW%qOV^6N
zT_?^G!YpAl88exVSk5w5^BJqzj54n%bV;W=x}+nC6QVF7_>}#4c|2Q#ao0f@BBdY*
zg{|a-k|+wPYGFk@{SObyqMI*mmqIsR+iY~L1$kaMWUg&w_MxPY$4zev?V=5Y7@VvX
z##pw=AP7mbjKE=;HiyntafEeyZ9XeUsn+6G#DJtfkPP~yUHN)*O#uW36q>TEsH&3D
z@k<bns4h;CG6EwZiiqL_8Ah%-Kq9mz3M5*$6xG_{dTiBHY0Yw7aP!tafwX#R#z@}!
z{#$(an_mT^+27qINfRcclq5=sWK0;xXx-6vnzOSD3^u8`Kixsbu?;_hfZ2S`XgsFt
zTIOdDi9?Ibp3fE>9`15*xW~cq9@)+g)v9K-niGdL#cD;IMx<HF?Hf1v-uJ#sHrc0Z
z2RWl!vAeTFs|@9O&B38!z9`t=dybcW>Q8d->wk}Nnj*reH}b|9*9J1!tb@O|O+&F-
zqgpF{riCO)BF0%p5=ugc)k#B1CMAhfgk4S437V#0xtud!E-0#s&ZN}QgbG2Fq&$D^
z29v#`UPWaXM)^q0U}eLKV$E_f<Lu-yj~{%(+2fO5RmW&UU6mF9+&1=;LOQ@^gp3oy
zFeL~r+`wLUNd4=c+f~f_3OIv>Jp8-o)N*qtUG)&{(EU=lD!1tklC^%eN+$`h!lE^T
z4vfNdEs<2jv1LApr6$xJs%oeeh`<m8nn2pnTpNvy64LS15?FQ`VI*}?(3Do(&uD{6
z1kvT@<4+mSWTfh<?!`JcRKj8JNe#j7{R!yj@Dg0$s>Gl-VpUa<7X@wGS-J4Gg>+MX
z_O5CFS@#*w*y(cyLy4lN4SOH9+Mv3QqFB*1l_P93z4}KS$5*N-Uq&!1Y3S;PCNF3L
z!SV6uI9;u=1bw!|9448?#RPKmELx*fi!mLckmReJFqABpOP0$eH?H4gwwQBxbZm14
z%5ZUhO1@rlba>2kXP1McBgW$qaTF7TiBshW2|_DxZtpeV{QLx@W;_~O7KAinv0RWw
z2?s~VgkenGG?e9n58pfCqYpw3_OJ2W^LNO`35W^BdPy8hisg!(-F-wlAxKi1@{~XP
zZ~hwp^l$w<f{=gnKYX8XtT~x2IoL`0-QW6E#-kl>-Mr1-3omf|@-HCf-=^tQzqahW
zUY#pcO0mv!+dt_RSQrKD?i`Xvl4&ZrzCT9S1wwbwRR~ql6-zG4l4V}7DoT`yP~ics
zj#w%XqX~O^2TXQ$3F63lAWqh4xSoxJJk(XiYO&<}^o*0or<|Oe(Kx`dF;?b&wOUcv
zja#$QVyS`{87Bl$?7+Jlnjzz};e+3YH%^<gyw6c>ui;C>8#ZbC^N^y)p>5}|Z11ho
z3axGa$>^G{t{6ojH;)h4*@>8@W5(kh+8}1uG?Zn7aNQq4ARtIESiz9dc!C7W-VjEF
zQAnAuX(~%6tvXHnDS+(D<C$|(ZQG)aRax=8;BjnG=-TC_%%#t4=4}{DZ~i{Sz(^?_
zo}}o*J*_(o7Sw0Uw`^Ursq5ZAdCZpgeynWL)k*!~HH^`gB1Tydpq*q(j+P{etMCO^
zmQII>{km%?^NOx*$gVw4B_{~EMrlPT1EM$~PDcb`XvI){jze31cPVT&&Ihwy*Y;B2
zhsW2+i^3j?ZOy~`_t>3GiK2*fG-fp2VKUvd;bfG6umH}9UFSA`V<bToLEBKQE3(N1
zr5fUKMlzbBMCc^D$4vK*2!epR%3025yz}0B>>phNW2mc!uB(tDBz2|cIEjhni1_Xq
zf9^l{M+C`;<2zH(5t(Rs=Lhfd@+)t!vwss2jZyiE$#XwR%@TQj&mQiV4zuldWXSW}
zQR~_vrDQsp5Mq#BO})qoM+#N0X{!RID(Y&@YChv)RZ$Zo(+NScLv2zTAlVxcCL_|x
z!~$xaSFC&@d27uW+Y7qRSuPju5Lwf<3gPtrd?MX<DD=-FEwg6edh<jOI4BK(==(jl
zLR|NZPdjKYhr#_^ZSK1NxBWE&;pB9D*eYD{qHQ%OiRui22noW3RtE&Zh~4Q9Sr(uJ
z&3F`1H-crQ$eWHPh!LW+UZM;M1Z3keL7Wn&3CSoUn#YvuoVqMgWlj{uPdv-h#xohI
zIEgXB(gyp|zBLSb^CJAzJhhT{b)|1e!_j*5u*3G~u5-j$_GM|ac{yF@i25Dxv_~j=
zsK90K?a9QW@A=voM&U}UNn$IQz61ZNCmLQ)YfV#>)KzO==FaP^suf*r!HT2Flq8#4
z<yoinQC2x<O{W@y(1PP+5RjxPO;fU5TB_jjXoNPJY?RTIC66B5C!d`Y$0-Nbjxj=5
zsUoF`9e;NahFFi^#-?>^5Clwj5134*q~i(37#=@<NSWu9c|i~a>>cbgnM`SQjENKC
zY{IRbUCQ;6x+&S2?2;uZi{+e04<7R9;m7>!&wQDOSxRXHU8G1x4RIC|29mBRdHby&
z@Z8;(h|?*p(Ildw%M}Z~Vtn{xbnA0WTMkzWK5_03v7#t!2Z2(SJ6%e~;|bCT^5qF-
zy`o*7v-&sy)v}zQ6GnpR&V*V+6HH)qfJt{Lm92OMafk>*hm?+-3#Snx*l^ts)~-^D
zswi16mn;`cmWu^-U0eDR-MYMJ?qs}dTNhv{SE8urC=UWJvay{Uw=!gdf`K<CgPGAh
z=}A4|3=9mH!NR77yr_Lv*M(_XIR)|7dQ)K%Ax0TZoHFkcC=I((GfpF7VHl;7og`vW
zwyf%cqU-2PNGD@-Anl<Q#{|)?O`0bOQ95V6TF~S<RXu#ypFW<A1;G<>i2#8_b(*H>
zsOy#}iY!)TcybtX<;M2kzv{&FW1U-!N#~59Y87!3S`A>cX>agehPq^+hP`=mD~zxZ
zHlI@xLeRFIrC@0q=XDv<II`^yVd(NCY?76I1_3l)ZpbyaRt3UHbT&mK*PvY>F^r}=
zB<UE!6d44xZG~1frt4fklR_&iM{b1W<<`1oxm;0I4eMgf(P%{5HOx*Q@!t2o$Mdhf
zL76Xk_~<c7X2G3S8mvQg4U_3EaW*27As9iHjL7qxJ9nNV9PiR<D{uAc;SODyQ>+Wt
z>w?8>!Fn+#iUYzZB}&Fjr@LsSSX`X*$p;^B?b;F7Za+^b1O7jM^Z&^o`pW0Id32q3
zKe><679MbZ_K<hq{SJG3dmJ1dB4kKeR_tXdfq-OZk8Bb$N_UyfZlccLweQDhv<yLn
z)?i@0++w|=Q`&g|hEWz1=#KTt8O7oe<;5e4`HcDLxy`i%hQsT}+(`FHrboo%U3562
z?lfAub{ip>jK_?!r1xkr!X}>0cH_C$nzAaGE#{ofE;u<o<9s%!s_OoLw1j=Oy6h9&
zGO$#-GKvWz%MXRCuNh(b%D=$aG((@Aa_lX}*j%QtwPYhe*pnKcp?$)eG7O%_xU_=x
zoTO{KwqH~zRa?+z*PvBHoWx|45tE%M)14i%(G)E;i&j%=m_~+4)R79!wUJ;q6Rc`j
zR4s*)bSj_|Acbv0PSO!U7!jojNj4&%&si_#l=<>A*_~$&_oTDZVGt5Uk;6*W#8E^g
z#HNh-8LE0$xvBOSLB7tZ9BG-yONqo$kh;%#V*Ivgd=i7$Q3?-x^RC23Y0F_{jB#wE
z4N_Rjiy#OXjW^0BKL54JuWiqXH7b@1Nh{kPG@9N)p)p!Ap6-!Nb_wFd65tuD5Zg61
zRk<b#LXd{0w%D$!ap8<B#|MGUTXbzjT`Ru!Tfff9M;}nM1#f=gi+thm77rht^6<eU
zZr{0062>fNGiGNOjCb}pJUS%HM(plRP^#m@4?kf4@YqtlL?KZiXxoy*>jzwa{x)@0
zQI{1>-StxGoiU`NG0(sJIqH1P>7xhy;M?D1?_i%V{p3&byTAP{e*6!Ah4Jj1kKTQo
zi`j+MvkO9+Rx^pl$aKQ=@ERtNWV<^gaRgEjAc^7wYOPU$M1~<k34~T4pi`RVYDv?z
z7;T7zU^>od>w=4m8O3srt`-EQrKt-Lnsl<y?)761Z`>d~x=AwLN63h(YS3OEQOG_m
z(1$||K_wg7Xs?*PUawfLR;-sRilRgy`sU@@r5!A`&gFs-j;B1Ztg^v|J!DHSQ1}X)
z?iZOiavOyef3jkkGhpBwZSSGswYPi#qaA&(HC|dPjJ9|vv?7Wu`^02-!gOcCbTW2Y
z1Fmw2VbLi{W!Q}ZCXpeNifk0KlM3dQW?44Wt)P)%9~p#Uz+}2h5?evnAPAYApIm7d
zE{|uc?qTR`Z5R<auPO`%tO(nk=EI4*%4mI>A*<_#BDWGm)|;~EFDWHT>z>%GaS2o_
z>>2gSF`tHTH{OTFdMtHSBD5luwx=x1GO}#c=NLRe+U8o@M$c1?)OC$C0u#qfc3)(r
zYsQlUvdIp~WCx`>+I&I2oKaLcT~m?7F-kxjCIqshQ#DzVkr!)J+p=0M7>`FJNle=*
zKKb}QRn?Ma8OzgC-ueE!Bww2Hg)e;xEd(EY@IFOe@!XwTpf$_ojCal~I{e`18l&+J
zbybot&p0?Z0wGYXrHB3Ax85er5^g<rha|~}Q;BK}T~o7OFKOz6Bu<IbF}GiOmDT)&
zV!q<;tFN#;zu<dc{U)!x@f^SVTfaseM~EOIO(#ru_AL%8P06yDFqAy}_(Rrt$^PLn
zFWkP#?u{2{^Lw<5Q!6SNfH4YfY_DQfRy1Y7D2dohGs@)!>(!jq=_yUIqA6GG?quxk
zP1!p*V08REqvJb_c6LazF-8OwMMbAt0$~+c!oJx&y!x<Y6H2vhL!Pf#&X+9CFSt0r
zutFS;AIcl3KA~>}Mo0_R4Q+TQT|>SQxWxQ+D7l4u8tjiP8{;eC(mD$gW7Vm}71i^>
z((k=3tk>^ZGA&F&TSaHBiPDTD%g9Dkt0kC4gwocr6&__DP`8THMpJZ}{X{a7kjjqZ
zk!CN6SQLtjwxv-mhS15ihgL*&I>o3T5#n5F%@9)QF;Za=5=u*icez1kh$DKMQMxMJ
zv!ameUOh%T^mq^kG_|sbaMwI65ri^eoLPeXPxGMczX5DNU|AM)szU}6A#4#K-D3TG
z6075{;s5TNTLD5TQRyBgIv@x`raK2jVZiF*gww~LaC-8<g563R62wW$bI-rP%^SCI
zU|)A~3d*u1j3i+w9j$W3!;c>_-8&+^e#~mQrd5g$@7+U*faB{o_|lJmnTy$sd-p$P
zG|D(Wyv9T;)~hw|z4tD06tla#hgOF)ZO3@J$7J`2y~!?<(S(ogeZcp>@f}w+ONb;0
z-O(s34URI5cMiCD=Q*z3xJH<bSwA>oG@bJPTR-5vZ-1LV^wU4b*T4Qvq9kEF-X)16
z0@HB%$$LDw_com|9PS>ne|(2HN^Iq7y2tp&n=IzPMj$$yu$6+ks4U)QJ!cXJ>_w79
z=j8c};^LHYenFY9S+16}`jERXze2WmNP2jSV6;OV+5X77EU29z14a-AmP$B^l3skl
zdkdGL4{ckM=WEAm;9Fj7U#K#gx@lN0myXBErxygmNl=ARLKMc1qkI#7U0(KE`BI;M
zWxGYjh>g)Q16eQ+-5A_B+rg^cvw^hTdfI4o*P$A#lh<_}Qb;-_s2iwjL)>VhNF!s3
z3<E-e)j8CfAe6LPuxg<yVK*}DBoT?|$O4F_2}z@wl?|0vXk$>?GJXgkP7^-scs8||
z*23ehp1ZW9>p?*1yr>N-{ORMW3f~hygXKK8{dX3*Yn=BKhLPnfcT(AbC+A92r=u1=
zVGOSf_nbsl(=--hB0xw<l4e9n0=+Z6x$n)%w#l{GFmH`g+V;^2O}29b6Q@jf4k*hd
z-~8Qg@x8D83(h{dM}gq*_!`%*Ut@QFkDZ-8+O}r4ctl%TX3{9K`7p~Q+R=8c>rHF9
z`}|GL=S$K_#@$yxN1iXJJ4LG+W*6t=%QZ*WZ}63$`dQ|)84n(Oj0|8hnXtRN!+N>k
z{Ny30Cl5I|ION)mTbArUzD6*fa{KOGP9HtAT;uuLo}XQfgN9>-;{4IaJo@N8cCTIM
z)i>T?`ut5!KKd9f1po9O{tvwT`8WC8=ReQ=Pae>9HS6_)e7!;n!QTD>`v=#EvN6l5
zA#OU-AYeVOA-zd-=jUj;f<&FO{^V^QJ$OLATCx)h%z8;{gr%qLa#k0oT%2Ff2|;#v
z%+8$`7(f3S$#|D28(Z8OnxblGs*1oFfhbJKvN1_7;$gPdh*d_mUV~DW8-KZ6GM`^?
zadE+Fy`t_qs<x$R8`kS}Z?yb62?NVdZh7lNizs)JUfNjihrtTl0&0SVBVf}TKBTj`
zE4wbWF(|yj+GZ!+w76_-s!fPzls)*{ny#@xON2m1F%Z+#uv#@JY;}1u&PWnN9JxXa
z0Z2nYgN&e4f|FKJw6GsbLeUY3h6xgpg)Cbr8ru(Oz(z&pvv;UG8{!;X)1g(wHSPow
zVE}=2RJ_|2JYPnAmSOpm4fDm4x@l>;mZobMq0m@;uuipzKwyL+3}TWrC5Y_f3=K-(
zvSe?l|6W};G)-rp%GgsYiES0azj*Myg0>PQjg^)*j$_&=DDow$t&k!i5QZ>~81LRB
z7*F}+gSU8a?+27c&hsz4$d~`nPg@*S+Y*McGm0@<H{{EO1rP=nf)OVvU0or9h(t&}
ze&;=2{;|*T!Yem8Tou$+#b}gq^UiY=MM;sb31vW|YR=9dQxt0sj;`_MkN*S-lGBsN
zT+B{UMsRR+MAOuqoS$&<_>>RcdYhx$cX;lFm$>`N9j@K^0(DhW7Y$`$fwi0YvtC}?
z)XZk5eDMA65ymOEZa>FQeCfw|`}e-eZ~UWw#$Wso{ypA#`(5&UMK;Pf*xP3`nUaoo
z3BrU%8=`K_<NHevcgCo$W$*eCxjbfkJLSf%=AE-O>&177bjj(XIis{gw-M)6L%v#b
zKD!_oA8>f{2FEvUv$KChHrcmifSt`}HQIVYB8&)y^Zum(p=>!^b#&S`!?tzHYMrw#
za_Xug4kKFCk>_jbBB$$|$Y|Zr)g7u+<m;SO?x=i(%})v$IxJK|5XZ<U!pOkpiL?Qw
zQG;xf@NSi`R3<`7SIL$HNP0`<_Y@K%UA4{wsKus2-iM85;0zd$!lpS?*J0Y0u4yb{
zA7SxMVTz0rL?CIKn)RYZH3?nUF&>W@Whr44^h_T{SfqStKxj#+HL6zZM*%yLM9Pi>
z(J%=^W`X3q(xAXdfqIJI$yMXoAPDZ{@$@IwV09R$HazU(yUJ7gtZgy^(1Ko)NOfK?
zv~{Ib??HhvR`Df@q9<@Qz$Wf@?ym$IcRFr*^B{_1vMePSc!N*7e_z;b+8Ti%i8K5B
zks_PyQE9>Ve)Csp^EsdU;+MJi;k$hF(FdF@&$w~x4!2)=iP3lrb<6zxjKyL>Y(Pd3
zhashEEye75Mb_qg;q_hSAJ6#c!;knA|JpB;7gnTlov+a*;QGzm=&oh8T2ht;>s4-*
znK4v##m??N2RCkT`^6V92G*+;%kvqdooiI<CF^p<hwr_^2XDX2-q#Mf`_c>Cx^;_e
zH{<AdLKJ2cg{`o5szqW&VsAh963WzcMzdb$1g7JU{7?Q2f9G%fKUmF98D}AN9WmKG
zBFiSEqY+7*fsh1BvATG~Y`NgIFTT#T+t*NKj;dzN-(N9XKB9T>F>zf|EKis}K4U3s
zvS~(Db`XRdTz{T)_c~%{4;_tJ6fN`ogjV^sDBBuk`_=@;6SO&s>cV-hd1)Cpx~`*B
zs*l5xFhUze+u1x;mRTK0>$zHn&0@8pt!rDE6T)JJqR3IN#7-bHu<9wsfYkOsV_aTF
z@c*Oj&x0*Vvh+Ue=kBrG?X9_2W>#kHRo&HlPtWwM0EPo#2!KNfAO#~uf;3U2NQp)*
zCdwocl=ufFnh`XEQDzcJfe=87AR$m7G=MM^IGn)@W-!w|J-t<TRo7mXd*<8k7Mm~n
z!`&m|-uJSq5@JPLcD;Asy>TNv+>f90o$q|7Ed%XQ2b?yzp0^P>gwgYp(raJ3V1R7!
z{91Or&kA+F`!u09%1wjOHO6S1Z~-sNgO*j&-oM9vON<fdx}+{GqjAn4PebJDv!Dop
zlNKvssquzLC0xi7hSD;0j!}{@PAvQLhN%^p<Wpjt_l=Oz1{^3AiI7dN-MAlA8yXgz
z3cf$)h~8L~mDO5fwWg_k_t-2=aMoa*zx*`K+HBq_zb$4Y5g-L-O(4yGJ|WLf8@lzw
zhVuy#3413V6AK#I>2QF}uJhZ!^F8j~{*bSJ@g<6?;Kq#~2d=Px`~zNl?FanuyFX<2
z{4VD%ohK2FTuN-7Q(KJ?3XupZQz1>kSDwGZ%TI6OzWO46{qMca{RfYD;l<A|oi7+|
zH5?zEP?QZ@TN`X{@35H7m<~@kJU(Q4a_lFf)tIJWF!1sv4#CB%Pmxcj=-u;NO;hqD
zp(>V4=X3TBj@dt+G8$!^KYyO*pL>q&tqI0vlsd;WzU*hVEOE950k*C%x*<zbzVziU
zp_F7e%o*2~y7YCJhlh`RImyB|l)O9<JT=KFKX{FUqjTK5bDx1JaEtpaAH74le+QZ=
zshN=_CE8hz4v(46YZjGexVhlYgV)$w`fk)V^o_UHVj;Bijz#eSC&J@0!{La*c*I~h
zW-u7}x=d6E**&oBJ0ba?DjN(DGNGz{2b!uXna!p?S%P1zUK=W?Yo>#?gD;3)GXkGM
z7S3yahv2T=_@wj^il<@(&G9(@T3HB(h*6C6o|g>o>bkS|rnTO<`Oj$GV044gl~;y!
z#uqmyd8@=ARYxubuOzW73bd|ibc(S#y2%*~GLp2T=Sb&2CuKtA97mN!3de;}%1~-@
zS93m>WI`bqhW*7SmqB>K=AgFi7^9RT$r6wPZ5`Se-vT`BQva|=R}Mhv>^i5Zf+y89
z3<d+Ax^@n2{KJhYT`A6TAw#M**q)*-xoHCap!_-WPGa`4hAO<Qjb^c2Qa23>iA;bb
zWw3FLfAH2ZZ~Wj5{@~C53_BOjvv)t|Z~ljuIo^MUpZ=u_y!zSi^5EWW-hb;&zV}<d
z!RF?W8=rZZS3dg%HaE8@tAe_zsf&UuI|C+zhWFpQO(GOeU%kM)Z~ur(m#%XD@-=3Q
zCBxAsdygJ)GF!5{v%|Ryml=&V+1x$Py?eJgIXOW$4Z;bA`Izm^Ey}WB|H1o=vLVHC
z$)kfKmdldOogFS;zQXkz*V)+KBpYmCtz_@GpfHNbWI&op(o~RU8FoWrUGC?DNNK8y
zt#dmRiv{N{UE^h&@L&AZ|BT_rkom$3qNQnuREklq7#qVy-B7-Hj}xOv^D*UOpT(nh
zNZgD}Xh<AR83v=2hsPC1vzo(M&HikTJDB3EpjZ}NI23<a<LOyU;6*7!qDTir2Ez$i
zmXi;B8z!T*=SK;F6E@VkC5%QnX_BBDuMc0<zV~8LEWkR16rMwvbbKh0q%BbHpd(~P
zo*umpMIwUVu7g4Yj3QZ$J|Yc3SQlwdZSnj{5i;H+dko)&VqxtGo4WDQh*Cj?!I!U~
zVn3Ro?-uDk!cuF4EPV!CAy5va^G!Rfz*ALdok1E4dzD3qhD&3`Ksi$9$P_3Xrur0H
zTKV=mVL*jIB?(!Ul4Y4^XlGd{2k7R?M;+@WGVzFE<$S)NZv1Yo5=EMNEC;Ee(k-gk
zW?j#SCgV$W(=ac*=6Ph1=XuU}JPBaw<E24luWGw2tB`6=kjf_#Ha53;a3uKNTkr77
zXJ6#{)6cVY<vRbb|LZP?4}B)%w}1IF{0o0B=fdZ<7!4FhCztr}jW_tt-}_D8_}1_9
zbHDgYT)1$aBpFbiJm7Eq=YNH2aSv@$gtc6J;VV4<W52-r@4d_B&LwuXwkew$m89(N
z?QwW`ggkei(RhO-A8`Ksc@7WudHdb>c>jajoE*<sK2>t_`4>2U?OB}G?CP55lwf(d
z&z%q6<3~UI9%Z?ttO|tHSm)T7Y_qd-k;~Vf=EC)-xpes)qwxTPWLebM+L0w1VH3uq
zEiPWa&a7ydY@Z`f1rHwF1C@{sCat9Gb^_z(gz5b|3^zAWQgVFnfTk=k%M<d%;Dlzp
zv4uwQ@NmJc!zGW7mMm*WZH4Ff7z06Ma0O!SFKqv0TdI*m_)WFZNSXMKCV_)x9jvgv
zh4Xcc%OyqOOLEGprYPn<QjijhCoLvPj!;<_=EZqUkV$c!0RiC~-L~M71?xSy5TJ$7
zZ*_hnKoUZ~NC;g+$^E>Q!FV~0Hdv!+>YAo1amFH4;=3*-zM`UC*Uq))CGfPQP=KSV
z99oyurJ>4ehU1)k=y5kIqhWL)EHX*3PH|FH43%U&NlCB>3>U|OT7OEld)gX=TlH3y
zQlv?Wl9EJutip$cvD|nyOLo?ACQcPzn$PE<i->5OASBs<#)7E22P8!Nc@0Mj|25W9
z76naJw{d@#WsJw;mbB?Fqs#L6baGH$aH}noj4|F|D#v8BNj@BL>(-Jleda}8dG&J`
zm+*sc{T@~I1>gn%gOOmkJtj%U48|Lz(-VxbTz~O1y!Gwh<~!f|CNF;Wb8PJFP@Noc
zy#Ihk&%h=K4ENuAjTgWC(=?{x&Ycf<?u8e<LCOXo6OIq|I5;{)Z*Gw%3G<?4b8DBc
zeeEI-@7?A02k-LETW|1S?*Z3uJj>N<*O+W<qpjoo)n~YP?O9%4&N$k;kFFNjrlG0|
zrqd&)CsSVg&hH|>{kvr2F*jd)iRWH^nJZT=GM^S4A5E#tnnWe+?4Aebs47h&9LEP!
zvLvHS8!qllxW2i=NFQKN?ohjm;z(fYf@Xe#H4CJ49M24LIKk?SN5@k>*jsY%sG>H0
zKB<tcvTe7bkDg}^ANC93YvhEpSZ4r>GY%`GDI+XWzi1@}t9>@im=IE0gA$$xg%iPp
z8X(dENtUD1wB6L4^As5`$KLJStL0fh1jyxTXPAJObn`Ma*NokPW4I;gC8`2W<(v-@
zBX(>|?fV`o89@Dkhe{H!Lms`k7<xv(N;uVCjJB{WHC1g`HiluX$%mR@o|2^@i%Ae5
zc2J9igUT_^9J`qe0KZ{-@W~;dCo)noMDhxS5txo?+ekx@fz<gCuFc0B?I0t({K~Rm
zxm=RxLy~mB0ErHYo$YsY&vE5&0cbB<Dc~?!L;LQtT3bDi2_G5mS~p=bSO(dIY?8v@
z0?)j(%hACBzx(UI!D3l*;o4RH<o|S!|Mow<!2J(X{s;fM;iWH4sFyD=-rnN&_cLz2
z`zG1WgmX_n!=t+&vU~Muq_W(-cZa54guZ<a3lxiE-u&i&!`J`ue}&ocJ_mc-3^!|b
zcFr-laFHy@+1q=_(a8~$p%07ex<WRFE7z}c?YW!m-G9LS+qZf12jAh|t@pX~)Ki?l
zc$w|ZbBx9tjK<q+Z0}N*1+(KB&T5`gjzr-Lb!P|c-QVNxd++eU>#y_DXFtQ|f9%IN
zfB7QVhH6=|m@lczlEGlel`Gfy;KSQw<1rVu$6VV=*p<ha<t^sN_ej#5+42BgE~!k#
z`CKB#yKKDtV{E<jbuMjg@Xde!f6v{+V;Ywr&J5{4+2I-8u3d~mc)ZG+C+{t^OEh?9
zanbvTA;TJrGZtq}z?q~E@6rKDHbA9$Perz;H`bd+{W<kGPTxnBjCCPl3=_M}IDV&6
z${Q!!;7pBaN{rSX!%9?WIx0h`(-kl~AzKW$4MD*Jvd%a4Ksp+2Fhzs0hDJMVt+B(L
zEYFZCcwok2m1LnU`-NeY3Q{Q%<Rr3OzpKX{PgWLV1xEYysgQys32D;cL8WOY-BQ;L
z{o>3gIzmkDY!nuoGoLNVvk_S~@O|nZ%UC^45UHDnqA1W!)9$9R^Kf4!;Vh5NI)_pM
zYdS**QK}|mW9-XahTG3@^6n!}PAY!<=l&2om(Fv1aEP#uKlMNUp~q~L!__xXV#G-0
z{H_1@|J{c>yVrQ`<<IlO@BSV|S>v2#@9w(@XM(%$u?k@uZvW`pT>sHGdG*Ub$-$#L
zOzIL_7wn$9%!LaV$cIDr_Hq^{Cs6xIB$3REB_i>P0blv*kFlsM2YU~Bc=tYc-g_S@
z1-ln6aN+V*wsv<of9@QcPhX_2EsLo)I@xf@g~>&(Tz#3v?2t$IZ}H}vZ}4FMfS>&7
zpTmQ*B^%qDj5h`x?eBB_=F|M{ul|?3{LBtFF6B(zg5~}#>e(T6*`TY2<D)~60w;2c
ztDooUmw$mvFaHD=o_>za?UX;>%=th6x&Jv$HTC1*It=FjpChL0J=lDfQiccIod^mY
ziSLbQt-(5rvB5m3*ljp%L!<S%6ih+ycVBlGuj8Yx@vNi8q5A*A^~(UHTRaZc=!&|j
zz)EE5Tezz%Cs8R#iE*wCN2QXjd`vv7wIE%nQg*>R^j=z7)Mx`%Cun0C7%vr^qzS@-
zHeTd!S!*io9UNyJP5mihQfZp@G@dO@V_E{ISL;k$jj7mf$Oqa_@-0!u)DMzxDYfWo
z_>50^DycB*{p8P>v)cIL;W}`Agma|5Fz>Ny4bWl|Ce&p`MsZXo9gQ{^j=(C#!-Hck
zU)tg6XP#qzbil12z2UiJQT0<BCcEb`R^#dlvW!3a$Nn^b^RN5|6!STkcDIn(gh7^3
zRhGjK-srFi*R^%Gy5jr4_SZRo@e;cipW^sH)0A_Xy5!u&t88xVlB5ZbP!x-)7hh{C
zbgco$;>dE}X|im<rOQ{i`Rq-QirMs->B%v7Zr|qpciv{0C7i!{o$F6O!+12Lu4X8e
zQdf>eU6N))o_gkGwstS^`fK0g*MH@2@iRaBbIi($*~y&w^nhWOaP!(#zIyWlJ9dxs
z=!mi@SezVER&yF%Ve1BC70KpPTzuhcTzvIs*tq^Gn^&%p<_T6ve&Ub(68jGy@|XYo
z{{dX{q|<3-f~-st=T!(i$<PFEYHc1ke3Q<AJEfVQpB5K-E()AVNYmUmtqlnIX?^x}
z1L)42uV<-*oZ3LcvReOqA2>Ir#@ZU)ROq?^3reJ|F;&Vl;3E#@vvW28`=%}AM|7j+
zv0#A`-T*n_LuO+Q%hIBarO}$<AoCjgc?wRDNiRr%NT{{NRwYHC|9q0;$@=a&ixi=D
z&Kiu>K1LR@P4y<hEf%6^i;d4ZP7Qo78`HYtcpAfEzCbs6r8$F2GY}1DS-@DD5XH%w
zx^CTI8N$3+-xpuEYsN~pZn~tZE0&8T)@rXx8Zv@|(GWq7Fcn9S?owNaBjsqmWb^z5
z@?=PsBpB=1*geN&W0U>+w<#7=%1U$YsTcU;|J=XIw|?(CIBN3q=TTY82S4}*#q5A?
zG}_@u3YZ^1;2->*|AIgEXZ|Iebriy2ousZCcF!%?xp0LW&)($G!$%bJ8S}*)Z5HH-
z7r?8k1w~!(=G$*F9FDnk`7(>y33X9&<N7tW&z)mY73|-8#5+HFljmM|j^S{??OPu*
zNK?k=b}8m_#6dxY;Y&aMHSWIsKEM5&zs<9seUWOhAghkqfAB6-TX5~-7(F{+d8C;Y
zhhQq|s-dbib>lc;i``3~hpV3@*?0<^WVu+7k2lC0#Yvg*@>l;XU-`<LeEXY!6`j?f
z71C9{OyfxpCqnoV)Ah9F8sLqPch>p@fB<VGio|6&AyH}QBPV><7ldfTqG&}V(tCc<
z-C4bmQ{J0h`D^!)l`v3re-+^>Js=;jHe+d;64TV^s>T`TyMG5jU!|$<coR<JVXr{n
ziD708eXZwU`5M1K7!@+KVGp69b69PtjrCG5+OJV-4TCI0N|2f2urQdWqAqHt%TI}d
zP!t8j;jj(wtg~K5CH$*v-(jU`8gGFEPPASMbILfKaoAq-#O2a!p8Mts9=uG2KuXc$
zj`UCI&lw9pb+57KVCY+{_w}pJIvk!)qqX+Mt)Vln5We}ua5yB-5{xq_3ERVld&>c)
zsc_a%E|!?au@I6=*KRV|+T!G3kK$+_m1S&hY*98Pm!ExwpBnA*&f7oa=C!NL9=y-3
zAN(%THtT*O!F=N!9Nc|}H@@@t`P@(bQR-reu05cu>zc)U!MV#%arvpsOpnJ*>WbNH
zPBEX6r;^RhElksudpx{zpKI5iLTozT{L$;o7IR+x!YjP;mCrDpPPy~&E;p~=;H4Ly
z<@MLT$Nr;7T)puu8#k}8IGVEe@Gh5cUSqN~;{7+i%jKbFc5oND+^48cn9oj_og5=<
zgAx@=2<qBViZQolOYGJ;hJ*8zm8C2eSS+TTql*jZ%RBtdznJp3|KbLloBt*oyYF&*
z^qWXB8nkEqQhmZ1+~KIcz_-J6LbTHg!MS#BcAib#<g(PaLAMsPF<v@X^wnac4;&Xj
z|JB~pT-z8|erst29ViZ*+09wDYt5?~7~No^7Mn4?ReG8t(==2yqy!|t*C*1CPvinR
z?8+t{N6#A5(m7&=++S``CaJyXq}Cd(Gu+UkMmba4aL_cAWlddaj_03}p44PxgEVwY
zL4ZP#B(6vP!VAz_OI0;pMR@NxD^_=Fzpr{Wg0d(BV6AC%jW!J#gH{nCoVa_i!i(X>
z_T6!?V>T}~_KdFax)&Ay773*ua3`dh6Go#6!{Hd|gfDNfij8s3U~mmx%$e+@j5anI
z498gK7z}cX#S9@OS8l!#^wJfz@r9n(FJ0o>zxm5xOWywW@1TpLGZO)xL>zeYrs6xl
z^LM%V@|PKnwy@SvEd;jFplj-;V(a`RcF$jAC)r{&nQ-rprC7`v4!4+WWb7)*xw1k^
z$;rtPgF(vm>(|+T^nmYu<M;U1Z+wgAUwn}(PhaEiy$4*la*5CX)K75lom;%|+V@Fw
z#VeotBCovq3is~ZXQ&)c4@-)>Zz1YAx}0L`0<)M=&Q5SrBUOsac3Dn#I57!(a-VA#
zZ{U)I*>u6(yB{*yoG>14p{s@;{to<|zjOr<Os5wZ4E}ANy7(t~^ymnq8gB@mpgmHY
zgZ+Pf@1b9U(g&ma^FoLa7)p?VA4CWNJ(Mn5PaH&VANCLR%w)D(B@*-v10S)Oc9j*@
z*SG2hQ`dph<!kbiBoBC+FVR7W){8oI9l=05*ZcDs&PBxX`Z$<Bufu3dWt;Y7#u%np
ztP?a1ESC*+{mFy&C!-)}k3k8c2;nQyWmQU|q^Bb!iEpvorCdMyyzOLo%-^fJYTHy9
zV}eYvLn_%$P)C%Y@H<*jmcCJ;)*iGBVO}4HI;+^ZPljs+^@2>Y5WS`hfy0uB6ax&l
zZg6e%3`we(Pmeg<f5fA`J&b9{rRDlF&vWDX8>oDO)gXkWG(e;I+kf*v;f-(n3eJ|k
zgM@Gx=UY?vk0=pjuqYh$V!=0l^>6Ttf9ik8+-S<;n5Hag>I_-VS<X*b93OJw(q+zH
zJkKD%!TayO&;H>7JG(o+%kTCU&N()v;NW17;b_8BPd~%OOIJ8NdC0-RZ9e?1AEL{O
z$!LSE3+K7|>@&RZ>Wdsdn)3VK`duz>4mqEuJb3@x)c1dYsS1DXrlKx<Pe83J3M()&
zLr$K;jL##qVKQH0<`2<Fl6riOrK_>F0tDG)Lgns}q+8V07=WMsqk^COnLo_G^}qTr
zQN;jbi%$rk_c?IxOhe`9Acr`oL=Pj1^Fj&_diz>VzcGUg^1nU^lR^gAqRSZy*FQ5W
znaY02RcF(}YeKYpRs^kAGKJxs)@rY?X?#0$fzL|H)T=R&q`r%@bH1buZ%2p*N#Mow
zZ164P*aXhv`kLue9!$)dBk;U6C~J+emc_Dc4Wkmi&03##_;JURHDahL5E3bDFjf-i
zx_WqVHm}u_rm620{}EjPR|<jqR=RrwINv3Cv0Tz<uX%1lU6#Lw5buWtY+YAQ(=<&e
zGq5XkgLXo+JLu`d7}Ejn;WJt0iJxie4V{w?6}k-9atKYqvJp(DN9cNvawTQ4q-ZP;
zzwu2r-+YTJPhI2Oh0AQ8+ab+Il#7y$VS>b=@p@LmX`f6vd(iMGfdZ$v{m%FJ{<puu
zXTSb4EEWr#ttiWy)YO=&#FPc|<0DRvFLUYIHC}!7dG6l1&))q#PL7Wm4u@R0c!90$
zEzVy&=R*OYsVbgoUI=9rUbV5&70bnxlY=8}z5O0iTP|;pFz<hd5BJ`rD34+3sRRPU
z)+D1Uj}U3fQKLCHDOoP(=qU^~_bBJboIJeEzB8!7CL25Fx$xAB>^$`hw{N}2xpU{a
zu_5>~fA%f@n_u~PzVhRH{A>T_-(%z4%Y1PAr})eN-mg)*9HT#3xaS-JLg<G${lD85
zi|O{)wK*CVVvtw@d=@$gpm-$8wV#Y9S&;}{rzgaMU=i!>+Ux0}gO!c7_d4K9#7sj|
zSLmj~gkE_nQ7Gj@Q<Y|+ac$=%b!5}-TmqlXw5=D?c}nYWjkP|B6$J#Xh$UhnY(yY}
z)OeJhF#?28So5is#W=LqEnXv)M2aL_B)&66;^TIeBwiIk$BPL9@N^E=<Bry7#L0p1
zEvg$|6p6K93?h*T>998LFdk?xf?8(;^|CBu61wB|J*M^V>=z}+;^qj#E2S8ZC*=7c
zz8mSlgmzT*(GRGGFXM*W>|VS=vwa@rU^zRXm><zJ6=hv<_ttydzx}?~WV-Pj<5AA5
zKk<k7&A<Iud2s7BoJ+9^oL-Ii+Mk`TpSQvyrN)r(oxk_XT)23J&1aq?EgC8pcXvxw
zF0f9sf3IXYJ!bdvQ(S%e3Qs?Glg0FeySE;)|L7s((U`66Z3Y`DgHg_CV}!tBw7|B+
zenX-R#yYB#L+*e7*Lm>$zl)vUr<|X#SQbc1szy^TOE$CX*x?oSYsve2cbJzA#({YF
zK1cW7_E?GX%`I#-<><kTqrDHgUj7s}UU&s_^nLVr%FQo0{;fauYwT`sP}Kv{@;v{{
z|K#WR-QRed@BP*-tdI!ha3(w#0Wl-mPNm@4cz-FjDYwCp7(nAhf?H_`6!}zZnK==j
zNc1EU5C)`;Mkf4C$~FOna4WEaU+iYZBZ&*V#jr$YQF`cDka6kF3x{84rokEQnF1<9
zBpEWzNz)vGS7C6@%Ln6}*Q2trsTv{6sr>b~)qp`_rDv%79+VRSNO#eo(-}}}u;G1D
zL{BoF^;1$o5W@2)Oh`2f1I{%hp%<SBh<}twl~Q6>`bI|svAQ!nt~91^v16Bzrt!@}
zoprwLI}#gI5rm@^>yL+0ij9qpwzbt+V0oVt0<@l3|6vRR;J#8e19&rKNo+|~+-J5p
z=Hiu?*ggL=8pp=YE_Hc{vRqJ?h3`A2z0c&TB{~H)n(*}VpW)%1*L}fqC~CDr1w-vZ
z2B(kz37=M0mZn*<_uvi}UU(T2pH#5M>jp_cHzn3K)J@53e#Ft?LoQ!^mdVBjSFc_t
zO&yJ~JbdtwhYubig=AxUhwbf6Hn)eQ<Akb#+1%31X4DVf;o%#<gTDU*^w9_CqC!`5
zrt>4}Is*mS`R8~zy2{;qkGON^HpS9{ND(0{kXeew`V6q}i$d50?3{Zae4o+4aOwJe
z)F`K1?y&!0&LDM+Hv~;FC*2tG&;Khw&%ge?zl5IqMI>a*PWfJZzQuX~!wJl9;mc77
zAh<(>R$9I$Et;`Vao5e<p7*^D4Jhky(zegU0oECUKtP;j;MNyrkiO}G^VXV<KNUj9
zxPf*CEVQSx&U;bXH&M``otu<NtHO|^nfF}cdTNsaKHggo8>3eTqLtT<e%2{GxhE-e
zR>&P|Y1_YvezyNp;z@aTSs{~zM*B>g!_w#mXF)f<d?c2yDCGs&ZM3p6HeWgCX`>Wx
ztnXeHfk^4OQl6V3g&+y+b^$9~rg)CgXvAnVTDy<ciPk@Sw~u9;5tP#FV2{V+PHCZU
zP(wSQRKkd3?%aFG{oA*>`rOMr^TOvDjdqY*+rD*G5bi6MCG(SGoM~`2AsG(HvVm7T
zlnISXyjG8i8O0D%yRI3j2;i*0l7?b>$Yi|9;p`DYdWjM#qLfF1HIORly}P3$_PIm3
zwM~)_5GbUuT)TFeEN>{5j>AVMJb1XzgNMFD$kt|#UCudr;~Sj3`#a3;y@f276w4!;
z+EMBn<G?cD(knlO8}9PPJ0Ec8-Mdu99#S?w6RJEV10#c+2oSs*SfZSwo*nVlcYc%U
zqfIVecoyOI8RnANNzUeGLzWjv>G;Y|UgM|#k<aqGf9Fl|;gG?2%xGi6cr;{lJR;9L
zcd2PKvtmwF)Y!&g8{fLDbYK$&LaliBevO{)6w{wC{dps$BMJ8rc}E@Y6;A-H@vvfz
zV8qLW5J9iZ!xkolTr1uSzS6-%MqYYGYp<^9MfW5^rKG7>HV{(bWZ2aDn(C{5Lwm2i
zXwu`#0nX{mz{LCajCIMGRRC>Fxdu%~jPB#YsowEq^1;S;G>L`2+S<^@$M|U$7NOUZ
zvc74cP!JUPy6gI67ucT&U84gNuEY9ja3w*<)>fTyan7AGE+6+@QB2?0<@wOCubaM(
zp~W!QIfb->M3r3H$+@>b=fgL@$Ns~IT)Fl%PAP_?A^Bj)WMhNfi<|6R*kn0h&@={Z
zG`p8?GTPW7$6=L2$_#;NcQv0p39hsB#Mv~>K`MmM46=+yTc(R8=e9=R{S=>pu@04_
z7^8i=ESXYF4;YS85+%S&7G=r9J0DUlPFU)So3B2{^Ea<@c>9ohZ+w@%=?5$i-lg8Z
zhdJ7#Sj=e5f@NdLl%vr#yU+dvH^2JFP!}%q-QW6kZd}>rkAMA7GgJ*uWu%D;*RL?b
zdkik%8g<<;FKebpCp<bl=B>A0XYav2zyDhgc=m-c<>HXti@Q`!%}5V0wIiFr|Ki{H
zvmE}@H`o|Y*qm&#y*Xht8X;r`!Ap`tVWjuKn?|#LvSk0k10LRbkN5B1<D1|5J@y{%
zdAwMxg6GkTIYk%fZJ-hmDi|8=(4paG{O3I&@4@szaZjHJ9)}27z#!$KqVbNPetSZ^
z1da2VJl)hZb?Jd}(F4j=h6u7euH|V30~<JDuJ>AJggu1P-Mv>ti0h-jUPSOjg%F{c
zF5&%QjMluhHhhd>;U_$v43lyeZ8bPwL@0zLNpq6KTUsS0jrI)hn2_xruHRima8u*J
zoH+qje(}FGjbT3XK(^5hO;fkZpbke}d##+v-4H^wz4!W5gOyz}f`MW8Y8wC0l>`wI
z1F?K28oqdR_?5>FjP)`1#<1k#<|cVPWI8|M&RYkh*_h*#6GW0Ro@{dU+2^_W{PReq
zDNl|$K0M^hf8ZDS13&vkp1HKi?(Tp=J|Z22CZA5W8tndl{bY1baCoxd-mQ0e=e>vA
zuM$)?03pFPSnEXsjWoX9c4LrAQCf?{^8P!=<oSd_K4fF}Jm;ReiZmBFe)N!E`HTM#
zUcPaKr!FKc@4UuDH`I6Tb8>t{Gdo~W&oN4p0EKW|{`?<g_tjtE`)|F=-rxCay!Y1c
z^M`-&Pw_|o$$#84wry`q#ra}g=SWkZ4W2EQ9PAx%u=kK3z4;bzy?2Xbv|#U{<nFB#
zoc(~&V22m}^ko{YkwuNRhRJqDrVcs&-kaQCI^L^}QCcCbAyHoLHcbSnNRY!Jj5j!G
zY97rO+*=mhfAom`z5SLl@NpczQ-2q2An84-Aa`P%M|GUT-+P$<-8tjWq4RfG<Py;;
zA;fvx5-nS|$6`&5!8-{)Ib{(t1r-xoLBq;}L9`*F>(1*9RWMF6ycWLOg>@B;%sNc6
z0((S|ea#7Rcmv!YMi&g<lhj~7k&!w&J|tPpy*QTiHB101h-%>^x@kxg-_=$dO;a1H
zs_MnwXUDgXJ>ulCuviq{(+SN&tud%9#p)&q;ss$d;i&?H!C(b%S_$piIN86yX&S1!
z#2DRbVP~Pmdfz~;y8Cg<@RPw>LA5-_)+K3jkvz{S7AGX>n9c1S#+y5wzi^S`!()E;
zSAT=et!*ZoV+Ml_?%&_z*feMwPTo0UfB%U2d_mE+NK%X^6DAuQOePagDHvodXGhGB
z9&mVYM7sSH=gyx)*GqDx5ap6a>$ba&N^_cW=^Fz~HaIsNk)%VCB*W;6dV0j8`}f%Y
z;59Co4;b8ko8w1<`FzS!8;%biad@!LvUU`@A`>a;=7ft^&T}-|<AcBdSJ}Jw9+z)^
zmEvdvcXE%tJ9~)H7O99HMVA#4LNXX+SX(ijPB}U{X7Avb_uhS<#cWQR4ng5FQ$@xv
z|L5Ol?_t3|`b(eTl~*T-Op*<2HeV?iA6~>wc2OH6-~2#G<f6b%k8$%krEaik#=<$Q
z_5gm<XdXUzgl-I}6h85@zG*slrm<?@7mSn-T`fX-??+m%OJ#ca#wpyUZ+CjXx^8NQ
z8P<h_ZM0I?*v1PDdi*HyjUSY6-HtUuN=Aeh>%N<{=g`~kOL{Ir8zn^J>fONBlRKO0
z;o4B~tc)qGItj#5w@)?Pd*!8<eJky3M4k_bTwN)mV<@P~lIh8m{lk48JlaE-`&4BW
zHVtpeSFP~ri|sA{?lmEp&*zju<<W<H4azCs`rR5-qI_*$nlKs-NfPxDgWDa}8tS_A
z;&8!RQ7U0D7?7svCmztMs;CwQqaS&k$L#=ua<O2vxrMWi58nHbbC<60g`fBw8{0dO
zNGy)six>IP@BKZd%bJIWOOQjh&u??(s+V+?Ds7892ZI6mFrzGHG-XXwYxedYVvc59
zd+xK4W+)uW`b?a$4Om0f)EJf0s05)DRyRy0ma;CWiY5QefBj#eAHB{iH!kwor!J!x
z_ju>+51F2nWQpWtdP1QSY6qkVmtTCAjh*wn`|bzq-I}v7DNUU)PYx)LYM#1sfoGrn
z995p6oTjbkbb#<JxrGoM9qwUIrf6+BIy&aoox7kCzgZ}7LZd59mR;fX-!FLZ{tx)W
z|JYUj_&@Pkw#Hl7aZYh_gZV*&DkQm4*x`^Q&q&UVa9a&YRnQniIh|6=iUP@TQS#{U
zD0oDHp3r}wXU)wsHahQ6`S3@?7{Y-O7Ax=tyrQ*7-N$?Xjo4K8d%T7=K=Rm>F}Tn~
z!8972Z~CCpoFogGJn0n>R<m-(7xlKSu)_C6uPXdA@%hB73HgA?6I@Sl#S+3ZJ<+xC
zYW;#3f_}o|$uKEhQ(3Q(T-QY#CktF_e44spF`rYF1-hw%@V#$d&^JD3g?TGCcg8D4
z;b{R{>%iU7P}hLR$s%5pWf@7LPHlke?rPkb%Ce*kbOZmlgwbHo)0uLoGJxyni<nu{
zXq3~K`h>w?iv^O6jR`2l$#llen=f$T>P?b#gwq1$EU9vse3Q?8;U^episxVOLTW+)
zFG0}0TJkjY!fZ~GCz3=eq!QSA&irJL0|oP=2i&`Rhr^ROTbo;K>|P2WDrA{LmrdKk
zXuh0NESD_a_#V^42Ryp{0ePNs>EbzV?k4>7pZqy&wWOS$a3UO7OEH}yCnJp36b86>
zj?34c=Ha5@dvENsyZao^J^vJovZkm?P7V*b{>qOreClPKO0Y$R3|d2;2x$>0QkhX#
z6}oJwtBTot&Ih+{Q&bIUK12y$0Bh|WEh|XIjK<qEmE~{zr}udMd&m4s|LRxx+~+Ui
zKD&$l-qX~=poSBM8)H&u!Dx)pB+G*7^ccHbV$zISBs@AeVX-JkQ-yQB06T)GLI_x!
zFDn`8-aK%vA?e;Ll?)1+)+h5L-nBM&`U}5>A?R(?@qKt*D61PUyVBHHqrG8N2?RAz
zCqmJ4qHscm@XdR7%7SsR?$)#|jinUr<{UGft;q})bA7$WF^KR9{jpg~hOIgYop1O_
zk0&cIwicBpC@B!mqNG3yMXF?oX(etjAW@!^QdTv~c^xi1R9UYbA7`xM;hb;K7(2vj
z<Aum%nvf53KiHx0(U-lf4nsSrYnIDr^A;go^zr;?Gzd*2)hR40oNeX0oO7P|+0<BJ
zP&y;aAk8gVl4G4F%O+eoe-4#qESC#XV-Rx8Ao0bjE@=?Taqj$pBuQwRhVdjt$&|&i
z0jbDRfo_~HDy}P>Suj1>=i$Tq93LK{8_n<jy<g>pS3b*yE7zz(?N*YHa0nVxP}dcQ
z2YWnt{|)v({1Ku$;VWNwi9hl4Kf|?)J2dlSP8JJ>a>@Ao1o7-8PNqj3RyED!9Cqs*
zc6XPz_Lp4Py}`|wCsb9z$!yNXaKO#$FY)N`5ho8n<n`BH<I7)pjT_H>31clyz%7MH
zKuWBDd9h$o7R(lN4v&twbN4R!U>GL8MG48q&KB>#bL8>Qfg+PbmgS6hUa$G{|K_*(
z*ZwzO=Ecto#yb}|zZj8@2B^A*Wr3|L9D-UnOkHEFFNB>JCHL+=AXPp=;bf4Re#~5K
z=gul>5#|=HRp~(ENN4mcYuC~T&Ys)x@JuBU0Q3%g##-M<B9w+%r@NZGEJucV!kAeU
zh(Nsv;5A^CK#Gt_bs(+voI`>(y0yV>_2NzXo?dS)t!MhqxR-T9>}^>zW@nGZS;rF@
zsqNkKBx&kHtx)J&*P5bg(Ar|H*EwIz=bTKZ%w}`6@j_cxhud|l+D)Gvn{q8Zpsp*N
z^T2Q-M<i;1ase2Z2}&hrhIqn%VK8XHi&`u(wQdJ8O|w>KDqdfloGUqxI029p#gd{Z
zJKl=Q7!Oips(IKn%xlestufXZ%4I<!AeSk+hC*u6Org>QCx*<Mim}PiMo`Wha1GXa
zc^K^+GLfN(R5P2`%okH`efR+aL7t{q;dt+j*RUshJpKGlE<JUL-}&utuxJ`KE?nfp
zci-gBkG_pwOu4+X!O#E9AK|CJ{$)0X1M1l^X7-SQs8}e$Vs^xGIinUS^UZVE^H-6h
zoMthH#f&09A<a@YMkBU1CRn4fw&Lp5%UJ99?)Sfm`p^DT{@H(igD-vYb6D4S^?<;o
zGY+gCGM!4ws^q=*KcuQON~LWx$s5<7;;Hkyy#2;&4AT^eMK?8Afw785`wf5o|M(65
zXaDOT<MW?6LN0C~W$G878{sK22?$GixP^G|h-LX6xAyK+mNQfm+dzfREm42m`JXJ-
zr|<(X96cNlu?9A^HE8D>Um6$4#rU5i&&W?9w>BHQ5GkTTLP#V@0N<?#wynX%MxON&
zV;Uq0O8J6em1M{y3t7SrQ}So*Xp=H#9b=>?Wk?WtxhPzaPY#BzA1?Z`m@aYSA_hko
z2iwMSJ%dDdf6;h~P}}%vg?qPd-|b;E5loZ8+vvH#QlXNRB*{pU5|t$0u=PO5bsp5m
z9r1nh`JB40vBB7-NkWnh5h1haIGb1&&(Z&Hu`H;o%4a51-;!%GnY68oRzKXs5F)lF
z1yxzKyeh)v7>ovthB-nx#;N9mI}g~{+#=cCMr4{ry+G7Cn|VgYrtfkj9aWYh4<50o
zzm2OMbLY7)T6+P(RE;qX=EaifY{CBi0UzGJ&+O!wJWtx}<<L3Kj|W`Z+~DZpBYyDh
z-(Y@xfKVy8n(YyM{TIH>*T42PE}!2)mrEAMk66qPak|FolJa<;_ir8Z`hLUF)(xsW
z#WqvaJg1z_psvW3B9VeYo?>l7o+a#Vji|~6&N`lX>2o|fIOgB|cm8dD_9wrBbQLm5
zamI#Xam{>DaWb89GCg57owGZ17*ipnz*vW<7n~eaWC=LaVCsaXYM4)tNRos!YY=M8
zpZ|Bh&7b`rH5@+p2H*Ovd!(t)DoLqOGNrMG#bU{#C|MLs_V)KEmIVS2C`$=4NdxN}
zz%#_X5ejT^>wwTfrZ;7UP;DYWB0z%Gp2ufJv<mUPelqzf6tj0R<n!yJReOPu$uro-
z7Y%FE_zhAjRF<O#zKMdA9nGez^rOYwa1LkN-dV(GR^aG+ISLn|9v7S}FVPaLaxkdg
zT6Z;s4xvOiZ^WDYSe=e&!fl<oLeD;)ka6ydi{k|d(OrZPv=7zO{{!D6U{aBM2z{6D
zlNqS^8RuN9Y+!?H>b)nBD)GgU?GCc)rmg~>j=HK@EazCGJ$@AmkSCMLDg0q|47&T(
zT2q$3-4A%SaF*u`hF(He3BywtQqp9T<>H7+jmYv5Dj8xL!?IXnRnA}|ClTNVL;l-e
z{XJg$)&B^YjEIzF5yn!4Mi^2FQmIhDAeTJ5;}FU>W^Ei8YuG<|pGV*LkT>3bAD!*u
zHd2Huc=qBI{*kYJjxT@lGe|UMd4j7Kq(U<mHOu)SZ{2#pYaj0OVWF72jB>F@Vja?H
z+;E7iG@>>vP7bKGW|&Dz)9`3H=SM%7BJ%;#7!D8iS!jtRA^GOF$gHmfkWRFdI)>9y
zNH&IP@Msdx=@JrKPT4zJ;;g1Dmf$Q+RWP3&ppul)NTSk=!EnTX@)uv@`4=X<`OaO^
zBx_M3B^5YBRo5(+OUkljzVMk(-#}zQmZexa9(3dc#gu*Ip?iS5qq%roGg4tZ(?4QK
zenEGbljtqj^=Th7Y0mZ`5$nBEU3<c&sd0c(URhJ+KFkwJwrZvA9ChxLA!v!B#OPqP
zbf|M2{OXY+Xig~+K4n(JKD?2G7~cBGLbbBQtM5M}q4<f8CuE!(pp!>kUB5owl}aK=
zFpZ*0Li?-mP+d%kt{tDmSc$MxRaMPm;Y(kJg9$}hBGVM(@H=nZmCi^V#{o0e<4{I7
zz6Mc>R;@F(X+PT=vmx~IjV65ekRljRgOq~Na7dCT2&b^lk>`f%mp0hnFL-qHE;3En
zICmMFr4&uV$VfOjqBaXgqX93x`Z9m*KYU18?SozAnAqUXxzGhyIj|~eyOt2RHbVG}
zWu6QvjbL&i=f=*6Kk$Vw^2IN_z~)ARsS2uUfvx8>^&EktT9kbMjSu<Jg9Qi41;#rg
zCay+G$Np@=v~Z~95?dIuEMt1`fGq@XzkWcLrHscD27?hoK;sfFzwmj45*$2u$Wdjv
z`^FoTi&<ERA*>QH6c9m#)P@blZ)(n1WFi@kHoQR@3|Y*N$%p>)wO+E7jH$|oqFD0(
z`5W)D_vju<C<u)bgbZ4A!h-NdOR0o3NfB7zmrzS^jn97hRA0baV)i<Oa?X*EAeAJ`
z64ESbm36hTlyyT@H)vE3(}~O3uB^}X877=B%?bUneVAvwvVk#mm}Alx42L#U-T<GS
zRrBHM+V_c`jByL5b%$(CY2Ua+E?C+E<wDSA^?QEX+PAIgO^NP(Hu=<Q_u{{}JA}ax
zF0qIOi`LhSsZdrTWpry<xt+5H_AF*^|8Kr2=e$)Y&^nZWAln8D&>OfOsMlum3SVO5
zvrWDjdShcF6c4XnM`x^p+b$xns`lNLO_Wm+42L6z1K*)Hy1dSL&7lk1L!3OJ(tDWl
z95NrHl;Qplex0Ga#f@hRMx!}fH%EN!=bz=b{;PdlDtna-Ce$f9kC7-9R1=&Rz7KDd
zLSUU?l)_6ludsFQ0@t3}<?CO1k=?Bv)0C9+BdTIbxt!BfCAzA4_~?*#9?W>-{*pSK
zfY4;FZoBV_Orwe+$M^pnrhbm$_+Lk*-zUpPOtyB&21A^a)JCGRA=~2=QaCPawk}>`
zoCuC?y-ichJrmk>^3qY60O4h5q*C$W;glx@>kuMJDY4E_6eZeN(nNSp)oxBxYoJET
z0mx1cSV&M-VlfCD3PmClgz~_1l4mHHcG<)h0T(DG$dZKNV901PW;hx$o{X7{M~p@}
zDv?Mhs7l|9@91R4{=pFs4`x)2FH>=TBA=d*baO10IEjFLdEq$U3cbtZDbh42%||3z
z-r-Q#c3ws(CVc5i<!me&jtkjm^i-VPA^Z#LNCceDS#<F3d^6Kvd}U{(;<JjCCB%?0
zfcB9W<+{uI<XALMToz+#noeS&Wu!*iD<M`|$C@55-6By+vfewP&$;+GjLOph6vjB_
z^JS}d?!jtEl*CFAy2H3r#wi|ERkBzt+Sg^G7>&nX#I7ILpYo(45606C`Uc8cOP&uH
zk0)J`tMGr5E(GsJGn9_S>;X-Fo<?X+-uiuBdPZ^Og(*T!5z<1o%`g5F&-2Y+`wrzC
zWTLSq!3iB6m-AsQz4$&-L8Qc3N9qh;e&Ggx=1=}SFTU_RjV>vdC(Mr@pqqlaEPM!2
z&Y7MRd~olWAHDa8+cV9HNU6;c4oRvK)F?wH3R%`nAN&RC;+FuJPycURz40$2<N=Mg
z2xm#N0YawOX3jzm*&Ghoog|cvX1Q!|YKV1)M95a3ULaaMaF4b0vZ>B?*(qx=b%hKe
z#A5DHGD8XN`C`pBX*NWu6sW-^w6$?gfCxWJ-zh~Y-;f|lyg~p{pi)Vm4;T&yOePaH
zHYRKgbJ9FxILJu@jUowNSekk+-DEsukmr<D#cVp|=;VYCZa?6Sx8CF6cpd}AFsC|C
zvBjaHVQPVEi!}{(U1919XMH9wNi))H=qnK-Jw^}=zMq@EoK6IfI`FfcH$Fnd`aD4_
z0PxZ*J{vCut)z-@KJ#adCKk4PkvoSH%D1_)F^3uPr?oX5Y=kwISVktodT~$CK=_F}
z7<aL{>q4sF{&iNU!+|_8(oYJZR~^Go8s@F@E{b9qyr@9xw8qyf37-M$e|Owjq?A-u
zMNxDjc9ByxnM}?u9_F;Miny3p>-6{QWn9J~BiLmxo%5Y$um-DAgi4Ujl*RM`*@tiN
zl^3=cKX->^Q}K8I%U@t?cg`RGr+>)vFJ9!2{;4nUH~z}+BV6MfAZUqFVN+P~?yR4H
zQem}0Im4?jJ<S(C^DKw=-{S*2WjsnKizUwa^!xE_&e8q}`+NI*aDUF*j|_JX=ZHKd
zOHzhF;~J<mW?5o_LRwuv*BgneIJPOPJCH~+H=tN9n2aZk#}h9us5L4{utiPdB<HR_
z&Gi0<;B=@T3)!f4v36tJPIA|U@O53cb(Owlll2u1rN6(;Y^$Y_L|$7v4@8HFc&T5D
zFvjcOCkdl0<HhG+Vtadwjg1N8$%w%q!{`P9Br52eD@Br~z9DX!kY_1*nlTs*P-+`}
z;T+THl=C|~T)J?9*Wdnt_dmQ#SvMh!6Rq1KW5zJhAEI!K*M~A-G|p+9bR<bmJ{Xe^
zMjrF)S5n};BAn+eM&l4%vha?M_w5{Q;3-ACD%W~iMw|E`d{<{|Fr2p2nX+DUs$WwW
zG?T*p1*};1x=AG|d7d*G4jE(_AH4b9k6iolcp@WJH%+IB)AQZ>*~GX*Xl+~2QAz1p
z)6%EU&jO=AlGxc>^x)-KJgL*G5E`SqL0OZ-kWrY-S<2Ffi17i(Y@S!^?1!<ZzdyoU
zMX{i+OI&O>g=iH&{rkLbPb8Mc<`qu(RBx6Dn!Pvq?AN|Py8RB*vf(fPg@2B3{O0FS
z%2E^;`S1UWzrrv4_g>+xAKl@*|KK*M^6kluu`5Q@Ik4e-gw>34$LC*qj-UUjFLURe
z*GRXo;PisU-1*walZsn+_IcyY54e3;FjpHaM2>b1Za74!l;vVcjz**jI6+ZWq<PL@
zG9tHsf#Uct`f*ME7G?Rnj5fw3iKMP;UtZJHG>w1GbyfR9=xoF!&#}V`TzU2t?!5b2
zD5Z*a>9jgVpL^5Jzq<BHtG#4Mlvcs`G`)-ZD58@JasBZMp$9SP<9t|ia8UAt*Iwi1
z%^N)P%uVt<C7q1e+T6f%T5FV)3<fz_mN6XW4DysTRisHyBtu$jQA&C)TUm4S>SgL$
zbL-w7(yemtWXv9lxGDM$zt;Fhp4Qah97?1lGDpeGdobE!g>54k7d@HiQ41GD_8it}
zto0c?V{M?`_>k5*ugex`2NH+2otD4vxD4arfQ`w7$#|2E%?-wrF=?K*b**`xlcyO<
zsn!VP`G7$_48-ar$l`QsY`I+WtN-7>{n$0%Kb|byvj!3Q5UsIVeUAwEf{hkD#=a4V
zjQGRK`&V(Nvy4`MVn#`pMmM23g9%2#*Uznf&Od<Va_LnRf-ZBCB#g)76^*F%@U#CK
zK+}{fVV-l2;c(dVqP&X@UD#-yXqmrCfRhSROPRcelZPDLeVT9l=Bqd-&|2{;fBlR6
zQ~$zma_Rbzf8*c%GkojU-sL~|_kM%rVhJu}nR?fxtib7tD;Kx;`LBP8uYKudP7WTj
zy)`1q63V7w@6ib#-g(HoxAwWSS5S!|ZhV<W4SX`AnUQ6R#u{7#Ius)-9L8yqbVxcF
z(j5F#?e{wKGq~nDRdt7pSDvD=(9{j{sz55o#>OT>3bH}w`@S8Wpavr@y!<ND<9%ia
z_dPks(N^XA0qi~NwPzC5FXGrHi3-g?e0}IJlr<ruXX3i&8LVK0fD@%il?w8?4kZO;
zRq*b+?@%lkeC9KsVUQ0<lZ36!4aTEw(!|ST=Xr*bUVX|K&GB+dQIyQ)GZym&RaG&Y
zEjd0q=49r1af3WzzAOT1GH$r0B~dz@r)D%YR@dN+$9SBNUK(pLO^KA=3$)gTZY_Ro
zwf7+eX`)b4c~Q#5%X$l?7$pgLmU&M!PZ{Jn!+b<O7&08>43d;|kdbC7X_}Fy8F{E*
zjpSyf)QU8?bwS1@l-snilJknXjV}DAO#rm<w58a;@{=4-nB8;T0|%c0fz{swpzc%4
z*7xB{6Otfh8Ua%8CVn&neU<bXf49_iL$RnZ+Mu<;gx+6L3PkA3=V3ALT6?DYjHdA_
z3BIs($YipCQho?zD3XqPSgXlFKN%N`IaO5>EpwKoOeW*DMzD7m?Rdp@UlKyCn!+04
ziLOb**6t&I{O7*QH-6)bNagq=fBM^`1IO0(g!9+N?0ryh`RemDddB>CNmUdS^$D1U
zq)AAn=8Io^i9hj2ewLe;cbOhNAXA!aS1z+yG<^5F@9_59A97T|T#nF_OEk#<V-2oZ
zvYgLIlLV|pIt@x-oFYk-=fPMTHiH@|c6)PSj3uRMN)GPd=i0N+v&d>pWoR1BV!5O!
zD`tx&ySqDVZBDQvV^)=1xbYm*g9jee6^f|I1?wA($8ua)2*#M8&{}oCuvFyP0P8HX
z*_5JKh6NPmO?vxi{9T6gS&K?k`!^jhpob6c(=-*I)t@D~ah>5HV~{CQl~R`_`;Q(m
zozAH1ATpOK-#utBWOw%hLMqClWHLEmaCppY`98Cw2P~HhS~>|WSQHw3C`2NWc}A8_
z7>$NZCS%6qF{9y#(Rjje=&2G}mXfAvD~X-w1C+{wgpX=H7w=7a<KR&m+l7Ly@sIy}
zXlU`CmeIcXsZS&Xj-&NremW#OVy9z)VkMl45mx&QRSV3=@9N9GeBANG%A8#$!h>%D
z!VcBPo16#CLyx*7QC@F3Nql#gXrQ9!<WF#B$k%!e!9WS&Md26oCAw*7bQ9VN8L}k9
z8bEs2RtSMMp2XKQx`#vMY;0_jrdd37d+qW2r0g3=A1an*fwhKM%jY}K1TecNJ<;7q
z;G{<^#5gd9*>uXS_orM>E$4ROpZ{O}WuAWikF$Gz$)EbOuXFk8c~WIKeyI5yf9-Wt
zDsXJFeQ}FK3677B*?V}G?Q?=Z`qMwg*T42E!$foc_S-CHGmO^UxxdHvUVoqadrO*R
zf*tN)vLQuPQPm~F)PN;T6}HhV7ju$9(oP^HKwEzgN=O=Av-8ydmV>+h7j$)*@y7oS
zA$~xT4p^222YY*r&ux<p$5c&$lnLj~ZD6eP&nJ;+Td_49GP!(>#f|5Ac<Vi+uz}>O
z5bkt18;y-~Uc3JA-~eL`Ns?lnCQ(U9cvz~sq$~?WJMwXE(t6O~+Ah}Gc)k4YI}JET
zqZ@>TH{X1N<D-2pTsX((<_4RSF?pUb9FEx7+75-^iB~?5F?sJx#j?$e&8;nNJOf|)
z%9r_54utRqBw}hwqS`qb3wXPG51d2Wf!1B3#<U)mZ<=e{#qXRSZy)*w&G9Ij*n=%v
z5xc%^ah)cq7u^dXAqK5AR+|tW2Ry5H-Cg(cgd~YlF&;0DQC0LwqO<Wyk0&xxN-0$D
zhP#d!0T;ow;gtuCRol@xQNDwWZ_eN+Oe-=LvxL2H?{vr);%zCtky<P%%gPU&u{dLi
z*-;maq;r<4EGUX)h_gL!Yjbmh@pu%B(|TCf8QbU$mVs3sCrRu)^SDs4dk`I)H&&$^
z>!pF;f8&J7ZpQA#9fli@U;1bM4x5`fP3>s{x!dGF{*Qi*2X{T$v@Qy4am4kj8+_^7
zQxwmf<E3Y|c=7q?I61!0t^2pB76pq%#odR;ymxP(lTuKsF>2>2OgdpYJw`OO2W)Yn
z-E~cpC<b{(ts7r<6M$Hy{LTrM#%R*yZLWUiU#6IAs^cG`j6+Gu?)i%x9vxEKhV$nx
zGuhbkZH#7fhC^SPQt1YnSd4A3HsjJWFLQ9Pk0_53mR8baJxS6ZOYpc<k|r42&@>H`
z$=DkS0d?K@?6Z*V96al2A%(TqoO3~q5*CX&8ygevnQ@NMXvEge4ugC^mgj(AFdX`l
zn=ogD6jW6iGGa1FBS%V%H(*+uHbZ_In?38k&$vb<*FG01195}(Mq;HWr3;3~*f77n
zCwFRyoF7Z2d}xbEf^b52t!EiiXYuu!wWcQOwsl4_WI@{ZchsGaKc2`)WdWg&7TyMd
zEE~Vn*><3X5P==-+1zRB)3&<tEvBraRUPcc=X7?hH+q2&<XzaZtb7H%j+*Blu=HtM
zf1aW!nJ;EuBT9Ldgz?zB(a}@tt~V0TV)oKo|GKW2&!<#X*$!}?=PS(gbpfx_iU}5#
zNCv~4jg2uywaILLjz@Rq-26<2a4;NYXl*fWNz)|!@-M&5_r7t=#^#X4OjDNkX-*E<
zeSVkEK7Rq}5-y$JW_Ea=>B$i%(}Mf=k9hmNI}|oyp%YFjOHISDEVz8>8qQSA52k=a
zJ0EQ*mHF#a3R5nT!gq`D9!BsktVAJciV~IFA{|MZg}@ri@?=gvT#%}SVmf2b2=dXG
z@z^V5mSstn<rrg0la!|P!0gt!3*3D172f*x?~o@E;9f6Wg+8aokGCO7Bt@YakA3-)
zSAzAPmhZ9~UHr3;xJX0@g=kauJ*+P%Hl#}WLi(`q9lq2?Dla+Q1RTs`OU8>r;zLR6
z97xr9iUAmln`3`Nj`tVuJ9@M}EcKg8yeVVW1tkf_r8`&Wp~dXFZ0BqA%B*!k3x8D_
zxb@;fc-&||zA@==)*2PWdf9%%3ut;6VpNxHd6S>;cw(NEM7KMT4F4wy4K`g!g@%;B
z3f(u1@#)Jt)iibO&)_`xW}RK3#jZ}$Zqi19JrRmii=yo6_^fTeBi?+JPN}PUrDxv8
z#zqh1?!2eYs}doc^VeNfWtfODbE8^G;&=`bb_o&60mRDn+g3rey&nl#KH$AK@AIWU
z2-{m*OlMP;bHna#hG`tX{MWzDVxF;)Bn(G_b350#e0_^&p1RJDzWqAaZk}hD)tnq1
z^XTB1Z-4(C?mnC&vN1<xP2&WWaR^~)$~on7N}6Xhc!MmJBvlz$LtPhSN#+fL^*zy?
zv7P~r#p#AD-Sj(UQ83IWs63&mYN}<)Vmf1IdzZ!--#%=RQwByq2nK2DpN~x07;j*m
zMpqT*FJ0yC&Uwm{10*$0B&Tu6-u#Omd|lU}zS6ZDP+e=Ps_sIY#~krF_=W{xGwCsB
zL@48uQgwfiI7&$C3oOD3>bk_5Zli0OX7$>o^bv$o0sHK-)GHf(`x-%spac9}Jl9H6
zr@IdkBAUoAi!^}_^9dwhDLRID`&l2NCZXIaUUyWQ#TXBpuxiL6*5q5Q2f$pomJ`|G
zV7~6qL}P0#E;J7=ichJ7v6No4UMUs$7`D4$C9yWvHi;G!NTL9j^&MtvT~aSAj0-yD
z)(~>Atp}u3LYigdg8}2wh{0gUaOm~QN0Tw5@tDDIz{PXt*xuS?I2w?oX#nyCwyvow
z?`dTLL{v$_*2V_I!61OKM4RLdNl%?2;$FTv=%Rf%N~u=w9B5l+Mb|beB?&28SG#T+
zn&5?1RmH*Kp)W>OhUR3+Fip{}X1t-ee*Gf5J2}7pUw?-?@17v7z*HrZt!=W&1hyo<
z|N0Nu)-^{DEVu67VRkU*gL`|NRDwdLEX$hOO796t&k(Mwf?{#PWakpQm!4vFxR3HR
zD83#{fK$Fu+n5rq41;6<5=JUzxm;q6CCyV%GT<MIMqAS1kh&<D$|<AC231)wo1L(=
zwaa2zG8hifre=_*<f%Y6HBL%!1~nM+;^)80_y7K{;Y0v+dpor^2<tnp|6OapaX9NR
z($`VqPVdkkWw2JxD&hte7hMR^q-e8*Di{Nye94Wkpz*RNRatUya6nNMffX+pj>hcl
zoFhpxtkD66=;1g?P}<ZU25{CZW(L+wi$y!(IzuIdmr4?S!`0VRw{6Jg9U(v0wRWM6
zyjOT28bE)7)=!h7>%iQ4VhN;#Dge=`M!4W)Sm)3!cIMS9V*Rei&N@#xUbp?UT3RV7
zNHX8P%7<NP*nRv!q^UPhiBe>FPM+m}@-5Go1&{XjP-)7>miB4Q@tCd6O|~Wz2E!p)
z?$e%0mf}ca80&8^reksFM$_m9Ypj<jZnUqtvzA1ql#Qk?N)!SSz{$zPhaXBxoLQ4!
z={)K7!5d>*94flV!@<BCxiBMaNFsC#E7p7YU8|}pj*pH)_hk=0mR03D?nuGr&VZeB
zBa+mzv$KtGGw$Es=Rf}se~2Qdu}h}Y6UwS)V^brNAx>^^baIaehaVylJX}<iHf7NW
zmbF)cY)ean=Yh4<#geHwWHO#$M<bR?4={UCLKviut&5x7epr#DL;qQA(a!m17m~J1
z5M@PEl;}z`7-kHH1B7rG)3Cj@$;oueC>yak9^(Mr7*1+Umin5%d^lh{%rHsH=7r0g
zyLz4dyKj38q;FIrZX{x{u0z<c=!|DS`^}-h2tJ-5yiPIcC=VS)#zi1n8{bpTI&@XD
zoG&<(N1hAiJmxZ)OjgLj(MZXFpBhtn3Fk0}<77)j$ES;0Xe#x|i>UKtaK2Ze=)5gs
z``%TnvDv%s-~qP{P+a@p{+f!8K|h=p!P|<>X+wY-&KEr?e?Has6oU;&LgBartg$V`
z+g(Qh_r*s)qyF(kMrtq`wv7#9wP3`65RfGa<I#xi?QJ$UH@(JqNcsC4?p)M3nfwFs
z6oN);nx>)A8eKIkizQ`IQdJdAqiM8hS;y8|wl~MXC8YG+74R<ia=C1~7fY$g2YK5}
zp`Yc6p_vdJ-y=?}*=*LTfYx>GNoh*a)HRXNe-_9~z_uE#na}6UW*!5Q(l>ROOeUDd
zQt2wl6Du}0GO(KEwBbMg3%|kbcMlnjeK=#SW4Tyh8;#>UqtPbSpx|h~z$}*3X~JM*
zlj3MbjSE`a>(^z`RY|QqfpBAE6Rj<ERfL^R^TG=^aAwZK{rl85@go0q<$+cusjE5^
zv<hYi$27BoK{`MxZ@8p#xT*jFo7<a|x*~NWpH`O&;ULYk06JTmrero-AkqQPe&$t<
z_U_?A=iT-YF@@o$n|1i!c<nBI8hrkwo>P49%KOh)IP2&4*lVXpCX@*C&(E=#$<oFW
zOT&EMyQrbm72Rs<!yl&|&S^gfh1VsML^7&h=sx@m>a&Sx1J7866|;3K;baK=V$t^M
zJ)Optobz>|645pw^*tm4#ULaoqM@-u(3nu$7uJ>Eq|XkypFl!t<#;krDjgz1)xDCC
zCyK$yFT9=IUA8y30+%E2l_fw`*Jy32>WZ?gsH)l<qq?E0N~*G=C<=<oH~(ym#x!0>
zsm)LbkV-Mg2W(7s*qm&zu`y<AYlD1{dr>&8SuRWFi-i~46M+JeCT%xje@(s2P6YTx
z=Y~f^I2d@9%vd_J8JdhLrO2|JJkMIUIC4{jP(I{r!AXmCy7k5m4-a|t@DU<WEKiPE
zEHqoYj=CuMmH+1t`K|x@9zu+%N`sODYl88vmN?r`jkhpX(_}+t%L-f9DCdEBX*9Ud
zd##_1aSmM-j5bFs=QETLY;JF{_izbVjIP-`*yA7h=`SFi=DTm+r*Z<41Sbq4@pW2F
zQKF=zUKA)NNb;O?Fd}-808Ld>=QW6gllckbjSWWmh&;_gV~&DBno`$>Tqu+hq~k4~
zefjge`Mqy=&K6<u*ht?9;ZCbl)IPd4MC@KWy%JVK$dCA2&uFz-DChg%MJ=>6^RPma
zB*fa(H1nj#ot>Tbyp6Ur+F%SAqp2H>>CZpsa31tq3Clb;DFXiv-<K(RF+M9ORiL)0
zP`nQaLTDCnQR_ei_=Pok1EJ{Ic`swdSVxb?M(<1DRY7}Rl=uEZSDLWFI{@PX?Z5@v
zTd3nD78eJ{K**IE;Ey|=j3?zRqrrgP-CZu6KgWghTWoLdkmr6eTWh^KKv^(fEGUYS
za#>PVHC0)82@B(;f35YY*rsVPv4jRinh4K<$z|(O4o5>aH^yvlZLqnqNtWepmQf`^
zuPOk4n>**Ya_y?8>SsBlk!OTwY1SG}8+DojmL?D@n`i_zo1mlKXdMWRv9F-!WIFSF
z6=TrhH4hZVUCp!}jl`$*turjAGmefA*xx^5|KSn8{dZ6J?B~z%tAFbU{N8WgN4O!{
z&aez|vaMTZI-62d3-ZMrB_(Mp8I3n+>XKqvV6|6*=zAY@_8nc9l#3HoHs<i~klpjU
z?C$Qe|L`8eO!8A-|0zz6AFw^i$t5h7l~<OKiJ!=grH(|_x<Mv6);Uy?gasl=(iCSc
zrm7iFa;mCEXC+!mlAvy&wMOQ-XEUpW#(0Xy&V?&v*$}I%uz<q?a~&?_vA>w)=x*?%
zys5s?=_a!9G#Ec(J?}a``-rW04?iwKp(LZ>By5VEw6v05JwXMTALkrNBE8;O;MJXV
zNPH<*8xtWwB;EtUV>=#bhFGJC6gZQB@P9X$_OmM+Z?9W$+(;v))n@MEn9?(xff3(^
zXns&SBjaMIX}p1oS;p{LZ-*jy-n*V)Ok?P2*3||+;&>t>_0_L@h27m<1m5kPEh_d8
zZ&55u?=6KcE>3ng4YF(I8QoHR+CIq!UT|OeMw~t+?&%7{(U8r}O?Gy6*xK4+G|U-f
zIa!_~18FSM3OZ7o$9<ghO+M?oVOdn1Oy|KE29GARF^}LPVb``R9bS8}HLH^>K4gJo
zos=d5-jt;|qe0GKkdoyBp;E6lSr!~kx4C?D#N~4tfB!fC0e|x^eV4{KWRj3#kTRpT
z4cd64s+3|h-e9?$<Lr_&QCQ=-T2^|EEn|f*GPW@g=)4Xe7C5Tq6sZj&8E|xPz+~dv
zWBp@)^yheV?;&sg;0HYO%z1Pzp`mTfEd-(aurEW&^Bn7hmo{;ZBu@hP>OlIasjh3X
z!HDHz;X_;pLi?(K##xlJRJx|pHAUr%h6kHlEDs*|Fe!#3(E#;0Ki#4is~!TU4OTxH
z60?JyA$RTh+vG>U6r;Bt7iPqWd}G|SHBkP2&XbgThqZAv`#G(g4gBQ@(09f&X0GF5
zx~uc*3{tfBZv#8OqaZ|(++y1s@#EF1W%S0)Ioh(Eh%KF_LHK_*7^|pn)n*l?2%AVy
z1rsi)eXQz7#eUmui}!$ud4T?t$9_*_q<-+Dx4fHc<HJtckV>fp5ww4B;H*VAruCX)
zacr6<t;B_=^yh&!?BjfubOg{W%P5M1<#I{?o+2+P?hX;O^bf?76n)mNdkFC%dB*!H
zF9gw5<j42LU-|(?+zD(*6MONt-d_bc+XPY{EanTGvou!VjGwsE`IMvM6CT~Y&Hml@
zNvvd&4LB(auM8=bkMAv^kQpCBrwL*-VzCI>Mv=2zHdM9GYB|6c&C*T=7ce*9%+*h_
zrY=Z@M5beAizP}Le)ZRXhoggIUb?Zt`Q1$xiv~Fy1~4?F>K(1}fW!mW7GEWwrhXnY
zbs3Td3S%m-n3?Cy=L;sA+eiVakYFrJThZvo%lHU^l8Vbu-{h@>homA}k;t|Qd&Wn=
zO=pn}&w7vR{Ve(hYW0l)=kfDE#B(~@pVy6Fzs$oM1A(!=Y)dzWs%g;Hg*n@CmipHl
zy)XeD8**`R_pZ&l7R2@*AYI%`+;k(n7S0>aEu}&b4IP31)m?v#AXctR2-22r6O2sc
zj5&udwD;#(KjIQw4<`vu#V0+UtaeW*k|dOp$j)^%R#P_>rV0h1#)aa?#21<7zJFbs
zrX;FQFjF48!GM#9klv1fUsNnmDykgDRDo`9IMzZz+#&kC#=55)!F?g5?-m<g7Xg%9
zpLAi3bh}kM@Li_lac+G4TS$U>(ARiH$UKqA>X?CE^6dGisFzdDJ?Hqw>+iDE4zFfo
zgYLF3rd}*`P+7pY5EqiNYG@kU`b@FLt-Xf!nfd>kFrgI=bcJmy#+zfxy5iCC+*2+r
zR8B$S@qZzcwuD55q(Pz*oHZnAG<*W1gJNXh7D<ufgrit47;kKNd@Wu2tSF8|Nl;$h
zau}tSrR37pYrOx1@8X&oCw#GNe~5hw{Yhr*>Cn%29Pbe+dnT#{ovjTUt8aJ_DKyTS
zo^g155qy9AVU=aAwOL6X#m}wspadA#KG)vl3!KFM91j~OQMIb~y|`8l(>aG|d;}3s
zXf2(p$;yRCoGy52EwR^s)anS*+jf2E_0zb{NJUSokN-8FmXXTRq$M%Z#@oKHn3T$Q
z&CNY{tW=5$T{@t%+9G5rI_kd9IK?izLYol3t}<v{0O@?zBQSu)8Xy=4CoWw9Dh#X>
z!t+h=m*i^;RWO47`uq_aSVm05r&5hEL06}X=hp#Q=fXk?;}b5%hX!fFayjSZ&TY!0
zDVeD;#R1iP%5=J5XOL0Om&h#feGMx^qPzq~(=_P1@eM+KaN+CK@HnBh4s~YRYS&KU
z7~%*(n1-g@r%W6o9g(CGDdGJ)OWwQnLqwjDhUXK#52H0w2(;C~!}8Z*tRYJVEe>G<
zX-`Pc{h3dvj5jvXbwQfu6s4DM$rFK09BL@S=A|&wapkF}xc&Zn$TXCd^stuBzdif-
zB=mzm2D9>lf3f%9up!!>mtpM3pFez7gx2V-5eeszD_+RyjZP|w?1}hFLEs(P)|2zp
zf<T;hz319`&JjcEN@JYx%_eB2jDn$by_QzaiwDHIK<A=CRUy=iSJ}<0ZpeE2Rz2gp
zet(ZM@E>(NS>HY3TPnqYNJZAhso{CHE}~F^B#|T{YB2ldCF8Vn78~MQV@=ov{GbUJ
zNOj?T{_DJElxzpXgN~jx8;dCME`YN*EJ_G3*d}@gQc2QewdIvp95t<L9eH2vBwBNS
z<IG+!fAL<oH<3p86tY6#tY-G$HpdU|;_8ZOcF6qX08>;<r*oFIp{bTk5=re0GR>%5
zg*7!oD)PY~bRu50g3%ZY5pXltO{8{V$7_#P%8AglQd7?kNjDWLA0nMaq|g`#w(_43
z5+7E3k3!&!q^%6>e<!dyq|k-wX;}Nkm<aEo>Z(FXi51#6Hx<IGK}tp4Siob2B#KMV
zeU`iT?t}v9P`n+k-`|>9OJzGE_HpKvJtM3NQ`tW+2SUcbgA*>$Wdb8SgniaUucX_c
z*1fcLQ+3{}H%5nSm&JErp~Vd2_4GGle<IoX?OyZAwci<lvvxypLAoc-&(_Pg5q7a%
zryCa;^P-y^tdQ+JYdyjoF)Zy*rf&3zye(^7s~Qqvk~TyYk(S^o6;=`)3s9ey#`*mZ
z-zQ1E7`s$Ho0B97gTauUoo%)!6UL(vX_jG~p{z<4%aZwg8PqI&ZIwW@6TwO0!><qv
z+u&+D=g88GEc0UXLcn0?!^SMjJZ~vgt)N$&xk7lIp*&5!a9Nt-tP7;insQNb5_Hk6
zHAvy_Ba$*@DBf&kC`5;St-Gu)2-+b;AjQ$WA8<x9pPg{B|A?k4NQI?b&RCutP%cYO
z77g=d1*XPIMUo6ymJOBGWN8l0V5}c26DIl!+!77HT%{a9SI@S3^no4!6qN7`d6{H1
zRY{U&J~U|>5+SkLcp|O!*-NP;Z6`OOy7}vRi3jV@+GnMt^c<_&ScYRkJ{b9imQvz`
z=M&|52>Z2Qys^#hg^L{Ay@gUSx#3o>e%+f|$D<y%OP|83oHt4xauXM8*ob<I-v)0#
zde!~5<B@I>zgNZv5CLPA){`S6th|1GtG}<)Ao{rHde{`7LyvP94vT5^(%S@LTkIU}
zeI-1MV`qE!9mB!SGm6(JqFbi|rz#k6T*QGTGIW*-OQhW>odsb(B_oB?Y>X%D?(TZR
z=GK<aFd5^sXT~r5>B%v*u30P!=JS%}av8FJA+`@G*0Qdt>pFn!UUI_YQ6{|RJMda-
zNfOER>(@Ac{u~<{n=l$7#SrJb!e*WgNV9|_5rI{1X_|_8x$xknF`-y^!LqCamB5Eh
zd7gRw>YisL*B)U15W<(u)OE#TvGicP^tjn%JY<~b2;HEUGn8#Gvtuk!2#Yff+E`3e
zVaj91;~|B~s0zbuUQiqF@s*V?RQ2Ij(6AP#uj5qM(8J6gckK?mfeCs+x-pFMv@PeE
zY>ZgUrqo4&z+rVm-IQcm3PgDlFIk+V3D$MZVwF<i8GECxLL&#M613K&){+FHE-X~K
zZj&Sefz~xQo_(H!dw0BXj8m_@pMGlrh&y-RIQ8%Sl&5lPPSb8KaY6gCkajqOClpAP
zO17lnrfJ&y23o$Hhbt7q+CWsc7;P{*N~{Dg_zZcUvuh9Csh|B@>|^gQX76HW7!QBQ
z_VpHbUHgo1ZZ%>Nf_~#G?<GZ*&G2^@#;ZNIk7Qy-6c=lGLvk!ts8GoMNscEnQa|?x
z{s5!V$YUhBp{^Ph^F6A%rdTeSP8S?cXB-|Kb8>RRe6hsTjnB3Pr(GqIG&JH=K|#<s
z>&5GWk#d0|poAb%iP!I$jM&}TVP|`Xot+&vH#f=h)FU*;cQ<b8lI5~)t+h1<?Pxp6
z8sl)X1MHn6s5<(}{49vu)5AeJOr&$s9o`*Ivz(!IZKfj6GlrQ$=?YUUk-DHN=G2QR
z!=a)q3r^-G(~}v}X-1;~3sN2&7c8q9hhNaG2QL(lu{G@>5n+63!vT1lx@o)(;gcIO
zgxkho$w&u7Uw`Q=gVBc9R<CP>N*N9Un_5beB=cFJ&?vJBT~VxcG)+S`7<doJIPc<P
zvAV`tO%V)&ah@0{r9^8(mU(>5H4WqKbL?Dvis_>Vt(O4(g?#p+Tz}8U;U1kw(s>Kf
zz*`H>*v{bE&fAaJQY(w=dU8bp>^Hi@FQa!Yh4p45gqLyguX^WaUGrH+b~WsbcMz|?
z4O-*Wa6o7@5<MSqEft`{euJUxzM=K#+6<=ixEdj2uVjkAa~89p{OR@IGauFDY3Og5
zhU|VySymtUZ2HF&8L3AP_CY{dloVyz(;D}rOka>&geHsy#u}6otr%7u3<n`Kt3e1~
zf?=Z@>%+B;jSVhbxWJ{07rAut5?h-S@;u0h7|UY0WHAq9J7Zf=yf)rAVMQ3!KA;Dw
z$`E<TB3d7^S!1Zm8i6EDQzG}J-7R{%jtD18vy?o`xOnLznNsAb!qg=P4<GmrzE#0u
zam>N~KJ~KT+ytbs9Pgj7xv@c$j41}^INCd*(o2t}g>r%pb&KEln8g;nm6&?*)Qi)e
z)cT^9Lij>q3|2R!*?`4-!S?Ph!_k<<Vu5i2lAKhfU>!-4c+Wvt5|M!M8<h#Wxz``^
z7!=yKUvgTbgRBVFk`IRn=X=R{rn+oD<D5f<dNXZuo_XmNzVqmzZw??@Xp*xRY-g1E
zfOxyXe3Vz|T!*{FXWc4P+CbE`ZALB5Qz^VO?^)h#|Lc5<M8A>MO%tA3c(StV&4=iy
z4rdvvh&}lT&a~lLyjd3vQk<Xu)y^Gkv8rIC+T_w%p=WQ-OCd;<rw(~TlaZtegFIz8
z9576CvNR_PHNCMjRIAJouR9vd09iD0tpw^Ljwdowul?}*aXlcUzzWgU1zBgX0%W2{
zNXW_2Mu&8+MeB-C739H0=ZI~lhod387tV3z@+GcZxy;227Z{C4ZK;danyM-(s|B$I
zbkzIS+ECR^TPWE+n07A5oz=C4!75Q?gB0Tob!|LYYHD2NA550z_%=ZnDV)zn1q^IF
znxHa8A`H4N(MGeJov<iMa$%ud(#-al9v`w;mefUwwiQhyNu=Q1g$v~8Z_uO@mWNYd
z78X+o=fZupBu2VL<W}yB80fbPtk2AD4V9oJ0(yHEa0lUcc|AWO-%vD-WxDr>-78l~
zvJ7oP%WYkeBpy)B1~~=|O^pqS3=t9|n84q`Zz5KJ%u=sXnP#C0K+0e+L^lm0(WoQ~
z8->F*4JPw)CaE)QojcFu!bRqL_W-amC9$TP>~zNWR5mFr!cTHUpgT4a#mP34rNj6c
ze0^HXew|G$U4JY-<de<@1lB(Rhoj}D;IrGl1)-mxlGwl`Y@|fsJ614+a|oYx^InmS
zO2om~IfBkkn~@fPi*iTO8xr4L+;1*f?i);uv)nhC7-YdPr6kHXobcmpyCfB~-c=Wt
z3%rlud~=L7tSpYX)}IpNTsJ0kR<U3O39^lIwRM!bLD$CfC~AvsIwq|OyX{~!W@BrU
z%U3RO`SN8hUAo9*<Qo~7#v7zY*UV<e%x80Evl)*bJz_RHVLTqPz3tTzhr<zRKJrX+
zt$jbW$W-opRYd#r+PnqUSTd;)S&p)vvtdO@$68IE415Qe+`Fp+0oHmCu{vQet$4Ke
zfKnT_C%b5?v9@7e&Z!RfSri3}qG2&#kVwbXogCe0vYjiOyYV?H6w~PpvLz(Kb1hrE
z@{C>7t=gtjM=);v5WwGiXRueoA?G}jiwC=XWSLSGa~|D#hn@46$kUM@#sp6w4B9{<
zi0V5$4-Vmrx(QsTEYEy)HnfG0<q=89s0yi&NVGBJqo8Uj1XWX!;S&a0JERp{d-_@4
zczECAy#k*x?F~^Mi&`;ME8}u@czAjMd*e+PMtRJs8&4^_u&!UBu)=dfpri}!q72vu
zt2IsIX&DwglN=okR``Bvi5(-n@$}%h3^jrfI9g76YJ@ZBm^};_mv6h3kSC5TNyzd6
zd7hJ{DWie!-jd}xgF#N3C4qh;gAjngndr%RTq*Xo?2L`WIo*y~H@2}>w+jzfqnI@w
z-^bw?z2k|D)b^ztfCzvyG`gU!8$a-&OeIMbveCFIfyM@&iLdj?hdEcTJjLaUmzYe(
z3<o*oqF_Fsvsf%BmIeC<hwSg|b9i_{v0PHs6?I*c=Zc#*u7gO~*bHSR6dud6-nyd_
z=!wz!q4N)0(8bwS`cAfK=KJKSMueRtC6R(ONl4H%MaiOyZmGbD3}dD&9y~xd70K2%
zf*M!NY35TFCnqe6Dbr<z&?T2P6f#Mf>x9wG&%@?L9_$}+ba24^hmSZsI-)F=Sfh#T
z{ZH#y!91c-3&z_yi%fFTEJf4>2lwwV-o1!Qb1#4vwY7DF2o?H5Dty#o$@1K{LzliB
zHpw!q6Rme^twAb5qczK-Buz790yv*_6-X2gYYZ|I7$>mCv3ubX*?8i+9*66Gg1Oxu
zpE%n`1dpTB((#-625UXb8$Z{6j{vK^@er&gHGn|7v3TgU^M0iep(G|Y(~OlA4k10C
zEX{KU5$hTF?024}WNAtg3DZfG{0%gj5K1aKJT;;OQg}?$Snb2I=xN7$veqX7T7W);
ziUE6zo^6C0{LO@wc7_+#AIndAJdu&wJD!uJ89vLAfEXcjNgfQth!o$xO0@%|(BU;n
zmZk_L&}qUH!E_lqQ&@$u8wj;Ue*O~IE(&f08Nn!>BCrewdDzW77(TZSC!P1Ar0?yQ
zpaLKqA8_#ILLO8IfePM)#334uYzmy!R8CMBNTfGbQYvijGde~$q}80}@Py-ohsZQ%
zYdqoj=plAFgQmpkf}(CHwZknIT)KFTy(ZyqVLAEMn;abe1D5k+bX|LPcA}6nA<gpt
zP7Tz@ol79X=Iy|RZZJ{?0AAs2!|dQ8<E<T3HV6v_oNcg;!zkg4%B4i>3Tt&6nqUpm
zI=_inhroo|G#7fXrR0MlbyG1tK4!A9K~q*Z1xYqQJI|T3rXkHUuHJlx_g{Z4bilTc
zh3EdeJUS416*O+Md=T8*RU;M0-Iq;)HjcWn)Q$1sVr;Uhd(cCW6H28bOA`je++#iA
zU!LbAi6TqWpl$ArwTcu72c75Af!%E)AgaPj<(r~f>z_+7erDYp3@{9YkX6c74u<&Z
zN(M5zs}ZOdE8go_^q6%6|8d6?8L2ma|2M(~TB5a&<KfzI>3z8nLGf6yy|u$&I3#9k
zq?F8NGp5s7=n@j2`)W3=4PAm_qG!AZ!F-vld`1Y*D({ozqOp=v`3}NT`f?BHb)SdB
zoP01OO@|~Z)SxwnWmSe=fs#y0E^dza{4-k|?mxf@L6)W1`2m=M3)`DCHN5fQ9w&<$
zk>uo?7kGDHb8Hk39_}-p9#Yi>1o;eKzbrgi61?I3=%DV*qt78~Kfb<@`}kJexuaVt
zLYa#a)D|y9tpw6)%K4P>mY3u(#)W>Gem6D7`0<hoj8_4TT(PEUNQCe9D^N&D(66dJ
z4jGg!>ZTzP0xdj(Q&$aADDo_)X=;R$+<5K<?!5m_O9+iBlpi@By#>$(!KcThBcHZQ
znil*E-hnj+XFTR31C-EuP1dywI~G+>r!&g3VzKZ&_R=gP&2lz3h71PVj7MVzgF#T~
zNKr`=@F5ixH$&E0o3@ip%qj|e9c7et@vYK)v9E1;f^qH>nji>>q)Z|4rJ=UfFPC)V
z<w0)QHlv8+>9B#|ymtTenCdH2*Uypo+9{e0u<l`f)bT_{DoF=D@5GCLt$vwceW8h-
z(eT}8RZG@%)-lMlQ$D<Z?w!lI`nUdf^cp{ax8S{<!Oyi0wrpP{l&aux8yjT1yEIMd
z)p{H(7fTNJA5qmM%E4DZbAw;{@t?vK1<Q$|X;L;Pj!bB(8Xi5k&um_DQdlaHqP63}
zV#d+3p)vwpd(l3BY5rlUAjj&g3&rD)=Vd+aI5Q=`x<*&=nYa|4C*%3f%3kzN7#En}
zF6aU|jH#%KCE0M(*KZ+l*0)GU`bL&%st}|YZ9G}gIp0!D*S_0s0H&o>L1WwDjHU`g
zf^}V^l=2u=n)&Z=z89U=hT&+$)tk@r?(46Se&P}8jK{~nb2Zr_B)+~>1dJ{5#)AMp
zdk@BsVOdm6r&IR!_PBTNK6mfk$2xfKxfgi)`c<x8z3MU7$v6;1SHr&Ou|%xLT8EG^
zW7}y(Iq7-}U9=m$bFNiP=u;s2&)x;lJsP|2`@-GF#6j;<eHj)1JrMx^s3Gm|kWs(=
z^cs%W_o;P@iQZF8fv&{c{mWee$^6fY#iC{E`biYu>p=8`&WHmc=<ob};P>&a;_3L;
zD@L!TqFrE>I|s`}NmW0>3P+=TsNuARy5{<&3IFKNf1NMC^bENzSvtYZD_dC8fGs%y
z$v`&0Ib87Y;mo%zuLVb?<D~Glhh<gbP&AE3_XPSs>Nw>9ea!pM9EZob4sP`=BEC;V
zC?SLl&I7PcNQBc=%Q-4Z5i&(vgA~rowuF{VjS;@4&_E)kFUYk7FVw>T)|ucXh@d&-
zP<Ww1qzBdO+EP~yzD`V$gtql%t-0~+^W1*#ZD<Ny4~+fDvFv-8z42OEpfM~F1dltm
zG4o98x^6f)I>8BrL2x{sg_dfX`Fz2_!4U`hM=Z*P_cVcX=Pw|HA`NZFT^vvA=m@zn
z6zZ_87OOF<bG`rQM5tu2VSRMshFwV7B@&|ALic@-C4D+t4|Uz!@%WJrj)rQDVgLGo
zGY!(ZA&U3<NjPl%;|YxvMB({(VXJB2(?x~|8Ol$5QLhM~ZA+`?Yy%Y8xT|jB<8LhP
zMyCuFUA1)-7;{cz@rzR%jkcj?$vMz9+oPPH`{^(8V_$fMk$~lcdo0b2lj0a%Ls1$Y
zKG<WqTu{m}d&`o8MMK#*5DKdspEWZYv~6ADkGl4#<j}kDf4F1qb7vX4xCaFG3WVNz
zzO#)%;sxA%nVF+r&PWGi-!l=<8EWdLZP{Ii&d_z;cmO!GPtr!WA%`&ysYv}I45{!$
zc}%X+ntU*zscY)WR|DkEw+_=<Gn#C0{ka!-_lMsmmD$JQ8ExeJQDYc9PJv%^Dp9R6
zM^vjc)-jtGoXl?X-UqkBxm{=v<wMydNf1d&61*!Hbflc~n3Qp@ot3g(oLxepn-={I
zI&9v(TZo&ihc)EdlNIfH=>YEj`PRGAoE>_tzusTlp3kPve1A;F_y{uIzgr={u3j6T
zqEt%=6GTim_BOSC7`6(k#f$T=_z)wyp{wD_<0fC<@_rQCc>Oc}Z_)U9_geT!#qXH0
znm-hMD%<eWU->M5<fp$(o+a3N!F+mzsh1d6uqYc2kLKLHcR+0*O;aAuN*>G_w2gxJ
zo`>a(MVsKPh`7V8UbAEMfv?DNgg~Sl#Wr&4?lHXo<94Mp9V-`^L>)LBIYSN?)GAV6
z#?uruO~YV30qL8(8m(!BK*~CRzmm~p<a^(_PAlFCK@v31+xAF7ciZcP2~Ubts~em#
z23=PqSw>Mcyz=T7x%1wen5yZo<dvnmG7e|l^eN~7(S=DN4WA@QXXTs&;o5dl(aV&9
zT_2M}HW(f$2#T{6br66w+ILBk!ZY;Su|IWve)jl~t=n+Vn)lY8zQJ_^v;MdKM)^_r
zP~VWOXD^|{1N)zmQu?BG;fUI1k$Te4j`lzI>9tu0e7`*CEOq||ZTN3{YhjUg^D3%h
zcK>8pAoR}HiE+jAtYT5!`TO2e>%O<=@dkCpzY1exLQ%4_4(T+XfAKnh><@mO8`mx|
zJ9x<K;R80eM=Z-@mWw$})v$kf#B6DpSC)npD~G)I=n=;HoliOEwWF+8a*$GB6^ORg
zQV3yt#wXs9j3MdzfV-Z%k6&0eglGcS=M|oLocU+rTMBjIum$JY|5o}oWDv`f{PVE7
zqA3J&FhnYab(*GeAca@nG!|=o8I!h6NWXXKZHL#g%JQHQpL;yY;Z;oQsv^l!|4c%&
zO=BcdTdZoxvJIYj?q%Nm!S_(YVSR-i${FmLGBy3%eLRahn(7cVVd=UUi?;Hu9sG7K
ziI9D6)h7a|-)Q5t&2@t@jjun|22?0I?=A9`1WzxyvGRqzIo@JYbmxfcW!+1P9+Y!E
zaJyH_;Qt<ZS!ZKf(U5qU($R@3#&$Ik>g+z_ZA5I&A_Ez<i@wClb$_zs$-}*NkmJpE
zzV%sdornV+V&nL^vyD#oCD8-w&iV0|);9#<6<Sv~s1Z76F;1YQMcIaDu3q40fAYuq
z%yZYs%!0jJZ*#c+fU?v)d{CpSDP>W!C=9ig+&$FXIj*Ukp)pe$>p&%8XN}gl!wULE
zpz+V5H;#TR-RYrN$p~~~-}+ZQ<MoN~<hAp7E@ynvtCb2uMlRqGak3|gSEA6?VC&jv
ztU`Dvg<w9NlIH`eru1b+(OU{(qqAKBhY94v<$O*)9ESJ%|CL3JlnJB6D=}16iI5Ic
z8|Fp9v#-3$`|rGosg|A=V({2mZ|C=mvFD+F#E`82&Gk2g;H9mnGfsK4Jp&jGf>%n=
z0iQFcu$hm-Sx@0nr(lfJMai%VQm-1Ro)?2|jie7*SI2Jse)=QTMm|FHJpNdY<GEG+
z@AVQh9bLoQ#OSnqk_p=#Pi~~b;lD_&7>^i7_g?A!JMFeU?!2GG(w-GNugB@F3yr5O
zI8$-?{0>*IT%cSo`0AIxz_U+ZAr%&i6J{q5XqpOJ&d}u&0W6D(gN0y!<|qx^oeTCV
zkTfU()(KyTD!gYVoa`pJaN(a_F`!}ZTM2nT$>G+1_~f7bsFS+(@kDaIb1vF+zv!?&
zMV|Oxh^g=$brK=5rtz6S=g6`gUDpU9d}V?lw7Zfm_TxekvJeWV4Yl=MgtBZ9Fb^L}
zrI`wBY>N<zMNx9`;(4CBev?~ozm6~#DU;x_t>4lo7@4&XMJVIU4TIhs6RT3Wt6;T%
z-8$DcYKR`nSzvSjI16|__1(P-Uriu6lv-Z{0en~q@m6JD;s)L~kn15~>u~gcrqjjh
z&fBLWtiv;+GjMBT^+}E=$E4OyzPgi~w!rIxWc`oFLHno+e*E7$!*@C?jDHs(1jf|3
zrskDrukla*$v?)0oei|s3<f#nV#>pN_gNn8kx>AaqOPz`FstGH1I>Gn<}~EYmYT)}
zDv{G(#Y;#MB}21HNlYyG9m67}!`iq@bsp%M_u1k^AAR(F#g;h!*u!zwSFQ&a&RLL&
zuc?a91uqa7u_Yk{S(>uBxyjMd5zYyW3(~T}Qa4o~)n;g|eOKwA$&FHB*Y$W+mMY59
zV^S7qnvyIZ&@`HvuGrX|Fd2`%ftL>5G)zyXJpIDUeE9BLBtp;_eM)$+9m}&uVjYuO
z^IlGmbC=2L8>u*W>o^bijFxViwlPQ>GW8bc<M1Ge*~Q3i?`r;pSdCE5!k7BqkPzMH
zA|~56O6!{J{V`b|KNAuiaZ`29qb4nA5$3LJ8!<c{^ZF=b{J7)EFsc3^_2P}r`-#6p
zf9b5av#rb5efAT6dk=F8gC9o9g?h5+_6n7BvA@+M7(V~Gm-y3v;tz9aXUL)`u}#Tr
zzK>bX*t`2agVC6DFlMn_GOZjBjtuWVoOAbRfyxGCd4@IyrJOIS7Ad|c)nc`6^>kS2
zEwdW$8xwb>(|6~HJI&)C;JU{5C)fOr&q}lw&h|Gr@Z}E*-&0O{-6*1?=eq1;V>Crs
zGMY>%mPH_P5*lszli6U1tkF1QeDV;lEhPi7P-{byB$R<C<>!_qNqo<^x~iEw5j4;(
zN+#4*&AD?IId}0gClBwDNEt$k&XB~wMy$=h_V4Z3eiHn^|LY4NI(mkoUHtUsrcV%f
zp3vFZw7#KP53|k&n)~O{r|t=&PXD^$wQKhq4J9p{(l;o5d~C(Y5Y}r@`SeX>PT>%}
zbsw+6uAH|o6!<Cfq-Y1g(S-)5AN=Fv+s;?$(xxu7pFV46>l?9C;9}S<I!&LBN1~*p
zsS3_-Z}M|r`y9Xc^FPJb=9qGM!oV$fuz#OoF{7?Zq%ahVW9C!CLZ|%bgMzy!H7AP-
zWJ)R(by*^0NEr)=B@XeSh=4!!kF(?bHIFn-PUqUz-H#i)zRw`|qy`FB)v*K}0PKy5
za30k5;&mEpd}!x{*PH5#mpSVwmP_B(%4n?i{oq{BKpAHciT@rM!VqUHby@lTj8bA%
zLXziXf%$KZrK*fqehs{$sw$Dvk){cx2Ao1`$CYQE<<b4y43xlx-g)?Fn}|uRKZjFZ
zl%G1Q?~S1hK}AytAFfC03~|nfQFTFDIAn{vjw=Sw0*}MCwg%4dGS&@QJU1)rrMr%J
zU+e3mpVhWDo`c?bIAt)Rag2uJ^!3Jh?%FfiIL<B(NW`aB!@|`Nv0ERX8U?|+uI9)a
z&5tH!JuW=#|NX4jd+&~+>=~cwc7*u5aUf)ZZAx}V89)2eKgCb{_*b~HJEUG5@$mRQ
zR+kjZCG+VaO&gv#7M0=lQO*004!QI22#AD6K~Yy2-4J;qtJfSR@|n1p3xvJLD=5-4
zO6z0Csj*#q&lw}OzQ?ui6a8tndcJ62rNEOkn<j*RK?|(C2awX27S%yj)oO!u0+9v-
zXu%4D)?TGEHjRzDxNoAG&{P#t3d*`>I2<w<46xF72bL=F;j=T?Mk5jlR+EU7qF8eE
z>6^Uu+IKO_Wx$c*y}0g%&zv)7jn+qvcN`|M_Fh_jDKAwdg4gL<#m**B5VSTtjwcsi
z>;B$(JN{hJ)9E~t&g(pz6xlZf@%?83^p6X>PKT7fhIQ2=bVLGui~!<!`Z(PBz4pKJ
zQ!`Ti8+q(w7yW(vV<t$?;Pl<)xNF4M+lH5GtIt}WhIZDGC`D5jY>y;g{mC!$vp@A!
zu3SDxDkaVAfWy59R9a(og)J8d(@@u%WdnEi3V!tAA@`1!7!rg~U8fQk$^^~?{Nhd*
zC)QcvTKBnb0(5S6|DCI!bDSRK$Al>Bmvn|Ha&1QD)Stt5`~P~4bA@rf$*OK@Uz;89
zzNqEhPD1B!R%4vkv{v9VgF!&hmqK+0L<oV_jTho~Fi10W-OxyhF(ood7>>tS=d)s3
zIE)G0A?qc4RhseqOP}TSZ+$a#3hti6#~r5@cK=6rI`r|415`JTE@rK|Vqe$J&n}dS
zcvP45dc614{e!-wa95qNN)1@w@E&Jq*57lwH{fZny)%&gWKVxXjzit6%{v2jSO?zM
z4Q$girytH^jwk0yMf}NM$w$6_ecGI5`MWbe-X(r4ZY6zYP3VTvAm^)J`!av%XTHYn
z)`VfEn9ol*oE~GAhnTv+T4<VzqPEPIn%j>S{OI-(xA*2)kpyJ`jg+Zhd?pwwK`T_Y
zg0H#0H~0AL%~_M?)L5P6QJwYnj|=h6{EE|cfh!{}e9>=h{Gv-$Xtv=(fo;&_PSZ4`
z)ot6_%md#Jg!Ps70d$ul;zhI`6m-~$1QyGJJj=<41M<;`swliPjC05&r7@O9J4CHX
zjX{b8rzFZ`+<5K<KK#*-D3?>eumvkS`o{sY>(BAbQG1LrScf>)J^iN9t)cR=*k<MX
z*1vbv0J*huuFjo!p7s0us5y2T<2vmztzXk>!XO$Oe?4t8T4<XuaB5wM6=~(ZF^=Bm
zCq14VIEPLnS*##Tw4TL^Lm)bL*+mc_4pir=ui{N1#)}1*I8Wd+4xzeo4ck<l-<j}}
zU->FO@wG3oyFDQlnqqpu?7<XW8){RKD=>}Ycs^%QS3Ej~ckevn-FpWtv_Pl?VZE*q
z2()bw$limAgWZk<@xgV!_inj25&GF00uJZI8pXo7sCCv2K^*7)_!B2Z_ks^odJ{i{
z9_`Z*{i_oz|B-kJVdvcw>SckH2_hSjCIhf0;BXypDp7v7SE2O9iqO;*f?BIYY29H^
zS>}Of7vw>l#aWNx4Tl4qOi_aoMy4crY&h!6Q(WdkmdsNi7R4NkCCgJLlQA#8`YPY~
z=HK^y>Us+yKBzd=&Vn7fVF(Kchjc{LuIG`umCyHq?KWJC^Eo~@8|&cw8KT0BH62f>
zcbV;6QEHW{(A`V7rexSq&N?1vpyJ>9Mq>RQ&Vn1d_jO*C5VQzv97JyyI(3u2v0C>c
zh48vqd7htn4`&_!aN(Z+HoU~{N^rO4U5T(rT$>eHjqc+h_Q{VS-Rd093ZL)C>+=Ye
z_#N5#?j}F`b3eg9`iFm>OS{|LxpfQCSSI6yqNzxf!%2&&C6A6zsOp0KS;ISbj(GFK
z`z$L9DkqiTj9%&G7P(G-y@!X0=w0us5sI%r?t%BS8_vb}z8wG`F2tUzu0NiuM}J~`
z)Ule|dfc0O&q+$0lenfvBo=EjoFsI9!SkS`A}TcsVQ?;L*3&K;Atb}eCT=;SEJ|X5
ztHE~(@!?vK9x(=EG=tEzK?#8`ee_shV?wB3H)vx>#sh>jl#2yQ35IFLwVTiK`VYTH
zy<7&u=;OWBbwkzuwB|&$LawZaFprA@;>I1vG~z8Lw7$~X&}dDjefJ;NRv(1cX|lWK
z{sQkd9smDP!652K#d|vI-ukcg!>;aepFL;Sy~eYQ%;V<ySs`ry^*)AXLU*Z8dOR6$
z-s614L&i2at;^T@KyVLZco;V)f3e}0vle6Ut_Urbg!C$pUfNgi>Ps*3i@)#(c<K44
zNu}fP!EK7!A=CYmqr_l!iFKB{5BEvZF-0wT@79zbynCPhS&dYwPr*9}PPg7tyq=E=
z%i{gFot#&$`|%;`+B>@)HJs1&P0Pnvn8%0B@$alGzSE=Zy2&pDLZxVJaN0K}6~a+f
zOOhlHO*9n7IB;Ij&=`x*!NYO}i6T`wI7O0V)Ob2V+-#zO_FDHE@QoT|qHuGInV$H&
zmy&EaWPA4<PWs|wDFnK%@qBo%{8?ANHQD){U9R1Dj(30Xy`VtpJq>GJ@G@f*xVl*T
z1|WJ(opI{GVb`0qXBmcd0}zM2F^0zYB69=?>3p55i{8lD^Z!hMz2~XKFIYEZkMn}o
z&)2{Iv#uXl3FFuAv+rH?$0EM}tWa_Ne&Tu4^iyhS)&G+AkvfRoSp{oWJwELC?%?)!
z*Zr8)D1`4V@4U>f)eXZ$@Z7UE_?e&lalY`G7dXE&rkWQV9zJ3*KjiS?UFxPrI2a8E
z6xzXHi(7j~y!OKnxbt91Bfv?AG=2xO0&LXnF^^yL>qp!z*TKxQ#%ewEvq9=u%Fgfi
z;Ray|hPfDi_$0W=8Lz^+i_pJ*|G}_|Lrs^{IOp}Zl~+dd_ZTvKtszw@f>nWiXFb;B
zHz^0E@w!tIv;kv0p4D%%YD@zzr_XVdw!u|`#qkl@V8X$}Kf%%dkFj(4zhe91cTkDK
z=!Q%P%Cg|-_=L+>o@$L$KlJ@1v{)C$KmM|^AnA-;fBYX8Qgs`Q@5mhzb#2B}$+-Ca
z0*;-E`ip=4d(I-N;?}+T$9p~TJbhz!mdA9Kv0FDl@rdKxzvg&Eu>CB|t&dXyXKwP3
zJN}`|V*Hm!UXSnLw$}gJAC5LNBv@q<^>L^ee~Vz!qyym$nQ+{E<~l$1mCy0xKlUXO
zCBfCqj~;RK=n;$QF-0+_S}xJna_Q1#PEO|Ros@k4-M4w?_5o$%5Q(ozQ%JP(+UpX5
zv=U+6W9{Yoan{6%5A3WRE#Bwp2@-a*XwPZ=I6T(ZHK^>5)w;WVw(*J~>PML>Pn~c<
zJIaJwJ0~O}$w^QU#Lgn{>urETDg`EVG!_aW6O_t4KgeoKUHP(QYkWfnC2`KI81|;N
z6jCx8Zz7Y7d@%GZ^(?~)!O_EC<ks8&EI7qtdIKr`Pn^H}Hcm*85-A*@$p>C8CgPp#
zJX#x@GZ$bFyXt(Q*2l!*PDKyvm`-P$!smo1b873uSz}Ew#H?J`I{4iW`TBDP>Ak3k
z>qU5K{W_qFbo$}kSw^W}zS6&*mGAA(Cg)b>egE_QYv`8(LFnWhPGwy;CqFJySwEiG
zNOce1r^x-6hOh61cfC)vgWCND1nPj04jH;)_~eq`en-l2^XYSZ^^0HR^RIq}3+K1U
zvxH)P!u({9<?M*2@)E0yqNZ^QjpB`44|wC92fTCpkkSZ{3ZY!@0gDiV#1yg%H}J^@
zZv6qTj>*T3S^U`-WQH^1FevL~b~_L1ac7IBr!#A3e&<J7C<lljAuA`yB*VHHLRtg{
zOoM4^Uk?^eB!$8WkFzA9k$}Krg&>i_w=5F^(P)CClJ|m}+81D}1SNe2)jET;7G2kr
zVoo;j94wLMAbm)AeDplN*ArMhVY#@9wQrCPhUCK$qddhm6@$T$;pQgA$%&_pNEuY&
zJ^?P*dnV4gwqs9JNsdP7ECVO{-_Y~igu}D!8;!6&L)>UhV=YQ}Rom!I_YK73Jd0Ca
zQxBi=U)}w^@8$FjVBgTL;-g)-_fhy3tn#nU^4QkLQ?#D?nwNHl2ljEt6V~qafxq>|
z+1rKM*F|@+@f8tv?4EVj4v%4&hCvFSedT$6;!B_5YhV5ho5LY_ma<qJv7AnspPX=X
za7eK%!5L<yW>&$&!;1HAKj6;8W9FrXM1iy+Thld7cl`lFe1tK0{L#Pvk1|&4;a>dj
zx(j~R`D4hj_Lq;nUa|JenQ!mCE>vtw<~Isw9Y*^eZ?Rc}3n~&0flM=S7GrB7@S~(Z
zhi8ua&&Kmh0j8<3Mq^~^o5V_AuV=MJ>lszKM5zRsq-28u`C!bYtACAR{ybH2oy~LK
z<@{6Mq;3r9z%#lV<8!f%F${)7mdD3Fgpc37wlREMXudjl5tFjmpe)oqlUNoKH|fq8
z#<}hsSSzqr1SljlZx7$w=L@ZSH0#&59?Ct=K=eJGv#_SKeqNuq>mJov*wgxV^^INp
zT<b}N#918M=n)DbK0Pw$S<(*iIQxdqK>34)wQz!_sTq%QoUmkx;Q8w}_}Q<2nd?uT
zLz;r^T%#BJJiI@ptV)XIg2nQL>1=^hIgOD#oYuVd_8s24v&XX1D3v0T6z6Pc;j@ZQ
z#OnT!i@#U!icdT~A6-1!$EEu3aRCqMJ4TNS<@`i?qRTpcoz4f|-ih=EN(zONDN?2(
zg!A~4b1BvGU&QLKl8^s=WcmhL*BEPj#~?f&Wx|GJbVE!>gEwHt_*yn4gzu>*6QmM2
zXF`3S1Epw6m`^QMB%C`}aP7H2hm@aT_wrqedI>HKb%CA`YU&C>MxN(>(-Sh_VxOEZ
zwLZ41o5Shn>3hD=8Y9`u^v3rZ<GX76e-3N@f9?HQtZhkp=ZF0wVy(6JKGU6Zs?4m)
z8dw8aB#XtVNs(-JlT?%4lB|~KcDDuFfB`?)Pl6u|3ovB87=E|?AV7d03_n<CNp1-;
zWUJj!tDD_yHCZI9SjA*jWmQ(rH}B27!#QUUYpsaz!xvvfthM*Oc`uEG_>e1c@|?Se
zwPN_<`~KfNRT&SdS%-|~bB3gUKN58vP}Szl=}{`+<IvNgvvP>G>%Xnh^3WO-J7wkv
z?QM^^p1KM`{rCQ%{OIq21wV3nwB++&`V7DNOFzqM)zTVx`T1vX^sLYCv0b0@-kTeu
zhs}0JzZ>ZL9T)2h`e1mt8Tixhy~}rAdz15B1S>t}V@S3V+hAe}D9om>Vi0}w_^Q{d
zc|M9SJk24P+onDWHXk0*<c`dM>Jtux)cHjuJjT(q3w*zot|LNISS<baf61`^Ujv5V
zzr^a~e@5Sb10Xk)3{f)Rt+T|j&xZemM{rF;zv=0BJ6zLZU5ky<CB-%lqNt7Caz$sM
z9|M?xiLY^Vd<ViQ(T_M|h`}%nBcb20yL^Z_KH9g}9XPCyUDbnVC8>O-d?rmx`>Iwd
z-^|t&g5*I)Ps9_AM4tQPoY4cCjs3H+)aPnmh;BaDIm0=`l<Ip=)@k<ndTjOIk31i7
zKjy051EHf5&Ldwsc&o==e<7RHWTE7b^!)TYL~^k=E7HvxC)wfSj$2nx`Hf%w1%B?Q
zzQpslu8L7`90}u=-R6REcY!m3-DbzOAL#p@-EPM)cs{)Jy!FAF?|lDV-n@Ix5REhi
zjXveUifhxyNFZcGJRl2~-P7^$jKZW-mQI_PX7-Ka`wEp(b-g0{_JFCKAV21b>?lG5
zCjb5D50<G^|CGN?)i}~Q+Qovtm(UDpz1<1}8vYeDrYHCp*<RknHJ)y{WDF)XRtPzD
zW*GW1YqdX_hOS#O_$?3#eqb1Tnzmulbp#}Az{Egf8tEL}wuI<;c<)2D!$7w@!ux<t
zu}7=cF@%xL<po2kVHHVpswS$hKWd?-);*)c7Rz6?8Tr7+v`6}8`c>DJlCr^4@Gb<8
zc7Tr7=FiHIri|v0@%nMhwoVd$)aM?h@jU99+lvk{!>PGsKfl&mo76g+lCui|K2BwQ
z)b$sxf{>nBp#oj_L=V=eeU=zU2$5CS@`W$F%-4VX3w-s<uW<de<MP3Me7~XZx7ZkP
z7NQSq#u2|A*zI;~cU$gX47~o%J^uX7_j&ijHCvzhiV+dRK}fyF5_7DXM7@e;I*&(5
z6m7~jF;gfvNiW2xZ&d3LYG|w@*q3;7DXOUBapvB`_N&Jeo+(2tFBq|vkO(P8uy)Du
z$ur`C!=$v9k=614nTrRXgLn(q{6A^Ce@WA>Fvbz_Vyg-Pmzdiw)gH?B7vqRP(>4UV
zl18sF(h?cRfo<?KO^0h*3Gg!w&N{lqip%o{v|UHjwbB)N3=*ah16^x?k#V$)V_{PB
zD^}}MuU=j=W3neNCfkz+Ox7;-N2`xX$i(TrV-j#tIbevBOoO?~S0k85nIiP6nbY){
zPj>j{L8FIE3x{k|^_+Fy#wZI3dChbj!l0+>o&NnXh9Jy52S|@HX?X(#siNp(#6*>{
z1IJ6t7eDhEe(Eb<<i+Q&bMwjx&i4#kX?4|Kp40C(h?Gsyy4JASZaLq0UVraHzWv&5
z-uvL3osU@8q{CS%4Hky+c=%jjkF!g~P|A8u)Wyf><R<F}hMI|Ggc+wsQoTp6F7cS_
zqilgsbtwW)ouy+848xYDb#x9w*b)3d7(C<nU(s~$(KKJ6TmEAr10hJ2g6O3`9?2l;
ztRwpI0GpLsRjb7rK3aSn2}4ikrP6|P4NcP#fW^r*7R!#^_JV$MLEGJsB6(fQ`R<&?
zNhxC!JpFnrfv8Ury*yvfL#y{x$b^0L8>3Vo9lcNWmxKl$XEV|=DaJ^G@{@jcZ$6K)
z73~%9W1sHl*W*;Z!eFQLJZF}3Gsn}|qu{BT)YOxu(Y@9-5)Xn+opA#$My{V7^Z8dk
z!&koe8J@d-ohwHjT^s1kj*IgLY%U_ZVc_E41>4P<Ku^;^*E;&a@S{8DeCvnr^Zj@4
z<Aam|mL35kmOwO4gCKYWO&#3mCpy@sW_Zc=6O8hobnP?5`a2psqfz;XY*Oiq)%V%^
z`_T2YhT(&<&^2HkHl&_;!-%m`pF4~rM2US_EMBE)ULyj#o!0CwDzt0G*zTN7bJ^bv
z-|v8yc6rKbc?#Z(d)_AxtTS{?%Z(e?d3gT=&hOpl>Xj=jmdDuW=-LHMyQA$|e3bBs
z;5}g3^{QefWb(wp2m6;sYR`p9-ZyJ<PGyWSmui`$nh^6CV+c~$T?GS*ezln1J_@Q0
z<R4`S*MHATv>f`}qcn#{sh5ZBW%IvdlwjRCikdT|I*w42ub=dK>ICC@Qr23Aeq?oY
z%3`tP=;Vl(U-}fjU-R7QiWjb*^3wCSSS&2(_aD$;Xb7w?E+IzNn;rKaK47!iCZl)9
zdfl_xta<(RJ^u7Zce%GqjU;0)OwcxhmrSy!D5O?<!c3ie)cBudj(eZY(pL{cq)za9
zl)dUvsO#}+#G}Wie?JZ-KJJP!<~B|t1e`J0q`Dd7=oV+h1q3`X1hGh^xTZMvG)_7b
z4<_PTse-yXTCu&nk1_j9LIOYn#;+MSBg^Frc-!C_$HH|OYZ-&|?t5_e9X96==-MM(
zdxCQdh9R=P+|ezLzy)^whK|S>Mg1Fx;iC<DNzJVm;UJ8f*jwWvub0UoH0cX_Ya64~
zcHi|qyS|q=sc}F<8S-_Y)z6$U9U8~uw1S#0=CA8D&)qv@3!C3}KAt&aod=fI^Ql$Z
zkFud1x}MIvR}I`mL-dhPz5Fsi_g8+6lWSLLyOw88TCTb+3r2REHLtvQlX1J|*a^Y!
z`W@S?)DXW|U*IgX+OGV;!1v#LmpdPxbMLZe2%@{1=rOKMiv-5H{P%oxd;3=rukbRn
zr&^inqcKK;k%Of-Guh~n3Y=xD>eTG|*mavh;W?_(_YM{IufO-CmmrD*3UD&d3A67c
zLOcxg+qEEIV;KQYz%~t5*oy=vh5)X`wXLMW_yOxWtaHRTuw7qBNoQMlcLtuuKnTPz
z&~GmZRyypsW{Y!<5IpU2iS16fa_bshyTG^>x9G%98Atpu;&(kE3=C1?td@(TR6#;*
zh{aX1M;>uHsHltah^yY4e6X2CPgl3d%pb(Vhr|GSABo-&8-unVm##8n7I^tth8%kp
z6Dk#YNc(1_w$}X6AMzB|pUH!V6a}xe4H1)2a3PQ<K~XhxQ)f`APAbT*E>k03_Y@oX
zaTdMHkS%zgd;WQT^Kbk%&%XG1OxMBgg5%*HS31k(#g@y9hm04F&34T&_T0Yv0Yi|y
zEgNC8*|FXZyz%}yuf6p?ckXZ5`qV_yqzLR-syfO7)!v7xd-l+k>)ntR_n>qn^O<bg
zGW(k%1Qf}=&+>hg$vk2V^`;(RPLE=6kAmihK6B``m_A>t6?$3QV(fypb;Ka?QRBD+
zFKWD$8}WD~_sq5qYYc5_`sux==@uAcXxm0?Q86TMZZ#Gt-o6ON8vNMf+D2Mu?RH=z
zt#d5fYn<MAhG-rAZi|!%k+`t+<vHzgl`2+5#!)I`geW<Nnl_TeNLkZ(V0E)hIlPvS
zv+K;iRWqpiKAWvnfZeRuT%2Ffx`yRykvqp&YiYZMej%%pArYX#qzLtD9W!c?GS*=X
zWpy)Mt3i9;7Nm9RjpnNwd7d#Me)If8iE1}VThaT?4gYghvE-)~<~}#{A4aH;zMhUr
zRgyjsIk|d;ThBdDw^-4K!1n(8-003R)^mP-pRwQY`kNQ5FVDfi!^<_Mk!p?S4>!F2
z_B~#E=RS8YHVhuH&4h{UAHYW=i9>3Sq9WP4q;||<f4)H?zE}=aeIH{CQtT^X6Eh{~
ztm4dWeaN6Vhsrvf`G$YgVR_U#98wGGzl{;Q%Fypvbw}VV&Mh#hy9*IqG=wk^f){r(
zDM&GlDF;u~ZRaeFg?@cO^paR>l?fQl)lA-26No0#4?D~#t*{7Qf-{X_=(h~xmMb@(
zrD+_E>)3AA7-!P~ix}I`G=|8Q5F^{|R_3ScmNxQnpwk#roVAbGt3!t{eQM?<oZd&*
zSOPk|pvS1|=4N|AYOs(x0~^3OoF3|ghUW)e8H~JUan|OBriIr0KAklh=^nK*e6pz(
z24jsrr{vu^+f3%BJqhNa6!mBHy=k|cLc<?l#Ces)zFe(%_J!vd#*q~;>^6M(;T^j8
z9^+}ly$>(hJbb|G>n&|#amnDg-d*tC-E&@h>khXc^bB<3rZ#~XJl3e1vVYJHAq%|@
z8}(w2Fjqp(YRF@lL9NzK)U2uqpWI(w6aHoL7w2-<m@p;&M|d|riYYyb1bQ5@D)Tdp
zNj+F^&RHxvmff*5Q3#&srTT<*ju;HVdxoKB=zED(YNP|P)IfLG7})MMQqMibU{5e&
zLI=_`Iu+0hf&v=X;+j?}a{7ULAKb?Kk*im4aOnfXIIvhQay0y+vG_PJ?0RDK>~>;+
z4XK(!6g@5rjS5IVa_#0enOaSnr<<6h=Ie^a(iC`lPsAC~I7`!5+Sbvvu2P9YW|YPz
zv};UGMRKMz@U*rXkH#J)W~^fu@Nk$W6v4+-&kJK1)o`6rv*bP6WI@cR9~jLzV@TSB
zWM2rDu^8#MXr%&#$qX>5rRHM_LHPXVK8LXuKaMnA%eAXlxqIh*8oQ$J*Vy)m_ue_@
z!3SIJ-rvyr9@`m=IpV{Mp7-xx^49J9+&S;r1dDBtXhV<W);T<yB2v-~(n_WUuxTU8
z5y%IkIvI1uqfAdcFBJhB@@u*w69)HKCRM=&OoqqW{-~zfG8U7yjLCsX_K`;!YgKv6
zhmh%5A9dwJr^a4l7X7w4mNs4EhTtQ95c^0!Z1H0cVU!4IBYR<hwsqp*NG--C)+I6y
zJE`JHN<-4t-PBsM$*Q-+Rdv{=#Wo$TThMh!INQ**7VkIQz55QXSz=uW7?wv1f{(2G
zOPr0Eall#Vce}I$iJ`>Q)?0Tfse2OkHb>eARJj?Ws%=p^_FT$ihM@GfKE)a<rCO{U
zSm)ftFrI@&Re)A%RZT}q^NONsGe@yyCHSc7XE;#UE;VUYyO+{oUFO$Bv73qaP1bMT
zKTyY0BQ-zD=QZSFvYs|Eyl4{Bwd{7ggEV{SdMYOMvtR!jZ3?>Uhk<6f!Vfz>eD8Jo
z-Hs0~FF84Sfwz9}O~U0p&Q6b{MfQV0?@s6!E6$#~%IR}<Vqg=-Bn#8`qGAlG!QwcM
zDa{~O&^Wm~qfaWRGAfgD&Z<M47N#(OVynpnTIuBaW63ohC}on9eM}h>fGL$A3^uEc
z5ps{gXkF^EqjVGtbBwX1Stu`z#h4nB#8nAv<SA)x<nhn1DHBYs18hv88wL}4f*Fx$
zaYG-rwC#$c<_g2G!-pMhNR>E=qM8vIM+rZWwY0Q0vfi$VQA_~Q#xjpIHrWe{4KhjV
z8jS4-&Oz%i*3dOzOklG)Cxnr<UC=I%=$1>m#<G5RAGbOI=dc9)Fyh>TFs!q!D@$<Q
zn`#`7W9M@(Rqn825QdyT2vSPg=RLB_m__<~^nu+tu<d(>ex%>_G*0SrX-O07QhE#~
zAAUoeFgj@hQSLk0Uy~L$KU+BzV2dvDI0P9dWTIgmQV=($MYN5<MX492@K3dqd|p}=
z;7J%5D-~|7j4K$j`f7}Hz<$!}shHHW&ps#Zvtz_M%ljYP<@|EP{qu+PyB&TE+}-t<
z<L8K><L$fe6C!;1Yd_7k=U!qv1VZq5uM9{Gus#)7^P`kr2_Z0yBSAZE$O;mJNJ%x^
z%I8C3UgC8TuJ0L!k#UrOx^YO%E;BbXAxemtl=FwFi8i0Gw9!Q}gGi$y{gCE~vitH?
z?GJ!#&>T^z2hotEZj!bWVvGhK3)Ftnt44x(yiq(&P1{LB1f)RQup0nRXnIzwQ=AdT
z6<voyTkl6<R7S?)8izBH-FAx~$Lbk{kn>m!fJZ%|F?zt$xFyaW5&Xb<eNNN0oE%@l
zwH-7DOJut~=c4biG^~!K0G}UsbZx`7Cv}cCdCs+cN~Oho)gEp0oPMSbrN{^J3CfHZ
z5L=WNChvpNwa;`!N`DtMPc#IRZJuPtGiQEK+Pr6m01&ERQJVzjW9F9UrRX+JwTSw+
zt;Qhp6Z)R!?bSW|64JIwHBB{$Og;@TjvrI`VB54b*&`8n;rZwJ{tsVgv)d8EfQ=)^
zCrd8Yk$8N@*=oVH8_)2}OP`@#tyt*<XS6ZT01rVTk?Yr`qmgu(==+{w7!m;lMz4%2
z7L-KU@G%*CL#jL7_dX%C6uOT@x+~KhhJnG4gcwW9>>$i8Gsxh<BdSu6_D|7-NmBqR
z8xm9VN|Q_6Mv$7}xf2W0L6|DqIGwOcTgnL0r27$LNg;cTtt#gKM~rp2<uOBOX<S3&
zmb5DaVUQ?ulSXVr>JuWak+JN?Q4pCCv+JTUoL_9^?xs4@Gc`z9w;}`?$KW>%ThD@s
zbA}LmcDsR-)fvvU3~}W0{2|c~xVB@_9)U4zH*0)<fm=yp^KKZjWH%ME(8DrU@#^c4
z&2Rn_yw9HtsW71Ih0O@fnsT5s1N1?J<Txa)KoaC@HBu{{I$MO(zWO;e8>j<EpH<Yu
zX+J*w{Cr+}^PP2BKIC)>{!q?pxd&($pfav(V~fg4V-*&rH1TBDQ!%M-v0!;5tcEB?
zxR*cuX@0otxx9Rc!QvXj^UvMHnrApWJ7aZxLff_kKW0wq8aFjaCM2b3$6Cqv$}uOh
z8dwW$+i`SsTp^n%%>vTmG|7WeQUHXQ>RxEu?S`Qr$kwGOz4atIIAnCC`uI3Vy=tk^
zE%#9GP4H5MF-GuVgdiKE^U==^<5*Q}^n}qf^gTmA(2oP-PRgGYI}((->&#R-wLcUa
z7qCZU^~bHXsS$$1r;H!%F6>-G2!f7GOz}H@l&YOXq6a^SYUrFq`7hd*?dF0QMl^OT
z##F%2%v2K)SQ*u>;&3!r=Lmj7@P=-&5;_Sz_wSu!T!+QbELIXP)pZQs<E)|SI>uqm
zFpOEHOq-p4l|y>SuzP5&CX6ZVnX1{3vRh?4PW8H8ejFKgJ-eMG$M=0t)3}_oRx^cQ
zr4E;tHp!m6y+rg>k4YI?T3xGZrw}ta{%fKYOp~DNd)8Nu;nL?Kz#`~6m`rSW9w3GY
zn^jiQvqP~@R;uM=<mtidshE^&TC{PdX2u^MpYo-z`~(jk-ea@g(zPwC)q<vJF}9J8
zwjtnblbblkSVHNAC<H1tZI@AkCcY{oniv_<6_Ra0k*zR?B>9p@&473%Atg3i+V-@P
zN)#fGV)N2_!4%RelB+!(AxhkoFe#}BplX4tsxcTs^f^u?0z*Hrxm<Ja{(T-ic)*7r
z-e<E}3&R>mOib}gh#DJabFWO3K+*Mu$xqA2`WUf6Y8y4nr37dWYc^XiO;w5bAf;D)
zzk}$d*;N$A6|^C!bHvn>uW2po^R2Xo7YZh%QI_RQ1d2_EMr;*QYp4~wX6IO}mN?rc
zreJ8gCCk;R1Q?oBPo9V$cJwiDcGAJ%h%pdSfxh~h7_sYpu2hjZqBoGdkMa|Ukb)0&
zZS#F?WgJIQgIX+w-l8>$!9_IIt8vo4UPA}0BiXeqR`M!RCIuNC=CXRqL=BMcXOmY@
zDxvy3H)1hf*Jx^9$Bb%Tb0^VczKZITT9?ODt+m=DE!A2WTb*Z<*8yX42*i_IPaT3_
zos+sg)ui2W$<6NBy+tq>9Lb=nz&olDwk9UgSrr;)qT(=AEek(Y6HQ`n09uT|B5-^P
zflDvE!i0RXHwdB{lYPn<K_CX3v$xChGkd7c$&{(d4`S1)Dn+aoE6z^OxN-9)7Z(?N
zc>g}{zyChBZ{OzO!v|^#%2G**CL}kqN?R`xjk$H$ypwDQ0i45x0pl8MyAp}oF6dei
zqfKDZwisg>oYZ%+R#23+mVOvgRRpmmI0x&?i)2r;*{)%t9B^x<-9og7VI;Y|V<5y4
z;~dep?0Qer3^;2D(b6oI!Z^o))*0eBu<=`P4U5GROlrWHnE70xJ5-+5<U_^OpkIh5
zBL!oXsS1P8*rD__K^BTuLQ?C~=d~xKC!{XSBVm+s%;PBS>x~gr&m|QNX9;OP9V0GP
z;D|0U$WmHwLTl<sDhx%0d5zKzou4^^od4ce8%1+6nUkxcTd8+x$ux#5tsH7cGwPYo
zPN*k`Q-6Wxpt67}U39Qb<A|{``o_YP>WJq(4+SLmeHG(k3?@#4!4B!vM!O2nrmR{p
z<ZYe7d8I*<7g&0=q_#+;!w)G*R8f^LRGs-~C7F#$!Fy9C<)h9$c|uYEp4Q3QZrAu>
z!?0ZwhCX?xk|#2qsO$l9Q^{uy;4*fRb%<)`is|5Zv6~pjnAGW}Y304(g=t{0Nrkl5
z(X}nnHe!Ej##|#Qj$4S4Vc4dK=%VgKt6tSw12dFmN(N$M2}E3D>6QzO5$|W)IEG=z
z7(GoZhW*i}B6rbXoMr4IolV(*-V>u#2(1yoVOmK6kH!>rC{9~6W%|*7n`BLyBaB(j
zT|rou#=0E$6uife1AZKnZF9hwn9_g(VQi#=K<H6w3I;q5V;!b(G|ovU9_u6)&zMR$
z*R(J_Ux%CnHOihhR;cGP)gqLv=LyV-foga`e!XUZA=7V;+eqi`akj$w^;Ar%wU#!y
zO{JjzRAN~uoIDLWm=fEh1g!5X^U~T=`tJ}z@`fqOip02IW+Fi(`7^mX(f>!1%H^rU
z7HT4vg_l?#j1r?9^(Kv2rLRI<;NXFf;Jlc_B@8K;Eioi9mWshpP@8AFb&O-fa%s4B
z^@LBq{2~_@YtGLvxLjYd*+^Q&X0xT=ZP^aRYnkUGD<@Skk-$7v1GI#OF_!29w(W3j
zM;txF`a+DiZG&s1M!DQ4(xhext%Vpy2}$Us;-=)A1zbv6i^&ExK`FJ{o-9WxDuXm}
z*ll`-UC(N@!Wq~N4!c-L6gn-zr_MC#a@No`4bDc!lmn**+s8Q$Nc3Qx!!{)XFGh&N
zAhtzPxtxSdQ(5=Dq*xK86&9$NLOl*^8(h=TEsp4xE0)V8tHn|RG_9<I4?Cidm=MyB
z2RkAuVGXu*`0kj{u4s&btUl%c<uwhVFroQaYQulcybc-655J!lkm-HPkvI%K%;Y}#
zP1G#&q}S7V_iW|<G+>hlNTpFNS7TzqCLP@9OH!jwfIL}kl^~3!5qL0rdc}m&_dK&i
z$*72ooJ1^vYKN;KHjPT)M3lzhV?inT$%?ot6*L<zD<%aIoSklZ8jCP7jnp+5<7ie#
z5aG(n85ozLyVznNgcOG1J!3zx8w0z3WV6|@Ua#41Hf+|HY_~nz-Invj;LaaD<nr=T
zjQ$~{MhJnnff!Of^Uh%m3}a8UEsHKC{JMswapK|Z2Zo`q(1yuT<JMY2l&EM$vbtvb
zF6$I-L1}>|&+LdesT=MGObocig0@@G8b@qI71HiB#K6$6X{|`+psH4w+7agmP6y!{
zu?YoA!t3ZqVi3e^g9s~P%+c8)227kL^F{+UWf}x)gqex1L)UdIkB(R!pWwO$yCJaI
zj&$vg#x*QgOBRb3V<XNxIp~c^0@o7lLOiS%MiX<S_7EgY=8&o;uT4v$JW2zZGxRy~
zJY-f8Taq@Duw_#d(~7nrrA#a7N6QBr<HW~Xz4xToQ!=TNLzD*^$&C-|$;D)waYUkp
zs8ScG0a(rT=k=WR{h}vNlNTL==Hz~+Fe+mzX&J@ykg~iZac`sQjcGA6n}*3hLWcz|
z%k!!d73o=J0}|1jNNKO0Y;sXlM-w7Mi}yQG3r0@}qUvcU7dZ?T*H{(_-*s(6JUhiT
zQrq1*C+P=7hTV?!dd+6DVY}VRK{SRx``$ae`u(?wPC7$3UB_Z^MAtMJBHnNC{f_>!
zCyYMTL>k4|>srP^R8<?jL=RhGyyE^Y-c(s1S+fvBg+K+RiUGJ0Iu8TRSzNooHZ5%{
zUOk(di4tRqsfi5z7Rv(fEZ&cd(G!hmyFw!q!kGvQ-9u-sh(BWp#t^I{#4$5ppGr<!
zW74sc1`-&l2B9?R6N0fAOw-XWmz<rPaC&mWV$rc$E;v0t;`Ho<D_2jsa(2S;(MrM&
z(ujOhsZKc`FV*ET>2OtMkVxD^L*)_=bB&&Bp|0nv(C(bh(;BKEPV}e2JesSiz(X}=
zUEZiD3t7hmsa2~kJD_LtqmXZPJvEZ5+b|0QsJQta*CyL`WTDdRo!MKsM=_2<Xc3GT
zt%R6~(it(_#!x=1(i%v4PO?O5%-^3^iEUDEjl`$aNs*N#t)Ot8yG~e@__Q*L4wIrM
z8*CsZgR|H?0MpRAhIWyR`iar&n%Wps{c)FV8c+gQ$8lumcldF@0gJ_g)oKaJW;TS#
zKmUVQrFlYX*y*g5vM0Nq-FD6Bd#n*e+$>r;+sVO52OthRuACn8Q$O)zY%bSa>@L`C
zw+y?1VHhQTZIpBxKU7W>iiXmpwo}e{`f<cM$CxUEwlSt~h?owX<h^O&pZ5W`fTn91
zM+YI&CVSKbG(F+~LaA>Qk<bh0MnT(Ig{orQV)EXl+%r|<oprQr!_jKR@$oUs<5O0v
zCCf#}@oK@@=?SMNr(8KZ<>+Wh(}-8O_PW+7GKKa{E<;47*~&zc?M{lub<9Nwlxfds
zlYMPs-VH_>rme<`DJMof%!bOS4r0^Hb~-&5CZp2&8TzS@t>~S#)A**n<jeHCRu4=w
z)sYQ_!|UMZ5OYFw%AX8b+SCU32*hY|{ph)#ZA{FMdlYk-=<mw)4TW$roGy~N%<+Kt
z5mE1?qi+PCyoXwUUDvW{QYqxA@pByIyFuZ76njZD)A@%O*={!Ug9HqUs_1Bwn_KtD
z7{j>P@xi@^gy8UFk1-DCff(poC-FzklD6rx7dQCS<8JJQRr!&h_{tag-QWE;@nc~0
z1ChWuc>2E2Kf7JecDrS>*>HY-&c($Amlta`+cmr0md$3(I2b;7u)!^kF^yxl-6W(c
zbEpgWXLEwF^t&w_FIX;*v841WCt(fQ777zDtkyAd&Sn)~ZD&DS$Y(i|0wXb*iRgV!
zPfl1JpKx+|g)3LiSau!VqGj1N9IcjIIX&j=<cQU3N!zrkGC|C=Wb8C|PY$xlAu{z`
z%h+RP08=%sP@8^lPSf-2P;K-%FbITH0%ra^vu|she0Pdi^)D(~t-tq?HpJ?BdhgzW
zo6irwO&$r^nE9d^a-!}`56-`dn3*p<_#oaTWrB#l-nOlHqT>Dr%iq4{i8B>o`kGab
zg)x-6Rdu+<eDmVo+?6P;O*Ow`geK;MPn2Mds&Yk@)DN_+1sY<M1EuLf%+`q*SuB?<
zS1V$SSvpx`CWpbghPPh*KD*J=bt`Nl{Khs?*L=00>3hNu#H(Z@5{fYpQ~$V#1bM#n
z(u?$dB#ep}Ep6M-b{!{4pwvQOZp1CIX7GRdzx+ph@Ni(+E?F(Xnjn=HY>+rRXBoy?
zJvYR0%Wkvf=xB)#DF!K&I4)ho+I&z9x0Ol3$nlOzRgN)8gVGw+7{}3S#kHH)xOVLt
zC&wo&mkZjqW3gyiEjmt4j<|BP5X9tKFp<VJsp5?`rxi3tQo;0WOq8{Zay0HKPUbQ<
zV$aPZP18(i115BtHHBHByi9dj8<_4P2~}zrp85C6Gxk+b&4ZkpY)lrWPkKF_cdu>I
zVkq=f7iH;8nunMYZ!=uX=qI)J)<sVXBpX>vm?^Wd)?r#PB5LW1h~)AGFOlL?buK->
zlE@}Q<5H6+lY$YdjaecU22zk+Wn`QYKv5UTh*C>nsG(Vmpe}_Gq~dUqoRcWbP*))4
z7##7SNnk=qxsA$@ePUk5T8U`3u99OWzxU_g{hkD4;#D`sIU$x_b0~dJ*m)!hXe5pv
za9||4v~5~mc=1y#7Aryv&PV{K1m23*6379(J~bjBmG+QzGZ@=o?T!$80;5#Wz+<gt
z7)H893I(wyH^dM%#L;gC&L2MD?DUx5{oUWi?=HB!Tyt@G!Dh2&y;*a4xnaBAvD-;`
znPD6mhf!)%5rkH)6+3AvOXG~#tD2_a^yG{ySFdsX`VCG`j#;c0EV_opqUC7Oa=hv|
zN$OqOT3Y8IdJI;AQVrO~p{^j&K!iyZ#wnqXYHtd;(kodvmK_wwghWrMr+Yt5rM6Qk
z6SgSLnM~itn4CS~eQIx?&XlN|^5eAYcTDyvk={wYi`h((4dFA!IY)FweH0;T-E{nq
zyPi(nGhj8=$Epfc7AaCxmIC>TYF6s#gtCxEQN&aNC^yfWnwvbR4Myw~+KEFRR@$kX
zglK|7YGaZrr@G(FhLsSGtq@2p6|5Dxh~_$l2tn(Do5^=n^|ehSEtqBqw4R^bTnK97
zm`L{-y=P1vbR@{K$qXulKx>L>D@Ds<ieFMD8;Qp9?%Q`#g7F3u0%08K36hRsA^1UB
zLnReD#7Jvfd^EVGVd#5~79H2FU6D$d)(TG0Xnv#-s|X?4@1%>THAyKMM1t>eZO4_X
zx484+TcTQqfOQSsVo3;|ZNFvZmKb9sXwz9j9BEn!4v8`F<*$5+RRctc7YKx$$QwqV
zli2%yVCX&lZb!e{a(Q{d#l=I;AKc^O;*y8w=WI3`Y}0ala?0`P6^@QiSREg8a(v9`
z@rtWgj=6Gn%xc-vIZKneB?l9U#$im8G>BFzH8__9p;o~x)ePi3sCO8G)*y~r;i8-d
z-9tg=1+~oUlq*e4n1dollW5J8=m8|mVdk}rVu2J*rHZsq=h_;Z1aCP<$$y{D$iCyN
z7UYFd&GRQ7e?4^v<J#b+1Iq?LTf96|+f=P)RpIbiV$?=ADiBbqQn8eDK^MtL?2O3^
zE%{-#u9`V(PDr(A^)-dmaev%N(yG_e7;!&`c(}dLoN>*cm8YK<7YG4A`sA@pVGT)w
zIV-6o`}Z!yneS^;Hdm6o@4kPJ+wXrs(=G9S2Qkoe4ULt8bJjxBb-2FCQhtbmW!nh~
zNdU$`IlrF1Ti32~{n|BYv7V4-P^KJHQO~qDnebDU^sL$@6JZFE^9K(Av276z-j6I6
zORQ~@0omfcq#+r=H63oza&&wQ$-r)H!w`$$5P8$kv>l6u#Wjw`HINP^sY<4a1Ndzb
zs1Y#oToHVdBWO&M1YW>IiMdn5aWpB-rD+6tqv+UFhWMhkP0bo6Rmrd=LAl~i#U_<_
zO*Ruq3>KM`r|MOqwSt6#-ZzRVC59B*kn%Y*LyJ*Re&TF;d-aHi81v_far}^4uj=GH
zyg|qN<R}WB7D+jaYZ@fQG(|<3eC`D;Rk2H|YQU5U=#`=1JePxM&1Jb4K%IU<vYk{=
zGOKH4p;XP(4<wvWUYRi_{m~GES}oM{2SqwIDsr3J3hJwtQq@Bz&#Hr0Wj-cJ;znsp
zptb_Wd;n76xtL#+1E%8*fv#Qf*4yv1?LDmvG;JrY_Baw!9$mD_xQj`ZB_deZY`6J2
z<IwZWty{!lBn%@YL$tBtst+b9*v6%hAIZ+zLy+mUe!J!2g9jK3u5}n(POZ=+=3GcH
zoqcDeQK&SLa1t!JTwwf2jG}&u3%@o?NOxs|l$wb#5Pc*>OLRxV)RG!TI*2COOhX9R
zI1rOMD4x+a8TK3bF)q=n?1`EzG`KQtx9;MMk!ZY~DS}r`3W<JQlZ3b_-_dh0(O71x
zmamYNl8bdUVw$Rm*)-t`k(pXEsf8d(4NW}Q7?Y}J6l04DqRe6IGeHFvpWu4B8s`ch
z3yZB>kK_s0pO0axKhISkIq;>)C)Ho|0naLsa%$@^HsnPiBnD(8>_M5nsh^WJEXH{)
zoLa3@$XdUz<4Vt#J7&qI>hJt})1`ZK0;dptQuoq0W0J@yn9kC6e+_lr1<@piW95)9
zx{mL?_5+%(1t)~+M_DIp4IzwJC(RW`KZ@7bOER=`4VV-H;5}dd(ib^8Ii3_O)1>i`
zBF{x~SFt?RTs5P>7{l>N%dK0tc=PuA7-MLgju0hvBr;;GWwks)yHyck90R_;WNaE5
zv*P0NlCxDuXB^RrgP>_zaiN=P?T{AdqGGk!&gHR*7glI15Coa$!v|t(5;`{s^OfM!
z0ONoO10;JRMq{hwz)q_U#7rx*jT8FG=<KixR@9|{p~*~YFVup9{wL?GIs^H$hGM>$
zjZ0g`VJxkkhEJ%2NsyJ?*K;WRXTJf1<{OT~$5i8d96aOTg~J*uHH|!oJXvFMQfJK!
z=Fl)3p><;VU9E+TNzF9@6N?SYWCWXFav}m_Q=N3Ze?fYAT)7Ofj=L0=mvSj{cdfFe
zg4p!?g02e0POnZkJ;YMSx=O&L8o#PdInNKBry{~qx~9x4N$Lv0Swt!}u3<D%9_B|s
zdPD4q7>sGKZItfAO#?Bcnpi$nI+aL#*S4uDsA1P{@xk-Lr(R(2fgy;oyg)mtVXs@E
zW6s7{D#uEcvN+CJu3o#s>u<l!IJAuYny%{@{V2uKBI$dMA&klWZKcSa^$-kw?^&)E
zL_g+;ZLKF?!i7R&K!q`dP|}lB0)!YO9VeQC{*$%MSd*AnDeEh1uFTLCgN4$(F$rlW
z-kpS~sm_D1669m1XP^HOV=9Gg()2QF*Xo(24!hMhZz|L&gg<9M;&jq~Oi{f_xrb`x
zx3)r4_4!TIE@qxIHdE9@YYfhq39XK)VC6Ut3?Ew~hFnP9q(!en;P=o`S)hD?4!SPM
z3lS&xF?j>#b{LFGsDe&R+GIOd6RC$*JSNkGNn0Sl9=1?axQQJsUBysY#%hd7HiV=K
zWhkGdj8TM9qKFlWyNVgg0IGFLpNSc{7S$oIOOnDMVFY6^G)>2wZ@tUyJ0H+A3qp_}
zOR+^nVi3=kAIDUsA*Px}A|PDTWY>7pIBwp&3D!zUomy3h31uYh!!#w}R}Z$S4xRL*
zi>y{l8YdC+tHp}T%XKo&ic8*)UK$1%%W}2CGz;u<LF2?@7JQ&vG=xx6LP7|+3PX&s
zAca(mBhx_spKL;d#Z*CzS+$uX*1ShH|2xE~5C(EJRZC&mraJhOd&Ux06QX1sXmGNU
zUX}-m;n%`nXd=;;)g<|t1zK&;o&bXbh)d@eWkHJ_3JIJpd8jrsOUXLW1`A}QrO(h{
z<1wgXBhY`GefP@3#Z=p-s6gg`DzaC;WzKLotiH?_wbtWnk6M-jAx`*+zGo{HB<8Xj
zMdi}kKFVmy=Fg*jBQcbPtqD^8T_n~V;tB&T_YAY=3&Ls=I)z*>D5To#Nk~chS|P3}
zk=az3j9dh;X&Zj{`dbXaVqJqDdR#(GP2-ZSCg2*9O{gsr+=~I<SVz}&oShzX^X5&A
z6K`M1(F$n1tR6_=R8+PW37zm>0z$`e#I_y7FtFQgaovK!_FxTNw~)i?wXS{0C?<qR
z(?~-2<>e(Gd~lb@XQZALKopcLab3k8qiQ*oP*W9BNv1}5Mw!<^dbC7rMUye3AY)8w
zHb6ElnCkng5*AxkW#Xu{31f2ob0@>b>e-SHOvV-y;frl3JujAQ1#Qq-M|syYIt2pL
zHNlWo(8iR{6#+@v?_2|pOdGk9ZK`G~wz6bOc*5)HDsy&Ss&TGy9(5j3Qk%lm?sBL(
z;Gu=mE6x^nE}@Jd(Y(0%Ug^D1JsX&IadDOEW{Qqn8`9BsLFuMhr8S~@<mhKrWlS-a
zs(L!L%jFxP`A>~eqBR*o#&jqZZHPz7Wt2qvyb_HH3KR4@<PyFn!THX$y!P4;6B;r!
zO`B>kk)A92)piZUF(FJZRS^<G@Lnp2e&MS>!D_i=h`v;>FeV$L4M<#9c2QR$Cza&3
z#?ZAbu5sLc?>(Gzc<<Tuy)e=wX|1)nj=k%eH2#JVM?&zNpI`9STW@3d1wt4qwSp8I
zMC#U3>?E}y0ZLGCHeWnCceLKnIJ4tq#G)z>F;$C52OKdhn_wsRt$XEVRD`UbHCk>u
zVgE!&YHOQ-Zq@86CZny^iY*K2^iz$yo|n#hVyn$(Lg>&}!Orp>JsUB{f~grGRIzc!
z2<=(g%&ZmdKFRfTHmOnjZ^=Lu(gC@kJk37SnIev|KpADk>F4Xu%;9lF4s@h2GN@3T
zn-YR`mB%5ELk^+bUtgc*BC0A>bb~S;MU&;Ms)QCPp;ryF6|xuaV_mE)c{n4IaX+tG
zXp%8HROqsvqqXYj7ez9N-C})l!H-`55v%2r7y@0_=Jj+Ksm(4hPY_Im(TEp!7zg5*
zipvcnKk?%~4iTw21<BGBMKvS^U8Y0f$9~dY+f@}Q#aJ)GVdw`m5OnB!TnB9@6*{#}
zk=6p&-5>fbZR=Pp7hGIyh;d3E!qk&>Cy1(8lEG^yh^RItXg4t_DzufD+Z^?4VC6v}
z8zctEwM||pQO_znVj6^>RR$@tu@ZXu8O31}qMw0*SzRxsgX-+nsJm8aYZVv(C-GeM
z=km}p{i=Z1GDt{M6r?(62-9chQKdR!aU2*vu0{+RMQlu%8tZ1T6%Ge>Wf+|dvMSw7
zY7{v4R7@&oruUq4Sl8wQQCNWdo@_uSHf1RLDFKQA6;Xw>=!%r2d#_1jiTShE7@}4v
zNXm|NO=e=c7V~0Z^B}xhb(0Tev54B-QZ3e05<`ex{&E>^i`_?T8P*B|*`5zPINv6F
zSD@cqAVuOFbSon2l`ID$;t0_)?6xeH9WFSQ3(t>z=@nw!VWQM>f|yj068xIecT5zU
zTttv*(z!+DQ-o!SG-VDJqWAb=5L=iJG>xQKp^5Dw;`_j|TX1~!3Zv;#U7|6suQ7sj
z)E1`e9fIzCK|igPphfa4H7&uMIVr*9wNe$yDa*FvLDdGgF_x?W6@Rh(D5J~hUlEk6
zgj*DWlVhErsIt`$Nd?t_XGQAyEa((7?Fdcf<S(O9{x*+GY0_nVqJ$KIe6|j$P!-TQ
zX(%%10nx}RgfQ)h^ikJS=kC?qRdw9Cac+K1(35;4Et1Wi0hPFVuod#>dpb99VT-Z0
zNV<Z!Y80|BO=66b_EvK{l1VkQa-@5RxyH8=gSEP4#kfw~pa0$_&7P^xt$!!}RO}@a
zVzg+mhN^|BABL={0Zb&Ch%=5i-g=$&c1L$~f^!C}#g98-N`9N8_8V)Z`ao24VN#Hy
zBW`!R^x_LV_sp{l{gx0sWAcRJHNdte5@;G*T<RnhBFxrVVjZxlY>YDvF$kg}8DYKm
z`B1Fa8x{*oj3e9iIn5HzAA0)ThHl{!Vo6@x9D-2RFqa9cfs2_oMUdB~`!Y2(B)d|D
zO2~6j^<TmSR1D`ErcKS-nlUAD*eVUl=cnUQ%okG<J(v{KI~TN*ahm<Ps4*YZen{>G
zn9o#|4rWv3+milV1n)#e4Iw{I(Xw;XLIQvwKm@6`?Z;%38V1iW_<V>pqErX3*CB{m
z-V^Up`hX0Ym|?OphQ?WNCNqE-NzgLp4JdVUlyK@&6y<xecyY40^=9tb%c@Lo?Jg5j
zE3ATc$VK<_ykPGcMK!EXu}*!e_Kx$zFcS!(CipymHIl33%=1mg@U~m>>i55ob&jTO
zX&T3RJ>uF0hDg^qY+?*CXc$7kG)_VhjHT<A46f%Vzxowodd<0VxTVM56w=2i_X(p^
zAH@nXXqqOMCkY`)>EK0!_am;8_F5`%8s`|j82Ftrj3HuOOVhSkd(7Bha{llh(XUu<
z)*K&QJD^_7*Lv#AP&z4=m)cXNv^r&U!j#gwXIN9jZc+I|QQef03I0jzChs+=LWriV
zL`7Lss*-g!lrr2*$7SvQyzAuZayMnF`)X<i1-U+(s*MoLRPayNHH9>24Rk{Ca4XZU
zmMNBz&r2{r?t01u<2Ks|j4AML&Y%vde`R4pZo~lPaF_Q>OQBGyc-o_EqQ)65Wf8I@
zBaV}?sE(}vR>tut<QEQ&-Ac@h#+O7T9Ez@9tCID6=MtyBcg?`EttfT3&9<r;qtdle
zrfz3M7m>g~KY0H92d@!KV7Gb5^2$|?k5}YcO<1h$iFkbMAxf+4##urP93P*i#N3{*
zfBmOr4P1js1>%UMGe{X*GLBnAv*;v;(5j(yDvb>irxjzM?|0a)G`a99kep!*BMZ}E
zt!4C{T|eUck>let+OEMHp^dp(eQ2HMDmGEloYE%`(}dB4bNS4~KBRmqWm*tjx?d<r
zxN@+ld7?a{Vm28Gn<>3$wT#SQ$o2otZ13we(Eg2+snqL2+<Q&eE5rGO&~x5p)^es!
z2tBuI-705ba_7p4=4jiFC%vB9CZ&i*jo$NvnJ1cuR&*D!6R43LHN9ezM<Y+u2wZec
zl?y00o{T4kl=O&t8zsCTWTRs8kR>mg5Q*8ZOQpoil*#6D5)n+3|I8H&^svRG@Cyr;
zPl+8(Rop3Rnlp8+byO4E9HGW2o*&ZpU1bm^>H0B5X<rp1m%9t@-M?UUd`ujAF3%t0
zx+RO{N&+Nz8$uWv#x33tSVYBZ+Lkd0gIJ$m@XRwedFjO$==+^y_{W%P$Cm)q={y{m
zWDiOxBG0*^+RbLi#rcJ#{}^eo+O#c=voy}J88$SwrEM2<?E=><Xu6ig(Gg)h!h2D(
z8rx>`z#)X7j35h)w1=X8rZPU6(S{VhQSvo4tW$@d?NKqIDUlS#tetZB%;fP>S3*0<
zB&9kFnwl2s-&*OT9;2~_#x`XwQ;?|Adqk5{<{>KN9LO9{oXQEQ^Fq|~n93<j`cVFN
zsCx0$&W6OaYtx<_*^<_j={RbuE$IaGq}Nk=_j<f<v#M7Qpf&)Pzf(Fv8n{Vn652pA
z&B|Ko%O@z!CWfVpQn2Sly{Z=98l+Q+u^7`#K2ul(Ii4rG`s@5AWAX+o+A4${vu#Bs
zsH#~pwUZ)^w6?k>!I;_fyi0A(Btsk%%(jz*U;piJ6E&i~j~t#<{3DL6cUyKn;FbXR
z^_Dn{ELO1FTr%`)X}1(SP1|B(lw{+k!v$HW-Q^`e|C_%?*R{m0e3qzg0-2m5zj=YU
zOc#}O*CT1I<>7;KKDhS)T$`K_5W@h?QXCSl!!-+>ZE>!n?K<()b_;ePU}B&#14muz
z0_<FF45BDPXfYvxIijzKNo&uSf`{T27WHT9Y0XT~6akd$*OPfr_JZnnf-xx@A_?8f
zxUXxaW3MmGgA@}<#Do-Q7eSJflSk4pabz7r^_j`of(R4GPIN6fxneRf;;l=jlSze+
z*?Jgb@Ty5<+hE1GDtgeclz9`E$6Zey%sHvLIf43h54*BK)@YKh&B#k7Pl;!d1@ygF
zHj$Hu_5KR7b+PKug!$p9$5ah6t7<_dn9hR|;y+ObGE6UrOKntYG@leC%y%nHDv#fk
z56ziS2qD+;F;xSKN7?%hA+oF@AmL!U-43FqaSfh^wp)~v!Peld6xM5+R*am{;MxV?
z=o&!~t7XSe|KwM>xHwPk#!D@)xmtRt6G{F%y>C;KigvN!Ti^N)F(zB$I7+>7AT&X`
zR4-3BS{||6ZN)y-_k#4Ss8UjqQ__uM+2VveYL%(-#?97D))SOFsSU<JuJKhnYtmEZ
zfIjcHbOVyhReP<6H>P_y45kXznbX~r5foaiI}!;J%3M_#%uHy_?R1%zbv*S`(xmhp
z&yC}Um|p!X^+-BnCNWTt(jJ_faj?f+Pid1{uvje9CTW~RQH^S5p+-c5TK^M6s8p{K
zUf^-orOJwi7=7;Ea>&M`-y6rN>v_(s4kce=DuUDRDM3+fO3unxj}44r$daI@)XcEz
z4W(jOdnyQG1M<Ew&y){WoWlCC!-O*b3>42w*EDQ4TS5#p#?iJ*f*0?~=zFY%ZYgn5
zHl;a@-eX-S-*Jx1hxd8;<>$F|>l%h21>xF7_IcFLJ~Xa6)B;0F`juc&V+^mp`dx-`
z#6(Yn$F?0~G}v)uckxivML%GgC2iMYT$7NwbX}Ib(2>jahUKbMrOY-J-Rxw`dDK2k
z4!e%MVlW|&St~FmHUF&8GRENQm?_mT@>mL~#2Bjvs)xsAwN#3k3EvckA{?|52KXcp
zXSS)LpQ`?abyRdJGthv``B|CQU!ODg`M5(84pgtz-_87B#!#Q+`U|AR<RMH6F-@$t
z4*+0OE{{po31e;7N)S0EIgX=GT)wKYqmxzrd~ZDa(y<;_f%hgOxjL3f7cGlX6onPT
zeMo_dv5K*g`NeV`Ag2%5Buyr*oeacWZ)j@7mGA0J_L0**^WV1v)hcV#7}o1csd9NF
z6+1VZjm+DR5F|dxX*rlE=Jqjox<$*PbKJRnz%Tuk-(YpLNacCuz$=qEq_RGWkfoJT
zQmN7+OP|0%^q$`Du)(wF4DHbou3cf91+H5q2a3?kVs%1{o~~=yc+YmzbMM3Z+`4hb
zFbsRoogmR_yi?Ct+53zp45<ObA!(n4N<znix>C?V5;(OgS<|1Rm`Ua+xO|>Wr}~05
zHpe_g8>MtJWGs~DkyS`?@y7bcXC|>9aJk<UqU>pXUgmo~Uvoq4WuGCrw0^>a55Un!
zTu&J%rF*Ta-ZXF#WU_7B6{E8e0u3RiIE~sipgOcJPRXImM_>ELR|u+p-s9kCj4^L;
z-T-UG0L_#T)EJyYA*`e=Sf|nH@_I;lNONYP_JwS#iUH4Tqw?lf`iUMEh1m%SXf$-K
z(rVykQA5)yA&RgRXhhSry!qB`XgmDw0_z18HjSfe7g8(U2(rRL=%u<vd$gciTDI$R
zj=Gkg`<b6E<$G<y7gg|P&1mN;-ei3f;)q7-99bVZIXk6WuJ8n6FSg6=ZVeZfThCo%
zd3FQqT4EfDA<%Do7=mOJ*p8-y-MQz^oe%ipk9|4|5?vduk5mwjBA+PCF~n5-tpx09
zjr6FDizymcW>ec^P1^H0YBs69y>`Hs_3;T1DZ?=t`DxseWR=8}37|&s3grZ{%|`-h
z%bAq<mN+MECuS0jNVl6ZFF|v%1|+pP+u#!8*4kQg6;%Gy(#~m(RlsJ+tq!TgL=!2@
z1@T8+Pl=?GszZT|Ku%pK5vUYfK=k==N+h(@-l+-Ora;ZQS6H5eDbG_0qn<3{406U|
zsMXiUSxADlQY-rq*sKldVu-YRXmL;dO3*xS4u|an@{jsi>3&)F_Y)P;=#MFp(}hV<
z#!^ddYlsBy-uaM!`g_06$(5Ug%_W=l1rblvtgto`td%g3<lgSb5gNy8wPf5p<b(I#
z;cxubZ*uGA%^dJKw{573P)p7F??YEEG2$%G-@3`z?{JGH?eds@^jHl2ZpWS5Z}Y5k
ztj?~{EEbI8C}o%}m~o)rjO_X?yUis(eEoHP`ImlX(&kGK?F0!;7;8|0E*{ZpJ?C>(
zq?#LBnBsWRIoCBCON9VL$PlxppkBj_=1l?c8vLr=jWL2XtO~t~d<t#V)S~oVnmpL5
z(QOFQ3!tVQ&AtyIN*l;1hIx6I$@HfIwRt@?z%&m)bXcL;bRH^3KN);};mNP33qgnh
z-BM{EUa1+e*oc%sB;Yc?R1#ok+}Rj0nCl1Y9(oyR7^woF(XC*KM!Oc6%(?A+L75&Z
zAV-kHVaTelc`s;4$DwwDY`8bYNLU}*Lx(h064j^&vu4z$+S6DOK}sQt5xn0&MA4Kn
z8zW%^zxVrp$OjKDxOL@O{9=Xa*MxBZ0{yTh1|jAUVDz479Vcfe4BJcId;N#}m7n_>
zfA{bGF2<Ck)cP57)$UP;JI}i!0%IcHk6gcY1wZVFMj9e4Qa5K~dUm@Fx8M08*M}{q
z*PkU~X`P|pZP-4%&t|uk<mcU%Kl`&+k+WBxEj7=kdz%AmlWV4!0A{FY_M_v^!Mqv0
zR^Qp%ZE{G73e-exWs{BAiX=xupQm8Z<oL3*h|{@Dn9?+8Q`3|p)3h?B{YlvtE4HLY
z5nGIsdX<@?$uw5aSI^@_?Ne1ZF;$F~+twFDIm<I}@ln^)A*pE4*6J#)>iCqxAgb64
z)jowSi>-z!@rGFW#8M8CbE?6r6o^?mLB*^pbXAj!J~K0zK<dbv))2KWkl|;2*bdo+
zG`>WMHcJkD<^i>HvV7@zK1{1c$QqRDo_Rg9swqrsu2z8%S?%rI?neH}KmQ|Wk1zy|
zj*hX>)9=<q1A|Y@&!)QEF|z7f7`EJf_YMBXf9K!k-}oQ=Eym~>lLj@h%~k7B&lQSY
z<KWz**!irH1AODwEnM45BChxBc3T#U70qIaX&ghp<^6Yl#JJmW^~SUG!@zcP!7%Iy
zexPj|Y-6~6`#t{afB291o!|aVf*&ASXi0UM_G|1<)~>dlG8Js~ck21+n(O;no2ZM#
z<@2HRr2B6uT#8)HLz$D(W{goh$n$W73fU{o&daap?-*mQ1*WeRlqd5FDHOoOQf<>v
zfr+Fk)IABytfp6HoK;=(681r^(h2NU>;0(fDVu{fXw4Wk93CpeqL4bKWYaW7)IR0w
zS*nVZm`MmKwGx4uqW9*@7#eFc?3^GMQ<a+t`x3t<)cjs%S}B6t%HhqMs-HEp!xzha
z4>z6A;pSadQ8^i}p_Jyx5<K5N&2eV)0CKE}sMqrM_rCuI?|*oXaV^`;C5vuB(=Nd}
zqyd32`+f{`iw;M`_m}wH1O8Y4i~l*l_=`WwcDDhQth2S6$B@<YP*uat46R}!NP0ws
zaTr*goX{>;5|@=wct7lDx&@2n5l!3ThYcUxzfE)wYaIRdk}=Vi_XF$oC1D)+um9mc
z;<x|CZ^{}3Rh2a4p{BF)x9(|1R8&)$37Y!fQ?y%tYl#_EkF*Sn{@Wfnlv;IDwy`j~
zXp0EVr>91Dp*SGs_OnTagiNnuQ+2LEs}_J{O*E!xJ4wr{07H51Txgf2tURij2g~1D
zoG*_wui2wb0*9`@K*YJ`rPUz^^Ee2Zq;;;;Lrla_A3WU%O+YPqX!T+g1Ee4Tn{?jH
zqw<f00!-~9hdeYj6UlZZrU#2C@P16i-(=Kc%a9JY@?Eu8WGOp~FFEw1wV~_r6v$s8
zsaf#c9%GOv5~9V#5p2VseEWMuYo+#fTuPZ%>j>kBkI8mu48$E>Yw-P=hAsc@|K{K2
z=8dbYFV?iKVPwd}Un9gisk|PSNj8=MaIl4TQC$p%)oRJv@f9|M2%E-w96f%B#Kv)U
z{W^mm8Fm}`U60>wFmc5Dp51mwKMb_Xj&`}^eBE>Vy$^Z**;Bkt%@ouL5|wXNCo!pC
znzAAlETov8oe3hPU!{S$*6^rHJZV-kWPA#OHM7u+3Xv5?sHLq11c*i;3ELR~!Nl~S
zHc@3VnB<*Ih)&*<=B_@d=27ikey)W7q(^3TQzOSSlhLu&YQE7km5C+`U<C|0<!q)6
zeAM++OiB-9tp}S5x#*>t;CTW_lx(fT=0wcu5_?VgTfZ+x#?0bH1x$pfOf4q;PSwGp
z9_4?gMayy`tCs0sWvnT<Qg5urKx;~^9fR~BQ#;%ovDJi^?L~=^naT}aat~vEc72b$
zE(e%bA&L=eVDyIX{qQa9@{H~JApy%UdLYntEv;#=&JexFhDaQ?Y%edkynlype&dh%
zm0$g320zgIo+f#x>H`^LaYxt8%9sQPl<`OQ=1{Lu2!W%c1<yWvlQ-VI$4I1GEb+A1
z4xDLd-4f$_@RzK&_t<VOAX=PlXcsMu<0DLSjBSr`ZO1?SZ~p~<?|=B)c)uZx5r-j+
zBkGWdP8>GYNzJn|ujJJ-A$iTJ&=F&@iawvqoNo1+Y88rNVx`k*x)7jctV10tQyqMi
zwI|z_Daa{!HKmOCJk<t%A?uQH5f4=9DD$eAxYDH$Vg!z&K~3gXGb8qe=Nx6`vs|s?
z#};v3AL4?ZA{p)Fes@WVAmvyybk%)IEZ-#bq9spjG%-hK8pcphxEQ-i8jf;({p{S1
zQlj6d2FJ)$g-M=wtg$p|#7yZ3F_Pw0qR{L0DXM{de$P%-!Ch-iO0?a#H`Pp`-tbzz
zj8R<39(eceeIBj{mPf}dJ4e?F;){M1dJ7|_u@F3Q+%b+LL%$;$!{7h=|0zHDlV9cP
zjWfn!m<W`*Y);L%CTdWOHQnxwS%|XK=OzYj+`PdL-u@7`=p;QRwsdWvp=a2B$oc(y
zY}XGM{Eo%yjFaPAxW$6Y%{j(7j!&*}<?IHni@f^kclpCVe2!oIg`Z-(8N{T}lC(@=
zK-Rfb6tOUoj8;_VH{@LP`Yb-GqQ#+qm)DXvJ253w-CqTra}FF?Jp(W3FMG#98jwh`
zb~eNZ)f`j+wh18{Via{(lc%GG{)Eb5P|@jh{5qB@#Kn}DFr7y|Z3#-Q#*<AubDB<*
zj|ya-?0RY@HRYNr0aX8<qM8^}ULA$A%ljcCo3c2m1a7`(*??k%i#h);#z@S?@~WJm
zRAFIGJc?otF$0}j{oB?ONv4`&q*PUKX|twvd~`PpfpM_*`#hzjes=wtdd&KnNjgv(
zL))~x_1^0YW59<C@B=1DpTs!cU^k4k?cG#DMLd0DFDP)aJm%qp5Bbmj!9U>N_?_S4
z+2@~Qy&EJ>>;NIx<Bw&-Q&>pMJripIpeNqPn33@D@d>tRXuB057&;q?<0T*7dyiq)
z({KMa-v1V7*Z)3OuKj?fJ7Nf)Yd4?g`qeX@f949q<^g~9tv{vT2L5mV*Vkyx|Afze
z{&SpNIb#?-#-u`op<3Tq_S{+{RZ+8=S25k$HQ%2{eWpC8+)t%y&A9F4?4*II@U|KH
z=lcBGeO1rWTtGyMGoVIh83{FWdu*ZaaxOHtE@nEO+vn=J=0MLnnK*Oy>Gw=3)-U%g
z>0cjpJ!J>uRJk(!`;a#zt64HUCasTj-Seb%0`l2xTZ;R+Lz5CEX`^h@=EGtQxl-o*
zK+hMp{#{EAqfnf(Y2t}fBfcl*(!+u-T-#3WHx5HVNmi0S_YKKYlo!*%mE|kM;m0xG
zS77|ClyWQx6$9_yxr=d*lV!uuOMcYm;&Z(7z5gA-FL~~j{{yR&-y`@DOJp|;oSvNG
z$AR^B#UKC4xA^lv`D1?V*Z(TN@!P+}(D&KqVKEY*2$I57A;^sE_6)XDR5>}INa~Kk
zV4{cNf`=C$GH$nmdg9+>?EjCzl8bx4#+4iYK1XLiq-~ch+rVd@J>_S<@(Lf^xy}0S
zd%XJHH}Ns>AO45`5jSt$;IIDbuky>k{7Vur>LHjB6B`W4u1bS5fzkWKKSK(@EVONm
z{<ah=!rTiSGFq<bNFhfyFDQ~yn^S7L7Sp+i38KdkF_fr&3CI-7m<D|6eRE$s%Hgk>
zjWndF98VJ_TTqVz4f%+bW)jujf!g*;T%0n2U<wVWmXd{*p9RCBV<m%$C%K+J;(U?~
zJxfAN+*wQSO5Gtd33ii?vQ<g~<aS32bL;8{`B}!~Ake&knHgr@og#<}q&G=rD|47u
z7D%g?_5J9}MofY`Rq|+g-)IOv;K!KvK1S)#GDko|VqCRUs_jfbiApBs0As0MXeKDI
ztYgd!L=x&O56&-%*3hjwy43;`7X0v={~lxi3;^%_@PEto=l(TpdyfW5ap_%Tu{?qh
z_{x{Rz|wkNef2w>Tsh&Fe)(4zh7pXYlVKQ2x=@^E#@9a@G+3lL`2g$Z1BT7!f^pbl
z?Z|fXkZ!SJb#k2;Ef3%SEg)5J7;g~#PjPbmKCQpt`Kv3QJF~p}!n6FrfA=rBefuu$
zV#USz1BUGukKzCEAO9!Zy>o}({H?#v$>|BkEoe+bj5|UYu#K1_v@xpGkFQXJkQILP
zSQ{CKTD|uYJW8oSl~tTfXLUWxOpbCcR4Yg^Nl7)x!azc9#xx$A^I9hesiH3xa1*ma
zYN`=J9qX0ozE0jx>SNX*QixFvI5eMB;fbzI0V67e491kOqg?(vT{fW<sd%xELLYTK
zb?zP%o0RINDsggPBPA+V3!{}oX)~ekS4~ikUcqF2X0jP*6;4t2iq>dS`imkXNS_H2
zjatu(H4A`~l$}&)ZC(SX`r9NGR_+(Dh^x7lo`D%cQg}YQ^!ZxcPwhVQ3Ioc#l_||3
zw?1d<;o;?$v$HdT-x0%@B91SrXKsjbVCZ-3h5_qGhTWEvYmU}N!tOq1X}Edgg!B9N
zaRzMHf^m#7VnY-)(pg47X30O_OH(nViI4#az%_=pHT=pie2wpZ=UYU-;<*=IrtMak
zqZ9fuuvq;=-u>>^8Tw~AyZ&d~diI|ZF5czE=Z|>q+LEKT<=fx>CSUxqPtz`1{?R}E
z1DeLtwkvL2zs@^vzr#QMXa9`PeENAz+|eDaI6Xe4-|pCrJ#DjKxjf?9^_whK$1Im8
z83opR2QVZXoiRqbJX_0{=Ib!IY>ldqE;ZQPv-w0rYElJezbIXdk?6HKNFw7@<EXCT
zv}dlEnNF0x)R3;9x4FLAG(D{r+973f)b~-_vB}@sXD}9%qW5D_iYl<AXHj{@JfZn<
z*HgwxrRs^4g|bzEBZ@{MrP@Jd>@H~((-OP77>V6bVmZolT-rH92^>>ntlS`dzJk3V
zfrE8@?fRa%EJb2?h7y!Fzw!ahl@{fhKH>F@x+=6(hP7V`IIFKZ&c<cRl)Sn~7{`pF
z^!s%e7PU#WjpG}C@@IVT;DY1Rt8Du%<JhxXf5>y6{SSEm_5XlzSaI{^|2<t7G4U4t
zZk?odgcx}K#tGMsEbH|-*REV+*&6<{zyE*dnP;El+2@|;<m?J(XIE+6g6(cggq*8b
z@8f*U=Jk<u?Ay!xy#2%P@DrcC#l!WEciw-U5548ebDyGJp3t`6<b}`vd*Hvsv(J5l
z&VQeqM~3a%bG}#+hQQFTi3Hw#>-}Wxk1X1Tzy52#!hi8!{=azt&K+L=(OZ1=%dhb6
zdvEjp_%_SMbxy9_;?KVOU7oqx@!dcDChg*gfAx2MhhRGbQ|S-TRE`ZrO%hEV)1E}E
zjX6duSgANrt5Gs}vN<EFO1D?~H&x=yW{{#4grN4PLjjyco2vzzd{0A%?8K%OLcphc
zDgCMkTkma*$%1&Q>h7C+LI{{p&aTpJKI7GRKlW}hs{ht@2PqInuGd@&%ZcT`geepl
z2~2*=pSn3yZ(AKhrN(t<OAekg&Rm!-bFJwd<_xN4D3g1bDY3>Fu`c5_NQsy>>FXv&
zv`S+A9r=uzJXcXs-aKn_Is%X*eyxS?eechc%l?RK*KTlfazfj-jH9$rm-gU+Z++*_
zX_rSd*0NYEIJ<I$d*&84Ms7X-@3X#Kv)O!z@huFIW$OsBBMv>cmhkMg1><fF-mq#q
z-gxcLxc_j?55N08u3WuJ(=>F;BYyT5euZaW`jjLK`!H#}t+a6Hsu`KJmdo7*@87w@
z{rmT@L|(ad#MXN*FW=+*PGIK^C&#xr>V6;laAdh`8Foh;wH<BSvKxCgyAeNl9^89C
z)3gu+W8d@E>)+>dpMC*w{NsP}FW9Zuyzu<<jB(44{rFGwOMmq@`1ZH|i2v_@{-1Js
zdX2yRZ~t37^Zc{C{^q;v`eFyHO&QjTTDV!S`QXm`{OFB0Sacn~{?~q;^`_5fLXokH
z(tOq0m>Wds%1Hc?2*$b4F-f3lbB35(t{+@hYErLht`^j4U(*k&aa;nIMF?tnZq4kd
z?=h-vG^?9T7#3)#ZsjnEpba(V=Llo)AD6lpq6h6|mh3YI#HAg<SEYNiiW1dh0HChu
zKn!N078QNl<<Hiaer9T)DP2euVosP&&5&ZqwSX)JPfT4^O0idAuKTKCzF*ew0}}KW
zlK~V|_tl`&nqh`YGD#wK)dniH$ufRqB!cO%AutY?eCM0r=8ymMRmSKDAz~ZH$?=M3
zZr$Le7eB@G&ppFpb<D-a!|s9;8|k+@d_OSy1&iYqr>938w>{@}$5D5}I1CJZ%jLO;
z;hd|T=jQPdEhBv&I6gV!^5TN?hifbnlz9H&0jI}jyz;phS)HuVM$RF6VwA=lY9cnq
z5N#v>isgW);0TzMuh+0#EO_CC7q~dTWP5o@zumzwa-|dNMPodzbsVi$ERT;kK04;~
zXvKpMKH%u&1nU~M!$3bqmhFNoXO1`D`hZt|@CHBmlV9b#-+!H#UU-H+!i`(c@teQ>
z8#IfK^@Dean{$5qZ~g21;xGL?i&ZQ3tYJ$(1fr3I_b`sQX2I>Z-{kGL-{zgWAM)<)
zJFFi*;4?43!2js){A~;aF^+&}7?_02y@yfN_ePqF7^PXE!+E+%K};z|$y9<{-;EF~
zf^Ni)Qp^IOD(0B%fXQN&7@(p-vig{{o0vWalyWi}^qHFC22s_e6tYjBb;;<T!Yz_2
zC-)MRo_qCS^Es(2wtm$0l!o(KFAh}o=0%FcSaDMdw5>TQnjl;wv2yQR)4W1zQ<aeX
zw-&||TXQMD0n-4}EK!V<3o#L#Ng?|m!O-XG%5*_Q#QC6;6=_me2e3w3O1la<$#~M-
zt$0&H5abkt7#NqUCB64FN2hEqHW<GlZZC-LI=A2RSkrO&aLwKOcX{TyXE|$k_{)2o
zt<Gr9jwSTK47BEwu5*NCq$RQ)cPw1b^^<^&4XfpXqoV~E7i(g4oVb?Pe((dDw&iGX
zLUf+fD_8l#m%hknKL2?d*Rb1dX_}7FOKWtz$pMhYfR$phsMl4R{|I3d1N~;?^86wF
zcEdRKoL)KQ#*J%$;o{<)ezzlpz+%y{TrOEG7OYkacDo%9A3mUQ4rjE(^2ny&anvn%
z{^l9Orf1O^-h1aQR-NV7e(C4=!k53o@BQBI^N0WZ_gQu=cR#qx-~8AAHD37iv-In8
z-hcaoAHDG&AKt&fC%e&|J9oJK-ffPrUZIJB5ANRP!Fz9Utvljx{>{J6*MIsmjDF4C
z``~R$WDulbRYj8eY1T@+Ek(Q&J8b&Ac!9~KRkR#&rrDUha2ib?O;+2~V6S%C`kd6L
z-k2(!raB)r(%ILJH0H|Cwr##gG^OCB3S(Q9@G14Drfunz;VLqH((CDLQbrSp<G~3Q
zW+bYQa9cZ>e2m&qKy5p^v8fhL@sCX1motO}NQ>`Of+EC}P?>8}r3I@tM__17J<$X~
z#bSD&%(b>b==10EkifY#-JoTFV@6Dh{0h+|!pW5!%DhHWBshj)=(+u)AMmx$-sJ0F
zc!qc0d6#eh=ni+^HQc=M0$=}EevNOx`W+r_FL~yL&(bsz{0Qp@*nY$D@hOX=BbKXU
zj173dW5GFt-QpKpLhJFyu)Fk}9-ne?vBot^KK;sPS+Cdp_*cKm(a{msS$Ygd%VYfL
zdF{{toKL^<1+L#Xq3=h@jq@YH2YT;Q{$0;zyW{fWl7|n^SzlhV+iuwPJ2smwyJ5hO
z0TVsfPg-`{E$5dPI1EQe$K1SegX5zkw%ZN&@7*JY5n@BX+hKsSvoqSJp>0~sII`Vt
z+3vP{`ct=f>DlYN_}q)wqa$8^<x4EPj#plMhHE!(@Nj+3&8w%l82Q@Qew|x4uj4<w
z<c-%}<G=f(KjHSnH7B=T;QEc5-231Tn~ODH_`(-?{mt(Io|`A|^S}HRp1*d=MSsa_
zuivJ<`U1hVG^Qk*YyC273^8~{uU&n8cDqlUA0BDJuCM8NNKTrV6T*`-s-o0%=7qsk
zf-#?0ThP)Nh6xj^oiw@jTXnXk_nE~sRhmyZ6A9Vn@g>{O<h~`N`cY<qL)TxR@1D9w
z%ft6|&WN!Qm!e-~Tm{mK0k5DVm(mnO8mQ!8lB9{*XCgk97^a9wCTFLa_!<`zf-jw9
zW9hbnNq)N7!W~+~;%zIor8-QZX_{n**tcou_ciXO*ar&ov_@dCbuG7E{xKfhc^CKK
zEx!2D4Swz`pXT=chrIUI9d7^N8(d$V(JgOrd~(bfb~x97b*%dx-bPNk!1`iMw^)I(
zcw@Q#%=274Jm=1xI}CmcUY_++U;8>ISFW(xths*uD$B(}^5vq(4}ovL`du#WpY!bV
z&#~(bWB9J9>qPvpV;l!Y?-_=H(Myit&`YcJkgBZ)V;K4YFtqK0)uO?;kzw?7?GhV2
zyX}@Y-uMxx$0xk_sTX+uxo3Ivjn{em&K<h8<Mi|tdv=7gD|Xuf9|ODfB@TG`g<D)b
zJ>~4yr}@m6e+*-fIXgY2W5?yiL%#Hd&+?_ue~xjtWqEYWo%i43kG}B@Zr|UsGb=v-
zl`rAP4d)-cO@HSdjve&xeZW_5thlilq3^i8@qFvfz}078;$(4*-)soYhUi*BPR6Es
zLJ*9nu#;5Zsi`{mMke;!n&xUJk}=S<4wquoGU^I}G5C_66bth!ri8F3Z0BfeZhx)m
zJ|llq;@}c!%Tvqrp9KEo>{aJQk(F>&Ge6Fcy`Bz9Ia}2zpWBj5)``_#BLj2QB8eQ(
zh6BZr9<s45d%K*?`gMpgweXIUMfcHTvc7F9KZvS)QJNjrjMV7(P0>|NWpJNurv6<U
z+v^S0o`7n5sT)I9pfLTag7+q>aZH+{5UZq3d5*3<$L{DV-@Er7i#xZuer3r|f8ixo
zUCZ73m%MZLP2T$6Lv~@LYot4D4AND`3~+L?qHPvjZZ3#^;Lhz!mfeDdbDVWo`1F@w
z<jS?Hk}6|8ZP##lamjYO<%2tSxpVgc_xcgnopAQtON6fF-o5+6^pQgsa(afr+ZNMw
zv>}4AxZoLw5kGp2AL$xL<Jv?kaP8Jj+SD^|;VjLf;q3GT6XER231=rKy!`UZSnGIj
z{{iduC70U^Mn5JySVPx!Jb&vZ*RNf}SjW+oYYfAV%kzgk{NOfA>v`#==XmDEP5k9U
z-hS&1{`gP6&7XYhRr+1zr@r<xbpAfWJ9lu_@YyTJ+<5i|&hOY<T=33^AMlUge#o2s
zim!bAmpSWJoSvQ#oRqk3gU3fPnD=9kA4jQImhC@|rfsHOb!>8sRJGCf!x2nIP(GCN
z;S-dK)Gfsdg3-9AIqj;7rF1v%50JGJy-$oK=1`4_YEY*b<Qu0cX8W;xc|?_+TA5TG
zyZBMp)7hjP*rtTR>ws!BAuzJZ2D<!RLox)$iWQ4WXlYWc3E8Wm@kT`&OgXeM!jzO4
zfZTbeFfx@4HcO%a67Q3)qVK0VRTTruAqtg&AJXQej=bnp@B}Xg%WSty|0oSdHI6Lg
zjn>Um(`f0`EVEm53yxm+Jl=cmK77ENZ@$Z7yv>!P$fsYp!lz%n$=!SBymS3J&%N+j
z-oO1eeZQlz79TwQ<vGK8hp~pPYk29!PvKrRoE^2a-2$xRy*qd5J#023AKrgJz~ZA}
zdE+@=>N=^T6-_euHrRkTP=M%rtdYi@oAsI)BVE(cb`I0DjK<NMk(f|~4*}aXxDa^e
zQ!nwQul_h^$1AQJAF({T!fLgkafaYWoQ+&uJmBKuf-5(#u-j}1gC}6v?s_(tYhHTk
zvoww4_+*8%aQ@(J-o5<>7n?Oh9JqD;D&ysCZnq6F473e={`nPO`@*kqyj<Z-pmo=X
z<BkUxm)!s01OEB9-{yPo+yi^WkN@1y^8fjRf502(Bge0NfgihelhLnejHJ_~?q7n$
z+eQ+YgD@v`<VfjdH6%yEn1m62Pw1)O%NYhqjZ&M`oS7M8CJZYk;!|X$!d1_y(4X3d
zHJ4Hi=2p<LHs;mhlW|@qTuzd)5o;Zd>u_x|L2DD{XT<huE0wTTF+b{h>NqK70`;QQ
zwwJkN$VM~+Ig~m=jFS6iT{j8(3SksO9AdZ&rb;P^8p&FoD~3H?xaoV=+T^v!?`D_z
z94V>2pl*0sGbNqlr0^JIPev_?RO0MsdW)Q@WlfQEYE+*)C`9Uc?^&F#IJ)*l?!0%0
zdk^mMM?ZL*+duj)pLykFp83q@dF4wlvc3EQ=MOH4VI&Sa#-XP<UeYa&xj4V1vmO_F
zHV+=~{qOyVd*@rW!O*lzy2X;C;~Ru1(!xehG|3KPBi*9K4+9}YoJliTNTr1sB3F-&
zQ*+P(Elm;ww04Q{k+B~bLy(FX(eV90{+Imp&;A_GefAZuT)oQ4s%0DpLJar;9^C(s
z%Zqb}2Jd0hd)DU<xwyP!jEK8@c)(&<^U!#P^#v|^c5l5)ck6k6V2}9W`;l8mBcFNc
zgcqMZ<JPq$ar2Oi_uu8M-N<Dhc;mefdG)n7csN9U_}Y7%U-tasU;SnN*6;ilci;FS
zZ~WkOhHlAsUVV*Me)cN6-4^d%E(HTBl>{NU6k?E#)q1ciJER6QPHdL3s$rJUSshDP
zpToKOR_DoSI#z9aiWS`lruS3iHI8GZKby|01W!8(BTX7g1^_7;fAo7zcjtM8#Rq)2
zgwms~r)v&642(^i&CEG(ZjvgqZ+ug3KYtBUm4u*ZOgBW4#LS$XlXQmg001>rNkl<Z
z-TnH%!|5WE=bPfb&CDuJKdWda#*p*iZ0c^Kj0uG4m<+0`tNuqS_9M)+{$$MRT%0k~
zqUw@VM!l(gE~hh$65?@s<pvKQ+~;rpTmL5i&cE|N#r?rI`P$EXjW7P#7kT^5A9C&1
zHU7@u{ky_EfX&5yKD_&Y%ZKN@`No_0-6cn>j_01g$){iLQe2K<2qVtf6oDVHUQ}wV
zVHkGAVSwIqIe5ls7>1F4gwPKRyB?x6d+<2I=u>)3OerK1q2Dc*5{3{w%WlCp{>gvC
z8{hgpZr!{>cY2*$&pgBN@d?IQ*6R)4i%oI6>)Gylh5;7c6<Q4A=n2*nmdA`ugS+kk
z$JvWNgJ~RJafTOeEO_?1W#JZl`0$cHdgCFzIVN^5(w&^rSi`L=Bd>k%dkk;CgZF{I
z_OJd5uYB<<oL_kU_?v%<i2)Zq=jRU@!^r3dydRStGqpamT2k7kv!P@fXGC0RPt64L
zvom)R&25`xgyhruOa&#BiX!syX%4G#m6{prY!R(UVo+z0iuips&L)*qX*LS-GMW#V
z=-5=2Ij#O_EFp|M>Gf1hs<BRbo~eYW7Iq#Ox#ty2x@b(uhN#irLxPoRJk<1$Vl>Z8
zVrCQ4XHN}|)7QPmfK)cGIZ^^r<a~p1(@+6p$fan7*_QbkG-UyZb<3?3%{v)u6g97c
z!c&8(cK%Q)RI36>vO1(KdhhwskKW|fXP@WV%}?>}8?W(=Z@tQ&z4iv%%L}foj=1yY
zU0!_dCTGV>Hof6|vtzK9U>zZza_@e?Jn&#P!c4u`4y>~&R6tXj<S;ZQ(mBh?$rA4Z
z$0sX}TEmeA6FqI)N|5Dl%W<o=Ns}52c#c=v`_JH`h?g*W`r%XTLg3-$C1JPb;?DbA
z+_@vR1rv#kG|tgFhqHl}Evs&cYb=dP0mRN?oCg!6UKW<safcrwA3Rv|&c%YJUC^wa
z#Vi||FyMTTzr4pT7;L-b=YRFrX_|&V{=+|@TQt1+gV(s%zQM_rQ$F{F&-26gA9DR<
z#d5J=7zVK~8Ap@U*xa=@sagb8>XW*|eq-B1z3mi1)pRWMX)9qV_qEQII_T9|%P~zA
zz1K$cFr?><9%~H8M@KBW1=cybR$`s{-N1I6R6@&vGbYkp4dRmgq0EpX%(cLvF`2%f
z#%K;t-DX|Xj$FQDrrsSwP$g?R#eQVe;?<0v;+?dxmM(Tt6+<WhRSoxf8D>nU{#)WV
zObS3ko^7od;wz>q1Yq)WRpraZ?5MV`sc|xBYwZ*#Wz^gzBh9H3>cugnq-)f^Q)*zP
ztfvW~jeZ0hX~&)+EZERutfb3)>8n4@pTGGAzxr4II{(Rke2361*({EEFv6XOm%R1+
z|AJrtxgY1Ze&y>tbM=I-bHw0jn}+qpCA)sh(eVkJ?T(0}?|1ldkdnfw)9}*4Fnani
z2*ZgG`W+XW9U-jon~{fup{JoAJbeg&Bia>X5CrAg1|I^>NfS%!Bv{uthc%th2nM^j
zPF%*kHWPy{7Dhp$#u~yd61Pd6^g-}+a@H6#V!bD@BeaXu(9{t7`<RFKv8`qB64w*M
z$l1v=oL#-f>S#sVIR4si{tX^%FZtf9f671nFaC3`T))X~dCbu*$BQ?wGi*0BbS%0>
zV&DTIfC~=SG^szNl($xUk8?@5E2F8I8SDw(xpz*_-o&I(pELcR7=Gosw%C8lJu7b;
zKqqHn^bx<^O3K?-pP6i=J}{zrhSIknd+}6Du_k-VH3wCnIag<MH@Cf6Jo<X-U`}i6
z_S)0H#ca^7l4Gqgct7HOS>%kaD*aU_0Tnl%ac`NJdeKn5D1Ubky%}X>qN;fvtC>nB
zxyt6$-DUOOYFEpKe^cg5YC|Leds7usq{*S^T2mNjL~KYg<W}5_BdvvT8@O*;ezd;i
z!Q~FKbUc6K250AiYpafTFE07)7e3GEJ#T;bAqygL+|sy?KmGF`ap&%boL;@j$?2&S
z!7~vP0@jaMYw7zPATe6~c9-kRkKSWigAWE{Td)q>G`J@H?K(`;(pgXIJQ#<yEyg%P
z+-7Mng3MysOw}G!ewn=9N)u$}UGH($i3>rd^-ne+jO0MZgj)JtI=4oQ|G{CLs3b0Q
z*wq=^>!-YTeuweGaO1{JUTm&%>Ki`)V_)VEU-}c?yL$&i;Kx7z3ZFVY#f<~RHX(mi
z_q6G!GLeW+l@fBYxFRSmGadJ}f-<dQKriRwFf%)>8m6UUnv`y4JrmQ4PR_5(jaKw?
zl0svQAw<=LO3}+Qwkp6As^H!VB_>p9t(CIYPk23LoAp+5r1FAH51+zaA4=V)ykZoD
zbO`w<{GBYIBAN+4jFWrR&#o6+VRNk_T5EAlF@EcJWFzVgS4pyeO$bAYLR)u5i-{^p
zOkq;7sE@+T<hvoQq5QL`cRHqV95|1TZ@vB=Z+`O|#QX0t1jDP(y~NM_`fqTvxyI3I
z!{<NuY0e+qXSuLExc4FE_*s^VC0DOp<>=^0Y#&WSpYA;j1AQOZ?$%Nfq2DqNgSgLg
zgVczNMKBo?b7x$gpR+c{{#a5;oEVC?T2Zkv;@PQ$m{z7xCX&(d9F@=ZIT}f7m*nQ8
zb$w?v9B<rqiijS)6TL?)RwqiZ%c^S;oz?4KOQMJ9CCU=3_r>Zp(R=Ug5`;*URU^Sh
z)V!YOeZRhQeZ9||Gjq<&+;jb|B4?Vz^pP<sL2Qa#w0b~_tw!P{sjYVs7Hi&lI}@6G
z<<;3YGyptW26kT`^xQFFe|c;1zl3}K`hM-YVK>p!l{NFb(=siLrT$)bVjgeh#Tf!#
zBIG%Q5r)xuI!_=~wbMHHVmd<?e8oa;Qi#<?URmXA|DrlQ**TSl+i|9!R^3z8VReAS
zg&Ryq0cIDo^{IaVH0(Gsy93e=1l@93n(g>No(8pSbR}<hf|01SNIQ5uS_z<T0FC}F
z5UKu=V!w?DVBa6vpL;eU8{@`9DJr3iMZhxZ4HGKeo0&-*pgZRZ_UXw0I}yW@vCCjp
zQVAepYW_#KYq=^@Lpb9DrMC@E0TTnKz)#ly*hh|fEqmZ`13mwwysp0=FNEGyA$Pr{
zsBb<{%#69Jtmf4_vn_sbUpx<wZ6d!B<Kg~Gq-MhyxYvQx9BVj}@<)IV+=em!U*ICe
zcR3C{6U-0aH?LDK77iZ7Ulu*CY@D+3IrMJz(4DL6uLd=H7q<?kyBa*cG_yWbD4kb`
zB7_#!9R2dNnTtJ|%L4|#J*LoHB|o<M>OgisENnpfU)4Y9iA}n-shqE-&&2pgMA|>r
zx_Ohj^!Zm@3=guK2P~d+roE;_a|Hlt3~^CzEWDTVhf=2pA$`>*9-4U)?IA3lOfD5a
zY3}k$Km8uw53c-AF9>m*BQBH3Q9(%;uG92<f9{3Un)_=-I>ZXz+yWXQ^3?zB0a({8
z7bmsv;!2m8=YDc{Ty0krc-=gi;F+k~NpUW{xHUj6<R`7}qK*|HyslJfY^B@w=Ecvv
z<SrslX&NtwEbB5enBv60KGiJ9N=x<`Rv=WhN!SdI&21$gw*Icwu`?2O;(I;u;xFN5
z;KarDXdHvR-@kRmewju;@+wbohjVFTg!kuk=|xHMO&Jg(VtA9D{>3Gec+$`!2a@2a
z>IZA>o3G*NR5l<_SH#2@NkKwQ;EEY`G|0SYZt}g|&u8h81yd9ZbcAh6OrT5`gn}!b
zoRm}&?=l0qP05gqHjcGV=HpSkqjaOJ8Fld{F6@N=;`U9?#6tzIP#n<n`UeyL4QN#g
zxu~)_FGVW1WXGbp+UjR1{d+cbSN>|y4qIQGU%u!*AI5^HN~k~!*Oe<)%s_c1am1IG
z=S8Qc1uE()VAjm0dt0jhSlJ0lFWwOh<LuY|U@JKF^x{sZDSl(`+;RbBW*#%asknca
z%)Q-JURMR#zvo|Xz!3R>CF3#2+&8f#_3VSu<oC6BV4Xv)Tre9AwJHeCMH6`W>(CRp
zr9>-RWu0?AoDgOEw6Vb6$k@otA&n19?vunzNIHx&)~AZCEhC=&40;n7we4Cs^lSgg
zH+%oUyrw3%zuhBenP&&uACf=p-f0WfV-C!imo|*(66~C)Q36yOcR5B|?I+AjJoDln
zX6P5C-je;NlZm^Q4di9oq%Y{XYjQzwNG)mfvkc)!^1MK+CvQXRR_X+b1lxD)H7d^Q
zn2d?(C5K~|7C~`x3(W+$*s5q)*rdcvckkU_4Ha&)cmGf><b`;(^k13zTH|VC%{vaQ
z^~jmIMMAchIs)xd?7!0j50tW>!A5^{wvv!cTjdE%hL5MXYCJh~Vet*Za<)G!QZ@5$
zN*5HGtR><AAE8M^hV6Df%;5rmX9d_vPUxK24V;NMiKP+%)+H#OLKOvbI_U!j7a{cO
zS?Sct-5)h9{1cW<a65)S3}#=ipX+oz_5~XJ3)!{uf3+akoVPNtleyl*sJ39iKYVfm
zT}AHS>(XD7C$6R~d+zj{;MbT;!2Gk%-bZ{-*cA#pzJKmEvojv~+*CDqlzkfWWVihT
z0{3F*Zrh_dG9qIl6dk|vg+aAyu|z6gH&QokBRs;!(s455WqIVsEuhw9v7_j#{{#bL
zE3BI<g5q~Dj%jxjd))6OXU?t+q}wgHf-i@#+7{xXr$Pd#{~hiaFgdX9iyxZt?zHmt
zm;TjP|9u?KeYpK-UA04@AE>=@XK*sB?6ukMb$cBp#gXMqiJiG_nPg1AU$G(XXgE|N
z6@x*S<iF3}d6EU%!4+}TVRWaB`jx4i{gaD<>GctE&qQM^m2Eo#AG@lZA;Yz0apR`>
zTY=-nZ&KVovhdWR`1EVUE2ug8PC+VqfV72I_Vo6i=3c5b<5)!Y<DRLc<N%QiFBpT6
zV|pySI}Qk6?s*dA{fzI(=L4*AHXB{w@0uhoj37JW<7z3X9(rQ6GNqXS);R0_HurX>
zpK;prt9{pjtYU{CF!lA5FI`vM#%mEy)&b$dOP=)>YzzJ#fP`Vpi)S%mDNHGi{IEBt
zUhyRQxg==EN^i+Z`*=PzgN2=QsLyPRh^#sO&50`_?m@<e)}e!L06Pjvx#a<~rM^la
zeijXSkw~`LJNqQE4<0fz<T&*k@22o~<G$uk#s$x}JAaw^wG)n{oB5sYER&69HKrnF
z*~8l$I>1R`VCH7Gp0qwvE5sRJ4$g}?xqP7kaw50o!$^3rQGN*o@6kCJjWtM=IKfeg
zk1UtYmeagI#gEq-L-37`Mu7MM&`jT7?|}0`=L%W5I9kWYdDpt*i63Elu|iDi3%TO(
zxCI^gLF9h>H~gU6@sL&76yw6_SWZcbxxqlN%2QtQg96!RDGmLYN_}jnpT<!a<}Uq@
znc_9zM^682=ToL(-qp|v)bhRXPsgF7#q$0B8%yUhq}<O#wuLP^E<_Tx(@-uAwN(Bk
zTIdfAG9-xxd>)7xZ$rvz!j~sqa@``}ApGoZu6R@J(&W@1Wtax^OQFH4A%w!ia-vhy
zp>98<a>g@siU*(NqYM{`uwx|r?RuGd<T>)k=NXJ)y<_0$({tH?+koiJ=xwU{wv9mF
z!6X*x&)zqD6Iit2F-Y>MI3g`~K4mZ9N!enB>@e~BG2fEHw6^6E)RAn+;(1+<u$$q#
zcZEF9C^mfJHb5er&jqqSx>gN^POw|<Uh1mcM>Np?=5l&o8kC_StQV@tmZpkD#LG_2
zCfrJ3<{5uGyVS#tuY}Z`+C7Z<?}V8;Imgl_W#ZT68kVI)&=$du{clk=s~f`2_TxnY
zzDR22n>V62%y1=70O&<`p>N{8r4e1J*+3zTR8Xj4t`(1eJsL7(>+m<*J5fmt8BU?S
z$wJ7YPl?Or{@Z@}(%%_<BHg}opYL@uWAMt7ohd#pwW6uGkQw&y(fsax>^UUrUE(4+
zB37I8mqpe8w5tu++ZVCh(#48rzMKE_6VWo?o}e{FHm1g$wpMe>CZ#F{czhDg7_c`{
zFrYE9<CEZK=I-{n)wIFFZ_t=)Wul8bBAeE7(XTSd`FyurW2c2)-)JWtW3?RH2*~Mg
zu7UEQKF{&Ibeh05i5crvRZ2P`gx$uox%a^FNqC6*)R2c(r%l&R6VZ8Ke)cu$9!G)p
zLFu2i<d-ytMLo%aeWtu9qVECOTp*F8V2!MFl&FjLDcNquES!fC_SB-a><c9K%9c3W
z$>?AqW}|@;84<>vp@s-Y4(j`wd;BR|DB&qk-HJd`hew`R{{ETKe7}niZQl~nvh`35
zfPa+`&Bt_VxlpgKvDiB=**40qTz^Gr9#^MP)fhAXPP}S-OvfT!ww}q7S2ts>emQ8E
zHBOhQQ%26O8X!=_|6^9!tVTDrncu<8M@Bi#1+(Ef{soDG3~_Xw<@z?a>cCh@JSqz5
zXc+)YwIuJg57J>LT2a6(D^|wLjxEfe3g%aseFg}P>-nEy{dPWn$DC$t=R2x0P4X+S
zlq_ty8XeEukPpiEm^KZLjg;9DYB$<6PS2z5ISDBf?46`s^E(DpOqWWb=Dz@VRx7d-
z1k7)EmKK1IYj_2rs+b|NIAOZYU!en2E73-eO|yk;O%*A|OO2!$sCi31cX|IuL?PT~
zY9VoxsKPjjVTU2V5O4;**qfCHit1*Wq!wOnqW@XCn_;iA+NubipFVr}X#PP*h3s}4
zA6wzz>wPumi5~t~88+&i$d~$utfbkagSU}nTDRo0Ddu;daeBh^xZ%UNXYfz5tpbL~
zZTE{<gTNnHq0iH+0%!=Y-`zZ4$tfvak19s5BE_r0m#II}3J_g5xll$i!x;n6ePM1H
zW0adH<+H|DDx!7US<HQMomZ9{r*CwYc4~$j*L4ZB^4XmYFoxAT*01OYHVGY}6mm$F
z&o~!A7l~7n<)d@Rz4NDU6!Y54^Tg1srI&>oHBm;|1<o{(0_tX8oezMxHsiMrcfp9@
zq)mIcvl*HPe8z#=uD=W->FP@dN@%^MqtX5c_2(b9bE_=#V^b>1<cLIi*;TCQaT7>>
z3c;fD0nteE@kg@AuZ>k=QQ4e=zZW6>ZG2U4@G|9<+E7&X67(ei06cFNs;R=~iSLv~
zi_s~!*e`m!`uGQym~}Q<^ZHu%(9FGMo${B|&F5YIS9w1+IWCosuKWix*D}^H`;W!a
z7eiQBQok5}4Cp#3HM}*KbjkKb*zZ3g?7;yIPfcslJ-CtHFEOeO|FTpcugpwQd1+*E
zsN;%C$hJ2ll}!uEW_~5;m>=GoP5gc?Wb}OUf%Qx^33wDDq!_gR=je`Xom|t7-eg4F
zlyp**YD5;fI??HK_bWWKczoEE8J(I!KIZeq$VrDxeJ+cgU^G%ZuYL@|S3(3L{O0$M
zAO?#n5*i_>y;7a^(VZ%_x=i02^dYKhpWY752+@rjFnX*;nze-?IJp`-5lh9;3^F(Y
zgAxl<va@#Aca<ieTUwVS2^h5Kx$zBQ@QW!A?fgPfO4mX%D6|d{fV;a{Ex_xHJ)Ugq
zp5Uets;HTY5$clGuHDGz!T(dZuRKm8{ddx9)JG;;YKJGnGd3yX*{1)6N_`VafYA9s
z6VF;)>~5joH1pjWnafuqywg{%?%N5q?x_7njVnrSk5HEI>UV}}AyDGOGgNiHcioNs
zmiLPb+f50DZzp&PpEa)2;N>*x%ed6!bdg&eXd6Zs;&aO1o3-Uqxf9A%m=V9t5mOHI
z?jMR(w2+A;g<?g(sk?1xmMjj|*xv@Hz4)$0ru(is+dd`zdz82W6QUxQ#8!BRT9CMx
zk6tk_S_A(}8yLxhH`I<%upM=c(&cpY7ex6~i<d|I*b0)$JKqI{mQCtOH3z?r02RLs
z(%fb+T##ISre&c$m22lnwq8C?>+&8(%`6<ATC_r1;#L|9DADJhJG%H8W$FuGJez(Q
zk_?c5yFF{2)5}tTIJBBoVwr`p&)qtz0I!@i1Tw_jnw{+n+%iBOhAX4o&s~^JHkpeR
zIf(qP<JSFpTFmR`wF%re(=;u5n1`Av%b*RgKpsB%OkE4?SFOUb7STe*hAT<ykkA@A
zobxf57t)rg=kQ^>HoN2%TbJuNr!lX6@k$$QNb0OZs?nXp^>6$<B7=k~W>VrV{nBiW
zWu(DyD1M*wVM^&l>k(hLRaS>dj5gxMeS`B@S+AO1_NFOsXh(9(P(l%z1e`*rPo?;r
zi+53Nq@c`6l9!XYLeU~nkW1j^53NI1FpmybEv4FcT$(o{?JY`%PVFQR(AGnpv7<TU
zOvL9u5|y}{2qQM7XySsvY9ch^<Q)v5xcAdDJ6pRl0hV_{5rV8S2OwvcQVCVj!<;zj
zY-%26VaX~%b(%|17z2x=!RcW&${??fn3n>KEs*;pB2mVK9%EMCXUja74<FX7a#?Yy
zsJ^lo0#>>~{#dA0ZZXlU)a)mO@2Iu^xbHbH;h2!yBH%zTwij6~o7k=k%!ZZ<xn_cE
zHZbm+r+?^S(BayHn!eGP{{PMbL>Ii+h@U^Yq?HZ)2YBSW$Vsp#i6Xx<U<n?xGW9FR
z-OFF&ewrwB#b3q9w|x^kSi}E35*@2d0oS!BYj528qbyr1>EJZ+U8gS=jCQ<AyWMKo
z*N7IGqb-RRAT%i++hkDXeo<b#!$j2)!;uY&zSj_OAu*N)Y+LRV%QiGrG%+9scTpMb
zQ`68nE;~DRKhp?od5RW8W{!aMrtXQRtAyu;ec+MYS~sNPWFui|il>F@$-|q4l$XDm
zMNLieVSGFa+At=TD%>cs@Jo&<_~GCm@lR_Rrs5ttW$O(JQa4sD8vdSnBqT2M1}Pc-
zuF?ATNx3;BU?J`tv3E>qX?JhS(m0&hjZL{*sk4z14-Sm1d1&IStjabYSVj2+D!--B
zCO3#o{0G}r<DwIqDxp_?YW~w|QbR7>i+aSUlo?h{4!PPHD|8^HnVBP$Zo7_wktPC=
z8VMU?rK+_2Nz~TAz$-ivO-U0P`mYq$M5CtpSk8G0-FGs?7X;pzZ~wv)9y9|h?o(p?
z*_zIN*(nRdrAuXyJkPSwtayn(6bR~iv^@<wmn~iK4`YgiEmek&KIFVP*E+&-R5#Uu
z{)x9&Z5h0&ySCYjfZXH?N||~8nAoh5uZXze+ZW0EYTj0mht~XwvNc%lJ%Db~5Hv_1
zHYOf978GEucEQs73V}1|QCdrGB&P8%4^99LwqX)qT9fyG&GR)S=LTHiOB0(ulY~so
z1t*S)N;>UIyo7#xbDOV`qW6M@=JE{6$prhH?i9tS9h)k6foyZ6hkE&TJq`?UcLA7c
zZyH|lxit+EVwK$I)e!e8jN~lIJkoYsLK^{Jh^C2I-@qPk2BxaD_b5%E7%MEGqsYTF
z7!Q4k7?B$7@vCQSqZLa3@-vf>G%eR`&?)bh0Ox>|$c5?&w59%AYhwY<UL)bITz957
z$X4clXNalouah@&vngaB37rh9O6?1T4xC0?zQY2AqRSbpTC6t+mLZD~*OQE@T7;D;
zI;y#x1l7fZrDQB{pW*Og8wkwWxt}%R((v0-cAKTgzF|QHd+TC!d7|}ECTFC@mEdum
zS+<`CA8650Zo$9$W5b<)B_&ZxgoAe-^}E6h26?OwOZ&}$m5W3hCDgJ-?M<e>tx?it
zR59!6N1kxoKUE!#fEm!D8(rA&haUO-941$=BTb^Ay;3K=y<NDPVZ!!Z+=K`<7cEyE
z!ja0{bN>hPX@tYB8w+?IlCYEd9;eoO725dgm9!_T#_BJgL0)rKA2wfPF)as+AGuXW
zF~eIok_MGr_e%xhfI!N0eQkbLH%s+<KrQZX+$A4K;@#5y5>x%!7Xq~kZS+yvvH~kE
zBRnVk1bg&~PH1I#DWKfqJ`Iez-J)vrLFifH1qbf~leioOUd~FLz*)Et4<<&Anu#V}
zJdV<?n#o7CH{jX@n8X`ZhWkBYbVS90q>Yr1Th2y@G3f}-$6n@x@8=J_cDc7V<)9y<
z<xL&SECQ20PfZjG9s279K!JDVRnvv4@JdbzD%6D<PIciq`ul0ZilyfVD4zcMb0+9i
z3V{v_Xggx-KB21kPL4jZvUn3*@1m-A8M}Ko2q!qXc9YKpXb83J28k2d{Sl2<mcU&9
zfU6p|l=2Rz3z*wDDHX<U{mL+!O336j`#!k%&bUrQM{RrFesDum#WD_Tum&x7;%KNv
z&>NXH&P-k!1N&Cw(-ti<hIO!FaV~EV!WLD`j+7`hJw9-H>>m2mz@)I^nXB-uzd=VH
z9Z(?>92HBP{z*N}fd?_vM4d-ZoO<3&)6Df5G`gQhLXpXsx?+p@`3NSFqy>v5aLKJC
z1L>*>X~`tJX}j5#6TXRho@~e<9s=71mxUxL)BYF@5`UaIt40J<w)58u=??DUSQX)h
zgC(>0yH^Kg5&<bcZjCqueIJXzuc6X2b~*jh^gyEM(}j#fM-&)wEZPG+Rb{f%DM6D|
z-IJzgULf+B76?EeVMalZ-WP=MMV5LPJ=h1=mb(v%xg973^?a0+rE1|4fb;x);`ZnS
zb-dMYzo_-&eM=T8OYe|etRZ(K{#Rx%1g=%r*!iE(MPfWzWudGC)6fUUnvbBh^DmRQ
z%HzG3H(~;t)z(#UKW4ikdtn*iXEnusYO;gBVjXmlT~Adx0{tJ9;w9es+sT72EQX!y
z-8?!3MgJ3?3(C?K5~shFo_VFchI&6UDQRR`<0j})^UMk=@q~(KOQsTJ$d(22Su=wl
zm^SuL!tyw`CKnnPefA}eDu4OtczbRIITA7m{shiSCes$G34RiBtF<zwRo~6~ai&x&
z8YI5<mDY&m<krUT8wvGeI#-7<5*m8ccgbJ$3Fx^=7w{5S)gqHtg0wzNm3^voo@A3t
zwtAFNyZS`>Me9x>)KeH3P0?j-E@78T_x_Vq%_|j2|CG{DiA-H}1|427pk5QI@E8Dg
z=u)Amt@3m?RL-qS(fE*C_c>^J!zEIhmaF2xbSepXdCuAFx9hL2clqi1%eDEE`CaTi
z7R*ihk*5$n)ve0nDv(5c!`Qr?Mr|I;9e8hkFGJ>2<Su62sO*=4Q7I2+RegMb4GR1?
z-uX26RDl*a>0JGl45VYDhU6sVpymm0G5dDmRQmN7?S&H%hD%A^zwO>H^A2c7ZFfah
zpUD%7Nj>T_e882V1iAbmxVuNWlD{%E|HX(Kbd@3!7N8(xbMn}yzM7zY#}h=%J5`W3
zL*1^#YhP})xYA@=-tAysEf{M!Vl>OB7km?yMf=V~FqP2CJ`19OTrdiZds$cb`XUfG
zh9ukcnd*{Q>E$R58eb=l0@n8JIo&;V+wXL6RJ@J3T!Q|&OzQgFeJ+z@oO~~V)2za3
zkGyESu^ufu%fH+C<X+=@Y%NhbY+<4>L_l*ErJmx?Go9Gt_^QhkrdDBme!5uKd7|J`
z;gBHetCCV_S&T}ycWj;^MVyYnK&gV_s-%l0^-6vb1=E{BIW^*by48Z(fP6&5kW1R8
zy-<HVZ={{>fd6k*=v?V-2gQ2LN3UT83~dUl#NNAUmjPwZ*Bps}q6zRgQZT#>NlOTz
z3zqWE^zmZL4A#qi;9@CQ>3BamI^T6En>C(><h=>4aAC1b4k}ljsikv=f6|&DImbj!
zj5nsvBR?0f;BN*Aq2n&4PAtwlxk$^iR3hEJhqk=qMGS;pQGTORlqFNL%D%vfdv6Wk
zuIYqPAbzy*3Qvf&lgSB+b~%0Eg)Ibo8OxlD)JjZSQC1=Ccel4le&6X3EKbj<o3mHS
z=bEVULtery=rH)8MRAa$DLX&M5%W?^QKa#F2wV4CX8FQ%aED(?)I141cOoD_xfhi=
zQtUz=z|NI-G?`$Z$a%-!?D^@P1XqROZWeApu|hUTMS3qSbi%ZzlDN8zk0`cCOowFw
zhecOI`0|)w3tUVA#_1_-`dN!)7mNbn&Zs#7l6c~VA!q6|KGGmDnN|Cv>6prqj0Z9a
zh2xtB?SgLT1CnUUVdAdTA7>UnNbh<AfNAr6yT{DbBMP&Pwv88;H_b(RzVjcQg}(3T
zhp<0POKP3wjfkgVB)5NW`I7hh**o@e?rI!=ObtfAZmq6m&G`eGL~Zro+ys)EtC0Mu
z#C+9HgBMj+32H&MlegcqDb4V07p4?z<|7Y4Aa&J`s!iDMM9dl+<$gA7!*MzhlA?lu
z;+dAX0H#C_ADTsd(s&|>(L<OM%4#y3Z^x~vTxl6CGIJ<@8Gt``Ff;b@W{zby&5Mx=
zt}u8$EY^Yl?r-Adn1AdivZsd8rIA$=%n&rjIdj?A)jCx;>91>FnekZO=9P?A0c)~O
zi<+z{Jw;_Qnwi-yL@+7_!Rh@U?ruTZn;G5Gu#k%IueJQspG$RDWaS&S@5%NXS4psQ
z)wiXaZ2nbHc2}(q{Lkq;kHCGJGSVvTAr{hYvYNfq7Xy6gd{VTt(WPGer8CtbM0pHN
zFoW&Cx8tpfxH2q7w6f)JmuvDd%%qG&Qc4uB{9A7>&#CrUkbDUn+uB<EHV)$F1Bz8W
zPm$+Oc&85*B{SC{5*j*iV;7TQcndsYB?-<A8s)1{z)EVyM2scMaT##)(v@mlF8Jq_
zyr45K<uz=Ujx&o@U#M_Ws1p!p$i@1UyzB1fcTvaPl|HSWX8+_o9CIsc4<8)T9e7)s
zrh4Pf8N%XZ|C?$F%3Y~e`>mS3jBYMae%GElQv9xUE1`(eyTWGcx_9>m$5LE3Gh)L*
z_eogn8`7l%cbQ#+xIjUOjN{yfJ5P`_a`WY^m$meVaE;khw@_1i5jqu<dC$P)&m)~>
zb`@jW()%+@uA|poYNxpp*J`nb9&}a{!_*ByzLE%=7l_zSF~)g0byz=jYI)%XI&7`T
zUp|rp_ni~}hGNddsv{o2gr)YI;1`;`1sc#f)d`B{iRFr^J@@r0PK{yDK99UFYsp{6
zg12F~FfI8j#uC=W9ut`xYH@DPWTdE{)c*WbxUVwVtWIs|-_!N3&ziX{%nXKyAJCS|
z*k=;LWnll}ZGxXFUm6(~)T^E&@|iRWc+&1q^=K7j1<mdUu(t2d==cmc#`yQ&gYCo6
zsgo~MiT>`|ay;(P5`-aedD4ntmApE}TX?H~T+1k~1q!C0+uaGQS!3VKElC1Vi>CMz
zc*G)RE;?hF&k=_agUgEQM6xvAD^L+#sL|+$MCJO@fIUtzyN#MZy4sB8KRy&5kC-at
z(4AJF*J_vr%61e;Az-acN7#KY9fk`h$C+z^?6H|D)lIsVo%huyBFCDT*UYpQD3Uhb
z5P^kLvY{!1k9w8b<GVvRH$+Hv;pa9=5jSWgIB|A^N8FN)6YMR@ter&-&~>c}!{8b0
z41wPlNsAuJMgE86%u)!rF(*EJH&|5`X?1GxTru6OR`s|tYYA^kQB)**nlrFxmZl(8
z`r^o}^raGu(4!$9a}7rx@3daoSI1)>I(u(WF~W;$4hw<IB&SOk)0CQuODFAFN>F04
zbT!_#w+o@KSqo!`nKdX3e<jOFexwXRe`0TB<6SHwY-q+9+FOt#3~hhZ`ze850H|pi
ztEoUJ&8{#G*xr|(auTzkzYPoi=$|=zAGY=uVab4V6lO-z@SS4=w4CKB<0j8f_ow!l
znfM0AYhRCw=AUi5n`C8rNMq(qA{%L0YrF*A?49K7?t-Dt|NhP>jTg$Al<}-_l@U$S
ztMSX8S07cnp8VW;)kfffKa`6G9k~Nf<?KI=&dB&tKJ^Bj9;1K1pXAgXW`e`0RC^d6
z=|M({3Ch#PHhj$w(c6j({@mE({6s)z$s7>dB0R&Y=1tFd^g(s`DQs2!Ez12KT!M@8
zze3mZuhCowl<>*8N#b}X0%r)7Qnuw&wZ^Dzg{E}5ERcb9&T-ythf>9^uF>3ghhTeG
zaZE4I4w@xk#!<*CWa^1u*JIIS8{N8O;|{kXyUFeiMZYN~ND*n;n^tEwbUZj}W4}|4
z7z=WHE?MX-B%7U<((K4i2<KD3|DyG)B7Cx$Y|NdCUv%-TJ+OZNd}nZ8E-utqFSl;P
zm%Jq*CtzIpL<0Wx32*G9*NdBm!=*0Lfni9`YlY6D{1|pj_Psx`yzxb1H1$od91wsq
z89E8P3`zeyy~#cgrF8J0`Fm!T^X&f_ZxzVA5Rt;Us^=wq`y}fE{xAu0I>rr@hOuyq
zDFHf1(>XCgW}7c{uI7+H;Gav(jfOZ?e8VDBj0?o-p9{%24|xqy?{fEE=nT$qVaH;s
zs&0kFJaZ*a0xi3Sok3f#nlHqw9CBy2JKWf!D!x_avEXkVX>(bD0Qzvc5+oT$H)D#&
zU=M8M9-!yUSn?sKQm~vIdIb2IP!u?c$E<hBa?d}?@;x=P+dbbu#^hpR-X#}(ktGu>
za6>OFndpEJTjQfIsTp__c?o%|vnyO@wu6-dpa_0(3EokudeCxniXjI?^>F}A*Mp?b
z(xN)NLNTVDw!7_`2AmRh@o1xX%qcv6szINo4pDC>IXxlEw7|N(2r-LbJ-TI{6kWr`
z*(x$vAjrk|IEy47x0%r0&wn6RL$g)3Uq8{;>;H;TzTw&8pI&i7edY24=HdST?9^#m
z9K~|Jtl!cyV*<ulTNt)WkP8-TLgylVoe{=b9Oa>o?4uA){t!d-d95lWkD_Teu<OsT
z;JH9VXC%$xBgTUOuytaCGaq1>+H6x%Y8d=?=K7@XuASj7j7Xj_lV_3GUa6kR7YLri
zaMFI{WqH6`(<0{U^&%d;=ZkX@YQuo+M0_8hN8XIG8)<J`vt@|w?-iL65Zq_1PKxB{
z$uM;V)vdf`O4gVIi^3V<Z3VBXz@Nb5^qh>%ToHs~JIw?~zlwKq8GYeVl%;Y%WCH(#
z7I_oN5*}+$X6bLy7Qx)iN|>b8{O!_{EC0T1kQm~rf#YT2klQ?d3Z|mU{c4gXbi0i+
zb2%ySK4asXXuF@_Njnrhix!D7M78qsGWMq(S|lRn4>zni*XR%EfUdk*Nm01{C-b)T
zbV_J9Dx?4N_4)x|8|%_tQWK@z9y~8(0n?DLQ>TL!u>5BK6}uE`#xUg!_WFantnj`1
z4cp1}#>)C;AN)K0=8J-c_1ymCeE#)XJ|_SEqoS$Q)gtmGz!ZKnup05V!nF8Momb*Q
z3{o}~ocpJN<#me6;qvO)&2kfMC6D>NN@|z39>4rWweMY9f08Pzjya8r!#Y^~va}k%
z{OGHhTL@0_?W$g0uef@p?}YEp`e{ioo^h|rg_A#=TaY_qI<m-~7O<+<@@e!2L76>O
z<SAxI+93<4Cp4Z>+%_WgiTmMf5#Ri+{M$wM(|Dag3&#Nqa|4w!(SL2kTw&|<zE{_W
zkY3fHW*n~0s!q*!S<N!}{)4+64;^Bj-sp1nLS?ban6#z^@!snBpKJX^0@*OH(BATw
z@<6FYQlr%*3Y9B0GGm;XH50keZ=sVSF-BlKgP{6t9_R1-#1fi$<2+}ImJeIxGmb~x
z`DW5Y!ZgW8qisQB1Mfa|1BYi@kJwkR6G6#(F681$jrraJN@(_~z>brAW?Ku_H<K}z
zS;MSzySERwDEf)NV7dq--sgW@0LS~RuQGC<zBO|A(*o|p+4%u-NX>93`}+s3|9uRL
zU$uH#IK$dCI1v?D?SmhG7-3O2QuS6sL$;)(?bB6?6yoYp_RTqGz%X=~E0e&6DAja1
z&FJq^qW)1tAX|K1)y43Fm0rzHHao@^l|zHvbIJ$vGxlSaY?Ep+ZH~b)KT8U-l08I|
zTnYSl*~MC3g|o+z+zj5}o(zbo7;Im?KIy#BV6!*nW6I}b;yHRbG^dG6lflY>frD_N
zaB+_!PhYK33o(ovDP`W66wl<@yT437DsWsG#uHX%CgbBfE4YTqJ~8#yb}y+ZIpJPD
zm<2d2vxx3SlwkI-FEDs`lB~tH=a-p>NbQSqG95-6h$0nvsd#T&qTjlF^P$RzfwLxG
z=t5uaD>l`1e2-h783B>AUw3eD;BmF#$QCl2`1C~fm2Bm90d>!f?V!<WHclaYKOjbb
z<Llh%=l|(hMwL_mOv|v{?WH~H-$K);oJ0lVyUi$+lhEu@@6Plk%q!Y3Su_JhL3_t2
z7gRz{1e6mHzf}gUWV0jw0F`twz^IC=uG*c0G`vfsjm%*h48&zBQXg|%^BhD0nfJ#2
zQ{+?g7I;CA>T57?XIzwPMk^;?3B@eP?J+TpRgm~D0fiHVOY5Upiwm#ZwUJ<hrEvw&
z<26)ukb``H4*rk@N_I&+M_F3+VI_6n%n8qFKn>Va&JlVm`7a(26g!sKP3`&eR};NU
z08IvXu%Av}PSjwBk(-la(kM`ZJUq+o{^h5YM3N9b<|kz`HsJV70Bl9{j`aS-)o)?I
z^DJ#qEIA&v?++nxh47s>Za}bD*)MOu%HBm=VL(Dbo{3qPft)%Pw&lYw4?CD~h8^pL
zvT!-cImI)mMES$fT?5Fzz5*t<Vd^-8iOb<VtwSZL)VHTTrzBYi%|))E#u>p?(qvg3
zSt59@h2NM>Xhi;EYFbroV`Y-0>kW=lVyVj|n-I4msk>}lu#XWg!OB+I`wu(LP^64(
zy?07vPBTPWthO$*e}sLip6A7DxuC4Vpw80=0(O&I`UK{s90{rtQB1r$sRVkus>90i
zW!x|T9cN}<{(Kzj#Z*?<imj~__RWrvRS4moEMo7i4@HDHa6Vh2ZG?b~2ix^}+@A0w
zc9@#4Sh-}8Yv-Nvapg?qfV3;HVe(w;M9xpqk~^AK?>l@jh6-YUf2k#Bt2O4y@m}b$
z^5cr0gTHCAO-Qw`3X!gAuqa9lm-c>^5{_~yU(Mv^s|A~U>EiT=(a-1hHBt8@HpRCk
zd^0qjM;$J_Grk}u?u*j@F|yyln%1tR%*S-DmiWp8%3?qqZM5fgVzsBYDs?A{ur84F
znrbr6Nv@7TbUV^K4631~jl@B%<tBC?4#*h#oW*Mu9;6)=6}ycgbXsOo4Kw3zx(Z~J
zRY4UH1Izdu?OiTMK?BULYUTpt^8kq^mULSu^sI#rK^HS(*?|H$);m$*JAk`7s(BOD
z>S}A3m<{;@%)DTlPZKozt_nL*3*woycPu9Ib~VS~^YXS=6eOJ5r*-a6-%b1AhO`Vd
z;fL{9bs_yvNMD?OUH1-YZ5><MAwFb3iE0KK@!P=mZ~ESD)03OY0uz?cPbxEk4Wthn
znT%PW<A|3?qxTg|b{?dLjybFE)E)EzkA11(cEk2L;59O6YW!SCoXc(MSUVa`Ni}9-
z^gv&u)Fd(W=ryT3anW0emidTD+rcOB={Y0Ec}dIsW6ts@Uy9Yw5X@!gFoda&{U_yv
z0kP3_7847+;F}rdJ@Isxxs}B3ef;wSOHM;B_%GzH1j6V^smMewn0R~?&Elfh825hi
zp4+F;YQ|tAzg-KB*ZUVd={)<YaNj17Ag|x(XzfhZc}OG0@^6D__1LS{I%*<0brsn|
zl$<6!iV?TJ%Zgvx3`#qQLmEXNh0M(QKMIBF7kH`~2+}g^Wod-y-Bc!Fe1{FaBCKed
z9VQ6^_ig89usygtt7CLvBCtANqIypzJ(^d8_b*6$<-hlSp8>8Ng_l-w<>BqO8IGkB
z-{Y7|83rY!qz~x{O{f5T<-Hi+e6c|IiAv=|XPg^fB}M<};B0NIa#qa9D_hD%p$BAV
zH=?V6VehZz@@^ir!Qayi@`y?RW78)24Vqt75@$K0e*)%oxh3>i`y(>4*H3j<vaJ02
z4RT3eg)9K}O@#<^yE*!~@SZ%wk?RRUuJh+n;IaXhkcOkN`{)!S-goe#HMQo<w-DpZ
zo#`hq72N9|{iA8wB1UoA>$A#`#FklGPf`ZdXfu6e;IV$uv`_xTsT+Qpalo<Z&N5Un
zG06WHWA4>8r2V|>ELHky>+*jQn0+uM{cUTA!8-WRDL)n*Mfnz&_eMrhF^?loulE0<
zXOs@Ub$y-7VD;Aa34W@M-_EeW;{W1bh^3yV-M{~LvdiP|myzHQpZ~#4S+|ny|63PU
wS*M4@@Gt!jtoILS`0!)@i_FXa-|~M?O+UB14H>41yLWeh)pcLis@g>SA1yZ^_y7O^

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_012.png b/0.16/_images/sphx_glr_plot_visualization_utils_012.png
new file mode 100644
index 0000000000000000000000000000000000000000..1a9960132017fcfc803ad93657db7d76102103b3
GIT binary patch
literal 146862
zcmZsCWl$VV*zMx(E+G(Hf-V}|-3cu2upziE?(Q1gB?J%d65J(_;1XDTahJ>ceYftf
z`=e{BX1aR1XP$oKoYPV2s&ZK9<mdnZ082q$S`z?(V}5I4sK{@33gwVU-Y$Y}GJ0-W
zj@E9TW-e9$6*D)dua0hC!RFK+RxYk!M+Y8ues&%<YFjrqCs!d34*UPtf!)!?hGR-x
zYX5BzG$(nGD*%9F_Fsdm5Gw}*-~a#xX$ftwoa1MMgs<c245)tAe6V?u#pM>QG{}Xd
zj&4l2Y+$0e%>3#CjvGea9lgBp8G_dq92&2rr105@+9nbWQ#(aFx{kXmv$U3WGAI)S
zZMbchZhh=j<JmEbHM4eoxYA>7IEo}(_E0^J+{JTnm}egtx*H<hP+#{+$X}U0f!?@2
z+AUwb)}oxbKDIrt`a9G%<%cc&-v?{{#F}C)|MjbNe+0?@_gJi@XYs^_=K6nnzv#F8
z|2vO&?<M58-u<ri54*g_EM0(kEC25}uhZGLR7*!hJ+F^FIMpUgY2AwQ&xg7_{qJA4
z-oIdwt^V(u+Y}ySdhRfu`rbeF{Rp_!kv#MjMF6~<y#9Tqbe{(RQZ=0xw_Uo-DN(#|
zNCE4ga;9x^R2iA7w^GA3(d8a85Ib0Rx#D<?jg2`uFj6l7!WV4AkGC5Im3Es2D6@w*
z5|@4(TNf@t?!<v1!IC;HcYE26rpm}(I2Dh#C&PPP!@Dkh*AKz&Mun*nn&jp4WU!4M
zx%6Ft_C8c2Kyh|s)q%v2gM{q}-R@&mlWjJ#8f;yH6$vU*f&u$32Bpk#ty?rVBJS+$
z+}s&XxSMngh!K097aKHr`kQZ-Y0XrfM1<s~LCeAVPba3KDrr}+agKAj&laf~m^ycM
zc82b>XqG25@f_%U%yolMRv*zqtj@CPBBQaH@OqbU*7AC@(GPolf$coPdft`=)A-HX
zIJ%iLaP2WLKF(8R1YokYKYMxR_qx48w4Yd`WG>~)f;?OT<GmpML!6j^zzN$0_c+fD
z|ClmxA>m|_tHYylo?SK3$Di<<L8l%A(CK_GWrC+KaIo10`n>fXsn3LG>u|<!ohI{s
zi8V9;FR0UzkYTN^05p*-M0ooA>gC-O)RR;8{V7r@5P$;}Nqo{@y!M6!p%gSpYNUF-
zzW52qHN3Z?eC*1PhvbnMvU9W=v7bFj$Y^Y@b$AMh<re<-@CCBqcs525aZa-4iTTNT
zw)H}AtO%pohWX#-d3ldm4Ww*_ee(97eq7Oy*?4+bx_ke0`|2an+`0H>L!*cbDk1GP
zg^;<2ZWO>j2%H7xCgoUBFaL&bND<RPA4vS2O6Akn>Ole38QH?A%DT}hC{JGAx|AIj
zk~>}b>2d>`(-iHvD`7jLxS!wS$!*0^7GK2is&3XT)s8^FPlWJv5>yb7&`@=D(y~zm
zR?Swq%<9!rG^3OgfF_Y3r7qSjei-NWeLQIyxfH(I^2=d_8s0fn51M0^IZJ|4nHnhD
z_|z4pz=J5p1Eg3Mn2n<ajX{gZnTs>Bv^zZ^x`Ab6Gl~N>f{mGG%CfH2gu>1JQg(4{
z<B3qgu?A@!or)uJoq2J)IOnX*E-6qtz=f$)V+pJK`Q{U$Og#uj+ehd6J+;5RhBO>*
z-ACIBt{N-4=vT{zf5?1?C+862mK*0^0HB(%$`#n`9O{m(Xr~Q<`|+`|;TQi#r`&HK
zU0n*shj&X!>IF?DerB-}=~j)p8(mOnKqpSKPuSv3`-Ia}5u1{&g=uVGhbOFGW*=`F
z%P-@+qQeiGGXPXMn<`{dVk%9vB3tkBi75AkC0e-^j*xv;l!!T#p9+1N(vLCo3OCSQ
z&_OB(SYW=*+dx=E#QugIJM^(z5PEiTfgiuI)GKE~Uoyv#TXUqCvL({E=rq+QUI^A-
z<OIy-3!m;PJ>c#Lgi~ngYoyv{X%VWM%;vc6w`0!(!>|f@XMk9hmB}c*88U=uQ!6Lo
zki3BpmZ0!vdaT`j!MT9)>{Y>tZWz}zPmZhJ5)mMdb-ZD6hzN`3g#Sqy-SiILGJbxt
zer-gNIdN6%jOE4tuJ%Ix(CcSOR(uojQ9E)UvkHRKQFQ&$W;#;aqLU)cG{)Nk3YpLW
z8aoQ-5yjJ~ZHMjHi=S*u3wT2I7K*@XX5DiLJutcR9B?;gdyUl6Ezkk>W<y!?XgWG6
zy*n|Rvo(LdFuVRC(YLd$)<~yUE$2CG58u7th=zbDP?AEUktPH%Ta9Y#F7Y4ivXG&d
znR!Hy7*4EJb6i4DPD~Mp7K(H$v<#(rpzR9Z#F~dPUBOe|&v8h*vAhLPvClSa{vfox
zoYC^{%CCr9DyfYfkFNfe>mM7py7-#wKFafvVDbU$v-RQ%#Z(;`>-8J`<<(Sf)4<R*
zfq1<oB*h@d3V4<I0;YZ0n|Rf7wg+R}k)Cp}`3a-LHXEyxQ0@TfmdN|%lYgbpv*tMa
zk5p4hk-xKmA<ZI5i~F#RfD+~rS;_~vGwgLiCV<LX!S~|2l>}%K+JQ(W#&>Y5$@|ix
zy#qKV3;8TteGCK{6yz$M^Q3LHR5eBAY3X_OD^DswW3W5%4yQS3gj13=D83A=p|d-1
zy==#H;RhmiGm5avGiwZAMAtFeaTBFwjs8TFDSS68Jx9}FWOgFSz!q3pGt>N0+1iUr
zrZ}T}lu}Ev&(gYl>-t-G(9n?e`u$UA5gRz3QY_`RsxiWdqU`gf-DT&dcb7RiqZbls
zwFV*5n}>)lYVD}#FHJd)bF2M09=viLZIw6hTPKY^FE1}RoN*?`lYJsHa?(x9u9Q=i
zpk|3R-71K$#GBte?zU>O%%bm<?Q|F`oh)M<Blr68xA&iQ7WA#I;Rd#<_)p#UBHSs3
z4*K7OLv&X@F%z*r8aR?7E2r3MSXL6s1XZ%!J3f~s6f;m%z#g4K043?s)-4)8pU*$C
z?_@*C{5A&kDH?wjG0irY&tyD3`%^wPvtc|KptRJSE%sFtXY$$p`HI}^+9VdBA@0TI
zF(@$oZ4a$w1?z1p>F!(iQJRKeKDegvMY7=2-2;vRqZ+~jB0@|60n8s}8-JxLHA_Bb
zS&{@s!bt#U&D@G~wzwtP7O>eho<xZ~`Ww#;q&r+$898^XabhEE!eNCqFgc6nz8P`c
zBkEDFM?h$dTi)!NgB}#g-v|KJ(BYjbCkEy9oxF&SbBk6|jm=wa<Hy;PB3~Kdz}Vc<
z`$$8yd?eR^oI_{y``fmBP;sPG31%7iU5{c{X^#GXWUA!^efS-I_54Y<@ctnAO{@Wc
z7fv-WT6p+g*|j`X=y)38e2$?5e~5JAasu8B6E->yUw)}|tTxrR7X5X)eY@$Tonp~F
zXc2wP*Y9dRJ2dr{cnWm&j5QCu5C`x5XqVN|F*tZIJ-Yyy??LiNVQ2*+x%&SgeUG!l
zgQ9g7xvuJp^>x{N6g_N8z}QSZJD5X}9|1)t%qO-}U@qXzJy)`zR?A7Uzsq;U_FF>1
zlyW#M1Mge!#L+UW7r!(v(j7lgEuDr#p*Bx<*&Cbl5eZSfZeOjB>%|b2IkVjJUO6gX
z>c6L$M_CG0O!h4OHqR_GeE<}USJv>C4cd`NvV;zs^`3apvW?m@S*W*s%xW%u*(IBQ
zNfk2GwfpGcYnw}}zG_L9braf91A8&Hb74RX`BM)WDhwsCBVx59QnX9e3A^z{w^p0+
zXzZur(xQq*lDT|+Gj}6h+v<DCIO|Nmz_xcru~UMWehH@@7$iSFqo3Rp?oE|0ACWx#
zT-^5V+V<!Ov*8}vbQ3BBQ@~%~$a&TcMKV0iB%D#rEQ|N_Y5Qt%mfuocQmU|McD-we
z<ty{zrGl)L$cQ`iHn2H!e%;8hLJ~)!4B;CW5%^eEMrn#C4M?>5i+rh|GAU=88(377
zG-{`#v-bDOM@RU%<w{INOMs59<g>O85h65$SpUIo>Mh>7`9&OE2fq(^`sja|y<wVF
zeg=t#bL2cFbncWM&^&K4#b+QL-V>CUw9iwZK}Y0@vf}W<VJe+nC_$^pJ-bU@Y-7V{
zXEhUXVLB}9J8-iuM*;*STprT^0+lnJY(XJ9ig?(+n`f54-VAilLVqGi&hILdTmozp
zGj$0Ca#fk4U4GW?Lm>JCx`d1C1jwxLit7gHR-BiVQt-7#6$he}!+0$#AFs+AWq$*0
zmx&q1Pj50e{5Ed?x*P4|P|w$b$+_Eb9()!o#n9ch?;Fmje202!x)f5fGRQ8uQJx}8
zKESzJY2cE6`}!b95yV-IHV?F?s$fFDpkjIVP}bwAKDLV1Bf^y1y<W%2ZC?Z@vYb;C
zJT8GpP+-3jh<B`#+E}`i(w**(W_oYzgnV?^CrenZ66zGtv9){2_;l0K(^uU?!^+*O
zMH+;P=Y`yV-KDO&6uS7u2aU2?>tO;c=GHecqewGP=GaN@zP9SOZv6OndZWFj)o65L
z{%IAI_vUYlEh8c>^5lFw^Z)h32ZFPlI}GcUc{=GS>uOkmCuk4}@`o+#v(cBZ;Nb9|
zmch}VX{KGisM)R%k8xUO2^#=QEQ59u8aOnMwGp+yL`8xhvxw0-6Q`{+J3qhqT{X40
z?c{~r1@Gqa>Llm+jD}4=#fTKvUWvNI5<UFowFW80htj#taD?00Z8X(AZkvc>zq?C8
z9?K8KUfH{^!4zDOT9G<TZ6fp+K3#57PKvnz{N*1+w%s|{!OkAA>gW;ZKo4>X6zyO_
z_1XxDor?!3;CX;e0P0J;c2RpV5h}*ClQzA;I46%VS5{%lv341aw9vowj!I#l+|_QX
zFX>kQ3+V288AL~&;;dmX|2ryuR^O+y%sO^haa;1YLB)Bytk>%MjH&sUP>$IrJF4-|
z1To~nrrN;)K$W-#raG^f$}tUIUS<ED(i^<N@7DBL;!A6&M_kB5OpWeg%{6Afwh9x;
zEj(Ro?icf*WG$!B{>NrrgNM3rq2b0?Rwq?GPsIp^mY}qlg7~ibyTwP>JnO3#Y>!Nu
zKgQ`K5(T|&J6xmwR-J~drcJC5d5wg2AAzXf9G8ht4nhg}m-Dk*r=_Kka+>x#;u{Gg
zk>#yHIy#7qk$*?_{O&MeTex9`dUQ+U^r6IX;>Q09duWx_Uk>h%hx5@G{UMOT7s``q
zHjbG*HBOUAZnc$_&;<cxLu6bI)sQEmN0-{u7-HUO*9&n(+qr&666x;uXk>8Xr5u31
zW*i21DvXM&CPFq639WSc<z}pU>VyZyd=2r5ZdiQ{Mvl8vy18jqGgjWTg}rKD8>#E|
zyGflT_bO(6+v2n5TQPAPvbVfP)X!^=iyB@%yKMj)R94LH33%s(*19&;p~#Lr9bv#{
zoM%%}i9dw^DSIo8z#<(|O(<UcYCp7lz0<7<1=y?K{_~WY&A)p-VOnA6TGe?<eWquZ
zas4vUBA%!gKkrz(wPOiJ*-bZVt^m5Qun}$x_NDMS3>e+;>$?cew5z~S0MnlfQ^SkZ
z^#OgQY45hg7G*oqL|G3Zd*|w_u9atU4u9w$S4dZX!;<EWe+vBlqu}y2v>xkstGm&+
z%fOUeRW3didWnW?QLrpMvXaxsi^NY<CgGR!WGeHfd5pgHJ}HU%dt_J}4pXeyhd*dc
zkNsy8fba2;;Yo#{r~wsrmY^wyzvquyGx`pjB6{rlsiBme;V}v(!3hc`7mv;MCho{3
z4B?$NRyb*>s-VX9%C|B?Of4Ge9cH{UCSHG$w@qf>y_@8GPMuIe2bO}b5;_BnFNsn?
zrWFrt;1VM=gf1+_L_y?8@aD#bPF!R>JGLX<cjJ5`7GyF{teqH~^(}?B16jMRVeQDT
z8<k=;6p8x?6|)t8Z-7ay2bAb2#o|6^O6nw)&&msDDRZ2LrC!xGVUvH|z}aeIV?A>d
zMIoh;?%<(flw|*A2l!D%sKaiZ+V~bEgz`tCw)uHp>qalfcj@6dt~`b8+4q+oG!!Kg
zM<uqK@N=r{|EF>Yw%OSoSRb=oIC!{0OT^)1Ti_F@@AYwGPgu?lbe$w=3YSPCs$fa~
za~$+Z|JqHf)U*w{f&{3nl-1B?E4((7^NCa521{S!J?%4zyNKm!AHm`FUI3%pYL2Ln
zZ!p||WsMx?dxFqS0rZZVqk_z{uacN-koB1ac(#z;ZL1Pn82X0)qe%DTDN3qLJuhAw
zDcHP{-n1rG1j)IJaef$*Xfw0cr^&on)HEc3d3P}Uo-2XiutBnnmqnBmOgS}cxK<k?
zwc)wnPLcCdotzUCK7TMF8KxP36D<E>dzZ)3{ud%WX^`-DQ2LIWP%)Dth?BJXz5VaF
z8~(`h*%}GzMI9N_tJ9fi1T^Xtxub8Yi0ilJ4RS$p8i%|F3_gXB#(>=>cO&O6V@W{c
zEn$JTP%e`y*Rc{^m&4@pS?<T|@!eoSwBVY`=ywQ%2r6C_g89u!&J_#&%w$W+F%Nj`
zz6<Lm3Q3BJW;ZgR2-|JW<J`TUTtwm&QSki5jl^}UP7X`Pc7Jb^w&gw?|GO4K3z6ro
zX*>xsoISo_DE<aHuvzDCQ^9>WY-g9YRytH#1I~Pyz%f__W(AE}<%Ni6gn;&BIFs@?
zIPiL+2mXF@Uu7;Waeke8+-%hneu<Q^-^fNcv->FOdxvJf@}Kd^<v~X-C{<R9*jW4N
z*3K{AfpvE{5qUn>r8ZQKG;S`$(em<Oy`r|3yY6>jBV->1e{8ZR*hFi)5{WSRY8F;)
z&2Yp-WsEgg$i?^%L<OHVV8e{{BhdZ#2=@R<i|{9i1V@I_f;>pcQGc0)y^DJ8a^NT<
zUahG2u^Xo@=y<8wo*899qx&%TAm7$5&j<lr5$;K)NQ*whvDq%S_t;*8n2Lj76s$<I
z<5it_U4L|>bfS`_t+iA!mK^-?hy|4xfP?MqYqS7$H!80D&db6VqKO6FhnquFFnXN3
z5XNb~qn%_tQF7?*hDlYW1L4MRmsiY8gu@neBOp{}C-6{P1b&r}Dy9%!mUX8U<KWM`
z?~b`G$i1uPWBb*+4Rv(9ZYO<!<sKBM!@<$G^cV8XyM<eD)Acg=O~t~Uft$ux9~A`{
z3|aE<;Pqsb5EtEtgyRH$(|14bi9pn{iB(+7N=^_t{?_UWLE(NLF}WD9&aqu|{)*zo
z803GlJ0i(K*>jc-=rGS^Wa{bdNb7{C#4fpf7c4zZ9XjJ@EtkTQLh(}C3mKisS*NKb
zSN*swn(~u)4C1AL*z&<tqzux3?a|knn91n#;wvhZfy9Pm`>6bBi+TL`-B6*Nr^V0M
zVWTB?!c&S9**RBpQ3*qXUWlp|-cMy#TRg9Ff<`_kD(C65$6Q+t->tr3;hs|usSWol
zVaiu`(fklvrM-ubl7}i;3vbXw5(!ZOrZRkb53(|bYLdAq*xD~w+PEC1=9qiM+om8L
zgKyS?GTyV$$yn!08gA<04GS^KN8cPKI1B-NE5jSWb-Zvl3dtmS#IHU(SFzL`KM;|+
z@O8UfbnntQPni6><{fA>9o;1sNioHY1GAkEi@PlVSD#;9g>F3VzTaGN>>0d&l4ke&
zywStJujZgXI(a!IWkU|iLsBuVJt%{T%q^fDwV9C$cueE7D+s3VI<VuzWueYf>%m>5
zyT8k!Ppb7g-AzOrVuF?a@(%iB=|IoyJ9y&V$9ZMec`y4zeap3N;Dm=(hND=4wKW};
ziEEtik`|sW^T(CVZjc!r7RTgV#V<CCAV%`?jB^MMmJAsRTSXx0I3sy+V;PtadXPl0
z+jPUA!oq=Wo8l}w@4?f3Q<yB9vMr~76sK@q%7AfJI&#5ros`94w1hN)B5&S3W<Ldb
zcwYB?GaHGewQoUvNaf<>fCqXk4gZ!JjBJR4DHTe}n4w5J;Z4*ZXDRjVN@C?N#W%2S
z*btMT{u^!?Ve$mO;rA*67FKzK<BDvYJPVJZM$14IOov*U!seQzMD^_W$63Ru0;j1h
zj}9(A1%cE>WP4V)(&1EQaKK~zwMdw^?V_s8kwSf{n@Fy!0y(MHksbz2{>Q#UY#TGt
zM8$j*0aRoscxAMG{z6(Q2}lm^V_Si{Kj?J4G5SEAZq;`kww<8~M-K)%i=Ew0iZw@v
zE|Sj!w{~}}iiJ~?`CCK6)|c1mN+`9}Kk!qX>hkrA{?SC#*3qFs*>4inyrj9t9D~vQ
z;T~_$McVCt4dP0ot@Gt@v*J0;)=Z9N`_92<t4XmJwkD9;PNLf2++KqLzvlD)47u;K
z=GK$^ei%of79H+yKRoUD-|iYK&*$VLUNQ(H<M=}3g+DRzDQ<ZyJ8Wz4N8}d+t*fB1
z4Q%}zB8&LNDUuGSE7b!HRmT-)6-VI$@r}qxa2v3wVx@@2RE)A9{64sXk(ju2mrfL>
zUkg4Muy;VpO}6lwf!%wap+c0Lu+OL&DuHxfSp9?b1nPaA=0`MGOjux+(cPzI(1rM>
zFFuYNx~nCK_(4gf<LoE=Urb@y{8-Q;@9&GBYy$)+sghkazkI)UB_Rb#y%?FNP^}gm
z&_s`pXb=%`7=F=w*VL#6Vv4S%ew1OA5di6kOq73u-(GT;*|^rrNaNL`u;|9Or{@*i
zZL0oT$T@9+%i^Za7A-7gz)u<ZZR`>e^DowKgBUPn(Qq<B=#O@EDkZpoVQ9T5X`+8q
z(F5n6c6X5gS^OAAJZ{gdUOoMvL@niUqC-A2cvJ%IHIZw!W#dT=+DUHiQoTF|W3tGw
zQpg%lG65#`{g6@xt-b{T?(2P1L#|0$%7m^8MIfUhd_>VV9<-@gc%=!NOfqbn0~EzF
z`VQpGLhAAM<-PmT>Bd~Q^ydPveyCc$Q^%T@J?2;?n!ci*D=1<iZhB;UXf!eO^9ZGG
z)J-OLJQJN)(dE}er8>^sjc(Xzi3&F-{2BWe+*T%%XY`|T9p`;^&C$$;5cUV_!7XV>
zh~fcf`MMm<YwxhjpBMBn(&m=cu6qI!ZN!+s-G*7|m<N1+n&(eyWG6cUy@D8-Rl?)5
zPUJ&U3nLplq=X>RFfNu-^(7rsV|m9vSC7V<NtsJXm-G54SOwLBy}uz$YV}N4L~Zpq
z`St8P4=Y%gK%^iW7yaWRHhv0c9;m470}H?y;$@5b0}cZh3D+@IUKxAeQa(*6vEHdA
zI|RXN@XT)J?zyn#A_+OwTQn+lx}eG~XG)EeA;G~n{qV4W1BAf++j0LfzEKysb!8sS
zN(W(c+>Dm#HY9r6hg|l`D8k)%;Bl<7|3KP}{pzKna9ylmqC41~YC!zl=Sn$nmh7++
zJJeNzR?8qp`Yq=-XiNomvy8hoP+dHEZTP&;<M0jFRLA}{jMj3*>-7^1BdAFmMS8a<
zBueF2sG=n<pX#v~UKL(h>Gb7oQZdfp$EK&s3eGQ%_?j&0KU-BWCyis0`KB+C#qhsM
z^<HSRdR3$G^of$Sfhh<lKv(jas|hL<GC4m3FO+}le=YLwDajgx(f550Mc=Z}ktewD
z*{=LEP*$s*o8kX!!Y8yzggdGqmQsiOkl~~*`1$!AKws@0kd{7r&O>y=>F5N9J7^a}
zi0e{3UpHL_h!S6B&F}s_odK==QrL0AF6&3JU_!8B)Jea7*0(?3ad9{O-kYby(_eMY
z-#m5?r=_^YanYdEi!^yk#6ITFyW)pAx5O-$1+FiQ)*aX^wUl%-0e%>rxj?W5s&a<Q
z%gg0)er{-e(gG3*Fk|m_hG>aT321lk3@mdD8*g?ku8#zZANwY5_Jl+zN@3GYO24Rz
z&uQ=a(c$`Xb4ecH?>M2R5_G$J%e*^G1QRxfN+{1GGU32JHhxyF(*;DfN=3&}qM317
zATgzSl{F_#-euLTI2`*Cf0Ehcr(rc-$fyM^J?4Ds^+AAPZt^}K6KM>3qwK!3;M7Rl
za4d+^J>KW>WN?0HZz7i3usHtITf>RY#2rH5!&=-dRg{td7y9GtMtR9B>X3q=iA-eU
z`i{`S1fnax2J7kLiz;el?@GXpoh!gJq)iaXt6F_g8k)?AS&A=}aa!GFN|KJo^}2hV
zg$B@~{89@-Aq{-gE)edi>2NtaewMjAqhv4{1#IPD^!>chkBq21;vDw;fm^aa0PeJN
z<hkt(;fmH|!GFAvS3j9o1|ViU&HC`YPhndf>+c0o@Kr0<s5&x`pxwngu)>%zgU`6?
zhLIxj`sAqZ*JIPN4f!<YfjMi{AOUzp(D3W5>GmJQUzFv{UUQX|{T-ewD<?@agp|zK
zgc21|p6bjf)$bZ_)T4LG^ons$nAhTbw5~9sPmPpREXL*-UFl1QMN02Y%{E8>ZBh}$
zerNF$IELh*#IZd+pI1oca}SZ~P_+#mwp+&@yHA@wRJkZbm22!7|H0Bs7Qm}I@F^_~
z7L^Wit@{ysa&Un5mz|gswKBI3US3^McwzSf9$HZl>6O0=GUDct4^Yk1GA4ne*ZX-J
zDwVsOjv-f}BAf|4*%9h2?Lrf1Ngp|`_O7ad;9_jd;GF#k(FI1~e{Ew#eM6;!AI#+Y
zuLt+&?Ul=0M&M>iao2u>$+7SfG}u3{<RwAzUFwKrSI(skEvOijrErFYC#Txfzfl7w
zFwyUdqqhBXnn*jSEH-`uG&n)=H{WGkqm2uk&agvkJ*0&z^ya)Ax6*b~N$O<#brd(v
z3QT(;aZ^%)YMD=C65zcP_<~e3xU#KInX*7-X429RyTh|8qpu;Cb>qRO-h-SsXZmom
zw6!b?<vfEc5<`@{|4ifQ|Kzgu{=sQotI=DcH%$H~W1krQi&Efx;3Wy+11VW11UL)<
z6$<8?M&|$vAJbM)1)Y03-1~r6`;NN3AW|eIVt5tTU`+-+xT3)k@1#9tw*0J^FY>Jt
zV<UKT#$((MQXcv{>kA05TnM>Sm!dFXA1<#J6I7<}o7Y0g?Q{po1}4=kzFE5+)3iH-
z?o}hcLtBGH=iT6g@!{H<My52d5x?a~gz2NEkJ%J6SlXJzF?GpsU~b6y*Ee}PKu$uF
zisoBh1#d9Hh!s4)%n4oCnR$Jb?Ta)D7-*O{Q!X`+mC(_l3!H}vCHF<4@u*l*p{^7Y
zBLlcDyE(-?S>l&{5#Y&#D!59K3qPM354fIw^VC>Lx&A8(56L@IH%OXB>@0C~dv`)s
zQ`x}JaawK=Hl0w6vKIwsQlm&qBDErFjr&I}?fBE|$PJgAy-#~g5=QZIrcM!95c;7*
z0$#SIN(a%A3&qzVSB5HjXQC5D&@kmst#*oVsepDR9+j}8_t0^>OP$^OP`8;gd|@Y8
z+?t*C1sTCk&;V!X(%Uu>WNycMjCEDZryPFb)vs*iM`O03uOxNjAtjT}nGkg58JO=4
z%@psX;Kr-vg&UvG5$bE417>ktBLv<@H#c|19w_yzX8B@sam1cYp$M54601Tbmy1Q0
z+$A;$5CZqSy{t60ef=ekTG$Lt)QDW<{9G*yKXk;|+KGUbH38z4L$*j&s@o=dK4x_r
zQ7MSC|56eNI(N<liTdCy6^-iIcv2b4l)eXlC}cG0(4b%YU?sQ))jb_PUnWNlNV8J;
z#$J_vbpL(kb=K$BhdvGue^rOcpKjaekkZ_y8(meFoG4RbM_|*V1JPUoU`x(dyui1;
z_Gqc%X>H*xh+`y|XZ>*txjYeiD{r7&FLK~$mC_7*<V2o5yZK7PJtbgr)p=&@c2eu$
z*k4lBqz)K<UG>?0OIK%boPHn)SJ2^E;4=TT62d$9RU%ey>2dK1k8Q47R=iX&<l<Hi
z3^Wnq1eT%ssp0*?2X(6@zGTS?pBp(9V(77x1_5A?(>(*FRGF*TXY0tp{9K-mnpO20
zSl!ALmEJ|lt+Rue3O)Po@02SeYP`sqvKN<n2P|V{K|@vEkUS3F!2rfQwezAc)hXgb
z)e%QZJ|E%)lc5Ac8N$<iC{MNlWz+2I!e|Z3f3ftV@Zb}-TUKn;kiHK9v04p{Nn^`w
zk`K8WbbF*DQl{29--yh}3?p`;*)!u}B|&dLBi&U}970UB#Vt1>B(v|yFatOlF)wSV
z;nMCY4^>l3rq7Qs&%6Wn#SR(Da5&nGwxSr2^KQmRo<6(D$C2Vx^z@~X<9O}g`!!U2
zfPCo=7+7!arKHq?sNoTkw<XwM2#za@TIiSCSr((Ht6sBKR>lHn@)*4ERg8Ante`$3
z+J>W4h*tzHf&i77XOY*AL?T{@I$<2sS+$z2JxAL3j8|jiLNKm)zXByb*$5U|E}#WR
z&@NB&<*A6Sui0l!NWLvgs!O9ttDGHf;H{N}=Mg9NENvPT^3M%Dp<TpD?%8|#52lY*
zQ!?coo#0X%b%pff>bKY^g-_E2H5shN>>LufN<jZZLh7&Maw!JvADVt|L!%zDp@LPq
zI_Yo94Tqm5QtwA|%AZH_vWM{a_pC_ZTDr7hxqt&Z;8&dL{p57tp<1xxT*>DSbN6jD
zxEuqxp)xv<XA@tY=pLKizZ+i_@zF{5(D!JZ8Clx3V6!<?$~ZGz@hY^I>F+-FDG0F%
zD%xRx^dBBs5?{zr#TJTJn7az;yT@!58#uU&TB=v}YGM~=Tet1DdE7d6iBMeO1OIiq
zf*HI7_@0G@zg{@>RR%N8X|1~A$}0BRJuJW9T%ZF8yxMP8T(E_3Ast2oESbrD`1vCz
z^ubDCY3Y~`$`@!5<9?p+0p7EeMwkt?;r`+Qv{~jjfaQWO`h+?%z5X8}Z%G&YKQ2Lr
zPA-BWlm(EJ@rI!{s#cf-;2~6k!@L|p*qN64!RB3=fA5J);fF#*6X&eY@1V=WQ`(dh
z;Z{1#M2E0AL)_m%l-%X>1VrMfE)I718%A@ua6%DvI41|eN*Hn@k3hSlIXiAcFEs`C
zsZUo@3FG6tP7ZItVdz2w8;v7&WF{>6g1Vmd`?#A4rWp^~C|xN5TZ!|9zM^$Yz(#Sp
zOZv~pV2h$L^~Q^)?zyEw9sv%FS%?=#8ilsKruJ6BNm#jU_F6SY9tlC^Wiw%6wyBt$
z-rpZ(#0Fm^oa)77`ue=a=I4#;lJ1D_X-?7`naPBc_3lFw{@KuCwGIfY{7OPBoM}^w
zzj;)3qZF@Hx%I>8kmacbFMpc4U%HW6s2Ij&W6k5E@mN8CfmbYI-~Nkx_Ud#_aWDPN
zk<!O-S@`+J>vRZAgRYRPkCv3-)ERR_bQ*`6_1_zFBbW=Q-qOLm=$xPMW4mEhMQ&pP
zc<1x&)oe1=5xK1kF7E&~eQ|-~4Vx})riyiY98E1?&7-WqX;cKI%gcXFsT&8j-18Cm
zL8-`riW7<7An=k@nU)8#X)t_(-B4KsanPP^mN!q?l7rH{h>jnP>wtcH*m$M4?ViY&
zem~J^S5BO<22ATb)QX<U0`xq78A0#p0dnA_jKW&y2#O+#OgTI=+WqGOfO;vQcw2Tw
z{)e<w31JjPnR}o4N6Ns;?;4952<b39j$+X{gyPg?ylI(2{B3Xi{Y|BeP-%y9Z3%q9
z!$@{UGhzOPs7nkAO{O`|aquw74Ln!}eUZI&ok>~hXn#Rr+(2Q|#;t_%HUr&abV^-_
zTX1!&4CUQPpX2!qG7FW3lTbdBY0A<`g)B9AI2tjHQG)jAZh%vG1XI8vfSzWzD`J+R
zGo71YBt+5jJ}HC#Id5A|bC1POtzcbbpYep&@1w2gC1xoTWRY}o)AoMwwmhFk;=6-U
zVHLvXm`WyurAND9_7g@GXZ6_Pe~|baY3sbLXn_)nm;!IV;`hAthuSV$KTOq}gX{4v
zaD!;JoW>(JZ<B>w3n0<VyqI1PNT0BpSpuQedE%PJ&$`f!(-3X((J-rO4C9i{Nb<jS
z44I80?zOi`f;)EbG4sun*rtoB)w9i$Mt$)<JMJ68=PGMV!o7myR60VRWHds>w@6Fb
zTUWan=o4eUSv4qINq||%pvlo5Qi<oEN6Md>WqKd4y~B(?%g_Ck?u@x%4(NalAXL@Q
z-lj>hnjbFdst5+Qz(XAd+b7PZX3MUrq{O}^69xHjcND9%w>)aW3-x!{S}2b6G1%@E
zl+KBLoDs)gL7C@~4Ss){^u_Y1-JBA+)jjAZ@l=hve&KftLdgAD*Q0;x?YCjLwQKa2
znvvLWK6D)wYP)os$R^mGU*zdl$BX4@9Zz<vMr&EPMwTLqGZAqgCSbZPqeSLBThHs<
zhB|mZ@4N<UC|H}oI8)8eTrQBVnrhewE;b+XURhgECmyUi)E(l<nzOS|+AN_#MDg(=
z3BsU)upfc<e2qMv2d)ijV4rO?AdZ(~&}R|O0Pm?sQuS9RX3#HRl0iVfsMjs&sjg8f
zq&GqG7dLJ8v&}Z7*kxhmPKk-l2snBu1KNV<|G=J8Z_K&F_K(I<CvTx)CF8vz`q2TD
zbTo13xl3E!ocMGa$*;Vuxi96gPwh`&70`IMp%mveNyP@=$>eYIbV#_-C}s9g)Y`tr
zS;^@$t&jh!)5V(*iUCPffrve`Qb70#ftTOj1w$@aBtlq?bwoY4(ak=7i1}_77*}_s
zC^`0N!9AptsV!lt7NHMs(W;E?(Tbs32EhvCG!}}uAd|F%^K1HL5tAV@SeopYzx<qd
zec~KCTOHvn`XyF5V$%h2B4xL!>KmBSFH;;uEl^xg;uK9a&P_Y*N^0>&Ed8hWY8F`&
z2?0Wl04Y{^ohU=jl1oMh+*icw2C42b?bU^o7L)F;S&KrbX^<?15JdP62@UCLXj1xA
z52m%h=}Br`QY>)mP3yp{pU@c$-YB|e8^@oZSNv)n8_TPFWYl~6t7?&aW*xVAWb0LX
z8b*H8GDm&_j!A-?4h&0#`Olv#2v!D4L97W?whQYzQs*(;#S(6rMSBL&hl&c7=R+G<
zxG>)hIg4v(J5h>c;}82)ldHJ3?J%f+oaSM7O{1?_<*>lfgjY0XB+~iA2Me3m&|+zz
zk0fKwTFaC3>*>2~>0JHbw;`1SV3?OD!!FTO`m%k}R3UPDGlDR9fwtMoP%0eD-Pien
zK*=k8WOcr(g*o4pDmqx?Hj9Ih8a>U`BSh}YUL!Gq$oRxw=8au*K%P6OqM*Nb0^heU
zXIjk{G&<4gpa_{J(t*b_0A9j(XI@?Gb7SGn<wasC806e@?`64BqMtXZj9DdRk|-@e
zAs!VFmQvH?hBKFuDAt(h{<0<vW=G)O{mDV@?MEW@i%Z&IktnZ>+tF2#Ly|wA&*2h8
z#<{h7exMH#1`NP`iMQ3<Lh|~}*&+2ro?ELt);*y^PK&=VrpssLjeu$(Dda4fn=I7!
zUZpBl!HKUBp_#2B_#o<0Ah|K&%Eo1M!*mT_e6h@|zd`+ke=vy(!_!9O_^@t*b;wL|
zqtr6kw?n~EahOg`Mx$FlXj0@|6L0>uE?QLKRaA9gFfzg!m4gu%wt<|SVhbzEnO?T+
zMW*4@lZ#A1hnYL^q6}p{8>?_xm!QL23314U&K1aitoCYUHvCid8qRS^@4fo>I1H**
z5;r-!Z{8atJo{UeGM!8AF=g>lr(4g_FDYM`t^`?Ax%JYC-sDc%R*l12n|>Sd?3$(q
zG2-a)V;c%{*UjitU1dG5e@~p4s<C2c1@hHsOm6JnQh?N3s_}*f(2${YR&4l$IuhBr
zyl`~vnplo>difJHSVo2;{u_8D9X-+yQ1LUp)mL>zylW}&bb&`6(I=h-_Y=vO6}Z~W
zFHrsTF?q#UV_kbA#hSXz>5N53CpN7x{&Y{sZ>Vwg;qsEl|M|#dr##=^QcL029kHoP
zUL9ktNREr1NEkI!rjnYz<iF+hGO1%|cqp4+`cLc~$#Y|K*^zd8oYNgtFwulagmW}6
za*XisSm0F@0xbe&Z$EeL?ej~rDd^k@q6TmoZ*!Y!YYps>WoTq$A4Jw<K{Y{@b@l7L
zF^ejbyCY6?>Etk*M5uO3*O2xoqMzS&k|sIfRAF^!!N|+chqEP3DIAIE^<kHEelP78
znjd|gIF;SRSHf*SR@LLeTrPI&moUDI|1W5ki}wr2{CGR#zx=5k|8G}oSWHo%1#dgs
zn0OOQh_cQ|u>yWZXDF{cfZ9yoGOtu`vRgrnnKw3B<_+T4L20@e`K<c-u6NdII}&_R
zr3W>=fY<Q}9*dR-xx3-<P)cv0vAu7{UQaq&{6?wcD!C>q0cJiWgPU9c;-?ue3&ih7
zQ03?6pRc?R2r+p+<fvyS04h^o>gqe+bcbEAZn{(%BlKsX;+(vT{lLSaMKwd58Nc;L
z_<WEk3&GWh2JTI;_Cl(HW@c!{v0^_*QBU7+-*z+e@i+BxtB`iUElyFhO%$?W04LFg
zq!alJ6>G0PR^<-NAYC@GjaXKVW$}I@&gr<0`D3%n>Z;b;q13sQtpDK}w?{SxKJwka
z{{)5QPJhNbd*2jpZ;qZP;a`l3n_kQ4>}B-o%i48t1{8{mRMRN22K<y)vlLb~@x$n|
z+Y-f|8J)M2G`VS%7fLh3#q8>+XJ27x5efjEL1>}@ew=&LZiOOj0p1=^Dty>3nNGws
z1Q!@SdhzM<boO7QsR9OYq@*lq7yMqXw;9-%P(|PG^PPOP@%C$E!Vs;bPUt(G_cXVT
z^9x^BE;q)Z8=2fm^rn;%o7j9qh-IPnXPef9fAU2$8JWf32;{SiC$gRw=qu&gwe0RG
z0a8k`9)y+=?`hG;t2Ppw!8aPG5Cxq}Uk|}eFIcd7Z~%Zv^|6|_!LV{PRU$?)|L9zT
z#`jHT7Q6E%6av{@e|-UFAGUpGIJe~(IaynC5DzW~DkPSsf`twl;pc?}T#J<Qfm>N^
zb;cOmBHj)#N|_)}zk=CWHpD}Q&&_DP%}lNi+0x2cr|Iz^(P9>*MmVUi2%Olv{G35`
z;x(Ud$KA&3FI{qCtTHro92}`3-_mQ9+M3qq*7w#77@2m>xWkIr)YGw2x%ApnXGBYP
zW<%iWU7Yl_#fv;l5I)>Kmwe4`Kx^WvSRsahgc}P>OC_C1AE=KjTFZASy22L|9{<GR
z^auY5li|iEa37Yv_s(Zd{OkU1&-vHP8fP5se~jw!;n~4G<Z??DkpKlwwji4?#@f;-
zUD0+L;Tck7?8Rp^TD1gs<p+VF-6yP5kmkA?8vcs!wtN{0ry9mvi^I;|y(8P+TXvD|
z;@iqfn*~$OZU{vmZ9xg;RAbcxs%r(TJ8B3W5>tg!q6K~KOkZOkZEn{sC}lDEFN9(?
zzh{%{Xd&nixadvOe&pgEAA+EFLNCbaH+W!e3BSC-pa)u3lt`=D!J=9HV*K^X+@%XD
zij{b^);e$1QM|1@{A0gKLB*A~-3(+WeIQdgG1J9KwM6-pnE0T1x$cO%f0MJ@&-$Dm
zktp1{aLPW;-L80UwT0j35`F*S(|S19pi{I6d)e)~zK&G5g@_^qLjF|txeUjYR7npI
z@r1`vqw3vWhofogkeZm^>XRR)(Vszl$Sk4OEsup{q$GNBXVyfFZwAG!OIt5#h{@A7
z3GOrjS}@_KXbSHhWN^NVK!rM!p!ry(qz-S+hbwou>@z|t*A7h@45T?j7S-jcF;r?4
z6{Oq8;4xjMs>|S)MYDZx7i^k_z9Gwd3jz+mkEj*Szs9eFlLSWyBQXqbKAZmiwPQr%
zT3^HJT_*KG7@tO<sd6lC>3){Hq$gj1&^$-hR$g&R<WlnCr5J<H9Ub+I9VU{iO0)fJ
z#}4)L(fV}{oFpwLo9{FQefA)r4LyR1`Mzcej{sFD@|dWR1N|q<QnRUTEsp!(Dgwa(
zn)jWsihwZ@x~kq!(Y%F~d|<0TR~V?_T5}u1YOUA&6G!la%W@iogowl6Ipi*Xz9-0(
z#K>?cpHxAq?caFn+K_s5Tc=BhN9_x+Je5h|0O5HVb8ts!{+dQ1NcT~(!3O9c-yaO9
z88oyC%dG)(v}5AXp2$@#{$i<}TVNc2LCCfOrBfL0R}2akkDR>e4UoW0h~)}8zX>O&
zv(j#F^?SQj<_Y_;0Gv?YT<vWgnraIUKs?Y7E#{nBQ-1P3OGe|X@Y1^A)ORHiyVuQM
zMbh%9ZzSpT(m@$Iag`2C90^Sosm20LpG%laD@vZO6}PyaTC@W=&;U9HGyXj`^R<2l
z)r}=jeB`h8C-x<t@$lnA;9m{NKE@@EmGP(V8J@dq>`(<0d(u=Avmhq46^LTNw}zub
z1?I@=g~a}Dc4aMB!qcR~ct>B*fQ{H74bFSRUkS|=pFUFK@=J-iy~Tq(6*}g2y(L~k
zxsG!Fy;<SWI6+JiIsTP`zkZi521Cx~gE(y(ME}AOA;+O(!l8WXVE3a6+z?&#MLbm2
zFcT}&l{%b$vNwIHNpYo}@9Wy_$R)ha?@c#<Nf5t;X*Bc%6_y0ULpX?h8P+;*h<b-m
z$87Xs`@H+N94x&-5S9BRXhjMfQ6|3R$I1VA11*e%yYkrKzAedBYJ~$6vrO#TSy#n=
zyr`%>TTjC17SS-0EoJtRTE}Q=(<7qZ#oG!hNSd6NkH<<YWMhq1pLX>Y!(i-N{I!}w
zZ2+DUyh>*OSE2rWk^K6*%$f@xx_5!k>x?QM9wm#_@zwe;`cSrhh>3;N*OhGG_nIT7
zi)BGJ!;0jv6QjC6cz+nOwKPBfGo!bnH(I4FO!iMG{cirxrC#Za|AOxFToL;EBW@>|
zMUgSdx2siAK{g=6;CCiol&3Is?$)LK_{M2)7GGLyS9ya9Y`tqo25Rdxh5b1?Cg6+A
z!gu{XFrrZ6*={7eg)`NnkDi>#)MJTZs~59E5etBWgX8Y%TnqTI)%Lqr+YOy_f_|%W
zt?tA;DYgzsmy17v|9A0St2S!5B(ZYhevTCAa}f6~ugcfg%R{!B77B;wxXR}*L@vlj
zWrY<kUpG80hZG4WKa`?ZetS|22$Pb_q9a2Yvw#)i2X{3Zm&7U|YCojK>*T;Aw#{^d
z-;Pfplm~8z{J5eysv9osyoKvgVhw&Y3NTq&pO|1aIco1+9&unox!xbxHKY3{enzXo
zq4i~pLuP19KvQ}A?+Bp1cETB$t{tdA_&oNVWBq4DeXO+KflrH=kC)nn@*?}bP`@-2
ze^lt+{ldD<G7M2$hykt~>3vFwZXJaYBH>!1P{qG7+jz4g$&x(&u&=oEfK)F)p{Yon
z{`7JxW89b=7rlAEPXZ)My#?!K5B5qm1=}ceFwnL~`045*%<W@|**D9jDCim?%r79)
zH1Y#~kzh1-ChGVzPHMR<ky=i%Kw(EEeYuu=N21a{4W;3gTaosy@ZU8w_T_KF{QRjv
zzB*w?llHqnc?JAvW`P724kyvvg3II0zwyHikBu;&sqy0ID53G09DVq89KbQ5+?$Rr
z7)f;i6#+@0filaL4Yz#|2s4XX$WhMJN(j8T>H`1w^(;?IBLYCgRUR^h75H3h#0FOE
zY0coeQsMn#;hv?1{ZydpP!W+&DEZhIH3X>y?3m6x%>gg92%wr`FHGf6F@wHz>n<U@
z)#-t>pbb|)8CT+2pKJn`$xb-A+=Llgu$E#}=ihD8Hx{g-L8*CO?w_Kw;d*u`eB?sB
z%7<D3qi{!)YQn6tq4%$|@3$Ud0jO84VEgu^J*HgO{`l}oD%-FJTF}w3A5ERH;$IK6
zh{ORJ!`(OJTz1I6vjk-LZ8UjZcj-67>gI@$z=e}S$@OQitj#sQE4RbmHb#<zT^F;D
zCUBJS3r-;ZJ{ql{MB|!+qwU<%IfLZn6i0Y*EC2q>cgH&SxzZIVUr0+Od-rI)Pg_1{
zITC1M5`SU3``yP*?BeO80P=H6iM~&+%waa@&){p%$+H5D9yK&j2S<!zES#V*Xuxjw
z7Wh06&(@1zj(X{M{UB_Z`l5}c3`8smPAlEHVf+u=4$y30sP#9p&}P|}nP$d)wqk)&
zK)AOib=`G$q^Er@<5dgm;-*!}Q&YPGlR+L*5`!IL+i1a0Oa^{VNVqUhl~qt9mOw8n
zDIn|eeb=p172Mq9@i!lR6wbwz652wWKv?%H^2Y0#k^0tn13(gw3yA|?wTzWFMA%GK
z3Kg#)S$oO==pNGiJ&7WY4p$WQ;*N~_)@@Ugy3Tb*upUlZC+_9|GsT80*+hn>+Np;F
zuD6;s%mee?AO{>%3^{CiBlKGrEx$h{vBghn4)6vJCEyFqf`I#cfrFYjK2B3TncgfG
zDF=!5w{o<#6QoPRT%!M6ZhP%i1HZ^F)HM2&B4xT0LbU7nY|e7;D7zyTxf*;Q0QGaC
zo-BSH8nr+aouuE!dbEs!kA*&F9r6|+e$2y^FEs-Gi^-(kL`JGTkKe{@oD9mi1*tg_
z*)%*Y@$N4p6Q9D;rZ*|yTkxi2woM_PMS7v_V%8s3v5%*IUV_AzA8^J^gzOSUS6SkW
z=FR(r+<o0sh+%f?FtZ$nPeamD_RNVm1WKEIt!Kv&q~0{GlHEJ#51cAc!LGKbv3m0I
z#>JTo9s!JrQ;d=0hGQrYmqoh7y`7H>=96peeN0TLsHQX$NER06p0ci%(MT1@&r|<L
zKUY;AXc9V5?_R9*0piweTNLI%K`4{v#U>TuwGg?s86JN*zDXAGMNJvY?`i$YeG*<r
z0;$v9h}f1GS7tKjSfLee@{chA0-*_|IwgYF5$6pdRo%nLk0uTr&(IAtV>*&SAZDxa
zTXvV%=u;h)TE~)dDBvO{J-K>PMOhzPb4-Q_=3tm5CeU)<=pHguUCHKTF6M1}<R!_z
zVx=S)!;^JNV1&T<_sDSRPig1s7yMNM7){%3w%PCW*jYbTIwicbH~G0(*mJE1C0d8~
zi2If<h_CKx@u6Uu_DiIB3`YGcU($6w8G8&nL;tlR`>GYlv>FTVZe@il{V1K>Z4vF1
z+E@=qp;DE#kBRG`V5`&XR-PxbjM?|p8e2ZgzIY|2Es}G9Sa0n#2;0ar>})&c{<0MH
zpM6@^H^l68*(J&!I_b3$b(AMRqmJ%B9?xYIk(SS+G4O09-8?$~<Okgm1o{pF<0s&(
z{-%1H@BQ{d6C<~f_w}HXT0)67htg669ak^u3mI4QmI=<4z5z&T`J(kG^R6REKOIk}
z)@MQZD0=$za>^$Q!v1bG8_RD}S%u9^l<&nqn(91cthsG*1k_2zg%W`chRQX5WEC7(
zt3C<8HFMkDb9W{O-N=}4@BKAzza_h3&kNMqjKWWpYNxxS()sz?j+7uVeRJH}&SS<9
z2QAuW^EAZHOZiqR$mCe61$<JW)lH<q4KT{UJ<G5H%}CD2siv7)N@FreNH7mrRsQ>{
zg*6<Zj>K_puUcU)uY`vs(`^F^cdySj?6<ejA?I!`$*L<?q)qOp@j`P>YvJ5WyCdVP
zx5~?jxr;UQR@8VWc&MZpH0!0wc5JJXVI?t~v;t5u(MUt2&aeu8cN*RlAOW7M_{5@R
zox5<UB~h%paqFav#=wh3J1jNWn9*)6*;%W64bRjWL8*d=ktzfsD)&i}0ru)gyh+fV
z!L?UBnl>3;$n)q!nhS4I9juW4E#I#Cx1$?YpeaP7WVL_6Gn)GLTL@p@>emp(K;5iH
zrjsTLC(80Vu4+$p^@_H}`wU5}BII&THJ<RGPP+9^dL7gpq3#k)z6sfZzt5;Vnhg6{
z%NQbA*vrZ-mi+axjfpv&%xE=^BH-d(k>f@sbK55;_XJ|aox*d&hZpglkBg8Qvs7`l
zpHQOfTXTclT1H0*?}es6KUBRjARZm&ImqIWIreB_YZ7rx8R8PmJxQ5$y0+?!us<Ij
zyP@~)K@!xQ90Kp}YUq6XVMBDT{vB&hI|4MNgbl(<fPYK3J0AY4dz)jAS!FbLrWxqT
z-2t^m%uH;oY3s?2B(R{$N|?ak1jN4Q_Iq65kqCXeUFrGS+lg*|taSIFCf3+d0Y=wd
z>OF4TJ3Q`4!a@9UsK}tysPUQ!X2L{Q%YR&YHI@vQ0X!lvO2ViLdszgggZ#iP<l^R+
zvm8!?NgyOouqVdWfO2IIKOi+2azL{j{C@ywL6^Rr6q(%hH;l^~0#lXNsOuJ4MrDYO
zTEInuvf!~y6S}^GgGXj)8k?pezVbW^8p9r}_1*{0a1lCFKM<g77dU63;Lx_GV8&X@
z&*K5ZKycjT0KAa-#lKT73L??yNb4eyo`7-^y2?|Kse=!JeV&;tt+mLq3@SeSP9p$=
z^ii8q%$^`cP=}#sO^nvBsauqF4I<42TQ?Fk7?4|dfFrwA&<4h?C;OB&JOXJGGmp8m
z6wgP{!qnDnPU%KhSi3O&yF$~P0`tPx2x1XF%e;g-it@gqZqS^c!d{5Ek^GFF#<bc0
z-osgo{(3{#^^iPBsVG!c)dn8|wnAw?KBXc>@v^Rf;!4>~gZ50k=UHBa3=n!A6Dssn
zRkexdvUWw+UD5Yl_-qOJNQo|A6Hc!?$Bx^ZFeys1wWJw&4l;8#<S~VV$v5aQ3>#)k
zv0Kw%v7;csG$~T-U+G7wSV1P`Et1dP3xrf?2?MC-<Y8+qin2mcY-6jA?C{i_k_x@T
z(x_DA1lpH?KU@?Q7c3MCT_T%kRT0_!eIDce!lzi1QVLKWUO9N>;6Ut_LCA)`RiTca
zQxx<m1bO1!6EkshGbf)JKl^^JMROU3k<&|}b`h9c^}X|8|8%6ykd{ZR5*Ipbh_3Dv
zb3|<}roDN#`y5GoNxAt2hdSc#KALlBj@b=11!x&N+E@!m-R<!%Ah2CXgt`O-hhXP}
zG3>Q5&jNp)W^|WptmUU-^t?bMQn5(JYV8c?EQWC)d~}ZawOxXd0WqI3RLR*t-tJtd
z`z9^i@}>4BP3n5RKx;6#N?MbU!iB&>L^Dq!%Vl6fE1MQ*IC>a)jj)<2+)yIig!1I5
zpGTW(o?{@X_bzbo)J+5PplD^PF$g{Zfb01UOl?9*w$15;s*%)}E%vZ)A_&rGY3GT#
zz+@Skq^b?QKWa@!0+c5~I;R^`b4m3OlAz^Hp}I6UXp#ay@3q-<?H&r;qh~{`(mYKQ
z&ACe|zU29a8S`~BT}bTYQXtH$sN!KulWO^$cYS-tBGz?@+5q!3nIa3ZT4Lfn+Eemx
z(W3gkM?Z9MZY6U{XK35V1s*gD=k7dk8lV&`-p9i@ZZ1CnAxSz2Vox4G9IQopIl7J3
zLvvtEq)*<AgCQ7W492nJU|AFD3syB<mJ$yo44`rjzZP4TMlo(*U%~ioEC}<6mlsOU
z5dLH-#gj1)vxM+d%{v!fD<ttZ@k+dAM+4i?1rbPD7P-$K+4*UcjUwqd&RKYCIY2SG
ztX1;2kD5{%<2dkbVqvljMjNzEL)Awzs`}#nm-~`2iMDCo0uPU-J|_&IG4@Oj>?<A0
zD9LrNG@uM`=4c|)ZA_^#EcFs@^Vc?6yan$lZj-qdKwb*7%B3KlseL)~^k7M!drc(-
z%JP#G1(BywERG8IM>L!6&oI@vv(YnX9Spn8Pp31QwheQA^lbOjog~!GOE7(#rpBp_
z2RFr<?q+lcX-MvyhY!!$CK(Ir6#*q<3c<yFk89TveybErrUOrT`uC~DlIJ<JZiAOs
zDKh_`pL@RGS2q2LKTK$Aj^->C(COy^l+6E@3dU~|Qe{yA-lMzS&~;r11}m!>IpY?V
zlV3%h^WKL9;xx7+eVF|N%A}=Qlwdd|bDtlXb!p$}<V@3qX&f0(qq5dTQKG48j!W_I
z$syeLCw(n;#jWp%{EvllUQ&tlXqw#vqLBQ;@H>ZjT{xI@g7<U_7X_J&Sy8a(E#fmQ
zX0~RPjFWFz5jsA6ckosL##4!OM6)7f1bYAuP%N_~dR_T=_yubM_@MAV<zUL+iSQbR
z5w{!7lV~|r)1Yo!fI~MFN4zVshP-%()BWKv4jsDN70y!J5e&ulYys(eFPTfUDZvE=
zf{+7glp@}w^Ept%?Du85$LMvMkKhpC^d9f2<WqBsJZ55c1+y*=kDiVNh~<)C<JiM<
zkgzfu&92%@>it79*8P~LX~Ok(#Vi_`VvQ^zfjqik2$C^DtD5H->oVi=`5Dj8&&Z1$
z=kp15(>@H?+|SqkJO<#MJ*r>R-Xx^X=O=*8E)g1zg9?2`_41}^#59coJMx~;e*xBx
z_?FY3MJVhu{o#pfwW;cd+Sq<Vq#x48Rsw{1A?ms&LN7;~LNJ*}Hz+Y2^yr5^m{Vk$
zQL=4}uiAo|-aBZBkYNDQJ9Ir&+OdwjD_R+}4ST5s{u1p#fy}X;85s@%IDbBX+gwE3
zI}q2VkHtMTX*z+bHgwB0CN%H_zfMd(dAdZa^Gq{L0*Vt?Zx<}f9L6YBnzLLyl_0M5
zAVd@gHUh~p#L6K8R0@nfyA+P7B<y70B6h}W=RNfm3dS-~Tuzz_(ac1A?PzbM`-`Yy
zI&|HQ937sREZb9yQn<jA!N|%xY=xFnWCBvmQJU<lQ8y>l%+Zx>sJmy~k5`1iG(|R1
z;nm+??)$lu`4?zLFjM6n&L^Tj%JU)wZ0_r79$63RXGp26ebV)p+M9$FKzvOm%K)X3
z7kRL^;VI#B96ENs(54q}o|M*k9e<!Ed08Q{+<Rz6c5_jdgrn?)+rBRmh$$Vg=N!Gd
zZbBd_;?di*{a7X5%dQrAX1IKHy<-wR05f@urW(hTcw8g_wq~YFrPPMGK>8vL=-`=I
zvZUNn@iVttRVosU)}(Fl<L@!4{fyjUQURwDL8V}=<1%MGIVW5+tMfbq;DT9pYyibZ
z5zRLY{X`KGN+Zt-oKED;1LWraGEOGWmsosOg`xm;-2@!*HBVW@dK|?tU9JQhc$$l*
z-d3U!?Jbp<o7YmU=Xwd%lG4<YBMzqHF^c2J1{{k1OeJP^g`<mnYih^-0K##HzQ^iZ
zVAjaX0wu>dere89;Uxm<a*5CSwe8J9%7fI@X<6aCXYZcFGB23M3F|x)*;|&wdt%8<
z?bb(vDt+LTZBrW26_-$o_>YeINe{I#CPYz70<m+B(kGf05_B4b0*>ux(Z+_p$2^QM
ztYu_Hf%bGFOv?+uklu_rHcr&l2Gca+%I<3C2*9Tsed{5K6^)ax6nDYy5CDN{MFdq8
zETU17fJ*zP>^^rVq{Oy3TC3E{1Rw~CfY+@Hyrlrq_SB+D0BOj$4rVDewMPyB4Vm>y
z^9|lWMCjM#e9A>7ps(zW)QMx0r0;Yr={(C^QVi2=bAbi`SwMWH?q(A#xD*t;VaLg`
zu;XM3pDAG#<%3dLj~dkmOz={Zfxztsb=?Nm)Yq&SesU<KS$N(s%_~fn5xZ<t<K6pr
zhcS-({`Rj8!w>{iifevNdmHH;=M-57Ov@aqsd!IbxMiNgZJ`x;meR=bM-P&=wB{X<
zmRqmblk??BVxc<`ZOpTIq7-HG?<3U3Zr?Zg{Wy;3IDpU?L(Ga?!)+^A4i;T0Hnk#6
zN#Iepk)pV&VpZgOLC2<0VuBsdW5h0$R+~@`IyG#LdJi}iVxbQEO0=s(@80I$?18&p
zG0zjHz5qo@tc;m4BnHEjc!=PJisP2Gm?SWA*+w+CZA&bh9wA6Jfu8R6g>x>H3zzca
zlDgD15-t+wBq})fpf8YcS22k1okw2eG3b^T$<xCmtCwsq?o)X1Ls@i0mww%lcK3+3
zDD9qgX3X=1%jE^@x&(P!*A<?gp1v5NzWp7Qt&0_m=^`w$45y|=Rg{6bHeItv!cS{W
zWPL9$p+K%u3QfX#@ul{*1Y^$M(#m6<W+J>V%8>6xwrl}wJ#u4ERn3O3A$_Q$wvSK;
z5H@HoW?u65dzCdkG}AP4?L5owVZug=x`NwbA8NaeLZ2SMGE>=aZ&%JyR49uIr}in7
zTT0(^H)r0%@wmo$qA{%t6qgXM8{z~a%}!=H0@y)70Te2rB2N@g0{xYWhHFHO5QH0D
z9Q?W_49i;z$km?e@j<>GtP`a}ZF!!I;&Mor8>V3l8kE+Ws6?_1&dXCt-zr%HL8NHx
z3nwD4(;N)KqNupKZ%Kg!-ox;CpzZ~VFmqTtqr2WPb{#}R$nyg4-n~avH(_1UCyd|q
z=5_BK`u@fNpq5Ly<rHqvwD8(QCqTLeJm+C+c)|n6en5A<!7Xd}#HMZ0oX^m&@Vjr!
zMV_>k=YO3RN{0f`C-3P}_Un4ozhe-;>#htDC@7;))D_X`C<SjBS5BY>#}|FtH-OD0
zA)haCSzp`UXzqECCqrRfW~}QPk_+=ZM_HCwt;M`p7@Z?eR90z^ye8iJK);dRhH0MP
z9-}$xhcp(8aU3?_EtMmZev&tCFXQBO!rJ<-k3@C`$P5*&tm<NeW4?M*0az9)6flk>
z;61Yd6lmHOSzf@&VtX*WnrdZfUp@$(@b7Tg<$yBj3JY$n#r%u?$-GED-Xjw%{7$4D
zZ_acrhCcE-dG3#L^J+jLiIToIp=5r57;X@W5jC`y0tkddZE(5tzsa7U<ZrGNe7h2@
zFf&3B)l9oHH;&gX9ixOBtm^_OVzDX9oGS?>te4Sh+Vrfn;)_CI9=Ql-5=<I{feT9D
zPv-Z2LAYl{wfs`<DtLcb6gOSV{dr8+l7#0KZe1an1H>+SA}h*T&NPXE*h(pQmIG40
zP1f^vyKO4b2zS*{b>&CxO+sp(Cge<^x~^*|LvD;gQ4~yErZA2Rp&$pPsU98XG_pLE
z#zN!k4;F0?0*Z<+F;Vn40H9IV4VPT{%7<kWw(mRQf6fe)0-Cl)%S3k8K32z5Q<Bm{
zdh`Qra$@tOB9nDhv3l;q&C5NI334Eq<r%&RKJpc=A!c9nE_ryulGK;`QEF@QrCRGe
zN4LHl{IaZYYRRqd{Qd2^$0t*>I+%I<doim6rlsg9ygVPG79E93>}#H<aO<b>eP09I
zDgbbFd3M(u=5d7NIp#%$)44?+ST;E|N!Dn08H?$*x=4s8P0YNqh<SNws@J3Dy$?8)
zSTafpE9U9Y%zO0QN=XIdE9QxWr`8H(RpE4cLRnS;Q?xD?J_!XWv_={~Cjpx;FVA5O
z1jIj|&##)Y`}QUwH4Z)Ox{!ocRWS2}LDD*hEGwXl<v5tFMsU<*(j;Xiy_F<%;V3~9
zIQDc%Q3atgan5t$%00K!>rFPMTO-3kkrz0%Y1Nn40nK>$2Hb9Rn|Myjt{X`Y;ZpL@
z(cF44#QhxHKMRK$uyQ3UKq(Pm^_+;+oYHVE9)cs6aGc*X^+vutmSx7!-7qgBIAIV)
zN!3Ous(d>SM;f)fwkdYWf0KHK(|#Nh^EN5dC4DZ-g6s8yX&zZX5xD*I^t4IX-ag|&
zQwo;%!N66CJUqMMPpHZUfPSEg-*v?%V~4)K1>8<B$-1tgjV5#c6-We#fxRO~sITiB
zgnFzqxXpa{lIN1<?25i0C{#izWE{F6jxQ0K5<4%vV*m+dl?bVE9Psk;%;88R%u<g2
zR7iiRy^*SgqD|*{!m`Y`HZ|VAe~&E7nDSwT*362?2j$*t6q`((faPfdC=vcAiO2#G
z1~UhD>;eLor-7ux6ZLgP3F-IFc?^AzX&i$Ylo^BebRy0o7|za*@7X_-G$K{&Fi(^d
z<-O%+J0Z{VsA)K#2p6AljUzmFG|?>6<m?5D%tgeaQk;%{rz49bYgVLDsqmRfZJ%SS
zL9x-rn$I{*!8-~7byXuT3Iq-)Y)yLUFZ>Jwk%zv=z+7PfF-o+j6UwR#cb)U%9prIA
zfk1f08jK3_-E~*MIcN=-EWb;F4KsFlfXO`ePAUn}fiflo6wow84MTycG3ZqU#o;dV
z9@9ACdWlrK-V?h)%br0Q&qjdmu?b$EWf87;;b~LIB_68CgJyo%SMt2Fp9qJ#!FqCf
zQ4Am@%qtZcJDj86D}W1lxagqaxty7?^K!X>vJJA3Gk6rx{iXIMA$2-E!FwMl9fqz)
z-*v&ITXJzp4Fx8rJ-lT$c!c*!xphjRARE2d@;kXG?npMogDe7Qnnuj?7=UsD>52`7
z#QyMY$3+^#Gz}Ps9)Js4RMpg|+ZHM_d|2c;j^|1#6&NJPe&j$>5-LsGqU6wpSb9ss
zAOswZkwo4*EOy4ooKv5lpYhXAKLs#Sup*ic2M?*B1Nx6vP)6dUEa3+ZFxtZ?3G~##
z+@zF|?L29%V;OcuL4Lz9V09~v+<8<b5ziT&1#n;1YCqp$Jv?zI4P6hvu5<vjM%A=5
z-^PI1mhU%`0=y^21L>P6iZ5{M%vk}%x{&hLr(@X&ECDD%f_RxG)_TO0QYgw2b=yE^
zSxn0IwC7)IcD3lPS1ePSNZzp4S+d{tK!AF=jAhQKs!i8*7`l$K4HN<ma9-uXYbzSC
z^d(8g@XQaT6`Z%M>0hy~3o_30Br`jSP={Pr2+D`}_?)xIK_=fkO?ZBO#_f8+y3PPJ
z$}+>#(-YBYCp6w)+ujD8YlX6`kd+0|qh>M0o=m<qVhqkgAxkDKPJCDZwu4jdb6{)c
z#loXoO)70^KcYohYlFG4PeeXyqC!iUfKc~Baq3GODO)8B;;1{(-T)2&?K;MxC>gt|
zE1Wn0bW0_>{SL6E!}B}#18&zVo}Zub;m4ow(~m!H$i-CHkPMu&U`fORG}D-9x&1*l
zlT*p(KbT(`n-txkIpVoqJ(UxyiGYyB?M@5WHo>swS(0j}aKp>`X9t+(ey)R_=p`_c
zQZ7V5l|>mizEX$G{@G}TsA;WXmz8N|BD(?612uf<ZRcyZ%Yj7TIOhV@uJo1GWAwYf
z$6a6X83~7?v?IrKJDD-4YY{5h=I{5<mOi(K-|&=+0IRApu%;fF#rvRQx{z1Q^NQ!^
z7kv8k3DZ0QTA?Tm2T#Ym-M0uR;I9zONfMZ|rZs`%d0sHrlurijptU<Z$Z{)5$bo?6
z-p4$@z3cRmfRaNqO%rJVAhRXc25nWN$d7~Kj=dZ)kkkXvMnf5mx@o9%5O$`^11v(c
z&93VhA0yn(u*=%y1@Rjx!74~VhEx~@c;!Pv=`yeA`w`d64WB-I#IJw#E6mdbqYbQ!
zjGZ#3m~KK}MpNY&Ch}ywF#g?fQ=ulF%#T(Y&H`OGP?Zvh*=<l&F*noFfWoE^gGqZ=
z3f3;@J1)ct3wR=)OApAam=S_FfVGbE->5E~2auTz=kpVL2JbJ(bS%^tWlmUY(cP}N
zTwX#Tutb7yBezolwcDG?-ikq%LU$Zv<0P0=ynpGKqL8xys7S&zjf|B#KvC8}T~%n?
z6S8beS+kkCL}1GK6C7$}SOWmEJlm4%-7nppIqRADlFDw~;F<4D8I7l>Cp<m93#`7P
zm49hn)AlAlDLAfD!Z~o(dCbeov@KK;YC|3RKw=8|FPZZ48W1DYfh|dy$b@vy3*I51
zyKr+m=QjQ9F_A}O3LNUZDYem*fXZ|fdw1<4p$U-7dKd;+7qO$JZcvp6!VZWy!ER%%
z>x%x?VH!rP%YtbfVYI>b-+hm!xetgV42?`6I*dMOS)yrqP*|sIKnYj4_j9Q5OtPk3
z*WvPV;X)BBobxEl3T2s;q(a0)6b|J$0II6e^&K%lcn59tU0&Fi8^tgU15x*`tO09f
z(6%QwG0eu~O!2++OIjKoEka6yNuw~?nP$5$@be4ZvEWbuLWH^$FO@#`rk*STt(_z_
z-#A%b&I<{sq9}GBzMrS9P+r#+x9+wPo>`XRe4=VVDc-x8ZSOZ915euiFbud{FD&o~
zzdoPe;hS&1;}kR%G{GYdx!2m8tHto05AOgC#}UR9FzGB6*fcsL?*dX>C2Cn>p2bqu
z^!tKod@~LJ3@&C*nnk#s!CHV;0ob&buo3GWv@s~lDx?DJ)wv@ritwWj%QO)#q7<jl
zG$@J^*m(x`9;*m86Q_57yJDUOcpC#rCC|@oc$V2?;)IrkE5dk6IU1&s%C>u|%2O6O
ze)-GahjR}9{4f7`G9M8G@@bljQcy@rkE!snP&S&aTG0g*=9WD?BH$!)cv)6FJ)O}s
zM6p|Bl=PivS-6vENbwH7H-*6NZX(3fFn|Jzl5jr)kqMuSxJndvK@;M9M9KPJ))`bn
ziZI0H6oW}=T%Mmxgr`o21dkYeI80-RA*;#)?fJx|%mM7g?y<5c1!J$|;Q|#x&59E3
z(-U->_JX*lJ!{ec5PSl>KfC^ho~f^uQZPEh)9Hk|suB3Yf>!A71Xb80p27nt?kmQl
zHBF1=bS4~-Zka>uPF|PSmSv3<jJ1L>84C8MN`X*kMKIAbKIY56Z*Pn^m%KTe3`sj!
z*9GgE!C6a^OcB^8Id(vLs7XT_$B`q;<3Sg0<?vzheJ6Fj_e4GGNZ3(+-kRKcO~`CN
ze2$&KlpAjxCn!+zsnP~bU8AWRzKI1BNCp?4EgsByrW;DAVW_&Y38@pYkVTXh)aSkV
zF)uTE2B)>5pv-T+d5?GRfzO|xDd#N6f~^3JW0FejQJ9Iw6wJ)K>zB+l(S`u>z<&7f
zfp6{xc3n{xIhwkrJhSle;&G!EJcba*h3E7VieUmTZ=M%uPiJCE*vy9f%kQ=U-Qa?c
zk(Z0irh-`)@}lHq->4#POupSIUhNXGBc6sV%ZMelER)HL>%Lt%gx4jIk-XLx3LZtC
z1#FWv+Vs-hOg%p#ODqKpBVm%(E|hY{UhAfAV6-{RzTC9rm(#?Isq-A3hb~d7VC<7;
zJ8J0#?VQ7{?{K}|Xs>5EdsELFY8Y*F;D09!dwKi;yY?n?Z<(j4Hp^?v5<Xnwr9Ao7
zxo88mtrj(g06<awqyQ!*=^n%t@6>Xoh!#Zh{G^s!W-?+$)UpNdRcU#@G~!Ud!LBQ1
z!~-bHA_$iK=STd94|!tmk0W_JWgZk1;}0p7?8c_sjYoyK90wBQnbOFZUgq2HzQgHs
zl0OZ?Xlo@>fd%b63|MPW8(rjX4vv~=lmPJI!v}o$@PSqN9L8j*>X>IHhuY)bd`jm$
zG7@w_6Ed0MiZ3xEAmT=k=gRw7ew;vL6H-;aSoSHI6!LOW<h+<aV)D1`38&LJP|?IW
zPM<LqfIft<$ez-S=QXtiS!P?j4yo|OW{(3gFm++?C$MG3{3_Flhb}l2#VN7Jl35qZ
z#JO+;n5Rv=bdtQit}C?BA=>@SF=;ZF5yNxI*VnZ-X;NKx!)jNQWre)R18AKJ3DFeH
z18^*jWs*&6Odbkgib@FZVu>b|0D<zkLarwg-57(mu8|u9yA!6;j?BEVAF$2~#SbYy
zG`q?ihxR$53Jt`7#Ce%Xi*gP{UWUULxwN(hLaNyT2pZ}<O)yHs8w#8|ozGb8ifLI6
zmRSY-g3=Q-GNX`b1!%grL(LBlI2FyfM@=pfp-cX~-}^n}MFwq%zPD-{s4NSVFQR$L
z`QJ~OXsmtTQ!x)(1e7%q;zi1o-E+xoD?k#Hy?2-oN-I?*F(qi7<_hwAcLNGJh~8tE
zMqF>#km@CsXh=Yr;L*L?1FRjA8A-3r(}dgQ3cC=8nNkX6U88Q#Fj?fl7o|K3UDk(0
zWAo9J3f?z}m$5~`@la92d)yQb(vGx&=4f|`A=^LOk+722r)k2=%L~SFM5c+sp=oNo
zd-o320RwC5-P`bFeWATcNJ(+faTqX7BZ{&_o@aq}M5srcwNM#t{xn^}ImhpvgKZ3u
znI~0G<slw5d0<L?8CheYe#&ACmhWSu(x1b4#t~iL1(g&EVbN%FlFhysZqy$_zf?)%
zM&+`?O>8!KN?{(-=TrMWU5G()%)#xHLRFUVTH$)@Hd;&Cj|fbyGzH{o5>iToUn9eZ
zf-hCjlpfud!W=ozGv2*>!Vf=uk9SW`6y(=5D9Y;2E)bBHX9nrv5#qUNn#k1h9z{uX
zy8`^C7@z2VK?|3(YcCPfTr*FBalZ=Z^J&Lj76Vsht>QRoYzvQTU#3YS)TO0+Ds3qQ
zKtjFav@Fg)Eeo7a=S`Jubs)u4>%n^pa_u?DRbH2}TTRnM{AoMQo3Y9oIPcMS9Se1)
z<5sbl#oZq8*qEc+XZD!qB_s|@{%&2@czSx;jBP92RD4kXx=>Bqn}k$*I>UQ%uY2n-
z-+C;W#<i~T?%h*R?N+;@%rcmOBhZ2@^Mal|UJ3vlP!&Y>cC-QHR(5;~YEK#2(>!6A
z26(#=HLr6hvK(#G0w$v%vj=k^{Zk#cDWnw`KG;Q3;?$gwnMen`!7-AK6)5GzdxxnX
zFm|LNIBVe@a5^<8tK#mtwgwDq9vpvQSBrTZiB?#VK*{q0Ro%kDqwB}*nD{@EkFLB8
z!I{Qn(ApB0w`OJl=irHEf!4+eUgZNdFbQm~zN~<vDDaEF@dJMN#rHU!pKxkUDD#5W
z!^c2B@OMS5gd4smyIPEWhiMp~7!EMT;MAN@6*)8jn;ui23QrMM9J=Bx#(u;wjO?1>
zA`j;VWtp=^7Q+ON1WO77h;|`mr<5`{H)k|uv%%#anXqtl2{^Z6xOEf=tO&a`TBB`i
z)J=6K2;{6P$eCsHmKUZU2K3z(-Y%5jt~KhaM%A8}!(NPc68Ogg1@N9ibo%asW$B?5
z0poR3JtWuDAZ>dq|6I6`uh(aE-33Z{7^9Gv6;4m@HvyaR^JT9_m_$NnzNWpg?jFU8
z6h(n$nX%3@R=eVQyWrMc0RVYXGDo-LTbt-*op%_RnIM9M3RPX9$a6rc!_DB6&xWTk
z{Bazx&ND3xz;RnOxpYM{`NFf17Da^fvH>IkvaYt}x>L1xFKm?3Nm(jkU7(7{k8qLT
z@jM*Uh;bNMJ7j>gZBSJeh67&cRB&oVpnq-@@=PJmHOfrEh}q{otn;wmV*-`33SiZN
z@+!Gl@7}$`58wX~QcKFRL}m<IP=2#^@?c7d#z-c+)VZUaHM$Vm<`gtm!Z9B|oAgg+
zK1E1dyF$VZnx;kDoI+Cb{RY@muh19;Q?YD*I_HSTTL|%X4vC{C?s^F*rZ60eYigP{
z@L=7cFe&oBn|~>NKF?JC9iI8vPftX9O_Ops9vQ0$>1A1Py<9MiJtVWBND6x1y;q<E
z#0w||%Db*(3Se>y38r;KQG9r&{o8=^H-;eigg1&zy;Z<wet>zNu`CO_?PEZo#0_yr
z3{ICQ>pCRa-f2;(s>rtbI1VAJS}0BOyx<Z-J5wrY`wbRDAjrliV7iGeQ}2$>?PhE-
zbfM>Zem<03m&3f*5M!7Ck1P3*X_|vM1Wm!6by)`w4sb6l_lKBdj)PGv3caGvG*$S}
zq;V`(V_rRs^H@2B#&hMEL(^E1hbzl6N+=Cx5>t;r-wRtvY!(2hP)=;-SRti6ilRzz
zNW6Nhq|ds}6T0h_l7~4uRm_Ym%gO1Y1jgKw;3c~u?17^gCNUYKqvH!>Q&br{R(o+I
zRH_Pf)v#9ndOzG56iDmrZyn|_UP>tr^$2o4IzQt5?E4<x(WPp#45#x`!0tACN=2<Y
z>OWz4A?|CQ3A-a4>a@|w?ln0UaQuB`Q~hh(n}n1cOxdJLYZOHWr2wsRiackpg_q2J
zc&saxHg|4iwV122HVv5-8xO$t;nK~u+6BYV!>%j9`FSJ|Y8XRa!ABcm4@?VqFF~3@
zFi-h{_wPNLho~V?_C6&LQ*lqZiH`!;!~zzK)O&<<9PxU3??O59yvX4#2ekTaiFFx|
zG_4eJ1Jt?4`wDnh8RP{WW(R{|_UJ|nw^}UT!n+kpLPwJ4lW|JVQD|+2qEG(nk)uSg
zI_E632C}?l?xZq)><$AtY6}-;m;^h5N1hdETgrb6qdgP|KKw#1+dU4IGj(rKd|kA=
zypNbHFSy-q5YBqpPqF^q7rUZM-8%GL7m}@&GOXc0p{kg5;VYY5pyIJqN&&{8VNa?!
zt~Spp5C>B61crGy?PwJ0l?+W%IYtEDe&45OTju@c<>jlN<IS4X(ZeXEFfS`SQszhs
z5S!gf+~?~GqbZx(7>%l~XaSS{kFf)4WC{(WCVGx6XMvFA!X|J(-eBsBiN!>ALr(#c
zqLCG4iK49TZmA;yC2d~jP+l9L7_z+JGV&5cs90&<bmPOB5zK??!F6Gi8$&D*VKZ@z
zcvx_GKabBT3K!H0$dp2rDV!>Ur^cWui3@A>4ma)Lt;aMoVVRF!KL-KTeKJhZCR7lN
za@++AIi4SK#gC&r7gZBdis(eDx=u=XKd?8GKILNRhYpr6J8cZkCklf=qDAj>;Yi@b
zxp$6O7y9nT8l($MqE+4Sv&v4m#}hY>6YtUWJucT9RyG;6QJk+9Luwv^q(}t5^O%Au
z8^P_0=Sw`OsfM_n^Q~;2rwQHdMxjHBFuj^X7qU<g_klUPIqIGZV_ea99hB0@G76ZM
zfX#GFj}#8aR5i~p&v<_M{BS<@?ajG+@`34-IM2)oiALl+nLX>UVi>tlC?Be<$WT?~
zog_a3sO7a~nJ|r14OdK)EYDF?Qay$=w=}{XV7uB&?7PM?&FF78*kysrG@P?+b~YTH
zd~ZV^HL}fbn5H3kH5C+^=8XJ6*hzSR2V&(ZZWOw3gSDjDc~~@cy&*1>i2L-pv}gRl
zfvJUoOe^GCA=4SOR`ABdYX_wP<v@V0c2Gt`=Q;Ay!>ta*>X9ugrfCeOi3-A%Jb{F`
z^K&TrOejibPBsV85i5m$8bi>gg5!!mbrj;FO1>hXWIqWr=A4wbYN{G-bAmEDVM>U?
z<H)5B#AtWBENmB2Qf=Kt)&;q2kH!{KjsT1Ug)YqNLd+4KQfSWSC$8lu13MlIfS`HI
z^AsvI%imNr)$+@VJdS2>>LZ~o;whb%8Qt}Qc^Jqk!8qx*lzcL2yRo_CGpq!rc0IhM
zD%)9B1TW*G#NoXrHS|5Mmlu5g^bwyweY*SHN9}E-cjFj{uEN{{fo|jW=>=nU!+7d3
zoLW?MjmxbAyeIdoqgzf2g$izP^nlVtkd5QSAfm^*kl9ccb%?hh6AZ_Fc|hq{)_k7F
za7z_M$z_thj&oUVjA0l!tsq9a$9R)tl{^ZVs5!`(rfCd0a#@yhVyrNUZf~b>TLmo#
zP@W>FH9U;=P#{wd7F5p58!SG9H#Mp@$G2^b+7}qc3Ek}i{PY3CG@%bSB5&f&K|dDY
zd(%j%D0rs;_G@Grkp`B<fe?Q%UrH%9^RAf3Az)pSW|SBCofz2*Ckb~EGcUv)@n=L3
zd7Gf?+);vj42h5wnsFRp3<;>JtnhSxcUP%7Emj)G6xVfKRx<O3A&g(Bb|d@7E;25~
zY?C!v*A-nK<DsAxGLxZg&#c|i{!7na+RuIt#-YdcdcnLbFh&Aq+c38D^Y8Ykk1<uZ
z+YO&TKjZV~XMFnfJJO_vp%0pe$#R&?Kr0RJIo>GKD2f~^SIA2P@If&;3ohoIy@LQY
zlfF@@)pjwdlJP>IsIy#wxJF)-h^48H1zidTQtHbXey5cVVEKL%9FNBXoUhpTH%v2W
zc9!v@s%>C`8Xe8EeLr#wi-tfOlYUQ}-X|_Yl(7I*9BGkY1dhxp2M5ERBr@GBa%GWg
zk4zh6ri3Y5<VAxruTi4J;thsX!7LsQFX-8oPh;DSQ>D+Z5}_p~p%12@0PWU$bmM@b
z9{`T^(ppo3HFK38t<#ZkvDRYfJB)n~I16P|APsDqW;2HUx}|fTg2+U_;^3VR!HiO}
z{oef9?<o>CX<u|fC}@+ReLAD$afu)j=8!w5j-u;|0*Wb9K`aZQaw&>3j9vP?vpnDd
zP9*mLhM~jlc8NS#P+`7gOj^gwXy4Lxn5GGrz_4jxvJ6#2j*67ZWiy^#r!rTUmkU09
z_=L|NKjZVq-;t2Y^9+)l2oSeaQ54V_X;-Gq@bt|)4$iE?d*<1KD*ooEnI}j#<&pg=
zjrfM<z<usG7cS2-G<6NF%stj~Z~zp5kv(V5S!hEGN4z<a;$lChd*D<kLAt<}Ae&0O
zdGY-+p5)!TKIPeH6Gyz}d_E(~GS~<9z4S#ub9ez-bE)=s-{9%H?{WVAJDl1Yc~t-^
zgM-1cI@rYqGfO+isWV&cOwOky7#m}Vx`qXj_s4=o1n-^-SXcx$rO}hO9)Z-MFRo&2
zG$^OCNxZO$v*dMc+y;};vZFvNg|=xzT|m;dq6R_lz0=d+z2~aKH@*aE9@-P7y(pyv
zOF?*+1G6zTWxK8;Zg&oc5pH)%)PJls?fVObXO{((NKI5jr_&jERsbLYMFQ+G2u}_H
zPNC^~x@e5xQ16oBy4x}BH7N;{{rveeK7RO!=g*(<@xuqq(|GrpkJ=lVIVopso+hlT
z#XOIg=LsgG`tjZo=@zAcSY%m_%;a!x;TWru?Q~H$w-2^=e+yW;EDM}n0cW9sz&GWT
z5BNy{#*@B0-*C=aOnr}ao~d$*(WL&enRj#>q&INX5)4y<3|N;H){#fAt{b#<MJX)2
z(c=l+nxQf#@w`ma7&duPFuP(|;6TP9&5I*U2*mNNfZXJ0+Hdjn!!Pml!}mCS^9`y~
z3zeI|o4t}`fmm_5T4A(>Q5KnUM2hEG03m8dp-Qj_Ek^%zGCrCkk9D=^x&hNTF(6Oc
zbKRbh6(wwZ!)BfVD!mR69-=hSRNpQyn5T)ws6tsbXix8;Oa|)_1DDe^1efpfq;V>x
zu}lkw+ZC2;Lpmswbw%t3S&kU|iQRq>dSDj>**q-)J5t)idxNTJ(4L+m^M;P=zZb^Q
zT&F1f;l_AmMvkYduF;&%RABdkyE?rf6aeiQI^c>#;!(`Y63zLELxSY_07@^yW^bk~
zO`!o0<Ds6OaB8>Jb{=U{68HD~{EW|^KjZV~Px$odBd)jWCa?EP?QJ0lnz+oftbkot
z%nNz#>`L}@SC@FdykjyigR-blloj&Q!da^NYm6{bMnUz+)hj$vfmdqB6BSUxIKqwD
zM_~d!iP54|6{QX8CVI|Z1IcB4N)e_t_5<9qLIW`D6+E>KvOAJKAQe<J9+Idce8ue0
zMzY$8OG#-RQfuza7Zx;W?{p|Y-ZG`(Vc<M5h&u0}Tof|!lysacz-yo|3Y7th3@~iX
zk!*^f?9o$&X%7NT)cPCx{szH3Os3GZ4a%wtp3MF1@I$Z=rcT#+MSr_t=A1R96f)*L
z&x<mW{xhe~J`PGgM=J$q7xL&117!nfjl3wioLGLS_3nlinqXSQdrsrcivp+f(=Ht+
zAb7jJ61;~mDr;9ZzZ49QD@~y}YH4K_y7ai9Cu{f$3VD{J76L&YicHgy6_c8$!l^|`
z`zEC0Jh7I)f1dq076Em+T=4QjY@HuIe8kJ;MghfNFQeWjq^t)r&%&xO$crqnAk6cO
zVH_|`6V)|S8OpLj)l^tli)pb?D#k>m516)8Af#rVXP0bMRY~S|szlS#5fzr~ZVkIG
zAqrSPe<7?-xzCTrlWJ4WI*eUUWsUe=&yAsij&(yf-cHuuyNc=cg@YR<nN=w=o!f>?
zPYQb5;vwGqSg3)&66YJ=%v3PlIx<UXGq4V#6w7oEqX44;V-%K5VaYTqqhUe>I%!J|
zJ_K*3Sje_{p7ZE|b6j~?aHzI9ZRYt=+JtyOVLirSK!4k^S)NWOv<dI_Yb}~>D_07>
z5eB6-JeyN7wue6aai8=)iTR?)c_xZ8L=?9-e;$u(8)SLA-Jm$WF5`;d;>3AmPVXTJ
zDVZjr3$j*i+u|f1Vm6DvFl}TqP9t7kp0TcE0cA-ks$8gCaBteBB$YzYM;g>;eEReW
z&(|x)Wu|C=<#!ZuuB1=`tJz9X5M-7W{M`Uv!&wS&Tb2pgnq#%m)<4{w(tl#^3Exz@
zAXQz5;taSuF!8;zcq?moQ*s-Ny`QpwJ_Otm$2?9WZo>_$Tj9Ke^9~g=Qc?mk64Ud(
z5BeU1HUnj=V$Pz|IRuh=AD<%?a%lzUbj0L-H!xWRttod+X+oNuBknB`Hb#+l1E-+9
zLa8-!1LQ`5OFD)NAYGSKU~eZ%HYH(x%d$W#(u(rDMBSdC&32nbyeHicB5`T1g+pCF
zyJm(|iLLsQkkX1s7sqiz*IyI<eI#X{q^~1@eA}ds-1T}zT9|<V>F9WQ6!x(1D=>D>
zGc;c!WhvS{3H_r>K|VJXV3Z_(#j-9G4x&7&s%4Tz6Zg!~vDuAF!XWyd3al8dHkDrX
z&vk_5${rU1_2I(@eE$47C<|nH{jMU;1sM>DtfF0;WhA7uHstlw(e10W=5o?bMDV8H
zbi%#&Sk{Hgk#n6h?;Wz7%Hqh)C%8iF1f&5ZFD=WAVHg9Ai&B)TARAN|4(>Iccp&B0
z8^;05x)MdQqmt`IQ%B}Us{f5pZV(4TnK$P&u{NeOw2qz^tqtKuz#Z@^kLVQ_PDs9c
z?@^XDo}Ru%dwNIJXzChSZs0u=jVUtVv?ddXO05H#;ny^oqj9|P;O;lJQVCt0iY3WL
zF0bT0#jrJP8_c}aG(9darl7pTx~#a~Zs^#YG9bPq(R}UheWD@80f~{Nuq<<6Byfx^
z6<nAhXk)x@;E$kf`3$LG+z}4N=5STj8)CPx-s!q0{c2{@*cLyaIK#TCHcq8Ibg`NH
zL@*FAedC&y^t-C8@bvBpd9h8$iR+lIar~XN4g)&`#Of$mKt=i+ex?UnlnC>PLD5&=
z_i&c5<+1B9+%6Ah`@X%2^v+rZkld#)PvD$&fb)R0%#h$>VNPc)VIAI*7ADga77=VK
z%MD>M3>_9rm44m2Kq;VVDwsT@$ZDmqdK!s*Dfpo20Ycq+!@SO9+h{5UPn(^}Nu{RY
zelw)NAf*=cSmv2?6IY5*mk<ajc+CNhI5<Rb6yG(^7`vW|Ich*@kGiUHYR}MF7Lr*V
zXUNKd-uJh0L3_X`g>mdKOjmg2u{e#@dnl7p-k4T!io7~TDP&3`*9LiJpferns6j<0
z1<Pxr*n}o6lxSa^0!6?p0N!C42lUq~nR(8`D=JCVoKDz#kRD-xq=^vk`PdJ{rLR0p
z#!25brG)G?Oz#|p0C<X=CsUY8HTB&ML*IpDcVmbUugFXCR(T5k74`!OO5I;l%MyjK
ztV@;~t{jsCI}ad@jmN?;fL6p9;W%=eFjCs9S&`%X?g=HEjP&te))oaH>p6SE12B1x
z_WTZcS@Wq7C?M@mppy7+Z!N~|ireKvVI$DU^Ac@)Mst3m5?(u4Bt16{4%U&=<mKfB
z&(9z5@zZbd@#9BaFGK?Atj97==&qL}e(uZdjlrB(XiZ6>bQ{`;iLhyrlGh8zlxA@*
z>hPo6fj)y+0G!adcnWGOnY)OzGs&HcW&=+(C=J1dT7F{UnKT|I(R*Jr9h6c5K%_zs
z^lVi{6_K^tVk3U{-t<-CsHTlBZvB`vS_ChS2wBPS+>dWRR>F-mln1)Y70Wb&IL!eC
z@?yd|z*}UBI9j~7Sf?37|ANb>pYZ(C2YmYRC$99txA<-@;>HY52tZ+B1_*L_)4NJG
z&8TeGW3xs2RZ0ad^Lo88en_tOrf$BvxNEAB54WDniAj7BnR#VZam<nfxjtkZ+x3gj
zOQ{@PN8IA@9=XZT)TdyY(R(+mPwxfeh{^}xT5zZopF5pSA;<4)no@L;^?irFBidf2
zDDuB;PmHximHr5GWu$@km&4ficzJmV>nP>m1c%zs@BO(^TvF_UPoF;X*E6n{TaX9S
zM8#42T?xA{wYLH1cEvdKFeXEm<y5JbGtr5WQwJgCybDa7N)d&KeFf4F-ZS>*dFNr5
zl>+zbCaB7!2Fp#b_YeVa4(l{y8b-i6WSqHQmIYN~<;{;|;f?@GKw><un5KaQqIJl%
z!PDspMOGvR(QAza>oAHpkIZEvzL%g)6Jny|vyX)55qPZ>X#$3+Y;IkTep=D29;<c8
zu%a*u^Dtn!zQ7L8n8z1P>jl#=;^lLP?q$Y&Ti{kpGpZ#1D%S6l2z?bkm-oXsU|AO!
zLmE@4qMNV??9J%?_yJ&^mw;&rzEd~Er~YL|f-S~Hu~kwjm>dCRS)*mQK4Hj5E_Bgk
z8C&um^EhJYZgA^DQ3WbB^Rnb!X^YP-sYMR#F}q#hEqR9(<D;qh`DH=nLlpS!a-qT@
zT!FgcG@+Q6w-roxj0ZfSp|sUB(S@!J%4&;4r3%K~Md=)tc}91;;^p}>o<DuY=Z`P=
z{Co)k&+{~+zuwUGJ*>Suxv#c238|{6;o(BjgL#>e7df&#3llKUGnRGbi0H^+uC#_<
z?Xd(^G1X*Q7Mci-RIck9MK1Id;$=DpBsX3`p#=euXE{#oiEQ)``ath_bdbnnLN}-g
zTYy<R>z3FYzG|*x4L-}oG?mh*s|Ib;hO$O!>{!sI9`%@P6irzV@Xj%#gvIUoGmITz
zuO5260@DkW9Rb_J&o?N{Sg`=B2Nn-_4FyLgdC&<wn<^3K3nrl=t$7}V)s%um$th87
zLOb@XrYUwK-z7CjLNWjZ9I7mL=}HH4p03@xvYF4pg3kLOIJ2VQ!VLHCrJsS|VFU%+
zUN4t`Ap%6@(zaBNDr5lYqfNj9iXn`dy<O|N!0=fqIdJziF7|t86R|<a9?uL_9L|dZ
z=kq%bU5H##$NfcRVYQZk@}Vb_SX0d7)A?zm!S21_>CbeT_jrDO#`C9-`1s)`eER$u
z!$|yU>#`8%dq-(x5FGI<+na<`-L%kJVOeJ)y<1m|qlKO|@|=o<T1SGyGD`uKyK>3o
z>~e-!?Hs3xDt=fC?}!qZm<fu&5Gjj9-<cIsY(gk_=P^tZ#%Y509>!?ONN*}s4W&JZ
zHnyMaz4vOK7L5JCMEaz+6;*}ye1@X(D;%>V`c%A`NQjb|;he+JcbLY3(x0`Mc}+Ng
zaYG&TbFX3x79+Kq<iFE!!^=kpcl!xrrP1dzGT)&<hpN!Xw1amJnV|$xr>TG|l!FA6
zs^oJ>I(blG%2?D<0kDfWH^>7AMKNO)SJz3ODSd_wo(2-G^EhJYJ7R3rG)KZn5R->#
zW2ZQZn4F5WNC}V~h7QhKz*8V<(=;f`BA9C=IGFJ@X`Zzn&BAWzJ1#a#aeapK--@c@
z1Z00;7-KU+4PwU_`T^rGglEX}9Ou(HDK-k;pRpfW|E*`-dyAp#Fb)IZnoxv~of}lu
zR_oaT5F3+SaZ%E~yW(~uk7aH$D$aU7p=qUp_`UU$Qn5&_ftMEo%0GVkjOXVU4BY^4
z9h|l76}=E!2rpZn84PzlAGbG_wDQm<L!pWAZdq5%(@Jqt(}ZQ2u&xD$VZizPgjgco
zLF<fd=`CltQVOdj?Z>W5IDA?Y**|5aOO$jlHuzQtpeQ;>d5vXV(GNXi8e|q0WsbTP
zck_0!fUR(n*R8b_x-gD3(T>;&np2A`r_%9qSoD6A1`io&MAJB6o<^$nEbO0W;ZSFc
z4HV;abo<Exx7G%8xVzo(`O{}ANyN~@vP`g-1<UY)DF?<<W2rov(xJ*rSf^1IK&~|$
z44E4oAuk)w`9vrtb=<_$!!SDBYPkmAj1WeV=dmP}m&LgkvZ*;kn)k$}*$l5FpRz{P
z)X*jiz;Ch%5rlcbBI!T|ri68dM8!8Xr@feXnFP&1ZpcmFAP~9l(cf;MiZ+VVaZaeJ
zhQm8(ZjKZJ-ceL9&pGt{g#Oltq6`s-iuLmL8fH2udd)ym(uRIu2}s!!WnH7JN<gd4
z`HCq$dqGN?@X+_TTwkb6mgGd9PH0c35S)AzHY9kn;7pgx1)n~D#HUZ6@bU9!+-@B#
z*WjFn0o_fiR1?W!nQ1(=B`(*4$vbFo=93Z?UMYnl*T_r>=Paga;9GIQ_2q*1Pw!CV
zIR|5UWcht%5}AN$9I(tIv0-umVpGQ&d=m3=rz5gQ!-H<*q3^IxGnWuD%o=z`QB)D&
zd;r*`o22i1u7)dh!h|0vR$}EV@!qq8DD;+Pp%4h`h#gaDG!0R}>MUn>`WA9M0?DNu
z{q=gq^Ydr?_Se6`haZ0ofF#dm*gTYs?66p1Sv@qUzM6x_nki%+D3HNhhiP?KoP(8l
z^kltIsDfcuPQ9t%JPbqd&M2*rWd=>X)eMiYmA%$V0gXMTY2tzt2#K}jCgNJ8N7`pJ
z8N4U{`~cwdTohGB2QJH!b(8npYpNo|#&HO7T3SWC=X8F8dJsr@FIecckvW}+!InIw
zbzPIDazJsi7vl69X;c)$6mdK6DG0c(TVz?WwNSP*k5o`8MdjeGmn)Weg=mRY(}Yq~
z`*Ec|3ydS76clbi%m!3?YoTj*C6dCvCt(joR6<X6iEmCd{^DSyN9{irpOjaWbZD{+
zCO5Flf@P+NQi?*KP?QDMbw!pHF(EXWtZAAs4x}1-?=f48Jj;W(Bbs_gSQdB%73O*5
z#AgSksjgYo)Wi`a#A6S%sPt#SkCtT)+D%auXqpzr<Qt(N0Ddx<&>yM66IJNd8MI=m
z)tb^ekl<=*PTuD6SSSdG`pffYeE98e@t1%3=R`NdxEMRq`59==Ru8ORVRjms1*}ql
z({M_|8U<g?fCHA*Vs*gkRNy2hOmM{thb*J=VICIFQ#mqWlE|_Q8VXfeqpn+gDPcAJ
zITX)i%ulFjh2`-Gi&1g~q{2!$EGucgX$qCe3zTh(tSBLQYSIs6+a$OOHAg|IJ;fct
zdpJ;>R(n1pFUt_ooxU#}tNc!b0$lssg*>iW!Fj5Ne0ujp)sT1ZIc_9ODGK}rUDp%j
z??L;!mU8XU%1F(+D4-I{WnR!<Z<xmk$^$wxK?tX`Fsa3}$D;&nmuT`&A3x#w`Gqbo
z=cugL^@iK+60umXaB3=i_w<BQ`*jiez+XW~i3Ar7!g21DHVS!BlEBIgQEC)foFJY<
z=Nz)k1QTYI$|&A*(jnmyc~0pe(Owtrev1Ps7-Gomh_O*<OY)*X%^cb?>5oh)X))<{
zf(P@tgH-b(FXIlExnR#7R3e}la~Q{o84e;-L9EcMLg{bE=J2B?S>(eoka~W(z<ZCU
zrzfi3JTBqTD`+@vuy|m09+M^JiG|~hibA2t90~(erhs)G{R}u|;XG?)1WM6n3|=!f
z&88Xa@)DlIT8pBni5)eggkN!N?B{v^?}_s%SG$H9jcFwKJ$I4>5=O>M1rKM*Yj(Ze
zLIrK56z66(8<6(y%<eQP>(&r$Eq|JG1>>@+X-p3^M2Ux@p8>nB=&m;m{Sa_IqL^tz
z*2Lba5C@6mVpf5km_kIzs>rj9D22-^XpZ;cJ@*&8vzD}*TL*6`NIJ_Y<DqF!R8-e*
zHLny41)^sE^70ad)aTF7%=7OEC?5t~uNO}JQh@hpstn(}tMTpo2Khk&h)3;BLdw|%
z@Hqev&xH5dD8`-u4=On5y#a-%5THrr##iJ<!L1Z@sg%Ju4HzaO#dGWptg410iv<+D
z8-;`r5Y?n}7V|j3t~0sH4V!tbq`pMe`;jM4)aha9F!ndVEf9L*wylv@B|XZ1K72gH
z$*Y*B8MoU7!$48MPHSXkfvRbtL+%sV=d^(n`V!c%>78|0#tHK<!mq@mmFGEr@r%EK
zWm)m#k3RxnZ>aOo%ENkK@j0ATm{x_WGEfuXZ-9p;Ar6O(=Vcl_*41NK73Nj2J{^Pt
z2^|Bw0-ry>koT^s2%pR|H0Lw2q9k9|E^bYlO4d;UT$tgCW$FW_7f9^x$}2_8CVW?(
z3b}e<Vs0;Mmw;PVZHuNp1u7pWrnwA&cEx6QIp;8s6D5#4Dx{#bLDjaXm>=qq8A}Tq
z%}4SK59qp{FEPTDPN%0sjWY_EC>rQB5_ZF_$8Z|}>!H*NV<;%Ott+Y`z@I^IMw-H8
zcuh!9D-U$Hj&l6m3S|JDQz6{b>4eO%0+M(Kx1Kx~BFZn9XFNZ%fco?qLqEV-54Uo3
z{`Cd()IoV-CoOW|eOus%_YI!f0xJ6*390L?!?N1IBjrHCC6T%0%~P7prClb!*0LoP
z=4HXq5A4mO=ya|0CZ~!fCT{wzNko%4&r@&_=XuUW3Y!2-i>4><m+xs7x%c2=5mfY4
zJd$~lN3AYsM5(E9uQgcqW(>m+(iDskn3d)B{hJnfs3^G6W1c1q1C@~%ppwYODt!9%
zc_VBUSCAFcJ1a71O$2mGDdsN+R=zZ>({Srb;a^^HL~txGE<m~j$A13&8SoCj|I5D_
zHh$0y#VmiVc@pN7`A?DHg2&}qewR-vrt48B-aB|jG$*&~mAz)P7b)#2)m8*tKQ{BF
z&GU@w^%g3}D5XMQL@yNg&R%07=YgypFfEY|TFkWoEkV-0I`BiKW#kVg2`Y@0fax)K
z#>l%W{7|P@-ag$$B7`^Rhx(r85!dSrmUV*C3MR`!z-COgPe!v5P@d<Bfbx$YKXSn4
zCtPngSVtAir;#+MzV9HoUuG2Eooamd{*3oe4XUz$H?Olir0q>YYMKauB`&E21q&-N
zr#xtLt#-w-ELheBc_!ewB{R=xI5<ob(Y&Os;oPM(0X3bV;5T9SYN+~RZMcz*&d}5?
zdnfkqzc)R|?2Jjj@KK1hO@pc|p#Z+I_&VCC!eTp+dwUv2BCBU@$QTpMm{c&vTPl4f
zdPdKf-Z=JHmKn~iaBfAOXLx#g!sT*}>=67s#t`36SygD-7N@5(nx;mc8Dx10C`|#Q
zZULNyTWP~}eUHoSim~f4^=ZJV3eJYx?S|X!iXVRXo^U8;qmY2cy=J;Uz>*H-T#UH3
zt3{S)P@qhJI#i67`BfWJIrhiZNLCx@ZiE9X<3m@LTi`c7&+!Ez8T(V<VWPzO)9JWy
z=DtO(OrokUA*mezioC-4^t6fGPRF&GGeJjG>}#H<5dKkB6;4lQWMvVe`NhPfet`X@
zEYEnoe#SKSP%7?AqRy^)zOi{<`kjFCiJdpkoH+awE|&||Wr4F6>pbChyP&%f9SfkK
zmBXni@csK1-@R{fs!No4iSFx35Z)%F{{4UdA8?h30!3M(u9}c)K;LszRSlylX6n0d
zzvBm_TSg+6yM7EF9e{54q9{-|5lp+^P-NGVhXeu7k{c#7vBXYUQEuK|J-;{eBtm)O
z;J6jwr0AxuiTx7_-bz)OC}5}$z3~i9^NhZ~!Py0QPTFAGHh6k^3Y-1@AbTEbG##Cz
z8+4kcNC}dWfSTtS<2VIxoAkFXD=zZ|?&TR^$WSSctJWIO8I;bDX&^Hi6(?&;Xv1nP
zR%=O9L9~vRWx<CJA2>+!4^R{(C+W8A1&TuZ$SENQh#UYD`kt%37}C~?A`c{nHaJzb
zH+tGg3vuX&J_Oo|nb$N8<-e)z+8+r|d0kSd2y~AZ3#FITfiUnWJYf58A1=FLqzc9&
z>;f12o$$WTcA$UzBno_BGZM^w{|;qQ9?Z2roHM1E+rPV^?>i2I$xs$0&ZnnP(dLmL
zPI0K)txE{(KI8fM8PhZoEzrDhn$DFdUYN<DDGk1VSL6G0fpb$3y|~UW_rI&+prDNR
zbzN|~bpSqx*F}J6MNA0aeESXlEC1Tx#UKCCAK=~l_jJn_CA@Q(rZG@wI8RzYEzzws
zwFj7AdZ=j;r)h)&sD!v8N{LGHslbaTplNCvaZd_#2#3C-#Ar~YJC&v4!j=4o&ikzg
zJdB0=_7t#3H)yO9+j3r(j6sB4cqC5>K9Ykp&okErTqxeeSyWX4=L!yskofzZV1VKx
zq6iz?dq6t}?-#7Ah4sL!fa}mFiM7((z1B&T%@*H%_Z{B9e~+dm>fSW6c^?O*Hm{Qq
zTh}Epk$O*@-+58s>D?0r><SU$omcPvVv*{;Vw!o7nOC<Bv=Q*ohf76*O|b=iKqa?E
z+^$#Hb%lZe_zb!qeExopz#d0yEyiJB-YUm|yF_HzG^em9j`~P-b6MM=@5v<Q@rUZ#
zbwvxt<8fj(`gF}D4h$3)#2$2|i8VLR3Y^YoG|b>|6tJ0&Pk?d>H~94F6Fz?WgpsQ^
zIcw4N9bR6ZF>&Euc!e_4`0mu=hjWYfCuXnB3XBV|zJc4|?I8$C^TZ`D0m&hA%;R(k
z=)PT7xYfd2Vl{Bf4CTPCZG*BZkQG9eDi}i2H~(8cGk66#Z;7-t3R}pEl1eR!s)~cC
za@xIc-eTx(6awMNQ>36!l@0Qu5?FgX5E+lBttCQl93~814|w6N%1~AfMNKOcumSIV
z5Dwl)f$12Fkc|6zo?$Ivby-H*#xPFs{QS<XOy?rFnBsH`hY-ZaGTO0Lm4Wj7U@eUI
zP}%@e{+rAG#=}U-$Vz!UJ)QBx58vVZbVk**D5@Gt8|(uS1<ML+ry_UMIF97)WEx(j
z4XV10(c3(q*l8E303_nMCGv6cfy~Mb%_(I^^$0f#A66CmVw@C?@&&I~ST2(VQ1zhm
z`Dw#G32bxM<k^JUMgdr-8Qt{;w{ZTW)*;6)5bHgFBjYZkb2aomx?2Y?{K!f%M?41r
zLj@<NlE?Mfe1KSe^E9Eqy}++Cbe2OKqB^eI_5qtL!hFkfQ{wlPW1*h$(}zzO`Y}{W
z8ipR1>oW#Uo(B}NT;tul3g3To!n=13s`qDPd4+ZH7#53hecgmk+g~Y8>fY6vJ{*++
zk_VfX8H-&ouM5VR9*j0BZSXA9cMO@k52J4oO(br3@e(@ciK@7$!tD>---6M0CqXif
zQ>bL@Jb4Nmu0H&F>`HiA0G?``F;FVlf#jY^vgcMk$DO9Al;42vq_Gla*EHnu9G8^}
zfNBk;hzLq+5>C!ptZs&K76uMlduZd4YjRiTnL(x%N@I|l93GB5dRAdtJo?3B@`xTU
z7OGWQ;)fr8U@eMRB=Rha^CQyt?is~t0gU4m0xh*6q1p(Eoq+T%jga21opUajrjk$T
zSX+^j_7uIXg7&ywCJ_Pgzs^Q8Pg;=|o2>u4)jrst5_fdHUN`h3MIn8??#0^g`)QtV
zxn40ReN|l0v*6srgUEvWF?u1NyJ8a1ow39;Say#;$<Oj{A?N$_=@UMF{D{w=KjX@@
z!rpr<ES$$aN%=7v?@tx}=KB`kJym$u)Tr_ts|ChYVXz9r@-^p++S|*lzxo$Z!?M=l
zcI&ZP@=Pu3jCsz8HIl*9tS~5wDuCrF0bSZ>`uRj;)^mVp1`maztWdR7*fITE;9z61
zIU-4#luzV|F+m$jp<%7=^=DP#01ZRWC4U$wEef=4!)D$VE1+Ba{v;{yu+9^fafDme
z5aLmkCGx6dap+_DQ%`}1ei=}lTD<?}JDi_R$g=`@R-&IrSUVxp9#h}bEqJ}ac@1k7
zyd?*QS9I$erMbYQXbGANBIpo=%P1Uaf-F?LiqDk_E)U9^n`fzY$SkYQqiU)^78fe@
zKH5>~0iGsKxv*h%>Y4&cBaX#6b)ra#hT#Fy45`nd?+8EiRMa)g463$9QCCDJxWkIl
z@$6&kZXLq~JXWcf7g%Hw6xG8E6JDsUR}8m~LQXigD9a1fr!$JGhUGnn!%Hrn8?7~#
zWy0n1g1)=aj3|Xd6&o*V&|V)jIU~dmMOf6!%S&+B5c9!}sAss~1-wID=6HW@a5^<;
zPG{sz3x^ETx?)}|)`juUuWWxMP3lpL6*UM9zFsdF`VQ`!_mnj3E$XU-JvCHE&x0t3
zfdB$W%w})!2`@AW76Fb;@aSxdG17Rp8%lmo3n57Ai&Df1U4|l|^bD_QR0O&cc0dIc
zs0gEYzK&eF^!<KoEGvQTgx3+6ZSo9tT~j6KbtQ5?zXI!wQ&ZtT{h#_z;+KE$OMLU~
zchFf5D2@N?fAAmRr%x~VgTM7J<NWRknewR047qlA{`gz`<Ny0V#&3W76J}>&1Po>t
z1MfvxBr#>e2_`{)d%P%#`Js8#K?tuaRUDa;@<mymv1V9C>|_TY?AJ>v6{7B?q`ikn
zS(HqXn*{+N2QQd-JmKQ#@Xm7i@f36iYG!Obil^I;MSe$IQv~2UK6FA)o${xLJ`>?5
zW}4#b(5)7I*N0MR9$*}b3)dzZrP@iUbew^U{dPrny96_}C<uqD>pG!qITpe|q`7{6
ze&!+yL;)<$96K)<I%4s3t6Un2*ceU?PECPRo1?4@Os27{4wEw&tit4h)vb4*{iyv5
z6;iS-WKsIQN8b&ACqUhp97S1?r%8=iR(q$Sheecm4^RaG?&CNF{v>U5h-FC+R@yvk
z0|7$<FwP^op(plFP(_uR%{;>afZ+mu0GyK7WJZ6xVqKOH!;+U38cwv`pbeXiPfiAk
z4O6hr;(F`w@^THK8QvpQdX9Jqkd+m_{k>n}-7kKLKmD`6z^YsP_%gvO3lELc4}XN0
z@t@%`EHIUW2JkY?$h^g=`Ud~?f8*c8pa1GF@E`uee~9PLFHq3X$Qb)TydU)TASh)Z
zz&1dCPEfaF(@;E}p4cG(Oyht-JaCHUunr!$RJh?zNTnBya}NFOhKa#`1P`24Fy3m-
zN_h^nRpE0i3)<CMT(4KG^9(TTP?a^>_RLuhuV~@2#aRdofN$->bD+WnMOdF=CGI{y
zQ#2yMj;3kC?MizIig>)5w!!&)LRzK!k;w_xyqzPQwYw33Z#b#_e15{|bcQi8G-rQi
zHaOHU;&!><g@x4T&z~`m6TBsg=b@)u(s>%`Ie0Idtib3p{N`rxaW*JES{O6HISsE|
z;OSp}SD2K5kB{)E<nBVKYRszeG%qNv!)ghCfCrm?2h}<l8wr~Z3d>3^PUkE#O{Pha
z=bPf5kD3p0R~{=fWKQ!$Ic|=K@dWn?Op*@{jeI8xmY!l**o4fI5Qq(kxx1M?*joq>
z>pg~X#I5VG%(U4l(!Is0ZD5Fx1IjB5{e(aN%irSC5BT-3e!}VeExv2MhcT2wG))VN
zbB%xLKl2@#Zy<cGa0|4va9Huv#}|C_?mPUq|E>QH{?R}Bhxn5}{Zr(59m~aonLe?L
zAm=qvo9Cc)l)8ML@e-|wv9s@bSh;1D;ux+IAC$etdp}?DGo=Ix4KQ7TZPRjmkH&uK
zgT|*KzM{0o>MU;E4dd98iN$MH*A3^tZ7-w!x`&afSZrXK#LV+3i#kNjJGEm-*zc$0
z=Xs_)w`FF+KkreN6&Ie+EEBg6x%KMC`(+#lstGu+P|83ngSI&lQDa`PMkd1i7>gQ)
z0hgB-eEjecKmPVdeE9HzG$u<<uz8#U1+rj?+GG@-Qnf(Qz!x=;S6HC2Sb63c>R|6(
z<qzAxP)#a5#8Q3uc70)bi5^v1V=fjz`4FKjrr~RSiicrh=0L4AOh#b~RaGXzpIgl*
zy@`e3ejEl2Ll=nqy!WVUsXnZs6yp{LV@Yp!0q?!{tR>RT-Xv9e?>DzRXB;C%I$t>i
zg2q}F6uOWVIW#m@>+m1_KmQ5-AOF!mg;fSc`-C6<;9tTo{>Cq{>JqDWu<HV44A#}d
z!C-Y%x+*gTlqsN%hjTOK+)X1sJzw!({JZ}e%Jzi+_n-VJGOdAm4(2ZuLhj@tHUqh|
z?iMW6las(us6_MG#fznLyI(he6TX>6fEuMvg+S6|KJV9wH9JMbb3;$~iE|YEtGOgu
zUDxoUtsJcT{=ubi<}^?eGQ1RcICH+BjuW(*n)_#7twne1FpNV`W<%kLykIZhv4Be9
z1EUm&2Hh}ER4l?6gR(3`mV!FK{=^|6{YiY*^UE`ypP%va^2}s?D_|X)-B(r^D3_7K
zNh)MzjjXJZS2gmy3cO-zaJHDbujmwa?O(9D7h4FFV+}#PRaB{!X+J{5wB9VB-+|5Z
zJYyUPn3iPJs;ba(PL+IK$WuaUHhL}u;V7w<*#YI&AUE*|T4y1JW{<V(TUz>Ev(_cK
zXCe$!56Pn@@h55azSf#t{WTHk&y&U9|NH+4|Kv~q3_dGye)m1vci*9`Tg=YGtv$S3
zDK<xe=##Cdc`}*;;WGtO6wp4yI`z<H1>-D!{hJ^02Y>j-`13#i=OGLu%qJ2?PT5?W
zLl!YZnrk0^o`L|6&(yZ30B~cszG+JZScn9lrm$DavgA}16Z_d|1OlQfG0osHPXqdn
zbNmz`K_u)>P+1h5kJc<=trQQ}UvCKxWw@|w!{Bk$GGA-nd8!gT^voFQp$z5HHMQ*R
zW9(=<wwN3}3=~MJb&jGa@$~eLOK#}^8_2Vy@loWIT4K`h$4~h2#~<<O(`U@{j6f00
zH4N>#AbRL?WJQUrs-cStSy987JmeM1*(8P2e`#9o+FwCPz5Wp!lgK*UE*%AZSqsY*
zV3dk3?1)Ki)mOo`7S5<%)&<&-fXee6O&xRmjwmI<I2f^5mxXKajqsjw^CYr(Uy*gM
z@yO5=6r6WKu!#AaOLqGMx9+|_Ij9lSV0vQj>1MB}it#*K{Jp>T@8X~R#cz<cXPm$J
z4$b)+WJO7#9rFaZ6@cfwy_|x}?FwZ*Gz^Noz%-5c>9;RfhXLKt;g|o3KSr(zf9Y=>
z{tN%P|2+P|Klu9)T_p-wwOOZ7w1DR%HS^@}=4ryMV_qjs@1V4xcE|L%-AAbqlmj#j
z12aib#2O`z6Q!{8Qc9h;^^cloz(;`OF#+(2ey4%SsigfQ%`W}7Xc|3Vbh7>pbB!63
zMJx=slP&wbISxJgtIQQ~$k%Oy=G4OQeA}H6j=*~|Ukp76fNzXOo)HvsK0lE#QOC5~
z@?4g&y3e0K<ENi~!f$^4Yy9}*k0hWJLY4B{?aEc9DZxI=a^z)&yr@uCHL|LLHiQ#K
z!B51FVxoM_1)jDy3aPjN6v+0v=gWJKYd>NbCoJp2ft?mb!R3QIh>CXye0l(uh0DO{
zAtt!V4eGi@RW+OUr4vGnCn_cZ;p5n`4Q;{lfD>eMHf?y|AzIySurjR-6l)1%*Hb`b
zMzk<hQ=@7dK*!|JM@=cp%3qcRUEg7zC(?xMjG`#fv^BD#0KCHg@&EAe;7|VipX2HK
zzloxKkGv>wyS$*kUEr1p3LZ}l<-HldqA4hPcv=?><AA)#p^XQI7yQ5e;on1D<@j6w
z%D;-XX)q6zT666tc)#NJfBF0PlRx?Y(0nS;vn1SK-ZL<(R6^M-&qSuux~}M$Dp|2D
zX!0Bl6VJV-tq6du*cPBNSr@>rw6~Q)+199<hN*kO2l!q<rJo}nxM3UtM{#wTA)1q-
zZoYi0b3a-G>pY{oy`Ucl;_`AHby1-{onVR*mYFRmC1fx3*hA_(w8|I?xnfx+_Od$E
zWsRmG%ubjaXkHQl1whMrceLA<X~yl=!z~s@b3xC#Li==v%1jKF-3tgM44Vqwt;6&4
zCw%<)0YClpQxH%}DQNF8P817Bp*EEAqK!gUmM}#PQ&h<F5*g#8QWugU?+r4qVIeS2
z#5=#H{l_r(js?_y19nWnE*Q|dt|SDQmpcCa_9m5EOK=UtF7Udpf*aK(#%-#>$&Hbk
z4U$GdL7SdqnG&->Mxl_5f&Q%+yk3_zlIa^Mk6xgaQr5|%?*Ylw9CglN=z6BNSOaF(
zwk^t%Z2AB2|MGvs=WCC@^LPFZe*E|ezy7m-f_3TvM<zzi=B3G77~^0RP#9odSIpfF
zFP}f5ZfYpEplwe0FaI0=Ccb%hLUDSB-g->ygymuZXYu*NZ}If*6ijsQspPk43K2_C
zJRd2@4Ty<MjHZ296QhM?v)HWd-A+=HFGzs*vcxi30Hsce*Njr9zI+P{nBDh1CA?b8
zdyFesCncmFj!7;*_)s{r@5IAM+G}%a15unDyks}s^{14=x@?0#opS^iaKL7I>F&qG
z9}vPqmU+RIwM3=J^sK5HPfzc*aaz0mC~N0EFb+LFKYzlfk00>i!$<t|(@z+u5gFrj
z6raXscQ|hbSy3P_Yvg5xvTBg!C1oPeEf08MNpXDUla#c>;uXj3jY3Lp2zlVgiz`iV
zh1C+3gP1mkOJPN`V%tYxI9OL?+CZr^%RQP^M;lDAB?=Q*Lls+qa3G~kNNU}a){|h`
z9zv}ZldJVq-U?u<3xl@dI2JK14>rZtv?XmURF83#CQ~;J8pbrV*7(yu{WE;~-4D3D
zyx{if1D5Lxnab=|w?+Y|slbBqXzLQ%FK~9mt$W724wx@Jl(Eq6f<OMFKg7TMxBnLY
z@-Kda)+$VSgJ-Q!>lOd%AO9XckIN>HOEz$r2jHP_@IF?=an6ycq7(;O)(_WeBP4hY
ztd;c|0maGHhSoj*3J#+5P|D@BS_;pA0*b0ceWKFuubCX-FoG_rG}2vi{F;W#q7xVx
z!1~y4dV$9E@B15up@Rp7;7EaorfxO~%`%RAERf>LB%*v6IzWTC)bboB#x(b5F`b*J
z2oU$#<>eWdmkTb>&lvlGO--uwa=XP^fpVefJVRbID2f(EQ6no#D4ntW6Y)4Pwluch
z&flTA_jPUWs+W?|%R(j}aefoIe@Ku_n)8(SNmX7$uc)f(kbimq+4P-oD)Pb+k(mgj
zinXzO?1(-`Gy^$15~D>NIUXj{Fh&O~LNSf+Q6#xE&(A4AiOV!o%*nc<EDE$tPwb#D
z4kL!X2mD8T`uG9!c*8o6oRdgin{`o`#~#D=8Lq6L?2LXGh&^-cF?Sajtubc`fA_!n
z-^Jhk*Z*tyvw!xdnE&!e+%7-i^FRG3_;wyp=><2O@h5-zA(RsfCFprR1uwZ<C~i`c
zO3VX*90{jWn`oH#Ih4Bzg6G<F9Wy16W}rEGpJJ=T<oQygz~5WsVpob&Ri2RNkzVL<
zEW0*MW8_aKSx32arxSFZK`H}9kK%okGcoi-@Rlh}p%HCMg*l+Mm)UWP`*RouE?VjX
zpV;|)M$?><fX(9zLU5?tt;5UnC%im=!lzFkae29rlL9_?o;#)n5{){`bL2&hqHK@_
z0Yw=M!ET3w4-PSP$36Qs^DgaQh>&_sU3Sjldc9#9r;ydZ+7($&w7%k%lFjEmgLHj|
zb*0F1=M~DLrmXEOBkvz#mV5Zj<g=kjP3`*|(-S+ac12MNUwIY;M=1XuKuQqM$Qs_d
zP>j}^RO!v~992`paIs16{T5fbn~xajK8_<CYm{1(H>W5$w#t)`n#Ufes=zY!Smznm
zQyoHQSF-rrf=PRP_7+)2MOs(8LcyU`8vmt#<G%qnJN(1H|9{3{NBmF!$NxjL-825M
zHTdJU#4rA{e+%z^ctV*O{4f63{|<ijm%oP5l&0bxu@unuQc*rBDvBXyy7!K&p>z~`
z#F(Yd4C<yvmgk8Eakp!d{X^d;*2W|_)_aGts#(KSA>Qpvcd&C7!_YAeF#WYQ47yi!
zz<IVB^IprN@08wk3%PXOBak5yJ+==!Z;Ork2v)OxvCI<z&m6D`07Xf$V_8n#<x~fY
z?)q(Hz5^ATy>?eDmNa0K=ct<o?dcpZnY}qH^vB-8JBy+3@yu0TK71e&yJ?zWonuyq
z5x2_?%RC1PU~MRIJ}(;-RgJ8ys0f5;jR1rwK#Eu>-ZS!K-v16YsV_;ty=qDmtI+o&
z#&IV0K1U3YQW+Q40^YHmuY;#zVh@5+9y-&gng&@>GBW{!fJ$%Y;DJ(#t1S;y#m2k9
zVN^G5NFXJ*ETu+hrLplldG@Z@71J<ao~R^`(T1v{G!1Qdw!OELe!uzCo6japn^gLp
z09?ySv-32<E;BCApE30VR=dEu1(r*?XiWrm!#u;f6<Jn7E3PqE7Wg0i+y4`szxxIL
z-GA@@g#X?D_J4za?|=2b#NYWF@A0qv{tx)0U;ZAx|Na-KbOvKG)MbwU)_?V1$N&Bx
z|5FOE^W-g6o^V5LghrTrXhJZjIjMOZ29B9wvVE;kw=JqhlC-zw%aYxa>=)0PPlsuq
zf@dht%l|KJe;#E^a^LrXpNQP=Qv0j7Z@qV;fd(3A5&%I05I2#uM4B=sN~R=hvF6N-
z#-3wK*0E;JnUTi!$Y(T~ku8~`7E87?lw^_;O^K@rZ~;jWWMgfh_w9ARrK(<4-MU-u
z88LrEMBcnr^&0(1Bl=X=tEzkRW=2N-;<tQ%-!DmTfaeF?t4MR&VmDP&gNJgt&Lp<s
zbh|E34?P#hH0L*;8%&V4y^Ed~ko5b6u}e9)Vv_TMmP19Gtv1ck*4SzLnkbCv4_5HQ
z&=3+HhZ|^X-mI72dgE2ibT+3bE2QTeMv1V;%5c?StF*apcZu!!Sb46xs!ZeX(TK^>
zn4{5@(RfB))NN6sDof_mDYaD~MuGGso*&@_=F1O5yue2x@U_`smD}y@^VppUJJX$r
z<=u_nmnYQ?>rPAbxz`j$ZWc|WY+lNUhb~H}x-Cz;vb;&&`j!qjX_MT?UG!yqm8qNw
zWvg&$X6a5EnnoF>&Cn$|-2$`k)!NS6axS1K3M2B+n97Ni9#NchL@<{PQTGApCY@bw
zuCzB@Db*(Lecw0dZ5+T>U$4w_@*+1Bz?g+9E%Sj=4OOMkYN2Haio=7FAOGop!_Ajo
z<4e!}7EeF0!TY!RoIkt8=GGZD*SG1%9lF>OhQn{6u0M{_Uqr}%28?hO96QVHb(rh4
z4W@NXmglzZ#Y0L#97ptfL)$b?=NKF_R(8)7q7}f%(-hs9-QFbKU6XEiMIC#PkF5ch
z8!OAIZCsAylS;v-t=F#6>~1Pb*-h2vp%%)%?(@0l5SYzpKA*SXUj$YlcCb)Ia%lQv
zHdhPRLZz1SP_Cg}N=Xz&23uzv+MRRIopX1xN~snt?=VrL!$Wg!)=ZX_4fE#nmJSFZ
zkbX!IMR<Nh5JraK!EWv@HF7Tl^log$y~h7$IjQ3hXLqM+^DuQqT^aH<hf^}%k4eZY
zRWX}Q&{m8S3_(|~H^BG1N!>-_@3f0GrSR0sYbokl+2izWiF8>xM%Y}BJ2oA|l4Tin
zRoQbjL~W+eD>TPBwir%3C+8((S+)qLgq4uD6&OWh7M2iro-yoFN^0wRyO2Q&3@<#!
z6GZs71;V)1Uf}bS|KuOxl{xRZy3L0lILlVA&zbWVSlKwkAX!DIlG-<X<DdQ71^&x_
z{70#33H%^Jd=E1E-qxFN55iKLQpd0~j*UWVoAX&E2%;07&q<GGk+XE|Q(Dz{)_}Q<
zU?Ftr?!9R5-SKe7v212phHJNSI)2Bgdrw<8-NxIhG7y2LtPBdv69iF2((fas-#HKG
zv2->jx326Zmw91Yge>!^yZ5f4d1>Q4B?@x%M~jB$neldoY@3jiq}M(Z#x^hFg07)*
zo8>~~WyI9x$W&t~f+yov)!4)n65;!JQA7|X#z2KexVLT;NNK#^V;e}D0&(-w$+^jQ
z081Yq-!Q~EEFM^sSTzuUMj6jWYu%#9VbN+(mZg!^!Jy0CZsQh@JMO75tu|g%mSqNn
zCZVY@DBgabFpN5Rt3{Y>@?mzd*%x&4M|LB@^L^v$M<EUs-+g$GU$^^r`w(^vncI=w
zSnS}Y9G&GEfgiNtgUj2wb2E82O;r^Hz6Zjzqcn9zPu8q$#k}u54>IT_tZi-Z`pyIV
z$}fEzUwrKe-hcHcdGEPT^Was@Klu4SMi_bgp+5|7z2))jv)_bjz8C5LJWD&GSl%D=
zJg0DTgTVI#D}dk_JD2C<lv`y&mzIUJM%rIG^!|wj-+C^`QwqmDAKh+%11n{I+6H=>
zCuPELlZPtGvONQXtr`%9r#I-HfD72`D=hP8-4%F97G!wGKc6;4YJY3eAwrO4Df9Uh
zr5n6fr*M_QU|<Lj(d~LUG4FaI+17ecmU^YKG&6NwGn-DCT19Dh4Brn3;)o!M@uQe9
z?jijEWrY^dh6<)@gic1T_inNJdgI-jv(7k?kb8&MxzK7pODXe$x~dFQr)rSW^c*Rz
zDXNk*OAQ6F?c9!&m|j1zUe)OaYT0u!y>R(_j;d>fT0HboFF^~15({t4urwO!DurH5
zs@7C_PM*%ubz>Ej4M5IS^R%+Lf_u(^X%(ref^<5iDvB0@VRqymcn0gGTr$yCL6Tl#
z^I?rW&f>h3<6GAn1IDX)?1_i?qd)w^WLb(=&smKnPrc_M25~@~tT8Tq{@JI#o1g!k
z4<n-$e&Gv$me0QUG1Bpb*e^JD4*uo8gunO~0cteHU<l!g0UE807o*S=g=q_FI^l90
zN5pZ|J|y?{cFnv?G&BY;R^|mLZK->FEAt&5yVr{l(Rm5o<2!HN&6(C}$y+(yL(#{+
z6j~c%Jr|lOVZ<~077uk=sM$4q&U?^8Q`RNZ`Hb8aK9tY|k>xxOPKrfzPHg!LlV_Vw
zr_{B=mp;Dq=qEkahC_VmEx|(!XTH5($|A~D!z{}fjVByhFY0JKrK}XSf~Gc!^x0%e
zU6eR0BRwB4jBFJz!3%rHAT&&}E_}1F2y<V|9NH(UJ^$aUfr4*9gK>9s7ualOC|{el
zhtDh~DI`r(lH~?&5cq}|&kLf~O~0q-1WUV->xHZ9sx=^yZGUk5>5I7dm?Dii;Wo@m
zZI#LG$Wa)zYBmdl*}k(qR(Hp!Z7YG1$x<|`v3*`f9C7Kh?M>Hcl3tJPvuD}c-No}H
z4W^ptz;vo<&_aR|XaZh(=~Z_3#z-_PNyL?lXX$&A#Dl=s+&H+-w|?*Y_)k6n&pZR4
z`yBk@&;3cRT)V;LS5jU$?9S*MA@?ag@NMzHu~j%?wxTRsvO0k&?K|G9({e^GbX=Nm
zt&Qq%mZ2Q4lcn5s!ja4CN#Obe7v6&Td$&p3y)TLuU|oSgi)CZrzIWG;%ZWN;=R)1Y
z;i1GL#5;L?i#32N1f+A)EHh~VVGOXN-?`T~ZJjS59u?B+2t;Abz*5z8DasdbUxdZR
zo3~0Y<58<NY}(BXn7GUhRZN~{3q46G@WYreP6*>3LEJZCp7e0-X6<LG<?}ZADba?v
zzn}4YR4{&L532iG_m!m?={#Kmbpm*vBrh_ms>Jg>JR#e#$_W^pT28g2yXf*fvkTR{
zR*Ge8RW~#^?V8;8EXyeDf_Hy+AX1Y<I6WEMHBef+-aM<DXY)#N+(FDA=gr21PIc38
z?b>yA?(6^x>6uCz!esA!-=nH)no5E4m`^i~MpFV^6Rq~?`;rT5tGxf|Z|1`Fhxx&0
z{s20JAOCUq^rvBK3;yYkCzQK?p6BIR%HQeE#x_bmfXA0@K4Rg**G)}2H@$Q+@R4Xd
z-*C%!H<M-YOJiIX;kCJ5BcM8KnJ9#yqn9XKmW<`9o5{<_PAiqm_c<D1A;hBk>ZGks
zU!Et{vIBpXoqK=0ar|CM;qH2HMxj-+5u$BEHrIC&NP0YM>~5xxpwUrNdqHTB?!6w3
zhSQpvyB?xrE*RS+`RM3~xiwH#RZ$g%VLO;k&HX~6rN9efgD8&;+#rZz+n;FDBb|FJ
z<>|BBy&mm;eg~)HwDEg}Ae@}c(?8VxzOHMgvuUd~Q<ar9Rz}T2N@<YgnC^79a~<a^
zzt{P=+=g59u6h<)!vYIYfO~m-&ajvsKLd7fOpCpKuX8@9;eM8nSC)>~yKLbGQ53cU
zk=^rf^XIhv>bmBEhaTj&f9tpL{h*~__B@HMa(tnQe90tzgf#!-D83ut`!Ra`U*?nF
z`4k`jmJjoZ@BIOKgCY6oF3so`|M16N<?sE=_weS<I^Xof9&39q@Z^&te(;Ab@fZFg
zbP8XfC0?U#GQgFR8`Aldsw@bjkkAtdPmm;WTcKKncHPNY^mr<zky0Dw7}tbsjd~b{
z^ap(_*RFUsnFzO`v{f?}j0=mT^m2aiwBfv7SKM&-xm$gGr@~108E~QJvY_WeXw>(^
zPBr%Qhqe1zX_~fZZ_lc84+cZ}gB7b&s7}rQxMR3PxuO5v=jhNR%8R13g0O}?!HA|*
zNa3LY&ku;K2NlK!v?(1mjs`j{E{l0QF|$k8)BXGL%-cJS`;)Xf<pzB{&&utfP17(w
z8dDSnWm!>GrLp3r?c(<g0zLFYqx#}6?>6_knQlR)7F(s72Gul3DTv|--|IHE($e4_
zU$|{}mKx7TV)~#=sd&L%wD1&;yZH8;&}gNtBvIPt1f-Po`X;Y&T24&}Q+{@2c;_!(
zWOHka*>u_(HKi1r8!LS0w|o;)C;aTM{b#)X_WQYh9j;#eTl`=DFE8`VhqkG*Ik%qu
z4Q4yH$qsfHBoW{LfdL=7@Ct9=Il~7Z8?n(VIePI9|Kb-u%KzhUZ3D2f0&$d~n>i{J
zcr09XS2bH-U9<27N4b0QAwIr7tY3=|M&WuspOI#1`!E@H14}#yr#^Hi*PF?W8KKd-
zJ<m9btheIX&A}b;<XD`ug9yyAd0@i8NSQBafKOkG%NsWfPA5xo4F+ytlI||dv(4Mb
z@dk8@14fK@KA*RI{;n$6A9Ue3(g>TdY}~ZgZTe$89&<DrF&Z5)w%s2}De9_XKASNf
zkEyG&ZGiTJfG|o3!k93MO~O0~z%lH1?+qutBRAy^i4PW6XFXw|^-?tQ^~N`}x+etf
zPFAai={%(@E0Yhh!ULX^^b&)G@w|XAhzPsVMD2}R_T0?z%|qlcC??lsypCaSh#!V+
z<m~v$UC3~}%F``h0|-wVPhpVYM<H5xo%1=lNSz_lLbqsV-6*h_s6j`6PvYn+%njFo
zv`J2ZL~2bE20ZrU6a4z8f7S59*ELO}SnUsJ@)<X7U*h$*pCSn0mwpM(oPmG)Z!hv+
z|JA?Bd2z@<&51mp;b2HM_etk-R^vI_5A3iOuOfs;Rtx@%zhZ#Mbqzo81MuD_KgaXW
z|11?I_kk2tRn6|sT~=0xY;3F>P+t%bC%qG!$>|FL9OyC2a<XiW(v96&J;Ed=N*2Ia
zbIZC3^J53Y6buXlWnsz<%6Kco{s3S4?V1rw$2x6|3hRx{&`kpxe9t7M`>h<=;#{23
zJ6Q~H;7`N+>E7e}0sa1vIPSH%G?T|aJu+~)(yFSN%_d8H{!Zy7@J%a@Nsu?RY<VmU
zm4S9-j4gTH;b_cgJSEKwn!2H>4G{O}=#Zi?tPWZ$r0*JsP1A4~#rS@JX!-s{>*OvO
zZk)WC>k3dfFI?F3F|avn$W9pOdyn5&b?==LVvE--{_S!H4$RoP)Rr(TjuOH!vdPS4
zgK#{T<c{fj{Hm(5pv|UD<St~*@9pZlVOnU?r5u85kTnmYy2mk=k<(R2FhE<3lUrPl
zw@Nt!Wv--sCfzkErO5J}2OoNf;p!02mn|vUbX1T}CWP`02FY!dg1`RP;lT&t&;R+~
z<fnh`7g2tQQj)q>jMJ3KJmc>E0XsW893Jj-G&*2bW?cHjchWzz#qRBt|M74CI(1d@
zE#LC*s5N*^yDb1lX?SyMgSEAFhQnbi&;4$aqVB<TK+C$WJ3z~*!$#3HCnr5KSAlb_
zR*oS<YfQ8EU_cm!r^Ov7js>)#*=#m7eT(Logi&PLURs{9dwYr9HPGEG-AEF{BI?fb
zor*xt>sHD@Ao4tK!^J3y7!HSRXzdI@_w&03swfI3lSwP2aCmq~np*ib3$GcEk8Cc_
zRM?Tep<Ia)<4Hv^-Z4q8?(=+n4q%~D?!LOiL*q%)dn)C-_m6&0M(XR0Mbd8h_ohiV
z1x%&G#r(t7HPiMIMF`(BoJF0)-@R|DLEn~T4jBBEqdYlnhhKP3p3RdO{ko=ZOjV<<
zP2wISk8MDtfpuF?H&9rpf#JP2RY50jq--9c*Xu3Q<_ghP)LWra3_L;$O`{vKB1KBU
z<BvURU<<yamYSOfBW~Q;=gjIZ*U$V6Nwmv9{ijf8FQc2k$L`%*+#StHk`<D;M_m-u
zr6LRy%BnK*<W<4uH-8&v-~S=%BIC7B|Br;D|Ane3_+x+K4;xAiX}GeaG~m*W&2`q-
z*A1b6e`wiE<<f(9d=u#|W|#9Za|61Q^KonxCk@i+8<c4pUo_StrEOKQLZio#M#qlj
zZk?(s=F<sfmLo8sVJoB&h9)U}N*KG84Av$#bydN-6MFV~JqClpyE5e6(anv~!Jruq
zHQ5@o_UqWSIylC7JT_2+qoX#B(KHQ}&6!Q7Q`7BX$ds)I)gz4iL`fgt8YwUETf#|q
z{muxv`z);_n%Qu5`z6d9T4^-Trgh{uj;iH;$MOv>ijs*H7IYqq5QYq2lr^rztQvK~
z%kSoHz@dvXt3}_m3G+CP+px;K?;dBS7p*v1nvrKl+)&zX@?OtyJDJeJ9pt!8@8-4)
zaypx*=+5~JdZ!36+IKH%R5(-#N-K)e$YQIi;*p0Q=H{Dku(rO&S6_UASzYkO*YA*(
z4ex#6AMp4CzfH+|QTfmC%B`1p^>)dxe(ra9@cX}m(DUi32CX3qLkL2IkX-u2_ptHA
z2dK+}ul(Fkk>7rs3!7*7nV<My_|EVD0Y3KOXZX#}{tk{oHHsrvS65ppsvr!loV#ho
zxz}(ZwE^lbIM0PJy^@Aoej!2c-qXEDOTLv}Af^okv_aV?Nz~z*y7!G=&PAHsSeB-3
z?!*{_UQ4~Rd>-zQCy(1S3l<1>4P66#uYBEU!+G>hzRJ+J3|eMWdq!@C+iTUBXTnuG
zk2)$ZlgX62s!@%mt{NuOsmbM8L^)a`gTTTKdPGSNKS~U1sWdWRE-$gjF)k4LT01bb
zZZ`<iOX{_ob(?o}XMP`c`@TJ?)2#QsNB1!_d7<glxWs9ek{3BeSy9!>8Uhz!bUm@<
zW7s26S=HdGDvhR}Y6!!iEn#<GyEq$b;XB5&C<=pRQ4I*gHW()nNfO)Rm@k*ha)34S
zrEL?d<5QYZR!$yW5riR0zlZcZ+neVY5_RiAnV$`)6>YlitFlBl6~0ictq%F-XWq|z
zazx+L3`0p#HGK8P9-n>b2DkQplcxM}0)NKY^({gvxp{Y=o%x)Ni|1)nLl^|8ro#6F
z9{avO!o?4Llq%2pg@5$l^W|UrG{5_sU*HR$`vSl7+n;59eVqp$d;lq3ON?j<_1o#x
zHo@P9Fj&v@bfmefysD-Ko8(P;J;QnAghx*pC%2Gw`=wMvIyX(a!Y<t)2<Q)nCb9bt
zRlM#dXYPC1LIe;Uy*ztPx;>6NmTvEP_jslPHzUtdvr;t#e$YX>s}oUy?%a15<^xo=
zXRhkBVfA`Rt6sdgHaFLcP0uCD(==r=nJ^w5ax^+*J~hVEF&kug#$-CBC=1XEgkijh
zq8@Q%HX+Xs4Q)`}psU&xqw3liE3FJ8r~<9@B6Z`AWxa3dq9H66O$h7dE+eeo&G<ce
zQeV$N9ba$@??+p)NJHCF)irga2z|37x+Y4OKkXW)ZtSa+K?OU4c-3e;;n7bPd4#21
z6Jwx+Hc45fJ77yEO9e>bnJ}+Cj%d%r&?PI58%!Z2P2;>evlx_C_(9OhWF6n#g*N9S
zE%Bc#BTFf%^9-SDhP}{K&*C0it084Jr+)YX4_vs&b6<OrSKhkCXgudDZ|@`CylvEg
z{g8NLgCq%g?&UYQytPhMDWVXRZs?yq%lRkYPno6syZ`a;@LNCozjEc=WiD)==iDQY
z@t^<QALXmheU<<0KmSYoi=X%jv%?A_=Fran^0#C2v7DcPMVMDrCE6m*qc9?ldyA6x
zJ3=}Zd6xNnW;|Xaj<+B|3}eua=J(_SaNKA)Z_M#$p>@l79{BzuTsbzDO6X#NsY~ak
zNa6smz1YdVpJaeo4r$Hx3ArF=EX$%brcq?L;k(|Ld*AWU*!4gz;0A|G$43--j%u_)
zlxG=7M@MAY+%h6qkH-%P;~r}pn@Az3wWh8LgfQ%^v|RRzx^@*iqX47@NKe>Xa$LC7
z{Ip#CbJ5T@vUhJ`K{~&BXBdJr7~O<mQ#Y-GOc0r-$fhy)n0G4oYE27nGgjPOeA4R?
zM^Q&k%HmME=K<DGxg9Vyf(kZK?)kpab4y}$cc(hGK)Q?79ml!o?pnJo87Fcz_CPM=
z^gL5VYpN1-&B`D_q6j=6p$y9SLr=e-eiAU*zr*_;yTa2?KhEo~-sJ9K#;sd7nWQP>
zG^0|Q!y;v#uJiJ1Z*VpU3H_3-OILX6kNz3L-hf*#e3|K6FY)n@d^2bdC4FY~45dn5
zf8%vN_OXv~>B41h-MV$+K6lN~Y5QS!Zj@5Yr!$Hyw@cT-4f>siUxz!>ZMc8h+Hw7l
z%H@2-g8B0_g16I>yFx73D6%YTAE3Yw8T5xOMBw;+Kb{<R)#1YOem7NP*PF%o^^#U3
zq<bAFCu?y$!`5jk%Zo6tf9e|QUZXQmT|AV@LyeiwbE?Yd3Y2-l_~?jqHp8=3Wu*mL
z1PoTz**d#z7(JzCcXyAX5GWZSy#Q&M7|i`Q2qz&0*d}K~ZnyC0%)K_bLFwjx>CP^P
z@n3KJo;)dWilKU^*LH8VqhfKw3?{eGSlhSkshoCWt7;)9rIg{Gjzc4JB)#C62X=3_
zvM7qS(&&st5Jg129!PIVgrIXZOUIT{w8enRazr`4YS$)m&(MQ~iw>p{wXzB%o-{jH
z;7N3uqtOID3<rH?hd!HU&T!%CHD>dY4}8;kUVZ6hE?>OJ7k}sT?Cg&S<CxHExV^uJ
zT-)ZtgHQ0xpZY(NtZj1Vl^6N!zx*diAIc`DENarKVs+~b?|JANfBXOYxB33>|9*bt
zM}CCA@i+cP=gQx$i*8y_G)~krHDQ9#ctQ{*J;EsIs7l+TpL%$+n2m&(;g)xm>o|$q
zs-T-gx7JQeatpCwZfL=uTDRaxC+yLcd1~)@$96j7VDL~A!&V^xy=aUMgOKJJaQL|4
zT>PxbGPlW@RwGyZae1~bp!xVQv|b369vvNVaB#qAG-94+)QxJ>2h-_<*>nPpGWk#G
zGgw(=bz__JmoKrlwgReVs~@o1kNC>hUS?J_DB&Tkm+$!oOzH;_o);i=0}{h9iUMn>
zwZ+6qVFNqu7&~pj$fi2p?f5+=&h_HM>^RMvI_&}N_F1*6xw<~KTb<&4SWj#|PZwdH
z(k)cL_XDt)s-<&j7s<lBbSq8nbAuqXGSp6g@7NVD<w?wY+@f!*c~(&2cs1|%E-0;$
zH#gpd7fn-=R}~UTQ=6*b(cu9-U(ny$qL_@Bk9MijITtTqV?Hl9ckvuWUhvQ(kFYiv
z@Ud_CW`x$1MZsHlZZmHL`?AL;{`_C0zj=n;*I(h&|I?2$pN)`cGEZ~i>{%|vJ(Ozr
zgP-^~|HDuJJ6?L}WuAHFL#(W<oV<spSd$YE52o$K<$PRj$9O)AoZ$ku*5z?po^J>K
zY^kJenAh)kW5@5&^6#AUzo4Y9>l#nWHhJIbE*?wLOX)Q9x<*>Jm^#;rgRKVl`vWU$
zW_^ic8?-ALbg!|hidLw^R8q})9t`?R9^$>C6_>K;G8-5M2SbTeS7xnGj*g609+JUe
zjrFr<+1T1<<IFZuFJ_oT$fm?sIq!SV;~b6-xqWxS(JVtFt&mD>m@k+C8qbqptG_`I
z<9i{}50KKgeVv9VPveaBqD}7I6cOCG|K2^iLDcd5%-8?$rqRsjIZbU4ud1%78)c0Q
zh>p=+uEyw=I;CkeN`cljaoi`26STwv1Rdd*=qwE5qH19PlUbHg6sD@<dErvyXZJm)
zjjrKwEg4Q+LTlZIAgAr1Zf+V&+TW-KRhjDIbT)%JCyI@J8+Arm%!vFxf%K43LJ}&5
z!vT3wad+>KWU$J3|A>nZpQFg;Ovgu5Rms&WR|$Qe|K(?Yg@5!DKf~6!bG-Vsukwq3
z|L@S0GsZ_pcyUZoX=aUPYh{&Q+-ICCZr!=dm%jWJ{>Tsh0BdV&t-6mECT-m41|>jf
z2iCJ>eQm#<q4~|`hL_#WJV9teD!(<}i=4z_s+W8I3?o5Go@EF(kDj2vGHBIaPHak@
zbtuJxn9lU&%_xgGzBKO(l9;65LrZ~LGEp55s15PlW+_S28ME2k{>~(Y<6dHzHZb(Q
zC$WUxoTjjzUDatE_S!s@<12F;bNBkXtzC6pk!5qH(}}@CjYbAhZX3Ysre<b6th^`?
z(qm<9lj{#X#@4xu^algt&__y$LJyCKd|Fc1CGUCkA=3OsCTT%!o5|bUq>wiJGvBp_
zs;H>6rFXG*L3+mMdcGlx^t`}G!#O_mcR7Afy?aYO&pQ~V;}5!<bR*|eqO1^sY81k^
zf7fQ~Qd%ul>AJ~ICyljeS&_#+Z{dwC-(bB2+BC9MSXJEKJ;O71?5@EvA;`Q{Lw2Ce
zZjR($hEBKu9>}xYRCE;h!lTiOsxc7o&CN}UG-Z&4y!`r$TsVK0OP8<ktH1ggu3UeZ
zMoTVVyhPI|2K_$Sut!~oG_pa<n!o&){}R`(U+0AvUf|FCsXxtm-{b0fpWVqWvbTyT
zHHXtx-d@j<y*^)h@fBvX5&z^L{{#O0ANa%kxj*-3`FlV1_i@WkF3qp`?mP}<86LD&
zWESV6m7#@pm9r>{Xgc$-nCfM&DEII7UbPM8Mu^jaNKI96`Q9((YwUT}O~Y(HBhM}P
zlZ7`H#g_l-lr?*r7qbv_t#QiRNz$X&?{^H_(#6?hSk)o5$j`c{z<C!2wlG+NK-?=r
zX_}@Lh&VbLF+MzCHa?=To}w{eDH99W6jHLfafYi8KE&m#*XggU6Gb7u)JQZ!RaCYL
zc;oGxT)TRW)wNyXamK9WLUXJb&Oj{yn*wd+;tc(fR+_3d2!6AP36RcMdGC~oIWc|@
zy?ZBre_G|@oyKf7Zw*Cd`K7c)npY0--C2CyPIniBgF*9#VX&0TTAbg)P`L-Gu4_w(
zSG1k<COk_l-K{!d*t*x}7E{yIWZ5!$-Wkkp2y!nYr7T}|(=<lmr7S@=Xk8H}tLQ3+
zs-*ElqA(=SQv#_uclHd2qeHeYoa2$l9^ngL{1Ph5DGP}h#H_EcBM{tp;Z;8TUH=}R
z{GmUE(uz<0tDoSN=fB2RqlEKot2}&WlQk{5n;+4z!BDL7xnF;lS6_dPMkl=V@=fwQ
z=kX^Vw>@Jf>~VQ*b6LyDPqPp;MUk5@Nb43w-gDAcf;QK8g1Xrp1*a!hmR63cw0zwN
zq9|^K23+p(WO5rEfuXT$nmkX*(m7h01h@2j`W7*7j^UsM^3;3gyqV*wI9juWhjR2a
zr!}%KJTWtOS(cL7{zcF8It-f&FT9%{J6;%YfX$<$BMuMtIXXHdozH2i(rm_A!E8Kc
zKA)l+&Gy;zTzv3Rw$GkrZGD{}iVZZSE@|q5Do@GgQ;IBSI+-H<9YXsW(aFmRx#&di
z0&tgeffoPhFFaY718kbgwyt%zt9Ks1FU0x19>TBpo;=UmR)o4XMv9K-as0ThveSx-
zYE2YHEo0~DSQPW#HpH^cmPX*gQtf)uB6*V~ukCp9wL1)TMoOA41W#R<=ddhp97j<v
zAIF8Srb?-mVsmqohaY^9?K5Xkx}hj4);BhYlRi(p_kC<^Y=97a;OVECr8A<)CoSe|
zZf$Vo(j}yly#A%<`RI52VLtvx{zH^f{L+vAeeS;YEZ_Hm$GPOe?#?bR?H}>{s9-mh
ztY3MC$G-P3@Mr(q|A~)({~xE9z@42N{0~3&_xa=}KEX53d}!g7NLUyr^R-+#GYM{G
zb3P_~iAXGV%~Wr@^SeCH&R7^@U>+E)%&zR&Cgd(Or0aDpc|Dyuc5AJv49ljFhMG8O
zn~=N4_THPUk$;-C8>PuZ1(xwOd`HiDaeV-SvUGW<4ri6E26uVbk8l3nM0qz4b#ydg
zmgZD-L#>_aj`6Uh&}^SQ!-WeMxNzwbE32!-aY!6W>N2IsrcB3&jK@ceCsPhaW2W<*
zH*VZybTmP0T1DpLUMsNF;-*apvB~3UEYFoKHVFwr+I*FdC*Hk(^uM`+@x4ZOF`8Xq
z-s<ofBh?Y<6fRLIsLGPmaxytL*HmQs_@0juWprMi3$r?BX^-hbhq|teWV;l^QN*C%
zGr%(oWmvwfr7Q)mI+<nee9T{5HP3lt?s{E`+r7|K^vw6Ns!6kK5z2^=S<&#;?ml&q
z)8KLa(k0UQlm>wgC9k}CgWEfM<dtN&w#M0W7l@(&-}CvkPyGsmxX(xaz5jrZ{gEF=
zH4UHo=Rd}?Km9Msi=6A1*SU6nle-5~Uc9}_-t>UogMD6l<5i+7w^==Zl|TF^|1v-L
zAN`-m=Tn}3^a7qAa{byhcJ_8b7(lAAQ<nMKd7Q%Lp|Z5)^U@gJt0d{Q(pD_pm+qQy
zn~H&rINd>Q$>O9{9d@j!0z;LnPc|mlc{f$0x~iGarmY;60K&l1uh;^K5DU+?8zOc+
zav@wx*Sz(rq}9voik}|OA)QwIsjS$h>0k8lJZUuwhn-$QIG&4gYte|*W=zIoMu#Jg
zMn_C06Y`><Zj@<Ep3WGLkLdRj*4Kud+djv}`Ub;(pQP7AtD3s3nT`)hXS245l-Z7n
zx~{3}hN@PSRSm)kM4W7}Zu5V-?K-A)x7(%Px&YKbE5Eg!N9r5Oac)2CUVr^M$#WOJ
zxf@kCC3Rz<2!0S^%&>VJlrn+}S)QX*V@l9MkR(w{Xyz_<v8V)H5S`Y_hG!{NRbXl5
ziGq-RuSaNg^Xw{Z4T@&5NZbyx0BzQQk5YA;Z}I&Q-RVYmbaA5H;jQbdp{W(wJfo^=
zjFHuZzBKR$FF*zj>-}xcoIlUeUCC@dWp~f;WDnQZ$%>5Wos`R0t`U?qZ@l&zH=ciq
z5C6e$=M#VAhe5+<e)50jh2Q#R)>qcqh(qd1;R(t5+6wFE&*EQynE&~I{@1+p(wFgv
z*SYXm%3H7cJoxzIY@NNvL9j{DX#Vt{`qMoBg%`{&DFs?cG`{JI*S6-RjOb-P&(V!W
zS^SLAXNz0R3~s>!rh|VG{x~wbBG1X@b97_GK{XJy;iBD<^&&s!P&vzs%u&KjDRPs%
zS3=@jbb1tpXc`1IjCCFScI|aNG{Ax+r#25|?q!n1r((2@M+4nw!Bv%uqBP!w=UEz;
zl~db;%)cGo%tEwcI+-##8Z#apF&$5+$_k}4s&2^VDaCxo*6NTmTbpdI4p|)x==VZ|
zDv_c`nWt1)#(bVyxI#*nnFM)RR+N<iS~pE&p%ezRiFE+F;a}^Ac32IxF!P8$o+ozV
zuI9Sl?f3@w?wuU2RPO+FO+9QeMlKPjZPMK7r!`I8hE9%A;P^sv<7DE~l%5@E)AM|g
zlCYN$_4+6&&|T@Jj*(eDZ^w{myfNc3>9)tOTSD)8a_;9YSzcb`hEK`H<wg%MA&d-9
zlP5zI@a7wDQs;BZs$^&9E_tcBa;--{f;z1kAI^CD_6~RQKG(nFPw<^T_+5B@z#Gqf
zk>`Hxmx=0tXdvltZgX@np{XlWUDE5V@VS@nP}G`ie~;1LEs}GW+1l7*JQ|sMp#`_r
zwzzuvGJpQR`G4};|K{h&j&7lapr!#C5Qzq%1zE9xm1wPL&}}P+^Da)VgKoIya)uSt
zsgeD1yKtPu?XgTzWx*gJglN4}&}~v$>kh!yDztc>T%yZ4exK}Foe0@n9;!>w=fb>q
zqNzDv-E|O%1$cOYhjMD#rzPzdgn@?HU}_mQ4-OAG98Z{KhLT%1HAR|I7X{m!8(i4l
zVr9^$*Nce50N<BX#gtiHk>)vBo--cL$mVnNv-QN9x^4|r`#^}3dB1M4<Cv3dv>7|s
z(Z0{QzNgOpy~lkUspSCZ_cnx}1+djNuT$4Gp-m#XpH20wvW86)hB1S|3PBK^y70}h
zmy@YwS+@*{t^+%aVv^nfA$%GuY;dn@aTDxH*yemp1tW@L`u#yC%sb6$E?;+1l+3f4
z5i1qu1#z6v>kk+ExfDj_HlLA=50P<?AX(?_yJNVMa`nn8VbbIF;UUkyI%0L}694Bv
z`zN`6=`4G<-r^Vk+26%eIa2wEAVAb9y#OXv#dJPL^!q&e_*1<4@@tg0Zy>07@~IDU
z=jPkgb<O_WeLOGVt-T%(Zx7hqyv)--^e1`yi=Sis=9h?c3DT$55>;1BW;4>lB+Iqd
zgi%1h-y;a#=^<V&#5lE}qRenTa_%9Hqu9)^%{Phmu~-g;q|KSR_K>PFxiQeh@dBN`
z7|Z9@h9+16)}}GxPf@n&FTQ7_%)9x=<JWjvxYunQcI&X$Gc+#8Uw_*7MWTGfXk_39
z^VuAlhC$>(-JpHV=Jp0_Yb&e`2lSGNDD;p5>at)uI$*yPq_w6|kj`@>pj%ZYTvW<%
zpmi3m%kiDa7duZ%>!lJ(OE+!3=Ox2tqd@6GzjrzAuex_%hrag^LXf6ehp5hSp4%#1
zRaF#4PUYAFH3WfOuyK6Pg@1fBO+%jNEq#WR(zw_CK2Z=F!!Cr~olhq0-EQCQ@l#fk
z%e5^SxpucRoTLc>+GAJM0z~T2xIq|_^m_PSh_-YyjmC;m45?-*vs|(}ZrCe*Mms6q
z>NZ~Bb1<&ixbP5v=|BDtc;NCm4vq?b^|LQ?FwVI+=%H!}m7-A|%J-P&BMv8Hl3<k!
z*B|G}4}TlK`Ex(P!L1iLe|C$r=Py!brs4X|&JNG_0^WbE$MqGsIf_`l{!s)m#jP(O
zlnJ3`mO}|l)3NXS42FGzz-K{|d)nL@Bjdt<L%%XJumw!H-eK3;#oh_WyF2+}V_K-j
z6h&b(2bw63T0XDE&VKSfwa=gR_OnjAlll7sN^vhw-o4*>U9f=9)OFn&C_^XQYs1BR
z7g$VgEz2?{lQE;wVJqf|YDfaXnPEii1#GV*cz#TnBt%h29GU7}ndi*LM~p@@_EJrz
zr6oEv)mY;cn98RXhLq9X(8auWuI)4{(EY9G{O`C&Da4}N;@yt>5BGGZv-y+{`g$Mk
zZf;f9%%*e05?Ix?$<`UEnj$MK8=!&MyX5pc<-9cJL2(J)GA~SHt9Hu29&y}J>k!y*
zM0D2w@fHn1o)=_U-hws(k|efyu|@BhJ>=33qH$FjMVimainMimJ*x*8g&{&iZQn0x
zsB4(&fO`D_(wQdS4texa!kxF@fc`~@dpvw?jX(AW-oxc{n;gz7e&GvusWzS<`rv=a
z-rcX#lLdM@!|MxbFQyq@W%G$ON@C{9=fmInUA*x5&$9FO>%8)n-{z^0d@pLS!Npjx
zapo+35Rz7KHW8##!DL#}yYvCde2O=_K`~Fsv%-Kog~rpGUcXN-8PEl5p1LlyX_L`f
z(^MttY)V~Lfa}Un==BE(OODpg=aPL{JQl{V81QG7W(fO2=?C;Wp=bBLE^=bsb6oTa
z&gOHfx-@xA=@BKd#X~K4v`!n{HR}5E>bkOsa%tH*`^y|-opb1h$W3D!t0$8&qr(G^
zMu&{26Y9EV&`W6MBV;4lI<o=7BQG0-7lKwaMapb4Axl%H^NhoJ$t=^PjiEzw)GkZc
zueH;=wFc_s^=X$0T*{DGAJ=WzTC-RJ!sh)Jp5ZslNNKH+o(HyM?skj&`9L@7reQXn
zQ&=rNC(Ixvlveu6UA(ChEqKrGRql08q*?eirR6+#<XmAGp6E_@-qP}Qn(tCd3);*m
zU2;-6#g_&6^E6DAIrGKOh7u-4>Bcpc8$G}{wz5?fN(cB#pc}wX@P?Q0gAG;(8m%(Y
zENAWP6;!yw^+(^w^(~)^+Z!Cs8~*L*NBr8e&+(<-{v2^r@ok@chTVG3d{7WKA(8K~
zetCt#AfhZv^1PuxJj;ha`91ubfBuj7-CzGD-uL0}VxBhid#kLhtkP5!KmM;3|HYs2
z*#BLRpZ?XD@BZV0N8kG~Ui;#V^x!3AqX?ui^n<~WWZ1(s+q#GTv>^d)OV@4pzl{fs
z*F3=Y7c86YPTYE{i;%%hzY_;Vn}^8rJbHb@`np_AJbAC(E)NS)hN`MeBQ1g;3K<NC
zEl%j&EP0n3Gwn0F^`ZuYL0c8P7g*E%F3U4!>D(v|9~?3rPYHdWjkR^M*~BD(RD&lx
zdi@?!=1gaq<=CGa2GgpcQj%$*$%=}?5{HUYa&S_f0)aSoPF?S65wdHr#T2tiYnCc_
z!quki9c;cqiSusH)xC_<*E3q}-?V0)8VZlPsZdQtT^W#^YDzQ>(gUd>jH0C+z@qAQ
zirZ;j_0*Qbm6e2YJ$}dgz3w#@RmfROT(507<9hR!<gq)Se>^YN)O8D^aIbEYuU7a}
zqw%$drbc*@Xze`7g=Yv*#92mNq%_`$qO3WZ6+Cir$hST+1T_EfAOB1K)_?ymI6B;=
z84ekoy+rcDS-$ndA7wiTP=1ClD`L+>X-UwRG|gSIs^+l|e3;LE`WM-G?Q6XG#;ZK^
z*i$G8MLFkiSMhIue3d`?Nx`+N@VQ@z7;aQt{0luUJ^gJwe|Q%)zlrAw;xH!Z^$2`%
zFS~Sng3-2GRhE`N$rF}|!Vu31=_o)>CX~BBYps`<F{LoDg@sbEV`5!IjBR+nw85C<
zeV6812y?FtWy-zv@=iOaE`zAciT3-wrHJ5k!?^%#rt>-D(UA%9=5yjGV%X~u_#U$I
z2;(&ZT~THkWsx&YQ^u1SN8>qZQBZ4-vVpQuWVNKK?YflaJ{{j+7tzE5i{`w`(<+ws
zo$Wn!KWS)zyKSV7Mj}k5zAAaQ<M&nF`@I?}_xHKOXjr^dS!Vd2hwlmM#)Nud6c9(o
z`#H@hbdL$N3wAzFjmIZV!nN0nmZ&&RJICY6LAMKbL7B&U{7xIk%g)(6r7DVcU-4a-
z7ey^Oz0|(xDi@lRh}HFr_`?<AFu<#8vTTZynljIL>f#C?dC+(VKl$%I$1i^NIUacI
z8i$ij(!3-qa=!ZIFESn<@sV%&Fz3%*!}kScHX;cX{a!#7ZL_yGChR3_UwMGHzWOXL
zJ^O21fB3ynH{`P^FMTQEH$El!#b1QK@E73nW%%`<k$n6IN35;|JpJ+S<9B}PCm@}Y
z#4%wUy6E?w>!O__OD|&BIxY4{NaCnRlJs%>P{$3)@%eRgJfGV<ls$GBJ2BU;Ec(I&
zS{SPCan%w_f}_ao^#;Uoa$?;a&$o2X&viF6jgcago+OAu6D~?=@NuVHkNdZ4UY{Nv
zG1}i_a=6dRa7e$OP)uity5{WW8cmr|rZaZ;_L*lHX;G3E74uS4YJn1xqSlmEL#-8!
zn}d5a(Hrl7alcMm;a!8(od<J}ZaglmfwJH29IK{AHQyj3)p{`Qd^$JlUJrivb;n1Z
z7o@2{T-S9))6_&kU~w#-$rbqqp?X}9^VD!pEE1=6UEv97Bzhu$DM{J7r*yxIc45M@
zFvD@(yoIpm>43EET$F`nrR*psI}uDF3`kB3Py&;{mp!(xeU#pX$H15PzChF-p1(od
z8}jr+=lJB4>v&S~na{t<FMswGu08T7aneKUh^i=<P7c{Q9CLeb#J~NSPw~(bkMf6p
z;Ct~lwg_l|95voSmpS9AV)MdPZhY-op8M>l_~!5Z5yCWJJ{fcFa-TC7RsaKy_@N(y
zU;O+ov)P#2uO4ynfo<ORZQsoczxInni3dU>R6twr=~RG@Kg3PlEGT#EJ@LcPGGiKr
zT&2}156MZ-)is{G;Rgnt9>?+GS=7r8p35D&dtyA7g3O+iOA$CB1(Sn4$!Kxfa0Got
zQQ&#DFVBKV4eLc`<+-$jJwdG$jjeQ!$7A;I?s9Z^m(@X!l|h2m3K@C?9wed9;jCml
znK2q2GRq3)MMI-JYCocpA!P%NhDO-NXDL|#13~L%$><2-b*y*$q<1EK`4%k1V_C}a
z35!>>6tt~(sG6F(%xUWC)VVut+?SCGe4nzYL7r~?-_7VQ)}kz!&2zL?_<>IlM1)C?
zAl7((NtRcJa?~ZYPjQdC6YNwlOfEs9g&^>w7PfHme7hEZu>j33Hm9lzoC1wg#r14>
z=JHzIoYqp_Lu*u3lcy<VUA8wivei7#7tm;YLJBIRE>Zp}$@bHD8xPTh9-(Kh9pA$b
zU?>Crz$2S@0-n2>^OYTsN8a-^RZ~-xF1J^*y0yXit2OT}3vS=M$qTQ%$p8A!XZ-M=
z_+eJoHkqb13>7jtLX?WFt@BLBbLJe9O>X1&&kzm<oEyT&zHiFe%L9an_~9RhVzWRi
zK_F^g{L0H*_`Z*F{X-vP^vbhH<xvZb)QV+=+r_-QFnT(lF-zwNVK~C$I3no{@PfdJ
zA)!0T*7kOZMJ@`AY}0%`ZF|9tzTuECP7I@@vY&el^r^wJBp3r(RVA~@gsLbFm9xTB
z1+CzOdC$G`DCWMKzo+SxBA*)?T|5HcXE<0P3?loIlh2OEwv{L-vW)roh{HR#ne6Yg
zwZ2Xg29!lc6#0bxkbHW?WOm5W;fTA36UMWgX<kq$+W@S=^EJMdAT2;u`7}zS#KKBC
z5=g`%A-;G-O(Olc`jpEwitb}1Y)w#FW2?I>K!DO{)zCBrTG>{(cRB9MNEzPcdnu>9
zyV3Q&>JEXup_VX;#=akbXb@i0_5&?1v6IIG7oKI97+M;`VqpslWUy|-i_?zdlAT$W
zw$KPEMN80sPd690DK3nZ>lJj*l&j_ikx!`&g@vkf!;~1T5}f%Udg~!%>=S9XWE9eu
zL=rx5R^SP^J%JbZ1J=%*Cod}MrZ!3Fx;ALyDyObWsxqhFU*qD13*3DB7XS1g{uu9l
z-+OuF@h8~6aD|nJ681(r%!(OLe&V}%;{BhbH#|?2NTO&3>A?^DImP2w%>_L67<@<8
z<1=5Kl9nZ-EakPEhdlY{`<U+U!QC51prK8;E^-oXI6TyJY7pF7o965!=~=aD7j~R}
zf4XIfJU0vyuHjs3-In$jj27+IehNU>7^pT6Rpc%8i0>Og&maiz8NQx29Gx)AP#H_#
zGfFm*6R%k0oZ8&3GDcKulPEtr8nJWxCe!f&YpZ<&C8!Dl(ID!Ebe=OgI%G1LG0jTu
z9*vphHECH<)f&$Wk$!_8goJTyDN+PVXyZ8n8oL2%uQfWyoFv-Ste10Ar;Ww^v8JgT
z0qO?bH0Y+JsS299Lf-@Dc-pw{#JP!6b(?`a?vtE0PB#eVMop(v(llc}O9{i+05vsK
zjVZ{xa^}5`=O%BV=~h{<s38pFmf>JINhz0d5?a%8W1D2>!a%u&Z-9Va=Y8FCIv$Eb
zZmW4(E0f5T(y#$U5uWzUuH0zQA&3C6{S?W@LwLOeBovLn7upbt3Hac7!TABCrRIgZ
z8YMjZNRmV$Xw%bGmo;&(kI+V_qG@W%qNL2$Sl!&FscR<V3D17+cUeF427|;$iGcIx
z9%KC8lHvA4$RI)Z4ME@&_5;umcJjr~Ja&e!yq&SXdmE`G<DDb2>sL7U$OpJJ-lNea
zQY{-3bDywWUMaKL)ZV8>_?sks45~%7?>jB5cN>fo;wa0ay|1l!W?LLNS>QSTvr-Fm
zx_eNiC+Q80M3%B%;X4hNCUD{oN^8RbCOy;4ob=nO*^;r-x|QG(N>djF<D&z1ZoWx2
zK4N{45ClGTRp4R7M&{EghoeKrlNo7VG07^9<`uKb^xA<yHWi|22y|_PIKr49jPRvI
z_yVO(S4X3CCp9H)y>xkd5YCv<HLlC6-lfa+@kc_quwQ|yp{We$v{5G$g6}Zy%SidY
zkCauLqj@(5>iFn}W<nVC;n{Rfx|$OA`b4qmYf@S)JRYUmi8}3gj>^R>So0pUJ1?qv
z_ezqxE6{qOwy|LTRJO9yYZ*ITwc}n%R4E<T)QqMvakpS0ape0xN>_L~0nt$Snzc*c
zOmzMUv{HCNpp>L+6k18b7@j!~=lYOUn%}&k*qt<}Mo?7^jVkek(GSFv1VMt(5GF~x
zXcT9ts)~FzV>p{|t}N;IdnhkpJi5b|zxa8Qq{q(f*SUQ89KN@T^aV;c7O`y%`R$oz
zD~8KkF*k4ane84><%;p$lndKu@HQ?{?Z1p9fTjJ}%z`m2^VUFtq1zb_S56TdIGv{)
zOn8{4^R{xP^@1PDF%p>f*xA3w1F@*Ur4}!w?Kv&;)1Q`aJznW@umzXTlENcSdJG0b
zLVIl|s(IQbl2uKeXN>mu*}L;L`S^&9l_5zSkfn2)GAAobs-k3XZ;$<>DbqA3uQjv0
zqR}4Gj|f8zNK|d|KDsPvYK3YVw5kcB$j-B%ZjATVpbUYegAj;ile~oaJ7!=_#m;$K
z$2OBTqEM)&qN#FJRa4i>C{=skpax^#HzI+J0)#-gY5?!vP%T&J9IJq7wykZI4@^3t
zQOfl?v{(%|zF?d77%nd$rQwKjvQ*vob-T-#%c>^7;Sji5HC6&XA$HX~?!5uH-Lz?H
z@+_k)3Y%cp_?}Oa!~}k54V{mYIW+-d@DRcFRkF0e4<)q{RAr9uX$S(Ic_852K$2CO
zUwfrtZ)!ktQREW@9;)$8Q)pSEDtlj0HrzG3d9^YMErabIYHN$SsyLcXsPl}K)h#~y
zZC6R>8DIP2XSjOK=k>ig(({Qu59u!!?OS8bS|oXJOY-^9V^(BjP04gL=BvA7)&>{w
z_g(~Hvp22Bxj^kZt|=i+GP&DgvdHP#)nuQ7?vCxa(;X6A2*Wi$7`U*0v7?)I$pyIg
zxUs71%H|%c_VeO6p+6Yl`3s)=)2eyP231+F&Pk$!7e(||2BxB{P0HeU*eH}0g3YGP
z4iDJ9`8IhvW3?YMh<(bkAkWO4HA-`PXOEq``%LnhEU(CG>rn>*o*yCuAC0DwH4>=n
z(soH0itf6piQ@!62=G0D_CPiQ)j*@5!9_UTg>FJ`0Vdbg`c2q8miexlwye8&pH@~5
ztE8!o)SDIp=?4h!8${LO`x4I=l$AbqvaZq7#6s^>>2t&U-ISs8Jfp5F>bkVe#5Ep*
zN;@?eOp@;u3t@Wd=C<|4R41Y+rq^>VE9N2To{tkN?E0r=S#){@8&I00m(Yt7JYi(c
zyOpXgT*3X$b*Vepf|P~>I_f7xNeIFNU8AW{BBqeryz<J3hm+fAZ3GyaMzHtxD!=qE
z*7yV8=J6wcOOTd|pZoHdQ63Nm0qHC!&q|^wMEHU*32k$-IS17!>c+4}C^SJ3P?v_D
zL;5~4h}hgZYrV&sx+oA{1S00Gw|6*m<uQs>Fq`J+sp6m)adj=Ct_5$86z_RpljdhX
zi&Pq2m&i(Suy>dE!a48~lY_7irYkU%c}|`hSct$}Yu^Gb-Oz0^!-e_g5z*T8dugC7
za`J3$5!@#I5XU`wJtsq^Euc_zic`zRMr+&llBF%5muZ;_>GuW%ez+vWaazf}>tW@&
z;RSQAGhSWK2(k+EkSjDO!^okP#*;opU6Q9WMt65PxP22<7Q}(XQzcEQsIr`@sz~#S
z+jn-^Js2^`GUi!DtpMrchXK+HkzQm%N&%7pg~am$Dv_h=iZV9^53LkY91+ElA={G@
z;Y+GoQ8XrbZ=x7CZ$@CjrCZy1=st^jfuTfWuU4b#8da63x<n~c#g)EQp!r4_DXlf0
z^p?Dn-_J0)g{npOx*npYQApw0vb{ks!!gCwj!`JA*8&bCIkeOli|S2#zQQq8EhgNA
z4=H6~T4=P=1YtnZ@8O35jb8S??iD&Z+)r&g<~4@7v6m!x!b4!Z#9By*F7cVqzRb19
z9wMEND6$kGeVY6-Kl)eC@X1dK{_p>#Stnn3Ic0xRQ&n@ke!|9jpZ&v>vTg`H&-VHG
zNKX@Z9$^r)T;d3`OKKlv$3<7QNuGK#M20>=pm^|+_wd5YxA^!2??v_v@*5OyO%?A`
z@bISKi+2={Y<qmyM;_yMU%W#cM;z|oq7jljuj%#AQ_tQs)m0@B!lNkirBFx+(E=?!
zPx6jSNiS?oQ`Zz(hH4r@n`aEeke<y$xio|)yi*U=^@P%NZqVuW_l_UR6-Vxc$?JM#
zLM+se+|Qn8s=z&Kd>v<%R;K-@DRcAidOo#o*uV7_quX!OlsSGN(1l`@rfr&`sA}H2
zy~|s-?l8+MirO@{OE1I=BBURhjmNjFq5=am@FUYwBs^-7(bNTHk)xGwJZ-Irk{I7N
zNp(+3yhc-39#yTVEnSijK8TYLlH;N8qDRyIQq^WdQVqIp%v$h#gcl&K@ael9_vcAj
z1BEjb_w!~umh(8KH)%GTx9TyD1WyKrm|oeWc6VYNPRYqqxll_8Q-w46gk_$e&Iz=?
zb4D*qQ_3s{TOspohkD@XK{y_!(jLS8ZqbBXwjpp^&GQ3L8YLACJ|fxT*{|MWG^_Z~
z_9n%=Kn5|Z8(aLVzk41j;qU!D*xrVp_zCzY{~EsUKLRNt4)(`XWzL0*XBf{as#MWb
zMl3-zg2wSAS-E&^jgluNzO5uRmV>&{3a!l?6j{dE3s<;(XP5b8L_An0FBAgBAo1GL
zbWsToQ+V=`D_l9hLF7yJ_jeFU%+WYyuyr0ac@s}AHlZwaWO!!Iq9|fG99s5>clNp^
zrlhVKMUm$%9gH(9NzX~OcRbsBtULGlrpS`EjpxQ2i(4f4al*NK-FLUa8oqodq+sRD
zgl~BI(WdfRm6Wq7bv~ylGK#W9Nx}SJ2bE7vQarDytBT1iB})1bB;2^U!y7m6FwF|8
z#$dL6FT@MN)*}(pH=d(5v9T5ygCH$VNrI>)sw^oid8D>n<Z+x3$01QDiE5~%q?DSX
zYABUPY58@dqQ&#%A)I$!m#E5A^OR~p`ba+_2*Xx%<=u|^GE)71pGKKvV;jfc=U9&A
zjlprix2mdJ@TA+7RIL`3#EvK97Kr1UR!UhhMO#VHt{i%-=g8&J?0Za-+v)hJ_91c&
zsm_?4mUwGR$XYKMO!oj8*g}L95+OY*YD6?3)f>G0##i~s$3B9tG+IlBE9>+ID{NlK
zdF-)-l@;?DyL+1T^EI9yAe1JFBaRLyq<O`e^XEv`L$b7?$ZCYbdQ`quKe4$fPZ;=$
z)+hm<?-55(D@ko7vxC5+sT3D4p5x}5H+b~P4I1sCk)&yDjG2N;N!~bur_aZH=SSbe
z=U=={sWh_4D8qGBFhF_{=-M{?&d9QP+c4X5!MBvV_x8#KCd_jNDo-;i3;vW+(oYh?
zFyc7VLYL2IvDi%LSXGsUEmZBjGt8I^Jk+~c!%o)97`jDoXbT#)VXawX1!XoTA0JXq
zN9d{~&E^zULsh0^MM@mSjE|<IWkqjwgINW4Uc1TbH*PbY7gV;&7)YZy?E4YI4^1x0
z^Ub|pTq;T^>j8O2YD!8PAyH*P)il(3LDOiO${@gpy@+1m)A&#af}--6SDLI+Xij>j
zOX06^-#`mf1ZZlbtJxGL9Uy&#AK(W@zTJSczd;ZLFA#XXQCF?&hAx@@X`>svAHOhn
zd0>#|d7c@Pb{qByG@g)@x^8!S7pR%e@U&Gic70KC99s^f<r3%d33W@cs%xgRiKz;<
z)x3~iZ-DPRrUb`cv@m$xUD=WI6`7&VaGH8<XZEZVkdg|aJ+zFuvtRIm4}O3wpK$lJ
zUCx}jNH2~lXA?f~u|v*&c)-?{Nhbdv{`c^nuQj}IyTOklG%;R~Gd((@%reeixWL-x
zfHZ`3){x~jeyE6(2+v<YErf?{H_~_>=tkoQ0#Abwk~r>BWHsB{=Xmb9FLC98lrT<E
zHAsVBwGTPGbtrgf9X|1a%Y5niH&L}hHG(2a$<+o?w1Ud_$%>MEKB2ad1SutPuSb-)
zChwC~*Y5l(ZK}FOSumcC%_d;I!NDM=m-Gp|;FNVJmUAdV0E(uzY@J1p<5M?!i-RTV
zUUwedHEPzevR=Juiq7-2U0;UTU}&&j62r|@Rj4ATN@wKL36q0evdIBL*M|5oZIDrb
z*IVJ%&K_A_A>%%iLUZTtA-lVK%+nMh&6<@$n$0S(`gVbD*PAkH+9vYa{ICI3AdpDU
zLnH8ohptVZY*rLhbxoyA@2u$e7$y<@z(f0rK_Hoya8xKvJ<CH&qc%?aOdI1Q&}dbo
z>JnX5$f`!vhSk%Aq{g5M&jZVB@@~g{gE@WQ$M-$*VmWl{j#EpPr@Y4fy(kRRfN>M8
zdQ6+3?i^DXZM`hdNYi<%&e7$YKK|J!%Gi2QDa^8rGA~ed(;mwx(1e`=0PQY({Mr`b
zj1zb;#>}%ieqj*U2Srlr5?_YQ=2N8hsfrmjK94;96l+`CjCSuZn~d4oSmBTU!IIkt
zHE-<9_~;X7c;B@lw~i)Ea#Okw;~1rCviY31UcW(+ZL@u0iy-ot&q}JICZA)gW*R^A
zsg+6YI9Vm%6L_BKkT86Gnkey!qjfG_y~69SzQ&`EJw~m3gbZwQR3Jr7mMdP|_4x3$
zA=l6K_{<mfiTW$nck}V%HD=@6WLb`?8?-hUoTvk|Y?I@idv$!hINF}sd~W#Q8`CD!
zOOm!Qa@ttboEOJ;hQ;*?YP}G0HS=UPz~xP-n|oj6Hl4_5>ot2GiDPKRLsb<N28L4R
zQ|jr6YCIylyUQ${Q`QxJ++(o5PNM~{y?KLqQIQPR$jgR<(TJVBeX=~WM!-}<Wt-Cp
zk)CHX>|7qo8hl%yYl-Dt6;5amAwq<ZC@B#L>dH`eH(6u*=^6wLTWf~Vv)|OL)v%Tc
z?xYptQsXJptgVD@;VF*dSgG2wzg8$^#8q6P-17n}W%c!K!|`#SU{0?!RC2jEcIuA!
z?uMYQ8!OCULIm@*$TPLavSFU1b9LjmoYXtvZ5*5<&n&TiT~k|%*)WL-li2Lm&On)W
zowzP{URjor7mL0zA;glwR9d6ZcmkwuKviT_Mt^OCM;?6&DJ5^-c$F$k8T1D{dS!!i
z8$EvRb8qqT%?S@*T;a@W#J4;)<mbOIA}@2|UV_#=%Ce-)#@u=<V>X*{=E62B>wWUP
zVLooC8&jRpb%XRIRjCOAb56A-Jre>081)50YO+dm<-yC`c=;xIs*p*J5&>?h1c4-u
zC9_hK*Kp<RDqi4|3|8>N7%e<xe~Z~{PM+mPqFV^UAZTfT@3l^jZvaiBnaxt>^BkoF
zAPIcYQk@7S%6bw@$Lv&dyBnct_;u|whFi>G;JRL2yFs2<L%K~5I1k=!Z03A`x<r>b
z)qKkI;4b<205v|On2i|EC#axDZ}Tice@G)FZ@qSdd9LvX8;r|}d|q%kn%G`MSD};$
z&ubIqz89DjLTA2~(yka-ME41F#Uju1ton;dK`2{Q&m9#}qo``dxy?SSiH{5_`Vk~a
z#O|bFlxiyFs;y?Dk!VxBt1IfNpsozgP<kF-7~}h)@c@Oic&z&?y8Rx&oCby<q%irS
z-^W-iBo@i@rm`@E2Hz$lwe<pAPhFPHF*~ue$}xkM$nNXWrwF=Co0Ux<+G<{6B$K*3
zu?v^F$8w`AOA~^$pV>vHJN-3dD3s6!))4o(wDB-ue@LFCj1G2avN8SsI!T=H_>~O+
zUb;DEUck?OexD!t_ysPl`#gE2$8&EPan)YW00k#HC!eL<y?vY6WXAgT7UwRXVe?GH
z_^6~R6++2YXHaXCMAyoyKr1M6OE%|e;y45^;Oxb1Zr{1V^=sGBMFaR~rI=0TXaQ*u
zvNMKHeDizw&95CWZVca&QVpfILQzx(N2djWZ~4WesAJ^yiE!_@XP)O-3yPIikaBf6
zpx<)>2=0x?a!H+y#CW?Uv)Qyw-h<)tGT>UPWV0OIEeun8I%b1Rk2N5NVaPCv@pOZ(
zbEtE4Hf3_K$H7}S8Sm~P>cWI|YugOh&k(O}Fq+P|eQSqt(GU#R$jXX5D><4>m`x|f
z6R|pg(sLnGfDA&U^i5P@<+xgsIu?dixJ0_W1a0R<2!W8MalMfy^zso@RZdaWpd?kD
zQB<1kwUDzLJ>p0aOT}6cb0?|To0McFH278)%oGOds%Yu8r6-YLXbe<f^Jez>bE%bg
zJMK$&&#7DZzTZ&|ey4Z58}G_6B&IfGa$0-_)VY8SptV(OowhtbU7l!I=-hR`j(@h+
ziXzLX3WH9sb%P8H*HjpV?M+>NPv>mA)x7zfvaoF#TDOT)*EhDnl-S&W3`mQLy`3FY
zmNGuvVLF@OM}5Mu&&F!P)pIMn@Yaa5C<&sN(Y)sKuT7Cs@YL1BhG{<Y>69P{h~h1J
zy%i)C)5BffdgT>f{K5;&4-eT|^%?ded%lfLltNt@{F%vBNNboqicB#+n3Ja&{nePk
zaE)S~a_zjyV>Pt_xC#W*$&{D(YT_v34}J0}l3vf`_UaNZig8LTo)ngNF0s|7lb9p1
za9P?aVpST$Qdfn|UugRMgu$TSa;IyXd(y}LTYyLnWy*AF3NtQG<JL_Ogq?dUmH@+@
z<=L*S1>9vJ$gnyb&<lJhGbpAMle^6B-eh+B21jq*U^W_|`+Wu%ud#LM5&9b!m}L!j
z_eNw{h4w?HR!}vaOqh%gX)4>~3<&8VJs;2WZN<*F=h>OR<8^jtm5x804k{o7LV9>X
zfEPyiQA89aNa2%LInz92e>~^r-7z=!#^jks#sR&a;>wEPp{*X9g9wj?x-#syRaH>e
zCE6-kdp1!nJi}t*&Z`Rrdbi`gJgKrM?ZT4;zE4$FOP_bT7q_5BallHBO{zwd<rzg*
zplZW@SvLj9#J)gLS2feA$wT=<;z<ZPRhi?`Lgu`dk>x@N@}i(9Qnad(THy;x68Gpu
zJ$yWrUOqeX{+2M+E(m2BdGicurF*=<Zv{e^A3_5#)B?Q5AaGZ=Hpq&c!-GAB!&Ump
z3R*%O2>P+l-AMsnK+}{+ug`D2H0Hr`30o@x4{Qlu-|usfXQV|=KMAP)Re~VoaI{aE
zr|jH%lkwpp7p`98%=t6)R}%7E(G<$!RU|4D_!2y6LeV(#(EWs>tSR$~$=!@Vz|A{v
zb7|``TE%2(${R1e#zT)kLcF%Z?or9@Nys-ny2*e04}OwcyCc@uxA>-~KY)mbG}C<s
zv4=lgF*~VneQ?r-XsR<$77-^*n^Ts~DD%vOD8j=JL&IhfC3X`r5Xs}CO}g8ix+pTr
zGPj1svx2a$SsYpdEMY5&9Z1A+z-E?Z*1J+fq0j2dfUVf0X)?5$GdtR6c5s)YyF1*y
zv%@4W*tq%t>lZH*udX9x$ap+qZ~u_#Jg2A{imJ3Jj47j|L!)yieY7yxCFuotfvw_6
zlPFixbbAPEtsJAN=uBh_5wfyl+9s#k851T~8u$T?^{$kZ$VSps6=jhzZhWc+=B4K1
z=781MLkdAJf-`+dReS7DQ`0bP4U~|cm1T?ZWQ6be3l?AdZ0YXC^e)DIc~VLl{dq&B
zyO3ydhUYjE+a*VZSiHu)Zn3D$Yl}Q*ZrLmAx~AzcVJfA_Gm}{NJ>Pf`VZzGF$|*+S
z<kh>yR#g@0+~mI8c}HPH5+_LCGn`I~H!UBp8|JzEMu#3G>V$dFaj`MzYfl9rDh9(<
zR{DJo4(<{qeZshBb{tjn?H|0tG^_ZX=l75p7)3gt5r!eZ`NAOpAA5Y0^`v6Z?^Dh)
zno83&g1EQBg$vi%+P=WnxeLs*l$V}=jxYc2Z}a9$FXQWsUSCp4O|BKXhNgz1&_>Eg
zD})D0KcF{=31XkXm%RDP4c@$)p@AR_sC2`#pZ@{}2Y0!<zr)Y|*4F^|PyhJ0bLHYi
zHrLltO0atVL3~dUub!djhwYAka*H~?#!P8{Hk-AZjqe+UmX@0ymNuT#?!BXa&a%w@
zY;vh_oG=&+Pk|dO-|xkDD9hX)$56-g<A}tMQF(#XIbJa#AKhVoaGTNI4sYMsVWwlw
zJp3eE4?fE3*~`Sqkm+p3{=os0$<$PYi-M*y1(5O4k(IjwP-vlT9xA}|OzFOrIM;en
z0CMNv*`T|XNfX|QHazX-v0UUKrNj$D(}Wx(_+f$;9z|6#n`hiToO5Gm#w&MHUb$QI
z=3&j9S;?$0sTrDzx~b4;geUQAajmPyY)(gG8m5*vsINEf8z&{DN23%~)linrvOX6l
zJs9njO*x&Y&b{7%It}ecT~(+?*@~M-S(a2)i7$*7AUxB&d0HOo<YySljJmE3zROBv
zX>As67)JPEU~af;WjSRyOixu+miD-6^JSiI!bRzMG@Ybtn;(!8(P)&E5DnSe*)a+B
z;VQk}fV!%A_`-mTTYWzLr5)-<wtbRyQ=@dl?$qPUuOIQ$18Y2Wwa3BNro>Uid_KjG
zVwwi}NlcFw`cksGy20W8E|bwduYc`%-gxm9E?s|s?emuytPD-)SDUSEJjqd*@H+53
z;y7YuJ)z!+`Q{Hj!B?MsmgXbhgA|gDwRQ4&%E8VZD?5I`{x@CW+Qn@ie&9N3sc1B8
zTzUdMevL4S5sfw-L1rTFxxQTG%+R=W_@OL7cR^aXe2-6CS1tPAR*FGb(!#!b*U}ns
z>!{n+;CUW(Z76(IUD~+`iKU>J9&vbdh(3}evSIJmE%rx8jE@RLf1R^euhTzsmSk-M
zf#mS+T@FV_9F521d0{t7O<5Go##3_328sY}jf&?72;1i-Z2pK2Pm@?WdmBM?o`W_o
zWvkOCo{^>Mr9y-f<poHuMtT8Fl~dOx^P-^A+B_>!Vw&x>Hkh@hqN!|z<@pA~6&QnR
zfW$5!$ANgd*7aSD`y;&ze5y(s#fu=o^8@Ne+2pdTvMen5J0(Tk&~5@HOsCUMdyGM!
zOHZ^C>z;4uFZ?hhX?wL;`mEE8l-7pS%=HNB+MJ_Nd@=1aN*MQ9wRsWQxt2v6PRkKx
z7la=M^ei~h<<1tL>=uDQ$O0u-ILth%Dq~|~izpeQfJRqru5KcP;9y#s60uUI_Y1=q
z8H6i*{*5`0U0&gliz{T6;MZROokD+g9VH}9Q4mEv%BrF_?9=P7Fk9QC%rd6432(je
z8n3<b5*IFB;o3uwkn{#<ByrL+TwY~KK1)fbb9#e>m5l)x9=b?XW@rsz95Gm5B_55~
z+ui5Fh4UPUn*a4vU*ymK2j9wj9=XW#ZyTWi5pK|5zlz9r(88#fxQFX_IPS)AfG$}2
zmHwdJm5+N)r&R|Xn?+F+t+5M&h&Xn6s4j@~^ruCLg&>1&YGPl|i#&X(8SUL69qqGs
zXNMzCkt6|!lZ<)ovvU3#{q=PQ>l>`BZcsIf!^1-k$0H7p#!TlaRasKk4P{j_9Z#4|
zr=~&I(gm5EPKfj)JTJ1Sa*5W?6Ka;u)6J=aBeZ!x`&{Z%=G}c4+ETIDYGL4k@(gLD
z=lN7+j;eCXqCzQ%5h%|VmsE+=4UN`FFCYkGyui@HV7nfc^2C5{1+%_^d{Pc4<w;jD
zau@y1=JDQ;ghdr4O>^?xR0vel&{UPR_KLta31cCBf+${s8!SN#j(v_mQdbr8x$7ox
zP_;>Nhf##|JzSr6dymb5?CxWZqP3zdOjo)?6AK|)S`TR-A}9adG9Dt3+En-oxydZA
zNe1hL@et4VDT|EI6MXEcOWZlkc=K+K@7cDWx+01~nx^5cH(o(##ZP-VKlrihT-%O$
zakpkid!+N6)%6W@5TWauDDE*&bA~G`cu|ispV436VC(EzvLa(=XP37={YzZC{s7k>
zeh4o#oZ`M85)K3AvoYiSLq>NF8Ep5t`uIZxfrnSa*~^<my?_^=)7-wh%aiYWk{|!k
zf5dlu^TT}nJzM<t^S8Nt=`5A@k*gO`#V($z(MXi7>`19wI%T(hjA6;h^P*+U45N@<
ze?Sn0oM!snd%L+KL#mfDoz5(j#o(cmB<`>+Sji}NuR6zay><dkna$B<hDS{=2#{sQ
z^k~GL*I#3I=WVK@!k0e%^$mub=ULl2N3y;_)K5s_1l?$+<54T<Ih#$bfii-%lj)Sn
zbYfU4ZQfQ$AK!u>rR#H(ivq!^#V8xnx!l@`=Rhn70G6JWZb?(OWe23^69|cFe6+<?
z)n!RtR-|QV_F^<>h3`pI-3%jwAjA)2loyg$=2*V$vPqb#p0=r?cRTJI%qgXzLk)bd
zt?;=?>E7&a+5hVex5)CMBC{<wb!C8Iz9&hdfVviVet;C7$>H21ymO!D!ZxG$GPk8;
zWBi2liQ*VP3|dBnV*y0x2Ac0iP+>+>SFM2x!_c70J+D)#K1Kl}Kxl=Q3Ar57?+N^5
zlOPVLE0b(IyV)o3JZ>LMOdigvI2UDU&atWSMZ;TfzKKV}2Oe1G(W{$$*OM#!v(N0%
z6eW2&Ww^Rd?MYPCu(Gm3Ru&}vm?++$DoX02ByMUp&s=0a+T-@^Tl~&%?r`bqH7;Mh
z4kDmY+E$YSR8z2jYoBy9<$Y`Cp{aQDrJFqP$YZ?!>G$#6SD)w3?Hz=w`5Qm_ll<s^
z^A|Z2&8X)qr2YYeWSc6vj85N1sTv^ylvhzH3xm+A&5g|Gb7u1tSD6lcpI(1J6eo*E
zX=y!m1M0<`J8pXOGJ|aH^?FMZWbM7dI=(_8aNd<d2u+pFDCc7WT@VF6YBuL+Z<oWJ
z9d5qy2IDMcWow&_i&t4abAgr3Eqbe~1c_nu%=3)Nbj)Zx;qdT?$#`O^fJ&OGW<F0D
zPsSA1Yt(3vvIE!*kiOSeMB7jhPSE+<$?5#_-f5NN&q+(fVyu{{nA$v*6oNns8c(35
zMAZeVDnT`d`^@tV1Y;00>?cH_X?e+On4p-K(5MEXy|%LGNi6-+yN~zoD{=11t_FhO
znHaQNRlA>qEe>kkR3pgfNYNa(xM|r)@WU9-5AnmuklDY(Il5sCU%}L}EU4O2uq4S6
zI{o-K4PBnwFt3};b<$Svblt9<hIk9*M?~>}Fb<GX(9|VT!Z$r}o<?cD^!kxaj+&iK
zVj6#|$^h-wHr6>BPxz@{e~rD-oYkJkH$5~&>Y8jerO4)lVL*_CXbJr|MAr>b5-?cp
z(_dXB?ho<&9%n9I=i$fR%k}FIu)lMMFFpG^%qIuvro;;(;^78<Jj9cd@!<hqe0hh<
z=hny%cKOohzl`((9(&?3f-vFX_dUtYx9*VVB@bLYLouIF<|%0b;pU^L{#8SCCJKc1
zke-L?xciPtpsH&W4M;)QOX&6c9U9=pX6C{)i~MK}Pf-+1r_;6x88k`KV=yq`;&F(|
z(mfXvr5k*yXo@-W!#(Qhh@dH{Mt2##^*VQ6d6AtPZ{qm@7q30Uwa1_0>;sRnb@?jm
z+vgdsZV-kMRar5ePK+G-;fV2gLYn3jMMY)Rh{xljHmNV9sjeALDC0qSeqiW^+X`N%
z^wcRDI0(mSg@kt+r)~1b6YJ*O%y|V-Oc*7EQDSITq(n(UqoHUNwK5IHfhS1<Sc@cs
z5JHpCXj3?jO!6CsIBy!&CKUZ3XvGd4ZfV)We0N?H(0U(ADbl&qDHHg<M;KbPs#lrV
zS1hB@PrE5gSf(_!Z8oMCEDob%9M~?OM!Qpi3){?MHo~5ZK0zmw9VZD`BQ1}8_R_y4
zl>RzeB=~-WLQ|CmswvqV#=Q5zExz*lh{JgU2%`urB~58WC5OX4c{U}ALhjz(r#f?%
z&pyA)_kH*p&pf!o^KTw8$uyJkgx;W!5P~=k@I8>exo+u6q$lvhh$6438$q<PMbI0v
zb@n_r-gu1{p8E<{uRp-{xr-=2!1D(LfyVQN&E3I;Gw1l(U;hf1E?z^`HRsM=APxn;
z{PVxSuYUF`{ITzRn*Z)!J&O`Cr358>`e)t;QAj>|6~7J<R8-n{w2t=KNY9&xBaR{l
z{UJdRE_k^(EoHE{e|5d!%QtJ$u&Nr?&xHip63vi(_@%9m`81T-jKiJV)UyeRlpO9K
zad>x!Vm@b<75Kw-&R)90@Z3eB)eVO0TSQ4>)+L~-lG%L1WOBrKJm%==i1}<zX~l2L
zs<f2ZR>V|D<I8vkTjcqX-Tb}9HQs~uvTK-@lIQn$9!||ko#I_BtzirkiB^Wf$k_m`
z6-}i`n}(*TDJn%;Dh9oTDDV*y`k_SPF)vh$22e^beb&1f_o=(5TkdbHvGQzaqLfXH
zzDrBq9nO=|S~H!Q<hk#A1Yuw*3PFex9$8+r6{6$e<Z)x=O4)gu8KnxX@O+<muuL&4
zoy5;EBkO!a)3P%&^fC5fiK337GatTm4vUJ3$%SeGwKpK@tq_ZVLKkFtMO7ADJ-dZc
znp+2kK|u?FL=c7nRW?U>9tZpTcv2eqa3M(3lrOy=@W_=Jk6jw_Js&*F&wk;6X#w;3
zoYnOebW@QRB~_{sjb@gnoH=&}jU)^uGH*b7lnAENIcK-e5(hp9ySFLwlyjFZv9dm-
z9-g7Mvc}4WxvF$NXMb-GkDQQ(Ng-L^KFc#7dWOID|Nf_Z@3%k0C!c<hPknI*RfKF`
zy^4%{>Wy`jh^a<jLDm9+f0C_Jo4FnKSBS%CiLAeycWTQKwz%O68m{5oL01NYVSBH;
zUu`v5gD1eFLFfimq#WM9$^7V$P-$-M?2;ahn9s%}tE+5Xe}w+lHiM0A;?*_c!I01o
zEKj(mE=%U=l<D+{!^1=N_xG7hCMJbZR#t9p%4{~paNk?Q>3Mizg!BS~(Dpp*A>R90
zTJm~2bL76i)8_5o)8?rV_lyR#)+W4bY+qhep=#TgC!}SlD5)BQWh~2*VXsFV`gl?j
z1%^oj8mh){HQbMZg8QxRMFB!evOGuW2CXUthI8n6FP|G<FFz$(7sf20`8;KwW=v-@
zi`$5agrKe~jQ+T^?>OB+Ex>nq%k`s`LTE^mKE0%Oa`&+D_)=VtMl}sZn$lEdd!CL<
z`(8@OwkAk7F<6R(AQ_TGJ@TeNWkx+iN}ngLZlM&sx|18I0@`W@f)@tVRl#U~kNIrM
z_StRHJZD~HM1vK6>%|FYR%6a@#9Y}5_~PqFWb+9(UU`nYdpqpk-Q~>I8LnP=h^#VP
zMJubDgn{WsuN#dYg!FquswyK{S*30i>2%6lFTc#{nJq3{zQEdQLSifT>t`?E2LZQl
z?y<SG#opfAY}D&qeB>d1`4>LTzxm8p_=6vLFVEjh**Qp=9FFL(44Kve;n}B%y^?hI
z6|_(kvy5_<p)KFG@U3265*lvD?xxYnMJ+s{roGp**{rSP3E^N1F|q}aYLt*@VN)Wi
zCXgOJ4S}x6riUEv++ucj%xpSCsFG|p;b?Tg%K1y2dEjCCXU@@ES)(^tC2AFj%-XN(
zio7fsPv?w}rc95f%*GSSJTtsXbxoQ%l^jdvs7)o$wY>O2WRm9A+q+lmA>2qU00961
zNkl<Z$Z4UP*2n%~H(m!rXji1I{%M<kk{tu3an)@H1T{K&O;yp<<~oEHctL{pJll<=
z!M?_5Xc}lszn9RDBmBT4LWA%qN<~rERL2q1UuWExk&5FO9d$XRji+%hZmHu2>Yj#5
zXuE4@vMe`Mo+76xO3KQ#huqI_O`s0C&@@e_1K&SAZa*HIqYZ~q?r1$|*%q9F@oB@A
zw;ktE-PEXPm8xv;)f6QI^7#=(k#qIj8ke?*Jona=qOo=o4a$VJXu}EQdy?HdcUW6r
zqfv@9%~)Bf*gKr^wYU3x?D2q)zh{#hH*WG<Uw(<#-gphq6GVZ>-91I%`K)bh@%Edq
zas9eS?nOjVL|HdR8a*~GK|vlMm0&JCiah1+&JMHbgmV|pac#H_NVd<P<IyLc;P&lZ
z2E!FpQ?h&OZ8p}|2vOX*J2sVAp{O-Xvm6y9C<%kjA@S;mm`)BUM{kj(Gm54F>EYuM
z^?LNI5U11L>)w<0-gM^Pg^S%a*y}l#gkS+W#yl6Es)>C?90_DoAXSMjQf8A~?!Ngd
zJ3B{A<{87aKK<T+tw*0=>&i7&&R%4&x=s`)c+y`oB3d_XwQf8fb8vXbXf!g#nyRur
zf0=>U)b)aM-7{gH^nE-T;7PObotCp%hHWg;D~p#Md$z2ncs#$_{!VLGd_MV1>IFZ!
z@t$((+C(u@c*r2a4?}ANwP|1%IaO7X<+&+(HOfF?`iaRUOO5hDA)zr;Ht%-aU+<of
zgnmFAMMPnAf*kI6LT_nuxOW`do#?VGmgt0=x@mXLlgB)nDAanvUF}Mq#`W*FiX``1
z0Nqg*rg_*2Lzrq_L=yLq!nosZ3tD<`EU&q=UQtn2h;R#qkEEuYr!-}TFD1`Bb{-IX
z<;Kk5M}$B%H4;r-m(;cC;hN8;cpkj`@=H`zO&Eks$43<ToKHXdGQa(`TSS4!GmmVs
zf9FkRlMz`u1q-1Vj}8p5^WZLZm66TIOpgu_x+0xVXmkU<M{i|?jdK?`d*vz{7cVgA
z52=fan>TK;w{sUTv@VZ4`WT1fDZU@l@Aa6^rsV0IYfn7NFZ{+!OePIaY+d3ve&#%3
zv`y6Ov$+!D)ivW(p^B2i_0tLSKqOJjup>t$-(eGQ_e&{d0i}7HKsKs83BK=HIDtgl
za#%pk(687|;Cvs}{FGjEgqrTqOz*P4bDN{lh-sR!JD$=%d!DOLJ;T=Jhgjb_PZafA
z!nTEhGRz3m>6FoE#Aq~PfB%5dctVz!6jjYUvunD@P2LKVa}<7n7Z?B)LF&E_%5!wT
z(xg6?zU&%v&w%S4bk}Thr|Q%%hwj!|p_S!;Q>uNBYSbbzuQAXW_x`3v*Cnc{KnQ{)
zp})S)>eeRfTbrzJZP6dD5=4pV*Yrb_kmOa#(QLvbO&L#9>tWS+K7_s?^o)MwyB+t<
zaqcP!Zo*upZ-G3&zoBhBzX0MH<&E01bT)QDEF}`#oW#lV%UiA~!=wO4q9}^u69%pO
zJ~vz;Ak7w(w?bH%?B1a5nm3J__T1Zdi`FBQQc@KKd0vxvJ(7NpVsh6M2I__|iFx3{
zDtV>2JIbuRGogd$1x7iisZm-`6d8f<v9`6%aBUNfhZHcG9B}8>O=Nb!6Ia)H-(#2g
z{_l92AN}#4qn`|HLRsMre2P3n1`&t*`>b!Br7jB&c1KKRQ_fwuL}2ydJU^npy25aC
zKy~IU)6s;({kv49G1jWC*xXv@;YS_=FJx`w464z*dE*Vf`4b=G^_TbffBjAPYyY<p
zZyfN~|1|v7e=ubyeT}*(84UX5rN>bzppp2VkylR!eZnX*{a>xeeR^WotP>++mSrhm
z)`;O3kFBRG(T&Cv3aAN$W;GV9M}ovFk*Y?ilx#j?e00FPDA~Ap5xJ_lb2R6HCm!SC
zgOAc*Stm{gZ7-c`h3cM%3pc0JDF+7!OePbG(r`7EWyw69ljWI(znJQ+wCHSUX-uT$
zT5^(Xt><)X|2$rSYo|vq$nLZ@46NP!oz^Dh(Ski&@6v|aEP%{fX`6#GO~U}a{*dA3
zCTp8ptgWw;^o$N=T~*Y1MVe>KrW0o45wpn=*?h)4O)1KXB#w!a$mEx$Pap-_fA?1E
zzKoQcAkSNXIvp#ft1y^Hbn*K8O_I61lw(9_Eb=@Iy`|SI1p`afyv+6%8cwP>ib;Bj
z1-sqrx}<fVwe5B{(pIf>w=lhVVH7Paqpkied3V}gyO7jPV~nkekbcB`W^SPH1b!Iv
z^rPok8$^8S%LgX^r5oz9B9Ks(CM1wjGMi19r*r!KKB`faMM)I*@I8U=X*^x?E5Gqs
zUVr6l{FA@=m-r*!{Vm+s8S%=^Lo|}6(I6zN>l+{>^E72xRiGL|&*+LA?eDU&b%tq{
z)9VeHjY=cyzPZKLxh>W<*4SP*p2Ok(KBFS!>h%XG@{GaiDjLcD!9HbGa_8k|_}p_o
zf9}t~U;k@@ANnE9PyOhGKm4Z!GU+qiJja>K7r8Qjg4cfY-%)4#40?TnB(hU$izo`K
zCcIq9azB@)ZC7aw?WgBS1_P6GRNBBdBpOduM1f)$!fN7^P|_5J(s(>MVltUiD1nYv
zh&TJJtomI3z$G@e&k-j>!cbZifTmPNn@`yEWmVNor&A6O@3OnQ%dK0txVyJ+i%S|+
zEAlKSx8f1ff-<GW8%i(2^DJAX1jmcjeP$MR<ye#=+C9<?xqw$Fr3|Ynx&=}$ZP1-C
z&~f-{Z5FQt8dbL>cv>|`-yq4QAAk@jT~Sw}ZPy8t2tSGF4f+g*Jtm`s>CuSkXhb&8
zNmWMO)YMT;6h;Jog!c~5<MDA{MoQH+_&!n!JYNt7KE5vr0-wAPM#R&-xKkEg{A#%r
zy7=AVL03vJpXU~tTeYtBVv#CzYYNSx^@Ql`tT|d6kJj^i;v~VdSf<6MbBf{A#&tIN
ztT|0n8T^Xqko0?UQ4O|>M0SoT08c}MBFl2}JfpY$0Qf!f!@EcsfHpiko2xxEa4@T=
zF>Nq`CsC>*&1V+ZW2&Z&QcPwk=P#_|Nu#HxwdAE2UgB$CeipPs<>&a7&whnZeB^!n
ziGT0g_?!Rumrz2YR83VY_V*9bn55?Va0MBLCj9U`Mn^~V`hDuU;%NU4p>L4cN24*D
zo2zVWu95UjUT}4NgFA1H2>pU|Iwgz);w0wcxpO@K{MUex{R6mt8@}~h;irB|@L&Dy
zAyu<aHk+}r*>E&YSzo)%10VWsZhZ0I(2rw;AGAhZYi-*=IuC2--&I+XPG>Zgkv>yG
z5=8<1Bq8!8zGpz3u`fxaB$5q&T@ciQvP_vwjyRf(N%Ndq#}vVU96=DpTsd=&!P-`<
zqS6g79RyKh!lX2vF&U58-`i#P_AT~z_b3Y^?x2;X$TI@~u1fPxAuV7tK!y>%ALIE3
zZeWhvrT(?oZ7b$&1)R=8?*7|xYT3Dywt9%N=zb||m0Pz4$;Eh57^r{|7A+9epc`~u
z5y*xxG|UH~RQRf<$xDg`gs1U5g(nS6L1~2yBI5qg5Lng<p(RC{Qf5ZnPb*Cm34-Iz
z$L}!i%SaV@(Ta60sD%C2lj?%q+a1u&;W4<tX?Y=M#PU36o~Be)ZREnM3gSii*}10e
zd+mE1W2Z|zb|s4LH5}*<VI-5%bV^y~mat8?>K|bko~ojJ9Kqa3O<k0f^OVvPY;Ql#
z-RTTN&}T}_YpVl3{P-De>}A|M$ZWD)(KHoW*Z4v*pU&}p$z(ENGMRAh?0H7xBR03T
zO|GEP93I?dKAW($xeZ>(Kl$m;@J;W3l*g{0=L1h%<EyXk06v~?<ju{wdK?_=fmHN+
zeQK=<<ACvGOdLdPY;EKFAw^k|Wn<oaZI8F!@EN}EDo<`-B8ozWD+AKmgfNt(lPN2!
z>xg)O7sWjO%$NDv=PvPQ{%p*@`YFNt-w*%Y-;DSNKUWcig5Uk@uhQ?WaN+z#)~;P+
z`@xSPj-I2en|5uPc|E<9-ZTxf`P}qR+6CqZ9;+*x#DQcOOU|zMQALVSHPktx$*I!`
zhgrsCo-$1{8WGU=8&oP_(tzj<SX<j*u(FC52F3%ivQFLgtS#iB$aALS2?uxg+1tI#
z-rhcCSs8<)jm-UYI;AK|yJn@qQh6aVjPQccHsUU5hP2CucRtVMrSxJiw@L4>lk0Vc
zO`85Zq-b$y%YatXd)uf6rAm~l(5j#=a(aQ!`Rxr>Rsx1`pMHOZ$_p8lC0SM?Z1;!f
z3GgB`;Q0Z){s0Mvy}=Lg1D|X@qs$GRbW<y;cK~D`ANQS;s;Ua5jmnDS1rI}mLRU5~
zrH{R5F>l>idhzcr2F6ir%w}mD?kQEHF`zzEzGdl{Wl^*S%3-#g_v5HuPA1rw&r54%
zC}J7|0yL9sG}NLfvPnqq)Y54;v0v0B**vGNN|G~I$mIYbXEdt8mmWbF@vZMYPvA*@
z^MwOLbpfCZzq=Hs8s~!9x~^I&@XhVB%+u6tidDg#n>Sb;3<-jOxYtK{A)or4m-+rr
zypKQn-5+9FeTlo1jDgL+W?2ea;spUzCFv|D84PHelCYl;^@cRUx02m`hHG1Ro=1_-
z@ol*6c$JF0q^@(M@QH1yIgCO!w={qJKlvrT_?aj8D}O!aAN~((T-+Y=XTIxU{y+ch
zQ=GqWiIw&9h@ek1pE9`oK?)|wgBxaZKem}I-y_X@KDX4mW=JU+4hHyWWL;2<=lH#b
zCYw>^DNU18<TIv6BMzrI1tBsX;6*DGI;I4owH|)dBOVO!{Gj!U8@;@m-G)|I74zAg
z$z*Idkr`FhAgtb>OQd%@6pX<z%$lC<%@dwyp)>%Z?e|;?aqTtUYok3L2DfvyIP3Oj
z^DzQp<#b%wYJOH!m4Zgn)EZCtcz#5sJUp+*>Trc5@lc+k-wP;8!6a{(mo=pqB1C4q
zMCs!RNcw%eFeZ#6qFzF96q3#66j?@-%?X0=#I@XO+?SCGqYy0&ZLllt8^fR$H^NQL
zJu7+o)W3IWI4Xx8w%8t3*OoZTR>1V)=G3(%>bJbp4x#Lz0>{0#lZi*y^R+OH!j@K}
z$W$<02mWbS)IFY3iZab8^2*%Il}DN8Q|iKi6?=mrgTV^xD+y`YP${piw$x3D=NoXG
z^gN<CrYtii6GIi;@AXhxk@OPEEMw>PP3EHm!Z2pz%r;sGin`&~zVI4PJaC?eubkoA
zK5(61eD*EFbuOT&Gy;#|>IQ?skhniUYt8QN9kThHY@XtI9%~!x3<g6g6{5q4Fd4A2
zX2QKP&*?=mNfa}l9I<oz4m)?=<|7||nmb8MrUkv#86W)i&l4sCe*9O?^TVHfn5*a3
z_)CBKyZG0iy~DUvM53ggH;mPk{^onAX9wsi>#h`B;@oXwX_}f2f~GOt=~B|~50F|g
zpX`wpQ>w`U)3-fnDket<_<>-!GN4dUdIRWfpyO4t2DBD<p^xx=i<Azm3#SmmTX5ZX
z)^5`@<Y~riGGQ{FFd2_2io(#BsLJL==T^qcv~>Z7awP~_j&jcnFygh#$#Ek?)+rb`
zW73%!#Ys=<gfq}FTp9zLl5LFF&Mw$6O*K|QyfWUD9|eeBLK(&!)e&TxRjKI50in?J
zV#!JrFwQEbMM_%N)Y_+(A<C0x(+WepVAUkeqlh3rVm2L9&gbMs_g=sAxIY#IVWe@4
z8n2|O6=hjd6cs@b7_3Tn=g|7p2iyJqUTATE+1y}EYHI|Wsv(SgqXDcJ?G4V$XpR{u
z*PG|O95>9OlWk5(2+{Ua6=i9?E=?Q;rrp8s(6x&0EjHJtaB@SoxiwD_TB4F6B02-r
z7>TAgTp^14Y^?TqU_0XTuTMFgrs%r13>^)sF>>Tu7+!9rDkhUDd0sM0XKeL)RCUSd
z?hda#|2$V7d5mm6;m*!3QL=(ou)9Cvr+)d1{O5n_+xhT&E-=n&KKH^NXh9P7n9t{2
zx^x-8ze=SHE%3veE7aMXbe1xkrHn^oX5%A*(8CX6f~Zfvbru*hogDGPi?=v)W{Wcy
zui#6MpZe$jKfdXiCpo`$mRE1xq*12D>0tj3ufF<K-hSgH{_=nHJzPGw&W9e(`OFJP
zcmkr8HIji(FJ5IdI!|-(s<}T}p`;JOHwFXa<)*VKb)&2Ypy?$co~oJc?URmo$PRZ%
zk47BbJutZzPqTS;n{)9x(Qu2fzl!pE6tzN0+ioKSgMOc0615%;TA0MMUT!>BN|EI$
zqwx{@qeJ%Y?sG6YBF~F<gEWMFrn>AB+|u(6&vy{w1%@9ATVK<{J|o&z+O%UM5cZj~
zi%*+1t1+RUFgrCaLwmxT(sZsz+Z2iMoTP2MHot6Wn!<oK>k`$Jgi%N`=rLFsGF(|9
z=?zg*F|HJuf?=Q;1U0cxoaqTx6T!5AabA%&lDhGz1xR6<kfXSV=LZCFM3nTHA006p
z8{qM~8TTFTNo%EBhRi4;3=6^_AQxg$#(Wo5z0=%O^Fc75&B-lknZrwo$WoBn_c>zx
zsw`a+gV0h6A8$-uIzFb%DlE;aVH+)x!cbm#o=300P&RS-uSI@sc}|Q`F<eNhG3`OU
z;UyXrT1ooDHIl(9UKnxhT#s65ZXIOQWkH_J2s|I8rYsD$D=%$0W6SZLXYv+xl~WWA
zpa1M<*n8^@(kkVN_r8z!ZeHNd&Ry=@-r?e<i@b8<E<gE8U*Lzo=R>^z!7ZM9b(gGE
ztga4dnwmG?e1rAPZ5j+mSl|h&Dr58P24}Baq{wrMET=5$Rw{jGT#=QE^OvqN*K6$U
z+~&pSzRcRjI!}G@gZ%Dqe}(sd>ofF62fX#_%N&jljh>z7^VZH0f9I#3<v;ynALW@x
zHaSQ&Z>JGa7=RRbNP=*KLMa+gB7Gmx2!v`tK;0-N(+Oo=p_L{Of?+?QDpC#)N2HS@
zR6fSj6=jivP{f0ER?lv;dF~wX)_J1-Izk5Id5LoRh(fk$fi@hX3o2RAMmxpq*=)*W
zI%PJQlBO8~(Kas^HtkT?wF#?iBFXTSd-$Gf%y)UNPU=Fq3Y%&dnKN<=g%!86VtAmS
z;~OnML)~L9`2bp3`d(walxon*;Gs|rL15S?2CD;xD+7juK7J6GxzI-WXk0gBjb=6Q
z7zCO`HYB}}l~{0;D<)Y<Q3*=vw~>MGdkluFM4=IM^*o=^!QQEM;rO^e`d~D*RtAIN
zS+C0XI;;q%XYAh%KI!-<ijs70B#4YRWpX!N3YIiYDQyDP2zv%g$}#u+T<1e$$nc6h
zL#PH{3PR6kkn~BCp3N}``+dRKc-*~qR3k-EAe2CdA%oTHRH~pq*q}GuARex>GDvvO
z#hABt#ytDg*BBk_qE$_!>ILFhLlPw@8%j^7V}igVibATo;nt0t<atG$Buwt!<&_s+
zWip-e-lv}8sSkXR=`3aU?g3wVb&vgV#`%pN-|_x)y#DIT?CtIlh9O0sF`w+y^C1v|
zrc%@uJpYxKdHyS3Bg+bcC}FU=O0Pd8@I7wrO_}F4?>X-y!i0+tJj{hFm!Z~NefS{;
z>zh3P?3a1);d9L9M}%R3@B-rgDy!?;9A%m>zIqoa1mFGsGh7|yeDy0|;qABYkmicj
zbJvL0&J!q!UuZn75v>YxN}lDESxO{o)?=S)a>#6dhw1K3vV%KJ_jl1v#oF4C3s*05
z>G3Bx``G(gx$+3TjdRGjM^-n~O=aL6o?&HmT?pMreHoDKsHaQj^C^>~3FCu94i64V
z?LDvSx;0R3LSGBCkl^{IaoBQ0nfwy%rYh_gEzi>O^KqiUgfo1GIYA23+S0`?c3#|m
z32nT6A%uO8X@4;=1!WYSRYPDSh@>|(T7pSH6bhoyBZxe_K;lIqh48ss3idM1yq4$&
zVo|f*tGTk~v7Kmovcgk2G=}0uH_C{p4u=fa*6;c3A0PL}f}j*#NTtI_`JRt24H4e)
z2AM97=w3$YG-F~!A!n_6jI!u)&-W;c#vsCN^RPtVOOJkH2=?E}gEBt@n0~-4OR1Y0
z=}CkzMSysT_46cF#~s7{+cmcW_)^luYv^DD&-WRwY!LV!)5ASB))X84hxq3|^CiCc
z>=$X8nlOsFeB~PF&s{Vnc;$9ZL6&9sfyDQvrB%+k^Y$HvYg@!;x0z06R87O1H*U}f
zkL|PPc<TL6b2u9D+{<t9*MDrz-}q0ypU1A9<pYnN<5zy?C4wMib#)EZY*JP={oxve
z)h*Trs|<Pr-n#JyFMR2%wrUm;ND#`X+}715z9)J0g|D!E;UZ_woxxA~%x>?|8xDE>
zr5Abax##$%Z~iD>{Nk4hqKJNfl_(7GbjeTu^0Sobm_PJwPx9^W*`z8IFYg<X3WjU+
z&ppBT=-2Q>Z4$OpP^7uR+suv_gdS^wBvNzc^AYLcU9zJ?viXe3bV8*zx%%KkBx{?*
zn-}nUD};fMMp9%cg%xB#3w+N|2?s&ciZ3{WbR7CnRR#0;%(5D|mKW0(+Gs^lmP{rS
z%VXuz3j$#!sQe%z@WYN2^g<Qq_;b*DLN5Qxa*K=>3v;SFvS1Fn(K&IJgH`j)dbSfn
zX@#n5n$qax)pd;&lDZKTrQyC0OGOYUWGIoohc7TXhf3l3l1d4t733ML2AY*9AQCl+
z2f;8RDix!wBv%beYZ|3dHiQo2h<7{gPc5c0@OY!=Zsd@Rk@8wXystM-tMHw84TkgF
z^xqleu2zQg-1h^+S8k=UJx9(J$5uzpKNa54{cQ6))rNV(tSN+)L~%k8MJ%?V#d$Aa
z6^mS(3G-$&jl%OJS}2m0bLcQ;xUxZ(P5AQfeubC5^f~_Pzw#s8**oOZzwtTBykd2I
zjg^%(s;Xc#-l572GieZ*-OMnF)>L(6d($eeUOCUf(S&%AaP{HGna{@*bwkyZj1CW&
zPiAbLJ;yU2`UunMgx`7oHXncb8h`fR|5k2~N_Kbea<F%YyL)%o*x2OExeHd>=nP(e
z$i=Hyxw~`6aE;GrNE&oqAS{Ea(u)0^x48A@E3BS5%fpX7#_-B{_TG9MB?O=T`CsP2
zC!gT)C!gfzt=rUf!E82WKAR$i;HN&7((4cSj%Tj(eb1aF{_2=X<1srbAU;oU>7$f&
zO4J-MyY(_Vw{J6_PFM*AdN!fb!qC&!bEb!PIXpO|7J_7Ro0Ut~=wEq+sJ}{(^bKwe
zMVgnCd5&j|K;TCtNuQ_{@z6_a#CTRF#A_PEjX#-8I66Ay@bHl7Y)Vnr<W)skmCR<d
z)@ZqP;(Lal-1Cj*ob)_H>w`%jNV|Y^XMHUjDeE!mMM!6IS7u!*t!Z!uYnh$Y0gI14
zRErQ#H)i9n3hL4TEfE44gg{7Hz;s&DV5-Z5enJ#!!oWueP<S9U9wjn>x)JPE4QU1I
zp~M$8o+ue05sA;Fg0!@)J7A(B{q8$d?hkPew&~C);%d7C3Eu-xS}NY<3Z5&Y-p?@n
z$(p0_grck{>x!~2>CvDNM&qlg5S~B_jqim-af}z3%jp^vZJ{T^oKX9>q9`fL+FX@3
zyH*sMDulc7*z&Nsw}uATlM-z?rgekqeXDY$i138Qj|2Lv=kfYOZoTmmw{N^imd?3)
z{W?!RdX;HbvT^Pj+kR+`Vu(^D^U2r%2t5PA2&0&~$Pr#ZBqVRY@){4m=W(t*bdJqw
zN|EREdI{$*T_#O4()kQudX!DU{=qJ3I%8w&48QTpl*b<|IJ+A0L!W$tAN#4_Vq<HI
zvMks;*yC{bE^oZ_GFulfarxQ<Tz%*gXD+>$A}=V?k}NgAS_vkXAO%l;;3E`8&S-R(
zH(q!iKa9C>@iHHH>ixX@g=hJVU;GvR#Q*L;=9QOUWj>#h^b$7K*69s~#Qjx%;q$lA
zGT^&EbeZpX?;4+bt>E*!6V}ddF_+u)FUFi(RlKr4V>bROfy%hMb3`w$QB}Y}UNWD~
zI2aw`^*7i&e~#^Q7g<@~A{ngX2Zrnq0i`mYknjV1VZDDTKs2Tt-qh5}G{aU!#dJ1j
zmd+{ioX`)bnwt51Mv=~`Yb!EZl+;B{Q#Z_KbEb1k<s(dfQb^xop(4C6L<Rv`x=OBw
zM$vS#Ny52R!cdtADQzWN;vs1*m3yX;Xlbi;4nQpy74o)SMt7V64bn8{8Gf~@qAp8=
z*hd(=lOH342;oVpvS2o@Xv&DXuIcyt^pY4q@LHx1EeulL*C3Q6Q;Mc&SPwi_0*RD0
z8=_?3`;0uvL9U<yEd|Xz1W!&I_Xk0+JCCD3F$SwfTeabCAKz)7(z|Vw5r7f~pqXVE
zP3;6jD_dH19u#P8atlEaoWRupt!wSvUkQ|UcU<)5K@fx_No?2<k00aM`MbhySrrHb
zQJ9$P4;qre8o3fY|C_%`H9z9<_dU&xH(%u=PhMnonDEbj{&(59c!6HO5C4DK{yR>V
z<ht)fzY&q9TzB8@>)p&9ZvhNu1|Z=GfDkMJI)D<*Qd~-+C{bEH@yYDt?#jH+K2ds0
zYDH<alDLv6I#HxZf&`67iXZ^Okg$U>z<7IYds8-DM7%#DA}g!9@65e=JNfzCneOVU
ztjx%W6X*QS@BB_)vb(j(-rgRG1sOvUMND+*BxZ*L+U1DX+_A*&g&oenc%JvX{{xK1
zu1V!^G{jiN@sp=9l~aS9OvVfcBiET3OO|K+>L*Hm;Jx?q;63Mfsh{w*=ho<N?a)~~
zN;d2>oD6vJxo3Fc>1R3gmBZY4^EpnPIz@Y_&5>h0;;7AJ>@qb{KyT?NhoXqnXKzN^
zoXQ%8!x73>yyJr(;x~Wc7a8nq&}~QLc}#EV2<>){R;NRnv_VKd@z5o7HsXD6zJmvE
zN)WB1JhwhVXFKdZKVYZ7N^$uDNj_oRUuSo1lfKMpFSN-f6{v`n<2TS+I*wR8go(TC
zjZ1b%>y$c_MY+-_SGOjxL7>e`RFA#qI+{4ejjF1sw65c@G>S1+Q&uHPI6Zs7d#)Av
zWWsnbpv-fZnG?csh2q#tuek2LLMhih#afWY{j(P94MbfA+Mo{DN@E+|#w_OK<$V#J
z0P67Krh0;OS)u9gE=}l+a#^CQ0<9~oupTc<y_Qu}-@n6sGqe^|MMjYsI^C338vDrA
zWkC@FD-A}%xWXAC6>&6)XiG!eS~^igH!^GvOZrBjqgRP>-ZVnG(qKWUK!hwC?Z(}p
z8sD&BU-113NASjitZY?Pn5v>IT=&={j<7~!th@X;PU>vlKEEwwBoQD5dQKqCeLf;h
z_ZzzT!}#+F5eR#G4iobl*>SstNzU@QKYfHt7hdAQ`|e<U&62IZhY!5H&0|ll^Q#~G
z48uX6l_M)0K7NEqSW+o5X-Z)#giwe`kn0?&Cp>WLN$$L15&Pin{OYHk;mYbdx7~g>
z{o#ntQpxVl9+RwOX=#DQL(7Z?1N!Yfws*Jb@9m;>#iLJEJbd$E-uAj1`M`rGSl`^E
z)jCDm?^CTD;dC66MiKdBOn*3JV{4bq-9DX8!jU6Kxb>D>IJDGr%wizzCX6Q;gK>s6
z1qd)jj;>0QIOg^DzaFI|?RHAHFcg`q!`$9pcjY7_*HCixDMhOLyz{LO^57}O`U_Wh
z@$wVs!6nA)&ysCkgtAYp2P9EOWeq#qyYz<zqr9TMIO5{!V{D9FH)`Yi#v9|riCsJI
zP!#WRA{;K$Zg*&PJG9zeTCI+&%S8E*-DQ7WCnOhCC9O6>M&yO-K$GVggF)YQG{!Af
zrw!%RHRF~yP!5!zHY4z$w{@D|KDetlJ}F&<Vr`%!Oo^F6)e<ZsL@@zC_wSlYB)g`y
z&bYbHR8@(tN_3SwWmv0SadVW`N(@p>$)(_wB*v2oRaH<{G1{b5WlF1+5XDn^j<oJ`
zQby#~vXdL6upI5gv}HwV3y!3cL@4A)vpK4+*&bY9?Vjf$1!EN@PFkWemP%{NszUi)
z>c91vmIL5ByACOH?@5&<tyasWuC1lg?%{@%t`rBkkUmxGZO^1GxhXyVpximrsl@D6
z4V8bn(iNlen4&CENMr;=F|CC&{K?b1Jn^+Bc;~y`#^T|l{KucTn_kl6$WIAA`CmIk
z=0R4DJj@GEKgA=T{bLpv+uV5fo!oWL>sVY|BFiTfWkE5SaB{gtuT}EG^B0JO;)c^l
zdG47fIez>!M^2n!FdEbDEV8k_%HCki%JMRYkDj2@S!8kL2$wHkU~g}asx0}{k3Y)F
zVuyQgKE^#~JN)kF9;X$znT*G*Z|yK1XDlu+bK=BF&fa*Ig+q%Ztp$vcZ0t^$XobG9
zL!_YHb#2AWg2dR=%?A-v<~d7;mzj)496o-AJ59td|I&Y?z0hVja)M}aoFJ8=lPX%J
z<_mxDF<R9wZ~o?Y^TD^=2J27q$>01G%IuMd3L=YDnpP)fZ8v9UP_R8H*c=S8TYan%
zOvaOGI23-E<LMc+=S3+*q=;K>TJ0W5l9IMu8zx;<jvpliRv2IF7SZXX#8E_57U)Wo
z7q0i>WHJI{UGq)hs1tz?C8D?n$}LO@nSrNAqeGDh;J2MZp&mw&%tjx91|W<LG^e_F
zzKLsjM9Uc{VSTZ1<p`Tq>7o&(yb6OWUqOX_G+y5=(0zoVs5CNj8E}O_S&-H>?JxpI
zRiSnUDGhAo1|dq0cNHyViLE73ps?utRcvYH+UxiM6#^ASBuPw?B#xn-B)%MAI#*uq
zm@knDM+hs2!x2U4c54+W;@Dw1NCkOSql)392St<NrJ^huPMqd>pi87_O1Imq)6iFA
z&4>0~09ls%RC9z>E|IXXc!<><!6Q#U%U$=}&e<Do<>{~A!p5Z@KmKpsXaCVZ5{#-h
z^PPXCqSI3B>>cN&C!XM|pZYXUeEBQ9{r%s@(W6I*q88cSDj)q%KSw^gOr>LlF&w+?
z0d9T6yLsXH=U7}m&hpYBva&!$F`FA3Y;W%%4<Dh^T_8$Z9654??X69odG2|heELOx
z=MyQnp8jUucHfQY^e9hX$yuoiZc&2q_9ho!e4Z!2{s>t%Cd(%XQ(>%Sp?8Rtm1BJH
zwhz%;=#mXHTAdV)WIQP_g(Zn9go)^OmN<6yEQ3i&@6ch=Sg^W!8B|2l>eZ66ha%{f
zd+c4l$RGXA8da6?_V<2>4}9wf7_C0Z=l|f>v7)5AutbGoZF|J|?J?^+W5$K0Fv4+s
zv<5FSI0G>^7q)w{HPy%>+@@MqNEx|~CZ2<3EX=UJg>`j|<1v$oE6K_7g2`m)A_XZi
zxKKEXQiMvTVP2Tmc$pMCH6S2dquUxhGGLqs=L0k!`pwR71W1BjnW$+5^J@yEH(nN9
zX^gHYi-I!Gu-YJ0<hm|JuA-t|*VfkO#q+eJP=IyauB(h9)8uJEyPJ}>9qwjiFpL(2
zK}HeAD)uHhZ6)dUVj>Iz&9Sbas9vSoy~=2Wo%L3fQp8b=l9EU%V&9-a>rieyn<YEw
zI1s1uFAax7%Brekaw#QI(xNmV^63K-BK$muBl-Vp->SXLi#qOi0%_fvwCOHmn&oln
zWUstnL-|ZzmuAFqf>4%TXOXns;r#hAufO|t?z;CiXdAJ1<r%EFpRFx-{v$^~Rf@RX
zrPW;^?(d<s;q2{q^YlZX<Evl(BDdf38Wxt9$@g~H-CU)t24Ertnk&yg#_jjNl~R{n
zy!axw+;+P&NJ$H1#O~GxTRYoSi%X<Y#Bh?axU|9>-*}9*%a{24BhT|6dMSVPyWhe6
zH?>*oXGB$(%2<w^zKLUJZsyMMfSrviRQU)~mgM<_et(BSe}Jw^MuP#r`5%9ov$x*H
zEqC6@$&<$z_9yJ_^eM7}NJT8K906;|^NL7VcDMQ@Q9>4%99!;jc5#_bwT0QcNMUm(
zn|u7m&;KlgtsUO~XTOgR|HZ$-m5Ud7?2)fvs)Y63J}+*JxxAB8Xg8mf&sG_)|Mg1q
z%wlVvkO-`Hs{w=679)cx!!J^|XeAm`xonu$KBP1nB|Hzx3hzO+5OK?OGm7JSQ?u4l
zWSkuPbmyKe&jJEKF750L7U4f3v+P2|&>c90TY^TS%Hvdyer0RySnI;|Jyg<z`Yi{Q
zL{5i1cyl5241Sfcs=k=Yz&NYO3(dIHw2O+gUC~ZslGtZ4@dCsa3K6lDTe^v5C6ONB
z*BokHI|OuHMk<6zQmUzEVME$B&=90&;J%D&^9n~j$OtFDGM+FVk4e)uQQV@1r1FZL
z^><94<IG__(4N&&z<#p&_TA0$ndbK&#DotF_o^!WWH7Xn9!W2T)=_S{V}+fqEk6Ir
zKV~$}ICADRXKuTVfBkQk`5S*j@}qyt@`Y76eba4p4=r)w(S-BQJw>wI<M0hPv3}_#
zR!-l5REA5JFH#mG-?z_VfXQf=r@r`yyy@G2gu(76TN{UHFBU8>AEtHm7)g|}v9ZR^
z-VVLC3yX^)N0yqCXU}rxmU9e>F_$*SoI2d*TyM;8{K?l@K75Qri-+lS7wB{kv2bXG
zESoUc9biqxDP@TjbbAXJU2@{oF)ls-EH6I(ICtE0H?Mue8#r>}7?_fLoG}`XDYA@K
ztIf%iXL#|Y3nbkxM-O#5vlOu+chTeX40kUR#VLdF7F9OpXa42?MwyTK)<6H3dGGiB
zFr%|iv(#$y#eeY+xU{`XX(PmeA^o)-))OehVidyRRn9y)Z=qhI-ZKk}-bD-vMjNd5
zDwoE2EHX~Kfoh@Rv<A4FQG#g9qvo8t8mI5&O2(!kVF(kan{oI~rIa&Hro`$3U1sR2
zau`;meA7|sL!BuJ)CyT+xOE5$4v@93sRz<hRvJAi(MD5N7E@H1c1n^aNaZ~+Z7@nQ
zstlVGO(zz_QX)u+WHx_SS3Rz;EXD|Q<<h4@3ZlrTNxcUZ$5ZK+q9|z=XI|IgV{&VP
zu-K5{U`(2HNRpQ8Q?IWU)9xpT6lKX|GNCHVdN&Q7hnp%12YGbHT9gtPJvDR?q-qit
zx~|-%eds3ko?B;cFXtQH{>>~OKf><THm|woJQttb<<jOc-uBKm<F8xxh8;R8<+nff
zALw7XNPFcBx7_(!zW(r+m}CXk8a6IHhp^ha`wpuRrsTqt4{`R%FLLkwZ)R)#BE2HR
z6cbhspWx`xqonON8yhL3y*()0B$7ymlQANaEOc6Y@{x7E_nkNL#=B4R>wow#4?pu9
zQVLd%9_8qX(=4qlbL8-07H>F4Q5Z)3F=d`(ktAu0*FNwz2K_D8ubk(pr=DPSbBi~>
z_3b!tHe=z?BHe`+JDZ!FJ$D12|M;(Q=S|Dpcs!+NM~pYmQw+8#vXUw<+1=R&DX=2t
zqyObI-1p88anEb+W%-7CczDzB!E(s|@%Mg|GVi-_uu}}?|DPkI>mAs9kWz*N+^z5m
z9Z^IS$1ZDV4cd6~phCCdxDJix=aM%Cjo)3}UAT^yyT*f#8i#8B^VcsukZy1|R9WQ|
zWe!FnW7oo6B`J}LK}xi>bvP=OtmR|EVT=K3eU-BH-l6l-vPnUu!Bi2IHng;Mswi=U
zuuy3y@;5Fj^2#|l)>z8oRl=m=IIiV&tu>UTt_hq@tuwAQrb4?R7ieG4w*(neH%P9f
z)M#46r(McZNrj&8Cx5`4t+Xp1E{eiENNb53Vcu2M8lc7`w6DtwjN+h7+UYFN?toE>
zwXI!F9AD;!n{Hvav&H!*pKx3<Az8U$i)YRo-utH8SZG;xw#Sepyzg5+!f*cEzhN>Q
za(rb8ne=ES33+bVe({MZHesi2-A$GA=qG-aBgc-ja_khlTNPzCq%1NHA3M$B(h5-=
zv5sOg>O1kZGN&pkz%tq~tX)3OEk_04_?jE|(I0p>zxVJqSJt=a@9lE&!UbM<_8Hnq
z#F5iyIeWuRbUST!^P!veTCle_M8$1R-E=2QE5~{Ku}Api@BKD!`=+-u$Z`gILx%k=
z+DXK@GpBj*+)<Xz2Jy}gSvg^}w@sE0DXSb)lxU-f7Ef{Pwl{L@-na2b&l<k_gyGbZ
zBuW(B4}2S&Yis=MPyR!&<@HXdnF%s8MXXaHa3n+Ry{Wl*U_nWR^|(`<xcO<YzUQLA
zs)#uDO=~?uZuVW|yaBWa&ewxfLiTN-ep$_bz6qRbU0_T>RpwMh2?msiYh$XEW55R-
z%4O$*-np&|`4Q7m6JEbx31@(;a3QlcnsH{Rw4tmj+O32njY-oOtRRt2kN^=;R0fl0
zOg#PPT8``MyJrnjNVG9%qtT{vF|v?#syFg(u@IBG*!ZAh-@rGrF|{j>qcMzzBdW5R
zY0jYH1VqU}7BG}11aY#`ceD+-)>I>E%F;D+#!A+1x+E`h#-lODR8E!DX9QcFHi8tP
zb9UA*Q5cIQW@k8N@yJoqs7(?@Xk&SLz2a@T=GMc8KmFpfOtOl@r*7ke|NZ~Km%sE?
zECuP2Bd8?i#jkyV$zY4=Xw<`x6foRf<xl?0FY_(m{li#enFx)slA<VCIXq(d=t*w8
z`5fzO>r93NhNB^s9+5^)058u+{K1!>=a#caICHqm$wkZKkFS$YGHyJ3hC_!BGs-7y
zUS8+fC!gY$+is!V_Vthw;%>}nFhFcg$kDw18{WvJXI|iQpZ**--*Y?pXhf3lvbp*k
zeKX<Au`bnMi}6myV6qK5r^rk4qM|4*dn~bX{IziU9-@U)tnO5N?4cdL?fzwscNBXz
z<&n?*AbD01%b&-zpMgptZSKl6uJ^FQhc6*rPhGA7&ImbYtw{jzutubaY=RXM75hGN
zLO2OvA?mOwSP_YIo?mcxW-nx)_hx4PI{nB@7$~NH75*w6An&m@Z79nOT^3Y%fz__N
zu=0SuieuOD#-GN)UY@=a!pso*8ppwMHGZBjDx`12?n6UsF_os!#!0<YZjBnFX(fs4
zO_C_KCmLPm6qADf_*J4HOePcB?RFjB8Ec%3ivL#>uER=Mmd*luoM^rjW}k67;4qEm
ziQ}=;Ja^3%9C#TE*EPfJ-?HlO3x%Imqp@epw^(oLR~>X%9G|ADDp%a<JL3xBnoqRb
zZPFw{TaA)%s9kb-+(MT*)@ZWP7+qRMLUR1fIjVSpNufF1RY)ONTv#G2Gfv!m7jNmT
z@a!{R=iHgo3|3#@{MSB@H08XXh&SI@3tN|-<%zF;hS$9LeH4>1Rp9{Lswfx@M;tzJ
ziW8?!(BJLSD{=;dA(P>NG?pwbE>V;fpLukZ@A#&hdC%)lQI_!NGgo-($;TOthTQwQ
zySVFtyXp7)TwJ@vO=oXJ$cU<#aQT_1IdbYo7S5eyw9{u}?Gh)>ouRkX;e{t2=0v+<
zuyqMJ-ei*RF&ykM*xN;z5+!n!5EO+a6J5>^#+aqUv|C5Wazi#5VK6+K4f(aP=7SF$
z<+d9n|KNY@@#h};n{01CNIUuvl9-apv3|*~aRyIuRHwW-#dLhMQwzb_dTvfVo3P0x
zv1@~F3{+Y>>0HrNiw!<-SOCpe8&7j?W1RV|rV&g*ahMj^&t#tKwdPa}bXB5*7Ms?t
zReBsF<Jea<#CRmX(I?W4Pv8Pt%*-YpM$c$n(>a=g0<|(p;Y25^s-mhAY}=qZDScD2
zRhDE~L6KMN4qqibsouf@vG10G0EHlmY=iuT6QDPSJTIpe;f>><Sl!P3KI_>CvPtFv
z>x!}}sB}q!YtE=c<+uY7>d#kIRk!aBgB`MYp>0xdjpx0n@V7uHb$~lQ%^cC`^k}!c
zNGn`<gHbGWQ(CPvRFfgS<(SUGBCU28V=b*#%5c!<6AxE>_^mhcuD8CPr?0q7m9d7i
z$B*;Sr+*7f#xoE7Db-}>z(jx}5qsXe&iU$}{#VZ3c|V=b62@rqk-(G{R0TzuvvlM*
zD@Ts894*o5^|*Y|Fc}SLx0mQGB%I$<Joe&{*WI+h=_SkKam3lPXW3j|<&iIZi7)^0
zm$~)!+c|l|87^L4<ugyhx8A?P+aI`<kN)9Th*QN~uep!A?!Al47q8G(mK)j`lS@w{
ziXm0j#}pIvXh1gD!^#S&Vr;U)xVOxnj@Xc!oH=$QHi{VZM_jt}61~M9-S!g2q~iPj
z*&+VzzfO7co8g`B6#Uw+y$Pw`iBx}tR;4op*S;y&b8tq@qTxL>OHi6%baP$^;R8bn
z(({AxA)tfOMWZE`Xzau8p&ppbwzEWnreWY^Hq&Wkg;P<`Rf#SN&*^eCc~O*lJk6Ep
zAVlp&?Yj<dpsj8E`2go4V0ld(%$?VwD?@I|`ea&b`WTEAlqHPEB}H-VLHqSl5Gsd3
z38C=eE5T(|N}{BrBSevFu{=$=zWjOX$#9jw=S5z(snS||nPQ7nvYwz8FGYdd(I#2u
z8WmQR11)`+*TkU?Dt2y?;c7uWFOw`pqbY;uutXw81FfYSIn%j`C{_&nJ8W;Rv%ax`
z)+MPloW1E*7M7POH5^~=vN~xoserQLcYgEddEyJdhc%h&AR#Q;y4KXqBS-{k3<^sz
z8u5jXf0Xxs_%AZl71?B$GRr871epyP5BC`DZgceb362~)LMy$I7hZUQ&Fw9gSC+A|
z!`H7A+<Hp$hPzKQ(UL!W*l_BGn>cp-B-?vyY;9fOrO$qyD$8O0PQLwtKTE3<aofGO
zv%B8sD_{ORCl=criDOn@c!=W4*U<UIUAxXHCL^kROjYGfj6ll-**k^q9zj%^-f)Z_
zu2Jnsitb^?w!oMi5X7AYo_JxK@A|Hj+;b0n^rNu2T<{%#?oNL5mp_G?w9sa9jR1O+
z17{C3RF)~^5ayI<U_@bFNZ~+lSIg-(X7G|L9>bABde~x`GZeOYW@a*#&62CBP4lk_
zQSVs+w4TWn`g0mnIfYH_+Mx?vR#L`Jje#h3U7W3TC1p4}LNG`IC#Gk-ZyCnMa~9{<
zOq24UL)ILD<E=qfRA^%ujkDS?D&g9!HF<}xbX;E}hCBx$k-~UmB?&r9fEQ<TT0L<b
zyI%43>f78*A#l@5Pu~F6bx9tL$CQ=RJlDQ1%Uy$y_x%Dkc3MtTmZdK<Ff(+6dP3AY
z=>9`%Jq6tTXOhGbKjYXLIx8*8cNs2Z8_EgeQqb@3P!&Uz&B!KWCZ%ER3t!~vr=I54
zAAA?L9$Vt!^JC(uO*YC{Xh*IgC;~x*u$4=u96V?^lt6)1TzK{o9)0Kw-1DZlF&d4q
zCMU}ZVqKu~44q9F?(VR+dxGO<&T#L&w{q#?6*jJHu(!L*?oOL-BKXz^mbvfNqwJVP
zHb+qAIj73od>MsPZ9IF#;Zpsb9UlM6SCEzA#GwxQg|G6`##2nPT^Ntab&f!@)Jw>-
z9Yh?nQ&w#4WsJu|sy?(9H^_#&?5$m3(`r;}k%i?W96fbA%cpMQ!ujVpeE0}=-TgeD
z|HQla(wC0Y?d5#ad!FM*|M3x)j=hDS{imNGvnjf|ws6l{JcQ5;ahiWOH5N17U+3ni
zUx?DB>j0O9_5vsliL&)4!%1c&g6HW8p&(d<db|31rqRL7#@cuta3x~8q%3l(GDrJf
zc`8yU<w8>xC$7<jn0iT5vg!0(0+-D+wHGq=l;+_YV_Xs|2nZSxO2mBFh(LI$@gO~;
z1qh)q`c*57u~b!6<26!Aq=@`Q;yP1AE^b#*<Wv!=a4`ZQ!oz;~@2QPOm>dY#qPlcN
zkr)G7BO-~A7GuH=<AC;wSL+O*UXmpAe0SWY$29+)&7$N`+#Eo7r4-$6k2GzCcOxxm
z-;S!-`5JlqPHg)SE5}Y!9y)@uFdpnN8SYS)Ia!hO&{G$<@x&t4&Ur46TXZ`q_kQCy
z^XcFDC05TrhP5$9fvsjEzWQhD>gSCxNLitY`0A&Ai=)R*vUt-i#FLWThTYweXCsWQ
z*u0!E?(ed4;uNQEILQsSoMY7A<I?#x9(w#mZaWHh-LTA?Z&m!^bB4}B2Z2FX0^^DO
zD~oRG-Ehn8yscc~%A=oT_0j)|8D1e9?lBrqkYwa#MK;b@OwMB3C)q3{FKk?7n3ZS?
zV(kTXE<fY266Km(n7q%<>VTb%mpGffg&S|X3%&CwH{SeN?tSogdGU$&bN6evdB^+L
z$g;zH|6hAMpa0`$c;vI^F+w7g#cKat1cZ!eJ(av?<NT%2rrdZ#qJau4BJ4~{pune^
z%ghP?M1m&~fzTjLFf#scQq~C|gq?vE++x==9!Xf>HHIao7NvoXc^Q|+ys)@+rc11@
z98*9gh$unEDRG=4a4HPeI{9F%b$V1DE^%B_xB0nrmz;i|ofa<7dr($*Al(Lo&eVV!
zjS)d;FL;vStgn&^f)I{Jp?#`RXt1^<^1b+kNBo0Cs+1C4cPk4+8XTnQ&TzHTn5M-J
zT|&yzH4C-IqJ%^uF<uoxSZcBUa44l%SXiiATO9<JH#s3d?TIxXM%x4KGi5V?GgF4h
zWaN`83`V;gJ9!5yM{c0Pvaq~DksT+?MikkEoh%_zk~?oa#-&GAsbWx_9yi>2H)|Ik
zcLmG7sMQGN4Yl<doF@J!Tv}Nf%5u!c>P3#;b|*wGsbI9z4U#}rW*Adalo^BJ4m;ax
zoH%_my@dr%pFT?*TOQt$+;qxvq$9YwHR4a69<sQ&z~Z4caW^6_LHp-7xcmaoe(JZW
zu6&JZ=S8YXPL&Vo4|gbv1QaAkZegu+noF10xp?scld%C2BYapOlNf_>8DQZSg)kAA
zA(vl#luk=?{OlE?PD&gXyy30C%bBxXs$z+9G9+GT^MgO~F8=N#KSMQii%7_jopQbS
zgz&8gtizjvf~8OnKatYuQ~CloRP9FvFKC+EYkJ@F(10=)D^2}OEMTn03j~B&26ld7
zTEB5m4S+Fv%AfL~W7t5a2F$moGS+!fm209<`F3tnMzsn<6erGe3G1m&2Do@{K5PtL
z9f(?9JNQ}q@Z^T1%t0Y{sHLs{CYss)SBWR(++~G~BFf5T+AM~$DzOHt(v^>d@)f0=
zAiD`xHpJ#L=iG0U!i_cE%ZVpaI<A!CW=J83JiA@M43{aKqtofo>2&7qV|Jo757+Hu
zTxSHObUN7GZg;A*&@`yN9Z)Qx!!8#uuW{wV1y0{`CpX>pS~{I&<kBJ6p4%8CmY23Q
zueo`JuRcFuR4Jl%izI0|#X}iU+Q@13=#WwLA+??AJqr)cx+^J}^tb7C7ug=HBczik
zks?TWL>LXRlFqx^*<q85WJ`yL(iQ@R6oxN$G;g}K%WH4!^3c;;tghv(uDK2&ON%K>
zOA0v3@cbj}e(evC*_g?AhqAC_Re`o(XmR|mw_w{VJn`&{Tzu{l`D6nrOP2{%j*@|v
zUQPt?-VKaVR#6Ogc>1fKroX<(v7<L5>?ZA0GT2L5Tr5e_2~t`fc=H+F`e*Lp^Z(^3
z(sr9xw@YWCN4L{vvD+a{9CxWKD+ZGx`J}*<8dJK~WtjyNDG+MLyLW4Jf2Ww{d}+=b
zDJ_w|kH9;c;$Gndz#I=N)CdN=ObFri${cK=eaJQAz2GVx9AxCAXR6AnYdX<AiBK_d
z?34|JR9NXZwWg-}tlv=IYa^O;xUz?Hnldoq{vBgoa$pre9aGLh(*dJ<rEscoT%UZf
zb{$PZp>Ji3Z{wrgH1q12PEX1>X%C^mEAmaR>)I}`IT5Cf4z#}x<Ep`xgetVQni>~t
z?LOo3O5YX4^j%$^3k{pP>GK#`2y?AfNF#_;#_{Ep%bP=9dg>82*VZ_B<_4@%v^#Cm
zR-4|!0;4M8hBRX3#1R-&RJvm2#5p<(%cNM0vPhXA(Dkn7k|*AEmX0_Z$0<mK5EZQ?
zp{xx3(U`+a9dLe%%fJ|miej{`T)Qk8GwE;9?!-h&fRT){j6e9yOPuQ7!EI-k`QCTm
z#HXGcaQTHzE<f=w8~qo_kGJ6UKTN)}$!L9(QjZvyhD2G)s$k{jZ{*yA--0@Nf`>o*
zNp3v3!Ux~<5!$N6s)RUF{`wVKIFG@4T%#yThFL*>XOH#mU7mjWaW+;r`O0Tkx%sv(
z*=U=UV=LrkL8oe=3ro_2|L6bm18jfajVyF~EcO;TwAiE5=^$hR-b<1~p{4V{%d%o~
zZ_MWEDr@JT=Y>m``Qn$q#Kzi&!;8f%cpki%eRP4w1}XueyrHQqm2bEi{y7iGc`!{-
z+|ehz$02+c&`Y_fV7vpUU!M>sL1SGeuPO`5B6C2wXaMCZL8ur<SrOD02%Jm|K5U(f
zV1#fG%JlA?B0^Xn&GjO@Cn|*S&2;ha_hU5Yt<~XU2n(<AxIRqE8Y)wP4Gjt;QJfM*
z&eAF=DJ#eD4hh-m!*#oh@NQ}tm;)!k%rEY@veXO*4#?J3Nm&-PGN{E;6izEAa5sby
zb??0<)nI0q3}B$&y~@%(bR|Lf#6T#Y35G8mQ~b(d2imxpd!e0iY;lpaXwx6=aPjFa
z;-t&&-X0=K==K)bx_leAe*0^=<z&L;wGDQ+x4Hiv@8=!gbRRbzUu0#aMJw$Pw_45+
zSy^kayZd#M(OSXw-iXWRpX1r**SL~LsH6o#f+;b^i3Dn?UAyg4Bb6dE28rQ?XLm`{
z9<8*^!pad2pE}KFpP8`OYjgBaO1Hns_SgS_@r6g}RVB#N0E~uxwszLgN)iDRVL9>I
zcd~NtyLt5K=h*l!zryoRe}(tF|HHiZ&wjgOY@5cGigm@h))L1q8$1||+1l7*Yh#Tk
zpL&|7pFdC38L_b@xpaOHYhI+&TIRNooS>{K<fNc7n%<#=L~XPC$WvSyTb|E%QI$d(
zO{ARMZ5#<=5h2@c=q|8V7OW3PTpmxjvcAsd#%4_!cqI<szQ2n)kaQlEmpjqcp*q&$
z?%mJ-={dvCzVo+V<PuRUA%uBb6D@1E$6$1U#yJTtIb{$s1{D%oUc<_PLDV6lou1bj
zDsP;mf6afl_3J7anR%FG2KERb`<xSEaR#_KjMf{z>#4!KE+e(GyG=A2I&m!NYM20&
z7uCW_s<I@GTvyvlSCoZzT~zAW`{4NYsz;bS21cWa^K^W(P@^>}i7{2_1@XMFnQ&Br
zR;x9GH_e20b)4+pUzR0#k)d@}Yhfq8#d^~~&AR(x%W#vy7(qVX#S|HFbc{4jnT+;`
z;x3Damgz1obL8kTiopnsVaXN@b_TRs3tYLf!LBZ;DA{{<ht16$hQkq)x<!(r+w0L=
zSfJPIIZ8n*VLaGjxU<Uk)(-KZQye~Wn5r0)Duu|#l$A>ZYpqdnN|}vaV}RZQhua;Z
zxJ?u#=qjh^@9^kTmx;C&-}@ck%sU@A%j5s`BIhqX&$!ZbIzGEsRICmrBqAnW>~ZYm
z5q1U}y!e@4V&n4joH+L&0Q^7$zmuH>?A~QIE^Z(?OT;306w|Db5Rz6a!I+$Wzt7Ij
zE*o3BJpbGaj0QvExD5)InVKa0)_;1GjkO6M__n*b>)swBktFSc#k(eSw~t|aE2xEz
zYknXkax}s8cd^4ESyf`<gpswF$^rOgS+TadPE~4RDO}=be$%w}Kx0+E&l@Qhx*CLZ
z-j6g+mr6JAjeWRH)9y5WwbQBTXV^3(tfQ5pz?4pC(BVgsYy6;G>voLxQZmA~SeH)F
z4wr?jzog+3)KNk(uFegtJ-LIa9zos(4$7EPtCK(|b$eCAy}RzXlO#<@k`8IwB5-x3
z2#%p3&ocUZeKxl@SzX_t$~MWf+;1Arl+Rk>{TJI<{?pflU^pC-d6h>O@|7s7T<dqE
zQIT@Bd2vLi(<X}4%M9-HFh)}p8C6-<-ilHYtyYUTj<0z@^E@XXX>_&ja2~r12(r<L
z&f*f*SYCYoC61ju&C?s2J5Da~4R5-SwXtR8*fE~`(x>Q;3)Z&BAln=|a)^_sousQ&
zab4WmYPCq)3E5;onH7{p#m2@OdS}3yTke54L19rwqm4zI5{xD<3$%(URfJFqQ<e03
zhOC&7PsaS=@BcE@`r{04J<Z)G6CQZW+xh1Ae~>SJ;^U0>wpm<01VB+~h$BwielrWp
zM|keJ7uh&JWTa!tB4QYAk?j<mI(d|vZ+;DV8lh}OUC(I&;aYMFA=ug8!0h#@D#Oms
zF6S>^a;>0*aA<+3sPc*=ImzQ+nXvl8*Lcsjp5}w!bq`D3B}_MEa&Cd)R*9NOQmruU
zHc^@o9qwS4N}_y1sWn-@Pa$(ANOmU~>)SitBl7fw<^w%wZXU4FIfu%HKO)2s7PN5f
z-?Rs#=Vzg4{6An*)8E52)B%#irnJWTCJMT&z`CXnDo%+KpUIO>0bw>Pr(IF6r>Dzc
z5`$Oe|LO32!qxbEK;Q|^C%8fh;enp$+<4V~K?p&w@wh%rsw&8h(?~9gNgXE(Y;AnX
zqGU82l4ldDGWWvwu6aSz_#71G&D`7puS_P67Eo1{=k8Qc6oA9Y0$!6O2~ni>ZGiLc
zYS@{wEF<%D1NXOxPOH_>nX>ybfb-`Im|0m?C|jY6JzA|LMo1PGdY}}0{Q>9B-Nw<=
z=hzubB5CPLLuxEKUF0>d`$oEn;?~=okeZN&G>)Axo0X)IB$5iL1g01=+}mJF!Ek4l
z%a<;)y*FfOafyYM;~qqXBvDjZS+_76j)zRfV@6LrLVtUe^$RbOrZLBl9p>Ch#9P1P
z?U;Pbqo4gfZoBJlUh{@;<gI_^ef-`p|6ILj<)KB6-ursiMkS9tvB}EHE!=YJDMner
zB+uB}-sbFGZ=ikZPOOSBlN{-_h8z*nAW+0Ip~!Qptfa_u2E!pQp1;5(FNxDOO85>n
zW=JJ-h`MyThbVKyM}Ofmk3X`@5C0Dj@S4{i!``z(^~enrLZjL}+6!G`Yd}|MT@j5Z
z^!vM*@faN^6e41MYmd=rLL4isbp_Y~JQYH~+<ckINH^wztqsZay;4c9py`q&1iWi*
z?lc#E4MWh_sKfhkx=^Mno$N|kV07gSql&<*fm-2<o+E`7%7<^xyHf_V4RyD=ZfPu~
zs5j@3>8wqrx0v(mHH<-kPiT(KTrzCdNtpVE*Lqx^IFB)&SuPOP5J^ZPMJG+@c3N~h
zZI%`nSUI%F(&8cuy#?YVaFY%;O0VSQgu`0bpfPlat+W#&v(}KdQ(~_Unk0!Ydzn4&
zfs=xe%^Oj8?P_BzX{+U%Myfi)H+R)D*JZqJdQlb_Q6Y3fBB7lck|@R4iYQ4qdFnJO
zP8g3zJa)cMe`2`zghUI4jVwZ04j*Z8@??i&M^et+u*jLyi!3e0tgIv~AC77F1W9VK
z<XAnTzqiTS+7))Ux2du*pa0aydGU#d$p$-&20IJ~dye!f6{RjIik$7O4PJWYQJ#6|
z)ATPs#tpp@A9>f^eEWO9fu)u}RwIV1PxFP}|BnEC_h0?%WJ!xGa#>Q{UE=A@9Nj+4
zxjXMCU0PvpFr*zx?z;63Zoc&vgsgb{vB!AvrN>CyU63)Qu~b$#K9W=z0mI3NQ8r;P
z8nV5!!^KONNLy`Moi1@Jq1|qC;`nKdaeV-zmLlzR(d0b)_%1*B_aEY+FRYWaF4A4j
zI2Fen?X*}Z3R-&ulI>kudn07!LL_ZmA?$FHarx3Ju}YjlkX~l`DjR2w5g}miEe$8A
z0U8H7qhnc1G4(|D@va&fyg)w)DxUSpjq{?kFAdYS0$-D-ywIMCVpoacF)V3afwQuT
zhzOML1DK}7tu|?^MH<J%DsqOvxZ-5%8<hrtW`5o5;}~A$zM8MQbtD3ecTfa<)#JK6
zsY5GAh~n5UW|uiCii$~IQdI_HoX+`ZIApKiXD}F2X(zO0D%@b(S?#84$EIve4=9Qp
zYaK8g$qtcfp{xhSWrR}EfyPg`FLb*{@S;XzbWzoV8OKSjGZn5+cyM4Q=Ml0tNG6jp
zlSww^t*C@<D@Mi@Yh}r>s5rXR#b`}7o)8H@#cU4^#}7$TWhmnaE7}apoUTr&v>+Rn
zU`vc~@-UUP$VmDk(u%>bU^wb?{-qZY2+}yl2+Q+NJcilZ;D%ezas1SA{`7NSU{sbY
z96iQM&ppM(Cm*63^*OP;z`Ne|Uf%kq`&nqWC<eRe!5S@*Gg5*_zwmMH`i6IN>fD|D
z_5b((#vgt8VE|VmGa4{S_lT32g-(a1g&sy%7?X4Q^a+fyJpAYvQUCE5`13!w!0YdO
z4aSyEJ-~D6v;|YO>GvgBp7Hz(FOlaJO2u_E$s5m};?$89o_XRi+Hs7;P?ZH3fi{Zu
z&61z|@h|W<|Hn7*+Pim<qZ^Sjc7$Ks@syYdgdyHuLaeSc&YtJ|#uc(`fQmvJDBrmy
z=#N|Xlfk$Yz6XYbhvOht&!#q-%DTpv+In&^?&rWWa#P68&Bken6wn|cB$CL3@5TY!
zMq@)G&ti<$C6WlGT*0u45@eM4Ea4PWa%Y^<CS}MvhDc9JpCEE_QCKgZTvI5f;i4;x
znI>**z~J!XVCq=Tucd~;4%tbcIFsIhUh8pPM(X^9OAU-BfN9bL%MBM;DpV8`MF~-q
zp`yqcwgw0}&x3lUBfM`o98weo#v8jhiinan!e<sO2NTP}Ihy~Uj3*R%?lP0HYsuB?
z_3G9|vmb6?2mxD@f;`J>UKM_Fv^p(1?G&Lb-MHe#i>oXwE)lI9LL?QVVuZ+3zJ9UA
z&Bqoww<LLT(~u`Ia%-JY@esDK46Wn7IFHd1t1jBYa5APp7_qsz#Y-2iFxcBAO`|${
z*|wG=-4@3e7uZ=_=W7rBF~i+0go?oy9O}TE-hDrBdgB{8ab%e)8#CHnXEfZ#Rt2`o
z$aXh*{`@Zg;$Qq9{LlaE|Hg+t^dX+vTw^#IvDEE?)}%@iNkJ=(F{UI<B370<<k<*o
zEH~Zp8rHXV`H6q>kNBoHKY+A3GK#U<_+s&j;V5UX-)FDC$DluCrESqVM@WG-7Ez4Y
z+sa8Iu)0JS5oKO7?C%go5phx?RF|LpClB!hKUA{4`USrH*~`SS%PL8!P%@@8n$c*?
zXp%9SjM>=SU^1Q{aDcLuAfwo`z5yIV+!-Oy7B?2~9b{&be5y_aNCZeQmE-Xl5v)RZ
zubWJ63QfJGIuRqpRP6<PCQoBZS2SGd(ru7ZIS;DknkYy)rP)j?{iyLati_tTch=OS
z!8VMF<C|M6e1zb}!Pd1temuOjp4Jdrgc1aB--d#~14OUF(!6%q{JI(EuDCc{fDqG*
z5Q6%lI{Lrodw5JLkPl(%JzcAT3ZJpo*2)IP7$=coEkc4+kt>d@caSC^IS+VRio9Sn
z9%6Lm@GD<{-0SuB;SaN8FumWZs>m|e?gt!OI7w4lZ6_hCgyz)Im^fNwJlY{w9g?(z
zirVN>b7gzP=AgrgMak1UU|Vhe=;L4Fv5)^>$f!f0EDJxD!Z*T@N)Sth0$Qo$=4Fde
zt}$zAK^x8H-V3aM;U%7U<^`%`1-lR<?1Y<-o#fBF@ip9k-`z+m^mq?jjEF@=R}_qg
z+dO@KmB(J%<fVyXc>d$u_TX>wm;Sr&=eytkEo|-V@ckeDF#qR2|34Y+ZBbMe?L?C4
zlJ)VBC%@K5rY)q_Y;SEcsw9Sp=!;(_F|HCoT2W8xkU5n?ve1sbM-w?tmynpO&&JLe
zYbvtI7_6bpCkzK$s3@k>k*GMK)$Z`~|L!qvy{*Sn&t4*qk{TsaQi0XvMZtJHCd)F0
zBbWJf4MbWbNsM92gASaakg^XvbO(@6X)X@e3{)7$^bc5)ThNU#X$B_PKfhg|(PUy9
z=bcuCBW&sd3n=B3HC37-<JdKHm@$NI6P;gKHBpoh9n6*vO&tffdZh50Q%d+x=~-MC
zXXGG+w=S|!wJh=M`wvJcUej@%jB^c4<xywXuRsc?OdyERr6P}f`z!xYZAgjE9oJ&4
zgx{(1ykInPr7!JPk4csx;}~smJ8#&P4oDq`0n^6eP`WBz4WbmaT4!j}ey}%YeCXvG
zO}OqM6K_B@QVKflHc=WOtio7JnrhA-Utn`{!urm0$T(u*@Ci&3voVP1Pb_CwKyO{4
z9Vgs&@16Y0Pp^^Xn_y-+CdRvS)_1{G7L1DOt|fS`jSwzlnMQ3gE$AIhxpBF}J6?Am
z_r2~m78fFPJ|WLXm|{p-3=vrJQO2WByu_2MBetTWbeB8yY{9Sp=U?Fd*WAkk_uUUT
zT`>vtw^ms~@XX^|BuPxS+oRR$AOw^);>2yQMJU16>KZ$_;nEXNkc|d@A^NaNgit_u
z5mMtf7`LfeZIF?q-C1x3rPXFM+$C+h&llC0rKn4ul}sjMe)pr#v9W#`B^3BZ2|{`;
zI$=OKqoq_t9K{F>)>P<9f-PP4%cc4}1{Jc`K9sYThzO}9NfHq!QLU_7X+u_&<V8t^
zYJlEhIh&dFIX}aMb)`AJKeh|=v{N?Fy6|&Mx`JWfhRPY>gR^QbT%G$q(U382f@%Jc
zwJB{Hm%s&!YoJ^R>a2dlZ=3tJIlZX{oM)m}t#&W`3%i3KeD4yAP_SrR&6x6_vy{QD
zW#)Dc8rXxFz0JS5rkulCp`faK2?(-oumFvL6Vv^x&gM;AiH*xPxnk&rg#}+cJbN8e
zW98j;5qMRF>#nSWoQj~`?$B<z4!yzUwbp449X-^>%02RG1Dzd4rd^cM{Lv@B%wPDf
zZ{YQ3)_8X9_gOmE;f?RQna}>lCN`FhN(Sxg6s^NZl=7+x)(PMHH%cKeR?~^$j&mnj
zI((Efr&f5=>u+aeDMgnV*>Hz^(kB}aDf5gfFIZdO=GoN&Ph1&O#61udi7gOHad~gV
zfB2<e;;!58q|@nOZOYc#FESkeWs+o<-qJEjtBtjiLQ7QA=1@0=7<RLQrDMnGMuMI5
z&rnW=jtOn2^3p+>0O4e4q*CGGVU;5VS3X3PQev!TGRdg4A&!LORIQ|xMFkW{*#bG0
z0~QjLkr*@ri$W2}2%#MC9Hj|L#?x$~&sKv#DM1oNv|DXDy)Nxen{Ka5uiK&1Nl}qR
zT0x$<UVJ-y12(sISlb$qm#$33x{17hJ~EwSp~Oje?8^zqxmM`YOr9c+Q{uEkl%!J}
z3RBO^0LAz(ovEA+CBtDMyNsR+v)hM%ejN#q(^*61y*t;;)Ei$pHB#YOg~}2_NJzkP
zrzJu*ETvbAC#AKaET<9+H6u0HUJ0SnI@I(y=@yAnlKI{VP0qzDVN{L=pwQYf9FA+f
za|c#Kq$EZP-yO#8Gfv^i^Ni7GRKG4GMW@?!B6iKVexD~5crcD`&@@oS7}B&wx7(W*
zxeE6;X?^gnE85aB8mv;LM<_+b-qT;<*xk!i`gJTV3MMa<EG+hU|99QW7eDb;vLVQ*
z!srMqD*w2w3u|e__kju`CE8eGtGWNS8~N_<co(<db}MC-k&X8l?ygdm6N+r&LWpcg
ze{aHzmv?#c`E@P~D)vN7p?9!IViln}2{Kar;xBH~vWB1h2{&1Ly<6Dr{}qJ%-zh7D
zu!cBkA!Ljxhm2&K#deF8UPM+_jK?KbwJ}x`30dpY3q-94?y!~yv)`IJn<k7w7dg_0
z5Tl_*$pj@T#}_LP5hrbwih%-bL|q$a1qlDMbe&R^at#Th$SDLM1uB-LX^VEdMX%Rm
zVWG!DJ0(sN+O3q>(<mbEg~hSs()GG+T4_p_=M4INcJ}so@xm%kJpDXdyTcGD`Z+cA
z6l)wR7^WJyHW;6Y)kTgqj%pCa331YPl?Z_zBk%^_%}rNMC%i}P`B~N(A0a}0o*)ze
zIB6EA5FrJ%q>8Y{XZtEbVY?H#vnZiln=2D?m;ry9Thrb~7;Ok;WWui(dmRmgue*cs
zG*-9MkjlG%Q!6yZfgCZ?O$t%XI)>L8=FRgiCX=!EqCBb7XjiW!Tn4Q9-C<{uQj+I6
zlgU)XE^w-Pz23pa!|XR!0T**>o$h{}j7!&N1gF_cYh9-qjK-)Kp(142XVhOMdFcrr
zxNQl&u*&a${0iUtzWewOKXs1dw|;@|`^#VEeII@uAN{$nAZ&?JF-0X&%5MrY-kot1
zP%2ENQC4&B9XD{_-8Zv+<#}E-eY%~9Y%<1Lmww+J4B6S-V{>DZ7q1L?W?ge}dx%J5
zk|?GPl(vMtLXR`F6cpJ}e*M=v{KdZjciss<_G5w{{;Ri;<r9b`iES|%kLY!Kbh|w#
zE?8BlD8ft%N-H^h_6GV_UIJVB`Z1r4su$~YjO$5mry+b%6m?ytYuRL6MMLK9Z!uY_
zb<P59q@D+2iitR>UxU!v>EA~Yoh0G*Tkqh|p(Pd;dUSgoTCD_Kl?WhGUf*0PqBwR9
zapQ<IiAm#xR;z_lhp3Jorr+;#WO<q6N00LOGcWSOOP9!s(uZ-Pc3Wh~7<&3c5RP&B
zP#ScFwG~!cq9`S8bxB(thxs)tDR5p9)^Qetaqupgf5(*fZ7p@+DMh#{TYFl%(&0h4
zuFe>5IL%aN$~eubW=)~jO!D{Vv0~ZiCKbh`X-cQvrj;bT_|(HMyY|CzT}G-X%c&+#
z!*^?D6T=QsRi*|Vm6VP(EnWKjATau6iJgr_4^EE7kvg3Uq1M%OQ0C+?WDq8^hAeX-
zVtBwIo9EOzn_=w!?+-B7WHO>CGHhr!g{T!j-TRzwPvC}x<`q`BRBsXq%8jSE=Z&u;
zKJ+Y~`_r4rS8wLM?-bm4dWB#9xwrF%w?56gzw<7he)1v@|H%bn<=T^LZDx$9wP5`B
z2vgBXEw8=f7T)!i`?>h+W5kC}Vyh9Op>?&7dpYMXZt}!aFLGgf!cZ+R5-F7}vF$cO
z#f(Q|QYu6offY>hoH$MCEd3VeFZ^YmdFB@Gx(hz?5zEi~{6FKlmn=~vDT>0C*OUci
z>7H|u7p@RG>Cj74O#3J&Z@!C*&pqZ#se)bFuZ|%ao^U<?ioz+aHj*JhS_SRW^fu_D
zh^bVFZH|`@JqSr3>%y9?t&Fcd_88~R-N;Qhog+<S;$DZP#RY@~V+xd%v|1@ilF)9a
zw9=S3R>V<CAVV6@CA7ve$qLS$K0#4boWHz*w6pb}GGq@$*c4rdU!z?kPooR47A0aL
znWALkJebO0gsCGK8$6ldQ48xu_AJI$7~?W_+89s0aUrd-PM0mv4kVV!Ott(SR|~rB
z77M)|z3w6liwksnUE(yZ>sr$^C5;o5Qne9E(-y6??TOWqm&KW`vGI7!$N%7WuDa%%
z$MuDKMk4|rqBd5u?-3qfFu{U{*f#)?0e_fz|19owkkM*R%pl29R;6#wpuJIW^>eeI
za}QuV9y=8Uuge@o5#4TgMk8uI{A|7kP?p(Dm}jk}-EKF$DCc6sG;B0h)XZNcz)A(N
zA&VZv%5DDi<KM?KUl)AO_rQY>!oU5u9lrF(_wna{beI3-C%&66f8sg*%|HEP#-lM<
zpJi%XlQIHZ<(xdW#Jk?~dfxc@JK5V><IqxvD2d3*l8yB}Ub?u(bLTg?xG^CYZEW`h
zrE0lkMmZo!6s6JF2r6HUtgvWX5yfq^c{j$Ks=v40h9CR8KfvGkJO2~Cg&w7WvM3qm
z6Qr^%EG#0VAZaD8@7vBEs@37>o%hn;-DI$J#gTI?bydC_z{azldnQ5sB8+VmDc=ml
z)rYoySra07CU*Kfy%h`)aHJHmQeHmSqNE_pCp`DuvrNV#?!NnOT4{?oidb4)pxZe_
z966cnG)+*_sZVKLu{-WFnPd!x14hFUd7d*EjM?4UWpChkaji6BIG(sr$Ak@6``>*r
zzIGfiQx#yf!+5NVUP@!oWrmc_3pB>~ZY^$Y)%PI;aimaEIZ?{U$$AT==tL2zCp)V&
zrj@3&(++8?O}mxSielncLY&0JaY7s?q`rPNkeiiKGt%J3dKnjAZd1of)+y?i)x>Su
zct9IRTMGRvujRPT?4F%IaBvwAnEl-Y>MpfxTpzwTBJxtE0U&j5;(p<)gN&BHXaBpQ
zC`u-y99?OuN~3-6FDV7W_vLf2m~*WiQ++^LI+X-h*xIJoTR<r{1kx8t2R*FW<e;03
zqtTE&&j^+|Nn(1vZe1hTxQlwcLc1>szE(|Pv~Wb%sN{`r{Uncl<u2a)UiX=Wr7>@P
z&(ka)>T%>umyH)EoH%_eWi?>9J0_n@m=t@UOQJF&mKFEicLyJO?>BMo#0veLRT5Qk
z=Hv-RqmqXoewJsRd5N7IhO$G|J5Cw3&_-j+G2`KYIEuhXq^&>+v{giray%I0^ukOP
zfsN~DF&qls`<{36zCZjpAN$xJaOS347$pTd*OX<&Xgp?;<qSq+R#ui-TI^v&!XVE$
zdgCqhw^kjfD-=PK3+6W%hvfuTzt-9-wB}P_SSr$_g|UXgpwDD7_6sP;n>6;(@VoTq
zGX@o@`fruTfYw&8P?kCORQC|wc$Ri6p_M3N6;osxo9pZJ`vZ!?i_FE!bq{K_Sy?%X
zkcw=Q(d%u|+TLX_eu2TxD&z5pnodFu7WoDr3K2<UnvldjI-NGXUYBmSOQ+qT)9ul2
zJE}yI#Kdu2OJb*K3#Ae-;iGEL#d(v?I5?EXOhdui_=kTkG&DF*OINP>sY@hyj-zp6
zew9ylgigl-#Y{LABCPrusv4LN-_?|Td8OmJl{u%G2nW6i@H<o!Z?X<B_dV*ONIAXb
zD01Chf`JO2lRLqIAz$q^cmpMb6NMiQ$5drWS(U!6kS2)|j0U80ZG{k2+L8FmvTEQ^
zDGLjW#BmZ%U0-|nJ}H|9(uImyHo+K8sO58=XFQnQke-;{hv%dREJRz-nnAzM`4{?}
zjSY9*_ZaW{$jABQ@4StLr7_?3eUI{{x4V92yK5C6{gua2vB0v(p<_!#O0c`L%f{Lz
z4jmS}@2zj(O>exHc2sfY!ZVBq19Vk!@yZ5|JpKY#HpY}u57S;oCv7HqPF`dPT>yqS
zR+zG4JQ@<UqIv=;0hMw0poApt{5s|Mhk@IeOawpv<C-7;amC;HoBt1f^EZE=t&I)3
zhYyjoyX55rDI*RaUO*e`o=+sHOwLlfP4C1RMmOHd+WF^^!g!LeLfHM`Y%n&~I_>)H
ztu3_HL{aR@ja1|l9)`Tg$g&Bd9{DggsXb`0br<VOJH7nrcWSVfvMLc0o_gvDc6T;8
zdi1bUDeZMh(}Z@r!@|-bU-%t4<pUX#_W~&t$zsCd(h@h`1h0SKem-nL2xmY7rWQrh
zva3+QJH2<$IW$wydYY)wwTI=J=9+r(TkFQ#g}z>MJcuSXV2fJBu4!BBRFl++?)i`q
zO;uHxO8fBG<5`XCo^~&fh@v1B<M84TRRy0UI2+e`T$hnjN}(DzJne|#aS?1CUOCWM
z)g6r^<vPf?<_vDa)FNXcOV|ka_J@2v-j>oCsnM7$%iX|fgVlzR9kt#_T5HI&36sg#
z$JvgzwYa!Ix7+c?X+Er*8r$FvW}a0ZCQ0Z#bG1-$`XHuk-cXgUjgtm`;fXzZD+#yU
zUGUvM^4q-kBcEe=CFba{9LynPwaCB!zdph0B}X<bCKJqLhqI>_c>S4EOl~^N9XBm;
z`>nUIw|j;2S1yo`CX7Zom)3T9{_-Y!nIKbL)bc5G+@s&$MU;gDY_Y!GbwLy<T4_R2
zm9Ff@1F=fEofC{RFlYWEapxz1*YkTH|2Dq=`~Mkt+;Ip0-yi>n{MbMIM-&?+M-HE$
zx3J{e7!8KB+paVvuS#TO(5A$ggyT2e$=22;BHKk6YDts%BuR5D!QoO-9HC7~S(fyA
zU1uZ&6h-N>&qCI7@SvlH6y{#D)_OHc7>$N3EcBdb#$h_04ol0+w9*zyngW7WyX{JD
z{G1U|kms4th)FMv94Il)fK`>QGvxcR*@N!;fNNB8?sI`M5H?t6BxZV2T5ou?@$=hx
za{GqJy0KKsg|>*u3nxsk^&n$v4QOL1i;9Av+cq_lJ_}O6e@fkX<>R`HRN@i(VBw7y
z$THzet*HlE2;te$j?Ep%E^S+tuEmtG)T)Ei@!6kUYmA=f134Eq%W_viUj@x`2Uxl^
zt~<|Uk}(_&oJN##Dhb`LbEAW&G`-$HJPX-NXZ?#JXE^MW=UF|#X`0S3)8_@e_7@}@
zR3vG&Qx+DwO!7qr!^5m!9CGgN1c~K_o5l!XDRN6$M*P-qJ;Nhk*kxg{&1g`OWmhQo
zwph7!g?ny2inI~Oj~rsKeTDws4txCxSFY^v%<~tSn21pov6mYPO4`|k6UWbBb<S|B
z4_GSeq79`IcYR8svoTV*kj{A+-n%dog}_uVp{qU!%cQ)*PyX-!jQ{;7e}a#E$A|c(
z|NK$D{FSe=p#^EDOSkJ3GP5irNm8^n#8FI{Ibe3_@KMg)eiu(a^rxgz0J!HX?V-u3
zapSFtBFSV@(e1kOC8q@IJT2E{H@Nr*A7POQ9|}>Y>>F60S8RxVL48>G7FTK`m6IGU
zJr3ruCGA8ZaiOHK7NnYbiXIpXn`3iB4)+)CJ9xA%EOnboxGAmY1tsyuWqPiuhZeHy
zvYxNOD>KG=E&N$&;M$Ax;c;2H@eN6bgVv}Z)XO#-UO?Hv5QDmG&6~W&<GOiLlB(W;
zr2mh^H`ug373x#|%6H$;+NCdxxS}iycLwXgH)G8VEp~R2PA6>;*b}}ubu!7Ob$rIu
z-w|#;NT(EeG1D_|VPT;Ga!<XdsaGX@IOncA&oe&}L*_=+lEmR0!tWBomjj5I>o>K6
zXnj8tlC;J1PhH{l?}S53OY{eQ#zW1@N`fvezxAsRF&ZW;L=o+d;PCPpPMlrhrc-Bm
z@}b8$bM6T3q+oAnm-Ve(9(wdyF0Bm_Ntd0hptOQqTZAx_*^q49Cr%SeoI#dK5~~D^
zrYI&PQQ{1OaXrzjHjV*~!B!<nyy&<(O7cga`C}$!&PPA?G2ZspxANcrz@O)D{>TsU
zLqGgCFvienrDUGb4}w-4yXPZg7P<=<TT$gXM~<K7(()0qy)7gKRz&-8$j1B&9(++0
zzP{4d8&FYH<asd-ZLV^J=inL^_|2rjoDsf^OG-8Ud%#hAT3=ugR!|ffMo%}ovMgt>
zT}l^0DCM!wX_k6sqpx4XO9M{<e;dv<lhm2shwu?i;FkrOz!dZGBws0}4DtH2E<}xd
zxmCFCpfrot4mM%dkOi#CwOaRpxv(`Sa*Bhwx<eg|tu|OaRSmh;<GPHLluoo>DdqVX
zW_rO&Voa!Q5;Z0e1pzMOI?NPRMlsIO*6WlTjnBc>9*|NIagva>T68-dTCFziw$m%`
z^tyDqU0UrH#||In(9$C9PKzjxJ&-RjMM04}Pb={tqKYDx78Yo?S{{rg*d%928tM!I
z_j1iaNA<%|O4WMjK;1GcxVAwlNl00{+C^DXdM_-`bGEj(T`{uKlzU^^aZF_kx(kZ4
zXOFS6obt)v|0)-s+d~?G&NF&The&!oSdx6@@vm{HD%e>yoWFFD!Pbx$FK@7y3nnsV
zoD~#CI!{PChH#Njn2h%5Egxs)_$db4n<!U<;_AT!Smg?}wa%!NrWLh7LPy1n$775(
zyz}jE=Uwl32mknA{1oF!Nm*(B=|B6Ie9w3PIlk#_Z{~;p%3tEA{?)%@FxX>hX@$`^
zqt$Lv>4H`olg5IoEU;37)u>jR+h6-2kACJ8Sm8n4#!l@F!u*cwey=LGaad!~($!Jc
z@+OVr<qX!$Sw+~O!lLsbniO@GP<dk@lq<P$6*R6XLY`%8ZEd-dAt^Cd(C&0uUOr3|
zB^X_K7@~pWL|$pLaxj23PBGK7W@;?j3OhAaLO7`;(KK95O?6X;Y|asKbG@ppZzJy%
z9tZ}|ouKy9q?mSKu062`Vtf^Vn5srt?_?NjscP)ZsaJ&hU5A}bJ>hWO=GAIxrKBKA
zT>B~)cEx`8aRU*@&Ok*<k)$bUk^;)LJReV3-`GIKF$+tTOKWz!EG;gw)a%h|w@Ff$
z_KcDkOBBLbcY`4vi=`?nmvPd@NfeirtGP3VNX2AjMKQ@x2!sbGdp#F^C@HaePJU(T
zN!JfvYhB||!9{MjTF%J%8DV@9VY;wFy_egy@;qmEXUBJ6cHm={=dR<96f7>cSU%h#
ziVe%lhY)tal?$8vr=R{hij>lf>G$`@@`8oM3K6xja)F(_%dBp{gha45%E?U3s1%F~
zrvzD-mU_<vV<;wL`eK`2w}<I;7>^ym>_iEn6<d~$EplNkCyLwdvz4K;)-}73_~+wP
z49ZDHl~-I@UFHAtkAH&y`N#ea-}~pjlRx<Qr+D(|XXqpy7Q0<6P?eg!q9Td4tG{cv
z=(ZDd6tj5r1cy(bW%JTA4g+Z#m4F+GP^@ba#xFYU*w1crXfA?RCJ4_{Os15FDMiKx
zAiC17r<}1=dBJ!%Vq5MwE|hhcORv|PAqNK|B|Uzsb?zjb{TvRHEfP~co%94$6}#kx
zPsgFLu2-R$dRy8wy{lSdvvJ+t1FjpO*!us?H5D9#W;iXpw-uVx`T*6RFL+Y!e5&aw
zdK(b=!f^{2ZEA>jdL16z7cc*en#Xk+saB_5H#P{>f&l|UKoUiCI~@)kI>h4QqSF}n
zDSvmvQx`Q%Cig%bg`lh|%Ce-aDyqC>JQ<TsGV(m9EGx=N*Q{e>42KrGz;UE>+!b&x
z_jo+6yBABTNLy*$Ore?O389$~Q@%%-ScAc!Rsk)F!jaOHqAUsmq5mL|7msa~RmE^P
zWH4|Tkd&^eL$B9EmxjE`y*#mEVIcuiG47ZA`=9z_E<CqQr|ZHQV=Uv*2vb&Aj?n2W
zlD8)8Y);UlF-08FT3BSVGoZkFt?l{iGE{j+Q8@zP!ong|WhnBA-{~rDyX{7-9<sK1
zg~CKm<iE%r(5fUwUQiaUGj2YbFxeZ@idz^9KmG52j*oojw{zd?Uc(38`^~IwZje<u
zvF*6Dx>N`YahiD0*-(}lgTV+9x48N4d)e8zjP;#&>qCSThMR82;(FuNyYznW`C2`v
z@ZOpCAFy!d&+o9;sUDe7!p}c9$3iBn(uPnP=K9_R4W(((ZDm~eV=Ifbm79aY>5@qT
z8PzRx7k+y6*+|rZXQ;vo*}9o<(uaMaXnXda_Tx&{y1Gz_s2h;F9ul5n;1d+V&=?^o
zwXf~->&k7?2Z!9(At5z$T%RWuw-LVTUP(wJMXTc$-tx){hZdJSmm_VIB|u&jRFxqw
za<VKZ&kJXaijq9f$g`ZuWWpqO%|A<Bp-ZQuRA(pzNTq0{Ef#vqEcO;ySm?5}v_RTQ
zohV#YF&<|OM<XY;Cp-lrj_Pj2?wT5zodEEQsT&>)VXM_5ilVx7X3;kpRZ5X0DQTM2
zZgJqI2%%iaS%Z@XW2@R5+uq)0eQg~PDaL!dj7AkpE0$uC@q7RI>wNb2FC#>kJkuyC
zFxngMe2g_Ed3OnID$1nIV4P!$0%aXAFLedh_g-seW2~jhCv+A&jE4i15G)>AVq<L#
z7<5&zv9-aUdF$(tw&LNZu8>;+QG^v55xF|8GIK`6YK^qQD?D}p!B73mpXT5Hn}5!a
z{P17k;+0i?>-YYMz2P3+g#|ikhcr%nV~z=}IHo8xsZb~-h`USNeCKO<>X9!v&K7?0
zm_Xm~;ZChnR6n{pM4Z01e<jR@kT3JMhS93CP}cRo3tDJ#;$Vd+iU_r-apFji%gf94
zdCSUBRvN8A>x!bR(9QX0E!Kg4GhvzICI#T%;`%ZLFUDmBrSjAk<%{<Lfe+0*E@~_Y
z55F)+Z@?ElTjyoW80+BinBaX0z|4A4&inISY5WFn0krkB1M6vTzK)krT<jeKA!llU
zU+K6$o|H9oS}j&qRyca(Fh`Fpap=%8Y3dfUF~+F_WD|y?5tB(qHqOZMf;`KegoSp}
zzs9&!Y+07*P(p(uj)ddDq_TD?+nqLx3tbK^EwH$-NRp&=mQh7sugU{|i_3>OdFHgE
z>L)3kj$?!;aZ(#j6Lgw9mL?E0n`i(vORu9|R+VSBhrWUpd;Njqt7uK-UvofV*wu9H
z(MVic-&oDKKVWBfi_Og)HrICe+<)EUp4T4c<G=kizVzuU2-~JI0}O4ftn1e45Bf~<
z5$R}%l9D)<bh-<aMaE=2!BkEOqUn7wweM6#MmE|*C0(|+w^=!|!ph1Dn`@V8Cz7|k
z=`HN-u5zfCl1dnjbEhmJBR7#tgU+1~DSYEmV=XF*5J*1rxi9duzxd1ig+KqjeB|4{
zh0lNfD{O4<Q6(9bl0;tJpsFflnmRVKiYT?Ccq|`1Ns_cNRqhwCUto5MOF8T>Bsr!x
z_;TJ<)96emvT!sQH)0L%Iz0P;tvC-qEJC3qop#S}imKMaQqrj>C@=G4tt%BWHgK8_
z_U<IEl&g-35FjGw0pYM62Q)*d(L@TYjzGA-YjpkDnT@y6EjVnX0n@58cVSFv7*5ZK
zuR}C9C{rV2L#U~pfeKkh|5;~;0(joI9&b!T=&9$`243d4E+h5e0}rsWvVy?5y@OHC
z=Jt6e<IH(WzKe^M)0+l4ZRQ!=Qe4_TN?J~EUpXCgmlAh$g?6XS;^HF9%gZb+EzxPG
zw33u0O_8277H9=iQk%nltaVL3i=t#a$=T}<y)pD2jc;Qfz(xG7ZDu;WHezdLCs}yN
z0?RxpO?bR1iBmeQlvXPyNd-d1PHi%qu+u-piJcuz98UPmr~icC{F#R-wM9k|F&Zfo
z3R6;PXH=C^bh-<S$3v_c6GsZ89aqapr?I7ta7D%@1Oij9!-WNweB4J?8WFYF+1jGl
zbM3Ld^?h$={qh=5eeG-9bkh;4LPAO1np+6aR@eTzQeHJcIHrA+MwGhb=YHu|_|W^`
z!<)bHjlA|Xck`Lge}(aA<U(8vqH<LMr8OvP$g6_9DwyQ1Xt=ew#CU7fg-IbC2?ogQ
zH#`9YY`(Uido)*S<rh!w7l@_-v^Ay}yzQ_s14iT;<Hogta__T_q}(`+39H%7X=RP)
zF9(2rYCJ>cIvjd>b#9!csPEr+cK(!t5I}N+sc*!MSFM)O7&mLF%W?v?w4Vmy{uytq
zg1%LqRg}VSB3>0tSg-ams~;6wDAX<9JtpP=`fHE<uFFV$?a8N|n`^?uj_Qy~sR-e<
zf3RSUp(=IlHHG5XIF4$G3rFcsJ!{y-`6`+cK$9e4GMO+Qk7?dh;3b9Kq1h}!;7E!t
zYd3ue;UPK3`z$X6F|EiC?+d>)1B|c}7@sC~;%$w;3b3a1q&^r8M_6kpjlgO*ar?tQ
zJG*<VU%J5Nr5A{dq?feV%O*}4QYshU8+;)%E`*LFM5n`O<g<+;Wjrp)3zyZhfGe7%
zo($GwZmyZDn`C7%Ar=xDcNvVvD5?4QC;yb4tzGW8ae*T%i;PAkvfcJzs87{fP_dgX
zQ4~8Ja$J>s9J_h&+_TT|s~`OsfAuf_1^(B+|3C60Kl)=l@yzoKha-B6hmZndAwe6)
zCa0`QC*vaoN-9p=aE_<9)`&&0B7tq=?c>Y9O$U(;4|<Pn{4AOVYW9sD=W+8ugmYTz
zpXZHVv&_R81A#WKY^y3Yd0A2!>*ws0v(&ua;Drfr*pLm2yK!yC)*!a?0GY<UgiSZV
zYyQ0c+)^q8!O-FPU(@Rk5yZ@O2|--bZM=~QoH1*0h4$_|^G8@>^Wh}kskqkT`fB%t
zB8q$|iJZEQ+Ef%}j?R4nsJ6cNF>*zwsq0@C$1#y=63mptZqQ&QLb;l`0Qd#P0;Ph=
zVMrBJ^$mwwC<r@5v)9=4=>~9L2<f`T`qzaACFdudUnA4qsvh`hrsQgFeE6GBf;!Mw
zJ4MJelE{3QRyF44Bc~|FeGcDZ`NHGRF|I65&Bl1$ZC6Y^8dY9p0oOuYNV2@7EKTh*
zg&Mc|8tP}}{;&OnR#>QVOqtVN?2;8Z>$^ioxiFAh1(C!5g^cPF5~Wm~Ha6B0#li3i
zbmbKzrQ-km>7U`YTW;k&?|uj0`hoZG<Ujw{<XMJ_$1W?1B~lWUQ@3mfsbwiSe)<eA
zeC=UuSzv`LmTeBPOQB!Kgq@oM*7`I&4I*X3MAe|PF@9rp4KD(P#v0u)4p%RNZ;n6A
zvaG90&m@n+=Vp0O0<^84Yh&_xPGWP8`;C*JTGjYoY^{cAtwoeBf(R!xhN-H_%!LP>
z&U<M!vDbanQ~{(n_4=shPh+P>DtJ;&{I7hqj8qawHHncr-gbS(q*ShJZtB2erDBxt
z(g9PeEqs<@O5JxEr_g0rRNBX{vkY46K{}W9@C;xv2MC722}@Ui@&jvyaC{ToCAr!{
z<&B`bK6iL6!@#+*E|sda_PRRLczzy`wbm~rKR*6qTxbwSjK@RvE?ywp>67RJJ=r23
z_UZRWEVmM};TV}juCHOPiIkJTD9e(nC|!e47hJe{H5^W8j71&Tw%XQ{ID|L?5W1wy
zHpwE3h&x2FL`rz!;+W^pe;tt~#QymN??YD=Qov|5;`tX}@Jqm9Hc8Z~afrca%+LSQ
zukogDe319O`<?vkFZ?oBHnymW32~e<$((#k8VO`%QElNjFNKztlc!E`;f3dsu`erW
zU@cSs_Tb}M=m$*<X66O=V&ngQLo^L9Lz`DVzyGZ8ZKY}>;?E&xypa7Hom3Lp5b>1)
z&pR@;C+DaIo;YnA&$afP1BNs$jj_Tto1m65@`lpG72*1X^`WE3Roxm?KGX|WIh|M2
zA#3PcHH`24{aww#zua+MMhf9tDuscFMN-G9{(06eqELb;l0+hCFuUa?!?d#o<KtUx
zwBH5Xpb6_qb^d+s>zrnktOvt^j*c}OiYRd|fHhbQN(gl5*cuF^lEl$$%PXfis%zIe
z@V@FvH0S<?nLS_r;=FEWBDHWSWQD+*ioxmycGoUpi=2G0&2VoEJ;~|!hl~qNnUCp3
zlEP|aoRHfbqYH#oq^*|kL_BK+gE1Du<7Rd`k?MsVu02>OD}2++iej)$yr@uV8)*$9
zhEiKFx%<4A_%M|dVf&%K_=BviZ}8i{^LtcUBdowwK7}rHy=(vRXFtPNA9;*#dh46`
z>3{i?e9sU36|$nlh{`oL6~d`ON<~o`z+r_%isQH3!==j?eF3yD-uBn;ZcWdnvMn#y
zKJuWj%1>qYyetSA{ti}HPnYqGa3A&=8@!U~1~u=c)tjnyUcJ_p&vqGH2Nr6~FkDY_
zBX%c}wcqYEpKSe|9yqHv1nZ@H!u+hgd=p?7GwpO^17luHCkHcRd&gQ2Fh>YWo0Dle
zdIa8<(Y96%2{1_=qKZIEaFhxo@s0(kS4-pk%1bX0#ZHV}DoGT_L{UVm)n<A55Qlm_
zx}6Sjl3=VR%QHsfjNx$X)hu0Yl|a-J!Ajx6FCPmV?`m6XN#cYgabojAK&$P-#?S;b
zid8M>6=tpwPG=~MV<%h|$5>-M>9Zgk<?MM~bYnD9xcdmCOzDd^YZ(eL#lGfUR_g`r
z5W<t<n7$vdx?(ukV{dbvGM^9&LpB~T-rFJ@XY7qihT|M`fsu+RYBA1A@~R?<Q?MFs
z+*oNp(P!Y6VEE-M<p8GjY_mrb*l|xm3CEC^Q9_w#L~-Imld>cd5>shMq%{>c-FSol
zTBpi{P}BKqr-=t^EI<0U{wIF!XZ{s;-f<i6d-pr}g<ts%+Fe1~>YxJfEF7OGO?}v3
z3Azi1SUGx(txM-oDkL}T%+=3(Q}cM#)pqH9Se3;$(xe_2w^;o~)L8t+d;7twZnhl{
zbd&JCGBkkj7^^gn92sEc`RkkgeVzu<#69Q3CK+yQpV2hvS}(m$AlAjs{@!Q8!!UNH
zao-^voO(v#8bxsHl;>1=BaRI?utfUKQhtfl8>KZM%&TOiuoVm49xE#=j<C73<T6ZJ
zyX={E3%|d&OHmb!MiYj^jPZEvvwuFe_bJw_C@6};gX~Un!r@WczgD`QW5y6glCx*e
zaOB8g78Vwv(?N<h);fjFG-(kh5m6*OtK3kQIm2w^z)P)tvG9m-R(L9b3!Bn3ar)H_
z&q&Taz~&)@E1M~boY83Pz<B9!vtGANH%$>$Ni`mzOo<-sVt|P-SY1+SgD!J)woA9$
zW}*}FiDodIP-y4zWw|R<b>WuRuonBTV_(?Nz|5|8?G~JY@p?g3sp+I~UCz^6=r9`e
zDJByH7E_f}o*DokRf3WhAsq+J*lDv^r4-f}E?&OOfByB4@elu(zspCy<3s%6r~ZVk
z-92Jsh`do32J)(?lOzH`RTbQL^Q~-MzUYi&n0od7G+PTm*twg=sd?}FJe7TOntF2y
z3)+>1)WhjLAy2AQvL*$WWm(@hQ1j&+T%i!gc%rgFR~lUfi52e!A0Y2@aP6Vp_p^J8
zx$6Ce>|N*#<KPcjzh2|6bDt5`&PFUk&}@9=yriJA>Hlv1c-7~2kxa;lVnZ#jPmaY3
z<qO%b<+v^*_4aqXgHEU8Fp{dI$V*1U4f3L3G9J_KkJ#-G*xuP?Z*Py`XpAmOmu>S-
zyNV>SZ^Ws*f}pm>iPw1}Wj#ee2|=VHr{B}-u(G_&@}XsxmzP;wTqH?jhm2^~-MB0=
z#^a*4)<$b8OWjFUYm1drz}{NCs-vsSPrSH24IE^OiA-H|i*twLB&BU^ovBFEgm$7(
zRgRvFk=2BJGNc&wX}1+wHeqj=(cc>&ZAGWkVnod9?u2n(U~%oxYY$#1u3~HILn8e6
zQilU@HFeW)8^R?wqz|{X#*h%V+OGc68d{wNr>$NT2o)2zyT044M8z&Vw7yYh>ARvB
zV=2p$q}4(Q_~^%em;e6zzlZx?dk_EU@BbbC(qI4EOuRwR))6D6lvI@_NgTdr%aZP)
z!z>>=MSp#@_7b4EkPlvz^Y3{j++*sIOudC*;Ee$rT8;-#zd?L#5_rRTrXDw*`xoj_
zy()n)&TRPbGA#aC@BE-^KFG+<hMnOK!u8idYnU1q@Qp@-=L5E;0!*=AZz!kVP<wQB
z2GcrR4WF@BGDhGyi-}kMbo%d!i)zx?_cx4vc0VS|@|S%!&EvX^)cWcs2*@TGlPqg!
zjXP4NE6APrCX5rb(I_QqF|05c7JO>90wG)ph6!%03)dDF7C3tJD94W-<M^@TEG_m(
zQ!gW;4Wsdx(a@9aw5dVyLObJx5q?mcfF8suL8Ly5Xk5r<v?k9A1d=$83EY=@w`lA-
z!k;KkV$vkx*zsc|N|DA2U1V&nt-21r`GnDEm#xiBit&WQJ&?k%ySc~W!UAQ~Wzss#
z&c+^jHFj8<FDICyZs8lRve<lx8B#Bfda>V=nqSmX2v8D>##AM7(qc3macE_QcBji|
zG(uYeQA(_0FxGK{e3msz+UO#4jWGOf?(~OTmXyl1Us}7e%3u7;{{z4AYrnu79(WyZ
zdE*0o@u5ety7EkQS%1b_i}Lkml}@?oj=T8k`kHGFAZlomgBR@7C^Z4`dV_g6uhQBn
z?h>ALtx#z^QP<QNwJ=YmaMHX7d9%&0b1f3xMpl%ie`fy4%4u&t#FXlAkf93LlZ)VV
z9j=9&wca3w`RQJrx`Q=V<&9LGTskQ9Y|MEn1d(#oA!le3qBx?J#<bfl+Hpz}rzF0n
zH*|)ol^Mcy2ZQM$i&D;&K)uXyT}JA$uRj{r140Un5OrOUu^J;lMv91tl$1(WKAmf*
zsvKVx<iJF032mm^oi-~+4|DRwaZa8*!Lg%9>2y1FsY_K=<atJxj|eTGgWh*lY4W11
z3nl9ZQ_tnFv)Z~aSVfAY6{D@DD6|7hb%D*@gGrJU*Cxmyg>~5|kAZbNJyfEIgr+Jo
zDqS%i>@k{Tq{2WprW|b0-`!?3&L}1smB}efNhAe_j~*pGawBEjW4zr527WR5aL(UX
zO=4uai0sUL5d!^sfi;=gwV@K!L_lZHJnkUeE?*6INEZ}kY3OgPvvTq@QIb$;-*UUk
zozkL^^_L3)#wRj_PmG{Fe+RdT7y&YgdG^Kg{PutOL%#RVeh2^EcYhbpJb!_~XoQF=
zRFwFQ!eYu2oj5s@*lLyzAE9^j7{iUr02mn)89klsOpWipY?5Du*K!1)J2Vo-$~u!(
z`SH`Z`m~V!I+$2G|5#keC#~@ata}0$OU+HeWw%`mLN`App@E6tND07qs9^Bt5H9QH
zydo2nh`q71c%7X(BP{?M<c_2>B(A%-+gy^=HJIooscSOPO1xo;iIi(N;l|lalT=W9
zSJSXu;Cuw@nq$miWns*#>Q!Q#SEcryRSXzGgskJ-%2=|hq$;%IQ4|JKPMNgI@3yT@
zmxZN8PMkc>i4!L{e*74{j%#F~OJ|VEs$ekKWjGiz7z|inUuQ7bquXtB=#W!KY_~ha
zX~!|mtIG9L3ryvyuZpOjUY)nVXhR|uB1uukaW;(b=~z>dMlII?CUx$rK!7pML(KPR
z^>fxYR>>;Oq23CWsW7Hwm<`FdHyKSPj3y<c;fP3DPA{iaWks@llEXK?h8#t|KY(Nm
zk#Jng8m~NH7Y(;?wZj{?W(eT!eQL00!XaxNl8XboU1S-PPll|Yf0pGV$4TRk8^(yb
zzAJo{TY!!Shj2yRc&<~DCN4Yc+rlFR|Lc$cQ(phNdwK7>-_G;rFY>Sc-7ivUP1^CQ
zmO_x1ISDRdP*oOb1ZQrznJ3n+)b>#e?-yI@dzts3nuN_@tzQ<Cs^8ccZ>C|C!<?q$
zDdjY*YgQ=C@SG4RX??pW4W`6Y6=mrpSqwNPIh8k9{`-v~bc}Gu(}CmC*9d~=Xj$o~
z5mr-$?4ifFT-&XPG_oX7M3S~h)08BR>9kz;mLyGSwNm0F^7I?&g#ZLr2T#t2^S%ym
zYHTdlR`r-o$2QdJPQ%05C}xhw_ewZM<G3y(b?Eqwfbf7bl+}cyDBZyOGL<M+$g;xb
z5tPRBOkAB$+D<ur@)ReI9jDjp(r%?>qY1;|kkM$wWISPWYn#oDO}4l9n2g8dMNUx^
zq^aWEjb}l`EG+sm6DAJJGS0fAB4~)wxuJ6pTQH5YjdY!C<HYsJQ>E}bNlYXKaTF0z
zQBE>Od2mYwRwQWMXSBLXRpvxXhY%Fld`LO$Guqo@Jn7RP=ZGrf_<}-4F~cgNbMCdU
zc#PG}9d@?1xU#m+_RbF3WQ@@jfxZ7~9W$6mFlyd-TWgR}N}R-qV#3yyi*#3xq2kmD
zpm`O`Y5Lu`P(_2qkff<=hb~<?Y?LGzD{Aj{GMVr%|IL5kU;p!e!gqc92l$gOewBwG
ze-f<Ax(Xx;i_sdH2(%R#ZCN>boTS@xU61{BU&q|8k895M5y0c%v`qO;O@lR`<qe;!
zzenI`8(uK3nbZJ0?Z)7s*Qxg_h43XYp_yi=q_7C-_+)XK(h68l%VodQBqoVtqCl9A
zg5<BK$@oxGF~w5@N+5;9Gyt?fOTV?LTv!%7?Ql=VxFkRg(ECu)V{gH;4N!x-8NbpF
z@S>Vyd9BBF8L5rkAxWIzvK$eJ4l<Rb-Y|5CaqX*AJwS@eza~lI7@-7J9MMP6AN$S}
zMxo6DLM@RVInJ45f*ZYzV31A`7+S5=?`94R9~z65)_GCV_4bQU9uN)>*n4w64=RK}
zd2d2u5oLueCs<RFTfsy_B%QI6Qeje;(a}{&oDV6t_t@Q9L&hmf-5$F;Ynbr>$_!gg
zm=q;hWwE0X$B&(1ql~yTG3<T$DYmx%gz<2fswiB`p-3TRM4Y7mw;HHdI+sBB&D(-?
z-C(qo9)MR^Q!?0Eqr0?>N?LBg_{#g;y*{O`h?0~@N~$Wyn5qs<FdAvB+eC~-;L4&t
z_xUgJN1yr}?|R2u_-jA({rt$^{s;7Tcj+xGP-Zz+L6o$ptmDiXT@t4Wr_bHQ3y(kM
zJ7Alu!gK$39yL5Q2O8H|KJf1CtdR=i?#ia1(w3q$6s30IVra5iHJ}GCClpIX5=XS!
zsl$5wzcfvWB1IC%UfbLmYZWLE7N(xd6zmS3POS<nm1~M>jC(HL`006X(8Dm`J!UCu
zIT*sLGa1O>t_Gl9%y_Q{(PQQf{3{*TWu%_^${+g+Gz4oO#>3X*()e;AyyCIo(9$xk
zcAJo`ky0`k4Cwa<zDr1W?z7pnI?V8jiH`B^1@mPz^BEx=tGr2)3&u)H<vIvU>B>E%
z(|vBYQ_@zOIBpXu=gpL*W}IifSD+-3l4FY<UVGCL+ncLcAxM%KGu#3_;pm}7$^xEP
zz0BUIKtw6&;!&R4tk~6x)wNCf{cZAM0$x7D)h`Q2mIP-wUp}Zi@Mwl%^~dKIaueSQ
zJ9ls^MT9k)NC^sq6QWiEX)3Z|pYD>&=-qeky&PUy=12eTKXhCnq1|{%1=^{A2Ci6H
zmPEpJ`xPjpgzfEJ{$GFppYX@O|DXBB2VcjV-*`Wt{?jiJ2|*=X09oWEQYg|Sr7R1C
zlH7R9ZCrfeS*kpD)Dmm3FJq|eobi|j!TZM~kWZ&W7d!y}KjXYRs~zSeJ(N&;O~%#>
zTUgg5qTe5oWjUjf>#-Ln32~CLxX`B6Iz+eArPXSAm5vw{MIIkgUUAcBohw~;vI$v5
zfvckovM#Pwnk)7-HBT_ieS8xH0fCe$B(5~n)cWO;>3BJiTh`4e!gyMYCpgdDKMkh(
z%+&RufM}d!&Z~OW!o4VNHM|p76+ioBg879e8b-r)pH($k(;7=FP4@Zl?zvB0&e^{;
zzoXIk0h|SI><n(MO<~LWMSQ6W7Q3)Ova&*1W=^fg!gw@hdvl$<$WRs@y!%GJ?Hk^L
zolF?_6lEE+*s~;}A}?Tl^$LSw#@@t`i<qjitd0ikj7xGYs0t_A=Pu1XEal}`t+Bp%
z{OY`{s~rcX<Y(9DEIt#KqV;r4A+f@V-U)3z6Wn@TAdA*H`D9GeUgUN6-pi50D^z9a
zq#>niWEsZ_L5!}nBP&|#T8dSL>$dBGX(<(3TU-3xFa0_{@<ZRx|MN$Fkc(H=xNzYT
zN-2j?#fb~ct?NZsRho9E!|8Lk^4#N(5nuBNO^wGZzjHR(A|$T9RCtUna>fG>K06P_
zjbS#)>G%6=Y;17(@)a&!zJjrE%PqHY!`aiEK7HC@uDz}&h|Y$6!D9(nk1-Y@L&kQh
z6=kJuEOb$C^w!#1F`-F;Xg+)ELHA(nrr+o9J|qqrpK8jexbF!7_{$AxbB7H2?fcho
zxV~4dTTJktLJD*y-fmv*3`pjF9*ssdQ`b$R@Lmg|A#_F<2tjk_ZvwwpcNO=?zfLiF
zHWh6>tK3=`k23ON9V0Ad<w6ZxRZtY1J>KI3?|KvW-*FSE$rxL~xsyv6U4qHj+a9sD
zHgGM=3&Boi*_*i9!z|CRD9W;;Y6$ed+_BFAdWH8NI1X2H9qjB|M0lSFP=XH`tOH=J
zkO*6mkB6uzuE~u;g0>aPw`?l4a5aS*A}L)#uEBeu4hAq*dqoAyFa64I@YXjy$ZPMt
zi*J4JJNfCK{bk=s!BFHSu1-u5`L^|0QE}tVw{qe6XP}&58({3q9Wx$gW4vY-Xb4LL
z-s28!%pB9YC`z_=_OL>s5$yH{zNK2la5!RXYlp4P9Y)y*?F)hrA32H;irBXuw_!Yu
zWlG2mq0kiDs<9e9JJ*|!PWVa&<JZSj+;AF_P7?`1ZDIO-S0#N~Y7ceZ+j00&<sA)G
z8^h-H0S6kSc|#QL^;$S=&Eq<a6a?Y<aAC7);Qd902<gjDTv4y^plwa7=U@XA*toN9
z;?-{~>_+<x71L_#ATVZag~2V(N>^0I*DP5JRl%W7%G=+1A8&ZwU33JDS1&Wx1NJ7n
zR0T{j&D!b)<MD`0cG(zbY>i5?(t=Q!s&rX1T|s4Pm-ywbJt#ReF8sgknETv8hA!*@
zp1lITx1Q^4qmeiPH&<q6DaHfW`WyEdV-00d)-AgW-x<0nN(TV@_DPkl>X1Y0l32uC
zym*;^@vnY{fBnyYobUed2l=hv`vjNQ)=66}%A%miT{S>zUF)!_s_66<ID5-&Jooj7
zh-LDMct#yLzuXvlk5eFgqYhOobA)WBHkQF~!rtHl&%b!ypWFKOP%e~>q6iViMBcly
zUPsDWhe>H`>scx5#TgdAZg?im4cc$sjavwttb;YCbdd0r>;8T<z}-CGd{>%-L$CSQ
zn`_(g+0=pW56Ku8L5BOcGvwFVYvWavYVl!$2<gVgrq&F@W<j-ZaqbltVgxsIHe9*d
z<ZD{qmtz~xf5!bS7(eG;3xTVpk`n88ck7Ejl_`1a1NZRWx86^hM3`d4u)l*Y#%McX
zl$C7n47qf9i$X&j$E*!9RtF`O34-{Jhh?>)(%xASVTYT&X3OjYpONM8fyi_e>&R((
zkN*9yv@0FxSlK#PM*v;}XUJl`T1D*2c*+SyHm*OXyvI}ufs};@e<ht>$MwFmQ>}O_
z1d-P`5ABgY`=?*xp@+Z58y~!nzyEjtI)D2g{1Zl#Z06Z%t*MHfC`p*)C3oF>9~Ym0
z%Bh1+SMtnKof(G%Zh9Yd;8Xn$r*vufD2k?5&RP(*ZWk52OzGM6Avt8c;gJHbIGa;=
z0Z3iBE=f`dERC$$zUy<d$AxUu4fmjVZ_NH1oHsD@e`{`(FNY5`4at1=5~g@y^D|OP
zSF|oHLE9`Sb2hW1&Ck7hZPo$TFAt`cx_N`f|G&O9zeuO^DyU*k|4F|<Xq;~<#ud&p
zi$zV(-}Ihp_q{%kGpIBEl^+}Ji;}IeNL%sR+t2bX?|c(Co;l87YmLF$DvO6YjI&+F
zqakHpvbnv(V5}MDhLRW~+dRL%jy7)RQ`S1|DC3kIq!bthqHeY1L)eD#33ntzNIE~@
zw&CvM7MAfLn!q-Bg$Eu7{#m$|Lep^AfOYJDBV8LZ2<1s(!G_}D3X8_rin6pIg;U<t
z2BTdWQ)Nn@exIhdEl$fSNm~H?%YXX|{F5`Mc;LQ!c+-Qg=l4GTDWW8H&%`&|)LJ4d
zgHa_(vcOHZ+{sg4djusc##Pv%ti~KDQ`5ZNtFyR+srEq=hG`dLQCGgzgI~`j;j_=R
z`h*Ad%St<K^QuJa($$|<8k8?OZ!Gee1WzNmG4q9uIbLH@OwSS4%e<EqJSb}$;C7>y
z!Tmk(vJS?yf+2A*rDZCr7~0hYsB`)mXCp#$7U{{T)96dgT=%sd*B|cHgB)&l>RTV=
z)`>9CJ~j@YJJ{$<za)4-({sM^r8NzKe}&o=_G*OI8nhKCX;7x*rqf6HrZ;~Bci(b`
zM32}w{|wukt7KWl+G;_S_sJ#&qlu<4l1tka7k3MCt0{G#(pXTD-&un-o?->fBGB+>
z(HKWJmUjP8%wz<nV_*AM4dZo9c=FtN9G5e^=+sIDJ|pLGh%nisC`LuDB*|sFG_eow
zq!0}IebTf=US_VW2wM{jRTQ~2xc?^Y$%kM5$|L;tfB85+^n>5S_kQGqeELtm$lhRt
zlo6fCDKX@EhL9FrXoiysH{W$HFFgAcIv+b)h{j=Ojh){u#)gObGD9-|H{0A0yq7kg
z&e-S8HVj}e2u>-X^7x$Iht0ek&bkke+6QCoFG_}4kb2fgHM|(A+DN*Pb$0CL@25FZ
zb>!nSvGsb!8O|VpSBIC@)}CVsQ!&A5xt0k#J+9wK`NMsYnlT<Bj&8it__yC}z1n%N
z#nK)WJI}}IwF`}-Em)m%;>a?mPaY*3k9hF@*KzX=Cy9lD$sU8fRmw8QWCN;fi~z=y
zoUM^yb709dTp9{Ca*&iL0mcefh$@_CCaj!JaAEyFGh;yg-ZvBSUdv(Uez^8ezT8P|
z{dfX7-&z}Nx?6OZ+T)NwEL=yONJxw>opPhKBuPqD6bK<)WrA>Q{?P0~dM=jh2(9?V
z|MVMt<l8^My?5QtkNwTR!vFk_ehP~VrQ$?+Hnu?s#b}ao?AQ@boju3-XC6mrgOri?
z*yeBP8b)UBLjlS-aKoT6$Hc5u?krgCUN_d(jT(Z7auC?uJPrb$_kDNc!kga*7NzFb
zfCnFDLcCeom#~314di@CSUViepPA}nP0!n;Bh14yVrt;##_C#*>&K+#PCmPn?6<)4
zf@IB)he3O}3%>f_riO2SSQ!2;KnS!huw}trH=p6pe#f_Pba{cQs%W)RvQeM4%U2lh
zY>+Sk43i?qSizux7q%*%UmsGEG8k8s##4!G<y5?cYd!9oRZ2o)!R;6ZDJ{l?U25uq
z9(bQMPW19e(^ssC<5xW_2YuyyaAB<h8M&IO@LX^LfzA_f&pC<|i;Iiw?CfBzKwB>@
zD-1=Mds1yeRaLI5wAbWDDZlGFyef$m8=D*a*x&!h{N#`SJ>K$-4{+D*H}UA#pJq_y
zEG+iub-T{MOG{Oj^!NJQaNC`{^xV@#LQv{zpYUKkmIsZ*JSH{gz3d<7X(p>_q{86M
z<2>LpT2)!rjX~;=sj)b(h6h2&E(Uh{wB}EU*$Cwze5vUT2{HX#z+{_7X<n1PIVSVt
zr+v*|*i@}`s7Vc4__-_VMhsWSyk5>2U+uU)OsY9Zjd)|~{e<74xpZdS+1llse)bx_
zy@5IT!S^F&t*OIG*RU~~#{Q-vLi5_!+`&ga^d62cw;4?`OqnqlZlcEnHZHwDtJ5WJ
zbs3Gv^mEJVj^>56A(wVWsH8=bCRAFZlyybbBE}V^8cbzsJsoCx%go06T2GZ5=MPKR
zX|DbN=QX~sUGqCUD^XiG)7;>|l|Lw4PdOO_kq?g{Qi^lu&he$MJj5qI{TYmgPOryg
zJn=+Me4`C_GUE-A(G^x}mpsIEnUS7YSXG)RiumMbKhGDx@-XlI=C|?ZzVrP&^ZX0s
zMMaSp46X1Q=msStioD?P;bR;=c7naN%S2N8kYZ{`LSQ52=3o8ydTg%+KXCt=0*ID|
zVW<~Bjk)O(1db<ka5k-JXy(JLgMsGexvG8l1VN{N-td~a`wfPY8ct~%lqNnlV`T8_
z)vJ8ECNleQh{n1P*I;JO+Y}0X6?szB17Mkk2KyiU)#KZ#uP{xUTHk*9pq;I0#P)%U
zez%xv`b>EwN=nLn!jVIZy#0-@;r;J=3rmY#vhf}*J7RV73X{=*BF~UQGa2nN>}y6<
z%#$xpxU^TWH_AcA#8OdY8AAG$v4BwG5FU!~_*3&ZINqQ0NW<i`wr<^hwXtjZ41#Ml
zP&6R<DV|`%v+#yaVZC@=h0)4kNy6z(bvtc-;QPOauYBcU`nv;^Qlx2$(G{k0{oriF
zKxwNHk^3I$!w{<tMV7n%j8bA$<eH00#n1oBZ}PS`zmac!-=E>7^H=!y|MAzc#$i%Y
zTH-i@Sc6qmmF473H?w}_0xc!bzIPrj+9qI9^Uq<Q7v-kT?0Z8fgICcM!iDRBIzyQA
zepIcO7WUcVX~z|fV}ZNR>{|l|cp3ACES#H}^)kJVa9{K5qnXt<CY*!DdDv$#f^iIn
zWB>Jrd2Z`7nJ~^a3`oGIX2ZhS5OKObUNs7WvD2C(XEa})ly$Z6u=)3cUT?fRgt7;G
zX1XJU-?arHBTSjG+)4PRx4wmM{DucOxzeT>?Xb3c1yf~A#$$&4ZOS@4v5a!fg`I*I
z*0;I1wgUu0qhM0x=*r<J`|rh1<O6Xr>j`^TS5Ty3l;+2dePcWKo&!c~evhl)Cz{i2
z_I$y>N`WJ3mZcB>f)-e~Mry33@T#h&a@~wY><yp+BQ#azR60Y`*szPcCaMu-o+G6o
zD+(TZ=n;PMU;JPEPk;Nb@xk|fGr#|-&vSWg17S6$tPqg|QxS=n$z;sw8_x0cV-KUp
zV~-<+d$H3SK5)()G+Hk=-eH);+<U3@rJPie@Ls2_6+25$L8vOt)p&B@wdvod-i|w0
z@N}jgNzmUq$m0p;nqIc338mrt<_*R{A@HoT;A&WBJwi($(1!>hoTrJy&EISDJ6|;;
z)x43bK6cUEw_jm`Gz?DDT@JfOc)f0TIk)<ZacO94ERj-_#e_p0$%AjcpKp50gPc5Z
zm{>~6!4}&atK?OMsd7v<Lg<pBs2G=Uabv=hFKu&qXN)F72sQ0gVttvwfq-Aw>B7W1
zC|sL<ZaM*`Zg%sXv!Ao<ALXlrDD#(efGM(dMrPli{dc?nJB@RNwyw!)RTf~|^+$u2
zcRdNM#hMCjou;({ml^Z|g02*5YCwb#R8{GO_${>JgfcJq!%uvg@BZ+Ixb>EEeDhl#
z<d=T^x7{Ka7Og#Z$T$gK6(`(!$2~m$<uCe9!PDpPO2@v1-Tcw+4}DzY05u&)8?si@
zVqaU&&uJ(V@Th6l>+0Uq^dB@Og{RdSv($k34ex4(X8t|<djszG+NTDxne1t9$YH1(
zwRs1?4)ehKyn!vta{t4*%5nWXsenJZD|y-X&rh3!EdTV(S9gi8id#vSSrb)Br<L;H
z8}H|v-}XjUmU^@k#c;UCc7GQ=-bNP_jDfPunG}Y>xZ=Y4h$k=XaA9MJ5s_C0sE{&t
zi%)w)C8&kUX7DxJ^yaRfy*X&o>>I0tJgS4<{z@U<fnTw|E^ubVg)77^v|Dts^367^
zFR%?Xmquuy8j?5$3&P2_7!SHj5%3~v4+^H(i3CRD32Bm&wp)y{l3)AS@AJ3*#$V#E
z|E2Ha{N+_X`PnZ}Ru)lI#9AXogjEuy6K=faHeP!2NwRU@Eo?EpzUFvd?U;X#2aeiR
zjKMs_G4JV@WmOw0CyT9TzHk0}XAO{@JLl}&3Fn!=&zGBH`!TNl9@G3a%_a<jv2oW^
zH=~8R`2zdag_x06ZW`m@ZC>kf{lGa)HIl^)!bI&^%s2#M>Mq*=0)&B@y6UrdlaKL2
z0VdWF__Rf+X}N|ebB-+cc=H1f@{MnN9V>@=#G+!--(s-Zrz$jso{%cgrDb<GWK`s=
z@4|Bz*Lm*p7NbfaRD>{2*9ZiaDG|uVg9(FOj|AbtO@D9Pa$_Pivo&}e&WbsTg|$KJ
zY&ryCoSWlMm=x0&T$s|B_&)TgpN42&otXKb#7PKS=bliECrB9~k`8gy0;3U9qoO1P
zB1FpV_R5#O7~z|`f>&#msH&P`P)Xu|XY1ubtic+G;kDZ>tc+2u4qC?i+He06r%#{a
z`~Tdx@;%@ALB9CVV@$FPtqn>lMw1~1Lz2eydR=b6_g=pG#m~4tb&Z7(9#oiW2f+^0
zVektFi?jsPuHli|na?+Y?K)fw^Eo^>6YAjH8G^!$(Nmt%JdEKyjajO~^j@ZGO8O1u
zpyO%=D*UZ!B<AnoAh==rzNuFw1T_L12GQAtsk%wiSj~HpLO5NlG))h@hl7s)ws6mV
z8%|<(Cb*mPu7qDCw$6&oM)zS5o8(8IZnYL`gv<Bi^m&Ae+>UHrcav{=`#17|_q>ba
zD~Gsv{yd^I^tus~GAB|ND-F7ktncoT7ZWxICC^^m;i;FdFwPCAlvsk*)l4t9z;$x#
z-9JP`<9cU}P<Z`n54@S(ur|c^^#HhVA@p3e&GDQ)niJ#Yj@jJS)!x*2PEum6#Fhmj
zG8mm;CBE~^)K~_bL<o(wL9?EE(Fh@F_ZG3^0of!Y6u4?!mk<}OdFc_Y(Ym7Ln>Hxn
zByxWD4?n~E{><CB@7_Cj;I()0>Cb)9Wdge`gw$lC5lRW#al)B%H}m+{AE6kJJz@0f
z-s-%es(+euqG};mW<!{(MFC;s4r3bd7VTSKRh6c!Dw4`||FL!Tfp48Ar`OzE;L}aV
z{WmBW1pTORPY2yw^R;H!6^^UT*?F(=AR}|Nd45m`+kCx=p=sY;>RONM1I`<qk8sG)
zCZ~4!8XxfPVF(Yy=HxEc|8me`^xhS|#gdRt<<Uv|3huq*cHaN)cW}q8HxNt9_UZ*D
zgKheo89R~2R2jw^F0E}6#a$+a<oWY`zV_S|HU|Y##V!SJE!e8|mcsSCTv!(FzwYEb
zbKO@DS?At4-BJDdY}2&7f`z$y*c|@O%;MWW%62;Wg+Qp7s?ylXH6|6plILThD0RTO
zSKI`Fkjl{wD&KP4Y9xwSrC=3Nlu+R42w}4c2HI)eSAc8OAR~nx8gzfp-My52_2E4}
z{kbpjogaEH|M>6yEk5|+pJ%)Orx=D#-UdODyVhh!mRC4)<1IY*wMV=HrSmk5vEIuJ
zQQ+)iZ5n{!HBF7v6dbmVNqdlCm^T1n$V;s$wJS15fRNVJx!T~396bLI1lSv%O8A0#
zLv}SUX#RZ7`#<RVftfIV{yv-DRdX!D`wt2g=kF(+H;!MWmR8L#sUN9@(4AE<XVt^Q
z4)6AE|Ma@AvKobOz2&Ww`88EZI}+S-^Nqai&ELT5?!Jv9%U$x}gzfEhM#F8k*Dg_%
z1;Rq7)nZavXf1JmV~5AS{vsDw$CLuBv`FoCFeAVO-5!1QML&Op-EtnxJZP-uLqFrC
zj-{OX{Vv>qU<!hbo(bcwVw4UxsB>0o{`%bq{VMh~UADqnr@yV7GMc-`#!innXo6XR
zeQO-n<Tfb_y2P3aA)(TsweL0~r(v1aCD@cE$4#0Nn+ty8pZy!ooI1f9?!S+3f8Sr@
z@BPE~bMls7<?!(@qpOlc2(oO#&h8#3PM)fbR5SFw7Fw+J;~##RS&&SPTyy-d7E(<&
z7}t^8+w?kPs$^LFZUKi*Ma{)O|2+p0Rk8D4{nfpmaGs_yJIG@?$k@#rpm2n7ZeDXZ
z0@!{K=GMfifCD%As~!K%Wijr{1Fy&RaGUFY?GEe0z2Fv`na|f#fDnHRZ_=a%VKs@c
zoV)2PZ+YM~e8U@FPoyN+g5l0OJL~I=`nyafL-O&M${3CxKf&JKkd3{JN1uC!XD@7#
zl@<}Xnly!^QcinaB9KNROk*~_!g0{V2@mX`9WC7F{t4oDvtZAu{Wu)fH*HYa9IJVE
z`(Wc0Le!TtRgOAgy>^uLwRToWM3fStz>A&5v708p@v+~<SnFC_=?Y^EQYeIsP%3f!
zAXA}>+?6dG?HV#DiM4vhu$P5lA|;*fA~H%yTW!ZuPZGAam-y#Db)2958Tjyr;otqc
zo4I`9y{sI53@bE9i4+#7NLx-WCg7d*JenJu0~cTeyPEn!wU3F#?u#DgF`cP#@}Coq
z%vl*1&T6B*A!g>f=E3h~$QQm(2xsL)2$#JPwg+4XOe39UICqdyYL>4wuV?0ao3qK<
znR(y*eDfNbr9j|2Ir~!?JDroS6sgP~*KMSx4?f7Ygl8>g!n?*N>Or0U_Wx_|PlGMX
zvim;lxA#8h+<V`Yb5&(^RrNeoqZ{35%wS_88U#R4G6{k-DOn8BwrqzSj*x?4$sy~D
zL;Bzd+hIy(NDPy*7_=;!Ga!?)OcDl-d1|21=<e#OuI}oZvofnPU%ufE=j^@xVePf|
zKIh)d%7mjr_)w?f)q8n|bM~<ITK_czDRjWth)rEF#JMEmch5!M_tX`>^V6T^Q{VP9
z*REb*)-^0ok2yWQ$@$4``n8l;JzH+*qhrAF)fZpqt6zVeufOybD{sI$jEmI<o1}uM
z3pu6<JlMibH+Z}zZ?|Upw<KhyI}-7@AZ2z7g1Xn;l=T$0w)2_y<4_SWsf4UOz&0Hr
zo?vXi;KBF!ek1u{(eS_cxBo5n_xJdX-~1h72sEjYfFTf!p|M6<mKlTT2U1BTLBah-
z60lu^v*JMw(PKkk*lbvtQ@XiS!7%MiRQ!0A|L4y?#;^Sv{Ll}<6Hma`pZ_p}H_Z3v
z%ocm>&01o=W<FoAI67i^d@Nxj)~2d(4`7$8_9VubJN5(=Xl~oNVtv>u^ptIv5s9+z
z2TTwTaX$?7A>fQuwN*4-Gl+X3i!Ibt*{9^u<^3Av)C{m@v}5~dQTOg=--2=c)h@&~
zT~Cvd`UIuzKwxioJ!I})ZTOoW&Pod9*G(C$zG9NZuDJ6M5gUf@nYZwXkAH~o{mj#R
z_jf$a(PF`@YdJf+&H3HCoE{%@`^~plp06N!PF4daYq;^&ns2=HIxpY2!|7^(#(@pV
zTT_~*s|~1(@1fQ!&7sWye#~mB?&<F-1mAUkRXHZ_yzO|+<R3dfUcGLdwj~om5ktfe
z(&J5=HOyzyC?XoLZ3i*n!v;0{I4k#X&Prp}SOQ9|0pD*3e!$ySn#5Wu=^TcEVK`yE
zI>)&N+q87^Ic>Kh+F#?p`ES0TANnD=&($aXDML5`7E-gxi!ZkKp7~<I`JFqW!t1jq
zg5m8{^YP-TO)3ylWudYsniir$x-dhf^%x-pLkK3BLQ?bg^xQgLXo_g2<D05-_hJw=
z(%EHe+V%I8Zl{Q9m+fi#teV;B_cD@7jx$|bMTEwfcP^N7S7`^mc28)Bl|R+6Hqp@c
z8}|2RL=)&5!-t-HFF*Kw-@%iQU%~n%7iR;**)49oc9+$9#q#`&v-4x_o}3ZgjNV&r
z+}-e{=U(O;FW=;RJ>Xo6X<A|osfEwjKB3wDZ>N8c>=h5ZKKD-^t?g3vXJWF4)WGOo
zs+<tgLk(+toKX~S0g=L>tZ_KoVr`pLII%Bn&i_{o{SO0%)ye;sF#PX{@i*|nOUEF@
z9_5oj@<WfhqahW@^ODQvjFFytw!u0>j6UV}MR1OO1*dld!MwrV;;(Y)>KX8P-uLmJ
z<CW82#5tGp0wo~SZ`K$(X0w?HJ!6wy?7`!urfWME4%_=FJzvPo$X1?S{jB%WRa<@{
z!H-qOU24`YqxqB}>EHK6T{~2@DRa7)3V1*CwCk+wqV4)`YmDrN+BD6WneVi>-Q#-b
zDhT!8+lTV)zXullz~SDU4}I)?{NVS0CyPZ(Yv8HtkKovFesY`D`3bMTv?O|1uGVbU
zJ)6y%)ALg{!Ekrk^9Rqp##f$yiIa5%D?R38NVXE&U}6d=%%-nm5WV;Ks@JP|-it3h
z%psWCrrr)V?;g?Uj!b~+0}ceKUuHzYV;oI8!*5p7b;Kl;K75RR{WHYy9sqp*eQeHt
z2J9~aAvcr^Q8M7Iv&3PO4gU#`;F^Zba>Hi5mU`)~#YX9pVjBlh)W&YUptG<U0*n1;
zY1(f9u-N|!ZTBiR8e%Z?{Xp0(S)bj-?C)*c>vkO0`>yIvw31XlQ$CZXrF~T^m2YBe
z3PJK9qbK5tMk3FBa?a=h&Bp%8Sn6{%Ekrk+>y+W_VoLSBN9#2Cd_A`M?|Yt)xE*s<
z?}5-!3Fnb7?Yz}}uRoMcYP3-DM|ytx9U{5dn-%G1jg#!~Va>IRhy2(Neh=UE?H}X%
zwTogD90tO$V!b?NSfAodV7*+k+VpHT8`kSJeeZee%=7Y_=X~Ysukq5Y6Z&YRDQNVm
ztg5&+eT)P`HpBz6aM?W_@6RYqI%Vm!iH&F5IKHh=DOJ}ivTY9-+X?bx9>|U&G+^@I
z_x@m+O7&0q+f<Dsoui%2*lZ*;Lt1aQ_{~R%;qlQe`}OzI%_H5sWe9R0i9y;L66yO*
znYG%VOhebr>HP|b1mDvS8=AIZ)^!9VY{0}oV;bok-L`pdXQy`|N_vt_u}6#6(T9QM
z*(rUhVHHVpswS$h-)o`9);*)c7Rz6?8Tr7+v`6}8`c>DJlCr^4@Gb-|<!VA8TAM#B
zLmD%hUB>JCG21#x_+Fp8m&S9iYicjr#SF*hlI{FjYi&~N?ASXT#{%I&uRnAZg!Ie`
z73jhzday?Av&1++h%CC64}bV6KKqFe^U06DpUa0GXUBK&n<bmg3L68?LiB;<FyL1`
z>-Cz|dc~d7o)=!d&DUOfgV)|VXXR60F(P6(2&wm2VvaSFs8`X9=W#EIqD|RGW(wsd
z>4g~ejcOf24UKgK+Y)aoMHO}2&)mD)eszDsGi8Y71tYc+5+TJ1*3Q{KctkvK2voP-
zZ}8N|{(wL8XFkc_`+M+>Z$LZy>(WZyNVGae4xmfSZJlZlW&4Y9M4)LKf?Y_X*BEJu
z3`5T<c$%ieHLV2r8Hvm4W(zi{`9ssS(iM3K5~dLYU2A}WVXzECVN%m8R_jx*US2a}
zvL`Pl+mi-N)-LsXtB*;@#PPjj5^zyDV2GnkgQ?0_Bba-cBJ`@6)A*SWcKD}3qq|HC
zyKGbSoORyDC<_UB&2${XpvUT+{{0SyAWS?5NRKjUc>@HgqUdA9M3u5V`*X`j-}gSg
z{ppYL#P!Quy>Ni@8~T;By4swbuvsq=DVw5otzo%ZakBKh@cLVP`S}~X{^kj5AF-}U
zhqF`~EDYoR@VUP3XP1hhl=T{^i}%sVP1X+#H51DSGmed<dXHRP;y%~g*#aNxQUshj
zOUD@K`xQ;==p2NwCitGv-{$eh|L<J6@*nZdZ{EbT{~5F97l=3pFKxn|Yakg!opnS%
z>|nD}t7<Vj!bgjb1EJs0d8xGETtm}z1YmY>iP^kky*gz`%^$3_bY08I`h>={IO{OM
zvpHW$AnHRzFY7h3J??n#fsqN@=r=~GK00`x>Mscm+|OpDWm1eOZS5cAvs?4IkF98{
zfFIj*KfUg!>J<h%rspZMoSHcv#vTO^&7{VjERF89wvl)cZ0d{~a4~ZEXrB+g|9w3D
z(f9G#<;z^y>*(4*XV#pa9J4%)toxqR+o!CS=L9x14Rozz(;L2Z^Mo&b^9{cK+8un5
zGQiR!V89ZH#%T})_n@hr8~s2B+tds%*?xji{*$h4hFE_`V`nrff0s=veX;sJTYvAm
z9@a3tQx>{LTGWTsGp`>oR_b$yVSpI<_y7Gr!O7_ffAeqs6pOvLSg+T)Qm5R=;94=Z
zJ7?2ewl~9X)<8=;KV&gK1n<Q??~@1C8M>zB%9YF9z4Im~x9>>j7QjSH*Uo6#HEq}8
zql8Zc?*Yqtqbg=XCJ!8ZuzhKy_EebUeY0kiCZwfWMyV#mJjNJ;)OA<EfTCY5rnilP
zY6JP(8N&77(-JMaK6fw8;a=+HE_>PZ?-(UmcZ#B>45^MI)a2_2y&gKjxSo`?mVPs^
z*gIr4o3nSY$5T(f2fsPzvBL$AUq0l?>(`jgEGKu4X)rVd&QH%EM$VUOZr?p-xm+cq
zcgOkphUM~{7jE3<559Ga+v|wKVB<!Zplt*%nPg8<NUipSi8^<$@ju8Mw?3PtuO5U*
zo#1sZd)2*A*ZtLqdyh^3z8^}w-xXubZJa^~IAgF$bu-4%&5npO2za{r9HJB{8#X<0
z*w8rXOx&A@Yo!Y6VsF9f><-3kGYJU*2^fFQupF4rAIIAU*EnXb!&u7@q<7!(tyfu|
z9MiRXG~FHm{l;^Cwx*lyfeWlROFAM$6!mZD`?ojfB{jENgq<*IWN(eTyj~`U(4;SH
zt!<1_+kL&+u-<GWPHO1UP=<W%X!SE?OuNQ$Kdqpqi|OlnO>_5l*}|sxosMV9Sf_!d
z^?YiT_U&wFyRL`x?o|Uf(GY#)Jx@KwcYWUvad7D(ZP)V1LCZzAV#dIFdCvQvxXQ3P
zXWw~Z2&^}2Rx7C?etLe2v(Re0^5dSbzw{b6-#X#;*@ixd?rNgPxHc^k80+%i)6s41
zUq!sa%gmN)WvY+H7zsuWmfp-{qeCiimaVE|v+I4=Z32a-s7l}4RoK7&-h*C(D0V2o
z$vh{_zLN;?&`V#vC;^89;0gS(&;2nz^@*qXFaFm5jg#Zs#NffTxVDv47~f-Ehjoq^
zdsgSCQqtKL-kpJ`F%SYV^lVmV1S=hOT(gqUg5YWAb8L6Og=?4S+8ML?J|H>fF%I~C
zz^^xi(9=hWvzpKLQUwXMAr@E37J0<+prS6uJ+69h^1)^jJzd=*Gk*{d9})xTeI$BA
zYz*3dT)N5(S>WYo8FK7ZOsG`oA?=%y+FH{?zsplte<lweQWU(>HbhK9!G%B`1Vz=%
zO`SodI;kMLx=f9D-BWDjJ6ZHDL$=^~?D};+{}VsSqfdMY({->uWxv18h0bzzy5j8g
zF2ku~xjLsGHr%-NCVh~+EgNCETywtadGU=Co`3lbZr)k3@~MfWNfFqwRCSaEs=W_m
z_w24K*SjGt?oR1Q<}=x}Wwtj(2q=<!pXK`~lexzj>P_9jobJWo?gh<veP-8dF@3&P
zEA+Cq#n>5b>xe<(qlRG(KIRmRC*JcO_V@RB_Uqpu8b>>CXj9Wq?>$W?L7r{fh%G9H
z<jt+d0>#@G!B~SIHn_Ht)>-Q{*huRf^Y#*lR~{i+$7a2v^mdP&pPkUo7pY=JWEi9}
zMu?JQsAEoIq^xP&vAS8N>|V?J*>$Ghsu@&$pXEv_z%I|vIXyY0bq({yEO(Bv*3xz}
z{X$kHLn1(fNfGMRI%d=&Wvs&x%IapiR)hAoEl9g$8_icW@;qZi{O0+G64h>!wxai&
z8vduMV#!Y}%zbL;-;Gf3eLWnLsw90Na&Yki*B-kro|q6=-Fbs6-3i8ePEPJHY?i$6
z(kbU>Ct%?2**T_>YK<p%m%Q-GZJvMi4!2I1^d7Lyh>2_;z<VQ!U22b_BH6m6c1&S^
zzCj|sSPoQuA7czs>?>gtGbZS);!JM6%b+-g$~v6shQHlmxz{@EQVZ+9jS;&_zge^B
z_P|-3n_*IS7b47D%fI_q{+B%W_2>DA|M;I`Y?G=@4k-su)NSW1jfKtmDbY(}tyLyq
zG*>fuTTLLE$fjRo25E&w@DiM94E<(BKdiWL^--Ee;;ulDlu7k!jcsTeLu5sWk=1G?
z^V4-p8+kv_X^bh(+I#HPu0xnUHSrRT@1tuh0i9mZW7KtXlf57{SV*0L4d5J34|PJr
z(}S)IMqaZxYjZ=>LTi4T&Kiw$k6IZ%+0+Vyu|}U$^6s2%Msw4i1k+H8`ZN08xLZ!4
z;dd_LyvkyqFBUxd_+#|Lzyj#kOWu0xCS81;;jrQMTW2iq9`nNaing)1WN<uRpYr;x
z6P|zhCO3{Z^mO8;Hh~yC)~K4Yeb9Cx3%zz5^<s`NS3=He$bFbWt=5jztf~l~++SW3
z{$=tPr*hYrFed(ccsJgTDLse;x*xJC^D_)dJvd*UFq?JEyM1Y*5TyC%2S4xx7cU<1
z%fI^T48uS_Z0I)|iB)PG-K<MB&>c1gR_mqIb5AkYBaE2Pfi#UyjYor^fW}GV(P#|b
z_uPK-2Hp=`ym*CvC9T)}FyK<#_gQ1{v8P{eh|#lNivc#IY6?;GxKYegob37Zn%ZP)
zHEEh|Vv?G#D;i5v;PE{XXGG&HO=D?WN7uSaB?_5Q8k^9r(KM1MAswlKr?u61H1;Sl
zV;#GIhn>^Q2tKBIUKqomhU@ftjFR_glLaxOeqb=egdu4Yl5HVa#$u%3qLm5|CNsdK
zmYR1c1mQy;{2<0!{4mgTEtf7{<krnMXzYT`<{aDZ@%pPL9KX5Z)}1A--(Wj~F?+mq
zy5WsGXS{sl4mVFWEQ7_id$eJL<kmSnnj%ut4bn=c1+Zx&$`Qy1qB<E<#-mJ6JTDai
z8}e(qAtMHNUnW(-1x$v=+Wx4f+A<cCwT#h$N%oO@8EaK}%ZHHhSnqY^L#M`GVix_j
z*_Sq5!;;`5z8Cw*reEQQ4L|YuAL828EBs&n=HFtyKEnWQ>%_s4T8vAqOJwNRQpJ;$
zhNP{#skLU4Rd0!_>ab0VZ8}^xqwDr?wxMe+-Y>az>s8k4Qvmv2s-^}XIp3V&Y{U#b
z&cbHBPCJkoN=$9Nb*GZLCt+_>q`gCxn=q={7Nuj)r937GN`LE9tg%w6#ma$o&W#M?
zIapK$Xr)%wbfh$|D5^Gd6kAq;kE(u#9fj>ulUB8RDIL~jeoYj+nRwr1{igi`bv!jv
z^Mia|LoP<^X%oYXCNW*hdcEF3vAeH_Vp8Av+0W3Xpu0`q)65t6e$892zrbd_=FPJ+
z4)z}B<!}5x;p{d?hkMc@`#7+12W(~wjvl+n;bV4WU=zk93$xkul0cdoEDpnv(hOn+
zjg!kW`lNy?qcSPytUAPLVG08%wwg?!l}@fdmR#e3QYJas$CNPvm{JMCV6)m7A@>-J
z)}<~xN=LCU#~4eRh4R8!jHw|>T$Qj!o|5K99{=<jGr`n4z{V81VK8AsFar`TZrFqs
zZM$Hvxxjqh@*_X;!+hndU*qLhUZ?98qM8vIhSDh17)xs-=c{vK6ca$SvCJclP4>cK
zgG|!824g#dbI>}BHFOOa6Id=!2w|XYXIR$(FzXstCwFj*18@#Y!1n{r%?SN@wsmC*
zu6t9B<9_UX>ZQsZRt&<B^9MmnN&CD<mI<>+e-A#e9(q=r4gF?dv)a(O24hSvX_D1h
zOg{XEIAV0t1ftw`w7(`TZhp3MD8Lq7<bDV;j>tsAI@DPqH?xhwMX492@K3dqd|p}=
z;7J%5D-~|7j4K$j`f7}Hz<$u{p_tU8k3J^tvtz_M%NuXr;^b_}os+w4)@%F_xV7G3
z_8%jLj#qBIL5T42&wQFok3Gq%4}{?HUKx-WV0|jE<_9Ug5<;LK27-3nkQF2bk&<e-
zmCuL7yu|AwT;J381H&Kzbwi(+U1n}(M3fLQDd!JU6Ky_WX`_o|29ZWZ`XS8|W%uQ)
z+8+SfpedqK527JS-6U;C#25`e7O4H8SB(Voc%yilnzoaM2uOjpe%%9}&}>*N4msRE
z<g=gt6#wEEe~BA6-^8^Y-Veg4jEu!K4re0k)e1ih)iVkq=dl<7k9tI7^nj;vbDZ5H
z_@49g6Pm8&VE+QH?VvHVGhtFE$H#0o8y0)gV#^O}y0&4pA$5*6dCs+cN~Ohg)$VQc
z9Dk+`rN{^J0m_UR5L=WFQdJ@-UHeQ&r1W=D^F%{1+2%=RJX7WerOkV02mql97PU!W
z-e+!kUW#txSc|BC+iDClKcVk&-d^3aFClH4RMS+0$mG)i!|)E354KH9lRXlF$FE=K
z>)(8V<$6sBJvI*PAIv#DkHq~Wjuta6U3r8@o_rteV!=WuIH8Sc26zY(iCn)f9gU>R
z#AdUh@B2gmfx#=IiUlQ6Hhf6N-jM1}Z#F(5v=q9JM7k@}?E9YH4}=&?%j_V`E;GpB
z!6T|tkoHf}gi%ufDH{?~^GcIT+(y7yCv}GuZ3VHZ<Y?n`!YXYkBSe$#M~o$f>^`=t
znExL!*5T&+^r5A34UL=AE)0ZTqRfrq$tR!07{iM%y@D|o*T`7b!yt&vh}m`47*0;b
zZWK**q$g^Sux>>NGLGIa=~tc^5$6mcZdk8-4i-l^*V4y<z8B-FYZ~VBJursV@*KZ8
z#VsVUdENI}vYQH7=wX?vc=ff*<~RKb-sjJSR2We9!e)eKO*zn+0s0_9a_Ey*APMrd
z8rd`nSy?+O-!dNK*leH<9DP<%56AuZ^z+ksZOwPmWqFs=CHP%AtK}Y`U4Y8CvW+b&
zD~(lHn9{_9T@S^iy4j5Rp0FCC7~!6J?|b>?dc)b-T?|6d*B`rzHIHy~bi`u+fVOQ3
ze#o5EHEwK>Oh`)6j<u5Sm19n1HLw=iwqtK^zd|-Kw_Gwt47s(zQ;DhWg|=EP={G&u
zx)ddORicAKMpvqj553f@mKxo15B1&zFI5<01Rn+nvN1Xz{cJxBRmDb67(D%EL%-?S
z3_ZhogAXMgLbkhVbTOh4wLcUa7qCTS^~bHXsS$$1r;H!%F6>-G2!f7G4E&q_gP+A(
z!{7YR|2Dz*L=V0f)zCSK@}IRWtK}&%3~20FjH!U3nW!cpurjJ$#o=hM&Jp~Q;0@hu
zA#@Tp+_`;%aUC(0I1#&5NAEq(8k(+S=+Ei<A*+;Wv(v9~NOu`_cdgZkF{M3IHTzz6
zt8B-qUf0VH1O0l#dM(NEo6Ux%aXDwLW(vVd9WE_xl0A7_iRh^wlQOilx>nUrA!c&?
z*GMZECqdWutgjryrO!oxMbLFHnb`6?KnxK!tE{AFhhm?sRLlFw)1B8tF)7!yXyZ)n
zcCo*I$j6@kHjeM!W_iA%Yg-nJ8BNn-Y$F|QL%`W4H*t)ygwhL92vls^E~5lZd{ss?
zGBTtqB-?-@TVW1K@+FU&0r5&gN^G>W?O`RAC`2B`=B4+7F{D)_S9>}_l(;EjQc@8>
z)dE#jV=#p1bDT;9`c2RB?3~+o?r?m3%v*2WVYxgPhBXYBD5^k2)YveYdu5sgimoqA
zeq27*$A}G5+o+k(u^2*s&T{3YsVWg4z&gkEYghQ_hd;=#|Hg0e`s=UKub0w$FG|_q
zn0oRxt>ye=C9UCwf(dDqWjPapV$-1!TZPmbYQ?VEIcAGF&bEmu7@BU*d~wKZwh+3r
zVrBCEnoSHG9dyt;Vhn^-ps&70M(lc@D^+BQ=nW+Aqx=LSq~Jqc+k9VJ8OKr7pcYG^
zw`fgba1m`v(Ku;eub~6hk?dL)D|r<ulY)#6b6GuQq6SF!v&kzcl~8@28?hL#Ycw^j
zV@5Tvsgr0lUq$svt;^%7)>>_nmTE1Gt<JN_>wqyi1mZ!ihYmro&fydWrb)Z`oU7fV
zTZ>>YIFdnCfp?6_2wM}A=&TA2Gf}Y{s+NTxtBEEtHvlb0U=cVzhQOs4USUE$*&75=
z4aq)bj35w$&Dq=K`I$Xb=VZ*(<Oi|oRFxtYvjs<oM_jpjmDAHx-nw&#H{N)I8#iun
z_wI4B5frJUL=%!5S*5L)h{oJHY}!dSgaFQALXU9`wq1xsZD({Xh|wl6Yg>%5yzc|=
z=jo?E#{cO*`OoNmWWIO6rted;Pq8I92j^#}$)0AjUBg5<;MR=0g=i1MNOF6}K!^jz
zIihV@Z#+%Y<E$Y>OEa6(BWW)H!S^her{Ef9vpJa5fHN`ksY17_Jgv!xim^ez5Kl%5
z#wt@42BEP->1%>46s?4$)~C;FPe_l{l#)D1DaSkv(!SmpQT1F>(QuZK_R}%oQU#9a
z5`!$I^+vR&j-<j+M3~noZP)pkBFO3QZM9J}CzCn3D!P?=mzGRpxYEj@b~K}&`Rs&x
za5(jcXbvh1sM19T+cb_CE2D2LOsS4|-ttgDa^F`mF2-P#Kz9U;<?m}(;mMS1gQC$p
zkKnx0pvem?y;@RRB+}uBlq9OC$``86e6*6x#-!lAF_ZF9=bk(vDF9FFWUW@`_<l*h
zIw$m-<ef^M$aJEz2gr>jpEZEX*hSVMs+}vQgX6_+VjM$Kr<<mg_ktIufx*Ve|Kxx4
zZ}8UbJN%=c`=@mCeX&0^L#~k&hZV#~->*_cbWwMrRj+ETff>p&B?Gas1R}1nbn_X;
zi1)K?9DToLh@Pf(5X89Z2aop=<1E7_(%F;^=sht?h0q!S?533z@Mug?hvK+JQ>GvN
zw@KENDZ-fK+!cgnX{^g}Pr-Zq(Bp?8**1HOi75>z5Qau72!stvO~HW2VXVV6j>b9Z
z#ABVr;u%v3=b9GA=WCa9phnr##tQXZ##)4u^)!JwF;ERJpq!`jypZWP$8Dr@cRyR<
z^m-^J)mlrN+@?}ce=M;q6i%K79ZZRBQUccZm3e9HDgAc{A$h|TWkq7#E;ErJlKh!m
zo#_7~N#*j?VGA`8%fd^n4@QYmj(U?utkPE@F0k`JNN`@v;Sz=v%$69E7)!-qD5%Y|
zS~-THVLmrpx_H2QpL&AR({oNvPB}Y2W4V;Hisf>}X1!w77q4ZWkF1<j#Y6)0R1MG)
z8pc?n57@TDwH<Nr^yjByylop?BQ?rrvpFAl|NA&TzRPPjZqiwZVW98V_>ENDlzg**
zOG#@n*`P)!rFPqs<tRmEkR}f6<%WK}VX;`?46J*Hoy{Z)omMI_<SXK=p=}zRjSMLV
zP7SvAa~!0vW1Yh`B?2!-h<z`%MN+w(giKA!Kb$7TiXg49K*bd5aah~nnvQO^M>k(E
zpU+v$<`ST3Q*2RK6Me*lkbXSa0Z9pKu&u*)`-FBuV+>^VDgQ68X$XZ0O~+Cj{%hv7
z%V56y{j`9L?^}+<Zt!77_sMUfW|;@Q9?rXGEBB`XYnrTr#2BgNYD_HHq=OrMNov#y
zkSD9H5`@t-0uLrnub5E!o@bIM85NO{lZYiyZFe=qrcnu;h|(B*EGQ*ESrJ#If@Z^I
z#iSsDv*S%qV-Y5%k-7$B9L-`6B3w8)0^>4tcTNyp2q_H1dxlNVdI+pH1Iy)-^Ye37
z%O%V6Gghk&tM!VL#NbZu-sSA<OpN{^q(%sVwh{Z7H4fJ~jDddG5N*q>YxwD(_%S~I
zu@Cd_{crw5&d!%a-_!S-3T>DiHEyjXM2U)4B&%z-?Xphc7L*of^30Bile*!)$Hah}
z&1kzBt#LG|pjz8D(pD=5`pr4570Db_)e2)f;`G4jAY3Chp+HG^9sEEHf|zX(VMUBN
zIy=OGiQ{D6XuzgSgJ6v?GtqVEx{mqY9*g}0TsLFg2bQaWu3gi(hWTR7Y}R6I#CazN
zy)jAPTB4m1j3HVWOw5tmeULDjU8<VAHZF;BFAZeM(5J|Amsv$@N!m=pmQ77eE82pT
zGOeT^Egx)*6CZE&-h*Bb$)rjSQ66X{H$JQ<7n5zq5s4O}N?o7^U^UmD)^pbPi=I4A
zUUUeWllz&%sEnzkWfadt%JPoHt&OTTro~Kb8Yce;9TvDO&#OvQq-T{4NJMWUrM-Hx
z$wg5eO^6UJ-mgV17(CL-oZQJp4ui!tmRZ7gUE2_k4sngtc6ZK6`T>!Cz2^M<oaJ)K
zYPFJsXbfL{?p2=o`YS{youQkqV>a8PYZ?p@=LGfLz5N!eyT_>}(jdlO*D~~?s@mu!
zde{o%758`Xrpo%rnuQoD1S%+148V=hdFbW2?F`$rv~4HG=(feCet+Kk#GqDKW_V}u
zeqe~6Xhho;8W|DJNLc6|I%`Gz8AC9JU>zY2nfdxua@rb`j-51+z!*X$Bz;0K7K3Rz
z+WDNLg98o^4w%h47V{a0`+FQ79dO~|As3Dg*xy@7*g+bRZz|O(=i{ZiTqYf^$_x^T
zTWF|U;$g1Qb1l^Md==WA(s^1#6~u}DG?+(oH5GWM#;nU56=fmom>{)k)ny0tY`z`x
zt*(bgQgs_<VE`33-{RV2+m0+$n!Phy3wJNZQ3x%9@zOhhn2FLEG2F&bKC99iNO?}O
zL~6|6o>z%&Qg4mKr_@Q2l_ag8aGtwPSe5v+GKvn9q9+?{ASQ#e*gOE!(7J|pmW=v|
z(d(Mp7*qXmmu(tQ0$7J(VC~oVp~nHU*^I?v4#{TLhsdw}_A}BvAvNrD)=Jrv^#-En
z&;87w;%m=7$1nfduhPsKI@`&?NCzPHYc3q_^X=dE5te7?oUTt<uU7Qyp1vO>er*`=
zKH~ezi9*p(n$&j68P8@Iu+A~0%AjqGDI7x9M3Yd5G(YbHZYJe~hrvOJw8<Vd0!;w%
zKn}n6IDk;<8$~4a!nr}vc2=RP7`K?bcPaNw)p%zeZQHQ7Sg^mp&wT%o#bVBU*0H~s
zadddV;lUvnjt<$|o6|JnRj$3Rb&5=(eUr-&k!iLvkz~7*qH!H_5dvk}6WU~3o0xWk
zQHE)&v0}=JQ4h1BGOB~vG_##f&xOgTw0?$u>K!Y3XYDw?X)F0M{jSvm(~NawL*ei`
z_$kC35uNfULzXtR!94;onp{76s%IM$^W*NtTt@o4a(zP~oD8RnWG-{u;eABZJL%{f
z!6)yb)?e4PY?@RGd2IX~2KjDKcpt@H63ux2Ax2iqC7WIX21QkLw8_n_dt{7ZSgv{V
z_FY17_+f)F4(EXw=vv3#{(=vG=mR|a?DL$Q^(+o9NIRzBMI}Tkc7EH_ALVC%_D|!7
zz~Fl#fuZ+oHk<sjUT;{fRxFoGPEJlZJw4^@^ql4Dob`Iea(T|s8{Rx#;%57p#<5;4
z6H=8q$X?5OY7nk9R!X?+&zR5mv841WCt(fQ778OTtkyAd&Sn)~ZD%0`RMtVU)<`tI
zG7-Jc;lTlm{R0jTFL2?)5%aF2o3+flhP}m{3y1q09qh4K%xRl8RVIj;mW&<e?#V$m
zIYg$Wt1|YO8NgUgE7Ydno6_|3+Ep8U3Jd}vm4KN(&urToN8cSIR{e{LR_pJ*tqrld
z9^SjR<L1-DZ<9wtHfFkLhMcH7(S!4EB4*-C4?c)@Ntqy`ueWVWmpY3~H(37mHBX$V
z0OQxJdMu2g)UB$+EvB0nx8|-yX>F?c9V0X`Cw!s=YgCmhs-%9PZ7t9cqZ}wr4`Q-T
z#K>$uXTDevW6aXY8Z$Z!)-}BR%-2~Do~~P96X7?uk-FxK8UOSD>0jpZrHlO4zxFp$
zKRqc>7gPVZh!kmi^2sOI_<=AeVzgp2>N*aRK&gem)QDSR&EP-zYkz|`@Ak~wIg2@1
z6Qt6D4Kz)gQ&5am#BGSfiuH2E-rgJ^QVdckaa_8FwfUeJZYz_5k>eecsvKjG2BkHs
zF^;{(f=gE~ap}?}4)zb2&u6r4$86TJm~|W+>~UdlCWy(kU?PocQpFo>PAh1Pq=M<$
z7%6KR<!Ibf9L;5F#GaZ*wAAu0fv<ZJ<Jhpiv-!hh4@sy}yYNiESDvx0f@&V**kog}
zFn!SL;k<iolNLjvr@AOhXVN^xlz5xrVn#oyy|*rUS|HiTTEdK(jkONbiV;yuS41S2
zFL;R*AFFfe`Bh6Dlc8~`$&*RJ2-W6;Q8AE$>?$MUgaC@VNJf-e0z(bWVgz+5j35<<
zi{zX{VTQT_F~{JD2TcMKLdtDahU^pbGS*5&vvrjmGx@!*J^P#lW8zge#yKGt*R(wK
z-lw>E^A<0@^dif%yF>=!-~k6ll1tmB<?$!p!)&%7wBU>ca7y5<crAe(!0S^Z0#a!Y
zSvP~R4c4v+aYJB`3L1E<we<Z!H!HP_WD8R&iF0h0JtudMIXc|uXMgr*@#|C0&dxbK
zJ7u{%=X`n2+1ZlSYR!5r<z@O|VCV-{%X17tXw_P=lX@@qI3xC|rfE1lIO4*^OI*Hu
zg~Nk=W{VlKu3<K7*_*ZOFFFpAde^p=);WkCgO#9E1GaIfD@ZgDVN!*0O6a57n?kPi
zO4f~K2gNZV(G%+F-cD1g?NrKyElP7H)3-4uXHR&a+S{iyCF-X9IPLl!lYL60cT(?S
zHdACn_=Isz5nWLq(|)?~`0sZ;oVsVgYOIe{6{svyq^K+f@)gyr)X@oLA&sJlu>??V
zo;Ni&c~BdS#L#Fb4tZE<r*0CW2@0u=NvfRceiIv3LO8ZUAhlGmR^%d@>lC82(lTR1
zd{#99+ceUGX@WrO`N_?Nphk{~bf3X{hSWiaC{0i63KFd;s;v|)iz$9dnQSB)%WJRP
zObNzIObCQwU_&5y&lA@l;ZOaEALZ};!=K~p&pn&hwY4oi8eG%RZ#L}BIxbzhAeAt!
z6`Y{a{755K5kj)xNf%FRl2S5=1i!(x9TzTM<K|m0OL|NQSl7_a=7iu`ZC1qWG4g~s
z5aK}7N^nSwfsa4^F%}IFC0-y9aw2aSd`@EDY<l{QXR}_jS+6)dJLUBBE+@ygIXyk&
z?#T(u<r3Sp>>nJme|Ukt{R0;J`yA}=bGX0Y;)Q)K9PP81w{*_Zq;AQ<M51vR(<BX|
zl}ZiHB|)fFFiSN9IS=X`hM+ZwqgJ>mr$P5n(0M^E(>mo!6C>uJh|wfk^CWrz33Hfu
zEu&Z<MN_FF?bEro#wNjA&QbE;r!%tc_^JhYVN}!niThs<-NCpvxaq*M0nipN&)7Cq
zt65bze3lrs(TxfORH{@gC0)=(GE%F@<b{^}Fk4s6oHQq-TD1C_8pOB8og}S#EsYWP
zbBMd!3r!i<^jUfOd2xXd@Pkht%M{j-B$%_3O0s?LLY(=&Hf3`q$@|(Hx4H4gn>5`V
z@7E9mP1n#^DLD6IKl;NAgXfoi`8SyF9S~!nkAZpH2?`lJJ_fE`y2#~Am!!pdLYhID
za!f@%)81slmn58`+9nfW2$7THV?b<MM1%JOv)LSL8=)m!o@4L<*L1j9%ijJzBm=v(
z4Sg(vL*z|E({{{e7S}i$*FZXyq$-&v_TX1Vphm#Rb4Bn;j-WA35_ka<CFV{I$I+xT
zm!=WqjiO^y8RCoDHa2URR3*cf1m%i56`NGzHQ7ucF<4|$9;;V{)(R2|dfzCflo(QI
zL(1pO3@t`M`H8dX?bRb5V$7c_#_?Thy{ePb@CF_4gQF;TSR~~vu4#}I(-akD^0^nZ
zRK+f-ssUprpjU>1(_9Y5HJ9aH0CoBa7+W~Fex~%xGiC9sn+Vq*r|<0g(GY`LE!6Y}
zMLISravR$U>Z_Jg)k8<ms)JW$J|;=xMrli+wgSd{08-((m|v6wrsEBPuATAnD{rvc
zcv=@|+D=^UaUi5Tx*z!d@5X!2ul(9?rpgi#EG$>6{G6fR@W{1m#C{<310+MVvEr%^
zCMnp)rH~)V&e}qd>9x&j#ogm$j0M*UL(<d=O=6xJgPrYYx;8bDa1tyzpJDt!jG}&u
z3%@o?NOxs|l$wb#5Pc*>OLTj})RG!TI*2COOhX9R*b|dFD4x+a8TK3bF)q=n?1`Ez
zG`KQtx9;MEk!ZY~DS}r`3W<JQlZ3b_-_dh0(pV;{mamYNl8bdUVj8Q6*)-t`k(pXE
zsf8d(4NW}Q7zyX7jV&sOGKa0t1Qk?#fa~FEoGW}REVgnzk|$h$K8CUWJXL+<z?UYU
zRDab6JgY#;sjb7<kQarJ7?6>$2W9%Eeoor37^k&xYPC)wYyG~CD?MB8m?fL4ztiuH
zm+syPoI><T-Am()Ng|_QI!oK_HPm$%M3cNp)(}Hv)^$Ah{5NR27Mu{OA7q`ZHSF)t
zIXXJvwHr5Btyhw2;w2f{xdx0C)!KVL{;`j;cd$PySf)wiAw`~x<gQ|Qs<~=LfiZ^t
zgO+R8uJO{1H!#M~HXR{K>PTcrd7_z&Ai^*N{N{|IX=u!X)3Y;<79E{&L@N%0rfJ26
zZmP9ITAYiD)nYrB$0lA_p{+m=WS$Qnh^<NJ+#t+Xf=>esJtp*!?2Q<WtumR(76BDA
ztxPsf=qIDI-6~j7mjZ?+GpVgm3kv!lowMo;<j)$4`C>9IZ5fBLw00amp$;ZNR&rm@
zuJE7j1`L{SIP~vOjq{=R480c)YpB#T@*whLjmb%!H8Yq(!)%1siRpK>7BVI^*91%~
zHY}47Y=X&&2#ig2(rdSVHArK~$&Pv4rLeq|OPRWBl`R#-rr#HIRUmeHb-d{zmO9o|
z0w&e?Rc*?7e&{?E5th<5WnM{AR|w7`Qn7IjgOTzu-}=^zVo$_iOoMHsbRTXSzVkc2
zo$vbWr}+>6cmGFj-?<G@BJo|@rmCQZ^=5?+p2y$wIK2<_L5$@E+DQ$2-3lFZHpWsp
zR-%-}an5q_(gj|4<rRj$W!Rk4bsd8rr0PMTbs=nH04qiAtcPIOY&`SDjOd3Pv90yw
zOSn)-45%=s5K4NIN`Mf9q~k<W(0{VF8EX>rDrJ3T&6OFtVz5w}HzpzN#JiIaHPw0W
zRf2rX^z8FLVoarwO`2Xt?OHvv)M2;U=1qk<h47~gNSsdkk1?t@Dfdu~{MJ@zsy@Go
z+QrP1#%79|XpO-cGosZY6|5YFp8g$c#E=WAo3!Xv2>cd0Dhrek&`#GSc_HHHKE+Ks
zwZmXkLKSpk(k9!fnn*pg;xQQ~OxgnZ^{|Da!j0@;=_-cGGFD?uvLPf@C`0)qWsD+>
z5=E>~+*Qm_22ib2`b^BowWtnxU6K@rfU%At7@DTzrI%mh#?3crni(NTkfqonBE-NS
z`@Zku`0fe6^Mx;FDTx&U;hHA9#+$}*_3BlyR!ZvBszOXCBWWL|DFMHFuuXO7q$gcu
zv6$01iI`u^7Mz`(C*!QR<Z~5FW0)@%m}Z8Z&uE-@%z_VevxX2#N=OJHS7C@T7Nn4B
zabz0E|C3E<u$U@{F{?IH#G3cW=6{Dc7Q#TTrfMk++f)aCbkA7AYC@EZ0}W1A(#!H7
zG5lKi3r!^2vYI3xvp}m2+9P1F199p6qAX~!Lm`3FB@fkRW+_<*+F*f<wDcJoY&-;Y
zYy>v%WZ%8Aa52@kDJqcJp^9vkZ<#V2cB?PbMXmMt+M|}GK!_tgqVL&C1&O(=Mp3!6
zwvRHJviZ|!-$)E)VQYeve;0{0g}A~%%RR&7`GT;TgiayX3ks<=dlFJ1DDy1Px;)FK
z!erzkfKA)*%@<y#4;Jehe80g3Pl!IXj57T3ANd@&@7&=FzxzdI-MmO}F9v*L9bMOP
zbhyvet5-2jynQ7{E1>bRdLV^UQQ2A~bi#WH2pxt2+jjJQ&w9PWbu)Uq0c+^GnH*MM
zJV4n<A0ka73E^jFXT165Eh6uedR72YP_lft*ke>JrxI#vu~3q!QJzuewUZt#5nIt@
z%qYkhlbQ{XO$(;_zN&=9R#lleYHh-pT>sq3u(5i!<O8Fz#YFgG8%ocMC0juobk<Sc
zHI0sez<5nCWEHeA<#R<qQuaI7fFsjJu4J34*@~?!nGzoGdbrA*U6*Q{Yn(@&N0ij2
zFt)qwY7V$-q4bKAg`G+$BS<taZoXG~FI3M4#$8-orMj7-BiDv>v|UiTX;x{CL?Y*8
zYE@-SF_x-&I=0K@8=?76jZvaC89_!LvP72lOcA5?hSHC*g=kbzn4sSwm+&<S&Udcm
z`RBix(2${N+Ejx{7|7>;^oM!R^~d->{Kx-ko=XT`sv;zW;Js83{qQHhjm3OUAAPA_
zVN5ng8<4oJ?4qthPAbW5jiGB>T;sU$`s+C7@ZPiDY=n^}NjnZSB#pn3P=w$)IXUI!
zmtVo~J%rF#Y6U4ah}5m6*hy+Z0+gWOY`(a6?r6QCabm~Gh(%Q%VyYIA4me_1Ho=bW
zTldP#s0dj<jSW)}VAC3?t!)IlRkN#@jJ8@Uwk)L6Pc`m(UOMlQtybSm!Kc0oc9!qx
z*@!t7Ow9nHij6ZyXwTASX02%VL9U0hNe$Y6O9rBl4#)-NX||cp7;%&Z$|xg_KVN@l
z3Xdalpd*EmL51Snm=LV1JPvsrrO2MO`uaE*F$M|pE4o1$kD|$PR#ifal+de&*$UZ<
z_pvTkmOPvh$+(|ZEi}m(9V&EL&(T_S^ot@H#BOnZddjz6_!f)#oEQRK*XH$XjO7FG
zdoNwr^7ZGxN!xYA2!jzXZa?(IAr+VF2fpnSpMVH9AI#)+vZX`}=}`D#GitBxs*04>
zYOC=`zv<CH(0;SQb<lQFq0`v1HYGYa(r;F@tz$NyaeBHWMmu3>1~q8;;3)x=MGMIL
zLGobfp`(}`YNRwp{Tf)gQ^*EEfG#CR7xk>NBc?&r^sD=<-%)xaKchHoLi7_bFsbXM
zbWokW8g<tyZLI>Q#ij_~F$f^juL^iAgOqB1A=N=c7(Y9YD%BB-Lr?!sHDb^xVq?PC
zST}*Kusg6T!{}s?Rq19@qrj=BVpKUZz2}_6x;7t(!UE*?Xah2_DMQ&#2~Y&6h$^H-
zSEOW(97h>T%%8Q!5Vb-<Qg*CsG85Cam=+702I1AJn|vsXMbze&YO$u07((pwm&+(a
z>^@@4uvQqz>ZIrRWR>h)fz9$1DH7+naP<-X$N%PE<Tro&clp&{|E+u=B90I({d&c0
z-r<5{KJ$F!WA7)%H6}_eCx}S}DZ#HfeaA>e1rrfunsjba`4nMUB2Afth3GxL@5L6D
zkY!ppVN@aFH-ULKWB=j>2GgawL_=O*V+84_Elk%t1l{|Bep)L*i{w{oT7oHaQi92A
zr7DtBmTkp@sts;qELj68{$lx2MwijQA}Ci0w<rQ9$2vh#Wvd^O3aSCmiq!L2&?#ox
zk#)_<U#V=dnm=lqbXlJ$Aw?jctwSnQ1$0gticEPxG_ndIj5{K|-SyD9do_1e9d~M+
zn_eUIB;QDjWV2^LC9WQ9h5Y%R&W&8yVyrEat{|=&g=|Zc7~`nD)zpq;Qq8O!=^kRP
z@vX#Qt!`N{t`oQCzqLt|XX<n7--$mJd&!6xEgGz$YGLY!A!}*?6Nx6`jN`?ZU*LST
zrrSHfIRn<>hjr3O*34$oLvg)2m+Avi(S=DtijKHi^W+nc^VlPg(r;FT;2Dx96t4lc
zHIYEm*y2(rsSshZ&L%?i*i<&gnT8kyQDN+8okEP9pD&rsEHMtO&QECOaB|nPSug2k
zE+LlWwap<2Weszgpc=TCaZ?0&ZM-jIQ$w;VRj7nK7ghfyOhCnOzG2$btgRVS5{IqQ
zkbHhT9>shyCDDUPLA_H!I~u3io{JjuLG6d+UV!;bRq0?hRlY6h&qeT#RMZgi^As&R
zH!dU~`zXyVtdo4t2fZGON%g&_?|nYR8d0i)*J~HVEboc;D1AT%P0TRa7(?SMIFlJb
zj3j6o^9Gc<IZ8NnDT?wvS-dz}+<G&&>}6G^w|19_DGR@#9dgnAJTKUKMo|sxQ>;@T
ztG&bYFiZr3s0lvLUybA{IrDri5CUyC=b5j69qSxT+tM_S^Ya1M&M-v&+@JZ=%x5$H
z?%)4;h%U9}3Yf-86uq%@-JITS_|zw#CZ^Y%D~DTp?2RFPjB=kaNcB-t_ui&ya(R*v
zf|L%PHF!VZI%%(^0;h3K+JLKC6r`bO)3#W<&#*b;<nC>vUvR!WXMgX~4)tQX)?;Ue
z(n+zr)Sfb>)iI+Jrj*t_!<r&?i^?C0>ZXiL@K0Jdd9O(oLNsk9D#{vDm8`R&l;LJP
zE^D{vT_;zUyD3xMR#P)5$o1J&ZG>RPf`7cODWpMbpc9gZTbXvXOtFl7T7vn0*Fz>4
zw>jI|m;&#n3~HD9R~9DZMhs96cX_|G6bhA!r#;F>YMjwh79mSA;xHPE>d5+UWgPcH
zeqq<xt;D=&d`VQoq3G(hDp}8WDsk$2*9=UNpmw*-wyFuE%I%gCQ@0bMi%4Lg?>%4p
z#`6RdSTFA~zi^TL{Y6>}!^b}IQ6linv)_=-jM6H*ah4DR`}>C}F?YjfKl^D}1J__u
zfjDC63{u9HjN{hO%sR;-v}!0FOJjq?X~h`WY}VMWG`a9eAd%NV)@1OW^=80t2KM)l
zXuAe)gf^yX^{#cEs@OzHb4s7wO%q0=-sLkR`;hXflxab9>3*Rg;mX0H=85u*irHi&
zY{vAW)iN@JA=m#mlfAFkK>Ih2rc$p9aqBf*uMFoCLeF`ZS<4waA@tnFJP$nsqdQkl
zG)LQZJm~e%HYr6kYV@8S%skONw4%F+oj{H3sOc4xJQ{hLM&P1rs$4+1@n}3Tq@+jG
z+bH1$AsZEwhb(!~gh<SGT`DDB#!NOFY$K#fwxYh&HPiF3gaEagqu9|@#T}!jDO1;4
zM>Vlc5o(O$`5}GZRR&>_t{+2`_Ej-*wm#+dol_S3`^0|3*~wj8H)l3q@X-%_ke~j^
zALSqX<A1_)-}pM4VTDChyryj#g5+?WpPcf@BUgFyiO1P&){@~LW2zlr0#L{EaA1->
zD4~cv=Zb2V%QdGbr;`3-q`_*_wlvPtILESI5|eSYZ9BSdPSdr__Vx(F9^Q+Z)z~(h
z2X-L@WdvDZq&*b%GnMhlj5eh3jgqgaVVyesY>$cwO^Ku^X6=;2XGV{g?lRg*CMnff
z(A2b0|JF($^%#vcG`1;gnSw->-XofvG7V8lpaM!jaVjUI&I?h`V=AXC=|lP7q3Xp~
zI~x+yu1$M#WKCA_GaW~5wH5L2l;M0BHhAA=Rj(dEZ2&HRr*wieaFf&|w1H%rm9^5B
zPf(go3`-ZKV9$|yRV}_XNT(8GF{T-PrmzTdJWqD@*ZEDx<PBD|RR}v~+loq1RkLDh
zCq)`*ZFNh6F|+Aum)e?1hBzdcZAS;c{@dY3YD9e>IXtQOM;tg`uUKyYHwVC<uZaD?
zYypoydKK3=o_p?jHtRJ_+hStGHLWz_46;z`vopT?^FKt_wZxUU{)y@)kjW|Xn-+-6
zbWurnJ(AX1?jE1;=Ivu}ZE`|D3_Ub+aY(oh+s@MdO1&!U45pj04gnJbjp^CzQWs$7
za$^uh5kiXz0Zb8nMNC?Iz7#wZudt{;V^3>lf~E+dT)&>ogR&P?zY~l}*$_$SR>pl@
zD;;}%X&$7QNFpYrIJ*dvoSZz8hLIy{7pl)p#uh}FICi3I$;lOyfe~+AGM$VnbWGO6
z7=u?$BHIQl##PaShNVoKxZLl0=wQxK)y)yquY1^)4YEd)bZtgnDtSsgi!7kucx4kg
zd06kSAX^u!4o#RIj(SYhAhW6#RD$U|C?WnMbs)p^a=6q+rAG5nLBf2u!ld%}P5IE2
z355`H9UoIQptzU4Zx<rVDgqJ?wqCCxS{m2jX=uAyE*bon|I(l1^z@AX>hJwyx=uQn
z5Dl)K0gkQ_1hJTReEL(L<n;6;wHq(ByrydDu1+NR@ASS+O)A>ij4yrZE5w*=iNhfE
z#(~fT=~6vEpj+$#(z9^A*$C3JqDo0cPDwY4Ws4K?s8y!Q8#h@qSx->zq&64>xyD!R
ztVvIq0{Xn))|Su^SM9YP-k9#;FqkS-XG(WtMo?(6?nop^D05X|FcYCQwbNx<*74L&
zNt4oZJT;E*VtVzn)FbJP8O1=|OM7r`!oluyJ)}))#%wlEn-sI^2W&}FH6j|+`kxp=
zrFxa{0*|vURaP{_=yUg$T{a&5p4Pywb@O^Lb|qh8DuUDRDM3+fO3unxj|~i=&yt{~
z)XcEz4W(jOTPg@*14_nW5oRH&3fr~0?@hqvGf+G$UDL2!t_U&E7)RUA30}M_?|tez
zk3aSZzy6zFz`BO{VviUNgZEh1$#<ON?Cu?&dg?mYu3f?qq##^7!|srdyT(<AT3|>?
zzm||%$v^YVv-HD&iJk_JZ99f&u*1Om^e)yI{PIj_s}(g=khyeSmb}n`v-2hMMX5@e
zZ7RCi$(D1keHa~f9ec%KLL9PIU`%TMS)pZ&!PPNSs$t}@6jF&XRt;2lkIQPQ6f+UN
zDGWt8XeA8rNgz&aQ$;^j{R``;=u~E)0hiOWGOfQpW$x2)ha?=RUaP;G`N4#tKFIZl
zNQ=os7!zWeSZyBwz-n4Us!kYdyH<k8DammdeB$y|jUAn=>gQYI*_Mv=xC*>C8OhbL
zOuA@UjG`#4816#~RE$-Ojm$5W^8h)0z$R%jYVBko=6XY8Bd&Z`Z?cV?wweF79jI1W
zo5pZ{ekN5e_oQOya=FA>%Lm{06i>Y8I{)e4_}dIaz_~6FyJZNTZq_pE95>I7`Tp<w
zF&2BXRGwE3yfT?xD(k%nSy~w-l`1W=^a%_^@7ee@Hh5;8q21fVwF_)B!?yFZ{*shB
zTO1Ihr|TM)-m_Y6xc$~0u3b5z@B6Lij*w_I-m&Ma?0rTPhSY#zm$XkpC81+MT`A}w
z37lG$tm)5D%p_A3Ts}{xQ+>f2n`54$jZ(T9G8W47$SNeccw_zJGn3d3xZG|EQTDVx
zFVj7ruDPN1vQLm)T0h~zJK*R&u7`}1(!EwyZydM?GTFB6iqY8!frb!MoJMUMP#szq
zr{vJ(qp$trD+E<P?|yJJ#+WxaZ-BL8fM!exY7EYy5LVI_tkdXpc|D{&q$#sd`$D!=
z#eiqBQF(JK{X`Fo!t8_uG#a{AX>D#!P!W?D6GT|X2v$-MUV8Zkv>kqZiuHmD+os|1
z#~x#~T621~Ou9IPja0X2_hxi+%j*1uy{_fEzT?w5LOwSOsDd|ZMmtsUM(djpM>JCB
z$oj~^(IMS@fhQ0*V!K?e&*9W^?XgSDkFMal1pr+;$3_Qzkc<M`(R8pr@!Y)mCLjIC
zd$S<XwbA-W1>q?2iNYL1OvT?yz^>LvkIJ|hqj6<6wLR9PJ)fdxlj_@R2W(j%p8%0E
z9Fvis#w|%!NfgyEsu8?GIich-N+4}HlQQ2D=cMh#OrjC#c2njhXinCEq&8<8Tw>f>
zTWhL<%75CmCatjw*eto#A(fbDB890Se!J@-kyKK3D6kO(xil&5L36jVEg<@QI3*HV
zYVXv9ZBw9T-773l!j$K!gi%ixaRNDGG1Tho!z3iZTB()23vAYgbTLHQEws3$ekEw0
zH;3K!0r^M$taQJu`<J*N{iq_gT;sdyppwT@OKoe21a94Yi(megU+3V$Rl@R&<@qTQ
z&wPH!U;2wb!wcVfiGTJl{w3<(-V6gYj>Tfmu)NEgufNJq{P^d&cJ*p5(>%3psESZa
z&HC?MS1vK)EZ47HW!S87vpMa2pUvR07&hxQH*dVcqt4+>G9ot)H}7P)L(gV8u->d#
zFVFbq3oq~k-~Sz>HeY&ZM@VSISc3|5@rYLIIi0H_)!f*^6vvCsxvtq*Dg+=xhL|-4
z^%`b0Zwi3d;8*Q#j1jD1Rp?dZQ)sKE7Nzgf<iS>rZbOjjrZw$o_I(IZ+CWAz%*(?}
zraz4`&+DlHrg;FO!wSu&^H4GR(ctq74}Lvd2to|#mP-5ZO3jGHMx+EH0hjruk^np5
z&c=wrTt8U%(91}}NEHB$ZUsv;rIbhQg@8%&akWuEjv$BKkX2vvUeJ(^L+u3FaBqr{
zus*cA4r#0;s!<PS&8SVar?Db}ltL6Ec)xv!qA6oGM#2bw<=20QH;+%bcHvR{Y=POF
z6Na8|d-@|B?C<d}e({$P)lF)sN9#B^I-p;j@%jtj<omwsGyH|W__G*Ol2YquOjWyk
z9qv5uiU^E}ct3FY(gl3KCK_poFiYK>joGkXFS+sRH(8$Ek$d<+>kOOqlGWWiEY~YZ
zeqOKm>Q|pZ&R%)8)I1ySZ4RtWu9;#2n4qG`kB&bF^Jer~eP?U8$sr{wP!qM4O*UdH
zk{k(ro`OM><IB<_j^{FBO5>nSO;d`D)5?_gM`c^A*peDWY%xmeRc4AN(^x%UJ&z-`
zPgUK-R54m^TVD+2EKk72+g%Tbq@qDvtE;rC<5LQQsA4Zv`xLe;wi?F78)D@XOF2Z&
zsRpZ3AZF<V6|<_)RZTAX%*<c{sUvGzL)5xJhM)Ce+hrHh_!1@BEIIU<JJin6@}=kb
zFs>FMYf!3t=Jm{~rZBCkS_ML6wYPJ%9{8nS`(0@FFa-AY_Oa2kS)cQL-}BvExNyXO
z`wPEBNPA>rWYM+IuekNvi~MVU{?GES{`Y>GA$o?SL5*y4)q2!(g<{v(Irk`bK5OIv
zU%7S-*S3;~>pkoBirH*IGn-=?N55HddU_)F@;#f=6IRPp`hHFDJ#E`y8^etouk+9T
z`7iKOKlyor?;%=fNp+dFYwVBKuC|>r7Hsu*>iOxK>-$-osEfqq^P%*l`)?>*id@Y@
znUm6Hj8Qzu({O|e*(=RX%dhC~7-Oylrmq#0C-Vv^6u`t%ZPQSJiKHpiJqgRK##d&X
zRbBHE_Cc=F32asCeY@)+n}arJ%@{Qt?kdBgkUFMh(=<iYKIQ9Ks*04DNeC*n5`meb
z_vXtO8f!D`93dA|m756L62B(Y{9a~SDT3R|;mw<>pEZ-i7t4KjH=WSo=3Q1%IU28_
zl;+41Jl{UeaboiTa;%A{*YfvsUw@G|-a5g!meul%SvR9;=Ulvak)QsFALDnw^i`hy
z<|}N5NH^<nMEvFqzdq*Q{<r>zeDC*uC#&@mRI*OiY8pdU&qGxWH!-w|i6H3_5r)2J
zad1F8Ur1b5LgAZ!P1DVo&G%^9mbMea`Pumi=jW$vR%Z-}w!H5-KR+W3J^$>V{{lbx
z6Q7qg2&yV+$U{wM<!{~7jHt@HTY{$k_ZaP#-&$fu)gvv#qW`u#4y9Jzlx-}`F4`hO
z^XaM4T__HSsr_tJAtBSN*i@Zs(5eL>Srd&Z+D_8)D!@>lI~CexDJzd^=E3r}7U#<&
z&1-gVlfbU)4-s*$d1-aX!88s+CTX23^$-&=)CW&DLK9F+9$LK^#Q-U2#wMLN^QinI
zp#WpM$Sw~}%|x<YiSfZA3cMdu@i!T@*fONUt$bJQ6<Nwo;!Ae@Xl>|vJO%PsNNN&1
zx5XIbiG*k|aRA%!#V<ccv{q_w$GMbgwT{K!K2JVzoqzU=zsC8hr*V$1HTcar4J-bg
zfA`<v>XnO}pPtjYhJk*>gmp}HQh7ZtlWZ&jU}p>MqPiFii^ZIy{R=F65jKtWI5zk`
z5*x?S<;%2-LjX3b6=%n9Vd8-I8`i5eo4%)=ceL|4C+8b(y#5x~A3em|)J#F0AW`{N
zbrO^Mr70^?!9t4Z*@+-h`c)d3Y7O_g#FJ(vL&m2dSThODs1R9Ugj(8KK!9ijlCYf-
z5R6O@Y7<o^gGt`Wgy`fwY3}NSY97_z<>yNHPkLllH#KrRGZ`IQt>zm&Q<-S809L?|
zQ_d#Zz}sC9#iaBw)_Sn1kc(cL37#i_M9J1VY)-_iF0t2?zxDfKWXvpHRKP@t%G6@g
z?^GQu>QVk@TC^-DvTB+BRmPfvEA_@|478@y+A&BEGPT1^5nD}o*<O?wnW@~+CHFAq
zXV>@0>#~D+6`~lh1_p0<?wc=T=SQs0?-H=|g9iftv%mc3vBvVB|5yJI*MJR?*soZf
zopN^PCcpoM-{W&X_yhF5XX7_C$vaga$QX+|x@K0!BsidqKe9E4dW}K|?Cs5X^wF!l
z_}Xm-BHe6`r^R;QOhfDDxa9rq*LOKTzXj3aY(qP1neFdkntg1$k83;r`TzGf_=|t;
zC-Hts7$Ocs7zWfK5uHuV9i--2nOE{^nUK6@Rp^K@Sw)}DWlFbtjkO9zF|pF=G+hYL
zGS;CEm8lLs%G#4{%M|1kyqZ#me4c6pzmRpwxQIKdbd-5jOkC;G2QdOi(V#|itC<no
z!gGo;^I5Le@g0jeuMcrSPmzrFa=*JIMUZl=8M^8|C6;dzdeM@nHJX^BGYw;?CtQqO
zB@IWpzJ7LUM=8<oV}oO4timMEJJwhlHDac8gcwQlDpBb5`V`ebKEGwBs^G3QCMDW#
z+nZ{pP;YpxUdAXcWDmS{>kfC%d**xl%sWTd3gU}?U~eudCg*pLVY4C*YldN<->iwo
z@b~}zKjc%N`Xm>x95M9$NTAeZb85ylQiEcw>2_<(LX@RGH!*PK>J`56%3HWuC+RV<
zrE3EX8~W8-te3|Ctk+ALZjXcgYq;5rv*iiKIra}OapC9+t&2SK%(MK??>xr$e$ThF
zTJ~a6Xh~Y8Fd*w(DvDT`NJcBF^BZ!mdVLn}RncPCzsqaMo1K`FsqU|W&N&B;te$}v
z^q0NkAPq<)SvwozgK7>c042RhN+gTAtI5+*Lw`c$FsSHsJboQZ72;w_OqkB2p0)%f
zSL4a1ohePH$wvh;4|Y8?lNxhPl>n-Lk5NsGDX)&g+2#F^kxf~gR022Ovur>y!o{3_
z7h@#mVtG|gP^z#nB_2gFyO@E_t^RFmi6m1^F;c23xU^a0IzGCag}~TZ`+b^HQa`)?
zOg(1(%qSfwjiGH?UVi-r`XS)MDfk|OgZL<qUHG$n=2IW%pZ&t`^7`v<uwI=4fng&k
za5mrP?(tjvqo4aZ{?(uQaUQ+?80YI=;>30kay|Z7Havxe#N0En765wUeT*3i@9!UA
zn})Vq5Q3qzkvN?3*6r6>pZ_J6r#}IJ@Bclz?(Z-=_!52aT)KLl%NLKh{>TOT<uPCV
z(jTx{1^)K`^?6$JZ}5Q+eUPIINA!com{f=`RO=hdo?C0ADr#2qDyBQRru%cR&y?qs
z`>9l|8Mhssois2N-Zn%3T%TXNuj*Nv3W!K?2GqzbBcWz$k1h0F&V}aI#Z1Rj`&>QO
z9Ozjm6KBpo{+>z2`sJP_{p;<nhwNaSDp#g|@A8IZHA{xasP&Pqd!DpTKt7voOL04Q
zXjGykZIo@=d|0d@SIV3o=;^}NziX*s6pB+eO}m!J=R7f&9u{=r+IDomq3;VyvXcC{
zZAc!YyqFHIEMFlGKMeW40^=v8lw(1t7<ldGEsS#<%p3ZR0S<WKi+`0TzV9-RJ!1H+
z-#Fs-odcYE6H8>>_Z%J^;)kB|)q>yq;+Oc^7k`f*`jH>z$A0q1={K8f^RO5RPy|U~
zst{yGc3TG9F{&ILP$YH7U@*}`f6Cp{w-{C{2#?dR{x$sYGyuN;K9;9}N1l9<3p2~S
z4ZQEsL%!qb_w(k>8=T*IooAkX2_FN0{jdKGu3o#s5C7l~@&iBc{Sq+hA(#OZ8w|;=
zN`o_j!TZENLkhqwv~7(3wiGME)C(LkTCV9xAxAbZD3Vf}Q);^w)47NVqQ?+1l&F3Q
z$P~+%27Ky$b6-2k;jfvEG^D5;Pa`H<P>%u)`G}Qf64l;;+V)CZoHBu63Js{1l7*I^
z1jC|ZC4-3vxgI{^e3T76NkUBASxfIq-61myc9V{>RZ0Tnc1H?x>*@#jS;pib(7b?|
z8D`#{B8Us5H%VnHa~M|^NUN9i{pia^jDkB=@@RSAXb3*whnV+1M(NNpML>OGT(wlH
z?My(4N+#w2W2s(fMkuhXW6TUh66!3+Cuc-!=oTH_V#bYc{8@(n692(}V7PGu{@VY!
z<7D~gIlB74ra@9%HZC%o??DJW{qYYox1ML7`3eUY4*32b_#AydfDv^v^nFPeisQ`q
z`bUEXi!>)6VEuf+uw0%p^ee0#SS|0;%@!;UF0)yFinF^P=c`|Zzw>wC#tm4Weww|-
zMds!P*Do%3?8x%e<B#&&zxi9-xN(biw&3*Sn0~dwWB7mk-G9KXn>YFVkN<HF4i7MH
zMq?UcToXc%ZNwa5ot1EsAV@DWS0O9>;ITF`^tF2LC3uulgDR^yna=8ZmYE#oT&Px%
zVv>?-kcEMS+ze?vHs`gD5K={7D&R(D1#H@@I@T-EeVx3Y)W@trq!6PTaA-cM!V_Ja
z0!CB_8H_1mN4flUx@<xzQt@INh2HCW=-fRhHYwFjRpR8pMoLt!7Dg+F(q=;6u9~16
zy@JX5%w#jrDx9M36|K>v^cO`&kUkS48nvDmYZ3q_DLbk1=d=b+^|whXtlTeP5m$39
zJp&Vjr0{%l>GQR?pW1z<6$X@fD^r?6Zhg+m!`-tLM@L5lzn1n}M%22$@>eWf2e)p)
z{NR*+vu54*SU=FOR~%e&v^EmfcQA9q)hh>_+_{4@V7nHKV~7zOBKRKTEQ24i<e%=P
zsTk5o$bbak8bjL}KKDJJ;n}ZziRc$R_V`n@-2$_Bz<Lu|pMQ~`|M^ez^FJ?hy#C}r
zf#o-O;_(@eU7EAkwtV@^zt2ZM@?P3m%P;)$Z__lEwq0=L@?~Cq<yC(9SAUiFz4tmM
zuIcs`9PS^oS*=+Q8`@^Ze7?t}%U78#_L<KQG77Bs4xmpqI%5ojm+s0#ny<s;vNftc
zy3}BE%jOdesYw->?V@xsMxxi|Ac>4qjib7T)0VknVmeX!QbW3a-sbvd<MgyzXor-^
zQQt>x$0mPko55I2ir$Y!DXPGdo<-#q(}d>xT@M*2m8vIF7Rpuujwl+5lxhc+vAd*E
zj7#k5VkCA$iRCEIacSoaC2&lQv2ug-`3m-e1P<2qwd;H4vJ{Eo8A?#z^vVY?S6Y;3
z`h?dr>Z;IE8P;|w;H<vtI2)HKGa6y?VHh%s((l(@SkxxfHjXcR@vFRfe9HdeMOK>?
z!?2+{{JV6E&-2CKe-A{1ZQkPY^?yLyTw}97Pf|NV3|zl*z@<IQ`S}T#E?i>X8vfDW
z|1Wvu(Z_i7vFjWhUEt{GBCVUTTCa$Za~11-oUYllK9Y`ob#{kWzWEit?E}}id%ot?
zH(uZ^Z@KW;duZneT)g^UVB?=)dHTIP^61ZT{qoQ8p8a!H=NnFD3ql|0H|InGuf6<6
zGWJJiZNne`q0jMO|C4{p8#izA!na=LlOKOSuf6^XZw#+6pIzqQ!Zp77?6W*_vE$i4
z_<h>h9)IGeeu`i_0%Pe9&{U2MMNJY-9n+pftc^KFD_E&GP^(cgd9pbps!F$2`ZrbL
z%w~|H6@;Mnr(FS@MVqSyn|x10hwR9v6+*zLdnx^@2V3uLjLCv{tm<x?dqN19P|mK>
zZ9e1Gc;E4EF{=O8b_XdCMy}Uf3d@P*zl13i7zvDi%b&VAV{cm>L#4)bXG;#AGR|C>
zE_1E%9HtDaW+<b3m@%=&7_lznHb{w>HtFjoMzl&|{T=y?89i4~QQkajb2<W$BYv%g
z=brmoa@p^3>CzPr4i0GBmSK<<>e3!O@TISOjds3AV=c4UjH3&CxJN8czV|=kmFGUk
z`I%vH@xR2yQy3ic))8Vy>^EGS!=smG4C{07hDFoy;`3kS&fRmq`RsFCxOkDKY3SyA
zeCPLkjz^z-k0cBGFlxQ6w6N={8JV?~v-K%&+`P%1JGZe!-hXY6mG_*Tz0S$afcMYw
z$Rq!f#dXVmd(Pq9(XaQ|YdhMuWj$<It_OVYIlg^N)3gu+!)C+FFMOR3zV~s!@h^Yr
zH&~ya^Z51a3~|LrKJjV3|A&8!FMs)W`4>O`4>>%%#Gm=IKf@!}ALWIYUSqu}cEH+{
zVXdfz%ky*Iy!i&-dhtbOUB{37=#Ox|+~hN%$k;`Pl=F|dL4>Z1#2<-ZoC+P21e!Ky
zh`HtZ&Sj-0^_u2tL9O;R{h%7hC2(1UpqA&>%#QjVgW5*3y2*%Pfrjc<4wDGeh8lT%
z#^B#6buUB@+RH52XAD3Px2dm6_huC(s>c98UD1IU%t$RN`nJoTtuOt|*gjLbkSN3)
zF`b$r#gJ<OSqz?-x~i07ufklnRl|J0tltMD=q)A#D5&nML8mpt43%V(MC_^!RBDrD
z{J=m2(_uqk=+F4d?|+%!`-5i~q9cTeZ5#*t3m&<4g(shQ57!@igxO-B)1`;?DF-&P
zS*`J#p25$U?Jqbyv>blh|A&*icUiA3L%*QkG@PAy=uf!Vd9Lp7(K4{v1ojV(I6FP%
z<nB2Z2}(RUKIU-$i1&Z+2^I$nw2^a&o*1Puhnk3uF+|(QzhXHcDmVfr<?A)fXEPpu
z{BcfC&RCtDv01I5AGpwo^`f!Iw2fnbZ^3+jkNv%U4)+!uzxgJ62M1W!u<Cm@LuB60
zxNzin>E$<h<{K~asZV{9XTSaePd@$#n+R8~J;vvM^2caq9p}fd5|=0Z<WK)AeDC*v
zH?u`6_N;!zrVm6T3GZPTaLtSxue`)7ue`#mx8CBl8#g(>d(8WudYu3MpZ_x$dSV;^
z(J(LxnR^d|s_%_77cok+LWk3IlY*F1jFPDYx4s)8SOndO9i^BBLRHK$*#VQqDltGs
zgJktFYd0}{4k+bhH0U!m#SNmWODSZZKI@XvKZRQ)RZi|DC_VS;!{&2RS8RQ|>md#2
zwO$;k>dlK3iLv6Q6lhy>QZzxhMq=gOxu$uA)W#|y`EM<ZC${ENegnn<rdgsGDHmcQ
zIFmy5--e-2)s^vrh=}t+Co9sVt`1<0w3K!ga+2|+xm)q3gdoT%1~D+s7jrh=)9f9x
zJY8b^lDIl0y35>n-D6G1+1+z)-MPghk3Gs!yT+g0=4f$5bF?q9SEi>mXLOw-%p)z4
z)v#vfHe5ak*x0a`&)D0Wae8`AjE)1>^87cxLDRPE%?^mpb9mt*AO6@!dEbXVMB^IP
z>lIDYF?eZ>jyE{~(ipH(OcwRJO7kBfY+_)u95_3<%VxD?7&aVUIONKeOMv0@^n}fN
zO$dS6tYbc(Gn>s=EM~0NYwq4Xrg08uw8Qeia<gWyn{oZ>5&d$*tTVj+>dP!T%MX43
zck|(oKh3ZF%CGY~zxM0QyOvvT-r`UGD}Rc|-}@+=^Ap~9<&<x|_&RUhImIWt(aoDT
zx$*i9_Ag$biGeq7-Qf83m$}sK@hAV}ALp~5ejkHB=hhwYwk6UF(y*!`N&PfyrQMby
z-jN+PeO|o4<kBiyjyTh7OkOyRrjI78ZECPrJ8gYVYE*Ab6;4x~j~eN0YeyP$WoX+r
z-y@n*@KS}btxEWm`cvb!bjolQnLg<Ca5gEUiNkT{1Pe0~)knCk9Zfz)Z786&9o^Ve
z3#a%;#_r1rLIR}4cPc><VoIpYwW-pARhuI)G^U<tf}mnCy^rQv+aUD$(|JhXT$*ms
zGQcq-CPjXQXcFP%N)BaS1BU43aQFR&8{hf{pZUO5KKtQEc=grS`0}@Ia_cq2)hmzl
z*+2F}eEFHLaCddaBaeT8ritJOI6ua2mhA5zGTYl@zSzgufcI-=oY319ezqdC9&Zfm
zGtc4vA*ZM3xMt3K-~R#5&(Hb9CqK#F-X7LjHW>Ei`}o20{MWw5d*A<IE?+rdvl%2e
z&JP41*m$4v?=~z~YtBy3xO?}6^RqM7t0n8rn&ooEy6^Erz(mjGgO>Gb#mU(z4#VEw
zK3A_?VSjIr)oRI|+qcC55F0k@H3m33I-+eF+NQ+}1FO}F)q2Hy-*b&8AHB>Ik3E6i
z+vBPCe~fw8@%|?s;nLMB+&w?x>cvA`jC|%ZKf<-Em+^0%@#6E(^P9i>MQ+?Z=iu7o
zT)uLZ+i%`vd3w%=Km1W%c<DL7bM*ke`v;!p`lUloH)lNm!VTJsj}u%=V@k5Q)-SWh
z5QCT2T$to6nndXXc%%ipzNY6PIccKTbg<-%swg#`d0}vsV9e*$7PK^ke#FFTCr#-P
z4K<)B_nE{rRhmyZ6A9Vn@g>{O=)NVR`d(&%UDqF?@1D9w%fq*I&WN!Qm!e-~Tm{mK
z0k5DVm(mnO8mQ!8l%$E-W+Fb87^a9wCTGW)_!<`zf-jw9W9hbn&#}@w002D$Nkl<Z
zNq)M?!tGkb;%zIor8-QZX_{n**tTit_ciXO*ar&ov_@dCbuHJP`UuB2U&9^0%txQR
z!goFWUT)mE%kwYa<i<C?z~#je-TWH+2m1_RjdKlH$N6TBw~>P`aDIADH(P+Qcw@Qz
z$aPNdo^bQ#O?tlqFVFh+&wQ4H3l~@}&$)d0BJ<gd&3c269^VJP{LHhQ-Z|mX>yNSC
z7>4kysOv<0zh)SE2Jh+np1}|JXz4f7YCWWCtHBuhO%E8_cE)1XVBA1IcqwWZJnPkp
z7hn7qhx-RS@t()I{@A0u^x_M=a`Pr#+i`e!h&|fF*#+xWkB@=%`56v)>hWt_JUry+
z+IxB5$3KFx`y3q}(y`|3^e!L!@CW$Vhd#)#UNPU>=jIzP^SfX80yplgSepeOdirDd
zVadsxuduoKD*FyLufNGBuPnGS8(`CMW9j+Q&7O;oKFPssAHQ4@nkCV-f}D&^^@Jc8
zPhlsizEe|m?v0G>wKdJvP9$TXX&o-bsAbd@0z>d6Jt-FES4;_EOW01)*3|x5(|tz%
zro_P|(3Yo`=|2hl(b=ociy|xGs%Cne?|VHQl5)1HQ9iXLnXD75y+#J+sznkxpbZC#
zAw6VcTlRK2oAv7uV`||YM~m*G$7FrmRDKXu`JyyCtQo1%@tdNnn#$ll*-ZVrHn!Ir
zsyzYK_EI;7tUzJ<RR!;jRO6U5M<G^8oAT^kdW`knMV`C;I<uQMxO`#Gr$788i>~F?
zoikp&^%5^Xcb9b-=o;w`8-sL}F+Cg{ENGh<XUkKf@40#7jCnU><{U@e1>XDdC%ACw
zBBrrmJ#E);c6!EYwc^d2x43!hnA@8H*Bx;5*pq~=<@W76!t{|t7jk-r!P^$obhIIY
zvAEzF`T;+9j34M4N8{Q=D{$%BRoc`uZ{{q`tl{YJ02ATp!U0DI2R!xEQ&{UbzH`j^
z`59-cQwBdIJ6J>4bzHx8mCKhdVXR~C!X^5C&DqIa?!I}0x%E8x<YPQ?<tqN{F0Z`&
zBER>=FZ0DOJ;P=l`S#Cz2c5q||LRSgHGJU0K35*Sg7a&Zr>DI7)|>pxSMKuCX2H{+
z{Q-`;1&4<R1Sch~+u-q04Cb3*gC7Q|SC;KRj;3wKU3F}7j8wJJx5E)kMo>PK^WhVe
ziqtK|3WCwNrz!2KiluZn?GKQ(6TMH2CFW3#ifT}&7~~tLDQ4TTe0fBbom!bx9lQ8;
z*TdPQ9N4CW!Rvr(G$Amu$p*UoT|+Vi#)=h-N@!_PtO?nxq47pV8jLx#F~XFT7=YY)
zrZ6&<3^q%mfF#%IRrLK-r>bH=IYgl{@I%_1)R7mR3ZCG_V43Zf=^v%xsK$|nywSRO
zY8ow_nq_vgZpPl@AHsXjt-Hs(^wMk0h8tYii@f*o3%vJ<tK7PM!mF1r^Vs7b;Efxv
zu-U9>ti=b<=In(2e2uY&u4{PmiTB{1G92x-wA~D><Mo@j*mzhj2j04KOu*u!VSeQ?
zp6oiQqZLgu_%_&pI8cD-H&`Q$JD2C@#2D$Cj<$1{re!dWjTwjuMfeb~U4sjON8a-!
zAN%AdIND!uVSkVL-USwm8I3apKj3WS^z@k1(^D>7y})|8B=nwuVYS|{JUi#fCqF>b
zIQ9<~I14AoukhN97dc&?)5o4`moGA$-QY&s5JOMfz=y6c_{@ht$NqeQGlAA!A`WYg
zPtUmX=9~Q5mtWzzS8s#e;}hTYo&3Ll`{#J^WMKdOALb*Mt}^%qjgfSk)cs44c-u$<
za}egFjvOhytcK)B7?Lo;ZwWmWd^y7)sZnZ^nldwE%!pydM0|>@RJiIn75Y=#u;x;#
z!Q2Wu*2cVAd@@eUgv&`XHe#)#aUHI0Mrdus{EXOMZKV>{D(2f=4;?3^OrTzr+V(P)
z4B3cgAcs;%h*5Iitm{TWUm*-)h(io_!Bi<FQ6pK)bH%Wy3pakxTAREU`Q7X?pCTo-
z7t{?eYo?@g92Fje?8&GlkxHEHOmC4>wX7+UPL1kQ2ZcyI?>)1_1$&o1%FWkra{Kr;
zzx$0>xbdxLdEfh=;*s}#i1&Z&307wx=H&Q{7zSd$X6QFG`*XV4KBp&Vbk^hIhUM`w
zUw`gf+&)>c3Wlbg)6M4W?O!28krp<3qDgiT8|h{(zV8Vk;!K*!Oe!tJ5V^RwpPGaA
zXlardptW<1j|`ihAq1(25e;Aez2D-~-}zlU_JQ|v;o?OO7A?cj6Jo&kaD3-2&Q4Dt
z8oY<)#&dphm(#N|hKRV!`(tMPId_exKR?Aq&-&%p=&oJo8+MOxz7e^$H}JkE4|w9y
zBd%SV6PI^6ed9G=UJsmY0x!P)7SBBYB6s`9H=lo<ld}!q`@=uLPyf`9bL+)#^5QpM
zpzr2<<(cPs|94(wy<XwH%cWpIrIH{7mqHA(v04vyWrx&&#*xi3RyE8LI;&&p>T@_%
z-|9R$O~<NjPqCu=!1#WOyoO=O^k>s~mEdVdVWdez$p9cF;}3qTb-iaaT<sgTEeV3?
zEjptGK@3Jq5WS6Fq6eebsL_*XQAQtS^g0-f=$&YzccUdp^e)6i3whk{|9w6_`~14s
zUTg0UdtKLg9p~}eI~Gm+a{2LnhECf8_T9MLO<ti>JzR2UJYLVPDc)n`sH{c~nL<t-
zzZb=ml9^dw-~RSr^Ggfc^I-k&&2W=}QwN&yemXCf9L*Is81R!2bE=%LFXo@>KxhPf
zJ|Ak58%GsoYG9EvQZAl0x#bfjC;y7{h566pQ|7zP$-ho_*blcaeumy$T#daCj!F#O
zEbcg)5pVT5qWp08;xC>x%h$7WQjZm>?Zacj7oC@sp&{tiqK(>vHvuR57O|tibL+2?
z+3deK=Ht}kLFczwix*!`bFg^;hU-u(k=Kjo=LaN;jtrj%De&`WAU5iNWP)3Fu?(Sz
zAMp1gUB5KChvME|)1WLmU%&T<yClBqy}NsWHlCQw@P8pTfC=n#?kqg9p0_9%5F*aA
zlD6)eoDyvrC4I*veaf7PK5Y<r6>5hr&JYk0->7bP_<J1pekn7=<^p%62M`Z3Zx{OY
zH@jQ@-$<oD>*ciUspRUv=`u-4?BCUse-F6v&ChdkzS|Kz3O_jwRrMR2yB>2^bIP8G
zN^!~|_bikKQGq8X=F~(R-BKu0v=2mRVLE)VLxY2oRcJ$J`>J}S2KE@aTsgk=W{gC(
zfnMnn&(Nr&jMe=}tI2l5by;nM(654f`&k8AYA*mmS!BBCWVM<8hJUXJ2w*Hy4!R&T
zLZ0f(%EnH=5{Ze?RLz$blSp`b!SG0LsRN-(sG9tn1Rbw{2L)sl(b$`9J50gw*yaVA
z^+Zn`t90i&h=>wN5X)Bzr8?C_oWuJCtVWmdJ7)JtjRQ;PL}}&8&&H?uJ5SyQM%Bpt
zW3Q~&7v%rbay286=YPnja^s!O*f;E@=^@Y~CI4#kPwq#_KUV3z;Lb(Ts(N07*ZNtT
zLZaL=lA}F6OeYU-9zXeG+Lwn%bx$FvCv61|gjP1Gr7}G6|MPVDX7WdIbcT?>*lm#U
z=dUaEBE8o+v)LZ+0)HbGX~CSrQxYl%OcVH~M*MA(s9I5OJo1-uo-E+6QVRWbizmNA
zHpe%-Ev0j(8$_}OiZcTgr~-4=eoVU8lj47yxH@M2OOb{6c@etPyk}4HB4i`Lz4?XL
zyPrv9p_dB5zvJ7@WsHUL$9Hcm4>i!!oFK9HN18}lxwPw&<1Z_Ves4TpD(QfjpM)%{
z8;^ukdzX?1`p$cS^_5N<DWuQ$B*O<8`tDX)>fNJ!6HsZ!W#erQjp}@bc7ya(PAGg&
zwL2YqRf=s>tEo)z8W}|N^Jn*ZL1z`gHK6W^AD}!1lxrg^?0o91tbM9wYOL|Jy*w5`
z{Sg^;(?Lvy(ElfJ%E*n!2AFIzB0i<jr+FxJ)rvf`gZLSkEi6jVH#GXirV2Z|rzUUD
z5n|Wr^kzWLYrLQ@dh@f<Q~0tU;~V;-WX6nVY^q8|gAY^lw_RRz4w2pcV%=f=yOJGY
zKjGChd5R-`l6QJm^ABJ4%p(8CpzMVSORN9u1xcG<4Vrydd-Mn6vG-H&>6IXgLHl|%
zStpEID>rI~^tz{>K1?X*KDKPzr<FUx`>)pzcaO!XFMHD2<Txq~PomfeMjA<EaC1_*
z)92N`pgc)kqzWmB{run(i2-iP_;2}%!xZB|5x$(L6sf)NAA6fsK>M$O+E$v54i$wJ
z0QY07<JF#~^i>UWZQ@5NkMDJhr<8G>cy1e))06*j2P;MLg0{&ncm81~y?>q=&i@Yi
zgObc^3516>mJ~wf$0;A35I!(}*Md;H1(bF)E(hgk^jgqsj*nSs{L<?54*a9c9eJIz
z_pkrg^U216JN;vmfNwZbl2qY|ml$=uRipb;0uJn_2je2JF1)3<GSgVlk{L>}{CccH
zD!aYaZG3%oLE*8tsaf*^%VR<NaKALB`&9Lt2PdQK4nBr^HByA-EvFGdqy2MnQTs|)
z?Pj~<JRDSp;|Ip$#=^sfVr9pIl@~_srec4yh2QEivJ);cRwl7)`YA1B^-6HqD8vGq
z*X`*X1&AxZGS8k>{}xG@oX>3CxK4Y{l#44UAZ7VddgBr69eMag$JxBuSaiR&(uUOC
z50|y|?doQU4*#_9J_X(|2X=1UD%LOabxTzbVy2g?=3Be|{*Xq1F0F)`@D9FchP%wH
zq54Dxa<L5);rG@q8z@J=Vfr8TA^!9&<0If!#_&P4sGR*Omc=2rh1}NkpY5YwRZ?4U
zxm~^1&V2N#tOS%Qtv`~qUBbrqDrrI6T{wpUW!^eMHLgS&()U^B%rkUn_{F`YI-Blc
z{^it7z~mjkeHYL~cvaa}rYV~iqZo{YC8Z9~$T`9HTB>PaPBf1KC7-s#m)tQz0yG;X
zJ|cVYOd}tkPz(2lqWn~91_&_#$!!p8nO<x7AWsMH!iWc0sP0k?YYhE*0QgqboWhaz
z)He8%`uJLBUuKJ@n61QONag2=;{%#$=r3!gmEusp&PAeZ7n*6)IG5N{vu+MXU5^&^
zlp_~KI(6#hw&$i`12r3>)L5kuYJ2WA5+{<ae5KA=Ac?}Hnosav<U+vSMfdBD-%=q$
zuYND42H&@2>E2{(915fd?}ndjZC;Z8hHqW|G0VJFflqFmcf3pP?3Atz5hcm?Iq5oC
za(!5dZ8cC_tfITN{JzsF=iEQQo(YX<6+ul*nq)&h@K<nZ5f7MEmoL(t9vaAlpX>8b
z4e$ELjy$_zisx+9@bcDUt{NxSfTEQ$xAC9OTW2TC(-%bwwfC>DgDfvGZ+}k2z3>XW
z$oaEN$@(vO^0YQF6so~^`TZjFX|sFB#6Y;hAA9DNsy-ht+l~u(RGr&!s{!%!Ch5>f
z3C!AI_R$8iQlILpV9nYk6ZGwxUGJ1<`U1CPaDPX~)nCm_V}h-}d>?hFROt0>tYF^O
z#0`zyun2!>GTl(A2g_#Q5W^0BbtYko8O%jxfG1Cq6<xtuxR_al)Q0NLWm<oKVx3$X
z9w;bJ3>PXCdUoqM@soitVw6}1Yir0=HD|-eSREz51lyB-Prn6$(#6d=*2&~g^@7J8
z>l^$FmMe6WUD={Ol35;_ta@E+<}ki^H5mHMd$b0=@#`MnjO+e=(2#%2$sL#zuh@;~
z^Zt@K?e}=h^mF5*(cdp$_b)|hu07+p#0e=l5ubTKNtOcH6W~UVeF>-6Td+>7jh0`|
z_$%l_d_EJXVH{3b^L+m^OUZ~>_;BsuQj?@|H&H$NG&*^iQXi?ryz3HD`#VYfH#GzJ
zQ&Ur4d!7kLlgj(Kd*xyFVRd?kUtk;n_irv&_rHJF9`KVFyC}=85t}H<MThzOER?ux
zwI?WxadG88izD^vJ82E!IyA$T@Fb}h=HGaxGwhq;Z#|yacRIh^hfTYEdpD}8;G0h_
zeOVDNZnUEj8MK@oyawyB#+8dKCveRP(M`PT+Otpn&(|DS-wWazI<x%ToVQs%5a3>%
zhcN0{oaWZPo8Y5G4knIv?4o!358i!%Xv**x%D;!ej11y!zOpiMx|r#*ywUpy7NuOq
zk*o|JgB_K7e)1LUyS4rW=e^=_7=y6k8cAqf4_SY+`im#nuoJwl&$`(hdh>*M=egDd
zK4}MA@ylips;Ez`F2+9I%O)IYG|;}%j#`8hU6f@OwJ3{t{bi4Kk`sRueh=p*xz}IQ
z{T<R0D2{@cezrLpp$3#dPR}EiSHBU(+p|!6dlYj1b@gj07aGD2zitC(Z|YV~)z&)k
z4|h4@Fbbeeq;?3<K#!-Q2cNc9bI6(3ZY#7;u@3oLbdwry1M;4)1@-sX7q~Hk4+`9z
zvgvbOYU7LD%-T?l0lTW_xgj>S&c?cv`YvL7-SNX1B?C;_!Ys?&P6cJwEN;kc&_Y1f
zWI<5@XTqonj#pvZevVyMa|*<n*wNiurUppp;Nopr75MA?2IZt*UA?lEbnGiloqD!f
zUr`C($MOB5`!wPqW7=c3={2DQrL6s-i*OVcRBeZn3TCFFlvjnYQw3iC+H?nPuF^<W
zzRmuFj*GNDsmr%DFfuT;OX0?lc*S$!KSA#p=}<(YN(rZ0RNwo5x+*Fg_%$Z}*2dEx
zUR<nkIgVd13Gog^->*C@#35n_?<_Pn7;=y{x`*IiYOaUoro@rvZ!9;m6ipGD8raXG
z3CjL?aPrRQO@%&AKL>3}(x{xE!@fOVO1(<0%1Y1&4<=6xZ)mtgl@@Zw2(WjcKE9qX
zG~MYk1Fx<_R{9Sn&*?9Vhrfr46MrAZt*zVcnlM?|*VFG#W?V{43DWY@1~nW(n~oeA
z8bu_(QidU&ihQ>lA;Ds{lbEP<M@tdWly97YiIA}*XJs;1Cr14sj9{BUp}eVYJ&c!c
z3`M}Ec7!D2N84;COz#Aq!UJqXM>WrE`Y!n$gpzTAt0MP83u)f_9cV!RJd{=m4x>!$
zN>?`bjaxF>LFlbpPWdfgXm&o<2kHLnJh1dNnFTg5f9peJtaj5W%$lp9&(8F}W$xp2
z>a4sZtfDS`;qU_Y*RVvu)Od^69^c^LpwAnojHljXHmvV`l3S#9Hjjb#ll|{*r4Nr-
zzchCzdY9Uim&a$zJKlr0=-K2cHOf~y<W-~4)|6e?R7!_(jLF;>*Aqb{G74-Qt*`cU
z&N6uBB>3Rf9fLqgGo|;qEFs2r_9KXx_lQ>&fS)ciHFnY3erg`P9)6gg@9nwr+rE3V
zcT4(kiFZ~0<ys%edl^-HWLojnYU`_uU!)ity%RZR^1N}34)&n@hp@fDK(W~8Ib?BO
zc-HoNxGG|&kgaA;Yr>%SHHD*ZVi5@TB}_^{@Uz8h>keRgXO$xujVg^9G5+uOT=YK4
zC7qEIh2Yky7A~h`>-nuJrwvS*HL<0&bvN^nuO7jqX5n{h#wP~wUwIfl<+F!HL)v0M
zkfkm%Gfx5TBd-L*im5DQzOQq<@bgsSBX0J_;;MlQ%jI#EI5p+-aeK`TX%UD-n^)Rz
zY*b%yr5=@2$k=tGNA0fttLwYL)^~eEzXxP27HI1o!MC$smyh8Oc!TtO#s4A7aRw?1
zoYNUpQq>WT!qgzQKv}NG7NaTkh992Qh{1Vu2vPl(+prz$SVFl~BO2~|XWx`fwCnY!
z(A|5TcHxC__fMaIiMuTgrtW~Tyl;aI8Z1lQ{XO4p#5I<~IsRObnefw&*$ZfHcqJhi
zw$rjkJ=YM7Jm6B9hU?jq359^$6ONMlC?Se{c#|K2YiG%MtT1wO5k6H9grLJvF4wNZ
z;+)gr?n_O9CL&0>0qc?-EB~^NAREy+rd&m$)NrlPXLib?$Y+Xa#3xQ#@Y?=U#E}iy
zcjY8-z45}P@}8XDyV@Mr9jY$3ZY#J*iNa1&n=*2g_ZgHsn@Sb`v2>PD1d}{&r#scO
zc3dSp@5{m157e1F{B*C_AE^4)=lf&Q!<!mEFZkL?&>{r36s$GdV7w6BrpskGXv~C%
z=1FZD?vvmnFnYv(Wy$<Dy`(eH(MJnzvf#4K9aO+q%<DQ;?#9dgxd~;rT3q)A+coDP
z-ff?94VEkb(#lO=bozsCzdCo?`}zqQKDF!~j(;<8FE+{z?Y8cE%&*=0{m^r3ga&2a
zH4<D~KKGJx{PS1|XPKyN;HrED$T`8Y-|Fl;Pr0r0pK_-l^?dB@!O2J#H5(g`THK=+
zb9M<zjiD#8h|*=}+O;~Rr)^yjov)5nN*zN>PxdTJr$@ry-UIkL*b*Bps`H|M#EqiL
zuS^*ZGcjEwQ*ife4516Z=Ir*3reOu~zY5)_PNxQLv86(Jyo+L&L;`Rxo)IQAuFk()
zjc#;iq8af_;)`9T;=redcop+Pju`pad(*dDNGG$`TTf$p83<E2**T)DZ{{*;6`#6a
za)+P`C8Nunh5Q?j3iz1JEP$W?r;*Mm#D;Ufil<x|^ZJu$;A^b3K7g&*w~upOkKon9
ze)`SRJ|TLOaus8|qA{+5^XePh>K;koA($|?j4B37{wVk43q@FrWRiIQP%7&`HZjw5
zhv(+AHfDXLOJP@4_n~#umd9@7b0xcV2#pnmb|qPma82XKBH^-<KO@Sp_$6wDTdpv0
zAf6i_WhRp!SsQ4oSt^-rZlXWWq^7OsTIIm!Ea<Mkj&}gcS8Q4MJ+~>2hjV3a`mAv>
zgflCN=(baV5X}cvBG8_tma}o%6d5TI^;I*=a(G=6!NSo~e*L3PV8DL;JE_Lht76mq
zRlLN=uhf4TE10r~`#GbszB$(m<Q{WfQL{hohY@hXPGaBJl#uL#^tdW%s<>D`##Bt2
zii!?ro@7PAh3oLQ5yMJ<aKcui7S)46rfJeJe3*%$miv0q*r)MbK2IbxYJ6S<q3SdC
zV_>=v5-&$6P4O9j#@RQvM4ERtrSCxsxETMvJ{m1y$A`aHx|37f=!RU^aa6#oE@+8v
z_0gp>H!Z75dBayVe7^p*<vUkdlL}_2$f(;YN2k3}JZc6c3H(-4y^ROuk#=6xt{bix
zu{Sj2kHIP;8(=<K@|^vEH`ApI%|(YLmuV-@H6PfA?V^LUz20#*0@Kh5fF)&&x#H6P
zJ>-XybQDmdrz#S(S2eGo``pavJ}*`x^>xO8uy5Z}hr6_{HHn9Qz%4%9;+-)edOhpZ
zfw<QS{inRQUzuR~M)&h-t*7T$M=b6N%bVL1n_1-Q<d_AQKimVrmqfd$G|oE-+QZ>$
z(r2A9zLS>k-%DEL<P5XfF`iQ7{`1&Y=gi|{)OS;UsRp3}u&BMZD}DQTtx|F?|FcmQ
z=nPejQIGGcvaTIpVJc?uT+*c^I94K9J~B8ica&WDJQj#+mnZ1!PbvxZ{JXTjT)J-C
zkbEEpu+u%;p<zzmIpN;|M-&xh<Sa<N@dMA!rYN*e)FZ<{;TU&yxS^By*04EF5gA;0
z#K9FIsg!_*c7NzT_&Fzi6rsDthItHf7AP^jR?N1#&Me4v@q((2HTFK=uu;}_W_re`
zFssL<R`eJs@ENlP*(E_#!F78pGmWqFkuQO?W*P#e2fqkAaRL3oR0^0-20%f@QKb<U
z3hZ~)5gh8-Ot*;Sog-F@X||`F0GV`F4oYfSFo<9}^4MczDq717oh+(PGJ(%P5w$o-
zuKEQ1)XG^#%zD}_AN=MzHyXX+h${9x;~&1$CV3qUhE|nJ&@7M-qta&@cG}SC$ZHqs
zs0wYAE~E*_o#;My^3|v?+@jp_Ei1sOkpz#fC@}0M8K>rXJS?ca!)K|rhYl+;O}hU)
zu`e2@5`>pJzZxXV@Ms=N2R&Bka-h8ov^P|hx7XWSnr-s<Ux2VM(_rQy>y|GYh@IH*
zW#vZ(3IzQRDXWuCQD}s#YI+|+%Eg8!cH*4vH8<=uLepNDUg@>)CF16pM3oqcp``M=
zm8W6p-p<O{nL+TgbOmFKui!=Hp|?5{LW<6Gh_j44X-zx*b8-m_pi81+34$9rH>P-@
zd|@59cg^6Y7=sKk72!ponxpQAFMo?*f4=Df8o9|w)1Q_yj|3H8=-9G7R;4c!CMk{o
znxvjOXORMOdg=L=P)1X<2<q{Ls&R}Ul_$tJ06*MdzCY}D>Xp5TEix(FrldxbYBv8`
z=FB`5BY~lX7SFAck?YrjSYyQvj|+Kaqv!-atz-4GMALwwvHrQKs3DGHB62*;fUxoD
zTkZ#L$yk<(1z?x}4gHK~FXv-jygpTpmdNOP<7C56bNtu>88oA8|3h7+=J$(6hOI@O
zyK>{%4o`lMj!D*}&`dBF50SmGdH2v3!x>&e`~voKI8$-UfWQoxp{j->R=Yb*C5rs~
z775_2$S{@KBvdeZ?n4Bu)BvVI(j;l9pRLvJb*R*wKDP_4E8ME*P`sbXddb(Fl!aJw
z*IhSt-li6T{E{(5s4`ZiOc@hGi7}RR792`Y`bdTt_sRD!`m-485$~1vgP?GFv-^ER
z=*XruUp+>>dfvDq)a08*{II-;d_{ywu@c9Pd8MLr>Wnf=fZA*e!vgscsa*Mn_BrCF
zdL=B)lHH0i<JI~#h6fSJlN(z-!vS1s^;t)58$A@L_I+>^vr=}o|Jwy4sS=dPXFbvx
z%;(Mj9>X-gziX&ZTFVEZ!c3Iz#zhFml~B&5^}d1SRT)Gt+U2#bVW;+2g^1Jmj10)=
zKXO_$xrq~<k9_ZxGSB);7~Q5a7T}_Ek?ESi_${PLz>?T5liu5rh?KJPDQ_){)E&Qc
zF)!it>!Y7D8WeJS&1F2Jn;cG*o{@BVS}<r|!7uTF3c8eqxar%pa#2}6Wd)&%mx8v6
z4pr)3y)tR6k<K)t>fd=&8D6sbgo-^=8Js?F{x;1u1Qx6$+`|&a6EH7=JEmc5)5e7+
zn!F6tD%b6cLUCCR&p`3}!J0gts>r*!Ry#ParjF)3j^0U2#~PvNs+s1jmV(_%86Qf`
z0#dibpw2a!IpuEnmROW^hKB52Y97#Jvr37#{-f?Q+i~kTU*vQSyP<A7*cq?Po-x0<
zxjakQ&%%R5d6g!rc+wjldZ6mhd+#itzIo4+Sx*X3Ed#ohNAzk0?5MH^s^k9JWVy03
zSdujv5aK;i;dGY{02nCYdeM2aL#EQWnjepNmh!$dBlRFs2J3;oWt=}tG!z%MjX}_X
z6oD<2nk?Pq3|ErApvmY(Y20Z=Vcck%v3%9^s$osisA*Wrerv3{>VABIUPflQ(5SM=
zZ#Upf<*iO+Ps1y%Kxh1Sg~UXM=iZ!!fQ$54f<2wZOv94Qz-9Vk!v`aG^L(!MVuk50
z-^NP2_K6sPk%V)Pofsg`nldVmQ2i-RE_~R8t=hc{Bnat!V4$N~A01Xzr7se}?y~va
zKvj=^Df$68308HI35b97dGGN{OAe*tGk~OauPbwl{3FW!q<OPQf4XmgvSADWZTq)I
zo|Upf$BCs8CvfmS<6CToFgR5L!Hy}nakKg^u8l9Ck?Efx)-t>$TV-km?GZJ_)#j!t
zl5m1uhH1}mQ{=dri-nD>@A>iih%L1|dtgUZ1EW`I{l@pXM$EXZT8`vn0qP*L^z+k4
zru6NAfgVn$VdJgx{94~k*hfPy##3^_N^+WJFJsQ9ZHpvPuvN;_yhC}vSpItdXftmC
z=4`BQX$Xg7%W{5L3$efUR0@btGx`>dAf$`^ZNqdaWi#efeXN$~>B`UYEgY$y4uGzX
z-Y|@;;Y?(`O1#FNhQc)IDqe1xDY8$L+H+nzlW!mA^f}YTETNv+<qR0^y2Nfnz<!0i
z3i)hV*Ilz0qG7>*molfySf31+_m;kT-ft`q1rHC7f$HgkvJa%E4wyCKdQ2HImc>o7
zxx~u<Sf{dN;l3`N*|Tv>Rjk3L#`h?{)B4d;DMhys$24<`H4z(ly_F@8w#}$wuceEW
zqbu^Cbl!WdjJ>*O2=0fKZ=Ga1Csd9rjD~aO*jpGKh4tRl$rC<6nC@-aj<)o*V91pr
zy+q*|-Ls#fY`=&~q?&SZ-58k2DxgSeX7qut28^q3AF2%5H_&c6pvm=wRU=&~*NFjE
zG+f+vi*%_Xq||s2<h4A<w((rXb=N%j$QhDxt&089DE7+!uX8ssU~^eJ-~(o+#6?8D
zPHYyAf;2)g*XS_Y2I~jBr(*Fc7^IC94>ASXT|N&mC^hvt;sHCv_Pby``w3>QY0xlM
zzzw-PZB=ZMPMYRHxr4`V(Fd)|HO{i~JX+7BgVa?{=u)!ApElG7B$F4C2jq&zEqduc
zUC4aOukiMpYKpz7Xd=@q)`zPe%1`orH_+A%B---kbpwC};Af!vgF3l|AC1ogKQ$xB
z8T+RZ+Nnu{kL3s(eR^x>4RLjfMbD&#_sa`fl+_05)XH8~m*EoO+qJ>Mi*o(I>>uQl
z+^6*sM54+J-Xk<Zx?F{$jj{Yc9#1Bb>9M~{YZ=z3Oj#~@EX}PVA5zYd3@GGDB><qa
z0MBr92J?+(jrEKJSw4U<EDyre<XVa|t81?_reW;y<tK3JA11CD_(`kg_`c=T&pws+
z==pjx4wtXLf&Qa~WJPpwC|1;&o_IIWaXxVN`NPmcSSr{$qY&8buFg&zlnulXt4%e0
zG7wZ{v0~$dbP-0v`J`CzvboPo67RuOe0jFX?j~<~NQqut*Ryao`;#I9QHAjrtM@8M
z4O1I%OdX!z_M=!;(zj}u4u6v(8ZB26N9@RqsDYrSGI1#RTv{eIMjCr%rCh3r1Lmoy
zntmN<jAiD_a2ZN5x7cp{<i}L0#~e!(^8G->>`SQESHuN{FJA0&QPV?<r;XoCbo&(O
z45R7f0?ar~l!Z-RJ)d?;CJ)elDfc`qE~09+;>WufiV9}sBTF67Y`H<CY7!bg@^MqN
zq$pyxrKy42&-LA#ORpEcfY~>Zc?Qc11d4!!j%VKQ2&E%*@HmYw=5|Iep45#C3#Til
zI<2>BLi^|*@g8gAr3k-g6Io<9wM`RClzuM65tyBz-(;34dUbqeCh`IXm={GNzYx8H
z6awSfcAzPP%UbbJ1EI@d2YJ<HWv+t%S{lm4sZA597cOe4RJzZ@gq6eoeKBJ5xYiN`
z+gEeS7-zSBd0<`@&I6J+Ci_4*CW<)c8+}riGFd*9M3e<d*FAKccaE8JOC0yfyUw#T
zdmJu2aow23<F}$gU~UMXo8eOzhglFO8qY@z-vDrn-aLD4D*MRzSyMP{(hiE>KHrqR
z{;-*{Go!VH9rE+TY)<2xo7CkPSOJ4+OP`rvUI&3@$v?eLNR8#hBqwt&RD+rlxSo%h
zW9a24UJ1Ael*!EI5xrQT-cjyyuz#1lZuv;_x9S5anAd3C_P(8xZp12Y!pX*d-Ls8x
zO}R$E37PUmHJhhSq0yf*A!};}tw$cU<qYGnbRU5_=iI^u|3`*go2y{HY9YatzF25&
zKi`Mn%r2x-+f8~9$Az0l5C>Wf=<hO{MWd+ySZ!b2d(XgU+8~AcQIP0>H+ibTJmcn%
zwtX}=UjG%5)8^1sY<mU?i;BMjAgnC`GZ$U);Y41e-SW+O$%oOz@owMFMz1Y{T+uo!
zwTH;%!{-f!Q8SrSiE?IwytR=M1&`TsEn~bjqP31R4T)Hll+-Efh$=Jj?X!FeqQxe~
zClz%2BG4n_ALnqzA_#48*ySjVE3vtu$|EF+$sdgLeM#vn+)pYk88XUTOjk9A)Vm=#
z5PfofE;hZrzLJy7uspxQqgo68cui1P1KIFd#$fS7?hkB_N7V~lj-y~YGv`ZJ{DlYl
z@_zev)zWdk`({l8Sz$cBV7&O#uM4bMY7hL=D4l*Nr4QRiNXD0DsGbu!XVkn&g$@e|
zF3i0wCLxw@D?7~9F}`oChALw^j7b;$S#;S)c~kTT;kvjWN~Xb_LmPR6jj>u`gI{74
zZ&r+4vY2aJiXd<PG0xj9&z<h*f#53H6sL0=T5AzJ$QC3S+`3)wM2&DZXfo-ncBWaS
zR#rW*Sb@2O{_`U5R?=dGFnDCPUNd!`tMG_$Kgm<2N(9DcUVFQjwIfMymcuQng<!8T
zmghl;nm>IZN~5{oeeh$&`HFi?LG!^dFqL!{CPutz|G1Zqs*R}>3a2!VRkJ3C^f*wx
z=e?Ns9KPu1O`1<jai;y{xLe^@JzOYJDCDnH`;afT-f7GPraS5q+W%MSSBq2WBYm&S
z<?SAmNrke3xi>(0z8;=s-uxF!OI!6iqy@`dL`aIpB760A*GItRoJXRD^~4{af)L71
zI?+BQ&ZU|&7ruU69_q=Ol7BhS2Gxn>JuwyiJQXC=4Qyk|XkXbOGxReE#d-r=%ee?m
z$Jr%P-*hw4Z28TCSh#Yp)k#?sccYIz=&kuJLt<*~iLz^`%|;z2lUjf*>|wH{e*k~h
ze9{hJW=X3<^*->trZPX_f^)L@`u(cxN@7I$!>D1%DWarpH+Yi+62Y;f$4s=*01UCL
z9vQKnf$M?<Qj#5%2;ooCoHUaXKc)1W(Zv@PWRjafxdBdfUml0oYtA1B(*>FHeyk@z
zzG5lOu0yn;tO%^>rWtv)Cb_!)!!bB`nW1+mfalmF0Y>iZnl)|4ESYC+F=$G6B1{>O
zZOQS#oV{EkZ2Gm>qLKk+0@*I=Rc~OZ*P=QJcKO!yTRfs1Db#Sp+6m<aJY4<-<mlXI
z!Pf6LE;c?zyRxDv3JvqK6YLf`efd8|7;xJ1ikdMe6j<Yif*;RY=PYunZlq9hT$^WS
zX7ns_i(8X47DRtl1?l+O9}ZMG*|dq7?Wb0coPoQF^WS`Ui?GAemVMg&gb8r$`^<o~
zAx<?rv3!rU-@Zy1Op#GM5=|C?>A`J$6O<ovZwE#N;%{{})F{nttx3jqn9gQNo=Dx`
z{oD@ISF4;()|O|c9#G*uyuAMefft9W*9)H+hK>T)BQ*9)a_c$FNRv2R8>Kb`A7-lu
zfD*JH^(*%yAHVln)BU=QE4RxfsWZMC)L8Ar5Ns^+k(<@oZ`F)5cAU{<{xPgd{g=4_
zu@%EJ(Y6;#QcmA$dA-^kyiRcgpaeh;<k3ldP}}gbn5Ke38d%Hywi@bVqxY!(`eCbg
z9RWewX73kFM1fvM7ve#<{--yy6R59)6`x%rM)WnkeRwj*$~h=OdU|rsC9;+ayx02H
zN~Y{eLB-{f#-j0o=-F6y{T>qmaiZOWpcEq?b+Y+Zfto$qUY1THB%`YNp#)Tgk9Tg8
zcBa`=p4P)#S~O*O>w4T-*Oa2VIb4fdj^`0^Oe-A`#b+XAEi&lphc>|VwGT1Ao@e{R
z-|_U#L8uyDDY*Mcv{hOQ_t^79$wH2h5!8uuj|?}ibSL7o5?YVKhwBJ^5p6bR#ut@k
z9$I`CTpa~o`v#ii?OTiF!5IB)bmgaL2dRh<{Sk>7dJ+z();HkdrwAY-LR~K$nKoMx
zcrmc(eQ<ahvej5gm#Qlu*Bo5((CsaUq32l^`z%EtA=SAsmYX497Zs35mo)3E*QQJ6
zmz)TSB}+YBIH=Va+8gxkPLsnt!j#_#cT?sB)?k-?Z~y&Vi8Jdrsq_R49G5SrS*!@x
z>tz`{Gu8S*8zB$}v~0K*-h*5JQe&r|`dfJpuXtRYH;Agd`_3co>Se4#tW#=P$}s$w
zBjRHsOQ6@6N|+ufU^QSim3r<f_NT19S%z_3u;t4%TZ(vh)WWhGaDsO^n)ByFNtLN6
ziR+xnpmI;yVL{cL@A?r*A{1rP1M3n>k%vPBxEfg*EDt&GSDjB9B0Twy39dt-1NA1r
z$-oomlG2R&Z%!g6cG**w)pzqg$sf4?j5Qa-|90?+TI#yU3<VCHg(|w0$@SPFlgws5
z3rlP3e5ba%g0+=EErYwnmIjGpvx$1@#FR4o2}rcqXE78h)H_ABFkXs{@}}!=79V)H
z4XNBa$)yS`2tE_045u!1r+xKP?J!l}1&w286e_m(&hxSYEE4UWGGqyN^X8zqhC>vt
zXHiP(oArfyI^a4xTV+ZW!kl6(%gQ_~)FH6NX2n9Rx~x#>L+%N#=bX1dP4=ls8jgab
zK}wwXvBMx~AYGyLCL7*|8U+a9bXnj%<DhH9>`bnqA)tiD2=UTz`)Tato>OrE0iz6`
zYy;<R9_ytg90ef@8rxZ4T9)m1!H-q*+%2$l-X7|=-=*;!X<z!bohASi)+d_ely9r(
z;(BBkqSO!a;$*aRuA`kwJ!5eEihmo*>OU9{x4eI0P}HA`e}|pKS^9?RGXi?CUu}Ut
zDf*Mk=}7LAguFD3DWgdgBzkVbNTkNXFYYhY;_)fVXVHHs2H7E|b;lf_5e?uxNcUI2
zKfn4rJ~**V^8m_pRm4(U@Ai0_&RdefLf@kOYz#o{@M-#(2X?F5Xc6Hsf-fY-@$kW8
zq@T7MPaP=`oimhE;II4<P5rZY2@B*jGoBF9s7Q^W#y0M)?sXr8#~v%vFk7Nk5ZnqR
zU>j6h@`Y>B+5E!!>xG~nX9S4zK##_DT0XSvlU0iaSPyj@TYy8{d?7N^fB-e|5~dU1
z@JIepEF$H;Iilu*uokj6p#H@?x*~#xjwXR=JvVJMYbqaChI**;SEueeVDA@4G7CEG
zE$~5NPFWH3kxrPgm4mdP9XQLc_~P>^TDy;NKVVm`n$RlFxQLEO<iX2qLos^&SbZQQ
zv6zkD&9hYNp=Y1N3z7m1*Ac0M(p}7KpN%)lXs-3rEju-Ra|r5l7Agb2ZU#QwxKGT6
z(#v<2fwbltSmRCfAvM}79-LH#bS}A`ndnU{YgqwwW-1mZ=P>*D!bcF~riJOaMt0Wv
zR;=(AXU$mLR$Vt%gR1HJ#k?xa=1=TnuRE7pbx%B$dxmwXAQZ;H*x}bl21{Vdv^>Ru
z^oxin;C>5zw%0xo!_rqxhrBT-J6g#WX17S<6|don3az=*8dnP;xTK5C?MpgqMN5T*
zU?N1-QBWwSr`$c6snd&$LH;q<+DR1$1+a_ghq*qAgK0kD#p~ttNfMuaJ{4`YF=J5V
z6+OThg$kmk5#~UYtCy-u%a$xMJpr9c5k6Pm=*US0YU;X=mv9YDyVvw=&V;bg&wKZP
zmT(}Hl<M9|L537RcyVT><Sdb-?M%ex8IiRd-)!8-_#rrWK5Zy#9^z+4qPi~lsWEOx
z+J$nt8DqLZx~p-LnwV;@Ny@S3j-|L>`woatW7J+V&q@PN`1<Gegkn$&E-pD<iR?of
z5<NyAYx6SB2k7GLDIxrYpj8NjDe@oq1?%5=my0%*x^Pc58)ms}=4_y0Dz)a3ytPRk
zKxfAE3Dw1)&P<rh`z%C?Mm#54n5!T#xcyMB5qDxum#{dGC06M*`Ny7RRf03Kn9`_o
zWPHY{BX#bka1<c@#_4IYPj-nbMVQ=kc1IkDmc|t63+IU#ptuo%9Fl`H{INJKLwgeB
zS66=yp;7{cX(IxZ<w<4mO!+D^R-5{zK6pmTC|?A<xfZaVM3IG@KnhhNpWG+#w&`$u
z1Oz$oZ0Gg!b34}-mRJpN!eX;6jsy`S>;z7-dP<|Zc8Lv^p5TLx`l17r?%fvg-j0hY
zFLczSlSSV$cl+hCDM88Tv+7<$yN2JU2D;c*iWAaA|M+s>X=wVneGuwGs`<p+Zi4Vl
zu5ek!o$Rf&ofP)s;N<wMSF^kRfL@1unY#68`5p7Dfh=lJDmF2eYsAt(IF+o=co{tC
zbkc79lD<E^Q*$l<l}5699bOV&a{7(Nj*~R?s)qYoQrq>hJlA#es9Z(`XPf}bcLr(A
z1&M{xti$-Wd^}&HlkC$~JAgjnmn>Q@N{JL<_FXWYx-v!TtwB)ScUi*1VWxe2j>a$9
z!uHly*zfvUc5E3u3FuPQ8@KD+;~Ugs$&H;DxRHj&VE-&Xi|uU34cM$bIYGd001gVg
z+s7`0G6d#)S=s(S2A9`Ou9Pg7#kSMAamw_i)GpRWo6Jr#Y4!r~I2hajfj~N6H?BB7
zZCD$y@<n0OrVA)**jb6fb&LIz^)}b&DNmAw^yezT@@umoop$}J&@ub_ql>fmb01xI
zZ(y&89rw&2#BSf*3oSZ3E-qp7Y#SU0^zG}#`?bjC=O+~u=u*GI;Kej2Fh%^a?CWE;
zoR69-8bTtwo|3sGJUgB7K;d{5L;JVc)WL$vPwQI?d<APzT_}8qtFo3O?1SsE9*quj
z)git&HQxb4*tYIiacdZrR90a7dnb|Bw*^a~tSJt!(geKJaH7a#(gf5ZKa3M#1j`pt
zi>+B6uMYp~I}1kXGWH)21iS~Z-GbA<V|u>(X1bu84`i<4dy9yfa%G+N;|Gvzn%(5`
zDedo}p=%c$)P`LiDwEmnb6Qb$DL?5`<Yp2Km_mF6WAx)w4HRA>D0wz$_r=O0z<3Is
zO*qqVSA||ES#&;ILZG0YQ{!Q&Z)ue)DuiW}i5OyAd1F6YRsYacZy@dbJ-zAli9F6Y
zOVyymSicTq(~}wudi~maWGdnTdX8%PHQ%G0QPrJJg;ADG39ziBIEWbx#|SE#{A;JJ
zQu_E&d1rKR@M40`7o1QLBlIFaH7(Wjj`#H2i88pcI$(sVJg@l)o))w8Mf<(G+es>x
zA>7c^ty1IK?EENbDHxGbv|G7_GgrQENuWvwN{!4?&*cRkwq@M8XW4hO@Ji577OPrH
zr7Z)}e@H0l(<%vYax@6|CE8Mp2P!lg+W92wFxWy`E9OnMDZL4kh!-Hj)Su5F)=W%{
z15wu13YA3tE!AEU)v0;O$xx)wdyD$s7I`AGAplX^WK#s3+P8Wv+-J&dr0q>p>Vm;!
z7jU2;pPWRlZNHqk0_AJOO36i$O4?8nrt-UnuKsi8*zWV6%DA#l$?4*{a7BLe2c;X7
zb_qbrhrNoW;v3|d6|@B(D#y#G9EwhLVjrcK_|xl`9#Lun1D~^x(J0NZJCXb@cSKqa
zDi)w^y<R+-Ft*4Ii@(Y33$h@KFR9x>XV%s{o$06Xa|ZV}ReJI@9*|@SBP^&^<Yn(!
zmQT>*ffBy~1x@;I=az7Qo|vj`dYjenTYz{?QfmSua@)ahxH+&dsWp38?}adnwzLe_
zt|v2Z>poxhvTE`C2S1?qN;bmj9ZGWpc8jgh>J&E3tyed4aQ$<L69AU+k17Z7Mh-W+
zWgl6a_>S^f=+(gN_p#!U+Z_q)XfUbL7bT)P9#Q>FT^km)Uz?n^#0?NBJsN!qucbX;
zuH*GA?5d2J)?Pov0lAKGC|J$NN>JHh<I(4+mMji}etoUC;!*~-w=!|#w`M02br!H8
zA~?^x&XEc(Hli90e$P>3vCs#yQ9Y%WQ4Xg2N^PiJQgSalKg1U<_rme{c-@}x0xlfK
z+dr;QE@ogzASX~)l6UIu=Xu2IgPGLC)LJ+n`GB|X<mXKZghT}OWkd{}&h>_5NFCGi
zlgP~pxI(K~Mg=P>acxsPW&1NRVvQ7OR;p=t+$U*w4Q>jv4Odv*-}A3r8n1=|CH7pz
zgug8C$;tW4U`y+XWv0j;ajDdl<F*MdvGBE}zgjZ5VdC|LFRtDsKjMst7@@|LE_O!<
z+N4t|#S8jKCY+{-i?qwp`yq(NF^W@_K)A)WPAYK+R7yO!Yi&O8jbkM(H@bY-*3288
z2M~Z4(FhuFtu?Cjg1Nb4$p@R1EY^liP~#%{h=cERKkl8&65L9j(+oo>V~di9zHyT8
ziLm=FZ9lG;1&1xy%2w%4Uvbf~(nKY~6J&xGCeH&JKLuq3nozY1v1MB~0!v@1Md*T}
zvjFo=Y<5kpB*q5#G~PnbfZX=rCivXA5KRT$N=D{3D}8k(l87QXScwbAPBCzR&ZOGt
zYb<%Jxc+m2AtwA_*-a)*J8sR{t*c`%Q;D4~@uP~>lP+VKr~5xi0aF*9hn=1ShWTC>
zRmp-1aN6qe-Ac=Ub*nV&T*A1CF)|`**w-U<U^uJ*TGf{PhcjDZi1>@&!C-8Uiqhu^
zJKM56ZF#jKK~XVK986EX9eQ}(5xAyuki^WoWggMucEk6n074Dxb4PlM8bo^{6SQ<c
zNO#?%2GMGEi(N}o?=3#9QO}iS>lhG4q9F*~HT@6H99UTGTz0_%fg8O@SYDv<xi|Kt
zV#GMHkOe|gSw`&ginP=^Mwd`#CPnN!8?!rfS+Kn8&AL+o>O(30o|k_S{vO^bRFkAV
zvd2&nI?OAC%~64_W(Mo*<1FuGsUgFKgn<pu*>$#mdK%G1G{)jQ${v^c28^;^G=1sf
zh6?_fqMJ()4ji7i4|i1hE(e`ri#i5OYjKEZGxdh0Wv$L?EyFE+dUbQ2n1sv%_l^1R
zbGq1i*|F}N=!n%gKIfT>NVU>F#*o^h;Rnbh1Fnyd!WD~#41cgu24aHDP_E2xy>}=j
zi{BtdV|D5^7}_{x?f&E`dZ>XmBJf0~aKbBZ^tTIkg09aiAg}x8KJ<X>S4GC#Q24X_
z;4?RmKgZbrmC%C5i_U&3?N^>_lds5Ai2%R$D{H*`{UzinPb~jeA2Oe#hlYm<6J_r%
zY4@`&$@)uHJO5XA?QJaI`|tkm!`6DRf~kQQ|J5}|NayJPf8U+z+Y{l6_ci~O?94(t
kcz@;Kze+s+f8G91_E$>pevzG5CeHnJFX&c8$y<f}A3K<%a{vGU

literal 0
HcmV?d00001

diff --git a/0.16/_images/sphx_glr_plot_visualization_utils_thumb.png b/0.16/_images/sphx_glr_plot_visualization_utils_thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..c1c15cac09b00fde22ba595d6e931080dc6ebabf
GIT binary patch
literal 61829
zcmeFY^;4VQ7cJbP!AgMO?g>ST7E;_@f_s5r#jQ9L4^kkIQrx9DMT@o+C|2Cb2X}X;
z?aTMQf5yFM=9zis{C4)7z0TfitsSeasYHZNga7Q=Ga?mbdDydO&y$}%Dge&Yj;C3^
z|FdUj>MHUwdVbl5f9$JjjMG+P{=5(PnWfcL4l8pMY0ugujM>!QRK~)_9zf#Ysel}G
zlAXsz{&c&4SmVPX1cVb&)$!0aKbNa}){|`ef>60w886=v$I)~5cu!p4{4CmQG%N30
zs?_05Zbk0iak(U$xZu`~ug=G=lVGoTr<10?2jArrkpE{fQr-pr!BYI+`ukP@C;9*1
z4dnlSKgj<Vgr~dke^C$$!fpfqu>CI)LTdNg61z@FWS=2Ls}f|NJvI+Vha%1%K96~u
zcAcDm{u~T~u)~Iu{<FvGRdQuKmHf~AAMVl$aiFtgz+IA`q}1*8S<BKRuI1}tfpc-C
ze~E@70s;c(|6(5h8~Z~i^^f2?K-%ti-48hgvg?HZS3U}3GW=c4u!RMmqC-1QwXk9q
zh%_?)0Q$ijO?AfXK7X>{+M;YbzQ|O}6fM2d6uFu)wv(NcLmKliQcoED;5E8dRRSJ9
zA<%T0%16mUd~W7vO^Si(+_>vYwtLsuvV>#wyO}s8!oSkn9~y(T5J>Ui<8YlRp`c^0
zyshsSbAK19r2lvBXN`mYS65ok-TghH^hlYp+LzM0PDCXnzMGo8&LT3O^YHPBLh;qM
z*t9RX`}ip2Qub0gqSLYtD1UDYqWJwN=Uqe|?`!{X23@_hwDt7tJKqdvnY(luY4BYs
zc6s-AbL<h)b=t?rjKytGrd}pk42%T&ovmZr8oT@Z`?P#n`Oj=hkKEp#5|E)u>SNx3
zKJvdkCE0@G%gdgM|5=6j=X&`2U>CprhCXuG&y0^Js29i0ui?s-3LU+>6T9dOr()$-
zTn>2%^ero5q>{Eyhy)^-=}jle!dEe^Z`*<EJ*@u`tA_X8MtIrm6<-G@u>)5^&rZb8
zJbWV`kN!O(H`qP=jwdjgq8?p{uh%{vW8B;<83)}>OLYXEArH>iJa{Eeyp6HsA5Vkh
zibLajF7CQG|DgZgPteOJ>^(4{q#ds!mLpGkaHR~sh7gnJ5UzY>)BvnK{N<IrZzqvj
zeRCT8ovxesj{T2ANI&sbNQb&PBI)sjc#HAoECf9aT5mD;&V#r6-O>JC$}@9&)7G`M
z{)q1~IXUk`?}GLW2vAJ32X>pN>Fa0b<`y<_{6gI1_N@I_SP-)nYkl}V*U|zvNu^XU
z(T!TY-^SL;3;qC0`3;9>bW?rwrKz68lXmPmc3<t@VsF?ZjGn9O>Qs2yHxm)lW5QjY
zsMLHQY-mV4aPDlBhw<<yZO?hsTLhB{Yrg6=RZ(kOn=E12iJy2q+SUK^mfzoac`2k}
z;Z>HVG4E$zw_tsJ(F3ELoM%#OA09vEJ&66S{n!0{>@LEKMceWuw!0xHC^0#_r)K?I
zIdC+UAj-lH3zDp}9wf~bHVQI<pbWqej|nna=`RQbDcU$;JYAf1=z|_9CwDh)T?*CI
zA7J+zt%O__uE-&T>ROc&&hyTmHU=H4LOpd3GF~kPW*z4W%FD0pSg#hDI1|C*E=wEc
zOx{5Adw$vyBp8FlhEP@-k$XSC+Nl%?l9mW7%ogC`6rCUwh)1O`VLk*$?UVaktUCYT
zW}7ljG@w}tMC5x$O=BeW<0zqTR<U9DdK|lnY4?u9-yhumxwQtkU&(zxG+#M0{5xqW
z({Xb~@f8_bKksn25PD~Sckr+0`(M^tUoWqo(2Lr?2j5n;O-Qo)Z(iuyDYKtRj<=*y
zVF-jzrhJJW^P2xqob$@}&64P&zX~CQ3Q{6jFGyWOe)~%MY;9Xf5q2*-)VM%fL)nQ7
zk;X5E!8!VsBd#QPMOjE<<-aJSkW|rT!I8T$AG3jA_@XkNf6McW?hSM0TIR@hq2Xck
z;N<7b(L!qO^0Y`8GmAVPR2O0nioQ{ZYbVm9dOzoZn6Wl`^M3Bl-^K5DbGO5xcYOb9
z@*Le0@0>^B0<D30&1fDI+IsGGg8PR*x%J$TVFBzYs@B$4nexWJk;xVtuHSq7hl4y)
zBqIRm5gZ7}@^+o_F6!SprG=}C%3jH3>+}CQ{1Oq!nL)<@i638~lZ`}4j8d%+>78Cx
zb1mFj2Ac{VEqC8LJK4wZ4Ti~UPIP`fVcnNRTm}jMl-VjOkI}D7&c->}$va~FSh4`;
z7FrZ6+tLget<iht2k*3KfEpLX5FHZ}GNF&6=>skNuk;g}Y;R`M8gSKegYNjp@@mU{
zoh0X%1qhHzI|Oa8aT6^&&nn{SnzbMir7@6B#AO_eGF_I`k)+eAVVNJII?Bs1j);59
zjoTvkp4!-Syp^V!l^-B`c!p+;55cGuJ6K+a7jt{wALN=hvEuifAEFGU{!n+_-bpEt
zUq5a)gd*~kQG*$=)V~9>kLpIP@%yeL5_eWaSDj)6!hr?cYIc=k%<>&43*XWCRryY)
zG-#N`r!OUT^_emxgP-@%=6e$8KP3OeSrmR4GSsl$H-%4mM(?uK#162ql##}402AfQ
zv^&}Ic6MXoH-+n+-|rH-4nj9SOOW1;8g})(bnBE=yl($1s`z)%b93NG%IBfoV$5XN
zcjb}y9g7Y}(b8P+vd8%@AtTGk*EPQ{XPq{7g!SV_83sYI#a?OLx>VK4s-=Cjv9J^-
zaf~n*dVVa9DPw$^$&I4^m=AGrtFnv5*fPN9Zalvf)eR;I!fqJK+2#>0rX5(0Lh*Mg
zTbEcl_J?RwXJ=_O?`Z&$j%auiKqY{OHvR-dHSY2;O4@#5iN&$H#4P{?^6pgF!M6gH
zK!A!Y-uU>>wdhTQXpx#Mg-i%#k=L6UG72i*#jJ5J01)@?HTNDnbry{{?(bLISsi|V
z*KLAJx4cz>jNAVX+DcCXdHMLLEim-;Cz_9LN^17ow*AD<SGyy;o2VQYGml(*JX_@*
zF#-2oWl*|r9UYqyWm;B#co=7YVzHio{!A(EcM5>`EdGv0-M6*1SwnM{a|qpPYGaw}
z%($f<T^U|X$FL?@_=oG^GfLDPFSxpiegBpZjst<L;lY9N{F$k#WC#yHP+)TwPi@CN
z7Yrgxh_$m&HTT<I@VxNHweYDN210~J7v#M>p+|JoWplqrtufPZ_g2Xp`8x~`zuGIH
z7QxH(QF@2OhK={y+o#wUNiCMdC=Ta0o?OHi<kJpjGt@2Z1Ci$wNqEkAwfd=BfW$4o
z$2*w@ow!%CFJEheoLzmVUucuklbi2mh=bED)H62KBE35dgFh*r=q&1{6YAi%?GH5O
zS;I%=m5M(M9g!?OUoojzPRYWkl^Df=8x($&RgS%&Xn5tltedMbP!?`-lCI9_V4NRW
zK+ID)h`Lp_&XV@9*0!e(DBX*Y1H^K(<?MgOjh8ME82pU_@Mx<t2gB=Mp~@Eehj6HD
zw0rW9|5geoe6eR3qktH?RN8Vh^NMlYN|2970rDvL&?+{Z1G73B^hMeMh*PW_Z=C>?
zg{RP-cl6@MpfFx?cSGZ<m=KMu$L2}}f&%8_?}bNABNO~i5TjPnC+e4HyyTP$(x0OX
zwZSd}P2Ix`TJf*K#O)C3Kr?x~SZ($JgdPU^P4+WpXVo5+uoN10cG!0_6m}Q!t~%*Q
zX}dUaHMQg+X_W|2+=ntIQgZpvxr{X41rsn<*|vYmQ2*Y6QzQI>81GN&WywJ)!Bt}r
zi8sfZpyj~Ce%9fw^O_GXIX3O4f)IO$5*YIm^nE)|{FeC9Y7E&Ay9Pxyg&Hr!p8v53
z?e;MC)E(Up-Ckp9^kwFUwXab?R0A~(0$XC?xirod%o;n<6z$&#Te9lkx?`6uuKIXD
zKeINlu%KN_Z83axMUWt(&Y>HA<13t1>gad7J7GT9{X*vW@Q~0(9SVs?jSgLe&kREJ
zeLB=G{{48yQY;UrLE{!D0~q((6A#K-rekp2f}yTTO!7fn4zz0CB<4R@4E<<D4SeF>
zE{@aDaNJ=DBQHx!uW>21up1U2HJ$?$dWN1bi&fbq3Ywj6k+cvr(pbicrLzf0ZtfIY
zUAD*|8Z5NYJ;)x37qgvqy(f4n&dPvY@LXRWUo~w?H(+6^Q7MT0OKy`g0a>b`aZced
z9y*KGdr5GNGyfU6STj%Mg;-g9igk)0#51$`*F6PKDOS)BU&wf!nWj-Elzob(RziRP
z4C3_z<a<R3kw&oqGzg!&uyuXZ26M238rC27U$KVPAywC{r4^8qe9Yr563w)*X8w{|
zSeHoYFxCKgoNFt$<KlxfUaaqPKKe;OVd8B~4EPnaM<M<uY?qCLZ=9<<viNi4>gjAm
z40vi#jtNh4hA77efR{1)ilfJ0Z8rc80#07(VXfinCjq}X#EPt<l`olj)?shYNu2_-
zFE)pm;EOH~RN6YFpj;5R1dn-D?#DpEG8JnV0T$!n4o*2JTpGow`mQG*oWfGl7zyg7
zdWM8qJTIp3El*35OR)MzJb!X6y7QD_zW*!?$-DcBwVw5`I9G2yK3e;EhBrGppwm)k
zARfuT7_I}Ks=)3dC2w;rLaQ%(kE4wD!|G{N@J=87uPY~hM^|&YouUfR*#qH*HTh%=
zLwJL6QKNNN_DTtW_<`QX{A|CncB87ZviA5U(4dn17uC<?dQ&&o!Ha~McRrun7Ef;I
zl9U*T%WIQnJ43(Vzq}v#+l+zx&KUM4UThkqXcZWH%D~JP(_A)$)%^xuFn}L%f-#CS
zR4!v1*E$I7*6|4p4=b$%M3sH!8GEgDRw%cXxmQ?xhWp}yr`d6DABP^B1u*-Nw$W{4
zKvqA9nqeywoApM&Fx6a5Z%EH|ZwOU%u&ywc*Fs~7%8zP};&ra?zRRpo6v}j+OLdIU
z$^5y=tu;0P)O+Ch>&KM&;r(+mHYg2#^@PR5F+I!?lv3gPszlzd4r=TkAflu}UXPo&
z>XHu$6wmb|Oy}70D8lm&PO8Q%83k}Lkx)5;f1##1<CO0im?U6o;ULkh_xY;$&L%f7
z&BYiij|AoF*yn{r$`GJQew2J1;wCL0s~abr5M1QwqJ&_8U@w9TPrYZ?s@X+9@8if$
zwtl!^_(U5?L$N@G*QrJDl^mQyBAXIOQ-ENJeP>2H4pqu`A6?Iu!wNfrL3w7J*nUwp
zPF|YJxl@$)n&&7bzhlFYIW7OoqOU492=4wc3+XIrQZ4h~jFlwyHd&1zMu3m8B)V}s
zE(X#kRULp9%<KKN#HA)qPpx5MKj!cZKKqn4T8OBB135KiW&h&RTuGnah}W=Oglr1O
zw;e5z$M=eZHHGa*))mnXB_&nrgpdn#4@lAz3L`4}y_N(LDA~>~f-?A>o-*5tsc=RC
zT3){fLyUP1e9O}QxQ6-`Qgy)^UDSL;Pgl=nph!U2QK0MG>!+fd;z9x6?#K3ZTGvyH
zwmSi2V{w@yTK{m_CI`4QPdcJqjloJ@#YC4WE&51<Ra`xx#Fr8cTah~t_;X>oPKo*v
z(3k1F0u-rq4Fe7VYX1(EYXlp42UH`H%AfIs|Gehu@}OHZef0ReoZUz0c0>2VY4M1}
zR5)^oS^KcoY0*617(y{D#%XBi+@kb25=IwG5?7|>HqZ0YARa?keH=y>!M^*#ST!*~
zX{E)xO9sUh6rDz00k|BkUs!sTGYa-rRO@Y~r->lNIg)vwyWyk8|57)9dJi@8i$XQx
z6uH%r&rO(%s7tB4@UHKvWos31DEs2eg9Wyurl;?WZ(J@?m!@OQ7@33|<b4RY+q}W;
zBJj>pz)wmD2V>1b6K1iH?-vy-K;nAxTlnPUi{Rw(x#!GJfRsYYhDD;oPd*(%fY{w@
zqbVXP9Uz(AaMrpE+wEBjKL|{6l2GMXQxt=o0-TphqRdsZujp(Pav|!aBw#EOK6*fy
zVJ>fT53e{&qbxw7;*XM9-9Q)XoLI_d><f?)8wT|c<InG)5qfEWW!&EJ#VWmu3&8UU
zOx}{uvIu1!q|W+-^p7&oCw0!(eAWT#jIko59slO@Qky6k>3n98TX#z;8zsN>oixP0
zpdx|JzHj2{--=`{m3$J(5?&6BD|{UInLD{unL72xCoMTbRxcb?ZXtvaMapu7J*!);
zlEF8q)GLzaN_?P4AEn_EGyGRRd|$KUC+hC`KAIZlIx6N`Q=6e~|C^FPHbI&B+tTvn
zEO}X(z0uCNNr|*qmFV{lX{Q_!!CB8gk7Ds#DR)7dgwMm2v9G=1f4iUeKK!_PaN=d)
z+Ez&PND+7m-WF@9gdAkibIIz!y$fx+xWuHy!thaXz2ZL|Wm+@CabHT*Bn{;2V6+{i
z{Vi3@8Oiv~DbGBAa&uY3ex0~&v=NvTm(taOf}QQ|gBr5ewdpd`m{{#f_u8b%O!m$5
z#dSfI<T8NbS#_^iCDwStV&(R7bpcCW$mg<K6K$94RVRTgL8O>cARE^yb28pXi08W%
zq@kF`94pGsKeJ2^gFzSHd)kOI8B3N-8#dWYk2~jn=wwZxkdf+xe-Ok4oefJCD^`Ao
zsA=lu9iNqONe%hU8hWbo5e_l`v_sAEffT<)7&>=r6_yGEvMcjIIQ_X8;*~)pW!Q_m
zdhf_rH|o5|dkPoY^vHh0D;C6(E_i(eO|`sL6w;>+kdBVyCM2j|_sML9s)SKz$|nJh
z(_gs1r<STgNE{+n<1GOy!fYQ|46y~EZ8q^q&*_Cs2!$z-npEYQ1pt=vuhzJl3g;P{
zTPAUIY$M04DZ!s|TRt-EJi$@1nsqo^EOGBE#OK~&sA7Pn_3A@gYv}6$wAZ{VLdPQ@
z#AA+WUWiE2FKxKb{OrC?YTTEel5)ql#Vs?lxh9=2P0mTSSdziLd28ZjEt6%4$f$=X
z-DIpfxG{wayYm!?^E}Ex$R~<8akw(ZH|gL$@lQyPwXhQLzVN}MtW-vtx}DrOYkgN|
zx6+ukvQi(F7(=3vM*rX*HdbA8nIliuRpwMf!y9ZG{(6hriC@L5{Qv!R7k^uA>)(bE
zBG0INNMC8IQZD>Hr?@J05QHjhNV}h3v&0M9AJ!Ug$TsQ;u}mFdrx}zdr*f&Q-!QuQ
z!d(;}woa97s?~^`g>uleje9*<N7T|6#@|--nX1_la!fhkWu!Fn)=B7@<vcMWrRsb#
ztshXC<6sU$c2^pYPjbu|V|Mb?4ratX#G?dSr%W8k8Co*W)P~MZE7+*N`#BCH-Ll=(
ze?ZKtZAwtUFWeIeg?CXTH0BO8Xx5MN+oaSo(1-*2xue94kqjd_eIMsVOmlQmXSWOM
zq|!Oerm^%RPw;_GC)@U0pJybyokcG9XISSdKg;y=?LY-zBZ2%dmQO#SYJlgotPKr#
z0%~DD1GU%+6Vx4Ff@DiTh80l&*=P5NQ(&*!AFj*bqysaPw>g~mjb0$G9B8o?e(f)4
zLR$Lox95Z$3|PHPuT)6{0n4m%un&6$X+7gRAbB?SO>8d8&4PTlz`_L30w^rsM<ZoZ
zO#%5ys7*td@wv{R-nWm>ee5%sfq;M=89xCw{trzrC{u^HOx!1m;Lws)%rpS=(QO2l
z{QhmT)HYr()rfOQ%A}s$^MDpKe~#vfCi$q?6ACrl(N&Kum0PzjC2OhGTZ2pA^gor!
zqoaJ_bCpGIKzD=vnBPVF3IqN%1hcS7i6Dusvt%Oh#RGj(o_N7P{90%1e;wa`fqSgG
zRJEW#L)x)1hWqw;lMP~6UCh1iPObLUlaMtBerJ!=Wo2dRLwL+Mr$|bzJftmk>R?P$
zPS4thmc+^!;^X_Yii-97`^>2+-FlhFQ%p<*DEvanezO5D#B{B?M{LZ>_s49m-BXJI
zx|k9Zc<G+gClV$;hsx;2P3qf#CVga(_#y*vcW~OT>3=iMtI*VSv{#W9vcf+@jZj_^
z0g&jlu3c+4v^SZB96vX#&8-W{w$e*PMfU~a(7>lvGS=z3{a=w#Von5P?{>?pX9P$j
zE`>EWgFJ`Wvg&+Iua|Od8VcNKoGH^$WT?SGWf2ViR4hClCAjiB26Z=q78z~$`(5?%
z_w=1Dm&`eAL1fxg7!0w+H!mo~DPrPDj@S!8w*oA~h#{H8!U`=yenoHBXZ~~f<Tkx;
zv>a?yT}TE6KEmv@P5hY(^()Af%a38=5I=BaGS-O)FP><xcJfE&lFDo!FcQWZ3_vLF
zE8&MV?+pwqiwqBP;M=mXsG>3mgUR>!vH8hiF}W$gXa*295L#j;OX%IKGHIQ`ig(%7
z*T%q_5^1CyJB?oe8^4q+rHQiJeR-EGO=aN)RWUk<#R$5@qD0C7`^LY}vW&i2-6~fv
z-l?Es>pth@x%6<zN&~VQTNJOV?_<XD3uF{OV~Xz6clz@NX145achQ$q=;-^OE=Yk(
zIh1Xn(Y$ZN(84P8mcv7o#~-)*yE8m7==2a`d4JAmCHYS~w;jzfGu_AZGLe6Bnml;Q
z)5Bl*OY`Kxx7OC}!%LL-KkfSVC0B49i!k)pbWFp_2@qB<F4o!L^dux4*$v;_EtpUI
zC)OX(X(ySdi^85TUX3ZQi_AAucbA;3N&^1?sF==0a+4o3NEv;F8c#S8VcAiVt@8u9
z`r(;V1ZvZnK=I0jxc^Xe?U|#dBKk>*aTk3cR?tWRHg1E;uZ{t{C|UU=JuYE=h}KK)
z5DtC<4P(1aj-twxbG3J|9CCM;8^fL-M}<B*C7a7Tgr&aV*SUQ*YPuM)b^m^NdX!&t
zqCN9O@()e3S}Xde`WLbF-9dZyPG4o4WSMn2>04=2(^qd5l%lN*2`%KSTD&!q03_sC
z{C)#^ZRV67ypw@3#3YrhlATeYMJ7QRy}u#mEv(=9bW};7%@t=xvTR*Msa?gZI}H^o
zmoxKk-hN`cR52>BtzeXAG2v<y0kQ-mZmB#ZVQDmwDPO@hJ@415Wf9KmVG2X24Lq#u
z1bD0Zclshckt;L2Px3DXRDzR>F?{JJZC|l{_HL^M_??_pD+DB+?lrquPDJbZ>Ievn
zUVqgh3XvoQQnB$Duh;|lg)!r-ycD^TW2hz_xnG2BmzZNhKJ+HX4KbJgAoIr3svz%9
z@=5U>op?{rj}xi9cTrJcRBhn9SJib=UKy8~n%d6m?Zoq>hb%a!ZH@yLOABj~#Qo2y
zT+{>TMRLqIQmIotMT9&{`>&rv8DmL7DU~sX*_P)2QRDmVh#od)XIGCKL+QobjNM$*
zHFUfG+V=0hZ&$R>+zUbuabP6sM>VAl$Ey`9mpqh~8@bJynOhW~o-q0}my$16FhgC}
zQeFA3f5|ll7T3@qSW(gD1pm_w;eJk$<etc_dJfYa^~!9v4Qx<q9q5Z7d+*Y7Y#cS4
zG6P!=o-FBy`eg0=WYb4X+{*CP5Fzm30!r#a8jaxU3~p75Qh9%^CC0sMg?H&Zf;ARW
z3QmIy8)|Hs$FpQ{X}m(cyCk@=DH6tM1(Dr;8gnA@<bz=M<%va#i7No8GFA7oW0ltO
zAI>a2WSAP}WRS|<3AHwVtDimbkZT^SX7R2$i`7el^T4xhoJ3sZ7nrWe#RIE2vheqh
zuuF`Ia0$O@d8;>Chs-xPzPsf~!u>TRq9fV{B-nJb!c$WT^N)U(kI0o3`N=_6pHPBE
z70s%g2ET(0Dop&$X3jVSKbJ5f{+SL*F!UpMJq9O~rSv{B4w6%$d-^<VTxGLBr3^(i
z<tVO8qw9cDaim}uo(NCkEfK_ld_kmu*<oC8<*P&{ajsHnpcxf@1!R1yG{5Fu8wWLB
zxb(}2Nfnn;g=kRRXfvb<Br&8`{2MS)RsvRNLkT0~?p;O0NMH)%Xzz2Nd=LU$fC%1i
z-wlj^wXPX8b3@_CJ`wzwla6vEB9JUOw|JlZy^0_b_=FWQv=M{#pn2K1ElbHe+q`|`
z9ro?ZPb?!O&Nx6T>}o_<75{vY@PvEf;quYiWIRIa+@_j{le0{2AF>*}d4H!{FJ8nw
z^v3_lE;xQVdchhQuwVoKjwbEcSc_1Sd!=^1aWz6zwzE@?Nq;(8Z(eIW`dMWOTkhWw
zhj7|E1>Lz#|53Y!f(kAM$soVaqXEXLcTcwn0``oE>md<&-77o!8X2&nBPt_Z7u;;M
zgCX#_%D6!^$Gy$OQ*>IrA>r-hsO{WQeiS#=eTB^4Gp@R&BYuU!g?IdP5=2c0j?PJ`
z$OSXqTncrAB%NiZ;%@YZeK<@EPs#G?_c``dkSYp;T;RUBPc42_ARI@%cf;eO9E)*w
zZqXy1*CVg0g2~Kk(`HkjoJrG0-_^rjI^E{D80}XLB}PFzyI&O@KkgET48Ao-sSE1s
zF8kl`q!E{H8jBb;ZT8k+k^p520|t9$#9FS>+NvEagnC7i$B4RrT+dI2j3b-YSpS-`
zhK?g)EHNNOR0<ed=7SeA4^!(86D|AHABB_@CG7slFt1*qjsQ<1ub;KDh$KH-)F<uL
z4!LAAc$7NKn9+Lij#L`2m}Jv98sQI}h}_&!Q2#?Y>R+7fP#vR-pgF&c*jao#N!d3p
zxQ1!+X;a-XD2Y&{1YW3QFV@QI{yx;+ku+Gfq>l>s?yJ_Doo5WbpA(1FWy}plt@vA2
zMiN=DV>ZVqNi?<LVwz$Ba~luO4D!wEL%uG{ZG8RoZ$9hms)P#~*iqz4n_0ScetxdG
zH!VaoYGP1HT+$G;k=s@+CZW>j&5U(*hM09Mx}VfK{U+ru${;2tQBZ?^^N=w;OG-k5
zp^`{f{BJx$9se3lud#Hxy2kp*R9l|)F8NSWb0qV}*Xq{J&b76Om@jMIPwaNXvNnoc
zL5Ibfjdb2s1YGi@#RT1&yf@w7A1LbAIu@_rudr)bSr+E#sv=nEhHS)Khy6AXW5?0i
zMu-<Bk<<!*rS$gcZ>w-u<<HwmFE2a@q~IhATCoXMqQ__^E{~X|X|qW=x?3zPQgl=(
z1*QIGqWi#<j<~_+hd_v^SiiK8tNITatIT=Na{y<uG5Tu6^2b3t1lGyk{URk@32b8=
zJtXr}Z`Abmup;sG%ttc(2F}-+B^#a5rfuVscw={aFX^sL-NUK|2J~OF+W6G7#H)-n
zacfOL%f8~~&YXC9{56*Jx-$ODc=GtrA@U$3qnI0f%24SkqomQ~v<r<nqF?Gg`<1Gt
zOHWhPUIG9@W!&=v!&25q$Lu##!(icG*knppN&3bKDN_HNhT_5s8s(j0IrB9<Q_q>n
zt6feb_63hB<bZ<#Kb3I)aH|gK{l{?6TB!b9%@)U+>6>z|O#)3&GAy+cM7kp(SsYLz
zy%xdQ+x?(g-tEac!*XmOnG<3uS*#!x{h@!Ovri$(fZEt#wNnl&S(dp(%fgEpo3$i*
zkr=s}2rg+glU&DqkX-gv|Bb4sOeSfBU3RIJah&Y!LE<4W@C$tj<!Jr=Fu>`?@y|%$
zpj(9)Y(`>gZ*#P)(#7;eOpp1-((v?*N9bD<)E*p+<${L27B(biWcxI&CU!U?$dP*P
zYLz6%sEjKA#ina<!7E%GEj`ghKjSgB*XMsvXI*?mGy8fjJ3G5j{~d;RcLUnRxY_^|
z!DyI(T(bK(nKHAtXRf_?EFz@Qw>8t#KDR6++~2mod3adkWvSYI6>p!NV;PjEcelNb
zQ#2L$Ag9IDpf4ke6CSpV{eCP_D2Br4P%HzLFZ?OIau_?6W%33GRoq0RakAI=b`4Z^
zM|kALwFmgQ7T(jGTQ~sL4|B#aUwSLpZT&TWT&y2()YT_K(NdF68cUhm)C4cwyqwm4
zEqtudGq}fE+6SpX4eU$1yyVriDEM9WpX<u6yJ%^qDQw<=MVr3nPK7=8&RKD2Am(+P
zjnAg@b#mpMm|N&6eSH<1<%rQ>^i}mnch5^_GMSe^Wa>F8{yh9!Q_yiYrQcQe%k)PS
zU+1}C$Mq`zzlCGBv&%p?76EMsJ;G_D`&gylUUs!3kgdqQL9D7sv%+e-`d7C;Z{8P%
z!VvihZ1V0uF_6tSDh3OIjmjw_k^p$kt5#Wsm}IUaMLD*CXJON7k-z$sfk-Mk=F$)#
zE78yR=?i8S)e^-nCag}4rnGS}*^_bGUVL|E>w3L!EjjGEG}ZOfoebi$K42Sg19hhp
zpqoTNK$lQXwSR%3($wq}z9od6Ly_<saslZ)V&Z6ZD<IexNmdDqV}T6Gr`*_aabD(f
z4;!m*n;l@V3o_I&)PFXuskH>^DerM3C&M)ZUX^D_WYH$?b+KSDv#ExchbCxx=Q{;f
zHL1skTSFhZSo{;tMy2`5r*6PSkkTpd2jF~_TZSg5`p{m?9~Lx^tV(Ia=2zVRuwV0E
zejud4{_oEzZrO*YSM67N_sHv+u=i-5jL{5@ZC{zS`#V9|glw}MB457wla|99QwH&%
z?C_U^kymw6?g0UDgY}mS=v*`<pHX)9D_RS|;_&Yx)D44Jb3RWUeFax$={EdTz@Kj6
zWT2RtAwp5xP~z(Ajyo;h#}WtbK2KFAdgZXAH&40{9HDN<a6&(dQ$z7G(yLt8^i6oO
zcc+rkU}U(zq>LD%dus7Tc7AoDA692eq~v(5Y4eHeUcwqy&y%XAr&$-~z*CrdIr??k
zT+oMLXppW<!UPT`?5~k?k|QCwFD+euUG=?1!IC>Y$M7k24SA(B5KL(M+nN%H34{1O
zT@o*&yHCjA)milSi<`f4G;;!ZxeeJ>+Ok$IF_du&he2RrSqMk=sHT5i+dACimoIun
zUD7g&^ZVqchpecVS=P;f1zFH_6X%y$3@jJ7vrlr@+DnfPtbIr+X~1dlk2}2+07<}s
z2XkRgzk@=8N?Im?%ABe7Dfnu#c)V;|y6BV_7^nuq$V*xN!W6?mg4Cqu&ml#T`nLZE
z#F$XJZ{vg}k&<RWrpMJtthqvzp3j<we;n*Ip^em5W|J(QO4fsxl!q6+xkiA&6p$@}
zk+N^AcoDkI7ZfF4L!PR+lwgSewp4z_dvK1<_EhmXwI;8s<EA@}4vUKSMHSPSDpM%~
zK8bdQdd+M`Z$;V{mWB%fJRnMa797$4naX0LpSgEnt*+l>*xElgpw}jI&nn0E3hg}7
zUSlh1c`Be}Fh{QX#7h$UKFpjc>oXeVRKZ5LijR*+>;wUrSgNmt-oLPQhfJ7we0d>H
zk}*QzYaBabr$LBph*v9orlMNpV?I37)9+i`_x^v<oM)3P@rmQgjhUsCy=0cIrhP<C
z+}jp&n|ki@p=iB>zjd6y2KxHQ`D`qj&>miaBMS?6Ia2e)$MVF;EZ^DRdMb5c&1%;>
zLjeMlt|d4LqmCAOPvUY0(t&TFG+M<~TjuS9G$S<oTA8rhe|<t1E@qonP{gF#H#_Th
z0N^4mk$>V;$Qy;GW74c4ul!{=IOmr)0<)$y<V{xcAEF20n$uyc^lqKE9Br?tkYB-L
z;ND2etQ&rfBv-<jzXKfYYlr88W)>2ge-3VdBA$VN2oG2-{Wb`yG(Y>1`@<Q&p1diR
za!;}R;2u;LJX9}kRk+h}zN_1)_*4j8@w`Ucr>i$Ctkgp#dMcvb7xTx6Z0XKZN|kIg
zOGaa%$)E7!%Xm%9$VG#K_U;E5pE;kTw4c`jzUvXk0&yEgi<9F-gfn8PWR$#fe*?%d
zNK%sYfYtyCwkCaLt}-jBERexynRy$i=QnOF3OO3}#<-dkT3%qL+_#^8ECqdixUO<V
zg|?O>CP^52HN=+m@&w7tGi$z52-nm}Ycwa2S(nvXGC=jA?z|Jpa@8HD-x9eAL})wW
zC;`HLN~e*hi)+2>1B_k?xo9lSDwj-QR6^yE7ISa$vEMn2f(HNLBirsn|Fh|srHSTv
z%P4N`5G~G-1hW0Rrk=z+?jQ_aSH|A0ko0^Xa&5(^4Cwe4BI7F1I2cV<vU7cpeUO!f
z!UM6y#-lm}5h!WXC+#)v;)2<S{plsGTd5Y43Qewh$KsbaZERn*1L|S=vMRw>lV{Ut
zC89;Y8xH4Vw1JC!X~0s!ajxE3-^b@4=*w!eZZP)p_wXqb@>)JyzY~lbjNF5(0?I9}
z<Y@7TV_4g!Uuje%z9fvgF#ANKt9<={Gj|0iMLZd!>I*lW{o{tAjpIghf?246atuEH
z?(tTM)CP<kd(oY*opSR7Rp0`=#^88L$28@-G)G=Bsh7x--slg8$UkQYiAJFqoX*TT
z(KoDFwnB$rYkHb_ZU<)0NzCm%x|qoXDad5ZkM}x@nu!c$3&T{>_+Q$g79wZh-QD`*
zs@F4ND*+F*&MSdQTAhC(FJdIBYjl{C_YwxZWh67sch52ATJ(+*eU~42<&~zx_=2|8
zI$Hnm+;8dD!*u1t>c_>h{xep91qlIq$Mwidw3M!pT3F$zT3=q{QYFYh_*>F$09Zi5
zR$>A#8#6lmHO$&O4f;Lijv6)~#0UT4ahapl)32A3h{Mv&Szar@>ikZR48|lY4TK1M
zQm0Z-&HJ`{*4b|opF(3ZnZUpY(_q6QUQW#-WoPU38OVRd!b+nxchh4{i&N<@Q<_Th
zu2cXx{*5B+ZK$$=f}$OQ$+2CeQ;!)r#Dpq*<u8zTnO5WWC6iThe)=;zd8ZZzrZUEd
z##qg=rfEw#0hwUxH=F&$EX!gQ8_1=Q;T|QeXY;NtM(75kT+TOC%%XI}k*ia@A<7;u
zvre9=f?1;{H7yIvEH>&yBN6}kwWE`_sB44wiMsw~eUD|ahI85AW$ZJ|`q#MC+{KQC
z=5oJ8@Rtbqwi6A{P;Uucd0;Vq3qA|40;ls*$YZdmJ`tv)ftq8yh%;NV)#qs6a!dHD
zT=7}q_9GtqGRpMNVL!#1(L@(_DTUlIdbX-18arN3onh~q`|1{Q7ngDZ`_;%<D>FA1
zd-N2;qECE0DO#+pnNvA7gXSDo6h~+4xj#E$kmeZ4Eof$L4r@(&Z)uYSHC8#;ly$O|
z*C6!Ix_^m9;^BT<BexxQJ+Ffj%k$rQZpy}2!C=c)bL0H_jULLZhM_=RbB87|hTdfc
z1}6(*{Nkd($OF|h=1w!Sbo=_%7TPYa)_1dfCqis!IA2o|Pd_}!UfCdM$EjQ8d)u|v
z*%Fd^+`KVJDW%3ryEso0H*h!eQrx6-sGwrtlcGJfoyDt+(Nh`|ljf1krI7ef4D=G!
zEq=5UWKn~{0j1LZ#U$xwYmxbq^PB&@IUkfED5Q{QPd2fp`%-}0OH33lnf!|xjKtsf
zRCXbfpbtldQmRh+nwumC(e&c1QWM@4Dwkz}qb13t5Gs=xA{m?DYDcsEI0}~asvhGp
zxJciwvuK}Db^}5*TQjJQ@dT42uQ*e+5zIa@G=6jR_JSV3Z(Ip38ayZD`?<PnK4n0}
z8DG5XD3`=k_|XcQF+mh3$TK-EeB|>k9rT<w)uZT!!B+RWyZyr?#v1ldaPMj#J4rs3
zS+rEhGac2nz~T1|n$0)ziTZtywxLSu?{77H;36uM|31?%&o-M`bO>DTDrIi)=MZX+
zYC~si@0$ybOf)q|fjv70@RS*Dg^nw2O#WY|n`axKgy^ZBkoGX35l<SQDUTT4dm)$p
z)Qk8v@?yp9(u!8m@87aMF8e--Ej(rR_HYg@jY#IY6<2Wlp0>Bve$@!?!sqPV7x)6t
z4@QW5(5HJ!Nm7$em#mshSlSOC#A7dMlk_qVzyO#0-gQ(#HWl1TTOn+SjrZppH_845
zX_4X9Z<$}yDZHWd8Xm}{z&UUI5pVtaAa~W3z<?cxW+m%WqXQWkV=AHW;FAnEb~;}0
zlj<hB8P|NYaLBbCAMVm%`>#rlHtdpt2Q;-1`OD{b+0JiH`^dMB{~?dA(`>xp0U?RR
zL@_;?ZDK&{vG)&>CU)XM=aDm)&)qD(r@ntP^!A4Sp(pHh$T~sC+5SOeH-DbG@o4E|
z(aJ4*$X(X=A-$b%*UuC$rY!$Xfl?v=?xb$fmp<4m5Ps-Ni5P%oTsX1jBNj#gl3!{9
z6lG^FwrDOj<R=9I<m2wi4}G~s$i0==rSf2AK`{-(R+C|*?A|#=t^v|)547xOEQPi-
zzK78GgMi;vY%c}t(>-s^0rIbnvJi+#Qfh41x`|{HGHZccn(XmCXDD9#>X~_jH%rN!
z-`+kgQdY;#t1wZXWVfT}Ah2-|wxZa0*~~hHa{!eE&RhSK*YN1-d+{7$B_yw-`OPG^
zMAsMr^4F{)KFp+-47l%N|MkOL#Qv#tI-WY2c&+h{;!y6LGa-bYGibjtktO5%yP|PB
z@FhAaG?SS5ZGAQntNjQEWe3|vr4n-zyNa|7&f2DbI;-zKxje)WRKL7n&S5(5khi0j
z+{&CkYR<0G1gw=vgu=zng6_k9#!)|T{d}8#H#ta|U-9CQ_)QHjBQP3nJJ+PwFJWkC
z2+q;7>O6vdwCR$z-@E>a>-uC|X~-5})Z}HZLst#fC(H~EzO4tK(ZViM?mkz$tisut
zE>mZ_Ya8%-DvXoPhZm0nAOi&5oDxdl(!|<4kFAAS0OBPndMOGqBG0ro7+03DY&m?5
zn^_@u;$?E+e_K&lWm+>z(W{bwVt(EY5%Lf%J)RA$2=23%WX4jmQeRZ)E4~?RD!ds?
z3;05|4vqI7Z=vUNwWvdIs#EGxRQh|qo<XW_R}B$ROnKG3KB-^&zWNPql|<qFV`}Oy
zbLXH1ubKJs$usZkf3tvhS5@p{%J7>nBTX%6Zep6e&wI2e=JWKjt{qJ^oBdXt{aEgr
zPxR82I)|D?d(e-gdzu3BE?)$Ax4I9ah5!;o?fy8Tx4>%j%4e+^PS+rcDWTRk%A>+s
z_IGx(8ZcFYSq`X)RTGmG=?37J)GZU6Pf$emMoH_?pe`Kc8z<+}#hW5Yed8GBOr7mj
zh<2>&u<RWucaA+NEfRa_**5sEm>Yv*5*bqMsj87&SCz`l6O+kV;mg-T%_=M@l5Yv+
zk#W1#)p-20cJY4u+tLQ``CW*uVof|eVcwUCO#xZ-=_k>N!mRytLFq|5oTnCKtCL?S
zU75f;3y|{89hOFwLv&d1s<q0wa|ppUdf<#q5h>pC+$kM>rNo)Eg~!Ofps;VIjc*o9
zZNA>trH9O)JWC%*VNq2bJoEdpF%>zm7Wc690>XZ=kTUW<OZ->y`sHQ#!r;rd24;s|
z#g8TeQFM0kx2sR+^}CEf-GXheme|?V4c9a!rTZZLPlyscy^2Z~cHu4>{vmZr5V&6R
zwa)b&FJJAXFXt~%a>bL^Z_pfCc@pUA<|Za3^<7ZUIWxL8^ysz!Ez?5z<LWcnp{PEq
zGlJma3#3w+*ZEyG8cj)eO%yW<%QfQQ)D3cX$T7Hm7g~IiYiKE9qpw-lR00*z>t`q;
zYS++M3MYMz7NSeyfez;r&$nsF?GAGr<3;|kYcG1wT=F}+06jx@kl8R3SM>^X5kA-<
zuOZxi^<v4$ZVR7M;`NMf*PnH#`M=VaFI)(>s+a&OT{U;dZhvTo_r&7QQ8q1oEj$ld
zrn6T>jMFQ&UVl&>+3x;#ky;WV*e;-wlqx&ytl+-(vE`laK32(pi)DA8+(JI}muWee
zjXjK6oDL0^x#PwIBm%iAeyvwZQ`jr>mDjwqW{8!X1Un4p3Ve1DWRQ|niKL2&{*G$6
zVY~Pt{+cdexTa}y)TaK!ppY+)ru8cl!spfO8I$FEqYqZO%A_6M8qlxHGbZqgc&vj|
z+_VW_2IFA2Gu2WBPSrw!Z_C7jbE(BfdqSkM{U46#S_iGtfBPT4xUAv8N&;b&T)AgS
z>mKbjap>N?;#*n^j7drhcGz9SiDesBwM&Hrq_5>>=vCD$3fZi~{Z*GlL>aYZJ)GY)
zzCzuhYMi7M7gzCkefkO55$4bl0x}W-Ykj5{?{66?OH%xBv0x4q@%o+^2ok5j>j(OV
z1&|GFsNBUg+t-Yc;7qNjyC<Z7kjp29083+P+*|lFCGl|$C5wY@ooFa#PVFmvgH|~}
zUPq4}tgPDYwQ~6r;XugCi#*z<IH9Mke2Fv6+l)3foAU?>;YV<@JoM!LaUI^;k}FoF
zIC7b)N0(9Wa;PcX)Ht<vhWxVjAM(zfwByH`J)oty5PKs3BdMM|D?wYYLn3NJj-0B}
zq%a>5ia+7Ym>EeRT@5s}UVY;BgCIVBPwD==|CqgI*_YT46mbp%(5N9S+nQ%%u7+O!
z9XebSy@RcGaEb^gcYU)y#UJ|kOI3+c{3o>3c65g!udtN81&!IQ(3mXLAKlfE9;*sH
zmE|`E?W&iQzq0@ej1^euKCmj{1x*78rc`1t>MCDU`O<oW-~n+0;WY=R!D9iTPo&;T
z0s5rQC+dLv#Qgj4QOF3~r0yV7=yyoFO@ryhZZ}zBWLx>5XV1H|L*k*J(S-%FLYIjp
zp0m|?`lM`L!*9zC=-M+V^$H2bJxhNJva6o)pQ+UOLteU>oxXCESFv=d;-(Oeff|(2
zF>u&SB~E<<V;7t{ixF#3Zn4(ce?7<``OB)x>C_h|q-T+T`gN~%s<LDDo=_1tq0+(0
zIl4Qr!)w>943vsx3xd5g`&<RJR$n@Cu_P;!V~UmZ55mhRzHDysbDH>&MlI>5ryQc8
z5!$=RTizv`|BlN94Z9Z@0jO$aHM@+<5;dhw`@1!=2+}~vO(^AJg7^L0pfMEqtb!eC
zUTbp7!c6*HbSggA9%suN9STZvE9mY?HayH1{SwTg>XKTFW~b9L$U9HuZ#>+=%c~Ku
zViZ*>%7{eE=RdD{XaRUwXw+(Vh_GAtweUz+Uq4Ylx&<$UDwKy^1FdbSRKp;tDkCGU
zTjm(EW3Q5<kZ|cTY}c@Mc}OTo<q18RKk-%P)J+(iRGBk&b*qJxv2QKuDTp4nfNj!$
z<LG=q><KWpcBpzSM-T_yw6CmHTHbNB@YSSNBr{S-vj}x=jbFhtWR0Q{>cC#Eo&BG&
z7lIUhJK3k{+kLP(j4>%&?K~_d-{)3Tx0<_wJAS#Dn%QPZirPx%yv^xboYpQW{$aq=
z&7S3$QDSyvEwc7_9$wwF``@^!mY#~B1P>u+nU^dwb;Qftaa^a20f_0-uvR^i>kK>J
z{51C^j@9&vke=0iX|=j%lgYGy`x^Tr*KW4W3In(YgW6hvCyU~8_Po0GAL&->npW`H
z*je|H?)`lIinBw~1B#XnSXFWmn1JlK-^5d>tsDK?<*(?S`I_Mm)>v*ArashhK_e+c
ztpg-9msXzyCyS!$tdrCcWpvE$2g{UVLBn+%eGggE1$x>)xCB&PfiortQ~};r$A`R7
zbdioctw*(7WdSZ6>>7&tOoo9Zg_|VDiOV@g(5b(`rH8-(=jxm-W;`US%u(O6J~4!s
zIlf>5DHtbbBoz7hO3*!@boM5}o1KiNIT4q23Q`}FlMkbrRMGI++`e|CQHsw$=8}&r
zm8{Fo`wvpFp8Rgu^U0@@U;Z)^I&P|F#o-_{YwAuicL&EyX;d!v=f~w{d}x6u$T;Qf
z)5J*3Uc*}p%AM|;ZJ=45zi;c7T#*>t7q%(%1W4x@xwlPF8N%nb?dbd_yJoYJb+3fO
z_olA75=Bm2AsdE&d0Q$I8!~LXs3d*aJ!WC_#MwD{oK|v#^IbdJ_t?WzWc39F{8(cF
zfy36V#bC~I%YIYWTFX}uRf<Y7YxNDU*J$BiV1sxv0AX1i5C6}*yRD^%x|L%w16&LL
zbDCu5<CVY@iI6SI(2SdvU`p%$cS!E-&|SRZrriI;B6a905>+Av#^z66X4*vLQhZC_
zrt<ZPK_zPm$tu*BbAV-d*$N5Z5zt;}%n+7B#@9)*To=~`=S0Uj)*rrU=_QFX>T2@2
zYb8JNl60b*_rC%%XP36L&C_%;3D=|vVv%v%?!DLL&tlV*OEN~wdb_U-1_pxkVFDWw
zC*lfy(#eUx#Gs!;o+UEG+s+y8njzJ?<9)k&+|sjcOs(26m)h3*?0R-W&&%$9MvT=#
zC7vI&S}?YDE&<{B%O~^VfvU9_mVQU^dNYYx=$$>(G_EWkQZOtLqbYIV7ju#oAyXbx
z5aLpVsX_T949Gu4$XMW`B1&p~s~z_)CcH)UU#u09QnJLJE#j(0Q_?hp--!9klXV4C
zk%O#DIrCpZB-23m5(Q9}rT6X26FIIzt|&c~KkPBsM#>qWH^JxAupw#nKct@NGrVj)
zRX~lY_l^~B)j8sZ$^w)TFXVe)n4g=dna$NNSr+Ljq74zr_|3*?x|`7*1AFMi<$$)X
zhgaBh&GTJ0^+r~WrRZwi9{}H9u)WUu^19I^t5Syt21X29owvb1VuWyRjTi^&t$~DW
zbIV=-8-h`1u#Bi4T&uJFYE*z?M!@AfMd|#{ZwK8461-i8kFKtAH14~II)c;B8g(3P
zhDlE7$MPgLHV+f(vwQN7poh&Mg{L7y4f-SgI<zxF=M{Y>QB-7)Rg72K)Wu6G=y+D!
z&#ScfVcbD%DjD;mg_=3>pi95qN)_|HC)w%glD&0e{MOpJJs0|C&i`cTX5Q7!|7a~G
z)poz@Ux8~!+pEn~Iv1gu;{idr82e7Kw=Kejd$F^V{fJkY>(0?ff%m~tqK#A~9@Qnn
zL_3`9-NucPK^0<?XS4C96%`zD14UXnMT^QekJ2Dg2{qLhL0U&|Oz1?u6;I+L@5fUh
zG9o#3>FWH|h2^yCLLc#6UJJt%wm+*CHj1|g<EA>w5?=DB#4G1ICbD=s^9*lgm-<Go
z4mNGhAC|YXP6Q@i<vO!y>#FFbm1ViX3}1ZAFLw%eDsPMP&66xyT$+5U+d3|?pV9>v
zZ5VWN<Yv<@Q5&@H^+lz7aF87Rhekf%G2nhRW0Tr=8lIECWV~qOYriUTj`uWqB?&GM
zTD&44`sj06Ew@)-^Pf^<+!vKJI^Ak^lO%y2fMH(MN^k-b%utlea~E2r5pQllB~XO-
zQKRUsPfBCJ!CA31F!~3xLS`~$FxOu@lcMCvWis2vceu1b=6r#Vli|xgy^@6WV|pt{
zQbfwKfAP3{21fuF1Tirc7|&$giAkNzQ8E-@N#?9@$&dX3=-xf)omN_|Bm2}U`It|%
zCO!qZQK_Cq>!Uw^zD6|Lm9^-B4y@plv_GEY`a=?5Z0P-=Sm?>g-{7=i_KU6Tt)T8c
z{r&Did&gWIkP3N|S++LCln@)2qD6;a^rT8N9p2OEG2{E}ygW{_9$<9aPvNO2s?=HW
zb`^86s87U)%Z!zI2wozP_tmUsaxZKkq@~|KYoCYtE}K3HH?d!CUp46iqHzS-?D|C0
zE}W1!W@TR8M_)|Eckggnn~%A+Y+P=7YbTXl?!Ir?-n<l&cn^4*nFu+f42^XQKFD%@
z-6AfTcfOdaR;5)o$QXd3qe>D<g_9)DQNj$CL0>7e^DD01OKR-2NoWsG4<}RiG=d$7
zdC+Yvqu2skv#dJZE^}W^-MTrWJ~h1%`-G~bYpG;b4kx5Lh%U!UmHDE}thwd*qW<)`
zIZZgi@4#W7S6F8+ZrKM+Bq`b^Nxl>8uy?S@WDLSK&Om<N6Lkxgr^1t?bU)$J`@D^@
zYZW3(#vPTAT+;Dw?n>L?MX&+@*|PkXNMvlq=HW0{(&&4Xp5yMQb!(xogzU>@zdyuh
ztB=exA##!s2VbS|BxTyzc7TAj4UHrEAC=`ja^>H^G@Zt`hrpEe#ZpcT6Z~u~>$sYm
zf!RWVC{h(|61iT5JZCu+b2XfVkJaaMl}0h$J4RBy8U!zovoi8fLH1h%B8k74lU$`Q
znE9=dbBbYV4EWST6fYj0#1by8nsiZ)NrTj6d^fN00D$!ppS--c`Xx_w-CfSd+Ql7!
zuzGR-wrt%dzlCV-fO7gl!r1XkCT4X5h@*_6JdBTzL;e2)#y~m0Co_86r7rCuHADpL
z&{QbWW4#>G^l=C=QrBtv>h;Yv&tE*_!NUg>MKNsXB6VG3tz&XNOFA|osH+Owb<EGs
z2R&3wgPIU57puX0Qc@kZ$xmxqGMP;I^iTX@oa>pClJ;=J2hVN^AyO1MXQ!t_FHkDr
zZ}(eH&Q9TFKZ<i44+rM+G}<;2@qN!`bIY`xVX}O%vGMHpJBq9zAMEA{k;8UJmgUSB
z%RwoYEKP#<EEY>38AyKn+u!D`-+Ghv^$kWVuC6XQbUAgG-q#=62hz4ZS!?NIV4*dS
zFVFB|m<K?l(ZjJ=2z=-0_bvKjN;WN+O{avT=i$XElU2#}`p9N|pqLlrvs4J1FDLlM
zvE8M@nLa6K@faC#t!Ml6$fHM>oX=*wXj-C(<ns)BbZ9MEEGINg%dziR=7LFHU^0oB
z8M187=Ea&&cPu9*^Rqe4v1NOx_|%gN9=!I5x@$0nL1l`%t_ZCsD>TP;%cPj2C*Xo5
z<{)Mn)3XUO8$7R;CG(RhoBa-4$D7~yRrc2v);Zejny-KGoZe3Om;U6xz<e=9Pc*N-
z_6E;xpW?N^%vVf`B{9A^bb2ONRfi{ph_j=H_mhOP0oMoc0tFN^J-D*=NCcMTnk$B}
zw7vCYGWB2>WAMF0Yek-C_z>w@3qmBfS0TuBs#$bZhYTQ$BC6z+E26}<4(XsY8C~>u
zo!d&L*%>BFZo9p;C==0IqX)&9DKxfA1!bi*T5FWi)Q1YG73Hi<Mv<;ZsbnE9th?)W
zwxL66K5XdIX<V){DP@vo1R1Qg#IC3B8lFD=9!=GfFHWfJmff!A)rYT979~fmn9b)y
z{U}L&v%ElTkEz%{yys!a`~Ns{cmacgO(i?07dM03ZL*(nHq}PX&d%-%*lpXe-mY1#
zPRNW&#r+}8c5(S2E%*d>)sf9^OYrcm?>t48Pq@5T5~D}P4t3QKvp4zfx4y~Ue2|z<
zWxLIW>3l(1Oo<^C-`BUPw_`e648{7uX1hkKG)T(Fv{BgawrHg|Jv|>Bpd;JOnyzcP
zeDHukGJ!lE_Jk04@Zb?brrPaSzw%Xn`Iml?58i#3x@kDQc#Y@BB~^9gbHg6@hn_%u
z^hw3w9h2FF#eB+I^c2gI{RcbxU5ySQxrYjgHj=)x^b&d-C`?IV3?eGl*ITY%9JzXB
z$?4e=tD#&>$yG+%_e_cju}Y0KqEJW|=o^dc!3BlNG(szKqY)(_1jQ^z45@+Xv?P-X
z<pqjJu`Jl#s&w=$GKwsxR~9AH7uH8lkdbJTTrpTn5Rv)ACFL?F=rH)!1fl`C(3q8h
z3^JuDiwVos6rH6LloyK7x#WbZEnU{|-qUya`t2pP-Eyf{eDCJlG%oV+@#|D7GU-6*
ze}FKOY+^7*)An7`d5_tvPYeR^fEzr<l}u{G04Occ3hK_I1Ze$X%|ejUKraV8X_?eO
zt`BsLCB}%!4Z@^(<t8_{5K;{zB%Ams)BUBfscGV1l0Bwzv@biArIN`?M+&r7cY%cv
z(sV?d#tjJ4;wlC|u9sQXC?YZGfDuDsSLPCv58dH|-AIhdu0!PtblQ9>rO3)O1{)&L
zg>=24Gdd_oXB&3ihR!u;3B7CZ1j=$k=XwfLp!6-)4)qu<v2K{}{$qVwk&oVDv)SOC
zLx>bm>bhc*mrTpqU4ehQStA6PA}0`u!BZa&si<8P>AlCZxxFUK3}5+;Z}aW%zDJCK
z#}6Okyr(Dxi+RDxqClu&e#s!^J|2$*0?YZ5p+SV|xTm)b%aez9f%$OQ;l1Ph?CS1&
zS5-&q<AKYoM?>%kbZtl5)T~a=(x_x2UDL4N?|Ah130Yp?hpB{L`|4M?xxS`p8YZ(9
z#blA{XCIjp{6qUd#LKB;A+nris7!HNbsVK75TJFK(r~gWaWT?{7Ak=<Q@Ue?w}IZz
zm{@_`H*`(UtFJ%c@#Q(SiL5Tqut!UL(DT~qBc8o@jxG(wY|8QWNL{xa>xNy^QRF3g
z-qN={&9S4L<t!epSf0#SO=jeaW%3I00oyzBsYK^skIC*J23XGLTs$}<&ke_FPtYmQ
zOixN;<B1x~dB)_t1O&dBP&YLJhl`HZ_Y@{47J`#UOXm3m+ohf5$t!cDfScz#q){yL
z8I$#%UHdjoeIP8S++2T`n3ohH@Xq&s6Kx$vOS<53Dj<!bzO^{#(b*V5J}E>%`H-Bd
zViI|1BU5v%2sjbJr7piRPtB=a^f(t$D57^bAJ9=?Vn7VN9z|i8O-j6os4`3Gjy{kZ
zgAmXPfzFakuXi5lBb{@IEOooP;F578kd+xG*BF^Bd7@w#{FKHq3zfc)agYidrOAsd
zbzOUp!y}Z$<Qk7lg?}xPg+eQh@FWTEU=QL$K#SA_DYPUnGn%SP<Ecx9GK&6XdQWRi
zp@_6t+YBOu1{WMkNxH74YYxfWPY65_lq7V1m<0Iagpib1yR%5YzM<_~%5u(byP>lk
zj~+b+Dek1b&N&vVljMX9>^9ppNp-eJ={X_TZZ^m$De{u{-u^AtFWygMWxEZZ`OK$y
zy)b-fb;i5TulQC4^70u<Nb06yf7o+=o<`de!{F{)adLWgXRlJ%HM@N(e7~GG(snh+
z!-3_=X-aWL=-P($`j*wnY4W)W$t8M!+;M(!Nl}yp?|JtAd;H=r{}S)L_bxX#H_R5N
z%$E<4$(Ri(^-zfaVKdQVQbidhi9RyVGr+OiH?;d6p~2)yi4i4eCGbtMkPfcr=C&h~
zlF)Vp5!v7Dkuvgdwd6rr(wB;?oblHC@6gsYld_<gOpwM<%qBG322JF0NE?NrGq-M9
zj)x{GsLl#v4*9BNezqjniso=2?jrN^3DvQ}buB7aEKiq6B&QcE&K4&`ZScY~pO@?}
zmmIf8oOfhGlTUJb=O`%&(PLYSuRSJea-D&a2o0F@Weq|gv>-Q<!e|s}oTj#lunYJ)
z6%lRUva2`r&6@i5efp{;FC@oJ$MMiJEizo^u}zOE#{lw4LXa#-gb?IPB1JOH>-sR9
zB;?D3!AAx+MIa=5cNYnnf}pXzCHO#|8A`43A|RE)I*+rKNtPkC0+E{0#Ng_XX9}eS
zeo*g*LGCF8m{NlTX(T#Jv(<@_s;krZT?MXBPfQz4o@MlT1}>tdfsh8Fm1Ra7(+rF#
zLMqs*p^rsqthGZuB+dPnQl_1;Oc5&r$fP9FYM7G|NJls;(jci2lPj_7I5rk<8*JAN
zSAxFtltr?y4=&tIGyHg;(}(|0AgHREs;<cL2~AzISgr;;kK~DUJRX@&r)aH)$L{Ic
zhKs8QNjDWEecy379ynR8K%_~6-ddatY&H$+_qY6+SN{#({<*(Mb`tnA*<a>2{9h#m
zPgPYc77OyM0McV_)@v5Cc^cpq(`oH+NJafT%Xv95aJ{}|T27fxQwO;qb<O8RL0L}j
zPDcCvma;7G3ie&s@%3+fgYUidHk-{Fr46&?1<GU}x(7@3qXP&6WS%~JBPF1Cu{pAS
zvBz7BoM>ce2)RI<XdW$Bm=_zi?;q)uWnv6QD0~Q1UC&|LkqtJ`+Ivo9iD(>e?kjX~
z^z8v#Rg?i%rD0zxnm!P+oZ0y)wyVew2b_SWb`;WZ_2dDY`bZ-j3lqqSg6*zF8Nunr
z1$E=d@|Fj$U$MG8VZYy_wPC(o;xd770cBuP<aD}0XhG}(Sy1T>&LhTy>Y%%6F`rGQ
za;|j<2a_^q`DjijA)jXWUF!5i3ud#D`KlP?hY=KmPmiY!O_~lW6=&y5dQlMzuqGYZ
zz5TFo`N`qsF7OleqCP%Kq<P(8n5t(z$951)C{Q}}ghV6AGD8+Lhhvh&YOU@TKo>o>
zPsMa81;Kl=Ort!h>pY~Bj<X&aB1Q{@NmeLcYE+gb_g?Q%At|j~=gG?gsU`TNxDrMU
zqiF+KUJlN$kmWgfnSm0+>~@p7+-)*IR4Unrpak9uginotQKz}?ee4KbfNa?5hXACM
zoyj%%#3aR;Ofu6D)4X?6=)1<dmjN~<kdi_P#Ly1|(RqR(P5KM*WnB4~r1x2fl<IEk
zmd)*&NjYV8vSK!0CQq>lH#au`=zK6IjFJ6e$Li!X6{~&n#CrbhDS45jj7CPmg9ndT
zub=ZN^)r0omH!4m^Dq7?PqIsz@+qJFOP}X|{XhOLw)I?JKj)R#-e5ML<AXyA#csQ0
zQshjg(>p`N_3bUj=rqe?I5};%8<f<kt9+R6vE6S$Nls4A?wSR6yB#3}PFCr%wr%;=
zx4y|YzxgemeeeMS&B?_pm~1w@Jt<&?@S#=y4{h{!Az*vY#2B(7XI-__ZA;JsS8B)%
zy2ue_PF)D{d5IB{LXc_}Cne&13d@op6|-4^G8tJmL5ZBM>hLyTR7rEJIUbJWO0bX-
z>IU06eBa?i>KzHWz!s8VB$tmac>T$1tjv^CS)v-xal7N$^)=0*;jleY9gfTw6HXQ@
za#P@gr?)+5fz;roqw0<v4@bIS5fgzMq?FcLh|u*t)$T}pTjQIavd~PXIa)}N5idPn
z1<F;4o-2$|6lIPm1o^CBvYg<xAS%eqq);o13?(IfA5gQL*;z?8HE5j{7VFY5oKFG*
zaKcc3W=i!(XM`gAF;fbk3<6+nKnj5|8W$o>pJcZvnHF^^kxJnepcPt56lph(QIIRh
zG(n1FHKNm;4<(YpU9u3D#0a*x2vU!h3V~pgHAQ%-s**_n6J$zr%_mdTykz_NGY*eu
z&^xTPsW~yEz7(w$MV@0+YHW-Vgc3+O%&`!W;8VI-n&J5ky#+4=U>pWp5hGekvdo}V
z8dxP|s>3Lw(YZ!yndY3RH2XsvH8@tpLHU*JTBMXoIVMs)RA=e^jDrFz{FsvwK0K`m
zdfRi@?<mWX*?d8&%BR`rbz70;c`A4Z!Etwh2$R_~#WWdHsyqZH<s>!kg~)4fyvAZa
zqmUkZgztXsJM5nQJwE*}{2KcF>wNAnd<!!ROs2EsxfRk0s%a{E*E3rz?^f-qu4wy~
z*>raI{M)9%+Mbh>)4SJFH%F@Df%$B4cUjl9IBPjQIlmji_0GHR@b#~MgJ;j4a;%O_
z7w1f-tMu6ji6GU+UOE=QkFs-@pfbVLa?Wa2^7OdJl14@bEt6$HCpp|WdQmgEGA!Mk
z*G$ffr!Cv2=DpK>GHL62PG(cKUB%PQ2D{&}sT<y1@3@#u_~3fO``ZIkm743e?{8Ta
znv0Ve()G0aTaLF|-aphdZ}0fr>yMFI@lXHspW}`5bF9c{tm7wMy-7LA`IMW>4a@n0
zhZhe~MaHYsSNW8AgVkz<80Jp*uI2L=UnEFSrJ~3toS&?co?L0_w&JH=`I8*4D;~an
z&f;Q8RDvwa$TH1i`6+DcxO{NN;bzN!@PGTiqoZbDx0qQ@F)2}{pbj0|_5e~*%%_M<
zP}eQlRSCHw3kvNdK1N)U_<thy>OG3(|9;C~`OCk=EbsYS|CcZG>;K(X$>{jqU!f#m
z5QIUe^amFyL8wTXD~jCEI)@!3wlACeWDwXsEwb4Z)J!o68XQ!uqty}z^u13CDG_OG
zpJe6U)ASbMBVE^1_ZIIXC8=Yb;3<m<TBb%Z+qJa4$99g1Py{9U+W+7u*#67|R2%v7
z|LPkofByYc@D`G~wFn_7R8A>#THhz7Z-@x*hto`??|Phf5F}9uvfO+aM6^iFidtZX
zV!bgLl8}PH%i4}qN$E8N)i@O3Rc=x`yNmQ8eIK;eNRw<&gpz4ep-e^eHb%UFA9pWy
zcUBex=R8lJKIcoplN*V5o)8?{?V81GL7wLj18vuFtSYXq9*|@_fy3dze!u7B>=dO_
zYP+*7pa0ZDUOD^c`QZJxdFS2#4eRHgpZb&kRm7{G=NJFu|B&qbbzXh#3Crawr3a&+
z?HV@gTUM(R<j^GA_YKFxp0m>nl-75FZL_)I?CgBFEYQ1-<Nm;Ed73OuLZI(EcKa>o
z=a)<-Q=D@=fBu|b`K4dst>1c+uCsjlb6=z^OJ3Zh(~L42qw}H9=66eBK~hm<n)$Rq
z7)hlZu9WByP>rMS9BMW}PBS{|X)G)*GF)B|Z#$X~DkjY`jkdG_W*OpS%B{0_7l#?q
zFf9y5O1i3{$P0plrm~#O3bKjew%)LQddv3tf$G>2_ZHu`a6IydzVs99cWaaiWU6Q1
z9+^$2s9cjNO@Uz&G`Tl8ualE!@0o~}S?0NEJ?|e{yokg~piUIxG&NDinC4CgAMnnj
zJcvHcrxpgfrl(q0gOg9>5C6oMSk3<q&9UXCK5%-nBr9{GfMeY;^F6cW98oG*NGK&P
zPqx^u@#vW0j6mk{6SGT?QuC+&nQ!uk7taa6PyVHE@eBWp&tjh0ANm>v5l0UHO2aFE
z{F|(+bE^N{75du#!F^X`Di#~9lGN1)q7jHll5e}DyfeAR35YJBj35ey>?7_N5H=un
z#Myw-3XvPUNS;R`3POZTNsN%h;0ZoSc%4rtpDar*zg^$a)i$NI8V7}>okglIE+NzS
zO!4)9?RCEV>+d7#WGN(HT8u}D4w4LocHpLxr~mvz>Q^=M-`t{qZcXqJKh$=Bq2n6F
z&;=u+K%&yTk`Pm!Bc?{jmpvN-LG;Og$@ZQQlDbNY^xnA;&_*GRBI@Ke5hZvTaf5Qr
z1DMnw^RXGMP+s<Y0Bkl}>gq_AO^FdU+bupsvOFIowW--|K3|aM#c=WoY&SddJi`n<
z8ohNib;VS7y#Dx%*PdMQM?RnWF5KA@{=NUx|B>Z<&To9{7kTBCSCW~nNRvOR!=AE8
z6A@!f>b&iCO_3Mm`B3)=p2IQKIPyXL6=P(--;?D9le-jG;Beei6a|xVa%ZjbwXc1J
zciw%M<FVo=fBMhz%BzoAzj&K>-u-~!Bg!bWPGXI&?{W4=d0~-u3-d|NYB8nvo~rB7
zKH|MV?j7_Vr!{$2q6$I(yyMx4<oQ{{>H>@{3D#q8dQ=2CE!l-Y-yGQ5j)I*2*ih-n
z`C`RsR`PH$=X5fo>m8*~2xI86th=5~-P3M5=0dW{4DI&FcE9HvU-=rr2WBUppa1GV
z!cJz)vVs?HU2}GI$>Z|}eChR1^WL*}`Nmhi!52UGkW2^gH98#mn}74$eEY?o@+9Z=
zH=giQKlMeFDiB)o=J(#@|NS?9p3VC^{`^1pWv<?MNFzKCKlPB_w)~C%^M8s`hM)bf
z{Bi!7KlbOS_dVC|Kj-b;Eobu+2!i(DnM^ga$%N(Q3au1bo*_acY+4R)9q22MSeA4?
zkj)I?@CSRx4Jk*RA$jZ@I=}&6K7Y>7{Ec7XtAFte@PYdsw-H3P<($|4`|tCw|2zL5
zp3q={f2aMgbNg?X=pQek{rd(XB}PgLom5da1jOLT8XW`|2`Z8m3aKU0MXZnPrvWJ<
zs?-PtzH>zDk%h(zP$AF<hZJdd?1*?J$V3WMg)!8=8PZ5ert4y|6!9*A3*=f6OeCU!
z9B_0$amn}ol{a|f|L`?}a|9RAnP$@*lS@U293E!;*8k+o9RBzj0Orr{`9uHUUBX*C
z;?PUsdykAmJ;O)3%F;ESIE@7B>7xKG2)W0$4qtm%q?3)cJ}I$GvTczFypKemEQ@qz
zQXrK>U~vI&EzRDcXJB1KnG{$)eltiCI*7qBS4zpVXV1C4x#jfq0sFe*a2N{WBvsmG
zw?PP4Emx^nE%?y?YBqzA(uXf@t|@cFWKz)f7F{ax*%g1}vv2V8|H=Q0_dfW5uP*)=
zU;Gn)kuQF(;H|fQnGhU@!ww-Ni{btwp51nXh+=30jKmn&tZ$G?vRa+q&G*=BHhAyv
z>K-8k4u?HOQF1z5R@XJZ_N%|jcfR#)W=8P&pZYSBvnAjC&bN5}?D=5qm~8+1HU*8g
zO?Lk9BTgXBd2%V4W`?6rAEvNA<??JmA<zP3A2A}wn}{$9D>R`L<mU=399|qSA<fi%
z{=qY5+H(vMZxwRia&&u|^^O<Mda|pI%QAUu6<L;+3t?z%+GS))kQ>FM%;{b7`Vk0X
zXAyqJpa1-4_?4%(eC6x!F!P%2wBow{4yshthb^Caq*+Z1Zr2dVIL#Z*bHg`!MWzch
z8Lkc}6N$>>_m2J$kTN4g=qpQ)M|pwkJ)ucz5n%<vL)SXu&6et>B7P>&=_XkfbQYsC
z9-cl#_`tiA_?^SrK)31eJsd?xjOp}4^9O&Vtw2yk$FH_m{50?5NI~VV{^?)i-~FF1
z`RZT)3{z?pHuXdR6%&^K^8^3J|L7m`god89q{YXUNcnsEdj4?;Lr0rYX_Tz<FKYoL
zyOF^_kB}_Zkr5#xg4Bd2a)lI<?qJiHT9V=?^<PAM5JVeMGSCD|*QI>BEex)=3_W!u
ziK-w9kxoJ)6`##@huC&-Mv7=car+lO!})*yjPNTr_+tli#lAlvy(cUSzWv{RgZhuZ
zoP_7G#vqoOXj2_TNij?<9Bk~J$F&~Yr^}r4;FF3g`eZ2|`lOr+D&l==+!Jk*Wd#?K
zcbrc<>%wG|rl6G#!b9jhK1O^OuqNO_OyhEgQPCfZ`2Bq&LI`%-9h>!<5Zq8ln3Lzp
zNYJ@te{^zkHq-<nb<<GS)lh>-$Dg{XfyiVw<JfzizX&WZ=KPT_{S?3Y%YO%<3O@Ho
ze}-RuU$H#>JWpPG17}aUe(_(h*=@MIe2_X+V)C@IeaqST#oY+ps;cR_mW!(gLvK!`
zZfcyhoSdAdQy$Q?4c1!D&o7Wt(sdngz4aE?AH2`&7ZXmF59!*DU;njVXS+Y9_y48h
zDrpsfBrte+pFdc}s|B;GoJo;p8ux7v-O$uD%mIMp_^GZNIt2a{lz~T-i1~!_WJ<rU
zvERH#3Fw0*?;?$dY&xZO5i`j+JJaktPgO?L#PZ(r@3A}7JbrD-HZ*K~z?FhWPZr!Z
z4%JHhGNYP!0wJj-*6)Cw@cL_kKmBCJ-}}#QIT*<@^IXjqT)+1YpFRzoFU~j|bL{Rp
zdO9QK5^p<V>oHw~E_K0c0={)n2UHLQAE*x<tC=O}0M=2hYua50xxjBMq4iuYOSU(+
zY`?ds`>C2;*Wj9-?YjpKpQ;H}vSG2?mhRwiP7-V&Sg?&n$cT{wr+>T-+TX7fG4fmg
z=!AFvxZw(mCy8SI)&Jhl^LPK~p1=Pey~?w{^%|#dX|7)j{ImbFU*KQ+SAT&VPe0s8
z1eU)Qxc$9*8UJ|1G`2JZl>$N-n&m>g>sylyb|rn9p`Z$d(2xs>l7?6cw30N3q4to5
zf}}CFMh%XJF*z|v!O=Rj5_Goa*tBV8x`fVKoQxC`4L)+L9ps8=J(G9$>^}Vvf(N%H
z6dB+AU;Zh|zxi$E|M6R7>%gSbyz`Sw-u<^egZuQH)DGT(grW_U2g@Nuln+Qn3X&=k
z9YmWfOGE^31Hnb2^q8R2>8VTe*Fzr(B4CsxL(m;9veY=+<E+DXp02h`XX$>ifzWwe
zABZA7rn4SlBO(`fJuDv!@%@!KA;9*Yo0}WD=186yvb^Nw9Q6Hu$9%S+EXu)4DsXdi
z%cLv@Z>yMuIMtD}(-p=P%-{GIS$wMCr#_$YxBvD(;vfEs;j@4Ezr>e+@(aBG_x~nB
zw7m0kirZiQNyLA7%lTPLW-<V5-?7_mI9Z*d^^h9tEyu&2)3Xb-GC(qMT(38rot_T{
zkVzQSbuH(Ym+5rq18=|eCePph9*<57s@*le`JFB6gJWO0yM7Ob6eBO|=0XsH!?_=k
z!V*d_c{HWO0Fu@^q8DURld_pZSLsyKHyvsVE=pXd=&VWEPVH&;HNNjLSw>V6@|<I|
z=qRuvopSc3V4o?1j;PBeyQe#x8=4vCn%RkFS{?CqkBG2b=D0GVCm?i0DTA+$tY5sz
zgNs+Vx_X8G+CTkMZ0ZJ?3r>rI$0w16zv1TAaQplg;cCpRplx<ktwlIbtXjImk*2Ls
zMq#4h(R{-B*@XR}PP+9xl1Yf2=XhP=dJkEo+jQ8*F`t!u?|Xap&ki)#M-IKk_MWCX
zvUz_;eO;4FO>^uCK2mRcx^}R;wgM*u+KV5@K=k)D8J4g9!_V>lzxj0@O`qaPiliki
z|MH*tRlf9RzR9!y)*CE8&}<)A{={d$PR@tQOAWkD!Si36p!tJ(DvRh`k;T-66^Qr%
zO31rT=@3AhH2u%^9%NDvsVLIXEP@NEs4EqN3P#{bwU0>3i#rr(i7zCXR#=jiyo#w`
zOBM<(1CfAp4kH!P2n2!r?HzYd=&%DhB~F&qf8h^P|JlcI-I4i-eRWDa4e6xP9-(hx
zIc4(B9<J+@GID_^9M0KPs}KSsC6mIi*Qq~34&GoA@O_$K7)kXXFMJv{6(aSqXQ~8!
z-Q%psxj<K2Z0iWxr$7`V-bVaj0q@<*Azo?H;>Se_d8bl(nZ`mycDp@NsbrTv*ybOO
z2Ym2J@*k4|?sz;Pr5NUeq}0-GyCussSyoV<{!vaJf0{QQK!0#7AO11^sXzBBhr^!Z
zQE>L?^9Uh0|5^C-pJ@1p&(Aqqex7FY9w7>L`yE9&Auq~1L(KK&maNE!Q;{Hq$l<tW
zQcOOq$pp{wuwy!#r)ELt*=}w)+`PwBw0!UB2YmZ`&*)>0&L)`5;9i=3enf$vGCp9t
zn&5xbNZSamOBMrB!0y-}VzOrsnlutK*YL&)_3n<asgbRsc^-K)J1{*dkVMqAMb@6|
zOmd<%4<}RJ+wJIg2kORQtz&I9b!M;<-rVikzrAHuW=tlA@?^r|Y{Kq(j}L)oZD4Yh
zBa|idJ$<{UK0d=sjV=sk{scXF%)j){ehJhAeEn5yXz32mc>0~U`TF1gM{GYh@Y$dJ
zS<14cxwhmw=g<D+5_I6jdP}x_PGLH7T~N*gSFcU@=6fDHag<k@&N{aJhIgJlqde2h
zE^?l|ct&5>L?O7nK4NRf;ms|#>pjPPi`I(6^8>pNwx|%fes;vVfV2@Ap;wXlMaIQK
zu`q@YcJ+_9M*Tj;18@94J@5TnSN!xp_w?@a05rho`2Vu^-@%rq_kAb!dD0tBy18>t
z#2^ug03m=N7(h`XMNw%jbEVo{S(2?S6=juVYu9qw-YVC+%C*5|*)rwQuDokfE6GdI
zD~XguQeq|%7?1`SU@&<a)6>)4w{JY*O;7M2zvtZBpfKQ!mjnTbPt~1!=k`74yyrdd
z`~1>3ta!_#pCdjh8%fW3@X)|A8vfq@`H!&tNR4|<$gQu<6{A9@OduQrAtaoVR#H@;
z92Aohl_LFeU}z&rDNa_NqL3amA&Fj_Qdq4SS<4V2z4wIZ>4!jH6oe#gvOY!H<p4<r
z=-O&gicJZ8WbgylW(Y2c*(dLE_uo5}Q)~?2fDsrCmmY*mD|%Ldhj0f@PvPY`>}{Y&
z8@%DiUL>`V&5eq(GAx=Nx)$wZ9-=i=BaJQ<J&{?dM3!C&c`XpBW)7bQ0Sp+0_mRH!
z)NN#S)X*+_ArkhUq3P*$ws`O&ldD1^YB}_AZK<4P{yIum>tv>~UTevg{k!|j=L^b8
z;Iducv05(K-Q7I{39MFgmdgcOTbn|#R*CsyP9idyY;(|EV)WbtyrG4|MawHevAuhZ
zo!#p~2^a>(qYJE72NVxzu3nj-e>`w-x=XEZaeQ<SNNjFyogLqNevDEuo6e*)nIb1A
z$0)7Y*xWh;GMvs&ab?M5GNC;^WdG$Cxcky`937wX_){-(ZyqShEkf)VRwqoxBkFNY
z*GBpv`e5YNSl=%R!J~~-MXxHRXiuRUSrm!Yq9F#4(N@IQoFPt2YHf+rlJ>YoEqV&0
zNJ*2Ho?a!)X-^#lw_O;;@kvW}(ot71??s~F&S}G7<bt_-ydoa9yy3zIdwXMoiu9|F
zAq3fI`vA4UqKV6%`Fugs9x%JO%WSqmQOy`U49cNwAho9`R}u9F#qJxx9#Pf-T~A1{
zVHq^ltYGi@j1wIQLx&A5lYT`%v>c~^+qNW=I2jsR+Yt!#rwdN+-D7KJ+1oPAV#8Lg
z&?qLAWoM(Lj-DzFyzTmojY&Z<)~MAn?W#jHJuyVCZaQwAL=KvPhc8t;^w5OE{hpUz
z2w$&z<!@OE#hlOns|Wf1Km54JZr}H>K7W0=RRTZx;kR-0f83-FOJVkYy)G{>U4>K;
zVF+j^X60fdE3GGmCThhn1k_r`DkOYLq&{K15bI1vBlUd{bW=()<n$q*Va*Vw4O>~w
zXyi~v)OCH35FJAl_PG%}T=ls_%#8!WqkAyh08_H2?8*0*vzo!oKzpzT&|Llh++gQ7
z?{QdLN^5Xqi%Nl{A~ppGQ3NAxLR&@aJwXMu)`Zm4wJlmDMwNjW89PJUX?#jlg~l0B
zAuuj9n<FE>@G%g&hBORBpC~NoLenlHRjJsX6=<z_<#kl9&e=1$bI5IErDUI3E~#p9
z>|4wijK>qItZo})<oNKAs;Wg~5s?k)+J^0|O?(=&Xhzg($>j$FPu%G_b#q45l(Lv$
ztf8oks9@Y`Xih!*Wy!TG8+_`sH`yC&R?7vu7cQMuMs3s3b-g$fW|3R*I$v^O?_z#O
zLJm|)v$?U!$tyRw_tI0`-M`13gCm}~vCp03CDZ8zjB&z4@|xvxg~GEl8?ovY^H%QU
zFtqr-#ksMxw|KP*Bqr>sWZZ;A>jS1LWdn%Nw*gZI;;mCmX(=yn;P)3yM;5m=B1}qx
zALwqj7@cSdJb(L`E-GSe*h~dYBS?vp69~pr){f?GkH6c{UM?_|!TW)iZXR*__95Z0
zMFFGDif9!U%kAR@AAkB;{^Ym48<Xao9G_B5XH>I-*c@Qe4AnOHZb|7X-u;K)$I0D>
zY9nx}5BNf5_@ap@=dqK5H@{_%O<PiAp6l4brP@$M#kA?U+7BonvCY8c2e$-CMMU8l
zTd}IVxHsnEt7G<NiL2Y&RHdRen!fK)t9z&+uw6!O_{gZ#?2HXoCEie2-l#2~ZyQqo
z4bsB`43RrOxxownr)}OiI(XH9asMiSpI$x0&-^Ff%p|P7zJrHSBp*d37irrntHst9
z7zP=b`_^N$qNqe+H>F5XT5zJ2+YN)91p|w2fZh|8!e&+55F-7MIYe59_FW_1>q-L}
zx~^xr?7(~aE-)SyQtD}iJ02MQlc$;g_08<NJ0Jn00AvV!UMB(@GZ>B8{`fuK{@;C$
zQw?(kg^HjP<4Hk<!YFAc8Ko$VqEJ#<R54;vs1T`*VSBU63JeKNpeJFCW_zQg%>0f?
zJ;LUSiR95kpf-_{B<+^cDki07q7}1Ds#LFo2;!{6S(7F0<ye$fv`xp+$q`#SdmJ4d
z5JF@&69B;)2CI#w8jS&k_XDR#$4sVUF~$l7bur=cBc3P1O&)mL76+dT?B9NgqN>=~
zy1-;MMO($*^^5p^!Ty~kvyCmrqkGI3k)55LvzAobHXIz>V{3ayW+4f>u3^<I+1{4<
z2q3=lUCUOjx%0$FxOZ^ON;#gr+wjGg4mmw-7*Dn_#*tD06vkF4WtlHmq;SghizB90
z$ty<#UAM%!F-oPgRPfb#AObk1X*&i$PYTT7Awckgvgqb5y4C1WL97bQc1^iAB@`vg
zqXqN{>kNaB9NeB$O-59kQ@RLo=%9_n(}0~AaEUYsqgJB{R%{)QPL4S_S`hAZjGQI9
z$g1lYmm<q>>*yXwcb=qhn#Z5K$s=z(=8f;#Vd$3}-aDZ*r&RicFs!Icjk!>Ay1K)i
zPN93xuxy#MH9M7M!<9^JiAjnwL`Hq!bm7r`BrSS2yM&1b=O*l39^w0e<zmI6m6=cs
z3S%u>8%`MSrNyVfY15;IjuOHAFIz9e;ac<9#VLL8>{Oc2^rUw9EqVIa_fjPMO~d}v
zo4om5udSb;fmeR=G5Sx8&@73sE|~Cju`(}L#OqQ5r^RVD$^1j*P1UO;k|J86l);sb
zWQ@#B!a$K-3YVQYa@KW0QW`=)M?{QpojpT_m5RVtOfU?dei%rtAfU8U1Og!HxYe)R
zX7Pjjqz5Ljil95FFUY?WAa<YwXF<d{cAq_=ZyI7*kb=zfswl=(h1Qrg10*H1Pv{Vd
zL&WB)L1EBNlWdaeEQu0>5C|286OVLdfFv+vVniot^Mw#acHSyxBg3SYU?T=q@;t9~
ziO-Rv>qH|==gDyhfzy){y1pg&fsKu=Ge^PYYC+$&OgA>~6E|GqoH^%SB#U*9z3mxK
zz1UJ0E8h6fm@hnQI5|0_UA@TibeGMYElg?fp~v?P2Ztwo(_<Sv_0$fdQIOd)Nq8Z3
z#dtiH${>1<kB=CSCt2r8uC>rE2&V_!y>*wvlMb`{053oF9JgNC=lJB9dbCMl>nzmB
z8Q{QJi#0Qr-IAwnESOFs-ggvbg>@smU&2=kASyv=G%*F1-9T@^hk#QWQy7ZElB_`o
z2+M(C=}Cp<<X}ZVB(b!d_qZW38kfw2pqqx!64NnpT2W4Gim{f93XArdxbRFX%cb2B
ztIl)dnO9iOPpDQshj%-+Hwrc@LtL)tf}ym~#fIna9&&kOlb`v8FYp6}<L&QG9PU5G
zkN(KNB*n;o@TcEIWju#BpQors9Q0Qx=BGU5J6?QcNl{l^uP3As@Nq?#r6EEHOd@>u
zo42qU4sYyJ)HQ{xh(SE5i+aMwWP^G-qYr_yvJ{nL-VXTK())-G1ACQ52T$Ae_z<v0
zQ#!@Y%#kQi8oEKOPrs2?LgCthPk!+N-|?>JUejX&@Zghyt!2za&9CPr7PK0<hXy4i
zH<jO)Eo2}bk%71Gg6Lc+g=z-04#Whsfx>B$)&ky0kwy&^!Qcvmt!3ye&#Z|)U=`Gr
z#Td=9>FC0M34y)|C=Z3z=$HsmEN6FbE;xDSfd2Xz?#!WB!nHl1h5egw`vk6Sz*Y%E
z2Wmi_41^|8r-4-;h*443S?b+KsW@|+C#A&pxS+KIL(@YMr4<r|KcTe28i!RHJ+uU$
zFjh46qC62PidTxxRa%S*Wn@%ijmi?`WS<?*<^Zo{Ip-LS#&rG3x_>2AV+b4`9HLTW
zcXtnCWtQZ<=iuG}lgUJ+@l~Q(E@_)a<|El#KDle`OlI^!;X~xppM8-Jec&3`u1?s#
zcm-tx%hLsi`}-Je*w`$YRDtCxu)O;cmmhiu{psfjt5>*pa1X6C+dFFrwc_OD1Zy01
zJthQ&iXEpfKg;2bXJ~@rfp@%*D-TZi@P|Lb^UpuW@$nHwIl&cU2?I*FO6MI*`MDm=
z@I%MFgMFNZYnLz6M@zdBSrd4*00NAZp|Fm5)6uOuC>?rMLNCnL;j|UP5aOHcS3SuG
z+PM%1L+dFM6xuNOUc|(lC9ZnBR+z#t+NjYn(JmTP^3Zhb&I)!m9s8$qW=+Q%X9cHi
z;^W<l!YamA5GMst*98whyvL<S-pS5r%Afdt#T&or50K0V6+Fo&Tp1bKhH5f`WUz6-
zdC%f-j#dIGyz$a0S2k*qhaLhcXF)#8*}v)9OzcNH52_^&9v>9OSxRk1OC~{K4N4i>
zWy7lL>1;s=k*!kEhs3fAI1Se~9k=fF5QP&!RN@&?Xd0ig#{bv*#(BOHiF7c2&3|uz
z8y^ZB{oTawUA}JdS0MY4K@dI(@Pja`Q*FiemCTojG7pg!h)ABNJ_w&8B=iumrDX^S
z6uEUJkX4;Et&Gux7>NBq@{z(QMzs;LNgJ40u!ZGb@TA;c(mDNf)}Tv=wqww{aLIrw
zfiZ9bd1itRkaYfvEO!bKFw8LBrt`GTKv765Xzf?AZVT0=p|Emo#~^5~wIiNJoeyl2
zR0N^~;j<`Mw#iU1L_t`UPT`E8$5JBubjaU?-@@zo?Uu2I(YWGtJ|~874y8|t!=ocY
zj8vmqjt5b}EnGo85=M6lk;Qz@crwBkVznEhfnv&=AKl>KNa20M(cwKl`f<a1zxg`9
z{I5U4WIE%KN8ibGvcc)e0WlitYK&G3?i_aPO*aVB?_fCn9~jzJe9g0B@~T<VXHBcR
zfU$0P?z5ldr5C=)-UE;E;A3wiS<7cX^Efx2e}Tip1C();^-N|bYxU%eU{kCw3Pp?o
zKXg>%DcTqgj!zknYKlUg?M|;AKnjC%Vri?pf!N4N(}f{+1F8!IrHIyuEXA@zg+MeS
z8kl-dRXDMC?FNdv!elmWf8rVDE$U*RtPNFc8Qa8A!O#jJZ_jw%wmD&Id&Xyu?}3N$
zsGu$imP!KGNNe7F;VRcJzn!jG@UCzDHl}+YV6ix)p6&8K|BD~QFYhqE_;zf$&9&>_
zM$x@QeRRSX<_qY0-u37<Z8LCqyvU|K0&Pq~V2BAYXzg%TW36Ou$qO@c*$%WJVVjPj
z?LccGdWD{ejvTcu76s1GwUOFs9Dzv%b#0i89ECHa7#KTCP%x_%m$&Wf%ACH|mqxSk
z!2WB#nLY5H@A^$1|JKWdpEzrReBCaSTTcN{A)>lSBw-7&EKQw2ScAzMXcy2v5Vb;E
zBPzZb`WiyOD23IzeUlPx=y5}!EG;?-dtIAE=w-%I7qI0tRp|&3?!^wRMOsBG=`=V6
z#l{GA?J`iq$O9LV@TcI?MYvcaj&8H8sM|9CmI}A?EOF5?9Thkm@G+43jEu_Yos>B*
zpt9Pow87V0m_Q<ifKEc7j4^_?#308u#7InJDVR(W)M5whj0h_lBftcM62)L2smj;h
z{xn*%wY|y7@hOYtN+=Ndm=E?3I5;?9JlO=LI66Mg{Epq693<w86RNUcI@v(y4Y>6i
z*dk(d#s0w&&FYBx{D@W4F~9c(uI+8|na@2<GheZ}z0KD4C4wIqjkej`2^>7P#Q*CJ
z`tP`nFCSy?Vn<Pkpk>px93LHV?eZQwBaiKFGe0^c6*Z52`}ec8a{(VbAN|!|;j^E4
zg1dL_5Tjx=+rk*}bJ1(_uxwpU4kQT({gP=7R~~yax|(rvc*w!6FVZ!qXf5~Tmk%Hc
zMx~~3(r>JdpgL^R;aq{W00ujDsEk@z#K^Fca@}U!_oUJkRYI>4HhG3lNL0p(H2QM?
zm|^62;G$)_fVqZch=iyZjU3aeL=AzH!;Y>4N5b_zV{MrfiAmC2vm@#I8b`7ZuvmE1
z(17=p<0<uIi;fC~V^nQm3%EF{=uUfHdHM!dwnnV#mN-OG2%>bVw1^reu@x0_wv<KR
z6h2DHR~StSk>=o-cHRhYI3`v}VU3{;0i6P)0V*km>BB%9728`QP@1mu0%l1Oli=aK
zl0l2i#W%)ESS){l)7KmZJn&Xl6z^<lexk<wZXVuOVQDQ^wlO4(76k`2L`-%dR8D2+
zr@(5xN(@axB@q}&x#BPev`;eQ4nc<VN_4iO5A-onmx{^)A<<;ciBbw@G#X7eh!&kK
zKxNV&N>PM_-Ws7cN~Go+Bf(=WtD`{8kS!x2I5|tV+v2Wn5udq>Mqw(>NRsDldgL)m
zpv+mtT`Gorov$mjI0Ql*^8H5E!4mHsp9J=<qrhYnNq8<+1QbJv(tXjOqiBOoEAyHY
zmlPvLD_r47i>z~@<nx;^RvaH4@!*3n41>%ZMq^>Rr$pCv^uFi9&MqnkvbG)Y<yEd-
z+++W!XVt7YIXYyuT5@>%Szg5RpZxFtBFgS?`%Z%^YnpaRRZWR8(03i1*K2O;$N9i}
zF7wP6-$+<C49yKegDn;~U~%&cV3+iE%xT<a`;iB!>oLCXdF9qko`2>!P7aUoAySPu
zv8Ip{pJiL*yT+i;+#r4EFoojMwF~Tw5+Tk}8zTz$2w!~WDF&YfEx+squM~Bms9j`t
zlrWbkcwlF1#CU8(AEysg<C3B-B@0^i4D+7kBepOol_*N-@9ykXn9|ZU9i|JEg^>#a
z&3NRvzFo3YYi<k;&kY^6gmNt218GR?A9f@KTa{(UTdKh_iH<RrV%gCxk7<SxzxqqR
z!m{^#+xy?bX1PF3D{QqX?D-TRG$=Z>3RKSFn;stnzKO)Bu{x207du}s5)s-3ab;5?
zqb(PI9yTQ{W>G0JBv0253`%jgR}9N8=WdZEBqmlv5h;A4X%rXtCXDKUiap*ZR?%aW
zqN)teeSHs&?_Ua7Y$aBYyuPMf0^Yj)0w3ocoDFTiPM6VO5TU+}$`DI~wVI-k>OE%3
zZtH|>7@{SPBgv<{fgwy_qbMs)%9~akB37oof^;&Ps?bzU6MZ7|Jz7V!mG+iVioTD8
zl&GDeKv5=XN4TiD@!*KQ7He7NOIX~6=@=#^OPq{3ZwiU7z%a3};UF6n5i@}=)4G!q
z9hI0sYY`{~ev8Rd?pY;iclsFdK4DG7C&jW4XOSk7=+KNbI3<!N+8V0Tppv|P%zhWf
zX!Pq~q9>rLYR01p$A>3pLCzRMvufDCyN_0i#e7a(RZOQ-Ii?s{EKZqDreeHAV!2u}
z_=?HyC6r$AsZT#kQBF9SA9H$g$SW_sz{Ts2@>4(i6d!#5qdfQglkBFe7+Z35bVB_G
zNANB4#UU>oJ;Ul34*CA?`v5QPe>;aSwd}wAIn4YvH=p?e)15ucTi#7j1zXjasv3#2
z;GJ81>XV=3&fR@Z7pK(YEsAo)TH{`6N#xUgCqOAhWT5XBZ13$-ZeO6a8W(!56%7k#
z=pTHL@pJ}!1r%24U}L1%eRvD}CPpbXHfqoouS=ZK5PN*z(zZQA?HFB|KoWSkQNsU7
zMiVe31v-k+P09!D3QJ+FxTZxooZO?H_c*6;t^i$f(yr(d>}?iI3X4k)mn<$Hx~jJ*
zrq{W5>xf_cxu0jgY`MB)c=XXRZBR@%BXtR+fN$;)+UMwl;&}f6Ra9s{P*&t2o!D{e
zq)-JUWw1^Wf+i+us*19x5}yR$p;hi5D@EVQ|F7D}%l8&+n8cNhf|+xSlmupN6~;td
zYAE8mu8;Vj7^0NkA-?Lz@^!iTNut>auPcCPpx8*6{hGe6p}b0wlLN~*n<F`6QCeir
zQ9489Mb5(qnGvofgCvueCrT<00aQ944&^?!F<RP2K$kq$`t#o_O+?X$NKqJZM$7|u
ztw5*9(e4OePJs&czX&hy!=-I_!)`7G(tTOL0L)3c=ctZsc+tXBItcR}VR5T0g<()<
zUfNOTEi1x$P@SSE601BQR|GM&QZR2khDc>Kr88(1&qNQ=Vi{}_f?>{8O^gzhQhI%f
zi`I&=Dj<8wuiIEDCXSAd=zWWIMmQfCb+ueBaju}Q$E2u;$<wt<`Z%TgOhb9ubMwYa
zZ0~O4STZU)zU_PdBrzp!-<|XFD@T0mw>-edAAgbA))pJv8%$?orW=X2^^9-d;tS8*
zWM|{!eAkEG&qtn(-1*!`Ik>ma?PJflt9a(%HweKq9!)_R+P0->mfXE_hxz;zSB`P@
z<m`G{zp)(7&rd*#^xYDzE$s?keqo=z%bOIHBSyuADiEih!hMAyy4G-cujcZFDOKeN
z!+`1sjJB+jA+!yny2Ml?R$b4iETJqg3{+)F-*#97g|ir~B=?F*5Ess7bCM`t2nnop
zgaB8rOqp&J%r?iAb%iM`7cXz|r@qJGy{9fK%Bqx>k}_iZIlhc4E-~5Ey!(9*apRfi
zsK<djuN-lFvd_EU_cmtZJq%0w<IiLJQ{3MBU^FG@&teKKCO;uSayVzuM7im-#m7L?
z^aP#OFm<#}Sfd!cAV-`v)Rkdxq_C?<F^X(&Dt4y@GaDhaq+sd2Wl}4?=~|5;kwT#J
ziC{H-7a4qF=+rq5{TpPh&)e%+5qST<bAzAy_4hFSyEPTf*KH6foThS66h*E$V$S>o
zm7{gSSW8F=O{8>|6oVjplC-jv&TSrroE#;_No$1j`I(W8+5y|sQp}jFl_2H}yIW(n
zW<peS#-5um0m@dOr+|gM4LCZ0v4J7L@f@}z*jav^1u!R?4L<SLzm>P&{ss1)I3yHe
zj~ipm1M!U4(U6kN1yYK6Sc|;3+`iIk-B!`m!JOOOS#dTjj6$a%*Fn@I1LION&hI}4
zk!*=st13qIx?sEsaHV6mF~wMm@1;$%hC?469n*CU<M9MP2qE)yKBw<HE?&5BCIdfT
zoH82ku{^zm`NgZ$-&^zC=l@^aa0T<llN{c?$q)X8_jC2Zw{o<9hv!~g@rOQig?FYI
zx9<g1Oth;N^Z6l{A9@Jyd!G2QAK_CcA7=67-{+gZ{rh-ga*?0^xnHAMC7Mq?A*aa}
zT)cdhot;Z)H2u(Ud~^?OOX|rE+FCi^SQ1%}Ex+?MY^U!QpcK_;hQhEq?Ra^$&v;g`
zbzzf;<B?6xrODYx^KUL|HP<fhur-^CjI?vKsUd|VM4{9WmP-gTtc8BrptZrrKo@-`
ztO|l64v36`b2)1eD6rN}M9+v=1<oce?@YM5HzR3Bv<^kWm5wV{FUi|WxsOcGcv|9|
z!75LT9$OW>?>pbeV{f{OHcRHmCtQ5sD%EJ4)Gz4gFEAXwh}*u%Xy;A5;RD}^b|a3z
z_-Tf|MQg#2D<l16C8St<)M!(PIGmE1kh9_et-Z!th4+rKuw2}7gyW9QY2<-%#iX_r
zR?~;9Q{)^IlQ2mnxQDi9tQ5u>7R!#d7nM}kz81RO*U!>}qG%Lsy!OWg;QH)@|Mma<
zr}_WdZ{_rdC)lsUR#r+-m5$1CUpqySLPU`;N`w((jHnziQWWdZ5Y9o&;`t#rywPUe
zMa}S6ogas^x)c;vWf49N`dkB`sEVBPOHzlDIf{*uvJ0pXNXEc)25)&lvifETl?6M>
zZ)yR44mN@s(O<p9lmD@%_|Jcq*)zvPC3X->tK4}?XXk?KK1d#}D`d8~-prI1{zYBM
z+)<GzM+R*a)+k6uOro;PxFMc(z@nC+a0ro*XO`K|>NS5@!AG2P6t*}EAnOBLE>;{J
zp0Kw!0j2QX)3q(r=@esRD_u4#LI_l&Evh&pe&2Ik{oXh7*5*BY?(t9X!c$MMTHNDf
zzwqDi-9PX{Y-~Nqa#``4pSsOAzvD7bJ@X<vd%F}xMYrmC@u@G+-+O|`W;f{OclqcO
zJ3RUw@8C~-_aEfv{>crN>Eo<VfU_2(vy-DzSPI6|4SFxWE()@1Wv=M&%g3~Wp<NRE
zKsnk#VdOd_;eF)vWF>lQTVuYs@@%cXf)<uS3DhnH)DUqXwwkJk;IUTIb%9}dLQ)Eg
zsAqa@2)?IuHbcWi4b!*+WwNG<(P#`t8<JKKgmYo6#T5nC8j7MMWkeKOW1JY7m~6eM
zRLx{_hjKh&R3;kV6a7F{)$Cn=n66zg-Mh}#c$1f&e3F|_ev}K<f(IVG!1U^j<X>hq
z9`UyC{<G{q^$-Woe^{Kplnna34-g~6Fc2fs+a-~F0F}^6<hV1~jiDc?Mq{pA-l6n6
z0?Umd_qiQ0M%q3wnoiKsvs#`K0@P#3%8_i;V>Yz35^nXc55NAkyb>_FleqJRG3wIm
zdu#x{xwy?k|Brvpk6n8&pZ`C+8RNhHbi$O*qvRky4f0D!$mSh@7`?!4N#~yDx&5N<
zdyu}*PIPS$el|}~>y0|0jBFyZSxdl>tY2DJQj!2fOcqsq^W#U{+@G_$x(y^4O@RQV
z2VFo2aPKC>6^sj*ZXu9|Q&#VIl;8M!yS(Fvf00N2*$ep6irG^(HcCmv4)Xz?eU}c`
zx)CqmwE|nLkz1MTp~Rd@s~lLhNSgS*Cnl*)wUM@-7G9(b#NBI$V8~EkQx=Z#WJKGp
zGSL&1PApdq2M7D?UARmladLXXXf$FnnVj)L7N-mL_I4<oW@B@QQx`Z~9`J{L;Lq^*
z$NxFYg?Op|=7;|W#?vj{{w;rmjjgMwdXF|#eAfrxz!T4|#Oq{vgIgc}=gjXt$rqk`
zfxD}kPz?O&-~SQbYyJrTlfU=_{N4Zd6LeX*qHjHY+Y<nVKzhIM(&aIl(RhOH0z;5<
zxQ4@NL^Cbr%f|Oh`k|w$W?AP-#2`alUz3{o(TZl#@?zg{;lZyEKp-)!R?v;)(^Lr^
z14<d9QMl5ej73|6(iRPwNTY~hz!>2}Xf1vNMNt#NAX$A%6s|xi5oaVRoYNTR^5Ya*
zi7jN|Dzq)+15|6H7x_Sovc}kgx=^eZFSDG_@y#5iG$U7H3>?1l1gqm$n7WS3Z@9+Z
z8{R}Q+9dRcD4xcunw<w8WmFuZgZQAQ7zx3jIeMWiJ`Rve@AK<C_Y$JQ4+9(9yKwyh
zrlmxA@4b}jMiYWu^u`EWu5PzwIA=B8YK8ATwy@}Y(QhnX`GyG~ph(H{!Y3}%eb?t-
z_oF*tis#?`n;+vJUsQbjKYxts3js&}_0iH&kU|hR{CXOm);F^Tl`zI))@Ydgo+Nu;
zzeANbF@@2Vl)cTR?UQE<>HfbHlO<^fAwK0=qcX{EjV?NCTquQBFgo^Zzi^M$)e2&p
z6I%t^fp4H&00Yb_2`;Ke+6^nf5JnR|{~x}W<&_z4{khvr?lc4gc^r`gnGD$>SM=~D
zB|^%$-jpF)DxJ5a*5_mm36}YelXIJig~6}qBLew#W%-&*kPsrPw!>M6b=i(LhV{ot
z@PU()6G)L})eu8qGMSw1O>NsSnNFzdF~k+IzsY3$Mj{1;?zwR3I*&g3CbZHNMM>cj
zZF5T3F4>7aH*Y`0hu(LQqq!$_$LxRh$GP>wXNl7{u=(I7lQ58C;@-Uv^Z2_z%D?mH
z-_5^UeK$v+g3Zy0X7C)W8t$B&%DKxQpR)qhmROLJj$E}U!26D_UE+!v##M6ubHGLk
zC}Hi19)ob_*0<iPz<3q2?LB5YyIAdTRgp7u$-I>Y?|UY*O)0y5#P=P>$~?dti#ArS
z8Lco%<qRujcQYwrjTXT(pOk!P+Mul=qyXNFjioKlKA`nmmVo6bT*+)aAvK3Q_v|w~
z{u3V~Y}ZV`=_2D<$pgEVM=nEk@gkE87cnW2h8D&b2|<yT&!UpYniFB2&#p;{JoH&;
z&Z1OterYQJjkZn(o<lDq@ZgzFx3K;KV<!P5=HH7=rxu2+5{U|>t+0<%B!mIq4``hT
z-jhOLb`RMkzCl)v=fj`6!dv4N{;<A9_c|XRfGO}7|NKXJ-w!^;Pyfh!dG>E!qh2&P
z243Ho^sBrQ80&C_Bu3ikGg!d7!bmBKVy!XKiir_&Ye(lHJDgV|>&<oTfVVa-(r97^
zYbbOc(o5S_E?o&=>i+Yk6e=f1)<-s9IOXOC7x4T|Xa`}PPbv}Ttu@#J$_go8JY>rU
z(O2&1HWN?(`ww&Ld$05GPu}7k|LSFmr6-o=98R{LY)b?baKIW+lF08(*!JtsHf1Q6
z#G&(K^6WX0iY|r7dXla*xbl$K8bOTF%%&5ns=^P#u~-eOv-1>F<Y51RX0;^5$mZ52
z){1Aws#&6d?X7JI)+GG$SzdbK4ny-+zT-RJ%OCr}zs9ioa}=&*b8Et<KKZMxx`Yoc
zFFx~Wo_q2)IC@(1wnt{%W{($-7u-r0c-O;k<S+f@zlK(pmtK02zw^U?i_iSjXPD_1
z`78hFcXIK$53$kJ+_?2c{<~lJC2s%bN6#u<g*x*vU$s654=n*f8B+8N-4a`la79TX
z<m*c`)C^|rD3Fs{bzepNssThZ+1O&XwTnXITtU)?VHijX%A$fKnUh*av0(+#V_ku@
z7Vihv;6)z;LqFiX$68D11Qu(p5n(x{gn1hi$xF6R@C?yoj6E|&iG3@?rx0-Ei272)
z>ga$z73l39X5ac2E<L!x&h9{Dhbyy=!fTwpgmN_@7Y$*4gti`~8ix7Hgw^qRC9!jz
zV!8!seQHv^q2*_#4aybhn9y#NE0TmM$Jog(1t$9tC}JE4G2~2An6_#iTon>jQpo*o
z$u6`~l#^FpN8|Que?=n2r}ikH+a*1A>;JFkSj)V>b9BH%KlJnb+7DdibN}E`p8Z?b
zDerZB?UD`UK5@=wgptlrNrZ=y1IilqN&;gdDGI|}X>*=+BYSF0QHIo6Fc3u9HY(|D
z$3*5NWk_z85s_kJ&1;ewjBMz#Q750+c=;4Curn1|_c}o;atphKa*PBGmA`!cZV<)Z
z4oEEa9H0FQk8t~gS9tT!-s1YN@8g!9VD)(mOT$_=<9<}H1XX+1+FBPXF|Ci8vvaoQ
zF|402tpiYuSfjDFc+Izx(MHgtKv9%9=f32aG;sU&UAk_?_U<K8p63jHV6|GZv%Q<u
ze1)YRe8<+Lrtb!}XC)6jxWkPnpN6P;`HM@Q`s_zZ+5abW_jvbXW2(~?cMsmhW8d<A
zy7Dr!-H!dEQ-1YVejck6Pe1h(i}@iJ_qO@MhmZKb{?%XRyFUQ-q~^xm+x+HdK27vu
z2$fj-yGMp0ClUESN-JXULJG7c<!B~<yAIN4-4?l~&J^7hs6m~b@-MH(3-m(}DS#C5
z%Q=0D4Blguq0y(r5JX}@K~Xwf;b0ihO@p-tA3Q#2qEd82qU|C@Q8Sv<tX6XjilP{Z
z^r#roG2vs%sEtVS;(REgWtk=EBSl$YouM|0geB^VXgAn<@BwDm-vIuIkzFyGyou7b
z80ILy0s=#O!s6g%`o%LCr75RdIHO5zi_s9`jK$(E(;bB=H*w|yN|mHENc%1Yw9+|4
z0_Cba3yj!uO06nvv5g{-l0hY(I}15miXpcV5-|;O{<H*PQbehQwT{Wgi{DTc2!QRu
z{0>;X_rREe?|a}yKJ>qPk&pieSNO^Q?FU)@qZ0QuIb@gYIwrBJ4S6$<*=IKTOp43c
zcB8ZiDyBs639V%A7PGiuj1cm`**QZ<BImAiAmOa4N$ZzC+Z41?2%m$qWJP4-^I%(p
zP8Jx!R)qwI20=e)1iHIc=?#z?2n%Fz;c_*QWR{~e=z;w=P1t|a+qv|IPI%wne}V0%
zP6*}sWLvFU!2<LJAnC+<(4#WcG$|oflGY()9Z2No^Y<$fD++0=o3eh*$=Kinr}G6x
zHxPWtz@PjX=Is04{ymP4Pq=dRVFoX4?d5XL_SQDj=?sOU8(J321+&?dtz9@edX|s>
z2Gp~LAO73_5r?-<(9V*Q;*l$x{L$}zH`gw1^Yl|U_y@oIBE7!NyWaRiOka5;FEl@c
z-@3#<{IMV7i%<UgSsT-6O+Bvp#IN4v@n3tM<HIAgHI!A!m8;jebom0a>6oqU4Ys#7
zc;(hUANjQ>Shd0pN<PpmkAcW|;{sY6R1&$GHP|VsMD&9wN9PWZQlLylS$+jSkTq#8
zWpa?ViXu76q6Cnv+E|f8a0Mx8NhEAd41r(_rqo#DP)bvU$RtLrb`)hnSxW|EjRXf`
zz>x3}(2f)mF+}R3L>VjXFA~leQHr#VYCNMDPnlkN023U`wx+-nhMLk<xN?e0OXASr
zn^XMih<1LPu$tql8s$oKS)&w$#R<h|!sOyZ1mDpv4k@RgQirj-Xl7@bP)uI>h{i$a
z&r9nt5Qm238&Cp|T=yFjSuQ~5$ek~6oyFf$^m6m1grE(+-nrNR=33{gzY{CMnZ4Av
zU%$oZ@BAV^a{Yr0|K$u9o8Q4#`kh(l?bqyet`rt?*3DU02WPSp2$&<SFCisJMwRP*
zdmkFhs&qC3$g;QV2*P?37CwW{W=Xldqof)_iCl_8VLaHfMkL&UGgtaD|9=EpNG+s3
z&q);7CK1_AdU*{-`Ek;)y#1yrzwp=J$OrzbXSw#NBZeZku2MXojjXH4xDG<=>LOQZ
z>HK_%6PMhRXVumE{i@vESFi7?^TE?J?RxBib1t_BbFfk($A5BsjMg&4JUl!`Da~xQ
zahB|^RtqK*QN~pYoZaV{Cl*XU^#lCjKl4F;<{$kUC-)Y-{jrPufj4i{p4{P|e)5<n
zZ=Q1Zv_&^}`Ru=5^1knVn2-PCju1Mgvu#Gx9p)$Zh{_X{##%w3rjQj45E+towP1Jq
zO}z6RZ)P?dF&d3{_`$0jpRV}gOLuWjV7!V=M)ibxG?F=-QD-etl~oap)y!ry>Z)Qi
z8c|mzec!Xa`4xzS3YgXflW|1^Y*C+yy~P-b8I5FJMM{CeOG#ZAn@#K`LpRQ0jUcF`
zrzXQbIk7T8XMJNOF>p`H=9?j6ixOumLh7Ygi$-aOwhmX;Xj5XV4O9;{ID!u-TjJaZ
z9|qdROQhyGVzVGLM`%3N#!l{|N^4-~78qyH(@h~rRyD)vO;(3Duqshj8`%0Hwtg7x
zGWa=1=QQ-BG(Z}N$rA~};V=>HQ|=Weh=GVgC6(RY6e;+uR@$MAyf4wLw%<_gE0F7H
zv7K0L{LYTO6Otdlb90~n^gsJa{@DlK!>|0g_b@v-;%nz5CK={BA@fNEnup#h-3MRI
zn_sR@6iQ^t<(IP2E0c$roK>}IofalY<~$=OrSq9bo;7GCiAhqKDW3A{=KIpb$f_=o
zZDHM-T5l08a3bH+00c>*)&WoShJgsmAuTN{f5#Gl`JUw$|HdP{>E{l4*FSz4;}cP@
z=NaoJrHB%)0R`)JinL_%EF>w!@+2dP@|@=^0IknaQm^ZJI}57FtZ6hJV~jbI?J(Bj
zoMX9MqBRU%ONav(FI>Vp(XTr=xJQhEjg5^gcnEzH=&yW+oe!;e>BE5sE?wsx@3E-V
z@=L$=EVoxBZLElcbRv>la%=XlxUg08?eBXJzxnB>*xD|+_Q+d#=JAi@{Z)h%tGb|R
zTKSA4B=0%6x6da({+nFCw#Vk?jOD7Q?*rfay&poELgpiJt#~g`>2;e_og=K}--DQt
z$tX_t%#+iTGd;5}kHS)lvYs)TY%zoZWu#&-Su`y9br}nO7^urJN?GwSULz%xf+YR`
zYvHanJJo1Ij3M`ZO{UZ<v@Oxr$&ED(U>#8j%n*`zA1FvkAY^Jy_RD=bXC-qA!PEDr
zplbB!Eoj@K-3t_Yfw3+J!4BW7Fa+Fe3sr403>_)?d^0;z2n>BoRClE(J$ewM1q&ZY
z;5{m{U)B2N1W1Mu0xC)tmDb`vcvO_=9%Ce~9o~jPSiV|xhNADj#&YQY@vB-S%qsk@
zZ0iFsA@C>u_$OHX6T|a={XzQAo~tnZUM>NW7>p_JYxJKFB9oG#uTny?6X<OrIk)=K
zc8w%H=Q1~Hef_MvH}^e1`G%KGIOb$cQg)TAl{JA+<zaNvV2AZNlY2m}+jUhj21cM4
z`FHCKKjg27Y$@xCB?p=_+;O0<HJ|-s7f3|j{o^mAhL|gcTp4kGT=@;l&-orodBcn6
zf4~}0Ac0h>m!!0i4%2IQ_q5isy|c~M<|dQbl%Wq?xO6GE!z?j@vzE)3FXm2N;L62|
zj3*P6*7QS<F^0{JP3o#dxFbtaR7^K27VarN_Aj2G)HTn(xa6hVr`$a3aU;!WvPBvK
zRjfFjuXx~16aK|d{tSQWPy8vKdghD5H`&-?WA75j`>(JD9&}Alisy5n6q9gpPz3Lp
zFBTYU>DrzUEKv>XvR6FhQ|1*Rz9KQV&d+cTF%e0E>_d{4rs)`!G9P-C7M8(Hb|}gz
zCT6&#29=B;M(U0H0VTrFpj-{H$J&aZK!t2~89-@=aSpsErGPCe$;gACm?9BZO;O4d
zvaGey2Ao2f0;L3xtaGbNS$PvfzmQEh1p3~Se2FbADMox~QD}x>h#_K=$f%@To>k?D
z!5F&!5Ifp{YL}==;$k1~muMBR#^OdhDC0;eGIWP%e?)YbQT76X2vN{rgn<-1Lp!H$
z8&EYUPYfMsC!%=9<<dNY%7b%iAt(+FN*5>#F?O`=;=ax1>wiTwJD)%1$zR{)Ly!IL
zyoLwH1pdl*e~jPwiI@0qzwP^QuPnYMf`~Q_Z3@059Au?Htu1S|o6@pLD)MHi^w~T?
zt_F49hC+ZEYTX8kNTAS03W~L?Ib`04&V`CXrS)Wboqdv}N;DhK&!O+3SXU8Wj?|hs
zuR2WrYV!AW%PVkhgPZd2iv07y6Mt-vlLsx|@qc*^)5Kgotpi7tq)UlKw)C0BMI|Zq
zrME`B?^u$S1Te|-#i)oO@LJbsJmG-{A7eJ15(qTQmPa3Z6IEH_tYb2ru)V!SRaRJQ
z*x1^^T6<P87Otc!#)RMna717<uGzV;$8Y}JC;7~0KF`semZTE5lpLQmhS)YN!#;Z#
zwz;%j^2yKMWKvG};)}QVr$6~G`L++dhkyNxA7d0su08TiERGKuhF;otaXp^_R3fN^
zPAOX#E5_3a)*8m+2}ef_ZRazHL-yKrv3h<z<jO9^j3AX{=RCk`nx-X8B$0kq(Iu5c
zhbHL?GAAQOj}M991KJsqv69h;fVLyN?}>P<G9)iZ2o)15_L5y$Cnb6q(6+$js1}iC
zuTlmKC|jY|wu#C~>ul%|ed-oS7&}IV0p(99Vj%f3T~c(bd$g-#Vv6)bPg!n&ze`*m
z5Dbh)hGL_@D$R-`qMP7T4gCP^A%sRer;R3|1O--%P^9dA@``9;F+wFv&;u#<6uu`$
z=-ZYMJDfG7G?3DYh{Kc)Q;dmzpkLk-#D`VH!4rIoHUZs>&W`$q;egz~Iu3q*o7KN>
zaQyC6BDq7)J08Bt5B|ij@Kb;3z4VWVvyJ=rcI671HKtyX)?ANVK`WCv9-;+8N}?Z@
z6la-t3hT`u$ea*`0;^C_<^?gW5jRpHteNU(0cefN$^-bdL~bN8yK~CM^9P)~c@LRi
zwE6fwtaXuN3j}!_2B}oK{C$mF>iIWCFxS<H1|l!LW0Oz*xvRYEN56<!C6bXKDDh1K
zV9fsTDkYK%5EV&Tlo7IW^u26sIsxIl73M?U-Cp}=A46oZT;PKM4TfGe;i30vtynCV
z%olUU<MA2zCB%UDL9%$Q&cqx;2xz0RDRS`23(UXp0{*Ze66LhVTsh?<jQx`1lajVG
zl+~2B>*<G{Pkj8-y!&16Vp3NeADyzfv(0$6$;tlh`($j^(C+n|NDz1-x1K`Ge6=vp
z^?jbnt!Gf#O%+I)>n6zvUjpPgLy{_;J1TNszdQ;nMNST1VN!S*xG2kF2zgTs_-=`|
zj&in3hzV;9Mhmnz`hh0(6xLzu8GXM(DTnV@=n(MDoEUr2;jtD}0anb0bXa0F=#b&B
z7>uh>YJygQNTlzVIISRBQj#ox>;@X|xp8nzzY@1Cr7cBKV2JqRr%C+*jJ9d|2HTwC
ziV>>ZAgT&{gJ13w{Yp-*mr^vD5=F%N7Nad15s0*AfiABBGnDs)D7sd@Tar3~#YL})
zeM^WvosSG0Fx$FH=$0%FZ&Ozl^>|DeJVU?4x<Xony}|cOUeECF>tSgkhd(#v?xNzs
z$&%jzK-QOCxv9Y)eDBlT{)YuW`)%LB#>v6&SqM>p_XB+|Geo6y9%@VV5QDe@YFj`Q
zgCB9#%jm;|fJ{tDB;+8HZ{Qe-F=Va?C}kzMC@tWZ95jd-tfI`>#vz(q(soi&viwNI
zHIaAyvuF9`zc~UW1i|xphK4v7LZK@_`}^Ld<xA%!YqpPU?71P5znin4pZuO(jy4p3
z@V|Hg<p;8f`dKBS(E0d7{sJFctgF%5Fbpe53G1ZYmV!T<pG4y8sz9`gbj>M?`5`aA
ze4CS#Q;3P8EZEuGWoLVf%a<--RG=TcNT!6qV!5L4I>zHMg)MM}MWukBAF==9lf-_(
zp@;dPh#?~*$k~4$1!xjs@c43v(RhlnC4-O5Pfz*fU-}h(=s)}m{1^YX{|4_PyO*zX
zy#ER+opWH$+jnvagdj>9(xQ?%SB#0i?Xw`~dfpcD!{nOKfHnqaEu+zxs&o{Eqi_XP
zRZ$cL!_X0e&=X#jiEd1Z8*RavBXviLo|HVMm;i$WgA`EO64ji94B1hZA-2MPOpy>}
zlZ~-K5kQv=NuvjkqR+DgBxBbaojDPah)NwHt^}#1qG)b~p47J_5+OxYN|dP~g`UB;
zbVGwOj`8RMg;NC8V(JlzK)*O(?pw-of~jYu&@wc~1mER(V36k>dThHw>4fhatS!XL
zp{zypGu}k$OQd87u_wlk(0BNuLk*so28f=hEu}3fLc?(U0?t^*WgxM_x8TD7B>d15
zGnyrR1GTT#O98_j`0(Rz;>x#u8b$nF+td@Xg?-;QJ;mXVZSuK4`zXcH?+rR@y=7`G
z>bnXcoAJoKljnPRGbiPZJ?{sSUqZ;Zvy?agl*x7ide>65rhzlWU+y4CB@%N05q?Qh
z>vn|7!%|V1$c+H@K6lLMl~Yz%48)Ezgq5m+0L}{XYUm}1n1ECs>of1~d-v-zu&!Y4
z4=7U6-2RqLp8c*3-uP2@3B`IdTR$}BDk~3swZ&>h3Joc0tjg%E?6R0r6m@Tfw7+=m
zz>TBh1y4Ww62=%-%O0f)b}#L)wJqep#j@eK8!t1iM~Km!IU6R20ISu2n-!$RT~v3<
z;VZYecX+@&7Th@<@L_!}IBQW$)hIKd=zE&eeP&xboSZC3(co;sGtYgIr=Na~_q_Y<
zeCpFrFrLnsZtXBXJd~<Oon0(vr#%GO$Bfd15a`=MfJx4ZI*7BBWr4FI5v8=EDhtY@
zpsH(Jp)uN`R6=9&d`*&``Fz20DZq$VsX!84w_tUA2ecBFta5a1kF#KHFT#Uptg-a%
zLL^L_Md=D-JjPU{I0%nKX}TCNy2PXbs~sVBX9g}Q29&d?R1%dU`Hq;YfJjH^PIF0U
z&xVaIIr^@lS<KN&;Sy{WiWnwzeZc#H<@^ve+hsg?1iCMxLQj8qi#XcCT^wWMKnep%
z8*DWr#r6y!ptPbJdWfFn2eeXD^_Zj#xP(oC<iCLDDyRxjG;Krg1BzZUq2w{uR9e}C
zhtT7L5XzLXwB4LAw3JoNvN@qFH#5`t8;b+7UT*05@L&B_-usKs^4P}dm+h|qPA#!S
z8u+2_{V@Oi|GVc`{`6a^{OWaVC%-dGA*8+FGm<dhh%w3#+~{<cG0X5c5PUD!h!Vus
z+Mp=qp?S<1qh`(QmiCj*yKX$2mB^z~^!E+Z#W;<#S;D%9F0+=X6xDvuLm$4&=YQ})
z2@q*LB=10v<-1DIw|TZc%WbtaxW;nE6**h$YdXW=)qPNPrTE1E;sKf+&0GKBEwENR
z%|j5QEipHjbCTZ`(F++<iCCn}L%k?lM32hgoG;l}f17oBddv&Yex9nTGn7@aJiW(r
zPfNA2Zrf-Cg}G+KClt0QdE@mx-toTcEEliv@h4y5ncE%5E6>sKF=bIwRikxiJi9ow
zKA+O16d1Y{I@Ub?_-9!)9YrxB_{fj`=#TMN{+EA=C%^D4F(@8<?CpI1;~zd7h_CCr
zgrLPzgin6rGd%P33!I$HdF0VY`Qv}`k1?5yS>xNJU6@cgu&)6cAxLE##Z^z*ZmfOn
zVYOQ2_ShOY{pFUe>4t%bCh3|^mze1;T}&LGo^U*0a&$Upz@SGvlv~$O<BND*&~%=~
z$uS`cBC2g$REmV56CzX?GAawADu^Uv@c5ylSuJT=k58JQYoe-1+L3IHu4W`#k#tFN
zH3>~hnxqPXs+c;<w5%yyLFp=D3?eZweG6*(Hlp36+PuJM>k<^33?UM5C^tsiQU>0(
z%%6#3=v(~Y30W=G554@(_XH&*CCw>qv%n7n)|F^&FxFwp8Q7ZW2Zmuqz@p3;%!r^$
zlqv~fAViUcN0$Uuo<aG)QCIP)1)uo&2fr%YctrBv5B;%Eq5e!yeD#sv@9t9KClg}=
zZ3KYgO3;ShdxpW&b-kcsw8a!7l&wk15|SVtW6Z?E;AP+)LmpwQw~a)~$`8^Ti9|1i
zI0-mfHf^QF`Or9-p$N%@^GT5+TFa%6AEJkZR2~_690#BTy$w`QR8Wp`0<~_ZtXo;?
zzTcgrqw-5|{^>~N<;#N4{mE+_UoWBSGpJw<eYPkqFxeQr%*~ZWX(tI?-o|2z8J(qw
zN?!Z*+)8T%n6wv@9M>SIC1X@py48@a9b$|WR&)L07T@uXYkc?HH`ssqS$^YlFZ0B$
zj+0gd7WI1O@D*QMu(bkoh@>>&hYUhciRE&^7ryWeAN<z$(s!P+o-*0k%x%aoUB+3A
zEqM9n9X|j0XSnssUFP$Js30UID(L}25JN8S1&GwP9jj$a+qU#sIm8c=jQb(~zaQ`+
zoV5mDl}M<y#?>QiQ8HgHh)GjBIhm^7W=I|F>59>`#MpvWJIA^j<!bu2=k6=7aKS@0
zn&Lw*y7VZ9ZpCbC3+)Puq9(<LvZ@%3H#j-DjfxRlj_@&MvB5c}s&Q^ih&=`)`diA9
zFf99oH36*?E+uA8(GQA_Yy=g1-1rKy-o?z`1hhooaD4AMEOW|gN<k7XhN>9iiXlc!
zZV?zQ($pwLKeR*werPGYM~!Fbc!-V=y~p=0zFmsfHA3>ywgfY#AC`c|#6U=cV61&0
zg-9GcMNvT*2;G1W{u^upa=r{LpZJfznQ!|;w|HoJ@Va%eUhfLPo*wwKKl}^)$cO(Z
zvlr)I^(q8~Twi%Z>-~T=3YCHk&JoC^@{yt}3DKX4*(s%1EgP)%6s1(U-5`VQ;62t@
zj1?WN6g$y>aw1+D!+_Qz36hj0<-WBs1wv{Bm4lAaRuUy5+G=*6IO0A3^;7)j54~Ah
zSSms6$)Z@Grr<K~gBHT6@IuaV;T&}E_f-Mw7LHU+=Y1;K>_cVw{P*tgfgip>G!k@F
zW)3Eh_MLX57_%yO4xDQJw?GuRA%i*}3cubDYkgRpTbZqqP9i86@i8zN)%=D3>4$h=
zx8$YgpW%~F-r!TupR(!_C=o_AR%6tCmC*Xq`;y2eFzb{Nr*$>v?wvcd%>r#4RZ*jr
z;v>KMYy9xv`OkR%#!KA1eV6TvS2;P{m+LE6k}5-@^;$@hL_Y-jz9$TxwK8&yqAVgo
zOJ~TqR2*wAOpGzh=47jYB+q*e(0G*q5@X_Y*)pAuXpashpoE4r-6NnFl?J6psJ5eP
zPS768qQ<wvvj_<`cXy~pYtBSubMGozB^HOb2w9+OakNjwuvpGHK03hJ0-XY$hH5lp
z7&>q@Atn53E*mnC!hnhcG4%Lhj!HF6KL;PNwq#@iAtaj6QkJ{8!qK)%5ottodU%J$
z!A&L;g;o_QB)kzLsTdPVm!#x{1Zl05%YJ|eeb-?804ZX$7W1F}i^SLy80h*1NzI5#
zB;1W|aYhu12|Y2i#N_FRQk0A{|3hiV&<l(GEK~aiTnY%!d;ZoBeJ_9NZ+whzf7c7Y
ztF0`MJoo(@clg-<>S<p5YY$TVu8r`&{E}g7BF5a}k;%BAiBuN5%L$7QoouIT6@w&(
zYc{jRLyA~epq0U^_H0fPyh!7q6@FNpwR=+VXXr;E^J)oDXB*<VDC3%kA{F<*13!I}
zlZoM(fA3+EUGI2nPxNzc6Wu8&FW$Ytd7#LMvNJa^b^cfd=lqd%uqk=vLp!|ks|V~p
zG0y{aok0Pzp&8wu@m-SGh~~XsT3IO34~sDf+3@w2^ZC4-g}RJXTJtV?KX7q(i}$|k
z5#IdZ4n;cUqrY*Jk3V^r{gZ)VEj_Qq&O$3AfSmIqz3*J`CFf>+cFLk)=sSkK!&RCP
z2Q-@HV$R?G;UD4s-})^)_v8P9+13taJrXC>v%|m7H8tx7%1EN8>w88)&{^wwhG?Uy
z^K+`w*?@d4uOo9iSZj}+{TsYW3ri_t3UqCY(u%rrj5aQ!M_UY7PEU{Ux?-~T04gr9
zBto~K(3<x2kZLr+6(elDfih6IiV&6*MZsvgNen%!#VIep{0cAMyu)hVGYkpqrf5^)
z`<}z2yL2JaC&jAQbUqNnfR6)%k0kFgS}}5xeYtuIjKLL-sSS)xPa0N)el9bUvc@{g
z(9XGg<4M}pF+-Q6ebBb7=EuyB@6xT7SUbX$V^mS2Oo_4;%9d!Ixg2O>NDR%As4Uf_
zz?lJ8Z{TK^ag!}VSO>eF5C{696^|1Iv0V^-i&hGw9eqC#hJdZNaicAat-q1GSZD(2
zA0+<i_kAayyK(VMIOBI~Nh2olwjX$g<bDs%tWt#FCFxKiL9T2`(xMp`V#1Uok&?*1
zAc`bGA*L)XZZn%)v*u(wCsqf+543GhvuYW7QA0>c{B&XzK*-=l2sI>iPMwG$+ZITs
zE>ads{2+`)C(Sqi!xwn>-~JqS(IK0?=bZl$1d-J&<h}?pV<Gks8}MEJTgy3`YyG9q
z$0BEg^?_kF;d4KDEw_SeIZp~fu90%}Ap>Hijc~A{oLjB*+3W;mQLa2A_`defN|REC
z*rp^VN<Px{p0aTKsqgzBfAxR%hj{NBX57C0GJo%%e1f0+$QQYJ*y5wiN7NdsoG&$P
z&3aV9Sp|01*%Rzj(utU)5=|Ngyd)2<B*YHyJBFd<)1Q2t`3suw+x*`Uo0{v7yx~jZ
zQm(=TrJeG%0A1fRgyFv9kZNLWp|P$?QW6ztYb|A2QdJ@<X^iNlS?6#?aV8r1<(tAO
zVPZ$+HgmF|Fs?)yjV%qSoYD6Uh3**o1<E+OW<fPpsA5D4J=Q3?ZjR9oQqQ<7@j;}6
zecwP*T)cFN5CXQWP{uL%6)FuR5>A(NgCjK!zPU?PR_L-onUZE$G8oAQ7AujFuF6Ub
zr!p>nR+coQnnethRf(oz>AiTLYe&)}LelhIR&os<t-SDGQo!hfM99)AMo8i)j5VSg
zV=btN@0Jwh2yFt*=@DrdNJi7PbK1c(t7<fk5PBBvk`z7TYDUxrgDo%-oDD=g&KmqM
z(00vztm1FXWr&13o`3Y;zn#DQHxIeS!0%?=<$=exI?N>lFMTx~9~Gp4))tL9v!zwq
zpkhEfC$jFLKT~H&2@$-@$S9Qu;ZX}rcby4}Z%rU3aiep_NRo0kK#%J*8Zh_($*;r%
z-DE|CB<&cRIVvfEPEp8+8hlbb^sjDn>s?#C@}b>40AIJNasyIXsQ~h@n>9}$qZ(0b
z0-^fS^P=-E%9<Q0k87{jbYm!1J=!^u9QOgMwGb~!gfK9qh}9+!(hV_XA=XTz7KM}7
z-ojeDK}s_x5kkaS%e9MJyy>Am-ub3WT-+J)#M4jlu}{Cqt$PiNl??7pZpW>W)Ax&6
zDs4c8FX49yL0X-~k`xg%YZ|y5-i^Jzf#OpV$o#<I6Fzu)-}32Se}X^zCqKw9{fwd<
zRhZ24S^Lkg<#5CoPd<|n0wHs1%zbsk{nf;oZN`0HDJ7iK=pX^nSW8(HXO+#X!g#Ur
z<%a>6Do&dQ8#UIiaIT^#N`|4QQ<|cfvblAg)pDOMB<kpit9{xoQqOj<^@O2aqHT?~
zB^%qDbbZUv4b)S|=I%v&u=GI@`ctNK^zDLv@f@irFnYkJ5vunXmdvJ`6r){|k94a!
zyB8iJgaI3a41IzI0=ripW4qYIn1nSoYH7Lk+*90s`374TE>Tt^VpM3=Qy48$^hJsH
z0i`3#77W2-mBQ!|!K|=$MBz$|$ppzT5C>@!6vc#g{vzFSz+#z9rW8d@j6KmzI9;7G
zDjd3!3N7L2hZUx9GNX(h<3#7{^`J+-URMMjQgD@FO;J1JEw9V@-^tgyymid<`z@#c
zRl=RgF@BYnLd?oi1Z^4ofE8nO$>tP<lrY9%T@5KR_+E^rVow+bj4jYv8DV{Pj4Lon
z;e(LfoONhd5<(KleF~5QAvAf005NG)(!?-et#GXfk*-}(RHHM2PeC4mP6=Jspe}OW
zZptQ8#|L2TS*s~ezh(Ef>%R@0BeT?(`FY*$l9txu(v<llQ}!O8<B??3S+PM_UmQM2
z_U^QiU~H^xYU^=z-q5vqeHBQELO@O=u3g;akABCyc;w0q{0Yz9yv>jQ^z+=@U(t3E
zYYbY6)$my-DyiSnvIA>0UI}pBz`vOsN{0-iOI!)NTf4mcxnBH`G<fo?Oj(8=mK!fH
zJop3y$212ASyC`pP3k-luG?lw!7vC|Bjl=c-JV;|yVu8b|L^WUK4UB)2;IWkbNKSB
zXklqoSwZE_xEU%0`sF=Fqk^av?edgnImZ@?)$)X{J0Z3mQCn78{5<-o=oZJA#R;Xe
z#5ku<8k!ZMUr4LQjM;zrc~F+V_YhlJ3%>91!PELK+u5F;B{lx;9cHsTOn3HBYT)GP
zCP^iHFWw87g5LLxw;hGi=%~@ou{b^C=8fk#JZy-j!IoPTg{CfR!LqA}(uK?=lp=Yl
zhEzEslCWlyVaz70Or%_i$c+<KFK|k2IG!)jJ~A3_aeT5O`cuX#VtmB6ioRKhQ(@)-
z5aog-2HNGEuj=OZd$bIJ+y82tfAO|=^2QI`q24^?3jG;dV9i-xH#q*stOQgf-uX9v
zosa(5x4`jN>->t2(AwNd<?NW?WuMLc`Y;SR$Rvio5jl<oJ_gZr0*2s8-edI|)~kp~
zL6Vt<HZ}dwqlP}WE_{}7m*K3BA(PV$0fD~ugJiANYLo##v}eaelH(2`u=A-yP9M0y
z;=-6zmVk%(F*LoXl(@XJ8<l6`NlKVA?XGo&u?`q(^wj-T2=aSn*nj&bd!If*q3|&h
zQliinr8IrtqDf~&%M>EXi{#S!^h)99@xxlP>$hJqMaJVX-~27_<KaEW^=(6KJ)eE@
zDL()F9bUM-;NY|st6VFADV~QciETOaS%Vbv7=g8}*;%EKQ=~6(oz$ArI@E)=c;=a3
z!VF^JOOm#4QPd<1qjCn((Dws<*Rp1b=ckclS+m@gu+zQw%oj^m%T_o?N>F8I0Ym9L
zMv~85r2ciD<v>a?&rT#5tt-1%MPUum&@NXPV`#f2zU`>SV-{UUH}6rg!-kecYe-fT
zntQmSAn1a=T~bz->Fg@Sc#G3J&(rk-x-3XTPrE!NsX*J@!`!<q!zQh$D?^F{jdlz%
z(RBmu{E)ussmh9SRMECQacBrvuCwZv+}gjz{CGuGz|ebK=?KF>+b7DWp5^ib53%#$
z+c_B&cW!)^VOVmyN;EGYf!m<2HM`qeC?)dLDFvuUf}M{NAbe;k%Nm!f4nMT0Oor0s
zlw!0kbCPtMp;@4kr?Q&G>C5al$FyxPnV**{Z0KI5DhoUV)o6=krX1dRiL%nf-m^HF
ze~IhYH|o+vj()-LbHDH|{^cWv>4iDFZ`>wK4WIw>FY=eSp5UPiE8KL>MY=r8INxd#
zUrJf~?V1${`1Z$c@U!3aEzCahJ6qv=<=53&?m`TpV(7&HDk<VHNDCnjpfn+fd#553
zbVC9au&%@uBcksa`VM6c3L&5><mMjw9+f!})*48WphbEjq6x#0xe_M7z5r5;G59FG
z@SzhUs}w`#QA9#a?EUB`x%AJU!9RMLes9X*uHx46Hp|B<!UI#{j)n^(aV)g=%et?F
zgm`|)TBit+^O|%0dvQB_<<T*3A`pF^ZG?fo?M2FjEGw6iyl1lVwE}Hwd@m$sYYctY
zz2;nOG9L5hx4n&byz^c3FMX67FCOuk8>ig7vt-o?5zm=(-bRdaZq89jYs8kkEv2*g
zY`!v6Oemu=LHfRoA4tlelx2Hoi_#Sw&*zj?N$3}>SsPuyq_7it#^4#cj<#9lVk@1;
zVh~StEvg4HLkjd=%g}csZK4%U>)euGcR<7gUFErm@`@DX_+t!sFSZkH(<1&w>Qzu!
zAwbihCWbIH=%jGt5pADH!%CDptzy}0R!7HdTo|DzyTstx+}vVu<e48YSe`sfHL6Lz
zr>ZR7@=!i}WiYnHxkT3vj3+gP5rj#y6MWy$cMYH!O}E(EEU}|0ZF@p#J++xql?CI`
z2G<^W8=ewdOo{y*ogzL&jCJ&br6>!G4$z&jw>4q*{_o<!qa`RsQI06bHQPH|)Gkt}
z7Lw1lqaNo<tg~o5);bhE=M7odF~&1EvMpzXVa{^7Wbl!~8mie9R>%9Y8DL1RU|2O&
zR&G#TZ*p>UL{g`uIAE$Rv=*Sizb#7>DQGF5Po(E9w|{L(0=E9^P5!z15T_3*Hm|Ih
zJ`z!HS@7<MZ?n20{qE`RA#d3_!0bdOdd^mU&ie&6ersSE(D>r3+a658qn##&L9$70
z2uTZlIYyM#L}dtrN2!FXDttc(+DZ=$eV}U^^w6VXLK}#^Tq6k}w^D)^?JR4J^w^cl
z?TkQLyL%dgaiz>#VgREtTAginwooKpVHBt&eQsw&lr*)0@sqb1KebQYH(W7}<-a^6
zj10{L{M8A|YctZd3Co8{mUaFRb_@KhAWW*v!e7IG({K684vIJi9ffm{TVTd!_-^X+
z97*gTRPwp15*V`gD`E&(SA3~-{%cr|KKceQj^o2aPQwNp7a!pL4{ZpFF7qG4;OT}A
zKS(z2L!j>mhM{NZd%C`(?|X*cXJOYs2$8;RnJ?xd{Hk*jlv`J*bG6@VSFdn%JeRpn
z@`M=Af{fCTqz6U}kzwfQ+N?Kr-`l)@uCXSJh9JgQUEfd?nSpL?9++#ac4xDj^_R#=
z$$=^bt|+kD5_}MT46iywk94ae25&hzK15>~Z|+c7LxZL?BQOdzE6`?2QP(sl$7nrp
zvVX+T1QedKfJ(Las-PPJ3<ItlvmAP~_Gl6gPro>&+Sw*}jkAu@8MLnP?UHI-V(bW|
z4VA0eyK<ejIiYC+_wIa=-D_{5ZC2QNiYp(L4ap^3QKQou{<dU#;Uc=ef^{Q|^2A|D
zY;H5O0s<Ifg!U9$j43xS5c`H#?i_IN$`PS!P+F=SQ)sHnQB>mRUYeRgX%6l@O~1ND
z=>+#ZYVUxt49e21=4fl^Q^Hmyq3b9|oAg~rv?X*WY;9~nSqqZP>DPSGf1|8S&BEsG
zz8E0jnLZ7})0+KXGAOqBmt3SbQdYad^Yb&q@=?!~i!IHcQ?%c4lmAG44!a%kBnk$$
zGi=uYO;a(uDVyb2r%fmoMU0vdJYfhFRgJTnK1P%ka-S(mpvP#Nhv|{ft)x{DlNcDO
zC@msHjX^*oO1KY#<sl3`rYZoDr_f3fbwL^e+K2<DF`{W#)DsjD?-Mch7;6cAPxL*y
z9HVW8Cg5z|TtZ9qvdvfZF2>E6rX;^VJtgcP!6%Q<+A^q)Ktxqz%4yA^?Kzq#jvsd1
zefK8mYDL<1#M(k>VeIm-T>{JIv7kGX)|l9%Ld4?)<i$Wz4va(jLA-P_)RvgAMlx%y
z4aU6o&$uirE?l_8g$oxUy;+n2(s~0IM*P`;KC|3qR+I>F$Ote|5>8QMme&6EUDxo`
zQ&02E)6cM6wX|JJ43Xf47-@8kbC&P_qu;}i|KooO2~qY&IBHUFMp7IEW}U*><Igx5
zMr4rI005gA^W?MIwhduu8A2o0<n<WJG2RQ+Vm;TmKZvBYFTEB3kw#|?E?Hk*%0xjQ
zJbls><1Nqy0YevreccZ|mc+1FQcQN~RxLV3HaB+Y`coFeoGuBb98-!Xv~l=s3{w`4
zqH+){<JlfvD)B1e>@rKEN5-{5?Y;qH944(W#&UcdY5FBT2BI2x=5rqfC6Uw>6(JNT
z6L2<QougZ=FuI^D96=YDVh?R=3>`5+(d{#&Knxy*X7C<s70Qh;%Fwn;+SQV~FTc#e
z!67aV7@IJK!Z=5v6V0kc;hBvsTf5gN>ML|@$FftzcnYCIIfpK4TvE8gpzH=+zhX#=
zYBIxZOrYyHefdSk(+%PfzdC!{*YJ|<eeK$vu<M~Qfhy#@s8xh^V00G_8MFEsVEicm
zPu|4QI|AKCWV7gbhgq<i8qy^3;%7He&kuZ6PjE_NLu(vbIg-*0Aqgptgo=^qo6M|M
zq`_ltfkFo9Q6(WAW(PsgSSx*YjkJGJ3S}Ke$&g*I8+<VXXR8urOHr`a8k6C!N^=%W
z{SeT`VVwdz$p@l}3>oGdW5ODnTP6c>IF(OGCv;&j<(Lpex2lL1r6NjMbd%T{VCT+2
z`_z)tpSnY;HK|gpE>w7{P#0=~h56JlX*?G{zod_OGmJ_yaoqz^V8$t{;5(Pwf&pb5
z###)SuVu`6CjJ_hF`Bw8^5f!}?XW^iTTdWPxh1qF+KEG?(uGtBDU<T@<5LDPq%`H?
z-Y)NZ-?uRIJ<H{a)vBRuR@^%{;Ugb=immN!#-kB;_xIV`-QvNAF3}G?T`Seie9>}r
z(ja!ik>m%AHN-&zl`{p}8oI6{#KdS)F`10Hc<}<4FJ0iumCGV9Ycxt3yfV-YpyYng
znNYY6BwC3G<NCUbJ~F8ECB7kFK7hnT(}<oG#^T%<KQs*8lBMeKJ~2b%oECIP^yqrR
z!O1Dh`2myCaba(R#o;M^+oD2bV{1y<-Dc5(=>{yGwpn5^gkq0QSB$%7(PT`wwruZS
zCWe;9$%58**utTW!#RyJpwtwn9om-kUB_ZMP>l>B2oj>6Twt<k=@tzP0cTH0{XLX&
z7-cA&A$Eb_W>_nkjx&i=&InOcCeL_Wu(esC)mvDe%rVa5s*;1-N4TQq=-?hF^Et|D
zR;vOEO)z8J$lz0;Drdx@l`mCNPIg5SWp#^_)0TdJpULiJF7E76x5tFw&oI`%zrOss
zpGm%MQz<$oEzrUOPx*_|O%GUplby4_w}h{{M~I-AjPQtz7I<m!L=v8ajl#iGh}D78
zhL9i(vSF5mBlx^YW+<gC3WgL39+Zwm??uvFTY{=dsTDN^0FodFZGk7`%~-UwMBver
zj6=o<GN0&ki$<!Pgw~G2NwD&rM=6g=IS?m}(g_t4kPYGtD1tELGx^Xs%^1rhjtC)=
zRuOyC;JbnJd1*bh{eV#sQyPXs?o+fv6V8q~Wrt6a!q^($H+UZi{eZHT(PRS?@@(XF
zd`{LHs<OZuL(2BkDc%pwm1<Bah0h2zviX-jlgXFrFhpt7#fW$!X-kY!p@bON-Wmuo
zF&<Ss^W05-`NL0e`O*b``e%NY(~}c^;0L~!@BYxY6EYt!gn>_f>a+ahPyD=~`9h$s
zOTP0vzK{L8cR4&fB6v|O9tO|F%h!1Lkq5YXaf`j(ZOVF#)(&kX6$>fJOeS$QZw*Pz
z`$EW7mj#`qs)-?DR2Gml2A#fwl!-Q$@#dcNx<U&vVa8JiJa&0R3YJjKC|triLrN{}
zVu>ys)o4T*dg^*g+#Asz-$hq7tKjL5Pf@O<E)(7>Vp-Ft0oOc-H3dW8Q07c&&?Wa?
z{tU`EY*AuRU@fki3NpYC;+39Yd2)z01sA54pb{sm6|>0*lUkMs9_2<T<0J?rNJ@m)
zFiv!~VoTdZl&SF{Vf_i^coRlb;;<sBDaty=TTZl$QpafQsAn70lZxv~GR&eVh~5*T
z!y<x_I%#4trt4a^H+IlzLAN-fEDXEbQ~IIe=<tX;tDbRbaaG0ZX3zfq0?xaBU&@D;
zHtFx>GKIJirVvOfqK%V5^U$J`BITaG)?(EfybRQp)vOV_t{yYAD@3L|(GMK~gI&K_
zWds--TMWG*lroYAr5XALKdjD}#~^JOZHmkRlwc5opqXq&=!6u6HmDMD@FXKjyn-ka
zFZQ*0m1N&p`T|hO$gIz)Eb!+AIA9doB$7+wiJ~J)M2Z+A5L9g~DoKW~l_BM%kfez!
z;<R)g{2++6!Z``ZKG2@dF*@%}Ygp84T3I;O2=Q+{qtGTJrZ6Z(3(IB3w_s?cLQB~~
z*d&b!hU~iw>qjS0F7tqLbzKyWTelu3Da9j?JjieS#&6(?lIz#5O4c70MOjl6GbZEN
zIoE?kRnK_;`#(TkSJJ%;GK+~ZqOHMMgEJ<#2G?!zL{bTzAgTl^Nh0wjPM3Ki6(DJi
zO__g_$%G-MGep*_NSQ>QfBFkdibtsHigL0+SRUh7$M|Nz7KWET|0EB*;S$ws$}lV`
z3WIVHZ2(tuaPI_kBq=HFoHJ0A5C;4ZXira=Ub;d|iGFd);hk5o<%oz!)z>J>UAp5h
zko=tKYzte~R7C;Fb8>J1X27``tu52d8l_8I>1bCc%+#30{vAv`Ve7(mPEU?0iV3Nz
zQQ-tXL=u5^b%aG@%MukMebX}D+@zXp2}&iFC_Bd429&0-H6$2@n(-ElCN=G%MyD0S
zu%w)9;+#YIhNe}thb`0XZJN_L+B)LU5M!d6ObG1)-!-`DCRN@XtR0i$imq*5;|2fw
z>uY@}1zppS`aoS*k_D!S_Z`tohTb<VLBnV=A|<g&Rz?t87|jq9x@%dt-J~^aFi|#=
z(03T)(AE+BKo|ymh!mwIc`((K=vSnn6QNII&-9P{&=EtAG6t<Iy0D~}Ft#Lw4vWUO
zE1IsO8colb%_7VvEcS>De976YLKA#MTQG&o5LOMQ5P$O+g`bfkB#kj(48$lfc9aA^
z=jsgA^$bx0lD3A>b|@ueLZz}Xn;$UBV2YA3^kO&#S?=Vu)gfzXkF_A>{TzuiZiUJk
zVoIY_&fqbmB*7|WmsI1jl5#}m#m1o4@gzG;rUYos&6oG7iyCbVM@J`&Mk6j<x=1}9
z<A$E-qs(>I&z~a3TKYj$maUTsNzw(z%K7GuRG7c3^&CwSh!CYED%Q#5W~fBXl<DeP
z>5RnIGWW@7t{CH4ZTDpZh%uTsz3r`R>}=4l8ctq$nWPMZHS}u4=4^}c&WhEtqZ(C&
zXz5ocsCr5Y5oZ*rh)I!T3bYy^Kvh}>A83aTH=Z(dEyZYqAvW|OGMdduX~4&p_VgAm
zL{x%hv!qOkva}?>!qcF238n^X8HSb2RH#VZF{V3Vwy}XKC#2O8vvEc5JNm`IpftXp
z)A^R;<0ab{AExL!v`&n-wo$s^<nT^ra+dU6Lu?P2Y+a$a_$UyuVNNxgf+{JB7EOt<
z0>xBbhFn$!#?uj`Ksgyvc=1#5LF3yFPf5LfiLPz1Wuh!A3TL1-Y|du?Ht-1ltyn1~
zs?mtr<<`c~<NZKH33O9COrh|7hY#7`BY44luVqVePkp`dh=`zAjfTD#rP}q>$T)o4
zqpg$8P=WUigaNH&kT1xbC{;!D*<(bipp>Q11=k6LUfR35g5<^YOe?&X%}WsoaS)iY
zDY4oTLys*Az~To_Oagfi5t46(m0yc&Q40N;&X>|DAB(6EujdD134M-HwS@!L%DXZ!
zHm(Ll&&#+X4}25WxPRj_`AVNvN-0-g(ngleSc;!@&}AJ&GEiZyKb4eh)S?oO_-I(s
z7K}s8d<%KLm=ankeBX0=GUw9eJ(|^$wjX%pp$FOB*&qxp1W^TW&Y?8omYC-ceb=#U
zS~j<)q?pcR$JPM=rO+m~{m%5qQr7#E8c1}z^5?yV6rWjmNSjbvW)k?7HUg^nN-Zp<
zHREcGMH5FOin@Yoo7*>@<IbHsT)g}smmj>!^6-G;lNBa;;t<%pa0!(LhW12Qqh3f!
z^|)Z@T9&8Bpc7GP(x50uC2i|zx*n@!Ko^IB)Spn*C8n%dty;Q%Afy$iOT}_N$C<?L
z_7>yKtF&Fj)iYvy#EYN5L9=X_UEHBk9p%iROvJ8weAm*c5v|+6SVd89GWEA8IU!Ms
z^I{yBpWI`$xJ}_IPL5A$=EwAH$N1njiUFGxeGuoejSCNu`j*NhTs4+{<!XUeo)iM(
z*_67T(k<t7eL@F?j{(M8I2{RSN)u>@fv7hax`lW@{98Ui|3+Iz8%*vXD5uf3zz?4J
z{DjeHMp=o4bI5ZCVni1O))j=%<;%gy;Mf|7p3wJMJP=Y;=#0qnL&$?)VMwo8>&6Ht
zAcVkLZlrB{NC9IFWnB^xSrCP%TOQM{8f;Z!wL)u8GJ=$o;W3~q0JL=&dJH++&*Ze=
z`!h-Mm|*By$$XPQ;jIyg6YmF%?VVLqTur!zlY9vT3l4(^_rTyz5_DjIWPsoh7+ixp
z1Q>j9_u#?ZEx<4^xJz)?;I4<N^Bd0H**9->?VGNyUA?PUKkr%zkFCo3KHqW%e^nnQ
zOuy2l{d9av-DCe1t4O7Y7oiGK23N~h?CXbpjokDR)m(?ylJIs8vv~{T>f-LmaJzBJ
zn}2aLLVGX&#{f6y6P}Y5yW!BuF6%)3`onXhi|bLV|2QMA<oyd7!_ZIK?hcVSO7m+{
zQS~HH%5qAx31&uWYO~SB_^7$`iFc<~0m$S!ba=FGgx0<*?N4|O8-(D$1f6W8W+}<+
zRWs}TrjF*4quc(Q{u0t1^`5O^8IthCfB$e?u9zC#z#U=*l&m@E%iYue;qn$+VG1ID
z<1}ZRxjUt;5Z^UKRlU;{?`$GitavLs<}rd=6=#~jQP>o2x}XD+Gm(}<NZy3suqw^-
zWDbR}RtiZlvQ)inbF5^Tl7n?lyp}n|a<Z9i@yd9{A$pmK0vL_RN@b&%ei!s$UjZTt
zf6qb-q4BIIz%f3~FlrO5QbM4V6=y%u;H69GqJP?m#PH|^14sM5P`hK=(6i0gbcn6H
zyps+ryRhksZ}YclgI9<e$xdaCWNY}u+WnR@E7+Hh$B57AfwmZ~X6x{d;OXH4r;@1M
z3P50_`z>DB4{U7>-Uz}GM=G??-u^7Jf;($M?Y{*jx7&eeflfwh^|r#<aUrT~5l7M)
z4W;b2^|aVO?bYgbK<<(uEKmjac2<`+-rHlt1Ki?{>~t{>B3+gNN1{ex^El@gctV7D
zd0D19H9}jSpz(fGQ3oL}Q*COXqw(w^t~ZW!Hx&+d)PI)qF!3$0C9<sp?UZ@Yvk{s}
zV{+oV*q;h|dekt4KIO@Xg}(f!aA@`Ksa~-p-F<uL6dcPi;sgmhXZ;nliv7C0Mz2&s
z%TtI@I8G!ul)Ws(@SjLZKn(unld4L`bK`OAMOoCz+=v7UiA-+j_y@$)z;hBJ6jFP>
z3C7!o(7+To#JR+-UbVDLEPRXoZZi3I3`kH1<4XXQj++i*yK*|*Ap36U2ZjsdpE0jo
zmePA|Mp}EOlPrUhDld0xd3f#6PgZuOcBs1>mJN3P#Ng4tnghG9BLb|&_&wIJVkU5b
z&I-MD0K4UFZcF1>NQ<Dp5kKI=Av{H$SbqxpKbA#&)x1)y(!IB6#mzZZGC!)02m)2V
zm<PGwl{(A-mzMfGH2gI5n?H4kZEa0(asN$r*u{zBDZ_+c!CkqGg`fycnbViyg`<g6
z7K%26IH_MEvV{xSDE|ukLs%O8>Oxbd;mlCk+%UdDp+-*%6AqpE7Px}qfk<f_#PhAV
zp))B}5{t%4Zgxr*Pc3`%i((cbF|iyDIXII)#SxxTDdg&P?wI|bMtoB-Zz_t(ql${%
z1hRIFfC;=?wAvnWrSr^oo7ctqebM72j0G$;D|*uVklA(Qs0;gFmtW)w@v;Kjcn$^<
zs09Rrp{o`2bc_Bce*PbFnGg(`*Xq!ST2*XWz!2-`AgT1PVmY{=>5JJMl6SAz=5b%p
z8mjesp-0iJMo5`B#qxa9=Ijs4B#kL9Kiw+b{rm1H|L>3KFk>aU5doOor-qLvWkq;|
z#a6Gq^LfI4M>RseKxscg-jxxGND~M&^h|$i>3E36(Tq2rkzlh5{*sX3E(Kp3O<E&+
zEt>z9Rznms9{=Nuk7_yAhi$bva9##t5bc=TNUy{t*=6|wPo|Q2bEf12b9?Y}6mykh
zucQsbPi-mjVr~}%p2%lShpR)Z`|%3Cdq2rzW%$?YXeDGkikNWx@pVjx7J4Foh0LbF
z%wJrF^8AMy$y+j*pzm$UKQ7x814A@|gdEk(dYhf7L%*_+f~_ArDbsH#L`k&<RmPMC
z%=cdb854sAF@Uk3qXdW*$hPi!5A@xVKZZJHPE7pwF4C4a)rO~tSZc;d?H8M=8zw>I
zoEmsHxZDIFLJ%{=X)bE+q(m<5u94!O8z4mx*1L4}VQ(yX!;|DLxU!Bpb~jrlLRN^k
zM=PMb;cdqDRs^Yef~nP3GIpjVl+=bOaOyn#jk-Ay35hfVWNZ?|uio~AyXB9`1|3=o
z&|<T{6*u8uJwB(lFbCM#_6=DmR`O%2Xes?zSNtxAJurX%4CO=|MTHp3h_}eAK}Bc)
zx63+{qem6XbHw-%t-ZF7%+5LJPMD3&Bmp@-zkXZAaFL1l@7QnGt9OoG2x+YL7l}c!
z`3ByYD;g?ovAizje9M|`-L*C<8*ZT$PZA6qA3g?NU4fFjVw*GZ=%$SQIWcQrE%SZ(
z#=`KHEEn|Gu(n$>VmfHb2qgNtbSlMHJ==*j4NN4Rf+}ulW_hzmPN=^6M>saouo=?T
zCJ!`<$Q=>+h9pZTuE5QwlHOg|t$v{wT^(jG&odL6Ej9njWFaNE&q_~wpZBB(uX;va
z-b|gqVv$L~2s6c~yq)iFI{W*==n(e_#!&Ur=~vy6dYM$31CMRcNEOc|Z69{Y3n&de
zOn92<+klKjB>XW9gN9>b4sCsor01UYcDtg6w?+OdiPr`_&GSFcH%h+7%275)R|5mc
z_zHt60s9;!{g;8;B<rw7XgEK<-;b-v4VkC{ItU3M&7*21hm+0%lJE!$6;*&=i73)`
z+OydxQq=>>>S-^;9*TO4Tnnv_tUuKvvdz>ejE*(U=2{*}WP9E&XB?lNh72SL{AQD*
zBr0x~_jL$Z&-GSrHL3Bh)XhOt(oz>vt<ue?_TBzDW<F@`m?m_d!v0dDLQl#foX)U}
z=~f;Rd4i-a@x;-qevPY^RQ1<TEqF$e+WR}WP;Nzb=P-V^k<Q7{XeO8SM=t)u=voUJ
z?-Fbb?VnL3j$?lpelUsNB!HMZztipQ^OOl~+PdkqE}5XkyNtz_Vf?S;(rK0ug24zp
zp*GjoQ(^sXt;K&l|0b{$++N=v?>{R;K^g>K!p1O{%}QQq#KtBz$saRNBFR%X!=eeM
zet)*0H~(GtK|QR|0sJ-cb@?3Uk5Bh@Au)MY5KBN%rUa(xG9IFC6k|REK!#?2R2wKt
zGg~OHl~~OicejEOYY))DYwI(S4`1{ngPMP`D$dE~SR^ADupIML$PUiqIeI^T%tT8f
z$7Le~C4wcRH@PQ9OZzI8$UBf!Y&?Dl!`If)WvaUw#Sd0SYh5zC$+=6OlzLA~2d^;0
zi{<H3VzysZ><oYE^Z1?x$t*MfR7QJN5dBpfBFh+0OxyMT>p>|oBSpirS!(bZ@uqpe
zp@u{s&ea<djiVEG$tNzBL7MIH1axUZ_Sauqe;MiOnCet|oxGxX(9vl<iV_PcO1$5Y
zyfaIuUS3tQ^NFtp*gyEm$twfhs&dQr*>t`Lb!xM&)*!F}(VBGXa&Y{t&|EeJ;`C|^
z?LK+GY>MG6NdtkOFo3Y|Udkk=RPiWA*p&EMPVVA7EIa=Pv1N0p&!#|c!xbx5;8&96
z%i4j+CFayP)pt=o<jXF<0gd-7>S|vdt}Z18lzHbdC+&YZCu&)|EkT3l>y&E_s=U$5
z{$AU}+MUC4I`LO%7iB5+)~{R(m=Y=9{6eh2J#cp`-jc=QE<^<*(beJ`PLCVm5n;mJ
zQ=O}+XspSM`{U)0pk#;NrP;xIp-`Q8a99opS&;SHAnXnp&8PO`jO9+1HLGRvpx505
z#-Hk0FbU?As)P3qQyt*D={3v-rGzn4#&H~I3@Oz!GavPxWYdN4$?BfkW{xX3my`h9
zr4o-;k%tR~jbzyjzu7GR^A2gegu6Hl*2wuPSc!sDR;*O$q~1E<0&a(5k+Rn9t6Qh(
ze0Oa14%Q9ZZkSBWMmxKLNeDo)5~ce3Z!aZQGP#lS>ku1QbMq*6-xE6^gf<S@lLv@W
zj3J<FipgkaG~B}wV7wg)TIp>MO-3Q*O^o2@0dOu(*0K5sKEfcsY!yv|-+D{|laJf>
z=-Ak8vvF~}^~DUvG2v=P!n-tcjP1~n@y6|YK=EFWAd25147k$&aYta*R2FHZfS3={
z<*=Al82h*wO+dia8<vu5uckhLmzdXHSYA>UzH}J(@P-w|&$wQejcpa?2;faxKsGk9
zvBs`(w|7hHf&N<;f)ne?)Kj-Ow)n6z5NXE7$nc?1C4YPdtv0FRKU12G2aj>}_mOz}
z<^y_17JkxD9EY)rJ??l*-0VOP+fSKPGP4{-TGsXq9+(UEX&f6Q)*eyml6N#u5w5-u
zqUnpi3CXX#nG-ytB=3@EOy8Uh`>v+kR4j}Tk4R{p9L_8m30Sya@0FxyEUkJ>)QI^G
zZ=R_D+t)HGyaVDfCBih#RFThVFT*hVr0#4YR#GIqNwbChkht4{T{c4aF`bQLn@#^|
zx&t<)+TRnEvhaktxaU9aW_#O*Af|rU@$PC~ubVkCR8`b1o3$mVq6r}*xjN*wp4Rm`
zWpxqBE~82!9XU}7ZwOk?Wvyf9h%{KeMG2RmuCNy<DUy1M7))@AGAR`#OLJrvMpX!l
zXY!<gkC(pk^RM2@BIl7{(WIa<rJZAv(h>_VIoA_!{LEL~&%w{4m5t%{a+=g9_iMr>
z$7IZ|hrqN!ATm}@Y3YOT)3s+__GtaMBS8cpN{Q_cSu5v;&->J_o++G(xHr?Cr>J0E
zpIiJP>tO_lU{ZjSh<$<!vf{=Q445k)(MnH;@Z)%Ia&I1aLK;tTWm)-Z=)Kl4l15Jc
z!FUqC1*!dJ>mIe)t*}j2w%jZ5AN$GG%OC+taB^_n=hO^8OElcECsI(DRH7H4ckbry
zizHJ$o(|7802wbd)+7O3X%KsQm@=|0H!Muqnkg!4rAg16@qa<GpFr->vGZe^#LGXe
zxPf>reWSRO`70%+gEjy#3TCUcQ~w3C-S#RUMW9Vn;-A;5AhL=DFaxnxwz&(C)c0)%
z%te$Z=(Pb?IP>r-r&luKXGP8iuwOM^v{bKG^)-2LVHbs@7|EW}d~0R~%EXQ}y;rw2
zO-gQe-o^nTjt-Wr(y51b?Wo756qkI;kjL<beNiw|g?(FC-iNbXT#2m*8XxHyyJr(`
zXK`j`tj87HN1Q8~1;d*5HI&(ikg(R-${|&x%28!y5_ZyvmV|p3uB6*wwa8{e`vccb
zaArCJeMkhMGXyAZ?(}3fgX=Qv3y553iv|(JZOD`+^-@6dEO<jtHc2u2@H`&ZqMkyQ
zuZQhMRSK-jryFet?u|QhJ05g~m)sIO?Y!P}rTWxmmy>(VhD`PC0)f)E&#H(ESl<`R
zQf?<Pxj4sw6QqvS{bsQ)Kfc9!xMH?cV8L^rcAyh)5m*<#QU67$Fjh1CWgL35bBwe;
zD=fZf9&K5naAOp7Pzg|>!C}ST;Qj)6tr0L_JuNsX=Sub=t6JM9Yr6BK?QYUp27sxO
zrN?W76W5dO;^JRrRd4zy;Uf?SNK(uO(F$R7JUhT}uVXUO;u2-L(l>M1PTp5pVh-{H
zcyt}SHKW9e3zd@FC%1+gYKC1!pPhT-USgWD-%B13+HM}<LpAjcK(w;z6Tu%<DTlQo
zGh-JirV_m~YlrDZNDY1K=`;^Y$+S@5IGAZgEdRAVLK$?hgb^Fvu+}$kCvaU5h^au<
z^M@#`qH;P)LYAOh{wuO_%bWr66=j=}j$)J-F4UX_871j~m+}4M3FZ#~&Fl&wwQ}-1
zzVB~&1`sp+)52myLVrnjd3skMoeIw7vqQK99~6Pmz0~5&*tM`x9LSpT%k>^g#@+|z
zzwu&Q6z`*$(U}(o?s;fAwikQ`OXB4f_0d$13+V2uNmE`mCd-bnYsF_YG&BXrqkwTC
z5J-mc-jTDpTYLIh$Kxqi5qN8nOY-%9f-_cK(dH;O3VXtU3_xkoaeF6V#MFJbuse(v
z8(T_6Rn*qDyipfNtWBg#gHx~@KZ)_j&3(|62M!o9H}`Q<tiJie^6e+^`+;ODV2FBq
z1IBiKJOGN|L9cpL*{ttB2yd&z6!SQzW$q<JGng%x_?{|QJEgZ}#@oDaH5^G&+RVSL
z!X=zKGh355Wu`J`>v`CA!&Z0}biqSFX0Af>U2F6fY=_DC&bmGnz%H;&k*Fv&W0PVt
z4URghUbJ?<-;r8+yq#}7xmaG?nnQ>E!x>i86$<7AY6pEF)al*xKp9Iw?LERUVrXsi
zULcI?=W*1<{j!<gVJyU73^@i}kRMs4qGDi4x6&ab6Us$QYhoZiq3tTuA=5U)YIBP{
zaEm5hIAb1#!=eLkseN2(ZcU;<#cyV;0)soW#S22Cj-JdmnFNvX%XG`}{x8{Lm0Gq%
z^JGAoQvvx`!$Ls@MsFieOCmbCVITD`CvC8hqc#HILY!exV9ZeLFZ~KnhvCNJ7_Ef?
z;`Va$RPuG>f)KI@sWKF_nit-#J?iR5WD=lE`9oc%D$%y7pcq@u^-u~0^HDN!G06ad
z340?q)9IzCe>3CMYeMjK9H2u>H&ZtX)=ILTaKjcfOy@;-gmufCX+y6zgD*p~Z&Oqm
zQh;;MJ7$%zyZvZFnWa{=O-{xz!wk@gNJDYryQU)9Kh3w%stgJCeX5~r!5RmRsy?WW
zsdkjgR)3mz`+7YDWyoSXBc@qdcW?WYnZJ~z<`NM=5|>SZ_L;I#)Ab(pY;9FjgI;?E
zmAE@@vyYBj-MjT4G7aF&jaGkmBsQIVa{gnUo{16RFmA6NaFtn{@--D&U&6V3o3XrQ
z*Z?E=;UlZ=zE`{y)v?Gp=4#UM0;st9lmA;-zeCJStqaEl3D!e8kEH7_wFxz^{j3MV
z+mgu2h0za@!fjVSO(pUvIQEB#m+!KcO<W+);_aq2T1gtkbXwog#UE>r&W$Di8EOt8
z3E{0>WaU2`=Qs0e$x5btguM=A!k+tSt*3X(D=0bN;ohUCHxIYM%r!x|Ty3BGV`#ju
zIvkHW$pqezo7vk8_4FhPVb6NJl0tsN0lvPyAfA}>ueU`puKSTo2_*jGt`u?)L~SH9
z(>xI4{Vpb(m;=^lR(U*zOzi$F;~4&yT*+wtEjis>>@~su3rg{gF*8Xz+-H-C>~b^k
zp_{F($n7wG3EFg0>*=x5;95wX&CURRw<*X<OqdB0I?Ew|@|HK&nF#hbu&F%bZ9eGD
zh&zqScy{2;B@;e}DIdt2p0T!{VhgX7nmO0Rr%(>_SWq#&`8)_qM!EmGL_5L`pMz;)
zOHtlYfssZh5%kz(%N)tni;O*-B_)N3+4T%Fk!QrCT1)?ndhGS$oFRgxkOG>HP5$|_
zpvBu+ZIX0r_=Y80M+2)Ma$K1N=EMJvwpw#Nk4QpRcQ+54rV6H@m2NRK+e4+;c}*s@
z$2vcC?M2@dXZxO!-7uSWz0Juei4ycJ=Y@b`ip7Gn9z7QIbgYuhyHJSEIM!K*FTODI
zXUbz`f(w(RX%=^bl-CVP0jAiFApX*MuHu%0?u=bSIR5&^xrW_;&-v(#LD}mWt{q)R
z_tSzXEy_;U?jqzRVt&Rw8)fI38GE{x0WHTyr?^E-HLv#@e2sHID|0nx;#`VoVzQO8
zhVs{{>6l=kK*TfGO9O|7rj#E|T~y-qW2zeeg~#n+3}=-J0b^Vb8%n=f9WFHuI87gq
zUHSZJrZ|qnyeHb%{$Pt%KkzPmeahg8&BPhSB%&Xsqg+f#UGZTyFO1-QsX*z2f1Ji)
zjrL>6&;|VmpVsUud^9ZQXQdc6+Bv~2_;F>=;+GcX1lTm*@K+{>`9hE-(2z$oCl67-
z&)xJTj2`)Mih6=iRpAk4lSSqab$9NpZ0qvuKHs2Kq4DmTy5To?nm!tzhBW9$1z4q*
zy)`BXwBYT@7}9LOb-&E*KqN-qSO=Z)+d9Ck=;9neJDhxS@pzh9lC;*gnuUo+zE>(Y
z<D#y~86K7zgvT31?=OU`>}#OYI=!O?U*`yQ#aqvl^lzq+aL_rAG*be@cpC`J%v-;Y
z(_LQ(pnq35JdnH|-$y$(XgdBPB!4*c@27Vb>8>A8TmHEzVEZQeo=98gNLR?V9d~tj
zT~Ij|&yV6hCn>MGE2mTa)Mu2}KSXQoq8gOIIB=VoE&`OMWDoz`!BqHhs&*q?OJ9#i
zQn*`Gu73r-=p5C@%0Qa#Jp1<*r(sJTf_j^BN{DexNG`!peq5z0tzgFP+jowEDV%9d
zk~2(R?UwMsR#;`J<HlNuG)ttB!BYQg`oNFXEJ-K40QU0I*)q7zdBeN5qG;^vF+oMl
z1o%A;bV-mxSzB>_I-O(7>n1|*55SAvU)P01zyyqZUgALSCTsFuC=o%uW_r>)r!efg
z*s+>GJo;&!y?#|sF8!#Rz?Z7>S{1<cS_GXDjC0n=#SS;UGlE$pC6I?{xh$sRX{4l!
zO}MJHapn3ESmONMNB8XfM74R%VuWO&0(;81t7qM@|AJ(K&$>dO^)|zIXp9}#?S4^`
zF0jutO0e}R59)LEZlVXHyCB!>m#}BwW{)Fy;ds^4xAV^1R77%ZcimC>h_li07HxVa
zeX(sI&tj)5=%48`S=R5F+U$QDVi*Ej)kbP4tNbX3KXShm2sIs#YY<Q1AJscOH<DVG
zQC1;IpCWED{x%~h_OP0z+CV5I(mhMRGkj9l+u#U3_)-47JJIsuKmE34hevBA1@Zl~
zmqseAegvMg)^pZel{NIA4@2^EQmIe2(jrYom__A)|8={7PE-faUn2ZI_Za(hA#2my
zZgG^erae2nH)$k+*vYYd?v#Fb|BYlBKuWVSCa?Pyk4Ct^v#F5N)u-|ZH!C)@zVDT<
zJU(#XkE}22sc*w=+=by0GG1cSRyMciS&|1yggi9y&Ng@5g*97pqkVo`t?0CVkA8#_
zCJg#BY}EZXOjnu%!PFgQn9ZLogy)d6PcW^?G!ev;=<c%ryy?a=$<c%EKa;S-$Jmx$
zz=&<27@E85A@aW&-Ut_MNMpWPXNSl6|1uJ@5fx}7tI|WU_UHT~H+E0h))XtZwXnpf
zXu}pBu)w~bojR->mKU}$<co?gW`LR$2(3@Kx<~V9*OOYfGaMCX=s({QxsdQxo3E#z
z?{A2STtt>!mjkN~it}CC7+Z{=t}DklXUU~>W*n46GjH7tZ?hets9Gvd?G3~(Jmz%M
ziz0~qeVkSVBU%cQ<%;FVnN;C!P1y;nKIoz#SOS<LP<r>Lr^a!~%jAPe?ZeLqs~~F&
zaQ2aiX$|HrzXi+S(O{F7O#Oc3CVCXWZE&rVqV@Jr@cO*%Xd=tEpS@yn_1b)RH|=O#
zPk$X8Wv3%BfOEDE1t)0PCN&ltRSVHxQ1yH~^bAsSyGN;dZ62#eExRm;W*XI)xI2Sb
z{X8HsMn<dR^U)#Rc`3>%lQ`2Jhf5s;wKgcm>fM<}ui)K{cU3y85pTBps<<sE!(C*P
zEEC%dw=!2sFeO{K+Vw349O_KCasK7y=M)}23jxDJLvNft!vurEpHDg>so!mSZfX=4
z?0Y7$lV>9esk(a#C0Fj_t}aai`Xzk^8&xBWtI%8%{}m*m(&`wx72#{a#3q;ZLdGG1
z9upocq0yK{`=Rsi*4egtdpF`|^$o`Pl-E}N!TMExmQ-EPf2|&z?9(FuqpPQi<N;S1
zjhuILdvJlYY&Hq&uCTk!ACq5^9aj(cf$_KPl}Y5cIsfH-$H@b18bjH!G2#LaN_z1^
zTp`kUzwIx{!_z6@b}aVq<K@}DxAp#|Giht5EBMGta>Q13{%eI@1=A1Dqqw;%{;hjy
z2&Q=V<<{ntGN4lVKtEblUp_RpXQ>^45uC*HjRUkV(m>&0znE`qZJ#>|deKjgq9O_v
zrKM%Vk5n7?Ij9E8kmEZR6t}8LM=<B=Vp{MO@M$T(zw<(dlXz(Pgh;^kwtt9jzd+d}
zTVFnExhx2606CbZG-V3RP3?s%FJ<Hmo5o3lD~)Jv>C$LF{#SD#)a2S<Tg2TRbQT_I
zOx4537Z>#QTxkIJX{*9*f548X&2`YBh4x)abGcS}UE|Gs>*;32njWk{pLsED-Kh7(
zs$y;1Osq*{`@FM+uSVD&up^M6$tx7qtx6_^i#_jWDbkyY<9e7?y@*{~LlzBcgZYpY
z@`f&|bCw$$H4vR4S_lhjFzlU`575J$XHme#%OD<W)QF_%*#UHKnvj{7jf*MTTQJ&R
z%~y@t)$O>hVxofX;YIhBKY_r4_c11|sqohg>4|haskOd$jn!R&e$?`F5O`tchxcV5
zbtye1`yT%^Lray%4zZ5UH!X~Eb16a39cH<g#Aks4fKb|)Fur2p?R>sTvgtHV^_HB>
z$-9$$<j)FDXa3wws?`V0dJiucioopZAKa%V+tyN-@d0ANM;7k~=5ds!bdKY2>oX!&
zKxku_ePsatEA8axd{eZ-njPH#HBGjFY3m~hL?Xh)JF7}r39M`r?rp9*?FGFAqx|sV
z;<4^Ni)Va=B%T}_{>t-fF3)zkgM3}@@mcDD@xDnD)e?%JP@-b@0>x7rPxmMH0)4h2
z_YMH7E2@R2zEA|b5FWqjRHO3=OJ2BQZc}zp0^=y-smKfCr#(SxELEVI<0c-!>K9T!
zw#sT5m19>!y9It9@vg%mYtGoZzF~&Oksm3K8+BhXPaTyJ60)+<hS`U*y@I-S{FGt4
z$7~!MU+N2H=mCG7{+%Tp^TW9&FlF5wQYGGzQ<oS%z^)vnV{KEk?buaKpP9w$enwm2
zTeQ07(CyNeFVWu1Ow%59lucJD)YQA^+77#joH`Ao4Oy(M;kMBJkmZFKtq^W{`l*YZ
zw&_?J44UYpv7H^A=P<ZTXYKwLS;D&7y1Jou`X1*?5NAz|DwuzB>%>u0Z#3a^pWn@W
zGDzJqFn5-fS<bO*lOH+lwx6s-S9J6iRkK*GRmeXr4Ms`ptXmB%iV_w~V2s#N1LK2?
zDboKT5*o@dkoaBKR|guC7iy`g9^B5f&)Vt0*cfoY2TL7Gz1yXUiFDv3rbpu@>H05w
z$6j7M2pGC44@Z<C`pS5p2_xi$t+rj2vy8Y+-@G69>%Y733l|wswIT-+Qb%a<0zCd*
zQ&tq`NDT4*9i}tm;(;L^?eVw4V+d0`3syKVdb75YeKTHGuPCva5wufe5h@jJl_0Iq
z(3GId5$f;C^TD+b1jDQRrcA(!B0}(FmP-HNs1uyDk=m(8B%`K3C5~(DV-J%-VlzBI
zFRjAVtZ6aMV2Fc!GRRUuRU}^!qL>Y|+^q;<*@(?QdUF~2&>BSylF0qzb5JrTd`jYO
z3o_}b)IoOYDl7jqNdnMBVIf7}gmE%J*4o^_5FriuLw#%2L6$E?(iX1*?KMdnnA0A-
zO+bxN%v4tY%y%U+KY)E`@%52oZFH7ReRX<Oc|dr33)di4b{PZXlmW~8B^`b5b$A2y
zQei(86XROt>Z1;Bkt4!ZK<v6-(EWV1@cw#C*t8E3ouaJcF%M~QP4?)$-$_OtgiHwf
zM!;r;1>@f6Nq)Jd_>8Ta=_@M`Y09&@wMDQHZ`)$@9A}pMW)7@XE|HC!GPf2In)LFb
zJOwhte|GaK3nzQIdWi!#th!CuyqUVTy&=<mJAu0s#5}l-B{EaJC=G8~xnGkqS~fgv
zAF_6dVk*;e9T654FDT+DNn?){r6*+>P%Mt~)HTp5&y)yAw?7HP^RUL$B&?duYkxL@
z!Mv`2sZq1Q#8byW>7C7oGrvN1+vnxd#Vrk?G1pP?wfDn$R#)V1rV4GvrkuhWc$x)*
zz~X5u*m!7ryRX(MhASF~aG6X`G8`fgEYqC<StsgJQ-j>?Og%rZz6Hq#f>{&sJV38!
zE;OgvhNl?O#rUTi5R>?EAwil!!DC&B(fWTv5nL$aA-Y6nRx}5JwNO%zE*JXE@^&Uh
z;UYvMjPj2MShP2ZZgFJziy$$pbZjk9^|ow7^Y>i2$DD8+cdk;4nw9YGCqNp`%hHk%
zGA-SgI>uWbRko7SZ0j4ZjVT%7g}F%&4d-j+HVs0&AC@jJXVCJpbK~t}9=%Ma%d1(5
z=Y*2+WQ2MNJIg)#$-_@fk6vL1Sz{ImfvF#g0XC`Cv&TB?$cZF?Lad2tEY+D7w}-7s
zO<J57iDKWBQR49JV$>qa%uKkWhBxMl57~nhY{!~Il_q9Qps%Ip(*aXdnyyxyC2sd=
zcv6%befgkgC(mI+_bQ#k*VnxxZb$Z=8P<*MOQ%*Hck}5IUU3&ayl*M=-S4s-FE1NH
z7ljM3h4yF5llaslfU02iyg6K{`44W-a#8uPWS*rC<osHpy>-I9n}Y%zD_fGxTgL_d
zi>89ONwjG&nJonl#Scw5E}SH}oe4g1+ShbpKau%*bw&R-k}2*#IZ+=O!n|_Q9ZOqL
zlpdouc^Z3n)Y^Y=z&|Y6LdRKQvE|sj6XNnIBvM4Cpf1~@U_n>R)Y-msBa*%cR?$;e
zOC%Q>8m*%XO;*Q{%=+5Oyd62`<&TjRk9{^V(|Gk)wDA$`s!fjT!dtNK{TGlj+fy{W
zW<h_iYM2}x1B%by2QrN-Yxle0s0Wl3mX?=m;shnaFjAT=gi^VT&C*`kr~*a3l^#D8
zzSMDtggMYGLHl<Q+L+INqznT4o6iPcjY2|P;%}htaZb%g%GA&~Q)Fz{NBc}lx@ir*
za8I?N^naW5#oAsqw?6c|U1cy6Na9pMUZk$EA))7Os^v5{OnT3ww4r*(3|-!QAnmid
zrx9ge_eUNy<WU`GVC%jC%QvQ@6rMzqnYF9U-qs!f(vk^luYY}zxz(EBq36eht(@H~
zu~9@}3~&q_6DWAJ5Zt?csKQi9gJ2k;zy(6ku?oJxH#G7{yZz#>c3PLS@M;<bmPEF?
zmPYNL9%^J3UiT*x0!<{zVqO*WZ;YL>ZMRmm&$r15Hj8YhUj~%Vw)xuwJUzplh9joj
zPm{TbqNhB%T_wHxqp~W5Q-s#Or!4wXN%&;d+X|g3ejD-m<LP@Q9u$8wA$bS0bm_YF
z^^HPZEkiZAQUv(ZX!MOFuc6aPSwHdFH!eIt3#BPaI_ZLW`fR1?&hcWER0Q5MKbC+G
z&!E|?C5I_TMVu@>*++E;KV9?4;XdPLTd7&Gu5Fm4@b<<7r{N9E@QhYiy7?m3RD={O
zo@n!0i)Vfa-HjeCR&S<C+ph7h9~j=^x;x`n`4_KJw;WBEzc&b{W|}*L55_>XE5HmF
zM_&(pKr5wv<_Y)*_W5hpSrcl`=M+!J7>Wz%dOt3kf!kJw7gdVdRx5n(;};PXSu(7F
zB}2sG;*VyNpILl;ot#?d+wQPvk{q{CH;jfmzcVwW`ma60nslW?C;z28w3|9Rf0DQ0
zYYK1yuHT!oQmwM`WKgZzNiYRw>e8XHE`tsKXo-H%4HbHB7T_ZQLs^R*ZO?_+a+8xE
z-hK<U8R%eKXDcO)Y9b1d9=1^u*D-XU?M*m`!j%G-G{#9^Rnv2&#M31>YGOTbf0-dx
zSM{^c(>Jw%sr4a2lIS38nwTFkWTf(?i%H{Qb<NDM#r<ac>P+<Ur85QDLv61Khz7sP
za0OYAK@NwH?xw3a-usrjwqNkVzz17NU;U4lPZbA2BJO$RmZa_EPc_irY%-u!dTQMk
z#C7;*QwLft+cdHa3Qh}sJYwleV_2vcN%?^?O@oY{3o{}&#fY^*$<O2<caC2<5tv@@
zx=bZAq}{SC;=S5`mbU!(Ze-+X@fzt`Y;j3L$H*(%Jiu=7u=N<xKgN6XfY=C#Ce5Vm
zXzRFKb1JkwO~n3jyB3}{uVC)7X?+@gav<bGlz$)@zj$`J$ko>tO*k~CH@Ju{c|zs>
zscq;HJ)g+*PYS->y>#acnr|<6{*Ts{!o_vJy&li&1*-+0mq)4dR6{C~TY)WX1Tzm?
zc25$nhYL6UB};2%+qh5bG!8XhH*AEPM~Qa6`|};g!0;Y_$@{A(KSO}T)zy+O1@cSR
z<MJ3*YomF>X_HyoA1q)H*DcfIB$iL-gb&Hx_96B4joraNnL$#qZDy`ZnX7BZ!#mQY
ziSU_g)Lo<Z2?Vp_%CqANlC)GVDAv-LQ(n?WeD^`vwwDY58LX_4NP;zJW>gy7Us<hO
zLz9dfh7a_97|yU0*bz?JGUJh$iPK`|rT?~Ov|@*J)>DLA_4ZdTiXS$wMpQxIFCE6o
zWWwo|EK|ft_zrHRY@gnt1)M+_NgZSHo7Xm8eQTW%NW*b%8+qy7*<ZjCcKSpXMI&k2
zqzi&Bs_c!CVyur4DPE?!v~U5sJ&*n0KW)ei`p|H6{8cw?13Wfwx~Ljnrrl?$xUX(+
zA63L>wLd;qaV?bBh8r`*Vc^n8sJ;A3B=hy1RLj(Ji4XYlzr)W8ub+oRn@Q3t@56Ff
z`ak{t$?{y$vtQ*rS4tRW$a8hTWBdP?|Gx<T7heM6>E-YIDq6c7ar_t04@gE?y8QEZ
G|Nj9=hs&1$

literal 0
HcmV?d00001

diff --git a/0.16/_modules/index.html b/0.16/_modules/index.html
new file mode 100644
index 00000000000..19e831020c9
--- /dev/null
+++ b/0.16/_modules/index.html
@@ -0,0 +1,800 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Overview: module code &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Overview: module code</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>All modules for which code is available</h1>
+<ul><li><a href="torchvision.html">torchvision</a></li>
+<ul><li><a href="torchvision/datasets/_optical_flow.html">torchvision.datasets._optical_flow</a></li>
+<li><a href="torchvision/datasets/_stereo_matching.html">torchvision.datasets._stereo_matching</a></li>
+<li><a href="torchvision/datasets/caltech.html">torchvision.datasets.caltech</a></li>
+<li><a href="torchvision/datasets/celeba.html">torchvision.datasets.celeba</a></li>
+<li><a href="torchvision/datasets/cifar.html">torchvision.datasets.cifar</a></li>
+<li><a href="torchvision/datasets/cityscapes.html">torchvision.datasets.cityscapes</a></li>
+<li><a href="torchvision/datasets/coco.html">torchvision.datasets.coco</a></li>
+<li><a href="torchvision/datasets/country211.html">torchvision.datasets.country211</a></li>
+<li><a href="torchvision/datasets/dtd.html">torchvision.datasets.dtd</a></li>
+<li><a href="torchvision/datasets/eurosat.html">torchvision.datasets.eurosat</a></li>
+<li><a href="torchvision/datasets/fakedata.html">torchvision.datasets.fakedata</a></li>
+<li><a href="torchvision/datasets/fer2013.html">torchvision.datasets.fer2013</a></li>
+<li><a href="torchvision/datasets/fgvc_aircraft.html">torchvision.datasets.fgvc_aircraft</a></li>
+<li><a href="torchvision/datasets/flickr.html">torchvision.datasets.flickr</a></li>
+<li><a href="torchvision/datasets/flowers102.html">torchvision.datasets.flowers102</a></li>
+<li><a href="torchvision/datasets/folder.html">torchvision.datasets.folder</a></li>
+<li><a href="torchvision/datasets/food101.html">torchvision.datasets.food101</a></li>
+<li><a href="torchvision/datasets/gtsrb.html">torchvision.datasets.gtsrb</a></li>
+<li><a href="torchvision/datasets/hmdb51.html">torchvision.datasets.hmdb51</a></li>
+<li><a href="torchvision/datasets/imagenet.html">torchvision.datasets.imagenet</a></li>
+<li><a href="torchvision/datasets/inaturalist.html">torchvision.datasets.inaturalist</a></li>
+<li><a href="torchvision/datasets/kinetics.html">torchvision.datasets.kinetics</a></li>
+<li><a href="torchvision/datasets/kitti.html">torchvision.datasets.kitti</a></li>
+<li><a href="torchvision/datasets/lfw.html">torchvision.datasets.lfw</a></li>
+<li><a href="torchvision/datasets/lsun.html">torchvision.datasets.lsun</a></li>
+<li><a href="torchvision/datasets/mnist.html">torchvision.datasets.mnist</a></li>
+<li><a href="torchvision/datasets/moving_mnist.html">torchvision.datasets.moving_mnist</a></li>
+<li><a href="torchvision/datasets/omniglot.html">torchvision.datasets.omniglot</a></li>
+<li><a href="torchvision/datasets/oxford_iiit_pet.html">torchvision.datasets.oxford_iiit_pet</a></li>
+<li><a href="torchvision/datasets/pcam.html">torchvision.datasets.pcam</a></li>
+<li><a href="torchvision/datasets/phototour.html">torchvision.datasets.phototour</a></li>
+<li><a href="torchvision/datasets/places365.html">torchvision.datasets.places365</a></li>
+<li><a href="torchvision/datasets/rendered_sst2.html">torchvision.datasets.rendered_sst2</a></li>
+<li><a href="torchvision/datasets/sbd.html">torchvision.datasets.sbd</a></li>
+<li><a href="torchvision/datasets/sbu.html">torchvision.datasets.sbu</a></li>
+<li><a href="torchvision/datasets/semeion.html">torchvision.datasets.semeion</a></li>
+<li><a href="torchvision/datasets/stanford_cars.html">torchvision.datasets.stanford_cars</a></li>
+<li><a href="torchvision/datasets/stl10.html">torchvision.datasets.stl10</a></li>
+<li><a href="torchvision/datasets/sun397.html">torchvision.datasets.sun397</a></li>
+<li><a href="torchvision/datasets/svhn.html">torchvision.datasets.svhn</a></li>
+<li><a href="torchvision/datasets/ucf101.html">torchvision.datasets.ucf101</a></li>
+<li><a href="torchvision/datasets/usps.html">torchvision.datasets.usps</a></li>
+<li><a href="torchvision/datasets/vision.html">torchvision.datasets.vision</a></li>
+<li><a href="torchvision/datasets/voc.html">torchvision.datasets.voc</a></li>
+<li><a href="torchvision/datasets/widerface.html">torchvision.datasets.widerface</a></li>
+<li><a href="torchvision/io/image.html">torchvision.io.image</a></li>
+<li><a href="torchvision/io/video.html">torchvision.io.video</a></li>
+<li><a href="torchvision/io/video_reader.html">torchvision.io.video_reader</a></li>
+<li><a href="torchvision/models/_api.html">torchvision.models._api</a></li>
+<li><a href="torchvision/models/alexnet.html">torchvision.models.alexnet</a></li>
+<li><a href="torchvision/models/convnext.html">torchvision.models.convnext</a></li>
+<li><a href="torchvision/models/densenet.html">torchvision.models.densenet</a></li>
+<li><a href="torchvision/models/detection/faster_rcnn.html">torchvision.models.detection.faster_rcnn</a></li>
+<li><a href="torchvision/models/detection/fcos.html">torchvision.models.detection.fcos</a></li>
+<li><a href="torchvision/models/detection/keypoint_rcnn.html">torchvision.models.detection.keypoint_rcnn</a></li>
+<li><a href="torchvision/models/detection/mask_rcnn.html">torchvision.models.detection.mask_rcnn</a></li>
+<li><a href="torchvision/models/detection/retinanet.html">torchvision.models.detection.retinanet</a></li>
+<li><a href="torchvision/models/detection/ssd.html">torchvision.models.detection.ssd</a></li>
+<li><a href="torchvision/models/detection/ssdlite.html">torchvision.models.detection.ssdlite</a></li>
+<li><a href="torchvision/models/efficientnet.html">torchvision.models.efficientnet</a></li>
+<li><a href="torchvision/models/feature_extraction.html">torchvision.models.feature_extraction</a></li>
+<li><a href="torchvision/models/googlenet.html">torchvision.models.googlenet</a></li>
+<li><a href="torchvision/models/inception.html">torchvision.models.inception</a></li>
+<li><a href="torchvision/models/maxvit.html">torchvision.models.maxvit</a></li>
+<li><a href="torchvision/models/mnasnet.html">torchvision.models.mnasnet</a></li>
+<li><a href="torchvision/models/mobilenetv2.html">torchvision.models.mobilenetv2</a></li>
+<li><a href="torchvision/models/mobilenetv3.html">torchvision.models.mobilenetv3</a></li>
+<li><a href="torchvision/models/optical_flow/raft.html">torchvision.models.optical_flow.raft</a></li>
+<li><a href="torchvision/models/quantization/googlenet.html">torchvision.models.quantization.googlenet</a></li>
+<li><a href="torchvision/models/quantization/inception.html">torchvision.models.quantization.inception</a></li>
+<li><a href="torchvision/models/quantization/mobilenetv2.html">torchvision.models.quantization.mobilenetv2</a></li>
+<li><a href="torchvision/models/quantization/mobilenetv3.html">torchvision.models.quantization.mobilenetv3</a></li>
+<li><a href="torchvision/models/quantization/resnet.html">torchvision.models.quantization.resnet</a></li>
+<li><a href="torchvision/models/quantization/shufflenetv2.html">torchvision.models.quantization.shufflenetv2</a></li>
+<li><a href="torchvision/models/regnet.html">torchvision.models.regnet</a></li>
+<li><a href="torchvision/models/resnet.html">torchvision.models.resnet</a></li>
+<li><a href="torchvision/models/segmentation/deeplabv3.html">torchvision.models.segmentation.deeplabv3</a></li>
+<li><a href="torchvision/models/segmentation/fcn.html">torchvision.models.segmentation.fcn</a></li>
+<li><a href="torchvision/models/segmentation/lraspp.html">torchvision.models.segmentation.lraspp</a></li>
+<li><a href="torchvision/models/shufflenetv2.html">torchvision.models.shufflenetv2</a></li>
+<li><a href="torchvision/models/squeezenet.html">torchvision.models.squeezenet</a></li>
+<li><a href="torchvision/models/swin_transformer.html">torchvision.models.swin_transformer</a></li>
+<li><a href="torchvision/models/vgg.html">torchvision.models.vgg</a></li>
+<li><a href="torchvision/models/video/mvit.html">torchvision.models.video.mvit</a></li>
+<li><a href="torchvision/models/video/resnet.html">torchvision.models.video.resnet</a></li>
+<li><a href="torchvision/models/video/s3d.html">torchvision.models.video.s3d</a></li>
+<li><a href="torchvision/models/video/swin_transformer.html">torchvision.models.video.swin_transformer</a></li>
+<li><a href="torchvision/models/vision_transformer.html">torchvision.models.vision_transformer</a></li>
+<li><a href="torchvision/ops/boxes.html">torchvision.ops.boxes</a></li>
+<li><a href="torchvision/ops/ciou_loss.html">torchvision.ops.ciou_loss</a></li>
+<li><a href="torchvision/ops/deform_conv.html">torchvision.ops.deform_conv</a></li>
+<li><a href="torchvision/ops/diou_loss.html">torchvision.ops.diou_loss</a></li>
+<li><a href="torchvision/ops/drop_block.html">torchvision.ops.drop_block</a></li>
+<li><a href="torchvision/ops/feature_pyramid_network.html">torchvision.ops.feature_pyramid_network</a></li>
+<li><a href="torchvision/ops/focal_loss.html">torchvision.ops.focal_loss</a></li>
+<li><a href="torchvision/ops/giou_loss.html">torchvision.ops.giou_loss</a></li>
+<li><a href="torchvision/ops/misc.html">torchvision.ops.misc</a></li>
+<li><a href="torchvision/ops/poolers.html">torchvision.ops.poolers</a></li>
+<li><a href="torchvision/ops/ps_roi_align.html">torchvision.ops.ps_roi_align</a></li>
+<li><a href="torchvision/ops/ps_roi_pool.html">torchvision.ops.ps_roi_pool</a></li>
+<li><a href="torchvision/ops/roi_align.html">torchvision.ops.roi_align</a></li>
+<li><a href="torchvision/ops/roi_pool.html">torchvision.ops.roi_pool</a></li>
+<li><a href="torchvision/ops/stochastic_depth.html">torchvision.ops.stochastic_depth</a></li>
+<li><a href="torchvision/transforms/autoaugment.html">torchvision.transforms.autoaugment</a></li>
+<li><a href="torchvision/transforms/functional.html">torchvision.transforms.functional</a></li>
+<li><a href="torchvision/transforms/transforms.html">torchvision.transforms.transforms</a></li>
+<li><a href="torchvision/transforms/v2/_augment.html">torchvision.transforms.v2._augment</a></li>
+<li><a href="torchvision/transforms/v2/_auto_augment.html">torchvision.transforms.v2._auto_augment</a></li>
+<li><a href="torchvision/transforms/v2/_color.html">torchvision.transforms.v2._color</a></li>
+<li><a href="torchvision/transforms/v2/_container.html">torchvision.transforms.v2._container</a></li>
+<li><a href="torchvision/transforms/v2/_deprecated.html">torchvision.transforms.v2._deprecated</a></li>
+<li><a href="torchvision/transforms/v2/_geometry.html">torchvision.transforms.v2._geometry</a></li>
+<li><a href="torchvision/transforms/v2/_meta.html">torchvision.transforms.v2._meta</a></li>
+<li><a href="torchvision/transforms/v2/_misc.html">torchvision.transforms.v2._misc</a></li>
+<li><a href="torchvision/transforms/v2/_temporal.html">torchvision.transforms.v2._temporal</a></li>
+<li><a href="torchvision/transforms/v2/_type_conversion.html">torchvision.transforms.v2._type_conversion</a></li>
+<li><a href="torchvision/transforms/v2/functional/_augment.html">torchvision.transforms.v2.functional._augment</a></li>
+<li><a href="torchvision/transforms/v2/functional/_color.html">torchvision.transforms.v2.functional._color</a></li>
+<li><a href="torchvision/transforms/v2/functional/_deprecated.html">torchvision.transforms.v2.functional._deprecated</a></li>
+<li><a href="torchvision/transforms/v2/functional/_geometry.html">torchvision.transforms.v2.functional._geometry</a></li>
+<li><a href="torchvision/transforms/v2/functional/_meta.html">torchvision.transforms.v2.functional._meta</a></li>
+<li><a href="torchvision/transforms/v2/functional/_misc.html">torchvision.transforms.v2.functional._misc</a></li>
+<li><a href="torchvision/transforms/v2/functional/_temporal.html">torchvision.transforms.v2.functional._temporal</a></li>
+<li><a href="torchvision/transforms/v2/functional/_type_conversion.html">torchvision.transforms.v2.functional._type_conversion</a></li>
+<li><a href="torchvision/transforms/v2/functional/_utils.html">torchvision.transforms.v2.functional._utils</a></li>
+<li><a href="torchvision/tv_tensors.html">torchvision.tv_tensors</a></li>
+<ul><li><a href="torchvision/tv_tensors/_bounding_boxes.html">torchvision.tv_tensors._bounding_boxes</a></li>
+<li><a href="torchvision/tv_tensors/_dataset_wrapper.html">torchvision.tv_tensors._dataset_wrapper</a></li>
+<li><a href="torchvision/tv_tensors/_image.html">torchvision.tv_tensors._image</a></li>
+<li><a href="torchvision/tv_tensors/_mask.html">torchvision.tv_tensors._mask</a></li>
+<li><a href="torchvision/tv_tensors/_torch_function_helpers.html">torchvision.tv_tensors._torch_function_helpers</a></li>
+<li><a href="torchvision/tv_tensors/_tv_tensor.html">torchvision.tv_tensors._tv_tensor</a></li>
+<li><a href="torchvision/tv_tensors/_video.html">torchvision.tv_tensors._video</a></li>
+</ul><li><a href="torchvision/utils.html">torchvision.utils</a></li>
+</ul></ul>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision.html b/0.16/_modules/torchvision.html
new file mode 100644
index 00000000000..062e1be30dd
--- /dev/null
+++ b/0.16/_modules/torchvision.html
@@ -0,0 +1,770 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="index.html">Module code</a> &gt;</li>
+        
+      <li>torchvision</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">modulefinder</span> <span class="kn">import</span> <span class="n">Module</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">_meta_registrations</span><span class="p">,</span> <span class="n">datasets</span><span class="p">,</span> <span class="n">io</span><span class="p">,</span> <span class="n">models</span><span class="p">,</span> <span class="n">ops</span><span class="p">,</span> <span class="n">transforms</span><span class="p">,</span> <span class="n">utils</span>
+
+<span class="kn">from</span> <span class="nn">.extension</span> <span class="kn">import</span> <span class="n">_HAS_OPS</span>
+
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">from</span> <span class="nn">.version</span> <span class="kn">import</span> <span class="n">__version__</span>  <span class="c1"># noqa: F401</span>
+<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+    <span class="k">pass</span>
+
+
+<span class="c1"># Check if torchvision is being imported within the root folder</span>
+<span class="k">if</span> <span class="ow">not</span> <span class="n">_HAS_OPS</span> <span class="ow">and</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">dirname</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">realpath</span><span class="p">(</span><span class="vm">__file__</span><span class="p">))</span> <span class="o">==</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
+    <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">realpath</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getcwd</span><span class="p">()),</span> <span class="s2">&quot;torchvision&quot;</span>
+<span class="p">):</span>
+    <span class="n">message</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="s2">&quot;You are importing torchvision within its own root folder (</span><span class="si">{}</span><span class="s2">). &quot;</span>
+        <span class="s2">&quot;This is not expected to work and may give errors. Please exit the &quot;</span>
+        <span class="s2">&quot;torchvision project source and relaunch your python interpreter.&quot;</span>
+    <span class="p">)</span>
+    <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="n">message</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getcwd</span><span class="p">()))</span>
+
+<span class="n">_image_backend</span> <span class="o">=</span> <span class="s2">&quot;PIL&quot;</span>
+
+<span class="n">_video_backend</span> <span class="o">=</span> <span class="s2">&quot;pyav&quot;</span>
+
+
+<div class="viewcode-block" id="set_image_backend"><a class="viewcode-back" href="../index.html#torchvision.set_image_backend">[docs]</a><span class="k">def</span> <span class="nf">set_image_backend</span><span class="p">(</span><span class="n">backend</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Specifies the package used to load images.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backend (string): Name of the image backend. one of {&#39;PIL&#39;, &#39;accimage&#39;}.</span>
+<span class="sd">            The :mod:`accimage` package uses the Intel IPP library. It is</span>
+<span class="sd">            generally faster than PIL, but does not support as many operations.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">global</span> <span class="n">_image_backend</span>
+    <span class="k">if</span> <span class="n">backend</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;PIL&quot;</span><span class="p">,</span> <span class="s2">&quot;accimage&quot;</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid backend &#39;</span><span class="si">{</span><span class="n">backend</span><span class="si">}</span><span class="s2">&#39;. Options are &#39;PIL&#39; and &#39;accimage&#39;&quot;</span><span class="p">)</span>
+    <span class="n">_image_backend</span> <span class="o">=</span> <span class="n">backend</span></div>
+
+
+<div class="viewcode-block" id="get_image_backend"><a class="viewcode-back" href="../index.html#torchvision.get_image_backend">[docs]</a><span class="k">def</span> <span class="nf">get_image_backend</span><span class="p">():</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Gets the name of the package used to load images</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">return</span> <span class="n">_image_backend</span></div>
+
+
+<div class="viewcode-block" id="set_video_backend"><a class="viewcode-back" href="../index.html#torchvision.set_video_backend">[docs]</a><span class="k">def</span> <span class="nf">set_video_backend</span><span class="p">(</span><span class="n">backend</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Specifies the package used to decode videos.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backend (string): Name of the video backend. one of {&#39;pyav&#39;, &#39;video_reader&#39;}.</span>
+<span class="sd">            The :mod:`pyav` package uses the 3rd party PyAv library. It is a Pythonic</span>
+<span class="sd">            binding for the FFmpeg libraries.</span>
+<span class="sd">            The :mod:`video_reader` package includes a native C++ implementation on</span>
+<span class="sd">            top of FFMPEG libraries, and a python API of TorchScript custom operator.</span>
+<span class="sd">            It generally decodes faster than :mod:`pyav`, but is perhaps less robust.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Building with FFMPEG is disabled by default in the latest `main`. If you want to use the &#39;video_reader&#39;</span>
+<span class="sd">        backend, please compile torchvision from source.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">global</span> <span class="n">_video_backend</span>
+    <span class="k">if</span> <span class="n">backend</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;pyav&quot;</span><span class="p">,</span> <span class="s2">&quot;video_reader&quot;</span><span class="p">,</span> <span class="s2">&quot;cuda&quot;</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid video backend &#39;</span><span class="si">%s</span><span class="s2">&#39;. Options are &#39;pyav&#39;, &#39;video_reader&#39; and &#39;cuda&#39;&quot;</span> <span class="o">%</span> <span class="n">backend</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;video_reader&quot;</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">io</span><span class="o">.</span><span class="n">_HAS_VIDEO_OPT</span><span class="p">:</span>
+        <span class="c1"># TODO: better messages</span>
+        <span class="n">message</span> <span class="o">=</span> <span class="s2">&quot;video_reader video backend is not available. Please compile torchvision from source and try again&quot;</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">message</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;cuda&quot;</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">io</span><span class="o">.</span><span class="n">_HAS_GPU_VIDEO_DECODER</span><span class="p">:</span>
+        <span class="c1"># TODO: better messages</span>
+        <span class="n">message</span> <span class="o">=</span> <span class="s2">&quot;cuda video backend is not available.&quot;</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">message</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">_video_backend</span> <span class="o">=</span> <span class="n">backend</span></div>
+
+
+<div class="viewcode-block" id="get_video_backend"><a class="viewcode-back" href="../index.html#torchvision.get_video_backend">[docs]</a><span class="k">def</span> <span class="nf">get_video_backend</span><span class="p">():</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Returns the currently active video backend used to decode videos.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        str: Name of the video backend. one of {&#39;pyav&#39;, &#39;video_reader&#39;}.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">return</span> <span class="n">_video_backend</span></div>
+
+
+<span class="k">def</span> <span class="nf">_is_tracing</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">_C</span><span class="o">.</span><span class="n">_get_tracing_state</span><span class="p">()</span>
+
+
+<span class="k">def</span> <span class="nf">disable_beta_transforms_warning</span><span class="p">():</span>
+    <span class="c1"># Noop, only exists to avoid breaking existing code.</span>
+    <span class="c1"># See https://github.com/pytorch/vision/issues/7896</span>
+    <span class="k">pass</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/_optical_flow.html b/0.16/_modules/torchvision/datasets/_optical_flow.html
new file mode 100644
index 00000000000..cea3d728c3d
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/_optical_flow.html
@@ -0,0 +1,1160 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets._optical_flow &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets._optical_flow</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets._optical_flow</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">itertools</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">abc</span> <span class="kn">import</span> <span class="n">ABC</span><span class="p">,</span> <span class="n">abstractmethod</span>
+<span class="kn">from</span> <span class="nn">glob</span> <span class="kn">import</span> <span class="n">glob</span>
+<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">..io.image</span> <span class="kn">import</span> <span class="n">_read_png_16</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_read_pfm</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<span class="n">T1</span> <span class="o">=</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]</span>
+<span class="n">T2</span> <span class="o">=</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">(</span>
+    <span class="s2">&quot;KittiFlow&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Sintel&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FlyingThings3D&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FlyingChairs&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;HD1K&quot;</span><span class="p">,</span>
+<span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">FlowDataset</span><span class="p">(</span><span class="n">ABC</span><span class="p">,</span> <span class="n">VisionDataset</span><span class="p">):</span>
+    <span class="c1"># Some datasets like Kitti have a built-in valid_flow_mask, indicating which flow values are valid</span>
+    <span class="c1"># For those we return (img1, img2, flow, valid_flow_mask), and for the rest we return (img1, img2, flow),</span>
+    <span class="c1"># and it&#39;s up to whatever consumes the dataset to decide what valid_flow_mask should be.</span>
+    <span class="n">_has_builtin_flow_mask</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+
+    <span class="k">def</span> <span class="nf">_read_img</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">img</span><span class="o">.</span><span class="n">mode</span> <span class="o">!=</span> <span class="s2">&quot;RGB&quot;</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span>
+
+    <span class="nd">@abstractmethod</span>
+    <span class="k">def</span> <span class="nf">_read_flow</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">):</span>
+        <span class="c1"># Return the flow or a tuple with the flow and the valid_flow_mask if _has_builtin_flow_mask is True</span>
+        <span class="k">pass</span>
+
+    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">T1</span><span class="p">,</span> <span class="n">T2</span><span class="p">]:</span>
+
+        <span class="n">img1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_read_img</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">img2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_read_img</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="mi">1</span><span class="p">])</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span><span class="p">:</span>  <span class="c1"># it will be empty for some dataset when split=&quot;test&quot;</span>
+            <span class="n">flow</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_read_flow</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_has_builtin_flow_mask</span><span class="p">:</span>
+                <span class="n">flow</span><span class="p">,</span> <span class="n">valid_flow_mask</span> <span class="o">=</span> <span class="n">flow</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">valid_flow_mask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">flow</span> <span class="o">=</span> <span class="n">valid_flow_mask</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">,</span> <span class="n">flow</span><span class="p">,</span> <span class="n">valid_flow_mask</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">,</span> <span class="n">flow</span><span class="p">,</span> <span class="n">valid_flow_mask</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_has_builtin_flow_mask</span> <span class="ow">or</span> <span class="n">valid_flow_mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># The `or valid_flow_mask is not None` part is here because the mask can be generated within a transform</span>
+            <span class="k">return</span> <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">,</span> <span class="n">flow</span><span class="p">,</span> <span class="n">valid_flow_mask</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">,</span> <span class="n">flow</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__rmul__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">v</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">ConcatDataset</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">ConcatDataset</span><span class="p">([</span><span class="bp">self</span><span class="p">]</span> <span class="o">*</span> <span class="n">v</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="Sintel"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Sintel.html#torchvision.datasets.Sintel">[docs]</a><span class="k">class</span> <span class="nc">Sintel</span><span class="p">(</span><span class="n">FlowDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Sintel &lt;http://sintel.is.tue.mpg.de/&gt;`_ Dataset for optical flow.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            Sintel</span>
+<span class="sd">                testing</span>
+<span class="sd">                    clean</span>
+<span class="sd">                        scene_1</span>
+<span class="sd">                        scene_2</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    final</span>
+<span class="sd">                        scene_1</span>
+<span class="sd">                        scene_2</span>
+<span class="sd">                        ...</span>
+<span class="sd">                training</span>
+<span class="sd">                    clean</span>
+<span class="sd">                        scene_1</span>
+<span class="sd">                        scene_2</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    final</span>
+<span class="sd">                        scene_1</span>
+<span class="sd">                        scene_2</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    flow</span>
+<span class="sd">                        scene_1</span>
+<span class="sd">                        scene_2</span>
+<span class="sd">                        ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the Sintel Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, either &quot;train&quot; (default) or &quot;test&quot;</span>
+<span class="sd">        pass_name (string, optional): The pass to use, either &quot;clean&quot; (default), &quot;final&quot;, or &quot;both&quot;. See link above for</span>
+<span class="sd">            details on the different passes.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in</span>
+<span class="sd">            ``img1, img2, flow, valid_flow_mask`` and returns a transformed version.</span>
+<span class="sd">            ``valid_flow_mask`` is expected for consistency with other datasets which</span>
+<span class="sd">            return a built-in valid mask, such as :class:`~torchvision.datasets.KittiFlow`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">pass_name</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;clean&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">pass_name</span><span class="p">,</span> <span class="s2">&quot;pass_name&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;clean&quot;</span><span class="p">,</span> <span class="s2">&quot;final&quot;</span><span class="p">,</span> <span class="s2">&quot;both&quot;</span><span class="p">))</span>
+        <span class="n">passes</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;clean&quot;</span><span class="p">,</span> <span class="s2">&quot;final&quot;</span><span class="p">]</span> <span class="k">if</span> <span class="n">pass_name</span> <span class="o">==</span> <span class="s2">&quot;both&quot;</span> <span class="k">else</span> <span class="p">[</span><span class="n">pass_name</span><span class="p">]</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;Sintel&quot;</span>
+        <span class="n">flow_root</span> <span class="o">=</span> <span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;training&quot;</span> <span class="o">/</span> <span class="s2">&quot;flow&quot;</span>
+
+        <span class="k">for</span> <span class="n">pass_name</span> <span class="ow">in</span> <span class="n">passes</span><span class="p">:</span>
+            <span class="n">split_dir</span> <span class="o">=</span> <span class="s2">&quot;training&quot;</span> <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span> <span class="k">else</span> <span class="n">split</span>
+            <span class="n">image_root</span> <span class="o">=</span> <span class="n">root</span> <span class="o">/</span> <span class="n">split_dir</span> <span class="o">/</span> <span class="n">pass_name</span>
+            <span class="k">for</span> <span class="n">scene</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">image_root</span><span class="p">):</span>
+                <span class="n">image_list</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">image_root</span> <span class="o">/</span> <span class="n">scene</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)))</span>
+                <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">image_list</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span> <span class="o">+=</span> <span class="p">[[</span><span class="n">image_list</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">image_list</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]]]</span>
+
+                <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span> <span class="o">+=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">flow_root</span> <span class="o">/</span> <span class="n">scene</span> <span class="o">/</span> <span class="s2">&quot;*.flo&quot;</span><span class="p">)))</span>
+
+<div class="viewcode-block" id="Sintel.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Sintel.html#torchvision.datasets.Sintel.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">T1</span><span class="p">,</span> <span class="n">T2</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3-tuple with ``(img1, img2, flow)``.</span>
+<span class="sd">            The flow is a numpy array of shape (2, H, W) and the images are PIL images.</span>
+<span class="sd">            ``flow`` is None if ``split=&quot;test&quot;``.</span>
+<span class="sd">            If a valid flow mask is generated within the ``transforms`` parameter,</span>
+<span class="sd">            a 4-tuple with ``(img1, img2, flow, valid_flow_mask)`` is returned.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="nf">_read_flow</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_read_flo</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="KittiFlow"><a class="viewcode-back" href="../../../generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow">[docs]</a><span class="k">class</span> <span class="nc">KittiFlow</span><span class="p">(</span><span class="n">FlowDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`KITTI &lt;http://www.cvlibs.net/datasets/kitti/eval_scene_flow.php?benchmark=flow&gt;`__ dataset for optical flow (2015).</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            KittiFlow</span>
+<span class="sd">                testing</span>
+<span class="sd">                    image_2</span>
+<span class="sd">                training</span>
+<span class="sd">                    image_2</span>
+<span class="sd">                    flow_occ</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the KittiFlow Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, either &quot;train&quot; (default) or &quot;test&quot;</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in</span>
+<span class="sd">            ``img1, img2, flow, valid_flow_mask`` and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_has_builtin_flow_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;KittiFlow&quot;</span> <span class="o">/</span> <span class="p">(</span><span class="n">split</span> <span class="o">+</span> <span class="s2">&quot;ing&quot;</span><span class="p">)</span>
+        <span class="n">images1</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;image_2&quot;</span> <span class="o">/</span> <span class="s2">&quot;*_10.png&quot;</span><span class="p">)))</span>
+        <span class="n">images2</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;image_2&quot;</span> <span class="o">/</span> <span class="s2">&quot;*_11.png&quot;</span><span class="p">)))</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">images1</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">images2</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span>
+                <span class="s2">&quot;Could not find the Kitti flow images. Please make sure the directory structure is correct.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">images1</span><span class="p">,</span> <span class="n">images2</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span> <span class="o">+=</span> <span class="p">[[</span><span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">]]</span>
+
+        <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;flow_occ&quot;</span> <span class="o">/</span> <span class="s2">&quot;*_10.png&quot;</span><span class="p">)))</span>
+
+<div class="viewcode-block" id="KittiFlow.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">T1</span><span class="p">,</span> <span class="n">T2</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img1, img2, flow, valid_flow_mask)``</span>
+<span class="sd">            where ``valid_flow_mask`` is a numpy boolean mask of shape (H, W)</span>
+<span class="sd">            indicating which flow values are valid. The flow is a numpy array of</span>
+<span class="sd">            shape (2, H, W) and the images are PIL images. ``flow`` and ``valid_flow_mask`` are None if</span>
+<span class="sd">            ``split=&quot;test&quot;``.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="nf">_read_flow</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="n">_read_16bits_png_with_flow_and_valid_mask</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="FlyingChairs"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs">[docs]</a><span class="k">class</span> <span class="nc">FlyingChairs</span><span class="p">(</span><span class="n">FlowDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`FlyingChairs &lt;https://lmb.informatik.uni-freiburg.de/resources/datasets/FlyingChairs.en.html#flyingchairs&gt;`_ Dataset for optical flow.</span>
+
+<span class="sd">    You will also need to download the FlyingChairs_train_val.txt file from the dataset page.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            FlyingChairs</span>
+<span class="sd">                data</span>
+<span class="sd">                    00001_flow.flo</span>
+<span class="sd">                    00001_img1.ppm</span>
+<span class="sd">                    00001_img2.ppm</span>
+<span class="sd">                    ...</span>
+<span class="sd">                FlyingChairs_train_val.txt</span>
+
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the FlyingChairs Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, either &quot;train&quot; (default) or &quot;val&quot;</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in</span>
+<span class="sd">            ``img1, img2, flow, valid_flow_mask`` and returns a transformed version.</span>
+<span class="sd">            ``valid_flow_mask`` is expected for consistency with other datasets which</span>
+<span class="sd">            return a built-in valid mask, such as :class:`~torchvision.datasets.KittiFlow`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;FlyingChairs&quot;</span>
+        <span class="n">images</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;data&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.ppm&quot;</span><span class="p">)))</span>
+        <span class="n">flows</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;data&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.flo&quot;</span><span class="p">)))</span>
+
+        <span class="n">split_file_name</span> <span class="o">=</span> <span class="s2">&quot;FlyingChairs_train_val.txt&quot;</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split_file_name</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span>
+                <span class="s2">&quot;The FlyingChairs_train_val.txt file was not found - please download it from the dataset page (see docstring).&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">split_list</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">loadtxt</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split_file_name</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">int32</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">flows</span><span class="p">)):</span>
+            <span class="n">split_id</span> <span class="o">=</span> <span class="n">split_list</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+            <span class="k">if</span> <span class="p">(</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span> <span class="ow">and</span> <span class="n">split_id</span> <span class="o">==</span> <span class="mi">1</span><span class="p">)</span> <span class="ow">or</span> <span class="p">(</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;val&quot;</span> <span class="ow">and</span> <span class="n">split_id</span> <span class="o">==</span> <span class="mi">2</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span> <span class="o">+=</span> <span class="p">[</span><span class="n">flows</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span> <span class="o">+=</span> <span class="p">[[</span><span class="n">images</span><span class="p">[</span><span class="mi">2</span> <span class="o">*</span> <span class="n">i</span><span class="p">],</span> <span class="n">images</span><span class="p">[</span><span class="mi">2</span> <span class="o">*</span> <span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]]]</span>
+
+<div class="viewcode-block" id="FlyingChairs.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">T1</span><span class="p">,</span> <span class="n">T2</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3-tuple with ``(img1, img2, flow)``.</span>
+<span class="sd">            The flow is a numpy array of shape (2, H, W) and the images are PIL images.</span>
+<span class="sd">            ``flow`` is None if ``split=&quot;val&quot;``.</span>
+<span class="sd">            If a valid flow mask is generated within the ``transforms`` parameter,</span>
+<span class="sd">            a 4-tuple with ``(img1, img2, flow, valid_flow_mask)`` is returned.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="nf">_read_flow</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_read_flo</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="FlyingThings3D"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D">[docs]</a><span class="k">class</span> <span class="nc">FlyingThings3D</span><span class="p">(</span><span class="n">FlowDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`FlyingThings3D &lt;https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html&gt;`_ dataset for optical flow.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            FlyingThings3D</span>
+<span class="sd">                frames_cleanpass</span>
+<span class="sd">                    TEST</span>
+<span class="sd">                    TRAIN</span>
+<span class="sd">                frames_finalpass</span>
+<span class="sd">                    TEST</span>
+<span class="sd">                    TRAIN</span>
+<span class="sd">                optical_flow</span>
+<span class="sd">                    TEST</span>
+<span class="sd">                    TRAIN</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the intel FlyingThings3D Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, either &quot;train&quot; (default) or &quot;test&quot;</span>
+<span class="sd">        pass_name (string, optional): The pass to use, either &quot;clean&quot; (default) or &quot;final&quot; or &quot;both&quot;. See link above for</span>
+<span class="sd">            details on the different passes.</span>
+<span class="sd">        camera (string, optional): Which camera to return images from. Can be either &quot;left&quot; (default) or &quot;right&quot; or &quot;both&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in</span>
+<span class="sd">            ``img1, img2, flow, valid_flow_mask`` and returns a transformed version.</span>
+<span class="sd">            ``valid_flow_mask`` is expected for consistency with other datasets which</span>
+<span class="sd">            return a built-in valid mask, such as :class:`~torchvision.datasets.KittiFlow`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">pass_name</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;clean&quot;</span><span class="p">,</span>
+        <span class="n">camera</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;left&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="n">split</span> <span class="o">=</span> <span class="n">split</span><span class="o">.</span><span class="n">upper</span><span class="p">()</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">pass_name</span><span class="p">,</span> <span class="s2">&quot;pass_name&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;clean&quot;</span><span class="p">,</span> <span class="s2">&quot;final&quot;</span><span class="p">,</span> <span class="s2">&quot;both&quot;</span><span class="p">))</span>
+        <span class="n">passes</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;clean&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;frames_cleanpass&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;final&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;frames_finalpass&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;both&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;frames_cleanpass&quot;</span><span class="p">,</span> <span class="s2">&quot;frames_finalpass&quot;</span><span class="p">],</span>
+        <span class="p">}[</span><span class="n">pass_name</span><span class="p">]</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">camera</span><span class="p">,</span> <span class="s2">&quot;camera&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;left&quot;</span><span class="p">,</span> <span class="s2">&quot;right&quot;</span><span class="p">,</span> <span class="s2">&quot;both&quot;</span><span class="p">))</span>
+        <span class="n">cameras</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;left&quot;</span><span class="p">,</span> <span class="s2">&quot;right&quot;</span><span class="p">]</span> <span class="k">if</span> <span class="n">camera</span> <span class="o">==</span> <span class="s2">&quot;both&quot;</span> <span class="k">else</span> <span class="p">[</span><span class="n">camera</span><span class="p">]</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;FlyingThings3D&quot;</span>
+
+        <span class="n">directions</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;into_future&quot;</span><span class="p">,</span> <span class="s2">&quot;into_past&quot;</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">pass_name</span><span class="p">,</span> <span class="n">camera</span><span class="p">,</span> <span class="n">direction</span> <span class="ow">in</span> <span class="n">itertools</span><span class="o">.</span><span class="n">product</span><span class="p">(</span><span class="n">passes</span><span class="p">,</span> <span class="n">cameras</span><span class="p">,</span> <span class="n">directions</span><span class="p">):</span>
+            <span class="n">image_dirs</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">pass_name</span> <span class="o">/</span> <span class="n">split</span> <span class="o">/</span> <span class="s2">&quot;*/*&quot;</span><span class="p">)))</span>
+            <span class="n">image_dirs</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">Path</span><span class="p">(</span><span class="n">image_dir</span><span class="p">)</span> <span class="o">/</span> <span class="n">camera</span> <span class="k">for</span> <span class="n">image_dir</span> <span class="ow">in</span> <span class="n">image_dirs</span><span class="p">)</span>
+
+            <span class="n">flow_dirs</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;optical_flow&quot;</span> <span class="o">/</span> <span class="n">split</span> <span class="o">/</span> <span class="s2">&quot;*/*&quot;</span><span class="p">)))</span>
+            <span class="n">flow_dirs</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">Path</span><span class="p">(</span><span class="n">flow_dir</span><span class="p">)</span> <span class="o">/</span> <span class="n">direction</span> <span class="o">/</span> <span class="n">camera</span> <span class="k">for</span> <span class="n">flow_dir</span> <span class="ow">in</span> <span class="n">flow_dirs</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">image_dirs</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">flow_dirs</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span>
+                    <span class="s2">&quot;Could not find the FlyingThings3D flow images. &quot;</span>
+                    <span class="s2">&quot;Please make sure the directory structure is correct.&quot;</span>
+                <span class="p">)</span>
+
+            <span class="k">for</span> <span class="n">image_dir</span><span class="p">,</span> <span class="n">flow_dir</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">image_dirs</span><span class="p">,</span> <span class="n">flow_dirs</span><span class="p">):</span>
+                <span class="n">images</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">image_dir</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)))</span>
+                <span class="n">flows</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">flow_dir</span> <span class="o">/</span> <span class="s2">&quot;*.pfm&quot;</span><span class="p">)))</span>
+                <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">flows</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
+                    <span class="k">if</span> <span class="n">direction</span> <span class="o">==</span> <span class="s2">&quot;into_future&quot;</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span> <span class="o">+=</span> <span class="p">[[</span><span class="n">images</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">images</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]]]</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span> <span class="o">+=</span> <span class="p">[</span><span class="n">flows</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span>
+                    <span class="k">elif</span> <span class="n">direction</span> <span class="o">==</span> <span class="s2">&quot;into_past&quot;</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span> <span class="o">+=</span> <span class="p">[[</span><span class="n">images</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">],</span> <span class="n">images</span><span class="p">[</span><span class="n">i</span><span class="p">]]]</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span> <span class="o">+=</span> <span class="p">[</span><span class="n">flows</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]]</span>
+
+<div class="viewcode-block" id="FlyingThings3D.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">T1</span><span class="p">,</span> <span class="n">T2</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3-tuple with ``(img1, img2, flow)``.</span>
+<span class="sd">            The flow is a numpy array of shape (2, H, W) and the images are PIL images.</span>
+<span class="sd">            ``flow`` is None if ``split=&quot;test&quot;``.</span>
+<span class="sd">            If a valid flow mask is generated within the ``transforms`` parameter,</span>
+<span class="sd">            a 4-tuple with ``(img1, img2, flow, valid_flow_mask)`` is returned.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="nf">_read_flow</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_read_pfm</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="HD1K"><a class="viewcode-back" href="../../../generated/torchvision.datasets.HD1K.html#torchvision.datasets.HD1K">[docs]</a><span class="k">class</span> <span class="nc">HD1K</span><span class="p">(</span><span class="n">FlowDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`HD1K &lt;http://hci-benchmark.iwr.uni-heidelberg.de/&gt;`__ dataset for optical flow.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            hd1k</span>
+<span class="sd">                hd1k_challenge</span>
+<span class="sd">                    image_2</span>
+<span class="sd">                hd1k_flow_gt</span>
+<span class="sd">                    flow_occ</span>
+<span class="sd">                hd1k_input</span>
+<span class="sd">                    image_2</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the HD1K Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, either &quot;train&quot; (default) or &quot;test&quot;</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in</span>
+<span class="sd">            ``img1, img2, flow, valid_flow_mask`` and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_has_builtin_flow_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;hd1k&quot;</span>
+        <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="c1"># There are 36 &quot;sequences&quot; and we don&#39;t want seq i to overlap with seq i + 1, so we need this for loop</span>
+            <span class="k">for</span> <span class="n">seq_idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">36</span><span class="p">):</span>
+                <span class="n">flows</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;hd1k_flow_gt&quot;</span> <span class="o">/</span> <span class="s2">&quot;flow_occ&quot;</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">seq_idx</span><span class="si">:</span><span class="s2">06d</span><span class="si">}</span><span class="s2">_*.png&quot;</span><span class="p">)))</span>
+                <span class="n">images</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;hd1k_input&quot;</span> <span class="o">/</span> <span class="s2">&quot;image_2&quot;</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">seq_idx</span><span class="si">:</span><span class="s2">06d</span><span class="si">}</span><span class="s2">_*.png&quot;</span><span class="p">)))</span>
+                <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">flows</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_flow_list</span> <span class="o">+=</span> <span class="p">[</span><span class="n">flows</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span> <span class="o">+=</span> <span class="p">[[</span><span class="n">images</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">images</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]]]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">images1</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;hd1k_challenge&quot;</span> <span class="o">/</span> <span class="s2">&quot;image_2&quot;</span> <span class="o">/</span> <span class="s2">&quot;*10.png&quot;</span><span class="p">)))</span>
+            <span class="n">images2</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;hd1k_challenge&quot;</span> <span class="o">/</span> <span class="s2">&quot;image_2&quot;</span> <span class="o">/</span> <span class="s2">&quot;*11.png&quot;</span><span class="p">)))</span>
+            <span class="k">for</span> <span class="n">image1</span><span class="p">,</span> <span class="n">image2</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">images1</span><span class="p">,</span> <span class="n">images2</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span> <span class="o">+=</span> <span class="p">[[</span><span class="n">image1</span><span class="p">,</span> <span class="n">image2</span><span class="p">]]</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_image_list</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span>
+                <span class="s2">&quot;Could not find the HD1K images. Please make sure the directory structure is correct.&quot;</span>
+            <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_flow</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="n">_read_16bits_png_with_flow_and_valid_mask</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span>
+
+<div class="viewcode-block" id="HD1K.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.HD1K.html#torchvision.datasets.HD1K.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">T1</span><span class="p">,</span> <span class="n">T2</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img1, img2, flow, valid_flow_mask)`` where ``valid_flow_mask``</span>
+<span class="sd">            is a numpy boolean mask of shape (H, W)</span>
+<span class="sd">            indicating which flow values are valid. The flow is a numpy array of</span>
+<span class="sd">            shape (2, H, W) and the images are PIL images. ``flow`` and ``valid_flow_mask`` are None if</span>
+<span class="sd">            ``split=&quot;test&quot;``.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">)</span></div></div>
+
+
+<span class="k">def</span> <span class="nf">_read_flo</span><span class="p">(</span><span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Read .flo file in Middlebury format&quot;&quot;&quot;</span>
+    <span class="c1"># Code adapted from:</span>
+    <span class="c1"># http://stackoverflow.com/questions/28013200/reading-middlebury-flow-files-with-python-bytes-array-numpy</span>
+    <span class="c1"># Everything needs to be in little Endian according to</span>
+    <span class="c1"># https://vision.middlebury.edu/flow/code/flow-code/README.txt</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file_name</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="n">magic</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="s2">&quot;c&quot;</span><span class="p">,</span> <span class="n">count</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span><span class="o">.</span><span class="n">tobytes</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">magic</span> <span class="o">!=</span> <span class="sa">b</span><span class="s2">&quot;PIEH&quot;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Magic number incorrect. Invalid .flo file&quot;</span><span class="p">)</span>
+
+        <span class="n">w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="s2">&quot;&lt;i4&quot;</span><span class="p">,</span> <span class="n">count</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+        <span class="n">h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="s2">&quot;&lt;i4&quot;</span><span class="p">,</span> <span class="n">count</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="s2">&quot;&lt;f4&quot;</span><span class="p">,</span> <span class="n">count</span><span class="o">=</span><span class="mi">2</span> <span class="o">*</span> <span class="n">w</span> <span class="o">*</span> <span class="n">h</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">data</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_read_16bits_png_with_flow_and_valid_mask</span><span class="p">(</span><span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]:</span>
+
+    <span class="n">flow_and_valid</span> <span class="o">=</span> <span class="n">_read_png_16</span><span class="p">(</span><span class="n">file_name</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">flow</span><span class="p">,</span> <span class="n">valid_flow_mask</span> <span class="o">=</span> <span class="n">flow_and_valid</span><span class="p">[:</span><span class="mi">2</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:],</span> <span class="n">flow_and_valid</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span>
+    <span class="n">flow</span> <span class="o">=</span> <span class="p">(</span><span class="n">flow</span> <span class="o">-</span> <span class="mi">2</span><span class="o">**</span><span class="mi">15</span><span class="p">)</span> <span class="o">/</span> <span class="mi">64</span>  <span class="c1"># This conversion is explained somewhere on the kitti archive</span>
+    <span class="n">valid_flow_mask</span> <span class="o">=</span> <span class="n">valid_flow_mask</span><span class="o">.</span><span class="n">bool</span><span class="p">()</span>
+
+    <span class="c1"># For consistency with other datasets, we convert to numpy</span>
+    <span class="k">return</span> <span class="n">flow</span><span class="o">.</span><span class="n">numpy</span><span class="p">(),</span> <span class="n">valid_flow_mask</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/_stereo_matching.html b/0.16/_modules/torchvision/datasets/_stereo_matching.html
new file mode 100644
index 00000000000..9930f07a3a6
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/_stereo_matching.html
@@ -0,0 +1,1893 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets._stereo_matching &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets._stereo_matching</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets._stereo_matching</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">functools</span>
+<span class="kn">import</span> <span class="nn">json</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">random</span>
+<span class="kn">import</span> <span class="nn">shutil</span>
+<span class="kn">from</span> <span class="nn">abc</span> <span class="kn">import</span> <span class="n">ABC</span><span class="p">,</span> <span class="n">abstractmethod</span>
+<span class="kn">from</span> <span class="nn">glob</span> <span class="kn">import</span> <span class="n">glob</span>
+<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_read_pfm</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+<span class="n">T1</span> <span class="o">=</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]</span>
+<span class="n">T2</span> <span class="o">=</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">()</span>
+
+<span class="n">_read_pfm_file</span> <span class="o">=</span> <span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span><span class="n">_read_pfm</span><span class="p">,</span> <span class="n">slice_channels</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">StereoMatchingDataset</span><span class="p">(</span><span class="n">ABC</span><span class="p">,</span> <span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Base interface for Stereo matching datasets&quot;&quot;&quot;</span>
+
+    <span class="n">_has_built_in_disparity_mask</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            root(str): Root directory of the dataset.</span>
+<span class="sd">            transforms(callable, optional): A function/transform that takes in Tuples of</span>
+<span class="sd">                (images, disparities, valid_masks) and returns a transformed version of each of them.</span>
+<span class="sd">                images is a Tuple of (``PIL.Image``, ``PIL.Image``)</span>
+<span class="sd">                disparities is a Tuple of (``np.ndarray``, ``np.ndarray``) with shape (1, H, W)</span>
+<span class="sd">                valid_masks is a Tuple of (``np.ndarray``, ``np.ndarray``) with shape (H, W)</span>
+<span class="sd">                In some cases, when a dataset does not provide disparities, the ``disparities`` and</span>
+<span class="sd">                ``valid_masks`` can be Tuples containing None values.</span>
+<span class="sd">                For training splits generally the datasets provide a minimal guarantee of</span>
+<span class="sd">                images: (``PIL.Image``, ``PIL.Image``)</span>
+<span class="sd">                disparities: (``np.ndarray``, ``None``) with shape (1, H, W)</span>
+<span class="sd">                Optionally, based on the dataset, it can return a ``mask`` as well:</span>
+<span class="sd">                valid_masks: (``np.ndarray | None``, ``None``) with shape (H, W)</span>
+<span class="sd">                For some test splits, the datasets provides outputs that look like:</span>
+<span class="sd">                imgaes: (``PIL.Image``, ``PIL.Image``)</span>
+<span class="sd">                disparities: (``None``, ``None``)</span>
+<span class="sd">                Optionally, based on the dataset, it can return a ``mask`` as well:</span>
+<span class="sd">                valid_masks: (``None``, ``None``)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># type: ignore</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># type: ignore</span>
+
+    <span class="k">def</span> <span class="nf">_read_img</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Path</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">img</span><span class="o">.</span><span class="n">mode</span> <span class="o">!=</span> <span class="s2">&quot;RGB&quot;</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span>
+
+    <span class="k">def</span> <span class="nf">_scan_pairs</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">paths_left_pattern</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">paths_right_pattern</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]]]:</span>
+
+        <span class="n">left_paths</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="n">paths_left_pattern</span><span class="p">)))</span>
+
+        <span class="n">right_paths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]</span>
+        <span class="k">if</span> <span class="n">paths_right_pattern</span><span class="p">:</span>
+            <span class="n">right_paths</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><span class="n">glob</span><span class="p">(</span><span class="n">paths_right_pattern</span><span class="p">)))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">right_paths</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="kc">None</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="n">left_paths</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">left_paths</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Could not find any files matching the patterns: </span><span class="si">{</span><span class="n">paths_left_pattern</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">right_paths</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Could not find any files matching the patterns: </span><span class="si">{</span><span class="n">paths_right_pattern</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">left_paths</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">right_paths</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Found </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">left_paths</span><span class="p">)</span><span class="si">}</span><span class="s2"> left files but </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">right_paths</span><span class="p">)</span><span class="si">}</span><span class="s2"> right files using:</span><span class="se">\n</span><span class="s2"> &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;left pattern: </span><span class="si">{</span><span class="n">paths_left_pattern</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;right pattern: </span><span class="si">{</span><span class="n">paths_right_pattern</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">paths</span> <span class="o">=</span> <span class="nb">list</span><span class="p">((</span><span class="n">left</span><span class="p">,</span> <span class="n">right</span><span class="p">)</span> <span class="k">for</span> <span class="n">left</span><span class="p">,</span> <span class="n">right</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">left_paths</span><span class="p">,</span> <span class="n">right_paths</span><span class="p">))</span>
+        <span class="k">return</span> <span class="n">paths</span>
+
+    <span class="nd">@abstractmethod</span>
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]:</span>
+        <span class="c1"># function that returns a disparity map and an occlusion map</span>
+        <span class="k">pass</span>
+
+    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">T1</span><span class="p">,</span> <span class="n">T2</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3 or 4-tuple with ``(img_left, img_right, disparity, Optional[valid_mask])`` where ``valid_mask``</span>
+<span class="sd">                can be a numpy boolean mask of shape (H, W) if the dataset provides a file</span>
+<span class="sd">                indicating which disparity pixels are valid. The disparity is a numpy array of</span>
+<span class="sd">                shape (1, H, W) and the images are PIL images. ``disparity`` is None for</span>
+<span class="sd">                datasets on which for ``split=&quot;test&quot;`` the authors did not provide annotations.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img_left</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_read_img</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">img_right</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_read_img</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="mi">1</span><span class="p">])</span>
+
+        <span class="n">dsp_map_left</span><span class="p">,</span> <span class="n">valid_mask_left</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">dsp_map_right</span><span class="p">,</span> <span class="n">valid_mask_right</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="mi">1</span><span class="p">])</span>
+
+        <span class="n">imgs</span> <span class="o">=</span> <span class="p">(</span><span class="n">img_left</span><span class="p">,</span> <span class="n">img_right</span><span class="p">)</span>
+        <span class="n">dsp_maps</span> <span class="o">=</span> <span class="p">(</span><span class="n">dsp_map_left</span><span class="p">,</span> <span class="n">dsp_map_right</span><span class="p">)</span>
+        <span class="n">valid_masks</span> <span class="o">=</span> <span class="p">(</span><span class="n">valid_mask_left</span><span class="p">,</span> <span class="n">valid_mask_right</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="p">(</span>
+                <span class="n">imgs</span><span class="p">,</span>
+                <span class="n">dsp_maps</span><span class="p">,</span>
+                <span class="n">valid_masks</span><span class="p">,</span>
+            <span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">imgs</span><span class="p">,</span> <span class="n">dsp_maps</span><span class="p">,</span> <span class="n">valid_masks</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_has_built_in_disparity_mask</span> <span class="ow">or</span> <span class="n">valid_masks</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">imgs</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">imgs</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dsp_maps</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">cast</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">valid_masks</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">imgs</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">imgs</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dsp_maps</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="CarlaStereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CarlaStereo.html#torchvision.datasets.CarlaStereo">[docs]</a><span class="k">class</span> <span class="nc">CarlaStereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Carla simulator data linked in the `CREStereo github repo &lt;https://github.com/megvii-research/CREStereo&gt;`_.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            carla-highres</span>
+<span class="sd">                trainingF</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        img0.png</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        disp0GT.pfm</span>
+<span class="sd">                        disp1GT.pfm</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                        img0.png</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        disp0GT.pfm</span>
+<span class="sd">                        disp1GT.pfm</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                    ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where `carla-highres` is located.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;carla-highres&quot;</span>
+
+        <span class="n">left_image_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;trainingF&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;im0.png&quot;</span><span class="p">)</span>
+        <span class="n">right_image_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;trainingF&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;im1.png&quot;</span><span class="p">)</span>
+        <span class="n">imgs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_image_pattern</span><span class="p">,</span> <span class="n">right_image_pattern</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">=</span> <span class="n">imgs</span>
+
+        <span class="n">left_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;trainingF&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;disp0GT.pfm&quot;</span><span class="p">)</span>
+        <span class="n">right_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;trainingF&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;disp1GT.pfm&quot;</span><span class="p">)</span>
+        <span class="n">disparities</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_disparity_pattern</span><span class="p">,</span> <span class="n">right_disparity_pattern</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="n">disparities</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="kc">None</span><span class="p">]:</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">_read_pfm_file</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">abs</span><span class="p">(</span><span class="n">disparity_map</span><span class="p">)</span>  <span class="c1"># ensure that the disparity is positive</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="CarlaStereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CarlaStereo.html#torchvision.datasets.CarlaStereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T1</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3-tuple with ``(img_left, img_right, disparity)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            If a ``valid_mask`` is generated within the ``transforms`` parameter,</span>
+<span class="sd">            a 4-tuple with ``(img_left, img_right, disparity, valid_mask)`` is returned.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T1</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="Kitti2012Stereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kitti2012Stereo.html#torchvision.datasets.Kitti2012Stereo">[docs]</a><span class="k">class</span> <span class="nc">Kitti2012Stereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    KITTI dataset from the `2012 stereo evaluation benchmark &lt;http://www.cvlibs.net/datasets/kitti/eval_stereo_flow.php&gt;`_.</span>
+<span class="sd">    Uses the RGB images for consistency with KITTI 2015.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            Kitti2012</span>
+<span class="sd">                testing</span>
+<span class="sd">                    colored_0</span>
+<span class="sd">                        1_10.png</span>
+<span class="sd">                        2_10.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    colored_1</span>
+<span class="sd">                        1_10.png</span>
+<span class="sd">                        2_10.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                training</span>
+<span class="sd">                    colored_0</span>
+<span class="sd">                        1_10.png</span>
+<span class="sd">                        2_10.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    colored_1</span>
+<span class="sd">                        1_10.png</span>
+<span class="sd">                        2_10.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    disp_noc</span>
+<span class="sd">                        1.png</span>
+<span class="sd">                        2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    calib</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where `Kitti2012` is located.</span>
+<span class="sd">        split (string, optional): The dataset split of scenes, either &quot;train&quot; (default) or &quot;test&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_has_built_in_disparity_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;Kitti2012&quot;</span> <span class="o">/</span> <span class="p">(</span><span class="n">split</span> <span class="o">+</span> <span class="s2">&quot;ing&quot;</span><span class="p">)</span>
+
+        <span class="n">left_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;colored_0&quot;</span> <span class="o">/</span> <span class="s2">&quot;*_10.png&quot;</span><span class="p">)</span>
+        <span class="n">right_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;colored_1&quot;</span> <span class="o">/</span> <span class="s2">&quot;*_10.png&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_img_pattern</span><span class="p">,</span> <span class="n">right_img_pattern</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="n">disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;disp_noc&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">disparity_pattern</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="nb">list</span><span class="p">((</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="kc">None</span><span class="p">]:</span>
+        <span class="c1"># test split has no disparity maps</span>
+        <span class="k">if</span> <span class="n">file_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">))</span> <span class="o">/</span> <span class="mf">256.0</span>
+        <span class="c1"># unsqueeze the disparity map into (C, H, W) format</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">disparity_map</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="Kitti2012Stereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kitti2012Stereo.html#torchvision.datasets.Kitti2012Stereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T1</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img_left, img_right, disparity, valid_mask)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            ``valid_mask`` is implicitly ``None`` if the ``transforms`` parameter does not</span>
+<span class="sd">            generate a valid mask.</span>
+<span class="sd">            Both ``disparity`` and ``valid_mask`` are ``None`` if the dataset split is test.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T1</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="Kitti2015Stereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kitti2015Stereo.html#torchvision.datasets.Kitti2015Stereo">[docs]</a><span class="k">class</span> <span class="nc">Kitti2015Stereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    KITTI dataset from the `2015 stereo evaluation benchmark &lt;http://www.cvlibs.net/datasets/kitti/eval_scene_flow.php&gt;`_.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            Kitti2015</span>
+<span class="sd">                testing</span>
+<span class="sd">                    image_2</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        img2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    image_3</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        img2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                training</span>
+<span class="sd">                    image_2</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        img2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    image_3</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        img2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    disp_occ_0</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        img2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    disp_occ_1</span>
+<span class="sd">                        img1.png</span>
+<span class="sd">                        img2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    calib</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where `Kitti2015` is located.</span>
+<span class="sd">        split (string, optional): The dataset split of scenes, either &quot;train&quot; (default) or &quot;test&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_has_built_in_disparity_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;Kitti2015&quot;</span> <span class="o">/</span> <span class="p">(</span><span class="n">split</span> <span class="o">+</span> <span class="s2">&quot;ing&quot;</span><span class="p">)</span>
+        <span class="n">left_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;image_2&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+        <span class="n">right_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;image_3&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_img_pattern</span><span class="p">,</span> <span class="n">right_img_pattern</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="n">left_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;disp_occ_0&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="n">right_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;disp_occ_1&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_disparity_pattern</span><span class="p">,</span> <span class="n">right_disparity_pattern</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="nb">list</span><span class="p">((</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="kc">None</span><span class="p">]:</span>
+        <span class="c1"># test split has no disparity maps</span>
+        <span class="k">if</span> <span class="n">file_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">))</span> <span class="o">/</span> <span class="mf">256.0</span>
+        <span class="c1"># unsqueeze the disparity map into (C, H, W) format</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">disparity_map</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="Kitti2015Stereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kitti2015Stereo.html#torchvision.datasets.Kitti2015Stereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T1</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img_left, img_right, disparity, valid_mask)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            ``valid_mask`` is implicitly ``None`` if the ``transforms`` parameter does not</span>
+<span class="sd">            generate a valid mask.</span>
+<span class="sd">            Both ``disparity`` and ``valid_mask`` are ``None`` if the dataset split is test.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T1</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="Middlebury2014Stereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Middlebury2014Stereo.html#torchvision.datasets.Middlebury2014Stereo">[docs]</a><span class="k">class</span> <span class="nc">Middlebury2014Stereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Publicly available scenes from the Middlebury dataset `2014 version &lt;https://vision.middlebury.edu/stereo/data/scenes2014/&gt;`.</span>
+
+<span class="sd">    The dataset mostly follows the original format, without containing the ambient subdirectories.  : ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            Middlebury2014</span>
+<span class="sd">                train</span>
+<span class="sd">                    scene1-{perfect,imperfect}</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                        im{0,1}.png</span>
+<span class="sd">                        im1E.png</span>
+<span class="sd">                        im1L.png</span>
+<span class="sd">                        disp{0,1}.pfm</span>
+<span class="sd">                        disp{0,1}-n.png</span>
+<span class="sd">                        disp{0,1}-sd.pfm</span>
+<span class="sd">                        disp{0,1}y.pfm</span>
+<span class="sd">                    scene2-{perfect,imperfect}</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                        im{0,1}.png</span>
+<span class="sd">                        im1E.png</span>
+<span class="sd">                        im1L.png</span>
+<span class="sd">                        disp{0,1}.pfm</span>
+<span class="sd">                        disp{0,1}-n.png</span>
+<span class="sd">                        disp{0,1}-sd.pfm</span>
+<span class="sd">                        disp{0,1}y.pfm</span>
+<span class="sd">                    ...</span>
+<span class="sd">                additional</span>
+<span class="sd">                    scene1-{perfect,imperfect}</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                        im{0,1}.png</span>
+<span class="sd">                        im1E.png</span>
+<span class="sd">                        im1L.png</span>
+<span class="sd">                        disp{0,1}.pfm</span>
+<span class="sd">                        disp{0,1}-n.png</span>
+<span class="sd">                        disp{0,1}-sd.pfm</span>
+<span class="sd">                        disp{0,1}y.pfm</span>
+<span class="sd">                    ...</span>
+<span class="sd">                test</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                        im{0,1}.png</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                        im{0,1}.png</span>
+<span class="sd">                    ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the Middleburry 2014 Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split of scenes, either &quot;train&quot; (default), &quot;test&quot;, or &quot;additional&quot;</span>
+<span class="sd">        use_ambient_views (boolean, optional): Whether to use different expose or lightning views when possible.</span>
+<span class="sd">            The dataset samples with equal probability between ``[im1.png, im1E.png, im1L.png]``.</span>
+<span class="sd">        calibration (string, optional): Whether or not to use the calibrated (default) or uncalibrated scenes.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">        download (boolean, optional): Whether or not to download the dataset in the ``root`` directory.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">splits</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;Adirondack&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Jadeplant&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Motorcycle&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Piano&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Pipes&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Playroom&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Playtable&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Recycle&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Shelves&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Vintage&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;additional&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;Backpack&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Bicycle1&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Cable&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Classroom1&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Couch&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Flowers&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Mask&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Shopvac&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Sticks&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Storage&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Sword1&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Sword2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Umbrella&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;Plants&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Classroom2E&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Classroom2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Australia&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;DjembeL&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;CrusadeP&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Crusade&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Hoops&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Bicycle2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Staircase&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Newkuba&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;AustraliaP&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Djembe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Livingroom&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Computer&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+    <span class="p">}</span>
+
+    <span class="n">_has_built_in_disparity_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">calibration</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;perfect&quot;</span><span class="p">,</span>
+        <span class="n">use_ambient_views</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;additional&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">split</span>
+
+        <span class="k">if</span> <span class="n">calibration</span><span class="p">:</span>
+            <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">calibration</span><span class="p">,</span> <span class="s2">&quot;calibration&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;perfect&quot;</span><span class="p">,</span> <span class="s2">&quot;imperfect&quot;</span><span class="p">,</span> <span class="s2">&quot;both&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">))</span>  <span class="c1"># type: ignore</span>
+            <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Split &#39;test&#39; has only no calibration settings, please set `calibration=None`.&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">split</span> <span class="o">!=</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Split &#39;</span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">&#39; has calibration settings, however None was provided as an argument.&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">Setting calibration to &#39;perfect&#39; for split &#39;</span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">&#39;. Available calibration settings are: &#39;perfect&#39;, &#39;imperfect&#39;, &#39;both&#39;.&quot;</span><span class="p">,</span>
+                <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download_dataset</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;Middlebury2014&quot;</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The </span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2"> directory was not found in the provided root directory&quot;</span><span class="p">)</span>
+
+        <span class="n">split_scenes</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">splits</span><span class="p">[</span><span class="n">split</span><span class="p">]</span>
+        <span class="c1"># check that the provided root folder contains the scene splits</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">any</span><span class="p">(</span>
+            <span class="c1"># using startswith to account for perfect / imperfect calibrartion</span>
+            <span class="n">scene</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="n">s</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">scene</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">split_scenes</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Provided root folder does not contain any scenes from the </span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2"> split.&quot;</span><span class="p">)</span>
+
+        <span class="n">calibrartion_suffixes</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="kc">None</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;perfect&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;-perfect&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;imperfect&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;-imperfect&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;both&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;-perfect&quot;</span><span class="p">,</span> <span class="s2">&quot;-imperfect&quot;</span><span class="p">],</span>
+        <span class="p">}[</span><span class="n">calibration</span><span class="p">]</span>
+
+        <span class="k">for</span> <span class="n">calibration_suffix</span> <span class="ow">in</span> <span class="n">calibrartion_suffixes</span><span class="p">:</span>
+            <span class="n">scene_pattern</span> <span class="o">=</span> <span class="s2">&quot;*&quot;</span> <span class="o">+</span> <span class="n">calibration_suffix</span>
+            <span class="n">left_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split</span> <span class="o">/</span> <span class="n">scene_pattern</span> <span class="o">/</span> <span class="s2">&quot;im0.png&quot;</span><span class="p">)</span>
+            <span class="n">right_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split</span> <span class="o">/</span> <span class="n">scene_pattern</span> <span class="o">/</span> <span class="s2">&quot;im1.png&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_img_pattern</span><span class="p">,</span> <span class="n">right_img_pattern</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="nb">list</span><span class="p">((</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">left_dispartity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split</span> <span class="o">/</span> <span class="n">scene_pattern</span> <span class="o">/</span> <span class="s2">&quot;disp0.pfm&quot;</span><span class="p">)</span>
+                <span class="n">right_dispartity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">split</span> <span class="o">/</span> <span class="n">scene_pattern</span> <span class="o">/</span> <span class="s2">&quot;disp1.pfm&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_dispartity_pattern</span><span class="p">,</span> <span class="n">right_dispartity_pattern</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_ambient_views</span> <span class="o">=</span> <span class="n">use_ambient_views</span>
+
+    <span class="k">def</span> <span class="nf">_read_img</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Path</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Function that reads either the original right image or an augmented view when ``use_ambient_views`` is True.</span>
+<span class="sd">        When ``use_ambient_views`` is True, the dataset will return at random one of ``[im1.png, im1E.png, im1L.png]``</span>
+<span class="sd">        as the right image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">ambient_file_paths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Path</span><span class="p">]]</span>  <span class="c1"># make mypy happy</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">file_path</span><span class="p">,</span> <span class="n">Path</span><span class="p">):</span>
+            <span class="n">file_path</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">file_path</span><span class="o">.</span><span class="n">name</span> <span class="o">==</span> <span class="s2">&quot;im1.png&quot;</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_ambient_views</span><span class="p">:</span>
+            <span class="n">base_path</span> <span class="o">=</span> <span class="n">file_path</span><span class="o">.</span><span class="n">parent</span>
+            <span class="c1"># initialize sampleable container</span>
+            <span class="n">ambient_file_paths</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">base_path</span> <span class="o">/</span> <span class="n">view_name</span> <span class="k">for</span> <span class="n">view_name</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;im1E.png&quot;</span><span class="p">,</span> <span class="s2">&quot;im1L.png&quot;</span><span class="p">])</span>
+            <span class="c1"># double check that we&#39;re not going to try to read from an invalid file path</span>
+            <span class="n">ambient_file_paths</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">p</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">p</span><span class="p">),</span> <span class="n">ambient_file_paths</span><span class="p">))</span>
+            <span class="c1"># keep the original image as an option as well for uniform sampling between base views</span>
+            <span class="n">ambient_file_paths</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+            <span class="n">file_path</span> <span class="o">=</span> <span class="n">random</span><span class="o">.</span><span class="n">choice</span><span class="p">(</span><span class="n">ambient_file_paths</span><span class="p">)</span>  <span class="c1"># type: ignore</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_read_img</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">],</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]:</span>
+        <span class="c1"># test split has not disparity maps</span>
+        <span class="k">if</span> <span class="n">file_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">_read_pfm_file</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">abs</span><span class="p">(</span><span class="n">disparity_map</span><span class="p">)</span>  <span class="c1"># ensure that the disparity is positive</span>
+        <span class="n">disparity_map</span><span class="p">[</span><span class="n">disparity_map</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">inf</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>  <span class="c1"># remove infinite disparities</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="p">(</span><span class="n">disparity_map</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># mask out invalid disparities</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+    <span class="k">def</span> <span class="nf">_download_dataset</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">base_url</span> <span class="o">=</span> <span class="s2">&quot;https://vision.middlebury.edu/stereo/data/scenes2014/zip&quot;</span>
+        <span class="c1"># train and additional splits have 2 different calibration settings</span>
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;Middlebury2014&quot;</span>
+        <span class="n">split_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span>
+
+        <span class="k">if</span> <span class="n">split_name</span> <span class="o">!=</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">split_scene</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">splits</span><span class="p">[</span><span class="n">split_name</span><span class="p">]:</span>
+                <span class="n">split_root</span> <span class="o">=</span> <span class="n">root</span> <span class="o">/</span> <span class="n">split_name</span>
+                <span class="k">for</span> <span class="n">calibration</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;perfect&quot;</span><span class="p">,</span> <span class="s2">&quot;imperfect&quot;</span><span class="p">]:</span>
+                    <span class="n">scene_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">split_scene</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="n">calibration</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="n">scene_url</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">base_url</span><span class="si">}</span><span class="s2">/</span><span class="si">{</span><span class="n">scene_name</span><span class="si">}</span><span class="s2">.zip&quot;</span>
+                    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Downloading </span><span class="si">{</span><span class="n">scene_url</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                    <span class="c1"># download the scene only if it doesn&#39;t exist</span>
+                    <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">split_root</span> <span class="o">/</span> <span class="n">scene_name</span><span class="p">)</span><span class="o">.</span><span class="n">exists</span><span class="p">():</span>
+                        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+                            <span class="n">url</span><span class="o">=</span><span class="n">scene_url</span><span class="p">,</span>
+                            <span class="n">filename</span><span class="o">=</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">scene_name</span><span class="si">}</span><span class="s2">.zip&quot;</span><span class="p">,</span>
+                            <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="n">split_root</span><span class="p">),</span>
+                            <span class="n">remove_finished</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;test&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="nb">any</span><span class="p">(</span><span class="n">s</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;test&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">splits</span><span class="p">[</span><span class="s2">&quot;test&quot;</span><span class="p">]):</span>
+                <span class="c1"># test split is downloaded from a different location</span>
+                <span class="n">test_set_url</span> <span class="o">=</span> <span class="s2">&quot;https://vision.middlebury.edu/stereo/submit3/zip/MiddEval3-data-F.zip&quot;</span>
+                <span class="c1"># the unzip is going to produce a directory MiddEval3 with two subdirectories trainingF and testF</span>
+                <span class="c1"># we want to move the contents from testF into the  directory</span>
+                <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">url</span><span class="o">=</span><span class="n">test_set_url</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span><span class="p">),</span> <span class="n">remove_finished</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="k">for</span> <span class="n">scene_dir</span><span class="p">,</span> <span class="n">scene_names</span><span class="p">,</span> <span class="n">_</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">walk</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;MiddEval3/testF&quot;</span><span class="p">)):</span>
+                    <span class="k">for</span> <span class="n">scene</span> <span class="ow">in</span> <span class="n">scene_names</span><span class="p">:</span>
+                        <span class="n">scene_dst_dir</span> <span class="o">=</span> <span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;test&quot;</span>
+                        <span class="n">scene_src_dir</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">scene_dir</span><span class="p">)</span> <span class="o">/</span> <span class="n">scene</span>
+                        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">scene_dst_dir</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                        <span class="n">shutil</span><span class="o">.</span><span class="n">move</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">scene_src_dir</span><span class="p">),</span> <span class="nb">str</span><span class="p">(</span><span class="n">scene_dst_dir</span><span class="p">))</span>
+
+                <span class="c1"># cleanup MiddEval3 directory</span>
+                <span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;MiddEval3&quot;</span><span class="p">))</span>
+
+<div class="viewcode-block" id="Middlebury2014Stereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Middlebury2014Stereo.html#torchvision.datasets.Middlebury2014Stereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T2</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img_left, img_right, disparity, valid_mask)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            ``valid_mask`` is implicitly ``None`` for `split=test`.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T2</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="CREStereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CREStereo.html#torchvision.datasets.CREStereo">[docs]</a><span class="k">class</span> <span class="nc">CREStereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Synthetic dataset used in training the `CREStereo &lt;https://arxiv.org/pdf/2203.11483.pdf&gt;`_ architecture.</span>
+<span class="sd">    Dataset details on the official paper `repo &lt;https://github.com/megvii-research/CREStereo&gt;`_.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            CREStereo</span>
+<span class="sd">                tree</span>
+<span class="sd">                    img1_left.jpg</span>
+<span class="sd">                    img1_right.jpg</span>
+<span class="sd">                    img1_left.disp.jpg</span>
+<span class="sd">                    img1_right.disp.jpg</span>
+<span class="sd">                    img2_left.jpg</span>
+<span class="sd">                    img2_right.jpg</span>
+<span class="sd">                    img2_left.disp.jpg</span>
+<span class="sd">                    img2_right.disp.jpg</span>
+<span class="sd">                    ...</span>
+<span class="sd">                shapenet</span>
+<span class="sd">                    img1_left.jpg</span>
+<span class="sd">                    img1_right.jpg</span>
+<span class="sd">                    img1_left.disp.jpg</span>
+<span class="sd">                    img1_right.disp.jpg</span>
+<span class="sd">                    ...</span>
+<span class="sd">                reflective</span>
+<span class="sd">                    img1_left.jpg</span>
+<span class="sd">                    img1_right.jpg</span>
+<span class="sd">                    img1_left.disp.jpg</span>
+<span class="sd">                    img1_right.disp.jpg</span>
+<span class="sd">                    ...</span>
+<span class="sd">                hole</span>
+<span class="sd">                    img1_left.jpg</span>
+<span class="sd">                    img1_right.jpg</span>
+<span class="sd">                    img1_left.disp.jpg</span>
+<span class="sd">                    img1_right.disp.jpg</span>
+<span class="sd">                    ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (str): Root directory of the dataset.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_has_built_in_disparity_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;CREStereo&quot;</span>
+
+        <span class="n">dirs</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;shapenet&quot;</span><span class="p">,</span> <span class="s2">&quot;reflective&quot;</span><span class="p">,</span> <span class="s2">&quot;tree&quot;</span><span class="p">,</span> <span class="s2">&quot;hole&quot;</span><span class="p">]</span>
+
+        <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">dirs</span><span class="p">:</span>
+            <span class="n">left_image_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="s2">&quot;*_left.jpg&quot;</span><span class="p">)</span>
+            <span class="n">right_image_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="s2">&quot;*_right.jpg&quot;</span><span class="p">)</span>
+            <span class="n">imgs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_image_pattern</span><span class="p">,</span> <span class="n">right_image_pattern</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">+=</span> <span class="n">imgs</span>
+
+            <span class="n">left_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="s2">&quot;*_left.disp.png&quot;</span><span class="p">)</span>
+            <span class="n">right_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="s2">&quot;*_right.disp.png&quot;</span><span class="p">)</span>
+            <span class="n">disparities</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_disparity_pattern</span><span class="p">,</span> <span class="n">right_disparity_pattern</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">+=</span> <span class="n">disparities</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="kc">None</span><span class="p">]:</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="c1"># unsqueeze the disparity map into (C, H, W) format</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">disparity_map</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span> <span class="o">/</span> <span class="mf">32.0</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="CREStereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CREStereo.html#torchvision.datasets.CREStereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T1</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img_left, img_right, disparity, valid_mask)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            ``valid_mask`` is implicitly ``None`` if the ``transforms`` parameter does not</span>
+<span class="sd">            generate a valid mask.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T1</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="FallingThingsStereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FallingThingsStereo.html#torchvision.datasets.FallingThingsStereo">[docs]</a><span class="k">class</span> <span class="nc">FallingThingsStereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`FallingThings &lt;https://research.nvidia.com/publication/2018-06_falling-things-synthetic-dataset-3d-object-detection-and-pose-estimation&gt;`_ dataset.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            FallingThings</span>
+<span class="sd">                single</span>
+<span class="sd">                    dir1</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            _object_settings.json</span>
+<span class="sd">                            _camera_settings.json</span>
+<span class="sd">                            image1.left.depth.png</span>
+<span class="sd">                            image1.right.depth.png</span>
+<span class="sd">                            image1.left.jpg</span>
+<span class="sd">                            image1.right.jpg</span>
+<span class="sd">                            image2.left.depth.png</span>
+<span class="sd">                            image2.right.depth.png</span>
+<span class="sd">                            image2.left.jpg</span>
+<span class="sd">                            image2.right</span>
+<span class="sd">                            ...</span>
+<span class="sd">                        scene2</span>
+<span class="sd">                    ...</span>
+<span class="sd">                mixed</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        _object_settings.json</span>
+<span class="sd">                        _camera_settings.json</span>
+<span class="sd">                        image1.left.depth.png</span>
+<span class="sd">                        image1.right.depth.png</span>
+<span class="sd">                        image1.left.jpg</span>
+<span class="sd">                        image1.right.jpg</span>
+<span class="sd">                        image2.left.depth.png</span>
+<span class="sd">                        image2.right.depth.png</span>
+<span class="sd">                        image2.left.jpg</span>
+<span class="sd">                        image2.right</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                    ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where FallingThings is located.</span>
+<span class="sd">        variant (string): Which variant to use. Either &quot;single&quot;, &quot;mixed&quot;, or &quot;both&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">variant</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;single&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;FallingThings&quot;</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">variant</span><span class="p">,</span> <span class="s2">&quot;variant&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;single&quot;</span><span class="p">,</span> <span class="s2">&quot;mixed&quot;</span><span class="p">,</span> <span class="s2">&quot;both&quot;</span><span class="p">))</span>
+
+        <span class="n">variants</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;single&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;single&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;mixed&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;mixed&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;both&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;single&quot;</span><span class="p">,</span> <span class="s2">&quot;mixed&quot;</span><span class="p">],</span>
+        <span class="p">}[</span><span class="n">variant</span><span class="p">]</span>
+
+        <span class="n">split_prefix</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;single&quot;</span><span class="p">:</span> <span class="n">Path</span><span class="p">(</span><span class="s2">&quot;*&quot;</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;mixed&quot;</span><span class="p">:</span> <span class="n">Path</span><span class="p">(</span><span class="s2">&quot;*&quot;</span><span class="p">),</span>
+        <span class="p">}</span>
+
+        <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">variants</span><span class="p">:</span>
+            <span class="n">left_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="n">split_prefix</span><span class="p">[</span><span class="n">s</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;*.left.jpg&quot;</span><span class="p">)</span>
+            <span class="n">right_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="n">split_prefix</span><span class="p">[</span><span class="n">s</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;*.right.jpg&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_img_pattern</span><span class="p">,</span> <span class="n">right_img_pattern</span><span class="p">)</span>
+
+            <span class="n">left_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="n">split_prefix</span><span class="p">[</span><span class="n">s</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;*.left.depth.png&quot;</span><span class="p">)</span>
+            <span class="n">right_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">s</span> <span class="o">/</span> <span class="n">split_prefix</span><span class="p">[</span><span class="n">s</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;*.right.depth.png&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_disparity_pattern</span><span class="p">,</span> <span class="n">right_disparity_pattern</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="kc">None</span><span class="p">]:</span>
+        <span class="c1"># (H, W) image</span>
+        <span class="n">depth</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">))</span>
+        <span class="c1"># as per https://research.nvidia.com/sites/default/files/pubs/2018-06_Falling-Things/readme_0.txt</span>
+        <span class="c1"># in order to extract disparity from depth maps</span>
+        <span class="n">camera_settings_path</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span><span class="o">.</span><span class="n">parent</span> <span class="o">/</span> <span class="s2">&quot;_camera_settings.json&quot;</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">camera_settings_path</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="c1"># inverse of depth-from-disparity equation: depth = (baseline * focal) / (disparity * pixel_constant)</span>
+            <span class="n">intrinsics</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
+            <span class="n">focal</span> <span class="o">=</span> <span class="n">intrinsics</span><span class="p">[</span><span class="s2">&quot;camera_settings&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">][</span><span class="s2">&quot;intrinsic_settings&quot;</span><span class="p">][</span><span class="s2">&quot;fx&quot;</span><span class="p">]</span>
+            <span class="n">baseline</span><span class="p">,</span> <span class="n">pixel_constant</span> <span class="o">=</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">100</span>  <span class="c1"># pixel constant is inverted</span>
+            <span class="n">disparity_map</span> <span class="o">=</span> <span class="p">(</span><span class="n">baseline</span> <span class="o">*</span> <span class="n">focal</span> <span class="o">*</span> <span class="n">pixel_constant</span><span class="p">)</span> <span class="o">/</span> <span class="n">depth</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+            <span class="c1"># unsqueeze disparity to (C, H, W)</span>
+            <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">disparity_map</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span>
+            <span class="n">valid_mask</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="FallingThingsStereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FallingThingsStereo.html#torchvision.datasets.FallingThingsStereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T1</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3-tuple with ``(img_left, img_right, disparity)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            If a ``valid_mask`` is generated within the ``transforms`` parameter,</span>
+<span class="sd">            a 4-tuple with ``(img_left, img_right, disparity, valid_mask)`` is returned.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T1</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="SceneFlowStereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SceneFlowStereo.html#torchvision.datasets.SceneFlowStereo">[docs]</a><span class="k">class</span> <span class="nc">SceneFlowStereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Dataset interface for `Scene Flow &lt;https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html&gt;`_ datasets.</span>
+<span class="sd">    This interface provides access to the `FlyingThings3D, `Monkaa` and `Driving` datasets.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            SceneFlow</span>
+<span class="sd">                Monkaa</span>
+<span class="sd">                    frames_cleanpass</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            left</span>
+<span class="sd">                                img1.png</span>
+<span class="sd">                                img2.png</span>
+<span class="sd">                            right</span>
+<span class="sd">                                img1.png</span>
+<span class="sd">                                img2.png</span>
+<span class="sd">                        scene2</span>
+<span class="sd">                            left</span>
+<span class="sd">                                img1.png</span>
+<span class="sd">                                img2.png</span>
+<span class="sd">                            right</span>
+<span class="sd">                                img1.png</span>
+<span class="sd">                                img2.png</span>
+<span class="sd">                    frames_finalpass</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            left</span>
+<span class="sd">                                img1.png</span>
+<span class="sd">                                img2.png</span>
+<span class="sd">                            right</span>
+<span class="sd">                                img1.png</span>
+<span class="sd">                                img2.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    disparity</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            left</span>
+<span class="sd">                                img1.pfm</span>
+<span class="sd">                                img2.pfm</span>
+<span class="sd">                            right</span>
+<span class="sd">                                img1.pfm</span>
+<span class="sd">                                img2.pfm</span>
+<span class="sd">                FlyingThings3D</span>
+<span class="sd">                    ...</span>
+<span class="sd">                    ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where SceneFlow is located.</span>
+<span class="sd">        variant (string): Which dataset variant to user, &quot;FlyingThings3D&quot; (default), &quot;Monkaa&quot; or &quot;Driving&quot;.</span>
+<span class="sd">        pass_name (string): Which pass to use, &quot;clean&quot; (default), &quot;final&quot; or &quot;both&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">variant</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;FlyingThings3D&quot;</span><span class="p">,</span>
+        <span class="n">pass_name</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;clean&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;SceneFlow&quot;</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">variant</span><span class="p">,</span> <span class="s2">&quot;variant&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;FlyingThings3D&quot;</span><span class="p">,</span> <span class="s2">&quot;Driving&quot;</span><span class="p">,</span> <span class="s2">&quot;Monkaa&quot;</span><span class="p">))</span>
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">pass_name</span><span class="p">,</span> <span class="s2">&quot;pass_name&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;clean&quot;</span><span class="p">,</span> <span class="s2">&quot;final&quot;</span><span class="p">,</span> <span class="s2">&quot;both&quot;</span><span class="p">))</span>
+
+        <span class="n">passes</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;clean&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;frames_cleanpass&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;final&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;frames_finalpass&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;both&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;frames_cleanpass&quot;</span><span class="p">,</span> <span class="s2">&quot;frames_finalpass&quot;</span><span class="p">],</span>
+        <span class="p">}[</span><span class="n">pass_name</span><span class="p">]</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">root</span> <span class="o">/</span> <span class="n">variant</span>
+
+        <span class="n">prefix_directories</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;Monkaa&quot;</span><span class="p">:</span> <span class="n">Path</span><span class="p">(</span><span class="s2">&quot;*&quot;</span><span class="p">),</span>
+            <span class="s2">&quot;FlyingThings3D&quot;</span><span class="p">:</span> <span class="n">Path</span><span class="p">(</span><span class="s2">&quot;*&quot;</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Driving&quot;</span><span class="p">:</span> <span class="n">Path</span><span class="p">(</span><span class="s2">&quot;*&quot;</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span><span class="p">,</span>
+        <span class="p">}</span>
+
+        <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">passes</span><span class="p">:</span>
+            <span class="n">left_image_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">p</span> <span class="o">/</span> <span class="n">prefix_directories</span><span class="p">[</span><span class="n">variant</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;left&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="n">right_image_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">p</span> <span class="o">/</span> <span class="n">prefix_directories</span><span class="p">[</span><span class="n">variant</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;right&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_image_pattern</span><span class="p">,</span> <span class="n">right_image_pattern</span><span class="p">)</span>
+
+            <span class="n">left_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;disparity&quot;</span> <span class="o">/</span> <span class="n">prefix_directories</span><span class="p">[</span><span class="n">variant</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;left&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.pfm&quot;</span><span class="p">)</span>
+            <span class="n">right_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;disparity&quot;</span> <span class="o">/</span> <span class="n">prefix_directories</span><span class="p">[</span><span class="n">variant</span><span class="p">]</span> <span class="o">/</span> <span class="s2">&quot;right&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.pfm&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_disparity_pattern</span><span class="p">,</span> <span class="n">right_disparity_pattern</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="kc">None</span><span class="p">]:</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">_read_pfm_file</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">abs</span><span class="p">(</span><span class="n">disparity_map</span><span class="p">)</span>  <span class="c1"># ensure that the disparity is positive</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="SceneFlowStereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SceneFlowStereo.html#torchvision.datasets.SceneFlowStereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T1</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3-tuple with ``(img_left, img_right, disparity)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            If a ``valid_mask`` is generated within the ``transforms`` parameter,</span>
+<span class="sd">            a 4-tuple with ``(img_left, img_right, disparity, valid_mask)`` is returned.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T1</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="SintelStereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SintelStereo.html#torchvision.datasets.SintelStereo">[docs]</a><span class="k">class</span> <span class="nc">SintelStereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Sintel `Stereo Dataset &lt;http://sintel.is.tue.mpg.de/stereo&gt;`_.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            Sintel</span>
+<span class="sd">                training</span>
+<span class="sd">                    final_left</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            img1.png</span>
+<span class="sd">                            img2.png</span>
+<span class="sd">                            ...</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    final_right</span>
+<span class="sd">                        scene2</span>
+<span class="sd">                            img1.png</span>
+<span class="sd">                            img2.png</span>
+<span class="sd">                            ...</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    disparities</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            img1.png</span>
+<span class="sd">                            img2.png</span>
+<span class="sd">                            ...</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    occlusions</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            img1.png</span>
+<span class="sd">                            img2.png</span>
+<span class="sd">                            ...</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    outofframe</span>
+<span class="sd">                        scene1</span>
+<span class="sd">                            img1.png</span>
+<span class="sd">                            img2.png</span>
+<span class="sd">                            ...</span>
+<span class="sd">                        ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where Sintel Stereo is located.</span>
+<span class="sd">        pass_name (string): The name of the pass to use, either &quot;final&quot;, &quot;clean&quot; or &quot;both&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_has_built_in_disparity_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">pass_name</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;final&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">pass_name</span><span class="p">,</span> <span class="s2">&quot;pass_name&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;final&quot;</span><span class="p">,</span> <span class="s2">&quot;clean&quot;</span><span class="p">,</span> <span class="s2">&quot;both&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;Sintel&quot;</span>
+        <span class="n">pass_names</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;final&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;final&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;clean&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;clean&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;both&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;final&quot;</span><span class="p">,</span> <span class="s2">&quot;clean&quot;</span><span class="p">],</span>
+        <span class="p">}[</span><span class="n">pass_name</span><span class="p">]</span>
+
+        <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">pass_names</span><span class="p">:</span>
+            <span class="n">left_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;training&quot;</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">p</span><span class="si">}</span><span class="s2">_left&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="n">right_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;training&quot;</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">p</span><span class="si">}</span><span class="s2">_right&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_img_pattern</span><span class="p">,</span> <span class="n">right_img_pattern</span><span class="p">)</span>
+
+            <span class="n">disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;training&quot;</span> <span class="o">/</span> <span class="s2">&quot;disparities&quot;</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;*.png&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">disparity_pattern</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_occlussion_mask_paths</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]:</span>
+        <span class="c1"># helper function to get the occlusion mask paths</span>
+        <span class="c1"># a path will look like  .../.../.../training/disparities/scene1/img1.png</span>
+        <span class="c1"># we want to get something like .../.../.../training/occlusions/scene1/img1.png</span>
+        <span class="n">fpath</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+        <span class="n">basename</span> <span class="o">=</span> <span class="n">fpath</span><span class="o">.</span><span class="n">name</span>
+        <span class="n">scenedir</span> <span class="o">=</span> <span class="n">fpath</span><span class="o">.</span><span class="n">parent</span>
+        <span class="c1"># the parent of the scenedir is actually the disparity dir</span>
+        <span class="n">sampledir</span> <span class="o">=</span> <span class="n">scenedir</span><span class="o">.</span><span class="n">parent</span><span class="o">.</span><span class="n">parent</span>
+
+        <span class="n">occlusion_path</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">sampledir</span> <span class="o">/</span> <span class="s2">&quot;occlusions&quot;</span> <span class="o">/</span> <span class="n">scenedir</span><span class="o">.</span><span class="n">name</span> <span class="o">/</span> <span class="n">basename</span><span class="p">)</span>
+        <span class="n">outofframe_path</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">sampledir</span> <span class="o">/</span> <span class="s2">&quot;outofframe&quot;</span> <span class="o">/</span> <span class="n">scenedir</span><span class="o">.</span><span class="n">name</span> <span class="o">/</span> <span class="n">basename</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">occlusion_path</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Occlusion mask </span><span class="si">{</span><span class="n">occlusion_path</span><span class="si">}</span><span class="s2"> does not exist&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">outofframe_path</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Out of frame mask </span><span class="si">{</span><span class="n">outofframe_path</span><span class="si">}</span><span class="s2"> does not exist&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">occlusion_path</span><span class="p">,</span> <span class="n">outofframe_path</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">],</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]:</span>
+        <span class="k">if</span> <span class="n">file_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="c1"># disparity decoding as per Sintel instructions in the README provided with the dataset</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="n">r</span><span class="p">,</span> <span class="n">g</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">disparity_map</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">axis</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">r</span> <span class="o">*</span> <span class="mi">4</span> <span class="o">+</span> <span class="n">g</span> <span class="o">/</span> <span class="p">(</span><span class="mi">2</span><span class="o">**</span><span class="mi">6</span><span class="p">)</span> <span class="o">+</span> <span class="n">b</span> <span class="o">/</span> <span class="p">(</span><span class="mi">2</span><span class="o">**</span><span class="mi">14</span><span class="p">)</span>
+        <span class="c1"># reshape into (C, H, W) format</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">disparity_map</span><span class="p">,</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="c1"># find the appropriate file paths</span>
+        <span class="n">occlued_mask_path</span><span class="p">,</span> <span class="n">out_of_frame_mask_path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_occlussion_mask_paths</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+        <span class="c1"># occlusion masks</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">occlued_mask_path</span><span class="p">))</span> <span class="o">==</span> <span class="mi">0</span>
+        <span class="c1"># out of frame masks</span>
+        <span class="n">off_mask</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">out_of_frame_mask_path</span><span class="p">))</span> <span class="o">==</span> <span class="mi">0</span>
+        <span class="c1"># combine the masks together</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">logical_and</span><span class="p">(</span><span class="n">off_mask</span><span class="p">,</span> <span class="n">valid_mask</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="SintelStereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SintelStereo.html#torchvision.datasets.SintelStereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T2</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img_left, img_right, disparity, valid_mask)`` is returned.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images whilst</span>
+<span class="sd">            the valid_mask is a numpy array of shape (H, W).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T2</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="InStereo2k"><a class="viewcode-back" href="../../../generated/torchvision.datasets.InStereo2k.html#torchvision.datasets.InStereo2k">[docs]</a><span class="k">class</span> <span class="nc">InStereo2k</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`InStereo2k &lt;https://github.com/YuhuaXu/StereoDataset&gt;`_ dataset.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            InStereo2k</span>
+<span class="sd">                train</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        left.png</span>
+<span class="sd">                        right.png</span>
+<span class="sd">                        left_disp.png</span>
+<span class="sd">                        right_disp.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                    ...</span>
+<span class="sd">                test</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        left.png</span>
+<span class="sd">                        right.png</span>
+<span class="sd">                        left_disp.png</span>
+<span class="sd">                        right_disp.png</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                    ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where InStereo2k is located.</span>
+<span class="sd">        split (string): Either &quot;train&quot; or &quot;test&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;InStereo2k&quot;</span> <span class="o">/</span> <span class="n">split</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="n">left_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;left.png&quot;</span><span class="p">)</span>
+        <span class="n">right_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;right.png&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_img_pattern</span><span class="p">,</span> <span class="n">right_img_pattern</span><span class="p">)</span>
+
+        <span class="n">left_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;left_disp.png&quot;</span><span class="p">)</span>
+        <span class="n">right_disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;right_disp.png&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_disparity_pattern</span><span class="p">,</span> <span class="n">right_disparity_pattern</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="kc">None</span><span class="p">]:</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="c1"># unsqueeze disparity to (C, H, W)</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">disparity_map</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span> <span class="o">/</span> <span class="mf">1024.0</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="InStereo2k.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.InStereo2k.html#torchvision.datasets.InStereo2k.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T1</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 3-tuple with ``(img_left, img_right, disparity)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            If a ``valid_mask`` is generated within the ``transforms`` parameter,</span>
+<span class="sd">            a 4-tuple with ``(img_left, img_right, disparity, valid_mask)`` is returned.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T1</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+
+
+<div class="viewcode-block" id="ETH3DStereo"><a class="viewcode-back" href="../../../generated/torchvision.datasets.ETH3DStereo.html#torchvision.datasets.ETH3DStereo">[docs]</a><span class="k">class</span> <span class="nc">ETH3DStereo</span><span class="p">(</span><span class="n">StereoMatchingDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ETH3D `Low-Res Two-View &lt;https://www.eth3d.net/datasets&gt;`_ dataset.</span>
+
+<span class="sd">    The dataset is expected to have the following structure: ::</span>
+
+<span class="sd">        root</span>
+<span class="sd">            ETH3D</span>
+<span class="sd">                two_view_training</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        im1.png</span>
+<span class="sd">                        im0.png</span>
+<span class="sd">                        images.txt</span>
+<span class="sd">                        cameras.txt</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                        im1.png</span>
+<span class="sd">                        im0.png</span>
+<span class="sd">                        images.txt</span>
+<span class="sd">                        cameras.txt</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                    ...</span>
+<span class="sd">                two_view_training_gt</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        disp0GT.pfm</span>
+<span class="sd">                        mask0nocc.png</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                        disp0GT.pfm</span>
+<span class="sd">                        mask0nocc.png</span>
+<span class="sd">                    ...</span>
+<span class="sd">                two_view_testing</span>
+<span class="sd">                    scene1</span>
+<span class="sd">                        im1.png</span>
+<span class="sd">                        im0.png</span>
+<span class="sd">                        images.txt</span>
+<span class="sd">                        cameras.txt</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                    scene2</span>
+<span class="sd">                        im1.png</span>
+<span class="sd">                        im0.png</span>
+<span class="sd">                        images.txt</span>
+<span class="sd">                        cameras.txt</span>
+<span class="sd">                        calib.txt</span>
+<span class="sd">                    ...</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the ETH3D Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split of scenes, either &quot;train&quot; (default) or &quot;test&quot;.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes in a sample and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_has_built_in_disparity_mask</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;ETH3D&quot;</span>
+
+        <span class="n">img_dir</span> <span class="o">=</span> <span class="s2">&quot;two_view_training&quot;</span> <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span> <span class="k">else</span> <span class="s2">&quot;two_view_test&quot;</span>
+        <span class="n">anot_dir</span> <span class="o">=</span> <span class="s2">&quot;two_view_training_gt&quot;</span>
+
+        <span class="n">left_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">img_dir</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;im0.png&quot;</span><span class="p">)</span>
+        <span class="n">right_img_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">img_dir</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;im1.png&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">left_img_pattern</span><span class="p">,</span> <span class="n">right_img_pattern</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="nb">list</span><span class="p">((</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">disparity_pattern</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span> <span class="o">/</span> <span class="n">anot_dir</span> <span class="o">/</span> <span class="s2">&quot;*&quot;</span> <span class="o">/</span> <span class="s2">&quot;disp0GT.pfm&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_disparities</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scan_pairs</span><span class="p">(</span><span class="n">disparity_pattern</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_read_disparity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">],</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]:</span>
+        <span class="c1"># test split has no disparity maps</span>
+        <span class="k">if</span> <span class="n">file_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">_read_pfm_file</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span>
+        <span class="n">disparity_map</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">abs</span><span class="p">(</span><span class="n">disparity_map</span><span class="p">)</span>  <span class="c1"># ensure that the disparity is positive</span>
+        <span class="n">mask_path</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span><span class="o">.</span><span class="n">parent</span> <span class="o">/</span> <span class="s2">&quot;mask0nocc.png&quot;</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">mask_path</span><span class="p">)</span>
+        <span class="n">valid_mask</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">valid_mask</span><span class="p">)</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="nb">bool</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">disparity_map</span><span class="p">,</span> <span class="n">valid_mask</span>
+
+<div class="viewcode-block" id="ETH3DStereo.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.ETH3DStereo.html#torchvision.datasets.ETH3DStereo.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">T2</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return example at given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index(int): The index of the example to retrieve</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: A 4-tuple with ``(img_left, img_right, disparity, valid_mask)``.</span>
+<span class="sd">            The disparity is a numpy array of shape (1, H, W) and the images are PIL images.</span>
+<span class="sd">            ``valid_mask`` is implicitly ``None`` if the ``transforms`` parameter does not</span>
+<span class="sd">            generate a valid mask.</span>
+<span class="sd">            Both ``disparity`` and ``valid_mask`` are ``None`` if the dataset split is test.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">cast</span><span class="p">(</span><span class="n">T2</span><span class="p">,</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">index</span><span class="p">))</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/caltech.html b/0.16/_modules/torchvision/datasets/caltech.html
new file mode 100644
index 00000000000..28d1f009775
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/caltech.html
@@ -0,0 +1,906 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.caltech &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.caltech</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.caltech</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="Caltech101"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Caltech101.html#torchvision.datasets.Caltech101">[docs]</a><span class="k">class</span> <span class="nc">Caltech101</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Caltech 101 &lt;https://data.caltech.edu/records/20086&gt;`_ Dataset.</span>
+
+<span class="sd">    .. warning::</span>
+
+<span class="sd">        This class needs `scipy &lt;https://docs.scipy.org/doc/&gt;`_ to load target files from `.mat` format.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``caltech101`` exists or will be saved to if download is set to True.</span>
+<span class="sd">        target_type (string or list, optional): Type of target to use, ``category`` or</span>
+<span class="sd">            ``annotation``. Can also be a list to output a tuple with all specified</span>
+<span class="sd">            target types.  ``category`` represents the target class, and</span>
+<span class="sd">            ``annotation`` is a list of points from a hand-generated outline.</span>
+<span class="sd">            Defaults to ``category``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">target_type</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;category&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;caltech101&quot;</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">target_type</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="n">target_type</span> <span class="o">=</span> <span class="p">[</span><span class="n">target_type</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="o">=</span> <span class="p">[</span><span class="n">verify_str_arg</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="s2">&quot;target_type&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;category&quot;</span><span class="p">,</span> <span class="s2">&quot;annotation&quot;</span><span class="p">))</span> <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">target_type</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">categories</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;101_ObjectCategories&quot;</span><span class="p">)))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">categories</span><span class="o">.</span><span class="n">remove</span><span class="p">(</span><span class="s2">&quot;BACKGROUND_Google&quot;</span><span class="p">)</span>  <span class="c1"># this is not a real class</span>
+
+        <span class="c1"># For some reason, the category names in &quot;101_ObjectCategories&quot; and</span>
+        <span class="c1"># &quot;Annotations&quot; do not always match. This is a manual map between the</span>
+        <span class="c1"># two. Defaults to using same name, since most names are fine.</span>
+        <span class="n">name_map</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;Faces&quot;</span><span class="p">:</span> <span class="s2">&quot;Faces_2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Faces_easy&quot;</span><span class="p">:</span> <span class="s2">&quot;Faces_3&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Motorbikes&quot;</span><span class="p">:</span> <span class="s2">&quot;Motorbikes_16&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;airplanes&quot;</span><span class="p">:</span> <span class="s2">&quot;Airplanes_Side_2&quot;</span><span class="p">,</span>
+        <span class="p">}</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">annotation_categories</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">name_map</span><span class="p">[</span><span class="n">x</span><span class="p">]</span> <span class="k">if</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">name_map</span> <span class="k">else</span> <span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">categories</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">y</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">categories</span><span class="p">):</span>
+            <span class="n">n</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;101_ObjectCategories&quot;</span><span class="p">,</span> <span class="n">c</span><span class="p">)))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">n</span> <span class="o">+</span> <span class="mi">1</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">n</span> <span class="o">*</span> <span class="p">[</span><span class="n">i</span><span class="p">])</span>
+
+<div class="viewcode-block" id="Caltech101.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Caltech101.html#torchvision.datasets.Caltech101.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where the type of target specified by target_type.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="kn">import</span> <span class="nn">scipy.io</span>
+
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span>
+                <span class="s2">&quot;101_ObjectCategories&quot;</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">categories</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="p">[</span><span class="n">index</span><span class="p">]],</span>
+                <span class="sa">f</span><span class="s2">&quot;image_</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">[</span><span class="n">index</span><span class="p">]</span><span class="si">:</span><span class="s2">04d</span><span class="si">}</span><span class="s2">.jpg&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="n">target</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;category&quot;</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+            <span class="k">elif</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;annotation&quot;</span><span class="p">:</span>
+                <span class="n">data</span> <span class="o">=</span> <span class="n">scipy</span><span class="o">.</span><span class="n">io</span><span class="o">.</span><span class="n">loadmat</span><span class="p">(</span>
+                    <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span>
+                        <span class="s2">&quot;Annotations&quot;</span><span class="p">,</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">annotation_categories</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="p">[</span><span class="n">index</span><span class="p">]],</span>
+                        <span class="sa">f</span><span class="s2">&quot;annotation_</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">[</span><span class="n">index</span><span class="p">]</span><span class="si">:</span><span class="s2">04d</span><span class="si">}</span><span class="s2">.mat&quot;</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="p">)</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">data</span><span class="p">[</span><span class="s2">&quot;obj_contour&quot;</span><span class="p">])</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">target</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">target</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">target</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="c1"># can be more robust and check hash of files</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;101_ObjectCategories&quot;</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="s2">&quot;https://drive.google.com/file/d/137RyRjvTBkBiIfeYBNZBtViDHQ6_Ewsp&quot;</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span>
+            <span class="n">filename</span><span class="o">=</span><span class="s2">&quot;101_ObjectCategories.tar.gz&quot;</span><span class="p">,</span>
+            <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;b224c7392d521a49829488ab0f1120d9&quot;</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="s2">&quot;https://drive.google.com/file/d/175kQy3UsZ0wUEHZjqkUDdNVssr7bgh_m&quot;</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span>
+            <span class="n">filename</span><span class="o">=</span><span class="s2">&quot;Annotations.tar&quot;</span><span class="p">,</span>
+            <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;6f83eeb1f24d99cab4eb377263132c91&quot;</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;Target type: </span><span class="si">{target_type}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Caltech256"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Caltech256.html#torchvision.datasets.Caltech256">[docs]</a><span class="k">class</span> <span class="nc">Caltech256</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Caltech 256 &lt;https://data.caltech.edu/records/20087&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``caltech256`` exists or will be saved to if download is set to True.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;caltech256&quot;</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">categories</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;256_ObjectCategories&quot;</span><span class="p">)))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">y</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">categories</span><span class="p">):</span>
+            <span class="n">n</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span>
+                <span class="p">[</span>
+                    <span class="n">item</span>
+                    <span class="k">for</span> <span class="n">item</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;256_ObjectCategories&quot;</span><span class="p">,</span> <span class="n">c</span><span class="p">))</span>
+                    <span class="k">if</span> <span class="n">item</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;.jpg&quot;</span><span class="p">)</span>
+                <span class="p">]</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">n</span> <span class="o">+</span> <span class="mi">1</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">n</span> <span class="o">*</span> <span class="p">[</span><span class="n">i</span><span class="p">])</span>
+
+<div class="viewcode-block" id="Caltech256.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Caltech256.html#torchvision.datasets.Caltech256.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span>
+                <span class="s2">&quot;256_ObjectCategories&quot;</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">categories</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="p">[</span><span class="n">index</span><span class="p">]],</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="p">[</span><span class="n">index</span><span class="p">]</span><span class="w"> </span><span class="o">+</span><span class="w"> </span><span class="mi">1</span><span class="si">:</span><span class="s2">03d</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">[</span><span class="n">index</span><span class="p">]</span><span class="si">:</span><span class="s2">04d</span><span class="si">}</span><span class="s2">.jpg&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">y</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="c1"># can be more robust and check hash of files</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;256_ObjectCategories&quot;</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="s2">&quot;https://drive.google.com/file/d/1r6o0pSROcV1_VwT4oSjA2FBUSCWGuxLK&quot;</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span>
+            <span class="n">filename</span><span class="o">=</span><span class="s2">&quot;256_ObjectCategories.tar&quot;</span><span class="p">,</span>
+            <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;67b4f42ca05d46448c6bb8ecd2220f6d&quot;</span><span class="p">,</span>
+        <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/celeba.html b/0.16/_modules/torchvision/datasets/celeba.html
new file mode 100644
index 00000000000..112779d3fe0
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/celeba.html
@@ -0,0 +1,858 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.celeba &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.celeba</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.celeba</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">csv</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">namedtuple</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">PIL</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_file_from_google_drive</span><span class="p">,</span> <span class="n">extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+<span class="n">CSV</span> <span class="o">=</span> <span class="n">namedtuple</span><span class="p">(</span><span class="s2">&quot;CSV&quot;</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;header&quot;</span><span class="p">,</span> <span class="s2">&quot;index&quot;</span><span class="p">,</span> <span class="s2">&quot;data&quot;</span><span class="p">])</span>
+
+
+<div class="viewcode-block" id="CelebA"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CelebA.html#torchvision.datasets.CelebA">[docs]</a><span class="k">class</span> <span class="nc">CelebA</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Large-scale CelebFaces Attributes (CelebA) Dataset &lt;http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images are downloaded to.</span>
+<span class="sd">        split (string): One of {&#39;train&#39;, &#39;valid&#39;, &#39;test&#39;, &#39;all&#39;}.</span>
+<span class="sd">            Accordingly dataset is selected.</span>
+<span class="sd">        target_type (string or list, optional): Type of target to use, ``attr``, ``identity``, ``bbox``,</span>
+<span class="sd">            or ``landmarks``. Can also be a list to output a tuple with all specified target types.</span>
+<span class="sd">            The targets represent:</span>
+
+<span class="sd">                - ``attr`` (Tensor shape=(40,) dtype=int): binary (0, 1) labels for attributes</span>
+<span class="sd">                - ``identity`` (int): label for each person (data points with the same identity are the same person)</span>
+<span class="sd">                - ``bbox`` (Tensor shape=(4,) dtype=int): bounding box (x, y, width, height)</span>
+<span class="sd">                - ``landmarks`` (Tensor shape=(10,) dtype=int): landmark points (lefteye_x, lefteye_y, righteye_x,</span>
+<span class="sd">                  righteye_y, nose_x, nose_y, leftmouth_x, leftmouth_y, rightmouth_x, rightmouth_y)</span>
+
+<span class="sd">            Defaults to ``attr``. If empty, ``None`` will be returned as target.</span>
+
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.PILToTensor``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">base_folder</span> <span class="o">=</span> <span class="s2">&quot;celeba&quot;</span>
+    <span class="c1"># There currently does not appear to be an easy way to extract 7z in python (without introducing additional</span>
+    <span class="c1"># dependencies). The &quot;in-the-wild&quot; (not aligned+cropped) images are only in 7z, so they are not available</span>
+    <span class="c1"># right now.</span>
+    <span class="n">file_list</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="c1"># File ID                                      MD5 Hash                            Filename</span>
+        <span class="p">(</span><span class="s2">&quot;0B7EVK8r0v71pZjFTYXZWM3FlRnM&quot;</span><span class="p">,</span> <span class="s2">&quot;00d2c5bc6d35e252742224ab0c1e8fcb&quot;</span><span class="p">,</span> <span class="s2">&quot;img_align_celeba.zip&quot;</span><span class="p">),</span>
+        <span class="c1"># (&quot;0B7EVK8r0v71pbWNEUjJKdDQ3dGc&quot;,&quot;b6cd7e93bc7a96c2dc33f819aa3ac651&quot;, &quot;img_align_celeba_png.7z&quot;),</span>
+        <span class="c1"># (&quot;0B7EVK8r0v71peklHb0pGdDl6R28&quot;, &quot;b6cd7e93bc7a96c2dc33f819aa3ac651&quot;, &quot;img_celeba.7z&quot;),</span>
+        <span class="p">(</span><span class="s2">&quot;0B7EVK8r0v71pblRyaVFSWGxPY0U&quot;</span><span class="p">,</span> <span class="s2">&quot;75e246fa4810816ffd6ee81facbd244c&quot;</span><span class="p">,</span> <span class="s2">&quot;list_attr_celeba.txt&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;1_ee_0u7vcNLOfNLegJRHmolfH5ICW-XS&quot;</span><span class="p">,</span> <span class="s2">&quot;32bd1bd63d3c78cd57e08160ec5ed1e2&quot;</span><span class="p">,</span> <span class="s2">&quot;identity_CelebA.txt&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;0B7EVK8r0v71pbThiMVRxWXZ4dU0&quot;</span><span class="p">,</span> <span class="s2">&quot;00566efa6fedff7a56946cd1c10f1c16&quot;</span><span class="p">,</span> <span class="s2">&quot;list_bbox_celeba.txt&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;0B7EVK8r0v71pd0FJY3Blby1HUTQ&quot;</span><span class="p">,</span> <span class="s2">&quot;cc24ecafdb5b50baae59b03474781f8c&quot;</span><span class="p">,</span> <span class="s2">&quot;list_landmarks_align_celeba.txt&quot;</span><span class="p">),</span>
+        <span class="c1"># (&quot;0B7EVK8r0v71pTzJIdlJWdHczRlU&quot;, &quot;063ee6ddb681f96bc9ca28c6febb9d1a&quot;, &quot;list_landmarks_celeba.txt&quot;),</span>
+        <span class="p">(</span><span class="s2">&quot;0B7EVK8r0v71pY0NSMzRuSXJEVkk&quot;</span><span class="p">,</span> <span class="s2">&quot;d32c9cbf5e040fd4025c592c306e6668&quot;</span><span class="p">,</span> <span class="s2">&quot;list_eval_partition.txt&quot;</span><span class="p">),</span>
+    <span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">target_type</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;attr&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">split</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">target_type</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="o">=</span> <span class="n">target_type</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="o">=</span> <span class="p">[</span><span class="n">target_type</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;target_transform is specified but target_type is empty&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="n">split_map</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span>
+            <span class="s2">&quot;valid&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
+            <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="mi">2</span><span class="p">,</span>
+            <span class="s2">&quot;all&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="p">}</span>
+        <span class="n">split_</span> <span class="o">=</span> <span class="n">split_map</span><span class="p">[</span><span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="o">.</span><span class="n">lower</span><span class="p">(),</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;valid&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;all&quot;</span><span class="p">))]</span>
+        <span class="n">splits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_csv</span><span class="p">(</span><span class="s2">&quot;list_eval_partition.txt&quot;</span><span class="p">)</span>
+        <span class="n">identity</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_csv</span><span class="p">(</span><span class="s2">&quot;identity_CelebA.txt&quot;</span><span class="p">)</span>
+        <span class="n">bbox</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_csv</span><span class="p">(</span><span class="s2">&quot;list_bbox_celeba.txt&quot;</span><span class="p">,</span> <span class="n">header</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">landmarks_align</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_csv</span><span class="p">(</span><span class="s2">&quot;list_landmarks_align_celeba.txt&quot;</span><span class="p">,</span> <span class="n">header</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">attr</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_csv</span><span class="p">(</span><span class="s2">&quot;list_attr_celeba.txt&quot;</span><span class="p">,</span> <span class="n">header</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">mask</span> <span class="o">=</span> <span class="nb">slice</span><span class="p">(</span><span class="kc">None</span><span class="p">)</span> <span class="k">if</span> <span class="n">split_</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="p">(</span><span class="n">splits</span><span class="o">.</span><span class="n">data</span> <span class="o">==</span> <span class="n">split_</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">mask</span> <span class="o">==</span> <span class="nb">slice</span><span class="p">(</span><span class="kc">None</span><span class="p">):</span>  <span class="c1"># if split == &quot;all&quot;</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">filename</span> <span class="o">=</span> <span class="n">splits</span><span class="o">.</span><span class="n">index</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">filename</span> <span class="o">=</span> <span class="p">[</span><span class="n">splits</span><span class="o">.</span><span class="n">index</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">torch</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nonzero</span><span class="p">(</span><span class="n">mask</span><span class="p">))]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">identity</span> <span class="o">=</span> <span class="n">identity</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">mask</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bbox</span> <span class="o">=</span> <span class="n">bbox</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">mask</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">landmarks_align</span> <span class="o">=</span> <span class="n">landmarks_align</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">mask</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">attr</span> <span class="o">=</span> <span class="n">attr</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">mask</span><span class="p">]</span>
+        <span class="c1"># map from {-1, 1} to {0, 1}</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">attr</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">div</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">attr</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="n">rounding_mode</span><span class="o">=</span><span class="s2">&quot;floor&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">attr_names</span> <span class="o">=</span> <span class="n">attr</span><span class="o">.</span><span class="n">header</span>
+
+    <span class="k">def</span> <span class="nf">_load_csv</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">header</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">CSV</span><span class="p">:</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="n">filename</span><span class="p">))</span> <span class="k">as</span> <span class="n">csv_file</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">csv</span><span class="o">.</span><span class="n">reader</span><span class="p">(</span><span class="n">csv_file</span><span class="p">,</span> <span class="n">delimiter</span><span class="o">=</span><span class="s2">&quot; &quot;</span><span class="p">,</span> <span class="n">skipinitialspace</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="n">header</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">headers</span> <span class="o">=</span> <span class="n">data</span><span class="p">[</span><span class="n">header</span><span class="p">]</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">data</span><span class="p">[</span><span class="n">header</span> <span class="o">+</span> <span class="mi">1</span> <span class="p">:]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">headers</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="n">indices</span> <span class="o">=</span> <span class="p">[</span><span class="n">row</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">data</span><span class="p">]</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="p">[</span><span class="n">row</span><span class="p">[</span><span class="mi">1</span><span class="p">:]</span> <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">data</span><span class="p">]</span>
+        <span class="n">data_int</span> <span class="o">=</span> <span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="n">i</span><span class="p">))</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">data</span><span class="p">]</span>
+
+        <span class="k">return</span> <span class="n">CSV</span><span class="p">(</span><span class="n">headers</span><span class="p">,</span> <span class="n">indices</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">data_int</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">for</span> <span class="p">(</span><span class="n">_</span><span class="p">,</span> <span class="n">md5</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">file_list</span><span class="p">:</span>
+            <span class="n">fpath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span>
+            <span class="n">_</span><span class="p">,</span> <span class="n">ext</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">splitext</span><span class="p">(</span><span class="n">filename</span><span class="p">)</span>
+            <span class="c1"># Allow original archive to be deleted (zip and 7z)</span>
+            <span class="c1"># Only need the extracted images</span>
+            <span class="k">if</span> <span class="n">ext</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;.zip&quot;</span><span class="p">,</span> <span class="s2">&quot;.7z&quot;</span><span class="p">]</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">fpath</span><span class="p">,</span> <span class="n">md5</span><span class="p">):</span>
+                <span class="k">return</span> <span class="kc">False</span>
+
+        <span class="c1"># Should check a hash of the images</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="s2">&quot;img_align_celeba&quot;</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="k">for</span> <span class="p">(</span><span class="n">file_id</span><span class="p">,</span> <span class="n">md5</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">file_list</span><span class="p">:</span>
+            <span class="n">download_file_from_google_drive</span><span class="p">(</span><span class="n">file_id</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">),</span> <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+
+        <span class="n">extract_archive</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="s2">&quot;img_align_celeba.zip&quot;</span><span class="p">))</span>
+
+<div class="viewcode-block" id="CelebA.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CelebA.html#torchvision.datasets.CelebA.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">X</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="s2">&quot;img_align_celeba&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">[</span><span class="n">index</span><span class="p">]))</span>
+
+        <span class="n">target</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;attr&quot;</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">attr</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="p">:])</span>
+            <span class="k">elif</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;identity&quot;</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">identity</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="mi">0</span><span class="p">])</span>
+            <span class="k">elif</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;bbox&quot;</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">bbox</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="p">:])</span>
+            <span class="k">elif</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;landmarks&quot;</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">landmarks_align</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="p">:])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="c1"># TODO: refactor with utils.verify_str_arg</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Target type &quot;</span><span class="si">{</span><span class="n">t</span><span class="si">}</span><span class="s1">&quot; is not recognized.&#39;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">X</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">X</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">target</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">target</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">target</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">target</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="k">return</span> <span class="n">X</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">attr</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">lines</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Target type: </span><span class="si">{target_type}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;Split: </span><span class="si">{split}</span><span class="s2">&quot;</span><span class="p">]</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">lines</span><span class="p">)</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/cifar.html b/0.16/_modules/torchvision/datasets/cifar.html
new file mode 100644
index 00000000000..fd991ca9ad2
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/cifar.html
@@ -0,0 +1,836 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.cifar &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.cifar</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.cifar</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">import</span> <span class="nn">pickle</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="CIFAR10"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CIFAR10.html#torchvision.datasets.CIFAR10">[docs]</a><span class="k">class</span> <span class="nc">CIFAR10</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`CIFAR10 &lt;https://www.cs.toronto.edu/~kriz/cifar.html&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``cifar-10-batches-py`` exists or will be saved to if download is set to True.</span>
+<span class="sd">        train (bool, optional): If True, creates dataset from training set, otherwise</span>
+<span class="sd">            creates from test set.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">base_folder</span> <span class="o">=</span> <span class="s2">&quot;cifar-10-batches-py&quot;</span>
+    <span class="n">url</span> <span class="o">=</span> <span class="s2">&quot;https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz&quot;</span>
+    <span class="n">filename</span> <span class="o">=</span> <span class="s2">&quot;cifar-10-python.tar.gz&quot;</span>
+    <span class="n">tgz_md5</span> <span class="o">=</span> <span class="s2">&quot;c58f30108f718f92721af3b95e74349a&quot;</span>
+    <span class="n">train_list</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">[</span><span class="s2">&quot;data_batch_1&quot;</span><span class="p">,</span> <span class="s2">&quot;c99cafc152244af753f735de768cd75f&quot;</span><span class="p">],</span>
+        <span class="p">[</span><span class="s2">&quot;data_batch_2&quot;</span><span class="p">,</span> <span class="s2">&quot;d4bba439e000b95fd0a9bffe97cbabec&quot;</span><span class="p">],</span>
+        <span class="p">[</span><span class="s2">&quot;data_batch_3&quot;</span><span class="p">,</span> <span class="s2">&quot;54ebc095f3ab1f0389bbae665268c751&quot;</span><span class="p">],</span>
+        <span class="p">[</span><span class="s2">&quot;data_batch_4&quot;</span><span class="p">,</span> <span class="s2">&quot;634d18415352ddfa80567beed471001a&quot;</span><span class="p">],</span>
+        <span class="p">[</span><span class="s2">&quot;data_batch_5&quot;</span><span class="p">,</span> <span class="s2">&quot;482c414d41f54cd18b22e5b47cb7c3cb&quot;</span><span class="p">],</span>
+    <span class="p">]</span>
+
+    <span class="n">test_list</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">[</span><span class="s2">&quot;test_batch&quot;</span><span class="p">,</span> <span class="s2">&quot;40351d587109b95175f43aff81a1287e&quot;</span><span class="p">],</span>
+    <span class="p">]</span>
+    <span class="n">meta</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;batches.meta&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;key&quot;</span><span class="p">:</span> <span class="s2">&quot;label_names&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;5ff9c542aee3614f3951f8cda6e48888&quot;</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="o">=</span> <span class="n">train</span>  <span class="c1"># training set or test set</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="p">:</span>
+            <span class="n">downloaded_list</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_list</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">downloaded_list</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">test_list</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="c1"># now load the picked numpy arrays</span>
+        <span class="k">for</span> <span class="n">file_name</span><span class="p">,</span> <span class="n">checksum</span> <span class="ow">in</span> <span class="n">downloaded_list</span><span class="p">:</span>
+            <span class="n">file_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="n">file_name</span><span class="p">)</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+                <span class="n">entry</span> <span class="o">=</span> <span class="n">pickle</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s2">&quot;latin1&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">entry</span><span class="p">[</span><span class="s2">&quot;data&quot;</span><span class="p">])</span>
+                <span class="k">if</span> <span class="s2">&quot;labels&quot;</span> <span class="ow">in</span> <span class="n">entry</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">entry</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">])</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">entry</span><span class="p">[</span><span class="s2">&quot;fine_labels&quot;</span><span class="p">])</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">vstack</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">transpose</span><span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>  <span class="c1"># convert to HWC</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_load_meta</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_load_meta</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;filename&quot;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;md5&quot;</span><span class="p">]):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset metadata file not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">infile</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">pickle</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">infile</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s2">&quot;latin1&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="n">data</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;key&quot;</span><span class="p">]]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="n">_class</span><span class="p">:</span> <span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">_class</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">)}</span>
+
+<div class="viewcode-block" id="CIFAR10.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CIFAR10.html#torchvision.datasets.CIFAR10.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+        <span class="c1"># doing this so that it is consistent with all other datasets</span>
+        <span class="c1"># to return a PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_list</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">test_list</span><span class="p">:</span>
+            <span class="n">fpath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">fpath</span><span class="p">,</span> <span class="n">md5</span><span class="p">):</span>
+                <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tgz_md5</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">split</span> <span class="o">=</span> <span class="s2">&quot;Train&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="ow">is</span> <span class="kc">True</span> <span class="k">else</span> <span class="s2">&quot;Test&quot;</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;Split: </span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">&quot;</span></div>
+
+
+<div class="viewcode-block" id="CIFAR100"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CIFAR100.html#torchvision.datasets.CIFAR100">[docs]</a><span class="k">class</span> <span class="nc">CIFAR100</span><span class="p">(</span><span class="n">CIFAR10</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`CIFAR100 &lt;https://www.cs.toronto.edu/~kriz/cifar.html&gt;`_ Dataset.</span>
+
+<span class="sd">    This is a subclass of the `CIFAR10` Dataset.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">base_folder</span> <span class="o">=</span> <span class="s2">&quot;cifar-100-python&quot;</span>
+    <span class="n">url</span> <span class="o">=</span> <span class="s2">&quot;https://www.cs.toronto.edu/~kriz/cifar-100-python.tar.gz&quot;</span>
+    <span class="n">filename</span> <span class="o">=</span> <span class="s2">&quot;cifar-100-python.tar.gz&quot;</span>
+    <span class="n">tgz_md5</span> <span class="o">=</span> <span class="s2">&quot;eb9058c3a382ffc7106e4002c42a8d85&quot;</span>
+    <span class="n">train_list</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;16019d7e3df5f24257cddd939b257f8d&quot;</span><span class="p">],</span>
+    <span class="p">]</span>
+
+    <span class="n">test_list</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">[</span><span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;f0ef6b0ae62326f3e7ffdfab6717acfc&quot;</span><span class="p">],</span>
+    <span class="p">]</span>
+    <span class="n">meta</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;meta&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;key&quot;</span><span class="p">:</span> <span class="s2">&quot;fine_label_names&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;7973b15100ade9c7d40fb424638fde48&quot;</span><span class="p">,</span>
+    <span class="p">}</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/cityscapes.html b/0.16/_modules/torchvision/datasets/cityscapes.html
new file mode 100644
index 00000000000..2c7a4dfbe46
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/cityscapes.html
@@ -0,0 +1,890 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.cityscapes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.cityscapes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.cityscapes</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">json</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">namedtuple</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">extract_archive</span><span class="p">,</span> <span class="n">iterable_to_str</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="Cityscapes"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Cityscapes.html#torchvision.datasets.Cityscapes">[docs]</a><span class="k">class</span> <span class="nc">Cityscapes</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Cityscapes &lt;http://www.cityscapes-dataset.com/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory ``leftImg8bit``</span>
+<span class="sd">            and ``gtFine`` or ``gtCoarse`` are located.</span>
+<span class="sd">        split (string, optional): The image split to use, ``train``, ``test`` or ``val`` if mode=&quot;fine&quot;</span>
+<span class="sd">            otherwise ``train``, ``train_extra`` or ``val``</span>
+<span class="sd">        mode (string, optional): The quality mode to use, ``fine`` or ``coarse``</span>
+<span class="sd">        target_type (string or list, optional): Type of target to use, ``instance``, ``semantic``, ``polygon``</span>
+<span class="sd">            or ``color``. Can also be a list to output a tuple with all specified target types.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in a PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes input sample and its target as entry</span>
+<span class="sd">            and returns a transformed version.</span>
+
+<span class="sd">    Examples:</span>
+
+<span class="sd">        Get semantic segmentation target</span>
+
+<span class="sd">        .. code-block:: python</span>
+
+<span class="sd">            dataset = Cityscapes(&#39;./data/cityscapes&#39;, split=&#39;train&#39;, mode=&#39;fine&#39;,</span>
+<span class="sd">                                 target_type=&#39;semantic&#39;)</span>
+
+<span class="sd">            img, smnt = dataset[0]</span>
+
+<span class="sd">        Get multiple targets</span>
+
+<span class="sd">        .. code-block:: python</span>
+
+<span class="sd">            dataset = Cityscapes(&#39;./data/cityscapes&#39;, split=&#39;train&#39;, mode=&#39;fine&#39;,</span>
+<span class="sd">                                 target_type=[&#39;instance&#39;, &#39;color&#39;, &#39;polygon&#39;])</span>
+
+<span class="sd">            img, (inst, col, poly) = dataset[0]</span>
+
+<span class="sd">        Validate on the &quot;coarse&quot; set</span>
+
+<span class="sd">        .. code-block:: python</span>
+
+<span class="sd">            dataset = Cityscapes(&#39;./data/cityscapes&#39;, split=&#39;val&#39;, mode=&#39;coarse&#39;,</span>
+<span class="sd">                                 target_type=&#39;semantic&#39;)</span>
+
+<span class="sd">            img, smnt = dataset[0]</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="c1"># Based on https://github.com/mcordts/cityscapesScripts</span>
+    <span class="n">CityscapesClass</span> <span class="o">=</span> <span class="n">namedtuple</span><span class="p">(</span>
+        <span class="s2">&quot;CityscapesClass&quot;</span><span class="p">,</span>
+        <span class="p">[</span><span class="s2">&quot;name&quot;</span><span class="p">,</span> <span class="s2">&quot;id&quot;</span><span class="p">,</span> <span class="s2">&quot;train_id&quot;</span><span class="p">,</span> <span class="s2">&quot;category&quot;</span><span class="p">,</span> <span class="s2">&quot;category_id&quot;</span><span class="p">,</span> <span class="s2">&quot;has_instances&quot;</span><span class="p">,</span> <span class="s2">&quot;ignore_in_eval&quot;</span><span class="p">,</span> <span class="s2">&quot;color&quot;</span><span class="p">],</span>
+    <span class="p">)</span>
+
+    <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;unlabeled&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;void&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;ego vehicle&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;void&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;rectification border&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;void&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;out of roi&quot;</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;void&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;static&quot;</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;void&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;dynamic&quot;</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;void&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">111</span><span class="p">,</span> <span class="mi">74</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;ground&quot;</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;void&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">81</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">81</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;road&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="s2">&quot;flat&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">128</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;sidewalk&quot;</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="s2">&quot;flat&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">244</span><span class="p">,</span> <span class="mi">35</span><span class="p">,</span> <span class="mi">232</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;parking&quot;</span><span class="p">,</span> <span class="mi">9</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;flat&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">250</span><span class="p">,</span> <span class="mi">170</span><span class="p">,</span> <span class="mi">160</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;rail track&quot;</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;flat&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">230</span><span class="p">,</span> <span class="mi">150</span><span class="p">,</span> <span class="mi">140</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;building&quot;</span><span class="p">,</span> <span class="mi">11</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="s2">&quot;construction&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">70</span><span class="p">,</span> <span class="mi">70</span><span class="p">,</span> <span class="mi">70</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;wall&quot;</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="s2">&quot;construction&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">102</span><span class="p">,</span> <span class="mi">102</span><span class="p">,</span> <span class="mi">156</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;fence&quot;</span><span class="p">,</span> <span class="mi">13</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="s2">&quot;construction&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">190</span><span class="p">,</span> <span class="mi">153</span><span class="p">,</span> <span class="mi">153</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;guard rail&quot;</span><span class="p">,</span> <span class="mi">14</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;construction&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">180</span><span class="p">,</span> <span class="mi">165</span><span class="p">,</span> <span class="mi">180</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;bridge&quot;</span><span class="p">,</span> <span class="mi">15</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;construction&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">150</span><span class="p">,</span> <span class="mi">100</span><span class="p">,</span> <span class="mi">100</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;tunnel&quot;</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;construction&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">150</span><span class="p">,</span> <span class="mi">120</span><span class="p">,</span> <span class="mi">90</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;pole&quot;</span><span class="p">,</span> <span class="mi">17</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="s2">&quot;object&quot;</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">153</span><span class="p">,</span> <span class="mi">153</span><span class="p">,</span> <span class="mi">153</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;polegroup&quot;</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;object&quot;</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">153</span><span class="p">,</span> <span class="mi">153</span><span class="p">,</span> <span class="mi">153</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;traffic light&quot;</span><span class="p">,</span> <span class="mi">19</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="s2">&quot;object&quot;</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">250</span><span class="p">,</span> <span class="mi">170</span><span class="p">,</span> <span class="mi">30</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;traffic sign&quot;</span><span class="p">,</span> <span class="mi">20</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="s2">&quot;object&quot;</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">220</span><span class="p">,</span> <span class="mi">220</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;vegetation&quot;</span><span class="p">,</span> <span class="mi">21</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="s2">&quot;nature&quot;</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">107</span><span class="p">,</span> <span class="mi">142</span><span class="p">,</span> <span class="mi">35</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;terrain&quot;</span><span class="p">,</span> <span class="mi">22</span><span class="p">,</span> <span class="mi">9</span><span class="p">,</span> <span class="s2">&quot;nature&quot;</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">152</span><span class="p">,</span> <span class="mi">251</span><span class="p">,</span> <span class="mi">152</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;sky&quot;</span><span class="p">,</span> <span class="mi">23</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="s2">&quot;sky&quot;</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">70</span><span class="p">,</span> <span class="mi">130</span><span class="p">,</span> <span class="mi">180</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;person&quot;</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">11</span><span class="p">,</span> <span class="s2">&quot;human&quot;</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">220</span><span class="p">,</span> <span class="mi">20</span><span class="p">,</span> <span class="mi">60</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;rider&quot;</span><span class="p">,</span> <span class="mi">25</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="s2">&quot;human&quot;</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">255</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;car&quot;</span><span class="p">,</span> <span class="mi">26</span><span class="p">,</span> <span class="mi">13</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">142</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;truck&quot;</span><span class="p">,</span> <span class="mi">27</span><span class="p">,</span> <span class="mi">14</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">70</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;bus&quot;</span><span class="p">,</span> <span class="mi">28</span><span class="p">,</span> <span class="mi">15</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">60</span><span class="p">,</span> <span class="mi">100</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;caravan&quot;</span><span class="p">,</span> <span class="mi">29</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">90</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;trailer&quot;</span><span class="p">,</span> <span class="mi">30</span><span class="p">,</span> <span class="mi">255</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">110</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="mi">31</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="mi">100</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;motorcycle&quot;</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">17</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">230</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;bicycle&quot;</span><span class="p">,</span> <span class="mi">33</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="p">(</span><span class="mi">119</span><span class="p">,</span> <span class="mi">11</span><span class="p">,</span> <span class="mi">32</span><span class="p">)),</span>
+        <span class="n">CityscapesClass</span><span class="p">(</span><span class="s2">&quot;license plate&quot;</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="s2">&quot;vehicle&quot;</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">142</span><span class="p">)),</span>
+    <span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;fine&quot;</span><span class="p">,</span>
+        <span class="n">target_type</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;instance&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">,</span> <span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">=</span> <span class="s2">&quot;gtFine&quot;</span> <span class="k">if</span> <span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;fine&quot;</span> <span class="k">else</span> <span class="s2">&quot;gtCoarse&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;leftImg8bit&quot;</span><span class="p">,</span> <span class="n">split</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">targets_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="n">split</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="o">=</span> <span class="n">target_type</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">split</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">images</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">mode</span><span class="p">,</span> <span class="s2">&quot;mode&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;fine&quot;</span><span class="p">,</span> <span class="s2">&quot;coarse&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;fine&quot;</span><span class="p">:</span>
+            <span class="n">valid_modes</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">valid_modes</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;train_extra&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">)</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="s2">&quot;Unknown value &#39;</span><span class="si">{}</span><span class="s2">&#39; for argument split if mode is &#39;</span><span class="si">{}</span><span class="s2">&#39;. Valid values are {{</span><span class="si">{}</span><span class="s2">}}.&quot;</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="n">mode</span><span class="p">,</span> <span class="n">iterable_to_str</span><span class="p">(</span><span class="n">valid_modes</span><span class="p">))</span>
+        <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_modes</span><span class="p">,</span> <span class="n">msg</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">target_type</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="o">=</span> <span class="p">[</span><span class="n">target_type</span><span class="p">]</span>
+        <span class="p">[</span>
+            <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="s2">&quot;target_type&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;instance&quot;</span><span class="p">,</span> <span class="s2">&quot;semantic&quot;</span><span class="p">,</span> <span class="s2">&quot;polygon&quot;</span><span class="p">,</span> <span class="s2">&quot;color&quot;</span><span class="p">))</span>
+            <span class="k">for</span> <span class="n">value</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span>
+        <span class="p">]</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets_dir</span><span class="p">):</span>
+
+            <span class="k">if</span> <span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train_extra&quot;</span><span class="p">:</span>
+                <span class="n">image_dir_zip</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;leftImg8bit_trainextra.zip&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">image_dir_zip</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;leftImg8bit_trainvaltest.zip&quot;</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;gtFine&quot;</span><span class="p">:</span>
+                <span class="n">target_dir_zip</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="si">}</span><span class="s2">_trainvaltest.zip&quot;</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;gtCoarse&quot;</span><span class="p">:</span>
+                <span class="n">target_dir_zip</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="si">}</span><span class="s2">.zip&quot;</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">image_dir_zip</span><span class="p">)</span> <span class="ow">and</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">target_dir_zip</span><span class="p">):</span>
+                <span class="n">extract_archive</span><span class="p">(</span><span class="n">from_path</span><span class="o">=</span><span class="n">image_dir_zip</span><span class="p">,</span> <span class="n">to_path</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+                <span class="n">extract_archive</span><span class="p">(</span><span class="n">from_path</span><span class="o">=</span><span class="n">target_dir_zip</span><span class="p">,</span> <span class="n">to_path</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                    <span class="s2">&quot;Dataset not found or incomplete. Please make sure all required folders for the&quot;</span>
+                    <span class="s1">&#39; specified &quot;split&quot; and &quot;mode&quot; are inside the &quot;root&quot; directory&#39;</span>
+                <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">city</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">):</span>
+            <span class="n">img_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">,</span> <span class="n">city</span><span class="p">)</span>
+            <span class="n">target_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets_dir</span><span class="p">,</span> <span class="n">city</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">file_name</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">img_dir</span><span class="p">):</span>
+                <span class="n">target_types</span> <span class="o">=</span> <span class="p">[]</span>
+                <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span><span class="p">:</span>
+                    <span class="n">target_name</span> <span class="o">=</span> <span class="s2">&quot;</span><span class="si">{}</span><span class="s2">_</span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                        <span class="n">file_name</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;_leftImg8bit&quot;</span><span class="p">)[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_target_suffix</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="n">t</span><span class="p">)</span>
+                    <span class="p">)</span>
+                    <span class="n">target_types</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">target_dir</span><span class="p">,</span> <span class="n">target_name</span><span class="p">))</span>
+
+                <span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">img_dir</span><span class="p">,</span> <span class="n">file_name</span><span class="p">))</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">target_types</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Cityscapes.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Cityscapes.html#torchvision.datasets.Cityscapes.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is a tuple of all target types if target_type is a list with more</span>
+<span class="sd">            than one item. Otherwise, target is a json object if target_type=&quot;polygon&quot;, else the image segmentation.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">[</span><span class="n">index</span><span class="p">])</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="n">targets</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">t</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">target_type</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;polygon&quot;</span><span class="p">:</span>
+                <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_json</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="n">i</span><span class="p">])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">target</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="n">i</span><span class="p">])</span>
+
+            <span class="n">targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="n">target</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">targets</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">targets</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">targets</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">lines</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Split: </span><span class="si">{split}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;Mode: </span><span class="si">{mode}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;Type: </span><span class="si">{target_type}</span><span class="s2">&quot;</span><span class="p">]</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">lines</span><span class="p">)</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_json</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path</span><span class="p">)</span> <span class="k">as</span> <span class="n">file</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">file</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">data</span>
+
+    <span class="k">def</span> <span class="nf">_get_target_suffix</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">target_type</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">target_type</span> <span class="o">==</span> <span class="s2">&quot;instance&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">mode</span><span class="si">}</span><span class="s2">_instanceIds.png&quot;</span>
+        <span class="k">elif</span> <span class="n">target_type</span> <span class="o">==</span> <span class="s2">&quot;semantic&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">mode</span><span class="si">}</span><span class="s2">_labelIds.png&quot;</span>
+        <span class="k">elif</span> <span class="n">target_type</span> <span class="o">==</span> <span class="s2">&quot;color&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">mode</span><span class="si">}</span><span class="s2">_color.png&quot;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">mode</span><span class="si">}</span><span class="s2">_polygons.json&quot;</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/coco.html b/0.16/_modules/torchvision/datasets/coco.html
new file mode 100644
index 00000000000..20790b2b53a
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/coco.html
@@ -0,0 +1,773 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.coco &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.coco</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.coco</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="CocoDetection"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection">[docs]</a><span class="k">class</span> <span class="nc">CocoDetection</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`MS Coco Detection &lt;https://cocodataset.org/#detection-2016&gt;`_ Dataset.</span>
+
+<span class="sd">    It requires the `COCO API to be installed &lt;https://github.com/pdollar/coco/tree/master/PythonAPI&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images are downloaded to.</span>
+<span class="sd">        annFile (string): Path to json annotation file.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.PILToTensor``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes input sample and its target as entry</span>
+<span class="sd">            and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">annFile</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">,</span> <span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">)</span>
+        <span class="kn">from</span> <span class="nn">pycocotools.coco</span> <span class="kn">import</span> <span class="n">COCO</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">coco</span> <span class="o">=</span> <span class="n">COCO</span><span class="p">(</span><span class="n">annFile</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ids</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">coco</span><span class="o">.</span><span class="n">imgs</span><span class="o">.</span><span class="n">keys</span><span class="p">()))</span>
+
+    <span class="k">def</span> <span class="nf">_load_image</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">id</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+        <span class="n">path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">coco</span><span class="o">.</span><span class="n">loadImgs</span><span class="p">(</span><span class="nb">id</span><span class="p">)[</span><span class="mi">0</span><span class="p">][</span><span class="s2">&quot;file_name&quot;</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">path</span><span class="p">))</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_target</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">id</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">coco</span><span class="o">.</span><span class="n">loadAnns</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">coco</span><span class="o">.</span><span class="n">getAnnIds</span><span class="p">(</span><span class="nb">id</span><span class="p">))</span>
+
+<div class="viewcode-block" id="CocoDetection.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="nb">id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ids</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_image</span><span class="p">(</span><span class="nb">id</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_target</span><span class="p">(</span><span class="nb">id</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ids</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CocoCaptions"><a class="viewcode-back" href="../../../generated/torchvision.datasets.CocoCaptions.html#torchvision.datasets.CocoCaptions">[docs]</a><span class="k">class</span> <span class="nc">CocoCaptions</span><span class="p">(</span><span class="n">CocoDetection</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`MS Coco Captions &lt;https://cocodataset.org/#captions-2015&gt;`_ Dataset.</span>
+
+<span class="sd">    It requires the `COCO API to be installed &lt;https://github.com/pdollar/coco/tree/master/PythonAPI&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images are downloaded to.</span>
+<span class="sd">        annFile (string): Path to json annotation file.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.PILToTensor``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes input sample and its target as entry</span>
+<span class="sd">            and returns a transformed version.</span>
+
+<span class="sd">    Example:</span>
+
+<span class="sd">        .. code:: python</span>
+
+<span class="sd">            import torchvision.datasets as dset</span>
+<span class="sd">            import torchvision.transforms as transforms</span>
+<span class="sd">            cap = dset.CocoCaptions(root = &#39;dir where images are&#39;,</span>
+<span class="sd">                                    annFile = &#39;json annotation file&#39;,</span>
+<span class="sd">                                    transform=transforms.PILToTensor())</span>
+
+<span class="sd">            print(&#39;Number of samples: &#39;, len(cap))</span>
+<span class="sd">            img, target = cap[3] # load 4th sample</span>
+
+<span class="sd">            print(&quot;Image Size: &quot;, img.size())</span>
+<span class="sd">            print(target)</span>
+
+<span class="sd">        Output: ::</span>
+
+<span class="sd">            Number of samples: 82783</span>
+<span class="sd">            Image Size: (3L, 427L, 640L)</span>
+<span class="sd">            [u&#39;A plane emitting smoke stream flying over a mountain.&#39;,</span>
+<span class="sd">            u&#39;A plane darts across a bright blue sky behind a mountain covered in snow&#39;,</span>
+<span class="sd">            u&#39;A plane leaves a contrail above the snowy mountain top.&#39;,</span>
+<span class="sd">            u&#39;A mountain that has a plane flying overheard in the distance.&#39;,</span>
+<span class="sd">            u&#39;A mountain view with a plume of smoke in the background&#39;]</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_load_target</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">id</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="p">[</span><span class="n">ann</span><span class="p">[</span><span class="s2">&quot;caption&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">ann</span> <span class="ow">in</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_target</span><span class="p">(</span><span class="nb">id</span><span class="p">)]</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/country211.html b/0.16/_modules/torchvision/datasets/country211.html
new file mode 100644
index 00000000000..02953139322
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/country211.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.country211 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.country211</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.country211</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">ImageFolder</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+
+
+<div class="viewcode-block" id="Country211"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Country211.html#torchvision.datasets.Country211">[docs]</a><span class="k">class</span> <span class="nc">Country211</span><span class="p">(</span><span class="n">ImageFolder</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`The Country211 Data Set &lt;https://github.com/openai/CLIP/blob/main/data/country211.md&gt;`_ from OpenAI.</span>
+
+<span class="sd">    This dataset was built by filtering the images from the YFCC100m dataset</span>
+<span class="sd">    that have GPS coordinate corresponding to a ISO-3166 country code. The</span>
+<span class="sd">    dataset is balanced by sampling 150 train images, 50 validation images, and</span>
+<span class="sd">    100 test images for each country.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default), ``&quot;valid&quot;`` and ``&quot;test&quot;``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and puts it into</span>
+<span class="sd">            ``root/country211/``. If dataset is already downloaded, it is not downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_URL</span> <span class="o">=</span> <span class="s2">&quot;https://openaipublic.azureedge.net/clip/data/country211.tgz&quot;</span>
+    <span class="n">_MD5</span> <span class="o">=</span> <span class="s2">&quot;84988d7644798601126c29e9877aab6a&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;valid&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span><span class="o">.</span><span class="n">expanduser</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">root</span> <span class="o">/</span> <span class="s2">&quot;country211&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="o">.</span><span class="n">exists</span><span class="p">()</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="o">.</span><span class="n">is_dir</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_URL</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_MD5</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/dtd.html b/0.16/_modules/torchvision/datasets/dtd.html
new file mode 100644
index 00000000000..29393635451
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/dtd.html
@@ -0,0 +1,769 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.dtd &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.dtd</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.dtd</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">pathlib</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="DTD"><a class="viewcode-back" href="../../../generated/torchvision.datasets.DTD.html#torchvision.datasets.DTD">[docs]</a><span class="k">class</span> <span class="nc">DTD</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Describable Textures Dataset (DTD) &lt;https://www.robots.ox.ac.uk/~vgg/data/dtd/&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default), ``&quot;val&quot;``, or ``&quot;test&quot;``.</span>
+<span class="sd">        partition (int, optional): The dataset partition. Should be ``1 &lt;= partition &lt;= 10``. Defaults to ``1``.</span>
+
+<span class="sd">            .. note::</span>
+
+<span class="sd">                The partition only changes which split each image belongs to. Thus, regardless of the selected</span>
+<span class="sd">                partition, combining all splits will result in all images.</span>
+
+<span class="sd">        transform (callable, optional): A function/transform that  takes in a PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again. Default is False.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_URL</span> <span class="o">=</span> <span class="s2">&quot;https://www.robots.ox.ac.uk/~vgg/data/dtd/download/dtd-r1.0.1.tar.gz&quot;</span>
+    <span class="n">_MD5</span> <span class="o">=</span> <span class="s2">&quot;fff73e5086ae6bdbea199a49dfb8a4c1&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">partition</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">partition</span><span class="p">,</span> <span class="nb">int</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">partition</span> <span class="o">&lt;=</span> <span class="mi">10</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Parameter &#39;partition&#39; should be an integer with `1 &lt;= partition &lt;= 10`, &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;but got </span><span class="si">{</span><span class="n">partition</span><span class="si">}</span><span class="s2"> instead&quot;</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_partition</span> <span class="o">=</span> <span class="n">partition</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;dtd&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_meta_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span> <span class="o">/</span> <span class="s2">&quot;labels&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span> <span class="o">/</span> <span class="s2">&quot;images&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">classes</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_meta_folder</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}{</span><span class="bp">self</span><span class="o">.</span><span class="n">_partition</span><span class="si">}</span><span class="s2">.txt&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">file</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">file</span><span class="p">:</span>
+                <span class="bp">cls</span><span class="p">,</span> <span class="n">name</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;/&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span><span class="o">.</span><span class="n">joinpath</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">name</span><span class="p">))</span>
+                <span class="n">classes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">cls</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">classes</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">))))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span><span class="p">[</span><span class="bp">cls</span><span class="p">]</span> <span class="k">for</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="n">classes</span><span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">)</span>
+
+<div class="viewcode-block" id="DTD.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.DTD.html#torchvision.datasets.DTD.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image_file</span><span class="p">,</span> <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_file</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">label</span></div>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;split=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}</span><span class="s2">, partition=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_partition</span><span class="si">}</span><span class="s2">&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span><span class="p">)</span> <span class="ow">and</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_URL</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_MD5</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/eurosat.html b/0.16/_modules/torchvision/datasets/eurosat.html
new file mode 100644
index 00000000000..3c270c93435
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/eurosat.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.eurosat &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.eurosat</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.eurosat</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">ImageFolder</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span>
+
+
+<div class="viewcode-block" id="EuroSAT"><a class="viewcode-back" href="../../../generated/torchvision.datasets.EuroSAT.html#torchvision.datasets.EuroSAT">[docs]</a><span class="k">class</span> <span class="nc">EuroSAT</span><span class="p">(</span><span class="n">ImageFolder</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;RGB version of the `EuroSAT &lt;https://github.com/phelber/eurosat&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where ``root/eurosat`` exists.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again. Default is False.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">expanduser</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;eurosat&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">,</span> <span class="s2">&quot;2750&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">expanduser</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_folder</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="s2">&quot;https://madm.dfki.de/files/sentinel/EuroSAT.zip&quot;</span><span class="p">,</span>
+            <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">,</span>
+            <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;c8fa014336c82ac7804f0398fcb19387&quot;</span><span class="p">,</span>
+        <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/fakedata.html b/0.16/_modules/torchvision/datasets/fakedata.html
new file mode 100644
index 00000000000..21012c8272d
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/fakedata.html
@@ -0,0 +1,736 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.fakedata &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.fakedata</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.fakedata</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">..</span> <span class="kn">import</span> <span class="n">transforms</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="FakeData"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FakeData.html#torchvision.datasets.FakeData">[docs]</a><span class="k">class</span> <span class="nc">FakeData</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;A fake dataset that returns randomly generated images and returns them as PIL images</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (int, optional): Size of the dataset. Default: 1000 images</span>
+<span class="sd">        image_size(tuple, optional): Size if the returned images. Default: (3, 224, 224)</span>
+<span class="sd">        num_classes(int, optional): Number of classes in the dataset. Default: 10</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        random_offset (int): Offsets the index-based random seed used to</span>
+<span class="sd">            generate each image. Default: 0</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">image_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">10</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">random_offset</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="kc">None</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>  <span class="c1"># type: ignore[arg-type]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_size</span> <span class="o">=</span> <span class="n">image_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">random_offset</span> <span class="o">=</span> <span class="n">random_offset</span>
+
+<div class="viewcode-block" id="FakeData.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FakeData.html#torchvision.datasets.FakeData.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is class_index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># create random image that is consistent with the index id</span>
+        <span class="k">if</span> <span class="n">index</span> <span class="o">&gt;=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">IndexError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2"> index out of range&quot;</span><span class="p">)</span>
+        <span class="n">rng_state</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">get_rng_state</span><span class="p">()</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">manual_seed</span><span class="p">(</span><span class="n">index</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">random_offset</span><span class="p">)</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randn</span><span class="p">(</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">image_size</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">set_rng_state</span><span class="p">(</span><span class="n">rng_state</span><span class="p">)</span>
+
+        <span class="c1"># convert to PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">ToPILImage</span><span class="p">()(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span><span class="o">.</span><span class="n">item</span><span class="p">()</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/fer2013.html b/0.16/_modules/torchvision/datasets/fer2013.html
new file mode 100644
index 00000000000..bd368498a1d
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/fer2013.html
@@ -0,0 +1,744 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.fer2013 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.fer2013</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.fer2013</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">csv</span>
+<span class="kn">import</span> <span class="nn">pathlib</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="FER2013"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FER2013.html#torchvision.datasets.FER2013">[docs]</a><span class="k">class</span> <span class="nc">FER2013</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`FER2013</span>
+<span class="sd">    &lt;https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``root/fer2013`` exists.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default), or ``&quot;test&quot;``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in an PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_RESOURCES</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;train.csv&quot;</span><span class="p">,</span> <span class="s2">&quot;3f0dfb3d3fd99c811a1299cb947e3131&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;test.csv&quot;</span><span class="p">,</span> <span class="s2">&quot;b02c2298636a634e8c2faabbf3ea9a23&quot;</span><span class="p">),</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_RESOURCES</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="n">base_folder</span> <span class="o">=</span> <span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;fer2013&quot;</span>
+        <span class="n">file_name</span><span class="p">,</span> <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_RESOURCES</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">]</span>
+        <span class="n">data_file</span> <span class="o">=</span> <span class="n">base_folder</span> <span class="o">/</span> <span class="n">file_name</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">data_file</span><span class="p">),</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">file_name</span><span class="si">}</span><span class="s2"> not found in </span><span class="si">{</span><span class="n">base_folder</span><span class="si">}</span><span class="s2"> or corrupted. &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;You can download it from &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">data_file</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">,</span> <span class="n">newline</span><span class="o">=</span><span class="s2">&quot;&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">file</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="nb">int</span><span class="p">(</span><span class="n">idx</span><span class="p">)</span> <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="n">row</span><span class="p">[</span><span class="s2">&quot;pixels&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">split</span><span class="p">()],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">48</span><span class="p">,</span> <span class="mi">48</span><span class="p">),</span>
+                    <span class="nb">int</span><span class="p">(</span><span class="n">row</span><span class="p">[</span><span class="s2">&quot;emotion&quot;</span><span class="p">])</span> <span class="k">if</span> <span class="s2">&quot;emotion&quot;</span> <span class="ow">in</span> <span class="n">row</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">csv</span><span class="o">.</span><span class="n">DictReader</span><span class="p">(</span><span class="n">file</span><span class="p">)</span>
+            <span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">)</span>
+
+<div class="viewcode-block" id="FER2013.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FER2013.html#torchvision.datasets.FER2013.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image_tensor</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">image_tensor</span><span class="o">.</span><span class="n">numpy</span><span class="p">())</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;split=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}</span><span class="s2">&quot;</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/fgvc_aircraft.html b/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
new file mode 100644
index 00000000000..e8aaa40b758
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.fgvc_aircraft &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.fgvc_aircraft</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.fgvc_aircraft</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">annotations</span>
+
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="FGVCAircraft"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FGVCAircraft.html#torchvision.datasets.FGVCAircraft">[docs]</a><span class="k">class</span> <span class="nc">FGVCAircraft</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`FGVC Aircraft &lt;https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/&gt;`_ Dataset.</span>
+
+<span class="sd">    The dataset contains 10,000 images of aircraft, with 100 images for each of 100</span>
+<span class="sd">    different aircraft model variants, most of which are airplanes.</span>
+<span class="sd">    Aircraft models are organized in a three-levels hierarchy. The three levels, from</span>
+<span class="sd">    finer to coarser, are:</span>
+
+<span class="sd">    - ``variant``, e.g. Boeing 737-700. A variant collapses all the models that are visually</span>
+<span class="sd">        indistinguishable into one class. The dataset comprises 100 different variants.</span>
+<span class="sd">    - ``family``, e.g. Boeing 737. The dataset comprises 70 different families.</span>
+<span class="sd">    - ``manufacturer``, e.g. Boeing. The dataset comprises 30 different manufacturers.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the FGVC Aircraft dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``train``, ``val``,</span>
+<span class="sd">            ``trainval`` and ``test``.</span>
+<span class="sd">        annotation_level (str, optional): The annotation level, supports ``variant``,</span>
+<span class="sd">            ``family`` and ``manufacturer``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_URL</span> <span class="o">=</span> <span class="s2">&quot;https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/archives/fgvc-aircraft-2013b.tar.gz&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;trainval&quot;</span><span class="p">,</span>
+        <span class="n">annotation_level</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;variant&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;trainval&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_annotation_level</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span>
+            <span class="n">annotation_level</span><span class="p">,</span> <span class="s2">&quot;annotation_level&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;variant&quot;</span><span class="p">,</span> <span class="s2">&quot;family&quot;</span><span class="p">,</span> <span class="s2">&quot;manufacturer&quot;</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_data_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;fgvc-aircraft-2013b&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="n">annotation_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_data_path</span><span class="p">,</span>
+            <span class="s2">&quot;data&quot;</span><span class="p">,</span>
+            <span class="p">{</span>
+                <span class="s2">&quot;variant&quot;</span><span class="p">:</span> <span class="s2">&quot;variants.txt&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;family&quot;</span><span class="p">:</span> <span class="s2">&quot;families.txt&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;manufacturer&quot;</span><span class="p">:</span> <span class="s2">&quot;manufacturers.txt&quot;</span><span class="p">,</span>
+            <span class="p">}[</span><span class="bp">self</span><span class="o">.</span><span class="n">_annotation_level</span><span class="p">],</span>
+        <span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">annotation_file</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">))))</span>
+
+        <span class="n">image_data_folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_path</span><span class="p">,</span> <span class="s2">&quot;data&quot;</span><span class="p">,</span> <span class="s2">&quot;images&quot;</span><span class="p">)</span>
+        <span class="n">labels_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_path</span><span class="p">,</span> <span class="s2">&quot;data&quot;</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;images_</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_annotation_level</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}</span><span class="s2">.txt&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">labels_file</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>
+                <span class="n">image_name</span><span class="p">,</span> <span class="n">label_name</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">image_data_folder</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">image_name</span><span class="si">}</span><span class="s2">.jpg&quot;</span><span class="p">))</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span><span class="p">[</span><span class="n">label_name</span><span class="p">])</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">)</span>
+
+<div class="viewcode-block" id="FGVCAircraft.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FGVCAircraft.html#torchvision.datasets.FGVCAircraft.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image_file</span><span class="p">,</span> <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_file</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">label</span></div>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Download the FGVC Aircraft dataset archive and extract it under root.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_URL</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_path</span><span class="p">)</span> <span class="ow">and</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_path</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/flickr.html b/0.16/_modules/torchvision/datasets/flickr.html
new file mode 100644
index 00000000000..36b4fc2ba7f
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/flickr.html
@@ -0,0 +1,835 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.flickr &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.flickr</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.flickr</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">glob</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">defaultdict</span>
+<span class="kn">from</span> <span class="nn">html.parser</span> <span class="kn">import</span> <span class="n">HTMLParser</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<span class="k">class</span> <span class="nc">Flickr8kParser</span><span class="p">(</span><span class="n">HTMLParser</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Parser for extracting captions from the Flickr8k dataset web page.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">root</span>
+
+        <span class="c1"># Data structure to store captions</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="c1"># State variables</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">in_table</span> <span class="o">=</span> <span class="kc">False</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">current_tag</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">current_img</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+
+    <span class="k">def</span> <span class="nf">handle_starttag</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">attrs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]]])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">current_tag</span> <span class="o">=</span> <span class="n">tag</span>
+
+        <span class="k">if</span> <span class="n">tag</span> <span class="o">==</span> <span class="s2">&quot;table&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">in_table</span> <span class="o">=</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">handle_endtag</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tag</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">current_tag</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="n">tag</span> <span class="o">==</span> <span class="s2">&quot;table&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">in_table</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="k">def</span> <span class="nf">handle_data</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">in_table</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">data</span> <span class="o">==</span> <span class="s2">&quot;Image Not Found&quot;</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">current_img</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_tag</span> <span class="o">==</span> <span class="s2">&quot;a&quot;</span><span class="p">:</span>
+                <span class="n">img_id</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;/&quot;</span><span class="p">)[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+                <span class="n">img_id</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">img_id</span> <span class="o">+</span> <span class="s2">&quot;_*.jpg&quot;</span><span class="p">)</span>
+                <span class="n">img_id</span> <span class="o">=</span> <span class="n">glob</span><span class="o">.</span><span class="n">glob</span><span class="p">(</span><span class="n">img_id</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">current_img</span> <span class="o">=</span> <span class="n">img_id</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="p">[</span><span class="n">img_id</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_tag</span> <span class="o">==</span> <span class="s2">&quot;li&quot;</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_img</span><span class="p">:</span>
+                <span class="n">img_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_img</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="p">[</span><span class="n">img_id</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">data</span><span class="o">.</span><span class="n">strip</span><span class="p">())</span>
+
+
+<div class="viewcode-block" id="Flickr8k"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Flickr8k.html#torchvision.datasets.Flickr8k">[docs]</a><span class="k">class</span> <span class="nc">Flickr8k</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Flickr8k Entities &lt;http://hockenmaier.cs.illinois.edu/8k-pictures.html&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images are downloaded to.</span>
+<span class="sd">        ann_file (string): Path to annotation file.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in a PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.PILToTensor``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">ann_file</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ann_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">expanduser</span><span class="p">(</span><span class="n">ann_file</span><span class="p">)</span>
+
+        <span class="c1"># Read annotations and store in a dict</span>
+        <span class="n">parser</span> <span class="o">=</span> <span class="n">Flickr8kParser</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ann_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">fh</span><span class="p">:</span>
+            <span class="n">parser</span><span class="o">.</span><span class="n">feed</span><span class="p">(</span><span class="n">fh</span><span class="o">.</span><span class="n">read</span><span class="p">())</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">annotations</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">ids</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="o">.</span><span class="n">keys</span><span class="p">()))</span>
+
+<div class="viewcode-block" id="Flickr8k.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Flickr8k.html#torchvision.datasets.Flickr8k.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: Tuple (image, target). target is a list of captions for the image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ids</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+        <span class="c1"># Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">img_id</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="c1"># Captions</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="p">[</span><span class="n">img_id</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ids</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Flickr30k"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Flickr30k.html#torchvision.datasets.Flickr30k">[docs]</a><span class="k">class</span> <span class="nc">Flickr30k</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Flickr30k Entities &lt;https://bryanplummer.com/Flickr30kEntities/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images are downloaded to.</span>
+<span class="sd">        ann_file (string): Path to annotation file.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in a PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.PILToTensor``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">ann_file</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ann_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">expanduser</span><span class="p">(</span><span class="n">ann_file</span><span class="p">)</span>
+
+        <span class="c1"># Read annotations and store in a dict</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="nb">list</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ann_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">fh</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">fh</span><span class="p">:</span>
+                <span class="n">img_id</span><span class="p">,</span> <span class="n">caption</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\t</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="p">[</span><span class="n">img_id</span><span class="p">[:</span><span class="o">-</span><span class="mi">2</span><span class="p">]]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">caption</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">ids</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="o">.</span><span class="n">keys</span><span class="p">()))</span>
+
+<div class="viewcode-block" id="Flickr30k.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Flickr30k.html#torchvision.datasets.Flickr30k.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: Tuple (image, target). target is a list of captions for the image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ids</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+        <span class="c1"># Image</span>
+        <span class="n">filename</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">img_id</span><span class="p">)</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">filename</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="c1"># Captions</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="p">[</span><span class="n">img_id</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ids</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/flowers102.html b/0.16/_modules/torchvision/datasets/flowers102.html
new file mode 100644
index 00000000000..e786a1673e8
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/flowers102.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.flowers102 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.flowers102</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.flowers102</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">download_url</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="Flowers102"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Flowers102.html#torchvision.datasets.Flowers102">[docs]</a><span class="k">class</span> <span class="nc">Flowers102</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Oxford 102 Flower &lt;https://www.robots.ox.ac.uk/~vgg/data/flowers/102/&gt;`_ Dataset.</span>
+
+<span class="sd">    .. warning::</span>
+
+<span class="sd">        This class needs `scipy &lt;https://docs.scipy.org/doc/&gt;`_ to load target files from `.mat` format.</span>
+
+<span class="sd">    Oxford 102 Flower is an image classification dataset consisting of 102 flower categories. The</span>
+<span class="sd">    flowers were chosen to be flowers commonly occurring in the United Kingdom. Each class consists of</span>
+<span class="sd">    between 40 and 258 images.</span>
+
+<span class="sd">    The images have large scale, pose and light variations. In addition, there are categories that</span>
+<span class="sd">    have large variations within the category, and several very similar categories.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default), ``&quot;val&quot;``, or ``&quot;test&quot;``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in an PIL image and returns a</span>
+<span class="sd">            transformed version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_download_url_prefix</span> <span class="o">=</span> <span class="s2">&quot;https://www.robots.ox.ac.uk/~vgg/data/flowers/102/&quot;</span>
+    <span class="n">_file_dict</span> <span class="o">=</span> <span class="p">{</span>  <span class="c1"># filename, md5</span>
+        <span class="s2">&quot;image&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;102flowers.tgz&quot;</span><span class="p">,</span> <span class="s2">&quot;52808999861908f626f3c1f4e79d11fa&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;label&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;imagelabels.mat&quot;</span><span class="p">,</span> <span class="s2">&quot;e0620be6f572b9609742df49c70aed4d&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;setid&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;setid.mat&quot;</span><span class="p">,</span> <span class="s2">&quot;a5357ecc9cb78c4bef273ce3793fc85c&quot;</span><span class="p">),</span>
+    <span class="p">}</span>
+    <span class="n">_splits_map</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="s2">&quot;trnid&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">:</span> <span class="s2">&quot;valid&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="s2">&quot;tstid&quot;</span><span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;flowers-102&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;jpg&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="kn">from</span> <span class="nn">scipy.io</span> <span class="kn">import</span> <span class="n">loadmat</span>
+
+        <span class="n">set_ids</span> <span class="o">=</span> <span class="n">loadmat</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">_file_dict</span><span class="p">[</span><span class="s2">&quot;setid&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="n">squeeze_me</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">image_ids</span> <span class="o">=</span> <span class="n">set_ids</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_splits_map</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">]]</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+
+        <span class="n">labels</span> <span class="o">=</span> <span class="n">loadmat</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">_file_dict</span><span class="p">[</span><span class="s2">&quot;label&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="n">squeeze_me</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">image_id_to_label</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">enumerate</span><span class="p">((</span><span class="n">labels</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">tolist</span><span class="p">(),</span> <span class="mi">1</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">image_id</span> <span class="ow">in</span> <span class="n">image_ids</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">image_id_to_label</span><span class="p">[</span><span class="n">image_id</span><span class="p">])</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;image_</span><span class="si">{</span><span class="n">image_id</span><span class="si">:</span><span class="s2">05d</span><span class="si">}</span><span class="s2">.jpg&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Flowers102.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Flowers102.html#torchvision.datasets.Flowers102.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image_file</span><span class="p">,</span> <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_file</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">label</span></div>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;split=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}</span><span class="s2">&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span><span class="o">.</span><span class="n">exists</span><span class="p">()</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span><span class="o">.</span><span class="n">is_dir</span><span class="p">()):</span>
+            <span class="k">return</span> <span class="kc">False</span>
+
+        <span class="k">for</span> <span class="nb">id</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;label&quot;</span><span class="p">,</span> <span class="s2">&quot;setid&quot;</span><span class="p">]:</span>
+            <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_file_dict</span><span class="p">[</span><span class="nb">id</span><span class="p">]</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="n">filename</span><span class="p">),</span> <span class="n">md5</span><span class="p">):</span>
+                <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_download_url_prefix</span><span class="si">}{</span><span class="bp">self</span><span class="o">.</span><span class="n">_file_dict</span><span class="p">[</span><span class="s1">&#39;image&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+            <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+            <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_file_dict</span><span class="p">[</span><span class="s2">&quot;image&quot;</span><span class="p">][</span><span class="mi">1</span><span class="p">],</span>
+        <span class="p">)</span>
+        <span class="k">for</span> <span class="nb">id</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;label&quot;</span><span class="p">,</span> <span class="s2">&quot;setid&quot;</span><span class="p">]:</span>
+            <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_file_dict</span><span class="p">[</span><span class="nb">id</span><span class="p">]</span>
+            <span class="n">download_url</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_download_url_prefix</span> <span class="o">+</span> <span class="n">filename</span><span class="p">,</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/folder.html b/0.16/_modules/torchvision/datasets/folder.html
new file mode 100644
index 00000000000..701163743e0
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/folder.html
@@ -0,0 +1,986 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.folder &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.folder</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.folder</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<span class="k">def</span> <span class="nf">has_file_allowed_extension</span><span class="p">(</span><span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">extensions</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="o">...</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Checks if a file is an allowed extension.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        filename (string): path to a file</span>
+<span class="sd">        extensions (tuple of strings): extensions to consider (lowercase)</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        bool: True if the filename ends with one of given extensions</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">return</span> <span class="n">filename</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="n">extensions</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">extensions</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="k">else</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">extensions</span><span class="p">))</span>
+
+
+<span class="k">def</span> <span class="nf">is_image_file</span><span class="p">(</span><span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Checks if a file is an allowed image extension.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        filename (string): path to a file</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        bool: True if the filename ends with a known image extension</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">return</span> <span class="n">has_file_allowed_extension</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">IMG_EXTENSIONS</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">find_classes</span><span class="p">(</span><span class="n">directory</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Finds the class folders in a dataset.</span>
+
+<span class="sd">    See :class:`DatasetFolder` for details.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">classes</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">entry</span><span class="o">.</span><span class="n">name</span> <span class="k">for</span> <span class="n">entry</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">scandir</span><span class="p">(</span><span class="n">directory</span><span class="p">)</span> <span class="k">if</span> <span class="n">entry</span><span class="o">.</span><span class="n">is_dir</span><span class="p">())</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">classes</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Couldn&#39;t find any class folder in </span><span class="si">{</span><span class="n">directory</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="n">cls_name</span><span class="p">:</span> <span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">cls_name</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">classes</span><span class="p">)}</span>
+    <span class="k">return</span> <span class="n">classes</span><span class="p">,</span> <span class="n">class_to_idx</span>
+
+
+<span class="k">def</span> <span class="nf">make_dataset</span><span class="p">(</span>
+    <span class="n">directory</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">class_to_idx</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">extensions</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="o">...</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">is_valid_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Generates a list of samples of a form (path_to_sample, class).</span>
+
+<span class="sd">    See :class:`DatasetFolder` for details.</span>
+
+<span class="sd">    Note: The class_to_idx parameter is here optional and will use the logic of the ``find_classes`` function</span>
+<span class="sd">    by default.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">directory</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">expanduser</span><span class="p">(</span><span class="n">directory</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">class_to_idx</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">class_to_idx</span> <span class="o">=</span> <span class="n">find_classes</span><span class="p">(</span><span class="n">directory</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="n">class_to_idx</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;&#39;class_to_index&#39; must have at least one entry to collect any samples.&quot;</span><span class="p">)</span>
+
+    <span class="n">both_none</span> <span class="o">=</span> <span class="n">extensions</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">is_valid_file</span> <span class="ow">is</span> <span class="kc">None</span>
+    <span class="n">both_something</span> <span class="o">=</span> <span class="n">extensions</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">is_valid_file</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="k">if</span> <span class="n">both_none</span> <span class="ow">or</span> <span class="n">both_something</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Both extensions and is_valid_file cannot be None or not None at the same time&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">extensions</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="k">def</span> <span class="nf">is_valid_file</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">has_file_allowed_extension</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">extensions</span><span class="p">)</span>  <span class="c1"># type: ignore[arg-type]</span>
+
+    <span class="n">is_valid_file</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">bool</span><span class="p">],</span> <span class="n">is_valid_file</span><span class="p">)</span>
+
+    <span class="n">instances</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="n">available_classes</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
+    <span class="k">for</span> <span class="n">target_class</span> <span class="ow">in</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">class_to_idx</span><span class="o">.</span><span class="n">keys</span><span class="p">()):</span>
+        <span class="n">class_index</span> <span class="o">=</span> <span class="n">class_to_idx</span><span class="p">[</span><span class="n">target_class</span><span class="p">]</span>
+        <span class="n">target_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">directory</span><span class="p">,</span> <span class="n">target_class</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">target_dir</span><span class="p">):</span>
+            <span class="k">continue</span>
+        <span class="k">for</span> <span class="n">root</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">fnames</span> <span class="ow">in</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">walk</span><span class="p">(</span><span class="n">target_dir</span><span class="p">,</span> <span class="n">followlinks</span><span class="o">=</span><span class="kc">True</span><span class="p">)):</span>
+            <span class="k">for</span> <span class="n">fname</span> <span class="ow">in</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">fnames</span><span class="p">):</span>
+                <span class="n">path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">fname</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">is_valid_file</span><span class="p">(</span><span class="n">path</span><span class="p">):</span>
+                    <span class="n">item</span> <span class="o">=</span> <span class="n">path</span><span class="p">,</span> <span class="n">class_index</span>
+                    <span class="n">instances</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">item</span><span class="p">)</span>
+
+                    <span class="k">if</span> <span class="n">target_class</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">available_classes</span><span class="p">:</span>
+                        <span class="n">available_classes</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">target_class</span><span class="p">)</span>
+
+    <span class="n">empty_classes</span> <span class="o">=</span> <span class="nb">set</span><span class="p">(</span><span class="n">class_to_idx</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span> <span class="o">-</span> <span class="n">available_classes</span>
+    <span class="k">if</span> <span class="n">empty_classes</span><span class="p">:</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;Found no valid file for the classes </span><span class="si">{</span><span class="s1">&#39;, &#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><span class="n">empty_classes</span><span class="p">))</span><span class="si">}</span><span class="s2">. &quot;</span>
+        <span class="k">if</span> <span class="n">extensions</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">msg</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;Supported extensions are: </span><span class="si">{</span><span class="n">extensions</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="nb">isinstance</span><span class="p">(</span><span class="n">extensions</span><span class="p">,</span><span class="w"> </span><span class="nb">str</span><span class="p">)</span><span class="w"> </span><span class="k">else</span><span class="w"> </span><span class="s1">&#39;, &#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">extensions</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">instances</span>
+
+
+<div class="viewcode-block" id="DatasetFolder"><a class="viewcode-back" href="../../../generated/torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder">[docs]</a><span class="k">class</span> <span class="nc">DatasetFolder</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;A generic data loader.</span>
+
+<span class="sd">    This default directory structure can be customized by overriding the</span>
+<span class="sd">    :meth:`find_classes` method.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory path.</span>
+<span class="sd">        loader (callable): A function to load a sample given its path.</span>
+<span class="sd">        extensions (tuple[string]): A list of allowed extensions.</span>
+<span class="sd">            both extensions and is_valid_file should not be passed.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in</span>
+<span class="sd">            a sample and returns a transformed version.</span>
+<span class="sd">            E.g, ``transforms.RandomCrop`` for images.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes</span>
+<span class="sd">            in the target and transforms it.</span>
+<span class="sd">        is_valid_file (callable, optional): A function that takes path of a file</span>
+<span class="sd">            and check if the file is a valid file (used to check of corrupt files)</span>
+<span class="sd">            both extensions and is_valid_file should not be passed.</span>
+
+<span class="sd">     Attributes:</span>
+<span class="sd">        classes (list): List of the class names sorted alphabetically.</span>
+<span class="sd">        class_to_idx (dict): Dict with items (class_name, class_index).</span>
+<span class="sd">        samples (list): List of (sample path, class_index) tuples</span>
+<span class="sd">        targets (list): The class_index value for each image in the dataset</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">loader</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Any</span><span class="p">],</span>
+        <span class="n">extensions</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="o">...</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">is_valid_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="n">classes</span><span class="p">,</span> <span class="n">class_to_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">find_classes</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+        <span class="n">samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">make_dataset</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">class_to_idx</span><span class="p">,</span> <span class="n">extensions</span><span class="p">,</span> <span class="n">is_valid_file</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">loader</span> <span class="o">=</span> <span class="n">loader</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">extensions</span> <span class="o">=</span> <span class="n">extensions</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="n">classes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="n">class_to_idx</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">samples</span> <span class="o">=</span> <span class="n">samples</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="p">[</span><span class="n">s</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">samples</span><span class="p">]</span>
+
+<div class="viewcode-block" id="DatasetFolder.make_dataset"><a class="viewcode-back" href="../../../generated/torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder.make_dataset">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">make_dataset</span><span class="p">(</span>
+        <span class="n">directory</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">class_to_idx</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">extensions</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="o">...</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">is_valid_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Generates a list of samples of a form (path_to_sample, class).</span>
+
+<span class="sd">        This can be overridden to e.g. read files from a compressed zip file instead of from the disk.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            directory (str): root dataset directory, corresponding to ``self.root``.</span>
+<span class="sd">            class_to_idx (Dict[str, int]): Dictionary mapping class name to class index.</span>
+<span class="sd">            extensions (optional): A list of allowed extensions.</span>
+<span class="sd">                Either extensions or is_valid_file should be passed. Defaults to None.</span>
+<span class="sd">            is_valid_file (optional): A function that takes path of a file</span>
+<span class="sd">                and checks if the file is a valid file</span>
+<span class="sd">                (used to check of corrupt files) both extensions and</span>
+<span class="sd">                is_valid_file should not be passed. Defaults to None.</span>
+
+<span class="sd">        Raises:</span>
+<span class="sd">            ValueError: In case ``class_to_idx`` is empty.</span>
+<span class="sd">            ValueError: In case ``extensions`` and ``is_valid_file`` are None or both are not None.</span>
+<span class="sd">            FileNotFoundError: In case no valid file was found for any class.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            List[Tuple[str, int]]: samples of a form (path_to_sample, class)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">class_to_idx</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># prevent potential bug since make_dataset() would use the class_to_idx logic of the</span>
+            <span class="c1"># find_classes() function, instead of using that of the find_classes() method, which</span>
+            <span class="c1"># is potentially overridden and thus could have a different logic.</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The class_to_idx parameter cannot be None.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">make_dataset</span><span class="p">(</span><span class="n">directory</span><span class="p">,</span> <span class="n">class_to_idx</span><span class="p">,</span> <span class="n">extensions</span><span class="o">=</span><span class="n">extensions</span><span class="p">,</span> <span class="n">is_valid_file</span><span class="o">=</span><span class="n">is_valid_file</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetFolder.find_classes"><a class="viewcode-back" href="../../../generated/torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder.find_classes">[docs]</a>    <span class="k">def</span> <span class="nf">find_classes</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">directory</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Find the class folders in a dataset structured as follows::</span>
+
+<span class="sd">            directory/</span>
+<span class="sd">            ├── class_x</span>
+<span class="sd">            │   ├── xxx.ext</span>
+<span class="sd">            │   ├── xxy.ext</span>
+<span class="sd">            │   └── ...</span>
+<span class="sd">            │       └── xxz.ext</span>
+<span class="sd">            └── class_y</span>
+<span class="sd">                ├── 123.ext</span>
+<span class="sd">                ├── nsdf3.ext</span>
+<span class="sd">                └── ...</span>
+<span class="sd">                └── asd932_.ext</span>
+
+<span class="sd">        This method can be overridden to only consider</span>
+<span class="sd">        a subset of classes, or to adapt to a different dataset directory structure.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            directory(str): Root directory path, corresponding to ``self.root``</span>
+
+<span class="sd">        Raises:</span>
+<span class="sd">            FileNotFoundError: If ``dir`` has no class folders.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            (Tuple[List[str], Dict[str, int]]): List of all classes and dictionary mapping each class to an index.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">find_classes</span><span class="p">(</span><span class="n">directory</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (sample, target) where target is class_index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">path</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">sample</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">loader</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">sample</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">sample</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">sample</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">)</span></div>
+
+
+<span class="n">IMG_EXTENSIONS</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;.jpg&quot;</span><span class="p">,</span> <span class="s2">&quot;.jpeg&quot;</span><span class="p">,</span> <span class="s2">&quot;.png&quot;</span><span class="p">,</span> <span class="s2">&quot;.ppm&quot;</span><span class="p">,</span> <span class="s2">&quot;.bmp&quot;</span><span class="p">,</span> <span class="s2">&quot;.pgm&quot;</span><span class="p">,</span> <span class="s2">&quot;.tif&quot;</span><span class="p">,</span> <span class="s2">&quot;.tiff&quot;</span><span class="p">,</span> <span class="s2">&quot;.webp&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">pil_loader</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="c1"># open path as file to avoid ResourceWarning (https://github.com/python-pillow/Pillow/issues/835)</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+
+<span class="c1"># TODO: specify the return type</span>
+<span class="k">def</span> <span class="nf">accimage_loader</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+    <span class="kn">import</span> <span class="nn">accimage</span>
+
+    <span class="k">try</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">accimage</span><span class="o">.</span><span class="n">Image</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+    <span class="k">except</span> <span class="ne">OSError</span><span class="p">:</span>
+        <span class="c1"># Potentially a decoding problem, fall back to PIL.Image</span>
+        <span class="k">return</span> <span class="n">pil_loader</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">default_loader</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+    <span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">get_image_backend</span>
+
+    <span class="k">if</span> <span class="n">get_image_backend</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;accimage&quot;</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">accimage_loader</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">pil_loader</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="ImageFolder"><a class="viewcode-back" href="../../../generated/torchvision.datasets.ImageFolder.html#torchvision.datasets.ImageFolder">[docs]</a><span class="k">class</span> <span class="nc">ImageFolder</span><span class="p">(</span><span class="n">DatasetFolder</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;A generic data loader where the images are arranged in this way by default: ::</span>
+
+<span class="sd">        root/dog/xxx.png</span>
+<span class="sd">        root/dog/xxy.png</span>
+<span class="sd">        root/dog/[...]/xxz.png</span>
+
+<span class="sd">        root/cat/123.png</span>
+<span class="sd">        root/cat/nsdf3.png</span>
+<span class="sd">        root/cat/[...]/asd932_.png</span>
+
+<span class="sd">    This class inherits from :class:`~torchvision.datasets.DatasetFolder` so</span>
+<span class="sd">    the same methods can be overridden to customize the dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory path.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        loader (callable, optional): A function to load an image given its path.</span>
+<span class="sd">        is_valid_file (callable, optional): A function that takes path of an Image file</span>
+<span class="sd">            and check if the file is a valid file (used to check of corrupt files)</span>
+
+<span class="sd">     Attributes:</span>
+<span class="sd">        classes (list): List of the class names sorted alphabetically.</span>
+<span class="sd">        class_to_idx (dict): Dict with items (class_name, class_index).</span>
+<span class="sd">        imgs (list): List of (image path, class_index) tuples</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">loader</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="n">default_loader</span><span class="p">,</span>
+        <span class="n">is_valid_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">root</span><span class="p">,</span>
+            <span class="n">loader</span><span class="p">,</span>
+            <span class="n">IMG_EXTENSIONS</span> <span class="k">if</span> <span class="n">is_valid_file</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span>
+            <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">,</span>
+            <span class="n">is_valid_file</span><span class="o">=</span><span class="n">is_valid_file</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">imgs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/food101.html b/0.16/_modules/torchvision/datasets/food101.html
new file mode 100644
index 00000000000..abcf0a5fc30
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/food101.html
@@ -0,0 +1,762 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.food101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.food101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.food101</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">json</span>
+<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="Food101"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Food101.html#torchvision.datasets.Food101">[docs]</a><span class="k">class</span> <span class="nc">Food101</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`The Food-101 Data Set &lt;https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/&gt;`_.</span>
+
+<span class="sd">    The Food-101 is a challenging data set of 101 food categories with 101,000 images.</span>
+<span class="sd">    For each class, 250 manually reviewed test images are provided as well as 750 training images.</span>
+<span class="sd">    On purpose, the training images were not cleaned, and thus still contain some amount of noise.</span>
+<span class="sd">    This comes mostly in the form of intense colors and sometimes wrong labels. All images were</span>
+<span class="sd">    rescaled to have a maximum side length of 512 pixels.</span>
+
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default) and ``&quot;test&quot;``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again. Default is False.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_URL</span> <span class="o">=</span> <span class="s2">&quot;http://data.vision.ee.ethz.ch/cvl/food-101.tar.gz&quot;</span>
+    <span class="n">_MD5</span> <span class="o">=</span> <span class="s2">&quot;85eeb15f3717b99a5da872d97d918f87&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;food-101&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_meta_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;meta&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;images&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_meta_folder</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">.json&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">metadata</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">loads</span><span class="p">(</span><span class="n">f</span><span class="o">.</span><span class="n">read</span><span class="p">())</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">metadata</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">))))</span>
+
+        <span class="k">for</span> <span class="n">class_label</span><span class="p">,</span> <span class="n">im_rel_paths</span> <span class="ow">in</span> <span class="n">metadata</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">+=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span><span class="p">[</span><span class="n">class_label</span><span class="p">]]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">im_rel_paths</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span> <span class="o">+=</span> <span class="p">[</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span><span class="o">.</span><span class="n">joinpath</span><span class="p">(</span><span class="o">*</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">im_rel_path</span><span class="si">}</span><span class="s2">.jpg&quot;</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;/&quot;</span><span class="p">))</span> <span class="k">for</span> <span class="n">im_rel_path</span> <span class="ow">in</span> <span class="n">im_rel_paths</span>
+            <span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Food101.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Food101.html#torchvision.datasets.Food101.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image_file</span><span class="p">,</span> <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_file</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">label</span></div>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;split=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}</span><span class="s2">&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">(</span><span class="n">folder</span><span class="o">.</span><span class="n">exists</span><span class="p">()</span> <span class="ow">and</span> <span class="n">folder</span><span class="o">.</span><span class="n">is_dir</span><span class="p">()</span> <span class="k">for</span> <span class="n">folder</span> <span class="ow">in</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_meta_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_URL</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_MD5</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/gtsrb.html b/0.16/_modules/torchvision/datasets/gtsrb.html
new file mode 100644
index 00000000000..75d1b2c02ad
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/gtsrb.html
@@ -0,0 +1,772 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.gtsrb &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.gtsrb</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.gtsrb</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">csv</span>
+<span class="kn">import</span> <span class="nn">pathlib</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">make_dataset</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="GTSRB"><a class="viewcode-back" href="../../../generated/torchvision.datasets.GTSRB.html#torchvision.datasets.GTSRB">[docs]</a><span class="k">class</span> <span class="nc">GTSRB</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`German Traffic Sign Recognition Benchmark (GTSRB) &lt;https://benchmark.ini.rub.de/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default), or ``&quot;test&quot;``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;gtsrb&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_target_folder</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;GTSRB&quot;</span> <span class="o">/</span> <span class="p">(</span><span class="s2">&quot;Training&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span> <span class="k">else</span> <span class="s2">&quot;Final_Test/Images&quot;</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="n">samples</span> <span class="o">=</span> <span class="n">make_dataset</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_target_folder</span><span class="p">),</span> <span class="n">extensions</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;.ppm&quot;</span><span class="p">,))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;GT-final_test.csv&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">csv_file</span><span class="p">:</span>
+                <span class="n">samples</span> <span class="o">=</span> <span class="p">[</span>
+                    <span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_target_folder</span> <span class="o">/</span> <span class="n">row</span><span class="p">[</span><span class="s2">&quot;Filename&quot;</span><span class="p">]),</span> <span class="nb">int</span><span class="p">(</span><span class="n">row</span><span class="p">[</span><span class="s2">&quot;ClassId&quot;</span><span class="p">]))</span>
+                    <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">csv</span><span class="o">.</span><span class="n">DictReader</span><span class="p">(</span><span class="n">csv_file</span><span class="p">,</span> <span class="n">delimiter</span><span class="o">=</span><span class="s2">&quot;;&quot;</span><span class="p">,</span> <span class="n">skipinitialspace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span> <span class="o">=</span> <span class="n">samples</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">transform</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="o">=</span> <span class="n">target_transform</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">)</span>
+
+<div class="viewcode-block" id="GTSRB.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.GTSRB.html#torchvision.datasets.GTSRB.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+
+        <span class="n">path</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">sample</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">path</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">sample</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">sample</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">sample</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_target_folder</span><span class="o">.</span><span class="n">is_dir</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">base_url</span> <span class="o">=</span> <span class="s2">&quot;https://sid.erda.dk/public/archives/daaeac0d7ce1152aea9b61d9f1e19370/&quot;</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">base_url</span><span class="si">}</span><span class="s2">GTSRB-Training_fixed.zip&quot;</span><span class="p">,</span>
+                <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+                <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;513f3c79a4c5141765e10e952eaa2478&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">base_url</span><span class="si">}</span><span class="s2">GTSRB_Final_Test_Images.zip&quot;</span><span class="p">,</span>
+                <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+                <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;c7e4e6327067d32654124b0fe9e82185&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">base_url</span><span class="si">}</span><span class="s2">GTSRB_Final_Test_GT.zip&quot;</span><span class="p">,</span>
+                <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+                <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;fe31e9c9270bbcd7b84b7f21a9d9d9e5&quot;</span><span class="p">,</span>
+            <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/hmdb51.html b/0.16/_modules/torchvision/datasets/hmdb51.html
new file mode 100644
index 00000000000..c16c147ceef
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/hmdb51.html
@@ -0,0 +1,820 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.hmdb51 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.hmdb51</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.hmdb51</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">glob</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">find_classes</span><span class="p">,</span> <span class="n">make_dataset</span>
+<span class="kn">from</span> <span class="nn">.video_utils</span> <span class="kn">import</span> <span class="n">VideoClips</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="HMDB51"><a class="viewcode-back" href="../../../generated/torchvision.datasets.HMDB51.html#torchvision.datasets.HMDB51">[docs]</a><span class="k">class</span> <span class="nc">HMDB51</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    `HMDB51 &lt;https://serre-lab.clps.brown.edu/resource/hmdb-a-large-human-motion-database/&gt;`_</span>
+<span class="sd">    dataset.</span>
+
+<span class="sd">    HMDB51 is an action recognition video dataset.</span>
+<span class="sd">    This dataset consider every video as a collection of video clips of fixed size, specified</span>
+<span class="sd">    by ``frames_per_clip``, where the step in frames between each clip is given by</span>
+<span class="sd">    ``step_between_clips``.</span>
+
+<span class="sd">    To give an example, for 2 videos with 10 and 15 frames respectively, if ``frames_per_clip=5``</span>
+<span class="sd">    and ``step_between_clips=5``, the dataset size will be (2 + 3) = 5, where the first two</span>
+<span class="sd">    elements will come from video 1, and the next three elements from video 2.</span>
+<span class="sd">    Note that we drop clips which do not have exactly ``frames_per_clip`` elements, so not all</span>
+<span class="sd">    frames in a video might be present.</span>
+
+<span class="sd">    Internally, it uses a VideoClips object to handle clip creation.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the HMDB51 Dataset.</span>
+<span class="sd">        annotation_path (str): Path to the folder containing the split files.</span>
+<span class="sd">        frames_per_clip (int): Number of frames in a clip.</span>
+<span class="sd">        step_between_clips (int): Number of frames between each clip.</span>
+<span class="sd">        fold (int, optional): Which fold to use. Should be between 1 and 3.</span>
+<span class="sd">        train (bool, optional): If ``True``, creates a dataset from the train split,</span>
+<span class="sd">            otherwise from the ``test`` split.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in a TxHxWxC video</span>
+<span class="sd">            and returns a transformed version.</span>
+<span class="sd">        output_format (str, optional): The format of the output video tensors (before transforms).</span>
+<span class="sd">            Can be either &quot;THWC&quot; (default) or &quot;TCHW&quot;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        tuple: A 3-tuple with the following entries:</span>
+
+<span class="sd">            - video (Tensor[T, H, W, C] or Tensor[T, C, H, W]): The `T` video frames</span>
+<span class="sd">            - audio(Tensor[K, L]): the audio frames, where `K` is the number of channels</span>
+<span class="sd">              and `L` is the number of points</span>
+<span class="sd">            - label (int): class of the video clip</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">data_url</span> <span class="o">=</span> <span class="s2">&quot;https://serre-lab.clps.brown.edu/wp-content/uploads/2013/10/hmdb51_org.rar&quot;</span>
+    <span class="n">splits</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;https://serre-lab.clps.brown.edu/wp-content/uploads/2013/10/test_train_splits.rar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;15e67781e70dcfbdce2d7dbb9b3344b5&quot;</span><span class="p">,</span>
+    <span class="p">}</span>
+    <span class="n">TRAIN_TAG</span> <span class="o">=</span> <span class="mi">1</span>
+    <span class="n">TEST_TAG</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">annotation_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">frames_per_clip</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">step_between_clips</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">frame_rate</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">fold</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">_precomputed_metadata</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">num_workers</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">_video_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_video_height</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_video_min_dimension</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_audio_samples</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">output_format</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;THWC&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">fold</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;fold should be between 1 and 3, got </span><span class="si">{</span><span class="n">fold</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="n">extensions</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;avi&quot;</span><span class="p">,)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="n">class_to_idx</span> <span class="o">=</span> <span class="n">find_classes</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">samples</span> <span class="o">=</span> <span class="n">make_dataset</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span>
+            <span class="n">class_to_idx</span><span class="p">,</span>
+            <span class="n">extensions</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="n">video_paths</span> <span class="o">=</span> <span class="p">[</span><span class="n">path</span> <span class="k">for</span> <span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">_</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">]</span>
+        <span class="n">video_clips</span> <span class="o">=</span> <span class="n">VideoClips</span><span class="p">(</span>
+            <span class="n">video_paths</span><span class="p">,</span>
+            <span class="n">frames_per_clip</span><span class="p">,</span>
+            <span class="n">step_between_clips</span><span class="p">,</span>
+            <span class="n">frame_rate</span><span class="p">,</span>
+            <span class="n">_precomputed_metadata</span><span class="p">,</span>
+            <span class="n">num_workers</span><span class="o">=</span><span class="n">num_workers</span><span class="p">,</span>
+            <span class="n">_video_width</span><span class="o">=</span><span class="n">_video_width</span><span class="p">,</span>
+            <span class="n">_video_height</span><span class="o">=</span><span class="n">_video_height</span><span class="p">,</span>
+            <span class="n">_video_min_dimension</span><span class="o">=</span><span class="n">_video_min_dimension</span><span class="p">,</span>
+            <span class="n">_audio_samples</span><span class="o">=</span><span class="n">_audio_samples</span><span class="p">,</span>
+            <span class="n">output_format</span><span class="o">=</span><span class="n">output_format</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="c1"># we bookkeep the full version of video clips because we want to be able</span>
+        <span class="c1"># to return the metadata of full version rather than the subset version of</span>
+        <span class="c1"># video clips</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">full_video_clips</span> <span class="o">=</span> <span class="n">video_clips</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fold</span> <span class="o">=</span> <span class="n">fold</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="o">=</span> <span class="n">train</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">indices</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_select_fold</span><span class="p">(</span><span class="n">video_paths</span><span class="p">,</span> <span class="n">annotation_path</span><span class="p">,</span> <span class="n">fold</span><span class="p">,</span> <span class="n">train</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span> <span class="o">=</span> <span class="n">video_clips</span><span class="o">.</span><span class="n">subset</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">indices</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">transform</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">metadata</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">full_video_clips</span><span class="o">.</span><span class="n">metadata</span>
+
+    <span class="k">def</span> <span class="nf">_select_fold</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">video_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">annotations_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">fold</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+        <span class="n">target_tag</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">TRAIN_TAG</span> <span class="k">if</span> <span class="n">train</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">TEST_TAG</span>
+        <span class="n">split_pattern_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;*test_split</span><span class="si">{</span><span class="n">fold</span><span class="si">}</span><span class="s2">.txt&quot;</span>
+        <span class="n">split_pattern_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">annotations_dir</span><span class="p">,</span> <span class="n">split_pattern_name</span><span class="p">)</span>
+        <span class="n">annotation_paths</span> <span class="o">=</span> <span class="n">glob</span><span class="o">.</span><span class="n">glob</span><span class="p">(</span><span class="n">split_pattern_path</span><span class="p">)</span>
+        <span class="n">selected_files</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">filepath</span> <span class="ow">in</span> <span class="n">annotation_paths</span><span class="p">:</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">filepath</span><span class="p">)</span> <span class="k">as</span> <span class="n">fid</span><span class="p">:</span>
+                <span class="n">lines</span> <span class="o">=</span> <span class="n">fid</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">:</span>
+                <span class="n">video_filename</span><span class="p">,</span> <span class="n">tag_string</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">()</span>
+                <span class="n">tag</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">tag_string</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">tag</span> <span class="o">==</span> <span class="n">target_tag</span><span class="p">:</span>
+                    <span class="n">selected_files</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">video_filename</span><span class="p">)</span>
+
+        <span class="n">indices</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">video_index</span><span class="p">,</span> <span class="n">video_path</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">video_list</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">video_path</span><span class="p">)</span> <span class="ow">in</span> <span class="n">selected_files</span><span class="p">:</span>
+                <span class="n">indices</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">video_index</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">indices</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">num_clips</span><span class="p">()</span>
+
+<div class="viewcode-block" id="HMDB51.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.HMDB51.html#torchvision.datasets.HMDB51.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">video_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">get_clip</span><span class="p">(</span><span class="n">idx</span><span class="p">)</span>
+        <span class="n">sample_index</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">indices</span><span class="p">[</span><span class="n">video_idx</span><span class="p">]</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">class_index</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">[</span><span class="n">sample_index</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">class_index</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/imagenet.html b/0.16/_modules/torchvision/datasets/imagenet.html
new file mode 100644
index 00000000000..e4db71fa2b7
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/imagenet.html
@@ -0,0 +1,887 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.imagenet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.imagenet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.imagenet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">shutil</span>
+<span class="kn">import</span> <span class="nn">tempfile</span>
+<span class="kn">from</span> <span class="nn">contextlib</span> <span class="kn">import</span> <span class="n">contextmanager</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Iterator</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">ImageFolder</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+
+<span class="n">ARCHIVE_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;ILSVRC2012_img_train.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;1d675b47d978889d74fa0da5fadfb00e&quot;</span><span class="p">),</span>
+    <span class="s2">&quot;val&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;ILSVRC2012_img_val.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;29b22e2961454d5413ddabcf34fc5622&quot;</span><span class="p">),</span>
+    <span class="s2">&quot;devkit&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;ILSVRC2012_devkit_t12.tar.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;fa75699e90414af021442c21a62c3abf&quot;</span><span class="p">),</span>
+<span class="p">}</span>
+
+<span class="n">META_FILE</span> <span class="o">=</span> <span class="s2">&quot;meta.bin&quot;</span>
+
+
+<div class="viewcode-block" id="ImageNet"><a class="viewcode-back" href="../../../generated/torchvision.datasets.ImageNet.html#torchvision.datasets.ImageNet">[docs]</a><span class="k">class</span> <span class="nc">ImageNet</span><span class="p">(</span><span class="n">ImageFolder</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`ImageNet &lt;http://image-net.org/&gt;`_ 2012 Classification Dataset.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Before using this class, it is required to download ImageNet 2012 dataset from</span>
+<span class="sd">        `here &lt;https://image-net.org/challenges/LSVRC/2012/2012-downloads.php&gt;`_ and</span>
+<span class="sd">        place the files ``ILSVRC2012_devkit_t12.tar.gz`` and ``ILSVRC2012_img_train.tar``</span>
+<span class="sd">        or ``ILSVRC2012_img_val.tar`` based on ``split`` in the root directory.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the ImageNet Dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``train``, or ``val``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        loader (callable, optional): A function to load an image given its path.</span>
+
+<span class="sd">     Attributes:</span>
+<span class="sd">        classes (list): List of the class name tuples.</span>
+<span class="sd">        class_to_idx (dict): Dict with items (class_name, class_index).</span>
+<span class="sd">        wnids (list): List of the WordNet IDs.</span>
+<span class="sd">        wnid_to_idx (dict): Dict with items (wordnet_id, class_index).</span>
+<span class="sd">        imgs (list): List of (image path, class_index) tuples</span>
+<span class="sd">        targets (list): The class_index value for each image in the dataset</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">root</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">expanduser</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">parse_archives</span><span class="p">()</span>
+        <span class="n">wnid_to_classes</span> <span class="o">=</span> <span class="n">load_meta_file</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">root</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">wnids</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">wnid_to_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="n">wnid_to_classes</span><span class="p">[</span><span class="n">wnid</span><span class="p">]</span> <span class="k">for</span> <span class="n">wnid</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">wnids</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="bp">cls</span><span class="p">:</span> <span class="n">idx</span> <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">clss</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">)</span> <span class="k">for</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="n">clss</span><span class="p">}</span>
+
+    <span class="k">def</span> <span class="nf">parse_archives</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">META_FILE</span><span class="p">)):</span>
+            <span class="n">parse_devkit_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+                <span class="n">parse_train_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;val&quot;</span><span class="p">:</span>
+                <span class="n">parse_val_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">split_folder</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;Split: </span><span class="si">{split}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">load_meta_file</span><span class="p">(</span><span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+    <span class="k">if</span> <span class="n">file</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">file</span> <span class="o">=</span> <span class="n">META_FILE</span>
+    <span class="n">file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">file</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">file</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="s2">&quot;The meta file </span><span class="si">{}</span><span class="s2"> is not present in the root directory or is corrupted. &quot;</span>
+            <span class="s2">&quot;This file is automatically created by the ImageNet dataset.&quot;</span>
+        <span class="p">)</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">file</span><span class="p">,</span> <span class="n">root</span><span class="p">))</span>
+
+
+<span class="k">def</span> <span class="nf">_verify_archive</span><span class="p">(</span><span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">file</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">md5</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">),</span> <span class="n">md5</span><span class="p">):</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="s2">&quot;The archive </span><span class="si">{}</span><span class="s2"> is not present in the root directory or is corrupted. &quot;</span>
+            <span class="s2">&quot;You need to download it externally and place it in </span><span class="si">{}</span><span class="s2">.&quot;</span>
+        <span class="p">)</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">file</span><span class="p">,</span> <span class="n">root</span><span class="p">))</span>
+
+
+<span class="k">def</span> <span class="nf">parse_devkit_archive</span><span class="p">(</span><span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Parse the devkit archive of the ImageNet2012 classification dataset and save</span>
+<span class="sd">    the meta information in a binary file.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (str): Root directory containing the devkit archive</span>
+<span class="sd">        file (str, optional): Name of devkit archive. Defaults to</span>
+<span class="sd">            &#39;ILSVRC2012_devkit_t12.tar.gz&#39;</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="kn">import</span> <span class="nn">scipy.io</span> <span class="k">as</span> <span class="nn">sio</span>
+
+    <span class="k">def</span> <span class="nf">parse_meta_mat</span><span class="p">(</span><span class="n">devkit_root</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="o">...</span><span class="p">]]]:</span>
+        <span class="n">metafile</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">devkit_root</span><span class="p">,</span> <span class="s2">&quot;data&quot;</span><span class="p">,</span> <span class="s2">&quot;meta.mat&quot;</span><span class="p">)</span>
+        <span class="n">meta</span> <span class="o">=</span> <span class="n">sio</span><span class="o">.</span><span class="n">loadmat</span><span class="p">(</span><span class="n">metafile</span><span class="p">,</span> <span class="n">squeeze_me</span><span class="o">=</span><span class="kc">True</span><span class="p">)[</span><span class="s2">&quot;synsets&quot;</span><span class="p">]</span>
+        <span class="n">nums_children</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="o">*</span><span class="n">meta</span><span class="p">))[</span><span class="mi">4</span><span class="p">]</span>
+        <span class="n">meta</span> <span class="o">=</span> <span class="p">[</span><span class="n">meta</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span> <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">num_children</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">nums_children</span><span class="p">)</span> <span class="k">if</span> <span class="n">num_children</span> <span class="o">==</span> <span class="mi">0</span><span class="p">]</span>
+        <span class="n">idcs</span><span class="p">,</span> <span class="n">wnids</span><span class="p">,</span> <span class="n">classes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="o">*</span><span class="n">meta</span><span class="p">))[:</span><span class="mi">3</span><span class="p">]</span>
+        <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="nb">tuple</span><span class="p">(</span><span class="n">clss</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;, &quot;</span><span class="p">))</span> <span class="k">for</span> <span class="n">clss</span> <span class="ow">in</span> <span class="n">classes</span><span class="p">]</span>
+        <span class="n">idx_to_wnid</span> <span class="o">=</span> <span class="p">{</span><span class="n">idx</span><span class="p">:</span> <span class="n">wnid</span> <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">wnid</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">idcs</span><span class="p">,</span> <span class="n">wnids</span><span class="p">)}</span>
+        <span class="n">wnid_to_classes</span> <span class="o">=</span> <span class="p">{</span><span class="n">wnid</span><span class="p">:</span> <span class="n">clss</span> <span class="k">for</span> <span class="n">wnid</span><span class="p">,</span> <span class="n">clss</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">wnids</span><span class="p">,</span> <span class="n">classes</span><span class="p">)}</span>
+        <span class="k">return</span> <span class="n">idx_to_wnid</span><span class="p">,</span> <span class="n">wnid_to_classes</span>
+
+    <span class="k">def</span> <span class="nf">parse_val_groundtruth_txt</span><span class="p">(</span><span class="n">devkit_root</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+        <span class="n">file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">devkit_root</span><span class="p">,</span> <span class="s2">&quot;data&quot;</span><span class="p">,</span> <span class="s2">&quot;ILSVRC2012_validation_ground_truth.txt&quot;</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file</span><span class="p">)</span> <span class="k">as</span> <span class="n">txtfh</span><span class="p">:</span>
+            <span class="n">val_idcs</span> <span class="o">=</span> <span class="n">txtfh</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+        <span class="k">return</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">val_idx</span><span class="p">)</span> <span class="k">for</span> <span class="n">val_idx</span> <span class="ow">in</span> <span class="n">val_idcs</span><span class="p">]</span>
+
+    <span class="nd">@contextmanager</span>
+    <span class="k">def</span> <span class="nf">get_tmp_dir</span><span class="p">()</span> <span class="o">-&gt;</span> <span class="n">Iterator</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="n">tmp_dir</span> <span class="o">=</span> <span class="n">tempfile</span><span class="o">.</span><span class="n">mkdtemp</span><span class="p">()</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="k">yield</span> <span class="n">tmp_dir</span>
+        <span class="k">finally</span><span class="p">:</span>
+            <span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span><span class="p">(</span><span class="n">tmp_dir</span><span class="p">)</span>
+
+    <span class="n">archive_meta</span> <span class="o">=</span> <span class="n">ARCHIVE_META</span><span class="p">[</span><span class="s2">&quot;devkit&quot;</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">file</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">file</span> <span class="o">=</span> <span class="n">archive_meta</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">md5</span> <span class="o">=</span> <span class="n">archive_meta</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+
+    <span class="n">_verify_archive</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+
+    <span class="k">with</span> <span class="n">get_tmp_dir</span><span class="p">()</span> <span class="k">as</span> <span class="n">tmp_dir</span><span class="p">:</span>
+        <span class="n">extract_archive</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">),</span> <span class="n">tmp_dir</span><span class="p">)</span>
+
+        <span class="n">devkit_root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">tmp_dir</span><span class="p">,</span> <span class="s2">&quot;ILSVRC2012_devkit_t12&quot;</span><span class="p">)</span>
+        <span class="n">idx_to_wnid</span><span class="p">,</span> <span class="n">wnid_to_classes</span> <span class="o">=</span> <span class="n">parse_meta_mat</span><span class="p">(</span><span class="n">devkit_root</span><span class="p">)</span>
+        <span class="n">val_idcs</span> <span class="o">=</span> <span class="n">parse_val_groundtruth_txt</span><span class="p">(</span><span class="n">devkit_root</span><span class="p">)</span>
+        <span class="n">val_wnids</span> <span class="o">=</span> <span class="p">[</span><span class="n">idx_to_wnid</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span> <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="n">val_idcs</span><span class="p">]</span>
+
+        <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">((</span><span class="n">wnid_to_classes</span><span class="p">,</span> <span class="n">val_wnids</span><span class="p">),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">META_FILE</span><span class="p">))</span>
+
+
+<span class="k">def</span> <span class="nf">parse_train_archive</span><span class="p">(</span><span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">folder</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Parse the train images archive of the ImageNet2012 classification dataset and</span>
+<span class="sd">    prepare it for usage with the ImageNet dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (str): Root directory containing the train images archive</span>
+<span class="sd">        file (str, optional): Name of train images archive. Defaults to</span>
+<span class="sd">            &#39;ILSVRC2012_img_train.tar&#39;</span>
+<span class="sd">        folder (str, optional): Optional name for train images folder. Defaults to</span>
+<span class="sd">            &#39;train&#39;</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">archive_meta</span> <span class="o">=</span> <span class="n">ARCHIVE_META</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">file</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">file</span> <span class="o">=</span> <span class="n">archive_meta</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">md5</span> <span class="o">=</span> <span class="n">archive_meta</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+
+    <span class="n">_verify_archive</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+
+    <span class="n">train_root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">folder</span><span class="p">)</span>
+    <span class="n">extract_archive</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">),</span> <span class="n">train_root</span><span class="p">)</span>
+
+    <span class="n">archives</span> <span class="o">=</span> <span class="p">[</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">train_root</span><span class="p">,</span> <span class="n">archive</span><span class="p">)</span> <span class="k">for</span> <span class="n">archive</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">train_root</span><span class="p">)]</span>
+    <span class="k">for</span> <span class="n">archive</span> <span class="ow">in</span> <span class="n">archives</span><span class="p">:</span>
+        <span class="n">extract_archive</span><span class="p">(</span><span class="n">archive</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">splitext</span><span class="p">(</span><span class="n">archive</span><span class="p">)[</span><span class="mi">0</span><span class="p">],</span> <span class="n">remove_finished</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">parse_val_archive</span><span class="p">(</span>
+    <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">wnids</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">folder</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;val&quot;</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Parse the validation images archive of the ImageNet2012 classification dataset</span>
+<span class="sd">    and prepare it for usage with the ImageNet dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (str): Root directory containing the validation images archive</span>
+<span class="sd">        file (str, optional): Name of validation images archive. Defaults to</span>
+<span class="sd">            &#39;ILSVRC2012_img_val.tar&#39;</span>
+<span class="sd">        wnids (list, optional): List of WordNet IDs of the validation images. If None</span>
+<span class="sd">            is given, the IDs are loaded from the meta file in the root directory</span>
+<span class="sd">        folder (str, optional): Optional name for validation images folder. Defaults to</span>
+<span class="sd">            &#39;val&#39;</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">archive_meta</span> <span class="o">=</span> <span class="n">ARCHIVE_META</span><span class="p">[</span><span class="s2">&quot;val&quot;</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">file</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">file</span> <span class="o">=</span> <span class="n">archive_meta</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">md5</span> <span class="o">=</span> <span class="n">archive_meta</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">wnids</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">wnids</span> <span class="o">=</span> <span class="n">load_meta_file</span><span class="p">(</span><span class="n">root</span><span class="p">)[</span><span class="mi">1</span><span class="p">]</span>
+
+    <span class="n">_verify_archive</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+
+    <span class="n">val_root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">folder</span><span class="p">)</span>
+    <span class="n">extract_archive</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">),</span> <span class="n">val_root</span><span class="p">)</span>
+
+    <span class="n">images</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">val_root</span><span class="p">,</span> <span class="n">image</span><span class="p">)</span> <span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">val_root</span><span class="p">))</span>
+
+    <span class="k">for</span> <span class="n">wnid</span> <span class="ow">in</span> <span class="nb">set</span><span class="p">(</span><span class="n">wnids</span><span class="p">):</span>
+        <span class="n">os</span><span class="o">.</span><span class="n">mkdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">val_root</span><span class="p">,</span> <span class="n">wnid</span><span class="p">))</span>
+
+    <span class="k">for</span> <span class="n">wnid</span><span class="p">,</span> <span class="n">img_file</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">wnids</span><span class="p">,</span> <span class="n">images</span><span class="p">):</span>
+        <span class="n">shutil</span><span class="o">.</span><span class="n">move</span><span class="p">(</span><span class="n">img_file</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">val_root</span><span class="p">,</span> <span class="n">wnid</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">img_file</span><span class="p">)))</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/inaturalist.html b/0.16/_modules/torchvision/datasets/inaturalist.html
new file mode 100644
index 00000000000..1953205206c
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/inaturalist.html
@@ -0,0 +1,910 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.inaturalist &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.inaturalist</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.inaturalist</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+<span class="n">CATEGORIES_2021</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;kingdom&quot;</span><span class="p">,</span> <span class="s2">&quot;phylum&quot;</span><span class="p">,</span> <span class="s2">&quot;class&quot;</span><span class="p">,</span> <span class="s2">&quot;order&quot;</span><span class="p">,</span> <span class="s2">&quot;family&quot;</span><span class="p">,</span> <span class="s2">&quot;genus&quot;</span><span class="p">]</span>
+
+<span class="n">DATASET_URLS</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;2017&quot;</span><span class="p">:</span> <span class="s2">&quot;https://ml-inat-competition-datasets.s3.amazonaws.com/2017/train_val_images.tar.gz&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2018&quot;</span><span class="p">:</span> <span class="s2">&quot;https://ml-inat-competition-datasets.s3.amazonaws.com/2018/train_val2018.tar.gz&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2019&quot;</span><span class="p">:</span> <span class="s2">&quot;https://ml-inat-competition-datasets.s3.amazonaws.com/2019/train_val2019.tar.gz&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2021_train&quot;</span><span class="p">:</span> <span class="s2">&quot;https://ml-inat-competition-datasets.s3.amazonaws.com/2021/train.tar.gz&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2021_train_mini&quot;</span><span class="p">:</span> <span class="s2">&quot;https://ml-inat-competition-datasets.s3.amazonaws.com/2021/train_mini.tar.gz&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2021_valid&quot;</span><span class="p">:</span> <span class="s2">&quot;https://ml-inat-competition-datasets.s3.amazonaws.com/2021/val.tar.gz&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+<span class="n">DATASET_MD5</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;2017&quot;</span><span class="p">:</span> <span class="s2">&quot;7c784ea5e424efaec655bd392f87301f&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2018&quot;</span><span class="p">:</span> <span class="s2">&quot;b1c6952ce38f31868cc50ea72d066cc3&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2019&quot;</span><span class="p">:</span> <span class="s2">&quot;c60a6e2962c9b8ccbd458d12c8582644&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2021_train&quot;</span><span class="p">:</span> <span class="s2">&quot;e0526d53c7f7b2e3167b2b43bb2690ed&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2021_train_mini&quot;</span><span class="p">:</span> <span class="s2">&quot;db6ed8330e634445efc8fec83ae81442&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;2021_valid&quot;</span><span class="p">:</span> <span class="s2">&quot;f6f6e0e242e3d4c9569ba56400938afc&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="INaturalist"><a class="viewcode-back" href="../../../generated/torchvision.datasets.INaturalist.html#torchvision.datasets.INaturalist">[docs]</a><span class="k">class</span> <span class="nc">INaturalist</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`iNaturalist &lt;https://github.com/visipedia/inat_comp&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where the image files are stored.</span>
+<span class="sd">            This class does not require/use annotation files.</span>
+<span class="sd">        version (string, optional): Which version of the dataset to download/use. One of</span>
+<span class="sd">            &#39;2017&#39;, &#39;2018&#39;, &#39;2019&#39;, &#39;2021_train&#39;, &#39;2021_train_mini&#39;, &#39;2021_valid&#39;.</span>
+<span class="sd">            Default: `2021_train`.</span>
+<span class="sd">        target_type (string or list, optional): Type of target to use, for 2021 versions, one of:</span>
+
+<span class="sd">            - ``full``: the full category (species)</span>
+<span class="sd">            - ``kingdom``: e.g. &quot;Animalia&quot;</span>
+<span class="sd">            - ``phylum``: e.g. &quot;Arthropoda&quot;</span>
+<span class="sd">            - ``class``: e.g. &quot;Insecta&quot;</span>
+<span class="sd">            - ``order``: e.g. &quot;Coleoptera&quot;</span>
+<span class="sd">            - ``family``: e.g. &quot;Cleridae&quot;</span>
+<span class="sd">            - ``genus``: e.g. &quot;Trichodes&quot;</span>
+
+<span class="sd">            for 2017-2019 versions, one of:</span>
+
+<span class="sd">            - ``full``: the full (numeric) category</span>
+<span class="sd">            - ``super``: the super category, e.g. &quot;Amphibians&quot;</span>
+
+<span class="sd">            Can also be a list to output a tuple with all specified target types.</span>
+<span class="sd">            Defaults to ``full``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">version</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;2021_train&quot;</span><span class="p">,</span>
+        <span class="n">target_type</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;full&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">version</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">version</span><span class="p">,</span> <span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="n">DATASET_URLS</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">version</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="c1"># map: category type -&gt; name of category -&gt; index</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="c1"># list indexed by category id, containing mapping from category type -&gt; index</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">target_type</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="n">target_type</span> <span class="o">=</span> <span class="p">[</span><span class="n">target_type</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="p">[:</span><span class="mi">4</span><span class="p">]</span> <span class="o">==</span> <span class="s2">&quot;2021&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="o">=</span> <span class="p">[</span><span class="n">verify_str_arg</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="s2">&quot;target_type&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;full&quot;</span><span class="p">,</span> <span class="o">*</span><span class="n">CATEGORIES_2021</span><span class="p">))</span> <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">target_type</span><span class="p">]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_init_2021</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span> <span class="o">=</span> <span class="p">[</span><span class="n">verify_str_arg</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="s2">&quot;target_type&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;full&quot;</span><span class="p">,</span> <span class="s2">&quot;super&quot;</span><span class="p">))</span> <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">target_type</span><span class="p">]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_init_pre2021</span><span class="p">()</span>
+
+        <span class="c1"># index of all files: (full category id, filename)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">dir_index</span><span class="p">,</span> <span class="n">dir_name</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span><span class="p">):</span>
+            <span class="n">files</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">dir_name</span><span class="p">))</span>
+            <span class="k">for</span> <span class="n">fname</span> <span class="ow">in</span> <span class="n">files</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">dir_index</span><span class="p">,</span> <span class="n">fname</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_init_2021</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Initialize based on 2021 layout&quot;&quot;&quot;</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">))</span>
+
+        <span class="c1"># map: category type -&gt; name of category -&gt; index</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="p">{}</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">CATEGORIES_2021</span><span class="p">}</span>
+
+        <span class="k">for</span> <span class="n">dir_index</span><span class="p">,</span> <span class="n">dir_name</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span><span class="p">):</span>
+            <span class="n">pieces</span> <span class="o">=</span> <span class="n">dir_name</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;_&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">pieces</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">8</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unexpected category name </span><span class="si">{</span><span class="n">dir_name</span><span class="si">}</span><span class="s2">, wrong number of pieces&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">pieces</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">dir_index</span><span class="si">:</span><span class="s2">05d</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unexpected category id </span><span class="si">{</span><span class="n">pieces</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">, expecting </span><span class="si">{</span><span class="n">dir_index</span><span class="si">:</span><span class="s2">05d</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="n">cat_map</span> <span class="o">=</span> <span class="p">{}</span>
+            <span class="k">for</span> <span class="n">cat</span><span class="p">,</span> <span class="n">name</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">CATEGORIES_2021</span><span class="p">,</span> <span class="n">pieces</span><span class="p">[</span><span class="mi">1</span><span class="p">:</span><span class="mi">7</span><span class="p">]):</span>
+                <span class="k">if</span> <span class="n">name</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">[</span><span class="n">cat</span><span class="p">]:</span>
+                    <span class="n">cat_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">[</span><span class="n">cat</span><span class="p">][</span><span class="n">name</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">cat_id</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">[</span><span class="n">cat</span><span class="p">])</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">[</span><span class="n">cat</span><span class="p">][</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">cat_id</span>
+                <span class="n">cat_map</span><span class="p">[</span><span class="n">cat</span><span class="p">]</span> <span class="o">=</span> <span class="n">cat_id</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cat_map</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_init_pre2021</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Initialize based on 2017-2019 layout&quot;&quot;&quot;</span>
+
+        <span class="c1"># map: category type -&gt; name of category -&gt; index</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;super&quot;</span><span class="p">:</span> <span class="p">{}}</span>
+
+        <span class="n">cat_index</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="n">super_categories</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">))</span>
+        <span class="k">for</span> <span class="n">sindex</span><span class="p">,</span> <span class="n">scat</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">super_categories</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">[</span><span class="s2">&quot;super&quot;</span><span class="p">][</span><span class="n">scat</span><span class="p">]</span> <span class="o">=</span> <span class="n">sindex</span>
+            <span class="n">subcategories</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">scat</span><span class="p">)))</span>
+            <span class="k">for</span> <span class="n">subcat</span> <span class="ow">in</span> <span class="n">subcategories</span><span class="p">:</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">version</span> <span class="o">==</span> <span class="s2">&quot;2017&quot;</span><span class="p">:</span>
+                    <span class="c1"># this version does not use ids as directory names</span>
+                    <span class="n">subcat_i</span> <span class="o">=</span> <span class="n">cat_index</span>
+                    <span class="n">cat_index</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">try</span><span class="p">:</span>
+                        <span class="n">subcat_i</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">subcat</span><span class="p">)</span>
+                    <span class="k">except</span> <span class="ne">ValueError</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unexpected non-numeric dir name: </span><span class="si">{</span><span class="n">subcat</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">subcat_i</span> <span class="o">&gt;=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="p">):</span>
+                    <span class="n">old_len</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="p">)</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="o">.</span><span class="n">extend</span><span class="p">([{}]</span> <span class="o">*</span> <span class="p">(</span><span class="n">subcat_i</span> <span class="o">-</span> <span class="n">old_len</span> <span class="o">+</span> <span class="mi">1</span><span class="p">))</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="s2">&quot;&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="n">subcat_i</span> <span class="o">-</span> <span class="n">old_len</span> <span class="o">+</span> <span class="mi">1</span><span class="p">))</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="p">[</span><span class="n">subcat_i</span><span class="p">]:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Duplicate category </span><span class="si">{</span><span class="n">subcat</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="p">[</span><span class="n">subcat_i</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;super&quot;</span><span class="p">:</span> <span class="n">sindex</span><span class="p">}</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span><span class="p">[</span><span class="n">subcat_i</span><span class="p">]</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">scat</span><span class="p">,</span> <span class="n">subcat</span><span class="p">)</span>
+
+        <span class="c1"># validate the dictionary</span>
+        <span class="k">for</span> <span class="n">cindex</span><span class="p">,</span> <span class="n">c</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="p">):</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">c</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Missing category </span><span class="si">{</span><span class="n">cindex</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="INaturalist.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.INaturalist.html#torchvision.datasets.INaturalist.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where the type of target specified by target_type.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">cat_id</span><span class="p">,</span> <span class="n">fname</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span><span class="p">[</span><span class="n">cat_id</span><span class="p">],</span> <span class="n">fname</span><span class="p">))</span>
+
+        <span class="n">target</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_type</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">t</span> <span class="o">==</span> <span class="s2">&quot;full&quot;</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cat_id</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">categories_map</span><span class="p">[</span><span class="n">cat_id</span><span class="p">][</span><span class="n">t</span><span class="p">])</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">target</span><span class="p">)</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">target</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">target</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">index</span><span class="p">)</span>
+
+<div class="viewcode-block" id="INaturalist.category_name"><a class="viewcode-back" href="../../../generated/torchvision.datasets.INaturalist.html#torchvision.datasets.INaturalist.category_name">[docs]</a>    <span class="k">def</span> <span class="nf">category_name</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">category_type</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">category_id</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            category_type(str): one of &quot;full&quot;, &quot;kingdom&quot;, &quot;phylum&quot;, &quot;class&quot;, &quot;order&quot;, &quot;family&quot;, &quot;genus&quot; or &quot;super&quot;</span>
+<span class="sd">            category_id(int): an index (class id) from this category</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            the name of the category</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">category_type</span> <span class="o">==</span> <span class="s2">&quot;full&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">all_categories</span><span class="p">[</span><span class="n">category_id</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">category_type</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid category type &#39;</span><span class="si">{</span><span class="n">category_type</span><span class="si">}</span><span class="s2">&#39;&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="nb">id</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">categories_index</span><span class="p">[</span><span class="n">category_type</span><span class="p">]</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                    <span class="k">if</span> <span class="nb">id</span> <span class="o">==</span> <span class="n">category_id</span><span class="p">:</span>
+                        <span class="k">return</span> <span class="n">name</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid category id </span><span class="si">{</span><span class="n">category_id</span><span class="si">}</span><span class="s2"> for </span><span class="si">{</span><span class="n">category_type</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">))</span> <span class="o">&gt;</span> <span class="mi">0</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The directory </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="si">}</span><span class="s2"> already exists. &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;If you want to re-download or re-extract the images, delete the directory.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">base_root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">dirname</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="n">DATASET_URLS</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="p">],</span> <span class="n">base_root</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="si">}</span><span class="s2">.tgz&quot;</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">DATASET_MD5</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="p">]</span>
+        <span class="p">)</span>
+
+        <span class="n">orig_dir_name</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">base_root</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">DATASET_URLS</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="p">])</span><span class="o">.</span><span class="n">rstrip</span><span class="p">(</span><span class="s2">&quot;.tar.gz&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">orig_dir_name</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unable to find downloaded files at </span><span class="si">{</span><span class="n">orig_dir_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">os</span><span class="o">.</span><span class="n">rename</span><span class="p">(</span><span class="n">orig_dir_name</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+        <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Dataset version &#39;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="si">}</span><span class="s2">&#39; has been downloaded and prepared for use&quot;</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/kinetics.html b/0.16/_modules/torchvision/datasets/kinetics.html
new file mode 100644
index 00000000000..f95faf23bd9
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/kinetics.html
@@ -0,0 +1,916 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.kinetics &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.kinetics</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.kinetics</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">csv</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">time</span>
+<span class="kn">import</span> <span class="nn">urllib</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">multiprocessing</span> <span class="kn">import</span> <span class="n">Pool</span>
+<span class="kn">from</span> <span class="nn">os</span> <span class="kn">import</span> <span class="n">path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">find_classes</span><span class="p">,</span> <span class="n">make_dataset</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">download_url</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.video_utils</span> <span class="kn">import</span> <span class="n">VideoClips</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<span class="k">def</span> <span class="nf">_dl_wrap</span><span class="p">(</span><span class="n">tarpath</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">videopath</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">line</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+    <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">line</span><span class="p">,</span> <span class="n">tarpath</span><span class="p">,</span> <span class="n">videopath</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="Kinetics"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kinetics.html#torchvision.datasets.Kinetics">[docs]</a><span class="k">class</span> <span class="nc">Kinetics</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Generic Kinetics &lt;https://www.deepmind.com/open-source/kinetics&gt;`_</span>
+<span class="sd">    dataset.</span>
+
+<span class="sd">    Kinetics-400/600/700 are action recognition video datasets.</span>
+<span class="sd">    This dataset consider every video as a collection of video clips of fixed size, specified</span>
+<span class="sd">    by ``frames_per_clip``, where the step in frames between each clip is given by</span>
+<span class="sd">    ``step_between_clips``.</span>
+
+<span class="sd">    To give an example, for 2 videos with 10 and 15 frames respectively, if ``frames_per_clip=5``</span>
+<span class="sd">    and ``step_between_clips=5``, the dataset size will be (2 + 3) = 5, where the first two</span>
+<span class="sd">    elements will come from video 1, and the next three elements from video 2.</span>
+<span class="sd">    Note that we drop clips which do not have exactly ``frames_per_clip`` elements, so not all</span>
+<span class="sd">    frames in a video might be present.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the Kinetics Dataset.</span>
+<span class="sd">            Directory should be structured as follows:</span>
+<span class="sd">            .. code::</span>
+
+<span class="sd">                root/</span>
+<span class="sd">                ├── split</span>
+<span class="sd">                │   ├──  class1</span>
+<span class="sd">                │   │   ├──  clip1.mp4</span>
+<span class="sd">                │   │   ├──  clip2.mp4</span>
+<span class="sd">                │   │   ├──  clip3.mp4</span>
+<span class="sd">                │   │   ├──  ...</span>
+<span class="sd">                │   ├──  class2</span>
+<span class="sd">                │   │   ├──   clipx.mp4</span>
+<span class="sd">                │   │    └── ...</span>
+
+<span class="sd">            Note: split is appended automatically using the split argument.</span>
+<span class="sd">        frames_per_clip (int): number of frames in a clip</span>
+<span class="sd">        num_classes (int): select between Kinetics-400 (default), Kinetics-600, and Kinetics-700</span>
+<span class="sd">        split (str): split of the dataset to consider; supports ``&quot;train&quot;`` (default) ``&quot;val&quot;`` ``&quot;test&quot;``</span>
+<span class="sd">        frame_rate (float): If omitted, interpolate different frame rate for each clip.</span>
+<span class="sd">        step_between_clips (int): number of frames between each clip</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in a TxHxWxC video</span>
+<span class="sd">            and returns a transformed version.</span>
+<span class="sd">        download (bool): Download the official version of the dataset to root folder.</span>
+<span class="sd">        num_workers (int): Use multiple workers for VideoClips creation</span>
+<span class="sd">        num_download_workers (int): Use multiprocessing in order to speed up download.</span>
+<span class="sd">        output_format (str, optional): The format of the output video tensors (before transforms).</span>
+<span class="sd">            Can be either &quot;THWC&quot; or &quot;TCHW&quot; (default).</span>
+<span class="sd">            Note that in most other utils and datasets, the default is actually &quot;THWC&quot;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        tuple: A 3-tuple with the following entries:</span>
+
+<span class="sd">            - video (Tensor[T, C, H, W] or Tensor[T, H, W, C]): the `T` video frames in torch.uint8 tensor</span>
+<span class="sd">            - audio(Tensor[K, L]): the audio frames, where `K` is the number of channels</span>
+<span class="sd">              and `L` is the number of points in torch.float tensor</span>
+<span class="sd">            - label (int): class of the video clip</span>
+
+<span class="sd">    Raises:</span>
+<span class="sd">        RuntimeError: If ``download is True`` and the video archives are already extracted.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_TAR_URLS</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;400&quot;</span><span class="p">:</span> <span class="s2">&quot;https://s3.amazonaws.com/kinetics/400/</span><span class="si">{split}</span><span class="s2">/k400_</span><span class="si">{split}</span><span class="s2">_path.txt&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;600&quot;</span><span class="p">:</span> <span class="s2">&quot;https://s3.amazonaws.com/kinetics/600/</span><span class="si">{split}</span><span class="s2">/k600_</span><span class="si">{split}</span><span class="s2">_path.txt&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;700&quot;</span><span class="p">:</span> <span class="s2">&quot;https://s3.amazonaws.com/kinetics/700_2020/</span><span class="si">{split}</span><span class="s2">/k700_2020_</span><span class="si">{split}</span><span class="s2">_path.txt&quot;</span><span class="p">,</span>
+    <span class="p">}</span>
+    <span class="n">_ANNOTATION_URLS</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;400&quot;</span><span class="p">:</span> <span class="s2">&quot;https://s3.amazonaws.com/kinetics/400/annotations/</span><span class="si">{split}</span><span class="s2">.csv&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;600&quot;</span><span class="p">:</span> <span class="s2">&quot;https://s3.amazonaws.com/kinetics/600/annotations/</span><span class="si">{split}</span><span class="s2">.csv&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;700&quot;</span><span class="p">:</span> <span class="s2">&quot;https://s3.amazonaws.com/kinetics/700_2020/annotations/</span><span class="si">{split}</span><span class="s2">.csv&quot;</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">frames_per_clip</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;400&quot;</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">frame_rate</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">step_between_clips</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">extensions</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="o">...</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;avi&quot;</span><span class="p">,</span> <span class="s2">&quot;mp4&quot;</span><span class="p">),</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">num_download_workers</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">num_workers</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">_precomputed_metadata</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">_video_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_video_height</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_video_min_dimension</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_audio_samples</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_audio_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_legacy</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">output_format</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;TCHW&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="c1"># TODO: support test</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">num_classes</span><span class="p">,</span> <span class="n">arg</span><span class="o">=</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;400&quot;</span><span class="p">,</span> <span class="s2">&quot;600&quot;</span><span class="p">,</span> <span class="s2">&quot;700&quot;</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">extensions</span> <span class="o">=</span> <span class="n">extensions</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_download_workers</span> <span class="o">=</span> <span class="n">num_download_workers</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">root</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_legacy</span> <span class="o">=</span> <span class="n">_legacy</span>
+
+        <span class="k">if</span> <span class="n">_legacy</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Using legacy structure&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span> <span class="o">=</span> <span class="n">root</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="s2">&quot;unknown&quot;</span>
+            <span class="n">output_format</span> <span class="o">=</span> <span class="s2">&quot;THWC&quot;</span>
+            <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Cannot download the videos using legacy_structure.&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">split</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="n">arg</span><span class="o">=</span><span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">])</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download_and_process_videos</span><span class="p">()</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="n">class_to_idx</span> <span class="o">=</span> <span class="n">find_classes</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">samples</span> <span class="o">=</span> <span class="n">make_dataset</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">,</span> <span class="n">class_to_idx</span><span class="p">,</span> <span class="n">extensions</span><span class="p">,</span> <span class="n">is_valid_file</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+        <span class="n">video_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span> <span class="o">=</span> <span class="n">VideoClips</span><span class="p">(</span>
+            <span class="n">video_list</span><span class="p">,</span>
+            <span class="n">frames_per_clip</span><span class="p">,</span>
+            <span class="n">step_between_clips</span><span class="p">,</span>
+            <span class="n">frame_rate</span><span class="p">,</span>
+            <span class="n">_precomputed_metadata</span><span class="p">,</span>
+            <span class="n">num_workers</span><span class="o">=</span><span class="n">num_workers</span><span class="p">,</span>
+            <span class="n">_video_width</span><span class="o">=</span><span class="n">_video_width</span><span class="p">,</span>
+            <span class="n">_video_height</span><span class="o">=</span><span class="n">_video_height</span><span class="p">,</span>
+            <span class="n">_video_min_dimension</span><span class="o">=</span><span class="n">_video_min_dimension</span><span class="p">,</span>
+            <span class="n">_audio_samples</span><span class="o">=</span><span class="n">_audio_samples</span><span class="p">,</span>
+            <span class="n">_audio_channels</span><span class="o">=</span><span class="n">_audio_channels</span><span class="p">,</span>
+            <span class="n">output_format</span><span class="o">=</span><span class="n">output_format</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">transform</span>
+
+    <span class="k">def</span> <span class="nf">download_and_process_videos</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Downloads all the videos to the _root_ folder in the expected format.&quot;&quot;&quot;</span>
+        <span class="n">tic</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_download_videos</span><span class="p">()</span>
+        <span class="n">toc</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+        <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Elapsed time for downloading in mins &quot;</span><span class="p">,</span> <span class="p">(</span><span class="n">toc</span> <span class="o">-</span> <span class="n">tic</span><span class="p">)</span> <span class="o">/</span> <span class="mi">60</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_make_ds_structure</span><span class="p">()</span>
+        <span class="n">toc2</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+        <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Elapsed time for processing in mins &quot;</span><span class="p">,</span> <span class="p">(</span><span class="n">toc2</span> <span class="o">-</span> <span class="n">toc</span><span class="p">)</span> <span class="o">/</span> <span class="mi">60</span><span class="p">)</span>
+        <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Elapsed time overall in mins &quot;</span><span class="p">,</span> <span class="p">(</span><span class="n">toc2</span> <span class="o">-</span> <span class="n">tic</span><span class="p">)</span> <span class="o">/</span> <span class="mi">60</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_download_videos</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;download tarballs containing the video to &quot;tars&quot; folder and extract them into the _split_ folder where</span>
+<span class="sd">        split is one of the official dataset splits.</span>
+
+<span class="sd">        Raises:</span>
+<span class="sd">            RuntimeError: if download folder exists, break to prevent downloading entire dataset again.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The directory </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="si">}</span><span class="s2"> already exists. &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;If you want to re-download or re-extract the images, delete the directory.&quot;</span>
+            <span class="p">)</span>
+        <span class="n">tar_path</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;tars&quot;</span><span class="p">)</span>
+        <span class="n">file_list_path</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;files&quot;</span><span class="p">)</span>
+
+        <span class="n">split_url</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_TAR_URLS</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">]</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">split</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">)</span>
+        <span class="n">split_url_filepath</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">file_list_path</span><span class="p">,</span> <span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">split_url</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">split_url_filepath</span><span class="p">):</span>
+            <span class="n">download_url</span><span class="p">(</span><span class="n">split_url</span><span class="p">,</span> <span class="n">file_list_path</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">split_url_filepath</span><span class="p">)</span> <span class="k">as</span> <span class="n">file</span><span class="p">:</span>
+            <span class="n">list_video_urls</span> <span class="o">=</span> <span class="p">[</span><span class="n">urllib</span><span class="o">.</span><span class="n">parse</span><span class="o">.</span><span class="n">quote</span><span class="p">(</span><span class="n">line</span><span class="p">,</span> <span class="n">safe</span><span class="o">=</span><span class="s2">&quot;/,:&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">file</span><span class="o">.</span><span class="n">read</span><span class="p">()</span><span class="o">.</span><span class="n">splitlines</span><span class="p">()]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_download_workers</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">list_video_urls</span><span class="p">:</span>
+                <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">line</span><span class="p">,</span> <span class="n">tar_path</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">part</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">_dl_wrap</span><span class="p">,</span> <span class="n">tar_path</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">)</span>
+            <span class="n">poolproc</span> <span class="o">=</span> <span class="n">Pool</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_download_workers</span><span class="p">)</span>
+            <span class="n">poolproc</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="n">part</span><span class="p">,</span> <span class="n">list_video_urls</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_make_ds_structure</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;move videos from</span>
+<span class="sd">        split_folder/</span>
+<span class="sd">            ├── clip1.avi</span>
+<span class="sd">            ├── clip2.avi</span>
+
+<span class="sd">        to the correct format as described below:</span>
+<span class="sd">        split_folder/</span>
+<span class="sd">            ├── class1</span>
+<span class="sd">            │   ├── clip1.avi</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">annotation_path</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;annotations&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">annotation_path</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="si">}</span><span class="s2">.csv&quot;</span><span class="p">)):</span>
+            <span class="n">download_url</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_ANNOTATION_URLS</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">]</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">split</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">),</span> <span class="n">annotation_path</span><span class="p">)</span>
+        <span class="n">annotations</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">annotation_path</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="si">}</span><span class="s2">.csv&quot;</span><span class="p">)</span>
+
+        <span class="n">file_fmtstr</span> <span class="o">=</span> <span class="s2">&quot;</span><span class="si">{ytid}</span><span class="s2">_</span><span class="si">{start:06}</span><span class="s2">_</span><span class="si">{end:06}</span><span class="s2">.mp4&quot;</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">annotations</span><span class="p">)</span> <span class="k">as</span> <span class="n">csvfile</span><span class="p">:</span>
+            <span class="n">reader</span> <span class="o">=</span> <span class="n">csv</span><span class="o">.</span><span class="n">DictReader</span><span class="p">(</span><span class="n">csvfile</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">reader</span><span class="p">:</span>
+                <span class="n">f</span> <span class="o">=</span> <span class="n">file_fmtstr</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                    <span class="n">ytid</span><span class="o">=</span><span class="n">row</span><span class="p">[</span><span class="s2">&quot;youtube_id&quot;</span><span class="p">],</span>
+                    <span class="n">start</span><span class="o">=</span><span class="nb">int</span><span class="p">(</span><span class="n">row</span><span class="p">[</span><span class="s2">&quot;time_start&quot;</span><span class="p">]),</span>
+                    <span class="n">end</span><span class="o">=</span><span class="nb">int</span><span class="p">(</span><span class="n">row</span><span class="p">[</span><span class="s2">&quot;time_end&quot;</span><span class="p">]),</span>
+                <span class="p">)</span>
+                <span class="n">label</span> <span class="o">=</span> <span class="n">row</span><span class="p">[</span><span class="s2">&quot;label&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">,</span> <span class="s2">&quot;_&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;&#39;&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;(&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;)&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span>
+                <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">,</span> <span class="n">label</span><span class="p">),</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">downloaded_file</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">,</span> <span class="n">f</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">downloaded_file</span><span class="p">):</span>
+                    <span class="n">os</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span>
+                        <span class="n">downloaded_file</span><span class="p">,</span>
+                        <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_folder</span><span class="p">,</span> <span class="n">label</span><span class="p">,</span> <span class="n">f</span><span class="p">),</span>
+                    <span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">metadata</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">metadata</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">num_clips</span><span class="p">()</span>
+
+<div class="viewcode-block" id="Kinetics.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kinetics.html#torchvision.datasets.Kinetics.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">info</span><span class="p">,</span> <span class="n">video_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">get_clip</span><span class="p">(</span><span class="n">idx</span><span class="p">)</span>
+        <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">[</span><span class="n">video_idx</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">label</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/kitti.html b/0.16/_modules/torchvision/datasets/kitti.html
new file mode 100644
index 00000000000..bbc4f2c4f79
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/kitti.html
@@ -0,0 +1,827 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.kitti &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.kitti</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.kitti</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">csv</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="Kitti"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kitti.html#torchvision.datasets.Kitti">[docs]</a><span class="k">class</span> <span class="nc">Kitti</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`KITTI &lt;http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark&gt;`_ Dataset.</span>
+
+<span class="sd">    It corresponds to the &quot;left color images of object&quot; dataset, for object detection.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images are downloaded to.</span>
+<span class="sd">            Expects the following folder structure if download=False:</span>
+
+<span class="sd">            .. code::</span>
+
+<span class="sd">                &lt;root&gt;</span>
+<span class="sd">                    └── Kitti</span>
+<span class="sd">                        └─ raw</span>
+<span class="sd">                            ├── training</span>
+<span class="sd">                            |   ├── image_2</span>
+<span class="sd">                            |   └── label_2</span>
+<span class="sd">                            └── testing</span>
+<span class="sd">                                └── image_2</span>
+<span class="sd">        train (bool, optional): Use ``train`` split if true, else ``test`` split.</span>
+<span class="sd">            Defaults to ``train``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in a PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.PILToTensor``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes input sample</span>
+<span class="sd">            and its target as entry and returns a transformed version.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">data_url</span> <span class="o">=</span> <span class="s2">&quot;https://s3.eu-central-1.amazonaws.com/avg-kitti/&quot;</span>
+    <span class="n">resources</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="s2">&quot;data_object_image_2.zip&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;data_object_label_2.zip&quot;</span><span class="p">,</span>
+    <span class="p">]</span>
+    <span class="n">image_dir_name</span> <span class="o">=</span> <span class="s2">&quot;image_2&quot;</span>
+    <span class="n">labels_dir_name</span> <span class="o">=</span> <span class="s2">&quot;label_2&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">root</span><span class="p">,</span>
+            <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span>
+            <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">,</span>
+            <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">images</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">root</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="o">=</span> <span class="n">train</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_location</span> <span class="o">=</span> <span class="s2">&quot;training&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="k">else</span> <span class="s2">&quot;testing&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You may use download=True to download it.&quot;</span><span class="p">)</span>
+
+        <span class="n">image_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_raw_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_location</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">image_dir_name</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="p">:</span>
+            <span class="n">labels_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_raw_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_location</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels_dir_name</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">img_file</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">image_dir</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">image_dir</span><span class="p">,</span> <span class="n">img_file</span><span class="p">))</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">labels_dir</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">img_file</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39;.&#39;</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">.txt&quot;</span><span class="p">))</span>
+
+<div class="viewcode-block" id="Kitti.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Kitti.html#torchvision.datasets.Kitti.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get item at a given index.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target), where</span>
+<span class="sd">            target is a list of dictionaries with the following keys:</span>
+
+<span class="sd">            - type: str</span>
+<span class="sd">            - truncated: float</span>
+<span class="sd">            - occluded: int</span>
+<span class="sd">            - alpha: float</span>
+<span class="sd">            - bbox: float[4]</span>
+<span class="sd">            - dimensions: float[3]</span>
+<span class="sd">            - locations: float[3]</span>
+<span class="sd">            - rotation_y: float</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_parse_target</span><span class="p">(</span><span class="n">index</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="k">else</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">_parse_target</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">:</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">])</span> <span class="k">as</span> <span class="n">inp</span><span class="p">:</span>
+            <span class="n">content</span> <span class="o">=</span> <span class="n">csv</span><span class="o">.</span><span class="n">reader</span><span class="p">(</span><span class="n">inp</span><span class="p">,</span> <span class="n">delimiter</span><span class="o">=</span><span class="s2">&quot; &quot;</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">content</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                    <span class="p">{</span>
+                        <span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="n">line</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+                        <span class="s2">&quot;truncated&quot;</span><span class="p">:</span> <span class="nb">float</span><span class="p">(</span><span class="n">line</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span>
+                        <span class="s2">&quot;occluded&quot;</span><span class="p">:</span> <span class="nb">int</span><span class="p">(</span><span class="n">line</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span>
+                        <span class="s2">&quot;alpha&quot;</span><span class="p">:</span> <span class="nb">float</span><span class="p">(</span><span class="n">line</span><span class="p">[</span><span class="mi">3</span><span class="p">]),</span>
+                        <span class="s2">&quot;bbox&quot;</span><span class="p">:</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">line</span><span class="p">[</span><span class="mi">4</span><span class="p">:</span><span class="mi">8</span><span class="p">]],</span>
+                        <span class="s2">&quot;dimensions&quot;</span><span class="p">:</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">line</span><span class="p">[</span><span class="mi">8</span><span class="p">:</span><span class="mi">11</span><span class="p">]],</span>
+                        <span class="s2">&quot;location&quot;</span><span class="p">:</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">line</span><span class="p">[</span><span class="mi">11</span><span class="p">:</span><span class="mi">14</span><span class="p">]],</span>
+                        <span class="s2">&quot;rotation_y&quot;</span><span class="p">:</span> <span class="nb">float</span><span class="p">(</span><span class="n">line</span><span class="p">[</span><span class="mi">14</span><span class="p">]),</span>
+                    <span class="p">}</span>
+                <span class="p">)</span>
+        <span class="k">return</span> <span class="n">target</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">_raw_folder</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">,</span> <span class="s2">&quot;raw&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Check if the data directory exists.&quot;&quot;&quot;</span>
+        <span class="n">folders</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">image_dir_name</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="p">:</span>
+            <span class="n">folders</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">labels_dir_name</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_raw_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_location</span><span class="p">,</span> <span class="n">fname</span><span class="p">))</span> <span class="k">for</span> <span class="n">fname</span> <span class="ow">in</span> <span class="n">folders</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Download the KITTI data if it doesn&#39;t exist already.&quot;&quot;&quot;</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_raw_folder</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="c1"># download files</span>
+        <span class="k">for</span> <span class="n">fname</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">resources</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span>
+                <span class="n">url</span><span class="o">=</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">data_url</span><span class="si">}{</span><span class="n">fname</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+                <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_raw_folder</span><span class="p">,</span>
+                <span class="n">filename</span><span class="o">=</span><span class="n">fname</span><span class="p">,</span>
+            <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/lfw.html b/0.16/_modules/torchvision/datasets/lfw.html
new file mode 100644
index 00000000000..ec7948db64d
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/lfw.html
@@ -0,0 +1,924 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.lfw &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.lfw</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.lfw</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">download_url</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<span class="k">class</span> <span class="nc">_LFW</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+
+    <span class="n">base_folder</span> <span class="o">=</span> <span class="s2">&quot;lfw-py&quot;</span>
+    <span class="n">download_url_prefix</span> <span class="o">=</span> <span class="s2">&quot;http://vis-www.cs.umass.edu/lfw/&quot;</span>
+
+    <span class="n">file_dict</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;original&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;lfw&quot;</span><span class="p">,</span> <span class="s2">&quot;lfw.tgz&quot;</span><span class="p">,</span> <span class="s2">&quot;a17d05bd522c52d84eca14327a23d494&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;funneled&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;lfw_funneled&quot;</span><span class="p">,</span> <span class="s2">&quot;lfw-funneled.tgz&quot;</span><span class="p">,</span> <span class="s2">&quot;1b42dfed7d15c9b2dd63d5e5840c86ad&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;deepfunneled&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;lfw-deepfunneled&quot;</span><span class="p">,</span> <span class="s2">&quot;lfw-deepfunneled.tgz&quot;</span><span class="p">,</span> <span class="s2">&quot;68331da3eb755a505a502b5aacb3c201&quot;</span><span class="p">),</span>
+    <span class="p">}</span>
+    <span class="n">checksums</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;pairs.txt&quot;</span><span class="p">:</span> <span class="s2">&quot;9f1ba174e4e1c508ff7cdf10ac338a7d&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;pairsDevTest.txt&quot;</span><span class="p">:</span> <span class="s2">&quot;5132f7440eb68cf58910c8a45a2ac10b&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;pairsDevTrain.txt&quot;</span><span class="p">:</span> <span class="s2">&quot;4f27cbf15b2da4a85c1907eb4181ad21&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;people.txt&quot;</span><span class="p">:</span> <span class="s2">&quot;450f0863dd89e85e73936a6d71a3474b&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;peopleDevTest.txt&quot;</span><span class="p">:</span> <span class="s2">&quot;e4bf5be0a43b5dcd9dc5ccfcb8fb19c5&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;peopleDevTrain.txt&quot;</span><span class="p">:</span> <span class="s2">&quot;54eaac34beb6d042ed3a7d883e247a21&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;lfw-names.txt&quot;</span><span class="p">:</span> <span class="s2">&quot;a6d0a479bd074669f656265a6e693f6d&quot;</span><span class="p">,</span>
+    <span class="p">}</span>
+    <span class="n">annot_file</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;10fold&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="s2">&quot;DevTrain&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="s2">&quot;DevTest&quot;</span><span class="p">}</span>
+    <span class="n">names</span> <span class="o">=</span> <span class="s2">&quot;lfw-names.txt&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">image_set</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">view</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_set</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">image_set</span><span class="o">.</span><span class="n">lower</span><span class="p">(),</span> <span class="s2">&quot;image_set&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">file_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
+        <span class="n">images_dir</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">file_dict</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">image_set</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">view</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">view</span><span class="o">.</span><span class="n">lower</span><span class="p">(),</span> <span class="s2">&quot;view&quot;</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;people&quot;</span><span class="p">,</span> <span class="s2">&quot;pairs&quot;</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="o">.</span><span class="n">lower</span><span class="p">(),</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;10fold&quot;</span><span class="p">,</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">view</span><span class="si">}{</span><span class="bp">self</span><span class="o">.</span><span class="n">annot_file</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">]</span><span class="si">}</span><span class="s2">.txt&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">images_dir</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_loader</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">img</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="n">st1</span> <span class="o">=</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">md5</span><span class="p">)</span>
+        <span class="n">st2</span> <span class="o">=</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">checksums</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">])</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">st1</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">st2</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">view</span> <span class="o">==</span> <span class="s2">&quot;people&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">names</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">checksums</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">names</span><span class="p">])</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+        <span class="n">url</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">download_url_prefix</span><span class="si">}{</span><span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">md5</span><span class="p">)</span>
+        <span class="n">download_url</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">download_url_prefix</span><span class="si">}{</span><span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">view</span> <span class="o">==</span> <span class="s2">&quot;people&quot;</span><span class="p">:</span>
+            <span class="n">download_url</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">download_url_prefix</span><span class="si">}{</span><span class="bp">self</span><span class="o">.</span><span class="n">names</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_path</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">identity</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">no</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">str</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">,</span> <span class="n">identity</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">identity</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="nb">int</span><span class="p">(</span><span class="n">no</span><span class="p">)</span><span class="si">:</span><span class="s2">04d</span><span class="si">}</span><span class="s2">.jpg&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;Alignment: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">image_set</span><span class="si">}</span><span class="se">\n</span><span class="s2">Split: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="si">}</span><span class="s2">&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="LFWPeople"><a class="viewcode-back" href="../../../generated/torchvision.datasets.LFWPeople.html#torchvision.datasets.LFWPeople">[docs]</a><span class="k">class</span> <span class="nc">LFWPeople</span><span class="p">(</span><span class="n">_LFW</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`LFW &lt;http://vis-www.cs.umass.edu/lfw/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``lfw-py`` exists or will be saved to if download is set to True.</span>
+<span class="sd">        split (string, optional): The image split to use. Can be one of ``train``, ``test``,</span>
+<span class="sd">            ``10fold`` (default).</span>
+<span class="sd">        image_set (str, optional): Type of image funneling to use, ``original``, ``funneled`` or</span>
+<span class="sd">            ``deepfunneled``. Defaults to ``funneled``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomRotation``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;10fold&quot;</span><span class="p">,</span>
+        <span class="n">image_set</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;funneled&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">split</span><span class="p">,</span> <span class="n">image_set</span><span class="p">,</span> <span class="s2">&quot;people&quot;</span><span class="p">,</span> <span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">,</span> <span class="n">download</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_classes</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_people</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_get_people</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+        <span class="n">data</span><span class="p">,</span> <span class="n">targets</span> <span class="o">=</span> <span class="p">[],</span> <span class="p">[]</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">))</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">lines</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+            <span class="n">n_folds</span><span class="p">,</span> <span class="n">s</span> <span class="o">=</span> <span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">lines</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="mi">1</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;10fold&quot;</span> <span class="k">else</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+            <span class="k">for</span> <span class="n">fold</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n_folds</span><span class="p">):</span>
+                <span class="n">n_lines</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">lines</span><span class="p">[</span><span class="n">s</span><span class="p">])</span>
+                <span class="n">people</span> <span class="o">=</span> <span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\t</span><span class="s2">&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">[</span><span class="n">s</span> <span class="o">+</span> <span class="mi">1</span> <span class="p">:</span> <span class="n">s</span> <span class="o">+</span> <span class="n">n_lines</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]]</span>
+                <span class="n">s</span> <span class="o">+=</span> <span class="n">n_lines</span> <span class="o">+</span> <span class="mi">1</span>
+                <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">identity</span><span class="p">,</span> <span class="n">num_imgs</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">people</span><span class="p">):</span>
+                    <span class="k">for</span> <span class="n">num</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">num_imgs</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span><span class="p">):</span>
+                        <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_path</span><span class="p">(</span><span class="n">identity</span><span class="p">,</span> <span class="n">num</span><span class="p">)</span>
+                        <span class="n">data</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+                        <span class="n">targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span><span class="p">[</span><span class="n">identity</span><span class="p">])</span>
+
+        <span class="k">return</span> <span class="n">data</span><span class="p">,</span> <span class="n">targets</span>
+
+    <span class="k">def</span> <span class="nf">_get_classes</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">names</span><span class="p">))</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">lines</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+            <span class="n">names</span> <span class="o">=</span> <span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">()[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">]</span>
+        <span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="n">name</span><span class="p">:</span> <span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">name</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">names</span><span class="p">)}</span>
+        <span class="k">return</span> <span class="n">class_to_idx</span>
+
+<div class="viewcode-block" id="LFWPeople.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.LFWPeople.html#torchvision.datasets.LFWPeople.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: Tuple (image, target) where target is the identity of the person.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_loader</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">extra_repr</span><span class="p">()</span> <span class="o">+</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">Classes (identities): </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span></div>
+
+
+<div class="viewcode-block" id="LFWPairs"><a class="viewcode-back" href="../../../generated/torchvision.datasets.LFWPairs.html#torchvision.datasets.LFWPairs">[docs]</a><span class="k">class</span> <span class="nc">LFWPairs</span><span class="p">(</span><span class="n">_LFW</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`LFW &lt;http://vis-www.cs.umass.edu/lfw/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``lfw-py`` exists or will be saved to if download is set to True.</span>
+<span class="sd">        split (string, optional): The image split to use. Can be one of ``train``, ``test``,</span>
+<span class="sd">            ``10fold``. Defaults to ``10fold``.</span>
+<span class="sd">        image_set (str, optional): Type of image funneling to use, ``original``, ``funneled`` or</span>
+<span class="sd">            ``deepfunneled``. Defaults to ``funneled``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomRotation``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;10fold&quot;</span><span class="p">,</span>
+        <span class="n">image_set</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;funneled&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">split</span><span class="p">,</span> <span class="n">image_set</span><span class="p">,</span> <span class="s2">&quot;pairs&quot;</span><span class="p">,</span> <span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">,</span> <span class="n">download</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">pair_names</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_pairs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_pairs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">images_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+        <span class="n">pair_names</span><span class="p">,</span> <span class="n">data</span><span class="p">,</span> <span class="n">targets</span> <span class="o">=</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[]</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">))</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">lines</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;10fold&quot;</span><span class="p">:</span>
+                <span class="n">n_folds</span><span class="p">,</span> <span class="n">n_pairs</span> <span class="o">=</span> <span class="n">lines</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\t</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="n">n_folds</span><span class="p">,</span> <span class="n">n_pairs</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">n_folds</span><span class="p">),</span> <span class="nb">int</span><span class="p">(</span><span class="n">n_pairs</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">n_folds</span><span class="p">,</span> <span class="n">n_pairs</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">lines</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="n">s</span> <span class="o">=</span> <span class="mi">1</span>
+
+            <span class="k">for</span> <span class="n">fold</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n_folds</span><span class="p">):</span>
+                <span class="n">matched_pairs</span> <span class="o">=</span> <span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\t</span><span class="s2">&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">[</span><span class="n">s</span> <span class="p">:</span> <span class="n">s</span> <span class="o">+</span> <span class="n">n_pairs</span><span class="p">]]</span>
+                <span class="n">unmatched_pairs</span> <span class="o">=</span> <span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\t</span><span class="s2">&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">[</span><span class="n">s</span> <span class="o">+</span> <span class="n">n_pairs</span> <span class="p">:</span> <span class="n">s</span> <span class="o">+</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="n">n_pairs</span><span class="p">)]]</span>
+                <span class="n">s</span> <span class="o">+=</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">n_pairs</span>
+                <span class="k">for</span> <span class="n">pair</span> <span class="ow">in</span> <span class="n">matched_pairs</span><span class="p">:</span>
+                    <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">,</span> <span class="n">same</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_path</span><span class="p">(</span><span class="n">pair</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pair</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_path</span><span class="p">(</span><span class="n">pair</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pair</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span> <span class="mi">1</span>
+                    <span class="n">pair_names</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">pair</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pair</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+                    <span class="n">data</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">))</span>
+                    <span class="n">targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">same</span><span class="p">)</span>
+                <span class="k">for</span> <span class="n">pair</span> <span class="ow">in</span> <span class="n">unmatched_pairs</span><span class="p">:</span>
+                    <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">,</span> <span class="n">same</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_path</span><span class="p">(</span><span class="n">pair</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pair</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_path</span><span class="p">(</span><span class="n">pair</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">pair</span><span class="p">[</span><span class="mi">3</span><span class="p">]),</span> <span class="mi">0</span>
+                    <span class="n">pair_names</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">pair</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pair</span><span class="p">[</span><span class="mi">2</span><span class="p">]))</span>
+                    <span class="n">data</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">))</span>
+                    <span class="n">targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">same</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">pair_names</span><span class="p">,</span> <span class="n">data</span><span class="p">,</span> <span class="n">targets</span>
+
+<div class="viewcode-block" id="LFWPairs.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.LFWPairs.html#torchvision.datasets.LFWPairs.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image1, image2, target) where target is `0` for different indentities and `1` for same identities.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_loader</span><span class="p">(</span><span class="n">img1</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">_loader</span><span class="p">(</span><span class="n">img2</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img1</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img2</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img1</span><span class="p">,</span> <span class="n">img2</span><span class="p">,</span> <span class="n">target</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/lsun.html b/0.16/_modules/torchvision/datasets/lsun.html
new file mode 100644
index 00000000000..0c84756b4d6
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/lsun.html
@@ -0,0 +1,836 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.lsun &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.lsun</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.lsun</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">io</span>
+<span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">import</span> <span class="nn">pickle</span>
+<span class="kn">import</span> <span class="nn">string</span>
+<span class="kn">from</span> <span class="nn">collections.abc</span> <span class="kn">import</span> <span class="n">Iterable</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">iterable_to_str</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<span class="k">class</span> <span class="nc">LSUNClass</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="kn">import</span> <span class="nn">lmdb</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">env</span> <span class="o">=</span> <span class="n">lmdb</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">max_readers</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">readonly</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">lock</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">readahead</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">meminit</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">begin</span><span class="p">(</span><span class="n">write</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> <span class="k">as</span> <span class="n">txn</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">length</span> <span class="o">=</span> <span class="n">txn</span><span class="o">.</span><span class="n">stat</span><span class="p">()[</span><span class="s2">&quot;entries&quot;</span><span class="p">]</span>
+        <span class="n">cache_file</span> <span class="o">=</span> <span class="s2">&quot;_cache_&quot;</span> <span class="o">+</span> <span class="s2">&quot;&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">c</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">root</span> <span class="k">if</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">string</span><span class="o">.</span><span class="n">ascii_letters</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">cache_file</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">keys</span> <span class="o">=</span> <span class="n">pickle</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="nb">open</span><span class="p">(</span><span class="n">cache_file</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span><span class="o">.</span><span class="n">begin</span><span class="p">(</span><span class="n">write</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> <span class="k">as</span> <span class="n">txn</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">keys</span> <span class="o">=</span> <span class="p">[</span><span class="n">key</span> <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">txn</span><span class="o">.</span><span class="n">cursor</span><span class="p">()</span><span class="o">.</span><span class="n">iternext</span><span class="p">(</span><span class="n">keys</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">values</span><span class="o">=</span><span class="kc">False</span><span class="p">)]</span>
+            <span class="n">pickle</span><span class="o">.</span><span class="n">dump</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">keys</span><span class="p">,</span> <span class="nb">open</span><span class="p">(</span><span class="n">cache_file</span><span class="p">,</span> <span class="s2">&quot;wb&quot;</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+        <span class="n">env</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">env</span>
+        <span class="k">with</span> <span class="n">env</span><span class="o">.</span><span class="n">begin</span><span class="p">(</span><span class="n">write</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> <span class="k">as</span> <span class="n">txn</span><span class="p">:</span>
+            <span class="n">imgbuf</span> <span class="o">=</span> <span class="n">txn</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">keys</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+
+        <span class="n">buf</span> <span class="o">=</span> <span class="n">io</span><span class="o">.</span><span class="n">BytesIO</span><span class="p">()</span>
+        <span class="n">buf</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">imgbuf</span><span class="p">)</span>
+        <span class="n">buf</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">buf</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">length</span>
+
+
+<div class="viewcode-block" id="LSUN"><a class="viewcode-back" href="../../../generated/torchvision.datasets.LSUN.html#torchvision.datasets.LSUN">[docs]</a><span class="k">class</span> <span class="nc">LSUN</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`LSUN &lt;https://www.yf.io/p/lsun&gt;`_ dataset.</span>
+
+<span class="sd">    You will need to install the ``lmdb`` package to use this dataset: run</span>
+<span class="sd">    ``pip install lmdb``</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory for the database files.</span>
+<span class="sd">        classes (string or list): One of {&#39;train&#39;, &#39;val&#39;, &#39;test&#39;} or a list of</span>
+<span class="sd">            categories to load. e,g. [&#39;bedroom_train&#39;, &#39;church_outdoor_train&#39;].</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">classes</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_verify_classes</span><span class="p">(</span><span class="n">classes</span><span class="p">)</span>
+
+        <span class="c1"># for each class, create an LSUNClassDataset</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dbs</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">dbs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">LSUNClass</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">c</span><span class="si">}</span><span class="s2">_lmdb&quot;</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">indices</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">count</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">db</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">dbs</span><span class="p">:</span>
+            <span class="n">count</span> <span class="o">+=</span> <span class="nb">len</span><span class="p">(</span><span class="n">db</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">indices</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">count</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">length</span> <span class="o">=</span> <span class="n">count</span>
+
+    <span class="k">def</span> <span class="nf">_verify_classes</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">classes</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="n">categories</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="s2">&quot;bedroom&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;bridge&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;church_outdoor&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;classroom&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;conference_room&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;dining_room&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;kitchen&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;living_room&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;restaurant&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;tower&quot;</span><span class="p">,</span>
+        <span class="p">]</span>
+        <span class="n">dset_opts</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">]</span>
+
+        <span class="k">try</span><span class="p">:</span>
+            <span class="n">classes</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="nb">str</span><span class="p">,</span> <span class="n">classes</span><span class="p">)</span>
+            <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">classes</span><span class="p">,</span> <span class="s2">&quot;classes&quot;</span><span class="p">,</span> <span class="n">dset_opts</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">classes</span> <span class="o">==</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span>
+                <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="n">classes</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="n">c</span> <span class="o">+</span> <span class="s2">&quot;_&quot;</span> <span class="o">+</span> <span class="n">classes</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">categories</span><span class="p">]</span>
+        <span class="k">except</span> <span class="ne">ValueError</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">classes</span><span class="p">,</span> <span class="n">Iterable</span><span class="p">):</span>
+                <span class="n">msg</span> <span class="o">=</span> <span class="s2">&quot;Expected type str or Iterable for argument classes, but got type </span><span class="si">{}</span><span class="s2">.&quot;</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">classes</span><span class="p">)))</span>
+
+            <span class="n">classes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">classes</span><span class="p">)</span>
+            <span class="n">msg_fmtstr_type</span> <span class="o">=</span> <span class="s2">&quot;Expected type str for elements in argument classes, but got type </span><span class="si">{}</span><span class="s2">.&quot;</span>
+            <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">classes</span><span class="p">:</span>
+                <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">c</span><span class="p">,</span> <span class="n">custom_msg</span><span class="o">=</span><span class="n">msg_fmtstr_type</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">c</span><span class="p">)))</span>
+                <span class="n">c_short</span> <span class="o">=</span> <span class="n">c</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;_&quot;</span><span class="p">)</span>
+                <span class="n">category</span><span class="p">,</span> <span class="n">dset_opt</span> <span class="o">=</span> <span class="s2">&quot;_&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">c_short</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]),</span> <span class="n">c_short</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+                <span class="n">msg_fmtstr</span> <span class="o">=</span> <span class="s2">&quot;Unknown value &#39;</span><span class="si">{}</span><span class="s2">&#39; for </span><span class="si">{}</span><span class="s2">. Valid values are {{</span><span class="si">{}</span><span class="s2">}}.&quot;</span>
+                <span class="n">msg</span> <span class="o">=</span> <span class="n">msg_fmtstr</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">category</span><span class="p">,</span> <span class="s2">&quot;LSUN class&quot;</span><span class="p">,</span> <span class="n">iterable_to_str</span><span class="p">(</span><span class="n">categories</span><span class="p">))</span>
+                <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">category</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="n">categories</span><span class="p">,</span> <span class="n">custom_msg</span><span class="o">=</span><span class="n">msg</span><span class="p">)</span>
+
+                <span class="n">msg</span> <span class="o">=</span> <span class="n">msg_fmtstr</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">dset_opt</span><span class="p">,</span> <span class="s2">&quot;postfix&quot;</span><span class="p">,</span> <span class="n">iterable_to_str</span><span class="p">(</span><span class="n">dset_opts</span><span class="p">))</span>
+                <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">dset_opt</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="n">dset_opts</span><span class="p">,</span> <span class="n">custom_msg</span><span class="o">=</span><span class="n">msg</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">classes</span>
+
+<div class="viewcode-block" id="LSUN.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.LSUN.html#torchvision.datasets.LSUN.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: Tuple (image, target) where target is the index of the target category.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="n">sub</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">ind</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">indices</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">index</span> <span class="o">&lt;</span> <span class="n">ind</span><span class="p">:</span>
+                <span class="k">break</span>
+            <span class="n">target</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">sub</span> <span class="o">=</span> <span class="n">ind</span>
+
+        <span class="n">db</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dbs</span><span class="p">[</span><span class="n">target</span><span class="p">]</span>
+        <span class="n">index</span> <span class="o">=</span> <span class="n">index</span> <span class="o">-</span> <span class="n">sub</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="n">img</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">db</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">length</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;Classes: </span><span class="si">{classes}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/mnist.html b/0.16/_modules/torchvision/datasets/mnist.html
new file mode 100644
index 00000000000..f7d461d157c
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/mnist.html
@@ -0,0 +1,1217 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.mnist &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.mnist</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.mnist</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">codecs</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">import</span> <span class="nn">shutil</span>
+<span class="kn">import</span> <span class="nn">string</span>
+<span class="kn">import</span> <span class="nn">sys</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+<span class="kn">from</span> <span class="nn">urllib.error</span> <span class="kn">import</span> <span class="n">URLError</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_flip_byte_order</span><span class="p">,</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="MNIST"><a class="viewcode-back" href="../../../generated/torchvision.datasets.MNIST.html#torchvision.datasets.MNIST">[docs]</a><span class="k">class</span> <span class="nc">MNIST</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`MNIST &lt;http://yann.lecun.com/exdb/mnist/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where ``MNIST/raw/train-images-idx3-ubyte``</span>
+<span class="sd">            and  ``MNIST/raw/t10k-images-idx3-ubyte`` exist.</span>
+<span class="sd">        train (bool, optional): If True, creates dataset from ``train-images-idx3-ubyte``,</span>
+<span class="sd">            otherwise from ``t10k-images-idx3-ubyte``.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">mirrors</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="s2">&quot;http://yann.lecun.com/exdb/mnist/&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;https://ossci-datasets.s3.amazonaws.com/mnist/&quot;</span><span class="p">,</span>
+    <span class="p">]</span>
+
+    <span class="n">resources</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">(</span><span class="s2">&quot;train-images-idx3-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;f68b3c2dcbeaaa9fbdd348bbdeb94873&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;train-labels-idx1-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;d53e105ee54ea40749a09fcbcd1e9432&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;t10k-images-idx3-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;9fb629c4189551a2d022fa330f9573f3&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;t10k-labels-idx1-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;ec29112dd5afa0611ce80d1b7f02629c&quot;</span><span class="p">),</span>
+    <span class="p">]</span>
+
+    <span class="n">training_file</span> <span class="o">=</span> <span class="s2">&quot;training.pt&quot;</span>
+    <span class="n">test_file</span> <span class="o">=</span> <span class="s2">&quot;test.pt&quot;</span>
+    <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="s2">&quot;0 - zero&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;1 - one&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;2 - two&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;3 - three&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;4 - four&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;5 - five&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;6 - six&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;7 - seven&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;8 - eight&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;9 - nine&quot;</span><span class="p">,</span>
+    <span class="p">]</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">train_labels</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;train_labels has been renamed targets&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">test_labels</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;test_labels has been renamed targets&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">train_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;train_data has been renamed data&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">test_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;test_data has been renamed data&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="o">=</span> <span class="n">train</span>  <span class="c1"># training set or test set</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_legacy_exist</span><span class="p">():</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_legacy_data</span><span class="p">()</span>
+            <span class="k">return</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_data</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_check_legacy_exist</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">processed_folder_exists</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">processed_folder</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">processed_folder_exists</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">False</span>
+
+        <span class="k">return</span> <span class="nb">all</span><span class="p">(</span>
+            <span class="n">check_integrity</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">processed_folder</span><span class="p">,</span> <span class="n">file</span><span class="p">))</span> <span class="k">for</span> <span class="n">file</span> <span class="ow">in</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">training_file</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">test_file</span><span class="p">)</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_legacy_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="c1"># This is for BC only. We no longer cache the data in a custom binary, but simply read from the raw data</span>
+        <span class="c1"># directly.</span>
+        <span class="n">data_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">training_file</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">test_file</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">processed_folder</span><span class="p">,</span> <span class="n">data_file</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_load_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">image_file</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="s1">&#39;train&#39;</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="w"> </span><span class="k">else</span><span class="w"> </span><span class="s1">&#39;t10k&#39;</span><span class="si">}</span><span class="s2">-images-idx3-ubyte&quot;</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="n">read_image_file</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">image_file</span><span class="p">))</span>
+
+        <span class="n">label_file</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="s1">&#39;train&#39;</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="w"> </span><span class="k">else</span><span class="w"> </span><span class="s1">&#39;t10k&#39;</span><span class="si">}</span><span class="s2">-labels-idx1-ubyte&quot;</span>
+        <span class="n">targets</span> <span class="o">=</span> <span class="n">read_label_file</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">label_file</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="n">data</span><span class="p">,</span> <span class="n">targets</span>
+
+<div class="viewcode-block" id="MNIST.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.MNIST.html#torchvision.datasets.MNIST.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="nb">int</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+
+        <span class="c1"># doing this so that it is consistent with all other datasets</span>
+        <span class="c1"># to return a PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">numpy</span><span class="p">(),</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;L&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">raw_folder</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">,</span> <span class="s2">&quot;raw&quot;</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">processed_folder</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">,</span> <span class="s2">&quot;processed&quot;</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">class_to_idx</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="p">{</span><span class="n">_class</span><span class="p">:</span> <span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">_class</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">)}</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">(</span>
+            <span class="n">check_integrity</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">splitext</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">url</span><span class="p">))[</span><span class="mi">0</span><span class="p">]))</span>
+            <span class="k">for</span> <span class="n">url</span><span class="p">,</span> <span class="n">_</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">resources</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Download the MNIST data if it doesn&#39;t exist already.&quot;&quot;&quot;</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="c1"># download files</span>
+        <span class="k">for</span> <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">resources</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">mirror</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">mirrors</span><span class="p">:</span>
+                <span class="n">url</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">mirror</span><span class="si">}{</span><span class="n">filename</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="k">try</span><span class="p">:</span>
+                    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Downloading </span><span class="si">{</span><span class="n">url</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                    <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span>
+                <span class="k">except</span> <span class="n">URLError</span> <span class="k">as</span> <span class="n">error</span><span class="p">:</span>
+                    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Failed to download (trying next):</span><span class="se">\n</span><span class="si">{</span><span class="n">error</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                    <span class="k">continue</span>
+                <span class="k">finally</span><span class="p">:</span>
+                    <span class="nb">print</span><span class="p">()</span>
+                <span class="k">break</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Error downloading </span><span class="si">{</span><span class="n">filename</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">split</span> <span class="o">=</span> <span class="s2">&quot;Train&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="ow">is</span> <span class="kc">True</span> <span class="k">else</span> <span class="s2">&quot;Test&quot;</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;Split: </span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">&quot;</span></div>
+
+
+<div class="viewcode-block" id="FashionMNIST"><a class="viewcode-back" href="../../../generated/torchvision.datasets.FashionMNIST.html#torchvision.datasets.FashionMNIST">[docs]</a><span class="k">class</span> <span class="nc">FashionMNIST</span><span class="p">(</span><span class="n">MNIST</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Fashion-MNIST &lt;https://github.com/zalandoresearch/fashion-mnist&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where ``FashionMNIST/raw/train-images-idx3-ubyte``</span>
+<span class="sd">            and  ``FashionMNIST/raw/t10k-images-idx3-ubyte`` exist.</span>
+<span class="sd">        train (bool, optional): If True, creates dataset from ``train-images-idx3-ubyte``,</span>
+<span class="sd">            otherwise from ``t10k-images-idx3-ubyte``.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">mirrors</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/&quot;</span><span class="p">]</span>
+
+    <span class="n">resources</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">(</span><span class="s2">&quot;train-images-idx3-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;8d4fb7e6c68d591d4c3dfef9ec88bf0d&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;train-labels-idx1-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;25c81989df183df01b3e8a0aad5dffbe&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;t10k-images-idx3-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;bef4ecab320f06d8554ea6380940ec79&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;t10k-labels-idx1-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;bb300cfdad3c16e7a12a480ee83cd310&quot;</span><span class="p">),</span>
+    <span class="p">]</span>
+    <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;T-shirt/top&quot;</span><span class="p">,</span> <span class="s2">&quot;Trouser&quot;</span><span class="p">,</span> <span class="s2">&quot;Pullover&quot;</span><span class="p">,</span> <span class="s2">&quot;Dress&quot;</span><span class="p">,</span> <span class="s2">&quot;Coat&quot;</span><span class="p">,</span> <span class="s2">&quot;Sandal&quot;</span><span class="p">,</span> <span class="s2">&quot;Shirt&quot;</span><span class="p">,</span> <span class="s2">&quot;Sneaker&quot;</span><span class="p">,</span> <span class="s2">&quot;Bag&quot;</span><span class="p">,</span> <span class="s2">&quot;Ankle boot&quot;</span><span class="p">]</span></div>
+
+
+<div class="viewcode-block" id="KMNIST"><a class="viewcode-back" href="../../../generated/torchvision.datasets.KMNIST.html#torchvision.datasets.KMNIST">[docs]</a><span class="k">class</span> <span class="nc">KMNIST</span><span class="p">(</span><span class="n">MNIST</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Kuzushiji-MNIST &lt;https://github.com/rois-codh/kmnist&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where ``KMNIST/raw/train-images-idx3-ubyte``</span>
+<span class="sd">            and  ``KMNIST/raw/t10k-images-idx3-ubyte`` exist.</span>
+<span class="sd">        train (bool, optional): If True, creates dataset from ``train-images-idx3-ubyte``,</span>
+<span class="sd">            otherwise from ``t10k-images-idx3-ubyte``.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">mirrors</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;http://codh.rois.ac.jp/kmnist/dataset/kmnist/&quot;</span><span class="p">]</span>
+
+    <span class="n">resources</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">(</span><span class="s2">&quot;train-images-idx3-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;bdb82020997e1d708af4cf47b453dcf7&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;train-labels-idx1-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;e144d726b3acfaa3e44228e80efcd344&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;t10k-images-idx3-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;5c965bf0a639b31b8f53240b1b52f4d7&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;t10k-labels-idx1-ubyte.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;7320c461ea6c1c855c0b718fb2a4b134&quot;</span><span class="p">),</span>
+    <span class="p">]</span>
+    <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;o&quot;</span><span class="p">,</span> <span class="s2">&quot;ki&quot;</span><span class="p">,</span> <span class="s2">&quot;su&quot;</span><span class="p">,</span> <span class="s2">&quot;tsu&quot;</span><span class="p">,</span> <span class="s2">&quot;na&quot;</span><span class="p">,</span> <span class="s2">&quot;ha&quot;</span><span class="p">,</span> <span class="s2">&quot;ma&quot;</span><span class="p">,</span> <span class="s2">&quot;ya&quot;</span><span class="p">,</span> <span class="s2">&quot;re&quot;</span><span class="p">,</span> <span class="s2">&quot;wo&quot;</span><span class="p">]</span></div>
+
+
+<div class="viewcode-block" id="EMNIST"><a class="viewcode-back" href="../../../generated/torchvision.datasets.EMNIST.html#torchvision.datasets.EMNIST">[docs]</a><span class="k">class</span> <span class="nc">EMNIST</span><span class="p">(</span><span class="n">MNIST</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`EMNIST &lt;https://www.westernsydney.edu.au/bens/home/reproducible_research/emnist&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where ``EMNIST/raw/train-images-idx3-ubyte``</span>
+<span class="sd">            and  ``EMNIST/raw/t10k-images-idx3-ubyte`` exist.</span>
+<span class="sd">        split (string): The dataset has 6 different splits: ``byclass``, ``bymerge``,</span>
+<span class="sd">            ``balanced``, ``letters``, ``digits`` and ``mnist``. This argument specifies</span>
+<span class="sd">            which one to use.</span>
+<span class="sd">        train (bool, optional): If True, creates dataset from ``training.pt``,</span>
+<span class="sd">            otherwise from ``test.pt``.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">url</span> <span class="o">=</span> <span class="s2">&quot;https://www.itl.nist.gov/iaui/vip/cs_links/EMNIST/gzip.zip&quot;</span>
+    <span class="n">md5</span> <span class="o">=</span> <span class="s2">&quot;58c8d27c78d21e728a6bc7b3cc06412e&quot;</span>
+    <span class="n">splits</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;byclass&quot;</span><span class="p">,</span> <span class="s2">&quot;bymerge&quot;</span><span class="p">,</span> <span class="s2">&quot;balanced&quot;</span><span class="p">,</span> <span class="s2">&quot;letters&quot;</span><span class="p">,</span> <span class="s2">&quot;digits&quot;</span><span class="p">,</span> <span class="s2">&quot;mnist&quot;</span><span class="p">)</span>
+    <span class="c1"># Merged Classes assumes Same structure for both uppercase and lowercase version</span>
+    <span class="n">_merged_classes</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;c&quot;</span><span class="p">,</span> <span class="s2">&quot;i&quot;</span><span class="p">,</span> <span class="s2">&quot;j&quot;</span><span class="p">,</span> <span class="s2">&quot;k&quot;</span><span class="p">,</span> <span class="s2">&quot;l&quot;</span><span class="p">,</span> <span class="s2">&quot;m&quot;</span><span class="p">,</span> <span class="s2">&quot;o&quot;</span><span class="p">,</span> <span class="s2">&quot;p&quot;</span><span class="p">,</span> <span class="s2">&quot;s&quot;</span><span class="p">,</span> <span class="s2">&quot;u&quot;</span><span class="p">,</span> <span class="s2">&quot;v&quot;</span><span class="p">,</span> <span class="s2">&quot;w&quot;</span><span class="p">,</span> <span class="s2">&quot;x&quot;</span><span class="p">,</span> <span class="s2">&quot;y&quot;</span><span class="p">,</span> <span class="s2">&quot;z&quot;</span><span class="p">}</span>
+    <span class="n">_all_classes</span> <span class="o">=</span> <span class="nb">set</span><span class="p">(</span><span class="n">string</span><span class="o">.</span><span class="n">digits</span> <span class="o">+</span> <span class="n">string</span><span class="o">.</span><span class="n">ascii_letters</span><span class="p">)</span>
+    <span class="n">classes_split_dict</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;byclass&quot;</span><span class="p">:</span> <span class="nb">sorted</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="n">_all_classes</span><span class="p">)),</span>
+        <span class="s2">&quot;bymerge&quot;</span><span class="p">:</span> <span class="nb">sorted</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="n">_all_classes</span> <span class="o">-</span> <span class="n">_merged_classes</span><span class="p">)),</span>
+        <span class="s2">&quot;balanced&quot;</span><span class="p">:</span> <span class="nb">sorted</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="n">_all_classes</span> <span class="o">-</span> <span class="n">_merged_classes</span><span class="p">)),</span>
+        <span class="s2">&quot;letters&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;N/A&quot;</span><span class="p">]</span> <span class="o">+</span> <span class="nb">list</span><span class="p">(</span><span class="n">string</span><span class="o">.</span><span class="n">ascii_lowercase</span><span class="p">),</span>
+        <span class="s2">&quot;digits&quot;</span><span class="p">:</span> <span class="nb">list</span><span class="p">(</span><span class="n">string</span><span class="o">.</span><span class="n">digits</span><span class="p">),</span>
+        <span class="s2">&quot;mnist&quot;</span><span class="p">:</span> <span class="nb">list</span><span class="p">(</span><span class="n">string</span><span class="o">.</span><span class="n">digits</span><span class="p">),</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">splits</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">training_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_training_file</span><span class="p">(</span><span class="n">split</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">test_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_test_file</span><span class="p">(</span><span class="n">split</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classes_split_dict</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">]</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">_training_file</span><span class="p">(</span><span class="n">split</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;training_</span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">.pt&quot;</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">_test_file</span><span class="p">(</span><span class="n">split</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;test_</span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">.pt&quot;</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">_file_prefix</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;emnist-</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="s1">&#39;train&#39;</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="w"> </span><span class="k">else</span><span class="w"> </span><span class="s1">&#39;test&#39;</span><span class="si">}</span><span class="s2">&quot;</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">images_file</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_file_prefix</span><span class="si">}</span><span class="s2">-images-idx3-ubyte&quot;</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">labels_file</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_file_prefix</span><span class="si">}</span><span class="s2">-labels-idx1-ubyte&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">read_image_file</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_file</span><span class="p">),</span> <span class="n">read_label_file</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">(</span><span class="n">check_integrity</span><span class="p">(</span><span class="n">file</span><span class="p">)</span> <span class="k">for</span> <span class="n">file</span> <span class="ow">in</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_file</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Download the EMNIST data if it doesn&#39;t exist already.&quot;&quot;&quot;</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">md5</span><span class="p">)</span>
+        <span class="n">gzip_folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="s2">&quot;gzip&quot;</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">gzip_file</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">gzip_folder</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">gzip_file</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;.gz&quot;</span><span class="p">):</span>
+                <span class="n">extract_archive</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">gzip_folder</span><span class="p">,</span> <span class="n">gzip_file</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">)</span>
+        <span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span><span class="p">(</span><span class="n">gzip_folder</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="QMNIST"><a class="viewcode-back" href="../../../generated/torchvision.datasets.QMNIST.html#torchvision.datasets.QMNIST">[docs]</a><span class="k">class</span> <span class="nc">QMNIST</span><span class="p">(</span><span class="n">MNIST</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`QMNIST &lt;https://github.com/facebookresearch/qmnist&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset whose ``raw``</span>
+<span class="sd">            subdir contains binary files of the datasets.</span>
+<span class="sd">        what (string,optional): Can be &#39;train&#39;, &#39;test&#39;, &#39;test10k&#39;,</span>
+<span class="sd">            &#39;test50k&#39;, or &#39;nist&#39; for respectively the mnist compatible</span>
+<span class="sd">            training set, the 60k qmnist testing set, the 10k qmnist</span>
+<span class="sd">            examples that match the mnist testing set, the 50k</span>
+<span class="sd">            remaining qmnist testing examples, or all the nist</span>
+<span class="sd">            digits. The default is to select &#39;train&#39; or &#39;test&#39;</span>
+<span class="sd">            according to the compatibility argument &#39;train&#39;.</span>
+<span class="sd">        compat (bool,optional): A boolean that says whether the target</span>
+<span class="sd">            for each example is class number (for compatibility with</span>
+<span class="sd">            the MNIST dataloader) or a torch vector containing the</span>
+<span class="sd">            full qmnist information. Default=True.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from</span>
+<span class="sd">            the internet and puts it in root directory. If dataset is</span>
+<span class="sd">            already downloaded, it is not downloaded again.</span>
+<span class="sd">        transform (callable, optional): A function/transform that</span>
+<span class="sd">            takes in an PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform</span>
+<span class="sd">            that takes in the target and transforms it.</span>
+<span class="sd">        train (bool,optional,compatibility): When argument &#39;what&#39; is</span>
+<span class="sd">            not specified, this boolean decides whether to load the</span>
+<span class="sd">            training set or the testing set.  Default: True.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">subsets</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;test10k&quot;</span><span class="p">:</span> <span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;test50k&quot;</span><span class="p">:</span> <span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;nist&quot;</span><span class="p">:</span> <span class="s2">&quot;nist&quot;</span><span class="p">}</span>
+    <span class="n">resources</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]]</span> <span class="o">=</span> <span class="p">{</span>  <span class="c1"># type: ignore[assignment]</span>
+        <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">(</span>
+                <span class="s2">&quot;https://raw.githubusercontent.com/facebookresearch/qmnist/master/qmnist-train-images-idx3-ubyte.gz&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;ed72d4157d28c017586c42bc6afe6370&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="p">(</span>
+                <span class="s2">&quot;https://raw.githubusercontent.com/facebookresearch/qmnist/master/qmnist-train-labels-idx2-int.gz&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;0058f8dd561b90ffdd0f734c6a30e5e4&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">(</span>
+                <span class="s2">&quot;https://raw.githubusercontent.com/facebookresearch/qmnist/master/qmnist-test-images-idx3-ubyte.gz&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;1394631089c404de565df7b7aeaf9412&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="p">(</span>
+                <span class="s2">&quot;https://raw.githubusercontent.com/facebookresearch/qmnist/master/qmnist-test-labels-idx2-int.gz&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;5b5b05890a5e13444e108efe57b788aa&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;nist&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">(</span>
+                <span class="s2">&quot;https://raw.githubusercontent.com/facebookresearch/qmnist/master/xnist-images-idx3-ubyte.xz&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;7f124b3b8ab81486c9d8c2749c17f834&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="p">(</span>
+                <span class="s2">&quot;https://raw.githubusercontent.com/facebookresearch/qmnist/master/xnist-labels-idx2-int.xz&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;5ed0e788978e45d4a8bd4b7caec3d79d&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">],</span>
+    <span class="p">}</span>
+    <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="s2">&quot;0 - zero&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;1 - one&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;2 - two&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;3 - three&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;4 - four&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;5 - five&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;6 - six&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;7 - seven&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;8 - eight&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;9 - nine&quot;</span><span class="p">,</span>
+    <span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">what</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">compat</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">what</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">what</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span> <span class="k">if</span> <span class="n">train</span> <span class="k">else</span> <span class="s2">&quot;test&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">what</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">what</span><span class="p">,</span> <span class="s2">&quot;what&quot;</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">subsets</span><span class="o">.</span><span class="n">keys</span><span class="p">()))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">compat</span> <span class="o">=</span> <span class="n">compat</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_file</span> <span class="o">=</span> <span class="n">what</span> <span class="o">+</span> <span class="s2">&quot;.pt&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">training_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_file</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">test_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_file</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">train</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">images_file</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="n">_</span><span class="p">),</span> <span class="n">_</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">resources</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">subsets</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">what</span><span class="p">]]</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">splitext</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">url</span><span class="p">))[</span><span class="mi">0</span><span class="p">])</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">labels_file</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">_</span><span class="p">,</span> <span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="n">_</span><span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">resources</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">subsets</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">what</span><span class="p">]]</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">splitext</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">url</span><span class="p">))[</span><span class="mi">0</span><span class="p">])</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">(</span><span class="n">check_integrity</span><span class="p">(</span><span class="n">file</span><span class="p">)</span> <span class="k">for</span> <span class="n">file</span> <span class="ow">in</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_file</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_load_data</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="n">read_sn3_pascalvincent_tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_file</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">data</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;data should be of dtype torch.uint8 instead of </span><span class="si">{</span><span class="n">data</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">data</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;data should have 3 dimensions instead of {data.ndimension()}&quot;</span><span class="p">)</span>
+
+        <span class="n">targets</span> <span class="o">=</span> <span class="n">read_sn3_pascalvincent_tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">labels_file</span><span class="p">)</span><span class="o">.</span><span class="n">long</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">targets</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;targets should have 2 dimensions instead of </span><span class="si">{</span><span class="n">targets</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">what</span> <span class="o">==</span> <span class="s2">&quot;test10k&quot;</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">data</span><span class="p">[</span><span class="mi">0</span><span class="p">:</span><span class="mi">10000</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+            <span class="n">targets</span> <span class="o">=</span> <span class="n">targets</span><span class="p">[</span><span class="mi">0</span><span class="p">:</span><span class="mi">10000</span><span class="p">,</span> <span class="p">:]</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">what</span> <span class="o">==</span> <span class="s2">&quot;test50k&quot;</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">data</span><span class="p">[</span><span class="mi">10000</span><span class="p">:,</span> <span class="p">:,</span> <span class="p">:]</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+            <span class="n">targets</span> <span class="o">=</span> <span class="n">targets</span><span class="p">[</span><span class="mi">10000</span><span class="p">:,</span> <span class="p">:]</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+
+        <span class="k">return</span> <span class="n">data</span><span class="p">,</span> <span class="n">targets</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Download the QMNIST data if it doesn&#39;t exist already.</span>
+<span class="sd">        Note that we only download what has been asked for (argument &#39;what&#39;).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">split</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">resources</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">subsets</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">what</span><span class="p">]]</span>
+
+        <span class="k">for</span> <span class="n">url</span><span class="p">,</span> <span class="n">md5</span> <span class="ow">in</span> <span class="n">split</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">raw_folder</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span>
+
+<div class="viewcode-block" id="QMNIST.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.QMNIST.html#torchvision.datasets.QMNIST.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="c1"># redefined to handle the compat flag</span>
+        <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">numpy</span><span class="p">(),</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;L&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">compat</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">target</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;Split: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">what</span><span class="si">}</span><span class="s2">&quot;</span></div>
+
+
+<span class="k">def</span> <span class="nf">get_int</span><span class="p">(</span><span class="n">b</span><span class="p">:</span> <span class="nb">bytes</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="n">codecs</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="s2">&quot;hex&quot;</span><span class="p">),</span> <span class="mi">16</span><span class="p">)</span>
+
+
+<span class="n">SN3_PASCALVINCENT_TYPEMAP</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="mi">8</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">,</span>
+    <span class="mi">9</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">int8</span><span class="p">,</span>
+    <span class="mi">11</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">int16</span><span class="p">,</span>
+    <span class="mi">12</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span>
+    <span class="mi">13</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span>
+    <span class="mi">14</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">float64</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<span class="k">def</span> <span class="nf">read_sn3_pascalvincent_tensor</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">strict</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Read a SN3 file in &quot;Pascal Vincent&quot; format (Lush file &#39;libidx/idx-io.lsh&#39;).</span>
+<span class="sd">    Argument may be a filename, compressed filename, or file object.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="c1"># read</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">read</span><span class="p">()</span>
+    <span class="c1"># parse</span>
+    <span class="n">magic</span> <span class="o">=</span> <span class="n">get_int</span><span class="p">(</span><span class="n">data</span><span class="p">[</span><span class="mi">0</span><span class="p">:</span><span class="mi">4</span><span class="p">])</span>
+    <span class="n">nd</span> <span class="o">=</span> <span class="n">magic</span> <span class="o">%</span> <span class="mi">256</span>
+    <span class="n">ty</span> <span class="o">=</span> <span class="n">magic</span> <span class="o">//</span> <span class="mi">256</span>
+    <span class="k">assert</span> <span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">nd</span> <span class="o">&lt;=</span> <span class="mi">3</span>
+    <span class="k">assert</span> <span class="mi">8</span> <span class="o">&lt;=</span> <span class="n">ty</span> <span class="o">&lt;=</span> <span class="mi">14</span>
+    <span class="n">torch_type</span> <span class="o">=</span> <span class="n">SN3_PASCALVINCENT_TYPEMAP</span><span class="p">[</span><span class="n">ty</span><span class="p">]</span>
+    <span class="n">s</span> <span class="o">=</span> <span class="p">[</span><span class="n">get_int</span><span class="p">(</span><span class="n">data</span><span class="p">[</span><span class="mi">4</span> <span class="o">*</span> <span class="p">(</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="p">:</span> <span class="mi">4</span> <span class="o">*</span> <span class="p">(</span><span class="n">i</span> <span class="o">+</span> <span class="mi">2</span><span class="p">)])</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">nd</span><span class="p">)]</span>
+
+    <span class="n">parsed</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">frombuffer</span><span class="p">(</span><span class="nb">bytearray</span><span class="p">(</span><span class="n">data</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch_type</span><span class="p">,</span> <span class="n">offset</span><span class="o">=</span><span class="p">(</span><span class="mi">4</span> <span class="o">*</span> <span class="p">(</span><span class="n">nd</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)))</span>
+
+    <span class="c1"># The MNIST format uses the big endian byte order, while `torch.frombuffer` uses whatever the system uses. In case</span>
+    <span class="c1"># that is little endian and the dtype has more than one byte, we need to flip them.</span>
+    <span class="k">if</span> <span class="n">sys</span><span class="o">.</span><span class="n">byteorder</span> <span class="o">==</span> <span class="s2">&quot;little&quot;</span> <span class="ow">and</span> <span class="n">parsed</span><span class="o">.</span><span class="n">element_size</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">parsed</span> <span class="o">=</span> <span class="n">_flip_byte_order</span><span class="p">(</span><span class="n">parsed</span><span class="p">)</span>
+
+    <span class="k">assert</span> <span class="n">parsed</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">prod</span><span class="p">(</span><span class="n">s</span><span class="p">)</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">strict</span>
+    <span class="k">return</span> <span class="n">parsed</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">*</span><span class="n">s</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">read_label_file</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">read_sn3_pascalvincent_tensor</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;x should be of dtype torch.uint8 instead of </span><span class="si">{</span><span class="n">x</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;x should have 1 dimension instead of </span><span class="si">{</span><span class="n">x</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">x</span><span class="o">.</span><span class="n">long</span><span class="p">()</span>
+
+
+<span class="k">def</span> <span class="nf">read_image_file</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">read_sn3_pascalvincent_tensor</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;x should be of dtype torch.uint8 instead of </span><span class="si">{</span><span class="n">x</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;x should have 3 dimension instead of </span><span class="si">{</span><span class="n">x</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">x</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/moving_mnist.html b/0.16/_modules/torchvision/datasets/moving_mnist.html
new file mode 100644
index 00000000000..a606375a723
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/moving_mnist.html
@@ -0,0 +1,762 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.moving_mnist &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.moving_mnist</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.moving_mnist</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision.datasets.utils</span> <span class="kn">import</span> <span class="n">download_url</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">torchvision.datasets.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="MovingMNIST"><a class="viewcode-back" href="../../../generated/torchvision.datasets.MovingMNIST.html#torchvision.datasets.MovingMNIST">[docs]</a><span class="k">class</span> <span class="nc">MovingMNIST</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`MovingMNIST &lt;http://www.cs.toronto.edu/~nitish/unsupervised_video/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where ``MovingMNIST/mnist_test_seq.npy`` exists.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``None`` (default), ``&quot;train&quot;`` and ``&quot;test&quot;``.</span>
+<span class="sd">            If ``split=None``, the full data is returned.</span>
+<span class="sd">        split_ratio (int, optional): The split ratio of number of frames. If ``split=&quot;train&quot;``, the first split</span>
+<span class="sd">            frames ``data[:, :split_ratio]`` is returned. If ``split=&quot;test&quot;``, the last split frames ``data[:, split_ratio:]``</span>
+<span class="sd">            is returned. If ``split=None``, this parameter is ignored and the all frames data is returned.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in an torch Tensor</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_URL</span> <span class="o">=</span> <span class="s2">&quot;http://www.cs.toronto.edu/~nitish/unsupervised_video/mnist_test_seq.npy&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">split_ratio</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">10</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_filename</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_URL</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;/&quot;</span><span class="p">)[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">split</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">split</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">split_ratio</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;`split_ratio` should be an integer, but got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">split_ratio</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">split_ratio</span> <span class="o">&lt;=</span> <span class="mi">19</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;`split_ratio` should be `1 &lt;= split_ratio &lt;= 19`, but got </span><span class="si">{</span><span class="n">split_ratio</span><span class="si">}</span><span class="s2"> instead.&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split_ratio</span> <span class="o">=</span> <span class="n">split_ratio</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it.&quot;</span><span class="p">)</span>
+
+        <span class="n">data</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_filename</span><span class="p">)))</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">data</span><span class="p">[:</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_ratio</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">data</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">split_ratio</span> <span class="p">:]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+
+<div class="viewcode-block" id="MovingMNIST.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.MovingMNIST.html#torchvision.datasets.MovingMNIST.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            torch.Tensor: Video frames (torch Tensor[T, C, H, W]). The `T` is the number of frames.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">data</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">data</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_filename</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">download_url</span><span class="p">(</span>
+            <span class="n">url</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_URL</span><span class="p">,</span>
+            <span class="n">root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">,</span>
+            <span class="n">filename</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_filename</span><span class="p">,</span>
+            <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;be083ec986bfe91a449d63653c411eb2&quot;</span><span class="p">,</span>
+        <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/omniglot.html b/0.16/_modules/torchvision/datasets/omniglot.html
new file mode 100644
index 00000000000..e7651c15dbe
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/omniglot.html
@@ -0,0 +1,771 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.omniglot &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.omniglot</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.omniglot</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">os.path</span> <span class="kn">import</span> <span class="n">join</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">list_dir</span><span class="p">,</span> <span class="n">list_files</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="Omniglot"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Omniglot.html#torchvision.datasets.Omniglot">[docs]</a><span class="k">class</span> <span class="nc">Omniglot</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Omniglot &lt;https://github.com/brendenlake/omniglot&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``omniglot-py`` exists.</span>
+<span class="sd">        background (bool, optional): If True, creates dataset from the &quot;background&quot; set, otherwise</span>
+<span class="sd">            creates from the &quot;evaluation&quot; set. This terminology is defined by the authors.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset zip files from the internet and</span>
+<span class="sd">            puts it in root directory. If the zip files are already downloaded, they are not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">folder</span> <span class="o">=</span> <span class="s2">&quot;omniglot-py&quot;</span>
+    <span class="n">download_url_prefix</span> <span class="o">=</span> <span class="s2">&quot;https://raw.githubusercontent.com/brendenlake/omniglot/master/python&quot;</span>
+    <span class="n">zips_md5</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;images_background&quot;</span><span class="p">:</span> <span class="s2">&quot;68d2efa1b9178cc56df9314c21c6e718&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;images_evaluation&quot;</span><span class="p">:</span> <span class="s2">&quot;6b91aef0f799c5bb55b94e3f2daec811&quot;</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">background</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">folder</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">background</span> <span class="o">=</span> <span class="n">background</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_folder</span> <span class="o">=</span> <span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_target_folder</span><span class="p">())</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_alphabets</span> <span class="o">=</span> <span class="n">list_dir</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">target_folder</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_characters</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span>
+            <span class="p">([</span><span class="n">join</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">list_dir</span><span class="p">(</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">target_folder</span><span class="p">,</span> <span class="n">a</span><span class="p">))]</span> <span class="k">for</span> <span class="n">a</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_alphabets</span><span class="p">),</span> <span class="p">[]</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_character_images</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="p">[(</span><span class="n">image</span><span class="p">,</span> <span class="n">idx</span><span class="p">)</span> <span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">list_files</span><span class="p">(</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">target_folder</span><span class="p">,</span> <span class="n">character</span><span class="p">),</span> <span class="s2">&quot;.png&quot;</span><span class="p">)]</span>
+            <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">character</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_characters</span><span class="p">)</span>
+        <span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_flat_character_images</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_character_images</span><span class="p">,</span> <span class="p">[])</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_flat_character_images</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Omniglot.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Omniglot.html#torchvision.datasets.Omniglot.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target character class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">image_name</span><span class="p">,</span> <span class="n">character_class</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_flat_character_images</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">image_path</span> <span class="o">=</span> <span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">target_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_characters</span><span class="p">[</span><span class="n">character_class</span><span class="p">],</span> <span class="n">image_name</span><span class="p">)</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_path</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;r&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;L&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">character_class</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">character_class</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">character_class</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="n">zip_filename</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_target_folder</span><span class="p">()</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">zip_filename</span> <span class="o">+</span> <span class="s2">&quot;.zip&quot;</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">zips_md5</span><span class="p">[</span><span class="n">zip_filename</span><span class="p">]):</span>
+            <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="n">filename</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_target_folder</span><span class="p">()</span>
+        <span class="n">zip_filename</span> <span class="o">=</span> <span class="n">filename</span> <span class="o">+</span> <span class="s2">&quot;.zip&quot;</span>
+        <span class="n">url</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">download_url_prefix</span> <span class="o">+</span> <span class="s2">&quot;/&quot;</span> <span class="o">+</span> <span class="n">zip_filename</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="n">zip_filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">zips_md5</span><span class="p">[</span><span class="n">filename</span><span class="p">])</span>
+
+    <span class="k">def</span> <span class="nf">_get_target_folder</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;images_background&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">background</span> <span class="k">else</span> <span class="s2">&quot;images_evaluation&quot;</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html b/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
new file mode 100644
index 00000000000..f7fd048bb00
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
@@ -0,0 +1,794 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.oxford_iiit_pet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.oxford_iiit_pet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.oxford_iiit_pet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">import</span> <span class="nn">pathlib</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="OxfordIIITPet"><a class="viewcode-back" href="../../../generated/torchvision.datasets.OxfordIIITPet.html#torchvision.datasets.OxfordIIITPet">[docs]</a><span class="k">class</span> <span class="nc">OxfordIIITPet</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Oxford-IIIT Pet Dataset   &lt;https://www.robots.ox.ac.uk/~vgg/data/pets/&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;trainval&quot;`` (default) or ``&quot;test&quot;``.</span>
+<span class="sd">        target_types (string, sequence of strings, optional): Types of target to use. Can be ``category`` (default) or</span>
+<span class="sd">            ``segmentation``. Can also be a list to output a tuple with all specified target types. The types represent:</span>
+
+<span class="sd">                - ``category`` (int): Label for one of the 37 pet categories.</span>
+<span class="sd">                - ``segmentation`` (PIL image): Segmentation trimap of the image.</span>
+
+<span class="sd">            If empty, ``None`` will be returned as target.</span>
+
+<span class="sd">        transform (callable, optional): A function/transform that  takes in a PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and puts it into</span>
+<span class="sd">            ``root/oxford-iiit-pet``. If dataset is already downloaded, it is not downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_RESOURCES</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="p">(</span><span class="s2">&quot;https://www.robots.ox.ac.uk/~vgg/data/pets/data/images.tar.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;5c4f3ee8e5d25df40f4fd59a7f44e54c&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;https://www.robots.ox.ac.uk/~vgg/data/pets/data/annotations.tar.gz&quot;</span><span class="p">,</span> <span class="s2">&quot;95a8c909bbe2e81eed6a22bccdf3f68f&quot;</span><span class="p">),</span>
+    <span class="p">)</span>
+    <span class="n">_VALID_TARGET_TYPES</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;category&quot;</span><span class="p">,</span> <span class="s2">&quot;segmentation&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;trainval&quot;</span><span class="p">,</span>
+        <span class="n">target_types</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Sequence</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;category&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;trainval&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">target_types</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="n">target_types</span> <span class="o">=</span> <span class="p">[</span><span class="n">target_types</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_target_types</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">target_type</span><span class="p">,</span> <span class="s2">&quot;target_types&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_VALID_TARGET_TYPES</span><span class="p">)</span> <span class="k">for</span> <span class="n">target_type</span> <span class="ow">in</span> <span class="n">target_types</span>
+        <span class="p">]</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;oxford-iiit-pet&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;images&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_anns_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;annotations&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_segs_folder</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_anns_folder</span> <span class="o">/</span> <span class="s2">&quot;trimaps&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="n">image_ids</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_anns_folder</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}</span><span class="s2">.txt&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">file</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">file</span><span class="p">:</span>
+                <span class="n">image_id</span><span class="p">,</span> <span class="n">label</span><span class="p">,</span> <span class="o">*</span><span class="n">_</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">()</span>
+                <span class="n">image_ids</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">image_id</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">label</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="s2">&quot; &quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">part</span><span class="o">.</span><span class="n">title</span><span class="p">()</span> <span class="k">for</span> <span class="n">part</span> <span class="ow">in</span> <span class="n">raw_cls</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;_&quot;</span><span class="p">))</span>
+            <span class="k">for</span> <span class="n">raw_cls</span><span class="p">,</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">sorted</span><span class="p">(</span>
+                <span class="p">{(</span><span class="n">image_id</span><span class="o">.</span><span class="n">rsplit</span><span class="p">(</span><span class="s2">&quot;_&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">],</span> <span class="n">label</span><span class="p">)</span> <span class="k">for</span> <span class="n">image_id</span><span class="p">,</span> <span class="n">label</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">image_ids</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">)},</span>
+                <span class="n">key</span><span class="o">=</span><span class="k">lambda</span> <span class="n">image_id_and_label</span><span class="p">:</span> <span class="n">image_id_and_label</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>
+            <span class="p">)</span>
+        <span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">))))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_images</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">image_id</span><span class="si">}</span><span class="s2">.jpg&quot;</span> <span class="k">for</span> <span class="n">image_id</span> <span class="ow">in</span> <span class="n">image_ids</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_segs</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_segs_folder</span> <span class="o">/</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">image_id</span><span class="si">}</span><span class="s2">.png&quot;</span> <span class="k">for</span> <span class="n">image_id</span> <span class="ow">in</span> <span class="n">image_ids</span><span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">)</span>
+
+<div class="viewcode-block" id="OxfordIIITPet.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.OxfordIIITPet.html#torchvision.datasets.OxfordIIITPet.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images</span><span class="p">[</span><span class="n">idx</span><span class="p">])</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="n">target</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">target_type</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_target_types</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">target_type</span> <span class="o">==</span> <span class="s2">&quot;category&quot;</span><span class="p">:</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">[</span><span class="n">idx</span><span class="p">])</span>
+            <span class="k">else</span><span class="p">:</span>  <span class="c1"># target_type == &quot;segmentation&quot;</span>
+                <span class="n">target</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_segs</span><span class="p">[</span><span class="n">idx</span><span class="p">]))</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">target</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">target</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="n">target</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">folder</span> <span class="ow">in</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_anns_folder</span><span class="p">):</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">folder</span><span class="p">)</span> <span class="ow">and</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">folder</span><span class="p">)):</span>
+                <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="k">for</span> <span class="n">url</span><span class="p">,</span> <span class="n">md5</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_RESOURCES</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/pcam.html b/0.16/_modules/torchvision/datasets/pcam.html
new file mode 100644
index 00000000000..2e9ac21fc89
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/pcam.html
@@ -0,0 +1,799 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.pcam &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.pcam</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.pcam</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">pathlib</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_decompress</span><span class="p">,</span> <span class="n">download_file_from_google_drive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="PCAM"><a class="viewcode-back" href="../../../generated/torchvision.datasets.PCAM.html#torchvision.datasets.PCAM">[docs]</a><span class="k">class</span> <span class="nc">PCAM</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`PCAM Dataset   &lt;https://github.com/basveeling/pcam&gt;`_.</span>
+
+<span class="sd">    The PatchCamelyon dataset is a binary classification dataset with 327,680</span>
+<span class="sd">    color images (96px x 96px), extracted from histopathologic scans of lymph node</span>
+<span class="sd">    sections. Each image is annotated with a binary label indicating presence of</span>
+<span class="sd">    metastatic tissue.</span>
+
+<span class="sd">    This dataset requires the ``h5py`` package which you can install with ``pip install h5py``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">         root (string): Root directory of the dataset.</span>
+<span class="sd">         split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default), ``&quot;test&quot;`` or ``&quot;val&quot;``.</span>
+<span class="sd">         transform (callable, optional): A function/transform that  takes in a PIL image and returns a transformed</span>
+<span class="sd">             version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">         target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">         download (bool, optional): If True, downloads the dataset from the internet and puts it into ``root/pcam``. If</span>
+<span class="sd">             dataset is already downloaded, it is not downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_FILES</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="p">{</span>
+            <span class="s2">&quot;images&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;camelyonpatch_level_2_split_train_x.h5&quot;</span><span class="p">,</span>  <span class="c1"># Data file name</span>
+                <span class="s2">&quot;1Ka0XfEMiwgCYPdTI-vv6eUElOBnKFKQ2&quot;</span><span class="p">,</span>  <span class="c1"># Google Drive ID</span>
+                <span class="s2">&quot;1571f514728f59376b705fc836ff4b63&quot;</span><span class="p">,</span>  <span class="c1"># md5 hash</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;targets&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;camelyonpatch_level_2_split_train_y.h5&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;1269yhu3pZDP8UYFQs-NYs3FPwuK-nGSG&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;35c2d7259d906cfc8143347bb8e05be7&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+        <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="p">{</span>
+            <span class="s2">&quot;images&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;camelyonpatch_level_2_split_test_x.h5&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;1qV65ZqZvWzuIVthK8eVDhIwrbnsJdbg_&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;d8c2d60d490dbd479f8199bdfa0cf6ec&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;targets&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;camelyonpatch_level_2_split_test_y.h5&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;17BHrSrwWKjYsOgTMmoqrIjDy6Fa2o_gP&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;60a7035772fbdb7f34eb86d4420cf66a&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+        <span class="s2">&quot;val&quot;</span><span class="p">:</span> <span class="p">{</span>
+            <span class="s2">&quot;images&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;camelyonpatch_level_2_split_valid_x.h5&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;1hgshYGWK8V-eGRy8LToWJJgDU_rXWVJ3&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;d5b63470df7cfa627aeec8b9dc0c066e&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;targets&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;camelyonpatch_level_2_split_valid_y.h5&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;1bH8ZRbhSVAhScTS0p9-ZzGnX91cHT3uO&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;2b85f58b927af9964a4c15b8f7e8f179&quot;</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="kn">import</span> <span class="nn">h5py</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">h5py</span> <span class="o">=</span> <span class="n">h5py</span>
+        <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="s2">&quot;h5py is not found. This dataset needs to have h5py installed: please run pip install h5py&quot;</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">))</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;pcam&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="n">images_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_FILES</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">][</span><span class="s2">&quot;images&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">h5py</span><span class="o">.</span><span class="n">File</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="n">images_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">images_data</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">images_data</span><span class="p">[</span><span class="s2">&quot;x&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+<div class="viewcode-block" id="PCAM.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.PCAM.html#torchvision.datasets.PCAM.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">images_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_FILES</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">][</span><span class="s2">&quot;images&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">h5py</span><span class="o">.</span><span class="n">File</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="n">images_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">images_data</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">images_data</span><span class="p">[</span><span class="s2">&quot;x&quot;</span><span class="p">][</span><span class="n">idx</span><span class="p">])</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="n">targets_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_FILES</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">][</span><span class="s2">&quot;targets&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">h5py</span><span class="o">.</span><span class="n">File</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="n">targets_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">targets_data</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">targets_data</span><span class="p">[</span><span class="s2">&quot;y&quot;</span><span class="p">][</span><span class="n">idx</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">])</span>  <span class="c1"># shape is [num_images, 1, 1, 1]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="n">images_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_FILES</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">][</span><span class="s2">&quot;images&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">targets_file</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_FILES</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">][</span><span class="s2">&quot;targets&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="o">.</span><span class="n">joinpath</span><span class="p">(</span><span class="n">h5_file</span><span class="p">)</span><span class="o">.</span><span class="n">exists</span><span class="p">()</span> <span class="k">for</span> <span class="n">h5_file</span> <span class="ow">in</span> <span class="p">(</span><span class="n">images_file</span><span class="p">,</span> <span class="n">targets_file</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="k">for</span> <span class="n">file_name</span><span class="p">,</span> <span class="n">file_id</span><span class="p">,</span> <span class="n">md5</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_FILES</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">]</span><span class="o">.</span><span class="n">values</span><span class="p">():</span>
+            <span class="n">archive_name</span> <span class="o">=</span> <span class="n">file_name</span> <span class="o">+</span> <span class="s2">&quot;.gz&quot;</span>
+            <span class="n">download_file_from_google_drive</span><span class="p">(</span><span class="n">file_id</span><span class="p">,</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span> <span class="n">filename</span><span class="o">=</span><span class="n">archive_name</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span>
+            <span class="n">_decompress</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="n">archive_name</span><span class="p">))</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/phototour.html b/0.16/_modules/torchvision/datasets/phototour.html
new file mode 100644
index 00000000000..470e6b6dc8a
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/phototour.html
@@ -0,0 +1,897 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.phototour &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.phototour</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.phototour</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_url</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="PhotoTour"><a class="viewcode-back" href="../../../generated/torchvision.datasets.PhotoTour.html#torchvision.datasets.PhotoTour">[docs]</a><span class="k">class</span> <span class="nc">PhotoTour</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Multi-view Stereo Correspondence &lt;http://matthewalunbrown.com/patchdata/patchdata.html&gt;`_ Dataset.</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        We only provide the newer version of the dataset, since the authors state that it</span>
+
+<span class="sd">            is more suitable for training descriptors based on difference of Gaussian, or Harris corners, as the</span>
+<span class="sd">            patches are centred on real interest point detections, rather than being projections of 3D points as is the</span>
+<span class="sd">            case in the old dataset.</span>
+
+<span class="sd">        The original dataset is available under http://phototour.cs.washington.edu/patches/default.htm.</span>
+
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images are.</span>
+<span class="sd">        name (string): Name of the dataset to load.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">urls</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;notredame_harris&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;http://matthewalunbrown.com/patchdata/notredame_harris.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;notredame_harris.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;69f8c90f78e171349abdf0307afefe4d&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;yosemite_harris&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;http://matthewalunbrown.com/patchdata/yosemite_harris.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;yosemite_harris.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;a73253d1c6fbd3ba2613c45065c00d46&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;liberty_harris&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;http://matthewalunbrown.com/patchdata/liberty_harris.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;liberty_harris.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;c731fcfb3abb4091110d0ae8c7ba182c&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;notredame&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;http://icvl.ee.ic.ac.uk/vbalnt/notredame.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;notredame.zip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;509eda8535847b8c0a90bbb210c83484&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;yosemite&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;http://icvl.ee.ic.ac.uk/vbalnt/yosemite.zip&quot;</span><span class="p">,</span> <span class="s2">&quot;yosemite.zip&quot;</span><span class="p">,</span> <span class="s2">&quot;533b2e8eb7ede31be40abc317b2fd4f0&quot;</span><span class="p">],</span>
+        <span class="s2">&quot;liberty&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;http://icvl.ee.ic.ac.uk/vbalnt/liberty.zip&quot;</span><span class="p">,</span> <span class="s2">&quot;liberty.zip&quot;</span><span class="p">,</span> <span class="s2">&quot;fdd9152f138ea5ef2091746689176414&quot;</span><span class="p">],</span>
+    <span class="p">}</span>
+    <span class="n">means</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;notredame&quot;</span><span class="p">:</span> <span class="mf">0.4854</span><span class="p">,</span>
+        <span class="s2">&quot;yosemite&quot;</span><span class="p">:</span> <span class="mf">0.4844</span><span class="p">,</span>
+        <span class="s2">&quot;liberty&quot;</span><span class="p">:</span> <span class="mf">0.4437</span><span class="p">,</span>
+        <span class="s2">&quot;notredame_harris&quot;</span><span class="p">:</span> <span class="mf">0.4854</span><span class="p">,</span>
+        <span class="s2">&quot;yosemite_harris&quot;</span><span class="p">:</span> <span class="mf">0.4844</span><span class="p">,</span>
+        <span class="s2">&quot;liberty_harris&quot;</span><span class="p">:</span> <span class="mf">0.4437</span><span class="p">,</span>
+    <span class="p">}</span>
+    <span class="n">stds</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;notredame&quot;</span><span class="p">:</span> <span class="mf">0.1864</span><span class="p">,</span>
+        <span class="s2">&quot;yosemite&quot;</span><span class="p">:</span> <span class="mf">0.1818</span><span class="p">,</span>
+        <span class="s2">&quot;liberty&quot;</span><span class="p">:</span> <span class="mf">0.2019</span><span class="p">,</span>
+        <span class="s2">&quot;notredame_harris&quot;</span><span class="p">:</span> <span class="mf">0.1864</span><span class="p">,</span>
+        <span class="s2">&quot;yosemite_harris&quot;</span><span class="p">:</span> <span class="mf">0.1818</span><span class="p">,</span>
+        <span class="s2">&quot;liberty_harris&quot;</span><span class="p">:</span> <span class="mf">0.2019</span><span class="p">,</span>
+    <span class="p">}</span>
+    <span class="n">lens</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;notredame&quot;</span><span class="p">:</span> <span class="mi">468159</span><span class="p">,</span>
+        <span class="s2">&quot;yosemite&quot;</span><span class="p">:</span> <span class="mi">633587</span><span class="p">,</span>
+        <span class="s2">&quot;liberty&quot;</span><span class="p">:</span> <span class="mi">450092</span><span class="p">,</span>
+        <span class="s2">&quot;liberty_harris&quot;</span><span class="p">:</span> <span class="mi">379587</span><span class="p">,</span>
+        <span class="s2">&quot;yosemite_harris&quot;</span><span class="p">:</span> <span class="mi">450912</span><span class="p">,</span>
+        <span class="s2">&quot;notredame_harris&quot;</span><span class="p">:</span> <span class="mi">325295</span><span class="p">,</span>
+    <span class="p">}</span>
+    <span class="n">image_ext</span> <span class="o">=</span> <span class="s2">&quot;bmp&quot;</span>
+    <span class="n">info_file</span> <span class="o">=</span> <span class="s2">&quot;info.txt&quot;</span>
+    <span class="n">matches_files</span> <span class="o">=</span> <span class="s2">&quot;m50_100000_100000_0.txt&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">name</span> <span class="o">=</span> <span class="n">name</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_down</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">.zip&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">.pt&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="o">=</span> <span class="n">train</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mean</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">means</span><span class="p">[</span><span class="n">name</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">std</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stds</span><span class="p">[</span><span class="n">name</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_datafile_exists</span><span class="p">():</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">cache</span><span class="p">()</span>
+
+        <span class="c1"># load the serialized data</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">matches</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_file</span><span class="p">)</span>
+
+<div class="viewcode-block" id="PhotoTour.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.PhotoTour.html#torchvision.datasets.PhotoTour.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (data1, data2, matches)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">data</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">data</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">data</span>
+        <span class="n">m</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">matches</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">data1</span><span class="p">,</span> <span class="n">data2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">m</span><span class="p">[</span><span class="mi">0</span><span class="p">]],</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">m</span><span class="p">[</span><span class="mi">1</span><span class="p">]]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">data1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">data1</span><span class="p">)</span>
+            <span class="n">data2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">data2</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">data1</span><span class="p">,</span> <span class="n">data2</span><span class="p">,</span> <span class="n">m</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">matches</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_datafile_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_file</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_downloaded</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_dir</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_datafile_exists</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;# Found cached data </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">data_file</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_downloaded</span><span class="p">():</span>
+            <span class="c1"># download files</span>
+            <span class="n">url</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">urls</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">name</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">filename</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">urls</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">name</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span>
+            <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">urls</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">name</span><span class="p">][</span><span class="mi">2</span><span class="p">]</span>
+            <span class="n">fpath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span>
+
+            <span class="n">download_url</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;# Extracting data </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">data_down</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+
+            <span class="kn">import</span> <span class="nn">zipfile</span>
+
+            <span class="k">with</span> <span class="n">zipfile</span><span class="o">.</span><span class="n">ZipFile</span><span class="p">(</span><span class="n">fpath</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">z</span><span class="p">:</span>
+                <span class="n">z</span><span class="o">.</span><span class="n">extractall</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_dir</span><span class="p">)</span>
+
+            <span class="n">os</span><span class="o">.</span><span class="n">unlink</span><span class="p">(</span><span class="n">fpath</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">cache</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># process and save as torch files</span>
+        <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;# Caching data </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">data_file</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="n">dataset</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">read_image_file</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_dir</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">image_ext</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">lens</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">name</span><span class="p">]),</span>
+            <span class="n">read_info_file</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_dir</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">info_file</span><span class="p">),</span>
+            <span class="n">read_matches_files</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_dir</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">matches_files</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_file</span><span class="p">,</span> <span class="s2">&quot;wb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">f</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">split</span> <span class="o">=</span> <span class="s2">&quot;Train&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="ow">is</span> <span class="kc">True</span> <span class="k">else</span> <span class="s2">&quot;Test&quot;</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;Split: </span><span class="si">{</span><span class="n">split</span><span class="si">}</span><span class="s2">&quot;</span></div>
+
+
+<span class="k">def</span> <span class="nf">read_image_file</span><span class="p">(</span><span class="n">data_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">image_ext</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">n</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Return a Tensor containing the patches&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="nf">PIL2array</span><span class="p">(</span><span class="n">_img</span><span class="p">:</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Convert PIL image type to numpy 2D array&quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">_img</span><span class="o">.</span><span class="n">getdata</span><span class="p">(),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">find_files</span><span class="p">(</span><span class="n">_data_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">_image_ext</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return a list with the file names of the images containing the patches&quot;&quot;&quot;</span>
+        <span class="n">files</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="c1"># find those files with the specified extension</span>
+        <span class="k">for</span> <span class="n">file_dir</span> <span class="ow">in</span> <span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">_data_dir</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">file_dir</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="n">_image_ext</span><span class="p">):</span>
+                <span class="n">files</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">_data_dir</span><span class="p">,</span> <span class="n">file_dir</span><span class="p">))</span>
+        <span class="k">return</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">files</span><span class="p">)</span>  <span class="c1"># sort files in ascend order to keep relations</span>
+
+    <span class="n">patches</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="n">list_files</span> <span class="o">=</span> <span class="n">find_files</span><span class="p">(</span><span class="n">data_dir</span><span class="p">,</span> <span class="n">image_ext</span><span class="p">)</span>
+
+    <span class="k">for</span> <span class="n">fpath</span> <span class="ow">in</span> <span class="n">list_files</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">fpath</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">y</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">img</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="mi">64</span><span class="p">):</span>
+            <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">img</span><span class="o">.</span><span class="n">width</span><span class="p">,</span> <span class="mi">64</span><span class="p">):</span>
+                <span class="n">patch</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">crop</span><span class="p">((</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">x</span> <span class="o">+</span> <span class="mi">64</span><span class="p">,</span> <span class="n">y</span> <span class="o">+</span> <span class="mi">64</span><span class="p">))</span>
+                <span class="n">patches</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">PIL2array</span><span class="p">(</span><span class="n">patch</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">ByteTensor</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">patches</span><span class="p">[:</span><span class="n">n</span><span class="p">]))</span>
+
+
+<span class="k">def</span> <span class="nf">read_info_file</span><span class="p">(</span><span class="n">data_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">info_file</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Return a Tensor containing the list of labels</span>
+<span class="sd">    Read the file and keep only the ID of the 3D point.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">data_dir</span><span class="p">,</span> <span class="n">info_file</span><span class="p">))</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="n">labels</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">()[</span><span class="mi">0</span><span class="p">])</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">]</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">read_matches_files</span><span class="p">(</span><span class="n">data_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">matches_file</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Return a Tensor containing the ground truth matches</span>
+<span class="sd">    Read the file and keep only 3D point ID.</span>
+<span class="sd">    Matches are represented with a 1, non matches with a 0.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">matches</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">data_dir</span><span class="p">,</span> <span class="n">matches_file</span><span class="p">))</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">line_split</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">()</span>
+            <span class="n">matches</span><span class="o">.</span><span class="n">append</span><span class="p">([</span><span class="nb">int</span><span class="p">(</span><span class="n">line_split</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="nb">int</span><span class="p">(</span><span class="n">line_split</span><span class="p">[</span><span class="mi">3</span><span class="p">]),</span> <span class="nb">int</span><span class="p">(</span><span class="n">line_split</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="n">line_split</span><span class="p">[</span><span class="mi">4</span><span class="p">])])</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">(</span><span class="n">matches</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/places365.html b/0.16/_modules/torchvision/datasets/places365.html
new file mode 100644
index 00000000000..0818d529945
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/places365.html
@@ -0,0 +1,839 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.places365 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.places365</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.places365</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">os</span> <span class="kn">import</span> <span class="n">path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+<span class="kn">from</span> <span class="nn">urllib.parse</span> <span class="kn">import</span> <span class="n">urljoin</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">default_loader</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="Places365"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Places365.html#torchvision.datasets.Places365">[docs]</a><span class="k">class</span> <span class="nc">Places365</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;`Places365 &lt;http://places2.csail.mit.edu/index.html&gt;`_ classification dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the Places365 dataset.</span>
+<span class="sd">        split (string, optional): The dataset split. Can be one of ``train-standard`` (default), ``train-challenge``,</span>
+<span class="sd">            ``val``.</span>
+<span class="sd">        small (bool, optional): If ``True``, uses the small images, i.e. resized to 256 x 256 pixels, instead of the</span>
+<span class="sd">            high resolution ones.</span>
+<span class="sd">        download (bool, optional): If ``True``, downloads the dataset components and places them in ``root``. Already</span>
+<span class="sd">            downloaded archives are not downloaded again.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        loader (callable, optional): A function to load an image given its path.</span>
+
+<span class="sd">     Attributes:</span>
+<span class="sd">        classes (list): List of the class names.</span>
+<span class="sd">        class_to_idx (dict): Dict with items (class_name, class_index).</span>
+<span class="sd">        imgs (list): List of (image path, class_index) tuples</span>
+<span class="sd">        targets (list): The class_index value for each image in the dataset</span>
+
+<span class="sd">    Raises:</span>
+<span class="sd">        RuntimeError: If ``download is False`` and the meta files, i.e. the devkit, are not present or corrupted.</span>
+<span class="sd">        RuntimeError: If ``download is True`` and the image archive is already extracted.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">_SPLITS</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;train-standard&quot;</span><span class="p">,</span> <span class="s2">&quot;train-challenge&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">)</span>
+    <span class="n">_BASE_URL</span> <span class="o">=</span> <span class="s2">&quot;http://data.csail.mit.edu/places/places365/&quot;</span>
+    <span class="c1"># {variant: (archive, md5)}</span>
+    <span class="n">_DEVKIT_META</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;standard&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;filelist_places365-standard.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;35a0585fee1fa656440f3ab298f8479c&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;challenge&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;filelist_places365-challenge.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;70a8307e459c3de41690a7c76c931734&quot;</span><span class="p">),</span>
+    <span class="p">}</span>
+    <span class="c1"># (file, md5)</span>
+    <span class="n">_CATEGORIES_META</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;categories_places365.txt&quot;</span><span class="p">,</span> <span class="s2">&quot;06c963b85866bd0649f97cb43dd16673&quot;</span><span class="p">)</span>
+    <span class="c1"># {split: (file, md5)}</span>
+    <span class="n">_FILE_LIST_META</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;train-standard&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;places365_train_standard.txt&quot;</span><span class="p">,</span> <span class="s2">&quot;30f37515461640559006b8329efbed1a&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;train-challenge&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;places365_train_challenge.txt&quot;</span><span class="p">,</span> <span class="s2">&quot;b2931dc997b8c33c27e7329c073a6b57&quot;</span><span class="p">),</span>
+        <span class="s2">&quot;val&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;places365_val.txt&quot;</span><span class="p">,</span> <span class="s2">&quot;e9f2fd57bfd9d07630173f4e8708e4b1&quot;</span><span class="p">),</span>
+    <span class="p">}</span>
+    <span class="c1"># {(split, small): (file, md5)}</span>
+    <span class="n">_IMAGES_META</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="p">(</span><span class="s2">&quot;train-standard&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span> <span class="p">(</span><span class="s2">&quot;train_large_places365standard.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;67e186b496a84c929568076ed01a8aa1&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;train-challenge&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span> <span class="p">(</span><span class="s2">&quot;train_large_places365challenge.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;605f18e68e510c82b958664ea134545f&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span> <span class="p">(</span><span class="s2">&quot;val_large.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;9b71c4993ad89d2d8bcbdc4aef38042f&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;train-standard&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">):</span> <span class="p">(</span><span class="s2">&quot;train_256_places365standard.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;53ca1c756c3d1e7809517cc47c5561c5&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;train-challenge&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">):</span> <span class="p">(</span><span class="s2">&quot;train_256_places365challenge.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;741915038a5e3471ec7332404dfb64ef&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">):</span> <span class="p">(</span><span class="s2">&quot;val_256.tar&quot;</span><span class="p">,</span> <span class="s2">&quot;e27b17d8d44f4af9a78502beb927f808&quot;</span><span class="p">),</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train-standard&quot;</span><span class="p">,</span>
+        <span class="n">small</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">loader</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="n">default_loader</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_verify_split</span><span class="p">(</span><span class="n">split</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">small</span> <span class="o">=</span> <span class="n">small</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">loader</span> <span class="o">=</span> <span class="n">loader</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">load_categories</span><span class="p">(</span><span class="n">download</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">imgs</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">load_file_list</span><span class="p">(</span><span class="n">download</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download_images</span><span class="p">()</span>
+
+<div class="viewcode-block" id="Places365.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.Places365.html#torchvision.datasets.Places365.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">file</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">imgs</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">loader</span><span class="p">(</span><span class="n">file</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">imgs</span><span class="p">)</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">variant</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;challenge&quot;</span> <span class="k">if</span> <span class="s2">&quot;challenge&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="k">else</span> <span class="s2">&quot;standard&quot;</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">images_dir</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="s2">&quot;256&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">small</span> <span class="k">else</span> <span class="s2">&quot;large&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">):</span>
+            <span class="nb">dir</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;data_</span><span class="si">{</span><span class="n">size</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">variant</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="nb">dir</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="n">size</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="k">return</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="nb">dir</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">load_categories</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+        <span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">line</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+            <span class="bp">cls</span><span class="p">,</span> <span class="n">idx</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">()</span>
+            <span class="k">return</span> <span class="bp">cls</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">idx</span><span class="p">)</span>
+
+        <span class="n">file</span><span class="p">,</span> <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_CATEGORIES_META</span>
+        <span class="n">file</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">(</span><span class="n">file</span><span class="p">,</span> <span class="n">md5</span><span class="p">,</span> <span class="n">download</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download_devkit</span><span class="p">()</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file</span><span class="p">)</span> <span class="k">as</span> <span class="n">fh</span><span class="p">:</span>
+            <span class="n">class_to_idx</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">process</span><span class="p">(</span><span class="n">line</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">fh</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">class_to_idx</span><span class="o">.</span><span class="n">keys</span><span class="p">()),</span> <span class="n">class_to_idx</span>
+
+    <span class="k">def</span> <span class="nf">load_file_list</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+        <span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">line</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;/&quot;</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+            <span class="n">image</span><span class="p">,</span> <span class="n">idx</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">()</span>
+            <span class="k">return</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="n">sep</span><span class="p">)</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="n">sep</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">sep</span><span class="p">)),</span> <span class="nb">int</span><span class="p">(</span><span class="n">idx</span><span class="p">)</span>
+
+        <span class="n">file</span><span class="p">,</span> <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_FILE_LIST_META</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">]</span>
+        <span class="n">file</span> <span class="o">=</span> <span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">(</span><span class="n">file</span><span class="p">,</span> <span class="n">md5</span><span class="p">,</span> <span class="n">download</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download_devkit</span><span class="p">()</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file</span><span class="p">)</span> <span class="k">as</span> <span class="n">fh</span><span class="p">:</span>
+            <span class="n">images</span> <span class="o">=</span> <span class="p">[</span><span class="n">process</span><span class="p">(</span><span class="n">line</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">fh</span><span class="p">]</span>
+
+        <span class="n">_</span><span class="p">,</span> <span class="n">targets</span> <span class="o">=</span> <span class="nb">zip</span><span class="p">(</span><span class="o">*</span><span class="n">images</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">images</span><span class="p">,</span> <span class="nb">list</span><span class="p">(</span><span class="n">targets</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download_devkit</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">file</span><span class="p">,</span> <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_DEVKIT_META</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">variant</span><span class="p">]</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">urljoin</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_BASE_URL</span><span class="p">,</span> <span class="n">file</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download_images</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The directory </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="si">}</span><span class="s2"> already exists. If you want to re-download or re-extract the images, &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;delete the directory.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">file</span><span class="p">,</span> <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_IMAGES_META</span><span class="p">[(</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">small</span><span class="p">)]</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="n">urljoin</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_BASE_URL</span><span class="p">,</span> <span class="n">file</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">):</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">rename</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="o">.</span><span class="n">rsplit</span><span class="p">(</span><span class="s2">&quot;_&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">images_dir</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">((</span><span class="s2">&quot;Split: </span><span class="si">{split}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;Small: </span><span class="si">{small}</span><span class="s2">&quot;</span><span class="p">))</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_verify_split</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">split</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_SPLITS</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">file</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">md5</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="n">integrity</span> <span class="o">=</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">file</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">md5</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">integrity</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">download</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The file </span><span class="si">{</span><span class="n">file</span><span class="si">}</span><span class="s2"> does not exist or is corrupted. You can set download=True to download it.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="n">integrity</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/rendered_sst2.html b/0.16/_modules/torchvision/datasets/rendered_sst2.html
new file mode 100644
index 00000000000..2afb77f7d44
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/rendered_sst2.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.rendered_sst2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.rendered_sst2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.rendered_sst2</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">make_dataset</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="RenderedSST2"><a class="viewcode-back" href="../../../generated/torchvision.datasets.RenderedSST2.html#torchvision.datasets.RenderedSST2">[docs]</a><span class="k">class</span> <span class="nc">RenderedSST2</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`The Rendered SST2 Dataset &lt;https://github.com/openai/CLIP/blob/main/data/rendered-sst2.md&gt;`_.</span>
+
+<span class="sd">    Rendered SST2 is an image classification dataset used to evaluate the models capability on optical</span>
+<span class="sd">    character recognition. This dataset was generated by rendering sentences in the Standford Sentiment</span>
+<span class="sd">    Treebank v2 dataset.</span>
+
+<span class="sd">    This dataset contains two classes (positive and negative) and is divided in three splits: a  train</span>
+<span class="sd">    split containing 6920 images (3610 positive and 3310 negative), a validation split containing 872 images</span>
+<span class="sd">    (444 positive and 428 negative), and a test split containing 1821 images (909 positive and 912 negative).</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default), `&quot;val&quot;` and ``&quot;test&quot;``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again. Default is False.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_URL</span> <span class="o">=</span> <span class="s2">&quot;https://openaipublic.azureedge.net/clip/data/rendered-sst2.tgz&quot;</span>
+    <span class="n">_MD5</span> <span class="o">=</span> <span class="s2">&quot;2384d08e9dcfa4bd55b324e610496ee5&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split_to_folder</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">:</span> <span class="s2">&quot;valid&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="s2">&quot;test&quot;</span><span class="p">}</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;rendered-sst2&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;negative&quot;</span><span class="p">,</span> <span class="s2">&quot;positive&quot;</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;negative&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span> <span class="s2">&quot;positive&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">}</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span> <span class="o">=</span> <span class="n">make_dataset</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_to_folder</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">]),</span> <span class="n">extensions</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;png&quot;</span><span class="p">,))</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">)</span>
+
+<div class="viewcode-block" id="RenderedSST2.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.RenderedSST2.html#torchvision.datasets.RenderedSST2.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image_file</span><span class="p">,</span> <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_file</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">label</span></div>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;split=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="si">}</span><span class="s2">&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">class_label</span> <span class="ow">in</span> <span class="nb">set</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">):</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_to_folder</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_split</span><span class="p">]</span> <span class="o">/</span> <span class="n">class_label</span><span class="p">)</span><span class="o">.</span><span class="n">is_dir</span><span class="p">():</span>
+                <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_URL</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_MD5</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/sbd.html b/0.16/_modules/torchvision/datasets/sbd.html
new file mode 100644
index 00000000000..98721d5627e
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/sbd.html
@@ -0,0 +1,792 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.sbd &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.sbd</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.sbd</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">shutil</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">download_url</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="SBDataset"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SBDataset.html#torchvision.datasets.SBDataset">[docs]</a><span class="k">class</span> <span class="nc">SBDataset</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Semantic Boundaries Dataset &lt;http://home.bharathh.info/pubs/codes/SBD/download.html&gt;`_</span>
+
+<span class="sd">    The SBD currently contains annotations from 11355 images taken from the PASCAL VOC 2011 dataset.</span>
+
+<span class="sd">    .. note ::</span>
+
+<span class="sd">        Please note that the train and val splits included with this dataset are different from</span>
+<span class="sd">        the splits in the PASCAL VOC dataset. In particular some &quot;train&quot; images might be part of</span>
+<span class="sd">        VOC2012 val.</span>
+<span class="sd">        If you are interested in testing on VOC 2012 val, then use `image_set=&#39;train_noval&#39;`,</span>
+<span class="sd">        which excludes all val images.</span>
+
+<span class="sd">    .. warning::</span>
+
+<span class="sd">        This class needs `scipy &lt;https://docs.scipy.org/doc/&gt;`_ to load target files from `.mat` format.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the Semantic Boundaries Dataset</span>
+<span class="sd">        image_set (string, optional): Select the image_set to use, ``train``, ``val`` or ``train_noval``.</span>
+<span class="sd">            Image set ``train_noval`` excludes VOC 2012 val images.</span>
+<span class="sd">        mode (string, optional): Select target type. Possible values &#39;boundaries&#39; or &#39;segmentation&#39;.</span>
+<span class="sd">            In case of &#39;boundaries&#39;, the target is an array of shape `[num_classes, H, W]`,</span>
+<span class="sd">            where `num_classes=20`.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes input sample and its target as entry</span>
+<span class="sd">            and returns a transformed version. Input sample is PIL image and target is a numpy array</span>
+<span class="sd">            if `mode=&#39;boundaries&#39;` or PIL image if `mode=&#39;segmentation&#39;`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">url</span> <span class="o">=</span> <span class="s2">&quot;https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/semantic_contours/benchmark.tgz&quot;</span>
+    <span class="n">md5</span> <span class="o">=</span> <span class="s2">&quot;82b4d87ceb2ed10f6038a1cba92111cb&quot;</span>
+    <span class="n">filename</span> <span class="o">=</span> <span class="s2">&quot;benchmark.tgz&quot;</span>
+
+    <span class="n">voc_train_url</span> <span class="o">=</span> <span class="s2">&quot;http://home.bharathh.info/pubs/codes/SBD/train_noval.txt&quot;</span>
+    <span class="n">voc_split_filename</span> <span class="o">=</span> <span class="s2">&quot;train_noval.txt&quot;</span>
+    <span class="n">voc_split_md5</span> <span class="o">=</span> <span class="s2">&quot;79bff800c5f0b1ec6b21080a3c066722&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">image_set</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;boundaries&quot;</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="k">try</span><span class="p">:</span>
+            <span class="kn">from</span> <span class="nn">scipy.io</span> <span class="kn">import</span> <span class="n">loadmat</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">_loadmat</span> <span class="o">=</span> <span class="n">loadmat</span>
+        <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Scipy is not found. This dataset needs to have scipy installed: pip install scipy&quot;</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_set</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">image_set</span><span class="p">,</span> <span class="s2">&quot;image_set&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;train_noval&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">mode</span><span class="p">,</span> <span class="s2">&quot;mode&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;segmentation&quot;</span><span class="p">,</span> <span class="s2">&quot;boundaries&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="mi">20</span>
+
+        <span class="n">sbd_root</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span>
+        <span class="n">image_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">sbd_root</span><span class="p">,</span> <span class="s2">&quot;img&quot;</span><span class="p">)</span>
+        <span class="n">mask_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">sbd_root</span><span class="p">,</span> <span class="s2">&quot;cls&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">md5</span><span class="p">)</span>
+            <span class="n">extracted_ds_root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;benchmark_RELEASE&quot;</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;cls&quot;</span><span class="p">,</span> <span class="s2">&quot;img&quot;</span><span class="p">,</span> <span class="s2">&quot;inst&quot;</span><span class="p">,</span> <span class="s2">&quot;train.txt&quot;</span><span class="p">,</span> <span class="s2">&quot;val.txt&quot;</span><span class="p">]:</span>
+                <span class="n">old_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">extracted_ds_root</span><span class="p">,</span> <span class="n">f</span><span class="p">)</span>
+                <span class="n">shutil</span><span class="o">.</span><span class="n">move</span><span class="p">(</span><span class="n">old_path</span><span class="p">,</span> <span class="n">sbd_root</span><span class="p">)</span>
+            <span class="n">download_url</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">voc_train_url</span><span class="p">,</span> <span class="n">sbd_root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">voc_split_filename</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">voc_split_md5</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">sbd_root</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="n">split_f</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">sbd_root</span><span class="p">,</span> <span class="n">image_set</span><span class="o">.</span><span class="n">rstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span> <span class="o">+</span> <span class="s2">&quot;.txt&quot;</span><span class="p">)</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">split_f</span><span class="p">))</span> <span class="k">as</span> <span class="n">fh</span><span class="p">:</span>
+            <span class="n">file_names</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">fh</span><span class="o">.</span><span class="n">readlines</span><span class="p">()]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">images</span> <span class="o">=</span> <span class="p">[</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">image_dir</span><span class="p">,</span> <span class="n">x</span> <span class="o">+</span> <span class="s2">&quot;.jpg&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">file_names</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">masks</span> <span class="o">=</span> <span class="p">[</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">mask_dir</span><span class="p">,</span> <span class="n">x</span> <span class="o">+</span> <span class="s2">&quot;.mat&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">file_names</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_get_target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_segmentation_target</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;segmentation&quot;</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_boundaries_target</span>
+
+    <span class="k">def</span> <span class="nf">_get_segmentation_target</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">filepath</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+        <span class="n">mat</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_loadmat</span><span class="p">(</span><span class="n">filepath</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">mat</span><span class="p">[</span><span class="s2">&quot;GTcls&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">][</span><span class="s2">&quot;Segmentation&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+
+    <span class="k">def</span> <span class="nf">_get_boundaries_target</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">filepath</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
+        <span class="n">mat</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_loadmat</span><span class="p">(</span><span class="n">filepath</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">(</span>
+            <span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">expand_dims</span><span class="p">(</span><span class="n">mat</span><span class="p">[</span><span class="s2">&quot;GTcls&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">][</span><span class="s2">&quot;Boundaries&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">toarray</span><span class="p">(),</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)],</span>
+            <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="p">)</span>
+
+<div class="viewcode-block" id="SBDataset.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SBDataset.html#torchvision.datasets.SBDataset.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">[</span><span class="n">index</span><span class="p">])</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_target</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">masks</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">lines</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Image set: </span><span class="si">{image_set}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;Mode: </span><span class="si">{mode}</span><span class="s2">&quot;</span><span class="p">]</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">lines</span><span class="p">)</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/sbu.html b/0.16/_modules/torchvision/datasets/sbu.html
new file mode 100644
index 00000000000..5e9df7bd2bb
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/sbu.html
@@ -0,0 +1,778 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.sbu &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.sbu</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.sbu</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">download_url</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="SBU"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SBU.html#torchvision.datasets.SBU">[docs]</a><span class="k">class</span> <span class="nc">SBU</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`SBU Captioned Photo &lt;http://www.cs.virginia.edu/~vicente/sbucaptions/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where tarball</span>
+<span class="sd">            ``SBUCaptionedPhotoDataset.tar.gz`` exists.</span>
+<span class="sd">        transform (callable, optional): A function/transform that takes in a PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">url</span> <span class="o">=</span> <span class="s2">&quot;https://www.cs.rice.edu/~vo9/sbucaptions/SBUCaptionedPhotoDataset.tar.gz&quot;</span>
+    <span class="n">filename</span> <span class="o">=</span> <span class="s2">&quot;SBUCaptionedPhotoDataset.tar.gz&quot;</span>
+    <span class="n">md5_checksum</span> <span class="o">=</span> <span class="s2">&quot;9aec147b3488753cf758b4d493422285&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="c1"># Read the caption for each photo</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">photos</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">captions</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="n">file1</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">,</span> <span class="s2">&quot;SBU_captioned_photo_dataset_urls.txt&quot;</span><span class="p">)</span>
+        <span class="n">file2</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">,</span> <span class="s2">&quot;SBU_captioned_photo_dataset_captions.txt&quot;</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">line1</span><span class="p">,</span> <span class="n">line2</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="nb">open</span><span class="p">(</span><span class="n">file1</span><span class="p">),</span> <span class="nb">open</span><span class="p">(</span><span class="n">file2</span><span class="p">)):</span>
+            <span class="n">url</span> <span class="o">=</span> <span class="n">line1</span><span class="o">.</span><span class="n">rstrip</span><span class="p">()</span>
+            <span class="n">photo</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">basename</span><span class="p">(</span><span class="n">url</span><span class="p">)</span>
+            <span class="n">filename</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">,</span> <span class="n">photo</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">filename</span><span class="p">):</span>
+                <span class="n">caption</span> <span class="o">=</span> <span class="n">line2</span><span class="o">.</span><span class="n">rstrip</span><span class="p">()</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">photos</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">photo</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">captions</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">caption</span><span class="p">)</span>
+
+<div class="viewcode-block" id="SBU.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SBU.html#torchvision.datasets.SBU.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is a caption for the photo.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">filename</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">photos</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">filename</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">captions</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;The number of photos in the dataset.&quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">photos</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Check the md5 checksum of the downloaded tarball.&quot;&quot;&quot;</span>
+        <span class="n">root</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span>
+        <span class="n">fpath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">fpath</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">md5_checksum</span><span class="p">):</span>
+            <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Download and extract the tarball, and download each individual photo.&quot;&quot;&quot;</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">md5_checksum</span><span class="p">)</span>
+
+        <span class="c1"># Download individual photos</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">,</span> <span class="s2">&quot;SBU_captioned_photo_dataset_urls.txt&quot;</span><span class="p">))</span> <span class="k">as</span> <span class="n">fh</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">fh</span><span class="p">:</span>
+                <span class="n">url</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">rstrip</span><span class="p">()</span>
+                <span class="k">try</span><span class="p">:</span>
+                    <span class="n">download_url</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">))</span>
+                <span class="k">except</span> <span class="ne">OSError</span><span class="p">:</span>
+                    <span class="c1"># The images point to public images on Flickr.</span>
+                    <span class="c1"># Note: Images might be removed by users at anytime.</span>
+                    <span class="k">pass</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/semeion.html b/0.16/_modules/torchvision/datasets/semeion.html
new file mode 100644
index 00000000000..a81802673b2
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/semeion.html
@@ -0,0 +1,760 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.semeion &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.semeion</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.semeion</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_url</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="SEMEION"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SEMEION.html#torchvision.datasets.SEMEION">[docs]</a><span class="k">class</span> <span class="nc">SEMEION</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;`SEMEION &lt;http://archive.ics.uci.edu/ml/datasets/semeion+handwritten+digit&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``semeion.py`` exists.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">url</span> <span class="o">=</span> <span class="s2">&quot;http://archive.ics.uci.edu/ml/machine-learning-databases/semeion/semeion.data&quot;</span>
+    <span class="n">filename</span> <span class="o">=</span> <span class="s2">&quot;semeion.data&quot;</span>
+    <span class="n">md5_checksum</span> <span class="o">=</span> <span class="s2">&quot;cb545d371d2ce14ec121470795a77432&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="n">fp</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">)</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">loadtxt</span><span class="p">(</span><span class="n">fp</span><span class="p">)</span>
+        <span class="c1"># convert value to 8 bit unsigned integer</span>
+        <span class="c1"># color (white #255) the pixels</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="p">(</span><span class="n">data</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">256</span><span class="p">]</span> <span class="o">*</span> <span class="mi">255</span><span class="p">)</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="s2">&quot;uint8&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">16</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">nonzero</span><span class="p">(</span><span class="n">data</span><span class="p">[:,</span> <span class="mi">256</span><span class="p">:])[</span><span class="mi">1</span><span class="p">]</span>
+
+<div class="viewcode-block" id="SEMEION.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SEMEION.html#torchvision.datasets.SEMEION.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="nb">int</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+
+        <span class="c1"># doing this so that it is consistent with all other datasets</span>
+        <span class="c1"># to return a PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;L&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="n">root</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span>
+        <span class="n">fpath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">fpath</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">md5_checksum</span><span class="p">):</span>
+            <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="n">root</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span>
+        <span class="n">download_url</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">md5_checksum</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/stanford_cars.html b/0.16/_modules/torchvision/datasets/stanford_cars.html
new file mode 100644
index 00000000000..73cf9ef13a5
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/stanford_cars.html
@@ -0,0 +1,790 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.stanford_cars &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.stanford_cars</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.stanford_cars</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">pathlib</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">download_url</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="StanfordCars"><a class="viewcode-back" href="../../../generated/torchvision.datasets.StanfordCars.html#torchvision.datasets.StanfordCars">[docs]</a><span class="k">class</span> <span class="nc">StanfordCars</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Stanford Cars &lt;https://ai.stanford.edu/~jkrause/cars/car_dataset.html&gt;`_ Dataset</span>
+
+<span class="sd">    The Cars dataset contains 16,185 images of 196 classes of cars. The data is</span>
+<span class="sd">    split into 8,144 training images and 8,041 testing images, where each class</span>
+<span class="sd">    has been split roughly in a 50-50 split</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        This class needs `scipy &lt;https://docs.scipy.org/doc/&gt;`_ to load target files from `.mat` format.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset</span>
+<span class="sd">        split (string, optional): The dataset split, supports ``&quot;train&quot;`` (default) or ``&quot;test&quot;``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If True, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="k">try</span><span class="p">:</span>
+            <span class="kn">import</span> <span class="nn">scipy.io</span> <span class="k">as</span> <span class="nn">sio</span>
+        <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Scipy is not found. This dataset needs to have scipy installed: pip install scipy&quot;</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">=</span> <span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span><span class="p">(</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;stanford_cars&quot;</span>
+        <span class="n">devkit</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;devkit&quot;</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_annotations_mat_path</span> <span class="o">=</span> <span class="n">devkit</span> <span class="o">/</span> <span class="s2">&quot;cars_train_annos.mat&quot;</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_images_base_path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;cars_train&quot;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_annotations_mat_path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;cars_test_annos_withlabels.mat&quot;</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_images_base_path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;cars_test&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="p">(</span>
+                <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_images_base_path</span> <span class="o">/</span> <span class="n">annotation</span><span class="p">[</span><span class="s2">&quot;fname&quot;</span><span class="p">]),</span>
+                <span class="n">annotation</span><span class="p">[</span><span class="s2">&quot;class&quot;</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span>  <span class="c1"># Original target mapping  starts from 1, hence -1</span>
+            <span class="p">)</span>
+            <span class="k">for</span> <span class="n">annotation</span> <span class="ow">in</span> <span class="n">sio</span><span class="o">.</span><span class="n">loadmat</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_annotations_mat_path</span><span class="p">,</span> <span class="n">squeeze_me</span><span class="o">=</span><span class="kc">True</span><span class="p">)[</span><span class="s2">&quot;annotations&quot;</span><span class="p">]</span>
+        <span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="n">sio</span><span class="o">.</span><span class="n">loadmat</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">devkit</span> <span class="o">/</span> <span class="s2">&quot;cars_meta.mat&quot;</span><span class="p">),</span> <span class="n">squeeze_me</span><span class="o">=</span><span class="kc">True</span><span class="p">)[</span><span class="s2">&quot;class_names&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="bp">cls</span><span class="p">:</span> <span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">)}</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">)</span>
+
+<div class="viewcode-block" id="StanfordCars.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.StanfordCars.html#torchvision.datasets.StanfordCars.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Returns pil_image and class_id for given index&quot;&quot;&quot;</span>
+        <span class="n">image_path</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_samples</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">pil_image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_path</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">pil_image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">pil_image</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">pil_image</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://ai.stanford.edu/~jkrause/cars/car_devkit.tgz&quot;</span><span class="p">,</span>
+            <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+            <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;c3b158d763b6e2245038c8ad08e45376&quot;</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span>
+                <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://ai.stanford.edu/~jkrause/car196/cars_train.tgz&quot;</span><span class="p">,</span>
+                <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+                <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;065e5b463ae28d29e77c1b4b166cfe61&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span>
+                <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://ai.stanford.edu/~jkrause/car196/cars_test.tgz&quot;</span><span class="p">,</span>
+                <span class="n">download_root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+                <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;4ce7ebf6a94d07f1952d94dd34c4d501&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">download_url</span><span class="p">(</span>
+                <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://ai.stanford.edu/~jkrause/car196/cars_test_annos_withlabels.mat&quot;</span><span class="p">,</span>
+                <span class="n">root</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span><span class="p">),</span>
+                <span class="n">md5</span><span class="o">=</span><span class="s2">&quot;b0a2b23655a3edd16d84508592a98d10&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_base_folder</span> <span class="o">/</span> <span class="s2">&quot;devkit&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">is_dir</span><span class="p">():</span>
+            <span class="k">return</span> <span class="kc">False</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_annotations_mat_path</span><span class="o">.</span><span class="n">exists</span><span class="p">()</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">_images_base_path</span><span class="o">.</span><span class="n">is_dir</span><span class="p">()</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/stl10.html b/0.16/_modules/torchvision/datasets/stl10.html
new file mode 100644
index 00000000000..2385a7b94c5
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/stl10.html
@@ -0,0 +1,843 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.stl10 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.stl10</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.stl10</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="STL10"><a class="viewcode-back" href="../../../generated/torchvision.datasets.STL10.html#torchvision.datasets.STL10">[docs]</a><span class="k">class</span> <span class="nc">STL10</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`STL10 &lt;https://cs.stanford.edu/~acoates/stl10/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset where directory</span>
+<span class="sd">            ``stl10_binary`` exists.</span>
+<span class="sd">        split (string): One of {&#39;train&#39;, &#39;test&#39;, &#39;unlabeled&#39;, &#39;train+unlabeled&#39;}.</span>
+<span class="sd">            Accordingly, dataset is selected.</span>
+<span class="sd">        folds (int, optional): One of {0-9} or None.</span>
+<span class="sd">            For training, loads one of the 10 pre-defined folds of 1k samples for the</span>
+<span class="sd">            standard evaluation procedure. If no value is passed, loads the 5k samples.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">base_folder</span> <span class="o">=</span> <span class="s2">&quot;stl10_binary&quot;</span>
+    <span class="n">url</span> <span class="o">=</span> <span class="s2">&quot;http://ai.stanford.edu/~acoates/stl10/stl10_binary.tar.gz&quot;</span>
+    <span class="n">filename</span> <span class="o">=</span> <span class="s2">&quot;stl10_binary.tar.gz&quot;</span>
+    <span class="n">tgz_md5</span> <span class="o">=</span> <span class="s2">&quot;91f7769df0f17e558f3565bffb0c7dfb&quot;</span>
+    <span class="n">class_names_file</span> <span class="o">=</span> <span class="s2">&quot;class_names.txt&quot;</span>
+    <span class="n">folds_list_file</span> <span class="o">=</span> <span class="s2">&quot;fold_indices.txt&quot;</span>
+    <span class="n">train_list</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">[</span><span class="s2">&quot;train_X.bin&quot;</span><span class="p">,</span> <span class="s2">&quot;918c2871b30a85fa023e0c44e0bee87f&quot;</span><span class="p">],</span>
+        <span class="p">[</span><span class="s2">&quot;train_y.bin&quot;</span><span class="p">,</span> <span class="s2">&quot;5a34089d4802c674881badbb80307741&quot;</span><span class="p">],</span>
+        <span class="p">[</span><span class="s2">&quot;unlabeled_X.bin&quot;</span><span class="p">,</span> <span class="s2">&quot;5242ba1fed5e4be9e1e742405eb56ca4&quot;</span><span class="p">],</span>
+    <span class="p">]</span>
+
+    <span class="n">test_list</span> <span class="o">=</span> <span class="p">[[</span><span class="s2">&quot;test_X.bin&quot;</span><span class="p">,</span> <span class="s2">&quot;7f263ba9f9e0b06b93213547f721ac82&quot;</span><span class="p">],</span> <span class="p">[</span><span class="s2">&quot;test_y.bin&quot;</span><span class="p">,</span> <span class="s2">&quot;36f9794fa4beb8a2c72628de14fa638e&quot;</span><span class="p">]]</span>
+    <span class="n">splits</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;train+unlabeled&quot;</span><span class="p">,</span> <span class="s2">&quot;unlabeled&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">folds</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">splits</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">folds</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_verify_folds</span><span class="p">(</span><span class="n">folds</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="c1"># now load the picked numpy arrays</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__loadfile</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_list</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_list</span><span class="p">[</span><span class="mi">1</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">__load_folds</span><span class="p">(</span><span class="n">folds</span><span class="p">)</span>
+
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train+unlabeled&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__loadfile</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_list</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_list</span><span class="p">[</span><span class="mi">1</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">__load_folds</span><span class="p">(</span><span class="n">folds</span><span class="p">)</span>
+            <span class="n">unlabeled_data</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__loadfile</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_list</span><span class="p">[</span><span class="mi">2</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">unlabeled_data</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">([</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">unlabeled_data</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">])))</span>
+
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;unlabeled&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__loadfile</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_list</span><span class="p">[</span><span class="mi">2</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">([</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>  <span class="c1"># self.split == &#39;test&#39;:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__loadfile</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">test_list</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">test_list</span><span class="p">[</span><span class="mi">1</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+
+        <span class="n">class_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">class_names_file</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">class_file</span><span class="p">):</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">class_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">read</span><span class="p">()</span><span class="o">.</span><span class="n">splitlines</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_verify_folds</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">folds</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="n">folds</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">folds</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">folds</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">folds</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">):</span>
+                <span class="k">return</span> <span class="n">folds</span>
+            <span class="n">msg</span> <span class="o">=</span> <span class="s2">&quot;Value for argument folds should be in the range [0, 10), but got </span><span class="si">{}</span><span class="s2">.&quot;</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">folds</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">msg</span> <span class="o">=</span> <span class="s2">&quot;Expected type None or int for argument folds, but got type </span><span class="si">{}</span><span class="s2">.&quot;</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">folds</span><span class="p">)))</span>
+
+<div class="viewcode-block" id="STL10.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.STL10.html#torchvision.datasets.STL10.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">target</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="nb">int</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="kc">None</span>
+
+        <span class="c1"># doing this so that it is consistent with all other datasets</span>
+        <span class="c1"># to return a PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)))</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="k">def</span> <span class="nf">__loadfile</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_file</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">labels_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]]:</span>
+        <span class="n">labels</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">labels_file</span><span class="p">:</span>
+            <span class="n">path_to_labels</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="n">labels_file</span><span class="p">)</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path_to_labels</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span>  <span class="c1"># 0-based</span>
+
+        <span class="n">path_to_data</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="n">data_file</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path_to_data</span><span class="p">,</span> <span class="s2">&quot;rb&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="c1"># read whole file in uint8 chunks</span>
+            <span class="n">everything</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+            <span class="n">images</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">everything</span><span class="p">,</span> <span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">96</span><span class="p">))</span>
+            <span class="n">images</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="n">images</span><span class="p">,</span> <span class="n">labels</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_list</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">test_list</span><span class="p">:</span>
+            <span class="n">fpath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">fpath</span><span class="p">,</span> <span class="n">md5</span><span class="p">):</span>
+                <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tgz_md5</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;Split: </span><span class="si">{split}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">__load_folds</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">folds</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># loads one of the folds if specified</span>
+        <span class="k">if</span> <span class="n">folds</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+        <span class="n">path_to_folds</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_folder</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">folds_list_file</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">path_to_folds</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">str_idx</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">read</span><span class="p">()</span><span class="o">.</span><span class="n">splitlines</span><span class="p">()[</span><span class="n">folds</span><span class="p">]</span>
+            <span class="n">list_idx</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">fromstring</span><span class="p">(</span><span class="n">str_idx</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">int64</span><span class="p">,</span> <span class="n">sep</span><span class="o">=</span><span class="s2">&quot; &quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">list_idx</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:]</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">[</span><span class="n">list_idx</span><span class="p">]</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/sun397.html b/0.16/_modules/torchvision/datasets/sun397.html
new file mode 100644
index 00000000000..bb88ca440e5
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/sun397.html
@@ -0,0 +1,745 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.sun397 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.sun397</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.sun397</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="SUN397"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SUN397.html#torchvision.datasets.SUN397">[docs]</a><span class="k">class</span> <span class="nc">SUN397</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`The SUN397 Data Set &lt;https://vision.princeton.edu/projects/2010/SUN/&gt;`_.</span>
+
+<span class="sd">    The SUN397 or Scene UNderstanding (SUN) is a dataset for scene recognition consisting of</span>
+<span class="sd">    397 categories with 108&#39;754 images.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image and returns a transformed</span>
+<span class="sd">            version. E.g, ``transforms.RandomCrop``.</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_DATASET_URL</span> <span class="o">=</span> <span class="s2">&quot;http://vision.princeton.edu/projects/2010/SUN/SUN397.tar.gz&quot;</span>
+    <span class="n">_DATASET_MD5</span> <span class="o">=</span> <span class="s2">&quot;8ca2778205c41d23104230ba66911c7a&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_data_dir</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;SUN397&quot;</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_dir</span> <span class="o">/</span> <span class="s2">&quot;ClassName.txt&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="n">c</span><span class="p">[</span><span class="mi">3</span><span class="p">:]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">f</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">))))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_dir</span><span class="o">.</span><span class="n">rglob</span><span class="p">(</span><span class="s2">&quot;sun_*.jpg&quot;</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">class_to_idx</span><span class="p">[</span><span class="s2">&quot;/&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">path</span><span class="o">.</span><span class="n">relative_to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_data_dir</span><span class="p">)</span><span class="o">.</span><span class="n">parts</span><span class="p">[</span><span class="mi">1</span><span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">])]</span> <span class="k">for</span> <span class="n">path</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span>
+        <span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">)</span>
+
+<div class="viewcode-block" id="SUN397.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SUN397.html#torchvision.datasets.SUN397.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">image_file</span><span class="p">,</span> <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_image_files</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">image_file</span><span class="p">)</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">:</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">label</span></div>
+
+    <span class="k">def</span> <span class="nf">_check_exists</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_data_dir</span><span class="o">.</span><span class="n">is_dir</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_exists</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_DATASET_URL</span><span class="p">,</span> <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_DATASET_MD5</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/svhn.html b/0.16/_modules/torchvision/datasets/svhn.html
new file mode 100644
index 00000000000..6557942cb98
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/svhn.html
@@ -0,0 +1,798 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.svhn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.svhn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.svhn</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os.path</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">check_integrity</span><span class="p">,</span> <span class="n">download_url</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="SVHN"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SVHN.html#torchvision.datasets.SVHN">[docs]</a><span class="k">class</span> <span class="nc">SVHN</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`SVHN &lt;http://ufldl.stanford.edu/housenumbers/&gt;`_ Dataset.</span>
+<span class="sd">    Note: The SVHN dataset assigns the label `10` to the digit `0`. However, in this Dataset,</span>
+<span class="sd">    we assign the label `0` to the digit `0` to be compatible with PyTorch loss functions which</span>
+<span class="sd">    expect the class labels to be in the range `[0, C-1]`</span>
+
+<span class="sd">    .. warning::</span>
+
+<span class="sd">        This class needs `scipy &lt;https://docs.scipy.org/doc/&gt;`_ to load data from `.mat` format.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the dataset where the data is stored.</span>
+<span class="sd">        split (string): One of {&#39;train&#39;, &#39;test&#39;, &#39;extra&#39;}.</span>
+<span class="sd">            Accordingly dataset is selected. &#39;extra&#39; is Extra training set.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">split_list</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;http://ufldl.stanford.edu/housenumbers/train_32x32.mat&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;train_32x32.mat&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;e26dedcc434d2e4c54c9b2d4a06d8373&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;http://ufldl.stanford.edu/housenumbers/test_32x32.mat&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;test_32x32.mat&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;eb5a983be6a315427106f1b164d9cef3&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;extra&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;http://ufldl.stanford.edu/housenumbers/extra_32x32.mat&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;extra_32x32.mat&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;a93ce644f1a588dc4d68dda5feec44a7&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_list</span><span class="o">.</span><span class="n">keys</span><span class="p">()))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">url</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_list</span><span class="p">[</span><span class="n">split</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">filename</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_list</span><span class="p">[</span><span class="n">split</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">file_md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_list</span><span class="p">[</span><span class="n">split</span><span class="p">][</span><span class="mi">2</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="c1"># import here rather than at top of file because this is</span>
+        <span class="c1"># an optional dependency for torchvision</span>
+        <span class="kn">import</span> <span class="nn">scipy.io</span> <span class="k">as</span> <span class="nn">sio</span>
+
+        <span class="c1"># reading(loading) mat file as array</span>
+        <span class="n">loaded_mat</span> <span class="o">=</span> <span class="n">sio</span><span class="o">.</span><span class="n">loadmat</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">loaded_mat</span><span class="p">[</span><span class="s2">&quot;X&quot;</span><span class="p">]</span>
+        <span class="c1"># loading from the .mat file gives an np.ndarray of type np.uint8</span>
+        <span class="c1"># converting to np.int64, so that we have a LongTensor after</span>
+        <span class="c1"># the conversion from the numpy array</span>
+        <span class="c1"># the squeeze is needed to obtain a 1D tensor</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">=</span> <span class="n">loaded_mat</span><span class="p">[</span><span class="s2">&quot;y&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">int64</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">()</span>
+
+        <span class="c1"># the svhn dataset assigns the class label &quot;10&quot; to the digit 0</span>
+        <span class="c1"># this makes it inconsistent with several loss functions</span>
+        <span class="c1"># which expect the class labels to be in the range [0, C-1]</span>
+        <span class="n">np</span><span class="o">.</span><span class="n">place</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">labels</span> <span class="o">==</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+
+<div class="viewcode-block" id="SVHN.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.SVHN.html#torchvision.datasets.SVHN.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="nb">int</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">labels</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+
+        <span class="c1"># doing this so that it is consistent with all other datasets</span>
+        <span class="c1"># to return a PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)))</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="n">root</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span>
+        <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_list</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">][</span><span class="mi">2</span><span class="p">]</span>
+        <span class="n">fpath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">check_integrity</span><span class="p">(</span><span class="n">fpath</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">md5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_list</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">][</span><span class="mi">2</span><span class="p">]</span>
+        <span class="n">download_url</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;Split: </span><span class="si">{split}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/ucf101.html b/0.16/_modules/torchvision/datasets/ucf101.html
new file mode 100644
index 00000000000..aefca6692a2
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/ucf101.html
@@ -0,0 +1,799 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.ucf101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.ucf101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.ucf101</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">.folder</span> <span class="kn">import</span> <span class="n">find_classes</span><span class="p">,</span> <span class="n">make_dataset</span>
+<span class="kn">from</span> <span class="nn">.video_utils</span> <span class="kn">import</span> <span class="n">VideoClips</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="UCF101"><a class="viewcode-back" href="../../../generated/torchvision.datasets.UCF101.html#torchvision.datasets.UCF101">[docs]</a><span class="k">class</span> <span class="nc">UCF101</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    `UCF101 &lt;https://www.crcv.ucf.edu/data/UCF101.php&gt;`_ dataset.</span>
+
+<span class="sd">    UCF101 is an action recognition video dataset.</span>
+<span class="sd">    This dataset consider every video as a collection of video clips of fixed size, specified</span>
+<span class="sd">    by ``frames_per_clip``, where the step in frames between each clip is given by</span>
+<span class="sd">    ``step_between_clips``. The dataset itself can be downloaded from the dataset website;</span>
+<span class="sd">    annotations that ``annotation_path`` should be pointing to can be downloaded from `here</span>
+<span class="sd">    &lt;https://www.crcv.ucf.edu/data/UCF101/UCF101TrainTestSplits-RecognitionTask.zip&gt;`_.</span>
+
+<span class="sd">    To give an example, for 2 videos with 10 and 15 frames respectively, if ``frames_per_clip=5``</span>
+<span class="sd">    and ``step_between_clips=5``, the dataset size will be (2 + 3) = 5, where the first two</span>
+<span class="sd">    elements will come from video 1, and the next three elements from video 2.</span>
+<span class="sd">    Note that we drop clips which do not have exactly ``frames_per_clip`` elements, so not all</span>
+<span class="sd">    frames in a video might be present.</span>
+
+<span class="sd">    Internally, it uses a VideoClips object to handle clip creation.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the UCF101 Dataset.</span>
+<span class="sd">        annotation_path (str): path to the folder containing the split files;</span>
+<span class="sd">            see docstring above for download instructions of these files</span>
+<span class="sd">        frames_per_clip (int): number of frames in a clip.</span>
+<span class="sd">        step_between_clips (int, optional): number of frames between each clip.</span>
+<span class="sd">        fold (int, optional): which fold to use. Should be between 1 and 3.</span>
+<span class="sd">        train (bool, optional): if ``True``, creates a dataset from the train split,</span>
+<span class="sd">            otherwise from the ``test`` split.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in a TxHxWxC video</span>
+<span class="sd">            and returns a transformed version.</span>
+<span class="sd">        output_format (str, optional): The format of the output video tensors (before transforms).</span>
+<span class="sd">            Can be either &quot;THWC&quot; (default) or &quot;TCHW&quot;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        tuple: A 3-tuple with the following entries:</span>
+
+<span class="sd">            - video (Tensor[T, H, W, C] or Tensor[T, C, H, W]): The `T` video frames</span>
+<span class="sd">            -  audio(Tensor[K, L]): the audio frames, where `K` is the number of channels</span>
+<span class="sd">               and `L` is the number of points</span>
+<span class="sd">            - label (int): class of the video clip</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">annotation_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">frames_per_clip</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">step_between_clips</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">frame_rate</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">fold</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">_precomputed_metadata</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">num_workers</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">_video_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_video_height</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_video_min_dimension</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">_audio_samples</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">output_format</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;THWC&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">fold</span> <span class="o">&lt;=</span> <span class="mi">3</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;fold should be between 1 and 3, got </span><span class="si">{</span><span class="n">fold</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="n">extensions</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;avi&quot;</span><span class="p">,)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fold</span> <span class="o">=</span> <span class="n">fold</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train</span> <span class="o">=</span> <span class="n">train</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">classes</span><span class="p">,</span> <span class="n">class_to_idx</span> <span class="o">=</span> <span class="n">find_classes</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">samples</span> <span class="o">=</span> <span class="n">make_dataset</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">class_to_idx</span><span class="p">,</span> <span class="n">extensions</span><span class="p">,</span> <span class="n">is_valid_file</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+        <span class="n">video_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">]</span>
+        <span class="n">video_clips</span> <span class="o">=</span> <span class="n">VideoClips</span><span class="p">(</span>
+            <span class="n">video_list</span><span class="p">,</span>
+            <span class="n">frames_per_clip</span><span class="p">,</span>
+            <span class="n">step_between_clips</span><span class="p">,</span>
+            <span class="n">frame_rate</span><span class="p">,</span>
+            <span class="n">_precomputed_metadata</span><span class="p">,</span>
+            <span class="n">num_workers</span><span class="o">=</span><span class="n">num_workers</span><span class="p">,</span>
+            <span class="n">_video_width</span><span class="o">=</span><span class="n">_video_width</span><span class="p">,</span>
+            <span class="n">_video_height</span><span class="o">=</span><span class="n">_video_height</span><span class="p">,</span>
+            <span class="n">_video_min_dimension</span><span class="o">=</span><span class="n">_video_min_dimension</span><span class="p">,</span>
+            <span class="n">_audio_samples</span><span class="o">=</span><span class="n">_audio_samples</span><span class="p">,</span>
+            <span class="n">output_format</span><span class="o">=</span><span class="n">output_format</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="c1"># we bookkeep the full version of video clips because we want to be able</span>
+        <span class="c1"># to return the metadata of full version rather than the subset version of</span>
+        <span class="c1"># video clips</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">full_video_clips</span> <span class="o">=</span> <span class="n">video_clips</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">indices</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_select_fold</span><span class="p">(</span><span class="n">video_list</span><span class="p">,</span> <span class="n">annotation_path</span><span class="p">,</span> <span class="n">fold</span><span class="p">,</span> <span class="n">train</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span> <span class="o">=</span> <span class="n">video_clips</span><span class="o">.</span><span class="n">subset</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">indices</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">transform</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">metadata</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">full_video_clips</span><span class="o">.</span><span class="n">metadata</span>
+
+    <span class="k">def</span> <span class="nf">_select_fold</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">video_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">annotation_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">fold</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+        <span class="n">name</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span> <span class="k">if</span> <span class="n">train</span> <span class="k">else</span> <span class="s2">&quot;test&quot;</span>
+        <span class="n">name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">list</span><span class="si">{</span><span class="n">fold</span><span class="si">:</span><span class="s2">02d</span><span class="si">}</span><span class="s2">.txt&quot;</span>
+        <span class="n">f</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">annotation_path</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
+        <span class="n">selected_files</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">as</span> <span class="n">fid</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">fid</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">data</span><span class="p">]</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="p">[</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="o">*</span><span class="n">x</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;/&quot;</span><span class="p">))</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">data</span><span class="p">]</span>
+            <span class="n">selected_files</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">data</span><span class="p">)</span>
+        <span class="n">indices</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">video_list</span><span class="p">))</span> <span class="k">if</span> <span class="n">video_list</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="ow">in</span> <span class="n">selected_files</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">indices</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">num_clips</span><span class="p">()</span>
+
+<div class="viewcode-block" id="UCF101.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.UCF101.html#torchvision.datasets.UCF101.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">info</span><span class="p">,</span> <span class="n">video_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">get_clip</span><span class="p">(</span><span class="n">idx</span><span class="p">)</span>
+        <span class="n">label</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">indices</span><span class="p">[</span><span class="n">video_idx</span><span class="p">]][</span><span class="mi">1</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">label</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/usps.html b/0.16/_modules/torchvision/datasets/usps.html
new file mode 100644
index 00000000000..4f729294c69
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/usps.html
@@ -0,0 +1,764 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.usps &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.usps</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.usps</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_url</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="USPS"><a class="viewcode-back" href="../../../generated/torchvision.datasets.USPS.html#torchvision.datasets.USPS">[docs]</a><span class="k">class</span> <span class="nc">USPS</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`USPS &lt;https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#usps&gt;`_ Dataset.</span>
+<span class="sd">    The data-format is : [label [index:value ]*256 \\n] * num_lines, where ``label`` lies in ``[1, 10]``.</span>
+<span class="sd">    The value for each pixel lies in ``[-1, 1]``. Here we transform the ``label`` into ``[0, 9]``</span>
+<span class="sd">    and make pixel values in ``[0, 255]``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset to store``USPS`` data files.</span>
+<span class="sd">        train (bool, optional): If True, creates dataset from ``usps.bz2``,</span>
+<span class="sd">            otherwise from ``usps.t.bz2``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">split_list</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass/usps.bz2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;usps.bz2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;ec16c51db3855ca6c91edd34d0e9b197&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;test&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="s2">&quot;https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass/usps.t.bz2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;usps.t.bz2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;8ea070ee2aca1ac39742fdd1ef5ed118&quot;</span><span class="p">,</span>
+        <span class="p">],</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">train</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span><span class="p">)</span>
+        <span class="n">split</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span> <span class="k">if</span> <span class="n">train</span> <span class="k">else</span> <span class="s2">&quot;test&quot;</span>
+        <span class="n">url</span><span class="p">,</span> <span class="n">filename</span><span class="p">,</span> <span class="n">checksum</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">split_list</span><span class="p">[</span><span class="n">split</span><span class="p">]</span>
+        <span class="n">full_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">full_path</span><span class="p">):</span>
+            <span class="n">download_url</span><span class="p">(</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="n">checksum</span><span class="p">)</span>
+
+        <span class="kn">import</span> <span class="nn">bz2</span>
+
+        <span class="k">with</span> <span class="n">bz2</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">full_path</span><span class="p">)</span> <span class="k">as</span> <span class="n">fp</span><span class="p">:</span>
+            <span class="n">raw_data</span> <span class="o">=</span> <span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">decode</span><span class="p">()</span><span class="o">.</span><span class="n">split</span><span class="p">()</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">fp</span><span class="o">.</span><span class="n">readlines</span><span class="p">()]</span>
+            <span class="n">tmp_list</span> <span class="o">=</span> <span class="p">[[</span><span class="n">x</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">)[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">data</span><span class="p">[</span><span class="mi">1</span><span class="p">:]]</span> <span class="k">for</span> <span class="n">data</span> <span class="ow">in</span> <span class="n">raw_data</span><span class="p">]</span>
+            <span class="n">imgs</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">tmp_list</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">16</span><span class="p">))</span>
+            <span class="n">imgs</span> <span class="o">=</span> <span class="p">((</span><span class="n">imgs</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span> <span class="o">*</span> <span class="mi">255</span><span class="p">)</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+            <span class="n">targets</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">d</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">-</span> <span class="mi">1</span> <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <span class="n">raw_data</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">imgs</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="n">targets</span>
+
+<div class="viewcode-block" id="USPS.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.USPS.html#torchvision.datasets.USPS.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is index of the target class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="nb">int</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+
+        <span class="c1"># doing this so that it is consistent with all other datasets</span>
+        <span class="c1"># to return a PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;L&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/vision.html b/0.16/_modules/torchvision/datasets/vision.html
new file mode 100644
index 00000000000..60beb10a328
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/vision.html
@@ -0,0 +1,779 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.vision &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.vision</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.vision</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch.utils.data</span> <span class="k">as</span> <span class="nn">data</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+
+<div class="viewcode-block" id="VisionDataset"><a class="viewcode-back" href="../../../generated/torchvision.datasets.VisionDataset.html#torchvision.datasets.VisionDataset">[docs]</a><span class="k">class</span> <span class="nc">VisionDataset</span><span class="p">(</span><span class="n">data</span><span class="o">.</span><span class="n">Dataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Base Class For making datasets which are compatible with torchvision.</span>
+<span class="sd">    It is necessary to override the ``__getitem__`` and ``__len__`` method.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of dataset.</span>
+<span class="sd">        transforms (callable, optional): A function/transforms that takes in</span>
+<span class="sd">            an image and a label and returns the transformed versions of both.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        :attr:`transforms` and the combination of :attr:`transform` and :attr:`target_transform` are mutually exclusive.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_repr_indent</span> <span class="o">=</span> <span class="mi">4</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="n">root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">expanduser</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">root</span>
+
+        <span class="n">has_transforms</span> <span class="o">=</span> <span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+        <span class="n">has_separate_transform</span> <span class="o">=</span> <span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">has_transforms</span> <span class="ow">and</span> <span class="n">has_separate_transform</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Only transforms or transform/target_transform can be passed as argument&quot;</span><span class="p">)</span>
+
+        <span class="c1"># for backwards-compatibility</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">transform</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="o">=</span> <span class="n">target_transform</span>
+
+        <span class="k">if</span> <span class="n">has_separate_transform</span><span class="p">:</span>
+            <span class="n">transforms</span> <span class="o">=</span> <span class="n">StandardTransform</span><span class="p">(</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            (Any): Sample and meta data, optionally transformed by the respective transforms.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">head</span> <span class="o">=</span> <span class="s2">&quot;Dataset &quot;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span>
+        <span class="n">body</span> <span class="o">=</span> <span class="p">[</span><span class="sa">f</span><span class="s2">&quot;Number of datapoints: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="fm">__len__</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">body</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Root location: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">body</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">extra_repr</span><span class="p">()</span><span class="o">.</span><span class="n">splitlines</span><span class="p">()</span>
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;transforms&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">body</span> <span class="o">+=</span> <span class="p">[</span><span class="nb">repr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">)]</span>
+        <span class="n">lines</span> <span class="o">=</span> <span class="p">[</span><span class="n">head</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="s2">&quot; &quot;</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">_repr_indent</span> <span class="o">+</span> <span class="n">line</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">body</span><span class="p">]</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">lines</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_format_transform_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transform</span><span class="p">:</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">head</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="n">lines</span> <span class="o">=</span> <span class="n">transform</span><span class="o">.</span><span class="fm">__repr__</span><span class="p">()</span><span class="o">.</span><span class="n">splitlines</span><span class="p">()</span>
+        <span class="k">return</span> <span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">head</span><span class="si">}{</span><span class="n">lines</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="s2">&quot;</span><span class="si">{}{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="s2">&quot; &quot;</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">head</span><span class="p">),</span> <span class="n">line</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">[</span><span class="mi">1</span><span class="p">:]]</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="s2">&quot;&quot;</span></div>
+
+
+<span class="k">class</span> <span class="nc">StandardTransform</span><span class="p">:</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">transform</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="o">=</span> <span class="n">target_transform</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">target</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="nb">input</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">input</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">def</span> <span class="nf">_format_transform_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transform</span><span class="p">:</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">head</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="n">lines</span> <span class="o">=</span> <span class="n">transform</span><span class="o">.</span><span class="fm">__repr__</span><span class="p">()</span><span class="o">.</span><span class="n">splitlines</span><span class="p">()</span>
+        <span class="k">return</span> <span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">head</span><span class="si">}{</span><span class="n">lines</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="s2">&quot;</span><span class="si">{}{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="s2">&quot; &quot;</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">head</span><span class="p">),</span> <span class="n">line</span><span class="p">)</span> <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">[</span><span class="mi">1</span><span class="p">:]]</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">body</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">body</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_format_transform_repr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">,</span> <span class="s2">&quot;Transform: &quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">body</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_format_transform_repr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">,</span> <span class="s2">&quot;Target transform: &quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">body</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/voc.html b/0.16/_modules/torchvision/datasets/voc.html
new file mode 100644
index 00000000000..1f3efb77f98
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/voc.html
@@ -0,0 +1,893 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.voc &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.voc</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.voc</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">collections</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">xml.etree.ElementTree</span> <span class="kn">import</span> <span class="n">Element</span> <span class="k">as</span> <span class="n">ET_Element</span>
+
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">from</span> <span class="nn">defusedxml.ElementTree</span> <span class="kn">import</span> <span class="n">parse</span> <span class="k">as</span> <span class="n">ET_parse</span>
+<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+    <span class="kn">from</span> <span class="nn">xml.etree.ElementTree</span> <span class="kn">import</span> <span class="n">parse</span> <span class="k">as</span> <span class="n">ET_parse</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+
+<span class="n">DATASET_YEAR_DICT</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;2012&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;VOCtrainval_11-May-2012.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;6cd6e144f989b92b3379bac3b3de84fd&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;base_dir&quot;</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="s2">&quot;VOCdevkit&quot;</span><span class="p">,</span> <span class="s2">&quot;VOC2012&quot;</span><span class="p">),</span>
+    <span class="p">},</span>
+    <span class="s2">&quot;2011&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;http://host.robots.ox.ac.uk/pascal/VOC/voc2011/VOCtrainval_25-May-2011.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;VOCtrainval_25-May-2011.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;6c3384ef61512963050cb5d687e5bf1e&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;base_dir&quot;</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="s2">&quot;TrainVal&quot;</span><span class="p">,</span> <span class="s2">&quot;VOCdevkit&quot;</span><span class="p">,</span> <span class="s2">&quot;VOC2011&quot;</span><span class="p">),</span>
+    <span class="p">},</span>
+    <span class="s2">&quot;2010&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;http://host.robots.ox.ac.uk/pascal/VOC/voc2010/VOCtrainval_03-May-2010.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;VOCtrainval_03-May-2010.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;da459979d0c395079b5c75ee67908abb&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;base_dir&quot;</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="s2">&quot;VOCdevkit&quot;</span><span class="p">,</span> <span class="s2">&quot;VOC2010&quot;</span><span class="p">),</span>
+    <span class="p">},</span>
+    <span class="s2">&quot;2009&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;http://host.robots.ox.ac.uk/pascal/VOC/voc2009/VOCtrainval_11-May-2009.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;VOCtrainval_11-May-2009.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;a3e00b113cfcfebf17e343f59da3caa1&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;base_dir&quot;</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="s2">&quot;VOCdevkit&quot;</span><span class="p">,</span> <span class="s2">&quot;VOC2009&quot;</span><span class="p">),</span>
+    <span class="p">},</span>
+    <span class="s2">&quot;2008&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;http://host.robots.ox.ac.uk/pascal/VOC/voc2008/VOCtrainval_14-Jul-2008.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;VOCtrainval_11-May-2012.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;2629fa636546599198acfcfbfcf1904a&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;base_dir&quot;</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="s2">&quot;VOCdevkit&quot;</span><span class="p">,</span> <span class="s2">&quot;VOC2008&quot;</span><span class="p">),</span>
+    <span class="p">},</span>
+    <span class="s2">&quot;2007&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;VOCtrainval_06-Nov-2007.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;c52e279531787c972589f7e41ab4ae64&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;base_dir&quot;</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="s2">&quot;VOCdevkit&quot;</span><span class="p">,</span> <span class="s2">&quot;VOC2007&quot;</span><span class="p">),</span>
+    <span class="p">},</span>
+    <span class="s2">&quot;2007-test&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;url&quot;</span><span class="p">:</span> <span class="s2">&quot;http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtest_06-Nov-2007.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;filename&quot;</span><span class="p">:</span> <span class="s2">&quot;VOCtest_06-Nov-2007.tar&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;md5&quot;</span><span class="p">:</span> <span class="s2">&quot;b6e924de25625d8de591ea690078ad9f&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;base_dir&quot;</span><span class="p">:</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="s2">&quot;VOCdevkit&quot;</span><span class="p">,</span> <span class="s2">&quot;VOC2007&quot;</span><span class="p">),</span>
+    <span class="p">},</span>
+<span class="p">}</span>
+
+
+<span class="k">class</span> <span class="nc">_VOCBase</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+    <span class="n">_SPLITS_DIR</span><span class="p">:</span> <span class="nb">str</span>
+    <span class="n">_TARGET_DIR</span><span class="p">:</span> <span class="nb">str</span>
+    <span class="n">_TARGET_FILE_EXT</span><span class="p">:</span> <span class="nb">str</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">year</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;2012&quot;</span><span class="p">,</span>
+        <span class="n">image_set</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">,</span> <span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">year</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">year</span><span class="p">,</span> <span class="s2">&quot;year&quot;</span><span class="p">,</span> <span class="n">valid_values</span><span class="o">=</span><span class="p">[</span><span class="nb">str</span><span class="p">(</span><span class="n">yr</span><span class="p">)</span> <span class="k">for</span> <span class="n">yr</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">2007</span><span class="p">,</span> <span class="mi">2013</span><span class="p">)])</span>
+
+        <span class="n">valid_image_sets</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;trainval&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">year</span> <span class="o">==</span> <span class="s2">&quot;2007&quot;</span><span class="p">:</span>
+            <span class="n">valid_image_sets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">&quot;test&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_set</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">image_set</span><span class="p">,</span> <span class="s2">&quot;image_set&quot;</span><span class="p">,</span> <span class="n">valid_image_sets</span><span class="p">)</span>
+
+        <span class="n">key</span> <span class="o">=</span> <span class="s2">&quot;2007-test&quot;</span> <span class="k">if</span> <span class="n">year</span> <span class="o">==</span> <span class="s2">&quot;2007&quot;</span> <span class="ow">and</span> <span class="n">image_set</span> <span class="o">==</span> <span class="s2">&quot;test&quot;</span> <span class="k">else</span> <span class="n">year</span>
+        <span class="n">dataset_year_dict</span> <span class="o">=</span> <span class="n">DATASET_YEAR_DICT</span><span class="p">[</span><span class="n">key</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">url</span> <span class="o">=</span> <span class="n">dataset_year_dict</span><span class="p">[</span><span class="s2">&quot;url&quot;</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">filename</span> <span class="o">=</span> <span class="n">dataset_year_dict</span><span class="p">[</span><span class="s2">&quot;filename&quot;</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">md5</span> <span class="o">=</span> <span class="n">dataset_year_dict</span><span class="p">[</span><span class="s2">&quot;md5&quot;</span><span class="p">]</span>
+
+        <span class="n">base_dir</span> <span class="o">=</span> <span class="n">dataset_year_dict</span><span class="p">[</span><span class="s2">&quot;base_dir&quot;</span><span class="p">]</span>
+        <span class="n">voc_root</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">base_dir</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="n">download_and_extract_archive</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">md5</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">voc_root</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download it&quot;</span><span class="p">)</span>
+
+        <span class="n">splits_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">voc_root</span><span class="p">,</span> <span class="s2">&quot;ImageSets&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_SPLITS_DIR</span><span class="p">)</span>
+        <span class="n">split_f</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">splits_dir</span><span class="p">,</span> <span class="n">image_set</span><span class="o">.</span><span class="n">rstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span> <span class="o">+</span> <span class="s2">&quot;.txt&quot;</span><span class="p">)</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">split_f</span><span class="p">))</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">file_names</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">f</span><span class="o">.</span><span class="n">readlines</span><span class="p">()]</span>
+
+        <span class="n">image_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">voc_root</span><span class="p">,</span> <span class="s2">&quot;JPEGImages&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">images</span> <span class="o">=</span> <span class="p">[</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">image_dir</span><span class="p">,</span> <span class="n">x</span> <span class="o">+</span> <span class="s2">&quot;.jpg&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">file_names</span><span class="p">]</span>
+
+        <span class="n">target_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">voc_root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_TARGET_DIR</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">targets</span> <span class="o">=</span> <span class="p">[</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">target_dir</span><span class="p">,</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">_TARGET_FILE_EXT</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">file_names</span><span class="p">]</span>
+
+        <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">)</span> <span class="o">==</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">targets</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="VOCSegmentation"><a class="viewcode-back" href="../../../generated/torchvision.datasets.VOCSegmentation.html#torchvision.datasets.VOCSegmentation">[docs]</a><span class="k">class</span> <span class="nc">VOCSegmentation</span><span class="p">(</span><span class="n">_VOCBase</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Pascal VOC &lt;http://host.robots.ox.ac.uk/pascal/VOC/&gt;`_ Segmentation Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the VOC Dataset.</span>
+<span class="sd">        year (string, optional): The dataset year, supports years ``&quot;2007&quot;`` to ``&quot;2012&quot;``.</span>
+<span class="sd">        image_set (string, optional): Select the image_set to use, ``&quot;train&quot;``, ``&quot;trainval&quot;`` or ``&quot;val&quot;``. If</span>
+<span class="sd">            ``year==&quot;2007&quot;``, can also be ``&quot;test&quot;``.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes input sample and its target as entry</span>
+<span class="sd">            and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_SPLITS_DIR</span> <span class="o">=</span> <span class="s2">&quot;Segmentation&quot;</span>
+    <span class="n">_TARGET_DIR</span> <span class="o">=</span> <span class="s2">&quot;SegmentationClass&quot;</span>
+    <span class="n">_TARGET_FILE_EXT</span> <span class="o">=</span> <span class="s2">&quot;.png&quot;</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">masks</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span>
+
+<div class="viewcode-block" id="VOCSegmentation.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.VOCSegmentation.html#torchvision.datasets.VOCSegmentation.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is the image segmentation.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">[</span><span class="n">index</span><span class="p">])</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">masks</span><span class="p">[</span><span class="n">index</span><span class="p">])</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div></div>
+
+
+<div class="viewcode-block" id="VOCDetection"><a class="viewcode-back" href="../../../generated/torchvision.datasets.VOCDetection.html#torchvision.datasets.VOCDetection">[docs]</a><span class="k">class</span> <span class="nc">VOCDetection</span><span class="p">(</span><span class="n">_VOCBase</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`Pascal VOC &lt;http://host.robots.ox.ac.uk/pascal/VOC/&gt;`_ Detection Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory of the VOC Dataset.</span>
+<span class="sd">        year (string, optional): The dataset year, supports years ``&quot;2007&quot;`` to ``&quot;2012&quot;``.</span>
+<span class="sd">        image_set (string, optional): Select the image_set to use, ``&quot;train&quot;``, ``&quot;trainval&quot;`` or ``&quot;val&quot;``. If</span>
+<span class="sd">            ``year==&quot;2007&quot;``, can also be ``&quot;test&quot;``.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+<span class="sd">            (default: alphabetic indexing of VOC&#39;s 20 classes).</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in an PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, required): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        transforms (callable, optional): A function/transform that takes input sample and its target as entry</span>
+<span class="sd">            and returns a transformed version.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_SPLITS_DIR</span> <span class="o">=</span> <span class="s2">&quot;Main&quot;</span>
+    <span class="n">_TARGET_DIR</span> <span class="o">=</span> <span class="s2">&quot;Annotations&quot;</span>
+    <span class="n">_TARGET_FILE_EXT</span> <span class="o">=</span> <span class="s2">&quot;.xml&quot;</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">annotations</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">targets</span>
+
+<div class="viewcode-block" id="VOCDetection.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.VOCDetection.html#torchvision.datasets.VOCDetection.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is a dictionary of the XML tree.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">images</span><span class="p">[</span><span class="n">index</span><span class="p">])</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">parse_voc_xml</span><span class="p">(</span><span class="n">ET_parse</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">annotations</span><span class="p">[</span><span class="n">index</span><span class="p">])</span><span class="o">.</span><span class="n">getroot</span><span class="p">())</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">parse_voc_xml</span><span class="p">(</span><span class="n">node</span><span class="p">:</span> <span class="n">ET_Element</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">voc_dict</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">children</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">node</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">children</span><span class="p">:</span>
+            <span class="n">def_dic</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="n">collections</span><span class="o">.</span><span class="n">defaultdict</span><span class="p">(</span><span class="nb">list</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">dc</span> <span class="ow">in</span> <span class="nb">map</span><span class="p">(</span><span class="n">VOCDetection</span><span class="o">.</span><span class="n">parse_voc_xml</span><span class="p">,</span> <span class="n">children</span><span class="p">):</span>
+                <span class="k">for</span> <span class="n">ind</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">dc</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                    <span class="n">def_dic</span><span class="p">[</span><span class="n">ind</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">v</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">node</span><span class="o">.</span><span class="n">tag</span> <span class="o">==</span> <span class="s2">&quot;annotation&quot;</span><span class="p">:</span>
+                <span class="n">def_dic</span><span class="p">[</span><span class="s2">&quot;object&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span><span class="n">def_dic</span><span class="p">[</span><span class="s2">&quot;object&quot;</span><span class="p">]]</span>
+            <span class="n">voc_dict</span> <span class="o">=</span> <span class="p">{</span><span class="n">node</span><span class="o">.</span><span class="n">tag</span><span class="p">:</span> <span class="p">{</span><span class="n">ind</span><span class="p">:</span> <span class="n">v</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">v</span> <span class="k">for</span> <span class="n">ind</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">def_dic</span><span class="o">.</span><span class="n">items</span><span class="p">()}}</span>
+        <span class="k">if</span> <span class="n">node</span><span class="o">.</span><span class="n">text</span><span class="p">:</span>
+            <span class="n">text</span> <span class="o">=</span> <span class="n">node</span><span class="o">.</span><span class="n">text</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">children</span><span class="p">:</span>
+                <span class="n">voc_dict</span><span class="p">[</span><span class="n">node</span><span class="o">.</span><span class="n">tag</span><span class="p">]</span> <span class="o">=</span> <span class="n">text</span>
+        <span class="k">return</span> <span class="n">voc_dict</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/datasets/widerface.html b/0.16/_modules/torchvision/datasets/widerface.html
new file mode 100644
index 00000000000..67d29ce8f97
--- /dev/null
+++ b/0.16/_modules/torchvision/datasets/widerface.html
@@ -0,0 +1,860 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.datasets.widerface &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.datasets.widerface</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.datasets.widerface</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">from</span> <span class="nn">os.path</span> <span class="kn">import</span> <span class="n">abspath</span><span class="p">,</span> <span class="n">expanduser</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">download_and_extract_archive</span><span class="p">,</span> <span class="n">download_file_from_google_drive</span><span class="p">,</span> <span class="n">extract_archive</span><span class="p">,</span> <span class="n">verify_str_arg</span>
+<span class="kn">from</span> <span class="nn">.vision</span> <span class="kn">import</span> <span class="n">VisionDataset</span>
+
+
+<div class="viewcode-block" id="WIDERFace"><a class="viewcode-back" href="../../../generated/torchvision.datasets.WIDERFace.html#torchvision.datasets.WIDERFace">[docs]</a><span class="k">class</span> <span class="nc">WIDERFace</span><span class="p">(</span><span class="n">VisionDataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;`WIDERFace &lt;http://shuoyang1213.me/WIDERFACE/&gt;`_ Dataset.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        root (string): Root directory where images and annotations are downloaded to.</span>
+<span class="sd">            Expects the following folder structure if download=False:</span>
+
+<span class="sd">            .. code::</span>
+
+<span class="sd">                &lt;root&gt;</span>
+<span class="sd">                    └── widerface</span>
+<span class="sd">                        ├── wider_face_split (&#39;wider_face_split.zip&#39; if compressed)</span>
+<span class="sd">                        ├── WIDER_train (&#39;WIDER_train.zip&#39; if compressed)</span>
+<span class="sd">                        ├── WIDER_val (&#39;WIDER_val.zip&#39; if compressed)</span>
+<span class="sd">                        └── WIDER_test (&#39;WIDER_test.zip&#39; if compressed)</span>
+<span class="sd">        split (string): The dataset split to use. One of {``train``, ``val``, ``test``}.</span>
+<span class="sd">            Defaults to ``train``.</span>
+<span class="sd">        transform (callable, optional): A function/transform that  takes in a PIL image</span>
+<span class="sd">            and returns a transformed version. E.g, ``transforms.RandomCrop``</span>
+<span class="sd">        target_transform (callable, optional): A function/transform that takes in the</span>
+<span class="sd">            target and transforms it.</span>
+<span class="sd">        download (bool, optional): If true, downloads the dataset from the internet and</span>
+<span class="sd">            puts it in root directory. If dataset is already downloaded, it is not</span>
+<span class="sd">            downloaded again.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">BASE_FOLDER</span> <span class="o">=</span> <span class="s2">&quot;widerface&quot;</span>
+    <span class="n">FILE_LIST</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="c1"># File ID                             MD5 Hash                            Filename</span>
+        <span class="p">(</span><span class="s2">&quot;15hGDLhsx8bLgLcIRD5DhYt5iBxnjNF1M&quot;</span><span class="p">,</span> <span class="s2">&quot;3fedf70df600953d25982bcd13d91ba2&quot;</span><span class="p">,</span> <span class="s2">&quot;WIDER_train.zip&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;1GUCogbp16PMGa39thoMMeWxp7Rp5oM8Q&quot;</span><span class="p">,</span> <span class="s2">&quot;dfa7d7e790efa35df3788964cf0bbaea&quot;</span><span class="p">,</span> <span class="s2">&quot;WIDER_val.zip&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;1HIfDbVEWKmsYKJZm4lchTBDLW5N7dY5T&quot;</span><span class="p">,</span> <span class="s2">&quot;e5d8f4248ed24c334bbd12f49c29dd40&quot;</span><span class="p">,</span> <span class="s2">&quot;WIDER_test.zip&quot;</span><span class="p">),</span>
+    <span class="p">]</span>
+    <span class="n">ANNOTATIONS_FILE</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="s2">&quot;http://shuoyang1213.me/WIDERFACE/support/bbx_annotation/wider_face_split.zip&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;0e3767bcf0e326556d407bf5bff5d27c&quot;</span><span class="p">,</span>
+        <span class="s2">&quot;wider_face_split.zip&quot;</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">split</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">target_transform</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">download</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">root</span><span class="o">=</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">BASE_FOLDER</span><span class="p">),</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">target_transform</span><span class="o">=</span><span class="n">target_transform</span>
+        <span class="p">)</span>
+        <span class="c1"># check arguments</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">=</span> <span class="n">verify_str_arg</span><span class="p">(</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;split&quot;</span><span class="p">,</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">,</span> <span class="s2">&quot;test&quot;</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="n">download</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">download</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dataset not found or corrupted. You can use download=True to download and prepare it&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">img_info</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]]]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;val&quot;</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">parse_train_val_annotations_file</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">parse_test_annotations_file</span><span class="p">()</span>
+
+<div class="viewcode-block" id="WIDERFace.__getitem__"><a class="viewcode-back" href="../../../generated/torchvision.datasets.WIDERFace.html#torchvision.datasets.WIDERFace.__getitem__">[docs]</a>    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            index (int): Index</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: (image, target) where target is a dict of annotations for all faces in the image.</span>
+<span class="sd">            target=None for the test split.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="c1"># stay consistent with other datasets and return a PIL Image</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">img_info</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="s2">&quot;img_path&quot;</span><span class="p">])</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+        <span class="n">target</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;test&quot;</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">img_info</span><span class="p">[</span><span class="n">index</span><span class="p">][</span><span class="s2">&quot;annotations&quot;</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span><span class="p">,</span> <span class="n">target</span></div>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">img_info</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">lines</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Split: </span><span class="si">{split}</span><span class="s2">&quot;</span><span class="p">]</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">lines</span><span class="p">)</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">parse_train_val_annotations_file</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">filename</span> <span class="o">=</span> <span class="s2">&quot;wider_face_train_bbx_gt.txt&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span> <span class="k">else</span> <span class="s2">&quot;wider_face_val_bbx_gt.txt&quot;</span>
+        <span class="n">filepath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;wider_face_split&quot;</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span>
+
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">filepath</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">lines</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+            <span class="n">file_name_line</span><span class="p">,</span> <span class="n">num_boxes_line</span><span class="p">,</span> <span class="n">box_annotation_line</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span>
+            <span class="n">num_boxes</span><span class="p">,</span> <span class="n">box_counter</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span>
+            <span class="n">labels</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">:</span>
+                <span class="n">line</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">rstrip</span><span class="p">()</span>
+                <span class="k">if</span> <span class="n">file_name_line</span><span class="p">:</span>
+                    <span class="n">img_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;WIDER_&quot;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">split</span><span class="p">,</span> <span class="s2">&quot;images&quot;</span><span class="p">,</span> <span class="n">line</span><span class="p">)</span>
+                    <span class="n">img_path</span> <span class="o">=</span> <span class="n">abspath</span><span class="p">(</span><span class="n">expanduser</span><span class="p">(</span><span class="n">img_path</span><span class="p">))</span>
+                    <span class="n">file_name_line</span> <span class="o">=</span> <span class="kc">False</span>
+                    <span class="n">num_boxes_line</span> <span class="o">=</span> <span class="kc">True</span>
+                <span class="k">elif</span> <span class="n">num_boxes_line</span><span class="p">:</span>
+                    <span class="n">num_boxes</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">line</span><span class="p">)</span>
+                    <span class="n">num_boxes_line</span> <span class="o">=</span> <span class="kc">False</span>
+                    <span class="n">box_annotation_line</span> <span class="o">=</span> <span class="kc">True</span>
+                <span class="k">elif</span> <span class="n">box_annotation_line</span><span class="p">:</span>
+                    <span class="n">box_counter</span> <span class="o">+=</span> <span class="mi">1</span>
+                    <span class="n">line_split</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">)</span>
+                    <span class="n">line_values</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">line_split</span><span class="p">]</span>
+                    <span class="n">labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">line_values</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">box_counter</span> <span class="o">&gt;=</span> <span class="n">num_boxes</span><span class="p">:</span>
+                        <span class="n">box_annotation_line</span> <span class="o">=</span> <span class="kc">False</span>
+                        <span class="n">file_name_line</span> <span class="o">=</span> <span class="kc">True</span>
+                        <span class="n">labels_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">img_info</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                            <span class="p">{</span>
+                                <span class="s2">&quot;img_path&quot;</span><span class="p">:</span> <span class="n">img_path</span><span class="p">,</span>
+                                <span class="s2">&quot;annotations&quot;</span><span class="p">:</span> <span class="p">{</span>
+                                    <span class="s2">&quot;bbox&quot;</span><span class="p">:</span> <span class="n">labels_tensor</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">:</span><span class="mi">4</span><span class="p">]</span><span class="o">.</span><span class="n">clone</span><span class="p">(),</span>  <span class="c1"># x, y, width, height</span>
+                                    <span class="s2">&quot;blur&quot;</span><span class="p">:</span> <span class="n">labels_tensor</span><span class="p">[:,</span> <span class="mi">4</span><span class="p">]</span><span class="o">.</span><span class="n">clone</span><span class="p">(),</span>
+                                    <span class="s2">&quot;expression&quot;</span><span class="p">:</span> <span class="n">labels_tensor</span><span class="p">[:,</span> <span class="mi">5</span><span class="p">]</span><span class="o">.</span><span class="n">clone</span><span class="p">(),</span>
+                                    <span class="s2">&quot;illumination&quot;</span><span class="p">:</span> <span class="n">labels_tensor</span><span class="p">[:,</span> <span class="mi">6</span><span class="p">]</span><span class="o">.</span><span class="n">clone</span><span class="p">(),</span>
+                                    <span class="s2">&quot;occlusion&quot;</span><span class="p">:</span> <span class="n">labels_tensor</span><span class="p">[:,</span> <span class="mi">7</span><span class="p">]</span><span class="o">.</span><span class="n">clone</span><span class="p">(),</span>
+                                    <span class="s2">&quot;pose&quot;</span><span class="p">:</span> <span class="n">labels_tensor</span><span class="p">[:,</span> <span class="mi">8</span><span class="p">]</span><span class="o">.</span><span class="n">clone</span><span class="p">(),</span>
+                                    <span class="s2">&quot;invalid&quot;</span><span class="p">:</span> <span class="n">labels_tensor</span><span class="p">[:,</span> <span class="mi">9</span><span class="p">]</span><span class="o">.</span><span class="n">clone</span><span class="p">(),</span>
+                                <span class="p">},</span>
+                            <span class="p">}</span>
+                        <span class="p">)</span>
+                        <span class="n">box_counter</span> <span class="o">=</span> <span class="mi">0</span>
+                        <span class="n">labels</span><span class="o">.</span><span class="n">clear</span><span class="p">()</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Error parsing annotation file </span><span class="si">{</span><span class="n">filepath</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">parse_test_annotations_file</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">filepath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;wider_face_split&quot;</span><span class="p">,</span> <span class="s2">&quot;wider_face_test_filelist.txt&quot;</span><span class="p">)</span>
+        <span class="n">filepath</span> <span class="o">=</span> <span class="n">abspath</span><span class="p">(</span><span class="n">expanduser</span><span class="p">(</span><span class="n">filepath</span><span class="p">))</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">filepath</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">lines</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">readlines</span><span class="p">()</span>
+            <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">lines</span><span class="p">:</span>
+                <span class="n">line</span> <span class="o">=</span> <span class="n">line</span><span class="o">.</span><span class="n">rstrip</span><span class="p">()</span>
+                <span class="n">img_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;WIDER_test&quot;</span><span class="p">,</span> <span class="s2">&quot;images&quot;</span><span class="p">,</span> <span class="n">line</span><span class="p">)</span>
+                <span class="n">img_path</span> <span class="o">=</span> <span class="n">abspath</span><span class="p">(</span><span class="n">expanduser</span><span class="p">(</span><span class="n">img_path</span><span class="p">))</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">img_info</span><span class="o">.</span><span class="n">append</span><span class="p">({</span><span class="s2">&quot;img_path&quot;</span><span class="p">:</span> <span class="n">img_path</span><span class="p">})</span>
+
+    <span class="k">def</span> <span class="nf">_check_integrity</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="c1"># Allow original archive to be deleted (zip). Only need the extracted images</span>
+        <span class="n">all_files</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">FILE_LIST</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+        <span class="n">all_files</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ANNOTATIONS_FILE</span><span class="p">)</span>
+        <span class="k">for</span> <span class="p">(</span><span class="n">_</span><span class="p">,</span> <span class="n">md5</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span> <span class="ow">in</span> <span class="n">all_files</span><span class="p">:</span>
+            <span class="n">file</span><span class="p">,</span> <span class="n">ext</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">splitext</span><span class="p">(</span><span class="n">filename</span><span class="p">)</span>
+            <span class="n">extracted_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">file</span><span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">extracted_dir</span><span class="p">):</span>
+                <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+    <span class="k">def</span> <span class="nf">download</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_integrity</span><span class="p">():</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Files already downloaded and verified&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+
+        <span class="c1"># download and extract image data</span>
+        <span class="k">for</span> <span class="p">(</span><span class="n">file_id</span><span class="p">,</span> <span class="n">md5</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">FILE_LIST</span><span class="p">:</span>
+            <span class="n">download_file_from_google_drive</span><span class="p">(</span><span class="n">file_id</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="p">,</span> <span class="n">md5</span><span class="p">)</span>
+            <span class="n">filepath</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">filename</span><span class="p">)</span>
+            <span class="n">extract_archive</span><span class="p">(</span><span class="n">filepath</span><span class="p">)</span>
+
+        <span class="c1"># download and extract annotation files</span>
+        <span class="n">download_and_extract_archive</span><span class="p">(</span>
+            <span class="n">url</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">ANNOTATIONS_FILE</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">download_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">md5</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">ANNOTATIONS_FILE</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/io/image.html b/0.16/_modules/torchvision/io/image.html
new file mode 100644
index 00000000000..f209ef54b85
--- /dev/null
+++ b/0.16/_modules/torchvision/io/image.html
@@ -0,0 +1,933 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.io.image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.io.image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.io.image</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">enum</span> <span class="kn">import</span> <span class="n">Enum</span>
+<span class="kn">from</span> <span class="nn">warnings</span> <span class="kn">import</span> <span class="n">warn</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">..extension</span> <span class="kn">import</span> <span class="n">_load_library</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+
+<span class="k">try</span><span class="p">:</span>
+    <span class="n">_load_library</span><span class="p">(</span><span class="s2">&quot;image&quot;</span><span class="p">)</span>
+<span class="k">except</span> <span class="p">(</span><span class="ne">ImportError</span><span class="p">,</span> <span class="ne">OSError</span><span class="p">)</span> <span class="k">as</span> <span class="n">e</span><span class="p">:</span>
+    <span class="n">warn</span><span class="p">(</span>
+        <span class="sa">f</span><span class="s2">&quot;Failed to load image Python extension: &#39;</span><span class="si">{</span><span class="n">e</span><span class="si">}</span><span class="s2">&#39;&quot;</span>
+        <span class="sa">f</span><span class="s2">&quot;If you don&#39;t plan on using image functionality from `torchvision.io`, you can ignore this warning. &quot;</span>
+        <span class="sa">f</span><span class="s2">&quot;Otherwise, there might be something wrong with your environment. &quot;</span>
+        <span class="sa">f</span><span class="s2">&quot;Did you have `libjpeg` or `libpng` installed before building `torchvision` from source?&quot;</span>
+    <span class="p">)</span>
+
+
+<div class="viewcode-block" id="ImageReadMode"><a class="viewcode-back" href="../../../generated/torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode">[docs]</a><span class="k">class</span> <span class="nc">ImageReadMode</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Support for various modes while reading images.</span>
+
+<span class="sd">    Use ``ImageReadMode.UNCHANGED`` for loading the image as-is,</span>
+<span class="sd">    ``ImageReadMode.GRAY`` for converting to grayscale,</span>
+<span class="sd">    ``ImageReadMode.GRAY_ALPHA`` for grayscale with transparency,</span>
+<span class="sd">    ``ImageReadMode.RGB`` for RGB and ``ImageReadMode.RGB_ALPHA`` for</span>
+<span class="sd">    RGB with transparency.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">UNCHANGED</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">GRAY</span> <span class="o">=</span> <span class="mi">1</span>
+    <span class="n">GRAY_ALPHA</span> <span class="o">=</span> <span class="mi">2</span>
+    <span class="n">RGB</span> <span class="o">=</span> <span class="mi">3</span>
+    <span class="n">RGB_ALPHA</span> <span class="o">=</span> <span class="mi">4</span></div>
+
+
+<div class="viewcode-block" id="read_file"><a class="viewcode-back" href="../../../generated/torchvision.io.read_file.html#torchvision.io.read_file">[docs]</a><span class="k">def</span> <span class="nf">read_file</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Reads and outputs the bytes contents of a file as a uint8 Tensor</span>
+<span class="sd">    with one dimension.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        path (str): the path to the file to be read</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        data (Tensor)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">read_file</span><span class="p">)</span>
+    <span class="n">data</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">read_file</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">data</span></div>
+
+
+<div class="viewcode-block" id="write_file"><a class="viewcode-back" href="../../../generated/torchvision.io.write_file.html#torchvision.io.write_file">[docs]</a><span class="k">def</span> <span class="nf">write_file</span><span class="p">(</span><span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">data</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Writes the contents of an uint8 tensor with one dimension to a</span>
+<span class="sd">    file.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        filename (str): the path to the file to be written</span>
+<span class="sd">        data (Tensor): the contents to be written to the output file</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">write_file</span><span class="p">)</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">write_file</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">data</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="decode_png"><a class="viewcode-back" href="../../../generated/torchvision.io.decode_png.html#torchvision.io.decode_png">[docs]</a><span class="k">def</span> <span class="nf">decode_png</span><span class="p">(</span><span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="n">ImageReadMode</span> <span class="o">=</span> <span class="n">ImageReadMode</span><span class="o">.</span><span class="n">UNCHANGED</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Decodes a PNG image into a 3 dimensional RGB or grayscale Tensor.</span>
+<span class="sd">    Optionally converts the image to the desired format.</span>
+<span class="sd">    The values of the output tensor are uint8 in [0, 255].</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[1]): a one dimensional uint8 tensor containing</span>
+<span class="sd">            the raw bytes of the PNG image.</span>
+<span class="sd">        mode (ImageReadMode): the read mode used for optionally</span>
+<span class="sd">            converting the image. Default: ``ImageReadMode.UNCHANGED``.</span>
+<span class="sd">            See `ImageReadMode` class for more information on various</span>
+<span class="sd">            available modes.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        output (Tensor[image_channels, image_height, image_width])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">decode_png</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">decode_png</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">mode</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="encode_png"><a class="viewcode-back" href="../../../generated/torchvision.io.encode_png.html#torchvision.io.encode_png">[docs]</a><span class="k">def</span> <span class="nf">encode_png</span><span class="p">(</span><span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">compression_level</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">6</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Takes an input tensor in CHW layout and returns a buffer with the contents</span>
+<span class="sd">    of its corresponding PNG file.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[channels, image_height, image_width]): int8 image tensor of</span>
+<span class="sd">            ``c`` channels, where ``c`` must 3 or 1.</span>
+<span class="sd">        compression_level (int): Compression factor for the resulting file, it must be a number</span>
+<span class="sd">            between 0 and 9. Default: 6</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[1]: A one dimensional int8 tensor that contains the raw bytes of the</span>
+<span class="sd">            PNG file.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">encode_png</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">encode_png</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">compression_level</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="write_png"><a class="viewcode-back" href="../../../generated/torchvision.io.write_png.html#torchvision.io.write_png">[docs]</a><span class="k">def</span> <span class="nf">write_png</span><span class="p">(</span><span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">compression_level</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">6</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Takes an input tensor in CHW layout (or HW in the case of grayscale images)</span>
+<span class="sd">    and saves it in a PNG file.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[channels, image_height, image_width]): int8 image tensor of</span>
+<span class="sd">            ``c`` channels, where ``c`` must be 1 or 3.</span>
+<span class="sd">        filename (str): Path to save the image.</span>
+<span class="sd">        compression_level (int): Compression factor for the resulting file, it must be a number</span>
+<span class="sd">            between 0 and 9. Default: 6</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">write_png</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">encode_png</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">compression_level</span><span class="p">)</span>
+    <span class="n">write_file</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">output</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="decode_jpeg"><a class="viewcode-back" href="../../../generated/torchvision.io.decode_jpeg.html#torchvision.io.decode_jpeg">[docs]</a><span class="k">def</span> <span class="nf">decode_jpeg</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="n">ImageReadMode</span> <span class="o">=</span> <span class="n">ImageReadMode</span><span class="o">.</span><span class="n">UNCHANGED</span><span class="p">,</span> <span class="n">device</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;cpu&quot;</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Decodes a JPEG image into a 3 dimensional RGB or grayscale Tensor.</span>
+<span class="sd">    Optionally converts the image to the desired format.</span>
+<span class="sd">    The values of the output tensor are uint8 between 0 and 255.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[1]): a one dimensional uint8 tensor containing</span>
+<span class="sd">            the raw bytes of the JPEG image. This tensor must be on CPU,</span>
+<span class="sd">            regardless of the ``device`` parameter.</span>
+<span class="sd">        mode (ImageReadMode): the read mode used for optionally</span>
+<span class="sd">            converting the image. The supported modes are: ``ImageReadMode.UNCHANGED``,</span>
+<span class="sd">            ``ImageReadMode.GRAY`` and ``ImageReadMode.RGB``</span>
+<span class="sd">            Default: ``ImageReadMode.UNCHANGED``.</span>
+<span class="sd">            See ``ImageReadMode`` class for more information on various</span>
+<span class="sd">            available modes.</span>
+<span class="sd">        device (str or torch.device): The device on which the decoded image will</span>
+<span class="sd">            be stored. If a cuda device is specified, the image will be decoded</span>
+<span class="sd">            with `nvjpeg &lt;https://developer.nvidia.com/nvjpeg&gt;`_. This is only</span>
+<span class="sd">            supported for CUDA version &gt;= 10.1</span>
+
+<span class="sd">            .. betastatus:: device parameter</span>
+
+<span class="sd">            .. warning::</span>
+<span class="sd">                There is a memory leak in the nvjpeg library for CUDA versions &lt; 11.6.</span>
+<span class="sd">                Make sure to rely on CUDA 11.6 or above before using ``device=&quot;cuda&quot;``.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        output (Tensor[image_channels, image_height, image_width])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">decode_jpeg</span><span class="p">)</span>
+    <span class="n">device</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">device</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s2">&quot;cuda&quot;</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">decode_jpeg_cuda</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">mode</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">device</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">decode_jpeg</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">mode</span><span class="o">.</span><span class="n">value</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="encode_jpeg"><a class="viewcode-back" href="../../../generated/torchvision.io.encode_jpeg.html#torchvision.io.encode_jpeg">[docs]</a><span class="k">def</span> <span class="nf">encode_jpeg</span><span class="p">(</span><span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">quality</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">75</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Takes an input tensor in CHW layout and returns a buffer with the contents</span>
+<span class="sd">    of its corresponding JPEG file.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[channels, image_height, image_width])): int8 image tensor of</span>
+<span class="sd">            ``c`` channels, where ``c`` must be 1 or 3.</span>
+<span class="sd">        quality (int): Quality of the resulting JPEG file, it must be a number between</span>
+<span class="sd">            1 and 100. Default: 75</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        output (Tensor[1]): A one dimensional int8 tensor that contains the raw bytes of the</span>
+<span class="sd">            JPEG file.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">encode_jpeg</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">quality</span> <span class="o">&lt;</span> <span class="mi">1</span> <span class="ow">or</span> <span class="n">quality</span> <span class="o">&gt;</span> <span class="mi">100</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Image quality should be a positive number between 1 and 100&quot;</span><span class="p">)</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">encode_jpeg</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">quality</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="write_jpeg"><a class="viewcode-back" href="../../../generated/torchvision.io.write_jpeg.html#torchvision.io.write_jpeg">[docs]</a><span class="k">def</span> <span class="nf">write_jpeg</span><span class="p">(</span><span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">quality</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">75</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Takes an input tensor in CHW layout and saves it in a JPEG file.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[channels, image_height, image_width]): int8 image tensor of ``c``</span>
+<span class="sd">            channels, where ``c`` must be 1 or 3.</span>
+<span class="sd">        filename (str): Path to save the image.</span>
+<span class="sd">        quality (int): Quality of the resulting JPEG file, it must be a number</span>
+<span class="sd">            between 1 and 100. Default: 75</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">write_jpeg</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">encode_jpeg</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">quality</span><span class="p">)</span>
+    <span class="n">write_file</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">output</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="decode_image"><a class="viewcode-back" href="../../../generated/torchvision.io.decode_image.html#torchvision.io.decode_image">[docs]</a><span class="k">def</span> <span class="nf">decode_image</span><span class="p">(</span><span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="n">ImageReadMode</span> <span class="o">=</span> <span class="n">ImageReadMode</span><span class="o">.</span><span class="n">UNCHANGED</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Detects whether an image is a JPEG or PNG and performs the appropriate</span>
+<span class="sd">    operation to decode the image into a 3 dimensional RGB or grayscale Tensor.</span>
+
+<span class="sd">    Optionally converts the image to the desired format.</span>
+<span class="sd">    The values of the output tensor are uint8 in [0, 255].</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor): a one dimensional uint8 tensor containing the raw bytes of the</span>
+<span class="sd">            PNG or JPEG image.</span>
+<span class="sd">        mode (ImageReadMode): the read mode used for optionally converting the image.</span>
+<span class="sd">            Default: ``ImageReadMode.UNCHANGED``.</span>
+<span class="sd">            See ``ImageReadMode`` class for more information on various</span>
+<span class="sd">            available modes.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        output (Tensor[image_channels, image_height, image_width])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">decode_image</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">decode_image</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">mode</span><span class="o">.</span><span class="n">value</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="read_image"><a class="viewcode-back" href="../../../generated/torchvision.io.read_image.html#torchvision.io.read_image">[docs]</a><span class="k">def</span> <span class="nf">read_image</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="n">ImageReadMode</span> <span class="o">=</span> <span class="n">ImageReadMode</span><span class="o">.</span><span class="n">UNCHANGED</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Reads a JPEG or PNG image into a 3 dimensional RGB or grayscale Tensor.</span>
+<span class="sd">    Optionally converts the image to the desired format.</span>
+<span class="sd">    The values of the output tensor are uint8 in [0, 255].</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        path (str): path of the JPEG or PNG image.</span>
+<span class="sd">        mode (ImageReadMode): the read mode used for optionally converting the image.</span>
+<span class="sd">            Default: ``ImageReadMode.UNCHANGED``.</span>
+<span class="sd">            See ``ImageReadMode`` class for more information on various</span>
+<span class="sd">            available modes.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        output (Tensor[image_channels, image_height, image_width])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">read_image</span><span class="p">)</span>
+    <span class="n">data</span> <span class="o">=</span> <span class="n">read_file</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">decode_image</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">mode</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_read_png_16</span><span class="p">(</span><span class="n">path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="n">ImageReadMode</span> <span class="o">=</span> <span class="n">ImageReadMode</span><span class="o">.</span><span class="n">UNCHANGED</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">data</span> <span class="o">=</span> <span class="n">read_file</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">decode_png</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">mode</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/io/video.html b/0.16/_modules/torchvision/io/video.html
new file mode 100644
index 00000000000..ca8c5f40c56
--- /dev/null
+++ b/0.16/_modules/torchvision/io/video.html
@@ -0,0 +1,1084 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.io.video &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.io.video</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.io.video</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">gc</span>
+<span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">re</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">fractions</span> <span class="kn">import</span> <span class="n">Fraction</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">_video_opt</span>
+
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">import</span> <span class="nn">av</span>
+
+    <span class="n">av</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">set_level</span><span class="p">(</span><span class="n">av</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">ERROR</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">av</span><span class="o">.</span><span class="n">video</span><span class="o">.</span><span class="n">frame</span><span class="o">.</span><span class="n">VideoFrame</span><span class="p">,</span> <span class="s2">&quot;pict_type&quot;</span><span class="p">):</span>
+        <span class="n">av</span> <span class="o">=</span> <span class="ne">ImportError</span><span class="p">(</span>
+<span class="w">            </span><span class="sd">&quot;&quot;&quot;\</span>
+<span class="sd">Your version of PyAV is too old for the necessary video operations in torchvision.</span>
+<span class="sd">If you are on Python 3.5, you will have to build from source (the conda-forge</span>
+<span class="sd">packages are not up-to-date).  See</span>
+<span class="sd">https://github.com/mikeboers/PyAV#installation for instructions on how to</span>
+<span class="sd">install PyAV on your system.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+        <span class="p">)</span>
+<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+    <span class="n">av</span> <span class="o">=</span> <span class="ne">ImportError</span><span class="p">(</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;\</span>
+<span class="sd">PyAV is not installed, and is necessary for the video operations in torchvision.</span>
+<span class="sd">See https://github.com/mikeboers/PyAV#installation for instructions on how to</span>
+<span class="sd">install PyAV on your system.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_check_av_available</span><span class="p">()</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">av</span><span class="p">,</span> <span class="ne">Exception</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="n">av</span>
+
+
+<span class="k">def</span> <span class="nf">_av_available</span><span class="p">()</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+    <span class="k">return</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">av</span><span class="p">,</span> <span class="ne">Exception</span><span class="p">)</span>
+
+
+<span class="c1"># PyAV has some reference cycles</span>
+<span class="n">_CALLED_TIMES</span> <span class="o">=</span> <span class="mi">0</span>
+<span class="n">_GC_COLLECTION_INTERVAL</span> <span class="o">=</span> <span class="mi">10</span>
+
+
+<div class="viewcode-block" id="write_video"><a class="viewcode-back" href="../../../generated/torchvision.io.write_video.html#torchvision.io.write_video">[docs]</a><span class="k">def</span> <span class="nf">write_video</span><span class="p">(</span>
+    <span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">video_array</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">fps</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">video_codec</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;libx264&quot;</span><span class="p">,</span>
+    <span class="n">options</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">audio_array</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">audio_fps</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">audio_codec</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">audio_options</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Writes a 4d tensor in [T, H, W, C] format in a video file</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        filename (str): path where the video will be saved</span>
+<span class="sd">        video_array (Tensor[T, H, W, C]): tensor containing the individual frames,</span>
+<span class="sd">            as a uint8 tensor in [T, H, W, C] format</span>
+<span class="sd">        fps (Number): video frames per second</span>
+<span class="sd">        video_codec (str): the name of the video codec, i.e. &quot;libx264&quot;, &quot;h264&quot;, etc.</span>
+<span class="sd">        options (Dict): dictionary containing options to be passed into the PyAV video stream</span>
+<span class="sd">        audio_array (Tensor[C, N]): tensor containing the audio, where C is the number of channels</span>
+<span class="sd">            and N is the number of samples</span>
+<span class="sd">        audio_fps (Number): audio sample rate, typically 44100 or 48000</span>
+<span class="sd">        audio_codec (str): the name of the audio codec, i.e. &quot;mp3&quot;, &quot;aac&quot;, etc.</span>
+<span class="sd">        audio_options (Dict): dictionary containing options to be passed into the PyAV audio stream</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">write_video</span><span class="p">)</span>
+    <span class="n">_check_av_available</span><span class="p">()</span>
+    <span class="n">video_array</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">video_array</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
+
+    <span class="c1"># PyAV does not support floating point numbers with decimal point</span>
+    <span class="c1"># and will throw OverflowException in case this is not the case</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fps</span><span class="p">,</span> <span class="nb">float</span><span class="p">):</span>
+        <span class="n">fps</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">round</span><span class="p">(</span><span class="n">fps</span><span class="p">)</span>
+
+    <span class="k">with</span> <span class="n">av</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;w&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">container</span><span class="p">:</span>
+        <span class="n">stream</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">add_stream</span><span class="p">(</span><span class="n">video_codec</span><span class="p">,</span> <span class="n">rate</span><span class="o">=</span><span class="n">fps</span><span class="p">)</span>
+        <span class="n">stream</span><span class="o">.</span><span class="n">width</span> <span class="o">=</span> <span class="n">video_array</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+        <span class="n">stream</span><span class="o">.</span><span class="n">height</span> <span class="o">=</span> <span class="n">video_array</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">stream</span><span class="o">.</span><span class="n">pix_fmt</span> <span class="o">=</span> <span class="s2">&quot;yuv420p&quot;</span> <span class="k">if</span> <span class="n">video_codec</span> <span class="o">!=</span> <span class="s2">&quot;libx264rgb&quot;</span> <span class="k">else</span> <span class="s2">&quot;rgb24&quot;</span>
+        <span class="n">stream</span><span class="o">.</span><span class="n">options</span> <span class="o">=</span> <span class="n">options</span> <span class="ow">or</span> <span class="p">{}</span>
+
+        <span class="k">if</span> <span class="n">audio_array</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">audio_format_dtypes</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;dbl&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;f8&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;dblp&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;f8&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;flt&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;f4&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;fltp&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;f4&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;s16&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;i2&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;s16p&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;i2&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;s32&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;i4&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;s32p&quot;</span><span class="p">:</span> <span class="s2">&quot;&lt;i4&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;u8&quot;</span><span class="p">:</span> <span class="s2">&quot;u1&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;u8p&quot;</span><span class="p">:</span> <span class="s2">&quot;u1&quot;</span><span class="p">,</span>
+            <span class="p">}</span>
+            <span class="n">a_stream</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">add_stream</span><span class="p">(</span><span class="n">audio_codec</span><span class="p">,</span> <span class="n">rate</span><span class="o">=</span><span class="n">audio_fps</span><span class="p">)</span>
+            <span class="n">a_stream</span><span class="o">.</span><span class="n">options</span> <span class="o">=</span> <span class="n">audio_options</span> <span class="ow">or</span> <span class="p">{}</span>
+
+            <span class="n">num_channels</span> <span class="o">=</span> <span class="n">audio_array</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">audio_layout</span> <span class="o">=</span> <span class="s2">&quot;stereo&quot;</span> <span class="k">if</span> <span class="n">num_channels</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="s2">&quot;mono&quot;</span>
+            <span class="n">audio_sample_fmt</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">audio</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">format</span><span class="o">.</span><span class="n">name</span>
+
+            <span class="n">format_dtype</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">dtype</span><span class="p">(</span><span class="n">audio_format_dtypes</span><span class="p">[</span><span class="n">audio_sample_fmt</span><span class="p">])</span>
+            <span class="n">audio_array</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">audio_array</span><span class="p">)</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">format_dtype</span><span class="p">)</span>
+
+            <span class="n">frame</span> <span class="o">=</span> <span class="n">av</span><span class="o">.</span><span class="n">AudioFrame</span><span class="o">.</span><span class="n">from_ndarray</span><span class="p">(</span><span class="n">audio_array</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="n">audio_sample_fmt</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="n">audio_layout</span><span class="p">)</span>
+
+            <span class="n">frame</span><span class="o">.</span><span class="n">sample_rate</span> <span class="o">=</span> <span class="n">audio_fps</span>
+
+            <span class="k">for</span> <span class="n">packet</span> <span class="ow">in</span> <span class="n">a_stream</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">frame</span><span class="p">):</span>
+                <span class="n">container</span><span class="o">.</span><span class="n">mux</span><span class="p">(</span><span class="n">packet</span><span class="p">)</span>
+
+            <span class="k">for</span> <span class="n">packet</span> <span class="ow">in</span> <span class="n">a_stream</span><span class="o">.</span><span class="n">encode</span><span class="p">():</span>
+                <span class="n">container</span><span class="o">.</span><span class="n">mux</span><span class="p">(</span><span class="n">packet</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">img</span> <span class="ow">in</span> <span class="n">video_array</span><span class="p">:</span>
+            <span class="n">frame</span> <span class="o">=</span> <span class="n">av</span><span class="o">.</span><span class="n">VideoFrame</span><span class="o">.</span><span class="n">from_ndarray</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;rgb24&quot;</span><span class="p">)</span>
+            <span class="n">frame</span><span class="o">.</span><span class="n">pict_type</span> <span class="o">=</span> <span class="s2">&quot;NONE&quot;</span>
+            <span class="k">for</span> <span class="n">packet</span> <span class="ow">in</span> <span class="n">stream</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">frame</span><span class="p">):</span>
+                <span class="n">container</span><span class="o">.</span><span class="n">mux</span><span class="p">(</span><span class="n">packet</span><span class="p">)</span>
+
+        <span class="c1"># Flush stream</span>
+        <span class="k">for</span> <span class="n">packet</span> <span class="ow">in</span> <span class="n">stream</span><span class="o">.</span><span class="n">encode</span><span class="p">():</span>
+            <span class="n">container</span><span class="o">.</span><span class="n">mux</span><span class="p">(</span><span class="n">packet</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_read_from_stream</span><span class="p">(</span>
+    <span class="n">container</span><span class="p">:</span> <span class="s2">&quot;av.container.Container&quot;</span><span class="p">,</span>
+    <span class="n">start_offset</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">end_offset</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">pts_unit</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">stream</span><span class="p">:</span> <span class="s2">&quot;av.stream.Stream&quot;</span><span class="p">,</span>
+    <span class="n">stream_name</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="o">...</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="s2">&quot;av.frame.Frame&quot;</span><span class="p">]:</span>
+    <span class="k">global</span> <span class="n">_CALLED_TIMES</span><span class="p">,</span> <span class="n">_GC_COLLECTION_INTERVAL</span>
+    <span class="n">_CALLED_TIMES</span> <span class="o">+=</span> <span class="mi">1</span>
+    <span class="k">if</span> <span class="n">_CALLED_TIMES</span> <span class="o">%</span> <span class="n">_GC_COLLECTION_INTERVAL</span> <span class="o">==</span> <span class="n">_GC_COLLECTION_INTERVAL</span> <span class="o">-</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">gc</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
+
+    <span class="k">if</span> <span class="n">pts_unit</span> <span class="o">==</span> <span class="s2">&quot;sec&quot;</span><span class="p">:</span>
+        <span class="c1"># TODO: we should change all of this from ground up to simply take</span>
+        <span class="c1"># sec and convert to MS in C++</span>
+        <span class="n">start_offset</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="n">start_offset</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">/</span> <span class="n">stream</span><span class="o">.</span><span class="n">time_base</span><span class="p">)))</span>
+        <span class="k">if</span> <span class="n">end_offset</span> <span class="o">!=</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">):</span>
+            <span class="n">end_offset</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">end_offset</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">/</span> <span class="n">stream</span><span class="o">.</span><span class="n">time_base</span><span class="p">)))</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;The pts_unit &#39;pts&#39; gives wrong results. Please use pts_unit &#39;sec&#39;.&quot;</span><span class="p">)</span>
+
+    <span class="n">frames</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="n">should_buffer</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="n">max_buffer_size</span> <span class="o">=</span> <span class="mi">5</span>
+    <span class="k">if</span> <span class="n">stream</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s2">&quot;video&quot;</span><span class="p">:</span>
+        <span class="c1"># DivX-style packed B-frames can have out-of-order pts (2 frames in a single pkt)</span>
+        <span class="c1"># so need to buffer some extra frames to sort everything</span>
+        <span class="c1"># properly</span>
+        <span class="n">extradata</span> <span class="o">=</span> <span class="n">stream</span><span class="o">.</span><span class="n">codec_context</span><span class="o">.</span><span class="n">extradata</span>
+        <span class="c1"># overly complicated way of finding if `divx_packed` is set, following</span>
+        <span class="c1"># https://github.com/FFmpeg/FFmpeg/commit/d5a21172283572af587b3d939eba0091484d3263</span>
+        <span class="k">if</span> <span class="n">extradata</span> <span class="ow">and</span> <span class="sa">b</span><span class="s2">&quot;DivX&quot;</span> <span class="ow">in</span> <span class="n">extradata</span><span class="p">:</span>
+            <span class="c1"># can&#39;t use regex directly because of some weird characters sometimes...</span>
+            <span class="n">pos</span> <span class="o">=</span> <span class="n">extradata</span><span class="o">.</span><span class="n">find</span><span class="p">(</span><span class="sa">b</span><span class="s2">&quot;DivX&quot;</span><span class="p">)</span>
+            <span class="n">d</span> <span class="o">=</span> <span class="n">extradata</span><span class="p">[</span><span class="n">pos</span><span class="p">:]</span>
+            <span class="n">o</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="sa">rb</span><span class="s2">&quot;DivX(\d+)Build(\d+)(\w)&quot;</span><span class="p">,</span> <span class="n">d</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">o</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">o</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="sa">rb</span><span class="s2">&quot;DivX(\d+)b(\d+)(\w)&quot;</span><span class="p">,</span> <span class="n">d</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">o</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">should_buffer</span> <span class="o">=</span> <span class="n">o</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">3</span><span class="p">)</span> <span class="o">==</span> <span class="sa">b</span><span class="s2">&quot;p&quot;</span>
+    <span class="n">seek_offset</span> <span class="o">=</span> <span class="n">start_offset</span>
+    <span class="c1"># some files don&#39;t seek to the right location, so better be safe here</span>
+    <span class="n">seek_offset</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">seek_offset</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">should_buffer</span><span class="p">:</span>
+        <span class="c1"># FIXME this is kind of a hack, but we will jump to the previous keyframe</span>
+        <span class="c1"># so this will be safe</span>
+        <span class="n">seek_offset</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">seek_offset</span> <span class="o">-</span> <span class="n">max_buffer_size</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+    <span class="k">try</span><span class="p">:</span>
+        <span class="c1"># TODO check if stream needs to always be the video stream here or not</span>
+        <span class="n">container</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="n">seek_offset</span><span class="p">,</span> <span class="n">any_frame</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">backward</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">stream</span><span class="o">=</span><span class="n">stream</span><span class="p">)</span>
+    <span class="k">except</span> <span class="n">av</span><span class="o">.</span><span class="n">AVError</span><span class="p">:</span>
+        <span class="c1"># TODO add some warnings in this case</span>
+        <span class="c1"># print(&quot;Corrupted file?&quot;, container.name)</span>
+        <span class="k">return</span> <span class="p">[]</span>
+    <span class="n">buffer_count</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="k">try</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">_idx</span><span class="p">,</span> <span class="n">frame</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">container</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="o">**</span><span class="n">stream_name</span><span class="p">)):</span>
+            <span class="n">frames</span><span class="p">[</span><span class="n">frame</span><span class="o">.</span><span class="n">pts</span><span class="p">]</span> <span class="o">=</span> <span class="n">frame</span>
+            <span class="k">if</span> <span class="n">frame</span><span class="o">.</span><span class="n">pts</span> <span class="o">&gt;=</span> <span class="n">end_offset</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">should_buffer</span> <span class="ow">and</span> <span class="n">buffer_count</span> <span class="o">&lt;</span> <span class="n">max_buffer_size</span><span class="p">:</span>
+                    <span class="n">buffer_count</span> <span class="o">+=</span> <span class="mi">1</span>
+                    <span class="k">continue</span>
+                <span class="k">break</span>
+    <span class="k">except</span> <span class="n">av</span><span class="o">.</span><span class="n">AVError</span><span class="p">:</span>
+        <span class="c1"># TODO add a warning</span>
+        <span class="k">pass</span>
+    <span class="c1"># ensure that the results are sorted wrt the pts</span>
+    <span class="n">result</span> <span class="o">=</span> <span class="p">[</span><span class="n">frames</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">frames</span><span class="p">)</span> <span class="k">if</span> <span class="n">start_offset</span> <span class="o">&lt;=</span> <span class="n">frames</span><span class="p">[</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">pts</span> <span class="o">&lt;=</span> <span class="n">end_offset</span><span class="p">]</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">frames</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">start_offset</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">start_offset</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">frames</span><span class="p">:</span>
+        <span class="c1"># if there is no frame that exactly matches the pts of start_offset</span>
+        <span class="c1"># add the last frame smaller than start_offset, to guarantee that</span>
+        <span class="c1"># we will have all the necessary data. This is most useful for audio</span>
+        <span class="n">preceding_frames</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">frames</span> <span class="k">if</span> <span class="n">i</span> <span class="o">&lt;</span> <span class="n">start_offset</span><span class="p">]</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">preceding_frames</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">first_frame_pts</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">preceding_frames</span><span class="p">)</span>
+            <span class="n">result</span><span class="o">.</span><span class="n">insert</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">frames</span><span class="p">[</span><span class="n">first_frame_pts</span><span class="p">])</span>
+    <span class="k">return</span> <span class="n">result</span>
+
+
+<span class="k">def</span> <span class="nf">_align_audio_frames</span><span class="p">(</span>
+    <span class="n">aframes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">audio_frames</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="s2">&quot;av.frame.Frame&quot;</span><span class="p">],</span> <span class="n">ref_start</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">ref_end</span><span class="p">:</span> <span class="nb">float</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">start</span><span class="p">,</span> <span class="n">end</span> <span class="o">=</span> <span class="n">audio_frames</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">pts</span><span class="p">,</span> <span class="n">audio_frames</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">pts</span>
+    <span class="n">total_aframes</span> <span class="o">=</span> <span class="n">aframes</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+    <span class="n">step_per_aframe</span> <span class="o">=</span> <span class="p">(</span><span class="n">end</span> <span class="o">-</span> <span class="n">start</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">total_aframes</span>
+    <span class="n">s_idx</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">e_idx</span> <span class="o">=</span> <span class="n">total_aframes</span>
+    <span class="k">if</span> <span class="n">start</span> <span class="o">&lt;</span> <span class="n">ref_start</span><span class="p">:</span>
+        <span class="n">s_idx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">((</span><span class="n">ref_start</span> <span class="o">-</span> <span class="n">start</span><span class="p">)</span> <span class="o">/</span> <span class="n">step_per_aframe</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">end</span> <span class="o">&gt;</span> <span class="n">ref_end</span><span class="p">:</span>
+        <span class="n">e_idx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">((</span><span class="n">ref_end</span> <span class="o">-</span> <span class="n">end</span><span class="p">)</span> <span class="o">/</span> <span class="n">step_per_aframe</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">aframes</span><span class="p">[:,</span> <span class="n">s_idx</span><span class="p">:</span><span class="n">e_idx</span><span class="p">]</span>
+
+
+<div class="viewcode-block" id="read_video"><a class="viewcode-back" href="../../../generated/torchvision.io.read_video.html#torchvision.io.read_video">[docs]</a><span class="k">def</span> <span class="nf">read_video</span><span class="p">(</span>
+    <span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">start_pts</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Fraction</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="n">end_pts</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Fraction</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">pts_unit</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;pts&quot;</span><span class="p">,</span>
+    <span class="n">output_format</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;THWC&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Reads a video from a file, returning both the video frames and the audio frames</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        filename (str): path to the video file</span>
+<span class="sd">        start_pts (int if pts_unit = &#39;pts&#39;, float / Fraction if pts_unit = &#39;sec&#39;, optional):</span>
+<span class="sd">            The start presentation time of the video</span>
+<span class="sd">        end_pts (int if pts_unit = &#39;pts&#39;, float / Fraction if pts_unit = &#39;sec&#39;, optional):</span>
+<span class="sd">            The end presentation time</span>
+<span class="sd">        pts_unit (str, optional): unit in which start_pts and end_pts values will be interpreted,</span>
+<span class="sd">            either &#39;pts&#39; or &#39;sec&#39;. Defaults to &#39;pts&#39;.</span>
+<span class="sd">        output_format (str, optional): The format of the output video tensors. Can be either &quot;THWC&quot; (default) or &quot;TCHW&quot;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        vframes (Tensor[T, H, W, C] or Tensor[T, C, H, W]): the `T` video frames</span>
+<span class="sd">        aframes (Tensor[K, L]): the audio frames, where `K` is the number of channels and `L` is the number of points</span>
+<span class="sd">        info (Dict): metadata for the video and audio. Can contain the fields video_fps (float) and audio_fps (int)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">read_video</span><span class="p">)</span>
+
+    <span class="n">output_format</span> <span class="o">=</span> <span class="n">output_format</span><span class="o">.</span><span class="n">upper</span><span class="p">()</span>
+    <span class="k">if</span> <span class="n">output_format</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;THWC&quot;</span><span class="p">,</span> <span class="s2">&quot;TCHW&quot;</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;output_format should be either &#39;THWC&#39; or &#39;TCHW&#39;, got </span><span class="si">{</span><span class="n">output_format</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">get_video_backend</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">filename</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;File not found: </span><span class="si">{</span><span class="n">filename</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">get_video_backend</span><span class="p">()</span> <span class="o">!=</span> <span class="s2">&quot;pyav&quot;</span><span class="p">:</span>
+        <span class="n">vframes</span><span class="p">,</span> <span class="n">aframes</span><span class="p">,</span> <span class="n">info</span> <span class="o">=</span> <span class="n">_video_opt</span><span class="o">.</span><span class="n">_read_video</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">start_pts</span><span class="p">,</span> <span class="n">end_pts</span><span class="p">,</span> <span class="n">pts_unit</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">_check_av_available</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">end_pts</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">end_pts</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">end_pts</span> <span class="o">&lt;</span> <span class="n">start_pts</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;end_pts should be larger than start_pts, got start_pts=</span><span class="si">{</span><span class="n">start_pts</span><span class="si">}</span><span class="s2"> and end_pts=</span><span class="si">{</span><span class="n">end_pts</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">info</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">video_frames</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">audio_frames</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">audio_timebase</span> <span class="o">=</span> <span class="n">_video_opt</span><span class="o">.</span><span class="n">default_timebase</span>
+
+        <span class="k">try</span><span class="p">:</span>
+            <span class="k">with</span> <span class="n">av</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">metadata_errors</span><span class="o">=</span><span class="s2">&quot;ignore&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">container</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">audio</span><span class="p">:</span>
+                    <span class="n">audio_timebase</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">audio</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">time_base</span>
+                <span class="k">if</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">video</span><span class="p">:</span>
+                    <span class="n">video_frames</span> <span class="o">=</span> <span class="n">_read_from_stream</span><span class="p">(</span>
+                        <span class="n">container</span><span class="p">,</span>
+                        <span class="n">start_pts</span><span class="p">,</span>
+                        <span class="n">end_pts</span><span class="p">,</span>
+                        <span class="n">pts_unit</span><span class="p">,</span>
+                        <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">video</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+                        <span class="p">{</span><span class="s2">&quot;video&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">},</span>
+                    <span class="p">)</span>
+                    <span class="n">video_fps</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">video</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">average_rate</span>
+                    <span class="c1"># guard against potentially corrupted files</span>
+                    <span class="k">if</span> <span class="n">video_fps</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">info</span><span class="p">[</span><span class="s2">&quot;video_fps&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">video_fps</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">audio</span><span class="p">:</span>
+                    <span class="n">audio_frames</span> <span class="o">=</span> <span class="n">_read_from_stream</span><span class="p">(</span>
+                        <span class="n">container</span><span class="p">,</span>
+                        <span class="n">start_pts</span><span class="p">,</span>
+                        <span class="n">end_pts</span><span class="p">,</span>
+                        <span class="n">pts_unit</span><span class="p">,</span>
+                        <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">audio</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+                        <span class="p">{</span><span class="s2">&quot;audio&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">},</span>
+                    <span class="p">)</span>
+                    <span class="n">info</span><span class="p">[</span><span class="s2">&quot;audio_fps&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">audio</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">rate</span>
+
+        <span class="k">except</span> <span class="n">av</span><span class="o">.</span><span class="n">AVError</span><span class="p">:</span>
+            <span class="c1"># TODO raise a warning?</span>
+            <span class="k">pass</span>
+
+        <span class="n">vframes_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">frame</span><span class="o">.</span><span class="n">to_rgb</span><span class="p">()</span><span class="o">.</span><span class="n">to_ndarray</span><span class="p">()</span> <span class="k">for</span> <span class="n">frame</span> <span class="ow">in</span> <span class="n">video_frames</span><span class="p">]</span>
+        <span class="n">aframes_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">frame</span><span class="o">.</span><span class="n">to_ndarray</span><span class="p">()</span> <span class="k">for</span> <span class="n">frame</span> <span class="ow">in</span> <span class="n">audio_frames</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">vframes_list</span><span class="p">:</span>
+            <span class="n">vframes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">vframes_list</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">vframes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">aframes_list</span><span class="p">:</span>
+            <span class="n">aframes</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">(</span><span class="n">aframes_list</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="n">aframes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">aframes</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">pts_unit</span> <span class="o">==</span> <span class="s2">&quot;sec&quot;</span><span class="p">:</span>
+                <span class="n">start_pts</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="n">start_pts</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">/</span> <span class="n">audio_timebase</span><span class="p">)))</span>
+                <span class="k">if</span> <span class="n">end_pts</span> <span class="o">!=</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">):</span>
+                    <span class="n">end_pts</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">end_pts</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">/</span> <span class="n">audio_timebase</span><span class="p">)))</span>
+            <span class="n">aframes</span> <span class="o">=</span> <span class="n">_align_audio_frames</span><span class="p">(</span><span class="n">aframes</span><span class="p">,</span> <span class="n">audio_frames</span><span class="p">,</span> <span class="n">start_pts</span><span class="p">,</span> <span class="n">end_pts</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">aframes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">output_format</span> <span class="o">==</span> <span class="s2">&quot;TCHW&quot;</span><span class="p">:</span>
+        <span class="c1"># [T,H,W,C] --&gt; [T,C,H,W]</span>
+        <span class="n">vframes</span> <span class="o">=</span> <span class="n">vframes</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">vframes</span><span class="p">,</span> <span class="n">aframes</span><span class="p">,</span> <span class="n">info</span></div>
+
+
+<span class="k">def</span> <span class="nf">_can_read_timestamps_from_packets</span><span class="p">(</span><span class="n">container</span><span class="p">:</span> <span class="s2">&quot;av.container.Container&quot;</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+    <span class="n">extradata</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">codec_context</span><span class="o">.</span><span class="n">extradata</span>
+    <span class="k">if</span> <span class="n">extradata</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">return</span> <span class="kc">False</span>
+    <span class="k">if</span> <span class="sa">b</span><span class="s2">&quot;Lavc&quot;</span> <span class="ow">in</span> <span class="n">extradata</span><span class="p">:</span>
+        <span class="k">return</span> <span class="kc">True</span>
+    <span class="k">return</span> <span class="kc">False</span>
+
+
+<span class="k">def</span> <span class="nf">_decode_video_timestamps</span><span class="p">(</span><span class="n">container</span><span class="p">:</span> <span class="s2">&quot;av.container.Container&quot;</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="n">_can_read_timestamps_from_packets</span><span class="p">(</span><span class="n">container</span><span class="p">):</span>
+        <span class="c1"># fast path</span>
+        <span class="k">return</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">pts</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">container</span><span class="o">.</span><span class="n">demux</span><span class="p">(</span><span class="n">video</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span> <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">pts</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">pts</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">container</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">video</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span> <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">pts</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">]</span>
+
+
+<div class="viewcode-block" id="read_video_timestamps"><a class="viewcode-back" href="../../../generated/torchvision.io.read_video_timestamps.html#torchvision.io.read_video_timestamps">[docs]</a><span class="k">def</span> <span class="nf">read_video_timestamps</span><span class="p">(</span><span class="n">filename</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">pts_unit</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;pts&quot;</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    List the video frames timestamps.</span>
+
+<span class="sd">    Note that the function decodes the whole video frame-by-frame.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        filename (str): path to the video file</span>
+<span class="sd">        pts_unit (str, optional): unit in which timestamp values will be returned</span>
+<span class="sd">            either &#39;pts&#39; or &#39;sec&#39;. Defaults to &#39;pts&#39;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        pts (List[int] if pts_unit = &#39;pts&#39;, List[Fraction] if pts_unit = &#39;sec&#39;):</span>
+<span class="sd">            presentation timestamps for each one of the frames in the video.</span>
+<span class="sd">        video_fps (float, optional): the frame rate for the video</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">read_video_timestamps</span><span class="p">)</span>
+    <span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">get_video_backend</span>
+
+    <span class="k">if</span> <span class="n">get_video_backend</span><span class="p">()</span> <span class="o">!=</span> <span class="s2">&quot;pyav&quot;</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_video_opt</span><span class="o">.</span><span class="n">_read_video_timestamps</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">pts_unit</span><span class="p">)</span>
+
+    <span class="n">_check_av_available</span><span class="p">()</span>
+
+    <span class="n">video_fps</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="n">pts</span> <span class="o">=</span> <span class="p">[]</span>
+
+    <span class="k">try</span><span class="p">:</span>
+        <span class="k">with</span> <span class="n">av</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">metadata_errors</span><span class="o">=</span><span class="s2">&quot;ignore&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">container</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">video</span><span class="p">:</span>
+                <span class="n">video_stream</span> <span class="o">=</span> <span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">video</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+                <span class="n">video_time_base</span> <span class="o">=</span> <span class="n">video_stream</span><span class="o">.</span><span class="n">time_base</span>
+                <span class="k">try</span><span class="p">:</span>
+                    <span class="n">pts</span> <span class="o">=</span> <span class="n">_decode_video_timestamps</span><span class="p">(</span><span class="n">container</span><span class="p">)</span>
+                <span class="k">except</span> <span class="n">av</span><span class="o">.</span><span class="n">AVError</span><span class="p">:</span>
+                    <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Failed decoding frames for file </span><span class="si">{</span><span class="n">filename</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="n">video_fps</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">video_stream</span><span class="o">.</span><span class="n">average_rate</span><span class="p">)</span>
+    <span class="k">except</span> <span class="n">av</span><span class="o">.</span><span class="n">AVError</span> <span class="k">as</span> <span class="n">e</span><span class="p">:</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;Failed to open container for </span><span class="si">{</span><span class="n">filename</span><span class="si">}</span><span class="s2">; Caught error: </span><span class="si">{</span><span class="n">e</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="n">msg</span><span class="p">,</span> <span class="ne">RuntimeWarning</span><span class="p">)</span>
+
+    <span class="n">pts</span><span class="o">.</span><span class="n">sort</span><span class="p">()</span>
+
+    <span class="k">if</span> <span class="n">pts_unit</span> <span class="o">==</span> <span class="s2">&quot;sec&quot;</span><span class="p">:</span>
+        <span class="n">pts</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span> <span class="o">*</span> <span class="n">video_time_base</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">pts</span><span class="p">]</span>
+
+    <span class="k">return</span> <span class="n">pts</span><span class="p">,</span> <span class="n">video_fps</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/io/video_reader.html b/0.16/_modules/torchvision/io/video_reader.html
new file mode 100644
index 00000000000..5ee4254f7d8
--- /dev/null
+++ b/0.16/_modules/torchvision/io/video_reader.html
@@ -0,0 +1,967 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.io.video_reader &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.io.video_reader</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.io.video_reader</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">io</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Iterator</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">._video_opt</span> <span class="kn">import</span> <span class="n">_HAS_VIDEO_OPT</span>
+
+<span class="k">if</span> <span class="n">_HAS_VIDEO_OPT</span><span class="p">:</span>
+
+    <span class="k">def</span> <span class="nf">_has_video_opt</span><span class="p">()</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="kc">True</span>
+
+<span class="k">else</span><span class="p">:</span>
+
+    <span class="k">def</span> <span class="nf">_has_video_opt</span><span class="p">()</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">return</span> <span class="kc">False</span>
+
+
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">import</span> <span class="nn">av</span>
+
+    <span class="n">av</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">set_level</span><span class="p">(</span><span class="n">av</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">ERROR</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">av</span><span class="o">.</span><span class="n">video</span><span class="o">.</span><span class="n">frame</span><span class="o">.</span><span class="n">VideoFrame</span><span class="p">,</span> <span class="s2">&quot;pict_type&quot;</span><span class="p">):</span>
+        <span class="n">av</span> <span class="o">=</span> <span class="ne">ImportError</span><span class="p">(</span>
+<span class="w">            </span><span class="sd">&quot;&quot;&quot;\</span>
+<span class="sd">Your version of PyAV is too old for the necessary video operations in torchvision.</span>
+<span class="sd">If you are on Python 3.5, you will have to build from source (the conda-forge</span>
+<span class="sd">packages are not up-to-date).  See</span>
+<span class="sd">https://github.com/mikeboers/PyAV#installation for instructions on how to</span>
+<span class="sd">install PyAV on your system.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+        <span class="p">)</span>
+<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+    <span class="n">av</span> <span class="o">=</span> <span class="ne">ImportError</span><span class="p">(</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;\</span>
+<span class="sd">PyAV is not installed, and is necessary for the video operations in torchvision.</span>
+<span class="sd">See https://github.com/mikeboers/PyAV#installation for instructions on how to</span>
+<span class="sd">install PyAV on your system.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+    <span class="p">)</span>
+
+
+<div class="viewcode-block" id="VideoReader"><a class="viewcode-back" href="../../../generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader">[docs]</a><span class="k">class</span> <span class="nc">VideoReader</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Fine-grained video-reading API.</span>
+<span class="sd">    Supports frame-by-frame reading of various streams from a single video</span>
+<span class="sd">    container. Much like previous video_reader API it supports the following</span>
+<span class="sd">    backends: video_reader, pyav, and cuda.</span>
+<span class="sd">    Backends can be set via `torchvision.set_video_backend` function.</span>
+
+<span class="sd">    .. betastatus:: VideoReader class</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        The following examples creates a :mod:`VideoReader` object, seeks into 2s</span>
+<span class="sd">        point, and returns a single frame::</span>
+
+<span class="sd">            import torchvision</span>
+<span class="sd">            video_path = &quot;path_to_a_test_video&quot;</span>
+<span class="sd">            reader = torchvision.io.VideoReader(video_path, &quot;video&quot;)</span>
+<span class="sd">            reader.seek(2.0)</span>
+<span class="sd">            frame = next(reader)</span>
+
+<span class="sd">        :mod:`VideoReader` implements the iterable API, which makes it suitable to</span>
+<span class="sd">        using it in conjunction with :mod:`itertools` for more advanced reading.</span>
+<span class="sd">        As such, we can use a :mod:`VideoReader` instance inside for loops::</span>
+
+<span class="sd">            reader.seek(2)</span>
+<span class="sd">            for frame in reader:</span>
+<span class="sd">                frames.append(frame[&#39;data&#39;])</span>
+<span class="sd">            # additionally, `seek` implements a fluent API, so we can do</span>
+<span class="sd">            for frame in reader.seek(2):</span>
+<span class="sd">                frames.append(frame[&#39;data&#39;])</span>
+
+<span class="sd">        With :mod:`itertools`, we can read all frames between 2 and 5 seconds with the</span>
+<span class="sd">        following code::</span>
+
+<span class="sd">            for frame in itertools.takewhile(lambda x: x[&#39;pts&#39;] &lt;= 5, reader.seek(2)):</span>
+<span class="sd">                frames.append(frame[&#39;data&#39;])</span>
+
+<span class="sd">        and similarly, reading 10 frames after the 2s timestamp can be achieved</span>
+<span class="sd">        as follows::</span>
+
+<span class="sd">            for frame in itertools.islice(reader.seek(2), 10):</span>
+<span class="sd">                frames.append(frame[&#39;data&#39;])</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        Each stream descriptor consists of two parts: stream type (e.g. &#39;video&#39;) and</span>
+<span class="sd">        a unique stream id (which are determined by the video encoding).</span>
+<span class="sd">        In this way, if the video container contains multiple</span>
+<span class="sd">        streams of the same type, users can access the one they want.</span>
+<span class="sd">        If only stream type is passed, the decoder auto-detects first stream of that type.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        src (string, bytes object, or tensor): The media source.</span>
+<span class="sd">            If string-type, it must be a file path supported by FFMPEG.</span>
+<span class="sd">            If bytes, should be an in-memory representation of a file supported by FFMPEG.</span>
+<span class="sd">            If Tensor, it is interpreted internally as byte buffer.</span>
+<span class="sd">            It must be one-dimensional, of type ``torch.uint8``.</span>
+
+
+
+<span class="sd">        stream (string, optional): descriptor of the required stream, followed by the stream id,</span>
+<span class="sd">            in the format ``{stream_type}:{stream_id}``. Defaults to ``&quot;video:0&quot;``.</span>
+<span class="sd">            Currently available options include ``[&#39;video&#39;, &#39;audio&#39;]``</span>
+
+<span class="sd">        num_threads (int, optional): number of threads used by the codec to decode video.</span>
+<span class="sd">            Default value (0) enables multithreading with codec-dependent heuristic. The performance</span>
+<span class="sd">            will depend on the version of FFMPEG codecs supported.</span>
+
+
+<span class="sd">        path (str, optional):</span>
+<span class="sd">            .. warning:</span>
+<span class="sd">                This parameter was deprecated in ``0.15`` and will be removed in ``0.17``.</span>
+<span class="sd">                Please use ``src`` instead.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">src</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span><span class="p">,</span>
+        <span class="n">stream</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;video&quot;</span><span class="p">,</span>
+        <span class="n">num_threads</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="kn">from</span> <span class="nn">..</span> <span class="kn">import</span> <span class="n">get_video_backend</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">=</span> <span class="n">get_video_backend</span><span class="p">()</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">src</span> <span class="o">==</span> <span class="s2">&quot;&quot;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;src cannot be empty&quot;</span><span class="p">)</span>
+                <span class="n">src</span> <span class="o">=</span> <span class="n">path</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;path is deprecated and will be removed in 0.17. Please use src instead&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="nb">bytes</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;cuda&quot;</span><span class="p">]:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                    <span class="s2">&quot;VideoReader cannot be initialized from bytes object when using cuda or pyav backend.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;pyav&quot;</span><span class="p">:</span>
+                <span class="n">src</span> <span class="o">=</span> <span class="n">io</span><span class="o">.</span><span class="n">BytesIO</span><span class="p">(</span><span class="n">src</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">with</span> <span class="n">warnings</span><span class="o">.</span><span class="n">catch_warnings</span><span class="p">():</span>
+                    <span class="c1"># Ignore the warning because we actually don&#39;t modify the buffer in this function</span>
+                    <span class="n">warnings</span><span class="o">.</span><span class="n">filterwarnings</span><span class="p">(</span><span class="s2">&quot;ignore&quot;</span><span class="p">,</span> <span class="n">message</span><span class="o">=</span><span class="s2">&quot;The given buffer is not writable&quot;</span><span class="p">)</span>
+                    <span class="n">src</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">frombuffer</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;cuda&quot;</span><span class="p">,</span> <span class="s2">&quot;pyav&quot;</span><span class="p">]:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                    <span class="s2">&quot;VideoReader cannot be initialized from Tensor object when using cuda or pyav backend.&quot;</span>
+                <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;`src` must be either string, Tensor or bytes object.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;cuda&quot;</span><span class="p">:</span>
+            <span class="n">device</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="s2">&quot;cuda&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_c</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">classes</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">GPUDecoder</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">device</span><span class="p">)</span>
+
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;video_reader&quot;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_c</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">classes</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">Video</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">stream</span><span class="p">,</span> <span class="n">num_threads</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_c</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">classes</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">Video</span><span class="p">(</span><span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_c</span><span class="o">.</span><span class="n">init_from_memory</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">stream</span><span class="p">,</span> <span class="n">num_threads</span><span class="p">)</span>
+
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;pyav&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">container</span> <span class="o">=</span> <span class="n">av</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">src</span><span class="p">,</span> <span class="n">metadata_errors</span><span class="o">=</span><span class="s2">&quot;ignore&quot;</span><span class="p">)</span>
+            <span class="c1"># TODO: load metadata</span>
+            <span class="n">stream_type</span> <span class="o">=</span> <span class="n">stream</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">stream_id</span> <span class="o">=</span> <span class="mi">0</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">stream</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">))</span> <span class="o">==</span> <span class="mi">1</span> <span class="k">else</span> <span class="nb">int</span><span class="p">(</span><span class="n">stream</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">)[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span> <span class="o">=</span> <span class="p">{</span><span class="n">stream_type</span><span class="p">:</span> <span class="n">stream_id</span><span class="p">}</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_c</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">container</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span><span class="p">)</span>
+
+            <span class="c1"># TODO: add extradata exception</span>
+
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Unknown video backend: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="fm">__next__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Decodes and returns the next frame of the current stream.</span>
+<span class="sd">        Frames are encoded as a dict with mandatory</span>
+<span class="sd">        data and pts fields, where data is a tensor, and pts is a</span>
+<span class="sd">        presentation timestamp of the frame expressed in seconds</span>
+<span class="sd">        as a float.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            (dict): a dictionary and containing decoded frame (``data``)</span>
+<span class="sd">            and corresponding timestamp (``pts``) in seconds</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;cuda&quot;</span><span class="p">:</span>
+            <span class="n">frame</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_c</span><span class="o">.</span><span class="n">next</span><span class="p">()</span>
+            <span class="k">if</span> <span class="n">frame</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">StopIteration</span>
+            <span class="k">return</span> <span class="p">{</span><span class="s2">&quot;data&quot;</span><span class="p">:</span> <span class="n">frame</span><span class="p">,</span> <span class="s2">&quot;pts&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">}</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;video_reader&quot;</span><span class="p">:</span>
+            <span class="n">frame</span><span class="p">,</span> <span class="n">pts</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_c</span><span class="o">.</span><span class="n">next</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">frame</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_c</span><span class="p">)</span>
+                <span class="n">pts</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">frame</span><span class="o">.</span><span class="n">pts</span> <span class="o">*</span> <span class="n">frame</span><span class="o">.</span><span class="n">time_base</span><span class="p">)</span>
+                <span class="k">if</span> <span class="s2">&quot;video&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span><span class="p">:</span>
+                    <span class="n">frame</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">frame</span><span class="o">.</span><span class="n">to_rgb</span><span class="p">()</span><span class="o">.</span><span class="n">to_ndarray</span><span class="p">())</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="s2">&quot;audio&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span><span class="p">:</span>
+                    <span class="n">frame</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">frame</span><span class="o">.</span><span class="n">to_ndarray</span><span class="p">())</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">frame</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="k">except</span> <span class="n">av</span><span class="o">.</span><span class="n">error</span><span class="o">.</span><span class="n">EOFError</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">StopIteration</span>
+
+        <span class="k">if</span> <span class="n">frame</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">StopIteration</span>
+
+        <span class="k">return</span> <span class="p">{</span><span class="s2">&quot;data&quot;</span><span class="p">:</span> <span class="n">frame</span><span class="p">,</span> <span class="s2">&quot;pts&quot;</span><span class="p">:</span> <span class="n">pts</span><span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__iter__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Iterator</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]:</span>
+        <span class="k">return</span> <span class="bp">self</span>
+
+<div class="viewcode-block" id="VideoReader.seek"><a class="viewcode-back" href="../../../generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader.seek">[docs]</a>    <span class="k">def</span> <span class="nf">seek</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">time_s</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">keyframes_only</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="s2">&quot;VideoReader&quot;</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Seek within current stream.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            time_s (float): seek time in seconds</span>
+<span class="sd">            keyframes_only (bool): allow to seek only to keyframes</span>
+
+<span class="sd">        .. note::</span>
+<span class="sd">            Current implementation is the so-called precise seek. This</span>
+<span class="sd">            means following seek, call to :mod:`next()` will return the</span>
+<span class="sd">            frame with the exact timestamp if it exists or</span>
+<span class="sd">            the first frame with timestamp larger than ``time_s``.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;cuda&quot;</span><span class="p">,</span> <span class="s2">&quot;video_reader&quot;</span><span class="p">]:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_c</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="n">time_s</span><span class="p">,</span> <span class="n">keyframes_only</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># handle special case as pyav doesn&#39;t catch it</span>
+            <span class="k">if</span> <span class="n">time_s</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">time_s</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="n">temp_str</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">offset</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">time_s</span> <span class="o">/</span> <span class="n">temp_str</span><span class="o">.</span><span class="n">time_base</span><span class="p">))</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">keyframes_only</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Accurate seek is not implemented for pyav backend&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">container</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="n">offset</span><span class="p">,</span> <span class="n">backward</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">any_frame</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">stream</span><span class="o">=</span><span class="n">temp_str</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_c</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">container</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span></div>
+
+<div class="viewcode-block" id="VideoReader.get_metadata"><a class="viewcode-back" href="../../../generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader.get_metadata">[docs]</a>    <span class="k">def</span> <span class="nf">get_metadata</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Returns video metadata</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            (dict): dictionary containing duration and frame rate for every stream</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;pyav&quot;</span><span class="p">:</span>
+            <span class="n">metadata</span> <span class="o">=</span> <span class="p">{}</span>  <span class="c1"># type:  Dict[str, Any]</span>
+            <span class="k">for</span> <span class="n">stream</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">container</span><span class="o">.</span><span class="n">streams</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">stream</span><span class="o">.</span><span class="n">type</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">metadata</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="n">stream</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s2">&quot;video&quot;</span><span class="p">:</span>
+                        <span class="n">rate_n</span> <span class="o">=</span> <span class="s2">&quot;fps&quot;</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">rate_n</span> <span class="o">=</span> <span class="s2">&quot;framerate&quot;</span>
+                    <span class="n">metadata</span><span class="p">[</span><span class="n">stream</span><span class="o">.</span><span class="n">type</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span><span class="n">rate_n</span><span class="p">:</span> <span class="p">[],</span> <span class="s2">&quot;duration&quot;</span><span class="p">:</span> <span class="p">[]}</span>
+
+                <span class="n">rate</span> <span class="o">=</span> <span class="n">stream</span><span class="o">.</span><span class="n">average_rate</span> <span class="k">if</span> <span class="n">stream</span><span class="o">.</span><span class="n">average_rate</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">stream</span><span class="o">.</span><span class="n">sample_rate</span>
+
+                <span class="n">metadata</span><span class="p">[</span><span class="n">stream</span><span class="o">.</span><span class="n">type</span><span class="p">][</span><span class="s2">&quot;duration&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">stream</span><span class="o">.</span><span class="n">duration</span> <span class="o">*</span> <span class="n">stream</span><span class="o">.</span><span class="n">time_base</span><span class="p">))</span>
+                <span class="n">metadata</span><span class="p">[</span><span class="n">stream</span><span class="o">.</span><span class="n">type</span><span class="p">][</span><span class="n">rate_n</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">rate</span><span class="p">))</span>
+            <span class="k">return</span> <span class="n">metadata</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_c</span><span class="o">.</span><span class="n">get_metadata</span><span class="p">()</span></div>
+
+<div class="viewcode-block" id="VideoReader.set_current_stream"><a class="viewcode-back" href="../../../generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader.set_current_stream">[docs]</a>    <span class="k">def</span> <span class="nf">set_current_stream</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">stream</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Set current stream.</span>
+<span class="sd">        Explicitly define the stream we are operating on.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            stream (string): descriptor of the required stream. Defaults to ``&quot;video:0&quot;``</span>
+<span class="sd">                Currently available stream types include ``[&#39;video&#39;, &#39;audio&#39;]``.</span>
+<span class="sd">                Each descriptor consists of two parts: stream type (e.g. &#39;video&#39;) and</span>
+<span class="sd">                a unique stream id (which are determined by video encoding).</span>
+<span class="sd">                In this way, if the video container contains multiple</span>
+<span class="sd">                streams of the same type, users can access the one they want.</span>
+<span class="sd">                If only stream type is passed, the decoder auto-detects first stream</span>
+<span class="sd">                of that type and returns it.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            (bool): True on success, False otherwise</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;cuda&quot;</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;GPU decoding only works with video stream.&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;pyav&quot;</span><span class="p">:</span>
+            <span class="n">stream_type</span> <span class="o">=</span> <span class="n">stream</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">stream_id</span> <span class="o">=</span> <span class="mi">0</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">stream</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">))</span> <span class="o">==</span> <span class="mi">1</span> <span class="k">else</span> <span class="nb">int</span><span class="p">(</span><span class="n">stream</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">)[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span> <span class="o">=</span> <span class="p">{</span><span class="n">stream_type</span><span class="p">:</span> <span class="n">stream_id</span><span class="p">}</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_c</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">container</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">pyav_stream</span><span class="p">)</span>
+            <span class="k">return</span> <span class="kc">True</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_c</span><span class="o">.</span><span class="n">set_current_stream</span><span class="p">(</span><span class="n">stream</span><span class="p">)</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/_api.html b/0.16/_modules/torchvision/models/_api.html
new file mode 100644
index 00000000000..dcc0ccf0f0c
--- /dev/null
+++ b/0.16/_modules/torchvision/models/_api.html
@@ -0,0 +1,946 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models._api &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models._api</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models._api</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">fnmatch</span>
+<span class="kn">import</span> <span class="nn">importlib</span>
+<span class="kn">import</span> <span class="nn">inspect</span>
+<span class="kn">import</span> <span class="nn">sys</span>
+<span class="kn">from</span> <span class="nn">dataclasses</span> <span class="kn">import</span> <span class="n">dataclass</span>
+<span class="kn">from</span> <span class="nn">enum</span> <span class="kn">import</span> <span class="n">Enum</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">inspect</span> <span class="kn">import</span> <span class="n">signature</span>
+<span class="kn">from</span> <span class="nn">types</span> <span class="kn">import</span> <span class="n">ModuleType</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Iterable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Mapping</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Set</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">TypeVar</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+
+<span class="kn">from</span> <span class="nn">.._internally_replaced_utils</span> <span class="kn">import</span> <span class="n">load_state_dict_from_url</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;WeightsEnum&quot;</span><span class="p">,</span> <span class="s2">&quot;Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;get_model&quot;</span><span class="p">,</span> <span class="s2">&quot;get_model_builder&quot;</span><span class="p">,</span> <span class="s2">&quot;get_model_weights&quot;</span><span class="p">,</span> <span class="s2">&quot;get_weight&quot;</span><span class="p">,</span> <span class="s2">&quot;list_models&quot;</span><span class="p">]</span>
+
+
+<span class="nd">@dataclass</span>
+<span class="k">class</span> <span class="nc">Weights</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    This class is used to group important attributes associated with the pre-trained weights.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        url (str): The location where we find the weights.</span>
+<span class="sd">        transforms (Callable): A callable that constructs the preprocessing method (or validation preset transforms)</span>
+<span class="sd">            needed to use the model. The reason we attach a constructor method rather than an already constructed</span>
+<span class="sd">            object is because the specific object might have memory and thus we want to delay initialization until</span>
+<span class="sd">            needed.</span>
+<span class="sd">        meta (Dict[str, Any]): Stores meta-data related to the weights of the model and its configuration. These can be</span>
+<span class="sd">            informative attributes (for example the number of parameters/flops, recipe link/methods used in training</span>
+<span class="sd">            etc), configuration parameters (for example the `num_classes`) needed to construct the model or important</span>
+<span class="sd">            meta-data (for example the `classes` of a classification model) needed to use the model.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">url</span><span class="p">:</span> <span class="nb">str</span>
+    <span class="n">transforms</span><span class="p">:</span> <span class="n">Callable</span>
+    <span class="n">meta</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__eq__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">other</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="c1"># We need this custom implementation for correct deep-copy and deserialization behavior.</span>
+        <span class="c1"># TL;DR: After the definition of an enum, creating a new instance, i.e. by deep-copying or deserializing it,</span>
+        <span class="c1"># involves an equality check against the defined members. Unfortunately, the `transforms` attribute is often</span>
+        <span class="c1"># defined with `functools.partial` and `fn = partial(...); assert deepcopy(fn) != fn`. Without custom handling</span>
+        <span class="c1"># for it, the check against the defined members would fail and effectively prevent the weights from being</span>
+        <span class="c1"># deep-copied or deserialized.</span>
+        <span class="c1"># See https://github.com/pytorch/vision/pull/7107 for details.</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">other</span><span class="p">,</span> <span class="n">Weights</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">NotImplemented</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">url</span> <span class="o">!=</span> <span class="n">other</span><span class="o">.</span><span class="n">url</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">False</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">meta</span> <span class="o">!=</span> <span class="n">other</span><span class="o">.</span><span class="n">meta</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">False</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">,</span> <span class="n">partial</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">other</span><span class="o">.</span><span class="n">transforms</span><span class="p">,</span> <span class="n">partial</span><span class="p">):</span>
+            <span class="k">return</span> <span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="o">.</span><span class="n">func</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">transforms</span><span class="o">.</span><span class="n">func</span>
+                <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="o">.</span><span class="n">args</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">transforms</span><span class="o">.</span><span class="n">args</span>
+                <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="o">.</span><span class="n">keywords</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">transforms</span><span class="o">.</span><span class="n">keywords</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">==</span> <span class="n">other</span><span class="o">.</span><span class="n">transforms</span>
+
+
+<span class="k">class</span> <span class="nc">WeightsEnum</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    This class is the parent class of all model weights. Each model building method receives an optional `weights`</span>
+<span class="sd">    parameter with its associated pre-trained weights. It inherits from `Enum` and its values should be of type</span>
+<span class="sd">    `Weights`.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        value (Weights): The data class entry with the weight information.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="nd">@classmethod</span>
+    <span class="k">def</span> <span class="nf">verify</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">obj</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">obj</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">obj</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">str</span><span class="p">:</span>
+                <span class="n">obj</span> <span class="o">=</span> <span class="bp">cls</span><span class="p">[</span><span class="n">obj</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="bp">cls</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="s2">&quot;.&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)]</span>
+            <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="bp">cls</span><span class="p">):</span>
+                <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Invalid Weight class provided; expected </span><span class="si">{</span><span class="bp">cls</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2"> but received </span><span class="si">{</span><span class="n">obj</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">.&quot;</span>
+                <span class="p">)</span>
+        <span class="k">return</span> <span class="n">obj</span>
+
+    <span class="k">def</span> <span class="nf">get_state_dict</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="n">load_state_dict_from_url</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">url</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_name_</span><span class="si">}</span><span class="s2">&quot;</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">url</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="o">.</span><span class="n">url</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">transforms</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="o">.</span><span class="n">transforms</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">meta</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="o">.</span><span class="n">meta</span>
+
+
+<div class="viewcode-block" id="get_weight"><a class="viewcode-back" href="../../../generated/torchvision.models.get_weight.html#torchvision.models.get_weight">[docs]</a><span class="k">def</span> <span class="nf">get_weight</span><span class="p">(</span><span class="n">name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">WeightsEnum</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Gets the weights enum value by its full name. Example: &quot;ResNet50_Weights.IMAGENET1K_V1&quot;</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        name (str): The name of the weight enum entry.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        WeightsEnum: The requested weight enum.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">try</span><span class="p">:</span>
+        <span class="n">enum_name</span><span class="p">,</span> <span class="n">value_name</span> <span class="o">=</span> <span class="n">name</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;.&quot;</span><span class="p">)</span>
+    <span class="k">except</span> <span class="ne">ValueError</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid weight name provided: &#39;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">&#39;.&quot;</span><span class="p">)</span>
+
+    <span class="n">base_module_name</span> <span class="o">=</span> <span class="s2">&quot;.&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">modules</span><span class="p">[</span><span class="vm">__name__</span><span class="p">]</span><span class="o">.</span><span class="vm">__name__</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;.&quot;</span><span class="p">)[:</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span>
+    <span class="n">base_module</span> <span class="o">=</span> <span class="n">importlib</span><span class="o">.</span><span class="n">import_module</span><span class="p">(</span><span class="n">base_module_name</span><span class="p">)</span>
+    <span class="n">model_modules</span> <span class="o">=</span> <span class="p">[</span><span class="n">base_module</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span>
+        <span class="n">x</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">inspect</span><span class="o">.</span><span class="n">getmembers</span><span class="p">(</span><span class="n">base_module</span><span class="p">,</span> <span class="n">inspect</span><span class="o">.</span><span class="n">ismodule</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">x</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="vm">__file__</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;__init__.py&quot;</span><span class="p">)</span>  <span class="c1"># type: ignore[union-attr]</span>
+    <span class="p">]</span>
+
+    <span class="n">weights_enum</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="n">model_modules</span><span class="p">:</span>
+        <span class="n">potential_class</span> <span class="o">=</span> <span class="n">m</span><span class="o">.</span><span class="vm">__dict__</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">enum_name</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">potential_class</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">potential_class</span><span class="p">,</span> <span class="n">WeightsEnum</span><span class="p">):</span>
+            <span class="n">weights_enum</span> <span class="o">=</span> <span class="n">potential_class</span>
+            <span class="k">break</span>
+
+    <span class="k">if</span> <span class="n">weights_enum</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The weight enum &#39;</span><span class="si">{</span><span class="n">enum_name</span><span class="si">}</span><span class="s2">&#39; for the specific method couldn&#39;t be retrieved.&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">weights_enum</span><span class="p">[</span><span class="n">value_name</span><span class="p">]</span></div>
+
+
+<div class="viewcode-block" id="get_model_weights"><a class="viewcode-back" href="../../../generated/torchvision.models.get_model_weights.html#torchvision.models.get_model_weights">[docs]</a><span class="k">def</span> <span class="nf">get_model_weights</span><span class="p">(</span><span class="n">name</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Callable</span><span class="p">,</span> <span class="nb">str</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Type</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Returns the weights enum class associated to the given model.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        name (callable or str): The model builder function or the name under which it is registered.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        weights_enum (WeightsEnum): The weights enum class associated with the model.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">get_model_builder</span><span class="p">(</span><span class="n">name</span><span class="p">)</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">name</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="k">else</span> <span class="n">name</span>
+    <span class="k">return</span> <span class="n">_get_enum_from_fn</span><span class="p">(</span><span class="n">model</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_get_enum_from_fn</span><span class="p">(</span><span class="n">fn</span><span class="p">:</span> <span class="n">Callable</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Type</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Internal method that gets the weight enum of a specific model builder method.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        fn (Callable): The builder method used to create the model.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        WeightsEnum: The requested weight enum.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">sig</span> <span class="o">=</span> <span class="n">signature</span><span class="p">(</span><span class="n">fn</span><span class="p">)</span>
+    <span class="k">if</span> <span class="s2">&quot;weights&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">sig</span><span class="o">.</span><span class="n">parameters</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The method is missing the &#39;weights&#39; argument.&quot;</span><span class="p">)</span>
+
+    <span class="n">ann</span> <span class="o">=</span> <span class="n">signature</span><span class="p">(</span><span class="n">fn</span><span class="p">)</span><span class="o">.</span><span class="n">parameters</span><span class="p">[</span><span class="s2">&quot;weights&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">annotation</span>
+    <span class="n">weights_enum</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">ann</span><span class="p">,</span> <span class="nb">type</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">ann</span><span class="p">,</span> <span class="n">WeightsEnum</span><span class="p">):</span>
+        <span class="n">weights_enum</span> <span class="o">=</span> <span class="n">ann</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="c1"># handle cases like Union[Optional, T]</span>
+        <span class="c1"># TODO: Replace ann.__args__ with typing.get_args(ann) after python &gt;= 3.8</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">ann</span><span class="o">.</span><span class="n">__args__</span><span class="p">:</span>  <span class="c1"># type: ignore[union-attr]</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="nb">type</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="n">WeightsEnum</span><span class="p">):</span>
+                <span class="n">weights_enum</span> <span class="o">=</span> <span class="n">t</span>
+                <span class="k">break</span>
+
+    <span class="k">if</span> <span class="n">weights_enum</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="s2">&quot;The WeightsEnum class for the specific method couldn&#39;t be retrieved. Make sure the typing info is correct.&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">weights_enum</span>
+
+
+<span class="n">M</span> <span class="o">=</span> <span class="n">TypeVar</span><span class="p">(</span><span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="n">bound</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">)</span>
+
+<span class="n">BUILTIN_MODELS</span> <span class="o">=</span> <span class="p">{}</span>
+
+
+<span class="k">def</span> <span class="nf">register_model</span><span class="p">(</span><span class="n">name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Callable</span><span class="p">[[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">M</span><span class="p">]],</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">M</span><span class="p">]]:</span>
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">fn</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">M</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">M</span><span class="p">]:</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">name</span> <span class="k">if</span> <span class="n">name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">fn</span><span class="o">.</span><span class="vm">__name__</span>
+        <span class="k">if</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">BUILTIN_MODELS</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;An entry is already registered under the name &#39;</span><span class="si">{</span><span class="n">key</span><span class="si">}</span><span class="s2">&#39;.&quot;</span><span class="p">)</span>
+        <span class="n">BUILTIN_MODELS</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="n">fn</span>
+        <span class="k">return</span> <span class="n">fn</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<div class="viewcode-block" id="list_models"><a class="viewcode-back" href="../../../generated/torchvision.models.list_models.html#torchvision.models.list_models">[docs]</a><span class="k">def</span> <span class="nf">list_models</span><span class="p">(</span>
+    <span class="n">module</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ModuleType</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">include</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Iterable</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">str</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">exclude</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Iterable</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="nb">str</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Returns a list with the names of registered models.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        module (ModuleType, optional): The module from which we want to extract the available models.</span>
+<span class="sd">        include (str or Iterable[str], optional): Filter(s) for including the models from the set of all models.</span>
+<span class="sd">            Filters are passed to `fnmatch &lt;https://docs.python.org/3/library/fnmatch.html&gt;`__ to match Unix shell-style</span>
+<span class="sd">            wildcards. In case of many filters, the results is the union of individual filters.</span>
+<span class="sd">        exclude (str or Iterable[str], optional): Filter(s) applied after include_filters to remove models.</span>
+<span class="sd">            Filter are passed to `fnmatch &lt;https://docs.python.org/3/library/fnmatch.html&gt;`__ to match Unix shell-style</span>
+<span class="sd">            wildcards. In case of many filters, the results is removal of all the models that match any individual filter.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        models (list): A list with the names of available models.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">all_models</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="n">k</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">BUILTIN_MODELS</span><span class="o">.</span><span class="n">items</span><span class="p">()</span> <span class="k">if</span> <span class="n">module</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">v</span><span class="o">.</span><span class="vm">__module__</span><span class="o">.</span><span class="n">rsplit</span><span class="p">(</span><span class="s2">&quot;.&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">module</span><span class="o">.</span><span class="vm">__name__</span>
+    <span class="p">}</span>
+    <span class="k">if</span> <span class="n">include</span><span class="p">:</span>
+        <span class="n">models</span><span class="p">:</span> <span class="n">Set</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">include</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="n">include</span> <span class="o">=</span> <span class="p">[</span><span class="n">include</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">include_filter</span> <span class="ow">in</span> <span class="n">include</span><span class="p">:</span>
+            <span class="n">models</span> <span class="o">=</span> <span class="n">models</span> <span class="o">|</span> <span class="nb">set</span><span class="p">(</span><span class="n">fnmatch</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="n">all_models</span><span class="p">,</span> <span class="n">include_filter</span><span class="p">))</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">models</span> <span class="o">=</span> <span class="n">all_models</span>
+
+    <span class="k">if</span> <span class="n">exclude</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">exclude</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="n">exclude</span> <span class="o">=</span> <span class="p">[</span><span class="n">exclude</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">exclude_filter</span> <span class="ow">in</span> <span class="n">exclude</span><span class="p">:</span>
+            <span class="n">models</span> <span class="o">=</span> <span class="n">models</span> <span class="o">-</span> <span class="nb">set</span><span class="p">(</span><span class="n">fnmatch</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="n">all_models</span><span class="p">,</span> <span class="n">exclude_filter</span><span class="p">))</span>
+    <span class="k">return</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">models</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">get_model_builder</span><span class="p">(</span><span class="n">name</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Gets the model name and returns the model builder method.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        name (str): The name under which the model is registered.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        fn (Callable): The model builder method.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">name</span> <span class="o">=</span> <span class="n">name</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+    <span class="k">try</span><span class="p">:</span>
+        <span class="n">fn</span> <span class="o">=</span> <span class="n">BUILTIN_MODELS</span><span class="p">[</span><span class="n">name</span><span class="p">]</span>
+    <span class="k">except</span> <span class="ne">KeyError</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unknown model </span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">fn</span>
+
+
+<div class="viewcode-block" id="get_model"><a class="viewcode-back" href="../../../generated/torchvision.models.get_model.html#torchvision.models.get_model">[docs]</a><span class="k">def</span> <span class="nf">get_model</span><span class="p">(</span><span class="n">name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="o">**</span><span class="n">config</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Gets the model name and configuration and returns an instantiated model.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        name (str): The name under which the model is registered.</span>
+<span class="sd">        **config (Any): parameters passed to the model builder method.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        model (nn.Module): The initialized model.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">fn</span> <span class="o">=</span> <span class="n">get_model_builder</span><span class="p">(</span><span class="n">name</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">fn</span><span class="p">(</span><span class="o">**</span><span class="n">config</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/alexnet.html b/0.16/_modules/torchvision/models/alexnet.html
new file mode 100644
index 00000000000..40ecd368134
--- /dev/null
+++ b/0.16/_modules/torchvision/models/alexnet.html
@@ -0,0 +1,788 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.alexnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.alexnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.alexnet</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;AlexNet&quot;</span><span class="p">,</span> <span class="s2">&quot;AlexNet_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;alexnet&quot;</span><span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">AlexNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">11</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">((</span><span class="mi">6</span><span class="p">,</span> <span class="mi">6</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">256</span> <span class="o">*</span> <span class="mi">6</span> <span class="o">*</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">4096</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">4096</span><span class="p">,</span> <span class="mi">4096</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">4096</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<div class="viewcode-block" id="AlexNet_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.alexnet.html#torchvision.models.AlexNet_Weights">[docs]</a><span class="k">class</span> <span class="nc">AlexNet_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/alexnet-owt-7be5be79.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">61100840</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">63</span><span class="p">,</span> <span class="mi">63</span><span class="p">),</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">56.522</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">79.066</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.714</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">233.087</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights reproduce closely the results of the paper using a simplified training recipe.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="alexnet"><a class="viewcode-back" href="../../../models/generated/torchvision.models.alexnet.html#torchvision.models.alexnet">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">AlexNet_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">alexnet</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">AlexNet_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">AlexNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;AlexNet model architecture from `One weird trick for parallelizing convolutional neural networks &lt;https://arxiv.org/abs/1404.5997&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        AlexNet was originally introduced in the `ImageNet Classification with</span>
+<span class="sd">        Deep Convolutional Neural Networks</span>
+<span class="sd">        &lt;https://papers.nips.cc/paper/2012/hash/c399862d3b9d6b76c8436e924a68c45b-Abstract.html&gt;`__</span>
+<span class="sd">        paper. Our implementation is based instead on the &quot;One weird trick&quot;</span>
+<span class="sd">        paper above.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.AlexNet_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.AlexNet_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.squeezenet.AlexNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/alexnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.AlexNet_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">AlexNet_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">AlexNet</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/convnext.html b/0.16/_modules/torchvision/models/convnext.html
new file mode 100644
index 00000000000..eb432c007bd
--- /dev/null
+++ b/0.16/_modules/torchvision/models/convnext.html
@@ -0,0 +1,1083 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.convnext &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.convnext</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.convnext</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.nn</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span><span class="p">,</span> <span class="n">Permute</span>
+<span class="kn">from</span> <span class="nn">..ops.stochastic_depth</span> <span class="kn">import</span> <span class="n">StochasticDepth</span>
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;ConvNeXt&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ConvNeXt_Tiny_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ConvNeXt_Small_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ConvNeXt_Base_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ConvNeXt_Large_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;convnext_tiny&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;convnext_small&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;convnext_base&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;convnext_large&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">LayerNorm2d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">layer_norm</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">normalized_shape</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">eps</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">CNBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dim</span><span class="p">,</span>
+        <span class="n">layer_scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">block</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="n">dim</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">Permute</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">]),</span>
+            <span class="n">norm_layer</span><span class="p">(</span><span class="n">dim</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_features</span><span class="o">=</span><span class="n">dim</span><span class="p">,</span> <span class="n">out_features</span><span class="o">=</span><span class="mi">4</span> <span class="o">*</span> <span class="n">dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">GELU</span><span class="p">(),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_features</span><span class="o">=</span><span class="mi">4</span> <span class="o">*</span> <span class="n">dim</span><span class="p">,</span> <span class="n">out_features</span><span class="o">=</span><span class="n">dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">Permute</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">]),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer_scale</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">layer_scale</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span> <span class="o">=</span> <span class="n">StochasticDepth</span><span class="p">(</span><span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="s2">&quot;row&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer_scale</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">block</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="n">result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
+        <span class="n">result</span> <span class="o">+=</span> <span class="nb">input</span>
+        <span class="k">return</span> <span class="n">result</span>
+
+
+<span class="k">class</span> <span class="nc">CNBlockConfig</span><span class="p">:</span>
+    <span class="c1"># Stores information listed at Section 3 of the ConvNeXt paper</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">input_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">=</span> <span class="n">input_channels</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">out_channels</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_layers</span> <span class="o">=</span> <span class="n">num_layers</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="s2">&quot;(&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="s2">&quot;input_channels=</span><span class="si">{input_channels}</span><span class="s2">&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="s2">&quot;, out_channels=</span><span class="si">{out_channels}</span><span class="s2">&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="s2">&quot;, num_layers=</span><span class="si">{num_layers}</span><span class="s2">&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="s2">&quot;)&quot;</span>
+        <span class="k">return</span> <span class="n">s</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ConvNeXt</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">block_setting</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">CNBlockConfig</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">layer_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-6</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">block_setting</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The block_setting should not be empty&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">block_setting</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">all</span><span class="p">([</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">s</span><span class="p">,</span> <span class="n">CNBlockConfig</span><span class="p">)</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">block_setting</span><span class="p">])):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;The block_setting should be List[CNBlockConfig]&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">CNBlock</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">LayerNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">)</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="c1"># Stem</span>
+        <span class="n">firstconv_output_channels</span> <span class="o">=</span> <span class="n">block_setting</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">input_channels</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="mi">3</span><span class="p">,</span>
+                <span class="n">firstconv_output_channels</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+                <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="n">total_stage_blocks</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">num_layers</span> <span class="k">for</span> <span class="n">cnf</span> <span class="ow">in</span> <span class="n">block_setting</span><span class="p">)</span>
+        <span class="n">stage_block_id</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">cnf</span> <span class="ow">in</span> <span class="n">block_setting</span><span class="p">:</span>
+            <span class="c1"># Bottlenecks</span>
+            <span class="n">stage</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">num_layers</span><span class="p">):</span>
+                <span class="c1"># adjust stochastic depth probability based on the depth of the stage block</span>
+                <span class="n">sd_prob</span> <span class="o">=</span> <span class="n">stochastic_depth_prob</span> <span class="o">*</span> <span class="n">stage_block_id</span> <span class="o">/</span> <span class="p">(</span><span class="n">total_stage_blocks</span> <span class="o">-</span> <span class="mf">1.0</span><span class="p">)</span>
+                <span class="n">stage</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">layer_scale</span><span class="p">,</span> <span class="n">sd_prob</span><span class="p">))</span>
+                <span class="n">stage_block_id</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">stage</span><span class="p">))</span>
+            <span class="k">if</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="c1"># Downsampling</span>
+                <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                        <span class="n">norm_layer</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">),</span>
+                        <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+                    <span class="p">)</span>
+                <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">lastblock</span> <span class="o">=</span> <span class="n">block_setting</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">lastconv_output_channels</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">lastblock</span><span class="o">.</span><span class="n">out_channels</span> <span class="k">if</span> <span class="n">lastblock</span><span class="o">.</span><span class="n">out_channels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">lastblock</span><span class="o">.</span><span class="n">input_channels</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">norm_layer</span><span class="p">(</span><span class="n">lastconv_output_channels</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">Flatten</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">lastconv_output_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">)):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward_impl</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_convnext</span><span class="p">(</span>
+    <span class="n">block_setting</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">CNBlockConfig</span><span class="p">],</span>
+    <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ConvNeXt</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">ConvNeXt</span><span class="p">(</span><span class="n">block_setting</span><span class="p">,</span> <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#convnext&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">        These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">        `new training recipe</span>
+<span class="s2">        &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">    &quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="ConvNeXt_Tiny_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_tiny.html#torchvision.models.ConvNeXt_Tiny_Weights">[docs]</a><span class="k">class</span> <span class="nc">ConvNeXt_Tiny_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/convnext_tiny-983f1562.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">236</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">28589128</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.520</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.146</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.456</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">109.119</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ConvNeXt_Small_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_small.html#torchvision.models.ConvNeXt_Small_Weights">[docs]</a><span class="k">class</span> <span class="nc">ConvNeXt_Small_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/convnext_small-0c510722.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">230</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">50223688</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.616</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.650</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">8.684</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">191.703</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ConvNeXt_Base_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_base.html#torchvision.models.ConvNeXt_Base_Weights">[docs]</a><span class="k">class</span> <span class="nc">ConvNeXt_Base_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/convnext_base-6075fbad.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88591464</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">84.062</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.870</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.355</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">338.064</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ConvNeXt_Large_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_large.html#torchvision.models.ConvNeXt_Large_Weights">[docs]</a><span class="k">class</span> <span class="nc">ConvNeXt_Large_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/convnext_large-ea097f82.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">197767336</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">84.414</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.976</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">34.361</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">754.537</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="convnext_tiny"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_tiny.html#torchvision.models.convnext_tiny">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ConvNeXt_Tiny_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">convnext_tiny</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ConvNeXt_Tiny_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ConvNeXt</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ConvNeXt Tiny model architecture from the</span>
+<span class="sd">    `A ConvNet for the 2020s &lt;https://arxiv.org/abs/2201.03545&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.convnext.ConvNeXt_Tiny_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.convnext.ConvNeXt_Tiny_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.convnext.ConvNext``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ConvNeXt_Tiny_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ConvNeXt_Tiny_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">block_setting</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+    <span class="p">]</span>
+    <span class="n">stochastic_depth_prob</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;stochastic_depth_prob&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_convnext</span><span class="p">(</span><span class="n">block_setting</span><span class="p">,</span> <span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="convnext_small"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_small.html#torchvision.models.convnext_small">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ConvNeXt_Small_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">convnext_small</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ConvNeXt_Small_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ConvNeXt</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ConvNeXt Small model architecture from the</span>
+<span class="sd">    `A ConvNet for the 2020s &lt;https://arxiv.org/abs/2201.03545&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.convnext.ConvNeXt_Small_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.convnext.ConvNeXt_Small_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.convnext.ConvNext``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ConvNeXt_Small_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ConvNeXt_Small_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">block_setting</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">27</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+    <span class="p">]</span>
+    <span class="n">stochastic_depth_prob</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;stochastic_depth_prob&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_convnext</span><span class="p">(</span><span class="n">block_setting</span><span class="p">,</span> <span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="convnext_base"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_base.html#torchvision.models.convnext_base">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ConvNeXt_Base_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">convnext_base</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ConvNeXt_Base_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ConvNeXt</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ConvNeXt Base model architecture from the</span>
+<span class="sd">    `A ConvNet for the 2020s &lt;https://arxiv.org/abs/2201.03545&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.convnext.ConvNeXt_Base_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.convnext.ConvNeXt_Base_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.convnext.ConvNext``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ConvNeXt_Base_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ConvNeXt_Base_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">block_setting</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">1024</span><span class="p">,</span> <span class="mi">27</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+    <span class="p">]</span>
+    <span class="n">stochastic_depth_prob</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;stochastic_depth_prob&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_convnext</span><span class="p">(</span><span class="n">block_setting</span><span class="p">,</span> <span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="convnext_large"><a class="viewcode-back" href="../../../models/generated/torchvision.models.convnext_large.html#torchvision.models.convnext_large">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ConvNeXt_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">convnext_large</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ConvNeXt_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ConvNeXt</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ConvNeXt Large model architecture from the</span>
+<span class="sd">    `A ConvNet for the 2020s &lt;https://arxiv.org/abs/2201.03545&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.convnext.ConvNeXt_Large_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.convnext.ConvNeXt_Large_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.convnext.ConvNext``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ConvNeXt_Large_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ConvNeXt_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">block_setting</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="mi">1536</span><span class="p">,</span> <span class="mi">27</span><span class="p">),</span>
+        <span class="n">CNBlockConfig</span><span class="p">(</span><span class="mi">1536</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+    <span class="p">]</span>
+    <span class="n">stochastic_depth_prob</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;stochastic_depth_prob&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_convnext</span><span class="p">(</span><span class="n">block_setting</span><span class="p">,</span> <span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/densenet.html b/0.16/_modules/torchvision/models/densenet.html
new file mode 100644
index 00000000000..5705c581fa7
--- /dev/null
+++ b/0.16/_modules/torchvision/models/densenet.html
@@ -0,0 +1,1117 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.densenet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.densenet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.densenet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">re</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">import</span> <span class="nn">torch.utils.checkpoint</span> <span class="k">as</span> <span class="nn">cp</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;DenseNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;DenseNet121_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;DenseNet161_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;DenseNet169_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;DenseNet201_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;densenet121&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;densenet161&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;densenet169&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;densenet201&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">_DenseLayer</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">num_input_features</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">growth_rate</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">bn_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">drop_rate</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">memory_efficient</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">num_input_features</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">num_input_features</span><span class="p">,</span> <span class="n">bn_size</span> <span class="o">*</span> <span class="n">growth_rate</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">bn_size</span> <span class="o">*</span> <span class="n">growth_rate</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">bn_size</span> <span class="o">*</span> <span class="n">growth_rate</span><span class="p">,</span> <span class="n">growth_rate</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">drop_rate</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">drop_rate</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">memory_efficient</span> <span class="o">=</span> <span class="n">memory_efficient</span>
+
+    <span class="k">def</span> <span class="nf">bn_function</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">concated_features</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">bottleneck_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relu1</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">norm1</span><span class="p">(</span><span class="n">concated_features</span><span class="p">)))</span>  <span class="c1"># noqa: T484</span>
+        <span class="k">return</span> <span class="n">bottleneck_output</span>
+
+    <span class="c1"># todo: rewrite when torchscript supports any</span>
+    <span class="k">def</span> <span class="nf">any_requires_grad</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">tensor</span> <span class="ow">in</span> <span class="nb">input</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">:</span>
+                <span class="k">return</span> <span class="kc">True</span>
+        <span class="k">return</span> <span class="kc">False</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>  <span class="c1"># noqa: T484</span>
+    <span class="k">def</span> <span class="nf">call_checkpoint_bottleneck</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">def</span> <span class="nf">closure</span><span class="p">(</span><span class="o">*</span><span class="n">inputs</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn_function</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">cp</span><span class="o">.</span><span class="n">checkpoint</span><span class="p">(</span><span class="n">closure</span><span class="p">,</span> <span class="o">*</span><span class="nb">input</span><span class="p">)</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">_overload_method</span>  <span class="c1"># noqa: F811</span>
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>  <span class="c1"># noqa: F811</span>
+        <span class="k">pass</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">_overload_method</span>  <span class="c1"># noqa: F811</span>
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>  <span class="c1"># noqa: F811</span>
+        <span class="k">pass</span>
+
+    <span class="c1"># torchscript does not yet support *args, so we overload method</span>
+    <span class="c1"># allowing it to take either a List[Tensor] or single Tensor</span>
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>  <span class="c1"># noqa: F811</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="n">prev_features</span> <span class="o">=</span> <span class="p">[</span><span class="nb">input</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">prev_features</span> <span class="o">=</span> <span class="nb">input</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">memory_efficient</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">any_requires_grad</span><span class="p">(</span><span class="n">prev_features</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+                <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s2">&quot;Memory Efficient not supported in JIT&quot;</span><span class="p">)</span>
+
+            <span class="n">bottleneck_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">call_checkpoint_bottleneck</span><span class="p">(</span><span class="n">prev_features</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">bottleneck_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn_function</span><span class="p">(</span><span class="n">prev_features</span><span class="p">)</span>
+
+        <span class="n">new_features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relu2</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">norm2</span><span class="p">(</span><span class="n">bottleneck_output</span><span class="p">)))</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">drop_rate</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">new_features</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">new_features</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">drop_rate</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">new_features</span>
+
+
+<span class="k">class</span> <span class="nc">_DenseBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">ModuleDict</span><span class="p">):</span>
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_input_features</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">bn_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">growth_rate</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">drop_rate</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">memory_efficient</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_layers</span><span class="p">):</span>
+            <span class="n">layer</span> <span class="o">=</span> <span class="n">_DenseLayer</span><span class="p">(</span>
+                <span class="n">num_input_features</span> <span class="o">+</span> <span class="n">i</span> <span class="o">*</span> <span class="n">growth_rate</span><span class="p">,</span>
+                <span class="n">growth_rate</span><span class="o">=</span><span class="n">growth_rate</span><span class="p">,</span>
+                <span class="n">bn_size</span><span class="o">=</span><span class="n">bn_size</span><span class="p">,</span>
+                <span class="n">drop_rate</span><span class="o">=</span><span class="n">drop_rate</span><span class="p">,</span>
+                <span class="n">memory_efficient</span><span class="o">=</span><span class="n">memory_efficient</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">add_module</span><span class="p">(</span><span class="s2">&quot;denselayer</span><span class="si">%d</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span> <span class="n">layer</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">init_features</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">features</span> <span class="o">=</span> <span class="p">[</span><span class="n">init_features</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="n">new_features</span> <span class="o">=</span> <span class="n">layer</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+            <span class="n">features</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">new_features</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">features</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">_Transition</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_input_features</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_output_features</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">num_input_features</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">num_input_features</span><span class="p">,</span> <span class="n">num_output_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AvgPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">DenseNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Densenet-BC model class, based on</span>
+<span class="sd">    `&quot;Densely Connected Convolutional Networks&quot; &lt;https://arxiv.org/pdf/1608.06993.pdf&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        growth_rate (int) - how many filters to add each layer (`k` in paper)</span>
+<span class="sd">        block_config (list of 4 ints) - how many layers in each pooling block</span>
+<span class="sd">        num_init_features (int) - the number of filters to learn in the first convolution layer</span>
+<span class="sd">        bn_size (int) - multiplicative factor for number of bottle neck layers</span>
+<span class="sd">          (i.e. bn_size * k features in the bottleneck layer)</span>
+<span class="sd">        drop_rate (float) - dropout rate after each dense layer</span>
+<span class="sd">        num_classes (int) - number of classification classes</span>
+<span class="sd">        memory_efficient (bool) - If True, uses checkpointing. Much more memory efficient,</span>
+<span class="sd">          but slower. Default: *False*. See `&quot;paper&quot; &lt;https://arxiv.org/pdf/1707.06990.pdf&gt;`_.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">growth_rate</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">32</span><span class="p">,</span>
+        <span class="n">block_config</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">16</span><span class="p">),</span>
+        <span class="n">num_init_features</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">64</span><span class="p">,</span>
+        <span class="n">bn_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="n">drop_rate</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">memory_efficient</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="c1"># First convolution</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">OrderedDict</span><span class="p">(</span>
+                <span class="p">[</span>
+                    <span class="p">(</span><span class="s2">&quot;conv0&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">num_init_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)),</span>
+                    <span class="p">(</span><span class="s2">&quot;norm0&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">num_init_features</span><span class="p">)),</span>
+                    <span class="p">(</span><span class="s2">&quot;relu0&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)),</span>
+                    <span class="p">(</span><span class="s2">&quot;pool0&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)),</span>
+                <span class="p">]</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="c1"># Each denseblock</span>
+        <span class="n">num_features</span> <span class="o">=</span> <span class="n">num_init_features</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">num_layers</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">block_config</span><span class="p">):</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">_DenseBlock</span><span class="p">(</span>
+                <span class="n">num_layers</span><span class="o">=</span><span class="n">num_layers</span><span class="p">,</span>
+                <span class="n">num_input_features</span><span class="o">=</span><span class="n">num_features</span><span class="p">,</span>
+                <span class="n">bn_size</span><span class="o">=</span><span class="n">bn_size</span><span class="p">,</span>
+                <span class="n">growth_rate</span><span class="o">=</span><span class="n">growth_rate</span><span class="p">,</span>
+                <span class="n">drop_rate</span><span class="o">=</span><span class="n">drop_rate</span><span class="p">,</span>
+                <span class="n">memory_efficient</span><span class="o">=</span><span class="n">memory_efficient</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="o">.</span><span class="n">add_module</span><span class="p">(</span><span class="s2">&quot;denseblock</span><span class="si">%d</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span> <span class="n">block</span><span class="p">)</span>
+            <span class="n">num_features</span> <span class="o">=</span> <span class="n">num_features</span> <span class="o">+</span> <span class="n">num_layers</span> <span class="o">*</span> <span class="n">growth_rate</span>
+            <span class="k">if</span> <span class="n">i</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">block_config</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="n">trans</span> <span class="o">=</span> <span class="n">_Transition</span><span class="p">(</span><span class="n">num_input_features</span><span class="o">=</span><span class="n">num_features</span><span class="p">,</span> <span class="n">num_output_features</span><span class="o">=</span><span class="n">num_features</span> <span class="o">//</span> <span class="mi">2</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="o">.</span><span class="n">add_module</span><span class="p">(</span><span class="s2">&quot;transition</span><span class="si">%d</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span> <span class="n">trans</span><span class="p">)</span>
+                <span class="n">num_features</span> <span class="o">=</span> <span class="n">num_features</span> <span class="o">//</span> <span class="mi">2</span>
+
+        <span class="c1"># Final batch norm</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="o">.</span><span class="n">add_module</span><span class="p">(</span><span class="s2">&quot;norm5&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">num_features</span><span class="p">))</span>
+
+        <span class="c1"># Linear layer</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">num_features</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="c1"># Official init from torch repo.</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">features</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adaptive_avg_pool2d</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">def</span> <span class="nf">_load_state_dict</span><span class="p">(</span><span class="n">model</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">WeightsEnum</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+    <span class="c1"># &#39;.&#39;s are no longer allowed in module names, but previous _DenseLayer</span>
+    <span class="c1"># has keys &#39;norm.1&#39;, &#39;relu.1&#39;, &#39;conv.1&#39;, &#39;norm.2&#39;, &#39;relu.2&#39;, &#39;conv.2&#39;.</span>
+    <span class="c1"># They are also in the checkpoints in model_urls. This pattern is used</span>
+    <span class="c1"># to find such keys.</span>
+    <span class="n">pattern</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">compile</span><span class="p">(</span>
+        <span class="sa">r</span><span class="s2">&quot;^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$&quot;</span>
+    <span class="p">)</span>
+
+    <span class="n">state_dict</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="nb">list</span><span class="p">(</span><span class="n">state_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">()):</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="n">pattern</span><span class="o">.</span><span class="n">match</span><span class="p">(</span><span class="n">key</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">res</span><span class="p">:</span>
+            <span class="n">new_key</span> <span class="o">=</span> <span class="n">res</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">+</span> <span class="n">res</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+            <span class="n">state_dict</span><span class="p">[</span><span class="n">new_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">state_dict</span><span class="p">[</span><span class="n">key</span><span class="p">]</span>
+            <span class="k">del</span> <span class="n">state_dict</span><span class="p">[</span><span class="n">key</span><span class="p">]</span>
+    <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">state_dict</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_densenet</span><span class="p">(</span>
+    <span class="n">growth_rate</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">block_config</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">num_init_features</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DenseNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">DenseNet</span><span class="p">(</span><span class="n">growth_rate</span><span class="p">,</span> <span class="n">block_config</span><span class="p">,</span> <span class="n">num_init_features</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_load_state_dict</span><span class="p">(</span><span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">29</span><span class="p">,</span> <span class="mi">29</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/116&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from LuaTorch.&quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="DenseNet121_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet121.html#torchvision.models.DenseNet121_Weights">[docs]</a><span class="k">class</span> <span class="nc">DenseNet121_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/densenet121-a639ec97.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">7978856</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">74.434</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.972</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">2.834</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">30.845</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="DenseNet161_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet161.html#torchvision.models.DenseNet161_Weights">[docs]</a><span class="k">class</span> <span class="nc">DenseNet161_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/densenet161-8d451a50.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">28681000</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.138</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.560</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">7.728</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">110.369</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="DenseNet169_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet169.html#torchvision.models.DenseNet169_Weights">[docs]</a><span class="k">class</span> <span class="nc">DenseNet169_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/densenet169-b2777c0a.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">14149480</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">75.600</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.806</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">3.36</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">54.708</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="DenseNet201_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet201.html#torchvision.models.DenseNet201_Weights">[docs]</a><span class="k">class</span> <span class="nc">DenseNet201_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/densenet201-c1103571.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">20013928</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">76.896</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.370</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.291</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">77.373</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="densenet121"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet121.html#torchvision.models.densenet121">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">DenseNet121_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">densenet121</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DenseNet121_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DenseNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Densenet-121 model from</span>
+<span class="sd">    `Densely Connected Convolutional Networks &lt;https://arxiv.org/abs/1608.06993&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.DenseNet121_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.DenseNet121_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.densenet.DenseNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.DenseNet121_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">DenseNet121_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_densenet</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">16</span><span class="p">),</span> <span class="mi">64</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="densenet161"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet161.html#torchvision.models.densenet161">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">DenseNet161_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">densenet161</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DenseNet161_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DenseNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Densenet-161 model from</span>
+<span class="sd">    `Densely Connected Convolutional Networks &lt;https://arxiv.org/abs/1608.06993&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.DenseNet161_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.DenseNet161_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.densenet.DenseNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.DenseNet161_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">DenseNet161_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_densenet</span><span class="p">(</span><span class="mi">48</span><span class="p">,</span> <span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">36</span><span class="p">,</span> <span class="mi">24</span><span class="p">),</span> <span class="mi">96</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="densenet169"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet169.html#torchvision.models.densenet169">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">DenseNet169_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">densenet169</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DenseNet169_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DenseNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Densenet-169 model from</span>
+<span class="sd">    `Densely Connected Convolutional Networks &lt;https://arxiv.org/abs/1608.06993&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.DenseNet169_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.DenseNet169_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.densenet.DenseNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.DenseNet169_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">DenseNet169_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_densenet</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">),</span> <span class="mi">64</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="densenet201"><a class="viewcode-back" href="../../../models/generated/torchvision.models.densenet201.html#torchvision.models.densenet201">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">DenseNet201_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">densenet201</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DenseNet201_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DenseNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Densenet-201 model from</span>
+<span class="sd">    `Densely Connected Convolutional Networks &lt;https://arxiv.org/abs/1608.06993&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.DenseNet201_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.DenseNet201_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.densenet.DenseNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.DenseNet201_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">DenseNet201_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_densenet</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">32</span><span class="p">),</span> <span class="mi">64</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/detection/faster_rcnn.html b/0.16/_modules/torchvision/models/detection/faster_rcnn.html
new file mode 100644
index 00000000000..8d2e8acb1db
--- /dev/null
+++ b/0.16/_modules/torchvision/models/detection/faster_rcnn.html
@@ -0,0 +1,1512 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.detection.faster_rcnn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.detection.faster_rcnn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.detection.faster_rcnn</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops</span> <span class="kn">import</span> <span class="n">MultiScaleRoIAlign</span>
+
+<span class="kn">from</span> <span class="nn">...ops</span> <span class="kn">import</span> <span class="n">misc</span> <span class="k">as</span> <span class="n">misc_nn_ops</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ObjectDetection</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_COCO_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..mobilenetv3</span> <span class="kn">import</span> <span class="n">mobilenet_v3_large</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span>
+<span class="kn">from</span> <span class="nn">..resnet</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">overwrite_eps</span>
+<span class="kn">from</span> <span class="nn">.anchor_utils</span> <span class="kn">import</span> <span class="n">AnchorGenerator</span>
+<span class="kn">from</span> <span class="nn">.backbone_utils</span> <span class="kn">import</span> <span class="n">_mobilenet_extractor</span><span class="p">,</span> <span class="n">_resnet_fpn_extractor</span><span class="p">,</span> <span class="n">_validate_trainable_layers</span>
+<span class="kn">from</span> <span class="nn">.generalized_rcnn</span> <span class="kn">import</span> <span class="n">GeneralizedRCNN</span>
+<span class="kn">from</span> <span class="nn">.roi_heads</span> <span class="kn">import</span> <span class="n">RoIHeads</span>
+<span class="kn">from</span> <span class="nn">.rpn</span> <span class="kn">import</span> <span class="n">RegionProposalNetwork</span><span class="p">,</span> <span class="n">RPNHead</span>
+<span class="kn">from</span> <span class="nn">.transform</span> <span class="kn">import</span> <span class="n">GeneralizedRCNNTransform</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;FasterRCNN&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FasterRCNN_ResNet50_FPN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FasterRCNN_ResNet50_FPN_V2_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FasterRCNN_MobileNet_V3_Large_FPN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FasterRCNN_MobileNet_V3_Large_320_FPN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;fasterrcnn_resnet50_fpn&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;fasterrcnn_resnet50_fpn_v2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;fasterrcnn_mobilenet_v3_large_fpn&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;fasterrcnn_mobilenet_v3_large_320_fpn&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_default_anchorgen</span><span class="p">():</span>
+    <span class="n">anchor_sizes</span> <span class="o">=</span> <span class="p">((</span><span class="mi">32</span><span class="p">,),</span> <span class="p">(</span><span class="mi">64</span><span class="p">,),</span> <span class="p">(</span><span class="mi">128</span><span class="p">,),</span> <span class="p">(</span><span class="mi">256</span><span class="p">,),</span> <span class="p">(</span><span class="mi">512</span><span class="p">,))</span>
+    <span class="n">aspect_ratios</span> <span class="o">=</span> <span class="p">((</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mf">2.0</span><span class="p">),)</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">AnchorGenerator</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">,</span> <span class="n">aspect_ratios</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">FasterRCNN</span><span class="p">(</span><span class="n">GeneralizedRCNN</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements Faster R-CNN.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each</span>
+<span class="sd">    image, and should be in 0-1 range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a Dict[Tensor] during training, containing the classification and regression</span>
+<span class="sd">    losses for both the RPN and the R-CNN.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as</span>
+<span class="sd">    follows:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the predicted labels for each image</span>
+<span class="sd">        - scores (Tensor[N]): the scores or each prediction</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            It should contain an out_channels attribute, which indicates the number of output</span>
+<span class="sd">            channels that each feature map has (and it should be the same for all feature maps).</span>
+<span class="sd">            The backbone should return a single Tensor or and OrderedDict[Tensor].</span>
+<span class="sd">        num_classes (int): number of output classes of the model (including the background).</span>
+<span class="sd">            If box_predictor is specified, num_classes should be None.</span>
+<span class="sd">        min_size (int): minimum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        max_size (int): maximum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        image_mean (Tuple[float, float, float]): mean values used for input normalization.</span>
+<span class="sd">            They are generally the mean values of the dataset on which the backbone has been trained</span>
+<span class="sd">            on</span>
+<span class="sd">        image_std (Tuple[float, float, float]): std values used for input normalization.</span>
+<span class="sd">            They are generally the std values of the dataset on which the backbone has been trained on</span>
+<span class="sd">        rpn_anchor_generator (AnchorGenerator): module that generates the anchors for a set of feature</span>
+<span class="sd">            maps.</span>
+<span class="sd">        rpn_head (nn.Module): module that computes the objectness and regression deltas from the RPN</span>
+<span class="sd">        rpn_pre_nms_top_n_train (int): number of proposals to keep before applying NMS during training</span>
+<span class="sd">        rpn_pre_nms_top_n_test (int): number of proposals to keep before applying NMS during testing</span>
+<span class="sd">        rpn_post_nms_top_n_train (int): number of proposals to keep after applying NMS during training</span>
+<span class="sd">        rpn_post_nms_top_n_test (int): number of proposals to keep after applying NMS during testing</span>
+<span class="sd">        rpn_nms_thresh (float): NMS threshold used for postprocessing the RPN proposals</span>
+<span class="sd">        rpn_fg_iou_thresh (float): minimum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training of the RPN.</span>
+<span class="sd">        rpn_bg_iou_thresh (float): maximum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as negative during training of the RPN.</span>
+<span class="sd">        rpn_batch_size_per_image (int): number of anchors that are sampled during training of the RPN</span>
+<span class="sd">            for computing the loss</span>
+<span class="sd">        rpn_positive_fraction (float): proportion of positive anchors in a mini-batch during training</span>
+<span class="sd">            of the RPN</span>
+<span class="sd">        rpn_score_thresh (float): during inference, only return proposals with a classification score</span>
+<span class="sd">            greater than rpn_score_thresh</span>
+<span class="sd">        box_roi_pool (MultiScaleRoIAlign): the module which crops and resizes the feature maps in</span>
+<span class="sd">            the locations indicated by the bounding boxes</span>
+<span class="sd">        box_head (nn.Module): module that takes the cropped feature maps as input</span>
+<span class="sd">        box_predictor (nn.Module): module that takes the output of box_head and returns the</span>
+<span class="sd">            classification logits and box regression deltas.</span>
+<span class="sd">        box_score_thresh (float): during inference, only return proposals with a classification score</span>
+<span class="sd">            greater than box_score_thresh</span>
+<span class="sd">        box_nms_thresh (float): NMS threshold for the prediction head. Used during inference</span>
+<span class="sd">        box_detections_per_img (int): maximum number of detections per image, for all classes.</span>
+<span class="sd">        box_fg_iou_thresh (float): minimum IoU between the proposals and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training of the classification head</span>
+<span class="sd">        box_bg_iou_thresh (float): maximum IoU between the proposals and the GT box so that they can be</span>
+<span class="sd">            considered as negative during training of the classification head</span>
+<span class="sd">        box_batch_size_per_image (int): number of proposals that are sampled during training of the</span>
+<span class="sd">            classification head</span>
+<span class="sd">        box_positive_fraction (float): proportion of positive proposals in a mini-batch during training</span>
+<span class="sd">            of the classification head</span>
+<span class="sd">        bbox_reg_weights (Tuple[float, float, float, float]): weights for the encoding/decoding of the</span>
+<span class="sd">            bounding boxes</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; import torch</span>
+<span class="sd">        &gt;&gt;&gt; import torchvision</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection import FasterRCNN</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection.rpn import AnchorGenerator</span>
+<span class="sd">        &gt;&gt;&gt; # load a pre-trained model for classification and return</span>
+<span class="sd">        &gt;&gt;&gt; # only the features</span>
+<span class="sd">        &gt;&gt;&gt; backbone = torchvision.models.mobilenet_v2(weights=MobileNet_V2_Weights.DEFAULT).features</span>
+<span class="sd">        &gt;&gt;&gt; # FasterRCNN needs to know the number of</span>
+<span class="sd">        &gt;&gt;&gt; # output channels in a backbone. For mobilenet_v2, it&#39;s 1280,</span>
+<span class="sd">        &gt;&gt;&gt; # so we need to add it here</span>
+<span class="sd">        &gt;&gt;&gt; backbone.out_channels = 1280</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s make the RPN generate 5 x 3 anchors per spatial</span>
+<span class="sd">        &gt;&gt;&gt; # location, with 5 different sizes and 3 different aspect</span>
+<span class="sd">        &gt;&gt;&gt; # ratios. We have a Tuple[Tuple[int]] because each feature</span>
+<span class="sd">        &gt;&gt;&gt; # map could potentially have different sizes and</span>
+<span class="sd">        &gt;&gt;&gt; # aspect ratios</span>
+<span class="sd">        &gt;&gt;&gt; anchor_generator = AnchorGenerator(sizes=((32, 64, 128, 256, 512),),</span>
+<span class="sd">        &gt;&gt;&gt;                                    aspect_ratios=((0.5, 1.0, 2.0),))</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s define what are the feature maps that we will</span>
+<span class="sd">        &gt;&gt;&gt; # use to perform the region of interest cropping, as well as</span>
+<span class="sd">        &gt;&gt;&gt; # the size of the crop after rescaling.</span>
+<span class="sd">        &gt;&gt;&gt; # if your backbone returns a Tensor, featmap_names is expected to</span>
+<span class="sd">        &gt;&gt;&gt; # be [&#39;0&#39;]. More generally, the backbone should return an</span>
+<span class="sd">        &gt;&gt;&gt; # OrderedDict[Tensor], and in featmap_names you can choose which</span>
+<span class="sd">        &gt;&gt;&gt; # feature maps to use.</span>
+<span class="sd">        &gt;&gt;&gt; roi_pooler = torchvision.ops.MultiScaleRoIAlign(featmap_names=[&#39;0&#39;],</span>
+<span class="sd">        &gt;&gt;&gt;                                                 output_size=7,</span>
+<span class="sd">        &gt;&gt;&gt;                                                 sampling_ratio=2)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # put the pieces together inside a FasterRCNN model</span>
+<span class="sd">        &gt;&gt;&gt; model = FasterRCNN(backbone,</span>
+<span class="sd">        &gt;&gt;&gt;                    num_classes=2,</span>
+<span class="sd">        &gt;&gt;&gt;                    rpn_anchor_generator=anchor_generator,</span>
+<span class="sd">        &gt;&gt;&gt;                    box_roi_pool=roi_pooler)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># transform parameters</span>
+        <span class="n">min_size</span><span class="o">=</span><span class="mi">800</span><span class="p">,</span>
+        <span class="n">max_size</span><span class="o">=</span><span class="mi">1333</span><span class="p">,</span>
+        <span class="n">image_mean</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_std</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># RPN parameters</span>
+        <span class="n">rpn_anchor_generator</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">rpn_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">rpn_pre_nms_top_n_train</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span>
+        <span class="n">rpn_pre_nms_top_n_test</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">rpn_post_nms_top_n_train</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span>
+        <span class="n">rpn_post_nms_top_n_test</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">rpn_nms_thresh</span><span class="o">=</span><span class="mf">0.7</span><span class="p">,</span>
+        <span class="n">rpn_fg_iou_thresh</span><span class="o">=</span><span class="mf">0.7</span><span class="p">,</span>
+        <span class="n">rpn_bg_iou_thresh</span><span class="o">=</span><span class="mf">0.3</span><span class="p">,</span>
+        <span class="n">rpn_batch_size_per_image</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span>
+        <span class="n">rpn_positive_fraction</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">rpn_score_thresh</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+        <span class="c1"># Box parameters</span>
+        <span class="n">box_roi_pool</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_predictor</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_score_thresh</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span>
+        <span class="n">box_nms_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_detections_per_img</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">box_fg_iou_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_bg_iou_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_batch_size_per_image</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
+        <span class="n">box_positive_fraction</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span>
+        <span class="n">bbox_reg_weights</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="s2">&quot;out_channels&quot;</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;backbone should contain an attribute out_channels &quot;</span>
+                <span class="s2">&quot;specifying the number of output channels (assumed to be the &quot;</span>
+                <span class="s2">&quot;same for all the levels)&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">rpn_anchor_generator</span><span class="p">,</span> <span class="p">(</span><span class="n">AnchorGenerator</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="kc">None</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;rpn_anchor_generator should be of type AnchorGenerator or None instead of </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">rpn_anchor_generator</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">box_roi_pool</span><span class="p">,</span> <span class="p">(</span><span class="n">MultiScaleRoIAlign</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="kc">None</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;box_roi_pool should be of type MultiScaleRoIAlign or None instead of </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">box_roi_pool</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">box_predictor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;num_classes should be None when box_predictor is specified&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">box_predictor</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;num_classes should not be None when box_predictor is not specified&quot;</span><span class="p">)</span>
+
+        <span class="n">out_channels</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span>
+
+        <span class="k">if</span> <span class="n">rpn_anchor_generator</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">rpn_anchor_generator</span> <span class="o">=</span> <span class="n">_default_anchorgen</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">rpn_head</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">rpn_head</span> <span class="o">=</span> <span class="n">RPNHead</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">rpn_anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">])</span>
+
+        <span class="n">rpn_pre_nms_top_n</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">training</span><span class="o">=</span><span class="n">rpn_pre_nms_top_n_train</span><span class="p">,</span> <span class="n">testing</span><span class="o">=</span><span class="n">rpn_pre_nms_top_n_test</span><span class="p">)</span>
+        <span class="n">rpn_post_nms_top_n</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">training</span><span class="o">=</span><span class="n">rpn_post_nms_top_n_train</span><span class="p">,</span> <span class="n">testing</span><span class="o">=</span><span class="n">rpn_post_nms_top_n_test</span><span class="p">)</span>
+
+        <span class="n">rpn</span> <span class="o">=</span> <span class="n">RegionProposalNetwork</span><span class="p">(</span>
+            <span class="n">rpn_anchor_generator</span><span class="p">,</span>
+            <span class="n">rpn_head</span><span class="p">,</span>
+            <span class="n">rpn_fg_iou_thresh</span><span class="p">,</span>
+            <span class="n">rpn_bg_iou_thresh</span><span class="p">,</span>
+            <span class="n">rpn_batch_size_per_image</span><span class="p">,</span>
+            <span class="n">rpn_positive_fraction</span><span class="p">,</span>
+            <span class="n">rpn_pre_nms_top_n</span><span class="p">,</span>
+            <span class="n">rpn_post_nms_top_n</span><span class="p">,</span>
+            <span class="n">rpn_nms_thresh</span><span class="p">,</span>
+            <span class="n">score_thresh</span><span class="o">=</span><span class="n">rpn_score_thresh</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">box_roi_pool</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">box_roi_pool</span> <span class="o">=</span> <span class="n">MultiScaleRoIAlign</span><span class="p">(</span><span class="n">featmap_names</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">,</span> <span class="s2">&quot;2&quot;</span><span class="p">,</span> <span class="s2">&quot;3&quot;</span><span class="p">],</span> <span class="n">output_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span> <span class="n">sampling_ratio</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">box_head</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">resolution</span> <span class="o">=</span> <span class="n">box_roi_pool</span><span class="o">.</span><span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">representation_size</span> <span class="o">=</span> <span class="mi">1024</span>
+            <span class="n">box_head</span> <span class="o">=</span> <span class="n">TwoMLPHead</span><span class="p">(</span><span class="n">out_channels</span> <span class="o">*</span> <span class="n">resolution</span><span class="o">**</span><span class="mi">2</span><span class="p">,</span> <span class="n">representation_size</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">box_predictor</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">representation_size</span> <span class="o">=</span> <span class="mi">1024</span>
+            <span class="n">box_predictor</span> <span class="o">=</span> <span class="n">FastRCNNPredictor</span><span class="p">(</span><span class="n">representation_size</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="n">roi_heads</span> <span class="o">=</span> <span class="n">RoIHeads</span><span class="p">(</span>
+            <span class="c1"># Box</span>
+            <span class="n">box_roi_pool</span><span class="p">,</span>
+            <span class="n">box_head</span><span class="p">,</span>
+            <span class="n">box_predictor</span><span class="p">,</span>
+            <span class="n">box_fg_iou_thresh</span><span class="p">,</span>
+            <span class="n">box_bg_iou_thresh</span><span class="p">,</span>
+            <span class="n">box_batch_size_per_image</span><span class="p">,</span>
+            <span class="n">box_positive_fraction</span><span class="p">,</span>
+            <span class="n">bbox_reg_weights</span><span class="p">,</span>
+            <span class="n">box_score_thresh</span><span class="p">,</span>
+            <span class="n">box_nms_thresh</span><span class="p">,</span>
+            <span class="n">box_detections_per_img</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">image_mean</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_mean</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">image_std</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_std</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">]</span>
+        <span class="n">transform</span> <span class="o">=</span> <span class="n">GeneralizedRCNNTransform</span><span class="p">(</span><span class="n">min_size</span><span class="p">,</span> <span class="n">max_size</span><span class="p">,</span> <span class="n">image_mean</span><span class="p">,</span> <span class="n">image_std</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">rpn</span><span class="p">,</span> <span class="n">roi_heads</span><span class="p">,</span> <span class="n">transform</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">TwoMLPHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Standard heads for FPN-based models</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of input channels</span>
+<span class="sd">        representation_size (int): size of the intermediate representation</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">representation_size</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc6</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">representation_size</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc7</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">representation_size</span><span class="p">,</span> <span class="n">representation_size</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">start_dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fc6</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fc7</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">FastRCNNConvFCHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">input_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">conv_layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">fc_layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            input_size (Tuple[int, int, int]): the input size in CHW format.</span>
+<span class="sd">            conv_layers (list): feature dimensions of each Convolution layer</span>
+<span class="sd">            fc_layers (list): feature dimensions of each FCN layer</span>
+<span class="sd">            norm_layer (callable, optional): Module specifying the normalization layer to use. Default: None</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">in_channels</span><span class="p">,</span> <span class="n">in_height</span><span class="p">,</span> <span class="n">in_width</span> <span class="o">=</span> <span class="n">input_size</span>
+
+        <span class="n">blocks</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">previous_channels</span> <span class="o">=</span> <span class="n">in_channels</span>
+        <span class="k">for</span> <span class="n">current_channels</span> <span class="ow">in</span> <span class="n">conv_layers</span><span class="p">:</span>
+            <span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">previous_channels</span><span class="p">,</span> <span class="n">current_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">))</span>
+            <span class="n">previous_channels</span> <span class="o">=</span> <span class="n">current_channels</span>
+        <span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Flatten</span><span class="p">())</span>
+        <span class="n">previous_channels</span> <span class="o">=</span> <span class="n">previous_channels</span> <span class="o">*</span> <span class="n">in_height</span> <span class="o">*</span> <span class="n">in_width</span>
+        <span class="k">for</span> <span class="n">current_channels</span> <span class="ow">in</span> <span class="n">fc_layers</span><span class="p">:</span>
+            <span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">previous_channels</span><span class="p">,</span> <span class="n">current_channels</span><span class="p">))</span>
+            <span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+            <span class="n">previous_channels</span> <span class="o">=</span> <span class="n">current_channels</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">blocks</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">layer</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">FastRCNNPredictor</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Standard classification + bounding box regression layers</span>
+<span class="sd">    for Fast R-CNN.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of input channels</span>
+<span class="sd">        num_classes (int): number of output classes (including background)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">cls_score</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bbox_pred</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_classes</span> <span class="o">*</span> <span class="mi">4</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">4</span><span class="p">:</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                <span class="nb">list</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:])</span> <span class="o">==</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+                <span class="sa">f</span><span class="s2">&quot;x has the wrong shape, expecting the last two dimensions to be [1,1] instead of </span><span class="si">{</span><span class="nb">list</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:])</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">start_dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">scores</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">cls_score</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">bbox_deltas</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bbox_pred</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">scores</span><span class="p">,</span> <span class="n">bbox_deltas</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_COCO_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="FasterRCNN_ResNet50_FPN_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights">[docs]</a><span class="k">class</span> <span class="nc">FasterRCNN_ResNet50_FPN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">41755286</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-resnet-50-fpn&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">37.0</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">134.38</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">159.743</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="FasterRCNN_ResNet50_FPN_V2_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html#torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights">[docs]</a><span class="k">class</span> <span class="nc">FasterRCNN_ResNet50_FPN_V2_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_v2_coco-dd69338a.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">43712278</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5763&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">46.7</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">280.371</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">167.104</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced using an enhanced training recipe to boost the model accuracy.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="FasterRCNN_MobileNet_V3_Large_FPN_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights">[docs]</a><span class="k">class</span> <span class="nc">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/fasterrcnn_mobilenet_v3_large_fpn-fb6a3cc7.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">19386354</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-fpn&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">32.8</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.494</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">74.239</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="FasterRCNN_MobileNet_V3_Large_320_FPN_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights">[docs]</a><span class="k">class</span> <span class="nc">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/fasterrcnn_mobilenet_v3_large_320_fpn-907ea3f9.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">19386354</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-320-fpn&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">22.8</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.719</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">74.239</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="fasterrcnn_resnet50_fpn"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">FasterRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">fasterrcnn_resnet50_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">FasterRCNN_ResNet50_FPN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FasterRCNN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Faster R-CNN model with a ResNet-50-FPN backbone from the `Faster R-CNN: Towards Real-Time Object</span>
+<span class="sd">    Detection with Region Proposal Networks &lt;https://arxiv.org/abs/1506.01497&gt;`__</span>
+<span class="sd">    paper.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each</span>
+<span class="sd">    image, and should be in ``0-1`` range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and a targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a ``Dict[Tensor]`` during training, containing the classification and regression</span>
+<span class="sd">    losses for both the RPN and the R-CNN.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a ``List[Dict[Tensor]]``, one for each input image. The fields of the ``Dict`` are as</span>
+<span class="sd">    follows, where ``N`` is the number of detections:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the predicted labels for each detection</span>
+<span class="sd">        - scores (``Tensor[N]``): the scores of each detection</span>
+
+<span class="sd">    For more details on the output, you may refer to :ref:`instance_seg_output`.</span>
+
+<span class="sd">    Faster R-CNN is exportable to ONNX for a fixed batch size with inputs images of fixed size.</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.fasterrcnn_resnet50_fpn(weights=FasterRCNN_ResNet50_FPN_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; # For training</span>
+<span class="sd">        &gt;&gt;&gt; images, boxes = torch.rand(4, 3, 600, 1200), torch.rand(4, 11, 4)</span>
+<span class="sd">        &gt;&gt;&gt; boxes[:, :, 2:4] = boxes[:, :, 0:2] + boxes[:, :, 2:4]</span>
+<span class="sd">        &gt;&gt;&gt; labels = torch.randint(1, 91, (4, 11))</span>
+<span class="sd">        &gt;&gt;&gt; images = list(image for image in images)</span>
+<span class="sd">        &gt;&gt;&gt; targets = []</span>
+<span class="sd">        &gt;&gt;&gt; for i in range(len(images)):</span>
+<span class="sd">        &gt;&gt;&gt;     d = {}</span>
+<span class="sd">        &gt;&gt;&gt;     d[&#39;boxes&#39;] = boxes[i]</span>
+<span class="sd">        &gt;&gt;&gt;     d[&#39;labels&#39;] = labels[i]</span>
+<span class="sd">        &gt;&gt;&gt;     targets.append(d)</span>
+<span class="sd">        &gt;&gt;&gt; output = model(images, targets)</span>
+<span class="sd">        &gt;&gt;&gt; # For inference</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # optionally, if you want to export the model to ONNX:</span>
+<span class="sd">        &gt;&gt;&gt; torch.onnx.export(model, x, &quot;faster_rcnn.onnx&quot;, opset_version = 11)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from</span>
+<span class="sd">            final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are</span>
+<span class="sd">            trainable. If ``None`` is passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.faster_rcnn.FasterRCNN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">FasterRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">FrozenBatchNorm2d</span> <span class="k">if</span> <span class="n">is_trained</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">FasterRCNN</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weights</span> <span class="o">==</span> <span class="n">FasterRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">:</span>
+            <span class="n">overwrite_eps</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="fasterrcnn_resnet50_fpn_v2"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html#torchvision.models.detection.fasterrcnn_resnet50_fpn_v2">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">FasterRCNN_ResNet50_FPN_V2_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">fasterrcnn_resnet50_fpn_v2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">FasterRCNN_ResNet50_FPN_V2_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FasterRCNN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs an improved Faster R-CNN model with a ResNet-50-FPN backbone from `Benchmarking Detection</span>
+<span class="sd">    Transfer Learning with Vision Transformers &lt;https://arxiv.org/abs/2111.11429&gt;`__ paper.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    It works similarly to Faster R-CNN with ResNet-50 FPN backbone. See</span>
+<span class="sd">    :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn` for more</span>
+<span class="sd">    details.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from</span>
+<span class="sd">            final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are</span>
+<span class="sd">            trainable. If ``None`` is passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.faster_rcnn.FasterRCNN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">FasterRCNN_ResNet50_FPN_V2_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">)</span>
+    <span class="n">rpn_anchor_generator</span> <span class="o">=</span> <span class="n">_default_anchorgen</span><span class="p">()</span>
+    <span class="n">rpn_head</span> <span class="o">=</span> <span class="n">RPNHead</span><span class="p">(</span><span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">rpn_anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">],</span> <span class="n">conv_depth</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+    <span class="n">box_head</span> <span class="o">=</span> <span class="n">FastRCNNConvFCHead</span><span class="p">(</span>
+        <span class="p">(</span><span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">[</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">],</span> <span class="p">[</span><span class="mi">1024</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+    <span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">FasterRCNN</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span><span class="p">,</span>
+        <span class="n">rpn_anchor_generator</span><span class="o">=</span><span class="n">rpn_anchor_generator</span><span class="p">,</span>
+        <span class="n">rpn_head</span><span class="o">=</span><span class="n">rpn_head</span><span class="p">,</span>
+        <span class="n">box_head</span><span class="o">=</span><span class="n">box_head</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<span class="k">def</span> <span class="nf">_fasterrcnn_mobilenet_v3_large_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span><span class="p">,</span> <span class="n">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span><span class="p">]],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Large_Weights</span><span class="p">],</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FasterRCNN</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">FrozenBatchNorm2d</span> <span class="k">if</span> <span class="n">is_trained</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">mobilenet_v3_large</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_mobilenet_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">)</span>
+    <span class="n">anchor_sizes</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="p">(</span>
+            <span class="mi">32</span><span class="p">,</span>
+            <span class="mi">64</span><span class="p">,</span>
+            <span class="mi">128</span><span class="p">,</span>
+            <span class="mi">256</span><span class="p">,</span>
+            <span class="mi">512</span><span class="p">,</span>
+        <span class="p">),</span>
+    <span class="p">)</span> <span class="o">*</span> <span class="mi">3</span>
+    <span class="n">aspect_ratios</span> <span class="o">=</span> <span class="p">((</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mf">2.0</span><span class="p">),)</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">FasterRCNN</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">rpn_anchor_generator</span><span class="o">=</span><span class="n">AnchorGenerator</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">,</span> <span class="n">aspect_ratios</span><span class="p">),</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<div class="viewcode-block" id="fasterrcnn_mobilenet_v3_large_320_fpn"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">fasterrcnn_mobilenet_v3_large_320_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FasterRCNN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Low resolution Faster R-CNN model with a MobileNetV3-Large backbone tuned for mobile use cases.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    It works similarly to Faster R-CNN with ResNet-50 FPN backbone. See</span>
+<span class="sd">    :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn` for more</span>
+<span class="sd">    details.</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn(weights=FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from</span>
+<span class="sd">            final block. Valid values are between 0 and 6, with 6 meaning all backbone layers are</span>
+<span class="sd">            trainable. If ``None`` is passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.faster_rcnn.FasterRCNN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="n">defaults</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="mi">320</span><span class="p">,</span>
+        <span class="s2">&quot;max_size&quot;</span><span class="p">:</span> <span class="mi">640</span><span class="p">,</span>
+        <span class="s2">&quot;rpn_pre_nms_top_n_test&quot;</span><span class="p">:</span> <span class="mi">150</span><span class="p">,</span>
+        <span class="s2">&quot;rpn_post_nms_top_n_test&quot;</span><span class="p">:</span> <span class="mi">150</span><span class="p">,</span>
+        <span class="s2">&quot;rpn_score_thresh&quot;</span><span class="p">:</span> <span class="mf">0.05</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="n">kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="o">**</span><span class="n">defaults</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">}</span>
+    <span class="k">return</span> <span class="n">_fasterrcnn_mobilenet_v3_large_fpn</span><span class="p">(</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span><span class="p">,</span>
+        <span class="n">weights_backbone</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span>
+        <span class="n">trainable_backbone_layers</span><span class="o">=</span><span class="n">trainable_backbone_layers</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="fasterrcnn_mobilenet_v3_large_fpn"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">fasterrcnn_mobilenet_v3_large_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FasterRCNN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a high resolution Faster R-CNN model with a MobileNetV3-Large FPN backbone.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    It works similarly to Faster R-CNN with ResNet-50 FPN backbone. See</span>
+<span class="sd">    :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn` for more</span>
+<span class="sd">    details.</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn(weights=FasterRCNN_MobileNet_V3_Large_FPN_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from</span>
+<span class="sd">            final block. Valid values are between 0 and 6, with 6 meaning all backbone layers are</span>
+<span class="sd">            trainable. If ``None`` is passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.faster_rcnn.FasterRCNN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="n">defaults</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;rpn_score_thresh&quot;</span><span class="p">:</span> <span class="mf">0.05</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="n">kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="o">**</span><span class="n">defaults</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">}</span>
+    <span class="k">return</span> <span class="n">_fasterrcnn_mobilenet_v3_large_fpn</span><span class="p">(</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span><span class="p">,</span>
+        <span class="n">weights_backbone</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span>
+        <span class="n">trainable_backbone_layers</span><span class="o">=</span><span class="n">trainable_backbone_layers</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/detection/fcos.html b/0.16/_modules/torchvision/models/detection/fcos.html
new file mode 100644
index 00000000000..ca11dbff4dd
--- /dev/null
+++ b/0.16/_modules/torchvision/models/detection/fcos.html
@@ -0,0 +1,1440 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.detection.fcos &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.detection.fcos</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.detection.fcos</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">...ops</span> <span class="kn">import</span> <span class="n">boxes</span> <span class="k">as</span> <span class="n">box_ops</span><span class="p">,</span> <span class="n">generalized_box_iou_loss</span><span class="p">,</span> <span class="n">misc</span> <span class="k">as</span> <span class="n">misc_nn_ops</span><span class="p">,</span> <span class="n">sigmoid_focal_loss</span>
+<span class="kn">from</span> <span class="nn">...ops.feature_pyramid_network</span> <span class="kn">import</span> <span class="n">LastLevelP6P7</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ObjectDetection</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_COCO_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..resnet</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">_utils</span> <span class="k">as</span> <span class="n">det_utils</span>
+<span class="kn">from</span> <span class="nn">.anchor_utils</span> <span class="kn">import</span> <span class="n">AnchorGenerator</span>
+<span class="kn">from</span> <span class="nn">.backbone_utils</span> <span class="kn">import</span> <span class="n">_resnet_fpn_extractor</span><span class="p">,</span> <span class="n">_validate_trainable_layers</span>
+<span class="kn">from</span> <span class="nn">.transform</span> <span class="kn">import</span> <span class="n">GeneralizedRCNNTransform</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;FCOS&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FCOS_ResNet50_FPN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;fcos_resnet50_fpn&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">FCOSHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A regression and classification head for use in FCOS.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of channels of the input feature</span>
+<span class="sd">        num_anchors (int): number of anchors to be predicted</span>
+<span class="sd">        num_classes (int): number of classes to be predicted</span>
+<span class="sd">        num_convs (Optional[int]): number of conv layer of head. Default: 4.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;box_coder&quot;</span><span class="p">:</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxLinearCoder</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_convs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="mi">4</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxLinearCoder</span><span class="p">(</span><span class="n">normalize_by_size</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span> <span class="o">=</span> <span class="n">FCOSClassificationHead</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">num_convs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span> <span class="o">=</span> <span class="n">FCOSRegressionHead</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_convs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">compute_loss</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">targets</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]],</span>
+        <span class="n">head_outputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">matched_idxs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+
+        <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">]</span>  <span class="c1"># [N, HWA, C]</span>
+        <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_regression&quot;</span><span class="p">]</span>  <span class="c1"># [N, HWA, 4]</span>
+        <span class="n">bbox_ctrness</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_ctrness&quot;</span><span class="p">]</span>  <span class="c1"># [N, HWA, 1]</span>
+
+        <span class="n">all_gt_classes_targets</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">all_gt_boxes_targets</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">targets_per_image</span><span class="p">,</span> <span class="n">matched_idxs_per_image</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">])</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">gt_classes_targets</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">((</span><span class="nb">len</span><span class="p">(</span><span class="n">matched_idxs_per_image</span><span class="p">),))</span>
+                <span class="n">gt_boxes_targets</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">((</span><span class="nb">len</span><span class="p">(</span><span class="n">matched_idxs_per_image</span><span class="p">),</span> <span class="mi">4</span><span class="p">))</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">gt_classes_targets</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">matched_idxs_per_image</span><span class="o">.</span><span class="n">clip</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">)]</span>
+                <span class="n">gt_boxes_targets</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">][</span><span class="n">matched_idxs_per_image</span><span class="o">.</span><span class="n">clip</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">)]</span>
+            <span class="n">gt_classes_targets</span><span class="p">[</span><span class="n">matched_idxs_per_image</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>  <span class="c1"># background</span>
+            <span class="n">all_gt_classes_targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">gt_classes_targets</span><span class="p">)</span>
+            <span class="n">all_gt_boxes_targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">gt_boxes_targets</span><span class="p">)</span>
+
+        <span class="c1"># List[Tensor] to Tensor conversion of  `all_gt_boxes_target`, `all_gt_classes_targets` and `anchors`</span>
+        <span class="n">all_gt_boxes_targets</span><span class="p">,</span> <span class="n">all_gt_classes_targets</span><span class="p">,</span> <span class="n">anchors</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">all_gt_boxes_targets</span><span class="p">),</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">all_gt_classes_targets</span><span class="p">),</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">anchors</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="c1"># compute foregroud</span>
+        <span class="n">foregroud_mask</span> <span class="o">=</span> <span class="n">all_gt_classes_targets</span> <span class="o">&gt;=</span> <span class="mi">0</span>
+        <span class="n">num_foreground</span> <span class="o">=</span> <span class="n">foregroud_mask</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+        <span class="c1"># classification loss</span>
+        <span class="n">gt_classes_targets</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">)</span>
+        <span class="n">gt_classes_targets</span><span class="p">[</span><span class="n">foregroud_mask</span><span class="p">,</span> <span class="n">all_gt_classes_targets</span><span class="p">[</span><span class="n">foregroud_mask</span><span class="p">]]</span> <span class="o">=</span> <span class="mf">1.0</span>
+        <span class="n">loss_cls</span> <span class="o">=</span> <span class="n">sigmoid_focal_loss</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">,</span> <span class="n">gt_classes_targets</span><span class="p">,</span> <span class="n">reduction</span><span class="o">=</span><span class="s2">&quot;sum&quot;</span><span class="p">)</span>
+
+        <span class="c1"># amp issue: pred_boxes need to convert float</span>
+        <span class="n">pred_boxes</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">bbox_regression</span><span class="p">,</span> <span class="n">anchors</span><span class="p">)</span>
+
+        <span class="c1"># regression loss: GIoU loss</span>
+        <span class="n">loss_bbox_reg</span> <span class="o">=</span> <span class="n">generalized_box_iou_loss</span><span class="p">(</span>
+            <span class="n">pred_boxes</span><span class="p">[</span><span class="n">foregroud_mask</span><span class="p">],</span>
+            <span class="n">all_gt_boxes_targets</span><span class="p">[</span><span class="n">foregroud_mask</span><span class="p">],</span>
+            <span class="n">reduction</span><span class="o">=</span><span class="s2">&quot;sum&quot;</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="c1"># ctrness loss</span>
+
+        <span class="n">bbox_reg_targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">anchors</span><span class="p">,</span> <span class="n">all_gt_boxes_targets</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">bbox_reg_targets</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">gt_ctrness_targets</span> <span class="o">=</span> <span class="n">bbox_reg_targets</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">(</span><span class="n">bbox_reg_targets</span><span class="o">.</span><span class="n">size</span><span class="p">()[:</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">left_right</span> <span class="o">=</span> <span class="n">bbox_reg_targets</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">]]</span>
+            <span class="n">top_bottom</span> <span class="o">=</span> <span class="n">bbox_reg_targets</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">]]</span>
+            <span class="n">gt_ctrness_targets</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span>
+                <span class="p">(</span><span class="n">left_right</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span> <span class="o">/</span> <span class="n">left_right</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">])</span>
+                <span class="o">*</span> <span class="p">(</span><span class="n">top_bottom</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span> <span class="o">/</span> <span class="n">top_bottom</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="n">pred_centerness</span> <span class="o">=</span> <span class="n">bbox_ctrness</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="n">loss_bbox_ctrness</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">binary_cross_entropy_with_logits</span><span class="p">(</span>
+            <span class="n">pred_centerness</span><span class="p">[</span><span class="n">foregroud_mask</span><span class="p">],</span> <span class="n">gt_ctrness_targets</span><span class="p">[</span><span class="n">foregroud_mask</span><span class="p">],</span> <span class="n">reduction</span><span class="o">=</span><span class="s2">&quot;sum&quot;</span>
+        <span class="p">)</span>
+
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;classification&quot;</span><span class="p">:</span> <span class="n">loss_cls</span> <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_foreground</span><span class="p">),</span>
+            <span class="s2">&quot;bbox_regression&quot;</span><span class="p">:</span> <span class="n">loss_bbox_reg</span> <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_foreground</span><span class="p">),</span>
+            <span class="s2">&quot;bbox_ctrness&quot;</span><span class="p">:</span> <span class="n">loss_bbox_ctrness</span> <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_foreground</span><span class="p">),</span>
+        <span class="p">}</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">cls_logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">bbox_regression</span><span class="p">,</span> <span class="n">bbox_ctrness</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;cls_logits&quot;</span><span class="p">:</span> <span class="n">cls_logits</span><span class="p">,</span>
+            <span class="s2">&quot;bbox_regression&quot;</span><span class="p">:</span> <span class="n">bbox_regression</span><span class="p">,</span>
+            <span class="s2">&quot;bbox_ctrness&quot;</span><span class="p">:</span> <span class="n">bbox_ctrness</span><span class="p">,</span>
+        <span class="p">}</span>
+
+
+<span class="k">class</span> <span class="nc">FCOSClassificationHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A classification head for use in FCOS.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of channels of the input feature.</span>
+<span class="sd">        num_anchors (int): number of anchors to be predicted.</span>
+<span class="sd">        num_classes (int): number of classes to be predicted.</span>
+<span class="sd">        num_convs (Optional[int]): number of conv layer. Default: 4.</span>
+<span class="sd">        prior_probability (Optional[float]): probability of prior. Default: 0.01.</span>
+<span class="sd">        norm_layer: Module specifying the normalization layer to use.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_anchors</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_convs</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="n">prior_probability</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.01</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_anchors</span> <span class="o">=</span> <span class="n">num_anchors</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">GroupNorm</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>
+
+        <span class="n">conv</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_convs</span><span class="p">):</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">norm_layer</span><span class="p">(</span><span class="n">in_channels</span><span class="p">))</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">())</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">conv</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="o">.</span><span class="n">children</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span> <span class="o">*</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="o">-</span><span class="n">math</span><span class="o">.</span><span class="n">log</span><span class="p">((</span><span class="mi">1</span> <span class="o">-</span> <span class="n">prior_probability</span><span class="p">)</span> <span class="o">/</span> <span class="n">prior_probability</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">all_cls_logits</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">features</span> <span class="ow">in</span> <span class="n">x</span><span class="p">:</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">)</span>
+
+            <span class="c1"># Permute classification output from (N, A * K, H, W) to (N, HWA, K).</span>
+            <span class="n">N</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">shape</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span>  <span class="c1"># Size=(N, HWA, 4)</span>
+
+            <span class="n">all_cls_logits</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">all_cls_logits</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">FCOSRegressionHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A regression head for use in FCOS, which combines regression branch and center-ness branch.</span>
+<span class="sd">    This can obtain better performance.</span>
+
+<span class="sd">    Reference: `FCOS: A simple and strong anchor-free object detector &lt;https://arxiv.org/abs/2006.09214&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of channels of the input feature</span>
+<span class="sd">        num_anchors (int): number of anchors to be predicted</span>
+<span class="sd">        num_convs (Optional[int]): number of conv layer. Default: 4.</span>
+<span class="sd">        norm_layer: Module specifying the normalization layer to use.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_anchors</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_convs</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">GroupNorm</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>
+
+        <span class="n">conv</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_convs</span><span class="p">):</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">norm_layer</span><span class="p">(</span><span class="n">in_channels</span><span class="p">))</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">())</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">conv</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">bbox_reg</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bbox_ctrness</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span> <span class="o">*</span> <span class="mi">1</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">bbox_reg</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">bbox_ctrness</span><span class="p">]:</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="o">.</span><span class="n">children</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">all_bbox_regression</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">all_bbox_ctrness</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">features</span> <span class="ow">in</span> <span class="n">x</span><span class="p">:</span>
+            <span class="n">bbox_feature</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">bbox_reg</span><span class="p">(</span><span class="n">bbox_feature</span><span class="p">))</span>
+            <span class="n">bbox_ctrness</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bbox_ctrness</span><span class="p">(</span><span class="n">bbox_feature</span><span class="p">)</span>
+
+            <span class="c1"># permute bbox regression output from (N, 4 * A, H, W) to (N, HWA, 4).</span>
+            <span class="n">N</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">shape</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>  <span class="c1"># Size=(N, HWA, 4)</span>
+            <span class="n">all_bbox_regression</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">bbox_regression</span><span class="p">)</span>
+
+            <span class="c1"># permute bbox ctrness output from (N, 1 * A, H, W) to (N, HWA, 1).</span>
+            <span class="n">bbox_ctrness</span> <span class="o">=</span> <span class="n">bbox_ctrness</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+            <span class="n">bbox_ctrness</span> <span class="o">=</span> <span class="n">bbox_ctrness</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+            <span class="n">bbox_ctrness</span> <span class="o">=</span> <span class="n">bbox_ctrness</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="n">all_bbox_ctrness</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">bbox_ctrness</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">all_bbox_regression</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">all_bbox_ctrness</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">FCOS</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements FCOS.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each</span>
+<span class="sd">    image, and should be in 0-1 range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a Dict[Tensor] during training, containing the classification, regression</span>
+<span class="sd">    and centerness losses.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as</span>
+<span class="sd">    follows:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the predicted labels for each image</span>
+<span class="sd">        - scores (Tensor[N]): the scores for each prediction</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            It should contain an out_channels attribute, which indicates the number of output</span>
+<span class="sd">            channels that each feature map has (and it should be the same for all feature maps).</span>
+<span class="sd">            The backbone should return a single Tensor or an OrderedDict[Tensor].</span>
+<span class="sd">        num_classes (int): number of output classes of the model (including the background).</span>
+<span class="sd">        min_size (int): minimum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        max_size (int): maximum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        image_mean (Tuple[float, float, float]): mean values used for input normalization.</span>
+<span class="sd">            They are generally the mean values of the dataset on which the backbone has been trained</span>
+<span class="sd">            on</span>
+<span class="sd">        image_std (Tuple[float, float, float]): std values used for input normalization.</span>
+<span class="sd">            They are generally the std values of the dataset on which the backbone has been trained on</span>
+<span class="sd">        anchor_generator (AnchorGenerator): module that generates the anchors for a set of feature</span>
+<span class="sd">            maps. For FCOS, only set one anchor for per position of each level, the width and height equal to</span>
+<span class="sd">            the stride of feature map, and set aspect ratio = 1.0, so the center of anchor is equivalent to the point</span>
+<span class="sd">            in FCOS paper.</span>
+<span class="sd">        head (nn.Module): Module run on top of the feature pyramid.</span>
+<span class="sd">            Defaults to a module containing a classification and regression module.</span>
+<span class="sd">        center_sampling_radius (int): radius of the &quot;center&quot; of a groundtruth box,</span>
+<span class="sd">            within which all anchor points are labeled positive.</span>
+<span class="sd">        score_thresh (float): Score threshold used for postprocessing the detections.</span>
+<span class="sd">        nms_thresh (float): NMS threshold used for postprocessing the detections.</span>
+<span class="sd">        detections_per_img (int): Number of best detections to keep after NMS.</span>
+<span class="sd">        topk_candidates (int): Number of best detections to keep before NMS.</span>
+
+<span class="sd">    Example:</span>
+
+<span class="sd">        &gt;&gt;&gt; import torch</span>
+<span class="sd">        &gt;&gt;&gt; import torchvision</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection import FCOS</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection.anchor_utils import AnchorGenerator</span>
+<span class="sd">        &gt;&gt;&gt; # load a pre-trained model for classification and return</span>
+<span class="sd">        &gt;&gt;&gt; # only the features</span>
+<span class="sd">        &gt;&gt;&gt; backbone = torchvision.models.mobilenet_v2(weights=MobileNet_V2_Weights.DEFAULT).features</span>
+<span class="sd">        &gt;&gt;&gt; # FCOS needs to know the number of</span>
+<span class="sd">        &gt;&gt;&gt; # output channels in a backbone. For mobilenet_v2, it&#39;s 1280,</span>
+<span class="sd">        &gt;&gt;&gt; # so we need to add it here</span>
+<span class="sd">        &gt;&gt;&gt; backbone.out_channels = 1280</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s make the network generate 5 x 3 anchors per spatial</span>
+<span class="sd">        &gt;&gt;&gt; # location, with 5 different sizes and 3 different aspect</span>
+<span class="sd">        &gt;&gt;&gt; # ratios. We have a Tuple[Tuple[int]] because each feature</span>
+<span class="sd">        &gt;&gt;&gt; # map could potentially have different sizes and</span>
+<span class="sd">        &gt;&gt;&gt; # aspect ratios</span>
+<span class="sd">        &gt;&gt;&gt; anchor_generator = AnchorGenerator(</span>
+<span class="sd">        &gt;&gt;&gt;     sizes=((8,), (16,), (32,), (64,), (128,)),</span>
+<span class="sd">        &gt;&gt;&gt;     aspect_ratios=((1.0,),)</span>
+<span class="sd">        &gt;&gt;&gt; )</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # put the pieces together inside a FCOS model</span>
+<span class="sd">        &gt;&gt;&gt; model = FCOS(</span>
+<span class="sd">        &gt;&gt;&gt;     backbone,</span>
+<span class="sd">        &gt;&gt;&gt;     num_classes=80,</span>
+<span class="sd">        &gt;&gt;&gt;     anchor_generator=anchor_generator,</span>
+<span class="sd">        &gt;&gt;&gt; )</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;box_coder&quot;</span><span class="p">:</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxLinearCoder</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">backbone</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="c1"># transform parameters</span>
+        <span class="n">min_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">800</span><span class="p">,</span>
+        <span class="n">max_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1333</span><span class="p">,</span>
+        <span class="n">image_mean</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_std</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># Anchor parameters</span>
+        <span class="n">anchor_generator</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">AnchorGenerator</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">head</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">center_sampling_radius</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.5</span><span class="p">,</span>
+        <span class="n">score_thresh</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">nms_thresh</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.6</span><span class="p">,</span>
+        <span class="n">detections_per_img</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">100</span><span class="p">,</span>
+        <span class="n">topk_candidates</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="s2">&quot;out_channels&quot;</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;backbone should contain an attribute out_channels &quot;</span>
+                <span class="s2">&quot;specifying the number of output channels (assumed to be the &quot;</span>
+                <span class="s2">&quot;same for all the levels)&quot;</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">anchor_generator</span><span class="p">,</span> <span class="p">(</span><span class="n">AnchorGenerator</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="kc">None</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;anchor_generator should be of type AnchorGenerator or None, instead  got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">anchor_generator</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">anchor_generator</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">anchor_sizes</span> <span class="o">=</span> <span class="p">((</span><span class="mi">8</span><span class="p">,),</span> <span class="p">(</span><span class="mi">16</span><span class="p">,),</span> <span class="p">(</span><span class="mi">32</span><span class="p">,),</span> <span class="p">(</span><span class="mi">64</span><span class="p">,),</span> <span class="p">(</span><span class="mi">128</span><span class="p">,))</span>  <span class="c1"># equal to strides of multi-level feature map</span>
+            <span class="n">aspect_ratios</span> <span class="o">=</span> <span class="p">((</span><span class="mf">1.0</span><span class="p">,),)</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">)</span>  <span class="c1"># set only one anchor</span>
+            <span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">AnchorGenerator</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">,</span> <span class="n">aspect_ratios</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">anchor_generator</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;anchor_generator.num_anchors_per_location()[0] should be 1 instead of </span><span class="si">{</span><span class="n">anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">head</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">head</span> <span class="o">=</span> <span class="n">FCOSHead</span><span class="p">(</span><span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">],</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head</span> <span class="o">=</span> <span class="n">head</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxLinearCoder</span><span class="p">(</span><span class="n">normalize_by_size</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">image_mean</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_mean</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">image_std</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_std</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">GeneralizedRCNNTransform</span><span class="p">(</span><span class="n">min_size</span><span class="p">,</span> <span class="n">max_size</span><span class="p">,</span> <span class="n">image_mean</span><span class="p">,</span> <span class="n">image_std</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">center_sampling_radius</span> <span class="o">=</span> <span class="n">center_sampling_radius</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">score_thresh</span> <span class="o">=</span> <span class="n">score_thresh</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">nms_thresh</span> <span class="o">=</span> <span class="n">nms_thresh</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">detections_per_img</span> <span class="o">=</span> <span class="n">detections_per_img</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">topk_candidates</span> <span class="o">=</span> <span class="n">topk_candidates</span>
+
+        <span class="c1"># used only on torchscript mode</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">eager_outputs</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">losses</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">losses</span>
+
+        <span class="k">return</span> <span class="n">detections</span>
+
+    <span class="k">def</span> <span class="nf">compute_loss</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">targets</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]],</span>
+        <span class="n">head_outputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">num_anchors_per_level</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">matched_idxs</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">anchors_per_image</span><span class="p">,</span> <span class="n">targets_per_image</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">anchors</span><span class="p">,</span> <span class="n">targets</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">matched_idxs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">anchors_per_image</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),),</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">anchors_per_image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="p">)</span>
+                <span class="k">continue</span>
+
+            <span class="n">gt_boxes</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span>
+            <span class="n">gt_centers</span> <span class="o">=</span> <span class="p">(</span><span class="n">gt_boxes</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span> <span class="o">+</span> <span class="n">gt_boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:])</span> <span class="o">/</span> <span class="mi">2</span>  <span class="c1"># Nx2</span>
+            <span class="n">anchor_centers</span> <span class="o">=</span> <span class="p">(</span><span class="n">anchors_per_image</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span> <span class="o">+</span> <span class="n">anchors_per_image</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:])</span> <span class="o">/</span> <span class="mi">2</span>  <span class="c1"># N</span>
+            <span class="n">anchor_sizes</span> <span class="o">=</span> <span class="n">anchors_per_image</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="n">anchors_per_image</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span>
+            <span class="c1"># center sampling: anchor point must be close enough to gt center.</span>
+            <span class="n">pairwise_match</span> <span class="o">=</span> <span class="p">(</span><span class="n">anchor_centers</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">-</span> <span class="n">gt_centers</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:])</span><span class="o">.</span><span class="n">abs_</span><span class="p">()</span><span class="o">.</span><span class="n">max</span><span class="p">(</span>
+                <span class="n">dim</span><span class="o">=</span><span class="mi">2</span>
+            <span class="p">)</span><span class="o">.</span><span class="n">values</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">center_sampling_radius</span> <span class="o">*</span> <span class="n">anchor_sizes</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span>
+            <span class="c1"># compute pairwise distance between N points and M boxes</span>
+            <span class="n">x</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">anchor_centers</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># (N, 1)</span>
+            <span class="n">x0</span><span class="p">,</span> <span class="n">y0</span><span class="p">,</span> <span class="n">x1</span><span class="p">,</span> <span class="n">y1</span> <span class="o">=</span> <span class="n">gt_boxes</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>  <span class="c1"># (1, M)</span>
+            <span class="n">pairwise_dist</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">([</span><span class="n">x</span> <span class="o">-</span> <span class="n">x0</span><span class="p">,</span> <span class="n">y</span> <span class="o">-</span> <span class="n">y0</span><span class="p">,</span> <span class="n">x1</span> <span class="o">-</span> <span class="n">x</span><span class="p">,</span> <span class="n">y1</span> <span class="o">-</span> <span class="n">y</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>  <span class="c1"># (N, M)</span>
+
+            <span class="c1"># anchor point must be inside gt</span>
+            <span class="n">pairwise_match</span> <span class="o">&amp;=</span> <span class="n">pairwise_dist</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">values</span> <span class="o">&gt;</span> <span class="mi">0</span>
+
+            <span class="c1"># each anchor is only responsible for certain scale range.</span>
+            <span class="n">lower_bound</span> <span class="o">=</span> <span class="n">anchor_sizes</span> <span class="o">*</span> <span class="mi">4</span>
+            <span class="n">lower_bound</span><span class="p">[:</span> <span class="n">num_anchors_per_level</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="n">upper_bound</span> <span class="o">=</span> <span class="n">anchor_sizes</span> <span class="o">*</span> <span class="mi">8</span>
+            <span class="n">upper_bound</span><span class="p">[</span><span class="o">-</span><span class="n">num_anchors_per_level</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="p">:]</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)</span>
+            <span class="n">pairwise_dist</span> <span class="o">=</span> <span class="n">pairwise_dist</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">values</span>
+            <span class="n">pairwise_match</span> <span class="o">&amp;=</span> <span class="p">(</span><span class="n">pairwise_dist</span> <span class="o">&gt;</span> <span class="n">lower_bound</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">])</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">pairwise_dist</span> <span class="o">&lt;</span> <span class="n">upper_bound</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">])</span>
+
+            <span class="c1"># match the GT box with minimum area, if there are multiple GT matches</span>
+            <span class="n">gt_areas</span> <span class="o">=</span> <span class="p">(</span><span class="n">gt_boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="n">gt_boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">])</span> <span class="o">*</span> <span class="p">(</span><span class="n">gt_boxes</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">-</span> <span class="n">gt_boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">])</span>  <span class="c1"># N</span>
+            <span class="n">pairwise_match</span> <span class="o">=</span> <span class="n">pairwise_match</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mf">1e8</span> <span class="o">-</span> <span class="n">gt_areas</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+            <span class="n">min_values</span><span class="p">,</span> <span class="n">matched_idx</span> <span class="o">=</span> <span class="n">pairwise_match</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># R, per-anchor match</span>
+            <span class="n">matched_idx</span><span class="p">[</span><span class="n">min_values</span> <span class="o">&lt;</span> <span class="mf">1e-5</span><span class="p">]</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>  <span class="c1"># unmatched anchors are assigned -1</span>
+
+            <span class="n">matched_idxs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">matched_idx</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">postprocess_detections</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]],</span> <span class="n">anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]],</span> <span class="n">image_shapes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]:</span>
+        <span class="n">class_logits</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">]</span>
+        <span class="n">box_regression</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_regression&quot;</span><span class="p">]</span>
+        <span class="n">box_ctrness</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_ctrness&quot;</span><span class="p">]</span>
+
+        <span class="n">num_images</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">image_shapes</span><span class="p">)</span>
+
+        <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">index</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_images</span><span class="p">):</span>
+            <span class="n">box_regression_per_image</span> <span class="o">=</span> <span class="p">[</span><span class="n">br</span><span class="p">[</span><span class="n">index</span><span class="p">]</span> <span class="k">for</span> <span class="n">br</span> <span class="ow">in</span> <span class="n">box_regression</span><span class="p">]</span>
+            <span class="n">logits_per_image</span> <span class="o">=</span> <span class="p">[</span><span class="n">cl</span><span class="p">[</span><span class="n">index</span><span class="p">]</span> <span class="k">for</span> <span class="n">cl</span> <span class="ow">in</span> <span class="n">class_logits</span><span class="p">]</span>
+            <span class="n">box_ctrness_per_image</span> <span class="o">=</span> <span class="p">[</span><span class="n">bc</span><span class="p">[</span><span class="n">index</span><span class="p">]</span> <span class="k">for</span> <span class="n">bc</span> <span class="ow">in</span> <span class="n">box_ctrness</span><span class="p">]</span>
+            <span class="n">anchors_per_image</span><span class="p">,</span> <span class="n">image_shape</span> <span class="o">=</span> <span class="n">anchors</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="n">image_shapes</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+            <span class="n">image_boxes</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">image_scores</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">image_labels</span> <span class="o">=</span> <span class="p">[]</span>
+
+            <span class="k">for</span> <span class="n">box_regression_per_level</span><span class="p">,</span> <span class="n">logits_per_level</span><span class="p">,</span> <span class="n">box_ctrness_per_level</span><span class="p">,</span> <span class="n">anchors_per_level</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span>
+                <span class="n">box_regression_per_image</span><span class="p">,</span> <span class="n">logits_per_image</span><span class="p">,</span> <span class="n">box_ctrness_per_image</span><span class="p">,</span> <span class="n">anchors_per_image</span>
+            <span class="p">):</span>
+                <span class="n">num_classes</span> <span class="o">=</span> <span class="n">logits_per_level</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+                <span class="c1"># remove low scoring boxes</span>
+                <span class="n">scores_per_level</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="n">logits_per_level</span><span class="p">)</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="n">box_ctrness_per_level</span><span class="p">)</span>
+                <span class="p">)</span><span class="o">.</span><span class="n">flatten</span><span class="p">()</span>
+                <span class="n">keep_idxs</span> <span class="o">=</span> <span class="n">scores_per_level</span> <span class="o">&gt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">score_thresh</span>
+                <span class="n">scores_per_level</span> <span class="o">=</span> <span class="n">scores_per_level</span><span class="p">[</span><span class="n">keep_idxs</span><span class="p">]</span>
+                <span class="n">topk_idxs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">keep_idxs</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+
+                <span class="c1"># keep only topk scoring predictions</span>
+                <span class="n">num_topk</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">_topk_min</span><span class="p">(</span><span class="n">topk_idxs</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">topk_candidates</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="n">scores_per_level</span><span class="p">,</span> <span class="n">idxs</span> <span class="o">=</span> <span class="n">scores_per_level</span><span class="o">.</span><span class="n">topk</span><span class="p">(</span><span class="n">num_topk</span><span class="p">)</span>
+                <span class="n">topk_idxs</span> <span class="o">=</span> <span class="n">topk_idxs</span><span class="p">[</span><span class="n">idxs</span><span class="p">]</span>
+
+                <span class="n">anchor_idxs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">div</span><span class="p">(</span><span class="n">topk_idxs</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">rounding_mode</span><span class="o">=</span><span class="s2">&quot;floor&quot;</span><span class="p">)</span>
+                <span class="n">labels_per_level</span> <span class="o">=</span> <span class="n">topk_idxs</span> <span class="o">%</span> <span class="n">num_classes</span>
+
+                <span class="n">boxes_per_level</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span>
+                    <span class="n">box_regression_per_level</span><span class="p">[</span><span class="n">anchor_idxs</span><span class="p">],</span> <span class="n">anchors_per_level</span><span class="p">[</span><span class="n">anchor_idxs</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">boxes_per_level</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">clip_boxes_to_image</span><span class="p">(</span><span class="n">boxes_per_level</span><span class="p">,</span> <span class="n">image_shape</span><span class="p">)</span>
+
+                <span class="n">image_boxes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">boxes_per_level</span><span class="p">)</span>
+                <span class="n">image_scores</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">scores_per_level</span><span class="p">)</span>
+                <span class="n">image_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">labels_per_level</span><span class="p">)</span>
+
+            <span class="n">image_boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_boxes</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_scores</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_scores</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_labels</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+            <span class="c1"># non-maximum suppression</span>
+            <span class="n">keep</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">batched_nms</span><span class="p">(</span><span class="n">image_boxes</span><span class="p">,</span> <span class="n">image_scores</span><span class="p">,</span> <span class="n">image_labels</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">nms_thresh</span><span class="p">)</span>
+            <span class="n">keep</span> <span class="o">=</span> <span class="n">keep</span><span class="p">[:</span> <span class="bp">self</span><span class="o">.</span><span class="n">detections_per_img</span><span class="p">]</span>
+
+            <span class="n">detections</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;boxes&quot;</span><span class="p">:</span> <span class="n">image_boxes</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                    <span class="s2">&quot;scores&quot;</span><span class="p">:</span> <span class="n">image_scores</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                    <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="n">image_labels</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">detections</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">images</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">targets</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            images (list[Tensor]): images to be processed</span>
+<span class="sd">            targets (list[Dict[Tensor]]): ground-truth boxes present in the image (optional)</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            result (list[BoxList] or dict[Tensor]): the output from the model.</span>
+<span class="sd">                During training, it returns a dict[Tensor] which contains the losses.</span>
+<span class="sd">                During testing, it returns list[BoxList] contains additional fields</span>
+<span class="sd">                like `scores`, `labels` and `mask` (for Mask R-CNN models).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+
+            <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;targets should not be none when in training mode&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">target</span> <span class="ow">in</span> <span class="n">targets</span><span class="p">:</span>
+                    <span class="n">boxes</span> <span class="o">=</span> <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="s2">&quot;Expected target boxes to be of type Tensor.&quot;</span><span class="p">)</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                        <span class="nb">len</span><span class="p">(</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">4</span><span class="p">,</span>
+                        <span class="sa">f</span><span class="s2">&quot;Expected target boxes to be a tensor of shape [N, 4], got </span><span class="si">{</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">,</span>
+                    <span class="p">)</span>
+
+        <span class="n">original_image_sizes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">img</span> <span class="ow">in</span> <span class="n">images</span><span class="p">:</span>
+            <span class="n">val</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                <span class="nb">len</span><span class="p">(</span><span class="n">val</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">,</span>
+                <span class="sa">f</span><span class="s2">&quot;expecting the last two dimensions of the Tensor to be H and W instead got </span><span class="si">{</span><span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">original_image_sizes</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">val</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">val</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+
+        <span class="c1"># transform the input</span>
+        <span class="n">images</span><span class="p">,</span> <span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">targets</span><span class="p">)</span>
+
+        <span class="c1"># Check for degenerate boxes</span>
+        <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">target_idx</span><span class="p">,</span> <span class="n">target</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">targets</span><span class="p">):</span>
+                <span class="n">boxes</span> <span class="o">=</span> <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span>
+                <span class="n">degenerate_boxes</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:]</span> <span class="o">&lt;=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">degenerate_boxes</span><span class="o">.</span><span class="n">any</span><span class="p">():</span>
+                    <span class="c1"># print the first degenerate box</span>
+                    <span class="n">bb_idx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">degenerate_boxes</span><span class="o">.</span><span class="n">any</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">))[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                    <span class="n">degen_bb</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[</span><span class="n">bb_idx</span><span class="p">]</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                        <span class="kc">False</span><span class="p">,</span>
+                        <span class="sa">f</span><span class="s2">&quot;All bounding boxes should have positive height and width. Found invalid box </span><span class="si">{</span><span class="n">degen_bb</span><span class="si">}</span><span class="s2"> for target at index </span><span class="si">{</span><span class="n">target_idx</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">,</span>
+                    <span class="p">)</span>
+
+        <span class="c1"># get the features from the backbone</span>
+        <span class="n">features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span><span class="p">(</span><span class="n">images</span><span class="o">.</span><span class="n">tensors</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">features</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="n">features</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">([(</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="n">features</span><span class="p">)])</span>
+
+        <span class="n">features</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">features</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+
+        <span class="c1"># compute the fcos heads outputs using the features</span>
+        <span class="n">head_outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+
+        <span class="c1"># create the set of anchors</span>
+        <span class="n">anchors</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">features</span><span class="p">)</span>
+        <span class="c1"># recover level sizes</span>
+        <span class="n">num_anchors_per_level</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span> <span class="o">*</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">3</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">features</span><span class="p">]</span>
+
+        <span class="n">losses</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;targets should not be none when in training mode&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="c1"># compute the losses</span>
+                <span class="n">losses</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">num_anchors_per_level</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># split outputs per level</span>
+            <span class="n">split_head_outputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">{}</span>
+            <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">head_outputs</span><span class="p">:</span>
+                <span class="n">split_head_outputs</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">head_outputs</span><span class="p">[</span><span class="n">k</span><span class="p">]</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">num_anchors_per_level</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+            <span class="n">split_anchors</span> <span class="o">=</span> <span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">a</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">num_anchors_per_level</span><span class="p">))</span> <span class="k">for</span> <span class="n">a</span> <span class="ow">in</span> <span class="n">anchors</span><span class="p">]</span>
+
+            <span class="c1"># compute the detections</span>
+            <span class="n">detections</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">postprocess_detections</span><span class="p">(</span><span class="n">split_head_outputs</span><span class="p">,</span> <span class="n">split_anchors</span><span class="p">,</span> <span class="n">images</span><span class="o">.</span><span class="n">image_sizes</span><span class="p">)</span>
+            <span class="n">detections</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="o">.</span><span class="n">postprocess</span><span class="p">(</span><span class="n">detections</span><span class="p">,</span> <span class="n">images</span><span class="o">.</span><span class="n">image_sizes</span><span class="p">,</span> <span class="n">original_image_sizes</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;FCOS always returns a (Losses, Detections) tuple in scripting&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span> <span class="o">=</span> <span class="kc">True</span>
+            <span class="k">return</span> <span class="n">losses</span><span class="p">,</span> <span class="n">detections</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">eager_outputs</span><span class="p">(</span><span class="n">losses</span><span class="p">,</span> <span class="n">detections</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="FCOS_ResNet50_FPN_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fcos_resnet50_fpn.html#torchvision.models.detection.FCOS_ResNet50_FPN_Weights">[docs]</a><span class="k">class</span> <span class="nc">FCOS_ResNet50_FPN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/fcos_resnet50_fpn_coco-99b0c9b7.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">32269600</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_COCO_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#fcos-resnet-50-fpn&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">39.2</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">128.207</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">123.608</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="fcos_resnet50_fpn"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.fcos_resnet50_fpn.html#torchvision.models.detection.fcos_resnet50_fpn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">FCOS_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">fcos_resnet50_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">FCOS_ResNet50_FPN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FCOS</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a FCOS model with a ResNet-50-FPN backbone.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    Reference: `FCOS: Fully Convolutional One-Stage Object Detection &lt;https://arxiv.org/abs/1904.01355&gt;`_.</span>
+<span class="sd">               `FCOS: A simple and strong anchor-free object detector &lt;https://arxiv.org/abs/2006.09214&gt;`_.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each</span>
+<span class="sd">    image, and should be in ``0-1`` range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a ``Dict[Tensor]`` during training, containing the classification and regression</span>
+<span class="sd">    losses.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a ``List[Dict[Tensor]]``, one for each input image. The fields of the ``Dict`` are as</span>
+<span class="sd">    follows, where ``N`` is the number of detections:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the predicted labels for each detection</span>
+<span class="sd">        - scores (``Tensor[N]``): the scores of each detection</span>
+
+<span class="sd">    For more details on the output, you may refer to :ref:`instance_seg_output`.</span>
+
+<span class="sd">    Example:</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.fcos_resnet50_fpn(weights=FCOS_ResNet50_FPN_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.FCOS_ResNet50_FPN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.FCOS_ResNet50_FPN_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The pretrained weights for</span>
+<span class="sd">            the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) resnet layers starting</span>
+<span class="sd">            from final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are</span>
+<span class="sd">            trainable. If ``None`` is passed (the default) this value is set to 3. Default: None</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.FCOS``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/fcos.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.FCOS_ResNet50_FPN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">FCOS_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">FrozenBatchNorm2d</span> <span class="k">if</span> <span class="n">is_trained</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="n">returned_layers</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="n">extra_blocks</span><span class="o">=</span><span class="n">LastLevelP6P7</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">)</span>
+    <span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">FCOS</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html b/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
new file mode 100644
index 00000000000..d397c36fc83
--- /dev/null
+++ b/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
@@ -0,0 +1,1141 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.detection.keypoint_rcnn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.detection.keypoint_rcnn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.detection.keypoint_rcnn</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops</span> <span class="kn">import</span> <span class="n">MultiScaleRoIAlign</span>
+
+<span class="kn">from</span> <span class="nn">...ops</span> <span class="kn">import</span> <span class="n">misc</span> <span class="k">as</span> <span class="n">misc_nn_ops</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ObjectDetection</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_COCO_PERSON_CATEGORIES</span><span class="p">,</span> <span class="n">_COCO_PERSON_KEYPOINT_NAMES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..resnet</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">overwrite_eps</span>
+<span class="kn">from</span> <span class="nn">.backbone_utils</span> <span class="kn">import</span> <span class="n">_resnet_fpn_extractor</span><span class="p">,</span> <span class="n">_validate_trainable_layers</span>
+<span class="kn">from</span> <span class="nn">.faster_rcnn</span> <span class="kn">import</span> <span class="n">FasterRCNN</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;KeypointRCNN&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;KeypointRCNN_ResNet50_FPN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;keypointrcnn_resnet50_fpn&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">KeypointRCNN</span><span class="p">(</span><span class="n">FasterRCNN</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements Keypoint R-CNN.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each</span>
+<span class="sd">    image, and should be in 0-1 range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">            ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the class label for each ground-truth box</span>
+<span class="sd">        - keypoints (FloatTensor[N, K, 3]): the K keypoints location for each of the N instances, in the</span>
+<span class="sd">          format [x, y, visibility], where visibility=0 means that the keypoint is not visible.</span>
+
+<span class="sd">    The model returns a Dict[Tensor] during training, containing the classification and regression</span>
+<span class="sd">    losses for both the RPN and the R-CNN, and the keypoint loss.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as</span>
+<span class="sd">    follows:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">            ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the predicted labels for each image</span>
+<span class="sd">        - scores (Tensor[N]): the scores or each prediction</span>
+<span class="sd">        - keypoints (FloatTensor[N, K, 3]): the locations of the predicted keypoints, in [x, y, v] format.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            It should contain an out_channels attribute, which indicates the number of output</span>
+<span class="sd">            channels that each feature map has (and it should be the same for all feature maps).</span>
+<span class="sd">            The backbone should return a single Tensor or and OrderedDict[Tensor].</span>
+<span class="sd">        num_classes (int): number of output classes of the model (including the background).</span>
+<span class="sd">            If box_predictor is specified, num_classes should be None.</span>
+<span class="sd">        min_size (int): minimum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        max_size (int): maximum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        image_mean (Tuple[float, float, float]): mean values used for input normalization.</span>
+<span class="sd">            They are generally the mean values of the dataset on which the backbone has been trained</span>
+<span class="sd">            on</span>
+<span class="sd">        image_std (Tuple[float, float, float]): std values used for input normalization.</span>
+<span class="sd">            They are generally the std values of the dataset on which the backbone has been trained on</span>
+<span class="sd">        rpn_anchor_generator (AnchorGenerator): module that generates the anchors for a set of feature</span>
+<span class="sd">            maps.</span>
+<span class="sd">        rpn_head (nn.Module): module that computes the objectness and regression deltas from the RPN</span>
+<span class="sd">        rpn_pre_nms_top_n_train (int): number of proposals to keep before applying NMS during training</span>
+<span class="sd">        rpn_pre_nms_top_n_test (int): number of proposals to keep before applying NMS during testing</span>
+<span class="sd">        rpn_post_nms_top_n_train (int): number of proposals to keep after applying NMS during training</span>
+<span class="sd">        rpn_post_nms_top_n_test (int): number of proposals to keep after applying NMS during testing</span>
+<span class="sd">        rpn_nms_thresh (float): NMS threshold used for postprocessing the RPN proposals</span>
+<span class="sd">        rpn_fg_iou_thresh (float): minimum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training of the RPN.</span>
+<span class="sd">        rpn_bg_iou_thresh (float): maximum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as negative during training of the RPN.</span>
+<span class="sd">        rpn_batch_size_per_image (int): number of anchors that are sampled during training of the RPN</span>
+<span class="sd">            for computing the loss</span>
+<span class="sd">        rpn_positive_fraction (float): proportion of positive anchors in a mini-batch during training</span>
+<span class="sd">            of the RPN</span>
+<span class="sd">        rpn_score_thresh (float): during inference, only return proposals with a classification score</span>
+<span class="sd">            greater than rpn_score_thresh</span>
+<span class="sd">        box_roi_pool (MultiScaleRoIAlign): the module which crops and resizes the feature maps in</span>
+<span class="sd">            the locations indicated by the bounding boxes</span>
+<span class="sd">        box_head (nn.Module): module that takes the cropped feature maps as input</span>
+<span class="sd">        box_predictor (nn.Module): module that takes the output of box_head and returns the</span>
+<span class="sd">            classification logits and box regression deltas.</span>
+<span class="sd">        box_score_thresh (float): during inference, only return proposals with a classification score</span>
+<span class="sd">            greater than box_score_thresh</span>
+<span class="sd">        box_nms_thresh (float): NMS threshold for the prediction head. Used during inference</span>
+<span class="sd">        box_detections_per_img (int): maximum number of detections per image, for all classes.</span>
+<span class="sd">        box_fg_iou_thresh (float): minimum IoU between the proposals and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training of the classification head</span>
+<span class="sd">        box_bg_iou_thresh (float): maximum IoU between the proposals and the GT box so that they can be</span>
+<span class="sd">            considered as negative during training of the classification head</span>
+<span class="sd">        box_batch_size_per_image (int): number of proposals that are sampled during training of the</span>
+<span class="sd">            classification head</span>
+<span class="sd">        box_positive_fraction (float): proportion of positive proposals in a mini-batch during training</span>
+<span class="sd">            of the classification head</span>
+<span class="sd">        bbox_reg_weights (Tuple[float, float, float, float]): weights for the encoding/decoding of the</span>
+<span class="sd">            bounding boxes</span>
+<span class="sd">        keypoint_roi_pool (MultiScaleRoIAlign): the module which crops and resizes the feature maps in</span>
+<span class="sd">             the locations indicated by the bounding boxes, which will be used for the keypoint head.</span>
+<span class="sd">        keypoint_head (nn.Module): module that takes the cropped feature maps as input</span>
+<span class="sd">        keypoint_predictor (nn.Module): module that takes the output of the keypoint_head and returns the</span>
+<span class="sd">            heatmap logits</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; import torch</span>
+<span class="sd">        &gt;&gt;&gt; import torchvision</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection import KeypointRCNN</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection.anchor_utils import AnchorGenerator</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # load a pre-trained model for classification and return</span>
+<span class="sd">        &gt;&gt;&gt; # only the features</span>
+<span class="sd">        &gt;&gt;&gt; backbone = torchvision.models.mobilenet_v2(weights=MobileNet_V2_Weights.DEFAULT).features</span>
+<span class="sd">        &gt;&gt;&gt; # KeypointRCNN needs to know the number of</span>
+<span class="sd">        &gt;&gt;&gt; # output channels in a backbone. For mobilenet_v2, it&#39;s 1280,</span>
+<span class="sd">        &gt;&gt;&gt; # so we need to add it here</span>
+<span class="sd">        &gt;&gt;&gt; backbone.out_channels = 1280</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s make the RPN generate 5 x 3 anchors per spatial</span>
+<span class="sd">        &gt;&gt;&gt; # location, with 5 different sizes and 3 different aspect</span>
+<span class="sd">        &gt;&gt;&gt; # ratios. We have a Tuple[Tuple[int]] because each feature</span>
+<span class="sd">        &gt;&gt;&gt; # map could potentially have different sizes and</span>
+<span class="sd">        &gt;&gt;&gt; # aspect ratios</span>
+<span class="sd">        &gt;&gt;&gt; anchor_generator = AnchorGenerator(sizes=((32, 64, 128, 256, 512),),</span>
+<span class="sd">        &gt;&gt;&gt;                                    aspect_ratios=((0.5, 1.0, 2.0),))</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s define what are the feature maps that we will</span>
+<span class="sd">        &gt;&gt;&gt; # use to perform the region of interest cropping, as well as</span>
+<span class="sd">        &gt;&gt;&gt; # the size of the crop after rescaling.</span>
+<span class="sd">        &gt;&gt;&gt; # if your backbone returns a Tensor, featmap_names is expected to</span>
+<span class="sd">        &gt;&gt;&gt; # be [&#39;0&#39;]. More generally, the backbone should return an</span>
+<span class="sd">        &gt;&gt;&gt; # OrderedDict[Tensor], and in featmap_names you can choose which</span>
+<span class="sd">        &gt;&gt;&gt; # feature maps to use.</span>
+<span class="sd">        &gt;&gt;&gt; roi_pooler = torchvision.ops.MultiScaleRoIAlign(featmap_names=[&#39;0&#39;],</span>
+<span class="sd">        &gt;&gt;&gt;                                                 output_size=7,</span>
+<span class="sd">        &gt;&gt;&gt;                                                 sampling_ratio=2)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; keypoint_roi_pooler = torchvision.ops.MultiScaleRoIAlign(featmap_names=[&#39;0&#39;],</span>
+<span class="sd">        &gt;&gt;&gt;                                                          output_size=14,</span>
+<span class="sd">        &gt;&gt;&gt;                                                          sampling_ratio=2)</span>
+<span class="sd">        &gt;&gt;&gt; # put the pieces together inside a KeypointRCNN model</span>
+<span class="sd">        &gt;&gt;&gt; model = KeypointRCNN(backbone,</span>
+<span class="sd">        &gt;&gt;&gt;                      num_classes=2,</span>
+<span class="sd">        &gt;&gt;&gt;                      rpn_anchor_generator=anchor_generator,</span>
+<span class="sd">        &gt;&gt;&gt;                      box_roi_pool=roi_pooler,</span>
+<span class="sd">        &gt;&gt;&gt;                      keypoint_roi_pool=keypoint_roi_pooler)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># transform parameters</span>
+        <span class="n">min_size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">max_size</span><span class="o">=</span><span class="mi">1333</span><span class="p">,</span>
+        <span class="n">image_mean</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_std</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># RPN parameters</span>
+        <span class="n">rpn_anchor_generator</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">rpn_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">rpn_pre_nms_top_n_train</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span>
+        <span class="n">rpn_pre_nms_top_n_test</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">rpn_post_nms_top_n_train</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span>
+        <span class="n">rpn_post_nms_top_n_test</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">rpn_nms_thresh</span><span class="o">=</span><span class="mf">0.7</span><span class="p">,</span>
+        <span class="n">rpn_fg_iou_thresh</span><span class="o">=</span><span class="mf">0.7</span><span class="p">,</span>
+        <span class="n">rpn_bg_iou_thresh</span><span class="o">=</span><span class="mf">0.3</span><span class="p">,</span>
+        <span class="n">rpn_batch_size_per_image</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span>
+        <span class="n">rpn_positive_fraction</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">rpn_score_thresh</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+        <span class="c1"># Box parameters</span>
+        <span class="n">box_roi_pool</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_predictor</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_score_thresh</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span>
+        <span class="n">box_nms_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_detections_per_img</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">box_fg_iou_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_bg_iou_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_batch_size_per_image</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
+        <span class="n">box_positive_fraction</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span>
+        <span class="n">bbox_reg_weights</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># keypoint parameters</span>
+        <span class="n">keypoint_roi_pool</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">keypoint_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">keypoint_predictor</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">num_keypoints</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">keypoint_roi_pool</span><span class="p">,</span> <span class="p">(</span><span class="n">MultiScaleRoIAlign</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="kc">None</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="s2">&quot;keypoint_roi_pool should be of type MultiScaleRoIAlign or None instead of {type(keypoint_roi_pool)}&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">min_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">min_size</span> <span class="o">=</span> <span class="p">(</span><span class="mi">640</span><span class="p">,</span> <span class="mi">672</span><span class="p">,</span> <span class="mi">704</span><span class="p">,</span> <span class="mi">736</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">800</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">num_keypoints</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">keypoint_predictor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;num_keypoints should be None when keypoint_predictor is specified&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">num_keypoints</span> <span class="o">=</span> <span class="mi">17</span>
+
+        <span class="n">out_channels</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span>
+
+        <span class="k">if</span> <span class="n">keypoint_roi_pool</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">keypoint_roi_pool</span> <span class="o">=</span> <span class="n">MultiScaleRoIAlign</span><span class="p">(</span><span class="n">featmap_names</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">,</span> <span class="s2">&quot;2&quot;</span><span class="p">,</span> <span class="s2">&quot;3&quot;</span><span class="p">],</span> <span class="n">output_size</span><span class="o">=</span><span class="mi">14</span><span class="p">,</span> <span class="n">sampling_ratio</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">keypoint_head</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">keypoint_layers</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="mi">512</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">8</span><span class="p">))</span>
+            <span class="n">keypoint_head</span> <span class="o">=</span> <span class="n">KeypointRCNNHeads</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">keypoint_layers</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">keypoint_predictor</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">keypoint_dim_reduced</span> <span class="o">=</span> <span class="mi">512</span>  <span class="c1"># == keypoint_layers[-1]</span>
+            <span class="n">keypoint_predictor</span> <span class="o">=</span> <span class="n">KeypointRCNNPredictor</span><span class="p">(</span><span class="n">keypoint_dim_reduced</span><span class="p">,</span> <span class="n">num_keypoints</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">backbone</span><span class="p">,</span>
+            <span class="n">num_classes</span><span class="p">,</span>
+            <span class="c1"># transform parameters</span>
+            <span class="n">min_size</span><span class="p">,</span>
+            <span class="n">max_size</span><span class="p">,</span>
+            <span class="n">image_mean</span><span class="p">,</span>
+            <span class="n">image_std</span><span class="p">,</span>
+            <span class="c1"># RPN-specific parameters</span>
+            <span class="n">rpn_anchor_generator</span><span class="p">,</span>
+            <span class="n">rpn_head</span><span class="p">,</span>
+            <span class="n">rpn_pre_nms_top_n_train</span><span class="p">,</span>
+            <span class="n">rpn_pre_nms_top_n_test</span><span class="p">,</span>
+            <span class="n">rpn_post_nms_top_n_train</span><span class="p">,</span>
+            <span class="n">rpn_post_nms_top_n_test</span><span class="p">,</span>
+            <span class="n">rpn_nms_thresh</span><span class="p">,</span>
+            <span class="n">rpn_fg_iou_thresh</span><span class="p">,</span>
+            <span class="n">rpn_bg_iou_thresh</span><span class="p">,</span>
+            <span class="n">rpn_batch_size_per_image</span><span class="p">,</span>
+            <span class="n">rpn_positive_fraction</span><span class="p">,</span>
+            <span class="n">rpn_score_thresh</span><span class="p">,</span>
+            <span class="c1"># Box parameters</span>
+            <span class="n">box_roi_pool</span><span class="p">,</span>
+            <span class="n">box_head</span><span class="p">,</span>
+            <span class="n">box_predictor</span><span class="p">,</span>
+            <span class="n">box_score_thresh</span><span class="p">,</span>
+            <span class="n">box_nms_thresh</span><span class="p">,</span>
+            <span class="n">box_detections_per_img</span><span class="p">,</span>
+            <span class="n">box_fg_iou_thresh</span><span class="p">,</span>
+            <span class="n">box_bg_iou_thresh</span><span class="p">,</span>
+            <span class="n">box_batch_size_per_image</span><span class="p">,</span>
+            <span class="n">box_positive_fraction</span><span class="p">,</span>
+            <span class="n">bbox_reg_weights</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">roi_heads</span><span class="o">.</span><span class="n">keypoint_roi_pool</span> <span class="o">=</span> <span class="n">keypoint_roi_pool</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">roi_heads</span><span class="o">.</span><span class="n">keypoint_head</span> <span class="o">=</span> <span class="n">keypoint_head</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">roi_heads</span><span class="o">.</span><span class="n">keypoint_predictor</span> <span class="o">=</span> <span class="n">keypoint_predictor</span>
+
+
+<span class="k">class</span> <span class="nc">KeypointRCNNHeads</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">layers</span><span class="p">):</span>
+        <span class="n">d</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">next_feature</span> <span class="o">=</span> <span class="n">in_channels</span>
+        <span class="k">for</span> <span class="n">out_channels</span> <span class="ow">in</span> <span class="n">layers</span><span class="p">:</span>
+            <span class="n">d</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">next_feature</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+            <span class="n">d</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+            <span class="n">next_feature</span> <span class="o">=</span> <span class="n">out_channels</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">d</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">children</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">KeypointRCNNPredictor</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">num_keypoints</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">input_features</span> <span class="o">=</span> <span class="n">in_channels</span>
+        <span class="n">deconv_kernel</span> <span class="o">=</span> <span class="mi">4</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">kps_score_lowres</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ConvTranspose2d</span><span class="p">(</span>
+            <span class="n">input_features</span><span class="p">,</span>
+            <span class="n">num_keypoints</span><span class="p">,</span>
+            <span class="n">deconv_kernel</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="n">deconv_kernel</span> <span class="o">//</span> <span class="mi">2</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">kps_score_lowres</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+        <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">kps_score_lowres</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">up_scale</span> <span class="o">=</span> <span class="mi">2</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">num_keypoints</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">kps_score_lowres</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">interpolate</span><span class="p">(</span>
+            <span class="n">x</span><span class="p">,</span> <span class="n">scale_factor</span><span class="o">=</span><span class="nb">float</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">up_scale</span><span class="p">),</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;bilinear&quot;</span><span class="p">,</span> <span class="n">align_corners</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">recompute_scale_factor</span><span class="o">=</span><span class="kc">False</span>
+        <span class="p">)</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_COCO_PERSON_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;keypoint_names&quot;</span><span class="p">:</span> <span class="n">_COCO_PERSON_KEYPOINT_NAMES</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="KeypointRCNN_ResNet50_FPN_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights">[docs]</a><span class="k">class</span> <span class="nc">KeypointRCNN_ResNet50_FPN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_LEGACY</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/keypointrcnn_resnet50_fpn_coco-9f466800.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">59137258</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/1606&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">50.6</span><span class="p">,</span>
+                    <span class="s2">&quot;kp_map&quot;</span><span class="p">:</span> <span class="mf">61.1</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">133.924</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">226.054</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were produced by following a similar training recipe as on the paper but use a checkpoint</span>
+<span class="s2">                from an early epoch.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/keypointrcnn_resnet50_fpn_coco-fc266e95.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">59137258</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#keypoint-r-cnn&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">54.6</span><span class="p">,</span>
+                    <span class="s2">&quot;kp_map&quot;</span><span class="p">:</span> <span class="mf">65.0</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">137.42</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">226.054</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="keypointrcnn_resnet50_fpn"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.keypointrcnn_resnet50_fpn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">KeypointRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_LEGACY</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;pretrained&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="s2">&quot;legacy&quot;</span>
+        <span class="k">else</span> <span class="n">KeypointRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">,</span>
+    <span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">keypointrcnn_resnet50_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">KeypointRCNN_ResNet50_FPN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">num_keypoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">KeypointRCNN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a Keypoint R-CNN model with a ResNet-50-FPN backbone.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    Reference: `Mask R-CNN &lt;https://arxiv.org/abs/1703.06870&gt;`__.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each</span>
+<span class="sd">    image, and should be in ``0-1`` range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the class label for each ground-truth box</span>
+<span class="sd">        - keypoints (``FloatTensor[N, K, 3]``): the ``K`` keypoints location for each of the ``N`` instances, in the</span>
+<span class="sd">          format ``[x, y, visibility]``, where ``visibility=0`` means that the keypoint is not visible.</span>
+
+<span class="sd">    The model returns a ``Dict[Tensor]`` during training, containing the classification and regression</span>
+<span class="sd">    losses for both the RPN and the R-CNN, and the keypoint loss.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a ``List[Dict[Tensor]]``, one for each input image. The fields of the ``Dict`` are as</span>
+<span class="sd">    follows, where ``N`` is the number of detected instances:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the predicted labels for each instance</span>
+<span class="sd">        - scores (``Tensor[N]``): the scores or each instance</span>
+<span class="sd">        - keypoints (``FloatTensor[N, K, 3]``): the locations of the predicted keypoints, in ``[x, y, v]`` format.</span>
+
+<span class="sd">    For more details on the output, you may refer to :ref:`instance_seg_output`.</span>
+
+<span class="sd">    Keypoint R-CNN is exportable to ONNX for a fixed batch size with inputs images of fixed size.</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.keypointrcnn_resnet50_fpn(weights=KeypointRCNN_ResNet50_FPN_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # optionally, if you want to export the model to ONNX:</span>
+<span class="sd">        &gt;&gt;&gt; torch.onnx.export(model, x, &quot;keypoint_rcnn.onnx&quot;, opset_version = 11)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        num_keypoints (int, optional): number of keypoints</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from final block.</span>
+<span class="sd">            Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If ``None`` is</span>
+<span class="sd">            passed (the default) this value is set to 3.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">KeypointRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="n">num_keypoints</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_keypoints&quot;</span><span class="p">,</span> <span class="n">num_keypoints</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;keypoint_names&quot;</span><span class="p">]))</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">2</span>
+        <span class="k">if</span> <span class="n">num_keypoints</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">num_keypoints</span> <span class="o">=</span> <span class="mi">17</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">FrozenBatchNorm2d</span> <span class="k">if</span> <span class="n">is_trained</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">KeypointRCNN</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">num_keypoints</span><span class="o">=</span><span class="n">num_keypoints</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weights</span> <span class="o">==</span> <span class="n">KeypointRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">:</span>
+            <span class="n">overwrite_eps</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/detection/mask_rcnn.html b/0.16/_modules/torchvision/models/detection/mask_rcnn.html
new file mode 100644
index 00000000000..143e1dff660
--- /dev/null
+++ b/0.16/_modules/torchvision/models/detection/mask_rcnn.html
@@ -0,0 +1,1256 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.detection.mask_rcnn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.detection.mask_rcnn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.detection.mask_rcnn</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops</span> <span class="kn">import</span> <span class="n">MultiScaleRoIAlign</span>
+
+<span class="kn">from</span> <span class="nn">...ops</span> <span class="kn">import</span> <span class="n">misc</span> <span class="k">as</span> <span class="n">misc_nn_ops</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ObjectDetection</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_COCO_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..resnet</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">overwrite_eps</span>
+<span class="kn">from</span> <span class="nn">.backbone_utils</span> <span class="kn">import</span> <span class="n">_resnet_fpn_extractor</span><span class="p">,</span> <span class="n">_validate_trainable_layers</span>
+<span class="kn">from</span> <span class="nn">.faster_rcnn</span> <span class="kn">import</span> <span class="n">_default_anchorgen</span><span class="p">,</span> <span class="n">FasterRCNN</span><span class="p">,</span> <span class="n">FastRCNNConvFCHead</span><span class="p">,</span> <span class="n">RPNHead</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;MaskRCNN&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MaskRCNN_ResNet50_FPN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MaskRCNN_ResNet50_FPN_V2_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;maskrcnn_resnet50_fpn&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;maskrcnn_resnet50_fpn_v2&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">MaskRCNN</span><span class="p">(</span><span class="n">FasterRCNN</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements Mask R-CNN.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each</span>
+<span class="sd">    image, and should be in 0-1 range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the class label for each ground-truth box</span>
+<span class="sd">        - masks (UInt8Tensor[N, H, W]): the segmentation binary masks for each instance</span>
+
+<span class="sd">    The model returns a Dict[Tensor] during training, containing the classification and regression</span>
+<span class="sd">    losses for both the RPN and the R-CNN, and the mask loss.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as</span>
+<span class="sd">    follows:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the predicted labels for each image</span>
+<span class="sd">        - scores (Tensor[N]): the scores or each prediction</span>
+<span class="sd">        - masks (UInt8Tensor[N, 1, H, W]): the predicted masks for each instance, in 0-1 range. In order to</span>
+<span class="sd">          obtain the final segmentation masks, the soft masks can be thresholded, generally</span>
+<span class="sd">          with a value of 0.5 (mask &gt;= 0.5)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            It should contain an out_channels attribute, which indicates the number of output</span>
+<span class="sd">            channels that each feature map has (and it should be the same for all feature maps).</span>
+<span class="sd">            The backbone should return a single Tensor or and OrderedDict[Tensor].</span>
+<span class="sd">        num_classes (int): number of output classes of the model (including the background).</span>
+<span class="sd">            If box_predictor is specified, num_classes should be None.</span>
+<span class="sd">        min_size (int): minimum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        max_size (int): maximum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        image_mean (Tuple[float, float, float]): mean values used for input normalization.</span>
+<span class="sd">            They are generally the mean values of the dataset on which the backbone has been trained</span>
+<span class="sd">            on</span>
+<span class="sd">        image_std (Tuple[float, float, float]): std values used for input normalization.</span>
+<span class="sd">            They are generally the std values of the dataset on which the backbone has been trained on</span>
+<span class="sd">        rpn_anchor_generator (AnchorGenerator): module that generates the anchors for a set of feature</span>
+<span class="sd">            maps.</span>
+<span class="sd">        rpn_head (nn.Module): module that computes the objectness and regression deltas from the RPN</span>
+<span class="sd">        rpn_pre_nms_top_n_train (int): number of proposals to keep before applying NMS during training</span>
+<span class="sd">        rpn_pre_nms_top_n_test (int): number of proposals to keep before applying NMS during testing</span>
+<span class="sd">        rpn_post_nms_top_n_train (int): number of proposals to keep after applying NMS during training</span>
+<span class="sd">        rpn_post_nms_top_n_test (int): number of proposals to keep after applying NMS during testing</span>
+<span class="sd">        rpn_nms_thresh (float): NMS threshold used for postprocessing the RPN proposals</span>
+<span class="sd">        rpn_fg_iou_thresh (float): minimum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training of the RPN.</span>
+<span class="sd">        rpn_bg_iou_thresh (float): maximum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as negative during training of the RPN.</span>
+<span class="sd">        rpn_batch_size_per_image (int): number of anchors that are sampled during training of the RPN</span>
+<span class="sd">            for computing the loss</span>
+<span class="sd">        rpn_positive_fraction (float): proportion of positive anchors in a mini-batch during training</span>
+<span class="sd">            of the RPN</span>
+<span class="sd">        rpn_score_thresh (float): during inference, only return proposals with a classification score</span>
+<span class="sd">            greater than rpn_score_thresh</span>
+<span class="sd">        box_roi_pool (MultiScaleRoIAlign): the module which crops and resizes the feature maps in</span>
+<span class="sd">            the locations indicated by the bounding boxes</span>
+<span class="sd">        box_head (nn.Module): module that takes the cropped feature maps as input</span>
+<span class="sd">        box_predictor (nn.Module): module that takes the output of box_head and returns the</span>
+<span class="sd">            classification logits and box regression deltas.</span>
+<span class="sd">        box_score_thresh (float): during inference, only return proposals with a classification score</span>
+<span class="sd">            greater than box_score_thresh</span>
+<span class="sd">        box_nms_thresh (float): NMS threshold for the prediction head. Used during inference</span>
+<span class="sd">        box_detections_per_img (int): maximum number of detections per image, for all classes.</span>
+<span class="sd">        box_fg_iou_thresh (float): minimum IoU between the proposals and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training of the classification head</span>
+<span class="sd">        box_bg_iou_thresh (float): maximum IoU between the proposals and the GT box so that they can be</span>
+<span class="sd">            considered as negative during training of the classification head</span>
+<span class="sd">        box_batch_size_per_image (int): number of proposals that are sampled during training of the</span>
+<span class="sd">            classification head</span>
+<span class="sd">        box_positive_fraction (float): proportion of positive proposals in a mini-batch during training</span>
+<span class="sd">            of the classification head</span>
+<span class="sd">        bbox_reg_weights (Tuple[float, float, float, float]): weights for the encoding/decoding of the</span>
+<span class="sd">            bounding boxes</span>
+<span class="sd">        mask_roi_pool (MultiScaleRoIAlign): the module which crops and resizes the feature maps in</span>
+<span class="sd">             the locations indicated by the bounding boxes, which will be used for the mask head.</span>
+<span class="sd">        mask_head (nn.Module): module that takes the cropped feature maps as input</span>
+<span class="sd">        mask_predictor (nn.Module): module that takes the output of the mask_head and returns the</span>
+<span class="sd">            segmentation mask logits</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; import torch</span>
+<span class="sd">        &gt;&gt;&gt; import torchvision</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection import MaskRCNN</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection.anchor_utils import AnchorGenerator</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # load a pre-trained model for classification and return</span>
+<span class="sd">        &gt;&gt;&gt; # only the features</span>
+<span class="sd">        &gt;&gt;&gt; backbone = torchvision.models.mobilenet_v2(weights=MobileNet_V2_Weights.DEFAULT).features</span>
+<span class="sd">        &gt;&gt;&gt; # MaskRCNN needs to know the number of</span>
+<span class="sd">        &gt;&gt;&gt; # output channels in a backbone. For mobilenet_v2, it&#39;s 1280</span>
+<span class="sd">        &gt;&gt;&gt; # so we need to add it here,</span>
+<span class="sd">        &gt;&gt;&gt; backbone.out_channels = 1280</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s make the RPN generate 5 x 3 anchors per spatial</span>
+<span class="sd">        &gt;&gt;&gt; # location, with 5 different sizes and 3 different aspect</span>
+<span class="sd">        &gt;&gt;&gt; # ratios. We have a Tuple[Tuple[int]] because each feature</span>
+<span class="sd">        &gt;&gt;&gt; # map could potentially have different sizes and</span>
+<span class="sd">        &gt;&gt;&gt; # aspect ratios</span>
+<span class="sd">        &gt;&gt;&gt; anchor_generator = AnchorGenerator(sizes=((32, 64, 128, 256, 512),),</span>
+<span class="sd">        &gt;&gt;&gt;                                    aspect_ratios=((0.5, 1.0, 2.0),))</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s define what are the feature maps that we will</span>
+<span class="sd">        &gt;&gt;&gt; # use to perform the region of interest cropping, as well as</span>
+<span class="sd">        &gt;&gt;&gt; # the size of the crop after rescaling.</span>
+<span class="sd">        &gt;&gt;&gt; # if your backbone returns a Tensor, featmap_names is expected to</span>
+<span class="sd">        &gt;&gt;&gt; # be [&#39;0&#39;]. More generally, the backbone should return an</span>
+<span class="sd">        &gt;&gt;&gt; # OrderedDict[Tensor], and in featmap_names you can choose which</span>
+<span class="sd">        &gt;&gt;&gt; # feature maps to use.</span>
+<span class="sd">        &gt;&gt;&gt; roi_pooler = torchvision.ops.MultiScaleRoIAlign(featmap_names=[&#39;0&#39;],</span>
+<span class="sd">        &gt;&gt;&gt;                                                 output_size=7,</span>
+<span class="sd">        &gt;&gt;&gt;                                                 sampling_ratio=2)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; mask_roi_pooler = torchvision.ops.MultiScaleRoIAlign(featmap_names=[&#39;0&#39;],</span>
+<span class="sd">        &gt;&gt;&gt;                                                      output_size=14,</span>
+<span class="sd">        &gt;&gt;&gt;                                                      sampling_ratio=2)</span>
+<span class="sd">        &gt;&gt;&gt; # put the pieces together inside a MaskRCNN model</span>
+<span class="sd">        &gt;&gt;&gt; model = MaskRCNN(backbone,</span>
+<span class="sd">        &gt;&gt;&gt;                  num_classes=2,</span>
+<span class="sd">        &gt;&gt;&gt;                  rpn_anchor_generator=anchor_generator,</span>
+<span class="sd">        &gt;&gt;&gt;                  box_roi_pool=roi_pooler,</span>
+<span class="sd">        &gt;&gt;&gt;                  mask_roi_pool=mask_roi_pooler)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># transform parameters</span>
+        <span class="n">min_size</span><span class="o">=</span><span class="mi">800</span><span class="p">,</span>
+        <span class="n">max_size</span><span class="o">=</span><span class="mi">1333</span><span class="p">,</span>
+        <span class="n">image_mean</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_std</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># RPN parameters</span>
+        <span class="n">rpn_anchor_generator</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">rpn_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">rpn_pre_nms_top_n_train</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span>
+        <span class="n">rpn_pre_nms_top_n_test</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">rpn_post_nms_top_n_train</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span>
+        <span class="n">rpn_post_nms_top_n_test</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">rpn_nms_thresh</span><span class="o">=</span><span class="mf">0.7</span><span class="p">,</span>
+        <span class="n">rpn_fg_iou_thresh</span><span class="o">=</span><span class="mf">0.7</span><span class="p">,</span>
+        <span class="n">rpn_bg_iou_thresh</span><span class="o">=</span><span class="mf">0.3</span><span class="p">,</span>
+        <span class="n">rpn_batch_size_per_image</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span>
+        <span class="n">rpn_positive_fraction</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">rpn_score_thresh</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+        <span class="c1"># Box parameters</span>
+        <span class="n">box_roi_pool</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_predictor</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">box_score_thresh</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span>
+        <span class="n">box_nms_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_detections_per_img</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">box_fg_iou_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_bg_iou_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">box_batch_size_per_image</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
+        <span class="n">box_positive_fraction</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span>
+        <span class="n">bbox_reg_weights</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># Mask parameters</span>
+        <span class="n">mask_roi_pool</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">mask_head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">mask_predictor</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">mask_roi_pool</span><span class="p">,</span> <span class="p">(</span><span class="n">MultiScaleRoIAlign</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="kc">None</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;mask_roi_pool should be of type MultiScaleRoIAlign or None instead of </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">mask_roi_pool</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">mask_predictor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;num_classes should be None when mask_predictor is specified&quot;</span><span class="p">)</span>
+
+        <span class="n">out_channels</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span>
+
+        <span class="k">if</span> <span class="n">mask_roi_pool</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">mask_roi_pool</span> <span class="o">=</span> <span class="n">MultiScaleRoIAlign</span><span class="p">(</span><span class="n">featmap_names</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">,</span> <span class="s2">&quot;2&quot;</span><span class="p">,</span> <span class="s2">&quot;3&quot;</span><span class="p">],</span> <span class="n">output_size</span><span class="o">=</span><span class="mi">14</span><span class="p">,</span> <span class="n">sampling_ratio</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">mask_head</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">mask_layers</span> <span class="o">=</span> <span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">)</span>
+            <span class="n">mask_dilation</span> <span class="o">=</span> <span class="mi">1</span>
+            <span class="n">mask_head</span> <span class="o">=</span> <span class="n">MaskRCNNHeads</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">mask_layers</span><span class="p">,</span> <span class="n">mask_dilation</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">mask_predictor</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">mask_predictor_in_channels</span> <span class="o">=</span> <span class="mi">256</span>  <span class="c1"># == mask_layers[-1]</span>
+            <span class="n">mask_dim_reduced</span> <span class="o">=</span> <span class="mi">256</span>
+            <span class="n">mask_predictor</span> <span class="o">=</span> <span class="n">MaskRCNNPredictor</span><span class="p">(</span><span class="n">mask_predictor_in_channels</span><span class="p">,</span> <span class="n">mask_dim_reduced</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">backbone</span><span class="p">,</span>
+            <span class="n">num_classes</span><span class="p">,</span>
+            <span class="c1"># transform parameters</span>
+            <span class="n">min_size</span><span class="p">,</span>
+            <span class="n">max_size</span><span class="p">,</span>
+            <span class="n">image_mean</span><span class="p">,</span>
+            <span class="n">image_std</span><span class="p">,</span>
+            <span class="c1"># RPN-specific parameters</span>
+            <span class="n">rpn_anchor_generator</span><span class="p">,</span>
+            <span class="n">rpn_head</span><span class="p">,</span>
+            <span class="n">rpn_pre_nms_top_n_train</span><span class="p">,</span>
+            <span class="n">rpn_pre_nms_top_n_test</span><span class="p">,</span>
+            <span class="n">rpn_post_nms_top_n_train</span><span class="p">,</span>
+            <span class="n">rpn_post_nms_top_n_test</span><span class="p">,</span>
+            <span class="n">rpn_nms_thresh</span><span class="p">,</span>
+            <span class="n">rpn_fg_iou_thresh</span><span class="p">,</span>
+            <span class="n">rpn_bg_iou_thresh</span><span class="p">,</span>
+            <span class="n">rpn_batch_size_per_image</span><span class="p">,</span>
+            <span class="n">rpn_positive_fraction</span><span class="p">,</span>
+            <span class="n">rpn_score_thresh</span><span class="p">,</span>
+            <span class="c1"># Box parameters</span>
+            <span class="n">box_roi_pool</span><span class="p">,</span>
+            <span class="n">box_head</span><span class="p">,</span>
+            <span class="n">box_predictor</span><span class="p">,</span>
+            <span class="n">box_score_thresh</span><span class="p">,</span>
+            <span class="n">box_nms_thresh</span><span class="p">,</span>
+            <span class="n">box_detections_per_img</span><span class="p">,</span>
+            <span class="n">box_fg_iou_thresh</span><span class="p">,</span>
+            <span class="n">box_bg_iou_thresh</span><span class="p">,</span>
+            <span class="n">box_batch_size_per_image</span><span class="p">,</span>
+            <span class="n">box_positive_fraction</span><span class="p">,</span>
+            <span class="n">bbox_reg_weights</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">roi_heads</span><span class="o">.</span><span class="n">mask_roi_pool</span> <span class="o">=</span> <span class="n">mask_roi_pool</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">roi_heads</span><span class="o">.</span><span class="n">mask_head</span> <span class="o">=</span> <span class="n">mask_head</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">roi_heads</span><span class="o">.</span><span class="n">mask_predictor</span> <span class="o">=</span> <span class="n">mask_predictor</span>
+
+
+<span class="k">class</span> <span class="nc">MaskRCNNHeads</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">layers</span><span class="p">,</span> <span class="n">dilation</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            in_channels (int): number of input channels</span>
+<span class="sd">            layers (list): feature dimensions of each FCN layer</span>
+<span class="sd">            dilation (int): dilation rate of kernel</span>
+<span class="sd">            norm_layer (callable, optional): Module specifying the normalization layer to use. Default: None</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">blocks</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">next_feature</span> <span class="o">=</span> <span class="n">in_channels</span>
+        <span class="k">for</span> <span class="n">layer_features</span> <span class="ow">in</span> <span class="n">layers</span><span class="p">:</span>
+            <span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                    <span class="n">next_feature</span><span class="p">,</span>
+                    <span class="n">layer_features</span><span class="p">,</span>
+                    <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                    <span class="n">padding</span><span class="o">=</span><span class="n">dilation</span><span class="p">,</span>
+                    <span class="n">dilation</span><span class="o">=</span><span class="n">dilation</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+            <span class="n">next_feature</span> <span class="o">=</span> <span class="n">layer_features</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">blocks</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">layer</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">,</span>
+        <span class="n">unexpected_keys</span><span class="p">,</span>
+        <span class="n">error_msgs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">version</span> <span class="o">=</span> <span class="n">local_metadata</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">version</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">version</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">num_blocks</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_blocks</span><span class="p">):</span>
+                <span class="k">for</span> <span class="nb">type</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;weight&quot;</span><span class="p">,</span> <span class="s2">&quot;bias&quot;</span><span class="p">]:</span>
+                    <span class="n">old_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">mask_fcn</span><span class="si">{</span><span class="n">i</span><span class="o">+</span><span class="mi">1</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="n">new_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}{</span><span class="n">i</span><span class="si">}</span><span class="s2">.0.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="k">if</span> <span class="n">old_key</span> <span class="ow">in</span> <span class="n">state_dict</span><span class="p">:</span>
+                        <span class="n">state_dict</span><span class="p">[</span><span class="n">new_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">state_dict</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">old_key</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span>
+            <span class="n">prefix</span><span class="p">,</span>
+            <span class="n">local_metadata</span><span class="p">,</span>
+            <span class="n">strict</span><span class="p">,</span>
+            <span class="n">missing_keys</span><span class="p">,</span>
+            <span class="n">unexpected_keys</span><span class="p">,</span>
+            <span class="n">error_msgs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">MaskRCNNPredictor</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">dim_reduced</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">OrderedDict</span><span class="p">(</span>
+                <span class="p">[</span>
+                    <span class="p">(</span><span class="s2">&quot;conv5_mask&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">ConvTranspose2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">dim_reduced</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+                    <span class="p">(</span><span class="s2">&quot;relu&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)),</span>
+                    <span class="p">(</span><span class="s2">&quot;mask_fcn_logits&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">dim_reduced</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+                <span class="p">]</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">param</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">():</span>
+            <span class="k">if</span> <span class="s2">&quot;weight&quot;</span> <span class="ow">in</span> <span class="n">name</span><span class="p">:</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">param</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+            <span class="c1"># elif &quot;bias&quot; in name:</span>
+            <span class="c1">#     nn.init.constant_(param, 0)</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_COCO_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="MaskRCNN_ResNet50_FPN_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html#torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights">[docs]</a><span class="k">class</span> <span class="nc">MaskRCNN_ResNet50_FPN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">44401393</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#mask-r-cnn&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">37.9</span><span class="p">,</span>
+                    <span class="s2">&quot;mask_map&quot;</span><span class="p">:</span> <span class="mf">34.6</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">134.38</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">169.84</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="MaskRCNN_ResNet50_FPN_V2_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html#torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights">[docs]</a><span class="k">class</span> <span class="nc">MaskRCNN_ResNet50_FPN_V2_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/maskrcnn_resnet50_fpn_v2_coco-73cbd019.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">46359409</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5773&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">47.4</span><span class="p">,</span>
+                    <span class="s2">&quot;mask_map&quot;</span><span class="p">:</span> <span class="mf">41.8</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">333.577</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">177.219</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced using an enhanced training recipe to boost the model accuracy.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="maskrcnn_resnet50_fpn"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html#torchvision.models.detection.maskrcnn_resnet50_fpn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MaskRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">maskrcnn_resnet50_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MaskRCNN_ResNet50_FPN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MaskRCNN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Mask R-CNN model with a ResNet-50-FPN backbone from the `Mask R-CNN</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1703.06870&gt;`_ paper.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each</span>
+<span class="sd">    image, and should be in ``0-1`` range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the class label for each ground-truth box</span>
+<span class="sd">        - masks (``UInt8Tensor[N, H, W]``): the segmentation binary masks for each instance</span>
+
+<span class="sd">    The model returns a ``Dict[Tensor]`` during training, containing the classification and regression</span>
+<span class="sd">    losses for both the RPN and the R-CNN, and the mask loss.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a ``List[Dict[Tensor]]``, one for each input image. The fields of the ``Dict`` are as</span>
+<span class="sd">    follows, where ``N`` is the number of detected instances:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the predicted labels for each instance</span>
+<span class="sd">        - scores (``Tensor[N]``): the scores or each instance</span>
+<span class="sd">        - masks (``UInt8Tensor[N, 1, H, W]``): the predicted masks for each instance, in ``0-1`` range. In order to</span>
+<span class="sd">          obtain the final segmentation masks, the soft masks can be thresholded, generally</span>
+<span class="sd">          with a value of 0.5 (``mask &gt;= 0.5``)</span>
+
+<span class="sd">    For more details on the output and on how to plot the masks, you may refer to :ref:`instance_seg_output`.</span>
+
+<span class="sd">    Mask R-CNN is exportable to ONNX for a fixed batch size with inputs images of fixed size.</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.maskrcnn_resnet50_fpn(weights=MaskRCNN_ResNet50_FPN_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # optionally, if you want to export the model to ONNX:</span>
+<span class="sd">        &gt;&gt;&gt; torch.onnx.export(model, x, &quot;mask_rcnn.onnx&quot;, opset_version = 11)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from</span>
+<span class="sd">            final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are</span>
+<span class="sd">            trainable. If ``None`` is passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.mask_rcnn.MaskRCNN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MaskRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">FrozenBatchNorm2d</span> <span class="k">if</span> <span class="n">is_trained</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">MaskRCNN</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weights</span> <span class="o">==</span> <span class="n">MaskRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">:</span>
+            <span class="n">overwrite_eps</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="maskrcnn_resnet50_fpn_v2"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html#torchvision.models.detection.maskrcnn_resnet50_fpn_v2">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MaskRCNN_ResNet50_FPN_V2_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">maskrcnn_resnet50_fpn_v2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MaskRCNN_ResNet50_FPN_V2_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MaskRCNN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Improved Mask R-CNN model with a ResNet-50-FPN backbone from the `Benchmarking Detection Transfer</span>
+<span class="sd">    Learning with Vision Transformers &lt;https://arxiv.org/abs/2111.11429&gt;`_ paper.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    :func:`~torchvision.models.detection.maskrcnn_resnet50_fpn` for more details.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from</span>
+<span class="sd">            final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are</span>
+<span class="sd">            trainable. If ``None`` is passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.mask_rcnn.MaskRCNN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MaskRCNN_ResNet50_FPN_V2_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">)</span>
+    <span class="n">rpn_anchor_generator</span> <span class="o">=</span> <span class="n">_default_anchorgen</span><span class="p">()</span>
+    <span class="n">rpn_head</span> <span class="o">=</span> <span class="n">RPNHead</span><span class="p">(</span><span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">rpn_anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">],</span> <span class="n">conv_depth</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+    <span class="n">box_head</span> <span class="o">=</span> <span class="n">FastRCNNConvFCHead</span><span class="p">(</span>
+        <span class="p">(</span><span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">[</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">],</span> <span class="p">[</span><span class="mi">1024</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+    <span class="p">)</span>
+    <span class="n">mask_head</span> <span class="o">=</span> <span class="n">MaskRCNNHeads</span><span class="p">(</span><span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="p">[</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">MaskRCNN</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span><span class="p">,</span>
+        <span class="n">rpn_anchor_generator</span><span class="o">=</span><span class="n">rpn_anchor_generator</span><span class="p">,</span>
+        <span class="n">rpn_head</span><span class="o">=</span><span class="n">rpn_head</span><span class="p">,</span>
+        <span class="n">box_head</span><span class="o">=</span><span class="n">box_head</span><span class="p">,</span>
+        <span class="n">mask_head</span><span class="o">=</span><span class="n">mask_head</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/detection/retinanet.html b/0.16/_modules/torchvision/models/detection/retinanet.html
new file mode 100644
index 00000000000..badf85db7eb
--- /dev/null
+++ b/0.16/_modules/torchvision/models/detection/retinanet.html
@@ -0,0 +1,1568 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.detection.retinanet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.detection.retinanet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.detection.retinanet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">...ops</span> <span class="kn">import</span> <span class="n">boxes</span> <span class="k">as</span> <span class="n">box_ops</span><span class="p">,</span> <span class="n">misc</span> <span class="k">as</span> <span class="n">misc_nn_ops</span><span class="p">,</span> <span class="n">sigmoid_focal_loss</span>
+<span class="kn">from</span> <span class="nn">...ops.feature_pyramid_network</span> <span class="kn">import</span> <span class="n">LastLevelP6P7</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ObjectDetection</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_COCO_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..resnet</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">_utils</span> <span class="k">as</span> <span class="n">det_utils</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_box_loss</span><span class="p">,</span> <span class="n">overwrite_eps</span>
+<span class="kn">from</span> <span class="nn">.anchor_utils</span> <span class="kn">import</span> <span class="n">AnchorGenerator</span>
+<span class="kn">from</span> <span class="nn">.backbone_utils</span> <span class="kn">import</span> <span class="n">_resnet_fpn_extractor</span><span class="p">,</span> <span class="n">_validate_trainable_layers</span>
+<span class="kn">from</span> <span class="nn">.transform</span> <span class="kn">import</span> <span class="n">GeneralizedRCNNTransform</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;RetinaNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RetinaNet_ResNet50_FPN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RetinaNet_ResNet50_FPN_V2_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;retinanet_resnet50_fpn&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;retinanet_resnet50_fpn_v2&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_sum</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">res</span> <span class="o">=</span> <span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">x</span><span class="p">[</span><span class="mi">1</span><span class="p">:]:</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="n">res</span> <span class="o">+</span> <span class="n">i</span>
+    <span class="k">return</span> <span class="n">res</span>
+
+
+<span class="k">def</span> <span class="nf">_v1_to_v2_weights</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">prefix</span><span class="p">):</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">):</span>
+        <span class="k">for</span> <span class="nb">type</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;weight&quot;</span><span class="p">,</span> <span class="s2">&quot;bias&quot;</span><span class="p">]:</span>
+            <span class="n">old_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">conv.</span><span class="si">{</span><span class="mi">2</span><span class="o">*</span><span class="n">i</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="n">new_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">conv.</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">.0.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="k">if</span> <span class="n">old_key</span> <span class="ow">in</span> <span class="n">state_dict</span><span class="p">:</span>
+                <span class="n">state_dict</span><span class="p">[</span><span class="n">new_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">state_dict</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">old_key</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_default_anchorgen</span><span class="p">():</span>
+    <span class="n">anchor_sizes</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">((</span><span class="n">x</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">x</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">**</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">/</span> <span class="mi">3</span><span class="p">)),</span> <span class="nb">int</span><span class="p">(</span><span class="n">x</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">**</span> <span class="p">(</span><span class="mf">2.0</span> <span class="o">/</span> <span class="mi">3</span><span class="p">)))</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">512</span><span class="p">])</span>
+    <span class="n">aspect_ratios</span> <span class="o">=</span> <span class="p">((</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mf">2.0</span><span class="p">),)</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">)</span>
+    <span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">AnchorGenerator</span><span class="p">(</span><span class="n">anchor_sizes</span><span class="p">,</span> <span class="n">aspect_ratios</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">anchor_generator</span>
+
+
+<span class="k">class</span> <span class="nc">RetinaNetHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A regression and classification head for use in RetinaNet.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of channels of the input feature</span>
+<span class="sd">        num_anchors (int): number of anchors to be predicted</span>
+<span class="sd">        num_classes (int): number of classes to be predicted</span>
+<span class="sd">        norm_layer (callable, optional): Module specifying the normalization layer to use. Default: None</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span> <span class="o">=</span> <span class="n">RetinaNetClassificationHead</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span> <span class="o">=</span> <span class="n">RetinaNetRegressionHead</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">compute_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">):</span>
+        <span class="c1"># type: (List[Dict[str, Tensor]], Dict[str, Tensor], List[Tensor], List[Tensor]) -&gt; Dict[str, Tensor]</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;classification&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">),</span>
+            <span class="s2">&quot;bbox_regression&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">),</span>
+        <span class="p">}</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="c1"># type: (List[Tensor]) -&gt; Dict[str, Tensor]</span>
+        <span class="k">return</span> <span class="p">{</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="s2">&quot;bbox_regression&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span><span class="p">(</span><span class="n">x</span><span class="p">)}</span>
+
+
+<span class="k">class</span> <span class="nc">RetinaNetClassificationHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A classification head for use in RetinaNet.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of channels of the input feature</span>
+<span class="sd">        num_anchors (int): number of anchors to be predicted</span>
+<span class="sd">        num_classes (int): number of classes to be predicted</span>
+<span class="sd">        norm_layer (callable, optional): Module specifying the normalization layer to use. Default: None</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">,</span>
+        <span class="n">num_anchors</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">,</span>
+        <span class="n">prior_probability</span><span class="o">=</span><span class="mf">0.01</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="n">conv</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">):</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">conv</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">layer</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span> <span class="o">*</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="o">-</span><span class="n">math</span><span class="o">.</span><span class="n">log</span><span class="p">((</span><span class="mi">1</span> <span class="o">-</span> <span class="n">prior_probability</span><span class="p">)</span> <span class="o">/</span> <span class="n">prior_probability</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_anchors</span> <span class="o">=</span> <span class="n">num_anchors</span>
+
+        <span class="c1"># This is to fix using det_utils.Matcher.BETWEEN_THRESHOLDS in TorchScript.</span>
+        <span class="c1"># TorchScript doesn&#39;t support class attributes.</span>
+        <span class="c1"># https://github.com/pytorch/vision/pull/1697#issuecomment-630255584</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">BETWEEN_THRESHOLDS</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">Matcher</span><span class="o">.</span><span class="n">BETWEEN_THRESHOLDS</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">,</span>
+        <span class="n">unexpected_keys</span><span class="p">,</span>
+        <span class="n">error_msgs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">version</span> <span class="o">=</span> <span class="n">local_metadata</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">version</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">version</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">_v1_to_v2_weights</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">prefix</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span>
+            <span class="n">prefix</span><span class="p">,</span>
+            <span class="n">local_metadata</span><span class="p">,</span>
+            <span class="n">strict</span><span class="p">,</span>
+            <span class="n">missing_keys</span><span class="p">,</span>
+            <span class="n">unexpected_keys</span><span class="p">,</span>
+            <span class="n">error_msgs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">compute_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">):</span>
+        <span class="c1"># type: (List[Dict[str, Tensor]], Dict[str, Tensor], List[Tensor]) -&gt; Tensor</span>
+        <span class="n">losses</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">]</span>
+
+        <span class="k">for</span> <span class="n">targets_per_image</span><span class="p">,</span> <span class="n">cls_logits_per_image</span><span class="p">,</span> <span class="n">matched_idxs_per_image</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">cls_logits</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">):</span>
+            <span class="c1"># determine only the foreground</span>
+            <span class="n">foreground_idxs_per_image</span> <span class="o">=</span> <span class="n">matched_idxs_per_image</span> <span class="o">&gt;=</span> <span class="mi">0</span>
+            <span class="n">num_foreground</span> <span class="o">=</span> <span class="n">foreground_idxs_per_image</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+
+            <span class="c1"># create the target classification</span>
+            <span class="n">gt_classes_target</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">cls_logits_per_image</span><span class="p">)</span>
+            <span class="n">gt_classes_target</span><span class="p">[</span>
+                <span class="n">foreground_idxs_per_image</span><span class="p">,</span>
+                <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">matched_idxs_per_image</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">]],</span>
+            <span class="p">]</span> <span class="o">=</span> <span class="mf">1.0</span>
+
+            <span class="c1"># find indices for which anchors should be ignored</span>
+            <span class="n">valid_idxs_per_image</span> <span class="o">=</span> <span class="n">matched_idxs_per_image</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">BETWEEN_THRESHOLDS</span>
+
+            <span class="c1"># compute the classification loss</span>
+            <span class="n">losses</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">sigmoid_focal_loss</span><span class="p">(</span>
+                    <span class="n">cls_logits_per_image</span><span class="p">[</span><span class="n">valid_idxs_per_image</span><span class="p">],</span>
+                    <span class="n">gt_classes_target</span><span class="p">[</span><span class="n">valid_idxs_per_image</span><span class="p">],</span>
+                    <span class="n">reduction</span><span class="o">=</span><span class="s2">&quot;sum&quot;</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_foreground</span><span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">_sum</span><span class="p">(</span><span class="n">losses</span><span class="p">)</span> <span class="o">/</span> <span class="nb">len</span><span class="p">(</span><span class="n">targets</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="c1"># type: (List[Tensor]) -&gt; Tensor</span>
+        <span class="n">all_cls_logits</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">features</span> <span class="ow">in</span> <span class="n">x</span><span class="p">:</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">cls_logits</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">)</span>
+
+            <span class="c1"># Permute classification output from (N, A * K, H, W) to (N, HWA, K).</span>
+            <span class="n">N</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">shape</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+            <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span>  <span class="c1"># Size=(N, HWA, 4)</span>
+
+            <span class="n">all_cls_logits</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">all_cls_logits</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">RetinaNetRegressionHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A regression head for use in RetinaNet.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): number of channels of the input feature</span>
+<span class="sd">        num_anchors (int): number of anchors to be predicted</span>
+<span class="sd">        norm_layer (callable, optional): Module specifying the normalization layer to use. Default: None</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;box_coder&quot;</span><span class="p">:</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxCoder</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="n">conv</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">):</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">conv</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">bbox_reg</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">bbox_reg</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">bbox_reg</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">layer</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxCoder</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_loss_type</span> <span class="o">=</span> <span class="s2">&quot;l1&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">,</span>
+        <span class="n">unexpected_keys</span><span class="p">,</span>
+        <span class="n">error_msgs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">version</span> <span class="o">=</span> <span class="n">local_metadata</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">version</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">version</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">_v1_to_v2_weights</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">prefix</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span>
+            <span class="n">prefix</span><span class="p">,</span>
+            <span class="n">local_metadata</span><span class="p">,</span>
+            <span class="n">strict</span><span class="p">,</span>
+            <span class="n">missing_keys</span><span class="p">,</span>
+            <span class="n">unexpected_keys</span><span class="p">,</span>
+            <span class="n">error_msgs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">compute_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">):</span>
+        <span class="c1"># type: (List[Dict[str, Tensor]], Dict[str, Tensor], List[Tensor], List[Tensor]) -&gt; Tensor</span>
+        <span class="n">losses</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_regression&quot;</span><span class="p">]</span>
+
+        <span class="k">for</span> <span class="n">targets_per_image</span><span class="p">,</span> <span class="n">bbox_regression_per_image</span><span class="p">,</span> <span class="n">anchors_per_image</span><span class="p">,</span> <span class="n">matched_idxs_per_image</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span>
+            <span class="n">targets</span><span class="p">,</span> <span class="n">bbox_regression</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span>
+        <span class="p">):</span>
+            <span class="c1"># determine only the foreground indices, ignore the rest</span>
+            <span class="n">foreground_idxs_per_image</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">matched_idxs_per_image</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">num_foreground</span> <span class="o">=</span> <span class="n">foreground_idxs_per_image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span>
+
+            <span class="c1"># select only the foreground boxes</span>
+            <span class="n">matched_gt_boxes_per_image</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">][</span><span class="n">matched_idxs_per_image</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">]]</span>
+            <span class="n">bbox_regression_per_image</span> <span class="o">=</span> <span class="n">bbox_regression_per_image</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">,</span> <span class="p">:]</span>
+            <span class="n">anchors_per_image</span> <span class="o">=</span> <span class="n">anchors_per_image</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">,</span> <span class="p">:]</span>
+
+            <span class="c1"># compute the loss</span>
+            <span class="n">losses</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">_box_loss</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_loss_type</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span><span class="p">,</span>
+                    <span class="n">anchors_per_image</span><span class="p">,</span>
+                    <span class="n">matched_gt_boxes_per_image</span><span class="p">,</span>
+                    <span class="n">bbox_regression_per_image</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_foreground</span><span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">_sum</span><span class="p">(</span><span class="n">losses</span><span class="p">)</span> <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">targets</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="c1"># type: (List[Tensor]) -&gt; Tensor</span>
+        <span class="n">all_bbox_regression</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">features</span> <span class="ow">in</span> <span class="n">x</span><span class="p">:</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bbox_reg</span><span class="p">(</span><span class="n">bbox_regression</span><span class="p">)</span>
+
+            <span class="c1"># Permute bbox regression output from (N, 4 * A, H, W) to (N, HWA, 4).</span>
+            <span class="n">N</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">shape</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+            <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">bbox_regression</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>  <span class="c1"># Size=(N, HWA, 4)</span>
+
+            <span class="n">all_bbox_regression</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">bbox_regression</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">all_bbox_regression</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">RetinaNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements RetinaNet.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each</span>
+<span class="sd">    image, and should be in 0-1 range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a Dict[Tensor] during training, containing the classification and regression</span>
+<span class="sd">    losses.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as</span>
+<span class="sd">    follows:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the predicted labels for each image</span>
+<span class="sd">        - scores (Tensor[N]): the scores for each prediction</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            It should contain an out_channels attribute, which indicates the number of output</span>
+<span class="sd">            channels that each feature map has (and it should be the same for all feature maps).</span>
+<span class="sd">            The backbone should return a single Tensor or an OrderedDict[Tensor].</span>
+<span class="sd">        num_classes (int): number of output classes of the model (including the background).</span>
+<span class="sd">        min_size (int): minimum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        max_size (int): maximum size of the image to be rescaled before feeding it to the backbone</span>
+<span class="sd">        image_mean (Tuple[float, float, float]): mean values used for input normalization.</span>
+<span class="sd">            They are generally the mean values of the dataset on which the backbone has been trained</span>
+<span class="sd">            on</span>
+<span class="sd">        image_std (Tuple[float, float, float]): std values used for input normalization.</span>
+<span class="sd">            They are generally the std values of the dataset on which the backbone has been trained on</span>
+<span class="sd">        anchor_generator (AnchorGenerator): module that generates the anchors for a set of feature</span>
+<span class="sd">            maps.</span>
+<span class="sd">        head (nn.Module): Module run on top of the feature pyramid.</span>
+<span class="sd">            Defaults to a module containing a classification and regression module.</span>
+<span class="sd">        score_thresh (float): Score threshold used for postprocessing the detections.</span>
+<span class="sd">        nms_thresh (float): NMS threshold used for postprocessing the detections.</span>
+<span class="sd">        detections_per_img (int): Number of best detections to keep after NMS.</span>
+<span class="sd">        fg_iou_thresh (float): minimum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training.</span>
+<span class="sd">        bg_iou_thresh (float): maximum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as negative during training.</span>
+<span class="sd">        topk_candidates (int): Number of best detections to keep before NMS.</span>
+
+<span class="sd">    Example:</span>
+
+<span class="sd">        &gt;&gt;&gt; import torch</span>
+<span class="sd">        &gt;&gt;&gt; import torchvision</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection import RetinaNet</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.models.detection.anchor_utils import AnchorGenerator</span>
+<span class="sd">        &gt;&gt;&gt; # load a pre-trained model for classification and return</span>
+<span class="sd">        &gt;&gt;&gt; # only the features</span>
+<span class="sd">        &gt;&gt;&gt; backbone = torchvision.models.mobilenet_v2(weights=MobileNet_V2_Weights.DEFAULT).features</span>
+<span class="sd">        &gt;&gt;&gt; # RetinaNet needs to know the number of</span>
+<span class="sd">        &gt;&gt;&gt; # output channels in a backbone. For mobilenet_v2, it&#39;s 1280,</span>
+<span class="sd">        &gt;&gt;&gt; # so we need to add it here</span>
+<span class="sd">        &gt;&gt;&gt; backbone.out_channels = 1280</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # let&#39;s make the network generate 5 x 3 anchors per spatial</span>
+<span class="sd">        &gt;&gt;&gt; # location, with 5 different sizes and 3 different aspect</span>
+<span class="sd">        &gt;&gt;&gt; # ratios. We have a Tuple[Tuple[int]] because each feature</span>
+<span class="sd">        &gt;&gt;&gt; # map could potentially have different sizes and</span>
+<span class="sd">        &gt;&gt;&gt; # aspect ratios</span>
+<span class="sd">        &gt;&gt;&gt; anchor_generator = AnchorGenerator(</span>
+<span class="sd">        &gt;&gt;&gt;     sizes=((32, 64, 128, 256, 512),),</span>
+<span class="sd">        &gt;&gt;&gt;     aspect_ratios=((0.5, 1.0, 2.0),)</span>
+<span class="sd">        &gt;&gt;&gt; )</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; # put the pieces together inside a RetinaNet model</span>
+<span class="sd">        &gt;&gt;&gt; model = RetinaNet(backbone,</span>
+<span class="sd">        &gt;&gt;&gt;                   num_classes=2,</span>
+<span class="sd">        &gt;&gt;&gt;                   anchor_generator=anchor_generator)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;box_coder&quot;</span><span class="p">:</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxCoder</span><span class="p">,</span>
+        <span class="s2">&quot;proposal_matcher&quot;</span><span class="p">:</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">Matcher</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">,</span>
+        <span class="c1"># transform parameters</span>
+        <span class="n">min_size</span><span class="o">=</span><span class="mi">800</span><span class="p">,</span>
+        <span class="n">max_size</span><span class="o">=</span><span class="mi">1333</span><span class="p">,</span>
+        <span class="n">image_mean</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_std</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># Anchor parameters</span>
+        <span class="n">anchor_generator</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">head</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">proposal_matcher</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">score_thresh</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span>
+        <span class="n">nms_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">detections_per_img</span><span class="o">=</span><span class="mi">300</span><span class="p">,</span>
+        <span class="n">fg_iou_thresh</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">bg_iou_thresh</span><span class="o">=</span><span class="mf">0.4</span><span class="p">,</span>
+        <span class="n">topk_candidates</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="s2">&quot;out_channels&quot;</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;backbone should contain an attribute out_channels &quot;</span>
+                <span class="s2">&quot;specifying the number of output channels (assumed to be the &quot;</span>
+                <span class="s2">&quot;same for all the levels)&quot;</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">anchor_generator</span><span class="p">,</span> <span class="p">(</span><span class="n">AnchorGenerator</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="kc">None</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;anchor_generator should be of type AnchorGenerator or None instead of </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">anchor_generator</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">anchor_generator</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">_default_anchorgen</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">anchor_generator</span>
+
+        <span class="k">if</span> <span class="n">head</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">head</span> <span class="o">=</span> <span class="n">RetinaNetHead</span><span class="p">(</span><span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">],</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head</span> <span class="o">=</span> <span class="n">head</span>
+
+        <span class="k">if</span> <span class="n">proposal_matcher</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">proposal_matcher</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">Matcher</span><span class="p">(</span>
+                <span class="n">fg_iou_thresh</span><span class="p">,</span>
+                <span class="n">bg_iou_thresh</span><span class="p">,</span>
+                <span class="n">allow_low_quality_matches</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">proposal_matcher</span> <span class="o">=</span> <span class="n">proposal_matcher</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxCoder</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="n">image_mean</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_mean</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">image_std</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_std</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">GeneralizedRCNNTransform</span><span class="p">(</span><span class="n">min_size</span><span class="p">,</span> <span class="n">max_size</span><span class="p">,</span> <span class="n">image_mean</span><span class="p">,</span> <span class="n">image_std</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">score_thresh</span> <span class="o">=</span> <span class="n">score_thresh</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">nms_thresh</span> <span class="o">=</span> <span class="n">nms_thresh</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">detections_per_img</span> <span class="o">=</span> <span class="n">detections_per_img</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">topk_candidates</span> <span class="o">=</span> <span class="n">topk_candidates</span>
+
+        <span class="c1"># used only on torchscript mode</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">eager_outputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">losses</span><span class="p">,</span> <span class="n">detections</span><span class="p">):</span>
+        <span class="c1"># type: (Dict[str, Tensor], List[Dict[str, Tensor]]) -&gt; Tuple[Dict[str, Tensor], List[Dict[str, Tensor]]]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">losses</span>
+
+        <span class="k">return</span> <span class="n">detections</span>
+
+    <span class="k">def</span> <span class="nf">compute_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">):</span>
+        <span class="c1"># type: (List[Dict[str, Tensor]], Dict[str, Tensor], List[Tensor]) -&gt; Dict[str, Tensor]</span>
+        <span class="n">matched_idxs</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">anchors_per_image</span><span class="p">,</span> <span class="n">targets_per_image</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">anchors</span><span class="p">,</span> <span class="n">targets</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">matched_idxs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">anchors_per_image</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),),</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">anchors_per_image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="p">)</span>
+                <span class="k">continue</span>
+
+            <span class="n">match_quality_matrix</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">box_iou</span><span class="p">(</span><span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">],</span> <span class="n">anchors_per_image</span><span class="p">)</span>
+            <span class="n">matched_idxs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">proposal_matcher</span><span class="p">(</span><span class="n">match_quality_matrix</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">postprocess_detections</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">image_shapes</span><span class="p">):</span>
+        <span class="c1"># type: (Dict[str, List[Tensor]], List[List[Tensor]], List[Tuple[int, int]]) -&gt; List[Dict[str, Tensor]]</span>
+        <span class="n">class_logits</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">]</span>
+        <span class="n">box_regression</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_regression&quot;</span><span class="p">]</span>
+
+        <span class="n">num_images</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">image_shapes</span><span class="p">)</span>
+
+        <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">index</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_images</span><span class="p">):</span>
+            <span class="n">box_regression_per_image</span> <span class="o">=</span> <span class="p">[</span><span class="n">br</span><span class="p">[</span><span class="n">index</span><span class="p">]</span> <span class="k">for</span> <span class="n">br</span> <span class="ow">in</span> <span class="n">box_regression</span><span class="p">]</span>
+            <span class="n">logits_per_image</span> <span class="o">=</span> <span class="p">[</span><span class="n">cl</span><span class="p">[</span><span class="n">index</span><span class="p">]</span> <span class="k">for</span> <span class="n">cl</span> <span class="ow">in</span> <span class="n">class_logits</span><span class="p">]</span>
+            <span class="n">anchors_per_image</span><span class="p">,</span> <span class="n">image_shape</span> <span class="o">=</span> <span class="n">anchors</span><span class="p">[</span><span class="n">index</span><span class="p">],</span> <span class="n">image_shapes</span><span class="p">[</span><span class="n">index</span><span class="p">]</span>
+
+            <span class="n">image_boxes</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">image_scores</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">image_labels</span> <span class="o">=</span> <span class="p">[]</span>
+
+            <span class="k">for</span> <span class="n">box_regression_per_level</span><span class="p">,</span> <span class="n">logits_per_level</span><span class="p">,</span> <span class="n">anchors_per_level</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span>
+                <span class="n">box_regression_per_image</span><span class="p">,</span> <span class="n">logits_per_image</span><span class="p">,</span> <span class="n">anchors_per_image</span>
+            <span class="p">):</span>
+                <span class="n">num_classes</span> <span class="o">=</span> <span class="n">logits_per_level</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+                <span class="c1"># remove low scoring boxes</span>
+                <span class="n">scores_per_level</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="n">logits_per_level</span><span class="p">)</span><span class="o">.</span><span class="n">flatten</span><span class="p">()</span>
+                <span class="n">keep_idxs</span> <span class="o">=</span> <span class="n">scores_per_level</span> <span class="o">&gt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">score_thresh</span>
+                <span class="n">scores_per_level</span> <span class="o">=</span> <span class="n">scores_per_level</span><span class="p">[</span><span class="n">keep_idxs</span><span class="p">]</span>
+                <span class="n">topk_idxs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">keep_idxs</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+
+                <span class="c1"># keep only topk scoring predictions</span>
+                <span class="n">num_topk</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">_topk_min</span><span class="p">(</span><span class="n">topk_idxs</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">topk_candidates</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="n">scores_per_level</span><span class="p">,</span> <span class="n">idxs</span> <span class="o">=</span> <span class="n">scores_per_level</span><span class="o">.</span><span class="n">topk</span><span class="p">(</span><span class="n">num_topk</span><span class="p">)</span>
+                <span class="n">topk_idxs</span> <span class="o">=</span> <span class="n">topk_idxs</span><span class="p">[</span><span class="n">idxs</span><span class="p">]</span>
+
+                <span class="n">anchor_idxs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">div</span><span class="p">(</span><span class="n">topk_idxs</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">rounding_mode</span><span class="o">=</span><span class="s2">&quot;floor&quot;</span><span class="p">)</span>
+                <span class="n">labels_per_level</span> <span class="o">=</span> <span class="n">topk_idxs</span> <span class="o">%</span> <span class="n">num_classes</span>
+
+                <span class="n">boxes_per_level</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span><span class="o">.</span><span class="n">decode_single</span><span class="p">(</span>
+                    <span class="n">box_regression_per_level</span><span class="p">[</span><span class="n">anchor_idxs</span><span class="p">],</span> <span class="n">anchors_per_level</span><span class="p">[</span><span class="n">anchor_idxs</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">boxes_per_level</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">clip_boxes_to_image</span><span class="p">(</span><span class="n">boxes_per_level</span><span class="p">,</span> <span class="n">image_shape</span><span class="p">)</span>
+
+                <span class="n">image_boxes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">boxes_per_level</span><span class="p">)</span>
+                <span class="n">image_scores</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">scores_per_level</span><span class="p">)</span>
+                <span class="n">image_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">labels_per_level</span><span class="p">)</span>
+
+            <span class="n">image_boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_boxes</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_scores</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_scores</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_labels</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+            <span class="c1"># non-maximum suppression</span>
+            <span class="n">keep</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">batched_nms</span><span class="p">(</span><span class="n">image_boxes</span><span class="p">,</span> <span class="n">image_scores</span><span class="p">,</span> <span class="n">image_labels</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">nms_thresh</span><span class="p">)</span>
+            <span class="n">keep</span> <span class="o">=</span> <span class="n">keep</span><span class="p">[:</span> <span class="bp">self</span><span class="o">.</span><span class="n">detections_per_img</span><span class="p">]</span>
+
+            <span class="n">detections</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;boxes&quot;</span><span class="p">:</span> <span class="n">image_boxes</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                    <span class="s2">&quot;scores&quot;</span><span class="p">:</span> <span class="n">image_scores</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                    <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="n">image_labels</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">detections</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">images</span><span class="p">,</span> <span class="n">targets</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="c1"># type: (List[Tensor], Optional[List[Dict[str, Tensor]]]) -&gt; Tuple[Dict[str, Tensor], List[Dict[str, Tensor]]]</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            images (list[Tensor]): images to be processed</span>
+<span class="sd">            targets (list[Dict[Tensor]]): ground-truth boxes present in the image (optional)</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            result (list[BoxList] or dict[Tensor]): the output from the model.</span>
+<span class="sd">                During training, it returns a dict[Tensor] which contains the losses.</span>
+<span class="sd">                During testing, it returns list[BoxList] contains additional fields</span>
+<span class="sd">                like `scores`, `labels` and `mask` (for Mask R-CNN models).</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;targets should not be none when in training mode&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">target</span> <span class="ow">in</span> <span class="n">targets</span><span class="p">:</span>
+                    <span class="n">boxes</span> <span class="o">=</span> <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="s2">&quot;Expected target boxes to be of type Tensor.&quot;</span><span class="p">)</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                        <span class="nb">len</span><span class="p">(</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">4</span><span class="p">,</span>
+                        <span class="s2">&quot;Expected target boxes to be a tensor of shape [N, 4].&quot;</span><span class="p">,</span>
+                    <span class="p">)</span>
+
+        <span class="c1"># get the original image sizes</span>
+        <span class="n">original_image_sizes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">img</span> <span class="ow">in</span> <span class="n">images</span><span class="p">:</span>
+            <span class="n">val</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                <span class="nb">len</span><span class="p">(</span><span class="n">val</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">,</span>
+                <span class="sa">f</span><span class="s2">&quot;expecting the last two dimensions of the Tensor to be H and W instead got </span><span class="si">{</span><span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">original_image_sizes</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">val</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">val</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+
+        <span class="c1"># transform the input</span>
+        <span class="n">images</span><span class="p">,</span> <span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">targets</span><span class="p">)</span>
+
+        <span class="c1"># Check for degenerate boxes</span>
+        <span class="c1"># TODO: Move this to a function</span>
+        <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">target_idx</span><span class="p">,</span> <span class="n">target</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">targets</span><span class="p">):</span>
+                <span class="n">boxes</span> <span class="o">=</span> <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span>
+                <span class="n">degenerate_boxes</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:]</span> <span class="o">&lt;=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">degenerate_boxes</span><span class="o">.</span><span class="n">any</span><span class="p">():</span>
+                    <span class="c1"># print the first degenerate box</span>
+                    <span class="n">bb_idx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">degenerate_boxes</span><span class="o">.</span><span class="n">any</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">))[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                    <span class="n">degen_bb</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[</span><span class="n">bb_idx</span><span class="p">]</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                        <span class="kc">False</span><span class="p">,</span>
+                        <span class="s2">&quot;All bounding boxes should have positive height and width.&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot; Found invalid box </span><span class="si">{</span><span class="n">degen_bb</span><span class="si">}</span><span class="s2"> for target at index </span><span class="si">{</span><span class="n">target_idx</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">,</span>
+                    <span class="p">)</span>
+
+        <span class="c1"># get the features from the backbone</span>
+        <span class="n">features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span><span class="p">(</span><span class="n">images</span><span class="o">.</span><span class="n">tensors</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">features</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="n">features</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">([(</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="n">features</span><span class="p">)])</span>
+
+        <span class="c1"># TODO: Do we want a list or a dict?</span>
+        <span class="n">features</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">features</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+
+        <span class="c1"># compute the retinanet heads outputs using the features</span>
+        <span class="n">head_outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+
+        <span class="c1"># create the set of anchors</span>
+        <span class="n">anchors</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">features</span><span class="p">)</span>
+
+        <span class="n">losses</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;targets should not be none when in training mode&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="c1"># compute the losses</span>
+                <span class="n">losses</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># recover level sizes</span>
+            <span class="n">num_anchors_per_level</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span> <span class="o">*</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">3</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">features</span><span class="p">]</span>
+            <span class="n">HW</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">num_anchors_per_level</span><span class="p">:</span>
+                <span class="n">HW</span> <span class="o">+=</span> <span class="n">v</span>
+            <span class="n">HWA</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+            <span class="n">A</span> <span class="o">=</span> <span class="n">HWA</span> <span class="o">//</span> <span class="n">HW</span>
+            <span class="n">num_anchors_per_level</span> <span class="o">=</span> <span class="p">[</span><span class="n">hw</span> <span class="o">*</span> <span class="n">A</span> <span class="k">for</span> <span class="n">hw</span> <span class="ow">in</span> <span class="n">num_anchors_per_level</span><span class="p">]</span>
+
+            <span class="c1"># split outputs per level</span>
+            <span class="n">split_head_outputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">{}</span>
+            <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">head_outputs</span><span class="p">:</span>
+                <span class="n">split_head_outputs</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">head_outputs</span><span class="p">[</span><span class="n">k</span><span class="p">]</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">num_anchors_per_level</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+            <span class="n">split_anchors</span> <span class="o">=</span> <span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">a</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">num_anchors_per_level</span><span class="p">))</span> <span class="k">for</span> <span class="n">a</span> <span class="ow">in</span> <span class="n">anchors</span><span class="p">]</span>
+
+            <span class="c1"># compute the detections</span>
+            <span class="n">detections</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">postprocess_detections</span><span class="p">(</span><span class="n">split_head_outputs</span><span class="p">,</span> <span class="n">split_anchors</span><span class="p">,</span> <span class="n">images</span><span class="o">.</span><span class="n">image_sizes</span><span class="p">)</span>
+            <span class="n">detections</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="o">.</span><span class="n">postprocess</span><span class="p">(</span><span class="n">detections</span><span class="p">,</span> <span class="n">images</span><span class="o">.</span><span class="n">image_sizes</span><span class="p">,</span> <span class="n">original_image_sizes</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;RetinaNet always returns a (Losses, Detections) tuple in scripting&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span> <span class="o">=</span> <span class="kc">True</span>
+            <span class="k">return</span> <span class="n">losses</span><span class="p">,</span> <span class="n">detections</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">eager_outputs</span><span class="p">(</span><span class="n">losses</span><span class="p">,</span> <span class="n">detections</span><span class="p">)</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_COCO_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="RetinaNet_ResNet50_FPN_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html#torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights">[docs]</a><span class="k">class</span> <span class="nc">RetinaNet_ResNet50_FPN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/retinanet_resnet50_fpn_coco-eeacb38b.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">34014999</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#retinanet&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">36.4</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">151.54</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">130.267</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="RetinaNet_ResNet50_FPN_V2_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html#torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights">[docs]</a><span class="k">class</span> <span class="nc">RetinaNet_ResNet50_FPN_V2_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/retinanet_resnet50_fpn_v2_coco-5905b1c5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">38198935</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5756&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">41.5</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">152.238</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">146.037</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced using an enhanced training recipe to boost the model accuracy.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="retinanet_resnet50_fpn"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html#torchvision.models.detection.retinanet_resnet50_fpn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RetinaNet_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">retinanet_resnet50_fpn</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RetinaNet_ResNet50_FPN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RetinaNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RetinaNet model with a ResNet-50-FPN backbone.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    Reference: `Focal Loss for Dense Object Detection &lt;https://arxiv.org/abs/1708.02002&gt;`_.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each</span>
+<span class="sd">    image, and should be in ``0-1`` range. Different images can have different sizes.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a ``Dict[Tensor]`` during training, containing the classification and regression</span>
+<span class="sd">    losses.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a ``List[Dict[Tensor]]``, one for each input image. The fields of the ``Dict`` are as</span>
+<span class="sd">    follows, where ``N`` is the number of detections:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (``Int64Tensor[N]``): the predicted labels for each detection</span>
+<span class="sd">        - scores (``Tensor[N]``): the scores of each detection</span>
+
+<span class="sd">    For more details on the output, you may refer to :ref:`instance_seg_output`.</span>
+
+<span class="sd">    Example::</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.retinanet_resnet50_fpn(weights=RetinaNet_ResNet50_FPN_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 400), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The pretrained weights for</span>
+<span class="sd">            the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from final block.</span>
+<span class="sd">            Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If ``None`` is</span>
+<span class="sd">            passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.RetinaNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/retinanet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RetinaNet_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">misc_nn_ops</span><span class="o">.</span><span class="n">FrozenBatchNorm2d</span> <span class="k">if</span> <span class="n">is_trained</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+    <span class="c1"># skip P2 because it generates too many anchors (according to their paper)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="n">returned_layers</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="n">extra_blocks</span><span class="o">=</span><span class="n">LastLevelP6P7</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">)</span>
+    <span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">RetinaNet</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weights</span> <span class="o">==</span> <span class="n">RetinaNet_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">:</span>
+            <span class="n">overwrite_eps</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="retinanet_resnet50_fpn_v2"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html#torchvision.models.detection.retinanet_resnet50_fpn_v2">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RetinaNet_ResNet50_FPN_V2_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">retinanet_resnet50_fpn_v2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RetinaNet_ResNet50_FPN_V2_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RetinaNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs an improved RetinaNet model with a ResNet-50-FPN backbone.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    Reference: `Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1912.02424&gt;`_.</span>
+
+<span class="sd">    :func:`~torchvision.models.detection.retinanet_resnet50_fpn` for more details.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The pretrained weights for</span>
+<span class="sd">            the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from final block.</span>
+<span class="sd">            Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If ``None`` is</span>
+<span class="sd">            passed (the default) this value is set to 3.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.RetinaNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/retinanet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RetinaNet_ResNet50_FPN_V2_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">is_trained</span> <span class="o">=</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span><span class="n">is_trained</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_resnet_fpn_extractor</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="n">returned_layers</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="n">extra_blocks</span><span class="o">=</span><span class="n">LastLevelP6P7</span><span class="p">(</span><span class="mi">2048</span><span class="p">,</span> <span class="mi">256</span><span class="p">)</span>
+    <span class="p">)</span>
+    <span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">_default_anchorgen</span><span class="p">()</span>
+    <span class="n">head</span> <span class="o">=</span> <span class="n">RetinaNetHead</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span>
+        <span class="n">anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">num_classes</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">GroupNorm</span><span class="p">,</span> <span class="mi">32</span><span class="p">),</span>
+    <span class="p">)</span>
+    <span class="n">head</span><span class="o">.</span><span class="n">regression_head</span><span class="o">.</span><span class="n">_loss_type</span> <span class="o">=</span> <span class="s2">&quot;giou&quot;</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">RetinaNet</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">anchor_generator</span><span class="o">=</span><span class="n">anchor_generator</span><span class="p">,</span> <span class="n">head</span><span class="o">=</span><span class="n">head</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/detection/ssd.html b/0.16/_modules/torchvision/models/detection/ssd.html
new file mode 100644
index 00000000000..6c3e468dbd9
--- /dev/null
+++ b/0.16/_modules/torchvision/models/detection/ssd.html
@@ -0,0 +1,1351 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.detection.ssd &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.detection.ssd</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.detection.ssd</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">...ops</span> <span class="kn">import</span> <span class="n">boxes</span> <span class="k">as</span> <span class="n">box_ops</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ObjectDetection</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_COCO_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..vgg</span> <span class="kn">import</span> <span class="n">VGG</span><span class="p">,</span> <span class="n">vgg16</span><span class="p">,</span> <span class="n">VGG16_Weights</span>
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">_utils</span> <span class="k">as</span> <span class="n">det_utils</span>
+<span class="kn">from</span> <span class="nn">.anchor_utils</span> <span class="kn">import</span> <span class="n">DefaultBoxGenerator</span>
+<span class="kn">from</span> <span class="nn">.backbone_utils</span> <span class="kn">import</span> <span class="n">_validate_trainable_layers</span>
+<span class="kn">from</span> <span class="nn">.transform</span> <span class="kn">import</span> <span class="n">GeneralizedRCNNTransform</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;SSD300_VGG16_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ssd300_vgg16&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<div class="viewcode-block" id="SSD300_VGG16_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.ssd300_vgg16.html#torchvision.models.detection.SSD300_VGG16_Weights">[docs]</a><span class="k">class</span> <span class="nc">SSD300_VGG16_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/ssd300_vgg16_coco-b556d3b4.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">35641826</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_COCO_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#ssd300-vgg16&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">25.1</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">34.858</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">135.988</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<span class="k">def</span> <span class="nf">_xavier_init</span><span class="p">(</span><span class="n">conv</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="n">conv</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">xavier_uniform_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">layer</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SSDHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span> <span class="o">=</span> <span class="n">SSDClassificationHead</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span> <span class="o">=</span> <span class="n">SSDRegressionHead</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;bbox_regression&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span><span class="p">(</span><span class="n">x</span><span class="p">),</span>
+            <span class="s2">&quot;cls_logits&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span><span class="p">(</span><span class="n">x</span><span class="p">),</span>
+        <span class="p">}</span>
+
+
+<span class="k">class</span> <span class="nc">SSDScoringHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">module_list</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">,</span> <span class="n">num_columns</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">module_list</span> <span class="o">=</span> <span class="n">module_list</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_columns</span> <span class="o">=</span> <span class="n">num_columns</span>
+
+    <span class="k">def</span> <span class="nf">_get_result_from_module_list</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        This is equivalent to self.module_list[idx](x),</span>
+<span class="sd">        but torchscript doesn&#39;t support this yet</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">num_blocks</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">module_list</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">idx</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">idx</span> <span class="o">+=</span> <span class="n">num_blocks</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">x</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">module</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">module_list</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="n">idx</span><span class="p">:</span>
+                <span class="n">out</span> <span class="o">=</span> <span class="n">module</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">out</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">all_results</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">features</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
+            <span class="n">results</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_result_from_module_list</span><span class="p">(</span><span class="n">features</span><span class="p">,</span> <span class="n">i</span><span class="p">)</span>
+
+            <span class="c1"># Permute output from (N, A * K, H, W) to (N, HWA, K).</span>
+            <span class="n">N</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">results</span><span class="o">.</span><span class="n">shape</span>
+            <span class="n">results</span> <span class="o">=</span> <span class="n">results</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_columns</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+            <span class="n">results</span> <span class="o">=</span> <span class="n">results</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+            <span class="n">results</span> <span class="o">=</span> <span class="n">results</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_columns</span><span class="p">)</span>  <span class="c1"># Size=(N, HWA, K)</span>
+
+            <span class="n">all_results</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">results</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">all_results</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SSDClassificationHead</span><span class="p">(</span><span class="n">SSDScoringHead</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">channels</span><span class="p">,</span> <span class="n">anchors</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">):</span>
+            <span class="n">cls_logits</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">channels</span><span class="p">,</span> <span class="n">num_classes</span> <span class="o">*</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+        <span class="n">_xavier_init</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SSDRegressionHead</span><span class="p">(</span><span class="n">SSDScoringHead</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]):</span>
+        <span class="n">bbox_reg</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">channels</span><span class="p">,</span> <span class="n">anchors</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">):</span>
+            <span class="n">bbox_reg</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">channels</span><span class="p">,</span> <span class="mi">4</span> <span class="o">*</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">))</span>
+        <span class="n">_xavier_init</span><span class="p">(</span><span class="n">bbox_reg</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">bbox_reg</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SSD</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements SSD architecture from `&quot;SSD: Single Shot MultiBox Detector&quot; &lt;https://arxiv.org/abs/1512.02325&gt;`_.</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each</span>
+<span class="sd">    image, and should be in 0-1 range. Different images can have different sizes, but they will be resized</span>
+<span class="sd">    to a fixed size before passing it to the backbone.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a Dict[Tensor] during training, containing the classification and regression</span>
+<span class="sd">    losses.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as</span>
+<span class="sd">    follows, where ``N`` is the number of detections:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the predicted labels for each detection</span>
+<span class="sd">        - scores (Tensor[N]): the scores for each detection</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            It should contain an out_channels attribute with the list of the output channels of</span>
+<span class="sd">            each feature map. The backbone should return a single Tensor or an OrderedDict[Tensor].</span>
+<span class="sd">        anchor_generator (DefaultBoxGenerator): module that generates the default boxes for a</span>
+<span class="sd">            set of feature maps.</span>
+<span class="sd">        size (Tuple[int, int]): the width and height to which images will be rescaled before feeding them</span>
+<span class="sd">            to the backbone.</span>
+<span class="sd">        num_classes (int): number of output classes of the model (including the background).</span>
+<span class="sd">        image_mean (Tuple[float, float, float]): mean values used for input normalization.</span>
+<span class="sd">            They are generally the mean values of the dataset on which the backbone has been trained</span>
+<span class="sd">            on</span>
+<span class="sd">        image_std (Tuple[float, float, float]): std values used for input normalization.</span>
+<span class="sd">            They are generally the std values of the dataset on which the backbone has been trained on</span>
+<span class="sd">        head (nn.Module, optional): Module run on top of the backbone features. Defaults to a module containing</span>
+<span class="sd">            a classification and regression module.</span>
+<span class="sd">        score_thresh (float): Score threshold used for postprocessing the detections.</span>
+<span class="sd">        nms_thresh (float): NMS threshold used for postprocessing the detections.</span>
+<span class="sd">        detections_per_img (int): Number of best detections to keep after NMS.</span>
+<span class="sd">        iou_thresh (float): minimum IoU between the anchor and the GT box so that they can be</span>
+<span class="sd">            considered as positive during training.</span>
+<span class="sd">        topk_candidates (int): Number of best detections to keep before NMS.</span>
+<span class="sd">        positive_fraction (float): a number between 0 and 1 which indicates the proportion of positive</span>
+<span class="sd">            proposals used during the training of the classification head. It is used to estimate the negative to</span>
+<span class="sd">            positive ratio.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;box_coder&quot;</span><span class="p">:</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxCoder</span><span class="p">,</span>
+        <span class="s2">&quot;proposal_matcher&quot;</span><span class="p">:</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">Matcher</span><span class="p">,</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">backbone</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span>
+        <span class="n">anchor_generator</span><span class="p">:</span> <span class="n">DefaultBoxGenerator</span><span class="p">,</span>
+        <span class="n">size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">image_mean</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_std</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">head</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">score_thresh</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.01</span><span class="p">,</span>
+        <span class="n">nms_thresh</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.45</span><span class="p">,</span>
+        <span class="n">detections_per_img</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">200</span><span class="p">,</span>
+        <span class="n">iou_thresh</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">topk_candidates</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">400</span><span class="p">,</span>
+        <span class="n">positive_fraction</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.25</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">anchor_generator</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">BoxCoder</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="mf">10.0</span><span class="p">,</span> <span class="mf">10.0</span><span class="p">,</span> <span class="mf">5.0</span><span class="p">,</span> <span class="mf">5.0</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="n">head</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="s2">&quot;out_channels&quot;</span><span class="p">):</span>
+                <span class="n">out_channels</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">out_channels</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">out_channels</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">retrieve_out_channels</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">out_channels</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">anchor_generator</span><span class="o">.</span><span class="n">aspect_ratios</span><span class="p">):</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;The length of the output channels from the backbone (</span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">out_channels</span><span class="p">)</span><span class="si">}</span><span class="s2">) do not match the length of the anchor generator aspect ratios (</span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">anchor_generator</span><span class="o">.</span><span class="n">aspect_ratios</span><span class="p">)</span><span class="si">}</span><span class="s2">)&quot;</span>
+                <span class="p">)</span>
+
+            <span class="n">num_anchors</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()</span>
+            <span class="n">head</span> <span class="o">=</span> <span class="n">SSDHead</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head</span> <span class="o">=</span> <span class="n">head</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">proposal_matcher</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">SSDMatcher</span><span class="p">(</span><span class="n">iou_thresh</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">image_mean</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_mean</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">image_std</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_std</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">GeneralizedRCNNTransform</span><span class="p">(</span>
+            <span class="nb">min</span><span class="p">(</span><span class="n">size</span><span class="p">),</span> <span class="nb">max</span><span class="p">(</span><span class="n">size</span><span class="p">),</span> <span class="n">image_mean</span><span class="p">,</span> <span class="n">image_std</span><span class="p">,</span> <span class="n">size_divisible</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">fixed_size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">score_thresh</span> <span class="o">=</span> <span class="n">score_thresh</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">nms_thresh</span> <span class="o">=</span> <span class="n">nms_thresh</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">detections_per_img</span> <span class="o">=</span> <span class="n">detections_per_img</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">topk_candidates</span> <span class="o">=</span> <span class="n">topk_candidates</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">neg_to_pos_ratio</span> <span class="o">=</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">positive_fraction</span><span class="p">)</span> <span class="o">/</span> <span class="n">positive_fraction</span>
+
+        <span class="c1"># used only on torchscript mode</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">eager_outputs</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">losses</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">losses</span>
+
+        <span class="k">return</span> <span class="n">detections</span>
+
+    <span class="k">def</span> <span class="nf">compute_loss</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">targets</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]],</span>
+        <span class="n">head_outputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">matched_idxs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_regression&quot;</span><span class="p">]</span>
+        <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">]</span>
+
+        <span class="c1"># Match original targets with default boxes</span>
+        <span class="n">num_foreground</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="n">bbox_loss</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">cls_targets</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="p">(</span>
+            <span class="n">targets_per_image</span><span class="p">,</span>
+            <span class="n">bbox_regression_per_image</span><span class="p">,</span>
+            <span class="n">cls_logits_per_image</span><span class="p">,</span>
+            <span class="n">anchors_per_image</span><span class="p">,</span>
+            <span class="n">matched_idxs_per_image</span><span class="p">,</span>
+        <span class="p">)</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">bbox_regression</span><span class="p">,</span> <span class="n">cls_logits</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">):</span>
+            <span class="c1"># produce the matching between boxes and targets</span>
+            <span class="n">foreground_idxs_per_image</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">matched_idxs_per_image</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">foreground_matched_idxs_per_image</span> <span class="o">=</span> <span class="n">matched_idxs_per_image</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">]</span>
+            <span class="n">num_foreground</span> <span class="o">+=</span> <span class="n">foreground_matched_idxs_per_image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span>
+
+            <span class="c1"># Calculate regression loss</span>
+            <span class="n">matched_gt_boxes_per_image</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">][</span><span class="n">foreground_matched_idxs_per_image</span><span class="p">]</span>
+            <span class="n">bbox_regression_per_image</span> <span class="o">=</span> <span class="n">bbox_regression_per_image</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">,</span> <span class="p">:]</span>
+            <span class="n">anchors_per_image</span> <span class="o">=</span> <span class="n">anchors_per_image</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">,</span> <span class="p">:]</span>
+            <span class="n">target_regression</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span><span class="o">.</span><span class="n">encode_single</span><span class="p">(</span><span class="n">matched_gt_boxes_per_image</span><span class="p">,</span> <span class="n">anchors_per_image</span><span class="p">)</span>
+            <span class="n">bbox_loss</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">smooth_l1_loss</span><span class="p">(</span><span class="n">bbox_regression_per_image</span><span class="p">,</span> <span class="n">target_regression</span><span class="p">,</span> <span class="n">reduction</span><span class="o">=</span><span class="s2">&quot;sum&quot;</span><span class="p">)</span>
+            <span class="p">)</span>
+
+            <span class="c1"># Estimate ground truth for class targets</span>
+            <span class="n">gt_classes_target</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span>
+                <span class="p">(</span><span class="n">cls_logits_per_image</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),),</span>
+                <span class="n">dtype</span><span class="o">=</span><span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
+                <span class="n">device</span><span class="o">=</span><span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">device</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">gt_classes_target</span><span class="p">[</span><span class="n">foreground_idxs_per_image</span><span class="p">]</span> <span class="o">=</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span>
+                <span class="n">foreground_matched_idxs_per_image</span>
+            <span class="p">]</span>
+            <span class="n">cls_targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">gt_classes_target</span><span class="p">)</span>
+
+        <span class="n">bbox_loss</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">bbox_loss</span><span class="p">)</span>
+        <span class="n">cls_targets</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">cls_targets</span><span class="p">)</span>
+
+        <span class="c1"># Calculate classification loss</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">cls_logits</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">cls_loss</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">cross_entropy</span><span class="p">(</span><span class="n">cls_logits</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">),</span> <span class="n">cls_targets</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">),</span> <span class="n">reduction</span><span class="o">=</span><span class="s2">&quot;none&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+            <span class="n">cls_targets</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+        <span class="p">)</span>
+
+        <span class="c1"># Hard Negative Sampling</span>
+        <span class="n">foreground_idxs</span> <span class="o">=</span> <span class="n">cls_targets</span> <span class="o">&gt;</span> <span class="mi">0</span>
+        <span class="n">num_negative</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">neg_to_pos_ratio</span> <span class="o">*</span> <span class="n">foreground_idxs</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># num_negative[num_negative &lt; self.neg_to_pos_ratio] = self.neg_to_pos_ratio</span>
+        <span class="n">negative_loss</span> <span class="o">=</span> <span class="n">cls_loss</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+        <span class="n">negative_loss</span><span class="p">[</span><span class="n">foreground_idxs</span><span class="p">]</span> <span class="o">=</span> <span class="o">-</span><span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)</span>  <span class="c1"># use -inf to detect positive values that creeped in the sample</span>
+        <span class="n">values</span><span class="p">,</span> <span class="n">idx</span> <span class="o">=</span> <span class="n">negative_loss</span><span class="o">.</span><span class="n">sort</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">descending</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># background_idxs = torch.logical_and(idx.sort(1)[1] &lt; num_negative, torch.isfinite(values))</span>
+        <span class="n">background_idxs</span> <span class="o">=</span> <span class="n">idx</span><span class="o">.</span><span class="n">sort</span><span class="p">(</span><span class="mi">1</span><span class="p">)[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">num_negative</span>
+
+        <span class="n">N</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_foreground</span><span class="p">)</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;bbox_regression&quot;</span><span class="p">:</span> <span class="n">bbox_loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span> <span class="o">/</span> <span class="n">N</span><span class="p">,</span>
+            <span class="s2">&quot;classification&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">cls_loss</span><span class="p">[</span><span class="n">foreground_idxs</span><span class="p">]</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span> <span class="o">+</span> <span class="n">cls_loss</span><span class="p">[</span><span class="n">background_idxs</span><span class="p">]</span><span class="o">.</span><span class="n">sum</span><span class="p">())</span> <span class="o">/</span> <span class="n">N</span><span class="p">,</span>
+        <span class="p">}</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">images</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">targets</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;targets should not be none when in training mode&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">target</span> <span class="ow">in</span> <span class="n">targets</span><span class="p">:</span>
+                    <span class="n">boxes</span> <span class="o">=</span> <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+                        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                            <span class="nb">len</span><span class="p">(</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">4</span><span class="p">,</span>
+                            <span class="sa">f</span><span class="s2">&quot;Expected target boxes to be a tensor of shape [N, 4], got </span><span class="si">{</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">,</span>
+                        <span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;Expected target boxes to be of type Tensor, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="c1"># get the original image sizes</span>
+        <span class="n">original_image_sizes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">img</span> <span class="ow">in</span> <span class="n">images</span><span class="p">:</span>
+            <span class="n">val</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                <span class="nb">len</span><span class="p">(</span><span class="n">val</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">,</span>
+                <span class="sa">f</span><span class="s2">&quot;expecting the last two dimensions of the Tensor to be H and W instead got </span><span class="si">{</span><span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">original_image_sizes</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">val</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">val</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+
+        <span class="c1"># transform the input</span>
+        <span class="n">images</span><span class="p">,</span> <span class="n">targets</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">targets</span><span class="p">)</span>
+
+        <span class="c1"># Check for degenerate boxes</span>
+        <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">target_idx</span><span class="p">,</span> <span class="n">target</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">targets</span><span class="p">):</span>
+                <span class="n">boxes</span> <span class="o">=</span> <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span>
+                <span class="n">degenerate_boxes</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:]</span> <span class="o">&lt;=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">degenerate_boxes</span><span class="o">.</span><span class="n">any</span><span class="p">():</span>
+                    <span class="n">bb_idx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">degenerate_boxes</span><span class="o">.</span><span class="n">any</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">))[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                    <span class="n">degen_bb</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[</span><span class="n">bb_idx</span><span class="p">]</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+                        <span class="kc">False</span><span class="p">,</span>
+                        <span class="s2">&quot;All bounding boxes should have positive height and width.&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot; Found invalid box </span><span class="si">{</span><span class="n">degen_bb</span><span class="si">}</span><span class="s2"> for target at index </span><span class="si">{</span><span class="n">target_idx</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">,</span>
+                    <span class="p">)</span>
+
+        <span class="c1"># get the features from the backbone</span>
+        <span class="n">features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span><span class="p">(</span><span class="n">images</span><span class="o">.</span><span class="n">tensors</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">features</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="n">features</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">([(</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="n">features</span><span class="p">)])</span>
+
+        <span class="n">features</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">features</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+
+        <span class="c1"># compute the ssd heads outputs using the features</span>
+        <span class="n">head_outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+
+        <span class="c1"># create the set of anchors</span>
+        <span class="n">anchors</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">anchor_generator</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">features</span><span class="p">)</span>
+
+        <span class="n">losses</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="n">matched_idxs</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">if</span> <span class="n">targets</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;targets should not be none when in training mode&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">anchors_per_image</span><span class="p">,</span> <span class="n">targets_per_image</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">anchors</span><span class="p">,</span> <span class="n">targets</span><span class="p">):</span>
+                    <span class="k">if</span> <span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">matched_idxs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                            <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">(</span>
+                                <span class="p">(</span><span class="n">anchors_per_image</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),),</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">anchors_per_image</span><span class="o">.</span><span class="n">device</span>
+                            <span class="p">)</span>
+                        <span class="p">)</span>
+                        <span class="k">continue</span>
+
+                    <span class="n">match_quality_matrix</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">box_iou</span><span class="p">(</span><span class="n">targets_per_image</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">],</span> <span class="n">anchors_per_image</span><span class="p">)</span>
+                    <span class="n">matched_idxs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">proposal_matcher</span><span class="p">(</span><span class="n">match_quality_matrix</span><span class="p">))</span>
+
+                <span class="n">losses</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">matched_idxs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">detections</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">postprocess_detections</span><span class="p">(</span><span class="n">head_outputs</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">images</span><span class="o">.</span><span class="n">image_sizes</span><span class="p">)</span>
+            <span class="n">detections</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="o">.</span><span class="n">postprocess</span><span class="p">(</span><span class="n">detections</span><span class="p">,</span> <span class="n">images</span><span class="o">.</span><span class="n">image_sizes</span><span class="p">,</span> <span class="n">original_image_sizes</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;SSD always returns a (Losses, Detections) tuple in scripting&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_has_warned</span> <span class="o">=</span> <span class="kc">True</span>
+            <span class="k">return</span> <span class="n">losses</span><span class="p">,</span> <span class="n">detections</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">eager_outputs</span><span class="p">(</span><span class="n">losses</span><span class="p">,</span> <span class="n">detections</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">postprocess_detections</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">head_outputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">image_anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">image_shapes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]:</span>
+        <span class="n">bbox_regression</span> <span class="o">=</span> <span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;bbox_regression&quot;</span><span class="p">]</span>
+        <span class="n">pred_scores</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">head_outputs</span><span class="p">[</span><span class="s2">&quot;cls_logits&quot;</span><span class="p">],</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">pred_scores</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">device</span> <span class="o">=</span> <span class="n">pred_scores</span><span class="o">.</span><span class="n">device</span>
+
+        <span class="n">detections</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">boxes</span><span class="p">,</span> <span class="n">scores</span><span class="p">,</span> <span class="n">anchors</span><span class="p">,</span> <span class="n">image_shape</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">bbox_regression</span><span class="p">,</span> <span class="n">pred_scores</span><span class="p">,</span> <span class="n">image_anchors</span><span class="p">,</span> <span class="n">image_shapes</span><span class="p">):</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">box_coder</span><span class="o">.</span><span class="n">decode_single</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">anchors</span><span class="p">)</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">clip_boxes_to_image</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">image_shape</span><span class="p">)</span>
+
+            <span class="n">image_boxes</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">image_scores</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">image_labels</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">label</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">):</span>
+                <span class="n">score</span> <span class="o">=</span> <span class="n">scores</span><span class="p">[:,</span> <span class="n">label</span><span class="p">]</span>
+
+                <span class="n">keep_idxs</span> <span class="o">=</span> <span class="n">score</span> <span class="o">&gt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">score_thresh</span>
+                <span class="n">score</span> <span class="o">=</span> <span class="n">score</span><span class="p">[</span><span class="n">keep_idxs</span><span class="p">]</span>
+                <span class="n">box</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[</span><span class="n">keep_idxs</span><span class="p">]</span>
+
+                <span class="c1"># keep only topk scoring predictions</span>
+                <span class="n">num_topk</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">_topk_min</span><span class="p">(</span><span class="n">score</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">topk_candidates</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="n">score</span><span class="p">,</span> <span class="n">idxs</span> <span class="o">=</span> <span class="n">score</span><span class="o">.</span><span class="n">topk</span><span class="p">(</span><span class="n">num_topk</span><span class="p">)</span>
+                <span class="n">box</span> <span class="o">=</span> <span class="n">box</span><span class="p">[</span><span class="n">idxs</span><span class="p">]</span>
+
+                <span class="n">image_boxes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">box</span><span class="p">)</span>
+                <span class="n">image_scores</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">score</span><span class="p">)</span>
+                <span class="n">image_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">full_like</span><span class="p">(</span><span class="n">score</span><span class="p">,</span> <span class="n">fill_value</span><span class="o">=</span><span class="n">label</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">))</span>
+
+            <span class="n">image_boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_boxes</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_scores</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_scores</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">image_labels</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+            <span class="c1"># non-maximum suppression</span>
+            <span class="n">keep</span> <span class="o">=</span> <span class="n">box_ops</span><span class="o">.</span><span class="n">batched_nms</span><span class="p">(</span><span class="n">image_boxes</span><span class="p">,</span> <span class="n">image_scores</span><span class="p">,</span> <span class="n">image_labels</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">nms_thresh</span><span class="p">)</span>
+            <span class="n">keep</span> <span class="o">=</span> <span class="n">keep</span><span class="p">[:</span> <span class="bp">self</span><span class="o">.</span><span class="n">detections_per_img</span><span class="p">]</span>
+
+            <span class="n">detections</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;boxes&quot;</span><span class="p">:</span> <span class="n">image_boxes</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                    <span class="s2">&quot;scores&quot;</span><span class="p">:</span> <span class="n">image_scores</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                    <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="n">image_labels</span><span class="p">[</span><span class="n">keep</span><span class="p">],</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="n">detections</span>
+
+
+<span class="k">class</span> <span class="nc">SSDFeatureExtractorVGG</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">backbone</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">highres</span><span class="p">:</span> <span class="nb">bool</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="n">_</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">maxpool3_pos</span><span class="p">,</span> <span class="n">maxpool4_pos</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="p">(</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">layer</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">))</span>
+
+        <span class="c1"># Patch ceil_mode for maxpool3 to get the same WxH output sizes as the paper</span>
+        <span class="n">backbone</span><span class="p">[</span><span class="n">maxpool3_pos</span><span class="p">]</span><span class="o">.</span><span class="n">ceil_mode</span> <span class="o">=</span> <span class="kc">True</span>
+
+        <span class="c1"># parameters used for L2 regularization + rescaling</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale_weight</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="mi">512</span><span class="p">)</span> <span class="o">*</span> <span class="mi">20</span><span class="p">)</span>
+
+        <span class="c1"># Multiple Feature maps - page 4, Fig 2 of SSD paper</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">backbone</span><span class="p">[:</span><span class="n">maxpool4_pos</span><span class="p">])</span>  <span class="c1"># until conv4_3</span>
+
+        <span class="c1"># SSD300 case - page 4, Fig 2 of SSD paper</span>
+        <span class="n">extra</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">(</span>
+            <span class="p">[</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>  <span class="c1"># conv8_2</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>  <span class="c1"># conv9_2</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">),</span>  <span class="c1"># conv10_2</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">),</span>  <span class="c1"># conv11_2</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="p">),</span>
+            <span class="p">]</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">highres</span><span class="p">:</span>
+            <span class="c1"># Additional layers for the SSD512 case. See page 11, footernote 5.</span>
+            <span class="n">extra</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">4</span><span class="p">),</span>  <span class="c1"># conv12_2</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+        <span class="n">_xavier_init</span><span class="p">(</span><span class="n">extra</span><span class="p">)</span>
+
+        <span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>  <span class="c1"># add modified maxpool5</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">6</span><span class="p">,</span> <span class="n">dilation</span><span class="o">=</span><span class="mi">6</span><span class="p">),</span>  <span class="c1"># FC6 with atrous</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>  <span class="c1"># FC7</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="n">_xavier_init</span><span class="p">(</span><span class="n">fc</span><span class="p">)</span>
+        <span class="n">extra</span><span class="o">.</span><span class="n">insert</span><span class="p">(</span>
+            <span class="mi">0</span><span class="p">,</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                <span class="o">*</span><span class="n">backbone</span><span class="p">[</span><span class="n">maxpool4_pos</span><span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>  <span class="c1"># until conv5_3, skip maxpool5</span>
+                <span class="n">fc</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">extra</span> <span class="o">=</span> <span class="n">extra</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="c1"># L2 regularization + Rescaling of 1st block&#39;s feature map</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">rescaled</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale_weight</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">F</span><span class="o">.</span><span class="n">normalize</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="p">[</span><span class="n">rescaled</span><span class="p">]</span>
+
+        <span class="c1"># Calculating Feature maps for the rest blocks</span>
+        <span class="k">for</span> <span class="n">block</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">extra</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">block</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+            <span class="n">output</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">OrderedDict</span><span class="p">([(</span><span class="nb">str</span><span class="p">(</span><span class="n">i</span><span class="p">),</span> <span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">output</span><span class="p">)])</span>
+
+
+<span class="k">def</span> <span class="nf">_vgg_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">:</span> <span class="n">VGG</span><span class="p">,</span> <span class="n">highres</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span> <span class="n">trainable_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">features</span>
+    <span class="c1"># Gather the indices of maxpools. These are the locations of output blocks.</span>
+    <span class="n">stage_indices</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">)][:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+    <span class="n">num_stages</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">stage_indices</span><span class="p">)</span>
+
+    <span class="c1"># find the index of the layer from which we won&#39;t freeze</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+        <span class="mi">0</span> <span class="o">&lt;=</span> <span class="n">trainable_layers</span> <span class="o">&lt;=</span> <span class="n">num_stages</span><span class="p">,</span>
+        <span class="sa">f</span><span class="s2">&quot;trainable_layers should be in the range [0, </span><span class="si">{</span><span class="n">num_stages</span><span class="si">}</span><span class="s2">]. Instead got </span><span class="si">{</span><span class="n">trainable_layers</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">freeze_before</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="k">if</span> <span class="n">trainable_layers</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">stage_indices</span><span class="p">[</span><span class="n">num_stages</span> <span class="o">-</span> <span class="n">trainable_layers</span><span class="p">]</span>
+
+    <span class="k">for</span> <span class="n">b</span> <span class="ow">in</span> <span class="n">backbone</span><span class="p">[:</span><span class="n">freeze_before</span><span class="p">]:</span>
+        <span class="k">for</span> <span class="n">parameter</span> <span class="ow">in</span> <span class="n">b</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+            <span class="n">parameter</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">(</span><span class="kc">False</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">SSDFeatureExtractorVGG</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">highres</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="ssd300_vgg16"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.ssd300_vgg16.html#torchvision.models.detection.ssd300_vgg16">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">SSD300_VGG16_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">VGG16_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_FEATURES</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">ssd300_vgg16</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">SSD300_VGG16_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG16_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">VGG16_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_FEATURES</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SSD</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The SSD300 model is based on the `SSD: Single Shot MultiBox Detector</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1512.02325&gt;`_ paper.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each</span>
+<span class="sd">    image, and should be in 0-1 range. Different images can have different sizes, but they will be resized</span>
+<span class="sd">    to a fixed size before passing it to the backbone.</span>
+
+<span class="sd">    The behavior of the model changes depending on if it is in training or evaluation mode.</span>
+
+<span class="sd">    During training, the model expects both the input tensors and targets (list of dictionary),</span>
+<span class="sd">    containing:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the ground-truth boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the class label for each ground-truth box</span>
+
+<span class="sd">    The model returns a Dict[Tensor] during training, containing the classification and regression</span>
+<span class="sd">    losses.</span>
+
+<span class="sd">    During inference, the model requires only the input tensors, and returns the post-processed</span>
+<span class="sd">    predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as</span>
+<span class="sd">    follows, where ``N`` is the number of detections:</span>
+
+<span class="sd">        - boxes (``FloatTensor[N, 4]``): the predicted boxes in ``[x1, y1, x2, y2]`` format, with</span>
+<span class="sd">          ``0 &lt;= x1 &lt; x2 &lt;= W`` and ``0 &lt;= y1 &lt; y2 &lt;= H``.</span>
+<span class="sd">        - labels (Int64Tensor[N]): the predicted labels for each detection</span>
+<span class="sd">        - scores (Tensor[N]): the scores for each detection</span>
+
+<span class="sd">    Example:</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.ssd300_vgg16(weights=SSD300_VGG16_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 300, 300), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.SSD300_VGG16_Weights`, optional): The pretrained</span>
+<span class="sd">                weights to use. See</span>
+<span class="sd">                :class:`~torchvision.models.detection.SSD300_VGG16_Weights`</span>
+<span class="sd">                below for more details, and possible values. By default, no</span>
+<span class="sd">                pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr</span>
+<span class="sd">            Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.VGG16_Weights`, optional): The pretrained weights for the</span>
+<span class="sd">            backbone</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers starting from final block.</span>
+<span class="sd">            Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If ``None`` is</span>
+<span class="sd">            passed (the default) this value is set to 4.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.SSD``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.SSD300_VGG16_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">SSD300_VGG16_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">VGG16_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="s2">&quot;size&quot;</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;The size of the model is already fixed; ignoring the parameter.&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span>
+        <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">4</span>
+    <span class="p">)</span>
+
+    <span class="c1"># Use custom backbones more appropriate for SSD</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">vgg16</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_vgg_extractor</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">)</span>
+    <span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">DefaultBoxGenerator</span><span class="p">(</span>
+        <span class="p">[[</span><span class="mi">2</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">]],</span>
+        <span class="n">scales</span><span class="o">=</span><span class="p">[</span><span class="mf">0.07</span><span class="p">,</span> <span class="mf">0.15</span><span class="p">,</span> <span class="mf">0.33</span><span class="p">,</span> <span class="mf">0.51</span><span class="p">,</span> <span class="mf">0.69</span><span class="p">,</span> <span class="mf">0.87</span><span class="p">,</span> <span class="mf">1.05</span><span class="p">],</span>
+        <span class="n">steps</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">100</span><span class="p">,</span> <span class="mi">300</span><span class="p">],</span>
+    <span class="p">)</span>
+
+    <span class="n">defaults</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="c1"># Rescale the input in a way compatible to the backbone</span>
+        <span class="s2">&quot;image_mean&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mf">0.48235</span><span class="p">,</span> <span class="mf">0.45882</span><span class="p">,</span> <span class="mf">0.40784</span><span class="p">],</span>
+        <span class="s2">&quot;image_std&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mf">1.0</span> <span class="o">/</span> <span class="mf">255.0</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="mf">255.0</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="mf">255.0</span><span class="p">],</span>  <span class="c1"># undo the 0-1 scaling of toTensor</span>
+    <span class="p">}</span>
+    <span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">{</span><span class="o">**</span><span class="n">defaults</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">}</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">SSD</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">anchor_generator</span><span class="p">,</span> <span class="p">(</span><span class="mi">300</span><span class="p">,</span> <span class="mi">300</span><span class="p">),</span> <span class="n">num_classes</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/detection/ssdlite.html b/0.16/_modules/torchvision/models/detection/ssdlite.html
new file mode 100644
index 00000000000..bb6e3dfd8c3
--- /dev/null
+++ b/0.16/_modules/torchvision/models/detection/ssdlite.html
@@ -0,0 +1,1000 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.detection.ssdlite &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.detection.ssdlite</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.detection.ssdlite</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">...ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ObjectDetection</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">..</span> <span class="kn">import</span> <span class="n">mobilenet</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_COCO_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..mobilenetv3</span> <span class="kn">import</span> <span class="n">mobilenet_v3_large</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span>
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">_utils</span> <span class="k">as</span> <span class="n">det_utils</span>
+<span class="kn">from</span> <span class="nn">.anchor_utils</span> <span class="kn">import</span> <span class="n">DefaultBoxGenerator</span>
+<span class="kn">from</span> <span class="nn">.backbone_utils</span> <span class="kn">import</span> <span class="n">_validate_trainable_layers</span>
+<span class="kn">from</span> <span class="nn">.ssd</span> <span class="kn">import</span> <span class="n">SSD</span><span class="p">,</span> <span class="n">SSDScoringHead</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;SSDLite320_MobileNet_V3_Large_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ssdlite320_mobilenet_v3_large&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="c1"># Building blocks of SSDlite as described in section 6.2 of MobileNetV2 paper</span>
+<span class="k">def</span> <span class="nf">_prediction_block</span><span class="p">(</span>
+    <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+        <span class="c1"># 3x3 depthwise with stride 1 and padding 1</span>
+        <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span>
+            <span class="n">groups</span><span class="o">=</span><span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU6</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="c1"># 1x1 projetion to output channels</span>
+        <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_extra_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">:</span>
+    <span class="n">activation</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU6</span>
+    <span class="n">intermediate_channels</span> <span class="o">=</span> <span class="n">out_channels</span> <span class="o">//</span> <span class="mi">2</span>
+    <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+        <span class="c1"># 1x1 projection to half output channels</span>
+        <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span> <span class="n">intermediate_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation</span>
+        <span class="p">),</span>
+        <span class="c1"># 3x3 depthwise with stride 2 and padding 1</span>
+        <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">intermediate_channels</span><span class="p">,</span>
+            <span class="n">intermediate_channels</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+            <span class="n">groups</span><span class="o">=</span><span class="n">intermediate_channels</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="c1"># 1x1 projetion to output channels</span>
+        <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">intermediate_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation</span>
+        <span class="p">),</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_normal_init</span><span class="p">(</span><span class="n">conv</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="n">conv</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.03</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">layer</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SSDLiteHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span> <span class="o">=</span> <span class="n">SSDLiteClassificationHead</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span> <span class="o">=</span> <span class="n">SSDLiteRegressionHead</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;bbox_regression&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">regression_head</span><span class="p">(</span><span class="n">x</span><span class="p">),</span>
+            <span class="s2">&quot;cls_logits&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">classification_head</span><span class="p">(</span><span class="n">x</span><span class="p">),</span>
+        <span class="p">}</span>
+
+
+<span class="k">class</span> <span class="nc">SSDLiteClassificationHead</span><span class="p">(</span><span class="n">SSDScoringHead</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span>
+    <span class="p">):</span>
+        <span class="n">cls_logits</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">channels</span><span class="p">,</span> <span class="n">anchors</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">):</span>
+            <span class="n">cls_logits</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">_prediction_block</span><span class="p">(</span><span class="n">channels</span><span class="p">,</span> <span class="n">num_classes</span> <span class="o">*</span> <span class="n">anchors</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">))</span>
+        <span class="n">_normal_init</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">cls_logits</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SSDLiteRegressionHead</span><span class="p">(</span><span class="n">SSDScoringHead</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">num_anchors</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]):</span>
+        <span class="n">bbox_reg</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">channels</span><span class="p">,</span> <span class="n">anchors</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">):</span>
+            <span class="n">bbox_reg</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">_prediction_block</span><span class="p">(</span><span class="n">channels</span><span class="p">,</span> <span class="mi">4</span> <span class="o">*</span> <span class="n">anchors</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">))</span>
+        <span class="n">_normal_init</span><span class="p">(</span><span class="n">bbox_reg</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">bbox_reg</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SSDLiteFeatureExtractorMobileNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">backbone</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span>
+        <span class="n">c4_pos</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">width_mult</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">min_depth</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">16</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">backbone</span><span class="p">[</span><span class="n">c4_pos</span><span class="p">]</span><span class="o">.</span><span class="n">use_res_connect</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;backbone[c4_pos].use_res_connect should be False&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="c1"># As described in section 6.3 of MobileNetV3 paper</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">backbone</span><span class="p">[:</span><span class="n">c4_pos</span><span class="p">],</span> <span class="n">backbone</span><span class="p">[</span><span class="n">c4_pos</span><span class="p">]</span><span class="o">.</span><span class="n">block</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span>  <span class="c1"># from start until C4 expansion layer</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">backbone</span><span class="p">[</span><span class="n">c4_pos</span><span class="p">]</span><span class="o">.</span><span class="n">block</span><span class="p">[</span><span class="mi">1</span><span class="p">:],</span> <span class="o">*</span><span class="n">backbone</span><span class="p">[</span><span class="n">c4_pos</span> <span class="o">+</span> <span class="mi">1</span> <span class="p">:]),</span>  <span class="c1"># from C4 depthwise until end</span>
+        <span class="p">)</span>
+
+        <span class="n">get_depth</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">d</span><span class="p">:</span> <span class="nb">max</span><span class="p">(</span><span class="n">min_depth</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">d</span> <span class="o">*</span> <span class="n">width_mult</span><span class="p">))</span>  <span class="c1"># noqa: E731</span>
+        <span class="n">extra</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">(</span>
+            <span class="p">[</span>
+                <span class="n">_extra_block</span><span class="p">(</span><span class="n">backbone</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">get_depth</span><span class="p">(</span><span class="mi">512</span><span class="p">),</span> <span class="n">norm_layer</span><span class="p">),</span>
+                <span class="n">_extra_block</span><span class="p">(</span><span class="n">get_depth</span><span class="p">(</span><span class="mi">512</span><span class="p">),</span> <span class="n">get_depth</span><span class="p">(</span><span class="mi">256</span><span class="p">),</span> <span class="n">norm_layer</span><span class="p">),</span>
+                <span class="n">_extra_block</span><span class="p">(</span><span class="n">get_depth</span><span class="p">(</span><span class="mi">256</span><span class="p">),</span> <span class="n">get_depth</span><span class="p">(</span><span class="mi">256</span><span class="p">),</span> <span class="n">norm_layer</span><span class="p">),</span>
+                <span class="n">_extra_block</span><span class="p">(</span><span class="n">get_depth</span><span class="p">(</span><span class="mi">256</span><span class="p">),</span> <span class="n">get_depth</span><span class="p">(</span><span class="mi">128</span><span class="p">),</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="p">]</span>
+        <span class="p">)</span>
+        <span class="n">_normal_init</span><span class="p">(</span><span class="n">extra</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">extra</span> <span class="o">=</span> <span class="n">extra</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="c1"># Get feature maps from backbone and extra. Can&#39;t be refactored due to JIT limitations.</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">block</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">block</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+            <span class="n">output</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">block</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">extra</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">block</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+            <span class="n">output</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">OrderedDict</span><span class="p">([(</span><span class="nb">str</span><span class="p">(</span><span class="n">i</span><span class="p">),</span> <span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">output</span><span class="p">)])</span>
+
+
+<span class="k">def</span> <span class="nf">_mobilenet_extractor</span><span class="p">(</span>
+    <span class="n">backbone</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">mobilenet</span><span class="o">.</span><span class="n">MobileNetV2</span><span class="p">,</span> <span class="n">mobilenet</span><span class="o">.</span><span class="n">MobileNetV3</span><span class="p">],</span>
+    <span class="n">trainable_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+<span class="p">):</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">features</span>
+    <span class="c1"># Gather the indices of blocks which are strided. These are the locations of C1, ..., Cn-1 blocks.</span>
+    <span class="c1"># The first and last blocks are always included because they are the C0 (conv1) and Cn.</span>
+    <span class="n">stage_indices</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="s2">&quot;_is_cn&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)]</span> <span class="o">+</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">]</span>
+    <span class="n">num_stages</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">stage_indices</span><span class="p">)</span>
+
+    <span class="c1"># find the index of the layer from which we won&#39;t freeze</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="mi">0</span> <span class="o">&lt;=</span> <span class="n">trainable_layers</span> <span class="o">&lt;=</span> <span class="n">num_stages</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;trainable_layers should be in the range [0, </span><span class="si">{num_stages}</span><span class="s2">], instead got </span><span class="si">{trainable_layers}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="n">freeze_before</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="k">if</span> <span class="n">trainable_layers</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">stage_indices</span><span class="p">[</span><span class="n">num_stages</span> <span class="o">-</span> <span class="n">trainable_layers</span><span class="p">]</span>
+
+    <span class="k">for</span> <span class="n">b</span> <span class="ow">in</span> <span class="n">backbone</span><span class="p">[:</span><span class="n">freeze_before</span><span class="p">]:</span>
+        <span class="k">for</span> <span class="n">parameter</span> <span class="ow">in</span> <span class="n">b</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+            <span class="n">parameter</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">(</span><span class="kc">False</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">SSDLiteFeatureExtractorMobileNet</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">stage_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">],</span> <span class="n">norm_layer</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="SSDLite320_MobileNet_V3_Large_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html#torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights">[docs]</a><span class="k">class</span> <span class="nc">SSDLite320_MobileNet_V3_Large_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/ssdlite320_mobilenet_v3_large_coco-a79551df.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">ObjectDetection</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">3440060</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_COCO_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/detection#ssdlite320-mobilenetv3-large&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;box_map&quot;</span><span class="p">:</span> <span class="mf">21.3</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.583</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">13.418</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were produced by following a similar training recipe as on the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_V1</span></div>
+
+
+<div class="viewcode-block" id="ssdlite320_mobilenet_v3_large"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html#torchvision.models.detection.ssdlite320_mobilenet_v3_large">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">SSDLite320_MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">COCO_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">ssdlite320_mobilenet_v3_large</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">SSDLite320_MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="n">trainable_backbone_layers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SSD</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;SSDlite model architecture with input size 320x320 and a MobileNetV3 Large backbone, as</span>
+<span class="sd">    described at `Searching for MobileNetV3 &lt;https://arxiv.org/abs/1905.02244&gt;`__ and</span>
+<span class="sd">    `MobileNetV2: Inverted Residuals and Linear Bottlenecks &lt;https://arxiv.org/abs/1801.04381&gt;`__.</span>
+
+<span class="sd">    .. betastatus:: detection module</span>
+
+<span class="sd">    See :func:`~torchvision.models.detection.ssd300_vgg16` for more details.</span>
+
+<span class="sd">    Example:</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.detection.ssdlite320_mobilenet_v3_large(weights=SSDLite320_MobileNet_V3_Large_Weights.DEFAULT)</span>
+<span class="sd">        &gt;&gt;&gt; model.eval()</span>
+<span class="sd">        &gt;&gt;&gt; x = [torch.rand(3, 320, 320), torch.rand(3, 500, 400)]</span>
+<span class="sd">        &gt;&gt;&gt; predictions = model(x)</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model</span>
+<span class="sd">            (including the background).</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The pretrained</span>
+<span class="sd">            weights for the backbone.</span>
+<span class="sd">        trainable_backbone_layers (int, optional): number of trainable (not frozen) layers</span>
+<span class="sd">            starting from final block. Valid values are between 0 and 6, with 6 meaning all</span>
+<span class="sd">            backbone layers are trainable. If ``None`` is passed (the default) this value is</span>
+<span class="sd">            set to 6.</span>
+<span class="sd">        norm_layer (callable, optional): Module specifying the normalization layer to use.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.detection.ssd.SSD``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">SSDLite320_MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="s2">&quot;size&quot;</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;The size of the model is already fixed; ignoring the parameter.&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">91</span>
+
+    <span class="n">trainable_backbone_layers</span> <span class="o">=</span> <span class="n">_validate_trainable_layers</span><span class="p">(</span>
+        <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">,</span> <span class="n">trainable_backbone_layers</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">6</span>
+    <span class="p">)</span>
+
+    <span class="c1"># Enable reduced tail if no pretrained backbone is selected. See Table 6 of MobileNetV3 paper.</span>
+    <span class="n">reduce_tail</span> <span class="o">=</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="kc">None</span>
+
+    <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.03</span><span class="p">)</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">mobilenet_v3_large</span><span class="p">(</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">reduced_tail</span><span class="o">=</span><span class="n">reduce_tail</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+    <span class="k">if</span> <span class="n">weights_backbone</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># Change the default initialization scheme if not pretrained</span>
+        <span class="n">_normal_init</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">_mobilenet_extractor</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">trainable_backbone_layers</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="n">size</span> <span class="o">=</span> <span class="p">(</span><span class="mi">320</span><span class="p">,</span> <span class="mi">320</span><span class="p">)</span>
+    <span class="n">anchor_generator</span> <span class="o">=</span> <span class="n">DefaultBoxGenerator</span><span class="p">([[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">6</span><span class="p">)],</span> <span class="n">min_ratio</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span> <span class="n">max_ratio</span><span class="o">=</span><span class="mf">0.95</span><span class="p">)</span>
+    <span class="n">out_channels</span> <span class="o">=</span> <span class="n">det_utils</span><span class="o">.</span><span class="n">retrieve_out_channels</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+    <span class="n">num_anchors</span> <span class="o">=</span> <span class="n">anchor_generator</span><span class="o">.</span><span class="n">num_anchors_per_location</span><span class="p">()</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">out_channels</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">anchor_generator</span><span class="o">.</span><span class="n">aspect_ratios</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;The length of the output channels from the backbone </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">out_channels</span><span class="p">)</span><span class="si">}</span><span class="s2"> do not match the length of the anchor generator aspect ratios </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">anchor_generator</span><span class="o">.</span><span class="n">aspect_ratios</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+    <span class="n">defaults</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;score_thresh&quot;</span><span class="p">:</span> <span class="mf">0.001</span><span class="p">,</span>
+        <span class="s2">&quot;nms_thresh&quot;</span><span class="p">:</span> <span class="mf">0.55</span><span class="p">,</span>
+        <span class="s2">&quot;detections_per_img&quot;</span><span class="p">:</span> <span class="mi">300</span><span class="p">,</span>
+        <span class="s2">&quot;topk_candidates&quot;</span><span class="p">:</span> <span class="mi">300</span><span class="p">,</span>
+        <span class="c1"># Rescale the input in a way compatible to the backbone:</span>
+        <span class="c1"># The following mean/std rescale the data from [0, 1] to [-1, 1]</span>
+        <span class="s2">&quot;image_mean&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">],</span>
+        <span class="s2">&quot;image_std&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">],</span>
+    <span class="p">}</span>
+    <span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="p">{</span><span class="o">**</span><span class="n">defaults</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">}</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">SSD</span><span class="p">(</span>
+        <span class="n">backbone</span><span class="p">,</span>
+        <span class="n">anchor_generator</span><span class="p">,</span>
+        <span class="n">size</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">,</span>
+        <span class="n">head</span><span class="o">=</span><span class="n">SSDLiteHead</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">num_anchors</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/efficientnet.html b/0.16/_modules/torchvision/models/efficientnet.html
new file mode 100644
index 00000000000..f025a3bfd6e
--- /dev/null
+++ b/0.16/_modules/torchvision/models/efficientnet.html
@@ -0,0 +1,1800 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.efficientnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.efficientnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.efficientnet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">copy</span>
+<span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">dataclasses</span> <span class="kn">import</span> <span class="n">dataclass</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops</span> <span class="kn">import</span> <span class="n">StochasticDepth</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span><span class="p">,</span> <span class="n">SqueezeExcitation</span>
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">InterpolationMode</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_make_divisible</span><span class="p">,</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;EfficientNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B0_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B1_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B2_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B3_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B4_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B5_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B6_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_B7_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_V2_S_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_V2_M_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;EfficientNet_V2_L_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b0&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b1&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b3&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b4&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b5&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b6&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_b7&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_v2_s&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_v2_m&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;efficientnet_v2_l&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="nd">@dataclass</span>
+<span class="k">class</span> <span class="nc">_MBConvConfig</span><span class="p">:</span>
+    <span class="n">expand_ratio</span><span class="p">:</span> <span class="nb">float</span>
+    <span class="n">kernel</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">input_channels</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">block</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">adjust_channels</span><span class="p">(</span><span class="n">channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">min_value</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_make_divisible</span><span class="p">(</span><span class="n">channels</span> <span class="o">*</span> <span class="n">width_mult</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="n">min_value</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">MBConvConfig</span><span class="p">(</span><span class="n">_MBConvConfig</span><span class="p">):</span>
+    <span class="c1"># Stores information listed at Table 1 of the EfficientNet paper &amp; Table 4 of the EfficientNetV2 paper</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">expand_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">kernel</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">input_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">width_mult</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">depth_mult</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">input_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">)</span>
+        <span class="n">out_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">)</span>
+        <span class="n">num_layers</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">adjust_depth</span><span class="p">(</span><span class="n">num_layers</span><span class="p">,</span> <span class="n">depth_mult</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">MBConv</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">expand_ratio</span><span class="p">,</span> <span class="n">kernel</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">input_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">num_layers</span><span class="p">,</span> <span class="n">block</span><span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">adjust_depth</span><span class="p">(</span><span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">depth_mult</span><span class="p">:</span> <span class="nb">float</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">num_layers</span> <span class="o">*</span> <span class="n">depth_mult</span><span class="p">))</span>
+
+
+<span class="k">class</span> <span class="nc">FusedMBConvConfig</span><span class="p">(</span><span class="n">_MBConvConfig</span><span class="p">):</span>
+    <span class="c1"># Stores information listed at Table 4 of the EfficientNetV2 paper</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">expand_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">kernel</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">input_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">FusedMBConv</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">expand_ratio</span><span class="p">,</span> <span class="n">kernel</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">input_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">num_layers</span><span class="p">,</span> <span class="n">block</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">MBConv</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">cnf</span><span class="p">:</span> <span class="n">MBConvConfig</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">se_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">SqueezeExcitation</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">&lt;=</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;illegal stride value&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">==</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">activation_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">SiLU</span>
+
+        <span class="c1"># expand</span>
+        <span class="n">expanded_channels</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">expand_ratio</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">expanded_channels</span> <span class="o">!=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                    <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span>
+                    <span class="n">expanded_channels</span><span class="p">,</span>
+                    <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="c1"># depthwise</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">expanded_channels</span><span class="p">,</span>
+                <span class="n">expanded_channels</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">kernel</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span>
+                <span class="n">groups</span><span class="o">=</span><span class="n">expanded_channels</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="c1"># squeeze and excitation</span>
+        <span class="n">squeeze_channels</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">//</span> <span class="mi">4</span><span class="p">)</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">se_layer</span><span class="p">(</span><span class="n">expanded_channels</span><span class="p">,</span> <span class="n">squeeze_channels</span><span class="p">,</span> <span class="n">activation</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">SiLU</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)))</span>
+
+        <span class="c1"># project</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">expanded_channels</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">block</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span> <span class="o">=</span> <span class="n">StochasticDepth</span><span class="p">(</span><span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="s2">&quot;row&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">block</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span><span class="p">:</span>
+            <span class="n">result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
+            <span class="n">result</span> <span class="o">+=</span> <span class="nb">input</span>
+        <span class="k">return</span> <span class="n">result</span>
+
+
+<span class="k">class</span> <span class="nc">FusedMBConv</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">cnf</span><span class="p">:</span> <span class="n">FusedMBConvConfig</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">&lt;=</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;illegal stride value&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">==</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">activation_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">SiLU</span>
+
+        <span class="n">expanded_channels</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">expand_ratio</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">expanded_channels</span> <span class="o">!=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">:</span>
+            <span class="c1"># fused expand</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                    <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span>
+                    <span class="n">expanded_channels</span><span class="p">,</span>
+                    <span class="n">kernel_size</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">kernel</span><span class="p">,</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+
+            <span class="c1"># project</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                    <span class="n">expanded_channels</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                    <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span>
+                    <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span>
+                    <span class="n">kernel_size</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">kernel</span><span class="p">,</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">block</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span> <span class="o">=</span> <span class="n">StochasticDepth</span><span class="p">(</span><span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="s2">&quot;row&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">block</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span><span class="p">:</span>
+            <span class="n">result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="n">result</span><span class="p">)</span>
+            <span class="n">result</span> <span class="o">+=</span> <span class="nb">input</span>
+        <span class="k">return</span> <span class="n">result</span>
+
+
+<span class="k">class</span> <span class="nc">EfficientNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inverted_residual_setting</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">MBConvConfig</span><span class="p">,</span> <span class="n">FusedMBConvConfig</span><span class="p">]],</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">last_channel</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        EfficientNet V1 and V2 main class</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            inverted_residual_setting (Sequence[Union[MBConvConfig, FusedMBConvConfig]]): Network structure</span>
+<span class="sd">            dropout (float): The droupout probability</span>
+<span class="sd">            stochastic_depth_prob (float): The stochastic depth probability</span>
+<span class="sd">            num_classes (int): Number of classes</span>
+<span class="sd">            norm_layer (Optional[Callable[..., nn.Module]]): Module specifying the normalization layer to use</span>
+<span class="sd">            last_channel (int): The number of channels on the penultimate layer</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">inverted_residual_setting</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The inverted_residual_setting should not be empty&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="p">(</span>
+            <span class="nb">isinstance</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span>
+            <span class="ow">and</span> <span class="nb">all</span><span class="p">([</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">s</span><span class="p">,</span> <span class="n">_MBConvConfig</span><span class="p">)</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">inverted_residual_setting</span><span class="p">])</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;The inverted_residual_setting should be List[MBConvConfig]&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="c1"># building first layer</span>
+        <span class="n">firstconv_output_channels</span> <span class="o">=</span> <span class="n">inverted_residual_setting</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">input_channels</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="mi">3</span><span class="p">,</span> <span class="n">firstconv_output_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">SiLU</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="c1"># building inverted residual blocks</span>
+        <span class="n">total_stage_blocks</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">num_layers</span> <span class="k">for</span> <span class="n">cnf</span> <span class="ow">in</span> <span class="n">inverted_residual_setting</span><span class="p">)</span>
+        <span class="n">stage_block_id</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">cnf</span> <span class="ow">in</span> <span class="n">inverted_residual_setting</span><span class="p">:</span>
+            <span class="n">stage</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">num_layers</span><span class="p">):</span>
+                <span class="c1"># copy to avoid modifications. shallow copy is enough</span>
+                <span class="n">block_cnf</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">copy</span><span class="p">(</span><span class="n">cnf</span><span class="p">)</span>
+
+                <span class="c1"># overwrite info if not the first conv in the stage</span>
+                <span class="k">if</span> <span class="n">stage</span><span class="p">:</span>
+                    <span class="n">block_cnf</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">=</span> <span class="n">block_cnf</span><span class="o">.</span><span class="n">out_channels</span>
+                    <span class="n">block_cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="mi">1</span>
+
+                <span class="c1"># adjust stochastic depth probability based on the depth of the stage block</span>
+                <span class="n">sd_prob</span> <span class="o">=</span> <span class="n">stochastic_depth_prob</span> <span class="o">*</span> <span class="nb">float</span><span class="p">(</span><span class="n">stage_block_id</span><span class="p">)</span> <span class="o">/</span> <span class="n">total_stage_blocks</span>
+
+                <span class="n">stage</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block_cnf</span><span class="o">.</span><span class="n">block</span><span class="p">(</span><span class="n">block_cnf</span><span class="p">,</span> <span class="n">sd_prob</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">))</span>
+                <span class="n">stage_block_id</span> <span class="o">+=</span> <span class="mi">1</span>
+
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">stage</span><span class="p">))</span>
+
+        <span class="c1"># building last several layers</span>
+        <span class="n">lastconv_input_channels</span> <span class="o">=</span> <span class="n">inverted_residual_setting</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">out_channels</span>
+        <span class="n">lastconv_output_channels</span> <span class="o">=</span> <span class="n">last_channel</span> <span class="k">if</span> <span class="n">last_channel</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="mi">4</span> <span class="o">*</span> <span class="n">lastconv_input_channels</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">lastconv_input_channels</span><span class="p">,</span>
+                <span class="n">lastconv_output_channels</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">SiLU</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">lastconv_output_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">GroupNorm</span><span class="p">)):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">ones_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">init_range</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">out_features</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="o">-</span><span class="n">init_range</span><span class="p">,</span> <span class="n">init_range</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward_impl</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_efficientnet</span><span class="p">(</span>
+    <span class="n">inverted_residual_setting</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">MBConvConfig</span><span class="p">,</span> <span class="n">FusedMBConvConfig</span><span class="p">]],</span>
+    <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">last_channel</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">EfficientNet</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">dropout</span><span class="p">,</span> <span class="n">last_channel</span><span class="o">=</span><span class="n">last_channel</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="k">def</span> <span class="nf">_efficientnet_conf</span><span class="p">(</span>
+    <span class="n">arch</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Sequence</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">MBConvConfig</span><span class="p">,</span> <span class="n">FusedMBConvConfig</span><span class="p">]],</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+    <span class="n">inverted_residual_setting</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">MBConvConfig</span><span class="p">,</span> <span class="n">FusedMBConvConfig</span><span class="p">]]</span>
+    <span class="k">if</span> <span class="n">arch</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;efficientnet_b&quot;</span><span class="p">):</span>
+        <span class="n">bneck_conf</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">MBConvConfig</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;width_mult&quot;</span><span class="p">),</span> <span class="n">depth_mult</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;depth_mult&quot;</span><span class="p">))</span>
+        <span class="n">inverted_residual_setting</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">last_channel</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="k">elif</span> <span class="n">arch</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;efficientnet_v2_s&quot;</span><span class="p">):</span>
+        <span class="n">inverted_residual_setting</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">15</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">last_channel</span> <span class="o">=</span> <span class="mi">1280</span>
+    <span class="k">elif</span> <span class="n">arch</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;efficientnet_v2_m&quot;</span><span class="p">):</span>
+        <span class="n">inverted_residual_setting</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">176</span><span class="p">,</span> <span class="mi">14</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">176</span><span class="p">,</span> <span class="mi">304</span><span class="p">,</span> <span class="mi">18</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">304</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">last_channel</span> <span class="o">=</span> <span class="mi">1280</span>
+    <span class="k">elif</span> <span class="n">arch</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;efficientnet_v2_l&quot;</span><span class="p">):</span>
+        <span class="n">inverted_residual_setting</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span>
+            <span class="n">FusedMBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">10</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">19</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">25</span><span class="p">),</span>
+            <span class="n">MBConvConfig</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">640</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">last_channel</span> <span class="o">=</span> <span class="mi">1280</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unsupported model type </span><span class="si">{</span><span class="n">arch</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span>
+
+
+<span class="n">_COMMON_META</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<span class="n">_COMMON_META_V1</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<span class="n">_COMMON_META_V2</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">33</span><span class="p">,</span> <span class="mi">33</span><span class="p">),</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="EfficientNet_B0_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b0.html#torchvision.models.EfficientNet_B0_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B0_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/rwightman/pytorch-image-models/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b0_rwightman-3dd342df.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5288548</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.692</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.532</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.386</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">20.451</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_B1_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b1.html#torchvision.models.EfficientNet_B1_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B1_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/rwightman/pytorch-image-models/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b1_rwightman-bac287d4.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">240</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">7794184</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.642</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.186</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.687</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">30.134</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b1-c27df63c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">240</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">255</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">7794184</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-lr-wd-crop-tuning&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">79.838</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.934</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.687</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">30.136</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_B2_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b2.html#torchvision.models.EfficientNet_B2_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B2_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/rwightman/pytorch-image-models/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b2_rwightman-bcdf34b7.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">288</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">288</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">9109994</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.608</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.310</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.088</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">35.174</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_B3_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b3.html#torchvision.models.EfficientNet_B3_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B3_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/rwightman/pytorch-image-models/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b3_rwightman-cf984f9c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">300</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">320</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">12233232</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.008</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.054</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.827</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">47.184</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_B4_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b4.html#torchvision.models.EfficientNet_B4_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B4_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/rwightman/pytorch-image-models/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b4_rwightman-7eb33cd5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">380</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">19341616</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.384</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.594</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.394</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">74.489</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_B5_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b5.html#torchvision.models.EfficientNet_B5_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B5_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/lukemelas/EfficientNet-PyTorch/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b5_lukemelas-b6417697.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">456</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">456</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">30389784</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.444</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.628</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">10.266</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">116.864</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_B6_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b6.html#torchvision.models.EfficientNet_B6_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B6_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/lukemelas/EfficientNet-PyTorch/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b6_lukemelas-c76e70fd.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">528</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">528</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">43040704</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">84.008</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.916</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">19.068</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">165.362</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_B7_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b7.html#torchvision.models.EfficientNet_B7_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_B7_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/lukemelas/EfficientNet-PyTorch/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_b7_lukemelas-dcc49843.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">600</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">600</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V1</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">66347960</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">84.122</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.908</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">37.746</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">254.675</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_V2_S_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_v2_s.html#torchvision.models.EfficientNet_V2_S_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_V2_S_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_v2_s-dd5fe13b.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V2</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">21458488</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">84.228</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.878</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">8.366</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">82.704</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_V2_M_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_v2_m.html#torchvision.models.EfficientNet_V2_M_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_V2_M_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_v2_m-dc08266a.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">480</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">480</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V2</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">54139356</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">85.112</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.156</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">24.582</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">208.01</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="EfficientNet_V2_L_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_v2_l.html#torchvision.models.EfficientNet_V2_L_Weights">[docs]</a><span class="k">class</span> <span class="nc">EfficientNet_V2_L_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="c1"># Weights ported from https://github.com/google/automl/tree/master/efficientnetv2</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/efficientnet_v2_l-59c71312.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">480</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">480</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META_V2</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">118515272</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">85.808</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.788</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">56.08</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">454.573</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b0"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b0.html#torchvision.models.efficientnet_b0">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b0</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B0_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B0 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B0_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B0_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B0_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B0_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b0&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">1.0</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">),</span> <span class="n">last_channel</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b1"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b1.html#torchvision.models.efficientnet_b1">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B1_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b1</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B1_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B1 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B1_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B1_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B1_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B1_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b1&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">1.1</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">),</span> <span class="n">last_channel</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b2"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b2.html#torchvision.models.efficientnet_b2">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B2_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B2_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B2 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B2_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B2_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b2&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">1.1</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">1.2</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">),</span> <span class="n">last_channel</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b3"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b3.html#torchvision.models.efficientnet_b3">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B3_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b3</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B3_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B3 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B3_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B3_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B3_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B3_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b3&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">1.2</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">1.4</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b4"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b4.html#torchvision.models.efficientnet_b4">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B4_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b4</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B4_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B4 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B4_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B4_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B4_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B4_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b4&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">1.4</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">1.8</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b5"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b5.html#torchvision.models.efficientnet_b5">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b5</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B5_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B5 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B5_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B5_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B5_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B5_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b5&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">1.6</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">2.2</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.01</span><span class="p">),</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b6"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b6.html#torchvision.models.efficientnet_b6">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B6_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b6</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B6_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B6 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B6_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B6_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B6_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B6_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b6&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">1.8</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">2.6</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.01</span><span class="p">),</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_b7"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_b7.html#torchvision.models.efficientnet_b7">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_B7_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_b7</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_B7_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;EfficientNet B7 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional</span>
+<span class="sd">    Neural Networks &lt;https://arxiv.org/abs/1905.11946&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_B7_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_B7_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_B7_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_B7_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_b7&quot;</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="mf">2.0</span><span class="p">,</span> <span class="n">depth_mult</span><span class="o">=</span><span class="mf">3.1</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.01</span><span class="p">),</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_v2_s"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_v2_s.html#torchvision.models.efficientnet_v2_s">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_V2_S_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_v2_s</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_V2_S_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs an EfficientNetV2-S architecture from</span>
+<span class="sd">    `EfficientNetV2: Smaller Models and Faster Training &lt;https://arxiv.org/abs/2104.00298&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_V2_S_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_V2_S_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_V2_S_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_V2_S_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_v2_s&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-03</span><span class="p">),</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_v2_m"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_v2_m.html#torchvision.models.efficientnet_v2_m">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_V2_M_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_v2_m</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_V2_M_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs an EfficientNetV2-M architecture from</span>
+<span class="sd">    `EfficientNetV2: Smaller Models and Faster Training &lt;https://arxiv.org/abs/2104.00298&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_V2_M_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_V2_M_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_V2_M_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_V2_M_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_v2_m&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-03</span><span class="p">),</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="efficientnet_v2_l"><a class="viewcode-back" href="../../../models/generated/torchvision.models.efficientnet_v2_l.html#torchvision.models.efficientnet_v2_l">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">EfficientNet_V2_L_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">efficientnet_v2_l</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">EfficientNet_V2_L_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EfficientNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs an EfficientNetV2-L architecture from</span>
+<span class="sd">    `EfficientNetV2: Smaller Models and Faster Training &lt;https://arxiv.org/abs/2104.00298&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.EfficientNet_V2_L_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.EfficientNet_V2_L_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.EfficientNet_V2_L_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">EfficientNet_V2_L_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_efficientnet_conf</span><span class="p">(</span><span class="s2">&quot;efficientnet_v2_l&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_efficientnet</span><span class="p">(</span>
+        <span class="n">inverted_residual_setting</span><span class="p">,</span>
+        <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;dropout&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">),</span>
+        <span class="n">last_channel</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-03</span><span class="p">),</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/feature_extraction.html b/0.16/_modules/torchvision/models/feature_extraction.html
new file mode 100644
index 00000000000..bb776a4450d
--- /dev/null
+++ b/0.16/_modules/torchvision/models/feature_extraction.html
@@ -0,0 +1,1232 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.feature_extraction &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.feature_extraction</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.feature_extraction</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">inspect</span>
+<span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">re</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">copy</span> <span class="kn">import</span> <span class="n">deepcopy</span>
+<span class="kn">from</span> <span class="nn">itertools</span> <span class="kn">import</span> <span class="n">chain</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torchvision</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">fx</span><span class="p">,</span> <span class="n">nn</span>
+<span class="kn">from</span> <span class="nn">torch.fx.graph_module</span> <span class="kn">import</span> <span class="n">_copy_attr</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;create_feature_extractor&quot;</span><span class="p">,</span> <span class="s2">&quot;get_graph_node_names&quot;</span><span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">LeafModuleAwareTracer</span><span class="p">(</span><span class="n">fx</span><span class="o">.</span><span class="n">Tracer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    An fx.Tracer that allows the user to specify a set of leaf modules, i.e.</span>
+<span class="sd">    modules that are not to be traced through. The resulting graph ends up</span>
+<span class="sd">    having single nodes referencing calls to the leaf modules&#39; forward methods.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">leaf_modules</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="k">if</span> <span class="s2">&quot;leaf_modules&quot;</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+            <span class="n">leaf_modules</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;leaf_modules&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">leaf_modules</span> <span class="o">=</span> <span class="n">leaf_modules</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">is_leaf_module</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">m</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">module_qualname</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">leaf_modules</span><span class="p">)):</span>
+            <span class="k">return</span> <span class="kc">True</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">is_leaf_module</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">module_qualname</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">NodePathTracer</span><span class="p">(</span><span class="n">LeafModuleAwareTracer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    NodePathTracer is an FX tracer that, for each operation, also records the</span>
+<span class="sd">    name of the Node from which the operation originated. A node name here is</span>
+<span class="sd">    a `.` separated path walking the hierarchy from top level module down to</span>
+<span class="sd">    leaf operation or leaf module. The name of the top level module is not</span>
+<span class="sd">    included as part of the node name. For example, if we trace a module whose</span>
+<span class="sd">    forward method applies a ReLU module, the name for that node will simply</span>
+<span class="sd">    be &#39;relu&#39;.</span>
+
+<span class="sd">    Some notes on the specifics:</span>
+<span class="sd">        - Nodes are recorded to `self.node_to_qualname` which is a dictionary</span>
+<span class="sd">          mapping a given Node object to its node name.</span>
+<span class="sd">        - Nodes are recorded in the order which they are executed during</span>
+<span class="sd">          tracing.</span>
+<span class="sd">        - When a duplicate node name is encountered, a suffix of the form</span>
+<span class="sd">          _{int} is added. The counter starts from 1.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="c1"># Track the qualified name of the Node being traced</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">current_module_qualname</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+        <span class="c1"># A map from FX Node to the qualified name\#</span>
+        <span class="c1"># NOTE: This is loosely like the &quot;qualified name&quot; mentioned in the</span>
+        <span class="c1"># torch.fx docs https://pytorch.org/docs/stable/fx.html but adapted</span>
+        <span class="c1"># for the purposes of the torchvision feature extractor</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">node_to_qualname</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">call_module</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">m</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">forward</span><span class="p">:</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Override of `fx.Tracer.call_module`</span>
+<span class="sd">        This override:</span>
+<span class="sd">        1) Stores away the qualified name of the caller for restoration later</span>
+<span class="sd">        2) Adds the qualified name of the caller to</span>
+<span class="sd">           `current_module_qualname` for retrieval by `create_proxy`</span>
+<span class="sd">        3) Once a leaf module is reached, calls `create_proxy`</span>
+<span class="sd">        4) Restores the caller&#39;s qualified name into current_module_qualname</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">old_qualname</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_module_qualname</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="n">module_qualname</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">path_of_module</span><span class="p">(</span><span class="n">m</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">current_module_qualname</span> <span class="o">=</span> <span class="n">module_qualname</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_leaf_module</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">module_qualname</span><span class="p">):</span>
+                <span class="n">out</span> <span class="o">=</span> <span class="n">forward</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+                <span class="k">return</span> <span class="n">out</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_proxy</span><span class="p">(</span><span class="s2">&quot;call_module&quot;</span><span class="p">,</span> <span class="n">module_qualname</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">finally</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">current_module_qualname</span> <span class="o">=</span> <span class="n">old_qualname</span>
+
+    <span class="k">def</span> <span class="nf">create_proxy</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">kind</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">target</span><span class="p">:</span> <span class="n">fx</span><span class="o">.</span><span class="n">node</span><span class="o">.</span><span class="n">Target</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">kwargs</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">type_expr</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="o">*</span><span class="n">_</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">fx</span><span class="o">.</span><span class="n">proxy</span><span class="o">.</span><span class="n">Proxy</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Override of `Tracer.create_proxy`. This override intercepts the recording</span>
+<span class="sd">        of every operation and stores away the current traced module&#39;s qualified</span>
+<span class="sd">        name in `node_to_qualname`</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">proxy</span> <span class="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">create_proxy</span><span class="p">(</span><span class="n">kind</span><span class="p">,</span> <span class="n">target</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">kwargs</span><span class="p">,</span> <span class="n">name</span><span class="p">,</span> <span class="n">type_expr</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="p">[</span><span class="n">proxy</span><span class="o">.</span><span class="n">node</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_node_qualname</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">current_module_qualname</span><span class="p">,</span> <span class="n">proxy</span><span class="o">.</span><span class="n">node</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">proxy</span>
+
+    <span class="k">def</span> <span class="nf">_get_node_qualname</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">module_qualname</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">node</span><span class="p">:</span> <span class="n">fx</span><span class="o">.</span><span class="n">node</span><span class="o">.</span><span class="n">Node</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">node_qualname</span> <span class="o">=</span> <span class="n">module_qualname</span>
+
+        <span class="k">if</span> <span class="n">node</span><span class="o">.</span><span class="n">op</span> <span class="o">!=</span> <span class="s2">&quot;call_module&quot;</span><span class="p">:</span>
+            <span class="c1"># In this case module_qualname from torch.fx doesn&#39;t go all the</span>
+            <span class="c1"># way to the leaf function/op, so we need to append it</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">node_qualname</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="c1"># Only append &#39;.&#39; if we are deeper than the top level module</span>
+                <span class="n">node_qualname</span> <span class="o">+=</span> <span class="s2">&quot;.&quot;</span>
+            <span class="n">node_qualname</span> <span class="o">+=</span> <span class="nb">str</span><span class="p">(</span><span class="n">node</span><span class="p">)</span>
+
+        <span class="c1"># Now we need to add an _{index} postfix on any repeated node names</span>
+        <span class="c1"># For modules we do this from scratch</span>
+        <span class="c1"># But for anything else, torch.fx already has a globally scoped</span>
+        <span class="c1"># _{index} postfix. But we want it locally (relative to direct parent)</span>
+        <span class="c1"># scoped. So first we need to undo the torch.fx postfix</span>
+        <span class="k">if</span> <span class="n">re</span><span class="o">.</span><span class="n">match</span><span class="p">(</span><span class="sa">r</span><span class="s2">&quot;.+_[0-9]+$&quot;</span><span class="p">,</span> <span class="n">node_qualname</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">node_qualname</span> <span class="o">=</span> <span class="n">node_qualname</span><span class="o">.</span><span class="n">rsplit</span><span class="p">(</span><span class="s2">&quot;_&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="c1"># ... and now we add on our own postfix</span>
+        <span class="k">for</span> <span class="n">existing_qualname</span> <span class="ow">in</span> <span class="nb">reversed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="o">.</span><span class="n">values</span><span class="p">()):</span>
+            <span class="c1"># Check to see if existing_qualname is of the form</span>
+            <span class="c1"># {node_qualname} or {node_qualname}_{int}</span>
+            <span class="k">if</span> <span class="n">re</span><span class="o">.</span><span class="n">match</span><span class="p">(</span><span class="sa">rf</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">node_qualname</span><span class="si">}</span><span class="s2">(_[0-9]+)?$&quot;</span><span class="p">,</span> <span class="n">existing_qualname</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">postfix</span> <span class="o">=</span> <span class="n">existing_qualname</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="n">node_qualname</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">postfix</span><span class="p">):</span>
+                    <span class="c1"># existing_qualname is of the form {node_qualname}_{int}</span>
+                    <span class="n">next_index</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">postfix</span><span class="p">[</span><span class="mi">1</span><span class="p">:])</span> <span class="o">+</span> <span class="mi">1</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="c1"># existing_qualname is of the form {node_qualname}</span>
+                    <span class="n">next_index</span> <span class="o">=</span> <span class="mi">1</span>
+                <span class="n">node_qualname</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;_</span><span class="si">{</span><span class="n">next_index</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="k">break</span>
+
+        <span class="k">return</span> <span class="n">node_qualname</span>
+
+
+<span class="k">def</span> <span class="nf">_is_subseq</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Check if y is a subsequence of x</span>
+<span class="sd">    https://stackoverflow.com/a/24017747/4391249</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">iter_x</span> <span class="o">=</span> <span class="nb">iter</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+    <span class="k">return</span> <span class="nb">all</span><span class="p">(</span><span class="nb">any</span><span class="p">(</span><span class="n">x_item</span> <span class="o">==</span> <span class="n">y_item</span> <span class="k">for</span> <span class="n">x_item</span> <span class="ow">in</span> <span class="n">iter_x</span><span class="p">)</span> <span class="k">for</span> <span class="n">y_item</span> <span class="ow">in</span> <span class="n">y</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_warn_graph_differences</span><span class="p">(</span><span class="n">train_tracer</span><span class="p">:</span> <span class="n">NodePathTracer</span><span class="p">,</span> <span class="n">eval_tracer</span><span class="p">:</span> <span class="n">NodePathTracer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Utility function for warning the user if there are differences between</span>
+<span class="sd">    the train graph nodes and the eval graph nodes.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">train_nodes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">train_tracer</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+    <span class="n">eval_nodes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">eval_tracer</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">train_nodes</span><span class="p">)</span> <span class="o">==</span> <span class="nb">len</span><span class="p">(</span><span class="n">eval_nodes</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">all</span><span class="p">(</span><span class="n">t</span> <span class="o">==</span> <span class="n">e</span> <span class="k">for</span> <span class="n">t</span><span class="p">,</span> <span class="n">e</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">train_nodes</span><span class="p">,</span> <span class="n">eval_nodes</span><span class="p">)):</span>
+        <span class="k">return</span>
+
+    <span class="n">suggestion_msg</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="s2">&quot;When choosing nodes for feature extraction, you may need to specify &quot;</span>
+        <span class="s2">&quot;output nodes for train and eval mode separately.&quot;</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">_is_subseq</span><span class="p">(</span><span class="n">train_nodes</span><span class="p">,</span> <span class="n">eval_nodes</span><span class="p">):</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="s2">&quot;NOTE: The nodes obtained by tracing the model in eval mode &quot;</span>
+            <span class="s2">&quot;are a subsequence of those obtained in train mode. &quot;</span>
+        <span class="p">)</span>
+    <span class="k">elif</span> <span class="n">_is_subseq</span><span class="p">(</span><span class="n">eval_nodes</span><span class="p">,</span> <span class="n">train_nodes</span><span class="p">):</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="s2">&quot;NOTE: The nodes obtained by tracing the model in train mode &quot;</span>
+            <span class="s2">&quot;are a subsequence of those obtained in eval mode. &quot;</span>
+        <span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="s2">&quot;The nodes obtained by tracing the model in train mode are different to those obtained in eval mode. &quot;</span>
+    <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="n">msg</span> <span class="o">+</span> <span class="n">suggestion_msg</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_get_leaf_modules_for_ops</span><span class="p">()</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">type</span><span class="p">]:</span>
+    <span class="n">members</span> <span class="o">=</span> <span class="n">inspect</span><span class="o">.</span><span class="n">getmembers</span><span class="p">(</span><span class="n">torchvision</span><span class="o">.</span><span class="n">ops</span><span class="p">)</span>
+    <span class="n">result</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">_</span><span class="p">,</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">members</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">inspect</span><span class="o">.</span><span class="n">isclass</span><span class="p">(</span><span class="n">obj</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+            <span class="n">result</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">obj</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">result</span>
+
+
+<span class="k">def</span> <span class="nf">_set_default_tracer_kwargs</span><span class="p">(</span><span class="n">original_tr_kwargs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+    <span class="n">default_autowrap_modules</span> <span class="o">=</span> <span class="p">(</span><span class="n">math</span><span class="p">,</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">ops</span><span class="p">)</span>
+    <span class="n">default_leaf_modules</span> <span class="o">=</span> <span class="n">_get_leaf_modules_for_ops</span><span class="p">()</span>
+    <span class="n">result_tracer_kwargs</span> <span class="o">=</span> <span class="p">{}</span> <span class="k">if</span> <span class="n">original_tr_kwargs</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">original_tr_kwargs</span>
+    <span class="n">result_tracer_kwargs</span><span class="p">[</span><span class="s2">&quot;autowrap_modules&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="nb">tuple</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">result_tracer_kwargs</span><span class="p">[</span><span class="s2">&quot;autowrap_modules&quot;</span><span class="p">]</span> <span class="o">+</span> <span class="n">default_autowrap_modules</span><span class="p">))</span>
+        <span class="k">if</span> <span class="s2">&quot;autowrap_modules&quot;</span> <span class="ow">in</span> <span class="n">result_tracer_kwargs</span>
+        <span class="k">else</span> <span class="n">default_autowrap_modules</span>
+    <span class="p">)</span>
+    <span class="n">result_tracer_kwargs</span><span class="p">[</span><span class="s2">&quot;leaf_modules&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="nb">list</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">result_tracer_kwargs</span><span class="p">[</span><span class="s2">&quot;leaf_modules&quot;</span><span class="p">]</span> <span class="o">+</span> <span class="n">default_leaf_modules</span><span class="p">))</span>
+        <span class="k">if</span> <span class="s2">&quot;leaf_modules&quot;</span> <span class="ow">in</span> <span class="n">result_tracer_kwargs</span>
+        <span class="k">else</span> <span class="n">default_leaf_modules</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">result_tracer_kwargs</span>
+
+
+<div class="viewcode-block" id="get_graph_node_names"><a class="viewcode-back" href="../../../generated/torchvision.models.feature_extraction.get_graph_node_names.html#torchvision.models.feature_extraction.get_graph_node_names">[docs]</a><span class="k">def</span> <span class="nf">get_graph_node_names</span><span class="p">(</span>
+    <span class="n">model</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span>
+    <span class="n">tracer_kwargs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">suppress_diff_warning</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Dev utility to return node names in order of execution. See note on node</span>
+<span class="sd">    names under :func:`create_feature_extractor`. Useful for seeing which node</span>
+<span class="sd">    names are available for feature extraction. There are two reasons that</span>
+<span class="sd">    node names can&#39;t easily be read directly from the code for a model:</span>
+
+<span class="sd">        1. Not all submodules are traced through. Modules from ``torch.nn`` all</span>
+<span class="sd">           fall within this category.</span>
+<span class="sd">        2. Nodes representing the repeated application of the same operation</span>
+<span class="sd">           or leaf module get a ``_{counter}`` postfix.</span>
+
+<span class="sd">    The model is traced twice: once in train mode, and once in eval mode. Both</span>
+<span class="sd">    sets of node names are returned.</span>
+
+<span class="sd">    For more details on the node naming conventions used here, please see the</span>
+<span class="sd">    :ref:`relevant subheading &lt;about-node-names&gt;` in the</span>
+<span class="sd">    `documentation &lt;https://pytorch.org/vision/stable/feature_extraction.html&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        model (nn.Module): model for which we&#39;d like to print node names</span>
+<span class="sd">        tracer_kwargs (dict, optional): a dictionary of keyword arguments for</span>
+<span class="sd">            ``NodePathTracer`` (they are eventually passed onto</span>
+<span class="sd">            `torch.fx.Tracer &lt;https://pytorch.org/docs/stable/fx.html#torch.fx.Tracer&gt;`_).</span>
+<span class="sd">            By default, it will be set to wrap and make leaf nodes all torchvision ops:</span>
+<span class="sd">            {&quot;autowrap_modules&quot;: (math, torchvision.ops,),&quot;leaf_modules&quot;: _get_leaf_modules_for_ops(),}</span>
+<span class="sd">            WARNING: In case the user provides tracer_kwargs, above default arguments will be appended to the user</span>
+<span class="sd">            provided dictionary.</span>
+
+<span class="sd">        suppress_diff_warning (bool, optional): whether to suppress a warning</span>
+<span class="sd">            when there are discrepancies between the train and eval version of</span>
+<span class="sd">            the graph. Defaults to False.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        tuple(list, list): a list of node names from tracing the model in</span>
+<span class="sd">        train mode, and another from tracing the model in eval mode.</span>
+
+<span class="sd">    Examples::</span>
+
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.resnet18()</span>
+<span class="sd">        &gt;&gt;&gt; train_nodes, eval_nodes = get_graph_node_names(model)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">tracer_kwargs</span> <span class="o">=</span> <span class="n">_set_default_tracer_kwargs</span><span class="p">(</span><span class="n">tracer_kwargs</span><span class="p">)</span>
+    <span class="n">is_training</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">training</span>
+    <span class="n">train_tracer</span> <span class="o">=</span> <span class="n">NodePathTracer</span><span class="p">(</span><span class="o">**</span><span class="n">tracer_kwargs</span><span class="p">)</span>
+    <span class="n">train_tracer</span><span class="o">.</span><span class="n">trace</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">())</span>
+    <span class="n">eval_tracer</span> <span class="o">=</span> <span class="n">NodePathTracer</span><span class="p">(</span><span class="o">**</span><span class="n">tracer_kwargs</span><span class="p">)</span>
+    <span class="n">eval_tracer</span><span class="o">.</span><span class="n">trace</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">())</span>
+    <span class="n">train_nodes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">train_tracer</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+    <span class="n">eval_nodes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">eval_tracer</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">suppress_diff_warning</span><span class="p">:</span>
+        <span class="n">_warn_graph_differences</span><span class="p">(</span><span class="n">train_tracer</span><span class="p">,</span> <span class="n">eval_tracer</span><span class="p">)</span>
+    <span class="c1"># Restore training state</span>
+    <span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">is_training</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">train_nodes</span><span class="p">,</span> <span class="n">eval_nodes</span></div>
+
+
+<span class="k">class</span> <span class="nc">DualGraphModule</span><span class="p">(</span><span class="n">fx</span><span class="o">.</span><span class="n">GraphModule</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A derivative of `fx.GraphModule`. Differs in the following ways:</span>
+<span class="sd">    - Requires a train and eval version of the underlying graph</span>
+<span class="sd">    - Copies submodules according to the nodes of both train and eval graphs.</span>
+<span class="sd">    - Calling train(mode) switches between train graph and eval graph.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">train_graph</span><span class="p">:</span> <span class="n">fx</span><span class="o">.</span><span class="n">Graph</span><span class="p">,</span> <span class="n">eval_graph</span><span class="p">:</span> <span class="n">fx</span><span class="o">.</span><span class="n">Graph</span><span class="p">,</span> <span class="n">class_name</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;GraphModule&quot;</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            root (nn.Module): module from which the copied module hierarchy is</span>
+<span class="sd">                built</span>
+<span class="sd">            train_graph (fx.Graph): the graph that should be used in train mode</span>
+<span class="sd">            eval_graph (fx.Graph): the graph that should be used in eval mode</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">fx</span><span class="o">.</span><span class="n">GraphModule</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">=</span> <span class="n">class_name</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">train_graph</span> <span class="o">=</span> <span class="n">train_graph</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">eval_graph</span> <span class="o">=</span> <span class="n">eval_graph</span>
+
+        <span class="c1"># Copy all get_attr and call_module ops (indicated by BOTH train and</span>
+        <span class="c1"># eval graphs)</span>
+        <span class="k">for</span> <span class="n">node</span> <span class="ow">in</span> <span class="n">chain</span><span class="p">(</span><span class="nb">iter</span><span class="p">(</span><span class="n">train_graph</span><span class="o">.</span><span class="n">nodes</span><span class="p">),</span> <span class="nb">iter</span><span class="p">(</span><span class="n">eval_graph</span><span class="o">.</span><span class="n">nodes</span><span class="p">)):</span>
+            <span class="k">if</span> <span class="n">node</span><span class="o">.</span><span class="n">op</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;get_attr&quot;</span><span class="p">,</span> <span class="s2">&quot;call_module&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">node</span><span class="o">.</span><span class="n">target</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+                    <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;node.target should be of type str instead of </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">node</span><span class="o">.</span><span class="n">target</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="n">_copy_attr</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="bp">self</span><span class="p">,</span> <span class="n">node</span><span class="o">.</span><span class="n">target</span><span class="p">)</span>
+
+        <span class="c1"># train mode by default</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">graph</span> <span class="o">=</span> <span class="n">train_graph</span>
+
+        <span class="c1"># (borrowed from fx.GraphModule):</span>
+        <span class="c1"># Store the Tracer class responsible for creating a Graph separately as part of the</span>
+        <span class="c1"># GraphModule state, except when the Tracer is defined in a local namespace.</span>
+        <span class="c1"># Locally defined Tracers are not pickleable. This is needed because torch.package will</span>
+        <span class="c1"># serialize a GraphModule without retaining the Graph, and needs to use the correct Tracer</span>
+        <span class="c1"># to re-create the Graph during deserialization.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">eval_graph</span><span class="o">.</span><span class="n">_tracer_cls</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_graph</span><span class="o">.</span><span class="n">_tracer_cls</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Train mode and eval mode should use the same tracer class. Instead got </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">eval_graph</span><span class="o">.</span><span class="n">_tracer_cls</span><span class="si">}</span><span class="s2"> for eval vs </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">train_graph</span><span class="o">.</span><span class="n">_tracer_cls</span><span class="si">}</span><span class="s2"> for train&quot;</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_tracer_cls</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">_tracer_cls</span> <span class="ow">and</span> <span class="s2">&quot;&lt;locals&gt;&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">_tracer_cls</span><span class="o">.</span><span class="vm">__qualname__</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_tracer_cls</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">_tracer_cls</span>
+
+    <span class="k">def</span> <span class="nf">train</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Swap out the graph depending on the selected training mode.</span>
+<span class="sd">        NOTE this should be safe when calling model.eval() because that just</span>
+<span class="sd">        calls this with mode == False.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># NOTE: Only set self.graph if the current graph is not the desired</span>
+        <span class="c1"># one. This saves us from recompiling the graph where not necessary.</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">graph</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_graph</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="n">mode</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">graph</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">eval_graph</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">mode</span><span class="o">=</span><span class="n">mode</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="create_feature_extractor"><a class="viewcode-back" href="../../../generated/torchvision.models.feature_extraction.create_feature_extractor.html#torchvision.models.feature_extraction.create_feature_extractor">[docs]</a><span class="k">def</span> <span class="nf">create_feature_extractor</span><span class="p">(</span>
+    <span class="n">model</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span>
+    <span class="n">return_nodes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">train_return_nodes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">eval_return_nodes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">tracer_kwargs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">suppress_diff_warning</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">fx</span><span class="o">.</span><span class="n">GraphModule</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Creates a new graph module that returns intermediate nodes from a given</span>
+<span class="sd">    model as dictionary with user specified keys as strings, and the requested</span>
+<span class="sd">    outputs as values. This is achieved by re-writing the computation graph of</span>
+<span class="sd">    the model via FX to return the desired nodes as outputs. All unused nodes</span>
+<span class="sd">    are removed, together with their corresponding parameters.</span>
+
+<span class="sd">    Desired output nodes must be specified as a ``.`` separated</span>
+<span class="sd">    path walking the module hierarchy from top level module down to leaf</span>
+<span class="sd">    operation or leaf module. For more details on the node naming conventions</span>
+<span class="sd">    used here, please see the :ref:`relevant subheading &lt;about-node-names&gt;`</span>
+<span class="sd">    in the `documentation &lt;https://pytorch.org/vision/stable/feature_extraction.html&gt;`_.</span>
+
+<span class="sd">    Not all models will be FX traceable, although with some massaging they can</span>
+<span class="sd">    be made to cooperate. Here&#39;s a (not exhaustive) list of tips:</span>
+
+<span class="sd">        - If you don&#39;t need to trace through a particular, problematic</span>
+<span class="sd">          sub-module, turn it into a &quot;leaf module&quot; by passing a list of</span>
+<span class="sd">          ``leaf_modules`` as one of the ``tracer_kwargs`` (see example below).</span>
+<span class="sd">          It will not be traced through, but rather, the resulting graph will</span>
+<span class="sd">          hold a reference to that module&#39;s forward method.</span>
+<span class="sd">        - Likewise, you may turn functions into leaf functions by passing a</span>
+<span class="sd">          list of ``autowrap_functions`` as one of the ``tracer_kwargs`` (see</span>
+<span class="sd">          example below).</span>
+<span class="sd">        - Some inbuilt Python functions can be problematic. For instance,</span>
+<span class="sd">          ``int`` will raise an error during tracing. You may wrap them in your</span>
+<span class="sd">          own function and then pass that in ``autowrap_functions`` as one of</span>
+<span class="sd">          the ``tracer_kwargs``.</span>
+
+<span class="sd">    For further information on FX see the</span>
+<span class="sd">    `torch.fx documentation &lt;https://pytorch.org/docs/stable/fx.html&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        model (nn.Module): model on which we will extract the features</span>
+<span class="sd">        return_nodes (list or dict, optional): either a ``List`` or a ``Dict``</span>
+<span class="sd">            containing the names (or partial names - see note above)</span>
+<span class="sd">            of the nodes for which the activations will be returned. If it is</span>
+<span class="sd">            a ``Dict``, the keys are the node names, and the values</span>
+<span class="sd">            are the user-specified keys for the graph module&#39;s returned</span>
+<span class="sd">            dictionary. If it is a ``List``, it is treated as a ``Dict`` mapping</span>
+<span class="sd">            node specification strings directly to output names. In the case</span>
+<span class="sd">            that ``train_return_nodes`` and ``eval_return_nodes`` are specified,</span>
+<span class="sd">            this should not be specified.</span>
+<span class="sd">        train_return_nodes (list or dict, optional): similar to</span>
+<span class="sd">            ``return_nodes``. This can be used if the return nodes</span>
+<span class="sd">            for train mode are different than those from eval mode.</span>
+<span class="sd">            If this is specified, ``eval_return_nodes`` must also be specified,</span>
+<span class="sd">            and ``return_nodes`` should not be specified.</span>
+<span class="sd">        eval_return_nodes (list or dict, optional): similar to</span>
+<span class="sd">            ``return_nodes``. This can be used if the return nodes</span>
+<span class="sd">            for train mode are different than those from eval mode.</span>
+<span class="sd">            If this is specified, ``train_return_nodes`` must also be specified,</span>
+<span class="sd">            and `return_nodes` should not be specified.</span>
+<span class="sd">        tracer_kwargs (dict, optional): a dictionary of keyword arguments for</span>
+<span class="sd">            ``NodePathTracer`` (which passes them onto it&#39;s parent class</span>
+<span class="sd">            `torch.fx.Tracer &lt;https://pytorch.org/docs/stable/fx.html#torch.fx.Tracer&gt;`_).</span>
+<span class="sd">            By default, it will be set to wrap and make leaf nodes all torchvision ops:</span>
+<span class="sd">            {&quot;autowrap_modules&quot;: (math, torchvision.ops,),&quot;leaf_modules&quot;: _get_leaf_modules_for_ops(),}</span>
+<span class="sd">            WARNING: In case the user provides tracer_kwargs, above default arguments will be appended to the user</span>
+<span class="sd">            provided dictionary.</span>
+<span class="sd">        suppress_diff_warning (bool, optional): whether to suppress a warning</span>
+<span class="sd">            when there are discrepancies between the train and eval version of</span>
+<span class="sd">            the graph. Defaults to False.</span>
+
+<span class="sd">    Examples::</span>
+
+<span class="sd">        &gt;&gt;&gt; # Feature extraction with resnet</span>
+<span class="sd">        &gt;&gt;&gt; model = torchvision.models.resnet18()</span>
+<span class="sd">        &gt;&gt;&gt; # extract layer1 and layer3, giving as names `feat1` and feat2`</span>
+<span class="sd">        &gt;&gt;&gt; model = create_feature_extractor(</span>
+<span class="sd">        &gt;&gt;&gt;     model, {&#39;layer1&#39;: &#39;feat1&#39;, &#39;layer3&#39;: &#39;feat2&#39;})</span>
+<span class="sd">        &gt;&gt;&gt; out = model(torch.rand(1, 3, 224, 224))</span>
+<span class="sd">        &gt;&gt;&gt; print([(k, v.shape) for k, v in out.items()])</span>
+<span class="sd">        &gt;&gt;&gt;     [(&#39;feat1&#39;, torch.Size([1, 64, 56, 56])),</span>
+<span class="sd">        &gt;&gt;&gt;      (&#39;feat2&#39;, torch.Size([1, 256, 14, 14]))]</span>
+
+<span class="sd">        &gt;&gt;&gt; # Specifying leaf modules and leaf functions</span>
+<span class="sd">        &gt;&gt;&gt; def leaf_function(x):</span>
+<span class="sd">        &gt;&gt;&gt;     # This would raise a TypeError if traced through</span>
+<span class="sd">        &gt;&gt;&gt;     return int(x)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; class LeafModule(torch.nn.Module):</span>
+<span class="sd">        &gt;&gt;&gt;     def forward(self, x):</span>
+<span class="sd">        &gt;&gt;&gt;         # This would raise a TypeError if traced through</span>
+<span class="sd">        &gt;&gt;&gt;         int(x.shape[0])</span>
+<span class="sd">        &gt;&gt;&gt;         return torch.nn.functional.relu(x + 4)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; class MyModule(torch.nn.Module):</span>
+<span class="sd">        &gt;&gt;&gt;     def __init__(self):</span>
+<span class="sd">        &gt;&gt;&gt;         super().__init__()</span>
+<span class="sd">        &gt;&gt;&gt;         self.conv = torch.nn.Conv2d(3, 1, 3)</span>
+<span class="sd">        &gt;&gt;&gt;         self.leaf_module = LeafModule()</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt;     def forward(self, x):</span>
+<span class="sd">        &gt;&gt;&gt;         leaf_function(x.shape[0])</span>
+<span class="sd">        &gt;&gt;&gt;         x = self.conv(x)</span>
+<span class="sd">        &gt;&gt;&gt;         return self.leaf_module(x)</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; model = create_feature_extractor(</span>
+<span class="sd">        &gt;&gt;&gt;     MyModule(), return_nodes=[&#39;leaf_module&#39;],</span>
+<span class="sd">        &gt;&gt;&gt;     tracer_kwargs={&#39;leaf_modules&#39;: [LeafModule],</span>
+<span class="sd">        &gt;&gt;&gt;                    &#39;autowrap_functions&#39;: [leaf_function]})</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">tracer_kwargs</span> <span class="o">=</span> <span class="n">_set_default_tracer_kwargs</span><span class="p">(</span><span class="n">tracer_kwargs</span><span class="p">)</span>
+    <span class="n">is_training</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">training</span>
+
+    <span class="k">if</span> <span class="nb">all</span><span class="p">(</span><span class="n">arg</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">for</span> <span class="n">arg</span> <span class="ow">in</span> <span class="p">[</span><span class="n">return_nodes</span><span class="p">,</span> <span class="n">train_return_nodes</span><span class="p">,</span> <span class="n">eval_return_nodes</span><span class="p">]):</span>
+
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="s2">&quot;Either `return_nodes` or `train_return_nodes` and `eval_return_nodes` together, should be specified&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="p">(</span><span class="n">train_return_nodes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">)</span> <span class="o">^</span> <span class="p">(</span><span class="n">eval_return_nodes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="s2">&quot;If any of `train_return_nodes` and `eval_return_nodes` are specified, then both should be specified&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="p">((</span><span class="n">return_nodes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">)</span> <span class="o">^</span> <span class="p">(</span><span class="n">train_return_nodes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;If `train_return_nodes` and `eval_return_nodes` are specified, then both should be specified&quot;</span><span class="p">)</span>
+
+    <span class="c1"># Put *_return_nodes into Dict[str, str] format</span>
+    <span class="k">def</span> <span class="nf">to_strdict</span><span class="p">(</span><span class="n">n</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">n</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="k">return</span> <span class="p">{</span><span class="nb">str</span><span class="p">(</span><span class="n">i</span><span class="p">):</span> <span class="nb">str</span><span class="p">(</span><span class="n">i</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">n</span><span class="p">}</span>
+        <span class="k">return</span> <span class="p">{</span><span class="nb">str</span><span class="p">(</span><span class="n">k</span><span class="p">):</span> <span class="nb">str</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">n</span><span class="o">.</span><span class="n">items</span><span class="p">()}</span>
+
+    <span class="k">if</span> <span class="n">train_return_nodes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">return_nodes</span> <span class="o">=</span> <span class="n">to_strdict</span><span class="p">(</span><span class="n">return_nodes</span><span class="p">)</span>
+        <span class="n">train_return_nodes</span> <span class="o">=</span> <span class="n">deepcopy</span><span class="p">(</span><span class="n">return_nodes</span><span class="p">)</span>
+        <span class="n">eval_return_nodes</span> <span class="o">=</span> <span class="n">deepcopy</span><span class="p">(</span><span class="n">return_nodes</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">train_return_nodes</span> <span class="o">=</span> <span class="n">to_strdict</span><span class="p">(</span><span class="n">train_return_nodes</span><span class="p">)</span>
+        <span class="n">eval_return_nodes</span> <span class="o">=</span> <span class="n">to_strdict</span><span class="p">(</span><span class="n">eval_return_nodes</span><span class="p">)</span>
+
+    <span class="c1"># Repeat the tracing and graph rewriting for train and eval mode</span>
+    <span class="n">tracers</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="n">graphs</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="n">mode_return_nodes</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="n">train_return_nodes</span><span class="p">,</span> <span class="s2">&quot;eval&quot;</span><span class="p">:</span> <span class="n">eval_return_nodes</span><span class="p">}</span>
+    <span class="k">for</span> <span class="n">mode</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;eval&quot;</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;eval&quot;</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+        <span class="c1"># Instantiate our NodePathTracer and use that to trace the model</span>
+        <span class="n">tracer</span> <span class="o">=</span> <span class="n">NodePathTracer</span><span class="p">(</span><span class="o">**</span><span class="n">tracer_kwargs</span><span class="p">)</span>
+        <span class="n">graph</span> <span class="o">=</span> <span class="n">tracer</span><span class="o">.</span><span class="n">trace</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+
+        <span class="n">name</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">)</span> <span class="k">else</span> <span class="n">model</span><span class="o">.</span><span class="vm">__name__</span>
+        <span class="n">graph_module</span> <span class="o">=</span> <span class="n">fx</span><span class="o">.</span><span class="n">GraphModule</span><span class="p">(</span><span class="n">tracer</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="n">graph</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
+
+        <span class="n">available_nodes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">tracer</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+        <span class="c1"># FIXME We don&#39;t know if we should expect this to happen</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">available_nodes</span><span class="p">))</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">available_nodes</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;There are duplicate nodes! Please raise an issue https://github.com/pytorch/vision/issues&quot;</span>
+            <span class="p">)</span>
+        <span class="c1"># Check that all outputs in return_nodes are present in the model</span>
+        <span class="k">for</span> <span class="n">query</span> <span class="ow">in</span> <span class="n">mode_return_nodes</span><span class="p">[</span><span class="n">mode</span><span class="p">]</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+            <span class="c1"># To check if a query is available we need to check that at least</span>
+            <span class="c1"># one of the available names starts with it up to a .</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="nb">any</span><span class="p">([</span><span class="n">re</span><span class="o">.</span><span class="n">match</span><span class="p">(</span><span class="sa">rf</span><span class="s2">&quot;^</span><span class="si">{</span><span class="n">query</span><span class="si">}</span><span class="s2">(\.|$)&quot;</span><span class="p">,</span> <span class="n">n</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">for</span> <span class="n">n</span> <span class="ow">in</span> <span class="n">available_nodes</span><span class="p">]):</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;node: &#39;</span><span class="si">{</span><span class="n">query</span><span class="si">}</span><span class="s2">&#39; is not present in model. Hint: use &quot;</span>
+                    <span class="s2">&quot;`get_graph_node_names` to make sure the &quot;</span>
+                    <span class="s2">&quot;`return_nodes` you specified are present. It may even &quot;</span>
+                    <span class="s2">&quot;be that you need to specify `train_return_nodes` and &quot;</span>
+                    <span class="s2">&quot;`eval_return_nodes` separately.&quot;</span>
+                <span class="p">)</span>
+
+        <span class="c1"># Remove existing output nodes (train mode)</span>
+        <span class="n">orig_output_nodes</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">n</span> <span class="ow">in</span> <span class="nb">reversed</span><span class="p">(</span><span class="n">graph_module</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">nodes</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">n</span><span class="o">.</span><span class="n">op</span> <span class="o">==</span> <span class="s2">&quot;output&quot;</span><span class="p">:</span>
+                <span class="n">orig_output_nodes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">n</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">orig_output_nodes</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;No output nodes found in graph_module.graph.nodes&quot;</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">n</span> <span class="ow">in</span> <span class="n">orig_output_nodes</span><span class="p">:</span>
+            <span class="n">graph_module</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">erase_node</span><span class="p">(</span><span class="n">n</span><span class="p">)</span>
+
+        <span class="c1"># Find nodes corresponding to return_nodes and make them into output_nodes</span>
+        <span class="n">nodes</span> <span class="o">=</span> <span class="p">[</span><span class="n">n</span> <span class="k">for</span> <span class="n">n</span> <span class="ow">in</span> <span class="n">graph_module</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">nodes</span><span class="p">]</span>
+        <span class="n">output_nodes</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">n</span> <span class="ow">in</span> <span class="nb">reversed</span><span class="p">(</span><span class="n">nodes</span><span class="p">):</span>
+            <span class="n">module_qualname</span> <span class="o">=</span> <span class="n">tracer</span><span class="o">.</span><span class="n">node_to_qualname</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">n</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">module_qualname</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="c1"># NOTE - Know cases where this happens:</span>
+                <span class="c1"># - Node representing creation of a tensor constant - probably</span>
+                <span class="c1">#   not interesting as a return node</span>
+                <span class="c1"># - When packing outputs into a named tuple like in InceptionV3</span>
+                <span class="k">continue</span>
+            <span class="k">for</span> <span class="n">query</span> <span class="ow">in</span> <span class="n">mode_return_nodes</span><span class="p">[</span><span class="n">mode</span><span class="p">]:</span>
+                <span class="n">depth</span> <span class="o">=</span> <span class="n">query</span><span class="o">.</span><span class="n">count</span><span class="p">(</span><span class="s2">&quot;.&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="s2">&quot;.&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">module_qualname</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;.&quot;</span><span class="p">)[:</span> <span class="n">depth</span> <span class="o">+</span> <span class="mi">1</span><span class="p">])</span> <span class="o">==</span> <span class="n">query</span><span class="p">:</span>
+                    <span class="n">output_nodes</span><span class="p">[</span><span class="n">mode_return_nodes</span><span class="p">[</span><span class="n">mode</span><span class="p">][</span><span class="n">query</span><span class="p">]]</span> <span class="o">=</span> <span class="n">n</span>
+                    <span class="n">mode_return_nodes</span><span class="p">[</span><span class="n">mode</span><span class="p">]</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">query</span><span class="p">)</span>
+                    <span class="k">break</span>
+        <span class="n">output_nodes</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">(</span><span class="nb">reversed</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="n">output_nodes</span><span class="o">.</span><span class="n">items</span><span class="p">())))</span>
+
+        <span class="c1"># And add them in the end of the graph</span>
+        <span class="k">with</span> <span class="n">graph_module</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">inserting_after</span><span class="p">(</span><span class="n">nodes</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]):</span>
+            <span class="n">graph_module</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">output</span><span class="p">(</span><span class="n">output_nodes</span><span class="p">)</span>
+
+        <span class="c1"># Remove unused modules / parameters</span>
+        <span class="n">graph_module</span><span class="o">.</span><span class="n">graph</span><span class="o">.</span><span class="n">eliminate_dead_code</span><span class="p">()</span>
+        <span class="n">graph_module</span><span class="o">.</span><span class="n">recompile</span><span class="p">()</span>
+
+        <span class="c1"># Keep track of the tracer and graph, so we can choose the main one</span>
+        <span class="n">tracers</span><span class="p">[</span><span class="n">mode</span><span class="p">]</span> <span class="o">=</span> <span class="n">tracer</span>
+        <span class="n">graphs</span><span class="p">[</span><span class="n">mode</span><span class="p">]</span> <span class="o">=</span> <span class="n">graph</span>
+
+    <span class="c1"># Warn user if there are any discrepancies between the graphs of the</span>
+    <span class="c1"># train and eval modes</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">suppress_diff_warning</span><span class="p">:</span>
+        <span class="n">_warn_graph_differences</span><span class="p">(</span><span class="n">tracers</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">],</span> <span class="n">tracers</span><span class="p">[</span><span class="s2">&quot;eval&quot;</span><span class="p">])</span>
+
+    <span class="c1"># Build the final graph module</span>
+    <span class="n">graph_module</span> <span class="o">=</span> <span class="n">DualGraphModule</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">graphs</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">],</span> <span class="n">graphs</span><span class="p">[</span><span class="s2">&quot;eval&quot;</span><span class="p">],</span> <span class="n">class_name</span><span class="o">=</span><span class="n">name</span><span class="p">)</span>
+
+    <span class="c1"># Restore original training mode</span>
+    <span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">is_training</span><span class="p">)</span>
+    <span class="n">graph_module</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">is_training</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">graph_module</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/googlenet.html b/0.16/_modules/torchvision/models/googlenet.html
new file mode 100644
index 00000000000..b5e30188e41
--- /dev/null
+++ b/0.16/_modules/torchvision/models/googlenet.html
@@ -0,0 +1,1014 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.googlenet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.googlenet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.googlenet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">namedtuple</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;GoogLeNet&quot;</span><span class="p">,</span> <span class="s2">&quot;GoogLeNetOutputs&quot;</span><span class="p">,</span> <span class="s2">&quot;_GoogLeNetOutputs&quot;</span><span class="p">,</span> <span class="s2">&quot;GoogLeNet_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;googlenet&quot;</span><span class="p">]</span>
+
+
+<span class="n">GoogLeNetOutputs</span> <span class="o">=</span> <span class="n">namedtuple</span><span class="p">(</span><span class="s2">&quot;GoogLeNetOutputs&quot;</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;logits&quot;</span><span class="p">,</span> <span class="s2">&quot;aux_logits2&quot;</span><span class="p">,</span> <span class="s2">&quot;aux_logits1&quot;</span><span class="p">])</span>
+<span class="n">GoogLeNetOutputs</span><span class="o">.</span><span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;logits&quot;</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="s2">&quot;aux_logits2&quot;</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="s2">&quot;aux_logits1&quot;</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]}</span>
+
+<span class="c1"># Script annotations failed with _GoogleNetOutputs = namedtuple ...</span>
+<span class="c1"># _GoogLeNetOutputs set here for backwards compat</span>
+<span class="n">_GoogLeNetOutputs</span> <span class="o">=</span> <span class="n">GoogLeNetOutputs</span>
+
+
+<span class="k">class</span> <span class="nc">GoogLeNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="n">__constants__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="s2">&quot;transform_input&quot;</span><span class="p">]</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">aux_logits</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform_input</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">init_weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">blocks</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">dropout_aux</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.7</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">blocks</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">blocks</span> <span class="o">=</span> <span class="p">[</span><span class="n">BasicConv2d</span><span class="p">,</span> <span class="n">Inception</span><span class="p">,</span> <span class="n">InceptionAux</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">init_weights</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;The default weight initialization of GoogleNet will be changed in future releases of &quot;</span>
+                <span class="s2">&quot;torchvision. If you wish to keep the old behavior (which leads to long initialization times&quot;</span>
+                <span class="s2">&quot; due to scipy/scipy#11299), please set init_weights=True.&quot;</span><span class="p">,</span>
+                <span class="ne">FutureWarning</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">init_weights</span> <span class="o">=</span> <span class="kc">True</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">blocks</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;blocks length should be 3 instead of </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">blocks</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">conv_block</span> <span class="o">=</span> <span class="n">blocks</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">inception_block</span> <span class="o">=</span> <span class="n">blocks</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">inception_aux_block</span> <span class="o">=</span> <span class="n">blocks</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span> <span class="o">=</span> <span class="n">aux_logits</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform_input</span> <span class="o">=</span> <span class="n">transform_input</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception3a</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception3b</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception4a</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">480</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">208</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception4b</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception4c</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception4d</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="mi">144</span><span class="p">,</span> <span class="mi">288</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception4e</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">528</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool4</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception5a</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">832</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inception5b</span> <span class="o">=</span> <span class="n">inception_block</span><span class="p">(</span><span class="mi">832</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">aux_logits</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">aux1</span> <span class="o">=</span> <span class="n">inception_aux_block</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">dropout</span><span class="o">=</span><span class="n">dropout_aux</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">aux2</span> <span class="o">=</span> <span class="n">inception_aux_block</span><span class="p">(</span><span class="mi">528</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">dropout</span><span class="o">=</span><span class="n">dropout_aux</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">aux1</span> <span class="o">=</span> <span class="kc">None</span>  <span class="c1"># type: ignore[assignment]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">aux2</span> <span class="o">=</span> <span class="kc">None</span>  <span class="c1"># type: ignore[assignment]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">init_weights</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">)</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">,</span> <span class="n">a</span><span class="o">=-</span><span class="mi">2</span><span class="p">,</span> <span class="n">b</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">):</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform_input</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform_input</span><span class="p">:</span>
+            <span class="n">x_ch0</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mf">0.229</span> <span class="o">/</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span><span class="mf">0.485</span> <span class="o">-</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">/</span> <span class="mf">0.5</span>
+            <span class="n">x_ch1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mf">0.224</span> <span class="o">/</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span><span class="mf">0.456</span> <span class="o">-</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">/</span> <span class="mf">0.5</span>
+            <span class="n">x_ch2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mf">0.225</span> <span class="o">/</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span><span class="mf">0.406</span> <span class="o">-</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">/</span> <span class="mf">0.5</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">x_ch0</span><span class="p">,</span> <span class="n">x_ch1</span><span class="p">,</span> <span class="n">x_ch2</span><span class="p">),</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]]:</span>
+        <span class="c1"># N x 3 x 224 x 224</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 64 x 112 x 112</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 64 x 56 x 56</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 64 x 56 x 56</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 192 x 56 x 56</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="c1"># N x 192 x 28 x 28</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception3a</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 256 x 28 x 28</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception3b</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 480 x 28 x 28</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 480 x 14 x 14</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception4a</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 512 x 14 x 14</span>
+        <span class="n">aux1</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux1</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+                <span class="n">aux1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception4b</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 512 x 14 x 14</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception4c</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 512 x 14 x 14</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception4d</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 528 x 14 x 14</span>
+        <span class="n">aux2</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux2</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+                <span class="n">aux2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception4e</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 832 x 14 x 14</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool4</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 832 x 7 x 7</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception5a</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 832 x 7 x 7</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inception5b</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1024 x 7 x 7</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1024 x 1 x 1</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># N x 1024</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1000 (num_classes)</span>
+        <span class="k">return</span> <span class="n">x</span><span class="p">,</span> <span class="n">aux2</span><span class="p">,</span> <span class="n">aux1</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">eager_outputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">aux2</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">aux1</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">GoogLeNetOutputs</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">_GoogLeNetOutputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux2</span><span class="p">,</span> <span class="n">aux1</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">x</span>  <span class="c1"># type: ignore[return-value]</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">GoogLeNetOutputs</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_transform_input</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span><span class="p">,</span> <span class="n">aux1</span><span class="p">,</span> <span class="n">aux2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">aux_defined</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">aux_defined</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scripted GoogleNet always returns GoogleNetOutputs Tuple&quot;</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">GoogLeNetOutputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux2</span><span class="p">,</span> <span class="n">aux1</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">eager_outputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux2</span><span class="p">,</span> <span class="n">aux1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">Inception</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">ch1x1</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">ch3x3red</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">ch3x3</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">ch5x5red</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">ch5x5</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">pool_proj</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">ch1x1</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">ch3x3red</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">ch3x3red</span><span class="p">,</span> <span class="n">ch3x3</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">ch5x5red</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+            <span class="c1"># Here, kernel_size=3 instead of kernel_size=5 is a known bug.</span>
+            <span class="c1"># Please see https://github.com/pytorch/vision/issues/906 for details.</span>
+            <span class="n">conv_block</span><span class="p">(</span><span class="n">ch5x5red</span><span class="p">,</span> <span class="n">ch5x5</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch4</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">pool_proj</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">branch1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch4</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch4</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">branch1</span><span class="p">,</span> <span class="n">branch2</span><span class="p">,</span> <span class="n">branch3</span><span class="p">,</span> <span class="n">branch4</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">outputs</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InceptionAux</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.7</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">2048</span><span class="p">,</span> <span class="mi">1024</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># aux1: N x 512 x 14 x 14, aux2: N x 528 x 14 x 14</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adaptive_avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">))</span>
+        <span class="c1"># aux1: N x 512 x 4 x 4, aux2: N x 528 x 4 x 4</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 128 x 4 x 4</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># N x 2048</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fc1</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># N x 1024</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1024</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1000 (num_classes)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">BasicConv2d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="GoogLeNet_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.quantization.googlenet.html#torchvision.models.GoogLeNet_Weights">[docs]</a><span class="k">class</span> <span class="nc">GoogLeNet_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/googlenet-1378be20.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">6624904</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">15</span><span class="p">,</span> <span class="mi">15</span><span class="p">),</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#googlenet&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">69.778</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">89.530</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.498</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">49.731</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="googlenet"><a class="viewcode-back" href="../../../models/generated/torchvision.models.googlenet.html#torchvision.models.googlenet">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">GoogLeNet_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">googlenet</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">GoogLeNet_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">GoogLeNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;GoogLeNet (Inception v1) model architecture from</span>
+<span class="sd">    `Going Deeper with Convolutions &lt;http://arxiv.org/abs/1409.4842&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.GoogLeNet_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.GoogLeNet_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.GoogLeNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/googlenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.GoogLeNet_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">GoogLeNet_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">original_aux_logits</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="s2">&quot;transform_input&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;transform_input&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;init_weights&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">GoogLeNet</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">original_aux_logits</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux_logits</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux1</span> <span class="o">=</span> <span class="kc">None</span>  <span class="c1"># type: ignore[assignment]</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux2</span> <span class="o">=</span> <span class="kc">None</span>  <span class="c1"># type: ignore[assignment]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;auxiliary heads in the pretrained googlenet model are NOT pretrained, so make sure to train them&quot;</span>
+            <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/inception.html b/0.16/_modules/torchvision/models/inception.html
new file mode 100644
index 00000000000..6f1dbb5efc3
--- /dev/null
+++ b/0.16/_modules/torchvision/models/inception.html
@@ -0,0 +1,1147 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.inception &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.inception</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.inception</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">namedtuple</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Inception3&quot;</span><span class="p">,</span> <span class="s2">&quot;InceptionOutputs&quot;</span><span class="p">,</span> <span class="s2">&quot;_InceptionOutputs&quot;</span><span class="p">,</span> <span class="s2">&quot;Inception_V3_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;inception_v3&quot;</span><span class="p">]</span>
+
+
+<span class="n">InceptionOutputs</span> <span class="o">=</span> <span class="n">namedtuple</span><span class="p">(</span><span class="s2">&quot;InceptionOutputs&quot;</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;logits&quot;</span><span class="p">,</span> <span class="s2">&quot;aux_logits&quot;</span><span class="p">])</span>
+<span class="n">InceptionOutputs</span><span class="o">.</span><span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;logits&quot;</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="s2">&quot;aux_logits&quot;</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]}</span>
+
+<span class="c1"># Script annotations failed with _GoogleNetOutputs = namedtuple ...</span>
+<span class="c1"># _InceptionOutputs set here for backwards compat</span>
+<span class="n">_InceptionOutputs</span> <span class="o">=</span> <span class="n">InceptionOutputs</span>
+
+
+<span class="k">class</span> <span class="nc">Inception3</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">aux_logits</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">transform_input</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">inception_blocks</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">init_weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">inception_blocks</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inception_blocks</span> <span class="o">=</span> <span class="p">[</span><span class="n">BasicConv2d</span><span class="p">,</span> <span class="n">InceptionA</span><span class="p">,</span> <span class="n">InceptionB</span><span class="p">,</span> <span class="n">InceptionC</span><span class="p">,</span> <span class="n">InceptionD</span><span class="p">,</span> <span class="n">InceptionE</span><span class="p">,</span> <span class="n">InceptionAux</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">init_weights</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;The default weight initialization of inception_v3 will be changed in future releases of &quot;</span>
+                <span class="s2">&quot;torchvision. If you wish to keep the old behavior (which leads to long initialization times&quot;</span>
+                <span class="s2">&quot; due to scipy/scipy#11299), please set init_weights=True.&quot;</span><span class="p">,</span>
+                <span class="ne">FutureWarning</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">init_weights</span> <span class="o">=</span> <span class="kc">True</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inception_blocks</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">7</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;length of inception_blocks should be 7 instead of </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">inception_blocks</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">conv_block</span> <span class="o">=</span> <span class="n">inception_blocks</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">inception_a</span> <span class="o">=</span> <span class="n">inception_blocks</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">inception_b</span> <span class="o">=</span> <span class="n">inception_blocks</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+        <span class="n">inception_c</span> <span class="o">=</span> <span class="n">inception_blocks</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span>
+        <span class="n">inception_d</span> <span class="o">=</span> <span class="n">inception_blocks</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span>
+        <span class="n">inception_e</span> <span class="o">=</span> <span class="n">inception_blocks</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span>
+        <span class="n">inception_aux</span> <span class="o">=</span> <span class="n">inception_blocks</span><span class="p">[</span><span class="mi">6</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span> <span class="o">=</span> <span class="n">aux_logits</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform_input</span> <span class="o">=</span> <span class="n">transform_input</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_1a_3x3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_2a_3x3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_2b_3x3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_3b_1x1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_4a_3x3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">80</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_5b</span> <span class="o">=</span> <span class="n">inception_a</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="n">pool_features</span><span class="o">=</span><span class="mi">32</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_5c</span> <span class="o">=</span> <span class="n">inception_a</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="n">pool_features</span><span class="o">=</span><span class="mi">64</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_5d</span> <span class="o">=</span> <span class="n">inception_a</span><span class="p">(</span><span class="mi">288</span><span class="p">,</span> <span class="n">pool_features</span><span class="o">=</span><span class="mi">64</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6a</span> <span class="o">=</span> <span class="n">inception_b</span><span class="p">(</span><span class="mi">288</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6b</span> <span class="o">=</span> <span class="n">inception_c</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="n">channels_7x7</span><span class="o">=</span><span class="mi">128</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6c</span> <span class="o">=</span> <span class="n">inception_c</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="n">channels_7x7</span><span class="o">=</span><span class="mi">160</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6d</span> <span class="o">=</span> <span class="n">inception_c</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="n">channels_7x7</span><span class="o">=</span><span class="mi">160</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6e</span> <span class="o">=</span> <span class="n">inception_c</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="n">channels_7x7</span><span class="o">=</span><span class="mi">192</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">AuxLogits</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">aux_logits</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">AuxLogits</span> <span class="o">=</span> <span class="n">inception_aux</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_7a</span> <span class="o">=</span> <span class="n">inception_d</span><span class="p">(</span><span class="mi">768</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_7b</span> <span class="o">=</span> <span class="n">inception_e</span><span class="p">(</span><span class="mi">1280</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_7c</span> <span class="o">=</span> <span class="n">inception_e</span><span class="p">(</span><span class="mi">2048</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">2048</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">init_weights</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">)</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                    <span class="n">stddev</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">stddev</span><span class="p">)</span> <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="s2">&quot;stddev&quot;</span><span class="p">)</span> <span class="k">else</span> <span class="mf">0.1</span>  <span class="c1"># type: ignore</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">stddev</span><span class="p">,</span> <span class="n">a</span><span class="o">=-</span><span class="mi">2</span><span class="p">,</span> <span class="n">b</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">):</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform_input</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transform_input</span><span class="p">:</span>
+            <span class="n">x_ch0</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mf">0.229</span> <span class="o">/</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span><span class="mf">0.485</span> <span class="o">-</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">/</span> <span class="mf">0.5</span>
+            <span class="n">x_ch1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mf">0.224</span> <span class="o">/</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span><span class="mf">0.456</span> <span class="o">-</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">/</span> <span class="mf">0.5</span>
+            <span class="n">x_ch2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mf">0.225</span> <span class="o">/</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span><span class="mf">0.406</span> <span class="o">-</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">/</span> <span class="mf">0.5</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">x_ch0</span><span class="p">,</span> <span class="n">x_ch1</span><span class="p">,</span> <span class="n">x_ch2</span><span class="p">),</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]]:</span>
+        <span class="c1"># N x 3 x 299 x 299</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_1a_3x3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 32 x 149 x 149</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_2a_3x3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 32 x 147 x 147</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_2b_3x3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 64 x 147 x 147</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 64 x 73 x 73</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_3b_1x1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 80 x 73 x 73</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Conv2d_4a_3x3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 192 x 71 x 71</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 192 x 35 x 35</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_5b</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 256 x 35 x 35</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_5c</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 288 x 35 x 35</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_5d</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 288 x 35 x 35</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6a</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 17 x 17</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6b</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 17 x 17</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6c</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 17 x 17</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6d</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 17 x 17</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_6e</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 17 x 17</span>
+        <span class="n">aux</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">AuxLogits</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+                <span class="n">aux</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">AuxLogits</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 17 x 17</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_7a</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1280 x 8 x 8</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_7b</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 2048 x 8 x 8</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">Mixed_7c</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 2048 x 8 x 8</span>
+        <span class="c1"># Adaptive average pooling</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 2048 x 1 x 1</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 2048 x 1 x 1</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># N x 2048</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1000 (num_classes)</span>
+        <span class="k">return</span> <span class="n">x</span><span class="p">,</span> <span class="n">aux</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">eager_outputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">aux</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">InceptionOutputs</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">InceptionOutputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">x</span>  <span class="c1"># type: ignore[return-value]</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">InceptionOutputs</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_transform_input</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span><span class="p">,</span> <span class="n">aux</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">aux_defined</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">aux_defined</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scripted Inception3 always returns Inception3 Tuple&quot;</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">InceptionOutputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">eager_outputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InceptionA</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">pool_features</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch1x1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch5x5_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch5x5_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">48</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">96</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch_pool</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">pool_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">branch1x1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch1x1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">branch5x5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch5x5_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch5x5</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch5x5_2</span><span class="p">(</span><span class="n">branch5x5</span><span class="p">)</span>
+
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_2</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">)</span>
+
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch_pool</span><span class="p">(</span><span class="n">branch_pool</span><span class="p">)</span>
+
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">branch1x1</span><span class="p">,</span> <span class="n">branch5x5</span><span class="p">,</span> <span class="n">branch3x3dbl</span><span class="p">,</span> <span class="n">branch_pool</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">outputs</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InceptionB</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">96</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_2</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">)</span>
+
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">max_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">branch3x3</span><span class="p">,</span> <span class="n">branch3x3dbl</span><span class="p">,</span> <span class="n">branch_pool</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">outputs</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InceptionC</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">channels_7x7</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch1x1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">c7</span> <span class="o">=</span> <span class="n">channels_7x7</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">c7</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">c7</span><span class="p">,</span> <span class="n">c7</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7_3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">c7</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">7</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">c7</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">c7</span><span class="p">,</span> <span class="n">c7</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">7</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">c7</span><span class="p">,</span> <span class="n">c7</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_4</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">c7</span><span class="p">,</span> <span class="n">c7</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">7</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_5</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">c7</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch_pool</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">branch1x1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch1x1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">branch7x7</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch7x7</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7_2</span><span class="p">(</span><span class="n">branch7x7</span><span class="p">)</span>
+        <span class="n">branch7x7</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7_3</span><span class="p">(</span><span class="n">branch7x7</span><span class="p">)</span>
+
+        <span class="n">branch7x7dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch7x7dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_2</span><span class="p">(</span><span class="n">branch7x7dbl</span><span class="p">)</span>
+        <span class="n">branch7x7dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_3</span><span class="p">(</span><span class="n">branch7x7dbl</span><span class="p">)</span>
+        <span class="n">branch7x7dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_4</span><span class="p">(</span><span class="n">branch7x7dbl</span><span class="p">)</span>
+        <span class="n">branch7x7dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7dbl_5</span><span class="p">(</span><span class="n">branch7x7dbl</span><span class="p">)</span>
+
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch_pool</span><span class="p">(</span><span class="n">branch_pool</span><span class="p">)</span>
+
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">branch1x1</span><span class="p">,</span> <span class="n">branch7x7</span><span class="p">,</span> <span class="n">branch7x7dbl</span><span class="p">,</span> <span class="n">branch_pool</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">outputs</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InceptionD</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_3</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">7</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_4</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2</span><span class="p">(</span><span class="n">branch3x3</span><span class="p">)</span>
+
+        <span class="n">branch7x7x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch7x7x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_2</span><span class="p">(</span><span class="n">branch7x7x3</span><span class="p">)</span>
+        <span class="n">branch7x7x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_3</span><span class="p">(</span><span class="n">branch7x7x3</span><span class="p">)</span>
+        <span class="n">branch7x7x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch7x7x3_4</span><span class="p">(</span><span class="n">branch7x7x3</span><span class="p">)</span>
+
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">max_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">branch3x3</span><span class="p">,</span> <span class="n">branch7x7x3</span><span class="p">,</span> <span class="n">branch_pool</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">outputs</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InceptionE</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch1x1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2a</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2b</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">448</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_2</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">448</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3a</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3b</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch_pool</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">branch1x1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch1x1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2a</span><span class="p">(</span><span class="n">branch3x3</span><span class="p">),</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2b</span><span class="p">(</span><span class="n">branch3x3</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">branch3x3</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_2</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3a</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">),</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3b</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch_pool</span><span class="p">(</span><span class="n">branch_pool</span><span class="p">)</span>
+
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">branch1x1</span><span class="p">,</span> <span class="n">branch3x3</span><span class="p">,</span> <span class="n">branch3x3dbl</span><span class="p">,</span> <span class="n">branch_pool</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">outputs</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InceptionAux</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">conv_block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">conv_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">conv_block</span> <span class="o">=</span> <span class="n">BasicConv2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv0</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">conv_block</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">5</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="o">.</span><span class="n">stddev</span> <span class="o">=</span> <span class="mf">0.01</span>  <span class="c1"># type: ignore[assignment]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">768</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="o">.</span><span class="n">stddev</span> <span class="o">=</span> <span class="mf">0.001</span>  <span class="c1"># type: ignore[assignment]</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># N x 768 x 17 x 17</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 5 x 5</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv0</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 128 x 5 x 5</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 768 x 1 x 1</span>
+        <span class="c1"># Adaptive average pooling</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adaptive_avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="c1"># N x 768 x 1 x 1</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># N x 768</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1000</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">BasicConv2d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="Inception_V3_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.Inception_V3_Weights">[docs]</a><span class="k">class</span> <span class="nc">Inception_V3_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/inception_v3_google-0cc3c7bd.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">299</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">342</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">27161264</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">75</span><span class="p">,</span> <span class="mi">75</span><span class="p">),</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#inception-v3&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.294</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.450</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">5.713</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">103.903</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights are ported from the original paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="inception_v3"><a class="viewcode-back" href="../../../models/generated/torchvision.models.inception_v3.html#torchvision.models.inception_v3">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Inception_V3_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">inception_v3</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Inception_V3_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Inception3</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Inception v3 model architecture from</span>
+<span class="sd">    `Rethinking the Inception Architecture for Computer Vision &lt;http://arxiv.org/abs/1512.00567&gt;`_.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        **Important**: In contrast to the other models the inception_v3 expects tensors with a size of</span>
+<span class="sd">        N x 3 x 299 x 299, so ensure your images are sized accordingly.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Inception_V3_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.Inception_V3_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.Inception3``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/inception.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Inception_V3_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Inception_V3_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">original_aux_logits</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="s2">&quot;transform_input&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;transform_input&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;init_weights&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">Inception3</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">original_aux_logits</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux_logits</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">AuxLogits</span> <span class="o">=</span> <span class="kc">None</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/maxvit.html b/0.16/_modules/torchvision/models/maxvit.html
new file mode 100644
index 00000000000..4a44dfa8be6
--- /dev/null
+++ b/0.16/_modules/torchvision/models/maxvit.html
@@ -0,0 +1,1501 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.maxvit &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.maxvit</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.maxvit</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torchvision.models._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">torchvision.models._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">torchvision.models._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span><span class="p">,</span> <span class="n">SqueezeExcitation</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops.stochastic_depth</span> <span class="kn">import</span> <span class="n">StochasticDepth</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">InterpolationMode</span>
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;MaxVit&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MaxVit_T_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;maxvit_t&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_get_conv_output_shape</span><span class="p">(</span><span class="n">input_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+    <span class="k">return</span> <span class="p">(</span>
+        <span class="p">(</span><span class="n">input_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="n">kernel_size</span> <span class="o">+</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">padding</span><span class="p">)</span> <span class="o">//</span> <span class="n">stride</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="p">(</span><span class="n">input_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">kernel_size</span> <span class="o">+</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">padding</span><span class="p">)</span> <span class="o">//</span> <span class="n">stride</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_make_block_input_shapes</span><span class="p">(</span><span class="n">input_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">n_blocks</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Util function to check that the input size is correct for a MaxVit configuration.&quot;&quot;&quot;</span>
+    <span class="n">shapes</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="n">block_input_shape</span> <span class="o">=</span> <span class="n">_get_conv_output_shape</span><span class="p">(</span><span class="n">input_size</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n_blocks</span><span class="p">):</span>
+        <span class="n">block_input_shape</span> <span class="o">=</span> <span class="n">_get_conv_output_shape</span><span class="p">(</span><span class="n">block_input_shape</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">shapes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block_input_shape</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">shapes</span>
+
+
+<span class="k">def</span> <span class="nf">_get_relative_position_index</span><span class="p">(</span><span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">coords</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">meshgrid</span><span class="p">([</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">height</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">width</span><span class="p">)]))</span>
+    <span class="n">coords_flat</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">coords</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="n">relative_coords</span> <span class="o">=</span> <span class="n">coords_flat</span><span class="p">[:,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">-</span> <span class="n">coords_flat</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span>
+    <span class="n">relative_coords</span> <span class="o">=</span> <span class="n">relative_coords</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+    <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">+=</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span>
+    <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+=</span> <span class="n">width</span> <span class="o">-</span> <span class="mi">1</span>
+    <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">*=</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">width</span> <span class="o">-</span> <span class="mi">1</span>
+    <span class="k">return</span> <span class="n">relative_coords</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">MBConv</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;MBConv: Mobile Inverted Residual Bottleneck.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): Number of input channels.</span>
+<span class="sd">        out_channels (int): Number of output channels.</span>
+<span class="sd">        expansion_ratio (float): Expansion ratio in the bottleneck.</span>
+<span class="sd">        squeeze_ratio (float): Squeeze ratio in the SE Layer.</span>
+<span class="sd">        stride (int): Stride of the depthwise convolution.</span>
+<span class="sd">        activation_layer (Callable[..., nn.Module]): Activation function.</span>
+<span class="sd">        norm_layer (Callable[..., nn.Module]): Normalization function.</span>
+<span class="sd">        p_stochastic_dropout (float): Probability of stochastic depth.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">expansion_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">squeeze_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">p_stochastic_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="n">proj</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span>
+
+        <span class="n">should_proj</span> <span class="o">=</span> <span class="n">stride</span> <span class="o">!=</span> <span class="mi">1</span> <span class="ow">or</span> <span class="n">in_channels</span> <span class="o">!=</span> <span class="n">out_channels</span>
+        <span class="k">if</span> <span class="n">should_proj</span><span class="p">:</span>
+            <span class="n">proj</span> <span class="o">=</span> <span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">)]</span>
+            <span class="k">if</span> <span class="n">stride</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+                <span class="n">proj</span> <span class="o">=</span> <span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">AvgPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)]</span> <span class="o">+</span> <span class="n">proj</span>  <span class="c1"># type: ignore</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">proj</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>  <span class="c1"># type: ignore</span>
+
+        <span class="n">mid_channels</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">out_channels</span> <span class="o">*</span> <span class="n">expansion_ratio</span><span class="p">)</span>
+        <span class="n">sqz_channels</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">out_channels</span> <span class="o">*</span> <span class="n">squeeze_ratio</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">p_stochastic_dropout</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span> <span class="o">=</span> <span class="n">StochasticDepth</span><span class="p">(</span><span class="n">p_stochastic_dropout</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;row&quot;</span><span class="p">)</span>  <span class="c1"># type: ignore</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>  <span class="c1"># type: ignore</span>
+
+        <span class="n">_layers</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+        <span class="n">_layers</span><span class="p">[</span><span class="s2">&quot;pre_norm&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">in_channels</span><span class="p">)</span>
+        <span class="n">_layers</span><span class="p">[</span><span class="s2">&quot;conv_a&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">mid_channels</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">inplace</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">_layers</span><span class="p">[</span><span class="s2">&quot;conv_b&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">mid_channels</span><span class="p">,</span>
+            <span class="n">mid_channels</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">groups</span><span class="o">=</span><span class="n">mid_channels</span><span class="p">,</span>
+            <span class="n">inplace</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">_layers</span><span class="p">[</span><span class="s2">&quot;squeeze_excitation&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">SqueezeExcitation</span><span class="p">(</span><span class="n">mid_channels</span><span class="p">,</span> <span class="n">sqz_channels</span><span class="p">,</span> <span class="n">activation</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">SiLU</span><span class="p">)</span>
+        <span class="n">_layers</span><span class="p">[</span><span class="s2">&quot;conv_c&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="n">mid_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">_layers</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Input tensor with expected layout of [B, C, H, W].</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Output tensor with expected layout of [B, C, H / stride, W / stride].</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="k">return</span> <span class="n">res</span> <span class="o">+</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">RelativePositionalMultiHeadAttention</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Relative Positional Multi-Head Attention.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        feat_dim (int): Number of input features.</span>
+<span class="sd">        head_dim (int): Number of features per head.</span>
+<span class="sd">        max_seq_len (int): Maximum sequence length.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">feat_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">head_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">max_seq_len</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">feat_dim</span> <span class="o">%</span> <span class="n">head_dim</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;feat_dim: </span><span class="si">{</span><span class="n">feat_dim</span><span class="si">}</span><span class="s2"> must be divisible by head_dim: </span><span class="si">{</span><span class="n">head_dim</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">n_heads</span> <span class="o">=</span> <span class="n">feat_dim</span> <span class="o">//</span> <span class="n">head_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span> <span class="o">=</span> <span class="n">head_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">max_seq_len</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len</span> <span class="o">=</span> <span class="n">max_seq_len</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">to_qkv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">feat_dim</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_heads</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span> <span class="o">*</span> <span class="mi">3</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale_factor</span> <span class="o">=</span> <span class="n">feat_dim</span><span class="o">**-</span><span class="mf">0.5</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">merge</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_heads</span><span class="p">,</span> <span class="n">feat_dim</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">parameter</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(((</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">-</span> <span class="mi">1</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_heads</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;relative_position_index&quot;</span><span class="p">,</span> <span class="n">_get_relative_position_index</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">))</span>
+        <span class="c1"># initialize with truncated normal the bias</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">get_relative_positional_bias</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">bias_index</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_index</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># type: ignore</span>
+        <span class="n">relative_bias</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span><span class="p">[</span><span class="n">bias_index</span><span class="p">]</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># type: ignore</span>
+        <span class="n">relative_bias</span> <span class="o">=</span> <span class="n">relative_bias</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+        <span class="k">return</span> <span class="n">relative_bias</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Input tensor with expected layout of [B, G, P, D].</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Output tensor with expected layout of [B, G, P, D].</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">B</span><span class="p">,</span> <span class="n">G</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">D</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">H</span><span class="p">,</span> <span class="n">DH</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span>
+
+        <span class="n">qkv</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">to_qkv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">chunk</span><span class="p">(</span><span class="n">qkv</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">q</span> <span class="o">=</span> <span class="n">q</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">G</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">DH</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+        <span class="n">k</span> <span class="o">=</span> <span class="n">k</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">G</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">DH</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+        <span class="n">v</span> <span class="o">=</span> <span class="n">v</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">G</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">DH</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+        <span class="n">k</span> <span class="o">=</span> <span class="n">k</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale_factor</span>
+        <span class="n">dot_prod</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s2">&quot;B G H I D, B G H J D -&gt; B G H I J&quot;</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">)</span>
+        <span class="n">pos_bias</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_relative_positional_bias</span><span class="p">()</span>
+
+        <span class="n">dot_prod</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">dot_prod</span> <span class="o">+</span> <span class="n">pos_bias</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s2">&quot;B G H I J, B G H J D -&gt; B G H I D&quot;</span><span class="p">,</span> <span class="n">dot_prod</span><span class="p">,</span> <span class="n">v</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">out</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">G</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">D</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">merge</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">SwapAxes</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Permute the axes of a tensor.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">a</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">b</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">a</span> <span class="o">=</span> <span class="n">a</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">b</span> <span class="o">=</span> <span class="n">b</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">swapaxes</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">a</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">b</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">res</span>
+
+
+<span class="k">class</span> <span class="nc">WindowPartition</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Partition the input tensor into non-overlapping windows.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Input tensor with expected layout of [B, C, H, W].</span>
+<span class="sd">            p (int): Number of partitions.</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Output tensor with expected layout of [B, H/P, W/P, P*P, C].</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">B</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">P</span> <span class="o">=</span> <span class="n">p</span>
+        <span class="c1"># chunk up H and W dimensions</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">H</span> <span class="o">//</span> <span class="n">P</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">W</span> <span class="o">//</span> <span class="n">P</span><span class="p">,</span> <span class="n">P</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># colapse P * P dimension</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="p">(</span><span class="n">H</span> <span class="o">//</span> <span class="n">P</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">W</span> <span class="o">//</span> <span class="n">P</span><span class="p">),</span> <span class="n">P</span> <span class="o">*</span> <span class="n">P</span><span class="p">,</span> <span class="n">C</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">WindowDepartition</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Departition the input tensor of non-overlapping windows into a feature volume of layout [B, C, H, W].</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">h_partitions</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">w_partitions</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Input tensor with expected layout of [B, (H/P * W/P), P*P, C].</span>
+<span class="sd">            p (int): Number of partitions.</span>
+<span class="sd">            h_partitions (int): Number of vertical partitions.</span>
+<span class="sd">            w_partitions (int): Number of horizontal partitions.</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Output tensor with expected layout of [B, C, H, W].</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">B</span><span class="p">,</span> <span class="n">G</span><span class="p">,</span> <span class="n">PP</span><span class="p">,</span> <span class="n">C</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">P</span> <span class="o">=</span> <span class="n">p</span>
+        <span class="n">HP</span><span class="p">,</span> <span class="n">WP</span> <span class="o">=</span> <span class="n">h_partitions</span><span class="p">,</span> <span class="n">w_partitions</span>
+        <span class="c1"># split P * P dimension into 2 P tile dimensionsa</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">HP</span><span class="p">,</span> <span class="n">WP</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">P</span><span class="p">,</span> <span class="n">C</span><span class="p">)</span>
+        <span class="c1"># permute into B, C, HP, P, WP, P</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+        <span class="c1"># reshape into B, C, H, W</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">HP</span> <span class="o">*</span> <span class="n">P</span><span class="p">,</span> <span class="n">WP</span> <span class="o">*</span> <span class="n">P</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">PartitionAttentionLayer</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Layer for partitioning the input tensor into non-overlapping windows and applying attention to each window.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): Number of input channels.</span>
+<span class="sd">        head_dim (int): Dimension of each attention head.</span>
+<span class="sd">        partition_size (int): Size of the partitions.</span>
+<span class="sd">        partition_type (str): Type of partitioning to use. Can be either &quot;grid&quot; or &quot;window&quot;.</span>
+<span class="sd">        grid_size (Tuple[int, int]): Size of the grid to partition the input tensor into.</span>
+<span class="sd">        mlp_ratio (int): Ratio of the  feature size expansion in the MLP layer.</span>
+<span class="sd">        activation_layer (Callable[..., nn.Module]): Activation function to use.</span>
+<span class="sd">        norm_layer (Callable[..., nn.Module]): Normalization function to use.</span>
+<span class="sd">        attention_dropout (float): Dropout probability for the attention layer.</span>
+<span class="sd">        mlp_dropout (float): Dropout probability for the MLP layer.</span>
+<span class="sd">        p_stochastic_dropout (float): Probability of dropping out a partition.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">head_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="c1"># partitioning parameters</span>
+        <span class="n">partition_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">partition_type</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="c1"># grid size needs to be known at initialization time</span>
+        <span class="c1"># because we need to know hamy relative offsets there are in the grid</span>
+        <span class="n">grid_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">mlp_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">p_stochastic_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">n_heads</span> <span class="o">=</span> <span class="n">in_channels</span> <span class="o">//</span> <span class="n">head_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span> <span class="o">=</span> <span class="n">head_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">n_partitions</span> <span class="o">=</span> <span class="n">grid_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="n">partition_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">partition_type</span> <span class="o">=</span> <span class="n">partition_type</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span> <span class="o">=</span> <span class="n">grid_size</span>
+
+        <span class="k">if</span> <span class="n">partition_type</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;grid&quot;</span><span class="p">,</span> <span class="s2">&quot;window&quot;</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;partition_type must be either &#39;grid&#39; or &#39;window&#39;&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">partition_type</span> <span class="o">==</span> <span class="s2">&quot;window&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">g</span> <span class="o">=</span> <span class="n">partition_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_partitions</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">g</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_partitions</span><span class="p">,</span> <span class="n">partition_size</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">partition_op</span> <span class="o">=</span> <span class="n">WindowPartition</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">departition_op</span> <span class="o">=</span> <span class="n">WindowDepartition</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">partition_swap</span> <span class="o">=</span> <span class="n">SwapAxes</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">3</span><span class="p">)</span> <span class="k">if</span> <span class="n">partition_type</span> <span class="o">==</span> <span class="s2">&quot;grid&quot;</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">departition_swap</span> <span class="o">=</span> <span class="n">SwapAxes</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">3</span><span class="p">)</span> <span class="k">if</span> <span class="n">partition_type</span> <span class="o">==</span> <span class="s2">&quot;grid&quot;</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">attn_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">norm_layer</span><span class="p">(</span><span class="n">in_channels</span><span class="p">),</span>
+            <span class="c1"># it&#39;s always going to be partition_size ** 2 because</span>
+            <span class="c1"># of the axis swap in the case of grid partitioning</span>
+            <span class="n">RelativePositionalMultiHeadAttention</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">head_dim</span><span class="p">,</span> <span class="n">partition_size</span><span class="o">**</span><span class="mi">2</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">attention_dropout</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="c1"># pre-normalization similar to transformer layers</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mlp_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">(</span><span class="n">in_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">in_channels</span> <span class="o">*</span> <span class="n">mlp_ratio</span><span class="p">),</span>
+            <span class="n">activation_layer</span><span class="p">(),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_channels</span> <span class="o">*</span> <span class="n">mlp_ratio</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">mlp_dropout</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="c1"># layer scale factors</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_dropout</span> <span class="o">=</span> <span class="n">StochasticDepth</span><span class="p">(</span><span class="n">p_stochastic_dropout</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;row&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Input tensor with expected layout of [B, C, H, W].</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Output tensor with expected layout of [B, C, H, W].</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="c1"># Undefined behavior if H or W are not divisible by p</span>
+        <span class="c1"># https://github.com/google-research/maxvit/blob/da76cf0d8a6ec668cc31b399c4126186da7da944/maxvit/models/maxvit.py#L766</span>
+        <span class="n">gh</span><span class="p">,</span> <span class="n">gw</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">%</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">%</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span>
+            <span class="s2">&quot;Grid size must be divisible by partition size. Got grid size of </span><span class="si">{}</span><span class="s2"> and partition size of </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span>
+            <span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">partition_op</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">partition_swap</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_dropout</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">attn_layer</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_dropout</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mlp_layer</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">departition_swap</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">departition_op</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">,</span> <span class="n">gh</span><span class="p">,</span> <span class="n">gw</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">MaxVitLayer</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    MaxVit layer consisting of a MBConv layer followed by a PartitionAttentionLayer with `window` and a PartitionAttentionLayer with `grid`.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): Number of input channels.</span>
+<span class="sd">        out_channels (int): Number of output channels.</span>
+<span class="sd">        expansion_ratio (float): Expansion ratio in the bottleneck.</span>
+<span class="sd">        squeeze_ratio (float): Squeeze ratio in the SE Layer.</span>
+<span class="sd">        stride (int): Stride of the depthwise convolution.</span>
+<span class="sd">        activation_layer (Callable[..., nn.Module]): Activation function.</span>
+<span class="sd">        norm_layer (Callable[..., nn.Module]): Normalization function.</span>
+<span class="sd">        head_dim (int): Dimension of the attention heads.</span>
+<span class="sd">        mlp_ratio (int): Ratio of the MLP layer.</span>
+<span class="sd">        mlp_dropout (float): Dropout probability for the MLP layer.</span>
+<span class="sd">        attention_dropout (float): Dropout probability for the attention layer.</span>
+<span class="sd">        p_stochastic_dropout (float): Probability of stochastic depth.</span>
+<span class="sd">        partition_size (int): Size of the partitions.</span>
+<span class="sd">        grid_size (Tuple[int, int]): Size of the input feature grid.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="c1"># conv parameters</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">squeeze_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">expansion_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="c1"># conv + transformer parameters</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="c1"># transformer parameters</span>
+        <span class="n">head_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">mlp_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">p_stochastic_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="c1"># partitioning parameters</span>
+        <span class="n">partition_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">grid_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">OrderedDict</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+
+        <span class="c1"># convolutional layer</span>
+        <span class="n">layers</span><span class="p">[</span><span class="s2">&quot;MBconv&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">MBConv</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="o">=</span><span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">out_channels</span><span class="o">=</span><span class="n">out_channels</span><span class="p">,</span>
+            <span class="n">expansion_ratio</span><span class="o">=</span><span class="n">expansion_ratio</span><span class="p">,</span>
+            <span class="n">squeeze_ratio</span><span class="o">=</span><span class="n">squeeze_ratio</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">p_stochastic_dropout</span><span class="o">=</span><span class="n">p_stochastic_dropout</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="c1"># attention layers, block -&gt; grid</span>
+        <span class="n">layers</span><span class="p">[</span><span class="s2">&quot;window_attention&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">PartitionAttentionLayer</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="o">=</span><span class="n">out_channels</span><span class="p">,</span>
+            <span class="n">head_dim</span><span class="o">=</span><span class="n">head_dim</span><span class="p">,</span>
+            <span class="n">partition_size</span><span class="o">=</span><span class="n">partition_size</span><span class="p">,</span>
+            <span class="n">partition_type</span><span class="o">=</span><span class="s2">&quot;window&quot;</span><span class="p">,</span>
+            <span class="n">grid_size</span><span class="o">=</span><span class="n">grid_size</span><span class="p">,</span>
+            <span class="n">mlp_ratio</span><span class="o">=</span><span class="n">mlp_ratio</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">mlp_dropout</span><span class="o">=</span><span class="n">mlp_dropout</span><span class="p">,</span>
+            <span class="n">p_stochastic_dropout</span><span class="o">=</span><span class="n">p_stochastic_dropout</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">layers</span><span class="p">[</span><span class="s2">&quot;grid_attention&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">PartitionAttentionLayer</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="o">=</span><span class="n">out_channels</span><span class="p">,</span>
+            <span class="n">head_dim</span><span class="o">=</span><span class="n">head_dim</span><span class="p">,</span>
+            <span class="n">partition_size</span><span class="o">=</span><span class="n">partition_size</span><span class="p">,</span>
+            <span class="n">partition_type</span><span class="o">=</span><span class="s2">&quot;grid&quot;</span><span class="p">,</span>
+            <span class="n">grid_size</span><span class="o">=</span><span class="n">grid_size</span><span class="p">,</span>
+            <span class="n">mlp_ratio</span><span class="o">=</span><span class="n">mlp_ratio</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">mlp_dropout</span><span class="o">=</span><span class="n">mlp_dropout</span><span class="p">,</span>
+            <span class="n">p_stochastic_dropout</span><span class="o">=</span><span class="n">p_stochastic_dropout</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">layers</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Input tensor of shape (B, C, H, W).</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Output tensor of shape (B, C, H, W).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">MaxVitBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    A MaxVit block consisting of `n_layers` MaxVit layers.</span>
+
+<span class="sd">     Args:</span>
+<span class="sd">        in_channels (int): Number of input channels.</span>
+<span class="sd">        out_channels (int): Number of output channels.</span>
+<span class="sd">        expansion_ratio (float): Expansion ratio in the bottleneck.</span>
+<span class="sd">        squeeze_ratio (float): Squeeze ratio in the SE Layer.</span>
+<span class="sd">        activation_layer (Callable[..., nn.Module]): Activation function.</span>
+<span class="sd">        norm_layer (Callable[..., nn.Module]): Normalization function.</span>
+<span class="sd">        head_dim (int): Dimension of the attention heads.</span>
+<span class="sd">        mlp_ratio (int): Ratio of the MLP layer.</span>
+<span class="sd">        mlp_dropout (float): Dropout probability for the MLP layer.</span>
+<span class="sd">        attention_dropout (float): Dropout probability for the attention layer.</span>
+<span class="sd">        p_stochastic_dropout (float): Probability of stochastic depth.</span>
+<span class="sd">        partition_size (int): Size of the partitions.</span>
+<span class="sd">        input_grid_size (Tuple[int, int]): Size of the input feature grid.</span>
+<span class="sd">        n_layers (int): Number of layers in the block.</span>
+<span class="sd">        p_stochastic (List[float]): List of probabilities for stochastic depth for each layer.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="c1"># conv parameters</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">squeeze_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">expansion_ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="c1"># conv + transformer parameters</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="c1"># transformer parameters</span>
+        <span class="n">head_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">mlp_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="c1"># partitioning parameters</span>
+        <span class="n">partition_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">input_grid_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="c1"># number of layers</span>
+        <span class="n">n_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">p_stochastic</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">len</span><span class="p">(</span><span class="n">p_stochastic</span><span class="p">)</span> <span class="o">==</span> <span class="n">n_layers</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;p_stochastic must have length n_layers=</span><span class="si">{</span><span class="n">n_layers</span><span class="si">}</span><span class="s2">, got p_stochastic=</span><span class="si">{</span><span class="n">p_stochastic</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="c1"># account for the first stride of the first layer</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span> <span class="o">=</span> <span class="n">_get_conv_output_shape</span><span class="p">(</span><span class="n">input_grid_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">p_stochastic</span><span class="p">):</span>
+            <span class="n">stride</span> <span class="o">=</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">idx</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="mi">1</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">+=</span> <span class="p">[</span>
+                <span class="n">MaxVitLayer</span><span class="p">(</span>
+                    <span class="n">in_channels</span><span class="o">=</span><span class="n">in_channels</span> <span class="k">if</span> <span class="n">idx</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">out_channels</span><span class="p">,</span>
+                    <span class="n">out_channels</span><span class="o">=</span><span class="n">out_channels</span><span class="p">,</span>
+                    <span class="n">squeeze_ratio</span><span class="o">=</span><span class="n">squeeze_ratio</span><span class="p">,</span>
+                    <span class="n">expansion_ratio</span><span class="o">=</span><span class="n">expansion_ratio</span><span class="p">,</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+                    <span class="n">head_dim</span><span class="o">=</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">mlp_ratio</span><span class="o">=</span><span class="n">mlp_ratio</span><span class="p">,</span>
+                    <span class="n">mlp_dropout</span><span class="o">=</span><span class="n">mlp_dropout</span><span class="p">,</span>
+                    <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+                    <span class="n">partition_size</span><span class="o">=</span><span class="n">partition_size</span><span class="p">,</span>
+                    <span class="n">grid_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">grid_size</span><span class="p">,</span>
+                    <span class="n">p_stochastic_dropout</span><span class="o">=</span><span class="n">p</span><span class="p">,</span>
+                <span class="p">),</span>
+            <span class="p">]</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Input tensor of shape (B, C, H, W).</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Output tensor of shape (B, C, H, W).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">layer</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">MaxVit</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements MaxVit Transformer from the `MaxViT: Multi-Axis Vision Transformer &lt;https://arxiv.org/abs/2204.01697&gt;`_ paper.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        input_size (Tuple[int, int]): Size of the input image.</span>
+<span class="sd">        stem_channels (int): Number of channels in the stem.</span>
+<span class="sd">        partition_size (int): Size of the partitions.</span>
+<span class="sd">        block_channels (List[int]): Number of channels in each block.</span>
+<span class="sd">        block_layers (List[int]): Number of layers in each block.</span>
+<span class="sd">        stochastic_depth_prob (float): Probability of stochastic depth. Expands to a list of probabilities for each layer that scales linearly to the specified value.</span>
+<span class="sd">        squeeze_ratio (float): Squeeze ratio in the SE Layer. Default: 0.25.</span>
+<span class="sd">        expansion_ratio (float): Expansion ratio in the MBConv bottleneck. Default: 4.</span>
+<span class="sd">        norm_layer (Callable[..., nn.Module]): Normalization function. Default: None (setting to None will produce a `BatchNorm2d(eps=1e-3, momentum=0.99)`).</span>
+<span class="sd">        activation_layer (Callable[..., nn.Module]): Activation function Default: nn.GELU.</span>
+<span class="sd">        head_dim (int): Dimension of the attention heads.</span>
+<span class="sd">        mlp_ratio (int): Expansion ratio of the MLP layer. Default: 4.</span>
+<span class="sd">        mlp_dropout (float): Dropout probability for the MLP layer. Default: 0.0.</span>
+<span class="sd">        attention_dropout (float): Dropout probability for the attention layer. Default: 0.0.</span>
+<span class="sd">        num_classes (int): Number of classes. Default: 1000.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="c1"># input size parameters</span>
+        <span class="n">input_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="c1"># stem and task parameters</span>
+        <span class="n">stem_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="c1"># partitioning parameters</span>
+        <span class="n">partition_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="c1"># block parameters</span>
+        <span class="n">block_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">block_layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="c1"># attention head dimensions</span>
+        <span class="n">head_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="c1"># conv + transformer parameters</span>
+        <span class="c1"># norm_layer is applied only to the conv layers</span>
+        <span class="c1"># activation_layer is applied both to conv and transformer layers</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">GELU</span><span class="p">,</span>
+        <span class="c1"># conv parameters</span>
+        <span class="n">squeeze_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.25</span><span class="p">,</span>
+        <span class="n">expansion_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="c1"># transformer parameters</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="n">mlp_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="c1"># task parameters</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="n">input_channels</span> <span class="o">=</span> <span class="mi">3</span>
+
+        <span class="c1"># https://github.com/google-research/maxvit/blob/da76cf0d8a6ec668cc31b399c4126186da7da944/maxvit/models/maxvit.py#L1029-L1030</span>
+        <span class="c1"># for the exact parameters used in batchnorm</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.99</span><span class="p">)</span>
+
+        <span class="c1"># Make sure input size will be divisible by the partition size in all blocks</span>
+        <span class="c1"># Undefined behavior if H or W are not divisible by p</span>
+        <span class="c1"># https://github.com/google-research/maxvit/blob/da76cf0d8a6ec668cc31b399c4126186da7da944/maxvit/models/maxvit.py#L766</span>
+        <span class="n">block_input_sizes</span> <span class="o">=</span> <span class="n">_make_block_input_shapes</span><span class="p">(</span><span class="n">input_size</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">block_channels</span><span class="p">))</span>
+        <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">block_input_size</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">block_input_sizes</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">block_input_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">%</span> <span class="n">partition_size</span> <span class="o">!=</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">block_input_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">%</span> <span class="n">partition_size</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Input size </span><span class="si">{</span><span class="n">block_input_size</span><span class="si">}</span><span class="s2"> of block </span><span class="si">{</span><span class="n">idx</span><span class="si">}</span><span class="s2"> is not divisible by partition size </span><span class="si">{</span><span class="n">partition_size</span><span class="si">}</span><span class="s2">. &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;Consider changing the partition size or the input size.</span><span class="se">\n</span><span class="s2">&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;Current configuration yields the following block input sizes: </span><span class="si">{</span><span class="n">block_input_sizes</span><span class="si">}</span><span class="s2">.&quot;</span>
+                <span class="p">)</span>
+
+        <span class="c1"># stem</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stem</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">input_channels</span><span class="p">,</span>
+                <span class="n">stem_channels</span><span class="p">,</span>
+                <span class="mi">3</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="n">inplace</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">stem_channels</span><span class="p">,</span> <span class="n">stem_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span>
+            <span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="c1"># account for stem stride</span>
+        <span class="n">input_size</span> <span class="o">=</span> <span class="n">_get_conv_output_shape</span><span class="p">(</span><span class="n">input_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">partition_size</span> <span class="o">=</span> <span class="n">partition_size</span>
+
+        <span class="c1"># blocks</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">blocks</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="n">in_channels</span> <span class="o">=</span> <span class="p">[</span><span class="n">stem_channels</span><span class="p">]</span> <span class="o">+</span> <span class="n">block_channels</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">out_channels</span> <span class="o">=</span> <span class="n">block_channels</span>
+
+        <span class="c1"># precompute the stochastich depth probabilities from 0 to stochastic_depth_prob</span>
+        <span class="c1"># since we have N blocks with L layers, we will have N * L probabilities uniformly distributed</span>
+        <span class="c1"># over the range [0, stochastic_depth_prob]</span>
+        <span class="n">p_stochastic</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="nb">sum</span><span class="p">(</span><span class="n">block_layers</span><span class="p">))</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+
+        <span class="n">p_idx</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">in_channel</span><span class="p">,</span> <span class="n">out_channel</span><span class="p">,</span> <span class="n">num_layers</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">block_layers</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">MaxVitBlock</span><span class="p">(</span>
+                    <span class="n">in_channels</span><span class="o">=</span><span class="n">in_channel</span><span class="p">,</span>
+                    <span class="n">out_channels</span><span class="o">=</span><span class="n">out_channel</span><span class="p">,</span>
+                    <span class="n">squeeze_ratio</span><span class="o">=</span><span class="n">squeeze_ratio</span><span class="p">,</span>
+                    <span class="n">expansion_ratio</span><span class="o">=</span><span class="n">expansion_ratio</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+                    <span class="n">head_dim</span><span class="o">=</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">mlp_ratio</span><span class="o">=</span><span class="n">mlp_ratio</span><span class="p">,</span>
+                    <span class="n">mlp_dropout</span><span class="o">=</span><span class="n">mlp_dropout</span><span class="p">,</span>
+                    <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+                    <span class="n">partition_size</span><span class="o">=</span><span class="n">partition_size</span><span class="p">,</span>
+                    <span class="n">input_grid_size</span><span class="o">=</span><span class="n">input_size</span><span class="p">,</span>
+                    <span class="n">n_layers</span><span class="o">=</span><span class="n">num_layers</span><span class="p">,</span>
+                    <span class="n">p_stochastic</span><span class="o">=</span><span class="n">p_stochastic</span><span class="p">[</span><span class="n">p_idx</span> <span class="p">:</span> <span class="n">p_idx</span> <span class="o">+</span> <span class="n">num_layers</span><span class="p">],</span>
+                <span class="p">),</span>
+            <span class="p">)</span>
+            <span class="n">input_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">blocks</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">grid_size</span>
+            <span class="n">p_idx</span> <span class="o">+=</span> <span class="n">num_layers</span>
+
+        <span class="c1"># see https://github.com/google-research/maxvit/blob/da76cf0d8a6ec668cc31b399c4126186da7da944/maxvit/models/maxvit.py#L1137-L1158</span>
+        <span class="c1"># for why there is Linear -&gt; Tanh -&gt; Linear</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Flatten</span><span class="p">(),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">(</span><span class="n">block_channels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">block_channels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">block_channels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Tanh</span><span class="p">(),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">block_channels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_init_weights</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stem</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">block</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">blocks</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">block</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">_init_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_maxvit</span><span class="p">(</span>
+    <span class="c1"># stem parameters</span>
+    <span class="n">stem_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="c1"># block parameters</span>
+    <span class="n">block_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">block_layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="c1"># partitioning parameters</span>
+    <span class="n">partition_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="c1"># transformer parameters</span>
+    <span class="n">head_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="c1"># Weights API</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="c1"># kwargs,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MaxVit</span><span class="p">:</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">assert</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;input_size&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">])</span>
+
+    <span class="n">input_size</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;input_size&quot;</span><span class="p">,</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">MaxVit</span><span class="p">(</span>
+        <span class="n">stem_channels</span><span class="o">=</span><span class="n">stem_channels</span><span class="p">,</span>
+        <span class="n">block_channels</span><span class="o">=</span><span class="n">block_channels</span><span class="p">,</span>
+        <span class="n">block_layers</span><span class="o">=</span><span class="n">block_layers</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">stochastic_depth_prob</span><span class="p">,</span>
+        <span class="n">head_dim</span><span class="o">=</span><span class="n">head_dim</span><span class="p">,</span>
+        <span class="n">partition_size</span><span class="o">=</span><span class="n">partition_size</span><span class="p">,</span>
+        <span class="n">input_size</span><span class="o">=</span><span class="n">input_size</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<div class="viewcode-block" id="MaxVit_T_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.maxvit_t.html#torchvision.models.MaxVit_T_Weights">[docs]</a><span class="k">class</span> <span class="nc">MaxVit_T_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># URL empty until official release</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/maxvit_t-bc5ab103.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">30919624</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#maxvit&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.700</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.722</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">5.558</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">118.769</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a similar training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="maxvit_t"><a class="viewcode-back" href="../../../models/generated/torchvision.models.maxvit_t.html#torchvision.models.maxvit_t">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MaxVit_T_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">maxvit_t</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MaxVit_T_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MaxVit</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a maxvit_t architecture from</span>
+<span class="sd">    `MaxViT: Multi-Axis Vision Transformer &lt;https://arxiv.org/abs/2204.01697&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MaxVit_T_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MaxVit_T_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.maxvit.MaxVit``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/maxvit.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MaxVit_T_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MaxVit_T_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_maxvit</span><span class="p">(</span>
+        <span class="n">stem_channels</span><span class="o">=</span><span class="mi">64</span><span class="p">,</span>
+        <span class="n">block_channels</span><span class="o">=</span><span class="p">[</span><span class="mi">64</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">512</span><span class="p">],</span>
+        <span class="n">block_layers</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">head_dim</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">partition_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/mnasnet.html b/0.16/_modules/torchvision/models/mnasnet.html
new file mode 100644
index 00000000000..b30ce88b745
--- /dev/null
+++ b/0.16/_modules/torchvision/models/mnasnet.html
@@ -0,0 +1,1103 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.mnasnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.mnasnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.mnasnet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;MNASNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MNASNet0_5_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MNASNet0_75_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MNASNet1_0_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MNASNet1_3_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mnasnet0_5&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mnasnet0_75&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mnasnet1_0&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mnasnet1_3&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="c1"># Paper suggests 0.9997 momentum, for TensorFlow. Equivalent PyTorch momentum is</span>
+<span class="c1"># 1.0 - tensorflow.</span>
+<span class="n">_BN_MOMENTUM</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="mf">0.9997</span>
+
+
+<span class="k">class</span> <span class="nc">_InvertedResidual</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_ch</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_ch</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">expansion_factor</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">bn_momentum</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">stride</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;stride should be 1 or 2 instead of </span><span class="si">{</span><span class="n">stride</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">kernel_size</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;kernel_size should be 3 or 5 instead of </span><span class="si">{</span><span class="n">kernel_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">mid_ch</span> <span class="o">=</span> <span class="n">in_ch</span> <span class="o">*</span> <span class="n">expansion_factor</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">apply_residual</span> <span class="o">=</span> <span class="n">in_ch</span> <span class="o">==</span> <span class="n">out_ch</span> <span class="ow">and</span> <span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="c1"># Pointwise</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_ch</span><span class="p">,</span> <span class="n">mid_ch</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">mid_ch</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="n">bn_momentum</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="c1"># Depthwise</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">mid_ch</span><span class="p">,</span> <span class="n">mid_ch</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">kernel_size</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">mid_ch</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">mid_ch</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="n">bn_momentum</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="c1"># Linear pointwise. Note that there&#39;s no activation.</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">mid_ch</span><span class="p">,</span> <span class="n">out_ch</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">out_ch</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="n">bn_momentum</span><span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">apply_residual</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span> <span class="o">+</span> <span class="nb">input</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_stack</span><span class="p">(</span>
+    <span class="n">in_ch</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_ch</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">exp_factor</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">repeats</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">bn_momentum</span><span class="p">:</span> <span class="nb">float</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Creates a stack of inverted residuals.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">repeats</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;repeats should be &gt;= 1, instead got </span><span class="si">{</span><span class="n">repeats</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="c1"># First one has no skip, because feature map size changes.</span>
+    <span class="n">first</span> <span class="o">=</span> <span class="n">_InvertedResidual</span><span class="p">(</span><span class="n">in_ch</span><span class="p">,</span> <span class="n">out_ch</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">exp_factor</span><span class="p">,</span> <span class="n">bn_momentum</span><span class="o">=</span><span class="n">bn_momentum</span><span class="p">)</span>
+    <span class="n">remaining</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">repeats</span><span class="p">):</span>
+        <span class="n">remaining</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">_InvertedResidual</span><span class="p">(</span><span class="n">out_ch</span><span class="p">,</span> <span class="n">out_ch</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">exp_factor</span><span class="p">,</span> <span class="n">bn_momentum</span><span class="o">=</span><span class="n">bn_momentum</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">first</span><span class="p">,</span> <span class="o">*</span><span class="n">remaining</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_round_to_multiple_of</span><span class="p">(</span><span class="n">val</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">divisor</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">round_up_bias</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.9</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Asymmetric rounding to make `val` divisible by `divisor`. With default</span>
+<span class="sd">    bias, will round up, unless the number is no more than 10% greater than the</span>
+<span class="sd">    smaller divisible value, i.e. (83, 8) -&gt; 80, but (84, 8) -&gt; 88.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;</span> <span class="n">round_up_bias</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;round_up_bias should be greater than 0.0 and smaller than 1.0 instead of </span><span class="si">{</span><span class="n">round_up_bias</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="n">new_val</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">divisor</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">val</span> <span class="o">+</span> <span class="n">divisor</span> <span class="o">/</span> <span class="mi">2</span><span class="p">)</span> <span class="o">//</span> <span class="n">divisor</span> <span class="o">*</span> <span class="n">divisor</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">new_val</span> <span class="k">if</span> <span class="n">new_val</span> <span class="o">&gt;=</span> <span class="n">round_up_bias</span> <span class="o">*</span> <span class="n">val</span> <span class="k">else</span> <span class="n">new_val</span> <span class="o">+</span> <span class="n">divisor</span>
+
+
+<span class="k">def</span> <span class="nf">_get_depths</span><span class="p">(</span><span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Scales tensor depths as in reference MobileNet code, prefers rounding up</span>
+<span class="sd">    rather than down.&quot;&quot;&quot;</span>
+    <span class="n">depths</span> <span class="o">=</span> <span class="p">[</span><span class="mi">32</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">320</span><span class="p">]</span>
+    <span class="k">return</span> <span class="p">[</span><span class="n">_round_to_multiple_of</span><span class="p">(</span><span class="n">depth</span> <span class="o">*</span> <span class="n">alpha</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span> <span class="k">for</span> <span class="n">depth</span> <span class="ow">in</span> <span class="n">depths</span><span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">MNASNet</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;MNASNet, as described in https://arxiv.org/abs/1807.11626. This</span>
+<span class="sd">    implements the B1 variant of the model.</span>
+<span class="sd">    &gt;&gt;&gt; model = MNASNet(1.0, num_classes=1000)</span>
+<span class="sd">    &gt;&gt;&gt; x = torch.rand(1, 3, 224, 224)</span>
+<span class="sd">    &gt;&gt;&gt; y = model(x)</span>
+<span class="sd">    &gt;&gt;&gt; y.dim()</span>
+<span class="sd">    2</span>
+<span class="sd">    &gt;&gt;&gt; y.nelement()</span>
+<span class="sd">    1000</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="c1"># Version 2 adds depth scaling in the initial stages of the network.</span>
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.2</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">alpha</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;alpha should be greater than 0.0 instead of </span><span class="si">{</span><span class="n">alpha</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span> <span class="o">=</span> <span class="n">alpha</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+        <span class="n">depths</span> <span class="o">=</span> <span class="n">_get_depths</span><span class="p">(</span><span class="n">alpha</span><span class="p">)</span>
+        <span class="n">layers</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="c1"># First layer: regular conv.</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">depths</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">momentum</span><span class="o">=</span><span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="c1"># Depthwise separable, no skip.</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">depths</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">momentum</span><span class="o">=</span><span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">momentum</span><span class="o">=</span><span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="c1"># MNASNet blocks: stacks of inverted residuals.</span>
+            <span class="n">_stack</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">_stack</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">_stack</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">4</span><span class="p">],</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">_stack</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">4</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">5</span><span class="p">],</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">_stack</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">5</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">6</span><span class="p">],</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">_stack</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">6</span><span class="p">],</span> <span class="n">depths</span><span class="p">[</span><span class="mi">7</span><span class="p">],</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="c1"># Final mapping to classifier input.</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="mi">7</span><span class="p">],</span> <span class="mi">1280</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="mi">1280</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">1280</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">))</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">ones_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_uniform_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;sigmoid&quot;</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># Equivalent to global avgpool and removing H and W dimensions.</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">mean</span><span class="p">([</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">])</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">:</span> <span class="n">Dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">:</span> <span class="n">Dict</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+        <span class="n">unexpected_keys</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+        <span class="n">error_msgs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">version</span> <span class="o">=</span> <span class="n">local_metadata</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">version</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;version shluld be set to 1 or 2 instead of </span><span class="si">{</span><span class="n">version</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">version</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span> <span class="o">==</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="c1"># In the initial version of the model (v1), stem was fixed-size.</span>
+            <span class="c1"># All other layer configurations were the same. This will patch</span>
+            <span class="c1"># the model so that it&#39;s identical to v1. Model with alpha 1.0 is</span>
+            <span class="c1"># unaffected.</span>
+            <span class="n">depths</span> <span class="o">=</span> <span class="n">_get_depths</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">)</span>
+            <span class="n">v1_stem</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="mi">16</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+                <span class="n">_stack</span><span class="p">(</span><span class="mi">16</span><span class="p">,</span> <span class="n">depths</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">_BN_MOMENTUM</span><span class="p">),</span>
+            <span class="p">]</span>
+            <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">layer</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">v1_stem</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span> <span class="o">=</span> <span class="n">layer</span>
+
+            <span class="c1"># The model is now identical to v1, and must be saved as such.</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_version</span> <span class="o">=</span> <span class="mi">1</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;A new version of MNASNet model has been implemented. &quot;</span>
+                <span class="s2">&quot;Your checkpoint was saved using the previous version. &quot;</span>
+                <span class="s2">&quot;This checkpoint will load and work as before, but &quot;</span>
+                <span class="s2">&quot;you may want to upgrade by training a newer model or &quot;</span>
+                <span class="s2">&quot;transfer learning from an updated ImageNet checkpoint.&quot;</span><span class="p">,</span>
+                <span class="ne">UserWarning</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span> <span class="n">prefix</span><span class="p">,</span> <span class="n">local_metadata</span><span class="p">,</span> <span class="n">strict</span><span class="p">,</span> <span class="n">missing_keys</span><span class="p">,</span> <span class="n">unexpected_keys</span><span class="p">,</span> <span class="n">error_msgs</span>
+        <span class="p">)</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/1e100/mnasnet_trainer&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="MNASNet0_5_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet0_5.html#torchvision.models.MNASNet0_5_Weights">[docs]</a><span class="k">class</span> <span class="nc">MNASNet0_5_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mnasnet0.5_top1_67.823-3ffadce67e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">2218512</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">67.734</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">87.490</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.104</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">8.591</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="MNASNet0_75_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet0_75.html#torchvision.models.MNASNet0_75_Weights">[docs]</a><span class="k">class</span> <span class="nc">MNASNet0_75_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mnasnet0_75-7090bc5f.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/6019&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">3170208</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">71.180</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.496</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.215</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">12.303</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="MNASNet1_0_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet1_0.html#torchvision.models.MNASNet1_0_Weights">[docs]</a><span class="k">class</span> <span class="nc">MNASNet1_0_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mnasnet1.0_top1_73.512-f206786ef8.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">4383312</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">73.456</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.510</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.314</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">16.915</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="MNASNet1_3_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet1_3.html#torchvision.models.MNASNet1_3_Weights">[docs]</a><span class="k">class</span> <span class="nc">MNASNet1_3_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mnasnet1_3-a4c69d6f.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/6019&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">6282256</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">76.506</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.522</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.526</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">24.246</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<span class="k">def</span> <span class="nf">_mnasnet</span><span class="p">(</span><span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MNASNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">MNASNet</span><span class="p">(</span><span class="n">alpha</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<div class="viewcode-block" id="mnasnet0_5"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet0_5.html#torchvision.models.mnasnet0_5">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MNASNet0_5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mnasnet0_5</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MNASNet0_5_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MNASNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;MNASNet with depth multiplier of 0.5 from</span>
+<span class="sd">    `MnasNet: Platform-Aware Neural Architecture Search for Mobile</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11626&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MNASNet0_5_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MNASNet0_5_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.mnasnet.MNASNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MNASNet0_5_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MNASNet0_5_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_mnasnet</span><span class="p">(</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="mnasnet0_75"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet0_75.html#torchvision.models.mnasnet0_75">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MNASNet0_75_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mnasnet0_75</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MNASNet0_75_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MNASNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;MNASNet with depth multiplier of 0.75 from</span>
+<span class="sd">    `MnasNet: Platform-Aware Neural Architecture Search for Mobile</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11626&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MNASNet0_75_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MNASNet0_75_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.mnasnet.MNASNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MNASNet0_75_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MNASNet0_75_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_mnasnet</span><span class="p">(</span><span class="mf">0.75</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="mnasnet1_0"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet1_0.html#torchvision.models.mnasnet1_0">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MNASNet1_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mnasnet1_0</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MNASNet1_0_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MNASNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;MNASNet with depth multiplier of 1.0 from</span>
+<span class="sd">    `MnasNet: Platform-Aware Neural Architecture Search for Mobile</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11626&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MNASNet1_0_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MNASNet1_0_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.mnasnet.MNASNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MNASNet1_0_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MNASNet1_0_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_mnasnet</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="mnasnet1_3"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mnasnet1_3.html#torchvision.models.mnasnet1_3">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MNASNet1_3_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mnasnet1_3</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MNASNet1_3_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MNASNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;MNASNet with depth multiplier of 1.3 from</span>
+<span class="sd">    `MnasNet: Platform-Aware Neural Architecture Search for Mobile</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11626&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MNASNet1_3_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MNASNet1_3_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.mnasnet.MNASNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MNASNet1_3_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MNASNet1_3_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_mnasnet</span><span class="p">(</span><span class="mf">1.3</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/mobilenetv2.html b/0.16/_modules/torchvision/models/mobilenetv2.html
new file mode 100644
index 00000000000..1842262536a
--- /dev/null
+++ b/0.16/_modules/torchvision/models/mobilenetv2.html
@@ -0,0 +1,929 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.mobilenetv2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.mobilenetv2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.mobilenetv2</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span>
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_make_divisible</span><span class="p">,</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;MobileNetV2&quot;</span><span class="p">,</span> <span class="s2">&quot;MobileNet_V2_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;mobilenet_v2&quot;</span><span class="p">]</span>
+
+
+<span class="c1"># necessary for backwards compatibility</span>
+<span class="k">class</span> <span class="nc">InvertedResidual</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">inp</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">oup</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">expand_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
+        <span class="k">if</span> <span class="n">stride</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;stride should be 1 or 2 instead of </span><span class="si">{</span><span class="n">stride</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+        <span class="n">hidden_dim</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">inp</span> <span class="o">*</span> <span class="n">expand_ratio</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">inp</span> <span class="o">==</span> <span class="n">oup</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="n">expand_ratio</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="c1"># pw</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">inp</span><span class="p">,</span> <span class="n">hidden_dim</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU6</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+            <span class="p">[</span>
+                <span class="c1"># dw</span>
+                <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                    <span class="n">hidden_dim</span><span class="p">,</span>
+                    <span class="n">hidden_dim</span><span class="p">,</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+                    <span class="n">groups</span><span class="o">=</span><span class="n">hidden_dim</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU6</span><span class="p">,</span>
+                <span class="p">),</span>
+                <span class="c1"># pw-linear</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">oup</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+                <span class="n">norm_layer</span><span class="p">(</span><span class="n">oup</span><span class="p">),</span>
+            <span class="p">]</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">oup</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_is_cn</span> <span class="o">=</span> <span class="n">stride</span> <span class="o">&gt;</span> <span class="mi">1</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">MobileNetV2</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">width_mult</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">inverted_residual_setting</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">round_nearest</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">8</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.2</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        MobileNet V2 main class</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            num_classes (int): Number of classes</span>
+<span class="sd">            width_mult (float): Width multiplier - adjusts number of channels in each layer by this amount</span>
+<span class="sd">            inverted_residual_setting: Network structure</span>
+<span class="sd">            round_nearest (int): Round the number of channels in each layer to be a multiple of this number</span>
+<span class="sd">            Set to 1 to turn off rounding</span>
+<span class="sd">            block: Module specifying inverted residual building block for mobilenet</span>
+<span class="sd">            norm_layer: Module specifying the normalization layer to use</span>
+<span class="sd">            dropout (float): The droupout probability</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">InvertedResidual</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+
+        <span class="n">input_channel</span> <span class="o">=</span> <span class="mi">32</span>
+        <span class="n">last_channel</span> <span class="o">=</span> <span class="mi">1280</span>
+
+        <span class="k">if</span> <span class="n">inverted_residual_setting</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inverted_residual_setting</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="c1"># t, c, n, s</span>
+                <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+                <span class="p">[</span><span class="mi">6</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+                <span class="p">[</span><span class="mi">6</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+                <span class="p">[</span><span class="mi">6</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+                <span class="p">[</span><span class="mi">6</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+                <span class="p">[</span><span class="mi">6</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+                <span class="p">[</span><span class="mi">6</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">]</span>
+
+        <span class="c1"># only check the first element, assuming user knows t,c,n,s are required</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="nb">len</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">!=</span> <span class="mi">4</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;inverted_residual_setting should be non-empty or a 4-element list, got </span><span class="si">{</span><span class="n">inverted_residual_setting</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="c1"># building first layer</span>
+        <span class="n">input_channel</span> <span class="o">=</span> <span class="n">_make_divisible</span><span class="p">(</span><span class="n">input_channel</span> <span class="o">*</span> <span class="n">width_mult</span><span class="p">,</span> <span class="n">round_nearest</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">last_channel</span> <span class="o">=</span> <span class="n">_make_divisible</span><span class="p">(</span><span class="n">last_channel</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">),</span> <span class="n">round_nearest</span><span class="p">)</span>
+        <span class="n">features</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">input_channel</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU6</span><span class="p">)</span>
+        <span class="p">]</span>
+        <span class="c1"># building inverted residual blocks</span>
+        <span class="k">for</span> <span class="n">t</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">n</span><span class="p">,</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">inverted_residual_setting</span><span class="p">:</span>
+            <span class="n">output_channel</span> <span class="o">=</span> <span class="n">_make_divisible</span><span class="p">(</span><span class="n">c</span> <span class="o">*</span> <span class="n">width_mult</span><span class="p">,</span> <span class="n">round_nearest</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n</span><span class="p">):</span>
+                <span class="n">stride</span> <span class="o">=</span> <span class="n">s</span> <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="mi">1</span>
+                <span class="n">features</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block</span><span class="p">(</span><span class="n">input_channel</span><span class="p">,</span> <span class="n">output_channel</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">expand_ratio</span><span class="o">=</span><span class="n">t</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">))</span>
+                <span class="n">input_channel</span> <span class="o">=</span> <span class="n">output_channel</span>
+        <span class="c1"># building last several layers</span>
+        <span class="n">features</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">input_channel</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">last_channel</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU6</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+        <span class="c1"># make it nn.Sequential</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">features</span><span class="p">)</span>
+
+        <span class="c1"># building classifier</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">last_channel</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="c1"># weight initialization</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">GroupNorm</span><span class="p">)):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">ones_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mf">0.01</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward_impl</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># This exists since TorchScript doesn&#39;t support inheritance, so the superclass method</span>
+        <span class="c1"># (this one) needs to have a name other than `forward` that can be accessed in a subclass</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># Cannot use &quot;squeeze&quot; as batch-size can be 1</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">adaptive_avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">3504872</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="MobileNet_V2_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.quantization.mobilenet_v2.html#torchvision.models.MobileNet_V2_Weights">[docs]</a><span class="k">class</span> <span class="nc">MobileNet_V2_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mobilenet_v2-b0353104.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">71.878</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.286</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.301</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">13.555</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mobilenet_v2-7ebf99e0.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">72.154</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.822</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.301</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">13.598</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="mobilenet_v2"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mobilenet_v2.html#torchvision.models.mobilenet_v2">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MobileNet_V2_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mobilenet_v2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V2_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MobileNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;MobileNetV2 architecture from the `MobileNetV2: Inverted Residuals and Linear</span>
+<span class="sd">    Bottlenecks &lt;https://arxiv.org/abs/1801.04381&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MobileNet_V2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MobileNet_V2_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.mobilenetv2.MobileNetV2``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MobileNet_V2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MobileNet_V2_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">MobileNetV2</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/mobilenetv3.html b/0.16/_modules/torchvision/models/mobilenetv3.html
new file mode 100644
index 00000000000..aef5fec9a5a
--- /dev/null
+++ b/0.16/_modules/torchvision/models/mobilenetv3.html
@@ -0,0 +1,1092 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.mobilenetv3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.mobilenetv3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.mobilenetv3</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span><span class="p">,</span> <span class="n">SqueezeExcitation</span> <span class="k">as</span> <span class="n">SElayer</span>
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_make_divisible</span><span class="p">,</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;MobileNetV3&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MobileNet_V3_Large_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MobileNet_V3_Small_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mobilenet_v3_large&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mobilenet_v3_small&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">InvertedResidualConfig</span><span class="p">:</span>
+    <span class="c1"># Stores information listed at Tables 1 and 2 of the MobileNetV3 paper</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">input_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">kernel</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">expanded_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">use_se</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">activation</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">width_mult</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">kernel</span> <span class="o">=</span> <span class="n">kernel</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">expanded_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">(</span><span class="n">expanded_channels</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_se</span> <span class="o">=</span> <span class="n">use_se</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_hs</span> <span class="o">=</span> <span class="n">activation</span> <span class="o">==</span> <span class="s2">&quot;HS&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dilation</span> <span class="o">=</span> <span class="n">dilation</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">adjust_channels</span><span class="p">(</span><span class="n">channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">:</span> <span class="nb">float</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">_make_divisible</span><span class="p">(</span><span class="n">channels</span> <span class="o">*</span> <span class="n">width_mult</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">InvertedResidual</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="c1"># Implemented as described at section 5 of MobileNetV3 paper</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">cnf</span><span class="p">:</span> <span class="n">InvertedResidualConfig</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">se_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">SElayer</span><span class="p">,</span> <span class="n">scale_activation</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">Hardsigmoid</span><span class="p">),</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">&lt;=</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;illegal stride value&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">==</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">activation_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Hardswish</span> <span class="k">if</span> <span class="n">cnf</span><span class="o">.</span><span class="n">use_hs</span> <span class="k">else</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span>
+
+        <span class="c1"># expand</span>
+        <span class="k">if</span> <span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span> <span class="o">!=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                    <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span>
+                    <span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span><span class="p">,</span>
+                    <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="c1"># depthwise</span>
+        <span class="n">stride</span> <span class="o">=</span> <span class="mi">1</span> <span class="k">if</span> <span class="n">cnf</span><span class="o">.</span><span class="n">dilation</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span><span class="p">,</span>
+                <span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">kernel</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+                <span class="n">dilation</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">dilation</span><span class="p">,</span>
+                <span class="n">groups</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">cnf</span><span class="o">.</span><span class="n">use_se</span><span class="p">:</span>
+            <span class="n">squeeze_channels</span> <span class="o">=</span> <span class="n">_make_divisible</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span> <span class="o">//</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">se_layer</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span><span class="p">,</span> <span class="n">squeeze_channels</span><span class="p">))</span>
+
+        <span class="c1"># project</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">cnf</span><span class="o">.</span><span class="n">expanded_channels</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">block</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">out_channels</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_is_cn</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride</span> <span class="o">&gt;</span> <span class="mi">1</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">block</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span><span class="p">:</span>
+            <span class="n">result</span> <span class="o">+=</span> <span class="nb">input</span>
+        <span class="k">return</span> <span class="n">result</span>
+
+
+<span class="k">class</span> <span class="nc">MobileNetV3</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inverted_residual_setting</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">InvertedResidualConfig</span><span class="p">],</span>
+        <span class="n">last_channel</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.2</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        MobileNet V3 main class</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            inverted_residual_setting (List[InvertedResidualConfig]): Network structure</span>
+<span class="sd">            last_channel (int): The number of channels on the penultimate layer</span>
+<span class="sd">            num_classes (int): Number of classes</span>
+<span class="sd">            block (Optional[Callable[..., nn.Module]]): Module specifying inverted residual building block for mobilenet</span>
+<span class="sd">            norm_layer (Optional[Callable[..., nn.Module]]): Module specifying the normalization layer to use</span>
+<span class="sd">            dropout (float): The droupout probability</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">inverted_residual_setting</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The inverted_residual_setting should not be empty&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="p">(</span>
+            <span class="nb">isinstance</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span>
+            <span class="ow">and</span> <span class="nb">all</span><span class="p">([</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">s</span><span class="p">,</span> <span class="n">InvertedResidualConfig</span><span class="p">)</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">inverted_residual_setting</span><span class="p">])</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;The inverted_residual_setting should be List[InvertedResidualConfig]&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">InvertedResidual</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="c1"># building first layer</span>
+        <span class="n">firstconv_output_channels</span> <span class="o">=</span> <span class="n">inverted_residual_setting</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">input_channels</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="mi">3</span><span class="p">,</span>
+                <span class="n">firstconv_output_channels</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">Hardswish</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="c1"># building inverted residual blocks</span>
+        <span class="k">for</span> <span class="n">cnf</span> <span class="ow">in</span> <span class="n">inverted_residual_setting</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block</span><span class="p">(</span><span class="n">cnf</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">))</span>
+
+        <span class="c1"># building last several layers</span>
+        <span class="n">lastconv_input_channels</span> <span class="o">=</span> <span class="n">inverted_residual_setting</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">out_channels</span>
+        <span class="n">lastconv_output_channels</span> <span class="o">=</span> <span class="mi">6</span> <span class="o">*</span> <span class="n">lastconv_input_channels</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">lastconv_input_channels</span><span class="p">,</span>
+                <span class="n">lastconv_output_channels</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">Hardswish</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">lastconv_output_channels</span><span class="p">,</span> <span class="n">last_channel</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Hardswish</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">last_channel</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">GroupNorm</span><span class="p">)):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">ones_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mf">0.01</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward_impl</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_mobilenet_v3_conf</span><span class="p">(</span>
+    <span class="n">arch</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">width_mult</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span> <span class="n">reduced_tail</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="n">dilated</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">):</span>
+    <span class="n">reduce_divider</span> <span class="o">=</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">reduced_tail</span> <span class="k">else</span> <span class="mi">1</span>
+    <span class="n">dilation</span> <span class="o">=</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">dilated</span> <span class="k">else</span> <span class="mi">1</span>
+
+    <span class="n">bneck_conf</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">InvertedResidualConfig</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="n">width_mult</span><span class="p">)</span>
+    <span class="n">adjust_channels</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">InvertedResidualConfig</span><span class="o">.</span><span class="n">adjust_channels</span><span class="p">,</span> <span class="n">width_mult</span><span class="o">=</span><span class="n">width_mult</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">arch</span> <span class="o">==</span> <span class="s2">&quot;mobilenet_v3_large&quot;</span><span class="p">:</span>
+        <span class="n">inverted_residual_setting</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">16</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">16</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>  <span class="c1"># C1</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">24</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">72</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">24</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">72</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>  <span class="c1"># C2</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">40</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">120</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">40</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">120</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">40</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">240</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>  <span class="c1"># C3</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">80</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">200</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">80</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">184</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">80</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">184</span><span class="p">,</span> <span class="mi">80</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">80</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">480</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">112</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">672</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">112</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">672</span><span class="p">,</span> <span class="mi">160</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="n">dilation</span><span class="p">),</span>  <span class="c1"># C4</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">160</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">960</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">160</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">dilation</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">160</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">960</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">160</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">dilation</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">last_channel</span> <span class="o">=</span> <span class="n">adjust_channels</span><span class="p">(</span><span class="mi">1280</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">)</span>  <span class="c1"># C5</span>
+    <span class="k">elif</span> <span class="n">arch</span> <span class="o">==</span> <span class="s2">&quot;mobilenet_v3_small&quot;</span><span class="p">:</span>
+        <span class="n">inverted_residual_setting</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">16</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>  <span class="c1"># C1</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">16</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">72</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>  <span class="c1"># C2</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">24</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">88</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;RE&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">24</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>  <span class="c1"># C3</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">40</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">240</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">40</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">240</span><span class="p">,</span> <span class="mi">40</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">40</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">120</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">48</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">144</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">48</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">288</span><span class="p">,</span> <span class="mi">96</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="n">dilation</span><span class="p">),</span>  <span class="c1"># C4</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">96</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">576</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">96</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">dilation</span><span class="p">),</span>
+            <span class="n">bneck_conf</span><span class="p">(</span><span class="mi">96</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">576</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="mi">96</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="s2">&quot;HS&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">dilation</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">last_channel</span> <span class="o">=</span> <span class="n">adjust_channels</span><span class="p">(</span><span class="mi">1024</span> <span class="o">//</span> <span class="n">reduce_divider</span><span class="p">)</span>  <span class="c1"># C5</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unsupported model type </span><span class="si">{</span><span class="n">arch</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span>
+
+
+<span class="k">def</span> <span class="nf">_mobilenet_v3</span><span class="p">(</span>
+    <span class="n">inverted_residual_setting</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">InvertedResidualConfig</span><span class="p">],</span>
+    <span class="n">last_channel</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MobileNetV3</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">MobileNetV3</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="MobileNet_V3_Large_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.quantization.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights">[docs]</a><span class="k">class</span> <span class="nc">MobileNet_V3_Large_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mobilenet_v3_large-8738ca79.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5483032</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">74.042</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.340</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.217</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">21.114</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were trained from scratch by using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mobilenet_v3_large-5c1a4163.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5483032</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">75.274</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.566</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.217</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">21.107</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve marginally upon the results of the original paper by using a modified version of</span>
+<span class="s2">                TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="MobileNet_V3_Small_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mobilenet_v3_small.html#torchvision.models.MobileNet_V3_Small_Weights">[docs]</a><span class="k">class</span> <span class="nc">MobileNet_V3_Small_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mobilenet_v3_small-047dcff4.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">2542856</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">67.668</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">87.402</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.057</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">9.829</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a simple training recipe.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="mobilenet_v3_large"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mobilenet_v3_large.html#torchvision.models.mobilenet_v3_large">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mobilenet_v3_large</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MobileNetV3</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a large MobileNetV3 architecture from</span>
+<span class="sd">    `Searching for MobileNetV3 &lt;https://arxiv.org/abs/1905.02244&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MobileNet_V3_Large_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.mobilenet.MobileNetV3``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MobileNet_V3_Large_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_mobilenet_v3_conf</span><span class="p">(</span><span class="s2">&quot;mobilenet_v3_large&quot;</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_mobilenet_v3</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="mobilenet_v3_small"><a class="viewcode-back" href="../../../models/generated/torchvision.models.mobilenet_v3_small.html#torchvision.models.mobilenet_v3_small">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MobileNet_V3_Small_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mobilenet_v3_small</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Small_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MobileNetV3</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a small MobileNetV3 architecture from</span>
+<span class="sd">    `Searching for MobileNetV3 &lt;https://arxiv.org/abs/1905.02244&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.MobileNet_V3_Small_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.MobileNet_V3_Small_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.mobilenet.MobileNetV3``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.MobileNet_V3_Small_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MobileNet_V3_Small_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_mobilenet_v3_conf</span><span class="p">(</span><span class="s2">&quot;mobilenet_v3_small&quot;</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_mobilenet_v3</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/optical_flow/raft.html b/0.16/_modules/torchvision/models/optical_flow/raft.html
new file mode 100644
index 00000000000..7ef1211b4b1
--- /dev/null
+++ b/0.16/_modules/torchvision/models/optical_flow/raft.html
@@ -0,0 +1,1616 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.optical_flow.raft &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.optical_flow.raft</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.optical_flow.raft</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.nn.modules.batchnorm</span> <span class="kn">import</span> <span class="n">BatchNorm2d</span>
+<span class="kn">from</span> <span class="nn">torch.nn.modules.instancenorm</span> <span class="kn">import</span> <span class="n">InstanceNorm2d</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">OpticalFlow</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">grid_sample</span><span class="p">,</span> <span class="n">make_coords_grid</span><span class="p">,</span> <span class="n">upsample_flow</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">(</span>
+    <span class="s2">&quot;RAFT&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;raft_large&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;raft_small&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Raft_Large_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Raft_Small_Weights&quot;</span><span class="p">,</span>
+<span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ResidualBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Slightly modified Residual block with extra relu and biases.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">always_project</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="c1"># Note regarding bias=True:</span>
+        <span class="c1"># Usually we can pass bias=False in conv layers followed by a norm layer.</span>
+        <span class="c1"># But in the RAFT training reference, the BatchNorm2d layers are only activated for the first dataset,</span>
+        <span class="c1"># and frozen for the rest of the training process (i.e. set as eval()). The bias term is thus still useful</span>
+        <span class="c1"># for the rest of the datasets. Technically, we could remove the bias for other norm layers like Instance norm</span>
+        <span class="c1"># because these aren&#39;t frozen, but we don&#39;t bother (also, we wouldn&#39;t be able to load the original weights).</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu1</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu2</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">out_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+
+        <span class="c1"># make mypy happy</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span>
+
+        <span class="k">if</span> <span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">always_project</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">in_channels</span><span class="p">,</span>
+                <span class="n">out_channels</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="n">x</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu1</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu2</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span> <span class="o">+</span> <span class="n">y</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">BottleneckBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Slightly modified BottleNeck block (extra relu and biases)&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="c1"># See note in ResidualBlock for the reason behind bias=True</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu1</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span> <span class="o">//</span> <span class="mi">4</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu2</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">out_channels</span> <span class="o">//</span> <span class="mi">4</span><span class="p">,</span> <span class="n">out_channels</span> <span class="o">//</span> <span class="mi">4</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu3</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">out_channels</span> <span class="o">//</span> <span class="mi">4</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">in_channels</span><span class="p">,</span>
+                <span class="n">out_channels</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="n">x</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu1</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu2</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu3</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span> <span class="o">+</span> <span class="n">y</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">FeatureEncoder</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The feature encoder, used both as the actual feature encoder, and as the context encoder.</span>
+
+<span class="sd">    It must downsample its input by 8.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">block</span><span class="o">=</span><span class="n">ResidualBlock</span><span class="p">,</span> <span class="n">layers</span><span class="o">=</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span> <span class="n">strides</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">layers</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">5</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The expected number of layers is 5, instead got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">layers</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="c1"># See note in ResidualBlock for the reason behind bias=True</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="mi">3</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">strides</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_2_blocks</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">layers</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">first_stride</span><span class="o">=</span><span class="n">strides</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_2_blocks</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">layers</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">first_stride</span><span class="o">=</span><span class="n">strides</span><span class="p">[</span><span class="mi">2</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_2_blocks</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">layers</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">first_stride</span><span class="o">=</span><span class="n">strides</span><span class="p">[</span><span class="mi">3</span><span class="p">])</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">layers</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">layers</span><span class="p">[</span><span class="mi">4</span><span class="p">],</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">InstanceNorm2d</span><span class="p">)):</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">weight</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="n">num_downsamples</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="nb">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">s</span><span class="p">:</span> <span class="n">s</span> <span class="o">==</span> <span class="mi">2</span><span class="p">,</span> <span class="n">strides</span><span class="p">)))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">output_dim</span> <span class="o">=</span> <span class="n">layers</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">downsample_factor</span> <span class="o">=</span> <span class="mi">2</span><span class="o">**</span><span class="n">num_downsamples</span>
+
+    <span class="k">def</span> <span class="nf">_make_2_blocks</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">block</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">,</span> <span class="n">first_stride</span><span class="p">):</span>
+        <span class="n">block1</span> <span class="o">=</span> <span class="n">block</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">first_stride</span><span class="p">)</span>
+        <span class="n">block2</span> <span class="o">=</span> <span class="n">block</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">block1</span><span class="p">,</span> <span class="n">block2</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convnormrelu</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">MotionEncoder</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The motion encoder, part of the update block.</span>
+
+<span class="sd">    Takes the current predicted flow and the correlation features as input and returns an encoded version of these.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">in_channels_corr</span><span class="p">,</span> <span class="n">corr_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">192</span><span class="p">),</span> <span class="n">flow_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">64</span><span class="p">),</span> <span class="n">out_channels</span><span class="o">=</span><span class="mi">128</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">flow_layers</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The expected number of flow_layers is 2, instead got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">flow_layers</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">corr_layers</span><span class="p">)</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The number of corr_layers should be 1 or 2, instead got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">corr_layers</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">convcorr1</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">in_channels_corr</span><span class="p">,</span> <span class="n">corr_layers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">corr_layers</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">convcorr2</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">corr_layers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">corr_layers</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">convcorr2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">convflow1</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="n">flow_layers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">7</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convflow2</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">flow_layers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">flow_layers</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+
+        <span class="c1"># out_channels - 2 because we cat the flow (2 channels) at the end</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">corr_layers</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">flow_layers</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">out_channels</span> <span class="o">-</span> <span class="mi">2</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">out_channels</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flow</span><span class="p">,</span> <span class="n">corr_features</span><span class="p">):</span>
+        <span class="n">corr</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convcorr1</span><span class="p">(</span><span class="n">corr_features</span><span class="p">)</span>
+        <span class="n">corr</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convcorr2</span><span class="p">(</span><span class="n">corr</span><span class="p">)</span>
+
+        <span class="n">flow_orig</span> <span class="o">=</span> <span class="n">flow</span>
+        <span class="n">flow</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convflow1</span><span class="p">(</span><span class="n">flow</span><span class="p">)</span>
+        <span class="n">flow</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convflow2</span><span class="p">(</span><span class="n">flow</span><span class="p">)</span>
+
+        <span class="n">corr_flow</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">corr</span><span class="p">,</span> <span class="n">flow</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">corr_flow</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">corr_flow</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">corr_flow</span><span class="p">,</span> <span class="n">flow_orig</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ConvGRU</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convolutional Gru unit.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">input_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">padding</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convz</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">hidden_size</span> <span class="o">+</span> <span class="n">input_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convr</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">hidden_size</span> <span class="o">+</span> <span class="n">input_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convq</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">hidden_size</span> <span class="o">+</span> <span class="n">input_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">hx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">h</span><span class="p">,</span> <span class="n">x</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">z</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">convz</span><span class="p">(</span><span class="n">hx</span><span class="p">))</span>
+        <span class="n">r</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">convr</span><span class="p">(</span><span class="n">hx</span><span class="p">))</span>
+        <span class="n">q</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tanh</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">convq</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">r</span> <span class="o">*</span> <span class="n">h</span><span class="p">,</span> <span class="n">x</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)))</span>
+        <span class="n">h</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">z</span><span class="p">)</span> <span class="o">*</span> <span class="n">h</span> <span class="o">+</span> <span class="n">z</span> <span class="o">*</span> <span class="n">q</span>
+        <span class="k">return</span> <span class="n">h</span>
+
+
+<span class="k">def</span> <span class="nf">_pass_through_h</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">_</span><span class="p">):</span>
+    <span class="c1"># Declared here for torchscript</span>
+    <span class="k">return</span> <span class="n">h</span>
+
+
+<span class="k">class</span> <span class="nc">RecurrentBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Recurrent block, part of the update block.</span>
+
+<span class="sd">    Takes the current hidden state and the concatenation of (motion encoder output, context) as input.</span>
+<span class="sd">    Returns an updated hidden state.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">input_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="mi">5</span><span class="p">,</span> <span class="mi">1</span><span class="p">)),</span> <span class="n">padding</span><span class="o">=</span><span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">))):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;kernel_size should have the same length as padding, instead got len(kernel_size) = </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span><span class="si">}</span><span class="s2"> and len(padding) = </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;kernel_size should either 1 or 2, instead got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">convgru1</span> <span class="o">=</span> <span class="n">ConvGRU</span><span class="p">(</span>
+            <span class="n">input_size</span><span class="o">=</span><span class="n">input_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="o">=</span><span class="n">hidden_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">convgru2</span> <span class="o">=</span> <span class="n">ConvGRU</span><span class="p">(</span>
+                <span class="n">input_size</span><span class="o">=</span><span class="n">input_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="o">=</span><span class="n">hidden_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">convgru2</span> <span class="o">=</span> <span class="n">_pass_through_h</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">hidden_size</span> <span class="o">=</span> <span class="n">hidden_size</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convgru1</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">x</span><span class="p">)</span>
+        <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convgru2</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">h</span>
+
+
+<span class="k">class</span> <span class="nc">FlowHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Flow head, part of the update block.</span>
+
+<span class="sd">    Takes the hidden state of the recurrent unit as input, and outputs the predicted &quot;delta flow&quot;.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">hidden_size</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)))</span>
+
+
+<span class="k">class</span> <span class="nc">UpdateBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The update block which contains the motion encoder, the recurrent block, and the flow head.</span>
+
+<span class="sd">    It must expose a ``hidden_state_size`` attribute which is the hidden state size of its recurrent block.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">motion_encoder</span><span class="p">,</span> <span class="n">recurrent_block</span><span class="p">,</span> <span class="n">flow_head</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">motion_encoder</span> <span class="o">=</span> <span class="n">motion_encoder</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">recurrent_block</span> <span class="o">=</span> <span class="n">recurrent_block</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">flow_head</span> <span class="o">=</span> <span class="n">flow_head</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">hidden_state_size</span> <span class="o">=</span> <span class="n">recurrent_block</span><span class="o">.</span><span class="n">hidden_size</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">hidden_state</span><span class="p">,</span> <span class="n">context</span><span class="p">,</span> <span class="n">corr_features</span><span class="p">,</span> <span class="n">flow</span><span class="p">):</span>
+        <span class="n">motion_features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">motion_encoder</span><span class="p">(</span><span class="n">flow</span><span class="p">,</span> <span class="n">corr_features</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">context</span><span class="p">,</span> <span class="n">motion_features</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">hidden_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">recurrent_block</span><span class="p">(</span><span class="n">hidden_state</span><span class="p">,</span> <span class="n">x</span><span class="p">)</span>
+        <span class="n">delta_flow</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">flow_head</span><span class="p">(</span><span class="n">hidden_state</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">hidden_state</span><span class="p">,</span> <span class="n">delta_flow</span>
+
+
+<span class="k">class</span> <span class="nc">MaskPredictor</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Mask predictor to be used when upsampling the predicted flow.</span>
+
+<span class="sd">    It takes the hidden state of the recurrent unit as input and outputs the mask.</span>
+<span class="sd">    This is not used in the raft-small model.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">multiplier</span><span class="o">=</span><span class="mf">0.25</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">convrelu</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+        <span class="c1"># 8 * 8 * 9 because the predicted flow is downsampled by 8, from the downsampling of the initial FeatureEncoder,</span>
+        <span class="c1"># and we interpolate with all 9 surrounding neighbors. See paper and appendix B.</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">hidden_size</span><span class="p">,</span> <span class="mi">8</span> <span class="o">*</span> <span class="mi">8</span> <span class="o">*</span> <span class="mi">9</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+        <span class="c1"># In the original code, they use a factor of 0.25 to &quot;downweight the gradients&quot; of that branch.</span>
+        <span class="c1"># See e.g. https://github.com/princeton-vl/RAFT/issues/119#issuecomment-953950419</span>
+        <span class="c1"># or https://github.com/princeton-vl/RAFT/issues/24.</span>
+        <span class="c1"># It doesn&#39;t seem to affect epe significantly and can likely be set to 1.</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">multiplier</span> <span class="o">=</span> <span class="n">multiplier</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convrelu</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">multiplier</span> <span class="o">*</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">CorrBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The correlation block.</span>
+
+<span class="sd">    Creates a correlation pyramid with ``num_levels`` levels from the outputs of the feature encoder,</span>
+<span class="sd">    and then indexes from this pyramid to create correlation features.</span>
+<span class="sd">    The &quot;indexing&quot; of a given centroid pixel x&#39; is done by concatenating its surrounding neighbors that</span>
+<span class="sd">    are within a ``radius``, according to the infinity norm (see paper section 3.2).</span>
+<span class="sd">    Note: typo in the paper, it should be infinity norm, not 1-norm.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">num_levels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span> <span class="n">radius</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_levels</span> <span class="o">=</span> <span class="n">num_levels</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">radius</span> <span class="o">=</span> <span class="n">radius</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">corr_pyramid</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mi">0</span><span class="p">)]</span>  <span class="c1"># useless, but torchscript is otherwise confused :&#39;)</span>
+
+        <span class="c1"># The neighborhood of a centroid pixel x&#39; is {x&#39; + delta, ||delta||_inf &lt;= radius}</span>
+        <span class="c1"># so it&#39;s a square surrounding x&#39;, and its sides have a length of 2 * radius + 1</span>
+        <span class="c1"># The paper claims that it&#39;s ||.||_1 instead of ||.||_inf but it&#39;s a typo:</span>
+        <span class="c1"># https://github.com/princeton-vl/RAFT/issues/122</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">num_levels</span> <span class="o">*</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="n">radius</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="nf">build_pyramid</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">fmap1</span><span class="p">,</span> <span class="n">fmap2</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Build the correlation pyramid from two feature maps.</span>
+
+<span class="sd">        The correlation volume is first computed as the dot product of each pair (pixel_in_fmap1, pixel_in_fmap2)</span>
+<span class="sd">        The last 2 dimensions of the correlation volume are then pooled num_levels times at different resolutions</span>
+<span class="sd">        to build the correlation pyramid.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="k">if</span> <span class="n">fmap1</span><span class="o">.</span><span class="n">shape</span> <span class="o">!=</span> <span class="n">fmap2</span><span class="o">.</span><span class="n">shape</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Input feature maps should have the same shape, instead got </span><span class="si">{</span><span class="n">fmap1</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2"> (fmap1.shape) != </span><span class="si">{</span><span class="n">fmap2</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2"> (fmap2.shape)&quot;</span>
+            <span class="p">)</span>
+
+        <span class="c1"># Explaining min_fmap_size below: the fmaps are down-sampled (num_levels - 1) times by a factor of 2.</span>
+        <span class="c1"># The last corr_volume most have at least 2 values (hence the 2* factor), otherwise grid_sample() would</span>
+        <span class="c1"># produce nans in its output.</span>
+        <span class="n">min_fmap_size</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">*</span> <span class="p">(</span><span class="mi">2</span> <span class="o">**</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_levels</span> <span class="o">-</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="k">if</span> <span class="nb">any</span><span class="p">(</span><span class="n">fmap_size</span> <span class="o">&lt;</span> <span class="n">min_fmap_size</span> <span class="k">for</span> <span class="n">fmap_size</span> <span class="ow">in</span> <span class="n">fmap1</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Feature maps are too small to be down-sampled by the correlation pyramid. &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;H and W of feature maps should be at least </span><span class="si">{</span><span class="n">min_fmap_size</span><span class="si">}</span><span class="s2">; got: </span><span class="si">{</span><span class="n">fmap1</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span><span class="si">}</span><span class="s2">. &quot;</span>
+                <span class="s2">&quot;Remember that input images to the model are downsampled by 8, so that means their &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;dimensions should be at least 8 * </span><span class="si">{</span><span class="n">min_fmap_size</span><span class="si">}</span><span class="s2"> = </span><span class="si">{</span><span class="mi">8</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="n">min_fmap_size</span><span class="si">}</span><span class="s2">.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">corr_volume</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_compute_corr_volume</span><span class="p">(</span><span class="n">fmap1</span><span class="p">,</span> <span class="n">fmap2</span><span class="p">)</span>
+
+        <span class="n">batch_size</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">corr_volume</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># _, _ = h, w</span>
+        <span class="n">corr_volume</span> <span class="o">=</span> <span class="n">corr_volume</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">batch_size</span> <span class="o">*</span> <span class="n">h</span> <span class="o">*</span> <span class="n">w</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">corr_pyramid</span> <span class="o">=</span> <span class="p">[</span><span class="n">corr_volume</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_levels</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
+            <span class="n">corr_volume</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">avg_pool2d</span><span class="p">(</span><span class="n">corr_volume</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">corr_pyramid</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">corr_volume</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">index_pyramid</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">centroids_coords</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Return correlation features by indexing from the pyramid.&quot;&quot;&quot;</span>
+        <span class="n">neighborhood_side_len</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">radius</span> <span class="o">+</span> <span class="mi">1</span>  <span class="c1"># see note in __init__ about out_channels</span>
+        <span class="n">di</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="o">-</span><span class="bp">self</span><span class="o">.</span><span class="n">radius</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">radius</span><span class="p">,</span> <span class="n">neighborhood_side_len</span><span class="p">)</span>
+        <span class="n">dj</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="o">-</span><span class="bp">self</span><span class="o">.</span><span class="n">radius</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">radius</span><span class="p">,</span> <span class="n">neighborhood_side_len</span><span class="p">)</span>
+        <span class="n">delta</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">meshgrid</span><span class="p">(</span><span class="n">di</span><span class="p">,</span> <span class="n">dj</span><span class="p">,</span> <span class="n">indexing</span><span class="o">=</span><span class="s2">&quot;ij&quot;</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">centroids_coords</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="n">delta</span> <span class="o">=</span> <span class="n">delta</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">neighborhood_side_len</span><span class="p">,</span> <span class="n">neighborhood_side_len</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="n">batch_size</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">centroids_coords</span><span class="o">.</span><span class="n">shape</span>  <span class="c1"># _ = 2</span>
+        <span class="n">centroids_coords</span> <span class="o">=</span> <span class="n">centroids_coords</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">batch_size</span> <span class="o">*</span> <span class="n">h</span> <span class="o">*</span> <span class="n">w</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="n">indexed_pyramid</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">corr_volume</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">corr_pyramid</span><span class="p">:</span>
+            <span class="n">sampling_coords</span> <span class="o">=</span> <span class="n">centroids_coords</span> <span class="o">+</span> <span class="n">delta</span>  <span class="c1"># end shape is (batch_size * h * w, side_len, side_len, 2)</span>
+            <span class="n">indexed_corr_volume</span> <span class="o">=</span> <span class="n">grid_sample</span><span class="p">(</span><span class="n">corr_volume</span><span class="p">,</span> <span class="n">sampling_coords</span><span class="p">,</span> <span class="n">align_corners</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;bilinear&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+                <span class="n">batch_size</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span>
+            <span class="p">)</span>
+            <span class="n">indexed_pyramid</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">indexed_corr_volume</span><span class="p">)</span>
+            <span class="n">centroids_coords</span> <span class="o">=</span> <span class="n">centroids_coords</span> <span class="o">/</span> <span class="mi">2</span>
+
+        <span class="n">corr_features</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">indexed_pyramid</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+
+        <span class="n">expected_output_shape</span> <span class="o">=</span> <span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">corr_features</span><span class="o">.</span><span class="n">shape</span> <span class="o">!=</span> <span class="n">expected_output_shape</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Output shape of index pyramid is incorrect. Should be </span><span class="si">{</span><span class="n">expected_output_shape</span><span class="si">}</span><span class="s2">, got </span><span class="si">{</span><span class="n">corr_features</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">corr_features</span>
+
+    <span class="k">def</span> <span class="nf">_compute_corr_volume</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">fmap1</span><span class="p">,</span> <span class="n">fmap2</span><span class="p">):</span>
+        <span class="n">batch_size</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">fmap1</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">fmap1</span> <span class="o">=</span> <span class="n">fmap1</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">h</span> <span class="o">*</span> <span class="n">w</span><span class="p">)</span>
+        <span class="n">fmap2</span> <span class="o">=</span> <span class="n">fmap2</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">h</span> <span class="o">*</span> <span class="n">w</span><span class="p">)</span>
+
+        <span class="n">corr</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">fmap1</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">fmap2</span><span class="p">)</span>
+        <span class="n">corr</span> <span class="o">=</span> <span class="n">corr</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">corr</span> <span class="o">/</span> <span class="n">torch</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">num_channels</span><span class="p">))</span>
+
+
+<span class="k">class</span> <span class="nc">RAFT</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">feature_encoder</span><span class="p">,</span> <span class="n">context_encoder</span><span class="p">,</span> <span class="n">corr_block</span><span class="p">,</span> <span class="n">update_block</span><span class="p">,</span> <span class="n">mask_predictor</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;RAFT model from</span>
+<span class="sd">        `RAFT: Recurrent All Pairs Field Transforms for Optical Flow &lt;https://arxiv.org/abs/2003.12039&gt;`_.</span>
+
+<span class="sd">        args:</span>
+<span class="sd">            feature_encoder (nn.Module): The feature encoder. It must downsample the input by 8.</span>
+<span class="sd">                Its input is the concatenation of ``image1`` and ``image2``.</span>
+<span class="sd">            context_encoder (nn.Module): The context encoder. It must downsample the input by 8.</span>
+<span class="sd">                Its input is ``image1``. As in the original implementation, its output will be split into 2 parts:</span>
+
+<span class="sd">                - one part will be used as the actual &quot;context&quot;, passed to the recurrent unit of the ``update_block``</span>
+<span class="sd">                - one part will be used to initialize the hidden state of the recurrent unit of</span>
+<span class="sd">                  the ``update_block``</span>
+
+<span class="sd">                These 2 parts are split according to the ``hidden_state_size`` of the ``update_block``, so the output</span>
+<span class="sd">                of the ``context_encoder`` must be strictly greater than ``hidden_state_size``.</span>
+
+<span class="sd">            corr_block (nn.Module): The correlation block, which creates a correlation pyramid from the output of the</span>
+<span class="sd">                ``feature_encoder``, and then indexes from this pyramid to create correlation features. It must expose</span>
+<span class="sd">                2 methods:</span>
+
+<span class="sd">                - a ``build_pyramid`` method that takes ``feature_map_1`` and ``feature_map_2`` as input (these are the</span>
+<span class="sd">                  output of the ``feature_encoder``).</span>
+<span class="sd">                - a ``index_pyramid`` method that takes the coordinates of the centroid pixels as input, and returns</span>
+<span class="sd">                  the correlation features. See paper section 3.2.</span>
+
+<span class="sd">                It must expose an ``out_channels`` attribute.</span>
+
+<span class="sd">            update_block (nn.Module): The update block, which contains the motion encoder, the recurrent unit, and the</span>
+<span class="sd">                flow head. It takes as input the hidden state of its recurrent unit, the context, the correlation</span>
+<span class="sd">                features, and the current predicted flow. It outputs an updated hidden state, and the ``delta_flow``</span>
+<span class="sd">                prediction (see paper appendix A). It must expose a ``hidden_state_size`` attribute.</span>
+<span class="sd">            mask_predictor (nn.Module, optional): Predicts the mask that will be used to upsample the predicted flow.</span>
+<span class="sd">                The output channel must be 8 * 8 * 9 - see paper section 3.3, and Appendix B.</span>
+<span class="sd">                If ``None`` (default), the flow is upsampled using interpolation.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">feature_encoder</span> <span class="o">=</span> <span class="n">feature_encoder</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">context_encoder</span> <span class="o">=</span> <span class="n">context_encoder</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">corr_block</span> <span class="o">=</span> <span class="n">corr_block</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">update_block</span> <span class="o">=</span> <span class="n">update_block</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">mask_predictor</span> <span class="o">=</span> <span class="n">mask_predictor</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">update_block</span><span class="p">,</span> <span class="s2">&quot;hidden_state_size&quot;</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The update_block parameter should expose a &#39;hidden_state_size&#39; attribute.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">image1</span><span class="p">,</span> <span class="n">image2</span><span class="p">,</span> <span class="n">num_flow_updates</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">12</span><span class="p">):</span>
+
+        <span class="n">batch_size</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">image1</span><span class="o">.</span><span class="n">shape</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span> <span class="o">!=</span> <span class="n">image2</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;input images should have the same shape, instead got (</span><span class="si">{</span><span class="n">h</span><span class="si">}</span><span class="s2">, </span><span class="si">{</span><span class="n">w</span><span class="si">}</span><span class="s2">) != </span><span class="si">{</span><span class="n">image2</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">h</span> <span class="o">%</span> <span class="mi">8</span> <span class="o">==</span> <span class="mi">0</span><span class="p">)</span> <span class="ow">and</span> <span class="p">(</span><span class="n">w</span> <span class="o">%</span> <span class="mi">8</span> <span class="o">==</span> <span class="mi">0</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;input image H and W should be divisible by 8, instead got </span><span class="si">{</span><span class="n">h</span><span class="si">}</span><span class="s2"> (h) and </span><span class="si">{</span><span class="n">w</span><span class="si">}</span><span class="s2"> (w)&quot;</span><span class="p">)</span>
+
+        <span class="n">fmaps</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">feature_encoder</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">image1</span><span class="p">,</span> <span class="n">image2</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">))</span>
+        <span class="n">fmap1</span><span class="p">,</span> <span class="n">fmap2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">chunk</span><span class="p">(</span><span class="n">fmaps</span><span class="p">,</span> <span class="n">chunks</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">fmap1</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span> <span class="o">!=</span> <span class="p">(</span><span class="n">h</span> <span class="o">//</span> <span class="mi">8</span><span class="p">,</span> <span class="n">w</span> <span class="o">//</span> <span class="mi">8</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The feature encoder should downsample H and W by 8&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">corr_block</span><span class="o">.</span><span class="n">build_pyramid</span><span class="p">(</span><span class="n">fmap1</span><span class="p">,</span> <span class="n">fmap2</span><span class="p">)</span>
+
+        <span class="n">context_out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">context_encoder</span><span class="p">(</span><span class="n">image1</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">context_out</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span> <span class="o">!=</span> <span class="p">(</span><span class="n">h</span> <span class="o">//</span> <span class="mi">8</span><span class="p">,</span> <span class="n">w</span> <span class="o">//</span> <span class="mi">8</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The context encoder should downsample H and W by 8&quot;</span><span class="p">)</span>
+
+        <span class="c1"># As in the original paper, the actual output of the context encoder is split in 2 parts:</span>
+        <span class="c1"># - one part is used to initialize the hidden state of the recurent units of the update block</span>
+        <span class="c1"># - the rest is the &quot;actual&quot; context.</span>
+        <span class="n">hidden_state_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">update_block</span><span class="o">.</span><span class="n">hidden_state_size</span>
+        <span class="n">out_channels_context</span> <span class="o">=</span> <span class="n">context_out</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">hidden_state_size</span>
+        <span class="k">if</span> <span class="n">out_channels_context</span> <span class="o">&lt;=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The context encoder outputs </span><span class="si">{</span><span class="n">context_out</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2"> channels, but it should have at strictly more than hidden_state=</span><span class="si">{</span><span class="n">hidden_state_size</span><span class="si">}</span><span class="s2"> channels&quot;</span>
+            <span class="p">)</span>
+        <span class="n">hidden_state</span><span class="p">,</span> <span class="n">context</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">context_out</span><span class="p">,</span> <span class="p">[</span><span class="n">hidden_state_size</span><span class="p">,</span> <span class="n">out_channels_context</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">hidden_state</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tanh</span><span class="p">(</span><span class="n">hidden_state</span><span class="p">)</span>
+        <span class="n">context</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">context</span><span class="p">)</span>
+
+        <span class="n">coords0</span> <span class="o">=</span> <span class="n">make_coords_grid</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">h</span> <span class="o">//</span> <span class="mi">8</span><span class="p">,</span> <span class="n">w</span> <span class="o">//</span> <span class="mi">8</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">fmap1</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="n">coords1</span> <span class="o">=</span> <span class="n">make_coords_grid</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">h</span> <span class="o">//</span> <span class="mi">8</span><span class="p">,</span> <span class="n">w</span> <span class="o">//</span> <span class="mi">8</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">fmap1</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="n">flow_predictions</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_flow_updates</span><span class="p">):</span>
+            <span class="n">coords1</span> <span class="o">=</span> <span class="n">coords1</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>  <span class="c1"># Don&#39;t backpropagate gradients through this branch, see paper</span>
+            <span class="n">corr_features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">corr_block</span><span class="o">.</span><span class="n">index_pyramid</span><span class="p">(</span><span class="n">centroids_coords</span><span class="o">=</span><span class="n">coords1</span><span class="p">)</span>
+
+            <span class="n">flow</span> <span class="o">=</span> <span class="n">coords1</span> <span class="o">-</span> <span class="n">coords0</span>
+            <span class="n">hidden_state</span><span class="p">,</span> <span class="n">delta_flow</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">update_block</span><span class="p">(</span><span class="n">hidden_state</span><span class="p">,</span> <span class="n">context</span><span class="p">,</span> <span class="n">corr_features</span><span class="p">,</span> <span class="n">flow</span><span class="p">)</span>
+
+            <span class="n">coords1</span> <span class="o">=</span> <span class="n">coords1</span> <span class="o">+</span> <span class="n">delta_flow</span>
+
+            <span class="n">up_mask</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask_predictor</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask_predictor</span><span class="p">(</span><span class="n">hidden_state</span><span class="p">)</span>
+            <span class="n">upsampled_flow</span> <span class="o">=</span> <span class="n">upsample_flow</span><span class="p">(</span><span class="n">flow</span><span class="o">=</span><span class="p">(</span><span class="n">coords1</span> <span class="o">-</span> <span class="n">coords0</span><span class="p">),</span> <span class="n">up_mask</span><span class="o">=</span><span class="n">up_mask</span><span class="p">)</span>
+            <span class="n">flow_predictions</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">upsampled_flow</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">flow_predictions</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">),</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="Raft_Large_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.optical_flow.raft_large.html#torchvision.models.optical_flow.Raft_Large_Weights">[docs]</a><span class="k">class</span> <span class="nc">Raft_Large_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The metrics reported here are as follows.</span>
+
+<span class="sd">    ``epe`` is the &quot;end-point-error&quot; and indicates how far (in pixels) the</span>
+<span class="sd">    predicted flow is from its true value. This is averaged over all pixels</span>
+<span class="sd">    of all images. ``per_image_epe`` is similar, but the average is different:</span>
+<span class="sd">    the epe is first computed on each image independently, and then averaged</span>
+<span class="sd">    over all images. This corresponds to &quot;Fl-epe&quot; (sometimes written &quot;F1-epe&quot;)</span>
+<span class="sd">    in the original paper, and it&#39;s only used on Kitti. ``fl-all`` is also a</span>
+<span class="sd">    Kitti-specific metric, defined by the author of the dataset and used for the</span>
+<span class="sd">    Kitti leaderboard. It corresponds to the average of pixels whose epe is</span>
+<span class="sd">    either &lt;3px, or &lt;5% of flow&#39;s 2-norm.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">C_T_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/princeton-vl/RAFT</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_large_C_T_V1-22a6c225.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5257536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/princeton-vl/RAFT&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Sintel-Train-Cleanpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">1.4411</span><span class="p">},</span>
+                <span class="s2">&quot;Sintel-Train-Finalpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">2.7894</span><span class="p">},</span>
+                <span class="s2">&quot;Kitti-Train&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;per_image_epe&quot;</span><span class="p">:</span> <span class="mf">5.0172</span><span class="p">,</span> <span class="s2">&quot;fl_all&quot;</span><span class="p">:</span> <span class="mf">17.4506</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">211.007</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">20.129</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were ported from the original paper. They</span>
+<span class="s2">            are trained on :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">            :class:`~torchvision.datasets.FlyingThings3D`.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="n">C_T_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_large_C_T_V2-1bb1363a.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5257536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/optical_flow&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Sintel-Train-Cleanpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">1.3822</span><span class="p">},</span>
+                <span class="s2">&quot;Sintel-Train-Finalpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">2.7161</span><span class="p">},</span>
+                <span class="s2">&quot;Kitti-Train&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;per_image_epe&quot;</span><span class="p">:</span> <span class="mf">4.5118</span><span class="p">,</span> <span class="s2">&quot;fl_all&quot;</span><span class="p">:</span> <span class="mf">16.0679</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">211.007</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">20.129</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were trained from scratch on</span>
+<span class="s2">            :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">            :class:`~torchvision.datasets.FlyingThings3D`.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="n">C_T_SKHT_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/princeton-vl/RAFT</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_large_C_T_SKHT_V1-0b8c9e55.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5257536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/princeton-vl/RAFT&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Sintel-Test-Cleanpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">1.94</span><span class="p">},</span>
+                <span class="s2">&quot;Sintel-Test-Finalpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">3.18</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">211.007</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">20.129</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were ported from the original paper. They are</span>
+<span class="s2">                trained on :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">                :class:`~torchvision.datasets.FlyingThings3D` and fine-tuned on</span>
+<span class="s2">                Sintel. The Sintel fine-tuning step is a combination of</span>
+<span class="s2">                :class:`~torchvision.datasets.Sintel`,</span>
+<span class="s2">                :class:`~torchvision.datasets.KittiFlow`,</span>
+<span class="s2">                :class:`~torchvision.datasets.HD1K`, and</span>
+<span class="s2">                :class:`~torchvision.datasets.FlyingThings3D` (clean pass).</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="n">C_T_SKHT_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_large_C_T_SKHT_V2-ff5fadd5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5257536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/optical_flow&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Sintel-Test-Cleanpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">1.819</span><span class="p">},</span>
+                <span class="s2">&quot;Sintel-Test-Finalpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">3.067</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">211.007</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">20.129</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch. They are</span>
+<span class="s2">                pre-trained on :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">                :class:`~torchvision.datasets.FlyingThings3D` and then</span>
+<span class="s2">                fine-tuned on Sintel. The Sintel fine-tuning step is a</span>
+<span class="s2">                combination of :class:`~torchvision.datasets.Sintel`,</span>
+<span class="s2">                :class:`~torchvision.datasets.KittiFlow`,</span>
+<span class="s2">                :class:`~torchvision.datasets.HD1K`, and</span>
+<span class="s2">                :class:`~torchvision.datasets.FlyingThings3D` (clean pass).</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="n">C_T_SKHT_K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/princeton-vl/RAFT</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_large_C_T_SKHT_K_V1-4a6a5039.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5257536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/princeton-vl/RAFT&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kitti-Test&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;fl_all&quot;</span><span class="p">:</span> <span class="mf">5.10</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">211.007</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">20.129</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were ported from the original paper. They are</span>
+<span class="s2">                pre-trained on :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">                :class:`~torchvision.datasets.FlyingThings3D`,</span>
+<span class="s2">                fine-tuned on Sintel, and then fine-tuned on</span>
+<span class="s2">                :class:`~torchvision.datasets.KittiFlow`. The Sintel fine-tuning</span>
+<span class="s2">                step was described above.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="n">C_T_SKHT_K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_large_C_T_SKHT_K_V2-b5c70766.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5257536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/optical_flow&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kitti-Test&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;fl_all&quot;</span><span class="p">:</span> <span class="mf">5.19</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">211.007</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">20.129</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch. They are</span>
+<span class="s2">                pre-trained on :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">                :class:`~torchvision.datasets.FlyingThings3D`,</span>
+<span class="s2">                fine-tuned on Sintel, and then fine-tuned on</span>
+<span class="s2">                :class:`~torchvision.datasets.KittiFlow`. The Sintel fine-tuning</span>
+<span class="s2">                step was described above.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">C_T_SKHT_V2</span></div>
+
+
+<div class="viewcode-block" id="Raft_Small_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.optical_flow.raft_small.html#torchvision.models.optical_flow.Raft_Small_Weights">[docs]</a><span class="k">class</span> <span class="nc">Raft_Small_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The metrics reported here are as follows.</span>
+
+<span class="sd">    ``epe`` is the &quot;end-point-error&quot; and indicates how far (in pixels) the</span>
+<span class="sd">    predicted flow is from its true value. This is averaged over all pixels</span>
+<span class="sd">    of all images. ``per_image_epe`` is similar, but the average is different:</span>
+<span class="sd">    the epe is first computed on each image independently, and then averaged</span>
+<span class="sd">    over all images. This corresponds to &quot;Fl-epe&quot; (sometimes written &quot;F1-epe&quot;)</span>
+<span class="sd">    in the original paper, and it&#39;s only used on Kitti. ``fl-all`` is also a</span>
+<span class="sd">    Kitti-specific metric, defined by the author of the dataset and used for the</span>
+<span class="sd">    Kitti leaderboard. It corresponds to the average of pixels whose epe is</span>
+<span class="sd">    either &lt;3px, or &lt;5% of flow&#39;s 2-norm.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">C_T_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/princeton-vl/RAFT</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_small_C_T_V1-ad48884c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">990162</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/princeton-vl/RAFT&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Sintel-Train-Cleanpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">2.1231</span><span class="p">},</span>
+                <span class="s2">&quot;Sintel-Train-Finalpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">3.2790</span><span class="p">},</span>
+                <span class="s2">&quot;Kitti-Train&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;per_image_epe&quot;</span><span class="p">:</span> <span class="mf">7.6557</span><span class="p">,</span> <span class="s2">&quot;fl_all&quot;</span><span class="p">:</span> <span class="mf">25.2801</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">47.655</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">3.821</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were ported from the original paper. They</span>
+<span class="s2">            are trained on :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">            :class:`~torchvision.datasets.FlyingThings3D`.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">C_T_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/raft_small_C_T_V2-01064c6d.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">OpticalFlow</span><span class="p">,</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">990162</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/optical_flow&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Sintel-Train-Cleanpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">1.9901</span><span class="p">},</span>
+                <span class="s2">&quot;Sintel-Train-Finalpass&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;epe&quot;</span><span class="p">:</span> <span class="mf">3.2831</span><span class="p">},</span>
+                <span class="s2">&quot;Kitti-Train&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;per_image_epe&quot;</span><span class="p">:</span> <span class="mf">7.5978</span><span class="p">,</span> <span class="s2">&quot;fl_all&quot;</span><span class="p">:</span> <span class="mf">25.2369</span><span class="p">},</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">47.655</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">3.821</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were trained from scratch on</span>
+<span class="s2">            :class:`~torchvision.datasets.FlyingChairs` +</span>
+<span class="s2">            :class:`~torchvision.datasets.FlyingThings3D`.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">C_T_V2</span></div>
+
+
+<span class="k">def</span> <span class="nf">_raft</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+    <span class="c1"># Feature encoder</span>
+    <span class="n">feature_encoder_layers</span><span class="p">,</span>
+    <span class="n">feature_encoder_block</span><span class="p">,</span>
+    <span class="n">feature_encoder_norm_layer</span><span class="p">,</span>
+    <span class="c1"># Context encoder</span>
+    <span class="n">context_encoder_layers</span><span class="p">,</span>
+    <span class="n">context_encoder_block</span><span class="p">,</span>
+    <span class="n">context_encoder_norm_layer</span><span class="p">,</span>
+    <span class="c1"># Correlation block</span>
+    <span class="n">corr_block_num_levels</span><span class="p">,</span>
+    <span class="n">corr_block_radius</span><span class="p">,</span>
+    <span class="c1"># Motion encoder</span>
+    <span class="n">motion_encoder_corr_layers</span><span class="p">,</span>
+    <span class="n">motion_encoder_flow_layers</span><span class="p">,</span>
+    <span class="n">motion_encoder_out_channels</span><span class="p">,</span>
+    <span class="c1"># Recurrent block</span>
+    <span class="n">recurrent_block_hidden_state_size</span><span class="p">,</span>
+    <span class="n">recurrent_block_kernel_size</span><span class="p">,</span>
+    <span class="n">recurrent_block_padding</span><span class="p">,</span>
+    <span class="c1"># Flow Head</span>
+    <span class="n">flow_head_hidden_size</span><span class="p">,</span>
+    <span class="c1"># Mask predictor</span>
+    <span class="n">use_mask_predictor</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="n">feature_encoder</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;feature_encoder&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">or</span> <span class="n">FeatureEncoder</span><span class="p">(</span>
+        <span class="n">block</span><span class="o">=</span><span class="n">feature_encoder_block</span><span class="p">,</span> <span class="n">layers</span><span class="o">=</span><span class="n">feature_encoder_layers</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">feature_encoder_norm_layer</span>
+    <span class="p">)</span>
+    <span class="n">context_encoder</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;context_encoder&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">or</span> <span class="n">FeatureEncoder</span><span class="p">(</span>
+        <span class="n">block</span><span class="o">=</span><span class="n">context_encoder_block</span><span class="p">,</span> <span class="n">layers</span><span class="o">=</span><span class="n">context_encoder_layers</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">context_encoder_norm_layer</span>
+    <span class="p">)</span>
+
+    <span class="n">corr_block</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;corr_block&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">or</span> <span class="n">CorrBlock</span><span class="p">(</span><span class="n">num_levels</span><span class="o">=</span><span class="n">corr_block_num_levels</span><span class="p">,</span> <span class="n">radius</span><span class="o">=</span><span class="n">corr_block_radius</span><span class="p">)</span>
+
+    <span class="n">update_block</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;update_block&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">update_block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">motion_encoder</span> <span class="o">=</span> <span class="n">MotionEncoder</span><span class="p">(</span>
+            <span class="n">in_channels_corr</span><span class="o">=</span><span class="n">corr_block</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span>
+            <span class="n">corr_layers</span><span class="o">=</span><span class="n">motion_encoder_corr_layers</span><span class="p">,</span>
+            <span class="n">flow_layers</span><span class="o">=</span><span class="n">motion_encoder_flow_layers</span><span class="p">,</span>
+            <span class="n">out_channels</span><span class="o">=</span><span class="n">motion_encoder_out_channels</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="c1"># See comments in forward pass of RAFT class about why we split the output of the context encoder</span>
+        <span class="n">out_channels_context</span> <span class="o">=</span> <span class="n">context_encoder_layers</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">recurrent_block_hidden_state_size</span>
+        <span class="n">recurrent_block</span> <span class="o">=</span> <span class="n">RecurrentBlock</span><span class="p">(</span>
+            <span class="n">input_size</span><span class="o">=</span><span class="n">motion_encoder</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">+</span> <span class="n">out_channels_context</span><span class="p">,</span>
+            <span class="n">hidden_size</span><span class="o">=</span><span class="n">recurrent_block_hidden_state_size</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="n">recurrent_block_kernel_size</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="n">recurrent_block_padding</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="n">flow_head</span> <span class="o">=</span> <span class="n">FlowHead</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="n">recurrent_block_hidden_state_size</span><span class="p">,</span> <span class="n">hidden_size</span><span class="o">=</span><span class="n">flow_head_hidden_size</span><span class="p">)</span>
+
+        <span class="n">update_block</span> <span class="o">=</span> <span class="n">UpdateBlock</span><span class="p">(</span><span class="n">motion_encoder</span><span class="o">=</span><span class="n">motion_encoder</span><span class="p">,</span> <span class="n">recurrent_block</span><span class="o">=</span><span class="n">recurrent_block</span><span class="p">,</span> <span class="n">flow_head</span><span class="o">=</span><span class="n">flow_head</span><span class="p">)</span>
+
+    <span class="n">mask_predictor</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;mask_predictor&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">mask_predictor</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">use_mask_predictor</span><span class="p">:</span>
+        <span class="n">mask_predictor</span> <span class="o">=</span> <span class="n">MaskPredictor</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="o">=</span><span class="n">recurrent_block_hidden_state_size</span><span class="p">,</span>
+            <span class="n">hidden_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span>
+            <span class="n">multiplier</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span>  <span class="c1"># See comment in MaskPredictor about this</span>
+        <span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">RAFT</span><span class="p">(</span>
+        <span class="n">feature_encoder</span><span class="o">=</span><span class="n">feature_encoder</span><span class="p">,</span>
+        <span class="n">context_encoder</span><span class="o">=</span><span class="n">context_encoder</span><span class="p">,</span>
+        <span class="n">corr_block</span><span class="o">=</span><span class="n">corr_block</span><span class="p">,</span>
+        <span class="n">update_block</span><span class="o">=</span><span class="n">update_block</span><span class="p">,</span>
+        <span class="n">mask_predictor</span><span class="o">=</span><span class="n">mask_predictor</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>  <span class="c1"># not really needed, all params should be consumed by now</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<div class="viewcode-block" id="raft_large"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.optical_flow.raft_large.html#torchvision.models.optical_flow.raft_large">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Raft_Large_Weights</span><span class="o">.</span><span class="n">C_T_SKHT_V2</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">raft_large</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Raft_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RAFT</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;RAFT model from</span>
+<span class="sd">    `RAFT: Recurrent All Pairs Field Transforms for Optical Flow &lt;https://arxiv.org/abs/2003.12039&gt;`_.</span>
+
+<span class="sd">    Please see the example below for a tutorial on how to use this model.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights(:class:`~torchvision.models.optical_flow.Raft_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.optical_flow.Raft_Large_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.optical_flow.RAFT``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/optical_flow/raft.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.optical_flow.Raft_Large_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Raft_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_raft</span><span class="p">(</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="c1"># Feature encoder</span>
+        <span class="n">feature_encoder_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+        <span class="n">feature_encoder_block</span><span class="o">=</span><span class="n">ResidualBlock</span><span class="p">,</span>
+        <span class="n">feature_encoder_norm_layer</span><span class="o">=</span><span class="n">InstanceNorm2d</span><span class="p">,</span>
+        <span class="c1"># Context encoder</span>
+        <span class="n">context_encoder_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+        <span class="n">context_encoder_block</span><span class="o">=</span><span class="n">ResidualBlock</span><span class="p">,</span>
+        <span class="n">context_encoder_norm_layer</span><span class="o">=</span><span class="n">BatchNorm2d</span><span class="p">,</span>
+        <span class="c1"># Correlation block</span>
+        <span class="n">corr_block_num_levels</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+        <span class="n">corr_block_radius</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+        <span class="c1"># Motion encoder</span>
+        <span class="n">motion_encoder_corr_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">192</span><span class="p">),</span>
+        <span class="n">motion_encoder_flow_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">64</span><span class="p">),</span>
+        <span class="n">motion_encoder_out_channels</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
+        <span class="c1"># Recurrent block</span>
+        <span class="n">recurrent_block_hidden_state_size</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
+        <span class="n">recurrent_block_kernel_size</span><span class="o">=</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="mi">5</span><span class="p">,</span> <span class="mi">1</span><span class="p">)),</span>
+        <span class="n">recurrent_block_padding</span><span class="o">=</span><span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+        <span class="c1"># Flow head</span>
+        <span class="n">flow_head_hidden_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span>
+        <span class="c1"># Mask predictor</span>
+        <span class="n">use_mask_predictor</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="raft_small"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.optical_flow.raft_small.html#torchvision.models.optical_flow.raft_small">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Raft_Small_Weights</span><span class="o">.</span><span class="n">C_T_V2</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">raft_small</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Raft_Small_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RAFT</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;RAFT &quot;small&quot; model from</span>
+<span class="sd">    `RAFT: Recurrent All Pairs Field Transforms for Optical Flow &lt;https://arxiv.org/abs/2003.12039&gt;`__.</span>
+
+<span class="sd">    Please see the example below for a tutorial on how to use this model.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights(:class:`~torchvision.models.optical_flow.Raft_Small_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.optical_flow.Raft_Small_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.optical_flow.RAFT``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/optical_flow/raft.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.optical_flow.Raft_Small_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Raft_Small_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_raft</span><span class="p">(</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="c1"># Feature encoder</span>
+        <span class="n">feature_encoder_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">128</span><span class="p">),</span>
+        <span class="n">feature_encoder_block</span><span class="o">=</span><span class="n">BottleneckBlock</span><span class="p">,</span>
+        <span class="n">feature_encoder_norm_layer</span><span class="o">=</span><span class="n">InstanceNorm2d</span><span class="p">,</span>
+        <span class="c1"># Context encoder</span>
+        <span class="n">context_encoder_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">160</span><span class="p">),</span>
+        <span class="n">context_encoder_block</span><span class="o">=</span><span class="n">BottleneckBlock</span><span class="p">,</span>
+        <span class="n">context_encoder_norm_layer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="c1"># Correlation block</span>
+        <span class="n">corr_block_num_levels</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+        <span class="n">corr_block_radius</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
+        <span class="c1"># Motion encoder</span>
+        <span class="n">motion_encoder_corr_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">96</span><span class="p">,),</span>
+        <span class="n">motion_encoder_flow_layers</span><span class="o">=</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">32</span><span class="p">),</span>
+        <span class="n">motion_encoder_out_channels</span><span class="o">=</span><span class="mi">82</span><span class="p">,</span>
+        <span class="c1"># Recurrent block</span>
+        <span class="n">recurrent_block_hidden_state_size</span><span class="o">=</span><span class="mi">96</span><span class="p">,</span>
+        <span class="n">recurrent_block_kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,),</span>
+        <span class="n">recurrent_block_padding</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,),</span>
+        <span class="c1"># Flow head</span>
+        <span class="n">flow_head_hidden_size</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
+        <span class="c1"># Mask predictor</span>
+        <span class="n">use_mask_predictor</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/quantization/googlenet.html b/0.16/_modules/torchvision/models/quantization/googlenet.html
new file mode 100644
index 00000000000..fb72b77ddb9
--- /dev/null
+++ b/0.16/_modules/torchvision/models/quantization/googlenet.html
@@ -0,0 +1,879 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.quantization.googlenet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.quantization.googlenet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.quantization.googlenet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.nn</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..googlenet</span> <span class="kn">import</span> <span class="n">BasicConv2d</span><span class="p">,</span> <span class="n">GoogLeNet</span><span class="p">,</span> <span class="n">GoogLeNet_Weights</span><span class="p">,</span> <span class="n">GoogLeNetOutputs</span><span class="p">,</span> <span class="n">Inception</span><span class="p">,</span> <span class="n">InceptionAux</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_fuse_modules</span><span class="p">,</span> <span class="n">_replace_relu</span><span class="p">,</span> <span class="n">quantize_model</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;QuantizableGoogLeNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;GoogLeNet_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;googlenet&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableBasicConv2d</span><span class="p">(</span><span class="n">BasicConv2d</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;conv&quot;</span><span class="p">,</span> <span class="s2">&quot;bn&quot;</span><span class="p">,</span> <span class="s2">&quot;relu&quot;</span><span class="p">],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInception</span><span class="p">(</span><span class="n">Inception</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">cat</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">cat</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInceptionAux</span><span class="p">(</span><span class="n">InceptionAux</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># aux1: N x 512 x 14 x 14, aux2: N x 528 x 14 x 14</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adaptive_avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">))</span>
+        <span class="c1"># aux1: N x 512 x 4 x 4, aux2: N x 528 x 4 x 4</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 128 x 4 x 4</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># N x 2048</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fc1</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="c1"># N x 1024</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1024</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># N x 1000 (num_classes)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableGoogLeNet</span><span class="p">(</span><span class="n">GoogLeNet</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>  <span class="c1"># type: ignore[misc]</span>
+            <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">blocks</span><span class="o">=</span><span class="p">[</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="n">QuantizableInception</span><span class="p">,</span> <span class="n">QuantizableInceptionAux</span><span class="p">],</span> <span class="o">**</span><span class="n">kwargs</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">quant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">QuantStub</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">DeQuantStub</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">GoogLeNetOutputs</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_transform_input</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span><span class="p">,</span> <span class="n">aux1</span><span class="p">,</span> <span class="n">aux2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">aux_defined</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">aux_defined</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scripted QuantizableGoogleNet always returns GoogleNetOutputs Tuple&quot;</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">GoogLeNetOutputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux2</span><span class="p">,</span> <span class="n">aux1</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">eager_outputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux2</span><span class="p">,</span> <span class="n">aux1</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Fuse conv/bn/relu modules in googlenet model</span>
+
+<span class="sd">        Fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization.</span>
+<span class="sd">        Model is modified in place.  Note that this operation does not change numerics</span>
+<span class="sd">        and the model after modification is in floating point</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">QuantizableBasicConv2d</span><span class="p">:</span>
+                <span class="n">m</span><span class="o">.</span><span class="n">fuse_model</span><span class="p">(</span><span class="n">is_qat</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="GoogLeNet_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.googlenet.html#torchvision.models.quantization.GoogLeNet_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">GoogLeNet_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/googlenet_fbgemm-c81f6644.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">6624904</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">15</span><span class="p">,</span> <span class="mi">15</span><span class="p">),</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;backend&quot;</span><span class="p">:</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">GoogLeNet_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">69.826</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">89.404</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.498</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">12.618</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized</span>
+<span class="s2">                weights listed below.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="googlenet"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.googlenet.html#torchvision.models.quantization.googlenet">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_googlenet&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">GoogLeNet_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">GoogLeNet_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">googlenet</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">GoogLeNet_QuantizedWeights</span><span class="p">,</span> <span class="n">GoogLeNet_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableGoogLeNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;GoogLeNet (Inception v1) model architecture from `Going Deeper with Convolutions &lt;http://arxiv.org/abs/1409.4842&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.GoogLeNet_QuantizedWeights` or :class:`~torchvision.models.GoogLeNet_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.GoogLeNet_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model. Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.QuantizableGoogLeNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/googlenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.GoogLeNet_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.GoogLeNet_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">GoogLeNet_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">GoogLeNet_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">original_aux_logits</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="s2">&quot;transform_input&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;transform_input&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;init_weights&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="s2">&quot;backend&quot;</span> <span class="ow">in</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;backend&quot;</span><span class="p">])</span>
+    <span class="n">backend</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">QuantizableGoogLeNet</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="n">_replace_relu</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">quantize</span><span class="p">:</span>
+        <span class="n">quantize_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">backend</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">original_aux_logits</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux_logits</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux1</span> <span class="o">=</span> <span class="kc">None</span>  <span class="c1"># type: ignore[assignment]</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux2</span> <span class="o">=</span> <span class="kc">None</span>  <span class="c1"># type: ignore[assignment]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;auxiliary heads in the pretrained googlenet model are NOT pretrained, so make sure to train them&quot;</span>
+            <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/quantization/inception.html b/0.16/_modules/torchvision/models/quantization/inception.html
new file mode 100644
index 00000000000..95543e950e2
--- /dev/null
+++ b/0.16/_modules/torchvision/models/quantization/inception.html
@@ -0,0 +1,942 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.quantization.inception &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.quantization.inception</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.quantization.inception</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torchvision.models</span> <span class="kn">import</span> <span class="n">inception</span> <span class="k">as</span> <span class="n">inception_module</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.inception</span> <span class="kn">import</span> <span class="n">Inception_V3_Weights</span><span class="p">,</span> <span class="n">InceptionOutputs</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_fuse_modules</span><span class="p">,</span> <span class="n">_replace_relu</span><span class="p">,</span> <span class="n">quantize_model</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;QuantizableInception3&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Inception_V3_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;inception_v3&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableBasicConv2d</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">BasicConv2d</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;conv&quot;</span><span class="p">,</span> <span class="s2">&quot;bn&quot;</span><span class="p">,</span> <span class="s2">&quot;relu&quot;</span><span class="p">],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInceptionA</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">InceptionA</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">myop</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">myop</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInceptionB</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">InceptionB</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">myop</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">myop</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInceptionC</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">InceptionC</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">myop</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">myop</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInceptionD</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">InceptionD</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">myop</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">myop</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInceptionE</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">InceptionE</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">myop1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">myop2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">myop3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">branch1x1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch1x1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2a</span><span class="p">(</span><span class="n">branch3x3</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3_2b</span><span class="p">(</span><span class="n">branch3x3</span><span class="p">)]</span>
+        <span class="n">branch3x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">myop1</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">branch3x3</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_2</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">)</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3a</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">),</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch3x3dbl_3b</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="n">branch3x3dbl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">myop2</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">branch3x3dbl</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">avg_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">branch_pool</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch_pool</span><span class="p">(</span><span class="n">branch_pool</span><span class="p">)</span>
+
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">branch1x1</span><span class="p">,</span> <span class="n">branch3x3</span><span class="p">,</span> <span class="n">branch3x3dbl</span><span class="p">,</span> <span class="n">branch_pool</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">outputs</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">myop3</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInceptionAux</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">InceptionAux</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">conv_block</span><span class="o">=</span><span class="n">QuantizableBasicConv2d</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInception3</span><span class="p">(</span><span class="n">inception_module</span><span class="o">.</span><span class="n">Inception3</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>  <span class="c1"># type: ignore[misc]</span>
+            <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+            <span class="n">inception_blocks</span><span class="o">=</span><span class="p">[</span>
+                <span class="n">QuantizableBasicConv2d</span><span class="p">,</span>
+                <span class="n">QuantizableInceptionA</span><span class="p">,</span>
+                <span class="n">QuantizableInceptionB</span><span class="p">,</span>
+                <span class="n">QuantizableInceptionC</span><span class="p">,</span>
+                <span class="n">QuantizableInceptionD</span><span class="p">,</span>
+                <span class="n">QuantizableInceptionE</span><span class="p">,</span>
+                <span class="n">QuantizableInceptionAux</span><span class="p">,</span>
+            <span class="p">],</span>
+            <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">quant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">QuantStub</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">DeQuantStub</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">InceptionOutputs</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_transform_input</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span><span class="p">,</span> <span class="n">aux</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">aux_defined</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">aux_logits</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">aux_defined</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scripted QuantizableInception3 always returns QuantizableInception3 Tuple&quot;</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">InceptionOutputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">eager_outputs</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">aux</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Fuse conv/bn/relu modules in inception model</span>
+
+<span class="sd">        Fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization.</span>
+<span class="sd">        Model is modified in place.  Note that this operation does not change numerics</span>
+<span class="sd">        and the model after modification is in floating point</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">QuantizableBasicConv2d</span><span class="p">:</span>
+                <span class="n">m</span><span class="o">.</span><span class="n">fuse_model</span><span class="p">(</span><span class="n">is_qat</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="Inception_V3_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.Inception_V3_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">Inception_V3_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/inception_v3_google_fbgemm-a2837893.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">299</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">342</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">27161264</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">75</span><span class="p">,</span> <span class="mi">75</span><span class="p">),</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;backend&quot;</span><span class="p">:</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">Inception_V3_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.176</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.354</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">5.713</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">23.146</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized</span>
+<span class="s2">                weights listed below.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="inception_v3"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.inception_v3">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_inception_v3&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">Inception_V3_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">Inception_V3_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">inception_v3</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Inception_V3_QuantizedWeights</span><span class="p">,</span> <span class="n">Inception_V3_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableInception3</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Inception v3 model architecture from</span>
+<span class="sd">    `Rethinking the Inception Architecture for Computer Vision &lt;http://arxiv.org/abs/1512.00567&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        **Important**: In contrast to the other models the inception_v3 expects tensors with a size of</span>
+<span class="sd">        N x 3 x 299 x 299, so ensure your images are sized accordingly.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.Inception_V3_QuantizedWeights` or :class:`~torchvision.models.Inception_V3_Weights`, optional): The pretrained</span>
+<span class="sd">            weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.Inception_V3_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr.</span>
+<span class="sd">            Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model.</span>
+<span class="sd">            Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.QuantizableInception3``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/inception.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.Inception_V3_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Inception_V3_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">Inception_V3_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">Inception_V3_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">original_aux_logits</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="s2">&quot;transform_input&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;transform_input&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;aux_logits&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="s2">&quot;backend&quot;</span> <span class="ow">in</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;backend&quot;</span><span class="p">])</span>
+    <span class="n">backend</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">QuantizableInception3</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="n">_replace_relu</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">quantize</span><span class="p">:</span>
+        <span class="n">quantize_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">backend</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">quantize</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">original_aux_logits</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux_logits</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">AuxLogits</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">quantize</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">original_aux_logits</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">aux_logits</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">AuxLogits</span> <span class="o">=</span> <span class="kc">None</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/quantization/mobilenetv2.html b/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
new file mode 100644
index 00000000000..255f4921153
--- /dev/null
+++ b/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
@@ -0,0 +1,823 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.quantization.mobilenetv2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.quantization.mobilenetv2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.quantization.mobilenetv2</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.ao.quantization</span> <span class="kn">import</span> <span class="n">DeQuantStub</span><span class="p">,</span> <span class="n">QuantStub</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.mobilenetv2</span> <span class="kn">import</span> <span class="n">InvertedResidual</span><span class="p">,</span> <span class="n">MobileNet_V2_Weights</span><span class="p">,</span> <span class="n">MobileNetV2</span>
+
+<span class="kn">from</span> <span class="nn">...ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_fuse_modules</span><span class="p">,</span> <span class="n">_replace_relu</span><span class="p">,</span> <span class="n">quantize_model</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;QuantizableMobileNetV2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MobileNet_V2_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mobilenet_v2&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInvertedResidual</span><span class="p">(</span><span class="n">InvertedResidual</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">skip_add</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">skip_add</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">)):</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">[</span><span class="n">idx</span><span class="p">])</span> <span class="ow">is</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">:</span>
+                <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">,</span> <span class="p">[</span><span class="nb">str</span><span class="p">(</span><span class="n">idx</span><span class="p">),</span> <span class="nb">str</span><span class="p">(</span><span class="n">idx</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableMobileNetV2</span><span class="p">(</span><span class="n">MobileNetV2</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        MobileNet V2 main class</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">           Inherits args from floating point MobileNetV2</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">quant</span> <span class="o">=</span> <span class="n">QuantStub</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span> <span class="o">=</span> <span class="n">DeQuantStub</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">Conv2dNormActivation</span><span class="p">:</span>
+                <span class="n">_fuse_modules</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">,</span> <span class="s2">&quot;2&quot;</span><span class="p">],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">QuantizableInvertedResidual</span><span class="p">:</span>
+                <span class="n">m</span><span class="o">.</span><span class="n">fuse_model</span><span class="p">(</span><span class="n">is_qat</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="MobileNet_V2_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.mobilenet_v2.html#torchvision.models.quantization.MobileNet_V2_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">MobileNet_V2_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_QNNPACK_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">3504872</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;backend&quot;</span><span class="p">:</span> <span class="s2">&quot;qnnpack&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">MobileNet_V2_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">71.658</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.150</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.301</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">3.423</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were produced by doing Quantization Aware Training (eager mode) on top of the unquantized</span>
+<span class="s2">                weights listed below.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_QNNPACK_V1</span></div>
+
+
+<div class="viewcode-block" id="mobilenet_v2"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.mobilenet_v2.html#torchvision.models.quantization.mobilenet_v2">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_mobilenet_v2&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">MobileNet_V2_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_QNNPACK_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">MobileNet_V2_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">mobilenet_v2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">MobileNet_V2_QuantizedWeights</span><span class="p">,</span> <span class="n">MobileNet_V2_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableMobileNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a MobileNetV2 architecture from</span>
+<span class="sd">    `MobileNetV2: Inverted Residuals and Linear Bottlenecks</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1801.04381&gt;`_.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.MobileNet_V2_QuantizedWeights` or :class:`~torchvision.models.MobileNet_V2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.MobileNet_V2_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, returns a quantized version of the model. Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.QuantizableMobileNetV2``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.quantization.MobileNet_V2_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    .. autoclass:: torchvision.models.MobileNet_V2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">MobileNet_V2_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">MobileNet_V2_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="s2">&quot;backend&quot;</span> <span class="ow">in</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;backend&quot;</span><span class="p">])</span>
+    <span class="n">backend</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="s2">&quot;qnnpack&quot;</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">QuantizableMobileNetV2</span><span class="p">(</span><span class="n">block</span><span class="o">=</span><span class="n">QuantizableInvertedResidual</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="n">_replace_relu</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">quantize</span><span class="p">:</span>
+        <span class="n">quantize_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">backend</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/quantization/mobilenetv3.html b/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
new file mode 100644
index 00000000000..60c9ba8f8b3
--- /dev/null
+++ b/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
@@ -0,0 +1,906 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.quantization.mobilenetv3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.quantization.mobilenetv3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.quantization.mobilenetv3</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.ao.quantization</span> <span class="kn">import</span> <span class="n">DeQuantStub</span><span class="p">,</span> <span class="n">QuantStub</span>
+
+<span class="kn">from</span> <span class="nn">...ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span><span class="p">,</span> <span class="n">SqueezeExcitation</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..mobilenetv3</span> <span class="kn">import</span> <span class="p">(</span>
+    <span class="n">_mobilenet_v3_conf</span><span class="p">,</span>
+    <span class="n">InvertedResidual</span><span class="p">,</span>
+    <span class="n">InvertedResidualConfig</span><span class="p">,</span>
+    <span class="n">MobileNet_V3_Large_Weights</span><span class="p">,</span>
+    <span class="n">MobileNetV3</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_fuse_modules</span><span class="p">,</span> <span class="n">_replace_relu</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;QuantizableMobileNetV3&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MobileNet_V3_Large_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mobilenet_v3_large&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableSqueezeExcitation</span><span class="p">(</span><span class="n">SqueezeExcitation</span><span class="p">):</span>
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;scale_activation&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Hardsigmoid</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">skip_mul</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">skip_mul</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_scale</span><span class="p">(</span><span class="nb">input</span><span class="p">),</span> <span class="nb">input</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;fc1&quot;</span><span class="p">,</span> <span class="s2">&quot;activation&quot;</span><span class="p">],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">,</span>
+        <span class="n">unexpected_keys</span><span class="p">,</span>
+        <span class="n">error_msgs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">version</span> <span class="o">=</span> <span class="n">local_metadata</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;qconfig&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="p">(</span><span class="n">version</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">version</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="n">default_state_dict</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;scale_activation.activation_post_process.scale&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mf">1.0</span><span class="p">]),</span>
+                <span class="s2">&quot;scale_activation.activation_post_process.activation_post_process.scale&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mf">1.0</span><span class="p">]),</span>
+                <span class="s2">&quot;scale_activation.activation_post_process.zero_point&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">0</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">),</span>
+                <span class="s2">&quot;scale_activation.activation_post_process.activation_post_process.zero_point&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+                    <span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int32</span>
+                <span class="p">),</span>
+                <span class="s2">&quot;scale_activation.activation_post_process.fake_quant_enabled&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">1</span><span class="p">]),</span>
+                <span class="s2">&quot;scale_activation.activation_post_process.observer_enabled&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">1</span><span class="p">]),</span>
+            <span class="p">}</span>
+            <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">default_state_dict</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                <span class="n">full_key</span> <span class="o">=</span> <span class="n">prefix</span> <span class="o">+</span> <span class="n">k</span>
+                <span class="k">if</span> <span class="n">full_key</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">state_dict</span><span class="p">:</span>
+                    <span class="n">state_dict</span><span class="p">[</span><span class="n">full_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span>
+            <span class="n">prefix</span><span class="p">,</span>
+            <span class="n">local_metadata</span><span class="p">,</span>
+            <span class="n">strict</span><span class="p">,</span>
+            <span class="n">missing_keys</span><span class="p">,</span>
+            <span class="n">unexpected_keys</span><span class="p">,</span>
+            <span class="n">error_msgs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInvertedResidual</span><span class="p">(</span><span class="n">InvertedResidual</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">se_layer</span><span class="o">=</span><span class="n">QuantizableSqueezeExcitation</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">skip_add</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_res_connect</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">skip_add</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">block</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">block</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableMobileNetV3</span><span class="p">(</span><span class="n">MobileNetV3</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        MobileNet V3 main class</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">           Inherits args from floating point MobileNetV3</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">quant</span> <span class="o">=</span> <span class="n">QuantStub</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span> <span class="o">=</span> <span class="n">DeQuantStub</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">Conv2dNormActivation</span><span class="p">:</span>
+                <span class="n">modules_to_fuse</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">]</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="o">==</span> <span class="mi">3</span> <span class="ow">and</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">[</span><span class="mi">2</span><span class="p">])</span> <span class="ow">is</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">:</span>
+                    <span class="n">modules_to_fuse</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">&quot;2&quot;</span><span class="p">)</span>
+                <span class="n">_fuse_modules</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">modules_to_fuse</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">QuantizableSqueezeExcitation</span><span class="p">:</span>
+                <span class="n">m</span><span class="o">.</span><span class="n">fuse_model</span><span class="p">(</span><span class="n">is_qat</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_mobilenet_v3_model</span><span class="p">(</span>
+    <span class="n">inverted_residual_setting</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">InvertedResidualConfig</span><span class="p">],</span>
+    <span class="n">last_channel</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableMobileNetV3</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="s2">&quot;backend&quot;</span> <span class="ow">in</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;backend&quot;</span><span class="p">])</span>
+    <span class="n">backend</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="s2">&quot;qnnpack&quot;</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">QuantizableMobileNetV3</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span><span class="p">,</span> <span class="n">block</span><span class="o">=</span><span class="n">QuantizableInvertedResidual</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="n">_replace_relu</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">quantize</span><span class="p">:</span>
+        <span class="c1"># Instead of quantizing the model and then loading the quantized weights we take a different approach.</span>
+        <span class="c1"># We prepare the QAT model, load the QAT weights from training and then convert it.</span>
+        <span class="c1"># This is done to avoid extremely low accuracies observed on the specific model. This is rather a workaround</span>
+        <span class="c1"># for an unresolved bug on the eager quantization API detailed at: https://github.com/pytorch/vision/issues/5890</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">fuse_model</span><span class="p">(</span><span class="n">is_qat</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">qconfig</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">get_default_qat_qconfig</span><span class="p">(</span><span class="n">backend</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">prepare_qat</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">if</span> <span class="n">quantize</span><span class="p">:</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<div class="viewcode-block" id="MobileNet_V3_Large_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.mobilenet_v3_large.html#torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">MobileNet_V3_Large_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_QNNPACK_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/mobilenet_v3_large_qnnpack-5bcacf28.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5483032</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;backend&quot;</span><span class="p">:</span> <span class="s2">&quot;qnnpack&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv3&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">73.004</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.858</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.217</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">21.554</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were produced by doing Quantization Aware Training (eager mode) on top of the unquantized</span>
+<span class="s2">                weights listed below.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_QNNPACK_V1</span></div>
+
+
+<div class="viewcode-block" id="mobilenet_v3_large"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.mobilenet_v3_large.html#torchvision.models.quantization.mobilenet_v3_large">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_mobilenet_v3_large&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">MobileNet_V3_Large_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_QNNPACK_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">mobilenet_v3_large</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">MobileNet_V3_Large_QuantizedWeights</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableMobileNetV3</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    MobileNetV3 (Large) model from</span>
+<span class="sd">    `Searching for MobileNetV3 &lt;https://arxiv.org/abs/1905.02244&gt;`_.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights` or :class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        quantize (bool): If True, return a quantized version of the model. Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv3.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    .. autoclass:: torchvision.models.MobileNet_V3_Large_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">MobileNet_V3_Large_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span> <span class="o">=</span> <span class="n">_mobilenet_v3_conf</span><span class="p">(</span><span class="s2">&quot;mobilenet_v3_large&quot;</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_mobilenet_v3_model</span><span class="p">(</span><span class="n">inverted_residual_setting</span><span class="p">,</span> <span class="n">last_channel</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/quantization/resnet.html b/0.16/_modules/torchvision/models/quantization/resnet.html
new file mode 100644
index 00000000000..846b097e67a
--- /dev/null
+++ b/0.16/_modules/torchvision/models/quantization/resnet.html
@@ -0,0 +1,1153 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.quantization.resnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.quantization.resnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.quantization.resnet</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.resnet</span> <span class="kn">import</span> <span class="p">(</span>
+    <span class="n">BasicBlock</span><span class="p">,</span>
+    <span class="n">Bottleneck</span><span class="p">,</span>
+    <span class="n">ResNet</span><span class="p">,</span>
+    <span class="n">ResNet18_Weights</span><span class="p">,</span>
+    <span class="n">ResNet50_Weights</span><span class="p">,</span>
+    <span class="n">ResNeXt101_32X8D_Weights</span><span class="p">,</span>
+    <span class="n">ResNeXt101_64X4D_Weights</span><span class="p">,</span>
+<span class="p">)</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_fuse_modules</span><span class="p">,</span> <span class="n">_replace_relu</span><span class="p">,</span> <span class="n">quantize_model</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;QuantizableResNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNet18_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNet50_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNeXt101_32X8D_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNeXt101_64X4D_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnet18&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnet50&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnext101_32x8d&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnext101_64x4d&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableBasicBlock</span><span class="p">(</span><span class="n">BasicBlock</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">add_relu</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">identity</span> <span class="o">=</span> <span class="n">x</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">identity</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">add_relu</span><span class="o">.</span><span class="n">add_relu</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="n">identity</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="p">[[</span><span class="s2">&quot;conv1&quot;</span><span class="p">,</span> <span class="s2">&quot;bn1&quot;</span><span class="p">,</span> <span class="s2">&quot;relu&quot;</span><span class="p">],</span> <span class="p">[</span><span class="s2">&quot;conv2&quot;</span><span class="p">,</span> <span class="s2">&quot;bn2&quot;</span><span class="p">]],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">:</span>
+            <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableBottleneck</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">skip_add_relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">identity</span> <span class="o">=</span> <span class="n">x</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu1</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn3</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">identity</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">skip_add_relu</span><span class="o">.</span><span class="n">add_relu</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="n">identity</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_fuse_modules</span><span class="p">(</span>
+            <span class="bp">self</span><span class="p">,</span> <span class="p">[[</span><span class="s2">&quot;conv1&quot;</span><span class="p">,</span> <span class="s2">&quot;bn1&quot;</span><span class="p">,</span> <span class="s2">&quot;relu1&quot;</span><span class="p">],</span> <span class="p">[</span><span class="s2">&quot;conv2&quot;</span><span class="p">,</span> <span class="s2">&quot;bn2&quot;</span><span class="p">,</span> <span class="s2">&quot;relu2&quot;</span><span class="p">],</span> <span class="p">[</span><span class="s2">&quot;conv3&quot;</span><span class="p">,</span> <span class="s2">&quot;bn3&quot;</span><span class="p">]],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">:</span>
+            <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableResNet</span><span class="p">(</span><span class="n">ResNet</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">quant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">QuantStub</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">DeQuantStub</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># Ensure scriptability</span>
+        <span class="c1"># super(QuantizableResNet,self).forward(x)</span>
+        <span class="c1"># is not scriptable</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Fuse conv/bn/relu modules in resnet models</span>
+
+<span class="sd">        Fuse conv+bn+relu/ Conv+relu/conv+Bn modules to prepare for quantization.</span>
+<span class="sd">        Model is modified in place.  Note that this operation does not change numerics</span>
+<span class="sd">        and the model after modification is in floating point</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">_fuse_modules</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;conv1&quot;</span><span class="p">,</span> <span class="s2">&quot;bn1&quot;</span><span class="p">,</span> <span class="s2">&quot;relu&quot;</span><span class="p">],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">QuantizableBottleneck</span> <span class="ow">or</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">QuantizableBasicBlock</span><span class="p">:</span>
+                <span class="n">m</span><span class="o">.</span><span class="n">fuse_model</span><span class="p">(</span><span class="n">is_qat</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_resnet</span><span class="p">(</span>
+    <span class="n">block</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">QuantizableBasicBlock</span><span class="p">,</span> <span class="n">QuantizableBottleneck</span><span class="p">]],</span>
+    <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableResNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="s2">&quot;backend&quot;</span> <span class="ow">in</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;backend&quot;</span><span class="p">])</span>
+    <span class="n">backend</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">QuantizableResNet</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">layers</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="n">_replace_relu</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">quantize</span><span class="p">:</span>
+        <span class="n">quantize_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">backend</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;backend&quot;</span><span class="p">:</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized</span>
+<span class="s2">        weights listed below.</span>
+<span class="s2">    &quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="ResNet18_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnet18.html#torchvision.models.quantization.ResNet18_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ResNet18_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/resnet18_fbgemm_16fa66dd.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">11689512</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ResNet18_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">69.494</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">88.882</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.814</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">11.238</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="ResNet50_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnet50.html#torchvision.models.quantization.ResNet50_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ResNet50_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/resnet50_fbgemm_bf931d71.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">25557032</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">75.920</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.814</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.089</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">24.759</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_FBGEMM_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/resnet50_fbgemm-23753f79.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">25557032</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V2</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.282</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.976</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.089</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">24.953</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V2</span></div>
+
+
+<div class="viewcode-block" id="ResNeXt101_32X8D_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnext101_32x8d.html#torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ResNeXt101_32X8D_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/resnext101_32x8_fbgemm_09835ccf.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88791336</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ResNeXt101_32X8D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.986</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.480</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">16.414</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">86.034</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_FBGEMM_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/resnext101_32x8_fbgemm-ee16d00c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88791336</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ResNeXt101_32X8D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V2</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.574</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.132</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">16.414</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">86.645</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V2</span></div>
+
+
+<div class="viewcode-block" id="ResNeXt101_64X4D_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnext101_64x4d.html#torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ResNeXt101_64X4D_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/resnext101_64x4d_fbgemm-605a1cb3.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">83455272</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5935&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ResNeXt101_64X4D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.898</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.326</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.46</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">81.556</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="resnet18"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnet18.html#torchvision.models.quantization.resnet18">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_resnet18&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ResNet18_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ResNet18_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">resnet18</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ResNet18_QuantizedWeights</span><span class="p">,</span> <span class="n">ResNet18_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNet-18 model from</span>
+<span class="sd">    `Deep Residual Learning for Image Recognition &lt;https://arxiv.org/abs/1512.03385&gt;`_</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ResNet18_QuantizedWeights` or :class:`~torchvision.models.ResNet18_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ResNet18_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model. Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.QuantizableResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ResNet18_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNet18_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ResNet18_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ResNet18_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">QuantizableBasicBlock</span><span class="p">,</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnet50"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnet50.html#torchvision.models.quantization.resnet50">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_resnet50&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ResNet50_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">resnet50</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ResNet50_QuantizedWeights</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNet-50 model from</span>
+<span class="sd">    `Deep Residual Learning for Image Recognition &lt;https://arxiv.org/abs/1512.03385&gt;`_</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ResNet50_QuantizedWeights` or :class:`~torchvision.models.ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ResNet50_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model. Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.QuantizableResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ResNet50_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNet50_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ResNet50_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ResNet50_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">QuantizableBottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnext101_32x8d"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnext101_32x8d.html#torchvision.models.quantization.resnext101_32x8d">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_resnext101_32x8d&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ResNeXt101_32X8D_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ResNeXt101_32X8D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">resnext101_32x8d</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ResNeXt101_32X8D_QuantizedWeights</span><span class="p">,</span> <span class="n">ResNeXt101_32X8D_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNeXt-101 32x8d model from</span>
+<span class="sd">    `Aggregated Residual Transformation for Deep Neural Networks &lt;https://arxiv.org/abs/1611.05431&gt;`_</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights` or :class:`~torchvision.models.ResNeXt101_32X8D_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ResNet101_32X8D_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model. Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.QuantizableResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNeXt101_32X8D_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ResNeXt101_32X8D_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ResNeXt101_32X8D_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;groups&quot;</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;width_per_group&quot;</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">QuantizableBottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">23</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnext101_64x4d"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.resnext101_64x4d.html#torchvision.models.quantization.resnext101_64x4d">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_resnext101_64x4d&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ResNeXt101_64X4D_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ResNeXt101_64X4D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">resnext101_64x4d</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ResNeXt101_64X4D_QuantizedWeights</span><span class="p">,</span> <span class="n">ResNeXt101_64X4D_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNeXt-101 64x4d model from</span>
+<span class="sd">    `Aggregated Residual Transformation for Deep Neural Networks &lt;https://arxiv.org/abs/1611.05431&gt;`_</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights` or :class:`~torchvision.models.ResNeXt101_64X4D_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ResNet101_64X4D_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model. Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.QuantizableResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNeXt101_64X4D_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ResNeXt101_64X4D_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ResNeXt101_64X4D_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;groups&quot;</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;width_per_group&quot;</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">QuantizableBottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">23</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/quantization/shufflenetv2.html b/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
new file mode 100644
index 00000000000..04db9ba72be
--- /dev/null
+++ b/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
@@ -0,0 +1,1096 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.quantization.shufflenetv2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.quantization.shufflenetv2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.quantization.shufflenetv2</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torchvision.models</span> <span class="kn">import</span> <span class="n">shufflenetv2</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..shufflenetv2</span> <span class="kn">import</span> <span class="p">(</span>
+    <span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="p">,</span>
+    <span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="p">,</span>
+    <span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="p">,</span>
+    <span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span> <span class="nn">.utils</span> <span class="kn">import</span> <span class="n">_fuse_modules</span><span class="p">,</span> <span class="n">_replace_relu</span><span class="p">,</span> <span class="n">quantize_model</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;QuantizableShuffleNetV2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X0_5_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X1_0_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X1_5_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X2_0_QuantizedWeights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x0_5&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x1_0&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x1_5&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x2_0&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableInvertedResidual</span><span class="p">(</span><span class="n">shufflenetv2</span><span class="o">.</span><span class="n">InvertedResidual</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">cat</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">quantized</span><span class="o">.</span><span class="n">FloatFunctional</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">x1</span><span class="p">,</span> <span class="n">x2</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">chunk</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">cat</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">x1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x2</span><span class="p">)],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">cat</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="bp">self</span><span class="o">.</span><span class="n">branch1</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x</span><span class="p">)],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="n">shufflenetv2</span><span class="o">.</span><span class="n">channel_shuffle</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">QuantizableShuffleNetV2</span><span class="p">(</span><span class="n">shufflenetv2</span><span class="o">.</span><span class="n">ShuffleNetV2</span><span class="p">):</span>
+    <span class="c1"># TODO https://github.com/pytorch/vision/pull/4232#pullrequestreview-730461659</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="n">inverted_residual</span><span class="o">=</span><span class="n">QuantizableInvertedResidual</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>  <span class="c1"># type: ignore[misc]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">quant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">QuantStub</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ao</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">DeQuantStub</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dequant</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">fuse_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">is_qat</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Fuse conv/bn/relu modules in shufflenetv2 model</span>
+
+<span class="sd">        Fuse conv+bn+relu/ conv+relu/conv+bn modules to prepare for quantization.</span>
+<span class="sd">        Model is modified in place.</span>
+
+<span class="sd">        .. note::</span>
+<span class="sd">            Note that this operation does not change numerics</span>
+<span class="sd">            and the model after modification is in floating point</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_modules</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="k">if</span> <span class="n">name</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;conv1&quot;</span><span class="p">,</span> <span class="s2">&quot;conv5&quot;</span><span class="p">]</span> <span class="ow">and</span> <span class="n">m</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">_fuse_modules</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">[[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">,</span> <span class="s2">&quot;2&quot;</span><span class="p">]],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">m</span><span class="p">)</span> <span class="ow">is</span> <span class="n">QuantizableInvertedResidual</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">branch1</span><span class="o">.</span><span class="n">_modules</span><span class="o">.</span><span class="n">items</span><span class="p">())</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">_fuse_modules</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">branch1</span><span class="p">,</span> <span class="p">[[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">],</span> <span class="p">[</span><span class="s2">&quot;2&quot;</span><span class="p">,</span> <span class="s2">&quot;3&quot;</span><span class="p">,</span> <span class="s2">&quot;4&quot;</span><span class="p">]],</span> <span class="n">is_qat</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">_fuse_modules</span><span class="p">(</span>
+                    <span class="n">m</span><span class="o">.</span><span class="n">branch2</span><span class="p">,</span>
+                    <span class="p">[[</span><span class="s2">&quot;0&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">,</span> <span class="s2">&quot;2&quot;</span><span class="p">],</span> <span class="p">[</span><span class="s2">&quot;3&quot;</span><span class="p">,</span> <span class="s2">&quot;4&quot;</span><span class="p">],</span> <span class="p">[</span><span class="s2">&quot;5&quot;</span><span class="p">,</span> <span class="s2">&quot;6&quot;</span><span class="p">,</span> <span class="s2">&quot;7&quot;</span><span class="p">]],</span>
+                    <span class="n">is_qat</span><span class="p">,</span>
+                    <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_shufflenetv2</span><span class="p">(</span>
+    <span class="n">stages_repeats</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">stages_out_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableShuffleNetV2</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="s2">&quot;backend&quot;</span> <span class="ow">in</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;backend&quot;</span><span class="p">])</span>
+    <span class="n">backend</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;backend&quot;</span><span class="p">,</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">QuantizableShuffleNetV2</span><span class="p">(</span><span class="n">stages_repeats</span><span class="p">,</span> <span class="n">stages_out_channels</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="n">_replace_relu</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">quantize</span><span class="p">:</span>
+        <span class="n">quantize_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">backend</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;backend&quot;</span><span class="p">:</span> <span class="s2">&quot;fbgemm&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized</span>
+<span class="s2">        weights listed below.</span>
+<span class="s2">    &quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X0_5_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html#torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X0_5_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/shufflenetv2_x0.5_fbgemm-00845098.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">1366792</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">57.972</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">79.780</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.04</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">1.501</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X1_0_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html#torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X1_0_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/shufflenetv2_x1_fbgemm-1e62bb32.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">2278604</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">68.360</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">87.582</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.145</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">2.334</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X1_5_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html#torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X1_5_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/shufflenetv2_x1_5_fbgemm-d7401f05.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5906&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">3503624</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">72.052</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.700</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.296</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">3.672</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X2_0_QuantizedWeights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html#torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X2_0_QuantizedWeights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_FBGEMM_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/quantized/shufflenetv2_x2_0_fbgemm-5cac526c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5906&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">7393996</span><span class="p">,</span>
+            <span class="s2">&quot;unquantized&quot;</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">75.354</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.488</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.583</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">7.467</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_FBGEMM_V1</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x0_5"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html#torchvision.models.quantization.shufflenet_v2_x0_5">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_shufflenet_v2_x0_5&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X0_5_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x0_5</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ShuffleNet_V2_X0_5_QuantizedWeights</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 with 0.5x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X0_5_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr.</span>
+<span class="sd">            Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model.</span>
+<span class="sd">            Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X0_5_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ShuffleNet_V2_X0_5_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span>
+        <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span> <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="o">=</span><span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x1_0"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html#torchvision.models.quantization.shufflenet_v2_x1_0">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_shufflenet_v2_x1_0&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X1_0_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x1_0</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ShuffleNet_V2_X1_0_QuantizedWeights</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 with 1.0x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X1_0_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr.</span>
+<span class="sd">            Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model.</span>
+<span class="sd">            Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X1_0_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ShuffleNet_V2_X1_0_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span>
+        <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">116</span><span class="p">,</span> <span class="mi">232</span><span class="p">,</span> <span class="mi">464</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span> <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="o">=</span><span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x1_5"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html#torchvision.models.quantization.shufflenet_v2_x1_5">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_shufflenet_v2_x1_5&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X1_5_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x1_5</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ShuffleNet_V2_X1_5_QuantizedWeights</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 with 1.5x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X1_5_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr.</span>
+<span class="sd">            Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model.</span>
+<span class="sd">            Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X1_5_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ShuffleNet_V2_X1_5_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span>
+        <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">176</span><span class="p">,</span> <span class="mi">352</span><span class="p">,</span> <span class="mi">704</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span> <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="o">=</span><span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x2_0"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html#torchvision.models.quantization.shufflenet_v2_x2_0">[docs]</a><span class="nd">@register_model</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">&quot;quantized_shufflenet_v2_x2_0&quot;</span><span class="p">)</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span>
+        <span class="s2">&quot;pretrained&quot;</span><span class="p">,</span>
+        <span class="k">lambda</span> <span class="n">kwargs</span><span class="p">:</span> <span class="n">ShuffleNet_V2_X2_0_QuantizedWeights</span><span class="o">.</span><span class="n">IMAGENET1K_FBGEMM_V1</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;quantize&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="k">else</span> <span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x2_0</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">ShuffleNet_V2_X2_0_QuantizedWeights</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">quantize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantizableShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 with 2.0x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Note that ``quantize = True`` returns a quantized model with 8 bit</span>
+<span class="sd">        weights. Quantized models only support inference and run on CPUs.</span>
+<span class="sd">        GPU inference is not yet supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights` or :class:`~torchvision.models.ShuffleNet_V2_X2_0_Weights`, optional): The</span>
+<span class="sd">            pretrained weights for the model. See</span>
+<span class="sd">            :class:`~torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr.</span>
+<span class="sd">            Default is True.</span>
+<span class="sd">        quantize (bool, optional): If True, return a quantized version of the model.</span>
+<span class="sd">            Default is False.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights</span>
+<span class="sd">        :members:</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X2_0_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">        :noindex:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="p">(</span><span class="n">ShuffleNet_V2_X2_0_QuantizedWeights</span> <span class="k">if</span> <span class="n">quantize</span> <span class="k">else</span> <span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="p">)</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span>
+        <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">244</span><span class="p">,</span> <span class="mi">488</span><span class="p">,</span> <span class="mi">976</span><span class="p">,</span> <span class="mi">2048</span><span class="p">],</span> <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">quantize</span><span class="o">=</span><span class="n">quantize</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/regnet.html b/0.16/_modules/torchvision/models/regnet.html
new file mode 100644
index 00000000000..e089328b13e
--- /dev/null
+++ b/0.16/_modules/torchvision/models/regnet.html
@@ -0,0 +1,2240 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.regnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.regnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.regnet</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span><span class="p">,</span> <span class="n">SqueezeExcitation</span>
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">InterpolationMode</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_make_divisible</span><span class="p">,</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;RegNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_400MF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_800MF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_1_6GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_3_2GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_8GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_16GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_32GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_Y_128GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_X_400MF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_X_800MF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_X_1_6GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_X_3_2GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_X_8GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_X_16GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RegNet_X_32GF_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_400mf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_800mf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_1_6gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_3_2gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_8gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_16gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_32gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_y_128gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_x_400mf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_x_800mf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_x_1_6gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_x_3_2gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_x_8gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_x_16gf&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;regnet_x_32gf&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">SimpleStemIN</span><span class="p">(</span><span class="n">Conv2dNormActivation</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Simple stem for ImageNet: 3x3, BN, ReLU.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">width_in</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">width_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">width_in</span><span class="p">,</span> <span class="n">width_out</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">BottleneckTransform</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Bottleneck transformation: 1x1, 3x3 [+SE], 1x1.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">width_in</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">width_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">group_width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">bottleneck_multiplier</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">se_ratio</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">layers</span><span class="p">:</span> <span class="n">OrderedDict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+        <span class="n">w_b</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">width_out</span> <span class="o">*</span> <span class="n">bottleneck_multiplier</span><span class="p">))</span>
+        <span class="n">g</span> <span class="o">=</span> <span class="n">w_b</span> <span class="o">//</span> <span class="n">group_width</span>
+
+        <span class="n">layers</span><span class="p">[</span><span class="s2">&quot;a&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">width_in</span><span class="p">,</span> <span class="n">w_b</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span>
+        <span class="p">)</span>
+        <span class="n">layers</span><span class="p">[</span><span class="s2">&quot;b&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">w_b</span><span class="p">,</span> <span class="n">w_b</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">g</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">activation_layer</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">se_ratio</span><span class="p">:</span>
+            <span class="c1"># The SE reduction ratio is defined with respect to the</span>
+            <span class="c1"># beginning of the block</span>
+            <span class="n">width_se_out</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">se_ratio</span> <span class="o">*</span> <span class="n">width_in</span><span class="p">))</span>
+            <span class="n">layers</span><span class="p">[</span><span class="s2">&quot;se&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">SqueezeExcitation</span><span class="p">(</span>
+                <span class="n">input_channels</span><span class="o">=</span><span class="n">w_b</span><span class="p">,</span>
+                <span class="n">squeeze_channels</span><span class="o">=</span><span class="n">width_se_out</span><span class="p">,</span>
+                <span class="n">activation</span><span class="o">=</span><span class="n">activation_layer</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="n">layers</span><span class="p">[</span><span class="s2">&quot;c&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+            <span class="n">w_b</span><span class="p">,</span> <span class="n">width_out</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span>
+        <span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">layers</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ResBottleneckBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Residual bottleneck block: x + F(x), F = bottleneck transform.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">width_in</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">width_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">group_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">bottleneck_multiplier</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">se_ratio</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="c1"># Use skip connection with projection if shape changes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">should_proj</span> <span class="o">=</span> <span class="p">(</span><span class="n">width_in</span> <span class="o">!=</span> <span class="n">width_out</span><span class="p">)</span> <span class="ow">or</span> <span class="p">(</span><span class="n">stride</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">should_proj</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">width_in</span><span class="p">,</span> <span class="n">width_out</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">f</span> <span class="o">=</span> <span class="n">BottleneckTransform</span><span class="p">(</span>
+            <span class="n">width_in</span><span class="p">,</span>
+            <span class="n">width_out</span><span class="p">,</span>
+            <span class="n">stride</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">group_width</span><span class="p">,</span>
+            <span class="n">bottleneck_multiplier</span><span class="p">,</span>
+            <span class="n">se_ratio</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">activation</span> <span class="o">=</span> <span class="n">activation_layer</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">f</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">f</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">activation</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">AnyStage</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;AnyNet stage (sequence of blocks w/ the same output shape).&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">width_in</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">width_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">depth</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">block_constructor</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">group_width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">bottleneck_multiplier</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">se_ratio</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">stage_index</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">depth</span><span class="p">):</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">block_constructor</span><span class="p">(</span>
+                <span class="n">width_in</span> <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">width_out</span><span class="p">,</span>
+                <span class="n">width_out</span><span class="p">,</span>
+                <span class="n">stride</span> <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="mi">1</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="p">,</span>
+                <span class="n">activation_layer</span><span class="p">,</span>
+                <span class="n">group_width</span><span class="p">,</span>
+                <span class="n">bottleneck_multiplier</span><span class="p">,</span>
+                <span class="n">se_ratio</span><span class="p">,</span>
+            <span class="p">)</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">add_module</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;block</span><span class="si">{</span><span class="n">stage_index</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="n">block</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">BlockParams</span><span class="p">:</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">depths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">widths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">group_widths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">bottleneck_multipliers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+        <span class="n">strides</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">se_ratio</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">depths</span> <span class="o">=</span> <span class="n">depths</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">widths</span> <span class="o">=</span> <span class="n">widths</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">group_widths</span> <span class="o">=</span> <span class="n">group_widths</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bottleneck_multipliers</span> <span class="o">=</span> <span class="n">bottleneck_multipliers</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">strides</span> <span class="o">=</span> <span class="n">strides</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">se_ratio</span> <span class="o">=</span> <span class="n">se_ratio</span>
+
+    <span class="nd">@classmethod</span>
+    <span class="k">def</span> <span class="nf">from_init_params</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">depth</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">w_0</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">w_a</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">w_m</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">group_width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">bottleneck_multiplier</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">se_ratio</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="s2">&quot;BlockParams&quot;</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Programmatically compute all the per-block settings,</span>
+<span class="sd">        given the RegNet parameters.</span>
+
+<span class="sd">        The first step is to compute the quantized linear block parameters,</span>
+<span class="sd">        in log space. Key parameters are:</span>
+<span class="sd">        - `w_a` is the width progression slope</span>
+<span class="sd">        - `w_0` is the initial width</span>
+<span class="sd">        - `w_m` is the width stepping in the log space</span>
+
+<span class="sd">        In other terms</span>
+<span class="sd">        `log(block_width) = log(w_0) + w_m * block_capacity`,</span>
+<span class="sd">        with `bock_capacity` ramping up following the w_0 and w_a params.</span>
+<span class="sd">        This block width is finally quantized to multiples of 8.</span>
+
+<span class="sd">        The second step is to compute the parameters per stage,</span>
+<span class="sd">        taking into account the skip connection and the final 1x1 convolutions.</span>
+<span class="sd">        We use the fact that the output width is constant within a stage.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">QUANT</span> <span class="o">=</span> <span class="mi">8</span>
+        <span class="n">STRIDE</span> <span class="o">=</span> <span class="mi">2</span>
+
+        <span class="k">if</span> <span class="n">w_a</span> <span class="o">&lt;</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">w_0</span> <span class="o">&lt;=</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">w_m</span> <span class="o">&lt;=</span> <span class="mi">1</span> <span class="ow">or</span> <span class="n">w_0</span> <span class="o">%</span> <span class="mi">8</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid RegNet settings&quot;</span><span class="p">)</span>
+        <span class="c1"># Compute the block widths. Each stage has one unique block width</span>
+        <span class="n">widths_cont</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">depth</span><span class="p">)</span> <span class="o">*</span> <span class="n">w_a</span> <span class="o">+</span> <span class="n">w_0</span>
+        <span class="n">block_capacity</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">round</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">widths_cont</span> <span class="o">/</span> <span class="n">w_0</span><span class="p">)</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">w_m</span><span class="p">))</span>
+        <span class="n">block_widths</span> <span class="o">=</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">round</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">divide</span><span class="p">(</span><span class="n">w_0</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">pow</span><span class="p">(</span><span class="n">w_m</span><span class="p">,</span> <span class="n">block_capacity</span><span class="p">),</span> <span class="n">QUANT</span><span class="p">))</span> <span class="o">*</span> <span class="n">QUANT</span><span class="p">)</span><span class="o">.</span><span class="n">int</span><span class="p">()</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+        <span class="n">num_stages</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">block_widths</span><span class="p">))</span>
+
+        <span class="c1"># Convert to per stage parameters</span>
+        <span class="n">split_helper</span> <span class="o">=</span> <span class="nb">zip</span><span class="p">(</span>
+            <span class="n">block_widths</span> <span class="o">+</span> <span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">block_widths</span><span class="p">,</span>
+            <span class="n">block_widths</span> <span class="o">+</span> <span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">block_widths</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">splits</span> <span class="o">=</span> <span class="p">[</span><span class="n">w</span> <span class="o">!=</span> <span class="n">wp</span> <span class="ow">or</span> <span class="n">r</span> <span class="o">!=</span> <span class="n">rp</span> <span class="k">for</span> <span class="n">w</span><span class="p">,</span> <span class="n">wp</span><span class="p">,</span> <span class="n">r</span><span class="p">,</span> <span class="n">rp</span> <span class="ow">in</span> <span class="n">split_helper</span><span class="p">]</span>
+
+        <span class="n">stage_widths</span> <span class="o">=</span> <span class="p">[</span><span class="n">w</span> <span class="k">for</span> <span class="n">w</span><span class="p">,</span> <span class="n">t</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">block_widths</span><span class="p">,</span> <span class="n">splits</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span> <span class="k">if</span> <span class="n">t</span><span class="p">]</span>
+        <span class="n">stage_depths</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">diff</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">d</span> <span class="k">for</span> <span class="n">d</span><span class="p">,</span> <span class="n">t</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">splits</span><span class="p">)</span> <span class="k">if</span> <span class="n">t</span><span class="p">]))</span><span class="o">.</span><span class="n">int</span><span class="p">()</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+
+        <span class="n">strides</span> <span class="o">=</span> <span class="p">[</span><span class="n">STRIDE</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_stages</span>
+        <span class="n">bottleneck_multipliers</span> <span class="o">=</span> <span class="p">[</span><span class="n">bottleneck_multiplier</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_stages</span>
+        <span class="n">group_widths</span> <span class="o">=</span> <span class="p">[</span><span class="n">group_width</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_stages</span>
+
+        <span class="c1"># Adjust the compatibility of stage widths and group widths</span>
+        <span class="n">stage_widths</span><span class="p">,</span> <span class="n">group_widths</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_adjust_widths_groups_compatibilty</span><span class="p">(</span>
+            <span class="n">stage_widths</span><span class="p">,</span> <span class="n">bottleneck_multipliers</span><span class="p">,</span> <span class="n">group_widths</span>
+        <span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">cls</span><span class="p">(</span>
+            <span class="n">depths</span><span class="o">=</span><span class="n">stage_depths</span><span class="p">,</span>
+            <span class="n">widths</span><span class="o">=</span><span class="n">stage_widths</span><span class="p">,</span>
+            <span class="n">group_widths</span><span class="o">=</span><span class="n">group_widths</span><span class="p">,</span>
+            <span class="n">bottleneck_multipliers</span><span class="o">=</span><span class="n">bottleneck_multipliers</span><span class="p">,</span>
+            <span class="n">strides</span><span class="o">=</span><span class="n">strides</span><span class="p">,</span>
+            <span class="n">se_ratio</span><span class="o">=</span><span class="n">se_ratio</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_expanded_params</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">widths</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">strides</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">depths</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">group_widths</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">bottleneck_multipliers</span><span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">_adjust_widths_groups_compatibilty</span><span class="p">(</span>
+        <span class="n">stage_widths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">bottleneck_ratios</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">group_widths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Adjusts the compatibility of widths and groups,</span>
+<span class="sd">        depending on the bottleneck ratio.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Compute all widths for the current settings</span>
+        <span class="n">widths</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">w</span> <span class="o">*</span> <span class="n">b</span><span class="p">)</span> <span class="k">for</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">stage_widths</span><span class="p">,</span> <span class="n">bottleneck_ratios</span><span class="p">)]</span>
+        <span class="n">group_widths_min</span> <span class="o">=</span> <span class="p">[</span><span class="nb">min</span><span class="p">(</span><span class="n">g</span><span class="p">,</span> <span class="n">w_bot</span><span class="p">)</span> <span class="k">for</span> <span class="n">g</span><span class="p">,</span> <span class="n">w_bot</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">group_widths</span><span class="p">,</span> <span class="n">widths</span><span class="p">)]</span>
+
+        <span class="c1"># Compute the adjusted widths so that stage and group widths fit</span>
+        <span class="n">ws_bot</span> <span class="o">=</span> <span class="p">[</span><span class="n">_make_divisible</span><span class="p">(</span><span class="n">w_bot</span><span class="p">,</span> <span class="n">g</span><span class="p">)</span> <span class="k">for</span> <span class="n">w_bot</span><span class="p">,</span> <span class="n">g</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">widths</span><span class="p">,</span> <span class="n">group_widths_min</span><span class="p">)]</span>
+        <span class="n">stage_widths</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">w_bot</span> <span class="o">/</span> <span class="n">b</span><span class="p">)</span> <span class="k">for</span> <span class="n">w_bot</span><span class="p">,</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">ws_bot</span><span class="p">,</span> <span class="n">bottleneck_ratios</span><span class="p">)]</span>
+        <span class="k">return</span> <span class="n">stage_widths</span><span class="p">,</span> <span class="n">group_widths_min</span>
+
+
+<span class="k">class</span> <span class="nc">RegNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">block_params</span><span class="p">:</span> <span class="n">BlockParams</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">stem_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">32</span><span class="p">,</span>
+        <span class="n">stem_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">block_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">activation</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">stem_type</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">stem_type</span> <span class="o">=</span> <span class="n">SimpleStemIN</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+        <span class="k">if</span> <span class="n">block_type</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block_type</span> <span class="o">=</span> <span class="n">ResBottleneckBlock</span>
+        <span class="k">if</span> <span class="n">activation</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">activation</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span>
+
+        <span class="c1"># Ad hoc stem</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stem</span> <span class="o">=</span> <span class="n">stem_type</span><span class="p">(</span>
+            <span class="mi">3</span><span class="p">,</span>  <span class="c1"># width_in</span>
+            <span class="n">stem_width</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">activation</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="n">current_width</span> <span class="o">=</span> <span class="n">stem_width</span>
+
+        <span class="n">blocks</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span>
+            <span class="n">width_out</span><span class="p">,</span>
+            <span class="n">stride</span><span class="p">,</span>
+            <span class="n">depth</span><span class="p">,</span>
+            <span class="n">group_width</span><span class="p">,</span>
+            <span class="n">bottleneck_multiplier</span><span class="p">,</span>
+        <span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">block_params</span><span class="o">.</span><span class="n">_get_expanded_params</span><span class="p">()):</span>
+            <span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;block</span><span class="si">{</span><span class="n">i</span><span class="o">+</span><span class="mi">1</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+                    <span class="n">AnyStage</span><span class="p">(</span>
+                        <span class="n">current_width</span><span class="p">,</span>
+                        <span class="n">width_out</span><span class="p">,</span>
+                        <span class="n">stride</span><span class="p">,</span>
+                        <span class="n">depth</span><span class="p">,</span>
+                        <span class="n">block_type</span><span class="p">,</span>
+                        <span class="n">norm_layer</span><span class="p">,</span>
+                        <span class="n">activation</span><span class="p">,</span>
+                        <span class="n">group_width</span><span class="p">,</span>
+                        <span class="n">bottleneck_multiplier</span><span class="p">,</span>
+                        <span class="n">block_params</span><span class="o">.</span><span class="n">se_ratio</span><span class="p">,</span>
+                        <span class="n">stage_index</span><span class="o">=</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span>
+                    <span class="p">),</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+
+            <span class="n">current_width</span> <span class="o">=</span> <span class="n">width_out</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">trunk_output</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">OrderedDict</span><span class="p">(</span><span class="n">blocks</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_features</span><span class="o">=</span><span class="n">current_width</span><span class="p">,</span> <span class="n">out_features</span><span class="o">=</span><span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="c1"># Performs ResNet-style weight initialization</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="c1"># Note that there is no bias due to BN</span>
+                <span class="n">fan_out</span> <span class="o">=</span> <span class="n">m</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">m</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">m</span><span class="o">.</span><span class="n">out_channels</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mf">2.0</span> <span class="o">/</span> <span class="n">fan_out</span><span class="p">))</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">ones_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stem</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">trunk_output</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">start_dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">def</span> <span class="nf">_regnet</span><span class="p">(</span>
+    <span class="n">block_params</span><span class="p">:</span> <span class="n">BlockParams</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;norm_layer&quot;</span><span class="p">,</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-05</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.1</span><span class="p">))</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">RegNet</span><span class="p">(</span><span class="n">block_params</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+<span class="p">}</span>
+
+<span class="n">_COMMON_SWAG_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/facebookresearch/SWAG&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;license&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/facebookresearch/SWAG/blob/main/LICENSE&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="RegNet_Y_400MF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_400mf.html#torchvision.models.RegNet_Y_400MF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_400MF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_400mf-c65dace8.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">4344144</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#small-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">74.046</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.716</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.402</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">16.806</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_400mf-e6988f5f.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">4344144</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">75.804</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.742</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.402</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">16.806</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_Y_800MF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_800mf.html#torchvision.models.RegNet_Y_800MF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_800MF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_800mf-1b27b58c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">6432512</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#small-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">76.420</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.136</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.834</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">24.774</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_800mf-58fc7688.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">6432512</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.828</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.502</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.834</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">24.774</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_Y_1_6GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_1_6gf.html#torchvision.models.RegNet_Y_1_6GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_1_6GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_1_6gf-b11a554e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">11202430</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#small-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.950</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.966</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.612</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">43.152</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_1_6gf-0d7bc02a.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">11202430</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.876</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.444</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.612</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">43.152</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_Y_3_2GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_3_2gf.html#torchvision.models.RegNet_Y_3_2GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_3_2GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_3_2gf-b5a9779c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">19436338</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#medium-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.948</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.576</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">3.176</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">74.567</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_3_2gf-9180c971.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">19436338</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.982</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.972</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">3.176</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">74.567</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_Y_8GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_8gf.html#torchvision.models.RegNet_Y_8GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_8GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_8gf-d0d0e4a8.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">39381472</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#medium-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.032</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.048</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">8.473</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">150.701</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_8gf-dc2b1b54.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">39381472</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.828</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.330</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">8.473</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">150.701</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_Y_16GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.RegNet_Y_16GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_16GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_16gf-9e6ed7dd.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">83590140</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#large-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.424</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.240</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.912</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">319.49</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_16gf-3e4a00f9.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">83590140</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.886</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.328</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.912</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">319.49</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_E2E_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_16gf_swag-43afe44d.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">83590140</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">86.012</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">98.054</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">46.735</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">319.49</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are learnt via transfer learning by end-to-end fine-tuning the original</span>
+<span class="s2">                `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ weights on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_LINEAR_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_16gf_lc_swag-f3ec0043.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5793&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">83590140</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.976</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.244</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.912</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">319.49</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are composed of the original frozen `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ trunk</span>
+<span class="s2">                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_Y_32GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.RegNet_Y_32GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_32GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_32gf-4dee3f7a.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">145046770</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#large-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.878</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.340</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">32.28</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">554.076</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_32gf-8db6d4b5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">145046770</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.368</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.498</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">32.28</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">554.076</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_E2E_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_32gf_swag-04fdfa75.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">145046770</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">86.838</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">98.362</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">94.826</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">554.076</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are learnt via transfer learning by end-to-end fine-tuning the original</span>
+<span class="s2">                `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ weights on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_LINEAR_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_32gf_lc_swag-e1583746.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5793&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">145046770</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">84.622</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.480</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">32.28</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">554.076</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are composed of the original frozen `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ trunk</span>
+<span class="s2">                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_Y_128GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_128gf.html#torchvision.models.RegNet_Y_128GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_Y_128GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_SWAG_E2E_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_128gf_swag-c8ce3e52.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">644812894</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">88.228</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">98.682</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">374.57</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">2461.564</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are learnt via transfer learning by end-to-end fine-tuning the original</span>
+<span class="s2">                `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ weights on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_LINEAR_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_y_128gf_lc_swag-cbe8ce12.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5793&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">644812894</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">86.068</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.844</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">127.518</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">2461.564</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are composed of the original frozen `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ trunk</span>
+<span class="s2">                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_SWAG_E2E_V1</span></div>
+
+
+<div class="viewcode-block" id="RegNet_X_400MF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_400mf.html#torchvision.models.RegNet_X_400MF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_X_400MF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_400mf-adf1edd5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5495976</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#small-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">72.834</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.950</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.414</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">21.258</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_400mf-62229a5f.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">5495976</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">74.864</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.322</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.414</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">21.257</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_X_800MF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_800mf.html#torchvision.models.RegNet_X_800MF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_X_800MF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_800mf-ad17e45c.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">7259656</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#small-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">75.212</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.348</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.8</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">27.945</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_800mf-94a99ebd.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">7259656</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.522</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.826</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.8</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">27.945</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_X_1_6GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_1_6gf.html#torchvision.models.RegNet_X_1_6GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_X_1_6GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_1_6gf-e3633e7f.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">9190136</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#small-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.040</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.440</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.603</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">35.339</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_1_6gf-a12f2b72.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">9190136</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">79.668</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.922</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.603</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">35.339</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_X_3_2GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_3_2gf.html#torchvision.models.RegNet_X_3_2GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_X_3_2GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_3_2gf-f342aeae.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">15296552</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#medium-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.364</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.992</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">3.177</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">58.756</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_3_2gf-7071aa85.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">15296552</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.196</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.430</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">3.177</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">58.756</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_X_8GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_8gf.html#torchvision.models.RegNet_X_8GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_X_8GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_8gf-03ceed89.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">39572648</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#medium-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">79.344</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.686</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">7.995</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">151.456</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_8gf-2b70d774.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">39572648</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.682</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.678</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">7.995</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">151.456</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_X_16GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_16gf.html#torchvision.models.RegNet_X_16GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_X_16GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_16gf-2007eb11.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">54278536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#medium-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.058</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.944</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.941</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">207.627</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_16gf-ba3796d7.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">54278536</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.716</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.196</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.941</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">207.627</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="RegNet_X_32GF_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_32gf.html#torchvision.models.RegNet_X_32GF_Weights">[docs]</a><span class="k">class</span> <span class="nc">RegNet_X_32GF_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_32gf-9d47f8d0.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">107811560</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#large-models&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.622</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.248</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">31.736</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">412.039</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/regnet_x_32gf-6eb8fdc6.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">107811560</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.014</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.288</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">31.736</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">412.039</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_400mf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_400mf.html#torchvision.models.regnet_y_400mf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_Y_400MF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_400mf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_400MF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_400MF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_400MF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_400MF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_400MF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_400MF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">48</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">27.89</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.09</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_800mf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_800mf.html#torchvision.models.regnet_y_800mf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_Y_800MF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_800mf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_800MF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_800MF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_800MF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_800MF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_800MF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_800MF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">14</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">56</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">38.84</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.4</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_1_6gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_1_6gf.html#torchvision.models.regnet_y_1_6gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_Y_1_6GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_1_6gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_1_6GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_1.6GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_1_6GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_1_6GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_1_6GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_1_6GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span>
+        <span class="n">depth</span><span class="o">=</span><span class="mi">27</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">48</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">20.71</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.65</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">24</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_3_2gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_3_2gf.html#torchvision.models.regnet_y_3_2gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_Y_3_2GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_3_2gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_3_2GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_3.2GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_3_2GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_3_2GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_3_2GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_3_2GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span>
+        <span class="n">depth</span><span class="o">=</span><span class="mi">21</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">80</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">42.63</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.66</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">24</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_8gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_8gf.html#torchvision.models.regnet_y_8gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_Y_8GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_8gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_8GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_8GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_8GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_8GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_8GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_8GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span>
+        <span class="n">depth</span><span class="o">=</span><span class="mi">17</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">192</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">76.82</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.19</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">56</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_16gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.regnet_y_16gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_Y_16GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_16gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_16GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_16GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_16GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_16GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_16GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_16GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span>
+        <span class="n">depth</span><span class="o">=</span><span class="mi">18</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">200</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">106.23</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.48</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">112</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_32gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.regnet_y_32gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_Y_32GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_32gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_32GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_32GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_32GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_32GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_32GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_32GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span>
+        <span class="n">depth</span><span class="o">=</span><span class="mi">20</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">232</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">115.89</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.53</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">232</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_y_128gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_y_128gf.html#torchvision.models.regnet_y_128gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_y_128gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_Y_128GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetY_128GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_Y_128GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_Y_128GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_Y_128GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_Y_128GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span>
+        <span class="n">depth</span><span class="o">=</span><span class="mi">27</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">456</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">160.83</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.52</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">264</span><span class="p">,</span> <span class="n">se_ratio</span><span class="o">=</span><span class="mf">0.25</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_x_400mf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_400mf.html#torchvision.models.regnet_x_400mf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_X_400MF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_x_400mf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_X_400MF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetX_400MF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_X_400MF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_X_400MF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_X_400MF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_X_400MF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">22</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">24</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">24.48</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.54</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_x_800mf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_800mf.html#torchvision.models.regnet_x_800mf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_X_800MF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_x_800mf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_X_800MF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetX_800MF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_X_800MF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_X_800MF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_X_800MF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_X_800MF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">56</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">35.73</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.28</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_x_1_6gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_1_6gf.html#torchvision.models.regnet_x_1_6gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_X_1_6GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_x_1_6gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_X_1_6GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetX_1.6GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_X_1_6GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_X_1_6GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_X_1_6GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_X_1_6GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">18</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">80</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">34.01</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.25</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">24</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_x_3_2gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_3_2gf.html#torchvision.models.regnet_x_3_2gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_X_3_2GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_x_3_2gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_X_3_2GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetX_3.2GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_X_3_2GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_X_3_2GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_X_3_2GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_X_3_2GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">25</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">88</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">26.31</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.25</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">48</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_x_8gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_8gf.html#torchvision.models.regnet_x_8gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_X_8GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_x_8gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_X_8GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetX_8GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_X_8GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_X_8GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_X_8GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_X_8GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">23</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">80</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">49.56</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.88</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_x_16gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_16gf.html#torchvision.models.regnet_x_16gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_X_16GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_x_16gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_X_16GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetX_16GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_X_16GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_X_16GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_X_16GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_X_16GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">22</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">216</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">55.59</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.1</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="regnet_x_32gf"><a class="viewcode-back" href="../../../models/generated/torchvision.models.regnet_x_32gf.html#torchvision.models.regnet_x_32gf">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">RegNet_X_32GF_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">regnet_x_32gf</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">RegNet_X_32GF_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">RegNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a RegNetX_32GF architecture from</span>
+<span class="sd">    `Designing Network Design Spaces &lt;https://arxiv.org/abs/2003.13678&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.RegNet_X_32GF_Weights`, optional): The pretrained weights to use.</span>
+<span class="sd">            See :class:`~torchvision.models.RegNet_X_32GF_Weights` below for more details and possible values.</span>
+<span class="sd">            By default, no pretrained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to either ``torchvision.models.regnet.RegNet`` or</span>
+<span class="sd">            ``torchvision.models.regnet.BlockParams`` class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py&gt;`_</span>
+<span class="sd">            for more detail about the classes.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.RegNet_X_32GF_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">RegNet_X_32GF_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">params</span> <span class="o">=</span> <span class="n">BlockParams</span><span class="o">.</span><span class="n">from_init_params</span><span class="p">(</span><span class="n">depth</span><span class="o">=</span><span class="mi">23</span><span class="p">,</span> <span class="n">w_0</span><span class="o">=</span><span class="mi">320</span><span class="p">,</span> <span class="n">w_a</span><span class="o">=</span><span class="mf">69.86</span><span class="p">,</span> <span class="n">w_m</span><span class="o">=</span><span class="mf">2.0</span><span class="p">,</span> <span class="n">group_width</span><span class="o">=</span><span class="mi">168</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_regnet</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/resnet.html b/0.16/_modules/torchvision/models/resnet.html
new file mode 100644
index 00000000000..469bca1f32e
--- /dev/null
+++ b/0.16/_modules/torchvision/models/resnet.html
@@ -0,0 +1,1654 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.resnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.resnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.resnet</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;ResNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNet18_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNet34_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNet50_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNet101_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNet152_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNeXt50_32X4D_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNeXt101_32X8D_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ResNeXt101_64X4D_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Wide_ResNet50_2_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Wide_ResNet101_2_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnet18&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnet34&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnet50&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnet101&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnet152&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnext50_32x4d&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnext101_32x8d&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;resnext101_64x4d&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;wide_resnet50_2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;wide_resnet101_2&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">conv3x3</span><span class="p">(</span><span class="n">in_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="n">dilation</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;3x3 convolution with padding&quot;&quot;&quot;</span>
+    <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span>
+        <span class="n">in_planes</span><span class="p">,</span>
+        <span class="n">out_planes</span><span class="p">,</span>
+        <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
+        <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+        <span class="n">padding</span><span class="o">=</span><span class="n">dilation</span><span class="p">,</span>
+        <span class="n">groups</span><span class="o">=</span><span class="n">groups</span><span class="p">,</span>
+        <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="o">=</span><span class="n">dilation</span><span class="p">,</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">conv1x1</span><span class="p">(</span><span class="n">in_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;1x1 convolution&quot;&quot;&quot;</span>
+    <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_planes</span><span class="p">,</span> <span class="n">out_planes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">BasicBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="n">expansion</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inplanes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">downsample</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">base_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">64</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+        <span class="k">if</span> <span class="n">groups</span> <span class="o">!=</span> <span class="mi">1</span> <span class="ow">or</span> <span class="n">base_width</span> <span class="o">!=</span> <span class="mi">64</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;BasicBlock only supports groups=1 and base_width=64&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">dilation</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;Dilation &gt; 1 not supported in BasicBlock&quot;</span><span class="p">)</span>
+        <span class="c1"># Both self.conv1 and self.downsample layers downsample the input when stride != 1</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">conv3x3</span><span class="p">(</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">stride</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">planes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">conv3x3</span><span class="p">(</span><span class="n">planes</span><span class="p">,</span> <span class="n">planes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn2</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">planes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">downsample</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">identity</span> <span class="o">=</span> <span class="n">x</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">identity</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">+=</span> <span class="n">identity</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">Bottleneck</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="c1"># Bottleneck in torchvision places the stride for downsampling at 3x3 convolution(self.conv2)</span>
+    <span class="c1"># while original implementation places the stride at the first 1x1 convolution(self.conv1)</span>
+    <span class="c1"># according to &quot;Deep residual learning for image recognition&quot; https://arxiv.org/abs/1512.03385.</span>
+    <span class="c1"># This variant is also known as ResNet V1.5 and improves accuracy according to</span>
+    <span class="c1"># https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch.</span>
+
+    <span class="n">expansion</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inplanes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">downsample</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">base_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">64</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+        <span class="n">width</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">planes</span> <span class="o">*</span> <span class="p">(</span><span class="n">base_width</span> <span class="o">/</span> <span class="mf">64.0</span><span class="p">))</span> <span class="o">*</span> <span class="n">groups</span>
+        <span class="c1"># Both self.conv2 and self.downsample layers downsample the input when stride != 1</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">conv1x1</span><span class="p">(</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">width</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">conv3x3</span><span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">groups</span><span class="p">,</span> <span class="n">dilation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn2</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">width</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span> <span class="o">=</span> <span class="n">conv1x1</span><span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">planes</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">expansion</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn3</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">planes</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">expansion</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">downsample</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">identity</span> <span class="o">=</span> <span class="n">x</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn3</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">identity</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">+=</span> <span class="n">identity</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">ResNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">]],</span>
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">zero_init_residual</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">width_per_group</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">64</span><span class="p">,</span>
+        <span class="n">replace_stride_with_dilation</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_norm_layer</span> <span class="o">=</span> <span class="n">norm_layer</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span> <span class="o">=</span> <span class="mi">64</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dilation</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="n">replace_stride_with_dilation</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># each element in the tuple indicates if we should replace</span>
+            <span class="c1"># the 2x2 stride with a dilated convolution instead</span>
+            <span class="n">replace_stride_with_dilation</span> <span class="o">=</span> <span class="p">[</span><span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">False</span><span class="p">]</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">replace_stride_with_dilation</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;replace_stride_with_dilation should be None &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;or a 3-element tuple, got </span><span class="si">{</span><span class="n">replace_stride_with_dilation</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">groups</span> <span class="o">=</span> <span class="n">groups</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">base_width</span> <span class="o">=</span> <span class="n">width_per_group</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">dilate</span><span class="o">=</span><span class="n">replace_stride_with_dilation</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">dilate</span><span class="o">=</span><span class="n">replace_stride_with_dilation</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer4</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">dilate</span><span class="o">=</span><span class="n">replace_stride_with_dilation</span><span class="p">[</span><span class="mi">2</span><span class="p">])</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">512</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">GroupNorm</span><span class="p">)):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="c1"># Zero-initialize the last BN in each residual branch,</span>
+        <span class="c1"># so that the residual branch starts with zeros, and each residual block behaves like an identity.</span>
+        <span class="c1"># This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677</span>
+        <span class="k">if</span> <span class="n">zero_init_residual</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">)</span> <span class="ow">and</span> <span class="n">m</span><span class="o">.</span><span class="n">bn3</span><span class="o">.</span><span class="n">weight</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bn3</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>  <span class="c1"># type: ignore[arg-type]</span>
+                <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">BasicBlock</span><span class="p">)</span> <span class="ow">and</span> <span class="n">m</span><span class="o">.</span><span class="n">bn2</span><span class="o">.</span><span class="n">weight</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bn2</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>  <span class="c1"># type: ignore[arg-type]</span>
+
+    <span class="k">def</span> <span class="nf">_make_layer</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">]],</span>
+        <span class="n">planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">blocks</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">dilate</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">:</span>
+        <span class="n">norm_layer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_norm_layer</span>
+        <span class="n">downsample</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">previous_dilation</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dilation</span>
+        <span class="k">if</span> <span class="n">dilate</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">dilation</span> <span class="o">*=</span> <span class="n">stride</span>
+            <span class="n">stride</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="n">stride</span> <span class="o">!=</span> <span class="mi">1</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span> <span class="o">!=</span> <span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">:</span>
+            <span class="n">downsample</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                <span class="n">conv1x1</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">,</span> <span class="n">stride</span><span class="p">),</span>
+                <span class="n">norm_layer</span><span class="p">(</span><span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">),</span>
+            <span class="p">)</span>
+
+        <span class="n">layers</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">block</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">downsample</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">groups</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_width</span><span class="p">,</span> <span class="n">previous_dilation</span><span class="p">,</span> <span class="n">norm_layer</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span> <span class="o">=</span> <span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">blocks</span><span class="p">):</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">block</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">,</span>
+                    <span class="n">planes</span><span class="p">,</span>
+                    <span class="n">groups</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">groups</span><span class="p">,</span>
+                    <span class="n">base_width</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">base_width</span><span class="p">,</span>
+                    <span class="n">dilation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dilation</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward_impl</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># See note [TorchScript super()]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bn1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer4</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_resnet</span><span class="p">(</span>
+    <span class="n">block</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">]],</span>
+    <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">ResNet</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">layers</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="ResNet18_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet18.html#torchvision.models.ResNet18_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNet18_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet18-f37072fd.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">11689512</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#resnet&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">69.758</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">89.078</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1.814</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">44.661</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ResNet34_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet34.html#torchvision.models.ResNet34_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNet34_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet34-b627a593.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">21797672</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#resnet&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">73.314</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.420</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">3.664</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">83.275</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ResNet50_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNet50_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet50-0676ba61.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">25557032</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#resnet&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">76.130</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.862</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.089</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">97.781</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet50-11ad3fa6.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">25557032</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#issuecomment-1013906621&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.858</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.434</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.089</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">97.79</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="ResNet101_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNet101_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet101-63fe2227.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">44549160</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#resnet&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.374</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.546</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">7.801</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">170.511</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet101-cd907fc2.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">44549160</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.886</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.780</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">7.801</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">170.53</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="ResNet152_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet152.html#torchvision.models.ResNet152_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNet152_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet152-394f9c45.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">60192808</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#resnet&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.312</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.046</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">11.514</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">230.434</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnet152-f82ba261.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">60192808</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.284</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.002</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">11.514</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">230.474</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="ResNeXt50_32X4D_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnext50_32x4d.html#torchvision.models.ResNeXt50_32X4D_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNeXt50_32X4D_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">25028904</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#resnext&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.618</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.698</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.23</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">95.789</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnext50_32x4d-1a0047aa.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">25028904</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.198</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.340</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.23</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">95.833</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="ResNeXt101_32X8D_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnext101_32x8d.html#torchvision.models.ResNeXt101_32X8D_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNeXt101_32X8D_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88791336</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#resnext&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">79.312</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.526</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">16.414</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">339.586</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnext101_32x8d-110c445d.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88791336</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.834</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.228</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">16.414</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">339.673</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="ResNeXt101_64X4D_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnext101_64x4d.html#torchvision.models.ResNeXt101_64X4D_Weights">[docs]</a><span class="k">class</span> <span class="nc">ResNeXt101_64X4D_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/resnext101_64x4d-173b62eb.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">83455272</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5935&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.246</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.454</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.46</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">319.318</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="Wide_ResNet50_2_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.wide_resnet50_2.html#torchvision.models.Wide_ResNet50_2_Weights">[docs]</a><span class="k">class</span> <span class="nc">Wide_ResNet50_2_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/wide_resnet50_2-95faca4d.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">68883240</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/912#issue-445437439&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.468</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.086</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">11.398</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">131.82</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/wide_resnet50_2-9ba9bcbe.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">68883240</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.602</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.758</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">11.398</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">263.124</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="Wide_ResNet101_2_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.wide_resnet101_2.html#torchvision.models.Wide_ResNet101_2_Weights">[docs]</a><span class="k">class</span> <span class="nc">Wide_ResNet101_2_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/wide_resnet101_2-32ee1156.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">126886696</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/912#issue-445437439&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.848</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.284</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">22.753</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">242.896</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_V2</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/wide_resnet101_2-d733dc28.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">126886696</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/issues/3995#new-recipe&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.510</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.020</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">22.753</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">484.747</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights improve upon the results of the original paper by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V2</span></div>
+
+
+<div class="viewcode-block" id="resnet18"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet18.html#torchvision.models.resnet18">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNet18_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnet18</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet18_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNet-18 from `Deep Residual Learning for Image Recognition &lt;https://arxiv.org/abs/1512.03385&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNet18_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNet18_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNet18_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet18_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnet34"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet34.html#torchvision.models.resnet34">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNet34_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnet34</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet34_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNet-34 from `Deep Residual Learning for Image Recognition &lt;https://arxiv.org/abs/1512.03385&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNet34_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNet34_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNet34_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet34_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnet50"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet50.html#torchvision.models.resnet50">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnet50</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNet-50 from `Deep Residual Learning for Image Recognition &lt;https://arxiv.org/abs/1512.03385&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">       The bottleneck of TorchVision places the stride for downsampling to the second 3x3</span>
+<span class="sd">       convolution while the original paper places it to the first 1x1 convolution.</span>
+<span class="sd">       This variant improves the accuracy and is known as `ResNet V1.5</span>
+<span class="sd">       &lt;https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNet50_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNet50_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnet101"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet101.html#torchvision.models.resnet101">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnet101</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet101_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNet-101 from `Deep Residual Learning for Image Recognition &lt;https://arxiv.org/abs/1512.03385&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">       The bottleneck of TorchVision places the stride for downsampling to the second 3x3</span>
+<span class="sd">       convolution while the original paper places it to the first 1x1 convolution.</span>
+<span class="sd">       This variant improves the accuracy and is known as `ResNet V1.5</span>
+<span class="sd">       &lt;https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNet101_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNet101_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNet101_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">23</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnet152"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnet152.html#torchvision.models.resnet152">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNet152_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnet152</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet152_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNet-152 from `Deep Residual Learning for Image Recognition &lt;https://arxiv.org/abs/1512.03385&gt;`__.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">       The bottleneck of TorchVision places the stride for downsampling to the second 3x3</span>
+<span class="sd">       convolution while the original paper places it to the first 1x1 convolution.</span>
+<span class="sd">       This variant improves the accuracy and is known as `ResNet V1.5</span>
+<span class="sd">       &lt;https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNet152_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNet152_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ResNet152_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet152_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">36</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnext50_32x4d"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnext50_32x4d.html#torchvision.models.resnext50_32x4d">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNeXt50_32X4D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnext50_32x4d</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNeXt50_32X4D_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNeXt-50 32x4d model from</span>
+<span class="sd">    `Aggregated Residual Transformation for Deep Neural Networks &lt;https://arxiv.org/abs/1611.05431&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNeXt50_32X4D_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNext50_32X4D_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.ResNeXt50_32X4D_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNeXt50_32X4D_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;groups&quot;</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;width_per_group&quot;</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnext101_32x8d"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnext101_32x8d.html#torchvision.models.resnext101_32x8d">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNeXt101_32X8D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnext101_32x8d</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNeXt101_32X8D_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNeXt-101 32x8d model from</span>
+<span class="sd">    `Aggregated Residual Transformation for Deep Neural Networks &lt;https://arxiv.org/abs/1611.05431&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNeXt101_32X8D_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNeXt101_32X8D_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.ResNeXt101_32X8D_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNeXt101_32X8D_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;groups&quot;</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;width_per_group&quot;</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">23</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resnext101_64x4d"><a class="viewcode-back" href="../../../models/generated/torchvision.models.resnext101_64x4d.html#torchvision.models.resnext101_64x4d">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ResNeXt101_64X4D_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">resnext101_64x4d</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNeXt101_64X4D_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;ResNeXt-101 64x4d model from</span>
+<span class="sd">    `Aggregated Residual Transformation for Deep Neural Networks &lt;https://arxiv.org/abs/1611.05431&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ResNeXt101_64X4D_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ResNeXt101_64X4D_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.ResNeXt101_64X4D_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNeXt101_64X4D_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;groups&quot;</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;width_per_group&quot;</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">23</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="wide_resnet50_2"><a class="viewcode-back" href="../../../models/generated/torchvision.models.wide_resnet50_2.html#torchvision.models.wide_resnet50_2">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Wide_ResNet50_2_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">wide_resnet50_2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Wide_ResNet50_2_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Wide ResNet-50-2 model from</span>
+<span class="sd">    `Wide Residual Networks &lt;https://arxiv.org/abs/1605.07146&gt;`_.</span>
+
+<span class="sd">    The model is the same as ResNet except for the bottleneck number of channels</span>
+<span class="sd">    which is twice larger in every block. The number of channels in outer 1x1</span>
+<span class="sd">    convolutions is the same, e.g. last block in ResNet-50 has 2048-512-2048</span>
+<span class="sd">    channels, and in Wide ResNet-50-2 has 2048-1024-2048.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Wide_ResNet50_2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Wide_ResNet50_2_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.Wide_ResNet50_2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Wide_ResNet50_2_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;width_per_group&quot;</span><span class="p">,</span> <span class="mi">64</span> <span class="o">*</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="wide_resnet101_2"><a class="viewcode-back" href="../../../models/generated/torchvision.models.wide_resnet101_2.html#torchvision.models.wide_resnet101_2">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Wide_ResNet101_2_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">wide_resnet101_2</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Wide_ResNet101_2_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Wide ResNet-101-2 model from</span>
+<span class="sd">    `Wide Residual Networks &lt;https://arxiv.org/abs/1605.07146&gt;`_.</span>
+
+<span class="sd">    The model is the same as ResNet except for the bottleneck number of channels</span>
+<span class="sd">    which is twice larger in every block. The number of channels in outer 1x1</span>
+<span class="sd">    convolutions is the same, e.g. last block in ResNet-101 has 2048-512-2048</span>
+<span class="sd">    channels, and in Wide ResNet-101-2 has 2048-1024-2048.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Wide_ResNet101_2_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Wide_ResNet101_2_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.resnet.ResNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+<span class="sd">    .. autoclass:: torchvision.models.Wide_ResNet101_2_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Wide_ResNet101_2_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;width_per_group&quot;</span><span class="p">,</span> <span class="mi">64</span> <span class="o">*</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resnet</span><span class="p">(</span><span class="n">Bottleneck</span><span class="p">,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">23</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/segmentation/deeplabv3.html b/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
new file mode 100644
index 00000000000..904ec4ec9c6
--- /dev/null
+++ b/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
@@ -0,0 +1,1059 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.segmentation.deeplabv3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.segmentation.deeplabv3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.segmentation.deeplabv3</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span> <span class="nn">torch.nn</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">SemanticSegmentation</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_VOC_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span><span class="p">,</span> <span class="n">IntermediateLayerGetter</span>
+<span class="kn">from</span> <span class="nn">..mobilenetv3</span> <span class="kn">import</span> <span class="n">mobilenet_v3_large</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="p">,</span> <span class="n">MobileNetV3</span>
+<span class="kn">from</span> <span class="nn">..resnet</span> <span class="kn">import</span> <span class="n">ResNet</span><span class="p">,</span> <span class="n">resnet101</span><span class="p">,</span> <span class="n">ResNet101_Weights</span><span class="p">,</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_SimpleSegmentationModel</span>
+<span class="kn">from</span> <span class="nn">.fcn</span> <span class="kn">import</span> <span class="n">FCNHead</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;DeepLabV3&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;DeepLabV3_ResNet50_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;DeepLabV3_ResNet101_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;DeepLabV3_MobileNet_V3_Large_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;deeplabv3_mobilenet_v3_large&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;deeplabv3_resnet50&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;deeplabv3_resnet101&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">DeepLabV3</span><span class="p">(</span><span class="n">_SimpleSegmentationModel</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements DeepLabV3 model from</span>
+<span class="sd">    `&quot;Rethinking Atrous Convolution for Semantic Image Segmentation&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1706.05587&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            The backbone should return an OrderedDict[Tensor], with the key being</span>
+<span class="sd">            &quot;out&quot; for the last feature map used, and &quot;aux&quot; if an auxiliary classifier</span>
+<span class="sd">            is used.</span>
+<span class="sd">        classifier (nn.Module): module that takes the &quot;out&quot; element returned from</span>
+<span class="sd">            the backbone and returns a dense prediction.</span>
+<span class="sd">        aux_classifier (nn.Module, optional): auxiliary classifier used during training</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">pass</span>
+
+
+<span class="k">class</span> <span class="nc">DeepLabHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">ASPP</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="p">[</span><span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">36</span><span class="p">]),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="mi">256</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ASPPConv</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">dilation</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">modules</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">dilation</span><span class="p">,</span> <span class="n">dilation</span><span class="o">=</span><span class="n">dilation</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">out_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(),</span>
+        <span class="p">]</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">modules</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ASPPPooling</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">out_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(),</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+        <span class="k">for</span> <span class="n">mod</span> <span class="ow">in</span> <span class="bp">self</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">mod</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">interpolate</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;bilinear&quot;</span><span class="p">,</span> <span class="n">align_corners</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ASPP</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">atrous_rates</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">256</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">modules</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">modules</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">out_channels</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">())</span>
+        <span class="p">)</span>
+
+        <span class="n">rates</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">atrous_rates</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">rate</span> <span class="ow">in</span> <span class="n">rates</span><span class="p">:</span>
+            <span class="n">modules</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">ASPPConv</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">rate</span><span class="p">))</span>
+
+        <span class="n">modules</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">ASPPPooling</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">convs</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">(</span><span class="n">modules</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">project</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">convs</span><span class="p">)</span> <span class="o">*</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">out_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="mf">0.5</span><span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">_res</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">conv</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">convs</span><span class="p">:</span>
+            <span class="n">_res</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">_res</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">project</span><span class="p">(</span><span class="n">res</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_deeplabv3_resnet</span><span class="p">(</span>
+    <span class="n">backbone</span><span class="p">:</span> <span class="n">ResNet</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">aux</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DeepLabV3</span><span class="p">:</span>
+    <span class="n">return_layers</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;layer4&quot;</span><span class="p">:</span> <span class="s2">&quot;out&quot;</span><span class="p">}</span>
+    <span class="k">if</span> <span class="n">aux</span><span class="p">:</span>
+        <span class="n">return_layers</span><span class="p">[</span><span class="s2">&quot;layer3&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;aux&quot;</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">IntermediateLayerGetter</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">return_layers</span><span class="o">=</span><span class="n">return_layers</span><span class="p">)</span>
+
+    <span class="n">aux_classifier</span> <span class="o">=</span> <span class="n">FCNHead</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span> <span class="k">if</span> <span class="n">aux</span> <span class="k">else</span> <span class="kc">None</span>
+    <span class="n">classifier</span> <span class="o">=</span> <span class="n">DeepLabHead</span><span class="p">(</span><span class="mi">2048</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">DeepLabV3</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">classifier</span><span class="p">,</span> <span class="n">aux_classifier</span><span class="p">)</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_VOC_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">        These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC</span>
+<span class="s2">        dataset.</span>
+<span class="s2">    &quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="DeepLabV3_ResNet50_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html#torchvision.models.segmentation.DeepLabV3_ResNet50_Weights">[docs]</a><span class="k">class</span> <span class="nc">DeepLabV3_ResNet50_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_WITH_VOC_LABELS_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/deeplabv3_resnet50_coco-cd0a2569.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">SemanticSegmentation</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">520</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">42004074</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet50&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017-VOC-labels&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;miou&quot;</span><span class="p">:</span> <span class="mf">66.4</span><span class="p">,</span>
+                    <span class="s2">&quot;pixel_acc&quot;</span><span class="p">:</span> <span class="mf">92.4</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">178.722</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">160.515</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_WITH_VOC_LABELS_V1</span></div>
+
+
+<div class="viewcode-block" id="DeepLabV3_ResNet101_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html#torchvision.models.segmentation.DeepLabV3_ResNet101_Weights">[docs]</a><span class="k">class</span> <span class="nc">DeepLabV3_ResNet101_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_WITH_VOC_LABELS_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/deeplabv3_resnet101_coco-586e9e4e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">SemanticSegmentation</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">520</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">60996202</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet101&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017-VOC-labels&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;miou&quot;</span><span class="p">:</span> <span class="mf">67.4</span><span class="p">,</span>
+                    <span class="s2">&quot;pixel_acc&quot;</span><span class="p">:</span> <span class="mf">92.4</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">258.743</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">233.217</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_WITH_VOC_LABELS_V1</span></div>
+
+
+<div class="viewcode-block" id="DeepLabV3_MobileNet_V3_Large_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html#torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights">[docs]</a><span class="k">class</span> <span class="nc">DeepLabV3_MobileNet_V3_Large_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_WITH_VOC_LABELS_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/deeplabv3_mobilenet_v3_large-fc3c493d.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">SemanticSegmentation</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">520</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">11029328</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_mobilenet_v3_large&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017-VOC-labels&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;miou&quot;</span><span class="p">:</span> <span class="mf">60.3</span><span class="p">,</span>
+                    <span class="s2">&quot;pixel_acc&quot;</span><span class="p">:</span> <span class="mf">91.2</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">10.452</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">42.301</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_WITH_VOC_LABELS_V1</span></div>
+
+
+<span class="k">def</span> <span class="nf">_deeplabv3_mobilenetv3</span><span class="p">(</span>
+    <span class="n">backbone</span><span class="p">:</span> <span class="n">MobileNetV3</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">aux</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DeepLabV3</span><span class="p">:</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">features</span>
+    <span class="c1"># Gather the indices of blocks which are strided. These are the locations of C1, ..., Cn-1 blocks.</span>
+    <span class="c1"># The first and last blocks are always included because they are the C0 (conv1) and Cn.</span>
+    <span class="n">stage_indices</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="s2">&quot;_is_cn&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)]</span> <span class="o">+</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">]</span>
+    <span class="n">out_pos</span> <span class="o">=</span> <span class="n">stage_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>  <span class="c1"># use C5 which has output_stride = 16</span>
+    <span class="n">out_inplanes</span> <span class="o">=</span> <span class="n">backbone</span><span class="p">[</span><span class="n">out_pos</span><span class="p">]</span><span class="o">.</span><span class="n">out_channels</span>
+    <span class="n">aux_pos</span> <span class="o">=</span> <span class="n">stage_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">4</span><span class="p">]</span>  <span class="c1"># use C2 here which has output_stride = 8</span>
+    <span class="n">aux_inplanes</span> <span class="o">=</span> <span class="n">backbone</span><span class="p">[</span><span class="n">aux_pos</span><span class="p">]</span><span class="o">.</span><span class="n">out_channels</span>
+    <span class="n">return_layers</span> <span class="o">=</span> <span class="p">{</span><span class="nb">str</span><span class="p">(</span><span class="n">out_pos</span><span class="p">):</span> <span class="s2">&quot;out&quot;</span><span class="p">}</span>
+    <span class="k">if</span> <span class="n">aux</span><span class="p">:</span>
+        <span class="n">return_layers</span><span class="p">[</span><span class="nb">str</span><span class="p">(</span><span class="n">aux_pos</span><span class="p">)]</span> <span class="o">=</span> <span class="s2">&quot;aux&quot;</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">IntermediateLayerGetter</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">return_layers</span><span class="o">=</span><span class="n">return_layers</span><span class="p">)</span>
+
+    <span class="n">aux_classifier</span> <span class="o">=</span> <span class="n">FCNHead</span><span class="p">(</span><span class="n">aux_inplanes</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span> <span class="k">if</span> <span class="n">aux</span> <span class="k">else</span> <span class="kc">None</span>
+    <span class="n">classifier</span> <span class="o">=</span> <span class="n">DeepLabHead</span><span class="p">(</span><span class="n">out_inplanes</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">DeepLabV3</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">classifier</span><span class="p">,</span> <span class="n">aux_classifier</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="deeplabv3_resnet50"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html#torchvision.models.segmentation.deeplabv3_resnet50">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">DeepLabV3_ResNet50_Weights</span><span class="o">.</span><span class="n">COCO_WITH_VOC_LABELS_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">deeplabv3_resnet50</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DeepLabV3_ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">aux_loss</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DeepLabV3</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Constructs a DeepLabV3 model with a ResNet-50 backbone.</span>
+
+<span class="sd">    .. betastatus:: segmentation module</span>
+
+<span class="sd">    Reference: `Rethinking Atrous Convolution for Semantic Image Segmentation &lt;https://arxiv.org/abs/1706.05587&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.segmentation.DeepLabV3_ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.segmentation.DeepLabV3_ResNet50_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        aux_loss (bool, optional): If True, it uses an auxiliary loss</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The pretrained weights for the</span>
+<span class="sd">            backbone</span>
+<span class="sd">        **kwargs: unused</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.segmentation.DeepLabV3_ResNet50_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">DeepLabV3_ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="n">aux_loss</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;aux_loss&quot;</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">21</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">replace_stride_with_dilation</span><span class="o">=</span><span class="p">[</span><span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">True</span><span class="p">])</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">_deeplabv3_resnet</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="deeplabv3_resnet101"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html#torchvision.models.segmentation.deeplabv3_resnet101">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">DeepLabV3_ResNet101_Weights</span><span class="o">.</span><span class="n">COCO_WITH_VOC_LABELS_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">deeplabv3_resnet101</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DeepLabV3_ResNet101_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">aux_loss</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet101_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DeepLabV3</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Constructs a DeepLabV3 model with a ResNet-101 backbone.</span>
+
+<span class="sd">    .. betastatus:: segmentation module</span>
+
+<span class="sd">    Reference: `Rethinking Atrous Convolution for Semantic Image Segmentation &lt;https://arxiv.org/abs/1706.05587&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.segmentation.DeepLabV3_ResNet101_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.segmentation.DeepLabV3_ResNet101_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        aux_loss (bool, optional): If True, it uses an auxiliary loss</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet101_Weights`, optional): The pretrained weights for the</span>
+<span class="sd">            backbone</span>
+<span class="sd">        **kwargs: unused</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.segmentation.DeepLabV3_ResNet101_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">DeepLabV3_ResNet101_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="n">aux_loss</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;aux_loss&quot;</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">21</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet101</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">replace_stride_with_dilation</span><span class="o">=</span><span class="p">[</span><span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">True</span><span class="p">])</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">_deeplabv3_resnet</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="deeplabv3_mobilenet_v3_large"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html#torchvision.models.segmentation.deeplabv3_mobilenet_v3_large">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">DeepLabV3_MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">COCO_WITH_VOC_LABELS_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">deeplabv3_mobilenet_v3_large</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DeepLabV3_MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">aux_loss</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DeepLabV3</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Constructs a DeepLabV3 model with a MobileNetV3-Large backbone.</span>
+
+<span class="sd">    Reference: `Rethinking Atrous Convolution for Semantic Image Segmentation &lt;https://arxiv.org/abs/1706.05587&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background)</span>
+<span class="sd">        aux_loss (bool, optional): If True, it uses an auxiliary loss</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The pretrained weights</span>
+<span class="sd">            for the backbone</span>
+<span class="sd">        **kwargs: unused</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">DeepLabV3_MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="n">aux_loss</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;aux_loss&quot;</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">21</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">mobilenet_v3_large</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">dilated</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">_deeplabv3_mobilenetv3</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/segmentation/fcn.html b/0.16/_modules/torchvision/models/segmentation/fcn.html
new file mode 100644
index 00000000000..e4a133f020b
--- /dev/null
+++ b/0.16/_modules/torchvision/models/segmentation/fcn.html
@@ -0,0 +1,901 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.segmentation.fcn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.segmentation.fcn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.segmentation.fcn</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">SemanticSegmentation</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_VOC_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span><span class="p">,</span> <span class="n">IntermediateLayerGetter</span>
+<span class="kn">from</span> <span class="nn">..resnet</span> <span class="kn">import</span> <span class="n">ResNet</span><span class="p">,</span> <span class="n">resnet101</span><span class="p">,</span> <span class="n">ResNet101_Weights</span><span class="p">,</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_SimpleSegmentationModel</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;FCN&quot;</span><span class="p">,</span> <span class="s2">&quot;FCN_ResNet50_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;FCN_ResNet101_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;fcn_resnet50&quot;</span><span class="p">,</span> <span class="s2">&quot;fcn_resnet101&quot;</span><span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">FCN</span><span class="p">(</span><span class="n">_SimpleSegmentationModel</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements FCN model from</span>
+<span class="sd">    `&quot;Fully Convolutional Networks for Semantic Segmentation&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1411.4038&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            The backbone should return an OrderedDict[Tensor], with the key being</span>
+<span class="sd">            &quot;out&quot; for the last feature map used, and &quot;aux&quot; if an auxiliary classifier</span>
+<span class="sd">            is used.</span>
+<span class="sd">        classifier (nn.Module): module that takes the &quot;out&quot; element returned from</span>
+<span class="sd">            the backbone and returns a dense prediction.</span>
+<span class="sd">        aux_classifier (nn.Module, optional): auxiliary classifier used during training</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">pass</span>
+
+
+<span class="k">class</span> <span class="nc">FCNHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">inter_channels</span> <span class="o">=</span> <span class="n">in_channels</span> <span class="o">//</span> <span class="mi">4</span>
+        <span class="n">layers</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">inter_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">inter_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="mf">0.1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">inter_channels</span><span class="p">,</span> <span class="n">channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+        <span class="p">]</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_VOC_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">        These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC</span>
+<span class="s2">        dataset.</span>
+<span class="s2">    &quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="FCN_ResNet50_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.fcn_resnet50.html#torchvision.models.segmentation.FCN_ResNet50_Weights">[docs]</a><span class="k">class</span> <span class="nc">FCN_ResNet50_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_WITH_VOC_LABELS_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/fcn_resnet50_coco-1167a1af.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">SemanticSegmentation</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">520</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">35322218</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet50&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017-VOC-labels&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;miou&quot;</span><span class="p">:</span> <span class="mf">60.5</span><span class="p">,</span>
+                    <span class="s2">&quot;pixel_acc&quot;</span><span class="p">:</span> <span class="mf">91.4</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">152.717</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">135.009</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_WITH_VOC_LABELS_V1</span></div>
+
+
+<div class="viewcode-block" id="FCN_ResNet101_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.fcn_resnet101.html#torchvision.models.segmentation.FCN_ResNet101_Weights">[docs]</a><span class="k">class</span> <span class="nc">FCN_ResNet101_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_WITH_VOC_LABELS_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/fcn_resnet101_coco-7ecb50ca.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">SemanticSegmentation</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">520</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">54314346</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet101&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017-VOC-labels&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;miou&quot;</span><span class="p">:</span> <span class="mf">63.7</span><span class="p">,</span>
+                    <span class="s2">&quot;pixel_acc&quot;</span><span class="p">:</span> <span class="mf">91.9</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">232.738</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">207.711</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_WITH_VOC_LABELS_V1</span></div>
+
+
+<span class="k">def</span> <span class="nf">_fcn_resnet</span><span class="p">(</span>
+    <span class="n">backbone</span><span class="p">:</span> <span class="n">ResNet</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">aux</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FCN</span><span class="p">:</span>
+    <span class="n">return_layers</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;layer4&quot;</span><span class="p">:</span> <span class="s2">&quot;out&quot;</span><span class="p">}</span>
+    <span class="k">if</span> <span class="n">aux</span><span class="p">:</span>
+        <span class="n">return_layers</span><span class="p">[</span><span class="s2">&quot;layer3&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;aux&quot;</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">IntermediateLayerGetter</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">return_layers</span><span class="o">=</span><span class="n">return_layers</span><span class="p">)</span>
+
+    <span class="n">aux_classifier</span> <span class="o">=</span> <span class="n">FCNHead</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span> <span class="k">if</span> <span class="n">aux</span> <span class="k">else</span> <span class="kc">None</span>
+    <span class="n">classifier</span> <span class="o">=</span> <span class="n">FCNHead</span><span class="p">(</span><span class="mi">2048</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">FCN</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">classifier</span><span class="p">,</span> <span class="n">aux_classifier</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="fcn_resnet50"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.fcn_resnet50.html#torchvision.models.segmentation.fcn_resnet50">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">FCN_ResNet50_Weights</span><span class="o">.</span><span class="n">COCO_WITH_VOC_LABELS_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">fcn_resnet50</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">FCN_ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">aux_loss</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet50_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FCN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Fully-Convolutional Network model with a ResNet-50 backbone from the `Fully Convolutional</span>
+<span class="sd">    Networks for Semantic Segmentation &lt;https://arxiv.org/abs/1411.4038&gt;`_ paper.</span>
+
+<span class="sd">    .. betastatus:: segmentation module</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.segmentation.FCN_ResNet50_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.segmentation.FCN_ResNet50_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background).</span>
+<span class="sd">        aux_loss (bool, optional): If True, it uses an auxiliary loss.</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet50_Weights`, optional): The pretrained</span>
+<span class="sd">            weights for the backbone.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.segmentation.fcn.FCN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.segmentation.FCN_ResNet50_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">FCN_ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="n">aux_loss</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;aux_loss&quot;</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">21</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">replace_stride_with_dilation</span><span class="o">=</span><span class="p">[</span><span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">True</span><span class="p">])</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">_fcn_resnet</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="fcn_resnet101"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.fcn_resnet101.html#torchvision.models.segmentation.fcn_resnet101">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">FCN_ResNet101_Weights</span><span class="o">.</span><span class="n">COCO_WITH_VOC_LABELS_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">fcn_resnet101</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">FCN_ResNet101_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">aux_loss</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ResNet101_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">FCN</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Fully-Convolutional Network model with a ResNet-101 backbone from the `Fully Convolutional</span>
+<span class="sd">    Networks for Semantic Segmentation &lt;https://arxiv.org/abs/1411.4038&gt;`_ paper.</span>
+
+<span class="sd">    .. betastatus:: segmentation module</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.segmentation.FCN_ResNet101_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.segmentation.FCN_ResNet101_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background).</span>
+<span class="sd">        aux_loss (bool, optional): If True, it uses an auxiliary loss.</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.ResNet101_Weights`, optional): The pretrained</span>
+<span class="sd">            weights for the backbone.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.segmentation.fcn.FCN``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.segmentation.FCN_ResNet101_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">FCN_ResNet101_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">ResNet101_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="n">aux_loss</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;aux_loss&quot;</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">21</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">resnet101</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">replace_stride_with_dilation</span><span class="o">=</span><span class="p">[</span><span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">True</span><span class="p">])</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">_fcn_resnet</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">aux_loss</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/segmentation/lraspp.html b/0.16/_modules/torchvision/models/segmentation/lraspp.html
new file mode 100644
index 00000000000..2f269ec0128
--- /dev/null
+++ b/0.16/_modules/torchvision/models/segmentation/lraspp.html
@@ -0,0 +1,847 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.segmentation.lraspp &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.segmentation.lraspp</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.segmentation.lraspp</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.nn</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">SemanticSegmentation</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_VOC_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_value_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span><span class="p">,</span> <span class="n">IntermediateLayerGetter</span>
+<span class="kn">from</span> <span class="nn">..mobilenetv3</span> <span class="kn">import</span> <span class="n">mobilenet_v3_large</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="p">,</span> <span class="n">MobileNetV3</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;LRASPP&quot;</span><span class="p">,</span> <span class="s2">&quot;LRASPP_MobileNet_V3_Large_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;lraspp_mobilenet_v3_large&quot;</span><span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">LRASPP</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements a Lite R-ASPP Network for semantic segmentation from</span>
+<span class="sd">    `&quot;Searching for MobileNetV3&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1905.02244&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        backbone (nn.Module): the network used to compute the features for the model.</span>
+<span class="sd">            The backbone should return an OrderedDict[Tensor], with the key being</span>
+<span class="sd">            &quot;high&quot; for the high level feature map and &quot;low&quot; for the low level feature map.</span>
+<span class="sd">        low_channels (int): the number of channels of the low level features.</span>
+<span class="sd">        high_channels (int): the number of channels of the high level features.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background).</span>
+<span class="sd">        inter_channels (int, optional): the number of channels for intermediate computations.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">backbone</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">low_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">high_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">inter_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">128</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">LRASPPHead</span><span class="p">(</span><span class="n">low_channels</span><span class="p">,</span> <span class="n">high_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">inter_channels</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+        <span class="n">features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backbone</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">features</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">interpolate</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:],</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;bilinear&quot;</span><span class="p">,</span> <span class="n">align_corners</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+        <span class="n">result</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+        <span class="n">result</span><span class="p">[</span><span class="s2">&quot;out&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">out</span>
+
+        <span class="k">return</span> <span class="n">result</span>
+
+
+<span class="k">class</span> <span class="nc">LRASPPHead</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">low_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">high_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">inter_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">cbr</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">high_channels</span><span class="p">,</span> <span class="n">inter_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">inter_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">high_channels</span><span class="p">,</span> <span class="n">inter_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Sigmoid</span><span class="p">(),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">low_classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">low_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">high_classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">inter_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">low</span> <span class="o">=</span> <span class="nb">input</span><span class="p">[</span><span class="s2">&quot;low&quot;</span><span class="p">]</span>
+        <span class="n">high</span> <span class="o">=</span> <span class="nb">input</span><span class="p">[</span><span class="s2">&quot;high&quot;</span><span class="p">]</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">cbr</span><span class="p">(</span><span class="n">high</span><span class="p">)</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">(</span><span class="n">high</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">*</span> <span class="n">s</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">interpolate</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">low</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:],</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;bilinear&quot;</span><span class="p">,</span> <span class="n">align_corners</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">low_classifier</span><span class="p">(</span><span class="n">low</span><span class="p">)</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">high_classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_lraspp_mobilenetv3</span><span class="p">(</span><span class="n">backbone</span><span class="p">:</span> <span class="n">MobileNetV3</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">LRASPP</span><span class="p">:</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">backbone</span><span class="o">.</span><span class="n">features</span>
+    <span class="c1"># Gather the indices of blocks which are strided. These are the locations of C1, ..., Cn-1 blocks.</span>
+    <span class="c1"># The first and last blocks are always included because they are the C0 (conv1) and Cn.</span>
+    <span class="n">stage_indices</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="s2">&quot;_is_cn&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)]</span> <span class="o">+</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">backbone</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">]</span>
+    <span class="n">low_pos</span> <span class="o">=</span> <span class="n">stage_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">4</span><span class="p">]</span>  <span class="c1"># use C2 here which has output_stride = 8</span>
+    <span class="n">high_pos</span> <span class="o">=</span> <span class="n">stage_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>  <span class="c1"># use C5 which has output_stride = 16</span>
+    <span class="n">low_channels</span> <span class="o">=</span> <span class="n">backbone</span><span class="p">[</span><span class="n">low_pos</span><span class="p">]</span><span class="o">.</span><span class="n">out_channels</span>
+    <span class="n">high_channels</span> <span class="o">=</span> <span class="n">backbone</span><span class="p">[</span><span class="n">high_pos</span><span class="p">]</span><span class="o">.</span><span class="n">out_channels</span>
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">IntermediateLayerGetter</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">return_layers</span><span class="o">=</span><span class="p">{</span><span class="nb">str</span><span class="p">(</span><span class="n">low_pos</span><span class="p">):</span> <span class="s2">&quot;low&quot;</span><span class="p">,</span> <span class="nb">str</span><span class="p">(</span><span class="n">high_pos</span><span class="p">):</span> <span class="s2">&quot;high&quot;</span><span class="p">})</span>
+
+    <span class="k">return</span> <span class="n">LRASPP</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">low_channels</span><span class="p">,</span> <span class="n">high_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="LRASPP_MobileNet_V3_Large_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html#torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights">[docs]</a><span class="k">class</span> <span class="nc">LRASPP_MobileNet_V3_Large_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">COCO_WITH_VOC_LABELS_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/lraspp_mobilenet_v3_large-d234d4ea.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">SemanticSegmentation</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">520</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">3221538</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_VOC_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/segmentation#lraspp_mobilenet_v3_large&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;COCO-val2017-VOC-labels&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;miou&quot;</span><span class="p">:</span> <span class="mf">57.9</span><span class="p">,</span>
+                    <span class="s2">&quot;pixel_acc&quot;</span><span class="p">:</span> <span class="mf">91.2</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">2.086</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">12.49</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained on a subset of COCO, using only the 20 categories that are present in the</span>
+<span class="s2">                Pascal VOC dataset.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">COCO_WITH_VOC_LABELS_V1</span></div>
+
+
+<div class="viewcode-block" id="lraspp_mobilenet_v3_large"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html#torchvision.models.segmentation.lraspp_mobilenet_v3_large">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span>
+    <span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">LRASPP_MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">COCO_WITH_VOC_LABELS_V1</span><span class="p">),</span>
+    <span class="n">weights_backbone</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained_backbone&quot;</span><span class="p">,</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">),</span>
+<span class="p">)</span>
+<span class="k">def</span> <span class="nf">lraspp_mobilenet_v3_large</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">LRASPP_MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="n">num_classes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">weights_backbone</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MobileNet_V3_Large_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">LRASPP</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Constructs a Lite R-ASPP Network model with a MobileNetV3-Large backbone from</span>
+<span class="sd">    `Searching for MobileNetV3 &lt;https://arxiv.org/abs/1905.02244&gt;`_ paper.</span>
+
+<span class="sd">    .. betastatus:: segmentation module</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        num_classes (int, optional): number of output classes of the model (including the background).</span>
+<span class="sd">        aux_loss (bool, optional): If True, it uses an auxiliary loss.</span>
+<span class="sd">        weights_backbone (:class:`~torchvision.models.MobileNet_V3_Large_Weights`, optional): The pretrained</span>
+<span class="sd">            weights for the backbone.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.segmentation.LRASPP``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/lraspp.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;aux_loss&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;This model does not use auxiliary loss&quot;</span><span class="p">)</span>
+
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">LRASPP_MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="n">weights_backbone</span> <span class="o">=</span> <span class="n">MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights_backbone</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">weights_backbone</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="n">_ovewrite_value_param</span><span class="p">(</span><span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="k">elif</span> <span class="n">num_classes</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">num_classes</span> <span class="o">=</span> <span class="mi">21</span>
+
+    <span class="n">backbone</span> <span class="o">=</span> <span class="n">mobilenet_v3_large</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights_backbone</span><span class="p">,</span> <span class="n">dilated</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">_lraspp_mobilenetv3</span><span class="p">(</span><span class="n">backbone</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/shufflenetv2.html b/0.16/_modules/torchvision/models/shufflenetv2.html
new file mode 100644
index 00000000000..e58affd8fbd
--- /dev/null
+++ b/0.16/_modules/torchvision/models/shufflenetv2.html
@@ -0,0 +1,1077 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.shufflenetv2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.shufflenetv2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.shufflenetv2</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;ShuffleNetV2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X0_5_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X1_0_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X1_5_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ShuffleNet_V2_X2_0_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x0_5&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x1_0&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x1_5&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;shufflenet_v2_x2_0&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">channel_shuffle</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">batchsize</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+    <span class="n">channels_per_group</span> <span class="o">=</span> <span class="n">num_channels</span> <span class="o">//</span> <span class="n">groups</span>
+
+    <span class="c1"># reshape</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batchsize</span><span class="p">,</span> <span class="n">groups</span><span class="p">,</span> <span class="n">channels_per_group</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+    <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+
+    <span class="c1"># flatten</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batchsize</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">InvertedResidual</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inp</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">oup</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">stride</span> <span class="o">&lt;=</span> <span class="mi">3</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;illegal stride value&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
+
+        <span class="n">branch_features</span> <span class="o">=</span> <span class="n">oup</span> <span class="o">//</span> <span class="mi">2</span>
+        <span class="k">if</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span><span class="p">)</span> <span class="ow">and</span> <span class="p">(</span><span class="n">inp</span> <span class="o">!=</span> <span class="n">branch_features</span> <span class="o">&lt;&lt;</span> <span class="mi">1</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Invalid combination of stride </span><span class="si">{</span><span class="n">stride</span><span class="si">}</span><span class="s2">, inp </span><span class="si">{</span><span class="n">inp</span><span class="si">}</span><span class="s2"> and oup </span><span class="si">{</span><span class="n">oup</span><span class="si">}</span><span class="s2"> values. If stride == 1 then inp should be equal to oup // 2 &lt;&lt; 1.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">depthwise_conv</span><span class="p">(</span><span class="n">inp</span><span class="p">,</span> <span class="n">inp</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">inp</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">inp</span><span class="p">,</span> <span class="n">branch_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span>
+                <span class="n">inp</span> <span class="k">if</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">)</span> <span class="k">else</span> <span class="n">branch_features</span><span class="p">,</span>
+                <span class="n">branch_features</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">depthwise_conv</span><span class="p">(</span><span class="n">branch_features</span><span class="p">,</span> <span class="n">branch_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">,</span> <span class="n">branch_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">depthwise_conv</span><span class="p">(</span>
+        <span class="n">i</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">o</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">i</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">x1</span><span class="p">,</span> <span class="n">x2</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">chunk</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            <span class="n">out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">x1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x2</span><span class="p">)),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">branch1</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x</span><span class="p">)),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="n">channel_shuffle</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">ShuffleNetV2</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">stages_repeats</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">stages_out_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">inverted_residual</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">InvertedResidual</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">stages_repeats</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;expected stages_repeats as list of 3 positive ints&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">stages_out_channels</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">5</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;expected stages_out_channels as list of 5 positive ints&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_stage_out_channels</span> <span class="o">=</span> <span class="n">stages_out_channels</span>
+
+        <span class="n">input_channels</span> <span class="o">=</span> <span class="mi">3</span>
+        <span class="n">output_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_stage_out_channels</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">output_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="n">input_channels</span> <span class="o">=</span> <span class="n">output_channels</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="c1"># Static annotations for mypy</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stage2</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stage3</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stage4</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span>
+        <span class="n">stage_names</span> <span class="o">=</span> <span class="p">[</span><span class="sa">f</span><span class="s2">&quot;stage</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">]]</span>
+        <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">repeats</span><span class="p">,</span> <span class="n">output_channels</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">stage_names</span><span class="p">,</span> <span class="n">stages_repeats</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_stage_out_channels</span><span class="p">[</span><span class="mi">1</span><span class="p">:]):</span>
+            <span class="n">seq</span> <span class="o">=</span> <span class="p">[</span><span class="n">inverted_residual</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">2</span><span class="p">)]</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">repeats</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
+                <span class="n">seq</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">inverted_residual</span><span class="p">(</span><span class="n">output_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+            <span class="nb">setattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">name</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">seq</span><span class="p">))</span>
+            <span class="n">input_channels</span> <span class="o">=</span> <span class="n">output_channels</span>
+
+        <span class="n">output_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_stage_out_channels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv5</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">output_channels</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">output_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_forward_impl</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># See note [TorchScript super()]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stage2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stage3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stage4</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv5</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">mean</span><span class="p">([</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">])</span>  <span class="c1"># globalpool</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_forward_impl</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_shufflenetv2</span><span class="p">(</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ShuffleNetV2</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">ShuffleNetV2</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/ericsun99/Shufflenet-v2-Pytorch&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X0_5_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.ShuffleNet_V2_X0_5_Weights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X0_5_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/ericsun99/Shufflenet-v2-Pytorch</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/shufflenetv2_x0.5-f707e7126e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">1366792</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">60.552</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">81.746</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.04</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">5.282</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were trained from scratch to reproduce closely the results of the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X1_0_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.ShuffleNet_V2_X1_0_Weights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X1_0_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/ericsun99/Shufflenet-v2-Pytorch</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/shufflenetv2_x1-5666bf0f80.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">2278604</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">69.362</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">88.316</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.145</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">8.791</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were trained from scratch to reproduce closely the results of the paper.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X1_5_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.ShuffleNet_V2_X1_5_Weights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X1_5_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/shufflenetv2_x1_5-3c479a10.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5906&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">3503624</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">72.996</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.086</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.296</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">13.557</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ShuffleNet_V2_X2_0_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.ShuffleNet_V2_X2_0_Weights">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNet_V2_X2_0_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/shufflenetv2_x2_0-8be3c8ee.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5906&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">7393996</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">76.230</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.006</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.583</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">28.433</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using TorchVision&#39;s `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x0_5"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.shufflenet_v2_x0_5">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x0_5</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 architecture with 0.5x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ShuffleNet_V2_X0_5_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ShuffleNet_V2_X0_5_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.shufflenetv2.ShuffleNetV2``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X0_5_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ShuffleNet_V2_X0_5_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x1_0"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.shufflenet_v2_x1_0">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x1_0</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 architecture with 1.0x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ShuffleNet_V2_X1_0_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ShuffleNet_V2_X1_0_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.shufflenetv2.ShuffleNetV2``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X1_0_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ShuffleNet_V2_X1_0_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">116</span><span class="p">,</span> <span class="mi">232</span><span class="p">,</span> <span class="mi">464</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x1_5"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.shufflenet_v2_x1_5">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x1_5</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 architecture with 1.5x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ShuffleNet_V2_X1_5_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ShuffleNet_V2_X1_5_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.shufflenetv2.ShuffleNetV2``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X1_5_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ShuffleNet_V2_X1_5_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">176</span><span class="p">,</span> <span class="mi">352</span><span class="p">,</span> <span class="mi">704</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="shufflenet_v2_x2_0"><a class="viewcode-back" href="../../../models/generated/torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.shufflenet_v2_x2_0">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">shufflenet_v2_x2_0</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ShuffleNetV2</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a ShuffleNetV2 architecture with 2.0x output channels, as described in</span>
+<span class="sd">    `ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1807.11164&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ShuffleNet_V2_X2_0_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.ShuffleNet_V2_X2_0_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.shufflenetv2.ShuffleNetV2``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ShuffleNet_V2_X2_0_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ShuffleNet_V2_X2_0_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_shufflenetv2</span><span class="p">(</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">244</span><span class="p">,</span> <span class="mi">488</span><span class="p">,</span> <span class="mi">976</span><span class="p">,</span> <span class="mi">2048</span><span class="p">],</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/squeezenet.html b/0.16/_modules/torchvision/models/squeezenet.html
new file mode 100644
index 00000000000..2fc54675f2e
--- /dev/null
+++ b/0.16/_modules/torchvision/models/squeezenet.html
@@ -0,0 +1,892 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.squeezenet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.squeezenet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.squeezenet</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">import</span> <span class="nn">torch.nn.init</span> <span class="k">as</span> <span class="nn">init</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;SqueezeNet&quot;</span><span class="p">,</span> <span class="s2">&quot;SqueezeNet1_0_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;SqueezeNet1_1_Weights&quot;</span><span class="p">,</span> <span class="s2">&quot;squeezenet1_0&quot;</span><span class="p">,</span> <span class="s2">&quot;squeezenet1_1&quot;</span><span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">Fire</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inplanes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">squeeze_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">expand1x1_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">expand3x3_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span> <span class="o">=</span> <span class="n">inplanes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">squeeze</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">squeeze_planes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">squeeze_activation</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">expand1x1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">squeeze_planes</span><span class="p">,</span> <span class="n">expand1x1_planes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">expand1x1_activation</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">expand3x3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">squeeze_planes</span><span class="p">,</span> <span class="n">expand3x3_planes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">expand3x3_activation</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">squeeze_activation</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">x</span><span class="p">))</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span>
+            <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">expand1x1_activation</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">expand1x1</span><span class="p">(</span><span class="n">x</span><span class="p">)),</span> <span class="bp">self</span><span class="o">.</span><span class="n">expand3x3_activation</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">expand3x3</span><span class="p">(</span><span class="n">x</span><span class="p">))],</span> <span class="mi">1</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SqueezeNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">version</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;1_0&quot;</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+        <span class="k">if</span> <span class="n">version</span> <span class="o">==</span> <span class="s2">&quot;1_0&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">7</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">96</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="k">elif</span> <span class="n">version</span> <span class="o">==</span> <span class="s2">&quot;1_1&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">ceil_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+                <span class="n">Fire</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># FIXME: Is this needed? SqueezeNet should only be called from the</span>
+            <span class="c1"># FIXME: squeezenet1_x() functions</span>
+            <span class="c1"># FIXME: This checking is not done for the other models</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unsupported SqueezeNet version </span><span class="si">{</span><span class="n">version</span><span class="si">}</span><span class="s2">: 1_0 or 1_1 expected&quot;</span><span class="p">)</span>
+
+        <span class="c1"># Final convolution is initialized differently from the rest</span>
+        <span class="n">final_conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">),</span> <span class="n">final_conv</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">m</span> <span class="ow">is</span> <span class="n">final_conv</span><span class="p">:</span>
+                    <span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.01</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">init</span><span class="o">.</span><span class="n">kaiming_uniform_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_squeezenet</span><span class="p">(</span>
+    <span class="n">version</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SqueezeNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">SqueezeNet</span><span class="p">(</span><span class="n">version</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/49#issuecomment-277560717&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a simple training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="SqueezeNet1_0_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.squeezenet1_0.html#torchvision.models.SqueezeNet1_0_Weights">[docs]</a><span class="k">class</span> <span class="nc">SqueezeNet1_0_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/squeezenet1_0-b66bff10.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">21</span><span class="p">,</span> <span class="mi">21</span><span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">1248424</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">58.092</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">80.420</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.819</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">4.778</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="SqueezeNet1_1_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.squeezenet1_1.html#torchvision.models.SqueezeNet1_1_Weights">[docs]</a><span class="k">class</span> <span class="nc">SqueezeNet1_1_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/squeezenet1_1-b8a52dc0.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">17</span><span class="p">,</span> <span class="mi">17</span><span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">1235496</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">58.178</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">80.624</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">0.349</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">4.729</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="squeezenet1_0"><a class="viewcode-back" href="../../../models/generated/torchvision.models.squeezenet1_0.html#torchvision.models.squeezenet1_0">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">SqueezeNet1_0_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">squeezenet1_0</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">SqueezeNet1_0_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SqueezeNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;SqueezeNet model architecture from the `SqueezeNet: AlexNet-level</span>
+<span class="sd">    accuracy with 50x fewer parameters and &lt;0.5MB model size</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1602.07360&gt;`_ paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.SqueezeNet1_0_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.SqueezeNet1_0_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.squeezenet.SqueezeNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/squeezenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.SqueezeNet1_0_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">SqueezeNet1_0_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_squeezenet</span><span class="p">(</span><span class="s2">&quot;1_0&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="squeezenet1_1"><a class="viewcode-back" href="../../../models/generated/torchvision.models.squeezenet1_1.html#torchvision.models.squeezenet1_1">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">SqueezeNet1_1_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">squeezenet1_1</span><span class="p">(</span>
+    <span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">SqueezeNet1_1_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SqueezeNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;SqueezeNet 1.1 model from the `official SqueezeNet repo</span>
+<span class="sd">    &lt;https://github.com/DeepScale/SqueezeNet/tree/master/SqueezeNet_v1.1&gt;`_.</span>
+
+<span class="sd">    SqueezeNet 1.1 has 2.4x less computation and slightly fewer parameters</span>
+<span class="sd">    than SqueezeNet 1.0, without sacrificing accuracy.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.SqueezeNet1_1_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.SqueezeNet1_1_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.squeezenet.SqueezeNet``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/squeezenet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.SqueezeNet1_1_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">SqueezeNet1_1_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_squeezenet</span><span class="p">(</span><span class="s2">&quot;1_1&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/swin_transformer.html b/0.16/_modules/torchvision/models/swin_transformer.html
new file mode 100644
index 00000000000..57f2f66d25f
--- /dev/null
+++ b/0.16/_modules/torchvision/models/swin_transformer.html
@@ -0,0 +1,1702 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.swin_transformer &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.swin_transformer</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.swin_transformer</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">MLP</span><span class="p">,</span> <span class="n">Permute</span>
+<span class="kn">from</span> <span class="nn">..ops.stochastic_depth</span> <span class="kn">import</span> <span class="n">StochasticDepth</span>
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">InterpolationMode</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;SwinTransformer&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin_T_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin_S_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin_B_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin_V2_T_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin_V2_S_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin_V2_B_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin_t&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin_s&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin_b&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin_v2_t&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin_v2_s&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin_v2_b&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_patch_merging_pad</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:]</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">W</span> <span class="o">%</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">H</span> <span class="o">%</span> <span class="mi">2</span><span class="p">))</span>
+    <span class="n">x0</span> <span class="o">=</span> <span class="n">x</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="p">:]</span>  <span class="c1"># ... H/2 W/2 C</span>
+    <span class="n">x1</span> <span class="o">=</span> <span class="n">x</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="p">:]</span>  <span class="c1"># ... H/2 W/2 C</span>
+    <span class="n">x2</span> <span class="o">=</span> <span class="n">x</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="p">:]</span>  <span class="c1"># ... H/2 W/2 C</span>
+    <span class="n">x3</span> <span class="o">=</span> <span class="n">x</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">::</span><span class="mi">2</span><span class="p">,</span> <span class="p">:]</span>  <span class="c1"># ... H/2 W/2 C</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">x0</span><span class="p">,</span> <span class="n">x1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">x3</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># ... H/2 W/2 4*C</span>
+    <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_patch_merging_pad&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_get_relative_position_bias</span><span class="p">(</span>
+    <span class="n">relative_position_bias_table</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">relative_position_index</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">N</span> <span class="o">=</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+    <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="n">relative_position_bias_table</span><span class="p">[</span><span class="n">relative_position_index</span><span class="p">]</span>  <span class="c1"># type: ignore[index]</span>
+    <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="n">relative_position_bias</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="n">N</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="n">relative_position_bias</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">relative_position_bias</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_get_relative_position_bias&quot;</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">PatchMerging</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Patch Merging Layer.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        dim (int): Number of input channels.</span>
+<span class="sd">        norm_layer (nn.Module): Normalization layer. Default: nn.LayerNorm.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dim</span> <span class="o">=</span> <span class="n">dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">reduction</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">4</span> <span class="o">*</span> <span class="n">dim</span><span class="p">,</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="mi">4</span> <span class="o">*</span> <span class="n">dim</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): input tensor with expected layout of [..., H, W, C]</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor with layout of [..., H/2, W/2, 2*C]</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">_patch_merging_pad</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">reduction</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># ... H/2 W/2 2*C</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">PatchMergingV2</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Patch Merging Layer for Swin Transformer V2.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        dim (int): Number of input channels.</span>
+<span class="sd">        norm_layer (nn.Module): Normalization layer. Default: nn.LayerNorm.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dim</span> <span class="o">=</span> <span class="n">dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">reduction</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">4</span> <span class="o">*</span> <span class="n">dim</span><span class="p">,</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="n">dim</span><span class="p">)</span>  <span class="c1"># difference</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): input tensor with expected layout of [..., H, W, C]</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor with layout of [..., H/2, W/2, 2*C]</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">_patch_merging_pad</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">reduction</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># ... H/2 W/2 2*C</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">def</span> <span class="nf">shifted_window_attention</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">qkv_weight</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">proj_weight</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">relative_position_bias</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="n">qkv_bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">proj_bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">logit_scale</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">training</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Window based multi-head self attention (W-MSA) module with relative position bias.</span>
+<span class="sd">    It supports both of shifted and non-shifted window.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, H, W, C]): The input tensor or 4-dimensions.</span>
+<span class="sd">        qkv_weight (Tensor[in_dim, out_dim]): The weight tensor of query, key, value.</span>
+<span class="sd">        proj_weight (Tensor[out_dim, out_dim]): The weight tensor of projection.</span>
+<span class="sd">        relative_position_bias (Tensor): The learned relative position bias added to attention.</span>
+<span class="sd">        window_size (List[int]): Window size.</span>
+<span class="sd">        num_heads (int): Number of attention heads.</span>
+<span class="sd">        shift_size (List[int]): Shift size for shifted window attention.</span>
+<span class="sd">        attention_dropout (float): Dropout ratio of attention weight. Default: 0.0.</span>
+<span class="sd">        dropout (float): Dropout ratio of output. Default: 0.0.</span>
+<span class="sd">        qkv_bias (Tensor[out_dim], optional): The bias tensor of query, key, value. Default: None.</span>
+<span class="sd">        proj_bias (Tensor[out_dim], optional): The bias tensor of projection. Default: None.</span>
+<span class="sd">        logit_scale (Tensor[out_dim], optional): Logit scale of cosine attention for Swin Transformer V2. Default: None.</span>
+<span class="sd">        training (bool, optional): Training flag used by the dropout parameters. Default: True.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, H, W, C]: The output tensor after shifted window attention.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">B</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">,</span> <span class="n">C</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">shape</span>
+    <span class="c1"># pad feature maps to multiples of window size</span>
+    <span class="n">pad_r</span> <span class="o">=</span> <span class="p">(</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">W</span> <span class="o">%</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="o">%</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+    <span class="n">pad_b</span> <span class="o">=</span> <span class="p">(</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="n">H</span> <span class="o">%</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">%</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">pad_r</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">pad_b</span><span class="p">))</span>
+    <span class="n">_</span><span class="p">,</span> <span class="n">pad_H</span><span class="p">,</span> <span class="n">pad_W</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+
+    <span class="n">shift_size</span> <span class="o">=</span> <span class="n">shift_size</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+    <span class="c1"># If window size is larger than feature size, there is no need to shift window</span>
+    <span class="k">if</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;=</span> <span class="n">pad_H</span><span class="p">:</span>
+        <span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="k">if</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;=</span> <span class="n">pad_W</span><span class="p">:</span>
+        <span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+
+    <span class="c1"># cyclic shift</span>
+    <span class="k">if</span> <span class="nb">sum</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">roll</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">shifts</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="n">dims</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span>
+
+    <span class="c1"># partition windows</span>
+    <span class="n">num_windows</span> <span class="o">=</span> <span class="p">(</span><span class="n">pad_H</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">*</span> <span class="p">(</span><span class="n">pad_W</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">pad_H</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pad_W</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">C</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span> <span class="o">*</span> <span class="n">num_windows</span><span class="p">,</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">C</span><span class="p">)</span>  <span class="c1"># B*nW, Ws*Ws, C</span>
+
+    <span class="c1"># multi-head attention</span>
+    <span class="k">if</span> <span class="n">logit_scale</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">qkv_bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">qkv_bias</span> <span class="o">=</span> <span class="n">qkv_bias</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+        <span class="n">length</span> <span class="o">=</span> <span class="n">qkv_bias</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">//</span> <span class="mi">3</span>
+        <span class="n">qkv_bias</span><span class="p">[</span><span class="n">length</span> <span class="p">:</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">length</span><span class="p">]</span><span class="o">.</span><span class="n">zero_</span><span class="p">()</span>
+    <span class="n">qkv</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">linear</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">qkv_weight</span><span class="p">,</span> <span class="n">qkv_bias</span><span class="p">)</span>
+    <span class="n">qkv</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="mi">3</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">C</span> <span class="o">//</span> <span class="n">num_heads</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+    <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="n">qkv</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">logit_scale</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># cosine attention</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">normalize</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span> <span class="o">@</span> <span class="n">F</span><span class="o">.</span><span class="n">normalize</span><span class="p">(</span><span class="n">k</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">logit_scale</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">logit_scale</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="mf">100.0</span><span class="p">))</span><span class="o">.</span><span class="n">exp</span><span class="p">()</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span> <span class="o">*</span> <span class="n">logit_scale</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">q</span> <span class="o">=</span> <span class="n">q</span> <span class="o">*</span> <span class="p">(</span><span class="n">C</span> <span class="o">//</span> <span class="n">num_heads</span><span class="p">)</span> <span class="o">**</span> <span class="o">-</span><span class="mf">0.5</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">q</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">k</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+    <span class="c1"># add relative position bias</span>
+    <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span> <span class="o">+</span> <span class="n">relative_position_bias</span>
+
+    <span class="k">if</span> <span class="nb">sum</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="c1"># generate attention mask</span>
+        <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">((</span><span class="n">pad_H</span><span class="p">,</span> <span class="n">pad_W</span><span class="p">))</span>
+        <span class="n">h_slices</span> <span class="o">=</span> <span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="o">-</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="p">(</span><span class="o">-</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="p">(</span><span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="kc">None</span><span class="p">))</span>
+        <span class="n">w_slices</span> <span class="o">=</span> <span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="o">-</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="p">(</span><span class="o">-</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="p">(</span><span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="kc">None</span><span class="p">))</span>
+        <span class="n">count</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">h</span> <span class="ow">in</span> <span class="n">h_slices</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">w</span> <span class="ow">in</span> <span class="n">w_slices</span><span class="p">:</span>
+                <span class="n">attn_mask</span><span class="p">[</span><span class="n">h</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="p">:</span> <span class="n">h</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">w</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="p">:</span> <span class="n">w</span><span class="p">[</span><span class="mi">1</span><span class="p">]]</span> <span class="o">=</span> <span class="n">count</span>
+                <span class="n">count</span> <span class="o">+=</span> <span class="mi">1</span>
+        <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">pad_H</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pad_W</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">num_windows</span><span class="p">,</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">-</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+        <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">masked_fill</span><span class="p">(</span><span class="n">attn_mask</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="o">-</span><span class="mf">100.0</span><span class="p">))</span><span class="o">.</span><span class="n">masked_fill</span><span class="p">(</span><span class="n">attn_mask</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="mf">0.0</span><span class="p">))</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">//</span> <span class="n">num_windows</span><span class="p">,</span> <span class="n">num_windows</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">))</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span> <span class="o">+</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">))</span>
+
+    <span class="n">attn</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">attn</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">attn</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">attn</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="n">training</span><span class="p">)</span>
+
+    <span class="n">x</span> <span class="o">=</span> <span class="n">attn</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">v</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">C</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">linear</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">proj_weight</span><span class="p">,</span> <span class="n">proj_bias</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="n">training</span><span class="p">)</span>
+
+    <span class="c1"># reverse windows</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">pad_H</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pad_W</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">C</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">pad_H</span><span class="p">,</span> <span class="n">pad_W</span><span class="p">,</span> <span class="n">C</span><span class="p">)</span>
+
+    <span class="c1"># reverse cyclic shift</span>
+    <span class="k">if</span> <span class="nb">sum</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">roll</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">shifts</span><span class="o">=</span><span class="p">(</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="n">dims</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span>
+
+    <span class="c1"># unpad features</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="p">[:,</span> <span class="p">:</span><span class="n">H</span><span class="p">,</span> <span class="p">:</span><span class="n">W</span><span class="p">,</span> <span class="p">:]</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+    <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;shifted_window_attention&quot;</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ShiftedWindowAttention</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`shifted_window_attention`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">qkv_bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">proj_bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">window_size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span> <span class="ow">or</span> <span class="nb">len</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;window_size and shift_size must be of length 2&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span> <span class="o">=</span> <span class="n">window_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">shift_size</span> <span class="o">=</span> <span class="n">shift_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span> <span class="o">=</span> <span class="n">num_heads</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">attention_dropout</span> <span class="o">=</span> <span class="n">attention_dropout</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">dropout</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="n">dim</span> <span class="o">*</span> <span class="mi">3</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="n">qkv_bias</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="n">dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="n">proj_bias</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">define_relative_position_bias_table</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">define_relative_position_index</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">define_relative_position_bias_table</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="c1"># define a parameter table of relative position bias</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">)</span>
+        <span class="p">)</span>  <span class="c1"># 2*Wh-1 * 2*Ww-1, nH</span>
+        <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">define_relative_position_index</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="c1"># get pair-wise relative position index for each token inside the window</span>
+        <span class="n">coords_h</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">coords_w</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="n">coords</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">meshgrid</span><span class="p">(</span><span class="n">coords_h</span><span class="p">,</span> <span class="n">coords_w</span><span class="p">,</span> <span class="n">indexing</span><span class="o">=</span><span class="s2">&quot;ij&quot;</span><span class="p">))</span>  <span class="c1"># 2, Wh, Ww</span>
+        <span class="n">coords_flatten</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">coords</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>  <span class="c1"># 2, Wh*Ww</span>
+        <span class="n">relative_coords</span> <span class="o">=</span> <span class="n">coords_flatten</span><span class="p">[:,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">-</span> <span class="n">coords_flatten</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span>  <span class="c1"># 2, Wh*Ww, Wh*Ww</span>
+        <span class="n">relative_coords</span> <span class="o">=</span> <span class="n">relative_coords</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>  <span class="c1"># Wh*Ww, Wh*Ww, 2</span>
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>  <span class="c1"># shift to start from 0</span>
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">*=</span> <span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="n">relative_position_index</span> <span class="o">=</span> <span class="n">relative_coords</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">flatten</span><span class="p">()</span>  <span class="c1"># Wh*Ww*Wh*Ww</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;relative_position_index&quot;</span><span class="p">,</span> <span class="n">relative_position_index</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">get_relative_position_bias</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_get_relative_position_bias</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_index</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span>  <span class="c1"># type: ignore[arg-type]</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Tensor with layout of [B, H, W, C]</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor with same layout as input, i.e. [B, H, W, C]</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_relative_position_bias</span><span class="p">()</span>
+        <span class="k">return</span> <span class="n">shifted_window_attention</span><span class="p">(</span>
+            <span class="n">x</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span>
+            <span class="n">relative_position_bias</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">shift_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">shift_size</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">,</span>
+            <span class="n">qkv_bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
+            <span class="n">proj_bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
+            <span class="n">training</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">,</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ShiftedWindowAttentionV2</span><span class="p">(</span><span class="n">ShiftedWindowAttention</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`shifted_window_attention_v2`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">qkv_bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">proj_bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">dim</span><span class="p">,</span>
+            <span class="n">window_size</span><span class="p">,</span>
+            <span class="n">shift_size</span><span class="p">,</span>
+            <span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">qkv_bias</span><span class="o">=</span><span class="n">qkv_bias</span><span class="p">,</span>
+            <span class="n">proj_bias</span><span class="o">=</span><span class="n">proj_bias</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">logit_scale</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="mi">10</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="n">num_heads</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))))</span>
+        <span class="c1"># mlp to generate continuous relative position bias</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">cpb_mlp</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">qkv_bias</span><span class="p">:</span>
+            <span class="n">length</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">bias</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">//</span> <span class="mi">3</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">bias</span><span class="p">[</span><span class="n">length</span> <span class="p">:</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">length</span><span class="p">]</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">zero_</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">define_relative_position_bias_table</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="c1"># get relative_coords_table</span>
+        <span class="n">relative_coords_h</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="o">-</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="n">relative_coords_w</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="o">-</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="n">relative_coords_table</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">meshgrid</span><span class="p">([</span><span class="n">relative_coords_h</span><span class="p">,</span> <span class="n">relative_coords_w</span><span class="p">],</span> <span class="n">indexing</span><span class="o">=</span><span class="s2">&quot;ij&quot;</span><span class="p">))</span>
+        <span class="n">relative_coords_table</span> <span class="o">=</span> <span class="n">relative_coords_table</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># 1, 2*Wh-1, 2*Ww-1, 2</span>
+
+        <span class="n">relative_coords_table</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">/=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="n">relative_coords_table</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">/=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+
+        <span class="n">relative_coords_table</span> <span class="o">*=</span> <span class="mi">8</span>  <span class="c1"># normalize to -8, 8</span>
+        <span class="n">relative_coords_table</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">sign</span><span class="p">(</span><span class="n">relative_coords_table</span><span class="p">)</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">log2</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">abs</span><span class="p">(</span><span class="n">relative_coords_table</span><span class="p">)</span> <span class="o">+</span> <span class="mf">1.0</span><span class="p">)</span> <span class="o">/</span> <span class="mf">3.0</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;relative_coords_table&quot;</span><span class="p">,</span> <span class="n">relative_coords_table</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">get_relative_position_bias</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="n">_get_relative_position_bias</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">cpb_mlp</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relative_coords_table</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">),</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_index</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="mi">16</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="n">relative_position_bias</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">relative_position_bias</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (Tensor): Tensor with layout of [B, H, W, C]</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor with same layout as input, i.e. [B, H, W, C]</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_relative_position_bias</span><span class="p">()</span>
+        <span class="k">return</span> <span class="n">shifted_window_attention</span><span class="p">(</span>
+            <span class="n">x</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span>
+            <span class="n">relative_position_bias</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">shift_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">shift_size</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">,</span>
+            <span class="n">qkv_bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
+            <span class="n">proj_bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
+            <span class="n">logit_scale</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">logit_scale</span><span class="p">,</span>
+            <span class="n">training</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">,</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SwinTransformerBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Swin Transformer Block.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        dim (int): Number of input channels.</span>
+<span class="sd">        num_heads (int): Number of attention heads.</span>
+<span class="sd">        window_size (List[int]): Window size.</span>
+<span class="sd">        shift_size (List[int]): Shift size for shifted window attention.</span>
+<span class="sd">        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4.0.</span>
+<span class="sd">        dropout (float): Dropout rate. Default: 0.0.</span>
+<span class="sd">        attention_dropout (float): Attention dropout rate. Default: 0.0.</span>
+<span class="sd">        stochastic_depth_prob: (float): Stochastic depth rate. Default: 0.0.</span>
+<span class="sd">        norm_layer (nn.Module): Normalization layer.  Default: nn.LayerNorm.</span>
+<span class="sd">        attn_layer (nn.Module): Attention layer. Default: ShiftedWindowAttention</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">4.0</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span>
+        <span class="n">attn_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">ShiftedWindowAttention</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm1</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">dim</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">attn</span> <span class="o">=</span> <span class="n">attn_layer</span><span class="p">(</span>
+            <span class="n">dim</span><span class="p">,</span>
+            <span class="n">window_size</span><span class="p">,</span>
+            <span class="n">shift_size</span><span class="p">,</span>
+            <span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span> <span class="o">=</span> <span class="n">StochasticDepth</span><span class="p">(</span><span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="s2">&quot;row&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm2</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">dim</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mlp</span> <span class="o">=</span> <span class="n">MLP</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">dim</span> <span class="o">*</span> <span class="n">mlp_ratio</span><span class="p">),</span> <span class="n">dim</span><span class="p">],</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">GELU</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">xavier_uniform_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">attn</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">norm1</span><span class="p">(</span><span class="n">x</span><span class="p">)))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mlp</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">norm2</span><span class="p">(</span><span class="n">x</span><span class="p">)))</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">SwinTransformerBlockV2</span><span class="p">(</span><span class="n">SwinTransformerBlock</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Swin Transformer V2 Block.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        dim (int): Number of input channels.</span>
+<span class="sd">        num_heads (int): Number of attention heads.</span>
+<span class="sd">        window_size (List[int]): Window size.</span>
+<span class="sd">        shift_size (List[int]): Shift size for shifted window attention.</span>
+<span class="sd">        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4.0.</span>
+<span class="sd">        dropout (float): Dropout rate. Default: 0.0.</span>
+<span class="sd">        attention_dropout (float): Attention dropout rate. Default: 0.0.</span>
+<span class="sd">        stochastic_depth_prob: (float): Stochastic depth rate. Default: 0.0.</span>
+<span class="sd">        norm_layer (nn.Module): Normalization layer.  Default: nn.LayerNorm.</span>
+<span class="sd">        attn_layer (nn.Module): Attention layer. Default: ShiftedWindowAttentionV2.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">4.0</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span>
+        <span class="n">attn_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">ShiftedWindowAttentionV2</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">dim</span><span class="p">,</span>
+            <span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">window_size</span><span class="p">,</span>
+            <span class="n">shift_size</span><span class="p">,</span>
+            <span class="n">mlp_ratio</span><span class="o">=</span><span class="n">mlp_ratio</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">stochastic_depth_prob</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">attn_layer</span><span class="o">=</span><span class="n">attn_layer</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">):</span>
+        <span class="c1"># Here is the difference, we apply norm after the attention in V2.</span>
+        <span class="c1"># In V1 we applied norm before the attention.</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">norm1</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">attn</span><span class="p">(</span><span class="n">x</span><span class="p">)))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">norm2</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mlp</span><span class="p">(</span><span class="n">x</span><span class="p">)))</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">SwinTransformer</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements Swin Transformer from the `&quot;Swin Transformer: Hierarchical Vision Transformer using</span>
+<span class="sd">    Shifted Windows&quot; &lt;https://arxiv.org/abs/2103.14030&gt;`_ paper.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        patch_size (List[int]): Patch size.</span>
+<span class="sd">        embed_dim (int): Patch embedding dimension.</span>
+<span class="sd">        depths (List(int)): Depth of each Swin Transformer layer.</span>
+<span class="sd">        num_heads (List(int)): Number of attention heads in different layers.</span>
+<span class="sd">        window_size (List[int]): Window size.</span>
+<span class="sd">        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4.0.</span>
+<span class="sd">        dropout (float): Dropout rate. Default: 0.0.</span>
+<span class="sd">        attention_dropout (float): Attention dropout rate. Default: 0.0.</span>
+<span class="sd">        stochastic_depth_prob (float): Stochastic depth rate. Default: 0.1.</span>
+<span class="sd">        num_classes (int): Number of classes for classification head. Default: 1000.</span>
+<span class="sd">        block (nn.Module, optional): SwinTransformer Block. Default: None.</span>
+<span class="sd">        norm_layer (nn.Module, optional): Normalization layer. Default: None.</span>
+<span class="sd">        downsample_layer (nn.Module): Downsample layer (patch merging). Default: PatchMerging.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">patch_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">depths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">4.0</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">downsample_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">PatchMerging</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">SwinTransformerBlock</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-5</span><span class="p">)</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="c1"># split image into non-overlapping patches</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span>
+                    <span class="mi">3</span><span class="p">,</span> <span class="n">embed_dim</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="n">patch_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">patch_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="n">patch_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">patch_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+                <span class="p">),</span>
+                <span class="n">Permute</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">]),</span>
+                <span class="n">norm_layer</span><span class="p">(</span><span class="n">embed_dim</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="n">total_stage_blocks</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="n">depths</span><span class="p">)</span>
+        <span class="n">stage_block_id</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="c1"># build SwinTransformer blocks</span>
+        <span class="k">for</span> <span class="n">i_stage</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">depths</span><span class="p">)):</span>
+            <span class="n">stage</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">dim</span> <span class="o">=</span> <span class="n">embed_dim</span> <span class="o">*</span> <span class="mi">2</span><span class="o">**</span><span class="n">i_stage</span>
+            <span class="k">for</span> <span class="n">i_layer</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="n">i_stage</span><span class="p">]):</span>
+                <span class="c1"># adjust stochastic depth probability based on the depth of the stage block</span>
+                <span class="n">sd_prob</span> <span class="o">=</span> <span class="n">stochastic_depth_prob</span> <span class="o">*</span> <span class="nb">float</span><span class="p">(</span><span class="n">stage_block_id</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="n">total_stage_blocks</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="n">stage</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                    <span class="n">block</span><span class="p">(</span>
+                        <span class="n">dim</span><span class="p">,</span>
+                        <span class="n">num_heads</span><span class="p">[</span><span class="n">i_stage</span><span class="p">],</span>
+                        <span class="n">window_size</span><span class="o">=</span><span class="n">window_size</span><span class="p">,</span>
+                        <span class="n">shift_size</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span> <span class="k">if</span> <span class="n">i_layer</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">w</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">for</span> <span class="n">w</span> <span class="ow">in</span> <span class="n">window_size</span><span class="p">],</span>
+                        <span class="n">mlp_ratio</span><span class="o">=</span><span class="n">mlp_ratio</span><span class="p">,</span>
+                        <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span>
+                        <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+                        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">sd_prob</span><span class="p">,</span>
+                        <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="p">)</span>
+                <span class="n">stage_block_id</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">stage</span><span class="p">))</span>
+            <span class="c1"># add patch merging layer</span>
+            <span class="k">if</span> <span class="n">i_stage</span> <span class="o">&lt;</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">depths</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
+                <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">downsample_layer</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+
+        <span class="n">num_features</span> <span class="o">=</span> <span class="n">embed_dim</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">**</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">depths</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">num_features</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">permute</span> <span class="o">=</span> <span class="n">Permute</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">])</span>  <span class="c1"># B H W C -&gt; B C H W</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">flatten</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Flatten</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">num_features</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">def</span> <span class="nf">_swin_transformer</span><span class="p">(</span>
+    <span class="n">patch_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">embed_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">depths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">num_heads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">SwinTransformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="n">patch_size</span><span class="p">,</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="n">embed_dim</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="n">depths</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="n">num_heads</span><span class="p">,</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="n">window_size</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">stochastic_depth_prob</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="Swin_T_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_t.html#torchvision.models.Swin_T_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin_T_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin_t-704ceda3.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">232</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">28288354</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#swintransformer&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.474</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.776</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.491</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">108.19</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a similar training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="Swin_S_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_s.html#torchvision.models.Swin_S_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin_S_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin_s-5e29d889.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">246</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">49606258</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#swintransformer&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.196</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.360</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">8.741</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">189.786</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a similar training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="Swin_B_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_b.html#torchvision.models.Swin_B_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin_B_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin_b-68c6b09e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">238</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">87768224</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#swintransformer&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.582</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.640</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.431</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">335.364</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a similar training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="Swin_V2_T_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_v2_t.html#torchvision.models.Swin_V2_T_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin_V2_T_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin_v2_t-b137f0e2.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">260</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">28351570</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">82.072</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.132</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">5.94</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">108.626</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a similar training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="Swin_V2_S_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_v2_s.html#torchvision.models.Swin_V2_S_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin_V2_S_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin_v2_s-637d8ceb.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">260</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">49737442</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">83.712</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.816</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">11.546</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">190.675</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a similar training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="Swin_V2_B_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_v2_b.html#torchvision.models.Swin_V2_B_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin_V2_B_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin_v2_b-781e5279.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">272</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">87930848</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">84.112</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.864</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">20.325</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">336.372</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights reproduce closely the results of the paper using a similar training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="swin_t"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_t.html#torchvision.models.swin_t">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin_T_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin_t</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin_T_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_tiny architecture from</span>
+<span class="sd">    `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows &lt;https://arxiv.org/abs/2103.14030&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Swin_T_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Swin_T_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Swin_T_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin_T_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">96</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="swin_s"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_s.html#torchvision.models.swin_s">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin_S_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin_s</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin_S_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_small architecture from</span>
+<span class="sd">    `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows &lt;https://arxiv.org/abs/2103.14030&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Swin_S_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Swin_S_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Swin_S_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin_S_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">96</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.3</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="swin_b"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_b.html#torchvision.models.swin_b">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin_B_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin_b</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin_B_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_base architecture from</span>
+<span class="sd">    `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows &lt;https://arxiv.org/abs/2103.14030&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Swin_B_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Swin_B_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Swin_B_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin_B_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="swin_v2_t"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_v2_t.html#torchvision.models.swin_v2_t">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin_V2_T_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin_v2_t</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin_V2_T_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_v2_tiny architecture from</span>
+<span class="sd">    `Swin Transformer V2: Scaling Up Capacity and Resolution &lt;https://arxiv.org/abs/2111.09883&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Swin_V2_T_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Swin_V2_T_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Swin_V2_T_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin_V2_T_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">96</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="n">block</span><span class="o">=</span><span class="n">SwinTransformerBlockV2</span><span class="p">,</span>
+        <span class="n">downsample_layer</span><span class="o">=</span><span class="n">PatchMergingV2</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="swin_v2_s"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_v2_s.html#torchvision.models.swin_v2_s">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin_V2_S_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin_v2_s</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin_V2_S_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_v2_small architecture from</span>
+<span class="sd">    `Swin Transformer V2: Scaling Up Capacity and Resolution &lt;https://arxiv.org/abs/2111.09883&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Swin_V2_S_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Swin_V2_S_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Swin_V2_S_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin_V2_S_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">96</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.3</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="n">block</span><span class="o">=</span><span class="n">SwinTransformerBlockV2</span><span class="p">,</span>
+        <span class="n">downsample_layer</span><span class="o">=</span><span class="n">PatchMergingV2</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="swin_v2_b"><a class="viewcode-back" href="../../../models/generated/torchvision.models.swin_v2_b.html#torchvision.models.swin_v2_b">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin_V2_B_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin_v2_b</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin_V2_B_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_v2_base architecture from</span>
+<span class="sd">    `Swin Transformer V2: Scaling Up Capacity and Resolution &lt;https://arxiv.org/abs/2111.09883&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.Swin_V2_B_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.Swin_V2_B_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.Swin_V2_B_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin_V2_B_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="n">block</span><span class="o">=</span><span class="n">SwinTransformerBlockV2</span><span class="p">,</span>
+        <span class="n">downsample_layer</span><span class="o">=</span><span class="n">PatchMergingV2</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/vgg.html b/0.16/_modules/torchvision/models/vgg.html
new file mode 100644
index 00000000000..1351eac9c43
--- /dev/null
+++ b/0.16/_modules/torchvision/models/vgg.html
@@ -0,0 +1,1180 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.vgg &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.vgg</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.vgg</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;VGG&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG11_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG11_BN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG13_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG13_BN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG16_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG16_BN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG19_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;VGG19_BN_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg11&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg11_bn&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg13&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg13_bn&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg16&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg16_bn&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg19&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vgg19_bn&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">VGG</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">features</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">init_weights</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">features</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">((</span><span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">512</span> <span class="o">*</span> <span class="mi">7</span> <span class="o">*</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">4096</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">4096</span><span class="p">,</span> <span class="mi">4096</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">4096</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">init_weights</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">):</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mf">0.01</span><span class="p">)</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">def</span> <span class="nf">make_layers</span><span class="p">(</span><span class="n">cfg</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]],</span> <span class="n">batch_norm</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">:</span>
+    <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="n">in_channels</span> <span class="o">=</span> <span class="mi">3</span>
+    <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">cfg</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">v</span> <span class="o">==</span> <span class="s2">&quot;M&quot;</span><span class="p">:</span>
+            <span class="n">layers</span> <span class="o">+=</span> <span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">v</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="n">v</span><span class="p">)</span>
+            <span class="n">conv2d</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">batch_norm</span><span class="p">:</span>
+                <span class="n">layers</span> <span class="o">+=</span> <span class="p">[</span><span class="n">conv2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">v</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">layers</span> <span class="o">+=</span> <span class="p">[</span><span class="n">conv2d</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)]</span>
+            <span class="n">in_channels</span> <span class="o">=</span> <span class="n">v</span>
+    <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+
+
+<span class="n">cfgs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;A&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">64</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;B&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;D&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;E&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">,</span> <span class="s2">&quot;M&quot;</span><span class="p">],</span>
+<span class="p">}</span>
+
+
+<span class="k">def</span> <span class="nf">_vgg</span><span class="p">(</span><span class="n">cfg</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">batch_norm</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;init_weights&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span>
+        <span class="k">if</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">VGG</span><span class="p">(</span><span class="n">make_layers</span><span class="p">(</span><span class="n">cfgs</span><span class="p">[</span><span class="n">cfg</span><span class="p">],</span> <span class="n">batch_norm</span><span class="o">=</span><span class="n">batch_norm</span><span class="p">),</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;These weights were trained from scratch by using a simplified training recipe.&quot;&quot;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="VGG11_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg11.html#torchvision.models.VGG11_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG11_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg11-8a719046.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">132863336</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">69.020</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">88.628</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">7.609</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">506.84</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="VGG11_BN_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg11_bn.html#torchvision.models.VGG11_BN_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG11_BN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg11_bn-6002323d.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">132868840</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">70.370</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">89.810</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">7.609</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">506.881</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="VGG13_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg13.html#torchvision.models.VGG13_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG13_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg13-19584684.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">133047848</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">69.928</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">89.246</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">11.308</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">507.545</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="VGG13_BN_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg13_bn.html#torchvision.models.VGG13_BN_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG13_BN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg13_bn-abd245e5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">133053736</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">71.586</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.374</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">11.308</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">507.59</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="VGG16_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg16.html#torchvision.models.VGG16_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG16_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg16-397923af.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">138357544</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">71.592</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.382</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.47</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">527.796</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_FEATURES</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="c1"># Weights ported from https://github.com/amdegroot/ssd.pytorch/</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg16_features-amdegroot-88682ab5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.48235</span><span class="p">,</span> <span class="mf">0.45882</span><span class="p">,</span> <span class="mf">0.40784</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">/</span> <span class="mf">255.0</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="mf">255.0</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="mf">255.0</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">138357544</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/amdegroot/ssd.pytorch#training-ssd&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;nan&quot;</span><span class="p">),</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;nan&quot;</span><span class="p">),</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.47</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">527.802</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights can&#39;t be used for classification because they are missing values in the `classifier`</span>
+<span class="s2">                module. Only the `features` module has valid values and can be used for feature extraction. The weights</span>
+<span class="s2">                were trained using the original input standardization method as described in the paper.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="VGG16_BN_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg16_bn.html#torchvision.models.VGG16_BN_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG16_BN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg16_bn-6c64b313.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">138365992</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">73.360</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.516</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.47</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">527.866</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="VGG19_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg19.html#torchvision.models.VGG19_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG19_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg19-dcbb9e9d.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">143667240</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">72.376</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">90.876</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">19.632</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">548.051</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="VGG19_BN_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg19_bn.html#torchvision.models.VGG19_BN_Weights">[docs]</a><span class="k">class</span> <span class="nc">VGG19_BN_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vgg19_bn-c79401a0.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">143678248</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">74.218</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">91.842</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">19.632</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">548.143</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="vgg11"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg11.html#torchvision.models.vgg11">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG11_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg11</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG11_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-11 from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG11_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG11_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG11_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG11_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;A&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vgg11_bn"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg11_bn.html#torchvision.models.vgg11_bn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG11_BN_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg11_bn</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG11_BN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-11-BN from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG11_BN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG11_BN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG11_BN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG11_BN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;A&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vgg13"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg13.html#torchvision.models.vgg13">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG13_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg13</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG13_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-13 from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG13_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG13_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG13_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG13_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;B&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vgg13_bn"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg13_bn.html#torchvision.models.vgg13_bn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG13_BN_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg13_bn</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG13_BN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-13-BN from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG13_BN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG13_BN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG13_BN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG13_BN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;B&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vgg16"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg16.html#torchvision.models.vgg16">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG16_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg16</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG16_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-16 from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG16_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG16_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG16_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG16_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;D&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vgg16_bn"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg16_bn.html#torchvision.models.vgg16_bn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG16_BN_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg16_bn</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG16_BN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-16-BN from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG16_BN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG16_BN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG16_BN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG16_BN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;D&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vgg19"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg19.html#torchvision.models.vgg19">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG19_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg19</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG19_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-19 from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG19_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG19_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG19_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG19_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;E&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vgg19_bn"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vgg19_bn.html#torchvision.models.vgg19_bn">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">VGG19_BN_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vgg19_bn</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">VGG19_BN_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VGG</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;VGG-19_BN from `Very Deep Convolutional Networks for Large-Scale Image Recognition &lt;https://arxiv.org/abs/1409.1556&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.VGG19_BN_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.VGG19_BN_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vgg.VGG``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.VGG19_BN_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">VGG19_BN_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vgg</span><span class="p">(</span><span class="s2">&quot;E&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/video/mvit.html b/0.16/_modules/torchvision/models/video/mvit.html
new file mode 100644
index 00000000000..d7fbf90b374
--- /dev/null
+++ b/0.16/_modules/torchvision/models/video/mvit.html
@@ -0,0 +1,1566 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.video.mvit &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.video.mvit</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.video.mvit</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">dataclasses</span> <span class="kn">import</span> <span class="n">dataclass</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+
+<span class="kn">from</span> <span class="nn">...ops</span> <span class="kn">import</span> <span class="n">MLP</span><span class="p">,</span> <span class="n">StochasticDepth</span>
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">VideoClassification</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_KINETICS400_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;MViT&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MViT_V1_B_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mvit_v1_b&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MViT_V2_S_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mvit_v2_s&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="nd">@dataclass</span>
+<span class="k">class</span> <span class="nc">MSBlockConfig</span><span class="p">:</span>
+    <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">input_channels</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">output_channels</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">kernel_q</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+    <span class="n">kernel_kv</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+    <span class="n">stride_q</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+    <span class="n">stride_kv</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_prod</span><span class="p">(</span><span class="n">s</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="n">product</span> <span class="o">=</span> <span class="mi">1</span>
+    <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">s</span><span class="p">:</span>
+        <span class="n">product</span> <span class="o">*=</span> <span class="n">v</span>
+    <span class="k">return</span> <span class="n">product</span>
+
+
+<span class="k">def</span> <span class="nf">_unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">target_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">expand_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+    <span class="n">tensor_dim</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span>
+    <span class="k">if</span> <span class="n">tensor_dim</span> <span class="o">==</span> <span class="n">target_dim</span> <span class="o">-</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">expand_dim</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">tensor_dim</span> <span class="o">!=</span> <span class="n">target_dim</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unsupported input dimension </span><span class="si">{</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">x</span><span class="p">,</span> <span class="n">tensor_dim</span>
+
+
+<span class="k">def</span> <span class="nf">_squeeze</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">target_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">expand_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">tensor_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">tensor_dim</span> <span class="o">==</span> <span class="n">target_dim</span> <span class="o">-</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">expand_dim</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_unsqueeze&quot;</span><span class="p">)</span>
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_squeeze&quot;</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">Pool</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">pool</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span>
+        <span class="n">norm</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">activation</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_before_pool</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pool</span> <span class="o">=</span> <span class="n">pool</span>
+        <span class="n">layers</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="n">norm</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">norm</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">activation</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">activation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm_act</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span> <span class="k">if</span> <span class="n">layers</span> <span class="k">else</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm_before_pool</span> <span class="o">=</span> <span class="n">norm_before_pool</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">thw</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+        <span class="n">x</span><span class="p">,</span> <span class="n">tensor_dim</span> <span class="o">=</span> <span class="n">_unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="c1"># Separate the class token and reshape the input</span>
+        <span class="n">class_token</span><span class="p">,</span> <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor_split</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">indices</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+        <span class="n">B</span><span class="p">,</span> <span class="n">N</span><span class="p">,</span> <span class="n">C</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[:</span><span class="mi">3</span><span class="p">]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="n">B</span> <span class="o">*</span> <span class="n">N</span><span class="p">,</span> <span class="n">C</span><span class="p">)</span> <span class="o">+</span> <span class="n">thw</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+
+        <span class="c1"># normalizing prior pooling is useful when we use BN which can be absorbed to speed up inference</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_before_pool</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_act</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_act</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="c1"># apply the pool on the input and add back the token</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">T</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">N</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">class_token</span><span class="p">,</span> <span class="n">x</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_before_pool</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_act</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_act</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="n">_squeeze</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">tensor_dim</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="n">T</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_interpolate</span><span class="p">(</span><span class="n">embedding</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">d</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">embedding</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">d</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">embedding</span>
+
+    <span class="k">return</span> <span class="p">(</span>
+        <span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">interpolate</span><span class="p">(</span>
+            <span class="n">embedding</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span>
+            <span class="n">size</span><span class="o">=</span><span class="n">d</span><span class="p">,</span>
+            <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;linear&quot;</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_add_rel_pos</span><span class="p">(</span>
+    <span class="n">attn</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">q</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">q_thw</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">k_thw</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">rel_pos_h</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">rel_pos_w</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">rel_pos_t</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># Modified code from: https://github.com/facebookresearch/SlowFast/commit/1aebd71a2efad823d52b827a3deaf15a56cf4932</span>
+    <span class="n">q_t</span><span class="p">,</span> <span class="n">q_h</span><span class="p">,</span> <span class="n">q_w</span> <span class="o">=</span> <span class="n">q_thw</span>
+    <span class="n">k_t</span><span class="p">,</span> <span class="n">k_h</span><span class="p">,</span> <span class="n">k_w</span> <span class="o">=</span> <span class="n">k_thw</span>
+    <span class="n">dh</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="n">q_h</span><span class="p">,</span> <span class="n">k_h</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="n">dw</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="n">q_w</span><span class="p">,</span> <span class="n">k_w</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="n">dt</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="n">q_t</span><span class="p">,</span> <span class="n">k_t</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+
+    <span class="c1"># Scale up rel pos if shapes for q and k are different.</span>
+    <span class="n">q_h_ratio</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">k_h</span> <span class="o">/</span> <span class="n">q_h</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">k_h_ratio</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">q_h</span> <span class="o">/</span> <span class="n">k_h</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">dist_h</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">q_h</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">q_h_ratio</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">k_h</span><span class="p">)[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">+</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">k_h</span><span class="p">))</span> <span class="o">*</span> <span class="n">k_h_ratio</span>
+    <span class="n">q_w_ratio</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">k_w</span> <span class="o">/</span> <span class="n">q_w</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">k_w_ratio</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">q_w</span> <span class="o">/</span> <span class="n">k_w</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">dist_w</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">q_w</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">q_w_ratio</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">k_w</span><span class="p">)[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">+</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">k_w</span><span class="p">))</span> <span class="o">*</span> <span class="n">k_w_ratio</span>
+    <span class="n">q_t_ratio</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">k_t</span> <span class="o">/</span> <span class="n">q_t</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">k_t_ratio</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">q_t</span> <span class="o">/</span> <span class="n">k_t</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">dist_t</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">q_t</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">q_t_ratio</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">k_t</span><span class="p">)[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">+</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">k_t</span><span class="p">))</span> <span class="o">*</span> <span class="n">k_t_ratio</span>
+
+    <span class="c1"># Interpolate rel pos if needed.</span>
+    <span class="n">rel_pos_h</span> <span class="o">=</span> <span class="n">_interpolate</span><span class="p">(</span><span class="n">rel_pos_h</span><span class="p">,</span> <span class="n">dh</span><span class="p">)</span>
+    <span class="n">rel_pos_w</span> <span class="o">=</span> <span class="n">_interpolate</span><span class="p">(</span><span class="n">rel_pos_w</span><span class="p">,</span> <span class="n">dw</span><span class="p">)</span>
+    <span class="n">rel_pos_t</span> <span class="o">=</span> <span class="n">_interpolate</span><span class="p">(</span><span class="n">rel_pos_t</span><span class="p">,</span> <span class="n">dt</span><span class="p">)</span>
+    <span class="n">Rh</span> <span class="o">=</span> <span class="n">rel_pos_h</span><span class="p">[</span><span class="n">dist_h</span><span class="o">.</span><span class="n">long</span><span class="p">()]</span>
+    <span class="n">Rw</span> <span class="o">=</span> <span class="n">rel_pos_w</span><span class="p">[</span><span class="n">dist_w</span><span class="o">.</span><span class="n">long</span><span class="p">()]</span>
+    <span class="n">Rt</span> <span class="o">=</span> <span class="n">rel_pos_t</span><span class="p">[</span><span class="n">dist_t</span><span class="o">.</span><span class="n">long</span><span class="p">()]</span>
+
+    <span class="n">B</span><span class="p">,</span> <span class="n">n_head</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">dim</span> <span class="o">=</span> <span class="n">q</span><span class="o">.</span><span class="n">shape</span>
+
+    <span class="n">r_q</span> <span class="o">=</span> <span class="n">q</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">:]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">n_head</span><span class="p">,</span> <span class="n">q_t</span><span class="p">,</span> <span class="n">q_h</span><span class="p">,</span> <span class="n">q_w</span><span class="p">,</span> <span class="n">dim</span><span class="p">)</span>
+    <span class="n">rel_h_q</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s2">&quot;bythwc,hkc-&gt;bythwk&quot;</span><span class="p">,</span> <span class="n">r_q</span><span class="p">,</span> <span class="n">Rh</span><span class="p">)</span>  <span class="c1"># [B, H, q_t, qh, qw, k_h]</span>
+    <span class="n">rel_w_q</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s2">&quot;bythwc,wkc-&gt;bythwk&quot;</span><span class="p">,</span> <span class="n">r_q</span><span class="p">,</span> <span class="n">Rw</span><span class="p">)</span>  <span class="c1"># [B, H, q_t, qh, qw, k_w]</span>
+    <span class="c1"># [B, H, q_t, q_h, q_w, dim] -&gt; [q_t, B, H, q_h, q_w, dim] -&gt; [q_t, B*H*q_h*q_w, dim]</span>
+    <span class="n">r_q</span> <span class="o">=</span> <span class="n">r_q</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">q_t</span><span class="p">,</span> <span class="n">B</span> <span class="o">*</span> <span class="n">n_head</span> <span class="o">*</span> <span class="n">q_h</span> <span class="o">*</span> <span class="n">q_w</span><span class="p">,</span> <span class="n">dim</span><span class="p">)</span>
+    <span class="c1"># [q_t, B*H*q_h*q_w, dim] * [q_t, dim, k_t] = [q_t, B*H*q_h*q_w, k_t] -&gt; [B*H*q_h*q_w, q_t, k_t]</span>
+    <span class="n">rel_q_t</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">r_q</span><span class="p">,</span> <span class="n">Rt</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="c1"># [B*H*q_h*q_w, q_t, k_t] -&gt; [B, H, q_t, q_h, q_w, k_t]</span>
+    <span class="n">rel_q_t</span> <span class="o">=</span> <span class="n">rel_q_t</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">n_head</span><span class="p">,</span> <span class="n">q_h</span><span class="p">,</span> <span class="n">q_w</span><span class="p">,</span> <span class="n">q_t</span><span class="p">,</span> <span class="n">k_t</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
+
+    <span class="c1"># Combine rel pos.</span>
+    <span class="n">rel_pos</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">rel_h_q</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span>
+        <span class="o">+</span> <span class="n">rel_w_q</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span>
+        <span class="o">+</span> <span class="n">rel_q_t</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+    <span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">n_head</span><span class="p">,</span> <span class="n">q_t</span> <span class="o">*</span> <span class="n">q_h</span> <span class="o">*</span> <span class="n">q_w</span><span class="p">,</span> <span class="n">k_t</span> <span class="o">*</span> <span class="n">k_h</span> <span class="o">*</span> <span class="n">k_w</span><span class="p">)</span>
+
+    <span class="c1"># Add it to attention</span>
+    <span class="n">attn</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">:,</span> <span class="mi">1</span><span class="p">:]</span> <span class="o">+=</span> <span class="n">rel_pos</span>
+
+    <span class="k">return</span> <span class="n">attn</span>
+
+
+<span class="k">def</span> <span class="nf">_add_shortcut</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">shortcut</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">residual_with_cls_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">):</span>
+    <span class="k">if</span> <span class="n">residual_with_cls_embed</span><span class="p">:</span>
+        <span class="n">x</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">shortcut</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">x</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">:,</span> <span class="p">:]</span> <span class="o">+=</span> <span class="n">shortcut</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">:,</span> <span class="p">:]</span>
+    <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_add_rel_pos&quot;</span><span class="p">)</span>
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_add_shortcut&quot;</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">MultiscaleAttention</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">input_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">output_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">kernel_q</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">kernel_kv</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">stride_q</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">stride_kv</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">residual_pool</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">residual_with_cls_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">rel_pos_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">embed_dim</span> <span class="o">=</span> <span class="n">embed_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">output_dim</span> <span class="o">=</span> <span class="n">output_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span> <span class="o">=</span> <span class="n">num_heads</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span> <span class="o">=</span> <span class="n">output_dim</span> <span class="o">//</span> <span class="n">num_heads</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">residual_pool</span> <span class="o">=</span> <span class="n">residual_pool</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">residual_with_cls_embed</span> <span class="o">=</span> <span class="n">residual_with_cls_embed</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">embed_dim</span><span class="p">,</span> <span class="mi">3</span> <span class="o">*</span> <span class="n">output_dim</span><span class="p">)</span>
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">output_dim</span><span class="p">,</span> <span class="n">output_dim</span><span class="p">)]</span>
+        <span class="k">if</span> <span class="n">dropout</span> <span class="o">&gt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">dropout</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">project</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">pool_q</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">_prod</span><span class="p">(</span><span class="n">kernel_q</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="ow">or</span> <span class="n">_prod</span><span class="p">(</span><span class="n">stride_q</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">padding_q</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">q</span> <span class="o">//</span> <span class="mi">2</span><span class="p">)</span> <span class="k">for</span> <span class="n">q</span> <span class="ow">in</span> <span class="n">kernel_q</span><span class="p">]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">pool_q</span> <span class="o">=</span> <span class="n">Pool</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">kernel_q</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="n">stride_q</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">padding</span><span class="o">=</span><span class="n">padding_q</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">groups</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="p">),</span>
+                <span class="n">norm_layer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">),</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">pool_k</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pool_v</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">_prod</span><span class="p">(</span><span class="n">kernel_kv</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="ow">or</span> <span class="n">_prod</span><span class="p">(</span><span class="n">stride_kv</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">padding_kv</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">kv</span> <span class="o">//</span> <span class="mi">2</span><span class="p">)</span> <span class="k">for</span> <span class="n">kv</span> <span class="ow">in</span> <span class="n">kernel_kv</span><span class="p">]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">pool_k</span> <span class="o">=</span> <span class="n">Pool</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">kernel_kv</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="n">stride_kv</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">padding</span><span class="o">=</span><span class="n">padding_kv</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">groups</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="p">),</span>
+                <span class="n">norm_layer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">),</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">pool_v</span> <span class="o">=</span> <span class="n">Pool</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">kernel_kv</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">stride</span><span class="o">=</span><span class="n">stride_kv</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">padding</span><span class="o">=</span><span class="n">padding_kv</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">groups</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,</span>
+                    <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="p">),</span>
+                <span class="n">norm_layer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">),</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_h</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_w</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_t</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">rel_pos_embed</span><span class="p">:</span>
+            <span class="n">size</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">input_size</span><span class="p">[</span><span class="mi">1</span><span class="p">:])</span>
+            <span class="n">q_size</span> <span class="o">=</span> <span class="n">size</span> <span class="o">//</span> <span class="n">stride_q</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">stride_q</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">size</span>
+            <span class="n">kv_size</span> <span class="o">=</span> <span class="n">size</span> <span class="o">//</span> <span class="n">stride_kv</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">stride_kv</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">size</span>
+            <span class="n">spatial_dim</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="n">q_size</span><span class="p">,</span> <span class="n">kv_size</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span>
+            <span class="n">temporal_dim</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">input_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_h</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">spatial_dim</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_w</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">spatial_dim</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_t</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">temporal_dim</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">))</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_h</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_w</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_t</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">thw</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+        <span class="n">B</span><span class="p">,</span> <span class="n">N</span><span class="p">,</span> <span class="n">C</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="n">N</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_k</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">k</span><span class="p">,</span> <span class="n">k_thw</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_k</span><span class="p">(</span><span class="n">k</span><span class="p">,</span> <span class="n">thw</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">k_thw</span> <span class="o">=</span> <span class="n">thw</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_v</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">v</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_v</span><span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="n">thw</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_q</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">q</span><span class="p">,</span> <span class="n">thw</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_q</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">thw</span><span class="p">)</span>
+
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">scaler</span> <span class="o">*</span> <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_h</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_w</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_t</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">attn</span> <span class="o">=</span> <span class="n">_add_rel_pos</span><span class="p">(</span>
+                <span class="n">attn</span><span class="p">,</span>
+                <span class="n">q</span><span class="p">,</span>
+                <span class="n">thw</span><span class="p">,</span>
+                <span class="n">k_thw</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_h</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_w</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">rel_pos_t</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">attn</span><span class="p">,</span> <span class="n">v</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">residual_pool</span><span class="p">:</span>
+            <span class="n">_add_shortcut</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">residual_with_cls_embed</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">B</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">output_dim</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">project</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span><span class="p">,</span> <span class="n">thw</span>
+
+
+<span class="k">class</span> <span class="nc">MultiscaleBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">input_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">cnf</span><span class="p">:</span> <span class="n">MSBlockConfig</span><span class="p">,</span>
+        <span class="n">residual_pool</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">residual_with_cls_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">rel_pos_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">proj_after_attn</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">proj_after_attn</span> <span class="o">=</span> <span class="n">proj_after_attn</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">pool_skip</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">_prod</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride_q</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">kernel_skip</span> <span class="o">=</span> <span class="p">[</span><span class="n">s</span> <span class="o">+</span> <span class="mi">1</span> <span class="k">if</span> <span class="n">s</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">s</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride_q</span><span class="p">]</span>
+            <span class="n">padding_skip</span> <span class="o">=</span> <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">k</span> <span class="o">//</span> <span class="mi">2</span><span class="p">)</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">kernel_skip</span><span class="p">]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">pool_skip</span> <span class="o">=</span> <span class="n">Pool</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool3d</span><span class="p">(</span><span class="n">kernel_skip</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride_q</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding_skip</span><span class="p">),</span> <span class="kc">None</span>  <span class="c1"># type: ignore[arg-type]</span>
+            <span class="p">)</span>
+
+        <span class="n">attn_dim</span> <span class="o">=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">output_channels</span> <span class="k">if</span> <span class="n">proj_after_attn</span> <span class="k">else</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm1</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm2</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">attn_dim</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">needs_transposal</span> <span class="o">=</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">norm1</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm1d</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">attn</span> <span class="o">=</span> <span class="n">MultiscaleAttention</span><span class="p">(</span>
+            <span class="n">input_size</span><span class="p">,</span>
+            <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span>
+            <span class="n">attn_dim</span><span class="p">,</span>
+            <span class="n">cnf</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">kernel_q</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">kernel_q</span><span class="p">,</span>
+            <span class="n">kernel_kv</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">kernel_kv</span><span class="p">,</span>
+            <span class="n">stride_q</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride_q</span><span class="p">,</span>
+            <span class="n">stride_kv</span><span class="o">=</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride_kv</span><span class="p">,</span>
+            <span class="n">rel_pos_embed</span><span class="o">=</span><span class="n">rel_pos_embed</span><span class="p">,</span>
+            <span class="n">residual_pool</span><span class="o">=</span><span class="n">residual_pool</span><span class="p">,</span>
+            <span class="n">residual_with_cls_embed</span><span class="o">=</span><span class="n">residual_with_cls_embed</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mlp</span> <span class="o">=</span> <span class="n">MLP</span><span class="p">(</span>
+            <span class="n">attn_dim</span><span class="p">,</span>
+            <span class="p">[</span><span class="mi">4</span> <span class="o">*</span> <span class="n">attn_dim</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">output_channels</span><span class="p">],</span>
+            <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">GELU</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span>
+            <span class="n">inplace</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span> <span class="o">=</span> <span class="n">StochasticDepth</span><span class="p">(</span><span class="n">stochastic_depth_prob</span><span class="p">,</span> <span class="s2">&quot;row&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">project</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span> <span class="o">!=</span> <span class="n">cnf</span><span class="o">.</span><span class="n">output_channels</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">project</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">output_channels</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">thw</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+        <span class="n">x_norm1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm1</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">needs_transposal</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x_attn</span><span class="p">,</span> <span class="n">thw_new</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">attn</span><span class="p">(</span><span class="n">x_norm1</span><span class="p">,</span> <span class="n">thw</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">project</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">proj_after_attn</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">project</span><span class="p">(</span><span class="n">x_norm1</span><span class="p">)</span>
+        <span class="n">x_skip</span> <span class="o">=</span> <span class="n">x</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_skip</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">pool_skip</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">thw</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x_skip</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="n">x_attn</span><span class="p">)</span>
+
+        <span class="n">x_norm2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm2</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">needs_transposal</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x_proj</span> <span class="o">=</span> <span class="n">x</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">project</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">proj_after_attn</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">project</span><span class="p">(</span><span class="n">x_norm2</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x_proj</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">stochastic_depth</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mlp</span><span class="p">(</span><span class="n">x_norm2</span><span class="p">)),</span> <span class="n">thw_new</span>
+
+
+<span class="k">class</span> <span class="nc">PositionalEncoding</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">embed_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">spatial_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">temporal_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">rel_pos_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">spatial_size</span> <span class="o">=</span> <span class="n">spatial_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">temporal_size</span> <span class="o">=</span> <span class="n">temporal_size</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_token</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">embed_size</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">spatial_pos</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">temporal_pos</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_pos</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">rel_pos_embed</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">spatial_pos</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">spatial_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">spatial_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">embed_size</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">temporal_pos</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">temporal_size</span><span class="p">,</span> <span class="n">embed_size</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">class_pos</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">embed_size</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">class_token</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">class_token</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">class_token</span><span class="p">,</span> <span class="n">x</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">spatial_pos</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">temporal_pos</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">class_pos</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">hw_size</span><span class="p">,</span> <span class="n">embed_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">spatial_pos</span><span class="o">.</span><span class="n">shape</span>
+            <span class="n">pos_embedding</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">repeat_interleave</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">temporal_pos</span><span class="p">,</span> <span class="n">hw_size</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">pos_embedding</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">spatial_pos</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">temporal_size</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">embed_size</span><span class="p">))</span>
+            <span class="n">pos_embedding</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">class_pos</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">pos_embedding</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">x</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">pos_embedding</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">MViT</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">spatial_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">temporal_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">block_setting</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">MSBlockConfig</span><span class="p">],</span>
+        <span class="n">residual_pool</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">residual_with_cls_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">rel_pos_embed</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">proj_after_attn</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">400</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">patch_embed_kernel</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span>
+        <span class="n">patch_embed_stride</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span>
+        <span class="n">patch_embed_padding</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        MViT main class.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            spatial_size (tuple of ints): The spacial size of the input as ``(H, W)``.</span>
+<span class="sd">            temporal_size (int): The temporal size ``T`` of the input.</span>
+<span class="sd">            block_setting (sequence of MSBlockConfig): The Network structure.</span>
+<span class="sd">            residual_pool (bool): If True, use MViTv2 pooling residual connection.</span>
+<span class="sd">            residual_with_cls_embed (bool): If True, the addition on the residual connection will include</span>
+<span class="sd">                the class embedding.</span>
+<span class="sd">            rel_pos_embed (bool): If True, use MViTv2&#39;s relative positional embeddings.</span>
+<span class="sd">            proj_after_attn (bool): If True, apply the projection after the attention.</span>
+<span class="sd">            dropout (float): Dropout rate. Default: 0.0.</span>
+<span class="sd">            attention_dropout (float): Attention dropout rate. Default: 0.0.</span>
+<span class="sd">            stochastic_depth_prob: (float): Stochastic depth rate. Default: 0.0.</span>
+<span class="sd">            num_classes (int): The number of classes.</span>
+<span class="sd">            block (callable, optional): Module specifying the layer which consists of the attention and mlp.</span>
+<span class="sd">            norm_layer (callable, optional): Module specifying the normalization layer to use.</span>
+<span class="sd">            patch_embed_kernel (tuple of ints): The kernel of the convolution that patchifies the input.</span>
+<span class="sd">            patch_embed_stride (tuple of ints): The stride of the convolution that patchifies the input.</span>
+<span class="sd">            patch_embed_padding (tuple of ints): The padding of the convolution that patchifies the input.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="c1"># This implementation employs a different parameterization scheme than the one used at PyTorch Video:</span>
+        <span class="c1"># https://github.com/facebookresearch/pytorchvideo/blob/718d0a4/pytorchvideo/models/vision_transformers.py</span>
+        <span class="c1"># We remove any experimental configuration that didn&#39;t make it to the final variants of the models. To represent</span>
+        <span class="c1"># the configuration of the architecture we use the simplified form suggested at Table 1 of the paper.</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="n">total_stage_blocks</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">block_setting</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">total_stage_blocks</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The configuration parameter can&#39;t be empty.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">MultiscaleBlock</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">)</span>
+
+        <span class="c1"># Patch Embedding module</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
+            <span class="n">out_channels</span><span class="o">=</span><span class="n">block_setting</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="n">patch_embed_kernel</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="n">patch_embed_stride</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="n">patch_embed_padding</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="n">input_size</span> <span class="o">=</span> <span class="p">[</span><span class="n">size</span> <span class="o">//</span> <span class="n">stride</span> <span class="k">for</span> <span class="n">size</span><span class="p">,</span> <span class="n">stride</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">((</span><span class="n">temporal_size</span><span class="p">,)</span> <span class="o">+</span> <span class="n">spatial_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">stride</span><span class="p">)]</span>
+
+        <span class="c1"># Spatio-Temporal Class Positional Encoding</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pos_encoding</span> <span class="o">=</span> <span class="n">PositionalEncoding</span><span class="p">(</span>
+            <span class="n">embed_size</span><span class="o">=</span><span class="n">block_setting</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">input_channels</span><span class="p">,</span>
+            <span class="n">spatial_size</span><span class="o">=</span><span class="p">(</span><span class="n">input_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">input_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span>
+            <span class="n">temporal_size</span><span class="o">=</span><span class="n">input_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+            <span class="n">rel_pos_embed</span><span class="o">=</span><span class="n">rel_pos_embed</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="c1"># Encoder module</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">blocks</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">stage_block_id</span><span class="p">,</span> <span class="n">cnf</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">block_setting</span><span class="p">):</span>
+            <span class="c1"># adjust stochastic depth probability based on the depth of the stage block</span>
+            <span class="n">sd_prob</span> <span class="o">=</span> <span class="n">stochastic_depth_prob</span> <span class="o">*</span> <span class="n">stage_block_id</span> <span class="o">/</span> <span class="p">(</span><span class="n">total_stage_blocks</span> <span class="o">-</span> <span class="mf">1.0</span><span class="p">)</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">block</span><span class="p">(</span>
+                    <span class="n">input_size</span><span class="o">=</span><span class="n">input_size</span><span class="p">,</span>
+                    <span class="n">cnf</span><span class="o">=</span><span class="n">cnf</span><span class="p">,</span>
+                    <span class="n">residual_pool</span><span class="o">=</span><span class="n">residual_pool</span><span class="p">,</span>
+                    <span class="n">residual_with_cls_embed</span><span class="o">=</span><span class="n">residual_with_cls_embed</span><span class="p">,</span>
+                    <span class="n">rel_pos_embed</span><span class="o">=</span><span class="n">rel_pos_embed</span><span class="p">,</span>
+                    <span class="n">proj_after_attn</span><span class="o">=</span><span class="n">proj_after_attn</span><span class="p">,</span>
+                    <span class="n">dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+                    <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">sd_prob</span><span class="p">,</span>
+                    <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">cnf</span><span class="o">.</span><span class="n">stride_q</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">input_size</span> <span class="o">=</span> <span class="p">[</span><span class="n">size</span> <span class="o">//</span> <span class="n">stride</span> <span class="k">for</span> <span class="n">size</span><span class="p">,</span> <span class="n">stride</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">input_size</span><span class="p">,</span> <span class="n">cnf</span><span class="o">.</span><span class="n">stride_q</span><span class="p">)]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">block_setting</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">output_channels</span><span class="p">)</span>
+
+        <span class="c1"># Classifier module</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">dropout</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">block_setting</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">output_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">)</span> <span class="ow">and</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">weight</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">PositionalEncoding</span><span class="p">):</span>
+                <span class="k">for</span> <span class="n">weights</span> <span class="ow">in</span> <span class="n">m</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="n">weights</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># Convert if necessary (B, C, H, W) -&gt; (B, C, 1, H, W)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">_unsqueeze</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="c1"># patchify and reshape: (B, C, T, H, W) -&gt; (B, embed_channels[0], T&#39;, H&#39;, W&#39;) -&gt; (B, THW&#39;, embed_channels[0])</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="c1"># add positional encoding</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pos_encoding</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="c1"># pass patches through the encoder</span>
+        <span class="n">thw</span> <span class="o">=</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">pos_encoding</span><span class="o">.</span><span class="n">temporal_size</span><span class="p">,)</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">pos_encoding</span><span class="o">.</span><span class="n">spatial_size</span>
+        <span class="k">for</span> <span class="n">block</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">blocks</span><span class="p">:</span>
+            <span class="n">x</span><span class="p">,</span> <span class="n">thw</span> <span class="o">=</span> <span class="n">block</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">thw</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="c1"># classifier &quot;token&quot; as used by standard language architectures</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">def</span> <span class="nf">_mvit</span><span class="p">(</span>
+    <span class="n">block_setting</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">MSBlockConfig</span><span class="p">],</span>
+    <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MViT</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">assert</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;spatial_size&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">])</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;temporal_size&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_temporal_size&quot;</span><span class="p">])</span>
+    <span class="n">spatial_size</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;spatial_size&quot;</span><span class="p">,</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">))</span>
+    <span class="n">temporal_size</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;temporal_size&quot;</span><span class="p">,</span> <span class="mi">16</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">MViT</span><span class="p">(</span>
+        <span class="n">spatial_size</span><span class="o">=</span><span class="n">spatial_size</span><span class="p">,</span>
+        <span class="n">temporal_size</span><span class="o">=</span><span class="n">temporal_size</span><span class="p">,</span>
+        <span class="n">block_setting</span><span class="o">=</span><span class="n">block_setting</span><span class="p">,</span>
+        <span class="n">residual_pool</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;residual_pool&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="n">residual_with_cls_embed</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;residual_with_cls_embed&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="n">rel_pos_embed</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;rel_pos_embed&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="n">proj_after_attn</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;proj_after_attn&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">stochastic_depth_prob</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<div class="viewcode-block" id="MViT_V1_B_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.mvit_v1_b.html#torchvision.models.video.MViT_V1_B_Weights">[docs]</a><span class="k">class</span> <span class="nc">MViT_V1_B_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mvit_v1_b-dbeb1030.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">VideoClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,),</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.45</span><span class="p">,</span> <span class="mf">0.45</span><span class="p">,</span> <span class="mf">0.45</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.225</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;min_temporal_size&quot;</span><span class="p">:</span> <span class="mi">16</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_KINETICS400_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/facebookresearch/pytorchvideo/blob/main/docs/source/model_zoo.md&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The weights were ported from the paper. The accuracies are estimated on video-level &quot;</span>
+                <span class="s2">&quot;with parameters `frame_rate=7.5`, `clips_per_video=5`, and `clip_len=16`&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">36610672</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">78.477</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.582</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">70.599</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">139.764</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="MViT_V2_S_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.mvit_v2_s.html#torchvision.models.video.MViT_V2_S_Weights">[docs]</a><span class="k">class</span> <span class="nc">MViT_V2_S_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mvit_v2_s-ae3be167.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">VideoClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,),</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.45</span><span class="p">,</span> <span class="mf">0.45</span><span class="p">,</span> <span class="mf">0.45</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.225</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;min_temporal_size&quot;</span><span class="p">:</span> <span class="mi">16</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_KINETICS400_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/facebookresearch/SlowFast/blob/main/MODEL_ZOO.md&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The weights were ported from the paper. The accuracies are estimated on video-level &quot;</span>
+                <span class="s2">&quot;with parameters `frame_rate=7.5`, `clips_per_video=5`, and `clip_len=16`&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">34537744</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">80.757</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.665</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">64.224</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">131.884</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="mvit_v1_b"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.mvit_v1_b.html#torchvision.models.video.mvit_v1_b">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MViT_V1_B_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mvit_v1_b</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MViT_V1_B_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MViT</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a base MViTV1 architecture from</span>
+<span class="sd">    `Multiscale Vision Transformers &lt;https://arxiv.org/abs/2104.11227&gt;`__.</span>
+
+<span class="sd">    .. betastatus:: video module</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.MViT_V1_B_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.MViT_V1_B_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.MViT``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/mvit.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.MViT_V1_B_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MViT_V1_B_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">config</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;num_heads&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
+        <span class="s2">&quot;input_channels&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">768</span><span class="p">],</span>
+        <span class="s2">&quot;output_channels&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">768</span><span class="p">],</span>
+        <span class="s2">&quot;kernel_q&quot;</span><span class="p">:</span> <span class="p">[[],</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[],</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[]],</span>
+        <span class="s2">&quot;kernel_kv&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;stride_q&quot;</span><span class="p">:</span> <span class="p">[[],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span> <span class="p">[],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span> <span class="p">[]],</span>
+        <span class="s2">&quot;stride_kv&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+        <span class="p">],</span>
+    <span class="p">}</span>
+
+    <span class="n">block_setting</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;num_heads&quot;</span><span class="p">])):</span>
+        <span class="n">block_setting</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">MSBlockConfig</span><span class="p">(</span>
+                <span class="n">num_heads</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;num_heads&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">input_channels</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;input_channels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">output_channels</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;output_channels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">kernel_q</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;kernel_q&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">kernel_kv</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;kernel_kv&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">stride_q</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;stride_q&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">stride_kv</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;stride_kv&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_mvit</span><span class="p">(</span>
+        <span class="n">spatial_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+        <span class="n">temporal_size</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">block_setting</span><span class="o">=</span><span class="n">block_setting</span><span class="p">,</span>
+        <span class="n">residual_pool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">residual_with_cls_embed</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;stochastic_depth_prob&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">),</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="mvit_v2_s"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.mvit_v2_s.html#torchvision.models.video.mvit_v2_s">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MViT_V2_S_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mvit_v2_s</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MViT_V2_S_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">MViT</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Constructs a small MViTV2 architecture from</span>
+<span class="sd">    `Multiscale Vision Transformers &lt;https://arxiv.org/abs/2104.11227&gt;`__ and</span>
+<span class="sd">    `MViTv2: Improved Multiscale Vision Transformers for Classification</span>
+<span class="sd">    and Detection &lt;https://arxiv.org/abs/2112.01526&gt;`__.</span>
+
+<span class="sd">    .. betastatus:: video module</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.MViT_V2_S_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.MViT_V2_S_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.MViT``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/mvit.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.MViT_V2_S_Weights</span>
+<span class="sd">            :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MViT_V2_S_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="n">config</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;num_heads&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
+        <span class="s2">&quot;input_channels&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">96</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">768</span><span class="p">],</span>
+        <span class="s2">&quot;output_channels&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">768</span><span class="p">,</span> <span class="mi">768</span><span class="p">],</span>
+        <span class="s2">&quot;kernel_q&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;kernel_kv&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;stride_q&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+        <span class="p">],</span>
+        <span class="s2">&quot;stride_kv&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+            <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
+        <span class="p">],</span>
+    <span class="p">}</span>
+
+    <span class="n">block_setting</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;num_heads&quot;</span><span class="p">])):</span>
+        <span class="n">block_setting</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">MSBlockConfig</span><span class="p">(</span>
+                <span class="n">num_heads</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;num_heads&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">input_channels</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;input_channels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">output_channels</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;output_channels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">kernel_q</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;kernel_q&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">kernel_kv</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;kernel_kv&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">stride_q</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;stride_q&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+                <span class="n">stride_kv</span><span class="o">=</span><span class="n">config</span><span class="p">[</span><span class="s2">&quot;stride_kv&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">],</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_mvit</span><span class="p">(</span>
+        <span class="n">spatial_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+        <span class="n">temporal_size</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">block_setting</span><span class="o">=</span><span class="n">block_setting</span><span class="p">,</span>
+        <span class="n">residual_pool</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">residual_with_cls_embed</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">rel_pos_embed</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">proj_after_attn</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;stochastic_depth_prob&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">),</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/video/resnet.html b/0.16/_modules/torchvision/models/video/resnet.html
new file mode 100644
index 00000000000..92ee1189093
--- /dev/null
+++ b/0.16/_modules/torchvision/models/video/resnet.html
@@ -0,0 +1,1172 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.video.resnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.video.resnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.video.resnet</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">VideoClassification</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_KINETICS400_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;VideoResNet&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;R3D_18_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MC3_18_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;R2Plus1D_18_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;r3d_18&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;mc3_18&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;r2plus1d_18&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">Conv3DSimple</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">midplanes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="o">=</span><span class="n">in_planes</span><span class="p">,</span>
+            <span class="n">out_channels</span><span class="o">=</span><span class="n">out_planes</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+            <span class="n">stride</span><span class="o">=</span><span class="n">stride</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span>
+            <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_downsample_stride</span><span class="p">(</span><span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span>
+
+
+<span class="k">class</span> <span class="nc">Conv2Plus1D</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">midplanes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span>
+                <span class="n">in_planes</span><span class="p">,</span>
+                <span class="n">midplanes</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+                <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span><span class="p">),</span>
+                <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="n">padding</span><span class="p">),</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="n">midplanes</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span>
+                <span class="n">midplanes</span><span class="p">,</span> <span class="n">out_planes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="n">stride</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="n">padding</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span>
+            <span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_downsample_stride</span><span class="p">(</span><span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span>
+
+
+<span class="k">class</span> <span class="nc">Conv3DNoTemporal</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">in_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">midplanes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="o">=</span><span class="n">in_planes</span><span class="p">,</span>
+            <span class="n">out_channels</span><span class="o">=</span><span class="n">out_planes</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span>
+            <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span><span class="p">),</span>
+            <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="n">padding</span><span class="p">),</span>
+            <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_downsample_stride</span><span class="p">(</span><span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span>
+
+
+<span class="k">class</span> <span class="nc">BasicBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+
+    <span class="n">expansion</span> <span class="o">=</span> <span class="mi">1</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inplanes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">conv_builder</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">downsample</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">midplanes</span> <span class="o">=</span> <span class="p">(</span><span class="n">inplanes</span> <span class="o">*</span> <span class="n">planes</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">*</span> <span class="mi">3</span><span class="p">)</span> <span class="o">//</span> <span class="p">(</span><span class="n">inplanes</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">+</span> <span class="mi">3</span> <span class="o">*</span> <span class="n">planes</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">conv_builder</span><span class="p">(</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">midplanes</span><span class="p">,</span> <span class="n">stride</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="n">planes</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">conv_builder</span><span class="p">(</span><span class="n">planes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">midplanes</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="n">planes</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">downsample</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">residual</span> <span class="o">=</span> <span class="n">x</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">residual</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">+=</span> <span class="n">residual</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">Bottleneck</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="n">expansion</span> <span class="o">=</span> <span class="mi">4</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inplanes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">conv_builder</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">downsample</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">midplanes</span> <span class="o">=</span> <span class="p">(</span><span class="n">inplanes</span> <span class="o">*</span> <span class="n">planes</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">*</span> <span class="mi">3</span><span class="p">)</span> <span class="o">//</span> <span class="p">(</span><span class="n">inplanes</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">*</span> <span class="mi">3</span> <span class="o">+</span> <span class="mi">3</span> <span class="o">*</span> <span class="n">planes</span><span class="p">)</span>
+
+        <span class="c1"># 1x1x1</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="n">planes</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="p">)</span>
+        <span class="c1"># Second kernel</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">conv_builder</span><span class="p">(</span><span class="n">planes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">midplanes</span><span class="p">,</span> <span class="n">stride</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="n">planes</span><span class="p">),</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="c1"># 1x1x1</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span><span class="n">planes</span><span class="p">,</span> <span class="n">planes</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">expansion</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="n">planes</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">expansion</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relu</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="o">=</span> <span class="n">downsample</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">residual</span> <span class="o">=</span> <span class="n">x</span>
+
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv2</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv3</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">residual</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">downsample</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">out</span> <span class="o">+=</span> <span class="n">residual</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">out</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">BasicStem</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The default conv-batchnorm-relu stem&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="mi">64</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">R2Plus1dStem</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;R(2+1)D stem is different than the default one as it uses separated 3D convolution&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">45</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="mi">45</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span><span class="mi">45</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="mi">64</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">VideoResNet</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">]],</span>
+        <span class="n">conv_makers</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Conv3DSimple</span><span class="p">,</span> <span class="n">Conv3DNoTemporal</span><span class="p">,</span> <span class="n">Conv2Plus1D</span><span class="p">]]],</span>
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">stem</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">400</span><span class="p">,</span>
+        <span class="n">zero_init_residual</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Generic resnet video generator.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            block (Type[Union[BasicBlock, Bottleneck]]): resnet building block</span>
+<span class="sd">            conv_makers (List[Type[Union[Conv3DSimple, Conv3DNoTemporal, Conv2Plus1D]]]): generator</span>
+<span class="sd">                function for each layer</span>
+<span class="sd">            layers (List[int]): number of blocks per layer</span>
+<span class="sd">            stem (Callable[..., nn.Module]): module specifying the ResNet stem.</span>
+<span class="sd">            num_classes (int, optional): Dimension of the final FC layer. Defaults to 400.</span>
+<span class="sd">            zero_init_residual (bool, optional): Zero init bottleneck residual BN. Defaults to False.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span> <span class="o">=</span> <span class="mi">64</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">stem</span> <span class="o">=</span> <span class="n">stem</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">conv_makers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mi">64</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">conv_makers</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="mi">128</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">conv_makers</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="mi">256</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer4</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">conv_makers</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="mi">512</span><span class="p">,</span> <span class="n">layers</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool3d</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="mi">512</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="c1"># init weights</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;fan_out&quot;</span><span class="p">,</span> <span class="n">nonlinearity</span><span class="o">=</span><span class="s2">&quot;relu&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mf">0.01</span><span class="p">)</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">zero_init_residual</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">):</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bn3</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>  <span class="c1"># type: ignore[union-attr, arg-type]</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">stem</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer4</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># Flatten the layer to fc</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">_make_layer</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">]],</span>
+        <span class="n">conv_builder</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Conv3DSimple</span><span class="p">,</span> <span class="n">Conv3DNoTemporal</span><span class="p">,</span> <span class="n">Conv2Plus1D</span><span class="p">]],</span>
+        <span class="n">planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">blocks</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">:</span>
+        <span class="n">downsample</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="n">stride</span> <span class="o">!=</span> <span class="mi">1</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span> <span class="o">!=</span> <span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">:</span>
+            <span class="n">ds_stride</span> <span class="o">=</span> <span class="n">conv_builder</span><span class="o">.</span><span class="n">get_downsample_stride</span><span class="p">(</span><span class="n">stride</span><span class="p">)</span>
+            <span class="n">downsample</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">ds_stride</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">(</span><span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="n">layers</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">conv_builder</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">downsample</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span> <span class="o">=</span> <span class="n">planes</span> <span class="o">*</span> <span class="n">block</span><span class="o">.</span><span class="n">expansion</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">blocks</span><span class="p">):</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inplanes</span><span class="p">,</span> <span class="n">planes</span><span class="p">,</span> <span class="n">conv_builder</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_video_resnet</span><span class="p">(</span>
+    <span class="n">block</span><span class="p">:</span> <span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">BasicBlock</span><span class="p">,</span> <span class="n">Bottleneck</span><span class="p">]],</span>
+    <span class="n">conv_makers</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Type</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Conv3DSimple</span><span class="p">,</span> <span class="n">Conv3DNoTemporal</span><span class="p">,</span> <span class="n">Conv2Plus1D</span><span class="p">]]],</span>
+    <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">stem</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VideoResNet</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">VideoResNet</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">conv_makers</span><span class="p">,</span> <span class="n">layers</span><span class="p">,</span> <span class="n">stem</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_KINETICS400_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/video_classification&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+        <span class="s2">&quot;The weights reproduce closely the accuracy of the paper. The accuracies are estimated on video-level &quot;</span>
+        <span class="s2">&quot;with parameters `frame_rate=15`, `clips_per_video=5`, and `clip_len=16`.&quot;</span>
+    <span class="p">),</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="R3D_18_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.r3d_18.html#torchvision.models.video.R3D_18_Weights">[docs]</a><span class="k">class</span> <span class="nc">R3D_18_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/r3d_18-b3b3357e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">VideoClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">112</span><span class="p">,</span> <span class="mi">112</span><span class="p">),</span> <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">171</span><span class="p">)),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">33371472</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">63.200</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">83.479</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">40.697</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">127.359</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="MC3_18_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.mc3_18.html#torchvision.models.video.MC3_18_Weights">[docs]</a><span class="k">class</span> <span class="nc">MC3_18_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/mc3_18-a90a0ba3.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">VideoClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">112</span><span class="p">,</span> <span class="mi">112</span><span class="p">),</span> <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">171</span><span class="p">)),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">11695440</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">63.960</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">84.130</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">43.343</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">44.672</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="R2Plus1D_18_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.r2plus1d_18.html#torchvision.models.video.R2Plus1D_18_Weights">[docs]</a><span class="k">class</span> <span class="nc">R2Plus1D_18_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/r2plus1d_18-91a641e6.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">VideoClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">112</span><span class="p">,</span> <span class="mi">112</span><span class="p">),</span> <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">171</span><span class="p">)),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">31505325</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">67.463</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">86.175</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">40.519</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">120.318</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="r3d_18"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.r3d_18.html#torchvision.models.video.r3d_18">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">R3D_18_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">r3d_18</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">R3D_18_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VideoResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Construct 18 layer Resnet3D model.</span>
+
+<span class="sd">    .. betastatus:: video module</span>
+
+<span class="sd">    Reference: `A Closer Look at Spatiotemporal Convolutions for Action Recognition &lt;https://arxiv.org/abs/1711.11248&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.R3D_18_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.R3D_18_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.resnet.VideoResNet`` base class.</span>
+<span class="sd">            Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.R3D_18_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">R3D_18_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_video_resnet</span><span class="p">(</span>
+        <span class="n">BasicBlock</span><span class="p">,</span>
+        <span class="p">[</span><span class="n">Conv3DSimple</span><span class="p">]</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">BasicStem</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="mc3_18"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.mc3_18.html#torchvision.models.video.mc3_18">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">MC3_18_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">mc3_18</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">MC3_18_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VideoResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Construct 18 layer Mixed Convolution network as in</span>
+
+<span class="sd">    .. betastatus:: video module</span>
+
+<span class="sd">    Reference: `A Closer Look at Spatiotemporal Convolutions for Action Recognition &lt;https://arxiv.org/abs/1711.11248&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.MC3_18_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.MC3_18_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.resnet.VideoResNet`` base class.</span>
+<span class="sd">            Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.MC3_18_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">MC3_18_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_video_resnet</span><span class="p">(</span>
+        <span class="n">BasicBlock</span><span class="p">,</span>
+        <span class="p">[</span><span class="n">Conv3DSimple</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="n">Conv3DNoTemporal</span><span class="p">]</span> <span class="o">*</span> <span class="mi">3</span><span class="p">,</span>  <span class="c1"># type: ignore[list-item]</span>
+        <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">BasicStem</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="r2plus1d_18"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.r2plus1d_18.html#torchvision.models.video.r2plus1d_18">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">R2Plus1D_18_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">r2plus1d_18</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">R2Plus1D_18_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VideoResNet</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Construct 18 layer deep R(2+1)D network as in</span>
+
+<span class="sd">    .. betastatus:: video module</span>
+
+<span class="sd">    Reference: `A Closer Look at Spatiotemporal Convolutions for Action Recognition &lt;https://arxiv.org/abs/1711.11248&gt;`__.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.R2Plus1D_18_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.R2Plus1D_18_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.resnet.VideoResNet`` base class.</span>
+<span class="sd">            Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.R2Plus1D_18_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">R2Plus1D_18_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_video_resnet</span><span class="p">(</span>
+        <span class="n">BasicBlock</span><span class="p">,</span>
+        <span class="p">[</span><span class="n">Conv2Plus1D</span><span class="p">]</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">R2Plus1dStem</span><span class="p">,</span>
+        <span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<span class="c1"># The dictionary below is internal implementation detail and will be removed in v0.15</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ModelURLs</span>
+
+
+<span class="n">model_urls</span> <span class="o">=</span> <span class="n">_ModelURLs</span><span class="p">(</span>
+    <span class="p">{</span>
+        <span class="s2">&quot;r3d_18&quot;</span><span class="p">:</span> <span class="n">R3D_18_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="o">.</span><span class="n">url</span><span class="p">,</span>
+        <span class="s2">&quot;mc3_18&quot;</span><span class="p">:</span> <span class="n">MC3_18_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="o">.</span><span class="n">url</span><span class="p">,</span>
+        <span class="s2">&quot;r2plus1d_18&quot;</span><span class="p">:</span> <span class="n">R2Plus1D_18_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="o">.</span><span class="n">url</span><span class="p">,</span>
+    <span class="p">}</span>
+<span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/video/s3d.html b/0.16/_modules/torchvision/models/video/s3d.html
new file mode 100644
index 00000000000..8d8d9d16cda
--- /dev/null
+++ b/0.16/_modules/torchvision/models/video/s3d.html
@@ -0,0 +1,888 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.video.s3d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.video.s3d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.video.s3d</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops.misc</span> <span class="kn">import</span> <span class="n">Conv3dNormActivation</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">VideoClassification</span>
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_KINETICS400_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;S3D&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;S3D_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;s3d&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">TemporalSeparableConv</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">Conv3dNormActivation</span><span class="p">(</span>
+                <span class="n">in_planes</span><span class="p">,</span>
+                <span class="n">out_planes</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">),</span>
+                <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="p">,</span> <span class="n">stride</span><span class="p">),</span>
+                <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="n">padding</span><span class="p">),</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="n">Conv3dNormActivation</span><span class="p">(</span>
+                <span class="n">out_planes</span><span class="p">,</span>
+                <span class="n">out_planes</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+                <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="n">stride</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+                <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="n">padding</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span>
+                <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">SepInceptionBlock3D</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_planes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">b0_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">b1_mid</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">b1_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">b2_mid</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">b2_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">b3_out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">],</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch0</span> <span class="o">=</span> <span class="n">Conv3dNormActivation</span><span class="p">(</span><span class="n">in_planes</span><span class="p">,</span> <span class="n">b0_out</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">Conv3dNormActivation</span><span class="p">(</span><span class="n">in_planes</span><span class="p">,</span> <span class="n">b1_mid</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">TemporalSeparableConv</span><span class="p">(</span><span class="n">b1_mid</span><span class="p">,</span> <span class="n">b1_out</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">Conv3dNormActivation</span><span class="p">(</span><span class="n">in_planes</span><span class="p">,</span> <span class="n">b2_mid</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">TemporalSeparableConv</span><span class="p">(</span><span class="n">b2_mid</span><span class="p">,</span> <span class="n">b2_out</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">branch3</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool3d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">Conv3dNormActivation</span><span class="p">(</span><span class="n">in_planes</span><span class="p">,</span> <span class="n">b3_out</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x0</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch0</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x1</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">x0</span><span class="p">,</span> <span class="n">x1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">x3</span><span class="p">),</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="k">class</span> <span class="nc">S3D</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;S3D main class.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_class (int): number of classes for the classification task.</span>
+<span class="sd">        dropout (float): dropout probability.</span>
+<span class="sd">        norm_layer (Optional[Callable]): Module specifying the normalization layer to use.</span>
+
+<span class="sd">    Inputs:</span>
+<span class="sd">        x (Tensor): batch of videos with dimensions (batch, channel, time, height, width)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">400</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">momentum</span><span class="o">=</span><span class="mf">0.001</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">TemporalSeparableConv</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool3d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)),</span>
+            <span class="n">Conv3dNormActivation</span><span class="p">(</span>
+                <span class="mi">64</span><span class="p">,</span>
+                <span class="mi">64</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="n">TemporalSeparableConv</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool3d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">192</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool3d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">480</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">208</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">24</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">112</span><span class="p">,</span> <span class="mi">144</span><span class="p">,</span> <span class="mi">288</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">528</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool3d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">832</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">160</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+            <span class="n">SepInceptionBlock3D</span><span class="p">(</span><span class="mi">832</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">384</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AvgPool3d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">),</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">classifier</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">))</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<div class="viewcode-block" id="S3D_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.s3d.html#torchvision.models.video.S3D_Weights">[docs]</a><span class="k">class</span> <span class="nc">S3D_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/s3d-d76dad2f.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">VideoClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;min_temporal_size&quot;</span><span class="p">:</span> <span class="mi">14</span><span class="p">,</span>
+            <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_KINETICS400_CATEGORIES</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/video_classification#s3d&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The weights aim to approximate the accuracy of the paper. The accuracies are estimated on clip-level &quot;</span>
+                <span class="s2">&quot;with parameters `frame_rate=15`, `clips_per_video=1`, and `clip_len=128`.&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">8320048</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">68.368</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">88.050</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">17.979</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">31.972</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="s3d"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.s3d.html#torchvision.models.video.s3d">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">S3D_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">s3d</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">S3D_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">S3D</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Construct Separable 3D CNN model.</span>
+
+<span class="sd">    Reference: `Rethinking Spatiotemporal Feature Learning &lt;https://arxiv.org/abs/1712.04851&gt;`__.</span>
+
+<span class="sd">    .. betastatus:: video module</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.S3D_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.S3D_Weights`</span>
+<span class="sd">            below for more details, and possible values. By default, no</span>
+<span class="sd">            pre-trained weights are used.</span>
+<span class="sd">        progress (bool): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.S3D`` base class.</span>
+<span class="sd">            Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/s3d.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.S3D_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">S3D_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">S3D</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/video/swin_transformer.html b/0.16/_modules/torchvision/models/video/swin_transformer.html
new file mode 100644
index 00000000000..a0faaf227ec
--- /dev/null
+++ b/0.16/_modules/torchvision/models/video/swin_transformer.html
@@ -0,0 +1,1412 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.video.swin_transformer &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.video.swin_transformer</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.video.swin_transformer</h1><div class="highlight"><pre>
+<span></span><span class="c1"># Modified from 2d Swin Transformers in torchvision:</span>
+<span class="c1"># https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py</span>
+
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">...transforms._presets</span> <span class="kn">import</span> <span class="n">VideoClassification</span>
+
+<span class="kn">from</span> <span class="nn">...utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">.._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+
+<span class="kn">from</span> <span class="nn">.._meta</span> <span class="kn">import</span> <span class="n">_KINETICS400_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+<span class="kn">from</span> <span class="nn">..swin_transformer</span> <span class="kn">import</span> <span class="n">PatchMerging</span><span class="p">,</span> <span class="n">SwinTransformerBlock</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;SwinTransformer3d&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin3D_T_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin3D_S_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Swin3D_B_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin3d_t&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin3d_s&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;swin3d_b&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_get_window_and_shift_size</span><span class="p">(</span>
+    <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">size_dhw</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">3</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">size_dhw</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">window_size</span><span class="p">[</span><span class="n">i</span><span class="p">]:</span>
+            <span class="c1"># In this case, window_size will adapt to the input size, and no need to shift</span>
+            <span class="n">window_size</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">size_dhw</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">shift_size</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+
+    <span class="k">return</span> <span class="n">window_size</span><span class="p">,</span> <span class="n">shift_size</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_get_window_and_shift_size&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_get_relative_position_bias</span><span class="p">(</span>
+    <span class="n">relative_position_bias_table</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">relative_position_index</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">window_vol</span> <span class="o">=</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+    <span class="c1"># In 3d case we flatten the relative_position_bias</span>
+    <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="n">relative_position_bias_table</span><span class="p">[</span>
+        <span class="n">relative_position_index</span><span class="p">[:</span><span class="n">window_vol</span><span class="p">,</span> <span class="p">:</span><span class="n">window_vol</span><span class="p">]</span><span class="o">.</span><span class="n">flatten</span><span class="p">()</span>  <span class="c1"># type: ignore[index]</span>
+    <span class="p">]</span>
+    <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="n">relative_position_bias</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">window_vol</span><span class="p">,</span> <span class="n">window_vol</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="n">relative_position_bias</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">relative_position_bias</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_get_relative_position_bias&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_compute_pad_size_3d</span><span class="p">(</span><span class="n">size_dhw</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">patch_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+    <span class="n">pad_size</span> <span class="o">=</span> <span class="p">[(</span><span class="n">patch_size</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">-</span> <span class="n">size_dhw</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">%</span> <span class="n">patch_size</span><span class="p">[</span><span class="n">i</span><span class="p">])</span> <span class="o">%</span> <span class="n">patch_size</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">3</span><span class="p">)]</span>
+    <span class="k">return</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_compute_pad_size_3d&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_compute_attention_mask_3d</span><span class="p">(</span>
+    <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">size_dhw</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">window_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">shift_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># generate attention mask</span>
+    <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">(</span><span class="o">*</span><span class="n">size_dhw</span><span class="p">)</span>
+    <span class="n">num_windows</span> <span class="o">=</span> <span class="p">(</span><span class="n">size_dhw</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">*</span> <span class="p">(</span><span class="n">size_dhw</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="o">*</span> <span class="p">(</span><span class="n">size_dhw</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">])</span>
+    <span class="n">slices</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">(</span>
+            <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="o">-</span><span class="n">window_size</span><span class="p">[</span><span class="n">i</span><span class="p">]),</span>
+            <span class="p">(</span><span class="o">-</span><span class="n">window_size</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="n">i</span><span class="p">]),</span>
+            <span class="p">(</span><span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="kc">None</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">3</span><span class="p">)</span>
+    <span class="p">]</span>
+    <span class="n">count</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <span class="n">slices</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+        <span class="k">for</span> <span class="n">h</span> <span class="ow">in</span> <span class="n">slices</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
+            <span class="k">for</span> <span class="n">w</span> <span class="ow">in</span> <span class="n">slices</span><span class="p">[</span><span class="mi">2</span><span class="p">]:</span>
+                <span class="n">attn_mask</span><span class="p">[</span><span class="n">d</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="p">:</span> <span class="n">d</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">h</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="p">:</span> <span class="n">h</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">w</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="p">:</span> <span class="n">w</span><span class="p">[</span><span class="mi">1</span><span class="p">]]</span> <span class="o">=</span> <span class="n">count</span>
+                <span class="n">count</span> <span class="o">+=</span> <span class="mi">1</span>
+
+    <span class="c1"># Partition window on attn_mask</span>
+    <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+        <span class="n">size_dhw</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">size_dhw</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>
+        <span class="n">size_dhw</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span>
+    <span class="p">)</span>
+    <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span>
+        <span class="n">num_windows</span><span class="p">,</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+    <span class="p">)</span>
+    <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">-</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+    <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">masked_fill</span><span class="p">(</span><span class="n">attn_mask</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="o">-</span><span class="mf">100.0</span><span class="p">))</span><span class="o">.</span><span class="n">masked_fill</span><span class="p">(</span><span class="n">attn_mask</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="mf">0.0</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">attn_mask</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;_compute_attention_mask_3d&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">shifted_window_attention_3d</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">qkv_weight</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">proj_weight</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">relative_position_bias</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="n">qkv_bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">proj_bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">training</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Window based multi-head self attention (W-MSA) module with relative position bias.</span>
+<span class="sd">    It supports both of shifted and non-shifted window.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[B, T, H, W, C]): The input tensor, 5-dimensions.</span>
+<span class="sd">        qkv_weight (Tensor[in_dim, out_dim]): The weight tensor of query, key, value.</span>
+<span class="sd">        proj_weight (Tensor[out_dim, out_dim]): The weight tensor of projection.</span>
+<span class="sd">        relative_position_bias (Tensor): The learned relative position bias added to attention.</span>
+<span class="sd">        window_size (List[int]): 3-dimensions window size, T, H, W .</span>
+<span class="sd">        num_heads (int): Number of attention heads.</span>
+<span class="sd">        shift_size (List[int]): Shift size for shifted window attention (T, H, W).</span>
+<span class="sd">        attention_dropout (float): Dropout ratio of attention weight. Default: 0.0.</span>
+<span class="sd">        dropout (float): Dropout ratio of output. Default: 0.0.</span>
+<span class="sd">        qkv_bias (Tensor[out_dim], optional): The bias tensor of query, key, value. Default: None.</span>
+<span class="sd">        proj_bias (Tensor[out_dim], optional): The bias tensor of projection. Default: None.</span>
+<span class="sd">        training (bool, optional): Training flag used by the dropout parameters. Default: True.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[B, T, H, W, C]: The output tensor after shifted window attention.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">b</span><span class="p">,</span> <span class="n">t</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">shape</span>
+    <span class="c1"># pad feature maps to multiples of window size</span>
+    <span class="n">pad_size</span> <span class="o">=</span> <span class="n">_compute_pad_size_3d</span><span class="p">((</span><span class="n">t</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">),</span> <span class="p">(</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]))</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="mi">0</span><span class="p">,</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="mi">0</span><span class="p">,</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+    <span class="n">_</span><span class="p">,</span> <span class="n">tp</span><span class="p">,</span> <span class="n">hp</span><span class="p">,</span> <span class="n">wp</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">padded_size</span> <span class="o">=</span> <span class="p">(</span><span class="n">tp</span><span class="p">,</span> <span class="n">hp</span><span class="p">,</span> <span class="n">wp</span><span class="p">)</span>
+
+    <span class="c1"># cyclic shift</span>
+    <span class="k">if</span> <span class="nb">sum</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">roll</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">shifts</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="o">-</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span> <span class="n">dims</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+
+    <span class="c1"># partition windows</span>
+    <span class="n">num_windows</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="p">(</span><span class="n">padded_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">*</span> <span class="p">(</span><span class="n">padded_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="o">*</span> <span class="p">(</span><span class="n">padded_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">])</span>
+    <span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+        <span class="n">b</span><span class="p">,</span>
+        <span class="n">padded_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">padded_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>
+        <span class="n">padded_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span>
+        <span class="n">c</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">7</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span>
+        <span class="n">b</span> <span class="o">*</span> <span class="n">num_windows</span><span class="p">,</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">c</span>
+    <span class="p">)</span>  <span class="c1"># B*nW, Wd*Wh*Ww, C</span>
+
+    <span class="c1"># multi-head attention</span>
+    <span class="n">qkv</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">linear</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">qkv_weight</span><span class="p">,</span> <span class="n">qkv_bias</span><span class="p">)</span>
+    <span class="n">qkv</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="mi">3</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">c</span> <span class="o">//</span> <span class="n">num_heads</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+    <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="n">qkv</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+    <span class="n">q</span> <span class="o">=</span> <span class="n">q</span> <span class="o">*</span> <span class="p">(</span><span class="n">c</span> <span class="o">//</span> <span class="n">num_heads</span><span class="p">)</span> <span class="o">**</span> <span class="o">-</span><span class="mf">0.5</span>
+    <span class="n">attn</span> <span class="o">=</span> <span class="n">q</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">k</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+    <span class="c1"># add relative position bias</span>
+    <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span> <span class="o">+</span> <span class="n">relative_position_bias</span>
+
+    <span class="k">if</span> <span class="nb">sum</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="c1"># generate attention mask to handle shifted windows with varying size</span>
+        <span class="n">attn_mask</span> <span class="o">=</span> <span class="n">_compute_attention_mask_3d</span><span class="p">(</span>
+            <span class="n">x</span><span class="p">,</span>
+            <span class="p">(</span><span class="n">padded_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">padded_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">padded_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span>
+            <span class="p">(</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span>
+            <span class="p">(</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">shift_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span>
+        <span class="p">)</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">//</span> <span class="n">num_windows</span><span class="p">,</span> <span class="n">num_windows</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">))</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span> <span class="o">+</span> <span class="n">attn_mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">attn</span> <span class="o">=</span> <span class="n">attn</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">))</span>
+
+    <span class="n">attn</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">attn</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">attn</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">attn</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="n">training</span><span class="p">)</span>
+
+    <span class="n">x</span> <span class="o">=</span> <span class="n">attn</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">v</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">c</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">linear</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">proj_weight</span><span class="p">,</span> <span class="n">proj_bias</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="n">training</span><span class="p">)</span>
+
+    <span class="c1"># reverse windows</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+        <span class="n">b</span><span class="p">,</span>
+        <span class="n">padded_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">padded_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>
+        <span class="n">padded_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">//</span> <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span>
+        <span class="n">c</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">7</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="n">tp</span><span class="p">,</span> <span class="n">hp</span><span class="p">,</span> <span class="n">wp</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span>
+
+    <span class="c1"># reverse cyclic shift</span>
+    <span class="k">if</span> <span class="nb">sum</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">roll</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">shifts</span><span class="o">=</span><span class="p">(</span><span class="n">shift_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">shift_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">shift_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]),</span> <span class="n">dims</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+
+    <span class="c1"># unpad features</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="p">[:,</span> <span class="p">:</span><span class="n">t</span><span class="p">,</span> <span class="p">:</span><span class="n">h</span><span class="p">,</span> <span class="p">:</span><span class="n">w</span><span class="p">,</span> <span class="p">:]</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+    <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;shifted_window_attention_3d&quot;</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">ShiftedWindowAttention3d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`shifted_window_attention_3d`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">shift_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">qkv_bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">proj_bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">window_size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">3</span> <span class="ow">or</span> <span class="nb">len</span><span class="p">(</span><span class="n">shift_size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;window_size and shift_size must be of length 2&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span> <span class="o">=</span> <span class="n">window_size</span>  <span class="c1"># Wd, Wh, Ww</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">shift_size</span> <span class="o">=</span> <span class="n">shift_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span> <span class="o">=</span> <span class="n">num_heads</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">attention_dropout</span> <span class="o">=</span> <span class="n">attention_dropout</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">dropout</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="n">dim</span> <span class="o">*</span> <span class="mi">3</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="n">qkv_bias</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="n">dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="n">proj_bias</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">define_relative_position_bias_table</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">define_relative_position_index</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">define_relative_position_bias_table</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># define a parameter table of relative position bias</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span>
+                <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">),</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">)</span>  <span class="c1"># 2*Wd-1 * 2*Wh-1 * 2*Ww-1, nH</span>
+        <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">define_relative_position_index</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># get pair-wise relative position index for each token inside the window</span>
+        <span class="n">coords_dhw</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="n">i</span><span class="p">])</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">3</span><span class="p">)]</span>
+        <span class="n">coords</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">meshgrid</span><span class="p">(</span><span class="n">coords_dhw</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">coords_dhw</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">coords_dhw</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">indexing</span><span class="o">=</span><span class="s2">&quot;ij&quot;</span><span class="p">)</span>
+        <span class="p">)</span>  <span class="c1"># 3, Wd, Wh, Ww</span>
+        <span class="n">coords_flatten</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">coords</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>  <span class="c1"># 3, Wd*Wh*Ww</span>
+        <span class="n">relative_coords</span> <span class="o">=</span> <span class="n">coords_flatten</span><span class="p">[:,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">-</span> <span class="n">coords_flatten</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span>  <span class="c1"># 3, Wd*Wh*Ww, Wd*Wh*Ww</span>
+        <span class="n">relative_coords</span> <span class="o">=</span> <span class="n">relative_coords</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>  <span class="c1"># Wd*Wh*Ww, Wd*Wh*Ww, 3</span>
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>  <span class="c1"># shift to start from 0</span>
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">*=</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">relative_coords</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">*=</span> <span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="c1"># We don&#39;t flatten the relative_position_index here in 3d case.</span>
+        <span class="n">relative_position_index</span> <span class="o">=</span> <span class="n">relative_coords</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># Wd*Wh*Ww, Wd*Wh*Ww</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;relative_position_index&quot;</span><span class="p">,</span> <span class="n">relative_position_index</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">get_relative_position_bias</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_get_relative_position_bias</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">relative_position_bias_table</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">relative_position_index</span><span class="p">,</span> <span class="n">window_size</span><span class="p">)</span>  <span class="c1"># type: ignore</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">t</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">size_dhw</span> <span class="o">=</span> <span class="p">[</span><span class="n">t</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">]</span>
+        <span class="n">window_size</span><span class="p">,</span> <span class="n">shift_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">window_size</span><span class="o">.</span><span class="n">copy</span><span class="p">(),</span> <span class="bp">self</span><span class="o">.</span><span class="n">shift_size</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+        <span class="c1"># Handle case where window_size is larger than the input tensor</span>
+        <span class="n">window_size</span><span class="p">,</span> <span class="n">shift_size</span> <span class="o">=</span> <span class="n">_get_window_and_shift_size</span><span class="p">(</span><span class="n">shift_size</span><span class="p">,</span> <span class="n">size_dhw</span><span class="p">,</span> <span class="n">window_size</span><span class="p">)</span>
+
+        <span class="n">relative_position_bias</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_relative_position_bias</span><span class="p">(</span><span class="n">window_size</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">shifted_window_attention_3d</span><span class="p">(</span>
+            <span class="n">x</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span>
+            <span class="n">relative_position_bias</span><span class="p">,</span>
+            <span class="n">window_size</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">shift_size</span><span class="o">=</span><span class="n">shift_size</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">,</span>
+            <span class="n">qkv_bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">qkv</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
+            <span class="n">proj_bias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
+            <span class="n">training</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">,</span>
+        <span class="p">)</span>
+
+
+<span class="c1"># Modified from:</span>
+<span class="c1"># https://github.com/SwinTransformer/Video-Swin-Transformer/blob/master/mmaction/models/backbones/swin_transformer.py</span>
+<span class="k">class</span> <span class="nc">PatchEmbed3d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Video to Patch Embedding.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        patch_size (List[int]): Patch token size.</span>
+<span class="sd">        in_channels (int): Number of input channels. Default: 3</span>
+<span class="sd">        embed_dim (int): Number of linear projection output channels. Default: 96.</span>
+<span class="sd">        norm_layer (nn.Module, optional): Normalization layer. Default: None</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">patch_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">embed_dim</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">96</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">tuple_patch_size</span> <span class="o">=</span> <span class="p">(</span><span class="n">patch_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">patch_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">patch_size</span><span class="p">[</span><span class="mi">2</span><span class="p">])</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">embed_dim</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tuple_patch_size</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tuple_patch_size</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">embed_dim</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Forward function.&quot;&quot;&quot;</span>
+        <span class="c1"># padding</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">t</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+        <span class="n">pad_size</span> <span class="o">=</span> <span class="n">_compute_pad_size_3d</span><span class="p">((</span><span class="n">t</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">tuple_patch_size</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="mi">0</span><span class="p">,</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="mi">0</span><span class="p">,</span> <span class="n">pad_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">proj</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># B C T Wh Ww</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>  <span class="c1"># B T Wh Ww C</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">class</span> <span class="nc">SwinTransformer3d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements 3D Swin Transformer from the `&quot;Video Swin Transformer&quot; &lt;https://arxiv.org/abs/2106.13230&gt;`_ paper.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        patch_size (List[int]): Patch size.</span>
+<span class="sd">        embed_dim (int): Patch embedding dimension.</span>
+<span class="sd">        depths (List(int)): Depth of each Swin Transformer layer.</span>
+<span class="sd">        num_heads (List(int)): Number of attention heads in different layers.</span>
+<span class="sd">        window_size (List[int]): Window size.</span>
+<span class="sd">        mlp_ratio (float): Ratio of mlp hidden dim to embedding dim. Default: 4.0.</span>
+<span class="sd">        dropout (float): Dropout rate. Default: 0.0.</span>
+<span class="sd">        attention_dropout (float): Attention dropout rate. Default: 0.0.</span>
+<span class="sd">        stochastic_depth_prob (float): Stochastic depth rate. Default: 0.1.</span>
+<span class="sd">        num_classes (int): Number of classes for classification head. Default: 400.</span>
+<span class="sd">        norm_layer (nn.Module, optional): Normalization layer. Default: None.</span>
+<span class="sd">        block (nn.Module, optional): SwinTransformer Block. Default: None.</span>
+<span class="sd">        downsample_layer (nn.Module): Downsample layer (patch merging). Default: PatchMerging.</span>
+<span class="sd">        patch_embed (nn.Module, optional): Patch Embedding layer. Default: None.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">patch_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">depths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">mlp_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">4.0</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">400</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">block</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">downsample_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">PatchMerging</span><span class="p">,</span>
+        <span class="n">patch_embed</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+
+        <span class="k">if</span> <span class="n">block</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">SwinTransformerBlock</span><span class="p">,</span> <span class="n">attn_layer</span><span class="o">=</span><span class="n">ShiftedWindowAttention3d</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">norm_layer</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-5</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">patch_embed</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">patch_embed</span> <span class="o">=</span> <span class="n">PatchEmbed3d</span>
+
+        <span class="c1"># split image into non-overlapping patches</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">patch_embed</span> <span class="o">=</span> <span class="n">patch_embed</span><span class="p">(</span><span class="n">patch_size</span><span class="o">=</span><span class="n">patch_size</span><span class="p">,</span> <span class="n">embed_dim</span><span class="o">=</span><span class="n">embed_dim</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pos_drop</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">dropout</span><span class="p">)</span>
+
+        <span class="n">layers</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">total_stage_blocks</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="n">depths</span><span class="p">)</span>
+        <span class="n">stage_block_id</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="c1"># build SwinTransformer blocks</span>
+        <span class="k">for</span> <span class="n">i_stage</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">depths</span><span class="p">)):</span>
+            <span class="n">stage</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">dim</span> <span class="o">=</span> <span class="n">embed_dim</span> <span class="o">*</span> <span class="mi">2</span><span class="o">**</span><span class="n">i_stage</span>
+            <span class="k">for</span> <span class="n">i_layer</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">depths</span><span class="p">[</span><span class="n">i_stage</span><span class="p">]):</span>
+                <span class="c1"># adjust stochastic depth probability based on the depth of the stage block</span>
+                <span class="n">sd_prob</span> <span class="o">=</span> <span class="n">stochastic_depth_prob</span> <span class="o">*</span> <span class="nb">float</span><span class="p">(</span><span class="n">stage_block_id</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="n">total_stage_blocks</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="n">stage</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                    <span class="n">block</span><span class="p">(</span>
+                        <span class="n">dim</span><span class="p">,</span>
+                        <span class="n">num_heads</span><span class="p">[</span><span class="n">i_stage</span><span class="p">],</span>
+                        <span class="n">window_size</span><span class="o">=</span><span class="n">window_size</span><span class="p">,</span>
+                        <span class="n">shift_size</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span> <span class="k">if</span> <span class="n">i_layer</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">w</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">for</span> <span class="n">w</span> <span class="ow">in</span> <span class="n">window_size</span><span class="p">],</span>
+                        <span class="n">mlp_ratio</span><span class="o">=</span><span class="n">mlp_ratio</span><span class="p">,</span>
+                        <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">,</span>
+                        <span class="n">attention_dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span>
+                        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">sd_prob</span><span class="p">,</span>
+                        <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span>
+                        <span class="n">attn_layer</span><span class="o">=</span><span class="n">ShiftedWindowAttention3d</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="p">)</span>
+                <span class="n">stage_block_id</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">stage</span><span class="p">))</span>
+            <span class="c1"># add patch merging layer</span>
+            <span class="k">if</span> <span class="n">i_stage</span> <span class="o">&lt;</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">depths</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
+                <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">downsample_layer</span><span class="p">(</span><span class="n">dim</span><span class="p">,</span> <span class="n">norm_layer</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">features</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_features</span> <span class="o">=</span> <span class="n">embed_dim</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">**</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">depths</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_features</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool3d</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">head</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_features</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># x: B C T H W</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">patch_embed</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># B _T _H _W C</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">pos_drop</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">features</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>  <span class="c1"># B _T _H _W C</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>  <span class="c1"># B, C, _T, _H, _W</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">def</span> <span class="nf">_swin_transformer3d</span><span class="p">(</span>
+    <span class="n">patch_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">embed_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">depths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">num_heads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">window_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">stochastic_depth_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer3d</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">SwinTransformer3d</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="n">patch_size</span><span class="p">,</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="n">embed_dim</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="n">depths</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="n">num_heads</span><span class="p">,</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="n">window_size</span><span class="p">,</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="n">stochastic_depth_prob</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_KINETICS400_CATEGORIES</span><span class="p">,</span>
+    <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="s2">&quot;min_temporal_size&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="Swin3D_T_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.swin3d_t.html#torchvision.models.video.Swin3D_T_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin3D_T_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin3d_t-7615ae03.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">VideoClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,),</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.4850</span><span class="p">,</span> <span class="mf">0.4560</span><span class="p">,</span> <span class="mf">0.4060</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.2290</span><span class="p">,</span> <span class="mf">0.2240</span><span class="p">,</span> <span class="mf">0.2250</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The weights were ported from the paper. The accuracies are estimated on video-level &quot;</span>
+                <span class="s2">&quot;with parameters `frame_rate=15`, `clips_per_video=12`, and `clip_len=32`&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">28158070</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">77.715</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.519</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">43.882</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">121.543</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="Swin3D_S_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.swin3d_s.html#torchvision.models.video.Swin3D_S_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin3D_S_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin3d_s-da41c237.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">VideoClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,),</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.4850</span><span class="p">,</span> <span class="mf">0.4560</span><span class="p">,</span> <span class="mf">0.4060</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.2290</span><span class="p">,</span> <span class="mf">0.2240</span><span class="p">,</span> <span class="mf">0.2250</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The weights were ported from the paper. The accuracies are estimated on video-level &quot;</span>
+                <span class="s2">&quot;with parameters `frame_rate=15`, `clips_per_video=12`, and `clip_len=32`&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">49816678</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">79.521</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.158</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">82.841</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">218.288</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="Swin3D_B_Weights"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.swin3d_b.html#torchvision.models.video.Swin3D_B_Weights">[docs]</a><span class="k">class</span> <span class="nc">Swin3D_B_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">KINETICS400_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin3d_b_1k-24f7c7c6.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">VideoClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,),</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.4850</span><span class="p">,</span> <span class="mf">0.4560</span><span class="p">,</span> <span class="mf">0.4060</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.2290</span><span class="p">,</span> <span class="mf">0.2240</span><span class="p">,</span> <span class="mf">0.2250</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The weights were ported from the paper. The accuracies are estimated on video-level &quot;</span>
+                <span class="s2">&quot;with parameters `frame_rate=15`, `clips_per_video=12`, and `clip_len=32`&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88048984</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">79.427</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.386</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">140.667</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">364.134</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">KINETICS400_IMAGENET22K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/swin3d_b_22k-7c6ae6fa.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">VideoClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="p">(</span><span class="mi">256</span><span class="p">,),</span>
+            <span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.4850</span><span class="p">,</span> <span class="mf">0.4560</span><span class="p">,</span> <span class="mf">0.4060</span><span class="p">),</span>
+            <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.2290</span><span class="p">,</span> <span class="mf">0.2240</span><span class="p">,</span> <span class="mf">0.2250</span><span class="p">),</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The weights were ported from the paper. The accuracies are estimated on video-level &quot;</span>
+                <span class="s2">&quot;with parameters `frame_rate=15`, `clips_per_video=12`, and `clip_len=32`&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88048984</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;Kinetics-400&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.643</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.574</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">140.667</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">364.134</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">KINETICS400_V1</span></div>
+
+
+<div class="viewcode-block" id="swin3d_t"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.swin3d_t.html#torchvision.models.video.swin3d_t">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin3D_T_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin3d_t</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin3D_T_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer3d</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_tiny architecture from</span>
+<span class="sd">    `Video Swin Transformer &lt;https://arxiv.org/abs/2106.13230&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.Swin3D_T_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.Swin3D_T_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.Swin3D_T_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin3D_T_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer3d</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">96</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="swin3d_s"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.swin3d_s.html#torchvision.models.video.swin3d_s">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin3D_S_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin3d_s</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin3D_S_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer3d</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_small architecture from</span>
+<span class="sd">    `Video Swin Transformer &lt;https://arxiv.org/abs/2106.13230&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.Swin3D_S_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.Swin3D_S_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.Swin3D_S_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin3D_S_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer3d</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">96</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">,</span> <span class="mi">24</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="swin3d_b"><a class="viewcode-back" href="../../../../models/generated/torchvision.models.video.swin3d_b.html#torchvision.models.video.swin3d_b">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">Swin3D_B_Weights</span><span class="o">.</span><span class="n">KINETICS400_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">swin3d_b</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Swin3D_B_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SwinTransformer3d</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a swin_base architecture from</span>
+<span class="sd">    `Video Swin Transformer &lt;https://arxiv.org/abs/2106.13230&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.video.Swin3D_B_Weights`, optional): The</span>
+<span class="sd">            pretrained weights to use. See</span>
+<span class="sd">            :class:`~torchvision.models.video.Swin3D_B_Weights` below for</span>
+<span class="sd">            more details, and possible values. By default, no pre-trained</span>
+<span class="sd">            weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the</span>
+<span class="sd">            download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.video.swin_transformer.SwinTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.video.Swin3D_B_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">Swin3D_B_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_swin_transformer3d</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span>
+        <span class="n">embed_dim</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
+        <span class="n">depths</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">18</span><span class="p">,</span> <span class="mi">2</span><span class="p">],</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">32</span><span class="p">],</span>
+        <span class="n">window_size</span><span class="o">=</span><span class="p">[</span><span class="mi">8</span><span class="p">,</span> <span class="mi">7</span><span class="p">,</span> <span class="mi">7</span><span class="p">],</span>
+        <span class="n">stochastic_depth_prob</span><span class="o">=</span><span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/models/vision_transformer.html b/0.16/_modules/torchvision/models/vision_transformer.html
new file mode 100644
index 00000000000..fc4c1d51823
--- /dev/null
+++ b/0.16/_modules/torchvision/models/vision_transformer.html
@@ -0,0 +1,1533 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.models.vision_transformer &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.models.vision_transformer</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.models.vision_transformer</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">NamedTuple</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span><span class="p">,</span> <span class="n">MLP</span>
+<span class="kn">from</span> <span class="nn">..transforms._presets</span> <span class="kn">import</span> <span class="n">ImageClassification</span><span class="p">,</span> <span class="n">InterpolationMode</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._api</span> <span class="kn">import</span> <span class="n">register_model</span><span class="p">,</span> <span class="n">Weights</span><span class="p">,</span> <span class="n">WeightsEnum</span>
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_IMAGENET_CATEGORIES</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_ovewrite_named_param</span><span class="p">,</span> <span class="n">handle_legacy_interface</span>
+
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;VisionTransformer&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ViT_B_16_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ViT_B_32_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ViT_L_16_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ViT_L_32_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ViT_H_14_Weights&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vit_b_16&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vit_b_32&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vit_l_16&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vit_l_32&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;vit_h_14&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">ConvStemConfig</span><span class="p">(</span><span class="n">NamedTuple</span><span class="p">):</span>
+    <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span>
+    <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span>
+
+
+<span class="k">class</span> <span class="nc">MLPBlock</span><span class="p">(</span><span class="n">MLP</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Transformer MLP block.&quot;&quot;&quot;</span>
+
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">mlp_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">in_dim</span><span class="p">,</span> <span class="p">[</span><span class="n">mlp_dim</span><span class="p">,</span> <span class="n">in_dim</span><span class="p">],</span> <span class="n">activation_layer</span><span class="o">=</span><span class="n">nn</span><span class="o">.</span><span class="n">GELU</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">dropout</span><span class="o">=</span><span class="n">dropout</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">xavier_uniform_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">,</span>
+        <span class="n">unexpected_keys</span><span class="p">,</span>
+        <span class="n">error_msgs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">version</span> <span class="o">=</span> <span class="n">local_metadata</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">version</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">version</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="c1"># Replacing legacy MLPBlock with MLP. See https://github.com/pytorch/vision/pull/6053</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">2</span><span class="p">):</span>
+                <span class="k">for</span> <span class="nb">type</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;weight&quot;</span><span class="p">,</span> <span class="s2">&quot;bias&quot;</span><span class="p">]:</span>
+                    <span class="n">old_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">linear_</span><span class="si">{</span><span class="n">i</span><span class="o">+</span><span class="mi">1</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="n">new_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}{</span><span class="mi">3</span><span class="o">*</span><span class="n">i</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="k">if</span> <span class="n">old_key</span> <span class="ow">in</span> <span class="n">state_dict</span><span class="p">:</span>
+                        <span class="n">state_dict</span><span class="p">[</span><span class="n">new_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">state_dict</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">old_key</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span>
+            <span class="n">prefix</span><span class="p">,</span>
+            <span class="n">local_metadata</span><span class="p">,</span>
+            <span class="n">strict</span><span class="p">,</span>
+            <span class="n">missing_keys</span><span class="p">,</span>
+            <span class="n">unexpected_keys</span><span class="p">,</span>
+            <span class="n">error_msgs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">EncoderBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Transformer encoder block.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">),</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span> <span class="o">=</span> <span class="n">num_heads</span>
+
+        <span class="c1"># Attention block</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ln_1</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">self_attention</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MultiheadAttention</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">num_heads</span><span class="p">,</span> <span class="n">dropout</span><span class="o">=</span><span class="n">attention_dropout</span><span class="p">,</span> <span class="n">batch_first</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">dropout</span><span class="p">)</span>
+
+        <span class="c1"># MLP block</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ln_2</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mlp</span> <span class="o">=</span> <span class="n">MLPBlock</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">mlp_dim</span><span class="p">,</span> <span class="n">dropout</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">3</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;Expected (batch_size, seq_length, hidden_dim) got </span><span class="si">{</span><span class="nb">input</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ln_1</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="n">x</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">self_attention</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">need_weights</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="nb">input</span>
+
+        <span class="n">y</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ln_2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mlp</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span> <span class="o">+</span> <span class="n">y</span>
+
+
+<span class="k">class</span> <span class="nc">Encoder</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Transformer Model Encoder for sequence to sequence translation.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">seq_length</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">),</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="c1"># Note that batch_size is on the first dim because</span>
+        <span class="c1"># we have batch_first=True in nn.MultiAttention() by default</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pos_embedding</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">seq_length</span><span class="p">,</span> <span class="n">hidden_dim</span><span class="p">)</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span><span class="n">std</span><span class="o">=</span><span class="mf">0.02</span><span class="p">))</span>  <span class="c1"># from BERT</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">dropout</span><span class="p">)</span>
+        <span class="n">layers</span><span class="p">:</span> <span class="n">OrderedDict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_layers</span><span class="p">):</span>
+            <span class="n">layers</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;encoder_layer_</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">EncoderBlock</span><span class="p">(</span>
+                <span class="n">num_heads</span><span class="p">,</span>
+                <span class="n">hidden_dim</span><span class="p">,</span>
+                <span class="n">mlp_dim</span><span class="p">,</span>
+                <span class="n">dropout</span><span class="p">,</span>
+                <span class="n">attention_dropout</span><span class="p">,</span>
+                <span class="n">norm_layer</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">layers</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ln</span> <span class="o">=</span> <span class="n">norm_layer</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">3</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;Expected (batch_size, seq_length, hidden_dim) got </span><span class="si">{</span><span class="nb">input</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="nb">input</span> <span class="o">=</span> <span class="nb">input</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">pos_embedding</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">ln</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="nb">input</span><span class="p">)))</span>
+
+
+<span class="k">class</span> <span class="nc">VisionTransformer</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Vision Transformer as per https://arxiv.org/abs/2010.11929.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">image_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">patch_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">attention_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
+        <span class="n">representation_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">),</span>
+        <span class="n">conv_stem_configs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">ConvStemConfig</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="n">image_size</span> <span class="o">%</span> <span class="n">patch_size</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span> <span class="s2">&quot;Input shape indivisible by patch size!&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_size</span> <span class="o">=</span> <span class="n">image_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">patch_size</span> <span class="o">=</span> <span class="n">patch_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hidden_dim</span> <span class="o">=</span> <span class="n">hidden_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mlp_dim</span> <span class="o">=</span> <span class="n">mlp_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">attention_dropout</span> <span class="o">=</span> <span class="n">attention_dropout</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">dropout</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">representation_size</span> <span class="o">=</span> <span class="n">representation_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">norm_layer</span> <span class="o">=</span> <span class="n">norm_layer</span>
+
+        <span class="k">if</span> <span class="n">conv_stem_configs</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># As per https://arxiv.org/abs/2106.14881</span>
+            <span class="n">seq_proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">()</span>
+            <span class="n">prev_channels</span> <span class="o">=</span> <span class="mi">3</span>
+            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">conv_stem_layer_config</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">conv_stem_configs</span><span class="p">):</span>
+                <span class="n">seq_proj</span><span class="o">.</span><span class="n">add_module</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;conv_bn_relu_</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+                    <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                        <span class="n">in_channels</span><span class="o">=</span><span class="n">prev_channels</span><span class="p">,</span>
+                        <span class="n">out_channels</span><span class="o">=</span><span class="n">conv_stem_layer_config</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span>
+                        <span class="n">kernel_size</span><span class="o">=</span><span class="n">conv_stem_layer_config</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">,</span>
+                        <span class="n">stride</span><span class="o">=</span><span class="n">conv_stem_layer_config</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span>
+                        <span class="n">norm_layer</span><span class="o">=</span><span class="n">conv_stem_layer_config</span><span class="o">.</span><span class="n">norm_layer</span><span class="p">,</span>
+                        <span class="n">activation_layer</span><span class="o">=</span><span class="n">conv_stem_layer_config</span><span class="o">.</span><span class="n">activation_layer</span><span class="p">,</span>
+                    <span class="p">),</span>
+                <span class="p">)</span>
+                <span class="n">prev_channels</span> <span class="o">=</span> <span class="n">conv_stem_layer_config</span><span class="o">.</span><span class="n">out_channels</span>
+            <span class="n">seq_proj</span><span class="o">.</span><span class="n">add_module</span><span class="p">(</span>
+                <span class="s2">&quot;conv_last&quot;</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="o">=</span><span class="n">prev_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span> <span class="o">=</span> <span class="n">seq_proj</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span>
+                <span class="n">in_channels</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">patch_size</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">patch_size</span>
+            <span class="p">)</span>
+
+        <span class="n">seq_length</span> <span class="o">=</span> <span class="p">(</span><span class="n">image_size</span> <span class="o">//</span> <span class="n">patch_size</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span>
+
+        <span class="c1"># Add a class token</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">class_token</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">hidden_dim</span><span class="p">))</span>
+        <span class="n">seq_length</span> <span class="o">+=</span> <span class="mi">1</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">encoder</span> <span class="o">=</span> <span class="n">Encoder</span><span class="p">(</span>
+            <span class="n">seq_length</span><span class="p">,</span>
+            <span class="n">num_layers</span><span class="p">,</span>
+            <span class="n">num_heads</span><span class="p">,</span>
+            <span class="n">hidden_dim</span><span class="p">,</span>
+            <span class="n">mlp_dim</span><span class="p">,</span>
+            <span class="n">dropout</span><span class="p">,</span>
+            <span class="n">attention_dropout</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">seq_length</span> <span class="o">=</span> <span class="n">seq_length</span>
+
+        <span class="n">heads_layers</span><span class="p">:</span> <span class="n">OrderedDict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">representation_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">heads_layers</span><span class="p">[</span><span class="s2">&quot;head&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">heads_layers</span><span class="p">[</span><span class="s2">&quot;pre_logits&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">representation_size</span><span class="p">)</span>
+            <span class="n">heads_layers</span><span class="p">[</span><span class="s2">&quot;act&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Tanh</span><span class="p">()</span>
+            <span class="n">heads_layers</span><span class="p">[</span><span class="s2">&quot;head&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">representation_size</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">heads</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="n">heads_layers</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+            <span class="c1"># Init the patchify stem</span>
+            <span class="n">fan_in</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">in_channels</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mi">1</span> <span class="o">/</span> <span class="n">fan_in</span><span class="p">))</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">conv_last</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">conv_last</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+            <span class="c1"># Init the last 1x1 conv of the conv stem</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">normal_</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">conv_last</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mf">2.0</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">conv_last</span><span class="o">.</span><span class="n">out_channels</span><span class="p">)</span>
+            <span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">conv_last</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="o">.</span><span class="n">conv_last</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="p">,</span> <span class="s2">&quot;pre_logits&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="o">.</span><span class="n">pre_logits</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+            <span class="n">fan_in</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="o">.</span><span class="n">pre_logits</span><span class="o">.</span><span class="n">in_features</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">trunc_normal_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="o">.</span><span class="n">pre_logits</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mi">1</span> <span class="o">/</span> <span class="n">fan_in</span><span class="p">))</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="o">.</span><span class="n">pre_logits</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="o">.</span><span class="n">head</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">):</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="o">.</span><span class="n">head</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">zeros_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="o">.</span><span class="n">head</span><span class="o">.</span><span class="n">bias</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_process_input</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">n</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">p</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">patch_size</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="n">h</span> <span class="o">==</span> <span class="bp">self</span><span class="o">.</span><span class="n">image_size</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;Wrong image height! Expected </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">image_size</span><span class="si">}</span><span class="s2"> but got </span><span class="si">{</span><span class="n">h</span><span class="si">}</span><span class="s2">!&quot;</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_assert</span><span class="p">(</span><span class="n">w</span> <span class="o">==</span> <span class="bp">self</span><span class="o">.</span><span class="n">image_size</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;Wrong image width! Expected </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">image_size</span><span class="si">}</span><span class="s2"> but got </span><span class="si">{</span><span class="n">w</span><span class="si">}</span><span class="s2">!&quot;</span><span class="p">)</span>
+        <span class="n">n_h</span> <span class="o">=</span> <span class="n">h</span> <span class="o">//</span> <span class="n">p</span>
+        <span class="n">n_w</span> <span class="o">=</span> <span class="n">w</span> <span class="o">//</span> <span class="n">p</span>
+
+        <span class="c1"># (n, c, h, w) -&gt; (n, hidden_dim, n_h, n_w)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv_proj</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="c1"># (n, hidden_dim, n_h, n_w) -&gt; (n, hidden_dim, (n_h * n_w))</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">n</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">hidden_dim</span><span class="p">,</span> <span class="n">n_h</span> <span class="o">*</span> <span class="n">n_w</span><span class="p">)</span>
+
+        <span class="c1"># (n, hidden_dim, (n_h * n_w)) -&gt; (n, (n_h * n_w), hidden_dim)</span>
+        <span class="c1"># The self attention layer expects inputs in the format (N, S, E)</span>
+        <span class="c1"># where S is the source sequence length, N is the batch size, E is the</span>
+        <span class="c1"># embedding dimension</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="c1"># Reshape and permute the input tensor</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_process_input</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">n</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="c1"># Expand the class token to the full batch</span>
+        <span class="n">batch_class_token</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">class_token</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">n</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">batch_class_token</span><span class="p">,</span> <span class="n">x</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">encoder</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="c1"># Classifier &quot;token&quot; as used by standard language architectures</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span>
+
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">heads</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="k">def</span> <span class="nf">_vision_transformer</span><span class="p">(</span>
+    <span class="n">patch_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">num_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">num_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">hidden_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">mlp_dim</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">WeightsEnum</span><span class="p">],</span>
+    <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VisionTransformer</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">weights</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;num_classes&quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">]))</span>
+        <span class="k">assert</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">_ovewrite_named_param</span><span class="p">(</span><span class="n">kwargs</span><span class="p">,</span> <span class="s2">&quot;image_size&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;min_size&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+    <span class="n">image_size</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;image_size&quot;</span><span class="p">,</span> <span class="mi">224</span><span class="p">)</span>
+
+    <span class="n">model</span> <span class="o">=</span> <span class="n">VisionTransformer</span><span class="p">(</span>
+        <span class="n">image_size</span><span class="o">=</span><span class="n">image_size</span><span class="p">,</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="n">patch_size</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="o">=</span><span class="n">num_layers</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="n">num_heads</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="o">=</span><span class="n">hidden_dim</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="o">=</span><span class="n">mlp_dim</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">weights</span><span class="p">:</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">get_state_dict</span><span class="p">(</span><span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span> <span class="n">check_hash</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">return</span> <span class="n">model</span>
+
+
+<span class="n">_COMMON_META</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;categories&quot;</span><span class="p">:</span> <span class="n">_IMAGENET_CATEGORIES</span><span class="p">,</span>
+<span class="p">}</span>
+
+<span class="n">_COMMON_SWAG_META</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+    <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/facebookresearch/SWAG&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;license&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/facebookresearch/SWAG/blob/main/LICENSE&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="ViT_B_16_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_b_16.html#torchvision.models.ViT_B_16_Weights">[docs]</a><span class="k">class</span> <span class="nc">ViT_B_16_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_b_16-c867db91.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">86567656</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#vit_b_16&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.072</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">95.318</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">17.564</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">330.285</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using a modified version of `DeIT</span>
+<span class="s2">                &lt;https://arxiv.org/abs/2012.12877&gt;`_&#39;s training recipe.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_E2E_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_b_16_swag-9ac1b537.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">384</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">86859496</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">384</span><span class="p">,</span> <span class="mi">384</span><span class="p">),</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">85.304</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.650</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">55.484</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">331.398</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are learnt via transfer learning by end-to-end fine-tuning the original</span>
+<span class="s2">                `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ weights on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_LINEAR_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_b_16_lc_swag-4e70ced5.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5793&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">86567656</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">81.886</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">96.180</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">17.564</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">330.285</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are composed of the original frozen `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ trunk</span>
+<span class="s2">                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ViT_B_32_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_b_32.html#torchvision.models.ViT_B_32_Weights">[docs]</a><span class="k">class</span> <span class="nc">ViT_B_32_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_b_32-d86f8d99.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">88224232</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#vit_b_32&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">75.912</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">92.466</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">4.409</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">336.604</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using a modified version of `DeIT</span>
+<span class="s2">                &lt;https://arxiv.org/abs/2012.12877&gt;`_&#39;s training recipe.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ViT_L_16_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_l_16.html#torchvision.models.ViT_L_16_Weights">[docs]</a><span class="k">class</span> <span class="nc">ViT_L_16_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_l_16-852ce7e3.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span> <span class="n">resize_size</span><span class="o">=</span><span class="mi">242</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">304326632</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#vit_l_16&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">79.662</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">94.638</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">61.555</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">1161.023</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using a modified version of TorchVision&#39;s</span>
+<span class="s2">                `new training recipe</span>
+<span class="s2">                &lt;https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/&gt;`_.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_E2E_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_l_16_swag-4f3808c9.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">305174504</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">),</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">88.064</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">98.512</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">361.986</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">1164.258</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are learnt via transfer learning by end-to-end fine-tuning the original</span>
+<span class="s2">                `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ weights on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_LINEAR_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_l_16_lc_swag-4d563306.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5793&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">304326632</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">85.146</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.422</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">61.555</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">1161.023</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are composed of the original frozen `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ trunk</span>
+<span class="s2">                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ViT_L_32_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_l_32.html#torchvision.models.ViT_L_32_Weights">[docs]</a><span class="k">class</span> <span class="nc">ViT_L_32_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_l_32-c7638314.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span><span class="n">ImageClassification</span><span class="p">,</span> <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">306535400</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/tree/main/references/classification#vit_l_32&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">76.972</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">93.07</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">15.378</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">1169.449</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights were trained from scratch by using a modified version of `DeIT</span>
+<span class="s2">                &lt;https://arxiv.org/abs/2012.12877&gt;`_&#39;s training recipe.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_V1</span></div>
+
+
+<div class="viewcode-block" id="ViT_H_14_Weights"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_h_14.html#torchvision.models.ViT_H_14_Weights">[docs]</a><span class="k">class</span> <span class="nc">ViT_H_14_Weights</span><span class="p">(</span><span class="n">WeightsEnum</span><span class="p">):</span>
+    <span class="n">IMAGENET1K_SWAG_E2E_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_h_14_swag-80465313.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">518</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">518</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">633470440</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">518</span><span class="p">,</span> <span class="mi">518</span><span class="p">),</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">88.552</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">98.694</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">1016.717</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">2416.643</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are learnt via transfer learning by end-to-end fine-tuning the original</span>
+<span class="s2">                `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ weights on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">IMAGENET1K_SWAG_LINEAR_V1</span> <span class="o">=</span> <span class="n">Weights</span><span class="p">(</span>
+        <span class="n">url</span><span class="o">=</span><span class="s2">&quot;https://download.pytorch.org/models/vit_h_14_lc_swag-c1eb923e.pth&quot;</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="o">=</span><span class="n">partial</span><span class="p">(</span>
+            <span class="n">ImageClassification</span><span class="p">,</span>
+            <span class="n">crop_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span>
+            <span class="n">resize_size</span><span class="o">=</span><span class="mi">224</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">meta</span><span class="o">=</span><span class="p">{</span>
+            <span class="o">**</span><span class="n">_COMMON_SWAG_META</span><span class="p">,</span>
+            <span class="s2">&quot;recipe&quot;</span><span class="p">:</span> <span class="s2">&quot;https://github.com/pytorch/vision/pull/5793&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;num_params&quot;</span><span class="p">:</span> <span class="mi">632045800</span><span class="p">,</span>
+            <span class="s2">&quot;min_size&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+            <span class="s2">&quot;_metrics&quot;</span><span class="p">:</span> <span class="p">{</span>
+                <span class="s2">&quot;ImageNet-1K&quot;</span><span class="p">:</span> <span class="p">{</span>
+                    <span class="s2">&quot;acc@1&quot;</span><span class="p">:</span> <span class="mf">85.708</span><span class="p">,</span>
+                    <span class="s2">&quot;acc@5&quot;</span><span class="p">:</span> <span class="mf">97.730</span><span class="p">,</span>
+                <span class="p">}</span>
+            <span class="p">},</span>
+            <span class="s2">&quot;_ops&quot;</span><span class="p">:</span> <span class="mf">167.295</span><span class="p">,</span>
+            <span class="s2">&quot;_file_size&quot;</span><span class="p">:</span> <span class="mf">2411.209</span><span class="p">,</span>
+            <span class="s2">&quot;_docs&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">                These weights are composed of the original frozen `SWAG &lt;https://arxiv.org/abs/2201.08371&gt;`_ trunk</span>
+<span class="s2">                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</span>
+<span class="s2">            &quot;&quot;&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+    <span class="p">)</span>
+    <span class="n">DEFAULT</span> <span class="o">=</span> <span class="n">IMAGENET1K_SWAG_E2E_V1</span></div>
+
+
+<div class="viewcode-block" id="vit_b_16"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_b_16.html#torchvision.models.vit_b_16">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ViT_B_16_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vit_b_16</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ViT_B_16_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VisionTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a vit_b_16 architecture from</span>
+<span class="sd">    `An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale &lt;https://arxiv.org/abs/2010.11929&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ViT_B_16_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.ViT_B_16_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vision_transformer.VisionTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ViT_B_16_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ViT_B_16_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vision_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="o">=</span><span class="mi">12</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="mi">12</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="o">=</span><span class="mi">768</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="o">=</span><span class="mi">3072</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vit_b_32"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_b_32.html#torchvision.models.vit_b_32">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ViT_B_32_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vit_b_32</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ViT_B_32_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VisionTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a vit_b_32 architecture from</span>
+<span class="sd">    `An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale &lt;https://arxiv.org/abs/2010.11929&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ViT_B_32_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.ViT_B_32_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vision_transformer.VisionTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ViT_B_32_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ViT_B_32_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vision_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="o">=</span><span class="mi">12</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="mi">12</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="o">=</span><span class="mi">768</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="o">=</span><span class="mi">3072</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vit_l_16"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_l_16.html#torchvision.models.vit_l_16">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ViT_L_16_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vit_l_16</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ViT_L_16_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VisionTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a vit_l_16 architecture from</span>
+<span class="sd">    `An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale &lt;https://arxiv.org/abs/2010.11929&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ViT_L_16_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.ViT_L_16_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vision_transformer.VisionTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ViT_L_16_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ViT_L_16_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vision_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="o">=</span><span class="mi">24</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="o">=</span><span class="mi">4096</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vit_l_32"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_l_32.html#torchvision.models.vit_l_32">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="n">ViT_L_32_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vit_l_32</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ViT_L_32_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VisionTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a vit_l_32 architecture from</span>
+<span class="sd">    `An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale &lt;https://arxiv.org/abs/2010.11929&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ViT_L_32_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.ViT_L_32_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vision_transformer.VisionTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ViT_L_32_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ViT_L_32_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vision_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="o">=</span><span class="mi">24</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="o">=</span><span class="mi">4096</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vit_h_14"><a class="viewcode-back" href="../../../models/generated/torchvision.models.vit_h_14.html#torchvision.models.vit_h_14">[docs]</a><span class="nd">@register_model</span><span class="p">()</span>
+<span class="nd">@handle_legacy_interface</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;pretrained&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">))</span>
+<span class="k">def</span> <span class="nf">vit_h_14</span><span class="p">(</span><span class="o">*</span><span class="p">,</span> <span class="n">weights</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ViT_H_14_Weights</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">progress</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">VisionTransformer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Constructs a vit_h_14 architecture from</span>
+<span class="sd">    `An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale &lt;https://arxiv.org/abs/2010.11929&gt;`_.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        weights (:class:`~torchvision.models.ViT_H_14_Weights`, optional): The pretrained</span>
+<span class="sd">            weights to use. See :class:`~torchvision.models.ViT_H_14_Weights`</span>
+<span class="sd">            below for more details and possible values. By default, no pre-trained weights are used.</span>
+<span class="sd">        progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True.</span>
+<span class="sd">        **kwargs: parameters passed to the ``torchvision.models.vision_transformer.VisionTransformer``</span>
+<span class="sd">            base class. Please refer to the `source code</span>
+<span class="sd">            &lt;https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py&gt;`_</span>
+<span class="sd">            for more details about this class.</span>
+
+<span class="sd">    .. autoclass:: torchvision.models.ViT_H_14_Weights</span>
+<span class="sd">        :members:</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">weights</span> <span class="o">=</span> <span class="n">ViT_H_14_Weights</span><span class="o">.</span><span class="n">verify</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_vision_transformer</span><span class="p">(</span>
+        <span class="n">patch_size</span><span class="o">=</span><span class="mi">14</span><span class="p">,</span>
+        <span class="n">num_layers</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span>
+        <span class="n">num_heads</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">hidden_dim</span><span class="o">=</span><span class="mi">1280</span><span class="p">,</span>
+        <span class="n">mlp_dim</span><span class="o">=</span><span class="mi">5120</span><span class="p">,</span>
+        <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span>
+        <span class="n">progress</span><span class="o">=</span><span class="n">progress</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">interpolate_embeddings</span><span class="p">(</span>
+    <span class="n">image_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">patch_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">model_state</span><span class="p">:</span> <span class="s2">&quot;OrderedDict[str, torch.Tensor]&quot;</span><span class="p">,</span>
+    <span class="n">interpolation_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;bicubic&quot;</span><span class="p">,</span>
+    <span class="n">reset_heads</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="s2">&quot;OrderedDict[str, torch.Tensor]&quot;</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;This function helps interpolate positional embeddings during checkpoint loading,</span>
+<span class="sd">    especially when you want to apply a pre-trained model on images with different resolution.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        image_size (int): Image size of the new model.</span>
+<span class="sd">        patch_size (int): Patch size of the new model.</span>
+<span class="sd">        model_state (OrderedDict[str, torch.Tensor]): State dict of the pre-trained model.</span>
+<span class="sd">        interpolation_mode (str): The algorithm used for upsampling. Default: bicubic.</span>
+<span class="sd">        reset_heads (bool): If true, not copying the state of heads. Default: False.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        OrderedDict[str, torch.Tensor]: A state dict which can be loaded into the new model.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="c1"># Shape of pos_embedding is (1, seq_length, hidden_dim)</span>
+    <span class="n">pos_embedding</span> <span class="o">=</span> <span class="n">model_state</span><span class="p">[</span><span class="s2">&quot;encoder.pos_embedding&quot;</span><span class="p">]</span>
+    <span class="n">n</span><span class="p">,</span> <span class="n">seq_length</span><span class="p">,</span> <span class="n">hidden_dim</span> <span class="o">=</span> <span class="n">pos_embedding</span><span class="o">.</span><span class="n">shape</span>
+    <span class="k">if</span> <span class="n">n</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unexpected position embedding shape: </span><span class="si">{</span><span class="n">pos_embedding</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="n">new_seq_length</span> <span class="o">=</span> <span class="p">(</span><span class="n">image_size</span> <span class="o">//</span> <span class="n">patch_size</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span> <span class="o">+</span> <span class="mi">1</span>
+
+    <span class="c1"># Need to interpolate the weights for the position embedding.</span>
+    <span class="c1"># We do this by reshaping the positions embeddings to a 2d grid, performing</span>
+    <span class="c1"># an interpolation in the (h, w) space and then reshaping back to a 1d grid.</span>
+    <span class="k">if</span> <span class="n">new_seq_length</span> <span class="o">!=</span> <span class="n">seq_length</span><span class="p">:</span>
+        <span class="c1"># The class token embedding shouldn&#39;t be interpolated, so we split it up.</span>
+        <span class="n">seq_length</span> <span class="o">-=</span> <span class="mi">1</span>
+        <span class="n">new_seq_length</span> <span class="o">-=</span> <span class="mi">1</span>
+        <span class="n">pos_embedding_token</span> <span class="o">=</span> <span class="n">pos_embedding</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">1</span><span class="p">,</span> <span class="p">:]</span>
+        <span class="n">pos_embedding_img</span> <span class="o">=</span> <span class="n">pos_embedding</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">:,</span> <span class="p">:]</span>
+
+        <span class="c1"># (1, seq_length, hidden_dim) -&gt; (1, hidden_dim, seq_length)</span>
+        <span class="n">pos_embedding_img</span> <span class="o">=</span> <span class="n">pos_embedding_img</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">seq_length_1d</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">seq_length</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">seq_length_1d</span> <span class="o">*</span> <span class="n">seq_length_1d</span> <span class="o">!=</span> <span class="n">seq_length</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;seq_length is not a perfect square! Instead got seq_length_1d * seq_length_1d = </span><span class="si">{</span><span class="n">seq_length_1d</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="n">seq_length_1d</span><span class="w"> </span><span class="si">}</span><span class="s2"> and seq_length = </span><span class="si">{</span><span class="n">seq_length</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="c1"># (1, hidden_dim, seq_length) -&gt; (1, hidden_dim, seq_l_1d, seq_l_1d)</span>
+        <span class="n">pos_embedding_img</span> <span class="o">=</span> <span class="n">pos_embedding_img</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">hidden_dim</span><span class="p">,</span> <span class="n">seq_length_1d</span><span class="p">,</span> <span class="n">seq_length_1d</span><span class="p">)</span>
+        <span class="n">new_seq_length_1d</span> <span class="o">=</span> <span class="n">image_size</span> <span class="o">//</span> <span class="n">patch_size</span>
+
+        <span class="c1"># Perform interpolation.</span>
+        <span class="c1"># (1, hidden_dim, seq_l_1d, seq_l_1d) -&gt; (1, hidden_dim, new_seq_l_1d, new_seq_l_1d)</span>
+        <span class="n">new_pos_embedding_img</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">interpolate</span><span class="p">(</span>
+            <span class="n">pos_embedding_img</span><span class="p">,</span>
+            <span class="n">size</span><span class="o">=</span><span class="n">new_seq_length_1d</span><span class="p">,</span>
+            <span class="n">mode</span><span class="o">=</span><span class="n">interpolation_mode</span><span class="p">,</span>
+            <span class="n">align_corners</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="c1"># (1, hidden_dim, new_seq_l_1d, new_seq_l_1d) -&gt; (1, hidden_dim, new_seq_length)</span>
+        <span class="n">new_pos_embedding_img</span> <span class="o">=</span> <span class="n">new_pos_embedding_img</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">hidden_dim</span><span class="p">,</span> <span class="n">new_seq_length</span><span class="p">)</span>
+
+        <span class="c1"># (1, hidden_dim, new_seq_length) -&gt; (1, new_seq_length, hidden_dim)</span>
+        <span class="n">new_pos_embedding_img</span> <span class="o">=</span> <span class="n">new_pos_embedding_img</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">new_pos_embedding</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">pos_embedding_token</span><span class="p">,</span> <span class="n">new_pos_embedding_img</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">model_state</span><span class="p">[</span><span class="s2">&quot;encoder.pos_embedding&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">new_pos_embedding</span>
+
+        <span class="k">if</span> <span class="n">reset_heads</span><span class="p">:</span>
+            <span class="n">model_state_copy</span><span class="p">:</span> <span class="s2">&quot;OrderedDict[str, torch.Tensor]&quot;</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+            <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">model_state</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">k</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;heads&quot;</span><span class="p">):</span>
+                    <span class="n">model_state_copy</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span>
+            <span class="n">model_state</span> <span class="o">=</span> <span class="n">model_state_copy</span>
+
+    <span class="k">return</span> <span class="n">model_state</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/boxes.html b/0.16/_modules/torchvision/ops/boxes.html
new file mode 100644
index 00000000000..92d6567e277
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/boxes.html
@@ -0,0 +1,1086 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.boxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.boxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.boxes</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torchvision</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torchvision.extension</span> <span class="kn">import</span> <span class="n">_assert_has_ops</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._box_convert</span> <span class="kn">import</span> <span class="n">_box_cxcywh_to_xyxy</span><span class="p">,</span> <span class="n">_box_xywh_to_xyxy</span><span class="p">,</span> <span class="n">_box_xyxy_to_cxcywh</span><span class="p">,</span> <span class="n">_box_xyxy_to_xywh</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_upcast</span>
+
+
+<div class="viewcode-block" id="nms"><a class="viewcode-back" href="../../../generated/torchvision.ops.nms.html#torchvision.ops.nms">[docs]</a><span class="k">def</span> <span class="nf">nms</span><span class="p">(</span><span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">scores</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">iou_threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Performs non-maximum suppression (NMS) on the boxes according</span>
+<span class="sd">    to their intersection-over-union (IoU).</span>
+
+<span class="sd">    NMS iteratively removes lower scoring boxes which have an</span>
+<span class="sd">    IoU greater than iou_threshold with another (higher scoring)</span>
+<span class="sd">    box.</span>
+
+<span class="sd">    If multiple boxes have the exact same score and satisfy the IoU</span>
+<span class="sd">    criterion with respect to a reference box, the selected box is</span>
+<span class="sd">    not guaranteed to be the same between CPU and GPU. This is similar</span>
+<span class="sd">    to the behavior of argsort in PyTorch when repeated values are present.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes (Tensor[N, 4])): boxes to perform NMS on. They</span>
+<span class="sd">            are expected to be in ``(x1, y1, x2, y2)`` format with ``0 &lt;= x1 &lt; x2`` and</span>
+<span class="sd">            ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">        scores (Tensor[N]): scores for each one of the boxes</span>
+<span class="sd">        iou_threshold (float): discards all overlapping boxes with IoU &gt; iou_threshold</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: int64 tensor with the indices of the elements that have been kept</span>
+<span class="sd">        by NMS, sorted in decreasing order of scores</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">nms</span><span class="p">)</span>
+    <span class="n">_assert_has_ops</span><span class="p">()</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">nms</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">scores</span><span class="p">,</span> <span class="n">iou_threshold</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="batched_nms"><a class="viewcode-back" href="../../../generated/torchvision.ops.batched_nms.html#torchvision.ops.batched_nms">[docs]</a><span class="k">def</span> <span class="nf">batched_nms</span><span class="p">(</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">scores</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">idxs</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">iou_threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Performs non-maximum suppression in a batched fashion.</span>
+
+<span class="sd">    Each index value correspond to a category, and NMS</span>
+<span class="sd">    will not be applied between elements of different categories.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes (Tensor[N, 4]): boxes where NMS will be performed. They</span>
+<span class="sd">            are expected to be in ``(x1, y1, x2, y2)`` format with ``0 &lt;= x1 &lt; x2`` and</span>
+<span class="sd">            ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">        scores (Tensor[N]): scores for each one of the boxes</span>
+<span class="sd">        idxs (Tensor[N]): indices of the categories for each one of the boxes.</span>
+<span class="sd">        iou_threshold (float): discards all overlapping boxes with IoU &gt; iou_threshold</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: int64 tensor with the indices of the elements that have been kept by NMS, sorted</span>
+<span class="sd">        in decreasing order of scores</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">batched_nms</span><span class="p">)</span>
+    <span class="c1"># Benchmarks that drove the following thresholds are at</span>
+    <span class="c1"># https://github.com/pytorch/vision/issues/1311#issuecomment-781329339</span>
+    <span class="k">if</span> <span class="n">boxes</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">&gt;</span> <span class="p">(</span><span class="mi">4000</span> <span class="k">if</span> <span class="n">boxes</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span> <span class="k">else</span> <span class="mi">20000</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">_is_tracing</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">_batched_nms_vanilla</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">scores</span><span class="p">,</span> <span class="n">idxs</span><span class="p">,</span> <span class="n">iou_threshold</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_batched_nms_coordinate_trick</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">scores</span><span class="p">,</span> <span class="n">idxs</span><span class="p">,</span> <span class="n">iou_threshold</span><span class="p">)</span></div>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">_script_if_tracing</span>
+<span class="k">def</span> <span class="nf">_batched_nms_coordinate_trick</span><span class="p">(</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">scores</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">idxs</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">iou_threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># strategy: in order to perform NMS independently per class,</span>
+    <span class="c1"># we add an offset to all the boxes. The offset is dependent</span>
+    <span class="c1"># only on the class idx, and is large enough so that boxes</span>
+    <span class="c1"># from different classes do not overlap</span>
+    <span class="k">if</span> <span class="n">boxes</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">((</span><span class="mi">0</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">max_coordinate</span> <span class="o">=</span> <span class="n">boxes</span><span class="o">.</span><span class="n">max</span><span class="p">()</span>
+    <span class="n">offsets</span> <span class="o">=</span> <span class="n">idxs</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">max_coordinate</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">boxes</span><span class="p">))</span>
+    <span class="n">boxes_for_nms</span> <span class="o">=</span> <span class="n">boxes</span> <span class="o">+</span> <span class="n">offsets</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span>
+    <span class="n">keep</span> <span class="o">=</span> <span class="n">nms</span><span class="p">(</span><span class="n">boxes_for_nms</span><span class="p">,</span> <span class="n">scores</span><span class="p">,</span> <span class="n">iou_threshold</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">keep</span>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">_script_if_tracing</span>
+<span class="k">def</span> <span class="nf">_batched_nms_vanilla</span><span class="p">(</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">scores</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">idxs</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">iou_threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># Based on Detectron2 implementation, just manually call nms() on each class independently</span>
+    <span class="n">keep_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">scores</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">bool</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">class_id</span> <span class="ow">in</span> <span class="n">torch</span><span class="o">.</span><span class="n">unique</span><span class="p">(</span><span class="n">idxs</span><span class="p">):</span>
+        <span class="n">curr_indices</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">idxs</span> <span class="o">==</span> <span class="n">class_id</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">curr_keep_indices</span> <span class="o">=</span> <span class="n">nms</span><span class="p">(</span><span class="n">boxes</span><span class="p">[</span><span class="n">curr_indices</span><span class="p">],</span> <span class="n">scores</span><span class="p">[</span><span class="n">curr_indices</span><span class="p">],</span> <span class="n">iou_threshold</span><span class="p">)</span>
+        <span class="n">keep_mask</span><span class="p">[</span><span class="n">curr_indices</span><span class="p">[</span><span class="n">curr_keep_indices</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="n">keep_indices</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">keep_mask</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="k">return</span> <span class="n">keep_indices</span><span class="p">[</span><span class="n">scores</span><span class="p">[</span><span class="n">keep_indices</span><span class="p">]</span><span class="o">.</span><span class="n">sort</span><span class="p">(</span><span class="n">descending</span><span class="o">=</span><span class="kc">True</span><span class="p">)[</span><span class="mi">1</span><span class="p">]]</span>
+
+
+<div class="viewcode-block" id="remove_small_boxes"><a class="viewcode-back" href="../../../generated/torchvision.ops.remove_small_boxes.html#torchvision.ops.remove_small_boxes">[docs]</a><span class="k">def</span> <span class="nf">remove_small_boxes</span><span class="p">(</span><span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">min_size</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Remove boxes which contains at least one side smaller than min_size.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes (Tensor[N, 4]): boxes in ``(x1, y1, x2, y2)`` format</span>
+<span class="sd">            with ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">        min_size (float): minimum size</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[K]: indices of the boxes that have both sides</span>
+<span class="sd">        larger than min_size</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">remove_small_boxes</span><span class="p">)</span>
+    <span class="n">ws</span><span class="p">,</span> <span class="n">hs</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span>
+    <span class="n">keep</span> <span class="o">=</span> <span class="p">(</span><span class="n">ws</span> <span class="o">&gt;=</span> <span class="n">min_size</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">hs</span> <span class="o">&gt;=</span> <span class="n">min_size</span><span class="p">)</span>
+    <span class="n">keep</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">keep</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="k">return</span> <span class="n">keep</span></div>
+
+
+<div class="viewcode-block" id="clip_boxes_to_image"><a class="viewcode-back" href="../../../generated/torchvision.ops.clip_boxes_to_image.html#torchvision.ops.clip_boxes_to_image">[docs]</a><span class="k">def</span> <span class="nf">clip_boxes_to_image</span><span class="p">(</span><span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Clip boxes so that they lie inside an image of size `size`.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes (Tensor[N, 4]): boxes in ``(x1, y1, x2, y2)`` format</span>
+<span class="sd">            with ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">        size (Tuple[height, width]): size of the image</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, 4]: clipped boxes</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">clip_boxes_to_image</span><span class="p">)</span>
+    <span class="n">dim</span> <span class="o">=</span> <span class="n">boxes</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span>
+    <span class="n">boxes_x</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">::</span><span class="mi">2</span><span class="p">]</span>
+    <span class="n">boxes_y</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">::</span><span class="mi">2</span><span class="p">]</span>
+    <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">size</span>
+
+    <span class="k">if</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">_is_tracing</span><span class="p">():</span>
+        <span class="n">boxes_x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">boxes_x</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">device</span><span class="p">))</span>
+        <span class="n">boxes_x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">boxes_x</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">device</span><span class="p">))</span>
+        <span class="n">boxes_y</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">boxes_y</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">device</span><span class="p">))</span>
+        <span class="n">boxes_y</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">boxes_y</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">boxes</span><span class="o">.</span><span class="n">device</span><span class="p">))</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">boxes_x</span> <span class="o">=</span> <span class="n">boxes_x</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">width</span><span class="p">)</span>
+        <span class="n">boxes_y</span> <span class="o">=</span> <span class="n">boxes_y</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">height</span><span class="p">)</span>
+
+    <span class="n">clipped_boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">((</span><span class="n">boxes_x</span><span class="p">,</span> <span class="n">boxes_y</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="n">dim</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">clipped_boxes</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="box_convert"><a class="viewcode-back" href="../../../generated/torchvision.ops.box_convert.html#torchvision.ops.box_convert">[docs]</a><span class="k">def</span> <span class="nf">box_convert</span><span class="p">(</span><span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">in_fmt</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">out_fmt</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Converts boxes from given in_fmt to out_fmt.</span>
+<span class="sd">    Supported in_fmt and out_fmt are:</span>
+
+<span class="sd">    &#39;xyxy&#39;: boxes are represented via corners, x1, y1 being top left and x2, y2 being bottom right.</span>
+<span class="sd">    This is the format that torchvision utilities expect.</span>
+
+<span class="sd">    &#39;xywh&#39; : boxes are represented via corner, width and height, x1, y2 being top left, w, h being width and height.</span>
+
+<span class="sd">    &#39;cxcywh&#39; : boxes are represented via centre, width and height, cx, cy being center of box, w, h</span>
+<span class="sd">    being width and height.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes (Tensor[N, 4]): boxes which will be converted.</span>
+<span class="sd">        in_fmt (str): Input format of given boxes. Supported formats are [&#39;xyxy&#39;, &#39;xywh&#39;, &#39;cxcywh&#39;].</span>
+<span class="sd">        out_fmt (str): Output format of given boxes. Supported formats are [&#39;xyxy&#39;, &#39;xywh&#39;, &#39;cxcywh&#39;]</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, 4]: Boxes into converted format.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">box_convert</span><span class="p">)</span>
+    <span class="n">allowed_fmts</span> <span class="o">=</span> <span class="p">(</span><span class="s2">&quot;xyxy&quot;</span><span class="p">,</span> <span class="s2">&quot;xywh&quot;</span><span class="p">,</span> <span class="s2">&quot;cxcywh&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">in_fmt</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">allowed_fmts</span> <span class="ow">or</span> <span class="n">out_fmt</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">allowed_fmts</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Unsupported Bounding Box Conversions for given in_fmt and out_fmt&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">in_fmt</span> <span class="o">==</span> <span class="n">out_fmt</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">boxes</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+
+    <span class="k">if</span> <span class="n">in_fmt</span> <span class="o">!=</span> <span class="s2">&quot;xyxy&quot;</span> <span class="ow">and</span> <span class="n">out_fmt</span> <span class="o">!=</span> <span class="s2">&quot;xyxy&quot;</span><span class="p">:</span>
+        <span class="c1"># convert to xyxy and change in_fmt xyxy</span>
+        <span class="k">if</span> <span class="n">in_fmt</span> <span class="o">==</span> <span class="s2">&quot;xywh&quot;</span><span class="p">:</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="n">_box_xywh_to_xyxy</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">in_fmt</span> <span class="o">==</span> <span class="s2">&quot;cxcywh&quot;</span><span class="p">:</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="n">_box_cxcywh_to_xyxy</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+        <span class="n">in_fmt</span> <span class="o">=</span> <span class="s2">&quot;xyxy&quot;</span>
+
+    <span class="k">if</span> <span class="n">in_fmt</span> <span class="o">==</span> <span class="s2">&quot;xyxy&quot;</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">out_fmt</span> <span class="o">==</span> <span class="s2">&quot;xywh&quot;</span><span class="p">:</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="n">_box_xyxy_to_xywh</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">out_fmt</span> <span class="o">==</span> <span class="s2">&quot;cxcywh&quot;</span><span class="p">:</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="n">_box_xyxy_to_cxcywh</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">out_fmt</span> <span class="o">==</span> <span class="s2">&quot;xyxy&quot;</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">in_fmt</span> <span class="o">==</span> <span class="s2">&quot;xywh&quot;</span><span class="p">:</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="n">_box_xywh_to_xyxy</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">in_fmt</span> <span class="o">==</span> <span class="s2">&quot;cxcywh&quot;</span><span class="p">:</span>
+            <span class="n">boxes</span> <span class="o">=</span> <span class="n">_box_cxcywh_to_xyxy</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">boxes</span></div>
+
+
+<div class="viewcode-block" id="box_area"><a class="viewcode-back" href="../../../generated/torchvision.ops.box_area.html#torchvision.ops.box_area">[docs]</a><span class="k">def</span> <span class="nf">box_area</span><span class="p">(</span><span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Computes the area of a set of bounding boxes, which are specified by their</span>
+<span class="sd">    (x1, y1, x2, y2) coordinates.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes (Tensor[N, 4]): boxes for which the area will be computed. They</span>
+<span class="sd">            are expected to be in (x1, y1, x2, y2) format with</span>
+<span class="sd">            ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N]: the area for each box</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">box_area</span><span class="p">)</span>
+    <span class="n">boxes</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+    <span class="k">return</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">])</span> <span class="o">*</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">])</span></div>
+
+
+<span class="c1"># implementation from https://github.com/kuangliu/torchcv/blob/master/torchcv/utils/box.py</span>
+<span class="c1"># with slight modifications</span>
+<span class="k">def</span> <span class="nf">_box_inter_union</span><span class="p">(</span><span class="n">boxes1</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+    <span class="n">area1</span> <span class="o">=</span> <span class="n">box_area</span><span class="p">(</span><span class="n">boxes1</span><span class="p">)</span>
+    <span class="n">area2</span> <span class="o">=</span> <span class="n">box_area</span><span class="p">(</span><span class="n">boxes2</span><span class="p">)</span>
+
+    <span class="n">lt</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">],</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">])</span>  <span class="c1"># [N,M,2]</span>
+    <span class="n">rb</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">2</span><span class="p">:],</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:])</span>  <span class="c1"># [N,M,2]</span>
+
+    <span class="n">wh</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">rb</span> <span class="o">-</span> <span class="n">lt</span><span class="p">)</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># [N,M,2]</span>
+    <span class="n">inter</span> <span class="o">=</span> <span class="n">wh</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">wh</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span>  <span class="c1"># [N,M]</span>
+
+    <span class="n">union</span> <span class="o">=</span> <span class="n">area1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">+</span> <span class="n">area2</span> <span class="o">-</span> <span class="n">inter</span>
+
+    <span class="k">return</span> <span class="n">inter</span><span class="p">,</span> <span class="n">union</span>
+
+
+<div class="viewcode-block" id="box_iou"><a class="viewcode-back" href="../../../generated/torchvision.ops.box_iou.html#torchvision.ops.box_iou">[docs]</a><span class="k">def</span> <span class="nf">box_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Return intersection-over-union (Jaccard index) between two sets of boxes.</span>
+
+<span class="sd">    Both sets of boxes are expected to be in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes1 (Tensor[N, 4]): first set of boxes</span>
+<span class="sd">        boxes2 (Tensor[M, 4]): second set of boxes</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, M]: the NxM matrix containing the pairwise IoU values for every element in boxes1 and boxes2</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">box_iou</span><span class="p">)</span>
+    <span class="n">inter</span><span class="p">,</span> <span class="n">union</span> <span class="o">=</span> <span class="n">_box_inter_union</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">)</span>
+    <span class="n">iou</span> <span class="o">=</span> <span class="n">inter</span> <span class="o">/</span> <span class="n">union</span>
+    <span class="k">return</span> <span class="n">iou</span></div>
+
+
+<span class="c1"># Implementation adapted from https://github.com/facebookresearch/detr/blob/master/util/box_ops.py</span>
+<div class="viewcode-block" id="generalized_box_iou"><a class="viewcode-back" href="../../../generated/torchvision.ops.generalized_box_iou.html#torchvision.ops.generalized_box_iou">[docs]</a><span class="k">def</span> <span class="nf">generalized_box_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Return generalized intersection-over-union (Jaccard index) between two sets of boxes.</span>
+
+<span class="sd">    Both sets of boxes are expected to be in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes1 (Tensor[N, 4]): first set of boxes</span>
+<span class="sd">        boxes2 (Tensor[M, 4]): second set of boxes</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, M]: the NxM matrix containing the pairwise generalized IoU values</span>
+<span class="sd">        for every element in boxes1 and boxes2</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">generalized_box_iou</span><span class="p">)</span>
+
+    <span class="n">inter</span><span class="p">,</span> <span class="n">union</span> <span class="o">=</span> <span class="n">_box_inter_union</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">)</span>
+    <span class="n">iou</span> <span class="o">=</span> <span class="n">inter</span> <span class="o">/</span> <span class="n">union</span>
+
+    <span class="n">lti</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">],</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">])</span>
+    <span class="n">rbi</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">2</span><span class="p">:],</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:])</span>
+
+    <span class="n">whi</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">rbi</span> <span class="o">-</span> <span class="n">lti</span><span class="p">)</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># [N,M,2]</span>
+    <span class="n">areai</span> <span class="o">=</span> <span class="n">whi</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">whi</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span>
+
+    <span class="k">return</span> <span class="n">iou</span> <span class="o">-</span> <span class="p">(</span><span class="n">areai</span> <span class="o">-</span> <span class="n">union</span><span class="p">)</span> <span class="o">/</span> <span class="n">areai</span></div>
+
+
+<div class="viewcode-block" id="complete_box_iou"><a class="viewcode-back" href="../../../generated/torchvision.ops.complete_box_iou.html#torchvision.ops.complete_box_iou">[docs]</a><span class="k">def</span> <span class="nf">complete_box_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-7</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Return complete intersection-over-union (Jaccard index) between two sets of boxes.</span>
+<span class="sd">    Both sets of boxes are expected to be in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        boxes1 (Tensor[N, 4]): first set of boxes</span>
+<span class="sd">        boxes2 (Tensor[M, 4]): second set of boxes</span>
+<span class="sd">        eps (float, optional): small number to prevent division by zero. Default: 1e-7</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, M]: the NxM matrix containing the pairwise complete IoU values</span>
+<span class="sd">        for every element in boxes1 and boxes2</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">complete_box_iou</span><span class="p">)</span>
+
+    <span class="n">boxes1</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">boxes1</span><span class="p">)</span>
+    <span class="n">boxes2</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">boxes2</span><span class="p">)</span>
+
+    <span class="n">diou</span><span class="p">,</span> <span class="n">iou</span> <span class="o">=</span> <span class="n">_box_diou_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">,</span> <span class="n">eps</span><span class="p">)</span>
+
+    <span class="n">w_pred</span> <span class="o">=</span> <span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span>
+    <span class="n">h_pred</span> <span class="o">=</span> <span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span>
+
+    <span class="n">w_gt</span> <span class="o">=</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span>
+    <span class="n">h_gt</span> <span class="o">=</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span>
+
+    <span class="n">v</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span> <span class="o">/</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">pi</span><span class="o">**</span><span class="mi">2</span><span class="p">))</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">pow</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">w_pred</span> <span class="o">/</span> <span class="n">h_pred</span><span class="p">)</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">w_gt</span> <span class="o">/</span> <span class="n">h_gt</span><span class="p">),</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+        <span class="n">alpha</span> <span class="o">=</span> <span class="n">v</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">iou</span> <span class="o">+</span> <span class="n">v</span> <span class="o">+</span> <span class="n">eps</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">diou</span> <span class="o">-</span> <span class="n">alpha</span> <span class="o">*</span> <span class="n">v</span></div>
+
+
+<div class="viewcode-block" id="distance_box_iou"><a class="viewcode-back" href="../../../generated/torchvision.ops.distance_box_iou.html#torchvision.ops.distance_box_iou">[docs]</a><span class="k">def</span> <span class="nf">distance_box_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-7</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Return distance intersection-over-union (Jaccard index) between two sets of boxes.</span>
+
+<span class="sd">    Both sets of boxes are expected to be in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes1 (Tensor[N, 4]): first set of boxes</span>
+<span class="sd">        boxes2 (Tensor[M, 4]): second set of boxes</span>
+<span class="sd">        eps (float, optional): small number to prevent division by zero. Default: 1e-7</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, M]: the NxM matrix containing the pairwise distance IoU values</span>
+<span class="sd">        for every element in boxes1 and boxes2</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">distance_box_iou</span><span class="p">)</span>
+
+    <span class="n">boxes1</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">boxes1</span><span class="p">)</span>
+    <span class="n">boxes2</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">boxes2</span><span class="p">)</span>
+    <span class="n">diou</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_box_diou_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">diou</span></div>
+
+
+<span class="k">def</span> <span class="nf">_box_diou_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-7</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+
+    <span class="n">iou</span> <span class="o">=</span> <span class="n">box_iou</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">)</span>
+    <span class="n">lti</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">],</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">])</span>
+    <span class="n">rbi</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="mi">2</span><span class="p">:],</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:])</span>
+    <span class="n">whi</span> <span class="o">=</span> <span class="n">_upcast</span><span class="p">(</span><span class="n">rbi</span> <span class="o">-</span> <span class="n">lti</span><span class="p">)</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># [N,M,2]</span>
+    <span class="n">diagonal_distance_squared</span> <span class="o">=</span> <span class="p">(</span><span class="n">whi</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span><span class="n">whi</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span> <span class="o">+</span> <span class="n">eps</span>
+    <span class="c1"># centers of boxes</span>
+    <span class="n">x_p</span> <span class="o">=</span> <span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">boxes1</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">])</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="n">y_p</span> <span class="o">=</span> <span class="p">(</span><span class="n">boxes1</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">boxes1</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">])</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="n">x_g</span> <span class="o">=</span> <span class="p">(</span><span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">])</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="n">y_g</span> <span class="o">=</span> <span class="p">(</span><span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">boxes2</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">])</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="c1"># The distance between boxes&#39; centers squared.</span>
+    <span class="n">centers_distance_squared</span> <span class="o">=</span> <span class="p">(</span><span class="n">_upcast</span><span class="p">((</span><span class="n">x_p</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">-</span> <span class="n">x_g</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]))</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span> <span class="o">+</span> <span class="p">(</span>
+        <span class="n">_upcast</span><span class="p">((</span><span class="n">y_p</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">-</span> <span class="n">y_g</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]))</span> <span class="o">**</span> <span class="mi">2</span>
+    <span class="p">)</span>
+    <span class="c1"># The distance IoU is the IoU penalized by a normalized</span>
+    <span class="c1"># distance between boxes&#39; centers squared.</span>
+    <span class="k">return</span> <span class="n">iou</span> <span class="o">-</span> <span class="p">(</span><span class="n">centers_distance_squared</span> <span class="o">/</span> <span class="n">diagonal_distance_squared</span><span class="p">),</span> <span class="n">iou</span>
+
+
+<div class="viewcode-block" id="masks_to_boxes"><a class="viewcode-back" href="../../../generated/torchvision.ops.masks_to_boxes.html#torchvision.ops.masks_to_boxes">[docs]</a><span class="k">def</span> <span class="nf">masks_to_boxes</span><span class="p">(</span><span class="n">masks</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Compute the bounding boxes around the provided masks.</span>
+
+<span class="sd">    Returns a [N, 4] tensor containing bounding boxes. The boxes are in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        masks (Tensor[N, H, W]): masks to transform where N is the number of masks</span>
+<span class="sd">            and (H, W) are the spatial dimensions.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, 4]: bounding boxes</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">masks_to_boxes</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">masks</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="mi">0</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="n">device</span><span class="o">=</span><span class="n">masks</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">)</span>
+
+    <span class="n">n</span> <span class="o">=</span> <span class="n">masks</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">n</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="n">device</span><span class="o">=</span><span class="n">masks</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">)</span>
+
+    <span class="k">for</span> <span class="n">index</span><span class="p">,</span> <span class="n">mask</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">masks</span><span class="p">):</span>
+        <span class="n">y</span><span class="p">,</span> <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">mask</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="n">bounding_boxes</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">bounding_boxes</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+        <span class="n">bounding_boxes</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">bounding_boxes</span><span class="p">[</span><span class="n">index</span><span class="p">,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">bounding_boxes</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/ciou_loss.html b/0.16/_modules/torchvision/ops/ciou_loss.html
new file mode 100644
index 00000000000..f9f1ecb1145
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/ciou_loss.html
@@ -0,0 +1,747 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.ciou_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.ciou_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.ciou_loss</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_upcast_non_float</span>
+<span class="kn">from</span> <span class="nn">.diou_loss</span> <span class="kn">import</span> <span class="n">_diou_iou_loss</span>
+
+
+<div class="viewcode-block" id="complete_box_iou_loss"><a class="viewcode-back" href="../../../generated/torchvision.ops.complete_box_iou_loss.html#torchvision.ops.complete_box_iou_loss">[docs]</a><span class="k">def</span> <span class="nf">complete_box_iou_loss</span><span class="p">(</span>
+    <span class="n">boxes1</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes2</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">reduction</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;none&quot;</span><span class="p">,</span>
+    <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-7</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Gradient-friendly IoU loss with an additional penalty that is non-zero when the</span>
+<span class="sd">    boxes do not overlap. This loss function considers important geometrical</span>
+<span class="sd">    factors such as overlap area, normalized central point distance and aspect ratio.</span>
+<span class="sd">    This loss is symmetric, so the boxes1 and boxes2 arguments are interchangeable.</span>
+
+<span class="sd">    Both sets of boxes are expected to be in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``, and The two boxes should have the</span>
+<span class="sd">    same dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes1 : (Tensor[N, 4] or Tensor[4]) first set of boxes</span>
+<span class="sd">        boxes2 : (Tensor[N, 4] or Tensor[4]) second set of boxes</span>
+<span class="sd">        reduction : (string, optional) Specifies the reduction to apply to the output:</span>
+<span class="sd">            ``&#39;none&#39;`` | ``&#39;mean&#39;`` | ``&#39;sum&#39;``. ``&#39;none&#39;``: No reduction will be</span>
+<span class="sd">            applied to the output. ``&#39;mean&#39;``: The output will be averaged.</span>
+<span class="sd">            ``&#39;sum&#39;``: The output will be summed. Default: ``&#39;none&#39;``</span>
+<span class="sd">        eps : (float): small number to prevent division by zero. Default: 1e-7</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: Loss tensor with the reduction option applied.</span>
+
+<span class="sd">    Reference:</span>
+<span class="sd">        Zhaohui Zheng et al.: Complete Intersection over Union Loss:</span>
+<span class="sd">        https://arxiv.org/abs/1911.08287</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="c1"># Original Implementation from https://github.com/facebookresearch/detectron2/blob/main/detectron2/layers/losses.py</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">complete_box_iou_loss</span><span class="p">)</span>
+
+    <span class="n">boxes1</span> <span class="o">=</span> <span class="n">_upcast_non_float</span><span class="p">(</span><span class="n">boxes1</span><span class="p">)</span>
+    <span class="n">boxes2</span> <span class="o">=</span> <span class="n">_upcast_non_float</span><span class="p">(</span><span class="n">boxes2</span><span class="p">)</span>
+
+    <span class="n">diou_loss</span><span class="p">,</span> <span class="n">iou</span> <span class="o">=</span> <span class="n">_diou_iou_loss</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">)</span>
+
+    <span class="n">x1</span><span class="p">,</span> <span class="n">y1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">y2</span> <span class="o">=</span> <span class="n">boxes1</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">x1g</span><span class="p">,</span> <span class="n">y1g</span><span class="p">,</span> <span class="n">x2g</span><span class="p">,</span> <span class="n">y2g</span> <span class="o">=</span> <span class="n">boxes2</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="c1"># width and height of boxes</span>
+    <span class="n">w_pred</span> <span class="o">=</span> <span class="n">x2</span> <span class="o">-</span> <span class="n">x1</span>
+    <span class="n">h_pred</span> <span class="o">=</span> <span class="n">y2</span> <span class="o">-</span> <span class="n">y1</span>
+    <span class="n">w_gt</span> <span class="o">=</span> <span class="n">x2g</span> <span class="o">-</span> <span class="n">x1g</span>
+    <span class="n">h_gt</span> <span class="o">=</span> <span class="n">y2g</span> <span class="o">-</span> <span class="n">y1g</span>
+    <span class="n">v</span> <span class="o">=</span> <span class="p">(</span><span class="mi">4</span> <span class="o">/</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">pi</span><span class="o">**</span><span class="mi">2</span><span class="p">))</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">pow</span><span class="p">((</span><span class="n">torch</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">w_gt</span> <span class="o">/</span> <span class="n">h_gt</span><span class="p">)</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">w_pred</span> <span class="o">/</span> <span class="n">h_pred</span><span class="p">)),</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+        <span class="n">alpha</span> <span class="o">=</span> <span class="n">v</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">iou</span> <span class="o">+</span> <span class="n">v</span> <span class="o">+</span> <span class="n">eps</span><span class="p">)</span>
+
+    <span class="n">loss</span> <span class="o">=</span> <span class="n">diou_loss</span> <span class="o">+</span> <span class="n">alpha</span> <span class="o">*</span> <span class="n">v</span>
+
+    <span class="c1"># Check reduction option and return loss accordingly</span>
+    <span class="k">if</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;none&quot;</span><span class="p">:</span>
+        <span class="k">pass</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;mean&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span> <span class="k">if</span> <span class="n">loss</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="mf">0.0</span> <span class="o">*</span> <span class="n">loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;sum&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Invalid Value for arg &#39;reduction&#39;: &#39;</span><span class="si">{</span><span class="n">reduction</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2"> Supported reduction modes: &#39;none&#39;, &#39;mean&#39;, &#39;sum&#39;&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">loss</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/deform_conv.html b/0.16/_modules/torchvision/ops/deform_conv.html
new file mode 100644
index 00000000000..c2949659bc9
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/deform_conv.html
@@ -0,0 +1,864 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.deform_conv &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.deform_conv</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.deform_conv</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.nn</span> <span class="kn">import</span> <span class="n">init</span>
+<span class="kn">from</span> <span class="nn">torch.nn.modules.utils</span> <span class="kn">import</span> <span class="n">_pair</span>
+<span class="kn">from</span> <span class="nn">torch.nn.parameter</span> <span class="kn">import</span> <span class="n">Parameter</span>
+<span class="kn">from</span> <span class="nn">torchvision.extension</span> <span class="kn">import</span> <span class="n">_assert_has_ops</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+
+<div class="viewcode-block" id="deform_conv2d"><a class="viewcode-back" href="../../../generated/torchvision.ops.deform_conv2d.html#torchvision.ops.deform_conv2d">[docs]</a><span class="k">def</span> <span class="nf">deform_conv2d</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">offset</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">weight</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">stride</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span>
+    <span class="n">dilation</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Performs Deformable Convolution v2, described in</span>
+<span class="sd">    `Deformable ConvNets v2: More Deformable, Better Results</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1811.11168&gt;`__ if :attr:`mask` is not ``None`` and</span>
+<span class="sd">    Performs Deformable Convolution, described in</span>
+<span class="sd">    `Deformable Convolutional Networks</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1703.06211&gt;`__ if :attr:`mask` is ``None``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[batch_size, in_channels, in_height, in_width]): input tensor</span>
+<span class="sd">        offset (Tensor[batch_size, 2 * offset_groups * kernel_height * kernel_width, out_height, out_width]):</span>
+<span class="sd">            offsets to be applied for each position in the convolution kernel.</span>
+<span class="sd">        weight (Tensor[out_channels, in_channels // groups, kernel_height, kernel_width]): convolution weights,</span>
+<span class="sd">            split into groups of size (in_channels // groups)</span>
+<span class="sd">        bias (Tensor[out_channels]): optional bias of shape (out_channels,). Default: None</span>
+<span class="sd">        stride (int or Tuple[int, int]): distance between convolution centers. Default: 1</span>
+<span class="sd">        padding (int or Tuple[int, int]): height/width of padding of zeroes around</span>
+<span class="sd">            each image. Default: 0</span>
+<span class="sd">        dilation (int or Tuple[int, int]): the spacing between kernel elements. Default: 1</span>
+<span class="sd">        mask (Tensor[batch_size, offset_groups * kernel_height * kernel_width, out_height, out_width]):</span>
+<span class="sd">            masks to be applied for each position in the convolution kernel. Default: None</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[batch_sz, out_channels, out_h, out_w]: result of convolution</span>
+
+<span class="sd">    Examples::</span>
+<span class="sd">        &gt;&gt;&gt; input = torch.rand(4, 3, 10, 10)</span>
+<span class="sd">        &gt;&gt;&gt; kh, kw = 3, 3</span>
+<span class="sd">        &gt;&gt;&gt; weight = torch.rand(5, 3, kh, kw)</span>
+<span class="sd">        &gt;&gt;&gt; # offset and mask should have the same spatial size as the output</span>
+<span class="sd">        &gt;&gt;&gt; # of the convolution. In this case, for an input of 10, stride of 1</span>
+<span class="sd">        &gt;&gt;&gt; # and kernel size of 3, without padding, the output size is 8</span>
+<span class="sd">        &gt;&gt;&gt; offset = torch.rand(4, 2 * kh * kw, 8, 8)</span>
+<span class="sd">        &gt;&gt;&gt; mask = torch.rand(4, kh * kw, 8, 8)</span>
+<span class="sd">        &gt;&gt;&gt; out = deform_conv2d(input, offset, weight, mask=mask)</span>
+<span class="sd">        &gt;&gt;&gt; print(out.shape)</span>
+<span class="sd">        &gt;&gt;&gt; # returns</span>
+<span class="sd">        &gt;&gt;&gt;  torch.Size([4, 5, 8, 8])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">deform_conv2d</span><span class="p">)</span>
+    <span class="n">_assert_has_ops</span><span class="p">()</span>
+    <span class="n">out_channels</span> <span class="o">=</span> <span class="n">weight</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="n">use_mask</span> <span class="o">=</span> <span class="n">mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+
+    <span class="k">if</span> <span class="n">mask</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="nb">input</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mi">1</span><span class="p">),</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">bias</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">bias</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="n">stride_h</span><span class="p">,</span> <span class="n">stride_w</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">stride</span><span class="p">)</span>
+    <span class="n">pad_h</span><span class="p">,</span> <span class="n">pad_w</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+    <span class="n">dil_h</span><span class="p">,</span> <span class="n">dil_w</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">dilation</span><span class="p">)</span>
+    <span class="n">weights_h</span><span class="p">,</span> <span class="n">weights_w</span> <span class="o">=</span> <span class="n">weight</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+    <span class="n">_</span><span class="p">,</span> <span class="n">n_in_channels</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">shape</span>
+
+    <span class="n">n_offset_grps</span> <span class="o">=</span> <span class="n">offset</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="n">weights_h</span> <span class="o">*</span> <span class="n">weights_w</span><span class="p">)</span>
+    <span class="n">n_weight_grps</span> <span class="o">=</span> <span class="n">n_in_channels</span> <span class="o">//</span> <span class="n">weight</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="n">n_offset_grps</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+            <span class="s2">&quot;the shape of the offset tensor at dimension 1 is not valid. It should &quot;</span>
+            <span class="s2">&quot;be a multiple of 2 * weight.size[2] * weight.size[3].</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;Got offset.shape[1]=</span><span class="si">{</span><span class="n">offset</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2">, while 2 * weight.size[2] * weight.size[3]=</span><span class="si">{</span><span class="mi">2</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="n">weights_h</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="n">weights_w</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">deform_conv2d</span><span class="p">(</span>
+        <span class="nb">input</span><span class="p">,</span>
+        <span class="n">weight</span><span class="p">,</span>
+        <span class="n">offset</span><span class="p">,</span>
+        <span class="n">mask</span><span class="p">,</span>
+        <span class="n">bias</span><span class="p">,</span>
+        <span class="n">stride_h</span><span class="p">,</span>
+        <span class="n">stride_w</span><span class="p">,</span>
+        <span class="n">pad_h</span><span class="p">,</span>
+        <span class="n">pad_w</span><span class="p">,</span>
+        <span class="n">dil_h</span><span class="p">,</span>
+        <span class="n">dil_w</span><span class="p">,</span>
+        <span class="n">n_weight_grps</span><span class="p">,</span>
+        <span class="n">n_offset_grps</span><span class="p">,</span>
+        <span class="n">use_mask</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="DeformConv2d"><a class="viewcode-back" href="../../../generated/torchvision.ops.DeformConv2d.html#torchvision.ops.DeformConv2d">[docs]</a><span class="k">class</span> <span class="nc">DeformConv2d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`deform_conv2d`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">in_channels</span> <span class="o">%</span> <span class="n">groups</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;in_channels must be divisible by groups&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">out_channels</span> <span class="o">%</span> <span class="n">groups</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;out_channels must be divisible by groups&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">in_channels</span> <span class="o">=</span> <span class="n">in_channels</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">out_channels</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">stride</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dilation</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">dilation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">groups</span> <span class="o">=</span> <span class="n">groups</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">weight</span> <span class="o">=</span> <span class="n">Parameter</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">in_channels</span> <span class="o">//</span> <span class="n">groups</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">bias</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">bias</span> <span class="o">=</span> <span class="n">Parameter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="n">out_channels</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">register_parameter</span><span class="p">(</span><span class="s2">&quot;bias&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">reset_parameters</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">reset_parameters</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">init</span><span class="o">.</span><span class="n">kaiming_uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">a</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mi">5</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">fan_in</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">init</span><span class="o">.</span><span class="n">_calculate_fan_in_and_fan_out</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">)</span>
+            <span class="n">bound</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">fan_in</span><span class="p">)</span>
+            <span class="n">init</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="o">-</span><span class="n">bound</span><span class="p">,</span> <span class="n">bound</span><span class="p">)</span>
+
+<div class="viewcode-block" id="DeformConv2d.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.DeformConv2d.html#torchvision.ops.DeformConv2d.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">offset</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            input (Tensor[batch_size, in_channels, in_height, in_width]): input tensor</span>
+<span class="sd">            offset (Tensor[batch_size, 2 * offset_groups * kernel_height * kernel_width, out_height, out_width]):</span>
+<span class="sd">                offsets to be applied for each position in the convolution kernel.</span>
+<span class="sd">            mask (Tensor[batch_size, offset_groups * kernel_height * kernel_width, out_height, out_width]):</span>
+<span class="sd">                masks to be applied for each position in the convolution kernel.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">deform_conv2d</span><span class="p">(</span>
+            <span class="nb">input</span><span class="p">,</span>
+            <span class="n">offset</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
+            <span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">,</span>
+            <span class="n">dilation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dilation</span><span class="p">,</span>
+            <span class="n">mask</span><span class="o">=</span><span class="n">mask</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">in_channels</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, kernel_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, stride=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, padding=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="o">!=</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, dilation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">dilation</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">dilation</span> <span class="o">!=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, groups=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">groups</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">groups</span> <span class="o">!=</span> <span class="mi">1</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="s2">&quot;, bias=False&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="s2">&quot;)&quot;</span>
+
+        <span class="k">return</span> <span class="n">s</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/diou_loss.html b/0.16/_modules/torchvision/ops/diou_loss.html
new file mode 100644
index 00000000000..b970a9c1b74
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/diou_loss.html
@@ -0,0 +1,763 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.diou_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.diou_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.diou_loss</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_loss_inter_union</span><span class="p">,</span> <span class="n">_upcast_non_float</span>
+
+
+<div class="viewcode-block" id="distance_box_iou_loss"><a class="viewcode-back" href="../../../generated/torchvision.ops.distance_box_iou_loss.html#torchvision.ops.distance_box_iou_loss">[docs]</a><span class="k">def</span> <span class="nf">distance_box_iou_loss</span><span class="p">(</span>
+    <span class="n">boxes1</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes2</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">reduction</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;none&quot;</span><span class="p">,</span>
+    <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-7</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Gradient-friendly IoU loss with an additional penalty that is non-zero when the</span>
+<span class="sd">    distance between boxes&#39; centers isn&#39;t zero. Indeed, for two exactly overlapping</span>
+<span class="sd">    boxes, the distance IoU is the same as the IoU loss.</span>
+<span class="sd">    This loss is symmetric, so the boxes1 and boxes2 arguments are interchangeable.</span>
+
+<span class="sd">    Both sets of boxes are expected to be in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``, and The two boxes should have the</span>
+<span class="sd">    same dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes1 (Tensor[N, 4]): first set of boxes</span>
+<span class="sd">        boxes2 (Tensor[N, 4]): second set of boxes</span>
+<span class="sd">        reduction (string, optional): Specifies the reduction to apply to the output:</span>
+<span class="sd">            ``&#39;none&#39;`` | ``&#39;mean&#39;`` | ``&#39;sum&#39;``. ``&#39;none&#39;``: No reduction will be</span>
+<span class="sd">            applied to the output. ``&#39;mean&#39;``: The output will be averaged.</span>
+<span class="sd">            ``&#39;sum&#39;``: The output will be summed. Default: ``&#39;none&#39;``</span>
+<span class="sd">        eps (float, optional): small number to prevent division by zero. Default: 1e-7</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: Loss tensor with the reduction option applied.</span>
+
+<span class="sd">    Reference:</span>
+<span class="sd">        Zhaohui Zheng et al.: Distance Intersection over Union Loss:</span>
+<span class="sd">        https://arxiv.org/abs/1911.08287</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="c1"># Original Implementation from https://github.com/facebookresearch/detectron2/blob/main/detectron2/layers/losses.py</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">distance_box_iou_loss</span><span class="p">)</span>
+
+    <span class="n">boxes1</span> <span class="o">=</span> <span class="n">_upcast_non_float</span><span class="p">(</span><span class="n">boxes1</span><span class="p">)</span>
+    <span class="n">boxes2</span> <span class="o">=</span> <span class="n">_upcast_non_float</span><span class="p">(</span><span class="n">boxes2</span><span class="p">)</span>
+
+    <span class="n">loss</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_diou_iou_loss</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">,</span> <span class="n">eps</span><span class="p">)</span>
+
+    <span class="c1"># Check reduction option and return loss accordingly</span>
+    <span class="k">if</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;none&quot;</span><span class="p">:</span>
+        <span class="k">pass</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;mean&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span> <span class="k">if</span> <span class="n">loss</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="mf">0.0</span> <span class="o">*</span> <span class="n">loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;sum&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Invalid Value for arg &#39;reduction&#39;: &#39;</span><span class="si">{</span><span class="n">reduction</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2"> Supported reduction modes: &#39;none&#39;, &#39;mean&#39;, &#39;sum&#39;&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">loss</span></div>
+
+
+<span class="k">def</span> <span class="nf">_diou_iou_loss</span><span class="p">(</span>
+    <span class="n">boxes1</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes2</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-7</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]:</span>
+
+    <span class="n">intsct</span><span class="p">,</span> <span class="n">union</span> <span class="o">=</span> <span class="n">_loss_inter_union</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">)</span>
+    <span class="n">iou</span> <span class="o">=</span> <span class="n">intsct</span> <span class="o">/</span> <span class="p">(</span><span class="n">union</span> <span class="o">+</span> <span class="n">eps</span><span class="p">)</span>
+    <span class="c1"># smallest enclosing box</span>
+    <span class="n">x1</span><span class="p">,</span> <span class="n">y1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">y2</span> <span class="o">=</span> <span class="n">boxes1</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">x1g</span><span class="p">,</span> <span class="n">y1g</span><span class="p">,</span> <span class="n">x2g</span><span class="p">,</span> <span class="n">y2g</span> <span class="o">=</span> <span class="n">boxes2</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">xc1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">x1</span><span class="p">,</span> <span class="n">x1g</span><span class="p">)</span>
+    <span class="n">yc1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">y1</span><span class="p">,</span> <span class="n">y1g</span><span class="p">)</span>
+    <span class="n">xc2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">x2</span><span class="p">,</span> <span class="n">x2g</span><span class="p">)</span>
+    <span class="n">yc2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">y2</span><span class="p">,</span> <span class="n">y2g</span><span class="p">)</span>
+    <span class="c1"># The diagonal distance of the smallest enclosing box squared</span>
+    <span class="n">diagonal_distance_squared</span> <span class="o">=</span> <span class="p">((</span><span class="n">xc2</span> <span class="o">-</span> <span class="n">xc1</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span> <span class="o">+</span> <span class="p">((</span><span class="n">yc2</span> <span class="o">-</span> <span class="n">yc1</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span> <span class="o">+</span> <span class="n">eps</span>
+    <span class="c1"># centers of boxes</span>
+    <span class="n">x_p</span> <span class="o">=</span> <span class="p">(</span><span class="n">x2</span> <span class="o">+</span> <span class="n">x1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="n">y_p</span> <span class="o">=</span> <span class="p">(</span><span class="n">y2</span> <span class="o">+</span> <span class="n">y1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="n">x_g</span> <span class="o">=</span> <span class="p">(</span><span class="n">x1g</span> <span class="o">+</span> <span class="n">x2g</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="n">y_g</span> <span class="o">=</span> <span class="p">(</span><span class="n">y1g</span> <span class="o">+</span> <span class="n">y2g</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span>
+    <span class="c1"># The distance between boxes&#39; centers squared.</span>
+    <span class="n">centers_distance_squared</span> <span class="o">=</span> <span class="p">((</span><span class="n">x_p</span> <span class="o">-</span> <span class="n">x_g</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span> <span class="o">+</span> <span class="p">((</span><span class="n">y_p</span> <span class="o">-</span> <span class="n">y_g</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="c1"># The distance IoU is the IoU penalized by a normalized</span>
+    <span class="c1"># distance between boxes&#39; centers squared.</span>
+    <span class="n">loss</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">iou</span> <span class="o">+</span> <span class="p">(</span><span class="n">centers_distance_squared</span> <span class="o">/</span> <span class="n">diagonal_distance_squared</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">loss</span><span class="p">,</span> <span class="n">iou</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/drop_block.html b/0.16/_modules/torchvision/ops/drop_block.html
new file mode 100644
index 00000000000..778f4819ac8
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/drop_block.html
@@ -0,0 +1,824 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.drop_block &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.drop_block</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.drop_block</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+
+<div class="viewcode-block" id="drop_block2d"><a class="viewcode-back" href="../../../generated/torchvision.ops.drop_block2d.html#torchvision.ops.drop_block2d">[docs]</a><span class="k">def</span> <span class="nf">drop_block2d</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-06</span><span class="p">,</span> <span class="n">training</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements DropBlock2d from `&quot;DropBlock: A regularization method for convolutional networks&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1810.12890&gt;`.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, C, H, W]): The input tensor or 4-dimensions with the first one</span>
+<span class="sd">                    being its batch i.e. a batch with ``N`` rows.</span>
+<span class="sd">        p (float): Probability of an element to be dropped.</span>
+<span class="sd">        block_size (int): Size of the block to drop.</span>
+<span class="sd">        inplace (bool): If set to ``True``, will do this operation in-place. Default: ``False``.</span>
+<span class="sd">        eps (float): A value added to the denominator for numerical stability. Default: 1e-6.</span>
+<span class="sd">        training (bool): apply dropblock if is ``True``. Default: ``True``.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, C, H, W]: The randomly zeroed tensor after dropblock.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">drop_block2d</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">p</span> <span class="o">&lt;</span> <span class="mf">0.0</span> <span class="ow">or</span> <span class="n">p</span> <span class="o">&gt;</span> <span class="mf">1.0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;drop probability has to be between 0 and 1, but got </span><span class="si">{</span><span class="n">p</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">input</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;input should be 4 dimensional. Got </span><span class="si">{</span><span class="nb">input</span><span class="o">.</span><span class="n">ndim</span><span class="si">}</span><span class="s2"> dimensions.&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">training</span> <span class="ow">or</span> <span class="n">p</span> <span class="o">==</span> <span class="mf">0.0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">input</span>
+
+    <span class="n">N</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+    <span class="n">block_size</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">W</span><span class="p">,</span> <span class="n">H</span><span class="p">)</span>
+    <span class="c1"># compute the gamma of Bernoulli distribution</span>
+    <span class="n">gamma</span> <span class="o">=</span> <span class="p">(</span><span class="n">p</span> <span class="o">*</span> <span class="n">H</span> <span class="o">*</span> <span class="n">W</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">block_size</span><span class="o">**</span><span class="mi">2</span><span class="p">)</span> <span class="o">*</span> <span class="p">((</span><span class="n">H</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">W</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)))</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">((</span><span class="n">N</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">H</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">W</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">noise</span><span class="o">.</span><span class="n">bernoulli_</span><span class="p">(</span><span class="n">gamma</span><span class="p">)</span>
+
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">noise</span><span class="p">,</span> <span class="p">[</span><span class="n">block_size</span> <span class="o">//</span> <span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="mi">4</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">max_pool2d</span><span class="p">(</span><span class="n">noise</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">block_size</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="n">block_size</span> <span class="o">//</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">noise</span>
+    <span class="n">normalize_scale</span> <span class="o">=</span> <span class="n">noise</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">/</span> <span class="p">(</span><span class="n">eps</span> <span class="o">+</span> <span class="n">noise</span><span class="o">.</span><span class="n">sum</span><span class="p">())</span>
+    <span class="k">if</span> <span class="n">inplace</span><span class="p">:</span>
+        <span class="nb">input</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">noise</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">normalize_scale</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="nb">input</span> <span class="o">=</span> <span class="nb">input</span> <span class="o">*</span> <span class="n">noise</span> <span class="o">*</span> <span class="n">normalize_scale</span>
+    <span class="k">return</span> <span class="nb">input</span></div>
+
+
+<div class="viewcode-block" id="drop_block3d"><a class="viewcode-back" href="../../../generated/torchvision.ops.drop_block3d.html#torchvision.ops.drop_block3d">[docs]</a><span class="k">def</span> <span class="nf">drop_block3d</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-06</span><span class="p">,</span> <span class="n">training</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements DropBlock3d from `&quot;DropBlock: A regularization method for convolutional networks&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1810.12890&gt;`.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, C, D, H, W]): The input tensor or 5-dimensions with the first one</span>
+<span class="sd">                    being its batch i.e. a batch with ``N`` rows.</span>
+<span class="sd">        p (float): Probability of an element to be dropped.</span>
+<span class="sd">        block_size (int): Size of the block to drop.</span>
+<span class="sd">        inplace (bool): If set to ``True``, will do this operation in-place. Default: ``False``.</span>
+<span class="sd">        eps (float): A value added to the denominator for numerical stability. Default: 1e-6.</span>
+<span class="sd">        training (bool): apply dropblock if is ``True``. Default: ``True``.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, C, D, H, W]: The randomly zeroed tensor after dropblock.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">drop_block3d</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">p</span> <span class="o">&lt;</span> <span class="mf">0.0</span> <span class="ow">or</span> <span class="n">p</span> <span class="o">&gt;</span> <span class="mf">1.0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;drop probability has to be between 0 and 1, but got </span><span class="si">{</span><span class="n">p</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">input</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">5</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;input should be 5 dimensional. Got </span><span class="si">{</span><span class="nb">input</span><span class="o">.</span><span class="n">ndim</span><span class="si">}</span><span class="s2"> dimensions.&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">training</span> <span class="ow">or</span> <span class="n">p</span> <span class="o">==</span> <span class="mf">0.0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">input</span>
+
+    <span class="n">N</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">D</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+    <span class="n">block_size</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">D</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">)</span>
+    <span class="c1"># compute the gamma of Bernoulli distribution</span>
+    <span class="n">gamma</span> <span class="o">=</span> <span class="p">(</span><span class="n">p</span> <span class="o">*</span> <span class="n">D</span> <span class="o">*</span> <span class="n">H</span> <span class="o">*</span> <span class="n">W</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">block_size</span><span class="o">**</span><span class="mi">3</span><span class="p">)</span> <span class="o">*</span> <span class="p">((</span><span class="n">D</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">H</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">W</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)))</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span>
+        <span class="p">(</span><span class="n">N</span><span class="p">,</span> <span class="n">C</span><span class="p">,</span> <span class="n">D</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">H</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">W</span> <span class="o">-</span> <span class="n">block_size</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span>
+    <span class="p">)</span>
+    <span class="n">noise</span><span class="o">.</span><span class="n">bernoulli_</span><span class="p">(</span><span class="n">gamma</span><span class="p">)</span>
+
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">noise</span><span class="p">,</span> <span class="p">[</span><span class="n">block_size</span> <span class="o">//</span> <span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="mi">6</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">max_pool3d</span><span class="p">(</span>
+        <span class="n">noise</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">block_size</span><span class="p">,</span> <span class="n">block_size</span><span class="p">),</span> <span class="n">padding</span><span class="o">=</span><span class="n">block_size</span> <span class="o">//</span> <span class="mi">2</span>
+    <span class="p">)</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">noise</span>
+    <span class="n">normalize_scale</span> <span class="o">=</span> <span class="n">noise</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">/</span> <span class="p">(</span><span class="n">eps</span> <span class="o">+</span> <span class="n">noise</span><span class="o">.</span><span class="n">sum</span><span class="p">())</span>
+    <span class="k">if</span> <span class="n">inplace</span><span class="p">:</span>
+        <span class="nb">input</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">noise</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">normalize_scale</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="nb">input</span> <span class="o">=</span> <span class="nb">input</span> <span class="o">*</span> <span class="n">noise</span> <span class="o">*</span> <span class="n">normalize_scale</span>
+    <span class="k">return</span> <span class="nb">input</span></div>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;drop_block2d&quot;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="DropBlock2d"><a class="viewcode-back" href="../../../generated/torchvision.ops.DropBlock2d.html#torchvision.ops.DropBlock2d">[docs]</a><span class="k">class</span> <span class="nc">DropBlock2d</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`drop_block2d`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-06</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span> <span class="o">=</span> <span class="n">block_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span> <span class="o">=</span> <span class="n">inplace</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">eps</span> <span class="o">=</span> <span class="n">eps</span>
+
+<div class="viewcode-block" id="DropBlock2d.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.DropBlock2d.html#torchvision.ops.DropBlock2d.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            input (Tensor): Input feature map on which some areas will be randomly</span>
+<span class="sd">                dropped.</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: The tensor after DropBlock layer.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">drop_block2d</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">eps</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">, block_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="si">}</span><span class="s2">, inplace=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;drop_block3d&quot;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="DropBlock3d"><a class="viewcode-back" href="../../../generated/torchvision.ops.DropBlock3d.html#torchvision.ops.DropBlock3d">[docs]</a><span class="k">class</span> <span class="nc">DropBlock3d</span><span class="p">(</span><span class="n">DropBlock2d</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`drop_block3d`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-06</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">block_size</span><span class="p">,</span> <span class="n">inplace</span><span class="p">,</span> <span class="n">eps</span><span class="p">)</span>
+
+<div class="viewcode-block" id="DropBlock3d.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.DropBlock3d.html#torchvision.ops.DropBlock3d.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            input (Tensor): Input feature map on which some areas will be randomly</span>
+<span class="sd">                dropped.</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: The tensor after DropBlock layer.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">drop_block3d</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">eps</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">)</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/feature_pyramid_network.html b/0.16/_modules/torchvision/ops/feature_pyramid_network.html
new file mode 100644
index 00000000000..6e55bab930a
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/feature_pyramid_network.html
@@ -0,0 +1,919 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.feature_pyramid_network &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.feature_pyramid_network</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.feature_pyramid_network</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..ops.misc</span> <span class="kn">import</span> <span class="n">Conv2dNormActivation</span>
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+
+<span class="k">class</span> <span class="nc">ExtraFPNBlock</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Base class for the extra block in the FPN.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        results (List[Tensor]): the result of the FPN</span>
+<span class="sd">        x (List[Tensor]): the original feature maps</span>
+<span class="sd">        names (List[str]): the names for each one of the</span>
+<span class="sd">            original feature maps</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        results (List[Tensor]): the extended set of results</span>
+<span class="sd">            of the FPN</span>
+<span class="sd">        names (List[str]): the extended set of names for the results</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">results</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">names</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+        <span class="k">pass</span>
+
+
+<div class="viewcode-block" id="FeaturePyramidNetwork"><a class="viewcode-back" href="../../../generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork">[docs]</a><span class="k">class</span> <span class="nc">FeaturePyramidNetwork</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Module that adds a FPN from on top of a set of feature maps. This is based on</span>
+<span class="sd">    `&quot;Feature Pyramid Network for Object Detection&quot; &lt;https://arxiv.org/abs/1612.03144&gt;`_.</span>
+
+<span class="sd">    The feature maps are currently supposed to be in increasing depth</span>
+<span class="sd">    order.</span>
+
+<span class="sd">    The input to the model is expected to be an OrderedDict[Tensor], containing</span>
+<span class="sd">    the feature maps on top of which the FPN will be added.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels_list (list[int]): number of channels for each feature map that</span>
+<span class="sd">            is passed to the module</span>
+<span class="sd">        out_channels (int): number of channels of the FPN representation</span>
+<span class="sd">        extra_blocks (ExtraFPNBlock or None): if provided, extra operations will</span>
+<span class="sd">            be performed. It is expected to take the fpn features, the original</span>
+<span class="sd">            features and the names of the original features as input, and returns</span>
+<span class="sd">            a new list of feature maps and their corresponding names</span>
+<span class="sd">        norm_layer (callable, optional): Module specifying the normalization layer to use. Default: None</span>
+
+<span class="sd">    Examples::</span>
+
+<span class="sd">        &gt;&gt;&gt; m = torchvision.ops.FeaturePyramidNetwork([10, 20, 30], 5)</span>
+<span class="sd">        &gt;&gt;&gt; # get some dummy data</span>
+<span class="sd">        &gt;&gt;&gt; x = OrderedDict()</span>
+<span class="sd">        &gt;&gt;&gt; x[&#39;feat0&#39;] = torch.rand(1, 10, 64, 64)</span>
+<span class="sd">        &gt;&gt;&gt; x[&#39;feat2&#39;] = torch.rand(1, 20, 16, 16)</span>
+<span class="sd">        &gt;&gt;&gt; x[&#39;feat3&#39;] = torch.rand(1, 30, 8, 8)</span>
+<span class="sd">        &gt;&gt;&gt; # compute the FPN on top of x</span>
+<span class="sd">        &gt;&gt;&gt; output = m(x)</span>
+<span class="sd">        &gt;&gt;&gt; print([(k, v.shape) for k, v in output.items()])</span>
+<span class="sd">        &gt;&gt;&gt; # returns</span>
+<span class="sd">        &gt;&gt;&gt;   [(&#39;feat0&#39;, torch.Size([1, 5, 64, 64])),</span>
+<span class="sd">        &gt;&gt;&gt;    (&#39;feat2&#39;, torch.Size([1, 5, 16, 16])),</span>
+<span class="sd">        &gt;&gt;&gt;    (&#39;feat3&#39;, torch.Size([1, 5, 8, 8]))]</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_version</span> <span class="o">=</span> <span class="mi">2</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">extra_blocks</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">ExtraFPNBlock</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inner_blocks</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">layer_blocks</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">in_channels</span> <span class="ow">in</span> <span class="n">in_channels_list</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">in_channels</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;in_channels=0 is currently not supported&quot;</span><span class="p">)</span>
+            <span class="n">inner_block_module</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span>
+            <span class="p">)</span>
+            <span class="n">layer_block_module</span> <span class="o">=</span> <span class="n">Conv2dNormActivation</span><span class="p">(</span>
+                <span class="n">out_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">norm_layer</span><span class="o">=</span><span class="n">norm_layer</span><span class="p">,</span> <span class="n">activation_layer</span><span class="o">=</span><span class="kc">None</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">inner_blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">inner_block_module</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">layer_blocks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">layer_block_module</span><span class="p">)</span>
+
+        <span class="c1"># initialize parameters now to avoid modifying the initialization of top_blocks</span>
+        <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">):</span>
+                <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_uniform_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">a</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">m</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">m</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">extra_blocks</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">extra_blocks</span><span class="p">,</span> <span class="n">ExtraFPNBlock</span><span class="p">):</span>
+                <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;extra_blocks should be of type ExtraFPNBlock not </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">extra_blocks</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">extra_blocks</span> <span class="o">=</span> <span class="n">extra_blocks</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">,</span>
+        <span class="n">unexpected_keys</span><span class="p">,</span>
+        <span class="n">error_msgs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">version</span> <span class="o">=</span> <span class="n">local_metadata</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;version&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">version</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">version</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">num_blocks</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inner_blocks</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">block</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;inner_blocks&quot;</span><span class="p">,</span> <span class="s2">&quot;layer_blocks&quot;</span><span class="p">]:</span>
+                <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_blocks</span><span class="p">):</span>
+                    <span class="k">for</span> <span class="nb">type</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;weight&quot;</span><span class="p">,</span> <span class="s2">&quot;bias&quot;</span><span class="p">]:</span>
+                        <span class="n">old_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}{</span><span class="n">block</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+                        <span class="n">new_key</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}{</span><span class="n">block</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">.0.</span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">&quot;</span>
+                        <span class="k">if</span> <span class="n">old_key</span> <span class="ow">in</span> <span class="n">state_dict</span><span class="p">:</span>
+                            <span class="n">state_dict</span><span class="p">[</span><span class="n">new_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">state_dict</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">old_key</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span>
+            <span class="n">prefix</span><span class="p">,</span>
+            <span class="n">local_metadata</span><span class="p">,</span>
+            <span class="n">strict</span><span class="p">,</span>
+            <span class="n">missing_keys</span><span class="p">,</span>
+            <span class="n">unexpected_keys</span><span class="p">,</span>
+            <span class="n">error_msgs</span><span class="p">,</span>
+        <span class="p">)</span>
+
+<div class="viewcode-block" id="FeaturePyramidNetwork.get_result_from_inner_blocks"><a class="viewcode-back" href="../../../generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork.get_result_from_inner_blocks">[docs]</a>    <span class="k">def</span> <span class="nf">get_result_from_inner_blocks</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        This is equivalent to self.inner_blocks[idx](x),</span>
+<span class="sd">        but torchscript doesn&#39;t support this yet</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">num_blocks</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inner_blocks</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">idx</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">idx</span> <span class="o">+=</span> <span class="n">num_blocks</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">x</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">module</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inner_blocks</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="n">idx</span><span class="p">:</span>
+                <span class="n">out</span> <span class="o">=</span> <span class="n">module</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">out</span></div>
+
+<div class="viewcode-block" id="FeaturePyramidNetwork.get_result_from_layer_blocks"><a class="viewcode-back" href="../../../generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork.get_result_from_layer_blocks">[docs]</a>    <span class="k">def</span> <span class="nf">get_result_from_layer_blocks</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        This is equivalent to self.layer_blocks[idx](x),</span>
+<span class="sd">        but torchscript doesn&#39;t support this yet</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">num_blocks</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">layer_blocks</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">idx</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">idx</span> <span class="o">+=</span> <span class="n">num_blocks</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">x</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">module</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">layer_blocks</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="n">idx</span><span class="p">:</span>
+                <span class="n">out</span> <span class="o">=</span> <span class="n">module</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">out</span></div>
+
+<div class="viewcode-block" id="FeaturePyramidNetwork.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Computes the FPN for a set of feature maps.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            x (OrderedDict[Tensor]): feature maps for each feature level.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            results (OrderedDict[Tensor]): feature maps after FPN layers.</span>
+<span class="sd">                They are ordered from the highest resolution first.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># unpack OrderedDict into two lists for easier handling</span>
+        <span class="n">names</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+
+        <span class="n">last_inner</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_result_from_inner_blocks</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">results</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">results</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">get_result_from_layer_blocks</span><span class="p">(</span><span class="n">last_inner</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+
+        <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="o">-</span> <span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">):</span>
+            <span class="n">inner_lateral</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_result_from_inner_blocks</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span> <span class="n">idx</span><span class="p">)</span>
+            <span class="n">feat_shape</span> <span class="o">=</span> <span class="n">inner_lateral</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+            <span class="n">inner_top_down</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">interpolate</span><span class="p">(</span><span class="n">last_inner</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">feat_shape</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;nearest&quot;</span><span class="p">)</span>
+            <span class="n">last_inner</span> <span class="o">=</span> <span class="n">inner_lateral</span> <span class="o">+</span> <span class="n">inner_top_down</span>
+            <span class="n">results</span><span class="o">.</span><span class="n">insert</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_result_from_layer_blocks</span><span class="p">(</span><span class="n">last_inner</span><span class="p">,</span> <span class="n">idx</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">extra_blocks</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">results</span><span class="p">,</span> <span class="n">names</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">extra_blocks</span><span class="p">(</span><span class="n">results</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">names</span><span class="p">)</span>
+
+        <span class="c1"># make it back an OrderedDict</span>
+        <span class="n">out</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">([(</span><span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">names</span><span class="p">,</span> <span class="n">results</span><span class="p">)])</span>
+
+        <span class="k">return</span> <span class="n">out</span></div></div>
+
+
+<span class="k">class</span> <span class="nc">LastLevelMaxPool</span><span class="p">(</span><span class="n">ExtraFPNBlock</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Applies a max_pool2d (not actual max_pool2d, we just subsample) on top of the last feature map</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">x</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">y</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">names</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+        <span class="n">names</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s2">&quot;pool&quot;</span><span class="p">)</span>
+        <span class="c1"># Use max pooling to simulate stride 2 subsampling</span>
+        <span class="n">x</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">max_pool2d</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">))</span>
+        <span class="k">return</span> <span class="n">x</span><span class="p">,</span> <span class="n">names</span>
+
+
+<span class="k">class</span> <span class="nc">LastLevelP6P7</span><span class="p">(</span><span class="n">ExtraFPNBlock</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    This module is used in RetinaNet to generate extra layers, P6 and P7.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p6</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p7</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">out_channels</span><span class="p">,</span> <span class="n">out_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">module</span> <span class="ow">in</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">p6</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">p7</span><span class="p">]:</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">kaiming_uniform_</span><span class="p">(</span><span class="n">module</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">a</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span><span class="n">module</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_P5</span> <span class="o">=</span> <span class="n">in_channels</span> <span class="o">==</span> <span class="n">out_channels</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">p</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">c</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">names</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+        <span class="n">p5</span><span class="p">,</span> <span class="n">c5</span> <span class="o">=</span> <span class="n">p</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">c</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">p5</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_P5</span> <span class="k">else</span> <span class="n">c5</span>
+        <span class="n">p6</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">p6</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">p7</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">p7</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">p6</span><span class="p">))</span>
+        <span class="n">p</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="n">p6</span><span class="p">,</span> <span class="n">p7</span><span class="p">])</span>
+        <span class="n">names</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="s2">&quot;p6&quot;</span><span class="p">,</span> <span class="s2">&quot;p7&quot;</span><span class="p">])</span>
+        <span class="k">return</span> <span class="n">p</span><span class="p">,</span> <span class="n">names</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/focal_loss.html b/0.16/_modules/torchvision/ops/focal_loss.html
new file mode 100644
index 00000000000..b21aedf25e1
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/focal_loss.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.focal_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.focal_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.focal_loss</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+
+<div class="viewcode-block" id="sigmoid_focal_loss"><a class="viewcode-back" href="../../../generated/torchvision.ops.sigmoid_focal_loss.html#torchvision.ops.sigmoid_focal_loss">[docs]</a><span class="k">def</span> <span class="nf">sigmoid_focal_loss</span><span class="p">(</span>
+    <span class="n">inputs</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">targets</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.25</span><span class="p">,</span>
+    <span class="n">gamma</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">2</span><span class="p">,</span>
+    <span class="n">reduction</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;none&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Loss used in RetinaNet for dense detection: https://arxiv.org/abs/1708.02002.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        inputs (Tensor): A float tensor of arbitrary shape.</span>
+<span class="sd">                The predictions for each example.</span>
+<span class="sd">        targets (Tensor): A float tensor with the same shape as inputs. Stores the binary</span>
+<span class="sd">                classification label for each element in inputs</span>
+<span class="sd">                (0 for the negative class and 1 for the positive class).</span>
+<span class="sd">        alpha (float): Weighting factor in range (0,1) to balance</span>
+<span class="sd">                positive vs negative examples or -1 for ignore. Default: ``0.25``.</span>
+<span class="sd">        gamma (float): Exponent of the modulating factor (1 - p_t) to</span>
+<span class="sd">                balance easy vs hard examples. Default: ``2``.</span>
+<span class="sd">        reduction (string): ``&#39;none&#39;`` | ``&#39;mean&#39;`` | ``&#39;sum&#39;``</span>
+<span class="sd">                ``&#39;none&#39;``: No reduction will be applied to the output.</span>
+<span class="sd">                ``&#39;mean&#39;``: The output will be averaged.</span>
+<span class="sd">                ``&#39;sum&#39;``: The output will be summed. Default: ``&#39;none&#39;``.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        Loss tensor with the reduction option applied.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="c1"># Original implementation from https://github.com/facebookresearch/fvcore/blob/master/fvcore/nn/focal_loss.py</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">sigmoid_focal_loss</span><span class="p">)</span>
+    <span class="n">p</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+    <span class="n">ce_loss</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">binary_cross_entropy_with_logits</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">targets</span><span class="p">,</span> <span class="n">reduction</span><span class="o">=</span><span class="s2">&quot;none&quot;</span><span class="p">)</span>
+    <span class="n">p_t</span> <span class="o">=</span> <span class="n">p</span> <span class="o">*</span> <span class="n">targets</span> <span class="o">+</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">targets</span><span class="p">)</span>
+    <span class="n">loss</span> <span class="o">=</span> <span class="n">ce_loss</span> <span class="o">*</span> <span class="p">((</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p_t</span><span class="p">)</span> <span class="o">**</span> <span class="n">gamma</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">alpha</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">alpha_t</span> <span class="o">=</span> <span class="n">alpha</span> <span class="o">*</span> <span class="n">targets</span> <span class="o">+</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">alpha</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">targets</span><span class="p">)</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">alpha_t</span> <span class="o">*</span> <span class="n">loss</span>
+
+    <span class="c1"># Check reduction option and return loss accordingly</span>
+    <span class="k">if</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;none&quot;</span><span class="p">:</span>
+        <span class="k">pass</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;mean&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;sum&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Invalid Value for arg &#39;reduction&#39;: &#39;</span><span class="si">{</span><span class="n">reduction</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2"> Supported reduction modes: &#39;none&#39;, &#39;mean&#39;, &#39;sum&#39;&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">loss</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/giou_loss.html b/0.16/_modules/torchvision/ops/giou_loss.html
new file mode 100644
index 00000000000..90dd1dfcb27
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/giou_loss.html
@@ -0,0 +1,745 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.giou_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.giou_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.giou_loss</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_loss_inter_union</span><span class="p">,</span> <span class="n">_upcast_non_float</span>
+
+
+<div class="viewcode-block" id="generalized_box_iou_loss"><a class="viewcode-back" href="../../../generated/torchvision.ops.generalized_box_iou_loss.html#torchvision.ops.generalized_box_iou_loss">[docs]</a><span class="k">def</span> <span class="nf">generalized_box_iou_loss</span><span class="p">(</span>
+    <span class="n">boxes1</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes2</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">reduction</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;none&quot;</span><span class="p">,</span>
+    <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-7</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Gradient-friendly IoU loss with an additional penalty that is non-zero when the</span>
+<span class="sd">    boxes do not overlap and scales with the size of their smallest enclosing box.</span>
+<span class="sd">    This loss is symmetric, so the boxes1 and boxes2 arguments are interchangeable.</span>
+
+<span class="sd">    Both sets of boxes are expected to be in ``(x1, y1, x2, y2)`` format with</span>
+<span class="sd">    ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``, and The two boxes should have the</span>
+<span class="sd">    same dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        boxes1 (Tensor[N, 4] or Tensor[4]): first set of boxes</span>
+<span class="sd">        boxes2 (Tensor[N, 4] or Tensor[4]): second set of boxes</span>
+<span class="sd">        reduction (string, optional): Specifies the reduction to apply to the output:</span>
+<span class="sd">            ``&#39;none&#39;`` | ``&#39;mean&#39;`` | ``&#39;sum&#39;``. ``&#39;none&#39;``: No reduction will be</span>
+<span class="sd">            applied to the output. ``&#39;mean&#39;``: The output will be averaged.</span>
+<span class="sd">            ``&#39;sum&#39;``: The output will be summed. Default: ``&#39;none&#39;``</span>
+<span class="sd">        eps (float): small number to prevent division by zero. Default: 1e-7</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: Loss tensor with the reduction option applied.</span>
+
+<span class="sd">    Reference:</span>
+<span class="sd">        Hamid Rezatofighi et al.: Generalized Intersection over Union:</span>
+<span class="sd">        A Metric and A Loss for Bounding Box Regression:</span>
+<span class="sd">        https://arxiv.org/abs/1902.09630</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="c1"># Original implementation from https://github.com/facebookresearch/fvcore/blob/bfff2ef/fvcore/nn/giou_loss.py</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">generalized_box_iou_loss</span><span class="p">)</span>
+
+    <span class="n">boxes1</span> <span class="o">=</span> <span class="n">_upcast_non_float</span><span class="p">(</span><span class="n">boxes1</span><span class="p">)</span>
+    <span class="n">boxes2</span> <span class="o">=</span> <span class="n">_upcast_non_float</span><span class="p">(</span><span class="n">boxes2</span><span class="p">)</span>
+    <span class="n">intsctk</span><span class="p">,</span> <span class="n">unionk</span> <span class="o">=</span> <span class="n">_loss_inter_union</span><span class="p">(</span><span class="n">boxes1</span><span class="p">,</span> <span class="n">boxes2</span><span class="p">)</span>
+    <span class="n">iouk</span> <span class="o">=</span> <span class="n">intsctk</span> <span class="o">/</span> <span class="p">(</span><span class="n">unionk</span> <span class="o">+</span> <span class="n">eps</span><span class="p">)</span>
+
+    <span class="n">x1</span><span class="p">,</span> <span class="n">y1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">y2</span> <span class="o">=</span> <span class="n">boxes1</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">x1g</span><span class="p">,</span> <span class="n">y1g</span><span class="p">,</span> <span class="n">x2g</span><span class="p">,</span> <span class="n">y2g</span> <span class="o">=</span> <span class="n">boxes2</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="c1"># smallest enclosing box</span>
+    <span class="n">xc1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">x1</span><span class="p">,</span> <span class="n">x1g</span><span class="p">)</span>
+    <span class="n">yc1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">y1</span><span class="p">,</span> <span class="n">y1g</span><span class="p">)</span>
+    <span class="n">xc2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">x2</span><span class="p">,</span> <span class="n">x2g</span><span class="p">)</span>
+    <span class="n">yc2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">y2</span><span class="p">,</span> <span class="n">y2g</span><span class="p">)</span>
+
+    <span class="n">area_c</span> <span class="o">=</span> <span class="p">(</span><span class="n">xc2</span> <span class="o">-</span> <span class="n">xc1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">yc2</span> <span class="o">-</span> <span class="n">yc1</span><span class="p">)</span>
+    <span class="n">miouk</span> <span class="o">=</span> <span class="n">iouk</span> <span class="o">-</span> <span class="p">((</span><span class="n">area_c</span> <span class="o">-</span> <span class="n">unionk</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="n">area_c</span> <span class="o">+</span> <span class="n">eps</span><span class="p">))</span>
+
+    <span class="n">loss</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">miouk</span>
+
+    <span class="c1"># Check reduction option and return loss accordingly</span>
+    <span class="k">if</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;none&quot;</span><span class="p">:</span>
+        <span class="k">pass</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;mean&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span> <span class="k">if</span> <span class="n">loss</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="mf">0.0</span> <span class="o">*</span> <span class="n">loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+    <span class="k">elif</span> <span class="n">reduction</span> <span class="o">==</span> <span class="s2">&quot;sum&quot;</span><span class="p">:</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Invalid Value for arg &#39;reduction&#39;: &#39;</span><span class="si">{</span><span class="n">reduction</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2"> Supported reduction modes: &#39;none&#39;, &#39;mean&#39;, &#39;sum&#39;&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">loss</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/misc.html b/0.16/_modules/torchvision/ops/misc.html
new file mode 100644
index 00000000000..3370851f775
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/misc.html
@@ -0,0 +1,989 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.misc &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.misc</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.misc</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span><span class="p">,</span> <span class="n">_make_ntuple</span>
+
+
+<span class="n">interpolate</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">interpolate</span>
+
+
+<div class="viewcode-block" id="FrozenBatchNorm2d"><a class="viewcode-back" href="../../../generated/torchvision.ops.FrozenBatchNorm2d.html#torchvision.ops.FrozenBatchNorm2d">[docs]</a><span class="k">class</span> <span class="nc">FrozenBatchNorm2d</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    BatchNorm2d where the batch statistics and the affine parameters are fixed</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_features (int): Number of features ``C`` from an expected input of size ``(N, C, H, W)``</span>
+<span class="sd">        eps (float): a value added to the denominator for numerical stability. Default: 1e-5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_features</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-5</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">eps</span> <span class="o">=</span> <span class="n">eps</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;weight&quot;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="n">num_features</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;bias&quot;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">num_features</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;running_mean&quot;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="n">num_features</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;running_var&quot;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="n">num_features</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_load_from_state_dict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">state_dict</span><span class="p">:</span> <span class="nb">dict</span><span class="p">,</span>
+        <span class="n">prefix</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">local_metadata</span><span class="p">:</span> <span class="nb">dict</span><span class="p">,</span>
+        <span class="n">strict</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">missing_keys</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+        <span class="n">unexpected_keys</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+        <span class="n">error_msgs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+    <span class="p">):</span>
+        <span class="n">num_batches_tracked_key</span> <span class="o">=</span> <span class="n">prefix</span> <span class="o">+</span> <span class="s2">&quot;num_batches_tracked&quot;</span>
+        <span class="k">if</span> <span class="n">num_batches_tracked_key</span> <span class="ow">in</span> <span class="n">state_dict</span><span class="p">:</span>
+            <span class="k">del</span> <span class="n">state_dict</span><span class="p">[</span><span class="n">num_batches_tracked_key</span><span class="p">]</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_load_from_state_dict</span><span class="p">(</span>
+            <span class="n">state_dict</span><span class="p">,</span> <span class="n">prefix</span><span class="p">,</span> <span class="n">local_metadata</span><span class="p">,</span> <span class="n">strict</span><span class="p">,</span> <span class="n">missing_keys</span><span class="p">,</span> <span class="n">unexpected_keys</span><span class="p">,</span> <span class="n">error_msgs</span>
+        <span class="p">)</span>
+
+<div class="viewcode-block" id="FrozenBatchNorm2d.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.FrozenBatchNorm2d.html#torchvision.ops.FrozenBatchNorm2d.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># move reshapes to the beginning</span>
+        <span class="c1"># to make it fuser-friendly</span>
+        <span class="n">w</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">b</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">bias</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">rv</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">running_var</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">rm</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">running_mean</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="n">w</span> <span class="o">*</span> <span class="p">(</span><span class="n">rv</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">eps</span><span class="p">)</span><span class="o">.</span><span class="n">rsqrt</span><span class="p">()</span>
+        <span class="n">bias</span> <span class="o">=</span> <span class="n">b</span> <span class="o">-</span> <span class="n">rm</span> <span class="o">*</span> <span class="n">scale</span>
+        <span class="k">return</span> <span class="n">x</span> <span class="o">*</span> <span class="n">scale</span> <span class="o">+</span> <span class="n">bias</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">, eps=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">eps</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<span class="k">class</span> <span class="nc">ConvNormActivation</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">kernel_size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="o">...</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="o">...</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">padding</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="o">...</span><span class="p">],</span> <span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="o">...</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">inplace</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">conv_layer</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="k">if</span> <span class="n">padding</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="nb">int</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">dilation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+                <span class="n">padding</span> <span class="o">=</span> <span class="p">(</span><span class="n">kernel_size</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">dilation</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">_conv_dim</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="k">else</span> <span class="nb">len</span><span class="p">(</span><span class="n">dilation</span><span class="p">)</span>
+                <span class="n">kernel_size</span> <span class="o">=</span> <span class="n">_make_ntuple</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">_conv_dim</span><span class="p">)</span>
+                <span class="n">dilation</span> <span class="o">=</span> <span class="n">_make_ntuple</span><span class="p">(</span><span class="n">dilation</span><span class="p">,</span> <span class="n">_conv_dim</span><span class="p">)</span>
+                <span class="n">padding</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">((</span><span class="n">kernel_size</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">dilation</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">_conv_dim</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">bias</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">bias</span> <span class="o">=</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="kc">None</span>
+
+        <span class="n">layers</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="n">conv_layer</span><span class="p">(</span>
+                <span class="n">in_channels</span><span class="p">,</span>
+                <span class="n">out_channels</span><span class="p">,</span>
+                <span class="n">kernel_size</span><span class="p">,</span>
+                <span class="n">stride</span><span class="p">,</span>
+                <span class="n">padding</span><span class="p">,</span>
+                <span class="n">dilation</span><span class="o">=</span><span class="n">dilation</span><span class="p">,</span>
+                <span class="n">groups</span><span class="o">=</span><span class="n">groups</span><span class="p">,</span>
+                <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">norm_layer</span><span class="p">(</span><span class="n">out_channels</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="n">activation_layer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">params</span> <span class="o">=</span> <span class="p">{}</span> <span class="k">if</span> <span class="n">inplace</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="p">{</span><span class="s2">&quot;inplace&quot;</span><span class="p">:</span> <span class="n">inplace</span><span class="p">}</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">activation_layer</span><span class="p">(</span><span class="o">**</span><span class="n">params</span><span class="p">))</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">out_channels</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span> <span class="o">==</span> <span class="n">ConvNormActivation</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;Don&#39;t use ConvNormActivation directly, please use Conv2dNormActivation and Conv3dNormActivation instead.&quot;</span>
+            <span class="p">)</span>
+
+
+<div class="viewcode-block" id="Conv2dNormActivation"><a class="viewcode-back" href="../../../generated/torchvision.ops.Conv2dNormActivation.html#torchvision.ops.Conv2dNormActivation">[docs]</a><span class="k">class</span> <span class="nc">Conv2dNormActivation</span><span class="p">(</span><span class="n">ConvNormActivation</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Configurable block used for Convolution2d-Normalization-Activation blocks.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): Number of channels in the input image</span>
+<span class="sd">        out_channels (int): Number of channels produced by the Convolution-Normalization-Activation block</span>
+<span class="sd">        kernel_size: (int, optional): Size of the convolving kernel. Default: 3</span>
+<span class="sd">        stride (int, optional): Stride of the convolution. Default: 1</span>
+<span class="sd">        padding (int, tuple or str, optional): Padding added to all four sides of the input. Default: None, in which case it will be calculated as ``padding = (kernel_size - 1) // 2 * dilation``</span>
+<span class="sd">        groups (int, optional): Number of blocked connections from input channels to output channels. Default: 1</span>
+<span class="sd">        norm_layer (Callable[..., torch.nn.Module], optional): Norm layer that will be stacked on top of the convolution layer. If ``None`` this layer won&#39;t be used. Default: ``torch.nn.BatchNorm2d``</span>
+<span class="sd">        activation_layer (Callable[..., torch.nn.Module], optional): Activation function which will be stacked on top of the normalization layer (if not None), otherwise on top of the conv layer. If ``None`` this layer won&#39;t be used. Default: ``torch.nn.ReLU``</span>
+<span class="sd">        dilation (int): Spacing between kernel elements. Default: 1</span>
+<span class="sd">        inplace (bool): Parameter for the activation layer, which can optionally do the operation in-place. Default ``True``</span>
+<span class="sd">        bias (bool, optional): Whether to use bias in the convolution layer. By default, biases are included if ``norm_layer is None``.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">kernel_size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">padding</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">,</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">inplace</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">out_channels</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="p">,</span>
+            <span class="n">stride</span><span class="p">,</span>
+            <span class="n">padding</span><span class="p">,</span>
+            <span class="n">groups</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">dilation</span><span class="p">,</span>
+            <span class="n">inplace</span><span class="p">,</span>
+            <span class="n">bias</span><span class="p">,</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Conv3dNormActivation"><a class="viewcode-back" href="../../../generated/torchvision.ops.Conv3dNormActivation.html#torchvision.ops.Conv3dNormActivation">[docs]</a><span class="k">class</span> <span class="nc">Conv3dNormActivation</span><span class="p">(</span><span class="n">ConvNormActivation</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Configurable block used for Convolution3d-Normalization-Activation blocks.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): Number of channels in the input video.</span>
+<span class="sd">        out_channels (int): Number of channels produced by the Convolution-Normalization-Activation block</span>
+<span class="sd">        kernel_size: (int, optional): Size of the convolving kernel. Default: 3</span>
+<span class="sd">        stride (int, optional): Stride of the convolution. Default: 1</span>
+<span class="sd">        padding (int, tuple or str, optional): Padding added to all four sides of the input. Default: None, in which case it will be calculated as ``padding = (kernel_size - 1) // 2 * dilation``</span>
+<span class="sd">        groups (int, optional): Number of blocked connections from input channels to output channels. Default: 1</span>
+<span class="sd">        norm_layer (Callable[..., torch.nn.Module], optional): Norm layer that will be stacked on top of the convolution layer. If ``None`` this layer won&#39;t be used. Default: ``torch.nn.BatchNorm3d``</span>
+<span class="sd">        activation_layer (Callable[..., torch.nn.Module], optional): Activation function which will be stacked on top of the normalization layer (if not None), otherwise on top of the conv layer. If ``None`` this layer won&#39;t be used. Default: ``torch.nn.ReLU``</span>
+<span class="sd">        dilation (int): Spacing between kernel elements. Default: 1</span>
+<span class="sd">        inplace (bool): Parameter for the activation layer, which can optionally do the operation in-place. Default ``True``</span>
+<span class="sd">        bias (bool, optional): Whether to use bias in the convolution layer. By default, biases are included if ``norm_layer is None``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">out_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">kernel_size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">stride</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">padding</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm3d</span><span class="p">,</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">,</span>
+        <span class="n">dilation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+        <span class="n">inplace</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">bias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">in_channels</span><span class="p">,</span>
+            <span class="n">out_channels</span><span class="p">,</span>
+            <span class="n">kernel_size</span><span class="p">,</span>
+            <span class="n">stride</span><span class="p">,</span>
+            <span class="n">padding</span><span class="p">,</span>
+            <span class="n">groups</span><span class="p">,</span>
+            <span class="n">norm_layer</span><span class="p">,</span>
+            <span class="n">activation_layer</span><span class="p">,</span>
+            <span class="n">dilation</span><span class="p">,</span>
+            <span class="n">inplace</span><span class="p">,</span>
+            <span class="n">bias</span><span class="p">,</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv3d</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="SqueezeExcitation"><a class="viewcode-back" href="../../../generated/torchvision.ops.SqueezeExcitation.html#torchvision.ops.SqueezeExcitation">[docs]</a><span class="k">class</span> <span class="nc">SqueezeExcitation</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    This block implements the Squeeze-and-Excitation block from https://arxiv.org/abs/1709.01507 (see Fig. 1).</span>
+<span class="sd">    Parameters ``activation``, and ``scale_activation`` correspond to ``delta`` and ``sigma`` in eq. 3.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input_channels (int): Number of channels in the input image</span>
+<span class="sd">        squeeze_channels (int): Number of squeeze channels</span>
+<span class="sd">        activation (Callable[..., torch.nn.Module], optional): ``delta`` activation. Default: ``torch.nn.ReLU``</span>
+<span class="sd">        scale_activation (Callable[..., torch.nn.Module]): ``sigma`` activation. Default: ``torch.nn.Sigmoid``</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">input_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">squeeze_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">activation</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">,</span>
+        <span class="n">scale_activation</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Sigmoid</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">squeeze_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fc2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">squeeze_channels</span><span class="p">,</span> <span class="n">input_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">activation</span> <span class="o">=</span> <span class="n">activation</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale_activation</span> <span class="o">=</span> <span class="n">scale_activation</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_scale</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc1</span><span class="p">(</span><span class="n">scale</span><span class="p">)</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">activation</span><span class="p">(</span><span class="n">scale</span><span class="p">)</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc2</span><span class="p">(</span><span class="n">scale</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale_activation</span><span class="p">(</span><span class="n">scale</span><span class="p">)</span>
+
+<div class="viewcode-block" id="SqueezeExcitation.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.SqueezeExcitation.html#torchvision.ops.SqueezeExcitation.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_scale</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">scale</span> <span class="o">*</span> <span class="nb">input</span></div></div>
+
+
+<div class="viewcode-block" id="MLP"><a class="viewcode-back" href="../../../generated/torchvision.ops.MLP.html#torchvision.ops.MLP">[docs]</a><span class="k">class</span> <span class="nc">MLP</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;This block implements the multi-layer perceptron (MLP) module.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        in_channels (int): Number of channels of the input</span>
+<span class="sd">        hidden_channels (List[int]): List of the hidden channel dimensions</span>
+<span class="sd">        norm_layer (Callable[..., torch.nn.Module], optional): Norm layer that will be stacked on top of the linear layer. If ``None`` this layer won&#39;t be used. Default: ``None``</span>
+<span class="sd">        activation_layer (Callable[..., torch.nn.Module], optional): Activation function which will be stacked on top of the normalization layer (if not None), otherwise on top of the linear layer. If ``None`` this layer won&#39;t be used. Default: ``torch.nn.ReLU``</span>
+<span class="sd">        inplace (bool, optional): Parameter for the activation layer, which can optionally do the operation in-place.</span>
+<span class="sd">            Default is ``None``, which uses the respective default values of the ``activation_layer`` and Dropout layer.</span>
+<span class="sd">        bias (bool): Whether to use bias in the linear layer. Default ``True``</span>
+<span class="sd">        dropout (float): The probability for the dropout layer. Default: 0.0</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">in_channels</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">hidden_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">norm_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">activation_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">,</span>
+        <span class="n">inplace</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">bias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="c1"># The addition of `norm_layer` is inspired from the implementation of TorchMultimodal:</span>
+        <span class="c1"># https://github.com/facebookresearch/multimodal/blob/5dec8a/torchmultimodal/modules/layers/mlp.py</span>
+        <span class="n">params</span> <span class="o">=</span> <span class="p">{}</span> <span class="k">if</span> <span class="n">inplace</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="p">{</span><span class="s2">&quot;inplace&quot;</span><span class="p">:</span> <span class="n">inplace</span><span class="p">}</span>
+
+        <span class="n">layers</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">in_dim</span> <span class="o">=</span> <span class="n">in_channels</span>
+        <span class="k">for</span> <span class="n">hidden_dim</span> <span class="ow">in</span> <span class="n">hidden_channels</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]:</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_dim</span><span class="p">,</span> <span class="n">hidden_dim</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">))</span>
+            <span class="k">if</span> <span class="n">norm_layer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">norm_layer</span><span class="p">(</span><span class="n">hidden_dim</span><span class="p">))</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">activation_layer</span><span class="p">(</span><span class="o">**</span><span class="n">params</span><span class="p">))</span>
+            <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">dropout</span><span class="p">,</span> <span class="o">**</span><span class="n">params</span><span class="p">))</span>
+            <span class="n">in_dim</span> <span class="o">=</span> <span class="n">hidden_dim</span>
+
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_dim</span><span class="p">,</span> <span class="n">hidden_channels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">))</span>
+        <span class="n">layers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">dropout</span><span class="p">,</span> <span class="o">**</span><span class="n">params</span><span class="p">))</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="o">*</span><span class="n">layers</span><span class="p">)</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Permute"><a class="viewcode-back" href="../../../generated/torchvision.ops.Permute.html#torchvision.ops.Permute">[docs]</a><span class="k">class</span> <span class="nc">Permute</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;This module returns a view of the tensor input with its dimensions permuted.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        dims (List[int]): The desired ordering of dimensions</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dims</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dims</span> <span class="o">=</span> <span class="n">dims</span>
+
+<div class="viewcode-block" id="Permute.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.Permute.html#torchvision.ops.Permute.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">dims</span><span class="p">)</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/poolers.html b/0.16/_modules/torchvision/ops/poolers.html
new file mode 100644
index 00000000000..fa08c7510bf
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/poolers.html
@@ -0,0 +1,996 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.poolers &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.poolers</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.poolers</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">import</span> <span class="nn">torchvision</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops.boxes</span> <span class="kn">import</span> <span class="n">box_area</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.roi_align</span> <span class="kn">import</span> <span class="n">roi_align</span>
+
+
+<span class="c1"># copying result_idx_in_level to a specific index in result[]</span>
+<span class="c1"># is not supported by ONNX tracing yet.</span>
+<span class="c1"># _onnx_merge_levels() is an implementation supported by ONNX</span>
+<span class="c1"># that merges the levels to the right indices</span>
+<span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+<span class="k">def</span> <span class="nf">_onnx_merge_levels</span><span class="p">(</span><span class="n">levels</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">unmerged_results</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">first_result</span> <span class="o">=</span> <span class="n">unmerged_results</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">dtype</span><span class="p">,</span> <span class="n">device</span> <span class="o">=</span> <span class="n">first_result</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">first_result</span><span class="o">.</span><span class="n">device</span>
+    <span class="n">res</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span>
+        <span class="p">(</span><span class="n">levels</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">first_result</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">first_result</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">first_result</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">3</span><span class="p">)),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span>
+    <span class="p">)</span>
+    <span class="k">for</span> <span class="n">level</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">unmerged_results</span><span class="p">)):</span>
+        <span class="n">index</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">levels</span> <span class="o">==</span> <span class="n">level</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="n">index</span> <span class="o">=</span> <span class="n">index</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span>
+            <span class="n">index</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span>
+            <span class="n">unmerged_results</span><span class="p">[</span><span class="n">level</span><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">unmerged_results</span><span class="p">[</span><span class="n">level</span><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span>
+            <span class="n">unmerged_results</span><span class="p">[</span><span class="n">level</span><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">3</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="n">res</span><span class="o">.</span><span class="n">scatter</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">index</span><span class="p">,</span> <span class="n">unmerged_results</span><span class="p">[</span><span class="n">level</span><span class="p">])</span>
+    <span class="k">return</span> <span class="n">res</span>
+
+
+<span class="c1"># TODO: (eellison) T54974082 https://github.com/pytorch/pytorch/issues/26744/pytorch/issues/26744</span>
+<span class="k">def</span> <span class="nf">initLevelMapper</span><span class="p">(</span>
+    <span class="n">k_min</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">k_max</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">canonical_scale</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">224</span><span class="p">,</span>
+    <span class="n">canonical_level</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+    <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-6</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="k">return</span> <span class="n">LevelMapper</span><span class="p">(</span><span class="n">k_min</span><span class="p">,</span> <span class="n">k_max</span><span class="p">,</span> <span class="n">canonical_scale</span><span class="p">,</span> <span class="n">canonical_level</span><span class="p">,</span> <span class="n">eps</span><span class="p">)</span>
+
+
+<span class="k">class</span> <span class="nc">LevelMapper</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Determine which FPN level each RoI in a set of RoIs should map to based</span>
+<span class="sd">    on the heuristic in the FPN paper.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        k_min (int)</span>
+<span class="sd">        k_max (int)</span>
+<span class="sd">        canonical_scale (int)</span>
+<span class="sd">        canonical_level (int)</span>
+<span class="sd">        eps (float)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">k_min</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">k_max</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">canonical_scale</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">224</span><span class="p">,</span>
+        <span class="n">canonical_level</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-6</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">k_min</span> <span class="o">=</span> <span class="n">k_min</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">k_max</span> <span class="o">=</span> <span class="n">k_max</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">s0</span> <span class="o">=</span> <span class="n">canonical_scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">lvl0</span> <span class="o">=</span> <span class="n">canonical_level</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">eps</span> <span class="o">=</span> <span class="n">eps</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">boxlists</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            boxlists (list[BoxList])</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Compute level ids</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">box_area</span><span class="p">(</span><span class="n">boxlist</span><span class="p">)</span> <span class="k">for</span> <span class="n">boxlist</span> <span class="ow">in</span> <span class="n">boxlists</span><span class="p">]))</span>
+
+        <span class="c1"># Eqn.(1) in FPN paper</span>
+        <span class="n">target_lvls</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lvl0</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">log2</span><span class="p">(</span><span class="n">s</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">s0</span><span class="p">)</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">eps</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">s</span><span class="o">.</span><span class="n">dtype</span><span class="p">))</span>
+        <span class="n">target_lvls</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">target_lvls</span><span class="p">,</span> <span class="nb">min</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">k_min</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">k_max</span><span class="p">)</span>
+        <span class="k">return</span> <span class="p">(</span><span class="n">target_lvls</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">)</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">k_min</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_convert_to_roi_format</span><span class="p">(</span><span class="n">boxes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">concat_boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">device</span><span class="p">,</span> <span class="n">dtype</span> <span class="o">=</span> <span class="n">concat_boxes</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">concat_boxes</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">ids</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span>
+        <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">full_like</span><span class="p">(</span><span class="n">b</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">1</span><span class="p">],</span> <span class="n">i</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">layout</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">strided</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">b</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">boxes</span><span class="p">)],</span>
+        <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">rois</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">ids</span><span class="p">,</span> <span class="n">concat_boxes</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">rois</span>
+
+
+<span class="k">def</span> <span class="nf">_infer_scale</span><span class="p">(</span><span class="n">feature</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">original_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="nb">float</span><span class="p">:</span>
+    <span class="c1"># assumption: the scale is of the form 2 ** (-k), with k integer</span>
+    <span class="n">size</span> <span class="o">=</span> <span class="n">feature</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+    <span class="n">possible_scales</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">s1</span><span class="p">,</span> <span class="n">s2</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">original_size</span><span class="p">):</span>
+        <span class="n">approx_scale</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">s1</span><span class="p">)</span> <span class="o">/</span> <span class="nb">float</span><span class="p">(</span><span class="n">s2</span><span class="p">)</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">**</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">approx_scale</span><span class="p">)</span><span class="o">.</span><span class="n">log2</span><span class="p">()</span><span class="o">.</span><span class="n">round</span><span class="p">())</span>
+        <span class="n">possible_scales</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">scale</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">possible_scales</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span>
+<span class="k">def</span> <span class="nf">_setup_scales</span><span class="p">(</span>
+    <span class="n">features</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">image_shapes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]],</span> <span class="n">canonical_scale</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">canonical_level</span><span class="p">:</span> <span class="nb">int</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">LevelMapper</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">image_shapes</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;images list should not be empty&quot;</span><span class="p">)</span>
+    <span class="n">max_x</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">max_y</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="k">for</span> <span class="n">shape</span> <span class="ow">in</span> <span class="n">image_shapes</span><span class="p">:</span>
+        <span class="n">max_x</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">max_x</span><span class="p">)</span>
+        <span class="n">max_y</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">max_y</span><span class="p">)</span>
+    <span class="n">original_input_shape</span> <span class="o">=</span> <span class="p">(</span><span class="n">max_x</span><span class="p">,</span> <span class="n">max_y</span><span class="p">)</span>
+
+    <span class="n">scales</span> <span class="o">=</span> <span class="p">[</span><span class="n">_infer_scale</span><span class="p">(</span><span class="n">feat</span><span class="p">,</span> <span class="n">original_input_shape</span><span class="p">)</span> <span class="k">for</span> <span class="n">feat</span> <span class="ow">in</span> <span class="n">features</span><span class="p">]</span>
+    <span class="c1"># get the levels in the feature map by leveraging the fact that the network always</span>
+    <span class="c1"># downsamples by a factor of 2 at each level.</span>
+    <span class="n">lvl_min</span> <span class="o">=</span> <span class="o">-</span><span class="n">torch</span><span class="o">.</span><span class="n">log2</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">scales</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+    <span class="n">lvl_max</span> <span class="o">=</span> <span class="o">-</span><span class="n">torch</span><span class="o">.</span><span class="n">log2</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">scales</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+    <span class="n">map_levels</span> <span class="o">=</span> <span class="n">initLevelMapper</span><span class="p">(</span>
+        <span class="nb">int</span><span class="p">(</span><span class="n">lvl_min</span><span class="p">),</span>
+        <span class="nb">int</span><span class="p">(</span><span class="n">lvl_max</span><span class="p">),</span>
+        <span class="n">canonical_scale</span><span class="o">=</span><span class="n">canonical_scale</span><span class="p">,</span>
+        <span class="n">canonical_level</span><span class="o">=</span><span class="n">canonical_level</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">scales</span><span class="p">,</span> <span class="n">map_levels</span>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span>
+<span class="k">def</span> <span class="nf">_filter_input</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span> <span class="n">featmap_names</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]:</span>
+    <span class="n">x_filtered</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">x</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+        <span class="k">if</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">featmap_names</span><span class="p">:</span>
+            <span class="n">x_filtered</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">v</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">x_filtered</span>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span>
+<span class="k">def</span> <span class="nf">_multiscale_roi_align</span><span class="p">(</span>
+    <span class="n">x_filtered</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">sampling_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">scales</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+    <span class="n">mapper</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">LevelMapper</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Args:</span>
+<span class="sd">        x_filtered (List[Tensor]): List of input tensors.</span>
+<span class="sd">        boxes (List[Tensor[N, 4]]): boxes to be used to perform the pooling operation, in</span>
+<span class="sd">            (x1, y1, x2, y2) format and in the image reference size, not the feature map</span>
+<span class="sd">            reference. The coordinate must satisfy ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">        output_size (Union[List[Tuple[int, int]], List[int]]): size of the output</span>
+<span class="sd">        sampling_ratio (int): sampling ratio for ROIAlign</span>
+<span class="sd">        scales (Optional[List[float]]): If None, scales will be automatically inferred. Default value is None.</span>
+<span class="sd">        mapper (Optional[LevelMapper]): If none, mapper will be automatically inferred. Default value is None.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        result (Tensor)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">scales</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">mapper</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;scales and mapper should not be None&quot;</span><span class="p">)</span>
+
+    <span class="n">num_levels</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">x_filtered</span><span class="p">)</span>
+    <span class="n">rois</span> <span class="o">=</span> <span class="n">_convert_to_roi_format</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">num_levels</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">roi_align</span><span class="p">(</span>
+            <span class="n">x_filtered</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+            <span class="n">rois</span><span class="p">,</span>
+            <span class="n">output_size</span><span class="o">=</span><span class="n">output_size</span><span class="p">,</span>
+            <span class="n">spatial_scale</span><span class="o">=</span><span class="n">scales</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+            <span class="n">sampling_ratio</span><span class="o">=</span><span class="n">sampling_ratio</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="n">levels</span> <span class="o">=</span> <span class="n">mapper</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+
+    <span class="n">num_rois</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">rois</span><span class="p">)</span>
+    <span class="n">num_channels</span> <span class="o">=</span> <span class="n">x_filtered</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+
+    <span class="n">dtype</span><span class="p">,</span> <span class="n">device</span> <span class="o">=</span> <span class="n">x_filtered</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">x_filtered</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">device</span>
+    <span class="n">result</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span>
+        <span class="p">(</span>
+            <span class="n">num_rois</span><span class="p">,</span>
+            <span class="n">num_channels</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="o">+</span> <span class="n">output_size</span><span class="p">,</span>
+        <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span>
+        <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="n">tracing_results</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">level</span><span class="p">,</span> <span class="p">(</span><span class="n">per_level_feature</span><span class="p">,</span> <span class="n">scale</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="n">x_filtered</span><span class="p">,</span> <span class="n">scales</span><span class="p">)):</span>
+        <span class="n">idx_in_level</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">levels</span> <span class="o">==</span> <span class="n">level</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">rois_per_level</span> <span class="o">=</span> <span class="n">rois</span><span class="p">[</span><span class="n">idx_in_level</span><span class="p">]</span>
+
+        <span class="n">result_idx_in_level</span> <span class="o">=</span> <span class="n">roi_align</span><span class="p">(</span>
+            <span class="n">per_level_feature</span><span class="p">,</span>
+            <span class="n">rois_per_level</span><span class="p">,</span>
+            <span class="n">output_size</span><span class="o">=</span><span class="n">output_size</span><span class="p">,</span>
+            <span class="n">spatial_scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+            <span class="n">sampling_ratio</span><span class="o">=</span><span class="n">sampling_ratio</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">_is_tracing</span><span class="p">():</span>
+            <span class="n">tracing_results</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">result_idx_in_level</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># result and result_idx_in_level&#39;s dtypes are based on dtypes of different</span>
+            <span class="c1"># elements in x_filtered.  x_filtered contains tensors output by different</span>
+            <span class="c1"># layers.  When autocast is active, it may choose different dtypes for</span>
+            <span class="c1"># different layers&#39; outputs.  Therefore, we defensively match result&#39;s dtype</span>
+            <span class="c1"># before copying elements from result_idx_in_level in the following op.</span>
+            <span class="c1"># We need to cast manually (can&#39;t rely on autocast to cast for us) because</span>
+            <span class="c1"># the op acts on result in-place, and autocast only affects out-of-place ops.</span>
+            <span class="n">result</span><span class="p">[</span><span class="n">idx_in_level</span><span class="p">]</span> <span class="o">=</span> <span class="n">result_idx_in_level</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">result</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">_is_tracing</span><span class="p">():</span>
+        <span class="n">result</span> <span class="o">=</span> <span class="n">_onnx_merge_levels</span><span class="p">(</span><span class="n">levels</span><span class="p">,</span> <span class="n">tracing_results</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">result</span>
+
+
+<div class="viewcode-block" id="MultiScaleRoIAlign"><a class="viewcode-back" href="../../../generated/torchvision.ops.MultiScaleRoIAlign.html#torchvision.ops.MultiScaleRoIAlign">[docs]</a><span class="k">class</span> <span class="nc">MultiScaleRoIAlign</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Multi-scale RoIAlign pooling, which is useful for detection with or without FPN.</span>
+
+<span class="sd">    It infers the scale of the pooling via the heuristics specified in eq. 1</span>
+<span class="sd">    of the `Feature Pyramid Network paper &lt;https://arxiv.org/abs/1612.03144&gt;`_.</span>
+<span class="sd">    They keyword-only parameters ``canonical_scale`` and ``canonical_level``</span>
+<span class="sd">    correspond respectively to ``224`` and ``k0=4`` in eq. 1, and</span>
+<span class="sd">    have the following meaning: ``canonical_level`` is the target level of the pyramid from</span>
+<span class="sd">    which to pool a region of interest with ``w x h = canonical_scale x canonical_scale``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        featmap_names (List[str]): the names of the feature maps that will be used</span>
+<span class="sd">            for the pooling.</span>
+<span class="sd">        output_size (List[Tuple[int, int]] or List[int]): output size for the pooled region</span>
+<span class="sd">        sampling_ratio (int): sampling ratio for ROIAlign</span>
+<span class="sd">        canonical_scale (int, optional): canonical_scale for LevelMapper</span>
+<span class="sd">        canonical_level (int, optional): canonical_level for LevelMapper</span>
+
+<span class="sd">    Examples::</span>
+
+<span class="sd">        &gt;&gt;&gt; m = torchvision.ops.MultiScaleRoIAlign([&#39;feat1&#39;, &#39;feat3&#39;], 3, 2)</span>
+<span class="sd">        &gt;&gt;&gt; i = OrderedDict()</span>
+<span class="sd">        &gt;&gt;&gt; i[&#39;feat1&#39;] = torch.rand(1, 5, 64, 64)</span>
+<span class="sd">        &gt;&gt;&gt; i[&#39;feat2&#39;] = torch.rand(1, 5, 32, 32)  # this feature won&#39;t be used in the pooling</span>
+<span class="sd">        &gt;&gt;&gt; i[&#39;feat3&#39;] = torch.rand(1, 5, 16, 16)</span>
+<span class="sd">        &gt;&gt;&gt; # create some random bounding boxes</span>
+<span class="sd">        &gt;&gt;&gt; boxes = torch.rand(6, 4) * 256; boxes[:, 2:] += boxes[:, :2]</span>
+<span class="sd">        &gt;&gt;&gt; # original image size, before computing the feature maps</span>
+<span class="sd">        &gt;&gt;&gt; image_sizes = [(512, 512)]</span>
+<span class="sd">        &gt;&gt;&gt; output = m(i, [boxes], image_sizes)</span>
+<span class="sd">        &gt;&gt;&gt; print(output.shape)</span>
+<span class="sd">        &gt;&gt;&gt; torch.Size([6, 5, 3, 3])</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="vm">__annotations__</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;scales&quot;</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span> <span class="s2">&quot;map_levels&quot;</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">LevelMapper</span><span class="p">]}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">featmap_names</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+        <span class="n">output_size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span>
+        <span class="n">sampling_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="o">*</span><span class="p">,</span>
+        <span class="n">canonical_scale</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">224</span><span class="p">,</span>
+        <span class="n">canonical_level</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output_size</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">output_size</span> <span class="o">=</span> <span class="p">(</span><span class="n">output_size</span><span class="p">,</span> <span class="n">output_size</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">featmap_names</span> <span class="o">=</span> <span class="n">featmap_names</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span> <span class="o">=</span> <span class="n">sampling_ratio</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scales</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">map_levels</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">canonical_scale</span> <span class="o">=</span> <span class="n">canonical_scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">canonical_level</span> <span class="o">=</span> <span class="n">canonical_level</span>
+
+<div class="viewcode-block" id="MultiScaleRoIAlign.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.MultiScaleRoIAlign.html#torchvision.ops.MultiScaleRoIAlign.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">x</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">boxes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">image_shapes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            x (OrderedDict[Tensor]): feature maps for each level. They are assumed to have</span>
+<span class="sd">                all the same number of channels, but they can have different sizes.</span>
+<span class="sd">            boxes (List[Tensor[N, 4]]): boxes to be used to perform the pooling operation, in</span>
+<span class="sd">                (x1, y1, x2, y2) format and in the image reference size, not the feature map</span>
+<span class="sd">                reference. The coordinate must satisfy ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">            image_shapes (List[Tuple[height, width]]): the sizes of each image before they</span>
+<span class="sd">                have been fed to a CNN to obtain feature maps. This allows us to infer the</span>
+<span class="sd">                scale factor for each one of the levels to be pooled.</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            result (Tensor)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">x_filtered</span> <span class="o">=</span> <span class="n">_filter_input</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">featmap_names</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">scales</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">map_levels</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">scales</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">map_levels</span> <span class="o">=</span> <span class="n">_setup_scales</span><span class="p">(</span>
+                <span class="n">x_filtered</span><span class="p">,</span> <span class="n">image_shapes</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">canonical_scale</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">canonical_level</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">_multiscale_roi_align</span><span class="p">(</span>
+            <span class="n">x_filtered</span><span class="p">,</span>
+            <span class="n">boxes</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">scales</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">map_levels</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(featmap_names=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">featmap_names</span><span class="si">}</span><span class="s2">, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;output_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="si">}</span><span class="s2">, sampling_ratio=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/ps_roi_align.html b/0.16/_modules/torchvision/ops/ps_roi_align.html
new file mode 100644
index 00000000000..ebd85e4513e
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/ps_roi_align.html
@@ -0,0 +1,759 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.ps_roi_align &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.ps_roi_align</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.ps_roi_align</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.nn.modules.utils</span> <span class="kn">import</span> <span class="n">_pair</span>
+<span class="kn">from</span> <span class="nn">torchvision.extension</span> <span class="kn">import</span> <span class="n">_assert_has_ops</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">check_roi_boxes_shape</span><span class="p">,</span> <span class="n">convert_boxes_to_roi_format</span>
+
+
+<div class="viewcode-block" id="ps_roi_align"><a class="viewcode-back" href="../../../generated/torchvision.ops.ps_roi_align.html#torchvision.ops.ps_roi_align">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span>
+<span class="k">def</span> <span class="nf">ps_roi_align</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">output_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+    <span class="n">sampling_ratio</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Performs Position-Sensitive Region of Interest (RoI) Align operator</span>
+<span class="sd">    mentioned in Light-Head R-CNN.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, C, H, W]): The input tensor, i.e. a batch with ``N`` elements. Each element</span>
+<span class="sd">            contains ``C`` feature maps of dimensions ``H x W``.</span>
+<span class="sd">        boxes (Tensor[K, 5] or List[Tensor[L, 4]]): the box coordinates in (x1, y1, x2, y2)</span>
+<span class="sd">            format where the regions will be taken from.</span>
+<span class="sd">            The coordinate must satisfy ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">            If a single Tensor is passed, then the first column should</span>
+<span class="sd">            contain the index of the corresponding element in the batch, i.e. a number in ``[0, N - 1]``.</span>
+<span class="sd">            If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i</span>
+<span class="sd">            in the batch.</span>
+<span class="sd">        output_size (int or Tuple[int, int]): the size of the output (in bins or pixels) after the pooling</span>
+<span class="sd">            is performed, as (height, width).</span>
+<span class="sd">        spatial_scale (float): a scaling factor that maps the box coordinates to</span>
+<span class="sd">            the input coordinates. For example, if your boxes are defined on the scale</span>
+<span class="sd">            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of</span>
+<span class="sd">            the original image), you&#39;ll want to set this to 0.5. Default: 1.0</span>
+<span class="sd">        sampling_ratio (int): number of sampling points in the interpolation grid</span>
+<span class="sd">            used to compute the output value of each pooled output bin. If &gt; 0,</span>
+<span class="sd">            then exactly ``sampling_ratio x sampling_ratio`` sampling points per bin are used. If</span>
+<span class="sd">            &lt;= 0, then an adaptive number of grid points are used (computed as</span>
+<span class="sd">            ``ceil(roi_width / output_width)``, and likewise for height). Default: -1</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[K, C / (output_size[0] * output_size[1]), output_size[0], output_size[1]]: The pooled RoIs</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">ps_roi_align</span><span class="p">)</span>
+    <span class="n">_assert_has_ops</span><span class="p">()</span>
+    <span class="n">check_roi_boxes_shape</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+    <span class="n">rois</span> <span class="o">=</span> <span class="n">boxes</span>
+    <span class="n">output_size</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">rois</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">rois</span> <span class="o">=</span> <span class="n">convert_boxes_to_roi_format</span><span class="p">(</span><span class="n">rois</span><span class="p">)</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">ps_roi_align</span><span class="p">(</span>
+        <span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="n">spatial_scale</span><span class="p">,</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">sampling_ratio</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="PSRoIAlign"><a class="viewcode-back" href="../../../generated/torchvision.ops.PSRoIAlign.html#torchvision.ops.PSRoIAlign">[docs]</a><span class="k">class</span> <span class="nc">PSRoIAlign</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`ps_roi_align`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">output_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">sampling_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span> <span class="o">=</span> <span class="n">output_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span> <span class="o">=</span> <span class="n">spatial_scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span> <span class="o">=</span> <span class="n">sampling_ratio</span>
+
+<div class="viewcode-block" id="PSRoIAlign.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.PSRoIAlign.html#torchvision.ops.PSRoIAlign.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">rois</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">ps_roi_align</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;output_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, spatial_scale=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, sampling_ratio=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/ps_roi_pool.html b/0.16/_modules/torchvision/ops/ps_roi_pool.html
new file mode 100644
index 00000000000..cba744ee505
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/ps_roi_pool.html
@@ -0,0 +1,739 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.ps_roi_pool &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.ps_roi_pool</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.ps_roi_pool</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.nn.modules.utils</span> <span class="kn">import</span> <span class="n">_pair</span>
+<span class="kn">from</span> <span class="nn">torchvision.extension</span> <span class="kn">import</span> <span class="n">_assert_has_ops</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">check_roi_boxes_shape</span><span class="p">,</span> <span class="n">convert_boxes_to_roi_format</span>
+
+
+<div class="viewcode-block" id="ps_roi_pool"><a class="viewcode-back" href="../../../generated/torchvision.ops.ps_roi_pool.html#torchvision.ops.ps_roi_pool">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span>
+<span class="k">def</span> <span class="nf">ps_roi_pool</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">output_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Performs Position-Sensitive Region of Interest (RoI) Pool operator</span>
+<span class="sd">    described in R-FCN</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, C, H, W]): The input tensor, i.e. a batch with ``N`` elements. Each element</span>
+<span class="sd">            contains ``C`` feature maps of dimensions ``H x W``.</span>
+<span class="sd">        boxes (Tensor[K, 5] or List[Tensor[L, 4]]): the box coordinates in (x1, y1, x2, y2)</span>
+<span class="sd">            format where the regions will be taken from.</span>
+<span class="sd">            The coordinate must satisfy ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">            If a single Tensor is passed, then the first column should</span>
+<span class="sd">            contain the index of the corresponding element in the batch, i.e. a number in ``[0, N - 1]``.</span>
+<span class="sd">            If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i</span>
+<span class="sd">            in the batch.</span>
+<span class="sd">        output_size (int or Tuple[int, int]): the size of the output (in bins or pixels) after the pooling</span>
+<span class="sd">            is performed, as (height, width).</span>
+<span class="sd">        spatial_scale (float): a scaling factor that maps the box coordinates to</span>
+<span class="sd">            the input coordinates. For example, if your boxes are defined on the scale</span>
+<span class="sd">            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of</span>
+<span class="sd">            the original image), you&#39;ll want to set this to 0.5. Default: 1.0</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[K, C / (output_size[0] * output_size[1]), output_size[0], output_size[1]]: The pooled RoIs.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">ps_roi_pool</span><span class="p">)</span>
+    <span class="n">_assert_has_ops</span><span class="p">()</span>
+    <span class="n">check_roi_boxes_shape</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+    <span class="n">rois</span> <span class="o">=</span> <span class="n">boxes</span>
+    <span class="n">output_size</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">rois</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">rois</span> <span class="o">=</span> <span class="n">convert_boxes_to_roi_format</span><span class="p">(</span><span class="n">rois</span><span class="p">)</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">ps_roi_pool</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="n">spatial_scale</span><span class="p">,</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="PSRoIPool"><a class="viewcode-back" href="../../../generated/torchvision.ops.PSRoIPool.html#torchvision.ops.PSRoIPool">[docs]</a><span class="k">class</span> <span class="nc">PSRoIPool</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`ps_roi_pool`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span> <span class="o">=</span> <span class="n">output_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span> <span class="o">=</span> <span class="n">spatial_scale</span>
+
+<div class="viewcode-block" id="PSRoIPool.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.PSRoIPool.html#torchvision.ops.PSRoIPool.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">rois</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">ps_roi_pool</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(output_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="si">}</span><span class="s2">, spatial_scale=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/roi_align.html b/0.16/_modules/torchvision/ops/roi_align.html
new file mode 100644
index 00000000000..900ed45524f
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/roi_align.html
@@ -0,0 +1,943 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.roi_align &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.roi_align</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.roi_align</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch._dynamo</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.jit.annotations</span> <span class="kn">import</span> <span class="n">BroadcastingList2</span>
+<span class="kn">from</span> <span class="nn">torch.nn.modules.utils</span> <span class="kn">import</span> <span class="n">_pair</span>
+<span class="kn">from</span> <span class="nn">torchvision.extension</span> <span class="kn">import</span> <span class="n">_assert_has_ops</span><span class="p">,</span> <span class="n">_has_ops</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">check_roi_boxes_shape</span><span class="p">,</span> <span class="n">convert_boxes_to_roi_format</span>
+
+
+<span class="c1"># NB: all inputs are tensors</span>
+<span class="k">def</span> <span class="nf">_bilinear_interpolate</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">,</span>  <span class="c1"># [N, C, H, W]</span>
+    <span class="n">roi_batch_ind</span><span class="p">,</span>  <span class="c1"># [K]</span>
+    <span class="n">y</span><span class="p">,</span>  <span class="c1"># [K, PH, IY]</span>
+    <span class="n">x</span><span class="p">,</span>  <span class="c1"># [K, PW, IX]</span>
+    <span class="n">ymask</span><span class="p">,</span>  <span class="c1"># [K, IY]</span>
+    <span class="n">xmask</span><span class="p">,</span>  <span class="c1"># [K, IX]</span>
+<span class="p">):</span>
+    <span class="n">_</span><span class="p">,</span> <span class="n">channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+
+    <span class="c1"># deal with inverse element out of feature map boundary</span>
+    <span class="n">y</span> <span class="o">=</span> <span class="n">y</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">y_low</span> <span class="o">=</span> <span class="n">y</span><span class="o">.</span><span class="n">int</span><span class="p">()</span>
+    <span class="n">x_low</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">int</span><span class="p">()</span>
+    <span class="n">y_high</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">y_low</span> <span class="o">&gt;=</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">y_low</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="n">y_low</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">y_low</span> <span class="o">&gt;=</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">y_low</span><span class="p">)</span>
+    <span class="n">y</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">y_low</span> <span class="o">&gt;=</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">y</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span> <span class="n">y</span><span class="p">)</span>
+
+    <span class="n">x_high</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">x_low</span> <span class="o">&gt;=</span> <span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">x_low</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="n">x_low</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">x_low</span> <span class="o">&gt;=</span> <span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">x_low</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">x_low</span> <span class="o">&gt;=</span> <span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">x</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span> <span class="n">x</span><span class="p">)</span>
+
+    <span class="n">ly</span> <span class="o">=</span> <span class="n">y</span> <span class="o">-</span> <span class="n">y_low</span>
+    <span class="n">lx</span> <span class="o">=</span> <span class="n">x</span> <span class="o">-</span> <span class="n">x_low</span>
+    <span class="n">hy</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">ly</span>
+    <span class="n">hx</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">lx</span>
+
+    <span class="c1"># do bilinear interpolation, but respect the masking!</span>
+    <span class="c1"># TODO: It&#39;s possible the masking here is unnecessary if y and</span>
+    <span class="c1"># x were clamped appropriately; hard to tell</span>
+    <span class="k">def</span> <span class="nf">masked_index</span><span class="p">(</span>
+        <span class="n">y</span><span class="p">,</span>  <span class="c1"># [K, PH, IY]</span>
+        <span class="n">x</span><span class="p">,</span>  <span class="c1"># [K, PW, IX]</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="n">ymask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">assert</span> <span class="n">xmask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+            <span class="n">y</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">ymask</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:],</span> <span class="n">y</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">xmask</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:],</span> <span class="n">x</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">input</span><span class="p">[</span>
+            <span class="n">roi_batch_ind</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">],</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">channels</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">],</span>
+            <span class="n">y</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">],</span>  <span class="c1"># prev [K, PH, IY]</span>
+            <span class="n">x</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:],</span>  <span class="c1"># prev [K, PW, IX]</span>
+        <span class="p">]</span>  <span class="c1"># [K, C, PH, PW, IY, IX]</span>
+
+    <span class="n">v1</span> <span class="o">=</span> <span class="n">masked_index</span><span class="p">(</span><span class="n">y_low</span><span class="p">,</span> <span class="n">x_low</span><span class="p">)</span>
+    <span class="n">v2</span> <span class="o">=</span> <span class="n">masked_index</span><span class="p">(</span><span class="n">y_low</span><span class="p">,</span> <span class="n">x_high</span><span class="p">)</span>
+    <span class="n">v3</span> <span class="o">=</span> <span class="n">masked_index</span><span class="p">(</span><span class="n">y_high</span><span class="p">,</span> <span class="n">x_low</span><span class="p">)</span>
+    <span class="n">v4</span> <span class="o">=</span> <span class="n">masked_index</span><span class="p">(</span><span class="n">y_high</span><span class="p">,</span> <span class="n">x_high</span><span class="p">)</span>
+
+    <span class="c1"># all ws preemptively [K, C, PH, PW, IY, IX]</span>
+    <span class="k">def</span> <span class="nf">outer_prod</span><span class="p">(</span><span class="n">y</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">y</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">x</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span>
+
+    <span class="n">w1</span> <span class="o">=</span> <span class="n">outer_prod</span><span class="p">(</span><span class="n">hy</span><span class="p">,</span> <span class="n">hx</span><span class="p">)</span>
+    <span class="n">w2</span> <span class="o">=</span> <span class="n">outer_prod</span><span class="p">(</span><span class="n">hy</span><span class="p">,</span> <span class="n">lx</span><span class="p">)</span>
+    <span class="n">w3</span> <span class="o">=</span> <span class="n">outer_prod</span><span class="p">(</span><span class="n">ly</span><span class="p">,</span> <span class="n">hx</span><span class="p">)</span>
+    <span class="n">w4</span> <span class="o">=</span> <span class="n">outer_prod</span><span class="p">(</span><span class="n">ly</span><span class="p">,</span> <span class="n">lx</span><span class="p">)</span>
+
+    <span class="n">val</span> <span class="o">=</span> <span class="n">w1</span> <span class="o">*</span> <span class="n">v1</span> <span class="o">+</span> <span class="n">w2</span> <span class="o">*</span> <span class="n">v2</span> <span class="o">+</span> <span class="n">w3</span> <span class="o">*</span> <span class="n">v3</span> <span class="o">+</span> <span class="n">w4</span> <span class="o">*</span> <span class="n">v4</span>
+    <span class="k">return</span> <span class="n">val</span>
+
+
+<span class="c1"># TODO: this doesn&#39;t actually cache</span>
+<span class="c1"># TODO: main library should make this easier to do</span>
+<span class="k">def</span> <span class="nf">maybe_cast</span><span class="p">(</span><span class="n">tensor</span><span class="p">):</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_autocast_enabled</span><span class="p">()</span> <span class="ow">and</span> <span class="n">tensor</span><span class="o">.</span><span class="n">is_cuda</span> <span class="ow">and</span> <span class="n">tensor</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">double</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">tensor</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">tensor</span>
+
+
+<span class="c1"># This is a slow but pure Python and differentiable implementation of</span>
+<span class="c1"># roi_align.  It potentially is a good basis for Inductor compilation</span>
+<span class="c1"># (but I have not benchmarked it) but today it is solely used for the</span>
+<span class="c1"># fact that its backwards can be implemented deterministically,</span>
+<span class="c1"># which is needed for the PT2 benchmark suite.</span>
+<span class="c1">#</span>
+<span class="c1"># It is transcribed directly off of the roi_align CUDA kernel, see</span>
+<span class="c1"># https://dev-discuss.pytorch.org/t/a-pure-python-implementation-of-roi-align-that-looks-just-like-its-cuda-kernel/1266</span>
+<span class="nd">@torch</span><span class="o">.</span><span class="n">_dynamo</span><span class="o">.</span><span class="n">allow_in_graph</span>
+<span class="k">def</span> <span class="nf">_roi_align</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="n">spatial_scale</span><span class="p">,</span> <span class="n">pooled_height</span><span class="p">,</span> <span class="n">pooled_width</span><span class="p">,</span> <span class="n">sampling_ratio</span><span class="p">,</span> <span class="n">aligned</span><span class="p">):</span>
+    <span class="n">orig_dtype</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">dtype</span>
+
+    <span class="nb">input</span> <span class="o">=</span> <span class="n">maybe_cast</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+    <span class="n">rois</span> <span class="o">=</span> <span class="n">maybe_cast</span><span class="p">(</span><span class="n">rois</span><span class="p">)</span>
+
+    <span class="n">_</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+
+    <span class="n">ph</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">pooled_height</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>  <span class="c1"># [PH]</span>
+    <span class="n">pw</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">pooled_width</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>  <span class="c1"># [PW]</span>
+
+    <span class="c1"># input: [N, C, H, W]</span>
+    <span class="c1"># rois: [K, 5]</span>
+
+    <span class="n">roi_batch_ind</span> <span class="o">=</span> <span class="n">rois</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">int</span><span class="p">()</span>  <span class="c1"># [K]</span>
+    <span class="n">offset</span> <span class="o">=</span> <span class="mf">0.5</span> <span class="k">if</span> <span class="n">aligned</span> <span class="k">else</span> <span class="mf">0.0</span>
+    <span class="n">roi_start_w</span> <span class="o">=</span> <span class="n">rois</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">spatial_scale</span> <span class="o">-</span> <span class="n">offset</span>  <span class="c1"># [K]</span>
+    <span class="n">roi_start_h</span> <span class="o">=</span> <span class="n">rois</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="n">spatial_scale</span> <span class="o">-</span> <span class="n">offset</span>  <span class="c1"># [K]</span>
+    <span class="n">roi_end_w</span> <span class="o">=</span> <span class="n">rois</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="n">spatial_scale</span> <span class="o">-</span> <span class="n">offset</span>  <span class="c1"># [K]</span>
+    <span class="n">roi_end_h</span> <span class="o">=</span> <span class="n">rois</span><span class="p">[:,</span> <span class="mi">4</span><span class="p">]</span> <span class="o">*</span> <span class="n">spatial_scale</span> <span class="o">-</span> <span class="n">offset</span>  <span class="c1"># [K]</span>
+
+    <span class="n">roi_width</span> <span class="o">=</span> <span class="n">roi_end_w</span> <span class="o">-</span> <span class="n">roi_start_w</span>  <span class="c1"># [K]</span>
+    <span class="n">roi_height</span> <span class="o">=</span> <span class="n">roi_end_h</span> <span class="o">-</span> <span class="n">roi_start_h</span>  <span class="c1"># [K]</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">aligned</span><span class="p">:</span>
+        <span class="n">roi_width</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">roi_width</span><span class="p">,</span> <span class="nb">min</span><span class="o">=</span><span class="mf">1.0</span><span class="p">)</span>  <span class="c1"># [K]</span>
+        <span class="n">roi_height</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">roi_height</span><span class="p">,</span> <span class="nb">min</span><span class="o">=</span><span class="mf">1.0</span><span class="p">)</span>  <span class="c1"># [K]</span>
+
+    <span class="n">bin_size_h</span> <span class="o">=</span> <span class="n">roi_height</span> <span class="o">/</span> <span class="n">pooled_height</span>  <span class="c1"># [K]</span>
+    <span class="n">bin_size_w</span> <span class="o">=</span> <span class="n">roi_width</span> <span class="o">/</span> <span class="n">pooled_width</span>  <span class="c1"># [K]</span>
+
+    <span class="n">exact_sampling</span> <span class="o">=</span> <span class="n">sampling_ratio</span> <span class="o">&gt;</span> <span class="mi">0</span>
+
+    <span class="n">roi_bin_grid_h</span> <span class="o">=</span> <span class="n">sampling_ratio</span> <span class="k">if</span> <span class="n">exact_sampling</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">roi_height</span> <span class="o">/</span> <span class="n">pooled_height</span><span class="p">)</span>  <span class="c1"># scalar or [K]</span>
+    <span class="n">roi_bin_grid_w</span> <span class="o">=</span> <span class="n">sampling_ratio</span> <span class="k">if</span> <span class="n">exact_sampling</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">roi_width</span> <span class="o">/</span> <span class="n">pooled_width</span><span class="p">)</span>  <span class="c1"># scalar or [K]</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    iy, ix = dims(2)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="n">exact_sampling</span><span class="p">:</span>
+        <span class="n">count</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">roi_bin_grid_h</span> <span class="o">*</span> <span class="n">roi_bin_grid_w</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>  <span class="c1"># scalar</span>
+        <span class="n">iy</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">roi_bin_grid_h</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>  <span class="c1"># [IY]</span>
+        <span class="n">ix</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">roi_bin_grid_w</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>  <span class="c1"># [IX]</span>
+        <span class="n">ymask</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">xmask</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">count</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">roi_bin_grid_h</span> <span class="o">*</span> <span class="n">roi_bin_grid_w</span><span class="p">,</span> <span class="nb">min</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>  <span class="c1"># [K]</span>
+        <span class="c1"># When doing adaptive sampling, the number of samples we need to do</span>
+        <span class="c1"># is data-dependent based on how big the ROIs are.  This is a bit</span>
+        <span class="c1"># awkward because first-class dims can&#39;t actually handle this.</span>
+        <span class="c1"># So instead, we inefficiently suppose that we needed to sample ALL</span>
+        <span class="c1"># the points and mask out things that turned out to be unnecessary</span>
+        <span class="n">iy</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>  <span class="c1"># [IY]</span>
+        <span class="n">ix</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>  <span class="c1"># [IX]</span>
+        <span class="n">ymask</span> <span class="o">=</span> <span class="n">iy</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">roi_bin_grid_h</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span>  <span class="c1"># [K, IY]</span>
+        <span class="n">xmask</span> <span class="o">=</span> <span class="n">ix</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">roi_bin_grid_w</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span>  <span class="c1"># [K, IX]</span>
+
+    <span class="k">def</span> <span class="nf">from_K</span><span class="p">(</span><span class="n">t</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">t</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+
+    <span class="n">y</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">from_K</span><span class="p">(</span><span class="n">roi_start_h</span><span class="p">)</span>
+        <span class="o">+</span> <span class="n">ph</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">from_K</span><span class="p">(</span><span class="n">bin_size_h</span><span class="p">)</span>
+        <span class="o">+</span> <span class="p">(</span><span class="n">iy</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">+</span> <span class="mf">0.5</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="o">*</span> <span class="n">from_K</span><span class="p">(</span><span class="n">bin_size_h</span> <span class="o">/</span> <span class="n">roi_bin_grid_h</span><span class="p">)</span>
+    <span class="p">)</span>  <span class="c1"># [K, PH, IY]</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">from_K</span><span class="p">(</span><span class="n">roi_start_w</span><span class="p">)</span>
+        <span class="o">+</span> <span class="n">pw</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">from_K</span><span class="p">(</span><span class="n">bin_size_w</span><span class="p">)</span>
+        <span class="o">+</span> <span class="p">(</span><span class="n">ix</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">+</span> <span class="mf">0.5</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="o">*</span> <span class="n">from_K</span><span class="p">(</span><span class="n">bin_size_w</span> <span class="o">/</span> <span class="n">roi_bin_grid_w</span><span class="p">)</span>
+    <span class="p">)</span>  <span class="c1"># [K, PW, IX]</span>
+    <span class="n">val</span> <span class="o">=</span> <span class="n">_bilinear_interpolate</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">roi_batch_ind</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">ymask</span><span class="p">,</span> <span class="n">xmask</span><span class="p">)</span>  <span class="c1"># [K, C, PH, PW, IY, IX]</span>
+
+    <span class="c1"># Mask out samples that weren&#39;t actually adaptively needed</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">exact_sampling</span><span class="p">:</span>
+        <span class="n">val</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">ymask</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">],</span> <span class="n">val</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="n">val</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">xmask</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:],</span> <span class="n">val</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">val</span><span class="o">.</span><span class="n">sum</span><span class="p">((</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">2</span><span class="p">))</span>  <span class="c1"># remove IY, IX ~&gt; [K, C, PH, PW]</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">count</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">output</span> <span class="o">/=</span> <span class="n">count</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">/=</span> <span class="n">count</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">orig_dtype</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<div class="viewcode-block" id="roi_align"><a class="viewcode-back" href="../../../generated/torchvision.ops.roi_align.html#torchvision.ops.roi_align">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span>
+<span class="k">def</span> <span class="nf">roi_align</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]],</span>
+    <span class="n">output_size</span><span class="p">:</span> <span class="n">BroadcastingList2</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+    <span class="n">sampling_ratio</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span>
+    <span class="n">aligned</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Performs Region of Interest (RoI) Align operator with average pooling, as described in Mask R-CNN.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, C, H, W]): The input tensor, i.e. a batch with ``N`` elements. Each element</span>
+<span class="sd">            contains ``C`` feature maps of dimensions ``H x W``.</span>
+<span class="sd">            If the tensor is quantized, we expect a batch size of ``N == 1``.</span>
+<span class="sd">        boxes (Tensor[K, 5] or List[Tensor[L, 4]]): the box coordinates in (x1, y1, x2, y2)</span>
+<span class="sd">            format where the regions will be taken from.</span>
+<span class="sd">            The coordinate must satisfy ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">            If a single Tensor is passed, then the first column should</span>
+<span class="sd">            contain the index of the corresponding element in the batch, i.e. a number in ``[0, N - 1]``.</span>
+<span class="sd">            If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i</span>
+<span class="sd">            in the batch.</span>
+<span class="sd">        output_size (int or Tuple[int, int]): the size of the output (in bins or pixels) after the pooling</span>
+<span class="sd">            is performed, as (height, width).</span>
+<span class="sd">        spatial_scale (float): a scaling factor that maps the box coordinates to</span>
+<span class="sd">            the input coordinates. For example, if your boxes are defined on the scale</span>
+<span class="sd">            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of</span>
+<span class="sd">            the original image), you&#39;ll want to set this to 0.5. Default: 1.0</span>
+<span class="sd">        sampling_ratio (int): number of sampling points in the interpolation grid</span>
+<span class="sd">            used to compute the output value of each pooled output bin. If &gt; 0,</span>
+<span class="sd">            then exactly ``sampling_ratio x sampling_ratio`` sampling points per bin are used. If</span>
+<span class="sd">            &lt;= 0, then an adaptive number of grid points are used (computed as</span>
+<span class="sd">            ``ceil(roi_width / output_width)``, and likewise for height). Default: -1</span>
+<span class="sd">        aligned (bool): If False, use the legacy implementation.</span>
+<span class="sd">            If True, pixel shift the box coordinates it by -0.5 for a better alignment with the two</span>
+<span class="sd">            neighboring pixel indices. This version is used in Detectron2</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[K, C, output_size[0], output_size[1]]: The pooled RoIs.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">roi_align</span><span class="p">)</span>
+    <span class="n">check_roi_boxes_shape</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+    <span class="n">rois</span> <span class="o">=</span> <span class="n">boxes</span>
+    <span class="n">output_size</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">rois</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">rois</span> <span class="o">=</span> <span class="n">convert_boxes_to_roi_format</span><span class="p">(</span><span class="n">rois</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">_has_ops</span><span class="p">()</span> <span class="ow">or</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">are_deterministic_algorithms_enabled</span><span class="p">()</span> <span class="ow">and</span> <span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">is_cuda</span> <span class="ow">or</span> <span class="nb">input</span><span class="o">.</span><span class="n">is_mps</span><span class="p">)):</span>
+            <span class="k">return</span> <span class="n">_roi_align</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="n">spatial_scale</span><span class="p">,</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">sampling_ratio</span><span class="p">,</span> <span class="n">aligned</span><span class="p">)</span>
+    <span class="n">_assert_has_ops</span><span class="p">()</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">roi_align</span><span class="p">(</span>
+        <span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="n">spatial_scale</span><span class="p">,</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">sampling_ratio</span><span class="p">,</span> <span class="n">aligned</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RoIAlign"><a class="viewcode-back" href="../../../generated/torchvision.ops.RoIAlign.html#torchvision.ops.RoIAlign">[docs]</a><span class="k">class</span> <span class="nc">RoIAlign</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`roi_align`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">output_size</span><span class="p">:</span> <span class="n">BroadcastingList2</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+        <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">sampling_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">aligned</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span> <span class="o">=</span> <span class="n">output_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span> <span class="o">=</span> <span class="n">spatial_scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span> <span class="o">=</span> <span class="n">sampling_ratio</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">aligned</span> <span class="o">=</span> <span class="n">aligned</span>
+
+<div class="viewcode-block" id="RoIAlign.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.RoIAlign.html#torchvision.ops.RoIAlign.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">rois</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">roi_align</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligned</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;output_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, spatial_scale=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, sampling_ratio=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sampling_ratio</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, aligned=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">aligned</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/roi_pool.html b/0.16/_modules/torchvision/ops/roi_pool.html
new file mode 100644
index 00000000000..23862fc70b8
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/roi_pool.html
@@ -0,0 +1,741 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.roi_pool &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.roi_pool</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.roi_pool</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+<span class="kn">from</span> <span class="nn">torch.jit.annotations</span> <span class="kn">import</span> <span class="n">BroadcastingList2</span>
+<span class="kn">from</span> <span class="nn">torch.nn.modules.utils</span> <span class="kn">import</span> <span class="n">_pair</span>
+<span class="kn">from</span> <span class="nn">torchvision.extension</span> <span class="kn">import</span> <span class="n">_assert_has_ops</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">check_roi_boxes_shape</span><span class="p">,</span> <span class="n">convert_boxes_to_roi_format</span>
+
+
+<div class="viewcode-block" id="roi_pool"><a class="viewcode-back" href="../../../generated/torchvision.ops.roi_pool.html#torchvision.ops.roi_pool">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span>
+<span class="k">def</span> <span class="nf">roi_pool</span><span class="p">(</span>
+    <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]],</span>
+    <span class="n">output_size</span><span class="p">:</span> <span class="n">BroadcastingList2</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Performs Region of Interest (RoI) Pool operator described in Fast R-CNN</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, C, H, W]): The input tensor, i.e. a batch with ``N`` elements. Each element</span>
+<span class="sd">            contains ``C`` feature maps of dimensions ``H x W``.</span>
+<span class="sd">        boxes (Tensor[K, 5] or List[Tensor[L, 4]]): the box coordinates in (x1, y1, x2, y2)</span>
+<span class="sd">            format where the regions will be taken from.</span>
+<span class="sd">            The coordinate must satisfy ``0 &lt;= x1 &lt; x2`` and ``0 &lt;= y1 &lt; y2``.</span>
+<span class="sd">            If a single Tensor is passed, then the first column should</span>
+<span class="sd">            contain the index of the corresponding element in the batch, i.e. a number in ``[0, N - 1]``.</span>
+<span class="sd">            If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i</span>
+<span class="sd">            in the batch.</span>
+<span class="sd">        output_size (int or Tuple[int, int]): the size of the output after the cropping</span>
+<span class="sd">            is performed, as (height, width)</span>
+<span class="sd">        spatial_scale (float): a scaling factor that maps the box coordinates to</span>
+<span class="sd">            the input coordinates. For example, if your boxes are defined on the scale</span>
+<span class="sd">            of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of</span>
+<span class="sd">            the original image), you&#39;ll want to set this to 0.5. Default: 1.0</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[K, C, output_size[0], output_size[1]]: The pooled RoIs.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">roi_pool</span><span class="p">)</span>
+    <span class="n">_assert_has_ops</span><span class="p">()</span>
+    <span class="n">check_roi_boxes_shape</span><span class="p">(</span><span class="n">boxes</span><span class="p">)</span>
+    <span class="n">rois</span> <span class="o">=</span> <span class="n">boxes</span>
+    <span class="n">output_size</span> <span class="o">=</span> <span class="n">_pair</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">rois</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">rois</span> <span class="o">=</span> <span class="n">convert_boxes_to_roi_format</span><span class="p">(</span><span class="n">rois</span><span class="p">)</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">torchvision</span><span class="o">.</span><span class="n">roi_pool</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="n">spatial_scale</span><span class="p">,</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="RoIPool"><a class="viewcode-back" href="../../../generated/torchvision.ops.RoIPool.html#torchvision.ops.RoIPool">[docs]</a><span class="k">class</span> <span class="nc">RoIPool</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`roi_pool`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">BroadcastingList2</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">spatial_scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span> <span class="o">=</span> <span class="n">output_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span> <span class="o">=</span> <span class="n">spatial_scale</span>
+
+<div class="viewcode-block" id="RoIPool.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.RoIPool.html#torchvision.ops.RoIPool.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">rois</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">roi_pool</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">rois</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(output_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">output_size</span><span class="si">}</span><span class="s2">, spatial_scale=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">spatial_scale</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/ops/stochastic_depth.html b/0.16/_modules/torchvision/ops/stochastic_depth.html
new file mode 100644
index 00000000000..65745e2ecaf
--- /dev/null
+++ b/0.16/_modules/torchvision/ops/stochastic_depth.html
@@ -0,0 +1,735 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.ops.stochastic_depth &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.ops.stochastic_depth</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.ops.stochastic_depth</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.fx</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span><span class="p">,</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+
+<div class="viewcode-block" id="stochastic_depth"><a class="viewcode-back" href="../../../generated/torchvision.ops.stochastic_depth.html#torchvision.ops.stochastic_depth">[docs]</a><span class="k">def</span> <span class="nf">stochastic_depth</span><span class="p">(</span><span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">training</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Implements the Stochastic Depth from `&quot;Deep Networks with Stochastic Depth&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1603.09382&gt;`_ used for randomly dropping residual</span>
+<span class="sd">    branches of residual architectures.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        input (Tensor[N, ...]): The input tensor or arbitrary dimensions with the first one</span>
+<span class="sd">                    being its batch i.e. a batch with ``N`` rows.</span>
+<span class="sd">        p (float): probability of the input to be zeroed.</span>
+<span class="sd">        mode (str): ``&quot;batch&quot;`` or ``&quot;row&quot;``.</span>
+<span class="sd">                    ``&quot;batch&quot;`` randomly zeroes the entire input, ``&quot;row&quot;`` zeroes</span>
+<span class="sd">                    randomly selected rows from the batch.</span>
+<span class="sd">        training: apply stochastic depth if is ``True``. Default: ``True``</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor[N, ...]: The randomly zeroed tensor.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">stochastic_depth</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">p</span> <span class="o">&lt;</span> <span class="mf">0.0</span> <span class="ow">or</span> <span class="n">p</span> <span class="o">&gt;</span> <span class="mf">1.0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;drop probability has to be between 0 and 1, but got </span><span class="si">{</span><span class="n">p</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">mode</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;batch&quot;</span><span class="p">,</span> <span class="s2">&quot;row&quot;</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;mode has to be either &#39;batch&#39; or &#39;row&#39;, but got </span><span class="si">{</span><span class="n">mode</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">training</span> <span class="ow">or</span> <span class="n">p</span> <span class="o">==</span> <span class="mf">0.0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">input</span>
+
+    <span class="n">survival_rate</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">p</span>
+    <span class="k">if</span> <span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;row&quot;</span><span class="p">:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">[</span><span class="nb">input</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span> <span class="o">+</span> <span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">ndim</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="nb">input</span><span class="o">.</span><span class="n">ndim</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="nb">input</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">noise</span> <span class="o">=</span> <span class="n">noise</span><span class="o">.</span><span class="n">bernoulli_</span><span class="p">(</span><span class="n">survival_rate</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">survival_rate</span> <span class="o">&gt;</span> <span class="mf">0.0</span><span class="p">:</span>
+        <span class="n">noise</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">survival_rate</span><span class="p">)</span>
+    <span class="k">return</span> <span class="nb">input</span> <span class="o">*</span> <span class="n">noise</span></div>
+
+
+<span class="n">torch</span><span class="o">.</span><span class="n">fx</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="s2">&quot;stochastic_depth&quot;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="StochasticDepth"><a class="viewcode-back" href="../../../generated/torchvision.ops.StochasticDepth.html#torchvision.ops.StochasticDepth">[docs]</a><span class="k">class</span> <span class="nc">StochasticDepth</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    See :func:`stochastic_depth`.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">=</span> <span class="n">mode</span>
+
+<div class="viewcode-block" id="StochasticDepth.forward"><a class="viewcode-back" href="../../../generated/torchvision.ops.StochasticDepth.html#torchvision.ops.StochasticDepth.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">stochastic_depth</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">, mode=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/autoaugment.html b/0.16/_modules/torchvision/transforms/autoaugment.html
new file mode 100644
index 00000000000..3fceb59a8ba
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/autoaugment.html
@@ -0,0 +1,1284 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.autoaugment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.autoaugment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.autoaugment</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">enum</span> <span class="kn">import</span> <span class="n">Enum</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">InterpolationMode</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;AutoAugmentPolicy&quot;</span><span class="p">,</span> <span class="s2">&quot;AutoAugment&quot;</span><span class="p">,</span> <span class="s2">&quot;RandAugment&quot;</span><span class="p">,</span> <span class="s2">&quot;TrivialAugmentWide&quot;</span><span class="p">,</span> <span class="s2">&quot;AugMix&quot;</span><span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_apply_op</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">op_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span><span class="p">,</span> <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span>
+<span class="p">):</span>
+    <span class="k">if</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span>
+        <span class="c1"># magnitude should be arctan(magnitude)</span>
+        <span class="c1"># official autoaug: (1, level, 0, 0, 1, 0)</span>
+        <span class="c1"># https://github.com/tensorflow/models/blob/dd02069717128186b88afa8d857ce57d17957f03/research/autoaugment/augmentation_transforms.py#L290</span>
+        <span class="c1"># compared to</span>
+        <span class="c1"># torchvision:      (1, tan(level), 0, 0, 1, 0)</span>
+        <span class="c1"># https://github.com/pytorch/vision/blob/0c2373d0bba3499e95776e7936e207d8a1676e65/torchvision/transforms/functional.py#L976</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+            <span class="n">img</span><span class="p">,</span>
+            <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+            <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+            <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+            <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="n">math</span><span class="o">.</span><span class="n">degrees</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">magnitude</span><span class="p">)),</span> <span class="mf">0.0</span><span class="p">],</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+            <span class="n">center</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+        <span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span>
+        <span class="c1"># magnitude should be arctan(magnitude)</span>
+        <span class="c1"># See above</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+            <span class="n">img</span><span class="p">,</span>
+            <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+            <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+            <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+            <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">math</span><span class="o">.</span><span class="n">degrees</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">magnitude</span><span class="p">))],</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+            <span class="n">center</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+        <span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+            <span class="n">img</span><span class="p">,</span>
+            <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+            <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">magnitude</span><span class="p">),</span> <span class="mi">0</span><span class="p">],</span>
+            <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+            <span class="n">img</span><span class="p">,</span>
+            <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+            <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">magnitude</span><span class="p">)],</span>
+            <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">rotate</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Color&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_sharpness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">posterize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">magnitude</span><span class="p">))</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">solarize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">autocontrast</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">equalize</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Invert&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">invert</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">op_name</span> <span class="o">==</span> <span class="s2">&quot;Identity&quot;</span><span class="p">:</span>
+        <span class="k">pass</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The provided operator </span><span class="si">{</span><span class="n">op_name</span><span class="si">}</span><span class="s2"> is not recognized.&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">img</span>
+
+
+<div class="viewcode-block" id="AutoAugmentPolicy"><a class="viewcode-back" href="../../../generated/torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy">[docs]</a><span class="k">class</span> <span class="nc">AutoAugmentPolicy</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;AutoAugment policies learned on different datasets.</span>
+<span class="sd">    Available policies are IMAGENET, CIFAR10 and SVHN.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">IMAGENET</span> <span class="o">=</span> <span class="s2">&quot;imagenet&quot;</span>
+    <span class="n">CIFAR10</span> <span class="o">=</span> <span class="s2">&quot;cifar10&quot;</span>
+    <span class="n">SVHN</span> <span class="o">=</span> <span class="s2">&quot;svhn&quot;</span></div>
+
+
+<span class="c1"># FIXME: Eliminate copy-pasted code for fill standardization and _augmentation_space() by moving stuff on a base class</span>
+<div class="viewcode-block" id="AutoAugment"><a class="viewcode-back" href="../../../generated/torchvision.transforms.AutoAugment.html#torchvision.transforms.AutoAugment">[docs]</a><span class="k">class</span> <span class="nc">AutoAugment</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;AutoAugment data augmentation method based on</span>
+<span class="sd">    `&quot;AutoAugment: Learning Augmentation Strategies from Data&quot; &lt;https://arxiv.org/pdf/1805.09501.pdf&gt;`_.</span>
+<span class="sd">    If the image is torch Tensor, it should be of type torch.uint8, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        policy (AutoAugmentPolicy): Desired policy enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.autoaugment.AutoAugmentPolicy`. Default is ``AutoAugmentPolicy.IMAGENET``.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">policy</span><span class="p">:</span> <span class="n">AutoAugmentPolicy</span> <span class="o">=</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">IMAGENET</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">policy</span> <span class="o">=</span> <span class="n">policy</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">policies</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_policies</span><span class="p">(</span><span class="n">policy</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_policies</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">policy</span><span class="p">:</span> <span class="n">AutoAugmentPolicy</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]]:</span>
+        <span class="k">if</span> <span class="n">policy</span> <span class="o">==</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">IMAGENET</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">[</span>
+                <span class="p">((</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mi">2</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+            <span class="p">]</span>
+        <span class="k">elif</span> <span class="n">policy</span> <span class="o">==</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">CIFAR10</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">[</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateX&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateX&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+            <span class="p">]</span>
+        <span class="k">elif</span> <span class="n">policy</span> <span class="o">==</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">SVHN</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">[</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mi">2</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+            <span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The provided policy </span><span class="si">{</span><span class="n">policy</span><span class="si">}</span><span class="s2"> is not recognized.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_augmentation_space</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">image_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]:</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="c1"># op_name: (magnitudes, signed)</span>
+            <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">image_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">image_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">30.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">8</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">4</span><span class="p">))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">255.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Invert&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="p">}</span>
+
+<div class="viewcode-block" id="AutoAugment.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.AutoAugment.html#torchvision.transforms.AutoAugment.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="n">transform_num</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get parameters for autoaugment transformation</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            params required by the autoaugment transformation</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">policy_id</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">transform_num</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+        <span class="n">probs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">((</span><span class="mi">2</span><span class="p">,))</span>
+        <span class="n">signs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="k">return</span> <span class="n">policy_id</span><span class="p">,</span> <span class="n">probs</span><span class="p">,</span> <span class="n">signs</span></div>
+
+<div class="viewcode-block" id="AutoAugment.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.AutoAugment.html#torchvision.transforms.AutoAugment.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be transformed.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: AutoAugmented image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span>
+        <span class="n">channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span> <span class="o">*</span> <span class="n">channels</span>
+            <span class="k">elif</span> <span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+
+        <span class="n">transform_id</span><span class="p">,</span> <span class="n">probs</span><span class="p">,</span> <span class="n">signs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">policies</span><span class="p">))</span>
+
+        <span class="n">op_meta</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_augmentation_space</span><span class="p">(</span><span class="mi">10</span><span class="p">,</span> <span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">))</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">op_name</span><span class="p">,</span> <span class="n">p</span><span class="p">,</span> <span class="n">magnitude_id</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">policies</span><span class="p">[</span><span class="n">transform_id</span><span class="p">]):</span>
+            <span class="k">if</span> <span class="n">probs</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">p</span><span class="p">:</span>
+                <span class="n">magnitudes</span><span class="p">,</span> <span class="n">signed</span> <span class="o">=</span> <span class="n">op_meta</span><span class="p">[</span><span class="n">op_name</span><span class="p">]</span>
+                <span class="n">magnitude</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="n">magnitude_id</span><span class="p">]</span><span class="o">.</span><span class="n">item</span><span class="p">())</span> <span class="k">if</span> <span class="n">magnitude_id</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="mf">0.0</span>
+                <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">signs</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mf">1.0</span>
+                <span class="n">img</span> <span class="o">=</span> <span class="n">_apply_op</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">op_name</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(policy=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">policy</span><span class="si">}</span><span class="s2">, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandAugment"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandAugment.html#torchvision.transforms.RandAugment">[docs]</a><span class="k">class</span> <span class="nc">RandAugment</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;RandAugment data augmentation method based on</span>
+<span class="sd">    `&quot;RandAugment: Practical automated data augmentation with a reduced search space&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1909.13719&gt;`_.</span>
+<span class="sd">    If the image is torch Tensor, it should be of type torch.uint8, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_ops (int): Number of augmentation transformations to apply sequentially.</span>
+<span class="sd">        magnitude (int): Magnitude for all the transformations.</span>
+<span class="sd">        num_magnitude_bins (int): The number of different magnitude values.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_ops</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">2</span><span class="p">,</span>
+        <span class="n">magnitude</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">9</span><span class="p">,</span>
+        <span class="n">num_magnitude_bins</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">31</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_ops</span> <span class="o">=</span> <span class="n">num_ops</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">magnitude</span> <span class="o">=</span> <span class="n">magnitude</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span> <span class="o">=</span> <span class="n">num_magnitude_bins</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+
+    <span class="k">def</span> <span class="nf">_augmentation_space</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">image_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]:</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="c1"># op_name: (magnitudes, signed)</span>
+            <span class="s2">&quot;Identity&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">image_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">image_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">30.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">8</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">4</span><span class="p">))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">255.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="p">}</span>
+
+<div class="viewcode-block" id="RandAugment.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandAugment.html#torchvision.transforms.RandAugment.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be transformed.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span>
+        <span class="n">channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span> <span class="o">*</span> <span class="n">channels</span>
+            <span class="k">elif</span> <span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+
+        <span class="n">op_meta</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_augmentation_space</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span><span class="p">,</span> <span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">))</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_ops</span><span class="p">):</span>
+            <span class="n">op_index</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">op_meta</span><span class="p">),</span> <span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+            <span class="n">op_name</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">op_meta</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="n">op_index</span><span class="p">]</span>
+            <span class="n">magnitudes</span><span class="p">,</span> <span class="n">signed</span> <span class="o">=</span> <span class="n">op_meta</span><span class="p">[</span><span class="n">op_name</span><span class="p">]</span>
+            <span class="n">magnitude</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">magnitude</span><span class="p">]</span><span class="o">.</span><span class="n">item</span><span class="p">())</span> <span class="k">if</span> <span class="n">magnitudes</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="mf">0.0</span>
+            <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,)):</span>
+                <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mf">1.0</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">_apply_op</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">op_name</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;num_ops=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">num_ops</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, magnitude=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">magnitude</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, num_magnitude_bins=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, interpolation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<div class="viewcode-block" id="TrivialAugmentWide"><a class="viewcode-back" href="../../../generated/torchvision.transforms.TrivialAugmentWide.html#torchvision.transforms.TrivialAugmentWide">[docs]</a><span class="k">class</span> <span class="nc">TrivialAugmentWide</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Dataset-independent data-augmentation with TrivialAugment Wide, as described in</span>
+<span class="sd">    `&quot;TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation&quot; &lt;https://arxiv.org/abs/2103.10158&gt;`_.</span>
+<span class="sd">    If the image is torch Tensor, it should be of type torch.uint8, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_magnitude_bins (int): The number of different magnitude values.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_magnitude_bins</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">31</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span> <span class="o">=</span> <span class="n">num_magnitude_bins</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+
+    <span class="k">def</span> <span class="nf">_augmentation_space</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]:</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="c1"># op_name: (magnitudes, signed)</span>
+            <span class="s2">&quot;Identity&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">32.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">32.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">135.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">8</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">6</span><span class="p">))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">255.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="p">}</span>
+
+<div class="viewcode-block" id="TrivialAugmentWide.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.TrivialAugmentWide.html#torchvision.transforms.TrivialAugmentWide.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be transformed.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span>
+        <span class="n">channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span> <span class="o">*</span> <span class="n">channels</span>
+            <span class="k">elif</span> <span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+
+        <span class="n">op_meta</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_augmentation_space</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span><span class="p">)</span>
+        <span class="n">op_index</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">op_meta</span><span class="p">),</span> <span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+        <span class="n">op_name</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">op_meta</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="n">op_index</span><span class="p">]</span>
+        <span class="n">magnitudes</span><span class="p">,</span> <span class="n">signed</span> <span class="o">=</span> <span class="n">op_meta</span><span class="p">[</span><span class="n">op_name</span><span class="p">]</span>
+        <span class="n">magnitude</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">),</span> <span class="p">(</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)]</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+            <span class="k">if</span> <span class="n">magnitudes</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">0</span>
+            <span class="k">else</span> <span class="mf">0.0</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,)):</span>
+            <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mf">1.0</span>
+
+        <span class="k">return</span> <span class="n">_apply_op</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">op_name</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;num_magnitude_bins=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, interpolation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<div class="viewcode-block" id="AugMix"><a class="viewcode-back" href="../../../generated/torchvision.transforms.AugMix.html#torchvision.transforms.AugMix">[docs]</a><span class="k">class</span> <span class="nc">AugMix</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;AugMix data augmentation method based on</span>
+<span class="sd">    `&quot;AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty&quot; &lt;https://arxiv.org/abs/1912.02781&gt;`_.</span>
+<span class="sd">    If the image is torch Tensor, it should be of type torch.uint8, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        severity (int): The severity of base augmentation operators. Default is ``3``.</span>
+<span class="sd">        mixture_width (int): The number of augmentation chains. Default is ``3``.</span>
+<span class="sd">        chain_depth (int): The depth of augmentation chains. A negative value denotes stochastic depth sampled from the interval [1, 3].</span>
+<span class="sd">            Default is ``-1``.</span>
+<span class="sd">        alpha (float): The hyperparameter for the probability distributions. Default is ``1.0``.</span>
+<span class="sd">        all_ops (bool): Use all operations (including brightness, contrast, color and sharpness). Default is ``True``.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">severity</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">mixture_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">chain_depth</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">all_ops</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span> <span class="o">=</span> <span class="mi">10</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">severity</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The severity must be between [1, </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span><span class="si">}</span><span class="s2">]. Got </span><span class="si">{</span><span class="n">severity</span><span class="si">}</span><span class="s2"> instead.&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">severity</span> <span class="o">=</span> <span class="n">severity</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mixture_width</span> <span class="o">=</span> <span class="n">mixture_width</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">chain_depth</span> <span class="o">=</span> <span class="n">chain_depth</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span> <span class="o">=</span> <span class="n">alpha</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">all_ops</span> <span class="o">=</span> <span class="n">all_ops</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+
+    <span class="k">def</span> <span class="nf">_augmentation_space</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">image_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="c1"># op_name: (magnitudes, signed)</span>
+            <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">image_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">/</span> <span class="mf">3.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">image_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">/</span> <span class="mf">3.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">30.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+            <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">4</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">4</span><span class="p">))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">255.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+            <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="p">}</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">all_ops</span><span class="p">:</span>
+            <span class="n">s</span><span class="o">.</span><span class="n">update</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+                    <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+                    <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+                    <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">_pil_to_tensor</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">_tensor_to_pil</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">to_pil_image</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_sample_dirichlet</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># Must be on a separate method so that we can overwrite it in tests.</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">_sample_dirichlet</span><span class="p">(</span><span class="n">params</span><span class="p">)</span>
+
+<div class="viewcode-block" id="AugMix.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.AugMix.html#torchvision.transforms.AugMix.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">orig_img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be transformed.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span>
+        <span class="n">channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">orig_img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">orig_img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">orig_img</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span> <span class="o">*</span> <span class="n">channels</span>
+            <span class="k">elif</span> <span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pil_to_tensor</span><span class="p">(</span><span class="n">orig_img</span><span class="p">)</span>
+
+        <span class="n">op_meta</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_augmentation_space</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span><span class="p">,</span> <span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">))</span>
+
+        <span class="n">orig_dims</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
+        <span class="n">batch</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">view</span><span class="p">([</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="mi">4</span> <span class="o">-</span> <span class="n">img</span><span class="o">.</span><span class="n">ndim</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="o">+</span> <span class="n">orig_dims</span><span class="p">)</span>
+        <span class="n">batch_dims</span> <span class="o">=</span> <span class="p">[</span><span class="n">batch</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)]</span> <span class="o">+</span> <span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">ndim</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="c1"># Sample the beta weights for combining the original and augmented image. To get Beta, we use a Dirichlet</span>
+        <span class="c1"># with 2 parameters. The 1st column stores the weights of the original and the 2nd the ones of augmented image.</span>
+        <span class="n">m</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_dirichlet</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">batch</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="c1"># Sample the mixing weights and combine them with the ones sampled from Beta for the augmented images.</span>
+        <span class="n">combined_weights</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_dirichlet</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">mixture_width</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">batch</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="p">)</span> <span class="o">*</span> <span class="n">m</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">view</span><span class="p">([</span><span class="n">batch_dims</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">])</span>
+
+        <span class="n">mix</span> <span class="o">=</span> <span class="n">m</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">)</span> <span class="o">*</span> <span class="n">batch</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mixture_width</span><span class="p">):</span>
+            <span class="n">aug</span> <span class="o">=</span> <span class="n">batch</span>
+            <span class="n">depth</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">chain_depth</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">chain_depth</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">low</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">high</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+            <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">depth</span><span class="p">):</span>
+                <span class="n">op_index</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">op_meta</span><span class="p">),</span> <span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+                <span class="n">op_name</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">op_meta</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="n">op_index</span><span class="p">]</span>
+                <span class="n">magnitudes</span><span class="p">,</span> <span class="n">signed</span> <span class="o">=</span> <span class="n">op_meta</span><span class="p">[</span><span class="n">op_name</span><span class="p">]</span>
+                <span class="n">magnitude</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">severity</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)]</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+                    <span class="k">if</span> <span class="n">magnitudes</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">0</span>
+                    <span class="k">else</span> <span class="mf">0.0</span>
+                <span class="p">)</span>
+                <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="p">(</span><span class="mi">1</span><span class="p">,)):</span>
+                    <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mf">1.0</span>
+                <span class="n">aug</span> <span class="o">=</span> <span class="n">_apply_op</span><span class="p">(</span><span class="n">aug</span><span class="p">,</span> <span class="n">op_name</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+            <span class="n">mix</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">combined_weights</span><span class="p">[:,</span> <span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">)</span> <span class="o">*</span> <span class="n">aug</span><span class="p">)</span>
+        <span class="n">mix</span> <span class="o">=</span> <span class="n">mix</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">orig_dims</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">img</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">orig_img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_tensor_to_pil</span><span class="p">(</span><span class="n">mix</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">mix</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;severity=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">severity</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, mixture_width=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mixture_width</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, chain_depth=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">chain_depth</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, alpha=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, all_ops=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">all_ops</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, interpolation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/functional.html b/0.16/_modules/torchvision/transforms/functional.html
new file mode 100644
index 00000000000..1ff0cf9e05e
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/functional.html
@@ -0,0 +1,2286 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.functional &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.functional</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.functional</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">numbers</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">enum</span> <span class="kn">import</span> <span class="n">Enum</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">import</span> <span class="nn">accimage</span>
+<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+    <span class="n">accimage</span> <span class="o">=</span> <span class="kc">None</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">_functional_pil</span> <span class="k">as</span> <span class="n">F_pil</span><span class="p">,</span> <span class="n">_functional_tensor</span> <span class="k">as</span> <span class="n">F_t</span>
+
+
+<span class="k">class</span> <span class="nc">InterpolationMode</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Interpolation modes</span>
+<span class="sd">    Available interpolation methods are ``nearest``, ``nearest-exact``, ``bilinear``, ``bicubic``, ``box``, ``hamming``,</span>
+<span class="sd">    and ``lanczos``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">NEAREST</span> <span class="o">=</span> <span class="s2">&quot;nearest&quot;</span>
+    <span class="n">NEAREST_EXACT</span> <span class="o">=</span> <span class="s2">&quot;nearest-exact&quot;</span>
+    <span class="n">BILINEAR</span> <span class="o">=</span> <span class="s2">&quot;bilinear&quot;</span>
+    <span class="n">BICUBIC</span> <span class="o">=</span> <span class="s2">&quot;bicubic&quot;</span>
+    <span class="c1"># For PIL compatibility</span>
+    <span class="n">BOX</span> <span class="o">=</span> <span class="s2">&quot;box&quot;</span>
+    <span class="n">HAMMING</span> <span class="o">=</span> <span class="s2">&quot;hamming&quot;</span>
+    <span class="n">LANCZOS</span> <span class="o">=</span> <span class="s2">&quot;lanczos&quot;</span>
+
+
+<span class="c1"># TODO: Once torchscript supports Enums with staticmethod</span>
+<span class="c1"># this can be put into InterpolationMode as staticmethod</span>
+<span class="k">def</span> <span class="nf">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">i</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">InterpolationMode</span><span class="p">:</span>
+    <span class="n">inverse_modes_mapping</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="mi">0</span><span class="p">:</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="mi">2</span><span class="p">:</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="mi">3</span><span class="p">:</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+        <span class="mi">4</span><span class="p">:</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BOX</span><span class="p">,</span>
+        <span class="mi">5</span><span class="p">:</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">HAMMING</span><span class="p">,</span>
+        <span class="mi">1</span><span class="p">:</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">LANCZOS</span><span class="p">,</span>
+    <span class="p">}</span>
+    <span class="k">return</span> <span class="n">inverse_modes_mapping</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+
+
+<span class="n">pil_modes_mapping</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">:</span> <span class="mi">2</span><span class="p">,</span>
+    <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">:</span> <span class="mi">3</span><span class="p">,</span>
+    <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST_EXACT</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BOX</span><span class="p">:</span> <span class="mi">4</span><span class="p">,</span>
+    <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">HAMMING</span><span class="p">:</span> <span class="mi">5</span><span class="p">,</span>
+    <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">LANCZOS</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
+<span class="p">}</span>
+
+<span class="n">_is_pil_image</span> <span class="o">=</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">_is_pil_image</span>
+
+
+<div class="viewcode-block" id="get_dimensions"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.get_dimensions.html#torchvision.transforms.functional.get_dimensions">[docs]</a><span class="k">def</span> <span class="nf">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Returns the dimensions of an image as [channels, height, width].</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): The image to be checked.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        List[int]: The image dimensions.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">get_dimensions</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="get_image_size"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.get_image_size.html#torchvision.transforms.functional.get_image_size">[docs]</a><span class="k">def</span> <span class="nf">get_image_size</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Returns the size of an image as [width, height].</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): The image to be checked.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        List[int]: The image size.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">get_image_size</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">get_image_size</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">get_image_size</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="get_image_num_channels"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.get_image_num_channels.html#torchvision.transforms.functional.get_image_num_channels">[docs]</a><span class="k">def</span> <span class="nf">get_image_num_channels</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Returns the number of channels of an image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): The image to be checked.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        int: The number of channels.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">get_image_num_channels</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">get_image_num_channels</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">get_image_num_channels</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+<span class="k">def</span> <span class="nf">_is_numpy</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+    <span class="k">return</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">)</span>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+<span class="k">def</span> <span class="nf">_is_numpy_image</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">img</span><span class="o">.</span><span class="n">ndim</span> <span class="ow">in</span> <span class="p">{</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">}</span>
+
+
+<div class="viewcode-block" id="to_tensor"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.to_tensor.html#torchvision.transforms.functional.to_tensor">[docs]</a><span class="k">def</span> <span class="nf">to_tensor</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a ``PIL Image`` or ``numpy.ndarray`` to tensor.</span>
+<span class="sd">    This function does not support torchscript.</span>
+
+<span class="sd">    See :class:`~torchvision.transforms.ToTensor` for more details.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        pic (PIL Image or numpy.ndarray): Image to be converted to tensor.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: Converted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">to_tensor</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">F_pil</span><span class="o">.</span><span class="n">_is_pil_image</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span> <span class="ow">or</span> <span class="n">_is_numpy</span><span class="p">(</span><span class="n">pic</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should be PIL Image or ndarray. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">_is_numpy</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">_is_numpy_image</span><span class="p">(</span><span class="n">pic</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should be 2/3 dimensional. Got </span><span class="si">{</span><span class="n">pic</span><span class="o">.</span><span class="n">ndim</span><span class="si">}</span><span class="s2"> dimensions.&quot;</span><span class="p">)</span>
+
+    <span class="n">default_float_dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">get_default_dtype</span><span class="p">()</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">):</span>
+        <span class="c1"># handle numpy array</span>
+        <span class="k">if</span> <span class="n">pic</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">pic</span> <span class="o">=</span> <span class="n">pic</span><span class="p">[:,</span> <span class="p">:,</span> <span class="kc">None</span><span class="p">]</span>
+
+        <span class="n">img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">pic</span><span class="o">.</span><span class="n">transpose</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)))</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+        <span class="c1"># backward compatibility</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ByteTensor</span><span class="p">):</span>
+            <span class="k">return</span> <span class="n">img</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">default_float_dtype</span><span class="p">)</span><span class="o">.</span><span class="n">div</span><span class="p">(</span><span class="mi">255</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">img</span>
+
+    <span class="k">if</span> <span class="n">accimage</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">accimage</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+        <span class="n">nppic</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">pic</span><span class="o">.</span><span class="n">channels</span><span class="p">,</span> <span class="n">pic</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="n">pic</span><span class="o">.</span><span class="n">width</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="n">pic</span><span class="o">.</span><span class="n">copyto</span><span class="p">(</span><span class="n">nppic</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">nppic</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">default_float_dtype</span><span class="p">)</span>
+
+    <span class="c1"># handle PIL Image</span>
+    <span class="n">mode_to_nptype</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;I&quot;</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span> <span class="s2">&quot;I;16&quot;</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">int16</span><span class="p">,</span> <span class="s2">&quot;F&quot;</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">}</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">mode_to_nptype</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">pic</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">),</span> <span class="n">copy</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+    <span class="k">if</span> <span class="n">pic</span><span class="o">.</span><span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;1&quot;</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="mi">255</span> <span class="o">*</span> <span class="n">img</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">pic</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">pic</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">get_image_num_channels</span><span class="p">(</span><span class="n">pic</span><span class="p">))</span>
+    <span class="c1"># put it from HWC to CHW format</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">permute</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ByteTensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">img</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">default_float_dtype</span><span class="p">)</span><span class="o">.</span><span class="n">div</span><span class="p">(</span><span class="mi">255</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+
+<div class="viewcode-block" id="pil_to_tensor"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.functional.pil_to_tensor.html#torchvision.transforms.functional.pil_to_tensor">[docs]</a><span class="k">def</span> <span class="nf">pil_to_tensor</span><span class="p">(</span><span class="n">pic</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a ``PIL Image`` to a tensor of the same type.</span>
+<span class="sd">    This function does not support torchscript.</span>
+
+<span class="sd">    See :class:`~torchvision.transforms.PILToTensor` for more details.</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        A deep copy of the underlying array is performed.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        pic (PIL Image): Image to be converted to tensor.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: Converted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">pil_to_tensor</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">_is_pil_image</span><span class="p">(</span><span class="n">pic</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should be PIL Image. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">accimage</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">accimage</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+        <span class="c1"># accimage format is always uint8 internally, so always return uint8 here</span>
+        <span class="n">nppic</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">pic</span><span class="o">.</span><span class="n">channels</span><span class="p">,</span> <span class="n">pic</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="n">pic</span><span class="o">.</span><span class="n">width</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+        <span class="n">pic</span><span class="o">.</span><span class="n">copyto</span><span class="p">(</span><span class="n">nppic</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">nppic</span><span class="p">)</span>
+
+    <span class="c1"># handle PIL Image</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">copy</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">pic</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">pic</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">get_image_num_channels</span><span class="p">(</span><span class="n">pic</span><span class="p">))</span>
+    <span class="c1"># put it from HWC to CHW format</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">permute</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">img</span></div>
+
+
+<div class="viewcode-block" id="convert_image_dtype"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.convert_image_dtype.html#torchvision.transforms.functional.convert_image_dtype">[docs]</a><span class="k">def</span> <span class="nf">convert_image_dtype</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a tensor image to the given ``dtype`` and scale the values accordingly</span>
+<span class="sd">    This function does not support PIL Image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        image (torch.Tensor): Image to be converted</span>
+<span class="sd">        dtype (torch.dtype): Desired data type of the output</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: Converted image</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        When converting from a smaller to a larger integer ``dtype`` the maximum values are **not** mapped exactly.</span>
+<span class="sd">        If converted back and forth, this mismatch has no effect.</span>
+
+<span class="sd">    Raises:</span>
+<span class="sd">        RuntimeError: When trying to cast :class:`torch.float32` to :class:`torch.int32` or :class:`torch.int64` as</span>
+<span class="sd">            well as for trying to cast :class:`torch.float64` to :class:`torch.int64`. These conversions might lead to</span>
+<span class="sd">            overflow errors since the floating point ``dtype`` cannot store consecutive integers over the whole range</span>
+<span class="sd">            of the integer ``dtype``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">convert_image_dtype</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Input img should be Tensor Image&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">convert_image_dtype</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">dtype</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="to_pil_image"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.functional.to_pil_image.html#torchvision.transforms.functional.to_pil_image">[docs]</a><span class="k">def</span> <span class="nf">to_pil_image</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a tensor or an ndarray to PIL Image. This function does not support torchscript.</span>
+
+<span class="sd">    See :class:`~torchvision.transforms.ToPILImage` for more details.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        pic (Tensor or numpy.ndarray): Image to be converted to PIL Image.</span>
+<span class="sd">        mode (`PIL.Image mode`_): color space and pixel depth of input data (optional).</span>
+
+<span class="sd">    .. _PIL.Image mode: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image: Image converted to PIL Image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">to_pil_image</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should be Tensor or ndarray. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">pic</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">{</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">}:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should be 2/3 dimensional. Got </span><span class="si">{</span><span class="n">pic</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span><span class="si">}</span><span class="s2"> dimensions.&quot;</span><span class="p">)</span>
+
+        <span class="k">elif</span> <span class="n">pic</span><span class="o">.</span><span class="n">ndimension</span><span class="p">()</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="c1"># if 2D image, add channel dimension (CHW)</span>
+            <span class="n">pic</span> <span class="o">=</span> <span class="n">pic</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+        <span class="c1"># check number of channels</span>
+        <span class="k">if</span> <span class="n">pic</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should not have &gt; 4 channels. Got </span><span class="si">{</span><span class="n">pic</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span><span class="si">}</span><span class="s2"> channels.&quot;</span><span class="p">)</span>
+
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">pic</span><span class="o">.</span><span class="n">ndim</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">{</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">}:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should be 2/3 dimensional. Got </span><span class="si">{</span><span class="n">pic</span><span class="o">.</span><span class="n">ndim</span><span class="si">}</span><span class="s2"> dimensions.&quot;</span><span class="p">)</span>
+
+        <span class="k">elif</span> <span class="n">pic</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="c1"># if 2D image, add channel dimension (HWC)</span>
+            <span class="n">pic</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">expand_dims</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="c1"># check number of channels</span>
+        <span class="k">if</span> <span class="n">pic</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;pic should not have &gt; 4 channels. Got </span><span class="si">{</span><span class="n">pic</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2"> channels.&quot;</span><span class="p">)</span>
+
+    <span class="n">npimg</span> <span class="o">=</span> <span class="n">pic</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">pic</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span> <span class="ow">and</span> <span class="n">mode</span> <span class="o">!=</span> <span class="s2">&quot;F&quot;</span><span class="p">:</span>
+            <span class="n">pic</span> <span class="o">=</span> <span class="n">pic</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="mi">255</span><span class="p">)</span><span class="o">.</span><span class="n">byte</span><span class="p">()</span>
+        <span class="n">npimg</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">pic</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">(),</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">npimg</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Input pic must be a torch.Tensor or NumPy ndarray, not {type(npimg)}&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">npimg</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">expected_mode</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">npimg</span> <span class="o">=</span> <span class="n">npimg</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+            <span class="n">expected_mode</span> <span class="o">=</span> <span class="s2">&quot;L&quot;</span>
+        <span class="k">elif</span> <span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">int16</span><span class="p">:</span>
+            <span class="n">expected_mode</span> <span class="o">=</span> <span class="s2">&quot;I;16&quot;</span>
+        <span class="k">elif</span> <span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">int32</span><span class="p">:</span>
+            <span class="n">expected_mode</span> <span class="o">=</span> <span class="s2">&quot;I&quot;</span>
+        <span class="k">elif</span> <span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">:</span>
+            <span class="n">expected_mode</span> <span class="o">=</span> <span class="s2">&quot;F&quot;</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">mode</span> <span class="o">!=</span> <span class="n">expected_mode</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Incorrect mode (</span><span class="si">{</span><span class="n">mode</span><span class="si">}</span><span class="s2">) supplied for input type </span><span class="si">{</span><span class="n">np</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">. Should be </span><span class="si">{</span><span class="n">expected_mode</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">mode</span> <span class="o">=</span> <span class="n">expected_mode</span>
+
+    <span class="k">elif</span> <span class="n">npimg</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="n">permitted_2_channel_modes</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;LA&quot;</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">mode</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">permitted_2_channel_modes</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Only modes </span><span class="si">{</span><span class="n">permitted_2_channel_modes</span><span class="si">}</span><span class="s2"> are supported for 2D inputs&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+            <span class="n">mode</span> <span class="o">=</span> <span class="s2">&quot;LA&quot;</span>
+
+    <span class="k">elif</span> <span class="n">npimg</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">==</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="n">permitted_4_channel_modes</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;RGBA&quot;</span><span class="p">,</span> <span class="s2">&quot;CMYK&quot;</span><span class="p">,</span> <span class="s2">&quot;RGBX&quot;</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">mode</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">permitted_4_channel_modes</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Only modes </span><span class="si">{</span><span class="n">permitted_4_channel_modes</span><span class="si">}</span><span class="s2"> are supported for 4D inputs&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+            <span class="n">mode</span> <span class="o">=</span> <span class="s2">&quot;RGBA&quot;</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">permitted_3_channel_modes</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;RGB&quot;</span><span class="p">,</span> <span class="s2">&quot;YCbCr&quot;</span><span class="p">,</span> <span class="s2">&quot;HSV&quot;</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">mode</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">permitted_3_channel_modes</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Only modes </span><span class="si">{</span><span class="n">permitted_3_channel_modes</span><span class="si">}</span><span class="s2"> are supported for 3D inputs&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">np</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+            <span class="n">mode</span> <span class="o">=</span> <span class="s2">&quot;RGB&quot;</span>
+
+    <span class="k">if</span> <span class="n">mode</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input type </span><span class="si">{</span><span class="n">npimg</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2"> is not supported&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">npimg</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">mode</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="normalize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.normalize.html#torchvision.transforms.functional.normalize">[docs]</a><span class="k">def</span> <span class="nf">normalize</span><span class="p">(</span><span class="n">tensor</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">mean</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">std</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Normalize a float tensor image with mean and standard deviation.</span>
+<span class="sd">    This transform does not support PIL Image.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        This transform acts out of place by default, i.e., it does not mutates the input tensor.</span>
+
+<span class="sd">    See :class:`~torchvision.transforms.Normalize` for more details.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        tensor (Tensor): Float tensor image of size (C, H, W) or (B, C, H, W) to be normalized.</span>
+<span class="sd">        mean (sequence): Sequence of means for each channel.</span>
+<span class="sd">        std (sequence): Sequence of standard deviations for each channel.</span>
+<span class="sd">        inplace(bool,optional): Bool to make this operation inplace.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor: Normalized Tensor image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">normalize</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;img should be Tensor Image. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">tensor</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">normalize</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="n">mean</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">std</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_compute_resized_output_size</span><span class="p">(</span>
+    <span class="n">image_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>  <span class="c1"># specified size only for the smallest edge</span>
+        <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">image_size</span>
+        <span class="n">short</span><span class="p">,</span> <span class="n">long</span> <span class="o">=</span> <span class="p">(</span><span class="n">w</span><span class="p">,</span> <span class="n">h</span><span class="p">)</span> <span class="k">if</span> <span class="n">w</span> <span class="o">&lt;=</span> <span class="n">h</span> <span class="k">else</span> <span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span>
+        <span class="n">requested_new_short</span> <span class="o">=</span> <span class="n">size</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="nb">int</span><span class="p">)</span> <span class="k">else</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="n">new_short</span><span class="p">,</span> <span class="n">new_long</span> <span class="o">=</span> <span class="n">requested_new_short</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">requested_new_short</span> <span class="o">*</span> <span class="n">long</span> <span class="o">/</span> <span class="n">short</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">max_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">max_size</span> <span class="o">&lt;=</span> <span class="n">requested_new_short</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;max_size = </span><span class="si">{</span><span class="n">max_size</span><span class="si">}</span><span class="s2"> must be strictly greater than the requested &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;size for the smaller edge size = </span><span class="si">{</span><span class="n">size</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="p">)</span>
+            <span class="k">if</span> <span class="n">new_long</span> <span class="o">&gt;</span> <span class="n">max_size</span><span class="p">:</span>
+                <span class="n">new_short</span><span class="p">,</span> <span class="n">new_long</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">max_size</span> <span class="o">*</span> <span class="n">new_short</span> <span class="o">/</span> <span class="n">new_long</span><span class="p">),</span> <span class="n">max_size</span>
+
+        <span class="n">new_w</span><span class="p">,</span> <span class="n">new_h</span> <span class="o">=</span> <span class="p">(</span><span class="n">new_short</span><span class="p">,</span> <span class="n">new_long</span><span class="p">)</span> <span class="k">if</span> <span class="n">w</span> <span class="o">&lt;=</span> <span class="n">h</span> <span class="k">else</span> <span class="p">(</span><span class="n">new_long</span><span class="p">,</span> <span class="n">new_short</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>  <span class="c1"># specified both h and w</span>
+        <span class="n">new_w</span><span class="p">,</span> <span class="n">new_h</span> <span class="o">=</span> <span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="k">return</span> <span class="p">[</span><span class="n">new_h</span><span class="p">,</span> <span class="n">new_w</span><span class="p">]</span>
+
+
+<div class="viewcode-block" id="resize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.resize.html#torchvision.transforms.functional.resize">[docs]</a><span class="k">def</span> <span class="nf">resize</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Resize the input image to the given size.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    .. warning::</span>
+<span class="sd">        The output image might be different depending on its type: when downsampling, the interpolation of PIL images</span>
+<span class="sd">        and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences</span>
+<span class="sd">        in the performance of a network. Therefore, it is preferable to train and serve a model with the same input</span>
+<span class="sd">        types. See also below the ``antialias`` parameter, which can help making the output of PIL images and tensors</span>
+<span class="sd">        closer.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be resized.</span>
+<span class="sd">        size (sequence or int): Desired output size. If size is a sequence like</span>
+<span class="sd">            (h, w), the output size will be matched to this. If size is an int,</span>
+<span class="sd">            the smaller edge of the image will be matched to this number maintaining</span>
+<span class="sd">            the aspect ratio. i.e, if height &gt; width, then image will be rescaled to</span>
+<span class="sd">            :math:`\left(\text{size} \times \frac{\text{height}}{\text{width}}, \text{size}\right)`.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode size as single int is not supported, use a sequence of length 1: ``[size, ]``.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`.</span>
+<span class="sd">            Default is ``InterpolationMode.BILINEAR``. If input is Tensor, only ``InterpolationMode.NEAREST``,</span>
+<span class="sd">            ``InterpolationMode.NEAREST_EXACT``, ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are</span>
+<span class="sd">            supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        max_size (int, optional): The maximum allowed for the longer edge of</span>
+<span class="sd">            the resized image. If the longer edge of the image is greater</span>
+<span class="sd">            than ``max_size`` after being resized according to ``size``,</span>
+<span class="sd">            ``size`` will be overruled so that the longer edge is equal to</span>
+<span class="sd">            ``max_size``.</span>
+<span class="sd">            As a result, the smaller edge may be shorter than ``size``. This</span>
+<span class="sd">            is only supported if ``size`` is an int (or a sequence of length</span>
+<span class="sd">            1 in torchscript mode).</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Resized image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">resize</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+            <span class="s2">&quot;Argument interpolation should be a InterpolationMode or a corresponding Pillow integer constant&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Size must be an int or a 1 or 2 element tuple/list, not a </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span><span class="si">}</span><span class="s2"> element tuple/list&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">max_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;max_size should only be passed if size specifies the length of the smaller edge, &quot;</span>
+                <span class="s2">&quot;i.e. size should be an int or a sequence of length 1 in torchscript mode.&quot;</span>
+            <span class="p">)</span>
+
+    <span class="n">_</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">[</span><span class="n">size</span><span class="p">]</span>
+    <span class="n">output_size</span> <span class="o">=</span> <span class="n">_compute_resized_output_size</span><span class="p">((</span><span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">),</span> <span class="n">size</span><span class="p">,</span> <span class="n">max_size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="p">[</span><span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">]</span> <span class="o">==</span> <span class="n">output_size</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">img</span>
+
+    <span class="n">antialias</span> <span class="o">=</span> <span class="n">_check_antialias</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">antialias</span><span class="p">,</span> <span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">antialias</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Anti-alias option is always applied for PIL Image input. Argument antialias is ignored.&quot;</span><span class="p">)</span>
+        <span class="n">pil_interpolation</span> <span class="o">=</span> <span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">resize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">output_size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">pil_interpolation</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">resize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">output_size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="pad"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.pad.html#torchvision.transforms.functional.pad">[docs]</a><span class="k">def</span> <span class="nf">pad</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span> <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Pad the given image on all sides with the given &quot;pad&quot; value.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means at most 2 leading dimensions for mode reflect and symmetric,</span>
+<span class="sd">    at most 3 leading dimensions for mode edge,</span>
+<span class="sd">    and an arbitrary number of leading dimensions for mode constant</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be padded.</span>
+<span class="sd">        padding (int or sequence): Padding on each border. If a single int is provided this</span>
+<span class="sd">            is used to pad all borders. If sequence of length 2 is provided this is the padding</span>
+<span class="sd">            on left/right and top/bottom respectively. If a sequence of length 4 is provided</span>
+<span class="sd">            this is the padding for the left, top, right and bottom borders respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode padding as single int is not supported, use a sequence of</span>
+<span class="sd">                length 1: ``[padding, ]``.</span>
+<span class="sd">        fill (number or tuple): Pixel fill value for constant fill. Default is 0.</span>
+<span class="sd">            If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            This value is only used when the padding_mode is constant.</span>
+<span class="sd">            Only number is supported for torch Tensor.</span>
+<span class="sd">            Only int or tuple value is supported for PIL Image.</span>
+<span class="sd">        padding_mode (str): Type of padding. Should be: constant, edge, reflect or symmetric.</span>
+<span class="sd">            Default is constant.</span>
+
+<span class="sd">            - constant: pads with a constant value, this value is specified with fill</span>
+
+<span class="sd">            - edge: pads with the last value at the edge of the image.</span>
+<span class="sd">              If input a 5D torch Tensor, the last 3 dimensions will be padded instead of the last 2</span>
+
+<span class="sd">            - reflect: pads with reflection of image without repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode</span>
+<span class="sd">              will result in [3, 2, 1, 2, 3, 4, 3, 2]</span>
+
+<span class="sd">            - symmetric: pads with reflection of image repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode</span>
+<span class="sd">              will result in [2, 1, 1, 2, 3, 4, 4, 3]</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Padded image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">pad</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="crop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.crop.html#torchvision.transforms.functional.crop">[docs]</a><span class="k">def</span> <span class="nf">crop</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crop the given image at specified location and output size.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If image size is smaller than output size along any edge, image is padded with 0 and then cropped.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be cropped. (0,0) denotes the top left corner of the image.</span>
+<span class="sd">        top (int): Vertical component of the top left corner of the crop box.</span>
+<span class="sd">        left (int): Horizontal component of the top left corner of the crop box.</span>
+<span class="sd">        height (int): Height of the crop box.</span>
+<span class="sd">        width (int): Width of the crop box.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Cropped image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">crop</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="center_crop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.center_crop.html#torchvision.transforms.functional.center_crop">[docs]</a><span class="k">def</span> <span class="nf">center_crop</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crops the given image at the center.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If image size is smaller than output size along any edge, image is padded with 0 and then center cropped.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be cropped.</span>
+<span class="sd">        output_size (sequence or int): (height, width) of the crop box. If int or sequence with single int,</span>
+<span class="sd">            it is used for both directions.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Cropped image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">center_crop</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output_size</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="n">output_size</span> <span class="o">=</span> <span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">output_size</span><span class="p">),</span> <span class="nb">int</span><span class="p">(</span><span class="n">output_size</span><span class="p">))</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output_size</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">output_size</span> <span class="o">=</span> <span class="p">(</span><span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+
+    <span class="n">_</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span> <span class="o">=</span> <span class="n">output_size</span>
+
+    <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="ow">or</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span><span class="p">:</span>
+        <span class="n">padding_ltrb</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="p">(</span><span class="n">crop_width</span> <span class="o">-</span> <span class="n">image_width</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+            <span class="p">(</span><span class="n">crop_height</span> <span class="o">-</span> <span class="n">image_height</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+            <span class="p">(</span><span class="n">crop_width</span> <span class="o">-</span> <span class="n">image_width</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+            <span class="p">(</span><span class="n">crop_height</span> <span class="o">-</span> <span class="n">image_height</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="p">]</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">pad</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">padding_ltrb</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>  <span class="c1"># PIL uses fill value 0</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">crop_width</span> <span class="o">==</span> <span class="n">image_width</span> <span class="ow">and</span> <span class="n">crop_height</span> <span class="o">==</span> <span class="n">image_height</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">img</span>
+
+    <span class="n">crop_top</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">((</span><span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">)</span> <span class="o">/</span> <span class="mf">2.0</span><span class="p">))</span>
+    <span class="n">crop_left</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">((</span><span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">)</span> <span class="o">/</span> <span class="mf">2.0</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">crop_top</span><span class="p">,</span> <span class="n">crop_left</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="resized_crop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.resized_crop.html#torchvision.transforms.functional.resized_crop">[docs]</a><span class="k">def</span> <span class="nf">resized_crop</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crop the given image and resize it to desired size.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    Notably used in :class:`~torchvision.transforms.RandomResizedCrop`.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be cropped. (0,0) denotes the top left corner of the image.</span>
+<span class="sd">        top (int): Vertical component of the top left corner of the crop box.</span>
+<span class="sd">        left (int): Horizontal component of the top left corner of the crop box.</span>
+<span class="sd">        height (int): Height of the crop box.</span>
+<span class="sd">        width (int): Width of the crop box.</span>
+<span class="sd">        size (sequence or int): Desired output size. Same semantics as ``resize``.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`.</span>
+<span class="sd">            Default is ``InterpolationMode.BILINEAR``. If input is Tensor, only ``InterpolationMode.NEAREST``,</span>
+<span class="sd">            ``InterpolationMode.NEAREST_EXACT``, ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are</span>
+<span class="sd">            supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Cropped image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">)</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">resize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">img</span></div>
+
+
+<div class="viewcode-block" id="hflip"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.hflip.html#torchvision.transforms.functional.hflip">[docs]</a><span class="k">def</span> <span class="nf">hflip</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Horizontally flip the given image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be flipped. If img</span>
+<span class="sd">            is a Tensor, it is expected to be in [..., H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading</span>
+<span class="sd">            dimensions.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor:  Horizontally flipped image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">hflip</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">hflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">hflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_get_perspective_coeffs</span><span class="p">(</span><span class="n">startpoints</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span> <span class="n">endpoints</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Helper function to get the coefficients (a, b, c, d, e, f, g, h) for the perspective transforms.</span>
+
+<span class="sd">    In Perspective Transform each pixel (x, y) in the original image gets transformed as,</span>
+<span class="sd">     (x, y) -&gt; ( (ax + by + c) / (gx + hy + 1), (dx + ey + f) / (gx + hy + 1) )</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        startpoints (list of list of ints): List containing four lists of two integers corresponding to four corners</span>
+<span class="sd">            ``[top-left, top-right, bottom-right, bottom-left]`` of the original image.</span>
+<span class="sd">        endpoints (list of list of ints): List containing four lists of two integers corresponding to four corners</span>
+<span class="sd">            ``[top-left, top-right, bottom-right, bottom-left]`` of the transformed image.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        octuple (a, b, c, d, e, f, g, h) for transforming each pixel.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">a_matrix</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">startpoints</span><span class="p">),</span> <span class="mi">8</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">)</span>
+
+    <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">p1</span><span class="p">,</span> <span class="n">p2</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="n">endpoints</span><span class="p">,</span> <span class="n">startpoints</span><span class="p">)):</span>
+        <span class="n">a_matrix</span><span class="p">[</span><span class="mi">2</span> <span class="o">*</span> <span class="n">i</span><span class="p">,</span> <span class="p">:]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">p1</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">p1</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="o">-</span><span class="n">p2</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">p1</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="n">p2</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">p1</span><span class="p">[</span><span class="mi">1</span><span class="p">]])</span>
+        <span class="n">a_matrix</span><span class="p">[</span><span class="mi">2</span> <span class="o">*</span> <span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="p">:]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">p1</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">p1</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="n">p2</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">p1</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="n">p2</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">p1</span><span class="p">[</span><span class="mi">1</span><span class="p">]])</span>
+
+    <span class="n">b_matrix</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">startpoints</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">8</span><span class="p">)</span>
+    <span class="n">res</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linalg</span><span class="o">.</span><span class="n">lstsq</span><span class="p">(</span><span class="n">a_matrix</span><span class="p">,</span> <span class="n">b_matrix</span><span class="p">,</span> <span class="n">driver</span><span class="o">=</span><span class="s2">&quot;gels&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">solution</span>
+
+    <span class="n">output</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">res</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<div class="viewcode-block" id="perspective"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.perspective.html#torchvision.transforms.functional.perspective">[docs]</a><span class="k">def</span> <span class="nf">perspective</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Perform perspective transform of the given image.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be transformed.</span>
+<span class="sd">        startpoints (list of list of ints): List containing four lists of two integers corresponding to four corners</span>
+<span class="sd">            ``[top-left, top-right, bottom-right, bottom-left]`` of the original image.</span>
+<span class="sd">        endpoints (list of list of ints): List containing four lists of two integers corresponding to four corners</span>
+<span class="sd">            ``[top-left, top-right, bottom-right, bottom-left]`` of the transformed image.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode single int/float value is not supported, please use a sequence</span>
+<span class="sd">                of length 1: ``[value, ]``.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: transformed Image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">perspective</span><span class="p">)</span>
+
+    <span class="n">coeffs</span> <span class="o">=</span> <span class="n">_get_perspective_coeffs</span><span class="p">(</span><span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+            <span class="s2">&quot;Argument interpolation should be a InterpolationMode or a corresponding Pillow integer constant&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">pil_interpolation</span> <span class="o">=</span> <span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">perspective</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">coeffs</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">pil_interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">perspective</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">coeffs</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="vflip"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.vflip.html#torchvision.transforms.functional.vflip">[docs]</a><span class="k">def</span> <span class="nf">vflip</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Vertically flip the given image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be flipped. If img</span>
+<span class="sd">            is a Tensor, it is expected to be in [..., H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading</span>
+<span class="sd">            dimensions.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor:  Vertically flipped image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">vflip</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">vflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">vflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="five_crop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.five_crop.html#torchvision.transforms.functional.five_crop">[docs]</a><span class="k">def</span> <span class="nf">five_crop</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crop the given image into four corners and the central crop.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    .. Note::</span>
+<span class="sd">        This transform returns a tuple of images and there may be a</span>
+<span class="sd">        mismatch in the number of inputs and targets your ``Dataset`` returns.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be cropped.</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">       tuple: tuple (tl, tr, bl, br, center)</span>
+<span class="sd">       Corresponding top left, top right, bottom left, bottom right and center crop.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">five_crop</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">),</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">))</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">(</span><span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+    <span class="n">_</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span> <span class="o">=</span> <span class="n">size</span>
+    <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="ow">or</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span><span class="p">:</span>
+        <span class="n">msg</span> <span class="o">=</span> <span class="s2">&quot;Requested crop size </span><span class="si">{}</span><span class="s2"> is bigger than input size </span><span class="si">{}</span><span class="s2">&quot;</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">msg</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="p">(</span><span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">)))</span>
+
+    <span class="n">tl</span> <span class="o">=</span> <span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">tr</span> <span class="o">=</span> <span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">bl</span> <span class="o">=</span> <span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">br</span> <span class="o">=</span> <span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+
+    <span class="n">center</span> <span class="o">=</span> <span class="n">center_crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="p">[</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">])</span>
+
+    <span class="k">return</span> <span class="n">tl</span><span class="p">,</span> <span class="n">tr</span><span class="p">,</span> <span class="n">bl</span><span class="p">,</span> <span class="n">br</span><span class="p">,</span> <span class="n">center</span></div>
+
+
+<div class="viewcode-block" id="ten_crop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.ten_crop.html#torchvision.transforms.functional.ten_crop">[docs]</a><span class="k">def</span> <span class="nf">ten_crop</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">vertical_flip</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Generate ten cropped images from the given image.</span>
+<span class="sd">    Crop the given image into four corners and the central crop plus the</span>
+<span class="sd">    flipped version of these (horizontal flipping is used by default).</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    .. Note::</span>
+<span class="sd">        This transform returns a tuple of images and there may be a</span>
+<span class="sd">        mismatch in the number of inputs and targets your ``Dataset`` returns.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be cropped.</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+<span class="sd">        vertical_flip (bool): Use vertical flipping instead of horizontal</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        tuple: tuple (tl, tr, bl, br, center, tl_flip, tr_flip, bl_flip, br_flip, center_flip)</span>
+<span class="sd">        Corresponding top left, top right, bottom left, bottom right and</span>
+<span class="sd">        center crop and same for the flipped image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">ten_crop</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">),</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">))</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">(</span><span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+    <span class="n">first_five</span> <span class="o">=</span> <span class="n">five_crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">vertical_flip</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">vflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">hflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="n">second_five</span> <span class="o">=</span> <span class="n">five_crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">first_five</span> <span class="o">+</span> <span class="n">second_five</span></div>
+
+
+<div class="viewcode-block" id="adjust_brightness"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.adjust_brightness.html#torchvision.transforms.functional.adjust_brightness">[docs]</a><span class="k">def</span> <span class="nf">adjust_brightness</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust brightness of an image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be adjusted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">        brightness_factor (float):  How much to adjust the brightness. Can be</span>
+<span class="sd">            any non-negative number. 0 gives a black image, 1 gives the</span>
+<span class="sd">            original image while 2 increases the brightness by a factor of 2.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Brightness adjusted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_brightness</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="adjust_contrast"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.adjust_contrast.html#torchvision.transforms.functional.adjust_contrast">[docs]</a><span class="k">def</span> <span class="nf">adjust_contrast</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust contrast of an image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be adjusted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">        contrast_factor (float): How much to adjust the contrast. Can be any</span>
+<span class="sd">            non-negative number. 0 gives a solid gray image, 1 gives the</span>
+<span class="sd">            original image while 2 increases the contrast by a factor of 2.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Contrast adjusted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_contrast</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="adjust_saturation"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.adjust_saturation.html#torchvision.transforms.functional.adjust_saturation">[docs]</a><span class="k">def</span> <span class="nf">adjust_saturation</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust color saturation of an image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be adjusted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">        saturation_factor (float):  How much to adjust the saturation. 0 will</span>
+<span class="sd">            give a black and white image, 1 will give the original image while</span>
+<span class="sd">            2 will enhance the saturation by a factor of 2.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Saturation adjusted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_saturation</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="adjust_hue"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.adjust_hue.html#torchvision.transforms.functional.adjust_hue">[docs]</a><span class="k">def</span> <span class="nf">adjust_hue</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">hue_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust hue of an image.</span>
+
+<span class="sd">    The image hue is adjusted by converting the image to HSV and</span>
+<span class="sd">    cyclically shifting the intensities in the hue channel (H).</span>
+<span class="sd">    The image is then converted back to original image mode.</span>
+
+<span class="sd">    `hue_factor` is the amount of shift in H channel and must be in the</span>
+<span class="sd">    interval `[-0.5, 0.5]`.</span>
+
+<span class="sd">    See `Hue`_ for more details.</span>
+
+<span class="sd">    .. _Hue: https://en.wikipedia.org/wiki/Hue</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be adjusted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            If img is PIL Image mode &quot;1&quot;, &quot;I&quot;, &quot;F&quot; and modes with transparency (alpha channel) are not supported.</span>
+<span class="sd">            Note: the pixel values of the input image has to be non-negative for conversion to HSV space;</span>
+<span class="sd">            thus it does not work if you normalize your image to an interval with negative values,</span>
+<span class="sd">            or use an interpolation that generates negative values before using this function.</span>
+<span class="sd">        hue_factor (float):  How much to shift the hue channel. Should be in</span>
+<span class="sd">            [-0.5, 0.5]. 0.5 and -0.5 give complete reversal of hue channel in</span>
+<span class="sd">            HSV space in positive and negative direction respectively.</span>
+<span class="sd">            0 means no shift. Therefore, both -0.5 and 0.5 will give an image</span>
+<span class="sd">            with complementary colors while 0 gives the original image.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Hue adjusted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_hue</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">adjust_hue</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">hue_factor</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">adjust_hue</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">hue_factor</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="adjust_gamma"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.adjust_gamma.html#torchvision.transforms.functional.adjust_gamma">[docs]</a><span class="k">def</span> <span class="nf">adjust_gamma</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">gamma</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">gain</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Perform gamma correction on an image.</span>
+
+<span class="sd">    Also known as Power Law Transform. Intensities in RGB mode are adjusted</span>
+<span class="sd">    based on the following equation:</span>
+
+<span class="sd">    .. math::</span>
+<span class="sd">        I_{\text{out}} = 255 \times \text{gain} \times \left(\frac{I_{\text{in}}}{255}\right)^{\gamma}</span>
+
+<span class="sd">    See `Gamma Correction`_ for more details.</span>
+
+<span class="sd">    .. _Gamma Correction: https://en.wikipedia.org/wiki/Gamma_correction</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): PIL Image to be adjusted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            If img is PIL Image, modes with transparency (alpha channel) are not supported.</span>
+<span class="sd">        gamma (float): Non negative real number, same as :math:`\gamma` in the equation.</span>
+<span class="sd">            gamma larger than 1 make the shadows darker,</span>
+<span class="sd">            while gamma smaller than 1 make dark regions lighter.</span>
+<span class="sd">        gain (float): The constant multiplier.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Gamma correction adjusted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_gamma</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">adjust_gamma</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">gamma</span><span class="p">,</span> <span class="n">gain</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">adjust_gamma</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">gamma</span><span class="p">,</span> <span class="n">gain</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_get_inverse_affine_matrix</span><span class="p">(</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">inverted</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]:</span>
+    <span class="c1"># Helper method to compute inverse matrix for affine transformation</span>
+
+    <span class="c1"># Pillow requires inverse affine transformation matrix:</span>
+    <span class="c1"># Affine matrix is : M = T * C * RotateScaleShear * C^-1</span>
+    <span class="c1">#</span>
+    <span class="c1"># where T is translation matrix: [1, 0, tx | 0, 1, ty | 0, 0, 1]</span>
+    <span class="c1">#       C is translation matrix to keep center: [1, 0, cx | 0, 1, cy | 0, 0, 1]</span>
+    <span class="c1">#       RotateScaleShear is rotation with scale and shear matrix</span>
+    <span class="c1">#</span>
+    <span class="c1">#       RotateScaleShear(a, s, (sx, sy)) =</span>
+    <span class="c1">#       = R(a) * S(s) * SHy(sy) * SHx(sx)</span>
+    <span class="c1">#       = [ s*cos(a - sy)/cos(sy), s*(-cos(a - sy)*tan(sx)/cos(sy) - sin(a)), 0 ]</span>
+    <span class="c1">#         [ s*sin(a - sy)/cos(sy), s*(-sin(a - sy)*tan(sx)/cos(sy) + cos(a)), 0 ]</span>
+    <span class="c1">#         [ 0                    , 0                                      , 1 ]</span>
+    <span class="c1"># where R is a rotation matrix, S is a scaling matrix, and SHx and SHy are the shears:</span>
+    <span class="c1"># SHx(s) = [1, -tan(s)] and SHy(s) = [1      , 0]</span>
+    <span class="c1">#          [0, 1      ]              [-tan(s), 1]</span>
+    <span class="c1">#</span>
+    <span class="c1"># Thus, the inverse is M^-1 = C * RotateScaleShear^-1 * C^-1 * T^-1</span>
+
+    <span class="n">rot</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">radians</span><span class="p">(</span><span class="n">angle</span><span class="p">)</span>
+    <span class="n">sx</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">radians</span><span class="p">(</span><span class="n">shear</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+    <span class="n">sy</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">radians</span><span class="p">(</span><span class="n">shear</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+
+    <span class="n">cx</span><span class="p">,</span> <span class="n">cy</span> <span class="o">=</span> <span class="n">center</span>
+    <span class="n">tx</span><span class="p">,</span> <span class="n">ty</span> <span class="o">=</span> <span class="n">translate</span>
+
+    <span class="c1"># RSS without scaling</span>
+    <span class="n">a</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">rot</span> <span class="o">-</span> <span class="n">sy</span><span class="p">)</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">sy</span><span class="p">)</span>
+    <span class="n">b</span> <span class="o">=</span> <span class="o">-</span><span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">rot</span> <span class="o">-</span> <span class="n">sy</span><span class="p">)</span> <span class="o">*</span> <span class="n">math</span><span class="o">.</span><span class="n">tan</span><span class="p">(</span><span class="n">sx</span><span class="p">)</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">sy</span><span class="p">)</span> <span class="o">-</span> <span class="n">math</span><span class="o">.</span><span class="n">sin</span><span class="p">(</span><span class="n">rot</span><span class="p">)</span>
+    <span class="n">c</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">sin</span><span class="p">(</span><span class="n">rot</span> <span class="o">-</span> <span class="n">sy</span><span class="p">)</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">sy</span><span class="p">)</span>
+    <span class="n">d</span> <span class="o">=</span> <span class="o">-</span><span class="n">math</span><span class="o">.</span><span class="n">sin</span><span class="p">(</span><span class="n">rot</span> <span class="o">-</span> <span class="n">sy</span><span class="p">)</span> <span class="o">*</span> <span class="n">math</span><span class="o">.</span><span class="n">tan</span><span class="p">(</span><span class="n">sx</span><span class="p">)</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">sy</span><span class="p">)</span> <span class="o">+</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">rot</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">inverted</span><span class="p">:</span>
+        <span class="c1"># Inverted rotation matrix with scale and shear</span>
+        <span class="c1"># det([[a, b], [c, d]]) == 1, since det(rotation) = 1 and det(shear) = 1</span>
+        <span class="n">matrix</span> <span class="o">=</span> <span class="p">[</span><span class="n">d</span><span class="p">,</span> <span class="o">-</span><span class="n">b</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="o">-</span><span class="n">c</span><span class="p">,</span> <span class="n">a</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+        <span class="n">matrix</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span> <span class="o">/</span> <span class="n">scale</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">matrix</span><span class="p">]</span>
+        <span class="c1"># Apply inverse of translation and of center translation: RSS^-1 * C^-1 * T^-1</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">+=</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cx</span> <span class="o">-</span> <span class="n">tx</span><span class="p">)</span> <span class="o">+</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cy</span> <span class="o">-</span> <span class="n">ty</span><span class="p">)</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">+=</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cx</span> <span class="o">-</span> <span class="n">tx</span><span class="p">)</span> <span class="o">+</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cy</span> <span class="o">-</span> <span class="n">ty</span><span class="p">)</span>
+        <span class="c1"># Apply center translation: C * RSS^-1 * C^-1 * T^-1</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cx</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cy</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">matrix</span> <span class="o">=</span> <span class="p">[</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">d</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+        <span class="n">matrix</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span> <span class="o">*</span> <span class="n">scale</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">matrix</span><span class="p">]</span>
+        <span class="c1"># Apply inverse of center translation: RSS * C^-1</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">+=</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cx</span><span class="p">)</span> <span class="o">+</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cy</span><span class="p">)</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">+=</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cx</span><span class="p">)</span> <span class="o">+</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="o">-</span><span class="n">cy</span><span class="p">)</span>
+        <span class="c1"># Apply translation and center : T * C * RSS * C^-1</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cx</span> <span class="o">+</span> <span class="n">tx</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cy</span> <span class="o">+</span> <span class="n">ty</span>
+
+    <span class="k">return</span> <span class="n">matrix</span>
+
+
+<div class="viewcode-block" id="rotate"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.rotate.html#torchvision.transforms.functional.rotate">[docs]</a><span class="k">def</span> <span class="nf">rotate</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Rotate the image by angle.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): image to be rotated.</span>
+<span class="sd">        angle (number): rotation angle value in degrees, counter-clockwise.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        expand (bool, optional): Optional expansion flag.</span>
+<span class="sd">            If true, expands the output image to make it large enough to hold the entire rotated image.</span>
+<span class="sd">            If false or omitted, make the output image the same size as the input image.</span>
+<span class="sd">            Note that the expand flag assumes rotation around the center and no translation.</span>
+<span class="sd">        center (sequence, optional): Optional center of rotation. Origin is the upper left corner.</span>
+<span class="sd">            Default is the center of the image.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode single int/float value is not supported, please use a sequence</span>
+<span class="sd">                of length 1: ``[value, ]``.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Rotated image.</span>
+
+<span class="sd">    .. _filters: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#filters</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">rotate</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+            <span class="s2">&quot;Argument interpolation should be a InterpolationMode or a corresponding Pillow integer constant&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">angle</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument angle should be int or float&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument center should be a sequence&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">pil_interpolation</span> <span class="o">=</span> <span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">rotate</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">pil_interpolation</span><span class="p">,</span> <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="n">center_f</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="c1"># Center values should be in pixel coordinates but translated such that (0, 0) corresponds to image center.</span>
+        <span class="n">center_f</span> <span class="o">=</span> <span class="p">[</span><span class="mf">1.0</span> <span class="o">*</span> <span class="p">(</span><span class="n">c</span> <span class="o">-</span> <span class="n">s</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">)</span> <span class="k">for</span> <span class="n">c</span><span class="p">,</span> <span class="n">s</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="p">[</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">])]</span>
+
+    <span class="c1"># due to current incoherence of rotation angle direction between affine and rotate implementations</span>
+    <span class="c1"># we need to set -angle.</span>
+    <span class="n">matrix</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center_f</span><span class="p">,</span> <span class="o">-</span><span class="n">angle</span><span class="p">,</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span> <span class="mf">1.0</span><span class="p">,</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">])</span>
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">rotate</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">matrix</span><span class="o">=</span><span class="n">matrix</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="affine"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.affine.html#torchvision.transforms.functional.affine">[docs]</a><span class="k">def</span> <span class="nf">affine</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Apply affine transformation on the image keeping image center invariant.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): image to transform.</span>
+<span class="sd">        angle (number): rotation angle in degrees between -180 and 180, clockwise direction.</span>
+<span class="sd">        translate (sequence of integers): horizontal and vertical translations (post-rotation translation)</span>
+<span class="sd">        scale (float): overall scale</span>
+<span class="sd">        shear (float or sequence): shear angle value in degrees between -180 to 180, clockwise direction.</span>
+<span class="sd">            If a sequence is specified, the first value corresponds to a shear parallel to the x-axis, while</span>
+<span class="sd">            the second value corresponds to a shear parallel to the y-axis.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode single int/float value is not supported, please use a sequence</span>
+<span class="sd">                of length 1: ``[value, ]``.</span>
+<span class="sd">        center (sequence, optional): Optional center of rotation. Origin is the upper left corner.</span>
+<span class="sd">            Default is the center of the image.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Transformed image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">affine</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+            <span class="s2">&quot;Argument interpolation should be a InterpolationMode or a corresponding Pillow integer constant&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">angle</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument angle should be int or float&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">translate</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument translate should be a sequence&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">translate</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument translate should be a sequence of length 2&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">scale</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument scale should be positive&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="p">(</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Shear should be either a single value or a sequence of two values&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">angle</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">angle</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">angle</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">translate</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
+        <span class="n">translate</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">translate</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="n">shear</span> <span class="o">=</span> <span class="p">[</span><span class="n">shear</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
+        <span class="n">shear</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">shear</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">shear</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">shear</span> <span class="o">=</span> <span class="p">[</span><span class="n">shear</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">shear</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">shear</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Shear should be a sequence containing two values. Got </span><span class="si">{</span><span class="n">shear</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument center should be a sequence&quot;</span><span class="p">)</span>
+
+    <span class="n">_</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="c1"># center = (width * 0.5 + 0.5, height * 0.5 + 0.5)</span>
+        <span class="c1"># it is visually better to estimate the center without 0.5 offset</span>
+        <span class="c1"># otherwise image rotated by 90 degrees is shifted vs output image of torch.rot90 or F_t.affine</span>
+        <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">center</span> <span class="o">=</span> <span class="p">[</span><span class="n">width</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">,</span> <span class="n">height</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">]</span>
+        <span class="n">matrix</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">)</span>
+        <span class="n">pil_interpolation</span> <span class="o">=</span> <span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">matrix</span><span class="o">=</span><span class="n">matrix</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">pil_interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="n">center_f</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="c1"># Center values should be in pixel coordinates but translated such that (0, 0) corresponds to image center.</span>
+        <span class="n">center_f</span> <span class="o">=</span> <span class="p">[</span><span class="mf">1.0</span> <span class="o">*</span> <span class="p">(</span><span class="n">c</span> <span class="o">-</span> <span class="n">s</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">)</span> <span class="k">for</span> <span class="n">c</span><span class="p">,</span> <span class="n">s</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="p">[</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">])]</span>
+
+    <span class="n">translate_f</span> <span class="o">=</span> <span class="p">[</span><span class="mf">1.0</span> <span class="o">*</span> <span class="n">t</span> <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">translate</span><span class="p">]</span>
+    <span class="n">matrix</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center_f</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translate_f</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">matrix</span><span class="o">=</span><span class="n">matrix</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span></div>
+
+
+<span class="c1"># Looks like to_grayscale() is a stand-alone functional that is never called</span>
+<span class="c1"># from the transform classes. Perhaps it&#39;s still here for BC? I can&#39;t be</span>
+<span class="c1"># bothered to dig.</span>
+<div class="viewcode-block" id="to_grayscale"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.to_grayscale.html#torchvision.transforms.functional.to_grayscale">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+<span class="k">def</span> <span class="nf">to_grayscale</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="mi">1</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert PIL image of any mode (RGB, HSV, LAB, etc) to grayscale version of image.</span>
+<span class="sd">    This transform does not support torch Tensor.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image): PIL Image to be converted to grayscale.</span>
+<span class="sd">        num_output_channels (int): number of channels of the output image. Value can be 1 or 3. Default is 1.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image: Grayscale version of the image.</span>
+
+<span class="sd">        - if num_output_channels = 1 : returned image is single channel</span>
+<span class="sd">        - if num_output_channels = 3 : returned image is 3 channel with r = g = b</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">to_grayscale</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">to_grayscale</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">)</span>
+
+    <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Input should be PIL Image&quot;</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="rgb_to_grayscale"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.rgb_to_grayscale.html#torchvision.transforms.functional.rgb_to_grayscale">[docs]</a><span class="k">def</span> <span class="nf">rgb_to_grayscale</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert RGB image to grayscale version of image.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Please, note that this method supports only RGB images as input. For inputs in other color spaces,</span>
+<span class="sd">        please, consider using meth:`~torchvision.transforms.functional.to_grayscale` with PIL Image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): RGB Image to be converted to grayscale.</span>
+<span class="sd">        num_output_channels (int): number of channels of the output image. Value can be 1 or 3. Default, 1.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Grayscale version of the image.</span>
+
+<span class="sd">        - if num_output_channels = 1 : returned image is single channel</span>
+<span class="sd">        - if num_output_channels = 3 : returned image is 3 channel with r = g = b</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">rgb_to_grayscale</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">to_grayscale</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">rgb_to_grayscale</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="erase"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.erase.html#torchvision.transforms.functional.erase">[docs]</a><span class="k">def</span> <span class="nf">erase</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">i</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">j</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">h</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">v</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Erase the input Tensor Image with given value.</span>
+<span class="sd">    This transform does not support PIL Image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (Tensor Image): Tensor image of size (C, H, W) to be erased</span>
+<span class="sd">        i (int): i in (i,j) i.e coordinates of the upper left corner.</span>
+<span class="sd">        j (int): j in (i,j) i.e coordinates of the upper left corner.</span>
+<span class="sd">        h (int): Height of the erased region.</span>
+<span class="sd">        w (int): Width of the erased region.</span>
+<span class="sd">        v: Erasing value.</span>
+<span class="sd">        inplace(bool, optional): For in-place operations. By default, is set False.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Tensor Image: Erased image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">erase</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;img should be Tensor Image. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">img</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">erase</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="gaussian_blur"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.gaussian_blur.html#torchvision.transforms.functional.gaussian_blur">[docs]</a><span class="k">def</span> <span class="nf">gaussian_blur</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Performs Gaussian blurring on the image by given kernel.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be blurred</span>
+<span class="sd">        kernel_size (sequence of ints or int): Gaussian kernel size. Can be a sequence of integers</span>
+<span class="sd">            like ``(kx, ky)`` or a single integer for square kernels.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode kernel_size as single int is not supported, use a sequence of</span>
+<span class="sd">                length 1: ``[ksize, ]``.</span>
+<span class="sd">        sigma (sequence of floats or float, optional): Gaussian kernel standard deviation. Can be a</span>
+<span class="sd">            sequence of floats like ``(sigma_x, sigma_y)`` or a single float to define the</span>
+<span class="sd">            same sigma in both X/Y directions. If None, then it is computed using</span>
+<span class="sd">            ``kernel_size`` as ``sigma = 0.3 * ((kernel_size - 1) * 0.5 - 1) + 0.8``.</span>
+<span class="sd">            Default, None.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode sigma as single float is</span>
+<span class="sd">                not supported, use a sequence of length 1: ``[sigma, ]``.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Gaussian Blurred version of the image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">gaussian_blur</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;kernel_size should be int or a sequence of integers. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">kernel_size</span> <span class="o">=</span> <span class="p">[</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">]</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If kernel_size is a sequence its length should be 2. Got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">ksize</span> <span class="ow">in</span> <span class="n">kernel_size</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">ksize</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">ksize</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;kernel_size should have odd and positive integers. Got </span><span class="si">{</span><span class="n">kernel_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">sigma</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="n">ksize</span> <span class="o">*</span> <span class="mf">0.15</span> <span class="o">+</span> <span class="mf">0.35</span> <span class="k">for</span> <span class="n">ksize</span> <span class="ow">in</span> <span class="n">kernel_size</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="n">sigma</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sigma should be either float or sequence of floats. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+        <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">sigma</span><span class="p">),</span> <span class="nb">float</span><span class="p">(</span><span class="n">sigma</span><span class="p">)]</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If sigma is a sequence, its length should be 2. Got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">sigma</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">s</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sigma should have positive values. Got </span><span class="si">{</span><span class="n">sigma</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="n">t_img</span> <span class="o">=</span> <span class="n">img</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">_is_pil_image</span><span class="p">(</span><span class="n">img</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;img should be PIL Image or Tensor. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">img</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="n">t_img</span> <span class="o">=</span> <span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">F_t</span><span class="o">.</span><span class="n">gaussian_blur</span><span class="p">(</span><span class="n">t_img</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">sigma</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">to_pil_image</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">img</span><span class="o">.</span><span class="n">mode</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="invert"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.invert.html#torchvision.transforms.functional.invert">[docs]</a><span class="k">def</span> <span class="nf">invert</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Invert the colors of an RGB/grayscale image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to have its colors inverted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Color inverted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">invert</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">invert</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">invert</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="posterize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.posterize.html#torchvision.transforms.functional.posterize">[docs]</a><span class="k">def</span> <span class="nf">posterize</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">bits</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Posterize an image by reducing the number of bits for each color channel.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to have its colors posterized.</span>
+<span class="sd">            If img is torch Tensor, it should be of type torch.uint8, and</span>
+<span class="sd">            it is expected to be in [..., 1 or 3, H, W] format, where ... means</span>
+<span class="sd">            it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+<span class="sd">        bits (int): The number of bits to keep for each channel (0-8).</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Posterized image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">posterize</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">0</span> <span class="o">&lt;=</span> <span class="n">bits</span> <span class="o">&lt;=</span> <span class="mi">8</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The number if bits should be between 0 and 8. Got </span><span class="si">{</span><span class="n">bits</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">posterize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">bits</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">posterize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">bits</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="solarize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.solarize.html#torchvision.transforms.functional.solarize">[docs]</a><span class="k">def</span> <span class="nf">solarize</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Solarize an RGB/grayscale image by inverting all pixel values above a threshold.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to have its colors inverted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+<span class="sd">        threshold (float): All pixels equal or above this value are inverted.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Solarized image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">solarize</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">solarize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">threshold</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">solarize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">threshold</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="adjust_sharpness"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.adjust_sharpness.html#torchvision.transforms.functional.adjust_sharpness">[docs]</a><span class="k">def</span> <span class="nf">adjust_sharpness</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust the sharpness of an image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image to be adjusted.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">        sharpness_factor (float):  How much to adjust the sharpness. Can be</span>
+<span class="sd">            any non-negative number. 0 gives a blurred image, 1 gives the</span>
+<span class="sd">            original image while 2 increases the sharpness by a factor of 2.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Sharpness adjusted image.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_sharpness</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">adjust_sharpness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">adjust_sharpness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="autocontrast"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.autocontrast.html#torchvision.transforms.functional.autocontrast">[docs]</a><span class="k">def</span> <span class="nf">autocontrast</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Maximize contrast of an image by remapping its</span>
+<span class="sd">    pixels per channel so that the lowest becomes black and the lightest</span>
+<span class="sd">    becomes white.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image on which autocontrast is applied.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: An image that was autocontrasted.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">autocontrast</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">autocontrast</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">autocontrast</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="equalize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.functional.equalize.html#torchvision.transforms.functional.equalize">[docs]</a><span class="k">def</span> <span class="nf">equalize</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Equalize the histogram of an image by applying</span>
+<span class="sd">    a non-linear mapping to the input in order to create a uniform</span>
+<span class="sd">    distribution of grayscale values in the output.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image on which equalize is applied.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            The tensor dtype must be ``torch.uint8`` and values are expected to be in ``[0, 255]``.</span>
+<span class="sd">            If img is PIL Image, it is expected to be in mode &quot;P&quot;, &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: An image that was equalized.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">equalize</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">equalize</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">F_t</span><span class="o">.</span><span class="n">equalize</span><span class="p">(</span><span class="n">img</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">elastic_transform</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">displacement</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Transform a tensor image with elastic transformations.</span>
+<span class="sd">    Given alpha and sigma, it will generate displacement</span>
+<span class="sd">    vectors for all pixels based on random offsets. Alpha controls the strength</span>
+<span class="sd">    and sigma controls the smoothness of the displacements.</span>
+<span class="sd">    The displacements are added to an identity grid and the resulting grid is</span>
+<span class="sd">    used to grid_sample from the image.</span>
+
+<span class="sd">    Applications:</span>
+<span class="sd">        Randomly transforms the morphology of objects in images and produces a</span>
+<span class="sd">        see-through-water-like effect.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        img (PIL Image or Tensor): Image on which elastic_transform is applied.</span>
+<span class="sd">            If img is torch Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">            where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">            If img is PIL Image, it is expected to be in mode &quot;P&quot;, &quot;L&quot; or &quot;RGB&quot;.</span>
+<span class="sd">        displacement (Tensor): The displacement field. Expected shape is [1, H, W, 2].</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`.</span>
+<span class="sd">            Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (number or str or tuple): Pixel fill value for constant fill. Default is 0.</span>
+<span class="sd">            If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            This value is only used when the padding_mode is constant.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">elastic_transform</span><span class="p">)</span>
+    <span class="c1"># Backward compatibility with integer value</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+            <span class="s2">&quot;Argument interpolation should be of type InterpolationMode instead of int. &quot;</span>
+            <span class="s2">&quot;Please, use InterpolationMode enum.&quot;</span>
+        <span class="p">)</span>
+        <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">displacement</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument displacement should be a Tensor&quot;</span><span class="p">)</span>
+
+    <span class="n">t_img</span> <span class="o">=</span> <span class="n">img</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">F_pil</span><span class="o">.</span><span class="n">_is_pil_image</span><span class="p">(</span><span class="n">img</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;img should be PIL Image or Tensor. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">img</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">t_img</span> <span class="o">=</span> <span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">t_img</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,)</span> <span class="o">+</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span> <span class="o">+</span> <span class="p">(</span><span class="mi">2</span><span class="p">,)</span>
+    <span class="k">if</span> <span class="n">shape</span> <span class="o">!=</span> <span class="n">displacement</span><span class="o">.</span><span class="n">shape</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Argument displacement shape should be </span><span class="si">{</span><span class="n">shape</span><span class="si">}</span><span class="s2">, but given </span><span class="si">{</span><span class="n">displacement</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="c1"># TODO: if image shape is [N1, N2, ..., C, H, W] and</span>
+    <span class="c1"># displacement is [1, H, W, 2] we need to reshape input image</span>
+    <span class="c1"># such grid_sampler takes internal code for 4D input</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">F_t</span><span class="o">.</span><span class="n">elastic_transform</span><span class="p">(</span>
+        <span class="n">t_img</span><span class="p">,</span>
+        <span class="n">displacement</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">to_pil_image</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">img</span><span class="o">.</span><span class="n">mode</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="c1"># TODO in v0.17: remove this helper and change default of antialias to True everywhere</span>
+<span class="k">def</span> <span class="nf">_check_antialias</span><span class="p">(</span>
+    <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]],</span> <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">antialias</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>  <span class="c1"># it should be &quot;warn&quot;, but we don&#39;t bother checking against that</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">)</span> <span class="ow">and</span> <span class="p">(</span>
+            <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span> <span class="ow">or</span> <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+        <span class="p">):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;The default value of the antialias parameter of all the resizing transforms &quot;</span>
+                <span class="s2">&quot;(Resize(), RandomResizedCrop(), etc.) &quot;</span>
+                <span class="s2">&quot;will change from None to True in v0.17, &quot;</span>
+                <span class="s2">&quot;in order to be consistent across the PIL and Tensor backends. &quot;</span>
+                <span class="s2">&quot;To suppress this warning, directly pass &quot;</span>
+                <span class="s2">&quot;antialias=True (recommended, future default), antialias=None (current default, &quot;</span>
+                <span class="s2">&quot;which means False for Tensors and True for PIL), &quot;</span>
+                <span class="s2">&quot;or antialias=False (only works on Tensors - PIL will still use antialiasing). &quot;</span>
+                <span class="s2">&quot;This also applies if you are using the inference transforms from the models weights: &quot;</span>
+                <span class="s2">&quot;update the call to weights.transforms(antialias=True).&quot;</span>
+            <span class="p">)</span>
+        <span class="n">antialias</span> <span class="o">=</span> <span class="kc">None</span>
+
+    <span class="k">return</span> <span class="n">antialias</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/transforms.html b/0.16/_modules/torchvision/transforms/transforms.html
new file mode 100644
index 00000000000..f4d87015c33
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/transforms.html
@@ -0,0 +1,2829 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.transforms &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.transforms</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.transforms</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">numbers</span>
+<span class="kn">import</span> <span class="nn">random</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">collections.abc</span> <span class="kn">import</span> <span class="n">Sequence</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
+
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">import</span> <span class="nn">accimage</span>
+<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+    <span class="n">accimage</span> <span class="o">=</span> <span class="kc">None</span>
+
+<span class="kn">from</span> <span class="nn">..utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+<span class="kn">from</span> <span class="nn">.</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+<span class="kn">from</span> <span class="nn">.functional</span> <span class="kn">import</span> <span class="n">_interpolation_modes_from_int</span><span class="p">,</span> <span class="n">InterpolationMode</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;Compose&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ToTensor&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;PILToTensor&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ConvertImageDtype&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ToPILImage&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Normalize&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Resize&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;CenterCrop&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Pad&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Lambda&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomApply&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomChoice&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomOrder&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomCrop&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomHorizontalFlip&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomVerticalFlip&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomResizedCrop&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;FiveCrop&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TenCrop&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;LinearTransformation&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ColorJitter&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomRotation&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomAffine&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;Grayscale&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomGrayscale&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomPerspective&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomErasing&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;GaussianBlur&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;InterpolationMode&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomInvert&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomPosterize&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomSolarize&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomAdjustSharpness&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomAutocontrast&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RandomEqualize&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ElasticTransform&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<div class="viewcode-block" id="Compose"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Compose.html#torchvision.transforms.Compose">[docs]</a><span class="k">class</span> <span class="nc">Compose</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Composes several transforms together. This transform does not support torchscript.</span>
+<span class="sd">    Please, see the note below.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transforms (list of ``Transform`` objects): list of transforms to compose.</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        &gt;&gt;&gt; transforms.Compose([</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.CenterCrop(10),</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.PILToTensor(),</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.ConvertImageDtype(torch.float),</span>
+<span class="sd">        &gt;&gt;&gt; ])</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        In order to script the transformations, please use ``torch.nn.Sequential`` as below.</span>
+
+<span class="sd">        &gt;&gt;&gt; transforms = torch.nn.Sequential(</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.CenterCrop(10),</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),</span>
+<span class="sd">        &gt;&gt;&gt; )</span>
+<span class="sd">        &gt;&gt;&gt; scripted_transforms = torch.jit.script(transforms)</span>
+
+<span class="sd">        Make sure to use only scriptable transformations, i.e. that work with ``torch.Tensor``, does not require</span>
+<span class="sd">        `lambda` functions or ``PIL.Image``.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transforms</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+            <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">t</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="s2">&quot;(&quot;</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;    </span><span class="si">{</span><span class="n">t</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">format_string</span></div>
+
+
+<div class="viewcode-block" id="ToTensor"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ToTensor.html#torchvision.transforms.ToTensor">[docs]</a><span class="k">class</span> <span class="nc">ToTensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a PIL Image or ndarray to tensor and scale the values accordingly.</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Converts a PIL Image or numpy.ndarray (H x W x C) in the range</span>
+<span class="sd">    [0, 255] to a torch.FloatTensor of shape (C x H x W) in the range [0.0, 1.0]</span>
+<span class="sd">    if the PIL Image belongs to one of the modes (L, LA, P, I, F, RGB, YCbCr, RGBA, CMYK, 1)</span>
+<span class="sd">    or if the numpy.ndarray has dtype = np.uint8</span>
+
+<span class="sd">    In the other cases, tensors are returned without scaling.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Because the input image is scaled to [0.0, 1.0], this transformation should not be used when</span>
+<span class="sd">        transforming target image masks. See the `references`_ for implementing the transforms for image masks.</span>
+
+<span class="sd">    .. _references: https://github.com/pytorch/vision/tree/main/references/segmentation</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pic</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            pic (PIL Image or numpy.ndarray): Image to be converted to tensor.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Converted image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">to_tensor</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">()&quot;</span></div>
+
+
+<div class="viewcode-block" id="PILToTensor"><a class="viewcode-back" href="../../../generated/torchvision.transforms.PILToTensor.html#torchvision.transforms.PILToTensor">[docs]</a><span class="k">class</span> <span class="nc">PILToTensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a PIL Image to a tensor of the same type - this does not scale values.</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Converts a PIL Image (H x W x C) to a Tensor of shape (C x H x W).</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pic</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        .. note::</span>
+
+<span class="sd">            A deep copy of the underlying array is performed.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            pic (PIL Image): Image to be converted to tensor.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Converted image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">pic</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">()&quot;</span></div>
+
+
+<div class="viewcode-block" id="ConvertImageDtype"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ConvertImageDtype.html#torchvision.transforms.ConvertImageDtype">[docs]</a><span class="k">class</span> <span class="nc">ConvertImageDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a tensor image to the given ``dtype`` and scale the values accordingly.</span>
+
+<span class="sd">    This function does not support PIL Image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        dtype (torch.dtype): Desired data type of the output</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        When converting from a smaller to a larger integer ``dtype`` the maximum values are **not** mapped exactly.</span>
+<span class="sd">        If converted back and forth, this mismatch has no effect.</span>
+
+<span class="sd">    Raises:</span>
+<span class="sd">        RuntimeError: When trying to cast :class:`torch.float32` to :class:`torch.int32` or :class:`torch.int64` as</span>
+<span class="sd">            well as for trying to cast :class:`torch.float64` to :class:`torch.int64`. These conversions might lead to</span>
+<span class="sd">            overflow errors since the floating point ``dtype`` cannot store consecutive integers over the whole range</span>
+<span class="sd">            of the integer ``dtype``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">dtype</span>
+
+<div class="viewcode-block" id="ConvertImageDtype.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ConvertImageDtype.html#torchvision.transforms.ConvertImageDtype.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">image</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_image_dtype</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="ToPILImage"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ToPILImage.html#torchvision.transforms.ToPILImage">[docs]</a><span class="k">class</span> <span class="nc">ToPILImage</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert a tensor or an ndarray to PIL Image</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Converts a torch.*Tensor of shape C x H x W or a numpy ndarray of shape</span>
+<span class="sd">    H x W x C to a PIL Image while adjusting the value range depending on the ``mode``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        mode (`PIL.Image mode`_): color space and pixel depth of input data (optional).</span>
+<span class="sd">            If ``mode`` is ``None`` (default) there are some assumptions made about the input data:</span>
+
+<span class="sd">            - If the input has 4 channels, the ``mode`` is assumed to be ``RGBA``.</span>
+<span class="sd">            - If the input has 3 channels, the ``mode`` is assumed to be ``RGB``.</span>
+<span class="sd">            - If the input has 2 channels, the ``mode`` is assumed to be ``LA``.</span>
+<span class="sd">            - If the input has 1 channel, the ``mode`` is determined by the data type (i.e ``int``, ``float``, ``short``).</span>
+
+<span class="sd">    .. _PIL.Image mode: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">=</span> <span class="n">mode</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pic</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            pic (Tensor or numpy.ndarray): Image to be converted to PIL Image.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image: Image converted to PIL Image.</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">to_pil_image</span><span class="p">(</span><span class="n">pic</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="s2">&quot;(&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;mode=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;)&quot;</span>
+        <span class="k">return</span> <span class="n">format_string</span></div>
+
+
+<div class="viewcode-block" id="Normalize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Normalize.html#torchvision.transforms.Normalize">[docs]</a><span class="k">class</span> <span class="nc">Normalize</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Normalize a tensor image with mean and standard deviation.</span>
+<span class="sd">    This transform does not support PIL Image.</span>
+<span class="sd">    Given mean: ``(mean[1],...,mean[n])`` and std: ``(std[1],..,std[n])`` for ``n``</span>
+<span class="sd">    channels, this transform will normalize each channel of the input</span>
+<span class="sd">    ``torch.*Tensor`` i.e.,</span>
+<span class="sd">    ``output[channel] = (input[channel] - mean[channel]) / std[channel]``</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        This transform acts out of place, i.e., it does not mutate the input tensor.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        mean (sequence): Sequence of means for each channel.</span>
+<span class="sd">        std (sequence): Sequence of standard deviations for each channel.</span>
+<span class="sd">        inplace(bool,optional): Bool to make this operation in-place.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">mean</span><span class="p">,</span> <span class="n">std</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mean</span> <span class="o">=</span> <span class="n">mean</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">std</span> <span class="o">=</span> <span class="n">std</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span> <span class="o">=</span> <span class="n">inplace</span>
+
+<div class="viewcode-block" id="Normalize.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Normalize.html#torchvision.transforms.Normalize.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tensor</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            tensor (Tensor): Tensor image to be normalized.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Normalized Tensor image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">normalize</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">mean</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">std</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(mean=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mean</span><span class="si">}</span><span class="s2">, std=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">std</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="Resize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Resize.html#torchvision.transforms.Resize">[docs]</a><span class="k">class</span> <span class="nc">Resize</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Resize the input image to the given size.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means a maximum of two leading dimensions</span>
+
+<span class="sd">    .. warning::</span>
+<span class="sd">        The output image might be different depending on its type: when downsampling, the interpolation of PIL images</span>
+<span class="sd">        and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences</span>
+<span class="sd">        in the performance of a network. Therefore, it is preferable to train and serve a model with the same input</span>
+<span class="sd">        types. See also below the ``antialias`` parameter, which can help making the output of PIL images and tensors</span>
+<span class="sd">        closer.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size. If size is a sequence like</span>
+<span class="sd">            (h, w), output size will be matched to this. If size is an int,</span>
+<span class="sd">            smaller edge of the image will be matched to this number.</span>
+<span class="sd">            i.e, if height &gt; width, then image will be rescaled to</span>
+<span class="sd">            (size * height / width, size).</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode size as single int is not supported, use a sequence of length 1: ``[size, ]``.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,</span>
+<span class="sd">            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        max_size (int, optional): The maximum allowed for the longer edge of</span>
+<span class="sd">            the resized image. If the longer edge of the image is greater</span>
+<span class="sd">            than ``max_size`` after being resized according to ``size``,</span>
+<span class="sd">            ``size`` will be overruled so that the longer edge is equal to</span>
+<span class="sd">            ``max_size``.</span>
+<span class="sd">            As a result, the smaller edge may be shorter than ``size``. This</span>
+<span class="sd">            is only supported if ``size`` is an int (or a sequence of length</span>
+<span class="sd">            1 in torchscript mode).</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="s2">&quot;warn&quot;</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Size should be int or sequence. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">size</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;If size is a sequence, it should have 1 or 2 values&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span> <span class="o">=</span> <span class="n">max_size</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span> <span class="o">=</span> <span class="n">antialias</span>
+
+<div class="viewcode-block" id="Resize.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Resize.html#torchvision.transforms.Resize.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be scaled.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Rescaled image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">resize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">detail</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;(size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="si">}</span><span class="s2">, interpolation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="si">}</span><span class="s2">, max_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">max_size</span><span class="si">}</span><span class="s2">, antialias=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}{</span><span class="n">detail</span><span class="si">}</span><span class="s2">&quot;</span></div>
+
+
+<div class="viewcode-block" id="CenterCrop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.CenterCrop.html#torchvision.transforms.CenterCrop">[docs]</a><span class="k">class</span> <span class="nc">CenterCrop</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crops the given image at the center.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If image size is smaller than output size along any edge, image is padded with 0 and then center cropped.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="CenterCrop.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.CenterCrop.html#torchvision.transforms.CenterCrop.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be cropped.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Cropped image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">center_crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="Pad"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Pad.html#torchvision.transforms.Pad">[docs]</a><span class="k">class</span> <span class="nc">Pad</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Pad the given image on all sides with the given &quot;pad&quot; value.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means at most 2 leading dimensions for mode reflect and symmetric,</span>
+<span class="sd">    at most 3 leading dimensions for mode edge,</span>
+<span class="sd">    and an arbitrary number of leading dimensions for mode constant</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        padding (int or sequence): Padding on each border. If a single int is provided this</span>
+<span class="sd">            is used to pad all borders. If sequence of length 2 is provided this is the padding</span>
+<span class="sd">            on left/right and top/bottom respectively. If a sequence of length 4 is provided</span>
+<span class="sd">            this is the padding for the left, top, right and bottom borders respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode padding as single int is not supported, use a sequence of</span>
+<span class="sd">                length 1: ``[padding, ]``.</span>
+<span class="sd">        fill (number or tuple): Pixel fill value for constant fill. Default is 0. If a tuple of</span>
+<span class="sd">            length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            This value is only used when the padding_mode is constant.</span>
+<span class="sd">            Only number is supported for torch Tensor.</span>
+<span class="sd">            Only int or tuple value is supported for PIL Image.</span>
+<span class="sd">        padding_mode (str): Type of padding. Should be: constant, edge, reflect or symmetric.</span>
+<span class="sd">            Default is constant.</span>
+
+<span class="sd">            - constant: pads with a constant value, this value is specified with fill</span>
+
+<span class="sd">            - edge: pads with the last value at the edge of the image.</span>
+<span class="sd">              If input a 5D torch Tensor, the last 3 dimensions will be padded instead of the last 2</span>
+
+<span class="sd">            - reflect: pads with reflection of image without repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode</span>
+<span class="sd">              will result in [3, 2, 1, 2, 3, 4, 3, 2]</span>
+
+<span class="sd">            - symmetric: pads with reflection of image repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode</span>
+<span class="sd">              will result in [2, 1, 1, 2, 3, 4, 4, 3]</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="s2">&quot;constant&quot;</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">padding</span><span class="p">,</span> <span class="p">(</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Got inappropriate padding arg&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Got inappropriate fill arg&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">padding_mode</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;constant&quot;</span><span class="p">,</span> <span class="s2">&quot;edge&quot;</span><span class="p">,</span> <span class="s2">&quot;reflect&quot;</span><span class="p">,</span> <span class="s2">&quot;symmetric&quot;</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Padding mode should be either constant, edge, reflect or symmetric&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">padding</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Padding must be an int or a 1, 2, or 4 element tuple, not a </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span><span class="si">}</span><span class="s2"> element tuple&quot;</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="o">=</span> <span class="n">padding</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span> <span class="o">=</span> <span class="n">padding_mode</span>
+
+<div class="viewcode-block" id="Pad.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Pad.html#torchvision.transforms.Pad.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be padded.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Padded image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(padding=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="si">}</span><span class="s2">, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">, padding_mode=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="Lambda"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Lambda.html#torchvision.transforms.Lambda">[docs]</a><span class="k">class</span> <span class="nc">Lambda</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Apply a user-defined lambda as a transform. This transform does not support torchscript.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        lambd (function): Lambda/function to be used for transform.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">lambd</span><span class="p">):</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">callable</span><span class="p">(</span><span class="n">lambd</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Argument lambd should be callable, got </span><span class="si">{</span><span class="nb">repr</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">lambd</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">lambd</span> <span class="o">=</span> <span class="n">lambd</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">lambd</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">()&quot;</span></div>
+
+
+<span class="k">class</span> <span class="nc">RandomTransforms</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Base class for a list of transformations with randomness</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transforms (sequence): list of transformations</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transforms</span><span class="p">):</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">transforms</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument transforms should be a sequence&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="s2">&quot;(&quot;</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;    </span><span class="si">{</span><span class="n">t</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">format_string</span>
+
+
+<div class="viewcode-block" id="RandomApply"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomApply.html#torchvision.transforms.RandomApply">[docs]</a><span class="k">class</span> <span class="nc">RandomApply</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Apply randomly a list of transformations with a given probability.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        In order to script the transformation, please use ``torch.nn.ModuleList`` as input instead of list/tuple of</span>
+<span class="sd">        transforms as shown below:</span>
+
+<span class="sd">        &gt;&gt;&gt; transforms = transforms.RandomApply(torch.nn.ModuleList([</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.ColorJitter(),</span>
+<span class="sd">        &gt;&gt;&gt; ]), p=0.3)</span>
+<span class="sd">        &gt;&gt;&gt; scripted_transforms = torch.jit.script(transforms)</span>
+
+<span class="sd">        Make sure to use only scriptable transformations, i.e. that work with ``torch.Tensor``, does not require</span>
+<span class="sd">        `lambda` functions or ``PIL.Image``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transforms (sequence or torch.nn.Module): list of transformations</span>
+<span class="sd">        p (float): probability</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transforms</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomApply.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomApply.html#torchvision.transforms.RandomApply.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">&lt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">):</span>
+            <span class="k">return</span> <span class="n">img</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">t</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="s2">&quot;(&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">    p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;    </span><span class="si">{</span><span class="n">t</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">format_string</span></div>
+
+
+<div class="viewcode-block" id="RandomOrder"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomOrder.html#torchvision.transforms.RandomOrder">[docs]</a><span class="k">class</span> <span class="nc">RandomOrder</span><span class="p">(</span><span class="n">RandomTransforms</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Apply a list of transformations in a random order. This transform does not support torchscript.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+        <span class="n">order</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">)))</span>
+        <span class="n">random</span><span class="o">.</span><span class="n">shuffle</span><span class="p">(</span><span class="n">order</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">order</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">[</span><span class="n">i</span><span class="p">](</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+
+<div class="viewcode-block" id="RandomChoice"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomChoice.html#torchvision.transforms.RandomChoice">[docs]</a><span class="k">class</span> <span class="nc">RandomChoice</span><span class="p">(</span><span class="n">RandomTransforms</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Apply single transformation randomly picked from a list. This transform does not support torchscript.&quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transforms</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">p</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument p should be a sequence&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+    <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">):</span>
+        <span class="n">t</span> <span class="o">=</span> <span class="n">random</span><span class="o">.</span><span class="n">choices</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">t</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__repr__</span><span class="p">()</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomCrop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomCrop.html#torchvision.transforms.RandomCrop">[docs]</a><span class="k">class</span> <span class="nc">RandomCrop</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crop the given image at a random location.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions,</span>
+<span class="sd">    but if non-constant padding is used, the input is expected to have at most 2 leading dimensions</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+<span class="sd">        padding (int or sequence, optional): Optional padding on each border</span>
+<span class="sd">            of the image. Default is None. If a single int is provided this</span>
+<span class="sd">            is used to pad all borders. If sequence of length 2 is provided this is the padding</span>
+<span class="sd">            on left/right and top/bottom respectively. If a sequence of length 4 is provided</span>
+<span class="sd">            this is the padding for the left, top, right and bottom borders respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode padding as single int is not supported, use a sequence of</span>
+<span class="sd">                length 1: ``[padding, ]``.</span>
+<span class="sd">        pad_if_needed (boolean): It will pad the image if smaller than the</span>
+<span class="sd">            desired size to avoid raising an exception. Since cropping is done</span>
+<span class="sd">            after padding, the padding seems to be done at a random offset.</span>
+<span class="sd">        fill (number or tuple): Pixel fill value for constant fill. Default is 0. If a tuple of</span>
+<span class="sd">            length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            This value is only used when the padding_mode is constant.</span>
+<span class="sd">            Only number is supported for torch Tensor.</span>
+<span class="sd">            Only int or tuple value is supported for PIL Image.</span>
+<span class="sd">        padding_mode (str): Type of padding. Should be: constant, edge, reflect or symmetric.</span>
+<span class="sd">            Default is constant.</span>
+
+<span class="sd">            - constant: pads with a constant value, this value is specified with fill</span>
+
+<span class="sd">            - edge: pads with the last value at the edge of the image.</span>
+<span class="sd">              If input a 5D torch Tensor, the last 3 dimensions will be padded instead of the last 2</span>
+
+<span class="sd">            - reflect: pads with reflection of image without repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode</span>
+<span class="sd">              will result in [3, 2, 1, 2, 3, 4, 3, 2]</span>
+
+<span class="sd">            - symmetric: pads with reflection of image repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode</span>
+<span class="sd">              will result in [2, 1, 1, 2, 3, 4, 4, 3]</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+<div class="viewcode-block" id="RandomCrop.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.RandomCrop.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get parameters for ``crop`` for a random crop.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be cropped.</span>
+<span class="sd">            output_size (tuple): Expected output size of the crop.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: params (i, j, h, w) to be passed to ``crop`` for random crop.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="n">th</span><span class="p">,</span> <span class="n">tw</span> <span class="o">=</span> <span class="n">output_size</span>
+
+        <span class="k">if</span> <span class="n">h</span> <span class="o">&lt;</span> <span class="n">th</span> <span class="ow">or</span> <span class="n">w</span> <span class="o">&lt;</span> <span class="n">tw</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Required crop size </span><span class="si">{</span><span class="p">(</span><span class="n">th</span><span class="p">,</span><span class="w"> </span><span class="n">tw</span><span class="p">)</span><span class="si">}</span><span class="s2"> is larger than input image size </span><span class="si">{</span><span class="p">(</span><span class="n">h</span><span class="p">,</span><span class="w"> </span><span class="n">w</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">w</span> <span class="o">==</span> <span class="n">tw</span> <span class="ow">and</span> <span class="n">h</span> <span class="o">==</span> <span class="n">th</span><span class="p">:</span>
+            <span class="k">return</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span>
+
+        <span class="n">i</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">h</span> <span class="o">-</span> <span class="n">th</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="n">j</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">w</span> <span class="o">-</span> <span class="n">tw</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="k">return</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">th</span><span class="p">,</span> <span class="n">tw</span></div>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">pad_if_needed</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="s2">&quot;constant&quot;</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="o">=</span> <span class="n">padding</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pad_if_needed</span> <span class="o">=</span> <span class="n">pad_if_needed</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span> <span class="o">=</span> <span class="n">padding_mode</span>
+
+<div class="viewcode-block" id="RandomCrop.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomCrop.html#torchvision.transforms.RandomCrop.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be cropped.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Cropped image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span><span class="p">)</span>
+
+        <span class="n">_</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="c1"># pad the width if needed</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pad_if_needed</span> <span class="ow">and</span> <span class="n">width</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
+            <span class="n">padding</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">width</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span><span class="p">)</span>
+        <span class="c1"># pad the height if needed</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pad_if_needed</span> <span class="ow">and</span> <span class="n">height</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="n">padding</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="n">height</span><span class="p">]</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span><span class="p">)</span>
+
+        <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="si">}</span><span class="s2">, padding=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomHorizontalFlip"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomHorizontalFlip.html#torchvision.transforms.RandomHorizontalFlip">[docs]</a><span class="k">class</span> <span class="nc">RandomHorizontalFlip</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Horizontally flip the given image randomly with a given probability.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading</span>
+<span class="sd">    dimensions</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being flipped. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomHorizontalFlip.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomHorizontalFlip.html#torchvision.transforms.RandomHorizontalFlip.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be flipped.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly flipped image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">hflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomVerticalFlip"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomVerticalFlip.html#torchvision.transforms.RandomVerticalFlip">[docs]</a><span class="k">class</span> <span class="nc">RandomVerticalFlip</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Vertically flip the given image randomly with a given probability.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading</span>
+<span class="sd">    dimensions</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being flipped. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomVerticalFlip.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomVerticalFlip.html#torchvision.transforms.RandomVerticalFlip.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be flipped.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly flipped image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">vflip</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomPerspective"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomPerspective.html#torchvision.transforms.RandomPerspective">[docs]</a><span class="k">class</span> <span class="nc">RandomPerspective</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Performs a random perspective transformation of the given image with a given probability.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        distortion_scale (float): argument to control the degree of distortion and ranges from 0 to 1.</span>
+<span class="sd">            Default is 0.5.</span>
+<span class="sd">        p (float): probability of the image being transformed. Default is 0.5.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (sequence or number): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. Default is ``0``. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">distortion_scale</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">distortion_scale</span> <span class="o">=</span> <span class="n">distortion_scale</span>
+
+        <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">fill</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="n">Sequence</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Fill should be either a sequence or a number.&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+
+<div class="viewcode-block" id="RandomPerspective.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomPerspective.html#torchvision.transforms.RandomPerspective.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be Perspectively transformed.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">fill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span>
+        <span class="n">channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span> <span class="o">*</span> <span class="n">channels</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">distortion_scale</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">perspective</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+<div class="viewcode-block" id="RandomPerspective.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.RandomPerspective.html#torchvision.transforms.RandomPerspective.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">distortion_scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get parameters for ``perspective`` for a random perspective transform.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            width (int): width of the image.</span>
+<span class="sd">            height (int): height of the image.</span>
+<span class="sd">            distortion_scale (float): argument to control the degree of distortion and ranges from 0 to 1.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            List containing [top-left, top-right, bottom-right, bottom-left] of the original image,</span>
+<span class="sd">            List containing [top-left, top-right, bottom-right, bottom-left] of the transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">half_height</span> <span class="o">=</span> <span class="n">height</span> <span class="o">//</span> <span class="mi">2</span>
+        <span class="n">half_width</span> <span class="o">=</span> <span class="n">width</span> <span class="o">//</span> <span class="mi">2</span>
+        <span class="n">topleft</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_width</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_height</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+        <span class="p">]</span>
+        <span class="n">topright</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_width</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_height</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+        <span class="p">]</span>
+        <span class="n">botright</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_width</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">height</span> <span class="o">-</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_height</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+        <span class="p">]</span>
+        <span class="n">botleft</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_width</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">height</span> <span class="o">-</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_height</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()),</span>
+        <span class="p">]</span>
+        <span class="n">startpoints</span> <span class="o">=</span> <span class="p">[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">]]</span>
+        <span class="n">endpoints</span> <span class="o">=</span> <span class="p">[</span><span class="n">topleft</span><span class="p">,</span> <span class="n">topright</span><span class="p">,</span> <span class="n">botright</span><span class="p">,</span> <span class="n">botleft</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomResizedCrop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop">[docs]</a><span class="k">class</span> <span class="nc">RandomResizedCrop</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crop a random portion of image and resize it to a given size.</span>
+
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    A crop of the original image is made: the crop has a random area (H * W)</span>
+<span class="sd">    and a random aspect ratio. This crop is finally resized to the given</span>
+<span class="sd">    size. This is popularly used to train the Inception networks.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (int or sequence): expected output size of the crop, for each edge. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square output size ``(size, size)`` is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode size as single int is not supported, use a sequence of length 1: ``[size, ]``.</span>
+<span class="sd">        scale (tuple of float): Specifies the lower and upper bounds for the random area of the crop,</span>
+<span class="sd">            before resizing. The scale is defined with respect to the area of the original image.</span>
+<span class="sd">        ratio (tuple of float): lower and upper bounds for the random aspect ratio of the crop, before</span>
+<span class="sd">            resizing.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,</span>
+<span class="sd">            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">size</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="p">(</span><span class="mf">0.08</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">),</span>
+        <span class="n">ratio</span><span class="o">=</span><span class="p">(</span><span class="mf">3.0</span> <span class="o">/</span> <span class="mf">4.0</span><span class="p">,</span> <span class="mf">4.0</span> <span class="o">/</span> <span class="mf">3.0</span><span class="p">),</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">scale</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Scale should be a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">ratio</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Ratio should be a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="ow">or</span> <span class="p">(</span><span class="n">ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">]):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scale and ratio should be of kind (min, max)&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span> <span class="o">=</span> <span class="n">antialias</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ratio</span> <span class="o">=</span> <span class="n">ratio</span>
+
+<div class="viewcode-block" id="RandomResizedCrop.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">scale</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">ratio</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get parameters for ``crop`` for a random sized crop.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Input image.</span>
+<span class="sd">            scale (list): range of scale of the origin size cropped</span>
+<span class="sd">            ratio (list): range of aspect ratio of the origin aspect ratio cropped</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: params (i, j, h, w) to be passed to ``crop`` for a random</span>
+<span class="sd">            sized crop.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="n">area</span> <span class="o">=</span> <span class="n">height</span> <span class="o">*</span> <span class="n">width</span>
+
+        <span class="n">log_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">ratio</span><span class="p">))</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">):</span>
+            <span class="n">target_area</span> <span class="o">=</span> <span class="n">area</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="n">aspect_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">log_ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">log_ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+            <span class="n">w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">target_area</span> <span class="o">*</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+            <span class="n">h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">target_area</span> <span class="o">/</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+
+            <span class="k">if</span> <span class="mi">0</span> <span class="o">&lt;</span> <span class="n">w</span> <span class="o">&lt;=</span> <span class="n">width</span> <span class="ow">and</span> <span class="mi">0</span> <span class="o">&lt;</span> <span class="n">h</span> <span class="o">&lt;=</span> <span class="n">height</span><span class="p">:</span>
+                <span class="n">i</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="n">h</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+                <span class="n">j</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">width</span> <span class="o">-</span> <span class="n">w</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+                <span class="k">return</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span>
+
+        <span class="c1"># Fallback to central crop</span>
+        <span class="n">in_ratio</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">width</span><span class="p">)</span> <span class="o">/</span> <span class="nb">float</span><span class="p">(</span><span class="n">height</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">in_ratio</span> <span class="o">&lt;</span> <span class="nb">min</span><span class="p">(</span><span class="n">ratio</span><span class="p">):</span>
+            <span class="n">w</span> <span class="o">=</span> <span class="n">width</span>
+            <span class="n">h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">w</span> <span class="o">/</span> <span class="nb">min</span><span class="p">(</span><span class="n">ratio</span><span class="p">)))</span>
+        <span class="k">elif</span> <span class="n">in_ratio</span> <span class="o">&gt;</span> <span class="nb">max</span><span class="p">(</span><span class="n">ratio</span><span class="p">):</span>
+            <span class="n">h</span> <span class="o">=</span> <span class="n">height</span>
+            <span class="n">w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">h</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="n">ratio</span><span class="p">)))</span>
+        <span class="k">else</span><span class="p">:</span>  <span class="c1"># whole image</span>
+            <span class="n">w</span> <span class="o">=</span> <span class="n">width</span>
+            <span class="n">h</span> <span class="o">=</span> <span class="n">height</span>
+        <span class="n">i</span> <span class="o">=</span> <span class="p">(</span><span class="n">height</span> <span class="o">-</span> <span class="n">h</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span>
+        <span class="n">j</span> <span class="o">=</span> <span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="n">w</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span>
+        <span class="k">return</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span></div>
+
+<div class="viewcode-block" id="RandomResizedCrop.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be cropped and resized.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly cropped and resized image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">resized_crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">interpolate_str</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="sa">f</span><span class="s2">&quot;(size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, scale=</span><span class="si">{</span><span class="nb">tuple</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">s</span><span class="p">,</span><span class="w"> </span><span class="mi">4</span><span class="p">)</span><span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">s</span><span class="w"> </span><span class="ow">in</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, ratio=</span><span class="si">{</span><span class="nb">tuple</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">r</span><span class="p">,</span><span class="w"> </span><span class="mi">4</span><span class="p">)</span><span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">r</span><span class="w"> </span><span class="ow">in</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, interpolation=</span><span class="si">{</span><span class="n">interpolate_str</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, antialias=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">format_string</span></div>
+
+
+<div class="viewcode-block" id="FiveCrop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.FiveCrop.html#torchvision.transforms.FiveCrop">[docs]</a><span class="k">class</span> <span class="nc">FiveCrop</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crop the given image into four corners and the central crop.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading</span>
+<span class="sd">    dimensions</span>
+
+<span class="sd">    .. Note::</span>
+<span class="sd">         This transform returns a tuple of images and there may be a mismatch in the number of</span>
+<span class="sd">         inputs and targets your Dataset returns. See below for an example of how to deal with</span>
+<span class="sd">         this.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">         size (sequence or int): Desired output size of the crop. If size is an ``int``</span>
+<span class="sd">            instead of sequence like (h, w), a square crop of size (size, size) is made.</span>
+<span class="sd">            If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">         &gt;&gt;&gt; transform = Compose([</span>
+<span class="sd">         &gt;&gt;&gt;    FiveCrop(size), # this is a list of PIL Images</span>
+<span class="sd">         &gt;&gt;&gt;    Lambda(lambda crops: torch.stack([PILToTensor()(crop) for crop in crops])) # returns a 4D tensor</span>
+<span class="sd">         &gt;&gt;&gt; ])</span>
+<span class="sd">         &gt;&gt;&gt; #In your test loop you can do the following:</span>
+<span class="sd">         &gt;&gt;&gt; input, target = batch # input is a 5d tensor, target is 2d</span>
+<span class="sd">         &gt;&gt;&gt; bs, ncrops, c, h, w = input.size()</span>
+<span class="sd">         &gt;&gt;&gt; result = model(input.view(-1, c, h, w)) # fuse batch size and ncrops</span>
+<span class="sd">         &gt;&gt;&gt; result_avg = result.view(bs, ncrops, -1).mean(1) # avg over crops</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="FiveCrop.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.FiveCrop.html#torchvision.transforms.FiveCrop.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be cropped.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple of 5 images. Image can be PIL Image or Tensor</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">five_crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="TenCrop"><a class="viewcode-back" href="../../../generated/torchvision.transforms.TenCrop.html#torchvision.transforms.TenCrop">[docs]</a><span class="k">class</span> <span class="nc">TenCrop</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Crop the given image into four corners and the central crop plus the flipped version of</span>
+<span class="sd">    these (horizontal flipping is used by default).</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading</span>
+<span class="sd">    dimensions</span>
+
+<span class="sd">    .. Note::</span>
+<span class="sd">         This transform returns a tuple of images and there may be a mismatch in the number of</span>
+<span class="sd">         inputs and targets your Dataset returns. See below for an example of how to deal with</span>
+<span class="sd">         this.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+<span class="sd">        vertical_flip (bool): Use vertical flipping instead of horizontal</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">         &gt;&gt;&gt; transform = Compose([</span>
+<span class="sd">         &gt;&gt;&gt;    TenCrop(size), # this is a tuple of PIL Images</span>
+<span class="sd">         &gt;&gt;&gt;    Lambda(lambda crops: torch.stack([PILToTensor()(crop) for crop in crops])) # returns a 4D tensor</span>
+<span class="sd">         &gt;&gt;&gt; ])</span>
+<span class="sd">         &gt;&gt;&gt; #In your test loop you can do the following:</span>
+<span class="sd">         &gt;&gt;&gt; input, target = batch # input is a 5d tensor, target is 2d</span>
+<span class="sd">         &gt;&gt;&gt; bs, ncrops, c, h, w = input.size()</span>
+<span class="sd">         &gt;&gt;&gt; result = model(input.view(-1, c, h, w)) # fuse batch size and ncrops</span>
+<span class="sd">         &gt;&gt;&gt; result_avg = result.view(bs, ncrops, -1).mean(1) # avg over crops</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">vertical_flip</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">vertical_flip</span> <span class="o">=</span> <span class="n">vertical_flip</span>
+
+<div class="viewcode-block" id="TenCrop.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.TenCrop.html#torchvision.transforms.TenCrop.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be cropped.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple of 10 images. Image can be PIL Image or Tensor</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">ten_crop</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">vertical_flip</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="si">}</span><span class="s2">, vertical_flip=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">vertical_flip</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="LinearTransformation"><a class="viewcode-back" href="../../../generated/torchvision.transforms.LinearTransformation.html#torchvision.transforms.LinearTransformation">[docs]</a><span class="k">class</span> <span class="nc">LinearTransformation</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Transform a tensor image with a square transformation matrix and a mean_vector computed</span>
+<span class="sd">    offline.</span>
+<span class="sd">    This transform does not support PIL Image.</span>
+<span class="sd">    Given transformation_matrix and mean_vector, will flatten the torch.*Tensor and</span>
+<span class="sd">    subtract mean_vector from it which is then followed by computing the dot</span>
+<span class="sd">    product with the transformation matrix and then reshaping the tensor to its</span>
+<span class="sd">    original shape.</span>
+
+<span class="sd">    Applications:</span>
+<span class="sd">        whitening transformation: Suppose X is a column vector zero-centered data.</span>
+<span class="sd">        Then compute the data covariance matrix [D x D] with torch.mm(X.t(), X),</span>
+<span class="sd">        perform SVD on this matrix and pass it as transformation_matrix.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transformation_matrix (Tensor): tensor [D x D], D = C x H x W</span>
+<span class="sd">        mean_vector (Tensor): tensor [D], D = C x H x W</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transformation_matrix</span><span class="p">,</span> <span class="n">mean_vector</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">!=</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;transformation_matrix should be square. Got &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">tuple</span><span class="p">(</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">())</span><span class="si">}</span><span class="s2"> rectangular matrix.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">mean_vector</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">!=</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;mean_vector should have the same length </span><span class="si">{</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot; as any one of the dimensions of the transformation_matrix [</span><span class="si">{</span><span class="nb">tuple</span><span class="p">(</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">())</span><span class="si">}</span><span class="s2">]&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">device</span> <span class="o">!=</span> <span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Input tensors should be on the same device. Got </span><span class="si">{</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2"> and </span><span class="si">{</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">mean_vector</span><span class="o">.</span><span class="n">dtype</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Input tensors should have the same dtype. Got </span><span class="si">{</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2"> and </span><span class="si">{</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span> <span class="o">=</span> <span class="n">transformation_matrix</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span> <span class="o">=</span> <span class="n">mean_vector</span>
+
+<div class="viewcode-block" id="LinearTransformation.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.LinearTransformation.html#torchvision.transforms.LinearTransformation.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tensor</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            tensor (Tensor): Tensor image to be whitened.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            Tensor: Transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">shape</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">n</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">n</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Input tensor and transformation matrix have incompatible shape.&quot;</span>
+                <span class="o">+</span> <span class="sa">f</span><span class="s2">&quot;[</span><span class="si">{</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span><span class="si">}</span><span class="s2"> x </span><span class="si">{</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span><span class="si">}</span><span class="s2"> x </span><span class="si">{</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2">] != &quot;</span>
+                <span class="o">+</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Input tensor should be on the same device as transformation matrix and mean vector. &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;Got </span><span class="si">{</span><span class="n">tensor</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2"> vs </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">flat_tensor</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">n</span><span class="p">)</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span>
+        <span class="n">transformation_matrix</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">flat_tensor</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="n">transformed_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">mm</span><span class="p">(</span><span class="n">flat_tensor</span><span class="p">,</span> <span class="n">transformation_matrix</span><span class="p">)</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="n">transformed_tensor</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">tensor</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(transformation_matrix=&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, mean_vector=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<div class="viewcode-block" id="ColorJitter"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ColorJitter.html#torchvision.transforms.ColorJitter">[docs]</a><span class="k">class</span> <span class="nc">ColorJitter</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Randomly change the brightness, contrast, saturation and hue of an image.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, mode &quot;1&quot;, &quot;I&quot;, &quot;F&quot; and modes with transparency (alpha channel) are not supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        brightness (float or tuple of float (min, max)): How much to jitter brightness.</span>
+<span class="sd">            brightness_factor is chosen uniformly from [max(0, 1 - brightness), 1 + brightness]</span>
+<span class="sd">            or the given [min, max]. Should be non negative numbers.</span>
+<span class="sd">        contrast (float or tuple of float (min, max)): How much to jitter contrast.</span>
+<span class="sd">            contrast_factor is chosen uniformly from [max(0, 1 - contrast), 1 + contrast]</span>
+<span class="sd">            or the given [min, max]. Should be non-negative numbers.</span>
+<span class="sd">        saturation (float or tuple of float (min, max)): How much to jitter saturation.</span>
+<span class="sd">            saturation_factor is chosen uniformly from [max(0, 1 - saturation), 1 + saturation]</span>
+<span class="sd">            or the given [min, max]. Should be non negative numbers.</span>
+<span class="sd">        hue (float or tuple of float (min, max)): How much to jitter hue.</span>
+<span class="sd">            hue_factor is chosen uniformly from [-hue, hue] or the given [min, max].</span>
+<span class="sd">            Should have 0&lt;= hue &lt;= 0.5 or -0.5 &lt;= min &lt;= max &lt;= 0.5.</span>
+<span class="sd">            To jitter hue, the pixel values of the input image has to be non-negative for conversion to HSV space;</span>
+<span class="sd">            thus it does not work if you normalize your image to an interval with negative values,</span>
+<span class="sd">            or use an interpolation that generates negative values before using this function.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">brightness</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">contrast</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">saturation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">hue</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">brightness</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">brightness</span><span class="p">,</span> <span class="s2">&quot;brightness&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">contrast</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">contrast</span><span class="p">,</span> <span class="s2">&quot;contrast&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">saturation</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">saturation</span><span class="p">,</span> <span class="s2">&quot;saturation&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hue</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">hue</span><span class="p">,</span> <span class="s2">&quot;hue&quot;</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">),</span> <span class="n">clip_first_on_zero</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+    <span class="k">def</span> <span class="nf">_check_input</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">value</span><span class="p">,</span> <span class="n">name</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bound</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)),</span> <span class="n">clip_first_on_zero</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">value</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If </span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> is a single number, it must be non negative.&quot;</span><span class="p">)</span>
+            <span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="n">center</span> <span class="o">-</span> <span class="nb">float</span><span class="p">(</span><span class="n">value</span><span class="p">),</span> <span class="n">center</span> <span class="o">+</span> <span class="nb">float</span><span class="p">(</span><span class="n">value</span><span class="p">)]</span>
+            <span class="k">if</span> <span class="n">clip_first_on_zero</span><span class="p">:</span>
+                <span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mf">0.0</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">value</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="nb">float</span><span class="p">(</span><span class="n">value</span><span class="p">[</span><span class="mi">1</span><span class="p">])]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> should be a single number or a list/tuple with length 2.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">bound</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">value</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">bound</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> values should be between </span><span class="si">{</span><span class="n">bound</span><span class="si">}</span><span class="s2">, but got </span><span class="si">{</span><span class="n">value</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="c1"># if value is 0 or (1., 1.) for brightness/contrast/saturation</span>
+        <span class="c1"># or (0., 0.) for hue, do nothing</span>
+        <span class="k">if</span> <span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">value</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="n">center</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">value</span><span class="p">)</span>
+
+<div class="viewcode-block" id="ColorJitter.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.ColorJitter.html#torchvision.transforms.ColorJitter.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span>
+        <span class="n">brightness</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+        <span class="n">contrast</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+        <span class="n">saturation</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+        <span class="n">hue</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get the parameters for the randomized transform to be applied on image.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            brightness (tuple of float (min, max), optional): The range from which the brightness_factor is chosen</span>
+<span class="sd">                uniformly. Pass None to turn off the transformation.</span>
+<span class="sd">            contrast (tuple of float (min, max), optional): The range from which the contrast_factor is chosen</span>
+<span class="sd">                uniformly. Pass None to turn off the transformation.</span>
+<span class="sd">            saturation (tuple of float (min, max), optional): The range from which the saturation_factor is chosen</span>
+<span class="sd">                uniformly. Pass None to turn off the transformation.</span>
+<span class="sd">            hue (tuple of float (min, max), optional): The range from which the hue_factor is chosen uniformly.</span>
+<span class="sd">                Pass None to turn off the transformation.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: The parameters used to apply the randomized transform</span>
+<span class="sd">            along with their random order.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fn_idx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randperm</span><span class="p">(</span><span class="mi">4</span><span class="p">)</span>
+
+        <span class="n">b</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="n">brightness</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">brightness</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">brightness</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="n">c</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="n">contrast</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">contrast</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">contrast</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="n">saturation</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">saturation</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">saturation</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="n">h</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="n">hue</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">hue</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">hue</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+
+        <span class="k">return</span> <span class="n">fn_idx</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">s</span><span class="p">,</span> <span class="n">h</span></div>
+
+<div class="viewcode-block" id="ColorJitter.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ColorJitter.html#torchvision.transforms.ColorJitter.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Input image.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Color jittered image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fn_idx</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">,</span> <span class="n">hue_factor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">brightness</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">contrast</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">saturation</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">hue</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">fn_id</span> <span class="ow">in</span> <span class="n">fn_idx</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">brightness_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">contrast_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">saturation_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">3</span> <span class="ow">and</span> <span class="n">hue_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_hue</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">hue_factor</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;brightness=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">brightness</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, contrast=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">contrast</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, saturation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">saturation</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;, hue=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">hue</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<div class="viewcode-block" id="RandomRotation"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomRotation.html#torchvision.transforms.RandomRotation">[docs]</a><span class="k">class</span> <span class="nc">RandomRotation</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Rotate the image by angle.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        degrees (sequence or number): Range of degrees to select from.</span>
+<span class="sd">            If degrees is a number instead of sequence like (min, max), the range of degrees</span>
+<span class="sd">            will be (-degrees, +degrees).</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        expand (bool, optional): Optional expansion flag.</span>
+<span class="sd">            If true, expands the output to make it large enough to hold the entire rotated image.</span>
+<span class="sd">            If false or omitted, make the output image the same size as the input image.</span>
+<span class="sd">            Note that the expand flag assumes rotation around the center and no translation.</span>
+<span class="sd">        center (sequence, optional): Optional center of rotation, (x, y). Origin is the upper left corner.</span>
+<span class="sd">            Default is the center of the image.</span>
+<span class="sd">        fill (sequence or number): Pixel fill value for the area outside the rotated</span>
+<span class="sd">            image. Default is ``0``. If given a number, the value is used for all bands respectively.</span>
+
+<span class="sd">    .. _filters: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#filters</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">degrees</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span> <span class="n">expand</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">degrees</span> <span class="o">=</span> <span class="n">_setup_angle</span><span class="p">(</span><span class="n">degrees</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">&quot;degrees&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="s2">&quot;center&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">center</span> <span class="o">=</span> <span class="n">center</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">expand</span> <span class="o">=</span> <span class="n">expand</span>
+
+        <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">fill</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="n">Sequence</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Fill should be either a sequence or a number.&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+
+<div class="viewcode-block" id="RandomRotation.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.RandomRotation.html#torchvision.transforms.RandomRotation.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="n">degrees</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="nb">float</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get parameters for ``rotate`` for a random rotation.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            float: angle parameter to be passed to ``rotate`` for random rotation.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">angle</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">degrees</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="nb">float</span><span class="p">(</span><span class="n">degrees</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+        <span class="k">return</span> <span class="n">angle</span></div>
+
+<div class="viewcode-block" id="RandomRotation.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomRotation.html#torchvision.transforms.RandomRotation.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be rotated.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Rotated image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span>
+        <span class="n">channels</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span> <span class="o">*</span> <span class="n">channels</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+        <span class="n">angle</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">rotate</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">expand</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">center</span><span class="p">,</span> <span class="n">fill</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">interpolate_str</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">+</span> <span class="sa">f</span><span class="s2">&quot;(degrees=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, interpolation=</span><span class="si">{</span><span class="n">interpolate_str</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, expand=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">expand</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, center=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">center</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="s2">&quot;)&quot;</span>
+        <span class="k">return</span> <span class="n">format_string</span></div>
+
+
+<div class="viewcode-block" id="RandomAffine"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomAffine.html#torchvision.transforms.RandomAffine">[docs]</a><span class="k">class</span> <span class="nc">RandomAffine</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Random affine transformation of the image keeping center invariant.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        degrees (sequence or number): Range of degrees to select from.</span>
+<span class="sd">            If degrees is a number instead of sequence like (min, max), the range of degrees</span>
+<span class="sd">            will be (-degrees, +degrees). Set to 0 to deactivate rotations.</span>
+<span class="sd">        translate (tuple, optional): tuple of maximum absolute fraction for horizontal</span>
+<span class="sd">            and vertical translations. For example translate=(a, b), then horizontal shift</span>
+<span class="sd">            is randomly sampled in the range -img_width * a &lt; dx &lt; img_width * a and vertical shift is</span>
+<span class="sd">            randomly sampled in the range -img_height * b &lt; dy &lt; img_height * b. Will not translate by default.</span>
+<span class="sd">        scale (tuple, optional): scaling factor interval, e.g (a, b), then scale is</span>
+<span class="sd">            randomly sampled from the range a &lt;= scale &lt;= b. Will keep original scale by default.</span>
+<span class="sd">        shear (sequence or number, optional): Range of degrees to select from.</span>
+<span class="sd">            If shear is a number, a shear parallel to the x-axis in the range (-shear, +shear)</span>
+<span class="sd">            will be applied. Else if shear is a sequence of 2 values a shear parallel to the x-axis in the</span>
+<span class="sd">            range (shear[0], shear[1]) will be applied. Else if shear is a sequence of 4 values,</span>
+<span class="sd">            an x-axis shear in (shear[0], shear[1]) and y-axis shear in (shear[2], shear[3]) will be applied.</span>
+<span class="sd">            Will not apply shear by default.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (sequence or number): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. Default is ``0``. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">        center (sequence, optional): Optional center of rotation, (x, y). Origin is the upper left corner.</span>
+<span class="sd">            Default is the center of the image.</span>
+
+<span class="sd">    .. _filters: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#filters</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">degrees</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">degrees</span> <span class="o">=</span> <span class="n">_setup_angle</span><span class="p">(</span><span class="n">degrees</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">&quot;degrees&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="k">if</span> <span class="n">translate</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">translate</span><span class="p">,</span> <span class="s2">&quot;translate&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+            <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">translate</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">t</span> <span class="o">&lt;=</span> <span class="mf">1.0</span><span class="p">):</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;translation values should be between 0 and 1&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">translate</span> <span class="o">=</span> <span class="n">translate</span>
+
+        <span class="k">if</span> <span class="n">scale</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">scale</span><span class="p">,</span> <span class="s2">&quot;scale&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+            <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">scale</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">s</span> <span class="o">&lt;=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;scale values should be positive&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">scale</span>
+
+        <span class="k">if</span> <span class="n">shear</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">shear</span> <span class="o">=</span> <span class="n">_setup_angle</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">&quot;shear&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">shear</span> <span class="o">=</span> <span class="n">shear</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+
+        <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">fill</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="n">Sequence</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Fill should be either a sequence or a number.&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+
+        <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="s2">&quot;center&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">center</span> <span class="o">=</span> <span class="n">center</span>
+
+<div class="viewcode-block" id="RandomAffine.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.RandomAffine.html#torchvision.transforms.RandomAffine.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span>
+        <span class="n">degrees</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+        <span class="n">translate</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+        <span class="n">scale_ranges</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+        <span class="n">shears</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+        <span class="n">img_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get parameters for affine transformation</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            params to be passed to the affine transformation</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">angle</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">degrees</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="nb">float</span><span class="p">(</span><span class="n">degrees</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+        <span class="k">if</span> <span class="n">translate</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">max_dx</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">translate</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">img_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="n">max_dy</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">translate</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">img_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="n">tx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="o">-</span><span class="n">max_dx</span><span class="p">,</span> <span class="n">max_dx</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()))</span>
+            <span class="n">ty</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="o">-</span><span class="n">max_dy</span><span class="p">,</span> <span class="n">max_dy</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()))</span>
+            <span class="n">translations</span> <span class="o">=</span> <span class="p">(</span><span class="n">tx</span><span class="p">,</span> <span class="n">ty</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">translations</span> <span class="o">=</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">scale_ranges</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">scale</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">scale_ranges</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">scale_ranges</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">scale</span> <span class="o">=</span> <span class="mf">1.0</span>
+
+        <span class="n">shear_x</span> <span class="o">=</span> <span class="n">shear_y</span> <span class="o">=</span> <span class="mf">0.0</span>
+        <span class="k">if</span> <span class="n">shears</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">shear_x</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">shears</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">shears</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">shears</span><span class="p">)</span> <span class="o">==</span> <span class="mi">4</span><span class="p">:</span>
+                <span class="n">shear_y</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">shears</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">shears</span><span class="p">[</span><span class="mi">3</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+
+        <span class="n">shear</span> <span class="o">=</span> <span class="p">(</span><span class="n">shear_x</span><span class="p">,</span> <span class="n">shear_y</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translations</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span></div>
+
+<div class="viewcode-block" id="RandomAffine.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomAffine.html#torchvision.transforms.RandomAffine.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be transformed.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Affine transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span>
+        <span class="n">channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span> <span class="o">*</span> <span class="n">channels</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+
+        <span class="n">img_size</span> <span class="o">=</span> <span class="p">[</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">]</span>  <span class="c1"># flip for keeping BC on get_params call</span>
+
+        <span class="n">ret</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">translate</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">shear</span><span class="p">,</span> <span class="n">img_size</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="o">*</span><span class="n">ret</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">center</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(degrees=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, translate=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">translate</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">translate</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, scale=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, shear=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">shear</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">shear</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, interpolation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">!=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">!=</span> <span class="mi">0</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, center=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">center</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">s</span> <span class="o">+=</span> <span class="s2">&quot;)&quot;</span>
+
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<div class="viewcode-block" id="Grayscale"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Grayscale.html#torchvision.transforms.Grayscale">[docs]</a><span class="k">class</span> <span class="nc">Grayscale</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Convert image to grayscale.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_output_channels (int): (1 or 3) number of channels desired for output image</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image: Grayscale version of the input.</span>
+
+<span class="sd">        - If ``num_output_channels == 1`` : returned image is single channel</span>
+<span class="sd">        - If ``num_output_channels == 3`` : returned image is 3 channel with r == g == b</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="mi">1</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_output_channels</span> <span class="o">=</span> <span class="n">num_output_channels</span>
+
+<div class="viewcode-block" id="Grayscale.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.Grayscale.html#torchvision.transforms.Grayscale.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be converted to grayscale.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Grayscaled image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">rgb_to_grayscale</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">num_output_channels</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(num_output_channels=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">num_output_channels</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomGrayscale"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomGrayscale.html#torchvision.transforms.RandomGrayscale">[docs]</a><span class="k">class</span> <span class="nc">RandomGrayscale</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Randomly convert image to grayscale with a probability of p (default 0.1).</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability that image should be converted to grayscale.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Grayscale version of the input image with probability p and unchanged</span>
+<span class="sd">        with probability (1-p).</span>
+<span class="sd">        - If input image is 1 channel: grayscale version is 1 channel</span>
+<span class="sd">        - If input image is 3 channel: grayscale version is 3 channel with r == g == b</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.1</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomGrayscale.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomGrayscale.html#torchvision.transforms.RandomGrayscale.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be converted to grayscale.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly grayscaled image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">num_output_channels</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">rgb_to_grayscale</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="n">num_output_channels</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomErasing"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomErasing.html#torchvision.transforms.RandomErasing">[docs]</a><span class="k">class</span> <span class="nc">RandomErasing</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Randomly selects a rectangle region in a torch.Tensor image and erases its pixels.</span>
+<span class="sd">    This transform does not support PIL Image.</span>
+<span class="sd">    &#39;Random Erasing Data Augmentation&#39; by Zhong et al. See https://arxiv.org/abs/1708.04896</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">         p: probability that the random erasing operation will be performed.</span>
+<span class="sd">         scale: range of proportion of erased area against input image.</span>
+<span class="sd">         ratio: range of aspect ratio of erased area.</span>
+<span class="sd">         value: erasing value. Default is 0. If a single int, it is used to</span>
+<span class="sd">            erase all pixels. If a tuple of length 3, it is used to erase</span>
+<span class="sd">            R, G, B channels respectively.</span>
+<span class="sd">            If a str of &#39;random&#39;, erasing each pixel with random values.</span>
+<span class="sd">         inplace: boolean to make this transform inplace. Default set to False.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Erased Image.</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        &gt;&gt;&gt; transform = transforms.Compose([</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.RandomHorizontalFlip(),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.PILToTensor(),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.ConvertImageDtype(torch.float),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.RandomErasing(),</span>
+<span class="sd">        &gt;&gt;&gt; ])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="p">(</span><span class="mf">0.02</span><span class="p">,</span> <span class="mf">0.33</span><span class="p">),</span> <span class="n">ratio</span><span class="o">=</span><span class="p">(</span><span class="mf">0.3</span><span class="p">,</span> <span class="mf">3.3</span><span class="p">),</span> <span class="n">value</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument value should be either a number or str or a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="ow">and</span> <span class="n">value</span> <span class="o">!=</span> <span class="s2">&quot;random&quot;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;If value is str, it should be &#39;random&#39;&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">scale</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Scale should be a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">ratio</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Ratio should be a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="ow">or</span> <span class="p">(</span><span class="n">ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">]):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scale and ratio should be of kind (min, max)&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Scale should be between 0 and 1&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">p</span> <span class="o">&lt;</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">p</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Random erasing probability should be between 0 and 1&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ratio</span> <span class="o">=</span> <span class="n">ratio</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">value</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span> <span class="o">=</span> <span class="n">inplace</span>
+
+<div class="viewcode-block" id="RandomErasing.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.RandomErasing.html#torchvision.transforms.RandomErasing.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span>
+        <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">scale</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span> <span class="n">ratio</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span> <span class="n">value</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Get parameters for ``erase`` for a random erasing.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            img (Tensor): Tensor image to be erased.</span>
+<span class="sd">            scale (sequence): range of proportion of erased area against input image.</span>
+<span class="sd">            ratio (sequence): range of aspect ratio of erased area.</span>
+<span class="sd">            value (list, optional): erasing value. If None, it is interpreted as &quot;random&quot;</span>
+<span class="sd">                (erasing each pixel with random values). If ``len(value)`` is 1, it is interpreted as a number,</span>
+<span class="sd">                i.e. ``value[0]``.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            tuple: params (i, j, h, w, v) to be passed to ``erase`` for random erasing.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">img_c</span><span class="p">,</span> <span class="n">img_h</span><span class="p">,</span> <span class="n">img_w</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">],</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">],</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">area</span> <span class="o">=</span> <span class="n">img_h</span> <span class="o">*</span> <span class="n">img_w</span>
+
+        <span class="n">log_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">ratio</span><span class="p">))</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">):</span>
+            <span class="n">erase_area</span> <span class="o">=</span> <span class="n">area</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="n">aspect_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">log_ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">log_ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+            <span class="n">h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">erase_area</span> <span class="o">*</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+            <span class="n">w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">erase_area</span> <span class="o">/</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">h</span> <span class="o">&lt;</span> <span class="n">img_h</span> <span class="ow">and</span> <span class="n">w</span> <span class="o">&lt;</span> <span class="n">img_w</span><span class="p">):</span>
+                <span class="k">continue</span>
+
+            <span class="k">if</span> <span class="n">value</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">([</span><span class="n">img_c</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span><span class="o">.</span><span class="n">normal_</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">value</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+
+            <span class="n">i</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">img_h</span> <span class="o">-</span> <span class="n">h</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="n">j</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">img_w</span> <span class="o">-</span> <span class="n">w</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="k">return</span> <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">v</span>
+
+        <span class="c1"># Return original image</span>
+        <span class="k">return</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">img_h</span><span class="p">,</span> <span class="n">img_w</span><span class="p">,</span> <span class="n">img</span></div>
+
+<div class="viewcode-block" id="RandomErasing.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomErasing.html#torchvision.transforms.RandomErasing.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (Tensor): Tensor image to be erased.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            img (Tensor): Erased Tensor image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+
+            <span class="c1"># cast self.value to script acceptable type</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+                <span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">)]</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+                <span class="n">value</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+                <span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span>
+
+            <span class="k">if</span> <span class="n">value</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">value</span><span class="p">)</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">])):</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;If value is a sequence, it should have either a single value or &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span><span class="si">}</span><span class="s2"> (number of input channels)&quot;</span>
+                <span class="p">)</span>
+
+            <span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">,</span> <span class="n">ratio</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="n">value</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">erase</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;scale=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="si">}</span><span class="s2">, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;ratio=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="si">}</span><span class="s2">, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;value=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="si">}</span><span class="s2">, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;inplace=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<div class="viewcode-block" id="GaussianBlur"><a class="viewcode-back" href="../../../generated/torchvision.transforms.GaussianBlur.html#torchvision.transforms.GaussianBlur">[docs]</a><span class="k">class</span> <span class="nc">GaussianBlur</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Blurs image with randomly chosen Gaussian blur.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., C, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        kernel_size (int or sequence): Size of the Gaussian kernel.</span>
+<span class="sd">        sigma (float or tuple of float (min, max)): Standard deviation to be used for</span>
+<span class="sd">            creating kernel to perform blurring. If float, sigma is fixed. If it is tuple</span>
+<span class="sd">            of float (min, max), sigma is chosen uniformly at random to lie in the</span>
+<span class="sd">            given range.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        PIL Image or Tensor: Gaussian blurred version of the input image.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">sigma</span><span class="o">=</span><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">2.0</span><span class="p">)):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="s2">&quot;Kernel size should be a tuple/list of two integers&quot;</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">ks</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">ks</span> <span class="o">&lt;=</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">ks</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Kernel size value should be an odd and positive number.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">sigma</span> <span class="o">&lt;=</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;If sigma is a single number, it must be positive.&quot;</span><span class="p">)</span>
+            <span class="n">sigma</span> <span class="o">=</span> <span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="n">sigma</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;sigma values should be positive and of the form (min, max).&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;sigma should be a single number or a list/tuple with length 2.&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span> <span class="o">=</span> <span class="n">sigma</span>
+
+<div class="viewcode-block" id="GaussianBlur.get_params"><a class="viewcode-back" href="../../../generated/torchvision.transforms.v2.GaussianBlur.html#torchvision.transforms.GaussianBlur.get_params">[docs]</a>    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="n">sigma_min</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">sigma_max</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">float</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Choose sigma for random gaussian blurring.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            sigma_min (float): Minimum standard deviation that can be chosen for blurring kernel.</span>
+<span class="sd">            sigma_max (float): Maximum standard deviation that can be chosen for blurring kernel.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            float: Standard deviation to be passed to calculate kernel for gaussian blurring.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">sigma_min</span><span class="p">,</span> <span class="n">sigma_max</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span></div>
+
+<div class="viewcode-block" id="GaussianBlur.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.GaussianBlur.html#torchvision.transforms.GaussianBlur.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): image to be blurred.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Gaussian blurred image</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">sigma</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">gaussian_blur</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">,</span> <span class="p">[</span><span class="n">sigma</span><span class="p">,</span> <span class="n">sigma</span><span class="p">])</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(kernel_size=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="si">}</span><span class="s2">, sigma=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">s</span></div>
+
+
+<span class="k">def</span> <span class="nf">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="p">):</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">),</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">error_msg</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">size</span>
+
+
+<span class="k">def</span> <span class="nf">_check_sequence_input</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">name</span><span class="p">,</span> <span class="n">req_sizes</span><span class="p">):</span>
+    <span class="n">msg</span> <span class="o">=</span> <span class="n">req_sizes</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">req_sizes</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">2</span> <span class="k">else</span> <span class="s2">&quot; or &quot;</span><span class="o">.</span><span class="n">join</span><span class="p">([</span><span class="nb">str</span><span class="p">(</span><span class="n">s</span><span class="p">)</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">req_sizes</span><span class="p">])</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> should be a sequence of length </span><span class="si">{</span><span class="n">msg</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">req_sizes</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> should be a sequence of length </span><span class="si">{</span><span class="n">msg</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_setup_angle</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">name</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,)):</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">x</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If </span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> is a single number, it must be positive.&quot;</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="o">-</span><span class="n">x</span><span class="p">,</span> <span class="n">x</span><span class="p">]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">name</span><span class="p">,</span> <span class="n">req_sizes</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">d</span><span class="p">)</span> <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <span class="n">x</span><span class="p">]</span>
+
+
+<div class="viewcode-block" id="RandomInvert"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomInvert.html#torchvision.transforms.RandomInvert">[docs]</a><span class="k">class</span> <span class="nc">RandomInvert</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Inverts the colors of the given image randomly with a given probability.</span>
+<span class="sd">    If img is a Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">    where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being color inverted. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomInvert.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomInvert.html#torchvision.transforms.RandomInvert.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be inverted.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly color inverted image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">invert</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomPosterize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomPosterize.html#torchvision.transforms.RandomPosterize">[docs]</a><span class="k">class</span> <span class="nc">RandomPosterize</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Posterize the image randomly with a given probability by reducing the</span>
+<span class="sd">    number of bits for each color channel. If the image is torch Tensor, it should be of type torch.uint8,</span>
+<span class="sd">    and it is expected to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        bits (int): number of bits to keep for each channel (0-8)</span>
+<span class="sd">        p (float): probability of the image being posterized. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">bits</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bits</span> <span class="o">=</span> <span class="n">bits</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomPosterize.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomPosterize.html#torchvision.transforms.RandomPosterize.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be posterized.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly posterized image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">posterize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">bits</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(bits=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">bits</span><span class="si">}</span><span class="s2">,p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomSolarize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomSolarize.html#torchvision.transforms.RandomSolarize">[docs]</a><span class="k">class</span> <span class="nc">RandomSolarize</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Solarize the image randomly with a given probability by inverting all pixel</span>
+<span class="sd">    values above a threshold. If img is a Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">    where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        threshold (float): all pixels equal or above this value are inverted.</span>
+<span class="sd">        p (float): probability of the image being solarized. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">threshold</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">threshold</span> <span class="o">=</span> <span class="n">threshold</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomSolarize.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomSolarize.html#torchvision.transforms.RandomSolarize.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be solarized.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly solarized image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">solarize</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">threshold</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(threshold=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">threshold</span><span class="si">}</span><span class="s2">,p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomAdjustSharpness"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness">[docs]</a><span class="k">class</span> <span class="nc">RandomAdjustSharpness</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust the sharpness of the image randomly with a given probability. If the image is torch Tensor,</span>
+<span class="sd">    it is expected to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        sharpness_factor (float):  How much to adjust the sharpness. Can be</span>
+<span class="sd">            any non-negative number. 0 gives a blurred image, 1 gives the</span>
+<span class="sd">            original image while 2 increases the sharpness by a factor of 2.</span>
+<span class="sd">        p (float): probability of the image being sharpened. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sharpness_factor</span> <span class="o">=</span> <span class="n">sharpness_factor</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomAdjustSharpness.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be sharpened.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly sharpened image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_sharpness</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharpness_factor</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(sharpness_factor=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sharpness_factor</span><span class="si">}</span><span class="s2">,p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomAutocontrast"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast">[docs]</a><span class="k">class</span> <span class="nc">RandomAutocontrast</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Autocontrast the pixels of the given image randomly with a given probability.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being autocontrasted. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomAutocontrast.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be autocontrasted.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly autocontrasted image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">autocontrast</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="RandomEqualize"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomEqualize.html#torchvision.transforms.RandomEqualize">[docs]</a><span class="k">class</span> <span class="nc">RandomEqualize</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Equalize the histogram of the given image randomly with a given probability.</span>
+<span class="sd">    If the image is torch Tensor, it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;P&quot;, &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being equalized. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+<div class="viewcode-block" id="RandomEqualize.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.RandomEqualize.html#torchvision.transforms.RandomEqualize.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">img</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            img (PIL Image or Tensor): Image to be equalized.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Randomly equalized image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">equalize</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">img</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">(p=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="si">}</span><span class="s2">)&quot;</span></div>
+
+
+<div class="viewcode-block" id="ElasticTransform"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ElasticTransform.html#torchvision.transforms.ElasticTransform">[docs]</a><span class="k">class</span> <span class="nc">ElasticTransform</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Transform a tensor image with elastic transformations.</span>
+<span class="sd">    Given alpha and sigma, it will generate displacement</span>
+<span class="sd">    vectors for all pixels based on random offsets. Alpha controls the strength</span>
+<span class="sd">    and sigma controls the smoothness of the displacements.</span>
+<span class="sd">    The displacements are added to an identity grid and the resulting grid is</span>
+<span class="sd">    used to grid_sample from the image.</span>
+
+<span class="sd">    Applications:</span>
+<span class="sd">        Randomly transforms the morphology of objects in images and produces a</span>
+<span class="sd">        see-through-water-like effect.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        alpha (float or sequence of floats): Magnitude of displacements. Default is 50.0.</span>
+<span class="sd">        sigma (float or sequence of floats): Smoothness of displacements. Default is 5.0.</span>
+<span class="sd">        interpolation (InterpolationMode): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (sequence or number): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. Default is ``0``. If given a number, the value is used for all bands respectively.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">50.0</span><span class="p">,</span> <span class="n">sigma</span><span class="o">=</span><span class="mf">5.0</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">alpha</span><span class="p">,</span> <span class="p">(</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;alpha should be float or a sequence of floats. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">alpha</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">alpha</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">alpha</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If alpha is a sequence its length should be 2. Got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">alpha</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">alpha</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">for</span> <span class="n">element</span> <span class="ow">in</span> <span class="n">alpha</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="nb">float</span><span class="p">):</span>
+                    <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;alpha should be a sequence of floats. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">element</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">alpha</span><span class="p">,</span> <span class="nb">float</span><span class="p">):</span>
+            <span class="n">alpha</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">alpha</span><span class="p">),</span> <span class="nb">float</span><span class="p">(</span><span class="n">alpha</span><span class="p">)]</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">alpha</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">alpha</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">alpha</span> <span class="o">=</span> <span class="p">[</span><span class="n">alpha</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">alpha</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span> <span class="o">=</span> <span class="n">alpha</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="p">(</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sigma should be float or a sequence of floats. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If sigma is a sequence its length should be 2. Got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">for</span> <span class="n">element</span> <span class="ow">in</span> <span class="n">sigma</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="nb">float</span><span class="p">):</span>
+                    <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sigma should be a sequence of floats. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">element</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="nb">float</span><span class="p">):</span>
+            <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">sigma</span><span class="p">),</span> <span class="nb">float</span><span class="p">(</span><span class="n">sigma</span><span class="p">)]</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span> <span class="o">=</span> <span class="n">sigma</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">interpolation</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+            <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+            <span class="n">fill</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fill</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;fill should be int or float or a list or tuple of them. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">get_params</span><span class="p">(</span><span class="n">alpha</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">dx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">size</span><span class="p">)</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="n">kx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">8</span> <span class="o">*</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="c1"># if kernel size is even we have to make it odd</span>
+            <span class="k">if</span> <span class="n">kx</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">kx</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">dx</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">gaussian_blur</span><span class="p">(</span><span class="n">dx</span><span class="p">,</span> <span class="p">[</span><span class="n">kx</span><span class="p">,</span> <span class="n">kx</span><span class="p">],</span> <span class="n">sigma</span><span class="p">)</span>
+        <span class="n">dx</span> <span class="o">=</span> <span class="n">dx</span> <span class="o">*</span> <span class="n">alpha</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">/</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="n">dy</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">size</span><span class="p">)</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="n">ky</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">8</span> <span class="o">*</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="c1"># if kernel size is even we have to make it odd</span>
+            <span class="k">if</span> <span class="n">ky</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">ky</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">dy</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">gaussian_blur</span><span class="p">(</span><span class="n">dy</span><span class="p">,</span> <span class="p">[</span><span class="n">ky</span><span class="p">,</span> <span class="n">ky</span><span class="p">],</span> <span class="n">sigma</span><span class="p">)</span>
+        <span class="n">dy</span> <span class="o">=</span> <span class="n">dy</span> <span class="o">*</span> <span class="n">alpha</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">/</span> <span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">concat</span><span class="p">([</span><span class="n">dx</span><span class="p">,</span> <span class="n">dy</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">])</span>  <span class="c1"># 1 x H x W x 2</span>
+
+<div class="viewcode-block" id="ElasticTransform.forward"><a class="viewcode-back" href="../../../generated/torchvision.transforms.ElasticTransform.html#torchvision.transforms.ElasticTransform.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tensor</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            tensor (PIL Image or Tensor): Image to be transformed.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            PIL Image or Tensor: Transformed image.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">_</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">(</span><span class="n">tensor</span><span class="p">)</span>
+        <span class="n">displacement</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_params</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">,</span> <span class="p">[</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">])</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">elastic_transform</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="n">displacement</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;(alpha=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, sigma=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, interpolation=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">format_string</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;, fill=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">)&quot;</span>
+        <span class="k">return</span> <span class="n">format_string</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_augment.html b/0.16/_modules/torchvision/transforms/v2/_augment.html
new file mode 100644
index 00000000000..3a47e87f898
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_augment.html
@@ -0,0 +1,994 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._augment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._augment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._augment</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">numbers</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch.nn.functional</span> <span class="kn">import</span> <span class="n">one_hot</span>
+<span class="kn">from</span> <span class="nn">torch.utils._pytree</span> <span class="kn">import</span> <span class="n">tree_flatten</span><span class="p">,</span> <span class="n">tree_unflatten</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span> <span class="k">as</span> <span class="n">_transforms</span><span class="p">,</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+<span class="kn">from</span> <span class="nn">._transform</span> <span class="kn">import</span> <span class="n">_RandomApplyTransform</span><span class="p">,</span> <span class="n">Transform</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_parse_labels_getter</span><span class="p">,</span> <span class="n">has_any</span><span class="p">,</span> <span class="n">is_pure_tensor</span><span class="p">,</span> <span class="n">query_chw</span><span class="p">,</span> <span class="n">query_size</span>
+
+
+<div class="viewcode-block" id="RandomErasing"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomErasing.html#torchvision.transforms.v2.RandomErasing">[docs]</a><span class="k">class</span> <span class="nc">RandomErasing</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Randomly select a rectangle region in the input image or video and erase its pixels.</span>
+
+<span class="sd">    .. v2betastatus:: RandomErasing transform</span>
+
+<span class="sd">    This transform does not support PIL Image.</span>
+<span class="sd">    &#39;Random Erasing Data Augmentation&#39; by Zhong et al. See https://arxiv.org/abs/1708.04896</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float, optional): probability that the random erasing operation will be performed.</span>
+<span class="sd">        scale (tuple of float, optional): range of proportion of erased area against input image.</span>
+<span class="sd">        ratio (tuple of float, optional): range of aspect ratio of erased area.</span>
+<span class="sd">        value (number or tuple of numbers): erasing value. Default is 0. If a single int, it is used to</span>
+<span class="sd">            erase all pixels. If a tuple of length 3, it is used to erase</span>
+<span class="sd">            R, G, B channels respectively.</span>
+<span class="sd">            If a str of &#39;random&#39;, erasing each pixel with random values.</span>
+<span class="sd">        inplace (bool, optional): boolean to make this transform inplace. Default set to False.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        Erased input.</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        &gt;&gt;&gt; from torchvision.transforms import v2 as transforms</span>
+<span class="sd">        &gt;&gt;&gt;</span>
+<span class="sd">        &gt;&gt;&gt; transform = transforms.Compose([</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.RandomHorizontalFlip(),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.PILToTensor(),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.ConvertImageDtype(torch.float),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),</span>
+<span class="sd">        &gt;&gt;&gt;   transforms.RandomErasing(),</span>
+<span class="sd">        &gt;&gt;&gt; ])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomErasing</span>
+
+    <span class="k">def</span> <span class="nf">_extract_params_for_v1_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_extract_params_for_v1_transform</span><span class="p">(),</span>
+            <span class="n">value</span><span class="o">=</span><span class="s2">&quot;random&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">scale</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.02</span><span class="p">,</span> <span class="mf">0.33</span><span class="p">),</span>
+        <span class="n">ratio</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.3</span><span class="p">,</span> <span class="mf">3.3</span><span class="p">),</span>
+        <span class="n">value</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument value should be either a number or str or a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="ow">and</span> <span class="n">value</span> <span class="o">!=</span> <span class="s2">&quot;random&quot;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;If value is str, it should be &#39;random&#39;&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">scale</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Scale should be a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">ratio</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Ratio should be a sequence&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="ow">or</span> <span class="p">(</span><span class="n">ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">]):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scale and ratio should be of kind (min, max)&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Scale should be between 0 and 1&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ratio</span> <span class="o">=</span> <span class="n">ratio</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">value</span><span class="p">)]</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">value</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">value</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span> <span class="o">=</span> <span class="n">inplace</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_log_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_call_kernel</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">functional</span><span class="p">:</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() is currently passing through inputs of type &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;tv_tensors.</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">. This will likely change in the future.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">img_c</span><span class="p">,</span> <span class="n">img_h</span><span class="p">,</span> <span class="n">img_w</span> <span class="o">=</span> <span class="n">query_chw</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">)</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">img_c</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;If value is a sequence, it should have either a single value or </span><span class="si">{</span><span class="n">img_c</span><span class="si">}</span><span class="s2"> (number of inpt channels)&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">area</span> <span class="o">=</span> <span class="n">img_h</span> <span class="o">*</span> <span class="n">img_w</span>
+
+        <span class="n">log_ratio</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_log_ratio</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">):</span>
+            <span class="n">erase_area</span> <span class="o">=</span> <span class="n">area</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="n">aspect_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span>
+                    <span class="n">log_ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">log_ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>  <span class="c1"># type: ignore[arg-type]</span>
+                <span class="p">)</span>
+            <span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+            <span class="n">h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">erase_area</span> <span class="o">*</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+            <span class="n">w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">erase_area</span> <span class="o">/</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">h</span> <span class="o">&lt;</span> <span class="n">img_h</span> <span class="ow">and</span> <span class="n">w</span> <span class="o">&lt;</span> <span class="n">img_w</span><span class="p">):</span>
+                <span class="k">continue</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">([</span><span class="n">img_c</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span><span class="o">.</span><span class="n">normal_</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">value</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+
+            <span class="n">i</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">img_h</span> <span class="o">-</span> <span class="n">h</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="n">j</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">img_w</span> <span class="o">-</span> <span class="n">w</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="k">break</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">img_h</span><span class="p">,</span> <span class="n">img_w</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">i</span><span class="o">=</span><span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="o">=</span><span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="o">=</span><span class="n">w</span><span class="p">,</span> <span class="n">v</span><span class="o">=</span><span class="n">v</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;v&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">erase</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="o">**</span><span class="n">params</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">inpt</span></div>
+
+
+<span class="k">class</span> <span class="nc">_BaseMixUpCutMix</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">labels_getter</span><span class="o">=</span><span class="s2">&quot;default&quot;</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">alpha</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_dist</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">distributions</span><span class="o">.</span><span class="n">Beta</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">alpha</span><span class="p">]),</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">alpha</span><span class="p">]))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span> <span class="o">=</span> <span class="n">num_classes</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels_getter</span> <span class="o">=</span> <span class="n">_parse_labels_getter</span><span class="p">(</span><span class="n">labels_getter</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">):</span>
+        <span class="n">inputs</span> <span class="o">=</span> <span class="n">inputs</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">flat_inputs</span><span class="p">,</span> <span class="n">spec</span> <span class="o">=</span> <span class="n">tree_flatten</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="n">needs_transform_list</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_needs_transform_list</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">has_any</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() does not support PIL images, bounding boxes and masks.&quot;</span><span class="p">)</span>
+
+        <span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels_getter</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">labels</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The labels must be a tensor, but got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span><span class="si">}</span><span class="s2"> instead.&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">labels</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;labels tensor should be of shape (batch_size,) &quot;</span> <span class="sa">f</span><span class="s2">&quot;but got shape </span><span class="si">{</span><span class="n">labels</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2"> instead.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">params</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="n">labels</span><span class="p">,</span>
+            <span class="s2">&quot;batch_size&quot;</span><span class="p">:</span> <span class="n">labels</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+            <span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">_get_params</span><span class="p">(</span>
+                <span class="p">[</span><span class="n">inpt</span> <span class="k">for</span> <span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">needs_transform</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">needs_transform_list</span><span class="p">)</span> <span class="k">if</span> <span class="n">needs_transform</span><span class="p">]</span>
+            <span class="p">),</span>
+        <span class="p">}</span>
+
+        <span class="c1"># By default, the labels will be False inside needs_transform_list, since they are a torch.Tensor coming</span>
+        <span class="c1"># after an image or video. However, we need to handle them in _transform, so we make sure to set them to True</span>
+        <span class="n">needs_transform_list</span><span class="p">[</span><span class="nb">next</span><span class="p">(</span><span class="n">idx</span> <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">inpt</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span> <span class="k">if</span> <span class="n">inpt</span> <span class="ow">is</span> <span class="n">labels</span><span class="p">)]</span> <span class="o">=</span> <span class="kc">True</span>
+        <span class="n">flat_outputs</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_transform</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">params</span><span class="p">)</span> <span class="k">if</span> <span class="n">needs_transform</span> <span class="k">else</span> <span class="n">inpt</span>
+            <span class="k">for</span> <span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">needs_transform</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">needs_transform_list</span><span class="p">)</span>
+        <span class="p">]</span>
+
+        <span class="k">return</span> <span class="n">tree_unflatten</span><span class="p">(</span><span class="n">flat_outputs</span><span class="p">,</span> <span class="n">spec</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_image_or_video</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">batch_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">expected_num_dims</span> <span class="o">=</span> <span class="mi">5</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span> <span class="k">else</span> <span class="mi">4</span>
+        <span class="k">if</span> <span class="n">inpt</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="n">expected_num_dims</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Expected a batched input with </span><span class="si">{</span><span class="n">expected_num_dims</span><span class="si">}</span><span class="s2"> dims, but got </span><span class="si">{</span><span class="n">inpt</span><span class="o">.</span><span class="n">ndim</span><span class="si">}</span><span class="s2"> dimensions instead.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">inpt</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="n">batch_size</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The batch size of the image or video does not match the batch size of the labels: &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">inpt</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2"> != </span><span class="si">{</span><span class="n">batch_size</span><span class="si">}</span><span class="s2">.&quot;</span>
+            <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_mixup_label</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">label</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">lam</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="n">label</span> <span class="o">=</span> <span class="n">one_hot</span><span class="p">(</span><span class="n">label</span><span class="p">,</span> <span class="n">num_classes</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">label</span><span class="o">.</span><span class="n">dtype</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">:</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="n">label</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+        <span class="k">return</span> <span class="n">label</span><span class="o">.</span><span class="n">roll</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">lam</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">label</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">lam</span><span class="p">))</span>
+
+
+<div class="viewcode-block" id="MixUp"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.MixUp.html#torchvision.transforms.v2.MixUp">[docs]</a><span class="k">class</span> <span class="nc">MixUp</span><span class="p">(</span><span class="n">_BaseMixUpCutMix</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Apply MixUp to the provided batch of images and labels.</span>
+
+<span class="sd">    .. v2betastatus:: MixUp transform</span>
+
+<span class="sd">    Paper: `mixup: Beyond Empirical Risk Minimization &lt;https://arxiv.org/abs/1710.09412&gt;`_.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        This transform is meant to be used on **batches** of samples, not</span>
+<span class="sd">        individual images. See</span>
+<span class="sd">        :ref:`sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py` for detailed usage</span>
+<span class="sd">        examples.</span>
+<span class="sd">        The sample pairing is deterministic and done by matching consecutive</span>
+<span class="sd">        samples in the batch, so the batch needs to be shuffled (this is an</span>
+<span class="sd">        implementation detail, not a guaranteed convention.)</span>
+
+<span class="sd">    In the input, the labels are expected to be a tensor of shape ``(batch_size,)``. They will be transformed</span>
+<span class="sd">    into a tensor of shape ``(batch_size, num_classes)``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        alpha (float, optional): hyperparameter of the Beta distribution used for mixup. Default is 1.</span>
+<span class="sd">        num_classes (int): number of classes in the batch. Used for one-hot-encoding.</span>
+<span class="sd">        labels_getter (callable or &quot;default&quot;, optional): indicates how to identify the labels in the input.</span>
+<span class="sd">            By default, this will pick the second parameter as the labels if it&#39;s a tensor. This covers the most</span>
+<span class="sd">            common scenario where this transform is called as ``MixUp()(imgs_batch, labels_batch)``.</span>
+<span class="sd">            It can also be a callable that takes the same input as the transform, and returns the labels.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lam</span><span class="o">=</span><span class="nb">float</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_dist</span><span class="o">.</span><span class="n">sample</span><span class="p">(())))</span>  <span class="c1"># type: ignore[arg-type]</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">lam</span> <span class="o">=</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;lam&quot;</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">inpt</span> <span class="ow">is</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_mixup_label</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">lam</span><span class="o">=</span><span class="n">lam</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">))</span> <span class="ow">or</span> <span class="n">is_pure_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_check_image_or_video</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;batch_size&quot;</span><span class="p">])</span>
+
+            <span class="n">output</span> <span class="o">=</span> <span class="n">inpt</span><span class="o">.</span><span class="n">roll</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">lam</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">inpt</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">lam</span><span class="p">))</span>
+
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)):</span>
+                <span class="n">output</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+            <span class="k">return</span> <span class="n">output</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">inpt</span></div>
+
+
+<div class="viewcode-block" id="CutMix"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.CutMix.html#torchvision.transforms.v2.CutMix">[docs]</a><span class="k">class</span> <span class="nc">CutMix</span><span class="p">(</span><span class="n">_BaseMixUpCutMix</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Apply CutMix to the provided batch of images and labels.</span>
+
+<span class="sd">    .. v2betastatus:: CutMix transform</span>
+
+<span class="sd">    Paper: `CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1905.04899&gt;`_.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        This transform is meant to be used on **batches** of samples, not</span>
+<span class="sd">        individual images. See</span>
+<span class="sd">        :ref:`sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py` for detailed usage</span>
+<span class="sd">        examples.</span>
+<span class="sd">        The sample pairing is deterministic and done by matching consecutive</span>
+<span class="sd">        samples in the batch, so the batch needs to be shuffled (this is an</span>
+<span class="sd">        implementation detail, not a guaranteed convention.)</span>
+
+<span class="sd">    In the input, the labels are expected to be a tensor of shape ``(batch_size,)``. They will be transformed</span>
+<span class="sd">    into a tensor of shape ``(batch_size, num_classes)``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        alpha (float, optional): hyperparameter of the Beta distribution used for mixup. Default is 1.</span>
+<span class="sd">        num_classes (int): number of classes in the batch. Used for one-hot-encoding.</span>
+<span class="sd">        labels_getter (callable or &quot;default&quot;, optional): indicates how to identify the labels in the input.</span>
+<span class="sd">            By default, this will pick the second parameter as the labels if it&#39;s a tensor. This covers the most</span>
+<span class="sd">            common scenario where this transform is called as ``CutMix()(imgs_batch, labels_batch)``.</span>
+<span class="sd">            It can also be a callable that takes the same input as the transform, and returns the labels.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">lam</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_dist</span><span class="o">.</span><span class="n">sample</span><span class="p">(()))</span>  <span class="c1"># type: ignore[arg-type]</span>
+
+        <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="n">r_x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">W</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span>
+        <span class="n">r_y</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">H</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span>
+
+        <span class="n">r</span> <span class="o">=</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">lam</span><span class="p">)</span>
+        <span class="n">r_w_half</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">r</span> <span class="o">*</span> <span class="n">W</span><span class="p">)</span>
+        <span class="n">r_h_half</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">r</span> <span class="o">*</span> <span class="n">H</span><span class="p">)</span>
+
+        <span class="n">x1</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">r_x</span> <span class="o">-</span> <span class="n">r_w_half</span><span class="p">,</span> <span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">))</span>
+        <span class="n">y1</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">r_y</span> <span class="o">-</span> <span class="n">r_h_half</span><span class="p">,</span> <span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">))</span>
+        <span class="n">x2</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">r_x</span> <span class="o">+</span> <span class="n">r_w_half</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">W</span><span class="p">))</span>
+        <span class="n">y2</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">r_y</span> <span class="o">+</span> <span class="n">r_h_half</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">H</span><span class="p">))</span>
+        <span class="n">box</span> <span class="o">=</span> <span class="p">(</span><span class="n">x1</span><span class="p">,</span> <span class="n">y1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">y2</span><span class="p">)</span>
+
+        <span class="n">lam_adjusted</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="p">(</span><span class="n">x2</span> <span class="o">-</span> <span class="n">x1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">y2</span> <span class="o">-</span> <span class="n">y1</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="n">W</span> <span class="o">*</span> <span class="n">H</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">box</span><span class="o">=</span><span class="n">box</span><span class="p">,</span> <span class="n">lam_adjusted</span><span class="o">=</span><span class="n">lam_adjusted</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">inpt</span> <span class="ow">is</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_mixup_label</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">lam</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;lam_adjusted&quot;</span><span class="p">])</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">))</span> <span class="ow">or</span> <span class="n">is_pure_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_check_image_or_video</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;batch_size&quot;</span><span class="p">])</span>
+
+            <span class="n">x1</span><span class="p">,</span> <span class="n">y1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">y2</span> <span class="o">=</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;box&quot;</span><span class="p">]</span>
+            <span class="n">rolled</span> <span class="o">=</span> <span class="n">inpt</span><span class="o">.</span><span class="n">roll</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">inpt</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+            <span class="n">output</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">y1</span><span class="p">:</span><span class="n">y2</span><span class="p">,</span> <span class="n">x1</span><span class="p">:</span><span class="n">x2</span><span class="p">]</span> <span class="o">=</span> <span class="n">rolled</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">y1</span><span class="p">:</span><span class="n">y2</span><span class="p">,</span> <span class="n">x1</span><span class="p">:</span><span class="n">x2</span><span class="p">]</span>
+
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)):</span>
+                <span class="n">output</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+            <span class="k">return</span> <span class="n">output</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">inpt</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_auto_augment.html b/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
new file mode 100644
index 00000000000..338ebfdc46d
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
@@ -0,0 +1,1304 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._auto_augment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._auto_augment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._auto_augment</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">torch.utils._pytree</span> <span class="kn">import</span> <span class="n">tree_flatten</span><span class="p">,</span> <span class="n">tree_unflatten</span><span class="p">,</span> <span class="n">TreeSpec</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span> <span class="k">as</span> <span class="n">_transforms</span><span class="p">,</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">_functional_tensor</span> <span class="k">as</span> <span class="n">_FT</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">AutoAugmentPolicy</span><span class="p">,</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">,</span> <span class="n">Transform</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2.functional._geometry</span> <span class="kn">import</span> <span class="n">_check_interpolation</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2.functional._meta</span> <span class="kn">import</span> <span class="n">get_size</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2.functional._utils</span> <span class="kn">import</span> <span class="n">_FillType</span><span class="p">,</span> <span class="n">_FillTypeJIT</span>
+
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_get_fill</span><span class="p">,</span> <span class="n">_setup_fill_arg</span><span class="p">,</span> <span class="n">check_type</span><span class="p">,</span> <span class="n">is_pure_tensor</span>
+
+
+<span class="n">ImageOrVideo</span> <span class="o">=</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">]</span>
+
+
+<span class="k">class</span> <span class="nc">_AutoAugmentBase</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="o">*</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_extract_params_for_v1_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">params</span> <span class="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_extract_params_for_v1_transform</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;fill&quot;</span><span class="p">],</span> <span class="nb">dict</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() can not be scripted for when `fill` is a dictionary.&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">params</span>
+
+    <span class="k">def</span> <span class="nf">_get_random_item</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dct</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Callable</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Callable</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]:</span>
+        <span class="n">keys</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">dct</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">keys</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">keys</span><span class="p">),</span> <span class="p">()))]</span>
+        <span class="k">return</span> <span class="n">key</span><span class="p">,</span> <span class="n">dct</span><span class="p">[</span><span class="n">key</span><span class="p">]</span>
+
+    <span class="k">def</span> <span class="nf">_flatten_and_extract_image_or_video</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+        <span class="n">unsupported_types</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="o">...</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">),</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">],</span> <span class="n">TreeSpec</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">ImageOrVideo</span><span class="p">]:</span>
+        <span class="n">flat_inputs</span><span class="p">,</span> <span class="n">spec</span> <span class="o">=</span> <span class="n">tree_flatten</span><span class="p">(</span><span class="n">inputs</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">needs_transform_list</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_needs_transform_list</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="n">image_or_videos</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">needs_transform</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">needs_transform_list</span><span class="p">)):</span>
+            <span class="k">if</span> <span class="n">needs_transform</span> <span class="ow">and</span> <span class="n">check_type</span><span class="p">(</span>
+                <span class="n">inpt</span><span class="p">,</span>
+                <span class="p">(</span>
+                    <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+                    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+                    <span class="n">is_pure_tensor</span><span class="p">,</span>
+                    <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">,</span>
+                <span class="p">),</span>
+            <span class="p">):</span>
+                <span class="n">image_or_videos</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">idx</span><span class="p">,</span> <span class="n">inpt</span><span class="p">))</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">unsupported_types</span><span class="p">):</span>
+                <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Inputs of type </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2"> are not supported by </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">()&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">image_or_videos</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Found no image in the sample.&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">image_or_videos</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Auto augment transformations are only properly defined for a single image or video, &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;but found </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">image_or_videos</span><span class="p">)</span><span class="si">}</span><span class="s2">.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">idx</span><span class="p">,</span> <span class="n">image_or_video</span> <span class="o">=</span> <span class="n">image_or_videos</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">return</span> <span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">spec</span><span class="p">,</span> <span class="n">idx</span><span class="p">),</span> <span class="n">image_or_video</span>
+
+    <span class="k">def</span> <span class="nf">_unflatten_and_insert_image_or_video</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">flat_inputs_with_spec</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">],</span> <span class="n">TreeSpec</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">image_or_video</span><span class="p">:</span> <span class="n">ImageOrVideo</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">flat_inputs</span><span class="p">,</span> <span class="n">spec</span><span class="p">,</span> <span class="n">idx</span> <span class="o">=</span> <span class="n">flat_inputs_with_spec</span>
+        <span class="n">flat_inputs</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span> <span class="o">=</span> <span class="n">image_or_video</span>
+        <span class="k">return</span> <span class="n">tree_unflatten</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">spec</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_apply_image_or_video_transform</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">image</span><span class="p">:</span> <span class="n">ImageOrVideo</span><span class="p">,</span>
+        <span class="n">transform_id</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">magnitude</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillTypeJIT</span><span class="p">],</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ImageOrVideo</span><span class="p">:</span>
+        <span class="n">fill_</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">image</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Identity&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span>
+            <span class="c1"># magnitude should be arctan(magnitude)</span>
+            <span class="c1"># official autoaug: (1, level, 0, 0, 1, 0)</span>
+            <span class="c1"># https://github.com/tensorflow/models/blob/dd02069717128186b88afa8d857ce57d17957f03/research/autoaugment/augmentation_transforms.py#L290</span>
+            <span class="c1"># compared to</span>
+            <span class="c1"># torchvision:      (1, tan(level), 0, 0, 1, 0)</span>
+            <span class="c1"># https://github.com/pytorch/vision/blob/0c2373d0bba3499e95776e7936e207d8a1676e65/torchvision/transforms/functional.py#L976</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+                <span class="n">image</span><span class="p">,</span>
+                <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+                <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+                <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="n">math</span><span class="o">.</span><span class="n">degrees</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">magnitude</span><span class="p">)),</span> <span class="mf">0.0</span><span class="p">],</span>
+                <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+                <span class="n">fill</span><span class="o">=</span><span class="n">fill_</span><span class="p">,</span>
+                <span class="n">center</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+            <span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span>
+            <span class="c1"># magnitude should be arctan(magnitude)</span>
+            <span class="c1"># See above</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+                <span class="n">image</span><span class="p">,</span>
+                <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+                <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+                <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">math</span><span class="o">.</span><span class="n">degrees</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">atan</span><span class="p">(</span><span class="n">magnitude</span><span class="p">))],</span>
+                <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+                <span class="n">fill</span><span class="o">=</span><span class="n">fill_</span><span class="p">,</span>
+                <span class="n">center</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span>
+            <span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+                <span class="n">image</span><span class="p">,</span>
+                <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">magnitude</span><span class="p">),</span> <span class="mi">0</span><span class="p">],</span>
+                <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+                <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+                <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span>
+                <span class="n">fill</span><span class="o">=</span><span class="n">fill_</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span>
+                <span class="n">image</span><span class="p">,</span>
+                <span class="n">angle</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">magnitude</span><span class="p">)],</span>
+                <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+                <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+                <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span>
+                <span class="n">fill</span><span class="o">=</span><span class="n">fill_</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">rotate</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">angle</span><span class="o">=</span><span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill_</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Color&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="o">=</span><span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">adjust_sharpness</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="o">=</span><span class="mf">1.0</span> <span class="o">+</span> <span class="n">magnitude</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">posterize</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">bits</span><span class="o">=</span><span class="nb">int</span><span class="p">(</span><span class="n">magnitude</span><span class="p">))</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span>
+            <span class="n">bound</span> <span class="o">=</span> <span class="n">_FT</span><span class="o">.</span><span class="n">_max_value</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="k">else</span> <span class="mf">255.0</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">solarize</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">threshold</span><span class="o">=</span><span class="n">bound</span> <span class="o">*</span> <span class="n">magnitude</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">autocontrast</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">equalize</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">transform_id</span> <span class="o">==</span> <span class="s2">&quot;Invert&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">invert</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;No transform available for </span><span class="si">{</span><span class="n">transform_id</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="AutoAugment"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.AutoAugment.html#torchvision.transforms.v2.AutoAugment">[docs]</a><span class="k">class</span> <span class="nc">AutoAugment</span><span class="p">(</span><span class="n">_AutoAugmentBase</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;[BETA] AutoAugment data augmentation method based on</span>
+<span class="sd">    `&quot;AutoAugment: Learning Augmentation Strategies from Data&quot; &lt;https://arxiv.org/pdf/1805.09501.pdf&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: AutoAugment transform</span>
+
+<span class="sd">    This transformation works on images and videos only.</span>
+
+<span class="sd">    If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        policy (AutoAugmentPolicy, optional): Desired policy enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.autoaugment.AutoAugmentPolicy`. Default is ``AutoAugmentPolicy.IMAGENET``.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">AutoAugment</span>
+
+    <span class="n">_AUGMENTATION_SPACE</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">width</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span>
+            <span class="kc">True</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">height</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span>
+            <span class="kc">True</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">30.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="p">(</span><span class="mi">8</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">4</span><span class="p">)))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span>
+            <span class="kc">False</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;Invert&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">policy</span><span class="p">:</span> <span class="n">AutoAugmentPolicy</span> <span class="o">=</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">IMAGENET</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">policy</span> <span class="o">=</span> <span class="n">policy</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_policies</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_policies</span><span class="p">(</span><span class="n">policy</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_policies</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">policy</span><span class="p">:</span> <span class="n">AutoAugmentPolicy</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]]:</span>
+        <span class="k">if</span> <span class="n">policy</span> <span class="o">==</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">IMAGENET</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">[</span>
+                <span class="p">((</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mi">2</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+            <span class="p">]</span>
+        <span class="k">elif</span> <span class="n">policy</span> <span class="o">==</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">CIFAR10</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">[</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateX&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Posterize&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">5</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateX&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Sharpness&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Brightness&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Color&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">0</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+            <span class="p">]</span>
+        <span class="k">elif</span> <span class="n">policy</span> <span class="o">==</span> <span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">SVHN</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">[</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">6</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Equalize&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Contrast&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mi">2</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.4</span><span class="p">,</span> <span class="mi">8</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Rotate&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">4</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateX&quot;</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;Solarize&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.6</span><span class="p">,</span> <span class="mi">7</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;TranslateY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">3</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearY&quot;</span><span class="p">,</span> <span class="mf">0.8</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;AutoContrast&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+                <span class="p">((</span><span class="s2">&quot;ShearX&quot;</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="s2">&quot;Invert&quot;</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="kc">None</span><span class="p">)),</span>
+            <span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The provided policy </span><span class="si">{</span><span class="n">policy</span><span class="si">}</span><span class="s2"> is not recognized.&quot;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="AutoAugment.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.AutoAugment.html#torchvision.transforms.v2.AutoAugment.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">image_or_video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_flatten_and_extract_image_or_video</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">get_size</span><span class="p">(</span><span class="n">image_or_video</span><span class="p">)</span>
+
+        <span class="n">policy</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_policies</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_policies</span><span class="p">),</span> <span class="p">()))]</span>
+
+        <span class="k">for</span> <span class="n">transform_id</span><span class="p">,</span> <span class="n">probability</span><span class="p">,</span> <span class="n">magnitude_idx</span> <span class="ow">in</span> <span class="n">policy</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(())</span> <span class="o">&lt;=</span> <span class="n">probability</span><span class="p">:</span>
+                <span class="k">continue</span>
+
+            <span class="n">magnitudes_fn</span><span class="p">,</span> <span class="n">signed</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_AUGMENTATION_SPACE</span><span class="p">[</span><span class="n">transform_id</span><span class="p">]</span>
+
+            <span class="n">magnitudes</span> <span class="o">=</span> <span class="n">magnitudes_fn</span><span class="p">(</span><span class="mi">10</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">magnitudes</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">magnitude</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="n">magnitude_idx</span><span class="p">])</span>
+                <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(())</span> <span class="o">&lt;=</span> <span class="mf">0.5</span><span class="p">:</span>
+                    <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mi">1</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">magnitude</span> <span class="o">=</span> <span class="mf">0.0</span>
+
+            <span class="n">image_or_video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_apply_image_or_video_transform</span><span class="p">(</span>
+                <span class="n">image_or_video</span><span class="p">,</span> <span class="n">transform_id</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_unflatten_and_insert_image_or_video</span><span class="p">(</span><span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">image_or_video</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="RandAugment"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandAugment.html#torchvision.transforms.v2.RandAugment">[docs]</a><span class="k">class</span> <span class="nc">RandAugment</span><span class="p">(</span><span class="n">_AutoAugmentBase</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;[BETA] RandAugment data augmentation method based on</span>
+<span class="sd">    `&quot;RandAugment: Practical automated data augmentation with a reduced search space&quot;</span>
+<span class="sd">    &lt;https://arxiv.org/abs/1909.13719&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: RandAugment transform</span>
+
+<span class="sd">    This transformation works on images and videos only.</span>
+
+<span class="sd">    If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_ops (int, optional): Number of augmentation transformations to apply sequentially.</span>
+<span class="sd">        magnitude (int, optional): Magnitude for all the transformations.</span>
+<span class="sd">        num_magnitude_bins (int, optional): The number of different magnitude values.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandAugment</span>
+    <span class="n">_AUGMENTATION_SPACE</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;Identity&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">width</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span>
+            <span class="kc">True</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">150.0</span> <span class="o">/</span> <span class="mf">331.0</span> <span class="o">*</span> <span class="n">height</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span>
+            <span class="kc">True</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">30.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="p">(</span><span class="mi">8</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">4</span><span class="p">)))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span>
+            <span class="kc">False</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_ops</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">2</span><span class="p">,</span>
+        <span class="n">magnitude</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">9</span><span class="p">,</span>
+        <span class="n">num_magnitude_bins</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">31</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_ops</span> <span class="o">=</span> <span class="n">num_ops</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">magnitude</span> <span class="o">=</span> <span class="n">magnitude</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span> <span class="o">=</span> <span class="n">num_magnitude_bins</span>
+
+<div class="viewcode-block" id="RandAugment.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandAugment.html#torchvision.transforms.v2.RandAugment.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">image_or_video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_flatten_and_extract_image_or_video</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">get_size</span><span class="p">(</span><span class="n">image_or_video</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_ops</span><span class="p">):</span>
+            <span class="n">transform_id</span><span class="p">,</span> <span class="p">(</span><span class="n">magnitudes_fn</span><span class="p">,</span> <span class="n">signed</span><span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_random_item</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_AUGMENTATION_SPACE</span><span class="p">)</span>
+            <span class="n">magnitudes</span> <span class="o">=</span> <span class="n">magnitudes_fn</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">magnitudes</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">magnitude</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">magnitude</span><span class="p">])</span>
+                <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(())</span> <span class="o">&lt;=</span> <span class="mf">0.5</span><span class="p">:</span>
+                    <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mi">1</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">magnitude</span> <span class="o">=</span> <span class="mf">0.0</span>
+            <span class="n">image_or_video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_apply_image_or_video_transform</span><span class="p">(</span>
+                <span class="n">image_or_video</span><span class="p">,</span> <span class="n">transform_id</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_unflatten_and_insert_image_or_video</span><span class="p">(</span><span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">image_or_video</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="TrivialAugmentWide"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.TrivialAugmentWide.html#torchvision.transforms.v2.TrivialAugmentWide">[docs]</a><span class="k">class</span> <span class="nc">TrivialAugmentWide</span><span class="p">(</span><span class="n">_AutoAugmentBase</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;[BETA] Dataset-independent data-augmentation with TrivialAugment Wide, as described in</span>
+<span class="sd">    `&quot;TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation&quot; &lt;https://arxiv.org/abs/2103.10158&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: TrivialAugmentWide transform</span>
+
+<span class="sd">    This transformation works on images and videos only.</span>
+
+<span class="sd">    If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_magnitude_bins (int, optional): The number of different magnitude values.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">TrivialAugmentWide</span>
+    <span class="n">_AUGMENTATION_SPACE</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;Identity&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">32.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">32.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">135.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="p">(</span><span class="mi">8</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">6</span><span class="p">)))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span>
+            <span class="kc">False</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">num_magnitude_bins</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">31</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span> <span class="o">=</span> <span class="n">num_magnitude_bins</span>
+
+<div class="viewcode-block" id="TrivialAugmentWide.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.TrivialAugmentWide.html#torchvision.transforms.v2.TrivialAugmentWide.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">image_or_video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_flatten_and_extract_image_or_video</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">get_size</span><span class="p">(</span><span class="n">image_or_video</span><span class="p">)</span>
+
+        <span class="n">transform_id</span><span class="p">,</span> <span class="p">(</span><span class="n">magnitudes_fn</span><span class="p">,</span> <span class="n">signed</span><span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_random_item</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_AUGMENTATION_SPACE</span><span class="p">)</span>
+
+        <span class="n">magnitudes</span> <span class="o">=</span> <span class="n">magnitudes_fn</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">magnitudes</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">magnitude</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_magnitude_bins</span><span class="p">,</span> <span class="p">()))])</span>
+            <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(())</span> <span class="o">&lt;=</span> <span class="mf">0.5</span><span class="p">:</span>
+                <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mi">1</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">magnitude</span> <span class="o">=</span> <span class="mf">0.0</span>
+
+        <span class="n">image_or_video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_apply_image_or_video_transform</span><span class="p">(</span>
+            <span class="n">image_or_video</span><span class="p">,</span> <span class="n">transform_id</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_unflatten_and_insert_image_or_video</span><span class="p">(</span><span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">image_or_video</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="AugMix"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.AugMix.html#torchvision.transforms.v2.AugMix">[docs]</a><span class="k">class</span> <span class="nc">AugMix</span><span class="p">(</span><span class="n">_AutoAugmentBase</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;[BETA] AugMix data augmentation method based on</span>
+<span class="sd">    `&quot;AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty&quot; &lt;https://arxiv.org/abs/1912.02781&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: AugMix transform</span>
+
+<span class="sd">    This transformation works on images and videos only.</span>
+
+<span class="sd">    If the input is :class:`torch.Tensor`, it should be of type ``torch.uint8``, and it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        severity (int, optional): The severity of base augmentation operators. Default is ``3``.</span>
+<span class="sd">        mixture_width (int, optional): The number of augmentation chains. Default is ``3``.</span>
+<span class="sd">        chain_depth (int, optional): The depth of augmentation chains. A negative value denotes stochastic depth sampled from the interval [1, 3].</span>
+<span class="sd">            Default is ``-1``.</span>
+<span class="sd">        alpha (float, optional): The hyperparameter for the probability distributions. Default is ``1.0``.</span>
+<span class="sd">        all_ops (bool, optional): Use all operations (including brightness, contrast, color and sharpness). Default is ``True``.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">        fill (sequence or number, optional): Pixel fill value for the area outside the transformed</span>
+<span class="sd">            image. If given a number, the value is used for all bands respectively.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">AugMix</span>
+
+    <span class="n">_PARTIAL_AUGMENTATION_SPACE</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;ShearX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;ShearY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;TranslateX&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">width</span> <span class="o">/</span> <span class="mf">3.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;TranslateY&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">height</span> <span class="o">/</span> <span class="mf">3.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Rotate&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">30.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Posterize&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="p">(</span><span class="mi">4</span> <span class="o">-</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">num_bins</span><span class="p">)</span> <span class="o">/</span> <span class="p">((</span><span class="n">num_bins</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">4</span><span class="p">)))</span><span class="o">.</span><span class="n">round</span><span class="p">()</span><span class="o">.</span><span class="n">int</span><span class="p">(),</span>
+            <span class="kc">False</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="s2">&quot;Solarize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;AutoContrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+        <span class="s2">&quot;Equalize&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
+    <span class="p">}</span>
+    <span class="n">_AUGMENTATION_SPACE</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]],</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="o">**</span><span class="n">_PARTIAL_AUGMENTATION_SPACE</span><span class="p">,</span>
+        <span class="s2">&quot;Brightness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Color&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Contrast&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+        <span class="s2">&quot;Sharpness&quot;</span><span class="p">:</span> <span class="p">(</span><span class="k">lambda</span> <span class="n">num_bins</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="n">num_bins</span><span class="p">),</span> <span class="kc">True</span><span class="p">),</span>
+    <span class="p">}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">severity</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">mixture_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+        <span class="n">chain_depth</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">all_ops</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span> <span class="o">=</span> <span class="mi">10</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">severity</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The severity must be between [1, </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span><span class="si">}</span><span class="s2">]. Got </span><span class="si">{</span><span class="n">severity</span><span class="si">}</span><span class="s2"> instead.&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">severity</span> <span class="o">=</span> <span class="n">severity</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mixture_width</span> <span class="o">=</span> <span class="n">mixture_width</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">chain_depth</span> <span class="o">=</span> <span class="n">chain_depth</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span> <span class="o">=</span> <span class="n">alpha</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">all_ops</span> <span class="o">=</span> <span class="n">all_ops</span>
+
+    <span class="k">def</span> <span class="nf">_sample_dirichlet</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># Must be on a separate method so that we can overwrite it in tests.</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">_sample_dirichlet</span><span class="p">(</span><span class="n">params</span><span class="p">)</span>
+
+<div class="viewcode-block" id="AugMix.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.AugMix.html#torchvision.transforms.v2.AugMix.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">orig_image_or_video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_flatten_and_extract_image_or_video</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">get_size</span><span class="p">(</span><span class="n">orig_image_or_video</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">orig_image_or_video</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="n">image_or_video</span> <span class="o">=</span> <span class="n">orig_image_or_video</span>
+        <span class="k">else</span><span class="p">:</span>  <span class="c1"># isinstance(inpt, PIL.Image.Image):</span>
+            <span class="n">image_or_video</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">orig_image_or_video</span><span class="p">)</span>
+
+        <span class="n">augmentation_space</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_AUGMENTATION_SPACE</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">all_ops</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">_PARTIAL_AUGMENTATION_SPACE</span>
+
+        <span class="n">orig_dims</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">image_or_video</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
+        <span class="n">expected_ndim</span> <span class="o">=</span> <span class="mi">5</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">orig_image_or_video</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span> <span class="k">else</span> <span class="mi">4</span>
+        <span class="n">batch</span> <span class="o">=</span> <span class="n">image_or_video</span><span class="o">.</span><span class="n">reshape</span><span class="p">([</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="n">expected_ndim</span> <span class="o">-</span> <span class="n">image_or_video</span><span class="o">.</span><span class="n">ndim</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="o">+</span> <span class="n">orig_dims</span><span class="p">)</span>
+        <span class="n">batch_dims</span> <span class="o">=</span> <span class="p">[</span><span class="n">batch</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)]</span> <span class="o">+</span> <span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="n">batch</span><span class="o">.</span><span class="n">ndim</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="c1"># Sample the beta weights for combining the original and augmented image or video. To get Beta, we use a</span>
+        <span class="c1"># Dirichlet with 2 parameters. The 1st column stores the weights of the original and the 2nd the ones of</span>
+        <span class="c1"># augmented image or video.</span>
+        <span class="n">m</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_dirichlet</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">batch</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="c1"># Sample the mixing weights and combine them with the ones sampled from Beta for the augmented images or videos.</span>
+        <span class="n">combined_weights</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sample_dirichlet</span><span class="p">(</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">mixture_width</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">batch</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="p">)</span> <span class="o">*</span> <span class="n">m</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">([</span><span class="n">batch_dims</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">])</span>
+
+        <span class="n">mix</span> <span class="o">=</span> <span class="n">m</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">)</span> <span class="o">*</span> <span class="n">batch</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mixture_width</span><span class="p">):</span>
+            <span class="n">aug</span> <span class="o">=</span> <span class="n">batch</span>
+            <span class="n">depth</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">chain_depth</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">chain_depth</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="k">else</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">low</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">high</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+            <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">depth</span><span class="p">):</span>
+                <span class="n">transform_id</span><span class="p">,</span> <span class="p">(</span><span class="n">magnitudes_fn</span><span class="p">,</span> <span class="n">signed</span><span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_random_item</span><span class="p">(</span><span class="n">augmentation_space</span><span class="p">)</span>
+
+                <span class="n">magnitudes</span> <span class="o">=</span> <span class="n">magnitudes_fn</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_PARAMETER_MAX</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">magnitudes</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">magnitude</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">magnitudes</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">severity</span><span class="p">,</span> <span class="p">()))])</span>
+                    <span class="k">if</span> <span class="n">signed</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(())</span> <span class="o">&lt;=</span> <span class="mf">0.5</span><span class="p">:</span>
+                        <span class="n">magnitude</span> <span class="o">*=</span> <span class="o">-</span><span class="mi">1</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">magnitude</span> <span class="o">=</span> <span class="mf">0.0</span>
+
+                <span class="n">aug</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_apply_image_or_video_transform</span><span class="p">(</span>
+                    <span class="n">aug</span><span class="p">,</span> <span class="n">transform_id</span><span class="p">,</span> <span class="n">magnitude</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span>
+                <span class="p">)</span>
+            <span class="n">mix</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">combined_weights</span><span class="p">[:,</span> <span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">batch_dims</span><span class="p">)</span> <span class="o">*</span> <span class="n">aug</span><span class="p">)</span>
+        <span class="n">mix</span> <span class="o">=</span> <span class="n">mix</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">orig_dims</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">image_or_video</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">orig_image_or_video</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)):</span>
+            <span class="n">mix</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">mix</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">orig_image_or_video</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">orig_image_or_video</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+            <span class="n">mix</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">to_pil_image</span><span class="p">(</span><span class="n">mix</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_unflatten_and_insert_image_or_video</span><span class="p">(</span><span class="n">flat_inputs_with_spec</span><span class="p">,</span> <span class="n">mix</span><span class="p">)</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_color.html b/0.16/_modules/torchvision/transforms/v2/_color.html
new file mode 100644
index 00000000000..c4da81c3969
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_color.html
@@ -0,0 +1,1049 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._color &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._color</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._color</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">collections.abc</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span> <span class="k">as</span> <span class="n">_transforms</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">Transform</span>
+
+<span class="kn">from</span> <span class="nn">._transform</span> <span class="kn">import</span> <span class="n">_RandomApplyTransform</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">query_chw</span>
+
+
+<div class="viewcode-block" id="Grayscale"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Grayscale.html#torchvision.transforms.v2.Grayscale">[docs]</a><span class="k">class</span> <span class="nc">Grayscale</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Convert images or videos to grayscale.</span>
+
+<span class="sd">    .. v2betastatus:: Grayscale transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor`, it is expected</span>
+<span class="sd">    to have [..., 3 or 1, H, W] shape, where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_output_channels (int): (1 or 3) number of channels desired for output image</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">Grayscale</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_output_channels</span> <span class="o">=</span> <span class="n">num_output_channels</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">rgb_to_grayscale</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">num_output_channels</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomGrayscale"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomGrayscale.html#torchvision.transforms.v2.RandomGrayscale">[docs]</a><span class="k">class</span> <span class="nc">RandomGrayscale</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Randomly convert image or videos to grayscale with a probability of p (default 0.1).</span>
+
+<span class="sd">    .. v2betastatus:: RandomGrayscale transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor`, it is expected to have [..., 3 or 1, H, W] shape,</span>
+<span class="sd">    where ... means an arbitrary number of leading dimensions</span>
+
+<span class="sd">    The output has the same number of channels as the input.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability that image should be converted to grayscale.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomGrayscale</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">num_input_channels</span><span class="p">,</span> <span class="o">*</span><span class="n">_</span> <span class="o">=</span> <span class="n">query_chw</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">num_input_channels</span><span class="o">=</span><span class="n">num_input_channels</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">rgb_to_grayscale</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;num_input_channels&quot;</span><span class="p">])</span></div>
+
+
+<div class="viewcode-block" id="ColorJitter"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ColorJitter.html#torchvision.transforms.v2.ColorJitter">[docs]</a><span class="k">class</span> <span class="nc">ColorJitter</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Randomly change the brightness, contrast, saturation and hue of an image or video.</span>
+
+<span class="sd">    .. v2betastatus:: ColorJitter transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor`, it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, mode &quot;1&quot;, &quot;I&quot;, &quot;F&quot; and modes with transparency (alpha channel) are not supported.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        brightness (float or tuple of float (min, max)): How much to jitter brightness.</span>
+<span class="sd">            brightness_factor is chosen uniformly from [max(0, 1 - brightness), 1 + brightness]</span>
+<span class="sd">            or the given [min, max]. Should be non negative numbers.</span>
+<span class="sd">        contrast (float or tuple of float (min, max)): How much to jitter contrast.</span>
+<span class="sd">            contrast_factor is chosen uniformly from [max(0, 1 - contrast), 1 + contrast]</span>
+<span class="sd">            or the given [min, max]. Should be non-negative numbers.</span>
+<span class="sd">        saturation (float or tuple of float (min, max)): How much to jitter saturation.</span>
+<span class="sd">            saturation_factor is chosen uniformly from [max(0, 1 - saturation), 1 + saturation]</span>
+<span class="sd">            or the given [min, max]. Should be non negative numbers.</span>
+<span class="sd">        hue (float or tuple of float (min, max)): How much to jitter hue.</span>
+<span class="sd">            hue_factor is chosen uniformly from [-hue, hue] or the given [min, max].</span>
+<span class="sd">            Should have 0&lt;= hue &lt;= 0.5 or -0.5 &lt;= min &lt;= max &lt;= 0.5.</span>
+<span class="sd">            To jitter hue, the pixel values of the input image has to be non-negative for conversion to HSV space;</span>
+<span class="sd">            thus it does not work if you normalize your image to an interval with negative values,</span>
+<span class="sd">            or use an interpolation that generates negative values before using this function.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">ColorJitter</span>
+
+    <span class="k">def</span> <span class="nf">_extract_params_for_v1_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="p">{</span><span class="n">attr</span><span class="p">:</span> <span class="n">value</span> <span class="ow">or</span> <span class="mi">0</span> <span class="k">for</span> <span class="n">attr</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_extract_params_for_v1_transform</span><span class="p">()</span><span class="o">.</span><span class="n">items</span><span class="p">()}</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">brightness</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">contrast</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">saturation</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">hue</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">brightness</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">brightness</span><span class="p">,</span> <span class="s2">&quot;brightness&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">contrast</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">contrast</span><span class="p">,</span> <span class="s2">&quot;contrast&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">saturation</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">saturation</span><span class="p">,</span> <span class="s2">&quot;saturation&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hue</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_check_input</span><span class="p">(</span><span class="n">hue</span><span class="p">,</span> <span class="s2">&quot;hue&quot;</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">),</span> <span class="n">clip_first_on_zero</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_input</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">value</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]],</span>
+        <span class="n">name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">center</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">bound</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)),</span>
+        <span class="n">clip_first_on_zero</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]:</span>
+        <span class="k">if</span> <span class="n">value</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+            <span class="k">if</span> <span class="n">value</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If </span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> is a single number, it must be non negative.&quot;</span><span class="p">)</span>
+            <span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="n">center</span> <span class="o">-</span> <span class="n">value</span><span class="p">,</span> <span class="n">center</span> <span class="o">+</span> <span class="n">value</span><span class="p">]</span>
+            <span class="k">if</span> <span class="n">clip_first_on_zero</span><span class="p">:</span>
+                <span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mf">0.0</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="n">collections</span><span class="o">.</span><span class="n">abc</span><span class="o">.</span><span class="n">Sequence</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">value</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">value</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">value</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">=</span><span class="si">{</span><span class="n">value</span><span class="si">}</span><span class="s2"> should be a single number or a sequence with length 2.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">bound</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">value</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">bound</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2"> values should be between </span><span class="si">{</span><span class="n">bound</span><span class="si">}</span><span class="s2">, but got </span><span class="si">{</span><span class="n">value</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="kc">None</span> <span class="k">if</span> <span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">value</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="n">center</span> <span class="k">else</span> <span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">value</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="nb">float</span><span class="p">(</span><span class="n">value</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">_generate_value</span><span class="p">(</span><span class="n">left</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">right</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">float</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="n">left</span><span class="p">,</span> <span class="n">right</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">fn_idx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randperm</span><span class="p">(</span><span class="mi">4</span><span class="p">)</span>
+
+        <span class="n">b</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">brightness</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">_generate_value</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">brightness</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">brightness</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="n">c</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">contrast</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">_generate_value</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">contrast</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">contrast</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">saturation</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">_generate_value</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">saturation</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">saturation</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="n">h</span> <span class="o">=</span> <span class="kc">None</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hue</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">_generate_value</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">hue</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">hue</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">fn_idx</span><span class="o">=</span><span class="n">fn_idx</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="n">b</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="n">c</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="o">=</span><span class="n">s</span><span class="p">,</span> <span class="n">hue_factor</span><span class="o">=</span><span class="n">h</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">inpt</span>
+        <span class="n">brightness_factor</span> <span class="o">=</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;brightness_factor&quot;</span><span class="p">]</span>
+        <span class="n">contrast_factor</span> <span class="o">=</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_factor&quot;</span><span class="p">]</span>
+        <span class="n">saturation_factor</span> <span class="o">=</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;saturation_factor&quot;</span><span class="p">]</span>
+        <span class="n">hue_factor</span> <span class="o">=</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;hue_factor&quot;</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">fn_id</span> <span class="ow">in</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;fn_idx&quot;</span><span class="p">]:</span>
+            <span class="k">if</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">brightness_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">,</span> <span class="n">output</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="n">brightness_factor</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">contrast_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="n">output</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="n">contrast_factor</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">saturation_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">,</span> <span class="n">output</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="o">=</span><span class="n">saturation_factor</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">fn_id</span> <span class="o">==</span> <span class="mi">3</span> <span class="ow">and</span> <span class="n">hue_factor</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_hue</span><span class="p">,</span> <span class="n">output</span><span class="p">,</span> <span class="n">hue_factor</span><span class="o">=</span><span class="n">hue_factor</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="RandomChannelPermutation"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomChannelPermutation.html#torchvision.transforms.v2.RandomChannelPermutation">[docs]</a><span class="k">class</span> <span class="nc">RandomChannelPermutation</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Randomly permute the channels of an image or video</span>
+
+<span class="sd">    .. v2betastatus:: RandomChannelPermutation transform</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">num_channels</span><span class="p">,</span> <span class="o">*</span><span class="n">_</span> <span class="o">=</span> <span class="n">query_chw</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">permutation</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">randperm</span><span class="p">(</span><span class="n">num_channels</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">permute_channels</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;permutation&quot;</span><span class="p">])</span></div>
+
+
+<div class="viewcode-block" id="RandomPhotometricDistort"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomPhotometricDistort.html#torchvision.transforms.v2.RandomPhotometricDistort">[docs]</a><span class="k">class</span> <span class="nc">RandomPhotometricDistort</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Randomly distorts the image or video as used in `SSD: Single Shot</span>
+<span class="sd">    MultiBox Detector &lt;https://arxiv.org/abs/1512.02325&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: RandomPhotometricDistort transform</span>
+
+<span class="sd">    This transform relies on :class:`~torchvision.transforms.v2.ColorJitter`</span>
+<span class="sd">    under the hood to adjust the contrast, saturation, hue, brightness, and also</span>
+<span class="sd">    randomly permutes channels.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        brightness (tuple of float (min, max), optional): How much to jitter brightness.</span>
+<span class="sd">            brightness_factor is chosen uniformly from [min, max]. Should be non negative numbers.</span>
+<span class="sd">        contrast tuple of float (min, max), optional): How much to jitter contrast.</span>
+<span class="sd">            contrast_factor is chosen uniformly from [min, max]. Should be non-negative numbers.</span>
+<span class="sd">        saturation (tuple of float (min, max), optional): How much to jitter saturation.</span>
+<span class="sd">            saturation_factor is chosen uniformly from [min, max]. Should be non negative numbers.</span>
+<span class="sd">        hue (tuple of float (min, max), optional): How much to jitter hue.</span>
+<span class="sd">            hue_factor is chosen uniformly from [min, max].  Should have -0.5 &lt;= min &lt;= max &lt;= 0.5.</span>
+<span class="sd">            To jitter hue, the pixel values of the input image has to be non-negative for conversion to HSV space;</span>
+<span class="sd">            thus it does not work if you normalize your image to an interval with negative values,</span>
+<span class="sd">            or use an interpolation that generates negative values before using this function.</span>
+<span class="sd">        p (float, optional) probability each distortion operation (contrast, saturation, ...) to be applied.</span>
+<span class="sd">            Default is 0.5.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">brightness</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.875</span><span class="p">,</span> <span class="mf">1.125</span><span class="p">),</span>
+        <span class="n">contrast</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">1.5</span><span class="p">),</span>
+        <span class="n">saturation</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">1.5</span><span class="p">),</span>
+        <span class="n">hue</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="o">-</span><span class="mf">0.05</span><span class="p">,</span> <span class="mf">0.05</span><span class="p">),</span>
+        <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">brightness</span> <span class="o">=</span> <span class="n">brightness</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">contrast</span> <span class="o">=</span> <span class="n">contrast</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hue</span> <span class="o">=</span> <span class="n">hue</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">saturation</span> <span class="o">=</span> <span class="n">saturation</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">num_channels</span><span class="p">,</span> <span class="o">*</span><span class="n">_</span> <span class="o">=</span> <span class="n">query_chw</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+        <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="n">key</span><span class="p">:</span> <span class="n">ColorJitter</span><span class="o">.</span><span class="n">_generate_value</span><span class="p">(</span><span class="nb">range</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="nb">range</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="k">else</span> <span class="kc">None</span>
+            <span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="nb">range</span> <span class="ow">in</span> <span class="p">[</span>
+                <span class="p">(</span><span class="s2">&quot;brightness_factor&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">brightness</span><span class="p">),</span>
+                <span class="p">(</span><span class="s2">&quot;contrast_factor&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">contrast</span><span class="p">),</span>
+                <span class="p">(</span><span class="s2">&quot;saturation_factor&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">saturation</span><span class="p">),</span>
+                <span class="p">(</span><span class="s2">&quot;hue_factor&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">hue</span><span class="p">),</span>
+            <span class="p">]</span>
+        <span class="p">}</span>
+        <span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_before&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">bool</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(())</span> <span class="o">&lt;</span> <span class="mf">0.5</span><span class="p">)</span>
+        <span class="n">params</span><span class="p">[</span><span class="s2">&quot;channel_permutation&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randperm</span><span class="p">(</span><span class="n">num_channels</span><span class="p">)</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="k">else</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">params</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;brightness_factor&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;brightness_factor&quot;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_factor&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_before&quot;</span><span class="p">]:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_factor&quot;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;saturation_factor&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;saturation_factor&quot;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;hue_factor&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_hue</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">hue_factor</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;hue_factor&quot;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_factor&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_before&quot;</span><span class="p">]:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;contrast_factor&quot;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;channel_permutation&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">permute_channels</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">permutation</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;channel_permutation&quot;</span><span class="p">])</span>
+        <span class="k">return</span> <span class="n">inpt</span></div>
+
+
+<div class="viewcode-block" id="RandomEqualize"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomEqualize.html#torchvision.transforms.v2.RandomEqualize">[docs]</a><span class="k">class</span> <span class="nc">RandomEqualize</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Equalize the histogram of the given image or video with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomEqualize transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor`, it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;P&quot;, &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being equalized. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomEqualize</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">equalize</span><span class="p">,</span> <span class="n">inpt</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomInvert"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomInvert.html#torchvision.transforms.v2.RandomInvert">[docs]</a><span class="k">class</span> <span class="nc">RandomInvert</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Inverts the colors of the given image or video with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomInvert transform</span>
+
+<span class="sd">    If img is a Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">    where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being color inverted. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomInvert</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">invert</span><span class="p">,</span> <span class="n">inpt</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomPosterize"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomPosterize.html#torchvision.transforms.v2.RandomPosterize">[docs]</a><span class="k">class</span> <span class="nc">RandomPosterize</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Posterize the image or video with a given probability by reducing the</span>
+<span class="sd">    number of bits for each color channel.</span>
+
+<span class="sd">    .. v2betastatus:: RandomPosterize transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor`, it should be of type torch.uint8,</span>
+<span class="sd">    and it is expected to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        bits (int): number of bits to keep for each channel (0-8)</span>
+<span class="sd">        p (float): probability of the image being posterized. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomPosterize</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">bits</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">bits</span> <span class="o">=</span> <span class="n">bits</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">posterize</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">bits</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">bits</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomSolarize"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomSolarize.html#torchvision.transforms.v2.RandomSolarize">[docs]</a><span class="k">class</span> <span class="nc">RandomSolarize</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Solarize the image or video with a given probability by inverting all pixel</span>
+<span class="sd">    values above a threshold.</span>
+
+<span class="sd">    .. v2betastatus:: RandomSolarize transform</span>
+
+<span class="sd">    If img is a Tensor, it is expected to be in [..., 1 or 3, H, W] format,</span>
+<span class="sd">    where ... means it can have an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        threshold (float): all pixels equal or above this value are inverted.</span>
+<span class="sd">        p (float): probability of the image being solarized. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomSolarize</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">threshold</span> <span class="o">=</span> <span class="n">threshold</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">solarize</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">threshold</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">threshold</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomAutocontrast"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomAutocontrast.html#torchvision.transforms.v2.RandomAutocontrast">[docs]</a><span class="k">class</span> <span class="nc">RandomAutocontrast</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Autocontrast the pixels of the given image or video with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomAutocontrast transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor`, it is expected</span>
+<span class="sd">    to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+<span class="sd">    If img is PIL Image, it is expected to be in mode &quot;L&quot; or &quot;RGB&quot;.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float): probability of the image being autocontrasted. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomAutocontrast</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">autocontrast</span><span class="p">,</span> <span class="n">inpt</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomAdjustSharpness"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomAdjustSharpness.html#torchvision.transforms.v2.RandomAdjustSharpness">[docs]</a><span class="k">class</span> <span class="nc">RandomAdjustSharpness</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Adjust the sharpness of the image or video with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomAdjustSharpness transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor`,</span>
+<span class="sd">    it is expected to have [..., 1 or 3, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        sharpness_factor (float):  How much to adjust the sharpness. Can be</span>
+<span class="sd">            any non-negative number. 0 gives a blurred image, 1 gives the</span>
+<span class="sd">            original image while 2 increases the sharpness by a factor of 2.</span>
+<span class="sd">        p (float): probability of the image being sharpened. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomAdjustSharpness</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sharpness_factor</span> <span class="o">=</span> <span class="n">sharpness_factor</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">adjust_sharpness</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">sharpness_factor</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_container.html b/0.16/_modules/torchvision/transforms/v2/_container.html
new file mode 100644
index 00000000000..06264bcad9e
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_container.html
@@ -0,0 +1,849 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._container &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._container</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._container</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span> <span class="k">as</span> <span class="n">_transforms</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">Transform</span>
+
+
+<div class="viewcode-block" id="Compose"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose">[docs]</a><span class="k">class</span> <span class="nc">Compose</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Composes several transforms together.</span>
+
+<span class="sd">    .. v2betastatus:: Compose transform</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+<span class="sd">    Please, see the note below.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transforms (list of ``Transform`` objects): list of transforms to compose.</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        &gt;&gt;&gt; transforms.Compose([</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.CenterCrop(10),</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.PILToTensor(),</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.ConvertImageDtype(torch.float),</span>
+<span class="sd">        &gt;&gt;&gt; ])</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        In order to script the transformations, please use ``torch.nn.Sequential`` as below.</span>
+
+<span class="sd">        &gt;&gt;&gt; transforms = torch.nn.Sequential(</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.CenterCrop(10),</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),</span>
+<span class="sd">        &gt;&gt;&gt; )</span>
+<span class="sd">        &gt;&gt;&gt; scripted_transforms = torch.jit.script(transforms)</span>
+
+<span class="sd">        Make sure to use only scriptable transformations, i.e. that work with ``torch.Tensor``, does not require</span>
+<span class="sd">        `lambda` functions or ``PIL.Image``.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Callable</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">transforms</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument transforms should be a sequence of callables&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="n">transforms</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Pass at least one transform&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+<div class="viewcode-block" id="Compose.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">needs_unpacking</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span>
+        <span class="k">for</span> <span class="n">transform</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">outputs</span> <span class="o">=</span> <span class="n">transform</span><span class="p">(</span><span class="o">*</span><span class="n">inputs</span><span class="p">)</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">outputs</span> <span class="k">if</span> <span class="n">needs_unpacking</span> <span class="k">else</span> <span class="p">(</span><span class="n">outputs</span><span class="p">,)</span>
+        <span class="k">return</span> <span class="n">outputs</span></div>
+
+<div class="viewcode-block" id="Compose.extra_repr"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose.extra_repr">[docs]</a>    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">format_string</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;    </span><span class="si">{</span><span class="n">t</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">format_string</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="RandomApply"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomApply.html#torchvision.transforms.v2.RandomApply">[docs]</a><span class="k">class</span> <span class="nc">RandomApply</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Apply randomly a list of transformations with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomApply transform</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        In order to script the transformation, please use ``torch.nn.ModuleList`` as input instead of list/tuple of</span>
+<span class="sd">        transforms as shown below:</span>
+
+<span class="sd">        &gt;&gt;&gt; transforms = transforms.RandomApply(torch.nn.ModuleList([</span>
+<span class="sd">        &gt;&gt;&gt;     transforms.ColorJitter(),</span>
+<span class="sd">        &gt;&gt;&gt; ]), p=0.3)</span>
+<span class="sd">        &gt;&gt;&gt; scripted_transforms = torch.jit.script(transforms)</span>
+
+<span class="sd">        Make sure to use only scriptable transformations, i.e. that work with ``torch.Tensor``, does not require</span>
+<span class="sd">        `lambda` functions or ``PIL.Image``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transforms (sequence or torch.nn.Module): list of transformations</span>
+<span class="sd">        p (float): probability of applying the list of transforms</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomApply</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Sequence</span><span class="p">[</span><span class="n">Callable</span><span class="p">],</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">],</span> <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">transforms</span><span class="p">,</span> <span class="p">(</span><span class="n">Sequence</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument transforms should be a sequence of callables or a `nn.ModuleList`&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">p</span> <span class="o">&lt;=</span> <span class="mf">1.0</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;`p` should be a floating point value in the interval [0.0, 1.0].&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="n">p</span>
+
+    <span class="k">def</span> <span class="nf">_extract_params_for_v1_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="p">{</span><span class="s2">&quot;transforms&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">,</span> <span class="s2">&quot;p&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">}</span>
+
+<div class="viewcode-block" id="RandomApply.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomApply.html#torchvision.transforms.v2.RandomApply.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">sample</span> <span class="o">=</span> <span class="n">inputs</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">&gt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">sample</span>
+
+        <span class="k">for</span> <span class="n">transform</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">sample</span> <span class="o">=</span> <span class="n">transform</span><span class="p">(</span><span class="n">sample</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">sample</span></div>
+
+<div class="viewcode-block" id="RandomApply.extra_repr"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomApply.html#torchvision.transforms.v2.RandomApply.extra_repr">[docs]</a>    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">format_string</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">:</span>
+            <span class="n">format_string</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;    </span><span class="si">{</span><span class="n">t</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">format_string</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="RandomChoice"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomChoice.html#torchvision.transforms.v2.RandomChoice">[docs]</a><span class="k">class</span> <span class="nc">RandomChoice</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Apply single transformation randomly picked from a list.</span>
+
+<span class="sd">    .. v2betastatus:: RandomChoice transform</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transforms (sequence or torch.nn.Module): list of transformations</span>
+<span class="sd">        p (list of floats or None, optional): probability of each transform being picked.</span>
+<span class="sd">            If ``p`` doesn&#39;t sum to 1, it is automatically normalized. If ``None``</span>
+<span class="sd">            (default), all transforms have the same probability.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">transforms</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Callable</span><span class="p">],</span>
+        <span class="n">p</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">transforms</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument transforms should be a sequence of callables&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">p</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">p</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">transforms</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Length of p doesn&#39;t match the number of transforms: </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">p</span><span class="p">)</span><span class="si">}</span><span class="s2"> != </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+        <span class="n">total</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">p</span> <span class="o">=</span> <span class="p">[</span><span class="n">prob</span> <span class="o">/</span> <span class="n">total</span> <span class="k">for</span> <span class="n">prob</span> <span class="ow">in</span> <span class="n">p</span><span class="p">]</span>
+
+<div class="viewcode-block" id="RandomChoice.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomChoice.html#torchvision.transforms.v2.RandomChoice.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">idx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">multinomial</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">p</span><span class="p">),</span> <span class="mi">1</span><span class="p">))</span>
+        <span class="n">transform</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">transform</span><span class="p">(</span><span class="o">*</span><span class="n">inputs</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="RandomOrder"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomOrder.html#torchvision.transforms.v2.RandomOrder">[docs]</a><span class="k">class</span> <span class="nc">RandomOrder</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Apply a list of transformations in a random order.</span>
+
+<span class="sd">    .. v2betastatus:: RandomOrder transform</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transforms (sequence or torch.nn.Module): list of transformations</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transforms</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Callable</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">transforms</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument transforms should be a sequence of callables&quot;</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+
+<div class="viewcode-block" id="RandomOrder.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomOrder.html#torchvision.transforms.v2.RandomOrder.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">sample</span> <span class="o">=</span> <span class="n">inputs</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="n">torch</span><span class="o">.</span><span class="n">randperm</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">)):</span>
+            <span class="n">transform</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+            <span class="n">sample</span> <span class="o">=</span> <span class="n">transform</span><span class="p">(</span><span class="n">sample</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">sample</span></div></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_deprecated.html b/0.16/_modules/torchvision/transforms/v2/_deprecated.html
new file mode 100644
index 00000000000..5f35389ce4f
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_deprecated.html
@@ -0,0 +1,719 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._deprecated &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._deprecated</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._deprecated</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">_F</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">Transform</span>
+
+
+<div class="viewcode-block" id="ToTensor"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ToTensor.html#torchvision.transforms.v2.ToTensor">[docs]</a><span class="k">class</span> <span class="nc">ToTensor</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] [DEPRECATED] Use ``v2.Compose([v2.ToImage(), v2.ToDtype(torch.float32, scale=True)])`` instead.</span>
+
+<span class="sd">    Convert a PIL Image or ndarray to tensor and scale the values accordingly.</span>
+
+<span class="sd">    .. v2betastatus:: ToTensor transform</span>
+
+<span class="sd">    .. warning::</span>
+<span class="sd">        :class:`v2.ToTensor` is deprecated and will be removed in a future release.</span>
+<span class="sd">        Please use instead ``v2.Compose([v2.ToImage(), v2.ToDtype(torch.float32, scale=True)])``.</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+
+<span class="sd">    Converts a PIL Image or numpy.ndarray (H x W x C) in the range</span>
+<span class="sd">    [0, 255] to a torch.FloatTensor of shape (C x H x W) in the range [0.0, 1.0]</span>
+<span class="sd">    if the PIL Image belongs to one of the modes (L, LA, P, I, F, RGB, YCbCr, RGBA, CMYK, 1)</span>
+<span class="sd">    or if the numpy.ndarray has dtype = np.uint8</span>
+
+<span class="sd">    In the other cases, tensors are returned without scaling.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Because the input image is scaled to [0.0, 1.0], this transformation should not be used when</span>
+<span class="sd">        transforming target image masks. See the `references`_ for implementing the transforms for image masks.</span>
+
+<span class="sd">    .. _references: https://github.com/pytorch/vision/tree/main/references/segmentation</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+            <span class="s2">&quot;The transform `ToTensor()` is deprecated and will be removed in a future release. &quot;</span>
+            <span class="s2">&quot;Instead, please use `v2.Compose([v2.ToImage(), v2.ToDtype(torch.float32, scale=True)])`.&quot;</span>
+        <span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_F</span><span class="o">.</span><span class="n">to_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_geometry.html b/0.16/_modules/torchvision/transforms/v2/_geometry.html
new file mode 100644
index 00000000000..f93af774b48
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_geometry.html
@@ -0,0 +1,2116 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._geometry &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._geometry</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._geometry</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">numbers</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Literal</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span> <span class="k">as</span> <span class="n">_transforms</span><span class="p">,</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops.boxes</span> <span class="kn">import</span> <span class="n">box_iou</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.functional</span> <span class="kn">import</span> <span class="n">_get_perspective_coeffs</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">,</span> <span class="n">Transform</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2.functional._geometry</span> <span class="kn">import</span> <span class="n">_check_interpolation</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2.functional._utils</span> <span class="kn">import</span> <span class="n">_FillType</span>
+
+<span class="kn">from</span> <span class="nn">._transform</span> <span class="kn">import</span> <span class="n">_RandomApplyTransform</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="p">(</span>
+    <span class="n">_check_padding_arg</span><span class="p">,</span>
+    <span class="n">_check_padding_mode_arg</span><span class="p">,</span>
+    <span class="n">_check_sequence_input</span><span class="p">,</span>
+    <span class="n">_get_fill</span><span class="p">,</span>
+    <span class="n">_setup_angle</span><span class="p">,</span>
+    <span class="n">_setup_fill_arg</span><span class="p">,</span>
+    <span class="n">_setup_number_or_seq</span><span class="p">,</span>
+    <span class="n">_setup_size</span><span class="p">,</span>
+    <span class="n">get_bounding_boxes</span><span class="p">,</span>
+    <span class="n">has_all</span><span class="p">,</span>
+    <span class="n">has_any</span><span class="p">,</span>
+    <span class="n">is_pure_tensor</span><span class="p">,</span>
+    <span class="n">query_size</span><span class="p">,</span>
+<span class="p">)</span>
+
+
+<div class="viewcode-block" id="RandomHorizontalFlip"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomHorizontalFlip.html#torchvision.transforms.v2.RandomHorizontalFlip">[docs]</a><span class="k">class</span> <span class="nc">RandomHorizontalFlip</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Horizontally flip the input with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomHorizontalFlip transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float, optional): probability of the input being flipped. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="n">inpt</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomVerticalFlip"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomVerticalFlip.html#torchvision.transforms.v2.RandomVerticalFlip">[docs]</a><span class="k">class</span> <span class="nc">RandomVerticalFlip</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Vertically flip the input with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomVerticalFlip transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        p (float, optional): probability of the input being flipped. Default value is 0.5</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomVerticalFlip</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="n">inpt</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Resize"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Resize.html#torchvision.transforms.v2.Resize">[docs]</a><span class="k">class</span> <span class="nc">Resize</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Resize the input to the given size.</span>
+
+<span class="sd">    .. v2betastatus:: Resize transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    .. warning::</span>
+<span class="sd">        The output image might be different depending on its type: when downsampling, the interpolation of PIL images</span>
+<span class="sd">        and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences</span>
+<span class="sd">        in the performance of a network. Therefore, it is preferable to train and serve a model with the same input</span>
+<span class="sd">        types. See also below the ``antialias`` parameter, which can help making the output of PIL images and tensors</span>
+<span class="sd">        closer.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size. If size is a sequence like</span>
+<span class="sd">            (h, w), output size will be matched to this. If size is an int,</span>
+<span class="sd">            smaller edge of the image will be matched to this number.</span>
+<span class="sd">            i.e, if height &gt; width, then image will be rescaled to</span>
+<span class="sd">            (size * height / width, size).</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode size as single int is not supported, use a sequence of length 1: ``[size, ]``.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,</span>
+<span class="sd">            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        max_size (int, optional): The maximum allowed for the longer edge of</span>
+<span class="sd">            the resized image. If the longer edge of the image is greater</span>
+<span class="sd">            than ``max_size`` after being resized according to ``size``,</span>
+<span class="sd">            ``size`` will be overruled so that the longer edge is equal to</span>
+<span class="sd">            ``max_size``.</span>
+<span class="sd">            As a result, the smaller edge may be shorter than ``size``. This</span>
+<span class="sd">            is only supported if ``size`` is an int (or a sequence of length</span>
+<span class="sd">            1 in torchscript mode).</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">Resize</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">size</span> <span class="o">=</span> <span class="p">[</span><span class="n">size</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="ow">in</span> <span class="p">{</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">}:</span>
+            <span class="n">size</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">size</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;size can either be an integer or a list or tuple of one or two integers, &quot;</span> <span class="sa">f</span><span class="s2">&quot;but got </span><span class="si">{</span><span class="n">size</span><span class="si">}</span><span class="s2"> instead.&quot;</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">size</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span> <span class="o">=</span> <span class="n">max_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span> <span class="o">=</span> <span class="n">antialias</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">resize</span><span class="p">,</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">max_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">max_size</span><span class="p">,</span>
+            <span class="n">antialias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CenterCrop"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.CenterCrop.html#torchvision.transforms.v2.CenterCrop">[docs]</a><span class="k">class</span> <span class="nc">CenterCrop</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Crop the input at the center.</span>
+
+<span class="sd">    .. v2betastatus:: CenterCrop transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    If image size is smaller than output size along any edge, image is padded with 0 and then center cropped.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">CenterCrop</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]]):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">center_crop</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">output_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomResizedCrop"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.v2.RandomResizedCrop">[docs]</a><span class="k">class</span> <span class="nc">RandomResizedCrop</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Crop a random portion of the input and resize it to a given size.</span>
+
+<span class="sd">    .. v2betastatus:: RandomResizedCrop transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    A crop of the original input is made: the crop has a random area (H * W)</span>
+<span class="sd">    and a random aspect ratio. This crop is finally resized to the given</span>
+<span class="sd">    size. This is popularly used to train the Inception networks.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (int or sequence): expected output size of the crop, for each edge. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square output size ``(size, size)`` is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode size as single int is not supported, use a sequence of length 1: ``[size, ]``.</span>
+<span class="sd">        scale (tuple of float, optional): Specifies the lower and upper bounds for the random area of the crop,</span>
+<span class="sd">            before resizing. The scale is defined with respect to the area of the original image.</span>
+<span class="sd">        ratio (tuple of float, optional): lower and upper bounds for the random aspect ratio of the crop, before</span>
+<span class="sd">            resizing.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,</span>
+<span class="sd">            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomResizedCrop</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span>
+        <span class="n">scale</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.08</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">),</span>
+        <span class="n">ratio</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">3.0</span> <span class="o">/</span> <span class="mf">4.0</span><span class="p">,</span> <span class="mf">4.0</span> <span class="o">/</span> <span class="mf">3.0</span><span class="p">),</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">scale</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Scale should be a sequence&quot;</span><span class="p">)</span>
+        <span class="n">scale</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span> <span class="n">scale</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">ratio</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Ratio should be a sequence&quot;</span><span class="p">)</span>
+        <span class="n">ratio</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span> <span class="n">ratio</span><span class="p">)</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span> <span class="ow">or</span> <span class="p">(</span><span class="n">ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">]):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Scale and ratio should be of kind (min, max)&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ratio</span> <span class="o">=</span> <span class="n">ratio</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span> <span class="o">=</span> <span class="n">antialias</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_log_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+        <span class="n">area</span> <span class="o">=</span> <span class="n">height</span> <span class="o">*</span> <span class="n">width</span>
+
+        <span class="n">log_ratio</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_log_ratio</span>
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">):</span>
+            <span class="n">target_area</span> <span class="o">=</span> <span class="n">area</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="n">aspect_ratio</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span>
+                    <span class="n">log_ratio</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>  <span class="c1"># type: ignore[arg-type]</span>
+                    <span class="n">log_ratio</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span>  <span class="c1"># type: ignore[arg-type]</span>
+                <span class="p">)</span>
+            <span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+            <span class="n">w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">target_area</span> <span class="o">*</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+            <span class="n">h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">target_area</span> <span class="o">/</span> <span class="n">aspect_ratio</span><span class="p">)))</span>
+
+            <span class="k">if</span> <span class="mi">0</span> <span class="o">&lt;</span> <span class="n">w</span> <span class="o">&lt;=</span> <span class="n">width</span> <span class="ow">and</span> <span class="mi">0</span> <span class="o">&lt;</span> <span class="n">h</span> <span class="o">&lt;=</span> <span class="n">height</span><span class="p">:</span>
+                <span class="n">i</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="n">h</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+                <span class="n">j</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">width</span> <span class="o">-</span> <span class="n">w</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+                <span class="k">break</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># Fallback to central crop</span>
+            <span class="n">in_ratio</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">width</span><span class="p">)</span> <span class="o">/</span> <span class="nb">float</span><span class="p">(</span><span class="n">height</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">in_ratio</span> <span class="o">&lt;</span> <span class="nb">min</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">):</span>
+                <span class="n">w</span> <span class="o">=</span> <span class="n">width</span>
+                <span class="n">h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">w</span> <span class="o">/</span> <span class="nb">min</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">)))</span>
+            <span class="k">elif</span> <span class="n">in_ratio</span> <span class="o">&gt;</span> <span class="nb">max</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">):</span>
+                <span class="n">h</span> <span class="o">=</span> <span class="n">height</span>
+                <span class="n">w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">h</span> <span class="o">*</span> <span class="nb">max</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ratio</span><span class="p">)))</span>
+            <span class="k">else</span><span class="p">:</span>  <span class="c1"># whole image</span>
+                <span class="n">w</span> <span class="o">=</span> <span class="n">width</span>
+                <span class="n">h</span> <span class="o">=</span> <span class="n">height</span>
+            <span class="n">i</span> <span class="o">=</span> <span class="p">(</span><span class="n">height</span> <span class="o">-</span> <span class="n">h</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span>
+            <span class="n">j</span> <span class="o">=</span> <span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="n">w</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">top</span><span class="o">=</span><span class="n">i</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">j</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">h</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">w</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">resized_crop</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="o">**</span><span class="n">params</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="FiveCrop"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.FiveCrop.html#torchvision.transforms.v2.FiveCrop">[docs]</a><span class="k">class</span> <span class="nc">FiveCrop</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Crop the image or video into four corners and the central crop.</span>
+
+<span class="sd">    .. v2betastatus:: FiveCrop transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a :class:`~torchvision.tv_tensors.Image` or a</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video` it can have arbitrary number of leading batch dimensions.</span>
+<span class="sd">    For example, the image can have ``[..., C, H, W]`` shape.</span>
+
+<span class="sd">    .. Note::</span>
+<span class="sd">         This transform returns a tuple of images and there may be a mismatch in the number of</span>
+<span class="sd">         inputs and targets your Dataset returns. See below for an example of how to deal with</span>
+<span class="sd">         this.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">         size (sequence or int): Desired output size of the crop. If size is an ``int``</span>
+<span class="sd">            instead of sequence like (h, w), a square crop of size (size, size) is made.</span>
+<span class="sd">            If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        &gt;&gt;&gt; class BatchMultiCrop(transforms.Transform):</span>
+<span class="sd">        ...     def forward(self, sample: Tuple[Tuple[Union[tv_tensors.Image, tv_tensors.Video], ...], int]):</span>
+<span class="sd">        ...         images_or_videos, labels = sample</span>
+<span class="sd">        ...         batch_size = len(images_or_videos)</span>
+<span class="sd">        ...         image_or_video = images_or_videos[0]</span>
+<span class="sd">        ...         images_or_videos = tv_tensors.wrap(torch.stack(images_or_videos), like=image_or_video)</span>
+<span class="sd">        ...         labels = torch.full((batch_size,), label, device=images_or_videos.device)</span>
+<span class="sd">        ...         return images_or_videos, labels</span>
+<span class="sd">        ...</span>
+<span class="sd">        &gt;&gt;&gt; image = tv_tensors.Image(torch.rand(3, 256, 256))</span>
+<span class="sd">        &gt;&gt;&gt; label = 3</span>
+<span class="sd">        &gt;&gt;&gt; transform = transforms.Compose([transforms.FiveCrop(224), BatchMultiCrop()])</span>
+<span class="sd">        &gt;&gt;&gt; images, labels = transform(image, label)</span>
+<span class="sd">        &gt;&gt;&gt; images.shape</span>
+<span class="sd">        torch.Size([5, 3, 224, 224])</span>
+<span class="sd">        &gt;&gt;&gt; labels</span>
+<span class="sd">        tensor([3, 3, 3, 3, 3])</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">FiveCrop</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_call_kernel</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">functional</span><span class="p">:</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() is currently passing through inputs of type &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;tv_tensors.</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">. This will likely change in the future.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">five_crop</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_inputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">has_any</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;BoundingBoxes&#39;es and Mask&#39;s are not supported by </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">()&quot;</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="TenCrop"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.TenCrop.html#torchvision.transforms.v2.TenCrop">[docs]</a><span class="k">class</span> <span class="nc">TenCrop</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Crop the image or video into four corners and the central crop plus the flipped version of</span>
+<span class="sd">    these (horizontal flipping is used by default).</span>
+
+<span class="sd">    .. v2betastatus:: TenCrop transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a :class:`~torchvision.tv_tensors.Image` or a</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video` it can have arbitrary number of leading batch dimensions.</span>
+<span class="sd">    For example, the image can have ``[..., C, H, W]`` shape.</span>
+
+<span class="sd">    See :class:`~torchvision.transforms.v2.FiveCrop` for an example.</span>
+
+<span class="sd">    .. Note::</span>
+<span class="sd">         This transform returns a tuple of images and there may be a mismatch in the number of</span>
+<span class="sd">         inputs and targets your Dataset returns. See below for an example of how to deal with</span>
+<span class="sd">         this.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+<span class="sd">        vertical_flip (bool, optional): Use vertical flipping instead of horizontal</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">TenCrop</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span> <span class="n">vertical_flip</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">vertical_flip</span> <span class="o">=</span> <span class="n">vertical_flip</span>
+
+    <span class="k">def</span> <span class="nf">_call_kernel</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">functional</span><span class="p">:</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() is currently passing through inputs of type &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;tv_tensors.</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">. This will likely change in the future.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_check_inputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">has_any</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;BoundingBoxes&#39;es and Mask&#39;s are not supported by </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">()&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">ten_crop</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="n">vertical_flip</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">vertical_flip</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Pad"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Pad.html#torchvision.transforms.v2.Pad">[docs]</a><span class="k">class</span> <span class="nc">Pad</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Pad the input on all sides with the given &quot;pad&quot; value.</span>
+
+<span class="sd">    .. v2betastatus:: Pad transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        padding (int or sequence): Padding on each border. If a single int is provided this</span>
+<span class="sd">            is used to pad all borders. If sequence of length 2 is provided this is the padding</span>
+<span class="sd">            on left/right and top/bottom respectively. If a sequence of length 4 is provided</span>
+<span class="sd">            this is the padding for the left, top, right and bottom borders respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode padding as single int is not supported, use a sequence of</span>
+<span class="sd">                length 1: ``[padding, ]``.</span>
+<span class="sd">        fill (number or tuple or dict, optional): Pixel fill value used when the  ``padding_mode`` is constant.</span>
+<span class="sd">            Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            Fill value can be also a dictionary mapping data type to the fill value, e.g.</span>
+<span class="sd">            ``fill={tv_tensors.Image: 127, tv_tensors.Mask: 0}`` where ``Image`` will be filled with 127 and</span>
+<span class="sd">            ``Mask`` will be filled with 0.</span>
+<span class="sd">        padding_mode (str, optional): Type of padding. Should be: constant, edge, reflect or symmetric.</span>
+<span class="sd">            Default is &quot;constant&quot;.</span>
+
+<span class="sd">            - constant: pads with a constant value, this value is specified with fill</span>
+
+<span class="sd">            - edge: pads with the last value at the edge of the image.</span>
+
+<span class="sd">            - reflect: pads with reflection of image without repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode</span>
+<span class="sd">              will result in [3, 2, 1, 2, 3, 4, 3, 2]</span>
+
+<span class="sd">            - symmetric: pads with reflection of image repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode</span>
+<span class="sd">              will result in [2, 1, 1, 2, 3, 4, 4, 3]</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">Pad</span>
+
+    <span class="k">def</span> <span class="nf">_extract_params_for_v1_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">params</span> <span class="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_extract_params_for_v1_transform</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;fill&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;fill&quot;</span><span class="p">],</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() can only be scripted for a scalar `fill`, but got </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">params</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">padding</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">padding_mode</span><span class="p">:</span> <span class="n">Literal</span><span class="p">[</span><span class="s2">&quot;constant&quot;</span><span class="p">,</span> <span class="s2">&quot;edge&quot;</span><span class="p">,</span> <span class="s2">&quot;reflect&quot;</span><span class="p">,</span> <span class="s2">&quot;symmetric&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="n">_check_padding_arg</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+        <span class="n">_check_padding_mode_arg</span><span class="p">(</span><span class="n">padding_mode</span><span class="p">)</span>
+
+        <span class="c1"># This cast does Sequence[int] -&gt; List[int] and is required to make mypy happy</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">padding</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">padding</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="o">=</span> <span class="n">padding</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span> <span class="o">=</span> <span class="n">padding_mode</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span><span class="p">)</span>  <span class="c1"># type: ignore[arg-type]</span></div>
+
+
+<div class="viewcode-block" id="RandomZoomOut"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomZoomOut.html#torchvision.transforms.v2.RandomZoomOut">[docs]</a><span class="k">class</span> <span class="nc">RandomZoomOut</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] &quot;Zoom out&quot; transformation from</span>
+<span class="sd">    `&quot;SSD: Single Shot MultiBox Detector&quot; &lt;https://arxiv.org/abs/1512.02325&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: RandomZoomOut transform</span>
+
+<span class="sd">    This transformation randomly pads images, videos, bounding boxes and masks creating a zoom out effect.</span>
+<span class="sd">    Output spatial size is randomly sampled from original size up to a maximum size configured</span>
+<span class="sd">    with ``side_range`` parameter:</span>
+
+<span class="sd">    .. code-block:: python</span>
+
+<span class="sd">        r = uniform_sample(side_range[0], side_range[1])</span>
+<span class="sd">        output_width = input_width * r</span>
+<span class="sd">        output_height = input_height * r</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        fill (number or tuple or dict, optional): Pixel fill value used when the  ``padding_mode`` is constant.</span>
+<span class="sd">            Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            Fill value can be also a dictionary mapping data type to the fill value, e.g.</span>
+<span class="sd">            ``fill={tv_tensors.Image: 127, tv_tensors.Mask: 0}`` where ``Image`` will be filled with 127 and</span>
+<span class="sd">            ``Mask`` will be filled with 0.</span>
+<span class="sd">        side_range (sequence of floats, optional): tuple of two floats defines minimum and maximum factors to</span>
+<span class="sd">            scale the input size.</span>
+<span class="sd">        p (float, optional): probability that the zoom operation will be performed.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">side_range</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">1.0</span><span class="p">,</span> <span class="mf">4.0</span><span class="p">),</span>
+        <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+
+        <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">side_range</span><span class="p">,</span> <span class="s2">&quot;side_range&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">side_range</span> <span class="o">=</span> <span class="n">side_range</span>
+        <span class="k">if</span> <span class="n">side_range</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span> <span class="ow">or</span> <span class="n">side_range</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">side_range</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid canvas side range provided </span><span class="si">{</span><span class="n">side_range</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">orig_h</span><span class="p">,</span> <span class="n">orig_w</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="n">r</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">side_range</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">side_range</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">side_range</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">canvas_width</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_w</span> <span class="o">*</span> <span class="n">r</span><span class="p">)</span>
+        <span class="n">canvas_height</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_h</span> <span class="o">*</span> <span class="n">r</span><span class="p">)</span>
+
+        <span class="n">r</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+        <span class="n">left</span> <span class="o">=</span> <span class="nb">int</span><span class="p">((</span><span class="n">canvas_width</span> <span class="o">-</span> <span class="n">orig_w</span><span class="p">)</span> <span class="o">*</span> <span class="n">r</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">top</span> <span class="o">=</span> <span class="nb">int</span><span class="p">((</span><span class="n">canvas_height</span> <span class="o">-</span> <span class="n">orig_h</span><span class="p">)</span> <span class="o">*</span> <span class="n">r</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+        <span class="n">right</span> <span class="o">=</span> <span class="n">canvas_width</span> <span class="o">-</span> <span class="p">(</span><span class="n">left</span> <span class="o">+</span> <span class="n">orig_w</span><span class="p">)</span>
+        <span class="n">bottom</span> <span class="o">=</span> <span class="n">canvas_height</span> <span class="o">-</span> <span class="p">(</span><span class="n">top</span> <span class="o">+</span> <span class="n">orig_h</span><span class="p">)</span>
+        <span class="n">padding</span> <span class="o">=</span> <span class="p">[</span><span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">right</span><span class="p">,</span> <span class="n">bottom</span><span class="p">]</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="o">**</span><span class="n">params</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomRotation"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomRotation.html#torchvision.transforms.v2.RandomRotation">[docs]</a><span class="k">class</span> <span class="nc">RandomRotation</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Rotate the input by angle.</span>
+
+<span class="sd">    .. v2betastatus:: RandomRotation transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        degrees (sequence or number): Range of degrees to select from.</span>
+<span class="sd">            If degrees is a number instead of sequence like (min, max), the range of degrees</span>
+<span class="sd">            will be (-degrees, +degrees).</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        expand (bool, optional): Optional expansion flag.</span>
+<span class="sd">            If true, expands the output to make it large enough to hold the entire rotated image.</span>
+<span class="sd">            If false or omitted, make the output image the same size as the input image.</span>
+<span class="sd">            Note that the expand flag assumes rotation around the center (see note below) and no translation.</span>
+<span class="sd">        center (sequence, optional): Optional center of rotation, (x, y). Origin is the upper left corner.</span>
+<span class="sd">            Default is the center of the image.</span>
+
+<span class="sd">            .. note::</span>
+
+<span class="sd">                In theory, setting ``center`` has no effect if ``expand=True``, since the image center will become the</span>
+<span class="sd">                center of rotation. In practice however, due to numerical precision, this can lead to off-by-one</span>
+<span class="sd">                differences of the resulting image size compared to using the image center in the first place. Thus, when</span>
+<span class="sd">                setting ``expand=True``, it&#39;s best to leave ``center=None`` (default).</span>
+<span class="sd">        fill (number or tuple or dict, optional): Pixel fill value used when the  ``padding_mode`` is constant.</span>
+<span class="sd">            Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            Fill value can be also a dictionary mapping data type to the fill value, e.g.</span>
+<span class="sd">            ``fill={tv_tensors.Image: 127, tv_tensors.Mask: 0}`` where ``Image`` will be filled with 127 and</span>
+<span class="sd">            ``Mask`` will be filled with 0.</span>
+
+<span class="sd">    .. _filters: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#filters</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomRotation</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">degrees</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">],</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">degrees</span> <span class="o">=</span> <span class="n">_setup_angle</span><span class="p">(</span><span class="n">degrees</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">&quot;degrees&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">expand</span> <span class="o">=</span> <span class="n">expand</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="s2">&quot;center&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">center</span> <span class="o">=</span> <span class="n">center</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">angle</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">rotate</span><span class="p">,</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">params</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">expand</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">expand</span><span class="p">,</span>
+            <span class="n">center</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">center</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomAffine"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomAffine.html#torchvision.transforms.v2.RandomAffine">[docs]</a><span class="k">class</span> <span class="nc">RandomAffine</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Random affine transformation the input keeping center invariant.</span>
+
+<span class="sd">    .. v2betastatus:: RandomAffine transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        degrees (sequence or number): Range of degrees to select from.</span>
+<span class="sd">            If degrees is a number instead of sequence like (min, max), the range of degrees</span>
+<span class="sd">            will be (-degrees, +degrees). Set to 0 to deactivate rotations.</span>
+<span class="sd">        translate (tuple, optional): tuple of maximum absolute fraction for horizontal</span>
+<span class="sd">            and vertical translations. For example translate=(a, b), then horizontal shift</span>
+<span class="sd">            is randomly sampled in the range -img_width * a &lt; dx &lt; img_width * a and vertical shift is</span>
+<span class="sd">            randomly sampled in the range -img_height * b &lt; dy &lt; img_height * b. Will not translate by default.</span>
+<span class="sd">        scale (tuple, optional): scaling factor interval, e.g (a, b), then scale is</span>
+<span class="sd">            randomly sampled from the range a &lt;= scale &lt;= b. Will keep original scale by default.</span>
+<span class="sd">        shear (sequence or number, optional): Range of degrees to select from.</span>
+<span class="sd">            If shear is a number, a shear parallel to the x-axis in the range (-shear, +shear)</span>
+<span class="sd">            will be applied. Else if shear is a sequence of 2 values a shear parallel to the x-axis in the</span>
+<span class="sd">            range (shear[0], shear[1]) will be applied. Else if shear is a sequence of 4 values,</span>
+<span class="sd">            an x-axis shear in (shear[0], shear[1]) and y-axis shear in (shear[2], shear[3]) will be applied.</span>
+<span class="sd">            Will not apply shear by default.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.NEAREST``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (number or tuple or dict, optional): Pixel fill value used when the  ``padding_mode`` is constant.</span>
+<span class="sd">            Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            Fill value can be also a dictionary mapping data type to the fill value, e.g.</span>
+<span class="sd">            ``fill={tv_tensors.Image: 127, tv_tensors.Mask: 0}`` where ``Image`` will be filled with 127 and</span>
+<span class="sd">            ``Mask`` will be filled with 0.</span>
+<span class="sd">        center (sequence, optional): Optional center of rotation, (x, y). Origin is the upper left corner.</span>
+<span class="sd">            Default is the center of the image.</span>
+
+<span class="sd">    .. _filters: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#filters</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomAffine</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">degrees</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">],</span>
+        <span class="n">translate</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">scale</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">shear</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">degrees</span> <span class="o">=</span> <span class="n">_setup_angle</span><span class="p">(</span><span class="n">degrees</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">&quot;degrees&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+        <span class="k">if</span> <span class="n">translate</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">translate</span><span class="p">,</span> <span class="s2">&quot;translate&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+            <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">translate</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">t</span> <span class="o">&lt;=</span> <span class="mf">1.0</span><span class="p">):</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;translation values should be between 0 and 1&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">translate</span> <span class="o">=</span> <span class="n">translate</span>
+        <span class="k">if</span> <span class="n">scale</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">scale</span><span class="p">,</span> <span class="s2">&quot;scale&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+            <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">scale</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">s</span> <span class="o">&lt;=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;scale values should be positive&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">scale</span>
+
+        <span class="k">if</span> <span class="n">shear</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">shear</span> <span class="o">=</span> <span class="n">_setup_angle</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">&quot;shear&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">shear</span> <span class="o">=</span> <span class="n">shear</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">_check_sequence_input</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="s2">&quot;center&quot;</span><span class="p">,</span> <span class="n">req_sizes</span><span class="o">=</span><span class="p">(</span><span class="mi">2</span><span class="p">,))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">center</span> <span class="o">=</span> <span class="n">center</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="n">angle</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">degrees</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">translate</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">max_dx</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">translate</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">width</span><span class="p">)</span>
+            <span class="n">max_dy</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">translate</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">height</span><span class="p">)</span>
+            <span class="n">tx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="o">-</span><span class="n">max_dx</span><span class="p">,</span> <span class="n">max_dx</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()))</span>
+            <span class="n">ty</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="o">-</span><span class="n">max_dy</span><span class="p">,</span> <span class="n">max_dy</span><span class="p">)</span><span class="o">.</span><span class="n">item</span><span class="p">()))</span>
+            <span class="n">translate</span> <span class="o">=</span> <span class="p">(</span><span class="n">tx</span><span class="p">,</span> <span class="n">ty</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">translate</span> <span class="o">=</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">scale</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">scale</span> <span class="o">=</span> <span class="mf">1.0</span>
+
+        <span class="n">shear_x</span> <span class="o">=</span> <span class="n">shear_y</span> <span class="o">=</span> <span class="mf">0.0</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">shear</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">shear_x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">shear</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">shear</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">shear</span><span class="p">)</span> <span class="o">==</span> <span class="mi">4</span><span class="p">:</span>
+                <span class="n">shear_y</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">shear</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">shear</span><span class="p">[</span><span class="mi">3</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+        <span class="n">shear</span> <span class="o">=</span> <span class="p">(</span><span class="n">shear_x</span><span class="p">,</span> <span class="n">shear_y</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">affine</span><span class="p">,</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">params</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+            <span class="n">center</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">center</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomCrop"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop">[docs]</a><span class="k">class</span> <span class="nc">RandomCrop</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Crop the input at a random location.</span>
+
+<span class="sd">    .. v2betastatus:: RandomCrop transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        size (sequence or int): Desired output size of the crop. If size is an</span>
+<span class="sd">            int instead of sequence like (h, w), a square crop (size, size) is</span>
+<span class="sd">            made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</span>
+<span class="sd">        padding (int or sequence, optional): Optional padding on each border</span>
+<span class="sd">            of the image. Default is None. If a single int is provided this</span>
+<span class="sd">            is used to pad all borders. If sequence of length 2 is provided this is the padding</span>
+<span class="sd">            on left/right and top/bottom respectively. If a sequence of length 4 is provided</span>
+<span class="sd">            this is the padding for the left, top, right and bottom borders respectively.</span>
+
+<span class="sd">            .. note::</span>
+<span class="sd">                In torchscript mode padding as single int is not supported, use a sequence of</span>
+<span class="sd">                length 1: ``[padding, ]``.</span>
+<span class="sd">        pad_if_needed (boolean, optional): It will pad the image if smaller than the</span>
+<span class="sd">            desired size to avoid raising an exception. Since cropping is done</span>
+<span class="sd">            after padding, the padding seems to be done at a random offset.</span>
+<span class="sd">        fill (number or tuple or dict, optional): Pixel fill value used when the  ``padding_mode`` is constant.</span>
+<span class="sd">            Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            Fill value can be also a dictionary mapping data type to the fill value, e.g.</span>
+<span class="sd">            ``fill={tv_tensors.Image: 127, tv_tensors.Mask: 0}`` where ``Image`` will be filled with 127 and</span>
+<span class="sd">            ``Mask`` will be filled with 0.</span>
+<span class="sd">        padding_mode (str, optional): Type of padding. Should be: constant, edge, reflect or symmetric.</span>
+<span class="sd">            Default is constant.</span>
+
+<span class="sd">            - constant: pads with a constant value, this value is specified with fill</span>
+
+<span class="sd">            - edge: pads with the last value at the edge of the image.</span>
+
+<span class="sd">            - reflect: pads with reflection of image without repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode</span>
+<span class="sd">              will result in [3, 2, 1, 2, 3, 4, 3, 2]</span>
+
+<span class="sd">            - symmetric: pads with reflection of image repeating the last value on the edge.</span>
+<span class="sd">              For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode</span>
+<span class="sd">              will result in [2, 1, 1, 2, 3, 4, 4, 3]</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomCrop</span>
+
+    <span class="k">def</span> <span class="nf">_extract_params_for_v1_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">params</span> <span class="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">_extract_params_for_v1_transform</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;fill&quot;</span><span class="p">]</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;fill&quot;</span><span class="p">],</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">))):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() can only be scripted for a scalar `fill`, but got </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">fill</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="n">padding</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span>
+        <span class="k">if</span> <span class="n">padding</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">pad_left</span><span class="p">,</span> <span class="n">pad_right</span><span class="p">,</span> <span class="n">pad_top</span><span class="p">,</span> <span class="n">pad_bottom</span> <span class="o">=</span> <span class="n">padding</span>
+            <span class="n">padding</span> <span class="o">=</span> <span class="p">[</span><span class="n">pad_left</span><span class="p">,</span> <span class="n">pad_top</span><span class="p">,</span> <span class="n">pad_right</span><span class="p">,</span> <span class="n">pad_bottom</span><span class="p">]</span>
+        <span class="n">params</span><span class="p">[</span><span class="s2">&quot;padding&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">padding</span>
+
+        <span class="k">return</span> <span class="n">params</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span>
+        <span class="n">padding</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">pad_if_needed</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">padding_mode</span><span class="p">:</span> <span class="n">Literal</span><span class="p">[</span><span class="s2">&quot;constant&quot;</span><span class="p">,</span> <span class="s2">&quot;edge&quot;</span><span class="p">,</span> <span class="s2">&quot;reflect&quot;</span><span class="p">,</span> <span class="s2">&quot;symmetric&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">error_msg</span><span class="o">=</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">pad_if_needed</span> <span class="ow">or</span> <span class="n">padding</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">padding</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">_check_padding_arg</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+            <span class="n">_check_padding_mode_arg</span><span class="p">(</span><span class="n">padding_mode</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">_geometry</span><span class="o">.</span><span class="n">_parse_pad_padding</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span> <span class="k">if</span> <span class="n">padding</span> <span class="k">else</span> <span class="kc">None</span>  <span class="c1"># type: ignore[arg-type]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pad_if_needed</span> <span class="o">=</span> <span class="n">pad_if_needed</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span> <span class="o">=</span> <span class="n">padding_mode</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">padded_height</span><span class="p">,</span> <span class="n">padded_width</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">pad_left</span><span class="p">,</span> <span class="n">pad_right</span><span class="p">,</span> <span class="n">pad_top</span><span class="p">,</span> <span class="n">pad_bottom</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding</span>
+            <span class="n">padded_height</span> <span class="o">+=</span> <span class="n">pad_top</span> <span class="o">+</span> <span class="n">pad_bottom</span>
+            <span class="n">padded_width</span> <span class="o">+=</span> <span class="n">pad_left</span> <span class="o">+</span> <span class="n">pad_right</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">pad_left</span> <span class="o">=</span> <span class="n">pad_right</span> <span class="o">=</span> <span class="n">pad_top</span> <span class="o">=</span> <span class="n">pad_bottom</span> <span class="o">=</span> <span class="mi">0</span>
+
+        <span class="n">cropped_height</span><span class="p">,</span> <span class="n">cropped_width</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pad_if_needed</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">padded_height</span> <span class="o">&lt;</span> <span class="n">cropped_height</span><span class="p">:</span>
+                <span class="n">diff</span> <span class="o">=</span> <span class="n">cropped_height</span> <span class="o">-</span> <span class="n">padded_height</span>
+
+                <span class="n">pad_top</span> <span class="o">+=</span> <span class="n">diff</span>
+                <span class="n">pad_bottom</span> <span class="o">+=</span> <span class="n">diff</span>
+                <span class="n">padded_height</span> <span class="o">+=</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">diff</span>
+
+            <span class="k">if</span> <span class="n">padded_width</span> <span class="o">&lt;</span> <span class="n">cropped_width</span><span class="p">:</span>
+                <span class="n">diff</span> <span class="o">=</span> <span class="n">cropped_width</span> <span class="o">-</span> <span class="n">padded_width</span>
+
+                <span class="n">pad_left</span> <span class="o">+=</span> <span class="n">diff</span>
+                <span class="n">pad_right</span> <span class="o">+=</span> <span class="n">diff</span>
+                <span class="n">padded_width</span> <span class="o">+=</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">diff</span>
+
+        <span class="k">if</span> <span class="n">padded_height</span> <span class="o">&lt;</span> <span class="n">cropped_height</span> <span class="ow">or</span> <span class="n">padded_width</span> <span class="o">&lt;</span> <span class="n">cropped_width</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Required crop size </span><span class="si">{</span><span class="p">(</span><span class="n">cropped_height</span><span class="p">,</span><span class="w"> </span><span class="n">cropped_width</span><span class="p">)</span><span class="si">}</span><span class="s2"> is larger than &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="s1">&#39;padded &#39;</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="w"> </span><span class="ow">is</span><span class="w"> </span><span class="ow">not</span><span class="w"> </span><span class="kc">None</span><span class="w"> </span><span class="k">else</span><span class="w"> </span><span class="s1">&#39;&#39;</span><span class="si">}</span><span class="s2">input image size </span><span class="si">{</span><span class="p">(</span><span class="n">padded_height</span><span class="p">,</span><span class="w"> </span><span class="n">padded_width</span><span class="p">)</span><span class="si">}</span><span class="s2">.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="c1"># We need a different order here than we have in self.padding since this padding will be parsed again in `F.pad`</span>
+        <span class="n">padding</span> <span class="o">=</span> <span class="p">[</span><span class="n">pad_left</span><span class="p">,</span> <span class="n">pad_top</span><span class="p">,</span> <span class="n">pad_right</span><span class="p">,</span> <span class="n">pad_bottom</span><span class="p">]</span>
+        <span class="n">needs_pad</span> <span class="o">=</span> <span class="nb">any</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+
+        <span class="n">needs_vert_crop</span><span class="p">,</span> <span class="n">top</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="p">(</span><span class="kc">True</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">padded_height</span> <span class="o">-</span> <span class="n">cropped_height</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">())))</span>
+            <span class="k">if</span> <span class="n">padded_height</span> <span class="o">&gt;</span> <span class="n">cropped_height</span>
+            <span class="k">else</span> <span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="p">)</span>
+        <span class="n">needs_horz_crop</span><span class="p">,</span> <span class="n">left</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="p">(</span><span class="kc">True</span><span class="p">,</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">padded_width</span> <span class="o">-</span> <span class="n">cropped_width</span> <span class="o">+</span> <span class="mi">1</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">())))</span>
+            <span class="k">if</span> <span class="n">padded_width</span> <span class="o">&gt;</span> <span class="n">cropped_width</span>
+            <span class="k">else</span> <span class="p">(</span><span class="kc">False</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">needs_crop</span><span class="o">=</span><span class="n">needs_vert_crop</span> <span class="ow">or</span> <span class="n">needs_horz_crop</span><span class="p">,</span>
+            <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span>
+            <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span>
+            <span class="n">height</span><span class="o">=</span><span class="n">cropped_height</span><span class="p">,</span>
+            <span class="n">width</span><span class="o">=</span><span class="n">cropped_width</span><span class="p">,</span>
+            <span class="n">needs_pad</span><span class="o">=</span><span class="n">needs_pad</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;needs_pad&quot;</span><span class="p">]:</span>
+            <span class="n">fill</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">pad</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;padding&quot;</span><span class="p">],</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">padding_mode</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;needs_crop&quot;</span><span class="p">]:</span>
+            <span class="n">inpt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+                <span class="n">F</span><span class="o">.</span><span class="n">crop</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">top</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;top&quot;</span><span class="p">],</span> <span class="n">left</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;left&quot;</span><span class="p">],</span> <span class="n">height</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;height&quot;</span><span class="p">],</span> <span class="n">width</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;width&quot;</span><span class="p">]</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">inpt</span></div>
+
+
+<div class="viewcode-block" id="RandomPerspective"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomPerspective.html#torchvision.transforms.v2.RandomPerspective">[docs]</a><span class="k">class</span> <span class="nc">RandomPerspective</span><span class="p">(</span><span class="n">_RandomApplyTransform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Perform a random perspective transformation of the input with a given probability.</span>
+
+<span class="sd">    .. v2betastatus:: RandomPerspective transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        distortion_scale (float, optional): argument to control the degree of distortion and ranges from 0 to 1.</span>
+<span class="sd">            Default is 0.5.</span>
+<span class="sd">        p (float, optional): probability of the input being transformed. Default is 0.5.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (number or tuple or dict, optional): Pixel fill value used when the  ``padding_mode`` is constant.</span>
+<span class="sd">            Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            Fill value can be also a dictionary mapping data type to the fill value, e.g.</span>
+<span class="sd">            ``fill={tv_tensors.Image: 127, tv_tensors.Mask: 0}`` where ``Image`` will be filled with 127 and</span>
+<span class="sd">            ``Mask`` will be filled with 0.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">RandomPerspective</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">distortion_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="n">p</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="mi">0</span> <span class="o">&lt;=</span> <span class="n">distortion_scale</span> <span class="o">&lt;=</span> <span class="mi">1</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument distortion_scale value should be between 0 and 1&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">distortion_scale</span> <span class="o">=</span> <span class="n">distortion_scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="n">distortion_scale</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">distortion_scale</span>
+
+        <span class="n">half_height</span> <span class="o">=</span> <span class="n">height</span> <span class="o">//</span> <span class="mi">2</span>
+        <span class="n">half_width</span> <span class="o">=</span> <span class="n">width</span> <span class="o">//</span> <span class="mi">2</span>
+        <span class="n">bound_height</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_height</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span>
+        <span class="n">bound_width</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">distortion_scale</span> <span class="o">*</span> <span class="n">half_width</span><span class="p">)</span> <span class="o">+</span> <span class="mi">1</span>
+        <span class="n">topleft</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound_width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound_height</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+        <span class="p">]</span>
+        <span class="n">topright</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="n">bound_width</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound_height</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+        <span class="p">]</span>
+        <span class="n">botright</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="n">bound_width</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">height</span> <span class="o">-</span> <span class="n">bound_height</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+        <span class="p">]</span>
+        <span class="n">botleft</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound_width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+            <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">height</span> <span class="o">-</span> <span class="n">bound_height</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,))),</span>
+        <span class="p">]</span>
+        <span class="n">startpoints</span> <span class="o">=</span> <span class="p">[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="n">width</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="mi">1</span><span class="p">]]</span>
+        <span class="n">endpoints</span> <span class="o">=</span> <span class="p">[</span><span class="n">topleft</span><span class="p">,</span> <span class="n">topright</span><span class="p">,</span> <span class="n">botright</span><span class="p">,</span> <span class="n">botleft</span><span class="p">]</span>
+        <span class="n">perspective_coeffs</span> <span class="o">=</span> <span class="n">_get_perspective_coeffs</span><span class="p">(</span><span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">coefficients</span><span class="o">=</span><span class="n">perspective_coeffs</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">perspective</span><span class="p">,</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="kc">None</span><span class="p">,</span>
+            <span class="kc">None</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">params</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="ElasticTransform"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ElasticTransform.html#torchvision.transforms.v2.ElasticTransform">[docs]</a><span class="k">class</span> <span class="nc">ElasticTransform</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Transform the input with elastic transformations.</span>
+
+<span class="sd">    .. v2betastatus:: RandomPerspective transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Given alpha and sigma, it will generate displacement</span>
+<span class="sd">    vectors for all pixels based on random offsets. Alpha controls the strength</span>
+<span class="sd">    and sigma controls the smoothness of the displacements.</span>
+<span class="sd">    The displacements are added to an identity grid and the resulting grid is</span>
+<span class="sd">    used to transform the input.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        Implementation to transform bounding boxes is approximative (not exact).</span>
+<span class="sd">        We construct an approximation of the inverse grid as ``inverse_grid = identity - displacement``.</span>
+<span class="sd">        This is not an exact inverse of the grid used to transform images, i.e. ``grid = identity + displacement``.</span>
+<span class="sd">        Our assumption is that ``displacement * displacement`` is small and can be ignored.</span>
+<span class="sd">        Large displacements would lead to large errors in the approximation.</span>
+
+<span class="sd">    Applications:</span>
+<span class="sd">        Randomly transforms the morphology of objects in images and produces a</span>
+<span class="sd">        see-through-water-like effect.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        alpha (float or sequence of floats, optional): Magnitude of displacements. Default is 50.0.</span>
+<span class="sd">        sigma (float or sequence of floats, optional): Smoothness of displacements. Default is 5.0.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.BILINEAR`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        fill (number or tuple or dict, optional): Pixel fill value used when the  ``padding_mode`` is constant.</span>
+<span class="sd">            Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.</span>
+<span class="sd">            Fill value can be also a dictionary mapping data type to the fill value, e.g.</span>
+<span class="sd">            ``fill={tv_tensors.Image: 127, tv_tensors.Mask: 0}`` where ``Image`` will be filled with 127 and</span>
+<span class="sd">            ``Mask`` will be filled with 0.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">ElasticTransform</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">alpha</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="mf">50.0</span><span class="p">,</span>
+        <span class="n">sigma</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="mf">5.0</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">_FillType</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">_FillType</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span> <span class="o">=</span> <span class="n">_setup_number_or_seq</span><span class="p">(</span><span class="n">alpha</span><span class="p">,</span> <span class="s2">&quot;alpha&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span> <span class="o">=</span> <span class="n">_setup_number_or_seq</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="s2">&quot;sigma&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fill</span> <span class="o">=</span> <span class="n">fill</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_fill</span> <span class="o">=</span> <span class="n">_setup_fill_arg</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">))</span>
+
+        <span class="n">dx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">size</span><span class="p">)</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="n">kx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">8</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="c1"># if kernel size is even we have to make it odd</span>
+            <span class="k">if</span> <span class="n">kx</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">kx</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">dx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="n">dx</span><span class="p">,</span> <span class="p">[</span><span class="n">kx</span><span class="p">,</span> <span class="n">kx</span><span class="p">],</span> <span class="nb">list</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">))</span>
+        <span class="n">dx</span> <span class="o">=</span> <span class="n">dx</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">/</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="n">dy</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">size</span><span class="p">)</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">-</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="n">ky</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">8</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="c1"># if kernel size is even we have to make it odd</span>
+            <span class="k">if</span> <span class="n">ky</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">ky</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="n">dy</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="n">dy</span><span class="p">,</span> <span class="p">[</span><span class="n">ky</span><span class="p">,</span> <span class="n">ky</span><span class="p">],</span> <span class="nb">list</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">))</span>
+        <span class="n">dy</span> <span class="o">=</span> <span class="n">dy</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">alpha</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">/</span> <span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">displacement</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">concat</span><span class="p">([</span><span class="n">dx</span><span class="p">,</span> <span class="n">dy</span><span class="p">],</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">])</span>  <span class="c1"># 1 x H x W x 2</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">displacement</span><span class="o">=</span><span class="n">displacement</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="n">_get_fill</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_fill</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">elastic</span><span class="p">,</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">params</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomIoUCrop"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomIoUCrop.html#torchvision.transforms.v2.RandomIoUCrop">[docs]</a><span class="k">class</span> <span class="nc">RandomIoUCrop</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Random IoU crop transformation from</span>
+<span class="sd">    `&quot;SSD: Single Shot MultiBox Detector&quot; &lt;https://arxiv.org/abs/1512.02325&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: RandomIoUCrop transform</span>
+
+<span class="sd">    This transformation requires an image or video data and ``tv_tensors.BoundingBoxes`` in the input.</span>
+
+<span class="sd">    .. warning::</span>
+<span class="sd">        In order to properly remove the bounding boxes below the IoU threshold, `RandomIoUCrop`</span>
+<span class="sd">        must be followed by :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes`, either immediately</span>
+<span class="sd">        after or later in the transforms pipeline.</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        min_scale (float, optional): Minimum factors to scale the input size.</span>
+<span class="sd">        max_scale (float, optional): Maximum factors to scale the input size.</span>
+<span class="sd">        min_aspect_ratio (float, optional): Minimum aspect ratio for the cropped image or video.</span>
+<span class="sd">        max_aspect_ratio (float, optional): Maximum aspect ratio for the cropped image or video.</span>
+<span class="sd">        sampler_options (list of float, optional): List of minimal IoU (Jaccard) overlap between all the boxes and</span>
+<span class="sd">            a cropped image or video. Default, ``None`` which corresponds to ``[0.0, 0.1, 0.3, 0.5, 0.7, 0.9, 1.0]``</span>
+<span class="sd">        trials (int, optional): Number of trials to find a crop for a given value of minimal IoU (Jaccard) overlap.</span>
+<span class="sd">            Default, 40.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">min_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.3</span><span class="p">,</span>
+        <span class="n">max_scale</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">min_aspect_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.5</span><span class="p">,</span>
+        <span class="n">max_aspect_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">2.0</span><span class="p">,</span>
+        <span class="n">sampler_options</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">trials</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">40</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="c1"># Configuration similar to https://github.com/weiliu89/caffe/blob/ssd/examples/ssd/ssd_coco.py#L89-L174</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">min_scale</span> <span class="o">=</span> <span class="n">min_scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_scale</span> <span class="o">=</span> <span class="n">max_scale</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">min_aspect_ratio</span> <span class="o">=</span> <span class="n">min_aspect_ratio</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_aspect_ratio</span> <span class="o">=</span> <span class="n">max_aspect_ratio</span>
+        <span class="k">if</span> <span class="n">sampler_options</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">sampler_options</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.7</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">options</span> <span class="o">=</span> <span class="n">sampler_options</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">trials</span> <span class="o">=</span> <span class="n">trials</span>
+
+    <span class="k">def</span> <span class="nf">_check_inputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span>
+            <span class="n">has_all</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">)</span>
+            <span class="ow">and</span> <span class="n">has_any</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">is_pure_tensor</span><span class="p">)</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() requires input sample to contain tensor or PIL images &quot;</span>
+                <span class="s2">&quot;and bounding boxes. Sample can also contain masks.&quot;</span>
+            <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">orig_h</span><span class="p">,</span> <span class="n">orig_w</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+        <span class="n">bboxes</span> <span class="o">=</span> <span class="n">get_bounding_boxes</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="k">while</span> <span class="kc">True</span><span class="p">:</span>
+            <span class="c1"># sample an option</span>
+            <span class="n">idx</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">low</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">high</span><span class="o">=</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">options</span><span class="p">),</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">1</span><span class="p">,)))</span>
+            <span class="n">min_jaccard_overlap</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">options</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+            <span class="k">if</span> <span class="n">min_jaccard_overlap</span> <span class="o">&gt;=</span> <span class="mf">1.0</span><span class="p">:</span>  <span class="c1"># a value larger than 1 encodes the leave as-is option</span>
+                <span class="k">return</span> <span class="nb">dict</span><span class="p">()</span>
+
+            <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">trials</span><span class="p">):</span>
+                <span class="c1"># check the aspect ratio limitations</span>
+                <span class="n">r</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">min_scale</span> <span class="o">+</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">max_scale</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">min_scale</span><span class="p">)</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+                <span class="n">new_w</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_w</span> <span class="o">*</span> <span class="n">r</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+                <span class="n">new_h</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_h</span> <span class="o">*</span> <span class="n">r</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+                <span class="n">aspect_ratio</span> <span class="o">=</span> <span class="n">new_w</span> <span class="o">/</span> <span class="n">new_h</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">min_aspect_ratio</span> <span class="o">&lt;=</span> <span class="n">aspect_ratio</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_aspect_ratio</span><span class="p">):</span>
+                    <span class="k">continue</span>
+
+                <span class="c1"># check for 0 area crops</span>
+                <span class="n">r</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+                <span class="n">left</span> <span class="o">=</span> <span class="nb">int</span><span class="p">((</span><span class="n">orig_w</span> <span class="o">-</span> <span class="n">new_w</span><span class="p">)</span> <span class="o">*</span> <span class="n">r</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+                <span class="n">top</span> <span class="o">=</span> <span class="nb">int</span><span class="p">((</span><span class="n">orig_h</span> <span class="o">-</span> <span class="n">new_h</span><span class="p">)</span> <span class="o">*</span> <span class="n">r</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+                <span class="n">right</span> <span class="o">=</span> <span class="n">left</span> <span class="o">+</span> <span class="n">new_w</span>
+                <span class="n">bottom</span> <span class="o">=</span> <span class="n">top</span> <span class="o">+</span> <span class="n">new_h</span>
+                <span class="k">if</span> <span class="n">left</span> <span class="o">==</span> <span class="n">right</span> <span class="ow">or</span> <span class="n">top</span> <span class="o">==</span> <span class="n">bottom</span><span class="p">:</span>
+                    <span class="k">continue</span>
+
+                <span class="c1"># check for any valid boxes with centers within the crop area</span>
+                <span class="n">xyxy_bboxes</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_bounding_box_format</span><span class="p">(</span>
+                    <span class="n">bboxes</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span>
+                    <span class="n">bboxes</span><span class="o">.</span><span class="n">format</span><span class="p">,</span>
+                    <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="n">cx</span> <span class="o">=</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="p">(</span><span class="n">xyxy_bboxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">xyxy_bboxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">])</span>
+                <span class="n">cy</span> <span class="o">=</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="p">(</span><span class="n">xyxy_bboxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">xyxy_bboxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">3</span><span class="p">])</span>
+                <span class="n">is_within_crop_area</span> <span class="o">=</span> <span class="p">(</span><span class="n">left</span> <span class="o">&lt;</span> <span class="n">cx</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">cx</span> <span class="o">&lt;</span> <span class="n">right</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">top</span> <span class="o">&lt;</span> <span class="n">cy</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">cy</span> <span class="o">&lt;</span> <span class="n">bottom</span><span class="p">)</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">is_within_crop_area</span><span class="o">.</span><span class="n">any</span><span class="p">():</span>
+                    <span class="k">continue</span>
+
+                <span class="c1"># check at least 1 box with jaccard limitations</span>
+                <span class="n">xyxy_bboxes</span> <span class="o">=</span> <span class="n">xyxy_bboxes</span><span class="p">[</span><span class="n">is_within_crop_area</span><span class="p">]</span>
+                <span class="n">ious</span> <span class="o">=</span> <span class="n">box_iou</span><span class="p">(</span>
+                    <span class="n">xyxy_bboxes</span><span class="p">,</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([[</span><span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">right</span><span class="p">,</span> <span class="n">bottom</span><span class="p">]],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">xyxy_bboxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">xyxy_bboxes</span><span class="o">.</span><span class="n">device</span><span class="p">),</span>
+                <span class="p">)</span>
+                <span class="k">if</span> <span class="n">ious</span><span class="o">.</span><span class="n">max</span><span class="p">()</span> <span class="o">&lt;</span> <span class="n">min_jaccard_overlap</span><span class="p">:</span>
+                    <span class="k">continue</span>
+
+                <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">new_h</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">new_w</span><span class="p">,</span> <span class="n">is_within_crop_area</span><span class="o">=</span><span class="n">is_within_crop_area</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">params</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">inpt</span>
+
+        <span class="n">output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">crop</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">top</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;top&quot;</span><span class="p">],</span> <span class="n">left</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;left&quot;</span><span class="p">],</span> <span class="n">height</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;height&quot;</span><span class="p">],</span> <span class="n">width</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;width&quot;</span><span class="p">]</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">):</span>
+            <span class="c1"># We &quot;mark&quot; the invalid boxes as degenreate, and they can be</span>
+            <span class="c1"># removed by a later call to SanitizeBoundingBoxes()</span>
+            <span class="n">output</span><span class="p">[</span><span class="o">~</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;is_within_crop_area&quot;</span><span class="p">]]</span> <span class="o">=</span> <span class="mi">0</span>
+
+        <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="ScaleJitter"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ScaleJitter.html#torchvision.transforms.v2.ScaleJitter">[docs]</a><span class="k">class</span> <span class="nc">ScaleJitter</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Perform Large Scale Jitter on the input according to</span>
+<span class="sd">    `&quot;Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation&quot; &lt;https://arxiv.org/abs/2012.07177&gt;`_.</span>
+
+<span class="sd">    .. v2betastatus:: ScaleJitter transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        target_size (tuple of int): Target size. This parameter defines base scale for jittering,</span>
+<span class="sd">            e.g. ``min(target_size[0] / width, target_size[1] / height)``.</span>
+<span class="sd">        scale_range (tuple of float, optional): Minimum and maximum of the scale range. Default, ``(0.1, 2.0)``.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,</span>
+<span class="sd">            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">target_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">scale_range</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">2.0</span><span class="p">),</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_size</span> <span class="o">=</span> <span class="n">target_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale_range</span> <span class="o">=</span> <span class="n">scale_range</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span> <span class="o">=</span> <span class="n">antialias</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">orig_height</span><span class="p">,</span> <span class="n">orig_width</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="n">scale</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale_range</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">scale_range</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale_range</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">r</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">target_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">/</span> <span class="n">orig_height</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">target_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">/</span> <span class="n">orig_width</span><span class="p">)</span> <span class="o">*</span> <span class="n">scale</span>
+        <span class="n">new_width</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_width</span> <span class="o">*</span> <span class="n">r</span><span class="p">)</span>
+        <span class="n">new_height</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_height</span> <span class="o">*</span> <span class="n">r</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">resize</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;size&quot;</span><span class="p">],</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomShortestSize"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomShortestSize.html#torchvision.transforms.v2.RandomShortestSize">[docs]</a><span class="k">class</span> <span class="nc">RandomShortestSize</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Randomly resize the input.</span>
+
+<span class="sd">    .. v2betastatus:: RandomShortestSize transform</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        min_size (int or sequence of int): Minimum spatial size. Single integer value or a sequence of integer values.</span>
+<span class="sd">        max_size (int, optional): Maximum spatial size. Default, None.</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,</span>
+<span class="sd">            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">min_size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">min_size</span> <span class="o">=</span> <span class="p">[</span><span class="n">min_size</span><span class="p">]</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">min_size</span><span class="p">,</span> <span class="nb">int</span><span class="p">)</span> <span class="k">else</span> <span class="nb">list</span><span class="p">(</span><span class="n">min_size</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span> <span class="o">=</span> <span class="n">max_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span> <span class="o">=</span> <span class="n">antialias</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">orig_height</span><span class="p">,</span> <span class="n">orig_width</span> <span class="o">=</span> <span class="n">query_size</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="n">min_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">min_size</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">min_size</span><span class="p">),</span> <span class="p">()))]</span>
+        <span class="n">r</span> <span class="o">=</span> <span class="n">min_size</span> <span class="o">/</span> <span class="nb">min</span><span class="p">(</span><span class="n">orig_height</span><span class="p">,</span> <span class="n">orig_width</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">r</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">r</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span> <span class="o">/</span> <span class="nb">max</span><span class="p">(</span><span class="n">orig_height</span><span class="p">,</span> <span class="n">orig_width</span><span class="p">))</span>
+
+        <span class="n">new_width</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_width</span> <span class="o">*</span> <span class="n">r</span><span class="p">)</span>
+        <span class="n">new_height</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">orig_height</span> <span class="o">*</span> <span class="n">r</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">resize</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;size&quot;</span><span class="p">],</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RandomResize"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.RandomResize.html#torchvision.transforms.v2.RandomResize">[docs]</a><span class="k">class</span> <span class="nc">RandomResize</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Randomly resize the input.</span>
+
+<span class="sd">    .. v2betastatus:: RandomResize transform</span>
+
+<span class="sd">    This transformation can be used together with ``RandomCrop`` as data augmentations to train</span>
+<span class="sd">    models on image segmentation task.</span>
+
+<span class="sd">    Output spatial size is randomly sampled from the interval ``[min_size, max_size]``:</span>
+
+<span class="sd">    .. code-block:: python</span>
+
+<span class="sd">        size = uniform_sample(min_size, max_size)</span>
+<span class="sd">        output_width = size</span>
+<span class="sd">        output_height = size</span>
+
+<span class="sd">    If the input is a :class:`torch.Tensor` or a ``TVTensor`` (e.g. :class:`~torchvision.tv_tensors.Image`,</span>
+<span class="sd">    :class:`~torchvision.tv_tensors.Video`, :class:`~torchvision.tv_tensors.BoundingBoxes` etc.)</span>
+<span class="sd">    it can have arbitrary number of leading batch dimensions. For example,</span>
+<span class="sd">    the image can have ``[..., C, H, W]`` shape. A bounding box can have ``[..., 4]`` shape.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        min_size (int): Minimum output size for random sampling</span>
+<span class="sd">        max_size (int): Maximum output size for random sampling</span>
+<span class="sd">        interpolation (InterpolationMode, optional): Desired interpolation enum defined by</span>
+<span class="sd">            :class:`torchvision.transforms.InterpolationMode`. Default is ``InterpolationMode.BILINEAR``.</span>
+<span class="sd">            If input is Tensor, only ``InterpolationMode.NEAREST``, ``InterpolationMode.NEAREST_EXACT``,</span>
+<span class="sd">            ``InterpolationMode.BILINEAR`` and ``InterpolationMode.BICUBIC`` are supported.</span>
+<span class="sd">            The corresponding Pillow integer constants, e.g. ``PIL.Image.BILINEAR`` are accepted as well.</span>
+<span class="sd">        antialias (bool, optional): Whether to apply antialiasing.</span>
+<span class="sd">            It only affects **tensors** with bilinear or bicubic modes and it is</span>
+<span class="sd">            ignored otherwise: on PIL images, antialiasing is always applied on</span>
+<span class="sd">            bilinear or bicubic modes; on other modes (for PIL images and</span>
+<span class="sd">            tensors), antialiasing makes no sense and this parameter is ignored.</span>
+<span class="sd">            Possible values are:</span>
+
+<span class="sd">            - ``True``: will apply antialiasing for bilinear or bicubic modes.</span>
+<span class="sd">              Other mode aren&#39;t affected. This is probably what you want to use.</span>
+<span class="sd">            - ``False``: will not apply antialiasing for tensors on any mode. PIL</span>
+<span class="sd">              images are still antialiased on bilinear or bicubic modes, because</span>
+<span class="sd">              PIL doesn&#39;t support no antialias.</span>
+<span class="sd">            - ``None``: equivalent to ``False`` for tensors and ``True`` for</span>
+<span class="sd">              PIL images. This value exists for legacy reasons and you probably</span>
+<span class="sd">              don&#39;t want to use it unless you really know what you are doing.</span>
+
+<span class="sd">            The current default is ``None`` **but will change to** ``True`` **in</span>
+<span class="sd">            v0.17** for the PIL and Tensor backends to be consistent.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">min_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">max_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+        <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">min_size</span> <span class="o">=</span> <span class="n">min_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span> <span class="o">=</span> <span class="n">max_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">antialias</span> <span class="o">=</span> <span class="n">antialias</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">min_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_size</span><span class="p">,</span> <span class="p">()))</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">[</span><span class="n">size</span><span class="p">])</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">resize</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;size&quot;</span><span class="p">],</span> <span class="n">interpolation</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">antialias</span>
+        <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_meta.html b/0.16/_modules/torchvision/transforms/v2/_meta.html
new file mode 100644
index 00000000000..59bdfa6bca2
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_meta.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._meta &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._meta</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._meta</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">Transform</span>
+
+
+<div class="viewcode-block" id="ConvertBoundingBoxFormat"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html#torchvision.transforms.v2.ConvertBoundingBoxFormat">[docs]</a><span class="k">class</span> <span class="nc">ConvertBoundingBoxFormat</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Convert bounding box coordinates to the given ``format``, eg from &quot;CXCYWH&quot; to &quot;XYXY&quot;.</span>
+
+<span class="sd">    .. v2betastatus:: ConvertBoundingBoxFormat transform</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        format (str or tv_tensors.BoundingBoxFormat): output bounding box format.</span>
+<span class="sd">            Possible values are defined by :class:`~torchvision.tv_tensors.BoundingBoxFormat` and</span>
+<span class="sd">            string values match the enums, e.g. &quot;XYXY&quot; or &quot;XYWH&quot; etc.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">format</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">format</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="nb">format</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">[</span><span class="nb">format</span><span class="p">]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">format</span> <span class="o">=</span> <span class="nb">format</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_bounding_box_format</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">format</span><span class="p">)</span>  <span class="c1"># type: ignore[return-value]</span></div>
+
+
+<div class="viewcode-block" id="ClampBoundingBoxes"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ClampBoundingBoxes.html#torchvision.transforms.v2.ClampBoundingBoxes">[docs]</a><span class="k">class</span> <span class="nc">ClampBoundingBoxes</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Clamp bounding boxes to their corresponding image dimensions.</span>
+
+<span class="sd">    The clamping is done according to the bounding boxes&#39; ``canvas_size`` meta-data.</span>
+
+<span class="sd">    .. v2betastatus:: ClampBoundingBoxes transform</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">clamp_bounding_boxes</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>  <span class="c1"># type: ignore[return-value]</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_misc.html b/0.16/_modules/torchvision/transforms/v2/_misc.html
new file mode 100644
index 00000000000..2f8067d84c9
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_misc.html
@@ -0,0 +1,1090 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._misc &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._misc</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._misc</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch.utils._pytree</span> <span class="kn">import</span> <span class="n">tree_flatten</span><span class="p">,</span> <span class="n">tree_unflatten</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span> <span class="k">as</span> <span class="n">_transforms</span><span class="p">,</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">Transform</span>
+
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_parse_labels_getter</span><span class="p">,</span> <span class="n">_setup_number_or_seq</span><span class="p">,</span> <span class="n">_setup_size</span><span class="p">,</span> <span class="n">get_bounding_boxes</span><span class="p">,</span> <span class="n">has_any</span><span class="p">,</span> <span class="n">is_pure_tensor</span>
+
+
+<span class="c1"># TODO: do we want/need to expose this?</span>
+<span class="k">class</span> <span class="nc">Identity</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">inpt</span>
+
+
+<div class="viewcode-block" id="Lambda"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Lambda.html#torchvision.transforms.v2.Lambda">[docs]</a><span class="k">class</span> <span class="nc">Lambda</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Apply a user-defined function as a transform.</span>
+
+<span class="sd">    .. v2betastatus:: Lambda transform</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        lambd (function): Lambda/function to be used for transform.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="nb">object</span><span class="p">,)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">lambd</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[[</span><span class="n">Any</span><span class="p">],</span> <span class="n">Any</span><span class="p">],</span> <span class="o">*</span><span class="n">types</span><span class="p">:</span> <span class="n">Type</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">lambd</span> <span class="o">=</span> <span class="n">lambd</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">types</span> <span class="o">=</span> <span class="n">types</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">_transformed_types</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">types</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">lambd</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">inpt</span>
+
+<div class="viewcode-block" id="Lambda.extra_repr"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Lambda.html#torchvision.transforms.v2.Lambda.extra_repr">[docs]</a>    <span class="k">def</span> <span class="nf">extra_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="n">extras</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">name</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lambd</span><span class="p">,</span> <span class="s2">&quot;__name__&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">name</span><span class="p">:</span>
+            <span class="n">extras</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">name</span><span class="p">)</span>
+        <span class="n">extras</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;types=</span><span class="si">{</span><span class="p">[</span><span class="nb">type</span><span class="o">.</span><span class="vm">__name__</span><span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="nb">type</span><span class="w"> </span><span class="ow">in</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">types</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="s2">&quot;, &quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">extras</span><span class="p">)</span></div></div>
+
+
+<div class="viewcode-block" id="LinearTransformation"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.LinearTransformation.html#torchvision.transforms.v2.LinearTransformation">[docs]</a><span class="k">class</span> <span class="nc">LinearTransformation</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Transform a tensor image or video with a square transformation matrix and a mean_vector computed offline.</span>
+
+<span class="sd">    .. v2betastatus:: LinearTransformation transform</span>
+
+<span class="sd">    This transform does not support PIL Image.</span>
+<span class="sd">    Given transformation_matrix and mean_vector, will flatten the torch.*Tensor and</span>
+<span class="sd">    subtract mean_vector from it which is then followed by computing the dot</span>
+<span class="sd">    product with the transformation matrix and then reshaping the tensor to its</span>
+<span class="sd">    original shape.</span>
+
+<span class="sd">    Applications:</span>
+<span class="sd">        whitening transformation: Suppose X is a column vector zero-centered data.</span>
+<span class="sd">        Then compute the data covariance matrix [D x D] with torch.mm(X.t(), X),</span>
+<span class="sd">        perform SVD on this matrix and pass it as transformation_matrix.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        transformation_matrix (Tensor): tensor [D x D], D = C x H x W</span>
+<span class="sd">        mean_vector (Tensor): tensor [D], D = C x H x W</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">LinearTransformation</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">is_pure_tensor</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">transformation_matrix</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mean_vector</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">!=</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;transformation_matrix should be square. Got &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">tuple</span><span class="p">(</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">())</span><span class="si">}</span><span class="s2"> rectangular matrix.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">mean_vector</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">!=</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;mean_vector should have the same length </span><span class="si">{</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot; as any one of the dimensions of the transformation_matrix [</span><span class="si">{</span><span class="nb">tuple</span><span class="p">(</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">size</span><span class="p">())</span><span class="si">}</span><span class="s2">]&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">device</span> <span class="o">!=</span> <span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Input tensors should be on the same device. Got </span><span class="si">{</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2"> and </span><span class="si">{</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">transformation_matrix</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">mean_vector</span><span class="o">.</span><span class="n">dtype</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Input tensors should have the same dtype. Got </span><span class="si">{</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2"> and </span><span class="si">{</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span> <span class="o">=</span> <span class="n">transformation_matrix</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span> <span class="o">=</span> <span class="n">mean_vector</span>
+
+    <span class="k">def</span> <span class="nf">_check_inputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">sample</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">has_any</span><span class="p">(</span><span class="n">sample</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() does not support PIL images.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">shape</span> <span class="o">=</span> <span class="n">inpt</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">n</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">n</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Input tensor and transformation matrix have incompatible shape.&quot;</span>
+                <span class="o">+</span> <span class="sa">f</span><span class="s2">&quot;[</span><span class="si">{</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span><span class="si">}</span><span class="s2"> x </span><span class="si">{</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span><span class="si">}</span><span class="s2"> x </span><span class="si">{</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2">] != &quot;</span>
+                <span class="o">+</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">inpt</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Input tensor should be on the same device as transformation matrix and mean vector. &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;Got </span><span class="si">{</span><span class="n">inpt</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2"> vs </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">flat_inpt</span> <span class="o">=</span> <span class="n">inpt</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">n</span><span class="p">)</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">mean_vector</span>
+
+        <span class="n">transformation_matrix</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transformation_matrix</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">flat_inpt</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">mm</span><span class="p">(</span><span class="n">flat_inpt</span><span class="p">,</span> <span class="n">transformation_matrix</span><span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)):</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">output</span></div>
+
+
+<div class="viewcode-block" id="Normalize"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.Normalize.html#torchvision.transforms.v2.Normalize">[docs]</a><span class="k">class</span> <span class="nc">Normalize</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Normalize a tensor image or video with mean and standard deviation.</span>
+
+<span class="sd">    .. v2betastatus:: Normalize transform</span>
+
+<span class="sd">    This transform does not support PIL Image.</span>
+<span class="sd">    Given mean: ``(mean[1],...,mean[n])`` and std: ``(std[1],..,std[n])`` for ``n``</span>
+<span class="sd">    channels, this transform will normalize each channel of the input</span>
+<span class="sd">    ``torch.*Tensor`` i.e.,</span>
+<span class="sd">    ``output[channel] = (input[channel] - mean[channel]) / std[channel]``</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        This transform acts out of place, i.e., it does not mutate the input tensor.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        mean (sequence): Sequence of means for each channel.</span>
+<span class="sd">        std (sequence): Sequence of standard deviations for each channel.</span>
+<span class="sd">        inplace(bool,optional): Bool to make this operation in-place.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">Normalize</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">mean</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">std</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mean</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">mean</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">std</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">std</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inplace</span> <span class="o">=</span> <span class="n">inplace</span>
+
+    <span class="k">def</span> <span class="nf">_check_inputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">sample</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">has_any</span><span class="p">(</span><span class="n">sample</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">() does not support PIL images.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">normalize</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">mean</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">std</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inplace</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="GaussianBlur"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.GaussianBlur.html#torchvision.transforms.v2.GaussianBlur">[docs]</a><span class="k">class</span> <span class="nc">GaussianBlur</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Blurs image with randomly chosen Gaussian blur.</span>
+
+<span class="sd">    .. v2betastatus:: GausssianBlur transform</span>
+
+<span class="sd">    If the input is a Tensor, it is expected</span>
+<span class="sd">    to have [..., C, H, W] shape, where ... means an arbitrary number of leading dimensions.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        kernel_size (int or sequence): Size of the Gaussian kernel.</span>
+<span class="sd">        sigma (float or tuple of float (min, max)): Standard deviation to be used for</span>
+<span class="sd">            creating kernel to perform blurring. If float, sigma is fixed. If it is tuple</span>
+<span class="sd">            of float (min, max), sigma is chosen uniformly at random to lie in the</span>
+<span class="sd">            given range.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">GaussianBlur</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">]],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">2.0</span><span class="p">)</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span> <span class="o">=</span> <span class="n">_setup_size</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="s2">&quot;Kernel size should be a tuple/list of two integers&quot;</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">ks</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">ks</span> <span class="o">&lt;=</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">ks</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Kernel size value should be an odd and positive number.&quot;</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span> <span class="o">=</span> <span class="n">_setup_number_or_seq</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="s2">&quot;sigma&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sigma values should be positive and of the form (min, max). Got </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_get_params</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">flat_inputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">sigma</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">uniform_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">sigma</span><span class="o">=</span><span class="p">[</span><span class="n">sigma</span><span class="p">,</span> <span class="n">sigma</span><span class="p">])</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">kernel_size</span><span class="p">,</span> <span class="o">**</span><span class="n">params</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="ToDtype"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ToDtype.html#torchvision.transforms.v2.ToDtype">[docs]</a><span class="k">class</span> <span class="nc">ToDtype</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Converts the input to a specific dtype, optionally scaling the values for images or videos.</span>
+
+<span class="sd">    .. v2betastatus:: ToDtype transform</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        ``ToDtype(dtype, scale=True)`` is the recommended replacement for ``ConvertImageDtype(dtype)``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        dtype (``torch.dtype`` or dict of ``TVTensor`` -&gt; ``torch.dtype``): The dtype to convert to.</span>
+<span class="sd">            If a ``torch.dtype`` is passed, e.g. ``torch.float32``, only images and videos will be converted</span>
+<span class="sd">            to that dtype: this is for compatibility with :class:`~torchvision.transforms.v2.ConvertImageDtype`.</span>
+<span class="sd">            A dict can be passed to specify per-tv_tensor conversions, e.g.</span>
+<span class="sd">            ``dtype={tv_tensors.Image: torch.float32, tv_tensors.Mask: torch.int64, &quot;others&quot;:None}``. The &quot;others&quot;</span>
+<span class="sd">            key can be used as a catch-all for any other tv_tensor type, and ``None`` means no conversion.</span>
+<span class="sd">        scale (bool, optional): Whether to scale the values for images or videos. See :ref:`range_and_dtype`.</span>
+<span class="sd">            Default: ``False``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">]]],</span> <span class="n">scale</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">dtype</span><span class="p">,</span> <span class="p">(</span><span class="nb">dict</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;dtype must be a dict or a torch.dtype, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span><span class="si">}</span><span class="s2"> instead&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="p">(</span>
+            <span class="nb">isinstance</span><span class="p">(</span><span class="n">dtype</span><span class="p">,</span> <span class="nb">dict</span><span class="p">)</span>
+            <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span> <span class="ow">in</span> <span class="n">dtype</span>
+            <span class="ow">and</span> <span class="nb">any</span><span class="p">(</span><span class="bp">cls</span> <span class="ow">in</span> <span class="n">dtype</span> <span class="k">for</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="p">[</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">])</span>
+        <span class="p">):</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;Got `dtype` values for `torch.Tensor` and either `tv_tensors.Image` or `tv_tensors.Video`. &quot;</span>
+                <span class="s2">&quot;Note that a plain `torch.Tensor` will *not* be transformed by this (or any other transformation) &quot;</span>
+                <span class="s2">&quot;in case a `tv_tensors.Image` or `tv_tensors.Video` is present in the input.&quot;</span>
+            <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">dtype</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="o">=</span> <span class="n">scale</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">):</span>
+            <span class="c1"># For consistency / BC with ConvertImageDtype, we only care about images or videos when dtype</span>
+            <span class="c1"># is a simple torch.dtype</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_pure_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)):</span>
+                <span class="k">return</span> <span class="n">inpt</span>
+
+            <span class="n">dtype</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span>
+        <span class="k">elif</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">:</span>
+            <span class="n">dtype</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">[</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)]</span>
+        <span class="k">elif</span> <span class="s2">&quot;others&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">:</span>
+            <span class="n">dtype</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">[</span><span class="s2">&quot;others&quot;</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;No dtype was specified for type </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="si">}</span><span class="s2">. &quot;</span>
+                <span class="s2">&quot;If you only need to convert the dtype of images or videos, you can just pass e.g. dtype=torch.float32. &quot;</span>
+                <span class="s2">&quot;If you&#39;re passing a dict as dtype, &quot;</span>
+                <span class="s1">&#39;you can use &quot;others&quot; as a catch-all key &#39;</span>
+                <span class="s1">&#39;e.g. dtype={tv_tensors.Mask: torch.int64, &quot;others&quot;: None} to pass-through the rest of the inputs.&#39;</span>
+            <span class="p">)</span>
+
+        <span class="n">supports_scaling</span> <span class="o">=</span> <span class="n">is_pure_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span> <span class="ow">or</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">dtype</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">scale</span> <span class="ow">and</span> <span class="n">supports_scaling</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                    <span class="s2">&quot;scale was set to True but no dtype was specified for images or videos: no scaling will be done.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">return</span> <span class="n">inpt</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">to_dtype</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">scale</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="ConvertImageDtype"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ConvertImageDtype.html#torchvision.transforms.v2.ConvertImageDtype">[docs]</a><span class="k">class</span> <span class="nc">ConvertImageDtype</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] [DEPRECATED] Use ``v2.ToDtype(dtype, scale=True)`` instead.</span>
+
+<span class="sd">    Convert input image to the given ``dtype`` and scale the values accordingly.</span>
+
+<span class="sd">    .. v2betastatus:: ConvertImageDtype transform</span>
+
+<span class="sd">    .. warning::</span>
+<span class="sd">        Consider using ``ToDtype(dtype, scale=True)`` instead. See :class:`~torchvision.transforms.v2.ToDtype`.</span>
+
+<span class="sd">    This function does not support PIL Image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        dtype (torch.dtype): Desired data type of the output</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        When converting from a smaller to a larger integer ``dtype`` the maximum values are **not** mapped exactly.</span>
+<span class="sd">        If converted back and forth, this mismatch has no effect.</span>
+
+<span class="sd">    Raises:</span>
+<span class="sd">        RuntimeError: When trying to cast :class:`torch.float32` to :class:`torch.int32` or :class:`torch.int64` as</span>
+<span class="sd">            well as for trying to cast :class:`torch.float64` to :class:`torch.int64`. These conversions might lead to</span>
+<span class="sd">            overflow errors since the floating point ``dtype`` cannot store consecutive integers over the whole range</span>
+<span class="sd">            of the integer ``dtype``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_v1_transform_cls</span> <span class="o">=</span> <span class="n">_transforms</span><span class="o">.</span><span class="n">ConvertImageDtype</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">dtype</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">to_dtype</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="SanitizeBoundingBoxes"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes">[docs]</a><span class="k">class</span> <span class="nc">SanitizeBoundingBoxes</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Remove degenerate/invalid bounding boxes and their corresponding labels and masks.</span>
+
+<span class="sd">    .. v2betastatus:: SanitizeBoundingBoxes transform</span>
+
+<span class="sd">    This transform removes bounding boxes and their associated labels/masks that:</span>
+
+<span class="sd">    - are below a given ``min_size``: by default this also removes degenerate boxes that have e.g. X2 &lt;= X1.</span>
+<span class="sd">    - have any coordinate outside of their corresponding image. You may want to</span>
+<span class="sd">      call :class:`~torchvision.transforms.v2.ClampBoundingBoxes` first to avoid undesired removals.</span>
+
+<span class="sd">    It is recommended to call it at the end of a pipeline, before passing the</span>
+<span class="sd">    input to the models. It is critical to call this transform if</span>
+<span class="sd">    :class:`~torchvision.transforms.v2.RandomIoUCrop` was called.</span>
+<span class="sd">    If you want to be extra careful, you may call it after all transforms that</span>
+<span class="sd">    may modify bounding boxes but once at the end should be enough in most</span>
+<span class="sd">    cases.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        min_size (float, optional) The size below which bounding boxes are removed. Default is 1.</span>
+<span class="sd">        labels_getter (callable or str or None, optional): indicates how to identify the labels in the input.</span>
+<span class="sd">            By default, this will try to find a &quot;labels&quot; key in the input (case-insensitive), if</span>
+<span class="sd">            the input is a dict or it is a tuple whose second element is a dict.</span>
+<span class="sd">            This heuristic should work well with a lot of datasets, including the built-in torchvision datasets.</span>
+<span class="sd">            It can also be a callable that takes the same input</span>
+<span class="sd">            as the transform, and returns the labels.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">min_size</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">labels_getter</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="n">Any</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]],</span> <span class="nb">str</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;default&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">min_size</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;min_size must be &gt;= 1, got </span><span class="si">{</span><span class="n">min_size</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">min_size</span> <span class="o">=</span> <span class="n">min_size</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">labels_getter</span> <span class="o">=</span> <span class="n">labels_getter</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_labels_getter</span> <span class="o">=</span> <span class="n">_parse_labels_getter</span><span class="p">(</span><span class="n">labels_getter</span><span class="p">)</span>
+
+<div class="viewcode-block" id="SanitizeBoundingBoxes.forward"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">inputs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">inputs</span> <span class="o">=</span> <span class="n">inputs</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels_getter</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">labels</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The labels in the input to forward() must be a tensor or None, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span><span class="si">}</span><span class="s2"> instead.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">flat_inputs</span><span class="p">,</span> <span class="n">spec</span> <span class="o">=</span> <span class="n">tree_flatten</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="n">boxes</span> <span class="o">=</span> <span class="n">get_bounding_boxes</span><span class="p">(</span><span class="n">flat_inputs</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="n">labels</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Number of boxes (shape=</span><span class="si">{</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">) and number of labels (shape=</span><span class="si">{</span><span class="n">labels</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">) do not match.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">boxes</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span>
+            <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span>
+            <span class="n">F</span><span class="o">.</span><span class="n">convert_bounding_box_format</span><span class="p">(</span>
+                <span class="n">boxes</span><span class="p">,</span>
+                <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">)</span>
+        <span class="n">ws</span><span class="p">,</span> <span class="n">hs</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">-</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span>
+        <span class="n">valid</span> <span class="o">=</span> <span class="p">(</span><span class="n">ws</span> <span class="o">&gt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">min_size</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">hs</span> <span class="o">&gt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">min_size</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">boxes</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># TODO: Do we really need to check for out of bounds here? All</span>
+        <span class="c1"># transforms should be clamping anyway, so this should never happen?</span>
+        <span class="n">image_h</span><span class="p">,</span> <span class="n">image_w</span> <span class="o">=</span> <span class="n">boxes</span><span class="o">.</span><span class="n">canvas_size</span>
+        <span class="n">valid</span> <span class="o">&amp;=</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">image_w</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">image_w</span><span class="p">)</span>
+        <span class="n">valid</span> <span class="o">&amp;=</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">image_h</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">image_h</span><span class="p">)</span>
+
+        <span class="n">params</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">valid</span><span class="o">=</span><span class="n">valid</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="n">labels</span><span class="o">=</span><span class="n">labels</span><span class="p">)</span>
+        <span class="n">flat_outputs</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="c1"># Even-though it may look like we&#39;re transforming all inputs, we don&#39;t:</span>
+            <span class="c1"># _transform() will only care about BoundingBoxeses and the labels</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_transform</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">params</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">inpt</span> <span class="ow">in</span> <span class="n">flat_inputs</span>
+        <span class="p">]</span>
+
+        <span class="k">return</span> <span class="n">tree_unflatten</span><span class="p">(</span><span class="n">flat_outputs</span><span class="p">,</span> <span class="n">spec</span><span class="p">)</span></div>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="n">is_label</span> <span class="o">=</span> <span class="n">inpt</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">inpt</span> <span class="ow">is</span> <span class="n">params</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span>
+        <span class="n">is_bounding_boxes_or_mask</span> <span class="o">=</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">is_label</span> <span class="ow">or</span> <span class="n">is_bounding_boxes_or_mask</span><span class="p">):</span>
+            <span class="k">return</span> <span class="n">inpt</span>
+
+        <span class="n">output</span> <span class="o">=</span> <span class="n">inpt</span><span class="p">[</span><span class="n">params</span><span class="p">[</span><span class="s2">&quot;valid&quot;</span><span class="p">]]</span>
+
+        <span class="k">if</span> <span class="n">is_label</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">output</span>
+
+        <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_temporal.html b/0.16/_modules/torchvision/transforms/v2/_temporal.html
new file mode 100644
index 00000000000..4e22b8a2b2e
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_temporal.html
@@ -0,0 +1,697 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._temporal &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._temporal</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._temporal</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">Transform</span>
+
+
+<div class="viewcode-block" id="UniformTemporalSubsample"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.UniformTemporalSubsample.html#torchvision.transforms.v2.UniformTemporalSubsample">[docs]</a><span class="k">class</span> <span class="nc">UniformTemporalSubsample</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Uniformly subsample ``num_samples`` indices from the temporal dimension of the video.</span>
+
+<span class="sd">    .. v2betastatus:: UniformTemporalSubsample transform</span>
+
+<span class="sd">    Videos are expected to be of shape ``[..., T, C, H, W]`` where ``T`` denotes the temporal dimension.</span>
+
+<span class="sd">    When ``num_samples`` is larger than the size of temporal dimension of the video, it</span>
+<span class="sd">    will sample frames based on nearest neighbor interpolation.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        num_samples (int): The number of equispaced samples to be selected</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">num_samples</span> <span class="o">=</span> <span class="n">num_samples</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Any</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_call_kernel</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">uniform_temporal_subsample</span><span class="p">,</span> <span class="n">inpt</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_samples</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/_type_conversion.html b/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
new file mode 100644
index 00000000000..0b5af12a030
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
@@ -0,0 +1,761 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2._type_conversion &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2._type_conversion</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2._type_conversion</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span><span class="p">,</span> <span class="n">Transform</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2._utils</span> <span class="kn">import</span> <span class="n">is_pure_tensor</span>
+
+
+<div class="viewcode-block" id="PILToTensor"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.PILToTensor.html#torchvision.transforms.v2.PILToTensor">[docs]</a><span class="k">class</span> <span class="nc">PILToTensor</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Convert a PIL Image to a tensor of the same type - this does not scale values.</span>
+
+<span class="sd">    .. v2betastatus:: PILToTensor transform</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Converts a PIL Image (H x W x C) to a Tensor of shape (C x H x W).</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="ToImage"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ToImage.html#torchvision.transforms.v2.ToImage">[docs]</a><span class="k">class</span> <span class="nc">ToImage</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Convert a tensor, ndarray, or PIL Image to :class:`~torchvision.tv_tensors.Image`</span>
+<span class="sd">    ; this does not scale values.</span>
+
+<span class="sd">    .. v2betastatus:: ToImage transform</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">is_pure_tensor</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">to_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="ToPILImage"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ToPILImage.html#torchvision.transforms.v2.ToPILImage">[docs]</a><span class="k">class</span> <span class="nc">ToPILImage</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Convert a tensor or an ndarray to PIL Image</span>
+
+<span class="sd">    .. v2betastatus:: ToPILImage transform</span>
+
+<span class="sd">    This transform does not support torchscript.</span>
+
+<span class="sd">    Converts a torch.*Tensor of shape C x H x W or a numpy ndarray of shape</span>
+<span class="sd">    H x W x C to a PIL Image while adjusting the value range depending on the ``mode``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        mode (`PIL.Image mode`_): color space and pixel depth of input data (optional).</span>
+<span class="sd">            If ``mode`` is ``None`` (default) there are some assumptions made about the input data:</span>
+
+<span class="sd">            - If the input has 4 channels, the ``mode`` is assumed to be ``RGBA``.</span>
+<span class="sd">            - If the input has 3 channels, the ``mode`` is assumed to be ``RGB``.</span>
+<span class="sd">            - If the input has 2 channels, the ``mode`` is assumed to be ``LA``.</span>
+<span class="sd">            - If the input has 1 channel, the ``mode`` is determined by the data type (i.e ``int``, ``float``,</span>
+<span class="sd">              ``short``).</span>
+
+<span class="sd">    .. _PIL.Image mode: https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">is_pure_tensor</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">mode</span> <span class="o">=</span> <span class="n">mode</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">],</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">to_pil_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="ToPureTensor"><a class="viewcode-back" href="../../../../generated/torchvision.transforms.v2.ToPureTensor.html#torchvision.transforms.v2.ToPureTensor">[docs]</a><span class="k">class</span> <span class="nc">ToPureTensor</span><span class="p">(</span><span class="n">Transform</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Convert all tv_tensors to pure tensors, removing associated metadata (if any).</span>
+
+<span class="sd">    .. v2betastatus:: ToPureTensor transform</span>
+
+<span class="sd">    This doesn&#39;t scale or change the values, only the type.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">_transformed_types</span> <span class="o">=</span> <span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span><span class="p">,)</span>
+
+    <span class="k">def</span> <span class="nf">_transform</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">params</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../" src="../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../" id="documentation_options" src="../../../../_static/documentation_options.js"></script>
+         <script src="../../../../_static/jquery.js"></script>
+         <script src="../../../../_static/underscore.js"></script>
+         <script src="../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../_static/doctools.js"></script>
+         <script src="../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_augment.html b/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
new file mode 100644
index 00000000000..8ded01f03b9
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._augment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._augment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._augment</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.functional</span> <span class="kn">import</span> <span class="n">pil_to_tensor</span><span class="p">,</span> <span class="n">to_pil_image</span>
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_get_kernel</span><span class="p">,</span> <span class="n">_register_kernel_internal</span>
+
+
+<div class="viewcode-block" id="erase"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.erase.html#torchvision.transforms.v2.functional.erase">[docs]</a><span class="k">def</span> <span class="nf">erase</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">i</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">j</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">h</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">w</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">v</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomErase` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">erase_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">i</span><span class="o">=</span><span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="o">=</span><span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="o">=</span><span class="n">w</span><span class="p">,</span> <span class="n">v</span><span class="o">=</span><span class="n">v</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">erase</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">erase</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">i</span><span class="o">=</span><span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="o">=</span><span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="o">=</span><span class="n">w</span><span class="p">,</span> <span class="n">v</span><span class="o">=</span><span class="n">v</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">erase</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">erase</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">erase_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">i</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">j</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">h</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">v</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">inplace</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+
+    <span class="n">image</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">i</span> <span class="p">:</span> <span class="n">i</span> <span class="o">+</span> <span class="n">h</span><span class="p">,</span> <span class="n">j</span> <span class="p">:</span> <span class="n">j</span> <span class="o">+</span> <span class="n">w</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span>
+    <span class="k">return</span> <span class="n">image</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">erase</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_erase_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">i</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">j</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">h</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">v</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">t_img</span> <span class="o">=</span> <span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">erase_image</span><span class="p">(</span><span class="n">t_img</span><span class="p">,</span> <span class="n">i</span><span class="o">=</span><span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="o">=</span><span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="o">=</span><span class="n">w</span><span class="p">,</span> <span class="n">v</span><span class="o">=</span><span class="n">v</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">to_pil_image</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">mode</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">erase</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">erase_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">i</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">j</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">h</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">v</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">erase_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">i</span><span class="o">=</span><span class="n">i</span><span class="p">,</span> <span class="n">j</span><span class="o">=</span><span class="n">j</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="o">=</span><span class="n">w</span><span class="p">,</span> <span class="n">v</span><span class="o">=</span><span class="n">v</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_color.html b/0.16/_modules/torchvision/transforms/v2/functional/_color.html
new file mode 100644
index 00000000000..ec8f12ab3b7
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_color.html
@@ -0,0 +1,1378 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._color &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._color</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._color</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch.nn.functional</span> <span class="kn">import</span> <span class="n">conv2d</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">_functional_pil</span> <span class="k">as</span> <span class="n">_FP</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms._functional_tensor</span> <span class="kn">import</span> <span class="n">_max_value</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">._misc</span> <span class="kn">import</span> <span class="n">_num_value_bits</span><span class="p">,</span> <span class="n">to_dtype_image</span>
+<span class="kn">from</span> <span class="nn">._type_conversion</span> <span class="kn">import</span> <span class="n">pil_to_tensor</span><span class="p">,</span> <span class="n">to_pil_image</span>
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_get_kernel</span><span class="p">,</span> <span class="n">_register_kernel_internal</span>
+
+
+<div class="viewcode-block" id="rgb_to_grayscale"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html#torchvision.transforms.v2.functional.rgb_to_grayscale">[docs]</a><span class="k">def</span> <span class="nf">rgb_to_grayscale</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.Grayscale` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">rgb_to_grayscale_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="n">num_output_channels</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">rgb_to_grayscale</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">rgb_to_grayscale</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="n">num_output_channels</span><span class="p">)</span></div>
+
+
+<span class="c1"># `to_grayscale` actually predates `rgb_to_grayscale` in v1, but only handles PIL images. Since `rgb_to_grayscale` is a</span>
+<span class="c1"># superset in terms of functionality and has the same signature, we alias here to avoid disruption.</span>
+<span class="n">to_grayscale</span> <span class="o">=</span> <span class="n">rgb_to_grayscale</span>
+
+
+<span class="k">def</span> <span class="nf">_rgb_to_grayscale_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span> <span class="n">preserve_dtype</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+
+    <span class="n">r</span><span class="p">,</span> <span class="n">g</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+    <span class="n">l_img</span> <span class="o">=</span> <span class="n">r</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="mf">0.2989</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">g</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.587</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.114</span><span class="p">)</span>
+    <span class="n">l_img</span> <span class="o">=</span> <span class="n">l_img</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">preserve_dtype</span><span class="p">:</span>
+        <span class="n">l_img</span> <span class="o">=</span> <span class="n">l_img</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">num_output_channels</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">l_img</span> <span class="o">=</span> <span class="n">l_img</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">l_img</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rgb_to_grayscale</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rgb_to_grayscale</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">rgb_to_grayscale_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">num_output_channels</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;num_output_channels must be 1 or 3, got </span><span class="si">{</span><span class="n">num_output_channels</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_rgb_to_grayscale_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="n">num_output_channels</span><span class="p">,</span> <span class="n">preserve_dtype</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rgb_to_grayscale</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_rgb_to_grayscale_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">num_output_channels</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;num_output_channels must be 1 or 3, got </span><span class="si">{</span><span class="n">num_output_channels</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_FP</span><span class="o">.</span><span class="n">to_grayscale</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="n">num_output_channels</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_blend</span><span class="p">(</span><span class="n">image1</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">image2</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">ratio</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">ratio</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">ratio</span><span class="p">)</span>
+    <span class="n">fp</span> <span class="o">=</span> <span class="n">image1</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span>
+    <span class="n">bound</span> <span class="o">=</span> <span class="n">_max_value</span><span class="p">(</span><span class="n">image1</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">image1</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">ratio</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">image2</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">ratio</span><span class="p">))</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">output</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">image1</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="adjust_brightness"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.adjust_brightness.html#torchvision.transforms.v2.functional.adjust_brightness">[docs]</a><span class="k">def</span> <span class="nf">adjust_brightness</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust brightness.&quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">adjust_brightness_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="n">brightness_factor</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_brightness</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">adjust_brightness</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="n">brightness_factor</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_brightness</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_brightness</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_brightness_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">brightness_factor</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;brightness_factor (</span><span class="si">{</span><span class="n">brightness_factor</span><span class="si">}</span><span class="s2">) is not non-negative.&quot;</span><span class="p">)</span>
+
+    <span class="n">c</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">c</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input image tensor permitted channel values are 1 or 3, but found </span><span class="si">{</span><span class="n">c</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="n">fp</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span>
+    <span class="n">bound</span> <span class="o">=</span> <span class="n">_max_value</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">brightness_factor</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">output</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_brightness</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_adjust_brightness_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">_FP</span><span class="o">.</span><span class="n">adjust_brightness</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="n">brightness_factor</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_brightness</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_brightness_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">adjust_brightness_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">brightness_factor</span><span class="o">=</span><span class="n">brightness_factor</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="adjust_saturation"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.adjust_saturation.html#torchvision.transforms.v2.functional.adjust_saturation">[docs]</a><span class="k">def</span> <span class="nf">adjust_saturation</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust saturation.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">adjust_saturation_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="o">=</span><span class="n">saturation_factor</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_saturation</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">adjust_saturation</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="o">=</span><span class="n">saturation_factor</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_saturation</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_saturation</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_saturation_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">saturation_factor</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;saturation_factor (</span><span class="si">{</span><span class="n">saturation_factor</span><span class="si">}</span><span class="s2">) is not non-negative.&quot;</span><span class="p">)</span>
+
+    <span class="n">c</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">c</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input image tensor permitted channel values are 1 or 3, but found </span><span class="si">{</span><span class="n">c</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">c</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>  <span class="c1"># Match PIL behaviour</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">grayscale_image</span> <span class="o">=</span> <span class="n">_rgb_to_grayscale_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">preserve_dtype</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">():</span>
+        <span class="n">grayscale_image</span> <span class="o">=</span> <span class="n">grayscale_image</span><span class="o">.</span><span class="n">floor_</span><span class="p">()</span>
+
+    <span class="k">return</span> <span class="n">_blend</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">grayscale_image</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">)</span>
+
+
+<span class="n">_adjust_saturation_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_saturation</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">adjust_saturation</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_saturation</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_saturation_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">adjust_saturation_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">saturation_factor</span><span class="o">=</span><span class="n">saturation_factor</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="adjust_contrast"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.adjust_contrast.html#torchvision.transforms.v2.functional.adjust_contrast">[docs]</a><span class="k">def</span> <span class="nf">adjust_contrast</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.RandomAutocontrast`&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">adjust_contrast_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="n">contrast_factor</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_contrast</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="n">contrast_factor</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_contrast_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">contrast_factor</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;contrast_factor (</span><span class="si">{</span><span class="n">contrast_factor</span><span class="si">}</span><span class="s2">) is not non-negative.&quot;</span><span class="p">)</span>
+
+    <span class="n">c</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">c</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input image tensor permitted channel values are 1 or 3, but found </span><span class="si">{</span><span class="n">c</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="n">fp</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span>
+    <span class="k">if</span> <span class="n">c</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">grayscale_image</span> <span class="o">=</span> <span class="n">_rgb_to_grayscale_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">num_output_channels</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">preserve_dtype</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">fp</span><span class="p">:</span>
+            <span class="n">grayscale_image</span> <span class="o">=</span> <span class="n">grayscale_image</span><span class="o">.</span><span class="n">floor_</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">grayscale_image</span> <span class="o">=</span> <span class="n">image</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">mean</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">grayscale_image</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">3</span><span class="p">,</span> <span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">),</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_blend</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">mean</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">)</span>
+
+
+<span class="n">_adjust_contrast_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">adjust_contrast</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_contrast</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_contrast_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">adjust_contrast_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">contrast_factor</span><span class="o">=</span><span class="n">contrast_factor</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="adjust_sharpness"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.adjust_sharpness.html#torchvision.transforms.v2.functional.adjust_sharpness">[docs]</a><span class="k">def</span> <span class="nf">adjust_sharpness</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.RandomAdjustSharpness`&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">adjust_sharpness_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="o">=</span><span class="n">sharpness_factor</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_sharpness</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">adjust_sharpness</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="o">=</span><span class="n">sharpness_factor</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_sharpness</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_sharpness</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_sharpness_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:]</span>
+    <span class="k">if</span> <span class="n">num_channels</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input image tensor can have 1 or 3 channels, but found </span><span class="si">{</span><span class="n">num_channels</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">sharpness_factor</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sharpness_factor (</span><span class="si">{</span><span class="n">sharpness_factor</span><span class="si">}</span><span class="s2">) is not non-negative.&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">height</span> <span class="o">&lt;=</span> <span class="mi">2</span> <span class="ow">or</span> <span class="n">width</span> <span class="o">&lt;=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">bound</span> <span class="o">=</span> <span class="n">_max_value</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">fp</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="c1"># The following is a normalized 3x3 kernel with 1s in the edges and a 5 in the middle.</span>
+    <span class="n">kernel_dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+    <span class="n">a</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="mf">13.0</span><span class="p">,</span> <span class="mf">5.0</span> <span class="o">/</span> <span class="mf">13.0</span>
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([[</span><span class="n">a</span><span class="p">,</span> <span class="n">a</span><span class="p">,</span> <span class="n">a</span><span class="p">],</span> <span class="p">[</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">a</span><span class="p">],</span> <span class="p">[</span><span class="n">a</span><span class="p">,</span> <span class="n">a</span><span class="p">,</span> <span class="n">a</span><span class="p">]],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">kernel_dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">kernel</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">num_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+
+    <span class="c1"># We copy and cast at the same time to avoid modifications on the original data</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">kernel_dtype</span><span class="p">,</span> <span class="n">copy</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+    <span class="n">blurred_degenerate</span> <span class="o">=</span> <span class="n">conv2d</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">kernel</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">num_channels</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">fp</span><span class="p">:</span>
+        <span class="c1"># it is better to round before cast</span>
+        <span class="n">blurred_degenerate</span> <span class="o">=</span> <span class="n">blurred_degenerate</span><span class="o">.</span><span class="n">round_</span><span class="p">()</span>
+
+    <span class="c1"># Create a view on the underlying output while pointing at the same data. We do this to avoid indexing twice.</span>
+    <span class="n">view</span> <span class="o">=</span> <span class="n">output</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+    <span class="c1"># We speed up blending by minimizing flops and doing in-place. The 2 blend options are mathematically equivalent:</span>
+    <span class="c1"># x+(1-r)*(y-x) = x + (1-r)*y - (1-r)*x = x*r + y*(1-r)</span>
+    <span class="n">view</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">blurred_degenerate</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">view</span><span class="p">),</span> <span class="n">alpha</span><span class="o">=</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">sharpness_factor</span><span class="p">))</span>
+
+    <span class="c1"># The actual data of output have been modified by the above. We only need to clamp and cast now.</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">fp</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_unsquash</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="n">_adjust_sharpness_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_sharpness</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">adjust_sharpness</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_sharpness</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_sharpness_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">adjust_sharpness_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">sharpness_factor</span><span class="o">=</span><span class="n">sharpness_factor</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="adjust_hue"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.adjust_hue.html#torchvision.transforms.v2.functional.adjust_hue">[docs]</a><span class="k">def</span> <span class="nf">adjust_hue</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">hue_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust hue&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">adjust_hue_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">hue_factor</span><span class="o">=</span><span class="n">hue_factor</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_hue</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">adjust_hue</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">hue_factor</span><span class="o">=</span><span class="n">hue_factor</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_rgb_to_hsv</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">r</span><span class="p">,</span> <span class="n">g</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+
+    <span class="c1"># Implementation is based on</span>
+    <span class="c1"># https://github.com/python-pillow/Pillow/blob/4174d4267616897df3746d315d5a2d0f82c656ee/src/libImaging/Convert.c#L330</span>
+    <span class="n">minc</span><span class="p">,</span> <span class="n">maxc</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">aminmax</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+
+    <span class="c1"># The algorithm erases S and H channel where `maxc = minc`. This avoids NaN</span>
+    <span class="c1"># from happening in the results, because</span>
+    <span class="c1">#   + S channel has division by `maxc`, which is zero only if `maxc = minc`</span>
+    <span class="c1">#   + H channel has division by `(maxc - minc)`.</span>
+    <span class="c1">#</span>
+    <span class="c1"># Instead of overwriting NaN afterwards, we just prevent it from occurring so</span>
+    <span class="c1"># we don&#39;t need to deal with it in case we save the NaN in a buffer in</span>
+    <span class="c1"># backprop, if it is ever supported, but it doesn&#39;t hurt to do so.</span>
+    <span class="n">eqc</span> <span class="o">=</span> <span class="n">maxc</span> <span class="o">==</span> <span class="n">minc</span>
+
+    <span class="n">channels_range</span> <span class="o">=</span> <span class="n">maxc</span> <span class="o">-</span> <span class="n">minc</span>
+    <span class="c1"># Since `eqc =&gt; channels_range = 0`, replacing denominator with 1 when `eqc` is fine.</span>
+    <span class="n">ones</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="n">maxc</span><span class="p">)</span>
+    <span class="n">s</span> <span class="o">=</span> <span class="n">channels_range</span> <span class="o">/</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">eqc</span><span class="p">,</span> <span class="n">ones</span><span class="p">,</span> <span class="n">maxc</span><span class="p">)</span>
+    <span class="c1"># Note that `eqc =&gt; maxc = minc = r = g = b`. So the following calculation</span>
+    <span class="c1"># of `h` would reduce to `bc - gc + 2 + rc - bc + 4 + rc - bc = 6` so it</span>
+    <span class="c1"># would not matter what values `rc`, `gc`, and `bc` have here, and thus</span>
+    <span class="c1"># replacing denominator with 1 when `eqc` is fine.</span>
+    <span class="n">channels_range_divisor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">eqc</span><span class="p">,</span> <span class="n">ones</span><span class="p">,</span> <span class="n">channels_range</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze_</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+    <span class="n">rc</span><span class="p">,</span> <span class="n">gc</span><span class="p">,</span> <span class="n">bc</span> <span class="o">=</span> <span class="p">((</span><span class="n">maxc</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span> <span class="o">-</span> <span class="n">image</span><span class="p">)</span> <span class="o">/</span> <span class="n">channels_range_divisor</span><span class="p">)</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+
+    <span class="n">mask_maxc_neq_r</span> <span class="o">=</span> <span class="n">maxc</span> <span class="o">!=</span> <span class="n">r</span>
+    <span class="n">mask_maxc_eq_g</span> <span class="o">=</span> <span class="n">maxc</span> <span class="o">==</span> <span class="n">g</span>
+
+    <span class="n">hg</span> <span class="o">=</span> <span class="n">rc</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="mf">2.0</span><span class="p">)</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">bc</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">mask_maxc_eq_g</span> <span class="o">&amp;</span> <span class="n">mask_maxc_neq_r</span><span class="p">)</span>
+    <span class="n">hr</span> <span class="o">=</span> <span class="n">bc</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">gc</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="o">~</span><span class="n">mask_maxc_neq_r</span><span class="p">)</span>
+    <span class="n">hb</span> <span class="o">=</span> <span class="n">gc</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mf">4.0</span><span class="p">)</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">rc</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">mask_maxc_neq_r</span><span class="o">.</span><span class="n">logical_and_</span><span class="p">(</span><span class="n">mask_maxc_eq_g</span><span class="o">.</span><span class="n">logical_not_</span><span class="p">()))</span>
+
+    <span class="n">h</span> <span class="o">=</span> <span class="n">hr</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">hg</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">hb</span><span class="p">)</span>
+    <span class="n">h</span> <span class="o">=</span> <span class="n">h</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">/</span> <span class="mf">6.0</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mf">1.0</span><span class="p">)</span><span class="o">.</span><span class="n">fmod_</span><span class="p">(</span><span class="mf">1.0</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">((</span><span class="n">h</span><span class="p">,</span> <span class="n">s</span><span class="p">,</span> <span class="n">maxc</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_hsv_to_rgb</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">h</span><span class="p">,</span> <span class="n">s</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+    <span class="n">h6</span> <span class="o">=</span> <span class="n">h</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="mi">6</span><span class="p">)</span>
+    <span class="n">i</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="n">h6</span><span class="p">)</span>
+    <span class="n">f</span> <span class="o">=</span> <span class="n">h6</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">i</span><span class="p">)</span>
+    <span class="n">i</span> <span class="o">=</span> <span class="n">i</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">)</span>
+
+    <span class="n">sxf</span> <span class="o">=</span> <span class="n">s</span> <span class="o">*</span> <span class="n">f</span>
+    <span class="n">one_minus_s</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">s</span>
+    <span class="n">q</span> <span class="o">=</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">sxf</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">v</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">t</span> <span class="o">=</span> <span class="n">sxf</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">one_minus_s</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">v</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">p</span> <span class="o">=</span> <span class="n">one_minus_s</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">v</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">i</span><span class="o">.</span><span class="n">remainder_</span><span class="p">(</span><span class="mi">6</span><span class="p">)</span>
+
+    <span class="n">vpqt</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">((</span><span class="n">v</span><span class="p">,</span> <span class="n">p</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">t</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+
+    <span class="c1"># vpqt -&gt; rgb mapping based on i</span>
+    <span class="n">select</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">]],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)</span>
+    <span class="n">select</span> <span class="o">=</span> <span class="n">select</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="n">img</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">non_blocking</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="n">select</span> <span class="o">=</span> <span class="n">select</span><span class="p">[:,</span> <span class="n">i</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">select</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="c1"># if input.shape is (B, ..., C, H, W) then</span>
+        <span class="c1"># select.shape is (C, B, ...,  H, W)</span>
+        <span class="c1"># thus we move C axis to get (B, ..., C, H, W)</span>
+        <span class="n">select</span> <span class="o">=</span> <span class="n">select</span><span class="o">.</span><span class="n">moveaxis</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="o">-</span><span class="mi">3</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">vpqt</span><span class="o">.</span><span class="n">gather</span><span class="p">(</span><span class="o">-</span><span class="mi">3</span><span class="p">,</span> <span class="n">select</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_hue</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_hue</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_hue_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">hue_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="o">-</span><span class="mf">0.5</span> <span class="o">&lt;=</span> <span class="n">hue_factor</span> <span class="o">&lt;=</span> <span class="mf">0.5</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;hue_factor (</span><span class="si">{</span><span class="n">hue_factor</span><span class="si">}</span><span class="s2">) is not in [-0.5, 0.5].&quot;</span><span class="p">)</span>
+
+    <span class="n">c</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">c</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input image tensor permitted channel values are 1 or 3, but found </span><span class="si">{</span><span class="n">c</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">c</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>  <span class="c1"># Match PIL behaviour</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="c1"># exit earlier on empty images</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">orig_dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">image</span> <span class="o">=</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="n">image</span> <span class="o">=</span> <span class="n">_rgb_to_hsv</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="n">h</span><span class="p">,</span> <span class="n">s</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">unbind</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+    <span class="n">h</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">hue_factor</span><span class="p">)</span><span class="o">.</span><span class="n">remainder_</span><span class="p">(</span><span class="mf">1.0</span><span class="p">)</span>
+    <span class="n">image</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">((</span><span class="n">h</span><span class="p">,</span> <span class="n">s</span><span class="p">,</span> <span class="n">v</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+    <span class="n">image_hue_adj</span> <span class="o">=</span> <span class="n">_hsv_to_rgb</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">image_hue_adj</span><span class="p">,</span> <span class="n">orig_dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="n">_adjust_hue_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_hue</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">adjust_hue</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_hue</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_hue_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">hue_factor</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">adjust_hue_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">hue_factor</span><span class="o">=</span><span class="n">hue_factor</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="adjust_gamma"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.adjust_gamma.html#torchvision.transforms.v2.functional.adjust_gamma">[docs]</a><span class="k">def</span> <span class="nf">adjust_gamma</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">gamma</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">gain</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Adjust gamma.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">adjust_gamma_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">gamma</span><span class="o">=</span><span class="n">gamma</span><span class="p">,</span> <span class="n">gain</span><span class="o">=</span><span class="n">gain</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">adjust_gamma</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">adjust_gamma</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">gamma</span><span class="o">=</span><span class="n">gamma</span><span class="p">,</span> <span class="n">gain</span><span class="o">=</span><span class="n">gain</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_gamma</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_gamma</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_gamma_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">gamma</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">gain</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.0</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">gamma</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Gamma should be a non-negative real number&quot;</span><span class="p">)</span>
+
+    <span class="c1"># The input image is either assumed to be at [0, 1] scale (if float) or is converted to that scale (if integer).</span>
+    <span class="c1"># Since the gamma is non-negative, the output remains at [0, 1] scale.</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">image</span><span class="p">):</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span><span class="o">.</span><span class="n">pow_</span><span class="p">(</span><span class="n">gamma</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">pow</span><span class="p">(</span><span class="n">gamma</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">gain</span> <span class="o">!=</span> <span class="mf">1.0</span><span class="p">:</span>
+        <span class="c1"># The clamp operation is needed only if multiplication is performed. It&#39;s only when gain != 1, that the scale</span>
+        <span class="c1"># of the output can go beyond [0, 1].</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">gain</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="n">_adjust_gamma_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_gamma</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">adjust_gamma</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">adjust_gamma</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">adjust_gamma_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">gamma</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">gain</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">adjust_gamma_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">gamma</span><span class="o">=</span><span class="n">gamma</span><span class="p">,</span> <span class="n">gain</span><span class="o">=</span><span class="n">gain</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="posterize"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.posterize.html#torchvision.transforms.v2.functional.posterize">[docs]</a><span class="k">def</span> <span class="nf">posterize</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">bits</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomPosterize` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">posterize_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">bits</span><span class="o">=</span><span class="n">bits</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">posterize</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">posterize</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">bits</span><span class="o">=</span><span class="n">bits</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">posterize</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">posterize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">posterize_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">bits</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">():</span>
+        <span class="n">levels</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">&lt;&lt;</span> <span class="n">bits</span>
+        <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">levels</span><span class="p">)</span><span class="o">.</span><span class="n">floor_</span><span class="p">()</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">levels</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">/</span> <span class="n">levels</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">num_value_bits</span> <span class="o">=</span> <span class="n">_num_value_bits</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">bits</span> <span class="o">&gt;=</span> <span class="n">num_value_bits</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span>
+
+        <span class="n">mask</span> <span class="o">=</span> <span class="p">((</span><span class="mi">1</span> <span class="o">&lt;&lt;</span> <span class="n">bits</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">&lt;&lt;</span> <span class="p">(</span><span class="n">num_value_bits</span> <span class="o">-</span> <span class="n">bits</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">image</span> <span class="o">&amp;</span> <span class="n">mask</span>
+
+
+<span class="n">_posterize_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">posterize</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">posterize</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">posterize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">posterize_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">bits</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">posterize_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">bits</span><span class="o">=</span><span class="n">bits</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="solarize"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.solarize.html#torchvision.transforms.v2.functional.solarize">[docs]</a><span class="k">def</span> <span class="nf">solarize</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomSolarize` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">solarize_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">threshold</span><span class="o">=</span><span class="n">threshold</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">solarize</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">solarize</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">threshold</span><span class="o">=</span><span class="n">threshold</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">solarize</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">solarize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">solarize_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">threshold</span> <span class="o">&gt;</span> <span class="n">_max_value</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Threshold should be less or equal the maximum value of the dtype, but got </span><span class="si">{</span><span class="n">threshold</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">image</span> <span class="o">&gt;=</span> <span class="n">threshold</span><span class="p">,</span> <span class="n">invert_image</span><span class="p">(</span><span class="n">image</span><span class="p">),</span> <span class="n">image</span><span class="p">)</span>
+
+
+<span class="n">_solarize_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">solarize</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">solarize</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">solarize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">solarize_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">threshold</span><span class="p">:</span> <span class="nb">float</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">solarize_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">threshold</span><span class="o">=</span><span class="n">threshold</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="autocontrast"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.autocontrast.html#torchvision.transforms.v2.functional.autocontrast">[docs]</a><span class="k">def</span> <span class="nf">autocontrast</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomAutocontrast` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">autocontrast_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">autocontrast</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">autocontrast</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">autocontrast</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">autocontrast</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">autocontrast_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">c</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">c</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input image tensor permitted channel values are 1 or 3, but found </span><span class="si">{</span><span class="n">c</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="c1"># exit earlier on empty images</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">bound</span> <span class="o">=</span> <span class="n">_max_value</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">fp</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span>
+    <span class="n">float_image</span> <span class="o">=</span> <span class="n">image</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+
+    <span class="n">minimum</span> <span class="o">=</span> <span class="n">float_image</span><span class="o">.</span><span class="n">amin</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">),</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+    <span class="n">maximum</span> <span class="o">=</span> <span class="n">float_image</span><span class="o">.</span><span class="n">amax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">),</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="n">eq_idxs</span> <span class="o">=</span> <span class="n">maximum</span> <span class="o">==</span> <span class="n">minimum</span>
+    <span class="n">inv_scale</span> <span class="o">=</span> <span class="n">maximum</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">minimum</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">/</span> <span class="n">bound</span><span class="p">)</span>
+    <span class="n">minimum</span><span class="p">[</span><span class="n">eq_idxs</span><span class="p">]</span> <span class="o">=</span> <span class="mf">0.0</span>
+    <span class="n">inv_scale</span><span class="p">[</span><span class="n">eq_idxs</span><span class="p">]</span> <span class="o">=</span> <span class="mf">1.0</span>
+
+    <span class="k">if</span> <span class="n">fp</span><span class="p">:</span>
+        <span class="n">diff</span> <span class="o">=</span> <span class="n">float_image</span><span class="o">.</span><span class="n">sub</span><span class="p">(</span><span class="n">minimum</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">diff</span> <span class="o">=</span> <span class="n">float_image</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">minimum</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">diff</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">inv_scale</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bound</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+
+<span class="n">_autocontrast_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">autocontrast</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">autocontrast</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">autocontrast</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">autocontrast_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">autocontrast_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="equalize"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.equalize.html#torchvision.transforms.v2.functional.equalize">[docs]</a><span class="k">def</span> <span class="nf">equalize</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomEqualize` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">equalize_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">equalize</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">equalize</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">equalize</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">equalize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">equalize_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="c1"># 1. The algorithm below can easily be extended to support arbitrary integer dtypes. However, the histogram that</span>
+    <span class="c1">#    would be needed to computed will have at least `torch.iinfo(dtype).max + 1` values. That is perfectly fine for</span>
+    <span class="c1">#    `torch.int8`, `torch.uint8`, and `torch.int16`, at least questionable for `torch.int32` and completely</span>
+    <span class="c1">#    unfeasible for `torch.int64`.</span>
+    <span class="c1"># 2. Floating point inputs need to be binned for this algorithm. Apart from converting them to an integer dtype, we</span>
+    <span class="c1">#    could also use PyTorch&#39;s builtin histogram functionality. However, that has its own set of issues: in addition</span>
+    <span class="c1">#    to being slow in general, PyTorch&#39;s implementation also doesn&#39;t support batches. In total, that makes it slower</span>
+    <span class="c1">#    and more complicated to implement than a simple conversion and a fast histogram implementation for integers.</span>
+    <span class="c1"># Since we need to convert in most cases anyway and out of the acceptable dtypes mentioned in 1. `torch.uint8` is</span>
+    <span class="c1"># by far the most common, we choose it as base.</span>
+    <span class="n">output_dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">image</span> <span class="o">=</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="c1"># The histogram is computed by using the flattened image as index. For example, a pixel value of 127 in the image</span>
+    <span class="c1"># corresponds to adding 1 to index 127 in the histogram.</span>
+    <span class="n">batch_shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[:</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+    <span class="n">flat_image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">start_dim</span><span class="o">=-</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)</span>
+    <span class="n">hist</span> <span class="o">=</span> <span class="n">flat_image</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">(</span><span class="n">batch_shape</span> <span class="o">+</span> <span class="p">(</span><span class="mi">256</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">)</span>
+    <span class="n">hist</span><span class="o">.</span><span class="n">scatter_add_</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span> <span class="n">index</span><span class="o">=</span><span class="n">flat_image</span><span class="p">,</span> <span class="n">src</span><span class="o">=</span><span class="n">hist</span><span class="o">.</span><span class="n">new_ones</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">expand_as</span><span class="p">(</span><span class="n">flat_image</span><span class="p">))</span>
+    <span class="n">cum_hist</span> <span class="o">=</span> <span class="n">hist</span><span class="o">.</span><span class="n">cumsum</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="c1"># The simplest form of lookup-table (LUT) that also achieves histogram equalization is</span>
+    <span class="c1"># `lut = cum_hist / flat_image.shape[-1] * 255`</span>
+    <span class="c1"># However, PIL uses a more elaborate scheme:</span>
+    <span class="c1"># https://github.com/python-pillow/Pillow/blob/eb59cb61d5239ee69cbbf12709a0c6fd7314e6d7/src/PIL/ImageOps.py#L368-L385</span>
+    <span class="c1"># `lut = ((cum_hist + num_non_max_pixels // (2 * 255)) // num_non_max_pixels) * 255`</span>
+
+    <span class="c1"># The last non-zero element in the histogram is the first element in the cumulative histogram with the maximum</span>
+    <span class="c1"># value. Thus, the &quot;max&quot; in `num_non_max_pixels` does not refer to 255 as the maximum value of uint8 images, but</span>
+    <span class="c1"># rather the maximum value in the image, which might be or not be 255.</span>
+    <span class="n">index</span> <span class="o">=</span> <span class="n">cum_hist</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">num_non_max_pixels</span> <span class="o">=</span> <span class="n">flat_image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">hist</span><span class="o">.</span><span class="n">gather</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span> <span class="n">index</span><span class="o">=</span><span class="n">index</span><span class="o">.</span><span class="n">unsqueeze_</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+
+    <span class="c1"># This is performance optimization that saves us one multiplication later. With this, the LUT computation simplifies</span>
+    <span class="c1"># to `lut = (cum_hist + step // 2) // step` and thus saving the final multiplication by 255 while keeping the</span>
+    <span class="c1"># division count the same. PIL uses the variable name `step` for this, so we keep that for easier comparison.</span>
+    <span class="n">step</span> <span class="o">=</span> <span class="n">num_non_max_pixels</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="mi">255</span><span class="p">,</span> <span class="n">rounding_mode</span><span class="o">=</span><span class="s2">&quot;floor&quot;</span><span class="p">)</span>
+
+    <span class="c1"># Although it looks like we could return early if we find `step == 0` like PIL does, that is unfortunately not as</span>
+    <span class="c1"># easy due to our support for batched images. We can only return early if `(step == 0).all()` holds. If it doesn&#39;t,</span>
+    <span class="c1"># we have to go through the computation below anyway. Since `step == 0` is an edge case anyway, it makes no sense to</span>
+    <span class="c1"># pay the runtime cost for checking it every time.</span>
+    <span class="n">valid_equalization</span> <span class="o">=</span> <span class="n">step</span><span class="o">.</span><span class="n">ne</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze_</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="c1"># `lut[k]` is computed with `cum_hist[k-1]` with `lut[0] == (step // 2) // step == 0`. Thus, we perform the</span>
+    <span class="c1"># computation only for `lut[1:]` with `cum_hist[:-1]` and add `lut[0] == 0` afterwards.</span>
+    <span class="n">cum_hist</span> <span class="o">=</span> <span class="n">cum_hist</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+    <span class="p">(</span>
+        <span class="n">cum_hist</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">step</span> <span class="o">//</span> <span class="mi">2</span><span class="p">)</span>
+        <span class="c1"># We need the `clamp_`(min=1) call here to avoid zero division since they fail for integer dtypes. This has no</span>
+        <span class="c1"># effect on the returned result of this kernel since images inside the batch with `step == 0` are returned as is</span>
+        <span class="c1"># instead of equalized version.</span>
+        <span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">step</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span> <span class="n">rounding_mode</span><span class="o">=</span><span class="s2">&quot;floor&quot;</span><span class="p">)</span>
+        <span class="c1"># We need the `clamp_` call here since PILs LUT computation scheme can produce values outside the valid value</span>
+        <span class="c1"># range of uint8 images</span>
+        <span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">255</span><span class="p">)</span>
+    <span class="p">)</span>
+    <span class="n">lut</span> <span class="o">=</span> <span class="n">cum_hist</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+    <span class="n">lut</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">lut</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">batch_shape</span> <span class="o">+</span> <span class="p">(</span><span class="mi">1</span><span class="p">,)),</span> <span class="n">lut</span><span class="p">],</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">equalized_image</span> <span class="o">=</span> <span class="n">lut</span><span class="o">.</span><span class="n">gather</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span> <span class="n">index</span><span class="o">=</span><span class="n">flat_image</span><span class="p">)</span><span class="o">.</span><span class="n">view_as</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">valid_equalization</span><span class="p">,</span> <span class="n">equalized_image</span><span class="p">,</span> <span class="n">image</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">output_dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+
+<span class="n">_equalize_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">equalize</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">equalize</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">equalize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">equalize_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">equalize_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="invert"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.invert.html#torchvision.transforms.v2.functional.invert">[docs]</a><span class="k">def</span> <span class="nf">invert</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :func:`~torchvision.transforms.v2.RandomInvert`.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">invert_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">invert</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">invert</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">invert</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">invert</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">invert_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">():</span>
+        <span class="k">return</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">image</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">bitwise_not</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>  <span class="c1"># signed integer dtypes</span>
+        <span class="c1"># We can&#39;t use `Tensor.bitwise_not` here, since we want to retain the leading zero bit that encodes the sign</span>
+        <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">bitwise_xor</span><span class="p">((</span><span class="mi">1</span> <span class="o">&lt;&lt;</span> <span class="n">_num_value_bits</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">))</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+
+
+<span class="n">_invert_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">invert</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">invert</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">invert</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">invert_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">invert_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="permute_channels"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.permute_channels.html#torchvision.transforms.v2.functional.permute_channels">[docs]</a><span class="k">def</span> <span class="nf">permute_channels</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">permutation</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Permute the channels of the input according to the given permutation.</span>
+
+<span class="sd">    This function supports plain :class:`~torch.Tensor`&#39;s, :class:`PIL.Image.Image`&#39;s, and</span>
+<span class="sd">    :class:`torchvision.tv_tensors.Image` and :class:`torchvision.tv_tensors.Video`.</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        &gt;&gt;&gt; rgb_image = torch.rand(3, 256, 256)</span>
+<span class="sd">        &gt;&gt;&gt; bgr_image = F.permutate_channels(rgb_image, permutation=[2, 1, 0])</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        permutation (List[int]): Valid permutation of the input channel indices. The index of the element determines the</span>
+<span class="sd">            channel index in the input and the value determines the channel index in the output. For example,</span>
+<span class="sd">            ``permutation=[2, 0 , 1]``</span>
+
+<span class="sd">            - takes ``ìnpt[..., 0, :, :]`` and puts it at ``output[..., 2, :, :]``,</span>
+<span class="sd">            - takes ``ìnpt[..., 1, :, :]`` and puts it at ``output[..., 0, :, :]``, and</span>
+<span class="sd">            - takes ``ìnpt[..., 2, :, :]`` and puts it at ``output[..., 1, :, :]``.</span>
+
+<span class="sd">    Raises:</span>
+<span class="sd">        ValueError: If ``len(permutation)`` doesn&#39;t match the number of channels in the input.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">permute_channels_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">permutation</span><span class="o">=</span><span class="n">permutation</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">permute_channels</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">permute_channels</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">permutation</span><span class="o">=</span><span class="n">permutation</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">permute_channels</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">permute_channels</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">permute_channels_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">permutation</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:]</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">permutation</span><span class="p">)</span> <span class="o">!=</span> <span class="n">num_channels</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Length of permutation does not match number of channels: &quot;</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">permutation</span><span class="p">)</span><span class="si">}</span><span class="s2"> != </span><span class="si">{</span><span class="n">num_channels</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+    <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="p">[:,</span> <span class="n">permutation</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span>
+    <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">permute_channels</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_permute_channels_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">permutation</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">to_pil_image</span><span class="p">(</span><span class="n">permute_channels_image</span><span class="p">(</span><span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">image</span><span class="p">),</span> <span class="n">permutation</span><span class="o">=</span><span class="n">permutation</span><span class="p">))</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">permute_channels</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">permute_channels_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">permutation</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">permute_channels_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">permutation</span><span class="o">=</span><span class="n">permutation</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html b/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
new file mode 100644
index 00000000000..29881743f27
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
@@ -0,0 +1,693 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._deprecated &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._deprecated</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._deprecated</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">_F</span>
+
+
+<div class="viewcode-block" id="to_tensor"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.to_tensor.html#torchvision.transforms.v2.functional.to_tensor">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+<span class="k">def</span> <span class="nf">to_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] [DEPREACTED] Use to_image() and to_dtype() instead.&quot;&quot;&quot;</span>
+    <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+        <span class="s2">&quot;The function `to_tensor(...)` is deprecated and will be removed in a future release. &quot;</span>
+        <span class="s2">&quot;Instead, please use `to_image(...)` followed by `to_dtype(..., dtype=torch.float32, scale=True)`.&quot;</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_F</span><span class="o">.</span><span class="n">to_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">get_image_size</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+        <span class="s2">&quot;The function `get_image_size(...)` is deprecated and will be removed in a future release. &quot;</span>
+        <span class="s2">&quot;Instead, please use `get_size(...)` which returns `[h, w]` instead of `[w, h]`.&quot;</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">_F</span><span class="o">.</span><span class="n">get_image_size</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html b/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
new file mode 100644
index 00000000000..26d15534520
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
@@ -0,0 +1,3049 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._geometry &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._geometry</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._geometry</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">numbers</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch.nn.functional</span> <span class="kn">import</span> <span class="n">grid_sample</span><span class="p">,</span> <span class="n">interpolate</span><span class="p">,</span> <span class="n">pad</span> <span class="k">as</span> <span class="n">torch_pad</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">_functional_pil</span> <span class="k">as</span> <span class="n">_FP</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms._functional_tensor</span> <span class="kn">import</span> <span class="n">_pad_symmetric</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.functional</span> <span class="kn">import</span> <span class="p">(</span>
+    <span class="n">_check_antialias</span><span class="p">,</span>
+    <span class="n">_compute_resized_output_size</span> <span class="k">as</span> <span class="n">__compute_resized_output_size</span><span class="p">,</span>
+    <span class="n">_get_perspective_coeffs</span><span class="p">,</span>
+    <span class="n">_interpolation_modes_from_int</span><span class="p">,</span>
+    <span class="n">InterpolationMode</span><span class="p">,</span>
+    <span class="n">pil_modes_mapping</span><span class="p">,</span>
+    <span class="n">pil_to_tensor</span><span class="p">,</span>
+    <span class="n">to_pil_image</span><span class="p">,</span>
+<span class="p">)</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">._meta</span> <span class="kn">import</span> <span class="n">_get_size_image_pil</span><span class="p">,</span> <span class="n">clamp_bounding_boxes</span><span class="p">,</span> <span class="n">convert_bounding_box_format</span>
+
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_FillTypeJIT</span><span class="p">,</span> <span class="n">_get_kernel</span><span class="p">,</span> <span class="n">_register_five_ten_crop_kernel_internal</span><span class="p">,</span> <span class="n">_register_kernel_internal</span>
+
+
+<span class="k">def</span> <span class="nf">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">InterpolationMode</span><span class="p">:</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_interpolation_modes_from_int</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Argument interpolation should be an `InterpolationMode` or a corresponding Pillow integer constant, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;but got </span><span class="si">{</span><span class="n">interpolation</span><span class="si">}</span><span class="s2">.&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">interpolation</span>
+
+
+<div class="viewcode-block" id="horizontal_flip"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.horizontal_flip.html#torchvision.transforms.v2.functional.horizontal_flip">[docs]</a><span class="k">def</span> <span class="nf">horizontal_flip</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomHorizontalFlip` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">horizontal_flip_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">horizontal_flip_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">flip</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_horizontal_flip_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">_FP</span><span class="o">.</span><span class="n">hflip</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">horizontal_flip_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">horizontal_flip_image</span><span class="p">(</span><span class="n">mask</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">horizontal_flip_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span> <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">shape</span>
+
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">format</span> <span class="o">==</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">]]</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">]]</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">neg_</span><span class="p">()</span>
+    <span class="k">elif</span> <span class="nb">format</span> <span class="o">==</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYWH</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">])</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">neg_</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>  <span class="c1"># format == tv_tensors.BoundingBoxFormat.CXCYWH:</span>
+        <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">neg_</span><span class="p">()</span>
+
+    <span class="k">return</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_horizontal_flip_bounding_boxes_dispatch</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">horizontal_flip_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">horizontal_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">horizontal_flip_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">horizontal_flip_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="vertical_flip"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.vertical_flip.html#torchvision.transforms.v2.functional.vertical_flip">[docs]</a><span class="k">def</span> <span class="nf">vertical_flip</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomVerticalFlip` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">vertical_flip_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">vertical_flip_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">flip</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_vertical_flip_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">_FP</span><span class="o">.</span><span class="n">vflip</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">vertical_flip_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">vertical_flip_image</span><span class="p">(</span><span class="n">mask</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">vertical_flip_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span> <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">shape</span>
+
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">format</span> <span class="o">==</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">]]</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">]]</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="o">.</span><span class="n">neg_</span><span class="p">()</span>
+    <span class="k">elif</span> <span class="nb">format</span> <span class="o">==</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYWH</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">])</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="o">.</span><span class="n">neg_</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>  <span class="c1"># format == tv_tensors.BoundingBoxFormat.CXCYWH:</span>
+        <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="o">.</span><span class="n">neg_</span><span class="p">()</span>
+
+    <span class="k">return</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_vertical_flip_bounding_boxes_dispatch</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">vertical_flip_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">vertical_flip</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">vertical_flip_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">vertical_flip_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<span class="c1"># We changed the names to align them with the transforms, i.e. `RandomHorizontalFlip`. Still, `hflip` and `vflip` are</span>
+<span class="c1"># prevalent and well understood. Thus, we just alias them without deprecating the old names.</span>
+<span class="n">hflip</span> <span class="o">=</span> <span class="n">horizontal_flip</span>
+<span class="n">vflip</span> <span class="o">=</span> <span class="n">vertical_flip</span>
+
+
+<span class="k">def</span> <span class="nf">_compute_resized_output_size</span><span class="p">(</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">[</span><span class="n">size</span><span class="p">]</span>
+    <span class="k">elif</span> <span class="n">max_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="s2">&quot;max_size should only be passed if size specifies the length of the smaller edge, &quot;</span>
+            <span class="s2">&quot;i.e. size should be an int or a sequence of length 1 in torchscript mode.&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">__compute_resized_output_size</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="resize"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.resize.html#torchvision.transforms.v2.functional.resize">[docs]</a><span class="k">def</span> <span class="nf">resize</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.Resize` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">resize_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">resize</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">resize</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resize</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">resize_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="n">antialias</span> <span class="o">=</span> <span class="n">_check_antialias</span><span class="p">(</span><span class="n">img</span><span class="o">=</span><span class="n">image</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="k">assert</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">antialias</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span>
+    <span class="n">antialias</span> <span class="o">=</span> <span class="kc">False</span> <span class="k">if</span> <span class="n">antialias</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">antialias</span>
+    <span class="n">align_corners</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="k">if</span> <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span> <span class="ow">or</span> <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">:</span>
+        <span class="n">align_corners</span> <span class="o">=</span> <span class="kc">False</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="c1"># The default of antialias should be True from 0.17, so we don&#39;t warn or</span>
+        <span class="c1"># error if other interpolation modes are used. This is documented.</span>
+        <span class="n">antialias</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">numel</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span>
+    <span class="n">num_channels</span><span class="p">,</span> <span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:]</span>
+    <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span> <span class="o">=</span> <span class="n">_compute_resized_output_size</span><span class="p">((</span><span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span><span class="p">),</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="p">(</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">)</span> <span class="o">==</span> <span class="p">(</span><span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">image</span>
+    <span class="k">elif</span> <span class="n">numel</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span><span class="p">)</span>
+
+        <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span>
+        <span class="n">acceptable_dtypes</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">float64</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span> <span class="ow">or</span> <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST_EXACT</span><span class="p">:</span>
+            <span class="c1"># uint8 dtype can be included for cpu and cuda input if nearest mode</span>
+            <span class="n">acceptable_dtypes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span><span class="p">:</span>
+            <span class="c1"># uint8 dtype support for bilinear and bicubic is limited to cpu and</span>
+            <span class="c1"># according to our benchmarks, non-AVX CPUs should still prefer u8-&gt;f32-&gt;interpolate-&gt;u8 path for bilinear</span>
+            <span class="k">if</span> <span class="p">(</span><span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span> <span class="ow">and</span> <span class="s2">&quot;AVX2&quot;</span> <span class="ow">in</span> <span class="n">torch</span><span class="o">.</span><span class="n">backends</span><span class="o">.</span><span class="n">cpu</span><span class="o">.</span><span class="n">get_cpu_capability</span><span class="p">())</span> <span class="ow">or</span> <span class="p">(</span>
+                <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span>
+            <span class="p">):</span>
+                <span class="n">acceptable_dtypes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+
+        <span class="n">strides</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">stride</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">is_contiguous</span><span class="p">(</span><span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">channels_last</span><span class="p">)</span> <span class="ow">and</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">numel</span> <span class="o">!=</span> <span class="n">strides</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="c1"># There is a weird behaviour in torch core where the output tensor of `interpolate()` can be allocated as</span>
+            <span class="c1"># contiguous even though the input is un-ambiguously channels_last (https://github.com/pytorch/pytorch/issues/68430).</span>
+            <span class="c1"># In particular this happens for the typical torchvision use-case of single CHW images where we fake the batch dim</span>
+            <span class="c1"># to become 1CHW. Below, we restride those tensors to trick torch core into properly allocating the output as</span>
+            <span class="c1"># channels_last, thus preserving the memory format of the input. This is not just for format consistency:</span>
+            <span class="c1"># for uint8 bilinear images, this also avoids an extra copy (re-packing) of the output and saves time.</span>
+            <span class="c1"># TODO: when https://github.com/pytorch/pytorch/issues/68430 is fixed (possibly by https://github.com/pytorch/pytorch/pull/100373),</span>
+            <span class="c1"># we should be able to remove this hack.</span>
+            <span class="n">new_strides</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">strides</span><span class="p">)</span>
+            <span class="n">new_strides</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="n">numel</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">as_strided</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span><span class="p">),</span> <span class="n">new_strides</span><span class="p">)</span>
+
+        <span class="n">need_cast</span> <span class="o">=</span> <span class="n">dtype</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">acceptable_dtypes</span>
+        <span class="k">if</span> <span class="n">need_cast</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+
+        <span class="n">image</span> <span class="o">=</span> <span class="n">interpolate</span><span class="p">(</span>
+            <span class="n">image</span><span class="p">,</span>
+            <span class="n">size</span><span class="o">=</span><span class="p">[</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">],</span>
+            <span class="n">mode</span><span class="o">=</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span>
+            <span class="n">align_corners</span><span class="o">=</span><span class="n">align_corners</span><span class="p">,</span>
+            <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">need_cast</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">interpolation</span> <span class="o">==</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span> <span class="ow">and</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+                <span class="c1"># This path is hit on non-AVX archs, or on GPU.</span>
+                <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="mi">255</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">dtype</span> <span class="ow">in</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">int8</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">int16</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">):</span>
+                <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">round_</span><span class="p">()</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">[:</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span> <span class="o">+</span> <span class="p">(</span><span class="n">num_channels</span><span class="p">,</span> <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">))</span>
+
+
+<span class="k">def</span> <span class="nf">_resize_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">width</span>
+    <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span> <span class="o">=</span> <span class="n">_compute_resized_output_size</span><span class="p">(</span>
+        <span class="p">(</span><span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span><span class="p">),</span>
+        <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span>  <span class="c1"># type: ignore[arg-type]</span>
+        <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="p">(</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">)</span> <span class="o">==</span> <span class="p">(</span><span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">resize</span><span class="p">((</span><span class="n">new_width</span><span class="p">,</span> <span class="n">new_height</span><span class="p">),</span> <span class="n">resample</span><span class="o">=</span><span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">])</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resize</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">__resize_image_pil_dispatch</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">antialias</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Anti-alias option is always applied for PIL Image input. Argument antialias is ignored.&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resize_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">resize_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">resize_image</span><span class="p">(</span><span class="n">mask</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_resize_mask_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">resize_mask</span><span class="p">(</span><span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="n">size</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">resize_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+    <span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span> <span class="o">=</span> <span class="n">canvas_size</span>
+    <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span> <span class="o">=</span> <span class="n">_compute_resized_output_size</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="p">(</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">)</span> <span class="o">==</span> <span class="p">(</span><span class="n">old_height</span><span class="p">,</span> <span class="n">old_width</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">canvas_size</span>
+
+    <span class="n">w_ratio</span> <span class="o">=</span> <span class="n">new_width</span> <span class="o">/</span> <span class="n">old_width</span>
+    <span class="n">h_ratio</span> <span class="o">=</span> <span class="n">new_height</span> <span class="o">/</span> <span class="n">old_height</span>
+    <span class="n">ratios</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">w_ratio</span><span class="p">,</span> <span class="n">h_ratio</span><span class="p">,</span> <span class="n">w_ratio</span><span class="p">,</span> <span class="n">h_ratio</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="k">return</span> <span class="p">(</span>
+        <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">ratios</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span>
+        <span class="p">(</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">),</span>
+    <span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_resize_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">resize_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">resize_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">max_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">resize_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">max_size</span><span class="o">=</span><span class="n">max_size</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="affine"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.affine.html#torchvision.transforms.v2.functional.affine">[docs]</a><span class="k">def</span> <span class="nf">affine</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomAffine` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">affine_image</span><span class="p">(</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+            <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span>
+            <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+            <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+            <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">affine</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">affine</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_affine_parse_args</span><span class="p">(</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">InterpolationMode</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">angle</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument angle should be int or float&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">translate</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument translate should be a sequence&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">translate</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument translate should be a sequence of length 2&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">scale</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument scale should be positive&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="p">(</span><span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Shear should be either a single value or a sequence of two values&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument interpolation should be a InterpolationMode&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">angle</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">angle</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">angle</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">translate</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
+        <span class="n">translate</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">translate</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="n">shear</span> <span class="o">=</span> <span class="p">[</span><span class="n">shear</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">shear</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
+        <span class="n">shear</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">shear</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">shear</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">shear</span> <span class="o">=</span> <span class="p">[</span><span class="n">shear</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">shear</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">shear</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Shear should be a sequence containing two values. Got </span><span class="si">{</span><span class="n">shear</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument center should be a sequence&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">center</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">c</span><span class="p">)</span> <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="n">center</span><span class="p">]</span>
+
+    <span class="k">return</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">center</span>
+
+
+<span class="k">def</span> <span class="nf">_get_inverse_affine_matrix</span><span class="p">(</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">inverted</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]:</span>
+    <span class="c1"># Helper method to compute inverse matrix for affine transformation</span>
+
+    <span class="c1"># Pillow requires inverse affine transformation matrix:</span>
+    <span class="c1"># Affine matrix is : M = T * C * RotateScaleShear * C^-1</span>
+    <span class="c1">#</span>
+    <span class="c1"># where T is translation matrix: [1, 0, tx | 0, 1, ty | 0, 0, 1]</span>
+    <span class="c1">#       C is translation matrix to keep center: [1, 0, cx | 0, 1, cy | 0, 0, 1]</span>
+    <span class="c1">#       RotateScaleShear is rotation with scale and shear matrix</span>
+    <span class="c1">#</span>
+    <span class="c1">#       RotateScaleShear(a, s, (sx, sy)) =</span>
+    <span class="c1">#       = R(a) * S(s) * SHy(sy) * SHx(sx)</span>
+    <span class="c1">#       = [ s*cos(a - sy)/cos(sy), s*(-cos(a - sy)*tan(sx)/cos(sy) - sin(a)), 0 ]</span>
+    <span class="c1">#         [ s*sin(a - sy)/cos(sy), s*(-sin(a - sy)*tan(sx)/cos(sy) + cos(a)), 0 ]</span>
+    <span class="c1">#         [ 0                    , 0                                      , 1 ]</span>
+    <span class="c1"># where R is a rotation matrix, S is a scaling matrix, and SHx and SHy are the shears:</span>
+    <span class="c1"># SHx(s) = [1, -tan(s)] and SHy(s) = [1      , 0]</span>
+    <span class="c1">#          [0, 1      ]              [-tan(s), 1]</span>
+    <span class="c1">#</span>
+    <span class="c1"># Thus, the inverse is M^-1 = C * RotateScaleShear^-1 * C^-1 * T^-1</span>
+
+    <span class="n">rot</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">radians</span><span class="p">(</span><span class="n">angle</span><span class="p">)</span>
+    <span class="n">sx</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">radians</span><span class="p">(</span><span class="n">shear</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+    <span class="n">sy</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">radians</span><span class="p">(</span><span class="n">shear</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+
+    <span class="n">cx</span><span class="p">,</span> <span class="n">cy</span> <span class="o">=</span> <span class="n">center</span>
+    <span class="n">tx</span><span class="p">,</span> <span class="n">ty</span> <span class="o">=</span> <span class="n">translate</span>
+
+    <span class="c1"># Cached results</span>
+    <span class="n">cos_sy</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">sy</span><span class="p">)</span>
+    <span class="n">tan_sx</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">tan</span><span class="p">(</span><span class="n">sx</span><span class="p">)</span>
+    <span class="n">rot_minus_sy</span> <span class="o">=</span> <span class="n">rot</span> <span class="o">-</span> <span class="n">sy</span>
+    <span class="n">cx_plus_tx</span> <span class="o">=</span> <span class="n">cx</span> <span class="o">+</span> <span class="n">tx</span>
+    <span class="n">cy_plus_ty</span> <span class="o">=</span> <span class="n">cy</span> <span class="o">+</span> <span class="n">ty</span>
+
+    <span class="c1"># Rotate Scale Shear (RSS) without scaling</span>
+    <span class="n">a</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">rot_minus_sy</span><span class="p">)</span> <span class="o">/</span> <span class="n">cos_sy</span>
+    <span class="n">b</span> <span class="o">=</span> <span class="o">-</span><span class="p">(</span><span class="n">a</span> <span class="o">*</span> <span class="n">tan_sx</span> <span class="o">+</span> <span class="n">math</span><span class="o">.</span><span class="n">sin</span><span class="p">(</span><span class="n">rot</span><span class="p">))</span>
+    <span class="n">c</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">sin</span><span class="p">(</span><span class="n">rot_minus_sy</span><span class="p">)</span> <span class="o">/</span> <span class="n">cos_sy</span>
+    <span class="n">d</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">rot</span><span class="p">)</span> <span class="o">-</span> <span class="n">c</span> <span class="o">*</span> <span class="n">tan_sx</span>
+
+    <span class="k">if</span> <span class="n">inverted</span><span class="p">:</span>
+        <span class="c1"># Inverted rotation matrix with scale and shear</span>
+        <span class="c1"># det([[a, b], [c, d]]) == 1, since det(rotation) = 1 and det(shear) = 1</span>
+        <span class="n">matrix</span> <span class="o">=</span> <span class="p">[</span><span class="n">d</span> <span class="o">/</span> <span class="n">scale</span><span class="p">,</span> <span class="o">-</span><span class="n">b</span> <span class="o">/</span> <span class="n">scale</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="o">-</span><span class="n">c</span> <span class="o">/</span> <span class="n">scale</span><span class="p">,</span> <span class="n">a</span> <span class="o">/</span> <span class="n">scale</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+        <span class="c1"># Apply inverse of translation and of center translation: RSS^-1 * C^-1 * T^-1</span>
+        <span class="c1"># and then apply center translation: C * RSS^-1 * C^-1 * T^-1</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cx</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">cx_plus_tx</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">cy_plus_ty</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cy</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="n">cx_plus_tx</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span> <span class="o">*</span> <span class="n">cy_plus_ty</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">matrix</span> <span class="o">=</span> <span class="p">[</span><span class="n">a</span> <span class="o">*</span> <span class="n">scale</span><span class="p">,</span> <span class="n">b</span> <span class="o">*</span> <span class="n">scale</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">c</span> <span class="o">*</span> <span class="n">scale</span><span class="p">,</span> <span class="n">d</span> <span class="o">*</span> <span class="n">scale</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+        <span class="c1"># Apply inverse of center translation: RSS * C^-1</span>
+        <span class="c1"># and then apply translation and center : T * C * RSS * C^-1</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cx_plus_tx</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">cx</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">cy</span>
+        <span class="n">matrix</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">+=</span> <span class="n">cy_plus_ty</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="n">cx</span> <span class="o">-</span> <span class="n">matrix</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span> <span class="o">*</span> <span class="n">cy</span>
+
+    <span class="k">return</span> <span class="n">matrix</span>
+
+
+<span class="k">def</span> <span class="nf">_compute_affine_output_size</span><span class="p">(</span><span class="n">matrix</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">w</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">h</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+    <span class="c1"># Inspired of PIL implementation:</span>
+    <span class="c1"># https://github.com/python-pillow/Pillow/blob/11de3318867e4398057373ee9f12dcb33db7335c/src/PIL/Image.py#L2054</span>
+
+    <span class="c1"># pts are Top-Left, Top-Right, Bottom-Left, Bottom-Right points.</span>
+    <span class="c1"># Points are shifted due to affine matrix torch convention about</span>
+    <span class="c1"># the center point. Center is (0, 0) for image center pivot point (w * 0.5, h * 0.5)</span>
+    <span class="n">half_w</span> <span class="o">=</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="n">w</span>
+    <span class="n">half_h</span> <span class="o">=</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="n">h</span>
+    <span class="n">pts</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+        <span class="p">[</span>
+            <span class="p">[</span><span class="o">-</span><span class="n">half_w</span><span class="p">,</span> <span class="o">-</span><span class="n">half_h</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">],</span>
+            <span class="p">[</span><span class="o">-</span><span class="n">half_w</span><span class="p">,</span> <span class="n">half_h</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">],</span>
+            <span class="p">[</span><span class="n">half_w</span><span class="p">,</span> <span class="n">half_h</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">],</span>
+            <span class="p">[</span><span class="n">half_w</span><span class="p">,</span> <span class="o">-</span><span class="n">half_h</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">],</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+    <span class="n">theta</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">matrix</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">new_pts</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">pts</span><span class="p">,</span> <span class="n">theta</span><span class="o">.</span><span class="n">T</span><span class="p">)</span>
+    <span class="n">min_vals</span><span class="p">,</span> <span class="n">max_vals</span> <span class="o">=</span> <span class="n">new_pts</span><span class="o">.</span><span class="n">aminmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="c1"># shift points to [0, w] and [0, h] interval to match PIL results</span>
+    <span class="n">halfs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">((</span><span class="n">half_w</span><span class="p">,</span> <span class="n">half_h</span><span class="p">))</span>
+    <span class="n">min_vals</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">halfs</span><span class="p">)</span>
+    <span class="n">max_vals</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">halfs</span><span class="p">)</span>
+
+    <span class="c1"># Truncate precision to 1e-4 to avoid ceil of Xe-15 to 1.0</span>
+    <span class="n">tol</span> <span class="o">=</span> <span class="mf">1e-4</span>
+    <span class="n">inv_tol</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">tol</span>
+    <span class="n">cmax</span> <span class="o">=</span> <span class="n">max_vals</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">inv_tol</span><span class="p">)</span><span class="o">.</span><span class="n">trunc_</span><span class="p">()</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">tol</span><span class="p">)</span><span class="o">.</span><span class="n">ceil_</span><span class="p">()</span>
+    <span class="n">cmin</span> <span class="o">=</span> <span class="n">min_vals</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">inv_tol</span><span class="p">)</span><span class="o">.</span><span class="n">trunc_</span><span class="p">()</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">tol</span><span class="p">)</span><span class="o">.</span><span class="n">floor_</span><span class="p">()</span>
+    <span class="n">size</span> <span class="o">=</span> <span class="n">cmax</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">cmin</span><span class="p">)</span>
+    <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>  <span class="c1"># w, h</span>
+
+
+<span class="k">def</span> <span class="nf">_apply_grid_transform</span><span class="p">(</span><span class="n">img</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">grid</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mode</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+    <span class="c1"># We are using context knowledge that grid should have float dtype</span>
+    <span class="n">fp</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">grid</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">float_img</span> <span class="o">=</span> <span class="n">img</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">img</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">grid</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">float_img</span><span class="o">.</span><span class="n">shape</span>
+    <span class="k">if</span> <span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="c1"># Apply same grid to a batch of images</span>
+        <span class="n">grid</span> <span class="o">=</span> <span class="n">grid</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="c1"># Append a dummy mask for customized fill colors, should be faster than grid_sample() twice</span>
+    <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="n">shape</span><span class="p">[</span><span class="mi">3</span><span class="p">]),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">float_img</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">float_img</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="n">float_img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">float_img</span><span class="p">,</span> <span class="n">mask</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="n">float_img</span> <span class="o">=</span> <span class="n">grid_sample</span><span class="p">(</span><span class="n">float_img</span><span class="p">,</span> <span class="n">grid</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">mode</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="s2">&quot;zeros&quot;</span><span class="p">,</span> <span class="n">align_corners</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+    <span class="c1"># Fill with required color</span>
+    <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">float_img</span><span class="p">,</span> <span class="n">mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor_split</span><span class="p">(</span><span class="n">float_img</span><span class="p">,</span> <span class="n">indices</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">3</span><span class="p">)</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">expand_as</span><span class="p">(</span><span class="n">float_img</span><span class="p">)</span>
+        <span class="n">fill_list</span> <span class="o">=</span> <span class="n">fill</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="k">else</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">)]</span>  <span class="c1"># type: ignore[arg-type]</span>
+        <span class="n">fill_img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">fill_list</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">float_img</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">float_img</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;nearest&quot;</span><span class="p">:</span>
+            <span class="n">bool_mask</span> <span class="o">=</span> <span class="n">mask</span> <span class="o">&lt;</span> <span class="mf">0.5</span>
+            <span class="n">float_img</span><span class="p">[</span><span class="n">bool_mask</span><span class="p">]</span> <span class="o">=</span> <span class="n">fill_img</span><span class="o">.</span><span class="n">expand_as</span><span class="p">(</span><span class="n">float_img</span><span class="p">)[</span><span class="n">bool_mask</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>  <span class="c1"># &#39;bilinear&#39;</span>
+            <span class="c1"># The following is mathematically equivalent to:</span>
+            <span class="c1"># img * mask + (1.0 - mask) * fill = img * mask - fill * mask + fill = mask * (img - fill) + fill</span>
+            <span class="n">float_img</span> <span class="o">=</span> <span class="n">float_img</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">fill_img</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">mask</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">fill_img</span><span class="p">)</span>
+
+    <span class="n">img</span> <span class="o">=</span> <span class="n">float_img</span><span class="o">.</span><span class="n">round_</span><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">float_img</span>
+
+    <span class="k">return</span> <span class="n">img</span>
+
+
+<span class="k">def</span> <span class="nf">_assert_grid_transform_inputs</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">matrix</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span><span class="p">,</span>
+    <span class="n">supported_interpolation_modes</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+    <span class="n">coeffs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">matrix</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">matrix</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument matrix should be a list&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">matrix</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">6</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument matrix should have 6 float values&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">coeffs</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">coeffs</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">8</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument coeffs should have 8 float values&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="n">length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span>
+            <span class="n">num_channels</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span>
+            <span class="k">if</span> <span class="n">length</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">length</span> <span class="o">!=</span> <span class="n">num_channels</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;The number of elements in &#39;fill&#39; cannot broadcast to match the number of &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;channels of the image (</span><span class="si">{</span><span class="n">length</span><span class="si">}</span><span class="s2"> != </span><span class="si">{</span><span class="n">num_channels</span><span class="si">}</span><span class="s2">)&quot;</span>
+                <span class="p">)</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument fill should be either int, float, tuple or list&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">interpolation</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">supported_interpolation_modes</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Interpolation mode &#39;</span><span class="si">{</span><span class="n">interpolation</span><span class="si">}</span><span class="s2">&#39; is unsupported with Tensor input&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_affine_grid</span><span class="p">(</span>
+    <span class="n">theta</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">w</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">h</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">ow</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">oh</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># https://github.com/pytorch/pytorch/blob/74b65c32be68b15dc7c9e8bb62459efbfbde33d8/aten/src/ATen/native/</span>
+    <span class="c1"># AffineGridGenerator.cpp#L18</span>
+    <span class="c1"># Difference with AffineGridGenerator is that:</span>
+    <span class="c1"># 1) we normalize grid values after applying theta</span>
+    <span class="c1"># 2) we can normalize by other image size, such that it covers &quot;extend&quot; option like in PIL.Image.rotate</span>
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">theta</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">device</span> <span class="o">=</span> <span class="n">theta</span><span class="o">.</span><span class="n">device</span>
+
+    <span class="n">base_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">oh</span><span class="p">,</span> <span class="n">ow</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">x_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">((</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">ow</span><span class="p">)</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">,</span> <span class="p">(</span><span class="n">ow</span> <span class="o">-</span> <span class="mf">1.0</span><span class="p">)</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="n">ow</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">x_grid</span><span class="p">)</span>
+    <span class="n">y_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">((</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">oh</span><span class="p">)</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">,</span> <span class="p">(</span><span class="n">oh</span> <span class="o">-</span> <span class="mf">1.0</span><span class="p">)</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="n">oh</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze_</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">y_grid</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">fill_</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="n">rescaled_theta</span> <span class="o">=</span> <span class="n">theta</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mf">0.5</span> <span class="o">*</span> <span class="n">w</span><span class="p">,</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="n">h</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">))</span>
+    <span class="n">output_grid</span> <span class="o">=</span> <span class="n">base_grid</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">oh</span> <span class="o">*</span> <span class="n">ow</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span><span class="o">.</span><span class="n">bmm</span><span class="p">(</span><span class="n">rescaled_theta</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output_grid</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">oh</span><span class="p">,</span> <span class="n">ow</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">affine</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">affine</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">affine_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">ndim</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">ndim</span>
+
+    <span class="k">if</span> <span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="o">-</span><span class="mi">1</span><span class="p">,)</span> <span class="o">+</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:])</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">elif</span> <span class="n">ndim</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+    <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">center</span> <span class="o">=</span> <span class="n">_affine_parse_args</span><span class="p">(</span><span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">interpolation</span><span class="p">,</span> <span class="n">center</span><span class="p">)</span>
+
+    <span class="n">center_f</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># Center values should be in pixel coordinates but translated such that (0, 0) corresponds to image center.</span>
+        <span class="n">center_f</span> <span class="o">=</span> <span class="p">[(</span><span class="n">c</span> <span class="o">-</span> <span class="n">s</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">)</span> <span class="k">for</span> <span class="n">c</span><span class="p">,</span> <span class="n">s</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="p">[</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">])]</span>
+
+    <span class="n">translate_f</span> <span class="o">=</span> <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">t</span><span class="p">)</span> <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">translate</span><span class="p">]</span>
+    <span class="n">matrix</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center_f</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translate_f</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">)</span>
+
+    <span class="n">_assert_grid_transform_inputs</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">matrix</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;nearest&quot;</span><span class="p">,</span> <span class="s2">&quot;bilinear&quot;</span><span class="p">])</span>
+
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">image</span><span class="p">)</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+    <span class="n">theta</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">matrix</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="n">_affine_grid</span><span class="p">(</span><span class="n">theta</span><span class="p">,</span> <span class="n">w</span><span class="o">=</span><span class="n">width</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">height</span><span class="p">,</span> <span class="n">ow</span><span class="o">=</span><span class="n">width</span><span class="p">,</span> <span class="n">oh</span><span class="o">=</span><span class="n">height</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">_apply_grid_transform</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">grid</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_unsquash</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">affine</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_affine_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">center</span> <span class="o">=</span> <span class="n">_affine_parse_args</span><span class="p">(</span><span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">interpolation</span><span class="p">,</span> <span class="n">center</span><span class="p">)</span>
+
+    <span class="c1"># center = (img_size[0] * 0.5 + 0.5, img_size[1] * 0.5 + 0.5)</span>
+    <span class="c1"># it is visually better to estimate the center without 0.5 offset</span>
+    <span class="c1"># otherwise image rotated by 90 degrees is shifted vs output image of torch.rot90 or F_t.affine</span>
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">_get_size_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+        <span class="n">center</span> <span class="o">=</span> <span class="p">[</span><span class="n">width</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">,</span> <span class="n">height</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">]</span>
+    <span class="n">matrix</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_FP</span><span class="o">.</span><span class="n">affine</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">matrix</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">],</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_affine_bounding_boxes_with_expand</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+    <span class="k">if</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">canvas_size</span>
+
+    <span class="n">original_shape</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">original_dtype</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span> <span class="k">if</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span> <span class="k">else</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">device</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">device</span>
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">convert_bounding_box_format</span><span class="p">(</span>
+            <span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+    <span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+    <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">center</span> <span class="o">=</span> <span class="n">_affine_parse_args</span><span class="p">(</span>
+        <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span> <span class="n">center</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">canvas_size</span>
+        <span class="n">center</span> <span class="o">=</span> <span class="p">[</span><span class="n">width</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">,</span> <span class="n">height</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">]</span>
+
+    <span class="n">affine_vector</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">,</span> <span class="n">inverted</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="n">transposed_affine_matrix</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+            <span class="n">affine_vector</span><span class="p">,</span>
+            <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span>
+            <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+        <span class="o">.</span><span class="n">T</span>
+    <span class="p">)</span>
+    <span class="c1"># 1) Let&#39;s transform bboxes into a tensor of 4 points (top-left, top-right, bottom-left, bottom-right corners).</span>
+    <span class="c1"># Tensor of points has shape (N * 4, 3), where N is the number of bboxes</span>
+    <span class="c1"># Single point structure is similar to</span>
+    <span class="c1"># [(xmin, ymin, 1), (xmax, ymin, 1), (xmax, ymax, 1), (xmin, ymax, 1)]</span>
+    <span class="n">points</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="p">[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">]]]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">points</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">points</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="n">points</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)],</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="c1"># 2) Now let&#39;s transform the points using affine matrix</span>
+    <span class="n">transformed_points</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">points</span><span class="p">,</span> <span class="n">transposed_affine_matrix</span><span class="p">)</span>
+    <span class="c1"># 3) Reshape transformed points to [N boxes, 4 points, x/y coords]</span>
+    <span class="c1"># and compute bounding box from 4 transformed points:</span>
+    <span class="n">transformed_points</span> <span class="o">=</span> <span class="n">transformed_points</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">out_bbox_mins</span><span class="p">,</span> <span class="n">out_bbox_maxs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">aminmax</span><span class="p">(</span><span class="n">transformed_points</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">out_bboxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">out_bbox_mins</span><span class="p">,</span> <span class="n">out_bbox_maxs</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">expand</span><span class="p">:</span>
+        <span class="c1"># Compute minimum point for transformed image frame:</span>
+        <span class="c1"># Points are Top-Left, Top-Right, Bottom-Left, Bottom-Right points.</span>
+        <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">canvas_size</span>
+        <span class="n">points</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+            <span class="p">[</span>
+                <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">],</span>
+                <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="n">height</span><span class="p">),</span> <span class="mf">1.0</span><span class="p">],</span>
+                <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">width</span><span class="p">),</span> <span class="nb">float</span><span class="p">(</span><span class="n">height</span><span class="p">),</span> <span class="mf">1.0</span><span class="p">],</span>
+                <span class="p">[</span><span class="nb">float</span><span class="p">(</span><span class="n">width</span><span class="p">),</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">],</span>
+            <span class="p">],</span>
+            <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span>
+            <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">new_points</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">points</span><span class="p">,</span> <span class="n">transposed_affine_matrix</span><span class="p">)</span>
+        <span class="n">tr</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">amin</span><span class="p">(</span><span class="n">new_points</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># Translate bounding boxes</span>
+        <span class="n">out_bboxes</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">tr</span><span class="o">.</span><span class="n">repeat</span><span class="p">((</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)))</span>
+        <span class="c1"># Estimate meta-data for image with inverted=True</span>
+        <span class="n">affine_vector</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">translate</span><span class="p">,</span> <span class="n">scale</span><span class="p">,</span> <span class="n">shear</span><span class="p">)</span>
+        <span class="n">new_width</span><span class="p">,</span> <span class="n">new_height</span> <span class="o">=</span> <span class="n">_compute_affine_output_size</span><span class="p">(</span><span class="n">affine_vector</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">)</span>
+        <span class="n">canvas_size</span> <span class="o">=</span> <span class="p">(</span><span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">)</span>
+
+    <span class="n">out_bboxes</span> <span class="o">=</span> <span class="n">clamp_bounding_boxes</span><span class="p">(</span><span class="n">out_bboxes</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+    <span class="n">out_bboxes</span> <span class="o">=</span> <span class="n">convert_bounding_box_format</span><span class="p">(</span>
+        <span class="n">out_bboxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span>
+    <span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">original_shape</span><span class="p">)</span>
+
+    <span class="n">out_bboxes</span> <span class="o">=</span> <span class="n">out_bboxes</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">original_dtype</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">out_bboxes</span><span class="p">,</span> <span class="n">canvas_size</span>
+
+
+<span class="k">def</span> <span class="nf">affine_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">out_box</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">_affine_bounding_boxes_with_expand</span><span class="p">(</span>
+        <span class="n">bounding_boxes</span><span class="p">,</span>
+        <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+        <span class="n">expand</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">out_box</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">affine</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_affine_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">affine_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span>
+        <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">affine_mask</span><span class="p">(</span>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">affine_image</span><span class="p">(</span>
+        <span class="n">mask</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">affine</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_affine_mask_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">affine_mask</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">affine</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">affine_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">translate</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">scale</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">shear</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">affine_image</span><span class="p">(</span>
+        <span class="n">video</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="n">translate</span><span class="p">,</span>
+        <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="n">shear</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+    <span class="p">)</span>
+
+
+<div class="viewcode-block" id="rotate"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.rotate.html#torchvision.transforms.v2.functional.rotate">[docs]</a><span class="k">def</span> <span class="nf">rotate</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomRotation` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">rotate_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">rotate</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">rotate</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rotate</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rotate</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">rotate_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:]</span>
+
+    <span class="n">center_f</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">center</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># Center values should be in pixel coordinates but translated such that (0, 0) corresponds to image center.</span>
+        <span class="n">center_f</span> <span class="o">=</span> <span class="p">[(</span><span class="n">c</span> <span class="o">-</span> <span class="n">s</span> <span class="o">*</span> <span class="mf">0.5</span><span class="p">)</span> <span class="k">for</span> <span class="n">c</span><span class="p">,</span> <span class="n">s</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">center</span><span class="p">,</span> <span class="p">[</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">])]</span>
+
+    <span class="c1"># due to current incoherence of rotation angle direction between affine and rotate implementations</span>
+    <span class="c1"># we need to set -angle.</span>
+    <span class="n">matrix</span> <span class="o">=</span> <span class="n">_get_inverse_affine_matrix</span><span class="p">(</span><span class="n">center_f</span><span class="p">,</span> <span class="o">-</span><span class="n">angle</span><span class="p">,</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span> <span class="mf">1.0</span><span class="p">,</span> <span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">])</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+        <span class="n">_assert_grid_transform_inputs</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">matrix</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="p">,</span> <span class="p">[</span><span class="s2">&quot;nearest&quot;</span><span class="p">,</span> <span class="s2">&quot;bilinear&quot;</span><span class="p">])</span>
+
+        <span class="n">ow</span><span class="p">,</span> <span class="n">oh</span> <span class="o">=</span> <span class="n">_compute_affine_output_size</span><span class="p">(</span><span class="n">matrix</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">)</span> <span class="k">if</span> <span class="n">expand</span> <span class="k">else</span> <span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">)</span>
+        <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">image</span><span class="p">)</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+        <span class="n">theta</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">matrix</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+        <span class="n">grid</span> <span class="o">=</span> <span class="n">_affine_grid</span><span class="p">(</span><span class="n">theta</span><span class="p">,</span> <span class="n">w</span><span class="o">=</span><span class="n">width</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">height</span><span class="p">,</span> <span class="n">ow</span><span class="o">=</span><span class="n">ow</span><span class="p">,</span> <span class="n">oh</span><span class="o">=</span><span class="n">oh</span><span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">_apply_grid_transform</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">grid</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+        <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">image</span>
+        <span class="n">new_width</span><span class="p">,</span> <span class="n">new_height</span> <span class="o">=</span> <span class="n">_compute_affine_output_size</span><span class="p">(</span><span class="n">matrix</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">)</span> <span class="k">if</span> <span class="n">expand</span> <span class="k">else</span> <span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">[:</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span> <span class="o">+</span> <span class="p">(</span><span class="n">num_channels</span><span class="p">,</span> <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">))</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rotate</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_rotate_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_FP</span><span class="o">.</span><span class="n">rotate</span><span class="p">(</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">],</span> <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="n">center</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">rotate_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+    <span class="k">return</span> <span class="n">_affine_bounding_boxes_with_expand</span><span class="p">(</span>
+        <span class="n">bounding_boxes</span><span class="p">,</span>
+        <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=-</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">translate</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span>
+        <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">shear</span><span class="o">=</span><span class="p">[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+        <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span>
+    <span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rotate</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_rotate_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">rotate_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span>
+        <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">rotate_mask</span><span class="p">(</span>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">rotate_image</span><span class="p">(</span>
+        <span class="n">mask</span><span class="p">,</span>
+        <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span>
+        <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rotate</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_rotate_mask_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">rotate_mask</span><span class="p">(</span><span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="n">angle</span><span class="o">=</span><span class="n">angle</span><span class="p">,</span> <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">rotate</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">rotate_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">angle</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span>
+    <span class="n">expand</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">center</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">rotate_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">angle</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">expand</span><span class="o">=</span><span class="n">expand</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">center</span><span class="o">=</span><span class="n">center</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="pad"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.pad.html#torchvision.transforms.v2.functional.pad">[docs]</a><span class="k">def</span> <span class="nf">pad</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.Pad` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">pad_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">pad</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_parse_pad_padding</span><span class="p">(</span><span class="n">padding</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">padding</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">pad_left</span> <span class="o">=</span> <span class="n">pad_right</span> <span class="o">=</span> <span class="n">pad_top</span> <span class="o">=</span> <span class="n">pad_bottom</span> <span class="o">=</span> <span class="n">padding</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">padding</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">pad_left</span> <span class="o">=</span> <span class="n">pad_right</span> <span class="o">=</span> <span class="n">pad_top</span> <span class="o">=</span> <span class="n">pad_bottom</span> <span class="o">=</span> <span class="n">padding</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">pad_left</span> <span class="o">=</span> <span class="n">pad_right</span> <span class="o">=</span> <span class="n">padding</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">pad_top</span> <span class="o">=</span> <span class="n">pad_bottom</span> <span class="o">=</span> <span class="n">padding</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span> <span class="o">==</span> <span class="mi">4</span><span class="p">:</span>
+            <span class="n">pad_left</span> <span class="o">=</span> <span class="n">padding</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">pad_top</span> <span class="o">=</span> <span class="n">padding</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+            <span class="n">pad_right</span> <span class="o">=</span> <span class="n">padding</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+            <span class="n">pad_bottom</span> <span class="o">=</span> <span class="n">padding</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Padding must be an int or a 1, 2, or 4 element tuple, not a </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span><span class="si">}</span><span class="s2"> element tuple&quot;</span>
+            <span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;`padding` should be an integer or tuple or list of integers, but got </span><span class="si">{</span><span class="n">padding</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="p">[</span><span class="n">pad_left</span><span class="p">,</span> <span class="n">pad_right</span><span class="p">,</span> <span class="n">pad_top</span><span class="p">,</span> <span class="n">pad_bottom</span><span class="p">]</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">pad_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># Be aware that while `padding` has order `[left, top, right, bottom]`, `torch_padding` uses</span>
+    <span class="c1"># `[left, right, top, bottom]`. This stems from the fact that we align our API with PIL, but need to use `torch_pad`</span>
+    <span class="c1"># internally.</span>
+    <span class="n">torch_padding</span> <span class="o">=</span> <span class="n">_parse_pad_padding</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">padding_mode</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;constant&quot;</span><span class="p">,</span> <span class="s2">&quot;edge&quot;</span><span class="p">,</span> <span class="s2">&quot;reflect&quot;</span><span class="p">,</span> <span class="s2">&quot;symmetric&quot;</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;`padding_mode` should be either `&#39;constant&#39;`, `&#39;edge&#39;`, `&#39;reflect&#39;` or `&#39;symmetric&#39;`, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;but got `&#39;</span><span class="si">{</span><span class="n">padding_mode</span><span class="si">}</span><span class="s2">&#39;`.&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="mi">0</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+        <span class="k">return</span> <span class="n">_pad_with_scalar_fill</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">fill</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_pad_with_scalar_fill</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_pad_with_vector_fill</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_pad_with_scalar_fill</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch_padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">],</span>
+    <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:]</span>
+
+    <span class="n">batch_size</span> <span class="o">=</span> <span class="mi">1</span>
+    <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">shape</span><span class="p">[:</span><span class="o">-</span><span class="mi">3</span><span class="p">]:</span>
+        <span class="n">batch_size</span> <span class="o">*=</span> <span class="n">s</span>
+
+    <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">padding_mode</span> <span class="o">==</span> <span class="s2">&quot;edge&quot;</span><span class="p">:</span>
+        <span class="c1"># Similar to the padding order, `torch_pad`&#39;s PIL&#39;s padding modes don&#39;t have the same names. Thus, we map</span>
+        <span class="c1"># the PIL name for the padding mode, which we are also using for our API, to the corresponding `torch_pad`</span>
+        <span class="c1"># name.</span>
+        <span class="n">padding_mode</span> <span class="o">=</span> <span class="s2">&quot;replicate&quot;</span>
+
+    <span class="k">if</span> <span class="n">padding_mode</span> <span class="o">==</span> <span class="s2">&quot;constant&quot;</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">torch_pad</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="nb">float</span><span class="p">(</span><span class="n">fill</span><span class="p">))</span>
+    <span class="k">elif</span> <span class="n">padding_mode</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;reflect&quot;</span><span class="p">,</span> <span class="s2">&quot;replicate&quot;</span><span class="p">):</span>
+        <span class="c1"># `torch_pad` only supports `&quot;reflect&quot;` or `&quot;replicate&quot;` padding for floating point inputs.</span>
+        <span class="c1"># TODO: See https://github.com/pytorch/pytorch/issues/40763</span>
+        <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">():</span>
+            <span class="n">needs_cast</span> <span class="o">=</span> <span class="kc">True</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">needs_cast</span> <span class="o">=</span> <span class="kc">False</span>
+
+        <span class="n">image</span> <span class="o">=</span> <span class="n">torch_pad</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">needs_cast</span><span class="p">:</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>  <span class="c1"># padding_mode == &quot;symmetric&quot;</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">_pad_symmetric</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">)</span>
+
+    <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+
+    <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">[:</span><span class="o">-</span><span class="mi">3</span><span class="p">]</span> <span class="o">+</span> <span class="p">(</span><span class="n">num_channels</span><span class="p">,</span> <span class="n">new_height</span><span class="p">,</span> <span class="n">new_width</span><span class="p">))</span>
+
+
+<span class="c1"># TODO: This should be removed once torch_pad supports non-scalar padding values</span>
+<span class="k">def</span> <span class="nf">_pad_with_vector_fill</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch_padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">padding_mode</span> <span class="o">!=</span> <span class="s2">&quot;constant&quot;</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Padding mode &#39;</span><span class="si">{</span><span class="n">padding_mode</span><span class="si">}</span><span class="s2">&#39; is not supported if fill is not scalar&quot;</span><span class="p">)</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">_pad_with_scalar_fill</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="s2">&quot;constant&quot;</span><span class="p">)</span>
+    <span class="n">left</span><span class="p">,</span> <span class="n">right</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">bottom</span> <span class="o">=</span> <span class="n">torch_padding</span>
+
+    <span class="c1"># We are creating the tensor in the autodetected dtype first and convert to the right one after to avoid an implicit</span>
+    <span class="c1"># float -&gt; int conversion. That happens for example for the valid input of a uint8 image with floating point fill</span>
+    <span class="c1"># value.</span>
+    <span class="n">fill</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">top</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">output</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="n">top</span><span class="p">,</span> <span class="p">:]</span> <span class="o">=</span> <span class="n">fill</span>
+    <span class="k">if</span> <span class="n">left</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">output</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:</span><span class="n">left</span><span class="p">]</span> <span class="o">=</span> <span class="n">fill</span>
+    <span class="k">if</span> <span class="n">bottom</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">output</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="o">-</span><span class="n">bottom</span><span class="p">:,</span> <span class="p">:]</span> <span class="o">=</span> <span class="n">fill</span>
+    <span class="k">if</span> <span class="n">right</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">output</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:,</span> <span class="o">-</span><span class="n">right</span><span class="p">:]</span> <span class="o">=</span> <span class="n">fill</span>
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="n">_pad_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">pad</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">pad_mask</span><span class="p">(</span>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">fill</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">fill</span> <span class="o">=</span> <span class="mi">0</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">fill</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Non-scalar fill value is not supported&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">pad_image</span><span class="p">(</span><span class="n">mask</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="k">def</span> <span class="nf">pad_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+    <span class="k">if</span> <span class="n">padding_mode</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;constant&quot;</span><span class="p">]:</span>
+        <span class="c1"># TODO: add support of other padding modes</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Padding mode &#39;</span><span class="si">{</span><span class="n">padding_mode</span><span class="si">}</span><span class="s2">&#39; is not supported with bounding boxes&quot;</span><span class="p">)</span>
+
+    <span class="n">left</span><span class="p">,</span> <span class="n">right</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">bottom</span> <span class="o">=</span> <span class="n">_parse_pad_padding</span><span class="p">(</span><span class="n">padding</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">format</span> <span class="o">==</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">:</span>
+        <span class="n">pad</span> <span class="o">=</span> <span class="p">[</span><span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">pad</span> <span class="o">=</span> <span class="p">[</span><span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span>
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">bounding_boxes</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+    <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">canvas_size</span>
+    <span class="n">height</span> <span class="o">+=</span> <span class="n">top</span> <span class="o">+</span> <span class="n">bottom</span>
+    <span class="n">width</span> <span class="o">+=</span> <span class="n">left</span> <span class="o">+</span> <span class="n">right</span>
+    <span class="n">canvas_size</span> <span class="o">=</span> <span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">clamp_bounding_boxes</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">),</span> <span class="n">canvas_size</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_pad_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">pad_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span>
+        <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">,</span>
+        <span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">,</span>
+        <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">pad</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">pad_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">padding_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;constant&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">pad_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="n">padding_mode</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="crop"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.crop.html#torchvision.transforms.v2.functional.crop">[docs]</a><span class="k">def</span> <span class="nf">crop</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomCrop` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">crop</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">crop</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">crop</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+
+    <span class="n">right</span> <span class="o">=</span> <span class="n">left</span> <span class="o">+</span> <span class="n">width</span>
+    <span class="n">bottom</span> <span class="o">=</span> <span class="n">top</span> <span class="o">+</span> <span class="n">height</span>
+
+    <span class="k">if</span> <span class="n">left</span> <span class="o">&lt;</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">top</span> <span class="o">&lt;</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">right</span> <span class="o">&gt;</span> <span class="n">w</span> <span class="ow">or</span> <span class="n">bottom</span> <span class="o">&gt;</span> <span class="n">h</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="nb">max</span><span class="p">(</span><span class="n">top</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="p">:</span> <span class="n">bottom</span><span class="p">,</span> <span class="nb">max</span><span class="p">(</span><span class="n">left</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="p">:</span> <span class="n">right</span><span class="p">]</span>
+        <span class="n">torch_padding</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="nb">max</span><span class="p">(</span><span class="nb">min</span><span class="p">(</span><span class="n">right</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="o">-</span> <span class="n">left</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span>
+            <span class="nb">max</span><span class="p">(</span><span class="n">right</span> <span class="o">-</span> <span class="nb">max</span><span class="p">(</span><span class="n">w</span><span class="p">,</span> <span class="n">left</span><span class="p">),</span> <span class="mi">0</span><span class="p">),</span>
+            <span class="nb">max</span><span class="p">(</span><span class="nb">min</span><span class="p">(</span><span class="n">bottom</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="o">-</span> <span class="n">top</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span>
+            <span class="nb">max</span><span class="p">(</span><span class="n">bottom</span> <span class="o">-</span> <span class="nb">max</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">top</span><span class="p">),</span> <span class="mi">0</span><span class="p">),</span>
+        <span class="p">]</span>
+        <span class="k">return</span> <span class="n">_pad_with_scalar_fill</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch_padding</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">padding_mode</span><span class="o">=</span><span class="s2">&quot;constant&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">image</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span><span class="n">bottom</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span><span class="n">right</span><span class="p">]</span>
+
+
+<span class="n">_crop_image_pil</span> <span class="o">=</span> <span class="n">_FP</span><span class="o">.</span><span class="n">crop</span>
+<span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">crop</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_crop_image_pil</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">crop_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+
+    <span class="c1"># Crop or implicit pad if left and/or top have negative values:</span>
+    <span class="k">if</span> <span class="nb">format</span> <span class="o">==</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">:</span>
+        <span class="n">sub</span> <span class="o">=</span> <span class="p">[</span><span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">sub</span> <span class="o">=</span> <span class="p">[</span><span class="n">left</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span>
+
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">bounding_boxes</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">sub</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">canvas_size</span> <span class="o">=</span> <span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">clamp_bounding_boxes</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">),</span> <span class="n">canvas_size</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_crop_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">crop_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">width</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">crop_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">mask</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">crop_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="perspective"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.perspective.html#torchvision.transforms.v2.functional.perspective">[docs]</a><span class="k">def</span> <span class="nf">perspective</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomPerspective` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">perspective_image</span><span class="p">(</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="n">startpoints</span><span class="o">=</span><span class="n">startpoints</span><span class="p">,</span>
+            <span class="n">endpoints</span><span class="o">=</span><span class="n">endpoints</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+            <span class="n">coefficients</span><span class="o">=</span><span class="n">coefficients</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">perspective</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">perspective</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="p">,</span>
+        <span class="n">startpoints</span><span class="o">=</span><span class="n">startpoints</span><span class="p">,</span>
+        <span class="n">endpoints</span><span class="o">=</span><span class="n">endpoints</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">coefficients</span><span class="o">=</span><span class="n">coefficients</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_perspective_grid</span><span class="p">(</span><span class="n">coeffs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">ow</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">oh</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># https://github.com/python-pillow/Pillow/blob/4634eafe3c695a014267eefdce830b4a825beed7/</span>
+    <span class="c1"># src/libImaging/Geometry.c#L394</span>
+
+    <span class="c1">#</span>
+    <span class="c1"># x_out = (coeffs[0] * x + coeffs[1] * y + coeffs[2]) / (coeffs[6] * x + coeffs[7] * y + 1)</span>
+    <span class="c1"># y_out = (coeffs[3] * x + coeffs[4] * y + coeffs[5]) / (coeffs[6] * x + coeffs[7] * y + 1)</span>
+    <span class="c1">#</span>
+    <span class="n">theta1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+        <span class="p">[[[</span><span class="n">coeffs</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">coeffs</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">coeffs</span><span class="p">[</span><span class="mi">2</span><span class="p">]],</span> <span class="p">[</span><span class="n">coeffs</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">coeffs</span><span class="p">[</span><span class="mi">4</span><span class="p">],</span> <span class="n">coeffs</span><span class="p">[</span><span class="mi">5</span><span class="p">]]]],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span>
+    <span class="p">)</span>
+    <span class="n">theta2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([[[</span><span class="n">coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">],</span> <span class="n">coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">],</span> <span class="mf">1.0</span><span class="p">],</span> <span class="p">[</span><span class="n">coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">],</span> <span class="n">coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">],</span> <span class="mf">1.0</span><span class="p">]]],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+
+    <span class="n">d</span> <span class="o">=</span> <span class="mf">0.5</span>
+    <span class="n">base_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">oh</span><span class="p">,</span> <span class="n">ow</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">x_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="n">d</span><span class="p">,</span> <span class="n">ow</span> <span class="o">+</span> <span class="n">d</span> <span class="o">-</span> <span class="mf">1.0</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="n">ow</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">x_grid</span><span class="p">)</span>
+    <span class="n">y_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="n">d</span><span class="p">,</span> <span class="n">oh</span> <span class="o">+</span> <span class="n">d</span> <span class="o">-</span> <span class="mf">1.0</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="n">oh</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze_</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">y_grid</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">fill_</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="n">rescaled_theta1</span> <span class="o">=</span> <span class="n">theta1</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mf">0.5</span> <span class="o">*</span> <span class="n">ow</span><span class="p">,</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="n">oh</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">))</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">oh</span> <span class="o">*</span> <span class="n">ow</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+    <span class="n">output_grid1</span> <span class="o">=</span> <span class="n">base_grid</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span><span class="o">.</span><span class="n">bmm</span><span class="p">(</span><span class="n">rescaled_theta1</span><span class="p">)</span>
+    <span class="n">output_grid2</span> <span class="o">=</span> <span class="n">base_grid</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span><span class="o">.</span><span class="n">bmm</span><span class="p">(</span><span class="n">theta2</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span>
+
+    <span class="n">output_grid</span> <span class="o">=</span> <span class="n">output_grid1</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">output_grid2</span><span class="p">)</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="mf">1.0</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">output_grid</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">oh</span><span class="p">,</span> <span class="n">ow</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_perspective_coefficients</span><span class="p">(</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="n">coefficients</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">startpoints</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">endpoints</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The startpoints/endpoints and the coefficients shouldn&#39;t be defined concurrently.&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">coefficients</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">8</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Argument coefficients should have 8 float values&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">coefficients</span>
+    <span class="k">elif</span> <span class="n">startpoints</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">endpoints</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">_get_perspective_coeffs</span><span class="p">(</span><span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Either the startpoints/endpoints or the coefficients must have non `None` values.&quot;</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">perspective</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">perspective</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">perspective_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">perspective_coeffs</span> <span class="o">=</span> <span class="n">_perspective_coefficients</span><span class="p">(</span><span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">,</span> <span class="n">coefficients</span><span class="p">)</span>
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">ndim</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">ndim</span>
+
+    <span class="k">if</span> <span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="o">-</span><span class="mi">1</span><span class="p">,)</span> <span class="o">+</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:])</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">elif</span> <span class="n">ndim</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">_assert_grid_transform_inputs</span><span class="p">(</span>
+        <span class="n">image</span><span class="p">,</span>
+        <span class="n">matrix</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">supported_interpolation_modes</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;nearest&quot;</span><span class="p">,</span> <span class="s2">&quot;bilinear&quot;</span><span class="p">],</span>
+        <span class="n">coeffs</span><span class="o">=</span><span class="n">perspective_coeffs</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="n">oh</span><span class="p">,</span> <span class="n">ow</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">image</span><span class="p">)</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="n">_perspective_grid</span><span class="p">(</span><span class="n">perspective_coeffs</span><span class="p">,</span> <span class="n">ow</span><span class="o">=</span><span class="n">ow</span><span class="p">,</span> <span class="n">oh</span><span class="o">=</span><span class="n">oh</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">_apply_grid_transform</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">grid</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_unsquash</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">perspective</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_perspective_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BICUBIC</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">perspective_coeffs</span> <span class="o">=</span> <span class="n">_perspective_coefficients</span><span class="p">(</span><span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">,</span> <span class="n">coefficients</span><span class="p">)</span>
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_FP</span><span class="o">.</span><span class="n">perspective</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">perspective_coeffs</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">pil_modes_mapping</span><span class="p">[</span><span class="n">interpolation</span><span class="p">],</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">perspective_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">bounding_boxes</span>
+
+    <span class="n">perspective_coeffs</span> <span class="o">=</span> <span class="n">_perspective_coefficients</span><span class="p">(</span><span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">,</span> <span class="n">coefficients</span><span class="p">)</span>
+
+    <span class="n">original_shape</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">shape</span>
+    <span class="c1"># TODO: first cast to float if bbox is int64 before convert_bounding_box_format</span>
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">convert_bounding_box_format</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">)</span>
+    <span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">)</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+    <span class="n">device</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">device</span>
+
+    <span class="c1"># perspective_coeffs are computed as endpoint -&gt; start point</span>
+    <span class="c1"># We have to invert perspective_coeffs for bboxes:</span>
+    <span class="c1"># (x, y) - end point and (x_out, y_out) - start point</span>
+    <span class="c1">#   x_out = (coeffs[0] * x + coeffs[1] * y + coeffs[2]) / (coeffs[6] * x + coeffs[7] * y + 1)</span>
+    <span class="c1">#   y_out = (coeffs[3] * x + coeffs[4] * y + coeffs[5]) / (coeffs[6] * x + coeffs[7] * y + 1)</span>
+    <span class="c1"># and we would like to get:</span>
+    <span class="c1"># x = (inv_coeffs[0] * x_out + inv_coeffs[1] * y_out + inv_coeffs[2])</span>
+    <span class="c1">#       / (inv_coeffs[6] * x_out + inv_coeffs[7] * y_out + 1)</span>
+    <span class="c1"># y = (inv_coeffs[3] * x_out + inv_coeffs[4] * y_out + inv_coeffs[5])</span>
+    <span class="c1">#       / (inv_coeffs[6] * x_out + inv_coeffs[7] * y_out + 1)</span>
+    <span class="c1"># and compute inv_coeffs in terms of coeffs</span>
+
+    <span class="n">denom</span> <span class="o">=</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span> <span class="o">-</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span>
+    <span class="k">if</span> <span class="n">denom</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Provided perspective_coeffs </span><span class="si">{</span><span class="n">perspective_coeffs</span><span class="si">}</span><span class="s2"> can not be inverted to transform bounding boxes. &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;Denominator is zero, denom=</span><span class="si">{</span><span class="n">denom</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+    <span class="n">inv_coeffs</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="p">(</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span> <span class="o">-</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+        <span class="p">(</span><span class="o">-</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+        <span class="p">(</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">-</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">4</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+        <span class="p">(</span><span class="o">-</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span> <span class="o">+</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+        <span class="p">(</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+        <span class="p">(</span><span class="o">-</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">5</span><span class="p">]</span> <span class="o">+</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">3</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+        <span class="p">(</span><span class="o">-</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">4</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">]</span> <span class="o">+</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+        <span class="p">(</span><span class="o">-</span><span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">]</span> <span class="o">+</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="n">perspective_coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">])</span> <span class="o">/</span> <span class="n">denom</span><span class="p">,</span>
+    <span class="p">]</span>
+
+    <span class="n">theta1</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+        <span class="p">[[</span><span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">2</span><span class="p">]],</span> <span class="p">[</span><span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">4</span><span class="p">],</span> <span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">5</span><span class="p">]]],</span>
+        <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span>
+        <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="n">theta2</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+        <span class="p">[[</span><span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">],</span> <span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">],</span> <span class="mf">1.0</span><span class="p">],</span> <span class="p">[</span><span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">6</span><span class="p">],</span> <span class="n">inv_coeffs</span><span class="p">[</span><span class="mi">7</span><span class="p">],</span> <span class="mf">1.0</span><span class="p">]],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span>
+    <span class="p">)</span>
+
+    <span class="c1"># 1) Let&#39;s transform bboxes into a tensor of 4 points (top-left, top-right, bottom-left, bottom-right corners).</span>
+    <span class="c1"># Tensor of points has shape (N * 4, 3), where N is the number of bboxes</span>
+    <span class="c1"># Single point structure is similar to</span>
+    <span class="c1"># [(xmin, ymin, 1), (xmax, ymin, 1), (xmax, ymax, 1), (xmin, ymax, 1)]</span>
+    <span class="n">points</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="p">[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">]]]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">points</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">points</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="n">points</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">points</span><span class="o">.</span><span class="n">device</span><span class="p">)],</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="c1"># 2) Now let&#39;s transform the points using perspective matrices</span>
+    <span class="c1">#   x_out = (coeffs[0] * x + coeffs[1] * y + coeffs[2]) / (coeffs[6] * x + coeffs[7] * y + 1)</span>
+    <span class="c1">#   y_out = (coeffs[3] * x + coeffs[4] * y + coeffs[5]) / (coeffs[6] * x + coeffs[7] * y + 1)</span>
+
+    <span class="n">numer_points</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">points</span><span class="p">,</span> <span class="n">theta1</span><span class="o">.</span><span class="n">T</span><span class="p">)</span>
+    <span class="n">denom_points</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">matmul</span><span class="p">(</span><span class="n">points</span><span class="p">,</span> <span class="n">theta2</span><span class="o">.</span><span class="n">T</span><span class="p">)</span>
+    <span class="n">transformed_points</span> <span class="o">=</span> <span class="n">numer_points</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">denom_points</span><span class="p">)</span>
+
+    <span class="c1"># 3) Reshape transformed points to [N boxes, 4 points, x/y coords]</span>
+    <span class="c1"># and compute bounding box from 4 transformed points:</span>
+    <span class="n">transformed_points</span> <span class="o">=</span> <span class="n">transformed_points</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">out_bbox_mins</span><span class="p">,</span> <span class="n">out_bbox_maxs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">aminmax</span><span class="p">(</span><span class="n">transformed_points</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+    <span class="n">out_bboxes</span> <span class="o">=</span> <span class="n">clamp_bounding_boxes</span><span class="p">(</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">out_bbox_mins</span><span class="p">,</span> <span class="n">out_bbox_maxs</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span>
+        <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="c1"># out_bboxes should be of shape [N boxes, 4]</span>
+
+    <span class="k">return</span> <span class="n">convert_bounding_box_format</span><span class="p">(</span>
+        <span class="n">out_bboxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span>
+    <span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">original_shape</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">perspective</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_perspective_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">perspective_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span>
+        <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">,</span>
+        <span class="n">startpoints</span><span class="o">=</span><span class="n">startpoints</span><span class="p">,</span>
+        <span class="n">endpoints</span><span class="o">=</span><span class="n">endpoints</span><span class="p">,</span>
+        <span class="n">coefficients</span><span class="o">=</span><span class="n">coefficients</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">perspective_mask</span><span class="p">(</span>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">perspective_image</span><span class="p">(</span>
+        <span class="n">mask</span><span class="p">,</span> <span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">coefficients</span><span class="o">=</span><span class="n">coefficients</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">perspective</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_perspective_mask_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">perspective_mask</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span>
+        <span class="n">startpoints</span><span class="o">=</span><span class="n">startpoints</span><span class="p">,</span>
+        <span class="n">endpoints</span><span class="o">=</span><span class="n">endpoints</span><span class="p">,</span>
+        <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span>
+        <span class="n">coefficients</span><span class="o">=</span><span class="n">coefficients</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">perspective</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">perspective_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">startpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">endpoints</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">coefficients</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">perspective_image</span><span class="p">(</span>
+        <span class="n">video</span><span class="p">,</span> <span class="n">startpoints</span><span class="p">,</span> <span class="n">endpoints</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">,</span> <span class="n">coefficients</span><span class="o">=</span><span class="n">coefficients</span>
+    <span class="p">)</span>
+
+
+<div class="viewcode-block" id="elastic"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.elastic.html#torchvision.transforms.v2.functional.elastic">[docs]</a><span class="k">def</span> <span class="nf">elastic</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.ElasticTransform` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">elastic_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">displacement</span><span class="o">=</span><span class="n">displacement</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">elastic</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">elastic</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">displacement</span><span class="o">=</span><span class="n">displacement</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span></div>
+
+
+<span class="n">elastic_transform</span> <span class="o">=</span> <span class="n">elastic</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">elastic</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">elastic</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">elastic_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">displacement</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument displacement should be a Tensor&quot;</span><span class="p">)</span>
+
+    <span class="n">interpolation</span> <span class="o">=</span> <span class="n">_check_interpolation</span><span class="p">(</span><span class="n">interpolation</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">ndim</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">ndim</span>
+
+    <span class="n">device</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">device</span>
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">image</span><span class="p">)</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+
+    <span class="c1"># Patch: elastic transform should support (cpu,f16) input</span>
+    <span class="n">is_cpu_half</span> <span class="o">=</span> <span class="n">device</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span> <span class="ow">and</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float16</span>
+    <span class="k">if</span> <span class="n">is_cpu_half</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+
+    <span class="c1"># We are aware that if input image dtype is uint8 and displacement is float64 then</span>
+    <span class="c1"># displacement will be casted to float32 and all computations will be done with float32</span>
+    <span class="c1"># We can fix this later if needed</span>
+
+    <span class="n">expected_shape</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,)</span> <span class="o">+</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span> <span class="o">+</span> <span class="p">(</span><span class="mi">2</span><span class="p">,)</span>
+    <span class="k">if</span> <span class="n">expected_shape</span> <span class="o">!=</span> <span class="n">displacement</span><span class="o">.</span><span class="n">shape</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Argument displacement shape should be </span><span class="si">{</span><span class="n">expected_shape</span><span class="si">}</span><span class="s2">, but given </span><span class="si">{</span><span class="n">displacement</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="o">-</span><span class="mi">1</span><span class="p">,)</span> <span class="o">+</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:])</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">elif</span> <span class="n">ndim</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_unsquash</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="k">if</span> <span class="n">displacement</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">dtype</span> <span class="ow">or</span> <span class="n">displacement</span><span class="o">.</span><span class="n">device</span> <span class="o">!=</span> <span class="n">device</span><span class="p">:</span>
+        <span class="n">displacement</span> <span class="o">=</span> <span class="n">displacement</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+
+    <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="n">_create_identity_grid</span><span class="p">((</span><span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">),</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">displacement</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">_apply_grid_transform</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">grid</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_unsquash</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">is_cpu_half</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">elastic</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_elastic_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">t_img</span> <span class="o">=</span> <span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">elastic_image</span><span class="p">(</span><span class="n">t_img</span><span class="p">,</span> <span class="n">displacement</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">to_pil_image</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">mode</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_create_identity_grid</span><span class="p">(</span><span class="n">size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">device</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">sy</span><span class="p">,</span> <span class="n">sx</span> <span class="o">=</span> <span class="n">size</span>
+    <span class="n">base_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">sy</span><span class="p">,</span> <span class="n">sx</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">x_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">((</span><span class="o">-</span><span class="n">sx</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">sx</span><span class="p">,</span> <span class="p">(</span><span class="n">sx</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">sx</span><span class="p">,</span> <span class="n">sx</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">x_grid</span><span class="p">)</span>
+
+    <span class="n">y_grid</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">((</span><span class="o">-</span><span class="n">sy</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">sy</span><span class="p">,</span> <span class="p">(</span><span class="n">sy</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">sy</span><span class="p">,</span> <span class="n">sy</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze_</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">base_grid</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">y_grid</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">base_grid</span>
+
+
+<span class="k">def</span> <span class="nf">elastic_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">expected_shape</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">canvas_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">canvas_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">displacement</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;Argument displacement should be a Tensor&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">displacement</span><span class="o">.</span><span class="n">shape</span> <span class="o">!=</span> <span class="n">expected_shape</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Argument displacement shape should be </span><span class="si">{</span><span class="n">expected_shape</span><span class="si">}</span><span class="s2">, but given </span><span class="si">{</span><span class="n">displacement</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">bounding_boxes</span>
+
+    <span class="c1"># TODO: add in docstring about approximation we are doing for grid inversion</span>
+    <span class="n">device</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">device</span>
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">)</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+
+    <span class="k">if</span> <span class="n">displacement</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">dtype</span> <span class="ow">or</span> <span class="n">displacement</span><span class="o">.</span><span class="n">device</span> <span class="o">!=</span> <span class="n">device</span><span class="p">:</span>
+        <span class="n">displacement</span> <span class="o">=</span> <span class="n">displacement</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+
+    <span class="n">original_shape</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">shape</span>
+    <span class="c1"># TODO: first cast to float if bbox is int64 before convert_bounding_box_format</span>
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">convert_bounding_box_format</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">)</span>
+    <span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+    <span class="n">id_grid</span> <span class="o">=</span> <span class="n">_create_identity_grid</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="c1"># We construct an approximation of inverse grid as inv_grid = id_grid - displacement</span>
+    <span class="c1"># This is not an exact inverse of the grid</span>
+    <span class="n">inv_grid</span> <span class="o">=</span> <span class="n">id_grid</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">displacement</span><span class="p">)</span>
+
+    <span class="c1"># Get points from bboxes</span>
+    <span class="n">points</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="p">[:,</span> <span class="p">[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">3</span><span class="p">]]]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">points</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">():</span>
+        <span class="n">points</span> <span class="o">=</span> <span class="n">points</span><span class="o">.</span><span class="n">ceil_</span><span class="p">()</span>
+    <span class="n">index_xy</span> <span class="o">=</span> <span class="n">points</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)</span>
+    <span class="n">index_x</span><span class="p">,</span> <span class="n">index_y</span> <span class="o">=</span> <span class="n">index_xy</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">index_xy</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span>
+
+    <span class="c1"># Transform points:</span>
+    <span class="n">t_size</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">canvas_size</span><span class="p">[::</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">displacement</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">displacement</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="n">transformed_points</span> <span class="o">=</span> <span class="n">inv_grid</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">index_y</span><span class="p">,</span> <span class="n">index_x</span><span class="p">,</span> <span class="p">:]</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">0.5</span> <span class="o">*</span> <span class="n">t_size</span><span class="p">)</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="mf">0.5</span><span class="p">)</span>
+
+    <span class="n">transformed_points</span> <span class="o">=</span> <span class="n">transformed_points</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">out_bbox_mins</span><span class="p">,</span> <span class="n">out_bbox_maxs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">aminmax</span><span class="p">(</span><span class="n">transformed_points</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">out_bboxes</span> <span class="o">=</span> <span class="n">clamp_bounding_boxes</span><span class="p">(</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">out_bbox_mins</span><span class="p">,</span> <span class="n">out_bbox_maxs</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span>
+        <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+        <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">convert_bounding_box_format</span><span class="p">(</span>
+        <span class="n">out_bboxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span>
+    <span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">original_shape</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">elastic</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_elastic_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">elastic_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">displacement</span><span class="o">=</span><span class="n">displacement</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">elastic_mask</span><span class="p">(</span>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">elastic_image</span><span class="p">(</span><span class="n">mask</span><span class="p">,</span> <span class="n">displacement</span><span class="o">=</span><span class="n">displacement</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">InterpolationMode</span><span class="o">.</span><span class="n">NEAREST</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">elastic</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_elastic_mask_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">elastic_mask</span><span class="p">(</span><span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="n">displacement</span><span class="o">=</span><span class="n">displacement</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">elastic</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">elastic_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">displacement</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">elastic_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">displacement</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="center_crop"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.center_crop.html#torchvision.transforms.v2.functional.center_crop">[docs]</a><span class="k">def</span> <span class="nf">center_crop</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomCrop` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">center_crop_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">output_size</span><span class="o">=</span><span class="n">output_size</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">center_crop</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">center_crop</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">output_size</span><span class="o">=</span><span class="n">output_size</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_center_crop_parse_output_size</span><span class="p">(</span><span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output_size</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+        <span class="k">return</span> <span class="p">[</span><span class="n">s</span><span class="p">,</span> <span class="n">s</span><span class="p">]</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output_size</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">return</span> <span class="p">[</span><span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">output_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">list</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_center_crop_compute_padding</span><span class="p">(</span><span class="n">crop_height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">image_height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">image_width</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">return</span> <span class="p">[</span>
+        <span class="p">(</span><span class="n">crop_width</span> <span class="o">-</span> <span class="n">image_width</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="p">(</span><span class="n">crop_height</span> <span class="o">-</span> <span class="n">image_height</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="p">(</span><span class="n">crop_width</span> <span class="o">-</span> <span class="n">image_width</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="p">(</span><span class="n">crop_height</span> <span class="o">-</span> <span class="n">image_height</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span> <span class="k">if</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span> <span class="k">else</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_center_crop_compute_crop_anchor</span><span class="p">(</span>
+    <span class="n">crop_height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">image_height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">image_width</span><span class="p">:</span> <span class="nb">int</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+    <span class="n">crop_top</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">((</span><span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">)</span> <span class="o">/</span> <span class="mf">2.0</span><span class="p">))</span>
+    <span class="n">crop_left</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">round</span><span class="p">((</span><span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">)</span> <span class="o">/</span> <span class="mf">2.0</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">crop_top</span><span class="p">,</span> <span class="n">crop_left</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">center_crop</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">center_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">center_crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span> <span class="o">=</span> <span class="n">_center_crop_parse_output_size</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">[:</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span> <span class="o">+</span> <span class="p">(</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">))</span>
+    <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+
+    <span class="k">if</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span> <span class="ow">or</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span><span class="p">:</span>
+        <span class="n">padding_ltrb</span> <span class="o">=</span> <span class="n">_center_crop_compute_padding</span><span class="p">(</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">)</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">torch_pad</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">_parse_pad_padding</span><span class="p">(</span><span class="n">padding_ltrb</span><span class="p">),</span> <span class="n">value</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+
+        <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+        <span class="k">if</span> <span class="n">crop_width</span> <span class="o">==</span> <span class="n">image_width</span> <span class="ow">and</span> <span class="n">crop_height</span> <span class="o">==</span> <span class="n">image_height</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">crop_top</span><span class="p">,</span> <span class="n">crop_left</span> <span class="o">=</span> <span class="n">_center_crop_compute_crop_anchor</span><span class="p">(</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">image</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">crop_top</span> <span class="p">:</span> <span class="p">(</span><span class="n">crop_top</span> <span class="o">+</span> <span class="n">crop_height</span><span class="p">),</span> <span class="n">crop_left</span> <span class="p">:</span> <span class="p">(</span><span class="n">crop_left</span> <span class="o">+</span> <span class="n">crop_width</span><span class="p">)]</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">center_crop</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_center_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span> <span class="o">=</span> <span class="n">_center_crop_parse_output_size</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+    <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">_get_size_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span> <span class="ow">or</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span><span class="p">:</span>
+        <span class="n">padding_ltrb</span> <span class="o">=</span> <span class="n">_center_crop_compute_padding</span><span class="p">(</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">)</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">_pad_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">padding_ltrb</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+        <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">_get_size_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">crop_width</span> <span class="o">==</span> <span class="n">image_width</span> <span class="ow">and</span> <span class="n">crop_height</span> <span class="o">==</span> <span class="n">image_height</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">crop_top</span><span class="p">,</span> <span class="n">crop_left</span> <span class="o">=</span> <span class="n">_center_crop_compute_crop_anchor</span><span class="p">(</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">crop_top</span><span class="p">,</span> <span class="n">crop_left</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">center_crop_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+    <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+    <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span> <span class="o">=</span> <span class="n">_center_crop_parse_output_size</span><span class="p">(</span><span class="n">output_size</span><span class="p">)</span>
+    <span class="n">crop_top</span><span class="p">,</span> <span class="n">crop_left</span> <span class="o">=</span> <span class="n">_center_crop_compute_crop_anchor</span><span class="p">(</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">,</span> <span class="o">*</span><span class="n">canvas_size</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">crop_bounding_boxes</span><span class="p">(</span>
+        <span class="n">bounding_boxes</span><span class="p">,</span> <span class="nb">format</span><span class="p">,</span> <span class="n">top</span><span class="o">=</span><span class="n">crop_top</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">crop_left</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">crop_width</span>
+    <span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">center_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_center_crop_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">center_crop_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">output_size</span><span class="o">=</span><span class="n">output_size</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">center_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">center_crop_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">True</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">needs_squeeze</span> <span class="o">=</span> <span class="kc">False</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">center_crop_image</span><span class="p">(</span><span class="n">image</span><span class="o">=</span><span class="n">mask</span><span class="p">,</span> <span class="n">output_size</span><span class="o">=</span><span class="n">output_size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">needs_squeeze</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">center_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">center_crop_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">output_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">center_crop_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">output_size</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="resized_crop"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.resized_crop.html#torchvision.transforms.v2.functional.resized_crop">[docs]</a><span class="k">def</span> <span class="nf">resized_crop</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.RandomResizedCrop` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">resized_crop_image</span><span class="p">(</span>
+            <span class="n">inpt</span><span class="p">,</span>
+            <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span>
+            <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span>
+            <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span>
+            <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span>
+            <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span>
+            <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+            <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="p">,</span>
+        <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span>
+        <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span>
+        <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span>
+        <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span>
+        <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+        <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">resized_crop_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">image</span> <span class="o">=</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">resize_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_resized_crop_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">image</span> <span class="o">=</span> <span class="n">_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resize_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_resized_crop_image_pil_dispatch</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">antialias</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Anti-alias option is always applied for PIL Image input. Argument antialias is ignored.&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">_resized_crop_image_pil</span><span class="p">(</span>
+        <span class="n">image</span><span class="p">,</span>
+        <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span>
+        <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span>
+        <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span>
+        <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span>
+        <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span>
+        <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span><span class="p">,</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">resized_crop_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+    <span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">crop_bounding_boxes</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="nb">format</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">resize_bounding_boxes</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_resized_crop_bounding_boxes_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="o">**</span><span class="n">kwargs</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">:</span>
+    <span class="n">output</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">resized_crop_bounding_boxes</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">resized_crop_mask</span><span class="p">(</span>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">mask</span> <span class="o">=</span> <span class="n">crop_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">resize_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_resized_crop_mask_dispatch</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="o">**</span><span class="n">kwargs</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">:</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">resized_crop_mask</span><span class="p">(</span>
+        <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="n">top</span><span class="o">=</span><span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="o">=</span><span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="o">=</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">resized_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">resized_crop_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">top</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">left</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">height</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
+    <span class="n">interpolation</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">InterpolationMode</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">InterpolationMode</span><span class="o">.</span><span class="n">BILINEAR</span><span class="p">,</span>
+    <span class="n">antialias</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="s2">&quot;warn&quot;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">resized_crop_image</span><span class="p">(</span>
+        <span class="n">video</span><span class="p">,</span> <span class="n">top</span><span class="p">,</span> <span class="n">left</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">antialias</span><span class="o">=</span><span class="n">antialias</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">interpolation</span><span class="o">=</span><span class="n">interpolation</span>
+    <span class="p">)</span>
+
+
+<div class="viewcode-block" id="five_crop"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.five_crop.html#torchvision.transforms.v2.functional.five_crop">[docs]</a><span class="k">def</span> <span class="nf">five_crop</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.FiveCrop` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">five_crop_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">five_crop</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">five_crop</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_parse_five_crop_size</span><span class="p">(</span><span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="n">numbers</span><span class="o">.</span><span class="n">Number</span><span class="p">):</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">size</span><span class="p">)</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">[</span><span class="n">s</span><span class="p">,</span> <span class="n">s</span><span class="p">]</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">size</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">s</span> <span class="o">=</span> <span class="n">size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="p">[</span><span class="n">s</span><span class="p">,</span> <span class="n">s</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Please provide only two dimensions (h, w) for size.&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">size</span>
+
+
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">five_crop</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">five_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">five_crop_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]:</span>
+    <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span> <span class="o">=</span> <span class="n">_parse_five_crop_size</span><span class="p">(</span><span class="n">size</span><span class="p">)</span>
+    <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+
+    <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="ow">or</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Requested crop size </span><span class="si">{</span><span class="n">size</span><span class="si">}</span><span class="s2"> is bigger than input size </span><span class="si">{</span><span class="p">(</span><span class="n">image_height</span><span class="p">,</span><span class="w"> </span><span class="n">image_width</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="n">tl</span> <span class="o">=</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">tr</span> <span class="o">=</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">bl</span> <span class="o">=</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">br</span> <span class="o">=</span> <span class="n">crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">center</span> <span class="o">=</span> <span class="n">center_crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="p">[</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">])</span>
+
+    <span class="k">return</span> <span class="n">tl</span><span class="p">,</span> <span class="n">tr</span><span class="p">,</span> <span class="n">bl</span><span class="p">,</span> <span class="n">br</span><span class="p">,</span> <span class="n">center</span>
+
+
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">five_crop</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_five_crop_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">]:</span>
+    <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span> <span class="o">=</span> <span class="n">_parse_five_crop_size</span><span class="p">(</span><span class="n">size</span><span class="p">)</span>
+    <span class="n">image_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">=</span> <span class="n">_get_size_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">crop_width</span> <span class="o">&gt;</span> <span class="n">image_width</span> <span class="ow">or</span> <span class="n">crop_height</span> <span class="o">&gt;</span> <span class="n">image_height</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Requested crop size </span><span class="si">{</span><span class="n">size</span><span class="si">}</span><span class="s2"> is bigger than input size </span><span class="si">{</span><span class="p">(</span><span class="n">image_height</span><span class="p">,</span><span class="w"> </span><span class="n">image_width</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="n">tl</span> <span class="o">=</span> <span class="n">_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">tr</span> <span class="o">=</span> <span class="n">_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">bl</span> <span class="o">=</span> <span class="n">_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">br</span> <span class="o">=</span> <span class="n">_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">image_height</span> <span class="o">-</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">image_width</span> <span class="o">-</span> <span class="n">crop_width</span><span class="p">,</span> <span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">)</span>
+    <span class="n">center</span> <span class="o">=</span> <span class="n">_center_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="p">[</span><span class="n">crop_height</span><span class="p">,</span> <span class="n">crop_width</span><span class="p">])</span>
+
+    <span class="k">return</span> <span class="n">tl</span><span class="p">,</span> <span class="n">tr</span><span class="p">,</span> <span class="n">bl</span><span class="p">,</span> <span class="n">br</span><span class="p">,</span> <span class="n">center</span>
+
+
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">five_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">five_crop_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]:</span>
+    <span class="k">return</span> <span class="n">five_crop_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="ten_crop"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.ten_crop.html#torchvision.transforms.v2.functional.ten_crop">[docs]</a><span class="k">def</span> <span class="nf">ten_crop</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">vertical_flip</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+<span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.TenCrop` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">ten_crop_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">vertical_flip</span><span class="o">=</span><span class="n">vertical_flip</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">ten_crop</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">ten_crop</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">,</span> <span class="n">vertical_flip</span><span class="o">=</span><span class="n">vertical_flip</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">ten_crop</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">ten_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">ten_crop_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">vertical_flip</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+<span class="p">]:</span>
+    <span class="n">non_flipped</span> <span class="o">=</span> <span class="n">five_crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">vertical_flip</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">vertical_flip_image</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">horizontal_flip_image</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+    <span class="n">flipped</span> <span class="o">=</span> <span class="n">five_crop_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">non_flipped</span> <span class="o">+</span> <span class="n">flipped</span>
+
+
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">ten_crop</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_ten_crop_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">vertical_flip</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+    <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span>
+<span class="p">]:</span>
+    <span class="n">non_flipped</span> <span class="o">=</span> <span class="n">_five_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">vertical_flip</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">_vertical_flip_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">_horizontal_flip_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+    <span class="n">flipped</span> <span class="o">=</span> <span class="n">_five_crop_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">non_flipped</span> <span class="o">+</span> <span class="n">flipped</span>
+
+
+<span class="nd">@_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">ten_crop</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">ten_crop_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">vertical_flip</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+<span class="p">]:</span>
+    <span class="k">return</span> <span class="n">ten_crop_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">size</span><span class="p">,</span> <span class="n">vertical_flip</span><span class="o">=</span><span class="n">vertical_flip</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_meta.html b/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
new file mode 100644
index 00000000000..1dc9c533959
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
@@ -0,0 +1,943 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._meta &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._meta</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._meta</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">_functional_pil</span> <span class="k">as</span> <span class="n">_FP</span>
+<span class="kn">from</span> <span class="nn">torchvision.tv_tensors</span> <span class="kn">import</span> <span class="n">BoundingBoxFormat</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_get_kernel</span><span class="p">,</span> <span class="n">_register_kernel_internal</span><span class="p">,</span> <span class="n">is_pure_tensor</span>
+
+
+<span class="k">def</span> <span class="nf">get_dimensions</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">get_dimensions_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">get_dimensions</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">get_dimensions</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_dimensions</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_dimensions</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_dimensions_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="n">chw</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:])</span>
+    <span class="n">ndims</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">chw</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">ndims</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">chw</span>
+    <span class="k">elif</span> <span class="n">ndims</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="n">chw</span><span class="o">.</span><span class="n">insert</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">chw</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input tensor should have at least two dimensions, but got </span><span class="si">{</span><span class="n">ndims</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+<span class="n">_get_dimensions_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">get_dimensions</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">get_dimensions</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_dimensions</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_dimensions_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">return</span> <span class="n">get_dimensions_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">get_num_channels</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">get_num_channels_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">get_num_channels</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">get_num_channels</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_num_channels</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_num_channels</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_num_channels_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="n">chw</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:]</span>
+    <span class="n">ndims</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">chw</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">ndims</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">chw</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="k">elif</span> <span class="n">ndims</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">return</span> <span class="mi">1</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input tensor should have at least two dimensions, but got </span><span class="si">{</span><span class="n">ndims</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+<span class="n">_get_num_channels_image_pil</span> <span class="o">=</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">get_num_channels</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)(</span><span class="n">_FP</span><span class="o">.</span><span class="n">get_image_num_channels</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_num_channels</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_num_channels_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">get_num_channels_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<span class="c1"># We changed the names to ensure it can be used not only for images but also videos. Thus, we just alias it without</span>
+<span class="c1"># deprecating the old names.</span>
+<span class="n">get_image_num_channels</span> <span class="o">=</span> <span class="n">get_num_channels</span>
+
+
+<span class="k">def</span> <span class="nf">get_size</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">get_size_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">get_size</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">get_size</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_size</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_size</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_size_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="n">hw</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:])</span>
+    <span class="n">ndims</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">hw</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">ndims</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">hw</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input tensor should have at least two dimensions, but got </span><span class="si">{</span><span class="n">ndims</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_size</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_get_size_image_pil</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="n">width</span><span class="p">,</span> <span class="n">height</span> <span class="o">=</span> <span class="n">_FP</span><span class="o">.</span><span class="n">get_image_size</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="k">return</span> <span class="p">[</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">]</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_size</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_size_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">return</span> <span class="n">get_size_image</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_size</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_size_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">return</span> <span class="n">get_size_image</span><span class="p">(</span><span class="n">mask</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_size</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_size_bounding_boxes</span><span class="p">(</span><span class="n">bounding_box</span><span class="p">:</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+    <span class="k">return</span> <span class="nb">list</span><span class="p">(</span><span class="n">bounding_box</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">get_num_frames</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">get_num_frames_video</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">get_num_frames</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">get_num_frames</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_num_frames</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">get_num_frames</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">get_num_frames_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">video</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">4</span><span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_xywh_to_xyxy</span><span class="p">(</span><span class="n">xywh</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">xyxy</span> <span class="o">=</span> <span class="n">xywh</span> <span class="k">if</span> <span class="n">inplace</span> <span class="k">else</span> <span class="n">xywh</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+    <span class="n">xyxy</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">:]</span> <span class="o">+=</span> <span class="n">xyxy</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span>
+    <span class="k">return</span> <span class="n">xyxy</span>
+
+
+<span class="k">def</span> <span class="nf">_xyxy_to_xywh</span><span class="p">(</span><span class="n">xyxy</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">xywh</span> <span class="o">=</span> <span class="n">xyxy</span> <span class="k">if</span> <span class="n">inplace</span> <span class="k">else</span> <span class="n">xyxy</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+    <span class="n">xywh</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">:]</span> <span class="o">-=</span> <span class="n">xywh</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span>
+    <span class="k">return</span> <span class="n">xywh</span>
+
+
+<span class="k">def</span> <span class="nf">_cxcywh_to_xyxy</span><span class="p">(</span><span class="n">cxcywh</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">inplace</span><span class="p">:</span>
+        <span class="n">cxcywh</span> <span class="o">=</span> <span class="n">cxcywh</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+
+    <span class="c1"># Trick to do fast division by 2 and ceil, without casting. It produces the same result as</span>
+    <span class="c1"># `torchvision.ops._box_convert._box_cxcywh_to_xyxy`.</span>
+    <span class="n">half_wh</span> <span class="o">=</span> <span class="n">cxcywh</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">:]</span><span class="o">.</span><span class="n">div</span><span class="p">(</span><span class="o">-</span><span class="mi">2</span><span class="p">,</span> <span class="n">rounding_mode</span><span class="o">=</span><span class="kc">None</span> <span class="k">if</span> <span class="n">cxcywh</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span> <span class="k">else</span> <span class="s2">&quot;floor&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">abs_</span><span class="p">()</span>
+    <span class="c1"># (cx - width / 2) = x1, same for y1</span>
+    <span class="n">cxcywh</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">half_wh</span><span class="p">)</span>
+    <span class="c1"># (x1 + width) = x2, same for y2</span>
+    <span class="n">cxcywh</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">:]</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">cxcywh</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">])</span>
+
+    <span class="k">return</span> <span class="n">cxcywh</span>
+
+
+<span class="k">def</span> <span class="nf">_xyxy_to_cxcywh</span><span class="p">(</span><span class="n">xyxy</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">inplace</span><span class="p">:</span>
+        <span class="n">xyxy</span> <span class="o">=</span> <span class="n">xyxy</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+
+    <span class="c1"># (x2 - x1) = width, same for height</span>
+    <span class="n">xyxy</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">:]</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">xyxy</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">])</span>
+    <span class="c1"># (x1 * 2 + width) / 2 = x1 + width / 2 = x1 + (x2-x1)/2 = (x1 + x2)/2 = cx, same for cy</span>
+    <span class="n">xyxy</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">xyxy</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">2</span><span class="p">:])</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="n">rounding_mode</span><span class="o">=</span><span class="kc">None</span> <span class="k">if</span> <span class="n">xyxy</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span> <span class="k">else</span> <span class="s2">&quot;floor&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">xyxy</span>
+
+
+<span class="k">def</span> <span class="nf">_convert_bounding_box_format</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">old_format</span><span class="p">:</span> <span class="n">BoundingBoxFormat</span><span class="p">,</span> <span class="n">new_format</span><span class="p">:</span> <span class="n">BoundingBoxFormat</span><span class="p">,</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+    <span class="k">if</span> <span class="n">new_format</span> <span class="o">==</span> <span class="n">old_format</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">bounding_boxes</span>
+
+    <span class="c1"># TODO: Add _xywh_to_cxcywh and _cxcywh_to_xywh to improve performance</span>
+    <span class="k">if</span> <span class="n">old_format</span> <span class="o">==</span> <span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYWH</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">_xywh_to_xyxy</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">inplace</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">old_format</span> <span class="o">==</span> <span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">CXCYWH</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">_cxcywh_to_xyxy</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">inplace</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">new_format</span> <span class="o">==</span> <span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYWH</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">_xyxy_to_xywh</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">inplace</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">new_format</span> <span class="o">==</span> <span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">CXCYWH</span><span class="p">:</span>
+        <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">_xyxy_to_cxcywh</span><span class="p">(</span><span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">inplace</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">bounding_boxes</span>
+
+
+<div class="viewcode-block" id="convert_bounding_box_format"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html#torchvision.transforms.v2.functional.convert_bounding_box_format">[docs]</a><span class="k">def</span> <span class="nf">convert_bounding_box_format</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">old_format</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">BoundingBoxFormat</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">new_format</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">BoundingBoxFormat</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :func:`~torchvision.transforms.v2.ConvertBoundingBoxFormat` for details.&quot;&quot;&quot;</span>
+    <span class="c1"># This being a kernel / functional hybrid, we need an option to pass `old_format` explicitly for pure tensor</span>
+    <span class="c1"># inputs as well as extract it from `tv_tensors.BoundingBoxes` inputs. However, putting a default value on</span>
+    <span class="c1"># `old_format` means we also need to put one on `new_format` to have syntactically correct Python. Here we mimic the</span>
+    <span class="c1"># default error that would be thrown if `new_format` had no default value.</span>
+    <span class="k">if</span> <span class="n">new_format</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;convert_bounding_box_format() missing 1 required argument: &#39;new_format&#39;&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">convert_bounding_box_format</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">or</span> <span class="n">is_pure_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">old_format</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;For pure tensor inputs, `old_format` has to be passed.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">_convert_bounding_box_format</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="n">old_format</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="n">new_format</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">old_format</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;For bounding box tv_tensor inputs, `old_format` must not be passed.&quot;</span><span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">_convert_bounding_box_format</span><span class="p">(</span>
+            <span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="n">old_format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="n">new_format</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="n">new_format</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Input can either be a plain tensor or a bounding box tv_tensor, but got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="si">}</span><span class="s2"> instead.&quot;</span>
+        <span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_clamp_bounding_boxes</span><span class="p">(</span>
+    <span class="n">bounding_boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="nb">format</span><span class="p">:</span> <span class="n">BoundingBoxFormat</span><span class="p">,</span> <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># TODO: Investigate if it makes sense from a performance perspective to have an implementation for every</span>
+    <span class="c1">#  BoundingBoxFormat instead of converting back and forth</span>
+    <span class="n">in_dtype</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span> <span class="k">if</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span> <span class="k">else</span> <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+    <span class="n">xyxy_boxes</span> <span class="o">=</span> <span class="n">convert_bounding_box_format</span><span class="p">(</span>
+        <span class="n">bounding_boxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span>
+    <span class="p">)</span>
+    <span class="n">xyxy_boxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">0</span><span class="p">::</span><span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+    <span class="n">xyxy_boxes</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">::</span><span class="mi">2</span><span class="p">]</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+    <span class="n">out_boxes</span> <span class="o">=</span> <span class="n">convert_bounding_box_format</span><span class="p">(</span>
+        <span class="n">xyxy_boxes</span><span class="p">,</span> <span class="n">old_format</span><span class="o">=</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span> <span class="n">new_format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">out_boxes</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">in_dtype</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="clamp_bounding_boxes"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html#torchvision.transforms.v2.functional.clamp_bounding_boxes">[docs]</a><span class="k">def</span> <span class="nf">clamp_bounding_boxes</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">BoundingBoxFormat</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :func:`~torchvision.transforms.v2.ClampBoundingBoxes` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">clamp_bounding_boxes</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">or</span> <span class="n">is_pure_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">):</span>
+
+        <span class="k">if</span> <span class="nb">format</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">canvas_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;For pure tensor inputs, `format` and `canvas_size` has to be passed.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">_clamp_bounding_boxes</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">format</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">canvas_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;For bounding box tv_tensor inputs, `format` and `canvas_size` must not be passed.&quot;</span><span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">_clamp_bounding_boxes</span><span class="p">(</span><span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="nb">format</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">inpt</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Input can either be a plain tensor or a bounding box tv_tensor, but got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="si">}</span><span class="s2"> instead.&quot;</span>
+        <span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_misc.html b/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
new file mode 100644
index 00000000000..d93715faee1
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
@@ -0,0 +1,946 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._misc &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._misc</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._misc</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch.nn.functional</span> <span class="kn">import</span> <span class="n">conv2d</span><span class="p">,</span> <span class="n">pad</span> <span class="k">as</span> <span class="n">torch_pad</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms._functional_tensor</span> <span class="kn">import</span> <span class="n">_max_value</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.functional</span> <span class="kn">import</span> <span class="n">pil_to_tensor</span><span class="p">,</span> <span class="n">to_pil_image</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_get_kernel</span><span class="p">,</span> <span class="n">_register_kernel_internal</span>
+
+
+<div class="viewcode-block" id="normalize"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.normalize.html#torchvision.transforms.v2.functional.normalize">[docs]</a><span class="k">def</span> <span class="nf">normalize</span><span class="p">(</span>
+    <span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">mean</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">std</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+    <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.Normalize` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">normalize_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="n">mean</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">std</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">normalize</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">normalize</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="n">mean</span><span class="p">,</span> <span class="n">std</span><span class="o">=</span><span class="n">std</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">normalize</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">normalize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">normalize_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mean</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">std</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">():</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input tensor should be a float tensor. Got </span><span class="si">{</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Expected tensor to be a tensor image of size (..., C, H, W). Got </span><span class="si">{</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">std</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+        <span class="n">divzero</span> <span class="o">=</span> <span class="ow">not</span> <span class="nb">all</span><span class="p">(</span><span class="n">std</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">std</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+        <span class="n">divzero</span> <span class="o">=</span> <span class="n">std</span> <span class="o">==</span> <span class="mi">0</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">divzero</span> <span class="o">=</span> <span class="kc">False</span>
+    <span class="k">if</span> <span class="n">divzero</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;std evaluated to zero, leading to division by zero.&quot;</span><span class="p">)</span>
+
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">device</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">device</span>
+    <span class="n">mean</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">mean</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">std</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">std</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">mean</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">mean</span> <span class="o">=</span> <span class="n">mean</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">std</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">std</span> <span class="o">=</span> <span class="n">std</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">inplace</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">mean</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">sub</span><span class="p">(</span><span class="n">mean</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">std</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">normalize</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">normalize_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mean</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">std</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">inplace</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">normalize_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">mean</span><span class="p">,</span> <span class="n">std</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="n">inplace</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="gaussian_blur"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.gaussian_blur.html#torchvision.transforms.v2.functional.gaussian_blur">[docs]</a><span class="k">def</span> <span class="nf">gaussian_blur</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.GaussianBlur` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">gaussian_blur_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">sigma</span><span class="o">=</span><span class="n">sigma</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">gaussian_blur</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">sigma</span><span class="o">=</span><span class="n">sigma</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_get_gaussian_kernel1d</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">sigma</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">lim</span> <span class="o">=</span> <span class="p">(</span><span class="n">kernel_size</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="mf">2.0</span> <span class="o">*</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mf">2.0</span><span class="p">)</span> <span class="o">*</span> <span class="n">sigma</span><span class="p">)</span>
+    <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="o">-</span><span class="n">lim</span><span class="p">,</span> <span class="n">lim</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">kernel1d</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">pow_</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">neg_</span><span class="p">(),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">kernel1d</span>
+
+
+<span class="k">def</span> <span class="nf">_get_gaussian_kernel2d</span><span class="p">(</span>
+    <span class="n">kernel_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="n">kernel1d_x</span> <span class="o">=</span> <span class="n">_get_gaussian_kernel1d</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="p">)</span>
+    <span class="n">kernel1d_y</span> <span class="o">=</span> <span class="n">_get_gaussian_kernel1d</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">sigma</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="p">)</span>
+    <span class="n">kernel2d</span> <span class="o">=</span> <span class="n">kernel1d_y</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">kernel1d_x</span>
+    <span class="k">return</span> <span class="n">kernel2d</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">gaussian_blur_image</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># TODO: consider deprecating integers from sigma on the future</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+        <span class="n">kernel_size</span> <span class="o">=</span> <span class="p">[</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">]</span>
+    <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If kernel_size is a sequence its length should be 2. Got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">ksize</span> <span class="ow">in</span> <span class="n">kernel_size</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">ksize</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">ksize</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;kernel_size should have odd and positive integers. Got </span><span class="si">{</span><span class="n">kernel_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">sigma</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="n">ksize</span> <span class="o">*</span> <span class="mf">0.15</span> <span class="o">+</span> <span class="mf">0.35</span> <span class="k">for</span> <span class="n">ksize</span> <span class="ow">in</span> <span class="n">kernel_size</span><span class="p">]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+            <span class="n">length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">length</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="n">s</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">sigma</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+                <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="n">s</span><span class="p">,</span> <span class="n">s</span><span class="p">]</span>
+            <span class="k">elif</span> <span class="n">length</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;If sigma is a sequence, its length should be 2. Got </span><span class="si">{</span><span class="n">length</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">sigma</span><span class="p">,</span> <span class="p">(</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">)):</span>
+            <span class="n">s</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span>
+            <span class="n">sigma</span> <span class="o">=</span> <span class="p">[</span><span class="n">s</span><span class="p">,</span> <span class="n">s</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sigma should be either float or sequence of floats. Got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">sigma</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">sigma</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">s</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;sigma should have positive values. Got </span><span class="si">{</span><span class="n">sigma</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">dtype</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="n">shape</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">ndim</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">ndim</span>
+    <span class="k">if</span> <span class="n">ndim</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">reshape</span><span class="p">((</span><span class="o">-</span><span class="mi">1</span><span class="p">,)</span> <span class="o">+</span> <span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">:])</span>
+
+    <span class="n">fp</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_gaussian_kernel2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">sigma</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">kernel</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">],</span> <span class="mi">1</span><span class="p">,</span> <span class="n">kernel</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">kernel</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+
+    <span class="n">output</span> <span class="o">=</span> <span class="n">image</span> <span class="k">if</span> <span class="n">fp</span> <span class="k">else</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+
+    <span class="c1"># padding = (left, right, top, bottom)</span>
+    <span class="n">padding</span> <span class="o">=</span> <span class="p">[</span><span class="n">kernel_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">//</span> <span class="mi">2</span><span class="p">]</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">torch_pad</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">padding</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s2">&quot;reflect&quot;</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">conv2d</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">kernel</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">3</span><span class="p">])</span>
+
+    <span class="k">if</span> <span class="n">ndim</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">ndim</span> <span class="o">&gt;</span> <span class="mi">4</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">shape</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">fp</span><span class="p">:</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">round_</span><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">output</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_gaussian_blur_image_pil</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+    <span class="n">t_img</span> <span class="o">=</span> <span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+    <span class="n">output</span> <span class="o">=</span> <span class="n">gaussian_blur_image</span><span class="p">(</span><span class="n">t_img</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">kernel_size</span><span class="p">,</span> <span class="n">sigma</span><span class="o">=</span><span class="n">sigma</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">to_pil_image</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">mode</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">gaussian_blur</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">gaussian_blur_video</span><span class="p">(</span>
+    <span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">sigma</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">gaussian_blur_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">kernel_size</span><span class="p">,</span> <span class="n">sigma</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="to_dtype"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.to_dtype.html#torchvision.transforms.v2.functional.to_dtype">[docs]</a><span class="k">def</span> <span class="nf">to_dtype</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">,</span> <span class="n">scale</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :func:`~torchvision.transforms.v2.ToDtype` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">to_dtype</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">to_dtype</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_num_value_bits</span><span class="p">(</span><span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+        <span class="k">return</span> <span class="mi">8</span>
+    <span class="k">elif</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">int8</span><span class="p">:</span>
+        <span class="k">return</span> <span class="mi">7</span>
+    <span class="k">elif</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">int16</span><span class="p">:</span>
+        <span class="k">return</span> <span class="mi">15</span>
+    <span class="k">elif</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">:</span>
+        <span class="k">return</span> <span class="mi">31</span>
+    <span class="k">elif</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">:</span>
+        <span class="k">return</span> <span class="mi">63</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Number of value bits is only defined for integer dtypes, but got </span><span class="si">{</span><span class="n">dtype</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">to_dtype</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">to_dtype</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">to_dtype_image</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">,</span> <span class="n">scale</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">dtype</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="n">scale</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
+
+    <span class="n">float_input</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="c1"># TODO: remove this branch as soon as `dtype.is_floating_point` is supported by JIT</span>
+        <span class="n">float_output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">float_output</span> <span class="o">=</span> <span class="n">dtype</span><span class="o">.</span><span class="n">is_floating_point</span>
+
+    <span class="k">if</span> <span class="n">float_input</span><span class="p">:</span>
+        <span class="c1"># float to float</span>
+        <span class="k">if</span> <span class="n">float_output</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
+
+        <span class="c1"># float to int</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span> <span class="ow">and</span> <span class="n">dtype</span> <span class="ow">in</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">))</span> <span class="ow">or</span> <span class="p">(</span>
+            <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float64</span> <span class="ow">and</span> <span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">int64</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The conversion from </span><span class="si">{</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2"> to </span><span class="si">{</span><span class="n">dtype</span><span class="si">}</span><span class="s2"> cannot be performed safely.&quot;</span><span class="p">)</span>
+
+        <span class="c1"># For data in the range `[0.0, 1.0]`, just multiplying by the maximum value of the integer range and converting</span>
+        <span class="c1"># to the integer dtype  is not sufficient. For example, `torch.rand(...).mul(255).to(torch.uint8)` will only</span>
+        <span class="c1"># be `255` if the input is exactly `1.0`. See https://github.com/pytorch/vision/pull/2078#issuecomment-612045321</span>
+        <span class="c1"># for a detailed analysis.</span>
+        <span class="c1"># To mitigate this, we could round before we convert to the integer dtype, but this is an extra operation.</span>
+        <span class="c1"># Instead, we can also multiply by the maximum value plus something close to `1`. See</span>
+        <span class="c1"># https://github.com/pytorch/vision/pull/2078#issuecomment-613524965 for details.</span>
+        <span class="n">eps</span> <span class="o">=</span> <span class="mf">1e-3</span>
+        <span class="n">max_value</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="n">_max_value</span><span class="p">(</span><span class="n">dtype</span><span class="p">))</span>
+        <span class="c1"># We need to scale first since the conversion would otherwise turn the input range `[0.0, 1.0]` into the</span>
+        <span class="c1"># discrete set `{0, 1}`.</span>
+        <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">max_value</span> <span class="o">+</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">eps</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="c1"># int to float</span>
+        <span class="k">if</span> <span class="n">float_output</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">/</span> <span class="n">_max_value</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">))</span>
+
+        <span class="c1"># int to int</span>
+        <span class="n">num_value_bits_input</span> <span class="o">=</span> <span class="n">_num_value_bits</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="n">num_value_bits_output</span> <span class="o">=</span> <span class="n">_num_value_bits</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">num_value_bits_input</span> <span class="o">&gt;</span> <span class="n">num_value_bits_output</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">bitwise_right_shift</span><span class="p">(</span><span class="n">num_value_bits_input</span> <span class="o">-</span> <span class="n">num_value_bits_output</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">bitwise_left_shift_</span><span class="p">(</span><span class="n">num_value_bits_output</span> <span class="o">-</span> <span class="n">num_value_bits_input</span><span class="p">)</span>
+
+
+<span class="c1"># We encourage users to use to_dtype() instead but we keep this for BC</span>
+<div class="viewcode-block" id="convert_image_dtype"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.convert_image_dtype.html#torchvision.transforms.v2.functional.convert_image_dtype">[docs]</a><span class="k">def</span> <span class="nf">convert_image_dtype</span><span class="p">(</span><span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] [DEPRECATED] Use to_dtype() instead.&quot;&quot;&quot;</span>
+    <span class="k">return</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">to_dtype</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">to_dtype_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">,</span> <span class="n">scale</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="k">return</span> <span class="n">to_dtype_image</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="n">scale</span><span class="p">)</span>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">to_dtype</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">to_dtype</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">_to_dtype_tensor_dispatch</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">dtype</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">scale</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># We don&#39;t need to unwrap and rewrap here, since TVTensor.to() preserves the type</span>
+    <span class="k">return</span> <span class="n">inpt</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html b/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
new file mode 100644
index 00000000000..618f64b4bbe
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
@@ -0,0 +1,696 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._temporal &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._temporal</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._temporal</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">_log_api_usage_once</span>
+
+<span class="kn">from</span> <span class="nn">._utils</span> <span class="kn">import</span> <span class="n">_get_kernel</span><span class="p">,</span> <span class="n">_register_kernel_internal</span>
+
+
+<div class="viewcode-block" id="uniform_temporal_subsample"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html#torchvision.transforms.v2.functional.uniform_temporal_subsample">[docs]</a><span class="k">def</span> <span class="nf">uniform_temporal_subsample</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.UniformTemporalSubsample` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">():</span>
+        <span class="k">return</span> <span class="n">uniform_temporal_subsample_video</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">num_samples</span><span class="o">=</span><span class="n">num_samples</span><span class="p">)</span>
+
+    <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">uniform_temporal_subsample</span><span class="p">)</span>
+
+    <span class="n">kernel</span> <span class="o">=</span> <span class="n">_get_kernel</span><span class="p">(</span><span class="n">uniform_temporal_subsample</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">))</span>
+    <span class="k">return</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">num_samples</span><span class="o">=</span><span class="n">num_samples</span><span class="p">)</span></div>
+
+
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">uniform_temporal_subsample</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+<span class="nd">@_register_kernel_internal</span><span class="p">(</span><span class="n">uniform_temporal_subsample</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">uniform_temporal_subsample_video</span><span class="p">(</span><span class="n">video</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+    <span class="c1"># Reference: https://github.com/facebookresearch/pytorchvideo/blob/a0a131e/pytorchvideo/transforms/functional.py#L19</span>
+    <span class="n">t_max</span> <span class="o">=</span> <span class="n">video</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">4</span><span class="p">]</span> <span class="o">-</span> <span class="mi">1</span>
+    <span class="n">indices</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">linspace</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">t_max</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">video</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">long</span><span class="p">()</span>
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">index_select</span><span class="p">(</span><span class="n">video</span><span class="p">,</span> <span class="o">-</span><span class="mi">4</span><span class="p">,</span> <span class="n">indices</span><span class="p">)</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html b/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
new file mode 100644
index 00000000000..818c8305b11
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._type_conversion &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._type_conversion</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._type_conversion</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">_F</span>
+
+
+<div class="viewcode-block" id="to_image"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.to_image.html#torchvision.transforms.v2.functional.to_image">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">unused</span>
+<span class="k">def</span> <span class="nf">to_image</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] See :class:`~torchvision.transforms.v2.ToImage` for details.&quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">):</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">((</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">inpt</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input can either be a numpy array or a PIL image, but got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">inpt</span><span class="p">)</span><span class="si">}</span><span class="s2"> instead.&quot;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">(</span><span class="n">output</span><span class="p">)</span></div>
+
+
+<span class="n">to_pil_image</span> <span class="o">=</span> <span class="n">_F</span><span class="o">.</span><span class="n">to_pil_image</span>
+<span class="n">pil_to_tensor</span> <span class="o">=</span> <span class="n">_F</span><span class="o">.</span><span class="n">pil_to_tensor</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_utils.html b/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
new file mode 100644
index 00000000000..e4eec91ffbf
--- /dev/null
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
@@ -0,0 +1,810 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional._utils &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../../../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional._utils</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.transforms.v2.functional._utils</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">functools</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+
+<span class="n">_FillType</span> <span class="o">=</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">float</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">Sequence</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="kc">None</span><span class="p">]</span>
+<span class="n">_FillTypeJIT</span> <span class="o">=</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]</span>
+
+
+<span class="k">def</span> <span class="nf">is_pure_tensor</span><span class="p">(</span><span class="n">inpt</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+    <span class="k">return</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span><span class="p">)</span>
+
+
+<span class="c1"># {functional: {input_type: type_specific_kernel}}</span>
+<span class="n">_KERNEL_REGISTRY</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="n">Type</span><span class="p">,</span> <span class="n">Callable</span><span class="p">]]</span> <span class="o">=</span> <span class="p">{}</span>
+
+
+<span class="k">def</span> <span class="nf">_kernel_tv_tensor_wrapper</span><span class="p">(</span><span class="n">kernel</span><span class="p">):</span>
+    <span class="nd">@functools</span><span class="o">.</span><span class="n">wraps</span><span class="p">(</span><span class="n">kernel</span><span class="p">)</span>
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="c1"># If you&#39;re wondering whether we could / should get rid of this wrapper,</span>
+        <span class="c1"># the answer is no: we want to pass pure Tensors to avoid the overhead</span>
+        <span class="c1"># of the __torch_function__ machinery. Note that this is always valid,</span>
+        <span class="c1"># regardless of whether we override __torch_function__ in our base class</span>
+        <span class="c1"># or not.</span>
+        <span class="c1"># Also, even if we didn&#39;t call `as_subclass` here, we would still need</span>
+        <span class="c1"># this wrapper to call wrap(), because the TVTensor type would be</span>
+        <span class="c1"># lost after the first operation due to our own __torch_function__</span>
+        <span class="c1"># logic.</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">),</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="k">def</span> <span class="nf">_register_kernel_internal</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">input_type</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
+    <span class="n">registry</span> <span class="o">=</span> <span class="n">_KERNEL_REGISTRY</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="p">{})</span>
+    <span class="k">if</span> <span class="n">input_type</span> <span class="ow">in</span> <span class="n">registry</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Functional </span><span class="si">{</span><span class="n">functional</span><span class="si">}</span><span class="s2"> already has a kernel registered for type </span><span class="si">{</span><span class="n">input_type</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">decorator</span><span class="p">(</span><span class="n">kernel</span><span class="p">):</span>
+        <span class="n">registry</span><span class="p">[</span><span class="n">input_type</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">_kernel_tv_tensor_wrapper</span><span class="p">(</span><span class="n">kernel</span><span class="p">)</span>
+            <span class="k">if</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">input_type</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span><span class="p">)</span> <span class="ow">and</span> <span class="n">tv_tensor_wrapper</span>
+            <span class="k">else</span> <span class="n">kernel</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">kernel</span>
+
+    <span class="k">return</span> <span class="n">decorator</span>
+
+
+<span class="k">def</span> <span class="nf">_name_to_functional</span><span class="p">(</span><span class="n">name</span><span class="p">):</span>
+    <span class="kn">import</span> <span class="nn">torchvision.transforms.v2.functional</span>  <span class="c1"># noqa</span>
+
+    <span class="k">try</span><span class="p">:</span>
+        <span class="k">return</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torchvision</span><span class="o">.</span><span class="n">transforms</span><span class="o">.</span><span class="n">v2</span><span class="o">.</span><span class="n">functional</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
+    <span class="k">except</span> <span class="ne">AttributeError</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Could not find functional with name &#39;</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">&#39; in torchvision.transforms.v2.functional.&quot;</span>
+        <span class="p">)</span> <span class="kn">from</span> <span class="kc">None</span>
+
+
+<span class="n">_BUILTIN_DATAPOINT_TYPES</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="n">obj</span> <span class="k">for</span> <span class="n">obj</span> <span class="ow">in</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="vm">__dict__</span><span class="o">.</span><span class="n">values</span><span class="p">()</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="nb">type</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span><span class="p">)</span>
+<span class="p">}</span>
+
+
+<div class="viewcode-block" id="register_kernel"><a class="viewcode-back" href="../../../../../generated/torchvision.transforms.v2.functional.register_kernel.html#torchvision.transforms.v2.functional.register_kernel">[docs]</a><span class="k">def</span> <span class="nf">register_kernel</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">tv_tensor_cls</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Decorate a kernel to register it for a functional and a (custom) tv_tensor type.</span>
+
+<span class="sd">    See :ref:`sphx_glr_auto_examples_transforms_plot_custom_tv_tensors.py` for usage</span>
+<span class="sd">    details.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+        <span class="n">functional</span> <span class="o">=</span> <span class="n">_name_to_functional</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="n">functional</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="p">(</span>
+        <span class="nb">callable</span><span class="p">(</span><span class="n">functional</span><span class="p">)</span>
+        <span class="ow">and</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="s2">&quot;__module__&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;torchvision.transforms.v2.functional&quot;</span><span class="p">)</span>
+    <span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Kernels can only be registered on functionals from the torchvision.transforms.v2.functional namespace, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;but got </span><span class="si">{</span><span class="n">functional</span><span class="si">}</span><span class="s2">.&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">tv_tensor_cls</span><span class="p">,</span> <span class="nb">type</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">tv_tensor_cls</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Kernels can only be registered for subclasses of torchvision.tv_tensors.TVTensor, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;but got </span><span class="si">{</span><span class="n">tv_tensor_cls</span><span class="si">}</span><span class="s2">.&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">tv_tensor_cls</span> <span class="ow">in</span> <span class="n">_BUILTIN_DATAPOINT_TYPES</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Kernels cannot be registered for the builtin tv_tensor classes, but got </span><span class="si">{</span><span class="n">tv_tensor_cls</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">_register_kernel_internal</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">tv_tensor_cls</span><span class="p">,</span> <span class="n">tv_tensor_wrapper</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_get_kernel</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">input_type</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">allow_passthrough</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+    <span class="n">registry</span> <span class="o">=</span> <span class="n">_KERNEL_REGISTRY</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">functional</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">registry</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;No kernel registered for functional </span><span class="si">{</span><span class="n">functional</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="k">for</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="n">input_type</span><span class="o">.</span><span class="vm">__mro__</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="n">registry</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">registry</span><span class="p">[</span><span class="bp">cls</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="bp">cls</span> <span class="ow">is</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span><span class="p">:</span>
+            <span class="c1"># We don&#39;t want user-defined tv_tensors to dispatch to the pure Tensor kernels, so we explicit stop the</span>
+            <span class="c1"># MRO traversal before hitting torch.Tensor. We can even stop at tv_tensors.TVTensor, since we don&#39;t</span>
+            <span class="c1"># allow kernels to be registered for tv_tensors.TVTensor anyway.</span>
+            <span class="k">break</span>
+
+    <span class="k">if</span> <span class="n">allow_passthrough</span><span class="p">:</span>
+        <span class="k">return</span> <span class="k">lambda</span> <span class="n">inpt</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">inpt</span>
+
+    <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+        <span class="sa">f</span><span class="s2">&quot;Functional F.</span><span class="si">{</span><span class="n">functional</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2"> supports inputs of type </span><span class="si">{</span><span class="n">registry</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span><span class="si">}</span><span class="s2">, &quot;</span>
+        <span class="sa">f</span><span class="s2">&quot;but got </span><span class="si">{</span><span class="n">input_type</span><span class="si">}</span><span class="s2"> instead.&quot;</span>
+    <span class="p">)</span>
+
+
+<span class="c1"># This basically replicates _register_kernel_internal, but with a specialized wrapper for five_crop / ten_crop</span>
+<span class="c1"># We could get rid of this by letting _register_kernel_internal take arbitrary functionals rather than wrap_kernel: bool</span>
+<span class="k">def</span> <span class="nf">_register_five_ten_crop_kernel_internal</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="n">input_type</span><span class="p">):</span>
+    <span class="n">registry</span> <span class="o">=</span> <span class="n">_KERNEL_REGISTRY</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="n">functional</span><span class="p">,</span> <span class="p">{})</span>
+    <span class="k">if</span> <span class="n">input_type</span> <span class="ow">in</span> <span class="n">registry</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Functional &#39;</span><span class="si">{</span><span class="n">functional</span><span class="si">}</span><span class="s2">&#39; already has a kernel registered for type &#39;</span><span class="si">{</span><span class="n">input_type</span><span class="si">}</span><span class="s2">&#39;.&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">wrap</span><span class="p">(</span><span class="n">kernel</span><span class="p">):</span>
+        <span class="nd">@functools</span><span class="o">.</span><span class="n">wraps</span><span class="p">(</span><span class="n">kernel</span><span class="p">)</span>
+        <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">kernel</span><span class="p">(</span><span class="n">inpt</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="n">container_type</span> <span class="o">=</span> <span class="nb">type</span><span class="p">(</span><span class="n">output</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">container_type</span><span class="p">(</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">o</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">inpt</span><span class="p">)</span> <span class="k">for</span> <span class="n">o</span> <span class="ow">in</span> <span class="n">output</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">wrapper</span>
+
+    <span class="k">def</span> <span class="nf">decorator</span><span class="p">(</span><span class="n">kernel</span><span class="p">):</span>
+        <span class="n">registry</span><span class="p">[</span><span class="n">input_type</span><span class="p">]</span> <span class="o">=</span> <span class="n">wrap</span><span class="p">(</span><span class="n">kernel</span><span class="p">)</span> <span class="k">if</span> <span class="nb">issubclass</span><span class="p">(</span><span class="n">input_type</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span><span class="p">)</span> <span class="k">else</span> <span class="n">kernel</span>
+        <span class="k">return</span> <span class="n">kernel</span>
+
+    <span class="k">return</span> <span class="n">decorator</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../../../" src="../../../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../../../" id="documentation_options" src="../../../../../_static/documentation_options.js"></script>
+         <script src="../../../../../_static/jquery.js"></script>
+         <script src="../../../../../_static/underscore.js"></script>
+         <script src="../../../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../../../_static/doctools.js"></script>
+         <script src="../../../../../_static/clipboard.min.js"></script>
+         <script src="../../../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors.html b/0.16/_modules/torchvision/tv_tensors.html
new file mode 100644
index 00000000000..e5ee49dcdcc
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors.html
@@ -0,0 +1,700 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">._bounding_boxes</span> <span class="kn">import</span> <span class="n">BoundingBoxes</span><span class="p">,</span> <span class="n">BoundingBoxFormat</span>
+<span class="kn">from</span> <span class="nn">._image</span> <span class="kn">import</span> <span class="n">Image</span>
+<span class="kn">from</span> <span class="nn">._mask</span> <span class="kn">import</span> <span class="n">Mask</span>
+<span class="kn">from</span> <span class="nn">._torch_function_helpers</span> <span class="kn">import</span> <span class="n">set_return_type</span>
+<span class="kn">from</span> <span class="nn">._tv_tensor</span> <span class="kn">import</span> <span class="n">TVTensor</span>
+<span class="kn">from</span> <span class="nn">._video</span> <span class="kn">import</span> <span class="n">Video</span>
+
+
+<div class="viewcode-block" id="wrap"><a class="viewcode-back" href="../../generated/torchvision.tv_tensors.wrap.html#torchvision.tv_tensors.wrap">[docs]</a><span class="k">def</span> <span class="nf">wrap</span><span class="p">(</span><span class="n">wrappee</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">like</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Convert a :class:`torch.Tensor` (``wrappee``) into the same :class:`~torchvision.tv_tensors.TVTensor` subclass as ``like``.</span>
+
+<span class="sd">    If ``like`` is a :class:`~torchvision.tv_tensors.BoundingBoxes`, the ``format`` and ``canvas_size`` of</span>
+<span class="sd">    ``like`` are assigned to ``wrappee``, unless they are passed as ``kwargs``.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        wrappee (Tensor): The tensor to convert.</span>
+<span class="sd">        like (:class:`~torchvision.tv_tensors.TVTensor`): The reference.</span>
+<span class="sd">            ``wrappee`` will be converted into the same subclass as ``like``.</span>
+<span class="sd">        kwargs: Can contain &quot;format&quot; and &quot;canvas_size&quot; if ``like`` is a :class:`~torchvision.tv_tensor.BoundingBoxes`.</span>
+<span class="sd">            Ignored otherwise.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">like</span><span class="p">,</span> <span class="n">BoundingBoxes</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">BoundingBoxes</span><span class="o">.</span><span class="n">_wrap</span><span class="p">(</span>
+            <span class="n">wrappee</span><span class="p">,</span>
+            <span class="nb">format</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;format&quot;</span><span class="p">,</span> <span class="n">like</span><span class="o">.</span><span class="n">format</span><span class="p">),</span>
+            <span class="n">canvas_size</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;canvas_size&quot;</span><span class="p">,</span> <span class="n">like</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">),</span>
+        <span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">wrappee</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">like</span><span class="p">))</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html b/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
new file mode 100644
index 00000000000..58f61e69b5e
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
@@ -0,0 +1,774 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors._bounding_boxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+          <li><a href="../tv_tensors.html">torchvision.tv_tensors</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors._bounding_boxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors._bounding_boxes</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">annotations</span>
+
+<span class="kn">from</span> <span class="nn">enum</span> <span class="kn">import</span> <span class="n">Enum</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Mapping</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch.utils._pytree</span> <span class="kn">import</span> <span class="n">tree_flatten</span>
+
+<span class="kn">from</span> <span class="nn">._tv_tensor</span> <span class="kn">import</span> <span class="n">TVTensor</span>
+
+
+<div class="viewcode-block" id="BoundingBoxFormat"><a class="viewcode-back" href="../../../generated/torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat">[docs]</a><span class="k">class</span> <span class="nc">BoundingBoxFormat</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Coordinate format of a bounding box.</span>
+
+<span class="sd">    Available formats are</span>
+
+<span class="sd">    * ``XYXY``</span>
+<span class="sd">    * ``XYWH``</span>
+<span class="sd">    * ``CXCYWH``</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">XYXY</span> <span class="o">=</span> <span class="s2">&quot;XYXY&quot;</span>
+    <span class="n">XYWH</span> <span class="o">=</span> <span class="s2">&quot;XYWH&quot;</span>
+    <span class="n">CXCYWH</span> <span class="o">=</span> <span class="s2">&quot;CXCYWH&quot;</span></div>
+
+
+<div class="viewcode-block" id="BoundingBoxes"><a class="viewcode-back" href="../../../generated/torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes">[docs]</a><span class="k">class</span> <span class="nc">BoundingBoxes</span><span class="p">(</span><span class="n">TVTensor</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] :class:`torch.Tensor` subclass for bounding boxes.</span>
+
+<span class="sd">    .. note::</span>
+<span class="sd">        There should be only one :class:`~torchvision.tv_tensors.BoundingBoxes`</span>
+<span class="sd">        instance per sample e.g. ``{&quot;img&quot;: img, &quot;bbox&quot;: BoundingBoxes(...)}``,</span>
+<span class="sd">        although one :class:`~torchvision.tv_tensors.BoundingBoxes` object can</span>
+<span class="sd">        contain multiple bounding boxes.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        data: Any data that can be turned into a tensor with :func:`torch.as_tensor`.</span>
+<span class="sd">        format (BoundingBoxFormat, str): Format of the bounding box.</span>
+<span class="sd">        canvas_size (two-tuple of ints): Height and width of the corresponding image or video.</span>
+<span class="sd">        dtype (torch.dtype, optional): Desired data type of the bounding box. If omitted, will be inferred from</span>
+<span class="sd">            ``data``.</span>
+<span class="sd">        device (torch.device, optional): Desired device of the bounding box. If omitted and ``data`` is a</span>
+<span class="sd">            :class:`torch.Tensor`, the device is taken from it. Otherwise, the bounding box is constructed on the CPU.</span>
+<span class="sd">        requires_grad (bool, optional): Whether autograd should record operations on the bounding box. If omitted and</span>
+<span class="sd">            ``data`` is a :class:`torch.Tensor`, the value is taken from it. Otherwise, defaults to ``False``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="nb">format</span><span class="p">:</span> <span class="n">BoundingBoxFormat</span>
+    <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]</span>
+
+    <span class="nd">@classmethod</span>
+    <span class="k">def</span> <span class="nf">_wrap</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">tensor</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="nb">format</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">BoundingBoxFormat</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span> <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">check_dims</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">BoundingBoxes</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">if</span> <span class="n">check_dims</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="n">tensor</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">tensor</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Expected a 1D or 2D tensor, got </span><span class="si">{</span><span class="n">tensor</span><span class="o">.</span><span class="n">ndim</span><span class="si">}</span><span class="s2">D&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">format</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="nb">format</span> <span class="o">=</span> <span class="n">BoundingBoxFormat</span><span class="p">[</span><span class="nb">format</span><span class="o">.</span><span class="n">upper</span><span class="p">()]</span>
+        <span class="n">bounding_boxes</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="bp">cls</span><span class="p">)</span>
+        <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">format</span> <span class="o">=</span> <span class="nb">format</span>
+        <span class="n">bounding_boxes</span><span class="o">.</span><span class="n">canvas_size</span> <span class="o">=</span> <span class="n">canvas_size</span>
+        <span class="k">return</span> <span class="n">bounding_boxes</span>
+
+    <span class="k">def</span> <span class="fm">__new__</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">data</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+        <span class="o">*</span><span class="p">,</span>
+        <span class="nb">format</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">BoundingBoxFormat</span><span class="p">,</span> <span class="nb">str</span><span class="p">],</span>
+        <span class="n">canvas_size</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span>
+        <span class="n">dtype</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">requires_grad</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">BoundingBoxes</span><span class="p">:</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_to_tensor</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">requires_grad</span><span class="o">=</span><span class="n">requires_grad</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_wrap</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+
+    <span class="nd">@classmethod</span>
+    <span class="k">def</span> <span class="nf">_wrap_output</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">output</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">(),</span>
+        <span class="n">kwargs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">BoundingBoxes</span><span class="p">:</span>
+        <span class="c1"># If there are BoundingBoxes instances in the output, their metadata got lost when we called</span>
+        <span class="c1"># super().__torch_function__. We need to restore the metadata somehow, so we choose to take</span>
+        <span class="c1"># the metadata from the first bbox in the parameters.</span>
+        <span class="c1"># This should be what we want in most cases. When it&#39;s not, it&#39;s probably a mis-use anyway, e.g.</span>
+        <span class="c1"># something like some_xyxy_bbox + some_xywh_bbox; we don&#39;t guard against those cases.</span>
+        <span class="n">flat_params</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">tree_flatten</span><span class="p">(</span><span class="n">args</span> <span class="o">+</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">kwargs</span><span class="o">.</span><span class="n">values</span><span class="p">())</span> <span class="k">if</span> <span class="n">kwargs</span> <span class="k">else</span> <span class="p">()))</span>  <span class="c1"># type: ignore[operator]</span>
+        <span class="n">first_bbox_from_args</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">x</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">flat_params</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">BoundingBoxes</span><span class="p">))</span>
+        <span class="nb">format</span><span class="p">,</span> <span class="n">canvas_size</span> <span class="o">=</span> <span class="n">first_bbox_from_args</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">first_bbox_from_args</span><span class="o">.</span><span class="n">canvas_size</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">BoundingBoxes</span><span class="p">):</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">BoundingBoxes</span><span class="o">.</span><span class="n">_wrap</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">check_dims</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="nb">type</span><span class="p">(</span><span class="n">output</span><span class="p">)(</span>
+                <span class="n">BoundingBoxes</span><span class="o">.</span><span class="n">_wrap</span><span class="p">(</span><span class="n">part</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span> <span class="n">check_dims</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> <span class="k">for</span> <span class="n">part</span> <span class="ow">in</span> <span class="n">output</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="n">output</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">tensor_contents</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_repr</span><span class="p">(</span><span class="nb">format</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">format</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">canvas_size</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html b/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
new file mode 100644
index 00000000000..995ef0e5832
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
@@ -0,0 +1,1323 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors._dataset_wrapper &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+          <li><a href="../tv_tensors.html">torchvision.tv_tensors</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors._dataset_wrapper</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors._dataset_wrapper</h1><div class="highlight"><pre>
+<span></span><span class="c1"># type: ignore</span>
+
+<span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">annotations</span>
+
+<span class="kn">import</span> <span class="nn">collections.abc</span>
+
+<span class="kn">import</span> <span class="nn">contextlib</span>
+<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">defaultdict</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">datasets</span><span class="p">,</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;wrap_dataset_for_transforms_v2&quot;</span><span class="p">]</span>
+
+
+<div class="viewcode-block" id="wrap_dataset_for_transforms_v2"><a class="viewcode-back" href="../../../generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html#torchvision.datasets.wrap_dataset_for_transforms_v2">[docs]</a><span class="k">def</span> <span class="nf">wrap_dataset_for_transforms_v2</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Wrap a ``torchvision.dataset`` for usage with :mod:`torchvision.transforms.v2`.</span>
+
+<span class="sd">    .. v2betastatus:: wrap_dataset_for_transforms_v2 function</span>
+
+<span class="sd">    Example:</span>
+<span class="sd">        &gt;&gt;&gt; dataset = torchvision.datasets.CocoDetection(...)</span>
+<span class="sd">        &gt;&gt;&gt; dataset = wrap_dataset_for_transforms_v2(dataset)</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">       For now, only the most popular datasets are supported. Furthermore, the wrapper only supports dataset</span>
+<span class="sd">       configurations that are fully supported by ``torchvision.transforms.v2``. If you encounter an error prompting you</span>
+<span class="sd">       to raise an issue to ``torchvision`` for a dataset or configuration that you need, please do so.</span>
+
+<span class="sd">    The dataset samples are wrapped according to the description below.</span>
+
+<span class="sd">    Special cases:</span>
+
+<span class="sd">        * :class:`~torchvision.datasets.CocoDetection`: Instead of returning the target as list of dicts, the wrapper</span>
+<span class="sd">          returns a dict of lists. In addition, the key-value-pairs ``&quot;boxes&quot;`` (in ``XYXY`` coordinate format),</span>
+<span class="sd">          ``&quot;masks&quot;`` and ``&quot;labels&quot;`` are added and wrap the data in the corresponding ``torchvision.tv_tensors``.</span>
+<span class="sd">          The original keys are preserved. If ``target_keys`` is omitted, returns only the values for the</span>
+<span class="sd">          ``&quot;image_id&quot;``, ``&quot;boxes&quot;``, and ``&quot;labels&quot;``.</span>
+<span class="sd">        * :class:`~torchvision.datasets.VOCDetection`: The key-value-pairs ``&quot;boxes&quot;`` and ``&quot;labels&quot;`` are added to</span>
+<span class="sd">          the target and wrap the data in the corresponding ``torchvision.tv_tensors``. The original keys are</span>
+<span class="sd">          preserved. If ``target_keys`` is omitted, returns only the values for the ``&quot;boxes&quot;`` and ``&quot;labels&quot;``.</span>
+<span class="sd">        * :class:`~torchvision.datasets.CelebA`: The target for ``target_type=&quot;bbox&quot;`` is converted to the ``XYXY``</span>
+<span class="sd">          coordinate format and wrapped into a :class:`~torchvision.tv_tensors.BoundingBoxes` tv_tensor.</span>
+<span class="sd">        * :class:`~torchvision.datasets.Kitti`: Instead returning the target as list of dicts, the wrapper returns a</span>
+<span class="sd">          dict of lists. In addition, the key-value-pairs ``&quot;boxes&quot;`` and ``&quot;labels&quot;`` are added and wrap the data</span>
+<span class="sd">          in the corresponding ``torchvision.tv_tensors``. The original keys are preserved. If ``target_keys`` is</span>
+<span class="sd">          omitted, returns only the values for the ``&quot;boxes&quot;`` and ``&quot;labels&quot;``.</span>
+<span class="sd">        * :class:`~torchvision.datasets.OxfordIIITPet`: The target for ``target_type=&quot;segmentation&quot;`` is wrapped into a</span>
+<span class="sd">          :class:`~torchvision.tv_tensors.Mask` tv_tensor.</span>
+<span class="sd">        * :class:`~torchvision.datasets.Cityscapes`: The target for ``target_type=&quot;semantic&quot;`` is wrapped into a</span>
+<span class="sd">          :class:`~torchvision.tv_tensors.Mask` tv_tensor. The target for ``target_type=&quot;instance&quot;`` is *replaced* by</span>
+<span class="sd">          a dictionary with the key-value-pairs ``&quot;masks&quot;`` (as :class:`~torchvision.tv_tensors.Mask` tv_tensor) and</span>
+<span class="sd">          ``&quot;labels&quot;``.</span>
+<span class="sd">        * :class:`~torchvision.datasets.WIDERFace`: The value for key ``&quot;bbox&quot;`` in the target is converted to ``XYXY``</span>
+<span class="sd">          coordinate format and wrapped into a :class:`~torchvision.tv_tensors.BoundingBoxes` tv_tensor.</span>
+
+<span class="sd">    Image classification datasets</span>
+
+<span class="sd">        This wrapper is a no-op for image classification datasets, since they were already fully supported by</span>
+<span class="sd">        :mod:`torchvision.transforms` and thus no change is needed for :mod:`torchvision.transforms.v2`.</span>
+
+<span class="sd">    Segmentation datasets</span>
+
+<span class="sd">        Segmentation datasets, e.g. :class:`~torchvision.datasets.VOCSegmentation`, return a two-tuple of</span>
+<span class="sd">        :class:`PIL.Image.Image`&#39;s. This wrapper leaves the image as is (first item), while wrapping the</span>
+<span class="sd">        segmentation mask into a :class:`~torchvision.tv_tensors.Mask` (second item).</span>
+
+<span class="sd">    Video classification datasets</span>
+
+<span class="sd">        Video classification datasets, e.g. :class:`~torchvision.datasets.Kinetics`, return a three-tuple containing a</span>
+<span class="sd">        :class:`torch.Tensor` for the video and audio and a :class:`int` as label. This wrapper wraps the video into a</span>
+<span class="sd">        :class:`~torchvision.tv_tensors.Video` while leaving the other items as is.</span>
+
+<span class="sd">        .. note::</span>
+
+<span class="sd">            Only datasets constructed with ``output_format=&quot;TCHW&quot;`` are supported, since the alternative</span>
+<span class="sd">            ``output_format=&quot;THWC&quot;`` is not supported by :mod:`torchvision.transforms.v2`.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        dataset: the dataset instance to wrap for compatibility with transforms v2.</span>
+<span class="sd">        target_keys: Target keys to return in case the target is a dictionary. If ``None`` (default), selected keys are</span>
+<span class="sd">            specific to the dataset. If ``&quot;all&quot;``, returns the full target. Can also be a collection of strings for</span>
+<span class="sd">            fine grained access. Currently only supported for :class:`~torchvision.datasets.CocoDetection`,</span>
+<span class="sd">            :class:`~torchvision.datasets.VOCDetection`, :class:`~torchvision.datasets.Kitti`, and</span>
+<span class="sd">            :class:`~torchvision.datasets.WIDERFace`. See above for details.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span>
+        <span class="n">target_keys</span> <span class="ow">is</span> <span class="kc">None</span>
+        <span class="ow">or</span> <span class="n">target_keys</span> <span class="o">==</span> <span class="s2">&quot;all&quot;</span>
+        <span class="ow">or</span> <span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">target_keys</span><span class="p">,</span> <span class="n">collections</span><span class="o">.</span><span class="n">abc</span><span class="o">.</span><span class="n">Collection</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">all</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">key</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">))</span>
+    <span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;`target_keys` can be None, &#39;all&#39;, or a collection of strings denoting the keys to be returned, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;but got </span><span class="si">{</span><span class="n">target_keys</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+    <span class="c1"># Imagine we have isinstance(dataset, datasets.ImageNet). This will create a new class with the name</span>
+    <span class="c1"># &quot;WrappedImageNet&quot; at runtime that doubly inherits from VisionDatasetTVTensorWrapper (see below) as well as the</span>
+    <span class="c1"># original ImageNet class. This allows the user to do regular isinstance(wrapped_dataset, datasets.ImageNet) checks,</span>
+    <span class="c1"># while we can still inject everything that we need.</span>
+    <span class="n">wrapped_dataset_cls</span> <span class="o">=</span> <span class="nb">type</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Wrapped</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="p">(</span><span class="n">VisionDatasetTVTensorWrapper</span><span class="p">,</span> <span class="nb">type</span><span class="p">(</span><span class="n">dataset</span><span class="p">)),</span> <span class="p">{})</span>
+    <span class="c1"># Since VisionDatasetTVTensorWrapper comes before ImageNet in the MRO, calling the class hits</span>
+    <span class="c1"># VisionDatasetTVTensorWrapper.__init__ first. Since we are never doing super().__init__(...), the constructor of</span>
+    <span class="c1"># ImageNet is never hit. That is by design, since we don&#39;t want to create the dataset instance again, but rather</span>
+    <span class="c1"># have the existing instance as attribute on the new object.</span>
+    <span class="k">return</span> <span class="n">wrapped_dataset_cls</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">)</span></div>
+
+
+<span class="k">class</span> <span class="nc">WrapperFactories</span><span class="p">(</span><span class="nb">dict</span><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">register</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset_cls</span><span class="p">):</span>
+        <span class="k">def</span> <span class="nf">decorator</span><span class="p">(</span><span class="n">wrapper_factory</span><span class="p">):</span>
+            <span class="bp">self</span><span class="p">[</span><span class="n">dataset_cls</span><span class="p">]</span> <span class="o">=</span> <span class="n">wrapper_factory</span>
+            <span class="k">return</span> <span class="n">wrapper_factory</span>
+
+        <span class="k">return</span> <span class="n">decorator</span>
+
+
+<span class="c1"># We need this two-stage design, i.e. a wrapper factory producing the actual wrapper, since some wrappers depend on the</span>
+<span class="c1"># dataset instance rather than just the class, since they require the user defined instance attributes. Thus, we can</span>
+<span class="c1"># provide a wrapping from the dataset class to the factory here, but can only instantiate the wrapper at runtime when</span>
+<span class="c1"># we have access to the dataset instance.</span>
+<span class="n">WRAPPER_FACTORIES</span> <span class="o">=</span> <span class="n">WrapperFactories</span><span class="p">()</span>
+
+
+<span class="k">class</span> <span class="nc">VisionDatasetTVTensorWrapper</span><span class="p">:</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+        <span class="n">dataset_cls</span> <span class="o">=</span> <span class="nb">type</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">datasets</span><span class="o">.</span><span class="n">VisionDataset</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;This wrapper is meant for subclasses of `torchvision.datasets.VisionDataset`, &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;but got a &#39;</span><span class="si">{</span><span class="n">dataset_cls</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">&#39; instead.</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;For an example of how to perform the wrapping for custom datasets, see</span><span class="se">\n\n</span><span class="s2">&quot;</span>
+                <span class="s2">&quot;https://pytorch.org/vision/main/auto_examples/plot_tv_tensors.html#do-i-have-to-wrap-the-output-of-the-datasets-myself&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">for</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="n">dataset_cls</span><span class="o">.</span><span class="n">mro</span><span class="p">():</span>
+            <span class="k">if</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="n">WRAPPER_FACTORIES</span><span class="p">:</span>
+                <span class="n">wrapper_factory</span> <span class="o">=</span> <span class="n">WRAPPER_FACTORIES</span><span class="p">[</span><span class="bp">cls</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">target_keys</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">cls</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">{</span>
+                    <span class="n">datasets</span><span class="o">.</span><span class="n">CocoDetection</span><span class="p">,</span>
+                    <span class="n">datasets</span><span class="o">.</span><span class="n">VOCDetection</span><span class="p">,</span>
+                    <span class="n">datasets</span><span class="o">.</span><span class="n">Kitti</span><span class="p">,</span>
+                    <span class="n">datasets</span><span class="o">.</span><span class="n">WIDERFace</span><span class="p">,</span>
+                <span class="p">}:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;`target_keys` is currently only supported for `CocoDetection`, `VOCDetection`, `Kitti`, &quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;and `WIDERFace`, but got </span><span class="si">{</span><span class="bp">cls</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">.&quot;</span>
+                    <span class="p">)</span>
+                <span class="k">break</span>
+            <span class="k">elif</span> <span class="bp">cls</span> <span class="ow">is</span> <span class="n">datasets</span><span class="o">.</span><span class="n">VisionDataset</span><span class="p">:</span>
+                <span class="c1"># TODO: If we have documentation on how to do that, put a link in the error message.</span>
+                <span class="n">msg</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;No wrapper exists for dataset class </span><span class="si">{</span><span class="n">dataset_cls</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">. Please wrap the output yourself.&quot;</span>
+                <span class="k">if</span> <span class="n">dataset_cls</span> <span class="ow">in</span> <span class="n">datasets</span><span class="o">.</span><span class="vm">__dict__</span><span class="o">.</span><span class="n">values</span><span class="p">():</span>
+                    <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">msg</span><span class="si">}</span><span class="s2"> If an automated wrapper for this dataset would be useful for you, &quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;please open an issue at https://github.com/pytorch/vision/issues.&quot;</span>
+                    <span class="p">)</span>
+                <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_dataset</span> <span class="o">=</span> <span class="n">dataset</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_target_keys</span> <span class="o">=</span> <span class="n">target_keys</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_wrapper</span> <span class="o">=</span> <span class="n">wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">)</span>
+
+        <span class="c1"># We need to disable the transforms on the dataset here to be able to inject the wrapping before we apply them.</span>
+        <span class="c1"># Although internally, `datasets.VisionDataset` merges `transform` and `target_transform` into the joint</span>
+        <span class="c1"># `transforms`</span>
+        <span class="c1"># https://github.com/pytorch/vision/blob/135a0f9ea9841b6324b4fe8974e2543cbb95709a/torchvision/datasets/vision.py#L52-L54</span>
+        <span class="c1"># some (if not most) datasets still use `transform` and `target_transform` individually. Thus, we need to</span>
+        <span class="c1"># disable all three here to be able to extract the untransformed sample to wrap.</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span><span class="p">,</span> <span class="n">dataset</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">transform</span><span class="p">,</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">target_transform</span><span class="p">,</span> <span class="n">dataset</span><span class="o">.</span><span class="n">target_transform</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">target_transform</span><span class="p">,</span> <span class="kc">None</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">,</span> <span class="n">dataset</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">transforms</span><span class="p">,</span> <span class="kc">None</span>
+
+    <span class="k">def</span> <span class="fm">__getattr__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">item</span><span class="p">):</span>
+        <span class="k">with</span> <span class="n">contextlib</span><span class="o">.</span><span class="n">suppress</span><span class="p">(</span><span class="ne">AttributeError</span><span class="p">):</span>
+            <span class="k">return</span> <span class="nb">object</span><span class="o">.</span><span class="fm">__getattribute__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">item</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_dataset</span><span class="p">,</span> <span class="n">item</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">):</span>
+        <span class="c1"># This gets us the raw sample since we disabled the transforms for the underlying dataset in the constructor</span>
+        <span class="c1"># of this class</span>
+        <span class="n">sample</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_dataset</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+
+        <span class="n">sample</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">)</span>
+
+        <span class="c1"># Regardless of whether the user has supplied the transforms individually (`transform` and `target_transform`)</span>
+        <span class="c1"># or joint (`transforms`), we can access the full functionality through `transforms`</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">sample</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="o">*</span><span class="n">sample</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">sample</span>
+
+    <span class="k">def</span> <span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_dataset</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">__reduce__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">wrap_dataset_for_transforms_v2</span><span class="p">,</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_dataset</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_target_keys</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">raise_not_supported</span><span class="p">(</span><span class="n">description</span><span class="p">):</span>
+    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+        <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">description</span><span class="si">}</span><span class="s2"> is currently not supported by this wrapper. &quot;</span>
+        <span class="sa">f</span><span class="s2">&quot;If this would be helpful for you, please open an issue at https://github.com/pytorch/vision/issues.&quot;</span>
+    <span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">identity</span><span class="p">(</span><span class="n">item</span><span class="p">):</span>
+    <span class="k">return</span> <span class="n">item</span>
+
+
+<span class="k">def</span> <span class="nf">identity_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">sample</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="k">def</span> <span class="nf">pil_image_to_mask</span><span class="p">(</span><span class="n">pil_image</span><span class="p">):</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">(</span><span class="n">pil_image</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">parse_target_keys</span><span class="p">(</span><span class="n">target_keys</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">available</span><span class="p">,</span> <span class="n">default</span><span class="p">):</span>
+    <span class="k">if</span> <span class="n">target_keys</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">target_keys</span> <span class="o">=</span> <span class="n">default</span>
+    <span class="k">if</span> <span class="n">target_keys</span> <span class="o">==</span> <span class="s2">&quot;all&quot;</span><span class="p">:</span>
+        <span class="n">target_keys</span> <span class="o">=</span> <span class="n">available</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">target_keys</span> <span class="o">=</span> <span class="nb">set</span><span class="p">(</span><span class="n">target_keys</span><span class="p">)</span>
+        <span class="n">extra</span> <span class="o">=</span> <span class="n">target_keys</span> <span class="o">-</span> <span class="n">available</span>
+        <span class="k">if</span> <span class="n">extra</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Target keys </span><span class="si">{</span><span class="nb">sorted</span><span class="p">(</span><span class="n">extra</span><span class="p">)</span><span class="si">}</span><span class="s2"> are not available&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">target_keys</span>
+
+
+<span class="k">def</span> <span class="nf">list_of_dicts_to_dict_of_lists</span><span class="p">(</span><span class="n">list_of_dicts</span><span class="p">):</span>
+    <span class="n">dict_of_lists</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="nb">list</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">dct</span> <span class="ow">in</span> <span class="n">list_of_dicts</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="n">dct</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="n">dict_of_lists</span><span class="p">[</span><span class="n">key</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">value</span><span class="p">)</span>
+    <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">dict_of_lists</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">wrap_target_by_type</span><span class="p">(</span><span class="n">target</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">target_types</span><span class="p">,</span> <span class="n">type_wrappers</span><span class="p">):</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">target</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="p">[</span><span class="n">target</span><span class="p">]</span>
+
+    <span class="n">wrapped_target</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span>
+        <span class="n">type_wrappers</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">target_type</span><span class="p">,</span> <span class="n">identity</span><span class="p">)(</span><span class="n">item</span><span class="p">)</span> <span class="k">for</span> <span class="n">target_type</span><span class="p">,</span> <span class="n">item</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">target_types</span><span class="p">,</span> <span class="n">target</span><span class="p">)</span>
+    <span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">wrapped_target</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">wrapped_target</span> <span class="o">=</span> <span class="n">wrapped_target</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="k">return</span> <span class="n">wrapped_target</span>
+
+
+<span class="k">def</span> <span class="nf">classification_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">return</span> <span class="n">identity_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">)</span>
+
+
+<span class="k">for</span> <span class="n">dataset_cls</span> <span class="ow">in</span> <span class="p">[</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">Caltech256</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">CIFAR10</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">CIFAR100</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">ImageNet</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">MNIST</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">FashionMNIST</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">GTSRB</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">DatasetFolder</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">ImageFolder</span><span class="p">,</span>
+<span class="p">]:</span>
+    <span class="n">WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">dataset_cls</span><span class="p">)(</span><span class="n">classification_wrapper_factory</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">segmentation_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">mask</span> <span class="o">=</span> <span class="n">sample</span>
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">pil_image_to_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="k">for</span> <span class="n">dataset_cls</span> <span class="ow">in</span> <span class="p">[</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">VOCSegmentation</span><span class="p">,</span>
+<span class="p">]:</span>
+    <span class="n">WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">dataset_cls</span><span class="p">)(</span><span class="n">segmentation_wrapper_factory</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">video_classification_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">video_clips</span><span class="o">.</span><span class="n">output_format</span> <span class="o">==</span> <span class="s2">&quot;THWC&quot;</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2"> with `output_format=&#39;THWC&#39;` is not supported by this wrapper, &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;since it is not compatible with the transformations. Please use `output_format=&#39;TCHW&#39;` instead.&quot;</span>
+        <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">label</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="n">video</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">video</span><span class="p">,</span> <span class="n">audio</span><span class="p">,</span> <span class="n">label</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="k">for</span> <span class="n">dataset_cls</span> <span class="ow">in</span> <span class="p">[</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">HMDB51</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">Kinetics</span><span class="p">,</span>
+    <span class="n">datasets</span><span class="o">.</span><span class="n">UCF101</span><span class="p">,</span>
+<span class="p">]:</span>
+    <span class="n">WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">dataset_cls</span><span class="p">)(</span><span class="n">video_classification_wrapper_factory</span><span class="p">)</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">Caltech101</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">caltech101_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">if</span> <span class="s2">&quot;annotation&quot;</span> <span class="ow">in</span> <span class="n">dataset</span><span class="o">.</span><span class="n">target_type</span><span class="p">:</span>
+        <span class="n">raise_not_supported</span><span class="p">(</span><span class="s2">&quot;Caltech101 dataset with `target_type=[&#39;annotation&#39;, ...]`&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">classification_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">)</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">CocoDetection</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">coco_dectection_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="n">target_keys</span> <span class="o">=</span> <span class="n">parse_target_keys</span><span class="p">(</span>
+        <span class="n">target_keys</span><span class="p">,</span>
+        <span class="n">available</span><span class="o">=</span><span class="p">{</span>
+            <span class="c1"># native</span>
+            <span class="s2">&quot;segmentation&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;area&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;iscrowd&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;image_id&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;bbox&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;category_id&quot;</span><span class="p">,</span>
+            <span class="c1"># added by the wrapper</span>
+            <span class="s2">&quot;boxes&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;masks&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;labels&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+        <span class="n">default</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;image_id&quot;</span><span class="p">,</span> <span class="s2">&quot;boxes&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">segmentation_to_mask</span><span class="p">(</span><span class="n">segmentation</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">canvas_size</span><span class="p">):</span>
+        <span class="kn">from</span> <span class="nn">pycocotools</span> <span class="kn">import</span> <span class="n">mask</span>
+
+        <span class="n">segmentation</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">mask</span><span class="o">.</span><span class="n">frPyObjects</span><span class="p">(</span><span class="n">segmentation</span><span class="p">,</span> <span class="o">*</span><span class="n">canvas_size</span><span class="p">)</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">segmentation</span><span class="p">,</span> <span class="nb">dict</span><span class="p">)</span>
+            <span class="k">else</span> <span class="n">mask</span><span class="o">.</span><span class="n">merge</span><span class="p">(</span><span class="n">mask</span><span class="o">.</span><span class="n">frPyObjects</span><span class="p">(</span><span class="n">segmentation</span><span class="p">,</span> <span class="o">*</span><span class="n">canvas_size</span><span class="p">))</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">mask</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">segmentation</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image_id</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">ids</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span>
+
+        <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">target</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="nb">dict</span><span class="p">(</span><span class="n">image_id</span><span class="o">=</span><span class="n">image_id</span><span class="p">)</span>
+
+        <span class="n">canvas_size</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">F</span><span class="o">.</span><span class="n">get_size</span><span class="p">(</span><span class="n">image</span><span class="p">))</span>
+
+        <span class="n">batched_target</span> <span class="o">=</span> <span class="n">list_of_dicts_to_dict_of_lists</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="k">if</span> <span class="s2">&quot;image_id&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;image_id&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">image_id</span>
+
+        <span class="k">if</span> <span class="s2">&quot;boxes&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_bounding_box_format</span><span class="p">(</span>
+                <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">(</span>
+                    <span class="n">batched_target</span><span class="p">[</span><span class="s2">&quot;bbox&quot;</span><span class="p">],</span>
+                    <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYWH</span><span class="p">,</span>
+                    <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">,</span>
+                <span class="p">),</span>
+                <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="s2">&quot;masks&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;masks&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">(</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span>
+                    <span class="p">[</span>
+                        <span class="n">segmentation_to_mask</span><span class="p">(</span><span class="n">segmentation</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="n">canvas_size</span><span class="p">)</span>
+                        <span class="k">for</span> <span class="n">segmentation</span> <span class="ow">in</span> <span class="n">batched_target</span><span class="p">[</span><span class="s2">&quot;segmentation&quot;</span><span class="p">]</span>
+                    <span class="p">]</span>
+                <span class="p">),</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="s2">&quot;labels&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">batched_target</span><span class="p">[</span><span class="s2">&quot;category_id&quot;</span><span class="p">])</span>
+
+        <span class="k">for</span> <span class="n">target_key</span> <span class="ow">in</span> <span class="n">target_keys</span> <span class="o">-</span> <span class="p">{</span><span class="s2">&quot;image_id&quot;</span><span class="p">,</span> <span class="s2">&quot;boxes&quot;</span><span class="p">,</span> <span class="s2">&quot;masks&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">}:</span>
+            <span class="n">target</span><span class="p">[</span><span class="n">target_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">batched_target</span><span class="p">[</span><span class="n">target_key</span><span class="p">]</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="n">WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">CocoCaptions</span><span class="p">)(</span><span class="n">identity_wrapper_factory</span><span class="p">)</span>
+
+
+<span class="n">VOC_DETECTION_CATEGORIES</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;__background__&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;aeroplane&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;bicycle&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;bird&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;boat&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;bottle&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;bus&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;car&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;cat&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;chair&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;cow&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;diningtable&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;dog&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;horse&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;motorbike&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;person&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;pottedplant&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;sheep&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;sofa&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;tvmonitor&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+<span class="n">VOC_DETECTION_CATEGORY_TO_IDX</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="n">VOC_DETECTION_CATEGORIES</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">VOC_DETECTION_CATEGORIES</span><span class="p">))))</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">VOCDetection</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">voc_detection_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="n">target_keys</span> <span class="o">=</span> <span class="n">parse_target_keys</span><span class="p">(</span>
+        <span class="n">target_keys</span><span class="p">,</span>
+        <span class="n">available</span><span class="o">=</span><span class="p">{</span>
+            <span class="c1"># native</span>
+            <span class="s2">&quot;annotation&quot;</span><span class="p">,</span>
+            <span class="c1"># added by the wrapper</span>
+            <span class="s2">&quot;boxes&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;labels&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+        <span class="n">default</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;boxes&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="n">batched_instances</span> <span class="o">=</span> <span class="n">list_of_dicts_to_dict_of_lists</span><span class="p">(</span><span class="n">target</span><span class="p">[</span><span class="s2">&quot;annotation&quot;</span><span class="p">][</span><span class="s2">&quot;object&quot;</span><span class="p">])</span>
+
+        <span class="k">if</span> <span class="s2">&quot;annotation&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="k">if</span> <span class="s2">&quot;boxes&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">(</span>
+                <span class="p">[</span>
+                    <span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">bndbox</span><span class="p">[</span><span class="n">part</span><span class="p">])</span> <span class="k">for</span> <span class="n">part</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;xmin&quot;</span><span class="p">,</span> <span class="s2">&quot;ymin&quot;</span><span class="p">,</span> <span class="s2">&quot;xmax&quot;</span><span class="p">,</span> <span class="s2">&quot;ymax&quot;</span><span class="p">)]</span>
+                    <span class="k">for</span> <span class="n">bndbox</span> <span class="ow">in</span> <span class="n">batched_instances</span><span class="p">[</span><span class="s2">&quot;bndbox&quot;</span><span class="p">]</span>
+                <span class="p">],</span>
+                <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+                <span class="n">canvas_size</span><span class="o">=</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">width</span><span class="p">),</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="s2">&quot;labels&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+                <span class="p">[</span><span class="n">VOC_DETECTION_CATEGORY_TO_IDX</span><span class="p">[</span><span class="n">category</span><span class="p">]</span> <span class="k">for</span> <span class="n">category</span> <span class="ow">in</span> <span class="n">batched_instances</span><span class="p">[</span><span class="s2">&quot;name&quot;</span><span class="p">]]</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">SBDataset</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">sbd_wrapper</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">mode</span> <span class="o">==</span> <span class="s2">&quot;boundaries&quot;</span><span class="p">:</span>
+        <span class="n">raise_not_supported</span><span class="p">(</span><span class="s2">&quot;SBDataset with mode=&#39;boundaries&#39;&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">segmentation_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">)</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">CelebA</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">celeba_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">if</span> <span class="nb">any</span><span class="p">(</span><span class="n">target_type</span> <span class="ow">in</span> <span class="n">dataset</span><span class="o">.</span><span class="n">target_type</span> <span class="k">for</span> <span class="n">target_type</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;attr&quot;</span><span class="p">,</span> <span class="s2">&quot;landmarks&quot;</span><span class="p">]):</span>
+        <span class="n">raise_not_supported</span><span class="p">(</span><span class="s2">&quot;`CelebA` dataset with `target_type=[&#39;attr&#39;, &#39;landmarks&#39;, ...]`&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="n">target</span> <span class="o">=</span> <span class="n">wrap_target_by_type</span><span class="p">(</span>
+            <span class="n">target</span><span class="p">,</span>
+            <span class="n">target_types</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">target_type</span><span class="p">,</span>
+            <span class="n">type_wrappers</span><span class="o">=</span><span class="p">{</span>
+                <span class="s2">&quot;bbox&quot;</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">item</span><span class="p">:</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_bounding_box_format</span><span class="p">(</span>
+                    <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">(</span>
+                        <span class="n">item</span><span class="p">,</span>
+                        <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYWH</span><span class="p">,</span>
+                        <span class="n">canvas_size</span><span class="o">=</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">width</span><span class="p">),</span>
+                    <span class="p">),</span>
+                    <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+                <span class="p">),</span>
+            <span class="p">},</span>
+        <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="n">KITTI_CATEGORIES</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Car&quot;</span><span class="p">,</span> <span class="s2">&quot;Van&quot;</span><span class="p">,</span> <span class="s2">&quot;Truck&quot;</span><span class="p">,</span> <span class="s2">&quot;Pedestrian&quot;</span><span class="p">,</span> <span class="s2">&quot;Person_sitting&quot;</span><span class="p">,</span> <span class="s2">&quot;Cyclist&quot;</span><span class="p">,</span> <span class="s2">&quot;Tram&quot;</span><span class="p">,</span> <span class="s2">&quot;Misc&quot;</span><span class="p">,</span> <span class="s2">&quot;DontCare&quot;</span><span class="p">]</span>
+<span class="n">KITTI_CATEGORY_TO_IDX</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="n">KITTI_CATEGORIES</span><span class="p">,</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">KITTI_CATEGORIES</span><span class="p">))))</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">Kitti</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">kitti_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="n">target_keys</span> <span class="o">=</span> <span class="n">parse_target_keys</span><span class="p">(</span>
+        <span class="n">target_keys</span><span class="p">,</span>
+        <span class="n">available</span><span class="o">=</span><span class="p">{</span>
+            <span class="c1"># native</span>
+            <span class="s2">&quot;type&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;truncated&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;occluded&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;alpha&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;bbox&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;dimensions&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;location&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;rotation_y&quot;</span><span class="p">,</span>
+            <span class="c1"># added by the wrapper</span>
+            <span class="s2">&quot;boxes&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;labels&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+        <span class="n">default</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;boxes&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">},</span>
+    <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="k">if</span> <span class="n">target</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+        <span class="n">batched_target</span> <span class="o">=</span> <span class="n">list_of_dicts_to_dict_of_lists</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+        <span class="n">target</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="k">if</span> <span class="s2">&quot;boxes&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">(</span>
+                <span class="n">batched_target</span><span class="p">[</span><span class="s2">&quot;bbox&quot;</span><span class="p">],</span>
+                <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+                <span class="n">canvas_size</span><span class="o">=</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">width</span><span class="p">),</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="s2">&quot;labels&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">KITTI_CATEGORY_TO_IDX</span><span class="p">[</span><span class="n">category</span><span class="p">]</span> <span class="k">for</span> <span class="n">category</span> <span class="ow">in</span> <span class="n">batched_target</span><span class="p">[</span><span class="s2">&quot;type&quot;</span><span class="p">]])</span>
+
+        <span class="k">for</span> <span class="n">target_key</span> <span class="ow">in</span> <span class="n">target_keys</span> <span class="o">-</span> <span class="p">{</span><span class="s2">&quot;boxes&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">}:</span>
+            <span class="n">target</span><span class="p">[</span><span class="n">target_key</span><span class="p">]</span> <span class="o">=</span> <span class="n">batched_target</span><span class="p">[</span><span class="n">target_key</span><span class="p">]</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">OxfordIIITPet</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">oxford_iiit_pet_wrapper_factor</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="k">if</span> <span class="n">target</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">target</span> <span class="o">=</span> <span class="n">wrap_target_by_type</span><span class="p">(</span>
+                <span class="n">target</span><span class="p">,</span>
+                <span class="n">target_types</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">_target_types</span><span class="p">,</span>
+                <span class="n">type_wrappers</span><span class="o">=</span><span class="p">{</span>
+                    <span class="s2">&quot;segmentation&quot;</span><span class="p">:</span> <span class="n">pil_image_to_mask</span><span class="p">,</span>
+                <span class="p">},</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">Cityscapes</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">cityscapes_wrapper_factory</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="k">if</span> <span class="nb">any</span><span class="p">(</span><span class="n">target_type</span> <span class="ow">in</span> <span class="n">dataset</span><span class="o">.</span><span class="n">target_type</span> <span class="k">for</span> <span class="n">target_type</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;polygon&quot;</span><span class="p">,</span> <span class="s2">&quot;color&quot;</span><span class="p">]):</span>
+        <span class="n">raise_not_supported</span><span class="p">(</span><span class="s2">&quot;`Cityscapes` dataset with `target_type=[&#39;polygon&#39;, &#39;color&#39;, ...]`&quot;</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">instance_segmentation_wrapper</span><span class="p">(</span><span class="n">mask</span><span class="p">):</span>
+        <span class="c1"># See https://github.com/mcordts/cityscapesScripts/blob/8da5dd00c9069058ccc134654116aac52d4f6fa2/cityscapesscripts/preparation/json2instanceImg.py#L7-L21</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="n">pil_image_to_mask</span><span class="p">(</span><span class="n">mask</span><span class="p">)</span>
+        <span class="n">masks</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">labels</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="nb">id</span> <span class="ow">in</span> <span class="n">data</span><span class="o">.</span><span class="n">unique</span><span class="p">():</span>
+            <span class="n">masks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">data</span> <span class="o">==</span> <span class="nb">id</span><span class="p">)</span>
+            <span class="n">label</span> <span class="o">=</span> <span class="nb">id</span>
+            <span class="k">if</span> <span class="n">label</span> <span class="o">&gt;=</span> <span class="mi">1_000</span><span class="p">:</span>
+                <span class="n">label</span> <span class="o">//=</span> <span class="mi">1_000</span>
+            <span class="n">labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
+        <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">masks</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Mask</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">masks</span><span class="p">)),</span> <span class="n">labels</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">labels</span><span class="p">))</span>
+
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="n">target</span> <span class="o">=</span> <span class="n">wrap_target_by_type</span><span class="p">(</span>
+            <span class="n">target</span><span class="p">,</span>
+            <span class="n">target_types</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">target_type</span><span class="p">,</span>
+            <span class="n">type_wrappers</span><span class="o">=</span><span class="p">{</span>
+                <span class="s2">&quot;instance&quot;</span><span class="p">:</span> <span class="n">instance_segmentation_wrapper</span><span class="p">,</span>
+                <span class="s2">&quot;semantic&quot;</span><span class="p">:</span> <span class="n">pil_image_to_mask</span><span class="p">,</span>
+            <span class="p">},</span>
+        <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+
+
+<span class="nd">@WRAPPER_FACTORIES</span><span class="o">.</span><span class="n">register</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">WIDERFace</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">widerface_wrapper</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="p">):</span>
+    <span class="n">target_keys</span> <span class="o">=</span> <span class="n">parse_target_keys</span><span class="p">(</span>
+        <span class="n">target_keys</span><span class="p">,</span>
+        <span class="n">available</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;bbox&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;blur&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;expression&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;illumination&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;occlusion&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;pose&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;invalid&quot;</span><span class="p">,</span>
+        <span class="p">},</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;all&quot;</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">wrapper</span><span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="n">sample</span><span class="p">):</span>
+        <span class="n">image</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">sample</span>
+
+        <span class="k">if</span> <span class="n">target</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+        <span class="n">target</span> <span class="o">=</span> <span class="p">{</span><span class="n">key</span><span class="p">:</span> <span class="n">target</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">}</span>
+
+        <span class="k">if</span> <span class="s2">&quot;bbox&quot;</span> <span class="ow">in</span> <span class="n">target_keys</span><span class="p">:</span>
+            <span class="n">target</span><span class="p">[</span><span class="s2">&quot;bbox&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_bounding_box_format</span><span class="p">(</span>
+                <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">(</span>
+                    <span class="n">target</span><span class="p">[</span><span class="s2">&quot;bbox&quot;</span><span class="p">],</span> <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYWH</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">height</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">width</span><span class="p">)</span>
+                <span class="p">),</span>
+                <span class="n">new_format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image</span><span class="p">,</span> <span class="n">target</span>
+
+    <span class="k">return</span> <span class="n">wrapper</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors/_image.html b/0.16/_modules/torchvision/tv_tensors/_image.html
new file mode 100644
index 00000000000..23238465a31
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors/_image.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors._image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+          <li><a href="../tv_tensors.html">torchvision.tv_tensors</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors._image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors._image</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">annotations</span>
+
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">._tv_tensor</span> <span class="kn">import</span> <span class="n">TVTensor</span>
+
+
+<div class="viewcode-block" id="Image"><a class="viewcode-back" href="../../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image">[docs]</a><span class="k">class</span> <span class="nc">Image</span><span class="p">(</span><span class="n">TVTensor</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] :class:`torch.Tensor` subclass for images.</span>
+
+<span class="sd">    .. note::</span>
+
+<span class="sd">        In the :ref:`transforms &lt;transforms&gt;`, ``Image`` instances are largely</span>
+<span class="sd">        interchangeable with pure :class:`torch.Tensor`. See</span>
+<span class="sd">        :ref:`this note &lt;passthrough_heuristic&gt;` for more details.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        data (tensor-like, PIL.Image.Image): Any data that can be turned into a tensor with :func:`torch.as_tensor` as</span>
+<span class="sd">            well as PIL images.</span>
+<span class="sd">        dtype (torch.dtype, optional): Desired data type. If omitted, will be inferred from</span>
+<span class="sd">            ``data``.</span>
+<span class="sd">        device (torch.device, optional): Desired device. If omitted and ``data`` is a</span>
+<span class="sd">            :class:`torch.Tensor`, the device is taken from it. Otherwise, the image is constructed on the CPU.</span>
+<span class="sd">        requires_grad (bool, optional): Whether autograd should record operations. If omitted and</span>
+<span class="sd">            ``data`` is a :class:`torch.Tensor`, the value is taken from it. Otherwise, defaults to ``False``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__new__</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">data</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+        <span class="o">*</span><span class="p">,</span>
+        <span class="n">dtype</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">requires_grad</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Image</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+            <span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+            <span class="n">data</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">data</span><span class="p">)</span>
+
+        <span class="n">tensor</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_to_tensor</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">requires_grad</span><span class="o">=</span><span class="n">requires_grad</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span>
+        <span class="k">elif</span> <span class="n">tensor</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="n">tensor</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">tensor</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="bp">cls</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">tensor_contents</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_repr</span><span class="p">()</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors/_mask.html b/0.16/_modules/torchvision/tv_tensors/_mask.html
new file mode 100644
index 00000000000..40368e3532f
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors/_mask.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors._mask &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+          <li><a href="../tv_tensors.html">torchvision.tv_tensors</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors._mask</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors._mask</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">annotations</span>
+
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">PIL.Image</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">._tv_tensor</span> <span class="kn">import</span> <span class="n">TVTensor</span>
+
+
+<div class="viewcode-block" id="Mask"><a class="viewcode-back" href="../../../generated/torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask">[docs]</a><span class="k">class</span> <span class="nc">Mask</span><span class="p">(</span><span class="n">TVTensor</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] :class:`torch.Tensor` subclass for segmentation and detection masks.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        data (tensor-like, PIL.Image.Image): Any data that can be turned into a tensor with :func:`torch.as_tensor` as</span>
+<span class="sd">            well as PIL images.</span>
+<span class="sd">        dtype (torch.dtype, optional): Desired data type. If omitted, will be inferred from</span>
+<span class="sd">            ``data``.</span>
+<span class="sd">        device (torch.device, optional): Desired device. If omitted and ``data`` is a</span>
+<span class="sd">            :class:`torch.Tensor`, the device is taken from it. Otherwise, the mask is constructed on the CPU.</span>
+<span class="sd">        requires_grad (bool, optional): Whether autograd should record operations. If omitted and</span>
+<span class="sd">            ``data`` is a :class:`torch.Tensor`, the value is taken from it. Otherwise, defaults to ``False``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__new__</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">data</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+        <span class="o">*</span><span class="p">,</span>
+        <span class="n">dtype</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">requires_grad</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Mask</span><span class="p">:</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">Image</span><span class="p">):</span>
+            <span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+            <span class="n">data</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">pil_to_tensor</span><span class="p">(</span><span class="n">data</span><span class="p">)</span>
+
+        <span class="n">tensor</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_to_tensor</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">requires_grad</span><span class="o">=</span><span class="n">requires_grad</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">tensor</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="bp">cls</span><span class="p">)</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html b/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
new file mode 100644
index 00000000000..37c46dcf7cd
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
@@ -0,0 +1,743 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors._torch_function_helpers &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+          <li><a href="../tv_tensors.html">torchvision.tv_tensors</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors._torch_function_helpers</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors._torch_function_helpers</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="n">_TORCHFUNCTION_SUBCLASS</span> <span class="o">=</span> <span class="kc">False</span>
+
+
+<span class="k">class</span> <span class="nc">_ReturnTypeCM</span><span class="p">:</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">to_restore</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">to_restore</span> <span class="o">=</span> <span class="n">to_restore</span>
+
+    <span class="k">def</span> <span class="fm">__enter__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span>
+
+    <span class="k">def</span> <span class="fm">__exit__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">):</span>
+        <span class="k">global</span> <span class="n">_TORCHFUNCTION_SUBCLASS</span>
+        <span class="n">_TORCHFUNCTION_SUBCLASS</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">to_restore</span>
+
+
+<div class="viewcode-block" id="set_return_type"><a class="viewcode-back" href="../../../generated/torchvision.tv_tensors.set_return_type.html#torchvision.tv_tensors.set_return_type">[docs]</a><span class="k">def</span> <span class="nf">set_return_type</span><span class="p">(</span><span class="n">return_type</span><span class="p">:</span> <span class="nb">str</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] Set the return type of torch operations on :class:`~torchvision.tv_tensors.TVTensor`.</span>
+
+<span class="sd">    This only affects the behaviour of torch operations. It has no effect on</span>
+<span class="sd">    ``torchvision`` transforms or functionals, which will always return as</span>
+<span class="sd">    output the same type that was passed as input.</span>
+
+<span class="sd">    .. warning::</span>
+
+<span class="sd">        We recommend using :class:`~torchvision.transforms.v2.ToPureTensor` at</span>
+<span class="sd">        the end of your transform pipelines if you use</span>
+<span class="sd">        ``set_return_type(&quot;TVTensor&quot;)``. This will avoid the</span>
+<span class="sd">        ``__torch_function__`` overhead in the models ``forward()``.</span>
+
+<span class="sd">    Can be used as a global flag for the entire program:</span>
+
+<span class="sd">    .. code:: python</span>
+
+<span class="sd">        img = tv_tensors.Image(torch.rand(3, 5, 5))</span>
+<span class="sd">        img + 2  # This is a pure Tensor (default behaviour)</span>
+
+<span class="sd">        set_return_type(&quot;TVTensor&quot;)</span>
+<span class="sd">        img + 2  # This is an Image</span>
+
+<span class="sd">    or as a context manager to restrict the scope:</span>
+
+<span class="sd">    .. code:: python</span>
+
+<span class="sd">        img = tv_tensors.Image(torch.rand(3, 5, 5))</span>
+<span class="sd">        img + 2  # This is a pure Tensor</span>
+<span class="sd">        with set_return_type(&quot;TVTensor&quot;):</span>
+<span class="sd">            img + 2  # This is an Image</span>
+<span class="sd">        img + 2  # This is a pure Tensor</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        return_type (str): Can be &quot;TVTensor&quot; or &quot;Tensor&quot; (case-insensitive).</span>
+<span class="sd">            Default is &quot;Tensor&quot; (i.e. pure :class:`torch.Tensor`).</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">global</span> <span class="n">_TORCHFUNCTION_SUBCLASS</span>
+    <span class="n">to_restore</span> <span class="o">=</span> <span class="n">_TORCHFUNCTION_SUBCLASS</span>
+
+    <span class="k">try</span><span class="p">:</span>
+        <span class="n">_TORCHFUNCTION_SUBCLASS</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;tensor&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;tvtensor&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">}[</span><span class="n">return_type</span><span class="o">.</span><span class="n">lower</span><span class="p">()]</span>
+    <span class="k">except</span> <span class="ne">KeyError</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;return_type must be &#39;TVTensor&#39; or &#39;Tensor&#39;, got </span><span class="si">{</span><span class="n">return_type</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span> <span class="kn">from</span> <span class="kc">None</span>
+
+    <span class="k">return</span> <span class="n">_ReturnTypeCM</span><span class="p">(</span><span class="n">to_restore</span><span class="p">)</span></div>
+
+
+<span class="k">def</span> <span class="nf">_must_return_subclass</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_TORCHFUNCTION_SUBCLASS</span>
+
+
+<span class="c1"># For those ops we always want to preserve the original subclass instead of returning a pure Tensor</span>
+<span class="n">_FORCE_TORCHFUNCTION_SUBCLASS</span> <span class="o">=</span> <span class="p">{</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="o">.</span><span class="n">clone</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="o">.</span><span class="n">to</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="o">.</span><span class="n">detach</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">}</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html b/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
new file mode 100644
index 00000000000..f721b955a42
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
@@ -0,0 +1,803 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors._tv_tensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+          <li><a href="../tv_tensors.html">torchvision.tv_tensors</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors._tv_tensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors._tv_tensor</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">annotations</span>
+
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Mapping</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">TypeVar</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torch._C</span> <span class="kn">import</span> <span class="n">DisableTorchFunctionSubclass</span>
+<span class="kn">from</span> <span class="nn">torch.types</span> <span class="kn">import</span> <span class="n">_device</span><span class="p">,</span> <span class="n">_dtype</span><span class="p">,</span> <span class="n">_size</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.tv_tensors._torch_function_helpers</span> <span class="kn">import</span> <span class="n">_FORCE_TORCHFUNCTION_SUBCLASS</span><span class="p">,</span> <span class="n">_must_return_subclass</span>
+
+
+<span class="n">D</span> <span class="o">=</span> <span class="n">TypeVar</span><span class="p">(</span><span class="s2">&quot;D&quot;</span><span class="p">,</span> <span class="n">bound</span><span class="o">=</span><span class="s2">&quot;TVTensor&quot;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="TVTensor"><a class="viewcode-back" href="../../../generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor">[docs]</a><span class="k">class</span> <span class="nc">TVTensor</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[Beta] Base class for all TVTensors.</span>
+
+<span class="sd">    You probably don&#39;t want to use this class unless you&#39;re defining your own</span>
+<span class="sd">    custom TVTensors. See</span>
+<span class="sd">    :ref:`sphx_glr_auto_examples_transforms_plot_custom_tv_tensors.py` for details.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="nd">@staticmethod</span>
+    <span class="k">def</span> <span class="nf">_to_tensor</span><span class="p">(</span>
+        <span class="n">data</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+        <span class="n">dtype</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">requires_grad</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">requires_grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">requires_grad</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">requires_grad</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="k">else</span> <span class="kc">False</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">as_tensor</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">(</span><span class="n">requires_grad</span><span class="p">)</span>
+
+    <span class="nd">@classmethod</span>
+    <span class="k">def</span> <span class="nf">_wrap_output</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">output</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">(),</span>
+        <span class="n">kwargs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+        <span class="c1"># Same as torch._tensor._convert</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="bp">cls</span><span class="p">):</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="bp">cls</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="c1"># Also handles things like namedtuples</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="nb">type</span><span class="p">(</span><span class="n">output</span><span class="p">)(</span><span class="bp">cls</span><span class="o">.</span><span class="n">_wrap_output</span><span class="p">(</span><span class="n">part</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">kwargs</span><span class="p">)</span> <span class="k">for</span> <span class="n">part</span> <span class="ow">in</span> <span class="n">output</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">output</span>
+
+    <span class="nd">@classmethod</span>
+    <span class="k">def</span> <span class="nf">__torch_function__</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">func</span><span class="p">:</span> <span class="n">Callable</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span>
+        <span class="n">types</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Type</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="o">...</span><span class="p">],</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">(),</span>
+        <span class="n">kwargs</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Mapping</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;For general information about how the __torch_function__ protocol works,</span>
+<span class="sd">        see https://pytorch.org/docs/stable/notes/extending.html#extending-torch</span>
+
+<span class="sd">        TL;DR: Every time a PyTorch operator is called, it goes through the inputs and looks for the</span>
+<span class="sd">        ``__torch_function__`` method. If one is found, it is invoked with the operator as ``func`` as well as the</span>
+<span class="sd">        ``args`` and ``kwargs`` of the original call.</span>
+
+<span class="sd">        Why do we override this? Because the base implementation in torch.Tensor would preserve the TVTensor type</span>
+<span class="sd">        of the output. In our case, we want to return pure tensors instead (with a few exceptions). Refer to the</span>
+<span class="sd">        &quot;TVTensors FAQ&quot; gallery example for a rationale of this behaviour (TL;DR: perf + no silver bullet).</span>
+
+<span class="sd">        Our implementation below is very similar to the base implementation in ``torch.Tensor`` - go check it out.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">all</span><span class="p">(</span><span class="nb">issubclass</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">t</span><span class="p">)</span> <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">types</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">NotImplemented</span>
+
+        <span class="c1"># Like in the base Tensor.__torch_function__ implementation, it&#39;s easier to always use</span>
+        <span class="c1"># DisableTorchFunctionSubclass and then manually re-wrap the output if necessary</span>
+        <span class="k">with</span> <span class="n">DisableTorchFunctionSubclass</span><span class="p">():</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">func</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span> <span class="ow">or</span> <span class="nb">dict</span><span class="p">())</span>
+
+        <span class="n">must_return_subclass</span> <span class="o">=</span> <span class="n">_must_return_subclass</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">must_return_subclass</span> <span class="ow">or</span> <span class="p">(</span><span class="n">func</span> <span class="ow">in</span> <span class="n">_FORCE_TORCHFUNCTION_SUBCLASS</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">args</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="bp">cls</span><span class="p">)):</span>
+            <span class="c1"># If you&#39;re wondering why we need the `isinstance(args[0], cls)` check, remove it and see what fails</span>
+            <span class="c1"># in test_to_tv_tensor_reference().</span>
+            <span class="c1"># The __torch_function__ protocol will invoke the __torch_function__ method on *all* types involved in</span>
+            <span class="c1"># the computation by walking the MRO upwards. For example,</span>
+            <span class="c1"># `out = a_pure_tensor.to(an_image)` will invoke `Image.__torch_function__` with</span>
+            <span class="c1"># `args = (a_pure_tensor, an_image)` first. Without this guard, `out` would</span>
+            <span class="c1"># be wrapped into an `Image`.</span>
+            <span class="k">return</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_wrap_output</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">kwargs</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">must_return_subclass</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="bp">cls</span><span class="p">):</span>
+            <span class="c1"># DisableTorchFunctionSubclass is ignored by inplace ops like `.add_(...)`,</span>
+            <span class="c1"># so for those, the output is still a TVTensor. Thus, we need to manually unwrap.</span>
+            <span class="k">return</span> <span class="n">output</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">output</span>
+
+    <span class="k">def</span> <span class="nf">_make_repr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="c1"># This is a poor man&#39;s implementation of the proposal in https://github.com/pytorch/pytorch/issues/76532.</span>
+        <span class="c1"># If that ever gets implemented, remove this in favor of the solution on the `torch.Tensor` class.</span>
+        <span class="n">extra_repr</span> <span class="o">=</span> <span class="s2">&quot;, &quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">key</span><span class="si">}</span><span class="s2">=</span><span class="si">{</span><span class="n">value</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">items</span><span class="p">())</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__repr__</span><span class="p">()[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2">, </span><span class="si">{</span><span class="n">extra_repr</span><span class="si">}</span><span class="s2">)&quot;</span>
+
+    <span class="c1"># Add properties for common attributes like shape, dtype, device, ndim etc</span>
+    <span class="c1"># this way we return the result without passing into __torch_function__</span>
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">shape</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">_size</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">with</span> <span class="n">DisableTorchFunctionSubclass</span><span class="p">():</span>
+            <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">shape</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">ndim</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">with</span> <span class="n">DisableTorchFunctionSubclass</span><span class="p">():</span>
+            <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">ndim</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">device</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">_device</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">with</span> <span class="n">DisableTorchFunctionSubclass</span><span class="p">():</span>
+            <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">device</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span> <span class="nf">dtype</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">_dtype</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">with</span> <span class="n">DisableTorchFunctionSubclass</span><span class="p">():</span>
+            <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">dtype</span>
+
+    <span class="k">def</span> <span class="nf">__deepcopy__</span><span class="p">(</span><span class="bp">self</span><span class="p">:</span> <span class="n">D</span><span class="p">,</span> <span class="n">memo</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">D</span><span class="p">:</span>
+        <span class="c1"># We need to detach first, since a plain `Tensor.clone` will be part of the computation graph, which does</span>
+        <span class="c1"># *not* happen for `deepcopy(Tensor)`. A side-effect from detaching is that the `Tensor.requires_grad`</span>
+        <span class="c1"># attribute is cleared, so we need to refill it before we return.</span>
+        <span class="c1"># Note: We don&#39;t explicitly handle deep-copying of the metadata here. The only metadata we currently have is</span>
+        <span class="c1"># `BoundingBoxes.format` and `BoundingBoxes.canvas_size`, which are immutable and thus implicitly deep-copied by</span>
+        <span class="c1"># `BoundingBoxes.clone()`.</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">)</span>  <span class="c1"># type: ignore[return-value]</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/tv_tensors/_video.html b/0.16/_modules/torchvision/tv_tensors/_video.html
new file mode 100644
index 00000000000..251273c2af4
--- /dev/null
+++ b/0.16/_modules/torchvision/tv_tensors/_video.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.tv_tensors._video &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../../torchvision.html">torchvision</a> &gt;</li>
+        
+          <li><a href="../tv_tensors.html">torchvision.tv_tensors</a> &gt;</li>
+        
+      <li>torchvision.tv_tensors._video</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.tv_tensors._video</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span> <span class="nn">__future__</span> <span class="kn">import</span> <span class="n">annotations</span>
+
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="kn">from</span> <span class="nn">._tv_tensor</span> <span class="kn">import</span> <span class="n">TVTensor</span>
+
+
+<div class="viewcode-block" id="Video"><a class="viewcode-back" href="../../../generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video">[docs]</a><span class="k">class</span> <span class="nc">Video</span><span class="p">(</span><span class="n">TVTensor</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;[BETA] :class:`torch.Tensor` subclass for videos.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        data (tensor-like): Any data that can be turned into a tensor with :func:`torch.as_tensor`.</span>
+<span class="sd">        dtype (torch.dtype, optional): Desired data type. If omitted, will be inferred from</span>
+<span class="sd">            ``data``.</span>
+<span class="sd">        device (torch.device, optional): Desired device. If omitted and ``data`` is a</span>
+<span class="sd">            :class:`torch.Tensor`, the device is taken from it. Otherwise, the video is constructed on the CPU.</span>
+<span class="sd">        requires_grad (bool, optional): Whether autograd should record operations. If omitted and</span>
+<span class="sd">            ``data`` is a :class:`torch.Tensor`, the value is taken from it. Otherwise, defaults to ``False``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span> <span class="fm">__new__</span><span class="p">(</span>
+        <span class="bp">cls</span><span class="p">,</span>
+        <span class="n">data</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span>
+        <span class="o">*</span><span class="p">,</span>
+        <span class="n">dtype</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">requires_grad</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Video</span><span class="p">:</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">_to_tensor</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span> <span class="n">requires_grad</span><span class="o">=</span><span class="n">requires_grad</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">data</span><span class="o">.</span><span class="n">ndim</span> <span class="o">&lt;</span> <span class="mi">4</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span>
+        <span class="k">return</span> <span class="n">tensor</span><span class="o">.</span><span class="n">as_subclass</span><span class="p">(</span><span class="bp">cls</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">tensor_contents</span><span class="p">:</span> <span class="n">Any</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>  <span class="c1"># type: ignore[override]</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_repr</span><span class="p">()</span></div>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
+         <script data-url_root="../../../" id="documentation_options" src="../../../_static/documentation_options.js"></script>
+         <script src="../../../_static/jquery.js"></script>
+         <script src="../../../_static/underscore.js"></script>
+         <script src="../../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../../_static/doctools.js"></script>
+         <script src="../../../_static/clipboard.min.js"></script>
+         <script src="../../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_modules/torchvision/utils.html b/0.16/_modules/torchvision/utils.html
new file mode 100644
index 00000000000..20e21b95987
--- /dev/null
+++ b/0.16/_modules/torchvision/utils.html
@@ -0,0 +1,1266 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.utils &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Module code</a> &gt;</li>
+        
+          <li><a href="../torchvision.html">torchvision</a> &gt;</li>
+        
+      <li>torchvision.utils</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <h1>Source code for torchvision.utils</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span> <span class="nn">collections</span>
+<span class="kn">import</span> <span class="nn">math</span>
+<span class="kn">import</span> <span class="nn">pathlib</span>
+<span class="kn">import</span> <span class="nn">warnings</span>
+<span class="kn">from</span> <span class="nn">itertools</span> <span class="kn">import</span> <span class="n">repeat</span>
+<span class="kn">from</span> <span class="nn">types</span> <span class="kn">import</span> <span class="n">FunctionType</span>
+<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">BinaryIO</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span><span class="p">,</span> <span class="n">ImageColor</span><span class="p">,</span> <span class="n">ImageDraw</span><span class="p">,</span> <span class="n">ImageFont</span>
+
+<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;make_grid&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;save_image&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;draw_bounding_boxes&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;draw_segmentation_masks&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;draw_keypoints&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;flow_to_image&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+
+
+<div class="viewcode-block" id="make_grid"><a class="viewcode-back" href="../../generated/torchvision.utils.make_grid.html#torchvision.utils.make_grid">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<span class="k">def</span> <span class="nf">make_grid</span><span class="p">(</span>
+    <span class="n">tensor</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]],</span>
+    <span class="n">nrow</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">8</span><span class="p">,</span>
+    <span class="n">padding</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">2</span><span class="p">,</span>
+    <span class="n">normalize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">value_range</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">scale_each</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">pad_value</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Make a grid of images.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        tensor (Tensor or list): 4D mini-batch Tensor of shape (B x C x H x W)</span>
+<span class="sd">            or a list of images all of the same size.</span>
+<span class="sd">        nrow (int, optional): Number of images displayed in each row of the grid.</span>
+<span class="sd">            The final grid size is ``(B / nrow, nrow)``. Default: ``8``.</span>
+<span class="sd">        padding (int, optional): amount of padding. Default: ``2``.</span>
+<span class="sd">        normalize (bool, optional): If True, shift the image to the range (0, 1),</span>
+<span class="sd">            by the min and max values specified by ``value_range``. Default: ``False``.</span>
+<span class="sd">        value_range (tuple, optional): tuple (min, max) where min and max are numbers,</span>
+<span class="sd">            then these numbers are used to normalize the image. By default, min and max</span>
+<span class="sd">            are computed from the tensor.</span>
+<span class="sd">        scale_each (bool, optional): If ``True``, scale each image in the batch of</span>
+<span class="sd">            images separately rather than the (min, max) over all images. Default: ``False``.</span>
+<span class="sd">        pad_value (float, optional): Value for the padded pixels. Default: ``0``.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        grid (Tensor): the tensor containing grid of images.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">make_grid</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_tensor</span><span class="p">(</span><span class="n">tensor</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">tensor</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_tensor</span><span class="p">(</span><span class="n">t</span><span class="p">):</span>
+                    <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;tensor or list of tensors expected, got a list containing </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">t</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;tensor or list of tensors expected, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">tensor</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="c1"># if list of tensors, convert to a 4D mini-batch Tensor</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>  <span class="c1"># single image H x W</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>  <span class="c1"># single image</span>
+        <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>  <span class="c1"># if single-channel, convert to 3-channel</span>
+            <span class="n">tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">tensor</span><span class="p">,</span> <span class="n">tensor</span><span class="p">,</span> <span class="n">tensor</span><span class="p">),</span> <span class="mi">0</span><span class="p">)</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">4</span> <span class="ow">and</span> <span class="n">tensor</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>  <span class="c1"># single-channel images</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">tensor</span><span class="p">,</span> <span class="n">tensor</span><span class="p">,</span> <span class="n">tensor</span><span class="p">),</span> <span class="mi">1</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">normalize</span> <span class="ow">is</span> <span class="kc">True</span><span class="p">:</span>
+        <span class="n">tensor</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>  <span class="c1"># avoid modifying tensor in-place</span>
+        <span class="k">if</span> <span class="n">value_range</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value_range</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;value_range has to be a tuple (min, max) if specified. min and max are numbers&quot;</span><span class="p">)</span>
+
+        <span class="k">def</span> <span class="nf">norm_ip</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">low</span><span class="p">,</span> <span class="n">high</span><span class="p">):</span>
+            <span class="n">img</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="nb">min</span><span class="o">=</span><span class="n">low</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="n">high</span><span class="p">)</span>
+            <span class="n">img</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">low</span><span class="p">)</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="nb">max</span><span class="p">(</span><span class="n">high</span> <span class="o">-</span> <span class="n">low</span><span class="p">,</span> <span class="mf">1e-5</span><span class="p">))</span>
+
+        <span class="k">def</span> <span class="nf">norm_range</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="n">value_range</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">value_range</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">norm_ip</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="n">value_range</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">value_range</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">norm_ip</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="n">t</span><span class="o">.</span><span class="n">min</span><span class="p">()),</span> <span class="nb">float</span><span class="p">(</span><span class="n">t</span><span class="o">.</span><span class="n">max</span><span class="p">()))</span>
+
+        <span class="k">if</span> <span class="n">scale_each</span> <span class="ow">is</span> <span class="kc">True</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">tensor</span><span class="p">:</span>  <span class="c1"># loop over mini-batch dimension</span>
+                <span class="n">norm_range</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="n">value_range</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">norm_range</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="n">value_range</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s2">&quot;tensor should be of type torch.Tensor&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">tensor</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+    <span class="c1"># make the mini-batch of images into a grid</span>
+    <span class="n">nmaps</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">xmaps</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">nrow</span><span class="p">,</span> <span class="n">nmaps</span><span class="p">)</span>
+    <span class="n">ymaps</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">nmaps</span><span class="p">)</span> <span class="o">/</span> <span class="n">xmaps</span><span class="p">))</span>
+    <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">tensor</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span> <span class="o">+</span> <span class="n">padding</span><span class="p">),</span> <span class="nb">int</span><span class="p">(</span><span class="n">tensor</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">3</span><span class="p">)</span> <span class="o">+</span> <span class="n">padding</span><span class="p">)</span>
+    <span class="n">num_channels</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">new_full</span><span class="p">((</span><span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span> <span class="o">*</span> <span class="n">ymaps</span> <span class="o">+</span> <span class="n">padding</span><span class="p">,</span> <span class="n">width</span> <span class="o">*</span> <span class="n">xmaps</span> <span class="o">+</span> <span class="n">padding</span><span class="p">),</span> <span class="n">pad_value</span><span class="p">)</span>
+    <span class="n">k</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="k">for</span> <span class="n">y</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">ymaps</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">xmaps</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">k</span> <span class="o">&gt;=</span> <span class="n">nmaps</span><span class="p">:</span>
+                <span class="k">break</span>
+            <span class="c1"># Tensor.copy_() is a valid method but seems to be missing from the stubs</span>
+            <span class="c1"># https://pytorch.org/docs/stable/tensors.html#torch.Tensor.copy_</span>
+            <span class="n">grid</span><span class="o">.</span><span class="n">narrow</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">y</span> <span class="o">*</span> <span class="n">height</span> <span class="o">+</span> <span class="n">padding</span><span class="p">,</span> <span class="n">height</span> <span class="o">-</span> <span class="n">padding</span><span class="p">)</span><span class="o">.</span><span class="n">narrow</span><span class="p">(</span>  <span class="c1"># type: ignore[attr-defined]</span>
+                <span class="mi">2</span><span class="p">,</span> <span class="n">x</span> <span class="o">*</span> <span class="n">width</span> <span class="o">+</span> <span class="n">padding</span><span class="p">,</span> <span class="n">width</span> <span class="o">-</span> <span class="n">padding</span>
+            <span class="p">)</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">tensor</span><span class="p">[</span><span class="n">k</span><span class="p">])</span>
+            <span class="n">k</span> <span class="o">=</span> <span class="n">k</span> <span class="o">+</span> <span class="mi">1</span>
+    <span class="k">return</span> <span class="n">grid</span></div>
+
+
+<div class="viewcode-block" id="save_image"><a class="viewcode-back" href="../../generated/torchvision.utils.save_image.html#torchvision.utils.save_image">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<span class="k">def</span> <span class="nf">save_image</span><span class="p">(</span>
+    <span class="n">tensor</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]],</span>
+    <span class="n">fp</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span><span class="p">,</span> <span class="n">BinaryIO</span><span class="p">],</span>
+    <span class="nb">format</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Save a given Tensor into an image file.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        tensor (Tensor or list): Image to be saved. If given a mini-batch tensor,</span>
+<span class="sd">            saves the tensor as a grid of images by calling ``make_grid``.</span>
+<span class="sd">        fp (string or file object): A filename or a file object</span>
+<span class="sd">        format(Optional):  If omitted, the format to use is determined from the filename extension.</span>
+<span class="sd">            If a file object was used instead of a filename, this parameter should always be used.</span>
+<span class="sd">        **kwargs: Other arguments are documented in ``make_grid``.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">save_image</span><span class="p">)</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="n">make_grid</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+    <span class="c1"># Add 0.5 after unnormalizing to [0, 255] to round to the nearest integer</span>
+    <span class="n">ndarr</span> <span class="o">=</span> <span class="n">grid</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="mi">255</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mf">0.5</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">255</span><span class="p">)</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
+    <span class="n">im</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">ndarr</span><span class="p">)</span>
+    <span class="n">im</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">fp</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="nb">format</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="draw_bounding_boxes"><a class="viewcode-back" href="../../generated/torchvision.utils.draw_bounding_boxes.html#torchvision.utils.draw_bounding_boxes">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<span class="k">def</span> <span class="nf">draw_bounding_boxes</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">boxes</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">labels</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">colors</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]],</span> <span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">fill</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+    <span class="n">font</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">font_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Draws bounding boxes on given image.</span>
+<span class="sd">    The values of the input image should be uint8 between 0 and 255.</span>
+<span class="sd">    If fill is True, Resulting Tensor should be saved as PNG image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        image (Tensor): Tensor of shape (C x H x W) and dtype uint8.</span>
+<span class="sd">        boxes (Tensor): Tensor of size (N, 4) containing bounding boxes in (xmin, ymin, xmax, ymax) format. Note that</span>
+<span class="sd">            the boxes are absolute coordinates with respect to the image. In other words: `0 &lt;= xmin &lt; xmax &lt; W` and</span>
+<span class="sd">            `0 &lt;= ymin &lt; ymax &lt; H`.</span>
+<span class="sd">        labels (List[str]): List containing the labels of bounding boxes.</span>
+<span class="sd">        colors (color or list of colors, optional): List containing the colors</span>
+<span class="sd">            of the boxes or single color for all boxes. The color can be represented as</span>
+<span class="sd">            PIL strings e.g. &quot;red&quot; or &quot;#FF00FF&quot;, or as RGB tuples e.g. ``(240, 10, 157)``.</span>
+<span class="sd">            By default, random colors are generated for boxes.</span>
+<span class="sd">        fill (bool): If `True` fills the bounding box with specified color.</span>
+<span class="sd">        width (int): Width of bounding box.</span>
+<span class="sd">        font (str): A filename containing a TrueType font. If the file is not found in this filename, the loader may</span>
+<span class="sd">            also search in other directories, such as the `fonts/` directory on Windows or `/Library/Fonts/`,</span>
+<span class="sd">            `/System/Library/Fonts/` and `~/Library/Fonts/` on macOS.</span>
+<span class="sd">        font_size (int): The requested font size in points.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        img (Tensor[C, H, W]): Image Tensor of dtype uint8 with bounding boxes plotted.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">draw_bounding_boxes</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Tensor expected, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">image</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Tensor uint8 expected, got </span><span class="si">{</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Pass individual images, not batches&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">{</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">}:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Only grayscale and RGB images are supported&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">])</span><span class="o">.</span><span class="n">any</span><span class="p">()</span> <span class="ow">or</span> <span class="p">(</span><span class="n">boxes</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">3</span><span class="p">])</span><span class="o">.</span><span class="n">any</span><span class="p">():</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="s2">&quot;Boxes need to be in (xmin, ymin, xmax, ymax) format. Use torchvision.ops.box_convert to convert them&quot;</span>
+        <span class="p">)</span>
+
+    <span class="n">num_boxes</span> <span class="o">=</span> <span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="n">num_boxes</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;boxes doesn&#39;t contain any box. No box was drawn&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">labels</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="kc">None</span><span class="p">]]</span> <span class="o">=</span> <span class="p">[</span><span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_boxes</span>  <span class="c1"># type: ignore[no-redef]</span>
+    <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span> <span class="o">!=</span> <span class="n">num_boxes</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Number of boxes (</span><span class="si">{</span><span class="n">num_boxes</span><span class="si">}</span><span class="s2">) and labels (</span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span><span class="si">}</span><span class="s2">) mismatch. Please specify labels for each box.&quot;</span>
+        <span class="p">)</span>
+
+    <span class="n">colors</span> <span class="o">=</span> <span class="n">_parse_colors</span><span class="p">(</span><span class="n">colors</span><span class="p">,</span> <span class="n">num_objects</span><span class="o">=</span><span class="n">num_boxes</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">font</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">font_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;Argument &#39;font_size&#39; will be ignored since &#39;font&#39; is not set.&quot;</span><span class="p">)</span>
+        <span class="n">txt_font</span> <span class="o">=</span> <span class="n">ImageFont</span><span class="o">.</span><span class="n">load_default</span><span class="p">()</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">txt_font</span> <span class="o">=</span> <span class="n">ImageFont</span><span class="o">.</span><span class="n">truetype</span><span class="p">(</span><span class="n">font</span><span class="o">=</span><span class="n">font</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">font_size</span> <span class="ow">or</span> <span class="mi">10</span><span class="p">)</span>
+
+    <span class="c1"># Handle Grayscale images</span>
+    <span class="k">if</span> <span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">image</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tile</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
+
+    <span class="n">ndarr</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
+    <span class="n">img_to_draw</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">ndarr</span><span class="p">)</span>
+    <span class="n">img_boxes</span> <span class="o">=</span> <span class="n">boxes</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">)</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+
+    <span class="k">if</span> <span class="n">fill</span><span class="p">:</span>
+        <span class="n">draw</span> <span class="o">=</span> <span class="n">ImageDraw</span><span class="o">.</span><span class="n">Draw</span><span class="p">(</span><span class="n">img_to_draw</span><span class="p">,</span> <span class="s2">&quot;RGBA&quot;</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">draw</span> <span class="o">=</span> <span class="n">ImageDraw</span><span class="o">.</span><span class="n">Draw</span><span class="p">(</span><span class="n">img_to_draw</span><span class="p">)</span>
+
+    <span class="k">for</span> <span class="n">bbox</span><span class="p">,</span> <span class="n">color</span><span class="p">,</span> <span class="n">label</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">img_boxes</span><span class="p">,</span> <span class="n">colors</span><span class="p">,</span> <span class="n">labels</span><span class="p">):</span>  <span class="c1"># type: ignore[arg-type]</span>
+        <span class="k">if</span> <span class="n">fill</span><span class="p">:</span>
+            <span class="n">fill_color</span> <span class="o">=</span> <span class="n">color</span> <span class="o">+</span> <span class="p">(</span><span class="mi">100</span><span class="p">,)</span>
+            <span class="n">draw</span><span class="o">.</span><span class="n">rectangle</span><span class="p">(</span><span class="n">bbox</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span> <span class="n">outline</span><span class="o">=</span><span class="n">color</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">fill_color</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">draw</span><span class="o">.</span><span class="n">rectangle</span><span class="p">(</span><span class="n">bbox</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span> <span class="n">outline</span><span class="o">=</span><span class="n">color</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">label</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">margin</span> <span class="o">=</span> <span class="n">width</span> <span class="o">+</span> <span class="mi">1</span>
+            <span class="n">draw</span><span class="o">.</span><span class="n">text</span><span class="p">((</span><span class="n">bbox</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">margin</span><span class="p">,</span> <span class="n">bbox</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">margin</span><span class="p">),</span> <span class="n">label</span><span class="p">,</span> <span class="n">fill</span><span class="o">=</span><span class="n">color</span><span class="p">,</span> <span class="n">font</span><span class="o">=</span><span class="n">txt_font</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">img_to_draw</span><span class="p">))</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="draw_segmentation_masks"><a class="viewcode-back" href="../../generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<span class="k">def</span> <span class="nf">draw_segmentation_masks</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">masks</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">alpha</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.8</span><span class="p">,</span>
+    <span class="n">colors</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]],</span> <span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Draws segmentation masks on given RGB image.</span>
+<span class="sd">    The values of the input image should be uint8 between 0 and 255.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        image (Tensor): Tensor of shape (3, H, W) and dtype uint8.</span>
+<span class="sd">        masks (Tensor): Tensor of shape (num_masks, H, W) or (H, W) and dtype bool.</span>
+<span class="sd">        alpha (float): Float number between 0 and 1 denoting the transparency of the masks.</span>
+<span class="sd">            0 means full transparency, 1 means no transparency.</span>
+<span class="sd">        colors (color or list of colors, optional): List containing the colors</span>
+<span class="sd">            of the masks or single color for all masks. The color can be represented as</span>
+<span class="sd">            PIL strings e.g. &quot;red&quot; or &quot;#FF00FF&quot;, or as RGB tuples e.g. ``(240, 10, 157)``.</span>
+<span class="sd">            By default, random colors are generated for each mask.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        img (Tensor[C, H, W]): Image Tensor, with segmentation masks drawn on top.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">draw_segmentation_masks</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The image must be a tensor, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">image</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The image dtype must be uint8, got </span><span class="si">{</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Pass individual images, not batches&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">()[</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Pass an RGB image. Other Image formats are not supported&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">masks</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="n">masks</span> <span class="o">=</span> <span class="n">masks</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span>
+    <span class="k">if</span> <span class="n">masks</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;masks must be of shape (H, W) or (batch_size, H, W)&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">masks</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">bool</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The masks must be of dtype bool. Got </span><span class="si">{</span><span class="n">masks</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">masks</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span> <span class="o">!=</span> <span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The image and the masks must have the same height and width&quot;</span><span class="p">)</span>
+
+    <span class="n">num_masks</span> <span class="o">=</span> <span class="n">masks</span><span class="o">.</span><span class="n">size</span><span class="p">()[</span><span class="mi">0</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="n">num_masks</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span><span class="s2">&quot;masks doesn&#39;t contain any mask. No mask was drawn&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">image</span>
+
+    <span class="n">out_dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span>
+    <span class="n">colors</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">color</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">out_dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">color</span> <span class="ow">in</span> <span class="n">_parse_colors</span><span class="p">(</span><span class="n">colors</span><span class="p">,</span> <span class="n">num_objects</span><span class="o">=</span><span class="n">num_masks</span><span class="p">)</span>
+    <span class="p">]</span>
+
+    <span class="n">img_to_draw</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+    <span class="c1"># TODO: There might be a way to vectorize this</span>
+    <span class="k">for</span> <span class="n">mask</span><span class="p">,</span> <span class="n">color</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">masks</span><span class="p">,</span> <span class="n">colors</span><span class="p">):</span>
+        <span class="n">img_to_draw</span><span class="p">[:,</span> <span class="n">mask</span><span class="p">]</span> <span class="o">=</span> <span class="n">color</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span>
+
+    <span class="n">out</span> <span class="o">=</span> <span class="n">image</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">alpha</span><span class="p">)</span> <span class="o">+</span> <span class="n">img_to_draw</span> <span class="o">*</span> <span class="n">alpha</span>
+    <span class="k">return</span> <span class="n">out</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">out_dtype</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="draw_keypoints"><a class="viewcode-back" href="../../generated/torchvision.utils.draw_keypoints.html#torchvision.utils.draw_keypoints">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<span class="k">def</span> <span class="nf">draw_keypoints</span><span class="p">(</span>
+    <span class="n">image</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">keypoints</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">connectivity</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">colors</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">radius</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">2</span><span class="p">,</span>
+    <span class="n">width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Draws Keypoints on given RGB image.</span>
+<span class="sd">    The values of the input image should be uint8 between 0 and 255.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        image (Tensor): Tensor of shape (3, H, W) and dtype uint8.</span>
+<span class="sd">        keypoints (Tensor): Tensor of shape (num_instances, K, 2) the K keypoints location for each of the N instances,</span>
+<span class="sd">            in the format [x, y].</span>
+<span class="sd">        connectivity (List[Tuple[int, int]]]): A List of tuple where,</span>
+<span class="sd">            each tuple contains pair of keypoints to be connected.</span>
+<span class="sd">        colors (str, Tuple): The color can be represented as</span>
+<span class="sd">            PIL strings e.g. &quot;red&quot; or &quot;#FF00FF&quot;, or as RGB tuples e.g. ``(240, 10, 157)``.</span>
+<span class="sd">        radius (int): Integer denoting radius of keypoint.</span>
+<span class="sd">        width (int): Integer denoting width of line connecting keypoints.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        img (Tensor[C, H, W]): Image Tensor of dtype uint8 with keypoints drawn.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_scripting</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">is_tracing</span><span class="p">():</span>
+        <span class="n">_log_api_usage_once</span><span class="p">(</span><span class="n">draw_keypoints</span><span class="p">)</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The image must be a tensor, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">image</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The image dtype must be uint8, got </span><span class="si">{</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Pass individual images, not batches&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">()[</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Pass an RGB image. Other Image formats are not supported&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">keypoints</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;keypoints must be of shape (num_instances, K, 2)&quot;</span><span class="p">)</span>
+
+    <span class="n">ndarr</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">numpy</span><span class="p">()</span>
+    <span class="n">img_to_draw</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">fromarray</span><span class="p">(</span><span class="n">ndarr</span><span class="p">)</span>
+    <span class="n">draw</span> <span class="o">=</span> <span class="n">ImageDraw</span><span class="o">.</span><span class="n">Draw</span><span class="p">(</span><span class="n">img_to_draw</span><span class="p">)</span>
+    <span class="n">img_kpts</span> <span class="o">=</span> <span class="n">keypoints</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">)</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+
+    <span class="k">for</span> <span class="n">kpt_id</span><span class="p">,</span> <span class="n">kpt_inst</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">img_kpts</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">inst_id</span><span class="p">,</span> <span class="n">kpt</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">kpt_inst</span><span class="p">):</span>
+            <span class="n">x1</span> <span class="o">=</span> <span class="n">kpt</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="n">radius</span>
+            <span class="n">x2</span> <span class="o">=</span> <span class="n">kpt</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+</span> <span class="n">radius</span>
+            <span class="n">y1</span> <span class="o">=</span> <span class="n">kpt</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">radius</span>
+            <span class="n">y2</span> <span class="o">=</span> <span class="n">kpt</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">radius</span>
+            <span class="n">draw</span><span class="o">.</span><span class="n">ellipse</span><span class="p">([</span><span class="n">x1</span><span class="p">,</span> <span class="n">y1</span><span class="p">,</span> <span class="n">x2</span><span class="p">,</span> <span class="n">y2</span><span class="p">],</span> <span class="n">fill</span><span class="o">=</span><span class="n">colors</span><span class="p">,</span> <span class="n">outline</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">connectivity</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">connection</span> <span class="ow">in</span> <span class="n">connectivity</span><span class="p">:</span>
+                <span class="n">start_pt_x</span> <span class="o">=</span> <span class="n">kpt_inst</span><span class="p">[</span><span class="n">connection</span><span class="p">[</span><span class="mi">0</span><span class="p">]][</span><span class="mi">0</span><span class="p">]</span>
+                <span class="n">start_pt_y</span> <span class="o">=</span> <span class="n">kpt_inst</span><span class="p">[</span><span class="n">connection</span><span class="p">[</span><span class="mi">0</span><span class="p">]][</span><span class="mi">1</span><span class="p">]</span>
+
+                <span class="n">end_pt_x</span> <span class="o">=</span> <span class="n">kpt_inst</span><span class="p">[</span><span class="n">connection</span><span class="p">[</span><span class="mi">1</span><span class="p">]][</span><span class="mi">0</span><span class="p">]</span>
+                <span class="n">end_pt_y</span> <span class="o">=</span> <span class="n">kpt_inst</span><span class="p">[</span><span class="n">connection</span><span class="p">[</span><span class="mi">1</span><span class="p">]][</span><span class="mi">1</span><span class="p">]</span>
+
+                <span class="n">draw</span><span class="o">.</span><span class="n">line</span><span class="p">(</span>
+                    <span class="p">((</span><span class="n">start_pt_x</span><span class="p">,</span> <span class="n">start_pt_y</span><span class="p">),</span> <span class="p">(</span><span class="n">end_pt_x</span><span class="p">,</span> <span class="n">end_pt_y</span><span class="p">)),</span>
+                    <span class="n">width</span><span class="o">=</span><span class="n">width</span><span class="p">,</span>
+                <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">img_to_draw</span><span class="p">))</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span></div>
+
+
+<span class="c1"># Flow visualization code adapted from https://github.com/tomrunia/OpticalFlow_Visualization</span>
+<div class="viewcode-block" id="flow_to_image"><a class="viewcode-back" href="../../generated/torchvision.utils.flow_to_image.html#torchvision.utils.flow_to_image">[docs]</a><span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<span class="k">def</span> <span class="nf">flow_to_image</span><span class="p">(</span><span class="n">flow</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Converts a flow to an RGB image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        flow (Tensor): Flow of shape (N, 2, H, W) or (2, H, W) and dtype torch.float.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        img (Tensor): Image Tensor of dtype uint8 where each color corresponds</span>
+<span class="sd">            to a given flow direction. Shape is (N, 3, H, W) or (3, H, W) depending on the input.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">if</span> <span class="n">flow</span><span class="o">.</span><span class="n">dtype</span> <span class="o">!=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Flow should be of dtype torch.float, got </span><span class="si">{</span><span class="n">flow</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="n">orig_shape</span> <span class="o">=</span> <span class="n">flow</span><span class="o">.</span><span class="n">shape</span>
+    <span class="k">if</span> <span class="n">flow</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">flow</span> <span class="o">=</span> <span class="n">flow</span><span class="p">[</span><span class="kc">None</span><span class="p">]</span>  <span class="c1"># Add batch dim</span>
+
+    <span class="k">if</span> <span class="n">flow</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">4</span> <span class="ow">or</span> <span class="n">flow</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Input flow should have shape (2, H, W) or (N, 2, H, W), got </span><span class="si">{</span><span class="n">orig_shape</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+    <span class="n">max_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">flow</span><span class="o">**</span><span class="mi">2</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">max</span><span class="p">()</span>
+    <span class="n">epsilon</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">finfo</span><span class="p">((</span><span class="n">flow</span><span class="p">)</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">eps</span>
+    <span class="n">normalized_flow</span> <span class="o">=</span> <span class="n">flow</span> <span class="o">/</span> <span class="p">(</span><span class="n">max_norm</span> <span class="o">+</span> <span class="n">epsilon</span><span class="p">)</span>
+    <span class="n">img</span> <span class="o">=</span> <span class="n">_normalized_flow_to_image</span><span class="p">(</span><span class="n">normalized_flow</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">orig_shape</span><span class="p">)</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>  <span class="c1"># Remove batch dim</span>
+    <span class="k">return</span> <span class="n">img</span></div>
+
+
+<span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<span class="k">def</span> <span class="nf">_normalized_flow_to_image</span><span class="p">(</span><span class="n">normalized_flow</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Converts a batch of normalized flow to an RGB image.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        normalized_flow (torch.Tensor): Normalized flow tensor of shape (N, 2, H, W)</span>
+<span class="sd">    Returns:</span>
+<span class="sd">       img (Tensor(N, 3, H, W)): Flow visualization image of dtype uint8.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">N</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">normalized_flow</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">device</span> <span class="o">=</span> <span class="n">normalized_flow</span><span class="o">.</span><span class="n">device</span>
+    <span class="n">flow_image</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">N</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+    <span class="n">colorwheel</span> <span class="o">=</span> <span class="n">_make_colorwheel</span><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>  <span class="c1"># shape [55x3]</span>
+    <span class="n">num_cols</span> <span class="o">=</span> <span class="n">colorwheel</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">normalized_flow</span><span class="o">**</span><span class="mi">2</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span>
+    <span class="n">a</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">atan2</span><span class="p">(</span><span class="o">-</span><span class="n">normalized_flow</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:],</span> <span class="o">-</span><span class="n">normalized_flow</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:])</span> <span class="o">/</span> <span class="n">torch</span><span class="o">.</span><span class="n">pi</span>
+    <span class="n">fk</span> <span class="o">=</span> <span class="p">(</span><span class="n">a</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span> <span class="o">*</span> <span class="p">(</span><span class="n">num_cols</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+    <span class="n">k0</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="n">fk</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)</span>
+    <span class="n">k1</span> <span class="o">=</span> <span class="n">k0</span> <span class="o">+</span> <span class="mi">1</span>
+    <span class="n">k1</span><span class="p">[</span><span class="n">k1</span> <span class="o">==</span> <span class="n">num_cols</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">f</span> <span class="o">=</span> <span class="n">fk</span> <span class="o">-</span> <span class="n">k0</span>
+
+    <span class="k">for</span> <span class="n">c</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">colorwheel</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]):</span>
+        <span class="n">tmp</span> <span class="o">=</span> <span class="n">colorwheel</span><span class="p">[:,</span> <span class="n">c</span><span class="p">]</span>
+        <span class="n">col0</span> <span class="o">=</span> <span class="n">tmp</span><span class="p">[</span><span class="n">k0</span><span class="p">]</span> <span class="o">/</span> <span class="mf">255.0</span>
+        <span class="n">col1</span> <span class="o">=</span> <span class="n">tmp</span><span class="p">[</span><span class="n">k1</span><span class="p">]</span> <span class="o">/</span> <span class="mf">255.0</span>
+        <span class="n">col</span> <span class="o">=</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">f</span><span class="p">)</span> <span class="o">*</span> <span class="n">col0</span> <span class="o">+</span> <span class="n">f</span> <span class="o">*</span> <span class="n">col1</span>
+        <span class="n">col</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">norm</span> <span class="o">*</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">col</span><span class="p">)</span>
+        <span class="n">flow_image</span><span class="p">[:,</span> <span class="n">c</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="mi">255</span> <span class="o">*</span> <span class="n">col</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">flow_image</span>
+
+
+<span class="k">def</span> <span class="nf">_make_colorwheel</span><span class="p">()</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Generates a color wheel for optical flow visualization as presented in:</span>
+<span class="sd">    Baker et al. &quot;A Database and Evaluation Methodology for Optical Flow&quot; (ICCV, 2007)</span>
+<span class="sd">    URL: http://vision.middlebury.edu/flow/flowEval-iccv07.pdf.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        colorwheel (Tensor[55, 3]): Colorwheel Tensor.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">RY</span> <span class="o">=</span> <span class="mi">15</span>
+    <span class="n">YG</span> <span class="o">=</span> <span class="mi">6</span>
+    <span class="n">GC</span> <span class="o">=</span> <span class="mi">4</span>
+    <span class="n">CB</span> <span class="o">=</span> <span class="mi">11</span>
+    <span class="n">BM</span> <span class="o">=</span> <span class="mi">13</span>
+    <span class="n">MR</span> <span class="o">=</span> <span class="mi">6</span>
+
+    <span class="n">ncols</span> <span class="o">=</span> <span class="n">RY</span> <span class="o">+</span> <span class="n">YG</span> <span class="o">+</span> <span class="n">GC</span> <span class="o">+</span> <span class="n">CB</span> <span class="o">+</span> <span class="n">BM</span> <span class="o">+</span> <span class="n">MR</span>
+    <span class="n">colorwheel</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">ncols</span><span class="p">,</span> <span class="mi">3</span><span class="p">))</span>
+    <span class="n">col</span> <span class="o">=</span> <span class="mi">0</span>
+
+    <span class="c1"># RY</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="mi">0</span><span class="p">:</span><span class="n">RY</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="mi">0</span><span class="p">:</span><span class="n">RY</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="mi">255</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">RY</span><span class="p">)</span> <span class="o">/</span> <span class="n">RY</span><span class="p">)</span>
+    <span class="n">col</span> <span class="o">=</span> <span class="n">col</span> <span class="o">+</span> <span class="n">RY</span>
+    <span class="c1"># YG</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">YG</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="mi">255</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">YG</span><span class="p">)</span> <span class="o">/</span> <span class="n">YG</span><span class="p">)</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">YG</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span>
+    <span class="n">col</span> <span class="o">=</span> <span class="n">col</span> <span class="o">+</span> <span class="n">YG</span>
+    <span class="c1"># GC</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">GC</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">GC</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="mi">255</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">GC</span><span class="p">)</span> <span class="o">/</span> <span class="n">GC</span><span class="p">)</span>
+    <span class="n">col</span> <span class="o">=</span> <span class="n">col</span> <span class="o">+</span> <span class="n">GC</span>
+    <span class="c1"># CB</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">CB</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="mi">255</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">CB</span><span class="p">)</span> <span class="o">/</span> <span class="n">CB</span><span class="p">)</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">CB</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span>
+    <span class="n">col</span> <span class="o">=</span> <span class="n">col</span> <span class="o">+</span> <span class="n">CB</span>
+    <span class="c1"># BM</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">BM</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">BM</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="mi">255</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BM</span><span class="p">)</span> <span class="o">/</span> <span class="n">BM</span><span class="p">)</span>
+    <span class="n">col</span> <span class="o">=</span> <span class="n">col</span> <span class="o">+</span> <span class="n">BM</span>
+    <span class="c1"># MR</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">MR</span><span class="p">,</span> <span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">floor</span><span class="p">(</span><span class="mi">255</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">MR</span><span class="p">)</span> <span class="o">/</span> <span class="n">MR</span><span class="p">)</span>
+    <span class="n">colorwheel</span><span class="p">[</span><span class="n">col</span> <span class="p">:</span> <span class="n">col</span> <span class="o">+</span> <span class="n">MR</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="mi">255</span>
+    <span class="k">return</span> <span class="n">colorwheel</span>
+
+
+<span class="k">def</span> <span class="nf">_generate_color_palette</span><span class="p">(</span><span class="n">num_objects</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+    <span class="n">palette</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="mi">2</span><span class="o">**</span><span class="mi">25</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="o">**</span><span class="mi">15</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="o">**</span><span class="mi">21</span> <span class="o">-</span> <span class="mi">1</span><span class="p">])</span>
+    <span class="k">return</span> <span class="p">[</span><span class="nb">tuple</span><span class="p">((</span><span class="n">i</span> <span class="o">*</span> <span class="n">palette</span><span class="p">)</span> <span class="o">%</span> <span class="mi">255</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_objects</span><span class="p">)]</span>
+
+
+<span class="k">def</span> <span class="nf">_parse_colors</span><span class="p">(</span>
+    <span class="n">colors</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]]],</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">num_objects</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Parses a specification of colors for a set of objects.</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        colors: A specification of colors for the objects. This can be one of the following:</span>
+<span class="sd">            - None: to generate a color palette automatically.</span>
+<span class="sd">            - A list of colors: where each color is either a string (specifying a named color) or an RGB tuple.</span>
+<span class="sd">            - A string or an RGB tuple: to use the same color for all objects.</span>
+
+<span class="sd">            If `colors` is a tuple, it should be a 3-tuple specifying the RGB values of the color.</span>
+<span class="sd">            If `colors` is a list, it should have at least as many elements as the number of objects to color.</span>
+
+<span class="sd">        num_objects (int): The number of objects to color.</span>
+
+<span class="sd">    Returns:</span>
+<span class="sd">        A list of 3-tuples, specifying the RGB values of the colors.</span>
+
+<span class="sd">    Raises:</span>
+<span class="sd">        ValueError: If the number of colors in the list is less than the number of objects to color.</span>
+<span class="sd">                    If `colors` is not a list, tuple, string or None.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="n">colors</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">colors</span> <span class="o">=</span> <span class="n">_generate_color_palette</span><span class="p">(</span><span class="n">num_objects</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">colors</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">colors</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">num_objects</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Number of colors must be equal or larger than the number of objects, but got </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">colors</span><span class="p">)</span><span class="si">}</span><span class="s2"> &lt; </span><span class="si">{</span><span class="n">num_objects</span><span class="si">}</span><span class="s2">.&quot;</span>
+            <span class="p">)</span>
+    <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">colors</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">str</span><span class="p">)):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;`colors` must be a tuple or a string, or a list thereof, but got </span><span class="si">{colors}</span><span class="s2">.&quot;</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">colors</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">colors</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;If passed as tuple, colors should be an RGB triplet, but got </span><span class="si">{colors}</span><span class="s2">.&quot;</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>  <span class="c1"># colors specifies a single color for all objects</span>
+        <span class="n">colors</span> <span class="o">=</span> <span class="p">[</span><span class="n">colors</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_objects</span>
+
+    <span class="k">return</span> <span class="p">[</span><span class="n">ImageColor</span><span class="o">.</span><span class="n">getrgb</span><span class="p">(</span><span class="n">color</span><span class="p">)</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">color</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="k">else</span> <span class="n">color</span> <span class="k">for</span> <span class="n">color</span> <span class="ow">in</span> <span class="n">colors</span><span class="p">]</span>
+
+
+<span class="k">def</span> <span class="nf">_log_api_usage_once</span><span class="p">(</span><span class="n">obj</span><span class="p">:</span> <span class="n">Any</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Logs API usage(module and name) within an organization.</span>
+<span class="sd">    In a large ecosystem, it&#39;s often useful to track the PyTorch and</span>
+<span class="sd">    TorchVision APIs usage. This API provides the similar functionality to the</span>
+<span class="sd">    logging module in the Python stdlib. It can be used for debugging purpose</span>
+<span class="sd">    to log which methods are used and by default it is inactive, unless the user</span>
+<span class="sd">    manually subscribes a logger via the `SetAPIUsageLogger method &lt;https://github.com/pytorch/pytorch/blob/eb3b9fe719b21fae13c7a7cf3253f970290a573e/c10/util/Logging.cpp#L114&gt;`_.</span>
+<span class="sd">    Please note it is triggered only once for the same API call within a process.</span>
+<span class="sd">    It does not collect any data from open-source users since it is no-op by default.</span>
+<span class="sd">    For more information, please refer to</span>
+<span class="sd">    * PyTorch note: https://pytorch.org/docs/stable/notes/large_scale_deployments.html#api-usage-logging;</span>
+<span class="sd">    * Logging policy: https://github.com/pytorch/vision/issues/5052;</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        obj (class instance or method): an object to extract info from.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">module</span> <span class="o">=</span> <span class="n">obj</span><span class="o">.</span><span class="vm">__module__</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">module</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;torchvision&quot;</span><span class="p">):</span>
+        <span class="n">module</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;torchvision.internal.</span><span class="si">{</span><span class="n">module</span><span class="si">}</span><span class="s2">&quot;</span>
+    <span class="n">name</span> <span class="o">=</span> <span class="n">obj</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="n">FunctionType</span><span class="p">):</span>
+        <span class="n">name</span> <span class="o">=</span> <span class="n">obj</span><span class="o">.</span><span class="vm">__name__</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_C</span><span class="o">.</span><span class="n">_log_api_usage_once</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">module</span><span class="si">}</span><span class="s2">.</span><span class="si">{</span><span class="n">name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+<span class="k">def</span> <span class="nf">_make_ntuple</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">Any</span><span class="p">,</span> <span class="n">n</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="o">...</span><span class="p">]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Make n-tuple from input x. If x is an iterable, then we just convert it to tuple.</span>
+<span class="sd">    Otherwise, we will make a tuple of length n, all with value of x.</span>
+<span class="sd">    reference: https://github.com/pytorch/pytorch/blob/master/torch/nn/modules/utils.py#L8</span>
+
+<span class="sd">    Args:</span>
+<span class="sd">        x (Any): input value</span>
+<span class="sd">        n (int): length of the resulting tuple</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">collections</span><span class="o">.</span><span class="n">abc</span><span class="o">.</span><span class="n">Iterable</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+    <span class="k">return</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">repeat</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">n</span><span class="p">))</span>
+</pre></div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/_sources/auto_examples/index.rst.txt b/0.16/_sources/auto_examples/index.rst.txt
new file mode 100644
index 00000000000..1ce555eba52
--- /dev/null
+++ b/0.16/_sources/auto_examples/index.rst.txt
@@ -0,0 +1,279 @@
+:orphan:
+
+.. _gallery:
+
+Examples and tutorials
+======================
+
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbnails">
+
+
+.. raw:: html
+
+    </div>
+
+
+Transforms
+----------
+
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbnails">
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_transforms_getting_started_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_transforms_getting_started.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_transforms_illustrations_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_transforms_illustrations.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_transforms_e2e_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_transforms_e2e.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy.">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_cutmix_mixup_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_custom_transforms_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_custom_transforms.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_tv_tensors_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_custom_tv_tensors_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_custom_tv_tensors.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+    </div>
+
+
+.. raw:: html
+
+    </div>
+
+Others
+------
+
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbnails">
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_optical_flow_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_optical_flow.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_repurposing_annotations_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_repurposing_annotations.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images.">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_scripted_tensor_transforms_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_scripted_tensor_transforms.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_video_api_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_video_api.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Video API</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_visualization_utils_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_visualization_utils.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+    </div>
+
+
+.. raw:: html
+
+    </div>
+
+
+.. toctree::
+   :hidden:
+   :includehidden:
+
+
+   /auto_examples/transforms/index.rst
+   /auto_examples/others/index.rst
+
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-gallery
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download all examples in Python source code: auto_examples_python.zip </auto_examples/auto_examples_python.zip>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download all examples in Jupyter notebooks: auto_examples_jupyter.zip </auto_examples/auto_examples_jupyter.zip>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/others/index.rst.txt b/0.16/_sources/auto_examples/others/index.rst.txt
new file mode 100644
index 00000000000..0d306f82cf0
--- /dev/null
+++ b/0.16/_sources/auto_examples/others/index.rst.txt
@@ -0,0 +1,113 @@
+
+
+.. _sphx_glr_auto_examples_others:
+
+Others
+------
+
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbnails">
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_optical_flow_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_optical_flow.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_repurposing_annotations_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_repurposing_annotations.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images.">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_scripted_tensor_transforms_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_scripted_tensor_transforms.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_video_api_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_video_api.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Video API</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ...">
+
+.. only:: html
+
+  .. image:: /auto_examples/others/images/thumb/sphx_glr_plot_visualization_utils_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_others_plot_visualization_utils.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+    </div>
+
+
+.. raw:: html
+
+    </div>
+
+
+.. toctree::
+   :hidden:
+
+   /auto_examples/others/plot_optical_flow
+   /auto_examples/others/plot_repurposing_annotations
+   /auto_examples/others/plot_scripted_tensor_transforms
+   /auto_examples/others/plot_video_api
+   /auto_examples/others/plot_visualization_utils
+
diff --git a/0.16/_sources/auto_examples/others/plot_optical_flow.rst.txt b/0.16/_sources/auto_examples/others/plot_optical_flow.rst.txt
new file mode 100644
index 00000000000..86955e0ea70
--- /dev/null
+++ b/0.16/_sources/auto_examples/others/plot_optical_flow.rst.txt
@@ -0,0 +1,380 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/others/plot_optical_flow.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_others_plot_optical_flow.py:
+
+
+=====================================================
+Optical Flow: Predicting movement with the RAFT model
+=====================================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_optical_flow.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_optical_flow.py>` to download the full example code.
+
+Optical flow is the task of predicting movement between two images, usually two
+consecutive frames of a video. Optical flow models take two images as input, and
+predict a flow: the flow indicates the displacement of every single pixel in the
+first image, and maps it to its corresponding pixel in the second image. Flows
+are (2, H, W)-dimensional tensors, where the first axis corresponds to the
+predicted horizontal and vertical displacements.
+
+The following example illustrates how torchvision can be used to predict flows
+using our implementation of the RAFT model. We will also see how to convert the
+predicted flows to RGB images for visualization.
+
+.. GENERATED FROM PYTHON SOURCE LINES 21-48
+
+.. code-block:: default
+
+
+    import numpy as np
+    import torch
+    import matplotlib.pyplot as plt
+    import torchvision.transforms.functional as F
+
+
+    plt.rcParams["savefig.bbox"] = "tight"
+
+
+    def plot(imgs, **imshow_kwargs):
+        if not isinstance(imgs[0], list):
+            # Make a 2d grid even if there's just 1 row
+            imgs = [imgs]
+
+        num_rows = len(imgs)
+        num_cols = len(imgs[0])
+        _, axs = plt.subplots(nrows=num_rows, ncols=num_cols, squeeze=False)
+        for row_idx, row in enumerate(imgs):
+            for col_idx, img in enumerate(row):
+                ax = axs[row_idx, col_idx]
+                img = F.to_pil_image(img.to("cpu"))
+                ax.imshow(np.asarray(img), **imshow_kwargs)
+                ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
+
+        plt.tight_layout()
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 50-58
+
+Reading Videos Using Torchvision
+--------------------------------
+We will first read a video using :func:`~torchvision.io.read_video`.
+Alternatively one can use the new :class:`~torchvision.io.VideoReader` API (if
+torchvision is built from source).
+The video we will use here is free of use from `pexels.com
+<https://www.pexels.com/video/a-man-playing-a-game-of-basketball-5192157/>`_,
+credits go to `Pavel Danilyuk <https://www.pexels.com/@pavel-danilyuk>`_.
+
+.. GENERATED FROM PYTHON SOURCE LINES 58-69
+
+.. code-block:: default
+
+
+
+    import tempfile
+    from pathlib import Path
+    from urllib.request import urlretrieve
+
+
+    video_url = "https://download.pytorch.org/tutorial/pexelscom_pavel_danilyuk_basketball_hd.mp4"
+    video_path = Path(tempfile.mkdtemp()) / "basketball.mp4"
+    _ = urlretrieve(video_url, video_path)
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 70-77
+
+:func:`~torchvision.io.read_video` returns the video frames, audio frames and
+the metadata associated with the video. In our case, we only need the video
+frames.
+
+Here we will just make 2 predictions between 2 pre-selected pairs of frames,
+namely frames (100, 101) and (150, 151). Each of these pairs corresponds to a
+single model input.
+
+.. GENERATED FROM PYTHON SOURCE LINES 77-86
+
+.. code-block:: default
+
+
+    from torchvision.io import read_video
+    frames, _, _ = read_video(str(video_path), output_format="TCHW")
+
+    img1_batch = torch.stack([frames[100], frames[150]])
+    img2_batch = torch.stack([frames[101], frames[151]])
+
+    plot(img1_batch)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_optical_flow_001.png
+   :alt: plot optical flow
+   :srcset: /auto_examples/others/images/sphx_glr_plot_optical_flow_001.png
+   :class: sphx-glr-single-img
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    /pytorch/vision/torchvision/io/video.py:161: UserWarning: The pts_unit 'pts' gives wrong results. Please use pts_unit 'sec'.
+      warnings.warn("The pts_unit 'pts' gives wrong results. Please use pts_unit 'sec'.")
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 87-93
+
+The RAFT model accepts RGB images. We first get the frames from
+:func:`~torchvision.io.read_video` and resize them to ensure their dimensions
+are divisible by 8. Note that we explicitly use ``antialias=False``, because
+this is how those models were trained. Then we use the transforms bundled into
+the weights in order to preprocess the input and rescale its values to the
+required ``[-1, 1]`` interval.
+
+.. GENERATED FROM PYTHON SOURCE LINES 93-111
+
+.. code-block:: default
+
+
+    from torchvision.models.optical_flow import Raft_Large_Weights
+
+    weights = Raft_Large_Weights.DEFAULT
+    transforms = weights.transforms()
+
+
+    def preprocess(img1_batch, img2_batch):
+        img1_batch = F.resize(img1_batch, size=[520, 960], antialias=False)
+        img2_batch = F.resize(img2_batch, size=[520, 960], antialias=False)
+        return transforms(img1_batch, img2_batch)
+
+
+    img1_batch, img2_batch = preprocess(img1_batch, img2_batch)
+
+    print(f"shape = {img1_batch.shape}, dtype = {img1_batch.dtype}")
+
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    shape = torch.Size([2, 3, 520, 960]), dtype = torch.float32
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 112-119
+
+Estimating Optical flow using RAFT
+----------------------------------
+We will use our RAFT implementation from
+:func:`~torchvision.models.optical_flow.raft_large`, which follows the same
+architecture as the one described in the `original paper <https://arxiv.org/abs/2003.12039>`_.
+We also provide the :func:`~torchvision.models.optical_flow.raft_small` model
+builder, which is smaller and faster to run, sacrificing a bit of accuracy.
+
+.. GENERATED FROM PYTHON SOURCE LINES 119-132
+
+.. code-block:: default
+
+
+    from torchvision.models.optical_flow import raft_large
+
+    # If you can, run this example on a GPU, it will be a lot faster.
+    device = "cuda" if torch.cuda.is_available() else "cpu"
+
+    model = raft_large(weights=Raft_Large_Weights.DEFAULT, progress=False).to(device)
+    model = model.eval()
+
+    list_of_flows = model(img1_batch.to(device), img2_batch.to(device))
+    print(f"type = {type(list_of_flows)}")
+    print(f"length = {len(list_of_flows)} = number of iterations of the model")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading: "https://download.pytorch.org/models/raft_large_C_T_SKHT_V2-ff5fadd5.pth" to /root/.cache/torch/hub/checkpoints/raft_large_C_T_SKHT_V2-ff5fadd5.pth
+    type = <class 'list'>
+    length = 12 = number of iterations of the model
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 133-145
+
+The RAFT model outputs lists of predicted flows where each entry is a
+(N, 2, H, W) batch of predicted flows that corresponds to a given "iteration"
+in the model. For more details on the iterative nature of the model, please
+refer to the `original paper <https://arxiv.org/abs/2003.12039>`_. Here, we
+are only interested in the final predicted flows (they are the most accurate
+ones), so we will just retrieve the last item in the list.
+
+As described above, a flow is a tensor with dimensions (2, H, W) (or (N, 2, H,
+W) for batches of flows) where each entry corresponds to the horizontal and
+vertical displacement of each pixel from the first image to the second image.
+Note that the predicted flows are in "pixel" unit, they are not normalized
+w.r.t. the dimensions of the images.
+
+.. GENERATED FROM PYTHON SOURCE LINES 145-151
+
+.. code-block:: default
+
+    predicted_flows = list_of_flows[-1]
+    print(f"dtype = {predicted_flows.dtype}")
+    print(f"shape = {predicted_flows.shape} = (N, 2, H, W)")
+    print(f"min = {predicted_flows.min()}, max = {predicted_flows.max()}")
+
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    dtype = torch.float32
+    shape = torch.Size([2, 2, 520, 960]) = (N, 2, H, W)
+    min = -3.8997180461883545, max = 6.400400161743164
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 152-162
+
+Visualizing predicted flows
+---------------------------
+Torchvision provides the :func:`~torchvision.utils.flow_to_image` utility to
+convert a flow into an RGB image. It also supports batches of flows.
+each "direction" in the flow will be mapped to a given RGB color. In the
+images below, pixels with similar colors are assumed by the model to be moving
+in similar directions. The model is properly able to predict the movement of
+the ball and the player. Note in particular the different predicted direction
+of the ball in the first image (going to the left) and in the second image
+(going up).
+
+.. GENERATED FROM PYTHON SOURCE LINES 162-173
+
+.. code-block:: default
+
+
+    from torchvision.utils import flow_to_image
+
+    flow_imgs = flow_to_image(predicted_flows)
+
+    # The images have been mapped into [-1, 1] but for plotting we want them in [0, 1]
+    img1_batch = [(img1 + 1) / 2 for img1 in img1_batch]
+
+    grid = [[img1, flow_img] for (img1, flow_img) in zip(img1_batch, flow_imgs)]
+    plot(grid)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_optical_flow_002.png
+   :alt: plot optical flow
+   :srcset: /auto_examples/others/images/sphx_glr_plot_optical_flow_002.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 174-182
+
+Bonus: Creating GIFs of predicted flows
+---------------------------------------
+In the example above we have only shown the predicted flows of 2 pairs of
+frames. A fun way to apply the Optical Flow models is to run the model on an
+entire video, and create a new video from all the predicted flows. Below is a
+snippet that can get you started with this. We comment out the code, because
+this example is being rendered on a machine without a GPU, and it would take
+too long to run it.
+
+.. GENERATED FROM PYTHON SOURCE LINES 182-194
+
+.. code-block:: default
+
+
+    # from torchvision.io import write_jpeg
+    # for i, (img1, img2) in enumerate(zip(frames, frames[1:])):
+    #     # Note: it would be faster to predict batches of flows instead of individual flows
+    #     img1, img2 = preprocess(img1, img2)
+
+    #     list_of_flows = model(img1.to(device), img2.to(device))
+    #     predicted_flow = list_of_flows[-1][0]
+    #     flow_img = flow_to_image(predicted_flow).to("cpu")
+    #     output_folder = "/tmp/"  # Update this to the folder of your choice
+    #     write_jpeg(flow_img, output_folder + f"predicted_flow_{i}.jpg")
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 195-199
+
+Once the .jpg flow images are saved, you can convert them into a video or a
+GIF using ffmpeg with e.g.:
+
+ffmpeg -f image2 -framerate 30 -i predicted_flow_%d.jpg -loop -1 flow.gif
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 10.096 seconds)
+
+
+.. _sphx_glr_download_auto_examples_others_plot_optical_flow.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_optical_flow.py <plot_optical_flow.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_optical_flow.ipynb <plot_optical_flow.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/others/plot_repurposing_annotations.rst.txt b/0.16/_sources/auto_examples/others/plot_repurposing_annotations.rst.txt
new file mode 100644
index 00000000000..12545e36300
--- /dev/null
+++ b/0.16/_sources/auto_examples/others/plot_repurposing_annotations.rst.txt
@@ -0,0 +1,445 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/others/plot_repurposing_annotations.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_others_plot_repurposing_annotations.py:
+
+
+=====================================
+Repurposing masks into bounding boxes
+=====================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_repurposing_annotations.py>` to download the full example code.
+
+The following example illustrates the operations available
+the :ref:`torchvision.ops <ops>` module for repurposing
+segmentation masks into object localization annotations for different tasks
+(e.g. transforming masks used by instance and panoptic segmentation
+methods into bounding boxes used by object detection methods).
+
+.. GENERATED FROM PYTHON SOURCE LINES 16-42
+
+.. code-block:: default
+
+
+
+    import os
+    import numpy as np
+    import torch
+    import matplotlib.pyplot as plt
+
+    import torchvision.transforms.functional as F
+
+
+    ASSETS_DIRECTORY = "../assets"
+
+    plt.rcParams["savefig.bbox"] = "tight"
+
+
+    def show(imgs):
+        if not isinstance(imgs, list):
+            imgs = [imgs]
+        fix, axs = plt.subplots(ncols=len(imgs), squeeze=False)
+        for i, img in enumerate(imgs):
+            img = img.detach()
+            img = F.to_pil_image(img)
+            axs[0, i].imshow(np.asarray(img))
+            axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 44-59
+
+Masks
+-----
+In tasks like instance and panoptic segmentation, masks are commonly defined, and are defined by this package,
+as a multi-dimensional array (e.g. a NumPy array or a PyTorch tensor) with the following shape:
+
+      (num_objects, height, width)
+
+Where num_objects is the number of annotated objects in the image. Each (height, width) object corresponds to exactly
+one object. For example, if your input image has the dimensions 224 x 224 and has four annotated objects the shape
+of your masks annotation has the following shape:
+
+      (4, 224, 224).
+
+A nice property of masks is that they can be easily repurposed to be used in methods to solve a variety of object
+localization tasks.
+
+.. GENERATED FROM PYTHON SOURCE LINES 61-67
+
+Converting Masks to Bounding Boxes
+-----------------------------------------------
+For example, the :func:`~torchvision.ops.masks_to_boxes` operation can be used to
+transform masks into bounding boxes that can be
+used as input to detection models such as FasterRCNN and RetinaNet.
+We will take images and masks from the `PenFudan Dataset <https://www.cis.upenn.edu/~jshi/ped_html/>`_.
+
+.. GENERATED FROM PYTHON SOURCE LINES 67-77
+
+.. code-block:: default
+
+
+
+    from torchvision.io import read_image
+
+    img_path = os.path.join(ASSETS_DIRECTORY, "FudanPed00054.png")
+    mask_path = os.path.join(ASSETS_DIRECTORY, "FudanPed00054_mask.png")
+    img = read_image(img_path)
+    mask = read_image(mask_path)
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 78-81
+
+Here the masks are represented as a PNG Image, with floating point values.
+Each pixel is encoded as different colors, with 0 being background.
+Notice that the spatial dimensions of image and mask match.
+
+.. GENERATED FROM PYTHON SOURCE LINES 81-86
+
+.. code-block:: default
+
+
+    print(mask.size())
+    print(img.size())
+    print(mask)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    torch.Size([1, 498, 533])
+    torch.Size([3, 498, 533])
+    tensor([[[0, 0, 0,  ..., 0, 0, 0],
+             [0, 0, 0,  ..., 0, 0, 0],
+             [0, 0, 0,  ..., 0, 0, 0],
+             ...,
+             [0, 0, 0,  ..., 0, 0, 0],
+             [0, 0, 0,  ..., 0, 0, 0],
+             [0, 0, 0,  ..., 0, 0, 0]]], dtype=torch.uint8)
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 87-98
+
+.. code-block:: default
+
+
+    # We get the unique colors, as these would be the object ids.
+    obj_ids = torch.unique(mask)
+
+    # first id is the background, so remove it.
+    obj_ids = obj_ids[1:]
+
+    # split the color-encoded mask into a set of boolean masks.
+    # Note that this snippet would work as well if the masks were float values instead of ints.
+    masks = mask == obj_ids[:, None, None]
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 99-104
+
+Now the masks are a boolean tensor.
+The first dimension in this case 3 and denotes the number of instances: there are 3 people in the image.
+The other two dimensions are height and width, which are equal to the dimensions of the image.
+For each instance, the boolean tensors represent if the particular pixel
+belongs to the segmentation mask of the image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 104-108
+
+.. code-block:: default
+
+
+    print(masks.size())
+    print(masks)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    torch.Size([3, 498, 533])
+    tensor([[[False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             ...,
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False]],
+
+            [[False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             ...,
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False]],
+
+            [[False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             ...,
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False],
+             [False, False, False,  ..., False, False, False]]])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 109-111
+
+Let us visualize an image and plot its corresponding segmentation masks.
+We will use the :func:`~torchvision.utils.draw_segmentation_masks` to draw the segmentation masks.
+
+.. GENERATED FROM PYTHON SOURCE LINES 111-120
+
+.. code-block:: default
+
+
+    from torchvision.utils import draw_segmentation_masks
+
+    drawn_masks = []
+    for mask in masks:
+        drawn_masks.append(draw_segmentation_masks(img, mask, alpha=0.8, colors="blue"))
+
+    show(drawn_masks)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_repurposing_annotations_001.png
+   :alt: plot repurposing annotations
+   :srcset: /auto_examples/others/images/sphx_glr_plot_repurposing_annotations_001.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 121-124
+
+To convert the boolean masks into bounding boxes.
+We will use the :func:`~torchvision.ops.masks_to_boxes` from the torchvision.ops module
+It returns the boxes in ``(xmin, ymin, xmax, ymax)`` format.
+
+.. GENERATED FROM PYTHON SOURCE LINES 124-131
+
+.. code-block:: default
+
+
+    from torchvision.ops import masks_to_boxes
+
+    boxes = masks_to_boxes(masks)
+    print(boxes.size())
+    print(boxes)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    torch.Size([3, 4])
+    tensor([[ 96., 134., 181., 417.],
+            [286., 113., 357., 331.],
+            [363., 120., 436., 328.]])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 132-135
+
+As the shape denotes, there are 3 boxes and in ``(xmin, ymin, xmax, ymax)`` format.
+These can be visualized very easily with :func:`~torchvision.utils.draw_bounding_boxes` utility
+provided in :ref:`torchvision.utils <utils>`.
+
+.. GENERATED FROM PYTHON SOURCE LINES 135-141
+
+.. code-block:: default
+
+
+    from torchvision.utils import draw_bounding_boxes
+
+    drawn_boxes = draw_bounding_boxes(img, boxes, colors="red")
+    show(drawn_boxes)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_repurposing_annotations_002.png
+   :alt: plot repurposing annotations
+   :srcset: /auto_examples/others/images/sphx_glr_plot_repurposing_annotations_002.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 142-145
+
+These boxes can now directly be used by detection models in torchvision.
+Here is demo with a Faster R-CNN model loaded from
+:func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`
+
+.. GENERATED FROM PYTHON SOURCE LINES 145-160
+
+.. code-block:: default
+
+
+    from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights
+
+    weights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT
+    model = fasterrcnn_resnet50_fpn(weights=weights, progress=False)
+    print(img.size())
+
+    tranforms = weights.transforms()
+    img = tranforms(img)
+    target = {}
+    target["boxes"] = boxes
+    target["labels"] = labels = torch.ones((masks.size(0),), dtype=torch.int64)
+    detection_outputs = model(img.unsqueeze(0), [target])
+
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading: "https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth" to /root/.cache/torch/hub/checkpoints/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth
+    torch.Size([3, 498, 533])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 161-169
+
+Converting Segmentation Dataset to Detection Dataset
+----------------------------------------------------
+
+With this utility it becomes very simple to convert a segmentation dataset to a detection dataset.
+With this we can now use a segmentation dataset to train a detection model.
+One can similarly convert panoptic dataset to detection dataset.
+Here is an example where we re-purpose the dataset from the
+`PenFudan Detection Tutorial <https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html>`_.
+
+.. GENERATED FROM PYTHON SOURCE LINES 169-212
+
+.. code-block:: default
+
+
+    class SegmentationToDetectionDataset(torch.utils.data.Dataset):
+        def __init__(self, root, transforms):
+            self.root = root
+            self.transforms = transforms
+            # load all image files, sorting them to
+            # ensure that they are aligned
+            self.imgs = list(sorted(os.listdir(os.path.join(root, "PNGImages"))))
+            self.masks = list(sorted(os.listdir(os.path.join(root, "PedMasks"))))
+
+        def __getitem__(self, idx):
+            # load images and masks
+            img_path = os.path.join(self.root, "PNGImages", self.imgs[idx])
+            mask_path = os.path.join(self.root, "PedMasks", self.masks[idx])
+
+            img = read_image(img_path)
+            mask = read_image(mask_path)
+
+            img = F.convert_image_dtype(img, dtype=torch.float)
+            mask = F.convert_image_dtype(mask, dtype=torch.float)
+
+            # We get the unique colors, as these would be the object ids.
+            obj_ids = torch.unique(mask)
+
+            # first id is the background, so remove it.
+            obj_ids = obj_ids[1:]
+
+            # split the color-encoded mask into a set of boolean masks.
+            masks = mask == obj_ids[:, None, None]
+
+            boxes = masks_to_boxes(masks)
+
+            # there is only one class
+            labels = torch.ones((masks.shape[0],), dtype=torch.int64)
+
+            target = {}
+            target["boxes"] = boxes
+            target["labels"] = labels
+
+            if self.transforms is not None:
+                img, target = self.transforms(img, target)
+
+            return img, target
+
+
+
+
+
+
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 2.507 seconds)
+
+
+.. _sphx_glr_download_auto_examples_others_plot_repurposing_annotations.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_repurposing_annotations.py <plot_repurposing_annotations.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_repurposing_annotations.ipynb <plot_repurposing_annotations.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/others/plot_scripted_tensor_transforms.rst.txt b/0.16/_sources/auto_examples/others/plot_scripted_tensor_transforms.rst.txt
new file mode 100644
index 00000000000..0bb9f283227
--- /dev/null
+++ b/0.16/_sources/auto_examples/others/plot_scripted_tensor_transforms.rst.txt
@@ -0,0 +1,267 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/others/plot_scripted_tensor_transforms.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_others_plot_scripted_tensor_transforms.py:
+
+
+===================
+Torchscript support
+===================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_scripted_tensor_transforms.py>` to download the full example code.
+
+This example illustrates `torchscript
+<https://pytorch.org/docs/stable/jit.html>`_ support of the torchvision
+:ref:`transforms <transforms>` on Tensor images.
+
+.. GENERATED FROM PYTHON SOURCE LINES 16-37
+
+.. code-block:: default
+
+    from pathlib import Path
+
+    import matplotlib.pyplot as plt
+
+    import torch
+    import torch.nn as nn
+
+    import torchvision.transforms as v1
+    from torchvision.io import read_image
+
+    plt.rcParams["savefig.bbox"] = 'tight'
+    torch.manual_seed(1)
+
+    # If you're trying to run that on collab, you can download the assets and the
+    # helpers from https://github.com/pytorch/vision/tree/main/gallery/
+    import sys
+    sys.path += ["../transforms"]
+    from helpers import plot
+    ASSETS_PATH = Path('../assets')
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 38-41
+
+Most transforms support torchscript. For composing transforms, we use
+:class:`torch.nn.Sequential` instead of
+:class:`~torchvision.transforms.v2.Compose`:
+
+.. GENERATED FROM PYTHON SOURCE LINES 41-55
+
+.. code-block:: default
+
+
+    dog1 = read_image(str(ASSETS_PATH / 'dog1.jpg'))
+    dog2 = read_image(str(ASSETS_PATH / 'dog2.jpg'))
+
+    transforms = torch.nn.Sequential(
+        v1.RandomCrop(224),
+        v1.RandomHorizontalFlip(p=0.3),
+    )
+
+    scripted_transforms = torch.jit.script(transforms)
+
+    plot([dog1, scripted_transforms(dog1), dog2, scripted_transforms(dog2)])
+
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_scripted_tensor_transforms_001.png
+   :alt: plot scripted tensor transforms
+   :srcset: /auto_examples/others/images/sphx_glr_plot_scripted_tensor_transforms_001.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 56-78
+
+.. warning::
+
+    Above we have used transforms from the ``torchvision.transforms``
+    namespace, i.e. the "v1" transforms. The v2 transforms from the
+    ``torchvision.transforms.v2`` namespace are the :ref:`recommended
+    <v1_or_v2>` way to use transforms in your code.
+
+    The v2 transforms also support torchscript, but if you call
+    ``torch.jit.script()`` on a v2 **class** transform, you'll actually end up
+    with its (scripted) v1 equivalent.  This may lead to slightly different
+    results between the scripted and eager executions due to implementation
+    differences between v1 and v2.
+
+    If you really need torchscript support for the v2 transforms, **we
+    recommend scripting the functionals** from the
+    ``torchvision.transforms.v2.functional`` namespace to avoid surprises.
+
+Below we now show how to combine image transformations and a model forward
+pass, while using ``torch.jit.script`` to obtain a single scripted module.
+
+Let's define a ``Predictor`` module that transforms the input tensor and then
+applies an ImageNet model on it.
+
+.. GENERATED FROM PYTHON SOURCE LINES 78-97
+
+.. code-block:: default
+
+
+    from torchvision.models import resnet18, ResNet18_Weights
+
+
+    class Predictor(nn.Module):
+
+        def __init__(self):
+            super().__init__()
+            weights = ResNet18_Weights.DEFAULT
+            self.resnet18 = resnet18(weights=weights, progress=False).eval()
+            self.transforms = weights.transforms(antialias=True)
+
+        def forward(self, x: torch.Tensor) -> torch.Tensor:
+            with torch.no_grad():
+                x = self.transforms(x)
+                y_pred = self.resnet18(x)
+                return y_pred.argmax(dim=1)
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 98-100
+
+Now, let's define scripted and non-scripted instances of ``Predictor`` and
+apply it on multiple tensor images of the same size
+
+.. GENERATED FROM PYTHON SOURCE LINES 100-111
+
+.. code-block:: default
+
+
+    device = "cuda" if torch.cuda.is_available() else "cpu"
+
+    predictor = Predictor().to(device)
+    scripted_predictor = torch.jit.script(predictor).to(device)
+
+    batch = torch.stack([dog1, dog2]).to(device)
+
+    res = predictor(batch)
+    res_scripted = scripted_predictor(batch)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 112-114
+
+We can verify that the prediction of the scripted and non-scripted models are
+the same:
+
+.. GENERATED FROM PYTHON SOURCE LINES 114-124
+
+.. code-block:: default
+
+
+    import json
+
+    with open(Path('../assets') / 'imagenet_class_index.json') as labels_file:
+        labels = json.load(labels_file)
+
+    for i, (pred, pred_scripted) in enumerate(zip(res, res_scripted)):
+        assert pred == pred_scripted
+        print(f"Prediction for Dog {i + 1}: {labels[str(pred.item())]}")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Prediction for Dog 1: ['n02113023', 'Pembroke']
+    Prediction for Dog 2: ['n02106662', 'German_shepherd']
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 125-126
+
+Since the model is scripted, it can be easily dumped on disk and re-used
+
+.. GENERATED FROM PYTHON SOURCE LINES 126-136
+
+.. code-block:: default
+
+
+    import tempfile
+
+    with tempfile.NamedTemporaryFile() as f:
+        scripted_predictor.save(f.name)
+
+        dumped_scripted_predictor = torch.jit.load(f.name)
+        res_scripted_dumped = dumped_scripted_predictor(batch)
+    assert (res_scripted_dumped == res_scripted).all()
+
+
+
+
+
+
+
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 1.683 seconds)
+
+
+.. _sphx_glr_download_auto_examples_others_plot_scripted_tensor_transforms.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_scripted_tensor_transforms.py <plot_scripted_tensor_transforms.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_scripted_tensor_transforms.ipynb <plot_scripted_tensor_transforms.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/others/plot_video_api.rst.txt b/0.16/_sources/auto_examples/others/plot_video_api.rst.txt
new file mode 100644
index 00000000000..29ac4466480
--- /dev/null
+++ b/0.16/_sources/auto_examples/others/plot_video_api.rst.txt
@@ -0,0 +1,637 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/others/plot_video_api.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_others_plot_video_api.py:
+
+
+=========
+Video API
+=========
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_video_api.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_video_api.py>` to download the full example code.
+
+This example illustrates some of the APIs that torchvision offers for
+videos, together with the examples on how to build datasets and more.
+
+.. GENERATED FROM PYTHON SOURCE LINES 15-20
+
+1. Introduction: building a new video object and examining the properties
+-------------------------------------------------------------------------
+First we select a video to test the object out. For the sake of argument
+we're using one from kinetics400 dataset.
+To create it, we need to define the path and the stream we want to use.
+
+.. GENERATED FROM PYTHON SOURCE LINES 22-35
+
+Chosen video statistics:
+
+- WUzgd7C1pWA.mp4
+    - source:
+        - kinetics-400
+    - video:
+        - H-264
+        - MPEG-4 AVC (part 10) (avc1)
+        - fps: 29.97
+    - audio:
+        - MPEG AAC audio (mp4a)
+        - sample rate: 48K Hz
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 35-49
+
+.. code-block:: default
+
+
+    import torch
+    import torchvision
+    from torchvision.datasets.utils import download_url
+    torchvision.set_video_backend("video_reader")
+
+    # Download the sample video
+    download_url(
+        "https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true",
+        ".",
+        "WUzgd7C1pWA.mp4"
+    )
+    video_path = "./WUzgd7C1pWA.mp4"
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/WUzgd7C1pWA.mp4 to ./WUzgd7C1pWA.mp4
+
    3.7%
    7.4%
    11.1%
    14.7%
    18.4%
    22.1%
    25.8%
    29.5%
    33.2%
    36.8%
    40.5%
    44.2%
    47.9%
    51.6%
    55.3%
    58.9%
    62.6%
    66.3%
    70.0%
    73.7%
    77.4%
    81.0%
    84.7%
    88.4%
    92.1%
    95.8%
    99.5%
    100.0%
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 50-54
+
+Streams are defined in a similar fashion as torch devices. We encode them as strings in a form
+of ``stream_type:stream_id`` where ``stream_type`` is a string and ``stream_id`` a long int.
+The constructor accepts passing a ``stream_type`` only, in which case the stream is auto-discovered.
+Firstly, let's get the metadata for our particular video:
+
+.. GENERATED FROM PYTHON SOURCE LINES 54-59
+
+.. code-block:: default
+
+
+    stream = "video"
+    video = torchvision.io.VideoReader(video_path, stream)
+    video.get_metadata()
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+
+    {'video': {'duration': [10.9109], 'fps': [29.97002997002997]}, 'audio': {'duration': [10.9], 'framerate': [48000.0]}, 'subtitles': {'duration': []}, 'cc': {'duration': []}}
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 60-67
+
+Here we can see that video has two streams - a video and an audio stream.
+Currently available stream types include ['video', 'audio'].
+Each descriptor consists of two parts: stream type (e.g. 'video') and a unique stream id
+(which are determined by video encoding).
+In this way, if the video container contains multiple streams of the same type,
+users can access the one they want.
+If only stream type is passed, the decoder auto-detects first stream of that type and returns it.
+
+.. GENERATED FROM PYTHON SOURCE LINES 69-76
+
+Let's read all the frames from the video stream. By default, the return value of
+``next(video_reader)`` is a dict containing the following fields.
+
+The return fields are:
+
+- ``data``: containing a torch.tensor
+- ``pts``: containing a float timestamp of this particular frame
+
+.. GENERATED FROM PYTHON SOURCE LINES 76-92
+
+.. code-block:: default
+
+
+    metadata = video.get_metadata()
+    video.set_current_stream("audio")
+
+    frames = []  # we are going to save the frames here.
+    ptss = []  # pts is a presentation timestamp in seconds (float) of each frame
+    for frame in video:
+        frames.append(frame['data'])
+        ptss.append(frame['pts'])
+
+    print("PTS for first five frames ", ptss[:5])
+    print("Total number of frames: ", len(frames))
+    approx_nf = metadata['audio']['duration'][0] * metadata['audio']['framerate'][0]
+    print("Approx total number of datapoints we can expect: ", approx_nf)
+    print("Read data size: ", frames[0].size(0) * len(frames))
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    PTS for first five frames  [0.0, 0.021332999999999998, 0.042667, 0.064, 0.08533299999999999]
+    Total number of frames:  511
+    Approx total number of datapoints we can expect:  523200.0
+    Read data size:  523264
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 93-101
+
+But what if we only want to read certain time segment of the video?
+That can be done easily using the combination of our ``seek`` function, and the fact that each call
+to next returns the presentation timestamp of the returned frame in seconds.
+
+Given that our implementation relies on python iterators,
+we can leverage itertools to simplify the process and make it more pythonic.
+
+For example, if we wanted to read ten frames from second second:
+
+.. GENERATED FROM PYTHON SOURCE LINES 101-114
+
+.. code-block:: default
+
+
+
+    import itertools
+    video.set_current_stream("video")
+
+    frames = []  # we are going to save the frames here.
+
+    # We seek into a second second of the video and use islice to get 10 frames since
+    for frame, pts in itertools.islice(video.seek(2), 10):
+        frames.append(frame)
+
+    print("Total number of frames: ", len(frames))
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Total number of frames:  10
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 115-119
+
+Or if we wanted to read from 2nd to 5th second,
+We seek into a second second of the video,
+then we utilize the itertools takewhile to get the
+correct number of frames:
+
+.. GENERATED FROM PYTHON SOURCE LINES 119-132
+
+.. code-block:: default
+
+
+    video.set_current_stream("video")
+    frames = []  # we are going to save the frames here.
+    video = video.seek(2)
+
+    for frame in itertools.takewhile(lambda x: x['pts'] <= 5, video):
+        frames.append(frame['data'])
+
+    print("Total number of frames: ", len(frames))
+    approx_nf = (5 - 2) * video.get_metadata()['video']['fps'][0]
+    print("We can expect approx: ", approx_nf)
+    print("Tensor size: ", frames[0].size())
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Total number of frames:  90
+    We can expect approx:  89.91008991008991
+    Tensor size:  torch.Size([3, 256, 340])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 133-137
+
+2. Building a sample read_video function
+----------------------------------------------------------------------------------------
+We can utilize the methods above to build the read video function that follows
+the same API to the existing ``read_video`` function.
+
+.. GENERATED FROM PYTHON SOURCE LINES 137-177
+
+.. code-block:: default
+
+
+
+    def example_read_video(video_object, start=0, end=None, read_video=True, read_audio=True):
+        if end is None:
+            end = float("inf")
+        if end < start:
+            raise ValueError(
+                "end time should be larger than start time, got "
+                f"start time={start} and end time={end}"
+            )
+
+        video_frames = torch.empty(0)
+        video_pts = []
+        if read_video:
+            video_object.set_current_stream("video")
+            frames = []
+            for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):
+                frames.append(frame['data'])
+                video_pts.append(frame['pts'])
+            if len(frames) > 0:
+                video_frames = torch.stack(frames, 0)
+
+        audio_frames = torch.empty(0)
+        audio_pts = []
+        if read_audio:
+            video_object.set_current_stream("audio")
+            frames = []
+            for frame in itertools.takewhile(lambda x: x['pts'] <= end, video_object.seek(start)):
+                frames.append(frame['data'])
+                audio_pts.append(frame['pts'])
+            if len(frames) > 0:
+                audio_frames = torch.cat(frames, 0)
+
+        return video_frames, audio_frames, (video_pts, audio_pts), video_object.get_metadata()
+
+
+    # Total number of frames should be 327 for video and 523264 datapoints for audio
+    vf, af, info, meta = example_read_video(video)
+    print(vf.size(), af.size())
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    torch.Size([327, 3, 256, 340]) torch.Size([523264, 1])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 178-183
+
+3. Building an example randomly sampled dataset (can be applied to training dataset of kinetics400)
+-------------------------------------------------------------------------------------------------------
+Cool, so now we can use the same principle to make the sample dataset.
+We suggest trying out iterable dataset for this purpose.
+Here, we are going to build an example dataset that reads randomly selected 10 frames of video.
+
+.. GENERATED FROM PYTHON SOURCE LINES 185-186
+
+Make sample dataset
+
+.. GENERATED FROM PYTHON SOURCE LINES 186-191
+
+.. code-block:: default
+
+    import os
+    os.makedirs("./dataset", exist_ok=True)
+    os.makedirs("./dataset/1", exist_ok=True)
+    os.makedirs("./dataset/2", exist_ok=True)
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 192-193
+
+Download the videos
+
+.. GENERATED FROM PYTHON SOURCE LINES 193-219
+
+.. code-block:: default
+
+    from torchvision.datasets.utils import download_url
+    download_url(
+        "https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true",
+        "./dataset/1", "WUzgd7C1pWA.mp4"
+    )
+    download_url(
+        "https://github.com/pytorch/vision/blob/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi?raw=true",
+        "./dataset/1",
+        "RATRACE_wave_f_nm_np1_fr_goo_37.avi"
+    )
+    download_url(
+        "https://github.com/pytorch/vision/blob/main/test/assets/videos/SOX5yA1l24A.mp4?raw=true",
+        "./dataset/2",
+        "SOX5yA1l24A.mp4"
+    )
+    download_url(
+        "https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi?raw=true",
+        "./dataset/2",
+        "v_SoccerJuggling_g23_c01.avi"
+    )
+    download_url(
+        "https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi?raw=true",
+        "./dataset/2",
+        "v_SoccerJuggling_g24_c01.avi"
+    )
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/WUzgd7C1pWA.mp4 to ./dataset/1/WUzgd7C1pWA.mp4
+
    3.7%
    7.4%
    11.1%
    14.7%
    18.4%
    22.1%
    25.8%
    29.5%
    33.2%
    36.8%
    40.5%
    44.2%
    47.9%
    51.6%
    55.3%
    58.9%
    62.6%
    66.3%
    70.0%
    73.7%
    77.4%
    81.0%
    84.7%
    88.4%
    92.1%
    95.8%
    99.5%
    100.0%
+    Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi to ./dataset/1/RATRACE_wave_f_nm_np1_fr_goo_37.avi
+
    12.4%
    24.9%
    37.3%
    49.7%
    62.1%
    74.6%
    87.0%
    99.4%
    100.0%
+    Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/SOX5yA1l24A.mp4 to ./dataset/2/SOX5yA1l24A.mp4
+
    5.8%
    11.7%
    17.5%
    23.4%
    29.2%
    35.1%
    40.9%
    46.8%
    52.6%
    58.5%
    64.3%
    70.2%
    76.0%
    81.9%
    87.7%
    93.6%
    99.4%
    100.0%
+    Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi to ./dataset/2/v_SoccerJuggling_g23_c01.avi
+
    6.4%
    12.9%
    19.3%
    25.8%
    32.2%
    38.7%
    45.1%
    51.6%
    58.0%
    64.5%
    70.9%
    77.3%
    83.8%
    90.2%
    96.7%
    100.0%
+    Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi to ./dataset/2/v_SoccerJuggling_g24_c01.avi
+
    5.3%
    10.5%
    15.8%
    21.0%
    26.3%
    31.6%
    36.8%
    42.1%
    47.3%
    52.6%
    57.9%
    63.1%
    68.4%
    73.6%
    78.9%
    84.2%
    89.4%
    94.7%
    99.9%
    100.0%
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 220-221
+
+Housekeeping and utilities
+
+.. GENERATED FROM PYTHON SOURCE LINES 221-239
+
+.. code-block:: default
+
+    import os
+    import random
+
+    from torchvision.datasets.folder import make_dataset
+    from torchvision import transforms as t
+
+
+    def _find_classes(dir):
+        classes = [d.name for d in os.scandir(dir) if d.is_dir()]
+        classes.sort()
+        class_to_idx = {cls_name: i for i, cls_name in enumerate(classes)}
+        return classes, class_to_idx
+
+
+    def get_samples(root, extensions=(".mp4", ".avi")):
+        _, class_to_idx = _find_classes(root)
+        return make_dataset(root, class_to_idx, extensions=extensions)
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 240-250
+
+We are going to define the dataset and some basic arguments.
+We assume the structure of the FolderDataset, and add the following parameters:
+
+- ``clip_len``: length of a clip in frames
+- ``frame_transform``: transform for every frame individually
+- ``video_transform``: transform on a video sequence
+
+.. note::
+  We actually add epoch size as using :func:`~torch.utils.data.IterableDataset`
+  class allows us to naturally oversample clips or images from each video if needed.
+
+.. GENERATED FROM PYTHON SOURCE LINES 250-294
+
+.. code-block:: default
+
+
+
+    class RandomDataset(torch.utils.data.IterableDataset):
+        def __init__(self, root, epoch_size=None, frame_transform=None, video_transform=None, clip_len=16):
+            super(RandomDataset).__init__()
+
+            self.samples = get_samples(root)
+
+            # Allow for temporal jittering
+            if epoch_size is None:
+                epoch_size = len(self.samples)
+            self.epoch_size = epoch_size
+
+            self.clip_len = clip_len
+            self.frame_transform = frame_transform
+            self.video_transform = video_transform
+
+        def __iter__(self):
+            for i in range(self.epoch_size):
+                # Get random sample
+                path, target = random.choice(self.samples)
+                # Get video object
+                vid = torchvision.io.VideoReader(path, "video")
+                metadata = vid.get_metadata()
+                video_frames = []  # video frame buffer
+
+                # Seek and return frames
+                max_seek = metadata["video"]['duration'][0] - (self.clip_len / metadata["video"]['fps'][0])
+                start = random.uniform(0., max_seek)
+                for frame in itertools.islice(vid.seek(start), self.clip_len):
+                    video_frames.append(self.frame_transform(frame['data']))
+                    current_pts = frame['pts']
+                # Stack it into a tensor
+                video = torch.stack(video_frames, 0)
+                if self.video_transform:
+                    video = self.video_transform(video)
+                output = {
+                    'path': path,
+                    'video': video,
+                    'target': target,
+                    'start': start,
+                    'end': current_pts}
+                yield output
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 295-309
+
+Given a path of videos in a folder structure, i.e:
+
+- dataset
+    - class 1
+        - file 0
+        - file 1
+        - ...
+    - class 2
+        - file 0
+        - file 1
+        - ...
+    - ...
+
+We can generate a dataloader and test the dataset.
+
+.. GENERATED FROM PYTHON SOURCE LINES 309-316
+
+.. code-block:: default
+
+
+
+    transforms = [t.Resize((112, 112))]
+    frame_transform = t.Compose(transforms)
+
+    dataset = RandomDataset("./dataset", epoch_size=None, frame_transform=frame_transform)
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 317-328
+
+.. code-block:: default
+
+    from torch.utils.data import DataLoader
+    loader = DataLoader(dataset, batch_size=12)
+    data = {"video": [], 'start': [], 'end': [], 'tensorsize': []}
+    for batch in loader:
+        for i in range(len(batch['path'])):
+            data['video'].append(batch['path'][i])
+            data['start'].append(batch['start'][i].item())
+            data['end'].append(batch['end'][i].item())
+            data['tensorsize'].append(batch['video'][i].size())
+    print(data)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    /pytorch/vision/torchvision/transforms/functional.py:1603: UserWarning: The default value of the antialias parameter of all the resizing transforms (Resize(), RandomResizedCrop(), etc.) will change from None to True in v0.17, in order to be consistent across the PIL and Tensor backends. To suppress this warning, directly pass antialias=True (recommended, future default), antialias=None (current default, which means False for Tensors and True for PIL), or antialias=False (only works on Tensors - PIL will still use antialiasing). This also applies if you are using the inference transforms from the models weights: update the call to weights.transforms(antialias=True).
+      warnings.warn(
+    {'video': ['./dataset/1/RATRACE_wave_f_nm_np1_fr_goo_37.avi', './dataset/2/SOX5yA1l24A.mp4', './dataset/2/SOX5yA1l24A.mp4', './dataset/2/v_SoccerJuggling_g24_c01.avi', './dataset/2/v_SoccerJuggling_g23_c01.avi'], 'start': [1.2933978369482158, 1.7918684185786267, 4.373791147055652, 3.9307513195312067, 3.6426948895065263], 'end': [1.7999999999999998, 2.3023, 4.9049, 4.437767, 4.170833], 'tensorsize': [torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112])]}
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 329-332
+
+4. Data Visualization
+----------------------------------
+Example of visualized video
+
+.. GENERATED FROM PYTHON SOURCE LINES 332-341
+
+.. code-block:: default
+
+
+    import matplotlib.pyplot as plt
+
+    plt.figure(figsize=(12, 12))
+    for i in range(16):
+        plt.subplot(4, 4, i + 1)
+        plt.imshow(batch["video"][0, i, ...].permute(1, 2, 0))
+        plt.axis("off")
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_video_api_001.png
+   :alt: plot video api
+   :srcset: /auto_examples/others/images/sphx_glr_plot_video_api_001.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 342-343
+
+Cleanup the video and dataset:
+
+.. GENERATED FROM PYTHON SOURCE LINES 343-347
+
+.. code-block:: default
+
+    import os
+    import shutil
+    os.remove("./WUzgd7C1pWA.mp4")
+    shutil.rmtree("./dataset")
+
+
+
+
+
+
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 5.277 seconds)
+
+
+.. _sphx_glr_download_auto_examples_others_plot_video_api.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_video_api.py <plot_video_api.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_video_api.ipynb <plot_video_api.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/others/plot_visualization_utils.rst.txt b/0.16/_sources/auto_examples/others/plot_visualization_utils.rst.txt
new file mode 100644
index 00000000000..4a3de070691
--- /dev/null
+++ b/0.16/_sources/auto_examples/others/plot_visualization_utils.rst.txt
@@ -0,0 +1,1092 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/others/plot_visualization_utils.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_others_plot_visualization_utils.py:
+
+
+=======================
+Visualization utilities
+=======================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_visualization_utils.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_others_plot_visualization_utils.py>` to download the full example code.
+
+This example illustrates some of the utilities that torchvision offers for
+visualizing images, bounding boxes, segmentation masks and keypoints.
+
+.. GENERATED FROM PYTHON SOURCE LINES 13-36
+
+.. code-block:: default
+
+
+
+    import torch
+    import numpy as np
+    import matplotlib.pyplot as plt
+
+    import torchvision.transforms.functional as F
+
+
+    plt.rcParams["savefig.bbox"] = 'tight'
+
+
+    def show(imgs):
+        if not isinstance(imgs, list):
+            imgs = [imgs]
+        fig, axs = plt.subplots(ncols=len(imgs), squeeze=False)
+        for i, img in enumerate(imgs):
+            img = img.detach()
+            img = F.to_pil_image(img)
+            axs[0, i].imshow(np.asarray(img))
+            axs[0, i].set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 38-43
+
+Visualizing a grid of images
+----------------------------
+The :func:`~torchvision.utils.make_grid` function can be used to create a
+tensor that represents multiple images in a grid.  This util requires a single
+image of dtype ``uint8`` as input.
+
+.. GENERATED FROM PYTHON SOURCE LINES 43-55
+
+.. code-block:: default
+
+
+    from torchvision.utils import make_grid
+    from torchvision.io import read_image
+    from pathlib import Path
+
+    dog1_int = read_image(str(Path('../assets') / 'dog1.jpg'))
+    dog2_int = read_image(str(Path('../assets') / 'dog2.jpg'))
+    dog_list = [dog1_int, dog2_int]
+
+    grid = make_grid(dog_list)
+    show(grid)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_001.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_001.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 56-61
+
+Visualizing bounding boxes
+--------------------------
+We can use :func:`~torchvision.utils.draw_bounding_boxes` to draw boxes on an
+image. We can set the colors, labels, width as well as font and font size.
+The boxes are in ``(xmin, ymin, xmax, ymax)`` format.
+
+.. GENERATED FROM PYTHON SOURCE LINES 61-71
+
+.. code-block:: default
+
+
+    from torchvision.utils import draw_bounding_boxes
+
+
+    boxes = torch.tensor([[50, 50, 100, 200], [210, 150, 350, 430]], dtype=torch.float)
+    colors = ["blue", "yellow"]
+    result = draw_bounding_boxes(dog1_int, boxes, colors=colors, width=5)
+    show(result)
+
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_002.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_002.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 72-77
+
+Naturally, we can also plot bounding boxes produced by torchvision detection
+models.  Here is a demo with a Faster R-CNN model loaded from
+:func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`
+model. For more details on the output of such models, you may
+refer to :ref:`instance_seg_output`.
+
+.. GENERATED FROM PYTHON SOURCE LINES 77-92
+
+.. code-block:: default
+
+
+    from torchvision.models.detection import fasterrcnn_resnet50_fpn, FasterRCNN_ResNet50_FPN_Weights
+
+
+    weights = FasterRCNN_ResNet50_FPN_Weights.DEFAULT
+    transforms = weights.transforms()
+
+    images = [transforms(d) for d in dog_list]
+
+    model = fasterrcnn_resnet50_fpn(weights=weights, progress=False)
+    model = model.eval()
+
+    outputs = model(images)
+    print(outputs)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    [{'boxes': tensor([[215.9767, 171.1661, 402.0078, 378.7391],
+            [344.6341, 172.6735, 357.6114, 220.1435],
+            [153.1306, 185.5567, 172.9223, 254.7014]], grad_fn=<StackBackward0>), 'labels': tensor([18,  1,  1]), 'scores': tensor([0.9989, 0.0701, 0.0611], grad_fn=<IndexBackward0>)}, {'boxes': tensor([[ 23.5964, 132.4331, 449.9359, 493.0222],
+            [225.8182, 124.6292, 467.2861, 492.2621],
+            [ 18.5248, 135.4171, 420.9786, 479.2225]], grad_fn=<StackBackward0>), 'labels': tensor([18, 18, 17]), 'scores': tensor([0.9980, 0.0879, 0.0671], grad_fn=<IndexBackward0>)}]
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 93-95
+
+Let's plot the boxes detected by our model. We will only plot the boxes with a
+score greater than a given threshold.
+
+.. GENERATED FROM PYTHON SOURCE LINES 95-103
+
+.. code-block:: default
+
+
+    score_threshold = .8
+    dogs_with_boxes = [
+        draw_bounding_boxes(dog_int, boxes=output['boxes'][output['scores'] > score_threshold], width=4)
+        for dog_int, output in zip(dog_list, outputs)
+    ]
+    show(dogs_with_boxes)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_003.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_003.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 104-119
+
+Visualizing segmentation masks
+------------------------------
+The :func:`~torchvision.utils.draw_segmentation_masks` function can be used to
+draw segmentation masks on images. Semantic segmentation and instance
+segmentation models have different outputs, so we will treat each
+independently.
+
+.. _semantic_seg_output:
+
+Semantic segmentation models
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+We will see how to use it with torchvision's FCN Resnet-50, loaded with
+:func:`~torchvision.models.segmentation.fcn_resnet50`. Let's start by looking
+at the output of the model.
+
+.. GENERATED FROM PYTHON SOURCE LINES 119-132
+
+.. code-block:: default
+
+
+    from torchvision.models.segmentation import fcn_resnet50, FCN_ResNet50_Weights
+
+    weights = FCN_ResNet50_Weights.DEFAULT
+    transforms = weights.transforms(resize_size=None)
+
+    model = fcn_resnet50(weights=weights, progress=False)
+    model = model.eval()
+
+    batch = torch.stack([transforms(d) for d in dog_list])
+    output = model(batch)['out']
+    print(output.shape, output.min().item(), output.max().item())
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading: "https://download.pytorch.org/models/fcn_resnet50_coco-1167a1af.pth" to /root/.cache/torch/hub/checkpoints/fcn_resnet50_coco-1167a1af.pth
+    torch.Size([2, 21, 500, 500]) -7.089669704437256 14.858257293701172
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 133-141
+
+As we can see above, the output of the segmentation model is a tensor of shape
+``(batch_size, num_classes, H, W)``. Each value is a non-normalized score, and
+we can normalize them into ``[0, 1]`` by using a softmax. After the softmax,
+we can interpret each value as a probability indicating how likely a given
+pixel is to belong to a given class.
+
+Let's plot the masks that have been detected for the dog class and for the
+boat class:
+
+.. GENERATED FROM PYTHON SOURCE LINES 141-154
+
+.. code-block:: default
+
+
+    sem_class_to_idx = {cls: idx for (idx, cls) in enumerate(weights.meta["categories"])}
+
+    normalized_masks = torch.nn.functional.softmax(output, dim=1)
+
+    dog_and_boat_masks = [
+        normalized_masks[img_idx, sem_class_to_idx[cls]]
+        for img_idx in range(len(dog_list))
+        for cls in ('dog', 'boat')
+    ]
+
+    show(dog_and_boat_masks)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_004.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_004.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 155-162
+
+As expected, the model is confident about the dog class, but not so much for
+the boat class.
+
+The :func:`~torchvision.utils.draw_segmentation_masks` function can be used to
+plots those masks on top of the original image. This function expects the
+masks to be boolean masks, but our masks above contain probabilities in ``[0,
+1]``. To get boolean masks, we can do the following:
+
+.. GENERATED FROM PYTHON SOURCE LINES 162-169
+
+.. code-block:: default
+
+
+    class_dim = 1
+    boolean_dog_masks = (normalized_masks.argmax(class_dim) == sem_class_to_idx['dog'])
+    print(f"shape = {boolean_dog_masks.shape}, dtype = {boolean_dog_masks.dtype}")
+    show([m.float() for m in boolean_dog_masks])
+
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_005.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_005.png
+   :class: sphx-glr-single-img
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    shape = torch.Size([2, 500, 500]), dtype = torch.bool
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 170-182
+
+The line above where we define ``boolean_dog_masks`` is a bit cryptic, but you
+can read it as the following query: "For which pixels is 'dog' the most likely
+class?"
+
+.. note::
+  While we're using the ``normalized_masks`` here, we would have
+  gotten the same result by using the non-normalized scores of the model
+  directly (as the softmax operation preserves the order).
+
+Now that we have boolean masks, we can use them with
+:func:`~torchvision.utils.draw_segmentation_masks` to plot them on top of the
+original images:
+
+.. GENERATED FROM PYTHON SOURCE LINES 182-191
+
+.. code-block:: default
+
+
+    from torchvision.utils import draw_segmentation_masks
+
+    dogs_with_masks = [
+        draw_segmentation_masks(img, masks=mask, alpha=0.7)
+        for img, mask in zip(dog_list, boolean_dog_masks)
+    ]
+    show(dogs_with_masks)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_006.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_006.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 192-199
+
+We can plot more than one mask per image! Remember that the model returned as
+many masks as there are classes. Let's ask the same query as above, but this
+time for *all* classes, not just the dog class: "For each pixel and each class
+C, is class C the most likely class?"
+
+This one is a bit more involved, so we'll first show how to do it with a
+single image, and then we'll generalize to the batch
+
+.. GENERATED FROM PYTHON SOURCE LINES 199-211
+
+.. code-block:: default
+
+
+    num_classes = normalized_masks.shape[1]
+    dog1_masks = normalized_masks[0]
+    class_dim = 0
+    dog1_all_classes_masks = dog1_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None]
+
+    print(f"dog1_masks shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}")
+    print(f"dog1_all_classes_masks = {dog1_all_classes_masks.shape}, dtype = {dog1_all_classes_masks.dtype}")
+
+    dog_with_all_masks = draw_segmentation_masks(dog1_int, masks=dog1_all_classes_masks, alpha=.6)
+    show(dog_with_all_masks)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_007.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_007.png
+   :class: sphx-glr-single-img
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    dog1_masks shape = torch.Size([21, 500, 500]), dtype = torch.float32
+    dog1_all_classes_masks = torch.Size([21, 500, 500]), dtype = torch.bool
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 212-224
+
+We can see in the image above that only 2 masks were drawn: the mask for the
+background and the mask for the dog. This is because the model thinks that
+only these 2 classes are the most likely ones across all the pixels. If the
+model had detected another class as the most likely among other pixels, we
+would have seen its mask above.
+
+Removing the background mask is as simple as passing
+``masks=dog1_all_classes_masks[1:]``, because the background class is the
+class with index 0.
+
+Let's now do the same but for an entire batch of images. The code is similar
+but involves a bit more juggling with the dimensions.
+
+.. GENERATED FROM PYTHON SOURCE LINES 224-238
+
+.. code-block:: default
+
+
+    class_dim = 1
+    all_classes_masks = normalized_masks.argmax(class_dim) == torch.arange(num_classes)[:, None, None, None]
+    print(f"shape = {all_classes_masks.shape}, dtype = {all_classes_masks.dtype}")
+    # The first dimension is the classes now, so we need to swap it
+    all_classes_masks = all_classes_masks.swapaxes(0, 1)
+
+    dogs_with_masks = [
+        draw_segmentation_masks(img, masks=mask, alpha=.6)
+        for img, mask in zip(dog_list, all_classes_masks)
+    ]
+    show(dogs_with_masks)
+
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_008.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_008.png
+   :class: sphx-glr-single-img
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    shape = torch.Size([21, 2, 500, 500]), dtype = torch.bool
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 239-258
+
+.. _instance_seg_output:
+
+Instance segmentation models
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Instance segmentation models have a significantly different output from the
+semantic segmentation models. We will see here how to plot the masks for such
+models. Let's start by analyzing the output of a Mask-RCNN model. Note that
+these models don't require the images to be normalized, so we don't need to
+use the normalized batch.
+
+.. note::
+
+    We will here describe the output of a Mask-RCNN model. The models in
+    :ref:`object_det_inst_seg_pers_keypoint_det` all have a similar output
+    format, but some of them may have extra info like keypoints for
+    :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`, and some
+    of them may not have masks, like
+    :func:`~torchvision.models.detection.fasterrcnn_resnet50_fpn`.
+
+.. GENERATED FROM PYTHON SOURCE LINES 258-272
+
+.. code-block:: default
+
+
+    from torchvision.models.detection import maskrcnn_resnet50_fpn, MaskRCNN_ResNet50_FPN_Weights
+
+    weights = MaskRCNN_ResNet50_FPN_Weights.DEFAULT
+    transforms = weights.transforms()
+
+    images = [transforms(d) for d in dog_list]
+
+    model = maskrcnn_resnet50_fpn(weights=weights, progress=False)
+    model = model.eval()
+
+    output = model(images)
+    print(output)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /root/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
+    [{'boxes': tensor([[219.7444, 168.1722, 400.7378, 384.0263],
+            [343.9716, 171.2287, 358.3447, 222.6263],
+            [301.0303, 192.6917, 313.8879, 232.3154]], grad_fn=<StackBackward0>), 'labels': tensor([18,  1,  1]), 'scores': tensor([0.9987, 0.7187, 0.6525], grad_fn=<IndexBackward0>), 'masks': tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              ...,
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.]]],
+
+
+            [[[0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              ...,
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.]]],
+
+
+            [[[0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              ...,
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.]]]], grad_fn=<UnsqueezeBackward0>)}, {'boxes': tensor([[ 44.6767, 137.9018, 446.5324, 487.3429],
+            [  0.0000, 288.0053, 489.9292, 490.2352]], grad_fn=<StackBackward0>), 'labels': tensor([18, 15]), 'scores': tensor([0.9978, 0.0697], grad_fn=<IndexBackward0>), 'masks': tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              ...,
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.]]],
+
+
+            [[[0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              ...,
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.],
+              [0., 0., 0.,  ..., 0., 0., 0.]]]], grad_fn=<UnsqueezeBackward0>)}]
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 273-289
+
+Let's break this down. For each image in the batch, the model outputs some
+detections (or instances). The number of detections varies for each input
+image. Each instance is described by its bounding box, its label, its score
+and its mask.
+
+The way the output is organized is as follows: the output is a list of length
+``batch_size``. Each entry in the list corresponds to an input image, and it
+is a dict with keys 'boxes', 'labels', 'scores', and 'masks'. Each value
+associated to those keys has ``num_instances`` elements in it.  In our case
+above there are 3 instances detected in the first image, and 2 instances in
+the second one.
+
+The boxes can be plotted with :func:`~torchvision.utils.draw_bounding_boxes`
+as above, but here we're more interested in the masks. These masks are quite
+different from the masks that we saw above for the semantic segmentation
+models.
+
+.. GENERATED FROM PYTHON SOURCE LINES 289-295
+
+.. code-block:: default
+
+
+    dog1_output = output[0]
+    dog1_masks = dog1_output['masks']
+    print(f"shape = {dog1_masks.shape}, dtype = {dog1_masks.dtype}, "
+          f"min = {dog1_masks.min()}, max = {dog1_masks.max()}")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    shape = torch.Size([3, 1, 500, 500]), dtype = torch.float32, min = 0.0, max = 0.9999862909317017
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 296-300
+
+Here the masks correspond to probabilities indicating, for each pixel, how
+likely it is to belong to the predicted label of that instance. Those
+predicted labels correspond to the 'labels' element in the same output dict.
+Let's see which labels were predicted for the instances of the first image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 300-304
+
+.. code-block:: default
+
+
+    print("For the first dog, the following instances were detected:")
+    print([weights.meta["categories"][label] for label in dog1_output['labels']])
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    For the first dog, the following instances were detected:
+    ['dog', 'person', 'person']
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 305-312
+
+Interestingly, the model detects two persons in the image. Let's go ahead and
+plot those masks. Since :func:`~torchvision.utils.draw_segmentation_masks`
+expects boolean masks, we need to convert those probabilities into boolean
+values. Remember that the semantic of those masks is "How likely is this pixel
+to belong to the predicted class?". As a result, a natural way of converting
+those masks into boolean values is to threshold them with the 0.5 probability
+(one could also choose a different threshold).
+
+.. GENERATED FROM PYTHON SOURCE LINES 312-322
+
+.. code-block:: default
+
+
+    proba_threshold = 0.5
+    dog1_bool_masks = dog1_output['masks'] > proba_threshold
+    print(f"shape = {dog1_bool_masks.shape}, dtype = {dog1_bool_masks.dtype}")
+
+    # There's an extra dimension (1) to the masks. We need to remove it
+    dog1_bool_masks = dog1_bool_masks.squeeze(1)
+
+    show(draw_segmentation_masks(dog1_int, dog1_bool_masks, alpha=0.9))
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_009.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_009.png
+   :class: sphx-glr-single-img
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    shape = torch.Size([3, 1, 500, 500]), dtype = torch.bool
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 323-326
+
+The model seems to have properly detected the dog, but it also confused trees
+with people. Looking more closely at the scores will help us plot more
+relevant masks:
+
+.. GENERATED FROM PYTHON SOURCE LINES 326-329
+
+.. code-block:: default
+
+
+    print(dog1_output['scores'])
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    tensor([0.9987, 0.7187, 0.6525], grad_fn=<IndexBackward0>)
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 330-334
+
+Clearly the model is more confident about the dog detection than it is about
+the people detections. That's good news. When plotting the masks, we can ask
+for only those that have a good score. Let's use a score threshold of .75
+here, and also plot the masks of the second dog.
+
+.. GENERATED FROM PYTHON SOURCE LINES 334-348
+
+.. code-block:: default
+
+
+    score_threshold = .75
+
+    boolean_masks = [
+        out['masks'][out['scores'] > score_threshold] > proba_threshold
+        for out in output
+    ]
+
+    dogs_with_masks = [
+        draw_segmentation_masks(img, mask.squeeze(1))
+        for img, mask in zip(dog_list, boolean_masks)
+    ]
+    show(dogs_with_masks)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_010.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_010.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 349-352
+
+The two 'people' masks in the first image where not selected because they have
+a lower score than the score threshold. Similarly, in the second image, the
+instance with class 15 (which corresponds to 'bench') was not selected.
+
+.. GENERATED FROM PYTHON SOURCE LINES 354-363
+
+.. _keypoint_output:
+
+Visualizing keypoints
+------------------------------
+The :func:`~torchvision.utils.draw_keypoints` function can be used to
+draw keypoints on images. We will see how to use it with
+torchvision's KeypointRCNN loaded with :func:`~torchvision.models.detection.keypointrcnn_resnet50_fpn`.
+We will first have a look at output of the model.
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 363-380
+
+.. code-block:: default
+
+
+    from torchvision.models.detection import keypointrcnn_resnet50_fpn, KeypointRCNN_ResNet50_FPN_Weights
+    from torchvision.io import read_image
+
+    person_int = read_image(str(Path("../assets") / "person1.jpg"))
+
+    weights = KeypointRCNN_ResNet50_FPN_Weights.DEFAULT
+    transforms = weights.transforms()
+
+    person_float = transforms(person_int)
+
+    model = keypointrcnn_resnet50_fpn(weights=weights, progress=False)
+    model = model.eval()
+
+    outputs = model([person_float])
+    print(outputs)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Downloading: "https://download.pytorch.org/models/keypointrcnn_resnet50_fpn_coco-fc266e95.pth" to /root/.cache/torch/hub/checkpoints/keypointrcnn_resnet50_fpn_coco-fc266e95.pth
+    [{'boxes': tensor([[124.3751, 177.9242, 327.6354, 574.7064],
+            [124.3625, 180.7574, 290.1061, 390.7958]], grad_fn=<StackBackward0>), 'labels': tensor([1, 1]), 'scores': tensor([0.9998, 0.1070], grad_fn=<IndexBackward0>), 'keypoints': tensor([[[208.0176, 214.2408,   1.0000],
+             [208.0176, 207.0375,   1.0000],
+             [197.8246, 210.6392,   1.0000],
+             [208.0176, 211.8398,   1.0000],
+             [178.6378, 217.8425,   1.0000],
+             [221.2086, 253.8590,   1.0000],
+             [160.6502, 269.4662,   1.0000],
+             [243.9929, 304.2822,   1.0000],
+             [138.4654, 328.8935,   1.0000],
+             [277.5698, 340.8990,   1.0000],
+             [153.4551, 374.5144,   1.0000],
+             [226.0053, 375.7150,   1.0000],
+             [226.0053, 370.3125,   1.0000],
+             [221.8082, 455.5516,   1.0000],
+             [273.9723, 448.9486,   1.0000],
+             [193.6275, 546.1932,   1.0000],
+             [273.3727, 545.5930,   1.0000]],
+
+            [[207.8327, 214.6636,   1.0000],
+             [207.2343, 207.4622,   1.0000],
+             [198.2590, 209.8627,   1.0000],
+             [208.4310, 210.4628,   1.0000],
+             [178.5134, 218.2642,   1.0000],
+             [219.7997, 251.8704,   1.0000],
+             [162.3579, 269.2736,   1.0000],
+             [245.5289, 304.6800,   1.0000],
+             [138.4238, 330.4848,   1.0000],
+             [278.4382, 346.0876,   1.0000],
+             [153.3826, 374.8929,   1.0000],
+             [233.5618, 368.2917,   1.0000],
+             [225.7832, 367.6916,   1.0000],
+             [289.8069, 357.4897,   1.0000],
+             [245.5289, 389.8956,   1.0000],
+             [281.4300, 349.0882,   1.0000],
+             [209.0294, 389.8956,   1.0000]]], grad_fn=<CopySlices>), 'keypoints_scores': tensor([[16.0164, 16.6672, 15.8312,  4.6510, 14.2053,  8.8280,  9.1136, 12.2084,
+             12.1901, 13.8453, 10.7090,  5.5852,  7.5005, 11.3378,  9.3700,  8.2987,
+              8.4479],
+            [12.9326, 13.8158, 14.9053,  3.9368, 12.9585,  6.4240,  6.8328, 10.4227,
+              9.2907, 10.1066, 10.1019,  0.1822,  4.3057, -4.9904, -2.7409, -2.7874,
+             -3.9329]], grad_fn=<CopySlices>)}]
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 381-388
+
+As we see the output contains a list of dictionaries.
+The output list is of length batch_size.
+We currently have just a single image so length of list is 1.
+Each entry in the list corresponds to an input image,
+and it is a dict with keys `boxes`, `labels`, `scores`, `keypoints` and `keypoint_scores`.
+Each value associated to those keys has `num_instances` elements in it.
+In our case above there are 2 instances detected in the image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 388-395
+
+.. code-block:: default
+
+
+    kpts = outputs[0]['keypoints']
+    scores = outputs[0]['scores']
+
+    print(kpts)
+    print(scores)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    tensor([[[208.0176, 214.2408,   1.0000],
+             [208.0176, 207.0375,   1.0000],
+             [197.8246, 210.6392,   1.0000],
+             [208.0176, 211.8398,   1.0000],
+             [178.6378, 217.8425,   1.0000],
+             [221.2086, 253.8590,   1.0000],
+             [160.6502, 269.4662,   1.0000],
+             [243.9929, 304.2822,   1.0000],
+             [138.4654, 328.8935,   1.0000],
+             [277.5698, 340.8990,   1.0000],
+             [153.4551, 374.5144,   1.0000],
+             [226.0053, 375.7150,   1.0000],
+             [226.0053, 370.3125,   1.0000],
+             [221.8082, 455.5516,   1.0000],
+             [273.9723, 448.9486,   1.0000],
+             [193.6275, 546.1932,   1.0000],
+             [273.3727, 545.5930,   1.0000]],
+
+            [[207.8327, 214.6636,   1.0000],
+             [207.2343, 207.4622,   1.0000],
+             [198.2590, 209.8627,   1.0000],
+             [208.4310, 210.4628,   1.0000],
+             [178.5134, 218.2642,   1.0000],
+             [219.7997, 251.8704,   1.0000],
+             [162.3579, 269.2736,   1.0000],
+             [245.5289, 304.6800,   1.0000],
+             [138.4238, 330.4848,   1.0000],
+             [278.4382, 346.0876,   1.0000],
+             [153.3826, 374.8929,   1.0000],
+             [233.5618, 368.2917,   1.0000],
+             [225.7832, 367.6916,   1.0000],
+             [289.8069, 357.4897,   1.0000],
+             [245.5289, 389.8956,   1.0000],
+             [281.4300, 349.0882,   1.0000],
+             [209.0294, 389.8956,   1.0000]]], grad_fn=<CopySlices>)
+    tensor([0.9998, 0.1070], grad_fn=<IndexBackward0>)
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 396-402
+
+The KeypointRCNN model detects there are two instances in the image.
+If you plot the boxes by using :func:`~draw_bounding_boxes`
+you would recognize they are the person and the surfboard.
+If we look at the scores, we will realize that the model is much more confident about the person than surfboard.
+We could now set a threshold confidence and plot instances which we are confident enough.
+Let us set a threshold of 0.75 and filter out the keypoints corresponding to the person.
+
+.. GENERATED FROM PYTHON SOURCE LINES 402-409
+
+.. code-block:: default
+
+
+    detect_threshold = 0.75
+    idx = torch.where(scores > detect_threshold)
+    keypoints = kpts[idx]
+
+    print(keypoints)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    tensor([[[208.0176, 214.2408,   1.0000],
+             [208.0176, 207.0375,   1.0000],
+             [197.8246, 210.6392,   1.0000],
+             [208.0176, 211.8398,   1.0000],
+             [178.6378, 217.8425,   1.0000],
+             [221.2086, 253.8590,   1.0000],
+             [160.6502, 269.4662,   1.0000],
+             [243.9929, 304.2822,   1.0000],
+             [138.4654, 328.8935,   1.0000],
+             [277.5698, 340.8990,   1.0000],
+             [153.4551, 374.5144,   1.0000],
+             [226.0053, 375.7150,   1.0000],
+             [226.0053, 370.3125,   1.0000],
+             [221.8082, 455.5516,   1.0000],
+             [273.9723, 448.9486,   1.0000],
+             [193.6275, 546.1932,   1.0000],
+             [273.3727, 545.5930,   1.0000]]], grad_fn=<IndexBackward0>)
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 410-414
+
+Great, now we have the keypoints corresponding to the person.
+Each keypoint is represented by x, y coordinates and the visibility.
+We can now use the :func:`~torchvision.utils.draw_keypoints` function to draw keypoints.
+Note that the utility expects uint8 images.
+
+.. GENERATED FROM PYTHON SOURCE LINES 414-420
+
+.. code-block:: default
+
+
+    from torchvision.utils import draw_keypoints
+
+    res = draw_keypoints(person_int, keypoints, colors="blue", radius=3)
+    show(res)
+
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_011.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_011.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 421-423
+
+As we see the keypoints appear as colored circles over the image.
+The coco keypoints for a person are ordered and represent the following list.\
+
+.. GENERATED FROM PYTHON SOURCE LINES 423-431
+
+.. code-block:: default
+
+
+    coco_keypoints = [
+        "nose", "left_eye", "right_eye", "left_ear", "right_ear",
+        "left_shoulder", "right_shoulder", "left_elbow", "right_elbow",
+        "left_wrist", "right_wrist", "left_hip", "right_hip",
+        "left_knee", "right_knee", "left_ankle", "right_ankle",
+    ]
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 432-451
+
+What if we are interested in joining the keypoints?
+This is especially useful in creating pose detection or action recognition.
+We can join the keypoints easily using the `connectivity` parameter.
+A close observation would reveal that we would need to join the points in below
+order to construct human skeleton.
+
+nose -> left_eye -> left_ear.                              (0, 1), (1, 3)
+
+nose -> right_eye -> right_ear.                            (0, 2), (2, 4)
+
+nose -> left_shoulder -> left_elbow -> left_wrist.         (0, 5), (5, 7), (7, 9)
+
+nose -> right_shoulder -> right_elbow -> right_wrist.      (0, 6), (6, 8), (8, 10)
+
+left_shoulder -> left_hip -> left_knee -> left_ankle.      (5, 11), (11, 13), (13, 15)
+
+right_shoulder -> right_hip -> right_knee -> right_ankle.  (6, 12), (12, 14), (14, 16)
+
+We will create a list containing these keypoint ids to be connected.
+
+.. GENERATED FROM PYTHON SOURCE LINES 451-457
+
+.. code-block:: default
+
+
+    connect_skeleton = [
+        (0, 1), (0, 2), (1, 3), (2, 4), (0, 5), (0, 6), (5, 7), (6, 8),
+        (7, 9), (8, 10), (5, 11), (6, 12), (11, 13), (12, 14), (13, 15), (14, 16)
+    ]
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 458-460
+
+We pass the above list to the connectivity parameter to connect the keypoints.
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 460-463
+
+.. code-block:: default
+
+
+    res = draw_keypoints(person_int, keypoints, connectivity=connect_skeleton, colors="blue", radius=4, width=3)
+    show(res)
+
+
+
+.. image-sg:: /auto_examples/others/images/sphx_glr_plot_visualization_utils_012.png
+   :alt: plot visualization utils
+   :srcset: /auto_examples/others/images/sphx_glr_plot_visualization_utils_012.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 14.555 seconds)
+
+
+.. _sphx_glr_download_auto_examples_others_plot_visualization_utils.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_visualization_utils.py <plot_visualization_utils.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_visualization_utils.ipynb <plot_visualization_utils.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/others/sg_execution_times.rst.txt b/0.16/_sources/auto_examples/others/sg_execution_times.rst.txt
new file mode 100644
index 00000000000..4eab93bf20f
--- /dev/null
+++ b/0.16/_sources/auto_examples/others/sg_execution_times.rst.txt
@@ -0,0 +1,21 @@
+
+:orphan:
+
+.. _sphx_glr_auto_examples_others_sg_execution_times:
+
+
+Computation times
+=================
+**00:34.118** total execution time for **auto_examples_others** files:
+
++------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_others_plot_visualization_utils.py` (``plot_visualization_utils.py``)               | 00:14.555 | 0.0 MB |
++------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_others_plot_optical_flow.py` (``plot_optical_flow.py``)                             | 00:10.096 | 0.0 MB |
++------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_others_plot_video_api.py` (``plot_video_api.py``)                                   | 00:05.277 | 0.0 MB |
++------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_others_plot_repurposing_annotations.py` (``plot_repurposing_annotations.py``)       | 00:02.507 | 0.0 MB |
++------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_others_plot_scripted_tensor_transforms.py` (``plot_scripted_tensor_transforms.py``) | 00:01.683 | 0.0 MB |
++------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/0.16/_sources/auto_examples/transforms/index.rst.txt b/0.16/_sources/auto_examples/transforms/index.rst.txt
new file mode 100644
index 00000000000..bedd351c281
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/index.rst.txt
@@ -0,0 +1,151 @@
+
+
+.. _sphx_glr_auto_examples_transforms:
+
+.. _transforms_gallery:
+
+Transforms
+----------
+
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbnails">
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_transforms_getting_started_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_transforms_getting_started.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_transforms_illustrations_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_transforms_illustrations.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_transforms_e2e_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_transforms_e2e.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy.">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_cutmix_mixup_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_custom_transforms_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_custom_transforms.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_tv_tensors_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+    </div>
+
+
+.. raw:: html
+
+    <div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to...">
+
+.. only:: html
+
+  .. image:: /auto_examples/transforms/images/thumb/sphx_glr_plot_custom_tv_tensors_thumb.png
+    :alt:
+
+  :ref:`sphx_glr_auto_examples_transforms_plot_custom_tv_tensors.py`
+
+.. raw:: html
+
+      <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+    </div>
+
+
+.. raw:: html
+
+    </div>
+
+
+.. toctree::
+   :hidden:
+
+   /auto_examples/transforms/plot_transforms_getting_started
+   /auto_examples/transforms/plot_transforms_illustrations
+   /auto_examples/transforms/plot_transforms_e2e
+   /auto_examples/transforms/plot_cutmix_mixup
+   /auto_examples/transforms/plot_custom_transforms
+   /auto_examples/transforms/plot_tv_tensors
+   /auto_examples/transforms/plot_custom_tv_tensors
+
diff --git a/0.16/_sources/auto_examples/transforms/plot_custom_transforms.rst.txt b/0.16/_sources/auto_examples/transforms/plot_custom_transforms.rst.txt
new file mode 100644
index 00000000000..ed5fe86f027
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/plot_custom_transforms.rst.txt
@@ -0,0 +1,233 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/transforms/plot_custom_transforms.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_transforms_plot_custom_transforms.py:
+
+
+===================================
+How to write your own v2 transforms
+===================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_transforms.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_transforms.py>` to download the full example code.
+
+This guide explains how to write transforms that are compatible with the
+torchvision transforms V2 API.
+
+.. GENERATED FROM PYTHON SOURCE LINES 15-20
+
+.. code-block:: default
+
+    import torch
+    from torchvision import tv_tensors
+    from torchvision.transforms import v2
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 21-45
+
+Just create a ``nn.Module`` and override the ``forward`` method
+===============================================================
+
+In most cases, this is all you're going to need, as long as you already know
+the structure of the input that your transform will expect. For example if
+you're just doing image classification, your transform will typically accept a
+single image as input, or a ``(img, label)`` input. So you can just hard-code
+your ``forward`` method to accept just that, e.g.
+
+.. code:: python
+
+    class MyCustomTransform(torch.nn.Module):
+        def forward(self, img, label):
+            # Do some transformations
+            return new_img, new_label
+
+.. note::
+
+    This means that if you have a custom transform that is already compatible
+    with the V1 transforms (those in ``torchvision.transforms``), it will
+    still work with the V2 transforms without any change!
+
+We will illustrate this more completely below with a typical detection case,
+where our samples are just images, bounding boxes and labels:
+
+.. GENERATED FROM PYTHON SOURCE LINES 45-73
+
+.. code-block:: default
+
+
+    class MyCustomTransform(torch.nn.Module):
+        def forward(self, img, bboxes, label):  # we assume inputs are always structured like this
+            print(
+                f"I'm transforming an image of shape {img.shape} "
+                f"with bboxes = {bboxes}\n{label = }"
+            )
+            # Do some transformations. Here, we're just passing though the input
+            return img, bboxes, label
+
+
+    transforms = v2.Compose([
+        MyCustomTransform(),
+        v2.RandomResizedCrop((224, 224), antialias=True),
+        v2.RandomHorizontalFlip(p=1),
+        v2.Normalize(mean=[0, 0, 0], std=[1, 1, 1])
+    ])
+
+    H, W = 256, 256
+    img = torch.rand(3, H, W)
+    bboxes = tv_tensors.BoundingBoxes(
+        torch.tensor([[0, 10, 10, 20], [50, 50, 70, 70]]),
+        format="XYXY",
+        canvas_size=(H, W)
+    )
+    label = 3
+
+    out_img, out_bboxes, out_label = transforms(img, bboxes, label)
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    I'm transforming an image of shape torch.Size([3, 256, 256]) with bboxes = BoundingBoxes([[ 0, 10, 10, 20],
+                   [50, 50, 70, 70]], format=BoundingBoxFormat.XYXY, canvas_size=(256, 256))
+    label = 3
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 74-75
+
+.. code-block:: default
+
+    print(f"Output image shape: {out_img.shape}\nout_bboxes = {out_bboxes}\n{out_label = }")
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Output image shape: torch.Size([3, 224, 224])
+    out_bboxes = BoundingBoxes([[224,   0, 224,   0],
+                   [136,   0, 173,   0]], format=BoundingBoxFormat.XYXY, canvas_size=(224, 224))
+    out_label = 3
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 76-93
+
+.. note::
+    While working with TVTensor classes in your code, make sure to
+    familiarize yourself with this section:
+    :ref:`tv_tensor_unwrapping_behaviour`
+
+Supporting arbitrary input structures
+=====================================
+
+In the section above, we have assumed that you already know the structure of
+your inputs and that you're OK with hard-coding this expected structure in
+your code. If you want your custom transforms to be as flexible as possible,
+this can be a bit limiting.
+
+A key feature of the builtin Torchvision V2 transforms is that they can accept
+arbitrary input structure and return the same structure as output (with
+transformed entries). For example, transforms can accept a single image, or a
+tuple of ``(img, label)``, or an arbitrary nested dictionary as input:
+
+.. GENERATED FROM PYTHON SOURCE LINES 93-105
+
+.. code-block:: default
+
+
+    structured_input = {
+        "img": img,
+        "annotations": (bboxes, label),
+        "something_that_will_be_ignored": (1, "hello")
+    }
+    structured_output = v2.RandomHorizontalFlip(p=1)(structured_input)
+
+    assert isinstance(structured_output, dict)
+    assert structured_output["something_that_will_be_ignored"] == (1, "hello")
+    print(f"The transformed bboxes are:\n{structured_output['annotations'][0]}")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    The transformed bboxes are:
+    BoundingBoxes([[246,  10, 256,  20],
+                   [186,  50, 206,  70]], format=BoundingBoxFormat.XYXY, canvas_size=(256, 256))
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 106-122
+
+If you want to reproduce this behavior in your own transform, we invite you to
+look at our `code
+<https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/_transform.py>`_
+and adapt it to your needs.
+
+In brief, the core logic is to unpack the input into a flat list using `pytree
+<https://github.com/pytorch/pytorch/blob/main/torch/utils/_pytree.py>`_, and
+then transform only the entries that can be transformed (the decision is made
+based on the **class** of the entries, as all TVTensors are
+tensor-subclasses) plus some custom logic that is out of score here - check the
+code for details. The (potentially transformed) entries are then repacked and
+returned, in the same structure as the input.
+
+We do not provide public dev-facing tools to achieve that at this time, but if
+this is something that would be valuable to you, please let us know by opening
+an issue on our `GitHub repo <https://github.com/pytorch/vision/issues>`_.
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 0.006 seconds)
+
+
+.. _sphx_glr_download_auto_examples_transforms_plot_custom_transforms.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_custom_transforms.py <plot_custom_transforms.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_custom_transforms.ipynb <plot_custom_transforms.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/transforms/plot_custom_tv_tensors.rst.txt b/0.16/_sources/auto_examples/transforms/plot_custom_tv_tensors.rst.txt
new file mode 100644
index 00000000000..04e7d90670d
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/plot_custom_tv_tensors.rst.txt
@@ -0,0 +1,253 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/transforms/plot_custom_tv_tensors.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_transforms_plot_custom_tv_tensors.py:
+
+
+====================================
+How to write your own TVTensor class
+====================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_custom_tv_tensors.py>` to download the full example code.
+
+This guide is intended for advanced users and downstream library maintainers. We explain how to
+write your own TVTensor class, and how to make it compatible with the built-in
+Torchvision v2 transforms. Before continuing, make sure you have read
+:ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.
+
+.. GENERATED FROM PYTHON SOURCE LINES 17-21
+
+.. code-block:: default
+
+    import torch
+    from torchvision import tv_tensors
+    from torchvision.transforms import v2
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 22-28
+
+We will create a very simple class that just inherits from the base
+:class:`~torchvision.tv_tensors.TVTensor` class. It will be enough to cover
+what you need to know to implement your more elaborate uses-cases. If you need
+to create a class that carries meta-data, take a look at how the
+:class:`~torchvision.tv_tensors.BoundingBoxes` class is `implemented
+<https://github.com/pytorch/vision/blob/main/torchvision/tv_tensors/_bounding_box.py>`_.
+
+.. GENERATED FROM PYTHON SOURCE LINES 28-37
+
+.. code-block:: default
+
+
+
+    class MyTVTensor(tv_tensors.TVTensor):
+        pass
+
+
+    my_dp = MyTVTensor([1, 2, 3])
+    my_dp
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+
+    MyTVTensor([1., 2., 3.])
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 38-46
+
+Now that we have defined our custom TVTensor class, we want it to be
+compatible with the built-in torchvision transforms, and the functional API.
+For that, we need to implement a kernel which performs the core of the
+transformation, and then "hook" it to the functional that we want to support
+via :func:`~torchvision.transforms.v2.functional.register_kernel`.
+
+We illustrate this process below: we create a kernel for the "horizontal flip"
+operation of our MyTVTensor class, and register it to the functional API.
+
+.. GENERATED FROM PYTHON SOURCE LINES 46-57
+
+.. code-block:: default
+
+
+    from torchvision.transforms.v2 import functional as F
+
+
+    @F.register_kernel(functional="hflip", tv_tensor_cls=MyTVTensor)
+    def hflip_my_tv_tensor(my_dp, *args, **kwargs):
+        print("Flipping!")
+        out = my_dp.flip(-1)
+        return tv_tensors.wrap(out, like=my_dp)
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 58-71
+
+To understand why :func:`~torchvision.tv_tensors.wrap` is used, see
+:ref:`tv_tensor_unwrapping_behaviour`. Ignore the ``*args, **kwargs`` for now,
+we will explain it below in :ref:`param_forwarding`.
+
+.. note::
+
+    In our call to ``register_kernel`` above we used a string
+    ``functional="hflip"`` to refer to the functional we want to hook into. We
+    could also have used the  functional *itself*, i.e.
+    ``@register_kernel(functional=F.hflip, ...)``.
+
+Now that we have registered our kernel, we can call the functional API on a
+``MyTVTensor`` instance:
+
+.. GENERATED FROM PYTHON SOURCE LINES 71-75
+
+.. code-block:: default
+
+
+    my_dp = MyTVTensor(torch.rand(3, 256, 256))
+    _ = F.hflip(my_dp)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Flipping!
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 76-78
+
+And we can also use the
+:class:`~torchvision.transforms.v2.RandomHorizontalFlip` transform, since it relies on :func:`~torchvision.transforms.v2.functional.hflip` internally:
+
+.. GENERATED FROM PYTHON SOURCE LINES 78-81
+
+.. code-block:: default
+
+    t = v2.RandomHorizontalFlip(p=1)
+    _ = t(my_dp)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Flipping!
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 82-104
+
+.. note::
+
+    We cannot register a kernel for a transform class, we can only register a
+    kernel for a **functional**. The reason we can't register a transform
+    class is because one transform may internally rely on more than one
+    functional, so in general we can't register a single kernel for a given
+    class.
+
+.. _param_forwarding:
+
+Parameter forwarding, and ensuring future compatibility of your kernels
+-----------------------------------------------------------------------
+
+The functional API that you're hooking into is public and therefore
+**backward** compatible: we guarantee that the parameters of these functionals
+won't be removed or renamed without a proper deprecation cycle. However, we
+don't guarantee **forward** compatibility, and we may add new parameters in
+the future.
+
+Imagine that in a future version, Torchvision adds a new ``inplace`` parameter
+to its :func:`~torchvision.transforms.v2.functional.hflip` functional. If you
+already defined and registered your own kernel as
+
+.. GENERATED FROM PYTHON SOURCE LINES 104-111
+
+.. code-block:: default
+
+
+    def hflip_my_tv_tensor(my_dp):  # noqa
+        print("Flipping!")
+        out = my_dp.flip(-1)
+        return tv_tensors.wrap(out, like=my_dp)
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 112-120
+
+then calling ``F.hflip(my_dp)`` will **fail**, because ``hflip`` will try to
+pass the new ``inplace`` parameter to your kernel, but your kernel doesn't
+accept it.
+
+For this reason, we recommend to always define your kernels with
+``*args, **kwargs`` in their signature, as done above. This way, your kernel
+will be able to accept any new parameter that we may add in the future.
+(Technically, adding `**kwargs` only should be enough).
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 0.004 seconds)
+
+
+.. _sphx_glr_download_auto_examples_transforms_plot_custom_tv_tensors.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_custom_tv_tensors.py <plot_custom_tv_tensors.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_custom_tv_tensors.ipynb <plot_custom_tv_tensors.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/transforms/plot_cutmix_mixup.rst.txt b/0.16/_sources/auto_examples/transforms/plot_cutmix_mixup.rst.txt
new file mode 100644
index 00000000000..c6a902badb2
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/plot_cutmix_mixup.rst.txt
@@ -0,0 +1,291 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/transforms/plot_cutmix_mixup.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py:
+
+
+===========================
+How to use CutMix and MixUp
+===========================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_cutmix_mixup.py>` to download the full example code.
+
+:class:`~torchvision.transforms.v2.CutMix` and
+:class:`~torchvision.transforms.v2.MixUp` are popular augmentation strategies
+that can improve classification accuracy.
+
+These transforms are slightly different from the rest of the Torchvision
+transforms, because they expect
+**batches** of samples as input, not individual images. In this example we'll
+explain how to use them: after the ``DataLoader``, or as part of a collation
+function.
+
+.. GENERATED FROM PYTHON SOURCE LINES 23-30
+
+.. code-block:: default
+
+    import torch
+    from torchvision.datasets import FakeData
+    from torchvision.transforms import v2
+
+
+    NUM_CLASSES = 100
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 31-35
+
+Pre-processing pipeline
+-----------------------
+
+We'll use a simple but typical image classification pipeline:
+
+.. GENERATED FROM PYTHON SOURCE LINES 35-49
+
+.. code-block:: default
+
+
+    preproc = v2.Compose([
+        v2.PILToTensor(),
+        v2.RandomResizedCrop(size=(224, 224), antialias=True),
+        v2.RandomHorizontalFlip(p=0.5),
+        v2.ToDtype(torch.float32, scale=True),  # to float32 in [0, 1]
+        v2.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),  # typically from ImageNet
+    ])
+
+    dataset = FakeData(size=1000, num_classes=NUM_CLASSES, transform=preproc)
+
+    img, label = dataset[0]
+    print(f"{type(img) = }, {img.dtype = }, {img.shape = }, {label = }")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    type(img) = <class 'torch.Tensor'>, img.dtype = torch.float32, img.shape = torch.Size([3, 224, 224]), label = 67
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 50-54
+
+One important thing to note is that neither CutMix nor MixUp are part of this
+pre-processing pipeline. We'll add them a bit later once we define the
+DataLoader. Just as a refresher, this is what the DataLoader and training loop
+would look like if we weren't using CutMix or MixUp:
+
+.. GENERATED FROM PYTHON SOURCE LINES 55-65
+
+.. code-block:: default
+
+
+    from torch.utils.data import DataLoader
+
+    dataloader = DataLoader(dataset, batch_size=4, shuffle=True)
+
+    for images, labels in dataloader:
+        print(f"{images.shape = }, {labels.shape = }")
+        print(labels.dtype)
+        # <rest of the training loop here>
+        break
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4])
+    torch.int64
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 68-77
+
+Where to use MixUp and CutMix
+-----------------------------
+
+After the DataLoader
+^^^^^^^^^^^^^^^^^^^^
+
+Now let's add CutMix and MixUp. The simplest way to do this right after the
+DataLoader: the Dataloader has already batched the images and labels for us,
+and this is exactly what these transforms expect as input:
+
+.. GENERATED FROM PYTHON SOURCE LINES 77-91
+
+.. code-block:: default
+
+
+    dataloader = DataLoader(dataset, batch_size=4, shuffle=True)
+
+    cutmix = v2.CutMix(num_classes=NUM_CLASSES)
+    mixup = v2.MixUp(num_classes=NUM_CLASSES)
+    cutmix_or_mixup = v2.RandomChoice([cutmix, mixup])
+
+    for images, labels in dataloader:
+        print(f"Before CutMix/MixUp: {images.shape = }, {labels.shape = }")
+        images, labels = cutmix_or_mixup(images, labels)
+        print(f"After CutMix/MixUp: {images.shape = }, {labels.shape = }")
+
+        # <rest of the training loop here>
+        break
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Before CutMix/MixUp: images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4])
+    After CutMix/MixUp: images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4, 100])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 92-106
+
+Note how the labels were also transformed: we went from a batched label of
+shape (batch_size,) to a tensor of shape (batch_size, num_classes). The
+transformed labels can still be passed as-is to a loss function like
+:func:`torch.nn.functional.cross_entropy`.
+
+As part of the collation function
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Passing the transforms after the DataLoader is the simplest way to use CutMix
+and MixUp, but one disadvantage is that it does not take advantage of the
+DataLoader multi-processing. For that, we can pass those transforms as part of
+the collation function (refer to the `PyTorch docs
+<https://pytorch.org/docs/stable/data.html#dataloader-collate-fn>`_ to learn
+more about collation).
+
+.. GENERATED FROM PYTHON SOURCE LINES 107-123
+
+.. code-block:: default
+
+
+    from torch.utils.data import default_collate
+
+
+    def collate_fn(batch):
+        return cutmix_or_mixup(*default_collate(batch))
+
+
+    dataloader = DataLoader(dataset, batch_size=4, shuffle=True, num_workers=2, collate_fn=collate_fn)
+
+    for images, labels in dataloader:
+        print(f"{images.shape = }, {labels.shape = }")
+        # No need to call cutmix_or_mixup, it's already been called as part of the DataLoader!
+        # <rest of the training loop here>
+        break
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4, 100])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 124-135
+
+Non-standard input format
+-------------------------
+
+So far we've used a typical sample structure where we pass ``(images,
+labels)`` as inputs. MixUp and CutMix will magically work by default with most
+common sample structures: tuples where the second parameter is a tensor label,
+or dict with a "label[s]" key. Look at the documentation of the
+``labels_getter`` parameter for more details.
+
+If your samples have a different structure, you can still use CutMix and MixUp
+by passing a callable to the ``labels_getter`` parameter. For example:
+
+.. GENERATED FROM PYTHON SOURCE LINES 135-151
+
+.. code-block:: default
+
+
+    batch = {
+        "imgs": torch.rand(4, 3, 224, 224),
+        "target": {
+            "classes": torch.randint(0, NUM_CLASSES, size=(4,)),
+            "some_other_key": "this is going to be passed-through"
+        }
+    }
+
+
+    def labels_getter(batch):
+        return batch["target"]["classes"]
+
+
+    out = v2.CutMix(num_classes=NUM_CLASSES, labels_getter=labels_getter)(batch)
+    print(f"{out['imgs'].shape = }, {out['target']['classes'].shape = }")
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    out['imgs'].shape = torch.Size([4, 3, 224, 224]), out['target']['classes'].shape = torch.Size([4, 100])
+
+
+
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 0.155 seconds)
+
+
+.. _sphx_glr_download_auto_examples_transforms_plot_cutmix_mixup.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_cutmix_mixup.py <plot_cutmix_mixup.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_cutmix_mixup.ipynb <plot_cutmix_mixup.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/transforms/plot_transforms_e2e.rst.txt b/0.16/_sources/auto_examples/transforms/plot_transforms_e2e.rst.txt
new file mode 100644
index 00000000000..a50908eafed
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/plot_transforms_e2e.rst.txt
@@ -0,0 +1,364 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/transforms/plot_transforms_e2e.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_transforms_plot_transforms_e2e.py:
+
+
+===============================================================
+Transforms v2: End-to-end object detection/segmentation example
+===============================================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_e2e.py>` to download the full example code.
+
+Object detection and segmentation tasks are natively supported:
+``torchvision.transforms.v2`` enables jointly transforming images, videos,
+bounding boxes, and masks.
+
+This example showcases an end-to-end instance segmentation training case using
+Torchvision utils from ``torchvision.datasets``, ``torchvision.models`` and
+``torchvision.transforms.v2``. Everything covered here can be applied similarly
+to object detection or semantic segmentation tasks.
+
+.. GENERATED FROM PYTHON SOURCE LINES 21-41
+
+.. code-block:: default
+
+    import pathlib
+
+    import torch
+    import torch.utils.data
+
+    from torchvision import models, datasets, tv_tensors
+    from torchvision.transforms import v2
+
+    torch.manual_seed(0)
+
+    # This loads fake data for illustration purposes of this example. In practice, you'll have
+    # to replace this with the proper data.
+    # If you're trying to run that on collab, you can download the assets and the
+    # helpers from https://github.com/pytorch/vision/tree/main/gallery/
+    ROOT = pathlib.Path("../assets") / "coco"
+    IMAGES_PATH = str(ROOT / "images")
+    ANNOTATIONS_PATH = str(ROOT / "instances.json")
+    from helpers import plot
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 42-47
+
+Dataset preparation
+-------------------
+
+We start off by loading the :class:`~torchvision.datasets.CocoDetection` dataset to have a look at what it currently
+returns.
+
+.. GENERATED FROM PYTHON SOURCE LINES 47-55
+
+.. code-block:: default
+
+
+    dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH)
+
+    sample = dataset[0]
+    img, target = sample
+    print(f"{type(img) = }\n{type(target) = }\n{type(target[0]) = }\n{target[0].keys() = }")
+
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    loading annotations into memory...
+    Done (t=0.00s)
+    creating index...
+    index created!
+    type(img) = <class 'PIL.Image.Image'>
+    type(target) = <class 'list'>
+    type(target[0]) = <class 'dict'>
+    target[0].keys() = dict_keys(['segmentation', 'iscrowd', 'image_id', 'bbox', 'category_id', 'id'])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 56-64
+
+Torchvision datasets preserve the data structure and types as it was intended
+by the datasets authors. So by default, the output structure may not always be
+compatible with the models or the transforms.
+
+To overcome that, we can use the
+:func:`~torchvision.datasets.wrap_dataset_for_transforms_v2` function. For
+:class:`~torchvision.datasets.CocoDetection`, this changes the target
+structure to a single dictionary of lists:
+
+.. GENERATED FROM PYTHON SOURCE LINES 64-72
+
+.. code-block:: default
+
+
+    dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=("boxes", "labels", "masks"))
+
+    sample = dataset[0]
+    img, target = sample
+    print(f"{type(img) = }\n{type(target) = }\n{target.keys() = }")
+    print(f"{type(target['boxes']) = }\n{type(target['labels']) = }\n{type(target['masks']) = }")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    type(img) = <class 'PIL.Image.Image'>
+    type(target) = <class 'dict'>
+    target.keys() = dict_keys(['boxes', 'masks', 'labels'])
+    type(target['boxes']) = <class 'torchvision.tv_tensors._bounding_boxes.BoundingBoxes'>
+    type(target['labels']) = <class 'torch.Tensor'>
+    type(target['masks']) = <class 'torchvision.tv_tensors._mask.Mask'>
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 73-87
+
+We used the ``target_keys`` parameter to specify the kind of output we're
+interested in. Our dataset now returns a target which is dict where the values
+are :ref:`TVTensors <what_are_tv_tensors>` (all are :class:`torch.Tensor`
+subclasses). We're dropped all unncessary keys from the previous output, but
+if you need any of the original keys e.g. "image_id", you can still ask for
+it.
+
+.. note::
+
+    If you just want to do detection, you don't need and shouldn't pass
+    "masks" in ``target_keys``: if masks are present in the sample, they will
+    be transformed, slowing down your transformations unnecessarily.
+
+As baseline, let's have a look at a sample without transformations:
+
+.. GENERATED FROM PYTHON SOURCE LINES 87-91
+
+.. code-block:: default
+
+
+    plot([dataset[0], dataset[1]])
+
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_e2e_001.png
+   :alt: plot transforms e2e
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_e2e_001.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 92-101
+
+Transforms
+----------
+
+Let's now define our pre-processing transforms. All the transforms know how
+to handle images, bouding boxes and masks when relevant.
+
+Transforms are typically passed as the ``transforms`` parameter of the
+dataset so that they can leverage multi-processing from the
+:class:`torch.utils.data.DataLoader`.
+
+.. GENERATED FROM PYTHON SOURCE LINES 101-117
+
+.. code-block:: default
+
+
+    transforms = v2.Compose(
+        [
+            v2.ToImage(),
+            v2.RandomPhotometricDistort(p=1),
+            v2.RandomZoomOut(fill={tv_tensors.Image: (123, 117, 104), "others": 0}),
+            v2.RandomIoUCrop(),
+            v2.RandomHorizontalFlip(p=1),
+            v2.SanitizeBoundingBoxes(),
+            v2.ToDtype(torch.float32, scale=True),
+        ]
+    )
+
+    dataset = datasets.CocoDetection(IMAGES_PATH, ANNOTATIONS_PATH, transforms=transforms)
+    dataset = datasets.wrap_dataset_for_transforms_v2(dataset, target_keys=["boxes", "labels", "masks"])
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    loading annotations into memory...
+    Done (t=0.00s)
+    creating index...
+    index created!
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 118-132
+
+A few things are worth noting here:
+
+- We're converting the PIL image into a
+  :class:`~torchvision.transforms.v2.Image` object. This isn't strictly
+  necessary, but relying on Tensors (here: a Tensor subclass) will
+  :ref:`generally be faster <transforms_perf>`.
+- We are calling :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` to
+  make sure we remove degenerate bounding boxes, as well as their
+  corresponding labels and masks.
+  :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes` should be placed
+  at least once at the end of a detection pipeline; it is particularly
+  critical if :class:`~torchvision.transforms.v2.RandomIoUCrop` was used.
+
+Let's look how the sample looks like with our augmentation pipeline in place:
+
+.. GENERATED FROM PYTHON SOURCE LINES 132-136
+
+.. code-block:: default
+
+
+    plot([dataset[0], dataset[1]])
+
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_e2e_002.png
+   :alt: plot transforms e2e
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_e2e_002.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 138-147
+
+We can see that the color of the images were distorted, zoomed in or out, and flipped.
+The bounding boxes and the masks were transformed accordingly. And without any further ado, we can start training.
+
+Data loading and training loop
+------------------------------
+
+Below we're using Mask-RCNN which is an instance segmentation model, but
+everything we've covered in this tutorial also applies to object detection and
+semantic segmentation tasks.
+
+.. GENERATED FROM PYTHON SOURCE LINES 147-170
+
+.. code-block:: default
+
+
+    data_loader = torch.utils.data.DataLoader(
+        dataset,
+        batch_size=2,
+        # We need a custom collation function here, since the object detection
+        # models expect a sequence of images and target dictionaries. The default
+        # collation function tries to torch.stack() the individual elements,
+        # which fails in general for object detection, because the number of bouding
+        # boxes varies between the images of a same batch.
+        collate_fn=lambda batch: tuple(zip(*batch)),
+    )
+
+    model = models.get_model("maskrcnn_resnet50_fpn_v2", weights=None, weights_backbone=None).train()
+
+    for imgs, targets in data_loader:
+        loss_dict = model(imgs, targets)
+        # Put your training logic here
+
+        print(f"{[img.shape for img in imgs] = }")
+        print(f"{[type(target) for target in targets] = }")
+        for name, loss_val in loss_dict.items():
+            print(f"{name:<20}{loss_val:.3f}")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    [img.shape for img in imgs] = [torch.Size([3, 512, 512]), torch.Size([3, 409, 493])]
+    [type(target) for target in targets] = [<class 'dict'>, <class 'dict'>]
+    loss_classifier     4.721
+    loss_box_reg        0.006
+    loss_mask           0.734
+    loss_objectness     0.691
+    loss_rpn_box_reg    0.036
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 171-182
+
+Training References
+-------------------
+
+From there, you can check out the `torchvision references
+<https://github.com/pytorch/vision/tree/main/references>`_ where you'll find
+the actual training scripts we use to train our models.
+
+**Disclaimer** The code in our references is more complex than what you'll
+need for your own use-cases: this is because we're supporting different
+backends (PIL, tensors, TVTensors) and different transforms namespaces (v1 and
+v2). So don't be afraid to simplify and only keep what you need.
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 4.813 seconds)
+
+
+.. _sphx_glr_download_auto_examples_transforms_plot_transforms_e2e.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_transforms_e2e.py <plot_transforms_e2e.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_transforms_e2e.ipynb <plot_transforms_e2e.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/transforms/plot_transforms_getting_started.rst.txt b/0.16/_sources/auto_examples/transforms/plot_transforms_getting_started.rst.txt
new file mode 100644
index 00000000000..2df8b5e8cd8
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/plot_transforms_getting_started.rst.txt
@@ -0,0 +1,428 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/transforms/plot_transforms_getting_started.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_transforms_plot_transforms_getting_started.py:
+
+
+==================================
+Getting started with transforms v2
+==================================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_getting_started.py>` to download the full example code.
+
+This example illustrates all of what you need to know to get started with the
+new :mod:`torchvision.transforms.v2` API. We'll cover simple tasks like
+image classification, and more advanced ones like object detection /
+segmentation.
+
+.. GENERATED FROM PYTHON SOURCE LINES 17-18
+
+First, a bit of setup
+
+.. GENERATED FROM PYTHON SOURCE LINES 18-34
+
+.. code-block:: default
+
+    from pathlib import Path
+    import torch
+    import matplotlib.pyplot as plt
+    plt.rcParams["savefig.bbox"] = 'tight'
+
+    from torchvision.transforms import v2
+    from torchvision.io import read_image
+
+    torch.manual_seed(1)
+
+    # If you're trying to run that on collab, you can download the assets and the
+    # helpers from https://github.com/pytorch/vision/tree/main/gallery/
+    from helpers import plot
+    img = read_image(str(Path('../assets') / 'astronaut.jpg'))
+    print(f"{type(img) = }, {img.dtype = }, {img.shape = }")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    type(img) = <class 'torch.Tensor'>, img.dtype = torch.uint8, img.shape = torch.Size([3, 512, 512])
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 35-41
+
+The basics
+----------
+
+The Torchvision transforms behave like a regular :class:`torch.nn.Module` (in
+fact, most of them are): instantiate a transform, pass an input, get a
+transformed output:
+
+.. GENERATED FROM PYTHON SOURCE LINES 41-47
+
+.. code-block:: default
+
+
+    transform = v2.RandomCrop(size=(224, 224))
+    out = transform(img)
+
+    plot([img, out])
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_001.png
+   :alt: plot transforms getting started
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_001.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 48-53
+
+I just want to do image classification
+--------------------------------------
+
+If you just care about image classification, things are very simple. A basic
+classification pipeline may look like this:
+
+.. GENERATED FROM PYTHON SOURCE LINES 53-64
+
+.. code-block:: default
+
+
+    transforms = v2.Compose([
+        v2.RandomResizedCrop(size=(224, 224), antialias=True),
+        v2.RandomHorizontalFlip(p=0.5),
+        v2.ToDtype(torch.float32, scale=True),
+        v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+    ])
+    out = transforms(img)
+
+    plot([img, out])
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_002.png
+   :alt: plot transforms getting started
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_002.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 65-90
+
+Such transformation pipeline is typically passed as the ``transform`` argument
+to the :ref:`Datasets <datasets>`, e.g. ``ImageNet(...,
+transform=transforms)``.
+
+That's pretty much all there is. From there, read through our :ref:`main docs
+<transforms>` to learn more about recommended practices and conventions, or
+explore more :ref:`examples <transforms_gallery>` e.g. how to use augmentation
+transforms like :ref:`CutMix and MixUp
+<sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py>`.
+
+.. note::
+
+    If you're already relying on the ``torchvision.transforms`` v1 API,
+    we recommend to :ref:`switch to the new v2 transforms<v1_or_v2>`. It's
+    very easy: the v2 transforms are fully compatible with the v1 API, so you
+    only need to change the import!
+
+Detection, Segmentation, Videos
+-------------------------------
+
+The new Torchvision transforms in the ``torchvision.transforms.v2`` namespace
+support tasks beyond image classification: they can also transform bounding
+boxes, segmentation / detection masks, or videos.
+
+Let's briefly look at a detection example with bounding boxes.
+
+.. GENERATED FROM PYTHON SOURCE LINES 90-111
+
+.. code-block:: default
+
+
+    from torchvision import tv_tensors  # we'll describe this a bit later, bare with us
+
+    boxes = tv_tensors.BoundingBoxes(
+        [
+            [15, 10, 370, 510],
+            [275, 340, 510, 510],
+            [130, 345, 210, 425]
+        ],
+        format="XYXY", canvas_size=img.shape[-2:])
+
+    transforms = v2.Compose([
+        v2.RandomResizedCrop(size=(224, 224), antialias=True),
+        v2.RandomPhotometricDistort(p=1),
+        v2.RandomHorizontalFlip(p=1),
+    ])
+    out_img, out_boxes = transforms(img, boxes)
+    print(type(boxes), type(out_boxes))
+
+    plot([(img, boxes), (out_img, out_boxes)])
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_003.png
+   :alt: plot transforms getting started
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_003.png
+   :class: sphx-glr-single-img
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    <class 'torchvision.tv_tensors._bounding_boxes.BoundingBoxes'> <class 'torchvision.tv_tensors._bounding_boxes.BoundingBoxes'>
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 112-120
+
+The example above focuses on object detection. But if we had masks
+(:class:`torchvision.tv_tensors.Mask`) for object segmentation or semantic
+segmentation, or videos (:class:`torchvision.tv_tensors.Video`), we could have
+passed them to the transforms in exactly the same way.
+
+By now you likely have a few questions: what are these TVTensors, how do we
+use them, and what is the expected input/output of those transforms? We'll
+answer these in the next sections.
+
+.. GENERATED FROM PYTHON SOURCE LINES 123-137
+
+.. _what_are_tv_tensors:
+
+What are TVTensors?
+--------------------
+
+TVTensors are :class:`torch.Tensor` subclasses. The available TVTensors are
+:class:`~torchvision.tv_tensors.Image`,
+:class:`~torchvision.tv_tensors.BoundingBoxes`,
+:class:`~torchvision.tv_tensors.Mask`, and
+:class:`~torchvision.tv_tensors.Video`.
+
+TVTensors look and feel just like regular tensors - they **are** tensors.
+Everything that is supported on a plain :class:`torch.Tensor` like ``.sum()``
+or any ``torch.*`` operator will also work on a TVTensor:
+
+.. GENERATED FROM PYTHON SOURCE LINES 138-144
+
+.. code-block:: default
+
+
+    img_dp = tv_tensors.Image(torch.randint(0, 256, (3, 256, 256), dtype=torch.uint8))
+
+    print(f"{isinstance(img_dp, torch.Tensor) = }")
+    print(f"{img_dp.dtype = }, {img_dp.shape = }, {img_dp.sum() = }")
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    isinstance(img_dp, torch.Tensor) = True
+    img_dp.dtype = torch.uint8, img_dp.shape = torch.Size([3, 256, 256]), img_dp.sum() = tensor(25087958)
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 145-165
+
+These TVTensor classes are at the core of the transforms: in order to
+transform a given input, the transforms first look at the **class** of the
+object, and dispatch to the appropriate implementation accordingly.
+
+You don't need to know much more about TVTensors at this point, but advanced
+users who want to learn more can refer to
+:ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py`.
+
+What do I pass as input?
+------------------------
+
+Above, we've seen two examples: one where we passed a single image as input
+i.e. ``out = transforms(img)``, and one where we passed both an image and
+bounding boxes, i.e. ``out_img, out_boxes = transforms(img, boxes)``.
+
+In fact, transforms support **arbitrary input structures**. The input can be a
+single image, a tuple, an arbitrarily nested dictionary... pretty much
+anything. The same structure will be returned as output. Below, we use the
+same detection transforms, but pass a tuple (image, target_dict) as input and
+we're getting the same structure as output:
+
+.. GENERATED FROM PYTHON SOURCE LINES 165-178
+
+.. code-block:: default
+
+
+    target = {
+        "boxes": boxes,
+        "labels": torch.arange(boxes.shape[0]),
+        "this_is_ignored": ("arbitrary", {"structure": "!"})
+    }
+
+    # Re-using the transforms and definitions from above.
+    out_img, out_target = transforms(img, target)
+
+    plot([(img, target["boxes"]), (out_img, out_target["boxes"])])
+    print(f"{out_target['this_is_ignored']}")
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_004.png
+   :alt: plot transforms getting started
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_getting_started_004.png
+   :class: sphx-glr-single-img
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    ('arbitrary', {'structure': '!'})
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 180-267
+
+We passed a tuple so we get a tuple back, and the second element is the
+tranformed target dict. Transforms don't really care about the structure of
+the input; as mentioned above, they only care about the **type** of the
+objects and transforms them accordingly.
+
+*Foreign* objects like strings or ints are simply passed-through. This can be
+useful e.g. if you want to associate a path with every single sample when
+debugging!
+
+.. _passthrough_heuristic:
+
+.. note::
+
+    **Disclaimer** This note is slightly advanced and can be safely skipped on
+    a first read.
+
+    Pure :class:`torch.Tensor` objects are, in general, treated as images (or
+    as videos for video-specific transforms). Indeed, you may have noticed
+    that in the code above we haven't used the
+    :class:`~torchvision.tv_tensors.Image` class at all, and yet our images
+    got transformed properly. Transforms follow the following logic to
+    determine whether a pure Tensor should be treated as an image (or video),
+    or just ignored:
+
+    * If there is an :class:`~torchvision.tv_tensors.Image`,
+      :class:`~torchvision.tv_tensors.Video`,
+      or :class:`PIL.Image.Image` instance in the input, all other pure
+      tensors are passed-through.
+    * If there is no :class:`~torchvision.tv_tensors.Image` or
+      :class:`~torchvision.tv_tensors.Video` instance, only the first pure
+      :class:`torch.Tensor` will be transformed as image or video, while all
+      others will be passed-through. Here "first" means "first in a depth-wise
+      traversal".
+
+    This is what happened in the detection example above: the first pure
+    tensor was the image so it got transformed properly, and all other pure
+    tensor instances like the ``labels`` were passed-through (although labels
+    can still be transformed by some transforms like
+    :class:`~torchvision.transforms.v2.SanitizeBoundingBoxes`!).
+
+.. _transforms_datasets_intercompatibility:
+
+Transforms and Datasets intercompatibility
+------------------------------------------
+
+Roughly speaking, the output of the datasets must correspond to the input of
+the transforms. How to do that depends on whether you're using the torchvision
+:ref:`built-in datatsets <datasets>`, or your own custom datasets.
+
+Using built-in datasets
+^^^^^^^^^^^^^^^^^^^^^^^
+
+If you're just doing image classification, you don't need to do anything. Just
+use ``transform`` argument of the dataset e.g. ``ImageNet(...,
+transform=transforms)`` and you're good to go.
+
+Torchvision also supports datasets for object detection or segmentation like
+:class:`torchvision.datasets.CocoDetection`. Those datasets predate
+the existence of the :mod:`torchvision.transforms.v2` module and of the
+TVTensors, so they don't return TVTensors out of the box.
+
+An easy way to force those datasets to return TVTensors and to make them
+compatible with v2 transforms is to use the
+:func:`torchvision.datasets.wrap_dataset_for_transforms_v2` function:
+
+.. code-block:: python
+
+   from torchvision.datasets import CocoDetection, wrap_dataset_for_transforms_v2
+
+   dataset = CocoDetection(..., transforms=my_transforms)
+   dataset = wrap_dataset_for_transforms_v2(dataset)
+   # Now the dataset returns TVTensors!
+
+Using your own datasets
+^^^^^^^^^^^^^^^^^^^^^^^
+
+If you have a custom dataset, then you'll need to convert your objects into
+the appropriate TVTensor classes. Creating TVTensor instances is very easy,
+refer to :ref:`tv_tensor_creation` for more details.
+
+There are two main places where you can implement that conversion logic:
+
+- At the end of the datasets's ``__getitem__`` method, before returning the
+  sample (or by sub-classing the dataset).
+- As the very first step of your transforms pipeline
+
+Either way, the logic will depend on your specific dataset.
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 0.838 seconds)
+
+
+.. _sphx_glr_download_auto_examples_transforms_plot_transforms_getting_started.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_transforms_getting_started.py <plot_transforms_getting_started.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_transforms_getting_started.ipynb <plot_transforms_getting_started.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/transforms/plot_transforms_illustrations.rst.txt b/0.16/_sources/auto_examples/transforms/plot_transforms_illustrations.rst.txt
new file mode 100644
index 00000000000..7ca0c507f60
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/plot_transforms_illustrations.rst.txt
@@ -0,0 +1,837 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/transforms/plot_transforms_illustrations.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_transforms_plot_transforms_illustrations.py:
+
+
+==========================
+Illustration of transforms
+==========================
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_transforms_illustrations.py>` to download the full example code.
+
+This example illustrates some of the various transforms available in :ref:`the
+torchvision.transforms.v2 module <transforms>`.
+
+.. GENERATED FROM PYTHON SOURCE LINES 14-34
+
+.. code-block:: default
+
+
+
+    from PIL import Image
+    from pathlib import Path
+    import matplotlib.pyplot as plt
+
+    import torch
+    from torchvision.transforms import v2
+
+    plt.rcParams["savefig.bbox"] = 'tight'
+
+    # if you change the seed, make sure that the randomly-applied transforms
+    # properly show that the image can be both transformed and *not* transformed!
+    torch.manual_seed(0)
+
+    # If you're trying to run that on collab, you can download the assets and the
+    # helpers from https://github.com/pytorch/vision/tree/main/gallery/
+    from helpers import plot
+    orig_img = Image.open(Path('../assets') / 'astronaut.jpg')
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 36-47
+
+Geometric Transforms
+--------------------
+Geometric image transformation refers to the process of altering the geometric properties of an image,
+such as its shape, size, orientation, or position.
+It involves applying mathematical operations to the image pixels or coordinates to achieve the desired transformation.
+
+Pad
+~~~
+The :class:`~torchvision.transforms.Pad` transform
+(see also :func:`~torchvision.transforms.functional.pad`)
+pads all image borders with some pixel values.
+
+.. GENERATED FROM PYTHON SOURCE LINES 47-50
+
+.. code-block:: default
+
+    padded_imgs = [v2.Pad(padding=padding)(orig_img) for padding in (3, 10, 30, 50)]
+    plot([orig_img] + padded_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_001.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_001.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 51-56
+
+Resize
+~~~~~~
+The :class:`~torchvision.transforms.Resize` transform
+(see also :func:`~torchvision.transforms.functional.resize`)
+resizes an image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 56-59
+
+.. code-block:: default
+
+    resized_imgs = [v2.Resize(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
+    plot([orig_img] + resized_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_002.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_002.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 60-65
+
+CenterCrop
+~~~~~~~~~~
+The :class:`~torchvision.transforms.CenterCrop` transform
+(see also :func:`~torchvision.transforms.functional.center_crop`)
+crops the given image at the center.
+
+.. GENERATED FROM PYTHON SOURCE LINES 65-68
+
+.. code-block:: default
+
+    center_crops = [v2.CenterCrop(size=size)(orig_img) for size in (30, 50, 100, orig_img.size)]
+    plot([orig_img] + center_crops)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_003.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_003.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 69-74
+
+FiveCrop
+~~~~~~~~
+The :class:`~torchvision.transforms.FiveCrop` transform
+(see also :func:`~torchvision.transforms.functional.five_crop`)
+crops the given image into four corners and the central crop.
+
+.. GENERATED FROM PYTHON SOURCE LINES 74-77
+
+.. code-block:: default
+
+    (top_left, top_right, bottom_left, bottom_right, center) = v2.FiveCrop(size=(100, 100))(orig_img)
+    plot([orig_img] + [top_left, top_right, bottom_left, bottom_right, center])
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_004.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_004.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 78-83
+
+RandomPerspective
+~~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomPerspective` transform
+(see also :func:`~torchvision.transforms.functional.perspective`)
+performs random perspective transform on an image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 83-87
+
+.. code-block:: default
+
+    perspective_transformer = v2.RandomPerspective(distortion_scale=0.6, p=1.0)
+    perspective_imgs = [perspective_transformer(orig_img) for _ in range(4)]
+    plot([orig_img] + perspective_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_005.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_005.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 88-93
+
+RandomRotation
+~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomRotation` transform
+(see also :func:`~torchvision.transforms.functional.rotate`)
+rotates an image with random angle.
+
+.. GENERATED FROM PYTHON SOURCE LINES 93-97
+
+.. code-block:: default
+
+    rotater = v2.RandomRotation(degrees=(0, 180))
+    rotated_imgs = [rotater(orig_img) for _ in range(4)]
+    plot([orig_img] + rotated_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_006.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_006.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 98-103
+
+RandomAffine
+~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomAffine` transform
+(see also :func:`~torchvision.transforms.functional.affine`)
+performs random affine transform on an image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 103-107
+
+.. code-block:: default
+
+    affine_transfomer = v2.RandomAffine(degrees=(30, 70), translate=(0.1, 0.3), scale=(0.5, 0.75))
+    affine_imgs = [affine_transfomer(orig_img) for _ in range(4)]
+    plot([orig_img] + affine_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_007.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_007.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 108-114
+
+ElasticTransform
+~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.ElasticTransform` transform
+(see also :func:`~torchvision.transforms.functional.elastic_transform`)
+Randomly transforms the morphology of objects in images and produces a
+see-through-water-like effect.
+
+.. GENERATED FROM PYTHON SOURCE LINES 114-118
+
+.. code-block:: default
+
+    elastic_transformer = v2.ElasticTransform(alpha=250.0)
+    transformed_imgs = [elastic_transformer(orig_img) for _ in range(2)]
+    plot([orig_img] + transformed_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_008.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_008.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 119-124
+
+RandomCrop
+~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomCrop` transform
+(see also :func:`~torchvision.transforms.functional.crop`)
+crops an image at a random location.
+
+.. GENERATED FROM PYTHON SOURCE LINES 124-128
+
+.. code-block:: default
+
+    cropper = v2.RandomCrop(size=(128, 128))
+    crops = [cropper(orig_img) for _ in range(4)]
+    plot([orig_img] + crops)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_009.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_009.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 129-135
+
+RandomResizedCrop
+~~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomResizedCrop` transform
+(see also :func:`~torchvision.transforms.functional.resized_crop`)
+crops an image at a random location, and then resizes the crop to a given
+size.
+
+.. GENERATED FROM PYTHON SOURCE LINES 135-139
+
+.. code-block:: default
+
+    resize_cropper = v2.RandomResizedCrop(size=(32, 32))
+    resized_crops = [resize_cropper(orig_img) for _ in range(4)]
+    plot([orig_img] + resized_crops)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_010.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_010.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 140-156
+
+Photometric Transforms
+----------------------
+Photometric image transformation refers to the process of modifying the photometric properties of an image,
+such as its brightness, contrast, color, or tone.
+These transformations are applied to change the visual appearance of an image
+while preserving its geometric structure.
+
+Except :class:`~torchvision.transforms.Grayscale`, the following transforms are random,
+which means that the same transform
+instance will produce different result each time it transforms a given image.
+
+Grayscale
+~~~~~~~~~
+The :class:`~torchvision.transforms.Grayscale` transform
+(see also :func:`~torchvision.transforms.functional.to_grayscale`)
+converts an image to grayscale
+
+.. GENERATED FROM PYTHON SOURCE LINES 156-159
+
+.. code-block:: default
+
+    gray_img = v2.Grayscale()(orig_img)
+    plot([orig_img, gray_img], cmap='gray')
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_011.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_011.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 160-164
+
+ColorJitter
+~~~~~~~~~~~
+The :class:`~torchvision.transforms.ColorJitter` transform
+randomly changes the brightness, contrast, saturation, hue, and other properties of an image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 164-168
+
+.. code-block:: default
+
+    jitter = v2.ColorJitter(brightness=.5, hue=.3)
+    jittered_imgs = [jitter(orig_img) for _ in range(4)]
+    plot([orig_img] + jittered_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_012.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_012.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 169-174
+
+GaussianBlur
+~~~~~~~~~~~~
+The :class:`~torchvision.transforms.GaussianBlur` transform
+(see also :func:`~torchvision.transforms.functional.gaussian_blur`)
+performs gaussian blur transform on an image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 174-178
+
+.. code-block:: default
+
+    blurrer = v2.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5.))
+    blurred_imgs = [blurrer(orig_img) for _ in range(4)]
+    plot([orig_img] + blurred_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_013.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_013.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 179-184
+
+RandomInvert
+~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomInvert` transform
+(see also :func:`~torchvision.transforms.functional.invert`)
+randomly inverts the colors of the given image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 184-188
+
+.. code-block:: default
+
+    inverter = v2.RandomInvert()
+    invertered_imgs = [inverter(orig_img) for _ in range(4)]
+    plot([orig_img] + invertered_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_014.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_014.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 189-195
+
+RandomPosterize
+~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomPosterize` transform
+(see also :func:`~torchvision.transforms.functional.posterize`)
+randomly posterizes the image by reducing the number of bits
+of each color channel.
+
+.. GENERATED FROM PYTHON SOURCE LINES 195-199
+
+.. code-block:: default
+
+    posterizer = v2.RandomPosterize(bits=2)
+    posterized_imgs = [posterizer(orig_img) for _ in range(4)]
+    plot([orig_img] + posterized_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_015.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_015.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 200-206
+
+RandomSolarize
+~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomSolarize` transform
+(see also :func:`~torchvision.transforms.functional.solarize`)
+randomly solarizes the image by inverting all pixel values above
+the threshold.
+
+.. GENERATED FROM PYTHON SOURCE LINES 206-210
+
+.. code-block:: default
+
+    solarizer = v2.RandomSolarize(threshold=192.0)
+    solarized_imgs = [solarizer(orig_img) for _ in range(4)]
+    plot([orig_img] + solarized_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_016.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_016.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 211-216
+
+RandomAdjustSharpness
+~~~~~~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomAdjustSharpness` transform
+(see also :func:`~torchvision.transforms.functional.adjust_sharpness`)
+randomly adjusts the sharpness of the given image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 216-220
+
+.. code-block:: default
+
+    sharpness_adjuster = v2.RandomAdjustSharpness(sharpness_factor=2)
+    sharpened_imgs = [sharpness_adjuster(orig_img) for _ in range(4)]
+    plot([orig_img] + sharpened_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_017.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_017.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 221-226
+
+RandomAutocontrast
+~~~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomAutocontrast` transform
+(see also :func:`~torchvision.transforms.functional.autocontrast`)
+randomly applies autocontrast to the given image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 226-230
+
+.. code-block:: default
+
+    autocontraster = v2.RandomAutocontrast()
+    autocontrasted_imgs = [autocontraster(orig_img) for _ in range(4)]
+    plot([orig_img] + autocontrasted_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_018.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_018.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 231-236
+
+RandomEqualize
+~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomEqualize` transform
+(see also :func:`~torchvision.transforms.functional.equalize`)
+randomly equalizes the histogram of the given image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 236-240
+
+.. code-block:: default
+
+    equalizer = v2.RandomEqualize()
+    equalized_imgs = [equalizer(orig_img) for _ in range(4)]
+    plot([orig_img] + equalized_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_019.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_019.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 241-251
+
+Augmentation Transforms
+-----------------------
+The following transforms are combinations of multiple transforms,
+either geometric or photometric, or both.
+
+AutoAugment
+~~~~~~~~~~~
+The :class:`~torchvision.transforms.AutoAugment` transform
+automatically augments data based on a given auto-augmentation policy.
+See :class:`~torchvision.transforms.AutoAugmentPolicy` for the available policies.
+
+.. GENERATED FROM PYTHON SOURCE LINES 251-260
+
+.. code-block:: default
+
+    policies = [v2.AutoAugmentPolicy.CIFAR10, v2.AutoAugmentPolicy.IMAGENET, v2.AutoAugmentPolicy.SVHN]
+    augmenters = [v2.AutoAugment(policy) for policy in policies]
+    imgs = [
+        [augmenter(orig_img) for _ in range(4)]
+        for augmenter in augmenters
+    ]
+    row_title = [str(policy).split('.')[-1] for policy in policies]
+    plot([[orig_img] + row for row in imgs], row_title=row_title)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_020.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_020.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 261-264
+
+RandAugment
+~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandAugment` is an alternate version of AutoAugment.
+
+.. GENERATED FROM PYTHON SOURCE LINES 264-268
+
+.. code-block:: default
+
+    augmenter = v2.RandAugment()
+    imgs = [augmenter(orig_img) for _ in range(4)]
+    plot([orig_img] + imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_021.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_021.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 269-274
+
+TrivialAugmentWide
+~~~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.TrivialAugmentWide` is an alternate implementation of AutoAugment.
+However, instead of transforming an image multiple times, it transforms an image only once
+using a random transform from a given list with a random strength number.
+
+.. GENERATED FROM PYTHON SOURCE LINES 274-278
+
+.. code-block:: default
+
+    augmenter = v2.TrivialAugmentWide()
+    imgs = [augmenter(orig_img) for _ in range(4)]
+    plot([orig_img] + imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_022.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_022.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 279-282
+
+AugMix
+~~~~~~
+The :class:`~torchvision.transforms.AugMix` transform interpolates between augmented versions of an image.
+
+.. GENERATED FROM PYTHON SOURCE LINES 282-286
+
+.. code-block:: default
+
+    augmenter = v2.AugMix()
+    imgs = [augmenter(orig_img) for _ in range(4)]
+    plot([orig_img] + imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_023.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_023.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 287-299
+
+Randomly-applied Transforms
+---------------------------
+
+The following transforms are randomly-applied given a probability ``p``.  That is, given ``p = 0.5``,
+there is a 50% chance to return the original image, and a 50% chance to return the transformed image,
+even when called with the same transform instance!
+
+RandomHorizontalFlip
+~~~~~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomHorizontalFlip` transform
+(see also :func:`~torchvision.transforms.functional.hflip`)
+performs horizontal flip of an image, with a given probability.
+
+.. GENERATED FROM PYTHON SOURCE LINES 299-303
+
+.. code-block:: default
+
+    hflipper = v2.RandomHorizontalFlip(p=0.5)
+    transformed_imgs = [hflipper(orig_img) for _ in range(4)]
+    plot([orig_img] + transformed_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_024.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_024.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 304-309
+
+RandomVerticalFlip
+~~~~~~~~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomVerticalFlip` transform
+(see also :func:`~torchvision.transforms.functional.vflip`)
+performs vertical flip of an image, with a given probability.
+
+.. GENERATED FROM PYTHON SOURCE LINES 309-313
+
+.. code-block:: default
+
+    vflipper = v2.RandomVerticalFlip(p=0.5)
+    transformed_imgs = [vflipper(orig_img) for _ in range(4)]
+    plot([orig_img] + transformed_imgs)
+
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_025.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_025.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 314-318
+
+RandomApply
+~~~~~~~~~~~
+The :class:`~torchvision.transforms.RandomApply` transform
+randomly applies a list of transforms, with a given probability.
+
+.. GENERATED FROM PYTHON SOURCE LINES 318-321
+
+.. code-block:: default
+
+    applier = v2.RandomApply(transforms=[v2.RandomCrop(size=(64, 64))], p=0.5)
+    transformed_imgs = [applier(orig_img) for _ in range(4)]
+    plot([orig_img] + transformed_imgs)
+
+
+
+.. image-sg:: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_026.png
+   :alt: plot transforms illustrations
+   :srcset: /auto_examples/transforms/images/sphx_glr_plot_transforms_illustrations_026.png
+   :class: sphx-glr-single-img
+
+
+
+
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 8.772 seconds)
+
+
+.. _sphx_glr_download_auto_examples_transforms_plot_transforms_illustrations.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_transforms_illustrations.py <plot_transforms_illustrations.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_transforms_illustrations.ipynb <plot_transforms_illustrations.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/transforms/plot_tv_tensors.rst.txt b/0.16/_sources/auto_examples/transforms/plot_tv_tensors.rst.txt
new file mode 100644
index 00000000000..053bb29a8d8
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/plot_tv_tensors.rst.txt
@@ -0,0 +1,439 @@
+
+.. DO NOT EDIT.
+.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
+.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
+.. "auto_examples/transforms/plot_tv_tensors.py"
+.. LINE NUMBERS ARE GIVEN BELOW.
+
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_auto_examples_transforms_plot_tv_tensors.py:
+
+
+=============
+TVTensors FAQ
+=============
+
+.. note::
+    Try on `collab <https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_tv_tensors.ipynb>`_
+    or :ref:`go to the end <sphx_glr_download_auto_examples_transforms_plot_tv_tensors.py>` to download the full example code.
+
+
+TVTensors are Tensor subclasses introduced together with
+``torchvision.transforms.v2``. This example showcases what these TVTensors are
+and how they behave.
+
+.. warning::
+
+    **Intended Audience** Unless you're writing your own transforms or your own TVTensors, you
+    probably do not need to read this guide. This is a fairly low-level topic
+    that most users will not need to worry about: you do not need to understand
+    the internals of TVTensors to efficiently rely on
+    ``torchvision.transforms.v2``. It may however be useful for advanced users
+    trying to implement their own datasets, transforms, or work directly with
+    the TVTensors.
+
+.. GENERATED FROM PYTHON SOURCE LINES 27-33
+
+.. code-block:: default
+
+    import PIL.Image
+
+    import torch
+    from torchvision import tv_tensors
+
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 34-38
+
+What are TVTensors?
+-------------------
+
+TVTensors are zero-copy tensor subclasses:
+
+.. GENERATED FROM PYTHON SOURCE LINES 38-45
+
+.. code-block:: default
+
+
+    tensor = torch.rand(3, 256, 256)
+    image = tv_tensors.Image(tensor)
+
+    assert isinstance(image, torch.Tensor)
+    assert image.data_ptr() == tensor.data_ptr()
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 46-63
+
+Under the hood, they are needed in :mod:`torchvision.transforms.v2` to correctly dispatch to the appropriate function
+for the input data.
+
+:mod:`torchvision.tv_tensors` supports four types of TVTensors:
+
+* :class:`~torchvision.tv_tensors.Image`
+* :class:`~torchvision.tv_tensors.Video`
+* :class:`~torchvision.tv_tensors.BoundingBoxes`
+* :class:`~torchvision.tv_tensors.Mask`
+
+What can I do with a TVTensor?
+------------------------------
+
+TVTensors look and feel just like regular tensors - they **are** tensors.
+Everything that is supported on a plain :class:`torch.Tensor` like ``.sum()`` or
+any ``torch.*`` operator will also work on TVTensors. See
+:ref:`tv_tensor_unwrapping_behaviour` for a few gotchas.
+
+.. GENERATED FROM PYTHON SOURCE LINES 65-74
+
+.. _tv_tensor_creation:
+
+How do I construct a TVTensor?
+------------------------------
+
+Using the constructor
+^^^^^^^^^^^^^^^^^^^^^
+
+Each TVTensor class takes any tensor-like data that can be turned into a :class:`~torch.Tensor`
+
+.. GENERATED FROM PYTHON SOURCE LINES 74-79
+
+.. code-block:: default
+
+
+    image = tv_tensors.Image([[[[0, 1], [1, 0]]]])
+    print(image)
+
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Image([[[[0, 1],
+             [1, 0]]]], )
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 80-82
+
+Similar to other PyTorch creations ops, the constructor also takes the ``dtype``, ``device``, and ``requires_grad``
+parameters.
+
+.. GENERATED FROM PYTHON SOURCE LINES 82-87
+
+.. code-block:: default
+
+
+    float_image = tv_tensors.Image([[[0, 1], [1, 0]]], dtype=torch.float32, requires_grad=True)
+    print(float_image)
+
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Image([[[0., 1.],
+            [1., 0.]]], grad_fn=<AliasBackward0>, )
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 88-90
+
+In addition, :class:`~torchvision.tv_tensors.Image` and :class:`~torchvision.tv_tensors.Mask` can also take a
+:class:`PIL.Image.Image` directly:
+
+.. GENERATED FROM PYTHON SOURCE LINES 90-94
+
+.. code-block:: default
+
+
+    image = tv_tensors.Image(PIL.Image.open("../assets/astronaut.jpg"))
+    print(image.shape, image.dtype)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    torch.Size([3, 512, 512]) torch.uint8
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 95-99
+
+Some TVTensors require additional metadata to be passed in ordered to be constructed. For example,
+:class:`~torchvision.tv_tensors.BoundingBoxes` requires the coordinate format as well as the size of the
+corresponding image (``canvas_size``) alongside the actual values. These
+metadata are required to properly transform the bounding boxes.
+
+.. GENERATED FROM PYTHON SOURCE LINES 99-107
+
+.. code-block:: default
+
+
+    bboxes = tv_tensors.BoundingBoxes(
+        [[17, 16, 344, 495], [0, 10, 0, 10]],
+        format=tv_tensors.BoundingBoxFormat.XYXY,
+        canvas_size=image.shape[-2:]
+    )
+    print(bboxes)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    BoundingBoxes([[ 17,  16, 344, 495],
+                   [  0,  10,   0,  10]], format=BoundingBoxFormat.XYXY, canvas_size=torch.Size([512, 512]))
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 108-115
+
+Using ``tv_tensors.wrap()``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+You can also use the :func:`~torchvision.tv_tensors.wrap` function to wrap a tensor object
+into a TVTensor. This is useful when you already have an object of the
+desired type, which typically happens when writing transforms: you just want
+to wrap the output like the input.
+
+.. GENERATED FROM PYTHON SOURCE LINES 115-121
+
+.. code-block:: default
+
+
+    new_bboxes = torch.tensor([0, 20, 30, 40])
+    new_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)
+    assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+    assert new_bboxes.canvas_size == bboxes.canvas_size
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 122-132
+
+The metadata of ``new_bboxes`` is the same as ``bboxes``, but you could pass
+it as a parameter to override it.
+
+.. _tv_tensor_unwrapping_behaviour:
+
+I had a TVTensor but now I have a Tensor. Help!
+-----------------------------------------------
+
+By default, operations on :class:`~torchvision.tv_tensors.TVTensor` objects
+will return a pure Tensor:
+
+.. GENERATED FROM PYTHON SOURCE LINES 132-142
+
+.. code-block:: default
+
+
+
+    assert isinstance(bboxes, tv_tensors.BoundingBoxes)
+
+    # Shift bboxes by 3 pixels in both H and W
+    new_bboxes = bboxes + 3
+
+    assert isinstance(new_bboxes, torch.Tensor)
+    assert not isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 143-149
+
+.. note::
+
+   This behavior only affects native ``torch`` operations. If you are using
+   the built-in ``torchvision`` transforms or functionals, you will always get
+   as output the same type that you passed as input (pure ``Tensor`` or
+   ``TVTensor``).
+
+.. GENERATED FROM PYTHON SOURCE LINES 151-157
+
+But I want a TVTensor back!
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+You can re-wrap a pure tensor into a TVTensor by just calling the TVTensor
+constructor, or by using the :func:`~torchvision.tv_tensors.wrap` function
+(see more details above in :ref:`tv_tensor_creation`):
+
+.. GENERATED FROM PYTHON SOURCE LINES 157-162
+
+.. code-block:: default
+
+
+    new_bboxes = bboxes + 3
+    new_bboxes = tv_tensors.wrap(new_bboxes, like=bboxes)
+    assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 163-166
+
+Alternatively, you can use the :func:`~torchvision.tv_tensors.set_return_type`
+as a global config setting for the whole program, or as a context manager
+(read its docs to learn more about caveats):
+
+.. GENERATED FROM PYTHON SOURCE LINES 166-171
+
+.. code-block:: default
+
+
+    with tv_tensors.set_return_type("TVTensor"):
+        new_bboxes = bboxes + 3
+    assert isinstance(new_bboxes, tv_tensors.BoundingBoxes)
+
+
+
+
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 172-211
+
+Why is this happening?
+^^^^^^^^^^^^^^^^^^^^^^
+
+**For performance reasons**. :class:`~torchvision.tv_tensors.TVTensor`
+classes are Tensor subclasses, so any operation involving a
+:class:`~torchvision.tv_tensors.TVTensor` object will go through the
+`__torch_function__
+<https://pytorch.org/docs/stable/notes/extending.html#extending-torch>`_
+protocol. This induces a small overhead, which we want to avoid when possible.
+This doesn't matter for built-in ``torchvision`` transforms because we can
+avoid the overhead there, but it could be a problem in your model's
+``forward``.
+
+**The alternative isn't much better anyway.** For every operation where
+preserving the :class:`~torchvision.tv_tensors.TVTensor` type makes
+sense, there are just as many operations where returning a pure Tensor is
+preferable: for example, is ``img.sum()`` still an :class:`~torchvision.tv_tensors.Image`?
+If we were to preserve :class:`~torchvision.tv_tensors.TVTensor` types all
+the way, even model's logits or the output of the loss function would end up
+being of type :class:`~torchvision.tv_tensors.Image`, and surely that's not
+desirable.
+
+.. note::
+
+   This behaviour is something we're actively seeking feedback on. If you find this surprising or if you
+   have any suggestions on how to better support your use-cases, please reach out to us via this issue:
+   https://github.com/pytorch/vision/issues/7319
+
+Exceptions
+^^^^^^^^^^
+
+There are a few exceptions to this "unwrapping" rule:
+:meth:`~torch.Tensor.clone`, :meth:`~torch.Tensor.to`,
+:meth:`torch.Tensor.detach`, and :meth:`~torch.Tensor.requires_grad_` retain
+the TVTensor type.
+
+Inplace operations on TVTensors like ``obj.add_()`` will preserve the type of
+``obj``. However, the **returned** value of inplace operations will be a pure
+tensor:
+
+.. GENERATED FROM PYTHON SOURCE LINES 211-225
+
+.. code-block:: default
+
+
+    image = tv_tensors.Image([[[0, 1], [1, 0]]])
+
+    new_image = image.add_(1).mul_(2)
+
+    # image got transformed in-place and is still a TVTensor Image, but new_image
+    # is a Tensor. They share the same underlying data and they're equal, just
+    # different classes.
+    assert isinstance(image, tv_tensors.Image)
+    print(image)
+
+    assert isinstance(new_image, torch.Tensor) and not isinstance(new_image, tv_tensors.Image)
+    assert (new_image == image).all()
+    assert new_image.data_ptr() == image.data_ptr()
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    Image([[[2, 4],
+            [4, 2]]], )
+
+
+
+
+
+.. rst-class:: sphx-glr-timing
+
+   **Total running time of the script:** (0 minutes 0.008 seconds)
+
+
+.. _sphx_glr_download_auto_examples_transforms_plot_tv_tensors.py:
+
+.. only:: html
+
+  .. container:: sphx-glr-footer sphx-glr-footer-example
+
+
+
+
+    .. container:: sphx-glr-download sphx-glr-download-python
+
+      :download:`Download Python source code: plot_tv_tensors.py <plot_tv_tensors.py>`
+
+    .. container:: sphx-glr-download sphx-glr-download-jupyter
+
+      :download:`Download Jupyter notebook: plot_tv_tensors.ipynb <plot_tv_tensors.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/0.16/_sources/auto_examples/transforms/sg_execution_times.rst.txt b/0.16/_sources/auto_examples/transforms/sg_execution_times.rst.txt
new file mode 100644
index 00000000000..b864e3e998a
--- /dev/null
+++ b/0.16/_sources/auto_examples/transforms/sg_execution_times.rst.txt
@@ -0,0 +1,25 @@
+
+:orphan:
+
+.. _sphx_glr_auto_examples_transforms_sg_execution_times:
+
+
+Computation times
+=================
+**00:14.597** total execution time for **auto_examples_transforms** files:
+
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_transforms_plot_transforms_illustrations.py` (``plot_transforms_illustrations.py``)     | 00:08.772 | 0.0 MB |
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_transforms_plot_transforms_e2e.py` (``plot_transforms_e2e.py``)                         | 00:04.813 | 0.0 MB |
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_transforms_plot_transforms_getting_started.py` (``plot_transforms_getting_started.py``) | 00:00.838 | 0.0 MB |
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py` (``plot_cutmix_mixup.py``)                             | 00:00.155 | 0.0 MB |
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py` (``plot_tv_tensors.py``)                                 | 00:00.008 | 0.0 MB |
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_transforms_plot_custom_transforms.py` (``plot_custom_transforms.py``)                   | 00:00.006 | 0.0 MB |
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_auto_examples_transforms_plot_custom_tv_tensors.py` (``plot_custom_tv_tensors.py``)                   | 00:00.004 | 0.0 MB |
++----------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/0.16/_sources/datasets.rst.txt b/0.16/_sources/datasets.rst.txt
new file mode 100644
index 00000000000..588c1f781ed
--- /dev/null
+++ b/0.16/_sources/datasets.rst.txt
@@ -0,0 +1,182 @@
+.. _datasets:
+
+Datasets
+========
+
+Torchvision provides many built-in datasets in the ``torchvision.datasets``
+module, as well as utility classes for building your own datasets.
+
+Built-in datasets
+-----------------
+
+All datasets are subclasses of :class:`torch.utils.data.Dataset`
+i.e, they have ``__getitem__`` and ``__len__`` methods implemented.
+Hence, they can all be passed to a :class:`torch.utils.data.DataLoader`
+which can load multiple samples in parallel using ``torch.multiprocessing`` workers.
+For example: ::
+
+    imagenet_data = torchvision.datasets.ImageNet('path/to/imagenet_root/')
+    data_loader = torch.utils.data.DataLoader(imagenet_data,
+                                              batch_size=4,
+                                              shuffle=True,
+                                              num_workers=args.nThreads)
+
+.. currentmodule:: torchvision.datasets
+
+All the datasets have almost similar API. They all have two common arguments:
+``transform`` and  ``target_transform`` to transform the input and target respectively.
+You can also create your own datasets using the provided :ref:`base classes <base_classes_datasets>`.
+
+Image classification
+~~~~~~~~~~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    Caltech101
+    Caltech256
+    CelebA
+    CIFAR10
+    CIFAR100
+    Country211
+    DTD
+    EMNIST
+    EuroSAT
+    FakeData
+    FashionMNIST
+    FER2013
+    FGVCAircraft
+    Flickr8k
+    Flickr30k
+    Flowers102
+    Food101
+    GTSRB
+    INaturalist
+    ImageNet
+    KMNIST
+    LFWPeople
+    LSUN
+    MNIST
+    Omniglot
+    OxfordIIITPet
+    Places365
+    PCAM
+    QMNIST
+    RenderedSST2
+    SEMEION
+    SBU
+    StanfordCars
+    STL10
+    SUN397
+    SVHN
+    USPS
+
+Image detection or segmentation
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    CocoDetection
+    CelebA
+    Cityscapes
+    Kitti
+    OxfordIIITPet
+    SBDataset
+    VOCSegmentation
+    VOCDetection
+    WIDERFace
+
+Optical Flow
+~~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    FlyingChairs
+    FlyingThings3D
+    HD1K
+    KittiFlow
+    Sintel
+
+Stereo Matching
+~~~~~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    CarlaStereo
+    Kitti2012Stereo
+    Kitti2015Stereo
+    CREStereo
+    FallingThingsStereo
+    SceneFlowStereo
+    SintelStereo
+    InStereo2k
+    ETH3DStereo
+    Middlebury2014Stereo
+
+Image pairs
+~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    LFWPairs
+    PhotoTour
+
+Image captioning
+~~~~~~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    CocoCaptions
+
+Video classification
+~~~~~~~~~~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    HMDB51
+    Kinetics
+    UCF101
+
+Video prediction
+~~~~~~~~~~~~~~~~~~~~
+
+.. autosummary::
+    :toctree: generated/
+    :template: class_dataset.rst
+
+    MovingMNIST
+
+.. _base_classes_datasets:
+
+Base classes for custom datasets
+--------------------------------
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    DatasetFolder
+    ImageFolder
+    VisionDataset
+
+Transforms v2
+-------------
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    wrap_dataset_for_transforms_v2
diff --git a/0.16/_sources/feature_extraction.rst.txt b/0.16/_sources/feature_extraction.rst.txt
new file mode 100644
index 00000000000..e83bc2fe4bc
--- /dev/null
+++ b/0.16/_sources/feature_extraction.rst.txt
@@ -0,0 +1,166 @@
+Feature extraction for model inspection
+=======================================
+
+.. currentmodule:: torchvision.models.feature_extraction
+
+The ``torchvision.models.feature_extraction`` package contains
+feature extraction utilities that let us tap into our models to access intermediate
+transformations of our inputs. This could be useful for a variety of
+applications in computer vision. Just a few examples are:
+
+- Visualizing feature maps.
+- Extracting features to compute image descriptors for tasks like facial
+  recognition, copy-detection, or image retrieval.
+- Passing selected features to downstream sub-networks for end-to-end training
+  with a specific task in mind. For example, passing a hierarchy of features
+  to a Feature Pyramid Network with object detection heads.
+
+Torchvision provides :func:`create_feature_extractor` for this purpose.
+It works by following roughly these steps:
+
+1. Symbolically tracing the model to get a graphical representation of
+   how it transforms the input, step by step.
+2. Setting the user-selected graph nodes as outputs.
+3. Removing all redundant nodes (anything downstream of the output nodes).
+4. Generating python code from the resulting graph and bundling that into a
+   PyTorch module together with the graph itself.
+
+|
+
+The `torch.fx documentation <https://pytorch.org/docs/stable/fx.html>`_
+provides a more general and detailed explanation of the above procedure and
+the inner workings of the symbolic tracing.
+
+.. _about-node-names:
+
+**About Node Names**
+
+In order to specify which nodes should be output nodes for extracted
+features, one should be familiar with the node naming convention used here
+(which differs slightly from that used in ``torch.fx``). A node name is
+specified as a ``.`` separated path walking the module hierarchy from top level
+module down to leaf operation or leaf module. For instance ``"layer4.2.relu"``
+in ResNet-50 represents the output of the ReLU of the 2nd block of the 4th
+layer of the ``ResNet`` module. Here are some finer points to keep in mind:
+
+- When specifying node names for :func:`create_feature_extractor`, you may
+  provide a truncated version of a node name as a shortcut. To see how this
+  works, try creating a ResNet-50 model and printing the node names with
+  ``train_nodes, _ = get_graph_node_names(model) print(train_nodes)`` and
+  observe that the last node pertaining to ``layer4`` is
+  ``"layer4.2.relu_2"``. One may specify ``"layer4.2.relu_2"`` as the return
+  node, or just ``"layer4"`` as this, by convention, refers to the last node
+  (in order of execution) of ``layer4``.
+- If a certain module or operation is repeated more than once, node names get
+  an additional ``_{int}`` postfix to disambiguate. For instance, maybe the
+  addition (``+``) operation is used three times in the same ``forward``
+  method. Then there would be ``"path.to.module.add"``,
+  ``"path.to.module.add_1"``, ``"path.to.module.add_2"``. The counter is
+  maintained within the scope of the direct parent. So in ResNet-50 there is
+  a ``"layer4.1.add"`` and a ``"layer4.2.add"``. Because the addition
+  operations reside in different blocks, there is no need for a postfix to
+  disambiguate.
+
+
+**An Example**
+
+Here is an example of how we might extract features for MaskRCNN:
+
+.. code-block:: python
+
+  import torch
+  from torchvision.models import resnet50
+  from torchvision.models.feature_extraction import get_graph_node_names
+  from torchvision.models.feature_extraction import create_feature_extractor
+  from torchvision.models.detection.mask_rcnn import MaskRCNN
+  from torchvision.models.detection.backbone_utils import LastLevelMaxPool
+  from torchvision.ops.feature_pyramid_network import FeaturePyramidNetwork
+
+
+  # To assist you in designing the feature extractor you may want to print out
+  # the available nodes for resnet50.
+  m = resnet50()
+  train_nodes, eval_nodes = get_graph_node_names(resnet50())
+
+  # The lists returned, are the names of all the graph nodes (in order of
+  # execution) for the input model traced in train mode and in eval mode
+  # respectively. You'll find that `train_nodes` and `eval_nodes` are the same
+  # for this example. But if the model contains control flow that's dependent
+  # on the training mode, they may be different.
+
+  # To specify the nodes you want to extract, you could select the final node
+  # that appears in each of the main layers:
+  return_nodes = {
+      # node_name: user-specified key for output dict
+      'layer1.2.relu_2': 'layer1',
+      'layer2.3.relu_2': 'layer2',
+      'layer3.5.relu_2': 'layer3',
+      'layer4.2.relu_2': 'layer4',
+  }
+
+  # But `create_feature_extractor` can also accept truncated node specifications
+  # like "layer1", as it will just pick the last node that's a descendent of
+  # of the specification. (Tip: be careful with this, especially when a layer
+  # has multiple outputs. It's not always guaranteed that the last operation
+  # performed is the one that corresponds to the output you desire. You should
+  # consult the source code for the input model to confirm.)
+  return_nodes = {
+      'layer1': 'layer1',
+      'layer2': 'layer2',
+      'layer3': 'layer3',
+      'layer4': 'layer4',
+  }
+
+  # Now you can build the feature extractor. This returns a module whose forward
+  # method returns a dictionary like:
+  # {
+  #     'layer1': output of layer 1,
+  #     'layer2': output of layer 2,
+  #     'layer3': output of layer 3,
+  #     'layer4': output of layer 4,
+  # }
+  create_feature_extractor(m, return_nodes=return_nodes)
+
+  # Let's put all that together to wrap resnet50 with MaskRCNN
+
+  # MaskRCNN requires a backbone with an attached FPN
+  class Resnet50WithFPN(torch.nn.Module):
+      def __init__(self):
+          super(Resnet50WithFPN, self).__init__()
+          # Get a resnet50 backbone
+          m = resnet50()
+          # Extract 4 main layers (note: MaskRCNN needs this particular name
+          # mapping for return nodes)
+          self.body = create_feature_extractor(
+              m, return_nodes={f'layer{k}': str(v)
+                               for v, k in enumerate([1, 2, 3, 4])})
+          # Dry run to get number of channels for FPN
+          inp = torch.randn(2, 3, 224, 224)
+          with torch.no_grad():
+              out = self.body(inp)
+          in_channels_list = [o.shape[1] for o in out.values()]
+          # Build FPN
+          self.out_channels = 256
+          self.fpn = FeaturePyramidNetwork(
+              in_channels_list, out_channels=self.out_channels,
+              extra_blocks=LastLevelMaxPool())
+
+      def forward(self, x):
+          x = self.body(x)
+          x = self.fpn(x)
+          return x
+
+
+  # Now we can build our model!
+  model = MaskRCNN(Resnet50WithFPN(), num_classes=91).eval()
+
+
+API Reference
+-------------
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    create_feature_extractor
+    get_graph_node_names
diff --git a/0.16/_sources/generated/classification_quant_table.rst.txt b/0.16/_sources/generated/classification_quant_table.rst.txt
new file mode 100644
index 00000000000..12913be9f31
--- /dev/null
+++ b/0.16/_sources/generated/classification_quant_table.rst.txt
@@ -0,0 +1,23 @@
+.. rst-class:: table-weights
+.. table::
+    :widths: 110 18 18 18 18 10 
+
+    ========================================================================================================  ===========  ===========  ============  ==========  ===============================================================================================================
+    **Weight**                                                                                                  **Acc@1**    **Acc@5**  **Params**      **GIPS**  **Recipe**
+    ========================================================================================================  ===========  ===========  ============  ==========  ===============================================================================================================
+    :class:`GoogLeNet_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <GoogLeNet_QuantizedWeights>`                          69.826       89.404  6.6M                1.5   `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <Inception_V3_QuantizedWeights>`                    77.176       93.354  27.2M               5.71  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`MobileNet_V2_QuantizedWeights.IMAGENET1K_QNNPACK_V1 <MobileNet_V2_QuantizedWeights>`                   71.658       90.15   3.5M                0.3   `link <https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv2>`__
+    :class:`MobileNet_V3_Large_QuantizedWeights.IMAGENET1K_QNNPACK_V1 <MobileNet_V3_Large_QuantizedWeights>`       73.004       90.858  5.5M                0.22  `link <https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv3>`__
+    :class:`ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ResNeXt101_32X8D_QuantizedWeights>`            78.986       94.48   88.8M              16.41  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V2 <ResNeXt101_32X8D_QuantizedWeights>`            82.574       96.132  88.8M              16.41  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`ResNeXt101_64X4D_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ResNeXt101_64X4D_QuantizedWeights>`            82.898       96.326  83.5M              15.46  `link <https://github.com/pytorch/vision/pull/5935>`__
+    :class:`ResNet18_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ResNet18_QuantizedWeights>`                            69.494       88.882  11.7M               1.81  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ResNet50_QuantizedWeights>`                            75.92        92.814  25.6M               4.09  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V2 <ResNet50_QuantizedWeights>`                            80.282       94.976  25.6M               4.09  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`ShuffleNet_V2_X0_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ShuffleNet_V2_X0_5_QuantizedWeights>`        57.972       79.78   1.4M                0.04  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`ShuffleNet_V2_X1_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ShuffleNet_V2_X1_0_QuantizedWeights>`        68.36        87.582  2.3M                0.14  `link <https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models>`__
+    :class:`ShuffleNet_V2_X1_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ShuffleNet_V2_X1_5_QuantizedWeights>`        72.052       90.7    3.5M                0.3   `link <https://github.com/pytorch/vision/pull/5906>`__
+    :class:`ShuffleNet_V2_X2_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1 <ShuffleNet_V2_X2_0_QuantizedWeights>`        75.354       92.488  7.4M                0.58  `link <https://github.com/pytorch/vision/pull/5906>`__
+    ========================================================================================================  ===========  ===========  ============  ==========  ===============================================================================================================
+
diff --git a/0.16/_sources/generated/classification_table.rst.txt b/0.16/_sources/generated/classification_table.rst.txt
new file mode 100644
index 00000000000..bd1595331b4
--- /dev/null
+++ b/0.16/_sources/generated/classification_table.rst.txt
@@ -0,0 +1,124 @@
+.. rst-class:: table-weights
+.. table::
+    :widths: 110 18 18 18 18 10 
+
+    ==================================================================================  ===========  ===========  ============  ============  =========================================================================================================
+    **Weight**                                                                            **Acc@1**    **Acc@5**  **Params**      **GFLOPS**  **Recipe**
+    ==================================================================================  ===========  ===========  ============  ============  =========================================================================================================
+    :class:`AlexNet_Weights.IMAGENET1K_V1 <AlexNet_Weights>`                                 56.522       79.066  61.1M                 0.71  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`ConvNeXt_Base_Weights.IMAGENET1K_V1 <ConvNeXt_Base_Weights>`                     84.062       96.87   88.6M                15.36  `link <https://github.com/pytorch/vision/tree/main/references/classification#convnext>`__
+    :class:`ConvNeXt_Large_Weights.IMAGENET1K_V1 <ConvNeXt_Large_Weights>`                   84.414       96.976  197.8M               34.36  `link <https://github.com/pytorch/vision/tree/main/references/classification#convnext>`__
+    :class:`ConvNeXt_Small_Weights.IMAGENET1K_V1 <ConvNeXt_Small_Weights>`                   83.616       96.65   50.2M                 8.68  `link <https://github.com/pytorch/vision/tree/main/references/classification#convnext>`__
+    :class:`ConvNeXt_Tiny_Weights.IMAGENET1K_V1 <ConvNeXt_Tiny_Weights>`                     82.52        96.146  28.6M                 4.46  `link <https://github.com/pytorch/vision/tree/main/references/classification#convnext>`__
+    :class:`DenseNet121_Weights.IMAGENET1K_V1 <DenseNet121_Weights>`                         74.434       91.972  8.0M                  2.83  `link <https://github.com/pytorch/vision/pull/116>`__
+    :class:`DenseNet161_Weights.IMAGENET1K_V1 <DenseNet161_Weights>`                         77.138       93.56   28.7M                 7.73  `link <https://github.com/pytorch/vision/pull/116>`__
+    :class:`DenseNet169_Weights.IMAGENET1K_V1 <DenseNet169_Weights>`                         75.6         92.806  14.1M                 3.36  `link <https://github.com/pytorch/vision/pull/116>`__
+    :class:`DenseNet201_Weights.IMAGENET1K_V1 <DenseNet201_Weights>`                         76.896       93.37   20.0M                 4.29  `link <https://github.com/pytorch/vision/pull/116>`__
+    :class:`EfficientNet_B0_Weights.IMAGENET1K_V1 <EfficientNet_B0_Weights>`                 77.692       93.532  5.3M                  0.39  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_B1_Weights.IMAGENET1K_V1 <EfficientNet_B1_Weights>`                 78.642       94.186  7.8M                  0.69  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_B1_Weights.IMAGENET1K_V2 <EfficientNet_B1_Weights>`                 79.838       94.934  7.8M                  0.69  `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-lr-wd-crop-tuning>`__
+    :class:`EfficientNet_B2_Weights.IMAGENET1K_V1 <EfficientNet_B2_Weights>`                 80.608       95.31   9.1M                  1.09  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_B3_Weights.IMAGENET1K_V1 <EfficientNet_B3_Weights>`                 82.008       96.054  12.2M                 1.83  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_B4_Weights.IMAGENET1K_V1 <EfficientNet_B4_Weights>`                 83.384       96.594  19.3M                 4.39  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_B5_Weights.IMAGENET1K_V1 <EfficientNet_B5_Weights>`                 83.444       96.628  30.4M                10.27  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_B6_Weights.IMAGENET1K_V1 <EfficientNet_B6_Weights>`                 84.008       96.916  43.0M                19.07  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_B7_Weights.IMAGENET1K_V1 <EfficientNet_B7_Weights>`                 84.122       96.908  66.3M                37.75  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1>`__
+    :class:`EfficientNet_V2_L_Weights.IMAGENET1K_V1 <EfficientNet_V2_L_Weights>`             85.808       97.788  118.5M               56.08  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2>`__
+    :class:`EfficientNet_V2_M_Weights.IMAGENET1K_V1 <EfficientNet_V2_M_Weights>`             85.112       97.156  54.1M                24.58  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2>`__
+    :class:`EfficientNet_V2_S_Weights.IMAGENET1K_V1 <EfficientNet_V2_S_Weights>`             84.228       96.878  21.5M                 8.37  `link <https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2>`__
+    :class:`GoogLeNet_Weights.IMAGENET1K_V1 <GoogLeNet_Weights>`                             69.778       89.53   6.6M                  1.5   `link <https://github.com/pytorch/vision/tree/main/references/classification#googlenet>`__
+    :class:`Inception_V3_Weights.IMAGENET1K_V1 <Inception_V3_Weights>`                       77.294       93.45   27.2M                 5.71  `link <https://github.com/pytorch/vision/tree/main/references/classification#inception-v3>`__
+    :class:`MNASNet0_5_Weights.IMAGENET1K_V1 <MNASNet0_5_Weights>`                           67.734       87.49   2.2M                  0.1   `link <https://github.com/1e100/mnasnet_trainer>`__
+    :class:`MNASNet0_75_Weights.IMAGENET1K_V1 <MNASNet0_75_Weights>`                         71.18        90.496  3.2M                  0.21  `link <https://github.com/pytorch/vision/pull/6019>`__
+    :class:`MNASNet1_0_Weights.IMAGENET1K_V1 <MNASNet1_0_Weights>`                           73.456       91.51   4.4M                  0.31  `link <https://github.com/1e100/mnasnet_trainer>`__
+    :class:`MNASNet1_3_Weights.IMAGENET1K_V1 <MNASNet1_3_Weights>`                           76.506       93.522  6.3M                  0.53  `link <https://github.com/pytorch/vision/pull/6019>`__
+    :class:`MaxVit_T_Weights.IMAGENET1K_V1 <MaxVit_T_Weights>`                               83.7         96.722  30.9M                 5.56  `link <https://github.com/pytorch/vision/tree/main/references/classification#maxvit>`__
+    :class:`MobileNet_V2_Weights.IMAGENET1K_V1 <MobileNet_V2_Weights>`                       71.878       90.286  3.5M                  0.3   `link <https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv2>`__
+    :class:`MobileNet_V2_Weights.IMAGENET1K_V2 <MobileNet_V2_Weights>`                       72.154       90.822  3.5M                  0.3   `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning>`__
+    :class:`MobileNet_V3_Large_Weights.IMAGENET1K_V1 <MobileNet_V3_Large_Weights>`           74.042       91.34   5.5M                  0.22  `link <https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small>`__
+    :class:`MobileNet_V3_Large_Weights.IMAGENET1K_V2 <MobileNet_V3_Large_Weights>`           75.274       92.566  5.5M                  0.22  `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning>`__
+    :class:`MobileNet_V3_Small_Weights.IMAGENET1K_V1 <MobileNet_V3_Small_Weights>`           67.668       87.402  2.5M                  0.06  `link <https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small>`__
+    :class:`RegNet_X_16GF_Weights.IMAGENET1K_V1 <RegNet_X_16GF_Weights>`                     80.058       94.944  54.3M                15.94  `link <https://github.com/pytorch/vision/tree/main/references/classification#medium-models>`__
+    :class:`RegNet_X_16GF_Weights.IMAGENET1K_V2 <RegNet_X_16GF_Weights>`                     82.716       96.196  54.3M                15.94  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_X_1_6GF_Weights.IMAGENET1K_V1 <RegNet_X_1_6GF_Weights>`                   77.04        93.44   9.2M                  1.6   `link <https://github.com/pytorch/vision/tree/main/references/classification#small-models>`__
+    :class:`RegNet_X_1_6GF_Weights.IMAGENET1K_V2 <RegNet_X_1_6GF_Weights>`                   79.668       94.922  9.2M                  1.6   `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres>`__
+    :class:`RegNet_X_32GF_Weights.IMAGENET1K_V1 <RegNet_X_32GF_Weights>`                     80.622       95.248  107.8M               31.74  `link <https://github.com/pytorch/vision/tree/main/references/classification#large-models>`__
+    :class:`RegNet_X_32GF_Weights.IMAGENET1K_V2 <RegNet_X_32GF_Weights>`                     83.014       96.288  107.8M               31.74  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_X_3_2GF_Weights.IMAGENET1K_V1 <RegNet_X_3_2GF_Weights>`                   78.364       93.992  15.3M                 3.18  `link <https://github.com/pytorch/vision/tree/main/references/classification#medium-models>`__
+    :class:`RegNet_X_3_2GF_Weights.IMAGENET1K_V2 <RegNet_X_3_2GF_Weights>`                   81.196       95.43   15.3M                 3.18  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_X_400MF_Weights.IMAGENET1K_V1 <RegNet_X_400MF_Weights>`                   72.834       90.95   5.5M                  0.41  `link <https://github.com/pytorch/vision/tree/main/references/classification#small-models>`__
+    :class:`RegNet_X_400MF_Weights.IMAGENET1K_V2 <RegNet_X_400MF_Weights>`                   74.864       92.322  5.5M                  0.41  `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres>`__
+    :class:`RegNet_X_800MF_Weights.IMAGENET1K_V1 <RegNet_X_800MF_Weights>`                   75.212       92.348  7.3M                  0.8   `link <https://github.com/pytorch/vision/tree/main/references/classification#small-models>`__
+    :class:`RegNet_X_800MF_Weights.IMAGENET1K_V2 <RegNet_X_800MF_Weights>`                   77.522       93.826  7.3M                  0.8   `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres>`__
+    :class:`RegNet_X_8GF_Weights.IMAGENET1K_V1 <RegNet_X_8GF_Weights>`                       79.344       94.686  39.6M                 8     `link <https://github.com/pytorch/vision/tree/main/references/classification#medium-models>`__
+    :class:`RegNet_X_8GF_Weights.IMAGENET1K_V2 <RegNet_X_8GF_Weights>`                       81.682       95.678  39.6M                 8     `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_E2E_V1 <RegNet_Y_128GF_Weights>`          88.228       98.682  644.8M              374.57  `link <https://github.com/facebookresearch/SWAG>`__
+    :class:`RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_LINEAR_V1 <RegNet_Y_128GF_Weights>`       86.068       97.844  644.8M              127.52  `link <https://github.com/pytorch/vision/pull/5793>`__
+    :class:`RegNet_Y_16GF_Weights.IMAGENET1K_V1 <RegNet_Y_16GF_Weights>`                     80.424       95.24   83.6M                15.91  `link <https://github.com/pytorch/vision/tree/main/references/classification#large-models>`__
+    :class:`RegNet_Y_16GF_Weights.IMAGENET1K_V2 <RegNet_Y_16GF_Weights>`                     82.886       96.328  83.6M                15.91  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_E2E_V1 <RegNet_Y_16GF_Weights>`            86.012       98.054  83.6M                46.73  `link <https://github.com/facebookresearch/SWAG>`__
+    :class:`RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_LINEAR_V1 <RegNet_Y_16GF_Weights>`         83.976       97.244  83.6M                15.91  `link <https://github.com/pytorch/vision/pull/5793>`__
+    :class:`RegNet_Y_1_6GF_Weights.IMAGENET1K_V1 <RegNet_Y_1_6GF_Weights>`                   77.95        93.966  11.2M                 1.61  `link <https://github.com/pytorch/vision/tree/main/references/classification#small-models>`__
+    :class:`RegNet_Y_1_6GF_Weights.IMAGENET1K_V2 <RegNet_Y_1_6GF_Weights>`                   80.876       95.444  11.2M                 1.61  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_Y_32GF_Weights.IMAGENET1K_V1 <RegNet_Y_32GF_Weights>`                     80.878       95.34   145.0M               32.28  `link <https://github.com/pytorch/vision/tree/main/references/classification#large-models>`__
+    :class:`RegNet_Y_32GF_Weights.IMAGENET1K_V2 <RegNet_Y_32GF_Weights>`                     83.368       96.498  145.0M               32.28  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_E2E_V1 <RegNet_Y_32GF_Weights>`            86.838       98.362  145.0M               94.83  `link <https://github.com/facebookresearch/SWAG>`__
+    :class:`RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_LINEAR_V1 <RegNet_Y_32GF_Weights>`         84.622       97.48   145.0M               32.28  `link <https://github.com/pytorch/vision/pull/5793>`__
+    :class:`RegNet_Y_3_2GF_Weights.IMAGENET1K_V1 <RegNet_Y_3_2GF_Weights>`                   78.948       94.576  19.4M                 3.18  `link <https://github.com/pytorch/vision/tree/main/references/classification#medium-models>`__
+    :class:`RegNet_Y_3_2GF_Weights.IMAGENET1K_V2 <RegNet_Y_3_2GF_Weights>`                   81.982       95.972  19.4M                 3.18  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_Y_400MF_Weights.IMAGENET1K_V1 <RegNet_Y_400MF_Weights>`                   74.046       91.716  4.3M                  0.4   `link <https://github.com/pytorch/vision/tree/main/references/classification#small-models>`__
+    :class:`RegNet_Y_400MF_Weights.IMAGENET1K_V2 <RegNet_Y_400MF_Weights>`                   75.804       92.742  4.3M                  0.4   `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_Y_800MF_Weights.IMAGENET1K_V1 <RegNet_Y_800MF_Weights>`                   76.42        93.136  6.4M                  0.83  `link <https://github.com/pytorch/vision/tree/main/references/classification#small-models>`__
+    :class:`RegNet_Y_800MF_Weights.IMAGENET1K_V2 <RegNet_Y_800MF_Weights>`                   78.828       94.502  6.4M                  0.83  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`RegNet_Y_8GF_Weights.IMAGENET1K_V1 <RegNet_Y_8GF_Weights>`                       80.032       95.048  39.4M                 8.47  `link <https://github.com/pytorch/vision/tree/main/references/classification#medium-models>`__
+    :class:`RegNet_Y_8GF_Weights.IMAGENET1K_V2 <RegNet_Y_8GF_Weights>`                       82.828       96.33   39.4M                 8.47  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`ResNeXt101_32X8D_Weights.IMAGENET1K_V1 <ResNeXt101_32X8D_Weights>`               79.312       94.526  88.8M                16.41  `link <https://github.com/pytorch/vision/tree/main/references/classification#resnext>`__
+    :class:`ResNeXt101_32X8D_Weights.IMAGENET1K_V2 <ResNeXt101_32X8D_Weights>`               82.834       96.228  88.8M                16.41  `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres>`__
+    :class:`ResNeXt101_64X4D_Weights.IMAGENET1K_V1 <ResNeXt101_64X4D_Weights>`               83.246       96.454  83.5M                15.46  `link <https://github.com/pytorch/vision/pull/5935>`__
+    :class:`ResNeXt50_32X4D_Weights.IMAGENET1K_V1 <ResNeXt50_32X4D_Weights>`                 77.618       93.698  25.0M                 4.23  `link <https://github.com/pytorch/vision/tree/main/references/classification#resnext>`__
+    :class:`ResNeXt50_32X4D_Weights.IMAGENET1K_V2 <ResNeXt50_32X4D_Weights>`                 81.198       95.34   25.0M                 4.23  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`ResNet101_Weights.IMAGENET1K_V1 <ResNet101_Weights>`                             77.374       93.546  44.5M                 7.8   `link <https://github.com/pytorch/vision/tree/main/references/classification#resnet>`__
+    :class:`ResNet101_Weights.IMAGENET1K_V2 <ResNet101_Weights>`                             81.886       95.78   44.5M                 7.8   `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`ResNet152_Weights.IMAGENET1K_V1 <ResNet152_Weights>`                             78.312       94.046  60.2M                11.51  `link <https://github.com/pytorch/vision/tree/main/references/classification#resnet>`__
+    :class:`ResNet152_Weights.IMAGENET1K_V2 <ResNet152_Weights>`                             82.284       96.002  60.2M                11.51  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`ResNet18_Weights.IMAGENET1K_V1 <ResNet18_Weights>`                               69.758       89.078  11.7M                 1.81  `link <https://github.com/pytorch/vision/tree/main/references/classification#resnet>`__
+    :class:`ResNet34_Weights.IMAGENET1K_V1 <ResNet34_Weights>`                               73.314       91.42   21.8M                 3.66  `link <https://github.com/pytorch/vision/tree/main/references/classification#resnet>`__
+    :class:`ResNet50_Weights.IMAGENET1K_V1 <ResNet50_Weights>`                               76.13        92.862  25.6M                 4.09  `link <https://github.com/pytorch/vision/tree/main/references/classification#resnet>`__
+    :class:`ResNet50_Weights.IMAGENET1K_V2 <ResNet50_Weights>`                               80.858       95.434  25.6M                 4.09  `link <https://github.com/pytorch/vision/issues/3995#issuecomment-1013906621>`__
+    :class:`ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1 <ShuffleNet_V2_X0_5_Weights>`           60.552       81.746  1.4M                  0.04  `link <https://github.com/ericsun99/Shufflenet-v2-Pytorch>`__
+    :class:`ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1 <ShuffleNet_V2_X1_0_Weights>`           69.362       88.316  2.3M                  0.14  `link <https://github.com/ericsun99/Shufflenet-v2-Pytorch>`__
+    :class:`ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1 <ShuffleNet_V2_X1_5_Weights>`           72.996       91.086  3.5M                  0.3   `link <https://github.com/pytorch/vision/pull/5906>`__
+    :class:`ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1 <ShuffleNet_V2_X2_0_Weights>`           76.23        93.006  7.4M                  0.58  `link <https://github.com/pytorch/vision/pull/5906>`__
+    :class:`SqueezeNet1_0_Weights.IMAGENET1K_V1 <SqueezeNet1_0_Weights>`                     58.092       80.42   1.2M                  0.82  `link <https://github.com/pytorch/vision/pull/49#issuecomment-277560717>`__
+    :class:`SqueezeNet1_1_Weights.IMAGENET1K_V1 <SqueezeNet1_1_Weights>`                     58.178       80.624  1.2M                  0.35  `link <https://github.com/pytorch/vision/pull/49#issuecomment-277560717>`__
+    :class:`Swin_B_Weights.IMAGENET1K_V1 <Swin_B_Weights>`                                   83.582       96.64   87.8M                15.43  `link <https://github.com/pytorch/vision/tree/main/references/classification#swintransformer>`__
+    :class:`Swin_S_Weights.IMAGENET1K_V1 <Swin_S_Weights>`                                   83.196       96.36   49.6M                 8.74  `link <https://github.com/pytorch/vision/tree/main/references/classification#swintransformer>`__
+    :class:`Swin_T_Weights.IMAGENET1K_V1 <Swin_T_Weights>`                                   81.474       95.776  28.3M                 4.49  `link <https://github.com/pytorch/vision/tree/main/references/classification#swintransformer>`__
+    :class:`Swin_V2_B_Weights.IMAGENET1K_V1 <Swin_V2_B_Weights>`                             84.112       96.864  87.9M                20.32  `link <https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2>`__
+    :class:`Swin_V2_S_Weights.IMAGENET1K_V1 <Swin_V2_S_Weights>`                             83.712       96.816  49.7M                11.55  `link <https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2>`__
+    :class:`Swin_V2_T_Weights.IMAGENET1K_V1 <Swin_V2_T_Weights>`                             82.072       96.132  28.4M                 5.94  `link <https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2>`__
+    :class:`VGG11_BN_Weights.IMAGENET1K_V1 <VGG11_BN_Weights>`                               70.37        89.81   132.9M                7.61  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`VGG11_Weights.IMAGENET1K_V1 <VGG11_Weights>`                                     69.02        88.628  132.9M                7.61  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`VGG13_BN_Weights.IMAGENET1K_V1 <VGG13_BN_Weights>`                               71.586       90.374  133.1M               11.31  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`VGG13_Weights.IMAGENET1K_V1 <VGG13_Weights>`                                     69.928       89.246  133.0M               11.31  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`VGG16_BN_Weights.IMAGENET1K_V1 <VGG16_BN_Weights>`                               73.36        91.516  138.4M               15.47  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`VGG16_Weights.IMAGENET1K_V1 <VGG16_Weights>`                                     71.592       90.382  138.4M               15.47  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`VGG16_Weights.IMAGENET1K_FEATURES <VGG16_Weights>`                              nan          nan      138.4M               15.47  `link <https://github.com/amdegroot/ssd.pytorch#training-ssd>`__
+    :class:`VGG19_BN_Weights.IMAGENET1K_V1 <VGG19_BN_Weights>`                               74.218       91.842  143.7M               19.63  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`VGG19_Weights.IMAGENET1K_V1 <VGG19_Weights>`                                     72.376       90.876  143.7M               19.63  `link <https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg>`__
+    :class:`ViT_B_16_Weights.IMAGENET1K_V1 <ViT_B_16_Weights>`                               81.072       95.318  86.6M                17.56  `link <https://github.com/pytorch/vision/tree/main/references/classification#vit_b_16>`__
+    :class:`ViT_B_16_Weights.IMAGENET1K_SWAG_E2E_V1 <ViT_B_16_Weights>`                      85.304       97.65   86.9M                55.48  `link <https://github.com/facebookresearch/SWAG>`__
+    :class:`ViT_B_16_Weights.IMAGENET1K_SWAG_LINEAR_V1 <ViT_B_16_Weights>`                   81.886       96.18   86.6M                17.56  `link <https://github.com/pytorch/vision/pull/5793>`__
+    :class:`ViT_B_32_Weights.IMAGENET1K_V1 <ViT_B_32_Weights>`                               75.912       92.466  88.2M                 4.41  `link <https://github.com/pytorch/vision/tree/main/references/classification#vit_b_32>`__
+    :class:`ViT_H_14_Weights.IMAGENET1K_SWAG_E2E_V1 <ViT_H_14_Weights>`                      88.552       98.694  633.5M             1016.72  `link <https://github.com/facebookresearch/SWAG>`__
+    :class:`ViT_H_14_Weights.IMAGENET1K_SWAG_LINEAR_V1 <ViT_H_14_Weights>`                   85.708       97.73   632.0M              167.29  `link <https://github.com/pytorch/vision/pull/5793>`__
+    :class:`ViT_L_16_Weights.IMAGENET1K_V1 <ViT_L_16_Weights>`                               79.662       94.638  304.3M               61.55  `link <https://github.com/pytorch/vision/tree/main/references/classification#vit_l_16>`__
+    :class:`ViT_L_16_Weights.IMAGENET1K_SWAG_E2E_V1 <ViT_L_16_Weights>`                      88.064       98.512  305.2M              361.99  `link <https://github.com/facebookresearch/SWAG>`__
+    :class:`ViT_L_16_Weights.IMAGENET1K_SWAG_LINEAR_V1 <ViT_L_16_Weights>`                   85.146       97.422  304.3M               61.55  `link <https://github.com/pytorch/vision/pull/5793>`__
+    :class:`ViT_L_32_Weights.IMAGENET1K_V1 <ViT_L_32_Weights>`                               76.972       93.07   306.5M               15.38  `link <https://github.com/pytorch/vision/tree/main/references/classification#vit_l_32>`__
+    :class:`Wide_ResNet101_2_Weights.IMAGENET1K_V1 <Wide_ResNet101_2_Weights>`               78.848       94.284  126.9M               22.75  `link <https://github.com/pytorch/vision/pull/912#issue-445437439>`__
+    :class:`Wide_ResNet101_2_Weights.IMAGENET1K_V2 <Wide_ResNet101_2_Weights>`               82.51        96.02   126.9M               22.75  `link <https://github.com/pytorch/vision/issues/3995#new-recipe>`__
+    :class:`Wide_ResNet50_2_Weights.IMAGENET1K_V1 <Wide_ResNet50_2_Weights>`                 78.468       94.086  68.9M                11.4   `link <https://github.com/pytorch/vision/pull/912#issue-445437439>`__
+    :class:`Wide_ResNet50_2_Weights.IMAGENET1K_V2 <Wide_ResNet50_2_Weights>`                 81.602       95.758  68.9M                11.4   `link <https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres>`__
+    ==================================================================================  ===========  ===========  ============  ============  =========================================================================================================
+
diff --git a/0.16/_sources/generated/detection_keypoint_table.rst.txt b/0.16/_sources/generated/detection_keypoint_table.rst.txt
new file mode 100644
index 00000000000..82dee0ca7e1
--- /dev/null
+++ b/0.16/_sources/generated/detection_keypoint_table.rst.txt
@@ -0,0 +1,11 @@
+.. rst-class:: table-weights
+.. table::
+    :widths: 110 18 18 18 18 10 
+
+    ==========================================================================================  =============  ==================  ============  ============  ==========================================================================================
+    **Weight**                                                                                    **Box MAP**    **Keypoint MAP**  **Params**      **GFLOPS**  **Recipe**
+    ==========================================================================================  =============  ==================  ============  ============  ==========================================================================================
+    :class:`KeypointRCNN_ResNet50_FPN_Weights.COCO_LEGACY <KeypointRCNN_ResNet50_FPN_Weights>`           50.6                61.1  59.1M               133.92  `link <https://github.com/pytorch/vision/issues/1606>`__
+    :class:`KeypointRCNN_ResNet50_FPN_Weights.COCO_V1 <KeypointRCNN_ResNet50_FPN_Weights>`               54.6                65    59.1M               137.42  `link <https://github.com/pytorch/vision/tree/main/references/detection#keypoint-r-cnn>`__
+    ==========================================================================================  =============  ==================  ============  ============  ==========================================================================================
+
diff --git a/0.16/_sources/generated/detection_table.rst.txt b/0.16/_sources/generated/detection_table.rst.txt
new file mode 100644
index 00000000000..2f0905fb206
--- /dev/null
+++ b/0.16/_sources/generated/detection_table.rst.txt
@@ -0,0 +1,18 @@
+.. rst-class:: table-weights
+.. table::
+    :widths: 110 18 18 18 10 
+
+    ==============================================================================================================  =============  ============  ============  ==================================================================================================================
+    **Weight**                                                                                                        **Box MAP**  **Params**      **GFLOPS**  **Recipe**
+    ==============================================================================================================  =============  ============  ============  ==================================================================================================================
+    :class:`FCOS_ResNet50_FPN_Weights.COCO_V1 <FCOS_ResNet50_FPN_Weights>`                                                   39.2  32.3M               128.21  `link <https://github.com/pytorch/vision/tree/main/references/detection#fcos-resnet-50-fpn>`__
+    :class:`FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.COCO_V1 <FasterRCNN_MobileNet_V3_Large_320_FPN_Weights>`           22.8  19.4M                 0.72  `link <https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-320-fpn>`__
+    :class:`FasterRCNN_MobileNet_V3_Large_FPN_Weights.COCO_V1 <FasterRCNN_MobileNet_V3_Large_FPN_Weights>`                   32.8  19.4M                 4.49  `link <https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-fpn>`__
+    :class:`FasterRCNN_ResNet50_FPN_V2_Weights.COCO_V1 <FasterRCNN_ResNet50_FPN_V2_Weights>`                                 46.7  43.7M               280.37  `link <https://github.com/pytorch/vision/pull/5763>`__
+    :class:`FasterRCNN_ResNet50_FPN_Weights.COCO_V1 <FasterRCNN_ResNet50_FPN_Weights>`                                       37    41.8M               134.38  `link <https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-resnet-50-fpn>`__
+    :class:`RetinaNet_ResNet50_FPN_V2_Weights.COCO_V1 <RetinaNet_ResNet50_FPN_V2_Weights>`                                   41.5  38.2M               152.24  `link <https://github.com/pytorch/vision/pull/5756>`__
+    :class:`RetinaNet_ResNet50_FPN_Weights.COCO_V1 <RetinaNet_ResNet50_FPN_Weights>`                                         36.4  34.0M               151.54  `link <https://github.com/pytorch/vision/tree/main/references/detection#retinanet>`__
+    :class:`SSD300_VGG16_Weights.COCO_V1 <SSD300_VGG16_Weights>`                                                             25.1  35.6M                34.86  `link <https://github.com/pytorch/vision/tree/main/references/detection#ssd300-vgg16>`__
+    :class:`SSDLite320_MobileNet_V3_Large_Weights.COCO_V1 <SSDLite320_MobileNet_V3_Large_Weights>`                           21.3  3.4M                  0.58  `link <https://github.com/pytorch/vision/tree/main/references/detection#ssdlite320-mobilenetv3-large>`__
+    ==============================================================================================================  =============  ============  ============  ==================================================================================================================
+
diff --git a/0.16/_sources/generated/instance_segmentation_table.rst.txt b/0.16/_sources/generated/instance_segmentation_table.rst.txt
new file mode 100644
index 00000000000..f72da571212
--- /dev/null
+++ b/0.16/_sources/generated/instance_segmentation_table.rst.txt
@@ -0,0 +1,11 @@
+.. rst-class:: table-weights
+.. table::
+    :widths: 110 18 18 18 18 10 
+
+    ====================================================================================  =============  ==============  ============  ============  ======================================================================================
+    **Weight**                                                                              **Box MAP**    **Mask MAP**  **Params**      **GFLOPS**  **Recipe**
+    ====================================================================================  =============  ==============  ============  ============  ======================================================================================
+    :class:`MaskRCNN_ResNet50_FPN_V2_Weights.COCO_V1 <MaskRCNN_ResNet50_FPN_V2_Weights>`           47.4            41.8  46.4M               333.58  `link <https://github.com/pytorch/vision/pull/5773>`__
+    :class:`MaskRCNN_ResNet50_FPN_Weights.COCO_V1 <MaskRCNN_ResNet50_FPN_Weights>`                 37.9            34.6  44.4M               134.38  `link <https://github.com/pytorch/vision/tree/main/references/detection#mask-r-cnn>`__
+    ====================================================================================  =============  ==============  ============  ============  ======================================================================================
+
diff --git a/0.16/_sources/generated/segmentation_table.rst.txt b/0.16/_sources/generated/segmentation_table.rst.txt
new file mode 100644
index 00000000000..9907827ad99
--- /dev/null
+++ b/0.16/_sources/generated/segmentation_table.rst.txt
@@ -0,0 +1,15 @@
+.. rst-class:: table-weights
+.. table::
+    :widths: 110 18 18 18 18 10 
+
+    ============================================================================================================  ==============  ===================  ============  ============  ===========================================================================================================
+    **Weight**                                                                                                      **Mean IoU**    **pixelwise Acc**  **Params**      **GFLOPS**  **Recipe**
+    ============================================================================================================  ==============  ===================  ============  ============  ===========================================================================================================
+    :class:`DeepLabV3_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1 <DeepLabV3_MobileNet_V3_Large_Weights>`            60.3                 91.2  11.0M                10.45  `link <https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_mobilenet_v3_large>`__
+    :class:`DeepLabV3_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1 <DeepLabV3_ResNet101_Weights>`                              67.4                 92.4  61.0M               258.74  `link <https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet101>`__
+    :class:`DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1 <DeepLabV3_ResNet50_Weights>`                                66.4                 92.4  42.0M               178.72  `link <https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet50>`__
+    :class:`FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1 <FCN_ResNet101_Weights>`                                          63.7                 91.9  54.3M               232.74  `link <https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet101>`__
+    :class:`FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1 <FCN_ResNet50_Weights>`                                            60.5                 91.4  35.3M               152.72  `link <https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet50>`__
+    :class:`LRASPP_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1 <LRASPP_MobileNet_V3_Large_Weights>`                  57.9                 91.2  3.2M                  2.09  `link <https://github.com/pytorch/vision/tree/main/references/segmentation#lraspp_mobilenet_v3_large>`__
+    ============================================================================================================  ==============  ===================  ============  ============  ===========================================================================================================
+
diff --git a/0.16/_sources/generated/torchvision.datasets.CIFAR10.rst.txt b/0.16/_sources/generated/torchvision.datasets.CIFAR10.rst.txt
new file mode 100644
index 00000000000..1308cfc9a9b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.CIFAR10.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+CIFAR10
+=======
+
+.. autoclass:: CIFAR10
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.CIFAR100.rst.txt b/0.16/_sources/generated/torchvision.datasets.CIFAR100.rst.txt
new file mode 100644
index 00000000000..b17aecebe35
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.CIFAR100.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+CIFAR100
+========
+
+.. autoclass:: CIFAR100
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.CREStereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.CREStereo.rst.txt
new file mode 100644
index 00000000000..00c6dbeb2b3
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.CREStereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+CREStereo
+=========
+
+.. autoclass:: CREStereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Caltech101.rst.txt b/0.16/_sources/generated/torchvision.datasets.Caltech101.rst.txt
new file mode 100644
index 00000000000..d8f90cd1165
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Caltech101.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Caltech101
+==========
+
+.. autoclass:: Caltech101
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Caltech256.rst.txt b/0.16/_sources/generated/torchvision.datasets.Caltech256.rst.txt
new file mode 100644
index 00000000000..ff6278851dd
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Caltech256.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Caltech256
+==========
+
+.. autoclass:: Caltech256
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.CarlaStereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.CarlaStereo.rst.txt
new file mode 100644
index 00000000000..17e1beaa003
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.CarlaStereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+CarlaStereo
+===========
+
+.. autoclass:: CarlaStereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.CelebA.rst.txt b/0.16/_sources/generated/torchvision.datasets.CelebA.rst.txt
new file mode 100644
index 00000000000..1d294226c21
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.CelebA.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+CelebA
+======
+
+.. autoclass:: CelebA
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Cityscapes.rst.txt b/0.16/_sources/generated/torchvision.datasets.Cityscapes.rst.txt
new file mode 100644
index 00000000000..c4fea0f5f4f
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Cityscapes.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Cityscapes
+==========
+
+.. autoclass:: Cityscapes
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.CocoCaptions.rst.txt b/0.16/_sources/generated/torchvision.datasets.CocoCaptions.rst.txt
new file mode 100644
index 00000000000..d0e759a41e5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.CocoCaptions.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+CocoCaptions
+============
+
+.. autoclass:: CocoCaptions
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.CocoDetection.rst.txt b/0.16/_sources/generated/torchvision.datasets.CocoDetection.rst.txt
new file mode 100644
index 00000000000..c21377a563e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.CocoDetection.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+CocoDetection
+=============
+
+.. autoclass:: CocoDetection
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Country211.rst.txt b/0.16/_sources/generated/torchvision.datasets.Country211.rst.txt
new file mode 100644
index 00000000000..ca037891450
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Country211.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Country211
+==========
+
+.. autoclass:: Country211
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.DTD.rst.txt b/0.16/_sources/generated/torchvision.datasets.DTD.rst.txt
new file mode 100644
index 00000000000..efca3601823
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.DTD.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+DTD
+===
+
+.. autoclass:: DTD
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.DatasetFolder.rst.txt b/0.16/_sources/generated/torchvision.datasets.DatasetFolder.rst.txt
new file mode 100644
index 00000000000..2e60348d5a5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.DatasetFolder.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+DatasetFolder
+=============
+
+.. autoclass:: DatasetFolder
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.EMNIST.rst.txt b/0.16/_sources/generated/torchvision.datasets.EMNIST.rst.txt
new file mode 100644
index 00000000000..365715a1e64
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.EMNIST.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+EMNIST
+======
+
+.. autoclass:: EMNIST
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.ETH3DStereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.ETH3DStereo.rst.txt
new file mode 100644
index 00000000000..8a54914d31c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.ETH3DStereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+ETH3DStereo
+===========
+
+.. autoclass:: ETH3DStereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.EuroSAT.rst.txt b/0.16/_sources/generated/torchvision.datasets.EuroSAT.rst.txt
new file mode 100644
index 00000000000..dd900588470
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.EuroSAT.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+EuroSAT
+=======
+
+.. autoclass:: EuroSAT
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.FER2013.rst.txt b/0.16/_sources/generated/torchvision.datasets.FER2013.rst.txt
new file mode 100644
index 00000000000..bf1eedf1700
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.FER2013.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+FER2013
+=======
+
+.. autoclass:: FER2013
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.FGVCAircraft.rst.txt b/0.16/_sources/generated/torchvision.datasets.FGVCAircraft.rst.txt
new file mode 100644
index 00000000000..515284f583a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.FGVCAircraft.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+FGVCAircraft
+============
+
+.. autoclass:: FGVCAircraft
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.FakeData.rst.txt b/0.16/_sources/generated/torchvision.datasets.FakeData.rst.txt
new file mode 100644
index 00000000000..3cd34fb9434
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.FakeData.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+FakeData
+========
+
+.. autoclass:: FakeData
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.FallingThingsStereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.FallingThingsStereo.rst.txt
new file mode 100644
index 00000000000..7e2a03c061b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.FallingThingsStereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+FallingThingsStereo
+===================
+
+.. autoclass:: FallingThingsStereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.FashionMNIST.rst.txt b/0.16/_sources/generated/torchvision.datasets.FashionMNIST.rst.txt
new file mode 100644
index 00000000000..a2fc4311f50
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.FashionMNIST.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+FashionMNIST
+============
+
+.. autoclass:: FashionMNIST
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Flickr30k.rst.txt b/0.16/_sources/generated/torchvision.datasets.Flickr30k.rst.txt
new file mode 100644
index 00000000000..df1ba2848a2
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Flickr30k.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Flickr30k
+=========
+
+.. autoclass:: Flickr30k
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Flickr8k.rst.txt b/0.16/_sources/generated/torchvision.datasets.Flickr8k.rst.txt
new file mode 100644
index 00000000000..25dc0196b71
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Flickr8k.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Flickr8k
+========
+
+.. autoclass:: Flickr8k
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Flowers102.rst.txt b/0.16/_sources/generated/torchvision.datasets.Flowers102.rst.txt
new file mode 100644
index 00000000000..3e429975bb5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Flowers102.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Flowers102
+==========
+
+.. autoclass:: Flowers102
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.FlyingChairs.rst.txt b/0.16/_sources/generated/torchvision.datasets.FlyingChairs.rst.txt
new file mode 100644
index 00000000000..245271ce0ae
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.FlyingChairs.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+FlyingChairs
+============
+
+.. autoclass:: FlyingChairs
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.FlyingThings3D.rst.txt b/0.16/_sources/generated/torchvision.datasets.FlyingThings3D.rst.txt
new file mode 100644
index 00000000000..4cbb087423f
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.FlyingThings3D.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+FlyingThings3D
+==============
+
+.. autoclass:: FlyingThings3D
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Food101.rst.txt b/0.16/_sources/generated/torchvision.datasets.Food101.rst.txt
new file mode 100644
index 00000000000..c7360608a17
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Food101.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Food101
+=======
+
+.. autoclass:: Food101
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.GTSRB.rst.txt b/0.16/_sources/generated/torchvision.datasets.GTSRB.rst.txt
new file mode 100644
index 00000000000..83a147d000e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.GTSRB.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+GTSRB
+=====
+
+.. autoclass:: GTSRB
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.HD1K.rst.txt b/0.16/_sources/generated/torchvision.datasets.HD1K.rst.txt
new file mode 100644
index 00000000000..52a62401ee9
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.HD1K.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+HD1K
+====
+
+.. autoclass:: HD1K
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.HMDB51.rst.txt b/0.16/_sources/generated/torchvision.datasets.HMDB51.rst.txt
new file mode 100644
index 00000000000..2f13270c1bf
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.HMDB51.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+HMDB51
+======
+
+.. autoclass:: HMDB51
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.INaturalist.rst.txt b/0.16/_sources/generated/torchvision.datasets.INaturalist.rst.txt
new file mode 100644
index 00000000000..72a9b28fe18
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.INaturalist.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+INaturalist
+===========
+
+.. autoclass:: INaturalist
+    :members:
+        __getitem__,
+         category_name 
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.ImageFolder.rst.txt b/0.16/_sources/generated/torchvision.datasets.ImageFolder.rst.txt
new file mode 100644
index 00000000000..cda59adee89
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.ImageFolder.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+ImageFolder
+===========
+
+.. autoclass:: ImageFolder
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.ImageNet.rst.txt b/0.16/_sources/generated/torchvision.datasets.ImageNet.rst.txt
new file mode 100644
index 00000000000..3101e3ddcb4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.ImageNet.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+ImageNet
+========
+
+.. autoclass:: ImageNet
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.InStereo2k.rst.txt b/0.16/_sources/generated/torchvision.datasets.InStereo2k.rst.txt
new file mode 100644
index 00000000000..7f277a52dfe
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.InStereo2k.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+InStereo2k
+==========
+
+.. autoclass:: InStereo2k
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.KMNIST.rst.txt b/0.16/_sources/generated/torchvision.datasets.KMNIST.rst.txt
new file mode 100644
index 00000000000..314f16040ca
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.KMNIST.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+KMNIST
+======
+
+.. autoclass:: KMNIST
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Kinetics.rst.txt b/0.16/_sources/generated/torchvision.datasets.Kinetics.rst.txt
new file mode 100644
index 00000000000..525f069fec6
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Kinetics.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Kinetics
+========
+
+.. autoclass:: Kinetics
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Kitti.rst.txt b/0.16/_sources/generated/torchvision.datasets.Kitti.rst.txt
new file mode 100644
index 00000000000..7ef67a5c518
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Kitti.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Kitti
+=====
+
+.. autoclass:: Kitti
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Kitti2012Stereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.Kitti2012Stereo.rst.txt
new file mode 100644
index 00000000000..25ca156311d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Kitti2012Stereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Kitti2012Stereo
+===============
+
+.. autoclass:: Kitti2012Stereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Kitti2015Stereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.Kitti2015Stereo.rst.txt
new file mode 100644
index 00000000000..ffa2c96414e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Kitti2015Stereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Kitti2015Stereo
+===============
+
+.. autoclass:: Kitti2015Stereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.KittiFlow.rst.txt b/0.16/_sources/generated/torchvision.datasets.KittiFlow.rst.txt
new file mode 100644
index 00000000000..77651ac4334
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.KittiFlow.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+KittiFlow
+=========
+
+.. autoclass:: KittiFlow
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.LFWPairs.rst.txt b/0.16/_sources/generated/torchvision.datasets.LFWPairs.rst.txt
new file mode 100644
index 00000000000..a9d228aa91a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.LFWPairs.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+LFWPairs
+========
+
+.. autoclass:: LFWPairs
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.LFWPeople.rst.txt b/0.16/_sources/generated/torchvision.datasets.LFWPeople.rst.txt
new file mode 100644
index 00000000000..e75433a8383
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.LFWPeople.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+LFWPeople
+=========
+
+.. autoclass:: LFWPeople
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.LSUN.rst.txt b/0.16/_sources/generated/torchvision.datasets.LSUN.rst.txt
new file mode 100644
index 00000000000..ac1360f2c3c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.LSUN.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+LSUN
+====
+
+.. autoclass:: LSUN
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.MNIST.rst.txt b/0.16/_sources/generated/torchvision.datasets.MNIST.rst.txt
new file mode 100644
index 00000000000..33acaa36c6a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.MNIST.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+MNIST
+=====
+
+.. autoclass:: MNIST
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Middlebury2014Stereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.Middlebury2014Stereo.rst.txt
new file mode 100644
index 00000000000..49cbf4ca1e6
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Middlebury2014Stereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Middlebury2014Stereo
+====================
+
+.. autoclass:: Middlebury2014Stereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.MovingMNIST.rst.txt b/0.16/_sources/generated/torchvision.datasets.MovingMNIST.rst.txt
new file mode 100644
index 00000000000..79eb581074a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.MovingMNIST.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+MovingMNIST
+===========
+
+.. autoclass:: MovingMNIST
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Omniglot.rst.txt b/0.16/_sources/generated/torchvision.datasets.Omniglot.rst.txt
new file mode 100644
index 00000000000..8cfee44a19a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Omniglot.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Omniglot
+========
+
+.. autoclass:: Omniglot
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.OxfordIIITPet.rst.txt b/0.16/_sources/generated/torchvision.datasets.OxfordIIITPet.rst.txt
new file mode 100644
index 00000000000..4cc85553219
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.OxfordIIITPet.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+OxfordIIITPet
+=============
+
+.. autoclass:: OxfordIIITPet
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.PCAM.rst.txt b/0.16/_sources/generated/torchvision.datasets.PCAM.rst.txt
new file mode 100644
index 00000000000..c2df2771e0e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.PCAM.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+PCAM
+====
+
+.. autoclass:: PCAM
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.PhotoTour.rst.txt b/0.16/_sources/generated/torchvision.datasets.PhotoTour.rst.txt
new file mode 100644
index 00000000000..7afafc32409
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.PhotoTour.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+PhotoTour
+=========
+
+.. autoclass:: PhotoTour
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Places365.rst.txt b/0.16/_sources/generated/torchvision.datasets.Places365.rst.txt
new file mode 100644
index 00000000000..6deedc358da
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Places365.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Places365
+=========
+
+.. autoclass:: Places365
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.QMNIST.rst.txt b/0.16/_sources/generated/torchvision.datasets.QMNIST.rst.txt
new file mode 100644
index 00000000000..3fc34d3d00a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.QMNIST.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+QMNIST
+======
+
+.. autoclass:: QMNIST
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.RenderedSST2.rst.txt b/0.16/_sources/generated/torchvision.datasets.RenderedSST2.rst.txt
new file mode 100644
index 00000000000..0aa7065c468
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.RenderedSST2.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+RenderedSST2
+============
+
+.. autoclass:: RenderedSST2
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.SBDataset.rst.txt b/0.16/_sources/generated/torchvision.datasets.SBDataset.rst.txt
new file mode 100644
index 00000000000..f8c027d8cb8
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.SBDataset.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+SBDataset
+=========
+
+.. autoclass:: SBDataset
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.SBU.rst.txt b/0.16/_sources/generated/torchvision.datasets.SBU.rst.txt
new file mode 100644
index 00000000000..f0c3601ebe1
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.SBU.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+SBU
+===
+
+.. autoclass:: SBU
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.SEMEION.rst.txt b/0.16/_sources/generated/torchvision.datasets.SEMEION.rst.txt
new file mode 100644
index 00000000000..816f68b9bb7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.SEMEION.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+SEMEION
+=======
+
+.. autoclass:: SEMEION
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.STL10.rst.txt b/0.16/_sources/generated/torchvision.datasets.STL10.rst.txt
new file mode 100644
index 00000000000..ff38a2aaf8c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.STL10.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+STL10
+=====
+
+.. autoclass:: STL10
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.SUN397.rst.txt b/0.16/_sources/generated/torchvision.datasets.SUN397.rst.txt
new file mode 100644
index 00000000000..91cfbbdd5d6
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.SUN397.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+SUN397
+======
+
+.. autoclass:: SUN397
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.SVHN.rst.txt b/0.16/_sources/generated/torchvision.datasets.SVHN.rst.txt
new file mode 100644
index 00000000000..dc1f45896cc
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.SVHN.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+SVHN
+====
+
+.. autoclass:: SVHN
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.SceneFlowStereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.SceneFlowStereo.rst.txt
new file mode 100644
index 00000000000..10ce9d3a644
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.SceneFlowStereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+SceneFlowStereo
+===============
+
+.. autoclass:: SceneFlowStereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.Sintel.rst.txt b/0.16/_sources/generated/torchvision.datasets.Sintel.rst.txt
new file mode 100644
index 00000000000..c5dd04c5680
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.Sintel.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+Sintel
+======
+
+.. autoclass:: Sintel
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.SintelStereo.rst.txt b/0.16/_sources/generated/torchvision.datasets.SintelStereo.rst.txt
new file mode 100644
index 00000000000..e26d91bca50
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.SintelStereo.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+SintelStereo
+============
+
+.. autoclass:: SintelStereo
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.StanfordCars.rst.txt b/0.16/_sources/generated/torchvision.datasets.StanfordCars.rst.txt
new file mode 100644
index 00000000000..3d6e2878c0d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.StanfordCars.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+StanfordCars
+============
+
+.. autoclass:: StanfordCars
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.UCF101.rst.txt b/0.16/_sources/generated/torchvision.datasets.UCF101.rst.txt
new file mode 100644
index 00000000000..a4951983c60
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.UCF101.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+UCF101
+======
+
+.. autoclass:: UCF101
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.USPS.rst.txt b/0.16/_sources/generated/torchvision.datasets.USPS.rst.txt
new file mode 100644
index 00000000000..1bdc9bdd921
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.USPS.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+USPS
+====
+
+.. autoclass:: USPS
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.VOCDetection.rst.txt b/0.16/_sources/generated/torchvision.datasets.VOCDetection.rst.txt
new file mode 100644
index 00000000000..caed621db23
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.VOCDetection.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+VOCDetection
+============
+
+.. autoclass:: VOCDetection
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.VOCSegmentation.rst.txt b/0.16/_sources/generated/torchvision.datasets.VOCSegmentation.rst.txt
new file mode 100644
index 00000000000..881c5780271
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.VOCSegmentation.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+VOCSegmentation
+===============
+
+.. autoclass:: VOCSegmentation
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.VisionDataset.rst.txt b/0.16/_sources/generated/torchvision.datasets.VisionDataset.rst.txt
new file mode 100644
index 00000000000..27d768a81b0
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.VisionDataset.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+VisionDataset
+=============
+
+.. autoclass:: VisionDataset
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.WIDERFace.rst.txt b/0.16/_sources/generated/torchvision.datasets.WIDERFace.rst.txt
new file mode 100644
index 00000000000..a5827131b1b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.WIDERFace.rst.txt
@@ -0,0 +1,13 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+WIDERFace
+=========
+
+.. autoclass:: WIDERFace
+    :members:
+        __getitem__,
+        
+    :special-members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.rst.txt b/0.16/_sources/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.rst.txt
new file mode 100644
index 00000000000..7f2bd5e5c9d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.datasets
+
+
+wrap_dataset_for_transforms_v2
+==============================
+
+.. autofunction:: wrap_dataset_for_transforms_v2
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.ImageReadMode.rst.txt b/0.16/_sources/generated/torchvision.io.ImageReadMode.rst.txt
new file mode 100644
index 00000000000..129b7fe399c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.ImageReadMode.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+ImageReadMode
+=============
+
+.. autoclass:: ImageReadMode
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.VideoReader.rst.txt b/0.16/_sources/generated/torchvision.io.VideoReader.rst.txt
new file mode 100644
index 00000000000..c42dce70a5c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.VideoReader.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+VideoReader
+===========
+
+.. autoclass:: VideoReader
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.decode_image.rst.txt b/0.16/_sources/generated/torchvision.io.decode_image.rst.txt
new file mode 100644
index 00000000000..b950d11ec24
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.decode_image.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+decode_image
+============
+
+.. autofunction:: decode_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.decode_jpeg.rst.txt b/0.16/_sources/generated/torchvision.io.decode_jpeg.rst.txt
new file mode 100644
index 00000000000..c8b7a0f09d6
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.decode_jpeg.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+decode_jpeg
+===========
+
+.. autofunction:: decode_jpeg
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.decode_png.rst.txt b/0.16/_sources/generated/torchvision.io.decode_png.rst.txt
new file mode 100644
index 00000000000..00b9b979eec
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.decode_png.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+decode_png
+==========
+
+.. autofunction:: decode_png
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.encode_jpeg.rst.txt b/0.16/_sources/generated/torchvision.io.encode_jpeg.rst.txt
new file mode 100644
index 00000000000..af6cb024a84
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.encode_jpeg.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+encode_jpeg
+===========
+
+.. autofunction:: encode_jpeg
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.encode_png.rst.txt b/0.16/_sources/generated/torchvision.io.encode_png.rst.txt
new file mode 100644
index 00000000000..05badd26a87
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.encode_png.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+encode_png
+==========
+
+.. autofunction:: encode_png
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.read_file.rst.txt b/0.16/_sources/generated/torchvision.io.read_file.rst.txt
new file mode 100644
index 00000000000..b56d976899e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.read_file.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+read_file
+=========
+
+.. autofunction:: read_file
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.read_image.rst.txt b/0.16/_sources/generated/torchvision.io.read_image.rst.txt
new file mode 100644
index 00000000000..54ae88643ba
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.read_image.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+read_image
+==========
+
+.. autofunction:: read_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.read_video.rst.txt b/0.16/_sources/generated/torchvision.io.read_video.rst.txt
new file mode 100644
index 00000000000..ca239e57936
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.read_video.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+read_video
+==========
+
+.. autofunction:: read_video
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.read_video_timestamps.rst.txt b/0.16/_sources/generated/torchvision.io.read_video_timestamps.rst.txt
new file mode 100644
index 00000000000..2272590bffd
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.read_video_timestamps.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+read_video_timestamps
+=====================
+
+.. autofunction:: read_video_timestamps
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.write_file.rst.txt b/0.16/_sources/generated/torchvision.io.write_file.rst.txt
new file mode 100644
index 00000000000..074cd2f7829
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.write_file.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+write_file
+==========
+
+.. autofunction:: write_file
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.write_jpeg.rst.txt b/0.16/_sources/generated/torchvision.io.write_jpeg.rst.txt
new file mode 100644
index 00000000000..bb0a779753a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.write_jpeg.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+write_jpeg
+==========
+
+.. autofunction:: write_jpeg
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.write_png.rst.txt b/0.16/_sources/generated/torchvision.io.write_png.rst.txt
new file mode 100644
index 00000000000..4b6327ae409
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.write_png.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+write_png
+=========
+
+.. autofunction:: write_png
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.io.write_video.rst.txt b/0.16/_sources/generated/torchvision.io.write_video.rst.txt
new file mode 100644
index 00000000000..25c387326e2
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.io.write_video.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.io
+
+
+write_video
+===========
+
+.. autofunction:: write_video
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.models.feature_extraction.create_feature_extractor.rst.txt b/0.16/_sources/generated/torchvision.models.feature_extraction.create_feature_extractor.rst.txt
new file mode 100644
index 00000000000..f0c04bde643
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.models.feature_extraction.create_feature_extractor.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.feature_extraction
+
+
+create_feature_extractor
+========================
+
+.. autofunction:: create_feature_extractor
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.models.feature_extraction.get_graph_node_names.rst.txt b/0.16/_sources/generated/torchvision.models.feature_extraction.get_graph_node_names.rst.txt
new file mode 100644
index 00000000000..cd9ef77fc21
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.models.feature_extraction.get_graph_node_names.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.feature_extraction
+
+
+get_graph_node_names
+====================
+
+.. autofunction:: get_graph_node_names
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.models.get_model.rst.txt b/0.16/_sources/generated/torchvision.models.get_model.rst.txt
new file mode 100644
index 00000000000..1cad02af92e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.models.get_model.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+get_model
+=========
+
+.. autofunction:: get_model
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.models.get_model_weights.rst.txt b/0.16/_sources/generated/torchvision.models.get_model_weights.rst.txt
new file mode 100644
index 00000000000..551c11340f1
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.models.get_model_weights.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+get_model_weights
+=================
+
+.. autofunction:: get_model_weights
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.models.get_weight.rst.txt b/0.16/_sources/generated/torchvision.models.get_weight.rst.txt
new file mode 100644
index 00000000000..697efe0c643
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.models.get_weight.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+get_weight
+==========
+
+.. autofunction:: get_weight
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.models.list_models.rst.txt b/0.16/_sources/generated/torchvision.models.list_models.rst.txt
new file mode 100644
index 00000000000..d8a60741957
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.models.list_models.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+list_models
+===========
+
+.. autofunction:: list_models
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.Conv2dNormActivation.rst.txt b/0.16/_sources/generated/torchvision.ops.Conv2dNormActivation.rst.txt
new file mode 100644
index 00000000000..327a14ab4d4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.Conv2dNormActivation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+Conv2dNormActivation
+====================
+
+.. autoclass:: Conv2dNormActivation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.Conv3dNormActivation.rst.txt b/0.16/_sources/generated/torchvision.ops.Conv3dNormActivation.rst.txt
new file mode 100644
index 00000000000..b14b62e8c8e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.Conv3dNormActivation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+Conv3dNormActivation
+====================
+
+.. autoclass:: Conv3dNormActivation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.DeformConv2d.rst.txt b/0.16/_sources/generated/torchvision.ops.DeformConv2d.rst.txt
new file mode 100644
index 00000000000..07f489a0731
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.DeformConv2d.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+DeformConv2d
+============
+
+.. autoclass:: DeformConv2d
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.DropBlock2d.rst.txt b/0.16/_sources/generated/torchvision.ops.DropBlock2d.rst.txt
new file mode 100644
index 00000000000..5934c89d36a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.DropBlock2d.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+DropBlock2d
+===========
+
+.. autoclass:: DropBlock2d
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.DropBlock3d.rst.txt b/0.16/_sources/generated/torchvision.ops.DropBlock3d.rst.txt
new file mode 100644
index 00000000000..64302f74e19
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.DropBlock3d.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+DropBlock3d
+===========
+
+.. autoclass:: DropBlock3d
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.FeaturePyramidNetwork.rst.txt b/0.16/_sources/generated/torchvision.ops.FeaturePyramidNetwork.rst.txt
new file mode 100644
index 00000000000..fa9e1020a7d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.FeaturePyramidNetwork.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+FeaturePyramidNetwork
+=====================
+
+.. autoclass:: FeaturePyramidNetwork
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.FrozenBatchNorm2d.rst.txt b/0.16/_sources/generated/torchvision.ops.FrozenBatchNorm2d.rst.txt
new file mode 100644
index 00000000000..53bb2d43d88
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.FrozenBatchNorm2d.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+FrozenBatchNorm2d
+=================
+
+.. autoclass:: FrozenBatchNorm2d
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.MLP.rst.txt b/0.16/_sources/generated/torchvision.ops.MLP.rst.txt
new file mode 100644
index 00000000000..08c70ce032a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.MLP.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+MLP
+===
+
+.. autoclass:: MLP
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.MultiScaleRoIAlign.rst.txt b/0.16/_sources/generated/torchvision.ops.MultiScaleRoIAlign.rst.txt
new file mode 100644
index 00000000000..65503977221
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.MultiScaleRoIAlign.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+MultiScaleRoIAlign
+==================
+
+.. autoclass:: MultiScaleRoIAlign
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.PSRoIAlign.rst.txt b/0.16/_sources/generated/torchvision.ops.PSRoIAlign.rst.txt
new file mode 100644
index 00000000000..1b1b30e9fbb
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.PSRoIAlign.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+PSRoIAlign
+==========
+
+.. autoclass:: PSRoIAlign
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.PSRoIPool.rst.txt b/0.16/_sources/generated/torchvision.ops.PSRoIPool.rst.txt
new file mode 100644
index 00000000000..f905b6a81c7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.PSRoIPool.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+PSRoIPool
+=========
+
+.. autoclass:: PSRoIPool
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.Permute.rst.txt b/0.16/_sources/generated/torchvision.ops.Permute.rst.txt
new file mode 100644
index 00000000000..29a4d8f0e11
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.Permute.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+Permute
+=======
+
+.. autoclass:: Permute
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.RoIAlign.rst.txt b/0.16/_sources/generated/torchvision.ops.RoIAlign.rst.txt
new file mode 100644
index 00000000000..926b6ccd8a6
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.RoIAlign.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+RoIAlign
+========
+
+.. autoclass:: RoIAlign
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.RoIPool.rst.txt b/0.16/_sources/generated/torchvision.ops.RoIPool.rst.txt
new file mode 100644
index 00000000000..3f2b4536650
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.RoIPool.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+RoIPool
+=======
+
+.. autoclass:: RoIPool
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.SqueezeExcitation.rst.txt b/0.16/_sources/generated/torchvision.ops.SqueezeExcitation.rst.txt
new file mode 100644
index 00000000000..54d7a1e3021
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.SqueezeExcitation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+SqueezeExcitation
+=================
+
+.. autoclass:: SqueezeExcitation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.StochasticDepth.rst.txt b/0.16/_sources/generated/torchvision.ops.StochasticDepth.rst.txt
new file mode 100644
index 00000000000..03b3ba953fa
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.StochasticDepth.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+StochasticDepth
+===============
+
+.. autoclass:: StochasticDepth
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.batched_nms.rst.txt b/0.16/_sources/generated/torchvision.ops.batched_nms.rst.txt
new file mode 100644
index 00000000000..7a132a373f0
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.batched_nms.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+batched_nms
+===========
+
+.. autofunction:: batched_nms
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.box_area.rst.txt b/0.16/_sources/generated/torchvision.ops.box_area.rst.txt
new file mode 100644
index 00000000000..e3821e3d1f5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.box_area.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+box_area
+========
+
+.. autofunction:: box_area
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.box_convert.rst.txt b/0.16/_sources/generated/torchvision.ops.box_convert.rst.txt
new file mode 100644
index 00000000000..1c0ef3e8718
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.box_convert.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+box_convert
+===========
+
+.. autofunction:: box_convert
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.box_iou.rst.txt b/0.16/_sources/generated/torchvision.ops.box_iou.rst.txt
new file mode 100644
index 00000000000..74b11d5f0ca
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.box_iou.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+box_iou
+=======
+
+.. autofunction:: box_iou
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.clip_boxes_to_image.rst.txt b/0.16/_sources/generated/torchvision.ops.clip_boxes_to_image.rst.txt
new file mode 100644
index 00000000000..fd800498ced
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.clip_boxes_to_image.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+clip_boxes_to_image
+===================
+
+.. autofunction:: clip_boxes_to_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.complete_box_iou.rst.txt b/0.16/_sources/generated/torchvision.ops.complete_box_iou.rst.txt
new file mode 100644
index 00000000000..7abc37659ba
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.complete_box_iou.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+complete_box_iou
+================
+
+.. autofunction:: complete_box_iou
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.complete_box_iou_loss.rst.txt b/0.16/_sources/generated/torchvision.ops.complete_box_iou_loss.rst.txt
new file mode 100644
index 00000000000..84d2f31419f
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.complete_box_iou_loss.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+complete_box_iou_loss
+=====================
+
+.. autofunction:: complete_box_iou_loss
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.deform_conv2d.rst.txt b/0.16/_sources/generated/torchvision.ops.deform_conv2d.rst.txt
new file mode 100644
index 00000000000..5c019e4fba1
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.deform_conv2d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+deform_conv2d
+=============
+
+.. autofunction:: deform_conv2d
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.distance_box_iou.rst.txt b/0.16/_sources/generated/torchvision.ops.distance_box_iou.rst.txt
new file mode 100644
index 00000000000..b0e327c2955
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.distance_box_iou.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+distance_box_iou
+================
+
+.. autofunction:: distance_box_iou
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.distance_box_iou_loss.rst.txt b/0.16/_sources/generated/torchvision.ops.distance_box_iou_loss.rst.txt
new file mode 100644
index 00000000000..4566a46f1a0
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.distance_box_iou_loss.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+distance_box_iou_loss
+=====================
+
+.. autofunction:: distance_box_iou_loss
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.drop_block2d.rst.txt b/0.16/_sources/generated/torchvision.ops.drop_block2d.rst.txt
new file mode 100644
index 00000000000..99187cf7fb7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.drop_block2d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+drop_block2d
+============
+
+.. autofunction:: drop_block2d
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.drop_block3d.rst.txt b/0.16/_sources/generated/torchvision.ops.drop_block3d.rst.txt
new file mode 100644
index 00000000000..ebf5bfd1a10
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.drop_block3d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+drop_block3d
+============
+
+.. autofunction:: drop_block3d
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.generalized_box_iou.rst.txt b/0.16/_sources/generated/torchvision.ops.generalized_box_iou.rst.txt
new file mode 100644
index 00000000000..70dba2e7409
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.generalized_box_iou.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+generalized_box_iou
+===================
+
+.. autofunction:: generalized_box_iou
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.generalized_box_iou_loss.rst.txt b/0.16/_sources/generated/torchvision.ops.generalized_box_iou_loss.rst.txt
new file mode 100644
index 00000000000..888346901fa
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.generalized_box_iou_loss.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+generalized_box_iou_loss
+========================
+
+.. autofunction:: generalized_box_iou_loss
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.masks_to_boxes.rst.txt b/0.16/_sources/generated/torchvision.ops.masks_to_boxes.rst.txt
new file mode 100644
index 00000000000..92baa43a427
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.masks_to_boxes.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+masks_to_boxes
+==============
+
+.. autofunction:: masks_to_boxes
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.nms.rst.txt b/0.16/_sources/generated/torchvision.ops.nms.rst.txt
new file mode 100644
index 00000000000..ff9ac2718b9
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.nms.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+nms
+===
+
+.. autofunction:: nms
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.ps_roi_align.rst.txt b/0.16/_sources/generated/torchvision.ops.ps_roi_align.rst.txt
new file mode 100644
index 00000000000..feb337ae487
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.ps_roi_align.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+ps_roi_align
+============
+
+.. autofunction:: ps_roi_align
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.ps_roi_pool.rst.txt b/0.16/_sources/generated/torchvision.ops.ps_roi_pool.rst.txt
new file mode 100644
index 00000000000..8b3504b3523
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.ps_roi_pool.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+ps_roi_pool
+===========
+
+.. autofunction:: ps_roi_pool
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.remove_small_boxes.rst.txt b/0.16/_sources/generated/torchvision.ops.remove_small_boxes.rst.txt
new file mode 100644
index 00000000000..46375c94465
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.remove_small_boxes.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+remove_small_boxes
+==================
+
+.. autofunction:: remove_small_boxes
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.roi_align.rst.txt b/0.16/_sources/generated/torchvision.ops.roi_align.rst.txt
new file mode 100644
index 00000000000..a772ae918bd
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.roi_align.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+roi_align
+=========
+
+.. autofunction:: roi_align
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.roi_pool.rst.txt b/0.16/_sources/generated/torchvision.ops.roi_pool.rst.txt
new file mode 100644
index 00000000000..a7e09d4c3af
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.roi_pool.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+roi_pool
+========
+
+.. autofunction:: roi_pool
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.sigmoid_focal_loss.rst.txt b/0.16/_sources/generated/torchvision.ops.sigmoid_focal_loss.rst.txt
new file mode 100644
index 00000000000..d7f7a17468d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.sigmoid_focal_loss.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+sigmoid_focal_loss
+==================
+
+.. autofunction:: sigmoid_focal_loss
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.ops.stochastic_depth.rst.txt b/0.16/_sources/generated/torchvision.ops.stochastic_depth.rst.txt
new file mode 100644
index 00000000000..d80fa907889
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.ops.stochastic_depth.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.ops
+
+
+stochastic_depth
+================
+
+.. autofunction:: stochastic_depth
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.AugMix.rst.txt b/0.16/_sources/generated/torchvision.transforms.AugMix.rst.txt
new file mode 100644
index 00000000000..e31abfc9c38
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.AugMix.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+AugMix
+======
+
+.. autoclass:: AugMix
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.AutoAugment.rst.txt b/0.16/_sources/generated/torchvision.transforms.AutoAugment.rst.txt
new file mode 100644
index 00000000000..452b15dd99b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.AutoAugment.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+AutoAugment
+===========
+
+.. autoclass:: AutoAugment
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.AutoAugmentPolicy.rst.txt b/0.16/_sources/generated/torchvision.transforms.AutoAugmentPolicy.rst.txt
new file mode 100644
index 00000000000..07950cf4519
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.AutoAugmentPolicy.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+AutoAugmentPolicy
+=================
+
+.. autoclass:: AutoAugmentPolicy
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.CenterCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.CenterCrop.rst.txt
new file mode 100644
index 00000000000..ac5acd7636e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.CenterCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+CenterCrop
+==========
+
+.. autoclass:: CenterCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.ColorJitter.rst.txt b/0.16/_sources/generated/torchvision.transforms.ColorJitter.rst.txt
new file mode 100644
index 00000000000..2e7b74ca187
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.ColorJitter.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+ColorJitter
+===========
+
+.. autoclass:: ColorJitter
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.Compose.rst.txt b/0.16/_sources/generated/torchvision.transforms.Compose.rst.txt
new file mode 100644
index 00000000000..f3c650219e9
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.Compose.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+Compose
+=======
+
+.. autoclass:: Compose
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.ConvertImageDtype.rst.txt b/0.16/_sources/generated/torchvision.transforms.ConvertImageDtype.rst.txt
new file mode 100644
index 00000000000..54850c32a35
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.ConvertImageDtype.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+ConvertImageDtype
+=================
+
+.. autoclass:: ConvertImageDtype
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.ElasticTransform.rst.txt b/0.16/_sources/generated/torchvision.transforms.ElasticTransform.rst.txt
new file mode 100644
index 00000000000..3d77938cb18
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.ElasticTransform.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+ElasticTransform
+================
+
+.. autoclass:: ElasticTransform
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.FiveCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.FiveCrop.rst.txt
new file mode 100644
index 00000000000..a9550854dde
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.FiveCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+FiveCrop
+========
+
+.. autoclass:: FiveCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.GaussianBlur.rst.txt b/0.16/_sources/generated/torchvision.transforms.GaussianBlur.rst.txt
new file mode 100644
index 00000000000..6ab95159ea4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.GaussianBlur.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+GaussianBlur
+============
+
+.. autoclass:: GaussianBlur
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.Grayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.Grayscale.rst.txt
new file mode 100644
index 00000000000..c301754d5eb
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.Grayscale.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+Grayscale
+=========
+
+.. autoclass:: Grayscale
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.Lambda.rst.txt b/0.16/_sources/generated/torchvision.transforms.Lambda.rst.txt
new file mode 100644
index 00000000000..a24ce81e21c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.Lambda.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+Lambda
+======
+
+.. autoclass:: Lambda
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.LinearTransformation.rst.txt b/0.16/_sources/generated/torchvision.transforms.LinearTransformation.rst.txt
new file mode 100644
index 00000000000..6ea3ef28ccd
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.LinearTransformation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+LinearTransformation
+====================
+
+.. autoclass:: LinearTransformation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.Normalize.rst.txt b/0.16/_sources/generated/torchvision.transforms.Normalize.rst.txt
new file mode 100644
index 00000000000..86745f24a38
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.Normalize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+Normalize
+=========
+
+.. autoclass:: Normalize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.PILToTensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.PILToTensor.rst.txt
new file mode 100644
index 00000000000..1fd6e7443b1
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.PILToTensor.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+PILToTensor
+===========
+
+.. autoclass:: PILToTensor
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.Pad.rst.txt b/0.16/_sources/generated/torchvision.transforms.Pad.rst.txt
new file mode 100644
index 00000000000..e8186bec58d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.Pad.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+Pad
+===
+
+.. autoclass:: Pad
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandAugment.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandAugment.rst.txt
new file mode 100644
index 00000000000..24bb6e2a5e8
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandAugment.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandAugment
+===========
+
+.. autoclass:: RandAugment
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomAdjustSharpness.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomAdjustSharpness.rst.txt
new file mode 100644
index 00000000000..f12dd541e98
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomAdjustSharpness.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomAdjustSharpness
+=====================
+
+.. autoclass:: RandomAdjustSharpness
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomAffine.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomAffine.rst.txt
new file mode 100644
index 00000000000..e818c9994db
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomAffine.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomAffine
+============
+
+.. autoclass:: RandomAffine
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomApply.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomApply.rst.txt
new file mode 100644
index 00000000000..55a607a8617
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomApply.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomApply
+===========
+
+.. autoclass:: RandomApply
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomAutocontrast.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomAutocontrast.rst.txt
new file mode 100644
index 00000000000..3a405a16af4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomAutocontrast.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomAutocontrast
+==================
+
+.. autoclass:: RandomAutocontrast
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomChoice.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomChoice.rst.txt
new file mode 100644
index 00000000000..ba0e7a3dc76
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomChoice.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomChoice
+============
+
+.. autoclass:: RandomChoice
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomCrop.rst.txt
new file mode 100644
index 00000000000..20c17fd62d0
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomCrop
+==========
+
+.. autoclass:: RandomCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomEqualize.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomEqualize.rst.txt
new file mode 100644
index 00000000000..4192d78f3a7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomEqualize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomEqualize
+==============
+
+.. autoclass:: RandomEqualize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomErasing.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomErasing.rst.txt
new file mode 100644
index 00000000000..852ab783d29
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomErasing.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomErasing
+=============
+
+.. autoclass:: RandomErasing
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomGrayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomGrayscale.rst.txt
new file mode 100644
index 00000000000..be08b08fb0a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomGrayscale.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomGrayscale
+===============
+
+.. autoclass:: RandomGrayscale
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomHorizontalFlip.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomHorizontalFlip.rst.txt
new file mode 100644
index 00000000000..6f82c01fdc6
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomHorizontalFlip.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomHorizontalFlip
+====================
+
+.. autoclass:: RandomHorizontalFlip
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomInvert.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomInvert.rst.txt
new file mode 100644
index 00000000000..447f0b4191a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomInvert.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomInvert
+============
+
+.. autoclass:: RandomInvert
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomOrder.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomOrder.rst.txt
new file mode 100644
index 00000000000..74b1804a465
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomOrder.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomOrder
+===========
+
+.. autoclass:: RandomOrder
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomPerspective.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomPerspective.rst.txt
new file mode 100644
index 00000000000..a50e8cc1c99
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomPerspective.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomPerspective
+=================
+
+.. autoclass:: RandomPerspective
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomPosterize.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomPosterize.rst.txt
new file mode 100644
index 00000000000..9889cd37141
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomPosterize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomPosterize
+===============
+
+.. autoclass:: RandomPosterize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomResizedCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomResizedCrop.rst.txt
new file mode 100644
index 00000000000..a38905a51db
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomResizedCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomResizedCrop
+=================
+
+.. autoclass:: RandomResizedCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomRotation.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomRotation.rst.txt
new file mode 100644
index 00000000000..298a295c536
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomRotation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomRotation
+==============
+
+.. autoclass:: RandomRotation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomSolarize.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomSolarize.rst.txt
new file mode 100644
index 00000000000..78aa0cd2ae9
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomSolarize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomSolarize
+==============
+
+.. autoclass:: RandomSolarize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.RandomVerticalFlip.rst.txt b/0.16/_sources/generated/torchvision.transforms.RandomVerticalFlip.rst.txt
new file mode 100644
index 00000000000..395b0fbdd01
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.RandomVerticalFlip.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+RandomVerticalFlip
+==================
+
+.. autoclass:: RandomVerticalFlip
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.Resize.rst.txt b/0.16/_sources/generated/torchvision.transforms.Resize.rst.txt
new file mode 100644
index 00000000000..76479700296
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.Resize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+Resize
+======
+
+.. autoclass:: Resize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.TenCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.TenCrop.rst.txt
new file mode 100644
index 00000000000..ba278a34d04
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.TenCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+TenCrop
+=======
+
+.. autoclass:: TenCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.ToPILImage.rst.txt b/0.16/_sources/generated/torchvision.transforms.ToPILImage.rst.txt
new file mode 100644
index 00000000000..74455aca85a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.ToPILImage.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+ToPILImage
+==========
+
+.. autoclass:: ToPILImage
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.ToTensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.ToTensor.rst.txt
new file mode 100644
index 00000000000..31f33097c59
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.ToTensor.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+ToTensor
+========
+
+.. autoclass:: ToTensor
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.TrivialAugmentWide.rst.txt b/0.16/_sources/generated/torchvision.transforms.TrivialAugmentWide.rst.txt
new file mode 100644
index 00000000000..39b8dc67021
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.TrivialAugmentWide.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms
+
+
+TrivialAugmentWide
+==================
+
+.. autoclass:: TrivialAugmentWide
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.adjust_brightness.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.adjust_brightness.rst.txt
new file mode 100644
index 00000000000..8903691f77e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.adjust_brightness.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+adjust_brightness
+=================
+
+.. autofunction:: adjust_brightness
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.adjust_contrast.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.adjust_contrast.rst.txt
new file mode 100644
index 00000000000..e2d4c0669df
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.adjust_contrast.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+adjust_contrast
+===============
+
+.. autofunction:: adjust_contrast
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.adjust_gamma.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.adjust_gamma.rst.txt
new file mode 100644
index 00000000000..f1cfcc7730e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.adjust_gamma.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+adjust_gamma
+============
+
+.. autofunction:: adjust_gamma
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.adjust_hue.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.adjust_hue.rst.txt
new file mode 100644
index 00000000000..e7bd6854cc2
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.adjust_hue.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+adjust_hue
+==========
+
+.. autofunction:: adjust_hue
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.adjust_saturation.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.adjust_saturation.rst.txt
new file mode 100644
index 00000000000..4907757bd0b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.adjust_saturation.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+adjust_saturation
+=================
+
+.. autofunction:: adjust_saturation
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.adjust_sharpness.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.adjust_sharpness.rst.txt
new file mode 100644
index 00000000000..f871a9548ae
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.adjust_sharpness.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+adjust_sharpness
+================
+
+.. autofunction:: adjust_sharpness
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.affine.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.affine.rst.txt
new file mode 100644
index 00000000000..7178303537a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.affine.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+affine
+======
+
+.. autofunction:: affine
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.autocontrast.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.autocontrast.rst.txt
new file mode 100644
index 00000000000..a221953e7ff
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.autocontrast.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+autocontrast
+============
+
+.. autofunction:: autocontrast
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.center_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.center_crop.rst.txt
new file mode 100644
index 00000000000..3e750d361c7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.center_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+center_crop
+===========
+
+.. autofunction:: center_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.convert_image_dtype.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.convert_image_dtype.rst.txt
new file mode 100644
index 00000000000..0afd20a873d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.convert_image_dtype.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+convert_image_dtype
+===================
+
+.. autofunction:: convert_image_dtype
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.crop.rst.txt
new file mode 100644
index 00000000000..c2b175f8d1a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+crop
+====
+
+.. autofunction:: crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.equalize.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.equalize.rst.txt
new file mode 100644
index 00000000000..62941edebb3
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.equalize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+equalize
+========
+
+.. autofunction:: equalize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.erase.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.erase.rst.txt
new file mode 100644
index 00000000000..784f86986eb
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.erase.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+erase
+=====
+
+.. autofunction:: erase
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.five_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.five_crop.rst.txt
new file mode 100644
index 00000000000..f3a3b89e009
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.five_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+five_crop
+=========
+
+.. autofunction:: five_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.gaussian_blur.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.gaussian_blur.rst.txt
new file mode 100644
index 00000000000..13a7d750704
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.gaussian_blur.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+gaussian_blur
+=============
+
+.. autofunction:: gaussian_blur
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.get_dimensions.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.get_dimensions.rst.txt
new file mode 100644
index 00000000000..2a684a4e8fe
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.get_dimensions.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+get_dimensions
+==============
+
+.. autofunction:: get_dimensions
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.get_image_num_channels.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.get_image_num_channels.rst.txt
new file mode 100644
index 00000000000..e5aed02d8b8
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.get_image_num_channels.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+get_image_num_channels
+======================
+
+.. autofunction:: get_image_num_channels
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.get_image_size.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.get_image_size.rst.txt
new file mode 100644
index 00000000000..5d25abb3f64
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.get_image_size.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+get_image_size
+==============
+
+.. autofunction:: get_image_size
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.hflip.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.hflip.rst.txt
new file mode 100644
index 00000000000..de3828ee5fb
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.hflip.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+hflip
+=====
+
+.. autofunction:: hflip
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.invert.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.invert.rst.txt
new file mode 100644
index 00000000000..74003f8e40d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.invert.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+invert
+======
+
+.. autofunction:: invert
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.normalize.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.normalize.rst.txt
new file mode 100644
index 00000000000..914c3deead4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.normalize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+normalize
+=========
+
+.. autofunction:: normalize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.pad.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.pad.rst.txt
new file mode 100644
index 00000000000..d61e98fc123
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.pad.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+pad
+===
+
+.. autofunction:: pad
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.perspective.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.perspective.rst.txt
new file mode 100644
index 00000000000..ae8fdb7bd09
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.perspective.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+perspective
+===========
+
+.. autofunction:: perspective
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.pil_to_tensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.pil_to_tensor.rst.txt
new file mode 100644
index 00000000000..89f4374844b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.pil_to_tensor.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+pil_to_tensor
+=============
+
+.. autofunction:: pil_to_tensor
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.posterize.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.posterize.rst.txt
new file mode 100644
index 00000000000..98c8b75b908
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.posterize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+posterize
+=========
+
+.. autofunction:: posterize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.resize.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.resize.rst.txt
new file mode 100644
index 00000000000..888e09c052e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.resize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+resize
+======
+
+.. autofunction:: resize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.resized_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.resized_crop.rst.txt
new file mode 100644
index 00000000000..cd82678b86d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.resized_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+resized_crop
+============
+
+.. autofunction:: resized_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.rgb_to_grayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.rgb_to_grayscale.rst.txt
new file mode 100644
index 00000000000..7c24bbb20bc
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.rgb_to_grayscale.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+rgb_to_grayscale
+================
+
+.. autofunction:: rgb_to_grayscale
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.rotate.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.rotate.rst.txt
new file mode 100644
index 00000000000..0aee22e5c42
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.rotate.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+rotate
+======
+
+.. autofunction:: rotate
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.solarize.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.solarize.rst.txt
new file mode 100644
index 00000000000..c539f43a3d0
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.solarize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+solarize
+========
+
+.. autofunction:: solarize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.ten_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.ten_crop.rst.txt
new file mode 100644
index 00000000000..eb698b2542a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.ten_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+ten_crop
+========
+
+.. autofunction:: ten_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.to_grayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.to_grayscale.rst.txt
new file mode 100644
index 00000000000..924598a8bb4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.to_grayscale.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+to_grayscale
+============
+
+.. autofunction:: to_grayscale
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.to_pil_image.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.to_pil_image.rst.txt
new file mode 100644
index 00000000000..8ba0eb77e86
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.to_pil_image.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+to_pil_image
+============
+
+.. autofunction:: to_pil_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.to_tensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.to_tensor.rst.txt
new file mode 100644
index 00000000000..d6243d110eb
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.to_tensor.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+to_tensor
+=========
+
+.. autofunction:: to_tensor
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.functional.vflip.rst.txt b/0.16/_sources/generated/torchvision.transforms.functional.vflip.rst.txt
new file mode 100644
index 00000000000..0f05c014884
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.functional.vflip.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.functional
+
+
+vflip
+=====
+
+.. autofunction:: vflip
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.AugMix.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.AugMix.rst.txt
new file mode 100644
index 00000000000..f31e5860972
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.AugMix.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+AugMix
+======
+
+.. autoclass:: AugMix
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.AutoAugment.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.AutoAugment.rst.txt
new file mode 100644
index 00000000000..105263c402c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.AutoAugment.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+AutoAugment
+===========
+
+.. autoclass:: AutoAugment
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.CenterCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.CenterCrop.rst.txt
new file mode 100644
index 00000000000..9ec0d32da4d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.CenterCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+CenterCrop
+==========
+
+.. autoclass:: CenterCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ClampBoundingBoxes.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ClampBoundingBoxes.rst.txt
new file mode 100644
index 00000000000..80182a7d6be
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ClampBoundingBoxes.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ClampBoundingBoxes
+==================
+
+.. autoclass:: ClampBoundingBoxes
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ColorJitter.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ColorJitter.rst.txt
new file mode 100644
index 00000000000..921bf6b2319
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ColorJitter.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ColorJitter
+===========
+
+.. autoclass:: ColorJitter
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.Compose.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.Compose.rst.txt
new file mode 100644
index 00000000000..f3004b1cb11
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.Compose.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+Compose
+=======
+
+.. autoclass:: Compose
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.rst.txt
new file mode 100644
index 00000000000..e6410e6bcd8
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ConvertBoundingBoxFormat
+========================
+
+.. autoclass:: ConvertBoundingBoxFormat
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ConvertImageDtype.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ConvertImageDtype.rst.txt
new file mode 100644
index 00000000000..8d7de33e7f4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ConvertImageDtype.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ConvertImageDtype
+=================
+
+.. autoclass:: ConvertImageDtype
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.CutMix.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.CutMix.rst.txt
new file mode 100644
index 00000000000..801a0365899
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.CutMix.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+CutMix
+======
+
+.. autoclass:: CutMix
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ElasticTransform.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ElasticTransform.rst.txt
new file mode 100644
index 00000000000..66daceda62d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ElasticTransform.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ElasticTransform
+================
+
+.. autoclass:: ElasticTransform
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.FiveCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.FiveCrop.rst.txt
new file mode 100644
index 00000000000..826655358e7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.FiveCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+FiveCrop
+========
+
+.. autoclass:: FiveCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.GaussianBlur.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.GaussianBlur.rst.txt
new file mode 100644
index 00000000000..7891a8fbe71
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.GaussianBlur.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+GaussianBlur
+============
+
+.. autoclass:: GaussianBlur
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.Grayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.Grayscale.rst.txt
new file mode 100644
index 00000000000..87f28317db2
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.Grayscale.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+Grayscale
+=========
+
+.. autoclass:: Grayscale
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.Lambda.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.Lambda.rst.txt
new file mode 100644
index 00000000000..4898edb1f5c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.Lambda.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+Lambda
+======
+
+.. autoclass:: Lambda
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.LinearTransformation.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.LinearTransformation.rst.txt
new file mode 100644
index 00000000000..b65b2c222b4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.LinearTransformation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+LinearTransformation
+====================
+
+.. autoclass:: LinearTransformation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.MixUp.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.MixUp.rst.txt
new file mode 100644
index 00000000000..d5fe24fb1d9
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.MixUp.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+MixUp
+=====
+
+.. autoclass:: MixUp
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.Normalize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.Normalize.rst.txt
new file mode 100644
index 00000000000..844f05acbd2
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.Normalize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+Normalize
+=========
+
+.. autoclass:: Normalize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.PILToTensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.PILToTensor.rst.txt
new file mode 100644
index 00000000000..5b9acb83426
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.PILToTensor.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+PILToTensor
+===========
+
+.. autoclass:: PILToTensor
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.Pad.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.Pad.rst.txt
new file mode 100644
index 00000000000..4edcc7da20e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.Pad.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+Pad
+===
+
+.. autoclass:: Pad
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandAugment.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandAugment.rst.txt
new file mode 100644
index 00000000000..1a87abe4b8a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandAugment.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandAugment
+===========
+
+.. autoclass:: RandAugment
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomAdjustSharpness.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomAdjustSharpness.rst.txt
new file mode 100644
index 00000000000..108c485cb27
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomAdjustSharpness.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomAdjustSharpness
+=====================
+
+.. autoclass:: RandomAdjustSharpness
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomAffine.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomAffine.rst.txt
new file mode 100644
index 00000000000..f180b9c5ce1
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomAffine.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomAffine
+============
+
+.. autoclass:: RandomAffine
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomApply.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomApply.rst.txt
new file mode 100644
index 00000000000..25e83819113
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomApply.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomApply
+===========
+
+.. autoclass:: RandomApply
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomAutocontrast.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomAutocontrast.rst.txt
new file mode 100644
index 00000000000..58eaa1e302a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomAutocontrast.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomAutocontrast
+==================
+
+.. autoclass:: RandomAutocontrast
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomChannelPermutation.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomChannelPermutation.rst.txt
new file mode 100644
index 00000000000..5218ae0d2b7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomChannelPermutation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomChannelPermutation
+========================
+
+.. autoclass:: RandomChannelPermutation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomChoice.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomChoice.rst.txt
new file mode 100644
index 00000000000..a936054619f
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomChoice.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomChoice
+============
+
+.. autoclass:: RandomChoice
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomCrop.rst.txt
new file mode 100644
index 00000000000..267b296c50e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomCrop
+==========
+
+.. autoclass:: RandomCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomEqualize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomEqualize.rst.txt
new file mode 100644
index 00000000000..31b9149ed53
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomEqualize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomEqualize
+==============
+
+.. autoclass:: RandomEqualize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomErasing.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomErasing.rst.txt
new file mode 100644
index 00000000000..59cf8521ba1
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomErasing.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomErasing
+=============
+
+.. autoclass:: RandomErasing
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomGrayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomGrayscale.rst.txt
new file mode 100644
index 00000000000..d971904679e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomGrayscale.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomGrayscale
+===============
+
+.. autoclass:: RandomGrayscale
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomHorizontalFlip.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomHorizontalFlip.rst.txt
new file mode 100644
index 00000000000..7368bde857a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomHorizontalFlip.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomHorizontalFlip
+====================
+
+.. autoclass:: RandomHorizontalFlip
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomInvert.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomInvert.rst.txt
new file mode 100644
index 00000000000..8c82dc136bc
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomInvert.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomInvert
+============
+
+.. autoclass:: RandomInvert
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomIoUCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomIoUCrop.rst.txt
new file mode 100644
index 00000000000..3bd8bb0c1ec
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomIoUCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomIoUCrop
+=============
+
+.. autoclass:: RandomIoUCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomOrder.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomOrder.rst.txt
new file mode 100644
index 00000000000..44af76c14ce
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomOrder.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomOrder
+===========
+
+.. autoclass:: RandomOrder
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomPerspective.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomPerspective.rst.txt
new file mode 100644
index 00000000000..b888709e030
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomPerspective.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomPerspective
+=================
+
+.. autoclass:: RandomPerspective
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomPhotometricDistort.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomPhotometricDistort.rst.txt
new file mode 100644
index 00000000000..a3dd3b96d63
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomPhotometricDistort.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomPhotometricDistort
+========================
+
+.. autoclass:: RandomPhotometricDistort
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomPosterize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomPosterize.rst.txt
new file mode 100644
index 00000000000..866987d36f3
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomPosterize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomPosterize
+===============
+
+.. autoclass:: RandomPosterize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomResize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomResize.rst.txt
new file mode 100644
index 00000000000..11c18cd2206
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomResize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomResize
+============
+
+.. autoclass:: RandomResize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomResizedCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomResizedCrop.rst.txt
new file mode 100644
index 00000000000..b1dac8987d5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomResizedCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomResizedCrop
+=================
+
+.. autoclass:: RandomResizedCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomRotation.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomRotation.rst.txt
new file mode 100644
index 00000000000..87a2f154daf
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomRotation.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomRotation
+==============
+
+.. autoclass:: RandomRotation
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomShortestSize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomShortestSize.rst.txt
new file mode 100644
index 00000000000..72b32db5c0c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomShortestSize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomShortestSize
+==================
+
+.. autoclass:: RandomShortestSize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomSolarize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomSolarize.rst.txt
new file mode 100644
index 00000000000..cb589f43822
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomSolarize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomSolarize
+==============
+
+.. autoclass:: RandomSolarize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomVerticalFlip.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomVerticalFlip.rst.txt
new file mode 100644
index 00000000000..f07764b6967
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomVerticalFlip.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomVerticalFlip
+==================
+
+.. autoclass:: RandomVerticalFlip
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.RandomZoomOut.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.RandomZoomOut.rst.txt
new file mode 100644
index 00000000000..68ed7ba4ce5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.RandomZoomOut.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+RandomZoomOut
+=============
+
+.. autoclass:: RandomZoomOut
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.Resize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.Resize.rst.txt
new file mode 100644
index 00000000000..3957bb049dd
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.Resize.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+Resize
+======
+
+.. autoclass:: Resize
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.rst.txt
new file mode 100644
index 00000000000..3c963f9b1a8
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+SanitizeBoundingBoxes
+=====================
+
+.. autoclass:: SanitizeBoundingBoxes
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ScaleJitter.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ScaleJitter.rst.txt
new file mode 100644
index 00000000000..f1536d7d24c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ScaleJitter.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ScaleJitter
+===========
+
+.. autoclass:: ScaleJitter
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.TenCrop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.TenCrop.rst.txt
new file mode 100644
index 00000000000..63bff4da050
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.TenCrop.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+TenCrop
+=======
+
+.. autoclass:: TenCrop
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ToDtype.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ToDtype.rst.txt
new file mode 100644
index 00000000000..6d3b98c6fcc
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ToDtype.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ToDtype
+=======
+
+.. autoclass:: ToDtype
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ToImage.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ToImage.rst.txt
new file mode 100644
index 00000000000..253e38f9363
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ToImage.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ToImage
+=======
+
+.. autoclass:: ToImage
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ToPILImage.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ToPILImage.rst.txt
new file mode 100644
index 00000000000..381756ba6f4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ToPILImage.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ToPILImage
+==========
+
+.. autoclass:: ToPILImage
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ToPureTensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ToPureTensor.rst.txt
new file mode 100644
index 00000000000..682df9d019d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ToPureTensor.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ToPureTensor
+============
+
+.. autoclass:: ToPureTensor
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.ToTensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.ToTensor.rst.txt
new file mode 100644
index 00000000000..02e9e879c5c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.ToTensor.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+ToTensor
+========
+
+.. autoclass:: ToTensor
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.TrivialAugmentWide.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.TrivialAugmentWide.rst.txt
new file mode 100644
index 00000000000..d50d0f65d74
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.TrivialAugmentWide.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+TrivialAugmentWide
+==================
+
+.. autoclass:: TrivialAugmentWide
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.UniformTemporalSubsample.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.UniformTemporalSubsample.rst.txt
new file mode 100644
index 00000000000..668ff14322c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.UniformTemporalSubsample.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2
+
+
+UniformTemporalSubsample
+========================
+
+.. autoclass:: UniformTemporalSubsample
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_brightness.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_brightness.rst.txt
new file mode 100644
index 00000000000..178d1dbc09b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_brightness.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+adjust_brightness
+=================
+
+.. autofunction:: adjust_brightness
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_contrast.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_contrast.rst.txt
new file mode 100644
index 00000000000..d37ffe38161
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_contrast.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+adjust_contrast
+===============
+
+.. autofunction:: adjust_contrast
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_gamma.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_gamma.rst.txt
new file mode 100644
index 00000000000..17e1d178735
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_gamma.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+adjust_gamma
+============
+
+.. autofunction:: adjust_gamma
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_hue.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_hue.rst.txt
new file mode 100644
index 00000000000..0c7c3d2a4b9
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_hue.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+adjust_hue
+==========
+
+.. autofunction:: adjust_hue
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_saturation.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_saturation.rst.txt
new file mode 100644
index 00000000000..da4742f4b27
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_saturation.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+adjust_saturation
+=================
+
+.. autofunction:: adjust_saturation
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_sharpness.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_sharpness.rst.txt
new file mode 100644
index 00000000000..7c9b533e03c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.adjust_sharpness.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+adjust_sharpness
+================
+
+.. autofunction:: adjust_sharpness
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.affine.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.affine.rst.txt
new file mode 100644
index 00000000000..7d8d2c15310
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.affine.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+affine
+======
+
+.. autofunction:: affine
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.autocontrast.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.autocontrast.rst.txt
new file mode 100644
index 00000000000..73b342d625e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.autocontrast.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+autocontrast
+============
+
+.. autofunction:: autocontrast
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.center_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.center_crop.rst.txt
new file mode 100644
index 00000000000..b0e4e43d196
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.center_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+center_crop
+===========
+
+.. autofunction:: center_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.rst.txt
new file mode 100644
index 00000000000..fd133d8c469
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+clamp_bounding_boxes
+====================
+
+.. autofunction:: clamp_bounding_boxes
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.rst.txt
new file mode 100644
index 00000000000..c030de05f4d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.rst.txt
@@ -0,0 +1,6 @@
+torchvision.transforms.v2.functional.convert\_bounding\_box\_format
+===================================================================
+
+.. currentmodule:: torchvision.transforms.v2.functional
+
+.. autofunction:: convert_bounding_box_format
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.convert_image_dtype.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.convert_image_dtype.rst.txt
new file mode 100644
index 00000000000..29c42e36bbc
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.convert_image_dtype.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+convert_image_dtype
+===================
+
+.. autoclass:: convert_image_dtype
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.crop.rst.txt
new file mode 100644
index 00000000000..c5f795a085a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+crop
+====
+
+.. autofunction:: crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.elastic.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.elastic.rst.txt
new file mode 100644
index 00000000000..456a55a6097
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.elastic.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+elastic
+=======
+
+.. autofunction:: elastic
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.equalize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.equalize.rst.txt
new file mode 100644
index 00000000000..40a39d06923
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.equalize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+equalize
+========
+
+.. autofunction:: equalize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.erase.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.erase.rst.txt
new file mode 100644
index 00000000000..59e755fb770
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.erase.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+erase
+=====
+
+.. autofunction:: erase
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.five_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.five_crop.rst.txt
new file mode 100644
index 00000000000..cde833dfa09
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.five_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+five_crop
+=========
+
+.. autofunction:: five_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.gaussian_blur.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.gaussian_blur.rst.txt
new file mode 100644
index 00000000000..29d4f979c7d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.gaussian_blur.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+gaussian_blur
+=============
+
+.. autofunction:: gaussian_blur
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.horizontal_flip.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.horizontal_flip.rst.txt
new file mode 100644
index 00000000000..1215134e7b1
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.horizontal_flip.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+horizontal_flip
+===============
+
+.. autofunction:: horizontal_flip
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.invert.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.invert.rst.txt
new file mode 100644
index 00000000000..dacf42d95ba
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.invert.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+invert
+======
+
+.. autofunction:: invert
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.normalize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.normalize.rst.txt
new file mode 100644
index 00000000000..7729e7fa400
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.normalize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+normalize
+=========
+
+.. autofunction:: normalize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.pad.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.pad.rst.txt
new file mode 100644
index 00000000000..18cf0465077
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.pad.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+pad
+===
+
+.. autofunction:: pad
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.permute_channels.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.permute_channels.rst.txt
new file mode 100644
index 00000000000..93040518a76
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.permute_channels.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+permute_channels
+================
+
+.. autofunction:: permute_channels
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.perspective.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.perspective.rst.txt
new file mode 100644
index 00000000000..412794c7fa7
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.perspective.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+perspective
+===========
+
+.. autofunction:: perspective
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.pil_to_tensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.pil_to_tensor.rst.txt
new file mode 100644
index 00000000000..d790208434e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.pil_to_tensor.rst.txt
@@ -0,0 +1,6 @@
+torchvision.transforms.v2.functional.pil\_to\_tensor
+====================================================
+
+.. currentmodule:: torchvision.transforms.v2.functional
+
+.. autofunction:: pil_to_tensor
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.posterize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.posterize.rst.txt
new file mode 100644
index 00000000000..1397c1c9580
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.posterize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+posterize
+=========
+
+.. autofunction:: posterize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.register_kernel.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.register_kernel.rst.txt
new file mode 100644
index 00000000000..44837920bfe
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.register_kernel.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+register_kernel
+===============
+
+.. autofunction:: register_kernel
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.resize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.resize.rst.txt
new file mode 100644
index 00000000000..95be11b8dc5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.resize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+resize
+======
+
+.. autofunction:: resize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.resized_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.resized_crop.rst.txt
new file mode 100644
index 00000000000..893a7f8957d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.resized_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+resized_crop
+============
+
+.. autofunction:: resized_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.rst.txt
new file mode 100644
index 00000000000..f8379282154
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+rgb_to_grayscale
+================
+
+.. autofunction:: rgb_to_grayscale
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.rotate.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.rotate.rst.txt
new file mode 100644
index 00000000000..bcd6e71617c
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.rotate.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+rotate
+======
+
+.. autofunction:: rotate
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.solarize.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.solarize.rst.txt
new file mode 100644
index 00000000000..5528bab4c05
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.solarize.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+solarize
+========
+
+.. autofunction:: solarize
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.ten_crop.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.ten_crop.rst.txt
new file mode 100644
index 00000000000..cdbfa0f48e8
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.ten_crop.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+ten_crop
+========
+
+.. autofunction:: ten_crop
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.to_dtype.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_dtype.rst.txt
new file mode 100644
index 00000000000..a3d32a99a4a
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_dtype.rst.txt
@@ -0,0 +1,6 @@
+torchvision.transforms.v2.functional.to\_dtype
+==============================================
+
+.. currentmodule:: torchvision.transforms.v2.functional
+
+.. autofunction:: to_dtype
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.to_grayscale.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_grayscale.rst.txt
new file mode 100644
index 00000000000..fe3300fcd48
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_grayscale.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+to_grayscale
+============
+
+.. autofunction:: to_grayscale
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.to_image.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_image.rst.txt
new file mode 100644
index 00000000000..d3dba56a9d4
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_image.rst.txt
@@ -0,0 +1,6 @@
+torchvision.transforms.v2.functional.to\_image
+==============================================
+
+.. currentmodule:: torchvision.transforms.v2.functional
+
+.. autofunction:: to_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.to_pil_image.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_pil_image.rst.txt
new file mode 100644
index 00000000000..5267dcb4f73
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_pil_image.rst.txt
@@ -0,0 +1,6 @@
+torchvision.transforms.v2.functional.to\_pil\_image
+===================================================
+
+.. currentmodule:: torchvision.transforms.v2.functional
+
+.. autofunction:: to_pil_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.to_tensor.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_tensor.rst.txt
new file mode 100644
index 00000000000..7142eb83039
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.to_tensor.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+to_tensor
+=========
+
+.. autoclass:: to_tensor
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.rst.txt
new file mode 100644
index 00000000000..f23bc49ed05
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+uniform_temporal_subsample
+==========================
+
+.. autofunction:: uniform_temporal_subsample
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.transforms.v2.functional.vertical_flip.rst.txt b/0.16/_sources/generated/torchvision.transforms.v2.functional.vertical_flip.rst.txt
new file mode 100644
index 00000000000..1cde4280a67
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.transforms.v2.functional.vertical_flip.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.transforms.v2.functional
+
+
+vertical_flip
+=============
+
+.. autofunction:: vertical_flip
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxFormat.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxFormat.rst.txt
new file mode 100644
index 00000000000..f2c2e7d643d
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxFormat.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+BoundingBoxFormat
+=================
+
+.. autoclass:: BoundingBoxFormat
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxes.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxes.rst.txt
new file mode 100644
index 00000000000..b9291636a6e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.BoundingBoxes.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+BoundingBoxes
+=============
+
+.. autoclass:: BoundingBoxes
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.Image.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.Image.rst.txt
new file mode 100644
index 00000000000..b9256e48843
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.Image.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+Image
+=====
+
+.. autoclass:: Image
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.Mask.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.Mask.rst.txt
new file mode 100644
index 00000000000..b84a0d1d5bb
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.Mask.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+Mask
+====
+
+.. autoclass:: Mask
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.TVTensor.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.TVTensor.rst.txt
new file mode 100644
index 00000000000..fd5d425e8d5
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.TVTensor.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+TVTensor
+========
+
+.. autoclass:: TVTensor
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.Video.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.Video.rst.txt
new file mode 100644
index 00000000000..a1dbeaab1a3
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.Video.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+Video
+=====
+
+.. autoclass:: Video
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.set_return_type.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.set_return_type.rst.txt
new file mode 100644
index 00000000000..b7a55f82cab
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.set_return_type.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+set_return_type
+===============
+
+.. autoclass:: set_return_type
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.tv_tensors.wrap.rst.txt b/0.16/_sources/generated/torchvision.tv_tensors.wrap.rst.txt
new file mode 100644
index 00000000000..846c954d160
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.tv_tensors.wrap.rst.txt
@@ -0,0 +1,10 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.tv_tensors
+
+
+wrap
+====
+
+.. autoclass:: wrap
+    :members:
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.utils.draw_bounding_boxes.rst.txt b/0.16/_sources/generated/torchvision.utils.draw_bounding_boxes.rst.txt
new file mode 100644
index 00000000000..fbd0c9e6d7e
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.utils.draw_bounding_boxes.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.utils
+
+
+draw_bounding_boxes
+===================
+
+.. autofunction:: draw_bounding_boxes
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.utils.draw_keypoints.rst.txt b/0.16/_sources/generated/torchvision.utils.draw_keypoints.rst.txt
new file mode 100644
index 00000000000..2036a9cd8ec
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.utils.draw_keypoints.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.utils
+
+
+draw_keypoints
+==============
+
+.. autofunction:: draw_keypoints
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.utils.draw_segmentation_masks.rst.txt b/0.16/_sources/generated/torchvision.utils.draw_segmentation_masks.rst.txt
new file mode 100644
index 00000000000..88b23e8989b
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.utils.draw_segmentation_masks.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.utils
+
+
+draw_segmentation_masks
+=======================
+
+.. autofunction:: draw_segmentation_masks
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.utils.flow_to_image.rst.txt b/0.16/_sources/generated/torchvision.utils.flow_to_image.rst.txt
new file mode 100644
index 00000000000..f8418cfaa49
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.utils.flow_to_image.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.utils
+
+
+flow_to_image
+=============
+
+.. autofunction:: flow_to_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.utils.make_grid.rst.txt b/0.16/_sources/generated/torchvision.utils.make_grid.rst.txt
new file mode 100644
index 00000000000..860e2b819dd
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.utils.make_grid.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.utils
+
+
+make_grid
+=========
+
+.. autofunction:: make_grid
\ No newline at end of file
diff --git a/0.16/_sources/generated/torchvision.utils.save_image.rst.txt b/0.16/_sources/generated/torchvision.utils.save_image.rst.txt
new file mode 100644
index 00000000000..0bca53e4a87
--- /dev/null
+++ b/0.16/_sources/generated/torchvision.utils.save_image.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.utils
+
+
+save_image
+==========
+
+.. autofunction:: save_image
\ No newline at end of file
diff --git a/0.16/_sources/generated/video_table.rst.txt b/0.16/_sources/generated/video_table.rst.txt
new file mode 100644
index 00000000000..278cd59f40a
--- /dev/null
+++ b/0.16/_sources/generated/video_table.rst.txt
@@ -0,0 +1,19 @@
+.. rst-class:: table-weights
+.. table::
+    :widths: 110 18 18 18 18 10 
+
+    =======================================================================  ===========  ===========  ============  ============  ==============================================================================================
+    **Weight**                                                                 **Acc@1**    **Acc@5**  **Params**      **GFLOPS**  **Recipe**
+    =======================================================================  ===========  ===========  ============  ============  ==============================================================================================
+    :class:`MC3_18_Weights.KINETICS400_V1 <MC3_18_Weights>`                       63.96        84.13   11.7M                43.34  `link <https://github.com/pytorch/vision/tree/main/references/video_classification>`__
+    :class:`MViT_V1_B_Weights.KINETICS400_V1 <MViT_V1_B_Weights>`                 78.477       93.582  36.6M                70.6   `link <https://github.com/facebookresearch/pytorchvideo/blob/main/docs/source/model_zoo.md>`__
+    :class:`MViT_V2_S_Weights.KINETICS400_V1 <MViT_V2_S_Weights>`                 80.757       94.665  34.5M                64.22  `link <https://github.com/facebookresearch/SlowFast/blob/main/MODEL_ZOO.md>`__
+    :class:`R2Plus1D_18_Weights.KINETICS400_V1 <R2Plus1D_18_Weights>`             67.463       86.175  31.5M                40.52  `link <https://github.com/pytorch/vision/tree/main/references/video_classification>`__
+    :class:`R3D_18_Weights.KINETICS400_V1 <R3D_18_Weights>`                       63.2         83.479  33.4M                40.7   `link <https://github.com/pytorch/vision/tree/main/references/video_classification>`__
+    :class:`S3D_Weights.KINETICS400_V1 <S3D_Weights>`                             68.368       88.05   8.3M                 17.98  `link <https://github.com/pytorch/vision/tree/main/references/video_classification#s3d>`__
+    :class:`Swin3D_B_Weights.KINETICS400_V1 <Swin3D_B_Weights>`                   79.427       94.386  88.0M               140.67  `link <https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400>`__
+    :class:`Swin3D_B_Weights.KINETICS400_IMAGENET22K_V1 <Swin3D_B_Weights>`       81.643       95.574  88.0M               140.67  `link <https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400>`__
+    :class:`Swin3D_S_Weights.KINETICS400_V1 <Swin3D_S_Weights>`                   79.521       94.158  49.8M                82.84  `link <https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400>`__
+    :class:`Swin3D_T_Weights.KINETICS400_V1 <Swin3D_T_Weights>`                   77.715       93.519  28.2M                43.88  `link <https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400>`__
+    =======================================================================  ===========  ===========  ============  ============  ==============================================================================================
+
diff --git a/0.16/_sources/index.rst.txt b/0.16/_sources/index.rst.txt
new file mode 100644
index 00000000000..dc5fdefaefb
--- /dev/null
+++ b/0.16/_sources/index.rst.txt
@@ -0,0 +1,69 @@
+torchvision
+===========
+This library is part of the `PyTorch
+<http://pytorch.org/>`_ project. PyTorch is an open source
+machine learning framework.
+
+Features described in this documentation are classified by release status:
+
+  *Stable:*  These features will be maintained long-term and there should generally
+  be no major performance limitations or gaps in documentation.
+  We also expect to maintain backwards compatibility (although
+  breaking changes can happen and notice will be given one release ahead
+  of time).
+
+  *Beta:*  Features are tagged as Beta because the API may change based on
+  user feedback, because the performance needs to improve, or because
+  coverage across operators is not yet complete. For Beta features, we are
+  committing to seeing the feature through to the Stable classification.
+  We are not, however, committing to backwards compatibility.
+
+  *Prototype:*  These features are typically not available as part of
+  binary distributions like PyPI or Conda, except sometimes behind run-time
+  flags, and are at an early stage for feedback and testing.
+
+
+
+The :mod:`torchvision` package consists of popular datasets, model
+architectures, and common image transformations for computer vision.
+
+.. toctree::
+   :maxdepth: 2
+   :caption: Package Reference
+
+   transforms
+   tv_tensors
+   models
+   datasets
+   utils
+   ops
+   io
+   feature_extraction
+
+.. toctree::
+   :maxdepth: 1
+   :caption: Examples and training references
+
+   auto_examples/index
+   training_references
+
+.. automodule:: torchvision
+   :members:
+
+.. toctree::
+   :maxdepth: 1
+   :caption: PyTorch Libraries
+
+   PyTorch <https://pytorch.org/docs>
+   torchaudio <https://pytorch.org/audio>
+   torchtext <https://pytorch.org/text>
+   torchvision <https://pytorch.org/vision>
+   TorchElastic <https://pytorch.org/elastic/>
+   TorchServe <https://pytorch.org/serve>
+   PyTorch on XLA Devices <http://pytorch.org/xla/>
+
+
+Indices
+-------
+
+* :ref:`genindex`
diff --git a/0.16/_sources/io.rst.txt b/0.16/_sources/io.rst.txt
new file mode 100644
index 00000000000..1da9bb6882a
--- /dev/null
+++ b/0.16/_sources/io.rst.txt
@@ -0,0 +1,89 @@
+Decoding / Encoding images and videos
+=====================================
+
+.. currentmodule:: torchvision.io
+
+The :mod:`torchvision.io` package provides functions for performing IO
+operations. They are currently specific to reading and writing images and
+videos.
+
+Images
+------
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    read_image
+    decode_image
+    encode_jpeg
+    decode_jpeg
+    write_jpeg
+    encode_png
+    decode_png
+    write_png
+    read_file
+    write_file
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    ImageReadMode
+
+
+
+Video
+-----
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    read_video
+    read_video_timestamps
+    write_video
+
+
+Fine-grained video API
+^^^^^^^^^^^^^^^^^^^^^^
+
+In addition to the :mod:`read_video` function, we provide a high-performance 
+lower-level API for more fine-grained control compared to the :mod:`read_video` function.
+It does all this whilst fully supporting torchscript.
+
+.. betastatus:: fine-grained video API
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    VideoReader
+
+
+Example of inspecting a video:
+
+.. code:: python
+
+    import torchvision
+    video_path = "path to a test video"
+    # Constructor allocates memory and a threaded decoder
+    # instance per video. At the moment it takes two arguments:
+    # path to the video file, and a wanted stream.
+    reader = torchvision.io.VideoReader(video_path, "video")
+
+    # The information about the video can be retrieved using the 
+    # `get_metadata()` method. It returns a dictionary for every stream, with
+    # duration and other relevant metadata (often frame rate)
+    reader_md = reader.get_metadata()
+
+    # metadata is structured as a dict of dicts with following structure
+    # {"stream_type": {"attribute": [attribute per stream]}}
+    #
+    # following would print out the list of frame rates for every present video stream
+    print(reader_md["video"]["fps"])
+
+    # we explicitly select the stream we would like to operate on. In
+    # the constructor we select a default video stream, but
+    # in practice, we can set whichever stream we would like 
+    video.set_current_stream("video:0")
diff --git a/0.16/_sources/models.rst.txt b/0.16/_sources/models.rst.txt
new file mode 100644
index 00000000000..15540778602
--- /dev/null
+++ b/0.16/_sources/models.rst.txt
@@ -0,0 +1,573 @@
+.. _models:
+
+Models and pre-trained weights
+##############################
+
+The ``torchvision.models`` subpackage contains definitions of models for addressing
+different tasks, including: image classification, pixelwise semantic
+segmentation, object detection, instance segmentation, person
+keypoint detection, video classification, and optical flow.
+
+General information on pre-trained weights
+==========================================
+
+TorchVision offers pre-trained weights for every provided architecture, using
+the PyTorch :mod:`torch.hub`. Instancing a pre-trained model will download its
+weights to a cache directory. This directory can be set using the `TORCH_HOME`
+environment variable. See :func:`torch.hub.load_state_dict_from_url` for details.
+
+.. note::
+
+    The pre-trained models provided in this library may have their own licenses or
+    terms and conditions derived from the dataset used for training. It is your
+    responsibility to determine whether you have permission to use the models for
+    your use case.
+
+.. note ::
+    Backward compatibility is guaranteed for loading a serialized
+    ``state_dict`` to the model created using old PyTorch version.
+    On the contrary, loading entire saved models or serialized
+    ``ScriptModules`` (serialized using older versions of PyTorch)
+    may not preserve the historic behaviour. Refer to the following
+    `documentation
+    <https://pytorch.org/docs/stable/notes/serialization.html#id6>`_
+
+
+Initializing pre-trained models
+-------------------------------
+
+As of v0.13, TorchVision offers a new `Multi-weight support API
+<https://pytorch.org/blog/introducing-torchvision-new-multi-weight-support-api/>`_
+for loading different weights to the existing model builder methods:
+
+.. code:: python
+
+    from torchvision.models import resnet50, ResNet50_Weights
+
+    # Old weights with accuracy 76.130%
+    resnet50(weights=ResNet50_Weights.IMAGENET1K_V1)
+
+    # New weights with accuracy 80.858%
+    resnet50(weights=ResNet50_Weights.IMAGENET1K_V2)
+
+    # Best available weights (currently alias for IMAGENET1K_V2)
+    # Note that these weights may change across versions
+    resnet50(weights=ResNet50_Weights.DEFAULT)
+
+    # Strings are also supported
+    resnet50(weights="IMAGENET1K_V2")
+
+    # No weights - random initialization
+    resnet50(weights=None)
+
+
+Migrating to the new API is very straightforward. The following method calls between the 2 APIs are all equivalent:
+
+.. code:: python
+
+    from torchvision.models import resnet50, ResNet50_Weights
+
+    # Using pretrained weights:
+    resnet50(weights=ResNet50_Weights.IMAGENET1K_V1)
+    resnet50(weights="IMAGENET1K_V1")
+    resnet50(pretrained=True)  # deprecated
+    resnet50(True)  # deprecated
+
+    # Using no weights:
+    resnet50(weights=None)
+    resnet50()
+    resnet50(pretrained=False)  # deprecated
+    resnet50(False)  # deprecated
+
+Note that the ``pretrained`` parameter is now deprecated, using it will emit warnings and will be removed on v0.15.
+
+Using the pre-trained models
+----------------------------
+
+Before using the pre-trained models, one must preprocess the image
+(resize with right resolution/interpolation, apply inference transforms,
+rescale the values etc). There is no standard way to do this as it depends on
+how a given model was trained. It can vary across model families, variants or
+even weight versions. Using the correct preprocessing method is critical and
+failing to do so may lead to decreased accuracy or incorrect outputs.
+
+All the necessary information for the inference transforms of each pre-trained
+model is provided on its weights documentation. To simplify inference, TorchVision
+bundles the necessary preprocessing transforms into each model weight. These are
+accessible via the ``weight.transforms`` attribute:
+
+.. code:: python
+
+    # Initialize the Weight Transforms
+    weights = ResNet50_Weights.DEFAULT
+    preprocess = weights.transforms()
+
+    # Apply it to the input image
+    img_transformed = preprocess(img)
+
+
+Some models use modules which have different training and evaluation
+behavior, such as batch normalization. To switch between these modes, use
+``model.train()`` or ``model.eval()`` as appropriate. See
+:meth:`~torch.nn.Module.train` or :meth:`~torch.nn.Module.eval` for details.
+
+.. code:: python
+
+    # Initialize model
+    weights = ResNet50_Weights.DEFAULT
+    model = resnet50(weights=weights)
+
+    # Set model to eval mode
+    model.eval()
+
+Listing and retrieving available models
+---------------------------------------
+
+As of v0.14, TorchVision offers a new mechanism which allows listing and
+retrieving models and weights by their names. Here are a few examples on how to
+use them:
+
+.. code:: python
+
+    # List available models
+    all_models = list_models()
+    classification_models = list_models(module=torchvision.models)
+
+    # Initialize models
+    m1 = get_model("mobilenet_v3_large", weights=None)
+    m2 = get_model("quantized_mobilenet_v3_large", weights="DEFAULT")
+
+    # Fetch weights
+    weights = get_weight("MobileNet_V3_Large_QuantizedWeights.DEFAULT")
+    assert weights == MobileNet_V3_Large_QuantizedWeights.DEFAULT
+
+    weights_enum = get_model_weights("quantized_mobilenet_v3_large")
+    assert weights_enum == MobileNet_V3_Large_QuantizedWeights
+
+    weights_enum2 = get_model_weights(torchvision.models.quantization.mobilenet_v3_large)
+    assert weights_enum == weights_enum2
+
+Here are the available public functions to retrieve models and their corresponding weights:
+
+.. currentmodule:: torchvision.models
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    get_model
+    get_model_weights
+    get_weight
+    list_models
+
+Using models from Hub
+---------------------
+
+Most pre-trained models can be accessed directly via PyTorch Hub without having TorchVision installed:
+
+.. code:: python
+
+    import torch
+
+    # Option 1: passing weights param as string
+    model = torch.hub.load("pytorch/vision", "resnet50", weights="IMAGENET1K_V2")
+
+    # Option 2: passing weights param as enum
+    weights = torch.hub.load("pytorch/vision", "get_weight", weights="ResNet50_Weights.IMAGENET1K_V2")
+    model = torch.hub.load("pytorch/vision", "resnet50", weights=weights)
+
+You can also retrieve all the available weights of a specific model via PyTorch Hub by doing:
+
+.. code:: python
+
+    import torch
+
+    weight_enum = torch.hub.load("pytorch/vision", "get_model_weights", name="resnet50")
+    print([weight for weight in weight_enum])
+
+The only exception to the above are the detection models included on
+:mod:`torchvision.models.detection`. These models require TorchVision
+to be installed because they depend on custom C++ operators.
+
+Classification
+==============
+
+.. currentmodule:: torchvision.models
+
+The following classification models are available, with or without pre-trained
+weights:
+
+.. toctree::
+   :maxdepth: 1
+
+   models/alexnet
+   models/convnext
+   models/densenet
+   models/efficientnet
+   models/efficientnetv2
+   models/googlenet
+   models/inception
+   models/maxvit
+   models/mnasnet
+   models/mobilenetv2
+   models/mobilenetv3
+   models/regnet
+   models/resnet
+   models/resnext
+   models/shufflenetv2
+   models/squeezenet
+   models/swin_transformer
+   models/vgg
+   models/vision_transformer
+   models/wide_resnet
+
+|
+
+Here is an example of how to use the pre-trained image classification models:
+
+.. code:: python
+
+    from torchvision.io import read_image
+    from torchvision.models import resnet50, ResNet50_Weights
+
+    img = read_image("test/assets/encode_jpeg/grace_hopper_517x606.jpg")
+
+    # Step 1: Initialize model with the best available weights
+    weights = ResNet50_Weights.DEFAULT
+    model = resnet50(weights=weights)
+    model.eval()
+
+    # Step 2: Initialize the inference transforms
+    preprocess = weights.transforms()
+
+    # Step 3: Apply inference preprocessing transforms
+    batch = preprocess(img).unsqueeze(0)
+
+    # Step 4: Use the model and print the predicted category
+    prediction = model(batch).squeeze(0).softmax(0)
+    class_id = prediction.argmax().item()
+    score = prediction[class_id].item()
+    category_name = weights.meta["categories"][class_id]
+    print(f"{category_name}: {100 * score:.1f}%")
+
+The classes of the pre-trained model outputs can be found at ``weights.meta["categories"]``.
+
+Table of all available classification weights
+---------------------------------------------
+
+Accuracies are reported on ImageNet-1K using single crops:
+
+.. include:: generated/classification_table.rst
+
+Quantized models
+----------------
+
+.. currentmodule:: torchvision.models.quantization
+
+The following architectures provide support for INT8 quantized models, with or without
+pre-trained weights:
+
+.. toctree::
+   :maxdepth: 1
+
+   models/googlenet_quant
+   models/inception_quant
+   models/mobilenetv2_quant
+   models/mobilenetv3_quant
+   models/resnet_quant
+   models/resnext_quant
+   models/shufflenetv2_quant
+
+|
+
+Here is an example of how to use the pre-trained quantized image classification models:
+
+.. code:: python
+
+    from torchvision.io import read_image
+    from torchvision.models.quantization import resnet50, ResNet50_QuantizedWeights
+
+    img = read_image("test/assets/encode_jpeg/grace_hopper_517x606.jpg")
+
+    # Step 1: Initialize model with the best available weights
+    weights = ResNet50_QuantizedWeights.DEFAULT
+    model = resnet50(weights=weights, quantize=True)
+    model.eval()
+
+    # Step 2: Initialize the inference transforms
+    preprocess = weights.transforms()
+
+    # Step 3: Apply inference preprocessing transforms
+    batch = preprocess(img).unsqueeze(0)
+
+    # Step 4: Use the model and print the predicted category
+    prediction = model(batch).squeeze(0).softmax(0)
+    class_id = prediction.argmax().item()
+    score = prediction[class_id].item()
+    category_name = weights.meta["categories"][class_id]
+    print(f"{category_name}: {100 * score}%")
+
+The classes of the pre-trained model outputs can be found at ``weights.meta["categories"]``.
+
+
+Table of all available quantized classification weights
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Accuracies are reported on ImageNet-1K using single crops:
+
+.. include:: generated/classification_quant_table.rst
+
+Semantic Segmentation
+=====================
+
+.. currentmodule:: torchvision.models.segmentation
+
+.. betastatus:: segmentation module
+
+The following semantic segmentation models are available, with or without
+pre-trained weights:
+
+.. toctree::
+   :maxdepth: 1
+
+   models/deeplabv3
+   models/fcn
+   models/lraspp
+
+|
+
+Here is an example of how to use the pre-trained semantic segmentation models:
+
+.. code:: python
+
+    from torchvision.io.image import read_image
+    from torchvision.models.segmentation import fcn_resnet50, FCN_ResNet50_Weights
+    from torchvision.transforms.functional import to_pil_image
+
+    img = read_image("gallery/assets/dog1.jpg")
+
+    # Step 1: Initialize model with the best available weights
+    weights = FCN_ResNet50_Weights.DEFAULT
+    model = fcn_resnet50(weights=weights)
+    model.eval()
+
+    # Step 2: Initialize the inference transforms
+    preprocess = weights.transforms()
+
+    # Step 3: Apply inference preprocessing transforms
+    batch = preprocess(img).unsqueeze(0)
+
+    # Step 4: Use the model and visualize the prediction
+    prediction = model(batch)["out"]
+    normalized_masks = prediction.softmax(dim=1)
+    class_to_idx = {cls: idx for (idx, cls) in enumerate(weights.meta["categories"])}
+    mask = normalized_masks[0, class_to_idx["dog"]]
+    to_pil_image(mask).show()
+
+The classes of the pre-trained model outputs can be found at ``weights.meta["categories"]``.
+The output format of the models is illustrated in :ref:`semantic_seg_output`.
+
+
+Table of all available semantic segmentation weights
+----------------------------------------------------
+
+All models are evaluated a subset of COCO val2017, on the 20 categories that are present in the Pascal VOC dataset:
+
+.. include:: generated/segmentation_table.rst
+
+
+.. _object_det_inst_seg_pers_keypoint_det:
+
+Object Detection, Instance Segmentation and Person Keypoint Detection
+=====================================================================
+
+The pre-trained models for detection, instance segmentation and
+keypoint detection are initialized with the classification models
+in torchvision. The models expect a list of ``Tensor[C, H, W]``.
+Check the constructor of the models for more information.
+
+.. betastatus:: detection module
+
+Object Detection
+----------------
+
+.. currentmodule:: torchvision.models.detection
+
+The following object detection models are available, with or without pre-trained
+weights:
+
+.. toctree::
+   :maxdepth: 1
+
+   models/faster_rcnn
+   models/fcos
+   models/retinanet
+   models/ssd
+   models/ssdlite
+
+|
+
+Here is an example of how to use the pre-trained object detection models:
+
+.. code:: python
+
+
+    from torchvision.io.image import read_image
+    from torchvision.models.detection import fasterrcnn_resnet50_fpn_v2, FasterRCNN_ResNet50_FPN_V2_Weights
+    from torchvision.utils import draw_bounding_boxes
+    from torchvision.transforms.functional import to_pil_image
+
+    img = read_image("test/assets/encode_jpeg/grace_hopper_517x606.jpg")
+
+    # Step 1: Initialize model with the best available weights
+    weights = FasterRCNN_ResNet50_FPN_V2_Weights.DEFAULT
+    model = fasterrcnn_resnet50_fpn_v2(weights=weights, box_score_thresh=0.9)
+    model.eval()
+
+    # Step 2: Initialize the inference transforms
+    preprocess = weights.transforms()
+
+    # Step 3: Apply inference preprocessing transforms
+    batch = [preprocess(img)]
+
+    # Step 4: Use the model and visualize the prediction
+    prediction = model(batch)[0]
+    labels = [weights.meta["categories"][i] for i in prediction["labels"]]
+    box = draw_bounding_boxes(img, boxes=prediction["boxes"],
+                              labels=labels,
+                              colors="red",
+                              width=4, font_size=30)
+    im = to_pil_image(box.detach())
+    im.show()
+
+The classes of the pre-trained model outputs can be found at ``weights.meta["categories"]``.
+For details on how to plot the bounding boxes of the models, you may refer to :ref:`instance_seg_output`.
+
+Table of all available Object detection weights
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Box MAPs are reported on COCO val2017:
+
+.. include:: generated/detection_table.rst
+
+
+Instance Segmentation
+---------------------
+
+.. currentmodule:: torchvision.models.detection
+
+The following instance segmentation models are available, with or without pre-trained
+weights:
+
+.. toctree::
+   :maxdepth: 1
+
+   models/mask_rcnn
+
+|
+
+
+For details on how to plot the masks of the models, you may refer to :ref:`instance_seg_output`.
+
+Table of all available Instance segmentation weights
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Box and Mask MAPs are reported on COCO val2017:
+
+.. include:: generated/instance_segmentation_table.rst
+
+Keypoint Detection
+------------------
+
+.. currentmodule:: torchvision.models.detection
+
+The following person keypoint detection models are available, with or without
+pre-trained weights:
+
+.. toctree::
+   :maxdepth: 1
+
+   models/keypoint_rcnn
+
+|
+
+The classes of the pre-trained model outputs can be found at ``weights.meta["keypoint_names"]``.
+For details on how to plot the bounding boxes of the models, you may refer to :ref:`keypoint_output`.
+
+Table of all available Keypoint detection weights
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Box and Keypoint MAPs are reported on COCO val2017:
+
+.. include:: generated/detection_keypoint_table.rst
+
+
+Video Classification
+====================
+
+.. currentmodule:: torchvision.models.video
+
+.. betastatus:: video module
+
+The following video classification models are available, with or without
+pre-trained weights:
+
+.. toctree::
+   :maxdepth: 1
+
+   models/video_mvit
+   models/video_resnet
+   models/video_s3d
+   models/video_swin_transformer
+
+|
+
+Here is an example of how to use the pre-trained video classification models:
+
+.. code:: python
+
+
+    from torchvision.io.video import read_video
+    from torchvision.models.video import r3d_18, R3D_18_Weights
+
+    vid, _, _ = read_video("test/assets/videos/v_SoccerJuggling_g23_c01.avi", output_format="TCHW")
+    vid = vid[:32]  # optionally shorten duration
+
+    # Step 1: Initialize model with the best available weights
+    weights = R3D_18_Weights.DEFAULT
+    model = r3d_18(weights=weights)
+    model.eval()
+
+    # Step 2: Initialize the inference transforms
+    preprocess = weights.transforms()
+
+    # Step 3: Apply inference preprocessing transforms
+    batch = preprocess(vid).unsqueeze(0)
+
+    # Step 4: Use the model and print the predicted category
+    prediction = model(batch).squeeze(0).softmax(0)
+    label = prediction.argmax().item()
+    score = prediction[label].item()
+    category_name = weights.meta["categories"][label]
+    print(f"{category_name}: {100 * score}%")
+
+The classes of the pre-trained model outputs can be found at ``weights.meta["categories"]``.
+
+
+Table of all available video classification weights
+---------------------------------------------------
+
+Accuracies are reported on Kinetics-400 using single crops for clip length 16:
+
+.. include:: generated/video_table.rst
+
+Optical Flow
+============
+
+.. currentmodule:: torchvision.models.optical_flow
+
+The following Optical Flow models are available, with or without pre-trained
+
+.. toctree::
+   :maxdepth: 1
+
+   models/raft
diff --git a/0.16/_sources/models/alexnet.rst.txt b/0.16/_sources/models/alexnet.rst.txt
new file mode 100644
index 00000000000..8e94b4eeed9
--- /dev/null
+++ b/0.16/_sources/models/alexnet.rst.txt
@@ -0,0 +1,28 @@
+AlexNet
+=======
+
+.. currentmodule:: torchvision.models
+
+The AlexNet model was originally introduced in the
+`ImageNet Classification with Deep Convolutional Neural Networks
+<https://papers.nips.cc/paper/2012/hash/c399862d3b9d6b76c8436e924a68c45b-Abstract.html>`__
+paper. The implemented architecture is slightly different from the original one,
+and is based on `One weird trick for parallelizing convolutional neural networks
+<https://arxiv.org/abs/1404.5997>`__.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an AlexNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.alexnet.AlexNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/alexnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    alexnet
diff --git a/0.16/_sources/models/convnext.rst.txt b/0.16/_sources/models/convnext.rst.txt
new file mode 100644
index 00000000000..f484bf63d94
--- /dev/null
+++ b/0.16/_sources/models/convnext.rst.txt
@@ -0,0 +1,26 @@
+ConvNeXt
+========
+
+.. currentmodule:: torchvision.models
+
+The ConvNeXt model is based on the `A ConvNet for the 2020s
+<https://arxiv.org/abs/2201.03545>`_ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a ConvNeXt model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.convnext.ConvNeXt`` base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py>`_ for
+more details about this class.
+
+.. autosummary::
+   :toctree: generated/
+   :template: function.rst
+
+   convnext_tiny
+   convnext_small
+   convnext_base
+   convnext_large
diff --git a/0.16/_sources/models/deeplabv3.rst.txt b/0.16/_sources/models/deeplabv3.rst.txt
new file mode 100644
index 00000000000..e6f21686081
--- /dev/null
+++ b/0.16/_sources/models/deeplabv3.rst.txt
@@ -0,0 +1,28 @@
+DeepLabV3
+=========
+
+.. currentmodule:: torchvision.models.segmentation
+
+The DeepLabV3 model is based on the `Rethinking Atrous Convolution for Semantic
+Image Segmentation <https://arxiv.org/abs/1706.05587>`__ paper.
+
+.. betastatus:: segmentation module
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a DeepLabV3 model with
+different backbones, with or without pre-trained weights. All the model builders
+internally rely on the ``torchvision.models.segmentation.deeplabv3.DeepLabV3`` base class. Please
+refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/deeplabv3.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    deeplabv3_mobilenet_v3_large
+    deeplabv3_resnet50
+    deeplabv3_resnet101
diff --git a/0.16/_sources/models/densenet.rst.txt b/0.16/_sources/models/densenet.rst.txt
new file mode 100644
index 00000000000..ee984886925
--- /dev/null
+++ b/0.16/_sources/models/densenet.rst.txt
@@ -0,0 +1,27 @@
+DenseNet
+========
+
+.. currentmodule:: torchvision.models
+
+The DenseNet model is based on the `Densely Connected Convolutional Networks
+<https://arxiv.org/abs/1608.06993>`_ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a DenseNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.densenet.DenseNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    densenet121
+    densenet161
+    densenet169
+    densenet201
diff --git a/0.16/_sources/models/efficientnet.rst.txt b/0.16/_sources/models/efficientnet.rst.txt
new file mode 100644
index 00000000000..cbc9718959a
--- /dev/null
+++ b/0.16/_sources/models/efficientnet.rst.txt
@@ -0,0 +1,31 @@
+EfficientNet
+============
+
+.. currentmodule:: torchvision.models
+
+The EfficientNet model is based on the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks <https://arxiv.org/abs/1905.11946>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an EfficientNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    efficientnet_b0
+    efficientnet_b1
+    efficientnet_b2
+    efficientnet_b3
+    efficientnet_b4
+    efficientnet_b5
+    efficientnet_b6
+    efficientnet_b7
diff --git a/0.16/_sources/models/efficientnetv2.rst.txt b/0.16/_sources/models/efficientnetv2.rst.txt
new file mode 100644
index 00000000000..3066c28ebd4
--- /dev/null
+++ b/0.16/_sources/models/efficientnetv2.rst.txt
@@ -0,0 +1,26 @@
+EfficientNetV2
+==============
+
+.. currentmodule:: torchvision.models
+
+The EfficientNetV2 model is based on the `EfficientNetV2: Smaller Models and Faster Training <https://arxiv.org/abs/2104.00298>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an EfficientNetV2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    efficientnet_v2_s
+    efficientnet_v2_m
+    efficientnet_v2_l
diff --git a/0.16/_sources/models/faster_rcnn.rst.txt b/0.16/_sources/models/faster_rcnn.rst.txt
new file mode 100644
index 00000000000..19ec9227886
--- /dev/null
+++ b/0.16/_sources/models/faster_rcnn.rst.txt
@@ -0,0 +1,31 @@
+Faster R-CNN
+============
+
+.. currentmodule:: torchvision.models.detection
+
+
+The Faster R-CNN model is based on the `Faster R-CNN: Towards Real-Time Object Detection 
+with Region Proposal Networks <https://arxiv.org/abs/1506.01497>`__
+paper.
+
+.. betastatus:: detection module
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a Faster R-CNN model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.detection.faster_rcnn.FasterRCNN`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    fasterrcnn_resnet50_fpn
+    fasterrcnn_resnet50_fpn_v2
+    fasterrcnn_mobilenet_v3_large_fpn
+    fasterrcnn_mobilenet_v3_large_320_fpn
+    
diff --git a/0.16/_sources/models/fcn.rst.txt b/0.16/_sources/models/fcn.rst.txt
new file mode 100644
index 00000000000..efcdb37c0d5
--- /dev/null
+++ b/0.16/_sources/models/fcn.rst.txt
@@ -0,0 +1,28 @@
+FCN
+===
+
+.. currentmodule:: torchvision.models.segmentation
+
+The FCN model is based on the `Fully Convolutional Networks for Semantic
+Segmentation <https://arxiv.org/abs/1411.4038>`__
+paper.
+
+.. betastatus:: segmentation module
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a FCN model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.segmentation.FCN`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    fcn_resnet50
+    fcn_resnet101
diff --git a/0.16/_sources/models/fcos.rst.txt b/0.16/_sources/models/fcos.rst.txt
new file mode 100644
index 00000000000..085f26549b8
--- /dev/null
+++ b/0.16/_sources/models/fcos.rst.txt
@@ -0,0 +1,24 @@
+FCOS
+=========
+
+.. currentmodule:: torchvision.models.detection
+
+The FCOS model is based on the `FCOS: Fully Convolutional One-Stage Object Detection
+<https://arxiv.org/abs/1904.01355>`__ paper.
+
+.. betastatus:: detection module
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a FCOS model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.detection.fcos.FCOS`` base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/fcos.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    fcos_resnet50_fpn
diff --git a/0.16/_sources/models/generated/torchvision.models.alexnet.rst.txt b/0.16/_sources/models/generated/torchvision.models.alexnet.rst.txt
new file mode 100644
index 00000000000..38d65ffb3b6
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.alexnet.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+alexnet
+=======
+
+.. autofunction:: alexnet
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.convnext_base.rst.txt b/0.16/_sources/models/generated/torchvision.models.convnext_base.rst.txt
new file mode 100644
index 00000000000..c31395b5948
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.convnext_base.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+convnext_base
+=============
+
+.. autofunction:: convnext_base
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.convnext_large.rst.txt b/0.16/_sources/models/generated/torchvision.models.convnext_large.rst.txt
new file mode 100644
index 00000000000..22d0097d7e1
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.convnext_large.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+convnext_large
+==============
+
+.. autofunction:: convnext_large
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.convnext_small.rst.txt b/0.16/_sources/models/generated/torchvision.models.convnext_small.rst.txt
new file mode 100644
index 00000000000..e10a8a6b4cb
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.convnext_small.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+convnext_small
+==============
+
+.. autofunction:: convnext_small
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.convnext_tiny.rst.txt b/0.16/_sources/models/generated/torchvision.models.convnext_tiny.rst.txt
new file mode 100644
index 00000000000..7ce1dd5b642
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.convnext_tiny.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+convnext_tiny
+=============
+
+.. autofunction:: convnext_tiny
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.densenet121.rst.txt b/0.16/_sources/models/generated/torchvision.models.densenet121.rst.txt
new file mode 100644
index 00000000000..f3fb17e30d2
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.densenet121.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+densenet121
+===========
+
+.. autofunction:: densenet121
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.densenet161.rst.txt b/0.16/_sources/models/generated/torchvision.models.densenet161.rst.txt
new file mode 100644
index 00000000000..dabe7817b80
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.densenet161.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+densenet161
+===========
+
+.. autofunction:: densenet161
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.densenet169.rst.txt b/0.16/_sources/models/generated/torchvision.models.densenet169.rst.txt
new file mode 100644
index 00000000000..f0127f9fec0
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.densenet169.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+densenet169
+===========
+
+.. autofunction:: densenet169
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.densenet201.rst.txt b/0.16/_sources/models/generated/torchvision.models.densenet201.rst.txt
new file mode 100644
index 00000000000..a9c7ae05fd8
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.densenet201.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+densenet201
+===========
+
+.. autofunction:: densenet201
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.rst.txt
new file mode 100644
index 00000000000..3770ba46352
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+fasterrcnn_mobilenet_v3_large_320_fpn
+=====================================
+
+.. autofunction:: fasterrcnn_mobilenet_v3_large_320_fpn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.rst.txt
new file mode 100644
index 00000000000..a5dfc1c1de7
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+fasterrcnn_mobilenet_v3_large_fpn
+=================================
+
+.. autofunction:: fasterrcnn_mobilenet_v3_large_fpn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.rst.txt
new file mode 100644
index 00000000000..65911055c1d
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+fasterrcnn_resnet50_fpn
+=======================
+
+.. autofunction:: fasterrcnn_resnet50_fpn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.rst.txt
new file mode 100644
index 00000000000..b1709b512d7
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+fasterrcnn_resnet50_fpn_v2
+==========================
+
+.. autofunction:: fasterrcnn_resnet50_fpn_v2
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.fcos_resnet50_fpn.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.fcos_resnet50_fpn.rst.txt
new file mode 100644
index 00000000000..15812328406
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.fcos_resnet50_fpn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+fcos_resnet50_fpn
+=================
+
+.. autofunction:: fcos_resnet50_fpn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.rst.txt
new file mode 100644
index 00000000000..c6de497eb99
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+keypointrcnn_resnet50_fpn
+=========================
+
+.. autofunction:: keypointrcnn_resnet50_fpn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.rst.txt
new file mode 100644
index 00000000000..5baaa70eadf
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+maskrcnn_resnet50_fpn
+=====================
+
+.. autofunction:: maskrcnn_resnet50_fpn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.rst.txt
new file mode 100644
index 00000000000..45c800e29bd
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+maskrcnn_resnet50_fpn_v2
+========================
+
+.. autofunction:: maskrcnn_resnet50_fpn_v2
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.rst.txt
new file mode 100644
index 00000000000..a598067604c
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+retinanet_resnet50_fpn
+======================
+
+.. autofunction:: retinanet_resnet50_fpn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.rst.txt
new file mode 100644
index 00000000000..3959eef84c0
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+retinanet_resnet50_fpn_v2
+=========================
+
+.. autofunction:: retinanet_resnet50_fpn_v2
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.ssd300_vgg16.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.ssd300_vgg16.rst.txt
new file mode 100644
index 00000000000..f208e69a8dc
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.ssd300_vgg16.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+ssd300_vgg16
+============
+
+.. autofunction:: ssd300_vgg16
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.rst.txt b/0.16/_sources/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.rst.txt
new file mode 100644
index 00000000000..d14e64c23ff
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.detection
+
+
+ssdlite320_mobilenet_v3_large
+=============================
+
+.. autofunction:: ssdlite320_mobilenet_v3_large
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b0.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b0.rst.txt
new file mode 100644
index 00000000000..27000dcfab8
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b0.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b0
+===============
+
+.. autofunction:: efficientnet_b0
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b1.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b1.rst.txt
new file mode 100644
index 00000000000..6426238f543
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b1.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b1
+===============
+
+.. autofunction:: efficientnet_b1
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b2.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b2.rst.txt
new file mode 100644
index 00000000000..a76513883df
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b2
+===============
+
+.. autofunction:: efficientnet_b2
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b3.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b3.rst.txt
new file mode 100644
index 00000000000..a9965f0ed49
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b3.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b3
+===============
+
+.. autofunction:: efficientnet_b3
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b4.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b4.rst.txt
new file mode 100644
index 00000000000..278d4efe28f
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b4.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b4
+===============
+
+.. autofunction:: efficientnet_b4
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b5.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b5.rst.txt
new file mode 100644
index 00000000000..652e6dd631c
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b5.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b5
+===============
+
+.. autofunction:: efficientnet_b5
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b6.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b6.rst.txt
new file mode 100644
index 00000000000..9d16f66606b
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b6.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b6
+===============
+
+.. autofunction:: efficientnet_b6
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_b7.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_b7.rst.txt
new file mode 100644
index 00000000000..4b1cfa9538b
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_b7.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_b7
+===============
+
+.. autofunction:: efficientnet_b7
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_l.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_l.rst.txt
new file mode 100644
index 00000000000..d88b56aedcf
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_l.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_v2_l
+=================
+
+.. autofunction:: efficientnet_v2_l
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_m.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_m.rst.txt
new file mode 100644
index 00000000000..a6ab65dc4e5
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_m.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_v2_m
+=================
+
+.. autofunction:: efficientnet_v2_m
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_s.rst.txt b/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_s.rst.txt
new file mode 100644
index 00000000000..c318a7c1d27
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.efficientnet_v2_s.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+efficientnet_v2_s
+=================
+
+.. autofunction:: efficientnet_v2_s
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.googlenet.rst.txt b/0.16/_sources/models/generated/torchvision.models.googlenet.rst.txt
new file mode 100644
index 00000000000..b8e4c6208c9
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.googlenet.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+googlenet
+=========
+
+.. autofunction:: googlenet
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.inception_v3.rst.txt b/0.16/_sources/models/generated/torchvision.models.inception_v3.rst.txt
new file mode 100644
index 00000000000..ca248178e52
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.inception_v3.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+inception_v3
+============
+
+.. autofunction:: inception_v3
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.maxvit_t.rst.txt b/0.16/_sources/models/generated/torchvision.models.maxvit_t.rst.txt
new file mode 100644
index 00000000000..3814979d49f
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.maxvit_t.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+maxvit_t
+========
+
+.. autofunction:: maxvit_t
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.mnasnet0_5.rst.txt b/0.16/_sources/models/generated/torchvision.models.mnasnet0_5.rst.txt
new file mode 100644
index 00000000000..343da5684ef
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.mnasnet0_5.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+mnasnet0_5
+==========
+
+.. autofunction:: mnasnet0_5
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.mnasnet0_75.rst.txt b/0.16/_sources/models/generated/torchvision.models.mnasnet0_75.rst.txt
new file mode 100644
index 00000000000..46a1c831393
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.mnasnet0_75.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+mnasnet0_75
+===========
+
+.. autofunction:: mnasnet0_75
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.mnasnet1_0.rst.txt b/0.16/_sources/models/generated/torchvision.models.mnasnet1_0.rst.txt
new file mode 100644
index 00000000000..9eb84550504
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.mnasnet1_0.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+mnasnet1_0
+==========
+
+.. autofunction:: mnasnet1_0
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.mnasnet1_3.rst.txt b/0.16/_sources/models/generated/torchvision.models.mnasnet1_3.rst.txt
new file mode 100644
index 00000000000..b40031677f7
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.mnasnet1_3.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+mnasnet1_3
+==========
+
+.. autofunction:: mnasnet1_3
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.mobilenet_v2.rst.txt b/0.16/_sources/models/generated/torchvision.models.mobilenet_v2.rst.txt
new file mode 100644
index 00000000000..b5445b3a43b
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.mobilenet_v2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+mobilenet_v2
+============
+
+.. autofunction:: mobilenet_v2
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.mobilenet_v3_large.rst.txt b/0.16/_sources/models/generated/torchvision.models.mobilenet_v3_large.rst.txt
new file mode 100644
index 00000000000..316223c4e39
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.mobilenet_v3_large.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+mobilenet_v3_large
+==================
+
+.. autofunction:: mobilenet_v3_large
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.mobilenet_v3_small.rst.txt b/0.16/_sources/models/generated/torchvision.models.mobilenet_v3_small.rst.txt
new file mode 100644
index 00000000000..027af5d83b9
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.mobilenet_v3_small.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+mobilenet_v3_small
+==================
+
+.. autofunction:: mobilenet_v3_small
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.optical_flow.raft_large.rst.txt b/0.16/_sources/models/generated/torchvision.models.optical_flow.raft_large.rst.txt
new file mode 100644
index 00000000000..92092c84df9
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.optical_flow.raft_large.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.optical_flow
+
+
+raft_large
+==========
+
+.. autofunction:: raft_large
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.optical_flow.raft_small.rst.txt b/0.16/_sources/models/generated/torchvision.models.optical_flow.raft_small.rst.txt
new file mode 100644
index 00000000000..6d954b05b1c
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.optical_flow.raft_small.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.optical_flow
+
+
+raft_small
+==========
+
+.. autofunction:: raft_small
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.googlenet.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.googlenet.rst.txt
new file mode 100644
index 00000000000..648744036db
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.googlenet.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+googlenet
+=========
+
+.. autofunction:: googlenet
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.inception_v3.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.inception_v3.rst.txt
new file mode 100644
index 00000000000..ca120392db4
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.inception_v3.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+inception_v3
+============
+
+.. autofunction:: inception_v3
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v2.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v2.rst.txt
new file mode 100644
index 00000000000..3c24964b1dc
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+mobilenet_v2
+============
+
+.. autofunction:: mobilenet_v2
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v3_large.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v3_large.rst.txt
new file mode 100644
index 00000000000..388fb3704c9
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.mobilenet_v3_large.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+mobilenet_v3_large
+==================
+
+.. autofunction:: mobilenet_v3_large
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.resnet18.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.resnet18.rst.txt
new file mode 100644
index 00000000000..8ad903d5269
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.resnet18.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+resnet18
+========
+
+.. autofunction:: resnet18
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.resnet50.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.resnet50.rst.txt
new file mode 100644
index 00000000000..049a95575be
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.resnet50.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+resnet50
+========
+
+.. autofunction:: resnet50
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.resnext101_32x8d.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.resnext101_32x8d.rst.txt
new file mode 100644
index 00000000000..811186f8795
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.resnext101_32x8d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+resnext101_32x8d
+================
+
+.. autofunction:: resnext101_32x8d
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.resnext101_64x4d.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.resnext101_64x4d.rst.txt
new file mode 100644
index 00000000000..d76e967bddc
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.resnext101_64x4d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+resnext101_64x4d
+================
+
+.. autofunction:: resnext101_64x4d
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.rst.txt
new file mode 100644
index 00000000000..99a4912bb2a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+shufflenet_v2_x0_5
+==================
+
+.. autofunction:: shufflenet_v2_x0_5
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.rst.txt
new file mode 100644
index 00000000000..216bb10abc6
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+shufflenet_v2_x1_0
+==================
+
+.. autofunction:: shufflenet_v2_x1_0
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.rst.txt
new file mode 100644
index 00000000000..20348c4b84f
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+shufflenet_v2_x1_5
+==================
+
+.. autofunction:: shufflenet_v2_x1_5
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.rst.txt b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.rst.txt
new file mode 100644
index 00000000000..507736dc42a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.quantization
+
+
+shufflenet_v2_x2_0
+==================
+
+.. autofunction:: shufflenet_v2_x2_0
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_x_16gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_x_16gf.rst.txt
new file mode 100644
index 00000000000..c0ab16b2d49
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_x_16gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_x_16gf
+=============
+
+.. autofunction:: regnet_x_16gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_x_1_6gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_x_1_6gf.rst.txt
new file mode 100644
index 00000000000..83f05fbd581
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_x_1_6gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_x_1_6gf
+==============
+
+.. autofunction:: regnet_x_1_6gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_x_32gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_x_32gf.rst.txt
new file mode 100644
index 00000000000..ac3859f8ece
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_x_32gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_x_32gf
+=============
+
+.. autofunction:: regnet_x_32gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_x_3_2gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_x_3_2gf.rst.txt
new file mode 100644
index 00000000000..121a267fb38
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_x_3_2gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_x_3_2gf
+==============
+
+.. autofunction:: regnet_x_3_2gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_x_400mf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_x_400mf.rst.txt
new file mode 100644
index 00000000000..1001e6fb76a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_x_400mf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_x_400mf
+==============
+
+.. autofunction:: regnet_x_400mf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_x_800mf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_x_800mf.rst.txt
new file mode 100644
index 00000000000..1e60e0057bd
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_x_800mf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_x_800mf
+==============
+
+.. autofunction:: regnet_x_800mf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_x_8gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_x_8gf.rst.txt
new file mode 100644
index 00000000000..43c695ffb97
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_x_8gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_x_8gf
+============
+
+.. autofunction:: regnet_x_8gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_128gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_128gf.rst.txt
new file mode 100644
index 00000000000..cb3d1988f6c
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_128gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_128gf
+==============
+
+.. autofunction:: regnet_y_128gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_16gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_16gf.rst.txt
new file mode 100644
index 00000000000..1ece5971c17
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_16gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_16gf
+=============
+
+.. autofunction:: regnet_y_16gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_1_6gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_1_6gf.rst.txt
new file mode 100644
index 00000000000..b01ee950e64
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_1_6gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_1_6gf
+==============
+
+.. autofunction:: regnet_y_1_6gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_32gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_32gf.rst.txt
new file mode 100644
index 00000000000..0c5fb9090d8
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_32gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_32gf
+=============
+
+.. autofunction:: regnet_y_32gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_3_2gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_3_2gf.rst.txt
new file mode 100644
index 00000000000..3d8cb219338
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_3_2gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_3_2gf
+==============
+
+.. autofunction:: regnet_y_3_2gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_400mf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_400mf.rst.txt
new file mode 100644
index 00000000000..931f81e55d7
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_400mf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_400mf
+==============
+
+.. autofunction:: regnet_y_400mf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_800mf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_800mf.rst.txt
new file mode 100644
index 00000000000..229cae2b03a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_800mf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_800mf
+==============
+
+.. autofunction:: regnet_y_800mf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.regnet_y_8gf.rst.txt b/0.16/_sources/models/generated/torchvision.models.regnet_y_8gf.rst.txt
new file mode 100644
index 00000000000..061761a5eb9
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.regnet_y_8gf.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+regnet_y_8gf
+============
+
+.. autofunction:: regnet_y_8gf
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnet101.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnet101.rst.txt
new file mode 100644
index 00000000000..097b249aa29
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnet101.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnet101
+=========
+
+.. autofunction:: resnet101
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnet152.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnet152.rst.txt
new file mode 100644
index 00000000000..ebc4bdb0a5e
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnet152.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnet152
+=========
+
+.. autofunction:: resnet152
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnet18.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnet18.rst.txt
new file mode 100644
index 00000000000..c9f1f0b0b5a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnet18.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnet18
+========
+
+.. autofunction:: resnet18
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnet34.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnet34.rst.txt
new file mode 100644
index 00000000000..cf7167f1ffc
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnet34.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnet34
+========
+
+.. autofunction:: resnet34
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnet50.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnet50.rst.txt
new file mode 100644
index 00000000000..0a9739f77ee
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnet50.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnet50
+========
+
+.. autofunction:: resnet50
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnext101_32x8d.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnext101_32x8d.rst.txt
new file mode 100644
index 00000000000..968862e0547
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnext101_32x8d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnext101_32x8d
+================
+
+.. autofunction:: resnext101_32x8d
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnext101_64x4d.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnext101_64x4d.rst.txt
new file mode 100644
index 00000000000..a4d60263654
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnext101_64x4d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnext101_64x4d
+================
+
+.. autofunction:: resnext101_64x4d
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.resnext50_32x4d.rst.txt b/0.16/_sources/models/generated/torchvision.models.resnext50_32x4d.rst.txt
new file mode 100644
index 00000000000..c637928a0b7
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.resnext50_32x4d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+resnext50_32x4d
+===============
+
+.. autofunction:: resnext50_32x4d
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.rst.txt b/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.rst.txt
new file mode 100644
index 00000000000..44818ebe870
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.segmentation
+
+
+deeplabv3_mobilenet_v3_large
+============================
+
+.. autofunction:: deeplabv3_mobilenet_v3_large
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.rst.txt b/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.rst.txt
new file mode 100644
index 00000000000..ee97f2f7a90
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.segmentation
+
+
+deeplabv3_resnet101
+===================
+
+.. autofunction:: deeplabv3_resnet101
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.rst.txt b/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.rst.txt
new file mode 100644
index 00000000000..ca3699ab17d
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.segmentation
+
+
+deeplabv3_resnet50
+==================
+
+.. autofunction:: deeplabv3_resnet50
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet101.rst.txt b/0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet101.rst.txt
new file mode 100644
index 00000000000..8984659d1a8
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet101.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.segmentation
+
+
+fcn_resnet101
+=============
+
+.. autofunction:: fcn_resnet101
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet50.rst.txt b/0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet50.rst.txt
new file mode 100644
index 00000000000..bb8cec4d64a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.segmentation.fcn_resnet50.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.segmentation
+
+
+fcn_resnet50
+============
+
+.. autofunction:: fcn_resnet50
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.rst.txt b/0.16/_sources/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.rst.txt
new file mode 100644
index 00000000000..b417770be13
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.segmentation
+
+
+lraspp_mobilenet_v3_large
+=========================
+
+.. autofunction:: lraspp_mobilenet_v3_large
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x0_5.rst.txt b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x0_5.rst.txt
new file mode 100644
index 00000000000..c2faa156e20
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x0_5.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+shufflenet_v2_x0_5
+==================
+
+.. autofunction:: shufflenet_v2_x0_5
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_0.rst.txt b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_0.rst.txt
new file mode 100644
index 00000000000..5f986a18924
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_0.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+shufflenet_v2_x1_0
+==================
+
+.. autofunction:: shufflenet_v2_x1_0
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_5.rst.txt b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_5.rst.txt
new file mode 100644
index 00000000000..ce1ada46397
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x1_5.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+shufflenet_v2_x1_5
+==================
+
+.. autofunction:: shufflenet_v2_x1_5
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x2_0.rst.txt b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x2_0.rst.txt
new file mode 100644
index 00000000000..dc107f77060
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.shufflenet_v2_x2_0.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+shufflenet_v2_x2_0
+==================
+
+.. autofunction:: shufflenet_v2_x2_0
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.squeezenet1_0.rst.txt b/0.16/_sources/models/generated/torchvision.models.squeezenet1_0.rst.txt
new file mode 100644
index 00000000000..76b501a90c7
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.squeezenet1_0.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+squeezenet1_0
+=============
+
+.. autofunction:: squeezenet1_0
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.squeezenet1_1.rst.txt b/0.16/_sources/models/generated/torchvision.models.squeezenet1_1.rst.txt
new file mode 100644
index 00000000000..c5cddfa3a70
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.squeezenet1_1.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+squeezenet1_1
+=============
+
+.. autofunction:: squeezenet1_1
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.swin_b.rst.txt b/0.16/_sources/models/generated/torchvision.models.swin_b.rst.txt
new file mode 100644
index 00000000000..7940dd0925b
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.swin_b.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+swin_b
+======
+
+.. autofunction:: swin_b
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.swin_s.rst.txt b/0.16/_sources/models/generated/torchvision.models.swin_s.rst.txt
new file mode 100644
index 00000000000..ebc375b8d95
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.swin_s.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+swin_s
+======
+
+.. autofunction:: swin_s
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.swin_t.rst.txt b/0.16/_sources/models/generated/torchvision.models.swin_t.rst.txt
new file mode 100644
index 00000000000..b95b7974a9a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.swin_t.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+swin_t
+======
+
+.. autofunction:: swin_t
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.swin_v2_b.rst.txt b/0.16/_sources/models/generated/torchvision.models.swin_v2_b.rst.txt
new file mode 100644
index 00000000000..5b402181af9
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.swin_v2_b.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+swin_v2_b
+=========
+
+.. autofunction:: swin_v2_b
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.swin_v2_s.rst.txt b/0.16/_sources/models/generated/torchvision.models.swin_v2_s.rst.txt
new file mode 100644
index 00000000000..76fbcbcd2fa
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.swin_v2_s.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+swin_v2_s
+=========
+
+.. autofunction:: swin_v2_s
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.swin_v2_t.rst.txt b/0.16/_sources/models/generated/torchvision.models.swin_v2_t.rst.txt
new file mode 100644
index 00000000000..118a9555284
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.swin_v2_t.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+swin_v2_t
+=========
+
+.. autofunction:: swin_v2_t
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg11.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg11.rst.txt
new file mode 100644
index 00000000000..38da98359fb
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg11.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg11
+=====
+
+.. autofunction:: vgg11
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg11_bn.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg11_bn.rst.txt
new file mode 100644
index 00000000000..dd36b77ec0c
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg11_bn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg11_bn
+========
+
+.. autofunction:: vgg11_bn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg13.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg13.rst.txt
new file mode 100644
index 00000000000..c653bf648be
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg13.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg13
+=====
+
+.. autofunction:: vgg13
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg13_bn.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg13_bn.rst.txt
new file mode 100644
index 00000000000..5b76febcb83
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg13_bn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg13_bn
+========
+
+.. autofunction:: vgg13_bn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg16.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg16.rst.txt
new file mode 100644
index 00000000000..24c071d7efb
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg16.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg16
+=====
+
+.. autofunction:: vgg16
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg16_bn.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg16_bn.rst.txt
new file mode 100644
index 00000000000..691e03da655
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg16_bn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg16_bn
+========
+
+.. autofunction:: vgg16_bn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg19.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg19.rst.txt
new file mode 100644
index 00000000000..4d12112e9eb
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg19.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg19
+=====
+
+.. autofunction:: vgg19
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vgg19_bn.rst.txt b/0.16/_sources/models/generated/torchvision.models.vgg19_bn.rst.txt
new file mode 100644
index 00000000000..f2d4e1300df
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vgg19_bn.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vgg19_bn
+========
+
+.. autofunction:: vgg19_bn
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.mc3_18.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.mc3_18.rst.txt
new file mode 100644
index 00000000000..1588aca5110
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.mc3_18.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+mc3_18
+======
+
+.. autofunction:: mc3_18
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.mvit_v1_b.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.mvit_v1_b.rst.txt
new file mode 100644
index 00000000000..23001050df2
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.mvit_v1_b.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+mvit_v1_b
+=========
+
+.. autofunction:: mvit_v1_b
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.mvit_v2_s.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.mvit_v2_s.rst.txt
new file mode 100644
index 00000000000..13a4b2caaf9
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.mvit_v2_s.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+mvit_v2_s
+=========
+
+.. autofunction:: mvit_v2_s
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.r2plus1d_18.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.r2plus1d_18.rst.txt
new file mode 100644
index 00000000000..dcd1490a4a6
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.r2plus1d_18.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+r2plus1d_18
+===========
+
+.. autofunction:: r2plus1d_18
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.r3d_18.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.r3d_18.rst.txt
new file mode 100644
index 00000000000..96a46712e0a
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.r3d_18.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+r3d_18
+======
+
+.. autofunction:: r3d_18
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.s3d.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.s3d.rst.txt
new file mode 100644
index 00000000000..47f41b2fb44
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.s3d.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+s3d
+===
+
+.. autofunction:: s3d
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.swin3d_b.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.swin3d_b.rst.txt
new file mode 100644
index 00000000000..c4c6aecd00d
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.swin3d_b.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+swin3d_b
+========
+
+.. autofunction:: swin3d_b
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.swin3d_s.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.swin3d_s.rst.txt
new file mode 100644
index 00000000000..fa8aeb497fc
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.swin3d_s.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+swin3d_s
+========
+
+.. autofunction:: swin3d_s
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.video.swin3d_t.rst.txt b/0.16/_sources/models/generated/torchvision.models.video.swin3d_t.rst.txt
new file mode 100644
index 00000000000..9bbdeb42d61
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.video.swin3d_t.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models.video
+
+
+swin3d_t
+========
+
+.. autofunction:: swin3d_t
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vit_b_16.rst.txt b/0.16/_sources/models/generated/torchvision.models.vit_b_16.rst.txt
new file mode 100644
index 00000000000..b172d0eb98c
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vit_b_16.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vit_b_16
+========
+
+.. autofunction:: vit_b_16
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vit_b_32.rst.txt b/0.16/_sources/models/generated/torchvision.models.vit_b_32.rst.txt
new file mode 100644
index 00000000000..339cb37a717
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vit_b_32.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vit_b_32
+========
+
+.. autofunction:: vit_b_32
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vit_h_14.rst.txt b/0.16/_sources/models/generated/torchvision.models.vit_h_14.rst.txt
new file mode 100644
index 00000000000..4fcffa8a720
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vit_h_14.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vit_h_14
+========
+
+.. autofunction:: vit_h_14
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vit_l_16.rst.txt b/0.16/_sources/models/generated/torchvision.models.vit_l_16.rst.txt
new file mode 100644
index 00000000000..02afd820cc4
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vit_l_16.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vit_l_16
+========
+
+.. autofunction:: vit_l_16
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.vit_l_32.rst.txt b/0.16/_sources/models/generated/torchvision.models.vit_l_32.rst.txt
new file mode 100644
index 00000000000..c5b1607756e
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.vit_l_32.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+vit_l_32
+========
+
+.. autofunction:: vit_l_32
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.wide_resnet101_2.rst.txt b/0.16/_sources/models/generated/torchvision.models.wide_resnet101_2.rst.txt
new file mode 100644
index 00000000000..05a8347b966
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.wide_resnet101_2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+wide_resnet101_2
+================
+
+.. autofunction:: wide_resnet101_2
\ No newline at end of file
diff --git a/0.16/_sources/models/generated/torchvision.models.wide_resnet50_2.rst.txt b/0.16/_sources/models/generated/torchvision.models.wide_resnet50_2.rst.txt
new file mode 100644
index 00000000000..7de5666ff97
--- /dev/null
+++ b/0.16/_sources/models/generated/torchvision.models.wide_resnet50_2.rst.txt
@@ -0,0 +1,9 @@
+.. role:: hidden
+    :class: hidden-section
+.. currentmodule:: torchvision.models
+
+
+wide_resnet50_2
+===============
+
+.. autofunction:: wide_resnet50_2
\ No newline at end of file
diff --git a/0.16/_sources/models/googlenet.rst.txt b/0.16/_sources/models/googlenet.rst.txt
new file mode 100644
index 00000000000..91ea03ddf3d
--- /dev/null
+++ b/0.16/_sources/models/googlenet.rst.txt
@@ -0,0 +1,24 @@
+GoogLeNet
+=========
+
+.. currentmodule:: torchvision.models
+
+The GoogleNet model is based on the `Going Deeper with Convolutions <https://arxiv.org/abs/1409.4842>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a GoogLeNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.googlenet.GoogLeNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/googlenet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    googlenet
diff --git a/0.16/_sources/models/googlenet_quant.rst.txt b/0.16/_sources/models/googlenet_quant.rst.txt
new file mode 100644
index 00000000000..4358389b3e5
--- /dev/null
+++ b/0.16/_sources/models/googlenet_quant.rst.txt
@@ -0,0 +1,24 @@
+Quantized GoogLeNet
+===================
+
+.. currentmodule:: torchvision.models.quantization
+
+The Quantized GoogleNet model is based on the `Going Deeper with Convolutions <https://arxiv.org/abs/1409.4842>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a quantized GoogLeNet
+model, with or without pre-trained weights. All the model builders internally
+rely on the ``torchvision.models.quantization.googlenet.QuantizableGoogLeNet``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/googlenet.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    googlenet
diff --git a/0.16/_sources/models/inception.rst.txt b/0.16/_sources/models/inception.rst.txt
new file mode 100644
index 00000000000..e162eef5d30
--- /dev/null
+++ b/0.16/_sources/models/inception.rst.txt
@@ -0,0 +1,23 @@
+Inception V3
+============
+
+.. currentmodule:: torchvision.models
+
+The InceptionV3 model is based on the `Rethinking the Inception Architecture for
+Computer Vision <https://arxiv.org/abs/1512.00567>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an InceptionV3 model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.inception.Inception3`` base class. Please refer to the `source
+code <https://github.com/pytorch/vision/blob/main/torchvision/models/inception.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    inception_v3
diff --git a/0.16/_sources/models/inception_quant.rst.txt b/0.16/_sources/models/inception_quant.rst.txt
new file mode 100644
index 00000000000..d26f1ab09da
--- /dev/null
+++ b/0.16/_sources/models/inception_quant.rst.txt
@@ -0,0 +1,24 @@
+Quantized InceptionV3
+=====================
+
+.. currentmodule:: torchvision.models.quantization
+
+The Quantized Inception model is based on the `Rethinking the Inception Architecture for
+Computer Vision <https://arxiv.org/abs/1512.00567>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a quantized Inception
+model, with or without pre-trained weights. All the model builders internally
+rely on the ``torchvision.models.quantization.inception.QuantizableInception3``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/inception.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    inception_v3
diff --git a/0.16/_sources/models/keypoint_rcnn.rst.txt b/0.16/_sources/models/keypoint_rcnn.rst.txt
new file mode 100644
index 00000000000..ba677c7f8f3
--- /dev/null
+++ b/0.16/_sources/models/keypoint_rcnn.rst.txt
@@ -0,0 +1,26 @@
+Keypoint R-CNN
+==============
+
+.. currentmodule:: torchvision.models.detection
+
+The Keypoint R-CNN model is based on the `Mask R-CNN
+<https://arxiv.org/abs/1703.06870>`__ paper.
+
+.. betastatus:: detection module
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a Keypoint R-CNN model,
+with or without pre-trained weights. All the model builders internally rely on
+the ``torchvision.models.detection.KeypointRCNN`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/keypoint_rcnn.py>`__
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    keypointrcnn_resnet50_fpn
diff --git a/0.16/_sources/models/lraspp.rst.txt b/0.16/_sources/models/lraspp.rst.txt
new file mode 100644
index 00000000000..312249c53e1
--- /dev/null
+++ b/0.16/_sources/models/lraspp.rst.txt
@@ -0,0 +1,24 @@
+LRASPP
+======
+
+.. currentmodule:: torchvision.models.segmentation
+
+The LRASPP model is based on the `Searching for MobileNetV3 <https://arxiv.org/abs/1905.02244>`_ paper.
+
+.. betastatus:: segmentation module
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a FCN model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.segmentation.LRASPP`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/lraspp.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    lraspp_mobilenet_v3_large
diff --git a/0.16/_sources/models/mask_rcnn.rst.txt b/0.16/_sources/models/mask_rcnn.rst.txt
new file mode 100644
index 00000000000..5887b6c71a6
--- /dev/null
+++ b/0.16/_sources/models/mask_rcnn.rst.txt
@@ -0,0 +1,27 @@
+Mask R-CNN
+==========
+
+.. currentmodule:: torchvision.models.detection
+
+The Mask R-CNN model is based on the `Mask R-CNN <https://arxiv.org/abs/1703.06870>`__
+paper.
+
+.. betastatus:: detection module
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a Mask R-CNN model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.detection.mask_rcnn.MaskRCNN`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    maskrcnn_resnet50_fpn
+    maskrcnn_resnet50_fpn_v2
diff --git a/0.16/_sources/models/maxvit.rst.txt b/0.16/_sources/models/maxvit.rst.txt
new file mode 100644
index 00000000000..29aaaaab334
--- /dev/null
+++ b/0.16/_sources/models/maxvit.rst.txt
@@ -0,0 +1,23 @@
+MaxVit
+===============
+
+.. currentmodule:: torchvision.models
+
+The MaxVit transformer models are based on the `MaxViT: Multi-Axis Vision Transformer <https://arxiv.org/abs/2204.01697>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an MaxVit model with and without pre-trained weights.
+All the model builders internally rely on the ``torchvision.models.maxvit.MaxVit`` 
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/maxvit.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    maxvit_t
diff --git a/0.16/_sources/models/mnasnet.rst.txt b/0.16/_sources/models/mnasnet.rst.txt
new file mode 100644
index 00000000000..fd9ea511585
--- /dev/null
+++ b/0.16/_sources/models/mnasnet.rst.txt
@@ -0,0 +1,28 @@
+MNASNet
+=======
+
+.. currentmodule:: torchvision.models
+
+
+The MNASNet model is based on the `MnasNet: Platform-Aware Neural Architecture
+Search for Mobile <https://arxiv.org/pdf/1807.11626.pdf>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an MNASNet model.
+All the model builders internally rely on the
+``torchvision.models.mnasnet.MNASNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    mnasnet0_5
+    mnasnet0_75
+    mnasnet1_0
+    mnasnet1_3
diff --git a/0.16/_sources/models/mobilenetv2.rst.txt b/0.16/_sources/models/mobilenetv2.rst.txt
new file mode 100644
index 00000000000..666dcce57ce
--- /dev/null
+++ b/0.16/_sources/models/mobilenetv2.rst.txt
@@ -0,0 +1,24 @@
+MobileNet V2
+============
+
+.. currentmodule:: torchvision.models
+
+The MobileNet V2 model is based on the `MobileNetV2: Inverted Residuals and Linear
+Bottlenecks <https://arxiv.org/abs/1801.04381>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a MobileNetV2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.mobilenetv2.MobileNetV2`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv2.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    mobilenet_v2
diff --git a/0.16/_sources/models/mobilenetv2_quant.rst.txt b/0.16/_sources/models/mobilenetv2_quant.rst.txt
new file mode 100644
index 00000000000..e5397378fab
--- /dev/null
+++ b/0.16/_sources/models/mobilenetv2_quant.rst.txt
@@ -0,0 +1,24 @@
+Quantized MobileNet V2
+======================
+
+.. currentmodule:: torchvision.models.quantization
+
+The Quantized MobileNet V2 model is based on the `MobileNetV2: Inverted Residuals and Linear
+Bottlenecks <https://arxiv.org/abs/1801.04381>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a quantized MobileNetV2
+model, with or without pre-trained weights. All the model builders internally
+rely on the ``torchvision.models.quantization.mobilenetv2.QuantizableMobileNetV2``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv2.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    mobilenet_v2
diff --git a/0.16/_sources/models/mobilenetv3.rst.txt b/0.16/_sources/models/mobilenetv3.rst.txt
new file mode 100644
index 00000000000..4322470286d
--- /dev/null
+++ b/0.16/_sources/models/mobilenetv3.rst.txt
@@ -0,0 +1,24 @@
+MobileNet V3
+============
+
+.. currentmodule:: torchvision.models
+
+The MobileNet V3 model is based on the `Searching for MobileNetV3 <https://arxiv.org/abs/1905.02244>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a MobileNetV3 model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.mobilenetv3.MobileNetV3`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    mobilenet_v3_large
+    mobilenet_v3_small
diff --git a/0.16/_sources/models/mobilenetv3_quant.rst.txt b/0.16/_sources/models/mobilenetv3_quant.rst.txt
new file mode 100644
index 00000000000..fe385b493e5
--- /dev/null
+++ b/0.16/_sources/models/mobilenetv3_quant.rst.txt
@@ -0,0 +1,23 @@
+Quantized MobileNet V3
+======================
+
+.. currentmodule:: torchvision.models.quantization
+
+The Quantized MobileNet V3 model is based on the `Searching for MobileNetV3 <https://arxiv.org/abs/1905.02244>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a quantized MobileNetV3
+model, with or without pre-trained weights. All the model builders internally
+rely on the ``torchvision.models.quantization.mobilenetv3.QuantizableMobileNetV3``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv3.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    mobilenet_v3_large
diff --git a/0.16/_sources/models/raft.rst.txt b/0.16/_sources/models/raft.rst.txt
new file mode 100644
index 00000000000..7ea477698b4
--- /dev/null
+++ b/0.16/_sources/models/raft.rst.txt
@@ -0,0 +1,25 @@
+RAFT
+====
+
+.. currentmodule:: torchvision.models.optical_flow
+
+The RAFT model is based on the `RAFT: Recurrent All-Pairs Field Transforms for
+Optical Flow <https://arxiv.org/abs/2003.12039>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a RAFT model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.optical_flow.RAFT`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/optical_flow/raft.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    raft_large
+    raft_small
diff --git a/0.16/_sources/models/regnet.rst.txt b/0.16/_sources/models/regnet.rst.txt
new file mode 100644
index 00000000000..aef4abd2544
--- /dev/null
+++ b/0.16/_sources/models/regnet.rst.txt
@@ -0,0 +1,37 @@
+RegNet
+======
+
+.. currentmodule:: torchvision.models
+
+The RegNet model is based on the `Designing Network Design Spaces
+<https://arxiv.org/abs/2003.13678>`_ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a RegNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.regnet.RegNet`` base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+   :toctree: generated/
+   :template: function.rst
+
+   regnet_y_400mf
+   regnet_y_800mf
+   regnet_y_1_6gf
+   regnet_y_3_2gf
+   regnet_y_8gf
+   regnet_y_16gf
+   regnet_y_32gf
+   regnet_y_128gf
+   regnet_x_400mf
+   regnet_x_800mf
+   regnet_x_1_6gf
+   regnet_x_3_2gf
+   regnet_x_8gf
+   regnet_x_16gf
+   regnet_x_32gf
diff --git a/0.16/_sources/models/resnet.rst.txt b/0.16/_sources/models/resnet.rst.txt
new file mode 100644
index 00000000000..9d777f2f6b1
--- /dev/null
+++ b/0.16/_sources/models/resnet.rst.txt
@@ -0,0 +1,33 @@
+ResNet
+======
+
+.. currentmodule:: torchvision.models
+
+The ResNet model is based on the `Deep Residual Learning for Image Recognition
+<https://arxiv.org/abs/1512.03385>`_ paper.
+
+.. note::
+    The bottleneck of TorchVision places the stride for downsampling to the second 3x3
+    convolution while the original paper places it to the first 1x1 convolution.
+    This variant improves the accuracy and is known as `ResNet V1.5
+    <https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch>`_.
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a ResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.resnet.ResNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    resnet18
+    resnet34
+    resnet50
+    resnet101
+    resnet152
diff --git a/0.16/_sources/models/resnet_quant.rst.txt b/0.16/_sources/models/resnet_quant.rst.txt
new file mode 100644
index 00000000000..5609990646c
--- /dev/null
+++ b/0.16/_sources/models/resnet_quant.rst.txt
@@ -0,0 +1,25 @@
+Quantized ResNet
+================
+
+.. currentmodule:: torchvision.models.quantization
+
+The Quantized ResNet model is based on the `Deep Residual Learning for Image Recognition
+<https://arxiv.org/abs/1512.03385>`_ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a quantized ResNet
+model, with or without pre-trained weights. All the model builders internally
+rely on the ``torchvision.models.quantization.resnet.QuantizableResNet``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    resnet18
+    resnet50
diff --git a/0.16/_sources/models/resnext.rst.txt b/0.16/_sources/models/resnext.rst.txt
new file mode 100644
index 00000000000..5d8325d9b4b
--- /dev/null
+++ b/0.16/_sources/models/resnext.rst.txt
@@ -0,0 +1,26 @@
+ResNeXt
+=======
+
+.. currentmodule:: torchvision.models
+
+The ResNext model is based on the `Aggregated Residual Transformations for Deep Neural Networks <https://arxiv.org/abs/1611.05431v2>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a ResNext model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.resnet.ResNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    resnext50_32x4d
+    resnext101_32x8d
+    resnext101_64x4d
diff --git a/0.16/_sources/models/resnext_quant.rst.txt b/0.16/_sources/models/resnext_quant.rst.txt
new file mode 100644
index 00000000000..916b9e4a39a
--- /dev/null
+++ b/0.16/_sources/models/resnext_quant.rst.txt
@@ -0,0 +1,25 @@
+Quantized ResNeXt
+=================
+
+.. currentmodule:: torchvision.models.quantization
+
+The quantized ResNext model is based on the `Aggregated Residual Transformations for Deep Neural Networks <https://arxiv.org/abs/1611.05431v2>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a quantized ResNeXt
+model, with or without pre-trained weights. All the model builders internally
+rely on the ``torchvision.models.quantization.resnet.QuantizableResNet``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    resnext101_32x8d
+    resnext101_64x4d
diff --git a/0.16/_sources/models/retinanet.rst.txt b/0.16/_sources/models/retinanet.rst.txt
new file mode 100644
index 00000000000..910692ef3a5
--- /dev/null
+++ b/0.16/_sources/models/retinanet.rst.txt
@@ -0,0 +1,25 @@
+RetinaNet
+=========
+
+.. currentmodule:: torchvision.models.detection
+
+The RetinaNet model is based on the `Focal Loss for Dense Object Detection
+<https://arxiv.org/abs/1708.02002>`__ paper.
+
+.. betastatus:: detection module
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a RetinaNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.detection.retinanet.RetinaNet`` base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/retinanet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    retinanet_resnet50_fpn
+    retinanet_resnet50_fpn_v2
diff --git a/0.16/_sources/models/shufflenetv2.rst.txt b/0.16/_sources/models/shufflenetv2.rst.txt
new file mode 100644
index 00000000000..2cbe328ca8b
--- /dev/null
+++ b/0.16/_sources/models/shufflenetv2.rst.txt
@@ -0,0 +1,27 @@
+ShuffleNet V2
+=============
+
+.. currentmodule:: torchvision.models
+
+The ShuffleNet V2 model is based on the `ShuffleNet V2: Practical Guidelines for Efficient
+CNN Architecture Design <https://arxiv.org/abs/1807.11164>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a ShuffleNetV2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.shufflenetv2.ShuffleNetV2`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    shufflenet_v2_x0_5
+    shufflenet_v2_x1_0
+    shufflenet_v2_x1_5
+    shufflenet_v2_x2_0
diff --git a/0.16/_sources/models/shufflenetv2_quant.rst.txt b/0.16/_sources/models/shufflenetv2_quant.rst.txt
new file mode 100644
index 00000000000..4fa236d2565
--- /dev/null
+++ b/0.16/_sources/models/shufflenetv2_quant.rst.txt
@@ -0,0 +1,27 @@
+Quantized ShuffleNet V2
+=======================
+
+.. currentmodule:: torchvision.models.quantization
+
+The Quantized ShuffleNet V2 model is based on the `ShuffleNet V2: Practical Guidelines for Efficient
+CNN Architecture Design <https://arxiv.org/abs/1807.11164>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a quantized ShuffleNetV2
+model, with or without pre-trained weights. All the model builders internally rely
+on the ``torchvision.models.quantization.shufflenetv2.QuantizableShuffleNetV2``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py>`_
+for more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    shufflenet_v2_x0_5
+    shufflenet_v2_x1_0
+    shufflenet_v2_x1_5
+    shufflenet_v2_x2_0
diff --git a/0.16/_sources/models/squeezenet.rst.txt b/0.16/_sources/models/squeezenet.rst.txt
new file mode 100644
index 00000000000..9771e5c623a
--- /dev/null
+++ b/0.16/_sources/models/squeezenet.rst.txt
@@ -0,0 +1,26 @@
+SqueezeNet
+==========
+
+.. currentmodule:: torchvision.models
+
+The SqueezeNet model is based on the `SqueezeNet: AlexNet-level accuracy with
+50x fewer parameters and <0.5MB model size <https://arxiv.org/abs/1602.07360>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a SqueezeNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.squeezenet.SqueezeNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/squeezenet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    squeezenet1_0
+    squeezenet1_1
diff --git a/0.16/_sources/models/ssd.rst.txt b/0.16/_sources/models/ssd.rst.txt
new file mode 100644
index 00000000000..68b0bb224df
--- /dev/null
+++ b/0.16/_sources/models/ssd.rst.txt
@@ -0,0 +1,26 @@
+SSD
+===
+
+.. currentmodule:: torchvision.models.detection
+
+The SSD model is based on the `SSD: Single Shot MultiBox Detector
+<https://arxiv.org/abs/1512.02325>`__ paper.
+
+.. betastatus:: detection module
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a SSD model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.detection.SSD`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    ssd300_vgg16
diff --git a/0.16/_sources/models/ssdlite.rst.txt b/0.16/_sources/models/ssdlite.rst.txt
new file mode 100644
index 00000000000..bac1575c966
--- /dev/null
+++ b/0.16/_sources/models/ssdlite.rst.txt
@@ -0,0 +1,27 @@
+SSDlite
+=======
+
+.. currentmodule:: torchvision.models.detection
+
+The SSDLite model is based on the `SSD: Single Shot MultiBox Detector
+<https://arxiv.org/abs/1512.02325>`__, `Searching for MobileNetV3
+<https://arxiv.org/abs/1905.02244>`__ and `MobileNetV2: Inverted Residuals and Linear
+Bottlenecks <https://arxiv.org/abs/1801.04381>__` papers.
+
+.. betastatus:: detection module
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a SSD Lite model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.detection.ssd.SSD`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    ssdlite320_mobilenet_v3_large
diff --git a/0.16/_sources/models/swin_transformer.rst.txt b/0.16/_sources/models/swin_transformer.rst.txt
new file mode 100644
index 00000000000..b302f5bd79d
--- /dev/null
+++ b/0.16/_sources/models/swin_transformer.rst.txt
@@ -0,0 +1,32 @@
+SwinTransformer
+===============
+
+.. currentmodule:: torchvision.models
+
+The SwinTransformer models are based on the `Swin Transformer: Hierarchical Vision
+Transformer using Shifted Windows <https://arxiv.org/abs/2103.14030>`__
+paper.
+SwinTransformer V2 models are based on the `Swin Transformer V2: Scaling Up Capacity
+and Resolution <https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Swin_Transformer_V2_Scaling_Up_Capacity_and_Resolution_CVPR_2022_paper.pdf>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an SwinTransformer model (original and V2) with and without pre-trained weights.
+All the model builders internally rely on the ``torchvision.models.swin_transformer.SwinTransformer``
+base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    swin_t
+    swin_s
+    swin_b
+    swin_v2_t
+    swin_v2_s
+    swin_v2_b
diff --git a/0.16/_sources/models/vgg.rst.txt b/0.16/_sources/models/vgg.rst.txt
new file mode 100644
index 00000000000..77b5686927c
--- /dev/null
+++ b/0.16/_sources/models/vgg.rst.txt
@@ -0,0 +1,30 @@
+VGG
+===
+
+.. currentmodule:: torchvision.models
+
+The VGG model is based on the `Very Deep Convolutional Networks for Large-Scale
+Image Recognition <https://arxiv.org/abs/1409.1556>`_ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a VGG model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.vgg.VGG`` base class. Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    vgg11
+    vgg11_bn
+    vgg13
+    vgg13_bn
+    vgg16
+    vgg16_bn
+    vgg19
+    vgg19_bn
diff --git a/0.16/_sources/models/video_mvit.rst.txt b/0.16/_sources/models/video_mvit.rst.txt
new file mode 100644
index 00000000000..cd23754b7bb
--- /dev/null
+++ b/0.16/_sources/models/video_mvit.rst.txt
@@ -0,0 +1,27 @@
+Video MViT
+==========
+
+.. currentmodule:: torchvision.models.video
+
+The MViT model is based on the
+`MViTv2: Improved Multiscale Vision Transformers for Classification and Detection
+<https://arxiv.org/abs/2112.01526>`__ and `Multiscale Vision Transformers
+<https://arxiv.org/abs/2104.11227>`__ papers.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a MViT v1 or v2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.video.MViT`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/video/mvit.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    mvit_v1_b
+    mvit_v2_s
diff --git a/0.16/_sources/models/video_resnet.rst.txt b/0.16/_sources/models/video_resnet.rst.txt
new file mode 100644
index 00000000000..ecb707b4eeb
--- /dev/null
+++ b/0.16/_sources/models/video_resnet.rst.txt
@@ -0,0 +1,28 @@
+Video ResNet
+============
+
+.. currentmodule:: torchvision.models.video
+
+The VideoResNet model is based on the `A Closer Look at Spatiotemporal
+Convolutions for Action Recognition <https://arxiv.org/abs/1711.11248>`__ paper.
+
+.. betastatus:: video module
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a VideoResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.video.resnet.VideoResNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    r3d_18
+    mc3_18
+    r2plus1d_18
diff --git a/0.16/_sources/models/video_s3d.rst.txt b/0.16/_sources/models/video_s3d.rst.txt
new file mode 100644
index 00000000000..0d66c55487c
--- /dev/null
+++ b/0.16/_sources/models/video_s3d.rst.txt
@@ -0,0 +1,25 @@
+Video S3D
+=========
+
+.. currentmodule:: torchvision.models.video
+
+The S3D model is based on the
+`Rethinking Spatiotemporal Feature Learning: Speed-Accuracy Trade-offs in Video Classification
+<https://arxiv.org/abs/1712.04851>`__ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate an S3D model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.video.S3D`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/video/s3d.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    s3d
diff --git a/0.16/_sources/models/video_swin_transformer.rst.txt b/0.16/_sources/models/video_swin_transformer.rst.txt
new file mode 100644
index 00000000000..e31e69759b4
--- /dev/null
+++ b/0.16/_sources/models/video_swin_transformer.rst.txt
@@ -0,0 +1,27 @@
+Video SwinTransformer
+=====================
+
+.. currentmodule:: torchvision.models.video
+
+The Video SwinTransformer model is based on the `Video Swin Transformer <https://arxiv.org/abs/2106.13230>`__ paper.
+
+.. betastatus:: video module
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a VideoResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.video.swin_transformer.SwinTransformer3d`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    swin3d_t
+    swin3d_s
+    swin3d_b
diff --git a/0.16/_sources/models/vision_transformer.rst.txt b/0.16/_sources/models/vision_transformer.rst.txt
new file mode 100644
index 00000000000..914caa9311e
--- /dev/null
+++ b/0.16/_sources/models/vision_transformer.rst.txt
@@ -0,0 +1,28 @@
+VisionTransformer
+=================
+
+.. currentmodule:: torchvision.models
+
+The VisionTransformer model is based on the `An Image is Worth 16x16 Words:
+Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>`_ paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a VisionTransformer model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.vision_transformer.VisionTransformer`` base class.
+Please refer to the `source code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py>`_ for
+more details about this class.
+
+.. autosummary::
+   :toctree: generated/
+   :template: function.rst
+
+   vit_b_16
+   vit_b_32
+   vit_l_16
+   vit_l_32
+   vit_h_14
diff --git a/0.16/_sources/models/wide_resnet.rst.txt b/0.16/_sources/models/wide_resnet.rst.txt
new file mode 100644
index 00000000000..9768355c77e
--- /dev/null
+++ b/0.16/_sources/models/wide_resnet.rst.txt
@@ -0,0 +1,25 @@
+Wide ResNet
+===========
+
+.. currentmodule:: torchvision.models
+
+The Wide ResNet model is based on the `Wide Residual Networks <https://arxiv.org/abs/1605.07146>`__
+paper.
+
+
+Model builders
+--------------
+
+The following model builders can be used to instantiate a Wide ResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+``torchvision.models.resnet.ResNet`` base class. Please refer to the `source
+code
+<https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py>`_ for
+more details about this class.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    wide_resnet50_2
+    wide_resnet101_2
diff --git a/0.16/_sources/ops.rst.txt b/0.16/_sources/ops.rst.txt
new file mode 100644
index 00000000000..7124c85bb79
--- /dev/null
+++ b/0.16/_sources/ops.rst.txt
@@ -0,0 +1,103 @@
+.. _ops:
+
+Operators
+=========
+
+.. currentmodule:: torchvision.ops
+
+:mod:`torchvision.ops` implements operators, losses and layers that are specific for Computer Vision.
+
+.. note::
+  All operators have native support for TorchScript.
+
+
+Detection and Segmentation Operators
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The below operators perform pre-processing as well as post-processing required in object detection and segmentation models.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    batched_nms
+    masks_to_boxes
+    nms
+    roi_align
+    roi_pool
+    ps_roi_align
+    ps_roi_pool
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    FeaturePyramidNetwork
+    MultiScaleRoIAlign
+    RoIAlign
+    RoIPool
+    PSRoIAlign
+    PSRoIPool
+
+
+Box Operators
+~~~~~~~~~~~~~
+
+These utility functions perform various operations on bounding boxes.
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    box_area
+    box_convert
+    box_iou
+    clip_boxes_to_image
+    complete_box_iou
+    distance_box_iou
+    generalized_box_iou
+    remove_small_boxes
+
+Losses
+~~~~~~
+
+The following vision-specific loss functions are implemented:
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    complete_box_iou_loss
+    distance_box_iou_loss
+    generalized_box_iou_loss
+    sigmoid_focal_loss
+
+
+Layers
+~~~~~~
+
+TorchVision provides commonly used building blocks as layers:
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    Conv2dNormActivation
+    Conv3dNormActivation
+    DeformConv2d
+    DropBlock2d
+    DropBlock3d
+    FrozenBatchNorm2d
+    MLP
+    Permute
+    SqueezeExcitation
+    StochasticDepth
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    deform_conv2d
+    drop_block2d
+    drop_block3d
+    stochastic_depth
diff --git a/0.16/_sources/training_references.rst.txt b/0.16/_sources/training_references.rst.txt
new file mode 100644
index 00000000000..fc22ac5eba6
--- /dev/null
+++ b/0.16/_sources/training_references.rst.txt
@@ -0,0 +1,29 @@
+Training references
+===================
+
+On top of the many models, datasets, and image transforms, Torchvision also
+provides training reference scripts. These are the scripts that we use to train
+the :ref:`models <models>` which are then available with pre-trained weights.
+
+These scripts are not part of the core package and are instead available `on
+GitHub <https://github.com/pytorch/vision/tree/main/references>`_. We currently
+provide references for
+`classification <https://github.com/pytorch/vision/tree/main/references/classification>`_,
+`detection <https://github.com/pytorch/vision/tree/main/references/detection>`_,
+`segmentation <https://github.com/pytorch/vision/tree/main/references/segmentation>`_,
+`similarity learning <https://github.com/pytorch/vision/tree/main/references/similarity>`_,
+and `video classification <https://github.com/pytorch/vision/tree/main/references/video_classification>`_.
+
+While these scripts are largely stable, they do not offer backward compatibility
+guarantees.
+
+In general, these scripts rely on the latest (not yet released) pytorch version
+or the latest torchvision version. This means that to use them, **you might need
+to install the latest pytorch and torchvision versions**, with e.g.::
+
+    conda install pytorch torchvision -c pytorch-nightly
+
+If you need to rely on an older stable version of pytorch or torchvision, e.g.
+torchvision 0.10, then it's safer to use the scripts from that corresponding
+release on GitHub, namely
+https://github.com/pytorch/vision/tree/v0.10.0/references.
diff --git a/0.16/_sources/transforms.rst.txt b/0.16/_sources/transforms.rst.txt
new file mode 100644
index 00000000000..2aa1fc5ba1e
--- /dev/null
+++ b/0.16/_sources/transforms.rst.txt
@@ -0,0 +1,665 @@
+.. _transforms:
+
+Transforming and augmenting images
+==================================
+
+.. currentmodule:: torchvision.transforms
+
+Torchvision supports common computer vision transformations in the
+``torchvision.transforms`` and ``torchvision.transforms.v2`` modules. Transforms
+can be used to transform or augment data for training or inference of different
+tasks (image classification, detection, segmentation, video classification).
+
+.. code:: python
+
+    # Image Classification
+    import torch
+    from torchvision.transforms import v2
+
+    H, W = 32, 32
+    img = torch.randint(0, 256, size=(3, H, W), dtype=torch.uint8)
+
+    transforms = v2.Compose([
+        v2.RandomResizedCrop(size=(224, 224), antialias=True),
+        v2.RandomHorizontalFlip(p=0.5),
+        v2.ToDtype(torch.float32, scale=True),
+        v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+    ])
+    img = transforms(img)
+
+.. code:: python
+
+    # Detection (re-using imports and transforms from above)
+    from torchvision import tv_tensors
+
+    img = torch.randint(0, 256, size=(3, H, W), dtype=torch.uint8)
+    boxes = torch.randint(0, H // 2, size=(3, 4))
+    boxes[:, 2:] += boxes[:, :2]
+    boxes = tv_tensors.BoundingBoxes(boxes, format="XYXY", canvas_size=(H, W))
+
+    # The same transforms can be used!
+    img, boxes = transforms(img, boxes)
+    # And you can pass arbitrary input structures
+    output_dict = transforms({"image": img, "boxes": boxes})
+
+Transforms are typically passed as the ``transform`` or ``transforms`` argument
+to the :ref:`Datasets <datasets>`.
+
+Start here
+----------
+
+Whether you're new to Torchvision transforms, or you're already experienced with
+them, we encourage you to start with
+:ref:`sphx_glr_auto_examples_transforms_plot_transforms_getting_started.py` in
+order to learn more about what can be done with the new v2 transforms.
+
+Then, browse the sections in below this page for general information and
+performance tips. The available transforms and functionals are listed in the
+:ref:`API reference <v2_api_ref>`.
+
+More information and tutorials can also be found in our :ref:`example gallery
+<gallery>`, e.g. :ref:`sphx_glr_auto_examples_transforms_plot_transforms_e2e.py`
+or :ref:`sphx_glr_auto_examples_transforms_plot_custom_transforms.py`.
+
+.. _conventions:
+
+Supported input types and conventions
+-------------------------------------
+
+Most transformations accept both `PIL <https://pillow.readthedocs.io>`_ images
+and tensor inputs. Both CPU and CUDA tensors are supported.
+The result of both backends (PIL or Tensors) should be very
+close. In general, we recommend relying on the tensor backend :ref:`for
+performance <transforms_perf>`.  The :ref:`conversion transforms
+<conversion_transforms>` may be used to convert to and from PIL images, or for
+converting dtypes and ranges.
+
+Tensor image are expected to be of shape ``(C, H, W)``, where ``C`` is the
+number of channels, and ``H`` and ``W`` refer to height and width. Most
+transforms support batched tensor input. A batch of Tensor images is a tensor of
+shape ``(N, C, H, W)``, where ``N`` is a number of images in the batch. The
+:ref:`v2 <v1_or_v2>` transforms generally accept an arbitrary number of leading
+dimensions ``(..., C, H, W)`` and can handle batched images or batched videos.
+
+.. _range_and_dtype:
+
+Dtype and expected value range
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The expected range of the values of a tensor image is implicitly defined by
+the tensor dtype. Tensor images with a float dtype are expected to have
+values in ``[0, 1]``. Tensor images with an integer dtype are expected to
+have values in ``[0, MAX_DTYPE]`` where ``MAX_DTYPE`` is the largest value
+that can be represented in that dtype. Typically, images of dtype
+``torch.uint8`` are expected to have values in ``[0, 255]``.
+
+Use :class:`~torchvision.transforms.v2.ToDtype` to convert both the dtype and
+range of the inputs.
+
+.. _v1_or_v2:
+
+V1 or V2? Which one should I use?
+---------------------------------
+
+**TL;DR** We recommending using the ``torchvision.transforms.v2`` transforms
+instead of those in ``torchvision.transforms``. They're faster and they can do
+more things. Just change the import and you should be good to go.
+
+In Torchvision 0.15 (March 2023), we released a new set of transforms available
+in the ``torchvision.transforms.v2`` namespace. These transforms have a lot of
+advantages compared to the v1 ones (in ``torchvision.transforms``):
+
+- They can transform images **but also** bounding boxes, masks, or videos. This
+  provides support for tasks beyond image classification: detection, segmentation,
+  video classification, etc. See
+  :ref:`sphx_glr_auto_examples_transforms_plot_transforms_getting_started.py`
+  and :ref:`sphx_glr_auto_examples_transforms_plot_transforms_e2e.py`.
+- They support more transforms like :class:`~torchvision.transforms.v2.CutMix`
+  and :class:`~torchvision.transforms.v2.MixUp`. See
+  :ref:`sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py`.
+- They're :ref:`faster <transforms_perf>`.
+- They support arbitrary input structures (dicts, lists, tuples, etc.).
+- Future improvements and features will be added to the v2 transforms only.
+
+These transforms are **fully backward compatible** with the v1 ones, so if
+you're already using tranforms from ``torchvision.transforms``, all you need to
+do to is to update the import to ``torchvision.transforms.v2``. In terms of
+output, there might be negligible differences due to implementation differences.
+
+.. note::
+
+    The v2 transforms are still BETA, but at this point we do not expect
+    disruptive changes to be made to their public APIs. We're planning to make
+    them fully stable in version 0.17. Please submit any feedback you may have
+    `here <https://github.com/pytorch/vision/issues/6753>`_.
+
+.. _transforms_perf:
+
+Performance considerations
+--------------------------
+
+We recommend the following guidelines to get the best performance out of the
+transforms:
+
+- Rely on the v2 transforms from ``torchvision.transforms.v2``
+- Use tensors instead of PIL images
+- Use ``torch.uint8`` dtype, especially for resizing
+- Resize with bilinear or bicubic mode
+
+This is what a typical transform pipeline could look like:
+
+.. code:: python
+
+    from torchvision.transforms import v2
+    transforms = v2.Compose([
+        v2.ToImage(),  # Convert to tensor, only needed if you had a PIL image
+        v2.ToDtype(torch.uint8, scale=True),  # optional, most input are already uint8 at this point
+        # ...
+        v2.RandomResizedCrop(size=(224, 224), antialias=True),  # Or Resize(antialias=True)
+        # ...
+        v2.ToDtype(torch.float32, scale=True),  # Normalize expects float input
+        v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+    ])
+
+The above should give you the best performance in a typical training environment
+that relies on the :class:`torch.utils.data.DataLoader` with ``num_workers >
+0``.
+
+Transforms tend to be sensitive to the input strides / memory format. Some
+transforms will be faster with channels-first images while others prefer
+channels-last. Like ``torch`` operators, most transforms will preserve the
+memory format of the input, but this may not always be respected due to
+implementation details. You may want to experiment a bit if you're chasing the
+very best performance.  Using :func:`torch.compile` on individual transforms may
+also help factoring out the memory format variable (e.g. on
+:class:`~torchvision.transforms.v2.Normalize`). Note that we're talking about
+**memory format**, not :ref:`tensor shape <conventions>`.
+
+Note that resize transforms like :class:`~torchvision.transforms.v2.Resize`
+and :class:`~torchvision.transforms.v2.RandomResizedCrop` typically prefer
+channels-last input and tend **not** to benefit from :func:`torch.compile` at
+this time.
+
+.. _functional_transforms:
+
+Transform classes, functionals, and kernels
+-------------------------------------------
+
+Transforms are available as classes like
+:class:`~torchvision.transforms.v2.Resize`, but also as functionals like
+:func:`~torchvision.transforms.v2.functional.resize` in the
+``torchvision.transforms.v2.functional`` namespace.
+This is very much like the :mod:`torch.nn` package which defines both classes
+and functional equivalents in :mod:`torch.nn.functional`.
+
+The functionals support PIL images, pure tensors, or :ref:`TVTensors
+<tv_tensors>`, e.g. both ``resize(image_tensor)`` and ``resize(boxes)`` are
+valid.
+
+.. note::
+
+    Random transforms like :class:`~torchvision.transforms.v2.RandomCrop` will
+    randomly sample some parameter each time they're called. Their functional
+    counterpart (:func:`~torchvision.transforms.v2.functional.crop`) does not do
+    any kind of random sampling and thus have a slighlty different
+    parametrization. The ``get_params()`` class method of the transforms class
+    can be used to perform parameter sampling when using the functional APIs.
+
+
+The ``torchvision.transforms.v2.functional`` namespace also contains what we
+call the "kernels". These are the low-level functions that implement the
+core functionalities for specific types, e.g. ``resize_bounding_boxes`` or
+```resized_crop_mask``. They are public, although not documented. Check the
+`code
+<https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/functional/__init__.py>`_
+to see which ones are available (note that those starting with a leading
+underscore are **not** public!). Kernels are only really useful if you want
+:ref:`torchscript support <transforms_torchscript>` for types like bounding
+boxes or masks.
+
+.. _transforms_torchscript:
+
+Torchscript support
+-------------------
+
+Most transform classes and functionals support torchscript. For composing
+transforms, use :class:`torch.nn.Sequential` instead of
+:class:`~torchvision.transforms.v2.Compose`:
+
+.. code:: python
+
+    transforms = torch.nn.Sequential(
+        CenterCrop(10),
+        Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),
+    )
+    scripted_transforms = torch.jit.script(transforms)
+
+.. warning::
+
+    v2 transforms support torchscript, but if you call ``torch.jit.script()`` on
+    a v2 **class** transform, you'll actually end up with its (scripted) v1
+    equivalent.  This may lead to slightly different results between the
+    scripted and eager executions due to implementation differences between v1
+    and v2.
+
+    If you really need torchscript support for the v2 transforms, we recommend
+    scripting the **functionals** from the
+    ``torchvision.transforms.v2.functional`` namespace to avoid surprises.
+
+
+Also note that the functionals only support torchscript for pure tensors, which
+are always treated as images. If you need torchscript support for other types
+like bounding boxes or masks, you can rely on the :ref:`low-level kernels
+<functional_transforms>`.
+
+For any custom transformations to be used with ``torch.jit.script``, they should
+be derived from ``torch.nn.Module``.
+
+See also: :ref:`sphx_glr_auto_examples_others_plot_scripted_tensor_transforms.py`.
+
+.. _v2_api_ref:
+
+V2 API reference - Recommended
+------------------------------
+
+Geometry
+^^^^^^^^
+
+Resizing
+""""""""
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.Resize
+    v2.ScaleJitter
+    v2.RandomShortestSize
+    v2.RandomResize
+
+Functionals
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    v2.functional.resize
+
+Cropping
+""""""""
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.RandomCrop
+    v2.RandomResizedCrop
+    v2.RandomIoUCrop
+    v2.CenterCrop
+    v2.FiveCrop
+    v2.TenCrop
+
+Functionals
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    v2.functional.crop
+    v2.functional.resized_crop
+    v2.functional.ten_crop
+    v2.functional.center_crop
+    v2.functional.five_crop
+
+Others
+""""""
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.RandomHorizontalFlip
+    v2.RandomVerticalFlip
+    v2.Pad
+    v2.RandomZoomOut
+    v2.RandomRotation
+    v2.RandomAffine
+    v2.RandomPerspective
+    v2.ElasticTransform
+
+Functionals
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    v2.functional.horizontal_flip
+    v2.functional.vertical_flip
+    v2.functional.pad
+    v2.functional.rotate
+    v2.functional.affine
+    v2.functional.perspective
+    v2.functional.elastic
+
+Color
+^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.ColorJitter
+    v2.RandomChannelPermutation
+    v2.RandomPhotometricDistort
+    v2.Grayscale
+    v2.RandomGrayscale
+    v2.GaussianBlur
+    v2.RandomInvert
+    v2.RandomPosterize
+    v2.RandomSolarize
+    v2.RandomAdjustSharpness
+    v2.RandomAutocontrast
+    v2.RandomEqualize
+
+Functionals
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    v2.functional.permute_channels
+    v2.functional.rgb_to_grayscale
+    v2.functional.to_grayscale
+    v2.functional.gaussian_blur
+    v2.functional.invert
+    v2.functional.posterize
+    v2.functional.solarize
+    v2.functional.adjust_sharpness
+    v2.functional.autocontrast
+    v2.functional.adjust_contrast
+    v2.functional.equalize
+    v2.functional.adjust_brightness
+    v2.functional.adjust_saturation
+    v2.functional.adjust_hue
+    v2.functional.adjust_gamma
+
+
+Composition
+^^^^^^^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.Compose
+    v2.RandomApply
+    v2.RandomChoice
+    v2.RandomOrder
+
+Miscellaneous
+^^^^^^^^^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.LinearTransformation
+    v2.Normalize
+    v2.RandomErasing
+    v2.Lambda
+    v2.SanitizeBoundingBoxes
+    v2.ClampBoundingBoxes
+    v2.UniformTemporalSubsample
+
+Functionals
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    v2.functional.normalize
+    v2.functional.erase
+    v2.functional.clamp_bounding_boxes
+    v2.functional.uniform_temporal_subsample
+
+.. _conversion_transforms:
+
+Conversion
+^^^^^^^^^^
+
+.. note::
+    Beware, some of these conversion transforms below will scale the values
+    while performing the conversion, while some may not do any scaling. By
+    scaling, we mean e.g. that a ``uint8`` -> ``float32`` would map the [0,
+    255] range into [0, 1] (and vice-versa). See :ref:`range_and_dtype`.
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.ToImage
+    v2.ToPureTensor
+    v2.PILToTensor
+    v2.ToPILImage
+    v2.ToDtype
+    v2.ConvertBoundingBoxFormat
+
+functionals
+
+.. autosummary::
+    :toctree: generated/
+    :template: functional.rst
+
+    v2.functional.to_image
+    v2.functional.pil_to_tensor
+    v2.functional.to_pil_image
+    v2.functional.to_dtype
+    v2.functional.convert_bounding_box_format
+
+
+Deprecated
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.ToTensor
+    v2.functional.to_tensor
+    v2.ConvertImageDtype
+    v2.functional.convert_image_dtype
+
+Auto-Augmentation
+^^^^^^^^^^^^^^^^^
+
+`AutoAugment <https://arxiv.org/pdf/1805.09501.pdf>`_ is a common Data Augmentation technique that can improve the accuracy of Image Classification models.
+Though the data augmentation policies are directly linked to their trained dataset, empirical studies show that
+ImageNet policies provide significant improvements when applied to other datasets.
+In TorchVision we implemented 3 policies learned on the following datasets: ImageNet, CIFAR10 and SVHN.
+The new transform can be used standalone or mixed-and-matched with existing transforms:
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.AutoAugment
+    v2.RandAugment
+    v2.TrivialAugmentWide
+    v2.AugMix
+
+
+CutMix - MixUp
+^^^^^^^^^^^^^^
+
+CutMix and MixUp are special transforms that
+are meant to be used on batches rather than on individual images, because they
+are combining pairs of images together. These can be used after the dataloader
+(once the samples are batched), or part of a collation function. See
+:ref:`sphx_glr_auto_examples_transforms_plot_cutmix_mixup.py` for detailed usage examples.
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    v2.CutMix
+    v2.MixUp
+
+Developer tools
+^^^^^^^^^^^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    v2.functional.register_kernel
+
+
+V1 API Reference
+----------------
+
+Geometry
+^^^^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    Resize
+    RandomCrop
+    RandomResizedCrop
+    CenterCrop
+    FiveCrop
+    TenCrop
+    Pad
+    RandomRotation
+    RandomAffine
+    RandomPerspective
+    ElasticTransform
+    RandomHorizontalFlip
+    RandomVerticalFlip
+
+
+Color
+^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    ColorJitter
+    Grayscale
+    RandomGrayscale
+    GaussianBlur
+    RandomInvert
+    RandomPosterize
+    RandomSolarize
+    RandomAdjustSharpness
+    RandomAutocontrast
+    RandomEqualize
+
+Composition
+^^^^^^^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    Compose
+    RandomApply
+    RandomChoice
+    RandomOrder
+
+Miscellaneous
+^^^^^^^^^^^^^
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    LinearTransformation
+    Normalize
+    RandomErasing
+    Lambda
+
+Conversion
+^^^^^^^^^^
+
+.. note::
+    Beware, some of these conversion transforms below will scale the values
+    while performing the conversion, while some may not do any scaling. By
+    scaling, we mean e.g. that a ``uint8`` -> ``float32`` would map the [0,
+    255] range into [0, 1] (and vice-versa). See :ref:`range_and_dtype`.
+    
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    ToPILImage
+    ToTensor
+    PILToTensor
+    ConvertImageDtype
+
+Auto-Augmentation
+^^^^^^^^^^^^^^^^^
+
+`AutoAugment <https://arxiv.org/pdf/1805.09501.pdf>`_ is a common Data Augmentation technique that can improve the accuracy of Image Classification models.
+Though the data augmentation policies are directly linked to their trained dataset, empirical studies show that
+ImageNet policies provide significant improvements when applied to other datasets.
+In TorchVision we implemented 3 policies learned on the following datasets: ImageNet, CIFAR10 and SVHN.
+The new transform can be used standalone or mixed-and-matched with existing transforms:
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    AutoAugmentPolicy
+    AutoAugment
+    RandAugment
+    TrivialAugmentWide
+    AugMix
+
+
+
+Functional Transforms
+^^^^^^^^^^^^^^^^^^^^^
+
+.. currentmodule:: torchvision.transforms.functional
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    adjust_brightness
+    adjust_contrast
+    adjust_gamma
+    adjust_hue
+    adjust_saturation
+    adjust_sharpness
+    affine
+    autocontrast
+    center_crop
+    convert_image_dtype
+    crop
+    equalize
+    erase
+    five_crop
+    gaussian_blur
+    get_dimensions
+    get_image_num_channels
+    get_image_size
+    hflip
+    invert
+    normalize
+    pad
+    perspective
+    pil_to_tensor
+    posterize
+    resize
+    resized_crop
+    rgb_to_grayscale
+    rotate
+    solarize
+    ten_crop
+    to_grayscale
+    to_pil_image
+    to_tensor
+    vflip
diff --git a/0.16/_sources/tv_tensors.rst.txt b/0.16/_sources/tv_tensors.rst.txt
new file mode 100644
index 00000000000..cb8a3c45fa9
--- /dev/null
+++ b/0.16/_sources/tv_tensors.rst.txt
@@ -0,0 +1,29 @@
+.. _tv_tensors:
+
+TVTensors
+==========
+
+.. currentmodule:: torchvision.tv_tensors
+
+TVTensors are :class:`torch.Tensor` subclasses which the v2 :ref:`transforms
+<transforms>` use under the hood to dispatch their inputs to the appropriate
+lower-level kernels. Most users do not need to manipulate TVTensors directly.
+
+Refer to
+:ref:`sphx_glr_auto_examples_transforms_plot_transforms_getting_started.py` for
+an introduction to TVTensors, or
+:ref:`sphx_glr_auto_examples_transforms_plot_tv_tensors.py` for more advanced
+info.
+
+.. autosummary::
+    :toctree: generated/
+    :template: class.rst
+
+    Image
+    Video
+    BoundingBoxFormat
+    BoundingBoxes
+    Mask
+    TVTensor
+    set_return_type
+    wrap
diff --git a/0.16/_sources/utils.rst.txt b/0.16/_sources/utils.rst.txt
new file mode 100644
index 00000000000..cda04de900a
--- /dev/null
+++ b/0.16/_sources/utils.rst.txt
@@ -0,0 +1,20 @@
+.. _utils:
+
+Utils
+=====
+
+The ``torchvision.utils`` module contains various utilities, mostly :ref:`for
+visualization <sphx_glr_auto_examples_others_plot_visualization_utils.py>`.
+
+.. currentmodule:: torchvision.utils
+
+.. autosummary::
+    :toctree: generated/
+    :template: function.rst
+
+    draw_bounding_boxes
+    draw_segmentation_masks
+    draw_keypoints
+    flow_to_image
+    make_grid
+    save_image
diff --git a/0.16/_static/_sphinx_javascript_frameworks_compat.js b/0.16/_static/_sphinx_javascript_frameworks_compat.js
new file mode 100644
index 00000000000..8549469dc29
--- /dev/null
+++ b/0.16/_static/_sphinx_javascript_frameworks_compat.js
@@ -0,0 +1,134 @@
+/*
+ * _sphinx_javascript_frameworks_compat.js
+ * ~~~~~~~~~~
+ *
+ * Compatability shim for jQuery and underscores.js.
+ *
+ * WILL BE REMOVED IN Sphinx 6.0
+ * xref RemovedInSphinx60Warning
+ *
+ */
+
+/**
+ * select a different prefix for underscore
+ */
+$u = _.noConflict();
+
+
+/**
+ * small helper function to urldecode strings
+ *
+ * See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/decodeURIComponent#Decoding_query_parameters_from_a_URL
+ */
+jQuery.urldecode = function(x) {
+    if (!x) {
+        return x
+    }
+    return decodeURIComponent(x.replace(/\+/g, ' '));
+};
+
+/**
+ * small helper function to urlencode strings
+ */
+jQuery.urlencode = encodeURIComponent;
+
+/**
+ * This function returns the parsed url parameters of the
+ * current request. Multiple values per key are supported,
+ * it will always return arrays of strings for the value parts.
+ */
+jQuery.getQueryParameters = function(s) {
+    if (typeof s === 'undefined')
+        s = document.location.search;
+    var parts = s.substr(s.indexOf('?') + 1).split('&');
+    var result = {};
+    for (var i = 0; i < parts.length; i++) {
+        var tmp = parts[i].split('=', 2);
+        var key = jQuery.urldecode(tmp[0]);
+        var value = jQuery.urldecode(tmp[1]);
+        if (key in result)
+            result[key].push(value);
+        else
+            result[key] = [value];
+    }
+    return result;
+};
+
+/**
+ * highlight a given string on a jquery object by wrapping it in
+ * span elements with the given class name.
+ */
+jQuery.fn.highlightText = function(text, className) {
+    function highlight(node, addItems) {
+        if (node.nodeType === 3) {
+            var val = node.nodeValue;
+            var pos = val.toLowerCase().indexOf(text);
+            if (pos >= 0 &&
+                !jQuery(node.parentNode).hasClass(className) &&
+                !jQuery(node.parentNode).hasClass("nohighlight")) {
+                var span;
+                var isInSVG = jQuery(node).closest("body, svg, foreignObject").is("svg");
+                if (isInSVG) {
+                    span = document.createElementNS("http://www.w3.org/2000/svg", "tspan");
+                } else {
+                    span = document.createElement("span");
+                    span.className = className;
+                }
+                span.appendChild(document.createTextNode(val.substr(pos, text.length)));
+                node.parentNode.insertBefore(span, node.parentNode.insertBefore(
+                    document.createTextNode(val.substr(pos + text.length)),
+                    node.nextSibling));
+                node.nodeValue = val.substr(0, pos);
+                if (isInSVG) {
+                    var rect = document.createElementNS("http://www.w3.org/2000/svg", "rect");
+                    var bbox = node.parentElement.getBBox();
+                    rect.x.baseVal.value = bbox.x;
+                    rect.y.baseVal.value = bbox.y;
+                    rect.width.baseVal.value = bbox.width;
+                    rect.height.baseVal.value = bbox.height;
+                    rect.setAttribute('class', className);
+                    addItems.push({
+                        "parent": node.parentNode,
+                        "target": rect});
+                }
+            }
+        }
+        else if (!jQuery(node).is("button, select, textarea")) {
+            jQuery.each(node.childNodes, function() {
+                highlight(this, addItems);
+            });
+        }
+    }
+    var addItems = [];
+    var result = this.each(function() {
+        highlight(this, addItems);
+    });
+    for (var i = 0; i < addItems.length; ++i) {
+        jQuery(addItems[i].parent).before(addItems[i].target);
+    }
+    return result;
+};
+
+/*
+ * backward compatibility for jQuery.browser
+ * This will be supported until firefox bug is fixed.
+ */
+if (!jQuery.browser) {
+    jQuery.uaMatch = function(ua) {
+        ua = ua.toLowerCase();
+
+        var match = /(chrome)[ \/]([\w.]+)/.exec(ua) ||
+            /(webkit)[ \/]([\w.]+)/.exec(ua) ||
+            /(opera)(?:.*version|)[ \/]([\w.]+)/.exec(ua) ||
+            /(msie) ([\w.]+)/.exec(ua) ||
+            ua.indexOf("compatible") < 0 && /(mozilla)(?:.*? rv:([\w.]+)|)/.exec(ua) ||
+            [];
+
+        return {
+            browser: match[ 1 ] || "",
+            version: match[ 2 ] || "0"
+        };
+    };
+    jQuery.browser = {};
+    jQuery.browser[jQuery.uaMatch(navigator.userAgent).browser] = true;
+}
diff --git a/0.16/_static/basic.css b/0.16/_static/basic.css
new file mode 100644
index 00000000000..7243282d260
--- /dev/null
+++ b/0.16/_static/basic.css
@@ -0,0 +1,930 @@
+/*
+ * basic.css
+ * ~~~~~~~~~
+ *
+ * Sphinx stylesheet -- basic theme.
+ *
+ * :copyright: Copyright 2007-2022 by the Sphinx team, see AUTHORS.
+ * :license: BSD, see LICENSE for details.
+ *
+ */
+
+/* -- main layout ----------------------------------------------------------- */
+
+div.clearer {
+    clear: both;
+}
+
+div.section::after {
+    display: block;
+    content: '';
+    clear: left;
+}
+
+/* -- relbar ---------------------------------------------------------------- */
+
+div.related {
+    width: 100%;
+    font-size: 90%;
+}
+
+div.related h3 {
+    display: none;
+}
+
+div.related ul {
+    margin: 0;
+    padding: 0 0 0 10px;
+    list-style: none;
+}
+
+div.related li {
+    display: inline;
+}
+
+div.related li.right {
+    float: right;
+    margin-right: 5px;
+}
+
+/* -- sidebar --------------------------------------------------------------- */
+
+div.sphinxsidebarwrapper {
+    padding: 10px 5px 0 10px;
+}
+
+div.sphinxsidebar {
+    float: left;
+    width: 230px;
+    margin-left: -100%;
+    font-size: 90%;
+    word-wrap: break-word;
+    overflow-wrap : break-word;
+}
+
+div.sphinxsidebar ul {
+    list-style: none;
+}
+
+div.sphinxsidebar ul ul,
+div.sphinxsidebar ul.want-points {
+    margin-left: 20px;
+    list-style: square;
+}
+
+div.sphinxsidebar ul ul {
+    margin-top: 0;
+    margin-bottom: 0;
+}
+
+div.sphinxsidebar form {
+    margin-top: 10px;
+}
+
+div.sphinxsidebar input {
+    border: 1px solid #98dbcc;
+    font-family: sans-serif;
+    font-size: 1em;
+}
+
+div.sphinxsidebar #searchbox form.search {
+    overflow: hidden;
+}
+
+div.sphinxsidebar #searchbox input[type="text"] {
+    float: left;
+    width: 80%;
+    padding: 0.25em;
+    box-sizing: border-box;
+}
+
+div.sphinxsidebar #searchbox input[type="submit"] {
+    float: left;
+    width: 20%;
+    border-left: none;
+    padding: 0.25em;
+    box-sizing: border-box;
+}
+
+
+img {
+    border: 0;
+    max-width: 100%;
+}
+
+/* -- search page ----------------------------------------------------------- */
+
+ul.search {
+    margin: 10px 0 0 20px;
+    padding: 0;
+}
+
+ul.search li {
+    padding: 5px 0 5px 20px;
+    background-image: url(file.png);
+    background-repeat: no-repeat;
+    background-position: 0 7px;
+}
+
+ul.search li a {
+    font-weight: bold;
+}
+
+ul.search li p.context {
+    color: #888;
+    margin: 2px 0 0 30px;
+    text-align: left;
+}
+
+ul.keywordmatches li.goodmatch a {
+    font-weight: bold;
+}
+
+/* -- index page ------------------------------------------------------------ */
+
+table.contentstable {
+    width: 90%;
+    margin-left: auto;
+    margin-right: auto;
+}
+
+table.contentstable p.biglink {
+    line-height: 150%;
+}
+
+a.biglink {
+    font-size: 1.3em;
+}
+
+span.linkdescr {
+    font-style: italic;
+    padding-top: 5px;
+    font-size: 90%;
+}
+
+/* -- general index --------------------------------------------------------- */
+
+table.indextable {
+    width: 100%;
+}
+
+table.indextable td {
+    text-align: left;
+    vertical-align: top;
+}
+
+table.indextable ul {
+    margin-top: 0;
+    margin-bottom: 0;
+    list-style-type: none;
+}
+
+table.indextable > tbody > tr > td > ul {
+    padding-left: 0em;
+}
+
+table.indextable tr.pcap {
+    height: 10px;
+}
+
+table.indextable tr.cap {
+    margin-top: 10px;
+    background-color: #f2f2f2;
+}
+
+img.toggler {
+    margin-right: 3px;
+    margin-top: 3px;
+    cursor: pointer;
+}
+
+div.modindex-jumpbox {
+    border-top: 1px solid #ddd;
+    border-bottom: 1px solid #ddd;
+    margin: 1em 0 1em 0;
+    padding: 0.4em;
+}
+
+div.genindex-jumpbox {
+    border-top: 1px solid #ddd;
+    border-bottom: 1px solid #ddd;
+    margin: 1em 0 1em 0;
+    padding: 0.4em;
+}
+
+/* -- domain module index --------------------------------------------------- */
+
+table.modindextable td {
+    padding: 2px;
+    border-collapse: collapse;
+}
+
+/* -- general body styles --------------------------------------------------- */
+
+div.body {
+    min-width: 360px;
+    max-width: 800px;
+}
+
+div.body p, div.body dd, div.body li, div.body blockquote {
+    -moz-hyphens: auto;
+    -ms-hyphens: auto;
+    -webkit-hyphens: auto;
+    hyphens: auto;
+}
+
+a.headerlink {
+    visibility: hidden;
+}
+
+a.brackets:before,
+span.brackets > a:before{
+    content: "[";
+}
+
+a.brackets:after,
+span.brackets > a:after {
+    content: "]";
+}
+
+h1:hover > a.headerlink,
+h2:hover > a.headerlink,
+h3:hover > a.headerlink,
+h4:hover > a.headerlink,
+h5:hover > a.headerlink,
+h6:hover > a.headerlink,
+dt:hover > a.headerlink,
+caption:hover > a.headerlink,
+p.caption:hover > a.headerlink,
+div.code-block-caption:hover > a.headerlink {
+    visibility: visible;
+}
+
+div.body p.caption {
+    text-align: inherit;
+}
+
+div.body td {
+    text-align: left;
+}
+
+.first {
+    margin-top: 0 !important;
+}
+
+p.rubric {
+    margin-top: 30px;
+    font-weight: bold;
+}
+
+img.align-left, figure.align-left, .figure.align-left, object.align-left {
+    clear: left;
+    float: left;
+    margin-right: 1em;
+}
+
+img.align-right, figure.align-right, .figure.align-right, object.align-right {
+    clear: right;
+    float: right;
+    margin-left: 1em;
+}
+
+img.align-center, figure.align-center, .figure.align-center, object.align-center {
+  display: block;
+  margin-left: auto;
+  margin-right: auto;
+}
+
+img.align-default, figure.align-default, .figure.align-default {
+  display: block;
+  margin-left: auto;
+  margin-right: auto;
+}
+
+.align-left {
+    text-align: left;
+}
+
+.align-center {
+    text-align: center;
+}
+
+.align-default {
+    text-align: center;
+}
+
+.align-right {
+    text-align: right;
+}
+
+/* -- sidebars -------------------------------------------------------------- */
+
+div.sidebar,
+aside.sidebar {
+    margin: 0 0 0.5em 1em;
+    border: 1px solid #ddb;
+    padding: 7px;
+    background-color: #ffe;
+    width: 40%;
+    float: right;
+    clear: right;
+    overflow-x: auto;
+}
+
+p.sidebar-title {
+    font-weight: bold;
+}
+
+div.admonition, div.topic, blockquote {
+    clear: left;
+}
+
+/* -- topics ---------------------------------------------------------------- */
+
+div.topic {
+    border: 1px solid #ccc;
+    padding: 7px;
+    margin: 10px 0 10px 0;
+}
+
+p.topic-title {
+    font-size: 1.1em;
+    font-weight: bold;
+    margin-top: 10px;
+}
+
+/* -- admonitions ----------------------------------------------------------- */
+
+div.admonition {
+    margin-top: 10px;
+    margin-bottom: 10px;
+    padding: 7px;
+}
+
+div.admonition dt {
+    font-weight: bold;
+}
+
+p.admonition-title {
+    margin: 0px 10px 5px 0px;
+    font-weight: bold;
+}
+
+div.body p.centered {
+    text-align: center;
+    margin-top: 25px;
+}
+
+/* -- content of sidebars/topics/admonitions -------------------------------- */
+
+div.sidebar > :last-child,
+aside.sidebar > :last-child,
+div.topic > :last-child,
+div.admonition > :last-child {
+    margin-bottom: 0;
+}
+
+div.sidebar::after,
+aside.sidebar::after,
+div.topic::after,
+div.admonition::after,
+blockquote::after {
+    display: block;
+    content: '';
+    clear: both;
+}
+
+/* -- tables ---------------------------------------------------------------- */
+
+table.docutils {
+    margin-top: 10px;
+    margin-bottom: 10px;
+    border: 0;
+    border-collapse: collapse;
+}
+
+table.align-center {
+    margin-left: auto;
+    margin-right: auto;
+}
+
+table.align-default {
+    margin-left: auto;
+    margin-right: auto;
+}
+
+table caption span.caption-number {
+    font-style: italic;
+}
+
+table caption span.caption-text {
+}
+
+table.docutils td, table.docutils th {
+    padding: 1px 8px 1px 5px;
+    border-top: 0;
+    border-left: 0;
+    border-right: 0;
+    border-bottom: 1px solid #aaa;
+}
+
+th {
+    text-align: left;
+    padding-right: 5px;
+}
+
+table.citation {
+    border-left: solid 1px gray;
+    margin-left: 1px;
+}
+
+table.citation td {
+    border-bottom: none;
+}
+
+th > :first-child,
+td > :first-child {
+    margin-top: 0px;
+}
+
+th > :last-child,
+td > :last-child {
+    margin-bottom: 0px;
+}
+
+/* -- figures --------------------------------------------------------------- */
+
+div.figure, figure {
+    margin: 0.5em;
+    padding: 0.5em;
+}
+
+div.figure p.caption, figcaption {
+    padding: 0.3em;
+}
+
+div.figure p.caption span.caption-number,
+figcaption span.caption-number {
+    font-style: italic;
+}
+
+div.figure p.caption span.caption-text,
+figcaption span.caption-text {
+}
+
+/* -- field list styles ----------------------------------------------------- */
+
+table.field-list td, table.field-list th {
+    border: 0 !important;
+}
+
+.field-list ul {
+    margin: 0;
+    padding-left: 1em;
+}
+
+.field-list p {
+    margin: 0;
+}
+
+.field-name {
+    -moz-hyphens: manual;
+    -ms-hyphens: manual;
+    -webkit-hyphens: manual;
+    hyphens: manual;
+}
+
+/* -- hlist styles ---------------------------------------------------------- */
+
+table.hlist {
+    margin: 1em 0;
+}
+
+table.hlist td {
+    vertical-align: top;
+}
+
+/* -- object description styles --------------------------------------------- */
+
+.sig {
+	font-family: 'Consolas', 'Menlo', 'DejaVu Sans Mono', 'Bitstream Vera Sans Mono', monospace;
+}
+
+.sig-name, code.descname {
+    background-color: transparent;
+    font-weight: bold;
+}
+
+.sig-name {
+	font-size: 1.1em;
+}
+
+code.descname {
+    font-size: 1.2em;
+}
+
+.sig-prename, code.descclassname {
+    background-color: transparent;
+}
+
+.optional {
+    font-size: 1.3em;
+}
+
+.sig-paren {
+    font-size: larger;
+}
+
+.sig-param.n {
+	font-style: italic;
+}
+
+/* C++ specific styling */
+
+.sig-inline.c-texpr,
+.sig-inline.cpp-texpr {
+	font-family: unset;
+}
+
+.sig.c   .k, .sig.c   .kt,
+.sig.cpp .k, .sig.cpp .kt {
+	color: #0033B3;
+}
+
+.sig.c   .m,
+.sig.cpp .m {
+	color: #1750EB;
+}
+
+.sig.c   .s, .sig.c   .sc,
+.sig.cpp .s, .sig.cpp .sc {
+	color: #067D17;
+}
+
+
+/* -- other body styles ----------------------------------------------------- */
+
+ol.arabic {
+    list-style: decimal;
+}
+
+ol.loweralpha {
+    list-style: lower-alpha;
+}
+
+ol.upperalpha {
+    list-style: upper-alpha;
+}
+
+ol.lowerroman {
+    list-style: lower-roman;
+}
+
+ol.upperroman {
+    list-style: upper-roman;
+}
+
+:not(li) > ol > li:first-child > :first-child,
+:not(li) > ul > li:first-child > :first-child {
+    margin-top: 0px;
+}
+
+:not(li) > ol > li:last-child > :last-child,
+:not(li) > ul > li:last-child > :last-child {
+    margin-bottom: 0px;
+}
+
+ol.simple ol p,
+ol.simple ul p,
+ul.simple ol p,
+ul.simple ul p {
+    margin-top: 0;
+}
+
+ol.simple > li:not(:first-child) > p,
+ul.simple > li:not(:first-child) > p {
+    margin-top: 0;
+}
+
+ol.simple p,
+ul.simple p {
+    margin-bottom: 0;
+}
+
+/* Docutils 0.17 and older (footnotes & citations) */
+dl.footnote > dt,
+dl.citation > dt {
+    float: left;
+    margin-right: 0.5em;
+}
+
+dl.footnote > dd,
+dl.citation > dd {
+    margin-bottom: 0em;
+}
+
+dl.footnote > dd:after,
+dl.citation > dd:after {
+    content: "";
+    clear: both;
+}
+
+/* Docutils 0.18+ (footnotes & citations) */
+aside.footnote > span,
+div.citation > span {
+    float: left;
+}
+aside.footnote > span:last-of-type,
+div.citation > span:last-of-type {
+  padding-right: 0.5em;
+}
+aside.footnote > p {
+  margin-left: 2em;
+}
+div.citation > p {
+  margin-left: 4em;
+}
+aside.footnote > p:last-of-type,
+div.citation > p:last-of-type {
+    margin-bottom: 0em;
+}
+aside.footnote > p:last-of-type:after,
+div.citation > p:last-of-type:after {
+    content: "";
+    clear: both;
+}
+
+/* Footnotes & citations ends */
+
+dl.field-list {
+    display: grid;
+    grid-template-columns: fit-content(30%) auto;
+}
+
+dl.field-list > dt {
+    font-weight: bold;
+    word-break: break-word;
+    padding-left: 0.5em;
+    padding-right: 5px;
+}
+
+dl.field-list > dt:after {
+    content: ":";
+}
+
+dl.field-list > dd {
+    padding-left: 0.5em;
+    margin-top: 0em;
+    margin-left: 0em;
+    margin-bottom: 0em;
+}
+
+dl {
+    margin-bottom: 15px;
+}
+
+dd > :first-child {
+    margin-top: 0px;
+}
+
+dd ul, dd table {
+    margin-bottom: 10px;
+}
+
+dd {
+    margin-top: 3px;
+    margin-bottom: 10px;
+    margin-left: 30px;
+}
+
+dl > dd:last-child,
+dl > dd:last-child > :last-child {
+    margin-bottom: 0;
+}
+
+dt:target, span.highlighted {
+    background-color: #fbe54e;
+}
+
+rect.highlighted {
+    fill: #fbe54e;
+}
+
+dl.glossary dt {
+    font-weight: bold;
+    font-size: 1.1em;
+}
+
+.versionmodified {
+    font-style: italic;
+}
+
+.system-message {
+    background-color: #fda;
+    padding: 5px;
+    border: 3px solid red;
+}
+
+.footnote:target  {
+    background-color: #ffa;
+}
+
+.line-block {
+    display: block;
+    margin-top: 1em;
+    margin-bottom: 1em;
+}
+
+.line-block .line-block {
+    margin-top: 0;
+    margin-bottom: 0;
+    margin-left: 1.5em;
+}
+
+.guilabel, .menuselection {
+    font-family: sans-serif;
+}
+
+.accelerator {
+    text-decoration: underline;
+}
+
+.classifier {
+    font-style: oblique;
+}
+
+.classifier:before {
+    font-style: normal;
+    margin: 0 0.5em;
+    content: ":";
+    display: inline-block;
+}
+
+abbr, acronym {
+    border-bottom: dotted 1px;
+    cursor: help;
+}
+
+/* -- code displays --------------------------------------------------------- */
+
+pre {
+    overflow: auto;
+    overflow-y: hidden;  /* fixes display issues on Chrome browsers */
+}
+
+pre, div[class*="highlight-"] {
+    clear: both;
+}
+
+span.pre {
+    -moz-hyphens: none;
+    -ms-hyphens: none;
+    -webkit-hyphens: none;
+    hyphens: none;
+    white-space: nowrap;
+}
+
+div[class*="highlight-"] {
+    margin: 1em 0;
+}
+
+td.linenos pre {
+    border: 0;
+    background-color: transparent;
+    color: #aaa;
+}
+
+table.highlighttable {
+    display: block;
+}
+
+table.highlighttable tbody {
+    display: block;
+}
+
+table.highlighttable tr {
+    display: flex;
+}
+
+table.highlighttable td {
+    margin: 0;
+    padding: 0;
+}
+
+table.highlighttable td.linenos {
+    padding-right: 0.5em;
+}
+
+table.highlighttable td.code {
+    flex: 1;
+    overflow: hidden;
+}
+
+.highlight .hll {
+    display: block;
+}
+
+div.highlight pre,
+table.highlighttable pre {
+    margin: 0;
+}
+
+div.code-block-caption + div {
+    margin-top: 0;
+}
+
+div.code-block-caption {
+    margin-top: 1em;
+    padding: 2px 5px;
+    font-size: small;
+}
+
+div.code-block-caption code {
+    background-color: transparent;
+}
+
+table.highlighttable td.linenos,
+span.linenos,
+div.highlight span.gp {  /* gp: Generic.Prompt */
+  user-select: none;
+  -webkit-user-select: text; /* Safari fallback only */
+  -webkit-user-select: none; /* Chrome/Safari */
+  -moz-user-select: none; /* Firefox */
+  -ms-user-select: none; /* IE10+ */
+}
+
+div.code-block-caption span.caption-number {
+    padding: 0.1em 0.3em;
+    font-style: italic;
+}
+
+div.code-block-caption span.caption-text {
+}
+
+div.literal-block-wrapper {
+    margin: 1em 0;
+}
+
+code.xref, a code {
+    background-color: transparent;
+    font-weight: bold;
+}
+
+h1 code, h2 code, h3 code, h4 code, h5 code, h6 code {
+    background-color: transparent;
+}
+
+.viewcode-link {
+    float: right;
+}
+
+.viewcode-back {
+    float: right;
+    font-family: sans-serif;
+}
+
+div.viewcode-block:target {
+    margin: -1px -10px;
+    padding: 0 10px;
+}
+
+/* -- math display ---------------------------------------------------------- */
+
+img.math {
+    vertical-align: middle;
+}
+
+div.body div.math p {
+    text-align: center;
+}
+
+span.eqno {
+    float: right;
+}
+
+span.eqno a.headerlink {
+    position: absolute;
+    z-index: 1;
+}
+
+div.math:hover a.headerlink {
+    visibility: visible;
+}
+
+/* -- printout stylesheet --------------------------------------------------- */
+
+@media print {
+    div.document,
+    div.documentwrapper,
+    div.bodywrapper {
+        margin: 0 !important;
+        width: 100%;
+    }
+
+    div.sphinxsidebar,
+    div.related,
+    div.footer,
+    #top-link {
+        display: none;
+    }
+}
\ No newline at end of file
diff --git a/0.16/_static/binder_badge_logo.svg b/0.16/_static/binder_badge_logo.svg
new file mode 100644
index 00000000000..327f6b639a9
--- /dev/null
+++ b/0.16/_static/binder_badge_logo.svg
@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="109" height="20"><linearGradient id="b" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="a"><rect width="109" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#a)"><path fill="#555" d="M0 0h64v20H0z"/><path fill="#579aca" d="M64 0h45v20H64z"/><path fill="url(#b)" d="M0 0h109v20H0z"/></g><g fill="#fff" text-anchor="middle" font-family="DejaVu Sans,Verdana,Geneva,sans-serif" font-size="110"><image x="5" y="3" width="14" height="14" xlink:href="data:image/png;base64,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"/> <text x="415" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="370">launch</text><text x="415" y="140" transform="scale(.1)" textLength="370">launch</text><text x="855" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="350">binder</text><text x="855" y="140" transform="scale(.1)" textLength="350">binder</text></g> </svg>
\ No newline at end of file
diff --git a/0.16/_static/broken_example.png b/0.16/_static/broken_example.png
new file mode 100644
index 0000000000000000000000000000000000000000..4fea24e7df4781c2c32c8d7995511ac89e953145
GIT binary patch
literal 21404
zcmaHTWmKEb({>2%?jE#QaVRdqp+KR9;KkkDwKxQKD^i>mcWI%x7Y&k9+}-8j_dn<T
z`tCW=b7$|J+1Z)7uANP^#(M>93<?YY0D!He2>t*7AV|HwhohmqKD$2joW8ywxydML
zqrEnNH1nv}&*;vI5H|n-lkmR>0w6n=?Db1>cR5{mEhkHNug|U)052~uE*nQXH?z;q
z7F<rQRyoIF6aWAnKnX0Z?VWp+=haDV(e&Ko<O@W{Lt_lVBk;uzUn{|!AtFL10{4G`
z%gG_9#+p18eRZuvPKw1Qf`UL@Ekq*pc~BV@G|u?T>UG-5?6Je82K9+|@BAhth9CST
zcHAcNv&0`(9}ayi)2wQc@&7-~jsgpX(%s?2qY+ic?=2k&^fkcbJ3TknEkKtDe&=cQ
z0sPtmK2NQh;D#Y$&7V2Y^6sv#*~LahMU`V>iICenIfqO4taPuTpf)ZngI?P{O4u`s
z0y&n$M&@6Xr%?l)|9V(8`Q9z9qwelr@uMg5WpxhDtsm=(Xg5w7AEldPGDW?Jy-Nb0
zrg(@%fC03Cai>T62zfwNHn*;$2*qH0d^@__p>5R6EErAj5#AlsOa^i9D#F^?Qw~A{
zkmHTp_)*CJu0q2SL>9z3HHd#g3=t2xx`O)jf=n3THGrF2mjgt7WbgLWh!%!QQ{byY
zs4-Exf&)U3dmE4+9sqvyqd4YET>w?7NpYH8tM78bHVA886=S{#+o|Ww%jna<xTxL2
z)BSj(Uf<pozzHc;1NL{?_IVw^Tf|8e-OP8E!r#LR{&}u9<9zqid~{Zm`F3q2^(y$N
z37)D;XChz}I5w1AaI!vM$nqVtePSJWjCKl6f21-C9_Xb+$w5Sdq?~l<cl`#n_r8t|
z4Fxe+Ja{i^>eJ@67jlAh$qiTgZ(5vF<r+5>UE9+*NvohRuUx<&0xEfDR#2V*@-%Wn
z)GmNqmvj+T7b9*$-=#YUnShrTMGRS0HVbYs?jMgB%&9?j>ayO9JjK_0E9k?3{bd)Z
z0Ose$YR}|tK}DQI^=2gPIWVn9lT1ja%we!|k_o=aQ3*@{KeR;j{RT)gE=%I8)Y(Ll
zn1I1yKmR%TA-+XndVBuRBJ~L-(p!aa%)dz2>sNJ_9Zh_TGlP=t6~DyWyE~uXqKVJh
z8x@5Trj%SSus`AIm;dM=<OSE*yyrqPLeix^cN2@O{@vZ19ysBpq3Rqjs~QRh^P}<l
zc2BUn{3tPbSyQ~`#bFXN0_59NRmKA_Rd?2q$t=0WduIS<YCn4YQQBk~{?pnK7UAC`
z3@ns<4p!rXIOPRT)OLg$244SC8W{_xZ0Fpb-?#7$+Y`nHF;>Jbc{+kuIYfUdS$LV)
zL$}h_C6R>jKDq_3Oy}rju2_$^Hv_H}vbIElV|)=~IMVZb=f2&6kSHWF{AnESfkhl}
zxFDP2TJrh}ed?Tmc7rGdnN~XUB>Y#o%l;^W{9W_7##akfa=UUNQ!H-KSM%17Erj(Y
z+nd){QocQTN}~%#KzWoTlPqrvVT?>WJfLpy0@fFbYQp*8e#J%r14uczC5wW}BwKo+
zb6{uC9+xUyaVl6R>G&kiRY(6vSpXExm-SR8n0SrT@}VTsL_z+)AMImlxtbZTIB+<7
z2`TV&%WCF2;G9|qzhSEVfjL#$JK|3QDT007b0O@~A0r40M6k?g`tAEaJ><c+Y$6da
zLSW~c8|i<f6;E2w@W;{q)RZ<dhAEv0QDw^2KB(YLUnykm=BZh@#wO`^7YxU@)Xy)^
zM{zlW!IZxkKcbf>o4sF)xO#dhJal5wZh>KXSC6%3tABw}3V^TFR*3(+WA-y(f$M^I
zQ^MwR8khCvwTmO<6&PQQLoob}&ksyzvv^@+b+}u(|2P!t1BRyQ&Qtgm-T$re$4)=q
zpoUBOR85SY<H&#7H#|l`WsLQy-tF>4>i^LC+Cgw%8Ok@1>4SoFu9Fr^eqi93KYRjk
zlO`94lDE45fCTD6g2-CSb|NK=AR3{rU7C5_IDBc075!s3ekC7=AP=raQ9bLBP2#>Y
z_930Wp^fA1)G0I{RAjV!Gj4k>5emEeRlB#1fwV3t_9kTXC<sNEVzJ+(0Z@mWY4g2Q
z`lOJYff^&}brSRtCbN$*v4Auf^$F(>H`iwEg`O9qs>HNMN&LDoyv=@F%`7Duax+Tz
zSt}uKSj@$X>yB%z9`9`XzvECt0mYb9_|r0(2Hlf@6O>j7^}vM5V&#Nd){c88-95qd
z1b|cVK^Nh`kC(^hXMy)upb_AW{0!O3`GI9q!R+_{*{3SmrQj`pOz{(E@8ZhyGs***
zeW<AoW=Kb{Byw~%j54Ld-7W|8bAfXl$I6SVPu~ea)b|6bJK^G*2-&xO4rg#bo5|F>
zUe~&?JRObhW}KLdBaD?;#X_M|Z_(5`0rV8R4$`!&^-Ztq6Vb<3QebD2sDcz?*8H~*
zBz#2C6)e-zQ$>p+ps%Xl^X@DwBWHy#xhu~BA#PBpFhEN)sQZC3{*mf!(aluq)eS<u
zkr42@aplMHZ<dS5b+<mD-7efq%NlCnf4anggVy<=Q!UD68xK7|st~T*mZZn6Bj1)v
zJ!)^3EX}l>%2T#_Ptl_7Di}$w74yFK6GA&WWB)C|cQaURdtej<9u<WpcI?3jYV7e@
z((w$ETrY@<B1K3nRC1s;bvF+tXEt2#9U0Gj2_zo&BsjF?7I=XmkvX_N3Gpt1%2kgY
z!fLSQs}vgMo8@2a6zx-UkqFU1(=pephl%F87S}X9blpUeATvi!Yn%VutMd738J-Au
zU$P7|h*^6G9&2_iXYiQTD7krBG*qJxwP01^rJ}7_8hV|lAD@0Tr27?OKK?CKwRz5z
zR=l3k!p-_J(r3AySo2Thfo{Q-H+w~JbzpE-&rRa31!W?mIOO2O18kkg2g6$YZSn!{
zuLTcD0rlaIFHbLO*@e?SF(Y;?8DgBS-QRd6$waIWyYtPRQoF+6$X!zdL1w{rqaew!
zT9gNbZw$Q4BoJ%ykbhVS*BtJ4ryTEENHRl~y|$*&Aw9&Wwn<=;2185EfA}_@j$6yh
z(yyX7;h<1@uO4&tdPs*weKNnU+xfJPce27ugbixvt$V7v5NVN(&g{ZhQkNOd7`C>z
zl=A2(5L{#`s4*lKE4#73hY~BE*F!AzNuJcYFfTGOI#ong(ChHBHVzzg!KA{bH6bZW
z^sKi*RvHoVfiHcEzgn<bDItM9nA|P>9@jt#Wr|2XFt=e6`y+46I@4}K<@;uVw5_hy
z(2ShmC{M&oAN$?9lp5XP_Mr~HY9~3%Z&@ST2>=vwdn!1mdNY<?pPsV9tCO6^$mA%U
zERMZd=C~tAQs?j|#D_DcF6t7%pX!GuO{leUolwCu5KMal*PwNIB*^BqBQV$8btwm#
zr9SD|8KokE)U}6@<6Ccy$4uV#Anv4YF}Aqa8v>L5<xq>>Ipi(797HR`0&sTA@UBdE
z&se)5OPto!iN9Zk)N;7VYo`ql^}FH9*9pxVx<<sEGiD;mJw$D-OAA?O{;XQSi~-DD
zHzQi73SxfQk`|j0!(a5TT~nZkNh|^UHT=F=v^}vP8q1MB@xtke2y=BjXMsUogGp5#
zJZb%clppo0!&cU7D&K4PJi=J{{Nnl|`G#}LPKA?!S6`d2T8NlJQGgx99B{<cM?ll$
zNw&}0`*B^_z5mQnzOOLW24`@%$fWkmi6qvopv~Q~N-r_OF+|N_)s~)>MySO1Lixsv
z1k~rj`*5BaVer>{J-zo$*>7Zl6h@<Fow-N48<D!ga6D`N?JvlK2@2pvY8G+Ho9b=7
z5)06OQXtmB#3?`085n*KQh@81LwiK)t839u7wAT-@VV1$zh;N7tTAHdDqP>{lrv5T
zCySy2X>XL%^)lmZgD-Z*Y8@0UgOWHb-<@_!j<t8Cgm6~(5!7KfWkyBK_iiRrV|JoE
zN;nPNJtrnL`xYefb{13A_;dDBa%rr+$IXg1wOvhmco@E%ViO8LaQ!SwdWbpnQ?!q$
z%8%K&fy$)9hU`A1ZHL^}s|o*&71`9Ji6S14WdZlE&9WY=J*g{VW`dPW1Fx)W@at;8
z!6aR!2?L%BA$z}Vk^EuW9cOAi8qD%K571<jn864!P5uxCJcyT^QsfaH0+UNGC>kS5
z*sJo#TWOY(IoJ&U80-4CQwwd>T(3x$uTW{44)V@y7A^QY;}iKZ5{Q#X8?K-(ORDfw
z?VH!N@m2Px1-lhUWqT5GK8k1TMTC10zhT#dmOZJDac99>I#^h3-++}+h%>5w_p5^S
zINc*Cc4hV*>yk!l9;dSHC5nWpvo13viP21cJHa^E049OFvo9xoL;4cb`DG)#LvKXj
zq-fX&`@bZb8iVT-N?Y3tB@ylD@jAk=lGLCh;E|`7_Wb9LNbQdf7SjZtq*IjVl!C!&
zB6)@?J2S9`(v<=UL?$P}H2VVMdMjyF2yOTC{x+5Wj=pwaE#VvM=(KkWIC5Lw@5?Ab
z8!J;S<H#g+AQA`#2tnQ*jm+NuRBlB8bN`EJ+mo@pq}ysGgv7)jvSr6rzJ@KJxQaWh
zd7nZs@ytAGm*$*t&b9g69inJX9TUw-hxuK|nOa$<dg!n?dZ+7);wb>GOQ;!U8Zrvq
zx?yHZtn#RxAnF(_oP>Wg2~eRdO^_>DB2vl-eRY!7)|rhqgLE4moK6F6Id}Nl81v+Y
zxBtRc`gjN5JK|795OAUMuq3A<t!lnmI~spyHAU3Mh?CjUn|x~|h4c=P^k);0n3M0u
zT~8OApJ75#VYf=lq+|r<=_E}O_II--w05oSLm^!C9W&*1^}{`P6Q?$bu231nOKox#
z0%LN2rT;ts2SKr1jK23{QM<oG5S-x8U>zm~{DtF$WWguTP?da_ufm=iL+>Pbr8%$W
zzP0QWFKrw*h8x+!&~IZVmCoz7KoM^6r%0qQXJBllJ%w>NT9JaE`DVE(3eP05^WzTW
znk#Uvs4UL2ep`C!*39p}Oz&486Opk3$!y&@SM^?KCqT{mAfD^b3+CwoT}ExShhq!{
z;4p&gkwb|?I$ix`Oc45U`4H!gqvll^@9*@0Jmjd`WILaQaC3vebD*>M!2M65fB?B-
z-UqstW?VIvPxZShYyu-Rw<w_HU>&IUttoKMBB=B~OoNSBM=FkQVe^9L+92Fna-ePE
zC2o$M{nOK!b$HbS(~1SG_$n@S!GQh{(<JJ{35pT#E^+p{(tW_CQ3vrMrGf=XGvCbF
zljst>;GZFL#sz>;ooDmeF-g>Tz9ui&;H0|4u;lHSBW%t@-Y7QOV4p#=+OwgsjHp4C
zbI>=^lcJL56@_?4+LT|Pl^>k1N?)~J@I7?+0iNjobS-H>pC6<hYDPdkA-$v;*k`aI
z0<~lmA5s_Qm>!FosyEWqDLRe8kI57;eB`<Ys;kR`G+0XnF~5K_^@b!=zXQm`f}&|Z
z&MC`iv{@^(g>oYmo%1-+#tKH@9-9$4%CJ^|sJ;fDWo$>aqlnO}n@9s!x`r2SxU2b+
zDykzI6Uw9-$!uiO7W$iax%2(AYaoxH3<PyLd{4VGDqDEFw59$GJ#!*}%KG#MTIGJ0
zGw_djV^6L0V^-648Qs%omTaddrg8x+8;(j;BT`8EKj4^9rqz^Ws<0`Y^ndpyG?rdk
zWe=sWtqs>v_eZ@t&57d8^#Vw8#y+aL1xX_1CMIJl<vO+3vHUQNAwc58M@T>nB!nl;
z{m>4p?ox2AOBv_gsj2Vg{|rHHn{%QY<?X4Jy4X*+fB{ckr#h3bFmUu_mT8?k;xzol
zj{dOH>jmSN{(Ku}6h&8y`OdEIrXb4Q9lm8JjqqT@RO{xKy1L4}+IW5wHbh~E;e?PF
zDQI6tVG>j7($U$^?6#;J>)@18MucPSb#N*-i@gQ-(BdN)<8uAgVt)Oh*Szk2vF}eP
zhw;q=hO%o-q{J^xt{b2ELUsK>qw)X5RStXe<%Y@RhZXCN`Lzd+FvTrTI~gLu<#S7l
zF~n}AAI`xa%V3O4sbLBan6bB7yU)R0p}C|vaN&`Ri?Rc$W&94=D62Zep*9KBaNU+j
z4OY(PA)?ZAgSl>Qob2|uf1Z9P_VAD^>GZvn7^E_7kaB#t^M~-2A}AN-;0z2R-Mi#y
z2@*bilgcB&nEa#wefST@d=*{~EFtw+yT1eA8%S+t2GEWsiQXx4##V9S!Kr5t+>6@}
zR5Y7CS;Bbeg?3thw|#4uYGZFlx0uWlZM20cg1uGLI(hRN-`PG7J7sMu=YSw%Pu=X~
zvv(_%^UI%JL%60;>-*yXgD6VZ3wl+TMlGdM#)}aBY5%pEW$i59(ZQ5<f4PvSV6-+C
z8?C8uH0C1vOE#r4c+-OP>IYjUoEQ?`nHeP28ZfvXH&v{?CTqtiluh6{Uh%tAdlfVO
z`|F-aP)iVbRX*~$SWMEjnj`f6u`KpH?%YM>LFjAbaKM?&4y3Xm57II@^N<H^hiQ^o
zzN_?Z6=+n?DRI<pj(=NUVk8&-OjPk8N_nf!N><j*rYmA~BSpm0(H-&48{c9uN6#%(
z8bww=S<IFFNU+3bB|eq?d?$!>6>tr9_<dgL+%(bCd)XNuY*YsA<4w(@NvPlU_zyq|
zh7rO(E4I6;{D(hRr}HG+yp4gixUOBxm<&k=u%6$6vPzMwNn)EOVCEnUWQQ9aWcAD3
zm<{7E{k{4vP(@=flbxKyUQu>A&S?p_VIfB;Srz%3DUUQaHwb-1)k|ShXOy5O;%w;W
z3eCsyIP~vR@KrF!>d%WNG9Gk-X=ZK+25AxI;UMy63*RY&wq6r<Jy9Sr5g|v!qk%@3
zV_=8zZc8aHdtlf~Tb^c<j8yuO$xQqFcK5S9^UQJ<e_wLv`x+bj2<LjA@@Rn+*gPt&
z-r5Hw?CACd<G1+Eb%1;{ZlT`8Pf%LNvvFCW{2ETqo1Acizm;b>4)L$5uR?xRJ#`zW
z4w5$=Z4Vz4X?GZ!8A&BtD|af<p`gI~-M@CMof={D+C(kcVfnDU#yu&iwb^M--Crgp
zeg<Hdrvvk<)en_;(r*5-)xY(?t!#S}l}1i%&Y1T-s}##p=|rg>fv5rbzx-jY=AU;S
zXq(@?#_mQ86md0|&e;%3#78E|o`#Xrk?z)kjq{8QS(nIU-)c9sQ%LHr?)Mn7f7I57
z5TKHtCh`ZM>h~_<#Bx=mj+6>nWiNM^x23JpKmCSy#rtDW^MxW5Fz0jWdXC0HA+9QT
z{2b4Bg6X%6vZ;asPU~C8Ctp6TL=}zWp>uz262`t-t#banx-5dUx69pTgknGd|58t4
zW!rlPG0%H1OVN+6CO+8EnrJDwh3>?bMsA|9uv2+3L<GAhsNYYgk5>HDkujkAXR1(?
zvpVYZ@T?&%Ck6-6{3EsL0C3+kEa-Ul@o#N<try1tHr+DTD;&*^&sDriEIR-B%lyp~
zn&GVyzz_d4EZv+=$1UDaEnexZbk=Z06r*6#7cS07RH#S8pCG3wt|2dIS~}^zBk(X!
zU)BoTim1WijyA%Z$m9{!02YYgZxE{?w;MAfYMXecP59T9D3DBX>rab(A6et?S2wL1
ze1kalLnPVc)yOUTHur1Cgf0L1GwW>i1glI(bRt>2A<U>4Ksf(Y>!JhpA51dhSO1E2
z0?L-=l3x9I3_hSuU%KM05XfZ7__ea3Y8iM5if>^$swANwR+N|(!$j{DMJSfVZ0|Rp
z?<5kCa2UDwqsDh-DeRfcKbz({)rB(#wVEYddbVubG%x?>VhGCwKjz$KoJ~Wk4lb+r
zm3r$myeu`+ehQrr&imvAh;MtNxQeI5TD(&xu*X!j4Qah6cXx1~$Ym#gpD$Nm$8w7O
zA#{bb#)`RKqUD?C-m<Pa#E5VvCW_=&!!h4%<VSC0$fx1%fSSAx9+p$xJ<pN?CoeW8
zA*b1&96oONuk(&q#uls+T<bWXHhQSCJpDp}y?-3CVU~fSq`$`H)23e>D|(VI5o_w4
zXPy6mS)H0GCT)QjPR=h*S<IL12cvdUNmR691r51qTLtaV9l^|s9GK_DQ~h!c53tz&
z#U9N3tacMOA;_SOs6MJyB)QP#GUr8JT-h(wPmtM-!pP)WoYnXVpMf469-P&vBz-h1
zH{Uzy4YO!#Z@WUeOi7)!K<-s9QgZzsgQQ3!&6$@k%gBRi!6K?d^WiGcSiy8JzuSqU
zU6VupBSTIyMiecge4~Y+80)BL$UZX?pXTjXJIC>kEZ+85m@W2rgjI253bQ#0gXUu@
zP^If_dvuKg`v`W93}Y{v6hw>BZZBbE!4~2oi@(Hy_>B12AmCCZ*uTyC*VWnaficy8
z7h?>gro;Id3_Lm1_$s!?1jE8N`S6#Qa^R8RM#$yF2+a9-Fb90*A&OKR;z~L(xAI`L
z3v9?6a?+WhoX2SF=DBNX^>ia!mH-9uIbwo=2BUu+2V90+@AQWt%8qA8WMUN|-K~>=
zg)am!ZGn(+Xaa|eI$@|UQB!RsNox<p8|9oHY_IU?xDxc!Wfw>d_331)c<3OfSh=*d
zo1Y04m=uC^KUfH=GaXrCkB~`{gU3pHIj_{9IDF{HV@UbfZNE{U^Hh0%{)6ed>AvC<
zJ03AR9}q^jajw_rClx=mkq^vBx4PEb2C(EQa;A6nXNM3)jTOqSZ=diB|9zHJuUR6C
zPP-BwIrXIX2)gOB`W!x|FROBfRd4V>r_%lHg+tC|+DfL4FXe%O{(f}#<Tb&qbVM=Y
zX@F>8HFIu!RRa;g^)SNRpiIscDtPq|DD($=JJ#FB>nQ=!A2G2Jypl*5s`%~F#Dk+K
zH!IaE13fSB{?vGjvkh}OJ*nYx|4MyAZtLb#>KLddS!yNj7Jc1Wpv-C8R?t*90FQ~2
zuXZZubX&H&942(Dk0vp-zj0O=60rE*81bTCf>^kp@G3FBYF5ucptMm23*XDn{7_p1
zg(Be@ZM{q>R(vJ@qua6OyIl}%RFv(={_g~!YJ<C?hUJB_jQ?1bG{N3);X*6u$5y@5
z4;{ZwKP#e$hFkK-U{`VmVnYm7&K)hq|DI>(O}~8Oi@AuuAzNo}GxGzDmlR#!de|}R
zhriw&VI@tR;v<UzQd)YhKAA$%kJM<`Z7$*9+-nxcQ0RuqNFOjQm+zj8Q!YLm$vfCc
z?6aQGzve?eCOWin5XwOfeeg@&3w8YXjq;@+H(XJOeehksi7Q{ti0JuzfcW3_2F|3p
zg<nEF$YN>gm3P=?qN#>KJiB}x)JcE^=hlxI+7t%5aNon25>VI`G>cWYM}&b&zhMF~
z3F-r)2w9>+Vr%kHNviTvwbajPoWq^DjhDA@dZo$G-RTphKBTm<zDun>NKfxf=(u{I
zi-)Iccfh&=u&WxmRG53>r`wY?5im8Xe^esP$a28wK|!t60wnYVa0(1uv4V*0qlA*>
zD>C>)O6wddYqz_;_$G?LJefxGIW}0&a2hLiRn}3}vd>LOnx~l&Mn3hh5}R7MJ>A@e
zaw31MH&e8UNCa9$ZR<gR1nJ-L%wqcru^ev~3^+6HWJy_U$x1x8whY^{d4%e5dRuSl
zJ`KBSFNiYNxwu!%ISwuRm2(O}YvihtAc&v$d%Ofsby#{HpVa|DbWDj4A#YKdn%H6=
zn~}N`ZR4<!2MKdmu;yMohn7{qJ@U3iOl0fn8;5jN+aZ_q4ET3l3Dw!*0I%ySRp9%U
zk3<)KV~4S<s?)KJ`z0jp$;>{>$YpN=JKDA~{zR!O<n<P?)AfjJTKc*NlTw5{(&+`;
zVXcG;Q$ZM0LPc)3Z|VqB8HLde!T&<mGpbQx0E^5U1v}r&Mn?7@&Vuso9WVSYl}1A&
zYL-56(oM|}gwslti$)Z6y5!Z~$(N7d@W|rHYXR^+pGKT3!1aW7s9;DEQ`R-D1hnz-
z)zdfCPr|Yc#@burCw~!TtYAU{|3No*C!q5nmO-XDN=(hNPRNDK2)Gfj0LZL&$S4*)
zW2axZ>L*HUCz#X1k@8v})tn9eFf^%s`m71Z0HY0iIWe|*e%LjYT(jjr8AG#ZIO~b5
zs78T+D-G@Wys@s<75%#~N1P{QNMO3!XVcWH2FVHRsJdkV-4(<Af1|`QJ}7y=)mVxs
zGeKAWF#X%&b5&@?!K91zQI}>%<p4q%33Od*Guj>fdpZ6v?pNf+qQ8*Yw}r90gUI#i
z6TL0abX;@lYk-7t^9o%r#%#oD*q`LA*wClwhnyy7UpkemMC5NUiUEH0KNVRaL%cTK
z6ws+=F4J`Yp(^R6&(PD+k))YY&5{di6;8c!B9h~2&KytR<6+^RHDbUx=2U!<Xxb;^
z`TLv@<vnEvqKwKzB??E~kbTbrPT^+U5#Xf4)!))^ZM|5<U}_@fBUni996igWfT0wu
zICjr@-9OO^S!EQ91pc=Cvn`9ew4~hfJ0|Jtuadp|X6ClXnMLJb&grdH1&ay_#X((l
zB(QxQ=->nq{Jubbvk~ZYlCU=Uo;iBrN6Cj`@@~l=fC2_;gH$h{b%pQ`+03(si+T7J
zEG3U@Y{jD)bslndCt>Dqv6BiR#2^Gu?#xF2S&i2r-O=?7V?Pp{>bMW)hh|eYwuvr&
zXf3>@pZS8okaSsE-u|=T=Qr%YRKEg4DDfh301hUUI1A5;x2o9@jbh`8@eM*k@yye>
zmpKHi_e#Zl88u2i%XIp>XnozqC7Ya~X~B%6T3`N@WZ@8XnveCyhGh_$$Zr*cNhqJq
zF5clas8A!7y<H>?`5~R&f4TDfy4Ug`=UyZhvH~@Z*cJkEA_{@_#t9g4hb@C*4Z3TR
zkk7G}Tk|E~M!$$F?r*S=t)?%nGdoLS@Ex8U47K+tXxvF8d)_YH2qO-dBZ3a?2W@1P
z(ou65if!1vO)Z0FI4(h#ONkX{;?Jw!r2k(nfU?Ns%ACo1gXJsV8}=(~$BO>o@6O$#
zy5$QR7->`^3r59;`pAz29200Ik8Wtl?xKiPnY5Z3|EDBMANr0c?US1PkKV(xk23|n
zbk<Q`R;fpM2`lS9iUq=qHxj=s=h3q@ZfZ-x5?>liPCVep^PiF$QPKrnS0gUns3JKg
z{cCR9%Vnp^3qN$=yqXF&#-w{7)oOel`UuYUD|j~;n&_2GI=Bhz%$W&pMa75vWH~f#
z(#$5MH8MD}?-dq>3;|)1H@P~p(MPvW9<<Wwwn9XMu$V@6tRVuSnMGqd!+%njLHei=
zeca8L1rH;Cse&Vkl6V7Sg?PgAExX~vwfKslxF#Yn&|WIDJMJ5z7sAe#A!Z7u501^z
z9%3)0bXX514MKVj*L7d@^1UC?+%*obz=JB+BJt47!>8lh57j+mAe~oAuDYjxhVqjS
z+FQ+49LXQY_Y9<N^rt-*mTGEq?5Ih<>2`qXL=X>}&?N-#H`QO2zcIWj9I=nc>m6mZ
zeh;y$Y-hmB0-iKQJ;yWB$_neq{Whrgfthoa+g3kFB}u?9+v$aZKmk`aeLi7mUpc=4
z=si+OcA*@DX!!%pqFOQyCP6a(KIBro`xr7WlvLQp4@nM^Ku!Kll>t!KGU%Je-Q!=E
zs_A009TpV<Aqb+WW=%6_{Jx%B>Gqh;xYkP*Q0;Hk+1Wy)#tM?->6&OeLRS)tr?6p$
zp%h^qRP>%`G>iH@db7+ogmkDKpP`vJqPMq+D#?u>KnZ+7s7fpANKUO+-dcGJO$7O@
zFv7jd<Vk<A`upvjgrkV4I>Jg=O)Gibn<%8k1ZPaTTf?g?D0{~F(;$P*C>R_#iO_?C
z@$6-d<e}R}dW$wN3}sP@q>yIw&W{Jf54Q68Mu-ZsBe>Rl$|FANXyrP{oX^gSGyzVs
z^em}!VFDzO%0M#A@)K34jCcJO06alAdXWUBw1X?<ecrX?<4v7PYZJ#u|DI=(teT6Q
zi|emJ$3czbPu=!LQkozmHf)o`qSlR?2u>SZpQq#;Bsg=)`6aEGQ9mpvsdRpjmIZuE
zf-%C<-p0t@MyULiDnm4or&3iI36h{tc)#2)cN}Z_?Ez@nemYo8Il9^GERD!f&;O89
zeZaK&%%u<=hI#>+^Zj|*ohb7nDU0+-h8z#_KkA^bnCltQd(C?YOhkMCAY0JyPu@s7
zu1gUeh=6gkVz_esWRwUyvF=+9EC6ZPnw%GYNHet_h}{m#5@J+(y82Xb#1zWgyOr9(
za(jD>`Z7-io*8LvquxmdQg=S{t#KLNoUe<Z-L5Nzf(zG_R(~L9;X4J-TvEw}*UsI+
zgv?n3@94t3+{tI-%RId=j#$MqwcsfwS4nOC$CJlr0rDh(N~rWVRl%D{dYv_kdd~@|
zrT&uQ;Z;^<;g0YtWD`o3?_JP-bNv~dAs-<5(E(LLe}h61*U%(#Yjwoi7d*BlmHMCT
zP>=LK_Z02H-#`mOZ%#%DV6(e2fs~MUJ}&Opw_^03G8P_&8Yq{sWf<8?fopA?-<=B-
z%z4ouK~5+Gjw&H4AFrhxZ#;&43*sE#ySWZBkL&mW6PrDB3mINnWpw9-4Gwq*;FbNr
zK4qVoKmHeN9Dd*E#-?X1<(k}9R!gJgZwp%~f^2vr)OfW#O3V%f0hMQ}XI3$7Q%`l^
ze%*UaoBLnJKBs8-pAf?v+O)>x!kbvw2JTtwgKZ%}awn*gH|ax_t27a`+dx{rq#?#l
zu7yr|<hF$unU@Qe(Xs68`9LHL7O@aVl9o-6qV)M&$9RsP?Fsw#=KY?!9^n;$XEQ~M
zAe+ADl8jY^-Hkw4qG$uwIUkt`Xv=Zvqc6%wAfv}G2qt$d`cOo!+E(w8`T8|0xpPG6
zhgfF(&i!6m=H(vfu_cG=G|uVZl1KdcAfjSTc4$WVtzqMd1sf`Dywv=P2rG{>fpgKG
z5K2HzeY4QJB|B-E4c_58T7rqX$c-#Be`=p|*h(TFV}YLCFfIsT9=q(gn}vSCOy(2Z
zc4~5yIe#<f3!V7-V3?_#M_q-x>H7^oQIHc{?hn65QOAe7H`c=^(?dU(aVE{J5-1he
z?pMiL|Cl@Qqb@Kc>nkr?2=|c@A&PUih~rau$?-IIVaywZe+`%o=Ja3ZW&A{wEh+)E
zlutzE;t?KNe$iWA4w6KCq(7OUsCH7O!bWz)wXHC2eqaB0YVD;_04F^`-e0(v33+L?
zI2fj(U$S}p*igt8922gXH@`5uX`3I$F<TP<s0*G+r|&rycz;0z%3XDZWiaUQ=wD4C
zwhmRY(^tZ4x73wt#MZ|QE+I#e)!Q6CT?r#wzC~Df#{=JLUv=js2)Z)QUQ(P%Lhw=L
z3lM}A(ZmsswAwIn3uGJ}1UVi2=nPh7KAE&^R2$ck4Cgp!3GL1(Wt<fOf)JYVIUIs6
zC0IMMU<E^B5}{o487@LC)$j(pP6;^?byMUw2omw8Nm@|-42PXmCaBQ#fFw1(NxF7f
z;my+9o6g$j`;W(_G}6SVvaUJCu6A2d`@hH$E}eCpkn2ITpoOAVxQ~yj^s&F`Q<rr1
z_;~jVQ<Qs)eEayQY_#6H#EYW{q;(kLuP_lVNqk=Fuh^k)5`00f5l)y#{J!-`_~_08
zpT9TtIi1R<7wKvC#8*_EJ$Ks#zz2JW>2>Ulc08`7(|cXA#t8$&b3j#&>|msW4|xD3
z?&U=NkkCVPVA`=0iBj?gY8IumR_i;r_rW}YKb+*5e8^q(NplQth^6xi>oA>pln)JJ
z73)EW*`h&(uBUgbuW_o**WSq<TG&Q3992R6G~8SETE2vL30c#1K9pDVTO@3KLljpQ
zhi6LCn#PSk*v7j1SPVUoN?v+#aWB53UH9;0ZL1z)5|or8ea{<(>Rxmbrq`e@#3a=$
zHLn$<?IlV+Ffa1ZZIxufle^EcFFb=YxfD>WN#)E3H_&=`F56H~WT+VZRARc}zb!-u
zG~OQUdVlP8b@~hKrH?UgjV9?qutbR!Di??aYRM{zSCZYn!O}kZA@cKtq|fjuOoL#1
zUHa%&wD?2&n^!zqJZXX`*?df&mhcLOYl4Rb?j=rP8dC@hX4a&|Mci~sKSu!;DFy#}
zprgLtmikiR-Z-am`j!{LS3p$(0kfgOrhJs|`=KvVm8<NaO3jjsNG~}q@9`LtjO0h+
zfZugrF(q7D8v6MNR=uyKsk2c9(&pTCJV;Kzvv8u&^HPOJIU$0(AGY&FNL$0sDQcK!
zO8)LP=%E?pcw;$+aE(#HkONRCk>;YjZh436@zB2`b0hCyLud*n3BlVys_>1Evzmmn
zk)uX-mTHk8(6jO8Gc)O_{mT*plf@SX{0l1GrR@s5_rMo!h@llj)({q0JLiLvHYOC=
z>;1j^d%igd(zeoJ>9;vUtiy+%wJYPisb>bk8?JuzEvM&Yl1XVSx*^Y!v!9uwA1RVX
zr5b~fV_4EZMf|fPk24*3N*W08DxYQ^C91t8yBDK<YZq^a@tSoYGB2F?YlmJPXspt=
zL|U!k0H4297aIeE*xBW^ktT8SRjzi2``KW#FM%Yx3q<t9^u1<flbex4_ptmCGDE3P
z|1^jQ6ct?ziHb+kmx%wOfMtve7gyKF+L{QH4$BMrj*Ez&_CwJGvKY=WaVgy=t<Db6
zwS>168C85SZ&hk;^Exg@p*atUQ+>-t0{@K&p{FB1tiM<ibB>s*D+&3sxkBH?HRIV(
zWw_+MF#a%;JcE#MtmcqNF4^XNbERl>$0+H;fw|v<g)s|khHc3!Qz`KiUl_%o&!v2G
z<a=A?%cinHol)*U*f&**mbvz`Z~~UyJHJOf<<3ro6`6Q)DrUO}6ZfOFkE6VC)-<mx
z0Z`j;&v=TeJ>l+|{>b}qf*=vOFZu9wewQD3*QVmd8-wZ~*OdAXd1mIvoXiyW^|vJC
z%%w5YKiyT<mljrRR_1?XSSp9_9A~gj>sRiCzWRbKWq+CZCKhc@HA&CTo&QaHUnabc
zZvD~&x~q`@A#9t}e|iqB?wOT!{0`kS?LfBpC<OL8<Uugv(!!m`It*Ev@Ka3TLC;qZ
zY4$klm#6$}aD3+0{bl^k{2KLlHitre*M)DmK8SlK$)7jsif*!*Zq39n*FWQ#Kx&X;
z-<So6)VtSV`cC3I1S<yPgoTuSL83*Rz&;JSYmKS!{pXu1tQ%g!Imo}8cgvl+xvQo(
zodoT-Hdh-fc0hG7jtrj}K8Pz^yjPiu*QSLK=G)LlR)5XGzw?lvH?o@P+dywGevZaQ
z4SOy7)t|NS9sWT{o*U{&w}9&1UCnX1{JI|$+(9CVQj2lgRd`-zZmb@5(K+*vaJiql
zS5tb+u2i7Kp59YBIqDzAb-dJv#GUMs>t>?-H&1zzs`ZUZUrXW6bVBQ`=#SU|#}Clv
z-+Hg!?b_qIX+&?^GeVkFEBVlW^J~ydr2+!u;5D~J2(BFY?gQl|Gwkn=d1mQHoKC~j
z+Ft3d+w+G+L@p`P>#Lz`kx~K3Z{EL-TPTX4WYtGy0=(Jj2q82J2#g59h<n(G<l*2*
zx2@x6XcmgSBf*s*xezw@FKyEne7y%kR;><&v%i4c8LVlK)z9}>3G`3KV_;(TQMW-4
znWiz(;_Ox3K259YG$iZ$&;ZoE_1{*U66-yfr|XK8sIDHQK$fP`0c`~=FoS^Phwn~|
z_0K>DD4-(bAc@9VIPJoPkP+-nTmNNyV~(P#m+>-)EEFve-4(Y1sbC;|$kcwK4Mt@&
zz(W@O3tQ>av226jxl%$AYlkSwtrqHCzK3`9N2Fbn{+6kSXG;m2zYjg}BMU&+AGOj)
z5h63%)%oS;o8ZdPHu*wKSBm5lduR_rH5V{Wz@lKXV#PpO15-23)s1M=m`Mpa(+ZSS
z{h;rxJ8*Egoa@%Yq{`%}0J8HV+r#?jC+m?ChyA#C*Sk(BTv9J@T0<wLC6wF%o!>t%
zJT&(V65>zUFqldDa)Tg=HJ+#`K*0PJ?n85de|8kgyQCCXt%g4JBX_x`0_!6u6&xnW
zC0t8lB)1NuDd~ENsqVSaC=Q<H6uz<TGhBSaE8<kpI`TvfBz9f3c@*{HuDovj)<=Y7
zhdqx01b<ou!~gh;4KblCKqrbbkqk&mM)%7XgB$Y&?4IJ~NzF-L{8+P}G;uFM)P1^O
z4Rs*tU3a`Sb0`X#hwSkdN!tbU@-NX4(NR^ZOxqaA7&d+>p{aNDI!7BnwpKL`&wmvD
zD8xSs7f1()m+^8B&D`2Z_qO${rzJ{}&iIx4AJy4OK{SQT<2RL9hM&3QR|{@rk2;MU
z4>D6asu~p0`-IGf#Mgi)go;PMtC}Xdre>mDhTmZ)-DA_wlGd%AsveubN4viaL_N}Q
zLNy%{__I<pFJ@-+wiSrX;u)k=?B3%tI2bmHwY=!FYM#Dhbh_iCkkQ7~`Cidf8@8b<
z%7FP5vp(07F4VU$jr1Z%d+<4(Vfv!29D}jqyD-YYw|#t;oG61%G5qz|%k<%UJz1o;
zg~}ORLp42Ijj!C}e(z>z5Lw%1jt!xDQ4-3X=VgF~jO{zn?uW>)3V(i6$OaqTdx+eB
z2yE}YQoJrmjG0Jz_RokQ`2F5;ej0tR?>;H{`-z?sk!aDz^9wJ!NziQa0-7zJNLXVr
zkK%Cyc{^EwO!x!A$d|Xp5=gg}TlL(js=l-q(S(hQAF;V-wS&Y-y-QRpVh`jK<%FtE
ziYCu>ihKc@q?bK3p(5@57zrr9DComIy><(R;vd{YI{ffrJf6@X!Lv!_>8}-fAG}Uv
z;L3mU#om6H!1VQ_lRn`}Z_aY!N(QSpvoS<tN<=*0giZT=Yd&2Wm(!FDRnRz<WY=@7
z3<yv|643H)<v8bS$~wW!jd5g55v%f(MYuZx`lCZm7r^Iss_xc6{>7iEy}}s642qIz
z9$tRgD=<oEl_+fzyia;wo!kzK+3s&pz^p{nV><o0O(i26-OZ&g91Ztbh}0+En0Ld)
z6FS<O+IaMUl)}C;uPDN06&-+-X2y=V&%QXtt#*14k!(4tA^YVeZ2x$wET#y-n#+~L
z*v`aV5y-DQ_*w%LdHfB53nQ!Bq0D!bM$yuRs2=MD;gA+C#FRm77@qCk?U?L`B@G!V
zb{iy4a|=N;d&@7imrT-moKo_mY-#9<3MjTUzeyPdAf?AVZB#Was(}`ei;=X&8oI#>
z9R4jh5R$nIEcWxYJQ-rPlMNTM)sH>ri7m=f-5@LJo8IY#75~C7d~+JWzzyoPk|}#?
zsj!%;D912;qZ}@VOhpavEM!ixsCFR?dwc(@|HwaEyY&H3d0~)t)dFC$T$2f1V^9n4
zRz~W2L4ExE;R)eh%(wV(Ykv3QAZ+N)Kr1qH?u&laQr|uTlVp9$Dv+CrUw1e||HyFq
z#oA2!ryvepdEou(*pEE%{&UD55_>u8U{7oOaoI?alvaeSF+998aKv4+Aa28saFw_l
zTbQf-NsIj8!10v$PpIwpx;OZgk`Wwj=tBgf%95|ZaguIzZMB56X2KzaJ-1)3fRXzA
zg1_Q{Qp44p^2JXU4x5jA&otiqR~v^qJ8Z)O(RFZj0i|xv(z#;9CkhZ*#BpMPVVRk^
zHoD|Z80AmJ{q=166v=vOp_Cm-myAD(^}yHOmO}I$R?~aq@I(x($~F#msX!u7<prK*
zqIAe9K=s0WbS)}Xa-_0G63^vGdh7d{nzd!;gjv81!-0hkBVkq|*W%NY2lIz_zzYD8
zo4tT4<XZiM(qZ&m4umGw8v2d|B|w!eBz9_?50HY`No%5X?Pvb|$9lL(v9PhtdP7vR
z+%JMc7Va<5-xcQ%ewmC^#hNS|7DH*vLPJ1yx0(eyh#D^*8aTM)e!X~Ak1RO(?Jr*S
zI*X@}>ffc>%-Dsk!_u6OKztV=$@P0y!PcA;T6K$2pWt_&+J0O9G91KA63AK_)6$+`
zLhu%AbUUi5P@79)&CgvS%E$Rmr)ZfX-u$hC;ezLVOkG6ii6wYYT9%-&BI@On`PpjH
zQ?if$Hh6QyFnFc9DkLM6xoM6ncJa5}p&j?i>Wcrl2@hL^ilsRt=?g=kJqy_WpQb6D
zs0#=E&jn5UW2Ko^<9a7uH!}~KR`q23VkP|{FNIP)r)0h0zMEGJ>r`?1oAauGj@m4J
zvEatxwI*6Zip#g7PwDwXYSU53rmzl5!1kTl13hH)w(#=-z8!!f24HyN$sVJWhgXR?
z6j_9|-=z3Ql(Ox$!o&4wMAwWoj89Y5+AURaNZ*AJFX?b2kr4PuYKwmr0Ph79V8!Uu
zd^sTCBMade5`r@2%2@>+ZErA_$Wq4sM1|aq%4_S{z5V*^ae|1673fg$(8cB-Hqu6i
zeq8GNe()tNEp_BqVUA7B&R6J9R7Exq$dr|Kcxe<BtUk;WELu#@CjAKNM1r{OgaZvT
z5B}6LQ181D?;r9C1=&+Cv~i8x222X6ftcDje$yqLq4fR%IJJqsFjnH%;Gv;r+MODL
z?W3E7fQGqu*$YSk*&U9OFG89!=H#!`__vvRzEqwyR3$A>dGlMl*=q$n1jc!rnTr!*
z^Q4nxQ5`86_|{OUa`P(b43>utcNz_IsgLM=I}Nhq=(?2QsXsr#ySLa(34?L0xJSyt
zG;c*HA*dP1hUQC8s{ps3!YGhxFNW3iQKmWpI?@yF4>~Rwdxy~+((&~1Me%|G6IMdR
zCm1GW(kgy*8&r{8Q<K`X?e)9eONx$cj^hsSolL2w0YMC<=s*Pu7OW_uly@Lnh0l@U
z;gg_vzU>r*n-A|&_ukAj-<9^xoc|)6R%@j6Jqf}^CZ<4%ES%h%e;j8s9hd27>n)Rd
z&Qsj+v`4t3jz69lMxLW>8z|t&c{C|qe(`Xp58>^IO~T9-O(j(#q`N^?d-Tot*jodo
ze7wskx;5l|;=nAvT5jv&4N{jvhi*RB6=iPwDuy=b_t_;X4Ch|?uOGYzDV_A#3!5Ob
z4Xa$o%Y5>)G1SOCeJd-{<)ImO8&UXfvd$s_vkXU|aPzlQ=ucdHJ8tC`fEw3ABJ^>i
zA9@h?*Kwqo)lmS{%iGc|F}LPr)-ppQpAH7~zBdg<*PN6+PYDO;Q}DH~1_enqQ8D@C
z4HsRR!gBDVZ!jpt^3!_7qN_V?hJKEJa%HR&&sVN;Nqyn64a|<}5iPp88YHk#qAb4M
zKlU905m^QBX}1)w13kX^5UegT6aM&A$Pc{4`g^0XvJ0cP>dRcIKLL^Mp<Oa{A%D^{
zj^BteNZ4>s@gHAOwUIh@@&a?z0UEG9?Ln>igB(P5`cVdh{)f1(=W*$DiW}!1aHN!v
zd9A+?2B>?KZJ&s*h^SQ9NO??BQ^x7mow5RMzJAt{#B948xe9%g<0pnu<<3MLx}KT%
zug&NT$@rwH%0(-Bw~#dgpcd}TcUlb9`jZ2Eu_|I5;^oh|(Jt9xFa0d7f@hLmAwH-5
z>NugBW~7%(2{NBj$=z|sqhM`m@2K&o<kLF1H=CdA1lE6b^a`M$S_)1oKyGTF-~HD|
zzE#^V2;ienr0m3%`ksmm1>}bOj_d9cIdH&PQPIV1&1@Wa9_?KZHEmil>nId(irnzm
zI_T<-`VEGVzzN_(uc-J2kMYl{Chi^msgXu0nBgBqDnOW{fs;P^bzhww)PShD-L6Dt
zzrm<){!}AnMgyxHdRBtt4k=T{^R|y_G)&i#P)QOGoFt!jTZw^xC|K4Yv`zJ}vwyJ!
z9Wf-$b)tiADT7%qxLWkPtO_-Lza4cQW}qPSpqEOlctjHBQyjB5ADVPWc#{0^HD@QW
z4vSg}a+1DcUv=zhIs{bZZ5fuz{}=3_IxnwwloK6{l9engx9N*bqs6~y?ZL^eF8?^r
z|6(3M1fz;1pbsl|w+cIfp+JsM;@s`nqP^ZDJ||cj{>jrdsC30hzIt3;2ol{VY-RO2
zv>Y-ujESm~gIq=l5IQIa=l8bawFQ%IThP`PcL1U@I_wC85W6YaI#9ic7=at%jAbCh
z4M$eZvxI0xg3wzO-xp%%o1=!_g(<mCS{u6;jtQBgY9oYWUYo;XFzI9Ce0MZmhF|+0
z^IW7mjVq0r5^7$Zlv2XDrtZIIjbz2eEzQBNy07)^e6Kkf%O=Ijz1K1La<#JzyJja!
z0N@}K9EcD^VOXdWuFaae8YJTCf|`us*h8}~l-q&zhyd|Yie9)FqSnV15C`m$650$Y
zs4X{;CKI+w-FL7)i179D&j#lqL4>?HTu*SM&|?C3-NeWLT*8As4IZfy2+%YHU_$(4
z@yTywc$N?~2`pXlmL{1TuJyb!CB98Ue=B?|lKzD0I_M!kPU@*3v!mOGuNl!?59lb8
zmPL)+=#!DvxYPSu+JT1s<Bj<9Mi#7S2&DUkO+e2g03|K?2{9>1YK~Cp@$Iw#jm{kr
z7w<6Nyj)F#MEo=E7x1jOhJ7P^s}&uRhe~=VN+mo64+@+AW83O?OVxHaoRITrXQd`0
z^5X<u8hlL7r5hE~oKv*ubR<R5Yq5;*1T;_8h>n1Q{kIKinEjcu*S+W(&1UHjDE0K#
z=GoEEDYKIIzUaK_o4+{v9hXp=KQ+>02eIBScBp7G4JhtyA*N&Fo6$_P(iGuk?vgMu
zv?Q0;8c_asXb}CSTgstT!*JfJ2aP;B1c>874%ahzQD(0Qk?Kp8mLla4m%^OX3&;0+
zL203k7R}s~u;dKN>Aoq;G&*Unb;tDgApP+BjS#LiYD5DAk~{KhyHjG^Q_0}!?s0t0
zyJUp6n!F^+8)zx{gk>M+8oOq20PjfDduO$!Gl6{1*8^ZX?*~{b^|x3a9Hl=^T@NXb
z9glACXjfjta>q(#`~KM63IDM?M7uN$pb?hrEofM0(8{UQg8J|OraJ*88~3wK&qLst
zn%6hlFQLp#b{!fsW1k53@*@Z^C_j)(!Ybfza6wh<h1JJBs<r+q+h@<jOrnD4>kF@S
zUS<_zU$;4%51VbGz__f78Dp2LG45aWrrXx!m4Zv!|8cs*uce8psL^+d>&0bV#w8<G
z4)GE>Y2zdsf>13&89>?bf;GsVf6qHUt<pMaZ9c!!{Ro0q^erd6Y;yf{Kt<k6NY!4T
z`5Pt}2?8QV6Oj)3g3xm~zn=&JB=qk3<ZfZYkuqY#dlmCqJc{d5q2kg0V#hbMaqdaX
zq@D#*%OgC=#yR*U!phN0;nPZiW<7^`Anfa=rg}*{hsB|Ee|CiLF5oSKBmb?lV(i)j
z!XPCT{ysP2Pa@iT^q4j#VoY@4sXQLf4ri~I#=j7bij_FVH(jhGh@_z24_bepTuWX$
zg17a9pagIX_aBM!Br7J27q@Jnt8TU21D+YryVqmSe;u7!;99+b%%t^2mO8Eewu*}E
zLDxt$GO~YnHWQ9Qlfhw0j*X44DZCxcaxTE|!IRRK!neuk{g>Ov8IuKQ6&i#(GwLsk
zc8byUnea=Ond3Tx;yw>u!_j)VmW$k-HlGW-vIfUjZ(n^*kgB4a`k%A1{U$ZYwVCVW
zXuchK#T6xV6f2GEijIhK6+z9+j)><paXOx*FNzLmfWGidov3(T>bs)IFfZ|^ihBM;
zLy+sin+8-^SA9IJ<`zVFaLE5+INF3ncCgex?)E^#?BIm#T8zT53JX9cF-t)skO@M?
zUqad2&OT5+_0~X#<Hin67Mu#6lA3>>tXV-W59z}xiT-9~aNgjyy=cM8^*|1v%nJXw
z$Gy0u?E<V^HQf98Wgj2yfx%&~N*cW^fi*kPTIx>(VD)DQ>xDOSav;FKG>vmH2YXhs
z<RlpYAVm1@e*sor4?V)#BZhrpJYKr^e;Lnu4j%+6(NCFYP`=#&Gn)7vw;5ZV&T6w^
z*`+W{6X|$7iIA054jS2cBPr889SyPXTH79FD3BQa;v8u!<u5jt=Mm+12Qn>(j%=<L
z;tVf_+C2|p(B`$6C~Lw$HT1T98CN<(_ltPr{nq#^7~WJKrb}dPs!NXl*T{80vh_xN
zGeoJPwMna5ZIwif*s8Q@kDwY`?Gby_u4s$C_DW)P5Vc!d&3u)hMy=XHtJI89#LOG-
zU+~^v?{m*R>z;G&bI$YWXVkTb_q|iH<L*oh5w+Kz?gVSZhBSf>Rk6QQ>)v7C78XUp
z*L5PnS`xL`M4B2@^=$Tse%}WS=pLJVYK{wsx_=<daMP{4=449C!IASWp9-IWi7f6{
zn?*xrYWlyd=P14cp}ODA&$2`~Q_r0mBeYzTVR<C>V2`y)Wj*9x#3!8G^%%Za{{ot_
zOFCLzA;-G3TbD!6i&K@(ulSjzFV3;QyHq2vdZpmOT3<Z_%y^1Y701k_{Kt0`Zh0(%
zWXq48ZaLJP3W_eCtDer$`7bZ}eus37)$-R)>q!k`Wis5=Iz2U7Im(j<{(NN5$ZLlZ
z>l`qW6z3|jW)K|-qcE4Z*pZkjWqUQTp*CRDC6%B<#(UaJ+!;G_Ne+H|OoE`XN5}CS
zN$)WD0xgCRqpwZ>owX**#HSh}$~o0Bv!JWWFz>UMSpI(+q`F5Uo@&G8*W-=7O?<aU
zpFB=WIf0tW1sVQ`E-=yQq^4s=odwUh`&+hjMCPu#Cc^u2Qjt1+v51JGTyBprbib?9
zF05*VcQyFr!^(!|b^^WvXN-BPxAWl@0K~iDw5@_VI+q@b!tsUAV>xF0zEdYOzSV03
zEcyCnn&Jic`ynBn2)Rfoh5sqPo>a(GVQ<bxm29uAi>Dt7!s|Wom!~mdV%^*CtrYon
ziGP14jwNkmp@7(+w-nic-vqc=VBeMOx4h;b<>#P|yvvU+v_q2{U~Xf#E$g;<P_6p=
zT_70QH;G0}{za=D)-vbH0tHjuFkn4;Y!Fhz7U|5%G6c;`$#dPt^&+hE4pn|=Z~dX;
z%ynHOy}@@GME@>_-mOuriujUg#o~f(2&*o^E(?pbQ|cYmLo~ll)4~r9RQJ|c8?Maz
za-Ieyu2`-9{C!odBbFh`;q#kx+lSG65?*TKv$8^4ORyO`Llhy8CK=W^4g72_;jL@4
z(!$Rm-)ma)7B4}472El9q(rbNub>^3^RTm(d*X5XSoiQ`sm%RHn&m=}Nn!@jK)B#H
zgFHdhzXwE|TF^iOoq;~=4U?)p91F^O7SE+8_SjGjZ27X2!%Hmf?ms4a*PNp&KC-HM
z`hB;&F9^G~rj~ybuZ`q8@J$2d-00w`$b`N>h{j)`M9q?5O*w2yH8GG_JOow1FX%Nq
zmuNAQQpc>v*DhrcGchKAUS<#t1hy>=<zR1glnb1;6q$*YA-*$<b&&YNCxF|bsGE9~
zJ9-UtM42u+B4mHYl8E5t33%9J`SK}dE&EWmX`d^XCdz61ZrK+1sFBaE>X0^xHvRjF
znQqs>)N>|unzAi){OGZ!41-SVC8m%;<<T{VRc~_TXhlfnMqS~wN{SuXj^+O25u~KP
zIkwTM3CL-T>YGTL5)Hl~UVX>%61JyaMQ1G>kIiau^kr&th>WUWD+%KAe&7^q>#^LV
z`-&JHdR=>T=daQ~neDE3_P_R^gaet@Z19Iw&!>#5+&4w|={6T%Z@5+x_d*9}QkzYk
zb{GQbn;<o-fWlkf1x;R&EEKtYZ?S&O`R7z{y&^d;Zg%>HV(iHs;T`nYT-!^!<2hsP
z=>tXrx5|`J;ueeAgu}>4)gtCl>RJRh3LCZswMVQE8Ql+#3eaejcjYGh>-WWvb&{Zk
z_!b${`5uuA2Gpa~=-Zs#;-`+{{yzRGe_rs3Ziz1de=V`UpVQvEMLduWZrjEYSDSz1
zR)^)VooD#Ta^Vt2Y+_1s*#L^&Grmvi=U*QLZ{He;p=cQkZCQTO2#67gHazxFv+Ei#
zp5Y>@@9ABB_}{WLYI?9LHoh?`=AO4C>#Ac*TpTk^y6LYE8^6s2+jH<!fP&BN`gW=U
z%0n3zQ{x3E+Ni@bmlp4^zKmD%BN+@YW->C4+6Z(mM}~6TA#c<U-vFuvtXv>tq9Mqw
zldToszUTTTIbQcOzbSi<h$QV;EM(n&deZYVqR$H@>S$lO*h;jc7tfOPa;$D^M@ppJ
ze56?`HZ9H-Os{i?m4O-KOk=LH{pOvV;ZtM1j7JxV9fbQjVGW%D!MfNRG$sE63W14A
z+7gu?{BQ5K#xu%uW@6`7*KM4XA)hSwz)xnjU4BPV)B}5{N3>_Ty6z2GjK7%SzddE*
zEmmFF8~|D<o3C^BU7p;+Z9$2evd!(Rr5~wbHl+3vZK&Qx*eCW`9kk>ztfDdPIpbR0
zO7R^<uP)Lb^&)#E$NF1B@CSbaKzk8f#s?u!68#^_(k4Ad)DP!>88}!gi+^`;bRuE`
z%17F)p2w`Ev1?B4fMr%%ycpAU&$R;l`C}(onG*52Xr+Ve3&o%G$6wtLdU=Io_{I+;
z;y$?B*1h@zXZeG(?MO497FxBcc`Gh$=@?JwqVLrfm}&N@@mS&?@F$^B<4?l-Jf_a#
z^a*M)%n&{Sg^P~jFjUtI5Ly7FZHp@(F2vv?T(^h7eSO%kZ^QVuEdy?bFs7cP7zs!x
zkv4n>qk5v#O%V|UUHe~TpO@0C9p+kwFGi=r=3X2<w;LT;%I#n|RL>{3;P+j4UDy0=
z2^)KQx@3|{c09eX+sJ0CbLlLvtKYMoiO_KqI*-9H()vz&FD|W}$UAHA7FKCu=ME3d
zA_x@7g#0&BsdDsu0tW19_i$gh!ol%H@v}^KhH_iKOxdBE_-7-1Glr?Y6cM^yZ=W<b
zz~hI}PcaNlB!TeCJ1Mr?h%hX}4{80~yn)l52K{RyDyNOjximR>b+WvUoN8*D-~9a8
zCfX$-*`?^UAT4a{VpMfsIdzVx4e-?BQnRC$CUrX3fOT({MY*h7ih}%)MUUBssljvk
zc*sSsD6se{hX>dCQ7hL%s6m`Jzo4TJnnZdInFDUGON3VuWyw~$`J?8LanoeK-R4o#
z3KMb@`aWr1;%Jr%E`$KWnNt7Y<SmEKH*-Zjp1mZYXUD~%4DyYc9tuCkI=J97rBB&@
zN+nxktqvv-8qrRjW3djFwiIT#`KcAWo|@=aH+OG}^gY&?8<yto9y0#0fD}jfE>eG(
zjhag6(pz4IrqSkeI8r;50TOSsjTo7qWbM+q%N!?^B#0_eKH7h8$z4LK7vP1Tbc81a
zTic{4RJt6yXnw;cz7~0~;AioS?oddS?<n3a&QMrQ(?6o-39}>y+>D|`_Ft`~=!yRv
zOAla~dK@Po`c77|*aW-hAU<fz{rbI|2iI_ainsIkQN8K*<i(;lPJxNk@54}rpEgLA
zt@U<C;**3OK2hD02S#E0x_9f>y0J_5v?F#t_Lz9+N40AVn<`z2SB)2yR!K3-6lI*h
zmDe@11u;vrN}H0o%EsByQrj&`XySZUo1N9zPepHHz(LEH(#Ywy-*lC4n3!=ouWjts
znq_wj+54c2($MO-!4VhCx7Gn`L)-ML%XCrUb%dd>Xh#TBPTwz^20b_1u>aC!dB`if
z&HV=-{^ytm?nZ>^X9`o1l{SI!0aCVfGG&25zMDnymJ;w!4W3?)jb}Nhxd_0n=r-py
zz^l3hL^(S5xJ1kxo(02D`uZwcHo<9DRRgNA<{jPuF#%=wQt`-n8ASEp$$WsV9QlR6
z>ILXj+QQHAU}!RIr+f>F^7l|)dCf}OwmcI_Vjv5^C;l|&(GVNVSVe-@cV57oPa8Pb
zM`cW;F>aiCosl-Wvj+z;cvT#&379-2bbF%p@8nHAY|m*w6(uC@@yil>_1!}ll>oFr
zJWVMm4j*O^6MR&!l}DrIG_slJg5`w~3fINgT%nT-L%1!7x22-ABM-6QiEWU-xck@f
zANR6oQ3Lul@NP6>H5tP8u5>M7Gj^_bOsKeXKrlYeqRehvYfsT8Sb?AbJjZqKWGm2~
z&w~Fo=<Rg8#L4Mfe$AAbSnvikV~5C)u6;?l_*~-M>m8A}O0m{ca)h2n^x$0**WK`;
zV2lpEErk4X{EnftS=YCby12L`I=IAE1%^P7r<H_GHZ70?^=lrgt_)e*%?D127Oid;
z%1r<l`Qd^aj&MqA@Yzb9Ag~RIVTif=zPhMRW27Vgi~-I!pJWg@@l5>%;%8q?40V@{
zgM;ldeE9&~|6564$Z@O8_tZr%eKtwDi=INr5F~@)oiAS?`H+qYbFD53z_t~&M+{`o
zE!l&r!;?zJQKR=#IEDc?Tz9D%^g2qT;@w9BDj*h9jHMU*Eh!23uKoM;1t23K*f02q
zPOQW5YgzMKH=ko=_{Xtiw8Kk(7G{WHOmG%uP<$R9g#sTDpJGDyugTfTPEKo6#LbP=
zxi~nC;~#T9eA*%QKWD(9b3k1!rIpmC9F@vs!C?n1rI8vdM$)*y3XBB>DUJI-!rn^_
ztBH>YL{Xa<3LSGD-3B4vG(4Ss=W*2W1>SHTvhwzCn)XYj$5e0l+v8KVCA??<DDq4D
zs`&z=Z2sQKw?NrIpQ(uqj!42BkuR&hS7-Uv)n_lkmg}Wj=9h3;w(QRo!G*3S)*^h0
z0~rv0n*hBFdEanNlBAQ{HyFs3MG|W@1a=v6Dy8)mzQ`aNmh-hiRLH@-34c)@V`nUv
zPiFB8q^!$^phaZ3@ihW1?mPkno(53^H~<emm^&p6;?~|4;TBgP5e^|vuaz*N^8>*i
zt5M3_=!;62;`I4>FUd;-1yWdT^-MdpvSXaGglB%w1gr>@TP`b7v(GdTH&R1!6bAib
zXq=Qu|M*t{UjQQlD+@{z%^pTSgOM3Wl~YFS+|Rceqc-oJ``4MVyC5+CAli0-?507g
zY5>tCs9kLg0j2;{MapvW;rikjan*J75?(1OR~7_t3Oq&BVe4P(V07)<D&AQa{fy#q
zkt6DZbqu)O8m57qAgs>9GDflc>cO!zxk8(BPXT4S#Jo){!EgLLQt@=!=1EUEl5030
zl&R9`6>8S{wT#&KA;`gDzYQGKTU^><&!egO<a;`}^$WSNQ+I@8T^IgS5$cpvN@}h<
z{G?4h)EJfTKm9_VS59fpauT5GF6+OsO-E@;bA$FzzsFpJ0>@Hf1T`v_1eo|9M!cqk
zf9#f0(Ki!+YKcVHP))`_gIh~da}h2;LS|*QUEcB!O%3Y|ViD7+{P0_S>E;l{(R5h?
zMilZ2+fJ1$E)zn|J9;3JK2}tw&2>oOxAPLn`PTc!xjxUeOQJDIPCLBKfzhk{0vxZ@
zF0FR<d1f9$vN{M`vtCtDyGx0|a~brT-4r`RM*=8AqK)&7vzd=IH3LG>*xX3DJu-#Q
zbNuo;GcF+!ME)(VAJ|lvM5)Ew>Zku7+#cd`a2N<uh3-L^;P~2{%I$;U&1cU5F30t)
zZ#g9Z&Bvrphai}Uh(n#dyH_S@D@NPG>XwvS=Y){m2=Cx-0iU^YyS&?0Xa(MFA7QAe
z`3;c}o;#3!<t?kK*_(+qWT_wqx%#8fgH6u+m%^J{)p-{Z%W0Yaa`;@Jz1{L2-ys*C
z_*-9}Hw`(Q%qwW*W5ojxB)s(6pQeTn4q1D5|7+0Xq2Nv8rdPl@5ixx?^aGyH+$et;
zn5u@J-XrW%qgUMV$Hya!4)6waa`|#o`|>x~cbQ_xv)f756VorAgkG2g&>lc1ILe>;
zMID5)%{gY6H_ce`q3)t2pf7w{2!|!Mod>HmOmhq5PV}#UkZui_b=Lg)m|Ynjg#vf?
zccB(DZ5R0?&b(vn-uXui!PcKH*GZq138rs)QtTFOQoR`~@_o<_jIz}G<6*sbLM1CQ
z&4$Z5qCX{7Zqb3EWH0=ebqt`m-DQo7Tzq>_>4%KZDi1(ld`E)Hpl0o59P)dEvoPq7
z`2yYEuC!RMwDnL!hm-Unucxg~n=pVV*Enpq0wwBRtbS|ngK=^WM<afLp^d~b{;0wd
z_&++-4nXXu{16a8`4Z!ve!=bQSaq;i^VWHYE?nUB5#Z9<iWdrGp<eI+Wsj4>@*sNm
z9$vVWxd5(Ge8okW(huCGhqJfodnki>PiLn;Cx}w-elV3!opbv^1XYnLw^bPl)Fvq9
z5tWGm3*HSkpYeN1M{y-Z<84kA;unjrqNhhm&pt<ZprRU^XPQ-ha#QVS{FK+z0V~RL
zx^s+6&4_~Wj?Ax9)#(Z<m$i;(YopiGqVE}^k!1JrZkD`vL4~yuedLP*Kl6lNyYPem
g*ERfphe9j<U3DF{x>-s^(xlOTWT093&@SeG0loJJ8vp<R

literal 0
HcmV?d00001

diff --git a/0.16/_static/check-solid.svg b/0.16/_static/check-solid.svg
new file mode 100644
index 00000000000..92fad4b5c0b
--- /dev/null
+++ b/0.16/_static/check-solid.svg
@@ -0,0 +1,4 @@
+<svg xmlns="http://www.w3.org/2000/svg" class="icon icon-tabler icon-tabler-check" width="44" height="44" viewBox="0 0 24 24" stroke-width="2" stroke="#22863a" fill="none" stroke-linecap="round" stroke-linejoin="round">
+  <path stroke="none" d="M0 0h24v24H0z" fill="none"/>
+  <path d="M5 12l5 5l10 -10" />
+</svg>
diff --git a/0.16/_static/clipboard.min.js b/0.16/_static/clipboard.min.js
new file mode 100644
index 00000000000..54b3c463811
--- /dev/null
+++ b/0.16/_static/clipboard.min.js
@@ -0,0 +1,7 @@
+/*!
+ * clipboard.js v2.0.8
+ * https://clipboardjs.com/
+ *
+ * Licensed MIT © Zeno Rocha
+ */
+!function(t,e){"object"==typeof exports&&"object"==typeof module?module.exports=e():"function"==typeof define&&define.amd?define([],e):"object"==typeof exports?exports.ClipboardJS=e():t.ClipboardJS=e()}(this,function(){return n={686:function(t,e,n){"use strict";n.d(e,{default:function(){return o}});var e=n(279),i=n.n(e),e=n(370),u=n.n(e),e=n(817),c=n.n(e);function a(t){try{return document.execCommand(t)}catch(t){return}}var f=function(t){t=c()(t);return a("cut"),t};var l=function(t){var e,n,o,r=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{container:document.body},i="";return"string"==typeof t?(e=t,n="rtl"===document.documentElement.getAttribute("dir"),(o=document.createElement("textarea")).style.fontSize="12pt",o.style.border="0",o.style.padding="0",o.style.margin="0",o.style.position="absolute",o.style[n?"right":"left"]="-9999px",n=window.pageYOffset||document.documentElement.scrollTop,o.style.top="".concat(n,"px"),o.setAttribute("readonly",""),o.value=e,o=o,r.container.appendChild(o),i=c()(o),a("copy"),o.remove()):(i=c()(t),a("copy")),i};function r(t){return(r="function"==typeof Symbol&&"symbol"==typeof Symbol.iterator?function(t){return typeof t}:function(t){return t&&"function"==typeof Symbol&&t.constructor===Symbol&&t!==Symbol.prototype?"symbol":typeof t})(t)}var s=function(){var t=0<arguments.length&&void 0!==arguments[0]?arguments[0]:{},e=t.action,n=void 0===e?"copy":e,o=t.container,e=t.target,t=t.text;if("copy"!==n&&"cut"!==n)throw new Error('Invalid "action" value, use either "copy" or "cut"');if(void 0!==e){if(!e||"object"!==r(e)||1!==e.nodeType)throw new Error('Invalid "target" value, use a valid Element');if("copy"===n&&e.hasAttribute("disabled"))throw new Error('Invalid "target" attribute. Please use "readonly" instead of "disabled" attribute');if("cut"===n&&(e.hasAttribute("readonly")||e.hasAttribute("disabled")))throw new Error('Invalid "target" attribute. You can\'t cut text from elements with "readonly" or "disabled" attributes')}return t?l(t,{container:o}):e?"cut"===n?f(e):l(e,{container:o}):void 0};function d(t){return(d="function"==typeof Symbol&&"symbol"==typeof Symbol.iterator?function(t){return typeof t}:function(t){return t&&"function"==typeof Symbol&&t.constructor===Symbol&&t!==Symbol.prototype?"symbol":typeof t})(t)}function p(t,e){for(var n=0;n<e.length;n++){var o=e[n];o.enumerable=o.enumerable||!1,o.configurable=!0,"value"in o&&(o.writable=!0),Object.defineProperty(t,o.key,o)}}function y(t,e){return(y=Object.setPrototypeOf||function(t,e){return t.__proto__=e,t})(t,e)}function h(n){var o=function(){if("undefined"==typeof Reflect||!Reflect.construct)return!1;if(Reflect.construct.sham)return!1;if("function"==typeof Proxy)return!0;try{return Date.prototype.toString.call(Reflect.construct(Date,[],function(){})),!0}catch(t){return!1}}();return function(){var t,e=m(n);return t=o?(t=m(this).constructor,Reflect.construct(e,arguments,t)):e.apply(this,arguments),e=this,!(t=t)||"object"!==d(t)&&"function"!=typeof t?function(t){if(void 0!==t)return t;throw new ReferenceError("this hasn't been initialised - super() hasn't been called")}(e):t}}function m(t){return(m=Object.setPrototypeOf?Object.getPrototypeOf:function(t){return t.__proto__||Object.getPrototypeOf(t)})(t)}function v(t,e){t="data-clipboard-".concat(t);if(e.hasAttribute(t))return e.getAttribute(t)}var o=function(){!function(t,e){if("function"!=typeof e&&null!==e)throw new TypeError("Super expression must either be null or a function");t.prototype=Object.create(e&&e.prototype,{constructor:{value:t,writable:!0,configurable:!0}}),e&&y(t,e)}(r,i());var t,e,n,o=h(r);function r(t,e){var n;return function(t){if(!(t instanceof r))throw new TypeError("Cannot call a class as a function")}(this),(n=o.call(this)).resolveOptions(e),n.listenClick(t),n}return t=r,n=[{key:"copy",value:function(t){var e=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{container:document.body};return l(t,e)}},{key:"cut",value:function(t){return f(t)}},{key:"isSupported",value:function(){var t=0<arguments.length&&void 0!==arguments[0]?arguments[0]:["copy","cut"],t="string"==typeof t?[t]:t,e=!!document.queryCommandSupported;return t.forEach(function(t){e=e&&!!document.queryCommandSupported(t)}),e}}],(e=[{key:"resolveOptions",value:function(){var t=0<arguments.length&&void 0!==arguments[0]?arguments[0]:{};this.action="function"==typeof t.action?t.action:this.defaultAction,this.target="function"==typeof t.target?t.target:this.defaultTarget,this.text="function"==typeof t.text?t.text:this.defaultText,this.container="object"===d(t.container)?t.container:document.body}},{key:"listenClick",value:function(t){var e=this;this.listener=u()(t,"click",function(t){return e.onClick(t)})}},{key:"onClick",value:function(t){var e=t.delegateTarget||t.currentTarget,t=s({action:this.action(e),container:this.container,target:this.target(e),text:this.text(e)});this.emit(t?"success":"error",{action:this.action,text:t,trigger:e,clearSelection:function(){e&&e.focus(),document.activeElement.blur(),window.getSelection().removeAllRanges()}})}},{key:"defaultAction",value:function(t){return v("action",t)}},{key:"defaultTarget",value:function(t){t=v("target",t);if(t)return document.querySelector(t)}},{key:"defaultText",value:function(t){return v("text",t)}},{key:"destroy",value:function(){this.listener.destroy()}}])&&p(t.prototype,e),n&&p(t,n),r}()},828:function(t){var e;"undefined"==typeof Element||Element.prototype.matches||((e=Element.prototype).matches=e.matchesSelector||e.mozMatchesSelector||e.msMatchesSelector||e.oMatchesSelector||e.webkitMatchesSelector),t.exports=function(t,e){for(;t&&9!==t.nodeType;){if("function"==typeof t.matches&&t.matches(e))return t;t=t.parentNode}}},438:function(t,e,n){var u=n(828);function i(t,e,n,o,r){var i=function(e,n,t,o){return function(t){t.delegateTarget=u(t.target,n),t.delegateTarget&&o.call(e,t)}}.apply(this,arguments);return t.addEventListener(n,i,r),{destroy:function(){t.removeEventListener(n,i,r)}}}t.exports=function(t,e,n,o,r){return"function"==typeof t.addEventListener?i.apply(null,arguments):"function"==typeof n?i.bind(null,document).apply(null,arguments):("string"==typeof t&&(t=document.querySelectorAll(t)),Array.prototype.map.call(t,function(t){return i(t,e,n,o,r)}))}},879:function(t,n){n.node=function(t){return void 0!==t&&t instanceof HTMLElement&&1===t.nodeType},n.nodeList=function(t){var e=Object.prototype.toString.call(t);return void 0!==t&&("[object NodeList]"===e||"[object HTMLCollection]"===e)&&"length"in t&&(0===t.length||n.node(t[0]))},n.string=function(t){return"string"==typeof t||t instanceof String},n.fn=function(t){return"[object Function]"===Object.prototype.toString.call(t)}},370:function(t,e,n){var f=n(879),l=n(438);t.exports=function(t,e,n){if(!t&&!e&&!n)throw new Error("Missing required arguments");if(!f.string(e))throw new TypeError("Second argument must be a String");if(!f.fn(n))throw new TypeError("Third argument must be a Function");if(f.node(t))return c=e,a=n,(u=t).addEventListener(c,a),{destroy:function(){u.removeEventListener(c,a)}};if(f.nodeList(t))return o=t,r=e,i=n,Array.prototype.forEach.call(o,function(t){t.addEventListener(r,i)}),{destroy:function(){Array.prototype.forEach.call(o,function(t){t.removeEventListener(r,i)})}};if(f.string(t))return t=t,e=e,n=n,l(document.body,t,e,n);throw new TypeError("First argument must be a String, HTMLElement, HTMLCollection, or NodeList");var o,r,i,u,c,a}},817:function(t){t.exports=function(t){var e,n="SELECT"===t.nodeName?(t.focus(),t.value):"INPUT"===t.nodeName||"TEXTAREA"===t.nodeName?((e=t.hasAttribute("readonly"))||t.setAttribute("readonly",""),t.select(),t.setSelectionRange(0,t.value.length),e||t.removeAttribute("readonly"),t.value):(t.hasAttribute("contenteditable")&&t.focus(),n=window.getSelection(),(e=document.createRange()).selectNodeContents(t),n.removeAllRanges(),n.addRange(e),n.toString());return n}},279:function(t){function e(){}e.prototype={on:function(t,e,n){var o=this.e||(this.e={});return(o[t]||(o[t]=[])).push({fn:e,ctx:n}),this},once:function(t,e,n){var o=this;function r(){o.off(t,r),e.apply(n,arguments)}return r._=e,this.on(t,r,n)},emit:function(t){for(var e=[].slice.call(arguments,1),n=((this.e||(this.e={}))[t]||[]).slice(),o=0,r=n.length;o<r;o++)n[o].fn.apply(n[o].ctx,e);return this},off:function(t,e){var n=this.e||(this.e={}),o=n[t],r=[];if(o&&e)for(var i=0,u=o.length;i<u;i++)o[i].fn!==e&&o[i].fn._!==e&&r.push(o[i]);return r.length?n[t]=r:delete n[t],this}},t.exports=e,t.exports.TinyEmitter=e}},r={},o.n=function(t){var e=t&&t.__esModule?function(){return t.default}:function(){return t};return o.d(e,{a:e}),e},o.d=function(t,e){for(var n in e)o.o(e,n)&&!o.o(t,n)&&Object.defineProperty(t,n,{enumerable:!0,get:e[n]})},o.o=function(t,e){return Object.prototype.hasOwnProperty.call(t,e)},o(686).default;function o(t){if(r[t])return r[t].exports;var e=r[t]={exports:{}};return n[t](e,e.exports,o),e.exports}var n,r});
\ No newline at end of file
diff --git a/0.16/_static/copy-button.svg b/0.16/_static/copy-button.svg
new file mode 100644
index 00000000000..9c074dae528
--- /dev/null
+++ b/0.16/_static/copy-button.svg
@@ -0,0 +1,5 @@
+<svg xmlns="http://www.w3.org/2000/svg" class="icon icon-tabler icon-tabler-copy" width="44" height="44" viewBox="0 0 24 24" stroke-width="1.5" stroke="#000000" fill="none" stroke-linecap="round" stroke-linejoin="round">
+  <path stroke="none" d="M0 0h24v24H0z" fill="none"/>
+  <rect x="8" y="8" width="12" height="12" rx="2" />
+  <path d="M16 8v-2a2 2 0 0 0 -2 -2h-8a2 2 0 0 0 -2 2v8a2 2 0 0 0 2 2h2" />
+</svg>
diff --git a/0.16/_static/copybutton.css b/0.16/_static/copybutton.css
new file mode 100644
index 00000000000..f1916ec7d1b
--- /dev/null
+++ b/0.16/_static/copybutton.css
@@ -0,0 +1,94 @@
+/* Copy buttons */
+button.copybtn {
+    position: absolute;
+    display: flex;
+    top: .3em;
+    right: .3em;
+    width: 1.7em;
+    height: 1.7em;
+	opacity: 0;
+    transition: opacity 0.3s, border .3s, background-color .3s;
+    user-select: none;
+    padding: 0;
+    border: none;
+    outline: none;
+    border-radius: 0.4em;
+    /* The colors that GitHub uses */
+    border: #1b1f2426 1px solid;
+    background-color: #f6f8fa;
+    color: #57606a;
+}
+
+button.copybtn.success {
+    border-color: #22863a;
+    color: #22863a;
+}
+
+button.copybtn svg {
+    stroke: currentColor;
+    width: 1.5em;
+    height: 1.5em;
+    padding: 0.1em;
+}
+
+div.highlight  {
+    position: relative;
+}
+
+/* Show the copybutton */
+.highlight:hover button.copybtn, button.copybtn.success {
+	opacity: 1;
+}
+
+.highlight button.copybtn:hover {
+    background-color: rgb(235, 235, 235);
+}
+
+.highlight button.copybtn:active {
+    background-color: rgb(187, 187, 187);
+}
+
+/**
+ * A minimal CSS-only tooltip copied from:
+ *   https://codepen.io/mildrenben/pen/rVBrpK
+ *
+ * To use, write HTML like the following:
+ *
+ * <p class="o-tooltip--left" data-tooltip="Hey">Short</p>
+ */
+ .o-tooltip--left {
+  position: relative;
+ }
+
+ .o-tooltip--left:after {
+    opacity: 0;
+    visibility: hidden;
+    position: absolute;
+    content: attr(data-tooltip);
+    padding: .2em;
+    font-size: .8em;
+    left: -.2em;
+    background: grey;
+    color: white;
+    white-space: nowrap;
+    z-index: 2;
+    border-radius: 2px;
+    transform: translateX(-102%) translateY(0);
+    transition: opacity 0.2s cubic-bezier(0.64, 0.09, 0.08, 1), transform 0.2s cubic-bezier(0.64, 0.09, 0.08, 1);
+}
+
+.o-tooltip--left:hover:after {
+    display: block;
+    opacity: 1;
+    visibility: visible;
+    transform: translateX(-100%) translateY(0);
+    transition: opacity 0.2s cubic-bezier(0.64, 0.09, 0.08, 1), transform 0.2s cubic-bezier(0.64, 0.09, 0.08, 1);
+    transition-delay: .5s;
+}
+
+/* By default the copy button shouldn't show up when printing a page */
+@media print {
+    button.copybtn {
+        display: none;
+    }
+}
diff --git a/0.16/_static/copybutton.js b/0.16/_static/copybutton.js
new file mode 100644
index 00000000000..2ea7ff3e217
--- /dev/null
+++ b/0.16/_static/copybutton.js
@@ -0,0 +1,248 @@
+// Localization support
+const messages = {
+  'en': {
+    'copy': 'Copy',
+    'copy_to_clipboard': 'Copy to clipboard',
+    'copy_success': 'Copied!',
+    'copy_failure': 'Failed to copy',
+  },
+  'es' : {
+    'copy': 'Copiar',
+    'copy_to_clipboard': 'Copiar al portapapeles',
+    'copy_success': '¡Copiado!',
+    'copy_failure': 'Error al copiar',
+  },
+  'de' : {
+    'copy': 'Kopieren',
+    'copy_to_clipboard': 'In die Zwischenablage kopieren',
+    'copy_success': 'Kopiert!',
+    'copy_failure': 'Fehler beim Kopieren',
+  },
+  'fr' : {
+    'copy': 'Copier',
+    'copy_to_clipboard': 'Copier dans le presse-papier',
+    'copy_success': 'Copié !',
+    'copy_failure': 'Échec de la copie',
+  },
+  'ru': {
+    'copy': 'Скопировать',
+    'copy_to_clipboard': 'Скопировать в буфер',
+    'copy_success': 'Скопировано!',
+    'copy_failure': 'Не удалось скопировать',
+  },
+  'zh-CN': {
+    'copy': '复制',
+    'copy_to_clipboard': '复制到剪贴板',
+    'copy_success': '复制成功!',
+    'copy_failure': '复制失败',
+  },
+  'it' : {
+    'copy': 'Copiare',
+    'copy_to_clipboard': 'Copiato negli appunti',
+    'copy_success': 'Copiato!',
+    'copy_failure': 'Errore durante la copia',
+  }
+}
+
+let locale = 'en'
+if( document.documentElement.lang !== undefined
+    && messages[document.documentElement.lang] !== undefined ) {
+  locale = document.documentElement.lang
+}
+
+let doc_url_root = DOCUMENTATION_OPTIONS.URL_ROOT;
+if (doc_url_root == '#') {
+    doc_url_root = '';
+}
+
+/**
+ * SVG files for our copy buttons
+ */
+let iconCheck = `<svg xmlns="http://www.w3.org/2000/svg" class="icon icon-tabler icon-tabler-check" width="44" height="44" viewBox="0 0 24 24" stroke-width="2" stroke="#22863a" fill="none" stroke-linecap="round" stroke-linejoin="round">
+  <title>${messages[locale]['copy_success']}</title>
+  <path stroke="none" d="M0 0h24v24H0z" fill="none"/>
+  <path d="M5 12l5 5l10 -10" />
+</svg>`
+
+// If the user specified their own SVG use that, otherwise use the default
+let iconCopy = ``;
+if (!iconCopy) {
+  iconCopy = `<svg xmlns="http://www.w3.org/2000/svg" class="icon icon-tabler icon-tabler-copy" width="44" height="44" viewBox="0 0 24 24" stroke-width="1.5" stroke="#000000" fill="none" stroke-linecap="round" stroke-linejoin="round">
+  <title>${messages[locale]['copy_to_clipboard']}</title>
+  <path stroke="none" d="M0 0h24v24H0z" fill="none"/>
+  <rect x="8" y="8" width="12" height="12" rx="2" />
+  <path d="M16 8v-2a2 2 0 0 0 -2 -2h-8a2 2 0 0 0 -2 2v8a2 2 0 0 0 2 2h2" />
+</svg>`
+}
+
+/**
+ * Set up copy/paste for code blocks
+ */
+
+const runWhenDOMLoaded = cb => {
+  if (document.readyState != 'loading') {
+    cb()
+  } else if (document.addEventListener) {
+    document.addEventListener('DOMContentLoaded', cb)
+  } else {
+    document.attachEvent('onreadystatechange', function() {
+      if (document.readyState == 'complete') cb()
+    })
+  }
+}
+
+const codeCellId = index => `codecell${index}`
+
+// Clears selected text since ClipboardJS will select the text when copying
+const clearSelection = () => {
+  if (window.getSelection) {
+    window.getSelection().removeAllRanges()
+  } else if (document.selection) {
+    document.selection.empty()
+  }
+}
+
+// Changes tooltip text for a moment, then changes it back
+// We want the timeout of our `success` class to be a bit shorter than the
+// tooltip and icon change, so that we can hide the icon before changing back.
+var timeoutIcon = 2000;
+var timeoutSuccessClass = 1500;
+
+const temporarilyChangeTooltip = (el, oldText, newText) => {
+  el.setAttribute('data-tooltip', newText)
+  el.classList.add('success')
+  // Remove success a little bit sooner than we change the tooltip
+  // So that we can use CSS to hide the copybutton first
+  setTimeout(() => el.classList.remove('success'), timeoutSuccessClass)
+  setTimeout(() => el.setAttribute('data-tooltip', oldText), timeoutIcon)
+}
+
+// Changes the copy button icon for two seconds, then changes it back
+const temporarilyChangeIcon = (el) => {
+  el.innerHTML = iconCheck;
+  setTimeout(() => {el.innerHTML = iconCopy}, timeoutIcon)
+}
+
+const addCopyButtonToCodeCells = () => {
+  // If ClipboardJS hasn't loaded, wait a bit and try again. This
+  // happens because we load ClipboardJS asynchronously.
+  if (window.ClipboardJS === undefined) {
+    setTimeout(addCopyButtonToCodeCells, 250)
+    return
+  }
+
+  // Add copybuttons to all of our code cells
+  const COPYBUTTON_SELECTOR = 'div.highlight pre';
+  const codeCells = document.querySelectorAll(COPYBUTTON_SELECTOR)
+  codeCells.forEach((codeCell, index) => {
+    const id = codeCellId(index)
+    codeCell.setAttribute('id', id)
+
+    const clipboardButton = id =>
+    `<button class="copybtn o-tooltip--left" data-tooltip="${messages[locale]['copy']}" data-clipboard-target="#${id}">
+      ${iconCopy}
+    </button>`
+    codeCell.insertAdjacentHTML('afterend', clipboardButton(id))
+  })
+
+function escapeRegExp(string) {
+    return string.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); // $& means the whole matched string
+}
+
+/**
+ * Removes excluded text from a Node.
+ *
+ * @param {Node} target Node to filter.
+ * @param {string} exclude CSS selector of nodes to exclude.
+ * @returns {DOMString} Text from `target` with text removed.
+ */
+function filterText(target, exclude) {
+    const clone = target.cloneNode(true);  // clone as to not modify the live DOM
+    if (exclude) {
+        // remove excluded nodes
+        clone.querySelectorAll(exclude).forEach(node => node.remove());
+    }
+    return clone.innerText;
+}
+
+// Callback when a copy button is clicked. Will be passed the node that was clicked
+// should then grab the text and replace pieces of text that shouldn't be used in output
+function formatCopyText(textContent, copybuttonPromptText, isRegexp = false, onlyCopyPromptLines = true, removePrompts = true, copyEmptyLines = true, lineContinuationChar = "", hereDocDelim = "") {
+    var regexp;
+    var match;
+
+    // Do we check for line continuation characters and "HERE-documents"?
+    var useLineCont = !!lineContinuationChar
+    var useHereDoc = !!hereDocDelim
+
+    // create regexp to capture prompt and remaining line
+    if (isRegexp) {
+        regexp = new RegExp('^(' + copybuttonPromptText + ')(.*)')
+    } else {
+        regexp = new RegExp('^(' + escapeRegExp(copybuttonPromptText) + ')(.*)')
+    }
+
+    const outputLines = [];
+    var promptFound = false;
+    var gotLineCont = false;
+    var gotHereDoc = false;
+    const lineGotPrompt = [];
+    for (const line of textContent.split('\n')) {
+        match = line.match(regexp)
+        if (match || gotLineCont || gotHereDoc) {
+            promptFound = regexp.test(line)
+            lineGotPrompt.push(promptFound)
+            if (removePrompts && promptFound) {
+                outputLines.push(match[2])
+            } else {
+                outputLines.push(line)
+            }
+            gotLineCont = line.endsWith(lineContinuationChar) & useLineCont
+            if (line.includes(hereDocDelim) & useHereDoc)
+                gotHereDoc = !gotHereDoc
+        } else if (!onlyCopyPromptLines) {
+            outputLines.push(line)
+        } else if (copyEmptyLines && line.trim() === '') {
+            outputLines.push(line)
+        }
+    }
+
+    // If no lines with the prompt were found then just use original lines
+    if (lineGotPrompt.some(v => v === true)) {
+        textContent = outputLines.join('\n');
+    }
+
+    // Remove a trailing newline to avoid auto-running when pasting
+    if (textContent.endsWith("\n")) {
+        textContent = textContent.slice(0, -1)
+    }
+    return textContent
+}
+
+
+var copyTargetText = (trigger) => {
+  var target = document.querySelector(trigger.attributes['data-clipboard-target'].value);
+
+  // get filtered text
+  let exclude = '.linenos';
+
+  let text = filterText(target, exclude);
+  return formatCopyText(text, '', false, true, true, true, '', '')
+}
+
+  // Initialize with a callback so we can modify the text before copy
+  const clipboard = new ClipboardJS('.copybtn', {text: copyTargetText})
+
+  // Update UI with error/success messages
+  clipboard.on('success', event => {
+    clearSelection()
+    temporarilyChangeTooltip(event.trigger, messages[locale]['copy'], messages[locale]['copy_success'])
+    temporarilyChangeIcon(event.trigger)
+  })
+
+  clipboard.on('error', event => {
+    temporarilyChangeTooltip(event.trigger, messages[locale]['copy'], messages[locale]['copy_failure'])
+  })
+}
+
+runWhenDOMLoaded(addCopyButtonToCodeCells)
\ No newline at end of file
diff --git a/0.16/_static/copybutton_funcs.js b/0.16/_static/copybutton_funcs.js
new file mode 100644
index 00000000000..dbe1aaad79c
--- /dev/null
+++ b/0.16/_static/copybutton_funcs.js
@@ -0,0 +1,73 @@
+function escapeRegExp(string) {
+    return string.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'); // $& means the whole matched string
+}
+
+/**
+ * Removes excluded text from a Node.
+ *
+ * @param {Node} target Node to filter.
+ * @param {string} exclude CSS selector of nodes to exclude.
+ * @returns {DOMString} Text from `target` with text removed.
+ */
+export function filterText(target, exclude) {
+    const clone = target.cloneNode(true);  // clone as to not modify the live DOM
+    if (exclude) {
+        // remove excluded nodes
+        clone.querySelectorAll(exclude).forEach(node => node.remove());
+    }
+    return clone.innerText;
+}
+
+// Callback when a copy button is clicked. Will be passed the node that was clicked
+// should then grab the text and replace pieces of text that shouldn't be used in output
+export function formatCopyText(textContent, copybuttonPromptText, isRegexp = false, onlyCopyPromptLines = true, removePrompts = true, copyEmptyLines = true, lineContinuationChar = "", hereDocDelim = "") {
+    var regexp;
+    var match;
+
+    // Do we check for line continuation characters and "HERE-documents"?
+    var useLineCont = !!lineContinuationChar
+    var useHereDoc = !!hereDocDelim
+
+    // create regexp to capture prompt and remaining line
+    if (isRegexp) {
+        regexp = new RegExp('^(' + copybuttonPromptText + ')(.*)')
+    } else {
+        regexp = new RegExp('^(' + escapeRegExp(copybuttonPromptText) + ')(.*)')
+    }
+
+    const outputLines = [];
+    var promptFound = false;
+    var gotLineCont = false;
+    var gotHereDoc = false;
+    const lineGotPrompt = [];
+    for (const line of textContent.split('\n')) {
+        match = line.match(regexp)
+        if (match || gotLineCont || gotHereDoc) {
+            promptFound = regexp.test(line)
+            lineGotPrompt.push(promptFound)
+            if (removePrompts && promptFound) {
+                outputLines.push(match[2])
+            } else {
+                outputLines.push(line)
+            }
+            gotLineCont = line.endsWith(lineContinuationChar) & useLineCont
+            if (line.includes(hereDocDelim) & useHereDoc)
+                gotHereDoc = !gotHereDoc
+        } else if (!onlyCopyPromptLines) {
+            outputLines.push(line)
+        } else if (copyEmptyLines && line.trim() === '') {
+            outputLines.push(line)
+        }
+    }
+
+    // If no lines with the prompt were found then just use original lines
+    if (lineGotPrompt.some(v => v === true)) {
+        textContent = outputLines.join('\n');
+    }
+
+    // Remove a trailing newline to avoid auto-running when pasting
+    if (textContent.endsWith("\n")) {
+        textContent = textContent.slice(0, -1)
+    }
+    return textContent
+}
diff --git a/0.16/_static/css/custom_torchvision.css b/0.16/_static/css/custom_torchvision.css
new file mode 100644
index 00000000000..07346d7b03f
--- /dev/null
+++ b/0.16/_static/css/custom_torchvision.css
@@ -0,0 +1,35 @@
+/* This rule should be removed once
+https://github.com/pytorch/pytorch_sphinx_theme/issues/125 is fixed.
+
+We override the rule so that the links to the notebooks aren't hidden in the
+gallery examples. pytorch_sphinx_theme is supposed to customize those links so
+that they render nicely (look at the nice links on top of the tutorials
+examples) but it doesn't work for repos that are not the tutorial repo, and in
+torchvision it just hides the links. So we have to put them back here */
+article.pytorch-article .sphx-glr-download-link-note.admonition.note,
+article.pytorch-article .reference.download.internal, article.pytorch-article .sphx-glr-signature {
+  display: block;
+}
+
+/* These 2 rules below are for the weight tables (generated in conf.py) to look
+ * better. In particular we make their row height shorter */
+.table-weights td, .table-weights th {
+  margin-bottom: 0.2rem;
+  padding: 0 !important;
+  line-height: 1 !important;
+}
+.table-weights p {
+    margin-bottom: 0.2rem !important;
+}
+
+/* Fix for Sphinx gallery 0.11
+See https://github.com/sphinx-gallery/sphinx-gallery/issues/990
+*/
+article.pytorch-article .sphx-glr-thumbnails .sphx-glr-thumbcontainer {
+    width: unset;
+    margin-right: 0;
+    margin-left: 0;
+}
+article.pytorch-article div.section div.wy-table-responsive tbody td {
+    width: 50%;
+}
diff --git a/0.16/_static/css/theme.css b/0.16/_static/css/theme.css
new file mode 100644
index 00000000000..43cf163a618
--- /dev/null
+++ b/0.16/_static/css/theme.css
@@ -0,0 +1,12482 @@
+@charset "UTF-8";
+/*!
+ * Bootstrap v4.0.0 (https://getbootstrap.com)
+ * Copyright 2011-2018 The Bootstrap Authors
+ * Copyright 2011-2018 Twitter, Inc.
+ * Licensed under MIT (https://github.com/twbs/bootstrap/blob/master/LICENSE)
+ */
+:root {
+  --blue: #007bff;
+  --indigo: #6610f2;
+  --purple: #6f42c1;
+  --pink: #e83e8c;
+  --red: #dc3545;
+  --orange: #fd7e14;
+  --yellow: #ffc107;
+  --green: #28a745;
+  --teal: #20c997;
+  --cyan: #17a2b8;
+  --white: #fff;
+  --gray: #6c757d;
+  --gray-dark: #343a40;
+  --primary: #007bff;
+  --secondary: #6c757d;
+  --success: #28a745;
+  --info: #17a2b8;
+  --warning: #ffc107;
+  --danger: #dc3545;
+  --light: #f8f9fa;
+  --dark: #343a40;
+  --breakpoint-xs: 0;
+  --breakpoint-sm: 576px;
+  --breakpoint-md: 768px;
+  --breakpoint-lg: 992px;
+  --breakpoint-xl: 1200px;
+  --font-family-sans-serif: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";
+  --font-family-monospace: SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;
+}
+
+*,
+*::before,
+*::after {
+  -webkit-box-sizing: border-box;
+          box-sizing: border-box;
+}
+
+html {
+  font-family: sans-serif;
+  line-height: 1.15;
+  -webkit-text-size-adjust: 100%;
+  -ms-text-size-adjust: 100%;
+  -ms-overflow-style: scrollbar;
+  -webkit-tap-highlight-color: rgba(0, 0, 0, 0);
+}
+
+@-ms-viewport {
+  width: device-width;
+}
+article, aside, dialog, figcaption, figure, footer, header, hgroup, main, nav, section {
+  display: block;
+}
+
+body {
+  margin: 0;
+  font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";
+  font-size: 1rem;
+  font-weight: 400;
+  line-height: 1.5;
+  color: #212529;
+  text-align: left;
+  background-color: #fff;
+}
+
+[tabindex="-1"]:focus {
+  outline: 0 !important;
+}
+
+hr {
+  -webkit-box-sizing: content-box;
+          box-sizing: content-box;
+  height: 0;
+  overflow: visible;
+}
+
+h1, h2, h3, h4, h5, h6 {
+  margin-top: 0;
+  margin-bottom: 0.5rem;
+}
+
+p {
+  margin-top: 0;
+  margin-bottom: 1rem;
+}
+
+abbr[title],
+abbr[data-original-title] {
+  text-decoration: underline;
+  -webkit-text-decoration: underline dotted;
+          text-decoration: underline dotted;
+  cursor: help;
+  border-bottom: 0;
+}
+
+address {
+  margin-bottom: 1rem;
+  font-style: normal;
+  line-height: inherit;
+}
+
+ol,
+ul,
+dl {
+  margin-top: 0;
+  margin-bottom: 1rem;
+}
+
+ol ol,
+ul ul,
+ol ul,
+ul ol {
+  margin-bottom: 0;
+}
+
+dt {
+  font-weight: 700;
+}
+
+dd {
+  margin-bottom: .5rem;
+  margin-left: 0;
+}
+
+blockquote {
+  margin: 0 0 1rem;
+}
+
+dfn {
+  font-style: italic;
+}
+
+b,
+strong {
+  font-weight: bolder;
+}
+
+small {
+  font-size: 80%;
+}
+
+sub,
+sup {
+  position: relative;
+  font-size: 75%;
+  line-height: 0;
+  vertical-align: baseline;
+}
+
+sub {
+  bottom: -.25em;
+}
+
+sup {
+  top: -.5em;
+}
+
+a {
+  color: #007bff;
+  text-decoration: none;
+  background-color: transparent;
+  -webkit-text-decoration-skip: objects;
+}
+a:hover {
+  color: #0056b3;
+  text-decoration: underline;
+}
+
+a:not([href]):not([tabindex]) {
+  color: inherit;
+  text-decoration: none;
+}
+a:not([href]):not([tabindex]):hover, a:not([href]):not([tabindex]):focus {
+  color: inherit;
+  text-decoration: none;
+}
+a:not([href]):not([tabindex]):focus {
+  outline: 0;
+}
+
+pre,
+code,
+kbd,
+samp {
+  font-family: monospace, monospace;
+  font-size: 1em;
+}
+
+pre {
+  margin-top: 0;
+  margin-bottom: 1rem;
+  overflow: auto;
+  -ms-overflow-style: scrollbar;
+}
+
+figure {
+  margin: 0 0 1rem;
+}
+
+img {
+  vertical-align: middle;
+  border-style: none;
+}
+
+svg:not(:root) {
+  overflow: hidden;
+}
+
+table {
+  border-collapse: collapse;
+}
+
+caption {
+  padding-top: 0.75rem;
+  padding-bottom: 0.75rem;
+  color: #6c757d;
+  text-align: left;
+  caption-side: bottom;
+}
+
+th {
+  text-align: inherit;
+}
+
+label {
+  display: inline-block;
+  margin-bottom: .5rem;
+}
+
+button {
+  border-radius: 0;
+}
+
+button:focus {
+  outline: 1px dotted;
+  outline: 5px auto -webkit-focus-ring-color;
+}
+
+input,
+button,
+select,
+optgroup,
+textarea {
+  margin: 0;
+  font-family: inherit;
+  font-size: inherit;
+  line-height: inherit;
+}
+
+button,
+input {
+  overflow: visible;
+}
+
+button,
+select {
+  text-transform: none;
+}
+
+button,
+html [type="button"],
+[type="reset"],
+[type="submit"] {
+  -webkit-appearance: button;
+}
+
+button::-moz-focus-inner,
+[type="button"]::-moz-focus-inner,
+[type="reset"]::-moz-focus-inner,
+[type="submit"]::-moz-focus-inner {
+  padding: 0;
+  border-style: none;
+}
+
+input[type="radio"],
+input[type="checkbox"] {
+  -webkit-box-sizing: border-box;
+          box-sizing: border-box;
+  padding: 0;
+}
+
+input[type="date"],
+input[type="time"],
+input[type="datetime-local"],
+input[type="month"] {
+  -webkit-appearance: listbox;
+}
+
+textarea {
+  overflow: auto;
+  resize: vertical;
+}
+
+fieldset {
+  min-width: 0;
+  padding: 0;
+  margin: 0;
+  border: 0;
+}
+
+legend {
+  display: block;
+  width: 100%;
+  max-width: 100%;
+  padding: 0;
+  margin-bottom: .5rem;
+  font-size: 1.5rem;
+  line-height: inherit;
+  color: inherit;
+  white-space: normal;
+}
+
+progress {
+  vertical-align: baseline;
+}
+
+[type="number"]::-webkit-inner-spin-button,
+[type="number"]::-webkit-outer-spin-button {
+  height: auto;
+}
+
+[type="search"] {
+  outline-offset: -2px;
+  -webkit-appearance: none;
+}
+
+[type="search"]::-webkit-search-cancel-button,
+[type="search"]::-webkit-search-decoration {
+  -webkit-appearance: none;
+}
+
+::-webkit-file-upload-button {
+  font: inherit;
+  -webkit-appearance: button;
+}
+
+output {
+  display: inline-block;
+}
+
+summary {
+  display: list-item;
+  cursor: pointer;
+}
+
+template {
+  display: none;
+}
+
+[hidden] {
+  display: none !important;
+}
+
+h1, h2, h3, h4, h5, h6,
+.h1, .h2, .h3, .h4, .h5, .h6 {
+  margin-bottom: 0.5rem;
+  font-family: inherit;
+  font-weight: 500;
+  line-height: 1.2;
+  color: inherit;
+}
+
+h1, .h1 {
+  font-size: 2.5rem;
+}
+
+h2, .h2 {
+  font-size: 2rem;
+}
+
+h3, .h3 {
+  font-size: 1.75rem;
+}
+
+h4, .h4 {
+  font-size: 1.5rem;
+}
+
+h5, .h5 {
+  font-size: 1.25rem;
+}
+
+h6, .h6 {
+  font-size: 1rem;
+}
+
+.lead {
+  font-size: 1.25rem;
+  font-weight: 300;
+}
+
+.display-1 {
+  font-size: 6rem;
+  font-weight: 300;
+  line-height: 1.2;
+}
+
+.display-2 {
+  font-size: 5.5rem;
+  font-weight: 300;
+  line-height: 1.2;
+}
+
+.display-3 {
+  font-size: 4.5rem;
+  font-weight: 300;
+  line-height: 1.2;
+}
+
+.display-4 {
+  font-size: 3.5rem;
+  font-weight: 300;
+  line-height: 1.2;
+}
+
+hr {
+  margin-top: 1rem;
+  margin-bottom: 1rem;
+  border: 0;
+  border-top: 1px solid rgba(0, 0, 0, 0.1);
+}
+
+small,
+.small {
+  font-size: 80%;
+  font-weight: 400;
+}
+
+mark,
+.mark {
+  padding: 0.2em;
+  background-color: #fcf8e3;
+}
+
+.list-unstyled {
+  padding-left: 0;
+  list-style: none;
+}
+
+.list-inline {
+  padding-left: 0;
+  list-style: none;
+}
+
+.list-inline-item {
+  display: inline-block;
+}
+.list-inline-item:not(:last-child) {
+  margin-right: 0.5rem;
+}
+
+.initialism {
+  font-size: 90%;
+  text-transform: uppercase;
+}
+
+.blockquote {
+  margin-bottom: 1rem;
+  font-size: 1.25rem;
+}
+
+.blockquote-footer {
+  display: block;
+  font-size: 80%;
+  color: #6c757d;
+}
+.blockquote-footer::before {
+  content: "\2014 \00A0";
+}
+
+.img-fluid {
+  max-width: 100%;
+  height: auto;
+}
+
+.img-thumbnail {
+  padding: 0.25rem;
+  background-color: #fff;
+  border: 1px solid #dee2e6;
+  border-radius: 0.25rem;
+  max-width: 100%;
+  height: auto;
+}
+
+.figure {
+  display: inline-block;
+}
+
+.figure-img {
+  margin-bottom: 0.5rem;
+  line-height: 1;
+}
+
+.figure-caption {
+  font-size: 90%;
+  color: #6c757d;
+}
+
+code,
+kbd,
+pre,
+samp {
+  font-family: SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;
+}
+
+code {
+  font-size: 87.5%;
+  color: #e83e8c;
+  word-break: break-word;
+}
+a > code {
+  color: inherit;
+}
+
+kbd {
+  padding: 0.2rem 0.4rem;
+  font-size: 87.5%;
+  color: #fff;
+  background-color: #212529;
+  border-radius: 0.2rem;
+}
+kbd kbd {
+  padding: 0;
+  font-size: 100%;
+  font-weight: 700;
+}
+
+pre {
+  display: block;
+  font-size: 87.5%;
+  color: #212529;
+}
+pre code {
+  font-size: inherit;
+  color: inherit;
+  word-break: normal;
+}
+
+.pre-scrollable {
+  max-height: 340px;
+  overflow-y: scroll;
+}
+
+.container {
+  width: 100%;
+  padding-right: 15px;
+  padding-left: 15px;
+  margin-right: auto;
+  margin-left: auto;
+}
+@media (min-width: 576px) {
+  .container {
+    max-width: 540px;
+  }
+}
+@media (min-width: 768px) {
+  .container {
+    max-width: 720px;
+  }
+}
+@media (min-width: 992px) {
+  .container {
+    max-width: 960px;
+  }
+}
+@media (min-width: 1200px) {
+  .container {
+    max-width: 1140px;
+  }
+}
+
+.container-fluid {
+  width: 100%;
+  padding-right: 15px;
+  padding-left: 15px;
+  margin-right: auto;
+  margin-left: auto;
+}
+
+.row {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  margin-right: -15px;
+  margin-left: -15px;
+}
+
+.no-gutters {
+  margin-right: 0;
+  margin-left: 0;
+}
+.no-gutters > .col,
+.no-gutters > [class*="col-"] {
+  padding-right: 0;
+  padding-left: 0;
+}
+
+.col-1, .col-2, .col-3, .col-4, .col-5, .col-6, .col-7, .col-8, .col-9, .col-10, .col-11, .col-12, .col,
+.col-auto, .col-sm-1, .col-sm-2, .col-sm-3, .col-sm-4, .col-sm-5, .col-sm-6, .col-sm-7, .col-sm-8, .col-sm-9, .col-sm-10, .col-sm-11, .col-sm-12, .col-sm,
+.col-sm-auto, .col-md-1, .col-md-2, .col-md-3, .col-md-4, .col-md-5, .col-md-6, .col-md-7, .col-md-8, .col-md-9, .col-md-10, .col-md-11, .col-md-12, .col-md,
+.col-md-auto, .col-lg-1, .col-lg-2, .col-lg-3, .col-lg-4, .col-lg-5, .col-lg-6, .col-lg-7, .col-lg-8, .col-lg-9, .col-lg-10, .col-lg-11, .col-lg-12, .col-lg,
+.col-lg-auto, .col-xl-1, .col-xl-2, .col-xl-3, .col-xl-4, .col-xl-5, .col-xl-6, .col-xl-7, .col-xl-8, .col-xl-9, .col-xl-10, .col-xl-11, .col-xl-12, .col-xl,
+.col-xl-auto {
+  position: relative;
+  width: 100%;
+  min-height: 1px;
+  padding-right: 15px;
+  padding-left: 15px;
+}
+
+.col {
+  -ms-flex-preferred-size: 0;
+      flex-basis: 0;
+  -webkit-box-flex: 1;
+      -ms-flex-positive: 1;
+          flex-grow: 1;
+  max-width: 100%;
+}
+
+.col-auto {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 auto;
+          flex: 0 0 auto;
+  width: auto;
+  max-width: none;
+}
+
+.col-1 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 8.3333333333%;
+          flex: 0 0 8.3333333333%;
+  max-width: 8.3333333333%;
+}
+
+.col-2 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 16.6666666667%;
+          flex: 0 0 16.6666666667%;
+  max-width: 16.6666666667%;
+}
+
+.col-3 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 25%;
+          flex: 0 0 25%;
+  max-width: 25%;
+}
+
+.col-4 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 33.3333333333%;
+          flex: 0 0 33.3333333333%;
+  max-width: 33.3333333333%;
+}
+
+.col-5 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 41.6666666667%;
+          flex: 0 0 41.6666666667%;
+  max-width: 41.6666666667%;
+}
+
+.col-6 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 50%;
+          flex: 0 0 50%;
+  max-width: 50%;
+}
+
+.col-7 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 58.3333333333%;
+          flex: 0 0 58.3333333333%;
+  max-width: 58.3333333333%;
+}
+
+.col-8 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 66.6666666667%;
+          flex: 0 0 66.6666666667%;
+  max-width: 66.6666666667%;
+}
+
+.col-9 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 75%;
+          flex: 0 0 75%;
+  max-width: 75%;
+}
+
+.col-10 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 83.3333333333%;
+          flex: 0 0 83.3333333333%;
+  max-width: 83.3333333333%;
+}
+
+.col-11 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 91.6666666667%;
+          flex: 0 0 91.6666666667%;
+  max-width: 91.6666666667%;
+}
+
+.col-12 {
+  -webkit-box-flex: 0;
+      -ms-flex: 0 0 100%;
+          flex: 0 0 100%;
+  max-width: 100%;
+}
+
+.order-first {
+  -webkit-box-ordinal-group: 0;
+      -ms-flex-order: -1;
+          order: -1;
+}
+
+.order-last {
+  -webkit-box-ordinal-group: 14;
+      -ms-flex-order: 13;
+          order: 13;
+}
+
+.order-0 {
+  -webkit-box-ordinal-group: 1;
+      -ms-flex-order: 0;
+          order: 0;
+}
+
+.order-1 {
+  -webkit-box-ordinal-group: 2;
+      -ms-flex-order: 1;
+          order: 1;
+}
+
+.order-2 {
+  -webkit-box-ordinal-group: 3;
+      -ms-flex-order: 2;
+          order: 2;
+}
+
+.order-3 {
+  -webkit-box-ordinal-group: 4;
+      -ms-flex-order: 3;
+          order: 3;
+}
+
+.order-4 {
+  -webkit-box-ordinal-group: 5;
+      -ms-flex-order: 4;
+          order: 4;
+}
+
+.order-5 {
+  -webkit-box-ordinal-group: 6;
+      -ms-flex-order: 5;
+          order: 5;
+}
+
+.order-6 {
+  -webkit-box-ordinal-group: 7;
+      -ms-flex-order: 6;
+          order: 6;
+}
+
+.order-7 {
+  -webkit-box-ordinal-group: 8;
+      -ms-flex-order: 7;
+          order: 7;
+}
+
+.order-8 {
+  -webkit-box-ordinal-group: 9;
+      -ms-flex-order: 8;
+          order: 8;
+}
+
+.order-9 {
+  -webkit-box-ordinal-group: 10;
+      -ms-flex-order: 9;
+          order: 9;
+}
+
+.order-10 {
+  -webkit-box-ordinal-group: 11;
+      -ms-flex-order: 10;
+          order: 10;
+}
+
+.order-11 {
+  -webkit-box-ordinal-group: 12;
+      -ms-flex-order: 11;
+          order: 11;
+}
+
+.order-12 {
+  -webkit-box-ordinal-group: 13;
+      -ms-flex-order: 12;
+          order: 12;
+}
+
+.offset-1 {
+  margin-left: 8.3333333333%;
+}
+
+.offset-2 {
+  margin-left: 16.6666666667%;
+}
+
+.offset-3 {
+  margin-left: 25%;
+}
+
+.offset-4 {
+  margin-left: 33.3333333333%;
+}
+
+.offset-5 {
+  margin-left: 41.6666666667%;
+}
+
+.offset-6 {
+  margin-left: 50%;
+}
+
+.offset-7 {
+  margin-left: 58.3333333333%;
+}
+
+.offset-8 {
+  margin-left: 66.6666666667%;
+}
+
+.offset-9 {
+  margin-left: 75%;
+}
+
+.offset-10 {
+  margin-left: 83.3333333333%;
+}
+
+.offset-11 {
+  margin-left: 91.6666666667%;
+}
+
+@media (min-width: 576px) {
+  .col-sm {
+    -ms-flex-preferred-size: 0;
+        flex-basis: 0;
+    -webkit-box-flex: 1;
+        -ms-flex-positive: 1;
+            flex-grow: 1;
+    max-width: 100%;
+  }
+
+  .col-sm-auto {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 auto;
+            flex: 0 0 auto;
+    width: auto;
+    max-width: none;
+  }
+
+  .col-sm-1 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 8.3333333333%;
+            flex: 0 0 8.3333333333%;
+    max-width: 8.3333333333%;
+  }
+
+  .col-sm-2 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 16.6666666667%;
+            flex: 0 0 16.6666666667%;
+    max-width: 16.6666666667%;
+  }
+
+  .col-sm-3 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 25%;
+            flex: 0 0 25%;
+    max-width: 25%;
+  }
+
+  .col-sm-4 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 33.3333333333%;
+            flex: 0 0 33.3333333333%;
+    max-width: 33.3333333333%;
+  }
+
+  .col-sm-5 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 41.6666666667%;
+            flex: 0 0 41.6666666667%;
+    max-width: 41.6666666667%;
+  }
+
+  .col-sm-6 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 50%;
+            flex: 0 0 50%;
+    max-width: 50%;
+  }
+
+  .col-sm-7 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 58.3333333333%;
+            flex: 0 0 58.3333333333%;
+    max-width: 58.3333333333%;
+  }
+
+  .col-sm-8 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 66.6666666667%;
+            flex: 0 0 66.6666666667%;
+    max-width: 66.6666666667%;
+  }
+
+  .col-sm-9 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 75%;
+            flex: 0 0 75%;
+    max-width: 75%;
+  }
+
+  .col-sm-10 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 83.3333333333%;
+            flex: 0 0 83.3333333333%;
+    max-width: 83.3333333333%;
+  }
+
+  .col-sm-11 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 91.6666666667%;
+            flex: 0 0 91.6666666667%;
+    max-width: 91.6666666667%;
+  }
+
+  .col-sm-12 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 100%;
+            flex: 0 0 100%;
+    max-width: 100%;
+  }
+
+  .order-sm-first {
+    -webkit-box-ordinal-group: 0;
+        -ms-flex-order: -1;
+            order: -1;
+  }
+
+  .order-sm-last {
+    -webkit-box-ordinal-group: 14;
+        -ms-flex-order: 13;
+            order: 13;
+  }
+
+  .order-sm-0 {
+    -webkit-box-ordinal-group: 1;
+        -ms-flex-order: 0;
+            order: 0;
+  }
+
+  .order-sm-1 {
+    -webkit-box-ordinal-group: 2;
+        -ms-flex-order: 1;
+            order: 1;
+  }
+
+  .order-sm-2 {
+    -webkit-box-ordinal-group: 3;
+        -ms-flex-order: 2;
+            order: 2;
+  }
+
+  .order-sm-3 {
+    -webkit-box-ordinal-group: 4;
+        -ms-flex-order: 3;
+            order: 3;
+  }
+
+  .order-sm-4 {
+    -webkit-box-ordinal-group: 5;
+        -ms-flex-order: 4;
+            order: 4;
+  }
+
+  .order-sm-5 {
+    -webkit-box-ordinal-group: 6;
+        -ms-flex-order: 5;
+            order: 5;
+  }
+
+  .order-sm-6 {
+    -webkit-box-ordinal-group: 7;
+        -ms-flex-order: 6;
+            order: 6;
+  }
+
+  .order-sm-7 {
+    -webkit-box-ordinal-group: 8;
+        -ms-flex-order: 7;
+            order: 7;
+  }
+
+  .order-sm-8 {
+    -webkit-box-ordinal-group: 9;
+        -ms-flex-order: 8;
+            order: 8;
+  }
+
+  .order-sm-9 {
+    -webkit-box-ordinal-group: 10;
+        -ms-flex-order: 9;
+            order: 9;
+  }
+
+  .order-sm-10 {
+    -webkit-box-ordinal-group: 11;
+        -ms-flex-order: 10;
+            order: 10;
+  }
+
+  .order-sm-11 {
+    -webkit-box-ordinal-group: 12;
+        -ms-flex-order: 11;
+            order: 11;
+  }
+
+  .order-sm-12 {
+    -webkit-box-ordinal-group: 13;
+        -ms-flex-order: 12;
+            order: 12;
+  }
+
+  .offset-sm-0 {
+    margin-left: 0;
+  }
+
+  .offset-sm-1 {
+    margin-left: 8.3333333333%;
+  }
+
+  .offset-sm-2 {
+    margin-left: 16.6666666667%;
+  }
+
+  .offset-sm-3 {
+    margin-left: 25%;
+  }
+
+  .offset-sm-4 {
+    margin-left: 33.3333333333%;
+  }
+
+  .offset-sm-5 {
+    margin-left: 41.6666666667%;
+  }
+
+  .offset-sm-6 {
+    margin-left: 50%;
+  }
+
+  .offset-sm-7 {
+    margin-left: 58.3333333333%;
+  }
+
+  .offset-sm-8 {
+    margin-left: 66.6666666667%;
+  }
+
+  .offset-sm-9 {
+    margin-left: 75%;
+  }
+
+  .offset-sm-10 {
+    margin-left: 83.3333333333%;
+  }
+
+  .offset-sm-11 {
+    margin-left: 91.6666666667%;
+  }
+}
+@media (min-width: 768px) {
+  .col-md {
+    -ms-flex-preferred-size: 0;
+        flex-basis: 0;
+    -webkit-box-flex: 1;
+        -ms-flex-positive: 1;
+            flex-grow: 1;
+    max-width: 100%;
+  }
+
+  .col-md-auto {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 auto;
+            flex: 0 0 auto;
+    width: auto;
+    max-width: none;
+  }
+
+  .col-md-1 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 8.3333333333%;
+            flex: 0 0 8.3333333333%;
+    max-width: 8.3333333333%;
+  }
+
+  .col-md-2 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 16.6666666667%;
+            flex: 0 0 16.6666666667%;
+    max-width: 16.6666666667%;
+  }
+
+  .col-md-3 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 25%;
+            flex: 0 0 25%;
+    max-width: 25%;
+  }
+
+  .col-md-4 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 33.3333333333%;
+            flex: 0 0 33.3333333333%;
+    max-width: 33.3333333333%;
+  }
+
+  .col-md-5 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 41.6666666667%;
+            flex: 0 0 41.6666666667%;
+    max-width: 41.6666666667%;
+  }
+
+  .col-md-6 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 50%;
+            flex: 0 0 50%;
+    max-width: 50%;
+  }
+
+  .col-md-7 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 58.3333333333%;
+            flex: 0 0 58.3333333333%;
+    max-width: 58.3333333333%;
+  }
+
+  .col-md-8 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 66.6666666667%;
+            flex: 0 0 66.6666666667%;
+    max-width: 66.6666666667%;
+  }
+
+  .col-md-9 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 75%;
+            flex: 0 0 75%;
+    max-width: 75%;
+  }
+
+  .col-md-10 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 83.3333333333%;
+            flex: 0 0 83.3333333333%;
+    max-width: 83.3333333333%;
+  }
+
+  .col-md-11 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 91.6666666667%;
+            flex: 0 0 91.6666666667%;
+    max-width: 91.6666666667%;
+  }
+
+  .col-md-12 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 100%;
+            flex: 0 0 100%;
+    max-width: 100%;
+  }
+
+  .order-md-first {
+    -webkit-box-ordinal-group: 0;
+        -ms-flex-order: -1;
+            order: -1;
+  }
+
+  .order-md-last {
+    -webkit-box-ordinal-group: 14;
+        -ms-flex-order: 13;
+            order: 13;
+  }
+
+  .order-md-0 {
+    -webkit-box-ordinal-group: 1;
+        -ms-flex-order: 0;
+            order: 0;
+  }
+
+  .order-md-1 {
+    -webkit-box-ordinal-group: 2;
+        -ms-flex-order: 1;
+            order: 1;
+  }
+
+  .order-md-2 {
+    -webkit-box-ordinal-group: 3;
+        -ms-flex-order: 2;
+            order: 2;
+  }
+
+  .order-md-3 {
+    -webkit-box-ordinal-group: 4;
+        -ms-flex-order: 3;
+            order: 3;
+  }
+
+  .order-md-4 {
+    -webkit-box-ordinal-group: 5;
+        -ms-flex-order: 4;
+            order: 4;
+  }
+
+  .order-md-5 {
+    -webkit-box-ordinal-group: 6;
+        -ms-flex-order: 5;
+            order: 5;
+  }
+
+  .order-md-6 {
+    -webkit-box-ordinal-group: 7;
+        -ms-flex-order: 6;
+            order: 6;
+  }
+
+  .order-md-7 {
+    -webkit-box-ordinal-group: 8;
+        -ms-flex-order: 7;
+            order: 7;
+  }
+
+  .order-md-8 {
+    -webkit-box-ordinal-group: 9;
+        -ms-flex-order: 8;
+            order: 8;
+  }
+
+  .order-md-9 {
+    -webkit-box-ordinal-group: 10;
+        -ms-flex-order: 9;
+            order: 9;
+  }
+
+  .order-md-10 {
+    -webkit-box-ordinal-group: 11;
+        -ms-flex-order: 10;
+            order: 10;
+  }
+
+  .order-md-11 {
+    -webkit-box-ordinal-group: 12;
+        -ms-flex-order: 11;
+            order: 11;
+  }
+
+  .order-md-12 {
+    -webkit-box-ordinal-group: 13;
+        -ms-flex-order: 12;
+            order: 12;
+  }
+
+  .offset-md-0 {
+    margin-left: 0;
+  }
+
+  .offset-md-1 {
+    margin-left: 8.3333333333%;
+  }
+
+  .offset-md-2 {
+    margin-left: 16.6666666667%;
+  }
+
+  .offset-md-3 {
+    margin-left: 25%;
+  }
+
+  .offset-md-4 {
+    margin-left: 33.3333333333%;
+  }
+
+  .offset-md-5 {
+    margin-left: 41.6666666667%;
+  }
+
+  .offset-md-6 {
+    margin-left: 50%;
+  }
+
+  .offset-md-7 {
+    margin-left: 58.3333333333%;
+  }
+
+  .offset-md-8 {
+    margin-left: 66.6666666667%;
+  }
+
+  .offset-md-9 {
+    margin-left: 75%;
+  }
+
+  .offset-md-10 {
+    margin-left: 83.3333333333%;
+  }
+
+  .offset-md-11 {
+    margin-left: 91.6666666667%;
+  }
+}
+@media (min-width: 992px) {
+  .col-lg {
+    -ms-flex-preferred-size: 0;
+        flex-basis: 0;
+    -webkit-box-flex: 1;
+        -ms-flex-positive: 1;
+            flex-grow: 1;
+    max-width: 100%;
+  }
+
+  .col-lg-auto {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 auto;
+            flex: 0 0 auto;
+    width: auto;
+    max-width: none;
+  }
+
+  .col-lg-1 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 8.3333333333%;
+            flex: 0 0 8.3333333333%;
+    max-width: 8.3333333333%;
+  }
+
+  .col-lg-2 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 16.6666666667%;
+            flex: 0 0 16.6666666667%;
+    max-width: 16.6666666667%;
+  }
+
+  .col-lg-3 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 25%;
+            flex: 0 0 25%;
+    max-width: 25%;
+  }
+
+  .col-lg-4 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 33.3333333333%;
+            flex: 0 0 33.3333333333%;
+    max-width: 33.3333333333%;
+  }
+
+  .col-lg-5 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 41.6666666667%;
+            flex: 0 0 41.6666666667%;
+    max-width: 41.6666666667%;
+  }
+
+  .col-lg-6 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 50%;
+            flex: 0 0 50%;
+    max-width: 50%;
+  }
+
+  .col-lg-7 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 58.3333333333%;
+            flex: 0 0 58.3333333333%;
+    max-width: 58.3333333333%;
+  }
+
+  .col-lg-8 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 66.6666666667%;
+            flex: 0 0 66.6666666667%;
+    max-width: 66.6666666667%;
+  }
+
+  .col-lg-9 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 75%;
+            flex: 0 0 75%;
+    max-width: 75%;
+  }
+
+  .col-lg-10 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 83.3333333333%;
+            flex: 0 0 83.3333333333%;
+    max-width: 83.3333333333%;
+  }
+
+  .col-lg-11 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 91.6666666667%;
+            flex: 0 0 91.6666666667%;
+    max-width: 91.6666666667%;
+  }
+
+  .col-lg-12 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 100%;
+            flex: 0 0 100%;
+    max-width: 100%;
+  }
+
+  .order-lg-first {
+    -webkit-box-ordinal-group: 0;
+        -ms-flex-order: -1;
+            order: -1;
+  }
+
+  .order-lg-last {
+    -webkit-box-ordinal-group: 14;
+        -ms-flex-order: 13;
+            order: 13;
+  }
+
+  .order-lg-0 {
+    -webkit-box-ordinal-group: 1;
+        -ms-flex-order: 0;
+            order: 0;
+  }
+
+  .order-lg-1 {
+    -webkit-box-ordinal-group: 2;
+        -ms-flex-order: 1;
+            order: 1;
+  }
+
+  .order-lg-2 {
+    -webkit-box-ordinal-group: 3;
+        -ms-flex-order: 2;
+            order: 2;
+  }
+
+  .order-lg-3 {
+    -webkit-box-ordinal-group: 4;
+        -ms-flex-order: 3;
+            order: 3;
+  }
+
+  .order-lg-4 {
+    -webkit-box-ordinal-group: 5;
+        -ms-flex-order: 4;
+            order: 4;
+  }
+
+  .order-lg-5 {
+    -webkit-box-ordinal-group: 6;
+        -ms-flex-order: 5;
+            order: 5;
+  }
+
+  .order-lg-6 {
+    -webkit-box-ordinal-group: 7;
+        -ms-flex-order: 6;
+            order: 6;
+  }
+
+  .order-lg-7 {
+    -webkit-box-ordinal-group: 8;
+        -ms-flex-order: 7;
+            order: 7;
+  }
+
+  .order-lg-8 {
+    -webkit-box-ordinal-group: 9;
+        -ms-flex-order: 8;
+            order: 8;
+  }
+
+  .order-lg-9 {
+    -webkit-box-ordinal-group: 10;
+        -ms-flex-order: 9;
+            order: 9;
+  }
+
+  .order-lg-10 {
+    -webkit-box-ordinal-group: 11;
+        -ms-flex-order: 10;
+            order: 10;
+  }
+
+  .order-lg-11 {
+    -webkit-box-ordinal-group: 12;
+        -ms-flex-order: 11;
+            order: 11;
+  }
+
+  .order-lg-12 {
+    -webkit-box-ordinal-group: 13;
+        -ms-flex-order: 12;
+            order: 12;
+  }
+
+  .offset-lg-0 {
+    margin-left: 0;
+  }
+
+  .offset-lg-1 {
+    margin-left: 8.3333333333%;
+  }
+
+  .offset-lg-2 {
+    margin-left: 16.6666666667%;
+  }
+
+  .offset-lg-3 {
+    margin-left: 25%;
+  }
+
+  .offset-lg-4 {
+    margin-left: 33.3333333333%;
+  }
+
+  .offset-lg-5 {
+    margin-left: 41.6666666667%;
+  }
+
+  .offset-lg-6 {
+    margin-left: 50%;
+  }
+
+  .offset-lg-7 {
+    margin-left: 58.3333333333%;
+  }
+
+  .offset-lg-8 {
+    margin-left: 66.6666666667%;
+  }
+
+  .offset-lg-9 {
+    margin-left: 75%;
+  }
+
+  .offset-lg-10 {
+    margin-left: 83.3333333333%;
+  }
+
+  .offset-lg-11 {
+    margin-left: 91.6666666667%;
+  }
+}
+@media (min-width: 1200px) {
+  .col-xl {
+    -ms-flex-preferred-size: 0;
+        flex-basis: 0;
+    -webkit-box-flex: 1;
+        -ms-flex-positive: 1;
+            flex-grow: 1;
+    max-width: 100%;
+  }
+
+  .col-xl-auto {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 auto;
+            flex: 0 0 auto;
+    width: auto;
+    max-width: none;
+  }
+
+  .col-xl-1 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 8.3333333333%;
+            flex: 0 0 8.3333333333%;
+    max-width: 8.3333333333%;
+  }
+
+  .col-xl-2 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 16.6666666667%;
+            flex: 0 0 16.6666666667%;
+    max-width: 16.6666666667%;
+  }
+
+  .col-xl-3 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 25%;
+            flex: 0 0 25%;
+    max-width: 25%;
+  }
+
+  .col-xl-4 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 33.3333333333%;
+            flex: 0 0 33.3333333333%;
+    max-width: 33.3333333333%;
+  }
+
+  .col-xl-5 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 41.6666666667%;
+            flex: 0 0 41.6666666667%;
+    max-width: 41.6666666667%;
+  }
+
+  .col-xl-6 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 50%;
+            flex: 0 0 50%;
+    max-width: 50%;
+  }
+
+  .col-xl-7 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 58.3333333333%;
+            flex: 0 0 58.3333333333%;
+    max-width: 58.3333333333%;
+  }
+
+  .col-xl-8 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 66.6666666667%;
+            flex: 0 0 66.6666666667%;
+    max-width: 66.6666666667%;
+  }
+
+  .col-xl-9 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 75%;
+            flex: 0 0 75%;
+    max-width: 75%;
+  }
+
+  .col-xl-10 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 83.3333333333%;
+            flex: 0 0 83.3333333333%;
+    max-width: 83.3333333333%;
+  }
+
+  .col-xl-11 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 91.6666666667%;
+            flex: 0 0 91.6666666667%;
+    max-width: 91.6666666667%;
+  }
+
+  .col-xl-12 {
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 100%;
+            flex: 0 0 100%;
+    max-width: 100%;
+  }
+
+  .order-xl-first {
+    -webkit-box-ordinal-group: 0;
+        -ms-flex-order: -1;
+            order: -1;
+  }
+
+  .order-xl-last {
+    -webkit-box-ordinal-group: 14;
+        -ms-flex-order: 13;
+            order: 13;
+  }
+
+  .order-xl-0 {
+    -webkit-box-ordinal-group: 1;
+        -ms-flex-order: 0;
+            order: 0;
+  }
+
+  .order-xl-1 {
+    -webkit-box-ordinal-group: 2;
+        -ms-flex-order: 1;
+            order: 1;
+  }
+
+  .order-xl-2 {
+    -webkit-box-ordinal-group: 3;
+        -ms-flex-order: 2;
+            order: 2;
+  }
+
+  .order-xl-3 {
+    -webkit-box-ordinal-group: 4;
+        -ms-flex-order: 3;
+            order: 3;
+  }
+
+  .order-xl-4 {
+    -webkit-box-ordinal-group: 5;
+        -ms-flex-order: 4;
+            order: 4;
+  }
+
+  .order-xl-5 {
+    -webkit-box-ordinal-group: 6;
+        -ms-flex-order: 5;
+            order: 5;
+  }
+
+  .order-xl-6 {
+    -webkit-box-ordinal-group: 7;
+        -ms-flex-order: 6;
+            order: 6;
+  }
+
+  .order-xl-7 {
+    -webkit-box-ordinal-group: 8;
+        -ms-flex-order: 7;
+            order: 7;
+  }
+
+  .order-xl-8 {
+    -webkit-box-ordinal-group: 9;
+        -ms-flex-order: 8;
+            order: 8;
+  }
+
+  .order-xl-9 {
+    -webkit-box-ordinal-group: 10;
+        -ms-flex-order: 9;
+            order: 9;
+  }
+
+  .order-xl-10 {
+    -webkit-box-ordinal-group: 11;
+        -ms-flex-order: 10;
+            order: 10;
+  }
+
+  .order-xl-11 {
+    -webkit-box-ordinal-group: 12;
+        -ms-flex-order: 11;
+            order: 11;
+  }
+
+  .order-xl-12 {
+    -webkit-box-ordinal-group: 13;
+        -ms-flex-order: 12;
+            order: 12;
+  }
+
+  .offset-xl-0 {
+    margin-left: 0;
+  }
+
+  .offset-xl-1 {
+    margin-left: 8.3333333333%;
+  }
+
+  .offset-xl-2 {
+    margin-left: 16.6666666667%;
+  }
+
+  .offset-xl-3 {
+    margin-left: 25%;
+  }
+
+  .offset-xl-4 {
+    margin-left: 33.3333333333%;
+  }
+
+  .offset-xl-5 {
+    margin-left: 41.6666666667%;
+  }
+
+  .offset-xl-6 {
+    margin-left: 50%;
+  }
+
+  .offset-xl-7 {
+    margin-left: 58.3333333333%;
+  }
+
+  .offset-xl-8 {
+    margin-left: 66.6666666667%;
+  }
+
+  .offset-xl-9 {
+    margin-left: 75%;
+  }
+
+  .offset-xl-10 {
+    margin-left: 83.3333333333%;
+  }
+
+  .offset-xl-11 {
+    margin-left: 91.6666666667%;
+  }
+}
+.table {
+  width: 100%;
+  max-width: 100%;
+  margin-bottom: 1rem;
+  background-color: transparent;
+}
+.table th,
+.table td {
+  padding: 0.75rem;
+  vertical-align: top;
+  border-top: 1px solid #dee2e6;
+}
+.table thead th {
+  vertical-align: bottom;
+  border-bottom: 2px solid #dee2e6;
+}
+.table tbody + tbody {
+  border-top: 2px solid #dee2e6;
+}
+.table .table {
+  background-color: #fff;
+}
+
+.table-sm th,
+.table-sm td {
+  padding: 0.3rem;
+}
+
+.table-bordered {
+  border: 1px solid #dee2e6;
+}
+.table-bordered th,
+.table-bordered td {
+  border: 1px solid #dee2e6;
+}
+.table-bordered thead th,
+.table-bordered thead td {
+  border-bottom-width: 2px;
+}
+
+.table-striped tbody tr:nth-of-type(odd) {
+  background-color: rgba(0, 0, 0, 0.05);
+}
+
+.table-hover tbody tr:hover {
+  background-color: rgba(0, 0, 0, 0.075);
+}
+
+.table-primary,
+.table-primary > th,
+.table-primary > td {
+  background-color: #b8daff;
+}
+
+.table-hover .table-primary:hover {
+  background-color: #9fcdff;
+}
+.table-hover .table-primary:hover > td,
+.table-hover .table-primary:hover > th {
+  background-color: #9fcdff;
+}
+
+.table-secondary,
+.table-secondary > th,
+.table-secondary > td {
+  background-color: #d6d8db;
+}
+
+.table-hover .table-secondary:hover {
+  background-color: #c8cbcf;
+}
+.table-hover .table-secondary:hover > td,
+.table-hover .table-secondary:hover > th {
+  background-color: #c8cbcf;
+}
+
+.table-success,
+.table-success > th,
+.table-success > td {
+  background-color: #c3e6cb;
+}
+
+.table-hover .table-success:hover {
+  background-color: #b1dfbb;
+}
+.table-hover .table-success:hover > td,
+.table-hover .table-success:hover > th {
+  background-color: #b1dfbb;
+}
+
+.table-info,
+.table-info > th,
+.table-info > td {
+  background-color: #bee5eb;
+}
+
+.table-hover .table-info:hover {
+  background-color: #abdde5;
+}
+.table-hover .table-info:hover > td,
+.table-hover .table-info:hover > th {
+  background-color: #abdde5;
+}
+
+.table-warning,
+.table-warning > th,
+.table-warning > td {
+  background-color: #ffeeba;
+}
+
+.table-hover .table-warning:hover {
+  background-color: #ffe8a1;
+}
+.table-hover .table-warning:hover > td,
+.table-hover .table-warning:hover > th {
+  background-color: #ffe8a1;
+}
+
+.table-danger,
+.table-danger > th,
+.table-danger > td {
+  background-color: #f5c6cb;
+}
+
+.table-hover .table-danger:hover {
+  background-color: #f1b0b7;
+}
+.table-hover .table-danger:hover > td,
+.table-hover .table-danger:hover > th {
+  background-color: #f1b0b7;
+}
+
+.table-light,
+.table-light > th,
+.table-light > td {
+  background-color: #fdfdfe;
+}
+
+.table-hover .table-light:hover {
+  background-color: #ececf6;
+}
+.table-hover .table-light:hover > td,
+.table-hover .table-light:hover > th {
+  background-color: #ececf6;
+}
+
+.table-dark,
+.table-dark > th,
+.table-dark > td {
+  background-color: #c6c8ca;
+}
+
+.table-hover .table-dark:hover {
+  background-color: #b9bbbe;
+}
+.table-hover .table-dark:hover > td,
+.table-hover .table-dark:hover > th {
+  background-color: #b9bbbe;
+}
+
+.table-active,
+.table-active > th,
+.table-active > td {
+  background-color: rgba(0, 0, 0, 0.075);
+}
+
+.table-hover .table-active:hover {
+  background-color: rgba(0, 0, 0, 0.075);
+}
+.table-hover .table-active:hover > td,
+.table-hover .table-active:hover > th {
+  background-color: rgba(0, 0, 0, 0.075);
+}
+
+.table .thead-dark th {
+  color: #fff;
+  background-color: #212529;
+  border-color: #32383e;
+}
+.table .thead-light th {
+  color: #495057;
+  background-color: #e9ecef;
+  border-color: #dee2e6;
+}
+
+.table-dark {
+  color: #fff;
+  background-color: #212529;
+}
+.table-dark th,
+.table-dark td,
+.table-dark thead th {
+  border-color: #32383e;
+}
+.table-dark.table-bordered {
+  border: 0;
+}
+.table-dark.table-striped tbody tr:nth-of-type(odd) {
+  background-color: rgba(255, 255, 255, 0.05);
+}
+.table-dark.table-hover tbody tr:hover {
+  background-color: rgba(255, 255, 255, 0.075);
+}
+
+@media (max-width: 575.98px) {
+  .table-responsive-sm {
+    display: block;
+    width: 100%;
+    overflow-x: auto;
+    -webkit-overflow-scrolling: touch;
+    -ms-overflow-style: -ms-autohiding-scrollbar;
+  }
+  .table-responsive-sm > .table-bordered {
+    border: 0;
+  }
+}
+@media (max-width: 767.98px) {
+  .table-responsive-md {
+    display: block;
+    width: 100%;
+    overflow-x: auto;
+    -webkit-overflow-scrolling: touch;
+    -ms-overflow-style: -ms-autohiding-scrollbar;
+  }
+  .table-responsive-md > .table-bordered {
+    border: 0;
+  }
+}
+@media (max-width: 991.98px) {
+  .table-responsive-lg {
+    display: block;
+    width: 100%;
+    overflow-x: auto;
+    -webkit-overflow-scrolling: touch;
+    -ms-overflow-style: -ms-autohiding-scrollbar;
+  }
+  .table-responsive-lg > .table-bordered {
+    border: 0;
+  }
+}
+@media (max-width: 1199.98px) {
+  .table-responsive-xl {
+    display: block;
+    width: 100%;
+    overflow-x: auto;
+    -webkit-overflow-scrolling: touch;
+    -ms-overflow-style: -ms-autohiding-scrollbar;
+  }
+  .table-responsive-xl > .table-bordered {
+    border: 0;
+  }
+}
+.table-responsive {
+  display: block;
+  width: 100%;
+  overflow-x: auto;
+  -webkit-overflow-scrolling: touch;
+  -ms-overflow-style: -ms-autohiding-scrollbar;
+}
+.table-responsive > .table-bordered {
+  border: 0;
+}
+
+.form-control {
+  display: block;
+  width: 100%;
+  padding: 0.375rem 0.75rem;
+  font-size: 1rem;
+  line-height: 1.5;
+  color: #495057;
+  background-color: #fff;
+  background-clip: padding-box;
+  border: 1px solid #ced4da;
+  border-radius: 0.25rem;
+  -webkit-transition: border-color 0.15s ease-in-out, -webkit-box-shadow 0.15s ease-in-out;
+  transition: border-color 0.15s ease-in-out, -webkit-box-shadow 0.15s ease-in-out;
+  transition: border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;
+  transition: border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out, -webkit-box-shadow 0.15s ease-in-out;
+}
+.form-control::-ms-expand {
+  background-color: transparent;
+  border: 0;
+}
+.form-control:focus {
+  color: #495057;
+  background-color: #fff;
+  border-color: #80bdff;
+  outline: 0;
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+}
+.form-control::-webkit-input-placeholder {
+  color: #6c757d;
+  opacity: 1;
+}
+.form-control::-moz-placeholder {
+  color: #6c757d;
+  opacity: 1;
+}
+.form-control:-ms-input-placeholder {
+  color: #6c757d;
+  opacity: 1;
+}
+.form-control::-ms-input-placeholder {
+  color: #6c757d;
+  opacity: 1;
+}
+.form-control::placeholder {
+  color: #6c757d;
+  opacity: 1;
+}
+.form-control:disabled, .form-control[readonly] {
+  background-color: #e9ecef;
+  opacity: 1;
+}
+
+select.form-control:not([size]):not([multiple]) {
+  height: calc(2.25rem + 2px);
+}
+select.form-control:focus::-ms-value {
+  color: #495057;
+  background-color: #fff;
+}
+
+.form-control-file,
+.form-control-range {
+  display: block;
+  width: 100%;
+}
+
+.col-form-label {
+  padding-top: calc(0.375rem + 1px);
+  padding-bottom: calc(0.375rem + 1px);
+  margin-bottom: 0;
+  font-size: inherit;
+  line-height: 1.5;
+}
+
+.col-form-label-lg {
+  padding-top: calc(0.5rem + 1px);
+  padding-bottom: calc(0.5rem + 1px);
+  font-size: 1.25rem;
+  line-height: 1.5;
+}
+
+.col-form-label-sm {
+  padding-top: calc(0.25rem + 1px);
+  padding-bottom: calc(0.25rem + 1px);
+  font-size: 0.875rem;
+  line-height: 1.5;
+}
+
+.form-control-plaintext {
+  display: block;
+  width: 100%;
+  padding-top: 0.375rem;
+  padding-bottom: 0.375rem;
+  margin-bottom: 0;
+  line-height: 1.5;
+  background-color: transparent;
+  border: solid transparent;
+  border-width: 1px 0;
+}
+.form-control-plaintext.form-control-sm, .input-group-sm > .form-control-plaintext.form-control,
+.input-group-sm > .input-group-prepend > .form-control-plaintext.input-group-text,
+.input-group-sm > .input-group-append > .form-control-plaintext.input-group-text,
+.input-group-sm > .input-group-prepend > .form-control-plaintext.btn,
+.input-group-sm > .input-group-append > .form-control-plaintext.btn, .form-control-plaintext.form-control-lg, .input-group-lg > .form-control-plaintext.form-control,
+.input-group-lg > .input-group-prepend > .form-control-plaintext.input-group-text,
+.input-group-lg > .input-group-append > .form-control-plaintext.input-group-text,
+.input-group-lg > .input-group-prepend > .form-control-plaintext.btn,
+.input-group-lg > .input-group-append > .form-control-plaintext.btn {
+  padding-right: 0;
+  padding-left: 0;
+}
+
+.form-control-sm, .input-group-sm > .form-control,
+.input-group-sm > .input-group-prepend > .input-group-text,
+.input-group-sm > .input-group-append > .input-group-text,
+.input-group-sm > .input-group-prepend > .btn,
+.input-group-sm > .input-group-append > .btn {
+  padding: 0.25rem 0.5rem;
+  font-size: 0.875rem;
+  line-height: 1.5;
+  border-radius: 0.2rem;
+}
+
+select.form-control-sm:not([size]):not([multiple]), .input-group-sm > select.form-control:not([size]):not([multiple]),
+.input-group-sm > .input-group-prepend > select.input-group-text:not([size]):not([multiple]),
+.input-group-sm > .input-group-append > select.input-group-text:not([size]):not([multiple]),
+.input-group-sm > .input-group-prepend > select.btn:not([size]):not([multiple]),
+.input-group-sm > .input-group-append > select.btn:not([size]):not([multiple]) {
+  height: calc(1.8125rem + 2px);
+}
+
+.form-control-lg, .input-group-lg > .form-control,
+.input-group-lg > .input-group-prepend > .input-group-text,
+.input-group-lg > .input-group-append > .input-group-text,
+.input-group-lg > .input-group-prepend > .btn,
+.input-group-lg > .input-group-append > .btn {
+  padding: 0.5rem 1rem;
+  font-size: 1.25rem;
+  line-height: 1.5;
+  border-radius: 0.3rem;
+}
+
+select.form-control-lg:not([size]):not([multiple]), .input-group-lg > select.form-control:not([size]):not([multiple]),
+.input-group-lg > .input-group-prepend > select.input-group-text:not([size]):not([multiple]),
+.input-group-lg > .input-group-append > select.input-group-text:not([size]):not([multiple]),
+.input-group-lg > .input-group-prepend > select.btn:not([size]):not([multiple]),
+.input-group-lg > .input-group-append > select.btn:not([size]):not([multiple]) {
+  height: calc(2.875rem + 2px);
+}
+
+.form-group {
+  margin-bottom: 1rem;
+}
+
+.form-text {
+  display: block;
+  margin-top: 0.25rem;
+}
+
+.form-row {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  margin-right: -5px;
+  margin-left: -5px;
+}
+.form-row > .col,
+.form-row > [class*="col-"] {
+  padding-right: 5px;
+  padding-left: 5px;
+}
+
+.form-check {
+  position: relative;
+  display: block;
+  padding-left: 1.25rem;
+}
+
+.form-check-input {
+  position: absolute;
+  margin-top: 0.3rem;
+  margin-left: -1.25rem;
+}
+.form-check-input:disabled ~ .form-check-label {
+  color: #6c757d;
+}
+
+.form-check-label {
+  margin-bottom: 0;
+}
+
+.form-check-inline {
+  display: -webkit-inline-box;
+  display: -ms-inline-flexbox;
+  display: inline-flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  padding-left: 0;
+  margin-right: 0.75rem;
+}
+.form-check-inline .form-check-input {
+  position: static;
+  margin-top: 0;
+  margin-right: 0.3125rem;
+  margin-left: 0;
+}
+
+.valid-feedback {
+  display: none;
+  width: 100%;
+  margin-top: 0.25rem;
+  font-size: 80%;
+  color: #28a745;
+}
+
+.valid-tooltip {
+  position: absolute;
+  top: 100%;
+  z-index: 5;
+  display: none;
+  max-width: 100%;
+  padding: .5rem;
+  margin-top: .1rem;
+  font-size: .875rem;
+  line-height: 1;
+  color: #fff;
+  background-color: rgba(40, 167, 69, 0.8);
+  border-radius: .2rem;
+}
+
+.was-validated .form-control:valid, .form-control.is-valid,
+.was-validated .custom-select:valid,
+.custom-select.is-valid {
+  border-color: #28a745;
+}
+.was-validated .form-control:valid:focus, .form-control.is-valid:focus,
+.was-validated .custom-select:valid:focus,
+.custom-select.is-valid:focus {
+  border-color: #28a745;
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.25);
+}
+.was-validated .form-control:valid ~ .valid-feedback,
+.was-validated .form-control:valid ~ .valid-tooltip, .form-control.is-valid ~ .valid-feedback,
+.form-control.is-valid ~ .valid-tooltip,
+.was-validated .custom-select:valid ~ .valid-feedback,
+.was-validated .custom-select:valid ~ .valid-tooltip,
+.custom-select.is-valid ~ .valid-feedback,
+.custom-select.is-valid ~ .valid-tooltip {
+  display: block;
+}
+
+.was-validated .form-check-input:valid ~ .form-check-label, .form-check-input.is-valid ~ .form-check-label {
+  color: #28a745;
+}
+.was-validated .form-check-input:valid ~ .valid-feedback,
+.was-validated .form-check-input:valid ~ .valid-tooltip, .form-check-input.is-valid ~ .valid-feedback,
+.form-check-input.is-valid ~ .valid-tooltip {
+  display: block;
+}
+
+.was-validated .custom-control-input:valid ~ .custom-control-label, .custom-control-input.is-valid ~ .custom-control-label {
+  color: #28a745;
+}
+.was-validated .custom-control-input:valid ~ .custom-control-label::before, .custom-control-input.is-valid ~ .custom-control-label::before {
+  background-color: #71dd8a;
+}
+.was-validated .custom-control-input:valid ~ .valid-feedback,
+.was-validated .custom-control-input:valid ~ .valid-tooltip, .custom-control-input.is-valid ~ .valid-feedback,
+.custom-control-input.is-valid ~ .valid-tooltip {
+  display: block;
+}
+.was-validated .custom-control-input:valid:checked ~ .custom-control-label::before, .custom-control-input.is-valid:checked ~ .custom-control-label::before {
+  background-color: #34ce57;
+}
+.was-validated .custom-control-input:valid:focus ~ .custom-control-label::before, .custom-control-input.is-valid:focus ~ .custom-control-label::before {
+  -webkit-box-shadow: 0 0 0 1px #fff, 0 0 0 0.2rem rgba(40, 167, 69, 0.25);
+          box-shadow: 0 0 0 1px #fff, 0 0 0 0.2rem rgba(40, 167, 69, 0.25);
+}
+
+.was-validated .custom-file-input:valid ~ .custom-file-label, .custom-file-input.is-valid ~ .custom-file-label {
+  border-color: #28a745;
+}
+.was-validated .custom-file-input:valid ~ .custom-file-label::before, .custom-file-input.is-valid ~ .custom-file-label::before {
+  border-color: inherit;
+}
+.was-validated .custom-file-input:valid ~ .valid-feedback,
+.was-validated .custom-file-input:valid ~ .valid-tooltip, .custom-file-input.is-valid ~ .valid-feedback,
+.custom-file-input.is-valid ~ .valid-tooltip {
+  display: block;
+}
+.was-validated .custom-file-input:valid:focus ~ .custom-file-label, .custom-file-input.is-valid:focus ~ .custom-file-label {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.25);
+}
+
+.invalid-feedback {
+  display: none;
+  width: 100%;
+  margin-top: 0.25rem;
+  font-size: 80%;
+  color: #dc3545;
+}
+
+.invalid-tooltip {
+  position: absolute;
+  top: 100%;
+  z-index: 5;
+  display: none;
+  max-width: 100%;
+  padding: .5rem;
+  margin-top: .1rem;
+  font-size: .875rem;
+  line-height: 1;
+  color: #fff;
+  background-color: rgba(220, 53, 69, 0.8);
+  border-radius: .2rem;
+}
+
+.was-validated .form-control:invalid, .form-control.is-invalid,
+.was-validated .custom-select:invalid,
+.custom-select.is-invalid {
+  border-color: #dc3545;
+}
+.was-validated .form-control:invalid:focus, .form-control.is-invalid:focus,
+.was-validated .custom-select:invalid:focus,
+.custom-select.is-invalid:focus {
+  border-color: #dc3545;
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.25);
+}
+.was-validated .form-control:invalid ~ .invalid-feedback,
+.was-validated .form-control:invalid ~ .invalid-tooltip, .form-control.is-invalid ~ .invalid-feedback,
+.form-control.is-invalid ~ .invalid-tooltip,
+.was-validated .custom-select:invalid ~ .invalid-feedback,
+.was-validated .custom-select:invalid ~ .invalid-tooltip,
+.custom-select.is-invalid ~ .invalid-feedback,
+.custom-select.is-invalid ~ .invalid-tooltip {
+  display: block;
+}
+
+.was-validated .form-check-input:invalid ~ .form-check-label, .form-check-input.is-invalid ~ .form-check-label {
+  color: #dc3545;
+}
+.was-validated .form-check-input:invalid ~ .invalid-feedback,
+.was-validated .form-check-input:invalid ~ .invalid-tooltip, .form-check-input.is-invalid ~ .invalid-feedback,
+.form-check-input.is-invalid ~ .invalid-tooltip {
+  display: block;
+}
+
+.was-validated .custom-control-input:invalid ~ .custom-control-label, .custom-control-input.is-invalid ~ .custom-control-label {
+  color: #dc3545;
+}
+.was-validated .custom-control-input:invalid ~ .custom-control-label::before, .custom-control-input.is-invalid ~ .custom-control-label::before {
+  background-color: #efa2a9;
+}
+.was-validated .custom-control-input:invalid ~ .invalid-feedback,
+.was-validated .custom-control-input:invalid ~ .invalid-tooltip, .custom-control-input.is-invalid ~ .invalid-feedback,
+.custom-control-input.is-invalid ~ .invalid-tooltip {
+  display: block;
+}
+.was-validated .custom-control-input:invalid:checked ~ .custom-control-label::before, .custom-control-input.is-invalid:checked ~ .custom-control-label::before {
+  background-color: #e4606d;
+}
+.was-validated .custom-control-input:invalid:focus ~ .custom-control-label::before, .custom-control-input.is-invalid:focus ~ .custom-control-label::before {
+  -webkit-box-shadow: 0 0 0 1px #fff, 0 0 0 0.2rem rgba(220, 53, 69, 0.25);
+          box-shadow: 0 0 0 1px #fff, 0 0 0 0.2rem rgba(220, 53, 69, 0.25);
+}
+
+.was-validated .custom-file-input:invalid ~ .custom-file-label, .custom-file-input.is-invalid ~ .custom-file-label {
+  border-color: #dc3545;
+}
+.was-validated .custom-file-input:invalid ~ .custom-file-label::before, .custom-file-input.is-invalid ~ .custom-file-label::before {
+  border-color: inherit;
+}
+.was-validated .custom-file-input:invalid ~ .invalid-feedback,
+.was-validated .custom-file-input:invalid ~ .invalid-tooltip, .custom-file-input.is-invalid ~ .invalid-feedback,
+.custom-file-input.is-invalid ~ .invalid-tooltip {
+  display: block;
+}
+.was-validated .custom-file-input:invalid:focus ~ .custom-file-label, .custom-file-input.is-invalid:focus ~ .custom-file-label {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.25);
+}
+
+.form-inline {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: horizontal;
+  -webkit-box-direction: normal;
+      -ms-flex-flow: row wrap;
+          flex-flow: row wrap;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+}
+.form-inline .form-check {
+  width: 100%;
+}
+@media (min-width: 576px) {
+  .form-inline label {
+    display: -webkit-box;
+    display: -ms-flexbox;
+    display: flex;
+    -webkit-box-align: center;
+        -ms-flex-align: center;
+            align-items: center;
+    -webkit-box-pack: center;
+        -ms-flex-pack: center;
+            justify-content: center;
+    margin-bottom: 0;
+  }
+  .form-inline .form-group {
+    display: -webkit-box;
+    display: -ms-flexbox;
+    display: flex;
+    -webkit-box-flex: 0;
+        -ms-flex: 0 0 auto;
+            flex: 0 0 auto;
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-flow: row wrap;
+            flex-flow: row wrap;
+    -webkit-box-align: center;
+        -ms-flex-align: center;
+            align-items: center;
+    margin-bottom: 0;
+  }
+  .form-inline .form-control {
+    display: inline-block;
+    width: auto;
+    vertical-align: middle;
+  }
+  .form-inline .form-control-plaintext {
+    display: inline-block;
+  }
+  .form-inline .input-group {
+    width: auto;
+  }
+  .form-inline .form-check {
+    display: -webkit-box;
+    display: -ms-flexbox;
+    display: flex;
+    -webkit-box-align: center;
+        -ms-flex-align: center;
+            align-items: center;
+    -webkit-box-pack: center;
+        -ms-flex-pack: center;
+            justify-content: center;
+    width: auto;
+    padding-left: 0;
+  }
+  .form-inline .form-check-input {
+    position: relative;
+    margin-top: 0;
+    margin-right: 0.25rem;
+    margin-left: 0;
+  }
+  .form-inline .custom-control {
+    -webkit-box-align: center;
+        -ms-flex-align: center;
+            align-items: center;
+    -webkit-box-pack: center;
+        -ms-flex-pack: center;
+            justify-content: center;
+  }
+  .form-inline .custom-control-label {
+    margin-bottom: 0;
+  }
+}
+
+.btn {
+  display: inline-block;
+  font-weight: 400;
+  text-align: center;
+  white-space: nowrap;
+  vertical-align: middle;
+  -webkit-user-select: none;
+     -moz-user-select: none;
+      -ms-user-select: none;
+          user-select: none;
+  border: 1px solid transparent;
+  padding: 0.375rem 0.75rem;
+  font-size: 1rem;
+  line-height: 1.5;
+  border-radius: 0.25rem;
+  -webkit-transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, -webkit-box-shadow 0.15s ease-in-out;
+  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, -webkit-box-shadow 0.15s ease-in-out;
+  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;
+  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out, -webkit-box-shadow 0.15s ease-in-out;
+}
+.btn:hover, .btn:focus {
+  text-decoration: none;
+}
+.btn:focus, .btn.focus {
+  outline: 0;
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+}
+.btn.disabled, .btn:disabled {
+  opacity: 0.65;
+}
+.btn:not(:disabled):not(.disabled) {
+  cursor: pointer;
+}
+.btn:not(:disabled):not(.disabled):active, .btn:not(:disabled):not(.disabled).active {
+  background-image: none;
+}
+
+a.btn.disabled,
+fieldset:disabled a.btn {
+  pointer-events: none;
+}
+
+.btn-primary {
+  color: #fff;
+  background-color: #007bff;
+  border-color: #007bff;
+}
+.btn-primary:hover {
+  color: #fff;
+  background-color: #0069d9;
+  border-color: #0062cc;
+}
+.btn-primary:focus, .btn-primary.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+}
+.btn-primary.disabled, .btn-primary:disabled {
+  color: #fff;
+  background-color: #007bff;
+  border-color: #007bff;
+}
+.btn-primary:not(:disabled):not(.disabled):active, .btn-primary:not(:disabled):not(.disabled).active, .show > .btn-primary.dropdown-toggle {
+  color: #fff;
+  background-color: #0062cc;
+  border-color: #005cbf;
+}
+.btn-primary:not(:disabled):not(.disabled):active:focus, .btn-primary:not(:disabled):not(.disabled).active:focus, .show > .btn-primary.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+}
+
+.btn-secondary {
+  color: #fff;
+  background-color: #6c757d;
+  border-color: #6c757d;
+}
+.btn-secondary:hover {
+  color: #fff;
+  background-color: #5a6268;
+  border-color: #545b62;
+}
+.btn-secondary:focus, .btn-secondary.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+}
+.btn-secondary.disabled, .btn-secondary:disabled {
+  color: #fff;
+  background-color: #6c757d;
+  border-color: #6c757d;
+}
+.btn-secondary:not(:disabled):not(.disabled):active, .btn-secondary:not(:disabled):not(.disabled).active, .show > .btn-secondary.dropdown-toggle {
+  color: #fff;
+  background-color: #545b62;
+  border-color: #4e555b;
+}
+.btn-secondary:not(:disabled):not(.disabled):active:focus, .btn-secondary:not(:disabled):not(.disabled).active:focus, .show > .btn-secondary.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+}
+
+.btn-success {
+  color: #fff;
+  background-color: #28a745;
+  border-color: #28a745;
+}
+.btn-success:hover {
+  color: #fff;
+  background-color: #218838;
+  border-color: #1e7e34;
+}
+.btn-success:focus, .btn-success.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+}
+.btn-success.disabled, .btn-success:disabled {
+  color: #fff;
+  background-color: #28a745;
+  border-color: #28a745;
+}
+.btn-success:not(:disabled):not(.disabled):active, .btn-success:not(:disabled):not(.disabled).active, .show > .btn-success.dropdown-toggle {
+  color: #fff;
+  background-color: #1e7e34;
+  border-color: #1c7430;
+}
+.btn-success:not(:disabled):not(.disabled):active:focus, .btn-success:not(:disabled):not(.disabled).active:focus, .show > .btn-success.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+}
+
+.btn-info {
+  color: #fff;
+  background-color: #17a2b8;
+  border-color: #17a2b8;
+}
+.btn-info:hover {
+  color: #fff;
+  background-color: #138496;
+  border-color: #117a8b;
+}
+.btn-info:focus, .btn-info.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+}
+.btn-info.disabled, .btn-info:disabled {
+  color: #fff;
+  background-color: #17a2b8;
+  border-color: #17a2b8;
+}
+.btn-info:not(:disabled):not(.disabled):active, .btn-info:not(:disabled):not(.disabled).active, .show > .btn-info.dropdown-toggle {
+  color: #fff;
+  background-color: #117a8b;
+  border-color: #10707f;
+}
+.btn-info:not(:disabled):not(.disabled):active:focus, .btn-info:not(:disabled):not(.disabled).active:focus, .show > .btn-info.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+}
+
+.btn-warning {
+  color: #212529;
+  background-color: #ffc107;
+  border-color: #ffc107;
+}
+.btn-warning:hover {
+  color: #212529;
+  background-color: #e0a800;
+  border-color: #d39e00;
+}
+.btn-warning:focus, .btn-warning.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+}
+.btn-warning.disabled, .btn-warning:disabled {
+  color: #212529;
+  background-color: #ffc107;
+  border-color: #ffc107;
+}
+.btn-warning:not(:disabled):not(.disabled):active, .btn-warning:not(:disabled):not(.disabled).active, .show > .btn-warning.dropdown-toggle {
+  color: #212529;
+  background-color: #d39e00;
+  border-color: #c69500;
+}
+.btn-warning:not(:disabled):not(.disabled):active:focus, .btn-warning:not(:disabled):not(.disabled).active:focus, .show > .btn-warning.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+}
+
+.btn-danger {
+  color: #fff;
+  background-color: #dc3545;
+  border-color: #dc3545;
+}
+.btn-danger:hover {
+  color: #fff;
+  background-color: #c82333;
+  border-color: #bd2130;
+}
+.btn-danger:focus, .btn-danger.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+}
+.btn-danger.disabled, .btn-danger:disabled {
+  color: #fff;
+  background-color: #dc3545;
+  border-color: #dc3545;
+}
+.btn-danger:not(:disabled):not(.disabled):active, .btn-danger:not(:disabled):not(.disabled).active, .show > .btn-danger.dropdown-toggle {
+  color: #fff;
+  background-color: #bd2130;
+  border-color: #b21f2d;
+}
+.btn-danger:not(:disabled):not(.disabled):active:focus, .btn-danger:not(:disabled):not(.disabled).active:focus, .show > .btn-danger.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+}
+
+.btn-light {
+  color: #212529;
+  background-color: #f8f9fa;
+  border-color: #f8f9fa;
+}
+.btn-light:hover {
+  color: #212529;
+  background-color: #e2e6ea;
+  border-color: #dae0e5;
+}
+.btn-light:focus, .btn-light.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+}
+.btn-light.disabled, .btn-light:disabled {
+  color: #212529;
+  background-color: #f8f9fa;
+  border-color: #f8f9fa;
+}
+.btn-light:not(:disabled):not(.disabled):active, .btn-light:not(:disabled):not(.disabled).active, .show > .btn-light.dropdown-toggle {
+  color: #212529;
+  background-color: #dae0e5;
+  border-color: #d3d9df;
+}
+.btn-light:not(:disabled):not(.disabled):active:focus, .btn-light:not(:disabled):not(.disabled).active:focus, .show > .btn-light.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+}
+
+.btn-dark {
+  color: #fff;
+  background-color: #343a40;
+  border-color: #343a40;
+}
+.btn-dark:hover {
+  color: #fff;
+  background-color: #23272b;
+  border-color: #1d2124;
+}
+.btn-dark:focus, .btn-dark.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+}
+.btn-dark.disabled, .btn-dark:disabled {
+  color: #fff;
+  background-color: #343a40;
+  border-color: #343a40;
+}
+.btn-dark:not(:disabled):not(.disabled):active, .btn-dark:not(:disabled):not(.disabled).active, .show > .btn-dark.dropdown-toggle {
+  color: #fff;
+  background-color: #1d2124;
+  border-color: #171a1d;
+}
+.btn-dark:not(:disabled):not(.disabled):active:focus, .btn-dark:not(:disabled):not(.disabled).active:focus, .show > .btn-dark.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+}
+
+.btn-outline-primary {
+  color: #007bff;
+  background-color: transparent;
+  background-image: none;
+  border-color: #007bff;
+}
+.btn-outline-primary:hover {
+  color: #fff;
+  background-color: #007bff;
+  border-color: #007bff;
+}
+.btn-outline-primary:focus, .btn-outline-primary.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+}
+.btn-outline-primary.disabled, .btn-outline-primary:disabled {
+  color: #007bff;
+  background-color: transparent;
+}
+.btn-outline-primary:not(:disabled):not(.disabled):active, .btn-outline-primary:not(:disabled):not(.disabled).active, .show > .btn-outline-primary.dropdown-toggle {
+  color: #fff;
+  background-color: #007bff;
+  border-color: #007bff;
+}
+.btn-outline-primary:not(:disabled):not(.disabled):active:focus, .btn-outline-primary:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-primary.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.5);
+}
+
+.btn-outline-secondary {
+  color: #6c757d;
+  background-color: transparent;
+  background-image: none;
+  border-color: #6c757d;
+}
+.btn-outline-secondary:hover {
+  color: #fff;
+  background-color: #6c757d;
+  border-color: #6c757d;
+}
+.btn-outline-secondary:focus, .btn-outline-secondary.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+}
+.btn-outline-secondary.disabled, .btn-outline-secondary:disabled {
+  color: #6c757d;
+  background-color: transparent;
+}
+.btn-outline-secondary:not(:disabled):not(.disabled):active, .btn-outline-secondary:not(:disabled):not(.disabled).active, .show > .btn-outline-secondary.dropdown-toggle {
+  color: #fff;
+  background-color: #6c757d;
+  border-color: #6c757d;
+}
+.btn-outline-secondary:not(:disabled):not(.disabled):active:focus, .btn-outline-secondary:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-secondary.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(108, 117, 125, 0.5);
+}
+
+.btn-outline-success {
+  color: #28a745;
+  background-color: transparent;
+  background-image: none;
+  border-color: #28a745;
+}
+.btn-outline-success:hover {
+  color: #fff;
+  background-color: #28a745;
+  border-color: #28a745;
+}
+.btn-outline-success:focus, .btn-outline-success.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+}
+.btn-outline-success.disabled, .btn-outline-success:disabled {
+  color: #28a745;
+  background-color: transparent;
+}
+.btn-outline-success:not(:disabled):not(.disabled):active, .btn-outline-success:not(:disabled):not(.disabled).active, .show > .btn-outline-success.dropdown-toggle {
+  color: #fff;
+  background-color: #28a745;
+  border-color: #28a745;
+}
+.btn-outline-success:not(:disabled):not(.disabled):active:focus, .btn-outline-success:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-success.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(40, 167, 69, 0.5);
+}
+
+.btn-outline-info {
+  color: #17a2b8;
+  background-color: transparent;
+  background-image: none;
+  border-color: #17a2b8;
+}
+.btn-outline-info:hover {
+  color: #fff;
+  background-color: #17a2b8;
+  border-color: #17a2b8;
+}
+.btn-outline-info:focus, .btn-outline-info.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+}
+.btn-outline-info.disabled, .btn-outline-info:disabled {
+  color: #17a2b8;
+  background-color: transparent;
+}
+.btn-outline-info:not(:disabled):not(.disabled):active, .btn-outline-info:not(:disabled):not(.disabled).active, .show > .btn-outline-info.dropdown-toggle {
+  color: #fff;
+  background-color: #17a2b8;
+  border-color: #17a2b8;
+}
+.btn-outline-info:not(:disabled):not(.disabled):active:focus, .btn-outline-info:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-info.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(23, 162, 184, 0.5);
+}
+
+.btn-outline-warning {
+  color: #ffc107;
+  background-color: transparent;
+  background-image: none;
+  border-color: #ffc107;
+}
+.btn-outline-warning:hover {
+  color: #212529;
+  background-color: #ffc107;
+  border-color: #ffc107;
+}
+.btn-outline-warning:focus, .btn-outline-warning.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+}
+.btn-outline-warning.disabled, .btn-outline-warning:disabled {
+  color: #ffc107;
+  background-color: transparent;
+}
+.btn-outline-warning:not(:disabled):not(.disabled):active, .btn-outline-warning:not(:disabled):not(.disabled).active, .show > .btn-outline-warning.dropdown-toggle {
+  color: #212529;
+  background-color: #ffc107;
+  border-color: #ffc107;
+}
+.btn-outline-warning:not(:disabled):not(.disabled):active:focus, .btn-outline-warning:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-warning.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(255, 193, 7, 0.5);
+}
+
+.btn-outline-danger {
+  color: #dc3545;
+  background-color: transparent;
+  background-image: none;
+  border-color: #dc3545;
+}
+.btn-outline-danger:hover {
+  color: #fff;
+  background-color: #dc3545;
+  border-color: #dc3545;
+}
+.btn-outline-danger:focus, .btn-outline-danger.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+}
+.btn-outline-danger.disabled, .btn-outline-danger:disabled {
+  color: #dc3545;
+  background-color: transparent;
+}
+.btn-outline-danger:not(:disabled):not(.disabled):active, .btn-outline-danger:not(:disabled):not(.disabled).active, .show > .btn-outline-danger.dropdown-toggle {
+  color: #fff;
+  background-color: #dc3545;
+  border-color: #dc3545;
+}
+.btn-outline-danger:not(:disabled):not(.disabled):active:focus, .btn-outline-danger:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-danger.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(220, 53, 69, 0.5);
+}
+
+.btn-outline-light {
+  color: #f8f9fa;
+  background-color: transparent;
+  background-image: none;
+  border-color: #f8f9fa;
+}
+.btn-outline-light:hover {
+  color: #212529;
+  background-color: #f8f9fa;
+  border-color: #f8f9fa;
+}
+.btn-outline-light:focus, .btn-outline-light.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+}
+.btn-outline-light.disabled, .btn-outline-light:disabled {
+  color: #f8f9fa;
+  background-color: transparent;
+}
+.btn-outline-light:not(:disabled):not(.disabled):active, .btn-outline-light:not(:disabled):not(.disabled).active, .show > .btn-outline-light.dropdown-toggle {
+  color: #212529;
+  background-color: #f8f9fa;
+  border-color: #f8f9fa;
+}
+.btn-outline-light:not(:disabled):not(.disabled):active:focus, .btn-outline-light:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-light.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(248, 249, 250, 0.5);
+}
+
+.btn-outline-dark {
+  color: #343a40;
+  background-color: transparent;
+  background-image: none;
+  border-color: #343a40;
+}
+.btn-outline-dark:hover {
+  color: #fff;
+  background-color: #343a40;
+  border-color: #343a40;
+}
+.btn-outline-dark:focus, .btn-outline-dark.focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+}
+.btn-outline-dark.disabled, .btn-outline-dark:disabled {
+  color: #343a40;
+  background-color: transparent;
+}
+.btn-outline-dark:not(:disabled):not(.disabled):active, .btn-outline-dark:not(:disabled):not(.disabled).active, .show > .btn-outline-dark.dropdown-toggle {
+  color: #fff;
+  background-color: #343a40;
+  border-color: #343a40;
+}
+.btn-outline-dark:not(:disabled):not(.disabled):active:focus, .btn-outline-dark:not(:disabled):not(.disabled).active:focus, .show > .btn-outline-dark.dropdown-toggle:focus {
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+          box-shadow: 0 0 0 0.2rem rgba(52, 58, 64, 0.5);
+}
+
+.btn-link {
+  font-weight: 400;
+  color: #007bff;
+  background-color: transparent;
+}
+.btn-link:hover {
+  color: #0056b3;
+  text-decoration: underline;
+  background-color: transparent;
+  border-color: transparent;
+}
+.btn-link:focus, .btn-link.focus {
+  text-decoration: underline;
+  border-color: transparent;
+  -webkit-box-shadow: none;
+          box-shadow: none;
+}
+.btn-link:disabled, .btn-link.disabled {
+  color: #6c757d;
+}
+
+.btn-lg, .btn-group-lg > .btn {
+  padding: 0.5rem 1rem;
+  font-size: 1.25rem;
+  line-height: 1.5;
+  border-radius: 0.3rem;
+}
+
+.btn-sm, .btn-group-sm > .btn {
+  padding: 0.25rem 0.5rem;
+  font-size: 0.875rem;
+  line-height: 1.5;
+  border-radius: 0.2rem;
+}
+
+.btn-block {
+  display: block;
+  width: 100%;
+}
+.btn-block + .btn-block {
+  margin-top: 0.5rem;
+}
+
+input[type="submit"].btn-block,
+input[type="reset"].btn-block,
+input[type="button"].btn-block {
+  width: 100%;
+}
+
+.fade {
+  opacity: 0;
+  -webkit-transition: opacity 0.15s linear;
+  transition: opacity 0.15s linear;
+}
+.fade.show {
+  opacity: 1;
+}
+
+.collapse {
+  display: none;
+}
+.collapse.show {
+  display: block;
+}
+
+tr.collapse.show {
+  display: table-row;
+}
+
+tbody.collapse.show {
+  display: table-row-group;
+}
+
+.collapsing {
+  position: relative;
+  height: 0;
+  overflow: hidden;
+  -webkit-transition: height 0.35s ease;
+  transition: height 0.35s ease;
+}
+
+.dropup,
+.dropdown {
+  position: relative;
+}
+
+.dropdown-toggle::after {
+  display: inline-block;
+  width: 0;
+  height: 0;
+  margin-left: 0.255em;
+  vertical-align: 0.255em;
+  content: "";
+  border-top: 0.3em solid;
+  border-right: 0.3em solid transparent;
+  border-bottom: 0;
+  border-left: 0.3em solid transparent;
+}
+.dropdown-toggle:empty::after {
+  margin-left: 0;
+}
+
+.dropdown-menu {
+  position: absolute;
+  top: 100%;
+  left: 0;
+  z-index: 1000;
+  display: none;
+  float: left;
+  min-width: 10rem;
+  padding: 0.5rem 0;
+  margin: 0.125rem 0 0;
+  font-size: 1rem;
+  color: #212529;
+  text-align: left;
+  list-style: none;
+  background-color: #fff;
+  background-clip: padding-box;
+  border: 1px solid rgba(0, 0, 0, 0.15);
+  border-radius: 0.25rem;
+}
+
+.dropup .dropdown-menu {
+  margin-top: 0;
+  margin-bottom: 0.125rem;
+}
+.dropup .dropdown-toggle::after {
+  display: inline-block;
+  width: 0;
+  height: 0;
+  margin-left: 0.255em;
+  vertical-align: 0.255em;
+  content: "";
+  border-top: 0;
+  border-right: 0.3em solid transparent;
+  border-bottom: 0.3em solid;
+  border-left: 0.3em solid transparent;
+}
+.dropup .dropdown-toggle:empty::after {
+  margin-left: 0;
+}
+
+.dropright .dropdown-menu {
+  margin-top: 0;
+  margin-left: 0.125rem;
+}
+.dropright .dropdown-toggle::after {
+  display: inline-block;
+  width: 0;
+  height: 0;
+  margin-left: 0.255em;
+  vertical-align: 0.255em;
+  content: "";
+  border-top: 0.3em solid transparent;
+  border-bottom: 0.3em solid transparent;
+  border-left: 0.3em solid;
+}
+.dropright .dropdown-toggle:empty::after {
+  margin-left: 0;
+}
+.dropright .dropdown-toggle::after {
+  vertical-align: 0;
+}
+
+.dropleft .dropdown-menu {
+  margin-top: 0;
+  margin-right: 0.125rem;
+}
+.dropleft .dropdown-toggle::after {
+  display: inline-block;
+  width: 0;
+  height: 0;
+  margin-left: 0.255em;
+  vertical-align: 0.255em;
+  content: "";
+}
+.dropleft .dropdown-toggle::after {
+  display: none;
+}
+.dropleft .dropdown-toggle::before {
+  display: inline-block;
+  width: 0;
+  height: 0;
+  margin-right: 0.255em;
+  vertical-align: 0.255em;
+  content: "";
+  border-top: 0.3em solid transparent;
+  border-right: 0.3em solid;
+  border-bottom: 0.3em solid transparent;
+}
+.dropleft .dropdown-toggle:empty::after {
+  margin-left: 0;
+}
+.dropleft .dropdown-toggle::before {
+  vertical-align: 0;
+}
+
+.dropdown-divider {
+  height: 0;
+  margin: 0.5rem 0;
+  overflow: hidden;
+  border-top: 1px solid #e9ecef;
+}
+
+.dropdown-item {
+  display: block;
+  width: 100%;
+  padding: 0.25rem 1.5rem;
+  clear: both;
+  font-weight: 400;
+  color: #212529;
+  text-align: inherit;
+  white-space: nowrap;
+  background-color: transparent;
+  border: 0;
+}
+.dropdown-item:hover, .dropdown-item:focus {
+  color: #16181b;
+  text-decoration: none;
+  background-color: #f8f9fa;
+}
+.dropdown-item.active, .dropdown-item:active {
+  color: #fff;
+  text-decoration: none;
+  background-color: #007bff;
+}
+.dropdown-item.disabled, .dropdown-item:disabled {
+  color: #6c757d;
+  background-color: transparent;
+}
+
+.dropdown-menu.show {
+  display: block;
+}
+
+.dropdown-header {
+  display: block;
+  padding: 0.5rem 1.5rem;
+  margin-bottom: 0;
+  font-size: 0.875rem;
+  color: #6c757d;
+  white-space: nowrap;
+}
+
+.btn-group,
+.btn-group-vertical {
+  position: relative;
+  display: -webkit-inline-box;
+  display: -ms-inline-flexbox;
+  display: inline-flex;
+  vertical-align: middle;
+}
+.btn-group > .btn,
+.btn-group-vertical > .btn {
+  position: relative;
+  -webkit-box-flex: 0;
+      -ms-flex: 0 1 auto;
+          flex: 0 1 auto;
+}
+.btn-group > .btn:hover,
+.btn-group-vertical > .btn:hover {
+  z-index: 1;
+}
+.btn-group > .btn:focus, .btn-group > .btn:active, .btn-group > .btn.active,
+.btn-group-vertical > .btn:focus,
+.btn-group-vertical > .btn:active,
+.btn-group-vertical > .btn.active {
+  z-index: 1;
+}
+.btn-group .btn + .btn,
+.btn-group .btn + .btn-group,
+.btn-group .btn-group + .btn,
+.btn-group .btn-group + .btn-group,
+.btn-group-vertical .btn + .btn,
+.btn-group-vertical .btn + .btn-group,
+.btn-group-vertical .btn-group + .btn,
+.btn-group-vertical .btn-group + .btn-group {
+  margin-left: -1px;
+}
+
+.btn-toolbar {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  -webkit-box-pack: start;
+      -ms-flex-pack: start;
+          justify-content: flex-start;
+}
+.btn-toolbar .input-group {
+  width: auto;
+}
+
+.btn-group > .btn:first-child {
+  margin-left: 0;
+}
+.btn-group > .btn:not(:last-child):not(.dropdown-toggle),
+.btn-group > .btn-group:not(:last-child) > .btn {
+  border-top-right-radius: 0;
+  border-bottom-right-radius: 0;
+}
+.btn-group > .btn:not(:first-child),
+.btn-group > .btn-group:not(:first-child) > .btn {
+  border-top-left-radius: 0;
+  border-bottom-left-radius: 0;
+}
+
+.dropdown-toggle-split {
+  padding-right: 0.5625rem;
+  padding-left: 0.5625rem;
+}
+.dropdown-toggle-split::after {
+  margin-left: 0;
+}
+
+.btn-sm + .dropdown-toggle-split, .btn-group-sm > .btn + .dropdown-toggle-split {
+  padding-right: 0.375rem;
+  padding-left: 0.375rem;
+}
+
+.btn-lg + .dropdown-toggle-split, .btn-group-lg > .btn + .dropdown-toggle-split {
+  padding-right: 0.75rem;
+  padding-left: 0.75rem;
+}
+
+.btn-group-vertical {
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+  -webkit-box-align: start;
+      -ms-flex-align: start;
+          align-items: flex-start;
+  -webkit-box-pack: center;
+      -ms-flex-pack: center;
+          justify-content: center;
+}
+.btn-group-vertical .btn,
+.btn-group-vertical .btn-group {
+  width: 100%;
+}
+.btn-group-vertical > .btn + .btn,
+.btn-group-vertical > .btn + .btn-group,
+.btn-group-vertical > .btn-group + .btn,
+.btn-group-vertical > .btn-group + .btn-group {
+  margin-top: -1px;
+  margin-left: 0;
+}
+.btn-group-vertical > .btn:not(:last-child):not(.dropdown-toggle),
+.btn-group-vertical > .btn-group:not(:last-child) > .btn {
+  border-bottom-right-radius: 0;
+  border-bottom-left-radius: 0;
+}
+.btn-group-vertical > .btn:not(:first-child),
+.btn-group-vertical > .btn-group:not(:first-child) > .btn {
+  border-top-left-radius: 0;
+  border-top-right-radius: 0;
+}
+
+.btn-group-toggle > .btn,
+.btn-group-toggle > .btn-group > .btn {
+  margin-bottom: 0;
+}
+.btn-group-toggle > .btn input[type="radio"],
+.btn-group-toggle > .btn input[type="checkbox"],
+.btn-group-toggle > .btn-group > .btn input[type="radio"],
+.btn-group-toggle > .btn-group > .btn input[type="checkbox"] {
+  position: absolute;
+  clip: rect(0, 0, 0, 0);
+  pointer-events: none;
+}
+
+.input-group {
+  position: relative;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  -webkit-box-align: stretch;
+      -ms-flex-align: stretch;
+          align-items: stretch;
+  width: 100%;
+}
+.input-group > .form-control,
+.input-group > .custom-select,
+.input-group > .custom-file {
+  position: relative;
+  -webkit-box-flex: 1;
+      -ms-flex: 1 1 auto;
+          flex: 1 1 auto;
+  width: 1%;
+  margin-bottom: 0;
+}
+.input-group > .form-control:focus,
+.input-group > .custom-select:focus,
+.input-group > .custom-file:focus {
+  z-index: 3;
+}
+.input-group > .form-control + .form-control,
+.input-group > .form-control + .custom-select,
+.input-group > .form-control + .custom-file,
+.input-group > .custom-select + .form-control,
+.input-group > .custom-select + .custom-select,
+.input-group > .custom-select + .custom-file,
+.input-group > .custom-file + .form-control,
+.input-group > .custom-file + .custom-select,
+.input-group > .custom-file + .custom-file {
+  margin-left: -1px;
+}
+.input-group > .form-control:not(:last-child),
+.input-group > .custom-select:not(:last-child) {
+  border-top-right-radius: 0;
+  border-bottom-right-radius: 0;
+}
+.input-group > .form-control:not(:first-child),
+.input-group > .custom-select:not(:first-child) {
+  border-top-left-radius: 0;
+  border-bottom-left-radius: 0;
+}
+.input-group > .custom-file {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+}
+.input-group > .custom-file:not(:last-child) .custom-file-label, .input-group > .custom-file:not(:last-child) .custom-file-label::before {
+  border-top-right-radius: 0;
+  border-bottom-right-radius: 0;
+}
+.input-group > .custom-file:not(:first-child) .custom-file-label, .input-group > .custom-file:not(:first-child) .custom-file-label::before {
+  border-top-left-radius: 0;
+  border-bottom-left-radius: 0;
+}
+
+.input-group-prepend,
+.input-group-append {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+}
+.input-group-prepend .btn,
+.input-group-append .btn {
+  position: relative;
+  z-index: 2;
+}
+.input-group-prepend .btn + .btn,
+.input-group-prepend .btn + .input-group-text,
+.input-group-prepend .input-group-text + .input-group-text,
+.input-group-prepend .input-group-text + .btn,
+.input-group-append .btn + .btn,
+.input-group-append .btn + .input-group-text,
+.input-group-append .input-group-text + .input-group-text,
+.input-group-append .input-group-text + .btn {
+  margin-left: -1px;
+}
+
+.input-group-prepend {
+  margin-right: -1px;
+}
+
+.input-group-append {
+  margin-left: -1px;
+}
+
+.input-group-text {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  padding: 0.375rem 0.75rem;
+  margin-bottom: 0;
+  font-size: 1rem;
+  font-weight: 400;
+  line-height: 1.5;
+  color: #495057;
+  text-align: center;
+  white-space: nowrap;
+  background-color: #e9ecef;
+  border: 1px solid #ced4da;
+  border-radius: 0.25rem;
+}
+.input-group-text input[type="radio"],
+.input-group-text input[type="checkbox"] {
+  margin-top: 0;
+}
+
+.input-group > .input-group-prepend > .btn,
+.input-group > .input-group-prepend > .input-group-text,
+.input-group > .input-group-append:not(:last-child) > .btn,
+.input-group > .input-group-append:not(:last-child) > .input-group-text,
+.input-group > .input-group-append:last-child > .btn:not(:last-child):not(.dropdown-toggle),
+.input-group > .input-group-append:last-child > .input-group-text:not(:last-child) {
+  border-top-right-radius: 0;
+  border-bottom-right-radius: 0;
+}
+
+.input-group > .input-group-append > .btn,
+.input-group > .input-group-append > .input-group-text,
+.input-group > .input-group-prepend:not(:first-child) > .btn,
+.input-group > .input-group-prepend:not(:first-child) > .input-group-text,
+.input-group > .input-group-prepend:first-child > .btn:not(:first-child),
+.input-group > .input-group-prepend:first-child > .input-group-text:not(:first-child) {
+  border-top-left-radius: 0;
+  border-bottom-left-radius: 0;
+}
+
+.custom-control {
+  position: relative;
+  display: block;
+  min-height: 1.5rem;
+  padding-left: 1.5rem;
+}
+
+.custom-control-inline {
+  display: -webkit-inline-box;
+  display: -ms-inline-flexbox;
+  display: inline-flex;
+  margin-right: 1rem;
+}
+
+.custom-control-input {
+  position: absolute;
+  z-index: -1;
+  opacity: 0;
+}
+.custom-control-input:checked ~ .custom-control-label::before {
+  color: #fff;
+  background-color: #007bff;
+}
+.custom-control-input:focus ~ .custom-control-label::before {
+  -webkit-box-shadow: 0 0 0 1px #fff, 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+          box-shadow: 0 0 0 1px #fff, 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+}
+.custom-control-input:active ~ .custom-control-label::before {
+  color: #fff;
+  background-color: #b3d7ff;
+}
+.custom-control-input:disabled ~ .custom-control-label {
+  color: #6c757d;
+}
+.custom-control-input:disabled ~ .custom-control-label::before {
+  background-color: #e9ecef;
+}
+
+.custom-control-label {
+  margin-bottom: 0;
+}
+.custom-control-label::before {
+  position: absolute;
+  top: 0.25rem;
+  left: 0;
+  display: block;
+  width: 1rem;
+  height: 1rem;
+  pointer-events: none;
+  content: "";
+  -webkit-user-select: none;
+     -moz-user-select: none;
+      -ms-user-select: none;
+          user-select: none;
+  background-color: #dee2e6;
+}
+.custom-control-label::after {
+  position: absolute;
+  top: 0.25rem;
+  left: 0;
+  display: block;
+  width: 1rem;
+  height: 1rem;
+  content: "";
+  background-repeat: no-repeat;
+  background-position: center center;
+  background-size: 50% 50%;
+}
+
+.custom-checkbox .custom-control-label::before {
+  border-radius: 0.25rem;
+}
+.custom-checkbox .custom-control-input:checked ~ .custom-control-label::before {
+  background-color: #007bff;
+}
+.custom-checkbox .custom-control-input:checked ~ .custom-control-label::after {
+  background-image: url("data:image/svg+xml;charset=utf8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 8 8'%3E%3Cpath fill='%23fff' d='M6.564.75l-3.59 3.612-1.538-1.55L0 4.26 2.974 7.25 8 2.193z'/%3E%3C/svg%3E");
+}
+.custom-checkbox .custom-control-input:indeterminate ~ .custom-control-label::before {
+  background-color: #007bff;
+}
+.custom-checkbox .custom-control-input:indeterminate ~ .custom-control-label::after {
+  background-image: url("data:image/svg+xml;charset=utf8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 4 4'%3E%3Cpath stroke='%23fff' d='M0 2h4'/%3E%3C/svg%3E");
+}
+.custom-checkbox .custom-control-input:disabled:checked ~ .custom-control-label::before {
+  background-color: rgba(0, 123, 255, 0.5);
+}
+.custom-checkbox .custom-control-input:disabled:indeterminate ~ .custom-control-label::before {
+  background-color: rgba(0, 123, 255, 0.5);
+}
+
+.custom-radio .custom-control-label::before {
+  border-radius: 50%;
+}
+.custom-radio .custom-control-input:checked ~ .custom-control-label::before {
+  background-color: #007bff;
+}
+.custom-radio .custom-control-input:checked ~ .custom-control-label::after {
+  background-image: url("data:image/svg+xml;charset=utf8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3E%3Ccircle r='3' fill='%23fff'/%3E%3C/svg%3E");
+}
+.custom-radio .custom-control-input:disabled:checked ~ .custom-control-label::before {
+  background-color: rgba(0, 123, 255, 0.5);
+}
+
+.custom-select {
+  display: inline-block;
+  width: 100%;
+  height: calc(2.25rem + 2px);
+  padding: 0.375rem 1.75rem 0.375rem 0.75rem;
+  line-height: 1.5;
+  color: #495057;
+  vertical-align: middle;
+  background: #fff url("data:image/svg+xml;charset=utf8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 4 5'%3E%3Cpath fill='%23343a40' d='M2 0L0 2h4zm0 5L0 3h4z'/%3E%3C/svg%3E") no-repeat right 0.75rem center;
+  background-size: 8px 10px;
+  border: 1px solid #ced4da;
+  border-radius: 0.25rem;
+  -webkit-appearance: none;
+     -moz-appearance: none;
+          appearance: none;
+}
+.custom-select:focus {
+  border-color: #80bdff;
+  outline: 0;
+  -webkit-box-shadow: inset 0 1px 2px rgba(0, 0, 0, 0.075), 0 0 5px rgba(128, 189, 255, 0.5);
+          box-shadow: inset 0 1px 2px rgba(0, 0, 0, 0.075), 0 0 5px rgba(128, 189, 255, 0.5);
+}
+.custom-select:focus::-ms-value {
+  color: #495057;
+  background-color: #fff;
+}
+.custom-select[multiple], .custom-select[size]:not([size="1"]) {
+  height: auto;
+  padding-right: 0.75rem;
+  background-image: none;
+}
+.custom-select:disabled {
+  color: #6c757d;
+  background-color: #e9ecef;
+}
+.custom-select::-ms-expand {
+  opacity: 0;
+}
+
+.custom-select-sm {
+  height: calc(1.8125rem + 2px);
+  padding-top: 0.375rem;
+  padding-bottom: 0.375rem;
+  font-size: 75%;
+}
+
+.custom-select-lg {
+  height: calc(2.875rem + 2px);
+  padding-top: 0.375rem;
+  padding-bottom: 0.375rem;
+  font-size: 125%;
+}
+
+.custom-file {
+  position: relative;
+  display: inline-block;
+  width: 100%;
+  height: calc(2.25rem + 2px);
+  margin-bottom: 0;
+}
+
+.custom-file-input {
+  position: relative;
+  z-index: 2;
+  width: 100%;
+  height: calc(2.25rem + 2px);
+  margin: 0;
+  opacity: 0;
+}
+.custom-file-input:focus ~ .custom-file-control {
+  border-color: #80bdff;
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+}
+.custom-file-input:focus ~ .custom-file-control::before {
+  border-color: #80bdff;
+}
+.custom-file-input:lang(en) ~ .custom-file-label::after {
+  content: "Browse";
+}
+
+.custom-file-label {
+  position: absolute;
+  top: 0;
+  right: 0;
+  left: 0;
+  z-index: 1;
+  height: calc(2.25rem + 2px);
+  padding: 0.375rem 0.75rem;
+  line-height: 1.5;
+  color: #495057;
+  background-color: #fff;
+  border: 1px solid #ced4da;
+  border-radius: 0.25rem;
+}
+.custom-file-label::after {
+  position: absolute;
+  top: 0;
+  right: 0;
+  bottom: 0;
+  z-index: 3;
+  display: block;
+  height: calc(calc(2.25rem + 2px) - 1px * 2);
+  padding: 0.375rem 0.75rem;
+  line-height: 1.5;
+  color: #495057;
+  content: "Browse";
+  background-color: #e9ecef;
+  border-left: 1px solid #ced4da;
+  border-radius: 0 0.25rem 0.25rem 0;
+}
+
+.nav {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  padding-left: 0;
+  margin-bottom: 0;
+  list-style: none;
+}
+
+.nav-link {
+  display: block;
+  padding: 0.5rem 1rem;
+}
+.nav-link:hover, .nav-link:focus {
+  text-decoration: none;
+}
+.nav-link.disabled {
+  color: #6c757d;
+}
+
+.nav-tabs {
+  border-bottom: 1px solid #dee2e6;
+}
+.nav-tabs .nav-item {
+  margin-bottom: -1px;
+}
+.nav-tabs .nav-link {
+  border: 1px solid transparent;
+  border-top-left-radius: 0.25rem;
+  border-top-right-radius: 0.25rem;
+}
+.nav-tabs .nav-link:hover, .nav-tabs .nav-link:focus {
+  border-color: #e9ecef #e9ecef #dee2e6;
+}
+.nav-tabs .nav-link.disabled {
+  color: #6c757d;
+  background-color: transparent;
+  border-color: transparent;
+}
+.nav-tabs .nav-link.active,
+.nav-tabs .nav-item.show .nav-link {
+  color: #495057;
+  background-color: #fff;
+  border-color: #dee2e6 #dee2e6 #fff;
+}
+.nav-tabs .dropdown-menu {
+  margin-top: -1px;
+  border-top-left-radius: 0;
+  border-top-right-radius: 0;
+}
+
+.nav-pills .nav-link {
+  border-radius: 0.25rem;
+}
+.nav-pills .nav-link.active,
+.nav-pills .show > .nav-link {
+  color: #fff;
+  background-color: #007bff;
+}
+
+.nav-fill .nav-item {
+  -webkit-box-flex: 1;
+      -ms-flex: 1 1 auto;
+          flex: 1 1 auto;
+  text-align: center;
+}
+
+.nav-justified .nav-item {
+  -ms-flex-preferred-size: 0;
+      flex-basis: 0;
+  -webkit-box-flex: 1;
+      -ms-flex-positive: 1;
+          flex-grow: 1;
+  text-align: center;
+}
+
+.tab-content > .tab-pane {
+  display: none;
+}
+.tab-content > .active {
+  display: block;
+}
+
+.navbar {
+  position: relative;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  -webkit-box-pack: justify;
+      -ms-flex-pack: justify;
+          justify-content: space-between;
+  padding: 0.5rem 1rem;
+}
+.navbar > .container,
+.navbar > .container-fluid {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  -webkit-box-pack: justify;
+      -ms-flex-pack: justify;
+          justify-content: space-between;
+}
+
+.navbar-brand {
+  display: inline-block;
+  padding-top: 0.3125rem;
+  padding-bottom: 0.3125rem;
+  margin-right: 1rem;
+  font-size: 1.25rem;
+  line-height: inherit;
+  white-space: nowrap;
+}
+.navbar-brand:hover, .navbar-brand:focus {
+  text-decoration: none;
+}
+
+.navbar-nav {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+  padding-left: 0;
+  margin-bottom: 0;
+  list-style: none;
+}
+.navbar-nav .nav-link {
+  padding-right: 0;
+  padding-left: 0;
+}
+.navbar-nav .dropdown-menu {
+  position: static;
+  float: none;
+}
+
+.navbar-text {
+  display: inline-block;
+  padding-top: 0.5rem;
+  padding-bottom: 0.5rem;
+}
+
+.navbar-collapse {
+  -ms-flex-preferred-size: 100%;
+      flex-basis: 100%;
+  -webkit-box-flex: 1;
+      -ms-flex-positive: 1;
+          flex-grow: 1;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+}
+
+.navbar-toggler {
+  padding: 0.25rem 0.75rem;
+  font-size: 1.25rem;
+  line-height: 1;
+  background-color: transparent;
+  border: 1px solid transparent;
+  border-radius: 0.25rem;
+}
+.navbar-toggler:hover, .navbar-toggler:focus {
+  text-decoration: none;
+}
+.navbar-toggler:not(:disabled):not(.disabled) {
+  cursor: pointer;
+}
+
+.navbar-toggler-icon {
+  display: inline-block;
+  width: 1.5em;
+  height: 1.5em;
+  vertical-align: middle;
+  content: "";
+  background: no-repeat center center;
+  background-size: 100% 100%;
+}
+
+@media (max-width: 575.98px) {
+  .navbar-expand-sm > .container,
+  .navbar-expand-sm > .container-fluid {
+    padding-right: 0;
+    padding-left: 0;
+  }
+}
+@media (min-width: 576px) {
+  .navbar-expand-sm {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-flow: row nowrap;
+            flex-flow: row nowrap;
+    -webkit-box-pack: start;
+        -ms-flex-pack: start;
+            justify-content: flex-start;
+  }
+  .navbar-expand-sm .navbar-nav {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-direction: row;
+            flex-direction: row;
+  }
+  .navbar-expand-sm .navbar-nav .dropdown-menu {
+    position: absolute;
+  }
+  .navbar-expand-sm .navbar-nav .dropdown-menu-right {
+    right: 0;
+    left: auto;
+  }
+  .navbar-expand-sm .navbar-nav .nav-link {
+    padding-right: 0.5rem;
+    padding-left: 0.5rem;
+  }
+  .navbar-expand-sm > .container,
+  .navbar-expand-sm > .container-fluid {
+    -ms-flex-wrap: nowrap;
+        flex-wrap: nowrap;
+  }
+  .navbar-expand-sm .navbar-collapse {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+    -ms-flex-preferred-size: auto;
+        flex-basis: auto;
+  }
+  .navbar-expand-sm .navbar-toggler {
+    display: none;
+  }
+  .navbar-expand-sm .dropup .dropdown-menu {
+    top: auto;
+    bottom: 100%;
+  }
+}
+@media (max-width: 767.98px) {
+  .navbar-expand-md > .container,
+  .navbar-expand-md > .container-fluid {
+    padding-right: 0;
+    padding-left: 0;
+  }
+}
+@media (min-width: 768px) {
+  .navbar-expand-md {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-flow: row nowrap;
+            flex-flow: row nowrap;
+    -webkit-box-pack: start;
+        -ms-flex-pack: start;
+            justify-content: flex-start;
+  }
+  .navbar-expand-md .navbar-nav {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-direction: row;
+            flex-direction: row;
+  }
+  .navbar-expand-md .navbar-nav .dropdown-menu {
+    position: absolute;
+  }
+  .navbar-expand-md .navbar-nav .dropdown-menu-right {
+    right: 0;
+    left: auto;
+  }
+  .navbar-expand-md .navbar-nav .nav-link {
+    padding-right: 0.5rem;
+    padding-left: 0.5rem;
+  }
+  .navbar-expand-md > .container,
+  .navbar-expand-md > .container-fluid {
+    -ms-flex-wrap: nowrap;
+        flex-wrap: nowrap;
+  }
+  .navbar-expand-md .navbar-collapse {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+    -ms-flex-preferred-size: auto;
+        flex-basis: auto;
+  }
+  .navbar-expand-md .navbar-toggler {
+    display: none;
+  }
+  .navbar-expand-md .dropup .dropdown-menu {
+    top: auto;
+    bottom: 100%;
+  }
+}
+@media (max-width: 991.98px) {
+  .navbar-expand-lg > .container,
+  .navbar-expand-lg > .container-fluid {
+    padding-right: 0;
+    padding-left: 0;
+  }
+}
+@media (min-width: 992px) {
+  .navbar-expand-lg {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-flow: row nowrap;
+            flex-flow: row nowrap;
+    -webkit-box-pack: start;
+        -ms-flex-pack: start;
+            justify-content: flex-start;
+  }
+  .navbar-expand-lg .navbar-nav {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-direction: row;
+            flex-direction: row;
+  }
+  .navbar-expand-lg .navbar-nav .dropdown-menu {
+    position: absolute;
+  }
+  .navbar-expand-lg .navbar-nav .dropdown-menu-right {
+    right: 0;
+    left: auto;
+  }
+  .navbar-expand-lg .navbar-nav .nav-link {
+    padding-right: 0.5rem;
+    padding-left: 0.5rem;
+  }
+  .navbar-expand-lg > .container,
+  .navbar-expand-lg > .container-fluid {
+    -ms-flex-wrap: nowrap;
+        flex-wrap: nowrap;
+  }
+  .navbar-expand-lg .navbar-collapse {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+    -ms-flex-preferred-size: auto;
+        flex-basis: auto;
+  }
+  .navbar-expand-lg .navbar-toggler {
+    display: none;
+  }
+  .navbar-expand-lg .dropup .dropdown-menu {
+    top: auto;
+    bottom: 100%;
+  }
+}
+@media (max-width: 1199.98px) {
+  .navbar-expand-xl > .container,
+  .navbar-expand-xl > .container-fluid {
+    padding-right: 0;
+    padding-left: 0;
+  }
+}
+@media (min-width: 1200px) {
+  .navbar-expand-xl {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-flow: row nowrap;
+            flex-flow: row nowrap;
+    -webkit-box-pack: start;
+        -ms-flex-pack: start;
+            justify-content: flex-start;
+  }
+  .navbar-expand-xl .navbar-nav {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-direction: row;
+            flex-direction: row;
+  }
+  .navbar-expand-xl .navbar-nav .dropdown-menu {
+    position: absolute;
+  }
+  .navbar-expand-xl .navbar-nav .dropdown-menu-right {
+    right: 0;
+    left: auto;
+  }
+  .navbar-expand-xl .navbar-nav .nav-link {
+    padding-right: 0.5rem;
+    padding-left: 0.5rem;
+  }
+  .navbar-expand-xl > .container,
+  .navbar-expand-xl > .container-fluid {
+    -ms-flex-wrap: nowrap;
+        flex-wrap: nowrap;
+  }
+  .navbar-expand-xl .navbar-collapse {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+    -ms-flex-preferred-size: auto;
+        flex-basis: auto;
+  }
+  .navbar-expand-xl .navbar-toggler {
+    display: none;
+  }
+  .navbar-expand-xl .dropup .dropdown-menu {
+    top: auto;
+    bottom: 100%;
+  }
+}
+.navbar-expand {
+  -webkit-box-orient: horizontal;
+  -webkit-box-direction: normal;
+      -ms-flex-flow: row nowrap;
+          flex-flow: row nowrap;
+  -webkit-box-pack: start;
+      -ms-flex-pack: start;
+          justify-content: flex-start;
+}
+.navbar-expand > .container,
+.navbar-expand > .container-fluid {
+  padding-right: 0;
+  padding-left: 0;
+}
+.navbar-expand .navbar-nav {
+  -webkit-box-orient: horizontal;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: row;
+          flex-direction: row;
+}
+.navbar-expand .navbar-nav .dropdown-menu {
+  position: absolute;
+}
+.navbar-expand .navbar-nav .dropdown-menu-right {
+  right: 0;
+  left: auto;
+}
+.navbar-expand .navbar-nav .nav-link {
+  padding-right: 0.5rem;
+  padding-left: 0.5rem;
+}
+.navbar-expand > .container,
+.navbar-expand > .container-fluid {
+  -ms-flex-wrap: nowrap;
+      flex-wrap: nowrap;
+}
+.navbar-expand .navbar-collapse {
+  display: -webkit-box !important;
+  display: -ms-flexbox !important;
+  display: flex !important;
+  -ms-flex-preferred-size: auto;
+      flex-basis: auto;
+}
+.navbar-expand .navbar-toggler {
+  display: none;
+}
+.navbar-expand .dropup .dropdown-menu {
+  top: auto;
+  bottom: 100%;
+}
+
+.navbar-light .navbar-brand {
+  color: rgba(0, 0, 0, 0.9);
+}
+.navbar-light .navbar-brand:hover, .navbar-light .navbar-brand:focus {
+  color: rgba(0, 0, 0, 0.9);
+}
+.navbar-light .navbar-nav .nav-link {
+  color: rgba(0, 0, 0, 0.5);
+}
+.navbar-light .navbar-nav .nav-link:hover, .navbar-light .navbar-nav .nav-link:focus {
+  color: rgba(0, 0, 0, 0.7);
+}
+.navbar-light .navbar-nav .nav-link.disabled {
+  color: rgba(0, 0, 0, 0.3);
+}
+.navbar-light .navbar-nav .show > .nav-link,
+.navbar-light .navbar-nav .active > .nav-link,
+.navbar-light .navbar-nav .nav-link.show,
+.navbar-light .navbar-nav .nav-link.active {
+  color: rgba(0, 0, 0, 0.9);
+}
+.navbar-light .navbar-toggler {
+  color: rgba(0, 0, 0, 0.5);
+  border-color: rgba(0, 0, 0, 0.1);
+}
+.navbar-light .navbar-toggler-icon {
+  background-image: url("data:image/svg+xml;charset=utf8,%3Csvg viewBox='0 0 30 30' xmlns='http://www.w3.org/2000/svg'%3E%3Cpath stroke='rgba(0, 0, 0, 0.5)' stroke-width='2' stroke-linecap='round' stroke-miterlimit='10' d='M4 7h22M4 15h22M4 23h22'/%3E%3C/svg%3E");
+}
+.navbar-light .navbar-text {
+  color: rgba(0, 0, 0, 0.5);
+}
+.navbar-light .navbar-text a {
+  color: rgba(0, 0, 0, 0.9);
+}
+.navbar-light .navbar-text a:hover, .navbar-light .navbar-text a:focus {
+  color: rgba(0, 0, 0, 0.9);
+}
+
+.navbar-dark .navbar-brand {
+  color: #fff;
+}
+.navbar-dark .navbar-brand:hover, .navbar-dark .navbar-brand:focus {
+  color: #fff;
+}
+.navbar-dark .navbar-nav .nav-link {
+  color: rgba(255, 255, 255, 0.5);
+}
+.navbar-dark .navbar-nav .nav-link:hover, .navbar-dark .navbar-nav .nav-link:focus {
+  color: rgba(255, 255, 255, 0.75);
+}
+.navbar-dark .navbar-nav .nav-link.disabled {
+  color: rgba(255, 255, 255, 0.25);
+}
+.navbar-dark .navbar-nav .show > .nav-link,
+.navbar-dark .navbar-nav .active > .nav-link,
+.navbar-dark .navbar-nav .nav-link.show,
+.navbar-dark .navbar-nav .nav-link.active {
+  color: #fff;
+}
+.navbar-dark .navbar-toggler {
+  color: rgba(255, 255, 255, 0.5);
+  border-color: rgba(255, 255, 255, 0.1);
+}
+.navbar-dark .navbar-toggler-icon {
+  background-image: url("data:image/svg+xml;charset=utf8,%3Csvg viewBox='0 0 30 30' xmlns='http://www.w3.org/2000/svg'%3E%3Cpath stroke='rgba(255, 255, 255, 0.5)' stroke-width='2' stroke-linecap='round' stroke-miterlimit='10' d='M4 7h22M4 15h22M4 23h22'/%3E%3C/svg%3E");
+}
+.navbar-dark .navbar-text {
+  color: rgba(255, 255, 255, 0.5);
+}
+.navbar-dark .navbar-text a {
+  color: #fff;
+}
+.navbar-dark .navbar-text a:hover, .navbar-dark .navbar-text a:focus {
+  color: #fff;
+}
+
+.card {
+  position: relative;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+  min-width: 0;
+  word-wrap: break-word;
+  background-color: #fff;
+  background-clip: border-box;
+  border: 1px solid rgba(0, 0, 0, 0.125);
+  border-radius: 0.25rem;
+}
+.card > hr {
+  margin-right: 0;
+  margin-left: 0;
+}
+.card > .list-group:first-child .list-group-item:first-child {
+  border-top-left-radius: 0.25rem;
+  border-top-right-radius: 0.25rem;
+}
+.card > .list-group:last-child .list-group-item:last-child {
+  border-bottom-right-radius: 0.25rem;
+  border-bottom-left-radius: 0.25rem;
+}
+
+.card-body {
+  -webkit-box-flex: 1;
+      -ms-flex: 1 1 auto;
+          flex: 1 1 auto;
+  padding: 1.25rem;
+}
+
+.card-title {
+  margin-bottom: 0.75rem;
+}
+
+.card-subtitle {
+  margin-top: -0.375rem;
+  margin-bottom: 0;
+}
+
+.card-text:last-child {
+  margin-bottom: 0;
+}
+
+.card-link:hover {
+  text-decoration: none;
+}
+.card-link + .card-link {
+  margin-left: 1.25rem;
+}
+
+.card-header {
+  padding: 0.75rem 1.25rem;
+  margin-bottom: 0;
+  background-color: rgba(0, 0, 0, 0.03);
+  border-bottom: 1px solid rgba(0, 0, 0, 0.125);
+}
+.card-header:first-child {
+  border-radius: calc(0.25rem - 1px) calc(0.25rem - 1px) 0 0;
+}
+.card-header + .list-group .list-group-item:first-child {
+  border-top: 0;
+}
+
+.card-footer {
+  padding: 0.75rem 1.25rem;
+  background-color: rgba(0, 0, 0, 0.03);
+  border-top: 1px solid rgba(0, 0, 0, 0.125);
+}
+.card-footer:last-child {
+  border-radius: 0 0 calc(0.25rem - 1px) calc(0.25rem - 1px);
+}
+
+.card-header-tabs {
+  margin-right: -0.625rem;
+  margin-bottom: -0.75rem;
+  margin-left: -0.625rem;
+  border-bottom: 0;
+}
+
+.card-header-pills {
+  margin-right: -0.625rem;
+  margin-left: -0.625rem;
+}
+
+.card-img-overlay {
+  position: absolute;
+  top: 0;
+  right: 0;
+  bottom: 0;
+  left: 0;
+  padding: 1.25rem;
+}
+
+.card-img {
+  width: 100%;
+  border-radius: calc(0.25rem - 1px);
+}
+
+.card-img-top {
+  width: 100%;
+  border-top-left-radius: calc(0.25rem - 1px);
+  border-top-right-radius: calc(0.25rem - 1px);
+}
+
+.card-img-bottom {
+  width: 100%;
+  border-bottom-right-radius: calc(0.25rem - 1px);
+  border-bottom-left-radius: calc(0.25rem - 1px);
+}
+
+.card-deck {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+}
+.card-deck .card {
+  margin-bottom: 15px;
+}
+@media (min-width: 576px) {
+  .card-deck {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-flow: row wrap;
+            flex-flow: row wrap;
+    margin-right: -15px;
+    margin-left: -15px;
+  }
+  .card-deck .card {
+    display: -webkit-box;
+    display: -ms-flexbox;
+    display: flex;
+    -webkit-box-flex: 1;
+        -ms-flex: 1 0 0%;
+            flex: 1 0 0%;
+    -webkit-box-orient: vertical;
+    -webkit-box-direction: normal;
+        -ms-flex-direction: column;
+            flex-direction: column;
+    margin-right: 15px;
+    margin-bottom: 0;
+    margin-left: 15px;
+  }
+}
+
+.card-group {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+}
+.card-group > .card {
+  margin-bottom: 15px;
+}
+@media (min-width: 576px) {
+  .card-group {
+    -webkit-box-orient: horizontal;
+    -webkit-box-direction: normal;
+        -ms-flex-flow: row wrap;
+            flex-flow: row wrap;
+  }
+  .card-group > .card {
+    -webkit-box-flex: 1;
+        -ms-flex: 1 0 0%;
+            flex: 1 0 0%;
+    margin-bottom: 0;
+  }
+  .card-group > .card + .card {
+    margin-left: 0;
+    border-left: 0;
+  }
+  .card-group > .card:first-child {
+    border-top-right-radius: 0;
+    border-bottom-right-radius: 0;
+  }
+  .card-group > .card:first-child .card-img-top,
+  .card-group > .card:first-child .card-header {
+    border-top-right-radius: 0;
+  }
+  .card-group > .card:first-child .card-img-bottom,
+  .card-group > .card:first-child .card-footer {
+    border-bottom-right-radius: 0;
+  }
+  .card-group > .card:last-child {
+    border-top-left-radius: 0;
+    border-bottom-left-radius: 0;
+  }
+  .card-group > .card:last-child .card-img-top,
+  .card-group > .card:last-child .card-header {
+    border-top-left-radius: 0;
+  }
+  .card-group > .card:last-child .card-img-bottom,
+  .card-group > .card:last-child .card-footer {
+    border-bottom-left-radius: 0;
+  }
+  .card-group > .card:only-child {
+    border-radius: 0.25rem;
+  }
+  .card-group > .card:only-child .card-img-top,
+  .card-group > .card:only-child .card-header {
+    border-top-left-radius: 0.25rem;
+    border-top-right-radius: 0.25rem;
+  }
+  .card-group > .card:only-child .card-img-bottom,
+  .card-group > .card:only-child .card-footer {
+    border-bottom-right-radius: 0.25rem;
+    border-bottom-left-radius: 0.25rem;
+  }
+  .card-group > .card:not(:first-child):not(:last-child):not(:only-child) {
+    border-radius: 0;
+  }
+  .card-group > .card:not(:first-child):not(:last-child):not(:only-child) .card-img-top,
+  .card-group > .card:not(:first-child):not(:last-child):not(:only-child) .card-img-bottom,
+  .card-group > .card:not(:first-child):not(:last-child):not(:only-child) .card-header,
+  .card-group > .card:not(:first-child):not(:last-child):not(:only-child) .card-footer {
+    border-radius: 0;
+  }
+}
+
+.card-columns .card {
+  margin-bottom: 0.75rem;
+}
+@media (min-width: 576px) {
+  .card-columns {
+    -webkit-column-count: 3;
+       -moz-column-count: 3;
+            column-count: 3;
+    -webkit-column-gap: 1.25rem;
+       -moz-column-gap: 1.25rem;
+            column-gap: 1.25rem;
+  }
+  .card-columns .card {
+    display: inline-block;
+    width: 100%;
+  }
+}
+
+.breadcrumb {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+  padding: 0.75rem 1rem;
+  margin-bottom: 1rem;
+  list-style: none;
+  background-color: #e9ecef;
+  border-radius: 0.25rem;
+}
+
+.breadcrumb-item + .breadcrumb-item::before {
+  display: inline-block;
+  padding-right: 0.5rem;
+  padding-left: 0.5rem;
+  color: #6c757d;
+  content: "/";
+}
+.breadcrumb-item + .breadcrumb-item:hover::before {
+  text-decoration: underline;
+}
+.breadcrumb-item + .breadcrumb-item:hover::before {
+  text-decoration: none;
+}
+.breadcrumb-item.active {
+  color: #6c757d;
+}
+
+.pagination {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  padding-left: 0;
+  list-style: none;
+  border-radius: 0.25rem;
+}
+
+.page-link {
+  position: relative;
+  display: block;
+  padding: 0.5rem 0.75rem;
+  margin-left: -1px;
+  line-height: 1.25;
+  color: #007bff;
+  background-color: #fff;
+  border: 1px solid #dee2e6;
+}
+.page-link:hover {
+  color: #0056b3;
+  text-decoration: none;
+  background-color: #e9ecef;
+  border-color: #dee2e6;
+}
+.page-link:focus {
+  z-index: 2;
+  outline: 0;
+  -webkit-box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+          box-shadow: 0 0 0 0.2rem rgba(0, 123, 255, 0.25);
+}
+.page-link:not(:disabled):not(.disabled) {
+  cursor: pointer;
+}
+
+.page-item:first-child .page-link {
+  margin-left: 0;
+  border-top-left-radius: 0.25rem;
+  border-bottom-left-radius: 0.25rem;
+}
+.page-item:last-child .page-link {
+  border-top-right-radius: 0.25rem;
+  border-bottom-right-radius: 0.25rem;
+}
+.page-item.active .page-link {
+  z-index: 1;
+  color: #fff;
+  background-color: #007bff;
+  border-color: #007bff;
+}
+.page-item.disabled .page-link {
+  color: #6c757d;
+  pointer-events: none;
+  cursor: auto;
+  background-color: #fff;
+  border-color: #dee2e6;
+}
+
+.pagination-lg .page-link {
+  padding: 0.75rem 1.5rem;
+  font-size: 1.25rem;
+  line-height: 1.5;
+}
+.pagination-lg .page-item:first-child .page-link {
+  border-top-left-radius: 0.3rem;
+  border-bottom-left-radius: 0.3rem;
+}
+.pagination-lg .page-item:last-child .page-link {
+  border-top-right-radius: 0.3rem;
+  border-bottom-right-radius: 0.3rem;
+}
+
+.pagination-sm .page-link {
+  padding: 0.25rem 0.5rem;
+  font-size: 0.875rem;
+  line-height: 1.5;
+}
+.pagination-sm .page-item:first-child .page-link {
+  border-top-left-radius: 0.2rem;
+  border-bottom-left-radius: 0.2rem;
+}
+.pagination-sm .page-item:last-child .page-link {
+  border-top-right-radius: 0.2rem;
+  border-bottom-right-radius: 0.2rem;
+}
+
+.badge {
+  display: inline-block;
+  padding: 0.25em 0.4em;
+  font-size: 75%;
+  font-weight: 700;
+  line-height: 1;
+  text-align: center;
+  white-space: nowrap;
+  vertical-align: baseline;
+  border-radius: 0.25rem;
+}
+.badge:empty {
+  display: none;
+}
+
+.btn .badge {
+  position: relative;
+  top: -1px;
+}
+
+.badge-pill {
+  padding-right: 0.6em;
+  padding-left: 0.6em;
+  border-radius: 10rem;
+}
+
+.badge-primary {
+  color: #fff;
+  background-color: #007bff;
+}
+.badge-primary[href]:hover, .badge-primary[href]:focus {
+  color: #fff;
+  text-decoration: none;
+  background-color: #0062cc;
+}
+
+.badge-secondary {
+  color: #fff;
+  background-color: #6c757d;
+}
+.badge-secondary[href]:hover, .badge-secondary[href]:focus {
+  color: #fff;
+  text-decoration: none;
+  background-color: #545b62;
+}
+
+.badge-success {
+  color: #fff;
+  background-color: #28a745;
+}
+.badge-success[href]:hover, .badge-success[href]:focus {
+  color: #fff;
+  text-decoration: none;
+  background-color: #1e7e34;
+}
+
+.badge-info {
+  color: #fff;
+  background-color: #17a2b8;
+}
+.badge-info[href]:hover, .badge-info[href]:focus {
+  color: #fff;
+  text-decoration: none;
+  background-color: #117a8b;
+}
+
+.badge-warning {
+  color: #212529;
+  background-color: #ffc107;
+}
+.badge-warning[href]:hover, .badge-warning[href]:focus {
+  color: #212529;
+  text-decoration: none;
+  background-color: #d39e00;
+}
+
+.badge-danger {
+  color: #fff;
+  background-color: #dc3545;
+}
+.badge-danger[href]:hover, .badge-danger[href]:focus {
+  color: #fff;
+  text-decoration: none;
+  background-color: #bd2130;
+}
+
+.badge-light {
+  color: #212529;
+  background-color: #f8f9fa;
+}
+.badge-light[href]:hover, .badge-light[href]:focus {
+  color: #212529;
+  text-decoration: none;
+  background-color: #dae0e5;
+}
+
+.badge-dark {
+  color: #fff;
+  background-color: #343a40;
+}
+.badge-dark[href]:hover, .badge-dark[href]:focus {
+  color: #fff;
+  text-decoration: none;
+  background-color: #1d2124;
+}
+
+.jumbotron {
+  padding: 2rem 1rem;
+  margin-bottom: 2rem;
+  background-color: #e9ecef;
+  border-radius: 0.3rem;
+}
+@media (min-width: 576px) {
+  .jumbotron {
+    padding: 4rem 2rem;
+  }
+}
+
+.jumbotron-fluid {
+  padding-right: 0;
+  padding-left: 0;
+  border-radius: 0;
+}
+
+.alert {
+  position: relative;
+  padding: 0.75rem 1.25rem;
+  margin-bottom: 1rem;
+  border: 1px solid transparent;
+  border-radius: 0.25rem;
+}
+
+.alert-heading {
+  color: inherit;
+}
+
+.alert-link {
+  font-weight: 700;
+}
+
+.alert-dismissible {
+  padding-right: 4rem;
+}
+.alert-dismissible .close {
+  position: absolute;
+  top: 0;
+  right: 0;
+  padding: 0.75rem 1.25rem;
+  color: inherit;
+}
+
+.alert-primary {
+  color: #004085;
+  background-color: #cce5ff;
+  border-color: #b8daff;
+}
+.alert-primary hr {
+  border-top-color: #9fcdff;
+}
+.alert-primary .alert-link {
+  color: #002752;
+}
+
+.alert-secondary {
+  color: #383d41;
+  background-color: #e2e3e5;
+  border-color: #d6d8db;
+}
+.alert-secondary hr {
+  border-top-color: #c8cbcf;
+}
+.alert-secondary .alert-link {
+  color: #202326;
+}
+
+.alert-success {
+  color: #155724;
+  background-color: #d4edda;
+  border-color: #c3e6cb;
+}
+.alert-success hr {
+  border-top-color: #b1dfbb;
+}
+.alert-success .alert-link {
+  color: #0b2e13;
+}
+
+.alert-info {
+  color: #0c5460;
+  background-color: #d1ecf1;
+  border-color: #bee5eb;
+}
+.alert-info hr {
+  border-top-color: #abdde5;
+}
+.alert-info .alert-link {
+  color: #062c33;
+}
+
+.alert-warning {
+  color: #856404;
+  background-color: #fff3cd;
+  border-color: #ffeeba;
+}
+.alert-warning hr {
+  border-top-color: #ffe8a1;
+}
+.alert-warning .alert-link {
+  color: #533f03;
+}
+
+.alert-danger {
+  color: #721c24;
+  background-color: #f8d7da;
+  border-color: #f5c6cb;
+}
+.alert-danger hr {
+  border-top-color: #f1b0b7;
+}
+.alert-danger .alert-link {
+  color: #491217;
+}
+
+.alert-light {
+  color: #818182;
+  background-color: #fefefe;
+  border-color: #fdfdfe;
+}
+.alert-light hr {
+  border-top-color: #ececf6;
+}
+.alert-light .alert-link {
+  color: #686868;
+}
+
+.alert-dark {
+  color: #1b1e21;
+  background-color: #d6d8d9;
+  border-color: #c6c8ca;
+}
+.alert-dark hr {
+  border-top-color: #b9bbbe;
+}
+.alert-dark .alert-link {
+  color: #040505;
+}
+
+@-webkit-keyframes progress-bar-stripes {
+  from {
+    background-position: 1rem 0;
+  }
+  to {
+    background-position: 0 0;
+  }
+}
+
+@keyframes progress-bar-stripes {
+  from {
+    background-position: 1rem 0;
+  }
+  to {
+    background-position: 0 0;
+  }
+}
+.progress {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  height: 1rem;
+  overflow: hidden;
+  font-size: 0.75rem;
+  background-color: #e9ecef;
+  border-radius: 0.25rem;
+}
+
+.progress-bar {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+  -webkit-box-pack: center;
+      -ms-flex-pack: center;
+          justify-content: center;
+  color: #fff;
+  text-align: center;
+  background-color: #007bff;
+  -webkit-transition: width 0.6s ease;
+  transition: width 0.6s ease;
+}
+
+.progress-bar-striped {
+  background-image: linear-gradient(45deg, rgba(255, 255, 255, 0.15) 25%, transparent 25%, transparent 50%, rgba(255, 255, 255, 0.15) 50%, rgba(255, 255, 255, 0.15) 75%, transparent 75%, transparent);
+  background-size: 1rem 1rem;
+}
+
+.progress-bar-animated {
+  -webkit-animation: progress-bar-stripes 1s linear infinite;
+          animation: progress-bar-stripes 1s linear infinite;
+}
+
+.media {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: start;
+      -ms-flex-align: start;
+          align-items: flex-start;
+}
+
+.media-body {
+  -webkit-box-flex: 1;
+      -ms-flex: 1;
+          flex: 1;
+}
+
+.list-group {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+  padding-left: 0;
+  margin-bottom: 0;
+}
+
+.list-group-item-action {
+  width: 100%;
+  color: #495057;
+  text-align: inherit;
+}
+.list-group-item-action:hover, .list-group-item-action:focus {
+  color: #495057;
+  text-decoration: none;
+  background-color: #f8f9fa;
+}
+.list-group-item-action:active {
+  color: #212529;
+  background-color: #e9ecef;
+}
+
+.list-group-item {
+  position: relative;
+  display: block;
+  padding: 0.75rem 1.25rem;
+  margin-bottom: -1px;
+  background-color: #fff;
+  border: 1px solid rgba(0, 0, 0, 0.125);
+}
+.list-group-item:first-child {
+  border-top-left-radius: 0.25rem;
+  border-top-right-radius: 0.25rem;
+}
+.list-group-item:last-child {
+  margin-bottom: 0;
+  border-bottom-right-radius: 0.25rem;
+  border-bottom-left-radius: 0.25rem;
+}
+.list-group-item:hover, .list-group-item:focus {
+  z-index: 1;
+  text-decoration: none;
+}
+.list-group-item.disabled, .list-group-item:disabled {
+  color: #6c757d;
+  background-color: #fff;
+}
+.list-group-item.active {
+  z-index: 2;
+  color: #fff;
+  background-color: #007bff;
+  border-color: #007bff;
+}
+
+.list-group-flush .list-group-item {
+  border-right: 0;
+  border-left: 0;
+  border-radius: 0;
+}
+.list-group-flush:first-child .list-group-item:first-child {
+  border-top: 0;
+}
+.list-group-flush:last-child .list-group-item:last-child {
+  border-bottom: 0;
+}
+
+.list-group-item-primary {
+  color: #004085;
+  background-color: #b8daff;
+}
+.list-group-item-primary.list-group-item-action:hover, .list-group-item-primary.list-group-item-action:focus {
+  color: #004085;
+  background-color: #9fcdff;
+}
+.list-group-item-primary.list-group-item-action.active {
+  color: #fff;
+  background-color: #004085;
+  border-color: #004085;
+}
+
+.list-group-item-secondary {
+  color: #383d41;
+  background-color: #d6d8db;
+}
+.list-group-item-secondary.list-group-item-action:hover, .list-group-item-secondary.list-group-item-action:focus {
+  color: #383d41;
+  background-color: #c8cbcf;
+}
+.list-group-item-secondary.list-group-item-action.active {
+  color: #fff;
+  background-color: #383d41;
+  border-color: #383d41;
+}
+
+.list-group-item-success {
+  color: #155724;
+  background-color: #c3e6cb;
+}
+.list-group-item-success.list-group-item-action:hover, .list-group-item-success.list-group-item-action:focus {
+  color: #155724;
+  background-color: #b1dfbb;
+}
+.list-group-item-success.list-group-item-action.active {
+  color: #fff;
+  background-color: #155724;
+  border-color: #155724;
+}
+
+.list-group-item-info {
+  color: #0c5460;
+  background-color: #bee5eb;
+}
+.list-group-item-info.list-group-item-action:hover, .list-group-item-info.list-group-item-action:focus {
+  color: #0c5460;
+  background-color: #abdde5;
+}
+.list-group-item-info.list-group-item-action.active {
+  color: #fff;
+  background-color: #0c5460;
+  border-color: #0c5460;
+}
+
+.list-group-item-warning {
+  color: #856404;
+  background-color: #ffeeba;
+}
+.list-group-item-warning.list-group-item-action:hover, .list-group-item-warning.list-group-item-action:focus {
+  color: #856404;
+  background-color: #ffe8a1;
+}
+.list-group-item-warning.list-group-item-action.active {
+  color: #fff;
+  background-color: #856404;
+  border-color: #856404;
+}
+
+.list-group-item-danger {
+  color: #721c24;
+  background-color: #f5c6cb;
+}
+.list-group-item-danger.list-group-item-action:hover, .list-group-item-danger.list-group-item-action:focus {
+  color: #721c24;
+  background-color: #f1b0b7;
+}
+.list-group-item-danger.list-group-item-action.active {
+  color: #fff;
+  background-color: #721c24;
+  border-color: #721c24;
+}
+
+.list-group-item-light {
+  color: #818182;
+  background-color: #fdfdfe;
+}
+.list-group-item-light.list-group-item-action:hover, .list-group-item-light.list-group-item-action:focus {
+  color: #818182;
+  background-color: #ececf6;
+}
+.list-group-item-light.list-group-item-action.active {
+  color: #fff;
+  background-color: #818182;
+  border-color: #818182;
+}
+
+.list-group-item-dark {
+  color: #1b1e21;
+  background-color: #c6c8ca;
+}
+.list-group-item-dark.list-group-item-action:hover, .list-group-item-dark.list-group-item-action:focus {
+  color: #1b1e21;
+  background-color: #b9bbbe;
+}
+.list-group-item-dark.list-group-item-action.active {
+  color: #fff;
+  background-color: #1b1e21;
+  border-color: #1b1e21;
+}
+
+.close {
+  float: right;
+  font-size: 1.5rem;
+  font-weight: 700;
+  line-height: 1;
+  color: #000;
+  text-shadow: 0 1px 0 #fff;
+  opacity: .5;
+}
+.close:hover, .close:focus {
+  color: #000;
+  text-decoration: none;
+  opacity: .75;
+}
+.close:not(:disabled):not(.disabled) {
+  cursor: pointer;
+}
+
+button.close {
+  padding: 0;
+  background-color: transparent;
+  border: 0;
+  -webkit-appearance: none;
+}
+
+.modal-open {
+  overflow: hidden;
+}
+
+.modal {
+  position: fixed;
+  top: 0;
+  right: 0;
+  bottom: 0;
+  left: 0;
+  z-index: 1050;
+  display: none;
+  overflow: hidden;
+  outline: 0;
+}
+.modal-open .modal {
+  overflow-x: hidden;
+  overflow-y: auto;
+}
+
+.modal-dialog {
+  position: relative;
+  width: auto;
+  margin: 0.5rem;
+  pointer-events: none;
+}
+.modal.fade .modal-dialog {
+  -webkit-transition: -webkit-transform 0.3s ease-out;
+  transition: -webkit-transform 0.3s ease-out;
+  transition: transform 0.3s ease-out;
+  transition: transform 0.3s ease-out, -webkit-transform 0.3s ease-out;
+  -webkit-transform: translate(0, -25%);
+          transform: translate(0, -25%);
+}
+.modal.show .modal-dialog {
+  -webkit-transform: translate(0, 0);
+          transform: translate(0, 0);
+}
+
+.modal-dialog-centered {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  min-height: calc(100% - (0.5rem * 2));
+}
+
+.modal-content {
+  position: relative;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-orient: vertical;
+  -webkit-box-direction: normal;
+      -ms-flex-direction: column;
+          flex-direction: column;
+  width: 100%;
+  pointer-events: auto;
+  background-color: #fff;
+  background-clip: padding-box;
+  border: 1px solid rgba(0, 0, 0, 0.2);
+  border-radius: 0.3rem;
+  outline: 0;
+}
+
+.modal-backdrop {
+  position: fixed;
+  top: 0;
+  right: 0;
+  bottom: 0;
+  left: 0;
+  z-index: 1040;
+  background-color: #000;
+}
+.modal-backdrop.fade {
+  opacity: 0;
+}
+.modal-backdrop.show {
+  opacity: 0.5;
+}
+
+.modal-header {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: start;
+      -ms-flex-align: start;
+          align-items: flex-start;
+  -webkit-box-pack: justify;
+      -ms-flex-pack: justify;
+          justify-content: space-between;
+  padding: 1rem;
+  border-bottom: 1px solid #e9ecef;
+  border-top-left-radius: 0.3rem;
+  border-top-right-radius: 0.3rem;
+}
+.modal-header .close {
+  padding: 1rem;
+  margin: -1rem -1rem -1rem auto;
+}
+
+.modal-title {
+  margin-bottom: 0;
+  line-height: 1.5;
+}
+
+.modal-body {
+  position: relative;
+  -webkit-box-flex: 1;
+      -ms-flex: 1 1 auto;
+          flex: 1 1 auto;
+  padding: 1rem;
+}
+
+.modal-footer {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  -webkit-box-pack: end;
+      -ms-flex-pack: end;
+          justify-content: flex-end;
+  padding: 1rem;
+  border-top: 1px solid #e9ecef;
+}
+.modal-footer > :not(:first-child) {
+  margin-left: .25rem;
+}
+.modal-footer > :not(:last-child) {
+  margin-right: .25rem;
+}
+
+.modal-scrollbar-measure {
+  position: absolute;
+  top: -9999px;
+  width: 50px;
+  height: 50px;
+  overflow: scroll;
+}
+
+@media (min-width: 576px) {
+  .modal-dialog {
+    max-width: 500px;
+    margin: 1.75rem auto;
+  }
+
+  .modal-dialog-centered {
+    min-height: calc(100% - (1.75rem * 2));
+  }
+
+  .modal-sm {
+    max-width: 300px;
+  }
+}
+@media (min-width: 992px) {
+  .modal-lg {
+    max-width: 800px;
+  }
+}
+.tooltip {
+  position: absolute;
+  z-index: 1070;
+  display: block;
+  margin: 0;
+  font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";
+  font-style: normal;
+  font-weight: 400;
+  line-height: 1.5;
+  text-align: left;
+  text-align: start;
+  text-decoration: none;
+  text-shadow: none;
+  text-transform: none;
+  letter-spacing: normal;
+  word-break: normal;
+  word-spacing: normal;
+  white-space: normal;
+  line-break: auto;
+  font-size: 0.875rem;
+  word-wrap: break-word;
+  opacity: 0;
+}
+.tooltip.show {
+  opacity: 0.9;
+}
+.tooltip .arrow {
+  position: absolute;
+  display: block;
+  width: 0.8rem;
+  height: 0.4rem;
+}
+.tooltip .arrow::before {
+  position: absolute;
+  content: "";
+  border-color: transparent;
+  border-style: solid;
+}
+
+.bs-tooltip-top, .bs-tooltip-auto[x-placement^="top"] {
+  padding: 0.4rem 0;
+}
+.bs-tooltip-top .arrow, .bs-tooltip-auto[x-placement^="top"] .arrow {
+  bottom: 0;
+}
+.bs-tooltip-top .arrow::before, .bs-tooltip-auto[x-placement^="top"] .arrow::before {
+  top: 0;
+  border-width: 0.4rem 0.4rem 0;
+  border-top-color: #000;
+}
+
+.bs-tooltip-right, .bs-tooltip-auto[x-placement^="right"] {
+  padding: 0 0.4rem;
+}
+.bs-tooltip-right .arrow, .bs-tooltip-auto[x-placement^="right"] .arrow {
+  left: 0;
+  width: 0.4rem;
+  height: 0.8rem;
+}
+.bs-tooltip-right .arrow::before, .bs-tooltip-auto[x-placement^="right"] .arrow::before {
+  right: 0;
+  border-width: 0.4rem 0.4rem 0.4rem 0;
+  border-right-color: #000;
+}
+
+.bs-tooltip-bottom, .bs-tooltip-auto[x-placement^="bottom"] {
+  padding: 0.4rem 0;
+}
+.bs-tooltip-bottom .arrow, .bs-tooltip-auto[x-placement^="bottom"] .arrow {
+  top: 0;
+}
+.bs-tooltip-bottom .arrow::before, .bs-tooltip-auto[x-placement^="bottom"] .arrow::before {
+  bottom: 0;
+  border-width: 0 0.4rem 0.4rem;
+  border-bottom-color: #000;
+}
+
+.bs-tooltip-left, .bs-tooltip-auto[x-placement^="left"] {
+  padding: 0 0.4rem;
+}
+.bs-tooltip-left .arrow, .bs-tooltip-auto[x-placement^="left"] .arrow {
+  right: 0;
+  width: 0.4rem;
+  height: 0.8rem;
+}
+.bs-tooltip-left .arrow::before, .bs-tooltip-auto[x-placement^="left"] .arrow::before {
+  left: 0;
+  border-width: 0.4rem 0 0.4rem 0.4rem;
+  border-left-color: #000;
+}
+
+.tooltip-inner {
+  max-width: 200px;
+  padding: 0.25rem 0.5rem;
+  color: #fff;
+  text-align: center;
+  background-color: #000;
+  border-radius: 0.25rem;
+}
+
+.popover {
+  position: absolute;
+  top: 0;
+  left: 0;
+  z-index: 1060;
+  display: block;
+  max-width: 276px;
+  font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol";
+  font-style: normal;
+  font-weight: 400;
+  line-height: 1.5;
+  text-align: left;
+  text-align: start;
+  text-decoration: none;
+  text-shadow: none;
+  text-transform: none;
+  letter-spacing: normal;
+  word-break: normal;
+  word-spacing: normal;
+  white-space: normal;
+  line-break: auto;
+  font-size: 0.875rem;
+  word-wrap: break-word;
+  background-color: #fff;
+  background-clip: padding-box;
+  border: 1px solid rgba(0, 0, 0, 0.2);
+  border-radius: 0.3rem;
+}
+.popover .arrow {
+  position: absolute;
+  display: block;
+  width: 1rem;
+  height: 0.5rem;
+  margin: 0 0.3rem;
+}
+.popover .arrow::before, .popover .arrow::after {
+  position: absolute;
+  display: block;
+  content: "";
+  border-color: transparent;
+  border-style: solid;
+}
+
+.bs-popover-top, .bs-popover-auto[x-placement^="top"] {
+  margin-bottom: 0.5rem;
+}
+.bs-popover-top .arrow, .bs-popover-auto[x-placement^="top"] .arrow {
+  bottom: calc((0.5rem + 1px) * -1);
+}
+.bs-popover-top .arrow::before, .bs-popover-auto[x-placement^="top"] .arrow::before,
+.bs-popover-top .arrow::after,
+.bs-popover-auto[x-placement^="top"] .arrow::after {
+  border-width: 0.5rem 0.5rem 0;
+}
+.bs-popover-top .arrow::before, .bs-popover-auto[x-placement^="top"] .arrow::before {
+  bottom: 0;
+  border-top-color: rgba(0, 0, 0, 0.25);
+}
+.bs-popover-top .arrow::after, .bs-popover-auto[x-placement^="top"] .arrow::after {
+  bottom: 1px;
+  border-top-color: #fff;
+}
+
+.bs-popover-right, .bs-popover-auto[x-placement^="right"] {
+  margin-left: 0.5rem;
+}
+.bs-popover-right .arrow, .bs-popover-auto[x-placement^="right"] .arrow {
+  left: calc((0.5rem + 1px) * -1);
+  width: 0.5rem;
+  height: 1rem;
+  margin: 0.3rem 0;
+}
+.bs-popover-right .arrow::before, .bs-popover-auto[x-placement^="right"] .arrow::before,
+.bs-popover-right .arrow::after,
+.bs-popover-auto[x-placement^="right"] .arrow::after {
+  border-width: 0.5rem 0.5rem 0.5rem 0;
+}
+.bs-popover-right .arrow::before, .bs-popover-auto[x-placement^="right"] .arrow::before {
+  left: 0;
+  border-right-color: rgba(0, 0, 0, 0.25);
+}
+.bs-popover-right .arrow::after, .bs-popover-auto[x-placement^="right"] .arrow::after {
+  left: 1px;
+  border-right-color: #fff;
+}
+
+.bs-popover-bottom, .bs-popover-auto[x-placement^="bottom"] {
+  margin-top: 0.5rem;
+}
+.bs-popover-bottom .arrow, .bs-popover-auto[x-placement^="bottom"] .arrow {
+  top: calc((0.5rem + 1px) * -1);
+}
+.bs-popover-bottom .arrow::before, .bs-popover-auto[x-placement^="bottom"] .arrow::before,
+.bs-popover-bottom .arrow::after,
+.bs-popover-auto[x-placement^="bottom"] .arrow::after {
+  border-width: 0 0.5rem 0.5rem 0.5rem;
+}
+.bs-popover-bottom .arrow::before, .bs-popover-auto[x-placement^="bottom"] .arrow::before {
+  top: 0;
+  border-bottom-color: rgba(0, 0, 0, 0.25);
+}
+.bs-popover-bottom .arrow::after, .bs-popover-auto[x-placement^="bottom"] .arrow::after {
+  top: 1px;
+  border-bottom-color: #fff;
+}
+.bs-popover-bottom .popover-header::before, .bs-popover-auto[x-placement^="bottom"] .popover-header::before {
+  position: absolute;
+  top: 0;
+  left: 50%;
+  display: block;
+  width: 1rem;
+  margin-left: -0.5rem;
+  content: "";
+  border-bottom: 1px solid #f7f7f7;
+}
+
+.bs-popover-left, .bs-popover-auto[x-placement^="left"] {
+  margin-right: 0.5rem;
+}
+.bs-popover-left .arrow, .bs-popover-auto[x-placement^="left"] .arrow {
+  right: calc((0.5rem + 1px) * -1);
+  width: 0.5rem;
+  height: 1rem;
+  margin: 0.3rem 0;
+}
+.bs-popover-left .arrow::before, .bs-popover-auto[x-placement^="left"] .arrow::before,
+.bs-popover-left .arrow::after,
+.bs-popover-auto[x-placement^="left"] .arrow::after {
+  border-width: 0.5rem 0 0.5rem 0.5rem;
+}
+.bs-popover-left .arrow::before, .bs-popover-auto[x-placement^="left"] .arrow::before {
+  right: 0;
+  border-left-color: rgba(0, 0, 0, 0.25);
+}
+.bs-popover-left .arrow::after, .bs-popover-auto[x-placement^="left"] .arrow::after {
+  right: 1px;
+  border-left-color: #fff;
+}
+
+.popover-header {
+  padding: 0.5rem 0.75rem;
+  margin-bottom: 0;
+  font-size: 1rem;
+  color: inherit;
+  background-color: #f7f7f7;
+  border-bottom: 1px solid #ebebeb;
+  border-top-left-radius: calc(0.3rem - 1px);
+  border-top-right-radius: calc(0.3rem - 1px);
+}
+.popover-header:empty {
+  display: none;
+}
+
+.popover-body {
+  padding: 0.5rem 0.75rem;
+  color: #212529;
+}
+
+.carousel {
+  position: relative;
+}
+
+.carousel-inner {
+  position: relative;
+  width: 100%;
+  overflow: hidden;
+}
+
+.carousel-item {
+  position: relative;
+  display: none;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  width: 100%;
+  -webkit-transition: -webkit-transform 0.6s ease;
+  transition: -webkit-transform 0.6s ease;
+  transition: transform 0.6s ease;
+  transition: transform 0.6s ease, -webkit-transform 0.6s ease;
+  -webkit-backface-visibility: hidden;
+          backface-visibility: hidden;
+  -webkit-perspective: 1000px;
+          perspective: 1000px;
+}
+
+.carousel-item.active,
+.carousel-item-next,
+.carousel-item-prev {
+  display: block;
+}
+
+.carousel-item-next,
+.carousel-item-prev {
+  position: absolute;
+  top: 0;
+}
+
+.carousel-item-next.carousel-item-left,
+.carousel-item-prev.carousel-item-right {
+  -webkit-transform: translateX(0);
+          transform: translateX(0);
+}
+@supports (transform-style: preserve-3d) {
+  .carousel-item-next.carousel-item-left,
+  .carousel-item-prev.carousel-item-right {
+    -webkit-transform: translate3d(0, 0, 0);
+            transform: translate3d(0, 0, 0);
+  }
+}
+
+.carousel-item-next,
+.active.carousel-item-right {
+  -webkit-transform: translateX(100%);
+          transform: translateX(100%);
+}
+@supports (transform-style: preserve-3d) {
+  .carousel-item-next,
+  .active.carousel-item-right {
+    -webkit-transform: translate3d(100%, 0, 0);
+            transform: translate3d(100%, 0, 0);
+  }
+}
+
+.carousel-item-prev,
+.active.carousel-item-left {
+  -webkit-transform: translateX(-100%);
+          transform: translateX(-100%);
+}
+@supports (transform-style: preserve-3d) {
+  .carousel-item-prev,
+  .active.carousel-item-left {
+    -webkit-transform: translate3d(-100%, 0, 0);
+            transform: translate3d(-100%, 0, 0);
+  }
+}
+
+.carousel-control-prev,
+.carousel-control-next {
+  position: absolute;
+  top: 0;
+  bottom: 0;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  -webkit-box-pack: center;
+      -ms-flex-pack: center;
+          justify-content: center;
+  width: 15%;
+  color: #fff;
+  text-align: center;
+  opacity: 0.5;
+}
+.carousel-control-prev:hover, .carousel-control-prev:focus,
+.carousel-control-next:hover,
+.carousel-control-next:focus {
+  color: #fff;
+  text-decoration: none;
+  outline: 0;
+  opacity: .9;
+}
+
+.carousel-control-prev {
+  left: 0;
+}
+
+.carousel-control-next {
+  right: 0;
+}
+
+.carousel-control-prev-icon,
+.carousel-control-next-icon {
+  display: inline-block;
+  width: 20px;
+  height: 20px;
+  background: transparent no-repeat center center;
+  background-size: 100% 100%;
+}
+
+.carousel-control-prev-icon {
+  background-image: url("data:image/svg+xml;charset=utf8,%3Csvg xmlns='http://www.w3.org/2000/svg' fill='%23fff' viewBox='0 0 8 8'%3E%3Cpath d='M5.25 0l-4 4 4 4 1.5-1.5-2.5-2.5 2.5-2.5-1.5-1.5z'/%3E%3C/svg%3E");
+}
+
+.carousel-control-next-icon {
+  background-image: url("data:image/svg+xml;charset=utf8,%3Csvg xmlns='http://www.w3.org/2000/svg' fill='%23fff' viewBox='0 0 8 8'%3E%3Cpath d='M2.75 0l-1.5 1.5 2.5 2.5-2.5 2.5 1.5 1.5 4-4-4-4z'/%3E%3C/svg%3E");
+}
+
+.carousel-indicators {
+  position: absolute;
+  right: 0;
+  bottom: 10px;
+  left: 0;
+  z-index: 15;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-pack: center;
+      -ms-flex-pack: center;
+          justify-content: center;
+  padding-left: 0;
+  margin-right: 15%;
+  margin-left: 15%;
+  list-style: none;
+}
+.carousel-indicators li {
+  position: relative;
+  -webkit-box-flex: 0;
+      -ms-flex: 0 1 auto;
+          flex: 0 1 auto;
+  width: 30px;
+  height: 3px;
+  margin-right: 3px;
+  margin-left: 3px;
+  text-indent: -999px;
+  background-color: rgba(255, 255, 255, 0.5);
+}
+.carousel-indicators li::before {
+  position: absolute;
+  top: -10px;
+  left: 0;
+  display: inline-block;
+  width: 100%;
+  height: 10px;
+  content: "";
+}
+.carousel-indicators li::after {
+  position: absolute;
+  bottom: -10px;
+  left: 0;
+  display: inline-block;
+  width: 100%;
+  height: 10px;
+  content: "";
+}
+.carousel-indicators .active {
+  background-color: #fff;
+}
+
+.carousel-caption {
+  position: absolute;
+  right: 15%;
+  bottom: 20px;
+  left: 15%;
+  z-index: 10;
+  padding-top: 20px;
+  padding-bottom: 20px;
+  color: #fff;
+  text-align: center;
+}
+
+.align-baseline {
+  vertical-align: baseline !important;
+}
+
+.align-top {
+  vertical-align: top !important;
+}
+
+.align-middle {
+  vertical-align: middle !important;
+}
+
+.align-bottom {
+  vertical-align: bottom !important;
+}
+
+.align-text-bottom {
+  vertical-align: text-bottom !important;
+}
+
+.align-text-top {
+  vertical-align: text-top !important;
+}
+
+.bg-primary {
+  background-color: #007bff !important;
+}
+
+a.bg-primary:hover, a.bg-primary:focus,
+button.bg-primary:hover,
+button.bg-primary:focus {
+  background-color: #0062cc !important;
+}
+
+.bg-secondary {
+  background-color: #6c757d !important;
+}
+
+a.bg-secondary:hover, a.bg-secondary:focus,
+button.bg-secondary:hover,
+button.bg-secondary:focus {
+  background-color: #545b62 !important;
+}
+
+.bg-success {
+  background-color: #28a745 !important;
+}
+
+a.bg-success:hover, a.bg-success:focus,
+button.bg-success:hover,
+button.bg-success:focus {
+  background-color: #1e7e34 !important;
+}
+
+.bg-info {
+  background-color: #17a2b8 !important;
+}
+
+a.bg-info:hover, a.bg-info:focus,
+button.bg-info:hover,
+button.bg-info:focus {
+  background-color: #117a8b !important;
+}
+
+.bg-warning {
+  background-color: #ffc107 !important;
+}
+
+a.bg-warning:hover, a.bg-warning:focus,
+button.bg-warning:hover,
+button.bg-warning:focus {
+  background-color: #d39e00 !important;
+}
+
+.bg-danger {
+  background-color: #dc3545 !important;
+}
+
+a.bg-danger:hover, a.bg-danger:focus,
+button.bg-danger:hover,
+button.bg-danger:focus {
+  background-color: #bd2130 !important;
+}
+
+.bg-light {
+  background-color: #f8f9fa !important;
+}
+
+a.bg-light:hover, a.bg-light:focus,
+button.bg-light:hover,
+button.bg-light:focus {
+  background-color: #dae0e5 !important;
+}
+
+.bg-dark {
+  background-color: #343a40 !important;
+}
+
+a.bg-dark:hover, a.bg-dark:focus,
+button.bg-dark:hover,
+button.bg-dark:focus {
+  background-color: #1d2124 !important;
+}
+
+.bg-white {
+  background-color: #fff !important;
+}
+
+.bg-transparent {
+  background-color: transparent !important;
+}
+
+.border {
+  border: 1px solid #dee2e6 !important;
+}
+
+.border-top {
+  border-top: 1px solid #dee2e6 !important;
+}
+
+.border-right {
+  border-right: 1px solid #dee2e6 !important;
+}
+
+.border-bottom {
+  border-bottom: 1px solid #dee2e6 !important;
+}
+
+.border-left {
+  border-left: 1px solid #dee2e6 !important;
+}
+
+.border-0 {
+  border: 0 !important;
+}
+
+.border-top-0 {
+  border-top: 0 !important;
+}
+
+.border-right-0 {
+  border-right: 0 !important;
+}
+
+.border-bottom-0 {
+  border-bottom: 0 !important;
+}
+
+.border-left-0 {
+  border-left: 0 !important;
+}
+
+.border-primary {
+  border-color: #007bff !important;
+}
+
+.border-secondary {
+  border-color: #6c757d !important;
+}
+
+.border-success {
+  border-color: #28a745 !important;
+}
+
+.border-info {
+  border-color: #17a2b8 !important;
+}
+
+.border-warning {
+  border-color: #ffc107 !important;
+}
+
+.border-danger {
+  border-color: #dc3545 !important;
+}
+
+.border-light {
+  border-color: #f8f9fa !important;
+}
+
+.border-dark {
+  border-color: #343a40 !important;
+}
+
+.border-white {
+  border-color: #fff !important;
+}
+
+.rounded {
+  border-radius: 0.25rem !important;
+}
+
+.rounded-top {
+  border-top-left-radius: 0.25rem !important;
+  border-top-right-radius: 0.25rem !important;
+}
+
+.rounded-right {
+  border-top-right-radius: 0.25rem !important;
+  border-bottom-right-radius: 0.25rem !important;
+}
+
+.rounded-bottom {
+  border-bottom-right-radius: 0.25rem !important;
+  border-bottom-left-radius: 0.25rem !important;
+}
+
+.rounded-left {
+  border-top-left-radius: 0.25rem !important;
+  border-bottom-left-radius: 0.25rem !important;
+}
+
+.rounded-circle {
+  border-radius: 50% !important;
+}
+
+.rounded-0 {
+  border-radius: 0 !important;
+}
+
+.clearfix::after {
+  display: block;
+  clear: both;
+  content: "";
+}
+
+.d-none {
+  display: none !important;
+}
+
+.d-inline {
+  display: inline !important;
+}
+
+.d-inline-block {
+  display: inline-block !important;
+}
+
+.d-block {
+  display: block !important;
+}
+
+.d-table {
+  display: table !important;
+}
+
+.d-table-row {
+  display: table-row !important;
+}
+
+.d-table-cell {
+  display: table-cell !important;
+}
+
+.d-flex {
+  display: -webkit-box !important;
+  display: -ms-flexbox !important;
+  display: flex !important;
+}
+
+.d-inline-flex {
+  display: -webkit-inline-box !important;
+  display: -ms-inline-flexbox !important;
+  display: inline-flex !important;
+}
+
+@media (min-width: 576px) {
+  .d-sm-none {
+    display: none !important;
+  }
+
+  .d-sm-inline {
+    display: inline !important;
+  }
+
+  .d-sm-inline-block {
+    display: inline-block !important;
+  }
+
+  .d-sm-block {
+    display: block !important;
+  }
+
+  .d-sm-table {
+    display: table !important;
+  }
+
+  .d-sm-table-row {
+    display: table-row !important;
+  }
+
+  .d-sm-table-cell {
+    display: table-cell !important;
+  }
+
+  .d-sm-flex {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+  }
+
+  .d-sm-inline-flex {
+    display: -webkit-inline-box !important;
+    display: -ms-inline-flexbox !important;
+    display: inline-flex !important;
+  }
+}
+@media (min-width: 768px) {
+  .d-md-none {
+    display: none !important;
+  }
+
+  .d-md-inline {
+    display: inline !important;
+  }
+
+  .d-md-inline-block {
+    display: inline-block !important;
+  }
+
+  .d-md-block {
+    display: block !important;
+  }
+
+  .d-md-table {
+    display: table !important;
+  }
+
+  .d-md-table-row {
+    display: table-row !important;
+  }
+
+  .d-md-table-cell {
+    display: table-cell !important;
+  }
+
+  .d-md-flex {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+  }
+
+  .d-md-inline-flex {
+    display: -webkit-inline-box !important;
+    display: -ms-inline-flexbox !important;
+    display: inline-flex !important;
+  }
+}
+@media (min-width: 992px) {
+  .d-lg-none {
+    display: none !important;
+  }
+
+  .d-lg-inline {
+    display: inline !important;
+  }
+
+  .d-lg-inline-block {
+    display: inline-block !important;
+  }
+
+  .d-lg-block {
+    display: block !important;
+  }
+
+  .d-lg-table {
+    display: table !important;
+  }
+
+  .d-lg-table-row {
+    display: table-row !important;
+  }
+
+  .d-lg-table-cell {
+    display: table-cell !important;
+  }
+
+  .d-lg-flex {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+  }
+
+  .d-lg-inline-flex {
+    display: -webkit-inline-box !important;
+    display: -ms-inline-flexbox !important;
+    display: inline-flex !important;
+  }
+}
+@media (min-width: 1200px) {
+  .d-xl-none {
+    display: none !important;
+  }
+
+  .d-xl-inline {
+    display: inline !important;
+  }
+
+  .d-xl-inline-block {
+    display: inline-block !important;
+  }
+
+  .d-xl-block {
+    display: block !important;
+  }
+
+  .d-xl-table {
+    display: table !important;
+  }
+
+  .d-xl-table-row {
+    display: table-row !important;
+  }
+
+  .d-xl-table-cell {
+    display: table-cell !important;
+  }
+
+  .d-xl-flex {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+  }
+
+  .d-xl-inline-flex {
+    display: -webkit-inline-box !important;
+    display: -ms-inline-flexbox !important;
+    display: inline-flex !important;
+  }
+}
+@media print {
+  .d-print-none {
+    display: none !important;
+  }
+
+  .d-print-inline {
+    display: inline !important;
+  }
+
+  .d-print-inline-block {
+    display: inline-block !important;
+  }
+
+  .d-print-block {
+    display: block !important;
+  }
+
+  .d-print-table {
+    display: table !important;
+  }
+
+  .d-print-table-row {
+    display: table-row !important;
+  }
+
+  .d-print-table-cell {
+    display: table-cell !important;
+  }
+
+  .d-print-flex {
+    display: -webkit-box !important;
+    display: -ms-flexbox !important;
+    display: flex !important;
+  }
+
+  .d-print-inline-flex {
+    display: -webkit-inline-box !important;
+    display: -ms-inline-flexbox !important;
+    display: inline-flex !important;
+  }
+}
+.embed-responsive {
+  position: relative;
+  display: block;
+  width: 100%;
+  padding: 0;
+  overflow: hidden;
+}
+.embed-responsive::before {
+  display: block;
+  content: "";
+}
+.embed-responsive .embed-responsive-item,
+.embed-responsive iframe,
+.embed-responsive embed,
+.embed-responsive object,
+.embed-responsive video {
+  position: absolute;
+  top: 0;
+  bottom: 0;
+  left: 0;
+  width: 100%;
+  height: 100%;
+  border: 0;
+}
+
+.embed-responsive-21by9::before {
+  padding-top: 42.8571428571%;
+}
+
+.embed-responsive-16by9::before {
+  padding-top: 56.25%;
+}
+
+.embed-responsive-4by3::before {
+  padding-top: 75%;
+}
+
+.embed-responsive-1by1::before {
+  padding-top: 100%;
+}
+
+.flex-row {
+  -webkit-box-orient: horizontal !important;
+  -webkit-box-direction: normal !important;
+      -ms-flex-direction: row !important;
+          flex-direction: row !important;
+}
+
+.flex-column {
+  -webkit-box-orient: vertical !important;
+  -webkit-box-direction: normal !important;
+      -ms-flex-direction: column !important;
+          flex-direction: column !important;
+}
+
+.flex-row-reverse {
+  -webkit-box-orient: horizontal !important;
+  -webkit-box-direction: reverse !important;
+      -ms-flex-direction: row-reverse !important;
+          flex-direction: row-reverse !important;
+}
+
+.flex-column-reverse {
+  -webkit-box-orient: vertical !important;
+  -webkit-box-direction: reverse !important;
+      -ms-flex-direction: column-reverse !important;
+          flex-direction: column-reverse !important;
+}
+
+.flex-wrap {
+  -ms-flex-wrap: wrap !important;
+      flex-wrap: wrap !important;
+}
+
+.flex-nowrap {
+  -ms-flex-wrap: nowrap !important;
+      flex-wrap: nowrap !important;
+}
+
+.flex-wrap-reverse {
+  -ms-flex-wrap: wrap-reverse !important;
+      flex-wrap: wrap-reverse !important;
+}
+
+.justify-content-start {
+  -webkit-box-pack: start !important;
+      -ms-flex-pack: start !important;
+          justify-content: flex-start !important;
+}
+
+.justify-content-end {
+  -webkit-box-pack: end !important;
+      -ms-flex-pack: end !important;
+          justify-content: flex-end !important;
+}
+
+.justify-content-center {
+  -webkit-box-pack: center !important;
+      -ms-flex-pack: center !important;
+          justify-content: center !important;
+}
+
+.justify-content-between {
+  -webkit-box-pack: justify !important;
+      -ms-flex-pack: justify !important;
+          justify-content: space-between !important;
+}
+
+.justify-content-around {
+  -ms-flex-pack: distribute !important;
+      justify-content: space-around !important;
+}
+
+.align-items-start {
+  -webkit-box-align: start !important;
+      -ms-flex-align: start !important;
+          align-items: flex-start !important;
+}
+
+.align-items-end {
+  -webkit-box-align: end !important;
+      -ms-flex-align: end !important;
+          align-items: flex-end !important;
+}
+
+.align-items-center {
+  -webkit-box-align: center !important;
+      -ms-flex-align: center !important;
+          align-items: center !important;
+}
+
+.align-items-baseline {
+  -webkit-box-align: baseline !important;
+      -ms-flex-align: baseline !important;
+          align-items: baseline !important;
+}
+
+.align-items-stretch {
+  -webkit-box-align: stretch !important;
+      -ms-flex-align: stretch !important;
+          align-items: stretch !important;
+}
+
+.align-content-start {
+  -ms-flex-line-pack: start !important;
+      align-content: flex-start !important;
+}
+
+.align-content-end {
+  -ms-flex-line-pack: end !important;
+      align-content: flex-end !important;
+}
+
+.align-content-center {
+  -ms-flex-line-pack: center !important;
+      align-content: center !important;
+}
+
+.align-content-between {
+  -ms-flex-line-pack: justify !important;
+      align-content: space-between !important;
+}
+
+.align-content-around {
+  -ms-flex-line-pack: distribute !important;
+      align-content: space-around !important;
+}
+
+.align-content-stretch {
+  -ms-flex-line-pack: stretch !important;
+      align-content: stretch !important;
+}
+
+.align-self-auto {
+  -ms-flex-item-align: auto !important;
+      align-self: auto !important;
+}
+
+.align-self-start {
+  -ms-flex-item-align: start !important;
+      align-self: flex-start !important;
+}
+
+.align-self-end {
+  -ms-flex-item-align: end !important;
+      align-self: flex-end !important;
+}
+
+.align-self-center {
+  -ms-flex-item-align: center !important;
+      align-self: center !important;
+}
+
+.align-self-baseline {
+  -ms-flex-item-align: baseline !important;
+      align-self: baseline !important;
+}
+
+.align-self-stretch {
+  -ms-flex-item-align: stretch !important;
+      align-self: stretch !important;
+}
+
+@media (min-width: 576px) {
+  .flex-sm-row {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: row !important;
+            flex-direction: row !important;
+  }
+
+  .flex-sm-column {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: column !important;
+            flex-direction: column !important;
+  }
+
+  .flex-sm-row-reverse {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: row-reverse !important;
+            flex-direction: row-reverse !important;
+  }
+
+  .flex-sm-column-reverse {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: column-reverse !important;
+            flex-direction: column-reverse !important;
+  }
+
+  .flex-sm-wrap {
+    -ms-flex-wrap: wrap !important;
+        flex-wrap: wrap !important;
+  }
+
+  .flex-sm-nowrap {
+    -ms-flex-wrap: nowrap !important;
+        flex-wrap: nowrap !important;
+  }
+
+  .flex-sm-wrap-reverse {
+    -ms-flex-wrap: wrap-reverse !important;
+        flex-wrap: wrap-reverse !important;
+  }
+
+  .justify-content-sm-start {
+    -webkit-box-pack: start !important;
+        -ms-flex-pack: start !important;
+            justify-content: flex-start !important;
+  }
+
+  .justify-content-sm-end {
+    -webkit-box-pack: end !important;
+        -ms-flex-pack: end !important;
+            justify-content: flex-end !important;
+  }
+
+  .justify-content-sm-center {
+    -webkit-box-pack: center !important;
+        -ms-flex-pack: center !important;
+            justify-content: center !important;
+  }
+
+  .justify-content-sm-between {
+    -webkit-box-pack: justify !important;
+        -ms-flex-pack: justify !important;
+            justify-content: space-between !important;
+  }
+
+  .justify-content-sm-around {
+    -ms-flex-pack: distribute !important;
+        justify-content: space-around !important;
+  }
+
+  .align-items-sm-start {
+    -webkit-box-align: start !important;
+        -ms-flex-align: start !important;
+            align-items: flex-start !important;
+  }
+
+  .align-items-sm-end {
+    -webkit-box-align: end !important;
+        -ms-flex-align: end !important;
+            align-items: flex-end !important;
+  }
+
+  .align-items-sm-center {
+    -webkit-box-align: center !important;
+        -ms-flex-align: center !important;
+            align-items: center !important;
+  }
+
+  .align-items-sm-baseline {
+    -webkit-box-align: baseline !important;
+        -ms-flex-align: baseline !important;
+            align-items: baseline !important;
+  }
+
+  .align-items-sm-stretch {
+    -webkit-box-align: stretch !important;
+        -ms-flex-align: stretch !important;
+            align-items: stretch !important;
+  }
+
+  .align-content-sm-start {
+    -ms-flex-line-pack: start !important;
+        align-content: flex-start !important;
+  }
+
+  .align-content-sm-end {
+    -ms-flex-line-pack: end !important;
+        align-content: flex-end !important;
+  }
+
+  .align-content-sm-center {
+    -ms-flex-line-pack: center !important;
+        align-content: center !important;
+  }
+
+  .align-content-sm-between {
+    -ms-flex-line-pack: justify !important;
+        align-content: space-between !important;
+  }
+
+  .align-content-sm-around {
+    -ms-flex-line-pack: distribute !important;
+        align-content: space-around !important;
+  }
+
+  .align-content-sm-stretch {
+    -ms-flex-line-pack: stretch !important;
+        align-content: stretch !important;
+  }
+
+  .align-self-sm-auto {
+    -ms-flex-item-align: auto !important;
+        align-self: auto !important;
+  }
+
+  .align-self-sm-start {
+    -ms-flex-item-align: start !important;
+        align-self: flex-start !important;
+  }
+
+  .align-self-sm-end {
+    -ms-flex-item-align: end !important;
+        align-self: flex-end !important;
+  }
+
+  .align-self-sm-center {
+    -ms-flex-item-align: center !important;
+        align-self: center !important;
+  }
+
+  .align-self-sm-baseline {
+    -ms-flex-item-align: baseline !important;
+        align-self: baseline !important;
+  }
+
+  .align-self-sm-stretch {
+    -ms-flex-item-align: stretch !important;
+        align-self: stretch !important;
+  }
+}
+@media (min-width: 768px) {
+  .flex-md-row {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: row !important;
+            flex-direction: row !important;
+  }
+
+  .flex-md-column {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: column !important;
+            flex-direction: column !important;
+  }
+
+  .flex-md-row-reverse {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: row-reverse !important;
+            flex-direction: row-reverse !important;
+  }
+
+  .flex-md-column-reverse {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: column-reverse !important;
+            flex-direction: column-reverse !important;
+  }
+
+  .flex-md-wrap {
+    -ms-flex-wrap: wrap !important;
+        flex-wrap: wrap !important;
+  }
+
+  .flex-md-nowrap {
+    -ms-flex-wrap: nowrap !important;
+        flex-wrap: nowrap !important;
+  }
+
+  .flex-md-wrap-reverse {
+    -ms-flex-wrap: wrap-reverse !important;
+        flex-wrap: wrap-reverse !important;
+  }
+
+  .justify-content-md-start {
+    -webkit-box-pack: start !important;
+        -ms-flex-pack: start !important;
+            justify-content: flex-start !important;
+  }
+
+  .justify-content-md-end {
+    -webkit-box-pack: end !important;
+        -ms-flex-pack: end !important;
+            justify-content: flex-end !important;
+  }
+
+  .justify-content-md-center {
+    -webkit-box-pack: center !important;
+        -ms-flex-pack: center !important;
+            justify-content: center !important;
+  }
+
+  .justify-content-md-between {
+    -webkit-box-pack: justify !important;
+        -ms-flex-pack: justify !important;
+            justify-content: space-between !important;
+  }
+
+  .justify-content-md-around {
+    -ms-flex-pack: distribute !important;
+        justify-content: space-around !important;
+  }
+
+  .align-items-md-start {
+    -webkit-box-align: start !important;
+        -ms-flex-align: start !important;
+            align-items: flex-start !important;
+  }
+
+  .align-items-md-end {
+    -webkit-box-align: end !important;
+        -ms-flex-align: end !important;
+            align-items: flex-end !important;
+  }
+
+  .align-items-md-center {
+    -webkit-box-align: center !important;
+        -ms-flex-align: center !important;
+            align-items: center !important;
+  }
+
+  .align-items-md-baseline {
+    -webkit-box-align: baseline !important;
+        -ms-flex-align: baseline !important;
+            align-items: baseline !important;
+  }
+
+  .align-items-md-stretch {
+    -webkit-box-align: stretch !important;
+        -ms-flex-align: stretch !important;
+            align-items: stretch !important;
+  }
+
+  .align-content-md-start {
+    -ms-flex-line-pack: start !important;
+        align-content: flex-start !important;
+  }
+
+  .align-content-md-end {
+    -ms-flex-line-pack: end !important;
+        align-content: flex-end !important;
+  }
+
+  .align-content-md-center {
+    -ms-flex-line-pack: center !important;
+        align-content: center !important;
+  }
+
+  .align-content-md-between {
+    -ms-flex-line-pack: justify !important;
+        align-content: space-between !important;
+  }
+
+  .align-content-md-around {
+    -ms-flex-line-pack: distribute !important;
+        align-content: space-around !important;
+  }
+
+  .align-content-md-stretch {
+    -ms-flex-line-pack: stretch !important;
+        align-content: stretch !important;
+  }
+
+  .align-self-md-auto {
+    -ms-flex-item-align: auto !important;
+        align-self: auto !important;
+  }
+
+  .align-self-md-start {
+    -ms-flex-item-align: start !important;
+        align-self: flex-start !important;
+  }
+
+  .align-self-md-end {
+    -ms-flex-item-align: end !important;
+        align-self: flex-end !important;
+  }
+
+  .align-self-md-center {
+    -ms-flex-item-align: center !important;
+        align-self: center !important;
+  }
+
+  .align-self-md-baseline {
+    -ms-flex-item-align: baseline !important;
+        align-self: baseline !important;
+  }
+
+  .align-self-md-stretch {
+    -ms-flex-item-align: stretch !important;
+        align-self: stretch !important;
+  }
+}
+@media (min-width: 992px) {
+  .flex-lg-row {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: row !important;
+            flex-direction: row !important;
+  }
+
+  .flex-lg-column {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: column !important;
+            flex-direction: column !important;
+  }
+
+  .flex-lg-row-reverse {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: row-reverse !important;
+            flex-direction: row-reverse !important;
+  }
+
+  .flex-lg-column-reverse {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: column-reverse !important;
+            flex-direction: column-reverse !important;
+  }
+
+  .flex-lg-wrap {
+    -ms-flex-wrap: wrap !important;
+        flex-wrap: wrap !important;
+  }
+
+  .flex-lg-nowrap {
+    -ms-flex-wrap: nowrap !important;
+        flex-wrap: nowrap !important;
+  }
+
+  .flex-lg-wrap-reverse {
+    -ms-flex-wrap: wrap-reverse !important;
+        flex-wrap: wrap-reverse !important;
+  }
+
+  .justify-content-lg-start {
+    -webkit-box-pack: start !important;
+        -ms-flex-pack: start !important;
+            justify-content: flex-start !important;
+  }
+
+  .justify-content-lg-end {
+    -webkit-box-pack: end !important;
+        -ms-flex-pack: end !important;
+            justify-content: flex-end !important;
+  }
+
+  .justify-content-lg-center {
+    -webkit-box-pack: center !important;
+        -ms-flex-pack: center !important;
+            justify-content: center !important;
+  }
+
+  .justify-content-lg-between {
+    -webkit-box-pack: justify !important;
+        -ms-flex-pack: justify !important;
+            justify-content: space-between !important;
+  }
+
+  .justify-content-lg-around {
+    -ms-flex-pack: distribute !important;
+        justify-content: space-around !important;
+  }
+
+  .align-items-lg-start {
+    -webkit-box-align: start !important;
+        -ms-flex-align: start !important;
+            align-items: flex-start !important;
+  }
+
+  .align-items-lg-end {
+    -webkit-box-align: end !important;
+        -ms-flex-align: end !important;
+            align-items: flex-end !important;
+  }
+
+  .align-items-lg-center {
+    -webkit-box-align: center !important;
+        -ms-flex-align: center !important;
+            align-items: center !important;
+  }
+
+  .align-items-lg-baseline {
+    -webkit-box-align: baseline !important;
+        -ms-flex-align: baseline !important;
+            align-items: baseline !important;
+  }
+
+  .align-items-lg-stretch {
+    -webkit-box-align: stretch !important;
+        -ms-flex-align: stretch !important;
+            align-items: stretch !important;
+  }
+
+  .align-content-lg-start {
+    -ms-flex-line-pack: start !important;
+        align-content: flex-start !important;
+  }
+
+  .align-content-lg-end {
+    -ms-flex-line-pack: end !important;
+        align-content: flex-end !important;
+  }
+
+  .align-content-lg-center {
+    -ms-flex-line-pack: center !important;
+        align-content: center !important;
+  }
+
+  .align-content-lg-between {
+    -ms-flex-line-pack: justify !important;
+        align-content: space-between !important;
+  }
+
+  .align-content-lg-around {
+    -ms-flex-line-pack: distribute !important;
+        align-content: space-around !important;
+  }
+
+  .align-content-lg-stretch {
+    -ms-flex-line-pack: stretch !important;
+        align-content: stretch !important;
+  }
+
+  .align-self-lg-auto {
+    -ms-flex-item-align: auto !important;
+        align-self: auto !important;
+  }
+
+  .align-self-lg-start {
+    -ms-flex-item-align: start !important;
+        align-self: flex-start !important;
+  }
+
+  .align-self-lg-end {
+    -ms-flex-item-align: end !important;
+        align-self: flex-end !important;
+  }
+
+  .align-self-lg-center {
+    -ms-flex-item-align: center !important;
+        align-self: center !important;
+  }
+
+  .align-self-lg-baseline {
+    -ms-flex-item-align: baseline !important;
+        align-self: baseline !important;
+  }
+
+  .align-self-lg-stretch {
+    -ms-flex-item-align: stretch !important;
+        align-self: stretch !important;
+  }
+}
+@media (min-width: 1200px) {
+  .flex-xl-row {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: row !important;
+            flex-direction: row !important;
+  }
+
+  .flex-xl-column {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: normal !important;
+        -ms-flex-direction: column !important;
+            flex-direction: column !important;
+  }
+
+  .flex-xl-row-reverse {
+    -webkit-box-orient: horizontal !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: row-reverse !important;
+            flex-direction: row-reverse !important;
+  }
+
+  .flex-xl-column-reverse {
+    -webkit-box-orient: vertical !important;
+    -webkit-box-direction: reverse !important;
+        -ms-flex-direction: column-reverse !important;
+            flex-direction: column-reverse !important;
+  }
+
+  .flex-xl-wrap {
+    -ms-flex-wrap: wrap !important;
+        flex-wrap: wrap !important;
+  }
+
+  .flex-xl-nowrap {
+    -ms-flex-wrap: nowrap !important;
+        flex-wrap: nowrap !important;
+  }
+
+  .flex-xl-wrap-reverse {
+    -ms-flex-wrap: wrap-reverse !important;
+        flex-wrap: wrap-reverse !important;
+  }
+
+  .justify-content-xl-start {
+    -webkit-box-pack: start !important;
+        -ms-flex-pack: start !important;
+            justify-content: flex-start !important;
+  }
+
+  .justify-content-xl-end {
+    -webkit-box-pack: end !important;
+        -ms-flex-pack: end !important;
+            justify-content: flex-end !important;
+  }
+
+  .justify-content-xl-center {
+    -webkit-box-pack: center !important;
+        -ms-flex-pack: center !important;
+            justify-content: center !important;
+  }
+
+  .justify-content-xl-between {
+    -webkit-box-pack: justify !important;
+        -ms-flex-pack: justify !important;
+            justify-content: space-between !important;
+  }
+
+  .justify-content-xl-around {
+    -ms-flex-pack: distribute !important;
+        justify-content: space-around !important;
+  }
+
+  .align-items-xl-start {
+    -webkit-box-align: start !important;
+        -ms-flex-align: start !important;
+            align-items: flex-start !important;
+  }
+
+  .align-items-xl-end {
+    -webkit-box-align: end !important;
+        -ms-flex-align: end !important;
+            align-items: flex-end !important;
+  }
+
+  .align-items-xl-center {
+    -webkit-box-align: center !important;
+        -ms-flex-align: center !important;
+            align-items: center !important;
+  }
+
+  .align-items-xl-baseline {
+    -webkit-box-align: baseline !important;
+        -ms-flex-align: baseline !important;
+            align-items: baseline !important;
+  }
+
+  .align-items-xl-stretch {
+    -webkit-box-align: stretch !important;
+        -ms-flex-align: stretch !important;
+            align-items: stretch !important;
+  }
+
+  .align-content-xl-start {
+    -ms-flex-line-pack: start !important;
+        align-content: flex-start !important;
+  }
+
+  .align-content-xl-end {
+    -ms-flex-line-pack: end !important;
+        align-content: flex-end !important;
+  }
+
+  .align-content-xl-center {
+    -ms-flex-line-pack: center !important;
+        align-content: center !important;
+  }
+
+  .align-content-xl-between {
+    -ms-flex-line-pack: justify !important;
+        align-content: space-between !important;
+  }
+
+  .align-content-xl-around {
+    -ms-flex-line-pack: distribute !important;
+        align-content: space-around !important;
+  }
+
+  .align-content-xl-stretch {
+    -ms-flex-line-pack: stretch !important;
+        align-content: stretch !important;
+  }
+
+  .align-self-xl-auto {
+    -ms-flex-item-align: auto !important;
+        align-self: auto !important;
+  }
+
+  .align-self-xl-start {
+    -ms-flex-item-align: start !important;
+        align-self: flex-start !important;
+  }
+
+  .align-self-xl-end {
+    -ms-flex-item-align: end !important;
+        align-self: flex-end !important;
+  }
+
+  .align-self-xl-center {
+    -ms-flex-item-align: center !important;
+        align-self: center !important;
+  }
+
+  .align-self-xl-baseline {
+    -ms-flex-item-align: baseline !important;
+        align-self: baseline !important;
+  }
+
+  .align-self-xl-stretch {
+    -ms-flex-item-align: stretch !important;
+        align-self: stretch !important;
+  }
+}
+.float-left {
+  float: left !important;
+}
+
+.float-right {
+  float: right !important;
+}
+
+.float-none {
+  float: none !important;
+}
+
+@media (min-width: 576px) {
+  .float-sm-left {
+    float: left !important;
+  }
+
+  .float-sm-right {
+    float: right !important;
+  }
+
+  .float-sm-none {
+    float: none !important;
+  }
+}
+@media (min-width: 768px) {
+  .float-md-left {
+    float: left !important;
+  }
+
+  .float-md-right {
+    float: right !important;
+  }
+
+  .float-md-none {
+    float: none !important;
+  }
+}
+@media (min-width: 992px) {
+  .float-lg-left {
+    float: left !important;
+  }
+
+  .float-lg-right {
+    float: right !important;
+  }
+
+  .float-lg-none {
+    float: none !important;
+  }
+}
+@media (min-width: 1200px) {
+  .float-xl-left {
+    float: left !important;
+  }
+
+  .float-xl-right {
+    float: right !important;
+  }
+
+  .float-xl-none {
+    float: none !important;
+  }
+}
+.position-static {
+  position: static !important;
+}
+
+.position-relative {
+  position: relative !important;
+}
+
+.position-absolute {
+  position: absolute !important;
+}
+
+.position-fixed {
+  position: fixed !important;
+}
+
+.position-sticky {
+  position: sticky !important;
+}
+
+.fixed-top {
+  position: fixed;
+  top: 0;
+  right: 0;
+  left: 0;
+  z-index: 1030;
+}
+
+.fixed-bottom {
+  position: fixed;
+  right: 0;
+  bottom: 0;
+  left: 0;
+  z-index: 1030;
+}
+
+@supports (position: sticky) {
+  .sticky-top {
+    position: sticky;
+    top: 0;
+    z-index: 1020;
+  }
+}
+
+.sr-only {
+  position: absolute;
+  width: 1px;
+  height: 1px;
+  padding: 0;
+  overflow: hidden;
+  clip: rect(0, 0, 0, 0);
+  white-space: nowrap;
+  -webkit-clip-path: inset(50%);
+          clip-path: inset(50%);
+  border: 0;
+}
+
+.sr-only-focusable:active, .sr-only-focusable:focus {
+  position: static;
+  width: auto;
+  height: auto;
+  overflow: visible;
+  clip: auto;
+  white-space: normal;
+  -webkit-clip-path: none;
+          clip-path: none;
+}
+
+.w-25 {
+  width: 25% !important;
+}
+
+.w-50 {
+  width: 50% !important;
+}
+
+.w-75 {
+  width: 75% !important;
+}
+
+.w-100 {
+  width: 100% !important;
+}
+
+.h-25 {
+  height: 25% !important;
+}
+
+.h-50 {
+  height: 50% !important;
+}
+
+.h-75 {
+  height: 75% !important;
+}
+
+.h-100 {
+  height: 100% !important;
+}
+
+.mw-100 {
+  max-width: 100% !important;
+}
+
+.mh-100 {
+  max-height: 100% !important;
+}
+
+.m-0 {
+  margin: 0 !important;
+}
+
+.mt-0,
+.my-0 {
+  margin-top: 0 !important;
+}
+
+.mr-0,
+.mx-0 {
+  margin-right: 0 !important;
+}
+
+.mb-0,
+.my-0 {
+  margin-bottom: 0 !important;
+}
+
+.ml-0,
+.mx-0 {
+  margin-left: 0 !important;
+}
+
+.m-1 {
+  margin: 0.25rem !important;
+}
+
+.mt-1,
+.my-1 {
+  margin-top: 0.25rem !important;
+}
+
+.mr-1,
+.mx-1 {
+  margin-right: 0.25rem !important;
+}
+
+.mb-1,
+.my-1 {
+  margin-bottom: 0.25rem !important;
+}
+
+.ml-1,
+.mx-1 {
+  margin-left: 0.25rem !important;
+}
+
+.m-2 {
+  margin: 0.5rem !important;
+}
+
+.mt-2,
+.my-2 {
+  margin-top: 0.5rem !important;
+}
+
+.mr-2,
+.mx-2 {
+  margin-right: 0.5rem !important;
+}
+
+.mb-2,
+.my-2 {
+  margin-bottom: 0.5rem !important;
+}
+
+.ml-2,
+.mx-2 {
+  margin-left: 0.5rem !important;
+}
+
+.m-3 {
+  margin: 1rem !important;
+}
+
+.mt-3,
+.my-3 {
+  margin-top: 1rem !important;
+}
+
+.mr-3,
+.mx-3 {
+  margin-right: 1rem !important;
+}
+
+.mb-3,
+.my-3 {
+  margin-bottom: 1rem !important;
+}
+
+.ml-3,
+.mx-3 {
+  margin-left: 1rem !important;
+}
+
+.m-4 {
+  margin: 1.5rem !important;
+}
+
+.mt-4,
+.my-4 {
+  margin-top: 1.5rem !important;
+}
+
+.mr-4,
+.mx-4 {
+  margin-right: 1.5rem !important;
+}
+
+.mb-4,
+.my-4 {
+  margin-bottom: 1.5rem !important;
+}
+
+.ml-4,
+.mx-4 {
+  margin-left: 1.5rem !important;
+}
+
+.m-5 {
+  margin: 3rem !important;
+}
+
+.mt-5,
+.my-5 {
+  margin-top: 3rem !important;
+}
+
+.mr-5,
+.mx-5 {
+  margin-right: 3rem !important;
+}
+
+.mb-5,
+.my-5 {
+  margin-bottom: 3rem !important;
+}
+
+.ml-5,
+.mx-5 {
+  margin-left: 3rem !important;
+}
+
+.p-0 {
+  padding: 0 !important;
+}
+
+.pt-0,
+.py-0 {
+  padding-top: 0 !important;
+}
+
+.pr-0,
+.px-0 {
+  padding-right: 0 !important;
+}
+
+.pb-0,
+.py-0 {
+  padding-bottom: 0 !important;
+}
+
+.pl-0,
+.px-0 {
+  padding-left: 0 !important;
+}
+
+.p-1 {
+  padding: 0.25rem !important;
+}
+
+.pt-1,
+.py-1 {
+  padding-top: 0.25rem !important;
+}
+
+.pr-1,
+.px-1 {
+  padding-right: 0.25rem !important;
+}
+
+.pb-1,
+.py-1 {
+  padding-bottom: 0.25rem !important;
+}
+
+.pl-1,
+.px-1 {
+  padding-left: 0.25rem !important;
+}
+
+.p-2 {
+  padding: 0.5rem !important;
+}
+
+.pt-2,
+.py-2 {
+  padding-top: 0.5rem !important;
+}
+
+.pr-2,
+.px-2 {
+  padding-right: 0.5rem !important;
+}
+
+.pb-2,
+.py-2 {
+  padding-bottom: 0.5rem !important;
+}
+
+.pl-2,
+.px-2 {
+  padding-left: 0.5rem !important;
+}
+
+.p-3 {
+  padding: 1rem !important;
+}
+
+.pt-3,
+.py-3 {
+  padding-top: 1rem !important;
+}
+
+.pr-3,
+.px-3 {
+  padding-right: 1rem !important;
+}
+
+.pb-3,
+.py-3 {
+  padding-bottom: 1rem !important;
+}
+
+.pl-3,
+.px-3 {
+  padding-left: 1rem !important;
+}
+
+.p-4 {
+  padding: 1.5rem !important;
+}
+
+.pt-4,
+.py-4 {
+  padding-top: 1.5rem !important;
+}
+
+.pr-4,
+.px-4 {
+  padding-right: 1.5rem !important;
+}
+
+.pb-4,
+.py-4 {
+  padding-bottom: 1.5rem !important;
+}
+
+.pl-4,
+.px-4 {
+  padding-left: 1.5rem !important;
+}
+
+.p-5 {
+  padding: 3rem !important;
+}
+
+.pt-5,
+.py-5 {
+  padding-top: 3rem !important;
+}
+
+.pr-5,
+.px-5 {
+  padding-right: 3rem !important;
+}
+
+.pb-5,
+.py-5 {
+  padding-bottom: 3rem !important;
+}
+
+.pl-5,
+.px-5 {
+  padding-left: 3rem !important;
+}
+
+.m-auto {
+  margin: auto !important;
+}
+
+.mt-auto,
+.my-auto {
+  margin-top: auto !important;
+}
+
+.mr-auto,
+.mx-auto {
+  margin-right: auto !important;
+}
+
+.mb-auto,
+.my-auto {
+  margin-bottom: auto !important;
+}
+
+.ml-auto,
+.mx-auto {
+  margin-left: auto !important;
+}
+
+@media (min-width: 576px) {
+  .m-sm-0 {
+    margin: 0 !important;
+  }
+
+  .mt-sm-0,
+  .my-sm-0 {
+    margin-top: 0 !important;
+  }
+
+  .mr-sm-0,
+  .mx-sm-0 {
+    margin-right: 0 !important;
+  }
+
+  .mb-sm-0,
+  .my-sm-0 {
+    margin-bottom: 0 !important;
+  }
+
+  .ml-sm-0,
+  .mx-sm-0 {
+    margin-left: 0 !important;
+  }
+
+  .m-sm-1 {
+    margin: 0.25rem !important;
+  }
+
+  .mt-sm-1,
+  .my-sm-1 {
+    margin-top: 0.25rem !important;
+  }
+
+  .mr-sm-1,
+  .mx-sm-1 {
+    margin-right: 0.25rem !important;
+  }
+
+  .mb-sm-1,
+  .my-sm-1 {
+    margin-bottom: 0.25rem !important;
+  }
+
+  .ml-sm-1,
+  .mx-sm-1 {
+    margin-left: 0.25rem !important;
+  }
+
+  .m-sm-2 {
+    margin: 0.5rem !important;
+  }
+
+  .mt-sm-2,
+  .my-sm-2 {
+    margin-top: 0.5rem !important;
+  }
+
+  .mr-sm-2,
+  .mx-sm-2 {
+    margin-right: 0.5rem !important;
+  }
+
+  .mb-sm-2,
+  .my-sm-2 {
+    margin-bottom: 0.5rem !important;
+  }
+
+  .ml-sm-2,
+  .mx-sm-2 {
+    margin-left: 0.5rem !important;
+  }
+
+  .m-sm-3 {
+    margin: 1rem !important;
+  }
+
+  .mt-sm-3,
+  .my-sm-3 {
+    margin-top: 1rem !important;
+  }
+
+  .mr-sm-3,
+  .mx-sm-3 {
+    margin-right: 1rem !important;
+  }
+
+  .mb-sm-3,
+  .my-sm-3 {
+    margin-bottom: 1rem !important;
+  }
+
+  .ml-sm-3,
+  .mx-sm-3 {
+    margin-left: 1rem !important;
+  }
+
+  .m-sm-4 {
+    margin: 1.5rem !important;
+  }
+
+  .mt-sm-4,
+  .my-sm-4 {
+    margin-top: 1.5rem !important;
+  }
+
+  .mr-sm-4,
+  .mx-sm-4 {
+    margin-right: 1.5rem !important;
+  }
+
+  .mb-sm-4,
+  .my-sm-4 {
+    margin-bottom: 1.5rem !important;
+  }
+
+  .ml-sm-4,
+  .mx-sm-4 {
+    margin-left: 1.5rem !important;
+  }
+
+  .m-sm-5 {
+    margin: 3rem !important;
+  }
+
+  .mt-sm-5,
+  .my-sm-5 {
+    margin-top: 3rem !important;
+  }
+
+  .mr-sm-5,
+  .mx-sm-5 {
+    margin-right: 3rem !important;
+  }
+
+  .mb-sm-5,
+  .my-sm-5 {
+    margin-bottom: 3rem !important;
+  }
+
+  .ml-sm-5,
+  .mx-sm-5 {
+    margin-left: 3rem !important;
+  }
+
+  .p-sm-0 {
+    padding: 0 !important;
+  }
+
+  .pt-sm-0,
+  .py-sm-0 {
+    padding-top: 0 !important;
+  }
+
+  .pr-sm-0,
+  .px-sm-0 {
+    padding-right: 0 !important;
+  }
+
+  .pb-sm-0,
+  .py-sm-0 {
+    padding-bottom: 0 !important;
+  }
+
+  .pl-sm-0,
+  .px-sm-0 {
+    padding-left: 0 !important;
+  }
+
+  .p-sm-1 {
+    padding: 0.25rem !important;
+  }
+
+  .pt-sm-1,
+  .py-sm-1 {
+    padding-top: 0.25rem !important;
+  }
+
+  .pr-sm-1,
+  .px-sm-1 {
+    padding-right: 0.25rem !important;
+  }
+
+  .pb-sm-1,
+  .py-sm-1 {
+    padding-bottom: 0.25rem !important;
+  }
+
+  .pl-sm-1,
+  .px-sm-1 {
+    padding-left: 0.25rem !important;
+  }
+
+  .p-sm-2 {
+    padding: 0.5rem !important;
+  }
+
+  .pt-sm-2,
+  .py-sm-2 {
+    padding-top: 0.5rem !important;
+  }
+
+  .pr-sm-2,
+  .px-sm-2 {
+    padding-right: 0.5rem !important;
+  }
+
+  .pb-sm-2,
+  .py-sm-2 {
+    padding-bottom: 0.5rem !important;
+  }
+
+  .pl-sm-2,
+  .px-sm-2 {
+    padding-left: 0.5rem !important;
+  }
+
+  .p-sm-3 {
+    padding: 1rem !important;
+  }
+
+  .pt-sm-3,
+  .py-sm-3 {
+    padding-top: 1rem !important;
+  }
+
+  .pr-sm-3,
+  .px-sm-3 {
+    padding-right: 1rem !important;
+  }
+
+  .pb-sm-3,
+  .py-sm-3 {
+    padding-bottom: 1rem !important;
+  }
+
+  .pl-sm-3,
+  .px-sm-3 {
+    padding-left: 1rem !important;
+  }
+
+  .p-sm-4 {
+    padding: 1.5rem !important;
+  }
+
+  .pt-sm-4,
+  .py-sm-4 {
+    padding-top: 1.5rem !important;
+  }
+
+  .pr-sm-4,
+  .px-sm-4 {
+    padding-right: 1.5rem !important;
+  }
+
+  .pb-sm-4,
+  .py-sm-4 {
+    padding-bottom: 1.5rem !important;
+  }
+
+  .pl-sm-4,
+  .px-sm-4 {
+    padding-left: 1.5rem !important;
+  }
+
+  .p-sm-5 {
+    padding: 3rem !important;
+  }
+
+  .pt-sm-5,
+  .py-sm-5 {
+    padding-top: 3rem !important;
+  }
+
+  .pr-sm-5,
+  .px-sm-5 {
+    padding-right: 3rem !important;
+  }
+
+  .pb-sm-5,
+  .py-sm-5 {
+    padding-bottom: 3rem !important;
+  }
+
+  .pl-sm-5,
+  .px-sm-5 {
+    padding-left: 3rem !important;
+  }
+
+  .m-sm-auto {
+    margin: auto !important;
+  }
+
+  .mt-sm-auto,
+  .my-sm-auto {
+    margin-top: auto !important;
+  }
+
+  .mr-sm-auto,
+  .mx-sm-auto {
+    margin-right: auto !important;
+  }
+
+  .mb-sm-auto,
+  .my-sm-auto {
+    margin-bottom: auto !important;
+  }
+
+  .ml-sm-auto,
+  .mx-sm-auto {
+    margin-left: auto !important;
+  }
+}
+@media (min-width: 768px) {
+  .m-md-0 {
+    margin: 0 !important;
+  }
+
+  .mt-md-0,
+  .my-md-0 {
+    margin-top: 0 !important;
+  }
+
+  .mr-md-0,
+  .mx-md-0 {
+    margin-right: 0 !important;
+  }
+
+  .mb-md-0,
+  .my-md-0 {
+    margin-bottom: 0 !important;
+  }
+
+  .ml-md-0,
+  .mx-md-0 {
+    margin-left: 0 !important;
+  }
+
+  .m-md-1 {
+    margin: 0.25rem !important;
+  }
+
+  .mt-md-1,
+  .my-md-1 {
+    margin-top: 0.25rem !important;
+  }
+
+  .mr-md-1,
+  .mx-md-1 {
+    margin-right: 0.25rem !important;
+  }
+
+  .mb-md-1,
+  .my-md-1 {
+    margin-bottom: 0.25rem !important;
+  }
+
+  .ml-md-1,
+  .mx-md-1 {
+    margin-left: 0.25rem !important;
+  }
+
+  .m-md-2 {
+    margin: 0.5rem !important;
+  }
+
+  .mt-md-2,
+  .my-md-2 {
+    margin-top: 0.5rem !important;
+  }
+
+  .mr-md-2,
+  .mx-md-2 {
+    margin-right: 0.5rem !important;
+  }
+
+  .mb-md-2,
+  .my-md-2 {
+    margin-bottom: 0.5rem !important;
+  }
+
+  .ml-md-2,
+  .mx-md-2 {
+    margin-left: 0.5rem !important;
+  }
+
+  .m-md-3 {
+    margin: 1rem !important;
+  }
+
+  .mt-md-3,
+  .my-md-3 {
+    margin-top: 1rem !important;
+  }
+
+  .mr-md-3,
+  .mx-md-3 {
+    margin-right: 1rem !important;
+  }
+
+  .mb-md-3,
+  .my-md-3 {
+    margin-bottom: 1rem !important;
+  }
+
+  .ml-md-3,
+  .mx-md-3 {
+    margin-left: 1rem !important;
+  }
+
+  .m-md-4 {
+    margin: 1.5rem !important;
+  }
+
+  .mt-md-4,
+  .my-md-4 {
+    margin-top: 1.5rem !important;
+  }
+
+  .mr-md-4,
+  .mx-md-4 {
+    margin-right: 1.5rem !important;
+  }
+
+  .mb-md-4,
+  .my-md-4 {
+    margin-bottom: 1.5rem !important;
+  }
+
+  .ml-md-4,
+  .mx-md-4 {
+    margin-left: 1.5rem !important;
+  }
+
+  .m-md-5 {
+    margin: 3rem !important;
+  }
+
+  .mt-md-5,
+  .my-md-5 {
+    margin-top: 3rem !important;
+  }
+
+  .mr-md-5,
+  .mx-md-5 {
+    margin-right: 3rem !important;
+  }
+
+  .mb-md-5,
+  .my-md-5 {
+    margin-bottom: 3rem !important;
+  }
+
+  .ml-md-5,
+  .mx-md-5 {
+    margin-left: 3rem !important;
+  }
+
+  .p-md-0 {
+    padding: 0 !important;
+  }
+
+  .pt-md-0,
+  .py-md-0 {
+    padding-top: 0 !important;
+  }
+
+  .pr-md-0,
+  .px-md-0 {
+    padding-right: 0 !important;
+  }
+
+  .pb-md-0,
+  .py-md-0 {
+    padding-bottom: 0 !important;
+  }
+
+  .pl-md-0,
+  .px-md-0 {
+    padding-left: 0 !important;
+  }
+
+  .p-md-1 {
+    padding: 0.25rem !important;
+  }
+
+  .pt-md-1,
+  .py-md-1 {
+    padding-top: 0.25rem !important;
+  }
+
+  .pr-md-1,
+  .px-md-1 {
+    padding-right: 0.25rem !important;
+  }
+
+  .pb-md-1,
+  .py-md-1 {
+    padding-bottom: 0.25rem !important;
+  }
+
+  .pl-md-1,
+  .px-md-1 {
+    padding-left: 0.25rem !important;
+  }
+
+  .p-md-2 {
+    padding: 0.5rem !important;
+  }
+
+  .pt-md-2,
+  .py-md-2 {
+    padding-top: 0.5rem !important;
+  }
+
+  .pr-md-2,
+  .px-md-2 {
+    padding-right: 0.5rem !important;
+  }
+
+  .pb-md-2,
+  .py-md-2 {
+    padding-bottom: 0.5rem !important;
+  }
+
+  .pl-md-2,
+  .px-md-2 {
+    padding-left: 0.5rem !important;
+  }
+
+  .p-md-3 {
+    padding: 1rem !important;
+  }
+
+  .pt-md-3,
+  .py-md-3 {
+    padding-top: 1rem !important;
+  }
+
+  .pr-md-3,
+  .px-md-3 {
+    padding-right: 1rem !important;
+  }
+
+  .pb-md-3,
+  .py-md-3 {
+    padding-bottom: 1rem !important;
+  }
+
+  .pl-md-3,
+  .px-md-3 {
+    padding-left: 1rem !important;
+  }
+
+  .p-md-4 {
+    padding: 1.5rem !important;
+  }
+
+  .pt-md-4,
+  .py-md-4 {
+    padding-top: 1.5rem !important;
+  }
+
+  .pr-md-4,
+  .px-md-4 {
+    padding-right: 1.5rem !important;
+  }
+
+  .pb-md-4,
+  .py-md-4 {
+    padding-bottom: 1.5rem !important;
+  }
+
+  .pl-md-4,
+  .px-md-4 {
+    padding-left: 1.5rem !important;
+  }
+
+  .p-md-5 {
+    padding: 3rem !important;
+  }
+
+  .pt-md-5,
+  .py-md-5 {
+    padding-top: 3rem !important;
+  }
+
+  .pr-md-5,
+  .px-md-5 {
+    padding-right: 3rem !important;
+  }
+
+  .pb-md-5,
+  .py-md-5 {
+    padding-bottom: 3rem !important;
+  }
+
+  .pl-md-5,
+  .px-md-5 {
+    padding-left: 3rem !important;
+  }
+
+  .m-md-auto {
+    margin: auto !important;
+  }
+
+  .mt-md-auto,
+  .my-md-auto {
+    margin-top: auto !important;
+  }
+
+  .mr-md-auto,
+  .mx-md-auto {
+    margin-right: auto !important;
+  }
+
+  .mb-md-auto,
+  .my-md-auto {
+    margin-bottom: auto !important;
+  }
+
+  .ml-md-auto,
+  .mx-md-auto {
+    margin-left: auto !important;
+  }
+}
+@media (min-width: 992px) {
+  .m-lg-0 {
+    margin: 0 !important;
+  }
+
+  .mt-lg-0,
+  .my-lg-0 {
+    margin-top: 0 !important;
+  }
+
+  .mr-lg-0,
+  .mx-lg-0 {
+    margin-right: 0 !important;
+  }
+
+  .mb-lg-0,
+  .my-lg-0 {
+    margin-bottom: 0 !important;
+  }
+
+  .ml-lg-0,
+  .mx-lg-0 {
+    margin-left: 0 !important;
+  }
+
+  .m-lg-1 {
+    margin: 0.25rem !important;
+  }
+
+  .mt-lg-1,
+  .my-lg-1 {
+    margin-top: 0.25rem !important;
+  }
+
+  .mr-lg-1,
+  .mx-lg-1 {
+    margin-right: 0.25rem !important;
+  }
+
+  .mb-lg-1,
+  .my-lg-1 {
+    margin-bottom: 0.25rem !important;
+  }
+
+  .ml-lg-1,
+  .mx-lg-1 {
+    margin-left: 0.25rem !important;
+  }
+
+  .m-lg-2 {
+    margin: 0.5rem !important;
+  }
+
+  .mt-lg-2,
+  .my-lg-2 {
+    margin-top: 0.5rem !important;
+  }
+
+  .mr-lg-2,
+  .mx-lg-2 {
+    margin-right: 0.5rem !important;
+  }
+
+  .mb-lg-2,
+  .my-lg-2 {
+    margin-bottom: 0.5rem !important;
+  }
+
+  .ml-lg-2,
+  .mx-lg-2 {
+    margin-left: 0.5rem !important;
+  }
+
+  .m-lg-3 {
+    margin: 1rem !important;
+  }
+
+  .mt-lg-3,
+  .my-lg-3 {
+    margin-top: 1rem !important;
+  }
+
+  .mr-lg-3,
+  .mx-lg-3 {
+    margin-right: 1rem !important;
+  }
+
+  .mb-lg-3,
+  .my-lg-3 {
+    margin-bottom: 1rem !important;
+  }
+
+  .ml-lg-3,
+  .mx-lg-3 {
+    margin-left: 1rem !important;
+  }
+
+  .m-lg-4 {
+    margin: 1.5rem !important;
+  }
+
+  .mt-lg-4,
+  .my-lg-4 {
+    margin-top: 1.5rem !important;
+  }
+
+  .mr-lg-4,
+  .mx-lg-4 {
+    margin-right: 1.5rem !important;
+  }
+
+  .mb-lg-4,
+  .my-lg-4 {
+    margin-bottom: 1.5rem !important;
+  }
+
+  .ml-lg-4,
+  .mx-lg-4 {
+    margin-left: 1.5rem !important;
+  }
+
+  .m-lg-5 {
+    margin: 3rem !important;
+  }
+
+  .mt-lg-5,
+  .my-lg-5 {
+    margin-top: 3rem !important;
+  }
+
+  .mr-lg-5,
+  .mx-lg-5 {
+    margin-right: 3rem !important;
+  }
+
+  .mb-lg-5,
+  .my-lg-5 {
+    margin-bottom: 3rem !important;
+  }
+
+  .ml-lg-5,
+  .mx-lg-5 {
+    margin-left: 3rem !important;
+  }
+
+  .p-lg-0 {
+    padding: 0 !important;
+  }
+
+  .pt-lg-0,
+  .py-lg-0 {
+    padding-top: 0 !important;
+  }
+
+  .pr-lg-0,
+  .px-lg-0 {
+    padding-right: 0 !important;
+  }
+
+  .pb-lg-0,
+  .py-lg-0 {
+    padding-bottom: 0 !important;
+  }
+
+  .pl-lg-0,
+  .px-lg-0 {
+    padding-left: 0 !important;
+  }
+
+  .p-lg-1 {
+    padding: 0.25rem !important;
+  }
+
+  .pt-lg-1,
+  .py-lg-1 {
+    padding-top: 0.25rem !important;
+  }
+
+  .pr-lg-1,
+  .px-lg-1 {
+    padding-right: 0.25rem !important;
+  }
+
+  .pb-lg-1,
+  .py-lg-1 {
+    padding-bottom: 0.25rem !important;
+  }
+
+  .pl-lg-1,
+  .px-lg-1 {
+    padding-left: 0.25rem !important;
+  }
+
+  .p-lg-2 {
+    padding: 0.5rem !important;
+  }
+
+  .pt-lg-2,
+  .py-lg-2 {
+    padding-top: 0.5rem !important;
+  }
+
+  .pr-lg-2,
+  .px-lg-2 {
+    padding-right: 0.5rem !important;
+  }
+
+  .pb-lg-2,
+  .py-lg-2 {
+    padding-bottom: 0.5rem !important;
+  }
+
+  .pl-lg-2,
+  .px-lg-2 {
+    padding-left: 0.5rem !important;
+  }
+
+  .p-lg-3 {
+    padding: 1rem !important;
+  }
+
+  .pt-lg-3,
+  .py-lg-3 {
+    padding-top: 1rem !important;
+  }
+
+  .pr-lg-3,
+  .px-lg-3 {
+    padding-right: 1rem !important;
+  }
+
+  .pb-lg-3,
+  .py-lg-3 {
+    padding-bottom: 1rem !important;
+  }
+
+  .pl-lg-3,
+  .px-lg-3 {
+    padding-left: 1rem !important;
+  }
+
+  .p-lg-4 {
+    padding: 1.5rem !important;
+  }
+
+  .pt-lg-4,
+  .py-lg-4 {
+    padding-top: 1.5rem !important;
+  }
+
+  .pr-lg-4,
+  .px-lg-4 {
+    padding-right: 1.5rem !important;
+  }
+
+  .pb-lg-4,
+  .py-lg-4 {
+    padding-bottom: 1.5rem !important;
+  }
+
+  .pl-lg-4,
+  .px-lg-4 {
+    padding-left: 1.5rem !important;
+  }
+
+  .p-lg-5 {
+    padding: 3rem !important;
+  }
+
+  .pt-lg-5,
+  .py-lg-5 {
+    padding-top: 3rem !important;
+  }
+
+  .pr-lg-5,
+  .px-lg-5 {
+    padding-right: 3rem !important;
+  }
+
+  .pb-lg-5,
+  .py-lg-5 {
+    padding-bottom: 3rem !important;
+  }
+
+  .pl-lg-5,
+  .px-lg-5 {
+    padding-left: 3rem !important;
+  }
+
+  .m-lg-auto {
+    margin: auto !important;
+  }
+
+  .mt-lg-auto,
+  .my-lg-auto {
+    margin-top: auto !important;
+  }
+
+  .mr-lg-auto,
+  .mx-lg-auto {
+    margin-right: auto !important;
+  }
+
+  .mb-lg-auto,
+  .my-lg-auto {
+    margin-bottom: auto !important;
+  }
+
+  .ml-lg-auto,
+  .mx-lg-auto {
+    margin-left: auto !important;
+  }
+}
+@media (min-width: 1200px) {
+  .m-xl-0 {
+    margin: 0 !important;
+  }
+
+  .mt-xl-0,
+  .my-xl-0 {
+    margin-top: 0 !important;
+  }
+
+  .mr-xl-0,
+  .mx-xl-0 {
+    margin-right: 0 !important;
+  }
+
+  .mb-xl-0,
+  .my-xl-0 {
+    margin-bottom: 0 !important;
+  }
+
+  .ml-xl-0,
+  .mx-xl-0 {
+    margin-left: 0 !important;
+  }
+
+  .m-xl-1 {
+    margin: 0.25rem !important;
+  }
+
+  .mt-xl-1,
+  .my-xl-1 {
+    margin-top: 0.25rem !important;
+  }
+
+  .mr-xl-1,
+  .mx-xl-1 {
+    margin-right: 0.25rem !important;
+  }
+
+  .mb-xl-1,
+  .my-xl-1 {
+    margin-bottom: 0.25rem !important;
+  }
+
+  .ml-xl-1,
+  .mx-xl-1 {
+    margin-left: 0.25rem !important;
+  }
+
+  .m-xl-2 {
+    margin: 0.5rem !important;
+  }
+
+  .mt-xl-2,
+  .my-xl-2 {
+    margin-top: 0.5rem !important;
+  }
+
+  .mr-xl-2,
+  .mx-xl-2 {
+    margin-right: 0.5rem !important;
+  }
+
+  .mb-xl-2,
+  .my-xl-2 {
+    margin-bottom: 0.5rem !important;
+  }
+
+  .ml-xl-2,
+  .mx-xl-2 {
+    margin-left: 0.5rem !important;
+  }
+
+  .m-xl-3 {
+    margin: 1rem !important;
+  }
+
+  .mt-xl-3,
+  .my-xl-3 {
+    margin-top: 1rem !important;
+  }
+
+  .mr-xl-3,
+  .mx-xl-3 {
+    margin-right: 1rem !important;
+  }
+
+  .mb-xl-3,
+  .my-xl-3 {
+    margin-bottom: 1rem !important;
+  }
+
+  .ml-xl-3,
+  .mx-xl-3 {
+    margin-left: 1rem !important;
+  }
+
+  .m-xl-4 {
+    margin: 1.5rem !important;
+  }
+
+  .mt-xl-4,
+  .my-xl-4 {
+    margin-top: 1.5rem !important;
+  }
+
+  .mr-xl-4,
+  .mx-xl-4 {
+    margin-right: 1.5rem !important;
+  }
+
+  .mb-xl-4,
+  .my-xl-4 {
+    margin-bottom: 1.5rem !important;
+  }
+
+  .ml-xl-4,
+  .mx-xl-4 {
+    margin-left: 1.5rem !important;
+  }
+
+  .m-xl-5 {
+    margin: 3rem !important;
+  }
+
+  .mt-xl-5,
+  .my-xl-5 {
+    margin-top: 3rem !important;
+  }
+
+  .mr-xl-5,
+  .mx-xl-5 {
+    margin-right: 3rem !important;
+  }
+
+  .mb-xl-5,
+  .my-xl-5 {
+    margin-bottom: 3rem !important;
+  }
+
+  .ml-xl-5,
+  .mx-xl-5 {
+    margin-left: 3rem !important;
+  }
+
+  .p-xl-0 {
+    padding: 0 !important;
+  }
+
+  .pt-xl-0,
+  .py-xl-0 {
+    padding-top: 0 !important;
+  }
+
+  .pr-xl-0,
+  .px-xl-0 {
+    padding-right: 0 !important;
+  }
+
+  .pb-xl-0,
+  .py-xl-0 {
+    padding-bottom: 0 !important;
+  }
+
+  .pl-xl-0,
+  .px-xl-0 {
+    padding-left: 0 !important;
+  }
+
+  .p-xl-1 {
+    padding: 0.25rem !important;
+  }
+
+  .pt-xl-1,
+  .py-xl-1 {
+    padding-top: 0.25rem !important;
+  }
+
+  .pr-xl-1,
+  .px-xl-1 {
+    padding-right: 0.25rem !important;
+  }
+
+  .pb-xl-1,
+  .py-xl-1 {
+    padding-bottom: 0.25rem !important;
+  }
+
+  .pl-xl-1,
+  .px-xl-1 {
+    padding-left: 0.25rem !important;
+  }
+
+  .p-xl-2 {
+    padding: 0.5rem !important;
+  }
+
+  .pt-xl-2,
+  .py-xl-2 {
+    padding-top: 0.5rem !important;
+  }
+
+  .pr-xl-2,
+  .px-xl-2 {
+    padding-right: 0.5rem !important;
+  }
+
+  .pb-xl-2,
+  .py-xl-2 {
+    padding-bottom: 0.5rem !important;
+  }
+
+  .pl-xl-2,
+  .px-xl-2 {
+    padding-left: 0.5rem !important;
+  }
+
+  .p-xl-3 {
+    padding: 1rem !important;
+  }
+
+  .pt-xl-3,
+  .py-xl-3 {
+    padding-top: 1rem !important;
+  }
+
+  .pr-xl-3,
+  .px-xl-3 {
+    padding-right: 1rem !important;
+  }
+
+  .pb-xl-3,
+  .py-xl-3 {
+    padding-bottom: 1rem !important;
+  }
+
+  .pl-xl-3,
+  .px-xl-3 {
+    padding-left: 1rem !important;
+  }
+
+  .p-xl-4 {
+    padding: 1.5rem !important;
+  }
+
+  .pt-xl-4,
+  .py-xl-4 {
+    padding-top: 1.5rem !important;
+  }
+
+  .pr-xl-4,
+  .px-xl-4 {
+    padding-right: 1.5rem !important;
+  }
+
+  .pb-xl-4,
+  .py-xl-4 {
+    padding-bottom: 1.5rem !important;
+  }
+
+  .pl-xl-4,
+  .px-xl-4 {
+    padding-left: 1.5rem !important;
+  }
+
+  .p-xl-5 {
+    padding: 3rem !important;
+  }
+
+  .pt-xl-5,
+  .py-xl-5 {
+    padding-top: 3rem !important;
+  }
+
+  .pr-xl-5,
+  .px-xl-5 {
+    padding-right: 3rem !important;
+  }
+
+  .pb-xl-5,
+  .py-xl-5 {
+    padding-bottom: 3rem !important;
+  }
+
+  .pl-xl-5,
+  .px-xl-5 {
+    padding-left: 3rem !important;
+  }
+
+  .m-xl-auto {
+    margin: auto !important;
+  }
+
+  .mt-xl-auto,
+  .my-xl-auto {
+    margin-top: auto !important;
+  }
+
+  .mr-xl-auto,
+  .mx-xl-auto {
+    margin-right: auto !important;
+  }
+
+  .mb-xl-auto,
+  .my-xl-auto {
+    margin-bottom: auto !important;
+  }
+
+  .ml-xl-auto,
+  .mx-xl-auto {
+    margin-left: auto !important;
+  }
+}
+.text-justify {
+  text-align: justify !important;
+}
+
+.text-nowrap {
+  white-space: nowrap !important;
+}
+
+.text-truncate {
+  overflow: hidden;
+  text-overflow: ellipsis;
+  white-space: nowrap;
+}
+
+.text-left {
+  text-align: left !important;
+}
+
+.text-right {
+  text-align: right !important;
+}
+
+.text-center {
+  text-align: center !important;
+}
+
+@media (min-width: 576px) {
+  .text-sm-left {
+    text-align: left !important;
+  }
+
+  .text-sm-right {
+    text-align: right !important;
+  }
+
+  .text-sm-center {
+    text-align: center !important;
+  }
+}
+@media (min-width: 768px) {
+  .text-md-left {
+    text-align: left !important;
+  }
+
+  .text-md-right {
+    text-align: right !important;
+  }
+
+  .text-md-center {
+    text-align: center !important;
+  }
+}
+@media (min-width: 992px) {
+  .text-lg-left {
+    text-align: left !important;
+  }
+
+  .text-lg-right {
+    text-align: right !important;
+  }
+
+  .text-lg-center {
+    text-align: center !important;
+  }
+}
+@media (min-width: 1200px) {
+  .text-xl-left {
+    text-align: left !important;
+  }
+
+  .text-xl-right {
+    text-align: right !important;
+  }
+
+  .text-xl-center {
+    text-align: center !important;
+  }
+}
+.text-lowercase {
+  text-transform: lowercase !important;
+}
+
+.text-uppercase {
+  text-transform: uppercase !important;
+}
+
+.text-capitalize {
+  text-transform: capitalize !important;
+}
+
+.font-weight-light {
+  font-weight: 300 !important;
+}
+
+.font-weight-normal {
+  font-weight: 400 !important;
+}
+
+.font-weight-bold {
+  font-weight: 700 !important;
+}
+
+.font-italic {
+  font-style: italic !important;
+}
+
+.text-white {
+  color: #fff !important;
+}
+
+.text-primary {
+  color: #007bff !important;
+}
+
+a.text-primary:hover, a.text-primary:focus {
+  color: #0062cc !important;
+}
+
+.text-secondary {
+  color: #6c757d !important;
+}
+
+a.text-secondary:hover, a.text-secondary:focus {
+  color: #545b62 !important;
+}
+
+.text-success {
+  color: #28a745 !important;
+}
+
+a.text-success:hover, a.text-success:focus {
+  color: #1e7e34 !important;
+}
+
+.text-info {
+  color: #17a2b8 !important;
+}
+
+a.text-info:hover, a.text-info:focus {
+  color: #117a8b !important;
+}
+
+.text-warning {
+  color: #ffc107 !important;
+}
+
+a.text-warning:hover, a.text-warning:focus {
+  color: #d39e00 !important;
+}
+
+.text-danger {
+  color: #dc3545 !important;
+}
+
+a.text-danger:hover, a.text-danger:focus {
+  color: #bd2130 !important;
+}
+
+.text-light {
+  color: #f8f9fa !important;
+}
+
+a.text-light:hover, a.text-light:focus {
+  color: #dae0e5 !important;
+}
+
+.text-dark {
+  color: #343a40 !important;
+}
+
+a.text-dark:hover, a.text-dark:focus {
+  color: #1d2124 !important;
+}
+
+.text-muted {
+  color: #6c757d !important;
+}
+
+.text-hide {
+  font: 0/0 a;
+  color: transparent;
+  text-shadow: none;
+  background-color: transparent;
+  border: 0;
+}
+
+.visible {
+  visibility: visible !important;
+}
+
+.invisible {
+  visibility: hidden !important;
+}
+
+@media print {
+  *,
+  *::before,
+  *::after {
+    text-shadow: none !important;
+    -webkit-box-shadow: none !important;
+            box-shadow: none !important;
+  }
+
+  a:not(.btn) {
+    text-decoration: underline;
+  }
+
+  abbr[title]::after {
+    content: " (" attr(title) ")";
+  }
+
+  pre {
+    white-space: pre-wrap !important;
+  }
+
+  pre,
+  blockquote {
+    border: 1px solid #999;
+    page-break-inside: avoid;
+  }
+
+  thead {
+    display: table-header-group;
+  }
+
+  tr,
+  img {
+    page-break-inside: avoid;
+  }
+
+  p,
+  h2,
+  h3 {
+    orphans: 3;
+    widows: 3;
+  }
+
+  h2,
+  h3 {
+    page-break-after: avoid;
+  }
+
+  @page {
+    size: a3;
+  }
+  body {
+    min-width: 992px !important;
+  }
+
+  .container {
+    min-width: 992px !important;
+  }
+
+  .navbar {
+    display: none;
+  }
+
+  .badge {
+    border: 1px solid #000;
+  }
+
+  .table {
+    border-collapse: collapse !important;
+  }
+  .table td,
+  .table th {
+    background-color: #fff !important;
+  }
+
+  .table-bordered th,
+  .table-bordered td {
+    border: 1px solid #ddd !important;
+  }
+}
+/*Github syntax highlighting theme via Rouge*/
+.highlight table td {
+  padding: 5px;
+}
+
+.highlight table pre {
+  margin: 0;
+}
+
+.highlight .cm {
+  color: #999988;
+  font-style: italic;
+}
+
+.highlight .cp {
+  color: #999999;
+  font-weight: bold;
+}
+
+.highlight .c1 {
+  color: #999988;
+  font-style: italic;
+}
+
+.highlight .cs {
+  color: #999999;
+  font-weight: bold;
+  font-style: italic;
+}
+
+.highlight .c, .highlight .cd {
+  color: #999988;
+  font-style: italic;
+}
+
+.highlight .err {
+  color: #a61717;
+  background-color: #e3d2d2;
+}
+
+.highlight .gd {
+  color: #000000;
+  background-color: #ffdddd;
+}
+
+.highlight .ge {
+  color: #000000;
+  font-style: italic;
+}
+
+.highlight .gr {
+  color: #aa0000;
+}
+
+.highlight .gh {
+  color: #999999;
+}
+
+.highlight .gi {
+  color: #000000;
+  background-color: #ddffdd;
+}
+
+.highlight .go {
+  color: #888888;
+}
+
+.highlight .gp {
+  color: #555555;
+}
+
+.highlight .gs {
+  font-weight: bold;
+}
+
+.highlight .gu {
+  color: #aaaaaa;
+}
+
+.highlight .gt {
+  color: #aa0000;
+}
+
+.highlight .kc {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .kd {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .kn {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .kp {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .kr {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .kt {
+  color: #445588;
+  font-weight: bold;
+}
+
+.highlight .k, .highlight .kv {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .mf {
+  color: #009999;
+}
+
+.highlight .mh {
+  color: #009999;
+}
+
+.highlight .il {
+  color: #009999;
+}
+
+.highlight .mi {
+  color: #009999;
+}
+
+.highlight .mo {
+  color: #009999;
+}
+
+.highlight .m, .highlight .mb, .highlight .mx {
+  color: #009999;
+}
+
+.highlight .sb {
+  color: #d14;
+}
+
+.highlight .sc {
+  color: #d14;
+}
+
+.highlight .sd {
+  color: #d14;
+}
+
+.highlight .s2 {
+  color: #d14;
+}
+
+.highlight .se {
+  color: #d14;
+}
+
+.highlight .sh {
+  color: #d14;
+}
+
+.highlight .si {
+  color: #d14;
+}
+
+.highlight .sx {
+  color: #d14;
+}
+
+.highlight .sr {
+  color: #009926;
+}
+
+.highlight .s1 {
+  color: #d14;
+}
+
+.highlight .ss {
+  color: #990073;
+}
+
+.highlight .s {
+  color: #d14;
+}
+
+.highlight .na {
+  color: #008080;
+}
+
+.highlight .bp {
+  color: #525252;
+}
+
+.highlight .nb {
+  color: #0086B3;
+}
+
+.highlight .nc {
+  color: #445588;
+  font-weight: bold;
+}
+
+.highlight .no {
+  color: #008080;
+}
+
+.highlight .nd {
+  color: #3c5d5d;
+  font-weight: bold;
+}
+
+.highlight .ni {
+  color: #800080;
+}
+
+.highlight .ne {
+  color: #990000;
+  font-weight: bold;
+}
+
+.highlight .nf {
+  color: #990000;
+  font-weight: bold;
+}
+
+.highlight .nl {
+  color: #990000;
+  font-weight: bold;
+}
+
+.highlight .nn {
+  color: #555555;
+}
+
+.highlight .nt {
+  color: #000080;
+}
+
+.highlight .vc {
+  color: #008080;
+}
+
+.highlight .vg {
+  color: #008080;
+}
+
+.highlight .vi {
+  color: #008080;
+}
+
+.highlight .nv {
+  color: #008080;
+}
+
+.highlight .ow {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .o {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .n {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .p {
+  color: #000000;
+  font-weight: bold;
+}
+
+.highlight .w {
+  color: #bbbbbb;
+}
+
+.highlight {
+  background-color: #f8f8f8;
+}
+
+@font-face {
+  font-family: FreightSans;
+  font-weight: 700;
+  font-style: normal;
+  src: url("../fonts/FreightSans/freight-sans-bold.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-bold.woff") format("woff");
+}
+@font-face {
+  font-family: FreightSans;
+  font-weight: 700;
+  font-style: italic;
+  src: url("../fonts/FreightSans/freight-sans-bold-italic.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-bold-italic.woff") format("woff");
+}
+@font-face {
+  font-family: FreightSans;
+  font-weight: 500;
+  font-style: normal;
+  src: url("../fonts/FreightSans/freight-sans-medium.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-medium.woff") format("woff");
+}
+@font-face {
+  font-family: FreightSans;
+  font-weight: 500;
+  font-style: italic;
+  src: url("../fonts/FreightSans/freight-sans-medium-italic.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-medium-italic.woff") format("woff");
+}
+@font-face {
+  font-family: FreightSans;
+  font-weight: 100;
+  font-style: normal;
+  src: url("../fonts/FreightSans/freight-sans-light.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-light.woff") format("woff");
+}
+@font-face {
+  font-family: FreightSans;
+  font-weight: 100;
+  font-style: italic;
+  src: url("../fonts/FreightSans/freight-sans-light-italic.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-light-italic.woff") format("woff");
+}
+@font-face {
+  font-family: FreightSans;
+  font-weight: 400;
+  font-style: italic;
+  src: url("../fonts/FreightSans/freight-sans-book-italic.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-book-italic.woff") format("woff");
+}
+@font-face {
+  font-family: FreightSans;
+  font-weight: 400;
+  font-style: normal;
+  src: url("../fonts/FreightSans/freight-sans-book.woff2") format("woff2"), url("../fonts/FreightSans/freight-sans-book.woff") format("woff");
+}
+@font-face {
+  font-family: IBMPlexMono;
+  font-weight: 600;
+  font-style: normal;
+  unicode-range: u+0020-007f;
+  src: local("IBMPlexMono-SemiBold"), url("../fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2") format("woff2"), url("../fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff") format("woff");
+}
+@font-face {
+  font-family: IBMPlexMono;
+  font-weight: 500;
+  font-style: normal;
+  unicode-range: u+0020-007f;
+  src: local("IBMPlexMono-Medium"), url("../fonts/IBMPlexMono/IBMPlexMono-Medium.woff2") format("woff2"), url("../fonts/IBMPlexMono/IBMPlexMono-Medium.woff") format("woff");
+}
+@font-face {
+  font-family: IBMPlexMono;
+  font-weight: 400;
+  font-style: normal;
+  unicode-range: u+0020-007f;
+  src: local("IBMPlexMono-Regular"), url("../fonts/IBMPlexMono/IBMPlexMono-Regular.woff2") format("woff2"), url("../fonts/IBMPlexMono/IBMPlexMono-Regular.woff") format("woff");
+}
+@font-face {
+  font-family: IBMPlexMono;
+  font-weight: 300;
+  font-style: normal;
+  unicode-range: u+0020-007f;
+  src: local("IBMPlexMono-Light"), url("../fonts/IBMPlexMono/IBMPlexMono-Light.woff2") format("woff2"), url("../fonts/IBMPlexMono/IBMPlexMono-Light.woff") format("woff");
+}
+html {
+  position: relative;
+  min-height: 100%;
+  font-size: 12px;
+}
+@media screen and (min-width: 768px) {
+  html {
+    font-size: 16px;
+  }
+}
+
+* {
+  -webkit-box-sizing: border-box;
+          box-sizing: border-box;
+}
+
+body {
+  font-family: FreightSans, Helvetica Neue, Helvetica, Arial, sans-serif;
+}
+
+a:link,
+a:visited,
+a:hover {
+  text-decoration: none;
+  color: #e44c2c;
+}
+
+a.with-right-arrow, .btn.with-right-arrow {
+  padding-right: 1.375rem;
+  position: relative;
+  background-image: url("../images/chevron-right-orange.svg");
+  background-size: 6px 13px;
+  background-position: center right 5px;
+  background-repeat: no-repeat;
+}
+@media screen and (min-width: 768px) {
+  a.with-right-arrow, .btn.with-right-arrow {
+    background-size: 8px 14px;
+    background-position: center right 12px;
+    padding-right: 2rem;
+  }
+}
+
+::-webkit-input-placeholder {
+  color: #e44c2c;
+}
+
+::-moz-placeholder {
+  color: #e44c2c;
+}
+
+:-ms-input-placeholder {
+  color: #e44c2c;
+}
+
+:-moz-placeholder {
+  color: #e44c2c;
+}
+
+.email-subscribe-form input.email {
+  color: #e44c2c;
+  border: none;
+  border-bottom: 1px solid #939393;
+  width: 100%;
+  background-color: transparent;
+  outline: none;
+  font-size: 1.125rem;
+  letter-spacing: 0.25px;
+  line-height: 2.25rem;
+}
+.email-subscribe-form input[type="submit"] {
+  position: absolute;
+  right: 0;
+  top: 10px;
+  height: 15px;
+  width: 15px;
+  background-image: url("../images/arrow-right-with-tail.svg");
+  background-color: transparent;
+  background-repeat: no-repeat;
+  background-size: 15px 15px;
+  background-position: center center;
+  -webkit-appearance: none;
+  -moz-appearance: none;
+  appearance: none;
+  border: 0;
+}
+
+.email-subscribe-form-fields-wrapper {
+  position: relative;
+}
+
+.anchorjs-link {
+  color: #6c6c6d !important;
+}
+@media screen and (min-width: 768px) {
+  .anchorjs-link:hover {
+    color: inherit;
+    text-decoration: none !important;
+  }
+}
+
+.pytorch-article #table-of-contents {
+  display: none;
+}
+
+code, kbd, pre, samp {
+  font-family: IBMPlexMono,SFMono-Regular,Menlo,Monaco,Consolas,"Liberation Mono","Courier New",monospace;
+}
+code span, kbd span, pre span, samp span {
+  font-family: IBMPlexMono,SFMono-Regular,Menlo,Monaco,Consolas,"Liberation Mono","Courier New",monospace;
+}
+
+pre {
+  padding: 1.125rem;
+}
+pre code {
+  font-size: 0.875rem;
+}
+pre.highlight {
+  background-color: #f3f4f7;
+  line-height: 1.3125rem;
+}
+
+code.highlighter-rouge {
+  color: #6c6c6d;
+  background-color: #f3f4f7;
+  padding: 2px 6px;
+}
+
+a:link code.highlighter-rouge,
+a:visited code.highlighter-rouge,
+a:hover code.highlighter-rouge {
+  color: #4974D1;
+}
+a:link.has-code,
+a:visited.has-code,
+a:hover.has-code {
+  color: #4974D1;
+}
+
+p code,
+h1 code,
+h2 code,
+h3 code,
+h4 code,
+h5 code,
+h6 code {
+  font-size: 78.5%;
+}
+
+pre {
+  white-space: pre-wrap;
+  white-space: -moz-pre-wrap;
+  white-space: -pre-wrap;
+  white-space: -o-pre-wrap;
+  word-wrap: break-word;
+}
+
+.header-holder {
+  height: 68px;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  left: 0;
+  margin-left: auto;
+  margin-right: auto;
+  position: fixed;
+  right: 0;
+  top: 0;
+  width: 100%;
+  z-index: 9999;
+  background-color: #ffffff;
+  border-bottom: 1px solid #e2e2e2;
+}
+@media screen and (min-width: 1100px) {
+  .header-holder {
+    height: 90px;
+  }
+}
+
+.header-container {
+  position: relative;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+}
+.header-container:before, .header-container:after {
+  content: "";
+  display: table;
+}
+.header-container:after {
+  clear: both;
+}
+.header-container {
+  *zoom: 1;
+}
+@media screen and (min-width: 1100px) {
+  .header-container {
+    display: block;
+  }
+}
+
+.header-logo {
+  height: 23px;
+  width: 93px;
+  background-image: url("../images/logo.svg");
+  background-repeat: no-repeat;
+  background-size: 93px 23px;
+  display: block;
+  float: left;
+  z-index: 10;
+}
+@media screen and (min-width: 1100px) {
+  .header-logo {
+    background-size: 108px 27px;
+    position: absolute;
+    height: 27px;
+    width: 108px;
+    top: 4px;
+    float: none;
+  }
+}
+
+.main-menu-open-button {
+  background-image: url("../images/icon-menu-dots.svg");
+  background-position: center center;
+  background-size: 25px 7px;
+  background-repeat: no-repeat;
+  width: 25px;
+  height: 17px;
+  position: absolute;
+  right: 0;
+  top: 4px;
+}
+@media screen and (min-width: 1100px) {
+  .main-menu-open-button {
+    display: none;
+  }
+}
+
+.header-holder .main-menu {
+  display: none;
+}
+@media screen and (min-width: 1100px) {
+  .header-holder .main-menu {
+    display: -webkit-box;
+    display: -ms-flexbox;
+    display: flex;
+    -webkit-box-align: center;
+        -ms-flex-align: center;
+            align-items: center;
+    -webkit-box-pack: end;
+        -ms-flex-pack: end;
+            justify-content: flex-end;
+  }
+}
+.header-holder .main-menu ul {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  margin: 0;
+}
+.header-holder .main-menu ul li {
+  display: inline-block;
+  margin-right: 40px;
+  position: relative;
+}
+.header-holder .main-menu ul li.active:after {
+  content: "•";
+  bottom: -24px;
+  color: #e44c2c;
+  font-size: 1.375rem;
+  left: 0;
+  position: absolute;
+  right: 0;
+  text-align: center;
+}
+.header-holder .main-menu ul li.active a {
+  color: #e44c2c;
+}
+.header-holder .main-menu ul li.docs-active:after {
+  content: "•";
+  bottom: -24px;
+  color: #e44c2c;
+  font-size: 1.375rem;
+  left: -24px;
+  position: absolute;
+  right: 0;
+  text-align: center;
+}
+.header-holder .main-menu ul li:last-of-type {
+  margin-right: 0;
+}
+.header-holder .main-menu ul li a {
+  color: #ffffff;
+  font-size: 1.3rem;
+  letter-spacing: 0;
+  line-height: 2.125rem;
+  text-align: center;
+  text-decoration: none;
+}
+@media screen and (min-width: 1100px) {
+  .header-holder .main-menu ul li a:hover {
+    color: #e44c2c;
+  }
+}
+
+.mobile-main-menu {
+  display: none;
+}
+.mobile-main-menu.open {
+  background-color: #262626;
+  display: block;
+  height: 100%;
+  left: 0;
+  margin-left: auto;
+  margin-right: auto;
+  min-height: 100%;
+  position: fixed;
+  right: 0;
+  top: 0;
+  width: 100%;
+  z-index: 99999;
+}
+
+.mobile-main-menu .container-fluid {
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  height: 68px;
+  position: relative;
+}
+.mobile-main-menu .container-fluid:before, .mobile-main-menu .container-fluid:after {
+  content: "";
+  display: table;
+}
+.mobile-main-menu .container-fluid:after {
+  clear: both;
+}
+.mobile-main-menu .container-fluid {
+  *zoom: 1;
+}
+
+.mobile-main-menu.open ul {
+  list-style-type: none;
+  padding: 0;
+}
+.mobile-main-menu.open ul li a, .mobile-main-menu.open .resources-mobile-menu-title {
+  font-size: 2rem;
+  color: #ffffff;
+  letter-spacing: 0;
+  line-height: 4rem;
+  text-decoration: none;
+}
+.mobile-main-menu.open ul li.active a {
+  color: #e44c2c;
+}
+
+.main-menu-close-button {
+  background-image: url("../images/icon-close.svg");
+  background-position: center center;
+  background-repeat: no-repeat;
+  background-size: 24px 24px;
+  height: 24px;
+  position: absolute;
+  right: 0;
+  width: 24px;
+  top: -4px;
+}
+
+.mobile-main-menu-header-container {
+  position: relative;
+}
+
+.mobile-main-menu-links-container {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  padding-left: 2.8125rem;
+  height: 90vh;
+  margin-top: -25px;
+  padding-top: 50%;
+  overflow-y: scroll;
+}
+.mobile-main-menu-links-container .main-menu {
+  height: 100vh;
+}
+
+.mobile-main-menu-links-container ul.resources-mobile-menu-items li {
+  padding-left: 15px;
+}
+
+.site-footer {
+  padding: 2.5rem 0;
+  width: 100%;
+  background: #000000;
+  background-size: 100%;
+  margin-left: 0;
+  margin-right: 0;
+  position: relative;
+  z-index: 201;
+}
+@media screen and (min-width: 768px) {
+  .site-footer {
+    padding: 5rem 0;
+  }
+}
+
+.site-footer p {
+  color: #ffffff;
+}
+.site-footer ul {
+  list-style-type: none;
+  padding-left: 0;
+  margin-bottom: 0;
+}
+
+.site-footer ul li {
+  font-size: 1.125rem;
+  line-height: 2rem;
+  color: #A0A0A1;
+  padding-bottom: 0.375rem;
+}
+.site-footer ul li.list-title {
+  padding-bottom: 0.75rem;
+  color: #ffffff;
+}
+.site-footer a:link,
+.site-footer a:visited {
+  color: inherit;
+}
+@media screen and (min-width: 768px) {
+  .site-footer a:hover {
+    color: #e44c2c;
+  }
+}
+
+.site-footer .privacy-policy {
+    background: #000000;
+    display: flex;
+    border-bottom: 1px solid white;
+    padding-bottom: 10px;
+}
+
+.site-footer .privacy-policy-links {
+    background: #000000;
+    display: flex; 
+    padding-top: 1rem;
+    padding-right: 1rem;
+    display: inline-flex;
+    color: white;
+}
+
+.site-footer .footer-links-wrapper {
+    display: flex;
+    flex-wrap: wrap;
+    border-bottom: 1px solid white;
+    padding-bottom: 1rem;
+}
+
+.site-footer .copyright {
+    padding-top: 1rem;
+    padding-right: 1rem;
+    display: inline-flex;
+    color: white;
+}
+
+.site-footer .copyright p {
+    color: white;
+}
+
+.site-footer .copyright a {
+    color: red;
+}
+
+.docs-tutorials-resources {
+  background-color: #262626;
+  color: #ffffff;
+  padding-top: 2.5rem;
+  padding-bottom: 2.5rem;
+  position: relative;
+  z-index: 201;
+}
+@media screen and (min-width: 768px) {
+  .docs-tutorials-resources {
+    padding-top: 5rem;
+    padding-bottom: 5rem;
+  }
+}
+.docs-tutorials-resources p {
+  color: #929292;
+  font-size: 1.125rem;
+}
+.docs-tutorials-resources h2 {
+  font-size: 1.5rem;
+  letter-spacing: -0.25px;
+  text-transform: none;
+  margin-bottom: 0.25rem;
+}
+@media screen and (min-width: 768px) {
+  .docs-tutorials-resources h2 {
+    margin-bottom: 1.25rem;
+  }
+}
+.docs-tutorials-resources .col-md-4 {
+  margin-bottom: 2rem;
+  text-align: center;
+}
+@media screen and (min-width: 768px) {
+  .docs-tutorials-resources .col-md-4 {
+    margin-bottom: 0;
+  }
+}
+.docs-tutorials-resources .with-right-arrow {
+  margin-left: 12px;
+}
+.docs-tutorials-resources .with-right-arrow:hover {
+  background-image: url("../images/chevron-right-white.svg");
+}
+.docs-tutorials-resources p {
+  font-size: 1rem;
+  line-height: 1.5rem;
+  letter-spacing: 0.22px;
+  color: #939393;
+  margin-bottom: 0;
+}
+@media screen and (min-width: 768px) {
+  .docs-tutorials-resources p {
+    margin-bottom: 1.25rem;
+  }
+}
+.docs-tutorials-resources a {
+  font-size: 1.125rem;
+  color: #e44c2c;
+}
+.docs-tutorials-resources a:hover {
+  color: #ffffff;
+}
+
+.footer-container {
+  position: relative;
+}
+
+@media screen and (min-width: 768px) {
+  .footer-logo-wrapper {
+    position: absolute;
+    top: 0;
+    left: 30px;
+  }
+}
+
+.footer-logo {
+  background-image: url("../images/logo-icon.svg");
+  background-position: center;
+  background-repeat: no-repeat;
+  background-size: 20px 24px;
+  display: block;
+  height: 24px;
+  margin-bottom: 2.8125rem;
+  width: 20px;
+}
+@media screen and (min-width: 768px) {
+  .footer-logo {
+    background-size: 29px 36px;
+    height: 36px;
+    margin-bottom: 0;
+    margin-bottom: 0;
+    width: 29px;
+  }
+}
+
+.footer-links-wrapper {
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -ms-flex-wrap: wrap;
+      flex-wrap: wrap;
+}
+@media screen and (min-width: 768px) {
+  .footer-links-wrapper {
+    -ms-flex-wrap: initial;
+        flex-wrap: initial;
+    -webkit-box-pack: end;
+        -ms-flex-pack: end;
+            justify-content: flex-end;
+  }
+}
+
+.footer-links-col {
+  margin-bottom: 3.75rem;
+  width: 50%;
+}
+@media screen and (min-width: 768px) {
+  .footer-links-col {
+    margin-bottom: 0;
+    width: 14%;
+    margin-right: 23px;
+  }
+  .footer-links-col.follow-us-col {
+    width: 18%;
+    margin-right: 0;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  .footer-links-col {
+    width: 18%;
+    margin-right: 30px;
+  }
+}
+
+.footer-social-icons {
+  margin: 8.5625rem 0 2.5rem 0;
+}
+.footer-social-icons a {
+  height: 32px;
+  width: 32px;
+  display: inline-block;
+  background-color: #CCCDD1;
+  border-radius: 50%;
+  margin-right: 5px;
+}
+.footer-social-icons a.facebook {
+  background-image: url("../images/logo-facebook-dark.svg");
+  background-position: center center;
+  background-size: 9px 18px;
+  background-repeat: no-repeat;
+}
+.footer-social-icons a.twitter {
+  background-image: url("../images/logo-twitter-dark.svg");
+  background-position: center center;
+  background-size: 17px 17px;
+  background-repeat: no-repeat;
+}
+.footer-social-icons a.youtube {
+  background-image: url("../images/logo-youtube-dark.svg");
+  background-position: center center;
+  background-repeat: no-repeat;
+}
+
+.site-footer .mc-field-group {
+  margin-top: -2px;
+}
+
+article.pytorch-article {
+  max-width: 920px;
+  margin: 0 auto;
+}
+article.pytorch-article h2,
+article.pytorch-article h3,
+article.pytorch-article h4,
+article.pytorch-article h5,
+article.pytorch-article h6 {
+  margin: 1.375rem 0;
+  color: #262626;
+}
+article.pytorch-article h2 {
+  font-size: 1.625rem;
+  letter-spacing: 1.33px;
+  line-height: 2rem;
+  text-transform: none;
+}
+article.pytorch-article h3 {
+  font-size: 1.5rem;
+  letter-spacing: -0.25px;
+  line-height: 1.875rem;
+  text-transform: none;
+}
+article.pytorch-article h4,
+article.pytorch-article h5,
+article.pytorch-article h6 {
+  font-size: 1.125rem;
+  letter-spacing: -0.19px;
+  line-height: 1.875rem;
+}
+article.pytorch-article p {
+  margin-bottom: 1.125rem;
+}
+article.pytorch-article p,
+article.pytorch-article ul li,
+article.pytorch-article ol li,
+article.pytorch-article dl dt,
+article.pytorch-article dl dd,
+article.pytorch-article blockquote {
+  font-size: 1rem;
+  line-height: 1.375rem;
+  color: #262626;
+  letter-spacing: 0.01px;
+  font-weight: 500;
+}
+article.pytorch-article table {
+  margin-bottom: 2.5rem;
+  width: 100%;
+}
+article.pytorch-article table thead {
+  border-bottom: 1px solid #cacaca;
+}
+article.pytorch-article table th {
+  padding: 0.625rem;
+  color: #262626;
+}
+article.pytorch-article table td {
+  padding: 0.3125rem;
+}
+article.pytorch-article table tr th:first-of-type,
+article.pytorch-article table tr td:first-of-type {
+  padding-left: 0;
+}
+article.pytorch-article table.docutils.field-list th.field-name {
+  padding: 0.3125rem;
+  padding-left: 0;
+}
+article.pytorch-article table.docutils.field-list td.field-body {
+  padding: 0.3125rem;
+}
+article.pytorch-article table.docutils.field-list td.field-body p:last-of-type {
+  margin-bottom: 0;
+}
+article.pytorch-article ul,
+article.pytorch-article ol {
+  margin: 1.5rem 0 3.125rem 0;
+}
+@media screen and (min-width: 768px) {
+  article.pytorch-article ul,
+  article.pytorch-article ol {
+    padding-left: 6.25rem;
+  }
+}
+article.pytorch-article ul li,
+article.pytorch-article ol li {
+  margin-bottom: 0.625rem;
+}
+article.pytorch-article dl {
+  margin-bottom: 1.5rem;
+}
+article.pytorch-article dl dt {
+  margin-bottom: 0.75rem;
+}
+article.pytorch-article pre {
+  margin-bottom: 2.5rem;
+}
+article.pytorch-article hr {
+  margin-top: 4.6875rem;
+  margin-bottom: 4.6875rem;
+}
+article.pytorch-article blockquote {
+  margin: 0 auto;
+  margin-bottom: 2.5rem;
+  width: 65%;
+}
+article.pytorch-article img {
+  width: 100%;
+}
+
+html {
+  height: 100%;
+}
+@media screen and (min-width: 768px) {
+  html {
+    font-size: 16px;
+  }
+}
+
+body {
+  background: #ffffff;
+  height: 100%;
+  margin: 0;
+}
+body.no-scroll {
+  height: 100%;
+  overflow: hidden;
+}
+
+p {
+  margin-top: 0;
+  margin-bottom: 1.125rem;
+}
+p a:link,
+p a:visited,
+p a:hover {
+  color: #e44c2c;
+  text-decoration: none;
+}
+@media screen and (min-width: 768px) {
+  p a:hover {
+    text-decoration: underline;
+  }
+}
+p a:link,
+p a:visited,
+p a:hover {
+  color: #ee4c2c;
+}
+
+.wy-breadcrumbs li a {
+  color: #ee4c2c;
+}
+
+ul.pytorch-breadcrumbs {
+  padding-left: 0;
+  list-style-type: none;
+}
+ul.pytorch-breadcrumbs li {
+  display: inline-block;
+  font-size: 0.875rem;
+}
+ul.pytorch-breadcrumbs a {
+  color: #ee4c2c;
+  text-decoration: none;
+}
+
+.table-of-contents-link-wrapper {
+  display: block;
+  margin-top: 0;
+  padding: 1.25rem 1.875rem;
+  background-color: #f3f4f7;
+  position: relative;
+  color: #262626;
+  font-size: 1.25rem;
+}
+.table-of-contents-link-wrapper.is-open .toggle-table-of-contents {
+  -webkit-transform: rotate(180deg);
+          transform: rotate(180deg);
+}
+@media screen and (min-width: 1100px) {
+  .table-of-contents-link-wrapper {
+    display: none;
+  }
+}
+
+.toggle-table-of-contents {
+  background-image: url("../images/chevron-down-grey.svg");
+  background-position: center center;
+  background-repeat: no-repeat;
+  background-size: 18px 18px;
+  height: 100%;
+  position: absolute;
+  right: 21px;
+  width: 30px;
+  top: 0;
+}
+
+.tutorials-header .header-logo {
+  background-image: url("../images/logo-dark.svg");
+}
+.tutorials-header .main-menu ul li a {
+  color: #262626;
+}
+.tutorials-header .main-menu-open-button {
+  background-image: url("../images/icon-menu-dots-dark.svg");
+}
+
+.rst-content footer .rating-hr.hr-top {
+  margin-bottom: -0.0625rem;
+}
+.rst-content footer .rating-hr.hr-bottom {
+  margin-top: -0.0625rem;
+}
+.rst-content footer .rating-container {
+  display: -webkit-inline-box;
+  display: -ms-inline-flexbox;
+  display: inline-flex;
+  font-size: 1.125rem;
+}
+.rst-content footer .rating-container .rating-prompt, .rst-content footer .rating-container .was-helpful-thank-you {
+  padding: 0.625rem 1.25rem 0.625rem 1.25rem;
+}
+.rst-content footer .rating-container .was-helpful-thank-you {
+  display: none;
+}
+.rst-content footer .rating-container .rating-prompt.yes-link, .rst-content footer .rating-container .rating-prompt.no-link {
+  color: #e44c2c;
+  cursor: pointer;
+}
+.rst-content footer .rating-container .rating-prompt.yes-link:hover, .rst-content footer .rating-container .rating-prompt.no-link:hover {
+  background-color: #e44c2c;
+  color: #ffffff;
+}
+.rst-content footer .rating-container .stars-outer {
+  display: inline-block;
+  position: relative;
+  font-family: FontAwesome;
+  padding: 0.625rem 1.25rem 0.625rem 1.25rem;
+}
+.rst-content footer .rating-container .stars-outer i {
+  cursor: pointer;
+}
+.rst-content footer .rating-container .stars-outer .star-fill {
+  color: #ee4c2c;
+}
+.rst-content footer div[role="contentinfo"] {
+  padding-top: 2.5rem;
+}
+.rst-content footer div[role="contentinfo"] p {
+  margin-bottom: 0;
+}
+
+h1 {
+  font-size: 2rem;
+  letter-spacing: 1.78px;
+  line-height: 2.5rem;
+  text-transform: uppercase;
+  margin: 1.375rem 0;
+}
+
+span.pre {
+  color: #6c6c6d;
+  background-color: #f3f4f7;
+  padding: 2px 0px;
+}
+
+pre {
+  padding: 1.375rem;
+}
+
+.highlight .c1 {
+  color: #6c6c6d;
+}
+
+.headerlink {
+  display: none !important;
+}
+
+a:link.has-code,
+a:hover.has-code,
+a:visited.has-code {
+  color: #4974D1;
+}
+a:link.has-code span,
+a:hover.has-code span,
+a:visited.has-code span {
+  color: #4974D1;
+}
+
+article.pytorch-article ul,
+article.pytorch-article ol {
+  padding-left: 1.875rem;
+  margin: 0;
+}
+article.pytorch-article ul li,
+article.pytorch-article ol li {
+  margin: 0;
+  line-height: 1.75rem;
+}
+article.pytorch-article ul p,
+article.pytorch-article ol p {
+  line-height: 1.75rem;
+  margin-bottom: 0;
+}
+article.pytorch-article ul ul,
+article.pytorch-article ul ol,
+article.pytorch-article ol ul,
+article.pytorch-article ol ol {
+  margin: 0;
+}
+article.pytorch-article h1,
+article.pytorch-article h2,
+article.pytorch-article h3,
+article.pytorch-article h4,
+article.pytorch-article h5,
+article.pytorch-article h6 {
+  font-weight: normal;
+}
+article.pytorch-article h1 a,
+article.pytorch-article h2 a,
+article.pytorch-article h3 a,
+article.pytorch-article h4 a,
+article.pytorch-article h5 a,
+article.pytorch-article h6 a {
+  color: #262626;
+}
+article.pytorch-article p.caption {
+  margin-top: 1.25rem;
+}
+
+article.pytorch-article .section:first-of-type h1:first-of-type {
+  margin-top: 0;
+}
+
+article.pytorch-article .sphx-glr-thumbcontainer {
+  margin: 0;
+  border: 1px solid #d6d7d8;
+  border-radius: 0;
+  width: 45%;
+  text-align: center;
+  margin-bottom: 5%;
+}
+@media screen and (max-width: 1100px) {
+  article.pytorch-article .sphx-glr-thumbcontainer:nth-child(odd) {
+    margin-left: 0;
+    margin-right: 2.5%;
+  }
+  article.pytorch-article .sphx-glr-thumbcontainer:nth-child(even) {
+    margin-right: 0;
+    margin-left: 2.5%;
+  }
+  article.pytorch-article .sphx-glr-thumbcontainer .figure {
+    width: 40%;
+  }
+}
+@media screen and (min-width: 1101px) {
+  article.pytorch-article .sphx-glr-thumbcontainer {
+    margin-right: 3%;
+    margin-bottom: 3%;
+    width: 30%;
+  }
+}
+article.pytorch-article .sphx-glr-thumbcontainer .caption-text a {
+  font-size: 1rem;
+  color: #262626;
+  letter-spacing: 0;
+  line-height: 1.5rem;
+  text-decoration: none;
+}
+article.pytorch-article .sphx-glr-thumbcontainer:hover {
+  -webkit-box-shadow: none;
+          box-shadow: none;
+  border-bottom-color: #ffffff;
+}
+article.pytorch-article .sphx-glr-thumbcontainer:hover .figure:before {
+  bottom: 100%;
+}
+article.pytorch-article .sphx-glr-thumbcontainer .figure {
+  width: 80%;
+}
+article.pytorch-article .sphx-glr-thumbcontainer .figure:before {
+  content: "";
+  display: block;
+  position: absolute;
+  top: 0;
+  bottom: 35%;
+  left: 0;
+  right: 0;
+  background: #8A94B3;
+  opacity: 0.10;
+}
+article.pytorch-article .sphx-glr-thumbcontainer .figure a.reference.internal {
+  text-align: left;
+}
+@media screen and (min-width: 768px) {
+  article.pytorch-article .sphx-glr-thumbcontainer:after {
+    content: "";
+    display: block;
+    width: 0;
+    height: 1px;
+    position: absolute;
+    bottom: 0;
+    left: 0;
+    background-color: #e44c2c;
+    -webkit-transition: width .250s ease-in-out;
+    transition: width .250s ease-in-out;
+  }
+  article.pytorch-article .sphx-glr-thumbcontainer:hover:after {
+    width: 100%;
+  }
+}
+@media screen and (min-width: 768px) {
+  article.pytorch-article .sphx-glr-thumbcontainer:after {
+    background-color: #ee4c2c;
+  }
+}
+
+article.pytorch-article .section :not(dt) > code {
+  color: #262626;
+  border-top: solid 2px #ffffff;
+  background-color: #ffffff;
+  border-bottom: solid 2px #ffffff;
+  padding: 0px 3px;
+  -webkit-box-decoration-break: clone;
+  box-decoration-break: clone;
+}
+article.pytorch-article .section :not(dt) > code .pre {
+  outline: 0px;
+  padding: 0px;
+}
+article.pytorch-article .function dt, article.pytorch-article .attribute dt, article.pytorch-article .class .attribute dt, article.pytorch-article .class dt {
+  position: relative;
+  background: #f3f4f7;
+  padding: 0.5rem;
+  border-left: 3px solid #ee4c2c;
+  word-wrap: break-word;
+  padding-right: 100px;
+}
+
+article.pytorch-article .class dt.field-odd, article.pytorch-article .class dt.field-even,
+article.pytorch-article .function dt.field-odd, article.pytorch-article .function dt.field-even,
+article.pytorch-article .attribute dt.field-odd, article.pytorch-article .attribute dt.field-even,
+article.pytorch-article .class .attribute dt.field-odd,article.pytorch-article .class .attribute dt.field-even,
+article.pytorch-article .class .method dt.field-odd, article.pytorch-article .class .staticmethod dt.field-odd,
+article.pytorch-article .class .method dt.field-even, article.pytorch-article .class .staticmethod dt.field-even
+{
+  background: none;
+  padding-right: -20px;
+  border-top: none;
+  border-left: none;
+  padding-left: 0.0rem;
+  padding-top: 0.0rem;
+  padding-bottom: 0.0rem;
+  font-weight: 700;
+}
+
+article.pytorch-article .function dt.sig {
+  position: relative;
+  background: #f3f4f7;
+  padding: 0.8rem;
+  border-left: 3px solid #ee4c2c;
+  word-wrap: break-word;
+  padding-right: 100px;
+  font-weight: 500;
+}
+
+article.pytorch-article .function dt {
+  background: #ffffff;
+  padding-right: -20px;
+  border-left: none;
+  border-top: none;
+  padding-left: 0.0rem;
+  padding-top: 0.0rem;
+  padding-bottom: 0.0rem;
+  font-weight: 700;
+}
+
+article.pytorch-article .function dt em.property, article.pytorch-article .attribute dt em.property, article.pytorch-article .class dt em.property {
+  font-family: inherit;
+}
+article.pytorch-article .function dt em, article.pytorch-article .attribute dt em, article.pytorch-article .class .attribute dt em, article.pytorch-article .class dt em, article.pytorch-article .function dt .sig-paren, article.pytorch-article .attribute dt .sig-paren, article.pytorch-article .class dt .sig-paren {
+  font-family: IBMPlexMono,SFMono-Regular,Menlo,Monaco,Consolas,"Liberation Mono","Courier New",monospace;
+  font-size: 87.5%;
+}
+article.pytorch-article .function dt a, article.pytorch-article .attribute dt a, article.pytorch-article .class .attribute dt a, article.pytorch-article .class dt a {
+  right: 30px;
+  padding-right: 0;
+  top: 50%;
+  -webkit-transform: perspective(1px) translateY(-50%);
+          transform: perspective(1px) translateY(-50%);
+}
+article.pytorch-article .function dt:hover .viewcode-link, article.pytorch-article .attribute dt:hover .viewcode-link, article.pytorch-article .class dt:hover .viewcode-link {
+  color: #ee4c2c;
+}
+article.pytorch-article .function .anchorjs-link, article.pytorch-article .attribute .anchorjs-link, article.pytorch-article .class .anchorjs-link {
+  display: inline;
+  position: absolute;
+  right: 8px;
+  font-size: 1.5625rem !important;
+  padding-left: 0;
+}
+article.pytorch-article .function dt > code, article.pytorch-article .attribute dt > code, article.pytorch-article .class .attribute dt > code, article.pytorch-article .class dt > code {
+  color: #262626;
+  border-top: solid 2px #f3f4f7;
+  background-color: #f3f4f7;
+  border-bottom: solid 2px #f3f4f7;
+  -webkit-box-decoration-break: clone;
+  box-decoration-break: clone;
+}
+article.pytorch-article .function .viewcode-link, article.pytorch-article .attribute .viewcode-link, article.pytorch-article .class .viewcode-link {
+  padding-left: 0.6rem;
+  position: absolute;
+  font-size: 0.875rem;
+  color: #979797;
+  letter-spacing: 0;
+  line-height: 1.5rem;
+  text-transform: uppercase;
+}
+article.pytorch-article .function dd, article.pytorch-article .attribute dd, article.pytorch-article .class .attribute dd, article.pytorch-article .class dd {
+  padding-left: 3.75rem;
+}
+article.pytorch-article .function dd p, article.pytorch-article .attribute dd p, article.pytorch-article .class .attribute dd p, article.pytorch-article .class dd p {
+  color: #262626;
+}
+article.pytorch-article .function table tbody tr th.field-name, article.pytorch-article .attribute table tbody tr th.field-name, article.pytorch-article .class table tbody tr th.field-name {
+  white-space: nowrap;
+  color: #262626;
+  width: 20%;
+}
+@media screen and (min-width: 768px) {
+  article.pytorch-article .function table tbody tr th.field-name, article.pytorch-article .attribute table tbody tr th.field-name, article.pytorch-article .class table tbody tr th.field-name {
+    width: 15%;
+  }
+}
+article.pytorch-article .function table tbody tr td.field-body, article.pytorch-article .attribute table tbody tr td.field-body, article.pytorch-article .class table tbody tr td.field-body {
+  padding: 0.625rem;
+  width: 80%;
+  color: #262626;
+}
+@media screen and (min-width: 768px) {
+  article.pytorch-article .function table tbody tr td.field-body, article.pytorch-article .attribute table tbody tr td.field-body, article.pytorch-article .class table tbody tr td.field-body {
+    width: 85%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  article.pytorch-article .function table tbody tr td.field-body, article.pytorch-article .attribute table tbody tr td.field-body, article.pytorch-article .class table tbody tr td.field-body {
+    padding-left: 1.25rem;
+  }
+}
+article.pytorch-article .function table tbody tr td.field-body p, article.pytorch-article .attribute table tbody tr td.field-body p, article.pytorch-article .class table tbody tr td.field-body p {
+  padding-left: 0px;
+}
+article.pytorch-article .function table tbody tr td.field-body p:last-of-type, article.pytorch-article .attribute table tbody tr td.field-body p:last-of-type, article.pytorch-article .class table tbody tr td.field-body p:last-of-type {
+  margin-bottom: 0;
+}
+article.pytorch-article .function table tbody tr td.field-body ol, article.pytorch-article .attribute table tbody tr td.field-body ol, article.pytorch-article .class table tbody tr td.field-body ol, article.pytorch-article .function table tbody tr td.field-body ul, article.pytorch-article .attribute table tbody tr td.field-body ul, article.pytorch-article .class table tbody tr td.field-body ul {
+  padding-left: 1rem;
+  padding-bottom: 0;
+}
+article.pytorch-article .function table.docutils.field-list, article.pytorch-article .attribute table.docutils.field-list, article.pytorch-article .class table.docutils.field-list {
+  margin-bottom: 0.75rem;
+}
+article.pytorch-article .attribute .has-code {
+  float: none;
+}
+article.pytorch-article .class dt {
+  border-left: none;
+  border-top: 3px solid #ee4c2c;
+  padding-left: 4em;
+}
+article.pytorch-article .class dt em.property {
+  position: absolute;
+  left: 0.5rem;
+}
+article.pytorch-article .class dd .docutils dt {
+  padding-left: 0.5rem;
+}
+article.pytorch-article .class em.property {
+  text-transform: uppercase;
+  font-style: normal;
+  color: #ee4c2c;
+  font-size: 1rem;
+  letter-spacing: 0;
+  padding-right: 0.75rem;
+}
+article.pytorch-article .class dl dt em.property {
+  position: static;
+  left: 0;
+  padding-right: 0;
+}
+
+article.pytorch-article .class .method dt,
+article.pytorch-article .class .staticmethod dt {
+  border-left: 3px solid #ee4c2c;
+  border-top: none;
+}
+
+article.pytorch-article .class .method dt,
+article.pytorch-article .class .staticmethod dt {
+  padding-left: 0.5rem;
+}
+article.pytorch-article .class .attribute dt {
+  border-top: none;
+}
+article.pytorch-article .class .attribute dt em.property {
+  position: relative;
+  left: 0;
+}
+article.pytorch-article table {
+  table-layout: fixed;
+}
+
+div.deprecated p {
+  display: inline;
+}
+div.deprecated,
+div.versionchanged {
+  margin-top: 0.5rem;
+  padding: 0.5rem;
+  margin-bottom: 0.5rem;
+  border: none;
+}
+div.versionadded {
+  margin: 1rem 0;
+}
+div.deprecated p:last-child,
+div.versionchanged p:last-child,
+div.versionadded p:last-child {
+  margin-bottom: 0
+}
+div.deprecated {
+  color: #b94a48;
+  background-color: #fdede9;
+}
+div.versionchanged {
+  background-color: #fffbe8;
+}
+
+article.pytorch-article .note,
+article.pytorch-article .warning,
+article.pytorch-article .tip,
+article.pytorch-article .seealso,
+article.pytorch-article .hint,
+article.pytorch-article .important,
+article.pytorch-article .caution,
+article.pytorch-article .danger,
+article.pytorch-article .attention,
+article.pytorch-article .error {
+  background: #f3f4f7;
+  margin-top: 1.875rem;
+  margin-bottom: 1.125rem;
+}
+article.pytorch-article .note .admonition-title,
+article.pytorch-article .warning .admonition-title,
+article.pytorch-article .tip .admonition-title,
+article.pytorch-article .seealso .admonition-title,
+article.pytorch-article .hint .admonition-title,
+article.pytorch-article .important .admonition-title,
+article.pytorch-article .caution .admonition-title,
+article.pytorch-article .danger .admonition-title,
+article.pytorch-article .attention .admonition-title,
+article.pytorch-article .error .admonition-title {
+  color: #ffffff;
+  letter-spacing: 1px;
+  text-transform: uppercase;
+  margin-bottom: 1.125rem;
+  padding: 3px 0 3px 1.375rem;
+  position: relative;
+  font-size: 0.875rem;
+}
+article.pytorch-article .note .admonition-title:before,
+article.pytorch-article .warning .admonition-title:before,
+article.pytorch-article .tip .admonition-title:before,
+article.pytorch-article .seealso .admonition-title:before,
+article.pytorch-article .hint .admonition-title:before,
+article.pytorch-article .important .admonition-title:before,
+article.pytorch-article .caution .admonition-title:before,
+article.pytorch-article .danger .admonition-title:before,
+article.pytorch-article .attention .admonition-title:before,
+article.pytorch-article .error .admonition-title:before {
+  content: "\2022";
+  position: absolute;
+  left: 9px;
+  color: #ffffff;
+  top: 2px;
+}
+article.pytorch-article .note p:nth-child(n + 2),
+article.pytorch-article .warning p:nth-child(n + 2),
+article.pytorch-article .tip p:nth-child(n + 2),
+article.pytorch-article .seealso p:nth-child(n + 2),
+article.pytorch-article .hint p:nth-child(n + 2),
+article.pytorch-article .important p:nth-child(n + 2),
+article.pytorch-article .caution p:nth-child(n + 2),
+article.pytorch-article .danger p:nth-child(n + 2),
+article.pytorch-article .attention p:nth-child(n + 2),
+article.pytorch-article .error p:nth-child(n + 2) {
+  padding: 0 1.375rem;
+}
+article.pytorch-article .note table,
+article.pytorch-article .warning table,
+article.pytorch-article .tip table,
+article.pytorch-article .seealso table,
+article.pytorch-article .hint table,
+article.pytorch-article .important table,
+article.pytorch-article .caution table,
+article.pytorch-article .danger table,
+article.pytorch-article .attention table,
+article.pytorch-article .error table {
+  margin: 0 2rem;
+  width: auto;
+}
+article.pytorch-article .note :not(dt) > code,
+article.pytorch-article .warning :not(dt) > code,
+article.pytorch-article .tip :not(dt) > code,
+article.pytorch-article .seealso :not(dt) > code,
+article.pytorch-article .hint :not(dt) > code,
+article.pytorch-article .important :not(dt) > code,
+article.pytorch-article .caution :not(dt) > code,
+article.pytorch-article .danger :not(dt) > code,
+article.pytorch-article .attention :not(dt) > code,
+article.pytorch-article .error :not(dt) > code {
+  border-top: solid 2px #ffffff;
+  background-color: #ffffff;
+  border-bottom: solid 2px #ffffff;
+  padding: 0px 3px;
+  -webkit-box-decoration-break: clone;
+  box-decoration-break: clone;
+  outline: 1px solid #e9e9e9;
+}
+article.pytorch-article .note :not(dt) > code .pre,
+article.pytorch-article .warning :not(dt) > code .pre,
+article.pytorch-article .tip :not(dt) > code .pre,
+article.pytorch-article .seealso :not(dt) > code .pre,
+article.pytorch-article .hint :not(dt) > code .pre,
+article.pytorch-article .important :not(dt) > code .pre,
+article.pytorch-article .caution :not(dt) > code .pre,
+article.pytorch-article .danger :not(dt) > code .pre,
+article.pytorch-article .attention :not(dt) > code .pre,
+article.pytorch-article .error :not(dt) > code .pre {
+  outline: 0px;
+  padding: 0px;
+}
+article.pytorch-article .note pre,
+article.pytorch-article .warning pre,
+article.pytorch-article .tip pre,
+article.pytorch-article .seealso pre,
+article.pytorch-article .hint pre,
+article.pytorch-article .important pre,
+article.pytorch-article .caution pre,
+article.pytorch-article .danger pre,
+article.pytorch-article .attention pre,
+article.pytorch-article .error pre {
+  margin-bottom: 0;
+}
+article.pytorch-article .note .highlight,
+article.pytorch-article .warning .highlight,
+article.pytorch-article .tip .highlight,
+article.pytorch-article .seealso .highlight,
+article.pytorch-article .hint .highlight,
+article.pytorch-article .important .highlight,
+article.pytorch-article .caution .highlight,
+article.pytorch-article .danger .highlight,
+article.pytorch-article .attention .highlight,
+article.pytorch-article .error .highlight {
+  margin: 0 2rem 1.125rem 2rem;
+}
+article.pytorch-article .note ul,
+article.pytorch-article .note ol,
+article.pytorch-article .warning ul,
+article.pytorch-article .warning ol,
+article.pytorch-article .tip ul,
+article.pytorch-article .tip ol,
+article.pytorch-article .seealso ul,
+article.pytorch-article .seealso ol,
+article.pytorch-article .hint ul,
+article.pytorch-article .hint ol,
+article.pytorch-article .important ul,
+article.pytorch-article .important ol,
+article.pytorch-article .caution ul,
+article.pytorch-article .caution ol,
+article.pytorch-article .danger ul,
+article.pytorch-article .danger ol,
+article.pytorch-article .attention ul,
+article.pytorch-article .attention ol,
+article.pytorch-article .error ul,
+article.pytorch-article .error ol {
+  padding-left: 3.25rem;
+}
+article.pytorch-article .note ul li,
+article.pytorch-article .note ol li,
+article.pytorch-article .warning ul li,
+article.pytorch-article .warning ol li,
+article.pytorch-article .tip ul li,
+article.pytorch-article .tip ol li,
+article.pytorch-article .seealso ul li,
+article.pytorch-article .seealso ol li,
+article.pytorch-article .hint ul li,
+article.pytorch-article .hint ol li,
+article.pytorch-article .important ul li,
+article.pytorch-article .important ol li,
+article.pytorch-article .caution ul li,
+article.pytorch-article .caution ol li,
+article.pytorch-article .danger ul li,
+article.pytorch-article .danger ol li,
+article.pytorch-article .attention ul li,
+article.pytorch-article .attention ol li,
+article.pytorch-article .error ul li,
+article.pytorch-article .error ol li {
+  color: #262626;
+}
+article.pytorch-article .note p,
+article.pytorch-article .warning p,
+article.pytorch-article .tip p,
+article.pytorch-article .seealso p,
+article.pytorch-article .hint p,
+article.pytorch-article .important p,
+article.pytorch-article .caution p,
+article.pytorch-article .danger p,
+article.pytorch-article .attention p,
+article.pytorch-article .error p {
+  margin-top: 1.125rem;
+}
+article.pytorch-article .note .admonition-title {
+  background: #54c7ec;
+}
+article.pytorch-article .warning .admonition-title {
+  background: #e94f3b;
+}
+article.pytorch-article .tip .admonition-title {
+  background: #6bcebb;
+}
+article.pytorch-article .seealso .admonition-title {
+  background: #6bcebb;
+}
+article.pytorch-article .hint .admonition-title {
+  background: #a2cdde;
+}
+article.pytorch-article .important .admonition-title {
+  background: #5890ff;
+}
+article.pytorch-article .caution .admonition-title {
+  background: #f7923a;
+}
+article.pytorch-article .danger .admonition-title {
+  background: #db2c49;
+}
+article.pytorch-article .attention .admonition-title {
+  background: #f5a623;
+}
+article.pytorch-article .error .admonition-title {
+  background: #cc2f90;
+}
+article.pytorch-article .sphx-glr-download-link-note.admonition.note,
+article.pytorch-article .reference.download.internal, article.pytorch-article .sphx-glr-signature {
+  display: none;
+}
+article.pytorch-article .admonition > p:last-of-type {
+  margin-bottom: 0;
+  padding-bottom: 1.125rem !important;
+}
+
+.pytorch-article div.sphx-glr-download a {
+  background-color: #f3f4f7;
+  background-image: url("../images/arrow-down-orange.svg");
+  background-repeat: no-repeat;
+  background-position: left 10px center;
+  background-size: 15px 15px;
+  border-radius: 0;
+  border: none;
+  display: block;
+  text-align: left;
+  padding: 0.9375rem 3.125rem;
+  position: relative;
+  margin: 1.25rem auto;
+}
+@media screen and (min-width: 768px) {
+  .pytorch-article div.sphx-glr-download a:after {
+    content: "";
+    display: block;
+    width: 0;
+    height: 1px;
+    position: absolute;
+    bottom: 0;
+    left: 0;
+    background-color: #e44c2c;
+    -webkit-transition: width .250s ease-in-out;
+    transition: width .250s ease-in-out;
+  }
+  .pytorch-article div.sphx-glr-download a:hover:after {
+    width: 100%;
+  }
+}
+@media screen and (min-width: 768px) {
+  .pytorch-article div.sphx-glr-download a:after {
+    background-color: #ee4c2c;
+  }
+}
+@media screen and (min-width: 768px) {
+  .pytorch-article div.sphx-glr-download a {
+    background-position: left 20px center;
+  }
+}
+.pytorch-article div.sphx-glr-download a:hover {
+  -webkit-box-shadow: none;
+          box-shadow: none;
+  text-decoration: none;
+  background-image: url("../images/arrow-down-orange.svg");
+  background-color: #f3f4f7;
+}
+.pytorch-article div.sphx-glr-download a span.pre {
+  background-color: transparent;
+  font-size: 1.125rem;
+  padding: 0;
+  color: #262626;
+}
+.pytorch-article div.sphx-glr-download a code, .pytorch-article div.sphx-glr-download a kbd, .pytorch-article div.sphx-glr-download a pre, .pytorch-article div.sphx-glr-download a samp, .pytorch-article div.sphx-glr-download a span.pre {
+  font-family: FreightSans, Helvetica Neue, Helvetica, Arial, sans-serif;
+}
+
+.pytorch-article p.sphx-glr-script-out {
+  margin-bottom: 1.125rem;
+}
+
+.pytorch-article div.sphx-glr-script-out {
+  margin-bottom: 2.5rem;
+}
+.pytorch-article div.sphx-glr-script-out .highlight {
+  margin-left: 0;
+  margin-top: 0;
+}
+.pytorch-article div.sphx-glr-script-out .highlight pre {
+  background-color: #fdede9;
+  padding: 1.5625rem;
+  color: #837b79;
+}
+.pytorch-article div.sphx-glr-script-out + p {
+  margin-top: unset;
+}
+
+article.pytorch-article .wy-table-responsive table {
+  border: none;
+  border-color: #ffffff !important;
+  table-layout: fixed;
+}
+article.pytorch-article .wy-table-responsive table thead tr {
+  border-bottom: 2px solid #6c6c6d;
+}
+article.pytorch-article .wy-table-responsive table thead th {
+  line-height: 1.75rem;
+  padding-left: 0.9375rem;
+  padding-right: 0.9375rem;
+}
+article.pytorch-article .wy-table-responsive table tbody .row-odd {
+  background-color: #f3f4f7;
+}
+article.pytorch-article .wy-table-responsive table tbody td {
+  color: #6c6c6d;
+  white-space: normal;
+  padding: 0.9375rem;
+  font-size: 1rem;
+  line-height: 1.375rem;
+}
+article.pytorch-article .wy-table-responsive table tbody td .pre {
+  background: #ffffff;
+  color: #ee4c2c;
+  font-size: 87.5%;
+}
+article.pytorch-article .wy-table-responsive table tbody td code {
+  font-size: 87.5%;
+}
+
+a[rel~="prev"], a[rel~="next"] {
+  padding: 0.375rem 0 0 0;
+}
+
+img.next-page,
+img.previous-page {
+  width: 8px;
+  height: 10px;
+  position: relative;
+  top: -1px;
+}
+
+img.previous-page {
+  -webkit-transform: scaleX(-1);
+          transform: scaleX(-1);
+}
+
+.rst-footer-buttons {
+  margin-top: 1.875rem;
+  margin-bottom: 1.875rem;
+}
+.rst-footer-buttons .btn:focus,
+.rst-footer-buttons .btn.focus {
+  -webkit-box-shadow: none;
+          box-shadow: none;
+}
+
+article.pytorch-article blockquote {
+  margin-left: 3.75rem;
+  color: #6c6c6d;
+}
+
+article.pytorch-article .caption {
+  color: #6c6c6d;
+  letter-spacing: 0.25px;
+  line-height: 2.125rem;
+}
+
+article.pytorch-article .math {
+  color: #262626;
+  width: auto;
+  text-align: center;
+}
+article.pytorch-article .math img {
+  width: auto;
+}
+
+.pytorch-breadcrumbs-wrapper {
+  width: 100%;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-breadcrumbs-wrapper {
+    float: left;
+    margin-left: 3%;
+    width: 75%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-breadcrumbs-wrapper {
+    width: 850px;
+    margin-left: 1.875rem;
+  }
+}
+.pytorch-breadcrumbs-wrapper .pytorch-breadcrumbs-aside {
+  float: right;
+}
+.pytorch-breadcrumbs-wrapper .pytorch-breadcrumbs-aside .fa.fa-github {
+  margin-top: 5px;
+  display: block;
+}
+
+.pytorch-article .container {
+  padding-left: 0;
+  padding-right: 0;
+  max-width: none;
+}
+
+a:link,
+a:visited,
+a:hover {
+  color: #ee4c2c;
+}
+
+::-webkit-input-placeholder {
+  color: #ee4c2c;
+}
+
+::-moz-placeholder {
+  color: #ee4c2c;
+}
+
+:-ms-input-placeholder {
+  color: #ee4c2c;
+}
+
+:-moz-placeholder {
+  color: #ee4c2c;
+}
+
+@media screen and (min-width: 768px) {
+  .site-footer a:hover {
+    color: #ee4c2c;
+  }
+}
+
+.docs-tutorials-resources a {
+  color: #ee4c2c;
+}
+
+.header-holder {
+  position: relative;
+  z-index: 201;
+}
+
+.header-holder .main-menu ul li.active:after {
+  color: #ee4c2c;
+}
+.header-holder .main-menu ul li.active a {
+  color: #ee4c2c;
+}
+@media screen and (min-width: 1100px) {
+  .header-holder .main-menu ul li a:hover {
+    color: #ee4c2c;
+  }
+}
+
+.mobile-main-menu.open ul li.active a {
+  color: #ee4c2c;
+}
+
+.version {
+  padding-bottom: 1rem;
+}
+
+.pytorch-call-to-action-links {
+  padding-top: 0;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+}
+@media screen and (min-width: 768px) {
+  .pytorch-call-to-action-links {
+    padding-top: 2.5rem;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  .pytorch-call-to-action-links {
+    padding-top: 0;
+  }
+}
+@media (min-width: 1100px) and (max-width: 1239px) {
+  .pytorch-call-to-action-links {
+    padding-top: 2.5rem;
+  }
+}
+.pytorch-call-to-action-links #tutorial-type {
+  display: none;
+}
+.pytorch-call-to-action-links .call-to-action-img, .pytorch-call-to-action-links .call-to-action-notebook-img {
+  height: 1.375rem;
+  width: 1.375rem;
+  margin-right: 10px;
+}
+.pytorch-call-to-action-links .call-to-action-notebook-img {
+  height: 1rem;
+}
+.pytorch-call-to-action-links a {
+  padding-right: 1.25rem;
+  color: #000000;
+  cursor: pointer;
+}
+.pytorch-call-to-action-links a:hover {
+  color: #e44c2c;
+}
+.pytorch-call-to-action-links a .call-to-action-desktop-view {
+  display: none;
+}
+@media screen and (min-width: 768px) {
+  .pytorch-call-to-action-links a .call-to-action-desktop-view {
+    display: block;
+  }
+}
+.pytorch-call-to-action-links a .call-to-action-mobile-view {
+  display: block;
+}
+@media screen and (min-width: 768px) {
+  .pytorch-call-to-action-links a .call-to-action-mobile-view {
+    display: none;
+  }
+}
+.pytorch-call-to-action-links a #google-colab-link, .pytorch-call-to-action-links a #download-notebook-link,
+.pytorch-call-to-action-links a #github-view-link {
+  padding-bottom: 0.625rem;
+  border-bottom: 1px solid #f3f4f7;
+  padding-right: 2.5rem;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+}
+.pytorch-call-to-action-links a #google-colab-link:hover, .pytorch-call-to-action-links a #download-notebook-link:hover,
+.pytorch-call-to-action-links a #github-view-link:hover {
+  border-bottom-color: #e44c2c;
+  color: #e44c2c;
+}
+
+#tutorial-cards-container #tutorial-cards {
+  width: 100%;
+}
+#tutorial-cards-container .tutorials-nav {
+  padding-left: 0;
+  padding-right: 0;
+  padding-bottom: 0;
+}
+#tutorial-cards-container .tutorials-hr {
+  margin-top: 1rem;
+  margin-bottom: 1rem;
+}
+#tutorial-cards-container .card.tutorials-card {
+  border-radius: 0;
+  border-color: #f3f4f7;
+  height: 98px;
+  margin-bottom: 1.25rem;
+  margin-bottom: 1.875rem;
+  overflow: scroll;
+  background-color: #f3f4f7;
+  cursor: pointer;
+}
+@media screen and (min-width: 1240px) {
+  #tutorial-cards-container .card.tutorials-card {
+    height: 200px;
+    overflow: inherit;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  #tutorial-cards-container .card.tutorials-card {
+    height: 200px;
+    overflow: scroll;
+  }
+}
+#tutorial-cards-container .card.tutorials-card .tutorials-image {
+  position: absolute;
+  top: 0px;
+  right: 0px;
+  height: 96px;
+  width: 96px;
+  opacity: 0.5;
+}
+#tutorial-cards-container .card.tutorials-card .tutorials-image img {
+  height: 100%;
+  width: 100%;
+}
+@media screen and (min-width: 768px) {
+  #tutorial-cards-container .card.tutorials-card .tutorials-image {
+    height: 100%;
+    width: 25%;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  #tutorial-cards-container .card.tutorials-card .tutorials-image {
+    height: 100%;
+    width: 198px;
+  }
+}
+#tutorial-cards-container .card.tutorials-card .tutorials-image:before {
+  content: '';
+  position: absolute;
+  top: 0;
+  left: 0;
+  bottom: 0;
+  right: 0;
+  z-index: 1;
+  background: #000000;
+  opacity: .075;
+}
+#tutorial-cards-container .card.tutorials-card .card-title-container {
+  width: 70%;
+  display: -webkit-inline-box;
+  display: -ms-inline-flexbox;
+  display: inline-flex;
+}
+@media screen and (min-width: 768px) {
+  #tutorial-cards-container .card.tutorials-card .card-title-container {
+    width: 75%;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  #tutorial-cards-container .card.tutorials-card .card-title-container {
+    width: 70%;
+  }
+}
+#tutorial-cards-container .card.tutorials-card .card-title-container h4 {
+  margin-bottom: 1.125rem;
+  margin-top: 0;
+  font-size: 1.5rem;
+}
+#tutorial-cards-container .card.tutorials-card p.card-summary, #tutorial-cards-container .card.tutorials-card p.tags {
+  font-size: 0.9375rem;
+  line-height: 1.5rem;
+  margin-bottom: 0;
+  color: #6c6c6d;
+  font-weight: 400;
+  width: 70%;
+}
+@media screen and (min-width: 768px) {
+  #tutorial-cards-container .card.tutorials-card p.card-summary, #tutorial-cards-container .card.tutorials-card p.tags {
+    width: 75%;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  #tutorial-cards-container .card.tutorials-card p.card-summary, #tutorial-cards-container .card.tutorials-card p.tags {
+    width: 70%;
+  }
+}
+#tutorial-cards-container .card.tutorials-card p.tags {
+  margin-top: 30px;
+  text-overflow: ellipsis;
+  white-space: nowrap;
+  overflow: hidden;
+}
+#tutorial-cards-container .card.tutorials-card h4 {
+  color: #262626;
+  margin-bottom: 1.125rem;
+}
+#tutorial-cards-container .card.tutorials-card a {
+  height: 100%;
+}
+@media screen and (min-width: 768px) {
+  #tutorial-cards-container .card.tutorials-card a {
+    min-height: 190px;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  #tutorial-cards-container .card.tutorials-card a {
+    min-height: 234px;
+  }
+}
+@media screen and (min-width: 768px) {
+  #tutorial-cards-container .card.tutorials-card:after {
+    content: "";
+    display: block;
+    width: 0;
+    height: 1px;
+    position: absolute;
+    bottom: 0;
+    left: 0;
+    background-color: #e44c2c;
+    -webkit-transition: width .250s ease-in-out;
+    transition: width .250s ease-in-out;
+  }
+  #tutorial-cards-container .card.tutorials-card:hover:after {
+    width: 100%;
+  }
+}
+#tutorial-cards-container .card.tutorials-card:hover {
+  background-color: #ffffff;
+  border: 1px solid #e2e2e2;
+  border-bottom: none;
+}
+#tutorial-cards-container .card.tutorials-card:hover p.card-summary {
+  color: #262626;
+}
+#tutorial-cards-container .card.tutorials-card:hover .tutorials-image {
+  opacity: unset;
+}
+#tutorial-cards-container .tutorial-tags-container {
+  width: 75%;
+}
+#tutorial-cards-container .tutorial-tags-container.active {
+  width: 0;
+}
+#tutorial-cards-container .tutorial-filter-menu ul {
+  list-style-type: none;
+  padding-left: 1.25rem;
+}
+#tutorial-cards-container .tutorial-filter-menu ul li {
+  padding-right: 1.25rem;
+  word-break: break-all;
+}
+#tutorial-cards-container .tutorial-filter-menu ul li a {
+  color: #979797;
+}
+#tutorial-cards-container .tutorial-filter-menu ul li a:hover {
+  color: #e44c2c;
+}
+#tutorial-cards-container .tutorial-filter {
+  cursor: pointer;
+}
+#tutorial-cards-container .filter-btn {
+  color: #979797;
+  border: 1px solid #979797;
+  display: inline-block;
+  text-align: center;
+  white-space: nowrap;
+  vertical-align: middle;
+  padding: 0.375rem 0.75rem;
+  font-size: 1rem;
+  line-height: 1.5;
+  margin-bottom: 5px;
+}
+#tutorial-cards-container .filter-btn:hover {
+  border: 1px solid #e44c2c;
+  color: #e44c2c;
+}
+#tutorial-cards-container .filter-btn.selected {
+  background-color: #e44c2c;
+  border: 1px solid #e44c2c;
+  color: #ffffff;
+}
+#tutorial-cards-container .all-tag-selected {
+  background-color: #979797;
+  color: #ffffff;
+}
+#tutorial-cards-container .all-tag-selected:hover {
+  border-color: #979797;
+  color: #ffffff;
+}
+#tutorial-cards-container .pagination .page {
+  border: 1px solid #dee2e6;
+  padding: 0.5rem 0.75rem;
+}
+#tutorial-cards-container .pagination .active .page {
+  background-color: #dee2e6;
+}
+
+article.pytorch-article .tutorials-callout-container {
+  padding-bottom: 50px;
+}
+article.pytorch-article .tutorials-callout-container .col-md-6 {
+  padding-bottom: 10px;
+}
+article.pytorch-article .tutorials-callout-container .text-container {
+  padding: 10px 0px 30px 0px;
+  padding-bottom: 10px;
+}
+article.pytorch-article .tutorials-callout-container .text-container .body-paragraph {
+  color: #666666;
+  font-weight: 300;
+  font-size: 1.125rem;
+  line-height: 1.875rem;
+}
+article.pytorch-article .tutorials-callout-container .btn.callout-button {
+  font-size: 1.125rem;
+  border-radius: 0;
+  border: none;
+  background-color: #f3f4f7;
+  color: #6c6c6d;
+  font-weight: 400;
+  position: relative;
+  letter-spacing: 0.25px;
+}
+@media screen and (min-width: 768px) {
+  article.pytorch-article .tutorials-callout-container .btn.callout-button:after {
+    content: "";
+    display: block;
+    width: 0;
+    height: 1px;
+    position: absolute;
+    bottom: 0;
+    left: 0;
+    background-color: #e44c2c;
+    -webkit-transition: width .250s ease-in-out;
+    transition: width .250s ease-in-out;
+  }
+  article.pytorch-article .tutorials-callout-container .btn.callout-button:hover:after {
+    width: 100%;
+  }
+}
+article.pytorch-article .tutorials-callout-container .btn.callout-button a {
+  color: inherit;
+}
+
+.pytorch-container {
+  margin: 0 auto;
+  padding: 0 1.875rem;
+  width: auto;
+  position: relative;
+}
+@media screen and (min-width: 1100px) {
+  .pytorch-container {
+    padding: 0;
+  }
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-container {
+    margin-left: 25%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-container {
+    margin-left: 350px;
+  }
+}
+.pytorch-container:before, .pytorch-container:after {
+  content: "";
+  display: table;
+}
+.pytorch-container:after {
+  clear: both;
+}
+.pytorch-container {
+  *zoom: 1;
+}
+
+.pytorch-content-wrap {
+  background-color: #ffffff;
+  display: -webkit-box;
+  display: -ms-flexbox;
+  display: flex;
+  position: relative;
+  padding-top: 0;
+}
+.pytorch-content-wrap:before, .pytorch-content-wrap:after {
+  content: "";
+  display: table;
+}
+.pytorch-content-wrap:after {
+  clear: both;
+}
+.pytorch-content-wrap {
+  *zoom: 1;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-content-wrap {
+    padding-top: 45px;
+    float: left;
+    width: 100%;
+    display: block;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-content-wrap {
+    width: 100%;
+  }
+}
+
+.pytorch-content {
+  background: #ffffff;
+  width: 100%;
+  max-width: 700px;
+  position: relative;
+}
+
+.pytorch-content-left {
+  min-height: 100vh;
+  margin-top: 2.5rem;
+  width: 100%;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-content-left {
+    margin-top: 0;
+    margin-left: 3%;
+    width: 75%;
+    float: left;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-content-left {
+    width: 850px;
+    margin-left: 30px;
+  }
+}
+.pytorch-content-left .main-content {
+  padding-top: 0.9375rem;
+}
+.pytorch-content-left .main-content ul.simple {
+  padding-bottom: 1.25rem;
+}
+.pytorch-content-left .main-content .note:nth-child(1), .pytorch-content-left .main-content .warning:nth-child(1) {
+  margin-top: 0;
+}
+
+.pytorch-content-right {
+  display: none;
+  position: relative;
+  overflow-x: hidden;
+  overflow-y: hidden;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-content-right {
+    display: block;
+    margin-left: 0;
+    width: 19%;
+    float: left;
+    height: 100%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-content-right {
+    width: 280px;
+  }
+}
+
+@media screen and (min-width: 1101px) {
+  .pytorch-side-scroll {
+    position: relative;
+    overflow-x: hidden;
+    overflow-y: scroll;
+    height: 100%;
+  }
+}
+
+.pytorch-menu-vertical {
+  padding: 1.25rem 1.875rem 2.5rem 1.875rem;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-menu-vertical {
+    display: block;
+    padding-top: 0;
+    padding-right: 13.5%;
+    padding-bottom: 5.625rem;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-menu-vertical {
+    padding-left: 0;
+    padding-right: 1.5625rem;
+  }
+}
+
+.pytorch-left-menu {
+  display: none;
+  background-color: #f3f4f7;
+  color: #262626;
+  overflow: scroll;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-left-menu {
+    display: block;
+    overflow-x: hidden;
+    overflow-y: hidden;
+    padding-bottom: 110px;
+    padding: 0 1.875rem 0 0;
+    width: 25%;
+    z-index: 200;
+    float: left;
+  }
+  .pytorch-left-menu.make-fixed {
+    position: fixed;
+    top: 0;
+    bottom: 0;
+    left: 0;
+    float: none;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-left-menu {
+    padding: 0 0 0 1.875rem;
+    width: 350px;
+  }
+}
+
+.expand-menu, .hide-menu {
+  color: #6c6c6d;
+  padding-left: 10px;
+  cursor: pointer;
+}
+
+.collapse {
+  display: none;
+}
+
+.left-nav-top-caption {
+  padding-top: 1rem;
+}
+
+.pytorch-left-menu p.caption {
+  color: #262626;
+  display: block;
+  font-size: 1rem;
+  line-height: 1.375rem;
+  margin-bottom: 1rem;
+  text-transform: none;
+  white-space: normal;
+}
+
+.pytorch-left-menu-search {
+  margin-bottom: 2.5rem;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-left-menu-search {
+    margin: 1.25rem 0.625rem 1.875rem 0;
+  }
+}
+
+.pytorch-left-menu-search ::-webkit-input-placeholder {
+  color: #262626;
+}
+.pytorch-left-menu-search ::-moz-placeholder {
+  color: #262626;
+}
+.pytorch-left-menu-search :-ms-input-placeholder {
+  color: #262626;
+}
+.pytorch-left-menu-search ::-ms-input-placeholder {
+  color: #262626;
+}
+.pytorch-left-menu-search ::placeholder {
+  color: #262626;
+}
+
+.pytorch-left-menu-search :focus::-webkit-input-placeholder {
+  color: transparent;
+}
+.pytorch-left-menu-search :focus::-moz-placeholder {
+  color: transparent;
+}
+.pytorch-left-menu-search :focus:-ms-input-placeholder {
+  color: transparent;
+}
+.pytorch-left-menu-search :focus::-ms-input-placeholder {
+  color: transparent;
+}
+.pytorch-left-menu-search :focus::placeholder {
+  color: transparent;
+}
+
+.pytorch-left-menu-search input[type=text] {
+  border-radius: 0;
+  padding: 0.5rem 0.75rem;
+  border-color: #ffffff;
+  color: #262626;
+  border-style: solid;
+  font-size: 1rem;
+  width: 100%;
+  background-color: #f3f4f7;
+  background-image: url("../images/search-icon.svg");
+  background-repeat: no-repeat;
+  background-size: 18px 18px;
+  background-position: 12px 10px;
+  padding-left: 40px;
+  background-color: #ffffff;
+}
+.pytorch-left-menu-search input[type=text]:focus {
+  outline: 0;
+}
+
+@media screen and (min-width: 1101px) {
+  .pytorch-left-menu .pytorch-side-scroll {
+    width: 120%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-left-menu .pytorch-side-scroll {
+    width: 340px;
+  }
+}
+
+.pytorch-right-menu {
+  min-height: 100px;
+  overflow-x: hidden;
+  overflow-y: hidden;
+  left: 0;
+  z-index: 200;
+  padding-top: 0;
+  position: relative;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-right-menu {
+    width: 100%;
+  }
+  .pytorch-right-menu.scrolling-fixed {
+    position: fixed;
+    top: 45px;
+    left: 83.5%;
+    width: 14%;
+  }
+  .pytorch-right-menu.scrolling-absolute {
+    position: absolute;
+    left: 0;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-right-menu {
+    left: 0;
+    width: 380px;
+  }
+  .pytorch-right-menu.scrolling-fixed {
+    position: fixed;
+    top: 45px;
+    left: 1230px;
+  }
+  .pytorch-right-menu.scrolling-absolute {
+    position: absolute;
+    left: 0;
+  }
+}
+
+.pytorch-left-menu ul,
+.pytorch-right-menu ul {
+  list-style-type: none;
+  padding-left: 0;
+  margin-bottom: 2.5rem;
+}
+.pytorch-left-menu > ul,
+.pytorch-right-menu > ul {
+  margin-bottom: 2.5rem;
+}
+.pytorch-left-menu a:link,
+.pytorch-left-menu a:visited,
+.pytorch-left-menu a:hover,
+.pytorch-right-menu a:link,
+.pytorch-right-menu a:visited,
+.pytorch-right-menu a:hover {
+  color: #6c6c6d;
+  font-size: 0.875rem;
+  line-height: 1rem;
+  padding: 0;
+  text-decoration: none;
+}
+.pytorch-left-menu a:link.reference.internal,
+.pytorch-left-menu a:visited.reference.internal,
+.pytorch-left-menu a:hover.reference.internal,
+.pytorch-right-menu a:link.reference.internal,
+.pytorch-right-menu a:visited.reference.internal,
+.pytorch-right-menu a:hover.reference.internal {
+  margin-bottom: 0.3125rem;
+  position: relative;
+}
+.pytorch-left-menu li code,
+.pytorch-right-menu li code {
+  border: none;
+  background: inherit;
+  color: inherit;
+  padding-left: 0;
+  padding-right: 0;
+}
+.pytorch-left-menu li span.toctree-expand,
+.pytorch-right-menu li span.toctree-expand {
+  display: block;
+  float: left;
+  margin-left: -1.2em;
+  font-size: 0.8em;
+  line-height: 1.6em;
+}
+.pytorch-left-menu li.on a, .pytorch-left-menu li.current > a,
+.pytorch-right-menu li.on a,
+.pytorch-right-menu li.current > a {
+  position: relative;
+  border: none;
+}
+.pytorch-left-menu li.on a span.toctree-expand, .pytorch-left-menu li.current > a span.toctree-expand,
+.pytorch-right-menu li.on a span.toctree-expand,
+.pytorch-right-menu li.current > a span.toctree-expand {
+  display: block;
+  font-size: 0.8em;
+  line-height: 1.6em;
+}
+.pytorch-left-menu li.toctree-l1.current > a,
+.pytorch-right-menu li.toctree-l1.current > a {
+  color: #ee4c2c;
+}
+.pytorch-left-menu li.toctree-l1.current > a:before,
+.pytorch-right-menu li.toctree-l1.current > a:before {
+  content: "\2022";
+  display: inline-block;
+  position: absolute;
+  left: -15px;
+  top: -10%;
+  font-size: 1.375rem;
+  color: #ee4c2c;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-left-menu li.toctree-l1.current > a:before,
+  .pytorch-right-menu li.toctree-l1.current > a:before {
+    left: -20px;
+  }
+}
+.pytorch-left-menu li.toctree-l1.current li.toctree-l2 > ul, .pytorch-left-menu li.toctree-l2.current li.toctree-l3 > ul,
+.pytorch-right-menu li.toctree-l1.current li.toctree-l2 > ul,
+.pytorch-right-menu li.toctree-l2.current li.toctree-l3 > ul {
+  display: none;
+}
+.pytorch-left-menu li.toctree-l1.current li.toctree-l2.current > ul, .pytorch-left-menu li.toctree-l2.current li.toctree-l3.current > ul,
+.pytorch-right-menu li.toctree-l1.current li.toctree-l2.current > ul,
+.pytorch-right-menu li.toctree-l2.current li.toctree-l3.current > ul {
+  display: block;
+}
+.pytorch-left-menu li.toctree-l2.current li.toctree-l3 > a,
+.pytorch-right-menu li.toctree-l2.current li.toctree-l3 > a {
+  display: block;
+}
+.pytorch-left-menu li.toctree-l3,
+.pytorch-right-menu li.toctree-l3 {
+  font-size: 0.9em;
+}
+.pytorch-left-menu li.toctree-l3.current li.toctree-l4 > a,
+.pytorch-right-menu li.toctree-l3.current li.toctree-l4 > a {
+  display: block;
+}
+.pytorch-left-menu li.toctree-l4,
+.pytorch-right-menu li.toctree-l4 {
+  font-size: 0.9em;
+}
+.pytorch-left-menu li.current ul,
+.pytorch-right-menu li.current ul {
+  display: block;
+}
+.pytorch-left-menu li ul,
+.pytorch-right-menu li ul {
+  margin-bottom: 0;
+  display: none;
+}
+.pytorch-left-menu li ul li a,
+.pytorch-right-menu li ul li a {
+  margin-bottom: 0;
+}
+.pytorch-left-menu a,
+.pytorch-right-menu a {
+  display: inline-block;
+  position: relative;
+}
+.pytorch-left-menu a:hover,
+.pytorch-right-menu a:hover {
+  cursor: pointer;
+}
+.pytorch-left-menu a:active,
+.pytorch-right-menu a:active {
+  cursor: pointer;
+}
+
+.pytorch-left-menu ul {
+  padding-left: 0;
+}
+
+.pytorch-right-menu a:link,
+.pytorch-right-menu a:visited,
+.pytorch-right-menu a:hover {
+  color: #6c6c6d;
+}
+.pytorch-right-menu a:link span.pre,
+.pytorch-right-menu a:visited span.pre,
+.pytorch-right-menu a:hover span.pre {
+  color: #6c6c6d;
+}
+.pytorch-right-menu a.reference.internal.expanded:before {
+  content: "-";
+  font-family: monospace;
+  position: absolute;
+  left: -12px;
+}
+.pytorch-right-menu a.reference.internal.not-expanded:before {
+  content: "+";
+  font-family: monospace;
+  position: absolute;
+  left: -12px;
+}
+.pytorch-right-menu li.active > a {
+  color: #ee4c2c;
+}
+.pytorch-right-menu li.active > a span.pre, .pytorch-right-menu li.active > a:before {
+  color: #ee4c2c;
+}
+.pytorch-right-menu li.active > a:after {
+  content: "\2022";
+  color: #e44c2c;
+  display: inline-block;
+  font-size: 1.375rem;
+  left: -17px;
+  position: absolute;
+  top: 1px;
+}
+.pytorch-right-menu .pytorch-side-scroll > ul > li > ul > li {
+  margin-bottom: 0;
+}
+.pytorch-right-menu ul ul {
+  padding-left: 0;
+}
+.pytorch-right-menu ul ul li {
+  padding-left: 0px;
+}
+.pytorch-right-menu ul ul li a.reference.internal {
+  padding-left: 0;
+}
+.pytorch-right-menu ul ul li ul {
+  display: none;
+  padding-left: 10px;
+}
+.pytorch-right-menu ul ul li li a.reference.internal {
+  padding-left: 0;
+}
+.pytorch-right-menu li ul {
+  display: block;
+}
+
+.pytorch-right-menu .pytorch-side-scroll {
+  padding-top: 20px;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-right-menu .pytorch-side-scroll {
+    width: 120%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-right-menu .pytorch-side-scroll {
+    width: 400px;
+  }
+}
+.pytorch-right-menu .pytorch-side-scroll > ul {
+  padding-left: 10%;
+  padding-right: 10%;
+  margin-bottom: 0;
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-right-menu .pytorch-side-scroll > ul {
+    padding-left: 25px;
+  }
+}
+.pytorch-right-menu .pytorch-side-scroll > ul > li > a.reference.internal {
+  color: #262626;
+  font-weight: 500;
+}
+.pytorch-right-menu .pytorch-side-scroll ul li {
+  position: relative;
+}
+
+#pytorch-right-menu .side-scroll-highlight {
+  color: #ee4c2c;
+}
+
+.header-container {
+  max-width: none;
+  margin-top: 4px;
+}
+@media screen and (min-width: 1101px) {
+  .header-container {
+    margin-top: 0;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .header-container {
+    margin-top: 0;
+  }
+}
+
+.container-fluid.header-holder {
+  padding-right: 0;
+  padding-left: 0;
+}
+
+.header-holder .container {
+  max-width: none;
+  padding-right: 1.875rem;
+  padding-left: 1.875rem;
+}
+@media screen and (min-width: 1101px) {
+  .header-holder .container {
+    padding-right: 1.875rem;
+    padding-left: 1.875rem;
+  }
+}
+
+.header-holder .main-menu {
+  -webkit-box-pack: unset;
+      -ms-flex-pack: unset;
+          justify-content: unset;
+  position: relative;
+}
+@media screen and (min-width: 1101px) {
+  .header-holder .main-menu ul {
+    padding-left: 0;
+    margin-left: 26%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .header-holder .main-menu ul {
+    padding-left: 38px;
+    margin-left: 310px;
+  }
+}
+
+.pytorch-page-level-bar {
+  display: none;
+  -webkit-box-align: center;
+      -ms-flex-align: center;
+          align-items: center;
+  background-color: #ffffff;
+  border-bottom: 1px solid #e2e2e2;
+  width: 100%;
+  z-index: 201;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-page-level-bar {
+    left: 0;
+    display: -webkit-box;
+    display: -ms-flexbox;
+    display: flex;
+    height: 45px;
+    padding-left: 0;
+    width: 100%;
+    position: absolute;
+    z-index: 1;
+  }
+  .pytorch-page-level-bar.left-menu-is-fixed {
+    position: fixed;
+    top: 0;
+    left: 25%;
+    padding-left: 0;
+    right: 0;
+    width: 75%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-page-level-bar {
+    left: 0;
+    right: 0;
+    width: auto;
+    z-index: 1;
+  }
+  .pytorch-page-level-bar.left-menu-is-fixed {
+    left: 350px;
+    right: 0;
+    width: auto;
+  }
+}
+.pytorch-page-level-bar ul, .pytorch-page-level-bar li {
+  margin: 0;
+}
+
+.pytorch-shortcuts-wrapper {
+  display: none;
+}
+@media screen and (min-width: 1101px) {
+  .pytorch-shortcuts-wrapper {
+    font-size: 0.875rem;
+    float: left;
+    margin-left: 2%;
+  }
+}
+@media screen and (min-width: 1600px) {
+  .pytorch-shortcuts-wrapper {
+    margin-left: 1.875rem;
+  }
+}
+
+.cookie-banner-wrapper {
+  display: none;
+}
+.cookie-banner-wrapper .container {
+  padding-left: 1.875rem;
+  padding-right: 1.875rem;
+  max-width: 1240px;
+}
+.cookie-banner-wrapper.is-visible {
+  display: block;
+  position: fixed;
+  bottom: 0;
+  background-color: #f3f4f7;
+  min-height: 100px;
+  width: 100%;
+  z-index: 401;
+  border-top: 3px solid #ededee;
+}
+.cookie-banner-wrapper .gdpr-notice {
+  color: #6c6c6d;
+  margin-top: 1.5625rem;
+  text-align: left;
+  max-width: 1440px;
+}
+@media screen and (min-width: 768px) {
+  .cookie-banner-wrapper .gdpr-notice {
+    width: 77%;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  .cookie-banner-wrapper .gdpr-notice {
+    width: inherit;
+  }
+}
+.cookie-banner-wrapper .gdpr-notice .cookie-policy-link {
+  color: #343434;
+}
+.cookie-banner-wrapper .close-button {
+  -webkit-appearance: none;
+     -moz-appearance: none;
+          appearance: none;
+  background: transparent;
+  border: 1px solid #f3f4f7;
+  height: 1.3125rem;
+  position: absolute;
+  bottom: 42px;
+  right: 0;
+  top: 0;
+  cursor: pointer;
+  outline: none;
+}
+@media screen and (min-width: 768px) {
+  .cookie-banner-wrapper .close-button {
+    right: 20%;
+    top: inherit;
+  }
+}
+@media (min-width: 768px) and (max-width: 1239px) {
+  .cookie-banner-wrapper .close-button {
+    right: 0;
+    top: 0;
+  }
+}
+
+.main-menu ul li .resources-dropdown a {
+  cursor: pointer;
+}
+.main-menu ul li .dropdown-menu {
+  border-radius: 0;
+  padding: 0;
+}
+.main-menu ul li .dropdown-menu .dropdown-item {
+  color: #6c6c6d;
+  border-bottom: 1px solid #e2e2e2;
+}
+.main-menu ul li .dropdown-menu .dropdown-item:last-of-type {
+  border-bottom-color: transparent;
+}
+.main-menu ul li .dropdown-menu .dropdown-item:hover {
+  background-color: #e44c2c;
+}
+.main-menu ul li .dropdown-menu .dropdown-item p {
+  font-size: 1rem;
+  color: #979797;
+}
+.main-menu ul li .dropdown-menu a.dropdown-item:hover {
+  color: #ffffff;
+}
+.main-menu ul li .dropdown-menu a.dropdown-item:hover p {
+  color: #ffffff;
+}
+
+.resources-dropdown-menu {
+  left: -75px;
+  width: 226px;
+  display: none;
+  position: absolute;
+  z-index: 1000;
+  display: none;
+  float: left;
+  min-width: 10rem;
+  padding: 0.5rem 0;
+  font-size: 1rem;
+  color: #212529;
+  text-align: left;
+  list-style: none;
+  background-color: #ffffff;
+  background-clip: padding-box;
+  border: 1px solid rgba(0, 0, 0, 0.15);
+  border-radius: 0.25rem;
+}
+
+.resources-dropdown:hover .resources-dropdown-menu {
+  display: block;
+}
+
+.main-menu ul li .resources-dropdown-menu {
+  border-radius: 0;
+  padding: 0;
+}
+.main-menu ul li.active:hover .resources-dropdown-menu {
+  display: block;
+}
+
+.main-menu ul li .resources-dropdown-menu .dropdown-item {
+  color: #6c6c6d;
+  border-bottom: 1px solid #e2e2e2;
+}
+
+.resources-dropdown .with-down-orange-arrow {
+  padding-right: 2rem;
+  position: relative;
+  background: url("../images/chevron-down-orange.svg");
+  background-size: 14px 18px;
+  background-position: top 7px right 10px;
+  background-repeat: no-repeat;
+}
+
+.with-down-arrow {
+  padding-right: 2rem;
+  position: relative;
+  background-image: url("../images/chevron-down-black.svg");
+  background-size: 14px 18px;
+  background-position: top 7px right 10px;
+  background-repeat: no-repeat;
+}
+.with-down-arrow:hover {
+  background-image: url("../images/chevron-down-orange.svg");
+  background-repeat: no-repeat;
+}
+
+.header-holder .main-menu ul li .resources-dropdown .doc-dropdown-option {
+  padding-top: 1rem;
+}
+
+.header-holder .main-menu ul li a.nav-dropdown-item {
+  display: block;
+  font-size: 1rem;
+  line-height: 1.3125rem;
+  width: 100%;
+  padding: 0.25rem 1.5rem;
+  clear: both;
+  font-weight: 400;
+  color: #979797;
+  text-align: center;
+  background-color: transparent;
+  border-bottom: 1px solid #e2e2e2;
+}
+.header-holder .main-menu ul li a.nav-dropdown-item:last-of-type {
+  border-bottom-color: transparent;
+}
+.header-holder .main-menu ul li a.nav-dropdown-item:hover {
+  background-color: #e44c2c;
+  color: white;
+}
+.header-holder .main-menu ul li a.nav-dropdown-item .dropdown-title {
+  font-size: 1.125rem;
+  color: #6c6c6d;
+  letter-spacing: 0;
+  line-height: 34px;
+}
+
+.header-holder .main-menu ul li a.nav-dropdown-item:hover .dropdown-title {
+  background-color: #e44c2c;
+  color: white;
+}
+
+/*# sourceMappingURL=theme.css.map */
diff --git a/0.16/_static/doctools.js b/0.16/_static/doctools.js
new file mode 100644
index 00000000000..c3db08d1c38
--- /dev/null
+++ b/0.16/_static/doctools.js
@@ -0,0 +1,264 @@
+/*
+ * doctools.js
+ * ~~~~~~~~~~~
+ *
+ * Base JavaScript utilities for all Sphinx HTML documentation.
+ *
+ * :copyright: Copyright 2007-2022 by the Sphinx team, see AUTHORS.
+ * :license: BSD, see LICENSE for details.
+ *
+ */
+"use strict";
+
+const _ready = (callback) => {
+  if (document.readyState !== "loading") {
+    callback();
+  } else {
+    document.addEventListener("DOMContentLoaded", callback);
+  }
+};
+
+/**
+ * highlight a given string on a node by wrapping it in
+ * span elements with the given class name.
+ */
+const _highlight = (node, addItems, text, className) => {
+  if (node.nodeType === Node.TEXT_NODE) {
+    const val = node.nodeValue;
+    const parent = node.parentNode;
+    const pos = val.toLowerCase().indexOf(text);
+    if (
+      pos >= 0 &&
+      !parent.classList.contains(className) &&
+      !parent.classList.contains("nohighlight")
+    ) {
+      let span;
+
+      const closestNode = parent.closest("body, svg, foreignObject");
+      const isInSVG = closestNode && closestNode.matches("svg");
+      if (isInSVG) {
+        span = document.createElementNS("http://www.w3.org/2000/svg", "tspan");
+      } else {
+        span = document.createElement("span");
+        span.classList.add(className);
+      }
+
+      span.appendChild(document.createTextNode(val.substr(pos, text.length)));
+      parent.insertBefore(
+        span,
+        parent.insertBefore(
+          document.createTextNode(val.substr(pos + text.length)),
+          node.nextSibling
+        )
+      );
+      node.nodeValue = val.substr(0, pos);
+
+      if (isInSVG) {
+        const rect = document.createElementNS(
+          "http://www.w3.org/2000/svg",
+          "rect"
+        );
+        const bbox = parent.getBBox();
+        rect.x.baseVal.value = bbox.x;
+        rect.y.baseVal.value = bbox.y;
+        rect.width.baseVal.value = bbox.width;
+        rect.height.baseVal.value = bbox.height;
+        rect.setAttribute("class", className);
+        addItems.push({ parent: parent, target: rect });
+      }
+    }
+  } else if (node.matches && !node.matches("button, select, textarea")) {
+    node.childNodes.forEach((el) => _highlight(el, addItems, text, className));
+  }
+};
+const _highlightText = (thisNode, text, className) => {
+  let addItems = [];
+  _highlight(thisNode, addItems, text, className);
+  addItems.forEach((obj) =>
+    obj.parent.insertAdjacentElement("beforebegin", obj.target)
+  );
+};
+
+/**
+ * Small JavaScript module for the documentation.
+ */
+const Documentation = {
+  init: () => {
+    Documentation.highlightSearchWords();
+    Documentation.initDomainIndexTable();
+    Documentation.initOnKeyListeners();
+  },
+
+  /**
+   * i18n support
+   */
+  TRANSLATIONS: {},
+  PLURAL_EXPR: (n) => (n === 1 ? 0 : 1),
+  LOCALE: "unknown",
+
+  // gettext and ngettext don't access this so that the functions
+  // can safely bound to a different name (_ = Documentation.gettext)
+  gettext: (string) => {
+    const translated = Documentation.TRANSLATIONS[string];
+    switch (typeof translated) {
+      case "undefined":
+        return string; // no translation
+      case "string":
+        return translated; // translation exists
+      default:
+        return translated[0]; // (singular, plural) translation tuple exists
+    }
+  },
+
+  ngettext: (singular, plural, n) => {
+    const translated = Documentation.TRANSLATIONS[singular];
+    if (typeof translated !== "undefined")
+      return translated[Documentation.PLURAL_EXPR(n)];
+    return n === 1 ? singular : plural;
+  },
+
+  addTranslations: (catalog) => {
+    Object.assign(Documentation.TRANSLATIONS, catalog.messages);
+    Documentation.PLURAL_EXPR = new Function(
+      "n",
+      `return (${catalog.plural_expr})`
+    );
+    Documentation.LOCALE = catalog.locale;
+  },
+
+  /**
+   * highlight the search words provided in the url in the text
+   */
+  highlightSearchWords: () => {
+    const highlight =
+      new URLSearchParams(window.location.search).get("highlight") || "";
+    const terms = highlight.toLowerCase().split(/\s+/).filter(x => x);
+    if (terms.length === 0) return; // nothing to do
+
+    // There should never be more than one element matching "div.body"
+    const divBody = document.querySelectorAll("div.body");
+    const body = divBody.length ? divBody[0] : document.querySelector("body");
+    window.setTimeout(() => {
+      terms.forEach((term) => _highlightText(body, term, "highlighted"));
+    }, 10);
+
+    const searchBox = document.getElementById("searchbox");
+    if (searchBox === null) return;
+    searchBox.appendChild(
+      document
+        .createRange()
+        .createContextualFragment(
+          '<p class="highlight-link">' +
+            '<a href="javascript:Documentation.hideSearchWords()">' +
+            Documentation.gettext("Hide Search Matches") +
+            "</a></p>"
+        )
+    );
+  },
+
+  /**
+   * helper function to hide the search marks again
+   */
+  hideSearchWords: () => {
+    document
+      .querySelectorAll("#searchbox .highlight-link")
+      .forEach((el) => el.remove());
+    document
+      .querySelectorAll("span.highlighted")
+      .forEach((el) => el.classList.remove("highlighted"));
+    const url = new URL(window.location);
+    url.searchParams.delete("highlight");
+    window.history.replaceState({}, "", url);
+  },
+
+  /**
+   * helper function to focus on search bar
+   */
+  focusSearchBar: () => {
+    document.querySelectorAll("input[name=q]")[0]?.focus();
+  },
+
+  /**
+   * Initialise the domain index toggle buttons
+   */
+  initDomainIndexTable: () => {
+    const toggler = (el) => {
+      const idNumber = el.id.substr(7);
+      const toggledRows = document.querySelectorAll(`tr.cg-${idNumber}`);
+      if (el.src.substr(-9) === "minus.png") {
+        el.src = `${el.src.substr(0, el.src.length - 9)}plus.png`;
+        toggledRows.forEach((el) => (el.style.display = "none"));
+      } else {
+        el.src = `${el.src.substr(0, el.src.length - 8)}minus.png`;
+        toggledRows.forEach((el) => (el.style.display = ""));
+      }
+    };
+
+    const togglerElements = document.querySelectorAll("img.toggler");
+    togglerElements.forEach((el) =>
+      el.addEventListener("click", (event) => toggler(event.currentTarget))
+    );
+    togglerElements.forEach((el) => (el.style.display = ""));
+    if (DOCUMENTATION_OPTIONS.COLLAPSE_INDEX) togglerElements.forEach(toggler);
+  },
+
+  initOnKeyListeners: () => {
+    // only install a listener if it is really needed
+    if (
+      !DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS &&
+      !DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS
+    )
+      return;
+
+    const blacklistedElements = new Set([
+      "TEXTAREA",
+      "INPUT",
+      "SELECT",
+      "BUTTON",
+    ]);
+    document.addEventListener("keydown", (event) => {
+      if (blacklistedElements.has(document.activeElement.tagName)) return; // bail for input elements
+      if (event.altKey || event.ctrlKey || event.metaKey) return; // bail with special keys
+
+      if (!event.shiftKey) {
+        switch (event.key) {
+          case "ArrowLeft":
+            if (!DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS) break;
+
+            const prevLink = document.querySelector('link[rel="prev"]');
+            if (prevLink && prevLink.href) {
+              window.location.href = prevLink.href;
+              event.preventDefault();
+            }
+            break;
+          case "ArrowRight":
+            if (!DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS) break;
+
+            const nextLink = document.querySelector('link[rel="next"]');
+            if (nextLink && nextLink.href) {
+              window.location.href = nextLink.href;
+              event.preventDefault();
+            }
+            break;
+          case "Escape":
+            if (!DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS) break;
+            Documentation.hideSearchWords();
+            event.preventDefault();
+        }
+      }
+
+      // some keyboard layouts may need Shift to get /
+      switch (event.key) {
+        case "/":
+          if (!DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS) break;
+          Documentation.focusSearchBar();
+          event.preventDefault();
+      }
+    });
+  },
+};
+
+// quick alias for translations
+const _ = Documentation.gettext;
+
+_ready(Documentation.init);
diff --git a/0.16/_static/documentation_options.js b/0.16/_static/documentation_options.js
new file mode 100644
index 00000000000..b938d517a79
--- /dev/null
+++ b/0.16/_static/documentation_options.js
@@ -0,0 +1,14 @@
+var DOCUMENTATION_OPTIONS = {
+    URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'),
+    VERSION: 'main',
+    LANGUAGE: 'en',
+    COLLAPSE_INDEX: false,
+    BUILDER: 'html',
+    FILE_SUFFIX: '.html',
+    LINK_SUFFIX: '.html',
+    HAS_SOURCE: true,
+    SOURCELINK_SUFFIX: '.txt',
+    NAVIGATION_WITH_KEYS: true,
+    SHOW_SEARCH_SUMMARY: true,
+    ENABLE_SEARCH_SHORTCUTS: false,
+};
\ No newline at end of file
diff --git a/0.16/_static/file.png b/0.16/_static/file.png
new file mode 100644
index 0000000000000000000000000000000000000000..a858a410e4faa62ce324d814e4b816fff83a6fb3
GIT binary patch
literal 286
zcmV+(0pb3MP)<h;3K|Lk000e1NJLTq000mG000mO1^@s6AM^iV0002xNkl<Zcmb`G
zgHi?o6ovOGdxdP*AltSE*&JruJwUGI3!FN?xxO>s`hMrGg#P~ix$^RISR_I47Y|r1
z_CyJOe}D1){SET-^Amu_i71Lt6eYfZjRyw@I6OQAIXXHD<M{a4P!N^sPbQKi=?mBx
zoos%BSoiGXjr-;%$QixXMOVNSUNp6L0a1Oz&cgu)wqE?07u5I7qrQIu4Fij)Y3c&0
z@0u_#NH6I?Mk(n;dT}d~^J<WkTLqp|RW-hV56tKpXqu)k@V{?amI+5DOlEU@funz+
kySsbM>fiX^GbOlHe=Ae4>0m)d(f|Me07*qoM6N<$f}vM^LjV8(

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-bold-italic.woff b/0.16/_static/fonts/FreightSans/freight-sans-bold-italic.woff
new file mode 100644
index 0000000000000000000000000000000000000000..e317248423c75c8291b49bf7ef10e792167a1623
GIT binary patch
literal 39560
zcmY&;W0WX8u<h8ktuwZ5<BV<Fwr$(CZQC}_*v6ah-uvgJ*It#XwW})W^h&zZag`Gl
z1poy2Eg4||$iD*6od19I|7`z%5fxRE`Q;h?T5<k^_uY4>n6QW_06+xFuP*QlC17gc
zb}@N5MF4=vGynj&2><}063BqcIx$6M!C#)*ZvtKb03hil-2QPnMLNdcI?1n}<}dXA
z)Z6<TTI<^Z0D!Fh^8A0XaNI5kV(8*T0001y_sfy|57@P+XC`*0)&KxtTECq7-x$V<
zO?0KE`i{T85iq|Ti2nnEsg=74002}Q0D!hG06_UAnOB;YnX&$F9{|gKIjsKy0im|t
z?AQ2f+xyjtenA3H18QVu?d0|w+wT|3006*cJ=a8=R<?$}Jc!F*?$j?<1U=FBto7Y~
z^8$hV`VssGAaD>W8+~hI001+L|5N|~a}wv4eO^0TM<)P)u-|(Jc=OBOV#nfmvU4#0
z^@TwA^^5sM&%qGpxvR0kZ<{mxes!?_u%Vz?2>|$80|pS3ez%a(h39?(!2hp{008qJ
z8=wEl)%l3I@B6P=2u!fuUmgU&Z`S{c+lHZ@k>1|kJ6Ev3|JO&a+)oDw3Fco6Va&LG
zDuBXq@VsBrf3eSj!T~^k<!{4&bszw=|IC0n{~7*^VWOb7G1xcJH()X4@ciQ)qE3*Z
z54NZdCPQWk21W)3XQBcICJUAV3IlTijG@+rnb)@sD^9^k@oA!O_zG_1V8?JwK~FCY
zHqM}9L<#a23Wgd0{|&D1H@0=6t~iPU#_;UU$8^;Y5@LS{8InH6Abk+cBob1YFbzpC
z!CxuNG2X{G5&6g1urM_eVL@aDglB;WJi&mJv;0t)05yt$Fz+xm;h>z$S%cx78C+LG
zWN@FXN!MP_)59#xC)Dn}LOj6um!(GbyxKfXm1#x)Kb?+aoRC9AB1O(<=uRtDXWJME
zda;t}n9UJil)-4uZ5;y|{62`G6bmYgDF28w#|WQqnoQeg;tGw#;SGjp_KObF&DBH5
zBiXOB3U|o9ZnFrDvh3X~0D64%cgAd2xc1s@31<wieGjzhEZ+k?+2$!lq&datFRezc
zobYb4;2yLl(o(zBmqzI)I<L=stX2R=jOVmgd3_|A3>?rmoiHUZ$?*~@fOI{klxL7s
zxx^bKBvv;zyrbE7jrDug>r9=1?7!xeoRQy;h&)6Huc!~BYSPTVrb(W%)aSCz{`Qet
zWWq_aEFgQ}m!f+=b}KzpM)RbSLdG)9bPSF;j6``IMXiW=;I9DDSB{aba3{0*j=5fm
z!A-Dar2;un=Jf(DP{t~J0k9oG^H8O&b3XQLzxMz#&G$)fP!Hf>)p0N%Oyzgx40>E@
zEzQtJf(=kxnMk}_dhY_T1*)Y5zwkZwG>b-AB|~0*qz-!M^@1zHHk$RB5Wb{=vT*t%
zUsF9hka@6(=Gy*4?E8XQQZ9ar<tDze%oFAj;hE<7z+Z|3L7Im`nixbGr<gq=3I3B1
zpEPbhbF?!`>#0J+a7fy;=1CgaJ&%-g2DFNQRu0X_DH?J61EeswwDEt>te&2UpB^Td
zC>R1e@DW468#N5JUlUj$P~338{OD~iLt1^nKa53OP+Cw#&_AFYproKIpx&TV!EV7U
z5t0?YK8>G0@Jso9b$$#G6$t_Y`~rdkU=<5UH3@e=KR>ZQ`0u*6yQeoly|s8azIJE*
z-{2p3mp*le{UP9!_{@Cty}oBFQY)qMs)(Q%>lk|znCX5X$Q?j=s_JU%YaHzDZSHPw
zub?2oK_bG#L!>0dMaD+QN2n;tNy<vgODxRIP0mhFkFYS&Q8LogQ#91oRn}Hlm$*3D
zS=w6LTRhy|UEW?_pTIzY0YZX<1H?py1%?KP2gpd68yj4m<g;n5Zu1GoQmM6C-4@du
zPZz9KTV2kREx4O6mph%Y`4prq4>rRC0hi>)>M*#C7!jBl3zPjdljZtq^_2`9@Iy1z
z-Kk4UoL_VMm8&+}JwArygZ;!XvH09x&sq8RC%7RgIVU?c-;l^3C9;NTjHcICA%&w*
z{uvd17yuDcVLtye2HFapvYG-2+pzG!{zGd792+Q&>2V{T`Jj+P%h3i#T*1c>=4=k}
zk@kT2{QknQ8?~5u`9<QHV1(%6UV^-YNKBSt1<YJrjgB+vOi~l^6|Vm*XBAIL;s&<C
zrdhva#hV)eUWzU*qNADu!8NQ*9w(eXzM;V~=aaZB0FHTV|G;C)aQSOAm445ELWO*M
z;y-X<G%+teE^$KsBLq3gw3iCJ!JO`xgntiM-Z!A9FG?#ppyez1``yZXYB5>Y96#R`
z$b|Sd&djf9X$_L`y+pYDV2>gdH(658nwHk^O@iDXJnp2d*gW;R(XZXmc%~OfIqeU=
zHmKqbg9isshO*E~GC>WC_-s%)1c8`{1I2N?HU3@1|0a$!(j_6ubr{kNHWI(um&YpC
z_`{tkV{MB&@T!i}@7jwg#rZlGvVnnUz-92q4^miARQW9gYqpyj?C9^uEPs;U@m;KM
zHbm3)nlN{2vojon3WSHx=GOgWM5jvoOr)TzZG}bI!k_66otd6d;2b>YK>@@5%-N4`
zN&k0l>KbM<I1nKZHjkfz=JT-1E_B`o^2OeIjsF?5&HmS3XJA9Rz!$I>zp&`2DDQae
z?X@Oi6gCK94_gFCAVw@AW41c;QDJ&3Jr%p{zN)JRf0=$~WJlse*0e#WqdE}*4r+e5
zN~okkxW6s<6Z-%TYfSk#vIL=n&;$BgeV{BUn*Gk`z^Y%qU}RU(Q#Qu@h<JIN*H-MQ
z=wrwB{`SOdVP?PpZeDDGzYPKbZUgxS1Oqq(cm?<dgaJeXL<i*hy+i?J0Vx0h0iFPP
z0oeih0T}=x0a*cs06_o|0GR-30OkNr0Vx6L0Eqxe05JiCK@tCV9n+ih9{>q~3n7nE
zmUfWVOsk=9WYDBnrN1%o86N02=s)Oh2s_5hMBhYo0e+@^ro%zPxnM75Ph?-U%h~f9
z9v)d9!5!QlB_2W^Y>ZBgUX68y&Be*Z*Tvq&_5uBN|KtM+0AYocgnWi%fLMp9M(!ZC
zBFG}@5`B$<2n~xA_a9@9hb1}{Mi!<Emk6UlIz^U?6Hjm>GbetNp%ko;;*@$1jS8J&
zStnUHbntNSJn}w@JYpN;`x~cttLT<@D^@Rc7juA1N>~hKjDAc^X+Y^x0aU?TakDhA
zRI!v&hWocxA-C{ba!}MTgDkWp#3c9xYT8o6OG8Jau?fzs^q<`s_ae=lcJYU_oQ!R7
zPnu8QH)J3{AZmzo2xN#{)KU~<R0{?b#w=q!eY)X<iG~50LCe@>mV4YAX(U3VX9{de
zYSmVidX;5W?|RZ&#(GbKrg5&`&k*6reuQDzAv`&BDO9QV4AhLPR!e(C8(RC0L#5rv
z;q_k6DDl8y=qYzCOs(?{>khWpk{8_D#w*#I%^+axtIoam-p8ce(Cl#Sn63XVyf5F6
z@pmi`jvqZW0XT#&ND7E7qysc3>L<++HUR2wHnrQvR5!F2%JLf<uPLqP?(_~@=7&k5
zRf0^zxJ`nw5Wzki;DEA?19*Ex7*s@<_#ctHFgMW1hzO<fW~NL~nqjQw70Tw%IjClf
zWR0e~8hNNR>77#!rcMchGH>56-*39O_4V-v$C(H2)ArmOdTORL{k69NRG3k-bf|Fp
z{P(t$7hk*-=7G*9`<P%w(SzFtztlqw839l4ti7kOX;Z4lb+5p6?smV+By>nst=D0_
zZ)dQc1oJcQ@oO|NCXbORC3<_@K@jAuocad=A`L+zM{dHjbZTCT;`lrgdREUZJ?2pZ
zhLQ$K4Q*_f&a8~j0%;;USXI6zw>u%#pI3{9%Yo<lMibW9Y>r5)W0la&o52s5Bf(Gk
z{Gb@VNJRJ7bMb&POdfocV0uJnLJ7Hm+iEzXIEYUelEQF>o;Pi``AN%mg!Va-*kW)T
zSJMK=_&SYSrgUjqrlfF7NaUm_^LU84LYao+pg|(+1TCGmzJ5?K6?Pm`4xzf@_~|8Z
z5%?_zI+C}&m;@EW#>i(i<K&h7SkZ!StX$%OeYv6IRnaEinVVbtq0`%F&Ne^qhmlnA
zW#rQw0w4A%o|cN_<Xvekc8cCqg7u+Z0Uuj(^`GT+x^d<E+1t~}M&0p?F%NfKuGlsG
zEv?hgY-)U@9g~POgvHSI@bFW>;bO^r8XX97dnV|ciFQ_^5y67}HWd1%aKK6>*r8e0
zCJE<TEc&%e8Z93IeSdN?l}#is9gjMs*~TjxbJkc~S<G@qc(UQ4A%{&SoSS>2#03hU
zA*<BO8H_2GJv&bvRIj{_KICU%VvW{NDkwXStjY0@2Vf1~rZD#H9J6>n-NUdTOUsH#
zAgkx5%@#VSSWMjua^Vj?;K~piY|qW)MU$Fz9=-3QxX*td$N6`grF^m{A{S)o4WTkK
zGQ9=@xjl_+2(t|kcb$`1A=5S9HPOr6ZVhf>Tf7C*-(;SBNy$Q*yvtq`st&7Y!j3a_
zfm~_%Uy@i4M~yJV^w6BUOI65}{3Z*2vw&m+-2FrT5d!t}2@v;D;qU=LO#Nu!^BMt=
zblnMM6UJefI!Uf@Ox~w?H(_eHNN<ObmHzeqTk6#bnIOnu$Jzaq#{HcJ$J+JE6@IR7
z6I`s+%K-njRWh{?_&<vy4A_Z->)?ti2(nI0`$cMjXJEJ&R)HC(lvh2@fbLGwMY3&z
zqxj_=(l4Nc`q<ESY&>m$fU9r54OXFTW3KtUzQf*6-Y$!JzYhFYUY>t0gIh&cW7{m0
zb`>5RgWsWh4|Cp@M~~t+4|BPDtjEr(RbermT|*__!!FlReynbu`wLrV@VU?Nk~l_|
z-2bjYSdxbRF{wy@lIDgsPs5A9e`95UfsO@tuY?TGI1EPuR_p`UbdX~1$Q$yzWgz^6
zt|M9X7h+7Gm7XIH+QJ`=w4HvPnidUNs0FRNx{N=*7A=%(TS|nU51M!HZUB6s#S|lY
zU0Yu1FIcZXsg5>wH37>qXuKVqJqHcQ5ZyU39HAZ|f1+SsIF|;9Ee=887bKK!{K(^n
zr>Gp!Q7~z*4v*!<nmTa~+2HZC?f5Kp=F@Wl`3~9laG`Ct^EeOk2{%iT`QsA9)8I+`
zz;y4Es!e9w=^&x&Ojd(&P>;>Ix_i&3{2^b;*5?%VV2NETFE4{FW@R}pE@c)AHDkh|
z34M_Hf#4;ulH4%^y}PmqVntmHxf)Yp@Ei0WnPuS&NUGhc3i8*VRoAQ|8x&ban~{Em
z`M@73#{LNz)o-chZa(ZMHFit5i+L5h%cnsVp2wpm??$}crLC@icj_M{Ja2Oq`bS6F
zLr<;fElN3Mn~p_7-zA4*IUUH!nYypo_JDj6r);?7>y38s=-r`QBv7Np%x$5eJAoIo
z>Qrm|%%MgA_Uh1;%~M`pi)jp*c)9H_9fo(sznQvxGd`_$Lt~u~q(<XIKXS@^dlKV^
zfn6tazpsL)lVNrO(}id3GJ~z!gyO;<>c)hvNo+u9uvY_b5#}K`P@BnZe=TNCP^Qp@
z!3tta`|}=9?Sj=pb|6iOV|qH@k8dB=$a3ELB!Lx2FVq)D&v8%GZ1uH<ySB%z?tfb^
zC!4I)cl6_TVdCQlAlrN)4Nqkq=omyzV31Gudir260)&HD>(K2(@@P$AJTM|4SxBkp
zgLeAd^y=^T+jjhVS1h0U+XP}u!9?+srl8;p^q{SI;qj!Hjfm#s?y%k<-+<J=Z5f4p
zE_d$@jlhZ@Y~1-cq)V7&=Xiyk6gV~gX7ZRxA_1UQ!c62hbA=$U<cQkJXtG><cFrK@
z_kkBu<kGr(W|NYs&Si^&!I@7(!H=WTo4t4ZSQ=}~3cZZp6E4<O*NzD+-zY4R=9~86
zp)!a1EDk-yHKS*jQ@{SS_jKd_*qWS5+#{reJeKQINX=Pbzv;a=N@B%nM5osWP%7yt
zQTpA*kR;2WEQ~9n(WP_$JKsh|*K!^n3Xa4e@+GD_7=ecGU&G+J>-V%FVrb?grCV6Z
z@r1o}FMB{)rDacdnwtzvZ$qwDBilt=EdZPp3;ru@ybC=sXqkNPXVeBGxyLdK9noJd
zvYnm_SsL|6G+EFW?-3B11m*J$)heF}>3ViW>EnS$YYaQR;r{fYX5CdenMR)NDZ`VD
zX46$U&FBWBfjOhSdp3#}kjmY7fr*a0-CTOHq%@WJYQy#C>ft?R?RL^U_uAtn7%-Q1
z!(9)2?h=S%usQ&YdeSY<e92#nWTt^)2CE4g$<!>;k$VDqG@#2&LcVI6eaZKf?E_PF
zP7l$&lA?*RKwe-^%xa1tYUqAxPDRByLC_9~_I1tjnC{51*6||T<|-I^OTVoa;Dv|q
zsG*ffnxwbI|8Z45xE6M)0{NsF>y{}X`_?r_fd1c;kT2KG%J&zyYV_+}i%;k-mgh@G
zNkROgZD%vbr7@Gk*pH^~+D*s6v)9$#fX>O>S5Ws+)5S64kB7IhCs*&$Blu_cYthG~
zF<fev&gV@+qYYi^=Ogz&mop!ad8zZ}i=tFlZC`d5ncnaEk5O7&=KONEwKI+%Ek*9<
z=Rbiq>ORW{|GYn4)K#oNHr*4rh>Qfc>XO1!3&Hl7t?#6l0BmG52NG0r%)w;FTbqkg
zP0#g`gZYE;QziLH+3#aEN$@FxB$<h{+7GHVtn3SD{LciX8>K`5-DFzBW75uAB74%(
z%#pk6=CB~zq|Z?O;wM){Eeo5sRgl^1^lnjV2Xbd)zNQCR7DH2_b5+nY<3DCAYqGYR
zOkAh;cZXE8*veb%*DLz@d|W2St?{KlE%ap_wbi|I(#>MqEh<HyFBU)lQH#CpRf4nD
zVk(*=UiW%$Gbv(V(H#$Hd`I;5VG!ta$jx^Z&=O&l*)!f-BiRPX4~k#KA|{261)CET
z^X1I_nNiHBP~PDhq|gwtMV+n=C~J(+vV^WZ#OX()!P`ft0qe%x$+#V3C}<DB*>kJz
zH>%0=<p-7E<e7^B)wT#~<0mPTC>$CUY4U+E0OA#+os=b?ylJa6GL`1lUX>MmkhrB7
zbt>|u?iC5RsV^a++dXyvhItJ$ME#LJ=wHARI^*E#FWV6Mi<MELH6b{6ZeNRbde$6J
zFjbhtpE4j!Oh11qLdJPn5X?v&Do@@TK#+;UTP;^f!3!1zzsdd{d;&^Uy36=pQocUh
zcmr-UofiVx$r_W)=j%KE@6NO`WHysUPte6g_qpa-4#==&+D$tJ8%1Y`bN5sh(q&{2
zh2Al}jo3VpnZjSAherRNgPFjej9Dyr38h5wM@70-l0z?i&YUt}HOnc?QU;DEnLD!#
z(BgfLv6#WG-zf1BY+0}#Q~|!$`A-QlDa@79MBu80!E$T2xP^{Y9QdczB5v!C!?CAk
z?bmXG#zDr4*YUu(%w#C%c%#(r>jm9vCold&Ip)V!Aoh2U^-%~p*i`%e^*S=yY7_wP
z<Av2J@F!ao*ry!nia$G=pc;{dP(lnX8rm849L6B}leq++i|yzl>6U}-C`yiH?`?l;
zZr-fz)n^<VB-x%RCbCfO?j%JBhYiR%NkLMja8;ow7SE?vrCLnnU+7c2N(T=}V7aK2
zIFsPp&7i~_DQ|zX<8~(;gZwK$wp}%>5dcL8l|{Xb_<}dq4?s~%=adlTTzs6}f(#3%
z#JDFQq&QF@2MM472hmu3977ho1{GP2IlI>uz8YNDpA*P?bg>cZ4X$SU?f3St@S6?$
z!&H>mGAf&l?x|fqjXKATTf|5Sv)krrmKJ09T?e~{Uh~DKS<&6sy{+xW3LIr(l^u4Q
zfhBuVAqxj<j@1OY2rMec#St4^R(*dWu|clglNq4uE`;PjoOF3FC<5mD0BkTiL5le5
z1S$k1=TT%#C;`T*G=A|llmrxD2}UP+Ph087^@D)$Ve0TCSf7F+gES?*L?_myLNfto
zKO0qa2I_zne<gY4bQ5;-?dfl`(UMYIH`j}Z(o|8VcbQ(H&-2n;8QgZgB+rdCD)b)i
zPtEBwbo`Cm?~{1!oGiZkozR=l@d5?&t<P{KOR=a7$3Yikot~M*s>IFlv;Ii@?c_~{
zh2t#xO5Osd^kcwNd1j)->WD&!?`|U_yH`$(9zDzbCX)>9vz+&OMLwi$@!f^eRBG?@
zbIVkjj4rpkBZ)mJ+BA0kc_Np-9ueqVH9rpQ+tPn-Sp><5qEm3B3(E|CD)#u`y}a?|
zTUgyzPM9<{YH1^nR;2q*;icQo4^m#&(dYCYcirvqONjY9nKNf?xO>IsnApgsknaK7
z<z-k0NP9O<HKMEuzVXdsW3Z=kZ95$RO}LX*$4FYR)27B0J6v+;aIq6)PpGg%;FC%h
z*p_=ESR1a653pAUE3>?YFTPMOIj}E1bndqKr`?4t!MCYxCW(y={*}wxZj;UjU7w4@
zW2?t{uB-f3CPso&&e}V@xLv-(uLj>`2{l&H?bX9Y`cD1A@klu4RSCEd&{R4V0jorb
z(?0UE#R79WF@RkoWXA9Qn=YM3G=ZyHnQ8EUFHij7|Fs`8Q1)&Mg2Nn_(}*r#mnj2z
zz_vAFr80#EBBWC5a++Kgh!~Di!&dhg&$sm(PSogd*O@Oz`@YU!UtKI{m%y0zoOFjM
zj7Je>vgr%xsJNRGWTl8m?=ZyD?cUl+2xP2RG<)K%2XN8DpYSR27P~8z>7S?TPhm%4
zK$BLOdw>fz2^SjFEXc<^Bv%RvhD#q?$f{Apl7g_HGiO2hjP+scsS;9!oQ3;9CBLf|
z3z9N{^d^dyp?xGTYR(x%a-gCyv?^m6@zB9p86UI*gZtBv(Qr&a!om~{_^^I31yC_n
zH*sPWzQvtmP=r#j&^I%bGvrcA+K18Ti_;kiihoLXrpYK5cefBAoou^^h@ky3kI%61
zB!SUyCM&1?dg}s93dOCqTms&rkdIPW^5--<d3TjC`k56UxpS6j<z{x@$6b%dhgEwR
zTOX8R3ipnu;`Es(Ugi6_fqj|O?Q#10)b;i}t>(MCZhN4+-$7q@FTB7y2VaKGbb@_5
zm-y>=PyTRl-;4J?9xgLTCU>^#cILktUPH2Ge^RmjW48UdE~IGtUHJLWdy^}Da1}%>
zG&bR_mQ}8nNX9u<9(U;tVDQ#f0H6powp_Z|48d7{gEoI&-$r&JuMJBm0kRrf0M!D3
z@|EFuLjNRLvoD8;OD6>$ee4=MEElE<G0hg!Ai38d`v_RgU!ZLivMn@l97t}@G=iLf
z5pn!a)BLL}31T_FnUs++ms*>C$~k-ye(IFoiPzF)dN~#U+vhhX&oPrCA(~QHl(GZX
zOeCmEcwgk=M#3>QKrS_Q>=s}Mdt(E1#}-3lY)o!^LzbOEO!zvh8x`7`2#qx5!L34Y
zt%8=)48FWqC_(s&=)FZnAc<Rym$&0&zyz+PDJ^A2*Ilw^u1)t>M)G1Vt@;JYR6Y+7
za*oS%p7KY;m#d6*m&ZxpfTPeT4E8;_%Pi*{`68zr6el72r5YdEwScGXYUNIx0=(i(
z9wp)Q5}7ek`4t{Uek7+=)J}L$><H~SisrO`nLU9h8U`8apJ=|=^cCN2f7=`0@6@&W
zuT0(auY;Yd4~t^HiQ37RhGVv`tJcjCmT~gY8jrh^(jonO)#)Pt4T8OBdMLNVk)lFt
z8AT4m;|})3n9r9G;Zu7FaKlOYcDv8VO$pi}MOroM(|LR1mIp29IzDYu4&q`%4VA*k
za*z`Nw)QLy=*bTb7S}8AhHKA|qi;s4kCxpD{IC0SG2O41{BA0clM~zAleY^rY7Vat
zi~-;8)eqMOezBxdVtNzeMzwe~3HU*g_++;l8bS&Ks=YcH!l_b<@6UfRM{e$%I^ADg
zV}ow{pxo{Pvw=%zw_$$=p*g)JEigFk;4VSFAaf3L%Ff%^gpb0eLe@kO{p0Dfk~itb
z(95O=?r?+d{`Ch?P)Xt|gqB)Moq=$2XQPFf=HaXYD@UJG>~emY8Yij5HyNDuw?1QN
zK*wAb=+F1kmvs{3%88AQBCc~Gi#mz&lK)7}h+soGr~)~p&N2tG8lgZZC3Fgln$UXe
zV~W%9;l|;DDftndiQ+SIkr|ul;f>)5>7zV#LlH$vKlxF!$`A7|Z$AQJ=UlaY4ED0%
z@w8o*8+((<e7zmb3%sepa@h}3mFmR(d`hv_o!rdYVgm<o_^efboH%@bDb_yXkXnsS
zolFnqE&N@Ra{o#aoNQAujS@!j7*Q7hK82j|BLSt9%Bqn$8f2~)jghJ<uO1#ll1|Pv
ztJs2|t3qXRNdWUHU52>jLa=DM&`5Lot0J4l8?ME^^n-8em4qv532kabc(SNY2!i5=
z>%o2C46Jji!LDQ57DjgtKlU8AezYac_?YEomSdF$JyjoP%P-P0$vuN8=;#7TmY+t*
z!%{aUYG$ID4Yle`i36i54*LR(Ln_u`qXWnG#dfYA;?x86JvMs#rE)8FfR+J1*E0OX
zTyHWsr^?uux^8}Jpb#M39UW5PPcd?Dh0+tZ(RNnxWLX)KQjGl^NBAMt`#rc!ZaTU8
z4hB<wU~t9t^SLii_UC87q%Ajmy{1MkTjDrv2nsPRk%NU?@r5H@4+G6v_9p9FL%+YX
zbhbZSff*25n*SV)81K+-YBQmAxYi$WR2Yt#(-eeCE_P|4agS4?3ozSjpiC6{4>BR|
zRN8wyU0_OGRd&_&n)J$q=Fx7WDzx-yrB#!m<)|d<!d(X=`gJRiN~h9JUYIk3SYnnq
zCFWb|qHigZP!-W4)Bu@7b=Hs_4?1<kP^1+n#)M1fC+GqUh=8M{p1(<`GRLZ!48>{M
zX6OgxxT7R@u^Avq|E46C5c)c+H&e5eui74RLH(2bMIH2#0TdB*4dN<gdKBmQ8K*5G
zY;`D><PRxjI9#>zQZv+EK!hOGXp<dhAR=i74hc1RnXM921?a(F5j|vg0{YpxR(c+_
zPmtl9*Qg+Jf%^%RdrybUw9k8Fhr-XXqC%ie-uTTQn!)fZ$9AzzF$MR3+{9h{Psg0e
zBUs&>n&I;93Qr``>bA*lRZiF)N+-RuR{wa7d1emefjv*)A1PyT5QO_FE%g5O@i>7(
zmP$}6gs&Px6c;1H*o6`VLZE*}Xo;O{%boYc-=yGs_@!<hZmmI^wwYhqTd<Y7_R!&=
zc|~R48TxD|EL8x(ae^zG1L8IgZjo_Tnaz;`x6$u4OQx%ZG(k|LaoALeGuG4AB6Y*l
zM~x4Gnye{wIxN<Dhr2ynu|Gw+&rW{)bnP!GK8QagzR2vdRC@e)6R+xW5TwHLh;!H0
z;JDt5Puq_1`5JF^`}R7X=8wB_bNur+HEdk5QL*M}&LkTduKLRo{U@Q}fwnGNlqnD|
z4N+0TMr1}>X~s(2iK$v&p@6EA(Y%y@N~^k#Ub~Ka+6{!un7Q*Wp5J8EiE!ChsZC>w
zBZ)Va8Z7va0JV(k_e0KUh<4}43VwB$NqL{wSLE=WOrg3vcBjGDBZFhy$vvE;Be`2V
za(WxAc}`A|J6@_ZQUCUaaH{fzig^rXkrg_KcmOIFv2dVFLlNQAqICpxn(0L_xQ0LR
zcR@gom0suCZEC8xUF_xjx8V0dJ*~>r8Vl1vhzl^e3L*J7*-g%K8FiwG4>kk7<h|j2
z3wdG>`VTK1lgVxOwX_H0i8#900*f_dVz9#y356_%^ehYxL*Qbi1IfSGueZl}8yp7B
z;l~xUL=IebgMMPWAtQ-*qlCq%fYKvay?w?6(B|TqvLdHWd&$D!CxEVqg|(df^TAfg
zhfHm;BkvE_C|zf{?aO-ON)xN&W40IGPnGgZLi<GSXm3B#nTC^$le@N_MlQK~HbHK!
zDQdj~k*EXk;j1y0TDTHrB&5X{sdV~kX_5Rlf!ve>Nr^%cZUsxOKJ3e#ETX7^PnSPD
z=<B2Ce_AFiEeXwS_`YXi;FyH_;8+UPQOfqra5G=*P)YCubXvu%3n<2}JI{0vYt&gC
z#TFBW6%yW7ucS|lI@W)@Jf4$Z*mT)SCBH9v*~WFF$k)9?ScWLEFC5-w)p_pQTQ36d
z`39n+my_#zIAD=$F};qK2rM41fsyC=56j5NYWkB&%g|!yoBO>6n?gmGa8g0ivUHBu
z=ND;`O!FL?tIG=A=+U}Z&8Zn7pRRCTmT!L-s$sgxP@R(=0zHO}t1a}1>TLKP+}H6s
zvTMTw7fJzsO}J<<+ZJJY_l~k9et=-8oKZa}FYtJ1i7rUG{6*l)Qe#EVvS%hf)k{EA
z)OD5OGMBO+>{4H`P^bW73sf2vhU#Q<eNaU#7Q}KU3y*1Th{PRP<%&YAZty&A`k<V<
z@2-Hi9BZR9l}*0Xn#W7BKzyii-*1iUxF1>8)hfMM=Okv-?|&=FS+!ql?T$BtI4|^F
zb|@!h`jn^}BldTGi9r&oS=j!@d4cNkT(Rr=qC$z+cCJmrL#?6G8UyEJ12I}}V0G<?
z*H5<Y!9JrHqNl~2_Poe!<vpB2-p<kI<z;O0G**`Lt+5J%XH;`ul7jE<fa@cEGJ=A)
zf)nscBZdQaX@3v*!x%~2ctlIg$7mbs@Y97)oGw$51CRUf2{CrTY-qmylDFo2vSJz^
z<R{>1|FX2oE+Y$)V4lz;O_uS(51o2=5cBMqQ)trt_*2Wqg&P@<GBAqoyd`vHZuiwB
zn<j|OuM>hFlLh15ZVxcF#m1zM4>h~t%M?X!d1z?!`wr$oP2$VlT%2%I^Z>^JRa5i)
zb)PevA41Y<ifREj<!?JIdV^$4ow``#lkx!sMo$^L!k<_#f<5(yax&t2f~8b~EeCxo
zsR1k%mOlfhPe{pMMuh+<hv@q(#0WLhe@`dXav5ELxS=q&HC$cF+H(lszj{`@$IQA>
zapz=HTZ6e^m2M0I+6_7OE-->LKS@4`Jp>9oM)6++78w`^DVjdL@1Iq9UcI+!5X-5d
zvF)sT$$jn4JQAsjT{$gfS_(b=-SF1XvjDV}M3<(2gtgXJIG3$p+ST~JA}FGW^{12q
zRzBc~Re(O_nHAKVb{Kaak*9oTMcg!bv)s>jtnK%vxp#VAKJsdk49Jo)=IK|LF-v+w
zfSn25^GZANsyi(J^P#3sUB9&D48cJwY0dUTK5Yz)b#R3S-}m$uCyxWO>dqh6XCbc;
z=gNT7s~W_=wRDa1NNE0<Oz#>J8|QTY^1S)e(AA<isGXNBhbf&!AcvVBoin!rRh%n`
z{nXA{_n85WPP_Ltd#~`EhM@ZH(<Xm@Qr?5_tx68I^+lcD@*Red%2trg;+%8v-Pa8|
zFPEe~IkdeNX@<Hnv>`88=~TYFprLNfflvs6;*8&MM%BMv)nu?CH4&ufs}|93PWr62
zftzN5D$@e2QQ66?R%rxYyaj|}Z%s2tNJXY$ZJ)@4%y-%%6$3-c0^P4f6EvkJ0Wnum
z_Sev^1Q9l57O#4w77_*Y6RUY1YQTI)iQ%Va7<Nz)F=v)ex6NAa<d*nB^~*v(yE$St
zV<I#ab>@80O3&8wtmH#N+0e+#(BJDI@;x~5Tv}u1qAsKnSg2<PHYv9A`gtC|b3qG?
z%T<zqW7zQ8^UPmf`>PS_Jd&M@@5mHt69f9txTPW^E?k_IdX=!7J0Ky?f%0F(l+xc~
zZJ7$SzKghpOM>!h8ZE35a~<xH{WT_8DL;@>cXJW}IC^c{r4U%`3P~yH$2BwleZt@@
z4SBuKH)cHLNO#!;Q81bcO`ek;zOh=bhJ#U9;u4*1W0%FTc|3h7(NS>a`dAGvWAocl
zA)U9otqVg6q4}AY_S!b&DDVecTG99b_SnV6!^Z{G7pSI8^y76XUSgHjrxy0YW@3DI
zk!HV42-;w=O`0Fk#m;$LVYkQbZpW6|Oef1+%~YM{8n+L{n=$+ZF<;TP4^v-;s|uPy
zbX~<K`$JX!8;Vz9TPP|5vRMP4@gdlX<9z#|L{+ilV+JOHAuCRL>^UB>;*Z@~Bl!+&
zN?~o#7SZ!<gz0~M-VV8;tu$DHXvD@IPK$-$vG#2X(bzTwjxew$f`u#{JWz5u^JdN-
zLig4YbULLQ;(@Y^Q#%UcUh_(Pc?nA`ac9}Pymbd1;YAAHqTk)^H!{EDkqncw&g?Z+
zR6-@Hr7OnDBS8H!Jhndty(y-=?EN^ib@7oP7CO_OJ3%lP#0CCo&(uEU&}=%MP$a^-
zn{tbh)y0u^TLv1yEr~{4B~ysHtydLoFfKRISHiY)@VM8pv~l238IATq%?^$&3Q&Mr
zBHKM`g+2XA$qo8v5sFgG-UhTtp5_hEwmsDMOT~S(uvORjBNq5+R`x@S7u2P5WdQ@b
zCyKisT8sZ*eI~l)g2^=Yy>ZD86=Nn1+$wCF?D&n{Xb|pRp^S>-ZVLj4pIVP<!o^-b
zM~svUF#9l^lCTT)h9P<;u8_z|xM6F5`PmR2483CkBJG7C^LI&j8yFfoB<eVGfXqCv
zSo?KY>DjO;o^G?(!WgK|C>z$>h0OhA=K5}rMo-i6+q|iP<w?#3#xNPi#!Hvz??<5j
zv$3?U2?n61C1?S82!(U{f_%YWj!6hotR;<$+c+~$+GXqwjfo2x?d>wPw3^n4_A}Ff
zLIR5<uw{=*<~V?hqy;MlwMVt$I{%i|isMWsWF0$-8e#lOwMEL~(rk2S4NX7D7y>Pe
z@+(DP1Ds()99oxfdz_kl=GBoj;J5FL4YOP*<xVGGG2D0uC0d-6X%cUUF;zKK=e3kE
zifSv208ocn{=ByBV3gUirgh{fvQ~%-Ap_vgRO(kcp-URF)Y0(NDpoy&3g_w#`bwOS
zM6)%;Ge`>3HUMSBcHt}caVo##f)FxEf$qwFx5MmK=B3>hqsJg;+pUgM3#3K9o7c;y
zeO2}bvMQLa_k)-iu_M_xse{go?P$Z(&6B8Em?2qKcuZdFr8s@29<1h>eXIrBGY8Z0
z6nm-z$1(XE)tZR{o>A@%Ez_Lc{PEv^00lhAt=wE@TX_+BYyHK==8u8v@Ytht>706(
ze}`Wl{AyV+oD!Il0~s^2(p(2NN{tRTgjthd$g}#mN)2$LLu%Ics~1O6@fIkQ{9^rh
z`ij;<CybGw7y2APXECK%vXEhefn)92b*w1Z1K0B#kS>MqqX#4tJ5r)qCGzIGV7K9M
zrv*?QBtey+C-cxWQJS+tF~<BX@qG0Lao;nKcWG9tb9n8W>Hjny4vXFB{KJ^551+2G
zt4>?5$9;+my*^$8><I~OZ=oWNTsECYl018i0!HV@V$Cub8G@C?HAU%I0RTs-FL7wR
zdCr4?tzZ`=*&{$!2}h2?0oz4m0aYHWXBkJb={Y69v~#=u{F8V(<(c8h*~l}o%VzVL
z^yhF?M&;a<VO?_c+{8;5fGfao3YwI0iQyzvc=k@16}Rw6&KT-D={$9*RT7RKQ#9&9
z14IY@a&E3L1<%<H1^4$2crdIT*Q0695Q|6S-@It7%kQiz)p4f)48;y+6+S3-%kv4m
zoQWocY#XC*x0F`0$mIq|9XD<z(YSGm%JUO<^8tVwY8Cr;b@1B)X?e*c4(zYl{EAvw
zN>P~l<HT0*dJ(~2>()fTEJC(YdN!u&qxDjWaKSzI`|t(No8ufnb%v?5^mLVvJ#RYh
zCS&H0)aSI5Ai<~cbIx_Ri=iQ_vz*-#WkFSON`cIY2_L(WEg=Z$iH6tQ#ZZRkuWk1$
z_WmSRG8%P{f(=8$<V7%hRfml%UBriOAdC{_d$)lJ&)t(!zqHm~an=c7V^|JK)#4RS
zF7%VW{c9ZMfq#4DLNcW6>e}+eKQ{?_A#j0nos+u#uyj_c*ZJ!(ti?$~NuDH^aJoL`
z{^SL(N!0Q(P<|SB0kgP8FEqiRN*2lY0@bf|kCw?eW?=hEuwWuiqD9P(nP$*Z#^-xG
zyk8Qr{mGO`-VcwJ6USbaXSTsNh>aR|gFi}wRH4dFEfFr2W|lH0m(jPnnr(lp?p~{C
z)_bV;#PRU%JRo&q2kXHt4{klVC&36;kY~UmT1comrR;J6W1iZCNm+VPFwpzRNF)@+
zJ8us>=t3Y9wyR5^nE|JAi~fKv3er6J@>HN#oim>UT54gbvZ1CD2sL2g+i19XnjAW+
zexAV%YP=<bmh^a!#u&U|AE%;W+N802(Yy%jDgM|qk+aaWpLR@4#9UQ*hjA>JvRjgI
z8jjkm=%n^Ixs9INJOjL*8hnd2gOfEvU-^efdI&c-mNVB+@;la;6vt{c5w%hyO2BV1
zhA65ZMz$aSZF^c>)TQ)d&=irX)MygF8wT~V-v_ckV1OHyx(u2>P!f-c99mBfb(klE
z8#hvZ5XACqkc}s?lq2>>!FfJP!?$MJA^EY%*FTObX9MbYfzPzRiO_>Zs+{Sz(jS#w
z_Qi*2`aQ0e_QN)B!TVaVq*P+qlwGX!;E!RhvQb$a$P&z9u8L)+N<R<NgnLoZVf11U
zWTM5uv0iP7n00>V+T;F^8-D#Pds2URul|?K`g35jggLg0Rr*=`r8HY{GxjpU_4}fo
zV=~4j*Kyjpvs~LvJ!fDx=uTbVt5^TP1SeMr_9V_(2aqP&ESdz|LZoIMR+YJLQibDr
zKm8I9r_n`MH?O^FERETtY2J|Rce|z-JF!pNluSL@Wb%^n-#8mo=h-t&G$v%AW-~=0
zdx8N<<POT|Sp>ib5-g@cDdJu>x(iLG`ND`11ce;p2-t}M#B0#!)7O(j6i*rqaJ(yj
zCM-YpR&&akD0UQ6oeaiWXgaMny0vr?>pLx3g~oQmIf4rn?tq45=4#p2!m}P*+)(f+
zdzLG4IJ8nh|8WUprLrsv(OEo%i5-<lHqDIXg85#;wfx`hE*eZf1Bei8ERX2Enht+v
zA%N#j^;F0{qIu|x@$|?8x{jT8RHc&E3fBIG<`wf$Ni5RP;!MxPzfBchYAA2)F7F~D
zywnuPCcjgQTevY3IeK;bWCLcplC{FBb@>7-uv#JbcwW?8HSp*oHn-U&jbd%+Q@gXc
zqdYHaTg_oyf^Gs~%ogxNp7!Wl(+<Gau%D<yWwpr3(hPd_hIObiQEmE$5(K8ETO}91
z_or4A0!H|(D#l(*GeRR%s>e?nP9KuHYPCOw4>s&zk_NEC#_d`N)en15gZ|nP&!#MI
z?<0O9`1im02DQVfQ5(Leh2>nBPgNs+2>~DaLZiid>+p>GMk?SQEnnH;5bPE{#Xv9f
zkhj}UmyyX&*{gyUGqcG60>6Rz;`@ZP*Es*|jI)H%Xq%10nqYax^lHqnC#MXSJxXc=
zk4l2O^iu&h&0!}?=*$_8fB}(c`p|c}w|5%K$y-P|A0oI>8)+j$Y3NG17s`{E<2p;S
z?M){8yg)fgmDax&M3Qyqh1jkv+U<!28Run%>cpbr%F7p6Z<%C5xgC1i`>?Gx;iWP`
zaV(F|<&F2Xk=e(9tExS@TPkb0)CyKhdUJ!0?WX;rxQd1sO%T@L13c#TJ4Pj>2y7?1
zq<HLbw6$nLiX+lb_z<jUqCxa7K>GBCtWD01Dixh`g$ND@kGn}49gvm;Xv-$W>Y7pJ
zWKLms7@>x2jF<us^#St+%F#)PTWd=Js}9W5B<XIH&rwCLW$7tqs{cUbKuEV4*P0R@
zHTAUIkH=VwZW`}x9Y1(DuF3h_>6WHuE?*1!TdcS{TfD9GX8Y1L<9TJW8ToL}3iZ9l
zHy|aJWs^dlA|pv<it@uBi3>S*;(e!iHtQ3$mw1yFJS0pzOG3Cnd~Bt+)%Bz1Y7^V8
zn~{cZu2eqfHZSc}=;uz9JYuDr0UClH=Ml;OK>F~5jXU3GCf_r>(X<th#oWBIa20ve
zd&-PzPh^bn4wI#c&Ax-b#k_;>pLyf(607s$dBwtLfAM0{?Y{PSnN61-p|~shgrZC+
zH-y5Zz*D`1O?<tYw1ICcmXkMe#$Z#qQ8O2@Ds3d<G9ob?M6-(kG9zp}@i&bXsjeok
zq9rYzC|{EUi^fA>k~S)q^wybIc6}(>c<2|Ue+Z3gxCnVZTzhBk_0R?Gc!OU{S>kwT
z9L|-g)5$J&kL2GJkN9iCcy08&RmtTj4M*R*erDrTc%&E~n{p+5_&GT+oucFzP^N1~
zsUZyXXigm~g^&QvxN1=7t0;0R6V`vSmm&Uv?ryi^uEbFmPXTq3@==@m+sL^jR{CL1
zUvCIMD&=@*Se$&zfRr-O_5$Fg^=DhE6#=bJ58BY6SZ0C}l$8H6fjsc~Fs`Wv!$ty(
zuwRD3iP&Do*_Zdc;16;YPiGmBh|GBf0)Kl!%Cb3Y4UQ7K#WU{aDYu8F)9y(0M?otO
z=f{k-qbMp|hdr?QZXP~;!RO|`d8&7`(#$p$#h;N+cvI@n@|E&l!~JAibXYty_9-kL
z{~lH0(%e7zyi;X!XQIE`DkPU7ZWNgkDyZxlX$rat!=Rb7F2wZ{&}Y0-14lsCfk`c{
z4E*yTg>x(T310k~VifzA1=S~QW^hs>{7qSQT8cr!i|l4!pcf`M9aAe1+dj#dF{$#*
zK}dX`JEaKCMG%(H9vUK*AO$Kz!hAyrr2N9U8lA3dBuqg_1&g#D4DK}xO{^M1*2oF9
zWMcB-8Ic8&|E%_#+hae=W)Jz^#t*%H9>=n{uG4rr^!6sAv8F??!GZEyw>dtl^!N4(
zT5Gl3LU+o){taN_pRhf%H9M3%8WbJ09utF|u~{+0Ve9ySjpN;7s1L&>vTTt&gOX*p
z2Ow7GS@%I*RI$@!OTgR2&r|?IhDOqIs-g#W26vBp^Me`V$4RWO#CY$d9k3K1qWM}%
zM?coP1awXpAR4#AuW8e0wz#NF5BLu*YJ;kLA4h1Ks19P=+8|1`T<xf1P*c;`b9XhC
zb(smnop^A<*HZ;qMd$-m!nckd%s@xf>KE&QfPv})W9z}>74)0yCttzgg^r4MJZ*T+
z_|{n+Ce$0IP-8N=gePhq*zOMDV{kdHq@AQQmG7=ox88w$6K;U=s8sV8`9sd~H%q5_
z>R6DY1SjTYQ<W`rVU9g9HxtC0O3|YwW_u8FgG7v|sVW2*h?qqJWm%{JYxEWdpm6fQ
zL9ccNn&Cq#M|T+BPx(t_VCBqab^dwT;W^GR-#oG1l&UDf#qxUU^SVEy9k^b?`6_Ka
zD8y174L%$FT6Y{$xJ&(NU5<w)VKwA9zT(nzH58<6%x#Yj=L>ju(-qzTwYWgkm<2Ae
z>BG80%crrQ+z?|;3p{uhKR_$q_Z4e1UL?l08{DU(4<_5elHDqPe6od>lIzd9LGqiq
zxbcj7@0lXz_Z!UtDc><hei<Foyj-{jC)C)f11-Ble(dx6YG&zI_7`_X*>v4%tM0cY
zY%^EAaI~mT$IgW3MpKFWr)CYR>NaSRxzD%GyXlx~S&OYqw_(1q$*aJP;Us5HE&Tem
zwVIjw*lRjBCBg|SHiCU5ghpHKWp4g;c1gUWyg9<-?s@r7CTJB(Y@r5Y*X^oBLg;6X
zA#c_+-@{??^IT%ngta)Df&#m}?&e(=j@)$xo~cxsMi>IC1pm%~?!l9PB*LGmEtW!V
zUR)#}q_(0UA?}PPXG|qUtp)WV0J5-QsP2Mn)3Om?FMH^Cyx@zK6~e0iVbzDfa5;0a
z%BfWtFRdTgBobAL$){|L%sskE(8W)mh0{Bwp`7|We^@pa?a(Tgl^>K5p_p=fk>+9B
z_Vhj9(KJ@W@#bCdCnY;`ugiS8o9#4H+p`vf)#+d;f3RlHnQIjtY-Ov}=YEVsh;_HO
ze6ICmhb1s&C6;1~&uk=0Ah1+U_0f&?6_O4T=Fmixf+~vEd>%o9eyA9fp-EI;tPo>k
z5K<>^RT{AfR;Mut5hb8NKwLv0S^!s}t^u&P5;2lWBNu%4NJGNC<dZ>)IX5x9Qi3;~
zRE)QE!qylGo~jo!SvLyl12?m!zVHacy_mS(dJ*>?eF%<QgT_~!722nRZG5)O8%)nd
zS&G;q!BpECj1`El(q0bq(`$@(n<1wYI#v>2{nWyzOMdLpn2f1+=!>;XX*df(@XRPr
zG#v@adKCG|DZ07s!aC38OH#&!y^gm?s}>G+6*Q4F(_)TS5m8}SR2!~-C#0FYm#05H
zs^0$huqcC}|KZ?OcvROx0wL5!G)7D~VP{(4RU5;w!bCkT!OX#VKJp#1yUTd=M=GP0
z6IC>M#)hU^f1!L?=Jl%)MDn0VQVdPB%66REp54hUg4#aq5ENh}1y(g#TY<Cf_+D3u
zPkqP!gJH3&HrMC->+&kn;q8U)Ho{pdZyU)uBeQI-YPa$JYdqU7i$k&3yPUgmAQkv|
zQzTPq8xW~?WPyg^PhMf}c7JboP2QNgTH3XP$?{SZEX{Tnuh+yD7dUJ=S3PHK-~Pj<
z7AG$gvz5lvmP|TvKP0|7cNEV#_4or96GzR|LTxV%9~<%D`2k4nq*8oN{^ir=Ae$W>
zq~)1TFBX`(P3&Gt9n4-UM^(=Jw9gUybIGA7z1=(g6aUxDL`BWC6KHuOkD^#i8l&-$
z`2FD@8RAk|-1VVd=iMO4ao;ul`aH)(lL9+pL*U+$jP)5kz`^P<4|bAd|11>$N+3FQ
zZR}c$GDsJS_kMxa1EO0P$}=RYl@Y}8!MLIH@%nX5YOfUN^^0a6L$s}QUSA67W?^*<
zBZ8pwe1+Opu7z$6X#r55IaBnn@sLUGj>L%NTX#^As%?u<)z0MLNw_Ri$~4&(qe!%^
zN36Tw^L1Z`GHhQ8-)cTG->=P?A-vFplavcTtGy|@Nv&ga)o*9bnXOS-?6wD6n2M)U
z)sQ<}uP^v@mcF8l-)$aG_{Xt6&wihG@tHQ9K)TElJ=#9|Y`uiAM`R0buXR)S-<ntR
zw+DZiQlH%?vmH@BWXIL7CHuiqs6pgH4ERZzl~NZdm=yAu<5A@4N(hj<(}*Moa3W+$
zK<%KJTB6Tj)0c|O>vjrT3tlV>IlR)!$u%NpA{#szqs_R1!dm!(1(x`k7x}oVr`NHz
zm2Pa=nVo-T9R+K?@6{vMC(c!9+YZisLq1^<cmF)!zH7fnVNXKkP*rWb^o<)wT5MfU
zOeIg+OjCt#)hZax54pX;w|D1xFJO!ylH&Cr&mT~gc~zH3RJ%^x$(2}F8H{Qik@uvF
zp5&6wulDToaR?J#8NFUKOyydNXme$83}qX^GXPzqDZ8kV`RlrQBQ*glV<HgFD4yR&
z1#IRYbiG;+vTSYVCa&_OVjjHi>et-z81ak+rJr_4d74uw)ZCUQiLw+Advxs6OVg0}
z6l_9;JvIOZOh8dUM+$&ijn``3zNjIEfuLRXkO;EYgRq<eHpcy<rBqP~#{l?4F??bx
z<^iM7M{r3s0GC+DWApNESbw97&5HlT)eNW0e{oQkOCjzvH{q6{^>dp@94c}D7QMI@
zZn+x!B`{gw%|&I~Wxoc?aUbrOVLL{aqREW@mb`cB*a-@MTj{6Z0!xV~|MFm?t=;D&
zN0iB9ba>j@<VTGJ+?m29{4(KT=ZeHyL#74A_n0r7F#*4ko=E@alYXA2VQn-7<i)@n
zA~W>~hn6NoYjw`%`nty*-ivN<*TvfN{OGz`j=EQWrodBWTEhvReCOf$o93mr*>`ga
zwG(Zo<GIJkm(?}kM#?bo&7v6oE}DRoS(N^b?0nE(C(i+v!9go;jyXW3b_g!z&}@7<
zBI#_vz7f>ddPF*#5T`ZmE;~5iohypAE(C0eu#TDt<$KH)+M<mo<2f9Qu_xPrpxO5h
z6#G5kXw3IVkXOQC9^Tee=+avd`4w>*MmQTX_uwjB5HeHp0|?4yzhp4)1AZ$lc6|Ir
zQ5obp3hVy?GeFG0c#|#8S==as@IH;EH7<d{(Shs3JA<q1M&}opYKunC2yOdm?Yt?~
zh75OQR%6yEcYsvovZ1GaQ-l5T%luA%e$7)&?uv=Aj7Vj1D7(T{Vez~KV?M)Xvs;Cs
zObK%}BNiw9QslQu;XH)R>FVe)(QsU}h7*NUjiA?xDm4i6D8yD$-2KcKcco;oqGqy7
zI6I7nUD?>tu0+FJGE)n4iiZ2MJtwE2l7Ot38Y*wED;fUNP>Cl!dPTIVWXAO1mQ_`A
zuIy7fbX3XsmCx;dgq{6FdTvqi5O$}KR$QMOe4>w|w7EJoI+UehXHB?zEknpqKFS?T
z_A4^I^ec90G2D$rh7u>4E>Gj5WR1KL*3}Bx&zrbsm7XE2P$=6SG+bNUMmslcAUSbD
zsBhc7DT57TUj{aqo5QpQO)EM|vjkWk?EAOSrWN!ml~sXoTQz7}1GW{U@2DXSSCGE)
zE3Wb7^D1S(aBs)ouNpCNKI?2BK0Q2We8(r$&)(d*oPUAt8}&1E0|pg&-SLE0fm^_E
zS*uernrd;&hi#$x72M|LyZ^a^{0;Qe$IG@2-SDH?eYx^EPu$jYVL4mkshhT)e*bJT
zo_xr@0(zl>LCH{X5q<!0mZRIMpMCY@DT@|Qp0tcLO<cI(iiryr((i8SILZ!&2;8dJ
ztpriEjE&kW3)L%eo~~A21NI6x<Xd>!T|!lu8=E(43M0xcVNH`|$IADK_o+YEEV}8V
zj-%T((b~z}l2>;5OjnJ)?hfoRhBdlIh=QHGg6U=zJ?KANie{Dd)g4OKnWz1~yj9|?
z=Ph*GqV+t}f=PCoMNUFo;fM-3zhkSHoOnULoiBW4WcxYR0s3nl_$@C$jBaGoyX#hG
zO1IJ#=1uBWPY>PdQ4=n`>(=g6?oq2;ckZ{pdawQYU;Y)`g8>tkjB4yNImj*h&A(sY
zclg4w9ph%~y=KhZ@oN=7ga<$5ounTUB{hL4(UNlYl_!(FnL++k)giJ&N%JRHlHXNz
zy!<g?*!Jh;XJz>}@0?~oJ1oC?R(=(}i|^bczoQ8OjXChJonT8oQmhd=)V7FfQ6!5y
z9F99(NOGqg6t5#3cR6Y5$*fnbj3U6u87-}=Xp#Tki#emK5H%IF+Y&hgp1R(8{!stB
zzqv~(3^*!8a|OB$_N?<*ACNp*=LOMjZL!!9Ey#vP?5S}&JlLLzM-E#sb$Yc06P#tU
z#b!r%#qM1$zyYVSA%@r_UZ9<4BtjJS6BSiR>yHK^j@iPapI?|6P(*_4V97L6e`T|8
zOyPTZ*FM<)3*n0kcEO<t%D00aA_wIuPwn5omt@jj!0xmxW1dkdSr$r^D#GJV0yYq_
zymyk}wh{-(QJx}46+eW4Obb##5pBIv8gYPzZ!vf?OKIq6i<4j(HCm-*V8J*eLwl=j
zadUbeo@G*OmtIM8i7C^9DeFn2n_WE&+g3DeR%dzc!2B^eC1BTj(CoslOO0H1U8trY
z;tku=UrNd2P@rKs(+9j-R-s~xiDX%52P^tWnE1nH^O{H^rbuYG4kWZs4C&aX7U&oa
z$76l)U93t^S)fx`U=Ag$s@AFpZ-fGaA`q$zaW#C3jqImVU6o5QS|L_dh*7Ili(*-2
zmHMJs)>>P_wN)U8HF$(yxAckjg|~Hh<QTDU1$m|_jD3`KeY?q`RD#LKB49?W4~_98
z9WX`YF{F_M%t>2Sh%k>$^7@ht`GGdLH(|b(+T~$ezsoMq?M=|R=I^%F9gBJ_0G282
z{E!QQ-YjIQa7QYwAZLT&EkOkuQj^2s?(m64AnRo`PuyZHE7BzuDCl}bXTl{A!SXVr
zbe|T1h?*#zgXGLdO4$yY9oe~gB_SF#hq6Js8FQ{01*iZn92T0<;kWQECxsNjJ@yr(
zyO&$Id;6@BrVK~E@J{2%n{KHHMm@GcW6Gwl9?*2-s#n*O`<ISwUb1-1xFuxhj~)tU
z<a)GruRWVjFP;~fupr-Lb~&>g#Sv?+e8xNGaqfZU*|VF+%$PxS0>uf@ya~G6$XG5#
z3Ndvdg`|bBIe?Lf(#YQ(Da7IDE<y`2;y#UY-5^6WZ^|3n50ekndkUlIJq^9xQ`gHq
zFGVLY5^%3BK`7}@T<C6f2Pz@+sq2-R>j`R??3dox+I3$?2)ALveXS}+b@IN+Eif0&
z_n)LM$=?6fOkRAs+nj|d%I=u%^X4^Z7xTco+H~>#eaZWax=@Qrmu?A4R=!Ql4k5lB
z_4h#~=P9vL9^EIJa4E48djEMyB412&+MXdSZxqt=>CQ=)o$jG02#?~AN!j?R%$Q`v
zZWO00)_rQdS6wce=}SF}Fl%Qei!h73PkhDh`wN4G!4k*AyNb$-hUc_Dh7>@1jx^9{
zc2n#n#G|_UVk+2EQ~<v-JCZOexwP~!MXS;u$1RPbMQLctG^ub^82J#pQ?(ns;85%a
z1ptG|+i8n^*HpVwZav;~pXY_CBb%m7X&PAx(|Tgc2u#9_NKLPWrs3KXAY9L@89C(d
zZp6dU+z9@sgUQ7`vFc826l3hN&JJOEDY?agJC3Ms$V?~Q<&exY;6?{0r$lcVwuITD
zaUlAasHeEMQnt-zKfvjY#Nna2ct0shb`@(=OQTy<JfD4Q@UL&Ze$VVyI<Z|OA7*s@
z^uZ$+P98FPF@Lgd+RAM!ufBfZx4EeWv#H;}34<F4jK_X6h?uo&X*^(O8oKrm^#EqE
zSWU<?3&m(Lr7&7ji-_-9L@BHlWCV#gmT*h>wTQXGG7*`nVwXhPr_{(h50YtTmvo`8
zC?jJt5o@R8D8<?-Wq!)U)00>`#S~aU1wO$(X`f2xU=n%<oAnAvP_2%{3$kfWuL!)r
zG6W>VVqBh2L5s^Mj!{r|jWJyTbx()I6VEOPpitRu6uQTrg6^?h9=az;)JQ_Bh8EGH
z-m!cP<Fgmt_|d0n5N~Y%R6gAstWTZIG=JQ6cWBfoWMMH0rS_U10bObY4t+0bjx~hw
z+kY7~2Q}g)$T=O)f&qA~i-V~CQ{z4J|Lb@$EWc#DvY0d+$q|F{Uej&NN5F0vVO@Xk
zn3aB#|NfY%*}0U?`%{*u$I#VXnVdTI-7t0+=%iiWJ$j6)aw3T|Mdf<w!y2%q&JOFx
zZpHp7Da%B+V*PLDCuMnT_5`Oj7&QvAVfWcGy;MXlo%vI$hECN;x)GqIN3{fbNU?Z8
z?yxpDwHGuV-R04E^u3_*C|yu|3@i)+P-wKVB+dw34-}<G{ujA&{Sh=8RQl-{xFhNq
zCj1x1z!AkWXK|~Wa(C(#(Yt%QJfV~*t7B*YS>`|@X7n-!)8)rt{th~boDJhw){cFP
zU9?EbtwN92NIl7Z688<Oqe|@^=IFRqXs}Pm3{n11E*TjP%pFN+716YXX;iO{)Idin
zm=H1)A&*3Z(sk2q@GWWdGfq=mD(de!4TKBTg7YppSva$~Q&Z?o7?tU~b8(T5&Z>v#
zvCKg5cQ#UtrfkFlv8bJl;0`F8-Mz2?31i`50a~*o6lY&zGZ0lGCnUO#OFDC#l-m=%
zN=&N|y%lKCY_{_lrLgdlW^J+JzFg5}ZOO2C5hxFj+(9<YYV&4fch5FZo*|xwqej7P
z^EeT-&F4rqy!y2fmv7*~xn%FNB=2cbN3q-FtgDE9&yjx}di0v`Ir#)@xIo%({`Tor
ze?ENs7}g+PpeSzVF3`I58Q}%+bM;I<^K!4qKcWFVTtkM1_SS=6Cz<&aNSTTTWKf<0
zg8orH#Rd}HVT1?5t%m3(75sNlg7OWtNmAr(ajiLnW+8G73OJe>oA})x72Zz854wz(
zd=aeB`m(jN!Al<Mbro9q-=UYh0KMb@+Z}J{5-%Co0mUEEyif6m0?e>p5r~R=Lm(<q
z5r|qWLO~IATYfkQ5$>%2ULc|+sLj3yjfkJC=$VV>(fps!BLVX$44~s{u+e<P51$TN
z`0t$#S*c{^zh_45$4T9poYM(9r{HDgRMy*^%Dc@;DHR%&QsIBVB2|-q@4Ti_hv3Uu
zy!d@Hd#AhpRpz#r&aEEz30yKa2tZp*1=X--vAZA^k^;D9`*oi$I@tBfnypXGzNONT
zsS75mlx5o=Nl|aJAGElIoIqK1Pf3+x1r)4R3Y5W>%BmkinBEuU_tL3Z31pQe+C7l{
zKC+#;WYuDgTA|-=O22X@%B<{-f*>5VkwieTVO1$W3YB^o#h04sJ~@?iVhF8RH5e^K
zs3<Lj)h)PlEV!S!8gbIAYX38^vEEuuzJJ{({j7~@82mj1;%a+3Ux>Bm<W6vh!Q*f;
zN@<Jh>g~odhn{%!M5`wO65K~Zq-vjhghckq$K*Hnkt)vfxcoYaJTAW>AKOR5@}KsR
zaz$n%oi_@f37f%R@Bz)QQIQ1-fZ{q<N$nA_g6<GV2U@0EX%}1?tj7>RfHBM&pn8vq
zrM4EQE|cyHNlQIwO)7BWv!-x^L$<9YjmyA?D2X~CwV6I?%28xfsO^bHIJXUp`Pc>L
zaC!)Z6vGWo2(A@iBJYSmWcR~w-1X?5cb~rYCAsb4mmhxgof_8BykZ6rdfnLTn}E`J
zwH=>rCYpzc?m^-`x_U~h+`d<qACylWT|McyZ0qtX*lYVt+V-S;XI*U0+RcNPPgZ*9
zSp_Q!{7QT(%%01kMR|W9T2#h=4K0e-Dmb@hTS9Mld(st0zgOHSRBo5Uj=COGiM=66
zX&h>N_W$Vo5iJX5?sDVI5#QnR^GAzynfbE|shzsAHwy4OQ!%PSgA+i5EavBzqe1o$
zq(RRA2QqK~Wxz{2Pc3#a{kup(4c+C$;mTCNdl`x(jT)!YWH-$-7BkOXL=}lQgChzg
z2}BSNA60UGm^RUVCyms+BSPBc0hvr;HM=BQ@Us-dFGPu0`W-}MBvwEqEj&^>WrD1<
z=<Vs*sN2)bpqlcPd=rRG;bm#nTWoSUx=p-TjrU5v(>--JtKf#g+R`y5L_NL)In10w
z4zv8H$YF7o=wrQNhh<_E+Y>+RsZ@R@1tZ2Sr_!(FXP%%jmY2CljU271$T5*-qxp?A
zp)J9sQ$S&FMB+FF_Xt8Py}1kb=$+H2lPL`aqs!``7+nNjtkzz(#d`6`Jwx%1AkgYj
z)ulL;%C6+(lUFQkkZ2tdx_^U$E)L0Gbi<3c4n>U2!-b=F-}<AQmM<2V_~7gA8Na0E
z_E9g8!(;=ZWKuj~kY`rd!PGH-X5ukgV~oTz>GMT|HnBP+X%X$HZ%B+Cm-;{y)yGPo
zH2@KDi`7sC;cpNw219g9aFvxlSOhJjRG&irRv|572v~@4EzwpuZ((DY$f?6)9F7<7
z6Xc_H>dIHoD2Ic6z<TE9fCyLF5}DfKe023BgO#S@87UvTh;y*DIBrN+OJWVAR%!G{
zh<2BgZfHRm{$y<tF&`;hrBau0lp9`kv4zC=@~d)W`0z+xb^nIyypNw+Aze9k<bt~T
zA$YN@e?!^m!NX>zy(%A}7&*(?U$;DW?}(*+N=m)fg2J-8A+w{q-l(elvMM;BwqRhN
zev`wYaeZUM0u7@ljBaOD9395g`7ZYdAx@*25Hm%=(UA;5_ljgzKVCr_cq)9@bD%s1
zqF8<0Lor0$Ehe+CfV#Vy3LADVK@boZOs<$xQ#CtPATiDpt-T0A4B^96>@W<~xNi7i
zZ>;0fruD!M)2EjYe_*$Q8}<#_&STxN!_sx$k<DGWVU>m4DeJiA`yLWP^2YX$lDJ`X
zzI;DA-yqC)a<}=Grsi8p=Udusz6I)h3$QB|=BsqYmj2&QHwl%7dg1%0n}SKZQ?Z%f
zIb#JRjlH)kzC#|rEgYnCj`ZMhI5p=mopYGZxth+o4CagvDJ|>C<M982$H^k&59afc
zlu>-(hcKH<sD;TS!;}2pK%>9G%poRTh9<I_Z)=H_RfFn)U@smE(2j)yeXxr%gHn_k
zL@6_fQf3IHOg}?gOF=(;2#OjM_Os%1$C9DoP*qSQ6FAgjKbz=_NtF;9HX1YrYff%J
zSKT-W-GwTf<jkakH0Uxk?H(?EKXort=lsFMe*1FCofSWb_OD-D8+~120#AA16qeBl
zvW98Er|zccT%DqGHKlX4kM@m-{ZsK|44jw7kXU#egUFVO(!ws>q6!6lB?<(u2hU^x
zHd&?!Ux9=dOYI43z))6&1drMdXmWCO5!BdXy*)|&>i>I*?pAPB$=v0-d_TcIQWLEf
zmA>RI@MES-8QJumgyxA-rTt7<=YLQ7G`ynAH5R(g`+&3`{2uKl)`SxEN(bw}P$Hz3
z7KbD?9zLR^48%I72zksZHNd-YPa#*ep%g<&HDLctgRN1o0XZ+xg=%OcRM;N(WTt<&
z%9IAe0+GtX|C#daY9*;4-8-iL1C8qc&ARI!es<l6HRIM^Ib_7*DMLps`T^RNTXMr4
ztCrsI)2c<I`VE{oxUpfJ`dnm9fI9)yuVMN#Elj)seVLMolmkiz&!v!x>wv&OdO^i_
zsOQ@fLZrY-9XPe6U#tbGZ%FE4vCK*{B_2@8jjBop!qX=k#%zs7L2q?xa|((wswlFj
z#~%0QDK;up0A%$Q0f8gf<j>^Ml>qlfz(JD-k>VsPlHP1j;(aQlS9f-JYA*p9%nNkG
z@F=Ru*OZea*@xeKVIOa~a&hBRgEB|_asHPpHncp85QLlOZ~k#n{eemHAJ$Fz{hGDA
zN!7r}t+T9;Kgh0A0SUhtwPr@C(K~unarW1_C9`KM(1bJRBQ)WWh4UY|w=%Y@dh)P0
zQW%BNj-wPCGl2E-6gFmIYG*1?7yA-&RG}_TLS<rmBR*v<L#dbhGH@~Y{Q4kG(3(+u
zWTSKEvq>K8&CY=Eg?A}?2wgzJc!sZl$}N46q~4x_+O%CBYEyf0xg4bR66nosj}G&z
z#qtmC*!#-GAe{ZUYFCd+`BnZ8QRXNHCHOOLCIi;}O&6n3`arfa6%@)$l@tZtQ7Cg%
z6v{&N*=&ffnPWS^Tb$_`nG}r@<`4cqoqta_%1qkB?)y$SO6t{pCnV)3=*jK9JS-)D
zQr_5cL1nu#U;cVJ-&~mQbw2=gv!FNB%|aD*6KCql^DDTUI8$FQGyRW%H_@Z}Zt%@n
zO33#e7@Xu0)p<+u_aMwS2V{NC#q%wM`Q|F~&FvxT>U{In`IdC!Z@%py@;56dS7Go0
zWH9CS#S_n^6EFO37AKv$?__fvOQh(?Nz*USYi@z!zIE}Nr~EDf-@(n~F(u%=K5;Px
zz=2En6bdNKG1yKOjv%5SdJpAc>d&Shm$LEUI@#%6-UE^-+e)`)RGPrrsZXn<PH#3M
zvN|(zQ5?hqTZ@Iy$C{0#4Px%gYAX--LM|UASKvFz)#Wibk_78=8G0q%Q8mF)ktAHB
zT9)mNf=ZQEW>h))LUj&UCL0bOy1zq<Ppj^Z5vgwIr5Jkp6fWi)--`mtQScIN3I!4c
zA^TY}@HzPyE^023*<PhUS}XVGR;c$M``^1iKFx*Roi+&2yHA4{X@)$F-d^3FLiZ3d
zoQ2|n8*%<R6%EEf?{C5uM2ZV)zWDw+gq#!<1W>rY9`28L`4k0Hhk0~m{;5>}bTFwf
z;)3`kjwgvF{_U%DNI(HQjHL!-T2Ffdu2IpH&5^hXT>~9$2;h1&1Q?V(wq^w0zz6>J
zDEOzwYCs%1LCIgjSdB1Nqe&U7#z40m_1Kn+W9@opSH-Lu$9$ifOXGh`1?bzO3N?S+
zl|QTQe4n2!Yy`fR;ob$Bt-)=%r3xgyF(f)dwB6oxyv1N*EI}y;hU|nPd%{woxlJq#
zscrURQ3f67tBhnaR=i3ihZE18j*3@tde%WNW>82c`sgaB1NVHim`x7E@5Ccscj5Io
zT<n&@vB%=@aTU6(#bCC(lw?lwj3#q%mn(vM7TQ__omEkRr9MSX3ZM;i;5Z=;DBWSE
zAl`YRMHCVsQ0PXW0K1J=q$mI$)}sOy1T25e6`OxPan}rf;N-6g_YNLJR)+G=`u8X9
zl)wJfQTb~!Xnym6rG5GItFBwzkL+A^WJA+(cKgkZ`-07be36odpI6BLklUY;--@4q
zX67O?f8V^EVt@JEyH%`LeiOvAE~bJM^h7z#!N|4<BHLh2k&c3K$AeyUZ-Q=!Y<plJ
zh7fftQK2~YT)4$m;qKCl2#r{AIcP-Oag%SPn@KSYsQk}?5>xnyDa6FyfD(J4B*qEr
zOjRE$5YOHlRGqLA+&&lAWECm#g@TtTdBIbod+IKOoM<Gy2g|(yDi$gQ)nQ#>v06~=
z4QY`d+e_~Q#a;y6XMvTUiycxFebJpnU!;$ZRItGv23j%os=$5NGV4%9)C(eNHQAL~
z9o-w3<-QCUV=9++89>H!sGk4SD<~uXF1TCMQ^<_U*w@gpXTsP=_lg{u*Bf%APeqPw
z5lmUs(a<4;!ViUzq+ZlzVI+5{9#n5=lI(=;z7x>>AdKA&zT#AM>>0_i=kckhUO`}m
z0N>fl*hPDoI*1;X^x3wAV0D_@YMs>PKvB!fxXe8C-LqVmgT&-hkD7R`S3YoaR9EO?
zUf|ldfG1eZJgm+q2=npa($9sZU*O5rDeNh*(@x+t+~<q4t-{&XB(t|IX~meo3Md+&
z*owNCP95xB7VgQ(RDP2gN@R6mCsYK$V>a3Ig&@tRWZEPbeX>ziP^ATZ2B?&o)r(Ai
zM|SFRVtToSx2boM-F(r~>ll=xo#IDM2b*hQ(wRaggwGSB)XPdrLq1q7Pe`nYB<zam
zEeR*Hzh0o1)eH2JTqe4GCRZg^0A0p1lad>DAq<URvU{?y;4bM96St&-LW*TwMpT%d
z>7yhax;cawi4Ch7X<s)AB+FjiN%1r_)E#qEi(-#rxj81-d6YZB^J+h0HG?~^)#gL>
znbYo2!lt5oq~q34(GsS*-w~#dnNaq`lhgydtzE6Gxcj?f<Mex`&$w&Gl_N&Zm;s}@
z{`1e}_Mua!4jndiD*LZHrcS+M3f)s6cizZv=WA$OQ;I&2L-ioy+LTTV(jEtOZQ_Ci
z6+_$FJ+#NcMYn&0>DD?ja@8IOy+P@5@Zp092;#M6VUL5iSP6AtB6MdL;<l*pQxvF$
z?cU;`p&bqcY<aat%eJoHM)=2eEL-=;gSl(B$ns;`mu-I2N~V<$UAARJsB!qp|5M%7
z$F@~n@pqr?=h$(cpW`_GO5*rCiSsf0kveWgp&x2X32Es^mO|KQ7g9t^2kp8csu42&
zfKm!VbX_}Iq?HZr4`A=vUTG1GZdEryDV2uAhbkrk+fXI8YS4|*CGPGy_dUl>UPuuV
z(-g%|zOkQP&i&nc&%NiI-|uSIvxB=o+I8wMF&$iR@TK^@#}Ah-9qv2AC;MXL$>ckw
z=c1!O8}1F0)1!3#M*c@SpL#ey_s$nEKnA`P17z@PXMj}Q)cVK)S+IG1?0jGjB1l-4
zjL9HBgA_a?Mo7ytJ--+1c1*L|x&LFk>CBmVTcbb62vx4YbdX13Y&zb|(*2CAXF5HV
zAI)%gl-jSE```=g*Yc(87y94gE(mD_)P#ATpQX6Yg+Og-Ha6v(<zUk+#$*{7Ce4<2
zV&y?Qk!H0f7_&_-`q}gv%^*9(T-K!)^uiyS9Oo-64Yo@_bC@=Vd(;G7bW&fh@x6@9
z&+MY!HO#B%XeXfBKH7E#wl~VrGwz4?r{|v=b%)tkm5<K2Q0e^OrmLC(8>+8|X(U|{
z8!C*p*%(oI2RmLcU`DlEVtE_LWj<uNn{#pj;4&|J%qSw@V?^dord|-7?mFA(sA`&p
z%iPH}Re${SB;%?^)hZERF-N`-Km{wYO2JCXDFL0!T$BP+T;}>4*1e1?p-l3wbD39a
zSyfY+50gbBx$W2g^tICK=kL5{@PYoptq<S2;<_74LqaJyIg!qN^xMm&f=*?=@q0N|
zOUU|Z7wxBJ>cwuH-cvweo{)=PHe|vXJkIpIr3NVM1I+P?IoNS#03=fFwKWiZiiI@p
z-Ye=~qx7QL`kORNFUK+Hd13vGx}2Li&iX9sTsCv}Y1*$MN=LAIJ~~w>flSd}=K?Kr
zJ8a5Zm@t83YFgC`k^J4jFhCto(FGMqYJgrusbCjvNLFrxLL}<>i2guDmX5tnuZBo0
z<YRJ0iDD!M(HpPqroPTFnScK)B=Edhk|%%PL(a;jbedlvPjnqCz5g7!?)cN2_V0W3
zGd_yt5PBWbz5h9J;3qu)@4s>SK1+DQ^fuSZb#e2!A9IW+phX6fdL@=&LL)X=?a+Pg
ztIX$%-OhleE>-N-POY=}&ubbQX?FCW_R(uP?5cmkVsi&toJpi&bGY&%rnhpc8}?w{
zCdSU&2xX4l)X`32=Bw&RQ(#yixNDs{yr~Lj&0}Y>!17qV`z%Ma(EmO;_Sl)@Z$TLA
zwxv(Va4LI#?2$LBcclDL(kot*TfU9C;bUX!)7$Tc09Jd`g<y1SY-)dc10K!p0K2gq
zv_o7CSI;$ZduL>!0l`a9E;`s?Xv|=uBg;xt?64R(8Sfq_>-sCw8?P)=%}$q4o;X~v
zSwXQhlcC%tZ!lP@Q`lA-6s%1pv%OY1Gzj;YIcs|SOX@N{Rz6Uz(q-fxFkfvtNB6@%
zZjfWC(GuMPe?^*J)l{Qp5YPuO-n2e2+s4%3^rxC5j;msT`zn9yB|FUmKSKZJWxR8i
z^YA;3x~x|)of62-3m5e7SYN$@?^qw2`pyCV2>+sG9i6X(9MV0q*b=T)7*KL)d`TwY
zs|Dt3T<s6^4Sj$6{JT20+`4?R^R6`;e`I=m(XNf_7c`DHEW4vKv!r#^Eqn3)!~8S+
zm}Q8L0Jy&!Hk{~L0;VC+m}w-rG1Wh*Ek+`59q!z+arxr<ZEH8)zvjV5uQw&N+pZpd
zbY#`OyKZa(uCd*vBjj09@$43+1p&`4T9BCrggY_yuHxQ52(4P1U9qTq^Y_H`q8lQa
zr4={l!7k=gXqP0N^QJ}J8Z?cD_9-JN4K7s>a}ZNgt?`_p(c*Jjk+);Z-T2=-x2@YW
zZeZ6<(_0<^PDuV4^KP_r9G^F#TN?R(fM<a|Z&EY5YIidtlmLXlTcK*qs1xFI^a?pR
zS1*mv>^AB@(S9=>B~`X%JqP><a5fszYzv*J);8qebX0AEc|TIGgbX`1K36Z3gL8KK
z_)H)mhO^viQ@?2h&g3~dXBRCtU`x|3DCR&=cMTfZ1uTxS7+91v*$s3}63Ik%aN(I$
z=jl*r-P+dBna<Ri<`DVS(CZ7=u4`$2qqFODYiQm2mN2Xz(aY~!Cdf*xfewShJaX~7
zLo$YsfKo^FXENBTHK4#6BO+@8x)(MNy4X7xCV>~57=L&apD<g=)>1z6iC45nw+;Y<
z_`p_P-g;=umP7OktSxs8tlGMN^~lKTRim2+;5x9GU002cGTvi9|0Z&2ag52+VhyGF
zonmPxiEkjcPLLI+FLUMz<n0X7&z@zU&63_Yux$m&?$WuN$iLE)dp-n?t&_;HML(Oq
z{I2<&1u~!>1+G3d%G~Xctg)sk%^8TW`Q)OFy>H9_A&0w1bqK7?&CI0Tn=|<!lAQ%Z
zHhA^fQc66NHwSfEQ2JV9OR*X_*wEXpt>~0Q+mL8T;UL3Zb-4BDQC7ZcUUptKnw0Cq
z8s~Qe8NUUq;0b%f$dTj0#Nf2>jFpI%9m0_($-dj)TQ-{8zijP&ak5V+T~lXrkU^pJ
z%i8Tf+_`DH`p?S(d1|jKSF^HyFM0Xut1hH^HxKSz+BhKWy=^pgG5zto%pX;FvsJP1
z)P`Q}FynSoZFG}b6cD9qjgumUF_mCLB_F0@%IG|)4I1y5fI8E_lWGxkJ(s+I)RF>F
zL|P=!>d<Oh;rLFX@0?Q%n35MP21!Sn`gx(&;qovN?p_e7j=BhSL1Z{Haz{<@qwpV>
z{&JST)<JIC<R7Ot_=DB^^A8{N3+C1N{A7{ur@y7izw`BzJfEHXfR9X`qIduY^_b}x
z9W6;MT1E(0<~3yOtkoo~X;;_<Uzu2;)o)v(s8<-EGv{BmTqAElmh*n#(v{pEe;t!&
zj`b8<P{vScp?()uO@LdUzzs};{d^^pTLad@6~j!7F%S;cq}V}cj~ZYTu@>5Wn|X3x
z2?TKH7D(|!XxB$QInb?v&~AeVG*=sO!Y<U+xFt>8xz^5jt79m)&=N(5*~9YhBot-!
zgR7_3#TZ9hiz{yrhP8}!9%(;xdDbXwo*>3PrA*jmf7R>myLr#I#%^D;WZ}2{?)QXC
zDAy_!$>NtsMuxt*Y2^9upm^(s`$y&8OG!<>b+X@dBsq~apFw@X4I5bPzNH`aW_^&~
zfhWNppDgRLv)-P9!z+k%Ed_h}4a8@G#Q|1zpM9#(sVoENVQN=CN%4a18maAorw(ja
z)~i)Ku0pNgf)Yf<SFhS?W|y$_&MskVBXI(TF_N+EB6hO4bYX^^?PNcbA7ogwo}+f?
zpzBDEJjLbPsU-@rc&ZyD2)h7oVzRx9!Y!JRWiXBEY|UhpzMf)+0Z+&%841jU#A<2)
z>@~kfX~&okgn8f;Jw{Czi(_;sLQNNA?7+PYiru|EQw0^>)SL@6bN=F!KjU6B*)1w{
zC)AFX&lS^&7Z?8S(;e6xmlWv)y_XU?0fL=pu`wknoz(K(lF~(6|8enlUAK_$K=r1i
zR1lLuE*O^z&bZ`;hbgqiLUdOfg;!`#Kzp8kp&~YXlsDB$p^nb7;-*L&M5h#LjV6*E
zom~ccoAEYq6$HkF1Kq2n>eT=Q2zp$q&x>0}r7x%Sy3|Y-wU?7jyrvfFJ#0R@m(0-J
zX2Dz2)61r*s{RjW%L7K_^&t~B95(fdlP^4d&v%Dbj`~Hin+~GkJ-6Jqb<L7%L*0Id
z#q%fY=MPc5l*R!y=HL99EO}}3ZK(D+0@xWWoEW$`f7if{p>Z5l-j1kO2a>_y(z!8e
z1d6L9o%l9$`2?UU9BE5YRMAOGZH?++YNlZ7pp(0d5UQn<4WP)a)qf}6QsWdC@v_`y
zJ{-a9$q;D1NNO8F;J}Jlg_syaOmZ4xB8V7G>5?gn+9^bpc8VAZoQ?{_P@M2dL=44;
z7z4yaC}JYgOz}(F9*rPkk~0v~lx&}ln5O2|Xe`;@VSt!OTt`e;Lrhz}>IcU-<x;(k
zh>;khv#O;Zpb%0!{Gf~5m@|S0DGROXvqi;>QH6;;Uixm${>-+){sr@EJah*61Q9WY
zi1>)qf@WzI77DMVA{TSlf1?rz+N{F;7ieCfwg7nA%{@zx<H&iO-Lcp1#TsiP@MGg3
z1s~SkfJXE7?oI*(>K1!O-fp!Zs%J1qN-C-8p-7oUc5m;|#~?Z7o@?wwe9SGkTyl(!
zMSvj3{DS-dBj2mYB1LI>#v3F%J1vT=s_*@*YL4f5Ke_ey-Yt*+`<q)I$@MMo!(0C<
z*Y9}m-}-O4e%I^0^DDW&?fv+jKgjhrybs>}M6SQ-{ovg~uHW<i^7eaj{Vne&xBptM
zzwLc++pfRo{qr{ixqjdK=)I?M{lTCA)A4`G^>@6F7IyrH-iOO?%Jp}>_m_9&`g`6F
zmn*>V*6nWt0Y4;c@%omhZvBH?zvF%X)_=(LyWX#E{Xwp8d!N4ZFLM12@1uABRIb12
zb>H>n`aSQ9cmGzdzvcCB_vQNA-bc6pQLguHeR=!e<@$Z^(R<&O>kt0wi}$8-{T=Vo
z;_u}8Lr*RKORm4`eSi4_x&EGaZ}}JAk(YXF@7l|}$Xj|BUhXOHH}Lm?cMtwN^6t>p
zeebc?fx9CZ5qYuqQ_sM+W0;%34C8g-+W@X-H0opc`*V0^V07Vy@U0I&GjHx`7^A!%
ztPsOz1=G+s4WAl5FX88$)*Qk7DU6M1ExfDX`!)Q0N-KsiUwexZ59OVE0j#gQhlKSV
z?^Ev+?=$%P4Cuf=%DbXJU6|RnLKkRAy_M%3rR(b~T3+PpH|oK?dynosxOe|?M~yC`
z_@_o4r%7(QYLLu2>SOiu)Layy>WA4}C#o06ie?&>X`{1CJ@2YfF^wa0q33FPt)7Ni
zxYP?=L!Adb_3+;PJD+~?*`3cGJb0k4uCBVvYCw0E0yf9Q9?Px(&`to^6SzNuCn*fb
zG=dDm#GD+DGPrXHw`1bi0H%b*#TndA<GC8-VI091S4ysoKCVf+l5<C--1VN2{C6Xz
zw&3u@(JzmA$E^F9xl7FL0B^fbRFf{{;QxdK<GnQ=-mb+x?>Q->fuSj>?L#2@9;uc0
zTxTXq6ZNor@812_BXFlg0Ksp7y@Ox}d2<~_YOlzEa#B?k&WgnJ3ht$#w%bjzQ<`M=
zwI*pXQjt-i%Cm5;S7G)_rHh)rcTnHEO?~%v+WN;gMA56;(8n!+cLDMmD7Cq@eE@XO
z>e{SRc|D?Er*oSSG*grF4AwL-2VIaV-4Z&j2tF_1YEJGdgmE!FK~+xZ?~-O9)I#7X
zQFKlE$nG;IgE)p2@ctCOrKC-$vva})S2g5#QTONYHzaDbIQkwuTMC{E_;-)^V|nQ~
z&i&az<vP*tZOHUyzH@Q`dP%A|D@>lQ6gozA73CLdt`}hu=PHhpC|RmSv@BpAe4XjU
zfV0h0bsWz0G)-TrFqx|;SEiVnXdZ=Gq``ytiY2V6?%(Su#L^>)<)AxHyu0w{hQZ<@
z&(}}x-feZ=qEzZf4W+tUi($v5sC^N{xhRYRN8$<*1+XLmqchMl3z5oN1-sV}*rS@b
z_UQQxMkX-+3KRy_Wa+GpB1iQ<1xn7mF{o|N8^TWoKhNNE0LJE%Pak>D;M)mM8W1LD
zG%g@YdoZ&HpT2WnsUHJvJbwt^VRiLq1PkZCz_sZ45ayw|oDnao@ji{6)SxL^=NPVD
z5KO<EiCDj+Sb#YKuIj^QKpY<u9TR$vG%9-Dho8^kZy#p#0CE#w-}4DygnS}%hXn5#
z!9AkyeOh_oJnIpjjg<()Bf2vZn(R6g_&Emt2SP=^b?u<MixOh``a%h>#e8jY5T!CE
z3{Hg;Eb(f_Kp<D^j|tZaanmPW`!u)0xs5u;yln*Xyjz3GFou-P9Xb5#!>3OpSo<ta
zfk^P2xW(88BsHJpa_m5UV?EuaUFuIn3tc`ndCVGkUVdXdcIj*YqYoj__`+vi;r~)2
z&{G_emXBZtO2t}ckJdvO_6aYhZ0z9XYPOyAwMqekyHBz{BF({ZNYk!42jmACPo@iX
z>jl~J8I0*Wp68j!qzSd+J@fGM<{a=CFG*iHg>ZOFp5u`8bt1aGFW3a+Z3o2Vap_-J
zesx|V{3E)1O0++tJ1EstN_BYb8ToGZe=dG)((i2bOrMj*y&x|4Nj5>vpRo<#H@uHD
z_C)sR#hW&Ga}BQMS^H{gXNPQSDB2eYU;2&@J*6>~Hrf?}(y|T-t3$Xxmr<+}c77M2
zAGh*m(<4>u*X=fVBg)$)aY|lfLiRHj$xbBt91+hMiuE;huGiP=m#bieI}$tJ1rzo1
zlx7Y{>b^kki}g)Pt6)FD`GWP`bzF8?^FTa0>kebeYc%8Uk4i6xoH!Tm?!(;!pay-`
zc0Y$cb|gCE6Agi*uW#<3e32&eetaX(8VHte4h^mAIZ@`fP*>?5N_ZfgJ|hlQv3w+1
z%XS@U!}(3D*FMEq{JhNAstECta2kupIFcw5l-a=7?JHENW6fMfa=a44VVMK6$K<Fv
zB)vbUHC&%!ZU2;T^-FpNq{S7EUD4#D8)pJdc~}p7Sms<K!skeypeA>AMy)~Vt%h`m
z=|Rd4$=W)v$<`QSzjdu+@t<}r*6urm2R1Bns4+sSUKCXr+$vF2+-fR-v7lN*wUSHq
zv!)ixk}9Q@j7X@9gGH~FqBW5<xRhQ~74%AKD~zmH^lU~YG`3K#W!4NTo`Q;MMWybN
z3gD26cpRf?OiCl>#F-VXeId}fw8z#=NS#VeV+@SIHS9fX)*wZ3Ijbt|ab`}%e@M93
zAzIoD(4^shMC;mmZ$)!ClgWuoS~!{2ng*dXCG5IJW@w@W+Xue{{Oo<RA*r0I$ylDd
zu=x(JiQG%%4oe9u(RDd$CD<Vjrq*(p6Az8>0O2#Gx$vN(<mP%}SGTm!Y1N{vYjJhk
z+CwU12cG4jCd#{DVX<InCB6`-+)jxI7vxG#aC4EH%PAMuQd&l~Z8M_PRz#)6f$M*l
zPeDTms-A({A+T#A9F%HASC(p1YE(dLS-kK*_xOKCKR=<LpE}S#Lt9q+-hHQeg*4m&
z>_ejbS~~U_X;tXp(j(1uy9wC3ye6wqrIv+)56=mm=4S<K?Z8^M!U{`RRaYdW7GEKw
zH0hX<XZUkTR7T#Hj-)ych8gLxZQ}9TmM?52St<9N@W~xreN)I4-mcf$&8-XJ*G#0z
zSh_90$~$iMkGd3zdHn@(weliiIohS||C4kO!X@FC5T*6jD^p*H?OK^swzC<Ru7pbT
zLz-632;P*ujVlS4GbxQ=dE@iM*{!lr+d6k_jger>cO&sV*CnT;n#Ej+cdAy_0<Vy+
zA=~N$lI&AQJ9*QfPh5UQSg$ws*{rPUeMf7#iKAugxwD$?GK4k7*osQNYq3peEc;RB
zR23=XMz?c8{OWAP>w4VR5&?KF!;e}dP^U>dTIgmE8O5lAEUdF-HqTm?yfrmTYljt<
zwBLd6UkEkG$4jBnKFwuaBax=!kGrL@soUEqa6L&SJtKItbTmPj5$u#KZdS`w+VcX=
zn=n6#C9krck#nB4Sj~twO>lc`Ml@hv@LS1LGs-?DFdHM>Z%gYxqwh)WNiNq5>2%vo
ziG*NVUZjo(W3QF8==~LK_rx2~Qc*@XC#z(==7<|_#M$~r&Mscrtdo62L~GbgQwzTm
zqF`Neb|IR98SG=~=p0eDZyde7i7j`_n7gTyjCtYs=xuuB=3y&>u4~-gj%{ZL<T~f8
zc$ESbHp^$rIh9SX2<)TU9eEpHR=0YV_gbiBFT%MX+9PI+Q-Q^HVyU`YV!=ANPdSzT
zS}`@GOKzsd8WhuR%|Y3jMQ{ItXh~XmoyrzvMk}`MJlXBSDLu(ZSKX5p$f=BMdPy3v
zZC}gX)$Np)40<CYZ5G9PpOqB^d!vjnD2PXylLu_eGUArZx5_bgYhLA3S>x&%%hxN%
zefW7mFx|}1P_4okM0ADqGD`FT=~E~YZfdU&u_G_3W^c~%`KGk8f7|BgzFz%p+FC_*
z<uB`=#?&%W>p2th>K)4XQp$m<+-y#F5{buL&vUkvwd54GHf_rxS1t3Knqj@ZZe(nm
zkg5p7et!vTGS*4$JX3qluq$H>fqM(_3{AbV+SJUuY>mr1!FofBxLd=zHx=Cr%3}*z
zS(UQancN_kuFGzB^vlZlT}8uCvg1VRe5;{6+EP&6--oeX!m!hrr-e^5$r>2Wf%w;o
zLd-|wvI?=?gBJwxm0;~w%~ee_6Kq`=e3XLot~wmcA?wYotilZ~;BG)dR@xPQXSKW+
zwR27RRVDK|QEdAPNiDkT(&E;FO<SDXPIFs*8n$RZ`<{gKx9Xr-3>MT=;z(w(Ly3me
zO^KZq3}r=YTA%r`DC0^jIdtBrL+V^HqEXn@ds1dYRiCGl$9*~;w2OMqpA)gYW{s^I
zlJ3;<CazH1vL`->v>mE;G}LTJbC(U;UR*(a3icw%2e+oZSzTQBWldXsKorcWR$T~R
zd9CeKvE^LK5u=pWR3x>i(XX;Tt+PefR^<tosaV0Z(Iu|eK-{;@%O2Y<tE=?w!A>dJ
zs@onm>)tq)Hlrp-^F61wXwn;&wqrIt>SjG&$+v#QFep3lC$x8BPfWU*pX-U)J4Vv0
zWG=P&j%I(>I99;&+<gSiy!H^<CJ&n&+T59~c-EG2T8`Nai7U#7lyz2qIuea>yAjqJ
zFQpG-b60C!bLu3SH^g8Y{o8B5!pU#Y!rjW)t%!@#CpPQeH|a6k+GA6aQG-_3N3u6!
zUzWd>W`?AFyg#!~Qs+4x(ShptY<1SRBYio<&%NFpwIXaI(&^^+&C!N3vxwG<CB8N9
zwPRvO?-~`$K34r(edIxz=cLjZoHp#YnSVFsWsfn|EKsU7)<M?bm3UKA!dXkLWlCMB
z{FM*h?u6{L>6y3n@v5DgcDpbe-poCU$tM|O$qp-B=RD_<)@)m+iESo_tfMmXRMc&<
zxi&(2DEqiq6k9T>f7a#6x$T{H*B98F*!JUcl4~KIhfp9lyM$an#Vy%=D!sv|%!Zr&
zo8A3Mx2kFDUACPPHMN~KOKm+bezOy>)+^<Xwpq=v`Fvb<DvKJWYw;@Aja`6_<PW*<
zt?E!VC0J$Q)@v?m-T61`kk-3{H^rc;%jQm3c}Da_WiH!h2W@@F?VUzTV>4E}^)&kL
zu(Z>9WUTjh=aiwc)>M5TSCTGl*Fvp_4R6q`ulAj$qLGEz-QFA9-O||?X@it(=uy=v
z3PnP>v)^6+g#)W~RX>9Jc+8Q_r-^h$nosO6e3<eagS*GiyUFY?b=g5p%SyKG`9T?_
zvQlEofzx@7Mm1@4cmJFGvO9d6IY@Jrg-By2Qq<1;c%!Ejm3!&-WW?q!)7x6>>n*D4
zjpwy0a9Y|^^OsNBvPGK@r(!`>UR~ek<W+4gqU9_=?o=;rzwUAZ&#yMRAYr4^mzQ<W
z^_E!Ksd;!?h0kg`cA-;+RQ&%1Rh+oioMtp@lA`P-y<LQ;s_c4<up?YO<XuMX<Uyu}
zyVP4yy-jxP6gGFp-!>vveS)0!zBrO)j^}JY7hu;l(wUr<_5HjhAK0bC_VbeXR@C~#
z+dNhUpEH()?EiBAGb|%?+o_k@FRR^H&RweAHuv-`_LT>B<5S!2N84^-*wSxpKP~X)
zJELjM5^NvNNOZSnNsLr6M!J&=IYV^Oa;B~Q)W)9d`eaUjs5+U#`$5%{x~I0|>RF<d
z*i52D)Zlr2x0m_8?{qms=a<=1${*I`;}VHGMcD&s=HR>awQfnN_)Al>r)q^Nnz~wB
zQS>%GC#;=RZ2Of=*FKv5xXSdff_OyIvwj;-i|VrGnbkT<^@^gI(MM%RvJOv+|8ecr
z&8XeIPR`J>@=Ro#{o0jwd55>R@pF<_A<-}+zo%=Qv@x@(Gwu9nvoWS>N<NFdR(HpK
z*}COKp3TFn&S5<ZeUrSrQ;W89=iDK#{G{Cdvuj;D7HDL5w92Prd0vk#k>Ggcbh%&0
zx#lm7y3wEagk0(h=Ui!9#?z)r(MI>IdB#)A3EMh%D=6)9tDmp9AfG)aiuk;xJ<)4(
zMtcUv#=S)&S7rP&(LB9zmgx=k9LBwU{p#ip*5boj|2=a~4i_c#Dkp95OR!#$M~Ef&
zsOr0^=w!1VbK_Xk65HYv=2MR)!`pts^=3Y9^H-TG&EMVEaCNwSKeBn+Dl2v0IGwuc
z3^0GsoKu$ec_MvCL-NWT8!~0&wf8prT&^@txx3!>3*1&qt8)>mb}k|;&!}G4_Jf->
zNB!)a-POxVIVBrsO8*!?k3G8WmTP6QBI)BU(sg;_{*~lm8R<qQSh1wsc-av<xR6>>
zJJobTv4A~yN6tQkbPlZ}^@H30&}CmFD$n(^#6sCUUO6ZJ=MoFA%k`>^%#`N@meL2Y
zP-cSXvh3fvc)0l|@2||A>TxQWS!ZYd)7hGhWLt8WcE4@UCT^?dx8)PIez!i@N;#4p
z^4I-7$LphLQrZjh^;Kyh9=>hhD|y?%U4G>XUMoMDt8=3>b;556P`zcQ^-9B|MVeLb
z3)qu)13cna1MKbfX0MX;D%SHQjqWNnI31mhC%xfB4bLWnBmapSJsX^S6{zF0;Uu{G
zv^VZO_32vu?gX9>gR}AP)gVwW1}7(j;Zrp^Q+(f7g9%Sl!P)WTMQ`lm!`|?vIy?^s
zLq7<VKb#CEFV)#t4Tk-}^FjZ-cVgG|{okvz;R!q)s>w579i73D$$-A~CKiaVdPhh8
z2q58@i9bFK07SnU9pL4p2k`m;1V#a_7Qx=rvG1P(5<oaTfd6|me%w2ns6&79!uJ_^
z1r8RkZk&y~JQHRN1DJ&P_0RkO*@r6yK;ZcVrm3KJ;;XY`n!J^#b$0z5u#j=G7y|>(
z@dMVV4-L)_f9?a$FyN14#NOWFOEo&_9icFLLy%419|N#4t;Wp8M>iy4-#-Dg#%II9
zQT@q52gMGs`e!eOCuhArDG#7tP9^LC!_V!L-@WvMjv6qwzXvUvOn}?_uuL#G9h~&W
zs4qaDVA&fU`Rdd^de$2bg3|-Mdv-bk{p0m2@;k0FeLg&UapL!%`T^@1NZ?s7P-DLb
z)SsMK3AQqro}s)fJ3H!p2<!_0mp*CeX@}W}0|FG*p|1|lU=VTwpLYkx6*i?J(Rh(<
z&<@-o;`BV2sL>dRoq+iq`8}9|S|g~-xIPZ)pC5tD>oW(Zr+$CXoA@xf58^}Z1*1j3
zfZ8pq)WtjiMF;M{qA<cP_Z6z#=?K(+ydNBN_x5T!7o3lPGkzaPM80m-Cm+ZJ?8lqR
z8o-iUr}g^J2Z0|{v;p@<;{hOjc0M61AH8gFq*}RMy!bWfJx9kz2J!Ojd<=?l1SS%6
z-#{xAa?E-e^msVo5IrC$CYa{1H#r~stSKn{Qy<0yoWZ0=D|Ug&M{Q2P(t!$lh@NY7
zJ_hv%CU@$lH9)l*j(vbL>>Zv6ESD9>el_YC^b<z`yYHwVUF26`rqxP^iBb85R<m@q
zE^-Jz>nvSn;R<dfc^D;nuA&4+L`JQ{*(-=vYX9mYnq8<>cnyzcaWU6w9xbChjN@yi
zlUX{~^A0?oUBGme>2MxSW8G0^7RGp`(jt%X5xp>H9_FD6jh%?l%v#T)FvhsI3UhEW
zdAihqWX5mT0XPcB8hD^*7fJNX0<VoFDo@iGzVmFjfRWKMfeRpMrq_6D1<R##!!O^N
zYXBb_lLF8jFd!hUJi*}zeV)z=jJ08ouiMIWO!E*!ou}PV3=1OR3z(XxDq5}6EDwR!
zC}E^{)-ufI_<Ep`&Q?*5D?t2boC4thWd@uO;K)q>vWPMY^9C6e=ozkM)Z>o&Sx0@+
zQJ>QE&$@f}=><jE{LU!M@@q9Jve`vQi}aR~B`nl+S}5dL2+P2eklwRJv(_|gUCj!Y
zHWLZew18XM7zN+bY#Aoemqb&C7R>Y-Uy=l)5fzX~hNR~gVNOilHksCD&1#?p7@+8I
zNq%<{X=9llb=6B4zbG=&kPvS(w6BMpi||sbB(?8<0`P^=L<tF}G^Hy5X;OWuvkt77
zGZBnTs-{49S|s$EJ)BzQb|A$lNA1r+uz({V9kF1pbjENr%mbKNcke-0Jtc0Y3)R==
zRi3WxEBufs$sAF(32_WmJ7ea;gmrjL5aAM6P0|E36mTMwlvh)!g$6?vs}{@Gp#Ja_
z=0_#|0!b4G**RqCyqM)3=P^85BDRKQ=FAEa`W^iRfHBb}K*X;(j5S~~gb%-}t7Ntd
zfOUk(Gq5(0qhqjxEKQ;ryA;P0fM+u67(5l~BKRG!XEjPq9?Y_6o&OqKYF6mU^;-WL
zrGK{834h7q=K@l0d>2=8O<1I)U!Yk@$@{AaUkt`Zslxd@LSBQ}#@1rxEn#)6&~bu7
zM2SI{32X-W6l+9w0nAYQj}O$*=-d!*H>KP~!!0=pix;HOE|1uQty0n-K)1GRr#4xT
zC$mbY=<q=}r$wAccc>U`T~$yVOtZyZg0pCXy9R=62o8ZDA-L-Ti$idS;O_43?(Vj@
zYaqZ9+}$qUSNFg7<)4~5T{Sh;)!j8uJ>935D@~}}vfVNDMI^eBOY_y`3<!r2MAipr
zef%F`VcMp1B>_(6-PwE{0^Q0^rvm}<O!>1I@hiop_Pw{Ia&;VSs!s=_PgJsDy8-s-
ztX?X5$R=dFdgoeeKGy(A<f&X1xG*{H3z*72hTY+p+#W$C<;}F_1k%`M8?5+dQM^ih
zU$cTXckrq23bwEAn9ge$nTUU)KYx=5qN9V#N#Id~DD_l+VlH>!I>P}d)-ORPU+riN
zy022(@64Y2V^l`UH>dt0s_df01PR<p*rX~?<qkWqS6KZsH|-?<gs^V_?q|;h+On|@
zQJ_|_PlE~0&d2T!MqBo7=Sve}DG>)cQX$P^3w6CfsGKTFBd$pFNOX=ZDeTm>*WCPi
zByiSe76S+m;a3cVCHxu=i{Ri-WPUSh3)1Jq8rRpz%=rdeDGPlsS`3m_G1GwI7seHl
z79`|{6<jR&FuEr&B>us`LaHjg(jaAY2OM2yMv{1heEhY(@CbuQJ)d)ln60P_MeQ&d
zDcP_ilDZq72*!>;pPWn-fq)~Hrqjz$HyD>9cl)Ko-SKQQ6E8Y5Vo@{rq}syIR@1fU
zG3xLK59gdZ*CB2{e&sNiV$5CYwxui$A#0tI=NWN0Sv;!gyoj$YTOW%Ix^0Fz%~t}j
z=Frv(S{rLMS(_9Dn;%BRFwLk{m5BWOSR^B#t#l7#u5LiO!o)A(^m}==z8HRFjhm*V
z9hKTfSZEbkg?<R_H_ljRlp}FHr(ZZOM`?T?v^J0i#+X-E*gVxHn}xF;B_V*tK=ocu
z9OD`1*GSNEx-yAsG!0yV3o0sxfLi<la2_-U$M&Rol(&g4FqD(rXX_hkMs4HX7+#Y{
z!-ZR19-S?bZG7(p3M@kbr#us9{Zxk7<P;E{`OWmWJ=L{0%v=_$FHc@HPM4Xb4qhIN
zY9QDw+6*Y_jRdU$g_Gh$>G%llHE*r<d!1cFVY;YL;r!LI3y}88KbQe_;Pw-|y=Isl
zm&ebFv+axx%2Ur*MQ?A;j*B3Uw){ykC4);QIW#>G*q6nf`Zjm$oX!pz;dQoL9$qLI
zAZP(AU<mnh!GJs<-oc6|ilSh*fBa(@k@~8ax6Rx*j?BDttri?)&noe?x%m(4xSDc7
z`cfc7_!Fg(j0T6|S#31B?F)7$_o&hMem-q9kgpW#=Wx%uUrgg~g|5huM!GQ0L*hg2
z&*V^E5RNjf$rH5Chq+h2`3uvu&A}73JZCN_WM`R0QF+Ohx}>Pr94(w>LFZhCwKi{U
zBPCv2kY|R(gPQgm_v40+F?j(_^cF1D?jiyg&Lg7UsbzIn_q<U2aCVYF-1`jW-`E29
ziq$MR{254jEaGg>th(yQ2^9TU8ZZFv_upqPm*`}0RR1moGoTmh7_Ps=-~Gcc)xGaH
zfH=gOONj(vyqH(A?N!8};$R$RY*nse6+*4DM8cJsNHRicbt<pzLyN-k7&OswF|j1L
z4WkN!08^KxmGool``$AchDmH<LO<Y#D<k$8X=ZFEYT>MidL)r6Nk*IMk4s3l?q9*$
zmisE(7nY38TBF|Ga)qN$1nP9;-fyZADn5zc00tW>!`cPgBX%lD5><rQ6duOk5_^;?
znCgR`O|;bwxN;gu$vwXQcW{C<NL1m|@f^Z3#%Y%FTJY#u;k6&Y`N9O^dg4XrnJ(fq
z7-m^gel2tDK954Q>Bf_FALSQ`%;5_C$A8WEe!624l@Vc;)8))8kJpg(X=Iike(pbT
zJZ8Zz8{D~Ub&qPzUjdhMp4k$Oz12wXQU4f+*A%ZjDr(t5@J`fO+zInPIH>;dXYZ@J
z4*_U8DK$pCuRh)Gte6HVyKtIJxlF5Ubio80TZgY8@Q=u<j%!dw;u;TeKOW1A?Hmv4
zlrX1^H3^KS1#Fd%K=eN5&XL8y@<R@mp)F#=SLUuCoQzy5aKiV^SR%5-I;nrA=*C0Q
zF;is9ljgQ1hJufY>N^7XRMfw~L(NwWR|YSuoC4Kuuo&*5>u$pzdJ*O6rNHNY1w;n;
zL~A>!A~Y+3oQ<5!{iB&T6wag-3xom8deidlHpV0n{i5J{)-sF~8-{|s4R#OETt)vW
zs&O=wDxWh)WObm7Yuk_sbKziqbt@)sE?99bxZ9d;qnB!F5OW`5kdgF3F2pMKMD@O-
zBh@k{Rt4>Pny2wd&li1+OWi&Z?z?3o`8!xXAml@^Zh5C~tZOq;Z*OuPc0`KwFCDM<
z7{znU;QPMkBa)#QO75k1KHBoz@qqG{Is7;~Os6cx3J4>gJoV^1?b}(q4rwSKu~b7%
z++QTuMLGSSQizhVzVL-DeKiI-&G^;(1Zeqg@LlHZ{@$#w0{+W4NM6)cn|l7zkk^+l
znf7tsPW?JLyIG`L1MRP{(u@lTFmE*gR#TrWf7rbnIzOe2%;4*|opN_SU6%J%JfbPQ
z-9V+O!Tf-Vg(Qj|tcxh>hbpcGGhHU!4HLW}j>Ew*JMfMeHt5!LJ2C3y(^^I(%H29#
zm(2d?>j9qhFLB(E0EY6VU^*nhntmJwxDc9{Z)hd?G4e<$#xZhOEjuv^uvTWWf-AGW
zkw0^}b$Z=fr!y$7@`*b3{R1O?;py4X7;}?xA7-MI9=LscF?u6{xJFC5$Mgd^;teoN
z5vTzJW`i7wcHyRQcq#qHg&Yay-%OFXQwA)dij0Jj)tQ*urtFkU!}5(aSw>_D42>6&
zjz>6UOG`aMybfWvxZ|*qm;Qua_Q`Q;du&C?V?d|x17f8uL-3gw!9&0;YM%B@z$AY}
z)aenFhut+GYtq96{Svj>Ctyx#NaR369A1P3wI-RC1=HIVFlKIuSf)NgfHe&PFH|S8
zdu%hA@{RFqoT(c2v(@0zpU1uriKQ8dr7DS~+mG$xpKu5nTZ6(8we-=+OVfR7X4Hrs
zcN&S&9*Of{tbPEekvz4A7@*H1ejKl09FzOZbjue{Tk$q{l+m#u=O;skOj50)Mm~$`
zOWGdq{JgeK(;|XLj=rOcN#5^pjH;3`ixL#X!&hE58;vsEC|4<Uv3*5>R~X*j%o7+G
zSQyyXR~T%U74mRdG#?K9{+WwxO<;0XX~(N>P7{YTiZLe!s(QV7VJw5dR}iS-6RHzd
z+gDrr<tBuTAuqtm_Uk4*#1RP!^UvFHR2-A(Q+)UNws1iyU%FgEl&B{FpGXIvh>8t=
zR+N(#_n@#Xn?W=pwz2F1)ytRiy;r#2R`1Znmty+a@8U;>6Tp)cZf+tZVE{a^7bDIt
z6qkg70&7KxM9MCUxLzTogMnu7d(=`?l;;s#9UyVZYBM`pDphMdj7Y;iuwrZ8#IJBu
zUZ7>7+Bq(xh>uv#TIw%KmiDx7_V;lzbiYKcmb2;*tNQ9F82X2qoTjT3r}Z-LcM@Ur
zK#N@oCM+Sj`Fbc1-qa^^+P6+|zbI_`*5ETXp?*>os&-tl<qFv{*{A};-Jmaf;{ZbD
za(T`{akdge@&-hepc*NMYvjb;4Vap>hCz*Y^2gZ7T^mG<{?0Ao2kq!97UDx@(BO*N
zzQ>@L$@OfNtMM4rA13NADFg4Ef0KH{4h!%o{1d<3baz|jc|doUWg=z>EP@(7iX?16
zurP|5csO<_enQYd@ihqn{`Ot(!BUE5`Rav|2X!bwxOhP+G<jxDGf*>Qb(jy!gvDw<
zG&EVY9AbP&9k*=fL?-XOx8`u0fT=-|rr-Aa>15>d?V5Yn<>v|hTedL#VCj%=F&6oo
z8?M-_(HmScqDg{k>5O@u+(n7UbIg@Sf(|zrjSiqmV=ciwa3hsWh)~;RrXa=KUPu$Y
z2&y7dfzA2evk(b~99n3Vp+bI%zHdy{%7tqOCJ=uK!8@lhFWJf08?lp(Oj})pugunK
zD@tT+rMIv*%8&MmhuH;GP2+;PyXJ@3Bjs=5e~R$snBXr@?}mv813r^jsXq9x_2`gT
znV8m+YDz0t<SXkH!CPSl0~2#8Qz``$b8Eq^$;*Ao3IZ?fHMU0v{#9{(Dy>Q4hYv5;
zep!Ur=mlP*N;`JUC5|Q%vtflq6W9reUj|guLg~R}u3Ff$vhH+=Yzdh5fHgD+kwxVn
zKjh?ic;@YM@Iz%gm_IPY%ARL34=BRXi!fgKc5=YT=g1Mu2l(2Oq+|K(b71&#F%%uA
z*b-v+Z`q=1qb2q1av)m6Dtu9c_E^~pLW*|DZRyt#Zxx-{pI|6{QGUPW*rh(>c9F=8
z{~YhW<zVK^9)7ZdR#cpUtEy#^4VKuhPM7DGmN4PieMb<-HL5Io&1)LKZ_kABYxaGE
zL>1P8rv#JbvAH(?B^Yx?ipQu27XUh$*N@^ICMuoC{3J}5pP@!Gpgkg}MBT^z^J)^+
zw4!2oDm%Gu9H+9Fhd8t03;dZD*s2D>JZ+B3a#C(F|6_#;<pa0yL0zLqz%#q@*FV-g
zZv@trvxYoY6Mw~`1@ZHme)kyw)D~}6%qMlmJG2DI=6RPaE04(6Eu*Y?3BmhU#S4PF
z9!Js3y-o28l_P7v54$3N(XEz_tkEV`oA4g)#Kw?z&I3w3y|W@#ADd@A*Y-Qbtz0#&
zrsfJ4^i3zOfHf0gAiTpv%P_~aPwBz=s_yd-SZX@b<Dy!1Q)`|%<Eh3Kr>qK9#tT2E
zR{BM2dE=dQ2XUVMX5fvx6)cpimJF}(KBIjm;3sWmwSP$hl}RtW4fO4e`=L((F%3N7
zn|%7lQ(H+A%zCBCB<>AAqcAbWjc$sSSYpNnkUNP_o8D#n*@>)Uo$xpfqDTdX4@SLr
z&i)FYQlq&rmS;9cBfBHF%v5GfS}VCBv5-6Keo=drIB9my@75WasW(h|Q9u?g$@up=
zVcsx%4#}E%jj17kGH)EETnMQYfJ~25L=^KOF620D-l8`6VW1izv6ssQW5MC*vpVFy
zG9BaDfFT{0+Bhy<&M6C?+5lg<yM=>f!k?3aa8QIcJcub725-xw8K!hgG?`{{RDvT#
zNbj45*t#;%0rC7(#6=qid8P3y(`Xr^Y4<k`#ySSmKAt&mY-o3bBKl&_wJp2{r8NSw
zwV%L5N6LlQLw&FiYanekeqxW&8J`<V5dG#?7qrYk(@hN*EY<GIYFu3(A_IbU1k7rZ
zg>Fq3Kz;e%ky^_w`zFsFt)-7%kK`SXrQf0fcROr^*%|2)Tvq_yffuq(@97rn9lT|L
z)tQtF=?lVXcbnPY-mDgH?#I)dK5Os`>i0Vop+JRA?z>p`C*)2)c-xF{MZN8N5};nt
zp2e95W@c5-rK3wH(&XYWG4!Y=emfg?P>f)RUQs?Dm~{bH9Efg!tBnx1WjJMzsMdj7
z8JNO>4(cP54AN|-u+CnOnB%v_^IrVE@$HCBbm%`pPMrONy7C!Sp=b6C<FC5=|2mj`
z99E5{>xZO{X~EY>sLOk<;ZJ07CdyC2Fa*Erl1+vkzwK)eRE!*8t~Np>tm^x-tKN=O
zrm8=WR<7x*!c6?BY-I4OaY0gw;0UC6PO&NT;`3h?bRCbaKj=hy&3C;#&VodeeXblx
zgS~vkzs?mA@${+Ev5ABm0Mq|}NkVH8WI#iWPR)x>ZES!*DRK7w4I-(~sTno?&S)jE
z{2USe51eO(y=p)0{rN)l+HyBJ;LY~p@b3}cHhcH~xgs2u$KEHPjxCiLfK$sMN+>zp
zJN(_);8&YuFzO;GXH|kwGB^q49QT7R6>#h@VeBiUq%VcYCc&@U3bcl;^aIdL!{NW|
zg23O5LfD6^jyAT*sk8X?<2;@m*7VN1e;|2H;J^CCqm2?591*6^Rb$I(g6a6yp}sh~
z>XuuQX;w;~dJt0RmOJtxl`MB;3+q3c{#Exc*zyFK*$318cSxP~H)=J?Gatl>*bfNj
zb;~z5dD{Wo5n^r_;KIg~70~5PF}6k0szw0<Q`YP>FX(8=d$THqim6a@j$YEK7q&sM
zLf9*}(}0Cd?e}}m`NA8A9T<yF)EOhn`A|B}hR=sJPP_&6W}hlLdkF96Xwcm2|0-F4
z4z=1@amBEyp~59yriVs_azZ~bOwC#qjt9%MfS<V|2f}aDmQR#z4y^th*$TBIBY(pW
zp_2%!o~DMLu6D^hry7qzQi0}#H^4lno**z$ZCW!}%n;0O`9X@ej-)7HLng#J&xk`}
z<5PRpOC@h9!*bpL`)wIA?&Ea58*|VIjOVfElk0*(2W2AO@ed~YgS}t#7z#sAHMEo4
z8r(4rB34RickbMrFPPhbksmYv%!}+&J&sHJCF3zYk=<~`LbBxy$%HyvStY1h9nW$|
zvcdBoc~;sZaBVzHNL2*qB9(328o7<^ES%M+PTN{6?pBjw*9{A7gU*jU1qF{s1q3|}
zO^d8tvodnY!8ggtHn7}Us|UE^l<UL98>OmcO?<5@5dBuUP|$&Yd>83+yEHVYRS<{!
zIA}aAuR+6M!)hU2z$Vsib$D_eT$Y)2#bUy;w=1Tuh3S5<qB`3-4@rB3XfKeLWs?V`
zsdY32HkMauYzziV+RR-Hvc|cu2pH@Z^Wp5izcza)PoS)wDbBMNs5qSe>Ck9BI~=@d
zt%~J#yjkrSK35X_&>gJjm0d`4Lv4Po+Db7iz;E9j*>U`4dwa3uPxA%S!8kB^d)2}R
zxxqdVp+23ah;C(fs@(x4hlS%h-9g3F(NXpEPIY1exJ}2|N};?PN`KQ-G;dly84hww
zDB6aK<t?*ugIP=$EDhVzA5S<gixMNuu0Ou<)m%5pTLk{){r++O(u&MnDvQr7q_|0r
z@K`qfaSHO=Td-my>1IbW#bXU5Ibk&@^*BD?;aI9W3l2MxHO|hIyQTje>Fn&fJ!Vh~
zWq2}^ieT0Jwj4Pgj#Rn;xo#MKRxgD`w*s}TEm@GntXD#7Be*6Gcm+<Jv!A_;PFpPZ
zNK*U)fQR?HG6zr3_vY7~{YPs=Jj`x3tW_bG7)RqFyXF=V#Tl$Jv+mF){>c02v0@!6
z`llgrBbHqNd#dNeF@;XK<lFR>R^$l6n?;g!v$W#5<u8W<<?m!8f-!2m>Gviib;fut
zqjvU=MaFTmlxN8821~1Q>zdFu+$rlw|M|A_C;y>&`TFz1YIr<zF}%$57PJOIwo6M9
zOU4=fp}Hd*QI+JzYE?G#6s|3G&N$MP)k<t`1jy5Lj;!Be<3W#0C>X(sSdsKD*ziv!
zujRz&S}6e9(Kg<@k&=w{fL<&WQvs$RVAwr#=_`>9)U2-Wiptw=ux?78y47+QZ~>cN
z+4tBbIVq}PTGOZVtnZdmZb}ABCjQivYWl{eTuNUFy2+4Ox=e6gi3DXiyXkLC6oGQW
zxf+Dxwa=4JMORv<SfHLwbNlVq+%ugDhc(v<L)Yo|@<qswTndAN5boMrXwG_PM6p<7
zGNH{qYsYiELz6``>ri>Xh};i~+cn^Usn64D=j?atU3H8{@L{tC*RJ#_k%G0uC<JQ{
zv```@ExV}D)B;{8u+E9(t=rQ~b8Wp_9LcHTzYn$|YMcc5&)H<+Ia)zbk>{-&=w}aI
zAAiOEb}F(s^rX*9dyN_fxyP>dHLj*f)Mc2bB3feqSqHKN+DB+nTh+7gXRbj~gMxO%
z@2(ntFNW&TU9DP$q}_eVe~H+;d>EGs!dN<#(gq=D8bVpGCdajo<o*KqYW_f0P8266
zVAnZGp&T<55WD<{y2xqj;|)uxC@>IxcZMyG?a*1b|E;g@WTdw8GpFT(J%o465oj^S
zh1}1*2XnPzu~6V1^*nN&tSYikZPk6#eF^3Cb}ndiTkXGn&1hKNcw@+UmG}m%U_0qT
z7FxLHHnkbFPpgTo+v8k4g11>!p?l+}YPMgCYgyrSkXW_T<cC(KM5M=`x;X<BPMYd9
z#@qQU4jXMklRq8H)YQ3hXC}{Y4;e*(@|KQgKx{_JK+g)_0_=ivp7tiTq46{+{W^G!
z!NusNtn%4O`5Ogq0_FG5J5~0X;c?R$+V0`ua;xn62s2%DhXdDPS5*`(3tyF(JoDKX
z1vDPUVE83^?MpG@XLo)A;+<~v0aPaOV!Pq}x9FxT7c785r{?hJI>yPE7%my6oi(zD
zFDJpQg|AJ%*$Eq}XmXSu23F?3Z-4Cv6uBz}&V=Qt(&TgvCe3;GXd0l*++2}8aLF^7
zv@0)HYyyr86spyHYjheXgMSfRn?`F}dD-QB=_o&Wa7ch;C;9!ZcN)05r)mF$e|jy~
zruhYE%eC>>pD1uMAjT!fj{zX}@rs@zFmD^+;tHdSwY}@he>L{#P0{*-A%pYvit0M=
zaWl;4v`eYXy)^wBB@kd<(6#T+Zun<D&i80vPv&NFE3E?Vk~zs|gv(|<b3etQ9HCnA
zivvCRWq?a5Qr5Qw^E(Ft3IJVQbe&10w(Q~I4$_fl-dcFYsT7k*%*S`g56(%myDF~!
zjdH<8UY@RT_t(4Hmxp7n(8Z*t5$k*9gI#Xp9&gyuyO)x;WWYa}g|{_Flid1`oPJz+
T`bREKhfcx3YL%@h!od6wNPH<r

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-bold-italic.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-bold-italic.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..cec2dc94fbb58a50a4400bf32fa2334d474811a6
GIT binary patch
literal 31812
zcmV)7K*zs#Pew8T0RR910DMFM6951J0ck)00DIE_0{}t*00000000000000000000
z0000#Mn+Uk92!&`frlm>giHot0E1EpN(qEz5eN!`(`<seD*-kFBm;*i3xiew1Rw>0
z(j*6kTnvFeKX+wfW(HeZS#bs3wqFs=D7{(beK~jIIai<@+fnXTfrUaaE5^1gc(;QD
zd6m{6`~Uy{)TAO~HmRiD1^|F^b#kr<5}B4(R1*SZsg^L-Vr6O#QZ_HvR+Xk=V@|NK
zwe2hp(Nlw&=<*D9BwdEEv3yZEu9^HbDp7%m9L-bg@RdQvgEXAe-Ur!_lZ6%gU{pGq
z9{q)H|G{WIEwRF_ux9;;SGT}X(Cl7MvMlozt5f<3LMW5OB%<6RQ3^o_t_YdPrch#R
zH%wDidM4;WtMF+fMEy|lx3g`Ir@N5Lwfa~1N8wKHc*H4kiidmmc9@BV^rls`i1xuO
zp&Z*$qM@an^Yr!M`ECBW?>$mlEJVdfh9LHo|2M=aR)vc!BXZrYU1Ym(u3U%Crz_tX
z4S`0)E3KCB@4?zfu|K6s%B1Q6pkrzZm$hH&gy8%A9^C`?eNX{%VI3jk2vlql9I>t?
z(sjVPQxB}_ND^LurtkM#-=?BWB+J{s4iCpMn;94ZlgLF|zTW`Mtr#$B)F_FPiUDH`
z7_~Laf{nq}L=HrRk&+`6+5!Q^OteqVL&55kQ;g(Z<vlP60RH*C;tomrpy5+N{?<iD
z2gPaFE>B;#3mT6!V7fmYL4XVX7XDwxe%~C|b(IHXLKfAFNeN<&Ol0Se%DPzv3R=?u
zv84^6i;!}^SC(#k%cr)iPno7o(-D-ZE&yC%Meu;(3TPj3<pAyTYC<c^dbwC>o-Awh
znj`<y-?!w8SaxS_Hu%e9TkyB;397wW1rvs<(t}H{Qk4sf<60SSKx&(M(||%-0treo
zReK#+w0V3>doxAhi)^ck!#yKKd){MLj^F)AqJ9&g0kkx97&+Q3^I<J@H{p?FC=vVu
z_{vD5J)OO__nb>HKzr8bu4VfhA}pd-YT310{`S=pwXG?^XxM1Yf+q36e|VnXZyu(4
zF;8Ub2x;?pCjQzFX<iy>Vu?h|Kon~#MT`Ytp(Erb+&0M98K5uYgRliuofg3Vsj7MA
zu>E$Ep#ZYn(>AA7A+o-V$0GXzi6nR@6B)<%AKmib5r{YmoYnl-EW^6dZSSx=`2RBV
zwFl09^sM(7*xm1%q?}0=MA<#B-vU((A}L6iWO4>Y9h*{7dWK|%l+9ji+ru4PL!VIU
zmBT26f2h`^c1dC*V(h`g%YzeERQ0sVy}I74mHc<3uYEfsFMN9mKY5Z-XaZQLYEm`S
z5>(+1?ymPSIY(49S8+7}@<83E{Br98E?5>UH*)eV$4$M)PrzDOQd?Z?lw$qFWy;^1
zYL&j<4UoS%mK4|7XwR9-<<3SviY&hAZngv12E=p&qy~o403aoyo1~TirD=emXr#HL
zxQ;rz#u(j11Av+vnl;X(cN?Db;aw735^rs^n{5zYkQy7`{OdGVD>)Z=1YE{LT=u?Q
z`uh6L&$sNvPWmD_J|9rRK6bAlHpqPB<dPyxgz=$MK-dm=D*#PHcR8h8r@l_zM&8?T
z>%4Ye>woS2|E0<0Das=;jx0UR`irn@R}*q`h>ALONS74$-re`SK7B7C<OwQd`wBg3
zgc|3t>dz>Y90J^B92^LgwiX2~Z+i@X5Ay2K!_R|H_rrTK|F6YD5M!3jnpu@1ld(V5
zjD^ssFJpwp7-144gb>0AV?2AkcHW<>76d!HPD5uzr6r_v5=cPz?o;1=-A}mb|ASCM
zjc)rvTA(537dv}ru=$U_X$v#^I$>r<+n^GAlEA(BpWn~6^nT8=mO}|41Q8Jt@p-LH
z(|H|D8X$iKfB<mwSlxaAxc1Mte*w7jNx|<e5de|^6h4B3;OdmZ6NiXT83_^*Nhu?l
z8PVxFqqFCT7g|tw0jQlG_>6Bw<j(7+jt7?ywuD>){gM`CgJ6Z>J%o=?@%|n^(AEY4
z<c^yrZ`ojW#>DJrh&E3??taD`HwkzLcEotSO>>QkSjC*08;y3+vkJ2{`{dK(Cta4V
z%r&%TGo{AG=Z^F4#s6qDx2pNff9~B4jPX6c7hi|3#s|gn*kN+~Voj+vz41-UXjjdf
zg;Q@0{W29P``70cF4C_kj&fCH-6d`r-+$d<t4v5*Zj_#uKQ$wP*NKu4F&hU^tP6DG
z(`}g|m1NKBN4d)?E~ths%5F@PxbYd3FleCx8g0587y@f`u+{);4YYPaW#C$>ma6@$
zs8z=e4}9=50R$037!gDfLo!mBR3?o{XEK;fCX2~ta+qAE0EH+*F-p(^Lo02x(}}bV
zksf;Kqu&geAv0pe7-yPUmdy(5Y;$q?%sXFv^C$bxMck(3rS?*W3!$P=a)*@zmBXr$
znk-3VVFL~?&Ks&g6=lKLvx)V8^*rqtgGehHt|n|*(%Hv8Y#L6MpzW6amZo`&fw#SZ
zt~|8fJUp>_yVov;UT*;N?orxqY2y5gW<mjcY*P>sy|MzY9%b2E(klEl1SV5<iZwVY
zL?4wcAzzpY0&Y57tWIQ*sy^PX&BazsAsI!5u#d<nIq@j~$<nRE7br<U7#Z&ki30nW
zjCVlN%_A|otMSMCUs1DsR<cw^W$w(=Z?kkpw!n)cA6am|7cyJs{CiM%{~Ayc@fflH
ziF&X<kcXZ5hl-JSP^O?H2`dX`%RqwYxh+cGJuI<Bp%TRuJ**^UC~YpvqS-P2)i;Kd
zYQQhk-v2oiLu4GIaRwP@sL(SepFSNFTQmm1CY+UsIpcF<q+l{kgb_NV7**Ea7-J1E
z!9;`1G|Lck)lY^=S7*3|mYHvb9kT4wpvHa`7CBP3*fA$9amynsJkey6XKKBrZ~WYf
zj8_8w-d6V<Tn)O{ePSNuG~`0(LgDj{_OnUEd@FLUmO(C)To$=(ayjI3$+hq=PntRP
zh#ezwjMOnsj&XLZ(4e4@u!yJ_vEmdjh?1<Rnr@iZ>uw#~_6}d`i{5O;QnO;ij-%$p
zRdcJ<)AQDR_4^4BBt)19QDVeNkR(OAmm#Ybl%i5nR#A;hvTQ|7T|-k#TSr$<-@wqw
z*woD2qSE!Pw$@l@gH5*BW`|uA_Solu!`cxXJE`ZCGtRl-k}Iyc<&JwEc;u<4s@rO!
zW3*n>SM^POS3lJ+^;i9``HGjyD=j5Gc+8(X8F;bqX5gdvrhY8c4BDldd4fL!EuDb8
zo<$&oAQr(4LX=Qp!nGG6vchXLU-6JAnOUdUm5c6rro9;S;j3@H9|H?tDdaNdf0;^_
zY&mk%d=>=^3Kc0<qEwl36)IyDi)uA$)ma!9F<5MgrIy7VT<o&j9(y&|ACK^LEPtHE
z2`)}Dc){08<&~DNX*8fVtWQ-X?*W>Z%QLrp<clg00D1rb0000006?2jYKw?y)*2K5
z00000000000L&U9nzf=(n`t1TCpo*8d|0v)ExMdJJsgxvR(7;xhhJ%8N7=P!-@)o|
zc}XErcSs~jk|arzBx%o{6MOe$JC`I$k|dHONs=T<k|ghSPqm%h-qW<5CIes4YzwN|
z&r*1ZwYWLDw(8>cZR)>03w1cZfA_fEeWL26);v=U_q~pb2gZbE5mc$Ez5iy0iu$Hv
znCRMj{;Gu)H0+9ht+8kd1?XIB+!S{5QkL*I#VkZ|Sj05^4JU7ljTOoM%MT@Xn4HKg
zFOc^st8V<x+PD;=e8@;#y8t7`G|lCY^m>b}>WFVdQRdFdMSVq|i|A+hZR69xgdgs^
zeJ)7Zalc83Y2(18_@3Xdhq~1xN8|3CM}v<ejUw5&mr}yF$*Q>vCw`U|)ia5jmQ8KN
z%B`%mh*xjM3efgotSqu%SG(U`5Zy!=%mw_^FRVsSCR~jcY@KOx!?!gqF$;d4x@h*a
zn<yu(m9aJOBm$60m^Kd$*7C_i$7XlbP<cIB@moJZiYL>7_+Y?a?U%j#wTcZ)&dVJu
z!=BU}<69Sk)_2BppQox4C%#NndPiR_rzRw|aePg4WUQ2~-Yry=I5WFxzy7)ZWr-P)
zNed>}V(^jn5L{c2aEE)HNu>jd`{Yc8eFjJ>64e_`-kd$SmNgkxVuGVc-f8JLs;)j;
z*Sl}YQpkmm<ZcAv1}C&#0~m3{l88Q{<j^DJXx!=Z=pi3RdZcPA$?C$k+q><_1Phwc
zn41DSiy<wcgG)pnXO}$zg1WDtGeT$fHFO4qMZkIHg=<(x>)L<wUeARnpx4Z!_k2pd
znK#;<(JMfoVfVFR2;+b6i^qv9iGjfDf`sV<1J+DBGY7&ngwI;cTCi_gaudFxZ!5?x
zT|lmxN5AzULM;Z9E_q!EYOqL0hCM3iXF)PqJ_)dpP?#KN*v}%5C346$jPi$?b+uOP
z-l`%rTnvY{Uw`k?ZIJO)1!pg5SHK5V7?iy4q|61DUaSq!{`578Lk{0uqIq<U4>3>Y
zo)3@|{%ha=Qh5x7ntcYNWk82QVXIK08`f-776Ds`tmftsV2ex!#9~hf-a$>S-DeaD
zrUuS{93-*~(0;tX9R8pjcw0VR7Q8uFq61T}zf0`wr1ZpccRq6fqa}XFhxCr9*to~2
z%OV%GC_yI6*#H}Yt*WVA@GOU6kz_?LTd7Ea$;u5xlO#uzm4!^V5UWyRTU211YirO`
z>!3?7R)!}NS?xm7V?+vTMTU2=deX%k6v~^wTN%SvL{F4SBS`B3EJ?G_Tu%d6hjVvA
z;bd===p;vxbmcHhzn#Q4J^$S@$;)eI3^!sAC8~-GIqL`)ZPf1@DZpxMSy8bDGM{7V
zO+AAoVnl%kU@kyNbJfV*r4?uIqsYv`QZ<*naMLV3L?_o{c<WGVg}o&-_e0S0Rd6mL
zPkV+zY99(<>ymZWLKtBATxe_<Qa`5Dg2Vt$<?I!~Ry)Nt+Z30T6FNb4u{A`LzR_gT
z1qlJlKr!?lQy25Ut)8^29$)t8>{trsZ{4o{Jmc0x#5{B>2wjw@mkw)*MVwpB08iV>
zqWbl&%m*+zG;%WH@azpL9jB==cjYmpE>(kJuD$Sse?8T@I7DF1nt#KL%*3CAM-L^`
zP5~o_2oQ@|358c~u>G64<Sb6cR9SVL<6>?97G`6b?BMvlKi8%~z&J*&Ocw@puKwKR
zE$nE`&c_}gT|-tG(#p(C`|mGLjt4zyii1mQ_&T6|6Z|0_W-_&!oG-28>0c9@?DaO>
zUxtb1rHk)?!+LfalgYSfn}&$t?#-M>R#z1*EU#<q=bW3I6Qy`W+4au4_G_NAhBpLD
zY{xK<M9i7Vb%~n5vq`zhqoaTmbKm@KzZ~5@a*rnZ40z*W4Kj-CVc(&1=8`42HgTW4
z6XWC6LWo_}j+R=DtKolh1ck<6VdLQb^F(Yleg#4zViHm^*5qv1vSZJ|v?c{dN`MpR
zrv+Z+K5JdM0o-};1bFf0!<QeHs1`RdD2%pq2`wK1!|-zO_Oc}qoG2LrO~jQD$^P;x
z@g+aQDT<C`%&DalKyQ73Izz|J11<=q+arN10RRVy7!%Ku{Z#`%f+RG618~K|Wrfd)
zfQTQgm?sH0LS7vC@aD%?J8BwI{=<7!%b7J98*+ARkJNd9Aksm*aR3>>3n2i|x#I=0
zWpto~(C$s7@(~D16@muCbe1N4)Kceif?%V-ZjG#1Ub;?^6YoOHeSew<0~V7QF5)C8
z%}iBKU3G)sL$9&1j}&SXD5{g8PQX98TpnkoEzEYn==%>Z#eU*IaS$3@Z8Y2?{9QR7
z0r2&;_CM$l!{!Nq)VVTO>waJ_?vVEK)zMvER{MVc`pz?6-YUAi{N|fi!U|#llyz#}
z4dU_y%_8&RP`G3i-1yOpWYXDtc?<@Hn7qL%;@d6622j?!NXFB}LfS_XhkiUHObCAj
z7lR*tkV*?$yiQ)Z{gb>z>Es1U5`s{kP_Dl~VH2(8Yj|)FMSwQPRMQT@n_L&3%V~rH
zAv^pB!$D9?x}p%)ln_3d?M8&Uikb(Qa<!CIZ;{kmvJVTDF)bml{0_mzG{PRK`aD5L
zx8~4;DSVW%=88R-ZdD${!i>Pa)Tf2+(^Hq1_+|u9xWxi^nAc$26)@*bU!~Yn09>j5
zNl~-{;BKk0hk!}|$S21nfC?kkK=SJVa8ceMK#5Tg@LBWa=bEpLWX#OdN8=Y?wMV5;
zIjh`MUMj7sy=s7Jv#XEi-~Y(}R{&I&N7wQDLbXF>uTl<&11I+a{}K^zdZqAJ^4CvK
z&OAB(<VX(ZKpIk?b;(TV<M{m+*7MfIaE2`S-Iq=hxYqywMSyM=n=^Y#8|!?CDVDW*
zquFYAy1o8jI2uo;v-x7VT5s;E`-jJ;=a=nHy092{Y+O7dAu%aA1&KmqusA#+l}I8}
zs5Cl*$zpT3Jib7fmM#)!NHVhwat>SCx(Bw7ZXeq*zH`@}-Fx>P*njZQkz+@XpFDA9
z^6WW-;-L!7v+|L`ocXeByFlIxm;xj2Q-7GpK8*DrrVkkp0OKBcBz-cw_V|fgx9>l=
zb9di5_7R%DYkKh#v%YzXtfw-I3rfq%OR8&7_0p0>IQ8))z-$1RYPebnMYEr1C7Q^>
zE+X{TQ%sn{CkVcgrwEG<5fD5Q5>derNd$(*RWHLYwBZ&W;lQ-14KIReu<{5-Ck=*=
zSunM^ftY7D+U3V~*3uX7^6G{UuIU%R^AXY`)22^ZsYeLw!wo_(X{7L1&H9LFxE@l;
zQ|52ZlNBI`n#G)dE0jdkGw45NOH(Ge9*F8}ddp~t>DwKAiE<-d-nLDEp_quj#U;Z*
z|Aa0TX>F&Eq^7e3w0H3bL<lb(QiRB|VX#T<_PgD5imnk-TH!7d)0CK*;;dK@1&OVk
z*7Qk~i3}e?C*s~VLo}fJcOM}``}V+75Iswkqyx#(fa~9_963cwSBPx{D<T`c3J~ls
z5O5<$W%<T$jUp-Nn^OUSm>(jrIlc_pW<IYKeIwjckr|m065JXm%*8{>DRM)XzPmMF
z(LC@`Fab(>k$HOBsubxipHSzfT~g50FKgmlzofFB!~2wrgC@-e@v8FBioVN&IOy)?
zBaaGfNc2TpI&&uKF>TZy0#n^01x~cSPcUunFk(?z_Z?0=S_;~W=KeN1epREwS74t-
z9~1~b5cm!s!fOD?=RlqYhc^Hp|9L5(f?jV8WrKS2X{?W@8<VJ~jwx0elsD*RK*EfG
zjc>#R%v%DwSI0x(gV~6|<M0&*Jt-^wmM+8YJeuYasqD$>943>ChOT28D;ka?KDj}c
zyz{t#D8X{J3DixXvCI%$<Ao4dA70Q-YMiCjC0OIar~hk_MP2S$XH99OOeDvc>+#WY
z6boKDt>skXz)&Nj$}18V)QTs}XaR)zChw8^Di2&Hl3m2uy~>oqn>3n0IqlENd4I=a
zdBUec=!XA0&t(`t)JBFnYn6j24z*=m(V1sli^RErYJ;+dD%A5%@mis|<`{8Arv~w&
zg-Rp?q#fD-(;;##MWa~}$5IAH7Eu=OHpQIljMTYgh*=ckiIf5d;`J!_$Z?~qte{46
zxe6G)m13C*kqMmf!!LzIR{+b_Pm73+Z?MlPD>9!n5=)tJo#MB<$5Sa|>{KjeB6=f)
zl)PYOLLAlFna_f=lbZ7_cR0f;!5ft=^FgSc+2&f87|p2>`iiLQ^~QGQw%S=?@p91s
zF@jharhj$8B0z>*gHgqE;s{ozkme(Q==O%jpzIq<3=65f(#5Z*CE5=>f<WHZznn-*
z0N2L)LhP?5eIqT)dm2H~+u<XCr+QKT`gp+2UxR5P?_&?LAhw;O6DCnfI+SaNvWXv3
zsHOEn28dvscT(JM8s7(xl63f`!jZ_W|D6tz_VTx_sHFI9^qZzH3fk<-E=q1S5--kl
z*F{N(xdaV~p<;-Td<Y+kTLa#*3W0G*4vQz4xH+Q6RU_alR^+VGnC|5^j780_T$b(*
zaJ#29GjncPZ@BTsaA2a|hzjn%bihw;Yt|4uM5B$Q59clK$Mqrwaz&n+c-$8-v`~~8
zg62kv6jC0m6nN1CX)|(k;PS749}Y_fN%mtjaOABz!dIXhYn!MQht9dlKKAf2JIJzK
zto?odqbFEk>gPZG@ZruiM}u?;Kpf6=3*ADbsf`-!oSz%JmtA+jz|Q<%_fKtN8+AmT
z5mBq`X~h<bv$!j+4g2$HBT$e&YeJYh0NqWSqRTgI*k^;Ed2du2WckJ8aA`dC1-DDT
zXOUUi$<^$Ee)mnZP-zoM2Ml~SQviRO`Dz)05;b1(!?Pu3g_vm+HVcPB`-lR2VuR8|
z3+SWT1I|UNs3i{icUbM0#E;^7ydSPZ6foPWn^0(u&H9Y=RvHOOQ=!X0b>gPE%p$?I
zOba7&Za7py<jSS&aO0VYOG^5b*+Mbz8U8Gk8!2*I)P7HLn^Qe`A+^j@RZO()-}F0+
z8CIG*Fy^@?XLg+m-QMY8VyB)Ce&D+mO_<XBgqPwhy<a^eGLAf>yq|T5xVG<I<_bB#
zH4IpDq2??%m<FYyTthnOB;(nB=mB|hB439&EPPfZRk?#w#eVn7H8()E$y|s!Xi98z
z#A_ll?fWH#<#BVZ$yP-I3J~eSsJ#QE{z3FU%h)Sd#(v_==;Z0qb;M=Si<du)hhR?f
z3Z_QwOPg6p{kPk!{I#vZue?Y2s3Q(ZlI;@Rf;k(hLSc!?a!y~&viXn!aZN+Hvr&A$
zz5YR3sAg+bJDHN1lTeiX5gGC_Pq7i`qp`{6^41WjPi9i>9z22Cv`w+r!`{X20ZLr|
zu*JFM(8R8AYda<fPc+0xs1*S$m(1k<k<oZquMUA0h3~&7$`-A=dAJZyhqhMtty@`D
zRJwTybRlmuRF5TnfOZvT`p80dPJPi!q0aDhBOwc0Y`Ct1g)Ni0;%;ETx6L@G5z|MF
z*94SKwT}W(^2v=t#hI#s+_DO?=I3t7C1vVe@=5elS%jepvLWu)v{hv#mdP@y$G*{!
zlM3s|DW7-Ha4cKV_MDxP+-^%F+&%+WOHVYSl2o;(;~x^tKS({TbcEOo&V`&>ah^aE
z^33>=$EC=ENakY13n}Df<PW5tGJQU=q9(y}6j!-%aggIFiAZRi#S(KPd=bmpWsZy&
zgaLQi?N-9cA;!lthoiITCzVm@76ijO9BH!pw@`&Z8)Lnd6>AXTd}tCs^DrH*LxH@G
zi|};80#<G8<qP-Wk_{9s$*?_`Bo<MWs2reB5yoz~SIWp?62z5gXDR#6V8~HOuDMBu
zi(x(AtcU0OiaobKn)H$PUy&en@+=3{G=1Rg(z|>6h~tib2@~c#mqqK-0X6$h-HVb<
z?>*1df3ef2r}s?T_D$~(xE?gYBQwoM#(KT>e}MHzGN*rW`|*>yM@{W7AJjj7s-J`D
zq(K!FT4Nm(p3U;g5O7n}=MQ@zI@z&e#Zm=m18}<EG;)UK%8p3lUaXW_hrv(~zJ>yG
zwc(7usxT^%`{qHGnx&mKZ7PS(IDMJKPzL`05iqo!&W6v!*|W=tKZEE;t=qi$;QEW(
z+@{9_w$t&%e?`DU)c1b)yz!AVZ#_hwwyn;qf)&FG3k@-X$NPHK9cI}_WzBgCb~dI}
z^V_I+h^RPYZb}|+CreolY3*<0D2o+!<E)wc3R;#&|3}oQDl|htAjQnJt{T{IbeB8k
zwmCH7uTM`KH_tD`G5{&MV&lgBw)Cl6I&|plHo9LgSV+(5wzeoRhw{DMxhP3JS_yk<
zh|D45q1R<uMf@c;Zi!Fk`|S*r7a<ueDUQ=9)P=eUo4?sDCg}NMsdQD4yKNrHgLV`7
z-=(UT>=6l^llF8%tW9m7uk|0PRal8UAF^lZ8IsdIPm#wjr7<(W&y{s|W=Xr(FNQ4Q
z5?FVgxMpZZ`)ti|wC5_kW@M!eT6?^8tMsZ2?MO@ARaxoA=@nI##t+RRB#G%_2)$VW
zC`{{eVRAFcjjWy@^)EPlJ`(UH9Z{qv#I{^LK)tp|zI56V5rMO_5rl>{b`GUP;Zj*{
zHO{$DqPs7B`lTchp%gooPgj5abr(&>oTpF}u<RHEgf^|(9gGfsvSiCaP~CA6T*YaH
zbuS!*Z&dyoX@gN2NLdG$!u`)D!x|~0c5>#;W&j<hL)m4JxFgyIHg{-4jGGC-)usCO
z&^<<v;_*Np3|T}uA01$ym=UWX5AH=*K9a6<C9&LpJOOYH>04>$=yB#;It21(P8Rpc
zOFdXxP-W+zCW+q;A^7rv2iD2GW{(Hi3UEHIo++Xv=d*Z{iCxWlfHPgl(<1z6O|Nab
zYEB@*ggl01l(raZm9y+2eZ5*sKAt*LiMhMzB%p5lUR^!oNlrgpMzQZ{IMgVBrmB9G
zoz+?p+XWgzROy?lUIOxtp><uEs-+rTHQSoh2L}e(#&^Y*-hnw*D_J^%Rc0g736obl
zqxxHYQR(rj;#g^VcY{o2T=g62wG_GqujDO1Sso*AQ{u3F#5`%sGDG{8DJDz8%&d}-
z<!+XJF8dyKP&~J9zwq3)ZIN-u9z&XY=2VQFc?H7$nL3fOxC_eCtco=@+J)Dk*!Mz{
z^57n3b(c9ta(S{O5|oY2Dj~yQ`219~v2VwM6w1&(B_7#2b_)5Lg7W!HMq(;Zw`PDy
zo0UqV1fHqKHZQi{fVT11&=FDa6jRe`cNKO7Q#U3WE}6*k{wVkaGcZ#JN$#NLWNcN7
zX$+NZU3aX$)#UQp+$fvT(GX)_EoF%xzy83}amTpvbqCzZ{l{vQfQ36{izgPMcFO-t
z;`>!dx^Sy>@hD&+{7B96?lfO*JML27I<PO^y-w&<o!bFb`8d)-naMGD9|`a*`|S+?
z+7x%Oojz9`n1&S^iBl^=C00mx8Is3`Xz@oJ-aik2#eC73{%f_?DZ0>4xrO_6yAy$j
z@m$Hd3S3ff^#O@_xx$lGn)v18NwMzUd*JDdT?RuB!_9}EwCowYu=;ipjzURJm9;8x
zspLeOY$LSr@cHigF1Ixue)?wjrVFd?mSDAwB@*SWax9sgK#_G%fc)~W9P0L=L~!+-
zVS^WBpTw4TW(YD`T2tv}8S$xteT#9^H{BhSnv(l6L-TO0dg9?8pIZbqpIT*DRizp0
zX0yC9bN2w@CJGTk@e^baqG_bVctfeE)_7D)T_Q@n<KD${8}SaM&$e7x-y)0{zu!N8
zY+K`W@=sCHgTCo!tMU7NA|{%Dbv5l7-of6AAV!RB{W{6~Q)J$iZEedPy+0}8=D9;F
zqkYG^ksPx1!iMH2+8!w5U=UD;@pEVU0*u)4^SPfs??<~w0({Dz%?~qz_>;bUyB`{q
z5Bz`k6caIPP`%dyy6paMN&pK9r}{t%XrzZVEkA>uS<@jy#a8eoT25hnj~k-+XVG6b
z*G-&k1z(_+b2%C%pO>SGvFZ;dz%4)iXA%R*6wmoM^a6iE+qi=!%O5Ao8zc8+lWYe9
z#^d`0O?lF81TSwqRbCgh*IoEw11@_2{>1eH>O*IoNEO>3aMY&DWHXxXZCX$X_BV+T
z+&ROt09wt~x8vf@_DShg*-7X`mJ>0EAg2%;^qGloCM)hSHU?JEiA@5G|2k5By1ZIl
zl}?Sdt9A7N9O9;KVo<3(kB9n{onlif$wEt_el(@d%8A}sF_q<7K9zqI7s{zLS56fc
z(uFKl${ndBRlATx|G>}`0BrigY2~(<bPpUA5jM&65?XeJi%dMPzHyMpq(zdGB(bD*
zo2an&lNpoog?2?@Tue+3nwC<LODm1QSk*HpYMQHCj?eXKpBv*D0+>x-DQLW@2zosb
z7h)Cf_v2}V4Ae)MDT@@{1p-Z}l}b?Qtaa|xa)Imgxed>~ek`L~QrH6l4K%u5JYxrV
zp8e%NM%;;<=v&o~F7AcZT$Cq&Jck|n&Cco6(Y*J}mU`_!PxWnu`$?L*n4J#DEMW;P
zUAZxn&Et^@a4`E+>E#g2&#pd81@_NF8l1dsAf|`{E>!`Wo6Ox-S$*tU&njl0PcxC#
znD!A&nJ(KRWo&B97>yJnb#^t8S*b}{?=ofqi^<Xmn8AVARM^%+qKBj)_gz55^Zev~
zdK*E=N+qflwS+&;6E-c|)KWGy?Yp?NzU1#G&22^dNZEDp2|3xNCvj_fXZ!jCss(bJ
zwe<-(M<6CE$uW+p6lQG<u|w+Zfj&F;YZM%IwrX#VssOnoSH<J2l@n?OpGOU2ZOqtn
zz;`G*z4BT#Q@ix_zT>yM%V2t|`3FWmNH}`8-26SmRbsmQ?XKhE{iWzE?KN<I2Q@o`
z&QMK{bgy90_pd6%i(Rf*=cRIEgoOlRqG;A(W7Tdb5Yf48&7ZW|v4Z+dTRVbuzB-Dn
z>7hEPJLjq6YqGRhys)bV_GAvae)lF9>Agcbi>*1t`|BFF;Q6X{o;xe7HU+hA_|(wM
zH`9X&;vjH{3gj>pE@7#MUmk5K{%~%~|9fZl+pfKIt=OTjQ?{~ya(P^tZheR|r7NoI
zVZYgnGU-(sB^{A+5KSB;@bZRl4~K9K$|G!NCU=|ngEl_aXxQ_xP@DDm-@PMBA&khw
zly!|)28w-D8-z&z9G<uVHRY$<_x)&d9*$1XVnzFdW-0*=7}n|GO~SD9q-VObNA=5{
zEAQVS#(CbP?{1lzCb?L#_u{s(&T)K<&B5gB(b>sK*q@3W_nvF_E^lRwC%Y-!HPw@6
zn_Ou7a*Vd0`~LB9|Ec~_K#{b~O-_^nb0CYqp+>*`m(+To=63z)t^l#1rf;AYZ^Uz%
z=>hAI|J0p3q5o;FeEE==;yKGap0@{|TNOJlH8QU4ArPeP<WrnNW+Pw#PTj~<@~exX
z&3nyhNP=+#{`XV-DdzhG_`yHEXZ_i%DX9>b_J<lGDgBi;p(+aJ9g19q>#{HkkG2>c
z8GPs87guxVa1oGq12YP42o70Wrmbu(i-#KKKs(W`lLFuO&M+{}*lxGspTPRf?y}vt
zgpNqCsIPZ7+ByIz+iv-F*TGO5<JQTx(KPcC2ExW)kmux^W<F}JPKRtMaJqEWNjqFW
z-+6bBbgt$0VF7pQ?uBj7L;Hu?IWTdmPc*>GXpf%Y4edJ7N3MxW%Rp9oHe=~N{NsPZ
ze>aUH9_nm-lhkttfG;j6+^nmu5z)1o`8j3jkj;p9F8x^keMfdl7&5XA@Rn-2n+6wc
zx1^rJUMAk#x>ygL;Dp*Vv^GClxjnY+^xeAP>9=XS4t0T;<E4Y$fUW`d^w3ah46T$Y
zZtHSqBHUH6-|$`XbCcsE#VK9!L=O!@&4-th&$>Q*-1$3#8xrp(iF%W=Uy=fgOpEJ7
zP(6y0Gue@O*lYgt-<zaprMc`g46KEM`z_g@@gFY^N!A?WRH(lh(|TV*^q+m3qpfn%
znqOXxHGdfa`5yZHe@0G^`YeFpu}3ed>Ssnk9DeF`8t*a<L*}`3SE-EdR5`8<KFeQW
zki=j%xLw|Wm*W%aVk<RAM%W{z5G}e@ed(C&SOCZ$v~|A*{Tt#b^T>FSB7U3U7IZ)A
zj#nwv+}`J5REgU<7zk1@YG4|^@7;0(plJB5-mG*h*sMRqMr1X4Q~B;d!{09lpym#t
z2|D*(x&`r5<AaT^bxY*eoZku?#o@G?ifi7WPl@458Fxdx8uXeF;S&e7g}P6v+P7g{
z|0Ot22gu9zhbbBHMa?#p$k#9%$5Dqa#m*n~&oG#m9B586tnS@f-nS)kFk&g1^~3!{
z5U>hvyVz=UQF(Qh$I<)#e12gij`qH&9+Uh!6)VhbnaYw%ib-TjTpN63Q<tG}&B|x1
zB$=N&gw?z%d2CCnrYbK-p^FTB-I22QiE<$c-Z;@|wOqZhs3?v-nc`~ust9389$phq
z@4NaO^H?e_CZ*C4+z}#<ODxWe@7o98IoP#ARWCbrw|=xDOf3f53}})nn(MIvpHH9g
zN&Dowz?H-R9zYQ($oSOUqJprqrdOxKoU;87Up#zbyp)pvwMQ6P>0Du#^@a6AoU=2b
z`3S#ZA*oZh&Tc*dn}n~<*Vt2nSS`#V-Jx3ps|=Qp;o_Xl7S1_D1Pp(;I+MzA3fTXV
zyjeb4JH^%27n`cpUE6+kTLG}m_dbEyz<An#hJ@0iWOjy`)=ca1k#U&07f31-mnzf<
zQDi!fAj)2eT4g)VL%QZAjg1V|w{I-9H$$#1PgJ{BITo!IvBl+uXI6n^`;o4o%YT{p
z)i7ByzJQ;}XECL`{e@FOS;5L4_Z~yvJy7##9@qjFBV8fAo|`>-{VsTJac!}t7zW)r
z@ncJvUEhOFbXKjH(Yyy1m}DTgp={q-z>?+etTBJKzG_nd1B>vhK$r<^fLr6QrJYHt
zCb26wg?L88t!wWtj@(vR7_R_>uvXFQ5*TEHp*xASDw8Ftz=O^?S`L9w*|-u3ySZ(`
z<0nM)x{jWr(6KdziAwOGQ(mc9pe$3}gpU_9XFB_D_pZ2W`s2c2)33H~^gi!f_?awW
zE=yHC+nL4Bqmc`WqBda>J8cyVYyddR#a^C?PV1$b1$k={xKj<s+{`uUK(9tc36u)A
zeN~;9@pXxh>bAfHI>4nBf~}&W>C+h3$FCNnQczTkYXhK?zoqG~Az6`c{e18tJeO@j
z`o@Cxxt-^8Ucz``sKra#?O~0py}pQ@wud*>X*=ilTps%lCZ(PQUOfPTK8%W-p8+n4
z;_>|s;bn->9>0YWZ(Zx5^)>#V*qT2uPa{B7tvE>m-Ve?zO(!ucDic8U^{4>PZHcG}
zF|%=f>n&!qA%&<)R8f4`rwOY~&b2qrly1)q4#7XiV%%kC!Xxq^d7<&j&GU`7@#TJh
z`>~p5AHBs_VBTW_uF7=tvwXi`&KE7!V)SfBavije;`Q<{e^dJSg`B3g(1g_|yaBs)
zzbI}eJOHQXx(YG<!OF-o(;|uC=w(Z&ABT*P1h%j00A(6p7y(r$A${O4#H#VO1`~%Y
zv+-h8I{aY(wi-2Jer5omldreU3?L)?II$x1S50aezRqHB`k;X+T=MaxI(aS*^J+U{
z17aVhtH#>bn>l2gj1{WHu`hg(wMoOE3z>%PR_%6@KnlVa`d*=yHAF44!3>zeYRIrC
z!>5L$zVh;-2I!qSEhK=AM6cdd*)NGv`k_Jye$I~Z|3R;WZ(Y})2~+x`!U+B@UjC^b
zT$xSx=;uBrch;J5jxjWz!=hkK2A!r6(?6bY2z6wz3t&>aA_7RB?clu;JEOuiuZXbu
zb3nJjjd0J`JxRmiS;FtdL+te7qYrd^UK7TjT8hG$FqCL@KErRi#JZeg%iLC|W+~`H
z>I#jLMxv@u0JL`VIYpa!K%CW|#y?L>uOq3knh$?>Ct$<|BW9S&*oIt|xjrVtiU_ez
zy__8yl1L%BE2k08HKlv?1ct$e!_KyS_(n@>9b0LPsWG~1)?ZswgY%PSk^?F#dV2SA
z{dnXoA8bo=e_vzcwOxtcFL2M=iq+_xM9Wm{=q?OLM8LB}7%V#-PvE52c!F)(HLi^=
z06qI7nNvQsdP;Jg{K-bIc5Pe*XxG5O$r(y4ci(TjA8fSI`*rXgEqr)hHx`WX4RH8I
z^!T%(o8TscH3ZWmayB#<tRfHu(&ElT=-fgJEFr%5{?+pOz#D${g&3b2IiY*GpT86z
zErdy18kMe0QaYJbv6i$UkX&kspP6;|?|hWcSCpbyst(=9PfP}*@>_OE<kGQ+t3zj2
z1@E&kuBZ{P$V5J?50OZ&-%TQs1uUN~NimAbO>12Ky5?LCKPs+WyXW6Dw8r_`wDTG`
z6@N|hr-kE|pGZzTU6yqEWODh*l%&bB#L0>u2cJKju4suN&9b9poGi5<LzFGd%8G^E
z9q;kLPCcLPOrNa%Q1^v9%A>3;wlVaG&zbpsk;2_O#ytiha}K#lDs$MaD=V&MTCI|s
zll!L1`>k(^k9XfZiwlxPE+)Y3|DyhnX?Z6+)c}qBL65b%&IcHO`XM}GK)49}*lqAB
zCE5mJbj8;qi<d>er!1RjSVJI{;5`lonKu0B!}>yUx&bN;)zVtVSrAt4e)zQQVL#i^
zgFVmg4z};v)!ILIb939yUHSTqlEN%?_kc=LS|FLNdTF7+Ps3S*xD>$*mgi4T*5~(M
z;Pul>ZUnL3H(V|lq9lS74m<cV1E?YxWRISxhf5v#;nD1P*NxlwzYqS~l^8BoMtNSM
zfh5U0zvnt5f(iaXIDJH|Q@^vEW35WG4o*oA3E9~aTNagRmlhmLg+&BYa1gTFdL}@d
z!OlHXxp^elK^|*gAsVmnE_`Y*dLuj)<WyV=hb?zVI2kWOG11|&2&NO?tXq>s-MO*)
z4ROQY(%ua&xk2stI;Yxri4f7@D*<tGlzf3o!R6>-+iR{8LX5QPC<AKYaQ@$h+0IjY
z@6;hxpfz$v@vHV|v!kBQQ};=h_T1mOGOp>@aR<JCKez4a#T{pU{5rqo(bbhMCP9Q^
zYew_w<Sb+yj~P-)GkVQkG5qvLba*B<@)zBNdrbX8wR}jMyc`m_h=3bN-lh3}{+44{
zYM<|?hl|5mz?MG864Q3ly!X)kc0F_1RWQ0G$@O*AvtBi>a=`gcZFxPGlK*&#f;!Wp
zZZ-YqK9Pst)X0o|3O^Xz7M|5#!){iHM|9n$Egc~lE+o%YO6P-I?T&f8>N{z0G<%I=
z-RdWM<uZ`H4;o6d3A2=YA~n^I&ziEUw3;!Ok|A2QghA6tXmquR&eVviwMNl)&mE~&
zw5c~~k7^pFm?U!;4NCr@*4e!gI*(nte<AW&9cUD=Qwo;BlXy&DeWi%8DhDKW1P_z$
zl-K?=`E1tgWF}5NYY~i$lES?^&#@awY3?d9S02v!2NR!GdQz_|*{01ck)SM%o05tH
zHmmD=cZ(a7(otBll7&U`kT|NEsmz9t8&qkeHiJV9BA2pX7bb#i)W>Xu_W#!?QZ(Q?
z3ugQUm86E5BRA(g%X&&*x(?<0KubuG+UxPl>E4A+Yy5qVECFU!pECT1E!9J#qq|6f
z5xWM}VP<ftelFp873nR{#eSO8S|Lc#c}%D~t`xrx%D6|^{Jm&InJ@1d^ka2{*V=u{
z*tZye#|W^%{7`c+Yxu&~&2=#X$dCH|-*Rx5%Ir0BoE)W@hvNKP97%otuq$i<DY~Lq
zizZ`(o1G#;YOnS+Eld#&d4-P$aJPkwaOIIk{Ypq2o>Wk(+EB)5$Vkd)U=dmurG`iB
z#>XJgNN9Bd3x`6+$01Q+QF`$s!yKT^k68znm~ENhd}-FPeB14E{>b^^C-lBmhFd1M
zUYd0(-*#mwcjWBjBLy&yT>&+=NIbyJ&__?`3h41Aq>^o!*RNMIwjhRQpO<ec5LdzN
zsa2xG(S!i)z`!l`SjAJW<Iw>^OL$%EMZ|vXtk`F*5k8B{1#@n<%+Ph-^^jV75mQx!
z>t<yW4|g`bN$fe}cY!jwxym4YM?_zfS)l0&J8ORRYM4Vl*zcsp;Zx^+ov9{5vk9?;
zpkNo<udpZYqOo3**umG=_7)wCOpBm@CPgYj5Q$hM+z*P1^0*SAJZMBH<MG>2d|(QN
zf+dJq))dR`*5KwNcWoHg&8<~B(>M`HNOdbaweB2)Hk&$|Q(e)<UyJ*ImA%4*kt#t}
zQcH-pvudDH#m_0%^2n|M{0Z<z>(^PU?n^8xU}_&&(vUjc###<wxdDUb)GWOln#LJ>
z^%p=UO?3GdWEePFd%C^4WNT@T_C#AXu_BV8sw5|+p>RYs8;j&8<0&e}anL6Up<<(G
z$#>i$PDn{3Ymj>a(eAGQraE5>k}niW>Lf=_nvf3d-z<~2*!;ibqdvSi-85f$MYgVL
z({N-@Bq}aimy!Fm*P6|E-7UjXB1g<ymqwpBVu+9S<GZ%!muI_t)Mx62-jpR;k2Ou(
z^ig1;$t<^kc{7IRaDv52rh$BiCyc~42OIkpIEd>qEx&U1^zfCW87ZH{@Cqk9K=r#&
zuAn<N>=sRwZq`L5a16@Z!~urENt5PWzypn<Z=s=QU;+HH${24vALZZ1T;4#>zD~=B
z^_Ov`o9lKEIfJ77C{zb;?Lvy*_-F`*ibiX_UpOx&`|Q{ie#FiAL}`fip=~!r{~7mW
z$W%0>!RR4~ahYdqu{^9;ot3!pwP7pAki!sk(RgB$qD5I**uO+^RwE9w&y1SElEk){
zc?Vd%91k^it^|{1BsMu$%)kgcE6$wVE|bQ}#I>y!8NT}mFT3}ar;PD!z1XTOq!{>d
zIFD6fgijar9EvB11RK9@T`cJ<Ad*Fb4iTSBgaMk5yiTjQvPExbRJUqFE<!gjvUzfT
z?%GpZT7ZDXt2%RIn@4VY#eEenoQlb7c@W2XflSpZ5{E@G27o>1dnodabD~Wt9-1Zk
zFq+RBa}9(g`s69sD*6HkNc3s!4q7|d!`#JMu{Q+zyss=1*xd|8>~x&^ja0~tY9_^z
zhyHqTaV47a&cb?l-7zwjcdRT)Nh*<{E58nc@m#sSJ}ceFpcHOX=kCIOTPP6Wzjy;`
zbR%*p0Sdfx^4?s&xFZ86`?dq5z2B>D%0=kL&B`z9bRD{xQ}^mei^B@pbfW^hF({Pk
zTn6uSx;EYHY!Hu8VD!h9eCC9_;$CpmLF3W$wQbXG_Ko^<+SLp;y9v)M69sPuObh&c
zCCgXsh4LDMqi9}VeiQh?W|ne_5WNL8N3wAeO&Ao9MtND&3Ph}|TD=?aa-g3*GcTB)
zm(I?rZkN-U-NxuoAcN%QR`&E0Es}+VQ+%KV3^G(_32Ek8kLiUqWziTeuRIY)TGIxX
zVbKR_ZW4p1aB&7|j(uI9KRA59fP3xPbyjHDH2c#>_DonfYkKALb#_?jb<WHj#HW{$
zBcstZGD|NgiaAOs^`d2Swv0i`R&G9{F8}OfpwV*_98XE)4{fZ~bn6P^@kemP!1YD*
z3blOU@kP7LV`)tWx-CHcTCm#GDjrqB$GURpOpp3$A}(6-xwRLDXIClxxIIAqF#*OO
zL^a{*YuDRCUZkih%?kNeq&u|O`T_W`17zI!@W&l)lpB;3GU!*|L132}Fc0|aj0xOs
z)Z=C2Y8N14iNQ&mfvWNc<dkhEh^<ajv_dngRWRm~(JNVUdkUV3zXt=4SYfuK0zh~E
zGQrMZIw%Xv1XF_7KkpXZvcrzt-T_p>v1sC(7|eZ+OGi$OuAz?h3McijqP92{+wo&-
z(ra)`tjp<-^t#`281Invx5@WkO>3h%sV4G^qagNnbim!Iv)UP!{N(K@*e?^TjO-lR
zh)?6;6Rh44a9#~Vl!SOvwZqB~?O3dxqZQ9Y;8FJ)7{&h+XxEEzcz=NG>1Rw`{`rOK
z|DPf869iD1Kw6BMtN%v^rk^w8EL26F%2S1H78oUw`ls+YLs)!FPQJB<@cbWyYoDVQ
zdOGhXQVR2o;9T9q7^h0LeC%lUgu!6Dm6u7<D}->=>usJioL{`x=Tg=YCa7<MnOrIq
z4ALa39%C{)`7z?+<JoC8SCSkqEG&SF5{<=I#}8~_>t#lt=8JhN>4J_ww5#WRW8<?=
ztuiSFjw9p;fW{8rkeXV9V)3OuViI99(L|sJv)~&OyArIK0aQ(-vVpz0Y4Fo96h7w{
zcSLnDdc?vP5*~Fn7WJd$fK<sPNK_}5utfXB@I<e0)0-4x35qSTUwvcvwHP>Ur+Dc3
z6iMVs#iLNa?|EP_a3+Mp=7vN^S2i}+e!1H`$uS@RknQr~3OJAh14EnJwX_=ITCdOn
zv!LZ%5A__i+Z|Qmq34sOwY|>x28^HQ%gud8&7TE%Z4{319HU;OQx9STL9t{NOVk18
zYe<10KasuPh6PoyYKl`8!9uki6G=k=o{U-0gCe%f5T^=|0s8<oAjJT8APi(sSi8i5
zc*RAa)+ym)BGh4F1C#y-OE!gJL`;x{bRkyfQp&MIAPWsXs9bLmz}ZpCxTGWHrc1$R
z#=_-blt#Kel`GWxSsaDR$`Izr#AqpurGB&`f&xvbs<c&vCuxbedNTm(dH_r=;6A9D
zC#c${wiXrz1S#PpB!Io&y3M|Sy7^uK`{$yvU=-;KaupvfdMS$-o*Dxb8`?F93{s_!
zZMsJ7^3=N)R$=n>DW0kDfy9O@3(@F@hu-2*DS%#v(#4hx+WX`N`HA`R_?~lABuc${
zQH3f*V|93yG+;laAcps?sj$LI<b{go0bz0*wR}Owe=I;8Z2AR9p+4AkrOND!loqq}
zzNqW?ki(eFRXiajjq}e3BDD|q%LcCnR5)eeC<==;P<aePy{3@x4R3TZLlLB?@`sCC
zYNAR2zmXcLk+>^IUKUq6GO<8`HCKPEos=2f+rrFkK0d3;f%Xp_4B9-x?MB)3dalGU
ziMBv*c<i$`D@}+Ho*Xax(#VEoct0RvkS4E)Tq_zSsOp_O`NFT<j0S;H2io-^FbZqz
zuBj)bL5uRF%^F;f&rx10u^ZxT)(rQ><Z`j~n0ey;p8EEIyGbnfz^3}$c9`lqTRs{n
z2%vV;*&7aEQQ@N{4X1>V?|G1+rBv3SxVT3nux>s&(8To;Pqw5H;UxYBKH%mRlAEEI
zjwdY=IZ_fB?rf7pj$%lVYb2DB^<%`w`fNj<Lqiu$%T0<`8|E8?2^_Pg&cWpO8s>4T
zW;_-y45dmPn>bdvlaY3+w}4U}aCOb*XqrhHxmO$Vd>+C=6bY3~W_vm=o4b1XDr5}K
z+C$nPN>tNzHT1ZhMMH-gBw!<~mh<65(o|hn^2_Wrr-b~)y>(wR=1>72C8}u4T9cDB
zU-(m1KIvd%7b^}xuH=tA!J|2w*hqDj&koVCcNM%(Zu1*=#k_Uz^>5x$x-(G_7mCoH
zm<Ku}njMej8`*mw79YQ)sRm!f)t%edZ-tk^wVL%&;PPCACsALl@@nkgMRjg&E_GCR
zhN`exWi;miw4#h0j8Q-}jtQAPft4|)b-#81W7R{lB`?PW(2oiujz%WWv7gJ^A6$&N
z02<q0jR4I1#E|ReyjJBVxy9iV;I%)PhZ@&><e(U+n^Dy?HCU``^kQtGY}#;7n20Ek
zP3Z!?e1MfwAnZUHiI*7zz5*4w<;(PwoYr&sZ&#4wayY<}|Hqf^i8*$N^3+&F58$cI
z?5zSh#1iVfjIBV8-~zEkiJAoh+A_Px|LC8Hu}BD$yg^_lUg$^XUo?7#=#|a{Qzk@I
zY*DUhNZcAy8%S7+)cDtK77^Hy-z?lWdPv&VUKb59RmY{4-BCE!OCj`W)X&i2s1!&8
zUB5=CvRDGrSMl2=l#alHFlRq@+!o{M;T$*$e+EcYkOR2f3mgWA9$=JMDF(p3n5Oax
z4<I(QdX$+Sap1sIm5#mCRC|B=qpqh%vcim}b*aSVW7qF>@kOfy`BonWq_oj$0_E}Q
zotn(f2?!8f%Or&-Q^^$qWE(^l5oxd;%+9tF7{W>b&GTSZ&}=GzEC<bi7e01{4b2lY
z(n(wR-X~E<fADkk(i(x2)fY1QTuOVoMr=?LY5_DoKNd0bC)k4rgF{OJ$i_ZnWfyXg
ze~7AgN@L%6+GL0V9q+;7r5pmO!&0bq{ViY}ip0-MF#ebupb_6LXb@@Fh4%?Z)$V;%
zs=x0Z*{4*~i+9~nI$Ave3_SD0sXw*9i;(?&-uid_y$&$Mf%YK%SQ$BEpAU<QXDGV9
zAcq-=0kzQ<5^e#^imLEg>N$0VeIbvizfb@I!JHvsaNYFR*H(7PPU9SnRX+ac7^|k?
zf}rRYpke)KP!1{p`&rhq^JLl%Jb^@7&1NKwqdG*W3S@-dM2SPC-(ahYfwuj=T{5_n
z)0M?RGN>xjfoGI~RzW*o1QfU=Di{p$u_Pnv*hw*&U~;nYE;$Tv@)PsRC3O%b1nIoi
zQxh$_k?=*30L=o;anpLh2wvYDk|gX>e=kRxf8zvO@nVYjvBvq@9eVb(rriV~111?K
z%b)p`D_|fW)p4s^kD$jv*xJO7hm>|g5-nh7hsP&y&f6!<F}6Y|6*X9ak1J<Z6B$pP
zwnn&W^p|R4w8vrW442}#HudDghI;rL3kVmO;WE#z>!H}9B_NuGPy-)0lxDn}3@#TP
z5`dy@$3C?aC@>&C?nlJfEJUHLsH=`%M{x{Hkpoa=5wPf#e>WolD1+g$h~eCDrslCC
z#{U1hb75scD6gwW_4nxuCkqAx8f=T-*Ld1Ra#N3t%}7RNMVj8UZFYbg0oB5+CIBmP
z&U7e3TL2i%gT}Qqw-bS7sH2=frW#LqH~n#yBQAzelbr9Yt|%=?s7elNI?wYL^e_m(
z;!|vjRBk;R@EDDxF$A`a)`}ZK;i1$kDR+NlEhFNNvlOmI;W1oD)Jh(|8DPt=$SjzS
ziFkP!#W}#LTZauSyG91khyYGNCdo6+ATYwTFHGc&Xpy5CFDs)_Pg88kP;`nNoG{}A
zJ1QdW$?s-AMnzH!xbPqfhPU8Npy5*kQYr&W^Yz3OmkfgQb-_{k8W=P+5#TVe1srA)
zV|s-A%7=x|Gy&#;#(`E`u_llb)HJFo<+yX-QuP9Ej$3kOmojekQGELl(wqz5Cu#O*
z;Rt9#CGSs~GS~1iijRefF0h9*?HM-js?T8h1EZUz6kxw&saoXFbn3-<GpQ%a(nKwR
zYM{~5N_bf;rJAu!maN8q^w+-bm9L*LTk3-r{;d}tvJNQhP;_zk__nKf0w=r=9D=bt
zZ<{g{V_p*BzzME{Ak3UqP6VJR23#vTL?J;2?AQtmPpzy;f@8uc8=DNA!ho)P;bw~i
zXSHoP_FEP2gS^BqNkkm)A~?b(IW1)xBrsu~m<pL@pW-bHQjwOnq6Mec&aHNuFTbf?
zK2_VAE1E_bAx_&0PBw-oX0~;d8nR|4O!}z;P)Mny%Trz2p|qMk3D30^6(bWEwP@XD
z+$Ru8k`rZxOw*tmf}%PBVXYusk?Na}qH;n@!epA^@RnK1aF_&mTZUTjKB*Y605#<Y
zep-~|WbBmJj}%Mdn#UuQ;xp4P{{njWV42;qD{Cb+NR%Ex9Ua%`he-{X@POW!neCuu
z&dPwZHGj(b=ac^%qC7`sxd+|gp|NZ%zUgFMt1g@J?cq<gD&KjUy~Kt{vq!QE>sG>k
znEqt|<MC~CamLu3zMGJMT02DKl=LufmYvbdt>>Tp2ID`B?2NP|Y~gZmg9CHAOun|B
zqoX(BkHW0?R{av(wlDz()hMXDH70zifd(&&HXK_sp4r<};AG=pyUhTBb_ECWx;$$(
z`+;6&ItnHG^&q2-6!pkM))75T8efQ`oUlbU(vzpG2m)?y3ZkeHF|#D_CTw=@sn)yT
zJ7fU0H`hFs(bR7N<#hVS=EnL9sP$NaR}n!`bbykQi#79GU6uipWC;UgVv}PEl`@wN
zpheM*NLE+8zL<&9uLku*Ch$f@ko9_bY8WJ1(HET<2%%;f#4)#=HnAq&ESR2?4LI;W
z0^4MobPGTCoi0b|LH>n`Egs(<QW3~{Kx#U_iEKg5!i2MNqUs^4@^SMo72;LU)%WQ$
z2cI6orF!KUTse=0SlwV*)D}dw%7bdTBBn!{Br=@p8mWy~*)W+*@8+!?hta$y4Pc?6
z;KCaR)?wDbpk#oq(gV>u3g2xy`r%O=@FNJAZP7pbTd&P^vkPB3=Lds5E%FFd7ZHV5
z_okNUin!PzC@9oI?J`+IMvn*0DrQq|#{F8ACmy{ftTaRIc+)J59{&N2gr5uE<@$KG
z=?{APC10TEF4@3J!&b{e)Fbq(O*t@e%Z;ZA+wvWzce@b_y%^JRLi7^6N~qN#EkR}?
z`%Y*aTo-{wBBzRW)Db0;vG07iH?^3o?OLna87M)!lVcBAXF4)&Lq#D|+b5ZQJ`^-c
z#W|W3Rf;Nx2G)09Za}Ox{}p_cAP!9nNuypuK+6K31=y?vV?{uPcE*)O3?Sn|M#%Ad
za}2yBo11^H6Am`Hu~NDW2%Bp7o)`#~W}o8CJ{yRt#_x)`$3~SCj@X?(M2&D)Pj0$e
zZ2>#1ZYL2O$3?e6;DDj=Un&A&rgujShJ^F{70r6$Y4#DXZ8VW!a&PvM$qpRpqJG${
zE#hR@Y*RPWn$Cj;CP*cK4R-7?8>Z-QwDet_aWWF@-23#o;~gu^y77>8ra?pgIG<C@
z-oZH#8p>h?nImgz*3FpvYZkMkQ*9j=^?s5~j55aNbgOq<yb{2lS{F_+!&paG!EK0c
z?YP$1=`JA}Ax<Z8aRJ0e3dn+}!xt;7%ErgeeJVldpI2Vo;@9sYL??&rorlfp*ic&k
z{^|TDzwbz>P-y1Gu>#&)_WZWPN*qsGRry&AKN^o!z_kqk8Cd^3el_c2#LL0pRcJ^7
zVH^Mp%mtBXT(2tDm=rF1>^1?6LuQ%niuBk^j~K6+_WR8~8hp%ZaZkGQki`+GMeb=E
z62*9inq!ifM3JF}J4H^caxz!FhdH-)a9G^b76ej=pMb;=*Pg+@73OKvk_LU+To)m?
z)>R>aE%57fvWuPrPAwJFGP9(=0U2ZX<lY(S{tT*X$L(SX{Om?)hEsS~Zd|C@dx@>y
z9c{S2nb*Lo7tU<DWMC_moogZ6diX4@pTAhJa8_cs6-)9orNA(YrDmcdQg5Xsn8-3H
z(ueJP-E@$mbpsHQ{9Dyz;b?*s!xZ%p9Tx)ygXYkV&~Fl50RTlPFBhQZTE(`~VbW@5
z_OeQ=#fbD0_zruRO7y|%9a|F-l6eiWkoJ)yxI#E)ggGqBNv(D!kdKk<94A3`yLrvg
zE+cQ|uA67fxLT~wSb|Xb^qHU#nJ)tla6+O$lIn3#uDQ{Qd|XDn<}(vCm{C}m4H2S%
zB5RoCYq8M@I+)!!1kK`vXhi6bRsI&9|8he=8U+gZX65k6KZyVDm%cpm&H9sm8cvSy
z-T!lb<-6}%?Wyp;!T20sIr}eQyeIycpZ}+oU&K_}*RS6n-!b^_li<%!Mf6L2_Ahz%
zrI6p;6XBbe(d#?KSqaG}6+Am9$2AM9N>QGs?4Uu@ghFfFq~X1{7425PCC{89q3I!&
z_S9-0lTR@)@e#h8nNz*lg1J<z16~Bgp!5;IOmWGIzbCRf;L~ZD<&He7^^hIw*Ps(P
zVoTwSs3<cqJ>AMKrfG0#!c(HLYSnSBG)bChNfV+RF!(|_D9=Obt;p%IcqXQ%Df-kr
zyf=EG-^;oyT%vvvK9OP%e|LI7#iFTq1h)MF$dx|#`S7Qqm!tJYiR`5d>MS~VwQCQx
zp`^LRJ>1qGf7?BDkBKLprlb;aQQDNoFgWtmb+e|#3>bIdxY-njB8syV!G<WBpklM9
zCDiLoYz@qbez&1Ug2kIe1QBWzD_Ef-AsDX+2DNO}QKNc;3Vbtq?oDP-+uFrjc68|1
zen02=vDWSk!~8E&x%Zmh8V1v44T{Z9Zbu1R8~f?V*c0c_LhQ>zi8hLli2#&~0V*Gn
zD9DhJBIEN4WoQa(_6!|ph_2y;J<JF+SAhl;FSPQSd>6a69)+d#8d~GrNoZ)iAW!u{
zVpG|&HasnJ`eO66WYOh>P3eGIjiQBnqYusHy0G6*EhMmR$C79xCGF4zTYis=&fdaw
zHnTD4+yq4&e{6F~G8enoM0yZOU(6F4Gzty+j8-%tbX^b93<~-t3Vk#~4RDsK(o+ag
z4C@WXb6`Zh7l_s4sbj}nG)*DYcrh~vVIxD5#cVv2EStrJpgLe`_M;N$WB?-wBTa<C
zOnS=YN88D)66mKLvO$xfdA`)-8h(pOFGn`oa#2%JmdvSGO;L92vj#-j9@+V9R0;WN
zy`*An!}m)7s?P;FQD%c8B<b(d`4n|6-~c}U6dT47DMXFH0*!-^H^|4pHY<_il?@2W
zX-$3aA*?r-1!&AA*5+go0Ml@^9Feiv%VSQ-qP5D3Oto1v&5h+jZX-s}y9_H*h%+);
z_=>G^#S@QrL)2Q0%Qn{gTM49?;~sDc;k#&6BlwO1z7(khx3+dWT+h|d8J?gMTUJYS
zB^Y)*ACrR67n-ogc588~o;WT;irA9oSs}U3_YYp0v^gWp>8#yb76Q0SFRL;|Ktasy
zf4cI%t321DT;^gBcGwF7Gm*_E79nJtuGLWLk!3N*N$XTk7r|l7t;_u4FHqct_aT?H
zx~o|RIw`uPYJvpTIjn5S*F~02c|j?zW0w_I%m(B7<johDea0M2&Dfb|q1xl9TF>&f
zcQ>Kc(csg4x5Vq6qM?aS!b;K1IX)T{MXZy|_0jz3#yU>6OvPHkJ?SR+wW_4^k&%0m
z?mxrb$n|jk`f{G%1P5cerV9^-;?qKOl>E4q(LZK+%%WUD7x^yNDX##yeT0+n8g56=
z2Ve9XM1qR>{?NAz{na8xSg-Tq<;14t?caP`DL$)~6PZikdU%?&;TUj)Ihg~IaxF))
zoXFT!jFZ`m8?CP<yme4XH=XIFFn4{sJUyS?&hqTo7Huqx7xIA;MP%ECxrCug6Y*p}
zz~QS|2cz-kX$HrK{!18$r$&-W$2tzc7ahlOx~Fcb-`mo;7G-f9LL|l+WQ=zZLmxjB
zc+T2wZ-g_@5K0jc$XGTfvlmQIoJgo>&?m(Y?nN=aK_P6vox<6Wgi#B?k5Nu>Xkz>|
zR)WG<61R$BD6PPen&3()b91H|t`I{Dz*?fRmAgiM8m4-KAIG{d5@b056W|hOMNq6c
zfo|9~r`S*hjmGyze6OevbnC@>x@#)Y%{-K{VIYvg4KCdyZ&#(o=9JODb|ToA#VgIe
zpCQh1!qhq$R_UDMhL@2_z#_GyT$)Lm-ca5a<0GSz`H;7<zu1GN&Gr@htiWVU1kIPL
zMC~7^(-!CB2dtT`l4}Z4iB_Rf=8{JeMxl|=W^=tII4btNKsSW71>-cT4emE=svr}q
z8xD~zq%x{baWBWJs5`NAWQ`rzyDAK~f#d@c(DqY@8=iE=6C*YD@Fu!D;ul9Ai6u9X
zI$o2`3L&ssGIbppXIx6Ut)#(Og`gml?@5%100HSC-7fJ%yW@WjvASI8kP6T#8lWZO
z%5i(pW_vBduLPR(aY&&lWH-VyuD|o%2=re4@DH?*U>>*V2#@{kUn+9>NPzEROY?RI
z`IucJQ7Q$HOYD5tE%koC!)I?vvC|zaYgiFoS3-UxPCe_ezrh(Up(Bi9fy#t9Q+ns1
zizf~M4k=|wxCc*tz5^JEa(J)<<TYV37Wc~?y}R6_5lo6DX=g%NjY4<IWpV{szxwX1
zPWM+e3FP)-waQD9GP0%}opO8pj_aohxdve~4v_TaBih1MSZsRSITaJ;LzH<q)QUOU
z;hFq&uRj$G#~#M3J&q@<`;P^Z=Q^XD`z&@oJzvLta*#GvZi6qnQRYOpk!7ILg1dIe
zq<naX!%TvU<MsNoc`P$IhBpQKT)NKAr-KZ~G<06e7G{+Kf=4~V1;wthchGA&f*h@p
zj}AE3j=|B=lC;Jg*M^8j#t++TVr<3|c!<pJg4vh*`vu|&84unCv+l>E_|_<oTx>h;
zl``ZaHCydL{+mMDuu-H|BaP<EFG~_X9?iE}TXw?U=-66cmQBiK1f7qa>ny^rz|gUc
zhKI6}SYE0IRIr(rE{dBY8xI-R<Q$#uy14Xo#n~?=HLgr8i_{*F2==3AgDMBsoJro9
ztvGCcTrJr1((OEln?LjJlX?^KHHyyn1u6navfog0TzuU3zR2vtJ6xi{-Sct)9t(wg
z_`-U-Aw}3uW)5>m7+q`J2dmHrJ;ZP!n8xNc=~~Xcg?k7N>Lmf=?Cwl#eB2Ge-0lg&
zxhT8H`4A2wvu)sN<DS&lUGY)MUOQ4wsfVm1(s)e$e51y!9P#FGf4V_RI-yt9*_6WO
zk*K2-1A{zY(TXl;^3)a?JbPYhus2s++dH%Hexi6oSxc`cyo{<@$Md;+bbwy5R4S5W
z1g2yP?_{=J4zI=a@$2M>Mr7Pw{E#27hR@<zJc}2!MK@b)Wj);@Z7a<e!(4Yho=La`
zxh@gD&SM80ek;Be--;i><AoWIhqGA_9&<-E_gNq35qSSw(j%%Dj>ls8bG|WYI9i<@
ze;!rGh)zh{ErZkf*I|BC(=~h%FX%!=<?s`_!izI2IpmAYerv%1GxvNRbXR?F_!&1d
zlT_F@Ywg`FFP1MAY`()ieK$Hgd_0+Y@TvU``Pmrix*>mfqpmjxKg3Fz!nl$V*edz;
z6t3v`BxkwtI9xG<<z0iG@#j2)*Jp3eHlOa^0cUSnx81^g-aNeAxn(>?QC5N0@R;>L
zX;?rU5L&eQU?{2c+Yw1d7MAFi3qQBTNw#R$?R;kOrCEY8U)@8l_;V|*Q~h)@Uv^*E
zz;l_mV$>DzVHT!eq{IjJ8eH(9^hvqo<jEK|Nq5!#O-xF}+{_H#B$cH0h-8~y@^n2+
z*r!?Q_#=HL^)s{*jMNOEm$EsOPZ4Umm6E|&Ii7=qC?^unH%T3+v#3v0$?AkL*;=y@
zQCAtl_TBJY+ufeV%07m%(}6=PRzv;TD7&W|lQUuOjc4byv{OwwU3U;@mgqKj1ul!^
zuyK{ku>oHT{o&|T*t0x$ycN6Cn4v|Y7yA|JNax^X|95vhLPD-J=x3<6ZCWpA&}QOD
zmN77y4Wm-CUW&8PWZ5`kw(4T_5Wt?`*UQ5AuuY@b4=vhwsWR5f7>Y^pi~8F2hXrEs
z0b4D5U3caQp9vh4C_x*%wp8O^jesM;n7jlbai<7axhY`k1PKOd>SR(pSat!J5(suo
z+3etR6JI?PNTk?;lOUw-RDfyl1kTu^#OV!tW#Jojl@$?O@M-8V66l6$9mYz^m{}cW
z1QX;n&vy5~pO$*6(q+{Az;QyW@g#Tl_};wqaD>`V8nb~?t8%!c*MvkfH?FAop;%Oz
z+7`(&j*Z82C1wJ*kt2UHS2A&_wR`Jvk9ZV9FxL{%Iu6)Qw8U+t)wy9dgXLKOL6A+#
z!0@Kofrbf(N?M-%^zmj;$w}U9_b9opJv%cZjSTB3FiH19=E&7lDVi#2bJeK*0v<B;
z46a!V(`uZB73czevz)~^WwSyth-jut-W*DWTAZNBh+u3cP;W&=kBU*eYIx$<<o3GF
z<y?&qddf+%hwL7W68cR^4Kb97p)xlyzKB|e%*>;YpgbXu$W|uby$yPye$_cG5+rH;
z(@rw9(X&Z`y2{uYok{JC9!cW|wol=dzEN(2@$Z3KV!^kCVTUhqRDHk6tloXGioB**
z;QcbZ*J*mJm-U4TNrs!n&<Zin@!eEC+0u2MVOT*%+6NB@gsBxTi=k*8mxePYafp%C
zr9@k}1-5R{tS8^c^IZsxt^U-8SL7_KJZ%^w7s!lsc09_PBa~C}-~}6h>UdyFJ9i?U
z6q!J;EcAwTgjnk_Hues=cJvt{OL{ZF$fA+G3A3lMwL7{~gpoc$yHJnp*1(dYJ0j$d
zU-|7-9oa-JX1CK>+-%kp=R<yT&XYp$%$qniRWXcXW$oGe@oT@0yXum91F|hxp4xoZ
zm^<oJmcfhO<W8A-vZmaY5GwQCT|074+oI3?Q2&2WPs|-|3d6Xm=a4W_4CW@AN2Iyi
zX}T(lj^tX%Pp42P3XyWwzv+jjY+211NL)`%{7?AfUpOHwhW#Zvp}fiLBESumKO>~B
z@`MY{DN%zK<fuaqS5TAY!Ly1+y|HvPO_mO`9Fe`zE#h(6dcfqQnx`3uEauh9MF<S*
zPs0&dO?(V4*=5WpWkyQ$9{}4D*EbZL4hj%ZU`rshayMZam9y53m*NMA;KQzK2eMTl
z7*y^-8;A^y9ISL!RMztmTrTgvF*q)>@AVe(P7d?2POD>~dwd$Vyb)iz!w*y1U|Ew{
zr*{yD1hSX)D9chhuJ1kI&LSR0|40d9D!3*KeO{q}R0Y=sj?#4FSagtc1^Ffo$=_Jm
zYp;T{5dvh#WnszuJU1h|;~Kq(836Em>I~H7X^^w8C=5qO<6C<;au;vO8!R!yL|Id*
zpg`V+$i!i|v~zMs1*x%rkhfYpNzzF>wPODaRX`Am5jwZmS=iU~59#?~JPfw1O%~Q-
zp9SQ~jM2)tjwZL#)WdKYJggdO4H2zPAJmqT3Z$MW<-x0qZli#lcsdpQn*~W2Frbn8
z@MCpFJ1kh)Xt>k1AYQ=UZntYbS<zFG6X(qS_D-~y%;1?BtQlz3BFnTMU};W=v1}K3
zb$h8{@vJ<!cn&GJxropPgER344&{ce;?f1x=-h+$63y7~YpN?j;5a}>x$da6XIh}a
zjDhrE0b(ZT)db2yv4Vx`v@Cr1(mHxYe$mDx%?Yj3C&Z8&)TdvFLnbh)-!Wx#A_({K
zMHj@?+}Q@7S?ah>4ey8}24+4h$PJzf0t`k|JrsBib=tm9PY}*B-wO3S=aNR}+P@w~
zqxC4Qs|9f+gHlCfKJwlaTiAX^jP-fm`?ijj;~rb0rZCH<$i(>9?l>9z($6F%LV<Gn
z(>fGMam)3NCOe;vE4SYHo(jxvV+?J;?X(hzTJA%`1zhU|>&Roibf;UWMU$^5NwGa|
zva#ERb>AkVtZz;#j9MqEb8xq(a9-S&TeM8|;WU%LGRXA|p_KkOAwe;Z@FpdgmB}}C
zhy3{zE<}7N!>gz^<GgzO*-y0VTYI;8xE9mJZdd6&^gnF3x+gzZON|ztp#hAM$5(NJ
zSH?=4qWTk>!5CzmH%c0(=18eFZz^in{=gGLqU}-g_1MGrf4#lPZ*02j!FxRd0r@vf
zd8;}Dxgwuk>WZycV54nu0()(DW#-SX>Tprld-Upp#l80Knxve}e%EUiY08t?s8;2v
zN3RKU%_7H5@&~<o(x=hgeQ>a5FP<yJG#byAt8dqO2o1)wRRLxAQMBW0+IULkr+z^4
z&Q!Rc?nO7q`A|v$d*_s3L*2l^;`BLZ^XX|>ZSx7utJ6=`)xa##NKB>`g5K8GBwDMX
zq)^31h+qqIG|SBVb%XR1B=sZc-K8pg%5>UGJu#kaAr5k&F85v5F8bNlY6+X;*JvT$
zp;7pPhUf9=)_a#+bElK&<aa-3f3^7~^#|kGe6_i|dzdGCSV~0#8#im0v-ri93e;|%
z@j5NRkK}$JE^z{&8LwuCbPf<l{+8?3sCa7opNIBr{dw^{*I{{###I!LM1Z^fR|U==
zfAGI6uVE+~_}stFw@d&~uLJoMsJs1Nvl;tW+tEf_^Q<#`1QWz%2{qi5{&7xIb3GgI
z&G5N6e<32nIcEI-CIX;U=r>{^%5rfi92t<L7MI~%w-p<(5bFH?4y@iLED!V_>bbC6
zsQbCTV<KtUfNt;O7|1fpsk0|d9qj1(19khT$8r%fHdBv#ajc^R-##y$&Z#e}8OeqY
zd&^flTpK=fe>dCFtI}r|h4^@Ls~b^im3QwsV5IS#DG9omyqbk}IfHti`~giz%|7F7
zcBxcziW%O9ugT>!WFHE$)@g73=e2^_)rJzqVeaWnO(9{l7+pWwhU!aOY^`hW48%=f
zhKstCcDBabwEK1qnnv4$RDJL+3C!wH>e5=i3^lwoEPaO{g@oDiyK3!fkZo0+tBRTZ
zZ0#bZ=bF?F@>84I6z@49S1GI|Qo9K=8d^~#?Yd{mSEJhEm3G$Tf#o1h6m2*=)-j?Z
z+=kBx%tM?QXuk5HG$~(=@nT`Tg&I#wB_|}fAnR&DIp_!-rQnvxL!2n2t|YEuRjuE0
zavhP=v$RS}BeNzzEv3OWb?)cfdq9pG8xJUY4PX$|fK1QP!Gc#mo|I~JVEHaHzr)pm
zCAGDZQ(9`dZ!1#Yb79>v=KH|hEMWCOeKi2{In7`rtxDmti&SbkAIlFLA=}ok0(r%x
zxqVn<d|Xi|J_I3&b0}8v8LeYxU?{T{i?Bkm+0`^0k<G#7)&-MC*s<{)TDq~KSj7?^
zd^e_XiYRVgOG0uBo0QcpC;nyZZy5&~X&yih03lSdE*`waeV}DTD@ItJ#G)uiZ2lFE
zV`dfOGPNmP>dF@2$LzktO3+G5c=CSEU?M>`SCUZO{EU?CEq3QJx?3U;rAyFVZb9u1
z1^&azLXVBv$Jx?W3Dgd1_!@%;&J=|ec+PFbIt5<%@O_vB+!3Lpx?pGMH}=ap+f$<O
zj(WVcHa!d)>F_*3yTN0$E_T#S!VX%W{T8eR-)AmexH2>6NLG4(ke9tWIN$jJtPFR0
z!PIq!?nmK2-WQ9_?0Pzv+Y;(S$JdMpx5^cstt~Kx7tdn~?j$-5yD`w|Atw@(@y@<N
zJs(5!V28f(C?Pwy?uF2P{w>yv<?Qm}YIe2qE|AmKL+z}q@$=;cuNtJP{GSPv{^3vl
zv%<dcJd(TF&tmzHKm8~C<ps?RZ8bbAqQSS?yr64zJeLe@FFhn%Mt{}Ik|n<^FWKrE
zFcc5yN4~k1bD}qxIKR>?o8TpBcfAIgBEdU(GfHYEJjDHi*!S|nvqWD~(Hmf`V)#9|
zG~umCb;G(^ESAJ~{sflFBuhSJ{cGVMu=!a5f!Hb4$(bGY1VQ6)+<6(a6>;+sf)WS`
z={zN#2%D>yEd5>Me;Mb=n6ejRGJedaL^TspfHi;dX}E~ZU;P^LC_0(6jAtWzO3~TJ
zvl>$>|Ko;FS`lQb96;l+jhIPFfgOu|$PT=y-K2#EX<TXg&LUhck*fK@xg!;1O+pxj
zz3w?#I+aZno2;2k{E|RrX`vhH(R$5~73Ou7LSEdTD}>FZD*HpzF2iPK`HB5wBWoS7
z++>bIuNDnEI^Aa%Qhc8qE!iQC2Wfn`KY`Xk?9cO&+EqfCE9~ZuuZHEK1P+Z!Z#9<7
zoB7+-m?++M^<UT9eD1epi~8yE%Y(4}+5E+-3GR)ct+jj--V8Xy&Tia7XN8uX91f$7
zl(Dv(rKqOUnkVBv3>)cpy#JZ?@rMvKOfD2=3Nfyz2I2g)>7sg4$%QzGq?CW31@?lf
zh0^t|FE`0W`gBq>f2rPdlSO}+6i1)}ux(Ki!ZDx9mtIDv)0w%E!;%;aWG0A<s;1%s
z=Pz%B`cPtFo!5&2i!;5O7%jPIO1l)n+H3WvvyvjM3Trny<spvU0%6<=W2@rLSuqus
zO+^aeflfp1753861VanxZ3B$2o^ea?DWyq)o(0_uzg5Y=b+vJF;SWg<_$QN8sJmG>
zG~#w#c!Z<L-n7&tp+4!SkGpf~x)N;`)}3hrbQFd3LVS7s0SL`<eRWz6e}H0i|JRD)
zrXO95ZtA2nLRAWnnq3e`K-C8<Cu(+_Jq$eR0X)x5iI&f=TKj`fo6>>lbm6{C=a@vA
z4hXMIG_-@h;k*4+>28q!eXz7%ZyIt)RqLb=AiGHoN^lNq0e1n7h1oA5mP&J--K%oQ
z&o*(}PT1bis>~m^nNoQ(YUU1vmu*^^=qb%gB7Dlce|p#Tw~}~G549BvbKa2EwLnNp
zda+L#myHnFjf?6QmaO^^%7MLQrw}T%RLsbX4mA7bCF%wi%COKaWC~WboM&l(C(vdB
z!h=y+1}tJ}8gHIg)FAxTLa~CWD$1kns#yf00&P2h(X3=uGly&d68)j~=G5S(VH*F-
z!$W%7?W)n@=CsTchya6>>dMxnU=p;y@i0|QfGbZbOoB;z9Zt?5yD|lprI4`u@^PJk
zrBNPLxT`6b7oJG$6V{A7$PFiblz^8Ks{J+YHjznQVT?Hw?`{~LkN5|(gku)NZ5lSo
zC+c1X%nU3=RWsqq{+@=tw+=!sXs-E)TQZ?&JG==vesip>MkRIzEPtF~hmNU95PHfh
z9HXao@+E}*?`x1myK<ZIK?H&UuanFmj&nSlS}s#1pP!F{+XZo6e6=Y;*)!Vofqp!X
zE|KIJ<f<78`M5O4%t_oGyUj55hD$BAqa2cPI?su}lTJ$7m||eE0LLskDJg}QZ{i_s
z=XrsECXWvDygqH(EdmgBFpaog&MwsGNM=1b`w(J82vU;xj&Q{QSTaUxHVaWa->C1+
zqcOAem~c+G<mTcG<BHmu%nL*JHhi%`2VlkZ*XDY|)Dd~IQ*xAA_zjAcB*MYOEp}j@
znKm)6>y?B|aTe(lqAshNL6z*iB(zQ36%GhSaWN8{ev%kTZVs*`<bKUY>OpcM<a}g*
zN71O^-_!-J9N@Twlu^+iM4VGh(!3+2DWTq%^DpOJ^p}%wzN+J|XH1q?u69HBNsUD7
z`dOJmP`$wTw4J?3c=-l?xH%jGQ-p$cN`acjoXf|nju)XRM4pX6n+=6>ERK_v9pT5-
z4=nWyxHRPJ3LeIi5-21G=OR$}<IUz&1KAnY9z^N!Jv2vv?s}gw#!7W0s|#babc7lM
z$=K*{{m)?L7d#U02t0S)707keE%`9Isd$oRgmZEnDlw~Qv)QoVND;pXuG6cxMHMGj
zOFJl02`CPrS5T<<(y2wWF`1$2W&xvyPS0EQkHW+agWWrno`B!=vSK4oLFGu*_9ifx
z_BLdT;lCM)8zBw|RD2sO6eJm!46auvC>=HM>cI1b?=<=_`;d|!;OQwq7aV>rxNd67
zB`cli$Ol@tsj`MS8F4S4IA4&Iqyp=g3o0WuUB*!3Zf8_v-2~p5h+$3>GMEuJT-w~?
ztks2-hDd<Ow4}?3R5V4z9r*JMHJ?>JYw(v#`F}|LS7D-W-7fTh<;&-vTRT+wrCZ6(
z8xmr3NkbpZ3x;*bB>!Nhw`!bI&VV{5N*ZoE?mafRiPKCI11Ic3sGlJdLOw6Z2nwWX
zCyLXcuyqF927S(XGeXCC_1lUZ0lFXUxk*AUg2^y>OJW;=9=l8pNH<>fVDJd|P>sId
z@cOv=v=~|4#sbBIYrUS5W$fGqZ#wzPp%Eg=KLY^yO@#*4Spu>od@~^p;+Y5cQEv#m
zPmNyw;5ckFYq&irF|4|D;;lk?9DY39Qf4}vgPX>499CLxp_{%y@kFYL0Sc1Okse?~
z64T36JthJ)1&oRs-DwZe89Rm8g0S-I;mh(3Rzna97bec>qCd(?u_;g~DY6?1_lbL7
z6k_pj@a9!;!rGcF?ktjSs567uSY==`+YF;L&B=2olcVrmEE9XC=EcbJ2ug<{PB52N
z<Kv;BHxva8f5fm>nPS`pmOmXh;H~X4HN=JHu(+VjW(uLCD5NlQG2coyj-7x|)IfF@
z=-k}1a92pAFq}8aiO~UXdpKVX?>0^-_(dtn<Paws3Emvd1+s!0l|4#|v;_vt&{u0U
z=*r08Xle$`$)9QZ1nd&C$wO*@y72{A&f{YS$)S;vr!fjZI^~nmls0<M5`_8S#fAOs
z4OqmzJ?{^3L3wf)kgCTu;;m-WLnwUMPVtC<<r$}!juR`c5CFTZ^MyG3(OuZ|=4ntW
zI9=$qPTz1DBp^ic{qxOmdkaFQ;xSyLKV~_8fn3ZFHQj~zdPJ7DOJqVitp+FDh158)
zcj6o;lg;{L4OgZ(E2b!Ld3J?N9N1W>Q(1wZEHx$0X+|XYQcw36iH{R7Hqxq!c6Q7n
zvpj6K0GNx8ED&A(NlYW$OL0bO$FCS-;^v)ZZ%*Z-p<}|ydzC9ftQmDPPBX1!al>tJ
ze<ANw6ueiw4?>?aVHl%oO^J6^+X$=KJ{15i_qhM;jph3aeIYrOrL3W=8a8=iI6|5%
zYY7l78uNo9f=bj=1;f}%Wu-+=FIi!F6ZXd3XOb^W&ZT0PhS5?<UMtqi%pDaLM}*Va
zBa8?(_U4x;7je?n%!7r#J-y5y8A}}1opeHTzbXwf02P&U_-CcDFapuX-M!WG1`up)
zZqTv93jPS1@@vE?4m6J*+6|+3$QAEl^y)j4)bIrH>S?4oXXur)a>~G1Rk`=w_DaQ+
zn<_lBq-oH^IE~ERhlwplbUT?N;>i2B>WC1iB<5H-;7jjfcG^XwSGdy$Gq!sVO5B=j
zEZLn6Fb=bdTZ$^Pq}5gm!U9ZECD?1I1+xa>u)5i!nskY#T3c2(MBQr2RN+A+UUB=C
zhSZnPC<;dTGy@aciBhhYC<WXCK$iEDD#}bb8M0>rI1RJk+-TKuG$;mFnQk&!-wS0i
z*l7L@8mLx|jU^-Tg+A@F^d*CjyDJ9OZn3B^ZSf3t<5+ZaojPf*A5awaX62+>j5;-U
z?J85-H%u9ugiJ}K!cJeS?%-hq3eHp!6JzA8UQl0%gok|C<F2&B;MV&~Lc3LFmZ!|o
zhwV~+b*LO{^_cNS7BPkR+~Pl^ip#x9{0IkV$AmTpd-jg|Iw{i{om&x=1=EQrbx$>!
zT6|jxFl#M^51v$mBBL-g)FV;{D^3h#4#5;{8bCz#QJ3HSMtl??BPsc^WT0bipWH6y
z&IG<~%er0?h@rlGn_wjIj$0ndV~unI?qCvj@WhKwf><z`RS-H2>@!*G{bsmdWhs6l
zZfGiksf07jY@%s4>;!P6aw(Zxwm;OPK(l=LO-__;XEoepR%SNBif*f4v3FVAxw`S<
zA5>e1<KMReAM1PqH|FiUsk->6E&?lgziCro<)!F0sDKrmrn7U8L^HuC7n30EJIZ07
ztzlzB)xxKbYRGJ*$kAjD1CwS;5D>{0IL}>6L=`N0L3v*koW`A>YPi7%ImW?rbbMcl
z-+M?kP?5S-Zc|I9+c>EctU<H7-Fy-ypOwckoi;)%HX(0bDKO?}*-yY1=AFL13)4gx
z=5bJ&1x^C3x-jk5oWjE#`;b-ba!9i=oRGes-7;c2p@ijaNh19+luH8@Reb@`@}Rn5
zz0cLE>^-7wJVtO+8U_J$wMf2QoYq8ZxWip<7JnSm-|7<vVPc*dy3kit;q|%T-ix&G
zVNvJ&SO+;|{oE<7hMlU2##Eez`<>$G938}(2@RYDyA{(B3zhVG+isheagqBk^t@3Q
zQvqE=bYt!@tts*A)|^mwb+LhJ-%9oVwy-sv=VTK{BdX2-o;yMraJxK;3K=ttIS&g}
zSoNmVOX)V;HEy@AU~iSjn2qT7s5G2>8&;9wrd1Bv_gvDM{a^ga)ke=tiO{2mR;izy
zuFLaE#pRgHyjZV3yiI|coro)f8=7q3Dmcs=!QLDfdO0|9n|>!J{Ts@`M%9uJ8JM5c
zww_8MeMT?|S!e(POU^6PYjP~!v4$|*a`_EDmQGx3-GuC_kIYIW>rOw|0*}IVhdG1~
z=6jS0W16tlYPrk_Y9Qht6lgt!;G!p1D}7In=jm}SQ-kPp+!92NiTKiB+1yKPO7cxn
zg4vz+ONPP!2TK3Sv8RI)5QLPU>Tm<}$$vDE{K@>;|F=R@^La1aTSuD=yuBb`8FqV+
zC(Czr9<K_p)N29EO)^l4+td7Gu<w9hv=tz#L&x+ewj%S=hQDZD+W${F-70^-IQWIL
zPyh6$=d|?rmvpbM_OAA#a;mC;v|ySk{j4tsY952_remcVG6{OmS~q!+n&?MmamR@P
z->vVK8Px@9UhkG(%y3cGLfjf=Wpb-Rj3hPz)bppQSPeqj?QYGBiC+N>3bj)y*%%DE
zS8yvjP+x#@jMR3MtAfYgi2&P{8_HWr9H}~r<?t-6RUoG14)PPSJAHOgRpoEe;z!}O
zR8A}3=|a13uHz3{)vluwntx<WXm)Of9LPo<f}qm_6IV1XzZ&fSRXotqD%`Yf8=f0^
z5S<qBuXSg!oNX8280Un$^t*E5KR0qC<@5vkmyT8CMh_249ojiuWlGH+4q|Jdfp)QJ
zj%UTaKBjr&iiy5yv!p%V=HCu^Azkb}Y0<?=0T$xkdvEoORnxvwRJnYkWrE0R#hlX{
zn+j<yxw_x!Kx*ECYJoQG+V$i_wIT*8_v>}!DlS0dSXnef(AMfVHE_aVWVYp4m-^Qx
ztJU6VxsCx*E^H!#j>3SYA%PH4Fmy0+Q7ua|7a-(@D)>K1+kc5wh(|v6A8iIvu8Bwp
zQm5UyQUDXquMw82CA#ilE@Am-f)vwS>kPFbUp2yanzuG$Si3MSEk=_+Ag#ghch~JJ
zD9}74cZbkPTS|bT!H%|;4&BpUq(;WI0!~7E9uH6&c`e5FI(Kqw9zc<STI=Kisx()$
z!fyiEhSf4nGPeMYS>vPAX;Z_t697*>mkCkz=SPG5-#p*yLxLD6mXL!meqOT}h~az4
zdG)ok5;b;^K@LqsX;0aSA|Kig40404joAsSyOxY`!&!6LeW}CkIFLw`E|0+lxXJjg
zA<P=(>}femt&{*0Y62_+Uu-I)sUc3O*K)gHXB~_e>kt|`){t0OGW~v@(=1R?+{Y#F
zNKORE*WIF!acjJ;7Y?k$dp>!`@04#p(+L)eoxtk)plvDTlJz#{*;XlaL*nQ{x!Z^3
zYdt7QByN9>5(k}^M~M~47~&lc*d@B!j+$|5buP3?eym1%_E5m@1}IR9^7m6?if>Ag
zp<%-_FtIJ+yo8GqE@8`8m0CWX=&<OYYKgj0&uE7r$nN$BgkBqdd;){Oj8v8|fo3Kh
z#CTj3kUfMH5qe%1nTd;Nh+Ry7kjGe%g%Yn|0#Y_gSt7&^<dDznBP7GiovIT{+A$4`
z8#?~lB<6_fy;185|9qmG;T$@aaZ29+qAWeiJK_D+Pc?CjiQJ@N)M3;sA5!UjXDryv
z5`Z((>d=JtY0ucBa4O@=KnAMB&mJ1A8<-%Y6f%QG;H=Y&#;R63JZkb$!iGEMv;2B~
z9~DS`Xr%pWn6@g~8Hx>s4N>mj^oL?>@ovFw+Q3wvOLZO{xr}Ie)E9{u4d)O=d6%ge
z)E@T?Z$7-wR!d#U0|!pc?5}tS9Vz>t+!qq{)ORk$774N37`!k|$C5VW+&!3RM`tC%
z1~=F$+_d?Vco1lT)x3t>ZunSX)-?;YXMJv=z>mxKG5PI%@yFd9J}+89ldl1rH0BCB
zyE!m|Yn_urOz&s2jxiu_2kc!--DGHTT5W$pSO<=%*R(Y^pIDZR!H!@UFNmYlccR#y
z!R>#(>u{~;;TK!sdDCK<5~DS@>4yg|cW~F)K@8>Q1yW|54XMMzqeKme_c=;G{Uy?+
zgUMzu5_qMlJVd1kKhFmPVI!me48EEx6iv2vgoOnAvoJJ=f4ttuz8cU$W7rST7aQ=8
zrq)ESkC3=k{Fg|`5kB_=e`WFK^fi_c0#qf-5|d*vO82&!75bRvk#J<yTc~}=x1IFH
ztaM7;r(6y3X20QA@PqraoP_AfRl(1S3Y1mRtbhNdqPJPyas`SvVej^KHB_q@jFDwF
zDZ7iAK9suU3h%YR(Lci2d$Imudq^jKe1?~Jv0#m%F*7tNil9z!(Wj{pY^+o@ad5Od
z8huFzpST3Kjji+~iOPQKie8FYaOxqCGemPp`wauDgO>KrY9F3u$vTZZ2s1}uPc)Df
z!(?Dfk+bunf>yVd)km;^tud4R6_{9Lw86~Bg1J8)&s>^02TdC>;Pc?wj#U(ch!Wm7
zjpAS4(VEr@4}BDgGcHf57g!eKe3>9CfA#Q-G2Ggz#%9tCsy;#qPd=FSk%Q=b4F5nC
zabM)TMr{kv>l(@qJwZ&*tm#8$Bb*2}?zaF`mxDnWev{>5LHZKoW^8k{MS;!5FKzd}
z*<rbbGn_KhDPo7-+tawU89}RY&#|jCD-NdgXwzy7R=15EI*@rcSU5?j&qQ>+qk47G
zlZ(dk!%S!4zL9L@$>*&MN-@pGbhgDx9DTJ-WL6uXGGve|SZhv`q>hG3r;DmHI1^QV
zCF<=OvBcY-P=|*-?!LDkZRn}RZW+pBtdZK^QSIrITV|?qr!b@b70bAag-p8bCl0h5
z>(r4=EAtM5X3-O^dt^_&Jl2`eZM56xJvQ+cY~0f@pMuG)I&pRLpiJitRX~d<a%+5h
z8n<^wDY8jU9XE=;(p?~L29*qL3*h2k>h+R=UFUgJ%WlCu_2Hv`#}UxD!}aWt_RQt%
zHm^Wdj?72jD!yBhX?8V!8qfGCZG!bv)DpaQ$<EXObE$Lp-~HpaZrt{8$Nrj$_d_~u
zjFHpUe_hOuq475p!{0TV6_idQL9%t>Job^RTJP}rVT?TVT7gLkD8|^?1o$Yt7mQn$
z)+_=IVDKDR2BUva2;J3WNH(1$NiJlRvu!|ADA%JH0GH`-H-^`GBC)#c2pj`s#WC1w
z7C3(~>X!YZ3y9WWrs>eQJi<2i$?msDl~7a$f~zKGe9Y@PEpNV*_*a)5;36$eA!yE{
zrQzv56J<#^V)?F(u!lF0r{ul)+6Q(OK<{J1XOXn$yjF&^+ORRXKAvE^-TF}>N`a-c
z`IEck!w?AZYRnegp7nC_p>|3<)3(4h{mT+}&0=uq4R#)SErA)0G7x{OrIVD>xY{qD
z77(ISc*b&YYZqee(4;F<x<4pEhQz$sl%=lwv~w>pIovGaw@u#|h||QDtabP-uxVM@
zB+)1bsk5Ug!4mtO#WQd6A$_ojx0V43O5Fr$E~A|aZ?ma%hVTzo_ba|<Xm)Kv<mDcd
z>ug}cT7IGz9Dn7}I@ynzIdhTKrVT1ec-JI2P!57CW~&*~GHwH)@%|>T&uE14P9mV+
z$|o7Kh3!*{rTt4Aw@o{(REEVx{<Gn$Huua59~yoEgHxN&CADF3{ZAG7ZRUXPv`{nP
z|AP?5nQ&jOVCJ~N1Pi%33~X_e232~Dkc4P+%K4z3$kUmAmYLz@>IM_+T5DO~B`-z+
zHW?`|Gd}z}+~$oou2kyX;Z^COH{<9vtJ4{Q@@{2?bUA7~%vLI-EDwb;h8epnuWGZ8
z>jn5?R!qlqiU@_plXo@E6#I~J7R}Yo*JMfDLrWuc7z^DG_Fk>YtiwZpU<J>p?O2J|
zAvRcl$<E`Uz0+2WEEDZ_S#ek>GtVyQxVaE4qt&HD>P_*ZF^s$Sid-J99jq{gPy#Iu
z|IWdR$?VGPgimxJUX(FSq%oEg>!59v+@HzfvrgJkRTe$oF4<7ga^&dfx!ZS}67zCP
zB|EF5s0+xpP`mxuHNLO^C*H~#!yF}yNbcc)v*t4?qoPWkNVapG=$Hnc^Z3wC9>&Q_
z>A@H&{bp*rn^{*IZ3-Sa-aAPdymR9Uq++hpqvH*RbIk5b_D@c<LOr?e9ul5{k}W-|
z(iU%_&U$+655rSzG9yF+9@ga<Jt;C|tK6e_Act1b45Vuq?>8znti8M!2JgDv%lYk}
zgp2Ti@d(569t$`-#2S~j`a|lyie0YSnM3l>DTU;9uu&1o<oP%$vBe#Sv9RVOob76i
z9x;d5PH{}rbJ|%xAMBhF^bWbup12%AQ&1Y}V#oC-7w{b3(TF=jjNX3|-b)+v?nX?U
zDVx8;<P+lf<NG6CYy8IyNAW^)^?x1@9r>42_WqkAE_GbQ9WW=xjP3`h2OVj@w>8qo
z|2kg-NkYR_Kvuy-WmUn<idsAD3e_mPfr11JA)n6S+KWI(-t^x~I%J*_737u}vAOD;
zAW@QtH6^mzrCkR4dhMa7UV7`JuYN32^f$mjgMz^-S+X5c<CPpG3N5k0>R_S3eVHnQ
zO*kl4E6)j!f{Ud#`i8IgPA~c6b<P=E4K_rXp)MHatn;q8=#tA?40pp-*K9MwN0htm
zrdvjO<DGn?jWWhKV~w}m1d~iO*%Z@EHQfxa%{1FAbIdbW16=cu+I(r!z4hLH^|m|U
zu6w~Fd}2ieL`XzLOe7%<Sy+cWY{J$KJMFR8NxSTJO0G>|=ddGT9}WZm>#L)}F_hsH
z&dZ29XH6U#kYBFnOuKuOPS;VFCQNr3^~pdX{AqffPL~Ww>XU)Vpk#0|BpI3vQ*b3Z
zW&?DQ7r_5-G%7;}O&NLZ>0{z(a10ZD&*TYNYU@#v7h~js9FzKF;0G4eK-$)?GW;qV
z*_W6!w>f|9?f(DUjz$#h2rAR3N4{X&5%abgMp0iM<n<y(eRskaLtX(z?}}|fk-MTx
z^mv=(OO&`Pei$Y1ObkzRq)<Y-=M$FVf0}NG`;YBVCYL^1MA^GDy?rby9wVu1za{_x
DkBp!-

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-bold.woff b/0.16/_static/fonts/FreightSans/freight-sans-bold.woff
new file mode 100644
index 0000000000000000000000000000000000000000..de46625edfc8c12d6d2594f630350a55e92d49b8
GIT binary patch
literal 32396
zcmY&;b95$8uy$-a8{4+AF*mktZR})Yys<a7ZQJG>I~&{h^1I*t=k}cE^i<7LRo!RK
z>FKH-cSQ*a5Kxfs<`)fu{@sD{d;EX-Ki&Ur5)!KN--Oa{E&hM-?|8S66cd;D)(w8k
zBHvI2rw4D7{H3S@0s;mD0s=n@0>W+xFtF~GR8bQJ0YRMo<_dm4lUpR{8&gzaVg&&~
zM)|h;=7ShgH#=dP*cmx~>rB6Sf!~-vY!ihrar;dS0)le#%@O+#Xv)|&=8hJ2ARwsp
z-<-N{e`YF;OeGdZF5k9jJKvn|*#83*3tLa~Z{4@8@-qlXjVheP7onw@(YJrI|9G7L
z0Y#ur*792k0%DHyEfar(99bK3)za>_$G49A+rJVB2zY6a60?M@gUL6~GWt92*f*Bh
zC17yuj6A;MT5JBVEio80RKLBEo!PhU+aA>)1Vn;9I!poI(ZS_62#ELdx83Ep&kb%$
zo(D%~vu|79|K^tQjb(waISh9*<L@#1{rQ%m{=>SzPQ`Z*`VEk!%cO_@`nk9DZ(krF
zD|1mF9zcR<S-1dZjKTTaTy6feobs+O{k#@tDHu~01X#^RtKtM|UTJglCN5Yvqz-9E
z=d~7yzaxIQm-}Bk;FGRUN8f%tfQOmBw#iEbBmOg84{(QI&H57^kP|wdN<N28Cpe^}
z;R8Y1HBe9eqvn*YF+;Y~;iN`*CZ+3AW@b1I%mshztH>rN7e%943dutCFfA60jP&fI
ztupWih_ZdBuXvuCS{QuWgq!{De5dc;r+H>x{%*hgU3pXGN2N)ja8C6=7P{}c8bwKG
z%wt;G;8iO2_mt;YDQc-cY`1=pskTwC+0*ViLaHVm4xhU4eZvC)efET;;~`*z(UAP`
zy;M7Ql*vhj;%2(yxZoKdQ@rRpOP|yn5?hAeP#a4h{LVN$avWR6-jH{Xw~B?G44n;~
zh<~i2t^T}}F9Zz??R^Ve2%I*gcGRZG3@QlL2>RD|*2&Q4sn1|#p(Of4huTIKiIYYJ
z##G>iqU~1UZ8XWVsn0X%&+<X$;S?O>qiBjgb$Z4tfc1e@owVM7)~r<5K2~}VKY<lJ
zRY^(`J(h(9SC)I4+?3x4l?Kfjl0LW~x1>$r(qs^cOI}&VMo%25D2L%$rmHCZGlC(R
zwoKUG47E#ryl8NcV$y|%#<m)6%M*U)OfSQ^O)N6*%^|#%-3d1(A#~SqoHiM;Y*i<&
zD(p=SVC<`c>GkZXO;ZJH7bV4#bX!TuX%SvxqL8V*wU9ASesEv8nB^x&!QzLgt(RR~
z$^W*Arzc3B{u0Wqhjv-jr0}G<oQaO>xnG9!c0|(5u(9s`vP^b_b-6O|F-T#9r!fKX
zAV0>1(e`96<0G3?q79mv=Rizx9GlWav4&$O+2xPM1h>O75HKn=@1;!QV%73ekMpHX
zJ7JnnpNc7^F>aksYjZ%DPVzJs%qmF;$~C6G_A^%7vE&@Q;!vDN3h<SPpAdGyyh`j4
z<RDXkA0Oi^YJ6MkLrj$LMpmmO2n86ovP!**#D1=OJjr9#x0d><>E-Xw5SBM2dzByN
zY}nPc70mfTpPK?-HbZ)c<|TGO^US^hl@qwWE4UZiTW5)zr=MIO;7GDdstM^3(lWre
zBhw1^@9&=qqog<AKM&W)*BsBhKIoog{QVzR9#Efn%{FI>!#U0UB|8g_cs@|}VKxyT
zy2Wwsv<^8E+<Hx={nbk8(%kUVl!r@yR(g4^wLJ)KR?xSEJ;-+D9T~Hrj|U(fNpv9?
zSkIc^D9;+90HvX4I|79r$XbY<Id9av;HwsNyn>K9q3b(rj(i&;kvVafCD#^?RPQ_+
zkp7_V0~Wh3*PipAXrFMOKQ05Ct|9jx`?q^}n;b9+RboeyCUT6x-{JWokO#Qt%22z-
zw<xm#ae^yHKd&RW1?NtE_HYDI6KPG~U;w?uXm1$DH1-0z3&7v<Qkij_yB|OMm~j;p
zOLL+5sSOgTD407IUri7D(w~fH*3&c5ikGrf@q;gpSQ_y7V>idAw8Z;;%z@!nCGL7m
z#~(9;bHqlmM=@uM&kFJ`SUMOI!8q@|{h`f+(XHy@V2BtBCPdwhYIgjs?c&btwk%YZ
zT`QGtdX0*`lpd4#jVr49d|^n&bbW3QXmyi|s_RVgi^Dw=j5@Az2+i9EmKf5_fmWO&
zm{iwUGs0l&6qCqPbkRdWx;4rOkeP0@Tk7~+-ty5~hVcmNlShj#^s<+;!4~UWEu)3q
zjjpN3cWc<_(pNL5;D3WPMbXt?{_+?iIex)iIAeLc?X4tEmJ}<^6>lNwBCR(Y9jnDO
znv-!Q2bV8!ve$+8o3o8oa^X!)RqBzi9k&fnb{uWWWpc;tjddH8&e3onQ8E@<+}HSH
z&C<p*`d-)(kN}>zMG2>INfePT=@ZCagniq@kD+%4hkA$YC62Whgl!oz8i8%0*TU0$
zr-|9wj~;hbNs7N8SNYy73WmXer8F*+%StWj*;~0EWkYMUc^Q(G$*MnA7Mrn~nax{J
znlXPD`?id*2<)!29Ig!oZ#kT@jS%oXmT{0qCuPdV7=!<~bbb~rP><1403w?6Y$*@u
zA?}*2PERXvASpFWcNR#6UZqi#9lL5dhi#1%5}F^i_P9;NH6!2h7gcEEm8UkB6+r;X
zRqz*67L_u!3Q{<(%;`PiY$g}ut79EoGT)H7ZMh^jv;5dOI{XD5eAs2P*`6vS+Cyv4
zke5zcA3qgoT*C4dIBuMJ&)ETI?ccY+-PTNe^@3ykY-Wd6zjy?lS?ZT}i4^{b>3f+;
z{N|m{+GYtntXb57DnKP92Vla;2L1L9ef<)E9z3_cp}SUi#;{Y<QL7YX$M!d{0ux}E
zF0gulS<P2F$@<=$_`)4&S3cL6aqV<&*`$BvNalC0V<SCo53e$lLZ^I7j%M!`X(bOh
zlbKEUh@;j09_>iI<kP{1P<Vsxm5X<s?x6oCA#u|y;4lYuCY7)&mLLM&xKPUZQV7{C
zTzYX+;*It0ghM&}FN|+tIY28C{!Fs6g|;ICuEhNf&3W|nxU7!k^J{&DF<PRtB~!hD
zN~~I(lr>iGsZdqNZei6f-K1IQ4q?lykjQ^1#0DT39T%@F0LLyW;k7jS35KTrJpa1r
z_55zlPCFoBbn&)(|2b4;0_T*U%ngkf(eR6V9dK?{SRDD~&;sNEmUe9FiM)qz4zYwF
z?w|K_23w}O)Oc;PzRXD9j((*R6dx83jLU^l@crpBw3sfN5HE>eUy$60aPnQ@aCIoc
zHWlD}{t+_@%XUbE`!=Y8)n#e=(N?CTSmREpm$h{Jac{qgSI;MR5%9W%e7i>Hda1UG
zDFxnbzuQvKV^LwTUCR4XV(oS5fGhvEN^23R)%L>X2a>h{-e|f_ApJ)67e3V;Hi_4l
zY@=APCyyfXiH;S)KN|y=!|}xi#mDWM298r{Di=yVxkKK44^EYd?q$X=kI5z-+2abA
zTleIw#9fu*P4-|m$g46v>zrm8{V0S<08&xSW}ycaK!VB43gh+x$I|=36Y1Yl!s(jR
zJb=d5VB$>Dg6SlUoy1e{ueX9-4|#3R;y>-H=oheUdliCV46md{i=K3&V*Zz?Kr0r!
zNu-7-wfVKEW3id$qUK{roCbL9nk~uEM;Ck&zWcLu^E>@)H68!q0@N-~r4(BSzs)(?
zSc#jcNr^g8VCmCMX~BvXc_Eb`l{}qE?+xfo<A+X+DIo3=0`)I%WfAMj*2nA#>{rH&
zPJ)XjA>PF~?}9UzC`m}O(&v)a)AB^4fxo(gUyTuyS9zG;&s4Rlv%Ix!o|qHujv%W8
zezFUaBMyQb#aC~dRCfqE&h<CGdA{zzYR8VHe*!BlTNBBAen;jgXDeE~_s4Vu#|3=e
zwR02EEPmeXB@7im*!ld*f5bR??}5I~@mTrN^2t^2pSu)K`oz1i9*#-9bWFyrZGK|0
zsF<4xc%#)C=uJ@A>v@$S%sCW?No$6x97(mw$rP$GmPFGP%1M|V5>D}2((iKPGNEUk
z%V4KWh*c*Z(3TNr(W{u2%QbxM&Yo?b3*t?Fx{JmCyk#nNmCA7lEeip#mU#czFCF5B
zmj36E{^#rbOI`&lzWE&glvSrX#l|$Y1C9YbZoTeIiry9Xy4C^qDg%TzPET4o5mTkt
zPScLoYzidxWPSKLg{NMw8gk;07bOX=x5ybijohX0S2}d)s1LIoR)Db<%<u#$%5qI%
z*n~235EsMG)b)UmJzt(5YYx68qj1dX>Op1oaB_(qy}w`JYPXeZM9w<;5Xzu5#<ZDL
zmIGm(ky~{yA&Z=eTKoMbVxc#YC8RNu#=s35{baAccOweE?g6BpNc5@4X2=iq1}o5q
zkYxpok9cpjoz&2?FHeSro%A*Nc3qvZq0k}`@2IM$*v}NNkbV*GYM-G<k9_Z}RuE>l
zSTHi^)bV7*^i&5#Cn(?~o<tA_KT#4AiE|D90TycRORiNV+9=#cg!v%n)*O;)w~W6m
z41+#rdd1q%RSokxr0cMB<c6S-bJMFXmKNQpy07UubfrEX$Iz@FXnE`i%GKVQ%6(%_
zeS1E?gfSR|5+2H@hQGijwUu2xfUdtLWCnhr2s{-pD;G*tii=|Rg<xRoVPKOoVl0TV
zP@tU_cjQuw#w0XXyu$hg5)u1F8|?OvObeA#Ehved8&C6Jrn9kOLkZ)8^Y<epIz^)W
zpe95D3jUCE%E4(=jOd~!A65da0sy|}U=6q_t9CoNagrH2Q&HGj&Y@MiOJ+gEI|{sQ
z)3xr2e@nP=<n1L-U~+Wl0&|t7@Zp0}blR?KBkUS$0U%Vb^LZ`z>28}9{wU&6RQxuF
z;<QMD@=J|J9O$!LRutt)W<daC=l1YxM!`WmHPTMBpKDdJ6|#UL<HIna<0)J$x=IC>
zVKK%sV}d3~wXix#mj~#S!!3}yjizBO6om_9=$<WXnqc>~=+jPkppDpw<xzN5{lHsT
z{LE&q+Wk~KtcscDXZjF|`!=3G=qN9T;(vLNiJ`fYA#k8~ytm(a(m5qC&<N{~IqyX7
zm?>f*E_a%iFvbKJROkaJ0{s@kuboRMats^{zlK_tcW_o7)AUFT930Bv0~{t+%pfx@
z5M65o?f>y<XlNc_hzlhFg~APfz!LOAkAN4@2o($#H`MoQ<T{5XtuE*v&H@1>BP1#$
z4I~dF1tbTgFC<-vM+irxOu4^b!}}NVVqR};01Hfcf{2K)h^Pou`8-;6!p+y$7vPKN
z)!=&j`0A^th7jN1@ucq)`i=11uXevL6ncV)U5L5I|72Nqx#X7yDkRPt&dxY)dH@7^
zJ6NuUmgd?j4|iLur^m}PBt*y`aj~I63i6^tGt<KZOpN3tHPxj>4)&%-*WVBKhzK#!
z@^Vv?4D?l%cD5Ji1o(fm^mI2j1o*o;eSJRO!NG!qeu@tClaUtWn;7r!p`+mffqWfQ
zGifaza|vd$sWrM?)>FWTGfvx$PS=TMg7v5KtqwpQH3i3={ZN0<IhC0f0zm^#BreYU
zL|^qpnUQ8)1q%=I;B-}2>f$2r$Lwy!iv4D{p9$4K9~lCGh~MWiEARG*AT%ZWXsh}Y
z7X7VQ!6c2<;?g#>U<58OqdWiyBvLlq@4rMD1iTdz1QY=j6XYD^9^@J19TW@{3KRuo
z6O`}!k^tET*#(ILxdEjBg#lRxIRc3Q*#|iRsQ|eI`2wW^nFkSrMD+mbDRn65h@fmD
zB7+AGt`hSAF&fh2hQ0G3VF#CDj7|AM4q>dgoa3XMK#7C{#SnotxVd?SQkhUB*it^C
zf+T2cHsSf~e0&Wq)9Gxo<MHM0|7>Pdj>+TtH=(9DKNO^z8bF>3&(30^8-t;>ZO!k;
zUEe<8p)%)^_^d$=1srIQapm~}wHr%5=icE$-`<Jt_;4E87w#8%VgHdp9A!Gm24CS$
zwND_w1}*IxGcy;a74I<$6)S(Xupe7b)HcP>bq2GczKpR8tJv5<WPC1?EZsR_$i__+
z*Kww$)qj$s_l1o8R#R!3d<ORE)i<0N2GdRjLaz>}dm<1bAyZ<^w~&w1BceVUR}8|S
zCgQ{K9Bzz#77D*ep$&IR%kb@oHbD)?uk_|}DmHxaXUf|-5cEH5;rF@s;L7qokA?zq
zP>uPFX#!xyL?zT-LVwJ3(L)_5zs(3I1svW0{Qtr<UapGqr#87Fv8cm%`K@o<PDFO7
zw@pWhx;vCxm(J5nz3ESPkAP<r!Vid;^kvSxeaZyB^3zwdTOvXH^y2dR%5ORiujs@U
ztfyM&snh=NwS$6yetm%;djQY7V0~1V)`a|~H_Z($>C<f!O<B_sNG9Ua=@DdC2fATk
zKtWOC<>zQ=N%O@cOJQ1<mW$M@=9eFvo)n-|bt6lg!n>OvF!gP!ik5+@;?`B-a`t{R
zE|-7D{*Z@XyuCeqd<h|%9&Pa@de3-HyJTPR&pe^TMpOOKVa%5IFeY`SrKm%tN&NMw
zq?<gr9W8Z(kB^o-GC``ZkY4!fI2>c>43>vnQq<X!U!z!U*1@J?utD+Dvh$iYP$z+|
z9Wak>%vOg2ttG5%E?U0ODa<d*p_auGg0B$8xQA69)8f`4HMwLJ7Cb82v=?4<aW3UL
zXN?)lAIVX+dgwN7dPD~%Eq+hkOpdF(y)lb1bN5zwE8`H3Y)~hm)upXvQgk-#*J$(D
zAJTA~<q~@>er#?FNC(`|)I~ltq*0R>!$h!ShC>ok4VM|mmSJflw$SI3%6o`gf5x>p
zt;G6C2>>g6qO-H|r;M2#YvwQ2`uU90FyfVgp-<LNkhf8^V6e$RJ|IsUw<c<M2~3lE
z3xE<b<upmChM5pRCtCT}JvC_$Tl_iSw2t>WDnGygL;hO4VG7BhP)19RhChpOwoGY9
zUA~dm&fJ0~5x2(<jbZ7+(;R(vvc`YnT;1-P;jeupRduLk)4Xb{!@(I*4cVvlS$BNT
zvj+tsb;zY#_3S~F{qb7GpPK79H`zd3q}z>)Wqv&T=gkYG9)j*2^BPv~hg~{9s@`sJ
z|KyoZxJ#|1Hbti+L6LU_Z|S4VB9FgB7e^lFZ8{S9n9#B^X9_P$64-mQoRk>@F(eK_
zBinz3Y-OH}qMfq3;^VkfU;CfkEpI%D=nL}92)>s(oUi7`5qQ33ONW}*YtbLYW^9cu
zf}R@QsntuDy$0Ni)rDO9SEw1#^l(`ymq*XRdzs{-*3#gcGRkJXMKaWR%jvwzn2uTX
zE{=Wazif7b5U@+<aa8ET7Lt*aLFMqO8+$>L_`)%z^#7@!8G!`7N?sZY_#Dg&IS<X-
zyA^W&3uZ0fbs*Z+gpp}ga9W`;YSF%Y(RJ!j`_7bo3*bIE5e1w24&4OL!G6F!(>C+n
z$kVvDOdQ%fJaj90`s0Oe+sq6sJc4vbUXb_<2vzkuSU4=}4_&gSZNgqDt-KK;dSnER
zByZ}vxoA3%XBU!^Czltpa;FC&-{70;>nB7v-vmwk#imBqK(rsl<>CcS!z=JGp_G^Y
z&|PsHk5gV!WAs+_mS>e-Y-dfICofwZy;$6X-%D!t#&AFj-OR+kUcQMzl#lSp<uMwE
zzfM-_7Q22Ec24V`O4c{f&uwGXy^P`F;8c|u7$e{sq%?50nve_|xSS`*t2&d=cNTxR
zE|MKlwc3=t>t81mvd-bAp<N^T6j{GrF1ssbE(uqM%D~#<?)lKzXsE%AU1@X@xj=8T
zo}6^u&*6Wa4ikTOHyMEc-qTP~M%NH1$%&!;^HT$oE~C>qC4iw(x{}K)a~g)Z)l_Y0
z{fz}i=*^}VZQ2~~$_REORKo(#kvtY=__5rMj4V_mSZGe>k1HmsUIRVJKjfS%LLAle
za@1Ov`W)5EWd})!Q9l@S3lkM7F{RZfrJM3mrlTfyeQGOL1G_mmTYqV0$d$hkawkZk
zNu7yj<Tuh++|NTC`g+1j3^+${vM2u|VsD5-6gHr7Hhg+|BAQhe`UpgMIf-gPn#^E!
zdpyUpa8?vjEV=MkkiGE#sD$0On8p{{rT<)l&hg(34Ab|ibE)CpJAi-B!LHQg@ZT+0
z)0>9Y_u1{c>a&2?sFbVH4soe2e=FryL`Xnm4QG$TlPbqK*7Xx*z~)ot%+3AL;$CYd
zfmBfxi&V5&rJ78T63>n)m7&=#BSh<3!9pChAR==gQIsOntLCE|qc)cPxOXdqcblwY
zpV|e9=_s{$YfLPW3vz;y`h+r;uAv3Kf@_v>W@+*hz62enwVKB?&v;RK;4m4rP}+qe
zi}<Rk_bHBF4Pc+v;&n29#bD#_a9qxkT3JHuXesLb*de>6)$#;j@ANuqmc3l#GzV@p
z%>3ZptDGY%UQk*Q&_{cFOsHF_wtWFyVEZ&w+8*iv&F$`Lh4lTF0u5#~ou8h5&}lf{
zjy4QnHY665Obg?RD>*oUE$En)S|KEhi=+}iIWqFA3{@V@H>Kp&TwkrY5*)5XNlFvm
zpp`@5XkEl2apZhNwDc|Xb3kY1rnCl>#U3xqciw&yF>d%?lC7Ihu0D>obzpy9f6`IS
z7qqB+IplPBj>a1y<Dbq8{^PyxttI<u?{(R-2(UZ<EIgxr!sQ>B$&SZX85}5w&qg(z
z+0pTEC&0F2%nMn(+Rw@wYdeUyW3#aLHXR#C{a|(79i;1dI}yEz*xeRb6#uIi*3I`m
z8lK~ZF2Cjy_Sp9w%_>|08u{Z76K2IO>>!?1G`n|Z7pNdgJQ-z81mKnsSlH$EB;vj^
z$mYhY`qt)pb<8+Jvehu{{eq`mB1Sc}9Loc2Ote$D7|{A%pKIkfV~w^0xK>2ROP5yW
z&nnO|1Y=-y3y_cdUcKSt)(KTA!!Q}Mo4KVx(iFo>=+)m>gRIxB`+m~Z1=Qs)H1UuJ
zdt$(h8MSHWH}L(lm3|YU(tASHD-?K-e`MB!m_fF-iVY^CnuKJIoW}}GbF|fC2@#6U
zz+csM1vgpuG_-~P!O#09u0w62%F=op)S#|iZLibybNyptY~mEE1u47Xb_;&nKXQ&<
zdc}#!IThq8iR`L$aU-y2l%F=RY@p-X6Y9wW@8>7EJG|b?T7-WG%z2CZ34g9UnY0$d
z*%XvA<8fL%Wb3ba2A+?pTwxc9)m{kg-$v{2OEyPMdLG*wm~=<epd!K@S99u4e1mU!
zb}mw(nVx@}J9Ua?d1L<?ou%R4osLAg#z>!HR4b6!Cv*8$J6`K89P9Gs;pZr^kGimn
zY0UQgQgE@Vkdvp<H|MnSeNy`5m{XYS6?c&yhY(P*BX;w^;3LG~>rN9z-p45jsB6W&
z5#KS5n=&NqCw7z`VxXfm54<H-;wQFTk04g2l;pQpOdpvFbatlNW6&+&)7Afi6}Ud`
z`TVrpi#Q2??eBuK+p8|J0|MH+^Dd4q=grj^UT&aX|4i?%=f#hR!Llo$JfmJ9kLOL1
zkO-CHHx3jct3l8^IPtl6KeAmA>=w&YDc;Vs?05tR$Va((Ay1+P|I7-q@_4$o0n4+D
z8azm$@qE0h;r3{BJUFna5lsE;dER}+cPi<2{`h+Ae0bUZlC22_{dy2Qc}uw4I`h9O
zP(f1UnKkh0%Dy#^=XKyfl$GzPtX9(JwR>|j`dH=Z@hE)>S~|AZ<&*WU@O(OOB{K&^
zH8Qp*+zdPJY+_tr65nzpM3Nrg_O^wLwh5|80?h$fyaAc<5f#RfzqtvP)49*pgK`Dc
za8=9jRf2oP4CMl3v6mQ_GqduQxzokK>CG77cjFb+!9_fn?x|mTmkLUasME&fV;DNC
zKjoB6{4z7*ri2xdYW!Z0)VS=s>`yr=dYrTdvUn1SezfhRyFY#Hc09cwl;Lf!KN#=l
zSWIRw1Bbo0u0z72E3FB+DvwTz!>&GnA8PYYJ$1VgF0h#es5`q)A&G`IJfrNJ4yu*8
zN@%NGA%GHFxpN?Q)TI;<neNC#!ddDD!(jyA+1KwQnGdY2%we{o4nGjaF;{^3<MM~F
zaKb*ZyU2I>VJ^o?KwvDfx*zERQTqeQLfi`g?IlF(1J=E#MXnZpk5!0d8FAX+URWte
zd3BLuU0AjNA%+}u#9av$&`Ci|2S&-vu4~Wk8Y234R-L$!EKZIUujS7vM#u}!Lr@0u
zzd;OP%~&S;H7@tHPzT-5o$>Ovjde<SS^|y(4I;NhlPH6V;zH8yLc7a(O@@}^1B>OK
zLS8o>swFy=Zv>yNbpg-9%S@y+pt#^!azp4Uce_>O#di{ExFCLpP%ilV{7o=F&i+Vf
z^m@k={jj?upEFm6M8z{mKu!&3i}1<MU}(quS7wPM6-1NUf*Y0{Woj#ZfI#y|#`}hB
z;$KZr!@ga2Hw|vH)%tLO&R*YW2TTxWJ57b0$~rFjy>T_rQ{!P&V)aquSAU4TGH<l4
z<vC$?+*DPHDB-HbjrP1Fp4$WArYJ)$$8|P7fuS<RU7y5k1E5(48GVbV{IphC837V;
z2YyxFxf|f`3xQvt1v>f2{!Z{w7BS{_zjdUKGD4Y>%GV|zN0CSZBaRGLp#X!1Y@f1Y
zlapr`Hq#>t29!vMJzCZ__dzo;Nw6&(DP;-QYS%J9q0CXHB9CjQB*)<DS<-^SkHWs@
zk;*a;l(So6w^@CJjR3lDQ!ALd--l7p9`!sm&x(sGX0$vmQ)EY)j`Z5i_uEt~m<soq
zTIsyz*o0nda{L_rRa*`a*vzzhPKAFk$@dsuHqxXfX;ki7o`skt)@0Y+<L^WPFUj7S
z^1h%9nAJ3(5R9VA9_hUgZXub)QB<qxs%4-Io?&l82-1kDD>|WHf;=%tw`J0^9vJx}
zTA>m64Pz`-#9dU!)28L8B(AK#KDQlmG(QU;ONanpJjn$|4h}sZAc&{@b&Ynf#(oDA
zi-ldkox_$tPOJM1FqzKdmcLde;2Cl7@zq<MM~b3n()IfN@y^kNPC9gLAd=jHW~&sQ
zs{})YT()T9)A+cYR@JV1<|7_yO)8AGnJrLTY)5ch=1=8n9UAq=Y?+AJ0DO+*0Fl3s
zI5YI&G?xQbJ)gooI(1|V*Sz32b<sS~et*Y<*$_AKpAVCUf=Tmgj7H4xl5O_n*?uff
zEbb$pRe@(Bc809?W^<sX7jj1y|KzMZv@7&O{bbm~<G}mC+x#7_|M`OYg-4aA$TsRG
zIK)-Is+ng`69kEHCx^o4Y|j$8$IljGuKA+1JpJ~3xU69I7DnU@@B^9Ahnwl9yR)r(
zYRJUvGY&guD$21^?;i8HMf{4NY0^O=5y0+$|K1_>{&h8FkY3vHP_O6C>#JBsI-(6j
zE~&VeGODGut(G{FgQUP15XljLiVQYdyK4<ey|Xo&tWhaHNLn}-eJwXjuBW0HgIL02
zqGOM^M2pkBev?zMzD=GDz>vVMZ4+NiU&<UHg};bu!Owq}RFTlJk%$6#7c==aby8Cv
zr!O<&)*D@)HbYiuFj?wM5tSUB@OAj}kfBVrEOA6p6s-VyOiGu|n>~GE2WN#XC@CX8
z^z*x@1jsoPP)$kOa;V2N+XIInG5qNTfC@f1=)zF6D>Vw?MN#(^6i#N22$lQ-;mi@Z
z(<}m$7%l}msf1;81M&78U>52E*X+1M%TjA>K+@MJ-C(?o#Qikcb88TNTbc3#k!QN>
zLRJw4ojWq}(0jUDY7t*^E;W#{TmJUz^2N<PiLC2D7^Z&-l}Li9<kMGoxz6B-pvnp>
z@KgGQ#|cwCc@dIVMIm7|na)!mcN&)nl+{EHH+?hN+4X<C)L$|!YP=jb6~DT_cYz}_
zH9DV(O!i%mp+W2R9}XA0;2SRsVKF~g>{kapu^0A*3S%T4xP^*7E<W*MZ8N$3r$^`B
zE<KYoD#|+qyA1ykJ#OFce!y4ziNEbs>unzVi>fFYF({Zop4L=VEJ5)8wUP)%VhnR1
z0ZSkKBZ0J3>luQraUmi77jr)OA}cjfdJ0R!&RyIqL}?ld(OwscelF9T!(5KFBGVjc
z6k#D7A=Vf+#QWubR$~72noM{ta`rBTc&_U-6%ErVe$Hb2UR)7fVeD_Ey3}Av%)Ij$
z=Cw2MH645nA6o>qu!^5FKg~>?T}zW4`;umWB8#L%_}mFHb<r$0aLC+bv)nuwOuO|z
zLXjoc(o`2D^d?jns&pTwW@@Utj^=HyXtO)*uhjf~&-RvjzJ@*zD`s@=;o}EWD>K`V
zj_Tj9bOp_)JWUF66f(SyC)lW36qa3YDlZHA3z#MEgy~9rR<TPAJ><>v$k5el+wMz)
zP6x5ZrHANC!Vk>ZrVGMHrR>onp*;?udZuxDrOZsRx`JZ)U+3$+{hvRuk>9y}zn;td
z+f}`#J7U(QGEat%twyjvtY^?2D(AP#WM+rBLxm0M;7)9ywAfTzP}QP_a;t3vu{<@S
z=!L}`_wY{!MfN*axz|X~Vn7p;QE{~gEu_T&^SJkL7hV_fRog)CYYXU>2(h>LN1Y4A
z3lduXyL)&=PzFa^o82|ew3cXZ)u^W6IsmOE-+cbGC@E)uH5b@0W&z|Mad=xpR%a&x
zTSJfW)@ZcbvmTunqTjxmewAYzv>uORGmX=UKxs~>=M6BA>X7K=92@?`E2&KJM>#g+
znMQNItvk6b<3Sq7K;k>$M+Z88ixrc5c?LM-H&wp84|%$&Xm_e;capZq!rM9k;}(uX
z)NK6Y>#e6pHujrlA#~D?r&|jxc;1~b<Amq4?i#!zeW+|bfoV<1jHJRScx=iq;*PkS
zyM*$YC=sK$iO7(brCwU)P_;k0`c{KbvLwNJoGJ@Jn1HLdn~Jw5-MQkfQHD2;=!Kak
zyg-YD`+lPs;0=dEAyTK(3&LT0b5+R%D?5gO?YRUl+jilMk0aB3XR=a!i<u-6&W~Z4
zy0*D32(S9K6Yh{d!eT%;g-wWyGDgJFlfs+;`jbTnk1bA*lL}G^fuB4k=3;!RgqNW`
z#*eY<mUJ>`0$Y$?PH^v4=#|S!e_y<tW0N92y5I@;@lB%@1L!PP1P~IdS2DWfhzCB>
zE^nhZyC2^htj>tG3_e8@WZ0Ot1|@lOuAV!yO<ztS_(v2%aoAZbtVA&--8|Y_PbW1u
zX&vNG*`%P#EJ6El%vEy%{pa!A((9!R2V(QC;Wp>MC0Cl%6EBWC`cFNNl(hKHO2svt
zFoi^nnbs8SKyNvo!C~lws6zBX`Yjtsl6ko0?~~DslzX>|thf`awqO>@4d{1*M%?*)
zLPob!r6TYKK%Fa2NQDdSyWf)er*7TlHrMT(NzYQSym_nr8dG`DlS}gc)HOlE(G*sN
zyur=tKwybg)=3n;+r136_HS6G1gJ&G*)RhoS%V^V>jI(gkAXDs4LAVIG2C*C-OG6k
zx^{4-e8J``h?dyQiCnVH1}u@O!m$khlrk-82_ECfW@Ur1Xc`aJRN+CNrjP8Va^RE*
zC%vKm<v?>>@&LT7e%dLUxhYWa8&|ya(O-#R-X0=g-Zn!c-<KW_B-@8B8_pUfW$%?)
zbrI-=Ce#xmxl8WnbAiAI6HILbq(n1;K!MYeBk2NpffGVmMf0x~4j(pNA>|*(4IMjw
z6}<NHt}Z=q`=IAHic>B3FPkOXul%LGU*7CGXx2U!JFfUY7w1|cNS;Ebgz@Mae@C(u
zu(Ar-)uAqyV!5clugA&Re)4*=2b?zcfu!ek)iX$<@4PCHlHX0q+1M93D5Enh`&Ici
zj&&Yh`EQdZ^x(g4j$}0lcl_Bm@FKJ-E*i23H21q!JVtBgF2&MB$Veuh60Jn_6mH<>
zN(`22aQ0xD2KPKk2q8!JD&LG)q>Ibrcq~*VeJaY6J}682De2C{D%cpzxVhmvdp<UM
z8~EEnTd=Il_pzl3oqd=kQt)HM{Uljf>2G#T;B9NPJux{6@)e;U=#Rk!SZ>ZRtS=_f
z;Ey3F?+_@Ly}*(f*rQ<(mmt5A81QUHG;T-y+L<emxt{E&je-rL?0tHJU+ahlC5ese
z?PFX;R|GE;@}!=3>1R1*E99?fdOH1^B<+5;-~ve0?2#e~#66^CUX(vYxS-?Cf-FJU
znKyqyS1y-EwVG`6)q<I&t2nibG?$@nACKpLIL_O<Xm>4LxILcsYEbOi*F+-(El_tC
z?{HhNaA(N%wOqrjE~bzdFg%K=XAa<&MIX_FKDoq?4e`Pj{uCjp=QXG~tH|9@@C|W2
zv4Y^25doZ9m+lw1Dx3Ti&!_KQ39Q+DN|pj`hC)ynx=z`CZ@=h&k1R8hTvpI$w)b&G
zE%Gn!){tQGJ>Iy|xdrFQi=P`^RZEqNwIX`=?Rb3QKkal@t<R;^+)KQ8Ne-55X0+e%
zepbF%7#xk#9eUBv8L(;bq8*v+^x0Fnijlz8g}>V)IU++KqoJ{}LB)e2o~sc@3+9)O
zm;}Y_>y?ovC6sA}Fy-C|_54Mrn^!fT4R(yeu2GUHRDHH7&m2n}$>Zp8lWD6DI1lw7
zB4L-A7^UR(8)(wdR~WJSD*DhnTlA9#++<e=tzWHrL4RO>IvSAlZ=Y4=IDVxLzt30}
zKc(M%$iK4bqjS4@?o4XHz8cV;FfdJ9UhMpig6Ud8d6mJDO;zKo?oyl@HBo17s_Y#>
zd$o%Oz>aL_A>`*f=jUdX+DpL@vrJk_{Yon#Wt)o_#Oe%4PLIK^pVwlSu=%HJmjHDs
zeeC!-#Dv2%Vpy)*XIn>yptEEwOTIdw+(}IF+ne6Rr70~IV91G$+>F^`(3ByFjk*5m
z3tdU%-&d`3s+zfkP(NN4>Gv$RI+KV&rLEL`4jC*vsjVwf{#~IKyAeWMv!s^5{&)u5
z-wJ)DP0tc8`T69lcq0JYvwL9u`<8)Ao;1xoaSxER-sk#^CQm?<$yp7aMmNY!U=y=S
z>Id>HRb$(|F^&toUcpqpq77dYpb5EcYQq(%{3^T>TkXnTB|mWd{F?y2`xIp}`wSNx
zUWZi}emXKGD?$lx)c#c54Fl;O-@5ouv(<j%aqmdn$=LK$Bk8N)hCjwePt01BYpA?k
zJ+3;8i&-Vh(L)*KCL?5+(y~CM7@2quy1s<dit7c&xt>mIjN~)esb9k#_yvv*U=izP
ze<)KJX;@437Y3i>zqC^&k(7avpbVOL(}P^nA@2hs7wZ-mMN;`j@nT?XBJvmP9VC)f
zMXQCP)?1O+tjzJS%Ca9a`D^8evhPLemBR)0b1rJ=5mN@`mj9mFNtlYxAi?DQ0EP<e
zeejF~m(!4jBBQht<qXYZu<0vmfI{xozdwTXhuNnY*Nv^E`-5lEVSmg8sV_s`LNz5w
zW?130m{-GnPopqMqc}r5cP&>jp6M7q?nevy?eo(0OYHp47;;B$>|1Ih%?dHi-%0C&
z32KYHHboG?c{cmi3^qJ;4UjQ1+Cq%sKFv}3uAPti);dud>gsEh1slMi7;qGh3B5$*
ztaRqG31J0~U_X2Mf~rU}rOO9L2g_>R3vH8#W!(zfEF-2s!Hoy;pc{JIDXK$zTant-
zkR(c3V=JPpIhdVbj-%Qj5Ofrv7*%%Ydtl6@#bMd=QKDgZ#NP>PebgcbYT8%l6O2U;
zuVQiNjO8PWg|oMhk9|Es(Q&%x92E!^3KC-vewa7u`5R8+NeT)2(`Y17$E>{jq_G<>
z^v4M0s+r);NN+i*K@*%kD%k10eir>+x(!Y!{BV*A^jQ2Yd)C;|qmzo?r9p`J(Sz~(
z8sSEHFMl5vb`-TOqRP-PnhlucswvlZX}DDM`?*-};!2_CQ)v1PC^J+0GQp(xooy<Y
zwekWVw#+Y>Bv-m>&1CKePO3EYT=vtxY-m#p?%6Z$e=px?Xw!U^t2+rr+8JiHuHcO`
z?euNF$8W6WFZLxksW#lJ8Al<Kqavqy?TLdP;Ic+S0ltnI7&2=<fgIb<KtzpU3I4X*
z)<;m^{ht|JX9Z(?HJy(~(j#PQ%$|;d(WibIFknKh|1@jIH92vd#Y<5=;>gac4RaBL
zETS_$W4<)KK0SZ`dF&tlp1>R`Hj^el=bD?~2L-EW_!g!>o6VZ*qPPZOkLUT99>9t?
ztjz!CX(Pg1D#a_IpH{+PFXMbFrKDy>nw(dvKlXE<WaL;8`JqUxV>px~29R&wFDeW=
z@PG?KIqoJ56Ps~Aa;fSFR|HizWCq-UVu<h826aLf>wXZ$C}|k9$LJJTY<1zDMYmL5
zWf(zxl<<9;n7<ol@Gm_^X)>!dZ>A66%0sQ^U3&*1eb?Jcj+NxUU%Z~Tg}4(Zc>ehz
z4wrg@TeOUNO(zSui(*tRTQA4@f%O^SbA%WU;Q#zcq%m`o5C@rUemae)!gg|TsqooD
z7qepwC4tGYtK8+uZh6r5fCx>@fMXXB6CuNgqpD9a1=z#P<_HHe>iDvj;+>SQGZRTT
zwPxUe`0JMgjG=?rP5U8WrhSm91lpdCDr9u4@y=sWVXcz|7xWLZ1vL5Il5>VnQsxTi
zPM#wc6n@L^3lIEuw-&>JG;xM(%RVRnl2&HKiN0re^;$HmyNCKXX;1sX>P{U}caL}u
zrsh`nj=&7tENlV8XrWhD69(M;hlUld(%ia~DZC4Pe+Js(%;I<P<KsR<^VQgQ?Oy%F
zF@gYG2kT@POKUAl)&7?Ml9o}|cP(_*buuhTZ7poydlL%hF={NC`<w)qy0WA`i98w4
zbV(V(qQwJv(H)RlHG3y<S#lF8tGz{Cou9iWqQ&Zy(B0Tis>KsnS$Z%PVGXNMZKvlo
z*7;&1=xw+1ck$Av|LAk)2DmoWkb=M+@iuy#1-mS3T2FKq>CF}Wn)f*7-hFerQg(%h
zUG=qWrgK!C-R$^!une0>6_Z*0H_fYTpeNUCBN<ZdDf_FJX#~1mdX}a5<S~aQICwlV
zN}@Ww-u^0Kc)|_oFSC)@9Wl+MtDKO<Sk^!IR90`WXyZH>)0}80NpoZ!G+P1<*W0!5
zAC#z0PU3ECQA>NXWY$=84Nl>-yfM+yf2w8g-g|B3LdJ82+>71$bGw-B@8`Alk}Ox?
zkv|UX5fEqW|1LJcnvd#Q4vM6aaqjhZ65W$3E69yC4m$d7N1Y!`o8le)TMNBjzDx*j
z(+1oPYtGX!inhk7(DUnP&ghLih;|%h3;Fleo53ID5ETxDJXy3fO$pk3zB4Uf3x&Ha
zqgTJ>9u2#NbL{;}Yx*vQ=Xf5ji1)&sAPGVwM$EJsxlH_lA?W_tEuWNcdJ00X(Np+x
z|8g`q>V7KFd0k@bN}8e}^l_MCi=+Bl(tZ7GTueQF^MyqAHzeu-#GY-hTqRPj8syd*
zOmWvn^2?*`f{;kaCVl#!hVjKTc^7wjm`4<TIoke1@VJrVxsi>oSb)f$o_yx@J^z<u
zI=nEF=ggu0Rd;y~cp2SJoJaln`CZER0n0n8VC`+p>cyxu<E(E^qCBY=9(_+Q7F4`z
znN~eVm@nJp-AIOSnK?ajvI&HlQpi~vKAp)E3O&9yG%??_ZGMM&1@~u@tw-&|d;b{I
zOMjGx??a-cz{c@wXNzNQB)5<!Uo<AP<5cgqP^}4QreD&S1ari4YzUH&Yem?ln5@Q_
z+CkvclW+^uL}-Q|TH@<6@LS{v)NgN0RqlF;cd<v@ym_A5Ued?sv$n5z+-se)Pz2Fy
zc9mQ|mrV9<Uv_$Mp=@S%7Z_if$!4!SOCZ!k{<t=E#=)>5rO2y4L~C7C#H{JwDsk@}
zpoO0sR`-3jH$f%Om(dIndy$iUSZvxDj`20^n}%+}9tP;J!)N^7ZsgQEoJkL$H2?^~
z4CE0Yq4l~-R+cK8U4KH5TM|lG+-mU!=tnM0*&?E|=Ywo@DIiR)&2CI2vPdo!)}$Me
zg=Z(p^s2L+W=d?$JH4$pUZ)6czUo{c|6U!p(5Uo!1baRD@b6ZBJD7R$bHr>ST5mNO
zDXwtX9D47SZQ?Y7YMhX#GW?y5C`!rrOZW{6%B`^{EHKbT!*mZ!JCmEP-KfFs-Tee9
z#rm%G?Tg^Yb7UUq{fH5M&^un+r-wd>4bkh<NSHmYT$C67t;?_l!L6vn{egUzn)|2&
z?w3Q@do;mqbmxwVZ8l|$n)@6Imb%=R@!>}K&(taUe_JxEKRQUqi6`XR>Z8r>nRl6=
zn%h()5b-=O<WLaTufnbU8Mn3`LO=N{7uvwGu}N&CXeamb9+8fImwTo$Yed*nxWwIU
zmP5?vA{2^Z=ovS@RVB5(d3!|B)lf%g8>Len-e9q9=OCZi06d-fn!G2jRj`i)j#PQX
z5QUhT4Chw}@4zTJ?JrkaN`<U(xr!Jb4(5(iI|Zy3wJ!E+(ggi?IVOy+2((}H_lc&<
zB30F2`dj?=hl!AAgnq=xmxp`#)=5`rw=}JY^lWTOwvqKSDvMQf8tXC2LWf3*t4PYQ
zTNt0H(!|od(EF2^nF#hwnU;li)8d;$L^`IMRK(zif(6YwASISXCX}a***U)4+bT3<
zC9r@*J;>1JNej@Uuu{kwZ5@CAcLL@WO{TrJI(zP#=?2@F%|~g~oNycevsjX`>o%fU
zYjH@xoB8pmgQ`(d+{w}jmf@M6-+SDBgS89gwGqm+bx<hp42|m)%{vH^gW6>a4Ob7$
z8WX<b+>0`HYL<fEC=X|r!hVv)y%y@nMBbF<!Qv=skIhMVoe5Uc{4OwxjL>%dlelWn
z&9#fS#1*5Zz(xP}aT*U?#ZUgZDp*v@s6Q|2uU^<w)$aeq!Pht#^bMF<7uJ~g6_2gA
z=P&njK<wHI8q|*agxyOp{&@dwFBX3@SRfsl@UF5t#I?~51Q?t)6k|S*+h~LGTHpQr
zJ^sX)m)O{2IdkEw>y4!ST_2lu<#TsB_ORDmK|24k$X(&Fbh+YS;JwaJU+vE}>I>8S
zCw9TZ`_nUNdR{-^xbU&~pZC4%1<r%zgLc5OP0cRb)~t32m`7-;YGP7I{mn2fSYQA6
zajkM}X^>=&8J1QNrw_{ZB(1)LQkNnn{?aLhJlQz<&nWmNos;v_oIfL<hNFcRS0ByI
zhaTPTqt7%x<u+9ATF01~`fWs|v92PSyKOvYx>y6sI}8@h)0m)l4va$Lc#1iG8QHo0
z3mJu0mBZ?peQ$$Ai6jU1`Kp7=$3OP9ej3v@xf$Mc4}@DRjU)jF(qJ_w_hQQSw6j8o
z6w>}4Iu)%>!HF;SUGEyIOGRSkQf2?pjWodHhv=5rx5j_&c4MxN#>f?;tqBN~NEk=a
zVjx67Pxe|ACh6=jT{;MJhc1RRiif8|v9z^B&G=A&Q-(!%eW(8wLOO0h6|^XNW0uma
z&%(%R_itM7Sm6HQXj7-<(*E=Hr>w)K(^lhX>w}KvG1f%dSd#(0wV?NNt`A?&_7-dV
z`{odx1Ir6NpGNkb`QdZjatV(PG4K2Nk$BFsLWbqS@qm{d<R;}7G^Y!5i++PxDd!fe
zSs}B{ib_j!keZ&dGF)Sod1KdK5b*`E5<BkV(17MGP2kE;oqx4^;T6fkjR_ty<)*Nh
z`psA+QuZOxK-$Yc+K@TeLn><ge)^c9Ge3EiDrl8nbynveP>8Mdlwl{zRYQz~0oG6w
zs-5+x)ctT=lLAv1dWPpU@vy$+1||`uiH#U*Npov7X);8eekWvH8`q<=XGHuC$J0d%
zbi?{}Y6;DiMDTyMoX+NFDosW-x;>}I|6=9*OG%E}W+2)$-kw>%gC@c3DgDboTyIuk
zw^s;Mj6a_Xh%sSzosaPe_m1_7_LAq6%m_0$I~X`jQ(#Ij=i(KxnX&nNjqK0Q{YSSJ
z<U7^1btE>Uvt2*@*$>w_ghn9fB=vABt>QC8QOw2_^I9kHW;Ip@nw(vnwI_A!=(k|P
zg-oQ*X1vQ$)hFdzXwD6c%xpKv9~mN!BAQ+}%c{)=F3{O)R*oR<R&CNe3}-r>p1?Un
zISfyi@6aQd^cA!!#E@gn{c&H&&|+JuOy{wQ@qAz3|08S8z~g=tO(EO8I%N<R$EI0P
z=z2W^_ml{xEK!9U!{k;7R+{0V$JlWGAXjflQtL*dGa#&rO;<rC3QlD=AH_Y6cprDY
z-Td!t*V8q!4CxO3<0XfV^4VVSB!%C-BaP}G45N7utcYObryD2ui{nn&>g50z!H{@A
z;KSd>lnDB!u?xaL#W5@KT`fVbZ@V9LjV>Mkv(r$Ta^I%L((M^+%_ZfW!ibjoO6EwD
zv@?s8lP7(Waf!tiS;Z^~o+}V`FW-nMsv5dep*<Xn5FR+NPM0E&v&GvL)8S=u;mV|O
zRq4COp)FKxwiPM6_DbM>E$_bs9sHq}bCU>=KUDlBcs~^Xk}2~Kds)Xdo#J;71&lF`
z73tgkIH{SGC}J?s(a`J9%3^$q90!tQlX9Gttk++8rB1zSr!*t6jJiFcza=%w7=wY@
zG7(C1n6q^4GWTS;eiR5wszmvDt(Svm^Ce>H;|VfF7JvaARUC5)X(g>>EiKDdDEtpf
zWs6Grn`9+jj8qo)l?-S}B&LSw&_%X3B1B2_WZa5H&3%MFawNLt^F<OMArh*rOA-AP
zbmx(WBsL>`FN%tU-L7XmvQ<e6%r-UB)9tQ^Cg$A8G6N}<cu#-dmhtz<TNcbbyPrFG
zell)3UH&ofQ6>1W9X2~YGC1F2u{~%sEp!_XQLDJr(S5t-a4qf~EzUFZ>UI=>gctS*
zz(^E1ID=^PZ<k9wO|EM+48Nn*9T8>5KiJFFK|EVZlFIMXtWYENesmN(MXo)Sj$0rE
zRRQJqCQlsNb#4F_vNDrBa%b6}(D{`qtHb{ahtC%HlQ@}T+|^&t)0=AyMoBKw`^Pm*
zNhL{5j}zKi-SV`jtIc(7-U_d$jR&p!pU(P@$M|puXDERua*jZq>vYy~ga!QfxofN#
z*H)9~=i(h@uLfye%=QMi(Ze-OyyOUjKEM5YzIb#^`!)n09VuFZWJ+8T8&wI_Matj7
zKywFR)m<V(_T$SM2TNc%Qce1vt=76Hj#37GC5A@Li<3EPz<XEp;vbp=NorX2whf#}
z%KXNQ)S7uu-~bf^(7~!r+*_6}x67=+40vVxmvk=SzNq>>){bw^ZN1G78n592*pb^a
zP&v#b7yVAaY=&@AC6!)~buVsx;w-1<lj>b*!|Bi0UH_)$?ttQB{3n{Sp$|D=SKY*|
zF<F((%hDxnkf1;jg+;|*aJWhVloRG(>QDP3G@-IE#Z!~g$ECJ*pSSDFV^0Ay*EAig
zjha}dXI-n!o-2<NkU6HA-*gL4!RhyH%8fk&LcbtHD+0Q(_UPW#yccRExb4hqq<z$e
zJJzY99YTWBFCTH)N`n6cFA&0}207&A+Q`sOjbUt{E5aqn7_mZbOZ0I8^)N=6sBLYn
z6_D_R@?2k|(@VfK=9s?V&;j+~dv#cvZTj|H9p##JdPP)bEo#hEm^U^_KiEN^!XL+h
z5e?TaY@*o@wA9njB@~n?cy#Fqn}g08Kk!cRr@GOy)45%JFz}N(j`bbyU&MQIdN<S)
zM^w~*sm$*XZZ$j0U9=tR-iD=WbaFKvQhn8}5tQK*WU6s=dC3)0Zu^qfod*#-h|h)I
zsc-O;bBE*cePtwQrEc6?%2T|yr~@W;Yq=*CL>Q$JGE>uB<X<Kynv2e$2=urIv`$jH
z+;TW(&=J&^kaV<_i5RMGUWSuG(2|RmK6|-`B@>t<YtXpr+vg;L%C%(*?Z9X-YJb6{
ze!?kbc`(bvyVzsS#o(9o#spQ9JP-~B+({o|t|e$$?f!+w-jZ%<mQUyP|6HW$zi^*Q
zPkaGrsSk9tNUQC6sb9?syRn=n^ZVKln^+@WJ5i*ST&r)WSdM{2%<JoAdv2x~1PF_b
zj;h%sNRwKJ2VVcb0Lncz!lhP~$ewnmQ0&KnE07;|h<Qb!a4gQGT1^~AH-_SDw8i4_
z<`<NZ3YG*N@+jd<xn&uvw&WzCtfG?hxIOUhadU8)86kc0Tq=iO^)fA>O|;;ZtFC%Q
zdtZB5d;jIDTA!dJ>6NtQ$tOWR+^jvn@sBTGee=}0=g(cXc)@wg8mqrK{%61AHr-1L
zU)iwX7447r-lM&L=&Gv@!Hwo=H)_v4^%NaNucD)$ep(ZrdAsSg`0BPL7Y=<RujDlt
zw<!G-#x2Au9JdoJe)c8Or8*JAu{3Xpx*p$&RXHlwc)D1(xr*J)oZ;hGSdYnmSmjuZ
zzNrNPn#$#riNPf(KrJ7TtrRqd>IKz^%7Vyx<`0KePpZYO`$<+~ei&(FEFc}EzJOR6
z4YSo+y&AnyivH^NEy>F_x3;{n<iY=#t!edM^P=mPZg?*~sC1n7HSXa0+pI;g<~C)>
z$Zy_$a?RSt<mmNR-WaVd^nS;SJR2qj!jJjgz_Tu}+#6{?^ou61tiOk;)~vlx3z{c$
zu7q}!mL}-(&uHU;W5Nf|(|Stiq=TB!`7^XnCXs#oQEnd*Ng`8qodF#ZEP73pz`QKP
z{S%nKBam!a6k-g8<w+8M$4B(lk2FHP{~YFz&}&{)lQ?-McV-NZrE7JZv%$>k9X!aw
zijm_?`{OrqW=o=<*;1{~mYTRyts{4Z^$X!%)29KrpgMQ@#AL*v3xgpb7UfkA(r1(s
z1&$b;ha8a)V_z$$ETj^dCGJUi3c`k&dD?j-hjd5=ZzyW=8?`QHG>aai>%RNLsS{Q0
z^Q{ZHxtli6I-@Dnxr@`buAg{$iFV|K^a<U@@vD~MctwQeyi*LocvTR+3R(A%bB}Yf
zKoRRAlE*!hCoFGoM_zo4+sMxb8`nboh*ZM-9<M$gW;XfkSv&-z6dbdp+(2+qCwKq|
z?zRHKd6SuA0?O;aWvL8gTrXfa-I982A04u9pZ1G=Yd5W0dCkUEtFPe??RxRWT|1tC
ze#g=qZ(Mr84L9gE0tnk9!f@!Di!3*yHk9QME4Bp7Cbq;=0t;mdPD^zT5GM24oR~U<
zE*f%)IgZsQ@Ua84bFJ}|(}8DBm=73`k@XcjL;~pry~RHub<XVp;5v4UiyfmS+V79y
z%w+l5T#Q>JHj@sR$p)YaZyNqeAQWwCZfb7OMM-%b4P3fs`<?mGvu3vb(@O=t{o(<a
zqx(;HIBj)8zutU5i!Rw$i_rCw@XNXhVyya{seE;%@Y<@KYsyDAPaW0x#9c~kVv{IV
z)XX^REZv_9b3yJ>pjA1!nV`??(XIacSh_&xxY8IV3NwxkgF;{&Co_&?MqpO-$B_9@
zs1>upFD?L~hDam6Vj1h%iRu9s#N@nuIc0VGaAFsDQsz8dN>EB=RYa&m&d6%kM~%S{
zNk-ysV?!9BiVUPaQ~0X;w%m2-YuD_%=Z4F|Uz@vaTk+y6&kGN!Ij>weK5^NyZE<PU
z?29+n@2;9S#r`eN;HsjMP(Jqeh|tV^#&WE)*n9owShMtQdgKtDt6^UEMoh1UzfWjk
z{x$vy;v@6*s!F{~P_lNUWR~(L1NZ8Ia%+Mr{wxjv$8Z3^%)xjVqV-``qeV}BC8~t;
z)9KBPm9ui+5DS(^<pdouZ9>c0bcFVje|GKpL$t8?&92{HFnP`eoUe1zqWHMkU4Lfn
zo6&uV5E2Q+8l;AbP<VLv@G`c#R#q{CR#V^qy+n^bg+bgv3uXW0O63_YlASIw;xyS0
z7VV?_&DwFTgP-y0dnYg9s!i`R>^wS~zQeslY^07D`D-@eOhEIzy+QLB*2{&Q{p=Mh
z&YrV!B^$H1x}N6lBC|lE4r2GF24<ie*Tgb$B8+00?7BBOq(0D`&~YKeC`0jj>reMS
zP<zQ;+A8<Qmv7j1<hH1I;W_WbZ?~O0w93MsF$H9z8RoeN9<G5tm#N6e0iZKYq0ees
zS~xbj^8{B~3i7@gY{C_wi)`?BGYnni*JT8>)MZEpmeHLN>khZ-(Q!}l?&t;M)Mr_q
zM29&&1-le0+Iv({b4PpXXh%wYw{z~4$+PB6natm`?>BEf_2h4UL(5*i`pOGdY(zcO
z4EEY{8GEf6(Icr$Ybn0_X)`U=jwU!q*I&614QL%r9MQ--oq~2yd!n0YPaXL{ICelg
z^qF>$pp7&|drREMXk&qS9RNwlit=%l%at&)l3~`SyndYEUK!D!y?E%iqgu<)Atn~E
zz{FNO@k|`j4N8t-CpbVIvOEV)M}AcDKoHveN!`vx3F-Ihpx7okMK~Q+Odl1QCM+C-
zI`fAcL(K`<7^aOMeX*7P0r<aap0MzUmQFvA8M%}1ipNfRfFyghSr0t+*!`dwd5~Ri
ziT6Q2dFbbHLIMaF%kEDzSXNmj*_2h131Q>0FSZ_G59Hpwbno^%uleEs&(!75gkRK7
z6sfK+qsNeP;%{_1c1>HgW6fDNejPMt?ZhTYGBl?^!UoTjlS)z#9%lzhRUlTdy@GRA
zqtm3XHYqiUblfnu<EbHWWP~Agz(DaKco?s;RQIQKnQ!`)bS+wd31GZfJUv+dR&6{r
zIKiy?YK!i|8md@LU=DqBP&}!F3s+$NJUXxqNyWqR=kq;T6)k5~w1=taGg8j{sr)*y
zw$AI1i3il+Cx6z5tw9!y?>@?DAfrmjU~)N01yLdAgF+6Yuo`467*xlUvcu_!jyL6x
zl$onlf*I^FL`X~dS+$$r4V0=uc(gJ!o~;~18;;0(9UfDDkiowZH9yD}=9NJ(OC5!V
zThiRZH#gMLXj1}(vcAX!jy#WuNnQic9)SQqsxGPbi?t(7kGp3rdbMrr+9fO2+E-k2
zNb5ejO1W^&g0@w3#p_d?HlNpSuWK@ugudQ-+t7lFNd3@=b|UgN-#&ZMqS<p6EMS#o
z4EIQS9b}-5IQt?W*?W+WTrnM1gxVQOSr+-oh81l5Dd<Nl>D)sv&dz~Al3v%gcJ8I0
z^?s(czh|2Jd1ha<B^&jIXdlF-i<qsIi@QYU8g^Iaxks41_w;?PyXU#y7^We>bKSC@
z1(NNnS1;E)wZB6@rKO+#_YcKK_?$kAZF(tR?|wf_Otg>!gJti@*7!JMjgRYD<2w~n
zs>O5Bc{R%?P4)ll7yay`=8yj0u;zC-m)vU~yD3r~sU30Ejz)N|9^UyN(p@m<exEV!
z#pI#^M;^;;oQZUv9%r=1Ssjt(FoWF!lLS*!3S4>-k;4Mho0=qCl2Cjyu4NT?+67LO
z*$h9Gb9l}ON5Y_bvJuH%!#jh8C%*|jYBeI1*`Bqt7-i>Uz|KE0!j)Nm+CbdIOM=;6
z*%LTzWHSp>N`@*ZVfKZwcqlW7&qCeAFybo++T!OiaT7}~3uRUEA`FQ~wJ{iyVp;a=
z;Vr=b0k*o5pjq@vQwIKw%VNX}^qFbj*}UccY4pjSsrkq`Q>HIoJbkKgbo7dq+izOE
zeDoK=>?D2unzPQjW;)@zKOsbVM<i^d+Vl(?g(P}|X*u0?951H@akcDOTrDuQ9EYKE
zS!gX539xW1E?^*C`ShAu+5@cx|9Edgc5)XB*L8iHY8uC_>RhPZdEH-5_CRg1cKWy?
z1NW@!X@+}NNml3Jp0SM~dce=3S*$3V8>yQv)CoNZ0qJcBxM${_d4b|!7%}Lv&<vxQ
zUkM`(5ejqc_h@rPU(B?Lzh3Jf$z6Pxwi>N^{|2p7JJJV0EgI)NO*F^cU;VWZJv)o7
zQwy>h*Ux8u8M2xC%aF~`0kP=I=%*l@p%TeKIqTet3Z!QMP<`}i&&2<U_MF-d&?-6Y
z=z7{vD4B+UOxM*(wk=1XEnbYVoxUx;FKdh6pp9zxCX;<z#netNg?<bw=dV~X-m7tX
z?Kp3=?LOAFAmX*1u63AL=t-mly4Z8ZGr&#*w^ku;#lhC^bSyn=e4&B;WI<2O<>}K2
z(q};?w>$SQymP^Vp2eA*PPO-1rU4nT0rAs-2XM`@ilhXVdCKX~qKa^s;j!UbkV#-k
z^PG;P%A=XW(A0plrqNbpsgpm@fz>#~WaD1%xn~Y4n>mZhW-&4oQ-WwpX1j+RvyHZD
z?`KVq93JWZ0lYlMYz`Y(mMp6yWnuRPci*g6Fe4D54J@2?j*1d$4`{40;26kh`;?aG
z&;YbGGr~s0`R;GS{r555hRC;za-M)3CnXU2R$nt?(O&(D?V`;Lwkw1Ah~es4yMtZu
zxY*_lsu{sPcqy4=IQFW`$2yzUU|1@e9m7=%P;1>qTj#;Z&6@`!r;pdk2N)^ZGsA}a
zBZ0nJ%kV%7$V!s3vKrnDHkKcxIRy#nU|0jx)S=`-lM7>tjg`Hk2KIv+Y-4V4K6pPi
zCf;or=ym8AgE;O*Sui@2%*^V>44`8bnEV+?M?dgm>7!X($M0(iI$_h3N8f&REfus*
z&U%t|W)L1Zt9Ek#lg~YS;-o(35r++II14R6Y^)598lPjsZNhMyVZ&XW1F*IN&A0Q>
zd^^3(bZY8&f7V;SZfsxstzWZ=vwAK*PdDC5P5+$j)=x9C@4cMn{>pxG0q3%|n{qQV
z-I#sD@aLp=hzy(^!pIRrpex8A(3K*ZT6uqHYGF*aA$o#Qv7B~z1;3A3Wo)r<04zG3
zarAU}bRQu`MmHH4ln!Lp)~-+nnJ%~e!JPK@M$jwo4?!<%An2tXLP==|%>xUj44pZa
z9)_JygQwTQMAg8kdfcF_&!qR^2=t+VM9A4jAA*@agspnsJtnESSdv<q(Fe@CFEqxW
zOy1-0`tk|{))kN9$3&;#z+K6DS*L{u3}zrxl!1qY-=NFOuG%bMV5UzG6Pbo-3}f7<
zKE9VjMspmp)3ztiv(}j3|8n<?P4LP8a__9_Gy}7<_k{1t&kT@YjL&@r&&9}P{cxR9
z7Xv+!uF|!eFP=f`2QF;)7Gmvppcg;3*H|!I6NQV>m|6o2ih%vm0BBu@5DYWva5|RE
z7%#ajNwP|?FEiMgGgp!!0R5g}*Yrb)|6(e#E0URf>vffh`w;nzeTaNSUk>7X^7V1e
z*FpmpYO+w6YrAQT*6-F1(Zp`;m)h&QX+0l!M0=Gc9?^cK{c<;rYyY{M*6RD)&F)?=
z?h<bz0no@%vP!SdP|Gva4dR?0#Hg9?Vzo38+-nF(R)um2=d>3iT6-k7M<^;YHfh7)
zGwuebdi-ohfOb16jN)g4N0lr+G-U~%9-XdH+_(({SAa5Pb)B-&*+|TLH!Zqz)AhF^
zJ@377$(@@wZKtiv>DJci5A1z*+&SOfvsx!){k?C@)&B5Pq-E0sZ_T3xFC19QJzV;B
z*DoUPJpS`bbi0FL>PfG2HQbPaQ1wvY^g-79CJOw2!|VMUUhmA|^~66Le(D+U96Ant
z(HzrEH~*XK{>^n~&UL~ihP{^IyCcAN4FmCA26C<s@1-&kg@1G3zq#)#ai1P_Lfb;;
zzx$B?xR!x9uy1N^A6|T@cR1+Z9QkjK{0bcTa_^8K$VG;?D!oq3#Ev^x_C{QFWD!@L
z2I6WvXR+FuQ_vT76`ku|(5tnX7;zSWmA_OM$UWHfc4)!WoVX$uy<qg!X6i`uK7ES`
z3={h^J1eN1BOb?13fx9yD-R=EK8Kv$*~On=@bqF>Pg7+(m9*YLB^uP;(mvc|-u(Hd
ztG9?G_1Bk=e#AXT_i@roOV4MxbKUQVf_MiRM!ro_N!+j4o(wv{1CTnD#+YhFgpJ`b
zCCLJBLmeHd<WLN?B{9@C)Xi!t9S(4_9C34Xx<-d&tBGM+FCN9+N(DxSF&sPszFIR>
z_N2;68<VWgvfZk$f$P}<dFV7G;h;3BG|FktQCK!iuihC&hcvecM&hN9)m~=v1hAyl
z*l*Bpz}9gMZ5wyYdE<Dbp&=4&nKrE@`r5OPe7Ac2(#@g9QKK6}!NG%r^ICqO9Yip@
z8S`#hG@|giM-sCes;g_NDr<&Mx^UF)7pHtOWlcCaI9ijKR#!W7Xw#UAmPJaZ#HCQK
zjqE<ipAb{Hw>ud`cIbG0YS`YvHu1^9MXfBz#YL^s<LJflID!!<Wz3gicnJE2i)tA|
zM#aqbh@6&eLEHu&u?JLt9xDe!3_m{>1*C`vx2(eSMyIjAGj1+-Dne>J&&1$S1p^!v
z9UHzMfYhJy2#~7%u~o-MHPN{*zcBPvjMUeDspF$Y9&Md=w>VVW+WC6}B$f20NktA#
zO8e5Jye~}<D;BGzi`{HP`W%WBXDJdwibV9>ZkH$J&x^7$m{Nl#<^L#6vVgHTlLw|r
z1~XP*(|Z7_7^tz_@7RRT&}1*rBur|^?FLPveQ7ed7fot2G^qfZRJhU8&d{U|Xi@<*
zNt+9bS)f9xlG`O)s9YyY+MicaTE;|G)F8~@f0Qs8%h3Df3^ODXd@DIE)9}w2DG)u3
zlcCWTFhr|poRHJ9Z%+NQ9#^M&T>T81*yOqByrI{vDE!|LZg0MBW&Xc5+TL;car)I`
zHw_<#(YWEms)u*JYK)d(jMZ9$pPR@}&Wxim_pCgT9-@cmO5&_;Hh5SKxpcr!#gbsG
z9?E5P#pPBKs5<Ig7~fM9ph2*1HWbWAggQAT*-#27iHw9C90m1~VRxCaJHwdi2R-^!
z=E|6%SzhW%%ov!fYFS|KA)T6kFxI>VpsHZi^EUpZlasH_7$gQ=B{$GDLWYneIbhJW
zp)XyVbLm<UOIPVfv6xX;MU`PlTcgrgiMrdZ(RyblH4Q0^rIb-O=I<GlZm)4QnOVgp
zrX&mp8aDwMAD3(epnq81p;|*^Jy74=a2f)$nnX2NocwZ9m*x5M5TL0ifO-ac>4xu6
z>3;5kE!S5pnlk;IbEZ#?pO%e<k-xa<#?>pg)A%*B&l=S-{fo7|_0>^H;*XFjk{~0H
zlgn9DS?tg^POi;>30Wi<13VaJ??^WkM7=<MV;4p-t~9vQ^#<KX8rzBL&>n8JS>NbP
zZGrc}Y<E=PRi#-TZMC?(Vz4|?5NklGo##pAmzFVQRuIIE!b=CUTF(Yqi6>P{JrLZM
z*NlbVE*KV7jxh2zv~am4Y{ovjxsb$_J)5gG<8F+#)QD7%9C7`7-&wd2NZj_vYu;Xc
z{f`d3?(2Hca{aoyzf*Gek=x!H`{2(|-;Ha!V0J!#=VmTWCvsCJPg~Y8aov(;UY;>w
z>HBqKmaq7&ngudmx#T6(eQjq=w~zZZC%71-G2(W2J<V{tD_GT07Op#zn++4B)#-el
z^{}|lhLkD_v8eGLGw&}fW2!A*FJ>SrZ<Ytb@|H|e?+lnadMdK_MNj3mT#QNW$FJDn
zyYi)XdSiT#wfdWK^Ik44&@#O8F8?mrHa?P1$`F^SH{x^&a7KZSIGxG14P_Ch3mD>b
zp@BG!TRU0x$)ps@dw7Q@FOZKNlsp4KPZuzdoRZT(Pcy6TH1N}7F~?!~KoHcz&$X>x
zodyhbuf7&TU#~a=bnVap=-TB5x;FMQSi+VDBrIiimofw8A4S+^cG_uBwsi+=K&);0
zSSBAM)7QPMuNCClQ~O#DeGTe;4H|uoWcwOqeJwHiT4{`{>PP745_vBh*@7Z{XtC4z
z{d79WI$eI6PG^m<(~R$ImTn5xmu;Xi=BFDApV@I1jJ~fm`d&xw`O@B_U0RgD{AX0-
zfjHZ8w}%>ehM$LIL2^{XVYOS4@N$WR?Jka#!cBM!^JFYk2S=ve;o;c?uV(luEM16E
zMrM@p3{07<zBVY6qptg-5mVH+=mh$tZ8FA@Cig?AdzezzIGt6#=!qG0uLFUWW_4aY
zc}5b=$un{coM#k~cc4o(mk!5WHNH5=KUUP)OgORuZk%&ZeXly2lV{}UKhG$_@{Bgn
z;o9$CMrLRC%hr@}-FLyueQeK>o;)Kf&B-%zd`X^BL|&6I>CJR_W*3r7!*T9(V?g5v
zz(2injqoW0<{CNv>0F~oPOi~LHl+8^;j=O&{NG^1^~GyCIlht25hJbLM}@iK4iW-B
zf0v|6k<TNh4y6WIk^xdMR(2n5x4H?2J`AM5A{S#<rKwP<ETpQ-U3ZYEZdctw#Gi2A
zbD&-ansB?7y!~jd7VK}gy9#ilH1_JmSNODDcISDkQ}CLDBmo~9yQDdiWRFwME3i8Z
zfQVja3=5*%SZ0la4{WjdO7(r0)j&R~i<$^(Ou`^yRI{<T+QQ@FYGpJoxo8yLBY6mZ
zLS!E&^?;s6ng7%ppL5Z=<n>F1Q8&{~qefNTb!9nSG-lYz>icNHwVyrmw)Q7F^0LP+
zEG}H!eD+23;5py9XnMswZtaec?A{$cuW(#_<CyP9wHLJK9@4t*rq<tGe$5S+pS_?c
zY~)4inDJYPhn8cGR38Xfz89!fk5jj?H#Y1ip0&FF_P-&WqWSCK!2N5pDf(dL3e$3(
zwycdSfXE>ZB7<)+!$8mVXV7yCF}7EnMe+5x*&IYchIo-4qHjPSkkf+8ChKTBh5i9>
zbdAtbe^|Qf(f;^WKX|&ri(q;@oCVd*ZEql_y}i+Xi~B?S4H{^_X~CASD_xVU6rLIG
zFS955!~Pwm8xs8?0MmD}IZ=9>_d=Wfq!e?aSepwnZ7#t!>rl_&3}9Ybk<sQ*CNB+W
zTqN%iTyBq_l?R#h3f_`}0h*q%2?ikd4H?ND?T6(26vGneA$ztDTZn@OX_TSp=8o(@
z6V%)P^Edr_ruLet>Hdb{+}z39wJSRqf-x3r4SuR3U+;&bFd@g6NQZR|>y8^+ZZQsX
zp}pmG<+7S$GEkH-;1Y_dl`efFE-#kO=9lO@px6*vtI`8LX!3N-mlGy22EQlqsTUhu
z!PN7GM&}DNOLx7f@GU*r=~1KmTus*A;z@P)Yy1&GfFqO!4FpV=v`O!<j)DSKMOrqH
zfAY};Q=D<Oskt|fRj~#R1~qEs$}8R3b~40=uC0siT(tQ1#pg_(x_B|zg6lr|Nb77}
zu%LC~f(6{)b}m@3vyHW5QulgcywFJUNC~MV>qyFlqY=XJKnNE?TpmlcJ}|}hY^8^R
z<Bo6zSecJy9swi$u|nP>39_rOG$&Omgad*5(kR84DpgQ<N~#p|@S79NyCr(26yo6Z
z&Xk&TW5*pAEu3-ZB?~Xme)HR<7tH$RWs5JMwN+H8ZCbi!Q&Yv#3nTQcdpoWQZ+>?4
z8{6L3J_>FB+1S@_TRDgB58kJp@4s*B;`zGHW_Z0Ihdt2iIRhZ}1_nUv%{S~-)r~bn
zb~)w4(A<}z_oDtd6~Q-*ilZ-#uNcbIT;#TM{y*4G^v0l-z|TpBEyu!Kr*A77-+j=A
z&pSltfVSw<T7Gv=jfmdfz0lr5M1||Oy+s4Gw>aYmxV=~-3T<xp`SS~qQ*0hJFYx6g
z&ghL(n?56(>Cs=~Gc_c7+t;$TSAgfzU+WRjqacwk(Wj|Dp7EElH$D_Vm_pdATZ+xC
zgz%GM7Z#qNgksKPzE2;SVp-os;Itj<n?Z1D7h8IRef87Pzn)rG!Z3`1x4a;I`av<y
zk1-*0W+=wUFaseO(|TrR^aE%t{w1?+e5p2z?LnB?Yd}8zUIUxA?)~E_dkxHzPDZ;r
zHW`}`mTEJ&-T&Tf;45x6z#BZV5qJV4^(#*03Z){Z6eiN8I{RbAN@Yy(AI2(CrGP~`
z)p8>lz5wj#cERc~<NOWE7z`<Ag>YxaP<=UmAoeQiojTJGn@Q99<VVaV#I!^t#7zpw
zG9A5AwOL^f$Mu><$%+LzX_M8-I~c|`tIff^zo4hGUI2#ST`8WjF^CJBogpy6fyGj-
zm*tqOXkqAOIjkNhTZ36fs4s6gl|i@7rkBwL!kDi6=?TruE&F1xrfNHA>(3AF_|D@m
zbL(j>U6ifLJe$_N@Dz9A86&p8i9Z4wdk~3XY#+gNmEwfzF^nk1oKT~pVNliDJJMCo
zfZ(V$R#P(R<A}^%%CvaeWuW-Q!Ov$4(<qWLX}qEofzpe0EN%%gxd(evW)B0Uv&kuk
z&(D;f2Xo8)h59<J6U#f5Do=`(B9<e#^)UXbP%CG`U8jra=Xfq+1Yr`G^?mT`+YbKx
zzx`bgTQ*;YF|Ri-|6%Jl{xip!;l4$u+&1HUW`Awk^xuy!on!>PMvdOw<JUwu@7^nP
z`%BM}xsRBE3kuQh$<2~MC_h&s?bA8k7H4Wp&zfm>N+xr)YQ;cc3=tUHJ?uijrRy*p
z_Y}3`3X<y4!A#YFgnKah3K*&S=Gbt3xV=}#&8NMEd#}`+S5D#W|FQjeiBlNKhr}(^
zeJU<D*W)2M|JS%eSfbCCo&UAIP$o!m)s99zN|B32NO~a+{xhE?=oNy%Er-tn8O%g|
zL+V}-c?pSsQKDCT{Bh=QOfQ|m-<WRf`I~C)KF%$UB2L6v0{?_FaF$#;c$2;>C8lgi
z*)nHuLnOe!JakCTYPxXy@>MfeTQ)9Oy`6vHPmf<Ueno8Xgz7mL+zs~+aw(1z+w}W8
zV~ACrxj%n6ZST3iT}J@I`!mcG9oEwC6N4*OoVC(&b=&G4+HStK_58ZQ6XD+Xta<+_
zY=2PO2V=DKtK2^@@cj?<zCDTx?mnTB(I-fzA%$@LKqrzSX9|lAIw=y{_X=SNDhvn$
zgK8jIQyhXlu~nj?eyA8Z4N<uWo~=G>RX-%Y&^Q@igY@Q*-a?eoo3+;&Q-TSDC`x`j
z1Ds&bY){Hisfv2qYfwq?$SIdUi0v)%DEa-2N@QRn(t9pgi6;wJZ$EX!HcmJ{#!l|O
z(4a4jpAco^_dnd;z(3aBL#Jyl@LMyP#ZTjxgKnuq6yLPSnrghUbVqtPZDru2#v4m?
zMEG9Z*NN@0%Iu%YXl3)fmRGB54z<jkQ&#$FP4%n6Qu>p&S6b%GX?dll_E1^r+_OtU
zsK=IpEijT!Ay)7a%g9^<qrqH~NTi(tgY)ML;&FBr!Pn&Mil(s%0P>{`=q9tRMA5nl
zz*#Z-43{_wF2);hSICJy5M+^T=waDKGiR(>bKhF7;i6e^Fk|LguJNM#;L1gJ&04;E
z*31<vXPokG$T3pDJ-{%@LK!q}Y!wbK&$gb|r-;B<%z&0naLrBHV{~>CU2+PJY=BGi
z59i(sadq|E_1uf%y)y7;ljv;iu|}@B>)<<`LgztkUpLX7#CCx%I7O@`Rb(;I(<MR@
zOD`z_WoqEO#uZac2_({H{oEf%BiWA2|AX0jS)s8Te4&7Kc2!LChh#)IRy}3|*52EU
zfcDMdXrieii)wEQnL?7^7YGGH8EE4&K4HC65bm`8xQ4n8t~fBNWy4L2))msb&Dugi
zqAShXW5a&3{fZ&Oo?WX=o=WfE6)!FywyWbeH5DIBxM0n$x{_FI*GsL(B31MmHm{~|
zf(h&hpjCi8sw1wd76)*Q#5h$oyMSYKv>0|F7l##yn!)Gw8DMVG;S9W%C}s9<N=Xc9
zOTsi-(T)2K;r7N>w!N_d!=hMX7({aG<|igdAlRE5EXkCW?N_buRMc*>%YHpiwh`^;
zFawB4U_iwi=JDSYu0#9NYuqRoU9`Mt6P+!+WX^y1-JA0yVMZ#|wU?U$znxvjId_);
zWBq%stm`G6|IiS8f`0;LQ8bgSOtTvy8YscN!Gi0-T*E)<oP=|a0~_)#@iy=xFeeo3
z@O-qxBZ*WL_lEZ7<wqTCiv^`TmJYK0dzGNEV$60Jtimm>FiR<*?~xN=8(jtbK)(C{
z+wvO87tm`G2Z8OG7b!%rG&WV4nU1?B!u(JQ<&@;Y4HMNt5ujhNxSY|?n}^1zjtUG`
zIGCcra!fC?Cv*13cJht-#@GqAF*fZ%-(m}rxKH*IdoEhX+wxsgnnxr$r9WveavRTo
zZ29?Z^Up7)BaAJxqjJ;b(`adC%WN)q)`A6R&2DS^VvD}5Fi;EW^$^<wyoB6f>;hht
z$ZP~2jJ4Z?k`q_}X;XS8dw^rTCI*2^V6v6<YzM9s0XfWiwn^WO!Bx)2(r=XF+zrB=
zw6U_+2H`Z*dA3P-$^2N(CgJ<Dy^hf>n`xPl-6_0lMrNx_oj**Apua=N2r`=dhNNOB
zO9qcjBvPZ0LwvQdQI2Y5P=fWpGOi4Zr5kiUkudiPDS^wc<;XDwNR=@$CB~>yr&A?X
zhop!hDeBxPZ$J>S1g%IsHMlwpQ9anLjs)_IMiv@b4=mKIv(Q+~W2gfj(W!;tcHqC&
zdr}4ExM7;wJQBF2VHncUSIbzb1-~|mu~Ne@Ox-FfO(qATV{?D}71DVu(TB-;(fONr
zzJ@W~#GxbNou&@$1#P|dqBsV*4tJU^qQh|vT6)G{!o=o<r}AO?G`wh90b>&bE%Z~8
zs>6|~8j?sTH8CZs4@d-oy%i~u8HP}H7#d<q9UF#PeHiK(8h9NJLoM^4aTsuqriyr~
zYB1ZQDQaOL`IsJ(0B&u>4Y2BTZXKNAR`8jT{zK$c0j7xI;`jfV9^{;SMC#ql#0_3H
zztKn$v>4HlK;VOgkW#fG&Xf%d@m*VcD&*x7ZEcJgIe_UL65y>k!Wqm{MGj^}+6Mq#
z&}E(g0RR910{}w5Dw1no4?Oh%9|!;d0002d``dH?0002j(Tn~6bN##r#|Ga30RRX9
z0ssI20001Z+GAj3VBnDb+sDAb{{4UP{|xry3_uYS@CpE&RtG|O+HI3fXj4HHg->Sg
zdkHQ=3B`h>1Wcoq5JCwhgc5=@XqrW7S}8?>)*@NBP-5|?LKh+rkuJKkD!9s`BBi)9
zh=>ajL?|LikRT<v$VN~KS_8uK%rlix8u&Q#b8lui=Z*S?Jor;j0pCaJWn9y@5ReWc
z0z~a~#H?|g@NFTWD_GPOcSj#Xh<kCB!Jz6OLoKTb_Yv1d$O_4jF%lzrk|f#gJmJ-|
zm*X;-B=dR>XYD4=Swj%{2}jzPvaV7cG6Tn2;Cu~^n#EoH1|bVvABJOZp<+Gb{ufNi
z>#lz1-W{I*0SRj!%T@@7{52%4VGPJu^h*snUBZf*b>Hh51a*?@bM6nFV9#OR^B$^t
z8#!4eL&&OKD9N~gc$caQ<BU4uzLS&QxN5&e)vU`+_N`Lcqet1li4iU6lk2!7VfVKk
z$AIp*zogIoq&E=LDFhU7gBs!aLEpIN{c_h6(JkYp``Oe;8gbUx_C<{Gj>n~eC~FGF
zdv7i2`MT+4QbRw9vaZ2-O6r(YNt9XB<a$v*Mp{}(TXhskibUys-uQkH{)D?@vhZ@~
zf!-~dET~V&1M3AMdpYdK+>1KxWzmC|#Y^@K(_iEJ0oW7B(u<rN<-7+Kb{FaY67@cD
z<RMP|&tX42pIY#;=!IE$nNj^4n6_WBmUlAW1N{r?D}cQX?5Qg9jXWo@f1pV{PrblB
z^A5VmC{q4iq*R=F#Q8>ulz4{{^)v?cZKUlrESO%^W?dTv^S@GIUy~WtIc`DmyH);m
z(t!F4pfkDA0001Z+GAi~(1F4dhFOetOp}-`m@hDYU{PWTVX0%;$MTMqht-C)jP)4n
zGqx-29_&5rPdI!y3OGJ+MsUvHe8$DZrNZUFwTHWcdliomPZZA#o-4duyk)%Wc+c=L
z@R{+&@SWgi;t%7W!~aj<hG3WA9-()_O2Re58-yQ-@QGN76p73eIV5sN<ezArn1fiG
zxQF;A2_Z=qNf*fu88?|KnQ1b=WbI_T<ox7?<h|s#DJUqUDfB2jQ<P8)QEXBYQ8G~4
zr<?+W>s0borBr9Beo>QAcTlg=kkOc>u}|ZLCY$B~%?Da~S`FG_+I`yVwEyX>(z&6_
zrK_e}q&q{8M{kpUh5i=<J%cQRRfcJXmyB$T`ixnOeT*j=KQQq#sW4e$a>Z23w8zZJ
zY>_#Od7g!x#UYD#mT6WCtnb*$*k#x)v3qCV;UMHN$5GFT#c6|cmWzx_gUdJ99M@NF
zO>S%4Zn$%~CwRzt-0?K=?DPEKwawegyUzQEkB`q1Unbuw-vfRven0#p{C@?c1#|`K
z;ea-QUV#yTMS(qmdjj7C$pn=Ioe5S6jtkxcgpWcvfUqLuPpC}jmN1>LU*W3&RE3rq
z000000RR91>IFFf1poj5000620RRF3761SN00D&n0001Z+Ra?ca$`psZcheCh9VUx
zDn(L2onkTAlt!6blF4kqwls;!_)?ZVQ&}EMN1nsTk~l{i$9WW1tf*p>6~!}fc>*>p
zc>)$J>Hhxz@7rm~vNKgsF3acA-T(dCEi0woU3*)-q1JA`rPSJwl;Z0(mFYXWzM<Y+
zd&JjosM*@D`1+>$#oC|v`djM#wSV&UxApV?@b!1pmp6XK*KexMjo<P0Tk8EAGXB=u
zS2tw*ch$q26<>evmtWre17E+b9(I1p*YBvTbI8}<S8sM?{14Ra&R2Z>7xixE_k8_V
zb+_|pzW$qfyYnx;{=0go^KZKT#~bSX?oG9=CTglKRHaIFs$Qs>%G4kA?|pSw|2$N8
zaCJ|8s9O5zfgVw+k@~r+_1zOacdTdBs;%#K^mQMjZtLG)>vy#tJy*HD>*&u)4OF4W
zWNK5dFw%b=%o=xv{#)q3uk_~tYaZzNCwgp&wan8@-(Tp@PqAXI=ND>d@yOJ%>gn|}
z^#EAkQIFKe>J$C<6HSNt$<#T1+InVcg|?<)qRupJ1|Pw%r^k);+o$^ZfqpYlTl#mT
z@$u>Osk|NV%5kd2d}^!5;NWs7nT7M%lga9{o`IjW9y<Z&A85FD!I{g}xMSA{M`n;k
z2C2@#KSS>sD1WJ+O*AhT>q*P%N&8YgS*Ag^SA?;6jg;Q6>p2@q-50I99OmoP{v$xE
z1D^Nwnzf!|bTqSiYqYx5e_!aU0rWZ7<3@O6tY?hhQ_L_>=M0bD_JLW=_BDeBJ<%(e
z=O_AZ0?8T;82}fvY7NUYnl;eBIjAYv6Kv{tr;KN&`F9um5ndXKXPWnq)d%`#HEU$P
zXYWOP{~+)KPb@dOjO#J;G}D^x(`j%#1rLnw7Cg(qt&+6gifDQmP;?9L`+DS9k3ZKu
zHJm*4);3%*8nUY?*;j{JvYTp8e`fk~U;o|F`q+hyAE?jt-BV3z518y@To06P>Y1DR
zZ`XUCWk1%m(fmDquUF500$67LbF&uS@9B9)qxQiIH@=IpM**}9>pam{&jGV*XBw=(
zfPXQ5)~wpqe|zBg9_Tp2dqZP}_Z|KDO#gQDj7<$W#<#1wz}G;2%5(Ps?=!&N#(f7X
z?|APvfoHT5Fx<wI1ExvVIntkpn*TkfqLW;^XP+7g5q%w|gw`Uy#vC+KIRpl~%n6dX
zo3X=?-TH^X^$6VTg4bQlbvUP}Bg|7H4A0vEOoFk8h`8<XucQBVF@m&DaO$xHkHIa%
zwg;(oA(tl}<SXmxGVRKK$Xe+0DdsV0;IX|UJeKKfPmkWxzPQVLrWO8|8eu%e7PS08
z&oELE%iP3zMur{WMU)+SxcQnbvtFwdbo}c=*4xk=GtSVoY|aM!AmK@L8QppgTi(}W
zI-ckGjAar-b-X7Yejd-!9^(t>D@9q4x9}WW(AOi@?RCbc2XDIrE+1O|Lh=iFY2a_;
z=`Lvh3{Q+ycaaUy*nRkJ@_#;lDe1QsdZy1|anHf!4rJ2{{2AGRxiilVjhifc<Hche
zTwR0RJh87>J6o`=J=VS+^QGhY&|Qpi+9)gZtYvKht1W$f%%eyrWPV#if0E=)(<8a{
zL%Z!=iSo20cHu>iU_XZ}*&~iV+u%7t5nmH>JzTG2S1}Q8o9%oVOrw{(n7IR~cNubr
zt?$TM1^Efe7sPjmxGb~g9roy?JA^5%5y#*6td}#KIA-py>!%x<8soDT`#Ixd4_Ifq
zprPmK>(%{Jm!(O(KU~SP_87~nLo?R(43u>f)H&TV65e4>?}I}wmLG7|vRFq_IA6tj
z<*kXu&uqr#BE%QK>5x6fHb;@3%?4h#ugI*9#kmaUcoxEz%>l_{Qq<go-XCKP-=~P}
z?*iAZrDq3P>~L(eCU38t3B>Xc54&Y^t^?q+&2Nk*x0Xh&9qX<3@Pz0wlx@M<T9;&N
zgmE{y))V%hvKDFg9pGW=>zS@Ga#SxFDh#UKGt_?tYD7~rO?9<0<LYOMn*J1((ix91
zbxDr8u;99G%xh4!IYkxpl4~myS<msVk1Eqt{1-f{uPdH8>dp++;3;av9CfZ4Q((*l
zBL?8i8P<Nm(5d=0orE0So&sZPJ;JOZ&rEN@P&Bf$T=i>a&QKrcz&(U0wAGrXLO(CD
zuGD*Hm`j<=3|uOhlT}bx=UlhTx;4)zK#8faf2;A6=XFC;GgOl!e(%GkC$y&F-k6_|
zluRW$wWH1$JA*?VROJKkux1_@_(bW9dEh8nT~B0np?!c==eDjjt4nXmP&xA82@eZU
z-qsd2Vhly%LxxHnhZ49LuFL>;#&YvH<-<DVu8MRv4zSu8s1zKi{`dLRi*&&C4Ezp(
ztZCpFsg}4Bs%xuJJ**{ondj6e{Rp2Q<MWXR{S#x$*}A&t^>+*ncQp1nD8JyYX9caw
zJzO@SnSM9HblNUp6`9qtobjPKx!0?6u%;fYX%bd$VY#kIjxN}oM-|YqjNh5hQ&3r|
zuRKY$JPa%7vGloUZQ%=9NoD0e13oiPSFZ}$;q7~^<=lF~{OYqb2}{2#=DcIGf7C`K
zruB#5s`Dbb9qrTh|0Ep);T7;3gVL}&Nz~8Tc10%6cA9bNoT)T^sKCk<;GMwR_>%BB
zGr<Uww|PJII{pGR)wv^UEE!vRTC(rCu$(Rfi#ccS<W?>i-kiIJWUKcf*-t&~q#kz%
zTz(0xr;WV|kyY4t6w4iZT1K9`3UrqsOc7%pmGo4w9b+u{QR0+~l%uHIIRw92jd&f#
z{emL^%_aC{LINR8g{Osn_D~^4&0%4!1+!_^spV}QSXvrZZfU=x@4sYf3?E-Hjq-NJ
z>((4;9Dn?-e^a*?QQ&(LC%u~SCg~J_&<E@ZEUq79D(Sp{@+Om?j5x0n&qz7Xl&$6f
z`wC{irVkoSUhp+%sui-2BR$(hxUa4C@8f<PyqWR!kUQPdw;uzx@M7Y5F!EaCgx;TH
zM`YB9mKid-0j!ernj-F~5og1_j%B=(tdo31i8UnCESO(oP%yQe9kOPa4E95GbheT0
z*Ph<4V#{qCbDKIzn9n^Qy-1JzJnRh6i-5a}v287Z9CF^ptBIyUvV5|fUEUWewSBnV
zk?Z)f(CSIvQ>K=@2<3vt9!<u0!m!9rgsKlL7Nmpg$f=6gifIj9@-sEkpb>U!Hf(3s
zc>Cv|Wt`;ogm>{OteDz)k{z`Zys4n8{_eElR0W$pg$69z*Yfu;Th@}b-tb7tqDb#6
zTS1UFs(`^9JgU4rU{RJaN@Tt+$5^g;&Zm;bg&E81mE*ep+yG2JGptdo5C$c#OucL*
zdLR0fvxH;q?dsSuvs9Ber}(@ot>oVp`MKAtzp<@3s-3?KJxv{Cq~bZp@aipOd=unA
zE;k$C$(Z9Y)$^1sRf(KJYSUB>d6t+TYle9JBFfmBkX(czzkjONB&^55nUHkOuq<QD
zHTQ<>8DhN>ZHn_Qsc~s1xM*k*b!$lXPFVMPc4xn`RVjI$qZM-LP<G3sUuEOBi-tL8
z$78PZMMG({P~e`g>#=QM*ox*A%%?tQ4FqR{{i~zU<f9{7g-G|{IUv4dto^Fl)kJ;9
z)`vl-4Jhvl;RuJsn~ALQh89paU<@m5Gry}K?={*vMSkUEJ^;niR~QG;y|5NHWo%M$
zZZXZN`ZP~yKlz?9^w)LJ1cM=ZN)*WiJ0sB?-IS5HLXE6wiuH*fLmO8{oI}r|Iwa&u
z%^GF8dXH^3<oY~L9`|s9W*POAKabh=;u^b%B;5+~CaO?V*%O`BSq!!CG_-C=bD0fF
zFK&)L1$hy}2fwCWtu79IS!}EKLBRmE>X7+LYb~aVEaw%U!kJ*r3{oqjewFyN)&gBS
zvnO0m*a}XfE^$}`QQvlEdu+?B&gt9zrBafrTh3#(SB|A|)TC&B%d0J7dUI<#b;F}J
z>v1RF@I+kCcHp04??g`F`I(>ZiOD;b+^ZxmrF=)+pRGMBAbBo7rBz>gPOHhom_yB-
zS;w<f#wi?=42der2go{|pDtNr{BDF;<16mNNbV}uH9#k+z9I&T=wDp>&At4_Sh!yq
z`xWur`oy^Iy-JT+YLBKQ%ZOIrN0K)pUpBjzX6Dd7+Mij6)M-wOb-<nWBzq|NY_H(w
zUpGgc0oxKf9p5)c*T~FDtT*EL7C%d4BBPg$8u30>co&Y`uzAjeJA)^Y{l@usEH62!
z7+4@zYovpu!DsAEYYS(}wN`C)!TGDM_I6KUr?F>V)W>r>HEFx38{W)6ZAm8?N1Pox
zU8g+f71m6x6JuNPnc&i9o*BALl4~1C4|pH<9I>V1`e$lS&Moe=`@TSOV(G`tAlEr}
z9&&~pcL}L}8YQy(3HJs|n+?bPo8|pUzp6>~E>ovO>tLr%s15VttDS%quQc<tO*BLD
z`H}5Z&I3xP>{TwJU4Rzn4>RVQ>rloLbXmA~&0)};zgmYh><+Gqfvd{~URSvf`bwM2
zrtBcqcU;_Q6dL1L?bp-Bf9KXt!^jx+cL&H&No!o+$CsoJ+qY2hu=xtzdbjU%!Wube
zySsH|yF2yvMN*K)4Lx$5qMRi(^Y**LGx?%*t{*{tJmN_6?U=hF@f-PmLT%46_<Q`c
zn@s+)upQKitz=X0H*A!uti&=8PU{klifQzB|6_j14qr0|iB}o2H2N&X!ka&8^pr}w
zmo6tGn!8Nbwbsim%Js%)K^1sn?J53#Mk-sBe0ahZ<nroppOaRVTBP8!05h+8Nqr;a
zseXUn=z`=?r*CHKpo;~uvK4rEslxlgj$Q6mA&&pQLB(m-8em3Tlgw=|>3R{uRoP*T
zkP*Hf(k^4+<w13TJJ(yfdOKROQyA}zUpFGUKEVuoUlhqI&vT~F1<1NJcP9I`z8@s=
zfn_=@e_w%b^PoSx$YVM9l(7tX|CjooxsA}NQ!l}Ht5%kC8?{?=PhYTi9^8*lsojs%
zZeX71x28`ETzzMB5?F%t;cAZVa+aj#DyHV{<dDx0y-1vCOP|`f#k)S0*B^2xQ)oZP
zy$L<Fj0>|wkyxLjMH%rt-0da4uX|n2-21-Q1o^`hK5oo$XKs5SaSpy*Uqefpu)nMW
zdvYtdXzFXNqv$$5Cl5|4roKVw+ehq=U8ZL$h=-7#_-%7qw6HZ#Kj<ih6-AuUm$oAr
z!YkPS`1a~&)c#&4WoVT>6PdDKS!tPfxW0`aLtb-^h86r?5pYssW~?)5exz)Ss5*hq
zBCpk6vR^h$oXC?r+;tAaEc7aQc`Jyv1Ml3SS^2Tu{gbsWJPWLOchu$6BYq#omXdKi
z_qyDdaW4K&r62uiPspck?wu=5WjrZODjVIi_>5=4Cu~FRIw)zmh36|?z-JFY5uLY`
z6TOl%${84md&4MKCHyPayrOZI>56);$6ddE^>YWY_&n&pSKi6txrOd>(uiM-^$;Fm
z#JPv7?@m}J<9f`GW3eTs;*-gzKI9B<@d?+}d|dO}zE1O(_ceSSPVYy?r>!ch`?c4p
zb7z3*zVc34%KI_*A#2F1@@%NKkyoBI`&_;>YrDIi`i`?`sXG^u1?M7idq(vl*bk0t
zj_~Z9?CMokPUFUz(r@DDO^<G|<%&$sAbqoobYV~2zvMiuf^Jld6-mmEmo2t~7hFpg
zUNt>NEFjO_;<FDq&Y`uqe(?Jrh3$)!_FO+nEa%<hGw;OzfMel>UC(7?wLK?r%6$-l
z(q}wRZU4^4!_Plye`VlRk0+d&wU*{Tt%cc0wIGK{`)xU!xTu<6lut<gE<RYKT=EY2
zW#8wxJc^F3y%@f}v=$=Y)vu#(>reUrAk^RBKWD7+Un~YXGdDeQhy6YL2aD&nf-ex&
z?yHY@_xG{>e560=e|^w5^dCk46QO3kpJGR%4)$wh9|zxN_%Bw~^)CQ;+HKHBOp|dK
z$MNq2rG>Kh2KOHC`?h5$juxnU?=1?nRRk<n#cjkrqA}5v#&uESp3$fWHSQHh)PrYt
zJg7%w;r)B_xqR~c4$t%CA%eF0Q${!bSD!&5B$7xhL}A57G%?tTC60I!NF<45Qb;9@
zbTY^!i)?bpC69axD5M?j=|D$1(U~rE4e#kr4|>vz-t?g_8af6JoVX~Wm=a3qM}Nu~
zz(58u7&jiush|=sLm0|1hBJbZjAArn7|S@uGl7XrVlq>h$~2}kgPF`?HglLul~_1R
z3mdt?K0dOEU2Nw7$2lfZY~wu}*eh1Cv6KC5;Wi&6nuDC+7eD#U5l-=ddz@w-^Vv-`
zkEr234|&F8p74}U)bf(&yx<Hzez1o(yy7)}zVMB$EFi!_7Ewn%hiG6ijRa{T#1fXW
zEd1r=te}~dtY#HgIm}wtu#WY7<vZ6n%Q>#|mUj{(c8QfZiI)UPlq5-(6iJmdNtX=H
zbAiiT;SLwM#9cOXQZl*8Ey<E>c1VuoN}l9Pfz?;n-00U$3|pu^plNPRSz+o?xyPX_
zZPm~;O{pslr9<gdx|BuAVr7Z)-#oXeuIB2xy*A(r1#7CSn*3&DcuieiOJyk7(AK!T
z<*j-><{e=Tr9<ib3#?R}`v3rV+BMCw3c@fH1<<5v)3mkKp_>SC^?|Ogg6+~poD%d8
z+#S`$QT>${C-LX%L!a9{mkakL`zy+u=+JCjkn-+BtHDii<U6dl6fS;(8um`0vJEO&
zV{DtJIz{kBLWpM!O-LS$l%RJELr7PQj9`|GoRBRT1tFg^283e#gT#{GE8Dz9dG197
eLA6I}hP0~=e|oIp1CgaX_5c6?0b8rvegFUhJnV=7

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-bold.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-bold.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..dc05cd82bc4d70251dc6af16d2a45ee3d6ba5b6f
GIT binary patch
literal 25672
zcmV(>K-j-`Pew8T0RR910Axr26951J0X%pB0Atqx0{}t*00000000000000000000
z0000#Mn+Uk92!&`=wKX#Oa@>8gHi}Y34~=42nvGURD*;m0X7081BWOJgH`|pAO(Sp
zEC+>L41rcZcV%K`23svmae(c52DgL#7QWsP1)P&>xm!9n5f!LtL=77UfMA!S+5i6u
z333d{0N)F$w%u<lKy;6xp+T@nre*c&W;kv)_~0Q6SgmtccY(_wlkDs5eD1qC_i)y!
zKmVMUABPUzokGAz%s`;s8u1nCWq)g9ihzLRm#PC9q6U8=G|8ixnh5c0ExPn1eEig2
zLT6MufAf{z@rlI$q-Ard6~-;Rkdk3;_5J5@(?9p+5u_ArYLr5%D^Rse8Fgc>&cyti
z3PH8Tzdf^a_lbh2QV|a5UVtS4I06-yO8Kw`LPw|o-FE?;0h(L4jgcZC6(WpDM6Fmd
zSim+m=xuC-L^&q3L`j!XAXVC=@roFQVkhD=rK_@~CQ^%N1Oq3{@X`ZP5*d)A*yZ^D
zZD#>+kl4Ggs=a3~6DB)(PgYu)u9l?F5D7^*lLQYSt?0(>SDycVO;2cdgX$IdN5+V%
z)&j{0*|P9Da~m9>zW+b{RaL!Bv6WgaCZ|@pW~CC(xn#xuiamyK<iZe!5AQ65$B?rE
z`1hCgtGB5r{TDqO-YRDT86Y})Chhl`N05D+1E4uYZBrIf=Fq7MGp_c-^WWP4TlFLE
zeuy9FK3+)$*~W+%HFBKY`_#E1Im5z?8exGFURwLlndPBBNfsV_hTLUJxwgtziV9UB
ztAOkira&uv|98LN-WL&u+;PG8po-dt+Ua5$&RV)aC>oMkip-l+U$T?tP0)N~5I9jg
zXLDc(addG;a6=^7E_`VD!{N3yAOKc%_`Y@|Gymw#oEcL5X=!gy41y9l9^P(z3+OGP
z?^ap+Srp*$LCZBe41xfg|20dS9JLORg#>baeZIaGGXMX9oteO9XR{YP$R1I4OE4d1
zlFUw)`<g8<?rn({N=%x-xc2$U_C!jg5Y@R#0#|8CRhSxC*;idH+W%A4YX1j7X&_tg
zS@y-GOQGmC=BB)&;>-Y`fB_@{15zeHN=s4pE=g&%BqjSq%DYtZNf&h?#-b8P?>cXc
zmr}Pyf2plf?C#3${IBKrij{FlR8WMa0GV26kHW26jTF147OSR4X_Z&u-rN0Vd*;2r
z3MnCF+jQ$=BLieajXp(dBlD1d(;ndge%IRCt^fo2*y13@uqDM&m}n5wh8&HL{U``+
ztQx1*hF`taLIRM1h=}_+5fKrQCP<JVA>@8}zt6mXN}|;<Cas8&MYhqq_kZ`l%*lU)
zI3}LstJY2zPC$?Z$Rf#2Qs4aFHsAh#@43r+U)xqn2|+|eWDp4wLb<-@fB%2qG=ELg
zcGR&|RYgQZ#1c{W@9kdz5CE@kUpoL^{73EwfY(1+as<!=$^ncHQ6m}^8LeWZByy$7
zkn+e?s6?vXaMeCVKJz7p16ZXGI-OOct1lQk92yBkjQkIBzKb&Mu{?P9>ce|{?pYl0
z|NjU;S6wnXXM^(*#{dHTc&|%B!-E4R_#+4<>R$WsL}~b<cZ`hjF;(D&Dsgk{jC1h`
zg+|6#*71AzsbjI(K7BbpH+q3MP>tldd}I7_{Cz?adHqL@uwkcR6GnQBAG!GbQ6sgK
zr15pk#zPsYMWm<0RpAJ{5HR1xs3R9L1~~<Z8XxUL)FmV)D*71_gf@qjWEUr`pbkj=
z5+RL3Nx~Y1H?TH}Xo9T~mr@ggfgu<wDAT6TWWWWsV4)Me+}lsuhgf6~Xd754Vj6uR
znTnH|c4^j=SQ;{pjA=qglAiGwJE(9Y+mkSb0s^*PA0F6MEc!<(9s6}XA$D<4b*LEW
z^_ryx?jQ(a!Ln?Tynu<#WWV-Aj>4H`1kq`*PwA~9-P={n%(*nl)9~=9pvQP`g`%4&
zuOAs`ZkANoUC}t76SzYAi4}H{HY(T@5sENdc}eG?3E8m&gJ)uCjfLj-ovz9fOf<{q
zbyjmQp#sz&fRP1>kwYa<dsZG^vSOuTl&cc2Ruh*N-K82KVz${f%CT9c9la{Mol@<L
z3wpUE^mqN)05{z+(5qfI)SKQi!aKq!@8#&B@rhT*pq!2av}{`4@liT!rOEFGGKwWh
zBFSlnyaxHL6*MaBLChwZLo%0S9?5)?1tdGdc}L%|$|1?)luuH?sZdaKRiaFJSwT_B
zsjBo*?BI0V34SNzI>pVn!|5}h`(pfZ=WA>Wvtf?8=9zCnHbl<ih@*(-(~cs6!5az}
zA2&rJMN$jN6e$#`6loOcFBxSfpDc=OIdbJiK1IP)NKu<Dspz+JT+5-JLnDVKiWZ?&
z+rE!eH<H;M#~f|0dFESCF5<J8Vu_`eS#E`uR#|OLtmU)LdK+xCIkr%2wQbzq+PUHs
zhcj^2Ip<w)DXw$6`S&e8w>jLQw0o_Q(<rZXKM3YPX|%)U9hrjHI&k==ySubDB}IU!
zw-K#+5+IO8jP|<gV-|MsR&a1W?Pi{TOSFnZb%_jg4Ig<OHJRxNWzqMiKbzo4D|F(@
z+B2A_;+e4GVv38u{d}~R36O|_wgxrSXd0VhANb@0Fp^K2mq}r1QMbzvU&M9zAtFfd
zF(c-^Ju0Dikp2KOumqN9@Ax{xJE3l^A<u4Xyw(2FtX@}BY<GW#sMLwaPG*FV@mo+x
z9>$^SI5*SveW4Mo3*R0$9YQ=YvO`7JwX(HlQf{uGB-HtM|L2P<KCFm`kM=f^$+8dK
zi0$t`3B`0e04Eod5UY;*{Q30fPuTH^oIWX~D!osu6V`2;MjdiwgK()~?cmvX@z~pP
zFbU0n(T2Lh;ML9zF|A7Kx#2j^*XQbWzDQvXo)YO}MV$>m@sJj18L}s4pQlXK=#{CX
zR1)6hnr<&ihbw6Nv?pIbi8L{;(ZEeyz5O|#ic;L!QCRSZfP#sMNR)-PXxJJJ$|-Ao
zX`H*Mg<@Cd)k4$=A2}*FKA>jO-?L(xmgd#afmL28pFm=MzJg^Sf22E^JQZrHswEc$
z>%y1mrk8|x(!-Ih=lEf1=1!1&2}P8+&_a8ls0J%CcR(qRcM%Qj;rx`neor5e_DrDx
zq7p!Hs`F^<?3wdiu-(ljH+?2VI`@^yP+h8ekX6+6gKvMYC(e@?U}2y#`tlPmFr1Bd
zUZ+u&cT*KleS0J<Yb$BFTi=E(IqD=;%4;8bO-D@cxvLRt9#1!?S%WT@M%-{jywwky
zc}`XZd8oKOol}(+<n$se-F5W611C+=`FrNB>edMxBDT^sc;XUcuTB*cXEw^RX&p;)
znI0<)S@7s&60k|CY&PI7#Ijop+?@MB(zcQ>c*5yC*I%j@R_^*QUi?$Q<QxYbQJdXE
zjDWY!eVebn1NNw#eY%GmkzOc(SUu4#<?*727D}{T%Ic;c(5W;>TpSB_z3?+pyKu~M
zkWIY;J2qPUkW%UI+%D9oBhZNB`UPSolQBx{IR==ykcXpbaj+vYD>efQu~J+>7)3cH
zM>&y9@Yo~bE-#$19WmeTaft2VY0!co>@|wolN~dHSeV<6$`LkD(Dujq_keWb_K3)#
zgJrjB$rq~$6}5g*tnT^LK4o&bP&>n>m>Vpz5~xbPPV1^{H)%UIW*cB3(T`_tROw*@
zd%0o(7+gGu(BS5GJsF*i?uB3$Z$Cz_b%wP{{Faw}sWi_s#qFVjCdyaw;l7~;Kau7S
z0qlK1Rm<w4V}`6fkObKG+s`H;u#gxI*@L^_>UE6ke&bLr))<;W^ifw=k(>x@nCDs&
zvjRaE=`_#K(2WtnjbLEdnoQlU@FNx6PsGvTRrNxSSt%RWG=$!aB=%6sOT&P=vcP7}
zgFDm?Z1&0+X~bsvu3p&D#u@8W3ErnlJB}mo)wiaJ-=bbjX2{YtuYByk*woMGGd3$D
za~2B;KP>n~B5KR?uRhg(Gp{&a=iIoqE%Xk@`65*8!x<M9ZC==nf;fs8+05OXg+AUr
z+lZ$UY|s1&I!94v_5_hT4{+IH_1N`OHVGGzIeSkBSu)3dplbLJSBi7WP&<V~T^Vf8
zoG%LzWux*MQNxU>gj<N=7dP&}xRS9G>&Wv9Eq^$*@HzJ}ruJmhS&ky34|}(FG<lKD
z=)q61x4ooZZkZ6s=CLx}4Ds6ivo7jsyuRTNbol<XdbvH)IK?bKWaX%1pJ!Do-`RLJ
z_u*vrkZvh3#TmOdF6IaF@67{w5_Uae^a7WpA44>BOT=;sBj8k5(ukVoxpYed=mjz`
zGBL9VA_*2EFzkJpa8^JBn@G{wexK<e1ENHW0mO<Eubl+z0-9tn$8aS;5%>!vL3=D%
z<PZQ57K&g62VqqlBoOCvP}<2D7g0pVHWr&CS&CG%z$@UylrSoGCd4d4q!{+mA-2RR
z5|4AGs?ZufO&nfnQn7S`OlsMp<j~5cktaYtodPz6sed4S1oD7^mr)T}F_RK;N?DbQ
zSD~G132HdiGOOcKFIoe)Mv<BXX=c$PSSv}JP(C5LW!N+NLUn;yi-23x7F%m0;bDC%
z*kU`)^ZKH%2N@iqp3z~)m>hRfyi<bN>2yB6AIz=UW!G4q;Fpj|g}6m<o7tVH_o~;B
z*NNW1yhZeOvIqJ@#G5!w2qQBflAR!SN2)^4Gp&(REhVc|SO4C`s8uIcgC_A>bmQD8
zCf6^MVzrIbYqK3X+wGWQ{-U0_+8|mvV_W+jKCgO5PbR-E2O@w@tNp*WNX=DL=7G1|
z%mC;aPp{|}s}CM<(VF{v`c<PYuD{i%#jF3CZSlbZl(1nD6TqW#HGpv3A4bE%Dw<2W
zLLSw@BfU6=jLz_u+S}Sp1TY6ZA@Py7Q1C;<kv^X$JV-L4!Rd<kl6^@~qRJQc<mEoc
zRQU{3ju6GfmOjsj45>zn5j`8DCzxWaafmSL))6>TK&*&m^Y=!LFdnrS62u1a^9tFF
z*t&|w03T0tHZtT1Dq+0GfS|vYiNqRREFfX%`V|wQxHpAv&FC;gO=Wv#q2#<mg-;pv
z6tLO83j2UF3+@Na-~huPf)l5KpJ;v5MaKY;$^R6=e*n#%IeH&p4FL4XI59HDxe7GD
z1^`d^^#W`-KOiCH$FDKn|8yS{^WOMj)5$x?R@p8`%kgrGoG<r~k4729{qs-V{|Cr+
zM(zB%kT=U=vaLT4f?jBT9Jewu<%j?OzWRFqgIgb5cg<xN_#H4u&il9C`cn0R(Z(@F
zGb~K|W>ErQ|Gx-?iPufr^}{&LnwH!T>l>R}+dI2^`v-?d$0w&}=NAtiK6?D*>9glA
zMweIDH@B*_cW`uac5!uc_we-cMj%mWj1LxvClE=#WIuljl|~N;WH4DlY))`UXjnLK
za6w5~ZBs{YPv7DGBgal0KY424^x&D{(UGx<@tNt_IUp&DD|nRAoe=XpO=wgsyayNq
z18!J91%}>=gD>Ys4hMi?x4qYU_PoXmw_knjt+!u)<6<}8#mC=$_}S-}^vye%dslu%
zY1N_X%K8Ri?S1X7z}@#p0hk;-4B^EIw4Q=};}?JUCwvmg<lS)FJ`}I_eQ1dvV-*^_
z<}-_Y><dSH#}9$lUgEjIR7%9;AWITyWJ1x<p(ci2Ir7x1R{j3aE;PK7P1faXvLt*{
zthLD@ho@98Y9Tu~<$eyRV?v3&s%>(~YzYFzsruuA<}jyK|NMjID9Bx_sP9n6rj6pd
z(JjMEK~?9G*QM4~$@Orxw8V+!G2*?eW&Iu9eKHY;#(JCl5-tj`pTc_t<SZXVAXmi@
zv5980ZPww^+`xs<+&pPhtS~n~St-V20$VvK;hB|@5ci>;ytPXTpdb6z4?w}zMs^0{
zgiEDSPcYDL`qt5a!J-_@*{6;%<&C{W)Uo|Q;YFmBrLFI|1i|6mh=6988w6srbP17W
zE-De;3r<8(!_!=lrJ07An-8KvFoV0Zp4u3*Q)E-D4v?HU@$B-p62LL15!|+{alxW|
zQUc}9rN|P8w$OM9CR7W*OFV5#-m(tzMs``mi5PP@vXsHB0tr(JH-H!-c?1@%^pkW<
zsT&5s+df@47Fgg=f=$YA7nYMxOG?Nik&Hu^BoONVj5hb_D*)(ophqD0M8I<?f_WJ_
zJ!iy%ZSw-vUS^#@&(0}S0hVt-Gce(jkd0en0_G*bZiKiEba1)G@i=@VC*Y9Srsoiy
zXY+lH5_=ja<V;Tui**4jzU)TMxS=Z9<zPyRJi>B4EB>tu{~n<ltr?X_3b_ypWofJ|
zQiw}(zkQOhNenHN%<s1nRxpIvs2-OL#*Am2D^V$AHPIqelBV*0JiTWBf23M*IGasR
z*z@2%H*@z{39G)6R4B>+YxYW`E;7R1=f;-Ix{j82D;8&Tz7J8Tq<4+V5PF&z#>A}V
zj1d$nLlI?N)!(s_`&1PWC6nqTMC31cs_!I2kiqgQkllwsFfN?u1Gs@tcVLW5A2HA(
zv=;Y4A);eEWSPGYR#8c)oKY>%4%rC=N18oaBVKb<vzEU%^TCNkEhL`>UnPho3@NgX
zeD@4GdJBVu5y-HXUbZ-;NqZ$@EJ9mgc@ZO~PZcxa_(D+51iP2erq6!A6R8AaGYe2D
zQH9P5oscdFtkZsP+JneyJT(N&lt|Mu(yCV;@4HBdC4y98C+LC!Vi*tysbJirRVP)*
zRTQ2aD#pHZ!z|=ZB%HHhaL<7j6b#8o2O*cU5dPE{DiRuwmD6Kd<ZE261R;)nttn<?
zHJ&bH%${n5Kna<<7iXdCu9D^pIbjSU@}Sbt2isjcmvf;R81Z>kZJy;NHybi03^=1Q
z-{i!=9((nLj43q&4c~qkOnb1m%-pTGbEjwZz9Yl3PdQ3b8(WkU3)wuT+bsW(oPuc}
zd7w+%sj!nYk8hsdIIaQVj={bRdpj3G5B5nar^#lU|Jg7cyolc-uE*i=SV@ZPcd2<L
zHif}haeXvteh`Z0#K-$(5UM{-V2&(e*Gig{@*)bL!#{G<oUqf1P;#ry?J76f!tSi|
zc|?;Jr0!v!S)k>uyiBxSPe4uBnytp#AkC#UELWSc)%EZeCgOH^m-ZLp=a+w8p^;+z
zoefu0-LV_7OBK{02YLZXOkL$xc8i2YP=P+s>TCB_fSEJ*1P8C)X^{Qxm%CT@aQ*yJ
z&gnBdv)fZMTjJD5k+^bnAh8FyS7%tUqfCu0rU3{;lXW<sLDS$U|GT<|?{~vVeUc@P
zOS4#~r-osb#0J$#5d{=uUASYrsof3D<^T-7%s}|9YkuhTPV4L-`?@FAw_3G;S_Gu6
zJ6giZx}`O0hQv-M44LMwv`tB5*IE?G*A@BkL~EF(uL%j<pY|C^`k};wtVu)AX`WQd
zJ2p^KPK_qGm<zw51?fGaNDUa3RK!g!WVL!qzE;3~IqT&Nn*r~R9scJ7Z<y-O8&)9M
zdU2)v+q7w46n(cV8dRX#@vS{HN^{G(YuE1Gd3NWUU)y&&+O&N;BQ+$YQZx~Sf3e>9
zG~uvh>+&MBZD1#UrT%1R+`k_!lxUy6q?p&_6)l$}K6H05)V22QD5tHAP6OgoP-$@P
z9;ceXv|p<g%^-mox*fOWdo0He9#g*Y+u_bb04!MhB|Cu-NRVp(9U8;6mx$R6r{q&Y
zUgUKD9)o#BRlBhz&J1Jm`GsN(yObE)z-`YbP%Kco|0|YoZeNmESfSF+6dM3!h(cG%
zW2fV;G@iTlfq7&Jth2t~5=0qFkfftV7_Uq$DR41UehI{x2lNLq?>zBh3u5-#P!^J$
zmhqz+<_|%#;J0j+(DKyp<?vHpjdMPZ;Q;5aa`C{+nghcs4d<vh{kWtf>Pd>o`ZU1D
zDG%&Q=Z&rSrX=>^+vNn4j$(JV-kn`2ES+L9`aGgw^fKH^b~$67^sWf>G(7pjlXI9m
z=WtcjV|j#NgN&VvLI5FGMUiHzK-ykXh8c1K;MwO(!ATOad)<Giq<UIbCTUF$nptUu
z*1JjwSfcgPd)5(DD$<OqS9_Ip9C`LE_GKv#lGzb^ipLVMZ$+M3N}8OI=GJ9pBrpa@
zgUUcsV5+>J4UJMQWmFw6kKa|oZ7K}sd{DyKSzw?tMM5Lt5PS3E1vw4U)qGMB8y*RS
z%a9R`ePNk1%{y4=57einj3FhM_WV{EGujd;fds}IuFcZ#PRhtqiXM3}i_@e-5w-`L
zHE)@P=V?hx3wTb>LV(F)hYO_9to||=bL9%L4CsRy0jBif5SvH}iNG0RlPZf8R|_g&
ztbgvxdGDPoPu>Y61Pr?uS&;aWpyut{cWzhI9s3dvZCWOSt1<lFuD_Oat?1+LtuZ&P
z86fnvn>l4yL#a#;f%eottqpo_a!h0E^!DblO>NU!GI4N4P2FQOx&ByQ4z`$;%f(r?
zfak8ghPD^UpwUl!9vTHp+<@APDSP;8fV?$<Y=A(=?IM1Xme8VEh6o0-2BwYEGR?IW
zM1fryYiPBAgj+x0mla!%SQHMP*NXV#`0*;#Aq1NeMR&;=;sc1s54OvYWE1uw%sZMI
zfu@7kumS27CUDDPK%II|+fD<|0R!If3e?)%Mq>b%*nCj!)SR%60CB1TNwQ`or5Q72
z37*A=T0;h}q6LsY5Q2OAS1M5pa|94165ITTQrYeqd5I?;csrlXHZK>DtYDU<pI?8s
z4qdw|#>LOEyEwLV>?Ixz)q;)ymjLRzW>ltG>xq=D4dNbN52)SiuCUDZpqu+B#V;7o
z4b}m^`<kP!I4yNvR=~a{n7TX4+wDvKTZA2T;-qbA&1}t;J8INggly}Yz%s<Yq6v;r
zDPlj(Z}2%p2w(um@MPfxs7)5L#eMMO-Y&$p#jM+$EcDj|C?IyMb^>48fgZ)K2X*YJ
zb?jP)<xg@;g?2h$PC72}l%cC`?uXXvi%O6Mu>$q|PxyPOWue6L1y;<9(rGf6l--Sy
zEL>Osd|$--pWv=P`k8bA=e>fSLD&NL!U%PQM1~RP7(+3TiGC$+-F8#~c1|nKA=8u;
zjh7&0RmQP1?h>wCI1n7Xs9)xl+rnd)to*eIDn5Sv7(bxJk9nU&9I}C@r3bmAx9ohv
ze30m&0!3%$K&TBF`xksl&<12~kjtU|{1PxnJkWQOfb<*_pB{qSdo*&mrqm%&;FS^R
z*3M%;k6I$(_8lGU%^pzPZ9rQ){g<%rDDoUZt+rI}0!*$Yvf|0~lX*h3Tz>%}zdV@(
zOq!K-;O-&e1s;H}BH})Hgsf+vyA*zRB&M2gzPPWYzUPI%O3pehL$k(SStQn7hi2V;
zs;F2^avkGMkhPar<5377C4(iHBL(~vjp|N6?dha0a4YpiB-9hlK8>m_DHy~z<aw3!
z@%4dM+ref}$H*DUB|U2(a~4sooTzn1+XNpi9dUvdWiy4HQUOv9svvk+J^^<kpcX8p
z_@NP2qnq7_+m?2ff)JM^!Z;=lqpjsV2YRqN5rrd)!BrrLcB6LZeiy=^;sR7#7v(h9
zRgVK2UM_oKzfUpYpGgN=AAFj1?>wYEfcwhV`_k1<Ve`zIb9H@g(-SoCIvQJ-Z!onH
zth@r`)s&KW8Pww;aBOl^eiT3UiBU!t01qylxc=^6%W9YgB!QA6U0RP9(?c$*Lp}S#
z)UBf1YPmT}jDdgH5vPm>d!PUy9wY~0+FSmuJx6NK-!(Unvz%8V3!={{^U=N_9ED2$
z75jeylKhdyN!Gj{GN`7jO_nK@Q~_%+nnk5W2Mr2>fWE(zxncGP7LcFv?vD{~J@B-e
zWyaeM=;(dfxP<!mWJ4z`_FZ`F9(TM67HwR(H~MHpOIY$oP-4h0c|Wht25b7hsz%%F
z=m2L(IkiyB(g>#(?~uUqC}Ir1|8LFki{5O7{<Q+lB$PWS9+{$JD&G}q9O~^&HcmoM
z0_z{D89Xk7y*sWYXV+=v<BHXQ|E1{rWb<*%NHp34<3`ZlE@rf=-RCbHK63VASI_xN
z<he_-`2=Ku`-erNI}G2l{J0n-h0h3*{VXQ)r@wSvxI{X8vAg%&1#;bZZgvLwb1`T#
zZ-~rCWBmCe#-7>7#xisRETbDkkWL49ydIB;;pZpIxD|E$-QK(7d1?2_uq@jwgiFVZ
z;N-{Y$zw0h8{8vD2m2MpH#PShwD&e5a%qYlbh^JmvKz(4+n?a=PVzmNcg@#NTCed@
zNqy9Q?q~_is(&rSrLXhglJC||cd6@eormdPV_6+7erqiCk&lY9JT}<1uj4UJ-r7kX
z7j(2Hd)E@(Zp@w8?pHNhdzSG#&K)amwlc_YmqU`Wb#l|{0_2Ap_7~-f(u@;B(wo&j
zs{Gr}fKy%q0j?zIdn#v(Q(I2v*;b`<gNKU>w>0&~@Jyts`*}=3c}vVI)|9bx(pLGF
z0lSqh<T4r2Tslp_4`d4Y4&27w>W<*WU&7O@9kRYKZoQ$4zp_4hYFj@zv3J5cSo6gG
ziLpVt^himB7=TLVU)1-|h18|f5l^Y9i|f~WoHV1{qj4bzMEhey{6KSawl-z27Fb8K
zBnJ$C;BRaJ0Z}^I6&?_}q%JAo0{oTK{n<gQ`Zqzm-*Bd%giB0%U{6s*8u^5>NoA#}
zK%eW)v~Ov25ryPaz{mm<GnH;fE9V;|TDhAzQIRx!x63kfeMY=-Zlkv~p`*Ro_%w9t
zaKZkf@#i5`mlI%r7PccktA@UAG?O(`e)HHymV`3Z-o7<yLyyP^X43ORQ@VMU<+PDU
zl7BY!e!Q{1`EKdN<?T-&gRj9UmP%)jBPmZe3=mJ}4HYgLRixFVRo0#VZy;ZAef7Yh
zN5Ar5ZT!$OCMh^EoXO^rZNkl0(m0S?KgGWy@yQ;QT{`-xpZYKqx41qH{w`+r#HU+N
zS3Y0s4clKC2T!&us)?hxl_Uu!-O|>8v-T^#(NXN65E_Y)7oV^PX^0W10^S6nV7CY-
zB8P+$sjQ#`@GXy}pW@v|zj8M#B7un~>Xj!nQS0-DiWlMKsf}r6HRo^XEV9^9$+*Q$
zR`zT0L(7@IoCHp#DMTEOLT2C!9v1L^mRzb_1Br`kahgSBi<j*vDSy7NseWJGH2apy
zyz}?Vm?QFdl+0Uytx&1WAFHZA*8XIKrz|T$dq2xb`QZ7>;Kzsmd+&=^+iH}X{fR!O
zyNjg9$wk+$IB6dFBuj~uw~KWOvCdEf_a;)KSxgp<cTHFP<cSOZbS%gI!MbhX@m9@<
z+h?+5I}YO4G7q-<F!&XLI#lyqFGOb7WXshb>KQ_?v63lYP9@4Qlp5W&ajqok-CX-m
z=A1^w{TG&#R2r%xw@+W*j9ZyiXsSUvWOL{)``Y=)hy^l<?dmm%#5ZFD<GMN|bW(xF
zh#!C_Sm6O%p-Z0J`@_5tfCBE)3pUZg=1D|hW=X>C1M2l{DF|%=E-VQ#3eWiZ?LmAr
zmRR@f#?rZ{Xfoa>49B?<=_hhbiSu}RB`jQ_g?;SErT<1J&aa*vY3HN#3C6E%6J6pt
z9_MZ4hm|RG1G59c>;g#R*uu=rrOQ)`Mb2BBIw(f}{3P*dbD)U*e2TBtQ>E|8-Fxco
zCvr&<dgMyFFzvYtV}hxZE2~2n&aRH8GK^;3z!CfJr5QM<hiAm^@BIq?idmKEx&_|y
zD*6T%ZH&@DKugJSk@UC&Mm`>IHhzB|b<x`D&;xS-AG1&Wlc1_B?6(EmS^qo*K5HnN
z^T4-(tES%bio@rxfJXrbpw9!P+Ifyr{$(Kl&BjBHYR^41FE1NFwWD><Fx~^{C!Qa)
zx9*=i^p$bCkNbbeWeNwI!*f-w>*sD`_f3yK>**YMJU%_K-rY6)5PkC9^-Z<*=jC;y
zkIG_DvQ&H?=qGso;;%V`i|qSkw?R2Qv@u7@(K2z@*_Jgk{-URA_;Fg;fo^~5P3_`H
z+hnsh+t^IoV;?a084-EG(m-og*cwCs6IEbL`?u`f*}TG;Y-{YFegB_YE7t?9>5*95
zIdJq#3E&#wA^#xtK2zxFqM6}N&_yVpTgIs~jgTfNcT<cO$6*Y({gujcf|p#V?PK@Q
zBe4o)=@G*_F2vvwN_?YuZyeb&89S}={MQmmubbdopG~QXO`kD8xQajQLcglXsL4u8
zE~o{~t+SsdEWkn<o7czd_2Uj1hp0niWC5aBV32gcAR-V#{W1fy@IVO3tCNvwyZlr%
z(~yx!mIBZZ|6SiM@49uFu3ZQ;Dc~P~PxfNT&d8FKoxLUNvx=<B1?UR-RSFlYsHy7X
zIVZB>L7mc_!hxje@+8G!<zYAo1F)t3Z2wurE$tUxudKc9?ER+s-1;5&_TEy>++|tX
z3rJZ^zUcy>oUYflkbn+xv=mH}pZ?*0Gri}(VNTnEQk=eKIlg7Re(b(P|NT9R8J{z(
zkkKLflIW7L`)j0~f-=z$!^}G$>iu#x;ps;pf*%^WC@D$d)<<D_>uTl;E@hH#7%?#@
z^GK6%hb6+<fn7CLqnPfMo1h1%1g>yxaZc6nvz<?qNf&eTPU6E8cyOh64UDLSsthr=
zD2$2qCjT_HG>-lmC$bt!_h>}MIKh`E6WXu7*HQg^t<2BkeGX2PkXezIm0yx@FcoY0
zX=K8f|5*j%;d!Hwo%w$zhDC<yH@Lw8Rib-@D64At`A#?9=eR|zU<88NFcdYd`M`$k
z`Az9lhPu=NMmD2Q4fuDNZXq)ga9@4BTl2;Iy}M&k$|$xBRNW(opWM*J*8A4vd;h0h
z(g}wHDKl^(IC6MOB4}1>q1hZnfi6$Xjg_Wbf^_pfeF6$&2IbvtQCN~J^!J}5ZbLB;
zsj}3B*c}*^P^Kp;J~coi&{)tC+4bYb^{QHwoSOXhxb38Zw|33j%i|Ul`B-CpKkblw
zCzEUn`vb=#g#MwJu78VGhP$VLkbHDhKDfPL#XdVxxxzUg9h=^Oc6UnLn@^915)D|d
zeMv!hA9OIy7stY(NTJsf0Vu&0?q1P7Qqg(3xys#+io|$_`C;>sQ1mNm1cSi}XHY5O
zfh<NuAZWGWs4M(B$x-Ve+6Q(Z3N*LGKC?C>bR5UJM1D!=GVx6V87MW<#Yc_fr;L4a
z^<v<x<Q6$q)}vQS#@fl+UMH`(^}DAw9}b#V$flLZFwp>$O_h)+r#CR9r>9|OmGcZ}
z_;p0dBt+{>?9+6a)D>~Q^E9uQ{_J%IH{5AR-2l}+C@R>XNf>-F^cyc>7CdLM0dRJ|
zPGU2AlnP7A``J|teM-%-@zRcgmI^l;8p6kmj`45`JeE#HV!Z><F4f$_meGbF_AywM
zUG#ej8VZf#TrgPoyB%e3UZ>VP(v*MOTK4u;`k}|#axNb-dF6B5@&h7%)m&61I^hTa
z^>>Bu7w1$AKUY{6=RTW<a`ldPyoj>e6wY~?HChWwcbEO_yoAL@p<W<)Ut`N?HPv)H
z+*7sHBa2*W<|5&-F&K2h6KQM&#~U-0&lpTp6%?-@y%_S)3cJU1-5)+=iB84ANbta2
zEagngaI7WYr|9|6GXJ@SP+?Ybq0$m^*9P?LBlOq*vH?(N?gQO()+7zskA&#72?oUv
z->wK4gd%})ZTfM7z#upn&kDEWuJB4X?=C8{tZM)Z<X@m(EWc1rSF4?h+5<|S@XE>P
z&LH%}#mC_kzs-1!CXFR|rONs|_q<6SP&%gM?{*FQnmlSfYz;RR2Y+G=8Q5iOqcrjA
zs3GcN;zXjqYJQoO7`r^%-r8hlG&B3_qy%AR`p*p5QtB@>uRuD*8*hQ!<mABUMAy*0
z73!68Vc>J!a@ii<29D2y`Tb=Dk}3t#{S|z+<6Ss98__K@6AG%5EmsSJ6J23+Cl2zP
z1uo4ePqsO&8IW>RT%4@BdOx~kauH~_(_J>ducA;|rBJf3j34au%)mR>qf72w7`Q^N
zLOpaZ$<^@a!-d$iz^8Z@NJD2s3-V;sd~g<sZ)v&T2QOEvk_`pb$QLV}iY*Cu%68?t
zXlXCUGys%3PQX1)mw-`tEfiPxn-~+cnihDbE5=91+AI==HO0eJbRGz66OHcB5<`@c
zG-DEL*iV5ms8{#-3xfUq1R;LpA^ClLK>!7<5~%D)5_Z{v?A7<0svmXF7%Y?--VIkl
z8@s;6n2=N$GT=#I5zhQW7GV)=`m53$Jg$FjGj)xD6h98w<Z2VDwyAw9UYi?6vzCh+
zlrHa)3BlZW{eF=#ZHj}t{wVM*WQQh(YRJSXV5G`zSRSOit4Zee2w%LsaKRYLcWDjg
zT=x&v-r6h3T(VFL+`9d$YU;1F-H*n`oE!`+B8&(ZiR$XLNmw8;+yGr%+|Y3F;I>im
z`i1p%aWODJ3*2_#>AlBLbPm>s%0Z#2Y#%J0odZq<m^ZGhTnS_oH>JxBiVUPHjCOhM
zGLo&(FVdGS-wGM8b6WAV5~Sw!NEw<U7=p|Y-6+3(CF{p7a-N!@jcV{!KsH=YXRCPT
zj>oqyhAz+6+r>P<juw&`v?rdNRJ1X?#CvW**z2)=nsq9dyz>vb`s+JC<r0!=x3>;y
zLj31fs9irC9;H`q2I+-+>+i7*Gvex%n0~*q7r7mesG0^Awa#JSDCoza5{QADOB*K=
z99+e%>6(kn8X1mKxV(HUxwR!-V|hs<-I2n@CBAtu0U6_-?o(+VQxl#U6J8$EY3|da
zzq2nME{RGU2-j#fk@WCrW(Y^f3J-U(U+u5cMPGSxtqQ)7{@<aeGTK&aE-5(-%)pb2
z?jhYL>uNTIHL)OjX)9@VT6S1Q>hq$#I>x(^ng;FHc9j`K{`_a}cwXa!gvMLP3pi`Q
zfVExF`zN9I*Y+}iU4g!@1wI7X0hdOn_@t8ENc=x}$I*gxEFsxPchFd>;1^5v6?BjW
z5`&{7a{Xt)^eEM#*)}_v?8b7{<R$CLt12chST{4aJWN~^&Ce_f<BAHS1(}5uGw6pd
z3A$de+WS%Uw)Cj=fS0*@Yj+eV<$3~tw5+y5XpqhGCy+fc_PmZzL)pY#26c*we=+bZ
z7WOaB{?1D)oOglL-xzp|_Q61*vG2^-1Yd}$zmY!xw*Q>&ZOg9@k0@HQ*$2(!Q{yK^
zEe8{=4f^*P(~Gc#A{~xgW2#b^iHWv_wRxDndZ~Ph(~^kJ66xDM1fccqdR!t7<4LD-
zRoo|BI0y>TBGQVY#*o+`Up0NM0-96R5mwD{lI1ZYBnW1qo>g++KP^<%%4TV~ifS(b
zwp@U1azLPf8$bgY&l(}m^8UeBYptD!uEa~!3)xQ^Z4|0M%7!{88)rxxOZ0u5#Yux5
z!+L6v<dUh^4~B=<?@v#T{T@2=?7?5h5^X;L@0ZHuQAP3~ksTlF{gu%BXS=Wi3+zDa
zuc^J2Q2XF;e%t=%wj*$nL_npB9sT%W(W(ta(5T5^Afz+s^qteB;r6cGrMBCok0>Ue
z`Y`SvB<>C;@W;w(%Y~H8qNCZh#o;=^^??<iQ6%Sc_^JbOT?_PRzd><ty~m%r4?z-l
zT)oU2(^d4-XJ`9VuQ)L@5QauJknqEU7Ymb<>IV}q0}l<#!&E+dP{^jz1WXo_&rGq@
zh9t+|GNH#WLk$TQTbdPDjj5})q|I^2Jmxoj{4RRq=zVD4eZcO-r9(cAQ#ut>9BwY9
ze4+_3RqYkQX^8<7Jj(;^y?amI`!AuuKA|}_7eJQ8r^B2(BtxU_t${htzxO*<LVNWo
z(Tm4#E6vSsFBU|tZ0)*@vT9M^1#yr+!I$GN=8$}eu{+@}fNVH?jQ=J6zl|PNjLlN`
zth}ks1ljk*2Y$~c7L2&87}tEvv0>+^L}%=S=o`}oW7TO4gG~IkL|_;u-k4y=MGhF`
z`D)2Aq3lo{jqcsSG-XOI#Tn9g8k4xpgyX$VyL_>^{w75xP@MGItwbdNCRW@fSvY(#
zzDNVyux$|)cli8{0B6ABIC?Uh$G7|n?h&l3vS{dm$@R07=O4~qjr%RJDZvNq1Cm@E
z?;Nq>AI1z=`}C%XGNohH5}D3C5LXgT-Yel-VMt4dIK-(TRc~Y+&J5w5Apqaiw*A<r
zp%Te15bEo_<*ws|Y^7WK;ltVnZl0b&z>)sQ$`+vZ@#Z!a6UI+=+@!XIS2;P)fh@&~
ztJJMF-TSB2=t-;YjU<7G)%*J_ZLMvy5qk?;%l>{#8*>}!ZEdbXx%^(Yimll#b>h`r
zT9x+Ms~p4DB8TA4x}5vsfKm<2#S4+K1-%$O->%VzvsOqL`%3S#FGR;jyF!Fz8s5^X
zXV?LviD;wu@uv~(<7~OueUkEscI;VIvaw6blo{Zd7XjZ&Y@!qg@8#n#R4b|Tb?GVW
zy)x*)d8x@9&~NB`m^ea|Pjnu7YyprjD8R0m!4|odYtn9iI@fOpqu99;%%g7;&4{iZ
zSh&#Z@U=^G{T+DIFKH!Vxk5u}{L8<cWum2p4!N;j#~bDu!1g>kxal<1<ia=j)8%X6
z8}563HDQ=z;Y|cmo#~6BOOSj$FTTpEI2w2VQz1{XYuKsT5Xe<Eq6TOq4f9n^1QK~u
zX;huF>`J*m83{XOx$;~zdfA3Gc7-`#0a@TxS+yct)==Xd4R)d|l>EvXxZu%}j`FzC
zRxH_GBmnU`OY`##ZNzK}!8gd?%QHy7k_~~?-r~S7xWC4_CFKS`cWeVIky+@VbMGI+
z;)S>PQFLT3_r!&pkIp#gIYLI)^vIo{xs9#6>^s%>d5GYyD_2PKQ5EUz$fYpU{lTPK
zxg4a;Js*&=0s|j{CxxdB_*O141~DZ=za3}vb$)WD>{iOrsc}~{9{;QH5e!SjF6dXP
zMM7K!D^28AsG6UuNg1Wak3wU7b%&_>q{?0^!>r>{TFSRt*}swh%jV33r)vZ7Ez-A#
znuw)C(N3pbwH8hGzs8lGJ1wTIfr(_FH?ay7>|c|q)JVa3!!ms~8yY|*<~Em)t8p+3
zY@0S}ObjW>&?PT4z{WGSq#PvHU;kv>Uk^_W(_|yPt?T}HyAHR)HQc;<4uF3&^*>UV
zaetDMn0Z=Qm=d7PY~dI<1;i9=IJra=U?fum-zc;L6JY&@R3V4W3>CxuCoDDgGQrkt
zV0!^&Om-D2_I#{4qQ}dY?n59D6ZpC3KH#;e)XSldt>)>3yqgC=wYtXIS5p`eyN0G&
zC6sfXel6(o>4_jdY-Zi$K3ER3W|$w9K}<YxJRVrdVG!~5*!!(TuV1Iv#<C22bL!$}
zSzb}AMXwBZJr0Y~_baVWyv#vF-84VMH>N_)t~iDXTE%U2SeUT2Fc=n?>?IQhJM1Af
zi?qGFrke8MB@vA!Vsx`5HVzB*7DEnNRgjHuV18!%eQqMks$a1CE|?_@0Fs-oaP=H<
z;n<6(96pnWa`lD|zvwIYRRtO&$=zoFkCJAQ$YvoVBYyst_`Mk+68w_r7>i5uQ;QY9
z+s7AYA}YrLEM|)jiWe8|+!+s3!Lmh$SPnml8SavMDMvjqUimrKsM;;n0K<cn$EwK@
zDFY(@s!~9$V3S$$?`&Q|fKRv(Z@uFs)Eu5D@Jzv9`=zHqKVEiOeHF^c0oMJ&!(Fdg
zzpyyEktjab%9M<bHkGD|qF92`CS%Yo3oeEmVQK2`!`o7Je-u+t-Yk4xzdkxk5#xmd
zFd=eAPwJ22n^_}WV~Ne8C^Rp^DwKj^!snJPk*>{l@X6tLni;VCXoYHKzC?XQTQ;|#
zE^U8Yp_!Yl(66prhS&SYL^_5^kq%-B+_qSro$uGq{sYK<u{`^(yA3t3Erv(84F(*s
z4hxij5|qCVm-k>F7g#sLe!&p~IQs0dZI`L6v(YhdsT)V_q}l784;Fdn%{DyVW}r(r
zNl}`DW~s`<9(q<Gd^xMAE>{N1QZQB@>Sm4O#tAdQ5`E;MH$xhP>jY2y&cB>K3ff`?
zD)jHXl%qzk?ncROHK_A(V8T-N{yyeF$LB_4li>RO>d8BaAiSQ_i(||AdOt)ulnCbO
zNG63E!C*3C7lAGvIO+z!KJe?Bt)ZI(3f#TX9eDN(044nL$$c<(#kyfRG}fse#M<l=
zreex@o_V2t`L*jr@&og+pMHL>(0k!wQuxei8u*wwW=HggZ$I7gPXGPFrEfm|`RmQf
zQT)~N^^v1Vqk$$rdgqiM_XqNc`Q3T(%<;9L&B;G*B<;V~0#@%)5&A4n!EHgi(_dc;
zFM1ro3_!w0!YOoN7_cVJ5=`NAFPdtFA$1Zr1vIC+zF2GF_hmpR3Y*{e)x~xN6H?37
zbpbG_q}K>XxeY8`FqT^+e@|Mx+1bC?zSWCtJ{}xiY~KX1Bid7-m`5bb-i@_@xIs8N
zPKEe6#Em+D8y(Xj7Utz$bWFKsL1J40Qj26yQP2UBP+V&1&@rMVB?d|eR}G#Oiv5n<
z?!SHfpMNDwo-NG1fBuW#D@thoes=v?gGYtX0!oeV=0Zq^#Igv<9z8F~7KEx(V)k9l
zD1w?(e4`bPzFKK=t%Xt+)ycoXM8$QGdx5ck<l=3OHW5#iReF->OaXoI&@-0ze5mRL
z4DQhVNV>RT_?qh^%PF++ZoWqjNkk3S;MpRZ_@UtRTC-)#8C*ke#q*peKl<(;H>+cO
zBk-L@aG;+sfws~gL%Z)jvlrt^4PLKu<Dg?S{br&Y8KAjnTz~fMlgoFnzyxpUW6-i=
zH2u~`ESs{-<Rsswi8?d%^EfEib`-__dxw*bU9zEqlu4O2KCoxVe!Eo5TLIQf-r7Dp
zfAaKhwE2B?e-*5gZEc@kJb6006Yh@Pi@oXx)l02KPI1u>0A+{Ga~p?R56X{pc;=7>
zphe+q1-Xf$Y6+BD6stEiugl%(;;bj*X;<jfFz(`Lfv#1_XIU@_lY~I9)RwTsSzX*+
zPZlQG22Jf1f-rTtA5&MI2Ie}Y#a)O<d7d|?YfQ|;^kRKl^5W1kWF=cKk$Q3pu)0WO
z0ZbQv>q$_S1<q-3tOa<qNRy=1jC<gsvjM0cAN!%zR}U8jQ*W@jarfd@Dh?-wl5Cg2
z1;?OETMwTn87{%}6>%4c|G>{rq1a`_U1Y?7%9+1H3K*@H-FZMOTL43gUQ4h~0dY^e
zP>fjdC3%u;3LpvYKRMi!1j5~j0HHds77QnEgc1Z6pio?o8*r$qwgK+zru7Ey8Rd-v
zi-Z})nrDN%S&%F4oSqr5CULpO$tv2a1Q$};20_mv{U??y_f>m;`&aNu>i}=~ZvYVX
zt6*^HI;3pCmX-c;8jv<9TLzS}5<(n!t2mfEB?rbLoTzra<<<s#B|xZi2Y3v_U_%2<
zQ`Fm=1~dTSRpO8e^7Rot$~10#VPQ(yHSU=*UhgeKSuRXKw6Qb-jgl?ZCF0ly1I`8E
zy|J}O46i${bvPaD;`CaH(d_hGE{U8FWDN)g7wk2lgsk6!^oIuSCbhz(I;%ti#!m#I
zKwHu-=@}j@OQMdr8jT}EOba3uqwP_fo*fNCSF=Q03$xw^E!0{>`&Byv!%N$>t_Z4F
zLslVV^Lf~^2*jw@kWjLcNx;fNlj!o?QqBFKq-sQwsKx8I+m#NEW+loTAlR5n*t3JO
z%%bQSV`Sej^{qK@7y+=4J5EJn;ovIAm0P$JPg*yPlj-i`o{3$xd9^TOONX-q%U&W0
zqs6^ZA36r@WM_{xf)xSXJd#!HuHchSpqayp1PmkwBQyh$xSL|)oT5W_)oDF*#GKZS
zl(dMFX%2U_e8U_WXP6R-nPAOiGMxa5PZdrA*~+ohA+#7J!QfPnPKk(bSL@VpfAbwE
zi&`eWx4`Lw*wGGa2+b;zo_-%jF;Uadn#c1w2qlW9)3`#jG4J9@bKU$EuWf7b#Btu7
z3BwEvNwwJB^+OOW4>cg-c~y8=T(>MsNI0KTAig@5QFlmy=e4T26=qHx#a+DQc-&(T
zVB`qq>f;q22Ukx*BnQ5%74T)!TUxxizj}A~Qhhf_EZUFyfm`*d<eU^L&$|@c2j%iZ
zjf!aM5kU+rzN0nCAf7^^zD#8}z(EyCMB+~IyDmFsb+SI8uu>{@g>ktSVp<#2nRI&H
zgJ!z6kZv)Riq&+fXr;!-8cTxauf$o?_{K@ru+iBqbD+rtu8CoQE?4ptf&ETV?r3Xr
zjvZk};7j44wh-e`qDkirMFrp$Zx(qb7J`GbYf2`zRjL75z?VkLTJSu~>`?Dwt<LH(
zK|u#7!U2<%#Js4j`3Bya(X=zEFd{*%h|1j+=qyYV!!3Axw?2b<2VJxwc)e4t8!1XV
zuSF~Fgj9{keo&cs!CMl5n@Ig@v<X}i`soA=j8nl=R(l{pW&x7AT%BdJt@MQ-*xqV)
zQ87$1XbP}`2x6ImMG;0P?NM7XNa<%&U1+Wk`l4~EpC<G!IhJ>(`7_Fw0fkH;&a{La
zuSPyizj+*sRHYO~!hB(vxJ%`lHa`9nb$f^VzJ&j_67ZlLR-rq5`}nv2e%$oiEGdZ}
zrbXjVvbz7?SP~AyIE9Cdx(~muAB^5CQDzEHdjwekyA!A=i%2Rfp>;yO@|+Bv4f?&;
zqSsGDI0!(|g~YRYYpOguT67X-H#9^hZhH}OcOK{l@P=(0H$u#q&2;i0W<9|N3({rc
z8nDT%^O`U`jA_vYV^UnUaFi&m+G8qH&{GC*AoY?Di_wFta8w<6Ef&}hD5XWp@Z2D*
z85usksl~Mx;laI;&~yIX`zKy0g@L-+?`pzNI6Sm9c{Hv+YwD=zdREZZw~v?32H3Ic
z{Bg-|G$7dZBubH5-g4?@Vbb&uJazuAtQEuSl^nX|7W-O@@V#a?HrTvrZ{&6%Q;RX;
zD*_ODNpTm98Nf1dgF`Tkg0urdK&I{DNC=vMr3)%n^juF-Yy=zz&<@~^fNlz3O|i_w
z<1O-Rz@~k)0ZY%H`$Yf$CqnRhihk7~58wVt_Fumw)$uo_?}b}h5zomj=rCD+@3*5s
z=0E+qZnQk}hp}bQeZExeJQnL4H;Bx5E9FV>ejO4NNPPU!1nAjNK!FC%xU-UieQ78)
ztplRkd|W$##VtX71TN?RtR@c2$-rhk36X)-`PV$7Vpju~{U<xkJ70KfC<|HwdniS`
z^|wxjSL5hS00^i782FbQ-eodZMz6B7>0;e`c472dUYGTuER@U-U5ZWV)tzjw{r%LJ
zF|Z=&dZvGWw5Y6_T@ht|W4PES(u4nySc@*vt%bem2iaMz_y|$tE}9?LT3nv>1EsRT
z0P|2Sl2qw1gWSxUeC!l^(RE6g)osQd%E(OI+N7dPvX#bHOs-D^ftNZ{Y)R^)fQhIE
z2nNe|+lrd{fk4FABv?r1fKno}REi*DMXH^EB{d^>Y$RCbT-qtxxLwQ>i6rvfK+m@Z
zify(iyqe^@29OrBH&M2wbkB?i#7(RX^H-wb%omRN^>wfB9_<HGa{lC7+{JHORV_sP
zVfr5+S-M{lqst1i<zmbEA$<RH`<&Uf=?wVHA=<af-#X@J#*f<H*=O5mgV1^9hxpz1
z^}C<bul8i`i_w#!enY)ZjJm1T;&;Dp9^y0LD{2eqQBo^`OIqb8_&q=QviMi@qnftY
z)e!<r3@^g?l&0n?=)ebuVmD7Z_lPI(kphJ(3S-aFzzG|PGCmrp?^;Ti3!3eOfu#%9
zPk;-w_mt(e@?g;;qixV316UHqdE$(4BNG=EQb5scZZp;B`^Q`By4<;0)HLDH%((#}
zB;MN#COnc6lYtJmY7*_g+uIJoLs65ZV}ehD@R)l?v4KDm;TIt>?ua5Sgb*bPCtleB
zA+gEDZqTdUI99Lg7|6agq%Ax04ogLhfn2Z9M2nRjjh2dn*J6tA+d!WTo=r9@iiVnN
zsYXT{UKi#_reF-O?$VE=FN}n}GDc^OwY-*;a*zbr;YQD+6t+q>j-gw&_Rq<#v7-pb
zgW3<_HDASaZ(KhCyD4l;Zr?h>Sll!4K--(TsF_@nks+%hk(Ayzz;C74c+-I25+Bst
zSm<x32P+jzIl7%hNQWHPz_nE{39L^sh*5<iNY9eGRYnMBAgBaPPa+)}B|!KvnOrcK
z0G7QQri68{@C#tCZT>x^mg`pk*Pk!S%>}lS+w3<Hvd9^?0+RzmL=WrRa?}gXy*-DE
z9{_YHSaORQWBU{>_QY~(>IY}o1H!tRNYt+nkkl(p3_=Z>6yJG=m`#(cI*x=4%;#jj
zs^Ezi6zre)ma@}_8(1Zu(E6!XJTlv_UTjf1e-%wAH4;K=E;Ae$PA*6>jMzhTju;4N
zTthoR?sO6X0D`bWP@IO<N@A+_>C?H}Csm*B287eWJ2=3olSk-v79i5~F~no3Sty$v
zbjnapt;YJjm=0;C&w-K#sb~#C*qK@1sx2ck*l6GfZ;@M~KPi(ad2g7BNkb$RUMBs*
zQ~_??c-r-Mzl9Viq$Jsb+9z8QU64Y<%Z$AzV)HBrvG4f1&mH|antqFyERK?;-}cSl
z-8)SuXzi;Vvrb_jTYa8AM){KTy18|DG@2jYed0kkw!d=WUMKU1!|ZebU=akPaky*c
zR2&^c2WN_UCh6~&6%-|j*5}NXH21RHql-I6teNs?zmQaCX?0;*=C&P3ajk9ep)jq>
z`!p@mDQg<*aewbzD^=Z~4tnV`MSE|}sE%O3yGLtZm({#tK}3&|pmss1bt7w3#0|ty
zxJ_w6gQF&4?q##<hp|6(o7=|7laFZFctz5f^n)%rEEPLC;MMd|-q_lPwO#~3i$K-(
z!+~_L*+~Bl8Q_ZO#xu;Iuxe5=o__+?Vo60#teKTixKGGL?=*U(D)h~!yk7)K^~^!m
z8A-c3vB%h`1Txm-ynAq%PZ#_sHjFQOI`9>5oBgKg9#{|Y9GtlphrWl!&-rKbpCS8G
zJi-}n4n9}>g!j{L|0{l1Ho=?lI?47uA`UIz?5{s-#j8@%6H4v&!~JJ*7bp+c^IXuL
zf6n^`Z|-^H%?dk$4m`Fa2A$P&tK4Z^j(vgxN*a@*>GBrr|GO0Fzdq5gul$=6>AyDd
zZ6(j&$!<5V0|Tyw*EY`BSiZe{tQ86sE~cW`ppw}s7RNYVz4)+tP=gVY%n|~Il$4E>
zL)$L}zUK0v@D$((3}uqlnCnb9$W0%5(uLhcvS$0`rdn&N-e?g-*lH6@SZEw?$#hn2
zra9P5R<8oT((Ts%>G|ai#*&=`90|3@TUx%hzO#E&F4TAEm2nvb6PH}1W@hAs>rIp=
zah>QaLb9f$vW6zE3?*H9c!$)chkREn0|vEKB;bfxi;LanI9h6@3IVoymiY=rIq;ut
zLQUTm*rp&L0&yg8G7#I?hJY??gEq6yye`k2`?ZXjtLpP(XvKwEu)Ur6e|}fmc4jAR
zK}&c1wskBIygt|VR7Ye|MjGpOf9v7nDHIJGEzV1!;I@8`!Ogn%7hNdA-FXwt&DjRq
zUGcG^eQdBnH%h!<B{rsD3{$%<5rCXb{KYqmNZ@M?_I-&XBud-G1~m2%nm~*!#)dMN
zK;k4prqg+X^IhMpoiKGV1WZg$bQjUE0U=@17u|pZNOUCv+6^f*<f1@xEI9qFl}9E{
zxav5~%HW>E_dwS&>g14mm)m_vd{Ph9Vwm0%tde60*r=0+f`Tox>&gPfS1Zpq8kmsr
zI*bdJNz%{-65@qZkclgXSd3yUfecjN=8@C_GTSL~xynxw%i&#SG+Tc3nj5y^vY4I+
zAw!T;tEgc+6w=5dC^Df_OS`j$5{H}QHE_@7u*d?l%tkaAy5uCrdRPJS_On}@0xZ~!
z&}tD8V~cL00WA^DMWRQ2kc{snS>mm-Oj0Di%rr?TB3U_*(7JzJ8!TKO4qrx!A;M;D
z7}(>~lMw{a9H@X$T#h%DK?3bE1nOCKB`;BRmN?0|$!QeD<=s2EJXsU-IK4f3_(u0H
z_sW8!<lL|BCFZiLFh84bmQ7+jhF9K(*gV`+Hws)jx>1H}86uJPQ$MQ2GD#ZKVD1@X
zDS8h?Uf4qCt}J~+$?XY%9so8;jKn+wSZj1=VRBK|hkQA;v%)Vm&F&n+tmDPW+lJqX
z2N)pjDe4(W?Hn<*40$NtJ7M6rOuP=X3OWZsq{cbu)!+4?WmJqQnetpI#54w+=p1Oc
zK*1u<h<4TH^=;1bbc}>_pw>{;ShVVG0#%YL-N1=cM6?*}CP+AF-7QJjr1rQVNu8vZ
zy;7oNT$MQ~VF0WbkD;52^=<hO)kgsoYu#Q0OM!8~UH=Txyk2oKR1^ArA{A%sj-S3X
z1>j@fCPG3=!z>ex56B{PdrQlzw6%S3cz#htvP7P}UTx(DY>gKx02j&?uz!GZu8p4K
zLd9N^qAyT2*8VbTCM=m@V9p3swa&BG0Du<04Ir1@$}6o|I!{?kw?tG3T-p0$&}yj!
z9!ncz+V;{DXIc#V7(Ymd<{7B%6sB@eF4IZd8WA~En{S)8F0350+&V}~TFI7+6UIpk
z8V-FAu*P8-S2q344o8qhGRrog<6b^R?UVu(IkiQtJ#kFWh+2*IzSl?1(nZNgvoov7
zBMAdK=sF*e31tBdeD8)#D_iPziZEhdp)B6+0Nmjirbh8?t1<OBwYkwGaB0<o?A8J$
z&v!_TkolwI(@T6+IMTtB{6c+=GOBAVp-gqS@yc}V>2O@$*xH>s;}DTt;qf%jfr~TD
zsoZk0c<I<B8jAL1Yj&-u?Huc{HycJa)a^1o+s|#AerUQmJkODpHyia#dZzBoG8yI^
zQb3aucRaYE%(Gb(S43rHV%dTSR<A~2nA$&a6`%J9Kyf%%d*gGIDIa}JL=tCT>(?n$
zYrY%j<gNy?89Zc{Oii+9!s(rB>_!v&{6hWoaD#GTW7wsnlW{Wd4t)5eP3hqZ96EBt
zigpT*iifeb;i1HM__G0dxWqgJ1}<Nxd?)E4%pF((J2k3u0hTm@H3aw=*?m@v?)<(Q
zy}~AVv>`4yY3_jDCJW^T(@%RRuW%GB%8rp9E}a&zOWo~?X;jBgZf@+RqAL@YWY>Di
zNxM$4s}E*6NjP+#UQ)4SveYpdcL9kqIU9ghU5xgOyJ!(fwlJQM^I4qFRS4I2${PHq
zJCo60E4_Q!a!LbHd{k>CWz!2goX_UmPknn_21_7O3*iBmkm8^!EMvk(!s%HA85|ws
zZ?U+UUd%HSM`wnTrwS}vF|ds^gnb6bVt7gBx%vtgxY`kU$P}RLRi=S0*PrI&o&yHU
zyBt^LN#^^^R7I7}h39j9&QE7qa!%{)k+lg?7mp;@)b-th4K#6oy&|Ks#N%0Rz$@H9
zwYJM&=8RG_cAbw}c$>1)S0wCWAvev$?o9E1zQwzb(#*{^1)Gxw1BV%#mae3ud+UwQ
zyZd2pxARD3Uhf{QFT>AmF&3^N&-dUx?iFa5-V5NJ?~?BmtLkj}!TdZGclef$FYD8X
zXAgES?=Yh&rO3}uFqIUi(+!+IP5)A^k&DXT4~FFOlCRBTL3f+f4h!;@mE4AnmgCxy
zSJkV_wg$(n-rPIEMmKL`ejP<P7`%+FR-}slUM*534z9@UE|fTiX>DtJ=LYW&Av?2L
z^tBrUwl7GZ8rbdj9Uh%94V|8kHv(?2azz}pCe`$>)F#z>>#R<joYV}6FQ?(|ewW?&
z{C3plq8J#J396#Pu}YoI#9`k`iI?&LmL!%Gd;31U1;{4ZCVyX+N<5Kr36KqOxoMhw
z@TDzbdmcA`$OcoDA}&>4lGqEHF6r9~nl35q1x_s1`q0IblttUBCp(rHzIl!|1Zxve
zn6zn+l~vBQ6LPy}z3*NujQ&~-axzW#^MY^wWCOwV=>bbaacC;<x?6Fi7`ms_9Bc(N
z=s<StFlV3_j}30^E~z1U<rl7X_6FgnWrIoFv~e8PRb;lUeSLSUS7sA&%|@9sv3V=y
zSPh)hftLl?grRXZUYKE`JQ;Ygy_~JzYgy3Gw@x*X?oJK(tQ0A2>HI={h0>}Wrf%wr
z&d_KtZG&mAaEx*uHW$xm=*Qmgc88cKITM#0NhTYJI}g7C^F{s$qkN+*6w+YK((Vyt
zXPXts1@kj+4_!9j)tW3$>#b)|ob)=TQ+7G!bl<pEtA<vLxWgWZ#6td6tkZKA=dwd=
zXe522jw-n6IDt!4Z02QBJ}kd2X>4`xSQOU#th-66FROUCHdKE>vQ*y1V5}8|5p(}l
zVf&+hK1tA{trUtv<Ya^AydYJoo{nJL3l=JD@w~J9w7@59x=C=v$d*dCEx}wOO+LGk
z6BEUI%kwp>NTA<}bx<?jaQYkOQ|7`Nol#a?byY_JnKLe|iB>sm8B3-txk#E!r5y+f
z7NTJESqDUlx!RClo4faNgGhS*zLnm97_(66Nh}&5n|!pTlGLY49Tj*RTY}XHp*fIE
z5D7WYQ29vSfEyE?`V_!yj=C7u%muLGru|@o)#99vwx<-bprv>k&w(+l&!)9x#e#iS
zgg<UPljV6{bkeI>l<ZfPnS#+|NkwAlX>%PF!**$y?1k^H28or40-RsOaW+MBQ1se!
zycNo#q|jTtagIwV#AhgvEzNXY4+Yz8W=&=bzlB-GMV9?icyU3YAuyXa>wXTwX?KOH
zxiNo7lb*1}N@vIQZ2C0C=|4j@3Y;xD+C+?KN9S3yqY_6bDaE*M4yWNVBMEW;VtP?;
zsRK3B(T>IN>%pr~Hd@`TNK=NB<f2tIq7Vf>Y^p9}reU0ry>>o?>K`nBDF>W*U@o~C
zpysV9@&+T|27rutbtyZ$y-m}I+gOIXo11}5_{|h%O(zsqJSc^8=#w}eEwkRvq}xr!
zqR(QptO)2eYvkj0q&w+WGwKModHM@E*>?)nfwrV)%Umu{CM(IlLednnsw3AF$d!e&
zL%nPWf2<b4We*Ea8V>FhhT+raDZ7$1H}+XZY>WAAg<Wz2_<KnuhCD}B=wxk&a0z}Q
zBh0LYn9abXo`>@%4q9qfDPSW*xVjpiWMAH7v`yDc-tcF9nT+yV?Kw)!Ps6Ij#>zIM
zW#`edb4!hZv+>5v(#YD}KM^L88+91hu7(oXHX_+(6#3@e$?w3;j9c~t8WN?qd-f>)
zxH)BGj*$UKHMucJ?%nb}xXpso{IZrJKUzMyp6$1BgZOImr0|Ea=$^W{stmg{+|=nz
zf<eOG>4l$eNG=gU%k_Px%dStA7I#qJ)^Oe}>j!5vqD}!G8=b$ry^SaH>{;>a<(n}N
zN_rNqg-1BAbIu`3GfP>O02uO6Dd1`K=pIrzaC~!)NrOYMwWP`0*7GZg`Pn+(GJ}s>
z_*9YVUtVi>_zc;;nW;}^Ml$70QC#g$TD<$RAH4o>Uow6zTIC7#$@D+^qbQG$fi658
z!|C1aEjwnD9auRR6~)R#e-GE{LE8@;UX_E`b25f-g5|*U6i<)l-8SqKl2_?TkAnNR
z;_sUyCQ3ayAyem>3cz-6IhF@BZi<V%TxvGd_xHq#C4ivw8?KrmJ&c9FajeFC#s4e*
z5t9#ZSKaTP{GWudv2-E=&7S|=4eCFVoPi&!|KkO|_h03E9st<GK%WA;*0?d5rvBt8
zAZ9)wZtB7r&uB(}W8!Y~Pr{m|0C(nz%s=JXWM7u_W)7qkutZ=U7GNBDFvU|`k(5gF
zz7GZX3aBig=V;bCzpeWei`TE_)#s|G(bAr$q0XOryRac48!Z@jSSExZ;g#lR`cM8v
z)4QtchJ^3G&oxIuy&!4oY1(9=TeU~ZlpwS|Ce<DtMvW!e*pDYPhLM6wbN?Xh)00Ce
zwaBRPz7DlQ-}_if{*F~%|KUG2DsAf1+6z;}=l)UiJy@tz6Dg_AsC0j1<$k}8OB)=f
z8gFpfIxcR7jz9C02faNxDQ;xYBrw4?J4OmabZNUctsSj=d?>B(|C{k=R}OcwIQLaB
zlC-PW#R#XExGz##y7YJf?%YGtc1%k0>jfaAE~v~DzPSB%Xou3VO-_9%XA-+qVul^l
zcE{85HEDWPQrmstd(cT~f}SL)wGYahHmWXh$W>7)z%51XK6Da?eI-3kz*@eRBy$OE
zL%u^ryPfo`wDmsUBx~Qn{As?4cMn!Y9&k%hybmjgg6|i#E&WNg^EC<|)(f{gk{HGa
z5WfXF20H=h1)u}9kGIi$9-tQ-0h3;-vSh6K{n_c}RwAEWHQak5tEgqnk~446JI`mo
zrbM4BF6r}2b58!Y(s2R!_t<2?Rvpmr`e_j0tI3550Joh90YZ1qX9q&U00EU2pk3fI
zPvqIaFr$gwVZqqwILx`?%&>|z-m}F}>W^Wyi5Wet;mOShwXB)?0$ohb;Q*O6dN}a#
z8~Oeb1A~iw*hF4x51MHf3|qu6lZPJ&S#}OTl)9W7e#Ef+^0LQF%P-%DHz$B~g`-Pm
zA+nibgHZwg0*)Q3_|)e8;T<{C05?V~STo=Qk(BwxE$~^mg!&cmg-j<ORS3L;lt-ch
zUJGh7)D!0PCWfdT;>*?Fs5*gI85U><of-I8ERP0^^&EPENB3R>CyaY<T)J>&<I1tx
z2d0nR8eWa=_Ld`}4^k{?m3t^nm2W<{dDV+IM3BA}PNUnFE1S%w%~o31=54@`;nrXK
zF;j;;Sy>EokKjbxf6I<8*#u1jJ_tA}#kvHEs(;@tAbLbDX>M8D49{__=o|a1$p8Ev
zXCHe4w;|uD0bd@NmS*>3PG#~wK?Y=>3MQ<)1URnkp_+o{1xa6IXigQia^XarD79#%
zvX6idNTEkMvdETM(pjoefQ9lrD;{eTCX@EFRC=+da^%n&+L$)`)FQifpk@f-p5=Ge
z&Xxxmz@q`Ym#L3-ei(eyID*#({e{E>Ek<rv-D^ud%hOGkXHRvu5`hF0`+@e~(*mFf
zRm0>$VWtq{ifRz=lcqP-+nE$0-Xf{;Z?nK&a5Pc7)O!z`<f2{iB#b|*pWSEpWl|iq
z3fh*%Bt$Xp%8xIj>o7A1vSsW+WgtHb9*U~7;)3$r8==0GSXh^OIbd?8_la>Gd;`+{
z6(Q~Q_@h~=NONIrr&A9p4i*SwFHBaY&3P~s)~2EYG_W)@?<DpT0LHzR+ol=cdZw0;
zDP^R<W<j5kTU7?p)ziU6VvrB?gGm-zHwuSSfJZr+?hQ*Z3H3?;^g*|?&WmU>SP!%U
z&{-7P3;6Q-2NIfbePvoUe_%1X|7&5m@1zbt<3OhF2tzqAi;RdQsOm$OgPI+u$AM=#
zgKm^W%O_W@egEPfbZ9zRxF6FwHj$|V!z+x&cG$Q2X@5n#bbuVN*4DdI<Jp^~)TU=4
z2a_7q;2hbKC`^dL?2nL3rFrijQ90#Hn|N*~EDvc_=XblOQu#1z#ty`nWoDV^r{t7G
z`jmbD=&r|aCU{K`qZNWV-jUTcK}bt_$xj*Q21It_R6WCzQ70h|>@80!p<+uVjSO`V
z+2c#q4KAF0SW+QVxT-aHrUsaSITIKjP-P*Q#L^<(%=d^IlR)Ie3RqQw$J|xR2vh~;
zb|9lU$f{`$Y#<W-;pok02{#Ng`45M)bn(oq#*7=&GC#oxI7n1y&L)MEu>H&<RW$|9
z%v6{ZleBD4&w$-H1$__#yRR{>v#>PcQI$JeaC!1XV&Af6hJze((t8QGFF_1Bg*lN4
zudv4K)w_Gh^H%?mmdTF9aGOR>@-t&ELuQ7SqN<tl<auLaZybZr3#Mz{>Xu9?+79mn
zcE1m-tVt!dgO*=s*rsc00z!AZLK@w(llK7oU*98%g(rH&1StX{n%hTEjC;&WQ&yG9
z=j@ph1BK%0E0Y4TH1nl*^kdR28O}49t46SAIxf!0IiR`Ypc7QS;aZOkrN9|CEf<i;
zHh4{EDgkBz9Va0VbpkrS&!&{meF35-hYmCOrmTOwKmZvVbsaVtahQ^`p2R+s7%2uR
znfN9(Q$tGzO0{Mo70*xDZ}n*1Tzae!LO650*nzIF!O7e>^oT<X8#D?l!atgceDFwl
zGJ_G7TJ#$gGa*7^>K0F!>r9&@*Gff^Q{<w3LaIya#zDo|ds3ulJXi?8GKx@2c=}<M
zOmaL}7G%HFkQ*Y23dxVz-&wSF@4s0Wxa9!p64H(e|Dxh-Fv<N!NrQnEshoei;G(}|
zzW?eNe-X}(Ryzlkw{fFH%Yvy?3RI6WKI*k$k#M*NW}=b<Fhwb7Ckl)-c&?vsI$wfj
zA@XblI<=t?qs4JDGAR5Y`{PpIh)Yw>D=yH|5(Ffx=TfBb*Krf8k?ay{1EX}jA#xPx
z7GfqDE2|@EhcKw6H`F9Z#zv>>e}*&XaD;XUNW@|i%C*Lpewf%)Mv_hnM<a+-l&f@V
zvvI+hB613^)6%D9KN+h_I|NZ#C=L);Sg3gK!lF|$8B!%KXjF~q1;_nIFhwZN;XA}^
z6(YN)pijZdk)iDqU`XwyVhi#gW@uK30}7SYcBvwmWQJt0&`nSpsKKZXM!wLd+6V4!
zY<{4pCjeb>IGYk)<&+^==|o2!Y+YK&8hA45o;Jvfl42OJzB$oCsWxPcHE!Ww!MaJj
zOA+Ionq)X5L@tiEh}7yzN<$?mz>^GFs8mcv)E$iH5!?{2v)^+R%ZmR?>OWHzc`(b!
z|G}4^|1Q&^mOt7rxw*2J#7i4`Nw4(&aVR0bKjF&^Hz-Gdye1-zVY?niY;eudY!eG7
z@<F(tB@-ZDAW}uaRPCWS3G6wu54<h<<T(zFac;+L1xJYPYd$w2<SLj%<|P^1D0J6`
zH89<bs(sNFe1t~7UkQDXZ7xP;&#@rz@LKcdWLP_236d)6cn}dP$_D~~->~Oi)hQ8K
z4E!)94e*SCnARHtw>0R>v$3aYv&seD&np9$PP|N$?(l<ZMw!#uf0`22U<6lM?)ox)
z!s3CICKf0lpLKdb5{a^xs`Qu&&{QxgVRWB1fX?J8U<=C1{}0`lJr;(b6ix<3cHv*8
zwRH2EbU@MF;M4QLdoPM&kslb%v*<*&HD%l>LN|<=QEaj@NSURTQKsgQ$RLvud{@iF
z?=+2K%<?Eo+q6z_F0IzbZLv3mf{DLIWv#+uh6^%(7>M9y`Z!2KU1%<g6YBM;5F$m8
z!pg<@R;+QI0K%dM*j?1+`%C=?)kKQKIn(+Y9r(7v;LAF?%^(!~raZ}Hg%gbwZ;r-2
ztQEJ`^a$M#$Q1@~=zg{ubx%}Dm_WEVkY_gC3U(8-nTON~b@UT3wQ>a<G>1t>pVkoo
z@s<w>Pibo(q69GaGXeYAJ>Ze<W&8kzE6Rh&LaHOKQEzoN-G<;JcS=GExI80!>4~xX
zViI;;@&d&^bQe=c^ANBVWEXzaIyT%45*Q+Odwvqku8WXa@fa`CbC2B*m5ckODR$v}
zy+$rCo601qrCOZuF`Y)np5R<JC!6EPHC(B5SCV3Zn`dX3iGv&qV=Bku2ai*UJk5#(
z?`20{n9vD0w#KU}=GiU{W;rHRxdnqc??`~?;!l!P!i@xHxVHN-K@8liVjWLwqoHBK
z$opli2xrYGn{`vuN)$KL24y%F=LB9Uyb4^OQ(-8hGN#1Ks&0g6wl@KQ$_)xyXJ)S|
z>kF4tNoCEts&13->5dR4e|SQ9jYj_fL{NyjLooEM6jt(jvXU9PH_hG{`%LhK%DGVN
z!Z2DY!D~jm%-B&OagFD6`UoY0jvaqt<%A;6X7=}l_OzJaGNd@lwO)?qepVVJ4l2y&
z@b^k%F*JyFWz*{44MVW8xj{pQ75o-7<$aG+1gv?jlWyp}L#kL(^y1qKYWM*5YE;r}
zJv7IxY&<YpRo;6nuT)gI3E`QPra=<xrev;YnJTlI>Ua*1BfImeXM}@_5=Z2K&pr9<
zR7Eo~++mWG?U4}^OLL8o-RS`RFj3r8u%edK+6o|i0vJpb-6B-5U}h&AqMHrSqzg1v
zY*{Ucx@^jz@W2z#Sbn7<^@V9<1-*Q7K*X|6DftPYfJ*?Rc{Q-IG6N?=@{9wgV)nfm
z%?d{YFgQzh6UllHl*P?P^>2_s6*)FSM$Q-7w2RVL2D0u=F+jV;BY~;&XV4o*qU-C#
zNp*df2-KU*Njo-bmE5hnG}ykO%X;FFfkcYr^d;*K>iA@wK@o!?&sn{p35bM;+)a4U
zR2W>oKPfV3WhQva7`;8+%KHYDm8>3moJqr`a6Xs(hpS?`SMeX=3EC5(j=-MoT@N@Z
zbB@kZL?vN55}|G=CWFP7B>?75O5tus)hW^oLqa`x>LBC9iHsrWqW2A0M0FeFr@uiH
z09ixGmrDfNWwyy>qwWmgOPAL57(mSG$6qQvM|^V_M{>nTM_>*nU<V&~({WJh@Fokw
zoo4L4sP#_aY<`tg{7B5uAcCP#W0u-P)oiE)@Jux*nO(Hs=Dk2OfBCiSAR`?bZc;1L
z8*WAW>{rsujOJV|y!gGUc_{w57Wk3pb8w^H&YG&pM>Pqo#QmnCz|5)WQba%mr)KP|
zk*Eb2H86>#eN8c{vo#bpbeZ_{iiS)nMaCv`7^pN`0*6SlKza63;#5J>i<S3#fYX@s
zQyme=B;$Fo9IdY^{(A$i1|m|?%B4xk+-;21G1g2oyWV^l8J(5wK_`t+i;c@0Tgycz
zan$T5z#WP^y?YskiO|jCpwbJJ1X6Ww+U1-=hgs{8Rq67SW@R`cc|ZG6f~paUYTlBT
zNV^RA(y)rEyfD$!VP3J`VJj<po2-mS4{o4g;6P{d<jclrjkkt5Ja{en<2n5-zQGi9
zOfQY;LR(QDub%+-MoSC#DV^~n4|0?BwNhGjJ5?5~GdK(DodZD~af>w_8Yl~@6~l;t
zN?P79DD%=Uvi?HLTkCx)MptFsI@Orw<oI1FM-*M1Z=l$hh~D3zO2fH!Hhwf()!x8k
z4>19*%A+Wdb&6un!a@;Nttnll^vvBgmRo1gx5}okVf_Y3!|}HviwuiaS$W?Zq%-wj
z{LXBnr(20|kB(BMe6pD?|E?e|M`X5Sy}I8_hMG#m4S@wsHZThg<3>=MW1^RpB3JY~
zI_}>C2OC98IwTT5us)wkG2MEwLFs_vzP^PjFl!?L_&Z_<LoL&9&@p)^ll5T~&_^aD
zlDX24XQ1(eB$oZry07N5j2lxMTN9Sc)Ir^d`vaofjNrT{vX$Nk#|yf3OH%{yvs>dt
zc2WP*pxLY?HU;@6FTv<e{UyPm|6Q3~jtvcz1A$3-D-Q1I$C%kGcjhdTc~TxVL-!+C
zY`OQ0bc0Jrq#Pa;dzFgv88~)+96EDtoR`wE61MZrE*}rUW1QdWj8&G^3K<TZz2X0f
zQ6FpJze6s(Wr<$_B?(vWqVw8axT|*IUJkZ%kgwK7^Kyw!E^dhz&eFCt=`v)>k}XHB
zJoySP$#7tM9dz6f3U%R`BVI)tTLVj#DZkvv!Bwi&>~gAnsy1lUq*;qrZG5`vu7{p_
z1;-w9&2!ayUzl&XCAK+aZ>X@?Tjp33BB`;=1`FNxUPxoR!+!OP-!47nsrJx)$Mx37
zN_{=n&jXJ<_rz1re5t>ez2HSB4Df?hUh|4q4fM5dEi%|3Lku(2a664KivL4njQhEa
zH^EmXnrxCOrkU!3U1peWrdhu6olE>qy6g>ah79Q_fc~(UWHOMEOkoa71jQd@_^XHz
zr<``qd3T&~)?Eu6iBQ*EkFW^0G^`O3wulV-JmBg{BL}8^IWk>6?%F}Ashv_^Xr95&
z>8T7Q@A>Jesi~>7RC+2Sm6^&)Wv6mdxz<@)r)*m4lVhgo`d=|%+~|INdyN_RX?MD}
zIm^qtUN?Hg)AVw>AoO+~f!b7hD&wAIUZDR^`$5Z;J*lZa{Up)yGfsN{hePW9CRUkW
zqQ31slNIR~-Dfw$Kjf9@<zD<;UO6R9PxmjV8ssh%2UROjJt(%JW>DP;wS$^m)D3D=
zFkn!ZjQZUH5k>NE=nRI$O9;H9!47&#@LuoP#+004$j2@%e#s1U)+29E2LJ#7&j)`t

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-book-italic.woff b/0.16/_static/fonts/FreightSans/freight-sans-book-italic.woff
new file mode 100644
index 0000000000000000000000000000000000000000..a50e5038a40569ac50f2cee39a0751cb26037eca
GIT binary patch
literal 33944
zcmY&<1CS<7*yKC5?cK3$duGSBZQHhO+qP|E$DSSAy!q~ri@U3ctfwnG^Qo%ti0-b4
zYFAlNQ2-F&rv#+|P<{ds2l4;3|KtAuA}XpR{Ua#*VR8S1Z|Cc|n6QZG4>$DF7Wjb@
z2sKE%n4GL4000aL0KmWl05lpb{rJ6Nipqij037CzmiOnJ)Dm9*xU3=_BLD!e_T$s~
zfgYKF^^~EtzTFSk^rQ3t!NO6yAgG~>(=Px3k^M&_{2yRm5WGz6Os#)7uOChQPd;PC
zCc08nea9a!GX0O{r|<s_L{lqwlOOKKs~!meG<Xdl<jj~E>l*<8mj8{J^*<n_)nc3d
zkU!kVPy6=|h~Xo_0A|)sZUBJQub=!%002mt!54>zm962A&c^Ad-^m}WjPhi}SnIp}
z^lQie<NMJ9egQ*(df4b&8~<>6005FN03ga4Q!G1TXY2Um^WXjPUH|0VWXI-sv~w{2
z@dnQQ`1*c$nNovqOjl!rpD_o;{czy_VckNr@+SiQ0I+<O^!Q(#e_x0FFRso<oVu;!
zMM*;Y(W4G6+~sL-r{z|3f9vG8(u+eHF~CBpHJO(rkaPYqF==Lla)IysW9P8e3YtCY
zgLQTAy$dqs41WBR$8CK%!_rR46w7bt!iIG`!466ilK7B+z}g&SPvJ^T6Z_ZZDe{$M
zag&H8R(#Bi+DViHlp2kanKG#eCOKiXUy{nSoOLKI-y0c|jOj1&WP7|44O8Ej!8L*}
z_1tT&)NYcx6zQn<8ONEN>o=e4_nhnZnX7n0B&ac9bs7<_oVOLWA|=v@?|Nw%vB{y)
z!D`%0jGCi1gJp-9J!LBPmXK6?)IX#m%d{R}YZyvA^)*9e_%pm%Y95UrwP?)>N}PdI
z@obSQrQ@v556qq}gHTk{j>Gr${|Iz$SKLkAQUen3w!OFSw>P{h@4ps3H`wMh|5;LM
zTxy<f>~ZaP?!FH_)0Ix|XAa-)x9xxkd~5ZN_V#}}e?PN(C}BG+`i|`QyM5%o9FXKB
zob9`fAHr6vc)FB&f3uT3(6xwClrOgJFWqaC<3-U@gqMO^Hx-rUi#Y#jb``bXbOrLf
zpsiL+sgFUI$jbQJpfxjIc!TyAx^b9lRlJwfhumL*JP-iu=TPS7F9#*`8*(rYIz)&i
z$}e6Hq*<@e3=`yW4xk%AH&}RsN$p40zi^zv4Ysjs?gpL{Tsz=+!+qgrI{<To=Z2&c
z&~2yDL*sk+Vfq3yv|F>H9qn$`pcpU?UoJ5+&rZIaB}Uw7MEozqNYH|#bfZ{l6*_?l
zwnC4+D3yGs{>U6^t$fg&qqDhCAwJ5W-*4%>CuY4}?8t`EV6`&o6zfr<C@Q52R}s6x
z$>|KVQyR|LTu#(FTPRW@FvMn5F??r1adJp(EeTzq)24*<2&<S9Q?}UZy5-Qk=9(15
z#@1xTt=h$)rDVQ};pHSBwmTbX#k9HRV%dDvy7XF2rlz6ac&j1Vb8V)(iU#$ZLG?6S
zg@DT_)sDKjsk7p}b)=SXuR{B*;)-N%y@Bie&`~Vsy|;%b`FhZ0Gvt&)N>-n3c})EA
zbm;;0LP~lP<x++DT+OysvT4zFF1n~~>+C5@MR~%7%eeE(yVHBMNCD4Q-J|#_1Gn0l
zx=7&@>#@P%rck-GawBDbOM_&|dTk2*OEuoHW|nX?@>t&kdeNgT6U)u^4w);=Vxn9-
zkl<D?5V{6}VVe<)#93>3ap0G^=mx0;56$G<DTjKn3fJGs3*kz7>~8Yvllz1!p~8cP
zoHkC8ZS(d3o9jQJxcfFel(l1_WBN{=v|@)LoAHKbxQQqyo~kkvrgt2V5nbOa*vWYt
zoU=?h-rJf@o$xdx2H&D-!(I#!zB42V$4DUeiKfN`r|)}!rhWl-(9*F^`1f*OY4AZr
zOfYA9behp|QY-`%J;ES8WNuIdAtI>|iaAh4eQc>vtQ*LwAa#12<1xZ?nESd}uNhMF
z>AyC=ThxZ!8JhZ#Yf>F_M%+27;;ttoW-9eqPGwKnw<s<1&{hUG)evy>h^zvp5%aIm
zXSlEyd1L<10Cq|rCT~Dqes=+f?m>1z^92;z!RiFA?v37{yinkVy10SJ^v&)PXBjA&
zp1e!5!|<)nH!J%9{oA8)!|{Q_*d^35i0TcPyn)yHeW~q+-~-p|zato`1JN?^e#7;`
z;{_mK+_t<y`V?#LiGgPmZ->&hgRvPbwS(X-jP32;*>fb1udmX4*W<(WV`Ya7)uXZ3
zN5mktNBT`GwLd`oQxOVkr40pBL?>lUxkEGeE2W8kIn6M!^9{I@P6e#dLmUrqry)q7
zFPtH(fm7*`R0Ys~w*E|5wmq7YQs`y9=m@a~e%};?59ejAZx-B#^`_B3mgwd0&?qS{
z>ur5_thATaz99-h=KoMJy9bcG|F`;o6<9B$eSF|w3^%d<zOWyLyRnEW@O!b4U-UPj
z;nPBYxr6?zhBg@^J)Pm7tDJXS9Clp7|EyEYtmcbfVyAghrTJo~G362)XXL<ePZ>8)
zN!>f2vKJprnW<=2WRsW6D-<r>&`VyT#nwwNS3O2?#+s(wpeBot0Q5`O%X^#23MJ}_
z^^fA%@J{a$3hQSzBHe6yCQd<kIxc$lB_JT!*!B*KB+M%*!_N%|#_9f&ojl<?TgNZP
zHo);-65|JKHOq@{m{+pT6JpUoXmc4qrc!R<=twP^#a7GDMU#IR{u^-`YCS*nyX;Oo
z5HZ%>NSrrPjk!1@rIEgiwu{O4d}W_~fM`_N_sKwK5AZxAlaS4hc}zOm>ZCawai5~$
zmGi7ulpNq4$!Y*wET>t^53)s?sW%F9lx!Hj1I0FeT@NQP4OwU(37iddf1xM;{8&`A
zUqUbyXA@26^iUfH4wKO56`agN`z}`&wmghFwz6<`(u`=k*Ws=$JN0FaC{Y3HG-bK+
z<H0%6hDH`|)acW8^dxHtSYs;c*zt_8;&}%3c!lpcyka73NflbBgltUslVKj2QCM|n
zlA=X8P-Uwx(QIx%Gt-*F%qZOaJZw_o`C67|pJccC<zjl-dQFcepS_Y5?bHx*Jz!%I
zx*AXF0oL%;KUV40s^K^Tb+N1EYFt$M#6OQ4UDdg^EdFe<rH!Y7K_XuqZ&M%X$xV_X
zBhOu0w788zw&@svJmbtw#Iu*umDKni!BJW6Ev4p~nw@!7&HvPq^4*bi-6OQ~sbX~9
zpsCs@$zOYCJ<hX^g13&1WS~0>T~souZcTLV3Cg9hy?ONJ-09g(Mbtg0z4!cgCDQcj
zW9!7_7^LvMnNOM5`qKBtH1A{O(eGh07RR6GBK&<Hi<S@1^IV;nZU^j6GZeZ7C%g?K
z-m{j?V;d4(t5i>>DD(9!2TiOS^?52hpSufv%x!;aGaB^GtaT&yy18d1Oh@op%NiET
zf0xaYF7dWgn%gjjPR@w$1`{Os`A^OP<ij2;R+p>|B*W$_IZp#}i|)HjA-E@ko>_Fk
z)xBa#v>@2?`X`~{Daynr*whO7FVX&gg+9m4UA`btPufQJRmIOwPp#~-N*1vM;~M&&
zUJ(;s3FJdQ4d3QPJjhMm$V=FVA0g|jmgc_8V{5Vb@e!M5PZsCAwj++J@!JtM5Mje=
zwPvWNCIkzo+#BU%u8WJg07H+ND0bITjI(wkvSI6Lf=x>iM)9v-iAD9J%csBH*$*^6
zUCXvZ$kBUyXyvb)5T!wiv?$$J35WxY!;~iE3>?yw#&4q5moJ({9ctn=u#8KW+bfsN
zVhoFPYD*1`v!YU=&bDUWeXzgJ^DCbPy<6A`YViY!fRkI4);dayLgbhJ5}Nk{Vb&DY
z8ly1tPN*C)q==?dD;ACOR(J=BrRc2+{B@vL2=We{TOcn2I$zdbtk)R%J*{TV?JX@B
z;cV%nVV)4oh-Myd9;RyYtC$f5i=r{kZ%TB+2ws8KF&>B!p*-bPJot;i=a)5G;V@y5
zC^}K0SuL0>tpNI%LXk`gQKS)AgPID_!6B4USJ314qh%b_LPS(b^Vq>#buvQ(J{BQP
zTz>;nd6Kkvx0PrBcf4p=Da{!DS~%)BFUOFZuBg`?_5xw;r2WWpjN1_$$GDozZ!`zd
znZkQb@#roZjQOX8uD}r02j9nQ!h^>Rwh*89k+|n05GT3o-7?qoB^+sj-#X#z-)p+1
z(>M+&cpY)b?ZIw`tO|s=g!t`&N~h0G=yu8?*FrZ19*Sd;GbyT4Rhv=h69JO-!xBY6
zFxB}(8dePWI9C1l7EOs*m-+^Km_%8nJJ=e1WwZ*4(fP`jy(*Tyf6lwrd$~{rCi%uy
zs;5`KKtwKc82<Ync9xrz7?eZgc`jK|C=#y{5ya>ZM#a!Z#UP|bT@+*>K|U+#%p(_!
zPH3rogYxkw_~jF&yEia8!&gSOC@*|rFvES7&cuWP&W{DcJph|%AAyWbjt{@ii!ElK
zi`gU_-c3$CA|LxVHrD$BxDhMmuk~(ToLGkDbR>qlLrC@BvT;Dkt_)}UOr2|Db}2ii
zw2jyaWUkh{f1dm_E=)j*X8TQDxOG!qEP(87A;0w@-DRuN7f~dVjN9@^gc6=tdb#PC
z1!a!SoTMVjI3O0mp(E^umUjq8nXn64XsueZQZlyK;3$;u-!v9FRh10Gh%jxr0bVnh
za%jDn<0C}M(Keudvw1`vQRWf>qIVmEBG9!x>Z}W9-BNhe>^Q8tVemaPes(KQ`C&R9
zO3_&DD}5NzbqB{6XpECZ_P^_AsAsZ0%<k`>>>s?Gc69Om4%^7jJP2FT2%8}l0t+hz
zTV+KJ3!4Q?4TFif0zqHr#>DU6i4&)1B=<5sFmi)1X4E@tRMgi;k3~G{7Ez8UPC;J>
z07#o4B>k6LPtU|p4+~rr9FZO5kRjle8WzW|2|N%uZn$4=^d^@ftv=usa}f`W77PiD
z0*nKU1dIjD8;mO0Etn-jqQcjw@#7m|DZj7Ij{&kGK|p|CKu`d@Vgb1(;r9FcJNBF4
zP4{N!-}QHIEk3TV-AVr!#5?|lPu)R(2*e}-Gar4g@5zefN~xSG5*X$h=I#Vmx*sS?
z2XLOMy4u=b4)*pocemFUFwo#25#iw>65`?_W22)(G}PoIWu@gM7Ut$AXQ#&pIN0bY
zX{qTc8tUpQYpcr(JlyOoZLO_M9`5cgZ?De}5a7T7A;G}`BErG~LxY2T6lARRb*@gb
z*|b)-`2=Ih)LN}>i|O^pb5^U(F6YS>yp89J?atVIauSw%o8f_g3o>JMSiDBe2rSHn
z$^M$ja(%V>N(K&up_%IL)TJfP&$+$IRhz9IA49Uiej?ad0&cITto*xUypWWf<L#O+
zD3tdS8N)P2(<`fx!cl1dj0!(YK!jwN&wmR~V2}#XFMvB>9H0R(0GI(R0k!}~fHOc2
z5D4f5cmsR^egOZU<6uB2AROQcum`vTbOA;HWq=95`e(f_42I;kt}WjwqbY#61&09Q
zKlJw($2w|bdfbR-J{Z){a<qXFSMU*}Ih#X#q&*M;zrQftdM#F7evx=4_-_nxFG1el
z$V`@D1<YJrjgB+vOp+7v6|Se2vx@(S;|8|Cr&&K`#G4xd&qe3w(NRr-5E@n{4-?Ms
zUohaA^GRG5fFmAT3Ir@^E`N=t(y#ds=#cjhf_pB^Cg#P5B~GZ*-=N2t_L70uSkoPo
z2yX$)dj|CMMQJ7bw0tEBU#-mlEGFxk<LA2qnUG$`nfVnhtwA%smVPha+oMXxO_tQN
zrlmD}5u@}6k2@(VHc!2*_iHyap6CTqPWwar9aM3L#fL{AMO|nmo}h+9dNQaSf<#Kh
zh2}Wg9RDife-%d_>G~tVbr8}FJ`%s$m&YpG_|2UuZEcG;@S={}@7jwc$@wxCvW|&l
zz-2(;2PG^hs{9&)J=;wUeyH$1%b(<TbQ|lN4cT<{SC~7s*%_We1=7Q3WAkn@qEn@P
zCQ{JVw!)%pfnxeyXQpQqBnKa6P{6Q1bN2m9!vBq%x`x>d9#qJK&EvbE`7Erm3xl_T
zY_Ydq<G=e31OWQ}1|YbtUwA`#p+2tNe3+X2tmU(<STh4j)EhF5)`*EWTeU>{!u%i|
zpo9D{kqo24fPm!XgoV%iQLASzz?<AqV^=gRtKTA;RjMDGS-mxN(Jb2;UvvKPHD<zr
z_P1Q$diNqsUtCx<-1}bLf5Q$k5@Cg8*LFnhVZterB9p4&EZ(^L!i8HbUt1XB!jwjh
z?HQx##F#H_Gc_{=tPZl)PG2{>i#50_0IiYGC}_K!-Nt|5z<QA^{=LVVEStdMF*Kn}
zON-G5CLDD|>PJQ$W01GLLS9mt7YEKJWmBAj5T=3(voL6k1b1YffTgUhD1#2&q%B=f
z&HclzWjNipVD{M8S7+VHM{8A9jfaRS@nje9C`BZ6jj<Q7DX#1{uxAWPN`A8^Gr+{j
z%MWTTw-gTm*<HH`pyVvu6yPR?<wI()Yi`<SYH*8g%7RH?Vo~7XnY+GluQf&bE|?fK
zVsMwjBB~RZ&qjie%&L%dG6@w7;3Qvn+zSAw#jMUpMCuY?Ah2OB(>l0O4%r($hiv3!
zx6_YC@af8$-8%|f-J^N^sCT`74>|OydjUzIe$(5#ermMcdPbxAe9x_U)mr}~+b3MT
zNO8F)^i7u*oguZAIC^2y^ss3;B=(KZx^M$d0x@&+bi2}A_LM@>b0=>M33o?YL=sZ1
zLX<HuLEjGE%?8;<(pQt4BS881UH^BT1XXj`g5M*iSP~b$cGx46X9NvZv`{{OomFWx
zv)cSrN4Kt_4+YKPf+2Gdo`AuC+@O(f+Rj529%`xIG;rcPeB*%pfQsr&oc%C7h{Yzh
zq%Lf7kNh!$xko&I@bqcPzDH2EXMGPow1ScfI+AiJW3%gsi=aA<n<LBf^S-iyG&YN~
zq`BX7t=e-$i-*VV?RBC*Rm$@3?Ztxe+vXmSf`Zd05ehVnjgQChMEUT!=0ET;RdbL~
z*jywHXT6H6<!6|X8iIeR0qsI1G&t@za9g>uU~QeFG?1pcuZ=E`Z%p!!A(o}pRljtb
z=_z-o*^~(?0{nPj5~0#ym^9Nc^_UQ6O+SwPFy)+~jszGmbfg64f#^;Jh#8S6{i<V}
zHF%;j5g=76Pr$#H7hP|uyESo6s8jeBG=~{q$a8y8sDL_csOnn&phQO)5Br(!&lZnS
zL&Tsq-t9SbfXqC2q27ddjXyT{>JOFY!Pbe*ICje$>o}fejyzv?9;81%+i)^hdVe|d
zwK>n9bzj6iKW*B-v?5r=ux)!xXF#w`>3#-kDPNYWwz)oeeVmj)@UHh-_roDntGgW(
zn)*EUUGEND-wX^|VpMr1<$ok^-KSTGC&M8l8daoQrm4a2{uzUe1nZ#Hx07Qw1EOIO
z@aG`)kk$@ZrL+KA7kHx8k4px&X>kc}zR}<qGvTw6J3>tY#T}D3(VR+7`4dmy2H-Vl
zI&PZcPh(6fEQpCtikARWVS>R0W4AYBXZq6z*DJ4PbR1U?ZG>uspu5a#|M3uuzxm;1
zbTy&Rb(|2zqp*>g`Aq%oeO{y6Y^%-eHatL4i~}F7#?g6}NM(X3Hd*!R8h3LT?HJN7
z*ebpKF*3*oYim>HGWGg9olH6&-Yx8MBwNh4uCCB%KWGjjQfoOewGKtBI%XzX*X7b0
zRw#`e!{}1!-(mj9p9&oO`T6mT+w8J}HvpPU615IRwcHZy*Qd$&j8FdhfEy-+P!Mwk
zW^@iML-tqrmIVYRk4cI%ha-DzNz?|n%km7KAp{$4o2koZz*F=fMC?|5VLB`c$IzXM
zS9NdwLq9?hg6q!9(S0oIYYn*?j&I~|9)dxhkkV}-Z>c7->;C0iG-EA!;?PD-b^b=z
zA?wTg^RwgD9o7Pmi!CnIXspF9!%_BA4c$c|IxqMA$z1M+iPe)(x|E)TuV3C^Y&87r
zHTs$CrIdJAasxxmdG?&u+=Y7}3^^jj2e1*eOFi6S{8Gg$qVc16U*hsZNG@*7_{k%+
z_GY*u6_(;q?*xuuj~U0gH^gzI-X+#%rppOK&CF-MoRXYE4xv7ktgW-ElGm7jNPH2m
zc@WC#VmJCQ(=;j`S@h!prWJw=AtxZW`B6F;VMySzP^E#(nB~w$LIpF6Bsah3i5JQC
zoBet2-K=>xxLMjgNYV>au5|yv*Zi6+!lNp(E`mDpe{d47Wi(Ao#iBYe1+}bGteo3K
zM{P4K=-(n)_g{m<#l*HwGQj-X28mWUHD+mQtx<u)t9KK${yJIVnKpO)b`~*k^RV78
z-8f<b!hJBV&zemGmNOydCOqK0hrpOT<|ncDW~W@d%97Ld2fkK2vaRmPYs6~|7hEgp
zL1Pnj(Xz+@x+L{$%uh_=A-uVT88I4M!m<=L58Q>k?>rv8-nQatzylLrxGpl2_j*de
zJig>?h^}lad*z<wxqV`y$RFK1D4gTgq`jji4INIlTxKP`Zxt&Km$AwK$pX|9!vZ{?
z@<4p+sg{kg#xDm$JY9V^CPIJXPyR<4f0ccTU5p=M%ZSjPgV--{+k4g{>$~!4R!?T|
z)!kqYVYM1=RNBICanoxYeKT;RNJjob&gL?G3CF!82BL#CsxzfM#C1}pf0J~EmHch|
znPrp$h+eNIJ{i^IoQC7si?lml=E;j(#?$f2skSy@RdfHujuCL$Jsh+JGE{PPS$x~P
zWLkE54)Q&@(3e)UozJI<YCYZ&3>>#MHTiNmyl+X8o$qZo+MLbvoZgUMy>RBjrog#P
zDx>s*Qy3WVohdM4>Xl@c%(R4O7SRse4pn28!Gxou$nQ5X2t<X`!B!~a3JWAW74%~8
z%eh=Wc)F+=DXaT5!ItIejf8{$%IP4X@S7mt2A&I;Y>XwAu6U27ve9uq!u5i?jewkU
zz&2ze+zDCdrG;(-E6kQboV^AFT&0~bIERDrbH>8ooJ@-)Vc%OlvDd5dC>e$lIa{d1
zSdusXL+q>7I9PUzg6p;1@Ln^zei~LJz&p-<xu~jqrkkP7UbFkRo=Vp64trS`WVvoi
zqlu~RwKX?U+&nmP-9g%5c06EiS-YtndXBSj&!MgBr#>+xVdve|LzNP8?s(F{DSe#4
ztO$UmqIUiz{>?Ct5Ov%Nt?oW*Re&}X$3kubG?=Uarvz*;E)14PBg*XK<TSNw3(Ouc
zNZQe`%Q8*dRxXw(RAx>C;#n9z1}>_oR16p~KY>{<_+4z6W6}00evm>XNwzkoYpd%v
zyZr$&tJUsmu6OZ4(!4RWqPt^vw84})(7drCtBt=z+i{4XI-T2lvVOL4yhNQLr)ku%
zy}11_wfv}*0?cFD=Bn&*lP*1FLRunnH5b&s7NW|!S<Yq#)i;6Ff3$SYALe}GkP#U0
zlvW0mY$I=UmVMA0r!HcNCMBjEFhW@;c90d~47zW@Jpf^!y#i_*?RJRSuOtdzpQT~S
zoJ?{UiUj8oLBp)m@D1H#IRw*NyHocqa$YG^i?h<*8~9Nkc=7Kzm!nOKQ)-0HG#4U6
z#Dn*USA{_jtv2-&EFL67J<i0XrgG7EtR8?%E1kNV$IG-{BxQ$kXAB65db8vL{-rhn
zDweKSAMw<bXwt8bHf>IZ)vqPGZxXo`ALj@u>tNJ??N?PIWuenYl6*Q!xP_~8(^Y9y
z{%_u&+~Wd5GdFTZU44U_>YToF?C+<w<Wtbzc?QnFLbxiOui^UoG027}EmlI}m8of~
zIMtGQeEDju@6ijQXt*`+KOLntfq$>*lj&(Vzo&q+OGP1B`jS(;TCsmM<=HBUC);ph
zW;#`-!7v&@_~I<<*>I)_MQ-qG=Y3EkQgeaen$?Gc6E(-7EqztGH;*LN5fs&~mddlu
zxF52St>Zrb(e+gs1R>*nNkpsNdfB{Q*=jv~BCyeXAH0;g8)Gea9o}wVNsc+^TN{S2
z@G)O~W7F)2b>jZ^8eNBrw4AfC(#O<6cu)vL&*KE_^ZEseG!{Y?2tlx$B+pSu`9X0T
z!Mr;^|7ETTyJo$`S()bJA)p@mn?rjg6sL@!I3+CCD{mQBu3RNRQea3gB8DAAV6Gx>
z3>ij}LYqHC88T;~(YenmBra)){F{t?9iS`^E6;C}yW%h-<@g&;46~kb(S&cP47Oz;
zZ=XR$1RCY5AR~fIt%)&#UnRX_ObYizET2b|hvjLlbLZu3cW@Mci`n*aCs)JX&H3Dr
zt(E&?qMRD*=z05LaAkiY<Saqib*s7SltH2CXv}XJ!ll>eCWqBt&S`QP;mc|%rg?I;
ziRKZW0w?L9RrBuJBpZZOS;XN2Of!YL%RF$3?A)UVz+uN|Gcv06`~J^VP}&V}tQq4b
z=`Io+VKKlPc^-t{vx+cCxd3fwE!NM|PmGx~<}Pf6rqMkFZi>S$0%ISH%0Z$iA{<qI
zyR52C5jKzRAO9SYgj?Me-&i%1vSCn2Xb#`wM`LgH=h(Y%ukm2ZIYH-SyXE4@sOb^n
z<+VSXkHfY6K(3chr_<eIw`0}jM6-3IrHRYOi@P6~7tF#K9K_DRaMbA|?Oc0fp6#B@
z3%Y$_1`zam-{1`ZR20^Z9gkYo8f99>E%+H70`6AP8{yY49_d@bl~tMA@*MR<JJG+U
z5*C({;4AXl;Xat!vB~3l7MQSeS#42MK{%v;=KyW`XzF#)nHRDCA&7XVb>M95;O6sR
zAC24254TI(iK`ARgCm`utHNm}BeTu+@0f6d?3BL)LccRO3|+Z4B76=<9v3WWPn(??
zZJQW(>~Q|Fne`9~Jw5bUgP9-9n<^lqmyPnT@Ow5}tl2i3%Gg_2w|-?NQ;ok%+x0rz
z^}@gU#7hKxx3AP!Z(U~1_r?fo5jW^qr~^G)oWDkW!WyviC{9eJUP2dZYsgKBV$s1s
zhZtgZtO(*}TI&4*9r=)|S9bCnP5-Aq;x~#lgz67u)sr$g2AYn7XMreSsnODF2S6KV
z?yFB@i(@1vsS;F`GYB)R2qzl%n}J0>lYfWau#kXQSpo-N0f!lp!tXMW7a81i*noAl
zf9QXfpzEgYuo_7=Tx%W!h8qlNcxtdu273w^M;Tb0SX{PT&TfrWW3Cv=(N$zQq&%mN
zSWV~Y^57iPVqwZ~j|nfQ$Hvk#qWn!B3(6&Qmrg={ETnWTj9#wj3PluPKpOIFXtt|D
zI(lQQGSSD9*KS!w9xYKs7o&V}v3IcCkd0@77J`T7$U+vJRVc+x=6XVZ7E3x?#&%W<
zgjdqQ&IA!eNrbAOf=F*HqJ%3C5w^sH@0*SXav+r%M^pJ-5nvbQx`CwYH%`ccX2&r)
zQ0{q&g8)>J4WSO<b&OHxL-`CJ=fRZTer3Jy>qf@VJI`pPJ{lB~5>*n@BBt`)<o-2q
ze(X7-?eiI{g8zDUoLW?x>lp;=<uWb$ewn$Qw9iNCYt!Q}j>bgBCcVKJ?)8dD;r@Nh
z5&KO4y!|q%Il>&Z`Lf@#va_wFO3|*e{qm2^=H2T#YPFb5*F(kQ(r$dLZ+7fqG<B;!
zEG(6lC(OYE)JhVH4k-34opgBc@WrlHwQz-TcNwyyuI!lJhA2SbDX{;Dc_NM1<HME`
zYjJn~tiNz?t`O;ArA=Zs5TkIA$c@3HkE3$o6a{}O&w&7n%L~_Jvu*xjwUgJf_T1+T
z&VY<8!~K+5@>tSy4B|p($}uX`IVAv&Iq>n0Q8PJNqm8yhZXU)*G6O=DvLs(KivSfj
z&s0nM)+!KWDSKj><n8TE7g&+hZ4z{ZQlpBDfkAM0*dW@IdRJ(8@n#IqAZ$4CSlYgW
z8(j|?guQ@uNmXa~n6s<gW%|1KiI1NUu?^9K^$S$}VY3q9Qp;ZPnd0=C__=)a+4_~U
z+w!ZYlFj-{X?Td!#*L8A^WQQ;&&oo}UBzQ%g^F~>mZ0-WGf(cVt>5&FUI=|Xv#85_
z+21gaz$!U`TFoq_SjOG({yQys0ozCh4Kj~8%Fet~W7SyjS9nZh_c;(&ir99Sz1Z-B
zd<@_CuZ37PFR#@lMO(f%{^KB3x<naCy1KPzlxIygDRn-_c}3r@(@)!ys@v2AO3G33
zSW`cI(Yl;yXcj3bi#B;n_kaLAT-~s?Mbx2Q&yOw8{Vmj6(lF93KevrPwD3)iWc3%U
zYlp@}CXRaTRT<A{!9rk96A=M~1bU0^_hPX#>CaUUGcdK^?}<aVWlm@Jc$6lam$o|&
zwN7Ukm6|q0Pye<*lZx4gnqz{SvD1bpB%H&O@A}v%C-!zP<`8iYY&zBvVhCibQc5eA
z)n$MKBSCeX`qnv(tZDG$rQMQMRwqfF_I4L1_zq<4yGlJjrC52oJ1gU@9V^q&%kMZT
zHV24TF+v_s)7tOe_}4vBOO@x3%#lhxB&_kmas<?m-beXTRm1hksw8(~pVT#QRabTz
zMsoFf9b9d5hwo0eOjF9!oD*!VpEJGPXhYlM!;R&0m>%x5N;eXVfurNlMLA(;=OEkv
zkWMM4Jf7Whoiu0}oAkRYk(8wW5)|i~2%B+Q>XK2_@UZpTwKsgQ4JX_vQ1XVDGzc1`
zuzH8J=ayP~YKgthSonN;#2Jy4ElPA%W$s#bW4-rTdHZTbYA~^WmW2q4_otXq2AV*5
zN6pnJWC=p+#xPX2V81?mfZ7_xqQ^=WpU3<a!U7=kd)41dN=~f6W?xogcqFQj)~4`!
zU@6!a-?EMM6|gs`8{r5<KEBmJqZH@fcjnRgw~w1!K|B~kA4&pEEgi)--xKiZB=R2q
zbbJ++Oj+^{=JH}5g17Mr+sBLq@Kg?7Z_wEzqAedyt|}c97@H<ZjfqqizTdI4#y#ym
z8bY}&RhlvvovG|zxyNEd>DdU%Y4F#8l8Gt%1zI9c3mT5?X`A$ZaA0VZnK#k|0p70h
zi|>vwUyOvAd5pJlofPqxyF6%E7Svu9?jJC=%};qHv-2t6&(LUDY26Lsh4yJv6PViN
z$BAheC5Wy3vF4RBIXNR`eYKSE0V~h}OwaLRsPN++Ju(^C8%z6{F#k57SO)edlV(ub
z2pCbojd(XWI+Z~e{+6#s*MgYwVelU{F|VOeP|eAPTHFg(B=gv$nHsQ+OOx7Dn@bzR
zf3DYm`ueBA*_Ecsg1}TBP)M>eNKfK4BpS-ER!4J-vxKXF9Z{KNHsxiE+IV3>G)g}a
ztR%7H?6x?ublVXHu_=0T6x?*G8A9HN(=rXD0;^yZviVn?b#XkCqp^I1=y7@O3jSqk
zbX6d$afq2Qn~4NM9BTVMh=2O^LNVgVfy<3=_A%JYRI<s@UC`@uB;SdexXT-YN}VB-
zFVE+(M*V@$4SazmPECTmuEUA#YTg^Zg5&b>0sgf)OvgEW(<sB&FU#pIH)1}^RzX;y
zxXQ=?H!xA={720ly0Sp}_sX)m2)>@Ox+mH>K`&MsbR~T1j%Tu1RwZb%vK$CAVW<@z
z+3wdfl4gK{WZSLphFFhx>En=h!W*92g^tg5x$VnQ>2m>|_1UXbvrICSk2E?{LO2%6
zI=bS$0cS?hMUca=R3wf{Iaq`q;#teDYpnR^g-g2%<D@3gj^f?qhAz5Xdv>ffE5?&B
zRy@K3pPt+k7yOi*uXN&1HxT_Dx3n_Oq8wcw>g+2$D&60yJ?)Bb$G54)qdTOjh=W`U
zsiER;Am5@=_NTH?odlp=8fnZX-37+f&?<|48tK(b8tL;I#gkO6i^*K_LID^l^QsT1
zgU0pH=3)mXlGtH6Xt8>2Q!7oNV(iAXFaem<i{@GLsH^;iPsb3D|8B@-FxVlIB?)vH
z-TuL!TBdk6+P|LneoUiuaw$t)24s$YlsuA}LuwA<cX&Q(FU6WCJb`>IV&RC77E4V9
zyttCw<ObIm;Rg@EzqD%BQ-ic#BX{3Kh{xWGM&cI&M@QGp8s|&!k2Qp6pgPdAk`>CE
zhjs8pM0r<dM#X=9AerY7E0?ShpPu@RF<z=JFeN-BH(6YSlNf>e>a4r@v#O9Lu9cY~
z7qax<r}N`QMk|a4cqNPMH3SGYO~j+VjjcgD`*?oPQ7qMK1Wy&(N3x)l84eQ3jT%f!
z$&CgIGsuZHsVBh$!6qtb88VduiRlh<5ISoGNDO~Jo%%7T($sw2YhUN4w?3sNvejZf
z1-=_l9xrDTtxRDu^v!)zShh$|Z%j4zAW@AnZv{A90&zWn%~6FdQ5L_N<fY&MO^KF)
z%hxg?xY5e4=z{Q<OvYQdBM#r1xVq?5L|-+yXLcw4eu9mP0c({OzfM#=j+kKQNOSaL
zPBT{N%;KmLD^JJpxHIv#dmJHSf{@9gcYF9)3)Jvwuvptj>^kuEK6;Erm+H2fYO}0?
zb>Z{fBEV=~kv&t>U?)VSm6$D0?`ZQ`_Ah_Te?%+Pz(By|DPmsbI&3&1@_nhaS2A@5
z{^S?nr)JFwFPCpng1*G8kvwEclUY#^sst9Qlni0wD=h+^uHcBk4u>%8vqZKSL%LmP
z37K)yxr{&KHGdD<sk7NSb^ZYNn0?$mckbuWE+eal5?Bhn%zqVW1WPW;Vps6Ofkzw%
zvFV2c&-5#Z&Js$!UBA_CC}1;G!2w%`e%dw?aSRu;pq=JMpJ(VKb{M~K76Dx<%6M~C
zen3_1JJ04?zjBB`xOVG*4{RZTjqBL?a7IYy(#P6Zw(hmK8Nm}HIIx(({o38o1;B=4
zsTh~M&f7C?8>CfXK!u*K&OM^UsaGRce|_0jqI6Y*Ps`A&8TYyQv#656Fv2sl54pMx
z%KnYk>}Yfi*uL6b%`rhTE%g&VFUPY^VPSnccbmlAo41uxli!!fm?;dD@kC~0A+_;1
zJlRZSU2{5{n9~x3bx;U?Fzf>nL5RaOaEDBbE{_M(#?r<W^D1R!txJEdCDC<t;#=zI
zd4E(<`{(!}{x*2yD0?lmb1AZ0NqcsVPKm8awc6y@i~Jjbs3B^9f?ush=oyR$6R+G9
zr+Cm61;+guVv8<O-Bpr6(!_<Uv*6v4wY#&>lLqJ12hQ#eb`!*jy3iQg$djtfILi+~
z0u+ztvrGW>5xVo^>_w5n$NvI7>l~jETMwBv;}=BTEtBg<^v7P$71_Vm83<Vmo(&&S
zV6^-a-ZRg-5Rs+hREl-kL5hs5#6q&BD_Ap=7Zv2-6KI`fNL$dvW9<Fqq_#GqEfSlW
zEguu=jia1NjOHn>Pk-dnZc>po1m#sfuN0=cRb*+55^!bg3>4vvN^V4od50E>LYO(u
z7A@!&cWs>)_h@Gc$8lbY3JPtfyZGS;Pl9;#c$%gVPm077nP1Es5MBX{tK$o<P`00D
zr>95`%IF4)DI+Ds2gD)2wMPD+6Pl+`QAr*d?GQO~nN&V4B+##10AeVh*vrc9u`@P|
zP}%E%V0V96Xrzd-w-Mt4gh@VNfiRC7Vy=265XwzyIB(&7u1C9aD^)K;uHoi1_{6xn
zU>!m+fQQrTFL$$a=%#>P+hYH`W+Tv^?)Dyk8bvpyvlsUQq+=hf_EzR(Qlo-wtF8;k
znht4K*S!5U<~JKnMOlcvjJU5&UCXaRdMQjmW1hS*-ygiwlAw9YAVr||u*Y7+;AiBw
zXtsh|m^=SD)}(o+V+{$Xr>3c4IXDTZRn0j=wdbd0>vah2ERTy#T#!DHb%L^({j-<=
zIohR4(nOM)w?0R<L`~`4JI75ws+3cVJk3emy)WC<*3vEpn#ArWY8%ODdZ0YRCQvvq
zFIr_7u_5lqMssSB**2h~C`_58^?;Gf7p1g%QakqJXU0a8E;FUEp&xX?`2b4!MX0<j
zhdBu+60T4VanS&dRtv#ws}@+TXKWQ;+Zu*upZ14!&717DbJN55c6VjNMZ@)qZ&S@S
ztzaR8i29u2#5=u*5Mol)RaV3CJEXG2rK+6!c9jm_EKTWb8XE(5Wp0z+^)NzDSR?gF
zip`XoP{UiEzS8EShsE%i8Hs3>_!&Gv+E;cN+DFo*V3Ep1RDsJ`n*pwX6D9GAL*Vyi
zT3IMNxoZE3*iKmzYBW~l73<_5jX%x$fYj(wOx$;HyX;@MEmrNn_YfD~MdjILzy^1_
zPs~eIgDMS=65#g90<QcO@)v}|%m?@VoneR&zB&#ftu<|DFx$-c_sL9mHoANQb(*;w
z-#@u~pJiov$UnWGAD;B4r+c-!kIF5}Kcs=&TJ89FxJt!(d}R<;U1TKadhI4&G<CFY
zm$ZAbH%hG1wVQn>?@Tmb@g8TrrJtz3Ua!_*RAzJM8Y~w+WtD3t#@=Q7zJ+gqM1j9(
z5CgeMS7YV`(=gk^2oCIxPuGp8fq99*e<gvA#(bwF0N~Qz$|RNyc1t-ZO~;n_A$WK?
zPUe5+s`V!y3UCY_RxtHDS)<UkJ=r9_bT3;a)hjm-!DWv^C$O7x_8<&VmpEB}UXdv=
z98Wjlqgl08@_Wq!YlZQ(pKaq>-TQue6@HW1qntjqiC4U-6I-41UTI_Ku7a<-;c{HW
z>Q7zGC0x7iF;fkiS>lLWeP+x`Om$X3pjO<AQ+FmrxUA}=x(R<MlowHA@=85D(0<@5
zHA)CLylV$CwjRCZ@Le9rI60$BoO~4>*%*0hD>G#&XL1yLuL53H%hMvw212q^Ydu5@
zb8TI!E!Nw`t*3f)yZd$pK|GJ9?6v1bOK@}-s~FMZ0~KQ^96Af_7DU$yprTQ!IoqDm
ze)x%N&>>#Y^5m?E6SEqs5w<{ihr~hDZ(Ky*L8F6yMM?alr>Ye*W87W?Y*L;*76sf<
zl^b8#j>X1VW49Y5mf?juc*HBIpSwsTg_FCH)1zD>SjLD$$$+#nOJ{XGNl8MUN0N$i
z>=3=G$sW%0TIu@U-L<8KSxF|_>l{E(Z&t*2aWL_y*ah*-OMkK6U&}M^bqp7BzD8KT
zxynCjP!fSP^AR7O9v3Fmzk&un?7vRikAhkmA45bwMWUi{3XUKW<<=1twop_6d{+^`
zuh2yd?=PEuL!s9nnhQ>&El5xPwmXcEEH**@ewU506ECl#k0vNKtTp)O_cb}n>@<w>
zSaInPRxqC=7{8-DLqTd~p1?-Flk8sKS^gLgZ>Jv6kQ+F#7()d$flg&hxCV71eDmm|
zODP;(x>Z3idM1lRu{LiIw~uVR)&R{wLJn6ddGqr&Z*YO2hUfLAvi0^7XSw}S1s}KF
z;L%X6PwmGuUz>rT{_rxJ=o9zSeP`~s={b|8E05#ibYd!v#{D%LwN8uSR>Uh;XHDCp
zDsQ=mj>f_E)^%g^sr87xBmPVDEGIh|>Nx%3R@2GK%crKPPUB0%+c?&-l`P8p`|G)D
zwU*rh?RqH`cL%{kY~*Wm!6%$3#<4}2MALcVktK%Qm~k<Ms3PS3uK-?DHZvNHD96sZ
zL>4tqqJ-9%-n2taNCWMqqZ-fa7qNzGqaynVAS#tK`FDot@AfMDnc(BW@c*6=83-3<
zi<wyw2@!)_rU_A^7<Kg%F;-Tl<RlFCXnThl72sK_#Z*C&0Y_#YMze6PSgS2H%-UOS
z8#UDxujyN}oAsw1{MyexE=Jw0`p0iq3UgwA0e2l1>@SQZ!r8$tML;HXuj-K^&W%}X
z%?nesCAp8s+uRquc)nC%FcQ`=hM>SPtoojb%1ARh>1EZcxE9~H4|nPLb5b_CBkA)W
z`9LJg%OoHcSS@7K<LC8()h}o$iwB&Hfd!eV)qw|TX(m_A@O)f7;TjM;Y7mq1@({EL
zzNiJGZ4s;bttdPeC-RYERlQi$9v5B(VkkTrx0r!^fAWK|A1KSh&0u}Nb}|bC7$PXr
z^1-WT1_B@HF{O_iwtbDtkS_lC<`Ux?*Te`fDPtktriJJDsn5q5HTL67sv0F0oLQtc
zZZYebz};GJ^IsKWQl^?DZ#iOoOEUVws&%m&zPu|WKiF2x6fU2|8W$}yLgDkla@!2$
zRBmgrIJhyMT3k*y<qUU<wu@YJGA@oqh7hY_dq|g1rl*4INF5i)gXJEMm)2kP#u-0L
zLcDPF`J1-H^DMZHi}NNj9<hK_Yx+Nnl5)_gi`6}$Bg|+}AJK;}swNC0_S*Rdv?hN+
zw{k9z5dUm<L~(M0svmzoX<B!<mw@JucxZgHQ=>lQR}Jke`_%wmM0w6_wG^<ZRe8b$
z3Thj6cA6CzXN+e3x6H8kl&18zpNR%6vFb-+(t}EP6;@IP8r`8f*tPtN#P5G6@o)aV
zo+{uoCTJJTXeRx=VU8r?Q3l*2yKX5NyVOs(w)F$<Wz5u1`>2trssR-Rh+TeFTI&Bc
ztO~XRD+&>#^H+#X*@>`~@N*oQA#Y4{gYO<f9xh#HI%p0;V|83p-KEbRHj-1fTGH!;
zcNW%-dhQxpqFo&zm{x0Z8TRt0rT6N$7+5{bXGoY<Dzz#e#GiFoIU61luq-<w6PWj&
zK%_S@U&-J8X3veDQ0H}0q6`X!?}1Zyp4$9Y>QlpR1PnBnY%kdA26a*loHKo~s*74j
zP=?IzXW_`74;aG6nBH=n=&B-DPq+($biU8`K_e^PmX=^nZuWK_*;q3Dvb;Jfydv<1
z&@($<nU*eI>_C`2NtavI4&>cK4Fl^9QkxqxDj=RaOQqSvC_PPRS7Pe3X)$2paRzPt
z*|Is}MIK8f(WsxSh$8*?qR=8aCJ>x1W&U}VQ_5%kezsleyVKEH_$RHs)G(PnakH`e
zy}y^Bz0z}*jc$9*eUfUj<=btr@^9~L{^7wkh=%^f11{0~1tx1bnu5Tg@Sot@HWsNs
z@`qjHdcfJ7fdgUA+aZqFP8D6m@W!aF;3DR~DseOzp7j$3oLGjo1i{~$N@FS~M)5*a
z>7R;&)EtqutiM$lvn9eqiIAcL(6BA+W0uq=#c*-725U3#`Zn-%ZF&z%f~`Sp31E6Y
zx~DJCNpdX=j~4{ca$ztnwAVhNJbmj*E>ov9b2Osr;dcMQT!CS4e1x6iEbf|sLR!*u
z6~mQ+vV~#8ZZ|*aNIq@f#v1D^tAnwn$<vJ7oUq~M`+x*-d1PmTbJ}3<x4j3k>yeui
z3+;y^hVBcP4%7D^ikD)Ule8eY;Yds92L)CO5B)nK8(4c-eyV1(O8zReKCL@1<-SpQ
z@3ypQzg=J07<~X|k1j>8BgvN_>A_+KJ-|_KneL-I7M@n34^5mh5T9r!>7Q%6jwRf^
zJfJJR2~j_eW{Vv)-liwwOS{1gJXm8<Of*QA>Q)*5mHJe7lNq;Alq*~O^Ebi2p;yCu
zGHKu45cIgImtsfBRWQD*p3L7oWFEYRy<nJLQ&yxYvW|lM@;DNOnq;k(8HqIA7EOo}
zoh8_UH9}e2$1@XRxf~c=?RxI7jN4l+><nCvw<y@<6h~%fgoX@x66*LPX86OAE8tJo
z?H?gjDscEbgxBp)p-;)&cIQjQsVueT9>!u``x(VdWbpX9@w|Fthc_MCjw-X6++8k8
z%F7dLY<yCW7auvYe+RvA5@)(<54Wc~Xfr5bA}Pa0!;r@ErLCg#KOwLL`n{#}1Pl6w
z9;v0rK%CN{laIX~)22#r(U<C^%DE_u)DNt%L*+GgdXkBpX>?8Ho`QG8M|Z;d8G_xY
z!x$<PHCe`fs}w8Gpgpm)dES>5q?<fWj}5X$?flv<uBgd@v!u6@6mD@y)f!^Z?QZQV
z&Qhu<9;~!RISZM#*KYAzyNofZF^PEILdfXsO|uGzf2zKr>+HB;<miQYd5iLl<brfL
zO?9|#(1TfY2u^+y2KhAPMyDBo>P#xr@MGAo?pB7eT%cXn{4$Z;kL01Wo#!w25tW$)
z$=pcN5ETT5-Gv;IW@Y*{AM^9}Q<O6iOe&Xp0-d1GlNLzV0@y!M_zBO@`k4`i6}`47
zFEK~A`z{@?4B{|8?W{H&B<^Efyj-cyvquvQ^mv$XYEVaHw(}rA!|sw6gA2~o_@^bX
z#9>J9FzY(O8QHw6I2cJjcUHCs{@wE^0lIE2`4?pY@;s!@g6LoaH?*V)g5_f2#m5qN
z>1%tg2{P|Hi#?X{gk-dM24-~3TJE(BY;KR@TDLLoGjDCu%gq%1S+Wt<>>rr-Yy#}_
zw!%N=y2KA~c7uS}G~ir&KOXTeGV5IWG9OJ`U^-O1t7uBEil-!@TcfS^yG9UqiSXrm
z8Z|?D<P>73{u|(M+ArYhpI7ZniMreC&fX3abUX=g5nVc&Hy<fnATUgHPE@&y-76EU
zP)#Wj=)AG<ZX&dE|Aec5;);`~t?9M8Uz#lbBMF>J{fgFlJ_RM?S@PB0nL-4w1A!YA
z%2Ki<_7H(FcuH5pt6*-5;FN3ED9hDR+L)e`ela@D)Y(1Vu%ku{_BXl$p0w{HxxaeF
zy@K{P;sr4-lRv`Ti57`12qiD+ok%`nh1q-#67=vwr2J$t4*%p)@c!EAFycIhUkH^S
z<SGIXKdVjdct9_?a!VffP4KVxvoFTqg;F<~!_<?1I{c>{?!>L{b7kfKOgD@$H8TFH
zdeA%w{vt$FVK1>#3`96(ZH}6v<~wm))5KtcLHRp-4-9-)DA-3@f+1W8TpuZ8w*XcM
z=LO=%GUlHwWRo}YS%-vSqJHrg<}Bg)SAM<Sp6OCD1*O7HJ16ef36cG1fwk*qeRaY9
z)!$KqkU`6q_-8SklBx3b7FVeaxRA9HsKWDNzG!Xt*qO4<((<>S&H)tC3=^yxTb%dZ
z>||9LCd$V$5n_pRQG8OYRG^X@u{A$4m<=7x^i6n3-vco0W}s+rdP_{mVqtKzB#eux
zc=JQZ2$>>zK;=}L;VlpnqYy(g!fBxkuy>kV)_r+1&<*J@A;O69KEDyR31GK^nDTH<
z*jXOcwJ1^}u*-2Wh~3cu9!Dj}Ob<yhISiF+zlbrsTuTsVw%mT%Ew{sTeuYM}g?637
z>qB2;c-`9Kv85Rejli4Onpp-{Q;gd%1HhBvsS-_+(@~Ei)hHd7W3Aj+iY*$$DwR!h
zpinx*Yy>shaG%EY`wN{GsAa(kYLTCvMZ4Y{7nm)&f2(vOa9<lT`U%<3L_+DXWbHyI
zw4A|E4@mmTu46N&$!RzBB5RY|wx<^GQqpF`ucv3OZYSqXt|hOIhJ{;QQ;w&g%IEa=
z28l>48lTB!)6E?LcFS$FAfdKgXny$Eu9;L0?~yi5%f<LJ4c3-CI_o}G8dvQjx0vpZ
zYU91vVeSoGOR$%MFU@_%wa6#>PP%+2;f|%q*jUos__Quxnx<H}_;Ij5HfQ@DoTE^p
z)V&Pa9i!e%iZqv#gOeEeUiPn8apJvf0em%{{;4Q@?BL*|&34+?_+RTs7N*L==w;q3
zzqcvkuf)E&eE)<D?6%b>{2b3%TjQ?0KO+<zZa8L?n-!#$l?K^ewkZ}UyKi2J4S{nZ
zwBbyYs2iMEjq`hArZJ;Y4umIXc)oErqCXp4IcG##3EdBCJ}#|wty+acWeIp}y^iXr
zdT8uwG-p^_!6L|{Ro6bv&#$a<SsvRsTwfEJ#&T#%36#2DRF30F;;XK#cpEJ0u4Y#z
z&<?~;ZGGy_`n{fno>DN5zQz$ecy|LUv`ELe2X;eZ_Up#@PxguxDV5k;1Pj6MFQ9@!
zxiE?Ftd09e(9TNOzf`iE6#Ik=Y8>aabhL)`L!=_blw5PDF}kMvh*;N7wbdP1z%(zk
za-yn|k}!xW1ZGE}Kbj28R>y+r5~b_%{%ML4DkJI2w^1hiY}CNsq(2+RSSJyxM75B@
zw*-H+)y<7J@yPI9KD+I8<v)S6y-?eT&oL*)(<om{C}~-~!eOyipU)})u{7mnXtCO%
zDO^l#ZmQ03{8wS>+wL~rg~ztl>@g#4=@l5tLamXuRUbN)&U~f7c4<zyYM1=M%n4Bt
z9RB*h0B=B$zX>B(TkUlI0}+aYLi(}JEO|>vmKFCXb{gb;ASGCcNylBVE)VXA5%O(u
zR&90Sslgu8=uC9mRxY-2nXOJYUiaJLI+Og{ygkmaYOO4~5f6>UJQ~a+cr-<nZ;CL-
z9`U4P&c}~ynXq8lxK(4<l!x2KwNAKZ{>0T|*Hl%MG|jnkRnzP_O}ut|OJ$^a?3&4Q
z=8tb1-&z@7J!Z|NS1lSVTz=#FnU~+Vj@nU%;kT(6(CUxM+%LGoRN@4ExRW7njVU0q
zHeeYBC!GMxbn+^^D44>C8<C^4PSP+`z?3;Qt<m71^LUW@Uv6w(#C*7wrdwDy?&Nk^
z<2IL-X1?Kw*FAWhJH&EUo72-<gm$^9R{(B;N(tQ^5ikBnJu=4U)zJSqpQFf57W<_K
zcrsG@{0@nHxU>6PmPGB+Za<mJOLvD`!^Al2&}?=?_a1hB_qy4KW=Y==qrCs>HR)5Z
zZ9cmHDw|On2yh+%7&k#pUy_{_VH-#7+{%^Gr|gs6)403vD1T2)&sy#|m7f{T%w;yn
zJG#Z8I=O>D@CC&jl(egYg7cU-0={{K6+<bp(9$Io%3a;Ha9V2>r{RbUxaU0lG!u7*
zn$xiqtHntyIlj`NmrR*G2id099xEGEO*zX4>hBQA^wv(HpXAttJgg}vi^$kyhErKs
zWou+2y3A2~a(H>HhLR$3ICE5<9L|HCX_}%kQlze_t=>1GeAejfqDyKkD~C@9AFc74
zs@YH98o#f4`^IrAwCRpu`tXe6s$nzB3o2$@UDYse*UVY3OmLJG4)ND%jip(I);}52
z(+6aS(#qAkYcl@Rx~yVY_SkSvl|PW__T*NCMi&pBJ3OEtw{YE-c`N1)A7l6Cy0a}g
zMq65LBs{((I(OIr!|3_bZy7q0J?ah=Warv4IF;S#(2Z)f=jUW(*gOU&%NJWl0UZd>
z&*Y{7Jq^IsK?YMUt}xx?W(mo|57H7F6qUI{qTQUBf{{8?#h-Fc_^JVc!Gi+>sz^sQ
z9N}Fx{{rs#!!3!*D{xRt&&R4xRW|MMtYB_oz-D#2jZ8jhjBrpa18q_q6n*fL9qi=k
zbs|(!vu2m|irtD0pPd1^(@qG`HqcI;tYxaO6EtA6KW%pL**<@HnC5cI!XO*vXosHG
z+0*m9OyMA<nCDM6g<|t6?i?#%BThS?hb^IOK_1UC*3?XcVx?N*r1EgbiP(}Qu@fDK
zrPt#%50eZsk)%KTu=KgqDgE(b;`qfwn`ZfDE?u)_>vhf3ombrU5WD_qQrWrW(20)Q
zPjoKnBo&7rmHznXqa@?e{Ur0T$E44sC&<jVYG+m7du#ii;S13A&SV+(WmP7~ToH2*
zM0Rw`w7Ls>5`bV{K$1Tw7mAB1hH~WHhb3qCxD;R?qG-=N2?4YU-o?2B%scoC=vJ^e
zNJfDiB6E1^Pq9kgOxc9{Z864^hPFw=$+;q<ToeW`!V{@VZun<4oEi;!O5+MdNM<tg
z#J76Kgjw@i-5u+O%pUJD)wpY}^{M%?8;8tTxp+cV<+TG!UVof@xqSy$+Bkbm<HYRY
z%ce;0FP-5j8DBP{xiK`PqTnrlA@;B_mHa*2M$kkShH*rQhPa3(@fyfvOVv_@O!GHo
zu+<u=lw9E_d)vvZrP9l+^ujc9<wi1px%3<>Ju{uoLwS|k%dTKlIG;q`RhpQD!s}9p
zOkne>GWY~Xo^Vx7L~4>|l8CbR;UAtu!uOGfQl^S|B2^*+K{06TiUoqWra&He{3Riw
zDo9D6A|H{n<RcI6Stj?a4GiJl%n)4g%L3!?BQS$OV9N5qF?3~v{tVgTW<F3%ZLd9+
z<sN`aHQff01S1APtc2pIP-W``A!7W>PV(=P9b!{<m}9rGjf*JT?=C`mnz}Y_y=27G
zYxh6Iw#{s8oW1UX!qa_%efPR0mn_MUen*U2=?t00lL;e6kDxthR(?F^=l27ByfS^j
zE)!K(N)wtkNgjf+wKbib(emSG&7yXTMA=ul-9RJMm!nZ%iWXy+;8J9dD$7ozjD4dO
zXWDeCW%Oc)y4?uq)zO9uY)BUwHD)x<H=2q>U*5QG&&!){xNrK(%ja#vAMTwEKizr5
z#-n$)&Rn$hs>>H`p|*4`W9RmQ9N=885y?H#0FWUps^Cl)it#FB8{VwLdTpW$k_cAJ
z$fDUcfsNIMur!z0Je@MwXRewHsn5KmCReQ_SE;4@R@zsrAU|m$qoik>r1wb?^1X!(
zvTIeN**8EE^FhvWFib)fptV$KYI<r0Y63lho2^96anVPoEehs04!>q@`tn;A?{tR4
zS5MeFi2MuwvFE$Mo^?zO;F8f-${E}DH6_fu(0{2fDWP0yVAvbmr9^qDKW88M9Pvf3
zC;!F&*z;pC6`B3vvL9rA2V(n2ZNE4@7r_^U=L=$+Tmaz;+LN!!S+G1UJ#hVy=`~~h
zHN!_vo49OF;i!?3OCs|N8gXA2>t)xdMuWw^oT26!RJ9=}8bM*$f}&dqlhhC<gY=oP
zYbgWlO9>qt;_n^JqRvO%YqX2Gc~M%Bc4W{|3}B4W4)snoIcMGun;7U(n47n3_0;CI
zGxMrUq-J{kyoKqjZmeH+Nnre?gJ!iHZXLRnoi}sDq%kAw`ALEGk*exxL%o;QmX0sa
z(`jc-U9!PPb!Q9j=VsCV-Q?Uh<9X(`=|~uUpVS<X+IT-%j=j4$x6Q|`p>={Wj9eR_
z#T7MHw`ku?7``>NcFl&ce2tYZV<=G`U__z@)oF9lDhB8%TBYSG;mlWw;h$}D>-MGj
zLx&E^X>D!XwP!ZgA-fN*TW2Y#x~z(wd1m^G+i&YWjO#hjvyS_mmlzXx63Waqv)mBx
z>58c}G_|JE1!&4dt)bl-jFPYL5}$((N;z=r@9!X!r9U;b`x{z*-pb9o`RKe;>)D;^
z@efdcU^ZDo0&ES~i+pM=z^6X{P%%e}C77KU=Rw|0OclfL+X%6ml^McVnL!ErMTe;I
z$2!f_s>gUURrL%k97n2QWdr*hDXSo(Xe#|-nJ+f-ts8G?A?9UYEx7)r?r&%E!-qr%
z8HT+7iu&?#vu~9BnfE}iXR3yPU$lx*yr51_zR|hUwoI42c#WL?W#!IaRhB$%Yv0$u
zl&|w;BA0!Vj@@FdqAL3&Lz{hhm6GJ`JAKPRr*FMyCRO_v-@)%~9ChYvc9$GK-ovt7
zyDFX5{>Lg^|7o&(*;I+jKcWOZAG!G7sgb+&|FQc2J=eT={f;AxUb=qA)AMg#*|>E~
z`-&^Jjo}(szkc&8JJ!73`qG}&cP@EwM)O_E9>n$Gz>F5IL-tMU38^IDmvRyL)u<-#
zNC*5RmXXer*CfGD_GeZ$kzp(uKV90jQ97`M)Uaga6zQQh@C$psmEKpq2Q=Uq9phv?
zM9supbYH(a92T9%F45`;g<?(@lEY~O&(I!<xt#dUg_(O}D5ghYBb39_u;PKJKoV+b
z6;q>z-_pq5icQ8*`q>0k13^_sMO_wlilq=#TPA9NpuA3t@u4pJ<dM3^zWz#%HSAWw
z13`uj%Sv&R&4_yl@dzKp9d-{k9^;W?4)e6D*pqAt)A1^u(PVYFJ*mxLG#M7tJ6%zl
zxzu14h;oQd9<-(LVY<ViSFKucYTIs^Yse__r|TE3v)oj&phRkICf<fGR9~F3tL~7Q
z>2wauNT!CL{<MK?#N7;l>kksRW`Dylz4sv0t5KiVtLb~!8lvL3ip^8zR)(C7!xzp?
zTi%9FPU!zTIsf0u`M=D`=_4<voZ#^k&##$KQCFRpaCt_KuAEeLjjyp+SB(bw2|xrJ
zWh5pRL0uIb47#ccbXBMlbX8RZ^-fh7^iD-67OBE_ky3b97K#l_=qe#CEM%Zp<WhBY
zmf*IIfLH>jRtr6~V!%~RCx8`#|76GC%<<6tU<D|$u&l@un|zW=slqBmK`Di;LfO%H
zFkA*&D1=A&AYNG!g|~z8$T32+j2tQphAIa~do@r+4rri(RS|ix7>-5car9hFrKj_s
zt#;8z3z9a`lS8|{ad+IsxoG_<v78IKZCm?QZ`_pc^ya#MkfQZiro879_HH^?q!_zk
z>0}WPIA!^0REyAADWTTXA*T%-Gy&Ubl5MNW8qYNqm;++2xl70miUr`!48YzVhs}OS
zt+MJpS(rr7+F}N85#q>XkEdtk7t(ZGkq2$BA)1(7Ooxg^Xre%_H-<wI4NWN~D^%qJ
zFT1pQ*^JrauV24<#iGV_BkBU4ssRuFhTKz)WveL39=T{#iNhO>jBK<vE}Fi;Uod6j
zh^tD}Yqf{?v_vtB>Uuh3N!>+fESWEM%915HWy$)lo3doUv$G~Gbv<%$PU^HJb|rez
ziFH4jTzAWV@47`y3<}2RS?dQrKWqJN+51m{0F>79R2Ty4FDVRmoM&?}FgOx02<A|t
zVdI|y!_pTFPN(d7JPQ=*&!l|Yz9pOZug(G|u^&=~r$xPv!n23jc`@(^GlQZ-(<NLP
z6wSE7^MNFV_Wtvri4HfaD923DSB5L*M&;I-B%tIxHSe0UA%F9~pawoD^SqY0|L^Iw
zT=1_C3DTO91@X$#p~H~3cuI?r>J~~1H^f~}Tog5&K_M>;;Zm275u)>Z4aaaABe<`0
z3PZNJE6!zmv;mPfVRDTUs^Fg^l+7SF^iYCj<VDXGs`J!W&Z5zOR1FA3BDC5P4Gf4x
z1_YvgaCHCKl^2x17pVL-fb6*#FY|zcA?(t_Xwn)M9e@>Q2sOIKC~v8Q$j>cjHr;er
zonQ$iM*Epf$uVxdxUg6S*HL-wfC|f)R+ojAF)fG-5gWuw++=mS64=PXE^!T^wO<D|
z;Oa>_O)dLK?@W7|L6)(Tk}4g}z8R^%bMb<X*^}M}#E6KPZ<|+MS~u#7hM^-b=U+{`
zvS#l3)mKco)^U2Yf|z}TiH@IKTQhMQ_G>@FFzVOo{1iKLx6(Psg!>F6gRrX8Y&9a+
z*upqTMKFarRazq*QYJT(C6rR#Z;t<wmUNADNSP5LBP$(JX1AU_qzp$JnT&R9-o_*M
zG)^otniAcKy|$uvZkgBN)e)~_7WW=Eh0ZVYc)I^|1>5dEb83`SA^j)0mUL{uS!R*z
zrEm)c_d`~Y<N0g5zoj$81~6?Ydt8gr=SmC`Lz?F)-A26=Zh*X4&c~?HmxcbCFBkl^
zLfT~{`Y@ZFUL2(ZR?=vre_9@Bo-7|V`F0!KCu>jcleK-<>^qCR6;6{#S40Z?orO2{
z=w%xgw7eAOTQ|119QeKTw{xfB6%1sT(Od>e<Th^4Bm<Lk^2UM<H2}pgIDO6<#LW1=
zc={a5`8gBlq-nU<Ju!w(-38`SUTeeu=vr}v{W)v>m>T%xVpZ#8>z=^vm#|*Ph1V-T
z<=|hu-h>@L7r#>ymOORc2{W$Dvm6Vy1M54!q#Zxc>Ty%9H;jXPh0GA`Z#N#p`ZyNe
za1Te0nfAB;v%rZtX`l`@dV?CB=CioyJT~H7$mg9~l<#FU=eah5rE}|E2fo8zkLmqP
zuG982xekj-x_*Y&i5w#bO9bZ+2`p~0mWOds-xr74O?l+tY#eMyxn8+HSFhZE+(q8M
z|C`I>;51R?mK=8J;w`t4+Q+0@@TLjwRqo3^1N_SY7tHAAzQ+FUYq|i?1_U|=#y=WO
z0<Lf=u4GMO%7(I-Bum07a6Rt;Ps-2TS$xilu%$aaENXapFnkvEBlW<B493K;&N$KY
zTA4h0nop+#p0oQ~u^#sfXiU9pJP&{2n#cCXQ9!?QnTb9xi+_{Oz{+LrRpxow5)+MZ
z1QuZ|97gr{U?`&R64@+lmdXnXo@1icO(%OPGxnZI47tywL+)*+lR~By+%Y;OmbGwn
zR2jBT*{ty_8_vec#UuAL%UP_p%&go#&32i{8d?txK`U`87EcE)^FD7Twga(ruqo$t
zc}G6nMC?xz`=hlt9qM83lkRIS`Q(<556qQL{PFIYWQo+uPF%$fzv0x6u7Cf?#!J|V
zQt9No|9$JbpR!V9mCRRk_8r6sI_7@nFN$Wt;di~W?}Qma(GDTOtf2g8hCTgF!h7If
zG6_%Qb(l<b%}@+-18&ZnftMQR&`~9`u*25+kM-3~n0U#g$y26IyR<=m%0+VI-g)DW
z)LDEakPJG?|BD)Lzi-fyIv!B9JvHd4A5LYI&I<%xWu{L(8`JN^KyQB&1M`)MfpOlL
zoh$D*yU3w|s68$)IFLj@r(R%sAa7P;^0Q!=^!;Jz1Q>Gi<aCda`#nIxa^Zz$AO@(r
zaRFe!t4B4SfaOaHOTl^Q;qMQY!W4SseJ^?>oQ@08q#jiLzNMh*FAPgEzo5Xgj>7XK
zb5%d^2up&ZA9TkG@ZM2(Jk=j`LJD*X`$8xBmw@zVU%M)LVa_|B6kiB*|NlC{z7P!#
zpCjwMLW>qki(;mNS<)Xh3WGu@Earht9vD*E78At40;CVjv(HL<p@VQ5>?IL4F8Z=5
ziPG)gPa&MgR$quJt>-2XE`Sy;dal1nnAuE}J4l}?#zb@k{`cqzxZ<-(qqKh_X(Wvk
zrAP2^K-#|n-p-sT?I*L~KtGlCQynzDXRYcHRR@y}-uFOek+RF65HPDp%;uupmQn{N
z!%3fvg}G)spsfsEGcc!@9u)@Ip4J;J?lh8JfXc{j$L)V&AvQ>-;r7x)#A4^O2I6OW
z%w^&(2)Y%BqZ<)8j6V^-MH+|GLuho0PhUht7fpZbvN<DLpZM_lpKdwW^zNls9Fm^-
z+2MQlm$$JmUAOzjW80>^vie>`?z#@5p7MuV*L-y3j>!w&k-oj;<6B;q4!(2Mo9xft
z*~m3bTl~bw-6y6XN`G}b_y%;|Bj{ciKz>926CX1!F!3?-`%Zj}>mBYi%3&86`iP3H
z-@%W~4<DS}@AOA?wc^94!qxWQ1Xm}(<)I^<EH)tc1;In>(;s;3SZZ?m1(5$`Z@eJG
z_$9!YeLmXdTmZ)0e}s1F7el**RqB_1*Cb5Q`C#k~LqIlVI9q7Ou$0-|AEY>PJ`fhu
zLGAfN^a&~bf{nT^3)G&xF+?nupPG_DM)3O}gJumK`imG{dOr2oFJ?BJ8{71YBh`Yl
zH6bg{+ytK2GRA(VZkm!)H_iXt)XhvRyQECsY$vreBhx#7lUqaCRL;q$mFIM9?wv58
zM36BLk%w`1nk)>4VJ0rw`fs+==@}_Q^&Pl{%-y?xB{4GEY)u_F5Vz2y$1IP|H(5M=
z9@Zi19P%ucutshh=Dj|RBkWELi*66ZmOim`u=K9<#hbxVufP7@!QF>d%m+iRX&ZB_
z=QA)WhviYxI7<!ql2vVEYM5gRWHsSfF7l>d*M*NC=8W{2t9sm1QcHoVvCyZ};KO-b
z9y+`VpSfxRlTwT^a~2qsA^nV5<B!5$3DwG&HJ%Ps<9j!*GBLte(>5mgl`?3sMtSz1
z;Ieh<bNvGcqh_kK$LwigT1NI`lE51kSX<ncm5<Nm5;(hHj|zirqA46z#;uX6h%)ie
zta3SVA4+oYA?`lG{VJR?b%su2%jP{R?!01?m>rBpgV|-Hrj9B*vSa^)fthU+rY|lX
zK6!X)<%9{9SI>B9;CpmL+g`SLN9}{_LYLNr0wv+>f-#XXqppeG{G$bHtHX;Xl}@c5
zSwE{{@bsGM>4O^<zH>T8pK<#+H$rtcMD)dAtM@XoY<Z5I=+j{)B4#A=R_0EwFx!Yv
z;8V%Q8}OeeTkO7j6Z~{<g4F|oJVdU3;3hvN#R~n(-UN%)pV*rKr-i-wu{?Q{5}0@&
zxUfumMnAg~A{7Oy-rWgq4X1WTH0pB)YvNeeY*I-^>)k7r!J3_gQ;gX)o<XC}nW(u0
z!mBSHIG`?FeM>`g!_gm0>xNX4MVzzmFipM!ErpD)d;VX;7aGrt#uu?t4Rv9Zu|HGd
zIK<g_)ozNjK^Kd&@+8he|2uh1vB?*Ww=pENjoRsp!dz<Z5b?K5<)`>N!n}A6{#Ky@
z7#J4UX}bEuWoP+ynpJRGjuGXvNwhxq--uSR{5trgq~)e)^!uRrd?D&Z<Gl1cW7TA^
z+TXVA-Zm_`iS1uh`MwC^|E`~UIOT!)Dj($meA@jVE=G!kP*A7~ivz)v8w{S@umtUK
zet$~Tie|IM+$9eGS963=g5z0*pgnHXWYg)Il{PUAJKt---=mXS{uPpV9-r`{xnxbB
z*ctuA=!MC|%{e!GR_HWB_0uGs{&tac5(k5SQx_BmVxj(QLUmTkHTy~{_QG`r<%0cx
z1-V2WZXBz$3qGKc?U2TpmN77dgbQAfU}D{1aFR-b(SMM1soSP5nsiTYU%KG7JC`iD
zeO5#Lh$|ZEM>bvr<!+yQ)2b`x+_3zr>WNc_R8N>9>)6j!hq<5Penh5@ImE;YkZL7i
zF^2;g3B}B~xxo<@HHI#sHq^;g6qro`tk@v=szA8vDa{4~%?8ruMSQ9+)->=Lu5MG8
zI22w7GV%o83R_%bE}>pxrajK7iYSVswoa?VTTmrW9;ZQXPJ!%JR@e?@GI>QZDL9$|
z(bzg!z11DXD`t5LLlW!CgQz{p*~fiG9~bpvJ5m_oOGy&xgUJZJd+ALhb`P}<d;HRG
zBO?wi-hMdH-Om1S%kpMx`5pDQ{4D&V-?coj;i^j~JwDl^mRxKr%bQnS-Soug4?eCB
zuW1`p(CjaNa_W@_He5ZMw`R*TldoU)?|asanK5hbl*Qaqwr~(mPo}fFyT7Hgx&ur*
z6H8C@F_h3g24!HkkVkh%CT3A6$+rR!f^4~up&%Rk7z*-q0US~x7C=-I%P1<rf=IS4
zuF@ycZ(huc78HMHCqsV|zXL>bRwqLl>3;+`zhS}kw3FeMmM3;}_A?IrtTu)_ay|vV
zqcgF=pTwEi{$#G~yi>6Y`<segq)f#=gj2Ee3;s28vC)VAuG!cDaGSqtJ~sOa_5BdO
z*OKrp`4_<Fm*Eq2zUV&*A$Yyt1>pzO&;2eKNwbnbO3-mPg|mdY_CjzD=nu|-0_S0w
zjzuz{F`d6K`p<((oz?FGSDo-rF931EGvz7F_kd55MsrC~=Hc&#8heo<VbKrXNO|&k
z4hjlVGDr$G1dTPGtEn&t#4@;E78EPY^l=YjzDZ6H7t3StgfhFx=i;<iMn+EsV60qj
z?$+Zh%<qk>^SBe=g}b_Qv+v(28^(To2bL)ZrPxaM=4t4DD29&Fq!T1*qXd6U)L47F
zBa+jn>0~|nta~gI3o1-FsZ++e`8Y>}D+hV~!=d&c<3l)3e?VtX<2~!)@R#0urp~}l
zHfQvA&kIhxM(L`D`;DhkB+t6`_vTx{DPq^t=b0FoDe`lf1w)w5yr#_=+GBLyxasWs
zYH{kKN}gv6_tn9Db@D_v(V(>{9VrUC8wji;8=HlYts7u5y^lXhqJTcpkzu#dc_`SM
z(R9uv@N*h7a7ZN10mrf6{WN1-1d`--(PWQC<h(r^h{7As^-pDdzHn~da$?V4IHYGD
z6J-9*(6LTp0XUwe^y%6Nj4Mc=<OPKFy1Ve%yqL0elR(_<P6uf#cG8g@>CP@?LXuFJ
zK}3_e>@`L-o624zqRWlvZdciB2(7jDr8GJa&SZ`2O*p32K#xW|Vmi@c#<59GJmT+o
z@Olg`rVMs;Mq}{nc&5iW;{sUhLq?0!lX!9=>GO_e<chsms@7?2In=(F7NA$EQAc~X
z0D8iOT!q>~iXw$Y5jV~kvODCbNhdc27SV@~O}zR2n-1N?dpmC!GQ{|D%@z05PWaF)
ze9Vn&d%5|@A3OGLSn+s+dZ2LmU9XUJo4Pi3>?91yvs!0|Ctu;NofDm1Dm^XzcGWwN
z%=_rsj)!i4^SbiD;Bmi49>;mGryy>3GY1IiJMDQchU^((aj~ZBoXO9fx%2IFL1pG*
zy#B)r41qrQ-#!GoQ)TyLETn^;aK@{kwa0Sv=SNcpLyNidff}BL%P>ZmyBMxzpMO~N
zd4v3ZrbxH^uWMxQcdj*Gb9MyJ$#bqjE^)7GF7`6?GwC|7ze(3#Wv6T0z-P(4evzA9
zQStXb@VWzCxBiA+bDOEp25aTb$+eFC{%h@owJMD;7r9u>K%Qqh_bj=9*WY^C?4<uC
z<99b*vk&`T`dM>9e{1%qtXb}R(a3!-7mH@awd#NM_a>dH^Q=EL&xd$-g1$f)euQ=?
z3Bv+Z7r9|^xuzGM&Vt2h3qVa2pqhBHKhT~^0j*%M0$NcjXkx))?C@UJ6ulsr=M&TZ
z;a>m0u1TbS0))?L4FSJa<|QAcLlG0kr#GI%P1tx;9u{-KEeL>HP?3s%^kDPBY0XcP
z0)4YGEJ{%yH;67R&AR1h0s3ue1`I=sZTKvoi)Op?_zV4$<48Z1(0=i8xSxF{%sgf~
zw~t!{XGX>aUMh~8MSn;cQsWGYR$hLNj0y*vcIwQGlMe{C5N^7W=PY@c#O6O1PjIR8
za3W3jsUeNEwT(k+8tdvx2a<y;R!DzssH+<g7|O1nUO%*f>P~IXTE3pIWIW*2lwb^r
z^QkeNCO^O|TPFup+%Yi?2#{u_0i#6;sM4?*mf)1sBHJOI)2+$$(yju#sHa2neK;iF
zjoE@j#Kc)LN~oO3!}mNmJWiezA6C<WacpWs32`L+2<@}YH-CLk`@F5Ix0J5A<hGwR
zwJqE{Z3Owl#96Cimv0zzb(47c$d3QG?%C~3&%U9t&gk{eUHgmOvxW++uXQ>*H+b96
zrM6)EPnX>?gZjpF7B}c$H<N+Y(C?ncoq2(2+*$ua)3|vobYpJH)~R=<rDyc*b2!gj
zZnRnb&F20RELOksxm7-;M+I=k+ZCL>`+qCW<fag(C!@c<m-C<uEoXlyBd=4t*)P_(
zLyFx@@Y-&`TMqN#h47Yp0lek?kMr6g<h9HkS;B>ki(xs%e)osNcN48O=fh%eZ>Nk$
zAKj~1#&mx-CeeEu7#1@?hZchI2&SSnu}_hw`xNPer-BsE*>wMwSYpP9^9UBSCe+>p
zDe4FP51~HWbJ0Wh&u6mw!~BVJ4OYK2;3u8CPeGm+54yk()aZ9&yeBy^KJ8yTF&?F{
z->LEJrBocx9UiZCBxF&3-i-3B&S%eh-V8o$Mt;tWzy>ewQ8UT>L-JuWFaIMCo9QIU
zC(K;@<7T+;dfd$74NLob+)T?x$)_2X+5JI(&zqUYa0*?0K-U~*0CW4fl*Q~!aTp(%
zCgg|c-bb0txbwmR$kBwvbTfT&37|^!gA+EeFa3F${^mXmAn0XK{kgPyb1`gizHCmv
ztU>lf<or&AY)a&J`r(r&`pp@(GmK`oiea1(sog41<N}wcOEjg2Lh^3VOn9fw2!UH>
zw5m*WA6z1QbEWLbdHwi9G=Y#_CTPnLMJCfpz=oIsFD7eTXVTK7ju!W};pW!#jFb&H
z#x!(VUD4!*(3Fj?5Gh2Qya%*wzL{;G_be%WfqTc@eRm_vySq;}vMbWhwBG#DqjNfl
z=JUH>`}CHNADhGa?<MMAtX4L^j=TNsJKz13<;|1t{&yu#`b@ozKBK0bsm4jFnW$^M
zVbKLFP!y7%Kco+fWjd^TcZRH)n1K#q-o1vxT>x@ZV5RP%-9vZ!(atgO7;E7*P=L)|
zJgpRonjBgWjqB4gL-MmA=_4aunQ|Yp)fQ(wIke;wvI~KzP-YWNsm}ZPYH<Cy^vOaT
z`tY|L894NpH}$Q)b^LtLun#x&aqPJOtCk;CithZ)D_6Alsl45CUDJjX_r6qlbSdCa
z?r&^iw7@xe*wuShhofK*3OQjMLocL-=#~-PG0_6Tqd~)?w$eHrMmL#ex_9KFTOdV?
z8e$&SXr~#Yb00#^okw(1S|sbw&9<lZOY-J(B!w{lr@3p1ZQ=;SJG<WXI>z>{oy4(m
z;%s8$B+i4_Ng*%VM5!7crGY$3X^;?24kaLjDnt=N2mz&m6GgNLl{Ct!LWu4<E)6}P
zNL)%G2c!ZAgj$4p@TIDXgajp~^UusWCUF9V9;ztbc)h!xeE<K?%s-FsyJratZLh4K
zeJ_`OHh0xhBb>nb`bnH4CRS72jE&Y5(?CG_7j?!dQ}H0db@#z|*y@c;_+!D0`s)8K
zcYc0A3#ZB19J$<c47l8DeAa;`g<igxXl{_?QcA&ZmKZr?k{yyDNoaCt$h^nueTjRd
zpRAdC&I)b(60r%lvth!{D_Vf{kWlc!PCZ;H0_mO&oWRN00(pqrf_xghXIlq{n`;I}
zw%a-?SFRPNn|5#MDsI97+T5@#-cSFRL7@xX5j*L3D0NAB_B(K%YG>bJ7s!Aq$Fw^l
zX1>o`2M3N1+dHaOtZ5zG(JUOQKe4K<AyVJpx^nkGS22g49ZTQjo^aDI{e0`3=kLpY
z{-ZgM$8`Y4NEjm+X~2=|tkZ*W29#bHGv%2vR?xPB8l=CF$}=<AbX84=5*gU=yJ0+S
zK8)c-VMV8)WHnnxhHV{H^HFZjdK-!L(_BXA5_S^INtqSc*v){^eYDIJ8yG;(hsjn}
zW(Jd<uO`$&4ayvVDaIUT2`^?Q^znzEZ@0Hst$4R>Xt-sr(VE!X_gg5){IiktP5vaY
z;eS?M>kG=eH)na*wGPd5g1N=TKHbT63!8*Nt^j8^;B(j=aFxjpOh-6rC-o>hm;*p$
zX$1(o5mG2<|7uZZ->T_o_4QzBPfuxZx;i>tT+)y3R^M3F(-SPYvH5ywu(z*7yMZHH
z!MWA%+$F+AP{O@URn!f4ah$N+8%{6o>ewhbOv+4B8QHx9qs0hdpt15hNs`JzjxyV{
zgy@h|S!nj0l40)rta(#&OF!yruC8fnu8wvyB2U!Y+S=RR-b>zLY#O;n{siGMFEkYu
zfKAnGQlJ{5oaT*7`4%}HL0iM<>n!$iz}Lsc`CW6KA!R?q<e>D)=uUb(gbZJN7h#;-
zls<Ee);rE;_br{`60VZF%q1cKN*N4RLZcnRsRZGi_t;WN4n<(`86Tih7kWu7DO<@Y
zsR*Bv@u`9s8CI*jZ3Odi*E(h@yuR7<=m*-&2Nc~uB^tcrv6C=LPCTrlX{J<b@D>@L
z6OAIrV<X^~@r9gW!F@)Q<SkN>e>Q^%R#I|P!y^-!6o4xJ22)50f!kPNnoeNyD4JJe
zlw#IW)hV479HM)VbY>0WFMV)%RYTMGy1)0L2U1#cS)EoiDy26I-+X&u8^7gvx&>Vv
zXjvavJ20JESvy?anEr8~t*YalfeU(ccZGn6M*S7uV)fyEvlX`|XE?g6+vWg}B%F!I
zOg?$Z7ByrSlLC=$Cq+V}hHE^+%PrWV35l?EOPnpLi<E55OUc%}ie8A#!6@5xTb;?s
z<0Ui6Nf*6Lg=?I%yyKw0Lx(IW=RA;l!yHfCL%S&ObA|sdeQO9k37-F1y+6M6Be(O<
zJ-3`b%lOu<Gu!x4{F|QH&mW!P`H`7J{J|O4?n0;fqHqyMT7=C~A~Fqc$?6!H{Uh3@
z)~lKdw>;eleaT+nArepGo@5bMhDz8o@(YS!rqclk%J$SmKz7Jc+^<e0{O~cXiX>ZB
zsBVbjRZ6Go3{M_m4E89Li@lu0a=r^R;Yk3q0bU;?<_KjJM1RX<yy6rH$h)S1z~??m
zX$q*UXc17Enmb+Wbo+<siHZ{zIN+Nmlzd<}^1y2Xf0nf=WCDq<LMr!zETSu$)s#4K
zK8V}#0@)yvkR2-2$wg$XpHC%AqXU{s7wN%15I&>(<3iYyf(Gj)GVFL{>qohE`ry9a
z#trWssVYaV^xZLZJ$ZPtHwLoUF(IC}UOPW}u6^gqYZpFwI!#vtWNBja&)H-qw?&eQ
zhm##->RHK40pJr}hb&{m`*Of;_!q_cmxt3Yfi;C%Toh}V1xW(Pg|)8&IBYJXt1P{A
z8F5a)aV@$4%-7J8qv=29BFt2^iLXl|7}qeaVH*(R+)8R?;blZu+Yi_l*i#H-oR7+i
zUIVpn96Np#tWXCp6|;_CK^$L&)-05lO7NxSu~-b!QxXAm({UJ^B1~6FX$o1i0bATF
zT|}oFcF!(%l1<0C3>2mY8vpASF;E$i+SU|3kQE`z6g`5iR;lO}#A>&5Mo1W2EsX7j
zpE049x$z&GGEuArupLX`70#EF^Z3e8czzUpkE6&PFvS4hauUm`?m%&Ac_>m*TRT@c
z-Y5$*!ERks4SOCwE^{Xe{3T%RN*Ev4t6<=lGiu_%dSNG$CbEN!2IsK+iaBmj?#aYn
zJUjS;*egEmYwYemQWZqMhPBfN_G2H}Kz-!i*uk-Ma12+8hbH^$qt!SdmlJpSa`ars
zjuqFw(z60HK0i45Y#%~<U+6(eB?Ms#j)9|>gsNH`i;)d+6?NuP!Ua7fqs#s;+`<gz
z#ofXT=GS!#2>46j3K`dUDc=ZZYJf78`G}kOh??;cx{TTZbAaF+pq-oNZF-?^Seb8_
z)x=^BA*CNMkEldT4zsFb0bgL%$Uj<tw2lA(0RR910{}wPQqb>T4?Oh%9|!;d0002d
z``Qcu0002j(TeH+7yb4JeFpFV0ssgA0ssI20001Z+GAj3U|_%h_Z*N6{4f4*GJ7pU
z3WF#EGI#|5s5=Ll0001Z+HI5HOH@G|#m}AZ_udr=6A9X&#gZ^Vd{`vpvMhma5SxL@
zB8v~pvMtd=Jwy~)WJnJpBGN-aLP(@+$kJ1Yh)6%4G#^5OCBpm%3hJRB1Tmdi+gPj%
zAI{wG%-oqdXRi5(6!@4$z}up>V8mQRkB-}l0<HcRjOZ`Kyt4?+CZ^1$-7!lz&%S~;
zW+%-s(%u-Fy^=l1+K|G_QC&MpI!O;1A=BhExmg|0xolq{w_VL_U_4krRv+0FJ%hBC
zamhE-CZwf8{m%DpuC;}J>9gw^XMGFlpoM$vK-!<fWo`eXr`Y$2d&uIMzQHZMfu%?S
zKXuH$;Jjr$U@H>Ex>v$HH7SZ6nXxO<>7`{7F>OPS+{1t#;aaO0GI?a=xt-QVgkBj5
zp0!)9v%P>Y=tk)3n&kL4yY1$fcL=4?{xD?}*;kYT@_rmwrQI${!FI}F``VOn+yLv;
zqteB1MXTNq(~G+b`(Cc`%-9_xFOd-S)PIXndY{z?RbLCvdsmY#ubIAwtmVyP>?f@}
z%P8Nc)ZhvI&T)+i)+<uPeVxNTf7yPkLGR;L-~VMTn6~pS3ztJZE{A&jC{nc?_QHMS
zsKeDP>Ty{xGaaPA^!<PMU#ZN(<xmfEh?A2f=2W}T|2K!d&?{<#%c35ah0Dxo<QVP*
z8yw3$(W`CpiWwvL4Aj&N@E=(v9lPKI^&GXpK4wuvX3!fML@#}xB}Lx#3Er_;QWUk5
z@(?MV;6K2*W)3IFA@^PGvnkStHF?T*12P&Q6I>ugWPSsV($zzF+GAi~(1F4fhCYS|
zj8#kmO!JubF@0cGVoqS5!F+=G8H)gm0?Q6o2G#)9V{BS%v)Gx~)!5f@h;YPkEaTY2
zagXB@X9kx6R|_`-w-5Is?oT{cJZpHxcwKn=ct7!}@y+4;!0*AoK|oBPL*SdBf#4*;
zQ$k8YAwtuHjtPAd))EdAZV;X&yho%+<dvwA=ngRjv3cS}5)2X^5>F)UBnzahqz*|7
zNH37lkm-_nC)*|`BIhHwMDB{bntYf7pMr+M5=9pvtWtcU)S&cBxkh<`N{Y%NRSVTa
zsxQ=d)U?#@sQpoQQ}56e(yY>)rTIupM9W62MC*XIfVP?T8{HngG`&aqcKQnpI1J7i
zCK$djiZOa%Y-ZeN{KllnWRuA|Q!~>UW>V%H=2_+!EG#VkS$0{jvC^=bV$Eh<Wy5C^
zVROv3$xg~{i+zYgn4^{xo6`YjC+AHrYAy?0JzSey_qnmSWw{%8n0fSgJoEJM;_;f|
zb;aAq`<YLZubS^0KM%hf{z?9K0_p<31u6v2!vWU>?g~5+_$csCkX}$+P+!oCV29ut
z!B;}mfY2i(D<mtFE3_qaNmyRERD?~W9RS9@ma6~&0RR91>H^sS1poj5000620RRF3
z761SN00Er<0001Z+Ra?qZd=D09@*{^Hx1gL2%4Y;&Xp?_G~+Fa<C{Vf=~#`mK$I+`
zcSq!*%$g$Eb4bg|1N1?9*PEaZ(Cebh6Z8do(Ff>tXTJacXFEfRR1CB*M4n}4{(YUH
zl~Nz9y{B%fwOj8fwf0k`_<Buc`i`z|sCU;M^Yu;jX0657w^VcO&wTwI_0ihD`TD!+
z&W&4q{XO;O#xME$`>J*04}ATO`sl{L`1<zR*Ea?357eVuU-R`3fA!|pH+=n`depkZ
z*FREO>o<J;WA$!J#=ozA+4?<S|5bg^`V(LOP2Fq#g|Gjv-fR7bum7RmYcrnzys18F
zzpu8{SWVQWD%C)psh4W1GWAFO_dwm#{~oEkxVo=CQ4Rg{K#v%xq56fY^xacEccf=j
zs;TdG^mP}b?&!Z?>vxqNJyW^9Yw6EY^;Dt9WNK5dFx1};W`(;#e;4}uwf^j3%>zCE
zRF55CE%P+f_m}$f3#^#y`Gx9RJTi5xI(q#~Jp|Tw)noOU`doj1uIVuUGIfD}O+7QU
zLQ~T)R_B^FgOA|X(c?z??Q{M7K))GlYNz^{!K@q5>Qj0r;Fsf8gL&0dPr${+kg@>Q
z6Hhj)&3pmgR{DAjDLmBh?}1BKt#cRagrhRZB!hIP;G?1W9Mr$k&&Hak^Yx`+^`&{G
zzAVzDJ4?ctzeY;`*Y%u@qz#DnT@CYXTEP*ZRRPbtdd*7DF}j*r{WaP>(BCihRS$Zd
z>v2Q8F<LOf-x+2Ys56F_v5pjwJlWY48uV1JV4k1qyD=ndw4(=H%&HYE)M!;t|K*^j
zV2`k=-<>g@ndaX;@JD!QD4uHGKT#j+f6G}V^F8}8;`_&eKX_`n(PmtanWvf7YM)Mn
z;~98hbhqGH25y<8{dPptqky8@c;D3{M|%8)=BeT2nYXs#iqVi=P07AG)RNs)d-^le
zpZofIN9$u7Hh!R<>AUBe(he}$$G8qC-PAKT^>^EQo@GDNw9))MeXm!~ehye>{)t%&
z@AvdPqfz_dg&W_-*rNbihIO9ms}sO%+nEOIuk`m8{H<BFt-m|q_#Wsu!h1twhW9P~
z`9lA-^o&gnImWlG+Q8R9e$I3E0Ph*#ZsWd%m3O>%o4_+#2^enU$pO<O>m2FNL(TsV
zQ_)JU-LX%NgowTtQ$lMIUt<m$sT=}>UFHNy+|Afw$Zq{Z;CcjZw!!N*<~p2H)Dh;X
z5r*f@04BlMLuB0c_}9|kZHyr86P!9M!DDcXu<bx<ZOG-R2YF>ZU8G&vPgo0mKE*sH
z4Lr7YgvTPC?dZ{4+84K(&$Pn-QX`C~*n*ZH=ov;TVwsy*&&aR^yoj<x4>w=4W!77j
zf{uS}$a))^W5yYp7R}j!A0#}9E~8r~u;qO{rsa8_XDpK#s^dNJ@XL6P_84D5Un$CZ
zyoKl3g1#QHZm%;o9eCRvaQV>s7m{DdO9Ou!Pj^B4GdwX;-9<J)WB1{^$^ZHIrKI1O
z>zTfU#hrl5Ey$)5_%pHrb7!6#8aG+?#*4=`xV#3td17C&b~a#Jd#rsO=1a@-p}QF4
zv{6>*Sj*Z1R$Kb|m`9OL$o!^;{xr#(x<_*Bhj!apiSo20cHu>iU_XZ}*&~iV+u%7t
z5nmH>JzTG4S1}Q8o9%oNOrw{(n7IR~w;6Jat?$TM1^Efe7sPjmxGb{f9roy?JA^5%
z5y#&Tt(P;LIA-py>!%x<8soF(`#Ixd4_If~prPaG>+1fg&C(>^A1>utJB;P((2RAx
z0A=k2bx!w;gm;+J`{0m^<p-R#%-4|=&Z}6jyfv}-h0WMpg!l?L9kR#R<|xv!*}&WO
z6`9qsIG5oZ&qCO;IUspVikf@S`(v!(`xLSLUEtcb^z1;39ga=b<n5(1fmj~mVYh6~
zbpU*}`Hj)!#=@wzW4+ZLo)A5TvMpF!<BDvJFm5N;ddmJ&)*|h`3p`AHJ=HZvj_M^t
zg+X<DhWf8Sjc977iLO?rT>VT?&!3@EI_D81^axB{Sa4l8;x(w+oS+JN#kG}*tQUCK
zMU`nP{!5<K)fLYSb!UcZ@C-F#jyl(jDKKV?5j}9`9BaR1=v4ihPC|~ZPk}L&9%0sy
zXQsDcC>q*XuKG1Ir>Ku};2uI0+A2*`p`Q=1uGD+ym`j<=6kIBplVwm>=UlhTx)skT
zK#8faf2Z-2=T%KoQ&f{fe(%GkC$y&F-iV)&luRW$v7^ozJA*?NROLPJuwot<_(bWH
zdEh8nUQc9op}mJyXSS|2t4m+WP&xGA2@eZU-qaR0WDG^(eTGUMhXHUgT$uvyl;!4g
z%7=BvT@~qU9ALF`P$@W2{qOUs6X}5K8TcInS<}EVQXSw*sIII=b+DG;Wu8;N^izC(
zhR??y^v{hgXY1;|*WWQT+|}6Up!|}%pCz;^_i)*SX8PR((`mbeRb*Dna>j?|<X*4N
z!J2rmCP`Sih2^>;Il5po9#ufcGJa=1&p_oseeFrA;bB-pkEPE=YYSh<N=hsDDe#$k
zy1FW4hqv#w7IW(*^Q+6!BrN@|nDdVH{!tT=nAY!utImt$cC=61|C4kKgxA1t1WLp1
zBvC(O+ZCBO+o{K;3#QWep#m$HfOiUS<4eNl%orm`-sb(t>--DUROgPYalqKp(*gUQ
zOUvm&U@;f$o!rU=!<%u}kZkn<B>RP@oz&xQfy=Le^`y3EA+ielj$*kZPs_-2mx1mQ
zgb8A-qmrHqwj+!sKT4c(k#ZPyJNw{QqZY5jxL<Gtpt%HpkdQz~Q{ic$pFNa_Q8QRr
zW6o@vb!K^61(ueEm0Q~H>iaiLjp5^KrcvHbdEJU5jpL8s^{?ypJPLeI;-ptG-Xxs@
z5W0XphQ)P*OeLKcP~K$nlOg9-;u$IDnXuIyU|+%P)pS9F$qT;WOtnPzainLP2=|S(
z{(an!f;UsX?sKPG`t~Eh7G8`!4@O>Vl+gPN?1&6&(K16u*Mn7(UQ@&!*5Yiq*RhCK
zl68`g7+?*_Gz;d}2oy{#XZx%fCWHM19i44t`<17+%h+<$#@xD266Q0{N6*tEKMy+x
z^djKyd~6#FAcvfH@oKE8kSw1pXP5Vd2HHN{?#Oj~S!nen?+H^&UW9T%V~-|dJY`s9
zCqmW777NnBb>vjVTg9}3F8P@nY0wb6H5;}wYrOplXc;AWJ?34!5-X;5o@7Vu6mLrC
zs=rHZI90-?&!7SG_O<*y%!ajOtv5VUvMAE~(pC`UjY?oJ1CL5C515x_3=^5J%P|&f
zp7W`sabd>tcICLPKQ{o=&kQTnDulrRSEgPz5`6%D%2~p(_O^BGm|CjIn^SyVmR9m_
z^ZeY~)!*3G9M#TWhMuMhGE(uJBY5=&GQKf#AeWo<@MOgCnCf}TmdZp<A+>2LhdfWr
zk2OQQei3DCbx1D4kl#PkYZBI@;A}`bXIPXm=9+tb_6)IJi8jS~m(;kl6I|4^h`KeT
zd#9{>9lNt%+NzYi&e0ONbSS&U(XX`e+eO2iv*Quh`J$mTS}1VO*Y(&YFl<Eg3g%On
zvj&2*!T!}zX!6matwN-GZ~}<07;C?3b~RC#vGrllX#>i;LO8-9@n#~cyru=z4H&^n
zo6PSr$a{@;PLN+YnfE}k^c6-ybT6#MO&FV0oSRQ`sy@vV+E2b`1pRd#G{K;co)See
z!OloDM>l2YtxzE=nqYn6N8iSkA?MK3s16CaQn5yvuHF-y4Y@v#lgE9Wpjkvc<<BFw
zy|~6MB1t!byooB*RQ5z?b>>4YJPoaC(p+SN(u<p+PeES9@WHQXm#d3IUl!Zy15nUI
zt=eb4(pvMWBFlNrr*OttGlSFyQNK!jT4Ro`o!b*Gr)&kMQI|NZfv9ggwLP{)R_FBX
z!9poX)h*|-noGyhIBHTfzwOl)F}=C9ovP+h>-D&kZ+IfEV>|HAv3DY;@chir_r&BK
z2i&V9E~R`&+@GyHD<F9;KBZM%c}}a&!<a+$omt1TRK_VBlMIO}%7@50ou3}C#`xU`
zvBuZjhmqV>tgDAkQnexm^XQ*n`^~)k##p#t8T%FS%=*N*?p>wFEVV~nl7om=-$#-+
zB40MWmS*PAKH8sIhtz3KgLS~2_H-vn8{C&O`1#lMQRl#R0G*ES>!T}VW&^A@<oFgp
zOJgFV7mXV7K2~@aj@+<$&X_xcr;+`}`FAWYIjI;}AXjUogQUUd>`f~RXTr5sWp%;%
ztG4!b&ta#rXP(!`b2~L@yQpg3%s*{OCmDyF9Xeg7Jm)pmOsx}RTk@IUfz3QKbekmC
zHjp0jKJEo#OUd=m#Gag+-)Z-Kf#k%}kDEfSGwwX(3_0!+QvEbcWcOq44GwHJ9QSV)
z_b2_TCe^!4of55roi?F1%!^k$0V`f<>S>#3hUD`@+o_xdlup>ITt>S94bC5?%s1De
zj3wx@aPgY`pgX@>hcxUCE{lPy%X(f{xeNLRHkVD=L8|XKztbo*#<AM3r;Y#4t(}IE
zG3@X5kfD;+xW11sNguXvq2gilCA#%)-|3V!a>jOddu6*j^Y%qjkVZ8<a-E`_B{cQ+
zyTdd2qIIqxL47>pNd4`IyCU%$`F%oV&oTIW{Ir`){<5$g)UmB(Q|~uylq#*nG7nDU
z3XO_s^mqSbe#s7BJqL+b>9aJtEXBf`KdkkX26iu9PDa#snXYTCS6h_pjZcFr@YLE<
z{QHblwkY}Vm@UZV)!{xTttz!h!Dj)cUiFgtMaXmg{-V|e$)ir+)Yd_lb7Ey9@bFTF
zcY__f+^a$y|9^{$)2!9QjJPJ5*<RB1B8025!x$kWd_AOH#=^^kssMMcw{-P(v|y(&
z-Wk7cM09<EDfYf7l9it4OrHyobt~>nc5Qv%OXLHKbXdM$gKx8-KRnN4Irx;Z^m+f6
z`k%Rt(5X`|!EdXUmU9!eTYXPouy-EZk58%HkJN5pp6IuxPYbNRGdd0|LHckNM|U|(
zQgIbiad)!MXNX=V&a|aZZQSNvpVI3OxsxfhALQPIo?6C*S)xd+%h6&G@jTq^CBCnF
zUC!M5z1JA|!vsEV#Bpb4dmwQRzF1#FOB%DktO9#-E4XOtYptW`IzA^4PAaB;LFn5@
z?2lchXDWzCke>K$b6T{pHBUF_D1{Y8oY4<#M>2$0u>bMx)z7H?y-v!|N_!?UWxul0
zBJXg08$X7;W*iMm_`M?Fq{PfvXVUyg*%(oE3ZF$@tGQsmY?3&UCwaK*9EMrwDtUP$
zh_*fN+@V?diQWB^wJtpitax|S<<mocAI6pe<9Ok9xv%0}{F_QY`qQ3}Pu<KrSDMOr
zQkpcVb<g56o&}$<4Y})}q~#W#uXqWc-2+8*-cnBVO3o-}U?lGKqg<8nFIn@7+F7O*
z^<0m;e*Nm_4r1|n(0?zzlfyF$-Q}bazZmO2Ji?H34_DuvvQEbJm><VtOH9QllTUrZ
z8Q%O8uFLtj`mcSR<}dDR_&S{4kBm=Sl~(sFuT$sF0MmWxowAhoBkn_1kXPy1P-P>p
zJgfJ)d}&s8cRlqRXVFr3E+PxgMdbF3>SeGW9M>G-**V$OE3KSHwKJvP#?PA`-F(Xx
znVduVW*6zwp16O-c~}YEC>bk~lpik}YzHs7mMpw#dW2X&p1Z+kA99>SYjFMG_dg2T
z7a7=d{UouRcaP7#6aPJqg_m|c0aF94{Fjkc_ME^O_dx_om+?Ha{W~8IKmVltm7Z5U
zo^oc^SeXAb=4K=1oE#?Yx8-c&ylQ@4J|XqH_+XLpfOp8R`aQ?hQFLVO#qjl&wGjEO
zeii*%f5QI{LiH{G2aV<bCyNvG${iK=@xNG{*h;=YbbFva<z3(>`tzawe1yG5`lbHk
z=>JBTFh?sy0e{z4c60EXhW`T2TJ`z>c-n2yH*8aJ5XbTFC61jqz4y?2d+*s!fl%xu
zruW_nB(V)42@VEA3y59?LV^iITL96kK)?Wq-b4{FFw?~l7y*a(XXBQS?l;_>jtKbO
z0>4V>%Kz#UNQ6WZiG?Vv*oY<uJF&zOPXdV~kxUAyq>)YrnPibo4!Pu!PXTRcOFP=r
zfsS;dGhM=ay3w5;^rRQP>4S!jfdeNl3MrzP68h4QQu;H1fegZphce2kz{_BUFqB~o
zX9Ob|#c0MbmT`<{0u!0UWTr5cX-sDZGnvI~=1?gXj<B8$TxTyI*vL+{v7cicl_<9I
zj&<x2tJv7VJ~nfU_Y%zkj`M>SesY+T+~+Q*n9DqNQN=^5xyJ*Z@`%Sg;UhJ?;2F<3
zjgRl_<~1*Qg`dxSWef8Ouz-csQpZ8+SwsUt8VRwOB`ghpc^S)TVg;*M$rTRK%xc!K
zmM?tcDrY#$HQw-6V#F@75-0JJAc>MB$&w<ek|ybr!8y)ziObyP0vEZ%CQe8uH@GQT
zlFfF>kzC1>d?~Q{YMUDTx`|;6)de)ottrb*Ju3G&lqIbinx-jrrJ-~vol2LoP+6ob
zR{opkHr3T!UANZ+e4$`<Rb`{!j0~@-%WJ6!1?&GdE^k?@UXOW4SVQShI{yGR!k+E`
zc-lS9y$-=p90l-RxB7)@tF1}Yd`BS$OxmRpVrmod6k@g*F?f-GO=9w9oN(Rk{7z1C
z9_)Q0cGhyT)<wnGrK~bT6a(_goWc#lvFwo{wuMGKNr{UE=}CI$QB8bN2lPJ-pnj7u
z0b?a$0p?P|2CRvM1K4AU2H=b&nm}Xz1EZFFchR=uny7eo0NbG3SM-wfskPZqQ;m;-
v_fiZ%|EkylgPmdsc9x0}7|s-9Fq$fM!Fc~?Nul@vDwJK^00CR8+Gzj)^$oj(

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-book-italic.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-book-italic.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..fe284db6614ad0c104691df0080f0fc0e234cb5a
GIT binary patch
literal 26832
zcmV)0K+eB+Pew8T0RR910BFzv6951J0XZZ90BB$U0{}t*00000000000000000000
z0000#Mn+Uk92!&`+FBfhOa@>8gHi}Y34~=42nvGae1g|b0X7081BWOJgH`|pAO(Sp
zItPVZ41!KScV%K`23t}uaRk+CH{x*H0F0aE*Mg1~D7T$(w-W0PQlN@)%ql>4a3>X4
z3z7Z*|DTsM##q~-+W`nxS$$tDJ4&M_9Juv?qtP5|)gY0^sqUK44WgPjZ>6Rq(8}H1
zADWZaexDA61G7y&do&CM!gOFL5)|nwXko?~YY!Q<pkEkCvsjTJbTYMGxt1Rz>IJvD
zf~;ZgDC4L0H}KeE!7i(TURyVwEL>Rpjz^(4`B;CvOkY-0Bo(E~K5*&KU6Dw7HgoEs
z$P&cD*A^rkW{W&bMoswiKQ1)YUq$20PJ=GvZEm80zJ`8u<`PE95@XkcfObe$dw7p%
zTeEvdDj`xP82!GH7@t^$nV*eOZ(u}x1|Nl=*$(OS-#Z<VP6U=@iXA&v>=2d^*Dljf
z*Z%H6JH9_I1>DL*wASaVuA1JxyD&`L2^--3L0};fHK2|HbbtySI^cfj%FbPuGU|Q9
z)&~<*&ID2*dXJShRBr&1$FuU8lIV^2W|o39Jy1*PstsTj?hvak*;m!xvzG}|&VEm3
zTA8kvq|XouNjOsk4<N1Rk6WqmUoBf#F-x8yq|gqsJL!=UVSaDAAl)lZ*!!Pbq{8c>
z69cw^{{}<-upb&}8L^rV2vs~<1N`?3{8QIs)=DzSg7>txwSWKaCmz2<3Tcuy>F9u{
z0&2ROo}dQoR2_m3p}`r_>j%I9Y|})Y&dNJypi{soi~>Q@20=hT1*Hr?L{uVejA1mJ
zvCwU<T)JwP{crs)JJ^&Vdkq9bwYCX@u)(xy5iXcM+o~n?(eOflvI*47<KubxxjZkI
z%k}Va_&gkYHhqB%M#B7KjMp7St*lnm`3iGF7XH7?YM&Y1BUBHSBU+A<z1z!HtxDv0
z2<hPpZCkkayqC2Tfkoh9I2eAx|E5A5=Ctfxo{5t^B>Igsa^zNpv|a=!`Xsl%!tW44
zmeTg1u<$+lT>wHAH%e{_IF+V6(U(qlGma=ib<(V<^dWjc!8r{YNV?b|7LUU?5`<Rm
zhIXC%|9`H0`|dz_I_I7$q7*p7_0+25HJfenW)?V;rP<=_lp!er`YM|2l(MCy$e&rU
zVifM%3i*w!vLDEOYrYWT&o}pKF(Z6_!m=lDL`w9y`PchtJ0iMVF5KS%pkqfU&7KEz
zVn*a=^jkI1#KV90jjt63_%Ix|>D!!uR$5jXY;y?1WyP@nQ?+dW015-`B}tbg=Q5<C
z)aGZG&Xr&9qSDL&pk@G4A^?g@kfN5NHi(q&7Q}`lxdEcWag)r=+I6}Z923b4In*w?
z7;87ZUGH6W+q%56wg0vJe){L~6y=c^mx6ROXAj|Ws|jhVYib>7YDkZ~gxk0K&Gz(r
ze+en6kYv^<gu*$j`k;VM_rpGL|HoD8P-2$McY7;eoY~_JMVz~YA|euo5Tv{J$KU#1
z^#3(IyL%tpx1axfuhvS5h=@2MB1&oVGdBN!2Wx8^^Qse6EMV=8_Prm&a9T%Y=5#7C
zNKkFodpX9|_lx_Yy>c(oO9N4SxE3fe1A0Gvudn`>x7_{nZA%+%v{Fimh=_<IqRd}k
z5da_naP8N&1Hh$!ZU+Ex<Kvo%ZnOb}0>GFM8idY#j3E#b#3uy{frRo&VZtF1)}%<!
zAkV#mQ3QZHT>{fMk<eA=m$aa>MioXN0K>r+30KtdXf1exRy@hJ9zNh59<D4yS6-|&
zAvkk|gaMG)%j=sw-qa9Ux`rJAmPaJ3S%@}y%bP-dbDOrcooufYh~3gPgF$)l>A5)9
z<n4Wdaeln275k*+7?BlT)2pF2)aClxu5`LSz3rJDPgeZy)iE`pPwgd1#d9$>z22Ag
z-J@da<^FboX<`;$j8wE@WXn6|4V;Xiw*WN~apQ@B*3795ORf%yx_b+r5p45vnk>Xs
z=}L8NuwrW7(%2N7@C2DwG5@fT>-7bTSw$(3sA%XIn6zkPVdK@!FDNW3u2;W7!;;b_
zO`9Ec^fAU7CykQlqD)z`<(OcSDW;ijwz-sY6(~KowCRWL9eQ}^(V?fh-7OELmH1u4
zM1)VAo9adSO%1ZgWC^2JD11>lC%E8FU|Wo)*f?V;8{ycZ*ywT1SOa*NO#2wkE1Eqm
za5Rrg%?cXEpm;1_X8LG(JYR^i={}t?B#w)SH@lXjc_g@!fgDmG_;m^uVeNvOXN|x!
zvB+F{GwW$jrzqcR#VD(N1^rXV!3E1#7iHQQT2+=!ThSGZ#nnwQ3^iAqQ@aAxEY}Vk
zpPB@faZPVb#WXZr(}8M_=NsdCH8$L<LNP@sMz&Z{tgsY9kCIZ-HZvKwg6~3Q&mf~Q
z7d?wpdMv%6LnPF&=M&+~n8+t~786&DBv7ZYCMl{FqcM;Wrq-y$7MN%#9Gb*RK$_H0
zRp;f<)Q?r6ErGpa9ED5+6d%6zs!KaC@$-NLlqvYBW+8C!p077zjK$PA9CB$KWyls{
z3SSptn&~3VGMj=@t{4U3n>eMq#OpC=mZi2@V7uK^G@(=Hz;*QwIjX@m_qBNFkv>mD
zf8fW`wS!Rpy#aP&abr*G8i$(09H0Gky+?~I@~U_zK5dKfl?CN0R8o~exg162KOFhd
z3T@C19nfd$2Mk~V2Y4U=5lDed5fBuj(&)k>qGIbfBY|s@YAFT*X__&guAy0)qj|GH
zi#Mf2X?a#Lt5KtM4NWa=9bG+rU~qmo8__AcQ=3W|l&erl6;+U=T8&zD>NRN8q*;qr
zZQ6C{)TKwSKK%yLLGZT>GAy^kN~^54##-yFx538P1Pg4Ywk5VghHbXnVW(ZK-9Tf=
zUi<8i15nNIw|h9yqjADBUhs-{eBc{D<~R8HcUyd6X=QC=Yj;iC$H|Xi3|Ryh@Tp)Z
zBBWFf3L+pVw5u9w2t5l&1Q8VzmyncdcwLl?m~1McVZMZYd-(Po$js%pZK$slHd65|
zCQ%lYt58XmR!dRCSFJkr8Z>IstVL_INztxDr!GCw%U7R%0|sL&iS2gSX_wtJae$3O
zmk&#Egv3$4p0n{Hc<I$Cokt?8)~u|_HyO<3C_RVJFlb@nF*8aU#QD$Dn#~YV+!WG8
zC~cY+1p<5v2$roB-aU&1&)PLkf`G239#6I04_CkppG3^GefG|;w}~nv1u<HJV$wAT
z_5q9)*>pfvSd<Eop^&)}S}YiE*<zBgLEU2w4U+>^Dx+X(+ad)9B}`MR7O8qkg^6=e
zR%5(YhiQhZ1zy9x;Q~OM6O~sO!o_l;$#G*Y!vlRN?Z?KOqhH#Rf8@&iWDTzlRSU|d
zanN+O(!|n$<^;<QpvPNeMy@(r=?~3yc6kmk(6;(ibTGq(eJY<RIy9Zo#crCz6Vrb{
zjYVR9$_s?jCLJw)iX3JT4l_Iz;G<}M!be{!kVT+M0=pR5k&SPa^^}hG1W*Wsbwy=K
z0w`Tsd`!b=UKgUElI&@0=TwKOB|w$nUqas|d46JuaF{Pc0C!Wbfvu>M(x`?Rr?oe~
z1$+|CGzR-*j~4=Ly4;wr?N4$>sF7Q=h==P3Smnnx;5OM<4?lvBcLMa~gMR`8GXF<J
zEYEWDJWNUY`4bH_cqKESj(BD9s{?WdYEZKy*(R)K8afd9bJ%j8<?eY{5~X9PqlW%K
zT>v0edS#IXn^TebO7rL|$(2j&Xkn<0@%U!J2F_j=_%)&;2L3<*f{pmL2Onod!%3QF
z-G1lcArV3vsRB|hwSES+(c=!NeQ8ua`YpNlXn=As_;mVk;@3A1<Rx+qaeL$v=L)DF
zI!GzK|5qH=;r3CEYVwjOzOq#aB@y|9vsQ>uJY#{Enn-;}(``(;#4NAG9H4Y;3k`-I
zgJ;}h@VnhHUt?BC)t4-^y^unB03A<qI_A;~(}pcc76pL{i>WX3Vci!1^Am?nsiQpw
zv_809?V%RNC~d#+fB%<|WKo(p=FR>i)bRvEJ>J#p$){<m_j^{jk}c$0l|PCl!pGk+
zZOVw50P8FM1iP#Er+HPQsU(JWF#tHUsufe-n<#{ll`e4y*`0g&JHfE<N=#1!?S{!(
zw`X}~TmHbGTUbYf;rZ|T_P#Kpe2>|ki)RSOMYimVX;3>Orff0<BakSlZobfI=(A50
zE#sqQVdLP^!J~^$KuCm+Bc?|mfB{2BmSN+=XRU-3fC*D(0L)piWW{>1OzoNdNf>F#
zz=+|?NC79zgRpSe2G9)XD-y_Np2W*PMW&ZAX5|!-446y>AH&-l$$_IcI)Du<2**MZ
zE+RT8WQ1mn>*z5D3zV}&k}ab!QmDFs=Eq6FK+Z^TMLKuX=YdR~$RI;)UZ}+z#e5j+
zCwPrb4#?q;1_IDXAQ}q7ae|rgEMM>!zyOhGAqp)iP$C+w#Gq6x8$V-3GZ7D{BLTS*
zktYdFB_m%7>Pkg{SsJo<r<tn;B(?Gbwdq3W;S+idU>M}HEVBm3S~~;lY-DJYE&Od=
zpu{$2IrGLV8Cqh<e%$#tper=gb-0Ld<mY+;KFc-NAvcibCStdd<_==_nQ)FyvT-mW
z4jnK{sEA;GMuzsKP0n9>v0zf+)|r@5h-NNUJWGjE*>DOY$FAp~MHlZGLqm_V%tnF3
zXE6i!#<HX#L-N>1rMTv<jDfsUPKE$5C0RAA+Uw~dDR;b_PMrXfPVf>%zu3Eqdt5m0
zTu-j4qb{zl)sPl1Z>-sHW8n_~PZl}?5Jv3c00;@^g3<BlGiA$}54p%f=6|miA#-bl
z>oDtK)*oBrtj->1h9!IqemEP)2svaVgpR~yo=oZ$LXZeAv3;l)un5m#p%8?0#E^a<
zM#ZL;Xr$6)m^MPb`4(^?O?6XHFg^j%hw!*(G+uQ%1y6BZhM>G_!Wp4#ikRvE<n&Zz
zNGewy0I`rl9e{9l$Hk{#Iz9mrL)gWPU>?p#6Jt!+bOnR0oJdVHp&%!hIyWBZ!Plm%
zonzCS07mH~00a3v*nFIkM=~FxfI}7li1eSL!hZm?Z<_P0pcMe1kJ=mnQsht>NIwn$
zaCUgf01T1T0IX>qc)B9P$V1tVx4ZLt={BfA9qQ49HVonjHtb>A`0t+@UI##GNhpjC
zNw+{1YO4;Ac!0(a<M5bW==+aOJUsmHfc=K-rnXuck8j?({fhEzKvQp)pbn%*{i09<
z0Q~<O0z}<hLKzoQX=7c8QJkb%UX)ecw3DtMrnC8Cd9+#|Z?-49@pOMUJHNQRy1u!+
zyMK6mdM=Dp$}}%4-?vgl+rN4A{W!NKBqk-Nq^6~3WJ0B?y0*1zVEM8Yt5&XFyMEn<
zjhi>Aw`|+7edq37dx!Szhq)I`m~v0HboR8T(y36@1M|S7TLA!I`dzR_Q<^;!0AR-5
z_p26HbRNdF>$mURxVh)hJMi|~N6%kC%hxYJUsqON)7aG9(B1)Uef_;~?7cYvkOu&;
zb@l)36sjzfO*6~9*?aZMU<e8oD}mDZ1*`xSR71Ww7D<;Q)<iR;kWWulF_on%UF6c5
z<H7|W<uX>qjD2pzXP?-0k1u4UtyM82i(l0HVy4)|4L2>`#4kv-=c>SD&6zXjW2V_^
zb&QxR?(gcY*EDzQ{;$6CS`5?oInG@0x&0!nHm2MBgZs;qVeiyOw>tLX*1?0Cs)>Qj
zx7^U4G|sXm#TaOdnQd{s1^x;>BrvscUcj`b1*)}#a4H-_;P2PQcpXwL8?-euXJDBk
zU1XSLb1goNWgw47(8kPt2OTggGsn)u;pFk^lt$!tO0=d3m<=<>Bp^szLK==TF$iz;
ziOb|vg3ucUofMCJt1?F8TMz-MFf|XV?O_>)WU9~#zO`Ie;1)NvVdPyH&8G7b1Y!OZ
z$B0#f`fkjkQ-HiE?mZ3mR01OS1M=DJC>K<Vmf8_m;ikyyE}WE*NvyeH{jBhA%j`Xs
zpk!hqT6|ojvcaQ-*0zH>8%~acN+z@h5s~_xBpdFof*PMaaW_H)TN5W(eX6#bM7d=_
zLYOpe#$F*00GDGvqSpWbeFo5@K=L!Nf65~Kekn+L;wBbs?Gh+XA3=+8KuXXugL;O>
z-I21=A|o(fBHFcTZh~QO*CW|v?1X&g50tP%+O1D)%6q7=MaB^_qEtoQKEV_v+o6YX
zg)I4+;{;N~gnlh9@_$LW2)B4;G}h)X&)^g;ibQ*?aaq&ZR-vIZtne;3T6)Pb=4y7Q
zIdVeQMrzq<Yz;Ni*-%8*t5cpaqXiY`t0%n|-qfKfMY^^+!!nizALrJ36B;+Ka{UR$
za*H>w2|1OgjV>^L%V9dQ=|pl)aTHwis^9hg|JlaMt@hMfmTRa&+3f$H*Pu`*Il>4y
zysXhWCJ~|yegEKmJOZ1XH{kb9zBr3*;3Os`6O|MhwtfN6v=-Qj`+hPTIJ{U|JN~T{
z+frZvuCgPXQCa2l7i+NsoW$hF8X<}n&9N2bd0*;bZ(^-eHMgRg&0lsAZ~o#;bH({F
z4{?g6JD1q5f=|ZQnghwwO6dHG{v7=TXe$bviv~LHBMmW5BpSQmW)u>@`4t6vXjOuj
zxfJ>vZ{B&_5KzHLr64a=xwyl~0oM|+x9w8g8{)yj2e{_zXo%}&$C0pDv^up^!g0^-
z@(87wO%dpR8Qh*)*562r-zdjMN_9ur-+Hd{(s9$WIaiI2%AMLiRhNrGw;K(S^p=<%
zDu_;GNzQX8ZtXVvdfCDKB&Vna?bA;Kd#9xL?bDph$9ajU%!Bz9LR|UaFuuX{I!+=M
zX7m@uUvjDwOhTXvg`l$lZ-mZSUkNhewq3x%u9gQ2u*jj9EFF>!s5)8YsYc9eT5%j8
z=93TwRP*W}19;sNdImk9tDKlxno42oQ%#Mk>dE1WY0`Y!5!8Ltj`j)sIVa<1=p~#g
z?;#2u4WALElOH|Z>MNVhzvJII_iWtxbNTfJ{EQrS^S%sI+^LEXdZ=<MzgljEnM~)`
z=*35Vk$bfHe^tjeT5khBwI*77P?h6N@y57$Rzk0|A}j`i;x`4}<NPKXa*l4{VBH)#
zBgApQ%OR>)daZJ@E1vE~6Ei4R$}&4S*mt4zKgYpf6Y5o!C%tGB(KvS^!O1X=hR`(K
z*@CLsE@HvO(tdYUG)m#z$uyieIyV(}Rs!15*{bF^Xz`(|8H*`h;Ur^I2eX>N0!%fF
zy@liw)d3qLEh)ooxOoMv1dTF{ViShIdbm|2`WI<~N{3JMX!B`cmu~M|iCxkN_4ani
zRe&fXMn_20pTy32&kpLR;m0az7w*<Q&`Zkj4N#S9{Su#ef8)f%ZsH{1?&v`=`P`GY
zycbmpu)4zrGa*=kxulo3&Jp#sy&quPN?fsu^M=bU_N>A}fdRh6IZ2@?wY8{{n?%G9
z64Jz0;##w6b4;NQ3ibdtx=juo5^;Yn?xnXP5|8h2byu{{-6$Tn0mr`a!6tq9>2A6;
z*3MmHIeYiVa1(3=NPfS(Fl=bztGtPQ<!;m;FfKF~z>b-1KA@s?SaX-A)$O~sd|m&D
zy3yfCR;b|wQ6jN0-Os8!IVuK)JS606KQTC=xTc((2jpOPcWA2S$cw5^9-e`DbPPNz
zNEiR3TlHgA2z*<Zn&O;u-R^$3zFsVTAuGom*AKxks6^5-ookI7(qv0W-gXzigFa!8
zL@??o6U4Na3iXKP6nRkc2g!^krr>h803+WpR#Rn{&LoS&{O0=93h6#9WHw}30;N2d
z-Q;X)Ii3w)sbG#Hb{e@N>8ccZmjmQ=OrQM<px#jYJ*~|xnqN%L0dQ@ra*a0I(LB-7
z^C+SVhfjb*(Zog%cI3BoxIFDdq>-4U6mhPYQmbNQ3&J#Qla`nGxu?hQEHs%=zn^~b
zz()iigd9Okz!flVN3MxYV%h6tE<2at&R1`Q$U`NBxdZ(wPKLmNL1NGZ*OA-LJG$@m
zk{LPS9Q<OQqmE_q6_*9cSg9cC07+6^4qy?qBsk$muUN^m=zn2EJ1bJ4(}lR8<&0dz
zF5H!}iLIv}T@Sus#N~Cr_7EUAb^3SjKECj9A@$ldSu^_>k&E#rO!eMV$FGeculB#u
z<6ca5$VUz=E}mGtzT>)g$D?rUbJl@|2c0Q2#8&Cxrp!+Ai&$oS(PnaXY_FZ$#g<U*
z8m1G(VIj(RiIS8P^nrx;^l8dv>^Q#0CTc`OC*7KRykUX?lB+%^tXLRDtd~Na=GEo)
z)<7&LWaC1U-+}~Nf!XS2<WYlzehk|4(-0BCmHCtL+k7!DeN)6Q0AP|~DUZ84#JA24
zcAy{F(&B=4iGKOxWxr&>MG$X4UM{em5O6ENyasr$4ZzeMF58_?@gT;fDEBx}W8LJA
zM{ynx^AM@q`DC$Zuyr#>g#-k~E~ww5y7;HhufM*!hC*EzErkqkQ3p%soi&m_BvI5-
z^gQ}Zds)yTr;TU!VhT@3uIJdu4ksV-%T$TS?5OJmB1gh}Q$V2K3`z$dgjDxdEz^}o
zo|t^OlFV3^?An>^cVWY`TbRQ0!R6RM^t8X$!8wC+fV!H7h3x{QSyK?}G}+k1Xp+&H
zg3z+nq&S-8f_L?2s3fXn*$s<jrGDOj5*E`B$9`Lm4Y7)JF20ba6wwiZYR+Fgx6egm
zh=v<Nr(mR<%2l;z`)NRuX$vdf<9o^$Yb7fh2a>|OLziO%ky_^y9i&-pgb2K#aDJWX
zN=T#5jf2z`W`_Mj9I^bPpT2wVrOGQQB}nzltRFeHqgNK(nd?t7ZiCcAF9*<TzGwoc
z>&+L<RRi?~2tyT*zML)s=@A9BgwvxR>*QqsN{*W5ZkAHbicz84Rsq>$C3~J;SDMl*
zdPT>+=jaPv+~|g@Z*Z+AZUNC5cLVC_&-4_EqE@Cjr&>r~Yxxa4*n<Rm*>P82E!jKs
za}8p`L|Un4vHFBVKPkgGmcmlTsc0AHLY8e#pM_R2%n3vqRyls2p48A1sg;`d5PwCH
znLqBX(RP5~nYxMY^1=)9E2XPM+HLJ*`4QS!jkl3S)dIwRYJ8-Fzf&y}_K;BpWxWcm
z;}10JUPp#G3?{2ayw>6Nj6iE+2+j2jm6L2zO=n(FOLZwZ(nj}Y#YuDxuAp+JSz*%^
ze#Le%ejmG(GjC4Y;#6PHZLoQ6gDt8*t<cO2_1Ci@Pk6d)yW^iaUFpOZ{A{6TZiBWP
znR_jlBEx6ZEr!yJkMg}0Us{mVZlNDQQOoE<2Rr&Ck!BNe!{lge{=~JYLc+oo27Az{
z2fdJOCSJTp|M`SOYBHf4f5@1U<CDpq<RJrWhW2VGgxcJEQlSfqT^%iqJzRNsr>?r|
z`@2p}k0EE9mLO3*Ag~z%hE3Z4qV9Ha`m)N^x-q8apLt#bXWvi$XTI0q*<kC@58te2
zJRC#(nkIR6eu^{jWD-nzeruH_{q7jzr`b2)<C}>en!qn7Kd5FLQ4<{~->j@-4uR%>
z^e6xqWW>V*N%7+5ovTISfC78O<zmzmWhfy#NKVA)fRlGnfmIo$#UKH4@$a*zl4N<G
zO9+3p+BzO;#q!yL_^H5h_$AO{OM{hcPHuDPL@7{xL|<1@-_Vd?S@ffotcem>^gd%j
z(WMN^+?A;(uey7W&Ua5gc{$L1c!Abet!-gTw{36Y=-RlF>G{z6&i#*`_w}wEZ~f`>
z{+@#ilvl2<rKV|Swz*p0%9d=KHwMHP!rMg*Ee!98jVm=2_Pksh>{*>{&4G%Ugoad@
zyNSWv<ySeST)su__MUXDYkNJU8oPN$yNzS)<fha|)Zhzo=k(7Gb8<WzQt1Uryf5*&
z)J9T@JM&PfthSupW2mt<IdK1(CEkZMs3toQ?JT-8bkHP41CuEP$j+JJXY`N>dDag(
ztJ?W?Av`3RiVihNO$WtyJfvH5LJUIf9P3o(4?lZG_USvrAbKZGrJ#XZr9%frcTmoO
zO%vZSHFfhbp&quZmI#X9*GJQ7%htB@1I5{=e{%yl`Dp!FO$3P6(bAP*x!b04H=v$X
zE39SKEv*&_MH$ld<hZWx>@<FY0Ui%Z!IfsDqzhRz0eUJ7yV88SImD$9bTSHj$yoJ+
zqs~Nr>P+H5tDIY75jD{1Y6{<;GD7O)vHR0YDBYPQ6-hb|6|yquZp2rT{l~@P?4=Tz
zP+k3pvJIudTxcFE5sqxkEMR3CLfd_ec6@MDQFUHY7dije?VI;{=foB7<+v<pw>dp%
zs`YWyMkAjutE<PgL|TC$s3w^=(8MdA&xR!i@$u;=VCAd@EO)dfIe-HplWPdY{ZeW9
z`Eh*aHU~*Y1{oqm#qUceb{_9j6jSPYqv=ukcCMUn^WIc;us9>rtYFU0+GF(TRK9_k
z4&tD3taM)ZYM;ygcAZFh^mbJ%{chGY&_e7S+u*<=opYISB|At8!%{ozK6S44K#Rkr
zJH*@jPVVeSrtlaU4Vy9=X}*V7=v+scgKXduKBv`33?&rBH{&kaeLg{1%TQA0`Wd!!
zrylqJkzV5B-Z19w)co#Cug*AB%)V5j`SiIlMg6U-O8M&Ze_Qu-jjeWf?pp0??$bB#
z9^j@?j4?pNF`w2g)~uMssStFy@x<}1k!$mIP5gLzaqi*n5zYMmu9(UcX)nwC`?2MQ
z6=b7HDXrfjQMb&h1;KeQ3ynx1+*uL4Ypc=&v?V{60G-x|LI->KoIypsXotSGmG<?f
zGCT5XI4Wj2l{mY4di3mH)~AM^{&}jIZaa7SN&oL@r7rf-!sb4;Z08_1t*mSGSa-+b
zD(xD1OUjQ%o+K?rLZ-|1p|7qICy6(Sqjd4^hS+l|g&gXogtl2If1x$}Zu{8Er?#}&
zzw5UG+mui*IaG1{<=0@LjsPu1L6(T=Ffimyv|Ds229;P4Dyk3V6yOsQiV3ky-#9lC
zz)RYcYJPO669M6rAc+wgK`<!|nps|fcClJWW0QZFZH>5l)4X_Y*om>WC|mC=fZY&i
zI#GC{jNpe5ws&df&i1PspI*#gYj#a$QzGkY?^&2Oifqs4o0Sb8j*tHsS8^V6{bcd^
zlXFn;%E_ko2scCrsQY%_rDLU^7i+~mbUu|s-d|86D(U6t?_Nd40RQh|6aO#sWfiZn
z+@Adu_ls~Y2AI5h`9N3e_NAfV`T)fc%lz-jzwKq4mlxH-O0(Sfvsm<*ZBD8Z6)?JF
zJ#YJJ114*{j@@z3ojrFg!dHHdeBAJgGq|tlz4GZ_zg%9JD)g7&fR{(a0^muRj2|6-
z`z2<+gK<|oS_wFD3Nr8Xb=kXp?uL1{)rWw`A5dEo&Se2ep-e}<?k8^Yp{!}VWV}kr
zKW*FEKzm^$mf8qnD>{UuRpvbHZVTV3p8osWWtkp|l@NWU59`wik$0Svy7kCQ%Vxsg
zTGfR^&P9J{KyS;_#ICMn^tq+cQoYdhDn;yQU0dpim<?XOeswK%UnZ?i<qNB=E>ti6
zT=K7|?*q$9%IfXmBUi=-8d`i>b2qyWTc}xy+z1Kxgp-#=g|=t;mczuMjrwpGecUq6
zrPN}Zgk9y7Sfz3?Xpc+g>y?|<EKIz(y##aX>MHx8HpitSD@4rQ%l}cKdWZX#oO5$F
zhpT+2A6RIgKgteP`5?Q$4Q{Iz(|PP<75d4L(xh;<!WSs#>(Z{HvyZo%Buj%6hZW6z
zS`?#iUi*g%u=+`0e2EtS5jN#|PNUoZfh;|QUkYpY_;@=aQi38?<ZvS=J_ocsar(ol
z{BPmh8`j~L-KO<9um-efn`ZoAhc{=ll9u%>jMcHXQxNW-a{wQY3zy|?Vx`fGVz}92
z+0N5N;@Z1eVR1onY{G3hT{f@eK|*tjGKdgVASS3fqnsFBhur<|xl_{j7=k0x;M;2J
z-Fn`cZ_qdwgOfAy9s<)R&ER()87Uq6O>;x1&IIjn@JQFhCEy~T;cpl2{J+-hSZ!g@
z`gM`Dq}dpq_<wuQS$K`9C`e@RVdSKp1i2ZA><W7i#dZ%WLWvcs#%#&}w%+9I`#`)j
zLrUv31wnsu4|g;`-PgsBwBuo+R;>SuvjN0NPiFf3md666&aE<CANE<GjJ|ptX!f4m
zZ5Gs&ZM-;#+?fjdQFjI*gPzIcdXzj&9&t{;TEt2QW+ePh51XHK!KjmsC;dGaO!K?z
zm=OWrL0>naKR?XNd#We^t4xNe+J^`(;{LDTjw2Li7NQ|$N1-0?>8E&_jID&6Z4RM3
zBWt3+urhXI3z{a(dI`FMQQRdFCOZCszZIV6xjvk!_veP6c~BFSY`*<*v$@m$yYc=$
zRjHr$bXQs)Z?B4eLt-4+*Qn6f8Ve_R*ejaxv-F1LL3T*v5Uana7e|*RjTTHU>@(Jj
z6J13cdL-GaKfTWkQJOpLei#jrQmWP$>$`cWRL*!<tSqU}eCy-l3;n};C2AF?HsW+w
zCQ7iZX0Gilp6rwU{n%_AJiMcw6G2g0+Z)+UuHI*CtWR_kZRnN2PX7F{BhA+G{+1*$
zze>EvTtnROHm-34Ai^9JvO8#cXiC*WfF;%zB^T-=ubQrRv7H;}VC`~9{E61puMEkL
zEl8tOB7H=ZnTfBS^jWr+r2~Bd>g$e<80Pfr$4hCr5N2{<d=<3Kk3YG2yMn9#f9u_=
z?|YQSvZK{~V@J8A*@^;{UxAY#fa5I@zI~n3BInhbMupX_Z4K<9(GRKIhdFwGA$qQ7
zs&%kS3O*&Js2UKrf@7-4)b>%i@!lP8Ey6afpQA|wV6T88H<bBBg1#uDuFcg&)ei>@
zT8H0|*^Kk{<a%(mejDMDPvWno+1Z2~>OP}S^&5Fjo8x$^m9uboE3$brdC+4;@Jny}
zJDaq3vZBOpNh)NjW8<RU-&a1+Z)drI<cHg}eL*_dElO}gOY&89DXt}f`Q@b2_HMb2
zr;Xo;tWqURaY968ht;%8`k$=xE1$sMxL}-U+1f3Naw!&LD>-M9D7h23+{~Vf=Ocy(
zmH&Pm=s5c3_Y*DZb!&S&kG}pzd2nF*P?u;~L%vWuIC-o~ykAegj0}09<JO_~_vFK4
zni7hp?|x8UW0m|cKG>@$uL)F7obE`&Tme}(8h25hoV$AIoNkxhkDbkZEfgcgt#neu
zkATHBCl*cLQJaqJZPK>YzpctZhxQKk9v^Zrp`@Rg$3{bfTl1qy(=D%LwY_$Dki^V4
zQd-D)DO9W|L6#etld-uVq3z_TZr03*Xq>9l7xN2pGKffVk|askz))Y99XiCYMWh1?
zr-Nc2<i<X8qlj^hZmtC`skr1UI3vWfo(Qb3?GUiS-HBqrQ`VOeT#`sU7U&5qGNb#l
z%8@l7#ZQJ$r|}k3_nIxSuFU?et-+kF(e%(B3D%CVsLT1L>pKbir}z*30l#dT%l@U~
z%ba<n={L8ItIU#c$)NoDoBP#!2haoI54L3Sx)Z@=dG$1Fx#?&&rnA3bZ!l+XG&7x(
zKOzox*A-J}Np9qwuj~clqweJ0uf7n9qr;z2{Q4al3z^XPzW8ah=4H!EhqV10O~I&a
zjt@rIH{4^t(E_(bl$J1-FqnYfia)Ti{FTGqW(_8OCcZz3E-CJF{O)R}1<hUPIPwXS
z?Q<Ls!DMl~(IVq;ul>e4Dj8zTK4unX2)DeE@oHnWqa@5E^g$WYqCN+}ctGwHRzC(b
zp)aB9kGK7T6j@LMTUHod4teJf@ehuJsVvPj?1~P<Rt~~=25hIegV9ZeY*~I&@P5-F
z-|LTU#I#C5rnM3Mqf(ZJID70u1MI(SC4PMb-dtt|M|M*u5(g%Y>c8L347%vYo+ahp
zsr1KYB=TTT>KDH~PT;rIK`lk%WDQuB|Cv<i39R?QShhm?A8WUXX){H^Ki|D8nYmX(
zI*SG7cKFH#elj2RM$u7g#ub-Fp$K^5(Tz3xJGt?^_R6m2PLr>~f6%e9%Z_1mR`^It
z?9yRoLT-tG6JEmz2=ClE*-W-rN<-&IVaz2OFN+oAJRO6T$#99)wHT;Kj7Cc&u?0+a
zJXWp7V3kT-VhtM~-`I#oNhC=nmD~h$#4<(L#-xknW%yzlOkrzFxT7152RzKOrXzjz
zz5iw3Weso8<5V|#&5uDVtZ8-ejw+A7a57+%aG#xZi}%a4X*peBGQ;q3w18Z6Lg1yG
zJkW_;6g;pr!7?N_A_dB%z^g$NI)>ABq*s4vtTMHJfZ2hqhojwvePQ1=z&A|_$*@2&
z)G8DknVP^P!Kxyugj_bMQZt%eR(=72JL3(~mEVFr6(2r#vgFSfNR|6`W7qgu?>y^b
zUG<ik?~w*q<mpHn2;3G4Y6F94osp4~ny`saolM5a<VO!CCLTJ{^~aPghk1WgS7pYQ
zZO>7wk0c&Cn2@%zFze?xpXWPYyoDXJ{=Pf%{QfxH31!#rEbFGFhtSBWf~MXPz0No(
zZ7hX*u&HV7vs&_b-5;&r@Dz`NqnB4HQC^fzd|M_x01NjNY%EL*_u{9KYP}->Seut^
zIho`JF_#jL%YY`;CYU1boF>*y905)HYauiM|7r#<h69U*<$mp{EZe$!2rDWYSZY!E
zA4Le=B})D%I?adw91_|+{H!uI@$>U|Iete%n-?6n(`-<(+L~-Z!68tT97fw7HyfNU
z?Oz%#*u!++aQ>pIZE~o!<NTSP*0HUn%CZIyrK-zVNx_5mqL^-g`gkWu`04;aSeYx#
zt*oD<Ia`EV5H8V{MOz}e=bMW&IHg|)2ZW_l;5ju+gl};jUdtO2FvC5N@rdxMJRH=>
z`URPDgrCd`do$r64cmrr78n~RmZ`E#zVX3fnu-h)i6J!ONxEi@YWFjfk!WeoAr_%g
zo4g<=r%kC;r@)xJKqn*JsTT6oiF$6(O7aX?3h()F*uR-9$dMF_^U+9YR#^~XV}2S)
zkzAUTT~(9nTP(_!XA6SzN;0BASZr*`=0B5ZGFmZzMX!PxLnuB&VQF={lyyPfFDxjk
zVp7{ct03h6FW-!a&@$Dcs)5omHbpn!&<?s3weHO8%HBdZgHEI!$|bg3!78V~^ar0V
zJj@~O8?57%*6m|6Q#CO%hfJp@wLczV1DeaQjB{<VPLdfXN)pn6<;?3D6>nfsCDqr@
z7R!c?Ul?lQTt8QYvm@tNlM-FOkZR<iX2^}&!3J8NAbw1k!vNtum=zfdEIq*SYxgBv
z^RA}hLIkjIf$p!iV7w!TeLsur^*v){UE@ddB)<7LtmOQh&n74%E>7@XC>0Ze3GM3)
z($nx*`POpGmWbqTR$&?whuD-F^nf57@bPK>YYiek*Rv`gmt?H|d69}w62G3%lOj{G
zcq2ofE)5+7Ai*3WvLs@wAc&NbjHFYrF3l7!NzQHre)@Nog?R`{MQa2iYr$Je2ALlp
z_G9~k&2j4Dx@G0JtWOPvgV~AR&Gxc1RkLq7e$!$l{bmWj%fseJUHOOiU#w4%Fo^fO
zTeI~^B9~LpQtCAYvuj5->+6YtkqCSO(nFsJTtd&D9LTRHI9cWC75#Evb!Q_DPp{2)
zcN)^v^c1@NE*@-Hmu+CPa;XIw=>$9sSynt<p81av(28Qoj;{I$n|-Nkx_8zV$y#&G
zu6@y^Up7@RABPS-114JPr?CMtgFf+~g|RQu-rUZP*ZxefrgAye|HCVG#_`^s=JZ?s
zvE%WPUjW*U9X~xGzYqMVIy+p|tf5eQp2#3Q-<kd`-@9(7q7J)0k`~)my~>i`Ixw=b
z+fM)<0xzBu!rk||waovuQJMav@c7~O_0VyS7$0%((c)JHJGV05_C2(F{2sN&U*(H6
zqxH|D-B$($vVda$s4`1WL7ly=&QA4^I8(UJ(BB^iGPAwT3P5$1`8>5ZO9b8x^s54{
z(7CT=o{-NkGP5w7PuEj4g^b60MKqJ9l~YW95iCkO?*L*78DHs}5HFR-mwH|l$S(D=
zci$5|*r3RxzS3SeT#CMTdFM#Q!mZe2S4!F;p`xbZ3Hd{A-%|0b3$%HKqZnNLt=0~1
ziF%HYd)nO9AT#%^U(|u+YjQC6n&TC{bJiqa8g#x!zOmJ+JnM-{c$vLRQ#vY)IPdYL
z**@DUSjym6C#%tqK>vBSM?RI)Z(N7#ok_BLaN<Z$UH9SwrsKTD^``!GTKE2$w)lXO
zuQlDD&KsFWp^(s`JPI<ZI#qpq&(Mw;Hb_dooyX%zTuwn0qzIl`fPiN~Xo)F_rMNU)
z>ip1Bh*dS@^We*@`2~*x-t{N$%mqvd8XGjl^LOE)GhxRH^{W@Rd2@dD(=C}tZ0B9I
zaz4&<knM7b7}FC(P?zG2xx};S+09Zyp6hbzda{)?LzvZV?`Okw=JWB;v{r3LU~0D{
zsLQeEO{W*-@V2vJKq06$h-l3I7JhX{9jzjbfX_xyi!!gr0+)@X6#i!TVK8|yXTw&3
zgs)a2#$$frh+)9RSi<k_dvl5om-7Z!b{be9y4-s6rpaB~QM@h{X)iU$3--Kykvlax
z;P(j<p<t0#!1OKzaKE-H;#*r@i%k;oS)wKL%Djwf6K8a-*`wKWmwHT_4!zOM!DNBH
z)PrL!EVqgMOf32f%RjCKh^2jD>wbDznT}R|Y|h%tG>x<~Fl;-`FedvxX)|wu>oqWz
z55gRWjsuY2YcKkW>@J>`v;VM*@U7jdPfse9D~e8=prJ$SXx@DLf>!GJc&`E~b=d19
zj`1ReMPG)z3bYCnWA3Tv&Q~^`c>l8MMA!6oL&IVR;g&RxdUP2>*?QsikF7LsS9WdQ
zG5S#&t~n&vzO{C_K<IBLX#r^rn<r2^5+GR%RM6yql{`KUTA(>oIIQC<{#-QgP({89
z{x1fAJMkv<6bO8Vnjt6cqrN6)!lEE?cAc89#Z4y%RsVk)?9f$a8xKf&4_%csuh!*V
zGmKxNi8M9Y-OI%xs+`iVnE2>&@c~2G6}=J_%@Xa`mR+pXB1rr>G}i@SCIJiZ=$pxa
zLn~(s@bGxC6JaF=T8Nj}9F{)(_(<1bt6ee->Ve<yF)ieUuoJHL=L6?RuWU<u<;j*W
z@VmSrQecQaPSBO^)B=AkbFcTkly#J=HiBjVWv#(iruj8Z-Om1pwvRucR_!77+jn3W
z?AXe9tCyqSA?<M$xFNg#hF0mheuuiQXkwaR!Wa5(NxMY6ItpNx_dF6Yenr+_*_Qfv
zjnjkQNW5KbOib54!(~oqHw(BCeiXpAweY!JTdTI9uFQtHbEz}Na5=x^DlN9NCp(Sz
zTQU^+FATFqXlSur-C3FZgdOl$P%?J+#lw^R_V#Ssv$v)mA%R@}e*A=XcpPY7Q9Yr=
zxGB`w4j?>$y-(zk(o-^oY#JZE2POUO2Y`ok5x#m9;u$37#>=w`Fyy-8`~OPLP)n|_
zzh8@nfY)Mf-if^ty|Z0HgGJ0_*(5OyMnc1uyo;}OZsMeWdm78G?g_$*`Ayo()LUy;
z{|IJ9$A*&kb9~#qo%T{_6QGUxcsE=VfHsC*vwV4kaxD3!%haCjiFyAl^m_|P{`)P@
z32uv;0nm0>Ee$O1QdEL0h2>kitS;T$<M?j2Vz~l5Q$6!nzwv+TPU+pgET8rNFI%V6
z`3Ggw1EsTX=a|JNDO*0UK*F4a?TnJy7^X&UXpIjLhSg~R$k&m2zvx)&FV;_f7P-bf
zwsXe@VXAfbYR$l5o{R`j6cj>>%8ff-12?~sqiAdHFNRP><-k`8xDQ@6Ve1SqWWCP+
z8GgRi6#;WMt(C|DdbLZF=gi!Nn}F4$H;jq>zkgEs_upt0MF#Yihm5sZefY=c4Y+>x
zn5cq-1cD>MrPKliG^1jsaNT!8=1>+B%H|wieH;WxI29<^T1`=$5R7^)uaZdq&}Bem
z+ol0YG}+g+C=ne$0ai`J7*PW;W!_en4gJTidS)4DW>+aVKzy{QH;s~uMPvmXX~8XX
zCXT)tkGfJD$<W1RmTTSO9iw((G(I_d^z?j~HSy-xDc>$|c(O|OfSbu-^tUwHBLF*}
zRu=<jBe5qspBRX<O2SeL=c8<G*)W9tu@wRt^CF*l=(h7l?XG%Ry$urW<msJr!JG`L
zglTeSchkd8czKiw?5jMbIm6{Tixe3VG<msHCFUKj3DqLpk-*ZyU;sEHPJPEIbnLOv
zdCxj&^hlLnD!DzsneR;ZgDhIQnMSWK^4EaY?asUz@2CLrOW>B>4eT=q!h>Wi1^2J1
zLudICXrJg#jIG$bgu+&8G8}uN5+Xek1@pUwky_>4*FzRdzz#?>zzq2_Sg*-o^?+fO
zjMAKi0E;k=Bc_Kbg;Y42;4gxfs;M(#YZQ-3Tm;8}9M^O--xR>8pmj*FgA#6Nk`iFo
z+EA7ypLx2D;j6$o>G<9VF0BkI;E3Rs?+E`Q>vH?k<EQ_r0^@u=6*v!sO3(05^}YbX
z16GaZIs-qX$2)UN(ti)tJF81f*|Nkh4=r3(pLMgisEO^9hI+1I9Br#a(lRi5avV9p
z3W$WWC|N*oLB+DLD0CYMT4E#|7;3%tt%v3wN+|&dmXh{gpnQ)@H{QCd08$kUBIPl_
zn&GuuI&{P|!=hI}Slu~VA<Rf(X>RCTFU-4*#Bhcn)+bXAT^MQ=S{Hy7rfBvu#~o|z
zW3=U%24PELOBV(UP;>Iz+t7D}MB_V8L<nvh2xJ*F*j)5FV?C-Jk76X<;cVfR0mLW7
zl0mgdK`OFWTy2E(&>%aMc4p?*EQv2=q0_Oh>$jbLh^26lDF+4$%0-PB&7DG$n{m<F
zq|l4|=$XtMyMVX4w{L9929!cmSKlazW69x&_oV&lHl()L*ikLQHJDQjdLN#Br+XlB
zv*K9}h&F$+#eltT1K2q>)}<Yp7&p)nQIJZRRX#zPho)nw7KKT=$DQpjE3P$`l0H-h
znv_7toX8=_W!*uQ%uO410#A3hOH}17hbWzmqh+YYO#+U<*qSp6pp}JJ73TE16&FLv
z*SLWGTm$KW^@KHWSd}U0j_;pVZFi{&;T^p(<^1;1?Bjeq9o(AjTMt|uR;dNDi<gfC
zijL7TGXx*T*Tc5^TM~mgNS*`aXfhNAq6@a=0lJx~95&JImMa8_oHckPj8eA?)P`;4
z1VUyW9^}<ETP3`uIdgPXB~VxMy%eznS32g_4som`l0~GpRTf^dc~%iMrZ-w1DI`AF
zLUwwx0pP+8<Q>HwoT)C{3@<2@mc!0ci_3vGeBSjoAa4yT6!-KYyn{x(>L;A@<y3eH
z;#!vP7j6zNE^iZoq9-Z(hp7u<%+Ej8_H<HLR1Wtgj+ja$_&0y7@s^QV65io8%$?7(
zcR-YayU=k51lK%cjCU&8+28q>7d+qTn_eC0nn9B?J~_S{JJKAC_sr_y#na{Y#Eqx7
z_0#7_Vi!F+&%EWa_wOgG;l(0ds2bG@=p{g6Q2$QavI2<&<z6QhrL5$*K{ZD2eD&oU
z0i2WNfK=a0f)iD*gn+lzGs!tNfz(FTL%k&($EuKlvqb>^4it(oCB=(s>`vQntvY1v
zoY|Yk0<(r$0C4c9Tj$XCIPlzEv?~hu1A84{el|KWGD_|WRaSxGf~#qtR3sw~|5C_a
z^a!%uAaR*Mw9z2-N;J@pvvCRH#po6><ACheLr#ho+0<EJeD3eNCTG|n%KB>?;sR!;
z=pCP#jZWJHV0#Hil91l}#xNb9`<K4R7z>9m@wqq^QYSd1^!U8GQCjb1kwjW8;Zot;
zkW|3JgjHW>|EU)l5G;)!FI)~qvzm5*nngrCZ5wPBW=ok=Bw}qNjZq!3!~J>p53749
z={!(UmW3ILvJA<AxW79YN8<)YK2-_eGJ7arr?Y0tSlbPPov}c+ak?NS+!*!4IEFI=
zied_Yj0hq#2pBm7V2A(>5?%V7n?3%y_jjb$h@;-=b>g-W0@C*+c{#lX2`z{VDV-Ki
zZd2qt5DK(cO@^ea!*l$nNTopBXGO&}hINiL%)98bzyGfQpcW~sF<u_nV??OA#FqXN
z6a_^u#x>1={Ln-h%&<1?`Z}<PJ=E{`28ruP2LnPuWPu>g70N?W9O#mQTZw?h1Xej`
zLDkYYX==boWYCzB(xsJk{i>nAh#&|Q0SJ{>z{!05Mw@;Zk31nZN{8-X(Z@b-$~pgS
zOQg#8hUogdld4J(v)-fb)vpSFvdQv2-Tg;RYDQ%9P`v0JN3H4J%q?OMS5@;wm4dBd
zIwpzf$^x+s25OC}j5S>x06-vILqIK$`~MvQ$fkyucLb)#nggMLmUlqhzjOn6tEg>S
z!ce9|tCuz|am&JoE<Ug_$0$kT(r<uQc|RtHvajH*1!E#vxwM_{2ql<}d#Gebvn$}B
z=(KJwLP+2o8%7R^DbEB7X!$fY2zL}okyjce_b7r0m58B%Vm#nj-nc@WVM`+I@Egml
z24PcKhlXY!)k_=-M;p1pp)o@*PMuY2yZFz&T|x+Tr-P~RkifYvPDMTnLlHCPU7dis
zXqiYS9CIaQqV3l>bu&Y~$Ls|c9|P<Yu#@Nc_rrF2aQU!NpcNab0G1|UOI{hOY!A_x
z$dmxM#Z8=Va07zKpa9ZIdISte0+*LCATv^&k{p4qWu2VLnPxZ<D!{%2gJx0YQ#!Gh
zyMA1kuM`V1EH4ksU|_L`kXZ%AiVnH1LdgO}-ZF4tLEfLtCgIS@TVDU_;cxSazmb+7
z5D20Q^xB}#)56*TBH$Vn4nnsJaX1=-cfWFcL=oN5<o7D$uL>Xma*Xu<6?F>70%_hB
z3D5P3Mw{N}nmi3nizI)(E)Y}(P`3e&C{j@&3Pa?5QxrNS?H2!{&E`g3&0yUX&{PG&
z9AjKROyuJ&)Ub$o1T&apgMO2Jp$TjN5Tt=<Adn5+CKU)wBr84WwGOo?V!+ti7Ua8K
zf@iuL>wRVajQ6DiW(5%ohFb^*P<=Ce0Tb0*8qFli#!YA+2EzkHv;)=>Y)5$k5~{4V
zY*1v*N0zNrtg&gH1r)ubi0f4kTd2;n^k(bUD3D7a+C6m56XzYeXc4nHMq%-ji?n66
z{Xk>X?LOWQutx%C*m~FXsn3UQDVOS1)H3|3F3L)kIDKMeYwlODsAtYL<7W%gjU9cc
zdhS7#ebp??5?HWHHw!JlW6yArbKvT+Dp`Fu&)veP5q8rx6rsWxiQAo*q+=KhpuXiW
zAr*_#pxya9AOC(Sd`1(c-VGb~B6RQ_EfTK+*u2116%Q5|SAi#*P=*tnV>P<^Krp_F
zSy*V4W2%~oHMoU%**Y;hG@gr%q<iT$5;<Cq+=Dx;gv_WlB$QJr<CY0d?C#U7EQ{pO
zl3JR{{^G)k<NpCmf38oGmMS%ILD>i)M~!zEN>8bF`%YP>KV*-Mu?$R>1#>WSX^giz
zXT3j0LJ22`L1N%pMIspbrj(?7AM^hFa2;0fMnNJ}6lsKXm<tm}Gm1#JlBIxX5t67Y
zvFor3X9b3QsUYLwu$>n1f?IKrBreV&1(3WIyZf53Lj=a{?Fo3!0R#D1gfm_-?5>^}
zEQxK4oVX>2{1$r+RS+bnu0qFE<qH#oh%hgsJd~gW5^07<2P2>P<5A?T6MX02uecBA
zcXudu{h+>#cPI4|ltcFnUEIjcgfdsE`cOJzTZYRAuiY`BXXfQ_{*ZGZ*JrQq>K5S#
zIqNxc>36FY0iP&f*O*!dH6d5&No2<)JRYffi7$nY6k>PF#6YVaGP-KLB6R*WG@-z8
zj{dqsi~?w+-l+HS58Y3uqTQ*bATA|ukI^2!LEt(Xm9--oiB|!jcHG$UVxKo^WPDH(
zdaM$}^UaPbI!w!9lz{mR!3FXV0R6B$29X+RQ<BBM>k`gd=^*otAgX~Y4=H44P)nK7
zmIZ1GtG(Sns0PV;u2h6S)wsaABD6uF%p~rW16R%@4Gm-?S#IP(8Whtkmc_Fjr_aHH
zE>St$ZBW|(A9wVi>h0p&j3aHO&AXM#!7QVq91mq4Rgi8};-#eLkUvr040c><>rP;h
z?1`XIZdnBhWu@LEWIW^#F(H8W4kg5+<D2uL#5NDq(6Hplqe0KP+p&tGo_A4y`kD2v
zA!?_064d~H#lvC<A)$6I-VHX@2jXS1J_^+Wys7k{p)?oO<b&4P5C0hgZ$z=RaNl+v
z-4aWzAhHTz%NDtGcyi4#D(H)Sx;9Npe~Zagiso3;>v=?moCC`$|3GdOxNVwLkX)8Z
z1vy8Vk<9{9jrF1c*Hi64VW`4H<!t<E`=5v@1$CDcyN;%XNFlNc;6g{rPNt(hYT%H1
zQdy*i@eVRxbeRmc05+Vu9Rg9(He8fwj0?;Hn!Fz|5kr`cpBxzQ36R1RKMaF~&xB)E
zNows*@UN?z8_{LwZ#q1-(IA+p;h^lLmRlw`>2*G<ox16w3I=xb5r&!{TGTA!^^jms
z4-j0jbodPk5>Jr=MHz`2?oCeDA9}#@>s(eAJYb{7Rn~y$l-*RMkXsyFKx$f8FdUnB
zf#HY5m}4_??4dwmL>OamroVBUXgsLIEaLbTGsJE2Z#qKK7JK$D+a_6IAhL?pn6R-T
zeXOJmQ5llCF<n3x^=$2A-S&e)J41qKxC5Mmw?!)M&rU~`Q-^B%Fwe$Tn4v>EK6;mh
zFrh>;X*<O1x0xUcg&G6f`A$}?gQ_fB01*`qS$0Xf3)j={ddsID-BM5u;V%XTiJ|}x
zHodL+Qdu<r)8i*UPxwYkrIMQ$Bnt#754wWmRO|w&9yPwzs3~W%u_>xz*Sq>Gj)YU*
z(vy@=Niq-Pgdye%oGM2eYoQ;bz7Grn*zn|jHbx}}29Y~+cLKW3f_w(83qjF_Jh;ID
zry_#Lg42@BxzZ*YkOhI7b}xTeCTllexHnk$YdNfs`BY>7e*cp+|F3k_sTF2|A7jy{
z<S9co;w_-mcC{oM-XQNcU`c~VvOKVMjR>{MY6~vSV@w&{!$&InzH1FcFR`HlIs2dp
zfM|7;bOFO^3}4svSyy7Tt!pc9dWcePtnH(>`<<>j^UyIT_+``3?xSu;ZJyBI`|z2+
zs_!P1#jI-90%+Xes_(D-?Eb^A{}vu-&Z>74QkL=M<<)_B0mAJF@z;d5`jY?MSDRPU
zQD3pI^rtYbzd_22oCQ5e!0*!t_hBWcgya)Dbg%Vw_$?h1KtfMtQh$i01U7d?7bY?b
z9d$H%UU!9#0i1=-mnRe^j#kesk8GI-z+IKJ&gv0*owFt$-rL26sm?%xl-{-8H^LkE
z@Ca%!A4a+24?{uJYxMP(Y!v?dAJGrqfL^YX?#5F8=CNj_#cUNvtIzf#01=H5ZV?uu
zPdaer`czkw*egB>)Gp<$I=}`K5&*HOK?W#(f~07L1=xOc(Pm!QOEKJRM_MI73j^M$
zgKVM3Jopq5rqesd*pEf$t^Z?f%hGX-RGD|3aK&2igbGh0-|Zzu6^WjrvScz7V^jDg
zC&S51RxS_Fs!BG$G*&oKqaOexyVQc2NyUiZM%#xKsa*bzGZh+`+C`Be)|Ap5RLS8u
zwBnX%iXPNq_k&M)W1Y2Z<Id;Y!tQ?B<N{roU;M_{J6N^Kex6jCo}D7eQ~;Z<?8X9Y
znywjbA|JP4`3+Yl(PtOIv_vseB~1bU!{(;0Z~e<5uMC71ZAK{HofS+ylFYgvUs;|J
zkETiaF1;?`q1qkt$RcgWt!NG~Dbr}CE*Fly$A+Flj&KP(&qCLg(~Ow3rT0stlEt@>
zT5kzw&&^mmENx%>K7xDn;h~Fw`0@|LA?iV~i@OlVJ4U@ddw3TXnU>Bi-@UipBd;|m
zo(W$I4^Q6sG&MUx_1y~uq!<ay#S`PrZU8hlN@Sr@4^oy&rtP%vIwN)xD`KW(TPip$
z8Fg`%m{IFPga%{d#J9CE_}fLKTU^F)Uw&`L{>j<J<3`a^c1i4e!;bgDkB?8n<*TcD
zv50e~S8RR;*y>8-K{^Z4o=7WuC*@E<tDwt+Iba+eA`kOUbkP@!gupv7rgMQl3|^4&
zz;Jkm7A(*}7@Y$S<1$!SYf3b{gN_M1pfsE*F4Kd~1>F^FTG$kfn%PihcysL2I_i?@
z8f@rgU7_CG>3Xx<s2xnjDGSaoy+GG^cha16pR*3=R?u~C+r(Ql-DW))m8f+NjQ3U2
zD(e!<*@%Y=?A;hp;&t%8&+3j6gwv2dGjd^SIzgkzpr28i{6u;qA@oqx>2payow-VL
z&!_|mb0Qj+9KqP7jR>u`8x=~uw1;tX80Qfmv##-rBq%~_@PxBLYJh^K!kVSrDs<iD
z5&?9HNmA;gN7~b!;P-hX-ioXv1mK4&w5ckUOnnw~c>wI2$Rq(;k6J;zg)l80CCl3h
zz&a-Um`(m{@-dl~y@NKGHtY%0CM}uy{gomCT%guw41}1)%UZEW8n%rIjJ63=F>Q(5
zv@Yx{`l%U^dA;WOLp6|8Ewn>+_o>hS^K6<A&!GLE5WTelKyCxf&ZpH)WuzWu)uCvu
zVJ}y~lujzr>O$Bq?6cX&QhJwWY8nzv!*yL-GSZHdsz6b+bcmF619Vna!Ch}mOf9}-
zPucBYEoQAF)?rOw?jB3#5e0(8QW!X57<WKHOlls#_Rmiv5=6PYEW8?Bat&^7%{@#=
zZpLFjOo}W@_3Y(;<zR4T-ug8db)$P5cJ_ycEjBw(nH=`AA|hSY8=q<#E;T(oE`g5A
zCzZ<v&O5H2P_7=ru5FuBkyM=(byLZ%oA;QBb+jaOB0QhrwnQ#nhK}f$>`9oV;M^%^
z&7-?hLaf3$x`8v;b&hEv4hhN4_{C!xiR-AJUqv@_>~>T?eUa;>oF8><pLl*}u-hsL
zXE`*Yu8>#gnJKg5NPhJEON}Sz%ybp_js#)dk<55JG~!9sE+%@gGT@!L9ypyvN6alf
zW=Xz&=KKIE!}{K=fTO9?tJ3_&d$~QmoV4&!xM~1>Msc;fw?cS$BNQf>9ltixxzM&a
z&Bn#r#TTy<cmMk2NZjCaBep?!o&DAP^M5GBQ_g|-wK~bv8!}|=-dqW(Pgns}gdu~P
z^cEhnaJQ}HqWy5y!3Ho@bv4c%-!EiYl^0hhx$&sRqp>_GOvD^X6nlqh$JaPsHk+=E
zEKbWsL}c6=C}5yM;GDEz2gFgEyTfr?Y)1kk=vJ{Yu)Ft&n<)unDnDAU!-_AIdr2P%
z670?^ip9d_YC=9?pMdEgtpm{t=)%?e_{W|%m7axXX+QuJ?-NEqj%3{EloUSMsW-+_
z>E4=M0zE8b4dtsK1fiq63iqqu05}I`bO}VKkB-?+nH?F+Rc|>nw~;7sGPDI3zC2Kz
zRLPq2F~yG1qT25FA908fzE%9;jb4cAiE(n1QYmRjIH9#*##7KZvSy?uw<nfdtW{{h
zWK$()s%vM9nIVVE<5ZsAn}}1OYWn*F1yLj`@M1a6;zq^%9Zr)oS+3q6Q!Po|03)5r
zS#B>wDicij6&TI<oRb!SLP1)}AE;_|VUuby!=UfAEfPzff|*#wpG7<niekljV6|O|
z(02<%)e-E<dQ!f~CnTDNV3c=f-K;v0p~#y-HGq7d0}ot<gCkynJa~g3GSQJ=vo{X7
z57l&bq(KoiPiuLY9w?eHaB&Oz#7NZOv!hqZUVjEHT>Y=R;7&e_jh~Gw1s}Z$M6dZ$
zd40Nzd86^{OUnJX!URX~N*KoF4y{vgXLIConHJA$7N|If$BNpKs>^A)i|QpsD|_6~
zw?`HYelXHMgM0m5hJAm(jPHJjSiD~+S^J^6vdnLFNjfBYfo(Li=y#~~%`9eMWaDF0
zHI^tYe_-s%Ar^@ZDML$1LYE9DeUMiT+ZD*bJZb{%a3$_aW!5p?OYNtlcyivzJ&~8e
z%c-%7tPWe~QvditI>^J)8Jf5<)heyQ?fp^xg7L{;Dk{<8Si6yOsG{(dnJjq6_DYVM
zt0Qd^rNAPF;LJ<|6`YaAV#0M;h2`j!t+J6xc_)a~=Tl`sWxH8zHZD7}q%3E+O5YmC
zde%ckH*nh4JJy2%1}Lr(lUY~?K_6hjQER(Pr2L!q8f5$|tmP7jb*2OEME8(ibO<zI
zO<WVQL%6SiedoN5_y{-+W*QJ1HG`n(<mWa+hoY*C28-fSBc@vdeQsXB3X~59#kXli
zeM`wmj8A(x7AwUfk1rm)(EYmb7HygpwX&#SC=Afk-RYwr+!FGNvaS)6?oeBxBr)qI
z>y&9`Kl3cQI@rvHDF+S!W<Od~(89_K90ugjLb^PCd3ml27bF_IBn~}<=C>GFo>$4h
z!Q}DHzGChBQ}}SeKMuZKZ4*CXsN}@%KSukBw%Ip?KTv$C{PB=r|FOcWWc|q^nOHV=
zAU!6w0@u+SU#0m#`YDt4KKssk<)1{1xLRYy{iAo?Y|eUKGo;w&tUFKa*1tG~hj91X
z;9BF6*lI4Fa<Q0)3x1a}$ETBy$uYz}k1SEmI4i-H>G|YA&F=n<Y?kx*RFzm$ew600
z#uE-BvJXpfg7t783wrE)kV$2#dHym+Jsr{3*qS28sG(Eb`AsyD8TyQ!Z3TAqouf1R
z80CDmcO-9>E4jfEu20UNeY7(_jW0ZYaxpBfe~xWk9beTVgVMJa!xE^lpT82;c018Z
zO+o3}V7#~BP$w{nisxl*v-!Ts{J5HvW$}E8Ro6G!SlOZzrMQ`=hoNsTu3Z3|d$}}@
zO4>tkdVcM18D$rj4RM~YdwFbZN?(pYzq-6yCN3}Y^>b}VSBZtd-k*S!DyvC^f``Gt
zyXIx!?1~&Tza!3VutAEcQKQWjF6oxE^Wfv0X{rkG^Ik4_^f{JLnZ7XIFM!iETjK++
zVX&|dA$u7xsUMTA2%{o!2vUbWMuI#vVCq5&gs|WdC|MdBqEn;5JP?Wq;sS)aCxC=0
zj+c#UqaH)UGFLVcB!T#FXar_~xzeoaLqqxkX_B9JI0;>+<=H8DcLBh|WDe4Aql~#{
z(|F*lq_hN4>hk^<Yq^bK?Axz{HFNA`I(@|1NGa+GpZVoexFG{B4KUGAM@f5S!JVBX
zMMCJ0rJP$%5G0mJvXRpF_aU=XR-Qwv-FO(txO@DJ$$fZr>MOXD@o?-=PA)D7G;M~@
zDqjqsI`Ov{ptOG{rF$pua~9DxgkPX)Ppw}bpjgAng8EOCLbX}kTX*)g3~5_T_fDN;
z`Fp14B>3XD)O~8vx6AEY2XJ%E#p&_9W`T-k>X_60a_4;)DrsS}$ccXqzJa>L_gSu`
zJAa?}8x}Ed&%OS(fWE&w{rrcej-t$^ctfoD;2at4%`Hg#L?^I`ThRoHpyMN++QDgQ
zIT&cS?qMTnp4eHGF}khCg!#?Jr*T;;6s}`Znnnf<FR(5l!rd<{vRDa9g=M>amydUW
zU2-C`AS<aQFCT+w!w-54yNGQufnjK(BW%U6zjAc4$WMAClv$dIY=-u>&{TeTV+@O-
zo9BQwU77k+Yz`PVhPGk)7K<E4<B3t>O}Iv?mgFlKh7&E#92i=e?$qwvNot+xp<o^*
zRgAgnX{q<aDuHE;8f&RrC1uRQhzw%eXzg63tS#0c*JD|qHi#3a;_2a#ZGcJ703vSB
zS7R|f@Pgi`0T*8b1#}ydPeR94eq^}7%Xzx2$PrruI>ni=r_rIQHjho1#<mfQly?6m
zcVV$-PUBlk+WmKFOJ0q{;&`Wk%FQQ*+ZCLegSvLC@;Zh2)>niMOWAd&Po~wWqo(_(
zErZhHGgaIGa8!Zm6h;Z1SiR1vTeFy4ZHeAg34pWW_2w2dBvOu}=!*i<+I6~u+bIW_
z_A_y)ceBHDfH~UWD6~YU8ILy?Nd~bzOTvy8!daJ_>9UdGXU7BSp{cwuL<iyU{0a{F
zP1X*#+nY&8dW(uklY|yf?85~j=Pa<Ehp6eM(Rgk5=9;?+b-QqNOcv-(1_S2Qc?;k+
z-&MVJe3${Uy;XRjum2RFoiXsz$xV;)t`9@%0l2Wn>*)dCv-#E~EETeoo(&dLEhUPm
zj*`vW?alosrYxyTrKOd{aA0z95P@?}HAPz4K6x}!t0(I#udbL84e8cVnS&ZTsI51`
zO{A4Oq=YU%oj~&zDi4$uSkX>esB3bS{e~Ka<ReKWqkV%APSl3mGO+jHZXHmVtaS(r
zN}+PvXP}EqpCFjBG_~v~XEBu*X-7`YJMg}A+r1vZZktZ919<yPV&Tn~3f&-Xr%`@`
zHSUJtCVDdVKE2}meS=aOP*s+}9!3<S*Y(YR*`8Gunh1o-vJH+h(QH?VG6UWozCjy^
zc0L(JW8Lj8F-qLZA+zLSVYUI(fVPN)J_2{R^DJGKAkE12L4SZr<r<YAFlCMi8m%SI
z<e*Ff8elgqK}a0PA#{rS8R2iLERc4@PnAL%TGx$2#vrO{uhmX^pHhFuO~%pkPY+RV
zc$+pWVh~LAJ<+~Vau&tap*#$)9RK_^Rv1<%29$4P=Ed$vI)Nrhhf5Si>-@xj=%H^r
zh3v?W%+#9?ikiIaVWbDXn@g9oCuG?z{j_Bqz)4m(Zl@HDo0xZEG!*O^$#ipC8S-Z2
z%nSkiTix6&D%x150{bB{i#mU%zw6w*1pG^d)q8#~1N#89<&@;haCUJSY#ggryZWd+
zDCEFFCub@?6L?cuHomD^brJbS_z~nHbHxNFXSL|0$%I9|KnruBIpYJ^1K8fR?H19o
zeQIskGxD!x(f!4LUNfd1$S_nl!URin1Z#P<{Of3eE=m22`?D0la8`Rdv%F7enmi*^
zhnQaq9BPNt%<IY?u^aw3S*6**@!9#+^<v?RJ`&SQKW2{xL@8Xb^p<W9TSKfDCYFw+
zdXc$-lLXLc*3ZGog5Pl4GE+Ca5Kzc(aNQyyo$L2?&Gng9WG>Z<nJ2hc+@@mef>7cs
zSt^{vOAa%RyT7y+U4Oz9o8It^5gYxdf@)U%l;=6(q37cIbYcn>0AbwyKb_!DFZB6s
zDOp0NyqW&XpyjaWzuy;y{Ub&X6oVr``{uv%>2)1HME^g;3gr-h-Rt(2901@K0Qw|f
z>{gQ6bRJbYZWMrI`jk}vj3SJzomHC-#)G!g!~)2QqEX%A1Y?66tB0rn6b0y<K##<?
zI@o~+Btbm{viwr9P6U~k1;ulh?X=RqQbH{qnhFtI8_bw+Zpwz>yrKL!CoUFoS!j+Q
zr^-npmoqz0Uw3PUn+DA(%S`-S=R2ZGu)QuStUNmx-hNlnI8<Z3yZGaAik*tpRv@8o
zlJFoKhbHX;>uN@2$|KuFmb8zPy%a4&OdcHd#ROSw;7xS6r1CiAO0w=kCS{gyX~cBA
zEVpd8$dWnyXWJkrTv|?M$(WohNXxO9S;H<3$Bnj{RRcJchc-!jvJl&Tyto2rE*QR(
z3k4S}TjpWjNg|SRRW9^|>b6wP0-f{HKDY&F&sauVjx*Uy(k^lfptdlqi6Uw36_sRh
zBXe}rAnoa&mB>-FojF-P87kl3>&FTos0SvZFroJ(CdVWZEr7Jl;(Ra>#gLYQD@u&z
zlcmh*TNWerz!XI&PJq#>QSMXrLTQqu0c4R%_rXLI)GwL`TbL{Br^`e=(X@AOiXBNM
zSW9V;4}y;a=uLn&QiX6IpbG(H0Uu<vmW{?OqUuJG9+K0>P$daHB&UMmuIwi-6>eP=
zKgSKKUP}igU@ykopPK-=gqWgIy-y}N+0aVuaSO#Z6QXg1071`qJCQKhKtRO+Ks$h=
zhDdKEaF}*@as*|_Inv13%TegZYmUmYksOV(*>ZI9woIT%zdZ-XjJyO+i)-a^to?6G
z+r+WCu9`S5>1uy(9onl>j>mR&gF8vMB6o`8>J4`qd-Z$wGt^qETW~8uf2{<@8!aN6
zDb^bl;CFE3fb|ob{iPi`)d1H9EPBC!_e4^tsRKSSUoiL{@R{7$4=M!SLMosT1FuB>
z2c(65VrRci3{g8Iw0rQisud};3BrERnt>0^c2SSmY4|Ph;La=HSn+RfTsU`W<I)kI
z1*W%l`s)3TCr0nJP}@CsP(p{Bd%k+_`gv7+QlO}%*O@(*dsdl;Hjm_78Mp?^65iS?
zUuK+=M_ZF&2_rb>C@$XNv5hkz;D>;-QY;LRQTcbf1w@C+spjIjYKRVFkG^ugi2hfv
zcx(I#-kN-8OaSDbX=!%f<5d=~6+}Sy!C=D5vjE3A9aMvOBue^A`r*(~D;F-ri&BeN
zXuAvEBS^a~d6F%&q^r;)0So65O95*WCKLBAm0oNp9XhavHtr5eyeGS|&ocya$MP$h
z%jJF+@N59@Wa@*RZv<a8j^LGEe<I_F79%&S?%q<*^5h}Qs|R~qi9n)>ZBP66H~{J#
zRyE25BmxjBu!Q3MTg{4<x#S{5N5obBG%{#GXA{i3D6{*8Q#Q9e0po#{t7`>c4}yYG
zp|NE!2~do)aPUQ%+K<E&n6h@23F7ZU9}|nKq>S>P7a7~JtV2cWS%fAf%A5$7teld*
zl}y_5`J1Cq5pEMoFWU5wVl@y`Uj@mEw9!w7fZ8;x01Yg4%}Z#pXdu-DxNU~<Pm9!I
zG9`=@vRP=($So>^=;Gy|EHTIj`aw7gts65%mjI7))ZH7pU=o;<@_fCTS?5Kx8mwFA
zZ2-(Ha_t3tv3~;z)wn+0b(1@?7}Nh+7_R-~G~sET6h|0JL8?)iED5Swla)cui8IH6
zM;MS|l-S7IXK8$IbO|~%{dT4fqEKieQwN3@7>(_)ue*&uZQjKb@q)FsUX>a$p{nuJ
zLCC6O6&oBaTi~5zV?y#pz@?O|FT2cd;y5On+X2g`w4(FdT~aA+#G0`n*_Wkfnea25
zB?tNxeSh<!#BU~e%?zXEf;nD*J~a{Jmb5uPrCu8lIgLxz3=2t>gg9_E-LQm$Erm2P
z)J|j%&sA5rkgBe30*$$<Q9M&Zkbybl817KTOfY3#ig-0&V%6}cXAA{m)f_zLu1ZFr
zD=@bm8P!2nPIF)bk(hU@H<k%ER4w`Un}HkM@~Sf9M%T3`jDUkgb>?hRauT+mzNM<V
zz?qp6%FQG^ZFbLqU24qA6aeftYFt}msl=lwcedd2;(mwrwr0vfo^rxI89Ky5`+vOU
zL<GFt9<$T$t|pID|3g}$Fq6PBwVdQTV=q-^BrS=>Fy+bZjfuT>3|udmu5s#H(xhlQ
zy_T^3rLZC=mB<UUd^3S2MN<_JI^*Ti=$xHA0QO&x0Y{q>onnF%0V7)3M^Fr}SeK@(
zDvQq_XG-)G8ZV!o6o{pjFYGaoMKfi%&S0$?!OC==yn%B-^TetbRbk+qfc4ltNPHe6
zk%t6RLS<+HW&s@|3UK2ibbYPegnGUvAZlWqFpF=fS}hPj*$x|Z?Kc_mn3AiWjB{u)
zQVdcu@daw7hL#AFYQsV*9-FXV>RI1hx~C9AxXO6(0=mR{Cu8H#u18NC&{0?t;U}|@
z4*@9;X)vNvi+-bGCPYX~-4Y06U1>w)PAMsJid?i$NOf|q9aLPsCq+8O)j|N4k%Ush
z(>JnYQpSUAK@3|Bxh9fPA^EZTXBPGD{5Si8Y&$@@fZ(X`Cn}BxliWQhX)v%Wm9u}2
zowpav*IlmT_s_%OZf8{m8`ny-Y?xA|K(#32&0gs@31bJ$frtQ@q7;l11x6Y?=gk+M
zPodd}JQ9Ic9Z1AzaUPQN3SZBDbE$2_g`1dnT%e^T2uM`VrAXnY!zNZEIZLcHjMCu^
zk)yzDBW9AZusaf0!l0JUP?I1L2Qyv&8P5EJ1GGaxA~usyt~Ivw!^EagNjfPUjUZA{
zuGFf-#)ZrjkxO_Tp4znaCu4DIhaf5o#R=jH3l)znEIK8VAywjnM%9=eJ8D0GiT(EK
zTf}Syzi4(}L7$wRBSYIJz>wNY#TMkB&(N$8Clo4a3>gZOOi4N$-2kORb*kE_e4&Tx
z9Jn{J`Hr5J0L+@hA1M)3UMb1KBqm~S>(WA2!4pyUv_YPgB*TEU&4m_9wURN`WE=ki
z)>Y!O6fw@JNrp2*<l=dgNTXRxsptggOq0oCr(!Ci@1UMXa80<<9WQ^%iho+_pQ(!6
zn`Pvm;LDG1m+4f?7wxv(+}TUwr46m57ZPhzLVj<;w;5ScjsOKsL>gt=j}`0Ob5z^J
z#)*6oo@dJh$j3ygD441p6u$*lPOX7AO`kl6p)roGZVNa<bX)Ve3Zbln$;iAUV=IMj
z+ps#Oo2uFqZNo=s^xK`#>)GaJWcD0$0uQe>zlX%$xeAhM^ebvas3`6W0Di;DovJfL
zWHRudX=#9GjKs9x5O@n8^zx)*bFx|G22Yf%c<aR5G-*#?uV$3FoQ+D<<k=4Hw6gol
zaD>GJD?@BhKtAg7fFu%SPgR*Q9iXaV6vCK2*MQFCDPVKj%KwLsv|?d!TH$2S$S(XT
zIZHQ1MM=?J7u@^bc~%sQ{D*3uMJKYYD&x)|blsR4#U?AAlv!FEWoizI6qyX*XT40q
zi>4~ZDvzeLN&5uX(qezyG|swEF!6iT)(R}9T#)%gPXsU1kA+n9h3d9Ap-z_yE>Z-^
z?Oa@M#RmHX5H>Zy?nE2!ZuM_d3n>!kbXz^z@okObi(1{L2nGL9USy)ei9w1tXX6TU
z;(E=Dl11tp12}Y@twvp{k{q?jz&Uwh)2-laVK(cKdZ8J80)}&c;Gj87GWyg<03=vG
zB)p{c-bV>w?qve@BRe1<-OKR@D6*zJh%BU7ag}<DtLY{LAGwnfQo!vQ*;7Z16$uE0
zotr#EV;iTFsntA0Yz5hcf6`hT*$UzqBK9wPd~6cUuAGoTJcf!;_t^emaPdDh$<B?}
zJtBG8cqW9?qH{t`r?O#7?Ccwn)$${TOO^hUk}`uV&dv~t9UXIJDo5Q90jm<*R1YM0
zue$<{$Z-OM^%zxA&$elJ7RPc6I?KEx0b&+^!UrnhT7olN+y00k25wfd4#(N3YnU+d
zb{Q+&Q6tKxZ)#eI;)YtM49C)(z@x&W<N7ichB7K+N_;NrhKpuf69A}Oqo8$q_E?yo
zJDox*YvxsSn|z}?LYPd)v?FRX+B+bELew3Cp>L(I!aprp$qdb_Mz4&0Cip_-vQV6b
zVeBacuNm<&V@H9+9*>#nBa{d_cKC&rg3LoUv$qoWhd1*_gcL`)-pk&yUxghqb}Gzg
z_s_!4WT=y=C8D*t|5kLtoz*)uW?0T2&aSxjIJXEW*lVThdT*C1mK1aG?F2QvgS|Q`
zX|xuaV^%aC7_Cb0Jd{@os$7NeOiEKBiM}aWR<uNk5m#qChsTlKam6FtPDP0$a=@3C
ze0FG}T82AJlCtd%IAUq860&DHKtD_rHx;a?CAGEy2ota=6`Z{a7L4qKU39Ysn#=-C
z5nC1uqFFX&P`KlXXDq)`k=nvEvVvYd`~f1Cbx6tgfC5<pAkE{z%FGO$bjdRgoQm0Y
zW;81tb->^(%}pe015g$}8r8o-0!8Fl2^l$_YtvbjK0A_rFhvLLrho*d-k(8lG7_`4
znK-GguMmNHlR0V2My-;&WS0ipS2XD-4jD+KNKRX_Zl{h<#u*eb81kIO3z~q4yUEpr
z+f0Qa%l9WmdaaBEPZ^^(r%Q2dP*KThp~dMmY;woTl7DwrO!o@@!#zMdBGePu(zX4F
zlOpHnEJYL&rUMb`nqo3od|3iuY^CI`W)z(wy)Y!yf~R&eP8`S>f+l*m1CuB&gM9f{
zXaXQTgnY3?plxQGTsG>?0KW8TU5f$4%s%~8yAd0=@8d|U80i4a!31pQofk6>B4IRH
z5Uw<R??kPaI}hepNW~At3=JX}3N>b|v`{rGDgiuF4N68I?KcH4P|aU{likdg6%ALZ
zm1zyPqFweY>2pMLE*4($wW@h2{wfE2&-2;2QSZ!}ipfVY2`t3@s-nQmspwKfKm?~|
z?5vTf1sF9jiK%@=F{rau6gG63__T_KOesZ%U1rl!Y3^|xBFO^fF-VC+IY}>O-mU<r
zG3N(+B9bc^j)UcBeJuFzHMr`CNJT4`CM9#VG182&W|-Oa=8eecEG!RZ(r~rdxV+I>
z(ut#H-w$q4+-d!1H%z#09vg*Lpu~}C=BBfpQ)n}49kM809#X9g2PE%DK1fhCLQ&0I
z(iCZzE?+8UQIr=3np!L<*2{FVvX{upc=X@~8afViHc!56jMjK-n8R&Xl0P2P&f*j6
z!uUK*!@0Ji++RBZ?zN`ou2W{lk31+FtgV&OqT4C5sL$Zct#=B|vo}oEG^n7=sa6aF
z0xIF{dA%|({UYlxw7g#DQ_;IB>-tn<nv>&K{>lMG*UUFi>`O%N&r_x0SQ{HZ>dk7W
z<FSL7fUL@+D3CryvCP6;5f-f}cqu(~ca`PV8T75N=xbQN2GYp*TaiV&MXRX1Zw=C!
z`Y--uw$ak1@PKP{kSgU9&2;&10dZwSMoZR<>&;}SsYIL~SkUASX2EXUaB6c*^rBLf
z75xs5`&YxkO3{)I8HpcQmrsRDy7gd#(hkLcdl6M&)<y#Gcf=5eTBhHiV+v3v>smZ!
z=pzymiCpOiS)frU?=YLN!8jjf+?ZP5ny_4?2I_mnf&#UNJ!IY!*-9IN<FU4HX{zIW
zwriZoHtJt0G@G@=svuwGB^ceQza$v+zaq2Bv8I8tBQPm$#lbcGo;_ZlIeV4N5A5a{
zGu_1E^G`@S08RXmbFd+W__0X_j+||4r>+t>?tS9alK!kNZi^7k)7;rqme|Tk4t;#9
z#1sh(hQEUU`<zn1$gcrvh&V5SsXQwI>3jsF&;5-_NY6vS@Ofk#w(N{JC4jgSXD)*z
zbLYX6j2G{-^NEi#+L$#VZ9Mt4?k`|%D;_LF=viF#j}R$pq4T*KD^9!wiIOBskt$8P
z44FZ~a^))Q)8VB`4QlmUYei6?#%*O<gNkTSuTzyH?gt$Ptn<xR->3dh^`C2}oUloj
zY)x{UF~LcvU2xVp=e;t~6&GEy*(4t|yY8xMCVS(ZYEw-y%?#7cw8SiPdI8R};LI0V
z<Ta%{x$+e%u-j6_ivCCGF5h}jLv4#8H{A*bVj?Z{ACNvD5F2q2H*|s*y1@@Y2yL~^
zPP-hn-44g7tPhd>4um-L)P;T+gkczkai4V6oQK2H_(!w_*YJVkc>ASqDlpOM@A3>Z
zJaTMr_jmX^{ayZUe~-V{-{<d_^YV@7cC?d&cH~UFyst_DZHQ}i&J5i%lje7`N+dk$
zB*}UIx(XupJNsQu&-HN(8(u^8LSAblPy3&thSTG@|CRdS>Mar%{~Hv+72KD2qV9HQ
zxrI8VYvgt-3hCNiE@*b2fUF{*4#-w$0(AtmfhG*PK<f!A(78h&P;Q`W^`nX+^`<3*
zt!9ML*VF$N16;n9d`e)Xrxhn$+!8Fg3kM1KAU47}=+j^buESsq4!$r2M;{2ar}-85
P{{}mS&V2FDQvd(}b#{e<

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-book.woff b/0.16/_static/fonts/FreightSans/freight-sans-book.woff
new file mode 100644
index 0000000000000000000000000000000000000000..6ab8775f00b11a80bfc92cbf736e395873a6751c
GIT binary patch
literal 31612
zcmY&fb8x0Vxc$1dx9!&4+P2NDZQHiDZnyku+qP}nwr$-0?)~d#=DeAloaad<nRzF9
zlDNu>iUL3Y-^tSpfbyMza6<iG`9JReCs9!)>2E>tH;eNheB0iF#e_vfzq$VJvcNZ#
zK&e4n#N=cZ0e~M+002xd0KiQDZ^IK+Oi@`70Dz16*7APW{h7z@8Ie__V*~);r@!r#
zzoADeX`O9ot#1baAdr9S{J$}K&>{$C=;A~G03eosYq<XdObJ4)iJhtSH+TH4srmM2
ztXNN1WUBA@ZHtutt@&>IfAQ1Q%H8Cf%Lf4DF#v$_{uEosIx}N^BLKjl{aeHO9}r`!
z%gnyXZ|?rPjQ<T{gumc!X4Xz_0D$4I|N096fEH?r3zJyc8h-1H^}gF3|Hk54Q+1=Y
zzT0=Zrqthd1pna&B&2|izP0f;_icyd0|1D!qkfLy+u1rg0RV3Q*<F77TxTccc(8LY
z{<ih_uP=??JTawp0Hv$3!FQiMzqJ5}|F9ONS@xZSd;_p>negymKJ&Kv_rH8;CKAqV
z6*p1}#*hA2|LkqH26u8+X~&mNP7}R2lo10Qv|7D+VH`Q9go#N58?*~Tn}nUiN)uSd
zkPqhN-q$whxHH7zw;#9Fi8NoUUqh%vTbG!I2v*%%>%<uU^yqQ)1O;JwmQhLg|NJ0j
zfdq3TC@n2kFrz9`77#4`)jeG=ZH6KlI~TzJJ|kMXCYP#Fq6ButN2#1xDjHQbtI|p-
zX?XHk{Ji%or+95@eDd<*quRT1s`9GrYVvaYd_3)GbFy8%Y5+O5+6su?&-poE>myxi
zRG{S3>G+1M{_;mGeXsH2js0@7F2Zmqa|34&cgpdXJ&wGqCY|?-v2#|ls={wpbMeE!
zo0aR7DrOWMgk<|D*SclBFq(JeoMoAN3X^?{l4UEa1`V3_Whj*C^_|w*^_LEdHXr`?
zptI4|Yn4rD7tJ3uFMrSCVID=^wA;2m>ppTesvK7@B0ow#%!?VBr@}){%j4&@G5OOo
zciqjF?)8tfT0W9DU$w#R%Hwl5Uj<CK@JbOEJe{ZeSi<bf;}lPii67k4cpFdmJiH@M
z4dHN=6^m2|Yv(1#LSzEoHBSE;B{B`GmodlKo?x9~#HS?W8zU?=%*(m5uw$<DKFBMQ
zQ8MKRcr^uumc%En_*~+sN$WN^jMHrhZpyn!k0CA6G!TY(%7$;oiE9c(soxh7%e$&L
zu;}HV<YXk!Nfg8~6<#?>86-3liKlqRaXe`;(WJf&LvN}Y8;vBZct)NFa?qImcq;rr
zUSNERKWp^qS->OFWaopal0S@Mtzbc!rE#ZGN_Gk4s3gtpM4iUGUG?EL<u)!QQ_j68
zp3=4Vpoq|mUEwBbP>|hZToFf1+jp!LZX7`{@3+j%OVX|tTNDq$iE(z2`!@{1L6cM<
z;?b?s0<-I&Z8OT-LgFp{f#)V+6S(If_Igmp2-Ui8yKVd;9=wljuu}9e#}SdqRdPqy
z7F0jSYk;5BB90tO+KdEmE4XPgSdV$4nT~yr8_RS}eI9(Y?d}Gw^?4@G3#AWFW{$TZ
zwf3*D<64K#S~3hDv2j?Fc<Kf_QkC%Bq1&n2=DKEman<8H!te3zi{6Q8d|+%pV&(>&
z>a}r0$P9VI6j4fav?Z#jBG4`=Fas)URgr3grWLm&-rF}L@PcIJd-0qo=4X*?0N8qq
zc1V}u*!s|lf8s3J97mM|e(u@var1Ws?mPDirIU#x=4H(#WcZuvvvEUC^<8B@sFplv
zpZXyB;CDjkhM;dFZX@nsUrW2iw(<oeZAb4IFN1F7o|9b{U578)Q)MC#DCXW`KhwU`
zoCjX@tPMLuZXRqBFMZ5UEqS8)#17Ia4AMbdM+lgcTv3|y!NulP&T18;M69=#>cp)R
zw^*p8gPTaPY;O5Y;m6XAP#Lf-P<BObXnhcJZj5Ne+bjGO@5h`r<&jYUxe(bv9$gHi
z_+ns<x1g8}Iw&=Q*~eh^E&#&eeauFx##IzFl*TL$MXKsF>h@ppoSfxm;}>9Car(Im
zhd0Q}Y?xQLZ4z?Oh<WiGJS0`{V5-20JB^C}y5WI&OP)=5e8$VbWnGkVS5bOpN&LeZ
z2!2c%ht6MuH_5;n7bP{GmE6dfC!t}2h*z0K%meSXs+g+ETOMK1nN<|f+U*01h3WXm
z-Fb{uv1zWTTRnjJz$s?#%&9GrKbKq6^K)-Pu9cbd#=RO0Thu7^t8g^Ba0l+>Ep(1$
zoDvO!GUNq{f~fM#R27LPmMs-~WnspaVrR<pB(B}$^@Kb{8}-kW#kxKd_Axp5Ur-I$
zH_OX?Hx**vV5O-yyUdEC%xhvkfexSaSbCLz<`saV33HOFh3pA)rU%Wt>4=aPQ{4H}
z#cPGDa++?O;>STY#VDh^+E@8O-8awo20p{Zh+m!Qtd{oxOE+|5%HA_r{wjVO95`CL
z%f`b3%=>L^#KV14!)zme;mf&n;4!|9l9a8t23k;l+;B_PTNjcal_K&O1zd76`(6>I
zh1=3e`g^2qT@%5pq<dKj$0r#^#eo`CTvsKqlNN~OdR(;ia85ht(VqeNfaN8rnJa5a
zl;f?dexZb5V}{#u1bviTbI7n&3H_x}Lp5}G3FA9zMwfy8^0s*~0Xeg4>D%a|caims
zS6uJRLp^tU<hf1HaHDj-BN=aB`$$N#LaVjPsnlpjyrWAjs$+ZKjzrP+C-#S|EsgRc
zH(x4om#Iu7P_Mym@*fi$7PPE<Jy%p#LrG$&jsa06t9%reRi0r=`V(vA0v?6xxLP4Y
z@R)5p@(@yHFD4_Sq&b}M#|!zfI;P=LNqli{KR}*Fes+uR;FMg&r7VhdkJ8;B$hJFD
zSU)V$n>W=&h3W-Y+~|}$tt0W@#e{k)E-?HC#hvQBx~;f9qrYRFIc=)apZlmY*Ri(+
ztTn`akNtvp7A*{;Ot>Bl{7SCKgNCL|n8n?kzKitwkqLO`Q&%jk!yPO<>KOl0R?)Tf
z2fdQe@WHOGPCEMCD*5SXZ*?t_el?8SF^rH%9?i-1vx7s{6b;m4D9O%sWzTyR)?vJ^
zaEkP~9B9m@t|QGiEFJi=!g!Q;Cl*8~+pYH6T6GwAv#wfS0lDqdxxN=`#$3{VYsK?4
z-nInHsa|X*ZY?_6NmZ~)jDPZ(nSnRQg_r(Qc=HlFGa9&iD>EEz!_g#Kap!{+R=F>(
zbTZ2D(u%nj>b+H(GOe32_*Ncu<4$?^j{mSaW~(YjC^YCH<X=S1If5LP`S8P>URP0q
zN!aT?)ltM#E$UBRR^!o^{$yr9l5`v9?6??_lv4SV6p^{NAhI+cE|57>J`ah<l*&Cn
zPP`4y^6RhZu;g6g8+vk1u5G=f^~+yJgZfY;U8+dY*1(xiU5z}36Iow*KCC<wsja1A
z?PD`X?pTYRRo^mkW(B7AQ7xCSNK+%8e)0VZ^Aam&1uaiw-#Jbr!up)|cWSYJT9RAQ
z$z5F0g?Vo)&8tfnfnHO16b4jn|H23ES%Q#XOtVHTyW?7J$<#@t`#PnyXe@6ps~21s
z8T<N+m<l9)Nq965#aVbJfbNXC7RLU<`@c)Z&TjPGh4DzV$scNm_h}x93P^D9;|pf_
zTmK|V<(12Q)zF_-nU9AneqG4vZ+{4Sz~O-{V$EZ&)Ts@VqhW~?%tpLIiTY46Qx$8f
z_Nt{CRdCaujmvi&H5ik5WJJ&&fxv!D78QtOgIv&v7L73{23ng{#CEFqW41O(YN(T~
z9Y9Z$$_B)6gO?hC%ISL9_WQ^Ns@Fn~$C}V<vId3}fGeHR6<C^iVk?{jU=P%*LmyYG
z^IsQT%eW!Y#%QaBTmnrGz^k=70ayh>zVaRD<uUAPgm2wUwfNkmklkMHNaOK=*(%;s
z|HyEDAW8UvJ791V>p~``C*DOqDa-w~4Pb2{Pr^)kE#)P)9YGz;?X0LuJiBEw`s7d}
zw;iL4X;j@5=uLZ-I@Jvn{W$Pe@^!}tZjD^U3V8c=@n3%XL;p)`-dTKEG|@E-Ih@@&
zmPO$U-uTJsTeuF|_>MAvLif<*pe%NESF=m#gd%9NL|I{`{*ly8_@VnlDj%0QJEvoQ
zIfNE#rbnN;O0+F;rZbqXy$6bosk5x!Gb6VCgrUFT{;mE#L*r8D^$P-7{Kb`{wBTj*
z3-C}Lw)@{Y)LCv!vQO?O&r{)|LcVyph#-1T&@XiDU+9FizvcuPNRUqp+p@_8qv9IN
zUZH*b@d<n)b$5D)Ci#lV=H!LX4JNrSQ<#|0A^0&txqIQ_?Zc6=$ng+Xd9lRovoPvq
z!#c=`2j!!eqocjgf7D?nE?aMB$B3nBPDG%qI|NtkEEoqAZp(1COjf(bXB4qxNZW`V
zLuF~r_-D&c;J^kXYPMWghgsKGM+3-SXLFkFQ(QL6e1D2Wka1fch)^Q%N-xwOvY<?}
znUj<z7zacnI<$sf)AIIXD-*UO3$0X0l}SYx7#xJ~9Zg`OQI*Ru3<}eh7~nR5D~HsG
zIX*xp9&7??)*A-Zf681SLUwJUQv|xUM4q<8u38EYnH`2!)b_oF#7=EwE8kDVLMs}p
zeWncjblt-C1sUdKk^OJ{4E0Pl2iX1nf&M`Y$p_S+pncR(6HbW&@gqOQnW}3U0?8A>
z;o#0dG1S^Iv%5Fo#o6h}-%a!lpCOHEeGO{m_4JbA;taV5l%Vo|VW<J(4B;LAcYk20
zr>AG)r-un43h|R2be|#Mg&GdquO1@sN6bKv+|YFvLvl^P3C0{QI4w95I0ZNdI0-ll
zxHmXekXsN-xMZoXPu=?$;(ShbwI2giX`Fxnzkr|sMCmMYW!%lz*H`ox{;Tfw*3s2h
zR}~(PuibIaC*&L6xli?8PcY;dJ~JPEm+$eS)MAmGDiS!x3dZ&*W{MveO6!koRduzM
zWe)b1W_P!jXK=8fzaqi|{UpQ%`Nl>E`>4Nw3Cc<f^DN8__0CQY_waC0k<x!A#%ZW4
z%B`&~&T(-v(zUfV)_J%)+P%F#-a&r^1_%lE^%4>0<r*67?V=!KuC8*mkxeBxxy{5G
zOC?olby!TSKAf>yt+zXmHR7&4oo}{9=a7@I+}RBD2Aq=_tHa^eVT5C1%#QU`j+N-E
z)s!)CAofpIbR^BsbAC+klr7n8bov;Q_4N?JMdNdOJ*MZ}9^wWkW*%-<enO+X70MVU
zGn!sn1?LUH_@|coVF1FVLcceW-wCu78~|_yGy~)SDgYyZIlva+2=D}u0ek^=0Qc{2
zARq(~0VoG(0dxS)0DpitKoMXK5C%tbTh*3tlhG9TxdD#|>fgUiz_I$PE+uBrGY1^H
ze<8}ih%4v-%ACz1Ho_hRpWj~?ezgiSJ11W}4dOStxR)UBZ)7IR&|GFNt~$rb6eg+B
z*izRM%PGYp;+WnIhzZsY8S#cXz*GL&SyW_wAf$$s$^EGF+b1kU+DrnM1>k_kmI4t|
zn#*6KzUXu29VYng9siCCqn>&0ex4Kh<Tu!1n!Qxu73M_i7~*Td!j1tweSUJ`E-hc7
z!e<lnk;PbbL+ng@AQRHd2s6K;r8QXU=lt)5JNsW!F=K@_tjWo>pTsCVK_gDeiVfq>
zt3BGab;o*vloS4t%Y7>DaCiuaq`zjHh)1d6ksb}o`k|2GabP$O)<-_``Cr752iqki
zx%Pq^AO>TXy0ckj>%O?tq^)gnd!N;DdR)6Or8u95gI6(-47dy^{Gf#eMU`KIv8Fnx
zA@&vCruY;54sN1-Gob1(mxZ~L8k`XrRG>V3*4A&w!rN3@CL;u0ZA&eRXDKG$bS688
zKr``R`veSo(x%=%CH-HysVkYy5Ws{y*gU>+8%{&Z+R=Gy$>zFhH2#}z5CF*67XZ<1
z^`ZmTO=+Qpt#fg+?kH~jkI58E{F@PD3UR!V9_a|Z1XPn9%MU_AQMl~TUrIKjkw^wG
zp?;yEO@zuy!u?|_o{ZTKDi)SB8vh#bY8E}X@uC;p<~7!)N@Fs<-YR*RQvQOSn0NV3
zJ#mXLd0w@gcPzGTK5Q=8BK>1V>7_xUCE%WkwMdE6GyGFD@<vT;8MMVR=!PH@Homv^
zqZ4<|{7#l-kh~KgjVE`C$Ro9du;A}z9yJQ>^J9(4g#C{u;<q=x{IVeW5n}cHsD8M=
zV7u+X%Pb(2>H)|cNK30Xl=jCAmSMa<rx;P1NC}L=_e>tftxoLIq!0b9>K`Nrv#n|$
zMhW4E52<G;v7gNnBC&8wN?=3M5!x0twJc!oHb3B%{a;`CjSVO9zUo&{qoPp#Z}!0N
zN-(D@KW$$KT;-suKRrT)1L8<=im(j2gHHl${UM-Y1nMOPo#xFlTg@;&>y_q8-p>`i
zyi*Wo72!wO9|hHPKR(8o=fLPjNP_M)+D)T3fEk**Fk?zfYkgtvFC>*Q`OuHoi2;IM
zntn|U!Eh+;(gnHM366x9tnm)VRcnN2*`6MbwttGIpzb4=%;Eg;igY0R?W?Q$ab&!<
z^qIhjJ47b){hMRG2|UgCOUK+z&Sk&kWO4lAq7oEoPRa^!BF<gB0=E`8I=`#9e3yf=
zBquFK)Mz+pYWrTIjQ(PCbRliG@o4><L_6MDDMQoEk>-JD!jnl)%!D%~O=qX(Aq_$5
z)YGpb^laFp@);WF1nn9%=?d^XB)y7HYM15pDJmj(nGp2KB9lG2PZ157%TUaC78}xJ
z$D=Stip_U_0S{Ub`jpsn^n<i&&`&E`1nrCV)elK}V&`sO3id^cp4js>?981`pB!77
z6Uj(&`8<wtPawy)X?M!I-Ih+Wxy+9^q<A}CXI)(v)m0ST(@ne8N;4?R+Qn~|^)obi
zvV7+0)yz)-llZG6YB12f5JyKHz3WNB`-{n*n7~xMProJn3aIV>Gx9PyJHcM?G0lUd
zON7?H=e9KRqG1@V8F_(fhH$C78a>(LpqMj!ukur>h%=mIAx@13sAg1k4ev8U*v$`A
z^LR#%pt&Az7V)j=d1CZrcB8mA+5x%TG<*o>*GKO}@S?z`3(7R$-sbTNypqq^B#LPt
zD(>m}w`~lEr&M2ImZoZnRv}+Ph!xm$L(L_bIjh!hWh)_Kr1u`#PC9lvO8ck`CMpsf
zi4CODNggbuOdfPtAfLe^WA2le+YOf_m9Oo4#}><=)odt4Gi|0qUgXejRp8Uj_8-B^
z=CsY9HwmFj@Q7$=v@5P`ES%?SE4B2zTaKW`*S&d5o{HlldwN@n?tEXpiLy@mNB1>1
z6VS^M^fu16K5yi*)qW;5Bpq$DT%Y|xNcORKYMtEb)yqR?Plw}}QRm{&h+O}gHre9P
z+qty$)j#lczB@*BBDMaixO4HnA%Xiz7-RQkKe5zn7AJ5{oaQEh+7j50(~ZdE69Typ
z)Lo-ehB)A6_ful%>p@F!x26M@rreJy9eEaxLIik_bSnjwAwR$bx8qd8%&*{reO4R&
zh;oy;8iv3cIUun+7@xRn-VPrIB4ih|5nutW=^1PJbNTr-rM2Y+nMKcKL<zG|;jxeH
z@c&*FrQ&ZceSEy3YhkuIh<F8M*|xpU`*>W%G`+amyuD1>wmq2AU8wmC;f|?ZZhGI^
z<lL-f%@qE&-Al(@`|>}e>XE+HUyLLi6Q@Q$f*a<K&S5j@hE}u&<VK3-5_~;DfoDS*
zbuCk&s_g@uR-xhv!G`>F>+x8xTkv$z975$2+NUjLX+tecl}gc~fA$;?KLYzg%C}-?
zS5rw#Eg-Z(oL92vI5+<+;Drb^Ds&B+xav>&`wmJrb$ajQja|LZJUS~rNxQ${9kA2r
ze1(8_UHQDbJyo9hp!t4M9LTD+zfHb2lHk_2oTZd5Ol7#f^)3ff;Xx&?Gbc@|(kl9F
z=jFL_yS-<QxhhQf%A97~;#*mAu~&U?&LjPLK)BHJD1u@UmOL=SV6$_HgEpl0yOjt2
zE44!|h47JWq*A$b)Q7xIjglL|O*n>LF}eggF3XmqS)!oZr|28}x~A1;>GD23<#Lg=
zwCXgUiz_T7NwcxANpF1nx8TH0ldax$s;;`KQETya=eu708FBPfMcLBqq<!vew&!^r
z8Fn%?of%`TK4N+;O^;aS?wjw|q`}TqbjNJF;<mC}lVrV~yXj5oS?GYe`5N(n-RTBm
zv!T&id89gu5ByT<^x|o(OqdDScd%?33UY7@hs-n-gYkQg#J&)Q6t>$=&iheVyl&m8
zq1%sI)8dviwKKQk)UKN(RWX{xw6!4C=3-{va+~OK_FDX?&FkSgcv-f2U{<a5qJ+MT
zGfkQGd}4NK{WvZ{Ji0Yx3*ng`YwnFpZYoqo{MUfMkv}UAPKZDhI#Xd20BR`QLu3DY
zMeh>G{ClN86fi0b);4s2)UiuQ(_w|GU95bmErM8Cs7mQtuHrHSw7}OMg{=>IPe}H^
zBXOj!T>8oL&s9ZkS7tb+dk*a(_>V=w^E_%*);L>aX_oI8)<GHe(raJKK@@Buwa+O|
z`_V6tzjcnv-9iJ)kSPI){(MG+DSr9e7sHCsw6zN!IEu$zvqWc$JGPoVcz5wcdJpq)
z$u`|sQhZ{dP|uXK#fd2%dT2AnGtMl|zl*dr*H`JgI`wr<)&~b$P-3dC6cT5qg5Kn~
z;)vPqO5b;B7=vo|{iK5anmfEBCKXFYDM@KH2vajwy@~E_N3_~D;|X%`mS8P&L(LuB
zN2`*H8mClVo~b5RuoQd|eVYG9->(e(BYz>el0E~QNSVGYpTNGd4uHwC6Va5X3SbP*
zgPK)5gfF%_C-|+u-5{S3OYh$MyUc8-eZcFsJ~Zt%DM|OWPnFO2#ln2(5mc5OpHFh?
z0BxoW*alomSORjj@N_TWD`GduaXC|E5k_tOg|I=7%*Lf9X28>caAB+umy?>A$)?^G
zHxESi{jm}@b_z6uisB)Tbe@ffW-N=;Aa+n9vee4n3(Dh;WnNGs6&xME0oj7Koh8Z`
zElZ_I>-wlOdHDg{+U~vraA^J+EOiOf8h#qIdiEmXUB*s;BQq*m#&*Pt8Oaa{2CNbZ
zv!aDB;>8);GxlJb&zAg<Pco9M6D?W*8KN_DyUm4{rpY|l&N#xgC{KN!LzPXplTCNK
z3m>C{b9k#S+SI`J-tVnENs>;}?6f&cF9hx#b$Dc7)&kKtYOQwCOI@rCJ-<TURWRBV
z97w`|df4t}dDv@a2cCPu0mx0z&(g5rIe;~@;r_5=Q-@*Dy%c39!Gya%9W~|L@9bkx
z+T?w-*V)h{$UwN<KRnrh3h-$Xdg42^r0!6;C{Wv)%EmB+lT0~CIxIPu3vgH7`j-<>
zE>jwxr;G|qQyx}Y_J^Vfyf+|n1g_$%kJk><OoL;iOZhT&W$31l@me^=4DjIyr~QDh
zavG;a-NXCW-o)ys;ya`u#C2iVA7TdYIM7|p6c394Xrx4@<nbH=B;pU~5hyr-7eJ5{
zwk<X#$0aa44%!(5K0=><yFC{&UxvnYWe~lg%k0YYTH}?0r<;z$`PkwBD#M=TiF>vY
z)NJkd&uanJ_+%~#O_Nv+zed);HL&`IoY@BWY1ZhOMd(>_K7AM3w=t{1&_atkBW}uJ
zv?VqZ32qoG!T^pXLog|B;|`7n`kZlf_8hmTZYcIVq0q}Fkhg1484a;V+_y%)&*2TO
zCA6s>74x^wP#xB@0oktSmv*9=$;CF?%T?TahYj3O!XdhY=PR)iUKQ;7D~WAfbc)|e
zG0C`|$*^}i5j^uJo_i+I+Li{jev-T<(ng?TMt}3IBDDPK@Hy4PSzTEFcvq|-DC#1M
zNF-JN_FxQ6PNBz5eGZk9a;nT%dPGxIG+`uu&9g4k{+KRm@_0?qRkA4LZP<n+<9(f3
zOhT{Vt3DfiyJ|gLxzKpssa5rP-G-P7uNgh7azHWpyw%P8^jh=Xu-jUTwmDsGX#osb
zB8)F>(CcU<*D&bws~8F1ISZ&~IS9Rw&vDY87C6?oZ*97^r<KiGU1~WvgqbkS2AkHm
zAezMsjfy8zlh#<}%YZ~WLu@)SOAcU(#%U3RA(+Z@r0;9&TP$?OaX)9mlPkg6DkViM
z5(UFKx(HAVQIE~gXCR<t%&1_A%MGheER{m!SNcIB^^inSX3Mc0u_>l$i%-_Go?<W+
z6`Z>|om6$%q<i{^Y@oi;dR{y{c)F~&mf4(ypQze6&WylTb?mKT+vXYS>Sm#2F0z_z
zNlk6O#b3T(b3~jL`T9zH$Rd<)w118slPHR|j5ax57ST~`EqgpnurYKdgL-d4dQXw^
zIxvK#B=b&_-TSlUv0V-`-E^ZP>_;ewg0cPKxUP%HE~9@l;=|bxH+NK`ME3mCE@ag4
zcnG3;ENmc+m9QfR64ZIi>5CioJQMS$T7+0%X!7NQiS5e^X)mf*vA$B0>EN0qiN{Oi
zXyZcmVjgdKp(EP%YsPf#^X!q>ahiZ}sNJ^(B&8R09pppbBQ*=kLxXrmAXTFlY%c|x
z$J%%|@b4~ZVV*z^aYKW3Oyl7V6pofw>FLGbpga7-9k$ZLgtzwQvYO5cGs4O6IjUPF
zCOJo-c%|NYs#_wLd5s@0@VuM+hWa5#y{B2*+YF)fg*sW)V<zZsly+2wEMfwV!Pcte
zF&~ptv_U@b0#{nU);^&g3EvmB3(-u_s$t&$tKnwwJ35NaQy*ueUWMaGKE$caQIq+e
zvNeordXs^(Ao(K}mPs<OjsR&};0BCoH1Do_$x-)2)_ctI!5h<8ry;zv11qx$JM$^~
zwOyQM;0vd%rLt4HeP&652mwpI%BlRXDb3Ya%-PRdeF)W97i8FNHDIX<kTZqJqo-e<
z41KDsGVAWI6-eN0gcG&uiJ3fj0@^w8;eL+ncpz5%=m3#u*+fQPVTO_(kkOw;f?||T
z6NE|Is-aW~+;Sc6?kBZxvXgTxX(j{lo;n#y%UIF4*ZtDw$eyUOYOr<2M7#!_E>3ZY
z<JJalF1#`jKFWVobV7S+`Fltkt+m4XF7IF0>(<hZRinAOD*lXDU?@7=M3Et?5G+Ni
zo^W^R=4C)~t0Jl_I}-HBO(;X<qrRd*4+{NIA7dA$g_uH55w)3pkc<y-pl$6|Oi59z
z800Ob8RxVFlrpOs;V;)<o`brlmd2(2dk-6^8=rNHVBRU#D*h6sWS8`Sff<t3V+fwd
zRFv$5n0ZZzF`lY`S`b|po`mVyuM8WFPvws^{+U#TRaBoqCHZ1ypm`?Q`#WkbHO9~<
zKk1aSm9U+aV!nL<Yf`_-La()L8=E9OTG|_qH~Hj!Qde<H$W`I)<{z}G4Uegf*yL%&
z##@!U`9&O6rS@*B&y2HdxyNBNTW2x%!@7^gi`5(2*~FTTqKA(#pStJBZN$-(29N2`
zpVxuF`u7WLblyeT%g&vott`$a_c2))mygH6LXCl%(_q<uTlX`Wu)4~bUN`Co%dxkI
z-BTy+LzV5un85F|UZ|nGn4i*Q$4Izp_&{RO>9NDU(wCeC6nZ`*np!;*IabDvT*zD_
z{;xhgV^4QP+q>(a`#zrgjS0=y%r)eqtao(+GH7jR+KG^Lyh78*J|^mGI3}|lLWh*s
zUmURKaD$g3dZtEg%Xzgo@_3}yb=~n(2omaOzoZ>VwGMuBLFpx+_hGveuoCk`#=*)m
z<|#BnB`L&W`GTubopj~lJ-REQo|i};8%+7M-fcA594~ji1?n7tp8A?y?F=Q@0ym<!
zyhCnV_;)=M&#A^cY$y8obX@yRl`Shf^cQZOk6|Bui&KW2AeKibjqsBP>Gk?H%6CNN
zAG{8z$?C_08>J(U%gMLg%JM}PYkrz#N<mWd<GB2={(+wE5iv~Y!O&3tt}ONPLDy06
z9%j<nw*NA6c~YfQ;8o7uQ&F_wq%7q8hB|unDobG-Rd#}hTCiov;Fx^{l3j&o^py<e
z3@>pSzRXBvnF11OR-G+2Ivh%<iw18HZX_c7H}R$KH{s&XXjeXM{vp33hoakaOIylq
z`mQ{$w^sfH&{4?Issu64;(zyF_!f#EP-dXMP^&Q#g$UOE5QD06>R~U8gGz?XzwP?t
zR7jHR>~cZZw476N1Q29*ZTYG$G^dQ8yvMD7iH9h7Rv=$uxLs;rYk^``@v_~Fk8d{X
z=uOS<F);<ZlQy4)L7E`AZ~I)1b!w*<UM28zqS}gWot@U%y|?MT-wihqCkiQ{sN;UY
z9p`Pt2BHPxW+hE!ng$%DP}ibDUH*9LJMBHvxVRSnwXT99Zp2Fj$<aqL%1LuaPeYGK
zV2neUI)JI7ZYns41j!)7BJCg}d_V8e4|1sf;3AcMa7+mF#OqXE5$I}fc3l8`MONsr
zl?oVVr#SVrc191%2;&(z&9C<j+G+{H94k=}K<}_;^^lJsM&eD1(f-Y3R7*cjI`zwm
z+Y5%yq;JLF{ft^SO=Epn0h@jL!FEHuQ~1DBF?&nUo&ZaZ%Wb}w7z#Gbot*GYmVfD1
zjXe<ly5=>VHPvk!MJ#TRB9SFB=3@&Z*$b_X*u}zxNT>KfSDz~!zZr!6PNZb>GKrfZ
z`4p|&-qdKxi#t)p)9_I66vtfvUHuj<IW3r#3+@@fIX3gfNOobJG8HPiP#Dbc=d?iQ
zoQLgAZwt+4FQ+pQXIc-jsovI6m#it@);7-d2^s@cMqJ0t#x?*!1BkHgdKanb4$>^j
z#;*stj`Q9apOHZ+-JOr_Qp4IrvRW=(ug)}WQauofMuBob4&ERU=un8cWza%Ib1yz6
zT%Lj?yJ)x`I7gsb$}|3bI_GA#e0Y!(@qXwhOyY4oay#K-+#_S$39I3hWbz`BYtH2^
z&h)j<3(WlRT|C!wbjr0zktD9L4l6M?s0|wm?%wW+OmGL~)g8*L>N`3^`BgZqt+d{i
z=2>PbmW7__F3hyI#%Fn$>6N;IK`$GGmRzL1z3{ybM53^inp9$Aw8!%-1owawO+b00
zWK>mg>AjDSO2$vvU}aw@o3TrG1<xMjfVa4Yn&ei{6w*VZ&6O&)o*RNh=gV$)Ra5!A
z>o!TfYilYNdU<y3lIQi8ILgC~z<)DYl1<o)#?Q{vd69Gn{`<V8!cx&8gI<~Fg0CMT
zP|2V41M3?O-Yk}+ce=DHAHkg7ACF1U*C!%Y^EBpi$X*~vQ1D8nJ}rzzt|?^Zf0~2V
zr5MHkyx&J!{S-tTPut>tShaPR!5m{TtDkRrKmeMIruAxMr@&JO+)0cFFguc%Cq^?<
z8NgQyfu5t8v=B{j+l|)*_VuOF979NXL<;z+G!|=c=yFH><rENWAd0(bAUpYd?c|fS
z%kWRMrIMBPb!2hImLV!Gg>B%MMX}HAnlcFxIAC3YCT2;+lu)-C10v5J$|=a?8H~$7
znR1}8%ODF^yM|h^5^{C_ib>zq!v-Y{0L+{CP{nzEUVXjgq#YNyQprBWBQ|lP@;%q`
z>v(vb#ol%{aYTsx)@svfZZP~1Wo0w*!}ma+?!fxED$S>M)N^#KiR#HVGKNO@V`-M7
zk4n!as?Xg!D!4(gf^bJFR#^p8qN@+oUKwt0;T{WDucMMwGLFNWE<OAhS;tT$(Lnn!
zU16}jiQGov+qicVaN8U}8KP0+Eji80^z|fU*UoIRcGW-5g(Zv9bQ{R;PU7>i6?kja
zOjMC6gGO7k8Za!XOm>@+qFM^hJzE=16H`Uyu<Eow(&91SIxCMBz=(&cV>pxelZICp
z^ZXMQ&r+6P$1ng*Bm)r6QHXl2Bn3CZ&p24X;DI-P3@nUFoyf=_<5DLGjBDekYl8Wk
z2IGSl3V|kzv4;31vi$?j?y?eIRw86~W9#$DHzVKn)?X2y*YSN>{<Z56pKX?e?K#<J
z8zKvVQ{s*5g8n%2gKzxAoK6T)NM`DK>!hd;CyuTUXX^fIyY6>U(MQxf<M`RhP=DBt
z(pA#v5i{t^B!X9m-@q<6bSEi}V-|#<y)bNDg3wEdM!d%b+sxaEcN*#!-n51$GcMT;
zuj3AyE#t;VTxo@YhmZ+7j!9F5qE6lduvtMPT|>aJtQHVOTX%ZVx%>`^2-}SuVZ?1Q
z_!r=MpHg$j`-ap>CGOSmhl8ixG?D3k3rgW|##G1artaXJEZBmBNr=w4+w=!wmicxF
z{?|^?b4oeVecU1;v>3RAY3#k2JM!LNxWrsGJ8mzF;bCS3n$qsh+RFf3W>;AFX^N*=
zG>vBYmAL{OCA?0M*Uv~3m^Y%-0<tGP8$>gAiaI_iNGn=eb4Pt}#JG4NF5x^4jZ-GL
z=NPTmLPm2sZIPu*`N5f{OVvt6&gKg(PhGmB)3t@BxQKh)K-1<OOq)ZpD20m{{~Jay
zk`3s@n-P9k;UQ7@U|BY^Xa4tQ*SQ6!my1y_JK<MWhqJe4UhveB{b{r5gQ1k~@OAZ(
z6Tuh@jzsb#rof^3_mQX+^uO3*&c8bvFedGR9yod5H)~*CCcHf^%S2Ldos%-XJTpA5
z3?&mhRn=e^nTK-~13J;?T$QGB6^GIF8VA5ks9-n`rAfd|dn-5wSAZf;S;z~@G;b!~
z5ycB%8TG*x$fIqgwwM)tdl?!6@%?QbxU9s$JPU+>__UCafm>eSAA86<ynnjo8PQhx
zxK@0UaluxnAdRJSKkmT>YYY-70O1!$iWg7x_(M!+tDT{B58j2`#oE>9XKYHX)JxCJ
zhP`;rwtfeu?kYfFinyeKg8T#mV&&ARF<N<&?&r;9H2PPZr}SRl#i)yg=}YtH=4Zz{
z2wKaq#Ogt-j!UmXVX5LScxmhNk*u^v?`jR~>JGX~pY14=6CAca4y!DkVrC2Rup;YK
z<%ez~bQ^Fj1%45(k)P@dnF7LBk0%Uuhk6)#tRPrt1Svd)T)!rnT~+=s2POiMfwI+f
zyI!xG`a;O~L69J|ssC>eIvI2!2w3RB`a;6Qwra_j?iJJmz4|aPsqx&}#J(%_+;Sg?
zKrW;RTe(&#vQ_PD<vvdOpYNOg8lkOF&zTciPA{t`A%jP<uOshb2=ZlWLyjJGC%kQ}
zPHX3vfoWCgT+Tw*bxEG)K-2M3AWGm_01CdB0w#9NTH;Jvs>x-iz3ACKr^Q4EmU(sC
zm-nY}PDHXf;%{08!T_KFxGFlRRRcIPP-^6As>{@MDH#9r*!y6gDi%A?A3pvv0nuu0
zMu*&L{DRA+;x@_GQ`NnzyQQJ|oGqHi?cMVWAU!Gns(NzKW5K1Z>KhV{_i1D#!rgw0
zTk0}ZY!x!o)R&%KX&8dD4drNxcm2V<?Fl)!U`34*m<f5y1-r=}{~HqSIe-M=UkY(<
z84{6(6k#Q4bquvYd}D%o{iJX=Fik8-7f0)=<*>2M^WS=dq4wpVch_fd=M94MgwEP+
zRW$EoS;|FS$?)M8ccQ82W<(&#hz$hXam7GeF*J5H#nr&tRNks;mFCRCg8PvpzAbnC
z-XRP6mWL@K&!;Ag#p}hoZQD^*!S46Z?$q-O13Qk3Xe*;fiF<>cn=0?R&8nwPznv+~
zt4PzMFU{4o@dP5pDcIbWbjXY1Ix}P6(AgigWh!C6Q)?U)tDG9K7*v#6&f5-U)bPHi
z4rL}>kln;ok9n$pw_Sswa7u8JxE6YY8CJ?JzT8Y{(wW$)vtn>mfu6I+-$436IgKtB
zIo<Vk=!w_zYHj>!{WH+{GAcxeqXjtUc%iD}%g6J{mHPDVeZ*qP_}DS(mh^9r5vFSM
zK1PENEW-lbrN_SZ`_7TGg1$QQ{ixD0IJN>BeL+eT1qm}U&NQ|Z)EQ)$m22%bQFA|r
zyS_=HHIC`SUo97=iUF4c!&}&Kf+QbEGP$Gm7xlyz_`1mgpNiW}k52tYQ??96nX@6e
zuiFJr)hrJnK0MY;uGupY|DH>2>^T@BkBR8+B>+~>TJF+MNLB9fx+9K@+*ZJIn>BJr
zBIS1rL7D36uhW)B{^`IrD0IDNt+WBH9czqohKI-X?RAK(it$sRYY*2NjPThr+@_WN
zS(V1x%tF*8UGo_$Mrb7hMiy{i0|!5v{6>Dn8W{AWd6jVVRg3+y`q5}L-C}jIU=d_d
zM9lyW%zx@Y)e(BYhY2SGo<;=0GnLS&ZxDI8gsQ@LVL%&QCYA9>7yq<9!J@v({YTY@
z-FbQUa-%bmtiCf0LZ<vlIp22l^=X8^&F16h;L%sxUD$@}`)#83(s-xd%Xt)4lIupb
zi-_yhIkxtKsxbNi*riq58<622ZQD`nE0=F773;HT)a&Uj<l#o$SpcfG_4hn!Cpm6W
zuQ}W)t8RKlSg7hF9eTtBl@uo=ApJ8hNxc&n%_}EG18=N#a7ahTh>ByXh#Uw@2bK##
z1!5>Wwsdp;@s4O~E9BwuJlm`sOWbTySBqOClTLd%Mtog#v|iG!YoRMWglWtE4e}f2
zu^!s8k#X99O~DC1fU)Ytc!A#RdDZi#YKT}5oP1J-zFvAdnc6Ip0yI#Olc!8v@8nsS
za0SBpv@}%_{Hu-CY?G>@gyKPrkgVt?#6gy&`sk^vs`>~v?ix-mFRV7X*Aj`I=k;z2
zF9Ara#cA0W_s{(FB4KeAe_iA&>y6Femq9TvCeA+$gK0Q-Oo^3rW2@b&D8<O1<rw+@
zK2MuPRtL7@uj+F}8w)38W->&S_Ra}$whEBNpj_a{$JW1|FrLpNZ+8qViSF)GmfeXH
zc@6um$(ej1siOP)eZlFnCMSG@%w1gIc-9Bh!A?aP4@B)qMD;d)r*ri#bazEQ*Fdl4
zKjTk(TW7yT=GzdUk;s)7j}reN6^!aQ;b<_ziy^^whW1Vx{c(ATeBAezK>At-35)=r
zb}r&9$^WBteh*(j`NC}xbaX=;o!<-BS0F&-R!6d}o4-nd186G^)rt*OOs?y*km~V(
z4Pme)S`i!x4<}UCM6VejEFWc9>vhUs$zN1zj!<(WmY0k*1~OTh^&ZC8?`!t^HwaX=
zIvMkdS6nT7&X7*&n2%*vO3`(_h)OT*j@jxUayGPiF1C0gUYwlmqB%9X8nHb(?~N#X
zXXaOcd{pO5q}o`DYpJoYsa5cyVHrSLfgVCEzMHNFGqB+Vj>FT`v1PKG4tZ-Q(Ph~$
zshLe>eh2+L^BgbgYsR<by#1rIaUrgSi%OlF_4tUrKr}^1QH)6;H2sKRNgG7h<7`a$
zh33#^V+bR4f7zS65I#qHc$`TbXda|~Ru)~h?_I}oU9k&pxg~KX@O3^_-fl6U1rBXJ
zH;HXWH&{099_SuM$5J0Al?uf~hBcp<3&fsX`Fs+=g*&N2vi@l$dOZF|a}ln(oZ03R
zb$g)MrR<Y+q!g*8YQLfhiT}CeJ1FwGG$`VSufN{(gD;gHt#FWtuzRL2gs3rKA`41v
z-AQ`wQJ6iq@n(&+GEs!cr(-v3CiJUgCKS`Qz=-pSGO{H_7<mC8A&?vvKpz23P*>_i
zUH+U!tWu;G6~Qb0TEmY4(<$GiHP6|{RC3UXO7QfLMhj0(_0xC@#7)KVD`b7_x_oAT
zi5Shu-X_Lnh5X`Hel;i{4s;%ZeO<(&vjlnlygC)(+3GW*ORkKHCfP2|TYXFXGtX|u
zuy(sthpAJNS4y^I$`-bN+gb0%6pptam{my5VbJ&-MXGD=uyGV)sgk0;FK*K6`F#FP
z@Olqa;UG(@z=tY*;CV4Xblz9lHC<7mAmdq~d6xPpKldVFRm-;5f8jv0Vnz4$eG*wp
zyC4x^ApuVjynlCJ2(j%nEUPMe>m8#Z89*5w#X6j<;`LZd|Co7ncQkE&cX!Y2Gu!5<
zaXKG>XFJcVb&(v)PY>U<YFZ(G9i*{3UDXWQJHGVI(kwk8wPB>gR31GxrU<L1F0M(&
zSU{x49@8i3OldC7JN`@FvZHBNi8#{2K0HDmrxR(+u`wg8M_PV&%-z1BFZ(z8y`X`m
zfwjpcq)|vJPvbysvC*;0j#jg=g~DQRc|7f^{L)jp$g|qw;P_-BjghIvZ`5&3;Hu#+
zMCpSn_md@vVeV&M!pUy_C%7UQx@n3ZDBSPHPUMHGFw-41LPZ*?;GsxGLTE&lnBp>H
z`M3*hsmf$$S0UA<c`_5*x8Dzr*l&34pQ+}CElIKyV;!ce=VCdyt3E-xfD7@lqS>qA
zB8#S32aRU+7Hb+`sIeN^iJ+`sk6FmQcKAf<IfsffK7E=4<W=w?Artwl5R4_7e2D3Z
z5+ul~(#Dg?dJ;6!N?FL|Fi7*_>g?zmMv8TrL7JPNcUY1>t=ca}&U`0iR4+o_E|>AV
zR8IpX9P6sy)V^>K5V}woIO%u!rjzc6jBSi4aG@p#54as&TRH6#X|~ETvri`s#%vvu
z=5h`vE(ICWpXk5X@``|9-KjlqCcz{O#iT^m0u4@~=nI(n=*FXXy<&F9gYqmvWqmVg
z$5{*op8+#_KyA&EEDu&H*Lvm)H3$D$FKKtnB@iFFy!TvHe3|8fZk6?<>-Bo>;vBX@
zftFtENm@j#?Er6lyO1}XqX4O9HdWnK)lQ9$CS7Pd&A%1bkyB914Gyu5sxV-C%<;d>
zeu&rofty!>D5yoL9mI#(DTQB2MS&SuH)da3S2Nf6bvojjbOP!CJzj&XM}v<wc0E8n
zo~cZNhWmN#C?W0kT8f0+V;Xz6zC}PCu=O#)6lKZxf`;BC<(`<i-ku(7jOzsj>A+6L
zHU$|Vj_T-OV@{|{iSC|#Hy}?zFnWm`vpY|@rY)~kM159aH?A~RVO`!SPBZ7K2R0jC
zAcoNReaH|?3+Xdx5%cbIVZHRX@9XpG7I(~L(_w5fHFcxrX%jwV&2_e^$Y@eWt?90_
zWMj#+F=aA^QRc;MwX_XRv`@S2%-7EIIUQ{-VEnALXv0_Yg2XX0to@D=%b3GJl~Qax
z&2vT9pojd=tznsh6wMAB9!8!GMjp(pOnx}QAB|~(9yfnZ(P1}sQ4s34*gwBlU;|^Y
z!_7O&cZU%6gamooRm%{?6RXcdWVL^{#7%s{WGi#fHpMo4lG9qtQPo=IS_!7ERK$Ur
z2N`Y0C^U%wqMFTT6)6$iH=hQ<m{@TCJg78ox?HwDtK}+q+N>y1Z&EDVXsIkb-1^d9
zF|bNm6xLQ$1$B3Ftw<5omY$nrXh@0S<~m!Q9J9)dbJt0elnPwELTgQbi)l-6if0kF
zKp5rJu43j19WkWz&N}cXD;_zHihN76r#YENi#cTE)2{!AA0KPZoKi=fyI|tVuDx=J
z1ED3qPw|)tI7<Mkf>D;02Ih%|^s+QK$1=>k1`pOq?+wbzp=0Gtns)N%=lr8#+?~FG
zp05wRn}T=ssmiWi=L{%doj1TTXvf%wForEkKcl6ri++Fm-E#_Z7JPKv`j3UFa4Bo8
zrm?k>b2YBpokK<y)+YFlF|eAkuXzWKosKm4E;qOBzFAt-k$+OspfwqSUfaq%nKpTU
zOIH`u%&NFN+$2esYKa>YgD)qyyk=*g`4g(Qm7nYJ-tw6ym3DfLWWLY@pTsk{#hVp{
zF@FLC?gVZ?t2jEDta!%1o*cs3296{N?$FO<TpySZU5`4V-*bYrRLb(JAWS(9*VkPd
zl|qKP#Gg)LHSfo6s#$tKcyz9#S&s)}tt!g*{jsv}XhMS0!O^buVL29<>)N824zwLh
zxJ?P6W%9r7$Rzdxu+#~F?{%1yX#57yi%7ZZcN#Lqf<aA)ky<q=z_Ao!!Ph#Q0u0{7
ztQZRGIG|aOgPA{~!IS*O-R5yUeT<Ggy+yCYeJpCS(wig5kq41`Yr&ZXKDrRob?_Lw
zt6*j7eI8Fo+wg*EN{i?1W2z(J>!yN_F8}PfRhOw_=8N_#R!+Qh*aexYFSZzN3{MLV
zwb#R}I?PVAMKF#f6tsU)#5i3KvpgWCpG@V2@C1o|fX&Z$haX%K2`F_de1;UUus`{j
zppWmJn*c7UbtZ7s(I3(fSBokY#f04OQ{Bp#Y4~T?5<V@ikx*;*qrn-5xx<HOjcyq>
z>T<34Tfk%SV^YY~$)i=X{u~5Phjx*Z*?4^ZpgPq?L@l0;@9y>dp5iFyT9sPeb?ZRa
zYjolQ6W%hKkH^kL{E;xWDa}ds*30mxvSAFH>2!ftg$cq4iO;Nn0q>v}fmF_0klMRE
z1TzQHlp-eWQ~*Od7%vnR_=uxWQ*j?KLOUq-pdO7Y(?r9NGce4U47R5dOmX8<sbt2G
zzGQ1r%@e9<p)7!8*MOXo7T0B|pwC&xCj)=Z5J^?zkKy3`KS$awhu2i|i@d&#xwc95
z*8GF-f-}oO8M;obiSt_T0@g?5zt!PY^NTt(+LRfdyfRj=iTTGnbh<QBZKuB17xazk
zum-lQ0O|W!{?PbwG6qMDD3x`>)G+7>cjE}Oi0nNVNKYZCJl<HR-zw9Vyua0jl0O92
zFz@O}9cvTEV(D?+O_sZ_Ih42|WWgd4x`sL<zVF~%%2np>)sO153YV=f6s+eJNF{V=
zaH<e^dw#zk%&-nYC-*HJ1P@u8$5~)F0G*E3X|$owA`bs&Cjdq+E-{8s!dYHD%%t67
z0lWoyGTOrVbn#NYVruOQ+@g9Oc4JKQ;oFG<npS&N;HiIblbVwHF{T6JNSFvTYt?Qz
z3DXmL%}Ycw5w1MCLsR0fr+MA&t4?!R)bdyJ92^hXW)t$Jyj;#AWEt*U3Pmh9D7m_A
z$GBGGFNQ%Gc1@}{&RNiPcZztN;T!BQ^{eC`HKMB&>IzI9?tZbSn4fUCn!pa@H~(Ss
z+SViw)*}?jRnTUJi0-y@{k6_S!G;yMm95Xbc{&^>T({BU-r)QIKM`M4pUzeHTJL**
z-Z>a~2W_UJB1#fLey4o>9P?1Usf3r#3Cq!@6lt7jM1P2wShmF!77{c*OZnjF-McQi
z?>pxuIvFG~`$GdB8hkH0UxhZ%xF|{wwLwFt0@wSl_Fqv`sn7n?=p+XpA(RIXex`%H
z5V7kADwU7Rn)_Rtmv)=xQW9M%xAu^#$l>49i1dpYW7SKzLCYlSmWESfn@^IpTe7N+
zYR}P937UBBhE6Zog6uzaS_P<vg{nHWEoQH&@SbH-OX+;B7UtgcYd!MuYRRhd8}q*~
zbhWh(((6iV-4lvtS@B;iD@=HwP`2m(RPR|pFLH7ACJO^=1ubY<xJsNd2>#TgX<eW<
zt%iaA4bwc>VqQ3v;?mVrMG6)Ef$HpJ$t;HyrUaIoif)HQOIS?u@6_Wi<>LxkGsufD
zPEbLp6{9n3PQ0)ev`?S~i_~6S?y>hbAI1{Kh*Ledg?e52X`F0X5Dhz2_dHB-j2MLT
zh$*gU6=?1QMV(Om)k73Q9lez^285fF5+zC&quPcDo&qOK$g(BS%K}7HBAG^_`yj0u
zr>B&H62vf;ielETU%}ObW@a2u!_(+&speqJP}i0xw|#Z)`sYLr&HTr-(5RTC*2{;`
z{Nrfg8k*ZtmD!g4S%c-#3SH-kfX#<uJ!|Lt^^&ITCub+OjO9P>^7qf35cA1CNV7|#
zXmWhY=((3FHrnDY&$+=46gpMMDdJ(IOT6DnUSVaKRxF*XSYX2ZrAJ?j|61MTP<1Lu
z#SNcy9HqA#FP7H19gR~0tD1d!b04WDE2=t5)i}mGK7o$+{);bLnq3!HKj*MXmAD*=
zFUMz^e}gn5{jlycWtwB+w6kdT$ZWH+c=+oqg-sUu*l_~D3YMraVeLXCc0|(zxy(f<
zmBQ=0=aD}t)ZWE-ZBTDX-LxruYdu~M8nusdXpfrQ6im`7<<67O?1)@>yc-#Nx3M8X
z3Z-k|<PHPmGZddSFqfL<OM!g44ZLV{x}YknP~DjmHr?|L?asiZ@+s&4<4)xb{2{<m
zd=Uj6uw2NG^!*a@_x}X5I!wi>E+~7ngXU9Y$b2e-XI85gL}duqRHH)i!36USnqXrP
z-=PRBBS$_;RI$gb38*HNHJ0bBz@i$r2(B4ORhWiS8AyCF4wH`f5-12~nrl={F7x;D
zI+0JVq&bV(wz_xAt>;d9YTbbcx$gO`t!<s(IobNY=H9$^`MJw;l;2UKUO7b<Xy`eU
zrc7e%NHjCJeVPMoO(Gi`Z7=?%;EC)A#grVHh{tu8mY&jUW-M3$yf&75j@t{iYbL`N
z0{JRu;R^Nyb{oqZCGZw;y`4DV8Zd+*W730>?Fdk*O=7U6gzNy;+Sr3PZ$0qH&08M5
zsQqW{@IU|QjXkmM4YAmUi@UlmZtLn&<;NlrxcBi}pf6d;tQtc!fUab+2&7~P#WZY@
zjmB&U!~{*Ua-8{JU`B!QVlp!`N>tO22qrQJ#)=PVPraz6i&xRbTIC0;R<5M4ucnig
zXICq4Q$Nr%$_2RvLNidaiAY63xd839pe!OqiV=xIwwrrnb`}M-Bd(B_1+)v$FE%BC
zR#-r>Nt-}~Nn#6eI;~C{3P)U?XvK!cd8L7@))~tdmR8<0{n}gni&rf6+36e0muvdo
z=eGN<s+|>VtjU|ua^Bq1&<#sNtt%GIDHJJf#W8Q@3RxR{WC@AoV~orgjvE9r9<dhL
z=%KlttyNJ#v?t^ua{^E(7uuecn*tQ_vtXKc0M4ce0fq8mH0@YimMyBGiAd65)}}Cp
zQoYh8N4wd+Wc8fY%PyG~+;Q97Y2RB|K5FyCYj1gbUH89i8r#LqnIE1psqq|*es<xU
zu{8}d<`!2jxp3Q<t7fmbF%M-7HEMj^gRB-kg&AiY0W%KcmDKp<e&rV$AHAB@p(AFT
zm&Jpfm<7d^fND{0Dk5wO!(Efh=bUO%*2(grh)k4aVk_v>c(5!wqG=J~CcYNeZem5P
zU)<icYen(+@uLd5x?1nq+opJhEq#YNJF`maE~w+?pPIMw=9~JCV0+s7*YnNb6TE_&
zup}Iy4bQ7-u+lcl|K@M5aXPKUV<b&BoGu$-8Pi$nFw|QsXSRIuCw~32zxZ}Ncc1Mw
z#9X7xsh?{=WL;b<k-%{|3$QioAh0$1SORSgd4lwP%5{;k#DaPUq*DfRv>233Y~cl4
z`lvOjp3<n%0yzA2Od|UkkLoD-bm{R`+a7)><C^PsZaDPOFNBNc&R=Jpr)`c*sLIAM
z>gIy<Ibkg5$GJojSj;a*n0O&Hp&t_fX6AGNv#i^H^64M|GshVK%yiwfVfk^*=d(_I
z$=y|>%DHBa;|m$Cn-kais&H;#LP8hKy3`s^T;`labu+6AJQ39`rbnThz;zocb&Jv~
zJGITf-go~izgAv+b=hs#uG@)!{Epb~-tLiq_nX684nDGJ>%oJpJuP5!EKAxP5n4+l
zU~Kpj#}>LvS!tnPDH&Wo?NYie^gYFM51p)`Q!Z8ZY*BvDPQw}+UZ_0Yr96)P(~xfE
zZQ%f;tAW_Tc7IAm3&jkLU5nOx%<f?F(+Wb|VP_D(nQ+RAC3WaJ6Sk&%$Z%v29u8c_
z11umpLx?QPloTroKI<WptsIauBh!%vk0m&;BARol=_)pWHgLRsDm<CPTZ`~yY)HMD
zma9)b#pTh?b)A~7OO@*{r=H|6f7>m5{;l;5jwq(NFTFIEZc%-SZsi?eb%HJdq{>Kr
z#Q<~}P|A_Okf|KD=W+o^^8zdWsGtI*dDUOr_anND>|AdGaYtmip|T3p>e?u-zl~+M
z5P3(+wMyBzq*&&;Ayheoic{HM)3Wm95pHgISK|f#imZ!fEM4rc+%i<%4W^o}<Z4y-
zO>CKSn(ot(X2x5n`^v!|i;y3aSS|8aRamY8iUmWlnp&htO$ZpOwgyjY18`b_yyXXa
z4^fzL-dmfG<xy!$bIh5PCgznxNa;{~bkYVzAY(ZMre>@nP>UoRElPfb)v5*Q6k$*p
zs9>$Cg7t%fg&B~m^4M7^*VB3MjOw;w7~VXi3O@DL=?u$}+5PWxZGsyt_X=_&SQ@BM
zvq7OEtgW1r(Hrw(1iUOE;sZ=-Im|4W39?<zLIPT(QiOjhQsJM9g7H#S_*cRgFi9@k
zkl764VO^%t>SgA6sa4L-M+sM!gI1Nps1gdH>R4DaqNYJuVg#IA2P4n`{L}25(s4DV
z7f$V%(A>IY@%(vr-#B`~lyeTf@nJ({prO99s-6Zz=hX#sEv0qASql@mPu_EyFTT-w
zjqMc;9~eJjE_(mGiZi6`1-Y0>%%>v_83&MtvVv-112rbn$tk2ET(dU_ZD{QNFWY{S
zhB&0{Rc=4EoZB<#c^Sh!Pe07_hCw#+gLyOr-KZXnrDN$3N2d~YQ}UT#FxhWC{h5}5
zXC{-c6k!sqY0pe7Ow6Q%`nmq2+df)!oO`W2t*=67a>*j=@xb#t;rUz1Em$!>`1#wC
z&zD-jI{_VeUQo`&^%gsV@~x~j+n_bKMCGg&RE@LdGG0khW2F4jU;wB#mTg-8w`7|x
z`rEQiGt-H*vt#Xy84c%ltgVVvMO#kI14?NUbm{;)*}!TjCRYumls!c$u#k$va;BzN
z$_X(rRQ-v#S&fmI8U9#2UyYHKU?NJ7R)n3w`NTNHvy4jlR?veC!o<!1r_%_UzWu=9
z%QSj+C!=d&=+>fSfGx%F_C$c~<VkK1y)zYUOY*{8B7)UWgf%EcwHm;=Lue7HASmku
z6>Lj$A}wi^tNQ%tnPN#OE?0_cb<}QT5j6yzh--}0p3U6)Fl7npkkiFrFxZ^Qq-j=D
zA$?{1Ih%jGk()7*z6AYwVQZXz`saHtm^69r+{u$J(7fQjV!}l?bhb{o!hLdLazatA
z{=AvdhB>oQZgPZZtHAch8Fg`@nuCM~flEo&P~2=W;*gmc)RZv=HDwD*#uJhlN+CZP
zPe+@wtKKagK%0U;o_a)%dzYWtx1MwNeZ*}%^{q1cRVvc0h(8s%P6>BlVt~HONeOJI
zzE2r8m>;>xk;De`v7})FAWX^&%K3@BO#(s4sKQCQ!6wCwO%aGZ#HGq|IP>yLe2nA{
z&~O+u&k`Aym^FnZCXR?Drp=;*tapE;ba-g7-7mGPM*X^5HyC#-%Im{`jI|lo3^Mk|
z?y}R%Nea8>5Xe}};m-Ol)Gy0-p?;xpZotsJF**eO3uR&&0$5*Q{~y~@FkmS2hqPtt
z-=i%<+W?v%tqqFnk4z0D8UezXDS>1=zJPYvzMCfY?`nr*WSZF273}GBnVnD*>CH+l
zYgcjjpluy22jE2vvop003}(p_ma{ZioI$0Mk(+8XU~1l~N<{-(6&WvJQW0ZFJ_OT2
zB&RG#^t4WlkWxtR)XdY<3j<?sA12Sh_3uP%G2jKv$(50q1}CAwP_^_dN~=@LC(*9b
zX;{rTLo5J#ieYgz*c64`$W`BNW{_hXVwG_-68G~DLMzj!%JlJOJgd~wCAt|F!D68Y
zyW26kf_F+V3D^G|+%K2e37L35hWBh>_cOAn7^1u*NM9!E6G6-F$0x=cBPq=Z)Pn8Z
z0|T6BERJ*kOK^WH8)rnonv%d2K}=`N`?%G-53UQyJ3Z%v;Tr^9#N8#Fs%pzuVf@q8
z6*MAUUo|Ah=!0$_3{o9jMpI++trJ_8x&(T9KsUX{@J`&gav?Ld<`%io0bEE)CT|Kz
z*<qPx^nz2$$|Ez`a#ciYLNJ6(<Umt4cq5stOe2#3Gn!NvA7N2)Q^0IGQAIpW0vwi8
z(UycbL_YL9K7ezmJh?H2bV!#fd$^W0+{7f{A=jc*B`^<_$Qo9|@@LHsu(u3gZ9RAz
z>Y6kQG%Z+JbAz%;wX%l6IlDE1bJoM@GFq6tr^(e8vUL`^)N?Mx0wL2}&<s(_UsV<~
zaZ>Z-DO0CSpK(r$s;yJVX{s&9oo>q|A=Ls{^0R2_%<O+$(NH?78=7&|W2#ej9Lz+9
z@dk4eIlK@h{xB$SMZ=-Ic@ijZJw`sSfKmMHfZlNaof+)yzWrY=JR{&6zXo;0KcMZX
zyw2QqdKmTD8C%b3R2`RW{}-(N-VxegGF<z8Y3(okE83rgYdLctUP`FH@9M>XN=)^o
zllA33#9%wEFVf<mTmtiWX;3c2d3^tHJ=>Sov$nxKlS>xEJkDG4z36vw*hsVT%-ueK
z4%6RZu1@`*&($+iJbTFGWpPcF?j4NoeiA0DhNHYMD2)!wg`lkhp(GT2g7A5upNdr@
zqu7GGImD~gtLlM4K~Va#VjL8koBwO3`UqypnaANcXuUCGn4way>cILl)8&V!8!wc!
zk|=+G?KeXr1`7YX3{+>Jt+Z7+(1M2+Wk0?gRN(2YbbgC+fG&U+_Nk(x7A@&tC(IPK
zk{p<wm1HG}Wg%rtla%%FDd3=7k-_<ws|d>sGbop+sKw>tc!QSDvO2T%g(Yg)7>nwb
zW+8A8=plU1BlcvHf=absP!{wl!m>@btSWK+I&Ty2Pzwer`YE_FgJ4na5+vx|w=7$+
z@w%HIQ2rzSgIkua(C5~dc5K~vc`KKH<@ZrC8CGui4Ygl3UnS<0`1|vf52d5Z@h7Il
z&B4ni0ugebzV*X%_ybYp^}f6)koT64rZfM9;R0%VxvAVaBOwE#R8K<(ycb0HKM?}|
zi4gerKnN7RWnsay<3rTY=_r-kQ62x26aUGHe-BR7+>!`Ur+D!T;Kf-Z@nX{2O5?|P
z!nFD)NB)x||4tm4vcmZUPj&!Lp3f~DQB#j#l%;X$bEk)j{>iZaWZ1upGWVTktaB=!
zEco$yVodIiIDjZ?LPS|yC#zLKdKsv!%o(G@!pdT(Z~$6X^W{SrTpqxe<vY^j%Wt#T
znVn2ei$%ulRE3$1XPO8C)@34Syd$+E31<q0Fx`Nl6*_A#`bb2!n^J%!VMjOt9Pbd5
z!0{Vj7z`Z$&f7mi<oFNiB2G*F2pms|3+Ap@*hR*X`$?<`TZ6Ervd_T4+7NLr8kCK=
z>T`Th8plF-jiz3y(ZWE?9a${Hjpf_8-%A$r<m7x)uRH;cqZ-k7^A<;OT}~ZKaMwgJ
z6A~-(RbmdL3^OA{_BE+XD2W(%<a<YrLwAHSCnhTPXmwdNY7Oc*T2oh3=csAo)B+3M
zVQ1z1;GrZ7`Gwk=#2!dmhl+AX>6pdWO}^`hKQhK&IrV}IrdIaseqfXL<{K_r9+*7$
z{K)}dO^t6w+tRfR{Z5}#Tg)xmHMjhUn?|=zh}OFUQ)|QZlb4R$vUhHK>-;&RXGKQW
zj$Ke6oK+v45x8W{jn|xfok!SU(%k<ae5<gV<P#4Gknbn<yLE&aCGF|xujRQZ^w&x@
zyv0B{)d!Y2R|5UD3NDJllZzQLSlvmPcvx%7EiCdNZA2_WA#%BoLBKLk?yMkSoe`Zj
zQFqRw(IX+imelK?a;o63<qv(?_RdHEuw`_aaaYsK-9oN%`>8huP+$ks!#|iFo}u*c
z4W&ncx(9H{P<j-k&~Qr_HOk5ZDmCQn3Y?Wb&;Ohr18BIo2z+FEB*AbuqJ2LiK~jjg
z-0@`0O{K>kpofnHaj(WCJ<5mDqiTR2qZmC(MYcSsGMyfyaoJEBe^iP`7=IM|k|go4
z&9%$_=Ojs$GMojQG$iax=XbFR|3s?lA)R58^zi|G_K`#6@d5hmSLw45<+3L%FVpv?
z(duB5R`Jryv<u<16l(?_V`O^($o7Cmek6@-(xQ<3I9xvfl<NV?tw3Tu!bs;Sy=+;M
zbmd68gZoD&om_eukiKB)3KkOlj+r)6RM2)7bj|z6#|9UF=aBwnaD4F8j|T=;lNeVo
zjB7awgO+)8U}UR?a{lP#XyV?lV?k7CdZkE+X&xBQ3bD7RpaN5sC2FLwLe)X3!4)l1
zJ=Ry%s2|q^<MnC+40TaW2@Iz;*TlBjs27<49bNS|2+;I&T52TI6QU!J>ZD|(abkkD
zbZRo<I3krbQxb;DsU=DB@=H^;Op?wor|29zOFGLDQ1N4fvPbAWozOLES;-JW_mqgE
z;KO?L!}?%i-+D&o`Wn%C7|86eVw$Sty9q6ijetHYI7@=Nh7dd*u61OBU$^<Ht2c3*
zS6;ozl%V+R^CwQe;DX5$&mWoK*JEKzYe&baMJa;sZ8&chD{Mh|^`WqZe+j5vOPa{j
zB<4YC`@?b`-z!zA)RwFahDB-&#ksHtXfC<JjO=y5)Z>!0M)D#k4_H=i6Mlx9ltxwK
zN*<3p#-#LKSnM(RwTy+`)>u}_C=~rsF=oy4;F-f3a}^Y_g6KR%r6|N-#7RL(#i%6V
zdkk3)G{@4IJW@A6^W@gI>7d4?{&WplO>51xiVl#R<5#Yme*Yxz<fkt_IVRe3r3!l-
zyM52P?2U)6f3d#d<tx_TyYaH~r#%?Wa(?m({6Mc+cjfi>e?V)T;p^9o8?`z#`S9Gu
zX^6-x+Adskd*`${7fdhP91_=jz}X#$j!enCzE2s>xDR(YO2UgvIf=sH($vOgnd-)x
zF0;DHYe6Pf<>+~{&7PB6>}81reJ++$YF&v+oDtB+Pp`u{4UPQjP3@_=dgC3>zB3q^
ze4=q^@r{idnZ9GV)nFglNj6xF?L(2P3p0A-ZWX!Oty7oVB!KJO47oZlk$?6gkRC?-
z9Z@L{cM`HDBKunFh!EC=?lVGIgVv`b!&tZO|8k-7^@v#3njb2+E0z?Vb$UP1GG6sv
z2;IdaAar{Z2;JDjf<h)+v|<nRu!s%k{}81cZM~6^x_^R&A0xwc(`%Eqo!ZAe(8oOB
z&2>Zjh`SNzt9{JZr5F|an9urHl<1>BZ7hq#BWx_)dFo(dSF?xhsyF{kT}>HrBai6(
zk9RWLEjZ$!^1-ChmNvAYbJTu&S%o6X!8?Zb+z&nXs6F@Sux3#ThvjSy3P>u|rO4n3
z^*Az6-{h*5EE%ft4>ai+r7eN?D-sGn7$2o-^azSY+`Zdm<^4#ZQ6e~yL^dFl2qTN%
zb4CtG*WO9eq$}^!$+RLFoGM7t>EdLm%rF%DMM5?AJ)jbd%NmF~^dnG$o2~C+^+=i|
zN?clfQ>wU$Ei_1#H$}wF`~TRM-Gy6@oP3Ahmvd?gVxYiH!-@NqC1MI3P1QFsG=sDf
zCrksXXrvgm*wfXCz6`rpY7H6J|2Vv4uzgtuYMe6Y+15Bk#JLHN;MT5zy;+hi#u}$M
z+@E9X+0$#B5LkYs8YjnJTjLZNRO8f=G`fD@Rg$!}QtRALUa9u~h4eZ!#^3J<yGSt^
zMP63-bCF9kdZlWH!H<M+1gSJy-NMDnws1j`VEP&}vbjJawmKq5D$Sv?(Nuj5rmwO3
z7?mAPJa;;(kKrm6%aIqyFf1ygr6<$qv|%AskJ;p~0YOZTG)M;c5Fr=jOfzCo;VT17
zDEErixS+RWtHQ9l6ow?1u&wW89abVaaDA#hJF#>Cd=Xc;nlcO!e_hz=!a0%L1~~<D
z>SsI4&6{@JU9%s$k@MWO;1BL+mz;MCw_wg&*$;Ae%h8ojUj6fvTjU-0Jm#x>VE*>U
z>A@>|){a|w-m!%f+QOHNE1o(h+7@_3x%a^8qdU(d{lEU<PhNle;-!^io>DPU89u#_
z*vMVvrUBI03((C2E2$0Kyfg&ac-}=>7l0JWL-~342w<}Jz|n@bl6^7aZi#WNEAJvK
z=D=-O3+F`#AZF#fi*VV=g2h;xbwQh`A`T7*JIjycy$!;no&S$_(@uxP)kIIP+fa4Y
zM<6Q@BiD@)$YD@&bB067%}t=>9!8X0XZG3Q<f1?r4lCEX|I4;v@p5@bVlBr1zwKm(
zz%cFniluRfYbbprnUhPlbq{N6{s?U?7_O~_gWBpo>$Wl}J6wAUKic^*%kz!U<Ydw&
z*=8SW^WEgSv^LMfDF_01p?;wK@%&cXyrlhd&v0G&LDG0}B5(`qP*E$+J;5qnnBFZp
zzZK@5^O8%WBX?08X|2j|eg20pSQ(z!&kkC-lCqaR0-c96_mC>wBkVLzkqU!SAS}B<
z`}sl)A!~rv2v3%$nn($>s4`V8mJN&p!e!+-u}TW;2B}yC!m~iTtYh#5-#ueTNYf2R
zdH*o^y5VTSjUSqP;mM_B9>0%Y24_Z~MudPijqBLzxdb|gWDmwIiRE-owxNtE6gQ|#
z=fn`(NlHa=Ajw@kICb^3xA;;-Ht73yY-?j<>)3|YrlzV&dT8ZJ<+Hg>O%;Lh-1>Rr
z#;fb9dD7ayj{j6s3o^z7HtaPjW|dr$G2n+$ushQ*;?PP*<0pVS=VAobii$NV`Ni=W
zTHY)=oGyKSaazGu38?Ov#lc+7LNT71P4dbI7hI{x&|$XOJxSDCd8=u}&dRN{wuC}%
zrMh3NY;C>r+S#qjH$T5>LF-le?3!XayLJ6lbWPKRcN}?oM#~*XuA*=JJGBSyeQo0L
z8(|cx?llE2Zk+JTO+Q}FZS?$G-$~EEJ@VXDs(r(7V?oB+h>Oe~0Xa5%1mswE(wMYk
zETIRxt(HHZ2}Bmv!C)ZSFe-x6!DIy?S?}<-wh_^7dqlefTPGPs5KrIHMy%u+0(R?T
zj^OF7<nJD+%TU|72iloSE*znqc_Xwle-JI>K+9}~;%3CnoW7&ENo_E=$t}rxk6{~~
zzNSQNx0mf39Y8%WOv|PGpi~}~T_BUZp=6c@y3u3j(QC_5h0vfr^^_V-@KKTak-F3s
z?R3oJ6y+eUgf{w}+%keF8q(F%%QlDUwPw`m`Ck=Z7kJA`oWmjPx`HYW#@Qh3BG-(7
zvP;J!`<#xot2HKb!n+ka+cPn_KY)1V{Q+*=arEP%`vcf@vvd32xKjDkZ`cL_eSO@m
z|J)$pfB6Oh%nw$1&<i|RKq|=6p<IYEqbw{tVd@o!*t&AtiMU&xRc;F!ka3?nvoJk~
z<<XJ>)r_2>iz2v_lo>rc1|Dg$jlk4bMO4)vq1%#CkYPCd;h~d`BTER;B0i)AH10(S
zZENQd4^k~|cEPz2it94W7&X%IDKlRbF~g(o7$PDxV=R*aXJi@-0d=*IE(3!<Vxoq*
zoDQr9Eh08M@q|ThF<C-a(8>H8eO;WZub1mS<%Owv`$k$yh1cIyl()EbbQHZPxk9v+
z`u~senZ~SqleF`-U2MOWN>WRHNMhM2D+|I{^e&Zzn8kqjmK9*eg+uYGtZeK<723>d
zu*-wNF4csvSTPEQmGGEaK`nqWEsR>pJZ@@d5oImn4vn(L3^rz97{!>`;bakATefOw
zXNfXGCrDM+81eY=OBJGQ)TWFI=g9H=K}Mh88`#(6<;}mW8}sspd!Kdm-NeUkzh_;}
zO^@F6%SiO~jXxi-{CF=0!LGjUerE68c+HycKY!o3(-PsY?zL;X2aLbz$%rAJsJH1L
zdKNLfXmG`iWDO!Ijlk*I>XtW#6Y(t7+Rd^c-V9daU<3$Q03(Ht#1iK-Z1&8Lz|;ek
zIv)(TI&q+GM`InX#{d7qp2eg^Q^kD<{@C|%57h}ON#q&&UTJbr)*}aPWOtsZ-gy(=
zSw9(b7}L(hAjX6l$#CaL0gIM@{F1QKBNFL&dSnn<-rw@sIQld{hs$S}=ggbX_2OUJ
z=FjLijMMR2Eq(g(%agy++0WuP+FuO(Cd`ke%Y<>n3|>F_FT4f;=8~IyU`XGRMBap%
zham6Ft_km09f{XZ<{L)U&*$G;d}Mlf;~IZBx4dxy)?@J#X^YUT-qRe6>j&=1sZrO&
zJ#oe+u~D=-SMr8CR`cx}*Z9J@fyM<>;NGmpOL>*HkmG;-J)=_+_f(XPYki@Cdve%!
z?o^`h4cL1rH=H1VWDO<b9Y6z|vA9W<BN6Y&(jr*1#$Zfs=$K9$-^iq*cO)(vN7Y|A
zm~K-B5s}&BKu7Mq1RWc&*HV5s(Q(P8womPQ5?3lG(LThL>dCUZ@)PYFOcURTWTy*~
zZ5uiS;lbm<;dpTGCC20bX4_K32W^9MwQYlkGCX)(DQPCN`T6`3kkHjcvIOIT1_e&y
zzzP!SiFgK+LWu(_MMUf}3@qb<EG;uAO8uqH7c?CYRKD6Y_xz&5<CTHq`Gxdf=Kr$k
z{PUZBSy}aJQQ_PR3OuS0{-GvFjxaorTryYnL1b-CI2_Mp5dfKUhC=M!6o^`08i<-C
zSa5d{1op72z#ga~+$y+^ni;4|AgiTofs(;!Yf~Gm0?}xos*!GHZ<I3hwG}A*1LMcj
z0W3B$j?{8f7)F>Ui5QIu3LHLl*LzuuLSMtiC;?YGj(#O7%ji8K-J6C?inye;+@itv
zf}136H*hbC_e#Q%&8K^mr55_NqEE*jQ`;8of1f`l6q2pzh2Dg8>}I8$(uOc%VM;fI
zr0)f#bzylqT(pO92j8o)x^o-LJY5frwjG{%J!YO(Y-P2s-SGI0QF-lhIId<WO^Bo(
z30&$AC$}jo;ypSK!&G;9te&Kq5!Bu)2xsLt3^D#sx8;Xwl=MG?_Cf!bPs`WlUweo0
zNfUivtK`CC4{DXR@zl8YI!9;h`TzFPCl_2|cQ0M={>M{RPf-poxWt)z=>lbc^Xlf`
zf0V$X(`fez9E^jD+(}eG1%oLZ2CD+x^-vNkRtv_aJ}VSrK?f}`AS-=|30j7}$r}PO
zEkiET;&S!`Te{Rj8J9(JVtJB^&LSCtacA-a7QI7=C=tJ4R?BHHdLd#YFT<p6jZll$
ztXD+jh;Sm)hW|BlPbzO-M!zjQ^s4#&2lFqtn!i~4mf59waM7Z^OS${tPwDIAHurJd
z*1m48t8X3S6EtX_<e!AumqJ9Ksf`O0229e9QT5+w(IetdPR$`qOA@}D?NjZAJ8Vkq
zRb3K}mEt}<95S9*YAHGzl<fw#0f60-wtsa`u3*v!u)vF<sif~@oq+$B{ifcY?2>GA
zKrXPW=qlNWN2e1IJw^7H0AW0A3#nXB0{3Zzt*+^iq~NeyUvvJd^0Mmcvhpg|hGDnC
zrcajE)cVqQ80HuamNOdU;!Z`30@>l@hTR$JhTS>Ao{}8F98h;W$UD8W*@;wP{@^YS
zvzxb+BTDWwbacedHErO>&K%UIwWp5r;Zw(xn^Cgy6h4Q(1V{vQ(k}=DiIaU}!eItH
z!4O3Ri{wBMxO^DEpd%j!uvXI>t3u`(jW6{BLtl?WkM){mKnCIsg)l}<vBs=1C{-{D
zl&cgdx3JwctRdM4Zxn?PxERrj8*u!`R082bF#cn)-nATK*WlO6l|CSo%TbJ+5E=vT
z25XUEVi^!D5QQt8H95m?e;yf1Isvy37=cfCnw=u$=RjD;y!lyKbDPd>p(R@7Kf4sm
zdd*!`{_?7-a(|U`JrcQ#5jl!P-tvQVCgN0z^I`fpmbnettD4dOEk^$!dX_5nE7PbS
z7t1Pe?^5CoGv<;!A;w?Tz+cr4b7IT4QD;G2j1&ZuHN?Zwbbyj3iqIArIV>D0h*H(r
zk<^nU=`b8Q0vR7aBlXrMi8sKq{~t<c3H<;8000030767l?>%1+JoNw{2mk;8007bZ
z+JpcA007s~j12#K{p<%62Jip?00;mA00000004N}V_;-pVE^>@2#^f?U;V$5U4#KB
zf&yLw0HC}F3jlc9ZIe%E6G0fo-_AF?DIy{gx)J}>dT0xwl(2XZ2{{-l+Ne<^gdiov
z;30Srq(u=cLBvB2L4pArM8qu>3F1MdhtPxIq0&>R2N4tx9>fZs3uXHze;||we!T3=
zH#6V7x2rah1b^x+V6~}X`07@$rOO!CK-xYJPj4e`T@1F=Ix1>C_^CeP2<P(F0$i2B
zG&N@}2X&53ldt4DnIosjJu*wK_O(;bwY}#Dq^7>0<SZeppJPPdgs<NtZ#&cj@MQzO
z&GYOt@Y&-Vxr}3ahU0Dc&M+R^9_O3L>#?4)PX6Ko_nFfFP}jE+4o%>iUSQub9M)mP
zB!V^e+@i)sqa@Yfk3{iQ9wDiRQIiHT`Y1B;2^m#@E2}uD_n~Gr@sR6&$zA?mLe23|
zGvhkJ`r}ltQQx>ef~0!IvxZQSA{=`LBQk~|DF&-@5>d5`aRsbVM|pnAigdlF7mFk+
z^XwnNrm0~MF`2|2djpr$3h(rcnE+Yiy&2<NooB{PCh9d#lcYqg{d}%yg<0dbOj_W0
zSqi*g11Wn2<sM0TpYHg+2TldbWMOg`#A*jslLfU!KHE`@_H)<``^Zs?oh$}1S)Aa0
zMfywM_rm^7ZFF-O#A*ijnZ+WtNj_+g{`Yg(4N#!wOcsNfEOdlfQA6iYbiT8eJ<-n+
zNikzM|AD@$G{2D+ndpKa)Em?S=a@wwsUi`&f&_i{%zJK-L-2^ptX+A*44&{C;9B((
z36eGYvd^i|S>E>@&)jAmaY}IE?f|p##!GnGV_;y=fx-%gMU2y!<}e#DZ(+W|!p367
zQo^!^<r>QuRt44w)@7_u*v#0Lu{~iAVc*04h$DgH1Sbn;5N8Hw7w0@K8LlVXV%$~S
z$9On+TzIN@UhpdL+VSS`F5*4GC&uT;w}kHnzYqTs0WkqL!6w0FLK;F*LR~^TgdPb?
z2)hX933mvu644P=6U`D!5j!I;CjLRfNTNtOPkNd3AsIE9ESYVxo8;EWJ(8aw|4czk
z!B1h4!X-sMMH|I0N_<LP$_hZ3qN1g;NmWO+MRkSR4RtB?4H`ij6&iCic4=m5c4%JE
z64FZ1I;Ztc+e~|%_6;2l9T%M*oh`Z!x{vf`=u7Aa=y&O#GKe#{U}$34VZ>zQVbo)E
z%h=Dj%y^#hITJaPCQ}E~d8Yr&Qp^?1*O;HP2(p}Jb;m}{w#jyn9gE!xdjtC&4jzss
zjxU@RI7c}ja4~UN=Bnhn&-Iy`fSZHcBzFh*KOO-d%RCu8cX>H@wRkgmhj_2>Vex75
zIp%xB&&F?-zm@+y{}TZ&IABmfLO@<XN5G<hGXdWM6#@$a_XY6;xdqJw!XrU%g5Cso
zgfN6Cge(f>2t5&&1OT*elzIRG0002$1-k$R0000000IC300ICO000310fGPk004N}
z&0NcFTgMq5+3qDOYNSOGG(iiTRVo2A<69EPn?e$)SdAqMqGTb<Bl3toHAS-LkXDp+
zpP+A1pva=DqEC>vPtaHBs>?1r^Zoxnw=<+JG0?&gc`h^a->);YQtG{xyXv-Dx$}-v
zD?e9?uUAy2@96rLy1Vj(uWzf_$~Ir$QTJE=%GcjhAFTX~ufMPE-@3!sKTu!Ydd$~9
zRGnLY;_G+R2e<ys*Q+aE-(KbGAF0Q8e$UrG{`FUP{=(OH)#J{6zJ6C_odI9}MBVMk
z_@C<Mf8gtH)q9;k^Y!1<{m!?1{dYb7KYaZUwc7nYUH|j8`k?!++ENoWRTrvKr5dVN
zYNj&vNB#Fu-Pivft9!V5pgvYD{d8ZCDAh>)O4a&qThATq8MSKbyB&Sq$EXkW-*5E0
zT92NqT;Fx{XQc+J&|@;Sp;s8`ZwIr+U7^1V{ryIN4zT9Fo`0gpmRQR?&Gh|+{`?#(
z=6ZgiPAwjpI#NBoex@D)>wD^n`c!?UzdzG-n17i%$G^6onOdQ(X_%-pO`E|-@aySu
zWBvBIe!j2YOw@sXJJi^Xu!>KuPwR&P#~i;}%&oS13O+7}m8nhrIZ>}MS|oG%tPsbS
z;BBpEOdy3v8v1>3>9UpXMJt8lGe{?coM+&uq5TXaU+ZTR&E3VC)3Tb=zEpFT>C}f;
zgt2&ylqIa`IqOMl5Iwvc=G(M~LqMwop7-^dwVq?NHnW;+^t{yHuk_Ub8lFRb=8dtZ
zG5&^_Q9`x^kKXoyS<MzVgEnpJ70mM!*2N5ZG5{`S)f%>H^lPC1a!^ySZ`jc9hKy&X
z`F9`u5ndXKXPWm<)kpf@)$Eh`p1mLO{iDD$Y+G)08P_A`X{Pntr_<m#1P_ev7Cg(q
zt&+5_Ml?MRC|bq)z8*Q&<Igou4JU`*+J-AeLv}SKFSLCf>hXK}Gt-|h^!JX|$1d!B
zUp>=z&o!kzVDbXvdZ2Ve&)m@8UGI68{X)}5^Y`?<UOoFIV43;HW-Ywm)ANi*y#O!V
z_%6mC2GBCBv#qa=0kdmo8mzyBhccemtlHJzJ#c&vbR6Qnp)tezj{bb9|2lfchK3yD
z+f`lQYal=8xqE>33~;w_-@(c|-n$Lp8Lb2ixA0`2X_9pg_2+@+e~+o?B-ifQr$$0V
zUxz87wTQ1V2aQw?fWa<vf+X%{>@Z}v{sC}31UI|jbr*9T&ME2$^VA5#^L7A}VC*4A
zZh8Fc=<hB@koE~qJ(l1RxJB6ZAhj;!vh6{>ww^B2uIy*5g+8BR9+L(h**n5xna=j~
z=uPd5yUb@=;eV+S##3xU%lGvRBNegC4XkHm*a2Qd*@1_fuh}x|tx7@1zb<6G1<f(z
z3{A`Ctiulyo<x_?tz+2o3q7Xed7fu1lNhSwJ@N31c#ifMUqW9g%6hzo=h%e49<pw)
zF*ZGT+Z}NE!1@=GU&u=Xe+y4{LHjd2F;d+{Hb7%vz;~1X^YKebzqQaaeF2L*2A4aK
zO)v0gWCP}|hm5WZ9%it?=M5XYx(2&>VqdX#wqRR(tbIM^OULt}yBOoNQC8?#%i081
zoBH~QN0Cm*{I-U^o#ai^Bf0fMyX{?z^0Xv&;YAK%KL;$?LykUM;5k7NUlVdYT(4tS
zF%fQy?R*(bqnEpwxdW+p8FGiM@6cKW`3cGw#CM0dEVJev_UNQLgek2N$KUs@mouC=
zV(zZ#r|X&;<Fgj~IpbsZS!cSSq37xA_5D+qrAfR$xRPh>F_zbdW~}QaDC;JubGm0F
zyu+M+0S>uXzRy|9VjW50d>!kRw<Z?9v>BU=5MKhP1NInO97TFI8+hBkBC|Rc=Q5n*
zSqPgp2PBV4QF9M^e}pxBpCY!u3tYRFo*ihh!?DepymjSFAeM)C*iD;r?E{}Jeq%Jb
zwKQt&SZ}q5Cq$2-Y!lYjx+Gg8jJwITw%LEmTBP0gfQPBCXS&A7QN3iSFsOphQ2!OE
z5lziB)z!+3tDh+<`ync&Gag~;k{oqm!FAo3*Pv>1iYn+e*H$L7p5t8~Ri>%<FL+j8
zS3Gmnof)dZA!@`Nb*>pxV9W#~2H?yY)_%p%srogYgdF{!0%K}DB1aE`o|)c)p=e}h
zx$4)<oS{C>fqMv1Xsb0%g??UQU8(oZFqbl!8Mst1C##^Y&be-vb!(ncfD%(-|4!p4
z&+CSyW~e4d{N9I6PiRfUy)i!_DVa)iYDb+hb_Rz!sLBW6Va+@+@QKnH^T1JZbv=>Q
zh4ukfo!h$BtS%iUL*>YWCp;`bd0Si9h%pq2pE6YHIF!J}aAgL#GnSjrDIeC5yDHMz
z*vD#Tpi*$4`rqeMFVX?mGw?eEvZjG!q*~%isIIL>^{|%UWu8;#^b>r3iq9t=^v{ef
zXKU(#*WWQT+|$_Sp!|aSpcS+#_i)*OX8PR((`mbaRb*Dna>j?|<X*4N!J2xorb$@2
zh2^>;Il5qT9#ufcGJa=1hoG`lUwe{jc^FpEW9f6z+QJvIlFG_`27G3ou3i_i!`t^-
z%enQ6`PFA>5|(~f%z4LV|EP^fOzWS5tImt$cC=61|C4kKgg3x%3`)c9BvC(S+ZCBO
z+iAw7bEeYxp#m#cfOi6K<4eNl%mgDy-sb(-+XE1&sm>i)W69Xk(~^D9h2?Y^Sj;(l
zC%1CJ@aEh#BwKw5$$su>C-t~H;PPu=J#Fk>h^)fCqgd|P(=zhhRiL{BVTu^*sHCTY
z?HFUpj}oU`q#Q-v&QtKK)ri+&+%Gr+&|HFFCL|EjRCrqGXAc!()EpMpS}>bt4J~i$
zz|zvNa!dO?eg73xWBB-nX_U7!Ubp5*<M`uu{hPYIhyve}IO)}lH%X@egg#(TU~&B*
zQ%UCqlsB3DWW;%uct*;3rffC)*jF&SHhs`w@`7(TQ>~DF9O~I7!hK_{{{`;H!J8Rh
zpK_;L`u1bM7G6v|4@O>VoY4Do?1+pS(K16uH-J@=UQ@&!HR5cz*RhOOl68`gD6xiQ
zng#P~3<{=}v!|>XCWHML9i1&?`?aUHSFz=`jk!&oB+TcYk6xrlejauP=taQY#n`r%
zKn^+Y;?+b`Az40I&Mxl@mD)bs?#NAiS!nen?<rGDUW9T%V~-|dJYiU5CqmUn77NnB
zHRM#qTg9}7F8P@nY0wC}HS4xBYrOq2Xc;GYJ>gxv3M-~|o@7Vu1aB(ns=r@tI90)>
zhtPmU`&#}UX3JW#)*BuvSrqAgWh)5sMinragGZH@2Q11mMv2VV<rvE~&-ql+xG-aR
zyK-F9pX-3>XNEOu6~dszm8qAFL?1$*a+Ywcy<HtUW|nI5<`kc=N-O!dMSkw>>Thgo
zj%w#GLr+r&8L4>AF}!*U8Q%mskju>mcrxaAO!YivOI0GLklHkrL!Kq($C@Etzlbun
zCL|YO$nOvJnuPT@I3<$K8J1;?x#r#}dxluAM4RHgOKM!&2`(C1MBN(Fy%W~Gp5586
zY*k8L=kN--bSS&!(XX=c+eO2iv*R(>`J$mTS}1VO*YwynFl<Hh3g%Ovvj&2*&i>U=
zX!6mKtwN-Ga14mA8Ee04b~RC-vGrllX#>i;LO8-9@n#~cyrBiu4H&~p+syAO$a{@;
zPLW?ZnGZm*^cBWIbT6#MO&Ob1oLfwDsy@vV+E2b`4E=Q-G{N8$Jtc}{f}N3Qj&91x
zTcJi)G{ySFk5e00Mw~;>qB<nxO3fN&x_VD-Hsty|P96_%f@T@@ls}Kz_Tn15h$P(#
z@+PWKQ`r-p)maR+@HDh;NOPGDN-u7XJ_UIZ!w0{ny;@xy`m)$oAA*7bYSmNbE3LJd
zDzcn6d<tiRH8V)9jQUmL(^?C3?aZEVIbka}iMqsL4Mct0neDMHvpT1550^?ws%|-t
z)xL5pjiV+-^Hs05i0RF(?bHpA+N{T&e8UrQJ==kQj=d8(k>_WAz9%N{SaPqDxRml8
zaeub<tbpXX{FGLG={c<?4`U8BcV->WQW>XkOfn>@C?6r~bbh*Ijq$q?VvTRO4<os&
zSl0lZr23i|ETVsL?Kk)G8)M;qW$ahPbL$i1y7xLgW~n`zk}M-yeIH5Qh<w@XMw*#J
z`)Ge=4N|8$E!F{d+LP>|<kP-_pMTvPbp~up=yZJF99<(bE3w{)<6HbJjfsq2HfqHC
zSm9kba^2=R6YdP2MD`o!-?6;pq+(!!T&<A~k_Ml#H?1w4Dc4%H)dlCTy4u@4hn>cr
zc~KwF?bM|0qHcII|Fk8YWE^pJ=yaX(oHtlAwN8v}#b<&`n|WsFHc75+AU)!J+;ha1
zitC@LJvq0y)9(8M$%&;OH-lW~+<C|ua@-}P`e~HN?kC(EENwO%_ivW>C;h4>)w@ic
z60L)sHla4mi?4SAR=m>8(>Boz$>&G5Q#lVPow8TCh;{*5oIlK%Z>~cbOVDNE;x$i$
z?)>#Sq+xgPsu;MsY~Xd3`=GD1xopY~Qhmq8okpQCj@5oWZTxp`?KF&xVSjgk43)IT
z^?iIv`mlWq6%U(Vp<D0volaOI=WKVY*S5Q%w=a@{G;ZjT>lEcIp_#Ye9iGV-t#kbd
z>f;edns3M46^Y--?-Oc!j=|sKr`=@omxb-1PHZKcdcSU?RAnWWd2m{nXjDw2zxyBa
zOLq90IY_+9DNCczQY^grqef4uw0r4tGNQT5bW>}++@f4>d=^xJC)S?g-)E$<MahRJ
zY(Xxs4)-}}RjEY^J_|7Os+ZI+LZ0jQ=Z!8%9(DR=whp>j5Gz}OhnFh6AMDuWUKQf_
z|65d?W~~8c#5Kv>_L6QEAzYOm#t0eV>mltj7G54y2e@;+rK`8YB|C-j&iG9uqU#gP
zu=hogtnxf(`domlTXSc!Z|nO(A|F_$!}9$Ge47XT;YA+H!KaMnl=pwB|C!qeojUar
z{I=@Ka&Dt`Ywqa__RfR*@hP?Yk=hN+6aCioX@TqSj7|bekUm_^(Ou4x)Lg~X+?_n-
zGeoZvXWG)IHdcApr}Fwk?qmw>2e~()r<QSHmM9YIbF?TUo`<`=#P>C?%b9z>_nIJo
zn8L@6IquAD4<ydPm+Nb2NfY*$bzo0!1s6?yt#uUL#OLI}NyXGJ2z~pA{jtmROa<{6
z(i6W;zkrmQryq2b!ipl!=u6v?4B-{*e|&rOGiraYlQOi*o{3D^udKApJKWsHk07r(
zN5cw!uLw9PF*DYgG(S={MpT`^XOY)xFWE1fCQjr@9_~7aVHSFwyu1}e+ktoP(5(E_
z?*7SI7oG*yygTah=@GvVV@t_6o_k&H%QzSRrqYl8v?t_KH}}q!rZS$CCY6ouS$xK`
z;1jkXcO8_p+`{t}ui&!>poq>}%86dd8RZO&#J$rfS0(%_*1V!|mg$;$uE*WHe)V$)
zvG_dbzgOPL;kkwGa?*%jjP)rz!iaMZSKpnmPR8|^AID-#OvNXYPkqc8-r^IkSMzbr
zU;8@EU*6a7bvV5r8K1VQtnSxdr_P-Lru)h}Whw8++=r|ougbHb+D2Y^*6efn(yZ<7
zdg?dMqNVO!L>8Qj$n6=`i(o%Et~tW9bF!;fSvidxXG*_|pEo_a#g;2FIfL}gF4BcP
zasQh0unM|SF;*lgKVG)j4qkCBS$Ng-7_opncZ<(H<T!`c;`+hwe-ySaQrdI<B(a=#
zkI%dl{{xPN7j`|Dk=6E`z>xbO0;SJ*4sHL=$HUJ*X@6zlRgWi}nYEVYKdpt?NVOn`
zN&9U%o4BZ&UzATs{VqOOq+Ie2`DMT7xIBuEt-Tn&zP1)3ztyj!U+Yi#|3Rp~^Z%uB
z+2~{R$Q||%@c&pG+X}uwRC}mC;oaY-`tynYe1s?T3;ie2|A{bRI~@l*wX%bQ-!c3b
zEq?Z_004N}ZO}(dlW`cw@$Unrg|hbs_a5*2wq+=e7N~pgEef<%1T0p?ZNxpIG0~I8
zby4G<(WnPC?iEMWgJ*X<s7GVr{d@DdeDeGb&-3IVg0}ioMmPRfpFtudl1MB>VZ}x?
zG1!SEj(8GCB#C5FNF|MQGRP#0Y;wpYk9-O!q#f<)Ku0>!nJ#n<@99ntdeV#D^r0^r
zItC7$xG18S5=!Yuf65rZKn5`wHy+BVpb{@b7|Jk)GlG$fVl-nI%Q(g}fr(6FGE<n!
zG^R6ynapA~bC^q&SU5@x8@a(gKC+2jZ07*SIVMqT<2@VLD^{_wll^SrHXkIKgPh<O
zKl#lOPVs<yoMs;L*-bT%sNp^jdB$U&@RU!~@{;Gg;0!)~u!lFi;x&H0@Qtl3AizQv
zQAa(8Xkamo1Zg6~5|*+o{N?4WpqZ7dW))XC%v#p4j`e)yJJ&eNIj-}TcM>CZiIq5s
zmjp?aBuSPONtHB7mkiExfy-Rs4i~w^T{d%4GP%hu$&ze#NRH%6p5#k`)mPWt=+{jQ
zTc|#uX>LtfVd_!2$Du53)zCCesVfboL+Mnylts#7Wr_0NJh!Q?=IXk=HsA{dYpSc7
z{AOf$O<i6~WhmIt*0{Xot$ID?9bpZnL+ShrtW=!)004N}Ma?}5!axv);oWSq`7`Dh
z6hVlh)gD2xG66{=R$1@>p1@WcORYV|#7fe5v^dIU?dEymV_vko0=<z6oL${edY4Vf
zMLmN^R*>!SxX3ARai|HfyZ|OPVCKb>jSTUFg)n<?g!NXi5qz!SAncWbi*S|-9>QHH
z_y}*VFhKYU6aw`<#2U1QXjpkd4w2zjGs=sW1m}+E5`RlXC;yV*Bz|b7$mFP*A=7bB
V37Pl+`?OL900001TdUiQ003jn30(jH

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-book.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-book.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..2688739f1f0bbaeb667259838a5ef06ecf9a5c1e
GIT binary patch
literal 25120
zcmV(@K-Rx^Pew8T0RR910Ae5j6951J0XhT#0AaBJ0{}t*00000000000000000000
z0000#Mn+Uk92!&`=rA0FOa@>8gHi}Q34~=42nvGPK!WsT0X7081BWIHgH`|pAO(Sp
zI0uDX41y#-cV%K`23v_laRN8TF*IB14W8C4*1D1?P;NWnc1&O7H}@$}Zo2@^gQA0v
zs!sO*|0gAtF*FS`X%SS$_wQ6TNmQYS3Wm_bnMjO@D@#PTed|dqu?wsD9Tt%flYBR5
zVh8=Se289f^Lw$eHgfhbc#mrYcg)iv!zhPpF{}DT?J!4ej8M)$k`qb#>)+__>l-TC
zp`sNY(;ftcLLrIq<%86DK^1I~QnN)*u}V2u7$^s^53{TmL;UPrY0Ti}-P|h-T*2Ho
zHB<PfJLy9gtcq%w5LVEEI{#jMkUZQHW7iB3W=jse56=^A+WWJEHYHBt-KtWPTtc0i
zsT~B>u+<ZG4@EAIG9hb(O+gs6L`jrE!T;W?t1rnQ+p;B9scm=jP7guXZt0U2`G_e5
zjyQ-I`~8hC^-+)jMfWDW0p4BzTDWaw%y2}?%9uGc#V(SRQC3zIqLfiG))~h>Yuzw!
zEpD#w-@m(mV-A1K-l~67QRZLtsBaT43Jzd0dw#ZGvLxQZ>}~na@7V&vY?0!De6vZF
zqj#w=Xc3ho_^7?^{n9S|W@+ju>U0!zC0a2C6@-b9%vT=ez}U<l<urE<@nuG`&xhXv
ze*tkp@+R>Ckh~wOWhwa;;dOw}0InoMA>nTu9Dt$zkgaM}T_&wEt>ZhZdZ^l)M?eHo
z%B4gx9t%dd9eflO4{y7-A7Z<t5SoxoNJ1l<(BRHy6iQ_2GWGN!KYsYJ@pGdzY#4<#
zr=Y_|-}u1OqLfW(3oT7b+my!h<n0Lz2cNt@OuB%hZb1$3FWa&J{{8<pd*AM`Fu*ie
z=J$5f8t%JPn=W)zj*qmDRA(a06_S0`d_Dg*%$WSo@9E@LrhY%^cAp^<l8_?_1|UyX
z#;wnR|Gwp$HA;E<XN3@EHuXA*8L$u`Fk)u!R@aw~7a7`s81RpTHxLachIO$uO5X#j
z35WpTlWT9b?{8D!H&sNLsa+%M>SQ(_^HU&HKx7s20m@T5XHTe`S~G2tWL=Hw__A>|
z`j>Nefd`;6FCQc+5XuUL+ElAeN|!1`A$w)2-={shAUVr-Rgcs;pBWp3ls{{O#0C+j
zXLo@$0J#R#+g(x;0O}qLNf7|4djOvSKuQNvrtmy<mS?O1NEwQ{Cx6EJ(D-kCko^2A
z|ITLL2jQ<Dgg(4H_<pslPrPTbk)sDim@|;6?(Mr+rBa1tCMr4>Pjs~0%8yKDR&{T6
z&-|5f4jE5Z7CD0j-~lT1DOw51fBx|2ALx^Q=Q#oY#x<*V?zy-1vM8gnMRQXaCJ;pu
zNJNQPSRknf|6Pgm{;;_u^c%4%Dpswi=%JTe@5KM_@1(o`b7>t_byP)+7%^gu5u+ly
z`+lD&LfMkB4Swvyc!7YWj)TM4bUIn2YZr}y(1L=>V-Vf?H;k#^4M05bD74a*?sQ}}
z%f9_T@BY94f7AFmO~<utOOho?k|f!_00MCB{nP`O{*<8sH!h7G0t5g)pl~AC5bSm<
z9F!uIc5BKgOYpfh<&-B>_!><`&k&yT3WWw#p`UP)s3K{5*R?%%GO$wQe~>sFo;!=h
z(>gC3y#0B(q~QNY0@CKGb51RXpFpDkg8*aoOe97s8tCDtFZv%;oDvl7h)t8~P`W5X
z6pVx@A627vX_(zJ`p48boL%MPWXH|Mm-$#7n+3CKcK7o&-g7x8r{?xNnpgAD9z0E!
z&+qD&Rv2nVP*g29-rPysV5;77R#mX&pHVoShfs-7@U|fAIxAAjt@#dS)ruvOo<}&F
zjj}TFua!YiABA)6tes?X{)t%-$!YXIC`4<OAr)Oai)X7am34-7{*=TtWAj<=jy$K!
zl13fMt`iK<Bx0tud{z*BCZ_JHXY9zq8R9yGpKW4Lzd)_C&{?KcoiH&ulTAlKo3lcI
z9AYo*H1KsJSvaRU)%85;oUktQKCPrDJujIWysK?QT)P#nNe0{Lt&u9V6vb8~k?gvf
za@Q#&LPJtuc_C)XnD=ZEis-22s$Z)&i;_(2A4GCz0>p}k^}P5oM?iu^L4^_`97Yrd
zCrljfDs=MI9i-X*5Z#bD(H88LYJV`(5qFkjmSy{gd2ZQ!2Y9a#D1HYDz`;bEWe=jT
zID8FGWZA?DoIN;?nOEiA_$1$*3UHO+6yTKLs=%qhZ6BNE`iKz^a9&Wn!TCV(4d6gT
zkPWWqZ=;0534;<2B_7UpC_AC-g0dURER-W~j=@=m@&w9LD9@lg*Llh+R0x$&K#3}-
z-W9_Nf;AkZcs6is1yF7~@z8KEZu{T>#}Q6&ei$pB3mjMR+~BweJm3j0c*6(2df8PI
zLL>D&90h?6AqD~i4l@u3mcQ{(*x)K4R6<ZdP(rAJpn_2Ss~YiY;ncyYhtmM35l&N-
zHN)A87PO+xwDa&b@jAro6t7#no<J}9Y`?rei!%pd0m34LB?x;U?1iu&!jZtS|9xD%
z6A+f+B;TyTLzzYAnI#|>sq5&c`YltP^a1Rn2B$Fw17HdYgMVjYCOUKD(VLBOQH^}a
zbvA{)Rfhu_qGa#P0mCFk>zHX>S#ztoAqGX5g3)xFTdZY(F(ZKoWQC&~0tpJ2fGbL-
z#+(<l88XOAY+;AR5ySG38p$?h!YFY_3zA4yibFc5AkSv%T0v&&@s<e8?AvAlAq3Qn
z*s8@)@%*Suv3Tae0zb5#Ae)tXX~)6~M&i3=UY*cjC@;c6(VqzmfDDVrOok@Ay>qf5
zRn=0469%0wmH{}>VMnLuStbIQI#cUVcS4CHl)#GU@1Wfl&P)|yp=#rf1>F;)7^H+3
z+zEKmaFX@83uFwal1O(kGRTLc^m?j}@<ebUkfIx`3r+xIcZ&qUs9&d%rsCw8?rK+u
z0Vs=pj?Qf^Jw`4l9ArU&7pvXC&4iX<lN#tL^X`o%f)waRr)c{Wrw7B0*+_hCecZ9o
zG@mmQi{f_5ieCdSka0Uo;qL^{<%4@*AaCP9fo*+09pf6~`k!R5!Hvv8_Nt==Yp~z}
z7awezK9`Pm0rTm?0y*i23%>RF#WBtp4d>y)hIXhUfHn5msfkJ@>PoY8B^jy2UImzX
zYRt+5w*jYXDg+B5GQp%LkFP<{$%9`#;aEw!1&ga2AK^0|^#&lS2&?A0erD{cx?sY!
z&tZ93wUjS-3k6{2)9HtuAKxtSb8-wZg>r&&Dbm{_tNZ1@&lKyF4v~*6b2XT*<rZ?$
z@~37SA!0!@*6>^>QVXHeolbRWvx;JKl2*mGv4ga9&xJnb_z;fd(Wf-2KDVrGgG)?B
z-FC-vl<=|($%~?x3bUz|^fp=p69wFqD&<L0els2IAs1GVyZyr3za>asR817KbN|63
zoHli{b6-#PE^hmKcHwfekfTf=*$|O%v2n4+&*w%zKL;+|^zo7z(^eTaEik?Y>=(f0
z{<Na5QFquR%{L*&&h)0)fDP@%0MN9u5tF|+kub}Ln6;S<wPmm^$lbY3AU1s0%*@%C
zuL7!vhyG}pqIv0*nuiDMFp4fclQu(pm=Gd3*F4AZ*Cxcc_~6NYDZ>1=V1}PV$WhaH
zUjlo0l_NGrjA)9(;ECqp@(P6gDRbd+2&}C4#6GMnsPY`a%Ex$CSP1U|h*M~gnb-(U
z<U=+~9{kT+0CQO^R?BR4<mNwd3Z>0#wvtQUpHEa$${l<GSLNVFD@$`?a~|9WA%>ol
z>SmX&Kk$)bYRBOF^Op(uwWCmF+>v<E3b<5LX`9F+_$+y~k;l9ivrc5K6@befXZjyq
z`?iCmV9pBNkKhARu=MkTdjFD*y2TUuC({y%C4%kA?Z#J6Hbu2G(!c@K$lp)o@2~wO
zCsUAZ&a#>MYH5{TXITTyUev8yR@gw|#Id7KO=(UWj3VS`&PMV?KKIlFf=yW@a_Dwz
zZcWO5zKDwrgP?Uqyy=8ME|N*so5UxUSr+tEg@MM>n+7)Qze0NDWCV~L7l8x`79upC
zSU7jFLr02A5{v-EFk$-E5&xTQ3g!S(va}gM8rG0(*upiM)$D2Dk{2}?27fhZK*hA-
z(dz+#NY9uNGN|)l3iq`bYPvon{c{+~m}FFksAI7`2acT9)c_fR1=(ck(<F!%^_X%q
zvs^<~uqFYLP<9qZWYn6Xi^w@ifhB}1HMrA=2iNlC8eUw_o9p<{fUn(<BI&M;2yhb;
zjge`JLK8s@M5UP!l7x~hg4QC56s0gUXoF63(Zq|P1qQXmQWKNf;%JFQxOnPdQ&+AV
zq=Jf-8bY<sr)bj&g8KOoJ^HO#JvK~D*kn=<8Jn_$iJfMV%moqVEmHRmVm`tFhiUd6
zI!e@Wybt1psTK1#xZx&H^1IZnV!_PsQuiMOyUX6tlm58`0@JgJ7JxWrAgrVmDaBcj
zfFxd)%3N5U$!1Y#tVCl#Vq!y8N*vCtfuk0ke019`(2yM{{0%yDG1`dZOjy&G#k=9Q
zL?%Bh(}4gu#Tom(LZYY8C=IV3BMz`6gHJxXt?f5sz^!&Y*P~q!b?f5G_i5|$uh|?A
zC4h2^KmcMwmj^+^xhU*}h?YtwxfJGAF?z`oGzy_}Mr`bMsM)N5Vx_aeczl{{@G*Gg
z`n-yG3E2pCG*0rROs_B~t%8@^%04epRPY>yk01(!So}_f6&F^<$jP^-2*99^uDTHn
zubV~Wk_JLFA*A{<u^}kDDpLrdVv$~#yOGqaNCVLRG#1N5o*@+S9)R(DVL`$~C#M<+
zk>r911cN-*CtOzHMA24Xw#PS&-7%2`ZSnLuFnhZ_bxy|2c@9{t1}OAGY`7J=tof5v
z@&-s?{<a$bgYC2OJqlwAKu%{KFlnv^;^zU*@>>dGX*Plkn2*1m{``Z8o_RcPA34oK
z%A_0$Pyq$0f~L*Gl!5%8U;WpQz_c(bf4+0fT}r2{u5yr%lmFJ?)zQA)-+!+ko_Kip
z;Q{;YvC~E?43Y1_;oGk)j~VU$p2ge53hZMg3Q#`_3Jky$tFp3n(>fnws#B-i>ko#b
z&8_X7-M#&T!=vMq)3fu7%d6{~+q?URriJV3=I-I?<?Z8(^V{Md5QrxbNn}b8H8><R
zEIcAIibju)VZ<`ySn+I5LShn7rK+iG?d%^O8W|fMpPZhW*|BTq?%Dapg{8fF4jwpk
z80gGP$+#mP6l6Xs<VRR_hY--B0sw7m6Vsq38vyN^EBQOkIE-~G=YEWi%^u#*`zg-1
zDcT|9Mo*YHY5cV5wDt4mu(BiodIJDs7=Rx3iV%Y*LoOi+8*&20htEbOl{z!xaTF+5
ziYD7SY>Gwz21x3JuTM~B-2zja4(ds@eSRBzmYx8UO4EIg&h!E8`Ea;%K$?x!ybZ*w
z>4b+q(Jg0vV9wS;8@(;x>e{L_tIDr>ZI#g=o}k3KeHnI!DnmGCVk`?Nhx+h+%!TIu
zCngj?5oyMu@v$qRJ?F#(Di8V)OnW70SK(<9P=321;DQd&C$5=mL`i9x(%LF_7aG9Q
zOi5WXUd3x<X~p!wN>p<eK`?JZh+0j{>>3Fz%w@+ky1~l8lH}S}S7O$<)(vIpreF&R
zW5ydk_()j1Vd)KvrM&tjU28=zP6RB&SUUj>(`9IwF_#U!l-$ygbPg&FXvHxzvE8_V
z+*_MI3oVcF==<<gKmy$-9+UZkKzGNedCZyN0)ZDYQLaBDvX&hS#xa<EQl%_#_IA}U
zJ|!U%E)u2eS5R0J2ZXe%y)bA|`Gu+HOpy8_27oZBuaNNUnwv2tJtS@xGlE?=n)4f@
z7Gv>v;fyz4YOqiGVM5LKal~i!8bGc>7Q*ul#oih0G?DaCdkw{3JOfLW+ZO26mVz2s
zo{(oS;X9Fy3yDCy7VNCdWANbnmZouZqsBlZv7}>|&&{N_B-+d8Go9tokh^V=B5|64
zvce6rWMAW9Q9)w+?dsWWtaZ1z!VAHb#2?~DNSx=9b}`3F(P5!bl^IrAl^ZTy$uZ`G
zy4D=ILgeM$(6ZLp7;2=a6_2zP`}}`e87<&6KAkW~_x}?cadMMcWIqZzl=#0NNQ1jb
zYHu42H5s{xpx?DnZt>=KIxEz!(kaG|K<mK9{}BtE;+e4mYpFK<9H)OBsnV|o6bklK
zu28EuMjT<g0P)}HCK~A=>!GE_8=q^<AmfJJlo}(S@HE~6XEPR;ltTPZixBM5oEF&d
zhZUBM#0~G!C-Ak@ySKU4Vjg1=3mo&KNeh&>8Eo`~<U`=r?_^Mhg(x_W@P>)S(;;TI
zaILj3IxV%7=1<_b)3svG&_w<`2b^j3;H`c)vHH3D>8ria^ndA(U^qk!LNKgx7gSdA
zWw>jjC@PEE3Y|e4B;eS@azG6Rq@9IQMqG8r@ygUsFq|)uMV!UL`70W<A`Acn8S5&{
zs^36hJPZ@+$3pu_*Qb0C%WCw@77XiNSH4UsKaqtqfbT69A)12Ama~#cQ;DI9#i*+C
zw5$vyeS7ocX5wokH#R{Fql$&yPQsvAU<>)qtAAAcerilFrAs7VsJJtKDOoblWu*m-
z!xm{w+1yG&+Rd<ISo}IdL2E^AlK3OxP}Kwu+3#!f(<X-X)JHW4IzFmn*i3)i=&b3b
z9<(l!BS|dNHu#?41H%Glo(kl%#V8Br8AHG*l(i_?YonYNDvd43ft07qOFNrk$}{Bx
z&pRkFmFp+WFPdcYlUK5=l=RDfr)81v0<HX|-3mm>yX-H;Rf$#>)h@!Aa&<9Xw)P{1
zftSTZG9wq5!d>~}N|lb@ae6_-U@$Dge*N+TP*jsmWHXVnz$-<Tmvk%LLOr6XpelMF
z+j**YAECr))KBEuo`Iuf7@p}OZ?9`u`ORbk!!hc0v6pq{LM|P{N*M&BHKIO7sK!4B
z<Bg?6>uPje1~qz6$~(&LKy8~GI}E`RTXMq|#l$OrJoA+ibO8jDXj89H=RD|~&UJ=e
zQ>l+8ErH<-^`|XBT_iSD*tw`rTk@IA5imujEox_`Nz4?ol=#_vh9dpJ>Q1F|*upA)
zsKA}LZ9D7$pwBAOLT}p(m0frvw3@Tfhdo8qMj6#*8AuQ-PvNd?l(AG+PNey11lSja
zEoNH`{KXAh4Qp5?D-EI_v{4(@l`5+svxRk#3`HzMn?TdK^ekvqtngfa79TU+VJ_Eb
z=sj~~c;shR1Ilpi8eF{=S81uUN{1ij3}lh&JT3Z3fw;OrEYBtt^uV>XMGIbDD|Dd*
z4itGB49Sl-Ix_PuSN!9!RQ3Md7$W-$0s(_Ivi!xgF<i?`(h2kslZ)af<h-r@J1G(e
z%Ac`Na-_5dX5&pREj^&Oa+mgZRW!z`V_kn3?7=KpN1{qhZ2Y#z5|v)jv_oF~?Ke8S
zB*|u03*>nJGA>)Ik)>Rm@1Bz4)F27}$i#YAzof>mVi+)p+}Q>#2m@9`EEPVj$THf$
z8eS$Y1y{VBZ1G_KJ`s)oagrnh{u@`d-4wNjEC0Mm!U{;;<UGtt;^yaaYv2F<*~omX
z`QVxNGCdQ$%fD>yY;+)jadk0j&+{M1R2FFR*ZrT7^)c1@$bT>lvdn)T-{voS&g|(v
zanCmkx~Dw4%b(Amj*dV4mmYhe7BbIsM_ZNfQ8h6up=FiDOnol>FpDjWnw08~TD!fx
z4MRKHQY%-W`zL4`N{)j&kg}3SXX%$ix-(A;WLtr#e`b)DB_%P2s1t?}D%f2Raq=F&
zTf)VnyBEEYQl~HVX4F7KhqDvjJ51T6*MbPcs}f_*S~yr9=@49E;A&rri2c?m@FVLV
zM;Wx)w&okGRaqW3n=s~?@eBkAEFX;*#V!$XTNKIfDgt4cuO)X8?i#xiW$3~wj3z(X
zEP2aoR3r=Rev|YTms!%LmayHj)G9xUv@BkjsP>nm=;g0pugt$5MSVk=d5#$R4l4cb
zD{(BoS#Y8|?(RQuL{^6))5qo3pZUa9EwS&H{(`-XvfrG*P=1S2yJUT=YoZX}h{ky5
zXl?4_M9`gH_Lov00wzezpHg05R+Y000?zSX*=J=`QDwn4vYZ_(SuDC|OHrF-V`6o-
z4|^9EN-S0C10B)Q=Cbj+dW`2qS>zRk!o|uLlab`&)j#obLR|D7_9L)U_k9(j<oU9E
zf6zRn6o|Nrv%n{5h09#sYGiK?TutcG0y@zI*ZA-E1Onje4#8xxC2ndG)^Uj}$@4(u
z#;!~+ZGkMy?pP;Z{R|^l(iL(!S+?<jUBXtf?q^R-AqN7~Di7Ytg%?HeTPr`gI?(f|
zXQ+6B@z0qC=#w#S%g}_XAJ%GgVV4SmX(K#Q(Ch}0Q}Q=!TerwjFD87e1TIC@=_62g
z$J9^4i01%JSG*(v;WQ-=KRazXEA6bFC4p6hYwnQo^yb<Mx^RxNP6R0X#NTzgRTzx7
zD_JmUFbH~|@0|<D1VBjkNSR_*5!7pC-T751n^@`@Qf3u9)z%O}6x6=ut<4H;kSk}C
z{_QH3{8GvY(I8j0Du8TsFJ~*puytyV>f8MUgKrNe31L2bjkHaxBmX`Vh=tPY^8gpn
z=yw5y%bpOqe>DJ=bf_<8)PdRh9wcbYG?%3J(o*_A`o9v3ICN#$n88anfuujKO-Qpa
zu-riXWMo7Hxe*z1_8%kXUuDMh1BqwM{KLRoL3;7ZF*O)5!OljomCipL7B8NHpv1fj
z4o3IFah8kN+(;1l>zH+CGMVLgp$k?4N5?`5e}McYrp9dCXtt;_2vKmE#T&K;uqx!2
z*&7&GBlwEYs_!SoUb+*6kXQM3Pl;=})q$LB!5<=k=K=sg09jQ_hB(HtoO7*6Fyhfc
zo<?F8dW^Ur4HpztHD0NT8agt>DytoFS?B<?*0@pI=^hsin6&uSjY<~z|8}?jzi+S+
z5j9kRcCPtJv3229h~pjk+I|R3dwFYUC{FBx6975@=cdwG2WFL)1Is-{i@Q%=Iy+UC
zqfU`Si3;Bn_@vHhRh_c26@tm?T4L0)iw&XR8G$isK_@hdmX%ISp%Q}q0ClzwUuzZ?
z$oI{_5dUd#nV=2o<@WWW%hH$LSNc+aWH&NR{E@uzht3-BTban#2iDvB3t5?)_?@a(
zs64(?NmQYw9A4uM3e?qFosNEvtNI?tt-6gStu-wWD+AfpWo*YEH-ipndzNUfI86rA
zTm7qO(P4r`-mJVlGfzv;xxP1Yzo#Uk*Zu6{5z3iysCcpj02Xb&`w0Dq@4V3IFdS|-
zZIAA;LvKf8j(|otc7ugW6b`_3g#0Y=_Bs0McoLpREs!+MH3lBcF)F#3BE6Vcqse+(
z`#9I4@c_rT;$?Cf;3)Z)SG+7?Cr0BF{}xwZVMManT3!72V~!=ydkYDb5zzT~dCvO1
z(V-}#)8%@_RHu9%p+JAj<ykf|3Acr6n;9_h;o>DTvwKly&$sIV50_aIo)V2)zG7J2
zEZwyEf^OABZ5pREr_%0ks^RiqbN>poccr<p??ib;|1ztdpV>OcZTPwTmWA(ouqear
zv2Tjoif=a#>YK357cbwnuK}|vI2akTb%Mih-#aTys~sqGAZ^~RB+^zHo!DNBRH@Y)
zCszIA!Sb2&=gT@$nj6K-^N@r0mjc&|Q&%T$vv08{Vx%GtPh6BPwRh3Hc8eKlqu^5O
zTuUL@+J}kXtjw((<{reola=D$j_{hG%*Ai^iQ8YarLc8Ya9PK^<@GkIaw~?D5Bk0>
zmBm2vA=9fqi_Erjf=i^0W~2>*%dNA>PFe>v@1?bM?S{#rS+{TJc{<2!>vdvm?UXvH
z=xSXl($gvO_Nqm~I84CJ5@aRh|0*fDf?Qv{HS=<zO$))ucG}6C!6GHqE#*;7iHk)S
zi!7Pe^y%0hy0?FfTSf;$-#raMc8?&2B!0<0welw2+3_>~<f-hgO}q3P>PEFS;WGvX
z6FQ^~bk(8)u&nRHcV#hH(uo~ou`#7G93?o}oUBhn6*X5&A}1pcVR8bt9FNZ2apUZm
zyFqfm*U#Fqsycoej}sI5Y5*oe;Y6OM`E8NI%01P^nWKuP@I=kb+n%?!W^t<d!tg2L
zGlv|K*Kum%^ri9od!ebbkkgCRyTO-_F!niSxsD=Tc$G27VcDmt1yoK!_OnBQVo&{&
zR^~w6^tNjGSW8D!zJqpTnE)o%^xU2d4?q6$-0Z^J$0uVJ9hK}y>8@u4$A<=AmNuB1
z^MCnzIH1oull+%L<GXI0+qL(}>wPsRYZ4wxCH<qTTg&<<6!P{)(Sce8n+>y4!b)s(
zxSL&j%Pp@fE!n<T;L_qqY%tk|?K$w@a_W`+J;%P59i(BSsHg<*Ah+^V?shNkrOQX9
zqrxslLu0cv%gZf3g~)fL=w;g0)an`RCB$=V3o7DrT2+N*P0A$L=7N2XU+u46!QR+H
z<n5aAfBL#PU`~(dzYxmijzp*KERo2!H5Tk?RIn|;qVnw%EM<PreZHa&#7v_weX%>e
z)V~c_MtnUfDtj{5>6ofi;;t)mmjKd_>bYOEvQyH(YU&-o7`0eQWbV2W_tN>NCl^o~
z*(M@thfQ8Nqrh3R34RYhT{~`RWf)Py#_I;!FtfMNc<e-*0dJ<tinUE%js@Vcw&u2&
zX}XX9hq8e2cBsVm%Bq!oxP1fX{FL}_$4C9oWtLZyg6U$KAzn~bQ$BpONiO_%IllCG
zskj1H?hCzKii9bid48cP_sPDlch$>|Q_dcUJbqx<9ldy<pvqPYud=JX6tc26`uS8=
zejS@1m&Vu~7L^xO&kvnEh%k!=x=n8YpH}4%>??cz(Tk5=m=VANOt^D%zMCi6E;g$*
zYU#@#eQ@}>NukYfk0ej9kT<PX_@{MRBH(W;Y}4i@NFCyGHFGa{2ix0<`cqf_S-<)3
zFKK&+chHaEVo*kLJnNZl5S5NAGFtaK9j2)hKbCk=eB+O;+TZ_;+G<xk-?p<BSDrH{
zG`LV%Oul`y+0Q!n4zw8R=#MxkSoIyWD-3o#;+;6&d9)!kURd<OFwXZ`fB*yvTSEe0
z2P#f&Ln*H*TP8qq(C!#1mvv?J49Op7VR_NY;k(Cs9N`X11&60hh)XIIQw%Pc=k)vp
z%3=vd30?#?8bkR;BmAA<c3>EYz%e4kG#M#kL83$sOxqND5tS-$;7>o!{yLo!m>{+0
ziggrE6UV0#zimA>fHj|9oHZRCbyS@BV(5DNAGp(pFZ$TqtGW(^obT?xFTejec}er%
z58xf36F!R&5zM;gQ0x4D6mQ_)i{u@Y;p<+4bq4+#`ltf5e_^Qk<i+w#&CsIKHVN55
z-t^b$_DH=7y3|lyI$tqyiXFsq5X}^xP)ygQVob@KqaETb0<<5nEHZopNf>@$=ac36
z>oYkH`4s2fo-ws8QVz-tuNjMpK?K4daI8}O0;zV{_<{EqFaDnd{-l&>NRBHjmEP^w
z(K6%xOXqh_&7VFBioMkB8fxks*H9NXSsLM?dkzn!G2X)?3fV{qq+u+SPBZ`^0(eO7
zB_M=R@Jbwm^gn#pdim(b8bF9ZmEosi!$-wR({^CC1&j!&v=>Nl9;pYuOw#&9s?<vX
z<hqJf-Gx-?it+W@gQDJCy?njyqd7O5x2xQ%+>xMmV2VJRA#FMQTUCOcoD&@tsQX3A
z<UFL!mm-JLCxM(jJbeJL#t&pTLW%upoq2^9;mdw1rt5qe(Q#B8v3@FQ(ob$l=XAgq
zb<p~Uc!!ME=(di$Z^yC4KeP8V_Y`F02!kdA{+JX7Ivfp%9iJQo)}8Rs0lP!MwicH+
z^y?_qOq)U@%Ewz2a%v`%PA(-p)_!GlGP|_J41iV2j66yU`u5$^xPAv}cGg2~`nYDW
z)CS+a`=9Bd(et$pqVxu`mYS%V1BjeRqx(ccYB-_QP@j1%mZnl7m@YbgkDx?R?dpF}
zz0(uYaPxd&aC67pg~{o~OWPU>Lybfg;R9E{Y2u>K--79#6Z5hE%8US*TpnAJ-+u6^
zX2t60<fVq{ylRD2>#s(vr4wwM-lnM>nOe09izT=z!a6A&k{ey#QJdV@K6f49=WqWo
zKGL)QV1J|zz67#XEtoso(p4C0P`#gqQ)ho@R{N=gq7XTv%B+(9p)8QlM?JjW9Ml4h
z)T1+OLC?Q}Xh@&%kc_-u8<@btt<;&g=~+asJWfw$pXL>l5NJrWi)`a`TsSi<N%DZe
zVX})lfL)Py`^DpZxG<xEv(`>*$D+#IT8RkPk-`$|nlbjjX>|<y(yV6SUUcQ57ab+d
z(KGzkwsSI0WF3C1C>%!D%ix87ORuQP<;zsrSrU~nv#f5U6|nGRMiFH4j?jpjo{6qX
zIj1C9nA)#UoC;L`^#464qD~-fPDY&siS|yF8m*AbMpInFI=6vhKXvP?8V$6HXiwE8
z3*W)ik6?Ne?#R}!Sb75zog*e4vOzu8dL0HHh^odjy^;9&i*@ufGG5}{rj;17N~7r|
zHRJolaH^S725hCt!q}?gF<6UWgUXX;5+CJ{FlLJ)=59%Z@;aPbxu?bc&?-A_0`C5J
zO4!J*fn$eGyc`<23cKe1BHXsiO<JeMyH{Rc*2}C6GY{$M*{$lD>@JtbOOiw>)pD_i
zul!09Q_`&e)6z}rCX;6(gB$u<jt`H7m(QoRlh2<bMqj<74sZP7`u`3LUjN4>doZ(f
zwY_cO+W5@k@I6|*MsB$cWb0%`rNwz2hn{Ts=N8oPtVfygjZbp__&;B@&EA|Cow(4T
z8gTFT+o;KL&F}$WLvpw1Kdi1aR0;Wmo_axb_D!Jaep#P!<kV#_<AUtW3j=LK`uXqU
zJn)_tqVnigW-*Pbf78e3ovY94P(WFfb+oU2oR4&&rMj(}Xq7J`mNR(4c%$#m`u+x`
zk^!0(!_S`MCqL5D95sx&3TpL%^_4#&u<OOJTV61M$3`c0^~JTJsTkdyG^cbWj^c?|
ztXr2nQ%o(=<1FV2)55orXHLn<8R#U<4`FyjZSMpT;7-+Ox2kD<R8#6hW`q`?k5j7E
zmmi;R`TP3t+d3`o&b+^$-``r_&eXu)b2#1bEbDPrgWWRzQRDB?KSm=14j_d=wj^&l
z(@C~vRPSSNjkZutTSrZ)cVZy;EN-N}j2DeY^Bk_5<}D^W+;q%yK(4}wO>2eW3GUcN
zY#+J&PVt?!oZahNsiZAw`V>K9d!5F5h4aIb=k3zzvehNHC3tpy{N1&;i^?h3er%H)
zTNK&2?)B*oz~(O4os_=0c;~wIV2WRw9yza}y<TJeSTzRnDfdc0N*i_8dd}VT*^c7U
z@6zbbDU4`b3rm6#d-yZTad&BS!TlSFMnQ0)rkIfz>QWMvPBBB&XoVPmT_dcg5t;+R
zFTngzTaG=^EyUi?T&xp3sF^5hIXZPibzV!@iQbMu`}o`0PGjtAL5jWL)+NgB|9)!M
z)mwkZCyn*@(G8FDbm&+!L8uE$E9L!0l{{2eAPXNbGP_4Ne?4G%5G*^~C|k)k17QoD
z3PEe9&0hK2+Yzg55=sivV0njOoTIR!q|4|2U+gRkHW!ozNfo67wJz#=TGTiyh>94s
z8tAht%r`uTYamY0+1NHRR<L7Xn3Q|o1E_<JXi$R0VybIO7R5a;pG+2soWc@#*r2j9
zN{~!Obxq|3V}(L8NhETMN=kPj+wNitBnj+1Y3%7;<0J@5ia|tRtALxyZD>(!fw6gu
zJQ$0=UcNrP;<XhU7kk=dr(mZ#Wb)xyrcMmDswT}uOc2WUwA9p!o?KZVq<^bfh^SUk
zCb|Cp`FZ9JM7=oc$uuJhhAma3BKVYi*@>R!$YJ{%R(Z0h);KARjI2p;Am~%zN#^Gh
zBh8Gve%BM>&`oLfnHG(h$McZ_j4Ph&$Te?3A3w?Xt1Mg(qgZz6Oq5+F+KyF?Le&}G
zoSYrH&Mj2BXU^4ieyQebtJ#H&uQk;%FgyPQ?OL<sIzQ*)y*N5#XgCsLFk&I2V?#y;
zV<Cng(%Q7xGna>ISw5C5hF4}m{We8qLr<YFIzWdP=Up`<cewmzR@Zr~M<it0F`Clr
zL@H^mwU$~`(giE*EOM-j16^PJZgBce4=PGI+J{Ex7!<)tkiKI7fJ9MYYKm`&zqebK
z6Z{4h4q&&-@6t3mGAvX#&}4grtG+hdkhGNu-AdSy3^0Df1a1=o59@eUist3;G6d>3
zo~s%P_@N;gfrd+)i$>QCszaqt8H52asdy4IkRzE~!=u7jztmY4kZZ5X#^<VQMrX@q
zW3$z&iP_jvmPodIlv$j8lavU{!Lp6Z{VOm*5>Y@*;%}2Q8-Wecv$m}?Yau#}GyTvc
zfEi3p=R{c37M0FHg>{ltERvbX=O-jahB*P(_!$6%iSWZm>M*$2yeL5s-ZzcPCPSV^
z)SER$;#_IolnhRgE$zJm9avb$Wz?>E6qRK~WbTwJa57__=Cbws>=BP_^V-`$Rwa*-
zG?!#r{`vUXl<poYpOWV?Rh_+TC6x?&Sb6|_Y8C5y2v_5&5J{Yi<k<*?&LB!TuvF&u
zhiZqs_;8W9Q~Jkea`-#>&nKzpPvt^uV(GCXg+p3vqJ^*`yhlJIrc~cj1OCQ4Y$Jc$
zc^5hq-!OOOzk%&%R_ABO0e)xyNn21gjTaU*jwx!MjxkOM?r0`Jn~9r}KBuTYH#ZR>
zp|?J{iF22Rk_~`AzcXMrIbr2Z4x#;x@zs*QMPDQkntUaBrHV(hQ0;Nk4G{kz%O#Ay
zp;=YRvKQ!KIXw?e0%L=zsfm$D+G168*c^MFJjgsMR%do3+yQGC5&gzHA;y~I<`^#K
z`0L{JsDyi4hGjIBkSud%X{lX*Ed)<pJ+qMxLGQw#z5VGu82g?w{iYAO`COLouh-e8
z=bTCRX?gub%PVrvzC1zwhUAH;S$5)iwZN5uTki~DkDIV5!^t!wXmiyHv5z7Q(LDs?
z=_gb0t%I~sLl2H2JcujodUHY>o;aCogvrR^Z)H*gm<q|QiV_erAZR5GOz4O2wF0Yi
z2VT8<9UyOd1fEC>_41)b67lr#7_j{f;GN>iGlpVA?d=bU8t>9suIBBU-7LNEE6~DT
z))3`Yeo{yp-hhKG&A&q(c-iYWz&_^r$h&r&aEFvxyX0MY%|l*a4h;kKygylJBpg=;
zh3b0fF%0@F(H_S$sF4F(w4~EoYy%}q)I@_Ci=GZj4_OLwc%$2ki@u71`wuL_dOVVO
zaQ;l7Dtu8?NX6Cp3ptaTpPHs)cEO9N^;Tz_U{CT0wKte=ZpHyxtn~KQJj~C{Ybb&2
z*<FAX=N<yh%aD@ZQ}?)|`nBzkNS1Cs(k5zTpYKI8keCwmjrUZpOpk!6V7kW$4&DJv
zV@<;lh68W|{_uCy5!2=~EAACBR7Sgx4qy>83te@@l=;lkonFb-0U$B6+NferBkvKD
zYH@8W)3YL?|L=gt!yHf9s*cl#`s61JO0M?bU$gB;>+tsnfC?2DEYxF=KS!`$xXxQo
zW&C*pw7N>itOb&J0OPjlKvMDh>|yt8H=ETd{~iAFZ}qf6=}jIdE`mgxSde|nPp>uB
z<N<*jdNe1}ifoD77H6uQkxX}gafTrpO(MC{v2-sI0iCi%HP_QRSV^-KshCn`1|!PZ
zE{xz#bBb~$_}a0&9j(H=JgtJzY<?amHD4}rEOp*)6L%?(X7#kCb-K;=a7l2x@vumB
z@P>JNySM{ZaeqCVI;Y}%=f3AYzYsFNqxN0_hzrRkQ~nzQI&qA1IM^B*kdjR{vk>Y^
z5+x3XZtnGGAzYZsT^z7UUp;k2i7dg(&Z(Skk5h<q=hKUUYXbrQA%H>V`b0$p66s-H
zKJ-XpU{v@V{(-Bft~R(hC$A`|-!IV7Qb2ItLty>%Sn#CHC1zP@!7OwO`X9BDV|*{R
zcg)GWw9@Xa6(yA9NE&63?n!y7tH3_jJd@}7Kpo`K%_%hB+Sk1nea|pEpmLpLLhYM-
zGaYPg-miUY9nDtEhmG{nw9V>oO~wvffq|KGw4k?P{{)d(m5#j;YK5NT5ZXo+EmWI7
z-QhHwa_(!6+bFR}wP{4R_Ptbs6xIVW!~HhhyFh)x`gyS}j9^`-T#ZM!7ebd0qm#dp
z*XFPgOgLPPehNv79?v4r73H`4MGVVxz~&L{r&*jg{s!#*SZ;iTU!4V*MT(5P>8Z~i
zv;%J&+szxx^E{H%xFaIx<c~j9SNe7zZRni3yq9H_a$vwCwr+aqlm)`SAuxqkZQZZo
zCJ{bpvu;n+eg%1PK93c_?g8Iy>`AYcTtd6&@o$U+{h9mK6Vj`M+(I~m(xgd+$Cn9A
zn|>kgkRAF+oUb%yXWajEoWDKBgL(C?n$GE5rX908tKedI0P}p=9X0sLY!-&uoh>bK
zi8BU;`!ml;E01Jl<W*c|gp=I^4Kfcoyr|;`v$90WY9HbTE4qD+xE^TJI-2*a*9)O1
zI%?KFGdx|TWv3pY{TA!K1Gw6HNUjsz@)F2J$c*genGHvuu3t69`G+w`<I`>z^|;#2
zQB-DgS(sUaYU}+}Zeh=Kt82yOXY|GT7s`w8$f_$Uke{u!VOKUz3}#bocS?Cq?k4oe
zJ6SHnFQcJwaxnxSL+8(%p-34%S#_?=J>N6`a#`ic8as@23c@xtM?6m-jr(M6%&lFY
z=VN|V<KGbIvX^Y!D5C701MDjN_NejHU`_Q8URD`!{I(!2|0mTi3Mk3*75M`YlcpeH
z!3Sb3+8HDlZ-}WLXud*dT~2hE#r;>%N?gH#CnJXC<Le*`3%Cav?A<3KbSIB$!0!-#
zXDt2Po%2y>%-N_rv~w8ysI!0nbC!ldpQm5A=P>B<x=4Jx1NP`TJ9HD;{)hs&bc%Gp
z9@Xb&nn@i*Xa_+v*@Ds0MSyncGU4kl@~_P;9CpB1jnH_wAvC3psMpMAsJC^CtLDd1
z9F{sm$b0Xfo|-Knnh45DZF)GyeHiO8G+xgbnly3aQ?-S)zPAVu7nI=V4!wA6`Gf%_
z3?=SNME7(1zt1!KJ#|*G@t$JH<adfz^~Sd{;Z4~_z3O&gz3->ZV1dITcOXxoY_D6}
z+_Xg**uRPtAR$gY^fP7o7A~bOn;*I$l3r@78PD{1a~U|&pz~XW>J6w98mSv|fjsT;
z3k`13YXuXSf?VV-?}8`}N}91=wm}(%Mp3#9h_`o2YP51XFg5Dk1BraIiA_BbD?xg7
z`heG?0Rpo7&Z$FVU(bVl1I_<m5=wjnT@pyfncDwy8gPA@E`%UZCc_LW^+;e?+P*0J
zvnW=HMYvYBXCRlt9*=IJ<kknEM-yqn@SfpIU|p9G>6q>Q!30P+1k9xN`*@rX5(X4q
z5{XyeqY{mWTgQVISa2NN-Z>sL!c-9K?p-3}rnGAOhpp0<R1A?@xt@V?fJwkDo0Sd9
zLs8TyNm;>a#Mga~dnej;ri%kgBJ}hm=_GVE`*8cz%k9>?S}L^Ih^wdo^6sOzD9hY;
z``4E*-^ggPk+!k~i2IMUFJaTJw`s+h&9AQt+6k1wlPqGo@CHe*?3W?#G2Mz&ypiY&
z5#AP4c88d;;NoQvot01zmd2AXJLI(dV`b}df3$h~nyYSM%H)!r>g>e7ZhXGmv9E_&
zk$jH47Y0p%gOm`vNd&mQZP`ni^sH$ry&%aU3quQFl!&3slLO$Lg?#X}lg(&jn9JO}
z#r@kaHH_fH=nwhq%JT9IgRuwgZ)88*zw&bK%3uD4PyN|n?w$plEWuu#Qa#;OxzMxI
zFnYho``0mr)}5Pq4(=4rH4g><dg&H}3SG)tN56!}Z!3%TjG-PL9bdi$EywR(V~W@h
zhKDDYSNCt__nuahw^t|3Mb5_Ft#t|@o4k8G$TFx%t53;eVJM64U=1u3ZHej#oLC3}
z1!dRNOr^`FlB!AWfRH>i!bnzPQ^>vqUR~yGQDXDZW3?aLHkof93|26<U7g5RITPA*
zK*UH7DNjdfpsUy-5wNRR4hcZ@WrwV6fmFbj4X}|w8|LWLZgW(pu}F6hPSzO&eTqXg
zx_O<HMD6ei<Sq|SE<VULI#S;a%-v0#)|Rsn30r>qmF6Tk#8{?y1bX+O)1!Q45QlDU
zeyYdRXaZk9v@C5|09HiUfo$YdNZ+BjP-*;1wGJ=KVT2{jn1UzDQt(XF7L~drXNWQs
zy;5ijm~w~27Xu^$wY;eFCT9+_=8E=hRB%}0Jmn*Icbpkn5GGP5APt7TV~LPkq8t3-
z<z)cC0YFg!P88-YzX5V-w2u=&L?KqQQIhCUH5!Rf86Rj!;1@t~LV7wYzwQ#K8nY4E
zE;*3ps1XF&4s%xknfCv=gRq>VBs2)?VvZxoF!F8|JaF@`Uo#3pVEYdqf4s={If^&c
z1akXQ#4Kfp#p)Uz43%O?f)h`w%Ax)K@falxeM_ARaH+37wvQPJ5^@OM#){SPXcY)m
zy2XlCZBXgLV2;%pa(Di(sSMj)e4@_TEYVW5pAm=>6vB1R1S>quYR0yMIOl!IkhNWS
z2nD1$l~-y>e4uBxqXy}2Vb@bS4N)d{Jf~DJ#Jg!&ZLdmU)n_DDkR(!D$~AAieD8x&
zvsqKUz468$U;fsj@%Fa*rA?~cjg+Wv>zdG%AG22>iHaySn-V}-nYZiioY#(^OwAjz
z6-ibGG!poRnS+ZqlR7!IL-sNGDAuUQiWJ?ld^*Vy05uCEco#09LeGSOhF4vtkPa*3
z1Rp7NRxPj^RRTMkV(XqEv~oIGa(7avC!xdE?&*~EqKB)gV^FyD=~el}3}Z`8Z#!si
zyjV3J#Jq`fn!98Uv=gQ*D~mwS$y>rkwOCA=>w@X97%L1djjC(!m^=zqkBva_X~1FD
zTZMTV3dfXE1d#FEf@CRB#Fraxj$61C5fqg;u1n5wAm^PC67)V2cVQ_;V6iE^BqU7O
zh;v2yqnZH7&P8bpQjrXJmIj=mmFX#Xh+4K4*jiDvo~{)>Y2O)bO0|s|FBJ`NSpA97
zt~=04UAWCUmoK{XoZ={#?`)*$6**glyMkU%l#1oeHqG351F$1$^qT3efk`6dOP9Ck
zdvN!&|9D%c8T0}4Xv?geHCvnJmdE=xw@dZR13KL?{h7F7>#}+Mw&JME1v-hm{%-xa
zJt;O%Zc|<*s;1iqLYl8!pL=O3HdIDwMQDR1Wxq_9#!U%IX;W@Ch88lT(<tg<E@Qc%
zatn2*SIlP#PUzQKRhdQN(MXf5ipuadp|*EB=t=Pc0>RA$fUB{c^Kko25Ka{6py;zd
zkB}FotyAcJMlk?UDghV?B2XJ2hhKFi*2x3dT_Oa5^y}>7m|Y0h<jLU#Dz$~=MJU;!
zK9A=DU!NQ{H6|U(ErctS`370JQMWnmL6l#tiD_?uK!Vh;gE1P*cbBNU6@lrJBfTS$
ze2OsDdI7AYG%7}cDCuG5Q!_$=1!5s85eb1b%?*fcRA7a}QUI4t`kBCR!m!d+r-KZ4
zYPTO(w%(u+V1N)l*-99ww#$ns3X;;rIkc>LhF?xkp;!dcW3^om)w~tPVHzinL`>6F
zGmOrJDB3<4OiF{4iB>{OkWls!GZjH`BL3PTmR<`UjpJlvrXo}m)+~Tnsc^%X9YK^D
zP!=c(aEjbQ5h{ruBvd}+y!!)C;o(FckYF~jE%ZoWMRl4q^3kP<3t7k-h(3VJn}F0}
zNMv5IB=B)0;wMBBM$j|_!HcI92({Rf$cexzV?a7uw5ft@>hzWoN114((5Nq8BSAwF
zVM>zowB<fdf2!4kmZ#W2W0^L4sB)AfC^@3G@!<7CB0*Z?d)lSPfl$J}5%}i=8VIFp
zLk?);Z9!ct9wU>tPHt!}SbNe*9@8={3LyL8<*T6>)!NOWZFvB3pVg)VfxZ)EU^F+r
z)BLhcN!@>ya}lsWDEI&pQ2kD2XPj5<W9MlK0;WY-{f1Y~l0t7n`mw>8wnMnt;|KqH
za2g6gdi#PPXZzhg|KESVDO)Mp024d9$QxPCm-=(F$2b4|=5iGql(s4(iKH6NwB~Ow
zPo3$6bE_0Ywig5O?Q?!QdDD_{cV4wLH=c3bw2n7c>uaauXlff?2af0xSl5lJe8>wD
z8)$7J9a`kZg$5|Hg@DPhl#P?J#oENXNf##2>;R`-yT6J6K}x~|ELk`N5Gx9(I8Re6
zR>%#VH%m7hF;N-c^T#0s*Nd=NF9AwI1P&V@q1{jkvJ2$}NJz#LYPpCb3vBhIJxbRG
z1->E7YOCr9@r>(t0We+R-K1SbMPOc0%x<I*_hYPcy1(jyqgeJ<B^hv{fp5;^ESiSi
zB|AEZ0HGP7v{Q3L_x5_#()!T$MITfuLlZ9q5(eo7AtY8uVRB3;CZ-ay-mYOsb&?;&
z(Sb^&Tu!$T#LXx5l7c<2>PaCswlpaUB%!OwQp3y+cBZjlx9M{fojYHXhVd_k#|2Q}
zwo-PLhf^J#e?FanUokKzLlDi9F^h;etHg#pCYLaxo}f;}V4ZN#n_5}UOXc5vpZMt>
zeoea+@}g(Tzn53Z-YHN_<LF8L@sC!s7J!pHAb#XJD)Kxe$@xY9r!#!}^zxcK<M|b)
zM{PnHEqr^=hd%MCt_hO1VKjdIAx!>l@uOt6>QE>u_mI_D_8I&kpC5L4$^Tf5`m{R%
zlO%&v$z-Wm!3hk3hd@){JZS=y&z%K^%efFa(X|waBcaT&E1$HJANgW(v3DPXB#dEP
z35+HZy@g%Yb{Le%?<7zyD-TdZjI3gNP9gNO=UFis0vNKBHn<#)e%#`&twN?u%{F~@
z3+2Z}JGMZ`Tx)1kDc^|`hV<KYJ#v(@Q40cHlX*jO!cvSvVV_Wl*E1q~7^AE8mQXhd
zo@KmWYY1X`BIbD6#)7ZzpCTX<3qgslXfc2th7xiNTQo)Z8M1GIA{H{Ch9ijm)-h)Y
z9oRa|<BsrVe|zE0fBm2ysVU6lW{d5KUkvok>UKAviJtkvndYEMYN+`Y+<>{%lZCSt
ze||cvch0$&!M4NBxq>D;cQe>_7jqX8n^;_9f!)@%^Ww~_XeS)amUJjU>hFSvrCiG>
zxIU^JaV+H?*sqttnNp%%n8MNBe#F&LU-&YitqdB<B3Fx<AOrCefdotaa3q<O_!3kN
zmUg%rKvDqM2=MkR?JzALx46I0=z)Lw?`M}doJ}=3O(rI&<gzGenjwrbhI5r2iKlgb
zV^jtCL~`AbC`1=w<}c<5^sN9^vGMdN096DUI@+qr4F#;_U<P5oC#qoCl9mJbsA<R@
z6U>^%^W$>Ur+gN)(fS@;pmKCH68xZEGgrEcqs+Y=x!LWv{{DlLDRkPX*k}T(=|=^y
z$n1CXo_~D0@g`i>3QD=yL{+w8-~t;LbNJ-+teOr_<sqRz#i?D_Nqy7?$@g;>^hD#s
zZ8gGuObN=74$86op@3zzCV--rqDepIECSf9MhAu&9F^>JkClXB1pzK-Z_YwqV6jT4
z%(eoN(?hPJT7(|utcER4g@}PC9eKr>4sW2Wu-{5_gkOFKmAj9a4dDsO9?Up+#25Xw
z1@suE#~c45D+R~sh>_h_p(GQHXfIWA;{tFMVtoy#9|A?8qqIYlAJV8dIblzQq4hi*
z0rw#Y-Yy=u^qMqF?nq%2;*z4-@)bh?I6==6%;kw|K6E+Y7n)GL%tcakdrw9R?q_vT
zy{wxv^iDKFBm;d`YjXW}Tvqkw)G4#E&}zP|)}WNnqS|OdR3<7)lU=K#a()<mh%l6i
z$*@F4Jr<x7YlUl3kY#&q1Dz(oqSTRL;pS3SK)J+aT#g**l^qQW_4Q)<$=EPyMfA9C
z9%~&L7j=5_{Ob1b;5?+t3FStEt}##V{>i0BP)uTyx)WV(8l}LBQA#rf%aMGDD!c6p
zaH^uN;Z&tnO@q2A>;_wyuA4b3bxV|YR(S@E%tniaZa)q~MHr7$u`eJgZRKjTqD1@f
za7{cU0-h3WgNV8hHVMrtz?jF(*+Ia}2zK~-wvAhnqq9?JaUbRe>!AAxH%ImZR+=po
zxc*+3t@FsyBY@Pvdk2pmNiv7I_Erbz5_umcy}>_jeiN?Yl)j-Jc+>V=^M%JhKQnz7
z=p9>PH><WK>$W>rCEtmKP_oY@%h?^TPmS}Dcqv>J>J3od_~65D@UEVBKK)6}N7Y>$
z^T0q0T}GWpokjN_&mo5bO6Hw`mFzL3xA`^?M_b#k-@O0oA6dr!eP#1+UJpR!h7%*{
zkvOewl!u^;oOR^K++5#%|BP}inWL-yrJ{HDPrOVvI41SQ&8-;kP}@t8;sDCX<vpj(
z<(38d8Bst}8bNEl88GRJ>gpy*BQIO(nAjDxB!>NxH3e?uwMM0<0hobA)(Jqep>2S@
zfw3C`z}YI~XLWkE-OT>c$wNzz(Op<chfzy>7JjsUV5j#^FGGXZ6#J@GeH&mdqcSmB
zI3P@eR-FRa6zwbahJ2%E`>$Us))~>M^sMV#Yq#dER*r00I|TJ7OFUVfKypEMJ3jHU
zh(@v44DCVId(~e)=ai#X#YE0v>78LDdW)SXGgCmtAL<%VpM^_G5U!#~rz<8Q4Z6Ya
zpt&M0-r#~T;F_V7FJ>wRZzy)Jh+xP0gl1u@lT7UA_{X+GU3QGE`=-CmsU@2B!YBNA
zKtGKSi7Y#!8)>_Lcm@eWt?J~!=zL_hj@EKX=@259TMZ3~nIs@GrD6e^XsxPEDs9(X
zT=dx!c3_z%nrHr_F;)$4kmIsCS;UUesN*IGkW|e~I1)&}E0Jbgvy~~-6n>ZI!(yb+
z*=@xvmLABG;**S&!%4t<%Uvnn1OmWmP!tIPE4-p1UCTbA`<1`^YlAvjYikz_dy6gP
zhE2*UZ%w;FU^}lc%1dM&i`jWQQte0JQCGx}e2?3Asaxj!t;7-?l-hhG5$nhtITFh;
zB^-VQ<`RNEld#uKkaWpqofV3dbS$y4O+I8+F+o<6EMu`|Y46*371LOD<3B-Hmnyxy
zhhjG6iRrDaBL$d&T!pker9u@eMwjDVxHF|!78yZM5@+dDwA~AwDB5Q~xjLm+8sAT^
zyh~TF3fIEmi{xjWru6ay?Y-VvMRVHV3rB5tHm);{!!FL^a#3-#Qzu`ay@s#dA!-qg
z;cA`mpM$XJg*;j8@Tl9$@t_u_+2VE}1r(Y+EE2J79L8qk#FnK3VdKK&f#yp$TCHNb
zrI;Trwks5EZNdzjCUDGZ*B^DNMRQNK%oz%^SIR-RGG@(OpT>2Gv)=+65p#B0yok3t
zpbGQ>mCutP>@5JDpDFYX!Y#=BEP*OnUZNkqd>8pl?#1{+_sPb0<<EF-DmvSx`J^_W
zGAh{KVT2EB$v9Y&hH4lsEc5}i)oi8QksC!>P<u$1=Mroil8}yPf;=KfsUcXdoovQ~
zAYhi{4uRGKlmwOpEf&KHK}S8ol)%km7&4ngA1n~GgvA053oA>*>I}Oaggf=X=jD2u
zjwl|hJn1ANde1QxQM1#E{>#+5`XIMtq-oor8qtfn-~VcyjA^_qQ6A-@inPnGAM2_*
zH_8M5Wg$eDq<it6fPU?+9~%`Ns2HM3F{ul<nTY*lRyqco)|FUB5>XKZ$;2ir%1lcs
zs1j}`rV`9e0XX3t!9!Rt9EbpxN&f0;OuIO`phygZ3%>*|3PV#RQH2$=&{Zu+5amYD
z#5fW{po_8|&%IPkF=Bp~Yc^*lwQ5eSq>US$AH=IK6s_@47J!-(mY*4_AKW#lNYLw?
zqGro?W!`Bmt_&F-VmOhj=`caf3|oOJ1?;kVHxU%{0m8H)Gbv0vUWVc{?<#l#imgxZ
z47@2l`vq?$nCD`8j2r)g%<3jyPX)W)wxfgM?CnBSyZs~T$9l)U)5h+#^;u#p=u5;b
z)A#x(AN&4!9e;R?uL93f_1ZT@-w!*H4g<4{Hf-CZte|qEZ$haaSc)X_S;H<`a@xJz
zeG;S5bYcXzbH1fZHCUt!m99eNm0JpXk>ywTSd}LB{o%slvBZJ1S#<8K;v)?HUeaMv
z;-GK&g(43rRhNPZ2wryBXM)p=b6&}ZJd*1ctBe>ARK4wwty(5I=20Wv=kGH=N&kgL
zWK<sHjw|sjriBAYHwBlHuHZquCpXRwD7l^3u_MBnfSADr)P{L<un+Fc(%!wp>~Kbp
zkSm2QdV6r4kNT2$j2P41a?Ub<5igJKzjNyd=Hk>#EVPJE8Ve}{#UW;MBK6|=CtKo{
z^m1v3=@^qe(t{fZA!l(+&2GKLu)r_MqOm$C9Z}4vV^kwwXSdD+(m=D$vA<Dp8sqZ~
z&M&T#WT$y_Uzk8ztP|D8UUYJhS~8(I7(kpGk?FraMr&!%(t0ac+!yRop%Iz@Mk2Q`
zUE_IdIwx*-0m)TM*nrNfKL_oehKmfsVBhiV?nLbD%t#yDl6gH78e@EYD+?qyT|nKN
z6C>a0ERw5MipiVTza?fk=iSS!&fBsUq}%d@%sFQiuLMKZI4R93PBzE(!dxSJB(5nR
zbj5m`C)c7U7*|J*^wP=xP3e5~ZlEhftNX&?#q4AGz2H2jTcXQ3HBacxdLTI?-d_n_
zC9cSFox@14u8^^FdD8&)b9^b~_IjPvKr3{xpj8^$^A&_;p&Je-94rH+VL5)rmEOaO
z@q^<cIZSSxN0kGYU5S;4MgO{ilpJF*M{{{NG57B7-)D+-i5@Htj<*+kPgi>tMzyY!
zW}bHmusXiHkNH>;F6RLNqErn85(+`Ke{Oun7SIflS|e4vsj1#VYr?v_xrN^mP(I(`
zKy6aGGujW3Stu9su3pSLDl$+ha*CrWJE2&Q8Lk@jh6Lx&rn_OZupxcNfMH#Kcke*L
zV(cevwuTFJq9Q}6-clDCh(N46+Qvl!uixU6JKCONnCl<jJHwJJLLGDi8gj>hEJMC+
zm7YbP>$t?aMwlf_YRYanN=eRvI&(7CtGw0}2S=N8l!JM$7(aa(x8)8{eKdsW5<KTG
z)^V&2uV%g@3*IMD^>V%=xval>+blM_&!xOKt=#CI(B|iwc2*}C*<ehAO`O3S=3umL
zBK0U{In?B+I-6-et0ryk{bw;vPgRNgUo`1Z#&c8eXOC~3!m?^8s@dpV$w09tcVgUZ
zvSNT;VIXBFtMJ4ff@P-FUY&xJX^4b$cj!?H0K{VD3yqs6LNqas2tJrd2XoM;N*YR3
zZuKJf-hc`%_vQCC6XXwC2#!JYs+grP=y?0*mp{D>>6tx0T2e&WWifg5QZM6~5563&
z0cNtUSZ~o$3gau;7|#F5`m)|;?b(Ko^bh4O<V~M0eH`EV#iJkU{a*hNbo`GSNE{C%
zu#bLOJst7-a5Ud1)Gx;5$;)0HV}EEf(k75+H1W_ZH6kt9m{t8&G=S!^4m2#GAp5lH
z%izhF)nmAv@_a_HbWHH5GzI&@fnxcejwrkjTzODI28KhlU`#%fHYdS0jk@bUsH}Cv
zu;9&^Y-k~D532BwowdM6U7FDK`h$i2LotDlHgK;aFmKQdLrrY^0?ceOvLRVK?`(>q
zD1#+5wy3op*|R}Wla3X??G3(qw!aD3kZNuX8wnhG%KUuDywby~CanT#Mgue7EclB}
zl!`?Z$WR+va^{V{9BO6BDLZoZTxw^UKdOzguBnQfl4@fnP{$pn_cqHwk}cPBr5jQ>
zL>{W0Llbmk3YlzC@U$fgd@`~av!-DnGeyABZBmqyV!;&B=?DkY!FYpKwf&$zx;|Kl
zvG+7(SGovV*RtG1V!8tencV~)+cF00NYQuGwT5GV&uwk>y;?7mTA`EC+x9#bcIDBZ
zSg43o5o@MFtE`cg=sbVmU5t>wM5EiTzj4D!N0D-o*ahsZ0qe6DvPQHLj(r6LBh{Ii
zS>BC^Vcm|P!L_M3XtI$Bi1p;JGg$%eF-uwVQP`Edt)w>+JR5{3+<l#E;YrFL$~btm
zT=au@^tk=74-5XNwXvPeB}41V)UQK2V4$b=+Z!r5`zvFbRE9veDq#G;wwP0zJL-t3
zt8xWpwVTO74yz7JSLx^GI$v#!6#&6YySVLO2jn~olI5$i=Y1M^IIL_h$RpmHp>_{r
z&{^9K8O!7JP2&(zIrgm_kuRK}HXzhRR59{1m1a__gF{=$WfSYy2^C!uj#rtZvaK44
zu+*Ox0{m90atiuY04n+g>#L+62flV<|B$VDOTJaHJWo-QpOF`YpGjZ;v<LW9P}wKq
z55NEBgYjKNBybwSJZ=HSpm9a}57dXd4P*&A3}2qbFakC8MtVD^6qInbrZ|lX;>kr$
zP*M4$`wPQf;)Fo=iR<O0VR{!sDbUMfA<6rQMwuk5Y79P?0(XP}|8S_Q9%a#DRs{CZ
zMN{V9f=lOmiW*C4#o7qJ7Ju={_J?brY4sL*rrhjhd0)z6zfN=(o?FVbhKff&m<5yU
zkS;B#jI-Imgm6Tu8&)i%i|46Bn8{u{uL6~aL)N2AH`;bUDv&G!igOW7V~;)i-Od6y
z&DeXw;D<jOu5v}V9eY`{C44Ec*EAFkgR`XEkk)Yvt(+AYoh1Wi`e3(gb^#YQ*I6Uo
zxF$#Bmd;N+NHL?^gOsP;VJ1mOH@f$)O8&96iDy`}&9j__zUPJOA79RT#K~y$FVOLl
z?1neJHPheyJ%AB0V0>RrH6vtI@K7Vu#nC%VLW-b#^RmY5bObO+e6O=;Q`r^%*+OhF
zU6^bBzt}DPeRi*G@c9wg1EEmB_W9q@JzD7vb+7vW@bM4jEA@&OpjJaJLxowIZP#yI
zE`3AAQJSa!aUA8&+tqnv>J-+4ToTjJF#3zg>Pqz1wPVGaj-f=uwIFp#H<E|+BK12f
zb1b{29N(RO4O8;8)^nMfJxhviZfS4Sa9Hy;E|%6+YXB5#5tps33gWu#2rl2wbQ?B2
zQ^r~iJ6&Ki$$F#AC*Hm6{m%66C~T8K)$T8*bBgLPIV;ks@74C^yIQ0)t!u)1RcMoH
z2yeShsX3~5tm<_*R?NmCYQLa%G9}g;aK(F6;-h$(Dbr(0%<5vZTD;i2mA9^DBYAVt
zJ@2laG1ZjSQdWve$8I!T`%x;HbA|K?r~2fSrSV0i?8j`F-H^GuE7SOyX7`_7&dRFG
zxT<{al)b0%G}H`uW@ByoedR?O8s-uOs=c=|M}USgt*9?`va$x*GpHJ0qUZ39odyNX
z``e9*lFt-Ua}1>eYETox*lWVXy0$mRj}5exQ(Z_l)J~(R4mAOx*tE2pijaK9V2x+p
zoTEWa2s5zyV42;uwd~xAJ*rQ2!Gp%B*VdsXAf#}lDN_L}sp-5k$G#_7S+%a5uX>`D
zrWV<*Q57M3AsND@+y`j_8bgN&RH+GyZ7StOf1W@@UWM>5qV(K(ygfqgydkwfG2%5T
zAJC9}lf+x=Vf>r%lrmd;zFC0RmIEvR|8)8dZU{2i{E|Cd7>IU&MB>#Ym`ezx!X^2o
zLDwiDH92Cpv=AMyrIj41rH!7MTiWsFT18)UeuO5c{HZbwG8wshs7zk$Q}2|nJ}%cf
zNlj4RxRx))8tsv2k*D>YlV{x8)8;JVTje==soi&SOD;xB({9NB58+QyY%nUopG*h#
zT#hXFe{J8f2Dmq3!B+;Xh@>n}w!rt|5O2E;{6O08A5{q4Ldv7qgDb&3loLDLP7F~S
z#M8a)y{Z$GXj0@5bY|ci*$xflp2Kb6#gi-Gkn!zHXHK14I(M)<0jAI28nO>tj)+!V
z$Zh`wrMCac%Kf$zkITg%2?VZE_i4Q?mieU3r_#ERa7LLuxF(i0lB&$i2>8jXti>>c
z5gcksUb_7kwG6NV$_QvIMa%>eQ9s@_AbQh0+Prjavz=~Yo4zr>+63Q!!P#j9T5x8$
zJJSmCTxe}_@14r>Ldam^w*?bcHh~hQhuWp+S&;OU0i6nJEro<QR%*?bXm?;mifLy<
z0gH&|F&4nu^vsIKL6}V1>ZH<(p~}8JX(;1vRC1qP*!2Rfc`o^@%4ogo3XBclj;FrJ
z`Pty78b@$tsK1aH(IVt=t$XW3J?o=S@*(xM&RQZ!5V0R<|2-`Ll_)Vx#wDf_Q6`BB
zt$y3|AbO^Pg&0Pn%Ae01yF+ZEOg;Bd1>;$kNie>Np1M$eFiD260^B+a2~muj^e&f?
zb;#6?6xY2f6vVqA6p^W7obvvwM!bk-VV&y}fXS$yCtB+21CV-@M%qjKbF-`>eJ-q?
zbk;+Ps|nmV3zJo8b3}&3wP~vYG>|kD?>2Vl00wBPZO~YIrWWTZDWpJVL7w4TtqiQI
z$-%f|upH<I6-6lB*cc>XJo3>{Z&<RBP@eS1XPT|(vKDPN*6TA2=p+i|h57P2fJ3ug
zKQJuCA4rVs|5`I#{Q9uik*O=frW}~Ij5JQDs&`(_)Xca$3cQm8=t7BOzHm|Tmq(XI
zM@_dk=1p~mMWpLM@Jc}=JLKDZ;y)nM4&s8fvOeV+?_4Y?Kk0$V)gT5DIJ>u`5IRI*
z`Vvy9RFBh6M9W8<m~SUo@7$`&pLTJj@}$y?9W-6m;bkI^bSQ!9Q`Y@+cfI`<GOx*D
zYlX%fFUZQ8z=b4z^G_LRV?<^nspes6RU@Gtm|JdyLq(Q~8yV`LWp6I3ZcyRZ4@*d7
za#b}io~Z&ZK%WT&52$h@n3|=wcyqa(7)`)(VhKgHH6DFe)g#a=(6<8~&5f+$=CBP|
zB0s!(^VH#nVLJbzI|xTNziRZjF)Z@|A)p|wy0AAXl!WZ}?yjoAaN$CQwaz3xj)!Kj
z-8eZ3LdNd%8rRveG}@!ecd^E0;w^!_x;1k-$j&G2#Q}<Ig4{u$$Yiguk2&w}o+i(w
z|ASkmZ6u0u>ORSLw!I9V8MPFVX!4WWXFB%mHVCz#yXMkw$>gFndI3(qTv%D0O7w(U
zemKR7UQ?4HbhB4TqnmdUVC=vBiUjpH(JLlM5fIVvK7wLAV_dq((lYq$GbKz4#pDMj
z1!8IBOFQy0XqF7e8H`mUSeW9)8A}dm&baCXl}9}3v7r<=;-<%gB=TwSn#@Fj!UA>N
zhCI{>D1Biq<?~#CRFgx78GKXLFJB;ltQhM$G#PQ2lB1r)JajQq1X2?5qtr-^S~5_o
zp@ph=_k{hf#<oqReIZ0hxw1HcuCPJL+$gkjXeB};vm*Rte#;X1;1Th(4n|07p>Iem
zBoS^VZSfm(&9rG^xl|M}MJS$6h;(V)GpLxoCq;U|tAzlJqX^vzPCuKaBRLq91=*Jx
za=%F;LBb>ZyAy3a@o(w^R}OGrf~2VM4-(EMCV6^8OoM^aR8GD;c-o%<zvjG+zcPod
z{c7i`@|Nz2XepRVrGn}a$LG2-6bXktFcWn<1g3}u>y!dD4VV|qFFrqtrVx2&1Uhx0
z5ZjB#X_bNCXR<$7>Kk<FQ_d9^_VOf9NLJ3JVBwFOCQyUf8P<$L>G&3!BSKe*8Dp$e
zN7A}5c1wRyV;~t99j*Tv%-qKj+8`hi#UzMp)h+ojx~X`QP6S6IXsakx>C|Pzf;)=H
zli)f%`c(fDuv*$diONXv0D8p<6`wn`=wwW0s}d9_s_OK?gZ`T^MRwfBTg0ptB5N+l
zQ&2flwS5v8Ona@e#rE&a&`b~yh*Z+rV~TJj<B~z4n?T7?1FsG|U+9hMH9#NZ;|Fwl
z5}*qXzfK9Sa*9h<F5*HS&bl_0HNa$~JuQ)UBE_)4`sN8OJJnpqoW>RY6<If-cP3((
zQ^O2qgtm)=Eh4SDkkSwd3h-NoEJP}rBH|AGc?9<a>eoMU6w8YLLh2t<6?rzx$bZV0
zU;kXEhgyDVUvjgumjuf*^kQCV|9mJ({&K=?hI7ghAWw;shTCojj}3Z`<})#H!XAXL
zGh{-@4~kS&II8xdxD6~E+W~HZK4ES`b(~wjt;i9m`)bck5^@ntBJi3H+lchk1~fRj
z@v2MF1$?MRf9`}nlP#N(#ceE5Jh;~UIjxMH9|)2v>R_M|63V*>AirVZiPRlISqywL
zAr0ag12NSb0&kPkmp9{;>X}s<yzj3Jv`%c(q<i?8YL+t7*?*c6)nEjjmg_Im3nv~}
zSz>^K<g=y+7?CJ=smc`-0h$6vMUC#$4AB`oh1i0y^8Z86%YvyP2!)eD5nT93X?<Yx
zd00S^-Qd%^gFEjEVUcg}=0#}2+L|ox4oNrEnIUYfGBBC7g;AR3kcgAXD0~;o#Q&qo
zi;?9Kl#XefU@oo3$7AtoC<+?>7KOD6h;bKK{xA^4>ogE)hzreOaYDV`6hcW+NMYn+
zz7<>CCIO+Sf$Sdc@>8q-29-!*IQL4A(E)EWoG+_)8z&U}qMT&1iW7@OZyt>ctVOrg
zToHOcAQu<_p*z-U$dxMLQPT{VlQ%G3BX)_|<RNuP-Tn!f8o2@p9*0Iop0-f{;!i##
zoYJ;dT7obyGZFSX_kc&V*Kt7Mg7P3TlB&lw(yeCGV<>#sPDzLW%QJ$Po*ed4Ovs*}
zyhE`Mmy4<1JOq3Sf(yUY>KiVD1P2j&`|e3Fy9h$2;xSyLU+;1H9&vF$H0dtP*IUH$
zx`|AZR;s}XGo41jp2WFrOg8I}HC(CmpOTalxIDW+CJt;Y)TtbxA3QcC!ZZ^Sd@h^)
zg~>hv#<sYsqMe=6$SlWk3y`_!$O6&jpES`3XDQA|?eqhNn7BE`x|!BTL&t=Z_aAaa
zNHwEww#`f{S=?|NjNv$+6L_oeHVA#rgkg-zH6`9uZ6mB^ds6_ooH5Y4H+x$-UwE8K
zD{IbG4V(PLaD+7ZorjdSXut<W1eK_p3Wl+j%1Wo6ykr6JChCp3&m>=%oJ++n4P&K}
zycVpNnL8>hZZS<~k1!(G*v%JFPAKAT=5isdr-%6?LyMz4^d&a;tFj;wsj!^GKP!vH
z5R$1Qg13eLHlW~Q^MZyBEBGT=$~zIK2srarr`<4mhg@-_=+$?g)bJVN)o7&IAT;N!
zY%(xbRqniuS1P95RN<MHra=?iW@K(?nF@27>ST_HBb$BI9U)Rt=2$u4OHVO76;YQC
zcbFt&+l^Qp%{7+n&ITBVS;b8SD`rWntrUcB0R|gIcL<eQFmn(NtD70sq)Rl_+Oj$j
zb={Py!h=Y>;P{n>)EA*q6pZre04COzq~zx)1zZCl&D(*MlbLcdWX}X}8fM?S(W2#O
zPz){tZZcW#3uSTCX#Ncvs8)`RB_rvDKJBvfvjf@ot{7Ci#Uq8Oi)XMK$D)IE>ZG}T
zlBlRRos)KA)T+6c?FUoaH-Ky>37L{ek)6JD-N786f-_abU?_4{pU_l@gb#T);SFZO
z;QIYZk-;i6m8Z<n$J0}J$Dy*a)dR%6G-3+rx#mAS72CZ^{0KK_&xAS#d%C(E>ZHs)
zI!6(ehUr?AI#W%i7GIYHm^&?ncQdL%kx>{L>LF4G9Va$2hX6#M7I28_ZIJK%6Eqbd
zTUhetl7UW{eRAEHI}`ZYwRJrv5OezRH<j)YzZ~X~Tx+Cja0ipHgU@`@Nl+?!(*@z8
z!F`^!KG7W9uhNQNiyNAXU?|j?r7qDl8zuqVsa8s6m+g=FD9|ilek&U&NLLLvnU#S?
zSkb=wmGmZ~J68uU{$ABG9DlzS_@&P$a%0}ko2t!6wF#``{idP7!ma39RKN;O&Dl94
zQA;pt#UzgQb&AoPtzodC9~z%tt06N=k#Wg93{09UK|my1;5>URkyOz1;^h6J;56?1
zwvGs-$+!=mqxEeie$S9<pduBkT$`56W#gnyu!hXycJtZD_^jL>blM25*o3^X^?@sM
zv~1jdWZvoBn=nm;VIB`EP~arcstePu=M;LFbq-n8F1Iur!!_yqonIuVI-#iLt?7vL
z%TO*2r>N=+gr*MjhV==yy0VXxjqw=4O=%be(8VJ8x^Y?)t>F%DxX}D@Pd}?~Fr_+X
zlm@xbSCsMfQ^7s!XyJWI=ltk{9I?JmN~>X~Dxz&J&cb=;K+s0qVGTe7XTh{$7*U~;
z9v?9n^D-`S{zA{&>SHRftBP)$W=u;;{M9qplwDnHpxW11y?=xn4d)7M;%IcLbHrm0
zF#~SOqo|N=%3{vLLKRlMDfLo%9PS#&tqa&&<)*I@{ftV($+w}442M=(W#3y#XXd~7
zlify7Po;n>I%<{r$>zHJw^DIACbKo`)%(p9sF_5(FmRyB1$Mz<-Uw!MZ1l2n<c5CN
zC;g|(!A8}R4~fhVtna5%O4ku=P>N98w~wI)%-U!G@s2fw;g;<;_?SGD*}A}V)kmgE
zB=e*nY=OtmBysGI)k`(+GGR;&x29Sy6M#C5mjmM5U~thB-AeD1<AdF{wW&e$IjspI
zr)Yd>ux!o}o05D}lwfwJ{gPqu|4Es<95Wr1h#;iARtN9tZ`s?|X30LK^J~1wpQAC$
z<Jnu^K|bzl?KwbrKO`AAaN@i>cCkA<lf&OIIDhNx&qGjYqU)rkrS>MxUl;zbBzz+O
z!zlc?Jl_2^Kmwu%_c8OKec(ZTU^baSFn=%~X7=tVcF-`7tK%<ohULPQ8+RT&dGY4+
znC(VCTlfzR+TjuAg@im*YorPm@~8u+a1kO0aXwR{#bAiV6o(}q8^?E*AW>3nn1a$F
z+dNaHQJp@MMj~q6QmHlC(x5}N6Yj?KoATKwUmlspGwqC1c1p(8jOUybr=4}lc^6#t
zT&im>yJD9#?`(C`bvLAY>9rbs88T(dl4C%wJjAD<P=POnio76JqL@S}sl5it$&@Mg
z${YLbHtT>@|0dd<{*s)3q1)?+`9`&M_2$f5vd6Loi&m&impbf7SyS~D&0Dl=H4PuK
z<~;82Hy_N`9NxRXa+)R*GOw}1CE!(baKE=F_QioX6i4D%oQP9#76~fxs%?J{T)Y3F
z&lswbM;o^p2dnwVD<}$7Mry_JpmlW(gLe>(Xg}@M|2ql)lkfvNVegZWgg4d-m3T}N
zv*FoJgblo%{~MbP>P<)(ikjmpuL+u(XW5U^@f1#wCuLBX`R`F=s=HBU7I&b^EDxYF
zt9{g&dKZnT`=pwP;Zn@6hCmO$e-7K>9y?8dWxo>U$Nwz@SOcn^1V8X2{3^gG+rJTQ
T)@X9ci+<3aS@reXfBhT)8I@;R

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-light-italic.woff b/0.16/_static/fonts/FreightSans/freight-sans-light-italic.woff
new file mode 100644
index 0000000000000000000000000000000000000000..beda58d4e2189693806b79b00cc3c29575a308fc
GIT binary patch
literal 29304
zcmY&<b8u%((C#nU*tTtRqfIupZF6JWwr$(CZQHhe^M1E()pt+Tb7uNE-P6;5ot~Z<
zXBlB(01)8kv2y_+|JObu{9pOM{{R073oA+i06--_EY5%MZht-(5fl>s;Rb$W{vRm*
z;{7!#A}gZ+002V(08rxq0Nn$hj%kF5f|39L0JHF;`=bXexuDehETcfr1OUKc{Mf1g
zKv&FKy4=7@&*q17_|Xgf!JepKQJ#U50|5Yl@S_4?`414fFayRmCRRV(j|^1%GoFz`
z1AVcHp8bz4;>nNBkMI8hf{CT8@eg+c04Tcv0Og9RoWp~rMtX(-fT`(^4%>f#_pXFA
z{ULw2oFAF+2PAO+K*mk29DZy}e_{hx0swy1$hq+iT3Q?Y=vn;44>a+EHCV;H8Y?}Q
zAHSCW`6c)dz>vUm|MaYk0066>c>tFJ0N@$@Q*dlH*7iSfSvUOHz5MXk1@r@@Hg-lo
zwzd^NcD+BmL~o`Vw6l@^&z$XkcmUXcSQFPM|7n4K09d?Ac>J%OeP5yZuU(!E8+BR3
zjgWxqAwU?+n-E@)H}ih_dPnOJ{0r&ZgB=7?(PmZ@N1;wyKXXC1+qE7HLY~Y+(&BoN
zeCqkZ_X3y$2Z#9?$7O9ZP0dDO8FQFVqRujWkfzTPEENET*aRT7^b1%7Vc1!z&=-R;
zTPv7InM;)%xj{;jB_O$Q89<{DBV;ubfY1{i!zFEicPy4+6%z|Tg^6c#JM$;EKD)-*
z6bgq0w%ix~e%++rw7=%ue(k!>vadQ4j|zD)Y@><GP|4bE;J-5}amiL=EEf!#Dk(AK
zvR6i32ZcVauT*XEu`(LuZX&vEDJi$t*6x<mA`jKsLqvCPc9^KX^boVVd$$oh@R(B9
zz(#goCR^rwsee6hItO>ge>Qb)b<ccfl?`j*v)jITc~QQiynW$TShcS*2`Ars?XzcI
zPUIRXO{ONkf_M*UUf|NfS3CY}RTB<v-)LWduX^6hXxu8_>Xt*Z7rE0+VvW)~W{UdQ
zfzS|6U*e=%Ua70P-&j!Q4CAP)j3ODw?x%XItWr#BnL2b&C`MWb8SQJ%JfykhqMq)D
zzG*lJBh(%V2EO43KY_^1VMY-^zxfMK;g>Z7?e-V{1bT1={;eP20t41ZfB_@Z&&+^p
z+#{I+YC)(MOUUQzSc9h>kliP7!|Z~o?W@xRwgS-|Aaz640hYe&xB_h5WA;M)0r>^@
z9k6~w-;)1c;xR|8O)hjhK1Yq0FXDCD-(}sQ9k%)$&Mxl#H0m|Jtb!^xhwR30Ey_#G
z-oIQ-YKnI$HJRFoI+WjV5zlXCwHX>bdH%t~R!AcATUH^Dgv5Wos=n%$-IXQ9m>;}q
ziFBca24lm}`vXaoQ=LG2%CX`meYr?|j`6TU;BBA0Nh;GdZQFT|x4ltT-5dFlB_&CC
z{%xl)5N^flkzq4_Ovmx4k+oqnMh>gqLYjV3#*DpFaW&I{N1U(!PBDMeEARUevyz2u
zG4?dm@@l1??~DsgX0AHHLu&kJK0%{ZU+7@a_!Fl^qpMhGHr-B39Wu6t#=NYE?0Txf
z9$(2}x+(p{I4a$Gz(S1_MQ+c?DkfUfWw2No)}`k!j`+<40S(PS9&DXzS#2H^w05G=
z%1s+h@gpr#(I=J~DIzCncVATaij!E?2wrR0(`*yrJ875^c!Bmm+bgSz-~LL@HG5pI
z+FS7%8{cQB;=U;DL9JsO!?(7z@9I}NcGmLlD&{9(n}h&TSt@En2$(<u82@2CjtHMv
z1}NcvWCpOIJ`NY8p?(DhAhJG^6c_~p5CuJ~iK0qQSup!uwG{AsJ)nes0acLjT;ME!
z@e)7M8kD47J5?ZRLYQ)0)(T(q74cKH6*#UPsTIiXAgvngryh+O^sOLgUHpsSUp5GE
zy?PwLBDr8hHZW2BkQ|W3J201DU{ihWH@qKcQ@yS)059-H8zA0*?S7FPM$G`39V8E^
z<Xz?#V3l3_7vL|zvI8B)fghpH+>^>hb$zvo_ph!X>m4!==;j@+7xoYEu3&CTqpTIE
zbzQCtA9g)*#aMKpJ0d~tfYERSS41LufdiqG4v1KELd?O`{8c_616`Vk{136G{Z+m}
zFJQ?4F$T;cpE<=8F#8fuJQ|<C8VIz%q?h!*Ml~7t%65`!bz#@#tJS+(YvQZ_FV1!j
zd-mf?-)3F<4%q7+-W7Q89Ri%ajeZ-DC;?c(9So!4U670b1)JD5`rcw2OkRI#8>MY2
z{yjtzJiR?c0z5(wyn$o>|9J@qg+|oHW7Ze{32F7B+COKB%GCG9O=}l{)z?FSUJT=>
zf%&iFuZl6yw}ZzT68OIf0yMB%hW|~en*EQel}9wzz!ZTKSS@o%CjSM@m7M<z=D;+8
z7uYRxNH+fm%$~`KtqmWJkG~BBTt6ZQ6f1(Sfensyj~EAN^c5cKg!@8EwflmUSII_5
zEH>j&vfxokMW`Li*eu0#3Frh}pgHT1M{AOr8IRflR7cK{uf9t6^Jqvb_cq%QHb9I)
zySRI?M#Q`@h9cLJ#{3@MllgfqOx3uR%Ip`{*<_Rc7#MVW!z`8U)SR|b*r`>%(w07c
z_omyzlaMgi82j%e*s0f1$A*II29*lS*o3strtG;j`>B7C6l7`*s|_R8CMdUz64(wy
z%P0N3%x<vCD2d*aqQd)+m}UR=RO9L<ZJK)92RX*~;g2ha+|}0lFgG;0<kZXRJBqoj
zx2^cOrm-*PfBg<~>tK5RT_^Q+-)pso)?G2s%<vxcvf>gbPlTkJfM!AMU4bZ`9E9hE
zGeH#OWp<q1G?wqgA+W}YRQ=BwO+hC33}PQq`3?E?dV*+;sGOZu*;DK>=yJ7;_x0sF
zHac&rLbs{n?UGr<UpRC3tXp{4g4cb@Ds$qlI6CIc%bKqKab@4)kgH@#+azGJmawGL
zZ_mhfDAi>|QLSTwn=heA{ciX-dCyCc;bk*pms<BP%3-O}H$2ONHjM~%mT0WY*6yLS
z_e#?^Imb{&>_x_4m2>ZU62_tQGG>W<^+h*j`LMNCWvlO~A{^(X&{>#pB<FgoSkL9f
zS^ZAV@Y1C+srM-oN<9VK*m5OA&z#t1HiO4S-N~XOwxatHDygU>K2Pm&FYPg>&D?9R
z`F7(=FR4G%z3I1;+8;R`h|1rC=OtFsvzV9p1J5;4>8RZY){9~VqRZ&^ZdVl_mlsc5
z)X_ypUdJB=t>3%ZzIEx#o^y13OZNQs7JE%DLE}?HOKxbRR8CzNN4aPkkpY+ZyY18L
zZW}kvEmB_*lJG_r?&^4r@>|SgVAs#uQaX_hv*yjX+ZI+<h?N^W=KP=5N$Rwxjgn*E
zY>f6ik2cx5(9g5WcZ{QVaF30<5U1TDX0p}xgKFO^sfjfdD?w<W6CC6;VQbW_iuj-!
z+~!~I>KD%Hj$c5`w0E4}KA)=jNsrzHUrpnd@A!`c%^1^~=8^c)>3Zp&Vd9=~rlF69
z@AV@STE~=?Z1;Q%O<5A_<w8qtFyV?v!t*8?@2<>AoZ$`1L41OAYz<*TyG2>vKe@qZ
z8{6TPL&5%d{QfUKBlbh+bM5(UyvNQYVy(k!CP`ryt~*}3eg&`N5Q&XF#%%+{(RdF^
zg7n{_BA;mbxemEg|K>6?E(nF@m=Dji=QQ4GuBiv-rn#54vxFp8;9`ai3EugcRK<-#
z+|Hk}vJsy*9Uq6Ph7%rb&v{jmPK?V&8aCy0+-8>*@2CiL$JJ3wrV%28?ZfL%#vw2%
z^9>tXmF)8*j`rrOUgXu{{jJ*-;?0c+c&U4f3r!KMU}95UEGevDMOqga!Z;%N>mPnM
zOAZk>M~J*K0X@csj*!R8`brAp8bO`qY<HiKD#($i2K^qlR20QEdN8CrBcIOGPYo^_
z_gK&wo3&I6IT8iMqM6Q>FI~7cSIi}>14$f_XZf3y3si^Wn21^uULm;Y0tFinSstUK
zoPSjZ_84QHK&)Xz{<#^*ZHJIJN?4k!gLsf+U>DbwA&_|xk<ov2UsCNsfL`<T_Z0)0
z{_L5Bm%s4sv((4Sn~ETNob^*V&8?XztZh7bgjOf@+CsB<%B|HGRwtkP6|D7uY%Jb7
zR^neS(Pz=3!{?BS%9*HPb9xh!Txw9Vi9zGtj)yo}uCGf;m^SSodf|&MY^+Kru9)?+
z-6W73hg$bDYo@l=j`?D+xRqyOlUHDqJE^kHu>0d3yjpps>#4UB9<|#~F6$UGEiNF3
zFywu#$s8MlK;0@T8Xkj2S0E%9PUL7{u5Gz59Y$26p;@o|1j54)kH8}-Gh4=&zSVOS
zRZKB5=y8ASaX&fH3ovuz)(87Mk3k`gbMI7-o%5POtyqczNx%Hno3n{>0TxN5!H~*8
zz6lmBh|{{!#_fvr7_Ek%k^M+X{Vym#Jmhlxgv8L9qYX=5rOQSMTxg%gK$1dzl`4E1
z3tMp|cP!VvG`x9_c-is#!=Fk$%f%@Xz0$=aEu&P@qKK-CCJ8CAL7L)@Far#CYq8k{
zIT5bfk9M<?P<aNAIxH3cD}mie6|La+QIJNbVJhk!b6M6vm+I1IPsf%<ak(>XZJ7qP
z-uL0nxn~lSor-du-$vkhB>O}w#Gwm~_~3Ui;`C&=(MEbP@=%oC;WQ1Q^ES3O5K=a4
z<NvG&40MgPCrbKz#(MgGhcBl1`pV%6FzXnK8tCil>gwwn8R+Tm80xC^^wjYB_ztE6
zL2YcJpkQENx_H;u6n;t==<0S)7J#E<!NeJG4=O?wpkSy1IE-M5|66PPeAG4e(ZvK4
z21981b-?KNN&^G#(*WlW95eXK@Rn(T@xuhhxI*{$kb{V?1iv7^IKMQ%7{3(1POmtj
zwq9gj+*$TE&G)ze<Syu$Mh^yv(qs-n4nZ!#FC{BTl}QN2$&c^%@8a+15Bj(3=jhw(
zbZ~Nf99Ns0USDu`{0q05)7~-gN_d)U{MN^`N6v*n!C%5`A<I3MWCXo_RX)DHiIJK9
z@!{#g$<aE0WJOg46(uzV<i*v6m8G@$<;2v4l%%x&DoALEsK{*h%#5rIEKF?l&5f-M
zElq9p&W^4QE>3Rt&yTMUFHiFkFi>!iu+VT5F;V>{W256GWTZ^Z&985FKC1wgNT$)K
z^*B6hv{<U#Zqb`Rs$6frTxqo5-ueW`5TBc?zgD588}n3-?Vn){&Q0(otVg<N>-Zg{
z<esJg{paH24V;qpk~tcu(`$cfGrie~8Hac1=3#apogTBydfUMC+~V>&mBZ{V;FKAo
zXfmBIdG{q$Kc@(b01VJK8*cgE0uu163={wm1F!@5073w1fEFMJpbSU@$O7a5@&E;Z
z;*X>PQ2kk<YXU?85&!{EM3*%!xprv{euPa}_+P#QD+HWtDD|l^L+-htkOPa6`i5MA
zhY)7$cCq2MK=^-r1!32!F>`VYMAN|t(M3H3cnOi1Ekg2GxVY-=r&5_E#$wByPc5bu
zj!9zrHo+#@KBYw)>jBRN=jV|T4gTQjmc|cbj_+U4VCl07T;_m79&1W?Oerp3^@ifF
z*$=3o_YeGgE{q12`G<wykf($oN9nc_{@0k3ZR7B7ev7;M3=9P+MSFC7Me<+GEXU^K
zHI1>e9sbOSucIt~6fCSj(!Lf57w>IRBx1&kYS~g!>b^*jdjm%uloT2#Ue<cG>grE)
z{i!B>!B_f~U19Lx;K@+tnn}iJU=g46%LgD3<8h!k4>v}?3jVx`A`NwjiF54-HG&Pr
zF8AcH$<%*yr%PE`<MzF%;q*FpV@mvf83|g$K-A~br}Tjo6cAQ=4Z@o4qyamSf1m!7
z;B$Bz?VSnHaJ3@Ho!saM$EXb9=C!_YHy+lm+&UF5;A~xHUNT2H`K~?HHT)|J54xY<
zpf`Q`{Y%{Ujhm*5#S{+YuN%ADcV6RJNO=c3Zyou3ckO@0m7nSX(Dye0-eqm06Ixes
zv60>D$#vqLCpDFuJXJDo)ZZ{Jm8749&=;_a%McKPfr#izOh7;+3nD0}sEC4yn5gct
zaK;($gQioPsNT@jh*qI)QTe!)?zL1NsLpS9SJ^n7Itl`oyS#P1^^HH7%4`zHX;Izy
z<aXKU9juos2h%@=Sj32vZ$w4X2ZF7yu6@58Q-J8Tjj*rEVMGE)5ADKPv}l-N7J}G;
z*Yti7$lGp>l1`2uvw7#Cw|0sjwYROmL!HX`nH_5&fNEe^Aqj*$&n%fbaOxgv3LQ3K
zD`s$!cu^oZ$KVuf8pf2U6nnO2Gjb_D|8gH*7ZRMV5SOl^o|ddpXSls#Aa@-@xU~^x
zH#gO3%)t!Y)tv=xHz^P^cWEF<Msc_MFVDom0~67aUrc5U0|U;Z9ig(q4hbHzUxsIu
zLreP$a@8fZBLcckvQ<+W9%gRi`}*_#wbL(qRNpQYO4!gpj+z)Ji9@RTU;-{sP)fGW
zeb*0-4x=&`0rB214*s8_TA||`vEIuqXFtwYK%>{ot*uvaG%;^6{XTv0iZkA45NRsD
z<VQ@!tL`57<^|XLbj@AyH*c*<*CAg>RK+U2^`=zUKf+AfgI}!(-s%U#>V~+l>Pxke
z1<0l#{15jVV!K*J0j0*m4Al&bd^XB7FcOGW?8=g@FyzW;(gHX&?VJ=?(c!)^ist92
z;R5=lQf1^udBh}#YHmU%qF9jE=FITy9iOS0wuzW7jIw%&uXx&2NYZKpyF3Cadb2hr
zsTehmsNgOpn8OaYLcY7GH_;BafrAf{I9Zg7$!oa%j~Tj+CdRMS7Es)BjMH>&c`yk2
z&fLQY`Q*&>s?p1pPNRoWewt|^>-#W@H>``1Bsh11eOSrs`Ls|r+EL~Lf6k*(IOjVm
zhSd(HmpL|fWr<5mt8{-^GMKzh9j}UxKgjlu9VZC}M`E#h{k511xaMKjcXK~Q*RLsa
zR9v6B^c~$l&oZUhe)*V)>uY+idTNd+M_ZO`%?qJ5VEz>;#oPytpJn*A7;HHkGD<A*
zCjAzq%~WmnZIL!Db8Ii8tj(0ioj!8J)pfeeV%1V$el{{z(ItcFo4Wd4bQ3EJH&^3V
z1(L#W&qD?_M|n<V1Vw8kpIW3+0VUpgqYo!8?J^3HH-}J*weZklX#X6>mYx9nrUO%a
z)v@n13Lqb3D+R7PY<Dlja&cV0fE%rd6fO+Y=b)a51g0<lQ&@Q6(UP(oz`W&SDR_7k
zcO*jsi=m{U5Xnt^+{&v!al!qm_t44X?9lzb^K?99t#j9jzvXrLR=hIe?*5NYlzy{I
zwKaj7CBrMy{kiuU-qQQV8Qh7c?KOB?qwX}=MBD4;bGVuB<;oW~{I>51U;dNk^cuA;
zK$9#RlCeLvRgwn&k(rqW)wWaLK$a;_J}QT;=lTNmlT8f#FKqE+m|j_G+~5o~Yrlo(
z$`ywp%3WBHB8Qv_%~J|B4|N!~DUDw=MkXmwc^(?C->KH-h%nV3#)MpT)k<u&y#+fo
zI^GLmI(d{Dn<<v&;#dV$Sthqf+IVu!+tW;T9u0OAqx=1F(NAF4{b>^(wuidL=BV}8
zw+Ha#U^L0DqNa~3l{9Nr-m`kwc<iHzsDEv4v-t&IaaPRfnu!|pt^398%TaAzol~_p
zj&*8y^n7|`zPrg&FkKHlj__#8P7BIk`{pK_4U3lN36Y0SVT|^{TCV+{Fn?#fYRwkB
z4s##HQ|R#2<|nZ*d34Ai7n(SbgCtnhUY?ytSTEB1*oOK!Fo}x5h3qt$G~5bd-pJw7
ze{t;9-c+#!C%ZJDZs5{NT({XP>b^{-AG2w-p7)UY$~kW}CP)&S4ctiBCVEF$ufn@^
zX+HaWe7(Pm5522|b{mR*liR@S-0`pslLpgz!B@prSM2~J(jV8<_S|LcvRk(mm7b<a
zvBMYZYzbUbSdVBttx)TDeHu-}C%q2sppWZ}W{>LZO4;H-uP$Z?*Yx1trKX@wC^t(@
zo5muWB_InEcg1T$kwX#thLG<)*x>PNlz_g11q6}=f7-Z28(Wl@5R$UJ5^s*2nqj?4
z{Cl$NuMMhJ+oOv$(z=Rpiu|C)E!}<2tt6KON8Y_UYVz|o&)NEmofOiff+D~egfZ3@
z#0=)C7Qx^m&cPF702LOrPlO%!pv(RmW2j6K_HwZU&coZjy_HKizCNNK>0yG+xgpLr
zf~|4*Xvo~D^ZnzxD1CJ*B~<{dLZDfB8p9?bDQ5&P8Kz2m6iVD$w^z?>7S~)a4Qz&D
zZ(*^DcdoQ_lG<ve{bC{xn+SB(XrGGSojYP5LuQL4yB;$~<dp*s2jnY<NDPMZ{3yaD
zyM=7=MRR}D;9zn|`9nkN+hhXInE58hu;(v%tGPm-(tb44jv;$xc>76<dyuW$!RtkM
zA8$_Zt}fJ*-8^|FC&BdclGi6#B=qn1?jv4g7)dsTVeY8`O&4Z97aCl*3I?oTw~lgA
z0eLu=3Y>JsAYmYuc79vM$oH?Ow!IPv4yr1VFm;*GSiIPkHq}1g{IIpQAYtOEI*g9V
zcG_#U?je0=K?56D?G*ojL!!U=s1Nw3crL%_B1DZibrrOJvoPuLWiTa`=xYbqKxoAU
zJ|C+<#In$Xg9!mdNM7z~1&VoO**#~cN|U$QTE~*`+8)}5tIf5Ua(TBaEtb)j=(8CP
zm&S)tOwK?pjD`9mMiMmWtSP4!Vv|L-+;yijp|OIe*BUCk=AzxQx;)Lj4CfN4$&68C
zTA$+Dk_w+MH|kqh+)oN(^UFGD$DJ-W+5-p;d2SX-OnacUCI?&^Dm6KVfbSs!$WJsm
zQgYM-m5R2zenYB}*7Z=%5kqcJ#-4I+xI7H}a#7O<$sAc1%mwd2M=bek0zo~{3N%L+
z`?GB@;)c6Q)$Bb*pTTkXVD<&MfD1cw+Ytj+t0H>qDPPsZek4elCKvX~p+y3ZVb3S5
zf27%-AHH&dXI|YOD_p7X*9as_P7gE%rJk*9VEX--#KT+-b9mYAbQVushv#K>w8i;q
zmCV!h$!Ai>)0Ud=&G79$e=L)YtKH@0A~haKeBEL`5Vly~Oz+w12$ebinQc!GEcG|#
zRI&oaY>&UbE|v^_dEJUB**JZi1o`)#UCQlnAwcrbM@pO-_%TNhRkSn|TpTkGHfNr3
zebBGu#CCv*srP-(XHa5~P7#|QwI}{h-}R2HjkyCz+Oy2o`#rfALq#8p$LJb$ae|eG
z>#~>iGW^wt?R;aibnE%<d@#LtV|`)SZmiXLvza=~X<+Wrcv-S$bB^n3mB#SK%<PU0
zZ-eW$b?_`F^2K);7nJ=FV99_Bc|LccQ4gYTHxbquJK{I+{+=X48Wu-Ng&T}iZ_P=R
zggP`Ujl<r@*#JLEyU@a_pnWufxfRj+{GgediCQvP^bLQqwn<Pk8MmEo3DpRH()@|k
zQP>NK9!Wk>1j^rhbJCeW0(9VQ0u|QJR22si;l{|b_V!5Hb&;#1rHs|z(q2Igk^2=V
z_iyg4@StEvU#4V9n9)j;6!_^)&FiJpEd3lACA^#qt1f?%J1J0jL<02%9}zR1Mh#W+
zhXZJUrSSvS%b0p;Xv?PEbhn?Q`P|RLQaR~srNB~G7g?-&>&Fl0Md&ZDOtrqw46yOV
z=?#b#4tA5owsoHFyO(Bhd{Q^RF`sL74VY5&i{e9DOO1aeVim|kE3QFpV&%*QsVfS=
z8j`J)N~R~kJB>lUX@?PN7HuNdw@R^SQVMReGfFcf(3=sKAUiIUP|&5#yX_g{SwaJ?
z_mP1Wrt(~;@@2P<*1W~*@Tm<-b8uhIW@o<LjvZH-(`U6f&3#t$cAU17XBuvM4!kXG
zVGgnN7o>YP#=^37I88%;mPc=J92L3tZ>=>RekD3`u=iZv!tHc|ebP*+j9lsp(zHQY
zA^0<h-m9`|K_MfQ1Yhh+dOp%k6>;5zeml5-vOnrB7C(FUc9L^ouqzC)For8ZXF=jH
z?L5Q&-Gwc-y#3>HT$@W97o_SwNXSMBE=}l#fXh}2Cq3Cyf%sC2_U}9_jPvOjMnJuY
zR+bBfV^Fl;ClnmBauu-zGvqF8k3UN|Jh($pQX{u;Hd7@Y;<|{PW`$$7f9n$-*LC-+
z`$Ng==<_uCa%IhE;wHglsWsS4U3nt%s-_r}Ik<huMhco9{<_Ocf&Esh^|kKzco}+g
z*V98&<tFL5m)&}~%yrAZ`OanIz;Wm>9#Awslk&8djLa-#%F+oMX7u`|nU_GR@TP11
zg^p+I4RVBO!#~=9w|L(Ty#{k(We8=_?07xU&)NJh6mTKrlW5<wX!u8Plr9i$&k{bj
z@;{Fv(P}iI;(<FEZ6NUqe_IiKeb{sfb52<o;e;+3q3nsch(_RFM43$<v$w$}F~9YP
zk?H-9aN5kbn6oolPop|D;J#)U@3LB-yVopDHw6-p3fpyb-&DHlJeoh2b+{c;W80m^
zqm*J_Rj$(ClbGm2PzfmM15|taBXC2|1XX0~^ebYdgZCJooXl{@a3fY%!gnFS;1KA<
z$s{#6cnWCYU^CRTX6x;}Ov-xnKUWdK-jTUS%S(DZaP_Ig<1t&b#zXK#r(+W)TtpX*
z7;FBBW6i__@&_PkikazlleCk&{Yy!d?Ef?JsFx9PuoBLNOdi0XkqT&zSsynw%c#g-
zqyK|ctm03%*Z192m;EqlfKRYzX{QnV;Tim)%5~#N&S)#K`cHj$Btpc7gV;4A)tu0O
zC>0MF?O$=yXJJ#d&u3>7y`W@e<w2-2Y1Sq@i=|i*kZO+93_>a4OST2yJ7DHUEGN_9
z_XsNP-DAC6llwcL!8WShpQgeIC4ZyzP{jnfRkSp>b|@*BgJuH7+8HuVpJL7(ouD3z
zTu7I)<2*SzgJka0n=>1nhBl`APoUq)TDm=j8AuIlCXhUSkUWn?kmTk-T2H1QLhj+H
zL+HiT+MRB<TjysMwEg1vuC)(wnEhxvY<y!?Vy2*cyJ>%JSu-Wr&&f`8RyjV>W_ug_
zW?6|%RJjs4&1NKhIXN>@d$@ZCKr+G(J*rm_5{Pc}VZ*9w4$$2O<p^yxWygj}Ib+2c
zFN$2ven;4VOw1v&m15*O?3|M<mh!$fD}*Un)rK0l6>vk;+PSAxZFD3tiieVT3Ga{Z
zAvz7)EB7X9;gwqCMAQl>QuQ@pfbSj={))eHL<v^1I4+@u2zJ53sqHC$F;t-<%>c7k
zN3$SA@DngS+tJ(lwR{I1wjy8-VmfL#71>SnCE|?KkIl|9wa($T7I}5#nE%hs<xr+d
ztEI$V+ISeX@qVp5x%1|5NU7WV>4=e!0F{2Kd$_^%9PjlYJbYxm{HQ8znbTmfXn6!Z
zIxf%Iy6tYXON;%>qyI!ASO)iTObeT@%T>HHqup5I6UpqhvJ+m)y7TjuugXpAYc+cO
zOkBA`8pl3)yE_pl^A>hu@ww+)wddQEFifpVI06SBFGQc^H&`Bq9DWiD3p{b2T&M|u
zza&(>yL9$PXP;$=IxIr34MuMWm`1Yh4x#xhwyCE$3t(N1-#SEuCQ-=?$?1)gX3!Ss
zX<w+*ium0LkC+B~gN9+7!LYI;5XZ>9)(dDplpT<X0rZl}r6hi-+Z!u83Xk~57)(Bl
zH#!6~hijNLzrLAz`2<*ln8UrWPq1TAD&M+D;6WuMLaI(fTCdl?Cwy}$QSbTJWU9*n
z9MeEn?5PY?9n(b)`cO%VS+FVpL%G-LsRB%58A>287G*+gh&;M~{ZjR4P4bq@Q7kIg
znabCJ`yDyS%{_4~x%7(kRCuNPO;0xH(W=JsaNPzxEl%^-nd$i__R%)0*GbV-U~@jK
z+Uz%`H^$yEjd96Y>s*8Rfq?*5jMQ-e?uHXw5VRTQAEE&RGWX=D+4g;2EP3K+w9e?j
zKv7A#F>Yg_Pgro?&ezHQrX?MnuKoK>h0vQ|k?Y27ch#$S@BX67w|yCB>GU;ajIOVr
zr2oPC?uF7IL_WUoZUyp=L12LKZ(<N79W1;2A!Hcbh&-Xcv<l|;RVd#8=5<j7JEnB#
zKi!U96mYqM(afDX?1`k3+>F3dunHsA4!#^o8nD#YL*{`d>nrNJ_UZ2DVqEg9bb9Z%
zXCB~sIWM<nkF$}xrKZmBm3K(~EyB#egeg#}#5aya-O+YO;6=>Lq@u+X2?g<Qo45hW
zL&`$8drH}Vp*`Vv3+mD4mHotJ5cDSxzo-Njf%LTxiyv9ub33gkaoC2}o{wEGl+$|}
zo-#95oaG<kMe#Ka9wul4c~i`e<ASNIa^#o%6C$Y+@iSi11qxy7L~|q#+`(d(0uu^}
z${#JBU?sMnUMqT+o}Z0?nG3JuAw8#=N86HyH%10J6qj3ow9cicZv+w|9s3>N`b2BM
zx&rPF%MCxj29^_O;0>Fv@c73XGU`IqcnB70SUUd67%LMcCeF>zoHFa0N#!)fN#jf9
zR>G)DL3na9G*T)#jaO0L?K|B)bl7&MCW+g>r=rz)3v7GClqx(V-)Yszb-D;fwm1GZ
zP;GQ)RG#T^Wj6^~=amj0EOa>?gy0>9SOf3|2P!C2CYkRlTJ*<76)--KIh?M9ZWD3I
zmb#3g=U0X^mzm*KBhUQVL#)QW`Mo@NJ1WdnQKhw$CUm}xOWm?@{`+gT_HH6u8B%gv
z!4f>>_NU;dD~YV9>kitGBvnRd3&~Zb%s>hka)4AoT$?bWlkEazwm;HCOP+Pb5vi<N
z8s7=}szUD3oZ6ZYidmX0YlTqI!BQUBfxr<nVy?qgt9=WJRbi+rU;YdAk0;4PNr;aB
zE%;yA6W<W7jAYPK>c?We?MruvWfG@>4{KiA!+K_NI&#>JC+2#CYF6?i*a?@RfeYE=
z#;F2v8T~+p7{j7{)?OE>U;B+DlC#&SxfFAZeb)}QWD_lsE)l)9Il`oN2naU91vt?x
z`UGT+WAt@S6);p5>R^M0wt7fQ$b7f?9m@3=sf#9C1`<b)pWy9Sf=%ICE6q*T&|KZt
z&I0JIZW!S{uk6UjRsuFy>Vm>&i*IuK*3vk+X@ZtS^S{o>?RYW<Z2VEcXk><UN;+SQ
zlp0xkz0^($xw&$x!c7~%6nx21*E15V)RmI=Ej6B-^MJw)e76xId4hEwvnG1VfjL}8
z4HtS@!v|cYArU}u{!x9sKcn*GU=dK9^)NC@n+$MVAn0-On=mMoz?2jS?N<-e09Ha8
zVY-ny#QX-jj})F}j<y{>Dy-icy_4iPHD3C+89uhGjqjs9m^F5dzQueCpBz|`=l42{
zS>9ecl3acZF2<?5^%{$F!^qXZU@XGRjEw$WAYKf{-sYN>sJ<Jq;775{0|O(pEt%2^
zn1^Ai*Ym4S^tm~%`I1?inz(SkHXv|<>wbO^`|4r0pY^Uf{gzCDby@;Taza679;T&*
z^{b!VB8Q;{A!SjSr({bCB8XG#L1DqJERG6=8z_v}qrQh;fk=}&W0(!8QnLv$b@CEA
zL&uAg*{;{Q>aO=!-ReAz2R9kf>1e-=2d6VAzamRwUmgjT*`)P;U21y|{Q|!p#jkYP
z9e=!z-8hHy24eB|++|$=Gh|3tf;Ld*Ob43kH7z`vLrRL!rIR8|yX(c6U;wL!U4|kd
zj6CufWfO7KF-}jJoi}1K>A`8(LEK0ljYCv{N>-DDqU?wIi>5j2TO(0^%m2HdlP9-^
zD{nvxSth<C|N81@233ng#&lP6=6u@D3Ky}>AV-Z_*V`>=iJz8`62_Bg%e)bFG1<j+
zHUfs6p`ojog$$)5L@Kc4dN^hbQ7lHD2u47+B&dIqb(_*n=XPr6mFyXpvjiyA^GIXB
z7>;Adho<ofu*(iYQBQS7tFu(tqDP0f=o6<KVsg1tZQN>n_G~eqSw}{u4*q<Vt0LF)
zRDE-*QrOR=^aT+{KU?C?`+3+Yy>>Y7@x>h&G>3BXXtBA&jSTyW#6(ZVA}7Nu{bYZM
zbH3;KWELpYBDz?poUUqV0s;;6HdCU+`0!z?q%u>Lc%GX@a^9J@r60E&i}>&wu2ve0
z@Y$Zyx<93?nYpIU`mL2`xbx;`&#9?{)rYg%aN&w}ky;bH8!>#J4RhUvTAPlljI0=S
zD_Kb{y2JjHJ5&W_v-3nEjxmq+eFAH#X<A5y*26p1St2vF)kGF`X~lTzmjR*{4M-7|
zDwgV<+dJ{Bi_w2vO*>h~bACqge|y(zla<jc-2RPrY&RV=e0VGk9ph%ney#Y$y+Jbf
zf$E4af7?zb#3pmQc<LH9u%w)h!oXO!ad+5k2iQ8sW0PH|?nWjgm?Uv%Zq3C`G84ah
zGl^mDpEDe{nR&5AhML(7L6*^ATG0=LsLC(N9aCfcF<@<BAhUU6qG2f69kwn(>Af;{
zpRu{!p?0y_Q0FA(U4y%A3w$3TT7GtVxT{@SvQ~dhvHONafbW8Pc4ll(NkMkC1mGD&
zG7V`|=&=<4D-%@5j`ZTtrH3y@pz)uNspzqd77$Yy+}>8?_Km}vLyhW><m0oYFP!!p
zw4OCbwu&jL_ggJ0IK5>)bD$+4AKeio-A5X{I)hNvK!8rdzD)>%$7Z_v*UhzGxZ-b(
z&O6uD{g6&<?Ios8zR~Bkk?f?<f8=@H(X*I+>h(OHP3qziv7|dp$sNTwqT9E7(%|3#
za?~z8>BRT>UU=)83k1(^!xAwf;Lu}^M9m`opgHjEHjI72lj#}Ts{6hETFjV61oAI8
zQ~L7;ip*Hvz^lV3;wJ6J2bruX;P|qX&r{^1sOUus{+Z^}t$5G%5y1l6jf5XsJ6Ys;
z3<`rk&YuPzvYC>TE0;GBX_!AG&n$bAjjk)OapHoD;V`PJxCoG|@zVGZyo>){gsk74
z4eT8?quMIk`1ae#3fh4&i;WR!oa7Df!!5#_0c0*-?*ivlmYkAMhyKhxQq!~W!+d-A
zLfc7tTaIncS!5f^$#~l<rdO|J7_KEa%1q^>sYK2Nn2`qY(LoJ@81b(_<+d0UbwbJB
z#!Xx=5?i>s-vh$TegAxD>f+?JyRq~zPRL+Uz0AX{r->@=Pzyc)7MI1TM3%M~ZI2;K
z379DYn?YQcq*DFmkdsy7QKW{Unv~ZnH}t(@TL16}V~T^L+!3SOZLh|<NlNiZz8c4f
z59z2+6K*an4?u~655jXUrATfIH$P~_UrwK2z9ybJdu9K+Q=IHUI!zQkB4_p5qF_JV
z@61a$O>S}0c11#lw~y>hPXUn`R>6h1|E%bNERSYNSZ-vdEWuLD;JV9Ftg5MZ5T}g6
z@BJy$V2Mw9(GRNMHvziFwclXjk6<N$hbZVTVMea_n^E8ro`0<s<noUT?8gLXn+eb-
zm}|Z9bW8QNoQqxD^dVHB)lH|7LuYpWraoDr1`IYV2Qiy&#yoqFBQYPR*x2!vg@;eF
z%uTCdZbvxC!E=Jg;r$mzbF5RBA=%)9i$L;DqN6jP&0_B*cYQ(P)VMNl@+M4!jU?kQ
ze+h)Rlh2wG>L&uG#Ww=n0>{<Zk{YS4z^`R+Ua{d$nEX<sEq#67g@d@71lotNUfBt8
zlwWI;i!T)jRFqnlf)TQ<u(h*?7!lNS`~jzIV*VEJOWI~7WslRDrt%HE0Zy8B>C&Ne
z5F&qLIszE3;e!{X8IrjKFOT`*Wcq*gI_@I0=dbIsCWF|H(rwZ$36d5MU4<CAdpaoP
zlVNW$1Pa}NQ3Tq@uVN~l^@H+4RGa_0O;my(($;ll#bfByv+CmGlU3X=K4Q;P-}ms&
zie8AhCF7y2T6ekki~@JMzQiy-ISA>+(o^eHp5L1^v3Qo!YHjztod|6wdQ_L%_I~1h
z`{P;b?eREpx^Fyz(N$9+4#!1<*%U!mZLQcJu({ksXJtjQD&Bz%>V=QS#(+kKy^m`y
zm0EK!(<3NgY+{<AB=Kx3R$QSjFWhCn2iw^5S7oO%Nx@-qF$`0e4e#e@Uj~p<KS-@H
z^(Q2FsxNz4>8On^#Xq^VE}{&17L!^PI=|ikb*W|3`~hBr)1*!KuEtlH=C2zv4I7>J
zyZ(FpErvtiA(q{siZ0@DUavRA*M-h;GUe!w!~95h?W~oOhru2Wb|39R1K1hdYR@@B
ztQj^VS=hQiFG-<L=Fa|pMM6UUof6|G8KI89E6M<vs-w>3j;DE1w%9sqME{<u*PBmn
z;)<PVyZP(S+`;(>xzIbhp2r3aj;-D{(>$w}8-@9bKmPf<<Y*o~-;bq%xcxfdGWwl7
z;gACX46r%l@C3j8u0-ZjL{W>DMp5Z%<PkTL`h~<WlOK+^`IGS_l9;vR<#9m;`EXMF
z0Xw%V*CO`cn)Ts_9|b;mjDo>B$_-ga0TQr@W)}EBl!|MbvqLr`?sA6Ts4;B&aBaUV
z0)PH!`;3K6z6hnOcCncY%Mvo26k{@`)kt!5+kY7Jb}^bBr<HTEd>pC7C$u*9iAeH~
z;EtJ}R8Y~ivZZvbbu)H=L~tvHFK{*{mx{E(luRxva8`>sb$}Kwrp7fXnWDq;aY%<2
zbru%N*V4yfVD3B(Jv2w5niAxXCs8Ptu_Di4bh5IXdaA1HGWw{0R+z<myC-r=IWG#t
z_7ZoZ^V;{F5?R({jE8T<d*d8=;S+N&xT?d>Qn6%ip;bn^3hFqaT)gJ$&m*R&(;FRo
z6vx>PFJ*c0rZp$HgwCfftm=)Q@Z%4kA^vTZR|r1U4&g6Ej~5NBd0F?HHbrrV5iFY&
zGA)`(b)bP_Z;-j1Sgb=DhEo-8GD)4K$B=kt;I|@A^WGW``)v={WK!KOzBbnu-)@{c
zm4s-T@Yg+1l@xmIcLvDW@Y2_=$^JBVpuy3b^C-+S=ZFmia$v^?eT06Uh`C!k&Qk2}
znxQuBS{4c=y9QSW45~NN-#2!m%wtv<1fq^b+O<<;ZnfTuzuu}L#kH}u<AMy{OIn-D
zXLxYBs4Y)X`}s=|wC4om!}}WP%d~m&ZLCV9DlTI^vJ~`r0|vT2)^f!~arp9SZu~C6
z&Pf8#Z9u7sC3Ms|D=<-uXXw^`e(<$_|L)TcSoB}S@j~A*73YtiRAEOY<2a3G1qn-N
zXqEqz0rxD_0<(PTa4>R?h>}T?I*>lgeDeU;5#6K9G#!^oA7c_CG&)cDYUjl<+@Hr%
z-mWd4nvXTq%F?bjIuC2d%G}>#!_e(d-JZw4CX>fZc-Puo_Ukpik={L)w1C!A=Gvr(
z58>%Np29-zq4pyzyll8*mM+S%vi^;?dpQ|?J*>U|%c?irOx&ltPn$nqNl)5B)aW#K
zZ_X?3|6`-kOV*mgzMx(;d(f8#pXmaZVI>~52Ta*w*<wkGY1I%Az9?=FG4DEy5N1LZ
z+0A(yAFwBIcAnSkv~MlbQ8*w&mR5#gTM1o`!B+$-i2kCB-WJWmJ+#BI;;HMO-kVXB
zRz~NNoP>}!8psL@E|rDps2_|hEeNewzUI;gOMHY$owen?pP7VuIZg|GwDH!b)t6Dx
zb#=L}oD_Y{HP946nNhow)rDhTp6<&D4syZIGQ4bRs)E(qCM7$!o7W*!3bk&kgdwpE
zw*LqA`4c&t{FYy9c^@P8PmI)xub)bQz0ckZY0byc)#<d34tJUU2*p+H1oPxa{HrS}
zs%sZj(B<@LJSC4?4MjXA6)ppLoLaRi{4W>ieV@md%TOITbXuSAr3cZD*MyNr2F9pm
zSunef9I-H!R$(Nd-!y^}dkH9>bzY87v!~X_mn(v96heh|3MXIfE(%BNB2Ugr=3DX<
z<q_vIZZm|)_J8eoFwG<v1i|NOhni(N(kcka3XRU!p++cd4FsNse0%ssQKI<wLi-q@
z-Y~NkE3V&BSJ`y7#&jU<c<UoCl8498?yl{E1ejf6Opj{(O2paaj+<doo~MbD#A9=T
zB|~Co%q;?TtPZ>TDy6L@CunyWYl4CFm9J*4>W^K1sj?K+hqL~H(a5O+MV;3Gw`_2q
z4!433FQ;glDcNLcX&o@m_k1!lua*-N*y+Y7umVa~|Kp|DdeIo#%5&qW&!$QH-7XK@
zye=pL?~!hL*;|RA2@sME7{<T>58px9t%$nVAZlEa%<3ngJM}Y{`^A9?-7{+ZcA|-L
z=``wNMJW&ThMlQa$|Kys!NYSrF<Vb|(Tj=M%;)l26d|HdZ97L-Kh?SSbiJO`f6;jq
zR_T7_D^~Y!vk<x4dmYN?OL`)`RyVoXLw&n^Oml1P7^B9P037}Ptd$Lp=PqyphgQvP
zs^5HBtwOra2A#!nlSFTMgn-!DYrg323P;)~*C@Y3jlpET+!@HaDPK>5!9XP&{u1jc
z-fO56LYWqtHM>f9m2+f4j?m1eup&g`&moQYrdl~olmuOlI@6t~^lxPv%*wW{k_IeT
z5_<EFe}uT$eA}-d*X&=%Q^$NKF2>7k%SXJO?w7HxhQ~af?q6?xJloti$yuEu|2UoY
zyYibq;oP%%lRh`}VvvW1=}Q_0;}KDi1BOzL)cbA$tNOWuZPQRO$d9l}0%(Kal%3E+
zbW?svjPv=4c=fj!{?uY~wZelJEfEu;8=!`U(o<R_>`X!o!uy0wV<E)0n2ZPjn<;)y
zNxB0Dyoks{zOMIyvqNmECu#NDvAke*%{yJiCunTuYMqZ@6PYo-uG(u-Cup!?cD;;>
z1B=U^Yy(a9#Laly)lygg5=e>D9(=8sD#|i}2Dz^pP-Sq`yt3YIX_m>hgd#QA!)d~^
zNVDMz=tQ(=Mh+Q$33U{tWES+sL99bTy?`mdQ@`0P>3L8g{s2=VJs*!=*MD6z()WQv
zcP7bnk;TyP!pWq?&_B0lOf06x&=xo+pX9YxtJbU~IEv&@2SXSeS*L(Uow8EP&j#1Y
zwSQ~SPtAeO_+gmxZvdr&<HW?~=GK?gk{gppelz&6A*Cewrg0Tbr0@K*ayK7`N{Qz?
z*Hq=@r0`xrHxT_!{hD5J)`baF^?x3=xPi-~$wjE8piqyNNF_7>s0GV5IZXtnkh~x|
zjD(9`uCB<WV0E!bqmDK%7hjX~pMZ*DOP2Rz7`+{6Z+-1+b041WwmN%06J|3Bx4`eJ
zw3(-6UY68y?(gV#*2X%wgk+;TF}1X~OM=1HY29m5%Qu-y`?+G3>m~L+=1J`PUH{H!
zoV~*?ApIHcn}L=5@>)|63?lMVUQa_><j|tXGS;AFDIMHBp<&rCs(U>RWerUuS<i2<
z52s0yQ4X)%a6uthKc7o?Tp5oX#_&iYA%Pu$q$y^kMt-b-vnEnA3q@s-pEIkq>Wfrh
z%Q{OZV0EAhOoKT3qepwh+f(f#&9Po@9BFLdLMyq(diT{><(^jEsM+<BcK$4QRKpnq
z?fG!~(q(Wz#||=-G@1L>ig363?}~EpSocz=Au#cdulKHdHBR%(87ASovf;z}=g?@u
zo%ho=y^P!!pVyvNp`X>zSU2e7YS>rDxM6>gGuKZtDXAX^d4WltQsK0iwc9$GH9%$`
zX5=S2#{`sRW_Hv))8%o-H2%T4uDvq56?tWHnc3-!d2!R#&k|+K$2lKL7jhOhaS6~S
zsYHyquSV=px7&+G?-XWeH^6_@HQsjLZz-PY7z~_lAiiFH9`B=R+1_CVw!6aeHPVgL
zT`NtOuR!E}gP-T?6yA~4CRW{E63#O1jH<{cgvad`$l`!H`G~pCc8!M4FgqTb?(lzH
zDkkKm1a||kMvSpmBL*z(;%3-{JKe2tF$>NyNo?I*GkXzji+p8%C~6BhjVaj?x~l&R
zzJ?zoP55>?vLh=JS~@4VL$Ts7WVz0P2Eo|lP=YMTu8@%wF6=7^&iUCf&Owh3yWF=_
z<P)saA~|^^%E6POGE$EK+oFqBo>1#9D>r~2JDFs?Tvs0&V-^P8a`j(#JfcA7WIn6J
zh4$%U*ZT<kqcgY0&rP)360X5j!dN@}##>XzCFyBoIe>Hsyq|Bfg<bsum905ch1kTH
zI3X!a10n7Ot4=i<0vb}5qiTmU5uX4K6wOL<v|3=d1B@SMx8p~QGEON%oTLB{p1_C`
zFb$MAxx^d|STA)Tm7}4Zr5O;z_D&S{fkyTQ*J^RxWur!mtcIRsgarlWo%~f!9J_zr
z-SNEEafFIYR$)5T?sPpKj<3qdf~58dc+DAdc>Nfe5*e;kGsd&ZWa(1#vQRY@wn#B~
zl|4V))w{D!BWqUIXlVUQZ-LeQoweqwl5(mVJ$f&OsWpGYAm%CYAhO9|0>2*ZNYYf@
zyL~W1dO@XE-j_X=GXU`B=7%At@Maknau=oYRqaC$GZT-&O%vA*CMQXfEe}M+pxHg3
z_@ETIO#t#v5$Avu7K+vyqb$w^z-3nU^1!o>K!P*n1^p9HC?jc*N^_&InxclMeWZ-I
zrQKYwcbOi`oz1m)y7|ar{hQzIuA^<9$<dXnNUR``K$*F#N5bJDU+(DOM|n@*yg+g*
zx2nOB{25}USHx70QOg~YMi%6U6z;Ypy-YQhu$ekvGL&e5q3{a{Zk03_i7N@G{~lI2
zeihhwepk}ImFE;QRc)1XYlB;-t2?KLLRM2=y=WoA_89J;N9e-wE-waJe|KHR?PNL>
zSK3g)u?Q5zhrubl1pL|enXiH4P5Gsi0rN+rQ0R6+H}_5Y7+>9o>fy|oDHmR_bib;t
z@^|&fx&YSZv{ufAfKnJJ+nTSXY%LTFx7NY0#v2YE%{y&hwP8Q_O&=~0o8mt{?m`^%
zJ>Hoba#_c?Odws}-h9)Yqt)QyN`o1O`f3q#ZCK;{_tN$2b|K(n0xaUBJ^E>zMUUco
zj@Dnv*_An9%`h|;+R=tGe57vf1TvR)E|)tU87?Sob~lGM{q(#BW~ON-=XZyRy~D2$
zC&k0^KUt$&T=sZ$*jI@vR%CDUuDeMBX4P|&--xW_ZAPOX4EvFb#Q0PfPP5Gw4T3jV
zLA`!!=lF!Gsgha`TjD_2AANW4M&Fh_(gUBR?U7cS{rL6249NP=GHtJeprBtzv-#C2
z2l%dolkDj2Rqg!s=6{8H3($JyQo4?!Fz=0#q<q^Ao;$$m>~GWCxmZQEjG5DhmW+^G
zQ#%rF94qBE(VNnY_D5r!@l}aYZ8$#?q-65GUuwSp^8UT?ecb4m?EtAIqB-=}XGYoP
z9}(~Xr{!3AkQJ4L7FTX~>&;Ph>1ZFBv;;M(R@{$Z{dS?2V`qARkVk9EZZn%#PL^ii
z7$2a~P;pDbMjs>bPp`s0fw_99D4?^J>e@u4rlIN=s$s!0=QIu1Ck|Z@{_wT25;T(q
z?Xtm&sAo~P{2FrIlY0g+{cPS$O4M`D`c5SsM$EK5FMMx)WwU<wP>ssmsEV(yH&UJK
z&&pb9t!U}m@1Met#H@RF59BQWr|$SXAoc8Msiw(p@Zs>9xv4?E)oHG{byBXZy&v;^
zUp9(dRJ7mW5?gW9bT^5zuS<wbEhu%^yTwV-1Vr0!h?88m*{R=D_c?i@+(d6=5MBvR
z-ShfHr)(hcSBv2{3D4#F=pZVMhB}JH&tY1(x*LTpL<82}ZGe*q-P(TS+Z{OiN)hZF
zb)+6Q+kyB1-5dSQvs!3$DOEf*IpkFYEk3WW4oGUEWg+4x?5eCzx5o?4a#yX*Gt<wy
zk+56)>MW3MId%sj!@&br=%K-?m4?%>RBU#?NN8r5ZUNnpNWBb1oq`L+9uapajBfbX
z%igTSnkX7d!tXfmi=1U&;&jXUJ$lQ&piIY?r0+N#u<oJBot*lC>DAztL6Qnza^wT=
zbZN_q-Lo$3B$(y#cJUEO?XL2+{IrxIeB9xytMvY>%XGf06PH&K#@Ubeln=-kWxmbj
z1mtbhz+7N9FpCq3yki2(<C5&llNdg|>{E;VtKR&$VfM%B-7Md0%-)H>3nfdaJ%OeZ
zcyaB89@mw3cm);npn>YuAj~)vhS@p-GO7O)h%tB04RG3EliK05+SDt}$MRq^D4%5<
zC||%HGmoDD8YgNIw*>OTXw$^WSD`XC*rnhV@Ikvns`R1C`V|C|%sgKM{S9ydkvJ9!
z2`HJP3PF)bX{meWJdj0lgSG}c)4V#50zy%tgTJ-#O(k3JjV@`MG<mK;7&&cb#jvZd
zjSi`ue)Hr#50va&TQjpIt6*|nogO~BvSQerYoq10uWy_xxs4XDYe-tPwU%92*m&jD
z5qw|jsd9w)t12odO|LI@j9gJTtv=FGGDKi=hBRMN*>FvBiPn30xb6Dt*`sFVPN~n2
z2FkNCc=&WAema7Fy5^4I>|uvLE7zH2PS?6gS(@1!o*aTJ-n2}MN9$sX(+%#zijg26
zQqFJS$^{3+qb{bFp__TJs^HW^nif*__R{*llD!GWUNDc?TqD_xG-bm^p+q6QFv+Kq
z@<3y8abtiiFU|E97khI{`8RV)OLKC|%X9J7i+cb$W^MmR!XY6C;`KbHhFQnNO3+Fb
zKulaj3mrkp&G(AcK`96}-W!y7UTN5a;Kk`?7V%}GQ?XrYQv_&Gx1N4LJNPn(fZ#1|
z-dj)<s;2dWU^PSmp{Q75AFkIsoSD9SCQudMu~iUO8xXNE)?vdvSp^Wdu?7Bo-eF6w
zEvVjaZ5OFIwC(HI!iBM~w;ht-?20}{vJN~>vL1U({`Bz!@~5j`+jsNjnbVi7xo_v%
z>!(?!-SRNI{y9?9z3|}I+iv~3dto=JCRfOZ<WCPAAXz_vpT{5XyNh)Ex?xQ8(K~PZ
z!H6mK$~@|1HS7y?Z%+U_*-dmivw(|>qBAVz!dZS0S6bu6^=$V*!TjKfY9+<-(QXhM
zgdbM$b3R)a=fiH0e4pA4Vzv3=-5_w9pXBIrajy`@8ul22&_7WKCUM80L8*GxdLp%n
zjr3Yh18j;Ub$B$9d<)lk`n174dBMyrj$3YAzcy!hTBIewaV2*(b}VY1+AwcR`OJH-
zY|LO^y!mcT)q?BBH%!>RLw<e1cwJyh*~BGnLq{+Ev!^K1fV>#SMEF(QX0S|FwE9F#
zTtpjx9cE>-Epm{|%->>RbF6YDndv8o$CC{+<QOa8TT3<%oMXPuyw2@omoft4dMLX$
z;|L<WegKrgLYwMUz!1NEwuY;1V|}gibTUK9EB@;^9{iv=Pyw^>5hhlPvmhTVr8)u1
z=~8m6yb>uZ_d!Nd8cIxu$XU|?mr;UXI?F?1(1t;%R0Qve22GBkh}sjhI}TdhUVcak
zTFoLmWHMr=142ngDt9Z51eJEFWC)njT1U6u!DJR)%uE4~pzg=@b6Pt`6U8!e!Kw1m
zTU&3xWb})j&pg9!xqSXxYcDua&V0ZgTRDH?{7m^6NjJ!UBW)V4ZR{S}qZQysaW?*L
zkcaPVd7J~#Lolxnq>*Dfe$*t|Ps@_ooPk>_WYPW7EsUs#NK&9XYB@pKEe&5{ax*=^
zaTDE2+pVGRXm_+DzMGbjT-b)Lp-tZBLvkXspS)qw9Z%l4@b=-0rgm;v*xtE;J9)#i
zH!ZyJM;n*6&AR!DsaM^iU>h(I?h$S)&2vU7qoSRGqA@H9kb&q!F<w9g;VpVdO!#<3
zk|siy<LI`{mR>9X*y*^Nh|fck0B}I7Lz*)`p)fd0H1Zv7k@4fn)2*aKeyUY|okUU3
zO>7YI=TVH8sZ*_xU7$O~>8{({pr}7CrP+F=JYYkb9ybkpkqxG7#1FAp(-i%QG-?JT
zZy<VzNjLP~j(O1<_w1I%SEXIM_L?28>d3tI`-YO2b!@P;bWzikq0No?O=HJ4PFye}
zFnV<Kl3|NWM=A5KjP<iCfUW}U3$@eDg+gyE4Sh&H6`E)aN+z5wg|rJg*Vfx*$<^pU
zU=UdpVFGMaKq@r31$0v+E{OpB&I~1{9e}Ec6bzzWlZXt;a^@}TT)S$PH)J4VXSB>-
z;9jzB)~$CfzH+6ww&^x@PV=ZsCykrH*O%Totf9WX`Lb#43tfiJORl-uPsc#E@wwa#
z+E<vslro(MQwqCO$u`oyUS6eiG$If7%6@JR_Y{-Cj8S$cD6K_0+$2QkssVg!e6vLc
z_)9h<X3(Wp>}{~w=#~pcB03nhBb^Q#bb|s~G_o~t>aU?TJz(G4Uez`$Ty|-Db$fg8
z?uk>kG{}BoYu}cptD-p>HFKKT^=CRZP8wTx8QpL8RR4PJQ(k5)OgST(g54SwA<9a%
z1uNq=tdQ#!C_zbMQWv`E_CBAom{Ram<jR|NOd{jt&)PNzI^LK2xmlOoHfv<w=?(1t
zx|gU9%p_ZxFW6j|v-xz!a{Au$10f~n|1;wV0pwT$o&dgYW8YWgGp2;Gp#{L%CeWSz
zMzjES>K81g-WYG8Q;yM>(H-^Jdrn)iv;=FI7FscrFg2oDggT_HDGV!0vyr{^zWde@
zw(Zze>yPxEZsI179aUrY{puO+(up%S(q3SgZw8?r{L@k<;f>tAF^<k|j)l9byS>b1
zuU@J2oy6gS1O+bWz~6e6>wJa8ai=J35Nks;A%^xR@`73*P*H^9RcDGF)N3T}$0pp(
zzc{M?%t`hTwL`mDp4%+g=-ON}qclFJ^K^#U`GHkc_rR*E`@g!Xx@-Q+tL}Sg!K>Hr
zJUsILbyq)h$<DQN9-hi|tbAkLOZUOy<=yM|E&I{zwfnC9krF3mF;Tva+oHq<O@!1C
zhzq!g@|tBKZ_E4h$*oo!J@b>Tmgo%9%915D^7`@e#({J2%l)V2cZGKZL`~5%Y2XKB
zwO7yV4u_>Q0o2wR3dPb~D2X(?9#I+TULX-BC%On7g>Z;zb@&Tf4g7*uL0z!BDXK=Y
z)I_0z$0Y!X2dU|krKYQZSnHI;>vf{W243aV;l{`R_JtDAIc%bbZn-h_2wi5o34%Er
z9^r#7r^AD7k$B{oL%c?-)0;erT9BR>^d_^->GmYofh334?uZ%ET(rg4l}1%r6zrlA
zJHN0BPnANi00(LP5d;p>2<tmfZ+=lx6cT3n_Qe~Fo69DY%gZ`QOXugp=cgUQQ;Pna
zKFL}@&)<Bri9AjF=uXQYBzRoR{FsR|5DVc&5dXM?QkI}V^`vB{6WHmsQPiIQB0HIG
z${TeG09z7WgIa8X&d4gnphxuDr9d%qr3m3iIPmDA+umhN&+t;0HozQ`(zy+0lApS|
zq3G%sUFEuV-8FZ{UEM4CLHs)PRcq<u#`F5UxQ>pZz35?x9Y1Gc)u^LGYCuQBprgSM
z=xDeac^1Y>c2!8(F`6EZxp3lU3ccOhqRLD&%_q8TR(fhv<4VJE$q$~=;WvP%%rQ~G
z$12<*<sN7@PiSKmHX@3ZHn9w}aY!ZBD}#8154u8CweYqEj~pYF)nb-WHY6CTsU1+h
zsvJ<hl5jOuz9P)iN~4k$q|gOn4X*x*HoKS?6^SF}@Z`{16SAGV7~h;qW!Z1{*V|HR
zGS_x~H{Q21NxQM<{1b?k(jZnUWtK3pY`R|%oN+$#y;!dxv*3O~3*9e>tqb5#z-V?A
zS`ddc7qac#pj2p7Fj{h9Iv&&rc7rEd>G+j0GC>P14k@21X0ZoNswo<WT0$cVe4uQI
zH*#xfYM$i!qGgE2G9A%aR$sq*$wSxG)mny?Jn|BGDqasK{^8fORK)REE?1CRuDuzN
zTKK|vk(|zDiV>puh+>3TE*2%knn<v0|3Z`yJUa_1ME-2~Ytdx55Vjn{CYA2}|2yao
zfIgo^cN&fIk_rDQ4fqE0+kv%{i=o>Yr(3jyu(j?xkPf$kIckj#9zvnI=jLqopCsk;
z9vZfXmrtHE7P=ir8P9ypOgf$`nY%A$JmSzGcC;ZfErl(dm5|C+-%To#V<^TdBVd9x
z+9g|qvJ^cq$g|eFIAy)->@k1dyt#1N@Ro~Q|ME6RYinE=TmVaE-@C_x(O|Uw7*^Gy
z)4(4f4|6p0DGw2y<LQgyqBkfy!cw}v7ZL%w<-m9x3w+ppiSb*OvA2uMLVO(FLbswJ
zgY&<e?zn8q0ypZAl<tiV)*@r-qT*Tn{drMiNl9Z7SyrZ$7|W7bLzXIK#-ws%V$QTN
zcFLUx<2#=%v`0X+h9Lq%SZ%DWXnAG(1t^ruO*<uV`-L9ETI|?2r{aOdCZ^Fn>JX5D
zP4BX%;g$s_9>MLI(y*(_WOc?{x}_`^xIB%PP?zU*psUl9bQ)3k30+)IR8kcCkW@Da
zut)1ge(#qX=3n;F084I|K6=ch(_2PQ=Z|H~s&8Mls(r#z!#DLRQ~uAp^=*^t>Mw1h
zc<J2?qj^kVAX0WP_Ydre8U%b}v6)a+>|x2GMeXPoba(;l4QNZ^7Nqfm6Mv_1mx&_b
zm>{yUo<fz79M)69Sy(s-qG2{Uux$sApd5}0@*JY$n9O!ZQaDOgVgMt@=k)1`&pDC%
zfOA>;ZkWU7*eO<yrO!WJ&XzwQSCK<I5jW?n*1nUi>X5%w>H{nKzM^<FLzr7s>`KH&
zLXgn}Xbs3(%7+xMSPePLlQ+HNW;-urdxc(%qzZkIcOlkyA*3-~nOVi9N{EOpgHmxm
zXtcKwEJqf^j3u}<n0c{COJzifSOW1(F;c{y4|>UP8zhYfXa@Phhi6iOFfFHPBp**J
zcV&{II3`me_zlGNzaZL?El3Ll|5~&ooag5t9?46V9jQye9%()TAE(gT^gl`~!XTeR
z>x-m)9Xe%|SYee?gGuJC^TPC|1rxg%{&n=~RsVY~e>dPRSk(g$B3$vmD(&-ub{FHr
z_4RL~9&3Hhu#}<4K!V~0n_#jS>14qs<!mE`4*sr#liXM+&>CSjW!oi<f%=I-s+7`E
zoICiv=lY7cruois6codR>;pOC`e#Q9lq6jWlq4*$>QewELAD6<pNGW7E(+LIqd<g*
zona%&3IUYrmZR!_!%(eP@6QF*`;WQE`=`D+OX$s2ydP{>!yCSsh$;tk7x(e)p!YN~
zQ&Zg2l<J=53k=vu2P`B8tZFCHFltaln52q80n)M`bdjSdGPBR224UljVFU~XuE%Sd
zDrvkCwAM&bdy=5H5N1M!5<MT3l|7{{A*pfYDa{)QG*aqKGt8kEsZaHyPqoDNc2cz7
zOqfB0v2ZYhtoR@jT$Et55Sk-b*?H`ug>!W?PLG0jQ6XklU?I(BIEB5@9K1k=$XYns
z(vOJgHd|MgEfd~)@yI>Lau%yCGs~OoNu^9IN9yQPJDfGiaCtr_mjKM;d@f(}^Q*pm
zmRKI9sCn|MtE?aIIPp9M&YJ@+bv4^Lk)6K&^wYcLe;i%GcJ}>E#n9{fY9h6Nh^quK
zyq=lK991m{0xJ#RDv6OG?!pj<1tk-P>{q5jS(^1PL0OV`T!Apu(DvsDo(_-F-Hq0I
zcr`hO=bx{sZAf`Vu(B!?u2!In2e6t*>cBr4!imgD74;`Icq9e$hUU!|AN41Ip@eON
z!+w}06ekMiF5*NLpv6dm6jhiCDXK_Cis~{8KBdF{;=!VDE?*EZicE*R`$DKue5I0$
z|IdsAS5qzwd^H74=E@6=g#DRfPR|-k@p;Bll4>laNn<Ja=f^?~?1kmB7|id!#s>uX
z<i(9i1<i<$YbG7nD%`<-&bY+Zpi~Sm?LmAv1QD-KHOkdVqntQ+lu~gkocS)r9v-_?
z)OIntn0jJfm<?_JRFCRK&ejWZ^!>ADF!li`e4S7ETFTTiOH*;SIw%&0r3&zS{!knc
zL}e_l`xhDMN<ThyqY6%|2H1@&{zX1WH_h2;*fmv#!N@-~+f#bS3p4ngvz_OKn8RIn
zwqqsC3?{-oMW6S=MDzsy_vi_Px--afc}ERdPFB^(+sX18d564X30X$Z=r#GSTKpBN
zw+s7M3(JLVOg0l>LWrYE_wtkiclE6D7!x9fX}*vB@HsXqh-)>#x@z2_inXrns2Ie)
zA*01jGSYp;nDwcvmi|(^<2k}x?CHk5kkVr2!~~wQ)DsDF6yY8D8VG5u#F=Q9b*ezn
zMS+E{PoFivn+Q8!zWqS^k&amly5(=~Idm_XT3Tvv9mD?Vrbm&VN7p}$+?~|2jWCn{
zwB>GM{l(hZ@AfmTf82Dp{Kc=<&t6<x!yaD#lTqA{rvo$l{x}^uyYwfcXk1Bgy1*v6
zAgX%cd`MlH7eMOD`j;Ygbs3y)WhdPQz`D?~r2_1l3&C?r)Gkdq%^{P={K)@~F?Yk5
zJsHZg)Wq})4;?*JilMVn8pfv_Jpa>pTt6`Wob$`vdx7!i{>x>aaWOKFyR%espPz7G
z-@*Vjw^evBA9#?<l;JvKDlVWm&IwC?Fs|ufT!);8C2pl8DrVVen~Q34<BTf*moX~K
zjr}Kvi-569ZJJZ@Z2#G5a7xw*DQc&zg1rWMq+?7e(bvog^fk+OLto2+ERiH5u&q|J
z)d3tfZiyOZsk!od6erKc+@WTQrh#g=(M)R=Mo7Lm7d@q0!Q3fue=$$AtMHbdv*1Su
zBmZ`p>Ct6UfS8SDS0d9w2vSLFS9br0wNxB@VL<EjxC>O&(T?&rj=n1&l|PA9HO2ly
zSPD7%??-mMe}ZGbBF!9k<k@u(E7&^ypK&h>cQbX&AJmc6g_V^=a98+<OGMX+(7iqC
zqqn3w1w&>CZA7C_YoW2R3t|WQq(a-kv$&M)zzO7?I-3|h8+j)+(oR%Gs8G+Sx(=y>
zap&gYvm@+X+H8LW&0Dpj%jznrqEnsqgW+>z<mL^*PdGd7JC8t8RuxU)@%WSlb5~j%
zIWw`nm$qoD&udbUdP+tU=Y|8hq56hUZu#iRqsyP#(fMp`$(Rl8v)2r}v}0IqWqm_s
z?uwbmD?XqYJ};8+-j$(_hMMy7+T1{E<Aj!3;hT1iC?7jCvh2#*t0qjETURzVR8bxp
zQ+CDm|Mty38Z#hH8$LrXhbe-XY^SpOM@$P#J`UFjVkV+&;p^rKvrMR3n#8g$wj~TX
z8<4F)X~8uqEw}}sQw3Q3DlAB};1&$D;1(1nCGdqlY)|u~d+B3{2D)z*ytU-+TX#gO
zIU8ZE0D<}xjJ2#4D#F^+{!V>voqo#5LC9)-f7%8K8)B=i-#vZYjDtUv!<Avu#5E^j
zs&U+tYL0{aS^xhB{-EW#NdAxrHAfd<5!<9#D1}F5Dvv5ImPf$^k1GH7=#c7+FOpAN
z67}YbVpSr)M1HLjbd+Cf(N~|vuM&*bF9X+|f^ppy?b;GJtw4LYF2Td~|J^*4D%#;J
z&<WP>GP^Y7(tn1lDfdo#FOta%|EXeAVSCdUb&F-v_N3$uMfBf=<TX{zUW0W(n!NH<
zW=(>F%B+%UVBSARYSpX)aajk^RmFLfepp;!qSr>zABGs|t`2NDiWqyP#($L<rGOWt
zsLwL^P>NMU;EmSgp?e-9cFCDWpZoS5Criqh>>{Q5MgYS!+&q)#4Fn)eyO2cw&p81b
zullZis@<tWl)4LB9<d@Qg@j&gtSA){F@-7;e0oyQ-HY1_@CnC5+ykM+cNa-<vPf!z
zkWQmWN=nNsP%J~fT`bbj5L^zF4JlThu6A)Ua-fVrM9ItQ7b&3+zxT@x8-7_EUv9sU
zd>(yb)v6~}9Hq<e9hWe1J10EFz08y_VVE&s=K?4lUzlzu&JM+l*tBgAOBxVXU8uX-
z6EK2|s&N-L26F7cjKPP4m)hu#bha`Sk}|Mp0RdlzjqZX<OQTaq3YB69nK7U*$AC=C
zuuD8{_^oz`7S#JLdv-1|Urcj|ZhY={ImBxjjJJTUZnJhV9F?j)v~5Gw*t>Lki_*>?
zu?b1Mjz11@FwkTmnej!Dg2FiPq>tTT$(+Ao+kMUl_FZ?cedJ3^Z|f@STh2doKixFE
z`AD$+m)ES^)4pWZlozJC4SgT6+f-P}kH~n(m<_vt=X2-GT$~I~xoPRz>+e}GarUy6
zQ|6d2QIRSM)6w@8#YQb>Zf9a1<aJqi5Iib&ec)EA!l0Cnx?d3oOU3mUin;0Tjpb)y
z2SC18Uq<Ug`G^JNNiU=^ZI%M2h);}^vht${oGTR+Mez}&QZa%j*zHmpZl^IgE`Yas
zP9s1g=dU2CAXv5Au3Gn+(gt8Eo*E%LyUu}RJ*yAEkUMfn=BJ7dQY=#XjA*R4FSrmI
zX<;ff(jpa&^bkcOrPcTEj7N&$>bGH%PQjYbd2vbEUn{wxGQPu!@h$pSjITiLzBJ@V
zztadI;QY1`%1UYHykjJD<MCy(42u)b^gQnvOHz%oR2}1UstgO`gDm*2gQOATw+*y5
z0nUA)!O{qkSI6t1<Hb7p{-hCCV3gqxOVto%6o*nXAd<|0q@uaX3$ZFB>TFUOTr3OH
zvbhg4dLup+vB(a*z~Z?hCU_lDrN;-i2#GpJ%uu1UE*tT|7XKulQU(j(#<g3{Dde1g
zHaN}>4sf3YL#EKZMp|Zy`Xp+tf+|Lv3wT=iN^}$?5I$)cONCM67*W7Njz)bR4^Lf#
z3JT9TBV5+9Bhh2s)1ODs^KU+;IG^eoIeSdK_t?RBqlw0bRQJ9hx+qR5#1*}kP8|hT
zG<Tp*grI<J6hq#KW2VnxHilycdS4UWU|_TjbTlOHOEG~21?CX$tB3nyA0v+0sEa@F
zJ+A3EZeK#9i-DQ}$oT{hILd>Wpu<Xk!}<X(;NCR^xbuJl8$&8nGO0kYr-4Dwk1(wI
zb2}O?f>BDlrN#sbtGCm(8ijS|>9<5?4vmMRplCD{RrvbmxjHqa|7-3e;cl3fmCV=b
zQ{fQuvw}Vi70ZLJAVt{$0sCO~`II5JNm*BcJHt{Yl^A-#I??DQM8_l+AxxL0;wU3o
z%oRt8<Z|P=+f{KCM{IlKCt-?}VzzY|%x<jB(W42E*d<}LP;5Uu;_rCydJHZO0{M%<
zuj82^ip?sXD09$cP4mPbHAM!!qlK;#h(Nn}z1HrfWvz+=3`n&Znz%Pd1c7i-pju-k
zfk+W}aKyrJ*po1kalI!%9}_cW-N&~*zJssWv1!Bz$ES63u57vFL+7i<Mz8<rirA;q
zWB1H{<eFODh^`Ll4`lO}KW|&q^=sm{JH|D(&an+&*f2gUACcc%{Oc#@{q~jn4_<c9
z!C4K#`cZ$Qd!!leEy$r=%tGQHj08#_d^48oO@;*R&Kc=Q2m6?V_Hoe#&_NgfkD!Bg
z3l4XB3q>Nai&;_8>X33qCgFv`J5X)+1W@g<b}6d`qc6ws{M20?wUMFGy84D;jZMQx
zs2v`uutv|5nx6|ndf>k!3r~e>%7@QNy}^P|oKvuCxF0VU`wUVbo#v%NI`yf0^16(C
zwl@jO@`Bs)(7>k#JQbI}7EK9y%H4o_7LZonl%RF&KTm5n(3+F`&k&21z4M^6?oeZ_
z6!fx|L?1^ofBlGGnS$nmR5TYR(d<RIO2m)L{U_M0`jphv|0+>l`<4O3`{{5DUx7Wq
zF)U+NCXK<1=AbocFx~!f>0YzC9=<%)xUNeYmw%i(uA<~|N&azgmVF6K!dAQFn|Q&I
zrk-f22j2bPG1*e{Fn{oD<5gX7P(BtiRm_?sPGVhB^oFG%ZO%wij!-fa@v(3|BlUi+
z#M3~<A?Vj>92&%;7e#IGWn<WBSHjNpe`eckYHxm?N#Hp*l1#;l!{;suf?myL?%*Eb
z$|)v67E{3>V7yk}8wV%I2qKgW-4e_n7i}Q~V-wvW`T%0(spvuyf%c?x3PvDj#&>2X
zc4)sie*DMkCT+G@eSSFHJ<z|J_wzN38>~t(Ghbcb%?GJa5XgA16?a_n0K+}XnwVAX
zwelc}pBvfyoK|mkx=1FStX5?scfJ!obz%SUL1N~t8MNy|t2{-*gQ!-H+Bi6xB8-66
znWi*%5T`1YK;Ws(H-EWv>os?@Kdqc>FPX)fE?uzp=L>eWEnWBef?3->y#B~NO#j}d
z=RacMOn$KG5SzW{YWDHkJKmSCs=f8y8y~(}iDN0=D%d0JXL#*f5LYuVfVi6VKZ3ZL
zqTM4k=qfsy6u_&0grp)B_^OZ=kMof~U#t9i<NuTVp`9TXyLb^?LhqJ>OXMYGjVcAB
zxTmOROV*<t=FM;6llKCA%KabV6SmN#XR3_KxCnJpooNbY{VtKOou6N7ABn;<huTNd
zJD6*JWn;BeqDO$x!Au--9wwgKOwwg8%F`*FE&qScSvRj#da{)I-354>+Rvm?N8{N}
zCKadojubwIRWN-)Pwl|<l&HS$1gfv+UyJIClPeXnFFTgbu5$o>g|F0}D+S+Qeg4(i
z`2GXn{vsQ^0KWe_Cyu7Z_wV|DhVRdP3%>t?8B0^(`%m88Ap5l=vigpm3-I3@ru#4z
znL_GD5zTI8V!5ocGF1q3CnJ<X$i;CXV{r~qG(1M?$T6PEHzv@|E=u%LezuCzazPbH
zt>USag1nGGyi*tL&e0Fe{&OJkzz*+b80{Q^VVsPYxtWRCRGa|GoE}mDMR4PYq{|3}
zy7eX-mNE5O1#Ahy1CsWP96#|p+NzXZA?hmdxqxQFDdbiLybQ*Z=MLg#%k+$-b!`)c
z4@etm1Wwv`nO^__CR0zG*s!{L?kl9^dG2SHzBSjdIkvt}=CWHYXSO~gf3dHVWc_{5
zdvwDHTeh2Mk1SU=V{VjVpY${QoVLCHi07=IX%28NGx<y<6UAr7=Aj+U3`<22HM*4G
zSu}*Da(EjFbywN)co=sThH-Ut+<pTEqrmXZVWwS9*rjVu>y_%^wG`p1^LW~eBx!T#
zB3G9oJ(I3`nX&VTk&5y_^yvsH(ZzUj=&D5(LQaU~cFCL^Vqb7YD-m}mR<$tHzVl~+
z#@}wn^6hfrS!Ffrj$am!=p7t>bFGP|tU=U$_xg4B(h_cyj+StDeQ*7JNkO<*UCZLs
zUcC8?j&Xq2Y&)kL4<l~yN^V+3j01C7z`1BFdYzhe1hH2SjShmSWEcnZ=wjOTBnv(b
zlEw5Sn(6_nT*`Tq+|(C5yRewt-^XX1+XMdJ^bAnuJ-&onD&P}kXEXG9wNjYtRROb!
zI4v0|yJG+;yB3rxY?p+p?}n7E?o)X0zhlo(qF5B)Wsr<h$P7Xu{4Sq0lQ+3?`kCh(
z>_Z`u=rwEDk2LBZ(ZA7raS?x`c~qPDo88<B@>3xfa_>S$GzPmlJz^@xpG~H5(Fc-A
z8Wg-Ssz7|fRCT3!=uF+ccUkU{PEBoIX#@A!Wsipom#nER$SxbFjA1#qh3plo;Jb@4
zhwau5d^dZ%+u&k~LHhXh1~yUghq)e5O@l_e{5AUB!ya13J+<Tp_~s$wD&a1OS-oBU
zh&jY;SLi<byCs84_rMox2YvA>mFT_8&Mdhhf)p#?%OXFxFy%)UGY&Z-W`tFyi1tt%
zhYR?ENJkb@)}5vdBWHk15Ie^>9&wv-(Ldu{t`(k-hh!O?t79I-&&soS%OY<%6Xh*F
zhY8mKvcf78z|>PZ6Fe3THi`z9?ouGbRIZ?rwm4h}HGK}}3(q45(w5-dAdwUf5)Z^`
zcgP>HB~(6H%!u>RhjO|o;<VpF+=>K6K-?9YTHo1XEyVMz1Z{*F%S_=qxj7K|Rxl6>
zbqf?bPkm9Fn7fVCIH@lxDzCcY=<yN?|AP*zf{dNk_*U7FH*@o*PR+}GYe?B!Io=NP
z!H{1xPHp#ke=+0DyxjI_-rQfptkglCV`4T_{L6f1wDJU3$(4ssBG!Y#W<qE}FHwL!
ze;i;BH4Rq`97gJr({bYrIF<rRQY6U{6b?R*pXw5Ip1NY?te8Yg8cR#yQdF(><R!o6
zm6xl}NQ$uYDJHBtj#iC`iE(#;=HzUASQQoa8CKq1t2`TmsbR-*8wY)dCiy$!cyc*$
z^f+rxGGZKlhrhp<k3w8qiqAp~s5L0D5UYC#YD^I`om_7>Lju@v#byIT@fq3PP`6Ht
zmqbrk($N@*_Q`6FOX-H*uH1Bls-$f`LBy^UvqR{X_4vImL!ytisq*i?Q#^OJP0Z{O
zB$lFZfb-@v5#5Z%7d82DBFv-Jrr;8WYgNGd0@N7<rO4w{xO$(*r%|Azv=Hw2|8I3?
z6VpT##PRMf-R&Z_yR_XFDs<cJmlULw7}SF%Sha96sDyAaF)<oF8e@zH51x!CgBS0n
zH_6f#B?JjE8WOL@s8OTDXynpFycxRj&FtIl2b9RkNU{_t+r!L%-oAPB-u(D`E?E?B
z@(=QibN<|uL+X)7W5(-S%un8Mh@zKgbip{m&M#lS%3ip>_Jw7Z^2aasr}7^wBk@zo
zVdHi_)qj30|Kg=Ku5y)cFg~oZEMFn>Ux8??=rB(Ti*7KCb6w}MRzF@|OqX3A$s33P
zLokT8aVjLGLYTOd6VpW=+n@3dF%501p`lhl!6X5q)>G@cckIXJ9h@m+cHk!ep(`tk
zE1}IZru8IAmr1ha0TL|-=p@+eF1lsK+N&nPWVwAWmQUe%2<=m$Dl$wb8|30v*}KN)
zlgzupD;qrHx-^hW^2YC}3LkG<9v!Xb*d=lq)jYdUW!XY?8ZJ{qFhQQ>W?@y6R_++a
zVfO*$%wRWP=(BQ2j>vG!<^(-E&wWB3Z@>{@eDpKCz5rDJLSS80Tral69~DTmJc^{Q
zn0=JkEQBp3_Nqi(HnulvVSAy&Fd!KN5?a;@acu{fuU(qfeI|v?v^U9OpJMrTx2)eX
z5Ptm@oDpfi7S6eu_^bQmqR;`8)VUz~Or$06M*44nq&cvGz@D^jWdRGu+XC~xJhgBm
z35?MF+@x!`o(oLN;x;SPJeTr;`r&(n7q-%omBCV<BDk>;4ec-7WNb^jEi6o_v;We<
zE$m80%of&s!*om=m1aXyswnE1_YwPPBo2m{XOt@%R<#&%Mbn_<3{f&1<18jn$q7`l
z1QZZh`)7M2*=$yinvy0D%S=tuklc7-{0s_qiAIw!sf2|%7-Yi|Jh?1TDQtcn*ke;&
zn*E@eet5X?82z9ka5OCGU3K!0NqXEQmL}?uz|GFlh;uaJ7C^vAYNt27c=Y+NYq;Jc
z!>{C1OWolvV8(>trPK@GO<)YXP416s+qtz!ZYMt#yP=&sa88M1<?guM6x5BLmgMeG
zZzLK|9BBALud9$(3$T483=4#Mtg|F~gFUbnHw8RY#ShtFI)ke#2-}_${fD=~ZI3jH
zMDuPNfJv0h_{7X$oLQs#IXy%4<L8Nf&K8WGLYo93naT~MbA70!GAQZbr{b%+(41@t
z&uAuNog3pjO?@%<o76?dJ5UdaRg6+8l+9WiN&t2e8%EXSliFE6b}Rl}XvPZY|6Vhi
zJyNJL)eg3mjme;$eOL-oI8!l+zf+7r6=Mx8JL%*r)k#NfuxTL?9C-c!MDlct00961
z00RI*v3#_RUk^O>03Qed00000(fime00000*U^f}{}=uC2V@4{00ICA00IC200000
zc-muNWME*w{r3`()c-I3uZlf|A%#Jd0U5jj0HFH^M*si-c-n1~PiPZC7{%ZG*-elj
z;-z4T5Jiee2_cAtut<ZV#1KTJNRg@`^bm@r^dwTGr&3A~3Dk(igSnUzLJy)viWh}|
z;zbY%MM}vb6#oQK5R5XuS&f9!g&%J-^Uco8dtc-yoB|)Y0{;<j|M3-hSwK?Ho3;j`
z&JT<`5=TW2Nm)lx)=f*+P*81*iiBB^F{DKTK~XXTtiP-wW~nY2wd+Ze<j4bZiRAry
zuI2kYDUpYA9m{SFd41nBbqra32Q#{btolfv@b`PnI0ujwRb<o|^G5IC_$snamuJ+}
z1>E3wGQP4-y~2P#j)q>vEuBLna7eg%!o1LX@l18ihB}RHQNknYgxth;b=`bXF_BdT
zL|Bthw|V{$E~q-j<O%LsH%qDmPZW_7izckj@qG@S8-i!Y^$6?tnJuolC@b*fXY)(m
zMUHa?<ss(;F{@x|$}>^<+q~vnL;~-qVHHOj-+b?{S}!!r2kJVB+Zy#8C*_tHcAB^;
z@6qET_oJuQdpoA$Vu-%y{ILo`<QVT_{aw~UtkCaq?lH&lZQgH*>j#}R^RoxNPx!v?
zf;(v*+bsMX`mj0l<Ajjv<**Z;V~kqxv*?Fe*v#~P`b*z;!)a5Qh0UQK<}m1IQAEV5
zK8W-GbJ&UPzYz@iS@gp!Y-Uyi!zjBg*4qAvvmfGUswp<|CeTxvW=2oRk-uPrdWCw8
zbIhWLnnWrP<=GM_lWF$)qO77!re)O>RRw2NjsF3zmFyuBw%_GB54#;Sct5C@^kj{`
zc1Z`BKZ951?EnA(c-muNV9-H=F$`@CHyCFzX)x_!y2H%J?7&>ZyoLDz^FJ0177Lae
ztTL==te4o#*gDw0uuHJ7;1J=6<5<LTfa3`#6K4^Z30DU<8+QQr67DZNHar`6rFcDf
zC-8pb)8d=M_krJwf17}ez$8HyK^MVog71Xfgert~3B3{)5%v%+5ndp?OZbk+6j2UQ
zAJJQ4PGWn+r%5PDWJ&Ch6q0n2;*jc-x+9$<{YfT7W|ORw><Kvrxd^!qxf}8v@^$hv
z6!;XHfN+yyg%Y1qgR+itfQo`jg31@wD%Az5M^qoF^{A~;`=)N8-lG0YgH0nyqfKL-
z#s^I+%?{0VS_#?=+UIoKbQbCS(@oRk(3_zzr$5C&%AmmDgrS<@8pAh6YDPImCyYHz
zSWGfZ&Y4=7J~JyZn`SOyKFLDJVv41iWs~Ixs}0r;)|YIuY+da5?C#l5b5L-Y;K<}y
z<s{=2<uuLdj<cH!k1LaFgzF|Z5w{)g4(=`PzdX7;<ve$I6?vWUj`4ov6XWy8x6F5u
zp9Bu5=4ayP<QL;t=C{P}lHWi7Apb@F9|Cj&T7Ym_z=?npfk}a{g1Cat1?Pp#3ta^O
z`bm*@009610PFz~00jU5000020096302TlM0RREL00000c-pO0O>a^`6g`iRr>&Ks
zi>9)g71jhOw6wOYY^<0X79`kqP2rUiDIkxkb?xsku1x$Rrs~Fp8#n$I&s;8_0YPJ4
z=FOS8bMM@H=F9+Cbs`AD2`vH68lc$0I!8Smz>4!oaS-pEg5nT%obQUmhy`55F46&`
z_y*ntii&UIAaJdC39;ac;$`P;Fs}F(vY`)(Z{vOFhvEpb;f~@cO!%YXJ6H>UR(uzm
z;a`fcvFb(@e?i*q$N!3md#v~yqHeW^e+ChAyLgNi+PFdob<}W<E(|`gZ6eJti=?0o
zwvpg$kt^zG;68kg^7L)e!$*pvL*k08inBeV&FAV%lsP)!zk@0~t})nWga-TJp1zDc
z_8$9J{8uGrk^ZM#TbEdN)^L2q|5J%rroV@?K^_K2C^5dlmSmm8E*@f!{T}PEpTTSS
zQuLe_A;lV6xL|EIpMHKNu4~d-;Czvm7ULAedPr+oGOUc%JX&#ljJF|c3FRz>oUpmr
zj=eJ)-8VOKUW-N<k8mu!`@}6dp)KZ=76#|VOO9k^BaW{(a&ksA&=s~VBNxKW3(mF3
z{A^Vv2C7QURn?#L7GH3W*)gWwl6Cq#n0izC8lAuA|GX>DWjB9}^9p17^s$OI1C`r5
zP-lNmR2B7?xvn7=tB0nvnt0e<FLghoYVkw^z1!FoP30NE&YyDBl9RPMsY)(3sxSB0
z>Z{6DmTEkGH}+|(X`Y7srG=k9m$u?AxzAyP-(q)Hc{leac;6U%HSz;Co@lN|%4zS8
z4C}Nx)`SDAT~EF2zMZMHFHh8zom!DUQBzyh;{U^IH0x#f``6$^Dj6yzBQL>;o_2|S
zTWZ-6+cs;+@U`gu`!()PsLGcrI`pgyCqw2-c*&VMZ8p!wZxmoNvj6}9c-n2zS8&Wx
z6vy%JNjAI5ruSa-&i?<qn?jV`B#Y?1myl$4i6k}_(Zc9;7&AOEf)|Wlhrt6zZ=(zz
zJnQ2jJYwU2y}2)+`JK7<oI7&}gWqlNtCa5i?|cCXlW;Lf1mT#7z>I}RqOcN846(!!
zPXdV~kxUAyq>)YrnPibo4!Pu!PXQh1NGCeeg|2i9&FVo<deNIc^ras*G<57Za8gJS
z#gx#W0hBV3K@4UHF5Hw+P6Zx@GK}F=GJ=teVl-nI%Q(g}fr(6FGE<n!G^R6ynapA~
zbC^q&m^s1*HgcVPd|(s1*v<itaa1g9;~neSE0Ge#PWH2fTf7%52RY6U+W5&~PI8~S
zoMImH*-bSMso@?Ec*-Ll^MsGo@`7hP=QLivvxnEb<P|<X^OdbEz|TS!QAa(8Xkamo
z1ZX125|*-zPb_Bz&8%cKtGGfdEv#WJ>q0yD##PR6mTSD>twc+V#7dmROM)ayk|axt
zq)M8kO9tmS&m}H%n+sgz4x2e4ncU!}WJxwVBu8>3Px7U}?5%5V^l1jVf!z|U_uFi?
zazmF*sj^#Fs(bE|cJ<H$rKZ%CcBMn<R2C|Wl*P(_{al8c>Z@tiTE90KsHv`M@);%F
zW9amlDuRKAzrpD#YuDp8{t;4F+Lew!cuAwRc-lqIu@b>R5P;#koLoYJNREOr6NTzL
z1@0P+Q6)3RV^Gp4c@fP<Wjuff@IowZR(Ib&v%7cWuOsjiTR^xhD$dVkWu-1gkdzBh
zD|{TvAxOP07>L|}&~uQvxii5d=IO8f++5|}p@#6SLx2!ZhY%4=9qNd%{{x4H+Z<tS
z3_M*4yG6#xXiRK=?#7;FR9c_X<dc7C<4Eg4Opxshv4c!@#1z@t5Hn<Ii3*t+Vi&1q
b#2&IsUoAoDH?E*uegFUf00CR8+DQNa9R=13

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-light-italic.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-light-italic.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..e2fa0134b1a5bc4b6805e6ec3dbf5acb17c5b976
GIT binary patch
literal 22720
zcmV)3K+C^(Pew8T0RR9109e2P6951J0QCp}09a7~0{}t*00000000000000000000
z0000#Mn+Uk92!&`-E<s;Oa@>8gJK9Y34~=42nvGMOoF^`0X7081BWmRgIWLtAO(*c
z2Ze78f=fSlWnyLq8wceV*tU$XVW^#;qKeyZQ}QSfHVz;ZhTSs}!NvjDGv5cY|Nl=*
zDl%poG-(Gb98iBxWta+fbTAT*nJ8pP49GzqPKe}MwXrGg&_t&7xP8Gm^wVKH@8s+A
zoxG22lZ?dW`wTUxurrB@N}?|`yVi9=L85}9f{4;qlx13i6%5}RQB#W?#2iGW*m&n&
zXeu^oEjX|-F{AwOi!1vRhr=%$CaJ|kypRk3@Lk&{{xMV;D*FA!Ox)c;^wpU7TwAEd
zF&3@BZRjx?=uuCDXwLuHyNeK_;0dxJ5ZXtAN+pFL$;%I(P-;}%(^6AvM)~bQ-Q+2_
zjgZZU=}}7wFD1f5{MdgxBlo_)D$RhbmFC{Gie*Y5G%|@sqLFAz-B1xtD-!?-o?Jh`
zZ?i`oVr0)qj06cmLRut@B1Hic48#u18MO>^ds7$v&&93ccDdc|{zdOlRU?yrCtA_h
zS`Ro<q=S4UTd9RM7BShJUsGELfz+~P`t1+r^3=$vQ_rLTc&b1trJ;c|)WN0&4lUT+
zIo6~57}GI7lSz$CDyE2<qorN}{6{eS6JyMSM5U|w$1IXasxp6@hoaG>&cu^^A`=+b
z%g+%BkIV1jak+jRK8_z}NLS`)h()|Bt(Nc%boLV@l^v*{0)V=|Pgwz#`G30q{~&}A
zLPGZDBpkhJ+4sRWzWbO}g;kAMWvibxE#sCMucsY{|9@{kn08J86oB5hMIq1bs7AbO
z4`U*g77RJv8%w|0fS9$_;rrStghl9P#D3G2y2Y<=k5JVF_@cdT=|7|%wOEFt<0wL!
z*=&l`PdE<#zxHN!E9Ku4l49DCG_$EUj#EI)2$&&P-Sb|LRJCZ0Tn5B|f990@a*jSN
zC%gqZggv4N>mtPo^FEC4|Cwr){%Vk7VJs=lj`Lh=a>|X1&ba8}3zB;eRn>;wg$C5!
zASrZ{)BvOm08t|VawMghM#BgIMGnRu<<?kB5E?_d6{WFjwP%e<FUT*a6jA5H%`ZqU
z2(1ai_;~pD#`A>;1~`$>%ibAaCyL@Ik7YP3oCW$pFz3(Hw~L#A9^tyJHXkSm=z!=5
z)W&ubUw+yShB#sr{<oY@;atw)BjFqoP0^n3_V!<6-<xG^@3xDGNH~-bf=H3h*1tK%
zTpdbtfQu9X@l-VWJ?6*M{*jm_jqN_WmKmamfUMcvAOFW@#t^S%E6G+YP20o9kiaZ9
z@;?K2m$#o300O|3KZd4&3wN#?0N~oKj=Zyx0;~Xr1mT8|=mi<>@(`96<XHWSu)QGX
zf9#0-`kVsK5YN5B&<8L~wuC8SHVTBgjNt_`o(As|MCQrk`DlA;a%p>cw{Cl5w`qIp
zxEcU6F&=@<Q-W}3fl9cXjabs74X3vvM{1r`>!alu>~)M~pR8{E^j-aj`MuN?UAAgd
zgP>!2Rn^t?mUz0YX-l{UL+<^wdRSNW-oCvUL1<3$sIw}`q^h=lTc7&c>`s2NHY(em
zyq9M{Z13JzW5esJIxrxv{_eB-sq8zEcj(bzF2`+Kh2#{gZv(kphbjfQqKb_DI>?u4
z5`_q)dBlk%?!|3R|F0Tu;9gV}RTeS2^yo8S$cQl$rp&CYZEWrA9UPsUUEDo9y}VPv
zz@<V!$^r9ZO}+w!iWKAF6A}?CAtfVMrG|o19X0b@1+D>4ftSD=;4Scedb4}<3a99i
zloVxiO`eKt%B{+gjL#F)=v0jxB`D1aR-ee+*K^O_(#-V1Z33AaPv5`M?o@l*XM)^Q
z<DMeNk4;ZCXYUQlYbA|+pUUoZq}N}V7J$w)4~WWTM0C?~?Czu5v+Dn1s@6<Zq^7#N
zz>iMzBHWYn6p>2Rl>Sk@`mnNkQstb}lPIz~m6RvRB88h<r0QfS@AN#v6kV);art%8
zg(<mK^)m9R9&IA4v1^^gyL_~|mx7hk6fbWNgUOp^Y<VV@wRpb$x<4O&U7ngL(0bpt
zVfYXb0>mmz3)9b%^1=y5$%N+VQo~edB}mH1&{UCUs-eVG*Tbf^pv0tg4dz{Gbj}0A
z9$C)>00zBq&5#dX8TZ9o3w~HxYJ%=9Gg$v{;Fy5iYJXabO}tVM>)5(L%@{1e5?sLz
ze26dkvk`w1K=zYB5=4SY2ni)&B%DN$ND@V$B$~vKSQ1C#NdieENhFz|0MGzw02lx)
zKso>p01rR_AOd6nkO0U46aXp!4S){70LTPj0%QTO0N4N=04_i_Kp8+2X(lbCm9&v|
z(m^^&7wINFq?h!OelkD?$q*SPBV?3l$QT(XlVplalLKUi%#sCIge6#p6*vfo;4mD4
zqi_t4!wEPEr{FZ4fwOQ9&cg**g*DiKP1u5qa0x)aunn*y%PznbSsnp=3Gf);3BWUe
z=K!w&z5)0a;5&ft0U-2)rq--Qt2XUVXndHn7s^2~f1#X=bMB&8kWg;9`yf;(PhPy+
zK8i&N<)>J*Pyr@TkYFJ*)x(4e7b>2=WTZ-$Av5K0m8alXC{YoAh<1rYrTmq#(#2di
zdi3hkpARb5CDgDHqsEMzFlowkno(?&#+Y%FX-cRStJbW?RmBZ%-MRO`qXd61_Ltap
zIlq?Z7k|I;$KRTflu`Utg4GizlZMHo*@-J$Do+3a000000001di4Fh&0000000000
zzDhR$;{l6~2DYG8VAITo37kN|lKHWWf)`s6a^xzcW_T@cpC6#wu6RlWS`hW>NgJQ%
zh?(4=L3H#6_|>P@&@CMqz!(Zq?_5XF$wMRp<zP@K(d&v}g3^&|&w}&D5F+&mGI(<g
zs6ofFigmkrL)q3!V3bN^4O1d7E?4B7Q_v05{Dx_!zFq-dX5TyjfGMZfai~F1F`zof
zvHaOhCBS3*dvHTuy$bg2JXe(XbIZJiI79GI3P(j(3Qcze_AJr*BdBYYbY!aLmvZO0
zxYjeqKt{4Ye<Z4Kx024Xb>!;A0(VkFB@z8^q1qO6ravVtbh@dIVV<kH7!rqSvH{eh
z!9wf1W{|Z2P9zk$+&M#bFK<t;@|0mlkXVOmvr0gHuTNxx$@O|Z%2Y0SS~|a|)DaIt
zzd1VeXtx-;)dlW?5WU2CfCfw{w%VXodxV|)BKa5)e(o!8pXBxhRLP1GzhHglj)fNV
zIYZJ=ET>@BYogZdn_xnHmq^b52dUm%Bp(C9eqRZnt3|917_pEQQocF)Fwv2c?r7zo
z6LIXksq}JGQHs>~NHeLl=dzF-$Wf{)=d=>dNkyroT*EMRvvY16A3^II|Bn!zmXczX
zFY%AWE4S~mb~)#A-{y`iSD7a5YWgK%T(v+y%2}1)^3D4<F2+yy=a7-SBq*pIp7YDM
z65!3rWk|6!9#d`}vAastE_*B1n)yW>uf+RBYLC4wG#4#@KdP1xO2JoCy*O2d(@dG2
zQZV;LRE+y?vU-Va%-sX-&U81Qe0p=s+ND6LAnV+&j^klKb%j}r%3&%I=1lSKX-5E9
z(~wV|x#Wsn+H*=I$5vBcX21QK>+*<!Q%SMX{QjeY@pIVJ%5T%fv!k=US7uz#X5LdD
zk43~z9K`-K1Jea8U&S+J_x}CqDK(FtmSE&oj6hB6wZwG)-oyheWWG5@((leoIY>5~
ze+4s!F){VTMRoc9ULGUU_G?JCz|Jd!TxRTiO4EnW5)mrr{li#wzi<S@<^+Sqtzn4*
zqC{@0JHN9satcZ+9z1#R=EJvtU_buU0F3~FADe^zpbbU=S|N0RP+|1K)$8t8qan&+
zS{!A^S$x^4Xx)Q!A8Z#zHPj)JJn&#AJGmi90YJFx<7l!JF^$cQB1MVb4+rXF<Wh<I
zf<y^VxDUf{9g6a&M-AQSlBph!fj|_Y*vV>fZ;p7d5{V^KOCgcUO&Y26abJ6J{IdPX
zs>qos<j}|!C{F+j6YB)6zh35&tw5MUdPRa23s-_F)r5nVlS&z-raY7j(TtboJhkAh
zXA)yf6lxOA?94I80<5ZJEwYrGWht{#HHGzm8Z|Zwz}v)ZvkA6k^lPVf@E2^4{VUSi
zc{;??;m10%R-;npn(K%gL~df-CUOVkz7LoCtg_nCgyBQQ)7X71y&~BSNtxw{m$0Jj
zIuV^B#X^;G(rc=P`V&Qwqo;~8%K~X@zOqeL+9(H~7lHeVc^NWlI{K(`*W86y1JS+p
z?q8PgetaEaQo+mjN(E@>aNUH|C&bmEw12dX&s4;r_c^y>=zpye0aJ+xV5)^ffOMh?
zF9i@BJQ{|W2elCYh?OprUCPZ5)jpz^g+7_{1!zNzwN8)u!`!B<_u&<>lS4!%A`&6S
zk)SuCUSV0}d&v_vFEB#%Ic7YXAS{Rg^V=9ga#PVnR5oHrv4-lW{~+ERAHMt~JF*Bh
z!ej7UG*ri|Xh<n<L7-lePdQWtp*jHf^wbxC`c?F;?iF*W-!+z!7xj2Y7U74G%Q=g#
z4<&bGNHDq?=Ir6wS`h{U+*+{WfL!A#_Lg=zKeAwRfE(^R*!UXvqU|?P_A{UqA-mJU
zcK3@wbUPyfGXWsKWKRIa%{jcE0)X4@H39-~_5%ziU)NHA%rO6v!WfW8^Y(QKy$BgP
zEO;OoaiT^W;{Lz-zyBDvQ%j@ZIYG}tgxWHo_=c<bHaBNyq+`3U_2;+y*I(_t0$yEu
zwQ0>M2Q`bmclgQo;AhU2W*Z^9U+)(HNCGcCQo%o4AlM?Yo5WozlPi?xXolr@L6l@g
z)pP@<Wjn6t2VoQ^X-4v*tm>xi`eB^rW!?79fx%M|NEA8^gH6Zb3B(K%nL?$}8JWy1
z7MsJ(&dJTo2Tf@2>>eJUSzK6JSw499=#gW`Po6k+`t1307uHs{HZNWR$=ekRU)Ib^
z#qX;`AlD?QFF*#8A1U!LQ1ynouoMIUI{Z~ZQ|s8y?xQcCJ%9XU9dzg@=&OfszxfuV
zxb*_m($n1A)ju%UH>v^6nLaQDI(A%!QUHbk0Jd5HxsuQjS!|kFs>A|z`3e*&629(A
zIg}}v!@{&wJPbTMExgt4;Pd)9dFUGY@xlm4hBN=fhC1nGo_j|68F^x$H~BWg$TJ2@
zfg?-j)V3*DLq|W#H?ljld(vs-IYYHNTAg-~<BsNxQ)6;S#~;n_PK{2>3nCo6xu3Hz
z5Cm(Tz#uCxn`uHrZ3vmrayp$*UasL$N2lo97uwQjda=PnMN^`TyqkdzZ9qLtj*Iea
zSgwxpoIa0-{XJJ(S1TH2Cu_7?r^Y2T70||l631dnjZ%v?;BKH7`_d`(2%>mVM7e{~
zY@NAPC+eVWDymZxP1Y)*4XKM}yUeY0F6z=XrRi!m_(l^lG<s(9joN02IhVIGJmadL
zR;!^`U6+7V7|NwQirwN`ogsxmN3S+qS4SO|LXDOuPKT>E8|Ug=I_k>ZwN~2VgfiR6
zE3W}^nyT|Os^$Tji*@Rr6)gurso`npf$185$?E2{(YZuCwL8)^`=y8UGHgPlXk;=G
z%XTGNJ5f!u*a3A49J&h1Lub-ZAjIo4G%9e^fuhOBraD3iY{}Esynow(cB0+}UJ}*u
zIL3`2xaXS+Y}u>sBarJLvjPA4fd1p?tENEN;{Z2#?xJNKst1%Or$jGPIVak=OgvYz
zVSzgtNkpy;)1oZL;)cAc);KwMPOIl+y>?Xew3oR{T`i>LF`?Gs5DHjUFH}*nYw;*t
zfG_3O9nGN!%B|U0*D&1Rtuk0|hmYRkqDZur9YfpE;h<1kIhLnMWwq`nmt)Lz|DLK*
zC}g9RmUE4*p+<(@>cmI-G0&LMLUPR8)8V~xw%oQ5$-$@$%RbEDb!M%z2Xk5KJ;6v`
z@Z~LTSL)R0|FiM?VNypfZ8YZ;zYWWs3RInb!j;n@wH9+FRcJKm@uX0%Il=@uyd%{T
zNPBcJ4y4Vs>{~!1j-~cazEgR;2Tsz2)Vfs;kzfj*NiDDuFS>*6fx|mP>UfmiN^vX%
z2H+wd=~+Pb+-L8s#Q|`Vl9JoB&V@j%8)0W7sS`C?QO)G<CPY`-PMvA3INx!P6D+HB
zf#s7^e^z|5!f4H}fwS$Ht)4Bwwpfc)*F+T$*H8?Mmj*G0ZAfr&A%u8E-(Y`o%(+na
z9kTY6^~*WTw~WU7+hk>u2X#FL_>O?Bu4h|+jQ34pTGnEm^UE6c(*>b1ZOyDivzX<2
zt{@@RDntfxAWXK+iK;-aL%FD7HKOu~jG`Op*PQ;`O0&7D=i;z)m1*DX^P|rrnBquT
z+;cW}m~6FB$8=d^zL>x9`0cZ!&U8MLuG}yG7Wc;@y!RjKc~3Ks?7ki`2ko(4APViE
zJvqPR<m1S^U5fp{8wsk0ePk$;8Ci>Dp*M>FYjXVywqz)pvn)X!z}&o~=1K}0*@-w<
zB;iPBTN^48b55d3@9_Tp#7ZKQL{2q!hcYt_;s!42YI3z5y{3xP5bp0-yRus_BNVMB
z(Y=l(;+U0#HeG>k;L?+LmN9cQ`V;<_f!ohowrty?WAl#P$Cf*;T)K7P(Yg7(`<Dz$
zmCL$-4hR)lK(7p5twUEi<SkdP9*ij0-`0=)q23@2MqFAHLk9SP*&pf;@dtYhrOWXs
z`oH1*G;96A7)R^@Lae@yxk~jlfPD)|LI(whW)7Kfi?cV(bM`8RpM>}|d#<uJIj)WX
z#ug|ygtcOK)B&9&n2{qOX_r`4Y>=wSn(`@wYI@E<#QQDucqobbu0ZQOaSeoefWhS6
z^;$?Lo5ckV5(8;?Qc{?L>Xwl7RC11295Fm;3&xQH!(jxhC|m<uU<=SGw5D$YS5Ofu
zcymmfaqS$}Zz9HT^(x`~;A9Fw3zZmyFJtOD>_es|YF-1}I+uvGDK$d(9*N?4cHxqR
z*Wb0x9{R{c={eS_cgx%A=*E2GhX-}lp{?c;6;jY~#tIYTacqe3mT{yH($#STjO>HG
z`{J<P&}W#&-Xn2X0t%j1gSt=rp+u1mJUuY*v}XxXpbLJ`QqVPwpbS+g4QHL)p0n%0
z{`mh_lKR^vPI{gQCwSdE5rZ4ROWS;%&}`fP9k;>#qvk}tZg<82l6U4}vH|l)oUJHe
zQ=nTa%j)f-*H@eZ!?M(_eWWAkeB7nqjBTX28s_F``84zSRg@_3RFX2s>*yL{8R)GD
zNlQ~6=(k1)Sy9Y*r_Amv>J9Bw#$+`mp!5kO8Jzx4YP({l?Pv>~<1=m)TD&y2AsLNB
zaZnR9L-;0u6N9}Wd3QSo_0}C{+Rg1m*ft;>?)^^P0A!+QO-V|^r6|^q%!mzPBKhd_
zyAD|=Ce9HFLqZ7SoUx*_hV-Qic^=A{FG;*_DKUuG${5+jMo_MXd=$gn`a_D00mKuE
zgGGv;)>f2#+ykOI1wEC}&MYIIMU$bdM)+FvqMiHH>)8K6J70x->qd{To_yj_w24pW
zF!{-RzfLpWI0kTj`+kA|yrR(Wrp8?q-SE3X$WV6Xj5n%{dXHfIYUezw<p{k_gAD)7
zId%*ZS(#&yW+*;j$ULx&@{eaWnbH{2e~WgT|0J#%DgKZKko1?(--U&b@^>>17;bDC
z^f7|Qf4YDwy*?TiLL7C(4y}joA6a46SUK)Kj_P;cZ!#OQw%U5-*8Jd)WAF8qy1v;c
z6^BvvsS_72+&K5(*n{`xABGGYPn)!8(WZHaa)<luM{A5ZVzdKnsjOJ`Rp+mxyP(d<
zJU_}Zueg5RJ34m_xN>o{8cHJbxK-xF)6N3)mc!%}8bR9_wn+KWnA?mwu_ET7RP~w~
zJw_Fm-WJk8u(%<+S~~hFSrL?}T{-x8_kc79q(k7>XD*3W|I=1{+}{m46D^LC4y2x#
z-S$gBiVxevH&0)0S3E*Zn~60VJJbWw41dH<;MapzcF;b<)oTiBP3^Uqrgr1gIov_=
zE7%bWSMROTnsmpvx#l;2-9+KzAZbjG_+_??!~~xZ0T(!kzbj&Lbf<B*POx$~Y-TL9
zpHXt;e%wcLlx^SBo*d2M4%J@14@U4-l^4woLOor(`(r6J-+mrDc4*~>J$6~tZSWzP
z5~r>p>K33?;F5Rv?rew))mor@%%I2@l8oMs?K$uA6<aZa_g_>EtBfMKpYx-vAsZPa
z2U87!h{}=o22;tN<f96JCoL*jtI69hmrpwH^gI+7=I*D)Xa{xVK#K!<UosTA(<z8$
z{+T%h*P8!ss{L7Q`lGojI8dNnL<AcP1}7=Um@mFa)Gci<!i{OvF3C{*-bC>2m#l_W
zM!7;-@*}Q<^pYTQ(kIzhdP<5#ZenzK#bOMXb>hU4BP+awUeY+>-x3kg*{$Av{P^y0
zlc&Ca=-+&BIfYYJe*c8A-hTK_l2GJs<jye>RiG<Ulw{Yr=D|$}3|jzKmEgb+a5-~h
z^15j=d@skN`}5V<BIzlYZfgTuzB4chDkQq{)R6DF@tfUo_-s_g6(1L~1CIr|NFZ@{
zQT-Jn=DiW5DU_C8%c>%OE<j(4DbrK}t;`IOWLqCZE?Cgnc4tOz(67gA6qz^nns=JF
z3Qp&AP4a@;^uE+C2$U;A7uJ4CVYCZ<+4J&T5<T9RcYF<txSb_UBE&37sxC%C)1$6;
zNk*PjFXjK18KFXeq;>MQ<VQbG1WRFu!@NPhlsw<=1;quE+;t3u>U3+nX1RV9XB~)H
zk>bsR!d0?@DQ=?xsgX%A_dF7sb}09)rA0Bl7CLkMs_}>kM`)azDfP^o=Iphrm1B0b
z0>R_^&X{(3l9?|q%<JZb1(VqArA0Q6Tcz^q5?nE&l2e9hVn)Z)9|ws(#6j|3>6nEc
z2GlU+5fg7}W@cK+(Ck?IMNf0kvA4oT?2C=i9eN$!c;a_1O{FXlT440t*`)%+E4^i2
zs}-O}Syyf>QI@vh^%a_IoIrpC#csh#dzz}btqIPLap`n*n{aWRa&zqxQS{&L6zUsO
zE_&&lj%xVKg`O|o&W8;Py<fQ7jeh(4IfC4qVg0>5qr;OYxMxqQ1EroRXq{#K^G~Vc
z``>ZTf3JD=J0x-7EUP*=xXOuG${z4OA7IW$`Y;k*Jed&!+4MP_IQC}h9R%i?cAYmq
z#fgB0*ZLhbOmSmGQ!@OElIMQhJ`n=Oe?XA_oT1e_y)M)zgOGv=oLK&M!>Bsn1%9X1
z97Y~W%Bv%Vd61Ftw2-*t4i`BtWLg579F?Fmcx~d#$K&0n7ds~&KAPw}w}|NxcBsX5
zhYr<=J5?pM2M>A5;N@*yXO@~CUM}iXi-g9XVs(CTBt%;Ur#XpWg$-f1kh<Sp?rAtS
zszAh+=Tl^{F!Lq8<Vsv4zwYc+Mz+zszCM#)!)Xo13|`q_Xr$xgW#K$8LP!CPcDMJp
ztqPb$Wo#)mwc>kxMo|MH@I-!4E-I!lF<VucFPs@E^7A*NTNl*|sb6xt_jtEF)x!t%
zX>|9-g~E$TpOo{t-ouX<r!xF}FsUBh^4{Ymw=JC-K98oV`GT#aPk*?zV4rH?-v3E&
zUe_Y^aZa+O?tFz)sr|G2e_#AY8fT67JFdm>b1`MQQOe<=42Wes#bhdkwhG3V)syQ*
z^N(MuCaV)F{yr?Kie`DTvfKjV(?XdOOOkMJ%niZ<!K3O9^CQXso6)FNC_ZW#3S#X^
zH5>F<j2&Zi3wQ&G6jpmOAv;uMDv8J29tpchKU~(iM9zha+_BI&oIBB)4N0fi`RNh8
z#Bmv=vFRDD$gC37kne!~n4U1e32lP2hX<veU^+)<G$XSl$YJR}j1v&ivh95VB)x@b
z`jsx#<lRcXu0Dr?xVz9)JXXtLaErqjuDKCqQ_2#`N^(l119G~0lRE*=@5+U_OGy=(
zx$_+zB_B6A_)8FCIF03lgoVz=6DO}8Xs=yvTTFali2SpQZ;cl>j0g+Jj1;$($@^w`
z&ahLG#hwepZRn&*F|9S8Xx?ivJ*l4lb8n4xDIqhsFT%1q)8K9OMM&CiDv4bGU6PQ}
z_Upej?M$UP;5DVw^A9d9sE$px$vCz*HlWG+M9}PuA87nzxYD`OYv3QQGfu}MbEk%j
zOpcy@JMlMjxPd;YLjMi!u6pH8XoER6cr~6jySqKG{>Oi-ea=<n>`<!FzV+!vH3|69
zOSeXvr$-vkugLt_)1f;@cC9F8HDUX9Js55o@oHXXLoC+#nS7)5o9(%&ox|;|Yo|y2
zzExcA_~8=y@nJ@fbY9ildTOx5ajq->>c}AHG8t(q4J}NMDeYl6>&Cnr`wM6Gb^nnc
zmW2uDCksbADjH7@i}GO=`fu^~n!5I)nBetz`pV`;|L5SWEN81sQ)F@U=9|gCGlv`K
z?Ugi7^faG6B)8#Z!>9TSi@ICEzV$?uKnA(pV|UseZnE*-Kjp^)Z8@Izrk$=~TWDBR
z%K<}O{GEzEU4F(NgTt&3Rl~B-u<38w^DHM~Tof(**H>w)lU`9!;Zxx&EMQ9}8l~|J
zW8AfXne{DIihn@{f*HIm^vpy=r4}=Lw5wRT`9IE9d*6BJQ9nDV&%M4<`Olm|iC2s2
z8W2cYH9Z?fR-_^trnmZ3vR8n*L?9rt;__eFSJ9d{(4shsDB^^uP4dm!<R`|+@~8M?
zo$B+bm*txau|XF$c>MvZjef1;hSK#b0`q%?W%7}t@}d^jH6~Nmxc%2kc(64{C?7_F
zsYZOiM(bKg<2Pjl(`IDsxqhsz_UOq24pQr$NdYEKDa!+5>YVVt*LFtgc_({U{%?x;
zrG0I*Ld6!0cP_o_ky@6=y{x>vIPi|zbaRz#O2T{H4QKU!C?CB{{xst0zJFiL4r0uR
zgQzu~AVbXfTjj%6G%Wh}$h<<`t;2q$7rw;5>HV3sb)xvUsz(l=k1~SkBJL>ht@(z|
z@^q&n!&U{JZ}SW5q>NZz>||hyHI$%pnw-VwyUoK$;w;k)d!)!VZU6sbX-x|?vnc6B
zb1kFMP0K+gy;&}D)(oyKwdC;0G>qlArFWK&>m;Qa>U5SEdM@@jd1HGe`xuO2b3Fa{
z+JE!=hfACNYKm=(wJVBy36rRNq0YHa6d7U>1*(=)*Y+2zsg1osmljM0G)<S7AZE-8
zrk5(OYZDSVX;R<%WYb+Eu7`Zd#A-ohRcJrY`0Pn=)!=V&S$|INXma*6STQgMn$xXE
zTX<~jcCF!!bbO8ZY3L18muspwWt!E@9}%wpsX$LW+#Y!{Q9W8=0fC$(oM%G6Q(6$n
zE0Cf^eeMax$?2Kf>Y9IFSvex$!()wqola=h>F7PX-;(lZf1+!6NB?19x~@bOW=B+_
z$cT<ImMSjQ;D^AhcOEWgh2|BUS2}>qRvbY>hYDcIL;c6kx7t)W5=rp;8T^5B;}yP^
z@NW(hg6-ck(LVI#B+sk}fqd#5R$RQ7hKzM0AmguMb5!ShWG*&5CwsJCSkqCgKNB7f
z^W;R&eF9Eayu=ZV<fn}BFDtQYGE_%DoD^CeKHaXl?-jN0d%OAs^~ThxiF;potr_UV
z7a4e3{U84IY4@h+h*7!MpTI(mzt<6V;4r9Nu2~_dNa!`#^PNJZSD<j4-?CiqMUY*5
zf5E$8Aw!m!KY__-`$2z)c8-_kb!8`UTsS_4i*QJnMavyXf464o`4Fo2M$)^=pXKQV
zlN3xpiQsRpfFQ=Nn57`$f~D9HgXYC3z4M6(e4#RDVt!Nn-!DXga1G=aP4Sa7KKqP(
zA}j_8gAP_-wU_SjWBuH|nT^z7*IGS6nBKAJs-_`^d76f)nWQoA3ikK0Zn>N29e5tC
zt=%KiMLjn|XtSoU6#3{VMNANuaB;(UahxMW*Zhpy+`5YCP!iW9(M;TWb!4Vd_w``g
z;d+T_`r9cuwQRMy_=ReGdb2)5lm5JwHo#B2(lD{b6`yDndY~0W)52Ai=u!3b;VUC$
zm9b#FJf`mKn-4Lhs$Z!2(NblND1hN$?p;qcWz-8N#_v9-Y>gk-tS5~j+e*=U<8Zv}
zVvFF#1?GSVeZ59=l2y=oYNnbK!0FWq^PW0(brkPppL_dJKuJW!G3zS@^_R~V;|%i2
zbr5=!S-{m~pXVi*I1e;7iZl@N+;7(kjB)aR0Qw>Rhn)7_W*^rOR6rb^m116+H6f53
zJHpPXB|=>-^w=K^lAMK$-#$1%P4kIJfVqdUlV!F<)A8V`^1SGnVlm@fl<dNrA)2~i
zO*ArhPmzu3&eS-=jOeyHv3KhwEnOnjv6GeMEgj#TJpEmw*@FV6gEakc$6QT!n>@%E
zBgIflk;W+REFQ0*ZLPoe%?-F=>H@>wP`wvYaL~i87V+H8{RATG%_U`6#fW3C(U6p3
zZ)}z3I@@zf72>3OTNyb(toacbysg{^o*f!`zU?R8fcN{j@2g~=ig{E$BGHhtz_hnh
zbSfrTD!N_zfw@!i?si!h9)$#HlJ1sO$m4X)4Z8%HQcIJZIzR!B(>m7D|KW2t+ir)B
zuDdk|97E%!;ocVjZD63`pATb$7oPsGezpGfQS+%k`IdHZ^61Wpa=W?ke$DjZYZ}E)
zd*LI5%QJIp71Q3P{N|qF_%E*JM=KT8rSQ6@y@?Up(i-kgTad<)ZR?b~$p;ivnIn4{
zC6QyIER#U+pp!rLeRd;RUHpt<Y@js(A51>f<RQg(>ae_!(}RgvUOSJTL?+cHO~L(O
z367Xh+3|M8f%a@Kgo@L`t;)uwl1YmA%D5sCRlYja!#D{i<c>w#<X1UWLFkQU;V;PH
zuc%-Qw<f@<$gX|>F7`1-X9b7~%kyvXzwtF=mL^jqmj#N`+SmGf-cYZq<T+6MXajUT
ztLgf0Bb-l1a6M#4Fc`?*Au`xg=b-i^Ks>{3_k6RQ)JL$82%}?a5JE<sfB(hhtm6a7
zft-dr@1JA=hXsdy?kxtUwgeX_zmQ)yjF>~$pHFf=9nJNmY*gK%^tAC%X{@LVK@?wa
z6HG5ZnbF3WV7Dh!jbTL=FB&=koxxN;kUTa*ZkkM3v2byCVz?_4hL?t8b+qeqDh-3-
zU9%z*#X-r$3H+uBN~o&|l14{oCS#$q(0*uf1zTgpy*~;PC6*ix;_KA6Win~O!b>6&
z!`<p=3Bzuzi%Wu2@U0h3kc-BGWUVo?(RFxLS!2eL@neQn?zBY@#GpsB;9^Q)_Zz)w
zAz!0H>^yj$6s;}S4!HJF+Dl;TXzS?$WovYZy(hz)8nT%S#BDSu+vmez_B;K(9Z1vn
zS8LoY{Ce`59T4Ges7KdXd@pa<4ia%sn{9_n@<795+4frMsp>v0nYHw3<{#er;Aq^B
zD58~@#1^AfwZH2jQ&U}Zg5znH?;K%6A0-y4;#Yq0n&{Lncc{0MoAzzjNxx#O3K%k{
z#WXR6t+9-`Hl`b<k<tD!A<${mtfdQ*-ecP5CpSsGn}9f+gJVl%z6G;1v*+w1IVi$@
z>?pa`cHD4BZ>(PhIz}MG#+6Enq|!)CQ9cw`SJ$RgVq$s4*l4M=s7x9|DCR@c%uaGD
zRrz@pDh{`zA+KP3><hvTAsDgWHLY!>Hg}!efs4t+`&ej@6L?UGdQgRU9wkcX6Ee_A
zOSd1*G@WppXl*m9ouZc}@Vmy<!IVZA(rY-#1RSFu1LqHr?<>x){opOqp;(a@i5ee`
zqY?ZH_Yc9~PSpJzSWrf2LP9|vOzhXP8I=`5W|XTVh~t`Eztn+jMHI&cNsRYcff548
zHEQE?xNv`d3cCGdtFLsaC?}GwS1snAynwGIM8Bt*@dHwyah>zRp@eYBNqU|-Cf2p_
z3d`|*^e4`n;dRs(-f40FrZYdO+Syf^Br0`r9&>eBhD+^-fMbBxD9@jYj68j={da!t
z8UF7RCyW)foBG2|JQsQTR74CLJMZAIxU@{2XM_Z{&ZWnX(h3QE`|fm8R=hom5hJc1
zn(65nSWr|@Q;dn!xRwJ-)_T*QeLui`-aJAXpWB3SY{u5q)^s@aB*$3tpco{iNSmnB
zF*QAjE)c;;F=RwSag#9cEqBm{>;15^H(}KT<5*|RyocB32(6nMjWJGjCd|8=2k-{v
zQZOF6LeBcoUGlDfw=({}_m$hNoiZLSqZnbo9>6|odsN><i_`?p(k>>Yjo((xSjZ8U
zxCaoNyN%~BRcnt`3A^eAH7iGI1f31ZC1gQ1g<YrSQ2062bK_=IE)<LjuPGv49|7ab
z=w!Nh{=Z!oOO8!73>M+52L-gGB4TSKjN_4<i$;lZ(23%7MPUyw%hw7Kf#4NoP!fiO
z)mxC%xPW{oSW#*zyM~U($gw#tmg8vuo<wFmDvuYJC4wn)S~F?XUcS(Zt1n}b3XiK5
z$?YCgyCGK!zqW(#Ht`Gu9$%zL4$UFOWIILdr`u5?Bl6+3)I(HNFs=`cqcp|FXR^@-
z^DRlxc5Ze(yvG%fF-NM%jU&bMC!mcW&i(h4C&i-y>@r?^$d-l}ic}c6lRM=5M{~55
zbR2i50cA3@+aT3CZhSi!rT3kc=;zbr#n(3#<-;2r!*bsBONxrY)lK+l?Z}iee^Qch
zBu_Cgr6`ydlaIhxZqgk@R;`UD7}Gkz@M^h%(?Nnaj0Ydt!6d`oLv{bYVHstlV@wr-
z>yN1eZlnj^=%Xc}ScgCRhPDONz~ik#TH=CA5CZ;-QWkcvy{MPETE|Np<XPbRdukIQ
z6=HQh#(1Oz6l5ToK_8%77%J=iHYx|vwI!2kY24fQpV`=QSn#$YV~OM4chxemY?~ij
z_AS3W^LguEBg4yq{NhP|;+W4qL!Ti({$gYP5O!bfO%4778R;Uwx?Er`U2KB2XXaS_
zy&dgpi8C{)-Pxr(i7jz)K4^e``;H#*4IFx)foI24Yvp(9<ka-yl1zQ$Uv>|h>buH^
z@eC3YO~=KZE9ofYI>-br@hr$&hHuN)cUIyP?|@&Fo?~j>dDY}xJ2&uYG}iAoXz#(*
zcUjaC_jhgQ@H6|ry~oA)QG4jik4E<t$N&06G~6wx@Rc3>m1AdJ9%qT{$lJOPHYj)U
zgvKNP7`ysjS+gPfdo~o*hS2q7E~|XbLxo@66;Dvl8*{6l2V>zb$QeIvlqskG{G_9<
z-qvwH>4juD%f$HgtYzD~p|!pO*FKsZ*e4rgyBN9cv0F4$1KrXrC;WQu!M1J+m6!=f
zA7>DE{dxa)@_`}Tz4R=W?gT6S(H7}(==%8<{0&MaC%kwx``{cQ<(zFKOi`(%v=mXT
zaycTf3}or{I~3G$!j;dw&3@h7Uu<&v)V`X*t<K!Vht!?VVw3%{v7cvp`s_aCz`4O(
zEwXTU$ZWL?51>+mJy+`)vx9B8=Bi{I7{z<MI}Sjm=44fDo1!4u82{Lgj_1&}+Khju
zo@~>!J*y>vz7G!G5>>9cTDWl_GVwS#Ra!p*!}}Z8AMT7if&T2ooFt)Nk{X}X{#ILS
z=eV+17Va)f{DAXYM}%9}FHXqAU1*LA8%|%j`Uc_=i3BVH8JHc>aB#SneJq9Thek!$
zL8*CFl5A)~4joIwB^1W(cR~O3|J^^inD=v}<=lrsUm2gNc#-XSIR~Y5bn|O`_>9LP
z)~((*eLX~NIIO^vbk^1zTqkRH(%{h+v;p=z3)D6>(5*WQ#{JAEYqIC`rUZYB3buDg
zEIBa-ySbBAks7@IoiksMm=E9T=E7ISa3?+K&UrSoVsYQjN}cph)HMr>XtT)p&g})Z
z&fFMfUspU_-#)~!z&zlzbg<AYAI{kjN;-;YLJu636Wc#F^^PjIgi>?6yih*YPvPN2
zxj7kY5~|FasGsj<)Mn4Hje9lIXF%Fsnvr5j+rOCZs3XILS_A&4-gOoBOyXSwF`@iJ
ztbYsnb3%o6ZB$`?bwaxwc`usH&4)9mbv8%aZ%lP9`E_T&pCh{OQi9#mw5>|J_rx^q
zx0_DWDUl-R68N*&T4H0f;_a)pj^!S~zM#6EyIQJ{6tfg_^R#z~vH1Lat|>A(ga0hv
z0&^++ew?oeaaXx=xkz^E-5z6W^yHa=yaSm9{ZVIf66#XoPNmXMKL6>D65!{pK|zG-
z-A|wOB}T_kr^<sU(cY0BW0IE$)ZQ^RSQBJB=5Ot3cM46Ayt)@QtD@;t=dLHkwO_qT
zDa-wS=RDD1pSFX~8q0SF)c<{&nAuv8)ODVE_=WbTh5vpUe_|bH{T#=$^9+|tqWFrd
zUA)Hw)1)Yg@=7c3Va<Vm<L2c`Z4<DSjEs9{KgV(J(A<stMm^bBT`gDRKDGNkn4O^$
zYJA^jV;!&N-_S~9xjeHOh5eK=E@;{NEhCP?POZb--SOO9W=L@;{cWQGr#+{;VvKJs
z(05kgoeQHZUVT#3V%4Su{j#ApQ9@`8$2?fiZl8}bQL#6u?sN|(K~KNQ;KSE_d)T)&
zMfuH;KuwB#<)Z(pj`&LZp#6)zzN*%KD(myrcqlF@;dwcz3Q)7I@kucLWofpLw+hTa
zx|gG`etpV*!@C_(HS~A<OmB!~^|q+$J+Kl9X`#%?@_Y1?{{xf!2S+3!pIf%~o{)y=
za23gG2guudLP&dIJ3UV{+cS&h{wD`RKMADXe140(Kkx?S!@uNHS>zj^@7$sU?!QI7
z{epHo_@6D*#=8D}K_bT0r8DUPuXYuNkr_Q8y<D-i%}&<mhldHq#$z3f?u5Y23OZY>
z!RR6M(|sS#(#}HVuBegua7<85!%tu|Xndyy7o2uN7rm~veP(BiC-Bh1<hshWU%&jI
z_3n_7d<)(Vci%v5JX#O_SMXt*5OThK;J*m{qAqyR@!?qKwdxdTL#V#5n}I$y`8T~b
zU5tPD@h@nD@|$krYCn5F-+xf_f!Qu0-$RDq?{}Wh>7NcJk~|pv%<JUqfa`~zl`(j>
zHCUw{(Ax^Lp?{IBMcj7tIF2~*c?Z0X&LJbT_UH4gUmUx6xD43WD`RE-|1VVje*CDg
z3-pyh)^%2`|G#)pf9siGA&^G#Brt~>B;<m!mZVd-y$b>)97~|ZP;xXU1m|9e(j=*x
zB_z`IcnqqLcsCk}65;+CVp#QQ(KBLhK@6YXMi!(g^+eF?zDeYF1zeAzynm9KtpZ!9
zSq9CidBLS9Ftdo8`hoB*@_6&^<*(<tYF(`yrq#4zoPV>F{d)7z-gG`$TxkWqzL-0|
znvKHeXOn50iiOY4pP!uGRJ^#Z(nJ@Pr~csR_Q`Q4Ylccw=)`m_Jpiuf{{GF&?=l>c
z`i$uc12Lm~%T6}LuEK@r85k6sB~5o$<&K2scwggzhKqHQmL2|dMMkkTaLmJC<Poo%
zL6tBnW{#IJo(^Jx3!}I>j#4flf}4TSY+6;3rNBL*sKwu#wPE9;W(4>wbsK7KBum3K
z$oqWS7qLqBN}irx4;$EITC{XE$b%U7=fx{Q?e50#DTuSJY&6{Q_5+O=l?-9u6fqT~
zV2-iM%a1_Mk6;>Og;4Q2PGI;FQ<--WQt*igkXR&8K!O3T*xvvt*#h>0gW}OdYm%J7
zfUtlA=D18lg=ujL7wpmD968o;oRI}3F}|gj^qx*46TrD3bAVhc-0xO`7ESM^p-fBL
z(ndk^B7&P;UhK4uAkae1Fxe3sZ1AOUUVi_jfBt}i8Bxx&zI&1hz#%V2@T~P|dZt>I
z{_jEQV>P3cEl=1+VHy<0_u|8wnz*i~)tf5jZm&wDHS>(yNfH5;I|SSSMT3P<G!?=1
z2<ou$T&jQf|FU5iLV$_L1o^xFv14ByDuHYh29dlB@Je2;06ed{V+el+y`p0(2u2$i
z^hlC7WaJ8%gDQCdRY#Oe+wP6A5E{rwj_&Ta*M*i8H~_;d3oD$MWN*n4hF_PL0N&gj
zMh3M=%AJ*)y^N{AfuUfSU@X$K8RB%PeIFwQLbcgRw<JD?&2`6=g}?Odan2G4nS3x1
z%nzx&zjE4;!kCNNW;MMy%52z7aRWsyjeTOk#R1h=0%(LPBPN((UNN!0TVH80ZFnmg
zDAzjR7lN7=0-GpCG)@I&27L_zMeGuP)?Uy&jrc|+S5yEhJa8O0upP~?sgilWCd=AH
z_e(T8U*8%ART4K@rXiRBR~BYV&N+Yz9WDwtsst&#65b3?zNpCoHXXL_u?N}&M*tCe
zo-_^eUW=+~njuYU@j=R=zjgNXwFA+>tIlmc1<H13PlQ1zyMWUA>I9dFz22rEU=7-;
znn?<C0JIV2i$q^x!4e7*)BRw`N!u^Lm6bRtC7d+#qN~nG<44`vzyJe)6dOtdjX_Kk
zF%wmBuN$-GfLsd?4OO}&Hf)mvB%7q>!fKf)NwbKA%OjkaIK3R=YZvJ<K$P4XzLuP)
zyd%LaV<30rT6M{H08edT*fC?26DA;<C0a}N+?d)W79vd)$yYEUa#J_<508{47D$CE
z(q3_xpA^FmC0-Iw*ufnsjf@f>UiI0DhWh+V4B)nc8&nK?Pr`ei^=wG`I$u+w3%>Es
zf8gW8ee#f*==xmVAAb6ZwUwU<#A2?Wdr5r(C&%qa&tANW2HfLZnw}*-8=rxmU9y|q
zbUqe!%p_F4%XS;BMM@$<?L}|2GHqa+^x(r^zdS<*a8l+2$#psjPMUJ2iRQs(!*YrR
zrBrgP^yY+$Dw9z$)f!NBc)VpnO6RCWJ$iGwB*>!9WJ-iH(;G}0CjlU|-4d$-OAEnE
zJZe^}n)f@q0i1gMBMss&abuNKp}5efh(g(wC^G=sDGjj02GxrNl657>-;qY=Y&4N9
zq$i+d$3PAdrlVdUd*L5^;mfdYP!^b5?FYD-M&{4jG75Ep0PGBxw9$AEt2JRDU-<W>
z<-=HX2ot7?8$r+-Zzw6gLRqy^?qx3$u|k4N0wAHQzApHSKgU2pu6=jme6CfMvq9Y?
z5<%9iDb~9SQ?CjYx8)&vPG^1SW6@vz<9PCE!&z>rY`^MUWLqx3GF$cs@yKE@2fz&U
z4KxX^J;xzqwSE`qm_b>m{RK5bjZilZ9!Oo)fC2)}z$6GmJCy-2L{thg$rt|K`;_*+
zeM1Ty8AsNJ>q-+1{Z5dtVzyAp|2;y=Ci#9_1vP+BpuMu*7XHfh?t|=5r0D1wQ5~`9
zb`kB+T<|?>ftn;&ah?~<;js~E=cl3uaEkT8+cO~bK8`k{Lb`))w!MGw85u-+sH^Iz
zqn4Pi8e}eSD^n;1km;Q+DCo<e^|&_^M6nd>luNB7sQ{x8r@k=ly_qfijk~`p4nQD-
zAe482pz^!wR4H@W!O6GK+3d<-(r5mq7jpG3M@Z%GE0^W}&9I{+C$wqDhq*z^e*|Fu
zUSIyPg4B%2!9(%HUMp1hbE9G021P+ruX0SJF=e+!3M<6c7^tMET3`ev_3N}iAY4PN
zEbjI<{{v*v$z6K?C%-j~$uQkPKs4`_XwF};0#Qrn0wa*gjX)J0Avo+HmrJHQM_^>k
zfkxtLIOCWunKQL~cw4GhFR(}<?W>RhgHnqOPCbXSC4&M6$~N7tnr#R?rFCmBX-BW^
zDEpQmWB-wm1-c>*dhb98pd^??g%qHPjyvU~qcxl8&+!d)Qjb$vjluC6wP$(v&f*oN
zqm9aj(cFK*Klu%W(6)Kp(V~XHg^+d_G0)PS8NmzmrdkD{D1>f)e7KknJca}f!GT}#
z8Pq3$*>Uz1_sibJcy08hxEbKXVEfimA=)K^4*)w@1?!v}3?O!sDhWiS2chGHFGiSx
zSVnZcx@}y7Bg9A|w9c^g@8Lve*A@f4iI9pDz{rA&B=7pt*V7-T^jXlrBB9iG|09Qb
z1Pcb-c0QLwX+wY`KQn-BX(JcJD?`x%#nUvfF#|CcA6p+Z2p@FK!&m<^W++MX6%74;
zm*5l=R<W0^9U}s-XF*xN>x`ZN%*2%s@n2;~Lt6hmNd9Mn>BO{QnM2fmh?iU|C>0*>
z+6I=DRK^+yc|_SS5-t792Ouj;2rNjE$`XpvBC4a>D=g=&e9sVxRUXZcfvOQaOEQJV
zm4%e`H&Mgt%=hocA~^Xt0t69&i9jF-r#$3i(6Azc9C(Y4tuqiI5QiS<7neQ+|FKzz
z2~2`M3=S%QooK#;p9gqeB*fc_5kLT<<6(I)ZM%aC2tf-<36M}mt!ZKl%y|hkc;(hy
zu7*BQgyqb|Dj_;rDbX~~0>VP<(EPyKL>h@xFvq21yFm{hY))BuJEuIkwDu6+QH`{s
z@Ss2x$9?eW^78v{{p&4U>)@QM?b6%?+fnejv|E!o8->UAsJ7z<y_AjJ3!ZK>itO#(
zP4at)0`({nS1nnye~{K0k9L_haCRC+im*?2eZ(*5Z9yceXu|bUR>?JFO`MIe@3-Cv
zKv|IkJ4+*cq1T!QnMVMyX$gwL9}dS`2Hr>_iAy&bZqxLxPz~QHIi;lCZ@V4pR9u13
zS4l~dL}zU16H}^26bH6h3;_UV<3Ip%aq0cJhY1d1fM(qAu|OKtTsFQ`$W+U@F*|aa
zlqHcArle}r*dAh-^NxZBDSAqd+OB7;>0~AG(b9_rUh6#OR&zN|H}b|Z3p<MH2p?=S
zuDgsvOyBCxGVtGFnF4;SlHw$V0GBh}Do1f#ilZVDj5vXy5+vyxjS<+F;25f;Om=1o
z1d|XOHm9}+%-_{;qs~iW?iL%O7A;hhyD4Mi6e5J6qX&Ji^&P-5ivdNPN}dq~Oo2}1
z1Bw^*e>D@=*b-+^02W^HkI-CkXuts}{Jf1n2X*I|LXr|4d^CBpow?&peo`|xiyJYq
z!6fkU6fgh#-G--!dz>?0{=*Nse181&xdW1)8NgkYA(Xg1o^C(mda(R<`})JACRE5)
zxKBE$j!!-x>=h!mIV*Oq{EJBiQEfOREDKdTvYFQ+cKb4%0;EV=3L+J7woLG9`pIBZ
z$*bY;Wd=dOQ(JWD{qMZ(9475jL76i9U1(KZ)vz6nvU*lBl9LBO(?Mm2;aaUQG;E@&
z#}E_uN&*oRELs=^o-IRgjr0V7g&@5_q(+)pW%8f>IA*Do5ZlWEsR;`ENNKW(BEo3#
z1bGu@7XHq=Ct@;fnC9Sahw?I1)@Z*9R91C)mBl->Qza&Fx%OAM|Fr-y1zkjXLx~gC
zWLlv25J}P+LX@bl3=%!B-UP(g^c5<&#+^|-?(CkWHH=$Uz(yB~xZ1b2R#58g@3?a3
zXfBoePzZhq4<x0<@P>+&x0unjf)uA|8FWUU>oU|xB6okpU=TtC-f&slc631^ZP_66
z0NdXO!gt$yr^6dF?p%QTgCtvUF$!{;Gu07qty73di5yt4;(Nh6wb?(xcus+`@to?C
zQg>aPwDgpqs*I9T=rvMlgo4IY)bzdv+a_pJrXb5l8KjNasPbrB;L8D2T+6PSPM&PG
z07y1kJj>dSs%1b46(!7lRw}i{R|{E8xoLb(XjIF&v0men8r;LywiBqZz|PEqZWC=@
z4v=j0NYO}9fuFKcXy5~Z`<ygq;d`eSno>^D7S4%RXw;HMwbVV@>?8#fLrYGB^oao@
zd0E_%AxkRLw_K*WV0F6PLMaqCx_cAPOA<A|$X~nZMZqmm3Kj^fJ33fF4T&U*^a_pX
zVI<<w@5-_S&Z=6)rU1{Zuo5=cnXa>Y)RbT0ht|IAWqo{_gd;*Aq(qa}pYP@&?6RB<
zn*_$K55~Z%?^+!Jj2Lt0b8ra20Bl1=qtPZhZ*yb#yH?XcHMx5MbiG4STe0!_6)3>(
z7{)RRS1}H~EA-3K2w3Ee5J!AWOa*?k&UJ9U9+{5zi`)Itbm8nHI{vtPJ{dZ~Wgc9^
z=_@U4kh!q|y?KGscp9!rS{o{`t0sY6M*fxemB06w!aLY6j@Czfxy}A>ZP_{S?cmM(
zSe~Ly9@U69#aDf0x6;HgTorH@)wB_TAHmxPNemT4va%k%4@dx5;IRVjR~1}ly_w;G
zSpyd|0}fomf%YsJ1#9o5!QURoRp6>#{HmjPK%Y1xxayTcEAKAqvU@q19C;={1v!~G
z)+0~|)y+CN4@R%x`{~O-9<<?8>Yvi^HrV#)KM1tL;C>pNl5(WI%h4*Zbs*N0t_0lj
zx8v2|4iBKcBbeRQ13a9wt_Zp3uOW1k<9~@tY9{=w&<&CWaI^Hw(g%C3W;PL{+eqx0
zWE~WCp_vb8x(*I?ZUuem%YVKW598gm;aN_&`iJjwjRcSOtHzbbCwsoyP3qmtpUqV5
zFl(D1UhyIHjlxznnfoRjJ|vAv+hj=e+w?eviQDmIPbqVu;ln!{zs=F^<<-yRYKF6I
zL(0>fbwD?4Hqsg(PFf~-%|`1O83X_{=Pa@Xh50I}gm9G{P6wLiXX(SY_=fXg?dwne
z|4g?vDf3C~^5B{I_Jo{E)nBOwFShN*)oi5nBg&`MYU&Oy4$$LuHl1JPut=1n<#Ons
zcnVh#VQ5&#5K7$mU}NEqC~x=0&Dta%pgi+{4y(`{P!3_F^!Lv1hs+6Gbo1a!Hg*Q9
zHl`w(S7=>9h|{xTh}q^ffhrh5cf#fSCq*+L@V<ukJrx(Lwy%nq`Z9r+=q=nC`2nA~
zdL9$@wUo6&Gq1nVjWvcWA}?vl{Pyd2bdq?9oIX*Qd>e5-qBuBWj0U#-gI8CX*;}+I
zWD{$!@U=KZ@nx78AV?w~%rl>~4G(p$kD-?=NS4S-+z}^XeMjKZM0+4GPwc}t94nTr
zr0IdmD&nV1Yim@uE{(E&?&<h+^!PPCJ#=JUlQuZfvTO7Y4#U|UwVHlyGCAf=D30{;
z76XAwbn|N(-*vy>oBw*u=Vl}9SID8caCGi)bYUM3NK$@Sl5oG)RBJXu17kbKN|+Eq
zG~SY<U4|0XU<rZo*;0Ul{EvD6P$5Wu7}wi<xQ`Ma4pwO!#JZ#>Xl|U^3TZ(n;3lAD
z<x<G#DZUX*W4m+j7vsZfiP|MqPTLeXcnt}h@#gl9dhCjDp?`|c%Ar2b%Pa}PNeDp%
zNdN><U#xw2Z%JbZtU|wGt55_Vhy~JqQA3#|nLc$&sR&UZg5mZc>~FA>Y^3agF?DO&
zxmKx$4DvvD6srabbpR}Y9z-I6I<skJqFbWfPZjIkkO$Tel}ZR9OLfA6N)%av5;UJI
z1dphPdf0=7fhk+E<O(Qc%~|!yPg|p!q|gzsGp=o!Cd^_ZYPt|d!4|SnfNVv$gaH^-
z4p42Ga4=J+iG#YAXq5zLox<L@W0m$rRS$?8(&KL0j<{()?JLqip0PXt8E1sbNgkHw
zEZd^7Ed~&@Fqzwd>DMsh17I6>v;kym=_VNPbw{~|Q7*eHFEuyc-{C`CSd;Ai>ObpC
zS$G189Uqnr0V4LLAw8$8^avMI`ghD+Zxv;J)r94sv4dX@FQf-Gemo|T#4ko8OU6v&
zq@o?3Y*Gj*h-I5|5RKhg{YyWqgA5{Lg&yRS!(iNGn(Lby!LXPBju^%oD2Vmn{n!5a
zv$}$a=dQDKozFA4Mm2vTg)xs^C#z-QL(Ik&S4}I;{KWk;&}Ai!4UKl^NxK1&YRTl#
z%5{zp#dezrN}+TPs~}1e%L*A$OAJ_kHc@PK00-L{#9BkMRFF@>Qra0SQ~jmu!O3Z|
zLspm6;$>3@#FkkZy7~o$RA>EnY^wdk<1sf$P9bc&J(NxV?KW}=PV-n1E#(cUB*V^c
z`|B-;Grs?<dzE@9IcJ1=L?<f))3yLtbg4OeYlM<JaDm5gM!UBC{grk)k;0f?A3P=6
zqcFhE-6QSF^XzVKMo*qL;IdaG+r-D2Guqz8wh7nsq+_Wr)(_}Wv!uXTwA`^qpeBP{
z(E}%uwZ>}!uo7$IMclM)bzkX|%>#+vpeH7S+`|2^>6A)Zx!yL9H=x|)EbAB~>|ZZm
zI-re+s1`RLd~jMxY(f0CV*|<IqXJY?l1u5X*Ha<_UJEAtUAAw;@7Y9;sXut`y+rqj
ze=@HRsE!mx3Jrr<0%dt)HbrT}+VIe}wIr~$i&8poyrGvAMygb7FHf<bchr>8P@d!>
zkwV-*5*VLL$x6Mp6V<9nG;0SL(@+DJCzy+MIv<HPH*9F}DIk`57*4L#ec#0OIABy|
zw=M462FE&Ex~G!6Fc-PjW??NuVZ<uI*@e{K#UTonORiGnuVubpxrVYKZTI`#bVg^j
z^OBUs+_YU*3s1k?*>E()9hR*$mQu4H4uD~tD0LetNHpT5%@`!ss=nd?>p)XR*V5Cz
zD#g=5I!7)h?CoH+J-<}VhLW*=eeCbfk~Olr8g}j41}#_^rVspl34>ETk)DKXGIN&D
zE^ac8l)8ap@OaQxXbKQzR%4LzJC%lCHW^k4W@1)@(B?+WMe!|-(~kOB;&@aLYe_gM
zK#?TqOHqOt3)r6xc>z%@RgcF|<x@*1mh!=h8EXFIZo3pW#>Nr`x8?>__E?VUE&$jV
zves`Efk6eA1Z-AlZkiJg)LC#F>`@4O9N93k=+|DF*-ok^S->}41t6AX!J#c&7vz*K
z4_-q*t99gOFuQ}o`OqKiDvh^K#an{=B59!WT<c$3Jt0DYSA-TQ-+}7g_}?7`)|8MB
zo4U0}KU>cxTe9Chq3TcWiu|wRC6W*Dx74Y52_8Ii7vvu2XcSA`blfiQal+S6tQow8
zaj)ff5qA9j626YIemlnV551f8He~tUB(}LDts_URdkm-^DFrmQmPQw$?udFx$}Z_@
zPp@<yS!ZU~%wcMG`+IRDWN0f0#sj~#>{Jik*+M7MkdeFT$hK@rP14C|^a9#V@wR(#
zu4@R4$Lk#*g3?iU^<`8J3PA5IPc?t)06f^XvAK_@e5Oz_Iz9Rp%b^O|mZzIvz0j7s
zM{lO9+sLTjoFYW;DGX*ESVs-IBB|2KbkC1NfNTFsh@0i!8_?KdZZumX-JL&6b=fF|
zLE~rUavCc`w?{x=L=79z0WwP21K)1BISJ`jvxSTu+YW!)-GpJUp2rbvY7~HGTiSkv
z*#yT(o<o$--ilhEFQz&N#+o~i)-v#T6z#|g6bqg&@mT@TD_>MTu^jR5AIJUS2u0{@
zcl4~vXWuz@<HcR|!tf-cMC-l%>x&0MD3|oaAs=ui)X5|=7vqp#)7<%%9CS8VlX18j
zs9PP3#67@`zD2~h8fvr<PS2lCI~?Ei_ykW!2?xvWF65e~Ps`b9Ih|?ez`rTr+j;3q
z%<=u|F4^iRD1XV~y#8wt<@o*>uRi`R>)Rpw4`mwNeR#BqJl8mib-^o#4ge9>-Z*67
z^7jN!47ukb%4$EnRBTsXjmvH364uV$k3l^_;vwe>lIciXCFYTytyCKlQ)rD(-a3KD
zd(GZA7dtszY>gk0@DRVad^CJDjy+>ho`#~;tDEcdz6V-4nn$ZJYJks1VT1~J9?ISF
zY<0@q3EH*cD}%*{wy;Il40t~pQFee*V@(MzH7gl%(&l<??7P<7EPvcV>Bpz>J11KR
zaSwNUOdMK?Q<ehPyDF8E?H-Tp0j6$=yQThROHp*LowY>GVR!8!CYipX?l_qIh4dXQ
zJ80k;JU77Z!&aBzA9m5nKiEt2)yZ|R28+^3#Zl^G0tH^2rzv-=ulO|~lHA@!I*@Y_
zVBw`s%}d`s>yj7ydM78+bZj68wzFO@b<@0Udgwkn+1wzf_*ncr&=|SXScha6d*bfg
zOB4j81#w_kVD0ENo_A5yiX@hZANi&`XJkKRT@iJZZ6fYWPG)edz?(W9PrdJ@T=*|=
z+;>yP{t9lE^~5Cj8zG3CO)3XgR^h4vnXBj(^2SvR4wuad0E){vxVDwlD9|P-Ap`9M
z4{(wYB&ddyZuhKeGI$8#jx>^MsE|cov78hJKto_57-4%F5{k{XXh_AhbtF)m<C>@n
z3(IlpcosuS36+xobLNYwn9?CLE?@ZGhS)UNklNwbq^H($6VuqI9}C+Z@gu4Kg0qok
z>=OT#zeS8#wE>si)OlnpL1)Q=I}1{zgkBeN)e3UaimK_-Pe+kSDk)E4s{5hiRahCn
zVcj~sIZ76+8)}H-f#YejJXx&^21yw-s};xHK1D#!{sx)$1=n2{t)ToENPFh}B!J?y
zyPslIewwH{$vN5=ptVG(kBRJ!l=W#y{W5<I3pY8w!yZXI`Cz$Aoi)$FYY%r(qdCV?
zMS;qF-`U$Nb<uJ1@9w{OVyHeQt;_lHuSdDrjIXVB*G&#%{xbHCzdHBmi^Z0m{30=K
z-t$iB>ml<u(?gq-y2#0N;$~*CD{8E{seMwaNer}a0$9_lQ74WeKg{_~(KccA!o{7C
zAy_jK&|-Z_#M)Ng5YK!-8pwjjzcP?U1H&oGW+^}@59Y%%-4UD%Q(2m$9+<X$E8a;v
z=nw*tVKH0Bs0OD&SNK&c6Jp*ig&|d!r#kcMpfF?2>6Cudd#%W#xLBU<%<BVGy1M<%
zGM2#+24J(%@KF!0mSvwuAGH3N^Zg}O&#Xg$GD~qMxr$^m50JPeyAktTe+t)re*}qq
z34ZJKGvag*>5D-mp7cGHK|l-E5gff^<)%?XP0wI}gvE=kFNYl*%3FMb-!xamtVS>*
zgG#`BomE5|=BI8w=F9=mF<CVGB^1D+Kn6G}J$wkv*^S+|X6bD$G?rtGVK%p5+R7v(
zBZurEpM-LO{kbV~5tQc?SYv~JL0c>8v~X0z?wFyT>n(eGQYB`ku519dm-h?hs-KWS
zxeZ<(u23P%ql4wn_-fR}X`6Jw1?ch_JV&V^`AF91m9cv}Sc6BcY@hUc3ECo}xTZ96
zDq1(1nYN;2uyw4+deV|jk=EgoY!bqJxAG2J-r*bQ5p=sDA9w;rcY)wY<)6KrqJn0p
z3xe5qC}GPg-m^*z*ulK=a6YF)@eADzDN-2|>{l4Dc4T8g2MCdp#9Y==Fj^K3m2J$1
zhESPOpXzvsD2+s+D<KK^nNy5)D5XxT(k0fl28I4}N^>dR40tL8Jt>_fZMd|2pcE?L
zXTM+r>EU>cqBPv*-t1b;Z^1}q0NNl*g3+v2ovqp|=He33pmOPQUs#+n^AH2mYg1p3
z!mb&{wSFuZz;D*34BR=Q0%fubg{$vV$|L+HXZl^B+ZPS5Uo8x*H;NNI=$n$FAQ5Lz
z+US2Te|`VQb!bvmZh;V#Y4QSZhQqjmXQ`#Yaovl98=1L;v9MbKB%nnWf!@CGJ~TSV
zozc>bcOibc4cn;vjH!r-fY_|uD`Diz9HjEH*Z<gR*8Hkhxt+57J~-||YQaNTj|17f
zSB5@ZQ^L{`Q&Id9YGnspV6a;mpBqHQEuIyat^n+K=nM~q8qc&Qe`04mXzShDe~;8M
zVaRLAi(H*uRAnmpf+0feXKLlZoLnw;b(~<X4Cy0I{QE?f3HZ*J)c;&4q0!@%y_Ws*
zJ-yIlO3)^qIV=jXEv7|z<nf>1JR{%w#o1h9V-;-6xmgIWlCSnDOB4=Mr9V76o?h(V
z73%qB7c$?{cdq#<E_9R>gZl@gRd6T)kaajGFi@Obl|qf~tP!4K(Tvu{&K_Dw{>s#d
z-E%OLgFNNbG%ot=KfmKMzhmgj%hCdi;52c$d^y?OGejdgA@<X`KTM6QclX~>_IqZ9
zUlYWbR|h>}R(&?TTZ?{2^jF089)}aLt2QUGYD>b*a4+~i>YuEg_-zZq2-TlX-tge?
z<aGGJ3dKJ=@v~Ozd%PpGa4r_z9h16$eSx!aAh5gdcd<DWCLu;=qmpyv-bZtW+*VmB
zUE{Z-QU37d!mvoY3T8Rh$`q4Q<%e9IPwyl=-Y8u&%om>%5N!_zGT&A<H#+K%hX7cn
zZ>1Zg#hLW~VI1Kh=UGtI{cL;h>!HoQhanzOl1@N$``*RVHV=>Q{XvH|@-E)OhOhIr
zOaPeg1zqIV_fPb1Uv`q#<L6leQR|yyF{mFE(sF~!_5|lq*638vltju*%our*LL4*4
zw}JA8$v1uoQ3;AdX{Z7%Mh@QQPl+7?81Ki-C$l_vtJ*c%^uoE(F>F0u?|#HquL9UQ
zi@v7!QnST6*2<nUEs+7AwtgXt<Ig(Lzy_?u<YDS0-%fC>dBeNHtPdd}<l4=lmSboA
zn)8+AMa;BzbaL9|J)f)Fysen9RVUh(Bc-M)X(<Y7gR)FwZ8OfCdM6_)nPSoemL**`
zQpT2|myB*~8dc5%^ZG@(LPoxse{!yh(d;@^hKeP|(87N;l`&EIXAa8~+61of?7)hX
z=abyEOAfKde|I3cyMegtGt^!U^FbKb4TE#Of7QyiY@lLAe|(XOi|_{dD7u35phgU1
z=7|kdwHVcrqIty`DNDSlPUhHd(5&tAb4C2lQ)HemjPEY9xfoaxxI;>E$eJaEiFlG&
z17)0dDu$epz(x&N+XIz(USh0goZJkw2y#{gxMpQn(mt;3AokGneJ7BBhZjT6iU2Jx
z`j~bwnztXB(W()R^ZAgtQQhGo6~f~Im=}Qj2;`w~THFHiKA;s8gG`>Tx-cE0E{PQL
zpyxBpq+Dkn^qgiYYlWPuA^vA7bN3-c3*3_P7&y@FcaGHtV1H(<?fmN(ygLE{cpWMN
z0RET4?R<bBKN*hX6LPRfy8+}XT%$z26sesdjYjRPF=1`zB*yu60d)%7g`$(TiwsD6
zyI5u(59m{ul-eaRsPygb;(KYk%;LV>E*HOgwJXuJx?RO@9c}jzyY99h=d+b-PKe&Q
z+fS13lIu@V?lzx;8wK^|h!RGq#5ay`c_j3YYASLXisvU+)l`7?JVYo>KvPsq#x*5q
z!Oe*D6tqOw%pOID#&n#;wgegoRgQdI+%>idHtvJX%9p;U7(%CcsAn`Xd_%KrQJqWJ
zL1=_7P6kjNIK8N&tvgOvZCr#t_g>;Ak1jD$ktvFdmN?PHE3r1UuM~yW0kq&cRhrYy
z`B0gX%=JMpd~iy7LY9C{)F+3tyr)%3GI%8OtK^PAuu`b5lxz8_<wYR+!6TNVY8>_F
zQfNv%lI7hSK$K&smcsIt3bxurr1BS-p0_~WbcQKgB(gn=<eM{P!p}^VLhh(Z6_;iX
zqrn@YPlV(_j%yq`MZGBG@s_*Mlu=h)Vt|`ax+p}NH;SfEU(`T1Rn^hOAx<)KQNl<#
z7BV$XvhmV)uJUsc<y2Lqk!6f}WF!(jWtD)o-bymRrR!;BSz*YR`p|?<{UT4Xez<r{
zp#hhF#}-}Cr6kAZd{=wpXY9tdnjm^oQ6^d_qJoq{i4_6Aap9^Yys8A~h0zvtVoUho
zt;?G?^;FeQGJn-N6^aWEQe1+_!zK06CyYXH8h^foLaGh3aQ}(>m42w-fFh%!W8FP<
zk#|ll-YvDb9WrCi;_j$j+pQfraprQ*&?n}}i#H#>{P+tHC`j<LsrqEXgo^+JmkI$1
z1uYE*R{FCi<b>x<DBpjrPTQ%>iOQP%qrt(IEk|w$t<a_0V>5nfF=EiNeOsZpe5N)T
zbsSS^R-YGM1X%3Af1iD|+_87AI;K*-KGkk2bj=NS-E!L<-zu`_p8q+n&L`8oci#g%
z-}~V0<q6>v5)&ygqh6U((mMJrAXlN%cdFE=rcg`if_1IxG-&tFM{CZUJEmQ(yAeqK
zZT9_z$ccg|iHdj-PvS+qi4XB5eyFp_R@)r5*%n8Pw?6!_*S=5_4JJnb2_!)zm{4K+
zE(3dXh&eaKZWJf#+pS}C1T!DcZ5%Cq+hvL+&2d7C8mldl=9uQ#=D6ng=7i?N=A`Ci
zIj*M0T`|$*fH9JG9ozTq)wWHm-W`tT#G11c%v@IAUfsim3Famdvxx>0G{-i_9bf!*
z^4~%#VfjI9chnLx#aV0qKkKIRN~eYQDu|{Y;SA)35#<i>MYF~sA~Ja)&Dsf;xFaLJ
z1wmh(aQGf^=L*b7bS`@liO&_8klQ(8bsym+v8wJ8EYZYjS#!TRlJ>ZVeXYsTTq|YF
zI12JFAR`7)tg4eA4UzJE#d@SVU)djdoUdw*JkR$qBd_y4vyk`sUQLis*gKokIC-xK
Lkf!TzE&u=kDziPT

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-light.woff b/0.16/_static/fonts/FreightSans/freight-sans-light.woff
new file mode 100644
index 0000000000000000000000000000000000000000..226a0bf83583627ed9d9575063e22d874b84e790
GIT binary patch
literal 26908
zcmZU4V{j&2xa}KfVodBzY-eIjY}>YN+qP}nwr$&XZoYGWojSW}ZLC@ky1Tk+Kf9})
zrG<n5K!BfSiVZ;iZ`>>S|Lgze|GNnZDM$hUKqWsc_J8nhe`*yL5ET0127cCjKTrVS
z2ALF=k(LJlfFS_@s5Ah8si2?AP)b-{ksknnDfrR;=zt{_6nmeg<!KoJ0N9fsJCz^k
zh`37K>RamC003~jKf2H#?1=~z<>@;)-~#~g6F(Z7{{XWDd23{2Z27}|{Ag-_{29tO
z&=wo(+W*)h*8FIGV*d{ij4fP^ez+C@K++8Wh|kc~`=@PUsA~WKXfyq2SpEaNO8KnG
z5BbAc{H*bQKm^ARCSzjh;PS&o{P-6I06=PFH$-17to46%x+_0%fARrZBjN&(S?ap{
z#MS?~HUR#A0R9E4W~FOs_{06&7vN$50K#G2?h4Ds+TH;GF#3rB^z_4D7jO)e+SnQX
z*qZ$01E~Dr(Qo@jSey;@esVVb{lkI(hc#dI@}Cjt2Y|(^gvbBJ+4rR%-vGe!Y}mKU
z8cu{bR1ZGFVBUn#dc3Lk)7Lv{hrqvIzCBn$pcQSVMR8<m#Pu^5G`n5v!JwqcTtqFd
z7s;odA3QIBIS2@tA3rW@hwADUzn3w-jFwa<^lW<kdpDMTz0~z;KqcrzTrL6a*)Znv
zwpjS!6GtwTG|q={7gz%!rLoXh;v0a024tHY%n%hai51Qfc-5~Zo5aV>RVnbLReTtz
zBEn~S9Z#gLrfxQNx-5sUUlm(VO|*U;e}CI!Mdt!R=EAP$cHT@xcSXxB{CT<b#43~h
zUWv-=sM@XxU#`SaciEh+O<d08V7Z4<K@~8)bf4}b8|(Gr|K$(_lmKQ0R?QDxVGMJl
z?@<ZX2wo?sUTry-x29Vu&yn{T@H@m5t)YC4$91{XbM(f%QXc*^`~j7#3~zS2#)0%6
z;Z5;V{<Dc29rCQ%P{LTpNe4(4PbC-USH-N}#Ci{^+~2;YsDfPixxg#;I(50r&8FSR
zc|Bidi%_|G4gVzslUYAMwVBb$FJ;xv6bO)HgkFP*hBsML&3+5YP>P``-M`CW`ndhC
zO9-fRRJu_F6NJX`iUlO%<{i<DjT;QA3=x78?(1O;TRmb;Oo64~X|nw;h{hx4pz$ej
z15Akh*m(-D=N9|-x{0I_5C{exn*O<WH>;7ELP_=1Ak(d-mJu8hT5)y(29lEr{WS49
z<|Ku%48y;hNQn@M{XNc<=LS@y`x1#t4kAY4oWugio#2tmB2FpoMHfrP5V!0K#gYgV
z%_T;0pBFg~$kQF<1tzJcRob7d)ZHaa_dj*-kmfVCND7UB>qv@&kQA#~hn56Z8i5^!
zTpd}noUw!Rh+_#a$*#4F&`whVOF`*0*DApy8`P1z0@+qVoD}qtn{)`8c10KTNT2o)
zFNO)Igwbj-g%SJn|B74m3+91vz4-k3qTn82d<k)ZTMoVM6Tr(7#?Iaa+ZZ^00kT5W
ze1W(n?CMc|Az=EHwiA#pa>f&mezFd_e=zeb?aI1ilItURa>;xMg;X0+y0f)oY)#w{
za-wtO=qDWo>F6H0fqKQd<GS;_W8asoi+<LQ`6lcp>W+IVFBHZmj*cB3RUT0uMLw`9
zB9R%cdkkT4W@Mg=7&HLMPNvfkqtTx_gz4t1!dH$fPx+DzD#zd+vw>X|sUl7vaV+44
zUKMkDU*u!=ToX;dB6*H+PqpUR4<M4h{{r;;FHFmhxkPYEb@6wJW#M~^rFRc)@aS)4
z$5+R19?*`&AJC3!$6Ch}kIYZJleh+Xgh#|0kJ?LcNOwtU#jslyZT2^~qj&$Qj_oGp
zA>tvdmZypD=&i%MZ;gJPf7A*Sax+FQ)DUK|7igZ5R>yFT)y2cmiEt6SBPah*I6`@q
z7@F%Jp*zEv=dCixQiM^CH+R`KY1FT@$PF)SQ=tg^To6tugoEs)+a@o{HhTF#8Q2Nf
zW*$o1fzQ7mbg-r7muf_~N9F!=L8YA^e<WrM&=%fO?|_g-s0pe76~)(k9Oq~v#22c_
z>}yCAnDxF$g)00#I^hs22g<frR#_0!o14ylP*7keTMR5(x*3p}pF@aYbzW2+9^n*V
zRX~kg`mx0hRmxF{GOWH+zH0M$U_L#Xo+@V(swIbnQ6)S}L%~eI#eGz(1x(;b>ZV7b
zo~FbJqN%#aIkMRA8W~}*kG+3CzcgbKhq~e)Y(8ut{u%WYaHNSB-QtyXlNeK<*ieW=
zlF<Pcf12Vb>EkRL^h--yZ^csG52<i<Et4Fn32k)Y=;OE-+rmW0?>p>J7}+Ms!83r;
z>q(JHAOw9%Sa15Wg7E3D3}ecCdWw8@@m5K@+F<AHw;(&i=egg-E5h-9OS@D^yaZVx
zv$G~#WV|uPDPsZpb-!zqwGmY0Jmi_9JC35gzh@Wn@vu|)eZ*qL%#<g|J;!ubg4SDf
zJn6)%WY4c?Rm#od8d@#B1;(aSGa@XMh3KG-MxLbFC(@H~(MF+Xd)RBFVj1_KWcv=t
zcf~UDSz={kp!v;R$JXg0TJA<^CPt#VVSBMWuf+On>Y+t;618)^s+Prko$XbI$9cQX
zK9C`%`zND|Zlo!*vMTTAt0j*mhPma&jpag(XB@i+%FD7fc_oe^=H-UAQi$Bhl84W~
z`$R6*+m9kRtIx8gtixVg?KJS7F^@#@T+){xn%7Q8{x>7$-HiVZ>!2reFQ+Xv;dS}H
zGVFV8-1qb{cI{$qKBfrSR%sH}*%fLwMHW0mqB^!lT>HsdB!q&BS&u2&r=y1^&)G<Y
zt}H9uwvk+pBEHXSl`KM6#zvIFN1W4jtEla0_r06{Tqa((SBdz(qf>N9+F|GB5>CYo
zk(DC0PCFj`*<L3hmn|4U>P;MJcy_~J5Bt4gHdKOyRjRb;KA#|vyWn-^y=cbP4MHDw
zyn2P@_)as*n7pztJg>B(D=lwI8XCYOF{ti;XSte_<2ctQk%4y<xgQWO@>;c<#q7qo
zor~&DO<f{jXqwBq6<Pmfv5APTN@&a@6dG`yc3N&Z)Jl=6CO9<SKkU^c`kV$V`g_vp
ztMKz8-n-h^V(8-en<3%paMX69_`;&iW+<&=+`a<kfJs$Fa{DoO>qP-`c|;ntF2cv%
z)LdqQSf4z(v8sJQr)@(&)`Y1v<3WM@eTv+CJSNrLq8ODeZ_KVXmG;VLX$w{a2ljw;
ze&s{N^w7%5_5SXU{e!IG<^<s^O11!>derUmWas0`V<m&aW^k<&+Vj2gVMW$ydC5j7
z4B0uXi;(JKi~AGl*9;adc>%`GSIKZ2Ay)+q@(JmsE8fwC?Zs?BI>Dj`Q=P!7S~q9h
z(tVuC-YHq}S410DtbJ1q1m9Hybj{N-*FSJYO=pYIySS4AN4X3_q1(oZ%@NmYs{u*s
zc=YGl?8WFI%PEB7nTbULPot^2r*u`PKZdW@f_6zFu0mt~5(K8woJbq7$fqJGg=JB*
z8<!ciJ+>&*9JO9fO4u<aF3UAHw<Mo02(MP^u207qi0!V83KVm6*}wdRYYbLc4Z~++
zl=y`K#Czq=ubjUMQzquWe>X9p%*#YUs2;;8^z=w5=p|J@SX>1k(Pf0Yrd02tme56*
zY2o$ttBlwnn#qK--)bY}e<f|y?+tjXB#`~>_PO|%YOIx(D^qG9txMxhuMAZ6;Rm;{
zZF5*HRTtn&X@4?~L^a4jrL24lCqIn5aO!X-Cz3DQj8Y-mP#42qUL<86Drk`pcV1fS
z*0jjW*@KxhD@7Rsdq=tMR6z)=BAK8ETMLL3fHV&sH6a|UiAo{-j-tn+9#CjNny-l^
zGsd~R=zQsN#$`dYuT3@SN7dtQLd2To<BY-)fbIGhVtqFFywM)QenC=p`}-FC2?1W0
zmLBIl1fn)O#h2`sE{g;P8=t}^CWX9vc-iJR>B{zU2vP>&nk^>%lekANUJsjwj`o1_
zoBXW*c*5)Q0CtDJGx;S-hx9L=m~Gqdq>@DK+u%dx58V$S?*f=T^)ILt0Wd2BAdiZ!
z<1774TMsb|?mr~E)d5k{p`tHXx#P5-l!wV*mL6pW-4qGV5GzTRiArXwAKSeiuig-g
zD^gSiK~A;G(z0lc3%z`5S`SX-?(35b)-}PB7u$kM>-|{`<J=bsbJuZm*YS4O$>076
zg$d0zB4YO+-T=eNBCY@Srj9YwV*IiQL*B<4jIl8Yl&vBn;W4N*1%d+MgpT@VS{D0K
zVT3j68udy~pj>?L2wW1<vt>N#TRlfn#bhIc9{0x{_mdO708=+kJ@C)-nBT;4?wx9}
zb6ztj6-!Z|>6akAIh(&Pz#|FO=~C%PH^HL?uv<6UI9)LxqgC-TvLDGQt%CByLoUZp
zhzy+nwqeRCciAXH2<|iKOOUCpQiM-qVkxZTj^(<ShBxp1S$4es@TXAAa&byTt90>5
z%P5sFFQVw8N<vC(kRrPy$N<OLT5NVfPK2xWqu#6}P@2J|3`@oPN?<iqK`mfE3R3Sh
zNJY70EXx|`Qd#=!>DW>)E_bG`EmOzR{XV=o_e^52Q&y_;+Xy_5WSwY*JanNF8~hGN
zoSqCf+(<7*9*WXEoTegh-p29<Lds@t{Ga{3zK)UBL`i?oSWh2&_+pB$uPiP<qqc#F
zzMhVbj-HO8KDaCuG&=ec1YL~_Bd>3Jx*xDV2*|ZHYEy%^DGba7Z7P*dUp${(L~#m~
zZvrp~G=Mh~=E-GE|Hs_OM+XC32ppjW<bdAql?n#Jrvc6%IA##U;Fe*5{=*o?s6vN*
z$U)dwoKJvHj8BSBluwdRyH|`rOE)qv?kszo>igS&au@7Oy$2msaWV%#2R|1dMA7n>
z@+73f<j42>cky@h2kqPSbM)<XIs_>mwyVueuP=l<-i2GuY3~?BB|OzNUh8ApBillt
z0EiGv$a0Sb34X6%m5;A)Vq~U&e0X|ra<n$DjEIVWf`kUYoS2%Bl9U#otf;D>qNL_O
zd2w}NW$EpnnUR%&g^7*6xv{mOrKzpn+0oU(#mUY7`SJDP<!K&#I<mh+OjMkN3>54n
zEHvB%^yI0z`StD2XBA-L$yDmK9*1X*=1Y~^ExPkZmFvxyD~<NsTb~f<Vsmr#*UHp1
zW1dQ}{WHwLxe2}m^+*?O9qd7h?rHMSycZvD5aiUCjM3PgUi(v<>CH}z*xW-m53~Dd
zv>0X9+xjNw=9ky0e@*WKP8rdQCewM6cV9yFa*8nV!2x}<;Xg&?PXj3f0|2A|-T-C*
z4?qwg1P}p;0h9nZ00n>qK<1}c0k{I^{Z#7zz!3j?uQX-brPTQlHeumGd<RzW+17s7
zr^XDq=YstjSd7#&;0Qc~G-b7m4Yvit<MkDQU8}~($t@5`2PZ%i@!;nsKw>lx$z$T+
zsJEX=WfUKaEpt9KpO!x+is{<~pJe%z5^1alJQtjwM@BUGL#SC8J&ZZNe?fz%&n9q~
z0S>vW$>A|1IegU`ioa$*pn~2%@a{R#8<^%F7TABC5`Z41+lu>NV@$S<!@v10?&{Ie
z6{HmH(eM<>eKj*3n~m2r#?E&5Ga|l@GV#irTY{#2Ef6f;+x`}h88513NlB^uB0}yB
z9Cc8XZ=85p>(#8QKhg20nDm8M=~r@v!G(h-`90T6G)4uB_@q}p0Erln4aIi2G5S@&
z`znGo)FCRyu^-e3J`}s$lfxoi|IL{$X=#nq_o9m3>)eeY&i*nIw1$qT$Dv2=^GkqV
zNbxlYbGnlX{6Ow~nm57c@HW~z6SCoIMSwH8(GiYb3DV7LedBICtX-*fDxBZhy3DL(
zj(qZ6YpQD)Bnub1pHIIxefs@N%=e9xs*1@34)mWJtJ`;8<5@^~2O4)B>3nys+JBE2
z2mti`4S;u9Tl9o>`F+@W^E`PmwN;m%*?H8pY7r+{Cz^B-M<tH`>)D4u2$G%<pAdKu
zT1F5wEF|PNIlkPDTnB-cS!w2~KtmX6V`ZZV*OFhcnmP4w(`l^|7jAhR%=fz&WBY0V
ztew>K>-Vv6J%@X`YxlGJbu$5Wssv=S3QzFO0I!*$Sl>Ppc;>D(qUy}Im4@H(PRs~r
zSLJXsGI;Mi4-Mb#XUPfL_(z^MgK|Z}TzjJ78=Lm0PTkdCU-`I!RQOW^PuSWRi@Fpd
ziGTn%R~+)0z!C#^@dJbEogU>i`DjV}AsOqY$b?gVIj5H8BHYyrE<=`b>V<UTxbPiE
zrIzKfD1+D_1NEkp7)cLKDWSzdAy$DuxXt2_2?-qe9{94SlvQOEPl|5I2f_P-i;IF?
zYnYwo-kf|dQ)C_~969jyj=~`VeQ`8v;*J9$08B()Rhy?#l9n*u9f}OND&QAHoH%ao
zvZ!J$%s`$f)G5WN=UU;ve@bP|RzUjsP9_*MGr4*1LZh|(mrrb6?eD}pWm7TV-Tv0*
z)GT1dI_%<ZX&v-8Ki)LyFsScZQlJDEcEsCNRZ5RFj)VL3o1BZIM;ysWhw~XjJ-S4I
z>x($`r(3~yojRrQ$Ia6G)Iq9E+^f?6<o<2d?`wa$L)ZpSu#;CSF)WF>I<YO^5PQw)
zyxgtccE7$yzLXx6Cf1n|M{7D^btlyAPz}*O-fZI1Oz*%v-dxD`2>TQbCzuC-M0XBf
z)b{wUZ|OT4_o7`~$Qne`cnn;-tUB)~vt20LOnn@<Ld6}E4_rU|Chk7%e|EiV=rtL&
zbGN%UYxHgyoVpFcZx2bKA`9Tcd6G681<|uv63+Z`<Cy{a65IsZx_-I`BK;ctkn%ox
z1QX+Vw8*H9t?~{2V?B}2en{Iaxm{+lxz_e5=eBP+rqpIbk;$6am7lP?t#G)bFeuX9
z!{dVYluRl<b@F2Q)+~ZziZAIvD4GD2I8JY$Js^yfXp=DlP8c&#!il*RE`Vg^5-m2V
zOMYH}UH2qP(7plLemH#<ufDKp*0qZ#z3a77kfUjboj9(kEjE?!=CarifipEzKN<dw
z7(}b$=I|o7{ip1j=1rKbi!yQD{cbGNXWYYMAbYWNJA4P(3%eg)j_1U*FZtJB)!WO#
zW(d&qlMr#oThNY3;a6L7S}8RN_FzWZZcdeD1&$UJB`s1jYhgXfepcVHT&f)u%w)_#
zt$$6$6?F?($scfR!{HH3>?G`2giPb<`@_%mf}<cBixON~#;!R^%Fz=Sk%C~&!{%i0
zcY^FX7YNLtm}ldz%81e16o@8Fb22<_oGk<1<6F>ws|w9eU&^+*l9ba~Yi;@!zr?Lg
zEHBjPSYC>Y?>v#i@Fpp-9c6|07oeuAKo%~xJYC+Gz!+(qY!Rn48LajaB{{TaI`7zz
zQ`eT{=x$2eJ50~wZk;>tlHu9Vyeo``PGr5%isv_1rJt%*yynUHABb#@TD}6h;11H$
z58@b*NSb2<AfB3yD@b*qM-zj%OIfJUFyD#fRi=VSl~s;lJrytq1E6il>M+FUf#AT}
zGa!NwZB(q1|JkU(P-#F+6Xw(o54hU_N!yiRYq>y!u-*8LG~yyixDMX7T=4Ue#r)lw
zNhN{Mr-AR63Ee4>MV@>S-D!2d@_hQ~xEc1RqOUhTPp`rqzFdpHM^<mCKTN<oAhK_+
zH*W8<w`=c<dDd&~zP}R7pRaj3)p@SUv*$uzP`yB{D}$nAXxFAbRG*`^fb6O>7Gfzy
z2+?i`Tl4Ft7n13K#Yc3J@d^77WC0#BI?0bU3;5XJmv+q9Y89S#zq88yc}az6mX*!)
z#h(s=Wn1^;R5&FbP>YLyk3ykoDv6)RpECkbdhK94BFn_V*h3Q`EEi7iSbCRZoYqbX
z1kb(Qj+cs^x9|4bxwVQqL@(ptj=wICK)Ns7Z1(50sXeZin>p(as<BuvTkH;-P4|<x
z%i52_9;dn6uFi&qCpq}7<=7McB{m#J;eF_x-+ek;zmE&*XYB})CM|4xy%wr>!Pfca
zdF7U;dXWo_X3J5t7rp<5sP%#K7erP}wmHQF^^$fXd)%P^WZ5H@;xEjo5s)5J;uWKL
z5ADA)+0)ZXhjZ<f8NW`ko1DUlHz(P08s94_8f@NRsWrNb%3PjdTM(427G+J4gsyOT
z36y^4@%-9F@m;FJDp*F3DBi*RdN7A1H55L>^oX}|pvH?pxrO5}-iHb7pW+VGlwSoD
zd*Mob0%y2kW~sdf-{s!HcTiOIPh0Nbg5sPIp@9DOuI3Yl34uV{13D<k08?9vqzI?8
z_3dk~=%PA`Wb5o`p3pS1!F<W-m1YwR{c6wEe`5CyN<i*wTg3_TJ|4$q=2?4V7dF33
zd=C6@b@jU&_E78ZjAGA<qe0qtzE}4Z9;S!21`Liri@u5Y$t<J{SU?ez_5K@62b5IQ
z7;ux@BCdial!T8c1wqcB6Rxmpb)zG+gSMHWcB;uZP328D%}0(EIT?a%R%lI^ckoRb
zaHwzwf?A@-N2M^k2SY@my~k|c21G#mliNK`rODxyLoZ0Gv1z|rY&1q5F0yDhSbDtc
z?=Rd><qV$)YtYQd{i*^%^Ys>yV93=`W(y@3l$>(wb_eGYiEeD(x8|7k5aGs84^3T&
z53rqq66Yz33zQ|Uu^><}IE&91l!7o)lrwTrVo_A|;~R`9-Alt1`ls_W+x+khV!;?H
zw8;Q37tuWa+bI!*yl5FFcHcX@8k4X<Y0r#UFWL1z83{t0$$Rt>VeD3Pvp0oMEO}OX
zEs@GFHFq~$JCBR`bwc=~-VN=d80|vINPcYfh`UB^x5x5IpGg9}NVbVxrYS5U!w^D5
zHkCE^%r?J*2sl2`NtkI5;Zh?fK%6IQ<9i<MyZ*Z{`P=?#8gIk<;PCy5XRNZAz^$@Q
znl1N3X;rwoU~#OQoYa(!I;XHf%r*UTei(3I)(Ke4jUjSb_)hp2NL%xH>{AbOhMEXO
ze9BX#&Yp<eMxEaECf00>_~iuzIXzj^1;CI-0@Bc=-%hsKJFBy47w?ZdEx4$CBIYC^
zQj-oPCra)mKdjK3fhJnN(w(DQGLQ+(AXpK=h+KsM&)Ei(oSxX;nA3{8@43tjt#w6?
z(s&#(Z_;rUzK=}iWJ+?EJ@K|yx@UxLYNSo4FTQKy+LV+pcEID3VY7ca5Uf0J!$V*9
zrL*F>C+No2fr+z<8xca^>}Uch^w74}m1VJp9Sv;3K+z?KU;$AX)%K7}08{ZM239`-
z*TGShQqYr&4G|J5%}<6jl9xd_!kqET<;(BTnaD4D;|<N_Chvt^C*|S!-kHXY^?o^@
zpPqE9SZ#j&;_xEB?tJNk-){5vJl>yd(~M-q-L%{M7g<z%^|Cw7Us%a_wb^DA>?t(v
zx!qy*SU}G!y+7?S)m_dCVy!YNkjSwT^_WI2WCknhp}>+Zqa`znniN4=SP^0YSOW)&
zjfAdE;WLPB5fB^=vzaeJe+1XD{=}yrvQKZ*Le<J-t8oNJD{6mj{8TTdl%7+aca<H~
z$Lp4m<`uS*d(58z!L1At=Gd7v9cu)}&E(xAA-!rG^#d;R88v}l2@g!`Y<a<Bp73g3
zS}Ul?cYC=_W)bCtA|mjMU5W=QvKu43hva>2vM>Nfre)`aa@(_Ayb#M}Jptw4g@6sb
zoZ95*a(=k^V{T1mer)s$#5?^al9;yRJ2|_V8}a($?Q4q*g3@0@Ian_KJ<a1}+bq>u
zFbOD<vuc{um6sTYDAvE661Z0#bEr`?sD{eq0cau({vp;{*jmu3{5RqIy`XS)jNvpr
zU8c4g=tG>~hLYqW>ut`fi$B_EL5RMNU0IxRGCoy!mB&jwM`!z&-83-i@>gGnX?k_m
z&2h2z+rMM^?e+Tw_~p(gI^pfMyG3ozrb2_D@1=h{s$-IuRaa}w&#64}>+U>n2d$mW
zoA)`Hqf$G=ku}1=Bbx!-qBsn-fSVe~KXZV4$k)H~NLrXVnZVZ0@EwbTua3d&SkkO#
zaEM2g6@Uy2h%R!%#)4ETqEO(0FgzGStSoy@OaW>b0&d<AFbonH2<~=H)Il`tq&@dm
zpS)i;Lo0MEf-K6u-C{a_F|UQ)*Y-v#%kT!|sHY|J2+y2O!(AqC>faHib5<-nkSl{p
zv<!<;7WZi~RTeiqZ70<sx^Gul6S}24T)bXbr$6rwrPf@&pDD1aQ<%A%Y@`NIu$@hx
zQ%sE2Pjc0*yQ7Epm8EdCsg4ATXfd6)?dPw$-ile#7AHIMOrLA}AriZ<JLnmTZ6!j7
z+1eiMtb{HFBOeRllnurT^Q}A1X>#JP81ZhLL8M}&+C&JU>GdSX?iqG~rDajYkl}eU
z)Wl>O%opVR7UBjuIF*7z846H)`7$tZ0YGpFKAa=`+^7QKHA;m^DPlDItZ_o%VW$QO
zenkD%UPseo#n)Mq4EHf|9=r?`n~a*{$-bPnu`dXglx+Q-EI6aWn&(DI(ZqSDs-sz)
zaa{e(RV6rZXId_0W@4r~SHw;-<+nKXJUefzJFye^IWX>-Ys~j?sE{Xuqp$%B=F2p=
zSB~Xq(+~=GPBBm<!f+yBqts(kwBZ#ntjZbWC37%n)ljQ+?}Em-pfPckL?g7v&&VM4
ztc&>JTSI<S6E?Y4AJ&x4W;*1~@QJ9i>+K1_IAgt1D(kpugVj53svRV>reFkUXEA>x
zb`L~jpg+OCaY3Fb`+3QqIr@16j2fHghu0O=&j39&#5mwrL}Lz@$F*)@->vXA&j6TW
zzf6ZL|EX&^E0us3Fx;c06X?+BfDy4i&W|l0Gs#jvApE^p@o<?A#bDxR(N(MSbMmJ4
z$lNVE)RQ=D#8E5b6_!3j^O;kS3OsPD334bpQa0^`%2u)hQ9x7w$%HZgn(B5=7zf>;
zWzXym8i?~SlV!@)V`U20TihmE@Mk>9lIC2S;lnd3RfogYKi0^5QV_j1w6zS$aBFnd
zr@3E>TVuDm;xtaQDr;>u41cYc7U8qB@&d<nI>V<^0^vF%8)BeG;_-oOBbnd0O*K@a
zdgYzaJ==U{NF}30=KJk{+?p*X;v&{YTvW^#XpVb;NdfaxHRC8cWofGrcM^*3%%0!#
zw2LOg3hhjnB^oeKIoD*Tw<yw?Lw+!&e~P*(NJ65D>mRTD(pFY%BUbD6@jZLLGvQL$
zElAh|qrysto}+^1Oay}KriXHW^?LNa&6Y5A&}#vJW(It*1EwwTF>%-tAhmk5Oxcu&
z3P8f?sI;EhVL2Ou)4W~v=jSBlk1FmZK9&yB@mjtL&6Bm6t!=fIKY|J~^Od=uM~fD_
z-)Af5adR7mlQvTuUC%qmjD@T6tShwY+)g`MmpAX^-){dkd{m~>zJK*}Id-r7J={xL
zjrX=2E_$}UG*R#{Z)Vc^qy2f=_1?DLeLsPx#8dt8F`WDJ;k=U{M_g&MF|mvNnQ8x$
zjDWoX4F+GcOw=z&cv4u=UXTunPT2s6try8Cs9hq`2ZK?E5Gu&Q=lNcW2m8a3&9mG6
zx~JE(Gei1SLRt#BO<YxQ`*4dN=kYJ62k3H5JocV&H%>KSxK1+N-IKq-)TD0m*r0eK
z9XP74z5Sm!6XAG)rG84ZI>1{QkW0Rdfeiv11bl@pxpF{qoHlH}pZeg0{FDzCG3eOG
zAYn{>_f|E9{rA4Z<9hcUvnJ!^6^M1VToj14*6SQ?qIBuyS2owvpF<F?*lVci_Ljjj
zH=9X++6MVy*ybkd^#-T%vGee0kCJkTPdUJf_`$APMh&Xa5UK_*fw~WlMgXDF6dPYG
zddTd)WSI<{$)F5$fd2l5lb<XoZHz3vs>kT(^~Og!-hrg2?JO&twRv$BkLK$J(Q-m|
zmBLq5$B5<1^C2>H(nLvA<Mmg+(;?#_+N~^3qy7nDQ?!r%rcqBV;=wdJY=03^P$neG
zeMBmRSzUf6`MKZ`0rILTlDr8eNnYlQe?-_W{>|SUBc{)yI$_tV$>yey^M9LT;bPnB
zZh4#)AEInU*k;)skIn65xsE<)8^_cD|3C>WI3SUEt^BOZT14MbF`)c#pZ>%KfnLxc
z5RewUvTztc>*w0`{zf~9d264q{Du;jDc(zk<qo6d)1G6s|9Kj3Ud_2tR}ou3b1&C2
zjR}TR?{KR=$U577&#~Jay^YyXPrbi8Px7idC6hX(=e2t$8<<;578+P1ttdEI3jOa`
z2;3x&=)r|fn?XG$Nm<4-YOoTFdx2FG-q#B-zu3FII%0@UTK`#Yq|E_=9240Q5jLJK
z@l}ECi%%xT(~07B8krYKhu9#VO-GCEH0EDAGQ>Uw(NA|qL3~~|%T!7ZibUZ!)QT|1
z=su+J?}4M~WNj3C`<flq<>lX=To^7H=iHJlt{@PK*xbJGZV5174WDAJuQn_n^Tksy
z0%&5tADaVnD5HC<EpsRp$D<&|r+U1yCp*wrbXKDbEO%xJicPa1Gr~>l_=m65ovzN^
zb_c_+EH5sS@b;A-uU!r=*NABT{`E%1DBSH!=>towFvIpC;B6wlbN~?GD)U9^aWe6O
z)VM{hdu;XqUyYuFEA{?@c(NUbV1c*TNrK&woO2P(PH3Bu7Jq%D!0lNTHrsvVr4Hf)
zZQniU0V(ESSIDs7$v>&qr=y*lB2LM}TxjDfQKXD)1^n)i#7v?#t69lW>x@I>;qM;}
zQVK8>aiZwOqMpzS(Su|<gSn}Z@Nzey=HlB{LB60+*juJ}3<NAHY}OZtT1}spJ2+=+
zRoxw@PCJBSc>A$h@Mgp@olKnCg)YO`NU+?J1o+s3GUF#jVkVKxp`+kwPe9kBKIFHl
zR<3_78n<;8FVm?)a+Xzt;5q^)OD*O7>WQUDMun{_!lK(#vLs;~;XfGmVAT+r)>tK^
z#ZGHnVXezDBg)OZ*5O|V0<Lqp_46;b>OB4QPo54whV}M1xU*tETiA+jqOc~@Z;f>J
zy7dnX_qvJ3arSaG*+mKd;=7jlTX@{gbZ%=Gq!p8}-T~S|57K}<Ux*fxs1+Z0n`J8F
zkv<z5E^@Ssv_S|b>(OSBCa#D@BY;=j6)i=X?GY!^@wJpLthE@g{rIs5I!z8lGjI!!
zv*sDIEAd*7dzc2{Vn-tS7s1$7G~;!WyfFjh?|5y#p8nr|iW&iWV%E+jUUFIg7KgoV
zsSa$xJAI-45ZoNdH?Wo%NTV|nqJB%dn(e(anl!N?Q!s0EfwVOAQ5>0Xn9GsFdUkG6
zVf`ZwcXeGg>q(2TKpZUte?-V+o1hHJYz8!ny2Y-dvqanxe2U}J<4pPirSN7)o!)3m
ziKd?}LMI9*I(M%)T@3!IJ|CSWZF)+A>lg#e1i`Zlk9Y%V-snoc6~ubEDhC`l8sLg9
z5E-Uo^-c;$1oy}77ifUep{^PN0m{d&;)Mx@aLk}25VH3wuRiK9nXb#R>sTRvv1S{y
z(FL;>hml7apo4qd8-PcUBlbpmv5WhHiLHG=cz3-I7wV&Gu;-S#o7$f|M|VEec&9ns
z;=n_k(Vk~4FgUA<mt2=NqfMwdZa2oRM#2_aIiI$Oh7Vu5=(?&Cbs3vpf<5MocU;{S
znU1Spxd|qU+r|{OPBdYQBlYQIfNZ8h+Q;!~_!}nAK$9Ik<U}C*;lhIfBTXlSap(td
zPJQ#4Ud)N&;1qK>AaQ)N%nb7V?H19)Oold63=%U!po%I$AIzaA+2FGpm%`NJqaaUj
z%_%{YD+Bu1u+Faib{m`R;B^<g?a8lweEufg@iFCIegFDd{8jb=vFL#C*Q5<SeEX%*
zQCALk;;8oq7I6m&TaHIe9JA!22r02GVFoEiDYNZg@h#{0&hKf4)=x&oWa0=C@!v$O
zSyjn*?7cE)lB3p88N!O}bRiN#`Q`U>e$i2Nyg0Sus=}5xu%R4p7cVQ@M!W59q1B1x
z#sSY3b*m4BT6Nf-PlY@mlRaK~Wr@^A`36DL51bxF{&-#A&%-y2s>Q5gewuT5i{jwz
z2ApH@pvA29pp~KfyYlg?am<_?QRar@U^p!oNbJ5eEkQIzyRBoNIAc<V$`BQv;Z|Jl
zf7<+uk8`;pG^7lhgkN*I4&&ZfC?{Z`Q~yzy%QsfI!6IuF#B^0lwD?^WIn&4u&GTrL
zHc-~LaBVR6Sk(G@f{eEYg20z8@9-El-I6JO!(1hkQ(;IkMCuDo>g%+FuN$M+dGdM%
zLZ8C5T{0K~uMmRRw+cf<fN0zJiNQbGd>N;Dl`lS2139!t{YhIjfMYzDV<{;mqm$TG
z?P*+)r`y?BxUM?Bhj93a`{KD}+jlfy8_M2VD#~h?4z<5NO$NXRsygsnU-du0Fh3R&
z;*9W4NX=lSfMptBPbOUEXH!C-`ipJIFn*2`GeUCR=ha30>9?btAe4EuR<k-R4C<c7
zNK(%=vkII}Q1{KW&JyLZoZtQ)qy>5+2E2C=UG!jpQlj_ON%?0o43LuPdxh29ldpVE
z)IHJywV(li7#Y^vTtComDcIJD9`7T)S=wwNjCbEw^pQ|1>iUOL21}vf`xpM4eVF~a
zN#|C1>n&h)v_A82XswF|Pu$k)I?ZSN<%Jwxa#XmgA5h{{Y7_wu$cOWi{${>}iQ)nT
zqwVa_LyQK?^JXN5Cz4MSzKKXP)fmGzHk~w+zsnd)g5l6*dHaHj;<UyQLB)cM<=I8e
zkOmU|;vi0YQ;uIl!JnYGu%o_+;wEfwFmM916HY=|kgrUkTOFTCwP8EhLqd&(xb#Rd
zP|f^`t4SLK^2EkU8%EUBLhzIQCz?Z~f(uhn+$@+$=eThs4`0#t)km7H8LySmc#oLO
zIArUWV_jM{YugC#xcOjmEO}46hM~yK<7jrbxM?fC90oft>=>2yvibg7Q`1Vy-u4u1
zpQh_Gh%KBw-GuDhsHps!+dhn^U|XcFH+)I2K?kL7AiyVlO47bS(HkU#3?9mw(!Sh~
zLzf?$sgBBEmp@bW46BupauZYW7)3uX=}tE`O$AYM=nkxZGjleWCRU1^C*8w6pt@Cm
z&7a5nsJ*@$i*mtw3g<6_ncKHH9vPm(i&3J&bLYp2J2k9D_i>-rVy{-~?6ost<O-c|
z;0^=Ce7r-n0j*WrEVx&97NiqB20Q<x`qb}gVQ6Su`*&@R2e;)&j%)A7m3*w_cB`$|
zaK?wo_yvyo%F62Fo9wl9ngXW*G%a_qopHFi3x+tl{NMEikTmh&{M>zU(;r}ejvT_|
zbbrrQiLqJ-80Hw_F>cM|HPO5G26`c>v386MAo(x&pFl7Cp|%lK(^2ZzI5bAKj&^Wb
zx;*zd%C9=UZq<>7xq0rPJ)!I_UPa-4i~gu)nJmu~2}I`k(2)aimJ~e!f&Ciw;Xv!@
zmZ+z#{J=lyw2BAOzq1lBkR4lcWK?GrzK24%!DeLtqXdD}icb=;`t<jYJtaxj3bfHc
zG4;SD!^u3|)q^<`-E+UO)5i)|&1ow{Q?;|#`EK^dJ7?8Yjq6|$UM+0Q6O-eq#Tnd+
ziKpP~*tHL%s7DP({A0T=4q6EleyA>yMtZ4q(HnO}d@WG|2Nj#@7@j=U5@dMqcwW3>
zs?T;{R%B=;F9TuDBsMfYq2q783=F0y>6MY&^{VTP-7JR3Uzi99@|hnUZuGY+U2{!!
z(OIUK-Wk^l_PlU{!#qGOn}S`!xPPSM2e?8s@oM5=Yu`h1tA!i5^g#QU=14Cg$K4}5
zJ6qv5^!t6VhLrH&#oZ>bM*;!oUn0{ufo?AhK#>5bV>lOzhH(hMBks$Je)4nQE=!F-
zH~cgNl%TLUs&zewePr`xi~Y3bGaczp7FSJ=rOuZ)x9pxTX>Tb-fHHhY9X(zaP84pq
z@i!2Yx}a<m2S$68RY>hjy^`%d^*#9FiKZyR-y&P`@DtO#QuL=8?LIMUxnJzPwMFjd
zgUD&dc<t_w*k>qjevZRenOkjdl5fo4Ee9aQpVeF@-^6$g->cbgbO+h5fm=A%)$Y5`
zm%*mu*>~^nq*`B<svOIeo%bAAOV@8T`%igy#I@pK{O-qF-b^G;sg%~Z-rM(Gwev}v
z9d}{ZjvJk(U$E~<ICF;<JK$CA*Y4&^o6bBlYnwH#(7wlm1;&s@l@vIZBg9d^{q3ow
z!no^I8EvTv^?__O<6H+!#4v#^S3!uuZ?C3OrKP{8p4qoOPkg*=zj{8=rrch(d?PoY
z#&FO}{-xH#2H+O6`f2JDSEfzqvN7dX(Z&kOGOsBh56lL8b18ArGoa5L^omfkmn!M3
z;7;ir#4@NHNsQn91<JGh3W|#Z@2;13rYf$WBCXEbdkFmK?f42+5r04O@moIjn*Ic%
zn-v(&_WhO`8X#!ROIys$9Jkh}&Jqr7=B$bS@_voR+C^dg@DcYOgmn+blA891*gzn}
za~d)9=!Mk}%?(b)3*v#@Ep%vld;4a<F>+x~mz?&Z?<;3Yn!0(rD7OEC<%aQeeA1P3
zGw={YrgpgL$#8z##(i<Ex6#l?&KnJ8h=Erv(4BFflb?W7hYZSnt>}`V7st&P1eYW!
zAWU+0K0BGNmi2qcL>W`T5efKWZcxUkKA794;LFpD_JCa8?==!n9ZKbIqW_DLu~RNd
z^vdxmG$YL&k;=Hv&Ourx$^eJ0)#`H{pJ8U&rc?qznlv^>QDRd<wW;gZZAw(k9&hC)
z*(kuCA+eevaSFDu%|n#oZJ;AKQ!JyInf!f@Fq%vQTzqN$^;!V!#RPk}Gu3%C<!rtE
z>&{r;e*D>}iF9zc*}MqGqTl}!<cMest<O!M0ch^9Zb0z+JkRxS1qK*X&%;e%X8!@>
zngSj$c8Yygu3OX2j5$l9iIRWNjv)qplew|1rVXkpMV}F9?gTV=RH^x_UnqEH7Q?^B
zUa7F&DXOD+g>wzFqGnmCR}IzY11t!<x5;WeuM^hqoX&;PkK(zakH{ykZ?A3no8q-6
zF(00R?lFtS@J6yuO4F5N-`Qb;Dv`gYC%(S4_Id%SFItlW!t&GR!)N+N8|_T{+_a_}
z9LyGMKeJ@G#;fzENw1?nH44pwGk-n7CKfpK#dy2_IE!Bj9F`+@D|46Lz<NwjQ<Q$q
z$Krk2@^kf*_H+W@gmRA(6X_bR;=pFB9Q?`=FS2P<*E9Ad9!2KEW^5h^pu0Z_#}I59
zca>_Wnlx5j*3k&Zh%~nmF^V+M(0*7<)X-?Gl$NYewlo)t-pB_u!<PS30c8`7pu$O=
z>}f0o_AwNq5-3*#U5hV8Ra*WFCMkm}Qhk>7io<B9UWC_%X;N+dI&{~>aJcCUI_G_!
z3f~H1(@EBvK}0yA3{Faf=G;62iH41n#7Sz%gC6C@tcB9_IQppKbBGb49@Z+Th{y!!
zhhyvZ0o&*my&_(a_wk}QvcI4APj<VJklz(w45irX5M;vWv6yUsObX<vzg)Z$y@Hxm
znBPEs;&HmKY(}yAIx<{-Z_h(^;li^zD|zJUeC!>1ni90XGxH>V^6;13slPY>W65l8
zow=~omczB6aBS<JRdc)))kS8vk5jW%$LPJY=9n(R?N&xZ((Qv3F>$qA?&R%<2{G;o
zt>T+UjD?`(fs}-3cxtW>%D5^p1@0_vhwn)=fL#fmpBj$Yb9St0bu-X8A&cmz1GCIA
z9st`%<=X^3raP<6{vJm;NUUsW5wbFi;&T1D<Qp*B;;>+Xna&c`B)OWAIM`Um;EXmB
zWcBq+QApQ9p$>qaqL(Kj4FetR=Mx7L%ZE`YC}QZzB3`dguBW6CnY>1SjgA64!A!Ni
z`1&3U=it=DGyj%(Ju7B4-(w}n*3-GgLwg+y-HxlIOP?+K;^sq4ttHk?qGgsPu5Qwy
z-kP82o%j70@A7&e?V2$RmlJ2k8LKe_=!An)5&m$RRKaa|`;9R`Ec?s7NsW?$X@lI}
zjp{ehJ0m5=Zh~l%(#K8&JCIm0=IMj;H;3a?PLp$cVzgSKbRs+n$i(&S*O&84-Ir8X
z(YK@zqKD&^zK^F#Mtl~>RBH7go25h9-4Lf>YI%ofu~wBwg@=Ijta1Xg1J?X>H05T6
z(VHQ0^6si{tLz^`cepEZBb1T@m~on-55dNk1;~tPnt^)46{E5sE7+FuefuH_s}4?f
z#nnp7{?E<GblO%GjBTZ+lcql(zF+ztp5JHD&(-qJtf1x`aLdfGzeZn3M?i2C!9c>j
zk$y)?Rk#|q%?PA`*-gcmgkbU5__gcr5Lb~;DK`zhZ<`5X%J6TfeX=lwQQ@n+V9-4g
zJW-ttF9jBk#g7;d@-D0*maarO%B1894KE>z{$9e5M@>rOKe-gD!77rIA5sO9xK;2I
zX$Uz(tuKJP7u2_+im<D3Lt;hcw6jVcJ|AKbX$+X8-KZx)<#>I9ey^s^a@hn!(_pQA
zS^K2b@SJwk*Q;fIo^d{_r_n@Rzn^(%q^y*$GoDH*!N$W&UB_LtZ4-GfoV^{ZPN%h)
z&N91+Y%Opd^5wuvnrdd~$dXB#MF^jF6E%{De!Ue?u8Ye&98MIoYV;}(%a-+V>4|2g
zIyQMhloHbus2W&172z*2Xibv!fwLAUGnmyLMTs@v+2=>$VQ!%%hzN=H^S9Bs3$xRe
z<g8M>wtb{Z2}!ZRCp<&s2%(COt}|me>cN8Yw%dm44nzMA`xLJ3dVZt7r8mD`JiXH7
zAo;4crlhP}H<BpGM<hoKN^qUMW!hk<=W-W=A|Iz|TrA#2TyBe@+1)xQLSi)~>UkA$
z#9uU+VnPqMH>>>%QuF|c84f4Aw7#2aKVk%}4q5K~4+Yz>RkQ&QRcS+k#d>WbBQDL)
z?Qe&JnY&d^TSyf<z1c(wz5NaSl3;_^0_&;45=Ts=tGq*WEsy?ymjdOvuwuK6`&fl)
zlZ>^K+J32H%`H>s1DyWyuZr4!H4bO37MbU;>ZYQWlD0Pol<&GFcQIV9{Qc!~mKaVs
z{;wQCvP~V+d`SxRa6-gIh6RkBN<A0KaWtTOju<UK_)+rkP#>pTW1s{L&Kh4yd`DxP
z45jO3-S7=DOUfy%_Svps^i;WWI}I#uFUHxZUbNEF$&m1e6r5LaaR}GjO)M#wTkpQ1
zqO0^djH;6Jl+Hadq^4>#i$5}14SIhmi5@OYG0U9FLJrHFX>Vr^0Rp9P_}F=XrN_fV
zc-IHHtGF?Hv9@jH_0i*0y7SSxhy3R=*LI_OKl9bQcQ3FfkLw_gDc0eYcYfz3|5bH%
zVjSc2)~6CZ-kH2Thw;|o_RPB|^{N;W3R$yxehU0BVdMrGiCc^_Z;P5Mv%Wbud3$}M
zo8$$b{W_w5r=giIp^^2bO0wMVu;Ifw;91+1vx%w6TV=X`8``;0#(WpM;^X`)X~zR|
z5R7sailB)Rn&>?)y}Wd4+$rPKsG3?ecytS+6FN%LLt)6Uj~q!*v`{>Z+>Z-W)<-Nf
z){0+7dF`Nvol+TZa`MNl(CHfQRcY=8^Ub$sb!(O9N}DTACVS=A(ivI@@^J54*6sUw
zZ2F`cPJ_)B>fId8`WM(d`StFnm|nG3ge(8}>|jL`=<^t!=oh14P8iD^7h6dWfw9i}
z`rFLHB`eEmMgqZH>7i5zpKSa#1DCSEMz+F)a;AZ@*EDZTQ&;rs-0pAXGyff>B%VdM
zs)7_R`-nec`dLi-qq9c1&Tk~T@0+V7YdiLqtvJH!vkb9Ls5CwJo$GB!3=d#ct27u*
z<_(%V4Aq+NMN;Z7H&SaYz-^;*9JQ@DqS1T~Jq&!PTcvjE%#$n><N0Mj=Vi&zy}Bqw
z5dr=>mBXrkrF3&MAw0CB&S*Nkdl+H}7h@)TI@I#5t`}a=&k%p-)92RraUDbMMqtOW
zDMuI%gp9J~;irP&P-*Dq$UhqBU^qYb)SXsO$#CbZGHPYQHQ4jBfw5EfoK-@{FRqp3
z%_5+z$WwS@ptu%ox3wk6Z^VPtzi`~fnsD8izjE1R88M_w5y;U|DTBzb_0|0<^d^lr
zrRMdZF2f*OQ^6#H2`;ozF-un8oq-@?6GpMd+j?dwPV=>GU$9ZgGI*CuvwK*D|NfZM
zPMut_N?wZsKKCL&#J#+2v*+_V-1+0tNs4OildOLun`UsG%4rIP-LUR%)YQ!O@0SbP
z3P`4pwJwMQRhCcuj;s_X(6;}dIX!!Kdpj&W{k?b+Y18yo?gpUDtap8-5vm5Cmy&^$
zHif{b#d#7eZKR4UHv$IPQZi%Jm_dt0e8+%_xiN}Ika#lHg`^;b6RuMo*&CrS(51IB
z@V&;MQZ;oLbKx5iw1(SU_oW62H@o>GYcwm3dYj8a`|28~#@p5W;nz-QjqansO8AY0
z=F=v@_WCd=%&{Ucjy}e+eXs6H&x^(74V15nGDF+v$X?mZ`tue+<%_Xr?Txytq|td2
zYLoJAI+=ofi>ST4G+E>^G|zR`#~>_cCyDNxu<(vV-wzLADJ3bVu3@|qE6Byk5YWD=
zaBFDE($X=?@uuYAB1X2t^^whTXCKY8AxL%>K41BWNoW;1>yR2}x5%`nr-DZcF}Ea!
z`8JG{U$MBYG~)7;&q=I`N-T-<iHd`QC@6wmF;^<Oii*0jmXXq)RfcS7QH<B8&F9d)
zc$6fM2kJz!KYwp7Y$THjm;MSkpMCZ{rV93r2>&@ff;ZUc;k7Fhyp~doPy)$TsIR)x
zLarw~Ku=2Q%i^}YKM4~X|C66e@hci!Eo7ijtai}A?(h8vF_B90LPnwgqD5gwf0W``
zp-zu&p)XQq|CWtYR$YGZ65iI6px*1<0NOen<*?fxgy~m)^z4T}L?S${yTYr^r1W+7
z<6Os;`O?*9{Z|lM6I(M^kEKi6(ld?sS-9o)SDv#gEso?dW$=%IXTlK=uy-^Uiw(r(
zTe9u^9{gp;XY4PWNDA#`TXwj2z&p7KGLdMdCyWjf^moYSuYdRRJAq4?FY?15iRPc{
z)}QX?>LXKNPEsuvhUUiiDU!X~A3hvj{YJG`vncnN$>iFf(<1s&bBKE982$%;#JBCm
zw|0S@MYck-PzaU2Q@du3D7#9`)?kwxB=FQfbM?vh8d3)w9Y0WGT2_7L-OdWF%tW|Z
z=Z`PPgxW^A!@EqCyZEl_=Ksi_hAAClNV&EpYH|5nnJL2Ajrl4W!BZ1f-#pt>hGZoW
zw;RT3PP50WwU^0v{{B}s%7#ya^b(|+(XDw}WdFUorN4=wRYnYZz!zu2Yehl-M2t~N
zLKO|XRfZ85v-r~6Um8}hLctPI`C%3!Ru(Fv%TZo3b3|M~zl!d{I+nK2DcEMVBrQ+l
zXH0a+#=C^nD#R>bD4VCD3cDl@vQh>Yr|iXz4Ur~w?gMoL9%~G)72-ub-s$Xd?o5+r
zXRsA7ZKqD7>E!Xb%%Jhju@QQn+(<nxFExW*e@C84d{eJ>ZTN00!kW6CyFO`F7kn8k
zApK+~MP(^!B|LG>MKpu#XLUD`%BziMuYIdpgZH;yk3!d;9Mq8}Fy>*P<_W2`GOFkV
zMZtWfd@@%?GpjnA<@R=~@x1GO)6x7{O6_(6>P6eQ>&q9SO8c<~q@1Sfb|}1kxk}M+
zgb-;T@1VI)t!|Xak-5T54BA(;NQ8fQ491Z~o-B8n*t|Gp*oY(4C`1u<rPT9ra^bqr
zbn=aN5;6Cic&W}iGUp-%C7s?M*Q{7`4Eb`5k}2v1dpfYy>eC?SN|s2sx1h9thBVQ~
zHX_Rfu<oU*ZAKnG{6_a>HST2YImSQ4a(=s*PNJcKQb(J{>}F`l_gGX~-jsgF3G2I>
zUqxL+djkG7R$2OCw0|LD|8MtUHFoZ3=cG~jXi3O#e~bo_nkuE@P7~tzm>sCfF2aTb
zp$gkSZ6Yx&uTGHnLPsMo^7pSgC`le6`I$^0JjC?Mkcs3nt1v+cng17HD4*BCk0lq_
z2cK}cwQy#Jw(8Vc2D`6291B);uf6uNhH<-h&!2hCWxm=?EvvR(yW)mhe{j>-bzJAf
z$n5EpW?NfIE}78SGNUO_vv~H@%l-Vdo!|IY0j`A>>U`XTto|xB3nOcUSr{1FqJ=bT
zjdDb-`$AmoR|5QE{t1$c{VF2HC2^1r*7B%PQd^WIDe{;xnX4vqav1?*MF(ar@{=u5
zGB!7A^`sa~XyUM<R4tz=4~qLc>t-#ChA!?L-Psx5-8Os6I3=LpI<Ter@_1ox{o;w-
zhW@TiGh4@AGH?)O?3oRGE2|5gsznscY`R6l%p<9q1te89(y4O-VcS{tK1yD@Oq{H0
zL(TJ<wLBkklV8g>lrK8{FEoEv(KmyC{SSV_bHDg%19!k=d!6+q)2-wK&d2H)!7#@j
z(|lb>d(sN^hYfSE46}?n)pkvKBCS7@>IOdM){sh&*Ud~in1Radnsji4CF@BRXK=D2
zAFTk>j4esVxMUY)LB!mNT0W--ho6xTk`qSf3{>~ttY@y;^~86L>uzb^)Y<!P!q`1;
z_D$yThK>obMjh+7-o=IKA$=q0tWF~7SxhWJm;fNul$@i0J_v_JkP*~e=n~lXy)=s1
z{?sm!qdHlKaSr`ao=Cb{DieLG>=nl@7<SRZw&S`FFYNz}dorfVp;k`d17ORTh>J{7
zcT#o=m^qNGf?6X)Ps(Pvs98;LA{wy9&2GkEB$J6wD~blapr+=SZRW89_oR|@#X{?w
z_w9fE#QytkefXA}9(ni-d4&5NUGVmAlmow`TlY#&@7&w-j0#UT*a0(=c0i2Q(-;^4
zW!iH$-KQ+K(>IkG&Pg{b8|?Ha%7jgH0Y~SxDjR1g8|TyM9Bpk?c28CA#PjLMF6Ax#
zL55dcp%*KUkbxv?RCWm%fdz?xSavx;%(`8y@yP;*237@buooW<<7l!WdyzAAFR;|q
zp=Q|7h0BUyPWZexM|N-^mdqA67K?#Xgb`=AOU-ELMc5FpDzDfEVnn}89kt*N)Z?fH
zwxzkcu6d`Q;OgjiXKm7LX;rSdfVL*L`Kzn-AARL`g|iKO)qLz&Gu3+XcPQ`ZuK~R9
zfR}0_6SXA0I?RSQxmiOivH^xnbZp>;PLE5k!CGY0IyksVAHmQXb|v_%eF>ifzvinl
zg;?!p`ug>DNeAmU;`&Xvej}`3XVTU$fW<=1xdoNgXtDT2#$1)vnEoTI8N0`vlk3Ac
zlc@mSG_dg{ID(F6IH4n`q8$t=a}ht>Ptw&=Vd$1=s#+?Y6}j3^U(gTqeS}ETzpbHc
zL%NP?Dc5vVWBH8g%L*r}bk)J=3UyQ!_)23ywkI02$Xh+o)u>3Kz5(%39|dA+sK>*G
zYB;O_G>@dydim0ZEG&sYy%s-9uVJPV3_mFrkt56!WaAM)XgU(1>INjAh$vOy&<Jqf
zr7<j-iwe+@RDc!#4Fz~M#hqIv)@LBabF0bz6K4}B&J%c_YXbpZOhROx*2ReZQLL3U
z7Y<5M=TvEaaWzvkF2IV5b^OUzTVGGmYBYv={i?I<4@>4~Z&1|;*m>^9ic-DElHi90
zi``qm=4lbRDi^CXU1}#8HU_D|%osz9jZ^(iHw(k3EI?2y*m8d9v@Oy7mn|IE8hPlZ
zmJ6^Iy*vE5HoAb;v!3(nvDkBd*^<_UmARv4$FF|SbpbY{^RE@xuvj_rChK<=-UAtE
zA-1zSp0fr!p0mShvYcYl&z9ct%%0M_p69NiS<Qb<Z+#ZtQ}*>Q!}jMP*EDCkrZK}c
zGjv%8Y4og4>l$X!r1x9D$c&rxYi1M2+1IoWUNe~(q;{5<9=xVjFmItl)_UkqOIZ2*
zxb)@fS0`)f)kP<-zko&gib>n>>lY=jkMc0v)CWeEY{sy$Ka5chfTjqbDUQq8*pO!|
zRNG^%%*aaNBtzn6=)P<}JB?FPU1S*o`L+|`czDd56XCJpF<t$gY|B6fsqimjd**@_
zEF5vyu!zMG?afi6O4bOjL9(1oO~PQkzrFb?ca$W1emP9O#xw3T4%EvKXd1rv%2&f9
zPw8|(uNfYNO76)l@X~Tf04sCYio__oY95|>Cl*1S2L;|q9&mh~ofYB$$KyOGInSt8
zEcwJ<vqhg{XFOd3p6^q<(Q3~rU^Ys$iRMu^izFs{<WvIY1_SibqRWqOn?CENU)?lo
zI(@AR)^zt_T+{uROqq6Z_tYuv3YSNhP}kB)ovT-OPFm_ZU7uQ*c>U~|V;e69Ip!in
z5Wr4G3hy55AsEthZnJ0MKH1<<Z)AshmFT4`mI_U_gJ+%Hg6>kQ+2~*kI{fj}FS@ut
z^7dpe`anP_{2evX!`PJmdHcX;?R34eg}%ytqV|gqe9UGARFkW;c}!SQRE5q)vS}QH
z*LjHdn)HV8QXFysc(28fHBd~8NpErH<`w%ZtJFTrpc?CCbu#BSV?fBJLyhNPtbBx-
z1Gm4n_Nraq|E_7>Ei*TD8V)GRI~m6hTxwb~)ZF$Tu+htsy_j1-{$O78!mpbKVE?*l
z0AjLcAS*s>Dgbg#+GK!%aW`OdO>#Z}is8ZD&?G$5{wLrW1_wHrX>ce^GC!XJL=})j
z;BbCD9Imf}!+jPU)TwqkrQ>9?CbEY?qqaV2^EtMFti$+?pImotsIYh18STwg#hopZ
zYz!V5O6=>-0ud;x0#RgrJiP^~Kn$F{f{fBHeLY))DENYORmfKq3^Ku&+;9Kcv3Ksw
zdSpbvIp&-yNJ5unK9+sboy7SCTHBm#&@l<8HM3#*X$$D~7FQQi(pTG)a|k$`VZhaE
z=L`;+Vqi?wY2XQYR?%aK$F=5toIm(PSGz_Xr-?=w49+6}=SP}t64d6#nb<kAm_)Xy
za~5O^1{?%EJ14L0H%1;N|7ppeCc{2I(KWcigD~@FPQdw3z&>L=e=eD>RrF(U3frL#
z(OyyAR4$?gJ%(mOSWwK$>PUNCso_otLt06HR7?-ug@R2L6-~htT~n)kOh;9zJLIdW
z`BFi^k{ecwIQ4^mI>h20`D7_cSW-PgNQGXYLP{*WH6-P!^EPuLtkllJ+H4RuZzN&0
zAo*GIz#Fhw@z6qm00C&HRtr{`8rooRs^gE6Lr*HH=uC}0L0a@ngR@VRx4H(WpU^GJ
zdTw4D*PR@K!p$4_G&u)l;PW;nZ*b-T*gq!X!yGx&v_fj?0hZ*5VD%J)<t(&%iZjhX
z@M|*=dN_mG&161J);G%{8&2Q@v#3Cnv-99N%OShmLkA>eGLVu(6J{k%qiQgXHY6)t
z(nl!_wTw_su(=9ApEy^c1f^Go$qIo?lNCy}$qGICY@eU4<ot#!AP=5vz`{3>H@{}e
z0_S1&@c(t#W5DR=g8eWv-M<F-Lz-|H3V-<s@CP%&Uy%lX$$tU<pe9((wT6hMSk7k^
zOxq;av4E{(G;;Y_>%c%sIq*hZSoR}t)Mr{%L)xmUhOSC3$MA-4H1I^7D3^x6Zk-vJ
z<Xmf>ozbWG8VnpA%1Yp0sCZb$@K6DmxG57RB4MdK+FPwkWq)KCF2FMHEmZ|{sU1se
zdc*3g@K8LJV!K2UW`YxXZ#Kp<!VanM{FyMLK0X(Ua+8ipNEP~_ufY!O9<H@1nM-2)
z6KuvZiJ2(;?=?|_iO!{~lwI|76<uAgY^SU0m0ilNYv@XL#NJcxXuz*%JhMh$r{4x1
zXgLYP8P%@SN|T1+Qd4SbdXdOxY~X}&tEkRLuLcM7X}!te$v2k-8HH8i)Qwy$q-X&v
z$Lh*6<om<v<~CNNIaCY=4S^=!t(KxuaCiuou@iB0d5|IQf!kMhY`yXJ`Rs62hbezt
z@W$J_S6;!jtlN1?`B!S$u>Fs;0$wx3&3x(4mneV!A>wZ4OWC8AOae6I$$P)(;NNXk
zp6);0N@q|4FImqX&cpyYA+oB-3{x>YnWkcRwW%11Y=<iz&pa6e!yakVF}7mNBg2Rc
z{eLh!_kRx#w6xQwbZny58<giWK%(2G+1!J$<YtB?FB2^JgRr<qf+aaqAPt_wnh`c4
zK$_VN^r^zN4B)~_x+~9yOe(`7!(N{%09=vev{DUH2e7<XWqGe%E`V1h>Z_8$Rj`UX
z<nwG{q<ziE-)Sp)T{Yt9s-z8<Y1ivDv(A@h&#}21fOjKUbdlkcH>|128#Zn7hRm^6
zpR;Fgpv^ouf5W52yauOm@GDX_bZVYb7SP=VX?A(9#mYPw9MZ<S2S=KDlHGAKXN^RX
zw@yGyk<|^mt*KVIs5`2SR&tA}fj?cZ4jrxZXLYvH=X7#%v=a7=pZSn~OTU|pBfChV
z5g}qB*Af#;)nV6oLs&MVG#Ve48kspe7It_n8}yJYUBj{jY$>tWStuqKAZ&u?+OePn
zjm5qOjNwN`(b1b-P*OFD)l`a-8}V5pI-bhT^%d9RY)jAx7=5ygN;aO9qv*oGKJRuF
z>4(n6t`$Zn$N0Lnj&wSD?uzD}4_7ueR#s2#oHMohiLLGT#`ABz<+64DhAGphH25pV
z#w%`Ee)W3Rt-fXc&N)?&Zi>vGJie)*sx{UyZpz|un|Fsw+dHpZRNG!Zy0X5ZV?jsb
z*qJp;H}1LR^zmd*JDZCEvfIz@fpHZ{_|SsRiK6&MLw|5GhAP3!t*m0BdT=rZ3h{EB
zT$W`mL8VtMK4r4|0;Oe$8<#`QV@*e<N)Yon9>bMiP&4v;3{T7;BrA_l5YiE+V^oY4
zTIXtWF=8kqmEWFYI!1&x^V4uThTf^{>pz}0B4Y>+LPK#-mKg^nfP)e?4>cVJC8_Bk
z6$}H_HDMJ4iM)bZb#_A8APy?e4+n$OK`@DNE*zw0gFMdo_FOnfO$gzBk+ORSaZpL3
z<k~?T)C|YLXoiC-z(JKA>#UPFz{azxG&&e19%gh<UYSBenTm#*K{V8z9~x32^7A5x
zDVn)le?A1ke?JvcA0&$;K8%2m7@0us8pKDViVsZP$fJ#YDa7<vhE1hbX7*5NBE!)b
zz|j~xiVqD(nD3edXd07-rY44_(#miwg{Ep1O^t(SYDS4t83qC?iNSA1i*hg$KQA82
z6uUn!B$j5hzjfz@(&vZ7`hPsgZ8{CF4#sQsfZ2U#aa){YDahVtRU1dsX>VdW|A7qZ
zty2|oT{<`RjB`~OL%j{^M-AcLacU3+G2c)JV%+X4Ga<6aiAmw~Pcddjy?$P-Y95|h
zI4`c{uRkmJmc;A*46jkrL>?Vcib>J1925FL2g+fLhBv3<6tjx;pa|=SAhZq<DvyQt
z$A;xP8@t=ppuCgE)&u%XzOveIB#OBT*;OYxMnJ7RJ_b8bJ*uwezlvuE;u+AKmfkoo
zBnNN$)lE13DyC&T&JV*?kFQ?+_;r+LDUdErKI*&q7r=*%fNlE}NfaY?%cD3`N2-n_
zEL6Qy7$&OXk>04c*kT2`jLK2IPfdo8V-@%&wJWUBjyeMeabe$mG=i=7=tH6)9S_fy
zM6JuI6MHP>0xK%W0VmuRU^=BiOo;i#>iDUIr?8003vWJFuoM;r@Ebf<Ajm=#>=dKP
z=4#N8U0N1v(5ierTCe_)p*`A|?c^<&>UCwwAwE;7jrw%vf8Tz0@P`kt+Utq;T)X4P
zE_JAn<L3L<{5UrL&F;0^XMJP#1&L{uN+lQN>V(ZJHa+><y$7tF{i~;Jx_s$^<<g#s
zv$SD9S1-C`_13F9X3g(jdUM{&l2RP|gZ}8i$87dfIay9pBcCv9nGYoE4`abgG5~|~
zaV_eIOs`E1z!X_9fn(%tPFG%jpp?0j`FW~4=@(^h2}79_z=S!DjB=NbG%_kf3TOBX
zsjF^TJ6I4U960>Pp#!BRpPjSGrWT|nHlqo2t&7d+UNwA5lMm-~t5cfN=5(9YIo)RF
zfG~R=XM&oMPTUxse2TX^JYIHB7QvBmTob$Z*2v?U(ApYtWK$=zuzot?;3nNJWuIbK
zhc}U-Yb=B{Uj22Gs7gkdL>15`QQ_)}SX>Z$wqSL|Y;XS?)2Psn8gU}kUsNA=q`6e|
zvXs4)zP_@rUtjqM>kFo@uefAzeg6OM`jUx?5!Yx-I?m@_C3By3+BzL<ok7qnd(&1L
zLQP#9?X6Qaby+0y+D8vtJ1}f-Kvm8G^{xW7>Z{s8yr3~z5deW^b@14lCg;?`>OKF>
zEzhW-M?{b_nf3y$%+g*XVv4hzodzS^nI-V}F{_sy%!3K^yz*c%@m}OKTPMyfJpH~l
zq6z%%{qW!$JSX$FIhx9YX`NlER2Y`vsS{$*so&q($(#aqx@280cLs+{oMQ7G2lHTN
zIzk>yFAU3r#l&{a&VICWaK_^x%>0>SaQ<VgF9`Es7if7fQELzKrrLwl5m*1+OjvA4
zCahjHsov<kAQ==*^$DHsRM&qjEoaH-`)B&yq=;0Lm&jnipel@U3!Gc(2q!BBeCiE-
z5HL)0m8ny6v`Rs#JdesIbM32GR9pKhl^sqzb~<WbWp#v)zciZN;nQUAF`Dq+o@RTG
z#q4l~67ZPTxCS3$lbUR_)Yek=$Xd?Ie4xsC!Rjth3%p8D*$X*(<P8T7@D|;vdVX)T
zJg5hS1!~^oMP6Xh(m};uKRT!mLHY8zJGXMP4k;fy|FrVrZtn8M?>Z9Por%kj?|$x+
zu9x;nKQ3!}xc$(J^vO3r??^l|eR|V1<8!7i9yg=zVP*d#_xyFo=~oWC^&kJT<f^JM
zzhd)f*c>$@5rLYv4^HfHu^_OqFMV`RZ?4bgLIr0^-5TiZDLxoe6Nx&-!y@(yt^yxM
zh$XWcRv$9ghYc%|9l<zjz|$1HTu>O#IPr(i;1;D#0jkaD`<yi!NEaQJZqW3?hm5D3
z#EbP<8Kwp0WSSO~t4#}f3a16RJm)zt2(@;mnL%|ZMKTQz;?5xN|6hj_BaE5gq%o#(
zGeJuCs&Z;5tob9rT966W!XdDF&lA>UusIXF^E8n+GQe8?O@-OWV6G(JOoQ2j>`-nT
z1im-lFM5C%@{t#+GOZ&tWF5Y<)*-_g9I^|iYLts|ZfSfZYRN$MnO51Dk^VET`TIjy
zK&P$S!PZ>}E8mi~az9FbPqa6vO8)#v+Wrk$w+rNbuBr%fReNDD|CWxET&#aJaB;j1
zAs_^K-y!p;sLqFVGFt^1(6M@Wo-HF|O~_yi@MIC$!9U2?upX&=62ti!cn{TlT`1Wv
zl^fPNWf>9cPT~#4%<8?SJnASTwG&LA2BnyG%(|qMyyV@XlF#(Wj2Z81-632>@}^Ll
zj9SBgsH-P#QbMY5N-P_2TbAUfV%(1qlgZj7(hBX4zL>Z3Xac!Hd|KdbqSL54oECM?
zOd04f$sU)A#ixuGC&{g1&Zs&V+Bur2$vYfYLewD=wG=aVP1MogW_zNkC)kCTS1(x9
zv8L;C<#RY%*m3=w%R8q}%$w7@rfu;<zxi3m!iRqI9s0)G)LHlV@#bIM^0xA~x<`*S
zAK!ZQ59@30qz%VusBhq8W#3c3U9ILT*qkPiA($uY8evkCXM{;j-lS2*^iWT>)1{V)
zeC^OCl+Njcn|v&nch>YK-B;?2>VFnSoF>(i?Qpu8k`yJ^`M`*hdB||7TxKes4JChv
zW;dq6nTz*-$OvcN2yo^P!DTPtlHC~_b-Md}z)cG0bdZyF=twZDnLu8H+o9zF*AIo<
zhr-Df?JZM<Q*LB%E$5vKOk3w*J`n52&Vy7kN~82uUN$^fn1PVSW%Nhbyeil)2k~T@
zScPMY&Ua!J9*i)xiq<p2IeU1OAZYnvHm^{dtvx@(yh5BY&;1kA3S~}UQ?A(5Lh-*o
zwa_prci?9ujV%O!M#ZB8@K^vo*kvS9%(45T_m-<PlNUJ)x0N`ax;ICqkQ{c8HO=>x
ztFZ7m2P;F^t#{4``<aPkg#Fk(oL(~QOd=D#ki~=uudw^b<zx4%I{uu;8{(u&1SNf>
z*I>4xk8j{r^9xOQk)=@Gs=APbNW#Jz(<~-aNUaDoV8<O1)tfhyN;+8bT}<E&e^kUm
z>c#XLI+d3OZe)e1-|nBJyr%5zrp?s*7OM{A)&Q|}Cd<I*(~>hpd0n@7^48xbZE!)I
zgBl{EF~d-#MZyDSoPVm$nKVV^N-X_}^j2pV;X0~$wFnB#5S(jPY?sTiMG(h8Fd7VN
znSKn@mKWn_AwI4)PUf>xpM=TDZoHW#CTwmGi@s$C@D4F<PkwPZOW;(Cjs(pT;+lPS
z&LC{MsiOA}jt71DJ8zF3_48XWba9Ise0|{dW3>(M-u`mRUg0Neu~%lN+;Q_wccuj|
z?!0;9o|MsYsTOzTwHktD#0b<~g!jowuOh(hwMTnBDw$g%?8Z)}lf5>Tu5E+2wh~wc
zsAjyeJ3XD_ULTn&o(3xr-bxlk4AvhH79;5L&#Hv@U%F=nS*WYz-UM6TjraBDSdY1}
zPaA~xzcT@CxM!I@$AJu)((uoGHjy6VYuIPrvt~mKP9T^laPJA)kJxX7kI&;b!b67C
zZ^Ha_G@!2m3%HQkzr2o^Q*+x`4TzX*Z^nUIIP{NELU`r;;Or^&W9kBRlllJk$HoOx
zH>YLi_3~?|Szo7~H#>~;;F9N6@0Oc9@30%-rrq|Izjw&}ZdJUn`|y(O|IB%Z-A7lI
ztAGb?55t2`ounrfMm1!ljb#RWNCve}*(<8m$#w%y5!RWlsnPC3@KkmVywO!F=a(np
z8ivI2!%>R&Am-O_>lscNjM-ty7u65|HW7m{bthsL1uM{C#M^@O*gdStH3M$SM#dGL
zkp%Xv{Rj-hBD+U2x1R0I{}cFz?dx#(C}|}f{5*a!!Mly&-I?`lqJZ|KwW9Z$ox})l
zs!!T&qaCTrP{E*1GbnGJ({wykeXOZ-cCqhxb?A7ZkN&9pwWis#n_jD~IaciJoKxgS
z+1-hDE!j-WVBeLHPHlFsu{avVn^uF!mbtu0guScX&swX&PK&l$)W(qlwQ&R_S#O~#
z!%K(X)H+>OrynP8aZTeX7j$fedQ_3T;dQa*hZd*#qaagAoGWAf^8VzcvR;7=LecJ2
zBnv30b#e7B(`=m1MEYw{Sw`;`>9c8*b0}!8IM+P%TyPTYbc5%jc&=n?{yciOvP`7E
zR%+Ax@U?qRoq3<{)%(F;0DG@gyWa;YmVzkW913hc$-wTTY&4;5frej}S&eCeD<5R}
zUD;U}VHsZ5co0V40MsE?3;W};Xk<1l%1xX!7z+fW$@$R5yx-s#+%B)*>v#B*#zDRC
zZN1@k;qKB8Uwmk_<HqkOM;hrTf@0O{>0v=>8%G=WuFYD1@TF6JdZhL8oW|DHKmMcX
zf@Wn^>*aa*OIno`@#$kf|A5IOjuX1redD~C(Pm9ZCR;Qr69MlWmz*~wm_ohAY`fE2
z>W#3VpuleXo%wEZNJ~aFBa8kKNU`2*8|ebvYXb$E`})7f8tRR9$#m2PUy?jWpFNt_
zf0a5G&Tf~)qk1XZF4=H8u;{_*z$P0$tG)JiR!wbHt1Pvf)?qhRRMp`W^>~=uru<<l
zeJgy`CjvDsshM1EP`>K^o1s+q!o-OKler`Crwp`m&kS&2PoPaPm$19e@lWzk0!O7f
zjc984J{k_l$?B7)>*_?mh(GC{g?#8^_X)sz)?n|$cJ013zGxx`F<YqXOZeGMMf_|c
z1NNm57Vl2KpG{!s{cI$M+8J76?dvJ=xozMlIn-VWw*!nx!H9L{atY40@fH;3sKYHO
z)_4}j)f?MNrV;LYLx<#0GTa4+6N(!Gfreto+4sYtYjR76-bDoI3D@sn>(9Yn9JYFQ
zG<oA5vwGtmZ@6cSSFo{&AeC~@Uanh9tK<z`WTw0NctzRhiI23t5muGc$uEeemNTRO
z3BIfU3GFTl?7l<#S+Ir*@&ri)k>86-qR~Vl+qgp9IGi(H28$uZz|>{x?wP{C7uazs
zIqi~wvkurqc<}y$Y<1=8>dNhy(3Y|ztlS1fM~S+-MX+3S2Lt8k+G5|4uLKtEbQiFN
z+sa_!6>;#Om<26cOe`0QL*<R`vzK2MOUFfw-6UB3V}+&Lub<V%(9x%&#x-Ty6ZFRm
zJC(*x#m#VY-(z=`wUzztAeTf^QU63llKPk0Ylf#XQb`_2O2+bN6q}n#VU+E+lpjfA
z2gISk#^m>msc_mb=dH$kOBs3sDtZET^cWC5J{3Ja)d#E8!o9_TBznqy>F5DmNo8>m
z?hHLxj|<4Da3qZGVEpUQV@TpAja)L|hmj7TN7tPhLo<@tNs-e30jo!@MF0Q*0RR91
z0{}wOh->g)4?Oh%9|!;d0002d`_{w&0002j(Te~7X8rI6@CM%i000O80ssI20001Z
z+GAj3U|@gw_XLpC|DX0hhwTFcPy_|M0RXBN2cQ6W+HI51YZE~f$G<yob|WG|L?{Sp
zNTmk}62j6$krD{4gxW)dASH;#5JD-%`~jYFDkX#<K@7DZSkR<|P>P_42>!TvQ+f~&
zMS3X3)(Rpb(yrgxBoIm$K7421yf-`hzO!l%li;J);6K)<9>1}n=Dj1mh^PjN_8(ld
z6=tlwcceO~s*bm<+K9*wZdl_OQgKXEGgjSeFn^N#B)3R~O!Rb)WJrn3`uVKoc#+h}
z6ZHxur;UPs1XoX?sqb^Vji!7j&pH2wf_(u^tA#b0LP)!eucPU3fBOuY_5<dlr_7TN
zn9<iUuGg`obGQ^3;Ta073nC={FvdN0dFHp&qS``P?s`8Yj=Id_svM#rHDt9$RyxS4
zD+p81$pIQx1xr@N`z+TuE~DWDnaeoOoYfAMHEI<N^%p5s#i;C~CJCIjhY?YS2ui|x
zPmQXFoCBNG5Z|A}uWs-84Nu7!S=R9OOpd%kR9r0EUyxRh5vMl^uIEkfO^j>pUf=V4
z_XJ5x+zN5NqE8`j`Yo=d8L!D4Qmmh~U!q2;B+cGu{Jx)nvw-u)3*UzUtd@7rctL$l
zHtn;B^nExE&yn@jcrggQFrMiV_Sf|N1nhUzTGxj`=s}TQtWY<|M?K7V--qL{FGhVY
z2B8<mGoyjaC^);!HTT2ieQ=SJ65wnBeO1%+=mi<=g0Ixu)H__G7ky*_$v_NA_C03q
zx&NQE5F<v1WEClC^B-WXDkDj5nthG$&H-ZlK9I0HL)hLYI|%;=_rlkl0001Z+GAi~
z&_RMV4C@%@G0kDtVqU|1hWQ_h0ZSgsGL}m$A6OMw!&sNG9%GYWt7F^4uEpNRzJ<e%
zV;RRYPBYFB&I-<HTzp(txY@Y#xOZ^>;xXgN;km}k$E(8|!#ja@8y_2=9p4nb8~j%M
zlLVLqj0ExoCJBlNx(SvEE)hH@#3ZCA6d_b3G(}iII7@hi2!}|H$TrbMqL0Ls#5ak*
zl3<g_lh`EjPclk!o8&vGG^qp9Y|>iN9n!m`zscyxERxwHYb09&guCQY<lf0ADex#5
zC^{*2D9I=tQ+lH;psc5SL;0JEol1ks8C4_IAk}%Qx766w+|-)XPN{uTuhWpzIH&2O
z*`|3wi$N<v>y);Ub_*~tJal?=uIakz7U<5>J*6k5*P!=G-$Q?e0gu51Lj^-8!#zd}
zMs7xvjG2sWjN42UOct4{n65F4FlRB}Vv%8S%hJbkkClPdDXTBma@Ky<3vAqMPT0!W
z7TKP$^RT;MFJ(W){+WZ7!!$=B$1uljPFhY~&MM9w&YPUSV23VDE&?ucE*369E_E)e
zT&}rlxK_BHaAR^y0>Tcr4Q?CU<=mIKAM<eYSm9~oc@6*}fs+UT0RR91>;mZk1poj5
z000620RRF3761SN00C_P0001Z+O1T}Zc<ScU6+>@BT)xL8J$tZ1SnXll!38YF&bis
zSUaYG^hU}>E>`PjICA9NKQIoBF+RS)U+^1@ah=WP84zO3&An%xv)}9NeF4Os7=my@
zYk+eRQ0ySd-2ev=b8adQ;@EkhID~@pNpTnn=a=FLQh~VQQ+OG;rT8><10NKxArbhk
zIO@C#4r84`KJ;4gS-cH>Rvbe<+)#WDCj3V6c|^nS6<<Iy{88~2#KS)oe?>Mjtm_+M
zk*wnH#5V@`M-Yif8Mon~ixc$VqJ=~BVepRcRb=_(krtH07E(N|P{TzVm*I0)qHTv3
zJ~G_hBd&|qI^X-0`P6=n8h5+=_Rv6+ItDxR(B?Xh>C3LkwaN9E--h_C(EfnBuK2R2
zhWitK?~2D7?VETq%EMp}Rr)vBl&sSz;5xRsZnF;i7`%`tL(6#{GOWSF5o@#g4D+i}
zuS02>=M_pk`YDU_A+=h}vMt^wtDdZReTum;Zz*LhgQD=cT1+b~PTbL|c4j8q$HKNx
znJ4$M$vm>c+cJ;oiA<(-jog)yTki|MmVqOQ_mpQIxnA5ysZk$gmiE!#I&6K#7>j+(
zn~NlEuFRc{VOK7X`9G)Qkwo*SIIq*EPaE6AX4KcV>s_vgL=EZpnskAc*y-xX+Y$>`
zI%E)UAgxPgzRq7?I;unu_WXc5Pj1%sNJDb5UVWKd+glC3HL0ele~KN-TAHUJe_7#Y
z$fd2gPwtD@;IrDaD(~jf4DTCL|B2FwjXRp_o^m>vka3+h$Chwld$*}pHg0cj?a`T<
z@^dTlXKLzdxA_0?|5-GDi>H5IG9GtCPNscdZ?(#`D>+{i*%ounv5q2j3Zq&IGu_jb
z|4B`5+n?aKif24=8Q<pC{sR?lE;;~s+HKKCY*RrH#_|6oj-5EY_d@UB_iU#?D0Y%Z
z@4XX9VjDsd91Mhx=v5#jIDu#vK=djQZ~#Paq6j!}ri%k`L^yfNjk)~vt#)TdS|Vtx
zKjn1gfAtk4LL$W?QAA=P3M)3EiNQ`Ram15AB1t5ZLMmyblR+j~WRpWKdE`?-Aw{&K
zJss#sCpyz5JgXbs=|N9=(VIRv(9m(>!c8$Hlu|}t`cY1Q1~8C8c<@p|B~=&<W(Y$W
z#&AY3l2MFi3}YF`cqTBBNla!6Q<=teW-yak%w`VNV&y0s*vJj`@sUmJVmk*o&M~pE
zjrXi)uS81>JK4_`Zu3Fx9OMMQ_{ndMaEb@q<1}-b$8KtPL@oDu$TJ@Egr|I>j+Z>=
z1!wT_gFU?A6|eF0g>P(SJ^>c6ka`+8L?er6B1khK7PEw<d}bNTX<-GcSjklmvxe2I
zWnK6WzH^PUoZ~ugc_*<FC-IUXiIODAk|L>+Ch3wPnUckME^wJE+~FdZxXWfvN;WsS
zB{`DI4#|^zDUd=bvij;<n*5rHZsN3s8UhZ7qtevlP^#?JmFk|itW`byK&dHprBmrr
zx|PMs5@o6KUq6qjruu4{y)NJj1#4@noBigIZkW0aOI0Y?*cRMIMXQF_+#{^6bShnc
z0m#y!!T<nx+C|N~4uU`s1<+j<mY2vw3L6u!-NsUTp`kHWyNTant1;0|zh$s8{v3{E
zfNsvsy_pa1?!cd{0q>w&&R?djrP3uNrh)v3$JMMrs*`3QiVB3<gI!c_QCg~pgRlok
zgmWXf2x$l&!qo&H;q3?k!ru}?M6e;u5Fvp>MBn40`m%^KW~o-xXH4eUiytGCzHsTL
mLi$e<zMPT_C+T|R0-3o34dm(v(w9v7000000b8rufB*pFx{6r<

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-light.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-light.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..6d8ff2c045b08b0c51b478810d0636b027a318b9
GIT binary patch
literal 21012
zcmV(=K-s@{Pew8T0RR9108$hH6951J0OF(o08yp@0{}t*00000000000000000000
z0000#Mn+Uk92!&`=_DM4Oa@>8gJK9S34~=42nvFy7=n=y0X7081BWmRgIWLtAO((e
z2Ze78f*3z{WnyLq8+X(faOVp!ZU?N%fBO@U3<MhofMlMtMh1e71HdT%G_(KzPdPfq
zs0|vRrk35mEr>&zsBDc2LQ_v}kf9U>jkv2YdrP4rS8-rFG_dKRFhpSpVaYKad!Wlz
zcC%&bl4i{{=%q|<6pk{(5$5ebIO7dUj*pT<RNrssB+-HnCT>LMM6PrV{q^ukR?6H;
z86(1wJV%%*e(}4aFcs$41KrTY{d0+h<&ZDvw)e4uR{m46SW6_5_ko=0TD<t6kKoVq
zShZw+mo2bu*pnUg8iZ48R+1qOkU|c~2`k77qr(nf>jGl90W1N;&6SAm@p|SV+K=?j
z%<jK90u|N~nn?k$0#tEzb*LjuQKL&0fn#?numiAdfZ|^wIE#<s93U&;0Kz+u^Eiu|
zv)gHTm#sE)HuwJHnnu^Cr&>ND)iLWNNBon4l=DArZEt!^bHlvw@y&8AqLQFCy;y&f
zu6jVxXscRv0RLZm)A}pr-=m3H%GQiiJ0tT=oFzc-dHqJJGBRC27I4SK2|oN2uX?Gp
zxumUZd_3KkP4wFg2LuBgKJ|YM#m~E?ZF;~10*gYp3Fk2YeEp@Ww1k9!;_xHDgnWGX
za1=UGP={4`{9!97a>sh8>nZ5uj?3wPN|r54vL)}z_AC9j_V(ZEnl2sD2511u@UQ@!
z02FGP#xmxCX@Gy(mId%Xq}S6AGGI6ncC%UOi6>QeMc1oq@c~gy%Tc(yRoGQJ>2nx6
zqje_8fZ)+5Gh#s6p|}X8iN;Ms!@wvVCnvp8`#$06=#M$|rNmC#_w1w@-agR@m?S+X
zTIVh}ggCk+>1}ZP1UN=R8;|I4|M0<|nxtccSE;&0=iQIJ`%L8w?f-A8z3saLKzvXn
zxlwYG{y%icI`-+Q<P<wsE=re}HvxF_W<U;R0PV~Wq(D#sfT-mFQ~*i&4+8lCNO`sA
zP(P$dX96U>qJo#Cq|Kpnh_Ur??LM}tC{t7UxTu{wSMF;UmH%A(v;|U~6Rt*eJn;A`
zeH*;x0;9qtFUisY(fE=^ZK_T7tG%jL*0@6ddh*Y*_z@0qVuhwj^RoDd#>-8i@qrEo
zG=b@0IETT%GgLh(**3{X7-e=(?mhK3th^cT`Pv%7R>Cz6RzySw8DtO<38AF#-l^&^
zrf}^s4lyEBsK<WB)IN`>8Ozk{YF(h<1P6+mp1uF~&v4owGs~Gx>M<fEm_As)JSqSL
zfZPAK&jPsq>beF1?!K%W*5&8{G5{1ALI)w!))ZZ7i1e))88Q)BTQjodAaa{C^1Ma7
z^ASZ7pc2_KaURRLb~-Z|Amv89;Rp@}w@oz0=-7-4?${GR={Oib?Kn&$_kotoAh~)z
zJw(N785k6RrdLm<MWhnE7?|WgAts)=SX3lLTBxEXM#sdMZe()z#;Q0FufiG~9}4|w
zo-)cQqf(U<vs>qH=QT?n$b0!Lzm#|go6P94EIS|RR{k-uLgDX6J)hm9LTEJ!Jjh1R
zVW`iQm_$Ur3sABRfh&{{<{uOB^udY>@th!H&zDI#XzQE})>=FrW5K+UWnzUTPpnHe
ziEZVUFt=Lea5&*&OCTl3=84e-ERIvV-Zu>~2taE}EP->{GWL?;Ni>Z!Fv@@}&Xt!|
z<KDs%(PFU10)HO1P=h?$FvC(iGI3K4Y%K-fIS8V^o7jLkgJea51W~bTL4>ZNjhdJu
zdk7M(z>DMqtcSn0HkuUjm9d(XhE@@^747<rT3calBEUa3ZiuY@%$$w~n(uD57Mm#(
zD~ygJ2Ww2sl8wkh4_5%R?2jlLzc;uf#L9FbZ^b`#lqG$=96}owaU+b#XJW|VkR*jd
znhcJz<Z_`@!d;mvzQ)+r!RG4~Vp&I))r)0YXP+Dg98v9<b*P<Q-XMMY4R+Uk!#(uK
zC{L_ojAv|Y(^%R#B+p;~I&3VoF8*P|&$3jccP5Z$Wx}u`@FKAUyePbAykuTFFO!$W
z%NFGDa(NYk)oPd8t@fzB>a?Iw(9e6vd(Zp8`&hcnGUg{NfQ7IK7DFv9yyfMhD}fso
zPvB1ALEuT{Mc_@~6TSp~1pWj81c7r1+B%pjgdh~cARHo`NW%2tqX?o2IYG9Z&*Dl>
zA*YyAA}BLZL3#5EUL|XOE=X7ii(oO-Hm~bnv%%14&;-q}1X`dK+MpdeXeYD_=!PEX
zg=Jfo8&<$dwu)c(uun+EIRG4lLvR?5z_Bf-@%nzHHB`>x^b<I5CY6(jD3Zm6LR2X_
zSv~a|r7w{cNRdPZKjK4i4Pa~FxD`BU05q^@aE=d6!s>)ahZdrBRHm#`g$hoJhK|x%
zw9BCzdY~7UZCPGcz)H4?g4>+#`k<P|`djA1x4h24S&S+J&GQ7cogw<rrYZv}gD0)U
z!oLDRUERdutlr#zc)j##WH5mWxscU$3UQ|KL<mVV5Gi66JWk{kidpnPU2do(a0c+8
zPH}^xGUn(jVq?PF9RxxtDry>z<W$TLu)0NNbDu^`mnt10#U8KtL+TRAmPbXPjN@ZY
zRc$MR1$y{008fME(za!XM(n#KUL8CJJPGCy>s;P(Dd4RDq*38A5|hx+33Z*}I9&rQ
zd=9@SS`?vQRA$~JLM~xBMrP1fSpNelSvZ5WkWi~>j%hBXjSZT@x6TWx72Pgu2O%{;
zn-|L@S;Z1#CbnjCYvrc|J>dl99J>)9+p1v67~?&;i84_M{A?|lRoXWM{pZ=u89d|A
zFvi}_(@u5gz<gbI8}PMgz2aBy3$ue_rk(U+U7Xn{P6lXXM(k^s#~lf6ss69}{zZ;<
z?WrxcW7vH<ws!<^I6_Q*%TT~fJ84@NVG~Bga*!dlEI=Gy7vh)*O$gfYc^>b9EQHSa
z=-AzKKPZF_KD3k4OW}1;lM7Xvo@S{h85N1Yr4aSzCKT-TVILPqCOS;Vz@r{c!4mF~
zd<yLoE1k6D+bmBwMow7^L;q<YLt)Z)==*dbaA0Mjr~ifwY)_yy4m-bpGbYEDHAi>F
zg=<i5c92wiKS`@;R_Le#hXswtDhatM_+f;52)Xe~?V>Ka#nf4?$P9%nSlDCDiyU^0
zQ5@@FmFY28lbWAf;<iBr582Gy?dI5(IDB3y+lu6aCj79LN@lC@&;aq{5}NMq!*E1_
z9+?Vcavr7Xz)!sjX6N`r+|xNNYJQkbmJ8aA_q^k34fA~LRqP^4Xaw8w)Al&ieC_h3
zI<bq_V#u75_OE1_0Qm7~t@@}tWXe_}+4%N4hwqaC88(V}Ala(+n8LA%ieRa2j%adp
zZWX8un@2JgmIxP*?D|uOXf@FVmv)4(9X7}Vt24Z02^oTO)zc6EMzRU^>(LW%d#WIQ
zTd++#We@_#@qf8DvdbKkF(SlVf)7zFjHCxV>??AZ(|NsDDBpp#kC}z<`02rSbc{71
zA+ksytIVK_sHlOok7Rbe{hx~gfF3Uvv*ogyAw5<qn>zbz|LMf4b1FONOIm{Vq_Cl%
zGULYr9!7hVY<S*qdTO+cbz_%$j;I}je^2|bT{{Xbra979v=4fa>fEJg4^Eu78h*@k
zF&jkgivggr;>6?E_idvhVJwD=&!G)u1szK*9<+zoD&Q_V?g;gi+m#QiW=`P!0R=jP
z3agHNe^+dR$~a4A^OxP7SU-#2!FUvPK;sIMl|YWonRD|L0BCnI<tO)Rq~^Sq{!i3(
z8P+t$`3gNU9Cl{{xQ{S)jXSSHvNI2!-aUoJ5RInaC8Tu7=+dLlfFUEsOqlxGj5&*E
zOIG9nYc_0Qw)<O-0$|U9BfyC>7p~mkP%qjURK_WxX!;2q(CZNbR{*etY|Um3NsF5k
z3J^iAP#D%xcI44}@#e$#G6FnGDr3olh!dvBJi+KlLBWa(*{KCc;0D7J5g)|-m3YF|
zQv?Dn7^zU~gdr1-jtKNbO7~|NZYc`%MPncdLn)G<?R6VzU@e_E8M6I-8A}R~EDvY-
zxKP3=z*Qj%MZ_z{UJ1rZF;a#J6%NW_RbZ-8W@qt@)LKsJfzi>X=+uioo!`=yTSe!L
zzUkL33~lu{xovi`w#!~g_8FZ0ayh|2aV_V#Q;g2>n`!PFSrdhJ_AGZ`sqe0Pi2F!9
zfbs~5$55WV2WEcC|M)l*bH+@Cous)GWPCI|NmAJU&Vi+nBkIVN3OFfJ!bO=%e5T-i
zea(Ed>lI;{)#9wPQwn{iB<?F<rIev;d_R@Do~R=78|7dGAX93??<e`n7I8$XczGRe
z0CQfLxn#?0Rl%8YxmoXq?q-(UoYS5i@9L4*B?c|1X5kP3jl|0V1Oqoikuo9Yz@5KH
z3399~saBK3*Z{4;u-<6Hlua((vlpAuv~R(N|HBm`0pX31V#uP;rSK6gU9Jx%ljS|?
z%zB5KjVFk7gpL2*6fpx|(MNR7MbSf>VWLS{yf5Y~SXz#!5psk{_DqtilSG+Q&Ws(n
zw`X1pS5;9RfN4)j8wy?z2MB^)9DttRn-P}Wc|47<L>Sl^9bKD&uS$`jj5L$%)#+Lr
zo<x}5+XH~CFgf}qbv|EPG|>Qs`3QFIXUs18MM~ZZQv@;eUp2A%ty+-$m9Bs`0LZtQ
z22fz08Q9MOz$1Ed0X7(?0G>Y2wjMy=Oxm-zn4JB2ccq+S5eiX+Vw53)IxNDrts0g7
zeK#$5b}6iiye|k)%tJonZM6{3Uyc8GIy7VK*!!0J{`cAWXPRfHpPhPk{P!cj@A-Z6
z?+bnpd!TvnQS_YL)M&yq$yBTv0e~bh_DCN8Z9#)p>y0bQs`jC2%UC(B*6NLBtKI4L
z`h($UJekhsXXh7}SJyYUkM0)B$4{O<d!FR#?&0a>?Sn#Ne6ct@fk^TrQ>gv{fk8An
zBbdozbGSUdAS5&_JR%aXxVpB!y=Q2AY-0DWJ^Q96_aB&<o}F7*T3lXT**JLUu<0=V
zLTti=ywS1Uvz$ag-Fuh-L++s;og*IuwtO*f;0ORP>dAK0&P6?%vv==3dVK%EQGH+H
z^`9@_y~nhl?qlIQ3z{2RTHBU%b))lxmCNzx*W&=dbO1oV=s=cPf<#MHf?KC4WJ@XI
z*-|PtCD(`*zm3@$gdwu}IWS_`x9b>4b3VnPZFKI3cH;0tm{px0kr;eYhY=~$KW00{
z>6nE!+xcM3h}BotMPy90CR+5N{Z+S4kzCRKtG7;xrYN-R+I5~G@~jSsOv)VBF0nav
zh$T~YBdMBG1dWK~>vL*rX)QIG8K_M})G;_ofZm`LfplrFfJ|8*gAvcK`(-=|Yn0Yb
zd53?DB&N!8mP%LYnps-c;HfAPJ={Qol5oxf-O{zZ1B<nl=!8VkPnFl_gjaO?u4VBG
z@0gS#ly$bO__C2NHe1$hJUKH+rMU4eUu%Uo&jUWfa4(3-d@+qPY-j{;2{#qk{&A&A
z9NQOWWUqKf_|!{T#uiuPW=Axt0*Dmj%#%kW6HxpuB1hd)DTuC{ngYc9BAzwvTN5uK
zMr?z+@Vz2y&e6hKFm18^<@Dggt^_kyV4+E#l@$=urgTJjp)V3MX3qiy?`qpw3^6#8
zSY^R_JRQr-(8m&ja~bps2$_$~W!}&y0Lb@1-Uaz@fXj1>Z$km`_W%?|oh_QkYD}P(
zx)Y?)q_Rtl=OHP3kO`|i%Zx<E^@Ma%F;9ei$cIJ@n-X@68TJXGcqY!W@#VWGyb<xO
zB`aWhh@*<QjcKN0$jOoZw+d0p6OI6hlAPbwm@85G;@H8pD91zCiNLy!Y7vmClZz-j
z+qtpdcJ=gzw@!cu`fOM_+OCCsUud#QA0upu+V0tJq>PAi^N26J&AH3TrZ1^?{GD^+
zpr^WQvw7LkmNfVy9%o^j_dEJu27C;TW1H#UWQE7BiHx;4KF3yiUwL??qM&jWM>!0l
zuq0&_&W~PKxuPw%!kI4-DVzmjU6_r}W8edk|15H8&#cG!((GM^wAUl}k>iL(=SQzA
zXJuTvolmgIAVe>_!u`6RDF7my6T4fQlMI^<=yGs%aUhEp<Gf0sHLWRh9C5)Glk4H*
z@(fxk&=i7-JLFJz1Li;W^j~1T!4|ZevW8X*%~3nt)#xo+AOp{3jt6?tORkHFBI&;l
zzt~&`cY?W%Fl3_?v6`_fT0Byw5cUERKD9g1rYL&K{~ZVeT$6Ce%HRv(Ib!NW9()L&
z`iYnMCUv)(2H8af4%b+#i*iWN5|dD@!qD%?1e;HV6k<<0mL}{Z87iF|@p<FXYA;c`
zub3h)Yoa$Ex;$a3yh3uimfqAVi<nXp<!Pd%EwY$#rIUediSpk4k9Ut`A*H#?xF>7W
zgU9ZVrI?9K83+2hH-SmBeIFP%)c}S@=}2OV`Afd2lF~@%(2E5*Y}|2?@F4v>MX2uw
z1dr#|26HGOOYfSe!T^PTi2MD5%zHc>(A|UwhL}oQjl9*q*u*FkUNuy_sRaf%anFTX
zCC9{QgrAP7=W2O}<M`>Fp}6(vP}TPADUongC$@%A$+Bscq{b_g-XQS`O|xI;UA>;a
zJ;wJ*Xugc4vNXn{f?Qe<dE(3XQoe-c^TLFf|6~|4$tN?dQC@0;)_t7rrpX!1DZ(@m
z-+bkF`3Xjn7xF@gMB^0=%_63WQy97&Cw)5ZS2M3@;~MzH0Lpn!7v`dIYfu#eN<G|l
z*>ue#=aUUbUz|L1P0+MiIoDo$WeB~Vk2-qYxlB3hToRtAKkw|(6j<J}JfD?mjH0Z=
z-OQ>9T162CU=K6M3~p%<&IWU}MuNuRDqOw-R)Ve!($z8e=zw;M*tj8WFpZi>L?N;?
zyTd8>`L$&oJ}E@T=5gH~|9u)4vqWQabSUP)+)-;2Znb=4s!^&$Dj#Dlt4H=HEik+Y
zXAp*TiNZyLaiJJS3vR-bFHjNAn~QVi;q2s$*#^yKfsNecc@Zr=GbQ6PCvaIC11P!1
z;cnm<jU_092Sp4UtOFHFrRJ(VE3Ln!u5<lY7(CYjcdu8lu@9Q~O7>(Z4Os_dkK}3K
zY5GsXc$qc2X~9Ho5JR#H3O+MLPh}|pzE>)(SM2q=8iu?XlbDYmFMQC|l2;P7h${e$
zE1Vn*$ZkR%K-<sDp{i8>3GbhS3YpLpWkM88!McRoK|y;&F@#UD#5uzf>@h%Co&8@y
z3I-_WB?&jrrNN=ig-^<1^Nl724anGX0S*w8=~CH<zW0b)VPWrIqLqwYSa+^>c5S(9
zp_0iijKzW)I=^M>A4T*O1z;<jaN<XZ{eWkw==*;E?e(9>pWpuZwBL8fo7J?SEwm}x
zC0~1O>TkQEO)I`%(ora*!WkX*^uA59-_i%QyAr=@8?7FOY|II+9xC_Vxn=`PnkPoh
zOWh%AhbV$v=*LkZBvW5GWWSO-AtnySCc8Eq4KqcOjVuI9SEU4mHx0ATmAe?Aq?*bj
z5E|)7UVhnKeA^R;w}Y51goX`ltikZ~g5;a<G<_fxQ(>*uTc@W9ptqkqX<&8uLaO@|
zC={N1cMLTnDW+V+bYoOX_83p{)sLYtuSf!mUJ@ZGAzxIUGb`~j#oQ#TT4|4O`bA)%
zvgzu-+#GT_m~Q2kY_!@|tv?jXgRc?EwrcO%_uonrqodr+UEhDb`|tSobHU&H*RTJW
zzCXC`y2~B!-`sWkzK|a2t*thIH}kvc87$Lj@N_@znUM4hd>8tsfs@C^5(|LfCYs^&
z;nP2HGR?;i)sf<F1^=2(t_MgMYW}^gpGAuwmQTe-{Pt<?vMG%Rw#!gZg0jez&C8Dg
z><zin*?d=`qHH^NcljF<_HEhn2_yC0>%ZZA#JfL!d+^`ze3$F=W!`tPqK`UaQL&ZB
zUmI_lSSuN{I+B7W=->)-z?oUGFv*?*us^-*(aWcA7UABexI&PP5vH00Tb!Vj_NW(z
z#ri^PLu&E{)mhY{AL4dIDmGg*4=uc0#(+EJqS8ejwO4D8><H}*?*ux)4rB1l&L`9}
z+j-v$2~|f-MsoRx)Ex1e?i;K_%X8vRl&=KFH&7c~+*oZSuVVk|1~a~Zb8;eOf}f0n
zZJX;0&MWvVx?3MEpLM*ZYrAY^f<$vJq4>PJHn)v+c#a1pB04bnHH<ZC#BUW14G%TI
z4hn+jd}s6^X`ov^N86zsb+kvWXIU<DTs)1rG;#cvG85~t-lsg8pjZvTXdCO;RhK96
zXVh%GQg6OgtH0Rj*(;JzL=(5)Py*igAxYTO*U!H;2F?Z%=0eu7ax@33!wRp^5TfIR
zR>D|vCzVjlC1mJnZ)Xz7sEXnW_5H%ZjZ80<M$w{khvV3~)lH5ZF%A~NwlLt$KHBue
zWnt{cVy@)Atl_SL-|PF?H9X^dZL4`5#<!25#++L__>2EL*F>Fgwd{DEx@W4C0Aj&X
zlU?@L`^G9*d71wg`R`Cf01_|N{U883QpH*=dSQY<)29B$q)ey!J9N?)Q`ZJ4hn9@c
zbJ?Sp&O&XAeHCLhswh?oj<O}fl6<1(aI5|Ph~V&4x0eLy1}*4g9aj0xY=S7(eTP?D
z>JXl$)tZWO^m)NPKh?b;mexh#*(>Ta-1{L(69%Wkk@*1qr-ACQ0+Ad*jLg7FxW|z1
zMNAwL>e6s8!9C!HXZd3nT;==v@Y?gbcg{Tw>0FPi`s^f4b`snRB2+`d(vfS(6l7}h
zFKtwifw3RILHC6r)4s<#2c?lBP7TDZ6}xt&dT3XwO({3^_C!)BM|!RtdKSBG6FU2b
z9TD}Ck(T_CjcVC?v^Aj**$)ash<f$3RLCT!#EA(vKOtufFEB@xP3QPHg7a_$X2NXi
zz0cqSMez(|-?HuxTxZ`)e8u3m6c`FT>#HY9n&w#5V@1W1(91l^!1A)?bCf<Z4-{SJ
zx@>1$4QipVUIS=93|a>iX~Y2M`s|_rjz0STInR!lU*2YSv<k&v5bz}NT1B`0#Imoc
zb^ImX*O5Bq-$@nUb2mi%kAjQK3E}1bPyImvBb6}Q)cqV(L0^ImV>~?3mQmjIcMk5-
zyP~^=(LWC=(2U2y$Xp-M4uaxXfvLdD4QCeZ)%kR5VTZwUH1suS+!Q!^(MF&2InXWu
zHj0zT&}tlRRv80mwk-6f#Mcy7*!{;9h&xTIo83Jtn@tU?$GW@Lj+5$An|nXR=2THr
z@23U98Yvh}Pd@sor!qQJt0y)|jXPDe{YhCtcK?aV54X?jip)Q@(o(rNo|o9PtHf@Z
z*Rl4#FwXjbG96>_+lSNmX0LeU)QRAV_`%-d*mIqsRjbc01(vvs?TJRrzKkw%5?6=w
z>UKt_R#rv|@S(u!7#dia#545=>O{UK)_Am0d2-1}<Uy~Gf3zoZy<}wGgEq1ztdFk~
z)gMSaqpn{X>R%5WT!lXyE5mA}I!N6VI4;wY{0XnlhX+{G^l}Y?t@QX-y1}`SDfEei
zuV%3bUIA0nifP4(emc_2Xl0&S@<AThBo8xk5)@?YfFXPNr%doYAf(g&WBx|;29fdZ
zHGtJ=|EuL3+X>Z1`a=mws&5|rchd1AUxaY;-@>EEzC>~0`kR(ReQT!rCPyv%2M(EH
zl_ksPiveIMMypjgL=bd;X^Af=t@5E#V+~2#49kjvmKw$kt6vv~!mhCr#?KrYu~ej(
z(M){<!o3OExop<ZK=%=FfT5Sne6z3nI!Q2ge7rnkY-rkjQs&Zp)6}oyM`;sQj|~W=
zq=BYW-IdwXb?q#FWxv-x*Za~F`ZMzxyD-n$o=q0&xy;;SVB#qnm6l5)+Nok{T-E4J
z!S%g|cDJ`KjE?7uEb!D|8aO~5bvaHaW@G!z%;fXOTf{v!C9&6|9Z#m57@7F6v)NFK
z)Bt;4*0oQtp%$Y`#T{hv38nCaU{nKl`Ps9jqYo76_Ycf}=xv=}ZLgT@$~!wU2n_^d
zC&6?TKcluyrjf>^QL?KWJj%nD>xqUoy<|BsiPqts{nuB=gcJC<&cx#?#b^lAo=>rj
z@NT|xdhPVl1^TF}$3Rbgj-v%*Z-h&b;bufAf<bTyQzzM)1+H?PSjK6_bO$olsu#)`
z>eyM+*AWITn~;}NzL)U?otIG!x~pT>w|eyTHuQVfyS0%4V5VH~Qv50oL_DfycJbA^
zcntLxc`537S$B5bz5iLLH*Ch};;D7<nUAU?)TO>`P-tpSoF*@3q^-!)9md6=(3Vnc
z3xX${vy<u6BZwsmLPIUMW;`A(acHo}6Qt3jnEt1g8u@-W*Yx~#qvGLK*^pDav;TDD
zd7X$U5@T_}xy(>BE#DU_9@uCq$=E(pmK+}+z;ln>*3AjtDL}@!(-RTXiB%~tR{K7%
zN<??4`)kXV@DD@>$J}g|*W;S4R!S`@d*svCw&GPUMyF+V#E4=lcqciWjM8Jj_9G0&
z0r3W!0I#iq-`+5H7b<_<dN%+J5ZyjE-DzjQ^I|6Y;1xPehX=EVZXJB=NOg`@%Q{ka
zTFGB&KvPwcJ5h09M`UC{2m43=p4$TP)xFWWBpm!OGD5gOt!kn^s8vvRc^19uim_7<
zz}g&hNmca@M^JwR&VgxsJ?BI^?lSd8`3>Kp3f9d=fVZ#q4cCxX_p&M#AQXr~WfsBy
zJ#{8W1d*f4;ZdnHTj}6e7n862ofCi3xRflD4P`U$Sw58SvsdS>>omNeN^fG7HndOg
zN@Hf6IC$$wP`3>n_e$}g(jghc1yiLN4Y8m&d-oEiD+8@-K?}%*FjgTNtJxq?Tn2cx
zRX?*mS%1C@c(o$}=OIPeN7dHS03wAD`{V7^n8>0Q!S~U<k3x4AZHM1QVc+Uo_&-MX
zIX4$Rj2PQ@c~;{;JIdqsx2+ZOyS0a%TUZKzf}mf1o!fg->FNj247mrN0iFk4ADV2w
zJi300+V7Dk*Zk(ubK7R54@xLmSDGC4;oH4?bZBqofw@xS1Y{V;%t%cm;=6Peb9%V6
z=5X8k1p#?TdkRKVx0;vxk{yj{c>-syF-qeq%Q(iv8)FseS%2@~uCpV;_~C_jIN-`^
z8fl5OHS)uTAn1k|tuGwYST7&El@QhWVr$_xU!rYS;CA5VG;`X$>bwI3e(cEYmGzDL
z7j`fpuRbVj-~I6aKHI_GAvyW+L#_}K3{V?Nm4l<9k+1nAKS>Llul6OMt7Qs!QQKrp
zTVpJ66SCd1kr>&8Y(_>L6+Uc_o(Y`_H3MD)gw>5#jfZZ0mtjiTJ9|0IXaAwZ_)(5!
z>riZ8YyqTgzLACiQa!8u>?FPpP@(2X3N^aG6(fBM8EGe@0evEj9e4U(Hk15M6Ry_k
z->N6;T%D~TL|Nb-VLcK`S&v)?<d`X`1(^X;#P{lSjNM^0tzoRS|Br*KxeEvsA_Ybb
z9+U6`-5Fvsi^i+zqTs@CK~O#21}>}js5rV)A7Gc|>tOAMh@9z1zt6t@tG9`*Ha7qT
z9chFUhBjXun^?NvMsM|rW^yiG|E+aoB7>q%&L%euoP062c>3u?aUNBjT<qU)@}Dw-
zvvvQk(k3SA)9{_H2e6A7e8u)y#`0Tm@YSjHC*9;0OmxzbMk~VLi6^0q{u<}C)^h#w
zfmUyg;35g7#UoD#YxAgT<b40pL$4HhP7bjKsin0kCHtr2fS>t{A{F2aHL<uolR+}p
zAY)DNd*3Y^g9#8egF#k(9hg<kf>fBsxG>o$$SYbKci{A)E)fH%r2-CqAsc63?A3pI
zy@y0F$aEvbT>+X=&_x4%NFZ4<NJ{~Us1$z4$ZhN`PK~VVxfPJNaj%mq7A%NRFWTcK
zjRw#XVcPixL4Pvp`|@&|d$RLt`wH_L`j@%@;3&FY1jT<bh*Z}z(o@1n0o@zs|M9(m
zk!icABZ<_|(bGms1D$5Z=(-z3>#a;5`alP8aWWuBL(roR1dTwhGMCb&SF5l&F=*vo
zRzg)=%HB#_iFY|!ts+#qRcL?}Y}M%A##-r60ab5#E@e=KwWL>Z3+vwJ{QyC7bY1;H
z;bykC^y1}4uL8Qp)v6^o=aOz7z5Og@^QPo2qzA8Pbp0-wxSwLYcdw7n7N*CKvY$$j
zpuC2*%0687Gh0)i_br8pj>_0}=wr#8Z}YNh6+pS`B2zC8UfdoVzVc$>;HB-+d$C{V
zT>!W~U#GlB13J6b)1j)EN`#u6e{k#R`qcWvZgMj!ilqceFZ!W_0s*34rP~%Btw%>G
zoP0dRFD->#MS?(*sSf!IX)?jc-;Lm|OD&_9hSi4$>%ILq6M)gf7lKQ#ydv8?gyoOW
z+=(*r&T@=Cxs;F|nj!FqD+b$}qBLM=*!!CSHdkPQjya2sl|$hmI-gOX0=`cJZ~KD@
zekxR1nOJ264Nww=RKZ7kBoqOH(jBm63JS^uW`l(Yxciim9oIP3e<_&Ao2QBlSi6_2
zA%ReDMX<FcCQ}SS>U-HrbM~qTLqC<rrixcg*{a5q#J+@$>YNR>SPA#Yc8mYF>Rg0t
zeH8aBTzQxAE;G-g+WmDitVdY5V>xb3>+=XT<h(>@ww7@%>%Dt#X)cEahr1bHv4j(w
z{FFYn^d_QVw;5gXS==}mWD9bLlzT1ZmICgeQg4PICdtvu5z0YR(;fXJK<Z6*=;cGX
z2(Kg?pMYNKK_z^GRF_@4iwoPw4>>#>H;v8WcPWHi?F2|8nbu{r^_w=7`qKR{$p|k;
z7>zau4CCFIY>T3{A5y~I(ruT~?mFz!Brb>Bt?1;`EuWL04wn{JxUrIu&{XrQMW#GE
zybHxn$LPckhO_fZTsxc#)lS!V5fbbnZ*t=OnZe$rpCKUQI&iel4tV7lb^PM>Us^|?
zdK)<1tr!i8gBqD=7oj2b*bXWl1^E&6KeQc}3|``4pZkp0{{^pfHC-=&N?n0#4xD(g
zjjgkZsmMzMXCDERH>LN*otocXXh=ir5T}F&M0G!?%yuJ~TrSa=8!U@M9SQMfq($iG
z-&3z>Z%yE?ap(arix>&8O4%`K^~5ZKYkY#AdumEddb%??D#C?YR8&<^Kyr(V^>ay2
zk4Z~+^^X-Ii7GRK?2^cctP(*;mN+6RA3_ok2dHT1tmzbVc7pIy#HdqIQltb0rb3ec
z{=}RbA<Fmu)%NIB_Rv);-0IY4wJNcSevho0%SjROfLk{Enft+!E!Y1rAy(m$T@6`!
zA23W*-c(6RS*5Q&z6g^YmLs4zbNx9+5s+g+_ey9`uXSZDlvzY$Wrjf(E-U9?edS<b
zfG$z&UlGw15v09*d6kRCYbOL~za!}_h(Yq#s8$Ycqy&dVqYm50pP3bKOYEOWOO|f7
zN%DXC4Gn2e{7<xF7a<()YKTF-H$TMy(hW1ng%}hXK&LPP1xvF|PIHUZ--R>t!td2p
z**i1$UaET0X>Jzb%J&=9(;J`I1v+};QD~1$jQ<~ymBKuBVPGfMLl@6-P0DQ>YN%-$
z%g<t<6ayIU6@vx#XS1%BK2>pu2#64*=J4cd2JL(=)dpS{5D+-VS01dZtn*3KciC$8
z>ZK!8srl5(3JGMg;sL`zkYzm{9hxE{uu*iPkE_@%;mrxvy)l5Oqxkh?WrCOh5~p}`
zP=eoO@#YR5sH1!yKs5|ZMee3?O=p0nyiT4v#m}bK&Q;cC#8CocvASzk{3-1zStVf5
zPB0e~8$|3mmp>$T91Esg)9i-Il5&}C1G#yv1DR>ro0t2Jz@)O0vzNEcEOv6v@@K<`
zBf%ucSr88=!PGP+jhW=1t{HALH($@JWpNtHok?rn{}n<Cr_r2J9LhEp#Rz;_O-{6K
zeKV71Li78T6#9EZ!dJ^afUb%bIHa+2Xe8exS}1ZI)nwSjMtksms7WEDx0l7%sI-bm
zIzA~Z872aijbI4(<`=o=?%=3ig~|)FeTA^z5%&kPQkm~#9~R^qs*iv(b?8orNK7GP
zFQdek(nO%pce;77`2;EB!nI!~uI&KWWqBhv(-h5LeL9~+zdBFyL%~*J?#W%+(oDQ`
zLlK|8r?6GFAXlm+8|^}M7faBUePq78vGG@+>_3ZS5blIrzSZA%{z}NzJAM6S740Jj
zN{hx88}3DC0b@%zmbhI2YA;hJTo%{gXIJBw5<#WOA7UdJzmWu>;Rd$1<D%d3t(&sn
zYE`63l|+@FA)aaV^CnI%)xGsjg2x}=zBc&RGNFFvh4B&u;Wrnqa}a4qLWsf!<EuTh
zc%3BZ@GM29A<qhJ;==K6pTnd#B!(E0w@k2ept<QoC_i=bJQA=R<;b&hRvfHB-b_sk
z@Wj(>d7+QeX!ndw#aSG7xPHrdA0f9XdMb?hh4-E$j>U9j7HUV_Z*7ZrrPzKkTbvu!
zR@Z3B0_>fPaw1BC7XVJ|36~UxfKbE!p0%oq1`^eSVtg5endjBZ@a@o?(m#HN8N0Ky
zVc`+<t7NiiI0BR6?!jW>aV!m6Q^0Umr5wcnr=Z^;?&UwelP7a>@gw^Or%~11Gcq|@
zZ4!3->wJ{&MA-1~UqzFUg#CBI8vh_mL5K?Qu^`SOQN9L@kgz+){2G)E80s*D2zh}M
zQ}5gZH4s9Qogu@lxxFYLPBb4Cx7hM9C7-5Yn@i%DhQzRtkQ*nm&8bC(cjiK%1^tRl
zP@LfyLr6DN)JOM|U_8n+`v<nSV95_wT)h{$@{ID>IjeKQruFVyT2%<fr#ER<D;B3&
z%+pJVV=U}zyi>e8$6;S}Ol{PALp#*;Pj4)}b97iTv3?Kqm?BPMYD#J<R_&RY*i>o?
zF4e37u>SrBt0^f*0xO<nv#WKc65u^h>a){uz$4lW1I*@B>$0ZXQ@%$8m`L2qB{V!X
zC#uYW3kbE3%t?UF$#7@kY%coTf1Hd*(05b5s9Hgog%Rr0RaAB3tcqYje{~k6)U#6!
zmS=<L_?-icOgYxfG2D9g^HV|4uk;P8k)u`5>`w3z^&CdBrBB1ic!{G0-`#hzUf&8%
zz#-|*3=bR@9_d-Or>WAv+p0{@*Zb31Xl_)5EsV;*B8V1Pr(jq>h<QLywK>HmR#X(9
znAcK@U@#h^ejh5WDTDyx;3Iyvgh+h`FILMyME*{5LL=p3hsLfGlQGq`TqH@~Bj6j*
zvHCJje!W1x2i3v3@q`x6Z0FEl2f*4)J#uC{>P>?(tadA(P`yd|c_c{Lpx{S3fxFqR
zIBcgNL6`mgy1p~h3_B?vO=}C&jBypr3X=n>-vFZ?L{+~>J{tqs%L&MEPxj?Zc5Y2Y
zP(ol_5S={3$FnSA#MgBlQyhX!Lands)bv7-cOc%kL`{`9pCW;He-@M;%EgCu9Wb_$
zoon7lcTZPZB-nT!!tJLkar?6sLj=H-?RActoL=OsRze%wWyfm=eE?^<Fl@}C@0$t6
zTB~nrA^8NltPQ!S&C|>!7!H`Ns4@9^y5#&tA^TS-WX{x;A79s6Dgr=3Gy9g5iE6~B
zUMY!8PnI?)3>O)B*F%*wV-IRxGZQ)99rp5UUGv(9U%^`=N3M5FFMqnIHi^GA;^IGf
z_}tOZ*6Ra$g5ZvmuMK|1?Ej7T<SCmCWE){7aep?iZEhy9-Mw+&-QS&0zkp1XLUYQ)
z*IXO(={CFxu(Rtepw63e;at2et9DEzun*rAm~}STJM3KPi_aE19++{Ov7EE?ibUch
z+NqDxJm;Ks7D%>=N-cZbGQc9N^S9|%ZDw3N%?5t?HI3E9+Z@%_re##!+Ub&yo2Kui
z1lzn0;4hxC>6G)MagI7u3IhtC^~BP?yXbMsf+k-FS*=-&tkB4;x4<-h+P}(-wyLrY
zP7Do}p&=NJx?`N9z0k3`TaAcfyTZ-pQZBA96m`GigEgm{ZHk?X?zYq%qljrvC?0Ci
zq2Di3%-mrIcx>TD{6yRZOG$DPF+rJ^A72Ofb|CgNnEF79_K!XBDuCSMfIj~L^vwub
znx8BDsC~&0U`9`dNHt|6)dZknq<Z=^ppl=92j%mSjADw*lkiBc8q7Z?!hSa6%)PTT
zxb<n;J^C4I8`|mrU(V94;b$4=wr&5E!`ZITuC~t&YLIWcL;(Z6;&(}`Lg?gyI?LT|
z$P-6E=v5&s_{|S-L#H(`p-M7L$3u!x115ACKYab|$m&clRljWSI;?xcM=;^DWSHI#
z$>>EeVcw(3qdmt}o|xx0cJ$F#*ppDumkmrc2DXl0zrJSsH{rvQl+2Ra`Y+aWTpB!H
z{H6{Vk1qiB^Hj|Lz;SMHmhNYyiH@E*@;lJv598j4THxmp8u7OyL6{7tVHxv}DYPWK
zFWrZUnvP5w+2l>$j=caz1$7{}S3)5h4CdyBzP6%!qwzSC11VK<tih#xZw7mTHZL_p
z@%ELma`XQWTK(=IF6;pI;>ZDGdiwvz;BJK`gs9_E4mY&0fCT2H@gUu9AX_a?-4AlL
zq4LS?v$=)52tKH53_m#2n%ux6Bb18mx`<FQg-3TmfaE2IrMLt*?h5TcA&7!h1W~C2
zUwtB(_z~q~*=SGdOQ7K(>~tsfg*?g_dVMb-8>ub*(^pteDrbs-0$@_%fihES{$P2b
zIW-%nIAgV2U{}eA0}2o6C1#MSyUmB!KmHEnF86XP&*I7}fudga5$5bq`1kLekOP*U
zrAJW;IJ3}W5E}F$8<){$d$`Q_%27TisVvqz50rMu#XfTFW;`%*fP5jBkZ@nd$MEGm
zyJ*E?Rn0Yt+!Ar!g6r~&nW`RZ&DM>XEq=6X8FylBDMa$0Q8l_*+WIJ~W)WM6pxirP
zw)JNd^9xahi5oY%fV<(s-bs4S)?*7iOJ(TD2iG>e|7f*Xuq7y(n1+9tPMeNq8Q;aq
z<%jq0-Jb0IM{n1+`55|ierx8o337Q`DUwoMNfh-=!d*xZcOS-%J-#7ol*oT3)MTzh
z7n9D}yvQ?B>9<nm@J{4?rPZFz@7~-9WN&79w|tF@XKzID5Q=%*c2O|v8J%_0^n6lJ
z8X~zblL9v%%D4iu-f0Z|27Of8R3DC|B4~bEjE|IcjRVmb2*O7wDZn-3)aGp+XfH~5
z;$MhD2@{%Ni+VlhI}J0D$wE!dR4lf)wvE|-+-My*EXqiEN{OV4IaI_PKyf{CiOy19
zh;p=yOOm+us0KopFSLXmQhKg;eRo%;0$%L(Cr8wAw{EQ@>xlQDs)pKx2V$2D%P{BV
zFN$G5SC*t16&IG1;2Z<B#aBeTMI#lEvr?pM!I_o<GtB7z6HXSH5WTE2TCxOGAZ$Q(
zS`4IZP`ohAZl*9zk~72z505eIk&lE{DzrR2N8=}fOH|34>_{1A0!Ve-hh6<F<9%Yu
zr&E|JG9sue-4f_x3gY<U!2uY6gP;ICp=Y`EpT7kNAcO$M4p9;4BR&VnG;4rD>LBv)
z9mUn?IEN4!C<JLeT*Q#0t%=$$f^AFz*BaIW4S1ZvLbNc-0gSw#wX(SgO0Q^4J*X8L
zk;qFuO)Y<MI_hPC`2zY>#tgc))(B$eM~pj_v&e|C+Z>KFuQZuNE69;=F`WJ?p`NT6
z1e<YV)D8_>ElTNeV|1X);)Z3E=eD)W#wA{b6Vg`xWeT$-KGO^f)VAW&!^*viKo&%q
z4v+^{)U0WctBtYUJ4CTv(H9Ae$|<aFlQ`|PMuByH@4Cp$`N;J2M7;my6;1(-H#fzj
z_HC&maH_A0?P3ZD)kOnxN*sqgpH^X{q=zwras&aWVik-wChpLTVOvJ0ZQIyhEEE_I
zWz1TPhx}MYvVex6mE=-KN)D&27J<Q_rIsN0YS~9Z&?!pz=$O#~PdggDl+ld}J)dj+
zi6sB?*r7AL*wSAwwugbHZ(ck}9drWI4D%O?O%DeZecn`+31-qE#tdGmRbM;g7H>*p
zs8q8gQ*$dOx1}6xXkktn#1%_cLcudm+Hv>h<S}Nnl<SN%$dxGWffGL9R&s}`S-?<6
z6S0vvN)CyaMu}{PtRcZG7RU@MUX|dqQrI$v=7lFCK~tJOmk=Ux1{LtcDlM_m4!Jpl
z!0Dm7qdvuf<|0-AI8n|9?RBQS72aDlG<suY^NI>FYYeXT8t>c7N+HeTeHD2Dy}wbP
z`gJ7c8ed)X{gJw6{3M<3yaxH2(G|#H&QF8SfR$92Z<_o}@0;F!`~PxX!M+Lf?eFw%
z!A&7|r}tBq;a>v9<JC9gAN*?UTt9p9ilEs4;L>B}`+4>cmtwaU&z46QK%a18<qm2j
z2Axo(KwCf)uSf2dCRAJ&GAOoKZOs*Ic8tb`9tA4|X4N4MC7I@o@nn878f?f8<Ucbe
zt3Z2PgFWfCQUoQY&k>}Y(KPu-8Q>>csR7lO(V1{l9s2|VBR-y_tOP}q5`Yf|L8Wyl
z5ZkVuGKV0~Sze8V<2`!+LmrzD(#R6x2inN`OQ@rY{dXjjS)n%p)vzAz53t#)v+0%0
zF}i^hN(7fzdg?aKU{}whmx^PWPSPBrGRSM4j7NDYGis#|E~;HiYCkIwNsirZ4IVG3
zV#HDamzwRj0`rVCtV8(#@<rCif(Qc4*cQx}-Bds%z%L=dR2Km<_M?sW6=u|?&~*YU
zJEwu&u7bh{D@w6zqj+91UWIig2a?o&Y`B{ziXcJQVi<-Xp)5U)A+PYaiKIdHVtC{A
z>@{pE7ROASEzoqGTyPo|ub?R28X$sTSicF_MAaP69MNmqV+~Gqg2$2giM+TP4Tu)Q
z?Q$AY%tv_9J1d$aStIBlWJIB%tlAnbik2Ac9vm0b98ex;3Fj*9Lh2(B^@LeQ_C0sN
zzVu0?TS(XFz-eCw23ge@Mg%E20*{q(L{K(h+bT?SQYm5?-7hP}A+AD46yesR7AW3y
z+=<|YU@^rN_G5c`o!dFJiDavk0+n(PP_JFEL=Dy5IR(s01kpU=DbY&{lgu_+{l-+n
zszo3mQ1WzyWGXCLlr{|85Ng&F->FD)zx*6p129cE*$P3Rdyp<aVBA(8DeRJQsPgs1
zBTU5rRXAdUA*B*MvFXo>|39w407y?S2tw2E^~1LxR}`%bpMixlj(sET@=(+Nw>*0*
zT_nyZ=S=zsiK1Sl&j0@2b-lA!IinylJs5~j_w|6P(o?f6%|gh(u3X+@mDzh0GBk;i
z89<_oc#e%c^N<rHHZu-LgVwXLKLF}!0pNlu>a9hwWk?dHCd?~|;Z3tfU7`DeICuoK
zSQ3P!KEXB;-D5q7GEI;RA`g3ggKz22;J<(I!yRHm(&K;wCcJJL9F|?E5eM6FPJo2t
zFKspPcOdo%Q?daOgyg5-97)o5K$s3L@A_Fl1wH`HHMB_N8e6tNiq!C3V71T+I7k&?
z5(gBoM02!{Ne46m)AVeXMEc8=0%A3__|TyB!*FzKkvam&rcY8SW`d;7MJ0m*nj%AZ
z<?av&2FX*3(6&J{ZeO6n5h@~x^RSBI8;(a)y>-oGQUg46(TP~MWiJ%@MC~O>Vh>TO
zz1dc6AU9B0YrODlf$ogu!d*4e$-i6~!9jGOVFI9uNf;NHC_&FMs?maVwqN(ImyAF%
zCx^9MGms{6?hY9CeWF_wjd$%c6%gKwfl3hkQWp?kQy~c?tR=95a$AsNv4af}GLSli
zeSlybDs&NKn~#Y<F4dl^2O@;qdtZY@Le@(-#z*UAqS$2ry+$9}q*J@jKY9#c8|_v^
z=Dv>YY4K(Ld5P_&R6ziUCAXCA32T6w2c^tU=y+vn4aw&Fb*!;1<2zMoE6tsS^{s7P
zJ-P$;v@K38auQdHj<Sk~jvSel#z+s4+azY)>DM)6A5z98Mx32`^@5~8Nu2G8NX7$y
z0;NfDYbczyX#wkn3Clq6ka~Fu(1OdF2j-CY=}Rt~O5EWNNoFhf*tl#)waa>5BG@pI
zznYF4S$PGw*f1D3yTz`mGMqS&^jFJ~&4ZecOKOEUbSgQDImGH9oM7Z$JwsslK04}9
z90z+-$`L-Mq#&aDlmfLdJKnP>3Qpr2SSa0k>H(HSIodZvXM7+C+~8nG8N@%$fi+#s
zSZ`7a(8I>@W}sXpK5R4*6{ysU9{mEuLMTWF;X-HxGV+Kski<nG=~i-PZR~8Z)aXS7
z$svA?gsqf&ESxE#K8(F1j$_7Ue-XfOJ%I4+amwfWtZ4Gal?eNc5=Ef7feY_(qqLBs
zw1#m*N@s?EL8;?oK)2Hv7N*QAu-Us<pyTRl2Q6}I>fCBM6{xCqE}=eRTvqtWv_cb<
zV|Dxos}7K^ihz)+ooU7d{Gp7B2HtPs%C*J5(50|)ib6^K<;)$Vrit-_ktz$)0g?#T
zXoKrs0BQda?joCC?+lgc^UZbr^LOX(|1oS7%ojBCM&z*)uC`b%2DH8|X<Jx))mbhE
zmy<EKLIOKROeFjcsDPO}2k5<EW=&6$T`*b@C`9z1M3~VGA>>45I|hw-ZjcVD8l7Y7
z`B=@@d-$v#*!yclKxZSi=7r3-nJgg1T%^)JurJnE6W|XiN#R(lkw^>bL;!<~$Xi4d
zg|N;JZ1kb|Abpq4SZjMJqlDQ9sm*zCuV9P8YYQE-GNgG)A8Z+#jSnFrkk#RV9y8wz
zFs7skA#fi?5!pmdiAj92A|>WCK09Mx_&TWj&{uaa+7r~N-HBL5%we&RW4c#=g&ynU
zMTO0SXpk9c&~++-3C@!-(Xwc6*;(|g3`pyR7%OD@*oU&W+=Pib*kX(6Z$860vYE?`
z+#;?kT9-$nr}~XKL-EV2Yd<#9MOl565FwF`bD%eV?hvAcm90ba?pmOvBEb~IIv5j1
z`B6y5T^ATLYVUxe(ey<mik?nEr*QjZG43XMpVRcmdyU~)Z@?n0yD4UTJ-u}26W3it
z5-e>{{uNuyr(t0HQHqGpgeFm8)B*Y+XsrskUN<M(rlc8Wq}x-PO{F5j#}w#dNH7@C
zVw3PYZRDv)yh`S?!(*|;==ZRpeaw>BIauTg%lP$vR6Z;|Bnk~*D1;-7XoT-jXXZEZ
z4*k9SFX{|E9(y<OrClWczUq8?^wrAjmq0)Lj#CS(DQ9D|a+&jQZaqT2&rj{j1M1b6
zu)g{d_dvh2|NIMu8hGF`Y6RrTy3a;|{6aZT!b5&|XUBS@`G3{$|3?k~zthy)emsR@
zoe(I*^y`&rofV=sS;zPwLUumCdjDzMJR-VD1?+Q0AEfQ&`Nf+<0;N<cE0i9I^P)pX
zj9(efZzN8Z1H_D2)KcJAU~O_RUdyqJj!q&dA+9T}a6=p&TEeRscXg<7q*RTmgYJOs
zl5&O%=GM6fT4u8N!7etAC$q(4{lNx~Z_B7l)*9bc^p)0d6y|pz**>5GutJ^?QSy0m
z7-_N)6uy)Dm>Dofdvvn9f2|YmhXdO1IJS4M4l2IDo^=HON`e@@qk;!^js9<0^fq3d
zUriHgn^i__a$wK#zkQvM^>a$Pi!Nel&4wsrKZD@Bs1FAhLITvU6B>f4x_=@BYB+sH
z&V_`hC^JjbnnD|blP#GBIfBg#mS8XR8!26FK~Tr+%#5lrHPo=WqNzfZsK$)wYD6vK
z$XzR=)POpcF@w;I0!@i_HS^kn&r_RFy3M%8R4!WvaL^Fr%QI9NS?Oi}^+<&V^;x0-
z%q1-uEgT6x-Kjcfp$H1MC*U@aVUvOCE7ihwX2rxgGHkI%t5~`Mz!BD7Or}_ElL77#
z5$<YkR?D?$-e9i>kH{^Beo^ox4#=HY<#OWW7-o^aOL53@$<G>fq(KG(6d7h#TPWt!
zkwPs&X&2;K0YWh);gEqLVhI=pN3|coB4s%*k3j*DVX|>k@XMdRV-149^o^(Weko+S
zFZ0gD45|>3;D}9(H8^3kA3Figkpa@RA-F+`182bL4#$392G!bW0GB6|NVFZq*CXj2
zPNHpY_#Jfn3hQJusN2EEMq~)v$qk|Kq@$G8z&lDa#h7LxD!OR08513aXv~>zFn)D{
zrPX$9X(422?5G@pIbet&q8$y-@(o%Q6t*WZD-Y9v2%~rrSI}-3x53?o5U45X9U|ay
zoD;y;c@}B+3c9_{`4u?okmR8$1P3hOw@j9+X5ZgJ>^w9ERcm)f1isbT6Wi_kw(;%K
zw!3$~gLhYN)2a~V%+ZdopB{MS&h0-M1`(LkJ#&$X0vI~xdQnE(K0lpGbb(cHe9s=-
zutOkAMK<1cB_>yJ=9Ka$6w1hP+D1KbJ3UD%ECD;_#RSgMJEuPAtkDBWCSP8lz0OdF
z4g_7of_vJu^-kGEDzI6ndK(z_%fM>gTIBlzXivdHaN*s`AP?#WvfB}aaH0sUI*?_C
zeTzyY>6Fi4b`Tsgea0WRCf|ceud`Vt{3d8gQSw~<#|$?#AT<NLLek3c>Y1Rng=AF+
zl4XT4jSG_!B31I=R65&%hk`Po1TWt<fwLIpehOZ;1umP_?#cY)+2SKxVdL9IC;nIp
z^L1J{p!T(f7?sSPA??`6DZ3N2pkE+~7U2ZE5#iJwu_=|Z&vpB>9gMUAIN%ZHde^EN
znyta`{Nlna?@Tbkai0{tSvJXdN&!q64PD9_9kqHj)-~BWb}89aH$g{=lNWTAu$-8x
zvd)PTfR^eZ<*_ibY<i8fd3#ZhSb}>n<jrA`8YCQ8Nl|4}oo$?Kfw36EtK&7teS+zQ
zs;YFUg)(RztvCs4f-(|I9*!uE*kUMSBnx&CGhwscp5#afwe+Y+GZc_@Kv9`0i?EEG
z!#Yfbf=M0OjLkIDgw4Kf%Kz4t9FWoC=t)nLY^`Hq)*hPPbSoo53)6jQLo~8nvNJHC
zY?R}&jzduygrFrlx>O_n=Kj9vNHPMUJDdkbl#P>N?WNV<(K>dx?7rI75z+E^vgoV~
z^LN7A8oS?Z=OV_Wn}glk_vws`)W+xwX-d9ZT<0735p|}IbcRN{M(XidgJTcyc$`gb
z)*6j0k0alsj#143xoF+m?jCIprO_fJ3p>+bwj|yi#y=3-*V=pM^I_q2{Ozg88u49H
z`fY3bV0fFoRtl@`Fb!nG=Eu{-XMcD0+hN(Q+No+xiztif9SIT%XJJY8&<Oc6ltJyE
z)y&!uA|Pa{<Gya8Oscfa1grrHfR}p|HTD|ik(#!6Ecy`Vg|xK~`45Y-S0WL(Y`62N
z3wRdIDPv|Gkuk8s9l?(|dS%b@2A<ec;!qBu5Y($W7z)&OZ-GE>O*niiW9Fy+(_J&J
zSVs4Eql-tzN~BOO2f$?+SN^&`xOuf1-^sm(ns??E*TwR*^oW4-Brts^WAbiv%?`mC
zCS>|~TD9+RHP|0drtR^kHRTRHa1VP=+nwu6xt*O@z>Libw@OgY3@T--hg+|%Z;=@{
z>8#t$_RhDv_uaq-ZMvP{wste3TTXbtdv;#=r`=w*cjtc&j>W)@q0NF_)9&RVdpqNc
z&fy{U0K}6JG11}6+K0;b_8F7&+h_i(ZlA&1-(A*JMrI#U(haiYG}p#s;c}S%&PR3E
zDXIHn_R-GtGYi$7NipHP3}-v)cW<Zf$X|X5ea3fhvN<?Tcd<;oAO%!=T*@K0D{-%B
z{b`PxEPM|(bPw--8*ux?opql4a!<rLl<0eC#p3B*!dr>s^yYT4Jb7P8-juOd7Ox-j
z-oJajqf@R&6Rs56fpQckL;b3c?APUB->c4vtHK3SUT`&ufPQWboMl!5%^kTbD+g6T
zs0?0VJ%@PK8MC_G@*E5Dm`S*x>(6nA7H$XRc;AxldpO(nW-uw^jn~8|l%c7N&TZH}
z?XjH5&sP$a#vj?8Xr*f?FF2qzvw&i4V?=*RmxbZ9)gDjGYNJYW1Cm{KpQm{i$4r2A
z<D?8w51qg8W-<CD4}luyiX1j_N?lU4sjt-=_e^8TUdYa$bZ)@T%mQ=L4(nMcAp|kA
z!7!`I512$ZJ2P#w{0f87t`G6HIH*4PnKLh7!5we!2DPTonNgh=y{d<2H3zJST*IrY
zOg3@-VE?fsKbKMIP1FNpCvDhoKvIGT!qwSgS6>q;zY@`0xF03YV&P(G!4#f{bt~F1
z3{m$hB5zXda>U8OG`7@534j#$nO)PSn1y@uXF%E-EE4s9a2$oGP{*%-1}<yg1Y~Z9
z%B3V^<@(B{Aurs(2k=lB;LV$x;#>d<@ZvbivG!vp)`2cF8R*^3?RSF6c9C!_&|rx<
zc#fgGAiKr!Xv=6lHY>eCt^ui(@Ek>B6gnL`{rQC|jnC5T10mxfO>~#Q3W??4ChC34
zOW0u^D7e)<Z)?|omz&vPy=}7#btRqavnEqxW~+;`cI2uPZkpb9yZt^T_ODnZc}PhU
zQe4j{9*ToXlP{_L?25-y+7(JnFBZRAaJ|R<D8U04LlLZqNiD9)5q&QatmBXknbVWd
zfebsu3j(2pc^u=-0(Kc6ZKeDmuoh>di&^J-wq1f!*^;(8;R<$IN`Dhq_&`<YDm&^+
zRq6TQBD4&$dR*~l2+fPtQCyx^IZt9|(r6}egvCCT{r=Q-E?+P)wGCpW2D2Y4O?_Y_
zQ3=|avcAF-TsZm7k&Nbg^)&~(Vkk+zpp~-<Zfk1%B5yxD)jd|R-Li!lDBGF5e9_Sd
z3m~B@3#6S4)~xhHM$SK-zhtaNuSxOj`2bPg)96K=4FsFjYFSsp+h^}qP#h3)BkatD
zO&X?n%7Hl{`eK|D65@s_kOZqjjQ<3gTMJNYZCyNfYRuOeiNxBTcVXXk$8~c%cFX9{
zqn=Ix0ozTLcJyP30jW<X=xbF>E5SB*j3)Z`{J1ol9c$0I1O8PRS75BQ2Q%*gJ--)&
z?K&0<lQ9hi$<s+S6c^ABiFxfT*4UzcMgv`v01|TQD-{+`Sq5$p70Jku!7_D)D`g?&
zlOU;P`7LJPy`HkjMR3edIXs23`ZdFb)TBneIfnen6a(GN3<G+>ppN14o!nY#GN~@z
znkx1SpW-E$C0cz(ig%qHEMTNp>sdM=Cs3*MU4jWSEBmnHA)d)6po#f!1^l#Jx?LJM
z&}Ok%5k5_{(cCIiRovL$fz033j_)tyi5X3wONHArLPZzl<+Z%8RC&&PFm7Yf?HGVm
zCm;mlu-GYLF>ZORS0OZgDq`B3ExoXG;{D=U2)C+qqNWaxC^uw6BX(pkaU3FiTmtl|
z;!<e?q6N;owo-pDzId|lg^+M+1DOVu`&mTxLU5-a<fjaPe#-m*|MwzpiRr}{qjdW4
zyZrzE6)OoAfWm2$53<vJ&EcZRE+)bc0T~4G!*hI$vhS9XY&UREVDWnxa)&spE?GXl
zOjFX~<X|t7xw+5+Sdb+p(c77uiV{dgGtZeFt4Av}hJvw+wowz~j4{hH=*xE9)3}3!
z<Pdo;)majG*a??z{jM;h{IJ{#cF}7YMYo!>)f<$ZRm{^a$Pm6r*W5C%y<9G>xs2KU
zU`eSaKZO|`a9S+17_ly>RtX@L6@zFxmEQ}Bg)MvGpU{Nnj(m<!#7e#;GlE({(*)<Y
zm>P@j5<23sGiSiwlu@fpBSocyvTm3uT4?N9*!Jym3EeqhznqqS6=TM?W=6AQv*+Qy
za^!G|Mj{*g=AWavzxQ`)r%trBeVFL4_QbC=9pMr!{KA&P`}S2dW1e6oo+bjI#r^*+
ze1<AV&XSu0UGjk@4rnjD`8Fx@6d;8tHv!3CZ%&SneEm?5Yty&)$1&fhd?Nt>>cw|)
z^4+}jS$lHmc-t27Xn?x^n`5Vt5jC4VxkDq%7%y{@eZ5r}VJy;5S`{m!ZFDyv0r&<L
zP_GF9gFy+SSg|kvI38F8ZOd7cM~s@b8PJvrXo)tpmSg6nP4zL+z7g%Y0bR~4vDubN
zWZMF)w{|gN0_wXSdhfWe2E*THnWd4@a)q^BYFib4#=+n=;4y9k*w+6AJ)XXQzw~rH
zRCPWzr$&zd%RyTWl~liVSF_T_ptN?_#*DxnYf@izN4?rijhjV2M&^Jg17g!#;(e*O
zC_`i|4=evGB^x={XU!E&*6rBmZw3`T`pojku5$~x4YhskzA8U?->Lx{8W5E2ZP3cS
ztlHpF0ebs=Ik}QwTO`_)8X#N#dTkr$wB~iGkgapdFKipItVvPBb*ch_xh~UOgQYw+
zQ``?}4rrt~-#b%sUI_=M;n*?=CGcX1*qEmHqQ~OU?XJRJm4$QzN);(X3G$3<z^3Ma
z<Gd2;{Y<INB`CoVc&K&Dg6?taVl%i-R%Ou!YEURc3G$F~5#yS%F<^RsKOzk)z?{rh
z)hMVb*~~-j1oAPEb=VRQft&<XLm6cFY&BfmjL*gh<HK2O%*}j=8f#H&rPMacuoU%7
zoGYrBO&JK9zVfLZKs#3(gZ(plYakVyeo5p`M|^@g5ac?2E%Pt{jWivAEQ2?TaGQ(J
zMKOq&F51M%>SBb3rb|Q~e|L#_rg@jd<c#Z*O2{K2lJfiq_Q_&urRmZY*jl^vt!VvS
z1|<il9@V7+y{k*rqAy!&*1x$3T_@B2s_WEg0(70WlRZ5Xe($H@R_RR?c4jd8GpiIz
z3jBk^lH)ln56|Y?gJ-2mYlXn2K~MP>1Kcp_;a-AU!))=A#=$vmmdY#xCtB4a9i!FU
z8IqZO%cSw#&orTR5{FpEgYXBgqFG!)PflSj`0(ZgxP;vYkx*Pt%$aV1pVw|Eg2x-A
zLFbZ+OfB<<>t~sJm!5lW0QU~}t4u+65~(EPlKc84Yo2>;fA1I)E~g+JanvT|nUau3
zAj&?Oc1T4f2QDMpuGzH*5@q}t%EVA`rnn`*x>O(A1`Kiuwe5G!s9Id7V;R57@QO9@
z-q)l}UC4r*WV*jImBHQ<)f3;Aq&rjc?SZ7{f*%#7gcMGW3xHUmW&O+wZ&ryzOeH#)
zB&AX9k+>FILTr{WG}a_$y^B>58C5Nild5uKRWxRNVmdw}FUMwaBlBpLRAfd8Q&Ui>
z66asl%H{QDi;!(exS*+h@F`Y@C&wI|DDrnQokdejo_Nl?(ITG{d-sh9v{5Lhn_5p*
zwO3{3voEq9F)+K*iDi5W-eg9)EU>$M37gjjEKwefBH=irtSF4K_HaA?%|xrrw4&|1
zfVz;k&ui4XuYc(lND`VbM0CY`le}6!@e29G`(ZmA#TD^+?vLiV^Wb@%c-QC0Uw}YC
zf`tebCS1gaY<V<MqQ!_6CtiX?Ns^^Vl_vcI4za9{Hk8YiC;!^xppX{Nq4ecbDOaIV
zmFf^<t@#!>uG2dUHL17UX6r(NIuFd#7E+>vMqL)$&yFV{hZVN?%P;@3#&+lZX@?qr
zSYm*S20G`0t1h|hiucrRy5_o_2KlDN9k<*z*hin$8fu7PMi_3SRYn<Ow6Udm8OEDv
zk`E@EYKm!QnC_6(W|?WWIX?N~h+TF&>Yn={PyRXiV}%$HLt+%hVG^dnhFO@0MOfNn
zuLBO!*k`}97TFqBPB<Czu+|VZVH<YA#l(O~#U|#a#yNcTJB~m|i;{)*TNP{8q3xx0
z9ODQ91m{K_?<_&UA>bH51e^lS0c5}>;3~)RqvH<tFt3ALj4)P|9vu~)5%*=SBfiiX
zFNqVSr(mZu-q*xqDDV+*3?NqKv<&<|0M>&%EfdI|0RkIJX7KM-%ss8bAx~@2cC(8)
zkZCD6pAFb00mJ27aO|`s3^yxb2!ekD;8l_!wGLbg((9x+AhS-I4YKQG+(2%ftSiW`
zld}hffP4}-t)CbvH(doN0s(-8&q(02E+M6XS>=jHrt-tDQ2bv}#c!ZObdcJ>tf~=E
T^;oYa-GVc%0S9p2zT08|&6ZP<

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-medium-italic.woff b/0.16/_static/fonts/FreightSans/freight-sans-medium-italic.woff
new file mode 100644
index 0000000000000000000000000000000000000000..a42115d63b39e9dc66c33d6294a81e0115569003
GIT binary patch
literal 19420
zcmZsBV{|6I6K-v{Tidp|wQak7YumPMZf%=eyWQHh-M($_?|;wzcqiw{Imyg3Nj@br
z6AwiR2}LzEMGz3#Ef6@6|D^PP-TzlaB_xPJK=8OhK*%FOK&U&fEJ7tDl@wJ$KtxbL
zKwu|8KnN%m*^O=#Rhd{pKtR?0bE5GdjIhDQ3{CBf9YH{_3qe4@0zp8)S;O+6wgBeF
zX8-A*{sYT@;5=l9vIP7$`5%}1zcKm$07C_x2(WYY1OX8m_>ZgfAGe|p182(4*z-Ss
zpmP7&K>c4_D1q1;+nN72R|5iq`3M4{Ui)_0Wb5eQ;`-m7iT|_t^&g0Td_gcgK~0U!
zj6n8}zjzVvft-donOTO+AcLUi5ns$70lWP_^hML!2yzJ1zY!ontd0;3|09Bc{<kvm
z4>yP`gwap@j~0YF_!-FZ&+PwH|4%MJK>pj_f4wIt!$>3IpfEP{{Ojip5M5W#1UwB^
z3NT&N9HbL#2F_N&P*_~J$Dw$fRGww9%6s_0=ON29$?kUY5jB!u+dqTLveO$kH(5uJ
zEKJ<aYf{1VaE#dWKfiLbEHdE2_!9T8bGA)5P2C-pW~|N1C=hgEgOigIB!z`M!NCT^
zYa3d++Eu-gVQFpoqGJ4Ed4-1i)&9re8dq?K^*G}a6LL2!CsOWLj-*%+H^+jPJ(J++
z@2Rp#@97aW8+})m*LsAgwj`QzS2}f~ZP1;@rVk`cUp_VgyZ9QOdz9YmfP=iPlIVn{
zL(;x+NFKWhKcvM<ej_%4DT?-US;D4G{mK0K-U%}2S4K@+McvL>%O!zMV}Brdd(BK9
z|F`i>boORP(i0H$8@fkNp{iuG6hk^*<+HHaG^2ebiZ1$<HenAzigXkN9fZvLWz{eM
zfSunsJ(=xcp`8mlzisFdunj_P<hlQ`aJelQ-F%`4+z$I99+5rQH6QUf*fz-dK{!Zj
zW}zR!+oT^kwtn_wxQRoRriZByen_aQHS`Kgg0>3rtkw~5J&;e8;HpVW^B<$tCZp*=
zR~*s0V9c6qKB!y?pe1T)Hqo!ws3zJKh+KwCxFmXI<eU{WDem*m|E2ePQ0>cUL<E{d
z{HaMj?};dM^V?1DqF;|iw;`di-{<AZk^3Lt`LTjsW1YPoSJpe3bc-~IDwg4AyQkmJ
z<=16_!Pmp119D7#eWr-l3vZU}+yT{Q^lK{rJ@Pg+UTVj#*B@T<LzDZxwtM1Uv^<f(
z0|VILdE}(@*PZL0E5Lw6Wvz8sTZ;(v!qf?PC6;AGul-AL3Yk2I1t1OZtf1;SL!D};
zSG_?Y&*g)LWssDCSCUT3zPKmdL4qqs!dRT942PqYSQ}=*p@W<KniqR2i@L3VD#e7v
zdjq+Ss)Yv-;qa;c!N0?AQf=A5@wT+`E==>b97bd0s2Wp}ywJl7Z82+5PtF%{G-Izy
zkMrfA)Iz$fwSf{-k#PJ+mvR)SB(BMXsffcaJA*TqVw$)X8=PwD>pG0ac}GmXDBcIN
z4$_i1#OqTes4@`ZuQv6BE6r@1b_#;GZ3V)@b7<*ETK<Dy1%>-pYa+l=A`7)BSe|KX
zFxmoULD??WL^QnyPI%t6-q$fbc@eqKLpV&Qr@i(;nrNh=L|yk}0)2=fMbzhge5I32
zelK-enK^=xO;_G#u5NB5#^V+o@@a;%vH{N6y#~$+7PNlkPfdvnVf3H8OJi(;PDwRF
z7gQ?a^;eAMzcaNeBqL<;y&E#D!pvFe$!7BZu#6Fu-9*+Uuk*q}THo^6RlT%4n5(ua
zE?G{6H}w~?YkJe-+LA7Wjz6|z6y<)=kX2EYiH-bgsrCPNO^nD+j3QSUuD3~D)D>s#
zGGp6Hf^H0uvO#&REqVVO?MB@9e0}(J!WJr7i!=ZM-b+RPq|zj_&qlWXzV4KbMtwpR
zK}h0i#0#yJZW|D1P--`X{M`6T3vr;UgsODV_mn>LtLf`A^(*gRCGrdgIG>WERWdql
zaQ%iexpFFiY~hl!_@cc_z0H_qu~MW|@3KMC1*|&BQ{`WUJ8>v`7hbHG9wm7QJ-+KE
zPHOCHVI1m;_|Rw2if0*sO1E`-dCAJLFBW`SYR$WLD1$d8^zn|&QPXLYA}rzaH;M)*
zh8Mzo0&nYPFJFb^aYmGvoGXt7yT%|&#(TJyH_H>nzLnBgqAy*ZBGNyi-fyhqEu_D!
z$dS{Z#2|p**Q>>)(KE@ia)VczYcBQQxr4-f95Xm3Wy)qf<w_*9=UKm+v47NZky$u&
zJDmW*E)S6_l^|);TA?Jyv;^f;+D}Iq$dqr%$l#<Eqt$cu>4A%~Jk$`jK0UCg0<Od|
z)tFJZSEsKo<b2w$J*>KGXxZV3Re|qY`QTema5?(?Lsim%F`Nm#>h?nwDfFZV4QQO;
zi$%)$WFk|djG+)@VOVpI4cgLq(?z#l&I?(aetFzia=Q&A8H6Xh<iI2n|5}*zmeE3m
zU>D3<P}lZxRR)Xf!HKx+=@@ydzu7)OjiF3*5NnCDqXhvd^@-RJ>|gcsxdDp40Joj#
z&WShf=;rF>PQWpC=vMO3wBTgIC2On>aL!;o3zePmn#Y4zZs0L|XhDV7g$ybGufE2Z
zF!DmTve9}{a0Y;vdo*#Se_SBQDIW%07YoR``p3?MSjYHb2eIe(i=IblZkv7Fo={;r
zSa9UA%x%pYkBR)oKR88nfz-1<2?K(lPzmFdjd*NRTl+jpsRD5*_ozwUR@xKa4sI>9
zjz38?JvgyX1-$4;R(<O!fm?u*!%QO`%gQMql`%*rfkklm0<hDiB1#6v2#PXrn?j{K
zxRPcQ5mj!0uunDalflUjf7`e!Q|sy5s|06kh+2y^e5YhQhPdgO*XNR=mGV+iqA%)2
zQYcS5&R}>DP8yi0UL+tx7ctE9@bRH#ig;*kx0oFu<VYMSi(zF2@S^gdO&v|$6(@@r
zIHuxy3O!ed@I?uJmDE97n-xPCrergt31&6hqRwJxsS7E~P=x-MRfWS4l|=4z$17ou
z(-TKQLA>ju1&DxUO-Nr=;#Rsp0(<KDmeV5uxj|ko&A<Ic1hHsOJA8E)<d(`%+;a8W
zL0(WhH`gLKH+u>dAJ+dTM?w_UFA5@1nn_R}IodEzSS>78M9;u*B_pF9tG)VJU0za9
z2sc?f`W5+gXk4_ec_2<=R#8Y%G5Hu6yZH}`RW^z95G6=@iLP(z-AWYHxdp_S6wL{5
z1TC~{nTEFYji;T7+2aUbkzx>k$J3bErQkrP?)Tm>aELoZ3M%(E!|lH%=~7ucm+Par
z(mT-KPClkp#X&R<qZ99D;YUfjG08Le`jsOA3BhLcA9Gh|@JO8@913mx<zR+VkJ320
zU=pxQVbfJ_%d+b!aaQS*STKiAjMPs+!X@N0M<cnZ%EXh$KSc&TjfL<o@;d7>3%?IP
z4&JTZoF0As?uVDJpH=3uHTB-u16qqY+J*GCfR29pF`5sD+h&e#&0QppWByJ`?!B*H
zjCrAi*}c}#C1)t%9fWDab7%$xiDVJO+{|(73~8L%&MWk<&Cbh62MC(2F8>lSU_z9G
zqgAlSMglNh)lQ9+vra_&xmBN$(pDlp82?6$UJB-~t`i8JDI6lO`QmI<BuBPkzM4yG
z<W?Qyz`jaIAVA`qTB!WiRcKdULdIDHadIt}R@cr}(yRXl_np|Trx~lPeIc=^UL%=2
zMiz`P?+YovZq=JvI4RLsPcxCg+HZ<OHyH|+*;|{t2Zctm`3;EWGg?o?nK#;n&4Lo4
z%H{~!f7E^0)$YT44X1M~+dM;A|4pNSfxERfsgQGT-`MfC5)gxjF@NmVw{Cf9+V~L$
z+m=AsBsjQ25v#n>x4w1^J6gBO?EZP3ub*dh#p%^1ND{wWU}WbvaiZXO^MT=VwS5d5
z)y)$|k=t2~-s59XaIm{<cXHhR*=|rUG_CG@camDw%{wuShXOU?7#vYvdY7*H27G})
zR-D>@)C;OcPxkD@q|)*|1|D&hULI3T9j|(rgBMaM0?0vb)TYNLS$yeo&iFYTq;~fu
zO6(4o_Mvk?F6s<&-5yGA!tEb?O3mo)6n(46cIU($!hhFbSO)my)KUi+EZzE<t>U>w
z2w8#@V)@`i%7>v0QcP=0&HNk<CM^ab4=vfrhF@89=;qc|2Iqr(FcrS%bo7u&nFL!k
z7TOFoiYn128OExUeLwwMoZBHw?B`u)92qVwLCdIqHsO4Kp@<Xc*Ir1f!E}|WW^&MQ
z_3-rX)mn93tUfQ*JS=T54B0aKe!&b-fHVA&MbFdJ5#MO9+ZmfxW1`i3Ue<b-(d@Q1
z9FbA8R9DSgdfl~RoL^It_V&@1)>+o5YI=Gh_zIA8>0*6~So`fzPb<lml%MhR5}u#)
z<f>hJ_UNa+^yp`PMn<_J;;s5rd-j>+-`4V=jyq)|rAfXSzLSGfMpJxi0VvbLo*-Dy
zvhgFf%DL*7*Wft6vS_=ZJUeYVh-}H6_+fj;I9vadSv%3J{sGkLB|`+8Gk29Q@zKGN
zM!O`65!2UKpX>kasEmRy8USY^EYt}09#77s>e`KQip+jeb^Qa+bn(+2e~7!s<mWxP
z%l3hPM-tQBBC^<C)h2G3SG-M-^loZW@?~2jr4QUJ5P!~I`>}CT+!aa^>Q@ue>{nL6
z+$}xSQp-<TIoLMSOJvY(efr9$i_QN|Vnt+P4$j`9C?lSuonSb74AfhC1odu2Y7P@l
zax{u9nOF5SS=m`q%wU&kpmV`NSbC}rgjCdqUW3Asz7#ZjIunXuU!7M>TH|WMh{JAL
zE>pKVXq&y_&;kcenB>qiXybaX@xEr~P9aWIv{0%cx~8(+;S*+&roAWEOq!u`pjlv+
z_dz~LFLZ?O(BHzso6V)nWEPjAgyDCB`v#4z6DdE1(-$RErLTNxBbxJ1vyF6}n=%Wv
z=SgZ&6<e~XG(l)QnX8YT1(Li%<OqcJOq+tfG?yotD)GN#=9^uQymDF*z)?y2oR{1!
z`nGABR3x8rynQ-67app+aHgHc{=&m!j!VjTHZ_@UdP|bLGpLGr3{EN&%jT<<PiF#L
znObl+J&s?|HzzSHZedDUAV8^h%Vb1mGFby_%uEcvP9c{E=%%{LpNY`QF*I?TtKOoh
zByAX+5%uS@UlD6|jOaC0fB$WAE%1=?98X)MvJ;34ZE2IZ6_IOn4o>j;XRg1gE%bhg
z)ShbwP)sMX?8{i1SUBUa<?#%3^GjMk!6*G#rCDT(R6iS@p66;QF8pl<m$Qwaegc1>
zVUb`N<>VUaAWYBVjz33_t3ONVaqQHGu1GJNeVb(&7S&J`=sJ`j#hZA{0Nmrz%5;;>
zy3STR#A4?I1Aj>4yCRG8dDCEpF#yhHKHeCkgUc`W<h&7{J}27V#2cCK?oh^QVR`*9
zxy1o`JQr0*N+pz5J%?kn-Q<03JVkcF*9@4=B&K<lQkvU>(lKE9TbupDU96k1f?7Cn
z=yq-wM4N#EGoC#ZdUppD+Bp*$)FAsmFZO)`Zfe5V{e}a#dw3~mppZw3(!R>N&+0ZN
zo7jdb5KTHo>tUR$inoG%kFxnQPe;zCTZwQ@mUCHN7AILkr_QY(5DEa|Q6xD}6%&pg
zMIg#YkIuFiJC4e1!OgShK0&VF<6aEzW`cQrV(lQ&fxZ<M3+4N8@@rvnyY;q;7sc~;
z@5I+b=IljL+uyw#twfGtv}9!eWrLd(Gnp#XW2Aj|#k!2+=Jei{rNl>>TQfb6Xb{@+
zJ!?|JsFsKxUgqgmE>$tZJtjoTLp^$GjE#!v`?Lg!R!6#2wB3H^f*mdPe5u`SubZfv
zD0Xv!@f!4YWb~V1N}g*3zcIq$8jS_K?sGo_WP~il-*9-)i_r>Ix)XG>-PvY0&v@K(
zCQ%>jjSJs^qjjLZ^SOap6YhQuhV|%i>q!bFvbNCV>vxt}x&`utxd8R-n}37MWl#_D
z=4yqAZVd!Hrq4t~ZQZvm0-TMwGu`Z6bG)2|9~{oTj7?MzqCe2Iwa2kxQp3p|D{T+z
zBq~~4h2)PD{f-g$GVZBpu56y|_n?mdoLOM3M^>wcpEq7)I+Yv-RF}YIs8Ir=UojS@
zSgP+q^PC5L)51b#^O+;TnwSINy_#wHF>wW<^X%%TeXpu*wQf_6^I3np<B<0gPF+&0
zGaq5W4_2NRQ~=he;$o7wJ%Wf;<<F}8%7K{iT+TwdxO#(fJ6BwEw*;PgNY_fke(!7p
zdMb_C+?Yl1_8L`n28oSxp4c)-_b!^ENmASNU1>+<eP8}^bjKsFfm>ZZsycC1)sVLH
zr(n3=IsTYcmKR0LSU>@g*dfc~#9-n{HLi}EAJ;ODkLOU!vejLF>}D(Do^0j>0JFbK
zIb*+>t>koWo4{9VM|?{V{h5A}<FeC%h3WONt;hTNR1<Ym%F}I^wa--HnmHO;^TToz
zhM*4qaPcJ&u$T^C7YtqrH}k@>neE}6ehc<^)@4_Gm)&F5OzfUO)2z@tVY{qbuJQJ7
z?U42%-WBW>%4CiDiM-zJ2(x~FyM*<Hw`aUO^4}jf#pa(@=7!D@4jv22ck`7lzsju4
z(&y+keqn8W_tjaOd9%foR$rn;ukf*<#a{%`W6yHdxhxlE6YGO!19Y~?6WZjMoN3Sw
z#-A02l|FR09Kdu>w7NX=vbVguUUvRO5o|pdgR^k+MJ}}|@7Gm<XoeRT6>N9Za3Y?p
z4hHNV@Wyqy&?*r^Cq*}!IT4Zkci&F!slDMR2bQ0*k3lf-o>akFj>gYjo?CqDk)ydR
zk#S4@Pw8eLfkRAn#Jd>Grh_)c;Pq_er2B2GXe6!Q>f`{Ex2hRyk6E>+3T1^|<%TT|
zekEH=*=kiyHi4J1K?%f(PYlvU7GCDFnpd>07c`GxZ6Lpp=ID$TSf>yf=bh8F2|DPk
zS@v$eZv)X|?HQxsnb!`skT!wb&-G}1o{3le!1{gMwv}F4LSG~dvx2vQJ#tlVC|X|?
z*PkGJhlH_!&+6g(OV?Je?<IlD1&mFm=4zjU({VzFvAhCvj*p4QEVtXO#5PNzgly@R
zcQy}IyjfU@61H@T!OA67$LdF_H)(-C?!;y1C%w4lSUoy@fId>7f!DKnZubkfeF5z0
z{C#%#=;zUsGdzS4y8||twk$3T`h)YD<ztRmMw7w>-%gU6KZx?pPOf_8K<FRuuoYlw
z*o0)g&XMGzjm0l3V2Jyyi+yFl23O;q($;lk#8JU_zQQx!BCBed?7Yq$B2&uutL|26
zD__}+@;knR?GC(5(~BR1y8as+{$qrsc!OrIRU$+cnGx10$5(3m!h@(8KX%rITmdXj
ze?hF!`A<Pzyu=HyhHvw%^On+%CKJ|YZv3rvL3#Dzy-mR4he-|e93EXi!Rkf9(atwI
zbnB`QpNWSlThXcr%jw8NN@nPO{bs?~+_KMoRK!HM+~}Z{J})|NcHb<=<(GEv@~jnR
zyEWeti8hu)WHvLK`9Ht>_X5Cn%O;YZ+U|>V|BK2m3ZcUj8EOjwnwk;9=t6>8$ICad
z_VYP_DOZTe6B_EYX8yAN37W3`dpM2P4tj&N5$=*8yzS+c2Gy9Njt$*S6WF;(RadpK
z6TifO=3UIHe~^I}>29PR|7sdx@ea`Vv|{L*aI0(8+eU#yG2=p`a>MocXd@1KG$`g6
z0@xQ<a5!j&i15qM!#(0{4<98dMFn{lN{qo4)30@Iclvv1@+R(BJRMxuIRqrLNO&gK
zjJ>8vJBOEmW$gX0j?d5Qs^=&yj#O*S?mE`?s`@ITPRI8@5@`5yw@3?AJ9vP5N20|T
zSvbJGD{>|tbMfk*t$RbFkbirk{o+`w?x^!vjMZrypC0vr0_GuFLT5>stw-C2;k5sv
z)oh0B+O1KgEpgm*FjdTPR|%vYb*%zF8@<_Z%1Bd)<o>#dvS$MTKYzxh@W3%^LwuaE
z^iYvT^$VereTRYfv;20007PH;-y<=0L12NYM*05zx^nM=<?+n=>AIm4)#Bz+ghHTi
zxz}jQeWe9ap29!9Du2QWj^>*QOE=@!yfeK&e$Icq?{77kPP+h>&pY>CMC&|!X)fQq
z;1K_v{l-HLM}(kwpI_*_>PwUGMAcP`6V={bmn2G8H4=UV+<9Iiv=1r=6-WDkT=}CZ
zD5VCk9adK%m_7!`+V>}4JGp08TNFr-aW`)0lHJKC;5F{A$7-Fyl<gq%fYo^;*W8qw
zTnQbooqdD%^x5ITjm~%8&a<>Rm$f6vCNPrQ({%rR{x)Jrg{b=_uR5)1mX)Ya_tLuh
zcJ*cYO?YJ}#kS4yU(Vaa$jB(%NP+@N0tzJ&^Z{Ggm)OGeH;4rDC`hc_G<}Wj6;7_+
z7JrW~5G+_^h_u+`04-T%k+sSB0WMl*lD5kF0ykS{gSX5311el%l(Nj?1T$S@g|p55
z1&Kppma)$MhA`h?N3hTT8#G9Gkfg|XA2msNfu%99?Tnm!k~{YB?UjJ?0ehK5<YE`U
z#ncT3E|{{kkk|Yae+q%&b%e%g>s8NrVX~>QQV(Bp%#S5w($7P0G#AH^ftrlhHIn$u
zE|N#;DU4|D+Qq95a^9$o-QyW4E;b=vUO`TNZf1IR6B9Ed{hF@Y=8`djZ4uj3fSa>y
zOYM`HCa(vqinszT5J(y=i|+PoD^%%gTQN@lE6g0VlLS_$-ukGWtIc5-7Cs?5KBixD
zSH;rK&gS;N8~S~Vi<uiZkeHjCot~7HmKUPa3VnA-->q~Q!UTH8Z1gu2hi;pC(dj(*
zhzU~0dxAKvVpL6`(GVXaB2?ln()6I^4*i4-cV?$xlBu;A1k5nLm<)xCqh~;buh(9j
zPCv=jo?@alw(82HfSb4{?5#R?qo|O2l2YfjXS=%);Xr!9>xSaZ&Q!9g-&a|XgeZN^
z_HVb0yTi|E4i(Rwc^w#%-uT`+rs4ajhyR^}VW2L=eXP-9B}IV}Ept+E5^Zt?c{qTT
zMr2Nv@G(~mx3VOk5|=q8B1|LOYEXvCu4QSEtm~{Khw$V4_!lGs+W@g>&pzyNEZY!u
zN-T>oi|RDX3_|l-re%=0K~rd+xrUZ$nzf8mV>~z&-alfO0yH?9LylXoF-j<BDT+9R
zCTYqT80UopaeNYT6h(67zfI!U`@VaJM3v^(JpZc|Ss1<jKt}4~h+#K?favCeC;V6N
z@dTxx)s@091R1+`nbotH)ythSus|Aj=Ul{aMXy9KY@%OaG^W8w8|jMYZM|mfwqFa;
zS3*Y{f)<4(jiE=4hKr45O#~xCD2_2oM1~FfB1--dwU@yxn=|wW3ZgW?{DzShUv$dJ
z)`j~odj98MJ}6&MX$>p6JuZV_?wVM{j4qb-EH`SxL=A|do|X^OKgg(KnElc@!PIFD
zvp@89HyVH-4ehD$x|Y(Fq)v~Qm_i=8P>pi@3rmHjZ0W&^!<0ZZOQCrKFd#3ui7L-t
zGVY?wwRJRM&(PrNUWlX7K}~2ag#)P5tNf0MiRO$riwC5B=llUIO7l!KD~JMS<TbZk
ze<^4l+dzWA7%`ox4iJNf%NB(iVZry2fq;U0qJV@I;$aS>y`jVp5dpxfMC~dFo(l!p
zV=Wd)o1@N6Br=7rgkLcDCTb6zJ5ky*Y>s$4+kMGyA54KQg>YRB;sA7e_>dvK{q~9|
zbCXUjlCA@NCk<dU>WL$Y=HcTT+Reb1Cydb8;TvN9(2}Qe@*jK}McPyYX`HOI_9=O1
zIe+?uam8cU_hwR>DwS${jqFr|<3Fr*JsQo+zR$(p+Pz8Q&EZWF9KzMBC#^qjrJqlJ
zUDY1v9#7C8@LwU{s)n4-v4;|W0UUL>=%cxfMmOwqGvF4rUjJNY=uQ|O^SyTQ*(~{+
z#_k{lIP$6T9d;yc*=_^+&<rbYTO4+Xo*bSFo?hR|YXZe<d8@ZCS_rk-+H5U1K|eG7
zhQ6gh)+HvTC@Lu>rG#*ZQ;r1Q#rhI?9@=HNKw}QphQdiyIJ8W8JEP=FH(k5Oi;Ih?
zD*|H*gDa@Viwg^h|JRzj>CnU=AkBm!AxJHL|5j7OP#d6N1nB)8sgUkriPE-r&h{Do
z{1TwteFW~ZEq!UME-w33DJ*-<owuI^Wc?^IqqYo{2NNS;hQqf7hcK&WGI+xE4XGIa
zX^@D-QVFdUDQnQ=CT53Y5XqUR*uv;TC4{dJB?{XOGgy%N7fBRV5Y+^rh({&(TX|r+
zp!$T46$5&p!lH$MI{Y{1;94I4#km8v?Ad?T=j?0l3L@@6Jp8=>*n<U0bU+*@dF;*x
zNSfee=|CgF52&EiNiKrFJ!N)OOAyXj2z3vkcz=FC_@Mfm-08|Ziso9Zt$uwid33O`
z_|ZhU)XDB7uA(JVdS2Yx7{e-#FJinRLWsY+i%3Azj*cYyD}d<&aKt)jIv5?5!)bA?
zkkKTgvng{7EP?DAf>J#zfp|zN8Nd1WuLc7(Dh%5HXsz>Uf!2#xBH|>1NK3^u|H_fQ
zk>&V^^6}Rj36E?-2<!y6R$H}P=y$&_xS!#ipt&ZK<)SXizlJ}8wf-Sq6DbwN>R}|G
z)6Ea@YBwn^(Jz>{3~U6kvpswtiJi-rv^k-8c81#%n;BVIOiqbM<J@Q9yq;q|oMY--
zu(k5#^=({XOj)Ph&i1)aWKa(=V5f{(WZv|4Hy3?t55*Gpar~#dN5(pb!@I&a<_pY@
zV2WDOHPp`BoNYWcBrTIb<B>0TrmzS$dxWtBiv$-fXEW_7>j_TxD_~LZwcn~x{!X`R
zujs*DDJ{4_e|FKYo{_Jf8-C#uBn)GOP*@bC@KEIlj3SF>NI-%rv+TgOQG&t%X2ncx
zvLjFZBUxcD0{Q@sJA7gox)EFx5{=#rrvV69{GcP~@O&aOC%f~9!h2wZ|E~G%MkHCm
zF=U|li#_(4IuN6l)x-P;F8f>_lsopSbk<044xPJunt<DcW8b6*6AVE#cR}8flzUFh
zGFOgG50?8dbyLkR>uY--A!9|D8957?c_|SxWbU2W_I?z^0O!Zhr77`Z>}M|5(U6X}
z;$(2^1N8wA<EJEo$Nn<{`w_yB83@6dJdC+xGRn-}r0aQpeCrYLc`06=k+*+8sF~zt
z*(;a7$R&eYr_<)dyYK6sL-SJ%i41#FLM#}FD)2kBoQ(<T`AH@kHZ@vbYPm9n!g8d}
zR5@tL@dF$Gg4#t;3FZZOIE-BH%LHV3Y6*$(a8B)4JCd2qQ6#X%QZ$S!5W=uq;kHX#
zvPirl)t?o<2gY<n#n;J}d(H7V(<>RU<%Q;XwtYv#5Q+4c-`(r~xHM0h)%`a9eBZU&
z7d4S4q8T1%uWDaVoO~p8%nvWF|JeH_Lv|bU3nD|+9>EtP$rpN>JW(8;2)##`;eI6u
zBU#_y%p>jXL7^h=Q0~X-5uEn>yzzY@yk2~7;B2=L(n$#qcS@Jb(*9TH!xGBy;lQoL
z5XjrnYr!DGhr&hM%Tr8@9_969AEXBJ8!tiy^X_-3kpj#|5g#QSQ`-ZF57KF>lq4Rb
znjS{IK9Z~Jle-6Y`<r{hGy`ur!qGnJ^5(^DvZh*A4L@1?GZ}l8w_%Wj7uNPcxcfe^
zGR327He$vQd+#=jyc7DNos==u6E=U4PQQ!^(6hL+_);Mk+z0(uw-|vHfMU??c4)aa
z=k-w0i8xrRoNOAslnXmFf6K4N^VI)(6;WNp@GVJEr#BJGPY4z77Xp5lVm!Qn@G0o_
z)%oZveobTLF-qSuH0a{^JNJD@8p0WO5P6G48M|Oi@>Z{(OgkFxIqnft0JH7sNe}q3
zUhTrRe6L`czji#lh~sT)$_=k-k#U#6z6J-jULWK_oGgjk3&}_h&#q?#&i*(FlD#DV
zBuah5p&UL>VC@ZZJY+57If90njMUl&r(ms%F=6qnK#hVlMlB}2AJEukl#-j>NvAK9
zC(;xh-5fpQQm?v|4q5hDHl81qf6q7+U3FLDv>4?_{;uTf);h0M5KyUgL4PCoA~6u=
z?Lh?9?sHif5@D1J^|e4`>IBzij8PYe3P7ft<2?D*DwT6FF`D>a!i#ivY;FK|YY<Oj
zko-;=nXi-^PuwGeW=9&43;vj`!9|0gCK7Qj0YjP38{411Z0Xo*DXK{#4_|AgrA;4r
zJ!9jT-uR6`6dXh@$jQhcqSpdEs+F`;BwSrCBgaqeWQpagLkPD4PEEqw2@EZTc{KEg
zeor8nP|NlIoY|dLu`&L7fNa>??h%oc0mxkr7W^ji{_(8n?}6(-I%R}!PE{%$LwqlL
zve*6VoAs&T^(Xa#52tIOJY=BqOWH}jI@dFylm%%zRS3k(ad57GED9}NiSMyU2@E$A
z4F32&$169S2uTtTR{qRG=Iz=q7H_Ttqv4u@-~TN>@wDYvH0)K`NCJ$-5)=&Syh4mO
z40HjtTN==uTa-Dbj2ksmO!E+;iN*K0jT!a*UVwPu@6SNL(ZL9v)dJTMeYi3p%9Q2d
z7TT@Py}F@F^~5N&Y3+(94B1Co@NrlsnCT3%g8XITsx!NS<Sy75G!_pmVK6J;55!3m
ztheGpdf%vJ_*s8(52spumdH>}SMrAV*`^m!EYj4LawdGjo18b!Y=3`JHn(IGwb%jA
z1KkMu8?Jx&B}5!`Mh%;)V97|8ef!c6b=-SwNU#AkMM3-WX$sa;yMaieGJowH-k&2>
zbp=4}<bO}wsw=OeU10+id!EA@Y!5LjwUftPp8nB&w0K2}IRJX7ty!(oZ^HR@tn}$*
z9+-<Sh&WGIP?_I-cMfm(wh)dZ3Bt{pm7PSUhK?I4U*a0+5a--UsdVD@2;tvg;NUo`
z2w&3*gKhoJy9ZcDRMCD$LS8Os6Xwjn3>0q%i_mz<*`&fYfgSn`iB(A3Q!5SYFSEv?
zY-aYjhqKgQX_SN}fCbZHrV^SI*uRT|ACwQV&SW^NkYmlBVb9r#3mEn1HluVLX8d8u
zJ6R|SI7sgnN);G-^H>$jMXebX=9#5)FPcpgj?so^&6<^!jZew6csCaWY+BHecd(d=
z1vI>a4D*A(SThf3r$rddD&>v9-1lL7<U8JDV~p{2_($CM)%3>-clcqt;?+}!G2ePt
zV+^JGHD3Fs)M}fF#vWV>FZ%g4W`YVmyl|2HRb*_}4j`AjD9-Wx7C&d$6tV{EO+v6W
zOKro4ff|%~VIrTdN4x7n0uA{S=$wz;S*d@`W>wD(2cX$neEA4I%vrVVP`7O6piP^Y
zHVM;a26|S*3L*XZVzk+izATvSIi7+leU(hK#41qKQ@yFxvvX07E*a!}u3(3a&JtD)
zc6H~C{FG#7C{?3HUN1wN7L$whtW?cMW_eL#Z-ssut#zNvOy0+8y9rCo;b>y?RZ)nG
z8=HG)ibadMuyF)y5A?|t^?KHwo30hEAH?fqGkm7rXrLa`Hv6`Ax+b11KJK6QMG7_l
zVMnZw>EI<0IjA!<_1+C!!CnV)ChIqS{w|t7+&nnD?51SbvX1!px`G1AS;yzm3Q@EX
z%Ic-&XQd}Gx~z3%8JB(s0C`7^o3s7DZ}0=0_nXmYGUx`TeJ_pAG(m}Ep>HsRX#B1d
z1&OWXCXd)Pm$<Jo0>2(Y#V#uny&B85NU;&C#Aw#`EJEDnLlMS9mmy$B00_k0XD;R&
z{$=i(=50|<g{uBCX8x!60=>f;^eH;%85%QK;OR9nS|nJm8|4X0(3Z_Pcknbq&M#6f
z{Evf^w(gMj7Fgz3Nf#UF5sIU1_Na&9)x+>Hsg=OKrHgsE+bscYRQxgaZ`iP)-ahnV
zzjcldp_n2Z>>T}|&C>TXC;e_M{Z58X3C3FzalB&ZlE1(E&jzI42fkjv&}PG&&R`^G
z7H<-3@A2BkqLr23-5~$_aO35SXNHO`+}uyXotXX;!M)^p>2#0d|LXp>hm@Su#l|6?
zx!8K)K6BsuSw(KoerO@$Nr^0(S_t!^LL|nG{ppSskTAsm_9_kJQdW(UVa5?(fHGu?
z2uL}{u-w0Fy$h+*-Eyqp^pRe`<{Z@1vaB!<S6~~IX7Gu*$BsO&vK~~rBi<q(6=y>#
z9~6)~V6vX4_7ESbjAy$>4*d+WymtQcm7T%EaC##I)A5gjor?gfg#SDlD#FeMj)<+(
zJSd`pn0~}RClxiBKqTLxTJa$pi#wo+`9v84VUyBHYqz9o?`PTpxT<G*c1Y>@xE_ZS
z_gB!v+-a<pjUXY6DED%nRP$jmaABqFdGHauw+u#Ak-@)JbfX5R-EKX(QdPGH6GoWs
zU>=ZjR0|rj712dR2<@tm%1FTEmrv&u4TB}gyaf@GdSgc~`V-z?^#*-ZC~+&#?GZ*(
z)%uNhUrr+{4P3Ev$}&lpiQ$am-T)KF50tQLCKQ{ZOC+mpArvI;M#3NGC+6fhPGcW`
z<(NP%(iI)4)op5On~dm^*_4^h)DWum&ZYThTPw$xT5(B;qKT>B;$2V~1D!DHnE*i)
zZFPg0-}(hQ+R0;Rb|R%mi4Z;pgz4Ip)PMD;S!;CGHS4nZmwkU;w%(o6l>Db=MjNkr
z(GVlIo`1vHdHqjXp7v`~UAso_Ge7qL1t%+`62g5oK##-{%ZzAEHM4rEa!y%0muY)R
zVm{}j<w_t))zrJZ;`~bgekiMzrM2k>bD{Q*rhp6yPk?CV@W?KSFT(+z_1M2NbCNE@
zCjv#Wsvl?%vu|B5%!14YsA3z$VV{Wzq|=WmNE&NF6J*mns?wJ$7e|iLscgtA<OtiI
zmzKcRI3SbSgTbk_hIdE97gvWznBJoyiZ{>&Tlwp0%f^)RQg$XMJ3;q&*boXL`MBI!
z|0k?nN>f-{J5Niwk)>6S@r%}SH^1jf?{VA5{dQuKlOaY|8=rh5;$wq+vXbpC<k`wU
z6kd7dH`~vQ>vj$9MDyh5JZw=T&Cgop?DOSCX;*hCC16$<*(o<>_wQa)M>K?&eV5H1
z{ou30bR36@rry}bACbfW>$l|2<1R-UU#KX6Z^KRa+gC~3E#4whb=mUsL0!uB&=H3J
zNc5UIjf#lz(9WChlDhb?jUs&9X#ze&GLjvC&*41fQdg=b3n<v>KtsC@&QO53F-z!N
z9a(PeLER0Gznn*Dje9|8^^8WNu8abOp*t+qFmv*sAh8CLRd!F@>lBG3bHH$h{%^iS
z1aTIl|Nc|hIJqur4HdCn)gH+*?Xx#9W0Zx+uPg^-SBAZ(@wW_wE)uaoUZr3TskIPi
z;mV2P4G{$WkqIDNnYa4*CJ*Crpz5&4kj*(C%vY+jfXaEW=dQ0Sh_N9F)vN(XKrbL`
z$d~Ah#6nws^(mMel7J&CKK(v@UB|TMG!U_L_=V3msLaq?N~n;3W>yhIHTVlny_?WK
z{0oTRBiU8iwL44BomKpt30onamJcaZKQ%W!o6+et2_bb{P1>hOorXlQeql0#-F=6M
zcZ7&HbY&8Hc8ET?A{%ASG+Z^3o#6(JJQ;y>ZlD2g&D1I2OWqvTDu5#g?MLLxl&ja8
zUP;PnCaCs=Q$zUS|2X_Jqd#*aTpJ%@%G~wDrw3n>;%kgA14S}|>r2kTwhP~IB`F8_
zDaMo;JB(m14Fx&`aop|FW3TT9INQjKj3l$c#ifPErp&x%&CSIw2cC~(wD;O%&b?5V
z&Ax`*@i=_N+~yP)Gd0DOf+I^=Sj$*cb$Gho$Eh-eVMQO$F2+ami7ah&jl)^JW2z)8
z5u%fYhB~~+B=Mu`o7O^T6?Qn7;;T_psN&eF%3NXhuxOFX7~Kw_A_%Iu@~bW7!b40^
zm($-oA6u%tcAz*2EpWXbtTsUt^pv<iC>|)T68*fVO$#SHgbHCA5}-B;TDTqUW9QZN
zbZH!5XdY)ls!BZhHw6c!;US*{pm9Z``8NbQ-6_!Q4-#15%CS=AN{?m@az9&xiNilf
zhXCaEJqoDsOjm59&`OHo@q7jzMoVzLXj|7qbDLY4PgTq!Zq~j@(*#AN>I>LZDRED<
zZuy4seV|iApeK#i1`jddCLFWQ#?BQV?pU}{wB$?XgQm^J6JRjujX%89h%A{w7m#R)
zWBOHdb&}j|4P(51!9WOd^OtycVRw)+2#PUkNg*w-5>=q<(j5Z1Ciu`{dF3AS&sJNW
zoL%h1XOwwzPX3;j9+hVgfuXB=B%^CmNk%R-nlHOa;tkSEo-_@<z}AB!JMxR)SeMxo
z=ZFk*Sw>$gWz1n3o&J3`3qP^2BDSXr*nEY$+(h?3G^{y)aY!pfhB3o*Yaky3>y6nK
zLm1lUhQx~F<NONlkCk}C>69k=kt^2SNU4uPffzQXeH%XmYxeZ5B9hbKvgHnt4>WcV
z4sEnluvOD!wO#3cxD&$>iw{r#@Pgx>#)Upu8%PX(fN?|V{6SL7_EJ}=%8I6_Q0~@^
zWW4KSkV{Q_zr|Ux^2P_k*eaTL=*xegd|A_<8sk2F;z&pviP$~wH#fn^tRM0X>g|ov
z5>C>7@467i4kQpPz{pD&<lLkC0^dC5rEYL~rWRpe?2Rq$-3$ERff8TI>uAuMK0=}O
zl&n_<;}MpU$0O4I-`|(ue+$Q>Rk7*`LXtWk1RM`_S3c7Dq6}P8DA|}e%eSxBYiF5m
z4bm)DcOyM<CoO!&pz0Gl%wva{0=aC$(2~{rxZ3pu^?%>kl#y=*a=n~a&vSU`J-kiJ
zJ;L6bRh4y5Go$H6H4(tQ#YEL6A*y((i_2?^e}g`20xm=>c@jmFj6XFaxkur|@sISo
zjsO+5aYV^{Pb+N6O1k~;p)hR@`)t^xhbV86W6N4{i&O3DMCb<kV2AxxlkMe1sl0cp
zJ1ahSg=-yKLlBksw)!r1Ay%8KSaks;7^9~W^_U1coS!~B>K+US;?ypg?OSYO7ts&a
z&KReS8HQZ?HtF}AnRl|<Fe$0KgbHT(K-sJ{zTAQtf6b~0*36pv7E8S#R`iei_We27
z1$b>67dQQoTyYl@S<Rg}nqCfKGtKh}zXw{Wf)TjXBv>GxdCVaB_LXF~oz{&d)5naE
z$}f`oT!msC`>{&&>4!;KY;UohDu5-6xx%f>_<sFD9vqwFKNinm+tPT)VB$ZUPjbg1
zJdROEORJvQTV(+vG1)5xbBD%eSl^a{<8RNuCI><0hYO(o)It~Neqt*O`wu&%hWY`(
z=_!~qOcK`3FG9w~f97aF>fYWxxTQz;|6$QZ!5@z{T_eyB3d9%jX%!#4hKp{z<xI`$
z$O57R^g&Zxo+4u8=Z-qN-wM|}3*@YnDM24KX8v4+8fpI-iP+=#*-l;MR>_t?{+75x
z{P%Vjz}e*TU)WVO%{{MvLfIgIfV2;0fc0q(06b+!MrMzPhga$3L1f6gF0>XEH722=
zs$WDKsNU_YFhJ0Kw>B{rtY2s*tyelZ4IXr<dr6WV0jSUdp7dmz*OuIH<)+M$AV*}@
z$NL|y8d~#@xEJe&AD8PDiGZZ<HwW98UNiZ4i}P;H*9S9oCk>s8LaD{i0{X6B1y>Wc
zS#-l5+<DLnE*^8>z);tvd*(>?47CfLu$w>$KG@GgLkvL(tR1$Lfqq_&Ysusg(bX}A
z@zhR%SUPnHVL>6=@4F-gP)nj8&#%F+y6U8)5tWs)?HrXb2jsgfqbvB)$Nm6M$AMXQ
z_V{(B)_unV;m7^+<NM>w$&%?j&MR$K=5qFOvji`DH#hsTHI7O*x!DW-EO3Pp#@Yhf
zVha@9LlkyzBWV6LU=?j-?eqw3ItMU|4FGK_EC4pUr`00LL4X26i@RtEhF~75x;y{q
z@oj`h69cz{MT(To2@Y$0GYeKSdg5o!@i2;%+`-7tAeW2k`0b=Pwcb5u-A=xO&9Bp-
zaNlN1t3kKJNlQB_(-#)K_s3JweEV*;K)CT>7+Uf#M1&-PHQ>9dtFw-iL~W>Og+KUv
z1pg%%Nj<;nfWIpGaa3d*xb707r-2_Ym`+>yswt>(6Kk|P^d1L%f`r!^{091cy2#7I
z#lSa*!mqc-=uV>FB1)O^;l%$8!_lLe+w+GkYhGo-$!prGDX|aeR=}jx%b9~-_32;^
z1ZqyHMQTi`@fTw<SoZC41Z%V4BRX8G;ln%FsVR+SA*h0`4sKS`rqi*It+Bgzwa;Y5
zncXSZXXJsNzIv}?H`I9ahefaY^p-WfU#%Ko@%=J#xJ81;94_o@U(P}_nKBB=gIjJG
zlz=>gXTYRYh(J_Te<SoaC;~%(!XD%|p1|Dk+5vO_;<wruJTeL=+R6Ub@sqb{wXG&-
zwg_ujmI&)F92SAJS>sJ(Fnx7cue8K=of-wXXPy@sxzJTzC*@E*=L$x&L{hRY7okku
zsJKJP*R2RVQQ}n>XI2&@8q~wqq1m3bWe@ptH#8gAc{el~=zswa4~`UsrMgZIwPs^e
zUYxv&XE|7mQ*)k{yAHvrp-$vS7||OpWUJ{tkPY2_diHXJN&h#jnKJkj!c}rkm&=Ky
z3$M<igvxWrH{5=1kN|U*^o^wF$asJspZOLaNu+Zbe8h{m_-Ypp4Ng<I;{us=Jf=SQ
zU7a2J9bb9I=X}02@n@;Np7dGs;jLIPE*ZgT^Ibw%qDW#h&N|Ljf<a;zPt_FO6!{S0
zFl1CRwgB(vUAO>QQ1!6mP%!g8=o7lr4^R-~ck%*z@n~L<F1YPy@bbQH7A+kr-8G?*
zB9a1^;zCJ4xtqz)@;(6pL-7>_eovfa#x_Cve#Fir$vfVDL<ROQ)7ce-KqO-6{$EDf
zlFY|1yRzCRv|dHSg#sEtEXNmz2Cj#EsYNZNcC&P|rfSu@%u*pw<@vmCiC2wR(fi=P
zH8mUFf3B&O)#v!Ch8(>34%ud1Gwj1pfmW60#!EU&=w;>d-Ey~;EQdsL0bG8o887C3
zoqb<^d$@n^o!(HTBM}vuukIzYnRZqN_wYOKIB!SzH@3WI^Q*0G6aV?y*lFK|ynbqg
z8}IT$wLwf*Z>c3k8$2dcv0YQA^TrwBGIa2U+C4oPZ7FfaTS>bZ+)rpUeoVQh!Jc5=
z%*;(_uJ83-Z3T=MzG&&mdwS|UT|`D_c6CKt=Y<YhjIAt(O4g`8(we%urCAe>A)B({
zjNj(beL=G=E9JoVKim?Ol*Qn5eP3gLjlt-W$L(H>8m%!AL$O}VyrB%e(I8014h3<8
zMD*pE+#N~ZAaHspnr)z&!(-gduijGu0fe7pzKN7KN1|&glveNW>`s!_XI;nD_S`Y{
zHUg*Qllm?6<rPH_e>LrqNAM(h>oX>{4Cj+4Lyns@nJZ-cOJgU@6})X&DHl9urJQ0d
zmlB{~Q?F@Ky3+*Tr^R<8uH@Em2<Qsm>s^ckH%9a?AoBmlA4olsC!7`B(X7k!c<<ey
zlx@QcPO|!=Jr7xgAk)yQJ9aj7oDPMRAkL_KOrM6;ry8+F%?@Ijxg8Z}5-$jS+)Xs)
zPF{&<Z`S_H+=s5}UkSfxy_Gv@xc`=e7kbsHRA}?yyuI?0g%Ywi{$VP>pGF#iAb|PB
zkkK6{m)Z)ov!@WQj30XwN;jW1V-CgEqSSf>tz1jHp*T&C6X39rkuOQU8@+enT~t(8
z)7i>n?iQi;9f_~s(Bas}J?NE`clkbE$K7e?=W-xptgSTE^v9oK3{&b}fxPTXWYX<k
ze(uk$`E|N|*j>h1{9l4796Z)!cM5aLoGPaU@zRLmWv(S-PZw2t>E5R}N)}RApGZJ!
z)7P^C4iF&0Hz>|t?df5qpzXe%AHri^mMGl3yd~yc-}Zb!aDR1Ny8Qg0*)FYLC|Ow^
z`QwZq<C?`E=FCrdMd~3tThL45epbxiZc%1wo7T#<y^xKZC9NWRfrj_D?qj2XabmR)
z`5$dH*;>1ILFe#z@LlPV7K<Q9JGJ%-W$0w5s<fA?u2b`y8WxL9NsT%bZTB0IMcbw}
zuR=pp@A>mVLVkIG)LU9#;w^`UAA9pTpXsvX5+CB$3Ue<H|AD@5j~;&pr(XB=%CcXr
zS6BoD)4dnQ(#48KMA+zP50}G50tPdKP>7(2;zRO~)E&73M%<lhQ;hQ`yQ32q4S#lZ
z^{<@{eZr#SdX?4J!?Eko7RH_7_o<Ig=O*x|o;T`~b)m*@qLbXl<q6XZ(fHh7_Ie)|
z8mqmnjkPKh=ql&8?rY}-x6qVZ2)>On{sb%7$3NfA;L`9nthphCgjB^m(HGj{D8|Jf
zY!N{LB~^~rGXZjkJ~9DdAv@K~M58U^dx)8vY}f8k8he;y^N92d*1SuU%u*#w&tF{3
z47UlNQX*z%CJkpNF8A-7c>Q1ARmP&WY146^W4A@eKa|RC<BD&f)2%zUi$K+Yah-?k
zK9;*VCaYh(DA#MxIR41(imL%KEt7?6;j7;#%8+WG$XUOy4;Su=+Mw)3D3NC-C83Dc
zaQS112DDkjga7p1qD_+|RC4Czg)^-**UG&NeEOno&X~nR*myg;PBe-fO+t*>T9q?_
z`YTGhl+WX9Tqe_5Njr%-7Rl(<EMR!yhm?6-4@l)Ci7*S!8E*iDb6yao6JWQoX{+FV
zh?Cy)Rs9~7s~r07UWAZtFYki!tL^GF@-}sLqMYEhb)Wa64=<+a^t81pa($fs+$UnF
zDq4sYT3i4~-m2$Qbi`Y4?=S|jtHh}gG<-ZHh-wEns2l~-tQt#quGT*XbQqV0w9bC6
zvDH>fUe(Ue&%KazY=Mim1WSYpe0_r66xDEKRi0Fg+|Of6Z9;<`&uY#`kex@UmG_1!
zM<fSQbU*~5)!;bi{){`a=Rer>skkP*R|H-^^1?XeXLJ(|7%jaz#mBpw&MxUQj+e1l
zJzm#;^(R*8x#_(}eCQhLx+?43MJxW|cxDeT&YtS9q)gn{!Y#qYjLpGIC$?NXaI5_q
zDSA3|VxiN1HfW<axel*jK$9F*(F^Z=EYkF&6LPQF*lRotUfVt_?0n=+ucRRW5c(2L
zsf9hN{knN;)wR>?=F=59mC^bB%(^~|;8n!8@Y%jxc2F$%VYGs-<xq{aTJQ_*rqPRt
zc+Y^LWp|?uvcsr96fNE^Fe-z<G0A^}S3jY3f;2?`GReM>+CyBPTdk+k*+05yjkSS^
z*2=){lwK@jj&Lo8<Z6Mian%e+2gKRyG`3p!DILVSS)Yl^s<_Ua%adw_(A+Ug*vu2U
zT~ILB;*NAe%#e2j<`ziz<J*3{B#E}Rn$H}>aXR=g_z|To3o4hi7U@)dptU%G!YB8L
zUpf2;?tM@J;qkwbI;`)>>BD}inx)~h>A~uRfKcFO-RwK&_w-kmg+Ui~zt7wnx%jV;
zjst)TsTU0mrI{5K$(bJ?$rmNb8u~w$RWtIIJz`|L9I%%|Ej?W}T<rLjE9Wa!Vs_iv
zFs+NYVT$|06cHP3zylpS%wcuzEtK<R|G^so;&(>_wwqVAfvUsZn3dv{k&a-bRg90+
zaScz{qgZWPPS|nNe299ZPH~m5Eg8h!B+M{0+lR55Y?2Xgr%X`bN1^f2?Y_GBP9X)7
z+SF?Q>p|BFsvRXL&*5emHJnabpWn-@V2)Lkz_Nd3N(=Kbslvi{fc8%em?x-RvZB<I
zG4@#h+#}t*KO??IQ=44-?p=h8b(pXgO0+Uz({xi#W;vRJunJ)@%*Ff!f<#cE+wJ_I
zXaw~zO}dhD%&!oM45Re|R$6cG{1u}(B%;2#Pr=)651*H!IS@1u^`d+olmpC(MdXfL
zak7L$d@J?^W?q1nvnxhzYFI)Ge<`QuM)hx*x;sL4>-v~vBbiAu_hWGR2{4BN%<PmF
ze*?8kjWw_CPqhCNZwrw0b+m-2<|Z1#Hm4nN?vf>rBgTE$%I7p)#OS@|n69rPh84tV
zLb6<=`#HoMMj2nz{yesG=x;V?&%S#zdiVr(aXSt=<2c2;CBqRM<8paNBj`m62Q=5v
zWgMqPnME4ilth5nB24$;X+|jpkJ{nWYnkQM2qUAsl3O&FubmLWw$GJzUUGJa#`OLr
z`A+?)*IqNtpqa0ujc&*@nvn?_m$%f5^>~9G&7i(trE`zdU3m0MKJlRZ<jsiDkgU*i
zMCdCvBB42rZtg^Yj-Xm`IRu|0@M#gX^f(XF`rbzzqI*`$1DZC_#QQ~f5*4ck(5SOs
zr*BRwx0ONVuyRJZpgdJhDhZ`i8B-kPqBDhBr$ZKK01eED^1QA`v4d`6A1@<-3~Hzh
z)m$lGoGy5L7CyUr&XsS@9i97m)H!{+_X0cr{8y*pQ&#C!*G93|=I8Y1`?k+Pc$T&{
zFG_e~X`YxP#h4?-6#OW~95FwNF+U29m<#_!-JsOJ*l$+?O3><p6Q!6J=0q{(M8ON2
ziOtC7002nj;brwnJ@^f+fa%Y80VL5+NK{c22JrKo_nhZ*Ea$zMH}BpnF3ikwO0!YN
zR|_jM?T?v-fk6~li?UgWi-@RJ5rYjbj6$T9h)p(`S(!~%lSD{Fm<t!>!bL^2Fo+NW
zZyz?pkMF#5E_ZPcDMVyYA%#ilJk=c|om9e4%GHT-?sOM=go*&OEJC8A6g01%yrSre
ztwfZ|S%X<gkK;MxnNt($s5+&-Qy0`_jcXNJo7SOS(?+yeZ7I7ed)VuGd%ZvOK7CSu
zuP@|KPQr*7%|@q@$*s@r_l13D^NRBt@|M?R*6zs9&cE&-^H2L{{mUjd!)Dw}n=|HT
z%e5M<nAK<XTi>mfKr|2!%mqt=-NBSSZ!g$O_DaYXY7b2~wsXPx>ilNASPMJBI@v|m
z!|t#_mSSl(!^7O=jl7L_@GgFpC-?wQ@(KQy&+$dR#8-q*a8W9%MZIVhCq+zL5^-@`
z42f}(5g!XG3+l*2mi$8j2!fZ2$e=Q)k{_r>{-mh<MYWi?@)Ma<BuA+X>M_{^4bTY9
zIB^JCP#;FNB9FjP)NODadgVOz;nrKwL_vHxWKa>+(K>9e7}i4xY=BbO2%De`@TbkN
z1<IiUwn8Org9ubXHPnzpQK-f3E^<41yD30>p$YcEek!B`$b-<1`aCiQah&ae8<2pT
z=nWu;;VvZM9*n?!NWlXbg@-T(k6;`g!xMN46YvZs;W?z?1x&$9cm=N^Lk_)xY19hd
zdyxd`crm~HMyCAoe^X?TjSS(19VC-~i6e_~rWDFiS0F1Pg1Sl$Q8ltgPEr(HY;Y&;
zJcb?oYsY?m<GQK<j$fR#b}06Yl7@sPikx7;t=o{A{BRr^DV9AMQpp~V$C)341~y%I
z(RzhOqw)tZ8)P9z2Ba=v!8#k(Y`S67KO#cPCYx;7me0F-Z{8c>SVkm4NY&GK&i!%k
zx#ymH?!E6#%LpBEN4}3U;WTY$@HWok3g(PBj|DBSk&B3$<!`B9!fh<m)(CHewsaS-
z*Y>N{yCj+}g*mbE1M-LX5$5IJCEvq>Uhk71;3|HMYjPiwi{wwp>$riNxP>L$#xm}p
z7)>7|i=>+I>u6wGyQZ=h+R`1eYxY^B?Q8RhJRqmSwE32B+FXm?KaYX(6IwhAB@8eT
z&hfH3ztiD+p@gzoCf*%LzZs^bN{FRey_4vXpNcgza>d%+w020+$c7TitUGW252R0p
zZ~L9*?Gnn;XVsk7`mS^j1LdqtG-H6Nw2u-*ZRDuE^r<YevZzWU%DIDOw6#25tI2Ry
z>PC!}=jpliG}hixHY*aJ>-kZ5%laqtFSC7C&7AZ+u3-_$$#;{yC1RJbj5~VX)%PBa
zE6Libe~2d9{M{kDWKWwNZFc4R#v77*+W%hECVt<{eXI2X$=>lDB!5@C9{AnNC(R)1
zIU^l=i>C|ZMO?-l=COclSj6>YF68Z0<0^aBP(f8$jjW@g-L`z357|NoEHDr8COstg
zwLQSu)S9PZMyf=X$qTp$r$yXh`Mj(HmgVoHmb8^e?b;9YACljVT4&Oz%hqHeUzc<Z
z%tLw&71U5i6P-+JVlVdP5AYrj{xh}tt2jc5ESu4+^kpravV{@<HX8A|2)Uu>Qfl{#
zvQ_CdR8Tdqb!?y}ze(1~c<V_gR<+QUjuBF@aKMHkxu<NOJiw`FrTNEOIIbNTJ*O;-
zij!%pqe`a6V@(9|WHGLsk`sQD*qS1v?|Gpt6}^Hh^v$WCM?8_uB8jP+<Wj+=Rds8q
zpsKz`){(_kQ~MS=h_g3NU-^)XF{SOh1#7-ctg%ZUvu1H*&7560k84=Obv<u<{ky~(
zarP$`{WUC;S(Mz<zqnTSQ9~Ug%Z%4T2Z?1n`h2SVGxXW9Pd<XRv4ejT-c0@M9w@|P
z50m5m=9z-8U(t@4O?~#f5Zlkv*nX00mtJshIt9iwisd6ugr?Z<AU_4fD=A)&JP9uS
z*YlbS6ULcTlb?D|GQS>rV#j@&!~r=vN9i!a_<qe}AV0y4(oW_|X3;swqLVy8DYrx_
zT0$;lBg?yo;@pf_m=7)FRdacoYWRz}xIxJ>e_oKjh|79jk<U-@$D>pHhB@ETdkM?P
zqHcwbRq?)t3aVJg25Q(u9b0JNA)08RjSjkq*0Nzh-;mr(=52o+s2e*)()Ap_o!(E+
z)^3;<WAUwmjPb`K7ZK-1Uw*!RmEW{_-o{=1SF~HzZVeSwu};qhYSNozo!lZD{26a7
zGTXy#bkJ4Sqc6VEuqEEWbjLYiG47aSSxn9&T1>`BQ=i=k%ZQd6an;w+8Y;N2yh^U~
z<_2ogn`E8bA{%sX)ANvQ>f1sa9dy~j`s@@MpN0B;I^u~$UW&e-Qe!r8CZaN0|9$Dq
z`e-?><5ueHd`}#*ABt}PsUQ0o7VfFV^yEoK{)8VNr#tj<8rA%m-L52GvZ(NmVm?>(
zO(Q(NVd+|N>aMzh*HA$f57d2(b?FV%u!%ah(7?9wACk?l{~lU=(?-X*U9v}S+|}$G
zlJPrK`GNMcqjSdDnffoc^s0Q($~#U1(f^;`EBR{54^g~V9{gW@3IBooUwsXq3|G?K
z*x<L%dF@18u@z_V9!_OnNJ>ljyB9j>VxVp+)*ki_nZ7R?C-$eYMSF)l7bc|QZVMks
zRpNJ0ylu%cncaD_$<43+o_6=;d(u17pGtp*Bm(;K1O0~5kCc5U-A%OfzB#1tsODi!
ztH$4z@2eXq8%m$h%Q)!aWXzQdCFf#PPf5-2nAv3C3fUX)3vb0HT`jcH!LI&&dWK}W
z`^QH=p4X4r>RWL=&qT|oodnUtWZCE|Mt<R4o_SWkf%ZI`S<8FT7f42K$oHjZl={UB
zq9J;qn|>J&^XM?bq*6Cy;am9Gn1*UW3(L~~0Qi*bpf7FxM;cG3+MJ*$vw20U!)%JH
zqb_Qaj+*qzZKElj)!W&cAF7ETQoB-ZFw>~9INLKXb*3{>+fiQcsz1z*Z%)bM|7YmE
z<j}LSxNFC9FZn;eb1(h>zT$s<uU7l{(LdZPa;yBeh6<_}z0pa>H#%|)4a7G(W!|60
z-$#`d|M`Bs|A?MQ{|kE>SLvm$bBv+Q>}%loxkK*Pe=A9Ef{CopA<>)bJO4^==Pz)?
z=&wBYP41I$7d?#SUTYLQ!4W5@vEK=Ac!IZsQ=<1R&&e<yE-Jemej3giZ8dyoq^euz
zfm>uL)WWu!X85`L{+FI=*zx>~9lKimD)hA)cxvHu&t@39YwGTr-`a(Ho@Uthw8DX>
i?aujQ=(u&hpzqI~p4;ck@C%mzBmDYb*-G0;0000+1zP+7

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-medium-italic.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-medium-italic.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..16a7713a451ad09b4351a550b1d54345ea7c08bc
GIT binary patch
literal 16000
zcmV-`K7YY?Pew8T0RR9106u^K4*&oF0K~`u06ri90RR9100000000000000000000
z0000#Mn+Uk92y=QtyUZ$24Db$5)lXrftMuoBnyLB00A}vBm;*m1Rw>0tYHU)6B}o+
zHF(ZAb$6>01TL_zVB1XU-43pE*grTMVZp|M@j$wk{r?{m=olh&vq9$7hZUe|RtaN7
zlR5}e<*$1ZogA&I8E^UN59!z^=8So%aH>{^Kmdxgg?s56KV>;B%`m6cXTt1+uGuAu
zKsZ_W!ByT%HnKe4P5<1yUXEqIa?%zH=`aEhvj}jfhe)xD?uq{xqwV`AY9WC~8s07x
zOp_;=PSzC2o~Tq3@ghUz55YD&5E~J7x~Zsu;!LDfEFw}HtU!C4+0*GF<8^8N(naOx
z_KW;20RrUyH>c)s`G>nYwAWbuUy8shuu}>YyuBO^6d>6Juw=s8-V@LYO~NovAy(l3
z_r23xwLB9QL&r{hgfP3Kc%4&x%2oHC%V4#CR{N&|rr<HGtP*_U|LImp2Kh)>7{q2R
zc!C7+)nGw(F}tIo=%`ML_G<5_n%km7-lm~C#Zs<$8xc?7%)|}J!ADx(5K|Q|2wa~Z
z>|{K7nVg)wR3?>5r4ormvgzKJVac#-YU4X_t}9#Kcj}k+c3}%ppOL9#CvgLC`Zas)
z+xx$5%DEya0wV}rt6cJyvhVF?;T>!a_j1g+4-45n%A}zJqY4xOT)!}Xd~y7ZulrNw
zswA+ju%=UZzNRVFn;YGop~e6S8X7Q<DO!DNC4a9sa?%yq9c_BmrAh`Y9gqwU^%$9l
zu%JqDZ~{;RB)-pnueKwS9!T2$o807c0Z5HdCGV9<4IBwGU$^7uSIbZF=oIidVwI4~
zgvmC~fS54#&6`(ut-f){TkG$TzjN5$nf;#qw>qlK9g3SLWKLa%*aTDa@Ojh!1&4!o
zw*g>T7!<UNvOoe&wet{Ur(G!1&X9(FD^Uv@U{*_&Jq~bbXf0s}HH?uaulzEMsZ}yR
z-rKP<jU|YH6N2DisCV0(Z+(%>vG@0JHK7T?7$FK#XzFQ&#}=|>Z}4Jr3LXgJZC`9=
zlbo2d%yhrfpRJk2EJg@ngb>3-m*f#40KZl}Tmazn#*>Bn2MPg90+s;^1jHX8U=a#D
zJYbwJVL%>Ut0)F`U;sTy&4XKIaSQ+?U^d-$Aj`{G`Cn-V@d%z2Xd*NndXe|p3T1WE
z>I-E9L&5@K444U416u>z1-s~P`vY>d9=1MblWcRq7H&IYd*4j_#|~+?#-3xJT6j+F
zYwdr6#^I%7fa60rfFE$Gb^2tSaAqM3k@J!3kW<L(E+m&smok@Dm(0bzo$vD0<uz$@
z#@x#7`F6RoY|2z%l^-ywB1H@8-YH9@F$CSdI0ZkLh|&u?pnk49>Id;Ps%Wd`5PP_o
z)KRh~wvY+2&s1Xo?$c%q`JfOLlTufMz2pKIcYIiWtN>OJD})usieQ0qAWDol33!sE
zNRuH;F3oc&99(pz#G#B;fi(<GTC>}T4(w_60+9l!2VtOrMw)1*g;v^Vr-M$q=%$BW
z`dG*SgDhf*VV1CzWh`e!SPABURViH!W`S{?1k-x5#3^3tE9?OQEKjE9W@&P)66+S=
zNU1Y$rS8n5{(2g2B$S7ysF@tHEjbDKj)Lr6YMYj3wFSbVa0IVNGc9mt#fBXw^#ft$
zny=@96kuGZ{DlHVgP(O9LFU;qi(=4x2tQ5$5k!O#VVnqpMR2t!B8G?~5(pkAiAdq3
z5g9}lkwfGW1w;{1LX;8N(k|p0dT0YO(s2t@W;)L-u3iEgv>Y~G&|<$@gKnH{`Vs{$
zJ(W~h1%{l?DmWqdBC=!7Od7MGtn_46RRCf*jd)n>02-*8CKCvDevgdMO2Smy8*@dc
zL`&ZXx^(89S;EZK<Z9%Om7Yps5aY%)_w9=v)b-sB)5dHgw($Z(N^YKPwZC1KDbG~9
zc|&=>3wmXSnrT9VT+xXQ;Yo=SG?zlYQV>d^4BUk!A*igZ6U)>{<jfLFUDenYaBe>|
zj?t?iefU?KfnFC_<HV6VpYE1BT`OIf7a7{{NKH9^m=IowUK2lIICxSn<7G{X!$LKx
z33U~hX)hY+$gYi(r&?x~g-q8?t$v9XIGFKPBhgF&CW0|I0#cdTxeW*OWfPJ@8GQ8)
z*aK>vZJ;-?R7BF@Vvz`}2aFBq`9TyCLv9!2NP=XA*4s0R1@uuI4dHVwGHXp6z4vDz
zS)gDUL91D7q@Ie=GhOXXT^}YAkT9}PG98uX8rd3*C5#hti1$?n#C8loI|&m}Q8$X0
z-857|_f62LYt*HnC==Il<=!JLzy=<QhK=PiR7@7=4ATq|JW~xlI&F0`pPnB5kV7Z!
zsFL(UMiN6CzTlWLgeRpLna=imd%j1ZKc!?@m<&T3LXSbq5C9dxPya~up^f_UYLUXP
ze*7!xe=e&Ki^?r|1<c*lsQVame9<LuF}j@*Iys6Hxj=a0l!hr}!GKIA8A|DVbUa(~
zI)e(dMLNePcb`O>vxC6)Lu%U`$Ni$q^(GLX*FeKEjxL?n(I)PJDZou_s2bqQqH!;D
z%B~s>@vul%6yq|ZUhh(7+7jz%pUJ}VTF6L=avmLv$ACH_&vfb0XTXpV6Q<0VvtXI~
zJ?OEmy;OFkJzjEZW!*;E*Wp>+G3Vgr!`?U;T3_8#-Y7(ONf9uI6h%=qP0=PXTpKkd
zQcdFgJaxKUhLyQZ+uAnz!FcW9)AH12Vax$FQ?rs5hOKFa%XL=^DyHyU>QZ}|z7B-p
z>``SLOmw<WXO=mOwPkI*xqgV7>0R@sG7b^!wq-altv=Yp8itaK9C4Y2$L3(NkgNGo
zOx7cjJTD#0noYBPAS?<GyjXm_p+4d$=`Kr;H6O1bb;PulwA1HaU3&BxFl1DZYvJXe
z%cZ!idCBfg_4;<97}2K1*E0EH92k7JWpoKxb(R7CTZoI`@AzpMbjoOg0NJwAE+5iC
zTJ2|KSxTJd&Er~J+1!??r(@_SD7*7^6v&mUt`~Wwk|VgQB5tZ0m2t7^2Z%KsaE{~^
zN!F0?5BLwS&yADyydjGHi#!o(XjdN*5^^+3`Z<QI9QCiR`Ni4!ZA{UHZR%y;e4^@;
z%5$^Y8$j!ThUpO|IXUq27IYs1zV{>{)#?NsMb>$&PkoSCF}fN-LNTL<WsXa1u9i@2
zNl8D&w9q0r$HP{99L8}04sZf<&?k?81cKT(BXsCr6#=#ae1?bI6Z8$c$vBkKDF|Tc
z3?6gHf&U$g%jHJcVOeEC<>hz$j--Tl8%c{PS8Q3YwUdy${Ua;>lq+PcS)X5{Fr_W)
z(b7jF*@A7DR)ZRKM18KhF1ZQxb#P(0uvleexwa{)<NpozeN^SJ7=%*wsx?waZ1ssa
z2qF%0H?^UDTL1kVYU6JaR#31}^Gq?3xkOHvC>Va#m1qRz+6_!l6;~%pYI%~;GgdJ5
z#!X;g#te@ND&RApVFH#|#mpv~<TGwu0h6W`a?&X!Tya||_uV&-7v7i;1TYST?~FD+
zhDujHum-E+)eU&F|1R?ziElu$8HNEc2KE{__yOP&066~*fI$DB6}qE{mmdfk2ta;T
zDPR{UU;r=x*n~Hez$*OQH-UKzI13t>In=Yueuw>%0+$6*Loy&Wkb1}>=xOK!^enU&
z+82#Q2cijRI+}ytg~4K7zbMZ8{eJzzF6-=Z$k7nPj4J7=N!_mfPSx#QH}o7d3XOi4
z8y)fErvJQC@8mn)W8K&FOPWgy)Bf-OzuFvg)Da_g+iVr;qlt%CY+nq$0P(z$VUTtz
zHD5gtU|`VN#`Z^+WYP~W)2}i7pMt^q2jBwdmgM<=4*LN=LK1SIw6WpeD96Oc#V0VB
zEH;PB;|qi$u|z79E0ij=Myt~sjEPCfDXD4c8KAkPtz+TvinVLjZCJl?^VTifw(s1r
zYZ-R$-M{a^p@WA<j~u;Inwgb%uX1%s!PD}5(E1jX0e7ANfQK)@rcZU|VsHj{^zy!`
zzG-N3>dMudx2|3P4V?J#_{qaZFJIlQx^eqKV`oEGdrxm)_u!)E1Iv~#9Xor;GIRby
z_2rA_0FYk=0FD516UZiDxdGUD1pqLw0qw23`B0lt(6$fhFcC>)0Hu(al`lCar{s1T
z;Z`tNw$<^2hyey(u0E>+_Z~eHI<^g$jZ~1HXB07`V>Qfu6h*K*pI#z7I%j@pf@n*2
zVIzBNr#z$DWS(UF+zh5??Ju$$s<zpQhzAjt@=)|WW27*d?uojA6S7dkMBbtuI%8=<
z*cF~fZCh=4dN$WACrlh2TC=244Tnjxds*Z{m%?<y^E7h*RupKa^rNWmK-UK<&>^kW
zNiYN6KYDK%eR>4|rGRMPQa_%Iv<oImt5P6Q3sTK!0RU=n*;#>>Y@jbC%H+5`U`|T_
zkQGrk)r|SLZ<qjpnc@-srp;y9Am&l#L%mR9uuvq%!VBXq4n2q>h@DHoWCN;<?kI%l
zLkcBYt*jz`SivSbMeKVl@5PjU^yW%&r86=vv_P@V3feK$ED88Pw1pv`9EQgR8$d=v
z0S{{fV=Ml;vwy6=f!!D;R`5t;w5dE=S6+T~4&CAr#xXEZ97(_>I^Pv?=_VvdFzi8l
zplC$v7$3!ayUchMe1uN4_O;*<M(APQZ?f`E`~(u*gW>~mQI>Q|iPMNo2J?)eyIMw)
zRF&G5O{J&6W5+R5%yke)qN#bjBS=tE)6T=hh0OIRt++}plNcUH1(Pb5#Fd@^fzd*N
zf4mdJi)___&D#WC;zVR{bzXwc@V%mMNI{8;Z$uNb1IUT|KaCWU&TpzDD!0kIdIO}Z
z>&r18V{dg2=d(#N4Pn41y?u{>%W8dekTQ+P;a*@(<tss?Qrj)d1qn-|sYq_G&xKO7
zPO~jmLEb-m1vyzEuM{-&SQ=CeDluagHwEu_<|@%wR<Dsa{v-tX`=cp7+Y#Ce>8-Fl
z(I9eK+0?+c>}Pq_3M(i40-f5mPfQtp&kF9(+q<`UW6`wkTuT#6^x2|HJt#!W941pO
zz9=)gl(4Fyt=W{G^dhJ-f;^Ige*G%It(UHXM;Q=Gssj%*KB7}C&#g82-u+?-Iw?@f
zX&!;DO=a{(pDqIl9C!;~COY}&F)-S?%mEjj<i{+kS+7Z3%9)hVsC{g#gACk+Hw>FT
zYMjEAQs=7*A5mSNMMCvT24RG#aC9}`xAr$kMG2w!XX?aZto|^n14XP4!4qZ$%%Do;
zue>Yuxv1i}APX+=i<D#9(@mqZiJHJa1<Z9bEXVph+Jjc7UxiigVc=&_a`m7RqYUh^
zL)tH6n;LDB&)NbGh5HG7D)DKs>}SgI3eLq)l4Xm$Lh1DwV=!FAJ0I!9(nOoqBN-iV
z1IzwonYrY$b8zoj-bCh8t7i3XNMvN5v(PEp^}^lUWDd2FMVus6-r!(!{{~*G$DJmX
zkIG|DbrTdx?t`>SGbc*;Ueca1EY<#A@Uo+Q;-;)<TcAn3%>Ovpxls=kI@PmUdS|NS
zq&Hdt6c;<-vXkj8S3AEo9x5IQj)qyy=fNL0C(Mce(^yCAHa)<)m?*N05iq5L-E;{F
z3SG{=sJGHn%F6?DQcZ8wIK>f${_#3HHZ)9-g#4b35iu4br}GN8w!{;jpVM4D4rkph
zLh8a%r}ZK3NnXPW$YtHPg0{K0uHGqduEhS`JyuuTeCTuv)D5wS5sKTONI)B<)i0=2
zKZ%Be=5AU-BSZ27zTTJv8OK%xtk5|0?8mTqkRnDJ$!sY)5z;=eooYnNDL?G6hpVpL
zlKYTgfWg9izyOK1056u3E<OXEIS;G1k_1n(QrDK01)v5+y7<Rbx!1;;6r2=fpwrU7
z^=Y@=T4*U|l^)@r9zzF4R<AR-i}KYN&J#DJ*pq;74W3h~)^4oohO|b6!=zrDbV@`7
zTSV+O!W5X!QD*@@!eMjFb&k;)&4$X3*B|?DWvJD)h!^zAy6V`3H4rzVaaokIAdMS4
zZS1Dg$PdA(LmX)CK^k^d{)ODCw5MI4Uog{K2OnX-1GklXb2DmEYN<_b;(@ilb@!Nj
z=Wkxr1P7Y-G*zzY_H4xd^SjSitZq*7g%;aG4_oOi!TLe``)%o)<h%ImJwd5x3UQce
zQkk}JQq8%hG+>Okp({Law2~;D5??#N_I{6`Z5!eifb#_=d5S$jLN+9-M)o=jzB!sT
z-8REo1y43?s69F@OXPZ+0EC;iQ@UVwO|*Y|zYxCro=cCkk->sR>BRwnn3bdfW6r2K
z-rr%$X&rGa5AyqNdVL(sg1uqLs~3R!DIK3FS&VIydbqS#%a$4@>GeIv-l;4-V?w`m
z{1kGG@5zz`xQJKq?UH=a<?IVtJU|XkHaGV=yZ4|KOdWMs+?l;=*SU3-U#k~|f6@>0
z5PsT>LV0;44UFv2VZNTn)jJEqu?bGiJE(&uD7G}EhIYC&01&Dtig?=D3Wt66?Ti_6
zr4Tw|rz?z(ddth=Hk^LBLK!_AFyc429xh%ks<;dXE+yFu1p!DY=W>eId;!mZ^EpLi
zp|(x0Z~$#_jEU{>yRELu3X!`~2Y-0Lh*DOeHkSX9PuKF#JVg^(O_Dr;GLT^)rj((e
zi*@&&x+N3tiekZ1*G42&0Tq7N09302apZ!6(ni`>o9NcfXgjn$nylS%$@PQiWJl^a
zl*1*3n~Y=3fysDJ=wnBxr_%|FT%4Y2*o4~o+J5++lXsCQyvSSF0?fiGJX8#F74y05
z0GMn>j}z=lu}`KT9|<sm+Y>O3_Ir_AH9N8wARqxbm%$Z5*A4pDus9!+Ri%G`so}2f
z(W&xUJ~EM8`*01#cII|h>P!7T=HB}E#jmqyH<No=)!t*ar5U8f2ac<~^zV%&;74A;
zH!$jbX#7KUfYR>_@PdJcBGY?6!?GJIjiCnmFo4m#TGk+nB`?{-`}yA<v-0ov8D~q$
z7k#Eq>=Chf)X3`dhK9_75F(Ma%<J7^_$m0zgNEjf4nj#7Y=2ycrPRk6o+>$L;EWa1
z8`b8m#_!IL)fPzaD4^~~9^ur4e7EdaMUzh}eTr)uPjJL+N6olFLVqQPI@cMSITp_E
zm0ZlBd2_wH#3(&u*x!XixJr&#!O<|^UvX``hVh1{0!O$`gC@MALIXGBv|SVPa*bYh
z4vLr3r~}R^aZgUzFsr(SYxpv3)@lB>gmVyCS-!V$;lcT!Z32o?;Nxlm+?r=iv^C?_
z4;(yw7Gl5yhvJie6-;?lb-i*!DJTaM*HE$g|Mn)E$xQYHSDsHbCgR!nH^Pd&6bacW
z?NFZ+GvFiitI%f_GRj!I5v)VTy!!+^36?E74C5`S=;CALlMH{96C7`^BuNUYr<S{<
z=OUcl^f;$2ehOKavo8OW%Zq$xr^&wT#QT_P8WrWO_4eYRe88T*4WD@azh6D}nICWp
z=>1S~(MWqj`-tg)Zp*uu(B%`-vv`9Rdavb8g27vNx`duP4<!iL`i}^}U4jwVTnN@B
zK^8^8y3!sJL#C{vy8}JMQAlOcD$Y5G*iSBF=o)iPo-kYqF{p-x{DD+YKM~$xdJDu+
zhFVkN9Mv3=(k6;aCi>3*06IR%@(N{ay@D;tD-v^z@;N+CVIK!}M_O1c;b;|XS#E)p
zZIrb_3EK!%zLXj<il^~KGI#uGJVhX*MvUO80txWUEe}=TN`mprf_xRdujA6?#Qq2~
z{tiJ`Ni3e1FU?No3UbmV9s-htlas3#n6xo*dNtGSY%puys?4L8v$yIIz`Jk5CpKSR
z$3*ZdMWwBztT!4c8?==v>R0|Lq~Yg-;;>#po>I02xSZdjlJWDxdc}i7&r?a{S1D>u
zC1r!bnDjH3toj6uMM1K~XxKdcJ!vZ}Vxcf7v&?l7a8`9hVIl_9wvJ0Mz~j^o-j3~1
z-j?mW?VF%HV9$FRiJ8F0RTS_rhuXAK3<~(qNIywIuY^YR_TkdlKHebm6y3|JjSf6r
zE0^yoGdGlFR4z;Z*J~_O=dlzrmNa{Tk|me1WZAV~vU|ZCN8$CbwAX3?b=L1U<*DMr
z0_?+?f>39^BwQ~`lf(yx@bzpFeriMo9Hqc&y^JZOCJt7l5exn9`Uur@WsHs!9q6_$
zw%SQR<{+fLpKu}U;NS##2!bbzNEVb!1&rimP-%~wR0mFm)-2s5mMzSJODmEghznr;
zO&AiERGX4!=&Vhl<of*KBUD6*W3&Q#f-!*?sCkqbLx(HAfbhH9etsMGu~z;FH=J;l
zMarZ(+1iY@<}~yCCS6K`NXLkmYU1d84oS|T^#fvut6~Xnv&gtI0Z=3l;<%w<Olk!6
za}X~wZ#A1Af)__d(|!-<!Nf+IPEeeaoLZGAjZ01bVWg>qMOlWl>_lOLMh8Cr%<>>h
za;wiY-h=Ad`jh3W%qvo)cW(|L2am@L9G3^RI|R3ZKn#r6-)-0V8;sYgg9-UE<WN-P
z1E0Sm!1A)V85{h4yBo@~Y*iHD(cow<<|Z;MDUj@P1tGzDNh_7dALaPBkGP*SucB9}
z^%2ln3Ca{+>ikw?D2?M0@t<41GEP~P(KNZSdEvRMgWU(mx}sQaoA@4#l{42?ZG7^2
z)3RHm-sS)*U1EwS7lnjw0<-j98{E@2q*RrpB_!?ezXy70dDsj^t*Q22jpg^GR6RRJ
zsBo$aS|3>wFkK{2L}H$X^GHz@xap2XKyGIW(bWuZtjjDf&5)~uT}7;d*`ZCFmo;rT
zcV%r-E)Jw-3&ZM+Swqq3lQotlO*?yZ9pjH-TwuIvsai1r{A(<|2w7CX1k71xg`0{*
z$iPZVGeV9JJon96_LXF#1zvLub90-vENj|u=F0k{T%5Dq`{oTjCUa$(iPnr>x7k_t
z1)QY1$tZ5x8jF_VEF(L=6o8|kIX(~kZ2_Yj(r$Q5STsh8M~(KV28?p!LkbO%&!e(#
z20@~O2(;*sa=e$dxxRR3CD;H#e=+FZm3}i0Su80(1AbOzS@eKqQB_rCKXFd8S65o{
zAK8J?q@i=S7B_5K-!gpm#*&6j>vY}b%G&g@zRjhXRki7*eOrL<k({39#xf9eQ6J~J
z{F)-Yjw)ver6g#HMymvM1!V4<jli~H)B0xeVvQTuHJ3FW%b7bUWwK3a{}#*hRHLo2
zY|JS&iBzUE8L}8;m_Oy)WJy#IQ(PQbKxJw1ufXfrZaEhBJ0B%LDW$QUt)7*JzuvBy
zqNnjG1pmDPlnA9zFuP~HEH~!#vsKCh4SsoknmoVz&I$*F0^kt2jplU!dHiB6uefp+
zSI!b})oO%5Hl&fQm{rhN15hBfs6?-Gqq(I3s3yvii23^ItBLgW`B#0gMMHrCK`uIW
z9oba%`TFO4Ps-L9Qymo0KSq31621>sBp1!SJvRfmUpn$wkC2~qWR5;eM96{aL6a+U
z=GP1pbe@3ux6xSx>{ng==DIey0=;c?Jv6UgzT~Jub`;Qsq8B27uq{56!#f0M53kx7
z(ZkqiW57fUJD$T0v=Q@g`s~CEw+!@3%XuFcj@}y1g)mq`x>!z@EH<vT27kdTmv1SK
z`b*{1xLyM%ST)u?^i*V)YcmqKQZoKD>DNCDVd{L);Fu^%t1Zhb=v!B$mxLl^3Ia!>
ztKx|zLMFukX4b3Wt$S3d7)hq*kjGVwgCeusK$W-{iA6c9E>LqtEKROB3D5rtq}EmB
z<aKN)6d6@~QUV_GR~+3n3=vKeuug%E!_df7qwvhmz!+zplb{JI>=iYrg!NKuf1|q-
z*I6pk@EM?ZN3B$tpRb4;8+3*!Qf)oMg}M0(vhT@Ar#XQP_|w&X*}P_5L0+|l9?InJ
ziS|cbSsxs_ixNhmgzP4TQN=TQtS?TFXK>{q{^UG2y({3*ya=;J8pcMJLDjro1E+b5
zIl?$TI|eDHhk<v^r2Ulrq)E!LP5LP{+;}PW0=l_nejxIl1Ws(%oMV{nPC_+2HTV{3
zfjKKhLdLs8Tv9pNzrgap!v11;dzzilJEuxN-q`p)Yw2IW`}zU2?EwagbD0NA5!Zyr
z$I(^Pizn;L^ii>mD#MHhevBK#$v{c9QgJMwP4hBEVE_kv7+Q$!Vn`E=hI1ucL977v
z!u$;_>0e)JwA3UO_iwT!S{A5Mm~2Iy*swsvWXt2l`gve{hs62&b^YO=RBeYeaDhc?
zXNG!Bw3Y4UYEd1Nej+SfRD|c&NnFMY3ZS0!kaq-gBBemr%F0rT^x9AlnSsc`5z$_c
z-89>XjC2`Wr(Riz=m|{h1r&0{>c3lv)mwqbMPigFGSbmReJ6tX8$3KrGLPI&sBj7O
zbDDERyl)+tI9u%x!&M;BpVP!<p3O`IW=Qyq-cz+-Eeh5wRi{3i5r=9HYk-?%q<_S8
z`i=RKp1ZYaYwGBO!ao6XsLgkl-_H1K?F8X{^0<*u93o&vhEp?u*S1K`TZA}s0AD{N
zfce}4D~e->;#+)z%bBpbb6M8up}zR$P<C;A2`pM}&B0($5p}_IE{9>|z<_Z%^x%m3
z-dIk690uek*llb;A7jE=f_)-dLfJ)e7Fd)V#_`8`<7<QHYz`CVz_enwKR@_76o!M5
zTSr@9Me*#=h-TlAQkL~x(2jQ?F!7l6myy12;!c}|rm$T8qJSPOjf(Q2(#Uk&Y_omY
z*auK~{znRy_hY9y(>HO=%AgwQJNx(`Nhiad6;MPDnm5ABmE{g>D1j^818ai*C+BZw
zz$Lm0ekKnKdimatIubIq#-vJSM`JlR__=dxT5Oo()G0fDl~pq{fgRvQIp`)ZNC)IV
zp$cWyA>H8-Uz-+}+lh?sfM*GmDrUB4aj6(?v)&xp{5<eOzD(yc5jYJMutWmRZBZom
z(E~-G!jta57$n$KE$<+uSdiC|8DB`lr}iMEyR-JqX;yaXz8lxiG_YFax4xCnL&-)+
z7~}fH>soLdLlUqhXadV5<CAhQcl-s*)qa7PWM98KOXU6);UNSCSG1WFmn7txg#>@W
z0l7EEchuVzIhH=m+#mZ&K|<K#dZy6B>o%+A20I05S6xM`o{oyJV6VvAie}~Q<wkLv
z@$`0kP4M=dMF{|Q&ikU{i1CT#miIpG-uHvF_Y-_@CpnVAiH&3Oi`O09$yJ$5T1{q_
zl5-CrKL0GYY9)rw3u0CYt4YoxCE@Ck4-BR#9wG8-+wz#dbDgz#mcwkK5`3TYV!caB
zz<S2BrI5`Fark@7U5p;fz3&b})^ve=)t2KvE*zb;BNM_&VCZ35PSqx^^1a$e>*7s~
zEN(Pa!uHbj@FjHut4#`!fG!A$tpoEtuRCzSqCZ3hSrWEF$r9w}h&U<*g@bR4AoB!d
zeC0s+0=l(hogM4rbadWdzaxI>@zQu6$r5j>5bzUTGPLBjbOiwa+a~lL$^lDe%_=RZ
z$TO*EIN@(9UtBy_jF1g=De=~0<rqK)Il|z>kwhmomlkd5+{Xude_Da>pG7Rr0!&0o
zR0@<*X+^`A*@Si#vBAi7V_9ROz>p?1l(r=D6HQDJg~lMp@Xaxl5@IYb19)IV?+rp$
zcg?!L+{K9B==_}9V}o9;cTENLLydTBHS?)4I{nxigb2c#D9~8ebIdoqNgI0jwQjtm
zx=MF;;?R(A6y}toh+AiXdYo`3`j6!f^$siO#D6U}hrHq>dFm^xP#$~KS=hV41T|(|
z*uw~POrEwuYUG%EVvo?)#rII<*!BA?os}(44ug<3r{=nSc#5GDs1hH{2pUy_!kp@7
z<;GH^2~;ANN)QQAIYK6-LGQa5!VALayYz>aBTn_YR)~_lz8h|W5XTZk=b}p6oOPqE
zuKNuX1*F2Dc3YK=%E?J5g?ohK1SPS1LWdv~z9)1-Wlwuoch-)V;xl6+*pdMo%-T$q
zjVJPx(lg6R7ak5gQ&PQrpl<vzi~|BDRtbaxiu2;;o!t#1mCV%H);_5{B|&*`$SKDT
zPPiQdkOwoRc8f)p&(z3ekZPQ%ty|n<#Y0^11e|$j45C)8iNdc?z3SI-Aqb8p51+fW
zI7t+)$lw$-YgH{n=WZ^}<Py}G+~PKcVh@ocEbHG~nk*}0aZCHRlp3AJ8zlII;9Mpm
z&L;wQL8|OT;4*0r_V0Dl)~6wV#br$#)#JT#;GWy1b;}YL>~=+yC5U81tBap8+ts;0
z#;%hrM(_QfV~O{l??i(CtD@)WQhop9y7$D@3(LArOE2m@o9xxpd+I%S#wocJX*`X{
zBS%nW3>uM3m7|J%(5L-SB`EZ5P&3EFyZfZboAx@)5LPM>t`q&u;+nj`8TRADC0io1
z4cR%dG6makL;+R^KTux{PTC0&hGK(9$`YQSkU{RaJ@GzC{ac0bBq`bawh?#{r2g|(
z?w+K`K|;PiqOTV!VP*i>ORRLYxM23+lmYuiA}^fUh1N0%41HJ?7#^+8<j1i&F<gDM
zHnpaAizT(BM3dXL$dE};BBR(f=OH5_gT@S+%<WWl2Ao=4=uMNlm2Mx6W-sn&vTnUp
z(yFeWTP&hh*W6{03e;S2i+ik;Tq}4E(j|iKn0pG!8|VtQ8)<r!PYx}M9^eoVH$uQ$
zhOvx)IBJ2Z#z{pNC0xYrN;;Sn-WNC(%=$KoBxDDPL}5^wV?uq>`=CGM&MziA;)Li9
z2fygT-36&$aRPLQnZ3MG2=V*er2k1FoMtI>N2?yUzcOC@Oints9M^^(C0MD%J1bl8
zZ`m$Fl{=lJrFlVxv&3$~En%QZwKjq0D$#lX%-;BPHE@<rGD~uA7Ysx-V*_v#?+~(g
zxG9{o>}b=@o_gEq;l3aYresv6=R-Ik@v5bI`2aYO$jhx)lPgt0&SGa(^?LQ~mOITx
z5;vpjWUG$A^SRB@c0=Kj_0WQQ<mFsU>LLd+TSK3KMw+(sn#XnU&!=EFi;~tk1^>P`
z2@9>S6}ck>2q9BgdjJ|quXXA6MP2QMbE+ML^KmEPEU|#^;3_iMI|+cUYny$M$yfL9
zlU^EgRyRG<O3Rqf!_nDPz{?>0=jHjIr`J;oSX?(yJ%Kwfqap}w5}wyxD7BNg7{fGT
z95prw22OWi8)640`P2OYxwfLM`}F#K!hwA{;CC^kFXRK5Z}#EtoJ3FbWlVf|0slK9
z#fo!Wsptu8-m#2LSA)FEFu+D0*zcS0;Tas-eKjgwmK6bvD9%{SBuoCR_3b+IO+cNx
zA7~hjLE{+HTmIP+fLz;VncuG|YWryH=#HozAQnu(K>4=VZ32PD*4VAvuTP%<*8K>X
zYyS0*ImpA^9Q@#Muqh<i{J<O(k`;9EFQAb4lPP2V-+hH=cEr9a{J=BQ!PumMr4>~P
zeoUqK5O4DC901##UN43081Q_#a!pMI^|6G$*x9L<gEH|05FGPxyyO5!SOndsvP_PK
zC!q_iD)XYE<D+}x(&IYy`wDnNoa4tinnUV?N7?M7hqQ-O?Bho{L+DT;P|R~s`6`y!
z-A#;djXUZllDgaCCy=_jNpbB7?ftn*>}iRcgGf8mFliq5jDrUg?v*~H^VB~QlN;`y
zMeFTL{U6bROLO*g$Nu|rw0ivyJ@01!|FC!<ACw&^0rTg`%B5F)Z8*J5+*#Hi<pcFE
z+Eu_uAPeS7;2EKx`<-Jw7g{lUE1)dIwy^(EkLYtj-0xG;1aKeRo$fEDQ1y781<l#M
znB%W^ZtR&n;62m#{SKQ4BKN_`2H%D8GHkl9K4u~>7A{_24#z(&QcY%iN8s<jLfsTB
zF2H9vm-O>|OW>a!4GQdP>8Q@3<hcMLbh)342#WH>6-&8cmBshtP>k9C`qN@+?vP`z
z4(Vq0LZ1}YVX{*pfVa~E;KzFBK^iN_hkY_@;LT&+Jm@Vsb5`aF2-8C1l3)MezZG@7
z?doqD%*`4^ctk|2&|KX&Jkm=t+%Hvg=NW}+E_QPh`+dcELV!Q}y`OmH$6s|n_m+7e
z-3q#mZn}^-oXQbkqocbma321-@cwfA^~L|FxGw^?#c?xXqW)E%f~F-fG@EIbh#f%p
zwXu6f?37~}2P*OemTG|kMF471Ne#5v(-Hvk8Lbs0;?m{KdW$7Qo)#6fe8DBBmoB*o
z07XFZ9!2&?v8M$9<byrf{4*%PY6oU-p2+b8s0(D~>H!(Op4p%6X2J<;d%86(*O@EQ
zDm87L7ofMXaGyBBy8hiw#1S^-aS0m4<Y()^2h)YBGuzybW$GVtzV;hqnUA;qO-^f`
z_Jxwy#)=-`c|?LO@NatO?S78uvG1B#Ixn;AENp#HY~o-*tUU&K`O}cO<ry}oX(QyT
zH41Y`W1I$MCV1T<lf3VRsr8f51*V1kelm;g;W2Z#K575+cs<)#fehLM;4Sa~0^vC*
zxF4+4=mLh~8P;TZvO#E&tu=L@9oEjX=XE5GD3=V5cDp&9ylV}Cp!r<~3F2^%3jt5G
zkuQQPl;^jbcF(dyqu6Rw08hqY9%N=yA#ek!hzRftGqqvKXAX#DqA)$?o(P+jDIhe!
zF_)4Aa>5o7C{c7>Hze|U?H$E;e=-LT*|v)Cf*MKS6FW)@$V4`YMg@<0hs?83@^G85
z*_^$$>qUWckwv>``u5#*-Ms=OEa+&z(?XZx{X^$p*bIH=!w=&~DC-H5)P>Q0{Qv4|
z53wEd=3*~}@z%vpe!O*I_4Eiaer-ZAcUs2S>X_-#>@9(iXK<TJXd%Euc%c6Zqcfbz
zPXfZ>(@!3on!x+?l|Y}YWkuLD-)?`FldRsqAUXJA;N5f633_?=b2_ebksT*rye5}6
z6=w$L7yT21Q_7E^jt(>{rLv!EDb!sHe)}K5#?mMjd&RfluD=1=&P_sI^#5}EE<u;p
zg;9HJY)9=B{?XzvI(pgpSbG&zX;||Ty?U+bcb%i|ob6~ohKegK(aK-Eui9HXk4+Q(
z^8WO1;xyih!5vTV^xqp8y87kn-K#g>ee>Jow0m6vTyZhXmEqU^3fvMAh+D`GJn88!
zk-HLbX;t!-c3Gc*S4~|PcevsNQ2)EuMi`IcdGAma#_KJWne}K#O87Qnat%?Ad*slS
z&u*dMqvUJg`G9x9mS>L|!6$>HiWr(}rm^m6LhlrsH+|F--WqI1Lq(MG?kowqa{_sa
zcVJVSoCs1gQy>x@oA#(6Vf6KN=Rg7;iKm^~pQ)RtEg&O^0<HVN|1$8BVWin@B^Ub=
zR{zwiEiNRC?X&dG%{^flC*V|GQb0mPK5p2`n9Psg&`jLPVfcDrv*FT;f4bNVBOFZM
zAzuZRM>$iGh{Gn$B0eS8>e-$84YW2fGw&(FO5+b^%3wO()h}9d?*T9J70F7G$fv^q
z86Ok6p8DRP@St0Qu^XA`jkVJ{vzT!73Z<(Z!`QaWw|RI*lb@X->u%v|Df{YpQc4+q
zqeuC(+qeg#*%%oZKlDPg)K3sKSn;84JA!<TDbekqaH!o2USpi{E3JVc+s8DEN~X)L
z?ZerxF~9ld|1RsmoY#TMRp+>GWqbl&73A*P4+Td$uei4u5GW+M&4$1Vkl_t2+?!Wj
zp-pCQ7>N}R!dAgWA#ZIT`keY+;&3oFP^Oj+T4SSYwSfBto&Y%%r6n^%I?D!5EnM5>
zy1rrRD}@*QsaY%z75X#{ZCBs3k_8ZW<Z6??bacoaE6re#>1$6t>$HJ<$O)Z<;)E%`
zE4$FxG;*2`8oYcg3ry1i0wK&ztQf{=$wBYbvW!~ROMTi*e+crgh~g(xFW#Pd0K4xv
z-w`GA{g&du<!DYpUY|5CVay~sNKT!OcoLX5I(PJLO>(^<aluwu9im)uhJ1<j7-B9!
z<iS{HDJlh?m(8><v^3m>YRwWkF|aTZLLjJB#&{l4l58ScZikR+UP8CnTsVmq4~qN9
zv@-o@P0%#bPBR(;w2Cj@(>2W*St&}4GZ`K=UIyaP3ZG~SVQuBKj}eUu6qg}x>1zjt
z2_uEOFc+F;(B<nTnZ@DwkW&<M>nU>Edh%QsFe?H;LcD+sb);siweFPcUZMJfIbwHq
zSY$NaxFWYyqc)V-c)EE@fKlV+zSkw{-*UHS|1;jNRgGl@=v3cY%gVL#DSDqMI%Pmk
zfIceW*LCW0{?zr*MU*izmbt{ZqOc5=?jdzk7(U`QbYw0GQ8H$^1=dRkZCMP#pwlZ!
zAfi&gvF%KDWU_&KyyEvYphyiiif?QU(_Omnhh$syxCBug#yrTsa!iw)#dq1In*N+L
z$L;TwCTk(c##c54Ugc)xNK4(Z6iw9=yYx$9n!Cc?7fp}ll5n{+Tf7A1U<t51igV4O
z9PZVCl!_-?d-$D1Lyi><jbM_mMQ?4poHY5W^^JA^ZqsmGIys(F0@jbnsY-&ADQFB7
zdzLKSpvUY|k)oyYQ@Jj+;d#^HyAqx_5==A-3UcEs3~Q2l<0PwvGjFnQdo>6teku1Y
zrc=e%XFU?~_Zzo1W{WwmJ!D5*c@TSW_5Zn0T(E7)TxO2OOVHhN;<>t3ft8^Jg9&<X
z%u$X$q}eqFZL-{R(sHngfz2N+^MHcJ5FmM)4VTrH5G(Y@M#$K)*YZo8t2!<%As%cK
z@@pk=55&Pnd!+Y3C2NLDT1ll82WnQ)50IB(F&=nCrQ!bezQ2<u<j$BNzd>XtEsbOJ
zTKebCg577Hn6}V!w`yPF7Ohudg^8V7`G=Ug1Tw5mdc3JISMCURxFgygX=`;$!=Z((
zk{1K#lL$UXaNaDjj4JR@M33&u6@=W?Gs`Rq(#)llUIlFvP|$$*&&#a&jVH5;m>-MZ
zVitwhnBHIStoy=kDl<FozfECrAHqhF^HxRGc4l+D=5D#J$B<vFZrFXJ`4E0KB?U=4
zXc-*#yk^Ev@b9pJKozt2#1&rCLtVw@w?T}3a}9}Frwmba&8!b5DmY871#*LSJgRVy
zmt))0N<)>RjC#a){rKBTV<vRqE+qs`b8P=caurwBhG&5c)H?!5R*oFXR-O>r?q9@?
zzc+yM(g2PF$*-gy+zM<9VDaK%`@#Q0zeB&&U`}<PzPVZZR%ckSXSfMv;741oi-V$8
z0*E$M=nn9_pcwdB!3frf$hEc+&{-OxecbhvYZBddkFvR?nPmz0na(4N?3uMx{RpPs
z5Q8>_FX96{EwLv{&11P6Mk2^FiKY!`XyK4H*^_R|IC;@;XKjTm$I)*HyPVE@KCC)c
z0MQ$!my2bkjIl{XGPIf2#l@1s@ofY~3f@35QM)-uW%}xgVnT_X)XFixB^E0RPp62u
z)vHb82$OVCS;EqTQ0vHXHHtlfTsi7^vIB(BLRuKi#ABwHWR<WqeIfI;5o;<zF;!c|
zAko1l$zNu1A!K;5WRU5|5Xo8SY4RtE=oQHJ_je><Ar#Q;(=Uc3Lvet4ins078)V$#
zRwH~qq!9Ri6mH~U{8ESXy;$DluMNtWw;%g<<v|Cp0KtkwiWCf1-S8+rj2oAB(5fG(
zI3?QNpoB0}V48*MtW;>Ly`2iR;1afEFoLD-9STbq`u&;1Epuy9>hzRf#(YrlsyI6c
z0Ffni;y^omZHFsEn5nWdCt)c_H}}gxwsO?v27b50Sh35{nCo--!rVwqc?@MOa9f$$
z>~PKxZIeEXcU0JLAj5kK`spn`_dz)SX0!!j*w0CP7%%;(@ik0effHOyg*lTv0hNjr
zCMnmA1rU$&LK4OSEnhDkr2ti<e3Gp;*N5A>_0JjV%$yTo$8<6vwPSD8UNHH}SO{YL
z)$iGs+_<DBLz7_^yTRFXZ(TDw)_*&=U4aU9_*^BZx_*4_quQMH6S&!#L7o+8rX|{3
zw9AhLanu!h-{>XCqtOUtMPYDsQWHouLQPgL1+i>|_K|ui$kRV8u3gCLmG>Y2LC7z6
zY=(|GMe28IcM;6`8Pu{7?2$t)+eUyELST&=i}q9e{pdAG@b%{{{PM1p{s~0L@(>Td
zes^+>lW+*5;mf9}JB_swtx_G*D)j*bFzVJE@Q|!~;NvCxZ;?3F5S+;AsCn>hTc<^z
zlr{WXr$%$gP<4ZMZ#`^>ly>}Q#+hyp+~7+EcbK_f{*Db<PXF+rod{Q?!F8Y|U%iq7
zdrv^4_?ak9K9ogIRWL2k64nC5B4LvhXc!<roJa&|f7!D5AOMmX!;2yF?_X6E7ke^b
zDRlKCFJF1j(vXn#9^+`rXEdxF)H;(~M4oqwpWy~HuA;X$Q#0|%UXX>i-h|=x*Y*9?
zU5Lk+PAk)|U-ezye;rER&%&!jNj~K1bN24R6HtH&4rnRjXacCMEL$vG#*I^s#Ckks
z_y|o#A@HDsf(bA;UBCm9OM-(8mcOL4Weh4y8KO2i3f(bG$uTqmCdvebh}0Om?SeLM
zFiAKRMJcFbrA&aO>L6xY<{cxTl`z@<p=5u8U%ku4(mbykbv@g-Z*Ii94HNoN?}Nvi
z2AHI#7H40QEQ635d}w4>n!6;gBKT$PTmdWknxjKLP`^*};*n2^=rx}2V#viLYNN5M
z+;ZqJlZ)wWe#wB3BiOX9(WeHeXl_oM#l@^a=>{$XS?%Jp^4ax|Kju$xzidx}F6<{m
zu&<l$JEC8Hq;UXR8?Cjf1rY3(R|iAv!QTZ+uD0P3H<0hQIiPAqr#07SrPUSOSiv*!
z!vjP_)vo|FSf6d>Jn-WgFc)z*&i=shg`JTLsy&5F-Op*CDK;13{~~Rv;GJH04Vq^x
z@Z;qT%!8*iSK}u062*Zyn>)cc7ZGDbX9pqwf-Nan(&f5^f(TWErxoCsQ9(QQZpEv~
z5B>zRif@Rtw%5o`7E$-2gfQAs^;{?*p<t+FwB^8-9H<jr3^r5}HH_l(QaqT{Xp}Y>
z2>Y_?C+wi34X1bK>h51dOu8D!at;m~7nxUrFWyvA8x3?Jx3vj1O2(=5W6|;fYzY|^
z0(0>}SvED)qp<M>QtW!C7kJnLfih_i5gTy;ix-oIiS$7Km?c4m<RIdV(W$%?Cr#M1
zGtP2WZ`W<g741=^C>3aqGb^LAY91gt-r3X+GIYmIquekdJIcT29}2@-kPp;4v$zF{
z#-Fl66|dn6+JOrhxN}FN9ksT(W>q8)YiWhb`MRNLn2GIvqxF!aQrr_c8+2fu2D;}#
z+fl5QV6cNdS#QLH5VI*owEnHtW|0|MoMSzM0lKsXw0f~YI0(w-TnUuXK@WijM^5;v
zA(&g?26yk^#f;+EyP7_h4=>V~wX*Ttrhsz#l;p&)C)XC{3oQ-K5`7M7;g!^bnW%9f
zPhkpyw6ka=>n-Nen@~~2b#yL1?hBg};S=jDqqd_;QyhI|w~}(FaD4<hp)`DugXv?b
zd}s#~Sat+0%)rG2_6tbXUPVPW=t8d;D`aNF-ATUqYOC+uCB?J`ih}~$M$9^$G@W=A
zY26Ov=A&`Goncm37WzJ|?MSUmhuSgM8akqpHGQO{x*$6vA9Uru#2|xS{q_mirm8@#
zYW!G5^bupf87;4Xh{WVBpzp!kf_36;5BCC+skpUV8F_t<6<lTjR7YAz1s~~DIdw2l
zu1R(8!_@X92WepXtYCi~#`XEJBbN*PIKFj$XQS|2PJ1MWfq25|Y-13JGZVo@$qZB*
zo#y={ofP77;saBbC6zRKf;jN~c}-MZ+dCdgR0I($nj1JMlY?W^$OD^Ts%0kD;}6Mn
zpQ)Q{*ri-89pqe^AQeU>GWJW)Qg28K{SUwbAV5M1E`$jTlH>=_>HT=R!V>RI3?%jI
zlZj9HB+{qOSV$FZ9%O8hh|{M|s7hilI`r{kgh>P$C1Iy<ptF3_PN+1FaR}8HIBhV^
z77Nof&c+$_xcD1G&;aAJ7oj=HdVER^DKbq>iYQV+n<zFdAGm7h%xMOT<h$#fh13(X
zQBIQ4ADDF;OJ`)cS1Kt+@G+-l;6Z4XFh<fIcw&9VFz4Z3bOqU0u@Utwuumwkh=I20
zWhjm6!_>EJRgXmUO%wh-REo0*ALAFgQODzPxX{1{yuSDV5jB}2ofl9%FK-vT^t|2J
za59@+Cw!XeOEc#et$XvdUdIL&g`A%B$Un^xzzSJ&3gqA7+({8EF|&f}aTJ+y(DAge
zmkBCH804OIs|s~pew%xhhBA5CD7Ads#xTCarhx9AIVjC)5+e$KhH|_lg?^F|uJa}w
ziYz-~0PoxAyc})(L}5yY-ARi^D!l+4Iweh$u+j6Vwz|>>@TxhhIFM*M-o8O+`pCwE
z1urU7$(o5;hBWAz;qz4if+xM8YtMjCi+V=f)gshrw|@-I=}R_6gN1j<D%_{VT8a)?
zqxNZuO{S7fD%__BB6-QCl<m_>Brn;b(DrF93NP96#xZXB?ZxsCR(OmPRZ5Ir(shGN
zzRcX(S{%|w?bCW~qYi0n_IWGW#+Pg?z*AZf^Q?f=Rn(7^z39%;`m>HqJAsM{Y%p_b
z5@`TCqH!_9xiA8bd|(d%n_N8y4}zM%42cPeJ(#>vLiA~(M-egabQFP3>b7z+?(le3
zMp>uUh$BmF5Zy(;>=tv-)g81=qM1>!xz955+SY(ajmI`N=<rU{DHA%J$-IjM;AL(E
zMAwh5NfK%fcoLhVli&edp%MH~FOc9H>gXl@c&$d5Zk|abE;Mo>tODW|k!wqQl6P8&
z@Ubezm_K)yH2VvB?2qKuM|5=;ypDY^=i4P60<Ya4*b83__ajS&)s6EHXS)C6k`8~?
za8>&&l`HhhjjRu3mBOBcpE#LAPyY9ih8ED#Gccl=oH4V&tZeKYoLt;IyohrbE?v3C
z@bL=>3JHr~aiU`45_m}|X&G5Lc?Cr!WjAi!x%c3aprWd#u0hn)($?|hSyxZrz|hFp
z#MI2(!qOXWtp->|pf+mN=7c3)YuB&GYTIqn?XDI>`YhFH+CyvX@X-gKY_Qu|r|n^4
zHXzfCEN7f^$$1xC^hUO8F1uo{9B&P}>AD+o{poLA^5rQ|q)@SSN|drFQ?62lD%Ji_
zW4?J7s8j2(^%~S`)Z{P!IO>+Cp4n%==UzDGrB}AN?TLHtd*HD<9y#Ek5u<*0$Zt++
uHl@Q(r;HhQJk#r6SEEdr3}rYYBadhFU<8xQWk9nRt03g9qlHlRF8}~ScC7ON

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-medium.woff b/0.16/_static/fonts/FreightSans/freight-sans-medium.woff
new file mode 100644
index 0000000000000000000000000000000000000000..5ea34539c6f54063a4a89213bfffebb244888eae
GIT binary patch
literal 32072
zcmY&eb8sh3wEe}ly>YU!ZQHhOn;YA<ZDV8GPByl!m+!s5URRyEbI+WMo|&qdo@zHa
zQBeR8;Ai-V15kbp5V!sR`TueMe~5}I%lrt+epuZ9;Me(fBPJ{&`oj(W_yRvr2B8LN
z7n7G$0sw#^0RU)20Du|a)UnJ`Oi4u$0D!gr(enPBlU~H@8<SI_V*~);s(#{r^Z<Rb
zK3jhy8v}a)0RG1X3i!eNVY?uxk*o7B001HQhm`yeD6_D8ruJqwKiuq(rv4|NiBc0?
zshNS(Pb|`Zx}Vzr3q&((57Qqm>ZiodENedCV4~5@O$>|y0OPnH4eNhEn5_F{{zLw7
zqdz{u4@lsLz*^01oZWx8lb`&;002mt$EyI7wVlzA&W!q}?&J@aDQZ8(Z4BIh>YBg&
z#QpjYzz`5}wgxsP0DuMNPwhYeKvX{dLy_Cw&dC`7@XY#&yZXtw$xg=cXzyt96YKrc
zAE4=n=U!s#iE}eC{OQ?;|Azzr4;xill|S~Ue*j?VD(Ue*JNLde`=4Ezi~8-pju#~b
z?N5(7IDeO?$(@#4(fzHP-%2kIY0Ll%rQT#wl0d;JVQSjU2IUIhDPixp)(V<E;)`{)
z|Gf(`=>mTAlgE91NlVKbbp`9obeVd>5JMeYXmSXCI0P&cQJi5IQ^7EdP+Jk0FpZR`
zIz!abyxZcCTb>fSPoPP{!ij3VDgPH=Q*wumxm30CZ=wk)d}@`5c^M&8i(>vv%OR!a
zyNw4fv+2p{i=oaIHw-sEA1Bi<r}r7ZFQ@cd24;Z3#h}q_j@KK+jtIbN_j~<s4SJQw
z%dCO!yDQ$F%Xgno7d~fO?#e(Y{vjH@OxWhSThC+HD8xCG%PiKL{Pp{0FHlBpEb=Z`
zl$CFF&}Pk}g;#Uf6P%4a&b+TU^k9>o#Yz*MJ6qXj=w2IZ8N{Q?n=4NN!YM|&SDXdp
zXU12!uh@6(Ozlp5g1`>)4zyvkT_NkiM!`^8OAw1O!(c~<BR$0_|1-A+OZnTCy8T2`
znjMT)!SduvzROFCm9M~*U8(6iMy3OE-(Qnp1_+ob2FxQ0>?vk+rH<e|A5&x**whRx
zY9sX#R4i&DoUx%IGAFApImVVg>0g7}nqy6RX-MNjtAiOw^sK?UO^UR}AzdL1nG6<-
z_Z)+$L~2wMn|;TLzh#(A^J5$<L4}GXjB1Fh$#K-xAZOU?xhS|KbW=P*O(eP(2gVZ(
zZJ;c*Mo^v9b7zGQ9_}rqEl`X&D8*2E=b5OTG#Lzr)oH}bxuN88CMtxJW0Bo@{O3sv
zC8H^~5_1xmYPu}QMU!w5eexxH^)+G{xZM2`x+d8>n=LG`a;e8?^JSCIQ`A$;T|EO>
z3wt_<kMX=z7fB3_`NrQ<PeeF*Olg58jWpq^PYioqjJQ545jy7reJ_fg3E&IxLhVtl
z9AM=PKzhWihq1JcV=x^6sbN8Fft9$oe*F|V!|#JkZL;1tZ1yNK|BOg(algXl(%+eg
zB4)ym9U%1nz0W>eItoLmziVjE4P&vM%DiER=rl>QZFD9ZAPm?T)3rx#3h_!_5kD(=
z`tJ!a_P<NP?jKx_@x@)ouTb_1Y#)5$-RT|lj?tN1R~PVQ5$ci6Bn%u-9dO)1s?HVI
zOm0lit1)R4H=h*i5UG#U1zi}AKf)6HVv9paz`^Pb-xGrE9k*v%6R{=i9IG})Xpcs<
zZ?)pHXgx(d=ac3Yz&GhIN`i|U@gOj84}9VnXiFVyJfMGt_K4x0LFF^-AV{oDsEy|%
zq>JMt^%L!p?Gfl1YqE>COXMS`OJKOe8~zOZAbqWAvZM3rO6S>*;!9i^?Tsruct3DC
za}N6O_dxOP8GC4FMaYZvU6ZNZl6<nKwkr&MQgjX9lbII_cbB}Uq~TUBK|4iGBBPL*
z^#XY~0Dp6!`h2GJMxhh9hJ}iS*3)?=scsLYk{Xtd2R6FHW)cH4Z#E^Jsmo>E%lVva
zS|rlpr;RHQ0GH@wMbdQs$LJbW7|$<BL{pvFFt4mkN<4$^(F{3}2K97?=`%o~EJ|ua
zerf0d!+nSylg-$osu508DN1)_VZ6GG)Lrl|cu98Q*r?`9iZJbwMsxOTA~=0gsL6lw
zk0W!^iam#qwxaG7lDr23t*TNWOSHx^ifgMVfuF(1j8e@(k|Y9w64PqhQmY6xr%HB6
zHOE|39A)E9mwrrIX|->js0YvGBpT&Jq@<T1hp%Es?A0GLX#>{UaDu77Sd1Ez_!TL?
z2uTze2o&T7bVaJ-^N#9*$1&*SRS&+mMWL><XteN}sL6}Lk(@WCbRDWWI>~&KsCk7v
zk4p5DMddd;arU&$)ORNrr8H^<gJ(X8x@jDShE5G!D=3cHfVgl=CVgW(51<RO%40tN
z`6w&!jUUu}PFuzoke4n&^N3OjPf^gMDl>ghD22(_@WY}YJD45sn%iAt-nj+a>P^p4
z6st0uu@P|$dnQ<-Nl`?zYuQKF>u+grMk}VaU33@SeH1kwM{&+b#BxWv$BdZbMk~H{
zHfQxPI<7Q3t`yJn9AbUi&ahr|BXwUf+c*VWVzh+K3lYzYmeSm8ci_acHe98PYD~cI
z<03sZ>MzDqjXKV+wH#l4WUuA5FRt6o$fIe#PiU+KrN#3d(;yE$4K}-nQi+Y6bw?)6
zx%G`z3!$w>uXg?sFMkxB#t|KI8Q0R7)2g|`=c~xraZ-g%H4oUc6>%+bd1@DpcI{+x
zty<D0UPjK*r)A;3b{V+xt=!%iW8T2*B7eiY^0cBC!bTbXL)vg3ztPB8!=p>X49!`{
zSJk6I0=-V5eXE<<79A7KttZ=v4Te<H<~LKPOLb2brXV$QWV^M#HFD4?*1NiAwzFh2
zf%<)wcu#g4&o((e)hTGYjm8Wyi4m?O(!(*$9<5zjPRKI)9;msI-1HCokE3AgW_<d}
zq!zG=B?&iMeCzt~{dVR2nub*aKJB*8W$Re^`?v7+!K<m_)b+T-qiz<eonK8oR%6fi
z1U=!-3(?btzKNz)Gz*rjSJl!#sXLMh+~*Km57rpFBQ{XzDF3lV7qv-Jr7^X-Du;iz
zHQP#aWl5^8-mKX!X2;FWqjD^y(Yy)w_GMYT_oKYjs~sEjqOGN+3yUA(8YW$JJqve*
zNKQq@83-`u&fA3wf9DI|TGF*6@pho|FY%+AC1Oe?r|mwP*DB4fe*bnOYE`-3a*b|i
z`E3Qtao#T2s;y}5l=)tWF?l~fAebh=x0d^?KV4%OX}SqpepDFF@>}^m5=FyD6MKpq
zCwVC+p%aP248{j>SGE(_>buEo)c^N+O0h;hkNX-aLK?_Fnu6fk!)sC5b~j>KYw?1!
z!rnXrnfZppYwIB`^RKfwE+!21cQNP-QNba#qX?@aZsK!7Nn$UV&g89V)E3=hw{J&O
zv>Od8p8L^AcllR8JSI6P0ve0L&NZKAqm5hbnH=*GRR<ij^@Y$o7RRfUBh)+t{I8hi
z6Y4h>2Ru$k$CH1P%%**Gxn@owNaQ)k1^93;&=Z-lU{wx;u+)OY=w6O=eLm(1p40=5
zq+0p?rcF1Ve5wV>d%v%;*6TGt`~=7Jj*H;PzgqKU`@UX{ajgAeDVcw7HL8!RQmgp?
zOb6~!P5V`gMq{KY+N7x%<@^@FT2}dTRbhm1$r;;{C1>7r4V==HWnTQKI$KgnvL42(
zS)J`tO=>zek6xbyxB3Sn+}E1K+dV8TdTJ2b8Of>c4`g0uT4c2_BF`YQHSyUY)+u<(
z)(HN#4D%k6A=bpXS>>R{PWGx4^$4#a+Y?IcE<*vF9{He!t3%40gCPq%3Jtp;cNBTp
zyF*f1EP1#GyD0TvKt1wE_@6NTkDY!fk9Ezg;1zrD<JIkZVZ7~MZ?|N$4Lbj<yAZB-
zby`HS;%Z0GJ~BAJL(Kc_Zn-RRM!XhpD!9cy#R_zCa_v%%k27!8KB76plpY8-e+j<$
z-w@_cEA1Gk6YLW#5=y+>Lgnu(^~Y)j)6T52jWa!Qy)KT<qt{`iF3j_uSwBkBNg<Jc
zrB0xH)3Eb5%?rjd4I!-;WV5pO<zQ6n%q;ha>bYmrTo}C5)^Ac-OY~X~7?gh^m4Es#
zzPE9I66B}k?iDQ_t$hQvL1_MbWV)}r$WKTP$RqMRmn<t5iC2jTV)Xq%#n3^;AfiQG
z5M&@lJ}c?WqY#WvXsLXI@(m#P<r}59*FQ4NS4O^|Abeps&3%>5#DoFPj|IZr51Z%^
zfs9Ro55LZfE#{Dm*(4X<O+hlO5W5;1>vI9zh?TNxvzr$umZ3EjiJ{>bQoXli5?Hb;
z%h^6%=a!gV%8n^xD|P~zt34Nxr!a*J6PTjaep452(^MA=Ab*?BZ+%F2-Kz9M6p19~
zwmKB~4bLmH)O5sxGRtN`T9ITD7>nT85q3k%JBXt~)P*dxRxMR26<ch07|M4%g@sO4
zCCe}@Oj~Y<*9@i-S}*4G2$6EQ4XEF29@apVy+nZM-Nv8{a%+z|>w;Og5*{%>3af4y
zcn^)A*~(LSn2Lu|GEx6ZA3}87!SMqc<z$ij@A?_(oA&g-{0T@7__LIDNDTrqKn*$N
zoFb4ohA7TdSI-bckpcz_dk%uB-i4L-cMDFM^25}?=mo;K?pyy(O<z9^7TK74NEtH0
zNFNB$E($mPzd7~wP5t$;z(v6k*+C8%0$-_Nar~RWgMj0P`s7D$av9R<15Ys*@W5!n
zkiaOxIKW83SipS1sQ$SBVTq8e@bhi__(oXF|6AwJ09lbBAiysmC;(nDk6e>*`~CeL
z`%Unsce8VR{oPxOkLzcD()R`Nj(_1>x8D~6F+sr0N8jsrvMjY+DzAnFhPj5hJC2p^
z4~o(OoTsLtzP8H2-rnZn{`vw2`X^XKcxaH6q`1h$`0xM?H91K|d1;Y_xw*;3`SAe`
zHabd1dTNq}y1L57`tkw~H#<v5dux-2ySvNB`||??I4Dp^aG;-<sIb7uaQ`0)GS>P!
zS10*QTC4k9f{9dWt#-HN)cWH&tMz7=%R~#_#`DE?XKX$NDa*a>P=DYBxrqiWUL$4%
z7Uuj!U(H0hfqH!<0|&z3bai*?;v(ng>|W)H?N*Pk5&1wLF>EXWxA#+4{@oE?NJ`Gp
zcFh+Q%6o~dQ5vJ!m32tr2y{S3g+C@BLMqJnzeOAXqyh{8C;;37JOQDAI6x904Uh#W
z0ayac09$}WKrW#0=coqI0O$Zj1DpY_07F0uAP7(ghy?rr=mYEket-;sFc{K*&xnpf
zr>vF$;uahNNWkFgFOGH8#`L&huY545!KG+JW3E4kkQQu?@sSQd1pEQQaO<^LdHF@+
znc##N;@*P1gvd-*VFk=wT#ZiC=}c1N@fB{TRx?V+Bys&);8U!hvf|B+fajv~^XRCi
zAP7xs(}!`F_b(Xm%()~kOTZzI9VG&m3|D|=Q|Z^-2Xx5$2f;lTW)t(m!y+ftDIw@l
zrh`<_HP%$e1j1Y3(w-qbeNkG;KU%&L#jjT8W6O!U=J>g;ASR^OF=l=xD;v;^uSLS8
zdk0jhxQUW_*0i*SFA|i#KV!}+O3jlm>wP*6jVJm+zo!BqRtHo)VDaG*$WZ57Nye$+
zke&=H2O*IXaiKX5H^;t;_+Q15hr1*sx%NYv!H45l{^qgDHGXqv%GlW9^}lG~_PO<9
zNpZf6hOA>E8FCp?`a=l|imJSZV9#_@gC8iq&+sStAKu3LWkWVytqOCeHoL$xs6u-B
zZfxF7M0BdQPe%&6*;QDU%~MXj>rVHKfaKuA3<wzYWzM{RNd~-eQ`az?!-EQWvUz?N
zG@pf4c46=~kT3MsYyNlpfdD|?-v9*n^~-LkG}NVbHeXql>kZ{c&Ub0K8Oa%Gge|58
zW_tfo+Dw0I;Ft@dh6+exA^}igWoDvZ$iji*kmY9<@(x0qL4%mYzQdR-;hyBS?Y`~d
z<I=Z&J4#DDJU9uNUf<8YGsNp#ODa4n$6d!f*S@}?_QsHeo7$gf!Q`TfG)l;41Z-!(
zdch_^U(cWW+z_F{N4HJ*awE|p|CYZTB94g*^r&wmzX-MS6{X#N!#t{aKQ|I~FoD`i
zxMqEaT`h@)|6v~$MrK1oPV&%!NoEv8=E$UH56gqGB$#pqVE{vcr*k+BOdw9GYb4gl
z6Ic@vE;Wvpa8+Qvtxq$ymu$f-KgFxAjt%a%70N4HMB4-58h*HBax~6eUF~0pU%NjS
z)oXFzWbdwbVQe}4*m*^CQ`YNyJ0VAglIRc*JPXsnC`zCt$+U>$U)d9QQ<t&ZJ}g9X
z*Wlf#&iX*!&N&u(=u{YYe+u!T;^>uUlOp)VQRKhv|7_*@&gc~3gOq-zmBGQlR=bLk
z;+$`)J4*L=!IRyFUhBdhd(<3?6=5D{s;_O#5+-y&`8TY@ni-0QE+bAyF6pQ)wzT_d
zYuiX<?^D~Pz!;VIaDSs=e~~Uj#!Z9*gc^6C^&h_adYFb>@Zz@%6gWBGL!z8K5ppar
z%J(8;<f5uXa%|tvB_eJr4X`mU#RjN^<09R0^o(}vLaNotMmLZiA)$;Z4KfVgwzYrF
z9XA|PuCZ{wzv|rLyE||3{PvA{1A51e{lK0lCAP`gJaV}|WSo5OQo-DF$fxk_R9#s<
z3WGpn!~~&;6q(uRdh<FK_$=gEqk1Z7|3NRl8XVlB-Gc5O9)r+&i+s}hjcuTyW^{H~
zu;2mNjqG66QuyiaH0gQ(Z@Vt+p(Kr*hT&M)qxlF}{9XBdc!`rN_(s^X%WW@t(G|m2
z<QEDm5&TdXo9GT;S3g6Gx1ymzCWuM>yV8~oQsWvHAGWzMPvm7~PyZfDGNI8vzfGh9
z?FbGVnb*+#Yt+X-7v_@2F^E$T#O`Ax@eI;LnX4!wkyB&zZJ986z+V}1%3CCa%H42x
z(og;+x>+Ms?{NzLBX0bUX}z|1@5MmvRQd%^WVhLVXHR?4{<Y*<Gove9?tJCGlHi$=
zWrAG!<;%+FjTLT&qPy509#D{w^Kg+oHGJzp;e6U69hu2UQB~SOq(nLc<7mwUVnf5k
zIF;b1cX}}%cmllR9cSoh%@g#}Vn2eUWMyS1mfdePO?}(>v?M~wAqEM1$?Y`lRWavr
zTQzFYgLQy58AEyNsx-}2zoC2E7G=9WEkcoQZl$K0n$}LND6wemjJl?()o9);XY2-I
zkpVlbu#}GNi%Q<;Zbb0;xuw*Ep^u-llhC(?6Zpg1zx0Gj4i2tmX6i1M6$868E<f|S
z%O%D$`ivt~?#()Tlk9!5)}CYA^~MXag=g8Zq8Sd5Pca(2yPHb6t~i#XfP9UI$5)u(
zc>$wZg13`z4eaPI(nvqgizdU4!;EDQHXRlVRjj$hu6wfwyAv-Is@phfML|JhzHqOA
zcKsSqT!HrfW#ngvm!**bwbp7<h3w<UwID|pT+_5QgAh1SD>#&@U!VnlZCC}+1D;{e
z*n)nU4YVpgYL)+j*wZ3>Cwm4jLA5DpT$}W2^mlfwL+yFF15UsBGrFYv%gNP%nyiy)
zca-0dUljwT2cOKR(3Y+cJ1$mqvT1|LL+*0j=U=%FSFWz(MD5D<M`p8%_xYPD-Y2r$
zUU4zo_t8;vukJ*z+sp*rj?+pcosHFIQ@O)-iOezn^7HqV2q&i89w|Eu0xeQnZvH*N
zDc&1*f|#U)9b@L8lfAF0sBv#OscoMJ79+4izc?z#&eD9<3<Mr=YP)8Uw?Iv>?1llE
z=x$?!v@?F=4aXO9fI|!gSl$j{?+Lf;gR7mY1ReX-8!k<CCEnliPo^WKT$nd2P_(fQ
zY$!3Q3D{A|RkPiaxTe$U^>n@*&)ssF6_M-igJO(iHS>oTv09%*_xsVe;u#)q%WJ*k
zX|0a~mxWTAg2D{hq$+pIzi?`gK8~MRp%sqHx}wD6jwVt-k5mL*4rLAlrG8;Gd46q%
z9*-|PsdVb%6lcuqgU*_SmU&5>3k<ZoxP#C}l@JOgznM!WnR&oD;}d9X^zPw5U3WP#
z@)Oud<{BGl3Pqmk`5d69fwUKociH=(@K{vuwM4h;tX59qmGC?LhNIpIG8w%mYqr>C
zEj1;#><F}UI?7CUI>+C~&&MHpH`E%-oFD6V`kAJCSDsi_AZJERXGWnULK+)ja^fpl
ztpJk~mOa!NGh95jq0{P&&J!W+aBxm+)>xdD4>dX~u~E`GEqluW<5Ba;5IL`BN6mUX
z_X}vn?)R<9GuKvc7wtkg*f8pPo0~*eZM>jzW1+`wOCGzwMDzS1z*kE$5X)F4&m;?a
z$|gqMCAmpOQ{(BFIsyj2S-)kJ*7|(FHA}jzR(<I8-Gc}^mm7TFF>s?7DfMvP2`uO-
z{NNTmuV!Al)`|#vb9VqvuE33CQF$-XGzq-8<fzGv_5O~fpa_(9Ap-UJvT<#7*3;(k
z&U<ZZfh&{dAnKFu=7gsD;-cuYdPM4~yuFx+W;iEBUMKEBjxYEfPF&eFxSKWHf@X#|
z-s1`#(cWU9X+Q~xl4;JXdcVU7!?~!m0$Q`rF4Kf@__0<v*$b~%g2=E%nPGKta~vLj
z9bR*#Bt218<2Y`;d{G@oBlP=(&VFk1+5WmwL3)xY$y?q16Ge&=husD8D!oswmc7Fi
z{1tTNcS2~bM7-*KKUO*jt4M8fTb%BTe$Wt8LV!(cU-*{9vck?e4BMVDlo|jhX74?7
zEsyS)<xAIOK_ubGeodOlpK=>0ov)H*`F6sxa6GBTq%tH~L0Y3p6gX+-OXYC0F=^Ls
zD#*cGt{I155lma~y<Nnf^k?|Z&Q~Jslk?w)*aylRFs2c~>Zk*gE78CvFggRSyYy_P
zG?}J#-ig2}aJ>l#A`c92_AeO|rr3<JAjU+@v`N&g_&Oz2er@~H5#q1U2fFXTS|Khw
z+x4FNBbH~!b8{M<stZL~s$OMZX^h#l0*W04Ne{~c@I&tX*V6p>_eWs!I7*V~LI%Nn
z6R3h?N(#_GVsPXzK9~zKC&5VJfd+-*n48sdPE%k-0U~|plgXVk(ocyT@ea%1C_dxE
zzk%Qv`DFOkO1Wz{8csWV&qAb4_XSM-0ozawtzocaw?swKK=Q+XlIqj<`U2vNd8xc~
zXdLMFZEFeIIx(ZIwE|9}5@GFNsXbCDp!icy0ENHlqA6*Yu;4_~08EB_>8|^ENFwQ(
zHF&Ll^50UrdPy(zPTD&@3wC7_hs)_&dVK1BJHcqgrZH*i^}EY+aL<pq56gSJ_qN!*
zX8%}NG&=LM8}(t`{$i@;`BE5JLtakyyK+v^Kvh;#Pv9p(6E!T&5@jlcJ{dArU?vHJ
zr>#5EMwugSOr@5ihp0p;2{AP3;u*`)J`hWRIt8a=SG*!pFfPO8WxpBIhum|-YC&`u
zs`eGeZadpQADue+K91nj;(e~>2@L4vQgwTs<ZC?3R4RUb^-t;zJx!{2n@>+oSF>mx
z>G?j^`EnT+5^VK+#8<3atgsCF5g2@61v0!S+W~!##}w*II`|J+0>j6>gjpLfh6f6f
zrcX<<>I33<e+4tviK#}e6aSWUJ&?slBJKf1wGkZ1h2eNVB<(F_+sN@qBg~S&KvkQy
zP?4Y1U2ZnBf58OQWIlkB%!sO710ysKGGXexrUZ{9BpR3{lGxtTpPGshrXd90Rb@eu
z(9B-!WPab84TqV3amt>Vn6aky)+7qMw*BR`b~R?#p+uoyfP6p~&Qk8cZ9YXuI&zAN
z=glRvKp=4gy1uzI9qkX4@@25t0^wqd?d|2HPwt$UeNr)BeZlztnwn6{jku($?fpDD
zXGJwdw-<Y&e1<G7M)89p_8}J{KzYNh8yk=F+FYIBK8XkNvS4l^!3Ams@;v2Nke?P7
zXjvuF=-+)A{bQ~aPU8rM{joM{ABux1f@%_k#DzF2R1WZo$1Sj~uRvb{@SZUt7SfH;
z2NYQ5q@#`bVv()T@!gCqNAOXZdEtQLW#8_m`+0ee``nA6UhZ^ToE~YV%C0VK2!6X>
zo|Zmu9`q8IWO;?yVSeR&46gW??(~#;Z`^a)YB!%IoGIe+)d60D=zYP{`W$=`oQ(U0
z)S?L2k{m-{F&D&mu>qumY;EhZJmW*o(eEue8x>Z<8h;aS`AZ8u`P2!MixAR9{=JHq
z8|rtz#aWA>P!EUgL;Op#msRNt(9=&z1%#To)54f9VaN(s{6%f-xR3NYO~2L9gWU<2
z7(G)dT+`Q2Rvy?4oEthl$Q$;?U(%chH5#85n1hy*?Ymtv8d0Ouye~aI=5|)O-eY=O
zipA5IKh6-t_&R!icdKnbeA<7W(#)aUVli0GhOz64hR1i`dMRu5J>L}bMayh-eJO)W
z#W-4u%|gTT+4%6S#mi_lRoL&UHZf^+SYu*KNn(*n+i(WC8$Y=x2&z?&J4l8t^4Fhr
z76~sW-J6)$4N&vO^8mvvhBOU?nFNkjYX~@T3;{$bkc@XBc#zIOlxmOik``Fow)SZB
zdKhW&gbRi$ea2dDGQOcCvni*%rm2=uJ9+kwlXu$c^`!UijroafrzIF;RhUuVlk6ul
z?qXixES!E&&z{=e1S^eeJBAo@ojD^g*@&|f()+*zlb5e`XESr{;MYt1ew`bWQ`vB;
zw%i5Jow&epui<wT_aid-W6Z`&%SIAzhQj-(!7I=}^lqOQ)Q-sHdFLOI1t~dvFb&MG
z6%BM?X(tMy6(-eE+^A3lL#p-1$qXIphU?{XP&}+M#`C6E!&_&A*A||qEyH%T*857K
zXR~%4F>>_%!}dy!`YNefQ;D8VKjhS2*F1n6RgZq)L-d=7D%Zy5lSR+S=JVjw>-rFO
zW6vXT&tqfpSmd1I>o|RK%64O91?)+wG_>QH%1UjZk)O;-i=R9-o<3dZ9}ByzK=yD@
zF?%C1Vflx3fhMUAnP>tiod*2KQXnq>{m=%{%D*aVh?sq;fc=W0eb!cTIzIIQQzHw+
z6U$#Ym`44aD*eC+f#!E@>8bp4d=UjYdg}&V7%>?)Cj^d?;E%M~@k*-N3f#Uo(%-wj
zOCm$@)+;)E72{Y7<gAv;RWq1dE(q<qtIAyUHRG8`_Qz_+a5d1zDO=11&nC^UvR<OH
zuIY9MSxz*U`#?@|3b0K4vd_c=W0c4%`OsB$@!PYt<`hV(vABe5c2@D)qtqaeH1o4@
zS#Sg%dAMTqRt=ibbIOgYnVUNsM2{G=h7xH@;s@d?MsV)0-?qlsR0zzfUlJbn)2;KQ
z<N`-JjF`mOQyyml74;GCk<%kC@M^9iZM5T%Cy+tX)zM=hIyaQVMfryP(p^JKC$6Sk
z#oDZ`*+cPl)8~stnpg=4WzZMrk$2!KgDL}P*+2)mN)WoG2)>Hi<H__Giz+e>Pv0T5
z@1O8z-i4?0Xd5FkT^x3EM~|^J?vEwqkyjb}So9W``6hC{F6XX?nWd@td3`8s+gF`l
z_d??sGjw}xhicN9ocVS2wo5_}wY`OGA3e9n5e?@lY_>QIu7hXC*ScO$DH+&94pR}a
zx@#}TI*DrCMISp}Im)_xstlaY`&Nss274V7Cszf!y}=Ikhy@$q9Yc^}$t1F~T*t%}
z;fC}`%!+X4MVdkl(Wf{lvxC5>Vg4NWyPsVXGg>DEdL`FGYOUSf9qGKcIXPLx&Rs-v
zbnP0_&wrU09z)F7!%|Uyt)Q3|-_um^QUSCE&y8a5{uM%}A4~!Er!i6(l{2@XcO2tr
z9%|5|vc@^|>!rvrsqI3r-Jim5RDR1F4eX|ocKa#Mk$ws4KmMv-XlnA~<9XO4NYU|{
zxXkqfL+?yI;_;L_$H=_iX;Xa_mBaTlf7&|y<5lBz@#<5nyU^ygrowb=!Fc!~)`7$C
zXPf{;a%HqWP%xzp!^$1$aqMbtz?vWlNdDD6_XVUjd?I+Ymz_WuV#^iqv?mNY(hmZR
zgBpm?B(z6;L)1{~WoMl`#+QAcB6w<d%+1H`c41ee@+Cdu*myGD(@k)uRO$WtiZMPi
zP#`wI;9^VYqX+Vq5eC_zQ(VI#p@U=+kr=-SE6faY(#y!$Q5RpJAst^87dBhhAl}wi
z)50$PyztmW(5<>wp2hyTtXC%V8vD&&O+Ick9y)H9EcWz0Lhkn(Dv9CF>j$1=_osd9
z^pq1*t9xA3(>jc13DLPW{oJM{BS`<M!nfr}TgZsp_p&(`p~Yr5%SO-9>e0B@alqH)
z_3P=F7ifkhQQ*vJBiCZS(JaG#Df{wH+Wxs2LX(Tfp6}XbLjvv3LL!Ld{(ucZYwV{l
z<~yz<Q@EKM{(JC7VsuJSrkYFcICz4NiEeH;>D9xYQ?{);(HU<ko^jKk8BvG{^1%EE
z)?fe`g{bisQN|QLx%$VXJGfDk*>rfs7iVvpVa}l+76y*r+3#WQ4Ldbg2drqnUfMC<
zPvUn=h>8m*sA`Y4OX*$#Y&;k8)fc!$5`TqcKmu>VE=^QI7b%M;J4Sn*G1@G!NbeHl
zWD+i$G^@xg57INiVGfUDkxgArQ}>WcHnFzOhDCA;#GXwcK4}o2&1O?rz~sQzO|jh8
z?Jg23<=M_rz}Fx{tI<XOa1|~l1o68U#67e@p2|viYx!5GLXvssKE$(-PHL*pMh58$
zuE*74P&a~1uMbmigUfy-fo0R*mNoHcQ5dBNbmfN1dQefoT6|M)K2&|$`Xc18DhRWG
z_c8RgRbty1L6Z-lYq15RDnQ1jvVFM4M`Bwf<2+GO?Wmb!02uf;-aqbrRb2#aBWv<H
z2-_Y=amQVM$wU|X%)+Xv{bp92Cf4)yjjCZjiZW^Jeq!PbWbvaY^oww7VX^cDLV^k(
z!!KD-qf#U_|4`Z6^T6DBu9Ffz6>-+$Q;cy-h^9TcjxHx*HHR9JRF^B8i<Q^j)0jpM
zsO=i04Z->K0vQpmM4>M9$N(%zcs|~24;@q<@$DjoBqBBihi{3io4wKvedB4LLMESQ
zw(R#!HtDCnuEm)@9_c_#PLSbR@d>oMkj(T4)RCNcd+ESY>3JD&qc!Xh-N4+V<DOYy
z8M>^Ns$Ax-aly8-5H)xhLmZ-KmJ)eklb*pTJQ7Z)fnrH+xn1{>1JTHbeqJGGNlv{y
zAizeKjm$(;&y;pNm0kF~XQS=ixs-kd90Y!%PDye~j@aP1%+)C63$*pKIBaXm;U<5`
zI@J?tW{s`X8|ofV6IX2y*rUvglCq}vmrZ7=vBB6RZ-UA>I-|4Hc)vIDQQqNyi|KEB
zKTK{A++{H)<5cgrZVI|YnU2G}Rr51pnD2*Y5F9XBF;7^j_1T*HXBLxSPRU=Ic8Zua
zPUjd1*np!<fDhiA&?oYbqjbj2ma!=c#Tt8`DtPXkwCtdnRx~YkFTv5wP>>IIKLKdw
z2?hAnzQpenKo1pN|3tfwPR2B<JOL!CT}2hChC8~-)2$G!G%!RJb5m5~X(&t{3yeeX
z*Q!V1>B~uD(@f<cJ=4HMlUL1UFG|n99@q{+I6vVIo1eEkF5`u`a+VWUVt(sQTfQf?
zaH5j8j^*if8lEzyVH7s8o5Xbq|Lab0oCueh5{^&{DT-=zgrAJU6=3Br9M?BhE<d`!
z*i*90=Q#gsUH4j~z5Atlmm5qsg<iRamINJq)DPTCg&F$%6s3nt+YiBE`lM9__D7So
zc7V<Fn^7uam?UU*l8cEUoG6*Yq!r>AY)1+$Ki!aN>LzKAg?yda!k~#Ur3Iuh^_1j+
z-km^d11v&xNS3?NbY@v7$W(`o`C-WEnvYjs+3+spkz)y^RF_kKIxs(Zr~7T_p<CwX
z&zF;Hu{yK<;v+G>ugfdQ9`A~_!n(Xh$~CGaL`=8uYT-A8i8fZ$BZ+D06Y1_|(U2es
zZ<H){gfHf7+xFkPr@Y;La2|YYdQ&^up*{v(y^=~raQwL`cKh8;pe<N3q2~bKoeJ_z
z`5O!c5&diI=>b2N8?biQo%mdyj`LZh!w<e5<>|TSmxSeBsTsD32fr@@w>~B9k(mdC
zngv3)J&cvpvd(WVZ&wc~pCUYG_KTz5s~+e3xjmZ8#(jg(?@ruepNE&T*XjYUyTS=_
ze09y<57E~s=ux*Dxz?Pamk(?`l$dt?OF~{le~Hn)EEENd1^z7CLN`)WzO;{H*^`iL
zqYo}b+gmj<7{8h|5mPIYe%obfkL>)_R7S8e2y1}KDCVW~YhfIR%EOY-q#0SDG1uwe
zDiF6j;ejbV)?JP6LA8k_raq2P>#=uLIt{rTc9-2qy?Q<dwF%WoiH5%{-C#ARvUeoY
zr!*Ay_{Z+q+t&5#^S{A%*D9XDm5|>DJHT8Q{*IJqs&<Q>m$|FAf0x|_-aQzc9HerR
zd*x@}!auOUn772NRTf_==1Q%bw;(c!n!s0MW?QD0&(QA+HJs2T6WT4*JD=f<^S=#u
zj{6=G+JdyI*3b5~+E{6gajf?YYm;&p2D~b>RNyV}(wd_G65ALHM30J(hKEg5z#&^E
zg*F`iM9vl-$yE;VcUX2!%iLVAI4LVww@$-w&s~#P%J?)pV1b*gl!dvM<}YMCg_Vsi
z^900mLWB2)zVQ%VUA9p4_EtxDIe~m{f5}v%W#~DphdW%K9tTyy)|;nn&jwluSnAJF
z8w1h5{qszqwl~C*Q{lI?oi(Uwofm(mWH4G9w%geeKY2wN7p$pLvW=}smQPA|%y#rD
zJ?rf#_>XAk>FM0WkNU17dvi(jrIWe<`picdNbvDb=g&j%7GJjT0YFRueDb7Ml`cSX
z&w_G2gnikXdgEbQKP5eu5m#TPw0_%=o&Ih^cnIC?5lBHf()g3EQcZ6JCB;IxRliIf
zy7m{yciwD-*UB@QtGY7pJn2!ty2)_+e&IgzWktP+n9E|~Tx%6bYNR4Y8bFab^j**5
z$6__&YA&YS3-MKw#Xq7LExrV%-gHqd?NJgGU#lH(X|`=Mf348M;N=Lpidwll<~tAL
z?e89-s-a*wI4ZFz3l+IV(sl>Tl_r2oHXUgF61hpE82>9(f%U-_5m?<^AsA&iN>wg>
z)w=E&Ew97Ps4H@m6pRR8{IK8x97m4T{zXeVdQ}7+J_4Hbw46D2W&bG;oh1-Y9-Y3(
zlI{%G84{zzbEqyC{uaUas=};PS7BEFwhcZdBW0^|PyS<kVtLa0el)d-mdk7?t7i#8
z86k`DG_9|!>g>2TQmfP8xkMm!M#ZsKaUXvp|C``+B_8$-K*-DR+psI!umdPZV<qQk
zf-BeU{z<6STG#LHEOWclX!`@|{OV=w^L&MtV%Bcd!s|VRz{h>1MUEAn+vjwG3*Xc7
z?R$r_v(WdbM~&F6@=vF)uKfN}kv^Sb{!^Yf7r|2*lh<<6I_@8bY#{GTLO7DE{?~)Q
z>M(bWV+jq1)fJC18saR{R>=E<zXlie!*gB<K`-YOO;bFZ5vke2HMd`0V*{yLce;=F
zHr({|?9c0;drIv5_<8o-VtIP5_sB=Cu%twW^;Tm*{^|j5L8<0rZ-UXnEapoIsrZ7!
zIMDFwBkxxU{Z7T(s{c9y8#Q^5IiU6=I_`W-44Aq44Cwtbd!6t;omg+P1^RMK)NV){
z%pNs(A_+Cg)I`Q_L@D&Z=b_iiIg1Q3GF2FW0IwIB9=Qns(>B<?ddN#-4s$>u1D*45
z2Ihih7^9ZNqK<(0Qfe7>W-}6xixTaIVA_i!2$w?1N-R4~9?5ZNwV5O6`7B*I-g#Tp
z8fSBOn=7zN-4T-W6Ta8=y|^RfyjO*KU62qvVMJ%!s-4E5Cut9QyBri7E@KIIK=S&>
z+A2;Gh0opGJjk&K?>Z}r-b^+;0iDe~OGg-WNH6T_rSYQB5~&%s!xHE8HozNK>azNm
zTiQ23gE-C2!BdmTC1-=F)Aa>W(?EWkZ}6C=LQ~{e&B2In|F@}5qk-9ZY8POAD*xmK
z!+)7+l*p0TBc-ln2yeCLT%cGEVJ!NFfsX`3LzCXC4HWI8BCd#!N@KvWXO>6@C*=;z
zf(-l{stNrfJ>69mw+$DfZ+o75q;nHBsWglx0W-xeF!9xsCXj>%BbKXyOvWy&7ZGCY
zJPAxSjI&q4dAKw#oq;B(6LCT)tAUq*H#Mw*5`wqZja5rWmt0kov!oHJ`t(xC3g$eq
z-9~Rwt;7^FVjLGW(dx6Gy{F@Wo2~OmvJ%mrx})MWVy}?w;|zms=jD3Xh=bJbx)u*3
z42@NHZi-s<IGK}g-6MVE+argq=e6SQy*2msXjjf|!}+4+y;K&H!{#IU-c-BQc`xda
zCUxhaN^jQ9R`up<*lVZi8SlZ2uD17`CAYeVzIV0tDP_x!yUy<6LPRa9ip^u#$uGCr
z@iD8i#?G(>DvhyI;HtPlIe`OtWj~JNMcH}W!c0NG^lz1ZW41`Cab%)~wD2-lQ%7)E
zMQS(xF76UPVk<!hanEbM&_XSCce^4VTj}Kio|tP3i-VC})&kustNhYsWgcPTL*!ps
zBLxdpHluvZ4ES;pOIB0RGd!9XOV(4$8-QsFL#F}D-Xb$5hxGB%gX{9LQD9?OXi3#s
zBRjfQmY?|Z=10EJF#=a>IA6vBw#f|lcX9t7)1%hoGY}Z~EGI%FpbHrc<e&F8RBK;a
zSEAF~s8FMtS-lr!rPrrB#HZf_fg6Kl;r<91+Sv0Tf@c_^^Bf%pmfVQ|ic4W6Em4a-
zkWO+<wdbrh29jM{+RcDS=&v4f*g1h`7v<p7_%K-r+~$K1zL930!Oky^?N77U7Bkt)
z$vhqQCMr2y(Ap>Oe}NZoa+n%CjYrS7#^fazV~FPrM9X`18`qbbwf7jj#c&+k592tg
zm^lW9I?a(Sjc+S1gQ6Ad`@&@-^NEII!a2M=@fbF82R3{~g{|Gy+OmgkNtrHrHJU5)
zh2w}r5<Bvqr8z=IH@Ns0V@Vq`itc%_H_6f_qI<UXq1Vf$U$TDS)`h@~a*;Uf-bgYF
zjw#~8(}!9lYV7GTl=3XL0l&;DBuvIiOG0p6SvV3k{wCF`LdOS);yzHZ(PX1P=KX#u
zJ~uKt17NqF?ha~!c;&pz1%;MOYPX_pU!}pkIefl37(L-`ZMDSUxGV%(8Rb>%HP+K)
zG@#`n&@w6)){`Q$<rN$essGDgdYzzczNKveg`cuey|Z(vFGG)q!@4{Q6r%x_YQ+Hm
zHvZT`AjshGI~kTVpV0{ts{|jYSo`#12oh*M<vWsds|FD>^qkz$f|$-!a@iegX40?Z
z;=z1Bo<HXzSNM9=^DS)ZdDA@a>&|4o_BLNQ(s{T<C|7&^Cr3b$t8kp_*86gi1EI%R
zK17iO_(qx)dn8-}5+9;yFqadEC}v6!fF{k`$#4onFB(<+-KN7gNcp8UNa=s9Z;kMG
zwc#ZZogfT&;|UR8y~9`NH+P-oZ4!&7`!B>7TTTpvxqxHSPx^}Q#JPaITGB(tXEjVU
zh;U1dw<7+=yRqas3TFj3$SUz%b7W!0yc3&LL+Y1c;40TkG#F8^(AZ8Vq|w?S6tX{0
z;Ckj8K^~d+1okvHRftD_Dic`VbdxmOO<kJvV1)VDTfr=+MCt0mR)7toJ}0haqgC$v
z;V0Kdf%de@4@kT%TDa`%?fA`U;gi||9o*$6U<O1|FkF*XF9QA<>#~31Ou%5YEq2u@
zSTB1xRn7)n9VRg~s$2LaXJAaDWtLM_rk-g_rJu5VEp;?l_E?#wsYQN5fWLnk4+dKW
z8-DxGC#)t><|jULJDSly@qf=PG)91n1b&vtRC^rujX65Ic}#_Y>sj=SST#ZxgW%xU
z`k1d?_0Na!uJvu=_We$R%iuBJ@55%bjkzaqICnmmKwoaHRKUw4RVpPzhY3HI89S&m
zwnT`Or^MMOY*6thq~WHL)Iice7%8dC3L40aI!P1eAf_q-7bOzyotx?8v89PpQpK*W
zcH)5ba(;(v8qDiTonx-3*Pml?gKrP!eCZO&U%0-oeor`F0rS;n_?WLILoriYc@EfK
zpC$m;!+9wEYsieW=3mU8bUI(u31<pQTL65m${4B;$>6pz-Ix;ag}VDUKt_tJwl5ed
z8nc{w%h4R$b3GK^2B<bO#U5wl--(6OOsnkZTEk(OW#q{;)d#q_m?nP0Q&OlhiZ+l=
zU`H8Y&F1GXzqaZquT~f2${EWgwtgS$1AH0;3qud_s5kj-#PFt4!mUdSaXBw+$E8%5
zP6?ri-0vDIY{uMTLJ#0!WPAI-{X~u72h>98gh$32armKci@%Z}9^q=+TaPQm<VkX_
z-;euSvjIrj1vaQ|1unqP&KroyWPfo!)K8>FxQV;%a~MLgVpI_|f|B+4?x)ie$GMzu
zm}m$V5Kb%(DYEkhlejyS*JMv#OOvTZoyegQ+MuqjqQwoxWdI$5mhA|16-sbvu??X&
z3b@+=$}*YQb8(o+IqS1#)-HO7J)mgilr`2^>@NX_Ovwresxq3***krrW)JzfcB=1a
z!M?A%?H27~RbTmM{gsj9Jm4ID8@*1wR`34zJ5$v=-Obe&t*&;Yqg$mrf)%IQn10e`
znvXP!4&v~C#azToR$Jx8p>vWSyZh<;g3;HKao|Us0@+!d>M)rRZD}W={{7yKLNBK#
z!I|hor=2v9{{#FIiEVScTSOOA4mEsdN!hCT8{+^t&aXXiQ|jeuu>8Ddb9{AFou<8l
zwD1?qN2|Hd*{hvgm;8qZ=E~T>scDeHb-&2>RKczzSZ4N%Z^3u@%-5)|{4eHXv;!hk
z;GTWS_lP^)U-ElAGI{+X+G@ENkpnXGDVqbNZ9L=a0`T2^kV=bx<e?7Hx;bmIOGorg
z4-eIvVDyL}A_9&`)i>yllL8F$W!GRD#2UgIpw!aUn-OU$MGW^w%R>MB;-L0=qDMi2
z@8(6Qd>RQXZpbrO7|+Py^Z6Ln`Pt+7Y`y7@;Jx`cn)|J-)aZ!f;}+;KnAE{VfZO8H
z{RqBs)7OnI-O=;&o{zG<=V4o{&SGx$UNuQcKGJH1^}3nvpb}MgsFI_AIYmrCd5TY(
z*tapDv)QAB%tvu+s$D4}0esY>y%Z+IkU1dSg<@b9GLoRYP(Y9U86{A$4B-J0?F3yR
zOa=DCl){V&!FUM@&U{5Cz6pgV#QVLAv37W~U>VI&^YPCaxkg~TW*GWTd!n^1LaYg3
zg!}!EcNjCo(ik`BS8BhAK0cndx50v$D(e+^zJ}c_A-VNORYdj}V#Xg>B==ux=F?+S
zAyRbG9~Vfj<;v!?R5TDZ{scu1U?0$YyNc@)ZB+XZnX-1Hh=UhdoAV8K@LjW27>?uX
zul~xw1fUve>*_%Jx+h6ftg2~bTyX?lm<<qut&IXY@BS<0IpB2yK+w>z^2BbSPdhqF
zPREz6I~qz(&tGz1s8>Fpra=bxkKF@)U+QaCSS-xm8&6VpT^kpy+fMs&UTn*{jQy@b
zDtIqr#8W6_>SEdf-VKMUOv^HJ=Pw$YoOkr$NEVFd&_Yt1%vho0%s?S$l5wV(b2|hH
z1OoAf8Bs7&r3ME&N41NC5il~v-tf#8y+{<(_FKUW?x(yXQCNLz10%+d7@n+#orCI}
z5Oka$OyqWJGfN9J3(>0&@5MXOW^_4)GMUS&ODfeHZ7mHoj@col_t0CFGe+*5#NnF1
zY?p7GIq52FTd>L~(_#vY3Mo<X8bb>9XNgwFmO{Fy{;dZ@+7Ca3tE^0wFHGrj+iLr4
zoj4);sSnK&ZcLxQT<SxZBmftbrNkXXjJhgCX5bm})=xt^k<Wl`+yDL>2gWLz;veJD
zUrpQO0K*=Sl+(WyqrMOm`mcQ0;sD9mt*@0m`2j31o=328I!JJdh8brI95ILW$=2+e
z)_Y{f{g#a!oc41B99$zI@@Ko4#|u&@-KA3q$pd&Rxmq?}?0FPx=R{^^Y7E!BZT?}n
zgW=e$;$PFHNINmqy*XtVMWbJu5wzViD#oj#aAw@;YRA+$@(d2+Q3SQqTlVUsR@Ce_
zoB8LaGi}ELZ%+he&*2zZ@dPVe>%%cy3{RhaF!BTC8^zwTocZI8_t*CkJ^soGUQ5jr
ziu6S9Q#9&z1^=dZkxhf@Cv6cVdFd~$`S<Bj98wyoKT`-&+-;DU5|C5f(SiU!%LFg^
zTh+k>z)7l+te0v@Zn8dl<S$D}^@h3K5oxi!HsE_@*{ZA-NJ|A#aIHo-!#rwvbjl@I
z2Ntv|kOj2HI7|2h5hB>(jK9?jurruBOdq$a0o7=&+rA4jrX5>(Rm&XUe2dqmhvzi0
zwOnbK_rse7CREt98XeB-Tm-UP46i*Un`S9BSoFN;csIJ+U;AFEqH*6T3-uqYTSN&$
z<D)X5slU+u?dhP#IJ-OQ7|*SV)T;5uO{}O;RGVEPS5%Ko1!BuaQiFJCkTW@gc87&3
z8`sQR%<8~2!RR-q;QKCzU`P!c*J{4BkFCkF`ig1<qTd!_vynQ_nEpXmGa6wXT^IB;
zxliA%$rK*xa(#phFR7^DnAV2wOR2HcRFq>mO)ZYWV94fpMo%+)$&@`mMpuqZo8!Xu
zLaOvVGThhTJ3ZVbQ_<zN8G;S)zGNY0%gBYLN|QQ>X?Cj%BH|6kP%>-O83?%bUmB1<
zAPI-ZNBxEEi92AY;ur{X3Dkuce!hq?)=3v*>hv(Z>09+sUWD>PjjCtteMimaAK|O6
zDq;v~2%To3;6nJ#1?ODTFv4%-Iqw=0*<_>BVpo30W~tR-myX@Zp8o;m?B-wXF-fd>
zi$Z$@IHbW^6{{elCc2Z3-Ej!LBfm1=0R7VSR=ZYm^JVel*At{AOkEfWQ$|X#`n%*9
zJiA{o4@#y64ns5b);EZn^qaT=GG<9%03S@}Vc=Eqn?n8o>Bf$jupQNVBRs>->uoW)
z!j6kS?ky`RRI@iZB$xX(`Fzp7q-O+bJ}r@6peOlGHjsJ@WbU8Q_5W?4Y<$Lj%89(T
zZ#-J{=e=pB^SYkd#<|hK{>_Yl!-)$&3^ipm5)H%vs<)ub_5mH;@x@SXCmk{L??(P(
z8Zik`l9G$;&|eHUa)5vd4}p8LwRa>lJvlpbg?yP_|D<luXgH?w>Wbjg%y+WODgLmn
zQolxQ&<I{hWGd{I9YiwQp|UtFJR}2pH&ar((^uM#rdqK!letkr%Gk|unxolqqpW(A
zF0$qQPW>2fh+v6{9#d#5!$L}S+XvwD6vD9JC>gr4f*GaiGjIq44vL|(_HuSLHc#5;
zix3cdO?}SNR<l`#!~T}x+s8BK5!n)~x}?kF=Cnj+`+?Sq5vA^MJ}}^lmh^PJPc<=y
zc9Ex1o`!S9)b6@&^MxeEieE#yqoT`X=wC`QbBt1j>tgy?lg(b;CikazoXyZxVn^Sg
z*eju9Sa~J$U46hD8l`Fhv^q3056^(b>_WzHJ^c4-0$k)bo&<fib5jwr^h3Xa!s&^D
z%j3XL({P0lA=LPs&bNp<T~7@YmgjUhI%;f|$~vDWsrBQ%bH{wMcEH_q{npTh*H#c%
zaX0JYIkcE7`<CO{u32T}P3M;a?Ze9nH>nLmy)rtFFfvZ%mKYyN-f<CVXB|7xKm&_o
zq&001Y%n2D4xM7T=LpJ`R=!kL`gN5CA43ABjSI=?$#M}Iqr=xp!0Qno;_Al_<`bG(
zfl6xX7pQIj1hc7h*xZ-wMNdyeZtYrdOI@a>#!vUPdcx@}-Gfc%6`0TC#ggQb_s4r_
zJoii8m{p~<+PwEm_++O#9MAh!+MBg{I9e$3vwT0=?mP_@{^4kX@qHYW@)fV$gmOS2
z{052XT62@Qs9|CIewkmYUs+32nNy{Oe`Cr-P1ItJoijJ@Gv|TT?fIVolss$0>J;!u
zp%NycEZ{PF*&+$FM)<HSfS@YH1)=DynxVsD+-S>06F_jvE-%i6{7jfDPLEpbK?K!0
z;4rc!n={a-YM&HpEWS)lkhsxkXs=zY$Ku7R*FU^$$nqO{-?{6?dnRqG$l~U%YFxc>
zXl-Ox-rg<TTQ}Tm4h<PGeZ@6v?o{5rc9B0caPdt`qtVj57mTWHRzV`VrTkWqx*2R`
zx{rFOG5J|R6+ziZn*=|XnW3zpcV*BGPtd+Al@D~v2P<gZNjkboc~hsn*+f))qrdWd
zxdTK`YLctglgLIov$~E<2Oe1l`1U-W)sfUXGDI&z+#<gI9?p5MGMX0sOPE5s4*cH>
zX28jwXF!EI0}wH-q7vIxkq@3iX*f|3W)}*2nh!zMW~c~DL4nyv0pNr{m)Q}_hvClk
zu!Zr>c71sX&Vzt6?nOtM;lPvvQ&b>enhO|3sssaql$7JI!p^ueD-er)-;7b07aLsx
z-ss^OmP3$!?~Hkh8aHm=8sENm2iJ6P!lb6}oiD9_;NHGz@t&yiN6MMMrE@v{)<OMn
zy`OF@7uTI&>nGVb*;??Gf?RSCQ~NMQ{l0Z3&Tr9;UA!3hPvX||V>P^_5t@JE0W-uL
zQqk!$JE7A#Q|NTvVkUKqGncZhq$hUleBz0n_dj`E)BJ@^O$!&S;osQy!VBB(Jay{M
z#@laioV#%&&WYI|K=<<-p+6bqYAqs#2n^62Yed$=kvSaKvBf+(hdG!A%}FJJnNDD^
z*fY>{irXD{&5jXXOE~VZ<2PViwoX-4)36AX(gD1SuQjxuY=`jLzMa0kofaq`ZC5^`
z1wij87vg5=hXAz)6L6K}e6-_2vWWC5LeL)B?QMy>S!~%|2y-leu5Nx|5{pQ-Gguid
zKs&&o37K}R3xm-Rjz)a`X!%WjFE6QZjK5^g?5gUmBiG#BYwqefW*z_X+_}28_qn}A
z3&NwSV&R<Fu<;8@!|N8*&b;=T%Z)sxO9^Xd9yb@-SxT0Wcs|AxZIOhTO*)LgffPks
zJPVs$#@0*uy@>H5<)ZtM0mm63ITw18k&8Xaut87CSqQRN?TJ~GsUI<w@;hNFxuVH2
zH*|xkWK6M!Z`rkD7F<0oFn!Cm;X}fUD|&~w3|)Qi+pBIlcKe`>+=NAuOZrELaF)ir
zk$tP`%8RZks+>FRj{3#p7H`PM{<Z2Vxc#hVK82;%ea$>8*r%=Uo>AV`RnV2JTps>@
ztrFlj@`p(lu2V;^n7PBaNL-D!XGGwxEE%9V!(15sVVDAFL5TJiH637bYLY{XwqxXC
z0FMic19i-q64xH89yv6ST{EcKf9O!r%3*hpRRa1=ZHF7C_Hr7tg5&DAsjZWz_itE*
zE21fxaegg77<`A9$uxCFO3y_SFEXIG_=`+(zjzXP{T0^F=hy-T={Z4u9)|4s5F90l
zdJbg4t1wBQC#i%s{~`?CDK<rmBSC$?V`_S`bOyD4@UJEhIZAs<5O!C#)qUcxoBFA4
z3r;#F@GN(YtZn(4U;EUn->v0#XS|_~ZzCN`=W~;Y1=m!VSe>A1_gTAQ88bACq~z)d
zc0=Dv0hdssw6$=X$as)UMeLpoehh?;?JNUg!DXZdo9e06^!7y~jM?>NY!Voizpwt|
zs*Q?dTXJnv)BdmDyK>Bv!D|ius(Y8a@!l)B5IwDL0R26QNP5t*pb-fZbA)R8TLXsk
zrUSz{DxZI#12A0S1%Tm<E9umYZ|J@p*ZMWLHl+G&gQ>#jfex^Mw>m`A0d76h0S@%g
zeXO+73J<la3uUpV=#z@p6l1v^7|5drrYoI7dw42M{15)sf&H(%x$og2YZotBbIal-
ztGRdS!aw}}c>I03z3HI`*53S+-D-QlCfby=iK4WQM!^^=(LM*$ZOTdqm)Z7jTm@}X
zHVTN%aFm90G}NRdPAZ8l)UTudmCDOcD6g>PbY!LSp8gP{E21I!$Z^5|0wlr!L^==|
zt5Y%}Oat1(aW{iOb2~xDc-#!o*^1+k*>N0j5fjP;Z#W^U=!P~{L4DC<BRe@nf3PA#
zk0gqc5w)pXT|})HVLW`Q1;WP91(K{pMSb%G>ip4&$VN1z8@T@5^PGjwdtjBWX|=Lt
zwUQj<@9x)s`JMAa`hS(Y`OiN;pUy@91ZlETqe+lFaRG|-3d!X_4o^tVv8fA_yXm}~
z4|*csiGP&e3-m<QKimDsjh4)ulG3sunqB!K%hZ+k0<xGb$Ws{K_F*}84yKiXDl*S?
zqR9pHD5H>|8?^r1Np46dow~YRr_KYN+L35|E}&p_B+B>-^=mIuNBWTGNW7du{+4Gs
z&@^i4taR%_QcXmz2WnP_<8}2&t-1&>SA88G)>oR7dYPfGU5rB3Alam7I1yHPuPUrU
zviX5_!OD81U09ULft<zlz*2p{*<o-ZW@jhmebH*WD+gX!kq_?5nitmXt?zVZSdNT6
z{}H!TpT{V@2<VK8Isl5g8iiFY3iW!XrVD{eg$yV-(OZSg>Mclaq@NA0*g|NU$x#fg
zSBlQ#228nB*CNmL7Ue*B5|uhm#FAJF-;V*_QjC5u+$@JqT|MMMVx&dl$L5q?(z~=c
zKjLu)Dgt9Ky(%lv5Gv8YvU<-PHu=#%yoFHrWBT@?!P<$n)!qz;E!!LL=3i1(lvQ@q
zi1JH<<z7b;FH3mH^#cE_8q8NQ`xrKp;cHd{??Akhj_6FbBsvr5H_R`nwJ6l48Ja18
zCS1DM0i>zq%)3nwr2#b=zE-xiE=91W^k-VSd#0)TXLbdJvQk%sb^?YPfd*MRIu!g`
zHO`XHRhSNNg8$g*xkBo>CUgqg)8iBiADn5A)|P5oX@@S4{XFq`(`(#mZ(6tY>ynE&
zS?7~&*a>YoOny!-@7#u;rP^>+h~WkbSK+i2u7qhhH-zA%R5}`xf5DpdGicV&W3q4+
zV%*K1&zLJko%Vd!$h3S{M<lJ<RH{no*$hh=U7Zc9E-5RPu`)Sk&In%pUzYw_l}^n2
zHr-NHpG<|dUdf=pKxV2m-2^msfwfUcuIo%ycZ#ZFNXn1MR-Kw4@uOpvaHz35hk!A<
zL@xV=83Wk?EElH+w+f&HUWk<IjB|Pq3w=^lrrvalr#YxD_En4Sr4f8^GSIgix;1$y
z#`=b)W{63P`L*T%*FK#J{3Us267Z~t6p*=E?;Na-!v&qn4<(E?G6da}LJY?wd}>I;
zCm_6b!jj6Y$v&NkVE9;5WuzWE$|wtBZGb1~5}gU1oqDoZW+Oti<wsth*I-P7uyM@r
zLJ@AEFT@7lv0~*Wj=Q8keKs}y?j1k)(g_nT9juGzEf_d!{jFCGSeSQiY-%pfT`_+A
zim|v3=C6bp5CaKk;+1L@k3=FJK)obqIN=a%ICKsM@nuUve7Vp##-&(E#4TG_lrOis
zB$~FPe8G#Seo^4w<2SXf=iF_dbL(2)RxUk3eY7!!`$fkpar#$onTE;L*23`E%5a@h
zI}DqO+^OMjA;fF8SA%%9i_Sr23om$mxp~DUEO5$Ol1b?9Fxd<lG~J#ri<XjZA=>VH
zZ5i76_9cs(n#9fLm6y1+G}IZcZH9eGngcL-lb0?`;kWT;K;~M3;u9{6pavqBxgQq>
zQA3;EjPX#m1xq@Z8VbiWK()4l*WX=}qAu!z&g~iXJ?*hwWP3Wd1GGb0I}}b;4O&A$
zGOyIuxT|eZpe-(P#rL)aZ24rt`30Ke>6tt(b)A0)5`Ctsj`Y?g7yt8G`!=z7nx8a%
zPYXREgt5*55tkiiv7-w%l-0l?0H{-yf@W6vpU6|CASQ~P2%DARiFN5mx2cTuUU}Qo
z8c9)?<N{OI{04Xa5Ms6gPw)&mA+E#diJoORLo9iyJBt`m8j^%;)Uk4YhU@?Y6(izQ
zusN!ZAcuXoUPW``5C@K%q204r8XlIU3nhUn5C{%$$g?6lqt?-g2v!X}cw|q~phz+d
zcOLQ9M=(3Wnnbsco0+Q<V#SO$7O%6q+7}AOsotPM(DK^oV^n8|W-u&0(mveDmcx*r
zPr!4=vT@HKSs1smB{!PEypvDOJ2}HFnq*^Wmd%C?n<<E)8j*!+l6j~M0gb4HM;+B1
zZ6gHL&4xCmYc6P)t7~HF>mvqr2vZ~XT@^OeU`h=j2fRsiUo(7|EV4kuNwf1woP6L?
zO02w{;YUzY3;4FI<yN*x7)7Zk9JgCpXf+cY7%PjbS~0F_S3?8{5SEhy#50O2_G}VV
zvVw}PB&;N|sQ+*~wxqJ@XcAqL4ppAw8g_71NsvjdL79g*lgbS{P!=Gz9N1@O@-wo!
z3(}p`7?Pb}l{p(3(&7wg+q{r$N2}~ecd&Cm*I?%ooYCTLR2OkeGgI()CiInXov>xj
z0<2cD&(jtGw}aC4P`8xCs?#N{t^bDlrM@nDkrrG}_reuZ>1cW#4LD|#Gi*JWpNOqj
zfSRBSx}H=JN)``w0@(9w0DH|=U2Z<pv=@i5hg0yvID6|rl3WmLkKe5M6c^F{-2dwK
zqhh;o`{`XMwJz8L&Xe+s)IKPz4*^`?-(4R-sZgo7yKa=Eb)(>)=tdIo=E7b1F~Z@z
z#{Kg9I+Mc6N%m(a>(8@@zR<NlD?@TI%;VyfIFE~$!aTkyBp2X3ey+PNo=ofFvW{Jp
zi&w$~&b@I>?EAQ`dynkGy<dId`TgJFq5ZQH{K5lM(P75xw6WR5#-@z)Cd;}Tosy6g
zipT|^-Fk(S`!;DKf(}iQs?>@EOdTbv)NfVoqlNnklozql%5wcPa-my`=E4JX=TKEc
z4yBK%3i86_*sE&K%j(<h9(50)!)y;B5;ar!-)E*qGo4L`DJSO8OX%R)xZ%cZ<yqyq
zy&b>K!Jn`>Joo%6{UZGqun-GL1xB+37R^dhHQ%D{Xaa=H1VUyCte(N<Vtbm%nM6gA
zE)^d)>I9eBlj}!aWm6Y71cYR1L%!H-BRLgn8Juh|r|22Rtb|v~W0$3hA89E{D=#B(
zY1v#P<d3&DZdkW^6Vh_W=BqZWTeJD-)bX_S%A3w8t<(q~luj9^(y;%*_ophK9D7DN
z9mhkO`^?7Q4&|3tDWA3tssf7M_S+$<Z^LjE4PSFZxsg2~E27BnfUcMT3I7Aa;tvRm
z|2o2=fvUORi{e96-5pUrccG^KffIk=#DAR=b$2D}5R$y8056W|i5HXhS4X+1p%4DR
zkw0+ce+fsX+2txvZUUaXoV&6|P2H^#)`3TlbS{_ufmwfG)_<K@^}a4aNlwM+1^?be
zY`9WC9i>+?huAt}4XD?l0rkc?Ts+q;ST9zLwFCA#)W~8xYA?UOU4AEV{B6*J%hKYO
z35$b8dM(zL1h3|`*iv3k5_?|e?BNvpJbFW3hP#I=J&bB($rI5a{XJq5(*J0p6Quu}
zDyIKCv?&Scj~IaGKhRyL-%k3IJtW=_Ta<@RMv<vSs}RL~-pshMSwl$b$HIDj?JZJY
zb%!^byQSG_i<p9P6`ZUJF?2U*Cx#EQ?Je>E_^j%Sn-3Qj)Z<bkn-hqcS5y&W8=gsC
zAZbxQ=te<IB4WasU0Bs0JrT-QWz~yu&emJ?!3WZs=m1{ZlFH-G*Q@AxS|viYzOHrv
zuNESBG^jz^aLuiw_q<vhi4<23zv7DFl~3*6b$9TN8z)V#9CpPO!z%qXHU2r5H(tXa
z2(O!S-^8-!J8Py4sIT``4hz@xx%8UBcm8zh?pY%Thek%KBLlB)2#t&N8PRL}{KfOm
zy~QvJhn#<xU!~tea+!a2o4UV?<cYBD@!aW1x@FlZBwf{E^SjwDd8I0nuBh%qDf<a4
zWXVJ7l#0g<_RQ@30=C~ZLs%qov9{AonJA0xaJoW|aJpWjilwVoqq8t|cWAoL{emxO
zu>7(Y!_v(JFKT;NJl&(uF)&?W>xSKWzp}0M-6W!JIvq+n(xI?39Ri){kOy?gQ}@YB
zqeET_QMgE@LTOorN(I@+YA*^kIt2bxbVwlz<J|3$4jQI#BhKg^2%(`0b8jZ+c#;-}
zfELB1irmpoi}KF22)5Irn$e;JXi*}_z1nFJ0$P+1oIaMkVg{rkKCUuFQ9z|f+*jyl
zf~j1iNbo;Jk)+w^`d;p+QnOmWv|Nxw_@k*3i7tFH7-;h>d4(Kqr_HI(w0S8>8w~1|
z7+T^bET~ta1@#IssP`%(+H*j(=LGr16w#z*Vfi(n+jAP-)*zu?ROwb&vht=RAC)21
zUOEMb-O)=f!3bcU76H8Qtm;0R|NW$N{;!Yo@3`<FpXyM$@sc*sI&GYL!#J0d-lQM-
zS^G#=b{^?aYNSg;QVc|fu0^U1Q>i~2-3k;Bn8^&N0l^Bkej_>3q98@_VpzQx4ke;$
zGOPjDc;u<;dctC}A=_Sw!ybiU&%|L5RrWCK-#66hNwwto(v(5l<46}LjoZ6af>N&<
zSL}hXx@fX$WTQ$`etLUoF-6}mfxe|A1oVBfoxZ&~)3+)`-;z)w!YWFnSSV4b)t0Da
zE(0=`2~rIbGhbc8ksnIbs7Gb2N-jzBdi+h1T780KuCKxJxa@*Ld#@68<*i2Js$M-!
ztyl!U=jDIz#A2YSK5w!(7-^jjhub5yuiyH~7LL2~`jul;YKO)RzV!0TFC9FtCvu0M
z-SDHvt8Q(5Tch^zzT?LAjbR}-%B+v{$NASluho$Tq%>Qr?T^S+K<QkS(vma63X%GS
z6ZM(6U|X)&vqb>)U?rh^AHb;G3#i;nEyDz2W{GUkJeC-G#1?rVLNuk+UIc;k%h*D^
zEN48UxCSkPsMzex^`jXjc|h>z7cjwJRi>IxnPP(E7qg9YWkWsM5gDvbxCfm5WU*!%
zqI7A8jb^JUu3Do>T3@<sl}2l2bS!>tJK%J|txvB#Juv#jD$tlSCtiAgxyAF&39dl}
zp#GBfb9>fY%_zQi<P~X9)C;ChT(W!9CF95S$$y~KJ?|qGmYR}DZ7mGLy_nVJq@deN
zQai|51-1#7Pg}m1XT{na6Yp@jv$Fj~YHf~5>=DSlOJPnIT=#|dEL(M(BHNc-yQWFp
z^!#r+V!RKsrHXOFrLM^voz%+oKEnnF8^}#E!5X}wGdg&_sU?x6qJw8)sk}cO9Xwk@
z2gmv}3x@4Ii2Qw=w>!PQ%q*5+cJ=@hp2Zf}<X;FT95wp|pu%T?x%W;_aN)Yglx>Pr
z9h>w%qQ$%H`w-Cm-65bCX$a`pJ5OGIp_)H$hTf@Z#{Vn|I+}79Ktg|zMPz$|L#JC(
zc~iBIhgctr$*RtM1ivVer}i;VpW;ickNH|3OVh?NUwoX6V^(gSI*{1a%x=1x$GTc{
zfv%=1R4*`^qcro()zgUb^-0StIj%ab&%L3~g`ly+<gU)WE`?tE)n5DcSeuCD>6xIZ
zgJHJf(sMSEqm`#%1+_)pjSNfYsvRv-wICVrbdksw<syY}xn&OG6Zm6Nh1hI&@}-O<
zW;>kAz{R*oj8f+X#ywpXYXoW6#NCl5RrfMqqfJB$G<BuXbDI8bdzO)ib0*a<cOEfY
z2eP$fMpCBLGZ`2(tX9lqZ%I`$v9$-OTBfMD6sB77qqq&rxp(+I*{!1pM|Kjz4Qsz2
ziwSVa`)N?HT??6_tdMEvqd!5mX18izug0Cn`l#M)uNo`fGlhYks$xomJD0JDsY4A@
zRGhBGi&s6`zF$qU#aP3%7w#Wr>*cYA=~AtR39<RR)-ZKkH*XEf8NeJ^x&((Fhg$p7
zxmqcjR;@HUIcncM%6D2ve{O@?^LSdl8RO-z_1lRbETi|;y=~+YGmA|aOb9R`c|!~x
zG2^VttVM2hGrw|VNqZ6_Q7X-*vLmDVH6mv?s$ZkB*N4YGZ}n@qUPbuLsX+|GiD4%C
zTtOA*$?FO>d+;64XX5!doJ>RZ@*s+$505fqq}eWbGF9xR<Pn2t!2NV2C+?ZkY_Pag
zu~=0CLzP1qQY14G)FlU4%L|&C;ggg_y$rR)se;6k<bv8V2GlUA_rygj`pn<fN0)PW
z_FH*(&m8jX%!ape?|V?5|J(b1t(;Hn+pvCLN$%dO7VoA9rtH3MeA$Rgxskt{+jn`{
zwEC*i4bf#~KU4NRy7TjGl)Sg|XU#vEHL0@iBN|pM!~GvjT;w>}+%1}~bZtn^1vf7D
zT6E)b7l0eLEF}BTjeEQYnBOPh(}hg$UV*S>V#46dy>^*e-7M!XjJ1RR%DLBqx#M24
z0?Vy@%ivsg4}idY_vIQLaDq4d@2}423T3Qo=vaiM>amYOu2@-B##jy6*p);!&N8>}
zaB5GL#b9|G8dri^<SKc^1aCKIINIwix?&uo+@Rg#9PfSocTLLs-GUvXVbOLpg?OCa
zUSC>!J0daXc8A2Ar>!ne@aD{{9IYHL`{FQ}RimLRE^~re3*CV;)8kCcCfj?6)fX3F
zg|Rug8!BsWp1FNDRC{VSRD%Y%?bhg<Egs|T4#8zsO#{$;M9%Yd+w7F`@1E3KK2LnM
z<~8p0tZu2@uJpU1`C#b7QF4N;(fSZi_TdQ-n{z_48|JNh4qM=O70la(Avpu*?Nh9C
zzktsDLXeMXoqHNSf1!2m7<4YfJ7-a{bB)lsC*oS?WcM7Hv$pw*VqN&Dhea;k_j~F8
z{?#s_?uTSl`iyPc-46b()5fJYn+pL_MQ+kYCMP*Eelsfx52_{MrQu`*G1K1CY|A93
zV6yT%Q!`58e1#z&&KIlai$jSDHB^Ej#$trUGV?ipHqZSz-OuyRv5)$$MptTB)+e=g
zNviYf)R0F<BBTth>01A}foW@GW|IPbKffQ2>_CYq0rMqc)Um~SDryPn@`T7vu*J6e
zvyvm6Fgr*FVt=T+6RAiRL~CTAh}RZwgJpsE4Q=m^zj5@KzWv7I3l8bd?c0@imy8`7
z?Q=QzU7yP*EMe=pc{2U{D*kg_9oRd5Qh}?H5UWcrPL4={ITg~*hAENFHpyd#7RIx)
zQI9#jY=^vK23~aJ6{*l<vWab7T@1#7EcoDhftavnl3W(;=D|2KrG^2kv2bA_VZ~Sj
z_y9tMvuZ4ZOH=gIVE`6Po3wfL=t;`CR~Js3w8@lJ8=#{n9UM)^*ACyj>qvC)mYt*7
z5R`3y^U^cxU=+%>+bgeZxa7?GN3P<AWu8=Wz0dA><SM4u8OAWkWGC_B4&mLR4EuUO
z8P3#fR$Q567n=pQ$IH0uLV&}lN;-fJkHCViF2KY33~i0azuG=TL2edYDjb|m1nca=
z?ZZN!j-bd@Oi^@cA-^ZJ4n=L_A!uVZnbJcWb9!iFZU-7BfQG(ot%lG0Jxxsp^gEI>
zSqIWxgVS>QYMXmQoAbdRT-$A{Q5ly;WG~1cKggeoR2N(rAk0$5Oo#4((w2(xVu5<G
zAf!nn%y=;C0-fOkQQg<!LTu7)H9?1-)^%Q`(zVOFlCH~Da2*I@D{DN}4Z5-~gq(A6
z=*q}0Js>R8Dabfoah45&wjPCSS4{1DRJzo?u$(<^|FC7{tAFme`-90iBd6_yYn9L5
zW7|JCxWo_JKm5;c|G;bfm7H7iJ9AjqY2`YboK0DRT!7jdyyoPLY39wM5k*1)MtP-T
z1cALc=InkF^}w!Iq)(?I;D&d@xw|?~CyvY|#AwnJB7oMMt!Yd*k64!K@*y7_>%$47
z#etC}BhD`sLq~LBa!1?g1z{!QR)(-)HJkC>rou9@WkW^_s7|6T8^ZNM?F&N-iOa3)
zN5w#qZZ2I&hw*=J+rYWnzTnoja>`4}{YR<wKTf~(7v*bi6|JN*lPg81(8|C3gMa!z
zRQuSljz3KbNhPTz50ZE$%E9~yTcBGUW||-IIeURQ77iz>GBU9rRTvknQ?vYbmgPrN
zKyWiJ3(?1HVvAf4pJjdw5g}NGAj$^ZJFpteFjr=ks*d6gkC#P!U72iEvXmjp9{g36
zGfw;rlB|~~+YBiS!P|3Czk}DKFKIWUZ+X7YfLB*P_Kc_Pj%29pwqFnE_hM6;H^g&V
zzzpq%Weesld-<_3BeY=I(z)}Nw0lQ=lABBNT8+&nVuGW59Q_VeHqMZg8<A}~+;EtJ
zCeZ3_nllVWFjCFhvc*s$BNaOGSc8ddRbAlHT;}?Bzk4>{EoZlv==`ng3dZ(w9i63%
zR`_q(`<cXaA?_RSz4jyi^%M>NdJ4n8o(SwcL&Lv*BqSeT*ilmuY-0K>hJQVyEB<x5
zs}%d+AHnDxuSmxbBpna}|CMVE^bo&_^D~@z<}v77@Rhc?Hu?q)M!w2G53OFE{2OEJ
zBL2qkRr}x6@g{miKLO-i9+4cOgq}fgQ3NIcAz%Va&Z&V$4d7r>BDy}ywAF=Y%o#N7
zje!#lBO3-U=3i*Ie(v~Q!*(nw?N?eiWU<-@A8(`E^pn+l3!#Lm{ob4!$<^+SlRFL4
z#`o(%6AdHaw&5#h58@|iH@$XT!}asV_rm)sFVJn&_PzJ@^?2VIR}cCo?Y7W#hU?3l
z_}PpWe$o%AAr(Y4g0RJ*MM;|AMk-j&mIaT(CCMW0y{R`?9qQhjooQ8aLE6#B@#|3h
zI-(m=j_XWW4(ZAdVRU8f6he|eq7i|?NE9XMfC)zz;!A7obXlRDqBqO0wX+j_N-i`G
zbm^OeCb0&GXLKEvVGmr_VUU>r0OQ0DVdM0#wbOUe?Sygd*v?L)JT$GDa~UrUCgb_Z
z{9KUq)r^<a;Z98h4@0oGFd3<RumVPer$?+zb?XT>s%52vuNZJ9Sb2KDq=|(EXDWkd
z@(SoLroKL4;=}>3S5}=aESPjfz902V<N1%^{+}|Wt^$k~VwOX8=0#ZcLAGax!|Evp
z<W`slqL13dLW2oG{U)fu^*JiKo1e9-K85a9(vI_1(iQTJt{>Bv9gL-KjOh~_h4A_{
zi!Zxu@wkmD6kNTA2nTY7M7a43tIVH7&BpeabPS9ZGdu=V96H8R8mwyrmnA3*=q`b-
zOG9v_WHQ`qUB`RDO|;v!-;3hClCWr#>277dz-6_4nvQ{{%9sl9_ukhRkZp*fxfO|a
zySXL40!_44=9UDZTUXeF(yb8$nv;GMPAs{OooI^4OW?$6;KI9D?&}s;i+ndc{T9r9
zE#1cIWp9UuY>vsRmcZu?3_A)z<)gv*Ws%fISG?coXYleqf6{I#<SPt6*6T0E%it*O
z0Y0#kV(Z4N*Y@u+9Pah)(I;=MxPH%P{pp(qrA%+6y9~;5|BglbZ_HbFLRmG8?prp&
zmo<CYhwqKuF<N<l*~H8~mn^&g-qAZp)8}xWPUUolPxTy(i4Sui48*}~2X4guN%*jO
z!fV96rsH0Qwd*yQgR&_j9A>dA!&zDHZHem*$&IGs1_rHWz_s#v0e4bD2$)F>0tQiQ
zhw9X98$$s=&@YCwD)?~ax!tJ(6QD*!s8WQd!2bfwAPMkOxi^(F>$%*FmIJ=0-?wbw
zyzk!gtk0s`xOQ#ZI&M9@ls1VQ-^Ovv+YWM5+qSEGf`;8u{wU18WTukR)TWi15tOur
zR9)Fy^o#f>trKw03h?$G+i$#p1ju%E&vB_R(%rV>MY!=(+TP<p11Sgdbp<$|i#zT^
zj{6^=RT!`%u5WX0AkP7c#lbcP&vU4&nX^1V8KWL4BL@mWHs<6PWV4V|PQiuuEhlN)
zKzF-`IX5b{xKE$O`Q2}3PJhvH$7qhx;dw@f9I{fQLT04L&DXO-&0@CR!SeAe3DUXi
znOf=+Qz4n{-R!b6I~G!8_1sU|WB=xL-?P@gCO0KzV{2FieF~BYiIS&Cycf&7)<)ER
zVZ~HA?rsqXhjG^&>|A|ni1R|S$AY_9NO~M%=%*CIg9wgwpODmxb+|(9aD`en?F`E$
zaH24bHM3Yx5Vf~7=hjvMoqEHN$KXt91?IPnFyNRzjr%-BIQYG5;cRt;)d6B9nk5@f
z>g{k;RTJx$I??W^^?jW2byqqnwnMs4w0*>ppmP{OM*~3@E?wV=xX$iaN97&E3`XY=
z*_otr<prqB2wWY)Z6koL&0@6&D3oVWw+Q@R!d3zqLuyt$jhvP3<Rr44qR7e4i80`8
zsp_92A-!nQ>B+?rP2<Y{52P5hG5`Po0096100RI*cAYR$Uk^O>03Qed00000(firZ
z00000*U^kB|B?Oa2RR1*0000800IC200000c-muNWME+b@pmJToc@2}{}y&j2A~KE
zcm)8f#s`Z4c-n1~L1+^}6o&tuncawp1SyDz3MxixC^<+dp@cw5QEX5Pp#%vb)FQ=0
zL3=P1BzOuvc-TuL2wD<ar3cADL`qO7Jrya^i--r0rPzuf7J}08f13nC>B7gq^WK};
zx9{IoyGVnVS^=Cb>QyYL8M|-haM1t>_Xqk-3o&Qd?yDWls~!7IeLx>`lg=7Wt2v~p
zMW<;O7`slIBu@$?MyAOGx!oZX)N^mA=Dy_BS4?{^kug;qHx<;(CJJtZ8beM0LCrnO
zbsjaRg=L*%Pm_%QM$H>U+08M(#<Rvc$~tojH~0>h+$a{!6Ce~rif5lO<Lo_(nA*b%
zdtRmX=?JpAVYjKievUMs>FXtAO`P#B$f^p$dJ`wjdHBu-9y#mwC-ZXedlT>_ZW2Mg
zAM827{c)t!yI{|}uE4k@^y^ztx?-F98luca6|hRZ!1rHthTF5}{Zk^b2L^UmYV<nd
z>@DMNA)`KH(9|&?^OxC6jJYz`jUZOXkighKt{>~a$jiLBK45%S7cgQP7;)cYw!;u-
ze>0f-5qPicFY!Wr=*4N-*Wv}WN#44nIM?;zFwz%wO1$VL@FK<Y>YOj<eiYs@>QdW>
zUYrdp^M6KNCNIn==il|=Fnm7sj(E|FcwsKnD{82SDen_&$2;Fc1PNURyaS-CD$9T5
zBe~KB->LVgcbTIXU33G<&_g6S_qcrLf-cd+5;ae%oPUx30QagUl4L^8WuF6~DsuV_
z@5c~kFJW(+>>>OgVNcT-c-muNV9<fWB8FLvJxm9fW0>zS|6$Q%Nn)ABa)*_JRgE=>
zwTJZ%n-p6f+Z}c(_8#_M910wZI0ZO;IA?IK<GjH6ge!qthdYk@5Dy=ZA5Rz00bV`c
zG~O28Exa%I`1nHjdiWmktMPa7KM-gT*dX{uNI<AfXot`XVKHF`;R@j;!smovh;WH?
ziI$1(5?dzDCmthVCQ%`=NqUy_Iq6q2J~9(z?#P~!+a~u$ev<qf1p|c;g<T4-6x9?%
zl)RKulzo)1C|^+#Q#qp=rg}k5L(NBhlKKUW9E~TMT$)Cj0a|QYDq4A3`?Q6$o3s~c
zztfqdvq$HZu9|M1?le6Gy+isH`o9cJ3<?Z38O9i%F)}miGJ0ohW?W-@#>BxS%Vdhl
z2~!2rCNnFuS!RFCQ!I=u_E_ArjIml^{mj<DuFLM2y@35X2P20aj&4p?POqF7xJ0;Y
zaaD7j;U?s^!tH|FFLy2X3J)WXKb|3;YrOcpZh6OfFY%G{$@4kjE9E=G_m1BQe>?wq
z0X6}P0?q`6;DAYi1%Y*eGXggSJ_zCovIy!4dJt?BoD+Nm2w#N=0pXTVwNSUvQ(-1y
zFT(2p%X6ET009610O|#l00jU5000020096302TlM0RREF00000c-qZe+in}j86Mg0
zA#Q7=MG-VXTI^LS0W{-N65BVeCDO4P>i|)*k;@gil((ixc9*oQJV+m)cfH9Y^s;Ev
zC+J<Dr0C4||NoqJNKvYR7KX_^%*=m2&QMCJ57us}>uT-BdrGbSLMgspQ|tPUuCJ+E
zYro~|>+0>=E??hJ57++6*WXhgt^J#?zpvWYWbO~t+iOqx`iH7>?Js=&p8DweHNL*N
z_RaN=`T9rd!Hqxi^-q5D_QrR7eM>#)Z1VN{D(k%A>z}G0cVzs})OzPneEm1|K}Y8P
zUES&Yi?6>^x4Q51^*_|j?yu<jpV!q#-F3C4CTgnARi#RGqF$+)%G96r|9k3={`WxL
z#?@W*nQG~$`+7vFM(Wq9)^|_!+_9cftG2$|*4KTE`b7W#oqkvA(Q}pSyN>>>)Ib$_
zOs1ab6-N5o!K`st=<h;*ztNuqthuk}AM3Ft)-q2seSfY$zr>2Uo?ob;#UoQks;Ae_
z)O}!mTRl`?s7Lzyk*34^%hVbEwe`%@3T;ipM4f8d3_gNiPmdexx6k$Sef?&l*VzF+
z2m13wV>!ZVKGi<Wp9Gw9d}}e++Uha*xf;w;uYZaeBBfPlW)Z$G!0%emoIo=7H70k!
zxr<i39pDm<&LFQ0lAnRch6|^F_*y@kXnrr&rIyvD_Jz8%Ov^r55ys**Qns<K=WHbH
zLp1YZnD5d;4gswWc;44*)_RW7<IL)|(fm?>ztUF&=zI=6FmH^7jq(2kGfL<P!K1f*
zU{<ph&7f0H^$O<svA&x?vPMe=z{RXu!+MQ&89m8CO~KyciGFv&cxIY^cfcRvrJ;DH
zdH+~_s{gHKvCQ}E!-(&n2EO8{<wlorJz}0_TDyHZ4UQ+^fzjQ9XBoIvlJ=VsO%DQ!
zZsL7kj~wgqXPT#mlPBKVhAT!xb~Gh>+D;Gk_+9;(>CZj=y{+}J3#;E(&-C4MO=%C9
z>|tCFls?fjpXl$d_dLr!*0j<5U45@t&wdG5X8udF7T)jbc}Ao5zza9Ni?N3Rv<&M!
z)mJY8vukG>tiRUZoAAYE)vo^Tf#bWN;}Gu+jTzo|^ydrxzoTb7(U4<&yQ&L(4dmxM
zcNg%U0qz#=J6L(!d-nu*Mk@isEj-z0nq-|r{du7I-(xB|$+dg-sgV%T*I`O%E#hm;
zK_is|V6elSAc?yf+YH&Qe*jz$!Obps-NjsobBa2`JT=1bydA(K7`up?TOR*9`n!t}
zq<w-@k0p2nZV|RUNUaOGJoO-7T2Gg0SN3z(LZ44Dk4XcM>>c5;OlNy~^rrU3UFI{b
z@W0du<0&?w<@<Vuk&0O66Rc-s*a2Qd*@1_fuh}x|ok~H+zb<6G1<f(z3{A`CY`_l^
zo<x_?t(UOnJw2x5d7fu1lNhSwJ@N3Xc#ifMUqfFh%6hzo=h%e49<pw)Gd4YV+ih_9
z!1@=GU&u=Xe+y4{K>IU1F;d+@Hb7(d;JeBH`S_)z-&*LIzJkTQ1eZIIO)v0gWCP}|
zhkULJ9%it?=M5WNU4z{`v9DM=Td=KN*1jI|rQ`X~9gK0>C@b`=Wo-hhO?`dDqev%Y
zep^F-n&eH>Bf0fMyX{?y^0Xv&;6)B$KL;$?LykUM;5k7NUlVdYT(4tSF%fQy?R*(b
zqnA6Fxecjz8FGiM@6cKW`3cGw#CM0dEVJfq_UNQLgek2N$KMaFmouC=V(zZ%ryH6Y
z<Fgj~IpbsZS!cSSq37xA<^5BarAfR$Sjn^Y7|Y8;GuHJ2lywu-Io&f7-eykkfkQ5q
z?{n6&SVvMgU&eant%=1iY{upy#Mi*-fIY?*N0FY*2Hv%=$gGaVxeVuc7Q&{@0m)-h
z)ZB&MA7KsOr-<$E0N1XiXB%4VaBQ<CZ>^jO#PSdiyJ>T-ec-djZ;U3lmPV~@>#cV2
zgy=DpZNl1G7i4RMaW}cvQ}&;-7HRiw;9=_PnXWN%R4*AS465HV)PDtPL{l?Ob+t0%
z>Sv0&{sfiMDUUF9NshX(;JR+iYf!a0MHTd#Ybz63&+x8~D$`W_=RB*gE1o&(&J5My
z32MX~b*>pxV9W#~2H?yo)_%p%srogYgd82A0%K}DB1bQSo|)c)p=e}hx$4)<oS{C>
zfqMv1Xsb0%g??UQU8(m@F_$u#8Mst1C##^Y&be-vb!(ncfD%(-|3Tv?&+CSyW~e4d
z{N9I6PiRfUy)i!_DVa)iYDb+ib_Rz!sLBW6Va+@+@QKnH^T1KEx}M1DLi+%#&TU<5
zR+s*gp>pKG6CM_zysa&4#2AXihYXcE4kd6gT$usxjOFHY%7=BrT@~qU>|?c4P$@W2
z{qOUs7wLfO8TcInS<}EVQY~>MRM%FcdRR;FGS8`(`Z+$oz~@5``Xgh@*}A&x^>+*n
zw>9=TC_m@UXa%jxJzSnZGyQIY>9n20Dl)5OIpaffa<5nCU`;(((<H3i!g5`a99^(E
zk1C*J8NV~1C!n%a-*}R0c^FpEW9f6z+QJvIlFG_`27G3ou3i?h!`t^-%enQ6`PFA>
z5|(~f%z4LV|EP^fOzRK9Rp&)=JKCr1|4BLq!W-Z>2Bl$llBl1v?TSpC?KI=k8B=Ne
zP=S>zz&nPw@g?DNW`YqUZ}Wca?FtCgROgPYv1DxNY019l+;X}MEar^8lUuo9cysO=
zlC9o@WWV&ZlX~18aQQW`o;LO~L{?$nQ7m`tX&HI$D$remFhz`YRMJzyc8sy)M~PD|
zQjVf-=MemAHR5#`_X~~yG?(C)2?>NW6`mIQ*+YdGHHU?@7R;tuCziK$U}<SsxuyNK
zzJJTq7(TvX8s+Vb*R46yIR5xu|E6v)qQLhgPI@)tP0}d<p%2&-SX@8IRML3?<xM6(
z8F5}Eo{@5%DO=4x_7%*YO&>Ivyx?2TR4Zg3hkCY&aNk<%-^2Ylcr)YcA$Pi^Z$AcX
z;l;%BVC1#N3B5nVj>xDHEi+_v16U>LHAUP}BhH3<9m{wnStt335^G4NSunrGpkQh_
zJ7mo;8SH21=xia|uRXn8#g^MP<~DVbFrRxqdXXOadDtnS7Xfz{W7}E+Ipn;HR})Qz
zWcg${JG?JcYWr}zBUkZdq1BVTr%Ww*5y}OPJ(`U1m|>Bf2vwh2EJz2}ky91#6w?~I
z<Y#K6K_l$eY}n4M@%Ar4%Q(sF3Gd=nSTVKpBs*%ycvC@F{T*z>sR}lI0u5NSujTJy
zwyY&<z2T9PMUmcDwt^sUQ~`rIcvN|Lz@jW;l*oKtj<H<xoKGc<3p18?E5~*HxdE7d
zW>}+EAq+}fnR?ks^d9snX9>sJ+tsmSW~nA`PVsqFTFJjH@^kN2e`8y7R6BnedYU@O
zNX2uG;niEn_$J7KTy8eNlQG9*s^=+NsuDSc)TXH%@-#6&)(r9bMU=5MA-M=ce*Z+T
zNm!4Ab0g`TVOhqQYwiu%GsJo&+7#zqQsdH2aM92r>ei6%9kcHB?9P5=t5Whhhb!dL
zq3o7NzskmM7Y%dHj>lZ*i-yu@p};*~*JInjuocZKm`{Dq8VJq?`&UPy$wx=F3X$%?
zOF(?hSo>A8tBLxItq+4v8&KXA!VwOMHxpUq4K1K<z!+BAW`0*e-fOgTiu}sSd;p52
zuP_dxdtohZ%Gjjh++vzj^=Y2ae)2tI=&$Rb2?j&-lqix3c1EH(x+x=Xg&JAW6zdZ|
zhBmH@IES7^bx6pSnl;LF^&Z=7$n|-gJnrHI%`)mKe;%{##Wi*jNxBu}O;n+#vL`yL
zvlwdOX=vS$<}w?UUfdjg3i2X`4}ML%T3sCave;Jdfr0^Q)gkkh)>=#zS<V|ig)_mL
z8KhQ5{VMTktp&PvYEQTvvlSdiUE;6?qQ33S_SlwLozu5_OQj@Lx17gnuN+I`s7cZM
zrdM0U^yb!f>V`*c*5gjT;fc7O?Z7|B-ie&c^D{r+6O(r=xmQVCO8JhsKU;fNK=NFE
zN~^x`oK};EF^8HvvyNw}j8ix!84^{L_mOovKV7oM_}vJx#y8xDk=#|RYk*EteMt-!
z(Z9I%n|t|<v2edK_ABDK^@(xadzl`y)E-SqmJzMKk0ftIzHD|S&CH>Fv_G>BsneVm
z>wr7$N%m0k`Cq}$ziy5?_0EsR_s!8YGP4rvjX1u=&(fI4=w+iuypI*$g(Ej?o-^Uj
z;BjQXasD04OHL{V7Rc2a=^$zFDSOk}!kKcdRa;$f{;I3J-E-J!?3ow!@!U>L+Aivb
zH}g+h(n-b<XNOMLDbIO>HB;-v*j9WdxU`vPhHjJO+6K~n-p4&dY^k{Znc9<ci#zSU
zFOZyA`f)SJb<UlKoFT_uLaLueiR^yDy}{CE!*Ty+d4JNcYEr$+)G5(A*l81L!@T%%
zCt$@Z%{*-r&5(S4WIL7ffYK>@mGfv9pvC#ajQQp|l(7U|7A{_M7<A_^*C7qNgR5fT
z>au~?Rqlhn(&n-$J4p2%7k3(k#yD2{^|bNdxwX?UGKT%#0Wwt58rS#nCF#TVEmS;g
zzCyR&?K>T_M$XypZeH5%PP~1Q6r^!Ok6foHX9>-`{qFEgzG$86M^GP+IMRGO=B`Nm
zMt+}A+j9*59zX3SlfNu%2X$gA+0^?D8>K2MvCM<hx<I318vWh>m|wEP*UUlURfa5$
zK1;Fi=8qaZrPA)D%gKo5F4I-5^<s;1z42L41s+>_ihrMx$`&Obp0EYEygJ<Hq*bLB
zDfleF%&T5fzX*A*-=8(QAbHg3o7p<(d_k;i1s-0i@P4pkmwQ!+<NxnbahkOTm=V_`
zbK6V0T7+;_b{HdMgs+FR%UF1MP#xgT^_H&Q4wvi{#yjIzjfk#KFvH##MY78Coau7`
zvTn_t$-b@c2Z?-OnGVbM8}Mx&^oJLDEC-)5mLc!|QvWlz5ju71CHQUC%5rX_c5CkG
z3--=~`|&BY`;poW%oF|A^l5?1?~G0YOOQTX&Cy-XlGI$q)ZCpM@)@F6i8F2KQyVvV
z*QfIOL+)e>?FYFxp{JH{VU{Qo>vOayBc6x5y~Ou*ugjTxzxSFTf0)9@jXCbjZ4V^Q
z!I$f6Xh{?HmvvxIZUq-jeXVsAUB&0*!AZr`F9?16i2bq4^h^cu0MZk`O}~GXnx`Lh
zl){Q4&ge_qkqqG#?0<ZF^)qUJuah#g%ASc#*{`g$%sX7&#*ZMcIY+|^ey<2PDKRtF
znKVCAHbzt(!)KA#YA@L@n<h@=NgnPxhhY|anY_FeMB9OP?$E6K*zW$xTIZex*1S9F
z^63%34`WNoIG%Z3?u$4V|EAK9{<J6LQ#bd{m8LSDlqQvp?pb`sv)~i9A$J{=wA{k;
z6|dm42cU?~Tgr)E$r<GgjKsZRl&cc{6>DD6ILmZNJ=f!|UcdUegIIhX^xrG*<nY`=
zcR6XqFUEQZk1*og!_{}ktdns)=Et$v5>xTX<Wrw<hPU{H>uNr(`D<UN`OEtnz7D7N
zBjeLnmDT;)>(se3z;s`Er!3|DnEQ}5<W+e#RNKfa&zgNMUz)YuT~GbSS+vxhi^zg=
z5xG61dLHZt$2CWIc20KnDl4aP<4oy)<L6C}Zn5QxOim$vvx{_YPu#!eJgkClRE!l#
z%8!>Vwu4t(OBP-=Jw_}b&)wp)4>``EwYYxp`yYkvi<I_UKS?a--Q!d5#Q%U};kjMU
zWn{HICvd`j5P{NXJWp)@&d0;gKWTqu;8l;uoSC(j=0B~4*+{h@he`WwIh(ksnqQPp
zNc}E8SfpI?4*5mD=eRhEj;*~IzP`2=BEQwIqhIS!`R^dqSNQ81i~qL7;=gV25}k5K
z!(IF>i<h>JFA&@Asn2;2__6+cq(ASdFYNF1A4Y!}VZzor4)$wh9|yl?_zzhz`N;qP
zc-n2yM@*A(7{~GN1Eqzs_XhVK@B6l8D2^7Wd+#j@v{eKwR>f_^J)$wulg4#X<DSu|
z2Q}^$N7RF7cRZ*^W8wXK^SONT{0`6a<ROB#`cp<X{#TztA|#SXEJR_&Ml><li6xGB
z5=bP8WKu{ajdU`|B#Ufv$R&?_3Mix<?dd>AI?<UfbPez6P7iw0i{A91FB&=q4xG3s
zqL>m&=|_Lc7{EXVF&H-<%Bi3dFGCp0ForXNk&I$AV;IXg#xsG5Oky%qn94M!GlQAT
zVm5P_OO;qSN(&pg!9G5+iCt{x0LM8dQEcNq8`vvWv9Xi=Y~eN^B$|Vq;1@sn%@I!V
zfP0)~9`o5vHIJy_J`Z`uW1jGoPt@{~=e*zyK7O!=H@xCCe!lRHtt=qGLKaa+J%?yu
zF^vRiBE%AwvMl`N<*cBYm8@nJS2@gD*07HCeC0dWILkS%^OkoKBX)_EIEj}8Nt7f>
zmJ~^qG)b2X&U1mwT;UEExx`&Ib5b(7$t}r}Y<5VF<Vv38OM%r_*WBpWO$=M8KA>rC
zO<7^;QMt#VEN#`$G)<{14W&crRJxQ!%3@`S^4~nSsjlYgy1h2w3k7SctD5{~WOz+o
zUQ1;t*wEIvyydNWJ?0%@4W&cr{0pp9ocjO(c-lqHJrBW96vpx1UcGHq^_4_IQt=XU
zR|Bi4#UiFQd=9h4$Y(h<`F1>!lWRM_Gd%yd_}oc6tOI$+suAK@H+!Y4F~#+P@)M3*
zJtCzql|`~BN$MT7it;5BTmRqytv*8m&Y9r?YQyjVcg^qtZ^>u_{(=zz?E#|$1l$o6
za=nXhIJ9*m_g63iyA#70j8=vT7|#t;Fqs)<U^+F-!R$ZP1M{t6AMBZKO}YL85^z@=
L00CR8+p7QoG7A`h

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/FreightSans/freight-sans-medium.woff2 b/0.16/_static/fonts/FreightSans/freight-sans-medium.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..c58b6a528bb611b9279266d909a7a49e7ee98ca8
GIT binary patch
literal 25460
zcmV(?K-a%_Pew8T0RR910Aq9j6951J0X%R30AmFJ0{}t*00000000000000000000
z0000#Mn+Uk92!&`=pGz|Oa@>8gHi}e34~=42nvGTe1iHE0X7081BWOJgH`|pAO(Sp
zPzQxv41zpAcV%K`23uxFaRGO}SQy<2MQYvJqx#9VVefX36tQ;+MF(owIDmosA7S?Y
z|0zkw7_~t+z_hgd9tu`KVPv8zL_!6(q`-+x^oD_?y6&3S*#eyceR>;BLo0u!J;sah
zO_paHI^eYRyGrq-_V@iy-noo?&u(N(EBxipeTf8_pWGo+JFEwy6wNC$A0s>;SG<sD
zWgumIh1Y0x@OZi8QgAqeWNwG7g`qCr&@mG_?ps@TP;lf#af)g@szEo0Fa}1479|FT
zYk}jukdm1BdkLbo@2&2h-FKfbKpSKgh3_uJEDR|m8Fs*+5{(Xc;d%1!^W2&F0IJ<c
zDgo%Fl2yg!W`Lht|K0)yiljt2YS5!b#S$26V!&WUj2t<}h(V7oW27jVc3|Y!JM~&G
z{a&iSv|rV~sVMyyJ)1n0lRyT@IXkGUQnD<uWtnk?7jO1}!$d`&Pswl(Skx)Xr=p3L
z%LG>Lv}^fG?ZeXv4@8ZCKL$d<r!+W#N_}|#v*l;?C+-^2j1Xp})y^!Hk>c+N05eC4
zBM14tuS$}SGXKmmQ0S};7%lvtPHV=t0{=_)X0c6q&;jy{zU4_)3jkK8k&{C^+abaU
z3<n9n>!@7*slP`L{#ksp76=6%6tH9gp`w=H`j2zfFG13znRXdomsQ=7P<M)@A*KH;
zI&;E2A&1bMcv@Oog7C{Rnw3lChWxnz{t=)qLUuZ;$XxjGL67L?P>N8>rl4~{*%7??
z<Nec+zpr4xOX`4Eh_B$X2Zb~&0H)FRI{)9KD-a+M%m~m8`K55@oze3@mUG^aW94?Y
zb5-ndwIOwE>_VjI-BjDKD*&<opDuOEGPF!53M9UrNyX<M=L{YnF0bR*_o}_qmdBK;
zQ|fe6`qbKtB`ot!1_i{GzYa0|y{Z6~urE2|EaR`xIoq=B$GT7fT!5INQwqqWa{^cZ
zJO68z_U4F-<Z991<yhFhDv|$(b~2gdk}M@TqU<q!n%QKMEYJ}a2#$HT2#!DjN<!OL
zVa}CiQ`+4k3qxQ@$a-5PTX?Gd)aGTf;R{Hjoryi~LR27<zqRaYEwZhu(RKB|_J04<
zKkG!WNQ_HCDTJl}n)PegY+4a+b)6bVI(2>;@+IuOyYHDkeeWAWN-AVK^%O}V6bk(j
z67{SaQ2d{&R{IAa#ewX5E3Caqm%9{SFQ?1hR+sxLD$Wc55;Gtr0-&@66qX_#2uf=+
zI8fo1q+~y8qjoPP?S&LYN)EKVLDq#3nkMA#iuS5L#B#5TU3*YwW7#4vKbLEp4uCT1
z|HgX0&K2q}?=eY)6ORZH5s`~rL_$KEty#7~=CtAZv894qR9=8U5(0$tChFdC%TTps
z?#;Sn@ATXCdzgUv0y%s0KWoqSXXbN{vph-&K}1AEL_{QnULy$b)vrr0@Yz3=hYaAG
z&pHQ$g|I(BauW{1srRb5Gy-b8SC}TqK+X0F(;OA3<%Ynt@($>;??D*IL?xK74m$As
zQ-%)&LH3PL?m^(NXVkjI6l;qET(>yX#}<djYwjQc#!*0=cgD~VC%Y0d6clg0?p=XZ
z4-BsIXQ)WAWl1#x9}&R`LF9%c$_A+WABy9#7H`R)8$Xyzf6h78RlW>LPD*mX8@Pc!
zhg{1i`Az;&Dw98ia?7yHsb9Nt4@vm|{1m6?h*BhJ%p_2A7a5Vc6Osr7-aN3QototX
z8H11{r`D;0LS;!6yXVyes-KQeBq5Ffv6w1S5f{ZH5~9Q)DN4?*B3^XN_24`=Rb`;U
z{>&M8<SB-P{<W@OMQjWQE^)v>hImxT8UjM%*T6Ca69Psz<eozGyZ6xG;|akUL3@%~
z6xlMgyofRC2o(hVP>(oNnQ<ui_ivaKZ^uD(7BEN=kXD%}S=BI2z)^)D!ODTazM4g<
zzjhQ&3~3e7L_9Q)sAb-fh#Kx;H3$JhQ;^8c1>rmZ+(O?4M@XhBMNzp?q+pE~6FUeZ
zxHSbBT4FLB8`yj#NuX8b3O~b#5*+cQ`kDX}ON4=Pd4sDgh%r1Oq%aYzM4~}qisc|q
z5?A6OpY5gE0B=Kas$);lC@rB|v4L-Eh^IR9wKV5kB=XQ>S)O`E?8Okz^S?oU)A(Eg
zFl&IpjI>(H_){8J{vA8UP|btz0e(aP2m%O{0+m2DPy^ILbwE8pp#$I;H~~(9GvEb!
zgFb;D;1~D<{tI_tOobOBL>Xh8iBGjM(l9}0lm)U1z_y|6P`aIw%7n7WCg<r~lml|2
zT#$RfgBKru0tD-HAt!3v`5ZLSXi>D;?NBGs^>jBxfgVJdMr0t$7~@RXQ>bZZhFRvA
zXMshQSZ1ZLidtix4K~>>>_EHh4fdsVqH`LT;EHQ*xaGd^0=)hAJL-eRCzSgt5l`gH
z`0;e0Di~mN5PS|j?xDVeeO9M4@_x`qQ46Q%U9pKs$uSJ_^coVv`8jx4_pj{b6=LPt
z0<~J#K(3%GLSys0fYLepB7<mDP?d3~)Q~6@Ymv$2R+$Y{s%e$N!bx;mLz%A(<OIXz
zL|P?N^X&|mpEc&L5vx)nMNIcf<Z?#HL7FUAKr$QG3Imj8b|_f-MLE;3N;jU7)RixN
zD`w20qopGX-R$@^Uocb4#cZ=GS*?0mFi0A!!a>P}lQ$SHR*i=lI=fiaEwnAcxXjZM
zbFyf6+YUWNHL4ZbHgr*fd}5L1Hlf-Ur&oav7$EuwF0TrM^C$&1lxUh`OproMpp4Ey
zsTFmL_5u-3f!MyPX%wLk_u^nf`HY<-8gFU4Rv@W0JE?_sUB=85c^FEl`h>Nm_urWF
zNIDUpD+tj0Db!J?#C@MYGNJbNQuYX@Znb3WxZ^%D*+e?3sWT&865U~$8nWSH0SCNf
z)sfe`*K)~YKh6Rl8-Mu#sH$AT7-24T3zTk9BI%T(Z1}T`bt3!{tQ8(mYs!YB4G=bg
zPp~AU<c~^7Mq!AwkAw|TCc;MSOd+J24PQH<Os0FK8FD4*ED|ObUe&Jci{{a77irsy
z4C$$u9SIg5DQr+Ettj9YNiz0yhdo@}$tHr25yrfUUBBE2T<?+T+7~xVK=-9y0S)9E
zWK8;D%irI0^OLr6W{z(PSB7@jCHThko%l1saP4Q_NWN5y#4m5WhOAlhg~{#^>S~m|
z1i8m858LTVuf%fBN*S91g@WPDyAApnw0|s%hD^<`{0$ee%a**Y-5g_IyLz=-E5fRZ
zU#)iYWs0VaLt8eWKcvf3hVG#6b}vJK__L-6n@VG?+D;?|Gva4JBG%i|Rh?4q?4b#;
z(A^{VbV(En<QoU!7Gd&k5SzDl^wyS+_1)+g=J65?n=luty2;5uds>PvQERa=I|j4K
z$4>3~(?+m?RZF^I0wta?GDp*~=_YvZm2Q&RVHWPmG~4*(-f3OM-{3Xe`oYZ);U#Es
z>!X3Yq9rmF&KYkq7numlou53KfY}gxr?xlhD5~Em)U0Jf2tAH65m+-XbDoVZMv@Eh
z5tI}rq^Ecouae#~9s*PI&A;c&)RR2fc;cgbrqqzIVZc?U0sIV-F67rUo5ml1&L41O
zF)&-XuBOrPQJHhjj66%0X>Hx`xj2az=wQ9msc_J^nzxnNOhhALHh!sm>ZiVSkC}a{
z+G6<PHEGutMS^LLVk>HfLsPNccxGdhJkiLbKNo$SNX7$P4t7tqYu{)hB-9?X`J)XG
zVNe&^x;|%{b_D8-g2v{uLp|EsC!em26xE=v7T!Q97XyLvYrOUwAF3i5%d)9<vUQTK
zGv^-|dr?bu{`?8T6~azh7^OLg;8(_4H&4CtgteYpgnA<jK@Qe#&031u&qq0IWY~J0
zaaJw_+~vUeXA&p2H7D<7E1F?r2PT}4otNz4<>MC+q)mq|`47_vh6S(5h%vxAOqq?^
z`ae9@ZU70eAxl;Oo3Mdm%Wf#rZOsqt8nr-^r1+E0Su~6{uMfbQI}^qLILih^5&(c~
zjtM3cr9mf=$I;GQpt-KKfGt5Zrm58gEfo|}^_i(w%y$^EMm8!QN`)yR;E~M3W0VYm
zn^MELt5D@sRTWfo-m02PDne!TfGVoGs;Z#?aqxL`MngTQhcr-)kyKM*s)=wFD?&9B
zsS;6C61r-Gp`tLOm@KHaqEw`46(>fu5UXlotD4%V+TvA9392rRsv}Y8J%UGIp@FqV
zs4&Ya!}J&ceHLE{wa!t9$MNW3VcH>ksGG4FvrNocM6@)hM=T4}(|P$w<q_NVXg;7v
zI)!s|oTo9U3s0$|JoZ74JOX;6s(7kGJXck`5Rm8fvi@g-1rety1h5`+Q27cnJJ3P(
z5Scyp&md8iX@!kgtVN535+{*;rpRS$t<bs*@HgZbETvgt{6!RVIa<VXMhUYo;&|u<
z9+O{{A|TAy*5n@yl6|C3Kk>#s)&SOw^Gt*6=KlGjf}6a4wy!R<XMJ<LNavTE*Oys^
zs0R@sn}r`hLgEz>3_cshDa@E92krvk(XI0vfKkNIfW&-ry0ugw?RJyI`$Kj45qjZ1
zZ4mW<N&q<-?&xbFeGjTJn(s{aNZ*2J=^GHufP{Qd*8i|#=l9ioH0PZ`cxmGlDgx52
zZV{IGP$CT!&VDNsLaB!dLPVGX+`Egn;#yQxAFVx2n?eG@M8wcGAGN%_AU(#;`9&mw
z{C1(ECrau15N=9;Wo$1Dt%vak8`G6-w*c-j50xLWPq)v4z1NV;Lx86a2DYEs*=r78
z@9+;)$b-qBNgDue1rYCN9>4&`RUm#5e8_JttP92h*uv+#?=;bu$dKmcd@;C@E8D!S
z+rFLKqxNe14&?p~H=6!`d%lFh@nBAm4036kwq@iB<h#`Wcz(?NZ}!`N-@9&fUGBPI
z&$bN<rqu9$^P>8<@s82SBi&9!kKgAj66F6^foe9p!|8H+ygq**xXcg2C{EHWFUqQJ
z+O8kQX`WPDJ9`I5CubM5tDC!rC&tU$$Jft4ATTI6BovFo6Nsd+@Cb4wg-W9{n5?Mi
zm{@jPd;%viiOb^y^$m^9-2+F*l;e{VQzuW&oSvPZTUcCPU0K^$-`U=!(`oWg@6qR#
zbj#p=A0>TJ+(QB2tg8TU{`E{hKc(A104})UwP5OuRp&2+QMDGAPj&81_U4D~fk9UC
z-3wNG^RxwX7cE}0aK%b`?%S}AOK;Bwa4i7!Xb13!U#)(Fj&{1;h)Efyz0|FkV}~E_
zoF!XY6_OC>Q)5B1Iy_=>Lv}KJr@a82ETG$z`_rBKN)lPs%t*lQrpqtvEaAnH!6nNH
zC*+q^!I{=XUiYabjaAT;o`JXNt|N%cX-~o0sOho)%558g+0g&xuZ@@zgUvZfvW<$M
zI)o<;PTMLW(_#-Q5-uGSZ{-L<6mj}3kCELvvq`6fW<4jLkY_QNH^N1M=p@`g`!s^+
zN5>gWry%#04N*-JCm|+{%@TxVi&9i%IXNMpXo)L6gE}cV4$N(~cLc}besBuIokWSe
zCZZ+99N(gC8sg#>af-XCByIG}<8CPy0xxtTwcWR*s^D!d3rK+Ja1cb3CDhNfJ6Yi^
z=C%f3+YlxE+B&E?GTdT7#)}5$`_@!ycsWfTAjduI>Hda%z{7~!zGOEQRL%rWAe{ez
zWo;u4EvSO-9l3DaFmyTxArVq%R3IN+3JC3}Py2Oj(B{=_+XsPF9gzW&Oz(1==}Sm(
zr;V1e4}xPk2)R~DIhjo4M2C!#SnnN*1CWeaV*45W2q3-$#6y6ccL2byH+nfd0OlU4
z$|7GgnoY-z1@y{Hf@)Y^Ae?~-*Mw|*B_bf+3hY54k3k04BZJ1Fn+?rC;x(Oy;b|K0
zkvR6E-e<C$G=v>%r0_URn&gJ8<iO)0Q9;sX*Qz7<Ao?*ez=4#b&B!8J{f4SISzXzD
z7sIGm%BRx&A$Y7t0^G!VF_dXtGx<O$r7U?xkqsp~@K5t3bApgZ&aCqP?yG$>Jv(oo
z>csEI2NNa@$$_|Jn@g|m`x{XAGmh7YZhjo74ZMQ9re@#C3^gSpqV6zTqxT&li?SPv
zp_7YD4126Hfg!PL0t;WM;Pp%;Y@nl=qUa)9la^VYl3y#r8xV@em#bs;rvRQyZb*C>
z!iI(+XZ#9D%Q=Q_+KJ01*d&N<%S4pfS*cMZV=1Xrp>3_hy+_vZ8l6M(ourhQ4^g$L
zXdVk}d#bB8Jx2126?ABlA5p$IXyK#Ic;l=_lL|4GLO+0Fq%jX|N=`t%gW{-wTMDx;
zDl+Idv!I5^s#2m*s@u`l<bk^(p$Mgj>_Q;KwGfyLSsN$?&CNiO_TkoqV&GvZ=7q@C
z@d+mT<cBuIgT#H~lN~2jA*MJIv6d3IP*vfCi6U2qDa8^+cA!M+n`krznOa8;`Wljf
zCd=)W5H|vPvoV1nz_y7<vm#SrbD!w(d>qM9N@holXXA!DA~#Z`vrQ5cmNXNPbb^dM
zwzpsq-H!W0af`*QL)jX)qlTqQqMf8GtE#<n*R^Cv6^dm&-h;+5^jjRqO)#AJ#*WcM
zDG_&{%E%!s)fn`8peVR6N0iuVcF|m_G}fU6lCo*Cj4(0zBLuI!=4NC%%SNuUjj~}y
zpBV9*9oaI*=|N_Hu~t))%!Br8y<VtVY$Rh6xAw_^-7FuyAaFHZPH5kHdft<;<3Sqv
zaOr-%VN=Jj!O@2sj#%5Goof}>Yz~{t2)jFJYZ60<&*$^F$29kQ<i)!;UkIK$NiID1
zrXYG&eCPEOC+_SykXUOznxhJ}UMzLzIXUC@JN11OmLz3^RGCqT+OD9?{7N^LGOYrN
zJyCTwpV?|Z`4>Qq9^}$lWsOE{6%18^0LEEMVoU?i)T~+ianO~PJ;DbW=Jz{Mo=ib!
zpe-)xPqg=$HkNvMp$7nU6Vr)|u6?{7i<zmwRg^S&CN<AuDr1o08R--`*7K#utH(N8
zsmBalj0+aRB9M7QlqvMyRA|dni_~a0xLHC7vHfz%qNRewaqS>443Oxf?kDOJB4JP*
z^9~dIz9V9+n77scbo-d8FiEBCi1GHI-E3GTnjkoQ=v)}eb%&zn!>9&mRCKT)e|X;#
z&dK#bae(8UJQ(aicz+W(eHu=jj8mq<v?&^=PRG1HTGDSOa-HT)zL;T-&M?!nSQ~oa
zSlZ+nFRc|ikb^8VG5R=49kcnF5aQXCatTxZJ`}ZK#0H@^gLn`~O{p0ZY{a953;BQ8
zu%oXEgB*w@iICH~Rka!vljH%HlkDo=8{$V=JZ!3t4Vf=yurL`LP~(-4I4N7RN^-?g
z&S%ULW*yW^?J-I6CQs7dwZ?o3+eaTVe_x2lL`e&)*Pk?9$as(rklk}E1;!zGBlV<!
zK}eyPucx|k=)VJT#*hD`L+oPQ4}?g@s8cw3Vj&BYAYlnStS8ofFOgh)Q2Y|9x0-Qy
zCZFsI8+0zITGy_l%Q4iydxsLnrDpp7>&OE`(K8r}speB6^*!BVmiM&&^y%BjU%Ku-
z2;Dwr->F;IUc8p@XegLMG5p)#;<L$m;u>~D1@0;XQ%qU$wt;nmr)1V&02*7}w+?w2
z-2ymy5<1h20pzk-g7t~ovT#v(*F(NI%`<F4hAHm`#ifc^5HP=fN3b94kt5k_)X#gc
z9gUuoxrM)I@jEKHQWQcZSH)hn{|PJ+9(e$G#ME0Gb}dibwq`zMwE2s9ghk7+2Oxl=
zwlF8%VlY+*Y_M_mw8WG-wLfHjvszEJsz4MpNrl}MQd~@KJ4G=m+;Wnll}(E|CAh`B
zwKXPL+_Wtxr_DA!;aUKV)B<6AnTjKBh2f)7*6Ai`(;4hPF}nWJ@{zZr=rPy$k3#WR
zZ&Jo1Zu45;#pdIC50lAm<hrTHyDQ9*_S4ncE8R<aYaP{|&Mv2~R!EW5zeO@Y*rmOP
zgfGSU=bQ(r_*r>xVP1+)mJ`@Sq4@ltewUgkJP?XotTvx#xb8+04&(&O(Lfy*Xv#<@
zJYuuJDOS|$J(_zV{w=<6WlQU{3pwmsm{>Y|X^Bv#UCJ4D)Mvf*8}<X2s9*rCaDt<&
zof4{{6L&7eeC`K+Q6AV%=|V5UvEs!GQF4SSM{`gT_S-kc9*?nw9$yEB_^V^c!*j`q
zG?`DLxopmCv&}>{AtjusvOn%HCQt$cZW1P6vc#NFDgYD=awWd=@`)Yp^WgcnhzpH+
z_U_3Sjvl?abFZ6(W8u+BFFC|BG&+!<6qo(uO0Ll5H$(>-`es?vqRmQ|;a*G8QwG?%
z)k)tL_3eO3H@PYJq~B_Ff}s=+uKm+HBxz<RTZY>K3`7c=nZc%L;M0t4=)gmMU~$3^
zF8-9xryvs8RKrw31X3=CcZ&580Jq+EbvYVAi*}~rT*LbmNGPFV)6$AL*a5kh`7LF7
zHZc>;9YglF6WnrVZ#tctJT$r80$o&_iDJ^=I7;oG7OqNed;Rz5D7bHGSkfl~B0IeE
zNdN~H(KiDWPPpKefY$>oB)YyO;})FSz)E3=)Tzgk9W_zmsKq&h4<BCiP6@Z&p=cB0
zNsUO#l`CfA%vb}|;~RXTxN!zoyp<=#H?U`NQ3k~&kbNheF7bP1CO9L*h(2JZmXsN>
zx7hRXfeVie7w&w5c@+rvnp^@He4aD1QGMILzdGXDY(SkB9YHotqgNq$$QcpY5|~2v
z4%OGvbBlDqUII31T5Bqk>E<nX(5|44@8Bs$06>TkF9}>a%TZC}XSCZJetYwgj~Z+T
z#U5u_E_=V18iRl;xHjVs3sS*?3ZJrIm;zYxL`iwX!HlIPU!wEt#MIhsY*C?aBMVDW
z)1=xz50F?^;KVbGj}Q;4_F26dD-Oh0;QDzKIsG;i1Lqd)&|!#t9e55b7>J-=(4MEu
z3R)0DaLV#XmA92Dz$A65=yM9E5mDdb{5l7e-;G!g#V(%7DDv@BTl0vlQZTp;C7Mfg
ztPSsv$X`Kawd2r_J?t9gYZ);Ak=&?5kg(^`WHdS<eSG8d$2Yz}@|#ID9KUTGk5und
zJEP3;%zm_dBjeFO5*#Fj4+4tt&R98sW15Wst&)B4#)t4yrMbM8)S1&f!^{*I3tenQ
zeNW&*3yhvEUf<Nm>xKUxCCBvktUOKL%k*q8D?N@U=-9%*BtjoBLpF<-UTsb{a09mn
zTzXS5kz_xgu!yWMs*f*)3W2NMfO^abSil(ihrb5HgXcdOiwl~!uqH_!G8{e^n@Z1@
zS`FKESSx-^`|cqzt$bD5)G*<1Ww40Ea0ve*trX`wS`|}j5U{t%wiNgWhTG@hPyBRc
z*T!I7Xnelcup}~|FAXm@1Wj&eq#R7WBeY2g##0;$2vh$)xoQh)<fKZ3=HDMTgzoQ9
z)7be|o{B@&a*@XJuwLnKF*iz_ScUp>T=t)LY+V{0+Pu`!wsm1}VEbZr3$MKXZdpAq
zrL_M3-aq64H~3OOm7elG)KhDP_U5!N?GB{y%6#7vl`i3>^PBb7_w%-cJPJ8U-M;Mc
zt&J*S{YaKmMrT<_#6|auK5br-RrA-#tH?iB?%qB%#jC6qpE>@`xtF6}Kj#TwMknWH
zI&aWM7+I<Dap`FS_7xAQa7Kmc7CsnQ8IY1~!qQ{V)5m(ABv^VTAbXyk_WSA+(<5yL
zq=PGk`?}U+cpF82^_Xxc;6qt?x3A%=#}#>aoA7k9`_(_NQt%*5x4YE+huA&wRCTqS
z`#~4PQITcp29+rtcYn%5xlec=(}+R&pGE~l4{A=hSz;Ks-<b3J;T=)&9r5(KGGh`u
z5}EpHcY1SBe=}{ONVq_3W)?lbEk7ZdNBi3hcVhI@21_nFiyYz06YGgi?>}PTReHqM
zi$$vN{3Y|o_3u+LUgDgld?5YsveFI!ig@|wMAix54_s_~R}a9oj^l%(V-qIoj&`vb
zMR~y_632pMz_d@RmuJ#WkpK1X3h}?e%2;`If5yiknOf{iJ2o80EELf;<cL8}Y@(0(
z@MOZ2bV8aT94uAD3$&YWp}$h3GKzSak<)~~gL{cSH^|8gPbNF_rj#R2NnIXnwo_K%
zlG^hNiMQMZF6-7-%sG$q@Q$m*0V3ge$9N_-&By#$i0vFEO5Rsjmfc^ip!Xbk>=Jvb
z4Sr19MNH77U;KD|Ve8Yk%lvb0b<w|PVy@?Ij}5;oZbI5MzVWe#x$ZFkoh5RHZ$4Yx
z`1aSG*e>rBKB>-P_=Iii;MoRoL$9P*UN2@2hiLQ);4mKiSfEam)!U*es~(;uEi~9U
zi$bG$Jxs@xG$oIHee0ZX#3xDMT1JW>RwvNsITTD-n02Ceq<4)dWztV{|M}GuQz?Tb
zvWmKNUdn1y7b+2zTVr69IA7ap&A2@v88f+ACX&<?Cc!QgZNB>jm&jPGw|nbj*Zuml
z6C3!Q7TeZ=v-Nomy@H%R1o<BfnS4?(sz*7Q2#^o$Z?ysvCrUgRHA=4TVS%<1NPnT;
zY3wC?KvZ1SJB*E^EtP!j!(3gZ`!!essMKbUhU+k=UY>98G-l%eg(oFgj%B-;S%tQ^
z(+omUoZ>JyWJIXP>^;*Uhe4D<c&nh=8C)@Fsv7$EhW|Z;Q+<6RsY3G4&+Zk9pTFqS
z=H5On0|zUxxAST`NHkU{#gvpTRa8x$Z7<Jzc%@v*PDi(*1;6Tv0`sEEr(R#F6TjIQ
z_z+#$sVdpNm~*gWbIikGWlK_Lr%!HkQH<&ooxPf^4_1qEE286KBN&6fcCyD5k6b&i
zZf*lyPF)3-q%_BNeO`LM@hs6v3BZBr*Du#b1>&|0+wvoI!`UZ)Jo_{Jn7-N7^hs{V
zzZ>5xdc@qs+T?PCFp-j!ns1!_%$MToJreCb?jhX&Py46$fbj$2Spw0^WR~J~&nMF~
zD8jX@^3RAtI;C-l*H7&Kpz+5yKdArz;|(9jxj8D!n{|{r?d3|6byWRv(eck)K4(G@
zFEsCaf2zCBF`CD@Anwc3I_pAQ(@to5aQ!|8XPcJEp*iP0=hNl@r!jk&;7J+i>tu1(
zXsRYomLsZD6r3)-xcBP(Fxuy`1Q2unbDA`|{uPo7;kjNGhm$exB!{d5-zJQk85Wwo
zf%F0kz(p{OGMK+-7y%#cbU(#8Ya2%(W&rrX9f!Ai2y9~6p3)t)uxXl_<$SER2WoM%
z<)@^Rr?`L5kNyZTeQ*nE{O(7X@d|*3_>d2|;J?qWcOUA$`W-R_PDgG+k3l9B6|kt&
zC>DL|B4Lu+b(P#_WAVFTRNMaqHQTRlTA6hT$>02bbxsGF7-x9L>z`*D9~$n%YhcQ<
zeDRdHZznN4&S!K}wkzvkmXz4UpK|;q(IP7lAMR2R)^io_%*(ahz`k&|W$&1pN>McG
z^m8UE)(ortN??;36uUbX5U2k7H~jr2NFit3dHfvbG{^7QGn>=u3o~b@2B(%5!ENl2
za~lxMFoPW9faqqB169H$vLG9%h^sV_I;Y9FfSQCkE|W#nV3Qf&aaj_{B9!N(ep=^G
zAo}S8OcUVT#Vi><AR4|Uu9$rUuIZrTdVfS4ucPxo=bCPmj<%l8jQ%Pxr~ANO4{tv+
zYjmx8yX+z5W9-K}XY+5jJWbRPXrMp>aH%<(7Jb&_0_@+p&!x~c&aCIxl*$HYkH`0v
zE+;Pos?Kh8#{m_q<8^3Q%FzzLODRh?^g?m>7{(n9$J_r<vFiV{D!l@QT!%UueO>ak
zvD)F0TTk*!?QrigE=DH#JV8Ej+!yDJ><e|?3*zp@#ze7JRa6`OUk9qShU!k*lp(~`
zT{TmKi}fM#bh3;i{CwBsRFb^I9I(|)89SC2Dyq&@@m{sH7QcH|+mw=GzQe%En~q-U
z8|~SuDo>58D3K5`K1em7^FGC7=X0h~ZA2+%DZRua8>cez=T!r+9=uZ1yE9uwa~RC0
zbhZz#9zSt>qkgD>Vm>UVtUmu$E0A@!(8k!BG6rn&>6qgDea>v>Xzx~4S@Aj5>QV^+
z*7dJcO|P8Xxpb1Di<3arU`;3I>n92*W`+0yYI|Q*dy(`Ss+pB8%gk%<$pWEfi1Yy`
zPy$PFWhXm9bSZif5Wbx!3742&V-N4Q(wWwxXne;=FHX{yVbc7fghaz|2SyLC>(<@*
zvLehc`YIDE$dlH&UhNOADm|(Cj&(@$-?{qHImW+<tU&N~K!qfz{!g*$y(?uELsqvR
z?tWT1aV&D2U)!*f9vxnT>18ui!}QoR+Q+1#{1i@>RLIRK5OH#4dz}D0dCEAXv2jT;
za%!M%nktDCa3dT1w$5Qy3(ur_<FkD5sXl}(@NLQ5l<~D-?fN+Tu<kzaC2mLsAQ<eG
zLQhA}(%F+2Ms)lvwTh<d9R#q}J4|J%g?a}Dy2lB|j1}~Z7jP**5MogG->wi)eae_;
zNoPe40Xf%Du4Uv?qD)pO9I01hRM<76dJM*Lsf!~?GE=X@Hg?62^i!lwCvG&T2;WW#
zafN?d8Upt1HqlO?Eni*@tY-G%-1frz$2GTMcbskP!!I}OY{1V~Rej|oOHQ?n93I}N
z9-1AhnvRghq$bL9GJNZf8BId&wG@9NwS|TrCDYrF_X-zVF^8W@sP*4GM_@g84LN-C
z`X`+?ub_HYul=Jv@n&fBLW81zcVu|v;<~E_&fq&<0PyB1<D&8E_ebIc!^eZNI2^qb
z;!4YEOTacc?IC@SvsYD|nfTH$hA8u11#05$e$s<c#)?i0cX1dcI?!6u&nJ6-<1pBN
z6dlGCkpc}q2EraDgty1o#00r;11ryrbc{6<90uF_Yyu+V0(^{{O#VjC%(GFk!DxO+
z-Q|hlm<9AD;7uw#eb@zhT?v0=1Mjyvqz#42OCz|J=H2Ob=`A>72o`8UAT$mnDXd3^
z)!523DS<|XV#K(30)U%Y(Rl)YvSgrpde{Qo|G0)PuRmEH{CV}%=9c!@+|bY4Z?}g!
zAff+5U*CPixXCzccRBRy-S2YV=Adw^ZbPw7fj$(Lv&6SUf4^@<>xpV4mt4V0&{XK@
zzJ|O+N5^(#rRLDHW@N3!kc-0;IHMksPm0B$o6$o=#n%tsSF=cYpp}SqFi*8+nHP9T
z+S_YY^lA~+u?S<He_mpt9kDe+saEg?@kZWAL=T~xG12*?di5{Wv3@|+t4Y7a3@rI4
z)Y`%c?&unH2Ts!5R)^55Q)zCO;L)AvTF=ln(uo5_@6^8z+t6*uwcnV6um%L2k^MBN
z7m%Vck7q@<A!6ej`CG)7*G)2w`h+r$Bj1VW8R=vcpjpA`<gSv~2jSbGnnrD;oZ|s=
zH1gL$6oP~)DpVf~3h`A-7Thpay(W<RhrFdjSu;Z6+7jOK;D7MLw+<Vu^Nr7EYBSvJ
zx|x_5jfBM>DaE9>nlhth`hsWv?U>~reCk|%);!-VVZLK%^GkMwatpRNp9dH1!gJ2}
zdKvflt5>VzO|f>p^5}YLaaXA+Vy6?H(*P|AqCW}s8VU5Duzgl(s&-;+RX({q8kuw3
z9ayTWhv4&NCE+f~nMBvZ0xnl7b)s;1PUPxpKEJZkX?v`bL}G8-qa#!J7h~isLzK^s
zNy}%j((+>1qC6=`0M0kM)7*Dj3Vq??<m1Wk9MQqagQ9{Ta}xWqH%;f7<`83DyCs@=
z(xOZ1*yh=!{K_-O+s7+S&4ISdIFQ2j_c54^Cz^2oW$=|DXXz%XZjT+JJ>$}A;*nfZ
zPTiH0LoxH{5L<_VsgXK(%>^CS%jyf&r^)!PUy%((u44|&VYYQzvRyh-;q;A!B(U-2
zIS7w5I9{k{eK~C|E_o6nFkZw6PyOQZ!Vt(~o)}zs7Cs!~g^$R>MYxm88q+^5u6*j#
zxaPUM<hj1K<gr#`t?^Tz#pU!*0ZtK0BZKh>!|~UxC_-s$I1ynmk$&#<?Yk3ATwl9v
zwvQ;UwpCVE-~Dwg)=(Pn(>!_OMu*e=q8pvB5NLb^37=OO=bG0n8kDq`wAf~{(p!*i
zZ4hB*F;7}to8yi5^7Kfwxp`q!MG;+LSH1CHWzBpU66SN%%tLyfH6S07wYtE<&U^sA
zHoVhorNh$oA(^olUu*}KEg!1ZbrfffG<hwWb&i6)h`&Ig44i;ehA{l1O3$kPK~>GP
zP}_z%0KL<?e!g#@FRry^{oFwB#<{F!o~$la6m#Y!%W6ewCADD!c!wtf-~k^#+S~F-
z#}Q!MHu+wodBv%n^zcZ<JriFVAymkxIi)^qatoKWRit^^bA%aEk$_(Bj=U@clA$?P
zSTBZty4XFFk{S}=&BMi6#HsNVrY&K9&O}eVd!}ubt4LenNGRzP#x*^_6qjexvR10=
zLu$=(G%wme$-m{0k1QJkJzv@|9TVsr8)lE!wle<zpU;p~410QJI^HBdfY?rvU`X%_
z<=g64UdhbKf`Fi0q*{~=@-7O*w4<xS1i7WC-<rHD+J?dod4B)v?KtlJ(ATxW?!OZs
zf0lgpthhFwvPp9|kdY0~@N;pK*~)Y+RXsn%QV-RY-*$-#eImN`a$I@&`P$Om7spTF
zI!en%vT{oL%FD`+u3@Bu_|VxF7npa~kQtEACfg~f@R;)d^WFHb5>QD0kdK+Tj~}Y^
z2Se>u@^AfbilHf`=St2MOH)24ucg@?glFR602Bo0y&#%U!lks3jmd*q*|*AnhT?Fy
z?@3~XIKz_ypW733^vx|pS}Z7rSX5GRn4MvW!<BcG>;Bz1nxh1rn0th6H9Yq43947T
zf=LDl7l6e4Hu>yV+wZ&^7Yz;2`AF_dl=OT*uy^twD^~0+y?Ptl3kpNpaQ8zA-JM*O
zat_lJ{<y+Mp*FHi;**XDjap7GS9W+8e$|4B6xd))42DARUcajDJ3rl{w=sAQZ1;no
zt#0bM)DoH+oYYvEUr;XxlgvWWc1+$bjpB2NPGLTchQ%>x`3wpUo2=!$3N$F?&3a0`
zcDd4#y1Ply|CyJYkXVKeaSqm%aP^=#n^Cxf)iO_uR&+!C=a^p_c7GaN^nbhgBwxmS
zA$a7Sd{Rv!_S3l>mj6J^qF>2p5+R4MAqbYj>uISi<e@wu>T-lelhrQOHGu<+KsTu>
zA2Ur94ZB=&TTPq&S0)5jL?uJTAp`nmoFU%nN&;txt`Db~&__rMojL$+<TzD%*3`P1
z!jDRrN&{LEK}$lt0oC6XEYG5d%*m~KGj9+{?$!%jfjpZ^*PxU<g_hclxJ7w@q5I~P
zHGT7_C!Sho#78JG3v~Eg{wN`Y5GvrmZ$aSmw+P`2+7bZahb8tj`Ew^~Dqcuq>y1`-
zm7Xzm>TB>%I$z6@*anl?ox0e48ve)yalDiC7SvGXjQPg`fbqN9n0+cP10W9^AcC#O
zI|TrG3L5pa$(Pp+?8iTVZ^O^LZeUSOf%+u_ADbC~{|fRU#M@NwXwIDXR*;r0JucPA
zksq5Y0LcOb0bL@d92GfoWvRv0D#qCcW8maSAi71Nqr8b&M-jGVYy9Y7Tdc!iW1=J~
zE)0uy3?sNlIY+q@f*g5)Xe9N|Pe_sjU68`yi8521r?p0+#(JHZn}zJ%biJl&-;@Nw
zItx2Lm#%H(SeDhkpMW{@aqsrtxZLd_`*X(SjMt#RSUZ|q)0Zb`b>5bnKuzf1&*nBW
zF$ms!VNQT}Q(Dr5+QN9VBg;;e{E(>P-S?WRzfn|yTrjCwURVIzcTCO^^81<}m5RmD
zseZn_H4b~-nN8e#f2{6dnV45x^y&Ca9`y-ZjLO7B!7px`*G~EM)Sx&o=6ZH99X$x4
z92yJPLoFWF>q9-#-AoR6jhBjb6DJo+3vNFk4i6?js!%Is?CxG-Bhxvt>TAzx-&4mT
zvjXt9Cu>2DNAD}kvLZB>Z%9G)aph(oA2#Mi>ESw6HB<4#U>DBJc8-+L_vUuj7E8f4
zm5&(G5g~Z{kx8R9+ksuO+8^%a32foxQK!kOg>UDnVht(;G{?&wCj>l~3f}yPV~xeS
z4l=XSX?ZB0)E1IA*3CHV<Ws)zzOQL+8;ZlmVRoH3<eT@sOvGKNp_v8Y)IiFQ`z_sf
zfiL&@z_FQ{s$)yXlB`+VqrTj%LFG%#z?Ue(Gtg0Bu849tZE9~%-h7&0YRSMl#7T{Z
z@GY;-6(y4&(d{)|w$%?6GSkFjZAX8Js{9_|cg@Qsr9ZZ1c_MU~AusERcjT@J5`PZq
z*KUD(Jj+F)a%dC@nd0peN%!-o{q<vHMh=ZCAyd8MDp`v(Hj34uklrO@SvTdcV0u?Y
zXKU0@*6b1Q1d>BoHzz8`$VoJUH)#<%^(F<|7$>-hEgSF-!0I%sV{G(SgF5>obIrIP
zA67HMw6VoznF45(jZZet@&J19utcjRt90!6go86>Y@x^~xcWx#IU?s{qu*98bWY}^
z;W2|0!Qm?70;xjDZFA-SQ~O)!Ds`35Q|Z20NB6m7XW<l_esiffEA~zeOuiy6>8bXQ
zqDLya8*A_61Pc9<FJzYrwDDY#J~GQ}Sb?5`0A793nhDOB_%Ma`NC(w7)ExnQA0c2$
zMd}Qf8kqi^Ux33BpsWjlZUqTK0qrjW;r7=Iaz9iKJ9MTQSisyw=*gh@YZ#x*pM;tI
zrx&+K5b{_HOI@Rq8u%)t82zgmEr)QoFB|nYs{r8++Q#Nze||am^p%BAlh8SK)Qy)n
znT}4^nJ-y4oSd20zqGq7C&!ynw_dw!sm!CYc9V4;w=riio)a~o;;bQAQ#RlNv%UJa
zh@KsEfeju!hGeIEwT0a=xVQ(#FffAgMN9!8ps*k>9U4;$SH)F>i)ZX#{rPNeQ%<q#
zEg+Bl=x=!sE`ADQ8v3Y&3?9X3NA~n~W+;ouru`M<twzEgA6{Emey`l4O2sm%JoMl9
zV>E+p+Do<nYQS9mCvhFSu?A{63Schcw%tOoi14{oAAFWKKGhqIA9rV-b|{us{HLoR
zYa@bc<Rd5&{{~hT;nDdcnu$~S^JfHP=XQzErO~0d6=vXXufSF<Ki8qSd0)`v=$|u5
z`u|kW#7iPO2_FiokKcU@xcqE`2}zVNTmp<sYH_z<Y@s?@N;K`$=oZk_oWqYoBcT>P
zBO?yV)LbAA71Q}E2|(DP!9=PQoidV3kVIfz99_>{!|BTE-tt_S@>;mIe$AX=Q$?_`
zX@~Gtq^J$Sc=a0fc3M;jR<`%rYNsuck1g}*t0~C9SfWKI@hA$3`_VeY0NwqeZ_Kfm
z=1P`kBT)L6)wjS-g2Uqa#_yliO^ee9sZt;xU_drp-}vXhzw5ANl|Ve8fM~cahiWbE
zhG|7Ri4ZJ;%IEh&tU(q2VMrodo&J%lFgw|kLaNCL2_dHhU9v#YQ%covt)Zl*=B7MF
zemptb2Hm^hjf6?I>ob%7r0*L#6FqJ(3Z#1sBE>p*ZU}Kq*7Jum5vXxK_r$RAUADsY
zc4F+-(iU`QmMQpk4R~3?*0ML4b-!Cb{&cL1dw8O&;nvOkt<N%RVAY2%pBAofed?2V
zdk>#%t9ScJ8;4jz^?HLFPeiAtJ2{i`S8)4jf{|xw!4lZbI&{=?yqk0Bni+V|U+>&z
z(AMX|QH-eQ;rC#UT#Mw9ERsahm_B6E&W$IBN5?O}cX{RDY+KpA3*Stw_sE{SeE#bE
z2KHrx_1}zcxivlmBQ7#Vj3bV=5BJ%yEhTw=ya7;if)$#@)?Un$jw7fig@{F}1a#Pa
z0#_9$nOK$SO$cjuzg}8?4EM)vZUL?z$JO{s<<lE;QJ^Q$hz11Zp&4ehk=rhwNa$5Q
zSFXk}0U0-Ei>s3L?n;;B-uKyjSBhbC@QIj-Lq#oVu**4OR_;;8IF0{{Y26MLZ_+)u
zXa{O=E1_fAGjpx&d7m$Y+S=ZKM(-w&AtsOZeZB;#8OQr4c3N8zYe|Hisw=Hpnu1s&
zt<T(POu(_KVUu$xfS{nKw89MOtEH$il2A2YMuXLn24O~+1IT6qi`ok)j#wsMr>a_X
zX0f~>n}Zu?UbITKC`4?;)k&%hLyaZLoW4_!u*OQu`YuC&unl#Zo^pXiLp6QIGS(=^
z_Ff1Yo3zfu%9P3WmB^>(b3q`KXjbDE#ykJ^?=N%#2do1CM6<V8aM3#EcY%zP(#jH-
zICcw;3RH=5IrWrfi2}bwV%cof(XV`#tGOg2&e#q#LaPOw2pUL}$OH$ejT0__MmB4~
z0Awk4{pM#6Vt!`|0Wdtbi`5GcXKEk1B}f}G)TeClk^)5)FaR&c?GSM>D1IA_tEjep
z4$zE6MFPZNg%&AMXek5#yeq4-@ng^grDk~@5^D}R&)5#QyFc`nmfaw8(RpPpkr+@P
zqtw!lqt!8FNE~xdqgxeaSp)?|*{bbS6Ktdwc0+FBF2e279E6Bdt5VwJ$u6j3WVW7~
zV(>@wr43ANr!_&xrXbNY%a^|W=_8XKN)PY-#oYu$QnzPk-IH$mQOktWY_0}NN!rW!
zD+CH>MU_Icij5P1slG}dFWdp`$C4Xxj-`Uh!FG;4L@T9TVBH;ohmwqJa#)`$071K=
zM4rS<I*ULtD9Hv3652}CBITITfs`5JN}Q<UELY6k-szx0G0l-LE+I>Kb(T82ZDAft
zb{l&qx3kH^$)XD6ZhUgyw9E@vNE%(nVft7$SwBq-Oj1W?&`%Sv94NU^T!6k8O_gfO
z1Bzj~0~d3t#{r*o6cM!|Il#?m<6xeL;(&|{p`grrV@qa}2_Vx|VOg@MF@fr=TG;1c
z-MEy}u-F-y$ZbaHqc18*$LUQ%C!bkeeMT(DwvXIQXSB0NKY=|OF!6AHQJRZBi_S`6
zkC&=jT;G^7FPogN6V4Mclyq{tv_EK7$=$8|1+~+|3{7?w1%%^UDLl{2jbg=}zOd1<
z_9G0jAIOz^-49%cs_Qgv>-r+DzVFSy{{I$L=c@<rF5hxz0HS^xTjIs4_8P$BW3#X9
zI!umduf0Jg)?C7Q^j2%O^CHi~#>KdI=29egl*D^LtfFf5RwcQtAu$X|s#1yaK>?T3
zRg^1j6gQswmSWMC8|=3EgmGI-Q=agG1qIv=c3Ov`@fjAA@xCA<39iyKBjtIKh?K?@
z7Z)f+*4M=yQsy9|nFfNeTwn#V2$TSdkYPD?;Q8Sbgm{*N&WsWP!F5Tm@Ch~v1Gc+3
z6F3eYQM3z`1vH~r+!q=v7DZj~Z7h}8p2S#EVHTD*m{UHvVLz#tZm24TmIW>jW6KiD
z@uz68TVSH%TUDsduZRStdM-(`=rc(qL_j2RIPKl-Wi#7+Xpkzvp&>)~2r|^uVyv|o
z2{MT^ydG&g(5?5&sUjDu8@;=%zzFI?=#IwZfb|61Q3bRG;u2wgrw&4)k{?z#mn%eO
zF1#>e{jDZCff$Wdwp>bP8iIC!t#$Hc&<{X55LjD+zuhpgjI@%ll^Dr}N<^0fQ0WCt
z0PqAFG$5&<qSwk4m~BTnk~sszwq(@IhyakWObbCXn&vN#!BG;RWbHC4QqMN5-*j2U
zd8m{FByy4w*Ep;ONFnXVbs)Lvq#&TBs?R7C;V_yBv<~CE3#x6qy6(-Oh>FRiI)k`v
zg%}xmm@?9~N$bmkgIGO3(+&^XnV3jM<+h@9&cq0hNvBLFm@y(GHM#bg^u&E2rLYAT
z7ZwUc@WNU<LE8Vc`-|T`gjES}+;pjKuyn1WtiqHAYP;PYO=DMQE}7v%HnY<O-#=*f
zw$6@Ay9bQ-GOpGJMhI5qR!-LIvT%ldDM@s-u**}duo(^);LG6>q6<_&+$m!Vz<yxu
z2tbKV(j(RqTiaMJ1YQOdln4dr5i7D=j+YPf498cf()@Ut)-miYI6#PobpL~v|NdoL
z1%Mhp5U_3uT>AEhz?w|aK{zQdPxvm@Z@uK7{s+gOzFn*QpfsflNx^1@J)AvSY0a#U
z{AQ^%tLmmT6GtU-C1_YANsHMXMV#d$XbGRS4V%^VS8y$mR?oP=mt?&@m;_n(C6K0y
zo-M?OnW*)pVzTyYD=+>w0FYRdO+w8{KmawefP$=y!8a@r3wmV)Wb<abXZE1izdsv-
zdqlJV=K)QKv}*^5NH5fqjG!WbK#r$SLvgW?x!24ErE^C}>7L?k2;~V7<&GJf;W{H!
zKn97%?6s$?ch7|*MN8OVt&7da23i$kGhDO#ZlDJja;W^!(h5F*FZwJub76s%*qu#<
zhhvZ+Ic@uWCzfa^N|8+95T+eeJfr|?!%?gw%y17&@{6&g)^VyC2GFiZHm&E9pLE{2
zah`LdqQ7ZilrpL^ri3LSW-&pj1`||d`$aqhg29?7H%L0J&zJ~DTxkj*oud>*u^nC#
zWnn>po6QTriem{0hs4IN6tQJ(M%tBz&2q`muhJ7bsBTddsnGAWepb_)$#)v?ey=wi
zF)JU}@P*a!NU}8ye1yT~Z5jdqoVQs=1o{-jpVfVj{|N%V0S7$xixSqU6l%w(@KVp8
z$9L}8dr>5I8gC7L|7WM;4dUml_QO|8$eegHgO6jjtl_h+%62J-@zoz6m>5F?$5>9N
zl^k+>t44rt3b2h^U*VF2Nn9I3%%;Rb(rUqc0+J%`oO<E1G%5vQwGdC47nQ1uxKW_c
z*c)WLp+D>v$5a=`@~c54Fs7&Tew@iN4|7XJ0k4o&(;sN8R4;QS>#VEodg-LwJ6+oa
zmkT;uS1JH&DF{S7tT4>hoj7iTk|9761oP)8vIR0BX-2Egi6O~-D=tCB5~4>KEf|){
zS5E05!aR*cQi^;wgc=$oxijB+9IsTuF$7+N15*$kS|ip8XY7CG?<e!D>W-$g1H5G#
zEM2Qpw_(Zvbv)i5O=DNT8WR^r9z~x%GDmw24;b0rv_Si9zjg}i-gLg*zD4Dw?f3-Z
z)|5}U1t(cB_8kwLOW|d(6WQxDufiot7_e$vfh>qW-UDL`2_7-UP<KA6$gF`AB#Hy&
zR2vL|kU;i{2j_xfJI7J^vq=J=Me7SM4v<n`W*{M6CV!YG1G)kI)fG3<)!;w=B5%%g
zwzSOkS@&!ynci^o@`7sRVJrJyOU^u|WWXaS?FOWtx(H|L4h)5=KBvP}M~YW#(K5YD
zsEI(C(nt;l>^Y3Q_K4Vw;;UIvb%4QOHJe3oUC7GvVcEd7T~4QRq3N<9>BW3*+puN}
zZ5@wU|BVm6U%e-(E)SFkmS6*KDZzCG{|B>Mv&XrV9taLl(5RIFz=i)q*8i9-;Ne9;
zC0Q1x7^ohMM$_u=ir3fw`1ilQ3Qb?NU9)Qq2>Q~xx;Pn41JpAGaUmc}6(fSWmrUj)
zY63c>57{g)T%SjeO~_6BCi{=jy%5an8^`xqSq>*MOzrn*ab6108PClJE0lE!xk|-E
zrsIdN2H_FgQ4CuIpqE-}r?#05;M0_~c?NpvsgjKVRCIc`#K!sDl_$<G((cE5=O^CZ
z02gv_acLk|JhNl%>28}GiX%=iXBSweYy);QJb3_!hjq%}9Y1-}8Qeb8_)<!!#cE=I
z--lTLV=(tE&W_A*tsiUxK@DPG{|rG$z;ByA)TUY9xqh;`c5rw!#BcG0VfAo$6>uXq
z>#9O8p;dSV5!Y-UjU7ROnXf0A=CBX5NXb86>|02r8NZ~T-zTq=?8BiP;A6e?gy<6i
zOe3@0OJ70>4teets!B)oK<6b&hPkYnSga-pIyqGw2~FDB!+gSYgjxwvHY)=oi53<*
zM2Lk!nU+xivJOzQqeEVmMPRsnE5G8n%-8BwyRQ=vs<gW!)OXpD0_rWr!S${0e43V1
z_ETx~Za-Cq8LPT^sZI^+#p2owe6%+^YfV^MTL-S{#Pv}X1~8ohImkWe>#+%T1q(F6
zL^_dv70J#XLsCI1%9yqtG3s^uYX-T}DQ&fbgrgtao2x|1XI9Z+6%`l0-B{T+3oAwp
z)12s-nCuG`9q%1G&C$<%R)9V%x$EPTO-d7Kr=Ht<Hn!3h5#_;cl2Cxlc~D`$yqZ9H
zkJZv{;rP(fRa%%Xn{1qSzlMb+m|Skt#%=b4NnSdTTeS(crDpJn7{G(3BFQc69|1<;
zMwB~E2tD#y$4W*>$8~B`t*6(`7mv;DhXeT$GjwLI$m4=Eb;w@4dih24E({evLkYG)
z4+idnFP5-#>YMkUIl?m>IAeOo!Pn><J8!)B9NdJ<_<3dd3x=&{bzgkaZfp7q;AO*N
zq*kh-#{K$eT9fZ`Dra&o7g9+g%GN7SylE9zck*@VMk&vO`uyh`?%@@EFMqSS8A83d
zAP@w!;d<n1<VuVf=_w1zI%E(n6qzzcK*m`}4z_>K`xcj0#*?-6H_Q7!@TaVi-(&4Y
zOB`>pE+x~Ngl7z(+nHy0q4SILPtm3F*@ZO6Ma9RL>#>b!yujt@>01f54tQ{CAIeZd
znho7Zqiv1l0a;KJdNp|2EgT`WxRRY)2t4_a%MIG`&Ew3#<BHpzC2m)F@Bvr>SAhW4
z*6JQsEtwCUYGy{}oGYily}iAA=D)|=yA9b#X8e#^)sI%j8);8BT+ulOWdTzy(r#Gw
zs>M;V(o{jPg?BJHowUZg7(X~mScpEuqd5YDR_jF<#yA6CO0u5yS-&@+TmV9)#)?ps
zH>;!?J1v-Dgk;^c@?|0$nqFHySt2J3tuAV8Lo522J$h1<_;RCynOnop-lS%vt&Mmo
zu9yvov9~?=fnnV7W@B5@Te`>Ur04r$z~CYtN3g(Q$sCg_51|Az*!FXAas;P`{rPSw
zqhVLnW3cbb)>3v(18k=%;uI0EyENi29K3BXdMM**0$CZvha%^&)A0sdkInHrH3AJ0
zsqw<8ov}^I5FFZ`D7EajS#O@8<#h|`ksglF&thXkKSXF<;BJfd4+oO<F5v3u1QZR$
zAJ?*o3XM;n!?+{^KAA-(7LLPsM?eRZosQ{#WL>gE-|McFrj?f4(Ft%2_peZY6Vla@
z2#atG8QK6-1BPYROwkPhGK+L2z_xe}K5Yv|JxjwV&D@L|b64bE_|?Ps0Vf`&aQx@9
zey5LBY2T}D+i(wl{(%0#1L%YMEAaICfP5Ux2LL-3qd(g`s5mf>0o7kgNW2$=xAZIw
z00-l!5duV>ODg$9Jplu71}uZLh)zg7sK1MUMt$?^U+mS6(d@P`Rdfo;TsT(x0SUD`
zevp(SUfDAo*82m<9DYR91{CsA+_VfK<pX1gz~KmcC){N(qV9<Mhm1uUTJK;Trii(D
zxC)`%Qm3XztLRUB4p9z8>);_YCZi4#NMTIt&01@5c1<p4QxREI&isWMS8nZxs9gZO
zIWfHZr>J>_D{9tPe?vq*ZDyPrZB@Ly@`9TzEg?dajX7C3Qldfa3N34DbuoSdeYqJU
zmR=%`gNJ#r=eYEMH2l!qi^D_5@dq#6KUH4?afM0sApaB>+)ii1-7$gvWNdC;Cp3GC
zEHyknsNix0Aw%!Zv8IDKuY=<3Cj!qC#$jF2)(^C3CAXaf#<M3tI8Uh@ZAI-<Cod#a
zH+{Y&*EIIotBkwu1RX9gUGG-_+%F}h+*1*WcPTvHNF-q;&=vzhH%Z&%P)8!bS(9Do
zHBcbjn4dZFuJaB;LfGarnXSO*3hyPQUgdsEKmT0;=-t9^;?K=zgjdc!-?MX(ZjMLP
zL75l<(B;{<TI&G~iPOplCH4|(dh!AFB2+t*I8wT>(yHSnVWnw_D-zl*vUXsorba9I
zl)~1^+v6sZ84^?u3InUsB=MQR%hgZi@EF9MgfUsSeQvgmhOt7A)XM`x2)vxnO~a$P
z&;jvC3&CME^7wWS)sk>+W<b6cU7`cnsC9Gy3AeVxht%rCEiJFD?+p^(fh00MsNW<w
z?{kT3pTq&Y19IV`c2A<TCx$xdQpRzE2Y0cDTP$kYni0_i-Gp*hAQoS4007jNxmZ38
z3hthETSd@-{UWd1muivCR74*{=2+eyOF?(wwv?Z42Sv!ANCsF&0foUxjA>^DN=)Jm
zuwp)Zivc%=*M2%+n$KyI3avW?$4V>-$7hW~s0foel{9UM2D&p0$g%(zjMJhAbd)io
zQVL=)`>>-hezO!8iY75ZEevxju0Tj&hL+J{#Nm`m9jrScZ=|x<0h%eAGQ(jK%P9(m
zI($I~-bwnJ<9{}&;y(ZJzK?V?<JXY_;om?TtruSoJdN<Sr~}At{Rc9t_(HD<oiD*H
z&Ep8hHRO?Lb_jP+o$U=~YdkDa%3n%(@qW^TPRHAU(>`y`{0up2&iI3DRZe~cUzC~~
zy;t_Y`*-u>Um(XHXy+(@EUR4487JZ>*d(-O%&)$qt*+HmAoOS{@0PKv7oBZe9=(u_
zQNJY9Gq|-uWCB2`lY({WsP=x;RA8Xm&3Hu<H@ZKpZAa#!9Ph&e7@FlZRwcHMo!^e_
zBn600Bo-FX!{}k13#({lP_t0F9^<LhQl%)lXys%6`ecOXIH+asuPNOSYkI2#Ek?R#
zR`2`mb6R&L4kXX#a6?=>`S4Y1U5#{Xr@Eo~)x$lwvvGD5KqQgVvNDugQa7bz%_DuT
ze7E^Jt*fY9pQAcDF4{B`631l!XrE~!0Gq$qI(?rX2k7gA!<bdw%{j__qJY3$62jb<
zoKLh#F*_hSLV@eq1M1=d-$``#jF-f&Pc}B=+!?q#mK@`;RxD62)V&KB_3yUb?J=sX
zDps?e3|3IP)qmW1aCq#PC~JxK!#)nXT@JXG_h^D?z#1K)397xb7whO6uLi;0*}cp~
zvfHMv#IhCpxY_N4J%6E10Gi|}d@8l_<KDgQy=%w8hnrXMN%o0(_@4Ch(8KrJ+W(o<
zGdWlABj*=xneEhGL@J||yWK`8I}=xO6%y<G{1X6tm4mc{4h>@EQ!!dO=>Zomq0`(c
zwTs_5Z%AN_JHaETSjpn?gfqI3W>3%2O+zzxyN=xECTeF|W|1q@+TTWN(PDA|9M7%V
z$&#<B9e)rDJHlu>Kg_INBc;UJhr3T89%iw*_4sl>-Q%aHXLptHN;rL)YE!)M&`s?!
zDjV(Lhk3bc+zzwDxpPR}4shUSWmLY2*4EmbBEZQIqM+9_g$M_4nS*PV={v(}pPkN*
z%v`u9=S4HP!0NBIU$=osO*WhP<nElOK<b0d9?sg6JgEP4Kvh55b-0zD$wvaFFOhf;
z?W!CEKVg~JurQu%@4N`PN3ozflu6kxT{uTQ7NP@DB*#l+`#+SJ;{T!$7tbPRJ@rY%
z&g`)Z_tNF)`o`8Rs+v4s{Ju$Ds1H>mRXS`s3D>Q22Y1wX#m8S~N0&6tJ*~*qUi4@V
zb<jC#yB?>BvtfrQM2SN4q}Rxnh&V*68T*>{0I_p=^cz$B=*`en(F5qr)k<|YPqRsG
zFTp*wLqKFUcanAlMoj-6vAki<h`Rfvm4<Hi&I&<vt$TpwN_EN8)Ut{99!lS}RJF-5
z3g;}DFk-$_u|8vx7c~(R3dD8CVRHK}!qZ$XhYO1ZS)F;&Ep>DKo1|OM$?oy=11FxH
zE2IAV#fPki*bubFz*-pX8ttf)l$tTC4}3#J{L;gmG^Xmn;tBw35DFMBt$-g{D!nFb
zklda2*+P^8L1D0%_?!+tLYA6SexuV-R^Nh8O36+68*T^v-3&GtF#Nu6L1H7*weRH`
z<}0kf>Wk)y`>UXxBm6c@wwQePvL#3O^aRVY0xcY|6i{Fplky`GV?(y;tj-xG3*m+K
zScWTTW9#EpljH_-2YQlYh*?R?s)9r;1r#L0cJ<|Wk+*JipIL`-&n#&yu;-8P_YU4e
zhME(8f#pi~CHSD^%Kt*Y+y6WG`Cn>6{|^5S{pA78__Yl5&inlR7eArl7qoMzFZZon
zYfpi?d_-(a@8Ox1pBGFmubQQMu&@)USRzj9O7Dio+(>Qpn`uYAqf^z^_KIY;>}>^#
z8cp1A`u3JwOwD`5e71mRX>PohF5RM&$XHL($+4*;O`#&<+(PayBvvlD+0LW*e(nM<
zi1JT%cEDgq7NM?;H8#^(G}z=Ut!a;dUR0dNGVM*0NIBA41Q7?T=@yrIT#srxwPj&P
zjaJi!s^3_JL}YJ@rDfYtOD|8%shbxUNh`~P=3ty0?zq!c`Scrr+?LIIV_g9Sj4q)j
z-4K1`TuYpqHiqe1bYES?O=G0&gfd}a(j?si%^MEJRzU{MLC2ak2Nf0mZrUpwL`23j
zM;EMtEw-vqkz|hwB&9GnYV6#WVyFdzu-B$Nh+C`}ARICz;gwajv)V8jD#oN{;~Yhf
zQhGxb2gV#VM+(*euM0T!cFnq`>{9uX(^-p*^0{L=QHb_fAv)m2zMW`O3I-)hvwH6;
zZ&yy2PcC%7qz);;4B|-_jUb9U7hn%szP&Z7v=Mz*K2Zi$LJ4=;nA2s|)kTSViu^%N
z&6S}LIW2#{JfTeBH^Hx;stE5iZ$JmhA3<NjZ-x8hUr8U?_VHg!K8bqwU`o9)E_^Z1
zt<co)MqdB}+2YRvUi>RuPYCIyG9O&BKe(6=!OS3Gu=T#g_x@Nn_9zU|g<_e{J=q^B
zNidMu%1VPNLHWG*{CHb$X>2GUvf+4JgxY97j17+3Ig!4<o-0&uS!Xtzo#{NVrDbAV
zNNkl9hFR!1({c{|mJ}yLp@+-#e1MNA5k6zbSBay3sN2c#+Y~<$BEMg}H=mve{6fJ6
z^7p4L_-6PysOa1LtB-%{5qt%Gy!a*V^ygYA{(6gkA$?u>#5?T-KlL0xj6eUvw@vXm
zw%`zDid$+Okp2$__Gng|k;f4u8?e7qlr{IW_3RxXW3k`z;(*UNqYpXc=T(jS^zl?j
zW&U1z_lyDi^qT|UDunHecDxcwLF51}8tq36#zyfgr44cWxR(vs{ZSaUH1ej{ZeYam
zyF@_^b2wRRfLb1}KY>||)sLQ<Zxd&|7c|8$<FMnY{$jH^nXuRbT&yvBPFXuDbE@9r
zwl!CnvevW(iBz|yL$;Q*w&t4*=qFt**><YY>3&`z{q%aBWx;~u*^FhagNI>&3Plh}
zm|$Q#+VivQfXnhaw6a$IG4Rbvvs<1hqSr#*E-Y_(fBFQT)a;cDd%nCDfs85Nlacg0
z48kY(p8J3~J_;Gd!bZXNWp{Vw<KEt|76*F*2TZ50cUXL~#qH)E@6pyeGF0l@H(;Ah
zI?|$w#cIGF*^QCY#rVU4Q0gqvnSII)1pLM$Gr-*Ne;aQ*F8O7>Z~qx)28z<a<j?=Q
zzVCOAo&UAD?|;+1+mHDZJb-)@5TBcSzO%l}8UGm>n+C8;_an3Rmy+TC+_d^$V}3tA
z&pmn#;M6$JAZ73<@M34|Lt*a*@FLLh=tQ(VIvX9CICv5J`;su-&ChjhGM^Il$3(V%
z>TMQfyDh0hnFpaM>tm`j-vg&j{AFjcnKgk8elO{r$vd`!^XW!b#Aj1YM?{mc?HnC%
zRe;I%g5$O~Y7BbK5jx+S;cITG$JQ&9s9G4medx{_hHvP?gs6tG#UHauHqffu4z(O=
z;=1CvIDT-^Pte)N=p?84`if3}^I&KZBRJPE)L5Jd>qavx8pu^L?gSD?d%-aQo;TH2
z4d9{hm%X9^+|qz!a7hFH|LopWU4$u=$nL(#yvw<bu^;%UM2Xmo@&@oUqHynp5c)T`
zq5jccYO*&!+fH9lNw94Jw%bs!nBKArs^kmNWWQ_tr3|zL4HU&5rxh{gj=FB;j=(}S
z1?5BxPcB1)V#owj=EAikUr?obgs~<{&_Gf6D7`sg%t@6w3IZ7=Xi6D?Qw@(VLjzHq
zY7=%kl_Td#?r-{yFlym2if_pQXQ{1j$-@Eh3Lv&xuJt`2&IC9OP=+=tYC+&6KN06)
z+y;q~!o&@V)9aUhk%t$ykuM6QeoV|K8@8PEW7iy@?MI8pf>!@6x!zw5gde*eaQHSs
za4Zf8@{IYd)o}<-nhRo0z-K`enwn6>AGK5)9aJ?c#bvcg$LC$U5*elrqmx)G$4u@)
zUD5?ncZ@Al^(fc+)VJ%^u<PnqZi>|)<W#a7a&YRaVRlpH?p69z=ryj>*LvNd*`?(h
z)MwZKg=Yn%XYIgc7HYULtl3Q9A8=GJd4}6p8Q%c$X2ZG%5LnWJnD0VHYzD={*+OjT
z`1B?qFo#S<L$&1nV4~|D(uwbGCXSdhQsX`MO;~hFagt>pq;9|}+7313mBUTo+fQ!*
zm+{R*kx*<X#`WVw^Nq7Act6Xi-A69u_WqxcjPL*Cqi^q7foJvHA4n7L1edWJxA#Re
zqo%QBiIgj(n_1YsS)f&dM!;46NPX@kT9o1%MSJbo_P`_X6eyKI(Z$G{AbYak><N(8
zI(j6%cDcF2uA>#dF}!*%esM?U(b-^C9+>XdoDTU)v~{`Pp9!wc6&dRKxuC)3TOdVQ
zVGShfeJ7Dcns-X5kqQnetkf$g#odABE7^7n)pANK;z}tBz>|BQmx^-}jwbmn5ntS^
zFjN}Kn6iEDG#rif3a9l<@;5ciw@=#yZ4+4V)NeVz7VLAJ!9>fyklmtF$m5px$jN8x
zjpDeTTbu}mbP@X&_n(syq_iZ4$%Vp9A;uNa;KbiFJqaB$$wJ&mGRr^A0(*dMp<!Jb
zx^I$;p-Pi5zX^4n8-6({j(i1v+e#!9W4<fzbQwYW%<Rb4hA#^P`4&(S<ti>>-h0!D
zf1+EMa(x`IIHO@=sV+YOX?y8Wd$E6gR#~K}uy)Xy4`W;|5XMnBS~YG?%Tzcw?O7lW
zBn`#8i9G~>0j**iZ>&AXmf$HRq(EjtnBiN^46Lip$;DsWF$jnClSvSAH+K$Cp*-Tz
z6mMDzNN7&<?W^4#bX|=$E9)J@N(DNOLVlsXy#4`(X1#uLS~h>cF{b}()o}i+_h0SE
zq=L|t6En+5#{^P+;Bu<w#O)E_LmYrIB{A~Ni#C37bTK+8-Q2l1qa3qH*8$@dK*Kxu
z8@|RrX;{||GGHyQPrAm3ZkAFvJs`Q9#2^gk@RmfOLlkCzghVRM^I<1)%*Qct-%i->
z*s8*xc5$WhV%*FMx-8rDGLg5mD~aeS%l`G1+HWQGni*OvROWm^*3<$aIqAE9%6M*+
z$Z0%P_pr38FQFbdTW$nHg_nvOnbm=1uP;eAh;a8|Nrg-ys#fqU4KM(GCSW{RmHUFF
zTbjk2;dWw-0+b6Y0M%|h`mUNsAS=+f102mpR&jIa1}HHfMsLm>ZW^ZZ?}vkQbn~l5
zkDJpn-@pU}q*hn<CWVmT{m#QxH4&~1R9H_Y>2*9MgYM=j$R{CX_Z8zh6HB8$s(4p3
zE}uM;*ne6x;UGJnw3UF5C3Hq^p-*JeSD0hY^4(L*^OpbMmdVV-@R){A@}0Jqfir_j
z5j9MFa=X*954J%l1>H4oc}pf1ZI|c3>6gID;#6V}pz^C3woyt=iqN~h!ZmvLPJV>4
z|8chBE#i|}O&7=za7NXA2FCV?y0i>o7N5H-Ml}ZQPd`~qh)Ly38|G0oi-L6qwQ2?(
z&2=d@$O+vhE*D*>124<5Rwk^t>GeP%dYh^=nPLD~fW}RzLW_=`pW~Rq`B;D?llz1z
zzA5NWpAY~JMq2x5AO){v)suY=DMk!KN-Tbksnnp7VWs-85Q&cz?4P(Aqe$;JloFmf
zSUg}|VNJ-K1lkq!;((r!6;(f*LOxYQJgt*ck~--d5-Vwhi%DDZg*hs1QtXzBMnVya
z{0WgR*_(ih)qBy%3V7L2K;S6KRzlLR6@(=RgSH^oRzuzv$xe{)sQ!+l(Wm^TFL3Pu
z*CnKa3jZSE>}rzF8!=4|dQ=I%JrMR0@LS_^{N2OmhIKoaO?B`_L}$ZPf+?y;9A9su
zH3|DG@JGdRC`<th#;FBb8ZfV@UcGt*Oe6B40Hk7|5MPVyv`QoR)$C5BKA}rv9vv6f
zk`ovto8)4!@Yi({sKM+U>j0*-zQb}v=o&HOjFs+4atLF!l#Ch&$uQ_>{m&5Q9@fwq
z0T-&71aY;tr5wgKl}OTt;OrDF6@x0N7&am}Rz#md>h$VmBS64vYX>8$K#B|46&xyj
zbY{`Unaolxpg>X0rVsFbrA(14?%^D9CzBd>8I&pL9BJBK6b9Gcnr^ZDrxkP~hzlZ>
z%h;oaz>*2cpwUfG3e+H~gUA=Usm%fU7_%SH=}CbuIo!>tDikv$E0dVW6|Y-2Sp!T4
z+RH)VN~AbeSf8G#S*ac}h8x%TS9INk-noboP8~CZkz%<N*rH_ArIdzDP(Z$98hWOp
zD<bb8o@ekzq3-@uVc0PJPpN;;1oF$xK>izFUjN<7rJ6sww%qLOMPSK8FWoC`-w283
zFB-hfa6vf(P$^N<2;1=~VuL+LGfzyM@COm|Oqo#f15zw1u&SjhZUP;?E@GDS33DB0
z<DBERqDP?atv@$u$Yn4&18<4hMx>`E(13LlRTt8d@L?K#-U)p*J8VW)_p!k6kXpxk
zTA4dP8C-<j0t1VXP=1O4`kOjFN!=2Z#la7g(x9HXQ<r{2$UQZB`K57OcviW=`^Cn=
z)`{CRSuVfY+_cPc_Mc`{Xg09Za{J5l0mqXthnS$C`E2C@P9z3i!kjS~pebQg%;??@
zP@VBps4Yk<|3CDxJcMQlQsIKo4lewwWR93GWrTvb!P`fIg{w(e^arAO6`Jt2CXc&C
z+6`@H2pg{qTxRQOl&(2lXpl**e3#24e^7~HRCy$&W0ogaORM?u*m*Wo1s#7xX{`cc
z!Udi`RTS~JCMcUUr^N-VR#yl$MJ0uqi}hA~a9k(|O$~JSP?|rt`tLM_6pnMJAI#B#
zYzG8iw&*rNDDX{1k;$e`3=+M$nmdGyZZwz?`Z$4-7yzM<yw#8^mf%sh41|+kFg+o5
zh1t|0&5*kF3qX;%0tm80C!<VbP=Hj<C&49cw6qeGc`>N6KXe6hM0>j*P`IQ#sez;#
zagB7V)$|xBAHGvAhydF&f|rsqd!a7KUcPXJ_P$RS6wy3&%mu+keAb8!S3v?s#P1&|
zx|?GXG98bRB0cvwof4Pqmx}GedOaeMw<$BpWu%&%aHi7;*ca*?yO7QHV-Ht?JrprH
zaCLTtN*wrDXj9?nryRQyVVVgEe3a|@!gQU0U?c9T=x3)So#pd!3y`_&$Oh5XpQPx7
z2N}+2?Q|{>3pY=(u4fdZVPGQ2d#6@}R5RFS>~31A;zruw7>>tt0(XUbkouep!x@!(
zO5Ccx5q7h^8vs%saG-T(ey{KsvQyd08osJ&lW$B%D3d?!(yB)TK4>CnM7?$}%&jz5
zI`kAJD}XmqZ>)W$_`>B}DRyNTBbDN{V!zDVQ4w(@XgYs{6T!!>f2wjpJKoJ)bezW1
z%lwmJi=%w{OKk2>W#F^UPDjIkD}x0Ql4YTKw1tliC^*<WpnZoG{1Xi2M;WIm;CZxI
zH_YClR6J7j<~u8Dc!zp*chc+-I%-yS9T=}F3%`$7Dz4nr;h8N>gC)l9WbV*1CFW_W
z(>XGZoYz%Xgh)jT$Ib!Y`XXniBpP<O(}g(OZo|Z*xyF{=`2h1UySQay(=F+>m4@&c
zfN-Pem{8dZW=_IkcXL2D=?YD?x2ztBy4{rN!h=k_;_)jTsh>)tDwyTd4nk~$q>PVf
z1>6Qe(tTo+Gt*9n;+X_a$L!ZITD2Vwn!#1TO(pAVqbwOUx_^TOs-0tF%Sd`*OuH)m
zqE5#(6oYQJ<j7!J@)`WbiRfURIq9xnB5LZ*&Ph`>YFO@d>zLlY0c0#1GA)s!IQ`sp
z2j}=yoarJaM47YtL@$j<c*$22-sw&l+<t%2Nb|}}?I~;Yae0;>1ynY%dVsi-q^6Lb
z+x&-Ial2Q^AK?Nmh0r2kPqX7wpOnQOoktOs4bznv^*}S3UVJ+XU{0+RzM4@@ip;{W
zP!E|p*l}VZYY0H}a{x?KUk3U4d+5@DjIia)rUIREi^*+rcNXw%+1B+~Kn(ZuL9jJa
zhr>FOZI5&X;b01O@Xn`BgJRK}T@WS>?z5`(#SWYBt8B%uBn(YQFijTB5<_&&hLZrU
zRBR=4srJWe7HHNlf0Qd!r0a&8+{!>Bt?0V@6}e>$=jy?Wf30dAiGNW>yzKLd+_<-k
zrt0RSx(Te*{iZ{KmA9f>X##d|>cP%467>S3*i3@8A24ms*%}Tu)Y<s-)(x4X6xoK%
zW#G~r2@)d3g5<eciKK!}FF5Zz4W|j`=Vnx=qipLCIco2f{C$8{0~4uw<<_larcIDK
z#TqiJ`_0z^m$UNnptDBk#U|y=nd4qKdNy8tWZmh@Ei4mZn#V;23X%j?b!pn|IfX8>
zokLcw%O%aua7Fh1&~E}o5Q=%;mRMw5hI(n>qG~S?S{AH2))$%W%D#YdjK>UaTEifL
zu9nHSDQHc$MmW6lX!FN4{mpz6n%yz8G{}XqqKvO^8t#Kw3ty*n!H<2ABi6@BX*KOs
zRWugOS$N(#70^c<vj(6+vf#90s!^koUZ2$*^D-~;{DqM>Dsw8ZtEp~GXH08q{4G9L
zR9#(epxL+By?>TD8qNW1@@T}>8REHwSO9m*qhOFRRWTP~p$V(elzJ)k(_Q0n>k9r>
zdC}LX{(w%y>9=7Q86H|?(|wOEojL!-zuax~^eho(bh9e$lRb3#Sn0T&kQq1Y)z_P;
zP;(OT%D@9n4saJ7){Wq7jvKvf61hXatJD5d?O>y6DTkbepV+#eN-jNRu*n!g$$oqu
zbHJ?00?2plA&j)#enXC>LPM-`P_FyP)QMz1>4#X5@vF&x?2pk4b6aK7m^yBiUM>@W
zW*9FTlH6c$*%P~!z81#^(y?z-gY0u!N+PFNerfP*o+UOF`KBzv>Q4Wqz>xn-a>U=t
zabOQhf>4t3sdn(&^hd6~O_kXv>-nlbSflaBN;Ljhi=Y0BEacD32z9_U{5}}~7dY>H
z$$DS>9xC)L`lINFBB<8XtvssbBb=nZz5jO|zL>oG=Roq2-;vCDA|(etjS@V85-7`~
zC*+Bf8090odin;>VdT#-oXg`2grYQw+50~^zBx#+kZ%y$;ozr2MEZuUIRaVa6FXdJ
z=osJ1`9h5rBNkhncnLTXCE?=X2aIiM)!EhOJN4Q$8+Fonz?tir8r`8H4i@OAaKZlq
z!A3Ja`s3fT0C}NZbHyA1p$^GzNO9G5cieQ#ZQo0^@2-31rTN7Yk38^@$dBG@lp$TF
zY*}&~Cze<KA1zSyFOewr11XsjrOK7raDrTgN>zUH!IlM!wmtDQ)Wj_w`CE1V9mFpI
z35Is)gl_1?;n25a*_w41typzQgHvH(&zUd`BQ0SZCSe+8LHxa}^*>cJ<CA$7BIw<O
z$_$>lnR0R6afaVfDonVz;)ySTgc7w;c_uTAf?Juha%FXq1J+uYFOT}m2ZWYF_3F>S
z^m0l<%?L~Q-u<Ke3jY6=d<VR$?5);^q5SuzpFjVJUpO%zBU|D7fBeBWKemQkf1Zt^
z-NzTu^~V<M-N%;@-hc&cY!5N`6Hr4U%+QD7hFu~Izd#x>M~%oNRzP(UM*E%BhUkTn
v6YiWowqm00JzmtD;YYnwf~bE&cqxd6C>lOPJQgK+O`H{8r|$c{|D6H=72yoL

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Light.woff b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Light.woff
new file mode 100644
index 0000000000000000000000000000000000000000..cf37a5c50bdb70f837f65c6233c0be5e4fc8bfa1
GIT binary patch
literal 50680
zcmY&;V{|25*X@bjQO8Ngwr$(CZ9D1MPCB-2yJPi<ZQI6|=lyl>9yP0K&6;a#tX-pO
zjqNTkCI$cje3zkT0P^<*q<#6@{vY!n`~Mbk5mB*kLBKa>{f*x^W;l}K3i8U|+&KUM
z8Uz49kl@hhiij(#3IPD%4&OF6007g$XcuTDuT0PMeP8wMr}YhkNNbA}V_U;-4J<SO
z0QLP0)~qlM<;2*{g#Z9R=K0p}{RdDQgkCcTbK7su=er%yZ#YA0X0e+aI)D431bl0d
z{s&9|p}CEx**6yt0Hpc=Knz*T>9Zv)Obtx{fO7b64eK}95Ok5YExyTbPUPDr{00ex
zFIa(vt&7JuX8{0!DFFZo3VJ6eI5zgi-@2+U006S^8x=ZVppdqP9^dUY8hrZ^{0E3g
zFgZIzThnhY;Csxj-(!qeo7X^fuy=L=09v2F>-q@*;39$@OH(>HnST2Y{nwZ8`2zWU
zf+gwgkwrSr2pgn;Bno^p^TUcqasVSF&{)$bLV0`Fm1x;M1b>Pr^X*gJ4|Lw?3li)L
zx%-V10_kYlByV;#gtl38?W#m!L+oFnZ0`NgQn9L9xzV(u+sW&a_xv|*l<mrd&#6wi
zXO#c^=EpL?Bda@^-(pVi=SZ)6?|rVCvNTQwEu)weDz<Qj|7spVu}VJ;I(h5v@+u+6
zyb>CTFf9=gX^;!WkNhB)_#d-^G~i*o1V45aaEpROenKH5=lfk{=C(vd1>s&fUu|A(
zbf%@bLxCCKi!s32yHt6==&IMGl^~^Ow6Vp2gT6;*<8jpLZkz*YyDCwOC0!zr8F#$T
z&h#GuGZ%l6MSQi37&CwKt7A91r#mfG{wvJhq#HiRAIlHMS{K2BOMWXefQhRcP|U9~
zLOAoNd4Samr@z6rzxlCNX8i%Nb=|mQNONJ`^K?}&VkzRl>frehj(ijS2hDp3)kuma
z@@Ob{neIm<vyo0!1QsK`Ta=~hVd{MmCW`o9xL2=1V9koFYMHX<V||Nnl?}Y4iDI8-
zbiTyK7s6xNqd&?-LvXQwG9L+Bl-Q%^$y`envl<pg&5tk9fiDt>^<kb%Lu5owdLt}V
zt2j#e>z^|UgJiDkR>@R)I|wb;zwocqE2)lZp9y7yWVgCHXUo#v=E|WnCmRion{5Pi
z+_QIGP?6=MBpC^FAzn!SCZ9tv(#S;y8LhW+&o~>k{z1MolEYQQpm}4&Gd5C0{uS<S
za>5Mc{h00VTnvo&8aC~n^5;IWs4!abbH|h0xMh3v23t2?F~0l(28ApF_ah(2_E<3T
zjOC2K@33AXUGVWO9*ccXA`_dN*Z8z#y2eDZ8}p=q#nBw8n<LckZ^9m550c(!6sX?%
z#7+SFsqX$o>v&Uyd5eR_rPW?kT`oD~v-FR`6h_$lm0gLGcS2Ma!5my>#+-AfF{-qz
zbY|z#OoYb3r17)lTWmGGk%!^tQ(vifP}{*g`R^D^N!7MVRz({5V!pGby?7kZvdCY%
zfGA4NEY>M5rY+V*LQX*v3MX7tBch=xf<S>`42BZ+Aoz<Z;laR%>A@Ho9$p5mWp^;E
zbm=Zg7Yg+Y9#KOj%l<I{j+M;B-T%(_XmT>$`7nzU4m3VArVB#}KbsGZQ}(KvMrfg6
z*wm3_a88fMC!7ZD_s@GFg<Rp|AfA2V-$w8vl%XUi34HveznJ4=j(`{+NMeLi*PxKb
z7Ky)z15}ELQM(g@bneALA{hWd3ZaIWo6Bt$Sf)64(W0`lQE9UV#_f(i&OT#C6|7=0
zb4RT#Y(SYk+`X>n+acO{8yyRsj-p~tIqK-jXlpIsFYB1I82o0s@R`!7VfkonE$s>M
zS@YqGVcl`qRbw0G7HV<D(wpd;L?;$q_D{2}N?hTm=nX!s>WQ=#r$|N%_-db0Q<x(x
z;^Dv!iz@zC`LT@joJ1BldFFpXl)T3e25yLtQY9|ba|by$HS+fHOvh+gZ#YoY%v~tP
z%g0Y<TJ|Mr#*@lso`lB#N?+tJ`kTlxXYrUpjTK|C{=|6xiQ`b)_Vo#PSPqY?hsZT9
zt^Vnm9+7gR){$P_%n9{RRxnb5bAj}}j1MRe{L5|cd`nI)&*?ca_9o{>UX;wOd0^6<
z)GXiQeRXP|J^XVXYZAyf=i>WDL{r$GNIjVHa$CSeL!68p!%}uPHRsH(Fl-^Mk2khd
z&XAojE`(J~BX`R>c|R^=AAW+siU6#4ePC6O@ST5ilvY%V!Z(#G;4K$rJ}4&Bv~20(
zjH_PGtoC+jHw*02>=+<C??Z!ChUhM5OtEM#>jR%R#5Oac+Z+~ZG9X<vz~lRC##i|Z
zBQJC&z3KptjQoZ)$@M*IY|LlFg!jv;A<#ZEI>UEzWhCqWKOr@zz*(|V?cCA*p0L}$
zo8CJJCePL^sc|5w|9UQnx{I9a+1c$l9`4;8?oGgySC_bWshI0I_4(v}Wo4hqoStFD
zc$y+Vw{PFnj?;Q^FS?moAAWlQ;O`6j{8@W}#$U*BUUiO+bB1!ig}K~K@`;qc!@5{0
z<ki1s`%G-TQ|}7uJ<NVbxSGw8Z$8rAoZ7m1dI!7O!S|%Ux|;#9e{8JQKa9b)2S7O>
zg*jkeaQrUh2;DmtAeYO7BX@)?b%Zo`{Bh(6g2CCR>`~$0H}92xams&#@i{E~g><qs
z4RtcT3-rq-md`yLSjwqnkFDju(H!>P&$ikaom7nWZuyh@NVs=Enof|SM2cFhWmP0c
zCf&^<p;9uc;;C6OM$M$Fmx6Vh(M4x}3@_wGj(%cCxn!kU+@4c`W}TW7&{}0yKTRck
zrymlL(QAxtjDxR!dGzA|<Wl9=KI*vedNI!=dCG-Wl*XO6mZyp{B}#lflbmU_$ruU4
zg;@BX29p3W^;`d&a#?=Lc%jO-@)o$T6K2Z|!rqEC=W^b+)>eJhd%Z-aY7x!F3|nW4
z#YTikW!w)-v}z%~>XHTMImp!j&{dI4r2Oh8w2-E_$NccsGK8UdK|G55H5VMP;~yJ9
zf3#2NA+1sr;^$Azy;I{S?dF<a3bnZb-Rh@MtJti(v+VjyjZ{4dZ>|@-kMR=-8BaW&
zZ?lg0{&_fvKE4%-A1bA}cFff*_&wEShhX=G*b7t){K$98WgnJf#((r`P7X5lmhWLn
zM41#QR%UUE*$SC5fk<EI^Yv7$m)P@0iPM%1(`6uvHzaeP_@*yU)gmll5zR=H3-BIh
z_|H&y`k^%6?@nU_!$^a9LjzL-gHxkO0|SHW(0|d6@T<`!0*ZF%a%yB|kYtAt@G>tT
z@URQ;3fm|9m3{yE`mQr0(vNy!!NA~%3+q6^_FUp|QQ_w32#Wjq-~k|Uu&03Ux~CRh
zUC`7)A+N9A%us#{Ln8@6Gb0jdKwr`gk+x|&AYS(CN(F+_{-XGox~otmLAhaCY^_U?
zm?i)>w4ptlDL#I3vU9%jVQ)Uodw1#0e0RzIw7I1NK=%<9B$#bcP%f;!oG~xSL6-MD
z+J5l-8Zyi0)-ncFGep`Hq`nuz9X4XpDj>|r=dZ})vdSZ~%4a*xqdU##(;`4`WWZ|#
zq9SY$1PM_@JBSdVC{p|`pa|Wi2((BRftSi>8PB5`&*!1aBcjS@@-haz8W3e@r5g_7
zV<8!==D?&UAgm;SaUzH7KnvOJ3+jOr;|s-a710JT{0t82!Tjy_2fU6yj480OEjKLv
zclL7F#&Xc$GwxVh<XT%SP7|g8|0{DcgytVQUYhY&n@LEUa7WCb1C0p7KXE%%fym(_
z64Xg(0qI$zs2ab)C;N4(4ykESAv86j6)|mdHblkpp*jo+(RX|&4X8Tus57+daYuLF
z6SwMkunS>U^LbbEXiE%1I&Uc<jUc%4nccRTqepmL1jOOZ_@_<rr}u<v4!rCPh!Ah%
zB|VZQWe(BC$BkA}dsC*y4G!vlS4FGYQB$7>Xv`%tm!y1JnDCjh%6FAms_M^$WpboP
zwa>3xlPjm_qs<G7-#62xznRp7PCdL-3#lc$u20*IEN`46KVo|~e$;o5@Y&=x^)_}t
zy1vza@My=>r5jog=5Pzjg`*yNWI>Y;m!HqOhH~y@wNFEm16Ov>9o)FzvV^A<kN#v%
zOznBCL$j~ip0v1ecj#`YUb?x*dH3<2=QzoG@83L`Jb`-&@fhY9<v`|K=4|em6FG~r
zQKsG<Hj^LeQL!$tvo&Vq(3y(q$<HmF)vu(P^78M&LA*8KE?Q<c8sv=CY7)tyFLRzN
zxpj5_)w8#6bzgqF*n4Du6n$&(9q}`3Ao)XcmaQXKhkDA<LzCj8DaHkJqS?aLnbker
zzh!E<)8e$&mc34T$>v_#xy5^*NyF)%L`ugi&6ad<E}cOm<NNN_P)t@b(-EN3t53dN
z>0G;*fvm*eoiR8W5N+xFq?-N3<B0D9S)&4(NZnIIjv-l-dM+8;r89V*(na4r(=7&m
zMSLZE<$I-jg?IJyip1XLR@9!Jz|?H~Jel)wc{&-=tTtYW&9c0GmoJ`$iIR_!uG~t=
zLuu!ZoZiRVa4NIc$9CFMr`VRNnKPK!H`t?U;BaiDi;#4+!eYTkxzbw7mwH-?g_oFn
zKDMs#eAZ31vt(oS(&DLcO6E3YEI_|W-?I1CUiu8<X3>6dHjc35=f{S9PHPtaF^0_}
z(z}1#6ThbU5VqxPkycDTrkPJ9JFF|?Bg+8v#1=(rp6;rxDU%$+2oJRccZ{O0tDd<l
z2ye86Z#2W8ubv!>5gm)!H0vr!i8b>av=@sk?%txI%mECsTn0C#&tK``A>ZM}jo}4e
zvBjw2#Wk@d7HzQTnqjqf0~cBwAJ>@bUoytYH_d(-b?1{_V%_~5eLc%VgdC$|61yrz
z4%l?guw7ZP)LDU7?U?SII7=M(?Cs!B98mmDXgSWLIX7Tk`&M1s*opWk>*F-*12Q_|
znfQ#%>`LpyRqw%^FO*>EQQVj&>l4OZqf@$*6C3b{zYh2K4v+Lg>i~!h(IZ%6Gbh)W
z#Is$8TN;612Be?aXfN_=hrOCHTMblSLsVUTTs-!}AMRB{XEtM%He&Q!J8?Gs_{+&l
z%UMq=fpIG#@-K<q&&=L;<ij^Vqi!ILIpM=NVA4DiMLkj%JQG~!cxP;7rY%CShhQ~4
ztGd(#sh7vi+f$80Gz`^>(CkDlGfK{kWfx^^foisMZr!B^uOd;+3f2z^18DgYqUGU{
z6@MZtg9}uZ7^#Y9sR|pYN^gg!A}fl<D+($qN=_^CGb)N#D+)Ur4r<|H6~s)H%uE&4
zOqJaZ6yzYN(tamvdwzE{Q=ah97Nr6dR#}kK=|GVXO^um#36oVxddRv+%(AJYJZ!ik
zsopC<%%XUgMKmTQk1V62B%^RKqqIe<VA@Mvw@F2}S!SbIh__kGw@D7HSudzbFsNBl
zIHk<+Rb`Jwzi~n8f<^IxRq%mD5`k5iwn=!8Rk}W-l1-~9Rx9t(OTW8Gs{0}E^;Gy_
z4*OwA3;}3{a3t)u%iD4U=C%(wKOi_i!pz#{@EN<|7=n3@G<Hn9;V>$C`(5Y|UFeua
z?2ruS5EbPRcD2iwwhNNBgPgX<sJ;!eH7IOnCT3>@`OA>duiqiZkn%I#^cf@dKwmS~
z-^kK$X!kt>_zot08yCG}nYwMtw)<o=kZCitW;4ibGtp@?qHQx}em+fpK74fEk8ysY
zt?Y7LO~;ktI_c5)Thn|q38xNNMbAd4nC2?W?xNi4D$Rdlz5&$TSmNdV$ILtLjI4aG
z689hySEZ%~Q)#jqxsMSj&&B!u7mop81W<hEGZ+B6fJgunzy^Q|U;%LdFLMF^#Q}f^
zkpw`3SbszQ8;$@-P;dY|Xva790pD2wk1XAIUb(*vNFgH}H2aVe=!vw{KdEbv%au41
z6fy)Tco6t!YeXJ@NLX6vIHcw?>(Yf9B*wHX4>Mm)9p5#g23Vu)+diZ3!X-Vw$!yp9
z6F|M(c1aRsEvjBgb~SJ~+f*UA&z@wMDDp!_ah-bf#&j%R*=vtSw%;0H+^C_t<R9Hp
zVp1vMt>NwBIX6TkO=X44EYk;n$Qbm|(`R>zC3RU23g`7xQ}I$6yNDX&Slo{j+%BH}
zNj?+zZ#L=Ko~&eLPlJ-eC{qE)+!NFk7LKO6<wnmo;9&&M!agv%mX{IHRFQ<g;K8m^
zOfrmN%|hkf%I++ILnCKxdx0W3wch`pLl6KM02%=MoyCFp`T`(&WHoie7#NEuFucD?
zHrPnFkVrR}j3tszq}pt>(rEX@-7r38<-Ez5k&TfgO$Ou2Tp}PzXbe0!(jo~Ftdk%)
zagnSWrX65{jtOBx`<NIKPB@v~L16Oxt8RScY&M>QJidKN7Cq^lF4cB+baYg7RA?cE
z1Q3M#ay90mMhkr<=3$(MKx#|BC<@y9svu26U$eHu^(bwQfD2k+uYQyXOb={bSN^I2
zuNiJUk(owsa)q+(4>(Re-=CO`pZ2*d{`1VhLkx?fria2UQi=3aX#grUSdWm_<e1xQ
z$FlPF2Yt^Vb<++vMcyxvhx20A>atcoG8fq4v{Z0vo_Wc$#u71vf5g;XeN^FZ_(AH1
zAiO&DeF*;!lX3LG<G6M(JOJ3+TtOS2GRFW3F4S{)aoBt_lv8+&Tj5;*<^Ev4oUuvx
zenYjK_Vh|N9tzaX5f{E=Q{s`ZAJ9M>bE{!1Wo_MmE_7!P7{0f?UPF90d)MDn_EtiR
zGtBKv+x;Rtmt9=v?BQZk&&$~69(8|VleU2IfIBI|8$q&dVRS%xi_@nV^^f#@Fwp>`
zBZT2136HUQd-OMEU75ptcz_t1KpdbO){++5!Z1MWxHR`s2hrUv5)TxKyUR+_0yG@|
zayH!#3rzI5nOc2xMF@HI%10@=^E0o;b*^t`f{vS~u3S}}u6HQy%CX(+wkKq0s2cc=
zqkosdepslTHX~Z~+4hvV*|z1?zk9z6A%U0%%FFQ=WAU<5^MtxT6t%zfzUT*tYei26
zvVq#eLMQ+SY+ewNx2e{wxRxuRu15>7oA29vWW&kU5zBS#OtKH3(ja<F746*!t@{v_
zmdfh;fpNSGdR6oM``)&=6?s*(co4Md6A*m{8(k@mDGu%(;^t9arx!L49(!S5a3Ox`
zqHxMWVQeHb&hlphUAo3hNQ|!YaCL9)4Oj9^#cytR_i(m~c-5=V=ZLm-oioFEC{ess
z)}A{<*28)4w~+l=v|!3+)iOb)75pckS5kSEl&6G!^f(e;^sUwL)FXc}d!nwLzN|w5
z1LD0|D5FJCk6s73FyYezm~XZE>-fr)Rw3b2LC%~?il2iN9~BE2qU8|cx^|vS9+#G|
z5#1RS6D(Is{;XmLET}o*%R~(#yv@u(YwE{rjCV_NYtyg}hCK}8@Vbc=@V!J^ELP6V
zT^^UBwb_<CW5#nLX`jFQBx1AU^@1TOzUR=8Va*U1Zy-fl;6*N9Ra75gIS_K;_@=+Y
z4HyW3IxEOyfW+&zuG`%=suK?(JI;x{wI2SkHJ?pKiB=|rDr6JLA~c`I0{z5Xf4Ro!
z#f;#4Id7ct6I{(8d@|YmDQ9vyBLT$(s?z01N;f=&oKFx9ryV?!8eCSVx-5o_6{00R
zU3V7DL}h1_uf4Dj00pHA_yNMz|2qI+zOPX$=g{b-T*$S94v1H-<H7x%nRJc(L=R(1
zfm&=jUL~VRP-nFtJe*e77zab#^rqOBndffTmT7D(oetsBtzxhxIWW9RS~w`56>Emf
zDaE{w5N}h-h#!mmUWoQdb!tK;Jf<DJ=`>gwS8AN@!;yYxR$?UVn09o`YO81OMzwfu
z_oJv)-`}8KBjeSBf{BljulFrid}nL)czJtZ@~;8Ul8XIIh&Rqd?38Z;P$hHZ<k3{Q
zCLw23qC$3l0jetkL3o2vnPy&lG=Y|Q#lX&k17k9lK@X|aIu3F1Xo<eRD0#`8KU$=a
z3FRXE$PH~*7uum5VD;E(Wf5;l0BJi8yWcQ=#(W#7H;&u|*r7%;9wy~?Fw?M8rxov5
znA~Ak*U?!yxw@Ar-l{cSprTWB2Ud1~@m}CPT}B^Xw!Nfs2N+5-8lYe38%`RTyoGR-
zV#bJxCKH7Q$4C@rGfbgVh;e}sr!${(6&egjY`9b)DGE1br6^#F5?K}EU4h3ZEleEO
zDwhD|@_=$ug~ke@lI+G+lbFT8??bMPO1S<cxRcMc`HyNfIx9HrT~jKX;msTh2@r`|
z(kh%8z3VG@t(tI3+in+JGFp}B;}g5cQjQC%Q^iD6^A91S3XIZgyG%)PZ+1W}KfvEw
zblMig%DM7+O+O9kd6cDEuQwTOQ3y&XnEJ=xm^fIY@#<6<8$ujZ@a;<xP5Tvw4qtyy
zr^(Eiwu!pO+>=voFM(HFtSz3k^<Q_ASA8!ooIWo@7=bmq10l;+dEhN7fQ93q6+3XI
zQ$7W@*7Zc)^WFEDqX&?D+8HO$yYy^cOYjSv0SHOc4Dh(q2$*eLc&6i{5o2Pt69*wP
zxjpX(y*wng=vOW)88&pU&F_TUlS)KE7rb?|;ys`jw!xv7I=>+bW9V%W)O1#ZAJ7qF
zr6;tapf=Mj-n8>RPNLzGgM|JamPH}kI>(_V-BUMkqI?ja<WJe!`lfmCG`_z&2(=cj
z>J^1p=hX+e^qqxJ%(KIxqtRYtsTD(jfx=fD-Y(Qh!O22f2mzD=1T6}((evj@n4=*>
zC8sSAPok&)v6_AW>Fu6xzE&I_iE!PYdgOhDy&~J5_&_dJe_YH8cVY)4P;9cmlBsrb
zEc#fp6rx+V)h$xOQOE8)&iH@!;-|5h|J7!mxWj5WMC^a->dWm2j-%^e`@_;`C8yG9
zD-xgxVId}P$E8%7F{cjV2EL5dH+q1RTq5Xg(kK?@rP_!#hT(4+Y|Z^vjC_RZ79vZL
z=BBq$ZWjv6<xGVxzZ5a`VnV749n#rwFFmjR4%FaCUS}cS5R`*IV%9uiy;a>oRvRX+
zvc3zy+b%45vuXzQwm#@?uJEe7Ov;d~+C?t#bEXaJ<79_ovP5lK0Z&3;0R#CfMNisW
z@#*~QLI8d8w9Th_7Cs<ByeXQ;?|s!eZqIzLMi*yCSQYf$1n~lA#XZ~)Muz?jtlI_;
zZ@z04{a*{;(m~c#1DV*;?1YFYePhKvQ>hw#<A<(Sx}G9zk^Z||vtB5Z{MJ|RYl|uG
zhq%Ih?%~6I?iZ9V_t)COD;qL;-k_x7j0^Z46`mn)&1dTgbc~9mZLr-!e~Nq@mkFd6
z9!MBUu~F7h-vat&`;oZUT1&*nNGW=5YD<2qgREnb{aP}K<F8iaLvgTZd5|L)2?|(J
z)lA-Hq8TpbWiVM-$3cZ~wumtiF*~^5D>C)?VBuk#FZnvru|b09oKX;cK9MA@GG;=)
z+F>zJ1!n|)cvA93P1h9`Bt8Z%M!>@>NBc)aGp_fc^yB-GY{R75R!r<nkTz?$z7M=&
z2xr0#thj@y%O7Eq=Cj)GD-vzBm858=Os$+5vzD;zI5eL!B(qz40=U73>ytz%N5~{N
zP-u!KxPR&<$oe}!X>@F57#vdjT{wdZCKw%mXY{~wsBw%r>0}a*S?njYwzMSc`Mn&7
z{8>-5wCuh!ESW&wSss_(7tO0w+esv)Ixl;%i(#8uiocu8QUh;NqlJ?7>0yC+u=%OH
zkZv@wqLBLw<%Uk=*FPyU({b0au6yvkj{VsK=jfV4K&k9MvqFYInruz(C0%MO6KGaB
zSAh*d-RU1+ClVj1ANas6Z9k!TX~|obo)~CB4|SMLQ{)=qcMS;b>Mo_ASnNA<w7zsj
zG{Y=N3V^Bf4g7d5h0rJ$cZ1>!zb^C_+@V=CMx3rq*7W5cekFIC9M(Uq8Shalzun}8
zJ8l16lI^ULmg!NOM?A~j_^kJu@m7zob9{+#k=Gj){*$<Z7LaubhG8A;TeXVY*J($A
z)lZ0be4UV&>ydxj1k(Gs-UfAah8rHdZKW0Qe;td5?<s0D>NHh%rW%9*Tbe^ts#{$Y
zky9C=nRygO)Ql`-z#^A8{{<Np#Z&JXB;_U-^RtP1x$E%WWCy-I6!^WEgwZ?Gdupx*
zU#vjsALa2P46x}eE+#`Pp-4IwEmWb^E72cQ_?^ZpQ2LLfK}!JTP+TsFJBy%3`K9oQ
zF_K5$m`G0~LTh%ApU1;T4`oMLPaPMEyUW8iF_T?mP+QzT#x(Fme;&dA&RLxOGte)-
zFxOX)8Vw#pNI;J=doYSWT%<l6=GSmCl4#O(J7k`NoR(Vp_pFGWxf|O7>=q)UysK_K
zM*|d@nMTB-`$7SSbFhZCk~GGOx)Ks{owCog6vw|*)uq&AEekpCSICvSAZHe&N<723
zSvo-)snK6E^~=W(ryB)*FNYv<-c4#rf85MJnBM!uDLXsk;(v?5q6(eS>n+z@T!c5%
z@TnisM~d_l+ofvA*(p9EZc{Zi8nynG3Q(1x=*FL3R1R-;9<l(HWt$SLud17dLgg~i
zWW#(;3`;f%>-Vh|6C(J(IN4GW(w$>AFJ-p-dVHZk)9vhyr*X$))5l2nQQYN$S5{F_
zT6VoReKD=A;6}gFdYAXi8jg^kZ^I>UX$No%Cg?aQI-*@<mWF!F_(SP%V%Eu43x*w|
zRUEzd$1>$}dtSz7@Yg$^wrO`~&G{crX7=^nFLJE)TYd2NM#A!GU5T8)Dm<C2bcf=W
zhOz32_!1ybcg7cAUVkbzM~^_RR%Fp0!{5#rI@R9OHORj*w2SDBOVb|}e@&{d0ZEKd
z4{xNo?Y~AY6nZ}xUXt(88)RQu{O=0yy9WiBT?Lqtcd(P4y~+q>{OB<<trwq9X<`Tn
z1^{Lq)gcHl{V6wf3X#<x7eDFx=r7=FWZ*RvTpNxhBVx+JZR7K>v5B@8**yk|FR5!g
zK0B60*i6RP*x|3X8ryn1drKd{z82lvSa<C`xBEx<#K-8^9p7TS?c3(?@Q+thQ~5@&
zJ5jHtk2&<Cd=$B-S)GHFwb1$yBCKhBZ9e|hKY+=a3<3yBw@ZRH2?pv4NuaH;FO`kL
zddPe`l?XMTWu)UGrDxN|{@o_GWIH!sRPAI~#Y8)_jFy2cF}CCoCKaLZGT8krcjLQ-
ze)&rrS2V<Xa(^p6d%eyUU+5=s&}Ep6+T`>LBXh9;gv_gOoM=YCXwQt7?(gbvh+AQO
zKQcc5rgj_bU)XB|-Jm?0QgTOKcgnKm{UI^?m7X2)5Fd1r(vvL>UQ@ML_&s*n=05W+
zDO{#LrxclCZrI3ZEjlowrAUm~6)Jxxa&z^_c7L`%FY5ON==f6uC1)hl4A<-q&O!w#
z@+Q^9A)b2;?Tn(9asO(1ScjYk^ysIlj|G+mWg8vxS60Z`5nb1mwA{&v*s0_LP&c@a
zx_Kv-E968GNKA+bNyL)J12A)|->$>Xq~PqCyIhSWc}PCM+C5+F1p39tE*DrlF?&Dx
zv$${B_^lg}heTHygj%;Z?I>sy*`874X?pP#p6G5+EaEBPL1u0NGp}bYQO)8*GUcf&
zbl6{!7rXZh>bb)+ITveBeIxJxf)M$Fb9rcN*w-CL5b8RBtOzfy+9HZWgfs;|N1AlV
zh8Ai!mGvrh!?gxpS}BLr5QjJLn9yzxs$=%|Aug|TpYXYwzQ)Ky#l=E%k<#c^I-gKZ
z*7lc0__A~JF0i%Q!*|Ca*P+E~A%cSld_|+o>_en07Tg@wgh!wkSgb3?4F#6@2RUTn
zLRe-IAR@<*R@DaCLS-v~>DnPP{{FU2N1frotDY;y*x4tN0W|WJNQ;z9=U*OC*!A$W
zzqsVr2P91p8*JMm*o;%u6m0p$@3q95Nz0^uLRSxE8<|~@TKU)=LTa!OD<y|<ObZTx
zFBDU$b@hUJY_(f>8odLxE8FyW%;I|loj4<XPfA^KFap#z0=~d+Sp_9vDHqmv)>y2S
zDRKPSHxz7&7K4At;i;{<Ci8BrYnI>;>@Z7=Nh&fD1gC!pH0xfuG%=kC;zg92QB!*J
zI2m}o<+P85yqnCseGEmB|5p5(E)U6`!#RXgR$N0kOpR?TrPltl#=vD_WpzBR0t)q*
zHoF1be_q~Z5a?)3+0Kv3&&8W(iyc@exMi-^({8l8*(oSg)~yHMmu2{HEGBA9*&68E
zYVpbT(9W5V;08$>4O<PS)TvsoO4|a4s<xDGDiU|8#Ejl1+O%@M7gYV7OQD8owYP^?
zacY;!t2dIi#lcxiZ2)0j=Wg@P>-IHxB9L&vu;pW2TjBJ!V)jI|Rfm7g$@A6MFRe}2
zw1#uW&enShN}xaOWqAfO@U2C&H|XN{aTOWWR_;tI2U{(te^-;?dfU}v7FdJYHa+Gu
z8gG<&DP~t^DW)k@x^PTi^Yd<af}F0FdY`(JzNQmXTROFf(uT2Je|2D(uBhs2i72+|
z>`h<)YX00};a#tw#~8AroQ(kSfXh0_YD!#KBt8TG9Jk%b@phu4Htu62jy%mJ7ypfs
za$vGM{yaH{*zhDtp&oUWTkhb7*MjBW_#W%}A<u;D!hV1?=^WF;`Xo}*&Em4eT`t+o
zaqh1#)SLLu;->`SS;7;H@K%<l!mt)&95<D<SZt*VBZxGEW}fi*@=`6H%p9?bnI{y@
zd<^d*uUo<s5>KsF@n4jYnRi16GktIi*0v%X;XKhN&~2+si8t@^YPv2#_0CNm0WeS}
zbQ`bK^}3Cg>Wy?)H%%>G+l`B_GSSs)mCY%BirfOt@^ahSOGdf-s*0D&XIr!6i#Uy9
z_F}^-;%T3x8m$N1_{k`3)#^x-4b|!=m9eg~q60ptJVGiBBb%Z@f$&c`e=F~c^@`(m
zb^l4~x1^kN*T2iTQTslFcNv(9UKVPQ7ahOj9J!dFJT9K1u@Y%l>nYa$LzbEgo}pPO
zs{}GfC!k!)UT_*e8yzh@`H|9JzL-2WHmq~QUDSbH`bx~W35bupxFE3dGNUO$^|s!{
zpd@P`9fnXRoV2ruaI9bw3v<m=g7l&jICRc``8x?S&ne&0h73;+*l=;{U)$kwO6&1I
zpWTO$z8$}t^z#10e+uYH{KLz8O}o0ubS&l{_?W8T)d1@&q`9DXnXP7P&kowrJ}+YX
zx!gmPd{Sad2U?uQ@sJiwq~D1TRv%y`4?Couxk-baJ-LZrQF9>!OBtWKQk@wZW~PmL
zcsXfd?^Cn9b3XmT_3E_|IR5D4Qh?b^=A%7w=uT_5BDz_ZzpjEcB??yE9f<C3)Wkl&
zw8wgiu=%V&!kG57!c}lNEn??4$;FnYZsQ3<HOIC$x#Ub6a#MTwVC!jhD87vmcs2}+
zsNjO!QlW6E15iaCBilcR6ix5a>&TR`<98JD;*U7pR3q?D#%8jAgMFcT-~?N>3wC?R
zn<68)SHh=iXI&>!v_-%t;UYt(E<7SVEuN6^5B!M=wH)5;sH=N4(9O>VTFDT%=06<H
zE8$1%+uS_enNQEDfA9XC3VMpgC|E4ApIaW-Z3%g_sOzTO5|ayWmE<Sd*yx!pv9Vhd
z2q^SiYH~Fb51KM9Cto72G0&5qB65oS<AB@{(H5`ZlNbDO;KEwdxq}h7Mua7PJJ++i
zjlmzXrWC!2t8)e7NX}1^>NLewSkYT)@+h)p)JxCkOwh<s=S~DHdOF9HoO*fz`&he!
z<94UBS#6(N=k=@WI*3ufygXj%dRr!tOnEPs#X7cSXmu_zH6#Bpa_alJrCcGdAL8}X
z>5J_RQhQ8f6`xm$TwfvZC_XPqDjE?f_x^;)Q|h(YV@<n9OQT0`>I}KobO&<(;bQ<W
zNOSaXv(Q;>eJ2<?!SO!rfF-Uyg8>pimSblZ5H%Ok11>F?CGY7nGFkp_iZ?$}%lo=w
z$Sws%Ym;}m-m;6w@1WjTeZW+CWc%_qH3+K3j@Q9qVsg=TZf>n5<KrcLdD1Ql*kSB(
z*B_J_=B^Qwx&at}bH)+ZDRg!&tVxL3|L5T^{l)RI>j$_tJ$P=17PU^lT&)VvcP+MV
z7FaY(LEa8yf(%{%N!8D^k*+K;IynsA2-mJ<A)y|*t%qGGSj1bnNDR#QvyrQLyv3ei
z!!*jmUdOxCs`2<?Ptp~l+c~Ptw19`Fq=*;4m&9#uBjLd&<xzkH;F1vLcOHnpW;pF1
z90)IfI#PcV$rd+|BP38|6*_7s*Nuulde~V647@%s0+{~E<;QJ?cyQh$+U_K=!ort@
z@g)Ro&D}4-&-a6eqJ#$!cC@wcpHO+vT22<@k+6$M_3>v$0MR1*AS_@RqxnBc>Ivt}
zE=NUu4CRg5sN=6}YvtL~_Fe*~{|&x+6@CuCbda5~qIN#S@(!(M-Hkc<^q)`2KDMMq
zS&CaIH+VA~+F4$;v_hZ;aVj5aoQrRSg)xDxd2e}0I&!3)e`p$bHx&D)0_V@i*F*w$
zy;Rx|L}M8Zv(i}sOM~HOH}lxPKRH;{2|Jl?Zbxf$%uR6L1gOo|T=@Ey0U;|ptss%5
z0D;?g+BN`Y`whv&y7IisAvl6_${^T-^R^xym<bx#`=V_l#(n2%Ai|^lSj*vkiHnbi
znHfh8+llRt6b&o7tHvH&Q0qlR$sACH&);)r*In1+KsRhJ;f9y<WV)gCX%tNz_r&sv
z%nYvQGUN;$F(!94)rTnCL!5}WBR(4+qQ`^1K1qBG)PV-$mV!51Y+!Fb8}2hYumP6m
z(r%aRF%TsF%EL4M%8Ww4I$o_ld-5aYh<d-Ys98ba<Hz__!9gZVv5f+(U!PFbW(wTj
z#}xMluQ9EM4vlX2gL1qlL~4Jck>ElFYm63SS^RdgmNm^8aK9K0{A&-;YeuiEsUNjV
z!1MdQOmVg82)i9<1pVV7-669-yo_9nQ}dd!y4;J==oR&`N&&-z(Zfi*R2(x5%9h9Q
z)2yzYPDd|iRJjueK{MX!x({}ljYAi9j;rg|8~M?4ju4fGn7k$c!fRPRSPmbz10tt#
z2UxF5l^9YtxR~-28Fq@zzb?lXugUjJi^3CgUM*#3glf0A>FD|aUhk`3Dh18<7VOx%
zPHaXOLAx;u#0&U6;Qn3OJ|L&-kD4I+g|?YkygXpndrsNA6<n2k@om-!G3W9aKO*S=
zaqX;p{WjNc)bqB=K@fy`Y5bRybba0ne5C2YSZf@8$XWyW<c8f6+~Jn1bN}e(=t3U<
z(&2!3IRE_KGxWAll@kX?yf3>i4sa|m{41eq0dnS%{1PMrOPU50t$Dk2zrcT69yOoH
zz36_@^SIRCX@5I73V&?p^050OB1KnvqtE}y_~zd0ebfE)IFPg)3t=eIlz=bD{kR!r
z+^+6HdvV0lCku#!ep9piSr_u=z>Q;Jfo?GhF53vFsWOebUf3D?YCO7putsP*`KbDZ
zg1vv{@QEhHnWkKt9U|-2+8s3WH_X4g>)qNhK~9v5@`O9VF723mM7>v8V~ebEgL6O(
z(}EZf##t(QmMRVUNZ6UqzUw{F`}uI|OqD|b3#A?^!h634CVd25bJWd~2)g8ag~c~}
z-<<utD!1rP?`uokm0G<?yMO@D(8$NK={5;u>QMDKS@G8tckF2GdRkszK33BP&g*8_
zU0i!{gX&$S#nF(l?Q@;V`AhKk_G@9~V%LRHF}8YWgmB{4;stc__pgRuZ2m?mT)*@x
z-k=f>qPp~OATrg!9PF#&*24}C*S%GcNrkO!+EObK#Ad~Nx!W|x85w5NjEd16I{6E-
zN{M(+2=IM5LOeMvJx{5RXy|U_KjmxqlQp2{D<0y*MW3VH9{R510|Bl}4(k<pg{B3s
z9nf>3Q=3O!L7hC$i$OCbN`3R<#GsKAr(WP(Oh7rc@mTQ@xP2)t9~Tpo-Q(iUQ1_+U
zTK`%b`f8te#;$tSdh<d&5kD`=n0R+KD(gQ27k=$s_JvlqMR89=NX@35Xp}tYws7D9
z35G_b!`2e?>TTrnxWetyN;dsL5a^2cyMwO4)m5EX1A@hRmn2>n4Vcs#yC`fU*nuxi
zm-u;q-ndJBWG8j(g;KReo>zmBUt>5QhaHS;n7n|te*rri<BSfnI0PXf4GV*1XzB8<
z!Xpo9I-?_i^1Tj+6ASQ>H*mvAklkripu?aakZNM*u^iW(<dlw&88o#{@uu#W8ylJL
zrea^wqsj~ScIa`Y&<hJ~DbjGPoH1$5_R_A7@9pg3+}f}={Ipq6;zVry+zTs-S9@00
z(wDC;ZJb#=b~iOP^Vs)|H`_}4fr}@by=u7|3w(&B4z~knuS_CuOQ7oL(X)w0^{6zE
zq<_8bk+sUoWGM2%LaqTQ<ZJH^FA2ZSkf11TM9YYVAL>zNQKBWG-)z{`-fWs`=ge<X
zz8p)jrU*WO50dJ-xzPIRIwayh5(+s_x&<tb>?(Z4%6x|bGx4>OS5pav?}W&QaRjju
zlVUy9QYVww1=?Fnp_%(=Z(M~ey09?oWq%r1lx&0Jmb1Q0K7T!jOV2*h$M?iht56^7
z826o>3n8)e>*bAbMw>~&;DmT;2}atJmUv*Y=ykrZ4%9NAt?_Xf*MfQ>|Eol~D}WAf
z(1X5~CeFm6M4uUVqf2i{4U5Dio-Wn+yMaWT5Y{Ql6?blhP_QU9aAUw>UoP+Vlc>5H
z`H#A1*C97iu~Jc~3X%P!Kg$=)bNrLbPgT-^mjInzCUYB>U-&Vc2iEfCxU^}<7%<9N
znKR1ZBf`Nq%Y(Krd<rz=A+`hZ2@^IL=(bK>2(jW|x>3suGqqsv9j3k$ZTKL5Xs^uB
zqoc>Qg~I_jNMDx=AmGw{lgn4^d{izwSA#!K1Dyufe++|8rV6y;l&rG+41WAOO#=e~
zkr=3?+;nL8Z2Jojc5y?A!L1YivEoPB&4ayOF&HGPe|J6;w|4pD)DO4Pxs8o-m3SW@
zQOvq_dU|>At1XvyyS0IvTVAEj+}e1yMhSH`87M70yM6jvm*)@ujTIMpF6rJHuevkA
zlP#oWkgxs93W0~HTNxuLJqAxcy#S(aUg|p#RnIA^;_=1+`BP<$bisc^f0Xi!n@t+`
ztLrqn&0n@|GRGM^)nX%wnxkg&-5rI}=U~e0Z{gFxDbT<ZV1Gx;!Ar^5$6KF_&J+8*
z(SZz2-rH#edIGp)uFR74?;K_@y9LkeMV`7Ma9tz{5f-?&Zc#7{4=k<;LpnpDIoEPB
zqf1)d!8eHOcglCpj|YaO41WZQi*aoXItItkP9ppq+@KXECMNS1=3wPvnzgVGC~m66
zb$ssP{UJ}7J*t6Q)BgEhC7ipyx`5$$|G=?Ev3$9@oSW--s$0lEk$Ub1EOS>YSm{Yc
z$J?xLu4{9me33sb%owRf$Ik0N$Whf%&;x`tQlKy~d$F7IgSp^upcCaxKK4&f_g9~o
z!(w3QT~^sln6#*J&G@Qea6BnD#YGB5wF$pt$_(fk{=Nv2g;jPT!CU9U9s=6PjQJ?<
z`!oh^a@O^&C@Rb{_IdJK>;NQ)4N$8_Fe$YP4lu>hC`)N!gk3PjUv!&SRBOt?a~%q0
zPI%Ixc7@4XNRqVW;K!DBKW`j7eEbviC-!iTMo<q9>nP8o%$aTk<dxG*>qy)8I$mOy
zt|Z9Zb$T5?JMhMGHctk9S9%8Ly2qy~pKMkdO39BMTGO2CeO9&>vBP0qI&_5}<G*B=
zTJ(;G80v`*|LiLb=_KgCEouxR1tQQJVoE?g_(AkwO1Fd9TLKb39E=m}e~h#D8nZ)s
zsLIq0i1V;o{k#m#oasA?x&~%r+x`Pg_lfu7KM@(PZ=VQsy21i1TG1_VcDoSf%WQ+k
zFv5UQgWh`zgt0j2zw39!(ohy1*Z~e6VJC5%ZK=m`jK}tYkRhMVgfRwHP^`ZTyBm^2
z^o{>Y?(W>kczZ*N$LbnRxwJcony3SNw=W!;i@5jlEeMcEy;tY<5=K2bbbH)5247-R
zV||v(*Wq~%@3^{OU$D|<j&VdkO-b>K{Y1s$<LyLwCx;3j(D+@$c=p=9Rz16@+T3J8
zj_CrKmBd@Av7g=23`~EdgT{F|{yQdVDd2KUp)!b98clBmwt=xIU=tG_nZD(0JuOaF
zeSg%VV6>i%iu-S^ym%XZ>3>PTPRaAV@8tHuZA!()`$KZw58itdYYiI->C{KNTHJ3}
zAmE2y!8>JsxZu7_L3bxS#W+m20Ee)Ss&FYoQ(4&V6q8LdTct6{|6aTsp0<(|#G)#Z
zHvmiOK{kjACbtEJJ~?;Y!nu?gRcW@`S$}0)omu>vMv!~-W6Fkh7YiFR{n5FykP@el
z6!Dn4e$gFtFq3=lDZ^(^5`2j~um%F9zvnbHtmWUQ64Rx$=D%pmeBp+7#fD&`x5VgR
zc9(i`LaSZk-0s7cIg>yP)oRP>;V{t>lu++N#>$pDUNn+tYfr$Qz?>{E@{hjmly~P}
zQ}sATo}>$@z;~cJ5zkeKsd3ijvNyODj6K@SpG)wt#9LTP!v}LwdsMr$h<>P5@~<AA
z3|Z5+m(%LOgizhEy=2(H{9Ro&3PgVyUn~iZ$1pa+PzXYh6uQwA_3io@_@yH-Xo0{i
zc6rkMfi=+>Voo!&j@(Ur_4RJzwY&A@OT_!)`IS^vlb*r{B<cYew?Nf5l;qxd-iDUl
zVI7~8)MSzW9c53$0pi7$KpEDX{QZovH|f2J@khUs`n)e4JAhNMt%GCFkC8ij(n;zx
z>At5yKipHUr4V(pK1F*9_p&QCqf@EMM&F!9w=oawXs5Onl?%8irVVBLVv*eZU7adN
z<{wGThV+p<I!Sb;FEl;0_okci9D!XMiyg<&OiTx%!VF*#XG2|j)z~IIE0H`YNiYT3
z#uIWYDa13f`LUt}^D4q`VLOGDuYZsFx}CXo;d|HV1-pbHQGvAM|4#F9c|VhyZ+s7F
zN|=bmO&Ryt=up(`3u_2_EYtx#4EVS?>nKH;iRF<-F^RiRP;;tw;a()}B5_MZ?ZY$E
zfF+|}8!QusQ2#g5wsT8UA>`_WWC;dGl(LYdHYqohdZHHHa|!t<27x?1r=tQW-W}EL
z>6xCk(Izm8S{t4_<=1P_uAX)%<yBR()8P4Fs1ns>KtW#6@xT>Z4KE4QSb&n>58lvK
z2p^KP*?QBv@!rDmqr1a>mp>!6EeM;esmCx;xqv`OXe^>l62vz=HTDMIv^V^mpRA0(
zof#<~CKynTGv*;-k;B(6z!pGVDxTTm0=t}SM3?#s%j*6}pJcxNy13Q@y&XcnB`3<q
zPYs32J2Vhk`<=jqHm)e~j+!q&`juq1fDd|iQosS<kLSe|M_fM6r`r(Ko-^d8dee72
z?}Dxw73j^_oHfLX;tNC)*Xk@^#D#x;t947O)O`^2;X7*3WHbl&Nmx?5DyMI$1K+!a
zi3D!4y3|=s{3_K(@G`A}x9KXP!)%;OpTq$8Qd0{oT|styhGL@I{d@#F*rf2$Ft~5R
za@YZkZ~QhBnqK-f@)EhJU4Qzgff3asS^%s!!{c0lITwh|*D_iW3X?h8pU9IPG~9~N
z^PsGq8qt|p%0BC_!iGPyp3W)Po5vd^lv##q&Ty}7YOPyf`w#fbjT_;5i>O_`BU?v0
z%l8N}hnYtmS&d(Me&+O~vG`KR9psnZ-}WCf_8YPCIcMBZuJoX@uDj^R)so?Rd?C0|
zWgDgcxH|L8$)S#045RX#HYqHr+v$^#brG@o*snM8OHbhA&)6GYBb4~_C4O49(ZDrS
zv8`gcK&FK59ql15`r2)u^p}QN`rP3M^CubaV}`O-J}Nolp1tr~>UqE2&*(ulF7B0p
zf$(lq-rC*DE5s?Q&Y%>tZeTaVn=pZ0dsblHh(z8;U7StyltCU(g<6q><2p24_QZK{
ztL^*gyBsQ-n{Ti6bp-~kMI5{wz2*9E-u*>&mFE$BO!}A&D?&t*3>GQEm*W4)VS9yw
z6WEG{Dd#~Wmr{`MrVzJ?FEKL8?(R!@2lbehc}O7eqZf@k@1gJj#ZNXi*43w}+=RYV
znlWoguCK^|Z>?lW_cO=;pMU+8_x?&zGKzkj-Wp5lFOwLy3~n<_ocs-pqr`IiW}*CX
z8|4vZ=pyEtteEr*OQy{0IAUP!M;rO;g8M1aj>zLW;vmd8B2Gx=Ux-`K>jnl2^HK+p
z2{;HYt?GYPTUc&W1WPa!J|OP6yvg()WiaTO*m-_Vw~-WKSH?>h<_tS4g+jORYFEgK
z6%X|;e1lq&iaV8VtcAH<=1y*Y!jAp<_BnELl&!$)_zk|Fv4>73uszGU?06bFJavYH
z`_VqKHkepA^cYD)OdkWi6jNT``xnE?{C*<W2$ZX)_GX^QeRU*Ys>Xp7MMIvh?H~R+
z`23ksgIzY57HbrSKsgt<?!sRHQp!0FPGh9D=ZA7m#{lg2R9*Y5&)LzbrCm;}nj{U2
zl@lkL6A!zlLK~Bu`2+$#a#{~dvNkyVrSovDsqwrn1_rDVQ?9Yw`1$#AdUK~_qj$U|
z-?NcKy7Aime*b`qnz+)r*g4crs-u;XQYJ5fRDMj*XsBa2B&>IwIS`zcQMjdCsmQUZ
zuvMNc5oRYE`}f|oVDHL}8hetz1El^zGPfcqKZzTi)|kJ%zxeAe-!pe*1<CSuWwZ~-
zn|Va<=flX!%k{_6n_0Jdw^!}37^7sfbxcc!sZ;s~SEu4TN{X+e{0{SR3bl8TMm5d9
zDXEcE-NDMIKI?+^nZ03sS>6J)e2{An0&rau#|DvSqDZao1DYsVE8dCtBQ=-tSnTXX
z1L$939VY_(z||5#A0mGGLnqN^iJb0QW|M7UyC3ZvgwN5RaGYJnS2;sNyx#5-_(*A~
z_87GgV=t=*YhxG1_Zl4{y>jA(Laa+OC@chlO~~7d5;nKki@9K``249NlX>TCOixrn
zi4iceZ6GaoWMRPTF4E#6y2LviHBuZK*n9gaw3a6*uJOC_TeG-LHBF9q?tMJ$^C(Y0
zwnq0Wc%N{`+k&y_nj7Ji@%)<_pSVBkT&j-R-mrVzbSZva?&_vXdatB)xJK@YXgAo~
zLfHeA8d52Andkh{$$X0>`br}a<)(SV%r+w}L8b^igKoi^hkfWZ9nCkDM>G`<>z7&1
zuQ00=#0+2xcnn;wv1)16=^T;<UqIuxMB}v!$q5H0Zo4=xH(5VBBO907@hDmQuEPxW
z47wf(_C&V_64%)#I-G}+5f_YvDd#Ypv7nf+qh&6lk!N=p_TAUgsqP1L<M+&D4TwY8
zEzo$ko?I>~BtAgE+fGn<l&w?)ds?1%L<X_${Pxjd=eU#`V0>Hv{XT-F5#JnCDmI|X
z*vsUtCv@xMCa%|dz17xaHDSRFh9X+ujbx)~A+hIr_6BqYML=zOz^h$eM9u|i=}fI(
zwK8*QQaOrdKs70>V8HzF2DWUZf%n6T3FtEXr!h>b9pw;;={Jux{0YOF*c+l7uhN%A
z7$f?J)gWRAeyzU`&br@0b^-012uH?;qM8+~xFgSN;Ij|IFUIJrq;B&sGW%KdOWWz;
z!ZdEA)|-R!9Ie^9YCQ8lAo)qd>i=YYnzF{_@+y}3I2(zsck`V2u-T6hZEdZDzoT24
zE9tZPqp@>!Qr=&+bS7*x(_A+*k`~0e#hWf+RpVYg5^rq&J@v1QddrGA-ufOer5qR7
zdTR!(jp~1wgCX4!mT;zl1pO#U$5hbRvKBqNUK+i#C-cxp^@*uh?2gg}y5wxH#ffQg
zmYg^^07bWP{(8h@i4Hc$m@AJP%OSRK^{)q|ESX0AI~IS4b^$Jn7cv(O+T#C70%!ki
zP=N9kE8V%L%PqFbh;$n{7?BRgg?iy^>Tg)F<TjM>m%DUi^8W}qtDriYU|mB95Ind;
z2=1<ng%jM}-QC^Y-52gGB)9~3cM0z9?r`$&+I7z5shX?rZf0t_-><vhXAZ%5bIrJK
z+%(CTY&C`MNFpsipB8|#R;%ZCmfo2RAG!;3%o!SbaVBp+^{?>0Ga(#gn!twps~FuV
zh(f`$hpBM5fRt_H15YGq@j|&CX2IGEyo&=(grPJvQ6JsEf8{q%J3&J(yG5oqnjbo8
zG*z~|jJtNTmqh&3^Mw1JF+=RVH9}due6~#2D(ea4reB9?c2=_|o(@DZhjfc06}X1p
z_Tu^~YWp1Llp3yv()IG(Xy$@7)Y#X@tVxKk<EDz7Xygsfp8Sf@ZnTEklFj(cx2Hu7
zf!HV3|Iblu?*#oG)UjbEiVfjSWHIuE7c`R?%h|j>3PQ5q{TVyhB2Qvmcz_TS<{N!v
zO|lA+P-xfSm*9Bw@tU+aAPjYULsnvUEUoR*Y=QHUv7@$rp6>4#t5n)wtW=6jCp)h0
zGNDz%6o=3nF)vl4>^P%s^31oB;LEHTjX7k$G4Z8Z3t|Z<q)LSlSGJ3_=b`qdwS$$H
zYnEX=<hZ`Gp-}U4xtOXFGDn8Gx8Rsg@?qEm7voTm(e2j|Vh7g1!6taCHq5qntP9h_
zquYeb5HY0}nwFxD&a_-0p7rLqtPr7SQb_25iupuS=Yed7TiaEe2EA6_sn>J4@E}-X
zsi^rWb0XrvelpYI%shJ-X_ci!zo2V18F3*i2EC_0dHJ@gb9>ZZ!p!4dtLZHi11*h^
zl;_=K2{1@H`B3EYTz}tJzWAWsUkE5+4Ij&OJn1*`?6{L_j*HT%&Wh^>(e;4N^7u}h
z3+}p|$#J{Z=-O1Z@gf6w){LU40j}bbaXsx!Hyyb+Y94*w9HAC1Uq2$9W(WG5)BO|j
zKf*h{K6#f5w7z+G0DS=pgWn=$%NYr<<Maxe=NS<f*nceL{vk`gv%^ncSJ8U}AL7xq
zF?#pJ!+$w+vlTy`JI^cTMza}yg_R<-*m}Q6CtXjqmOFLITH`4aWZW{A?|*7T{$AP{
z8$3NwsA)ete?5Y($?q9xrVNlHuw#c>*!AZlY-Fq}f-|pUoujLEzQ|2*5@E7zQY;HX
zc)nnbc6C@V-?oN5?QTSmxq$re3NM-a0t#W8;A2-(R4R_jqK)|xj?Y9sm^(D&M*Jym
z>#g1Cdr9v?rt>J9D9%_@Nmq(eI#pE(k|HJt6vut0o*D9}^ADS{{)F+3hJV~jBpU+m
zZ7-zUI|(SS8(Wqz+9r<m{4Hd2pB7J>RG4&PjZEJiFgC_GkQuQAK~>}Okb7;pu{O&X
zXgatE!4*nnJH%HeF+(&wVkp$9^vHascwx`(N1Yw#*jH@KuI!zk;Zk`jPBtC*_hH|?
zt~I%aS@bzOO!^eiC*mJ@smN;cv|Cxm?LX((yZt?}b>Z*|f&yL}fawh_kv&&Qufn@X
z_h@7WrMgm5hrzE>5lOSe={}tPhrv`0Pnc}FBMs(z4~T`o=H5>m)RW?xB|&ir?mr{;
zc^k~Pl^>RlYXj!;b2;L;#fP2Ei<pnO297BYcT80bQI><QNNS>^*@^y|t@!0RSc~W#
z+tDp1<-d3v`H8;tKS)g!jd>rtl1bdi9zQw0n7^cSz3rzo+!;8Ex8d`rn_*lX6g_|O
z?p-*lo#L>`Aen6Jw9|>9Dn%|-X5@(!bNsUtZK!}~TwPBWR|_)CXP{U~4sH|>-lWml
zyFhA4ewf;HdDXdnmD)Xy>Q5$9e=E=X+wN_9sq0`drF+bAlQ}bbvyhD3B;+TwgOGPP
z5W-d+rJK&J@#|4_95h;OQEaBIS)s{`_D%e^>uS~b&J=+K-qdV(=dX%`U}`GCAgAMv
zon-}5DlyM(gsW#{&D_V`0)mo=V`Ii%)Bb!JS7A3t8v9&9P6|djgF}gh1Fb6C_k}$j
z_5$YHLeE!gi+`gHf28VN-*wzDOG_eaZcF}XYW<nrKte!hPLcy+N`iQOmq(0`a?y;8
z;6~<6u0_{s9K^g+V+w_h@cBX89x}{wV$q{HpZ>n9el=$rNvg$}sBsbbVD_rUetE)1
z!r40-wy}nosQh%J{iKwMEZW_Fe>>OM{9Ya}#c*}>Qt-1~JCw9Onbw6q@g4zEPuNA~
zO|PxhDW#cKdk&U0tnyN&gcq8}ZyZ!P8&#4Y<_&DERJ-%kZZ2h^vw()K)z<p(%HmoA
zb-Dy;C>L2_)_8(zYhci%He#%GYmOPd>kUu9Pkn9{EXg}zRu^t5SyJy|>wTN6%Px^e
zzn=ZHuD7M~wY}?+vpu_;=gZ#xC*|1Xfdbx7Ny^_ZE%D^d`J6hjV*U9-n0t{b!Fs!=
zQ$F`U3UdWf^qf7Pjus2YJK!K-w@GsfVkPeDj1+-PaKWnR8+gw~FZ1S5yd_~ncR%t^
zL$Vyg5C~JR4koUiA-wG)k01NA`?w!?y?vvckfb01FgFW9>E*B%#8S*lo-gR8TEfi~
z(p@k{O}WS86W$uX9cb<xPj36%jq?et)Un<(Y+)eIdnsDDzY(w0Uj9hFXByLH{t9nE
zOXnpK-bFU3hjz#~LPa)t;Fbz+6T9fTV~u?ZVpB$U*Jpy;Z0R{1#gY;{grzZgdc1AT
zPOXI$f*GU(9P&x>;D<&&CfXj0Jz|C773Ie7{_t=2*mc)jvbI?4QA=U~b|%n?Aansd
z?_xrZ1e;7_1CbSAd4f@XF~m^nwPNZXA7BJ5ZQKJrTP5U2Tu}Ec-qyUWFx{Z04G&c5
za46h!vWsWF6ff(b>5sS4WEd{<|BYImD$t*mQi2Xt92G@r42|Dig&B+ou9bgnEBK1W
z()3|&I~TLmsD4ImX8N0ajsPjS%lX+MsbeGOMNC*}Jpk%??QCslb=NfeGs(_p4|08=
zRMQk~QeLnA?p~(GE1OEUU@pRK&3S7pW+N4WA*QR+=&csrxaH>&q{%jH*#%R!)L<OS
z%(j-&lqZ4d_makZPl3p4k~_NJzq-8(c_r{n&;qlRtZhjiA*SuN6<)L?JCD1wl<3NV
zS~=d5kqI0vniwHSNIxf(;reK1^pfBRBUq(J#;(aQkc48hE;R^vZ1;9_m%lqr?&Kh{
z>A|e846I9>kkA4+jFVWOUBSo0gt-YhKWi*4L3kJIIFbFa2ZiXkHRabw=eZ;%#}inq
zgAq?n>TY+A^_1+3bEwV5ai_D3@nAnPoMXv{79(K;yJqMB`i^r_r34W-3i{tI1|_#J
z0a+cP#X|a>)_g}YOD|Yo`RF23>Z6p(<FfqA$+zi*^MXDWzfA$CY*z9b<UAy{=wM6%
zN0LYwxWDjAf0?8-a|vZ9Q{YOp<@gQd-qZ7V%_`v3nq?;GsI>qKSsxHrxQ%>kNT1_E
zLlIE4%?T(N9N|c?bqv<iokPba#vbQo_OxbpG~CK%<@|Ao=}C$9ycJvQI!xYuPVs<3
z*%;pkyxEXRLXLG4cL(mwTJ)f%>M+p<a@BkpB6slXmLDZuZ?fo^I1S=2Z()KN3mGaW
zNVSq~MQp5oYqHnPW8Uy~XJzes*q6Av4ui`s=uEngZYuhEVMZrVf7C5NqX<sVz^40K
z)YjPy|DuihVUgEVO5TPO?)?3DgU%5#&v~!G*`~^J@`sV%TUw2mxqHwZyCPQUEAdZ8
z-zuP{+l%FTQw&lUMfus2fDQ+1;>5|sfO_gUL3LVZjQ1B52DoWp$|Xr`dd)~%jvC>&
zBYkfnW#jmYSP~ir>pF&7k=BHiu^&I1E04A%N(OP=*)&ubH46T;P~b~({n%u`rj+MM
zv8uh!RKPYMiEm+RxS~6>jowHSO(4J)j+UkAJ1iqP3!>cC+mE`k6Lu6{Wc`k_!vEws
zt2~Sqi|zP)*j2~b{P@(lwvpC0^!ON~T}vOrmc7Nc4qw)mLyw#tnRljpPFve_XPU=n
zp_zfiUSENsIyD06;tNX&yM&vH_Cp|LLs8{i?S{HnkwBsD#Jj2v<>TG(bGz@A|9U4X
zw|mbw*L?`EkpRpg2gF^jdR<?iH4i!Zv^tDaBE5~~3&URM>epTB1@et}oe5U6!IW$N
zzIdtN2>ZTa+px6Y5F4^I$Xcm)z*@PfkkeHXJ49>nkx8KT2$ej*dJW|=$<xVT^uStb
zxKxDYkYa(3v?KXxGfIfN3~D(l=s(!|>#XPLJ`3#)cxFzGeL>C4%*17r)3I3ru(DX8
zXEn<{O7HPFt0ivERL+k*&d@E4bIb<m?L@Ib?q^N!QN{5iE1X*94pcO$?q=rmZPZXS
z>6xXxlUJ|>v@m6RC_8j4D&%(Z;IT@*w=0Vc8${}k^A$OSx4O=ih~GV+BnH%xrHshN
zjqP8zY43TR>9_7X?RhB*wbwU>5Ms{H(mbZMfAg%s{IxF&3U%b0rU0M4f>+}1s!AR+
zStq<tX?IR!PH0G(HTOxUL}DQfP2k?<2}@`wN0x;ek}lgXJ6}hqr|Q;P5%11I_8(F(
zT|+Bm8AB}|vUzlN?Y`orO-`!%hz4-hKyg}aZ7Ofw!@qsuEllZ)X=8Fw8v}1x&CcJ}
zDe{#uFfvf@pZfM4+4V`vj)$w_-9Ip89+d<mIrF}h$&$s1VL4=|<adp>Ha2bH0m4IC
z-L_5n9TyCUDpL2itx#RrtiNKle_PixskJ`0Cp5KXJwMV-QlT3<Qi|Vu*!~G1F4R!N
z9DLgLOZYBMeh6S}(#n?BIy;hfmp(@hYwp>^LHH##0AKs+or!KN#-Jb@4{SE+b0^OE
zxw>-fl(p^}HDYIRj_ii)a@4Q?su?!wn?Xb=g_5c=SgnJL0QNj{ex$a~GL_C%-SU)q
zo*Pwvq%radh#c%qe&E`#eAZSLR-(jkmxkgm=-$P9LW1Y*n0%c8ZxE~BYpW`4N<in#
zbu+4MK_i!%70+_(>pNX`h0uUHb9=Iu0<zez>}ak$p?x89MEke@26bm%vRj;ZXiyU?
zr<FRzrPRFIaXeD!UqE7`^$MI_{`2i%y(|~~uBaDVB&ZyzKURT!Tgz@d*x<R4wifbC
zjuwE-LH!1u@NFT4`G!p@yJa!Q0L%(q<?vucHM6Lmt?V<B?-J+l9J;bPk6vEy>6nNy
zO9U#~>eA^cRf+NA@xxgw!A&d0Y|&NvXkgLp7<JdlD@bB0Cgyvw;=^8;9)N=;n(L0E
zx^l;4XglU*x_YjE_;0J5!lz)qG?k7+M&}Qlr#FSJPe*9oJobQSq)XybGwv7!{17vF
zztpL@emXOp<>)1g5VU7i!BtD7TVAvAvbn%U8UXYcvyHO;4EN5VHjH}MChGWP6%NkA
zP#(cCSV)#0kr9kcbEf|4VI#&9cUxF25kOzN5Osv(L)<f*#RJNPcRAm?T4{?x#T4og
z!Z+yc=EtdvRrZvqCK#ZrQPIX~$^*GrmP>5H-@CxO!w+e>YHrm{tjQ<VIeMO3<@wdQ
z4PF)=B66tK53{j@=i`v$>*EiKw9Nxo=U3|pX9mxk3d7T$ARmW6faQIHN@W=Ihsh!{
zwE}$^j1jfwvpcu*Op{%ip@^HT`sgVKo;)*krIv)ADw>v-g6^!!@lYpya<QrQ$!!m#
zYZb3ncCjlm)c8sYXCnoA39$EuNut)x^-5?K1Qt#k@>9`P;p%7EcUVE7x9at!7OlvF
zkoG_YdL@l|25!gj)rnL6qfC|yJ%?TqKb~Z!H&TGA+8<7W-)8{VD)R+UNj4#RQ~}A~
zne|i43nr~|WGk>Br-|n-d-Ga(?+h(`>^R<<k5~YH)wn4agpqw>?$|o%+Z-B_y-8$f
z_9>eycqKQ;%DQtw`;|AWAcqsEmc3)DhSX`H92?$DCANs$Fyw-bk*Ml$0Ad#Dhf33S
z%(L6bXH2_~eI~MF*l`te+-Q}_8s0t@U(*J*;JZ7>js*=+678F%1f2~(fQY&YDVLiU
zrBota+%hd={dQ2(7h=`qfQA6^!e5pI4ZW~2q~ba;AnrWDZmEuU`4p7f@jwo$fu^6<
z)E?-wgK()fA|E<gGAqI#8=oIH{rxMj@4(c*EzGrVu`BhH88ye31`hOjhq`e)g!4u!
zLn2r6geuzQQ%ZuDMuW#7>Vv{wHuol!Z5VNsTXuF;)?L+i8Hs+Jo=Z?}izIjw<6|~|
zOMXSog-<7Mvb3}OFY{%7Y;ASNgZb&qY0GQ5yJ?-#?KYTVFitjDS+)!v`#~CZXmtm<
zzQ)9}HP1mf#6xt-v$VY;;F?AOy7V@_=ilqbgx+Y7^{9vWkb%~%3Gn-O^q5c3?hm|S
zyC<1?_1P3MKt|gAAhPm6#(xn~gkp6f$U4ky)yzP&%>~YV;Jcc;r$M{asJjzk<3aKe
zUM_4@`*A;k9ii&-AtdLA)LK`7O%Z~!8kz?l9r-e<N>oFH=)x&6bA`>yL!dW^a(Q!A
zt?YZ!X3nZET;8j+uhlZ+dW)Hqhm=)i>st0$LA2VXU9_z(2;raidh-Ms78W4Vk?(ag
zJbX1Xa&Zv>wcpfaBX-MeXErdGVs)5E^lbC~fYAQWi_OCxo2zPHuC~o5y`K%kKuNA*
zL^UUrKKq4O;>9Z=*5tmm(sR5FG~c~X#{Tv%76@=fljnF?1lC$ZZTQWcuuAxVxqT-_
zBHzw%ms!F91$-u)Da(JEHK&Ke82TLz${&d|_m_Fawd*GF$Tkgxx7yryO_s)XSz5T3
zfEZ3vQbiFY?Yz(btm1pj!w$~EddnIo{s0};9Xu4cVy?2@3HGa&Nr)EanOX~lzT-DT
z`@)~3L#>Eby2U{j+otTG#i=d&%$rSxt-(*zI_czG{tK~*5hSM{d5I>LvSC$lWEC2g
z;K845i|YClzq>V!@-GnY-7&Ul9*ue@;yf>oc1naTg(y|=kq}sF`@+E_xmE;{ynMQy
zrwr#k{b@V!H{$2p{j{{Se%*XZ|7w=Ar5^ivS4>-h*;dv3m)uKMxi(KrQ>~pKLWP_J
zIr1Uq{7YdX36a4uC>pL<`mq0(Pgw{K8uh#t`iJp23OTEvb(2bR(zppc>MM9}Om#aM
zy4`R0yR+v0sfml-2`bSM2jUpj#wA^L&Z5O{e>m)P_wVzY8aSv&eD*{X={&{vx}eJz
z*b8`$^DzuEuI#_fw>?($M@=b6rQB%}kT&Q)6Ck5dt<Fqc-R^e{L@Rn<%`eid>)xYY
zz*py}+839$%+9eMeIm99p@@u6Jh>J*XxxYn311`rX?tGiTB-2MQR4HMm&f!wxwRY5
zln#=?5JSd`Tk8j65vU%#$A!5lSYUI~H&DS_I?bkz`Oxm$-|?@Y3KBJEJ#8aYE&g)^
zuKnqMteZOrQ&=@#YcguL&r%*ZdqyHO=;_5VtJb}*j5-+XtD3Ygg3>*5<}?;`^g=*1
zM0uKsr4`3cMe_k3N+};M&cx$>q~26D7phicVZh6vxJrbFgjW&XZj*FE6eB+3sV&8b
z;#4#LP{Iai1i_Wq-F~w7Nb6XU&nJd!cbLslmkPc*>BFj!>nNkT>IKR{Dm{IB)-b(#
z!|!8kZdEaphu<<vY&x;bo&9sDy-05M*r$5I+fFLopI3Ce&NEG6;g7S98g6e9Nc<iq
zk?ANtlJinsZC5>|#dgA8&BA}~c#<Kt%)J!&{nSI}e3@Ew+*2Ry0C#doaq(P1!p|Uh
zTa289jW6#9_YfQ#QExgUK8+8P@h^0XCpmEmLvLlQSeeOP($vd3SZ)Rr-ZjxPgfD}q
zDep=hb$&Ey=uZz4)M0?aWQ!EgFKs}R-$xMf+ne8}&*ScIU(lG@jWILny29{YF~6y~
zpWO64qKOS`Awq|Ih|=;uHDde}3Av?;EY@DEvY)FudU_u{wd8;cX=JY##Z#9C>!QKt
z1@`l?z3GUdiCv&HtWlgxFR|SoZjisQCzZA1lwDyD_K!!NzeYDX7hH~-YYi{kY}Z8m
zuIC<o0DUgZE8r%(z15&Z;<&B?KJd1im{Fy<t)-D1pHyo2G5eSAIfhE5RU^4^&0eb!
z!1bE@7Uez4c`8Nra_g3Gm|SwZfe5MO&s_MzMCesISTb<8r-tEv%VYn)k+qu<q9vtI
zAv7Gps$;3B{LrM>TV6@GG}&%CSx<4RtT!*atl`Fpnd+Cn<`qOLN?y|svF(*-8jDgs
z)6-wNs*NzIKUQVm&4bpqjEaykrPvO?{ju6N`MXGgF8n@PLkM9&Q4sF9)aMs6Lr3j<
zPM(%v>#Nkv0=V5wh|`ITUz{x(b2`7?vwjtqZ+-n3)L{@RsJBm0Er?kaejROIJHEBG
zRJssxG~XOHk47Xbvhbwzq3^j8yB1jr&S4+?w1BR0rlEaE0z=>lofb=6J91h)jBwuq
zQ0~EI6*39~h%G7O1mcQoiq5IjHdWGug~+HZf$99Z)xT*xm(H$S#L$<B_<heChYItB
zb2nP7mVP~js}VXh!OGdT?@t2kTK8pdNA_VMFkg{p80lpa|I%pxK++r7wwh*Q)D!M&
zZx0Ps2;135s<_cn<~L>gclmmGnO4&1dv>K=hsUqrDHHZ{p7L%dyr+Hmb^)tdIC}5T
ze7JaRR3f2Xf}#kN3>CyaI8V01)exhlV!UyjFNTevw$as!R<-?w-%_<ELoIeJtJ0=n
z7DXvgQnS<?Pv(ZGWkUp$B&1WO@g=4N>i2A9(>hztu1y)=HB-)e#Do@*!sI9;2Sf_~
z%`q~P*B3|apwsiLiW|E6`f{+FE;LG0lpYj0wo#hI;;!qIcgOZ!5x;95udx3yDb@2V
zk*C4Da6h-&P`|(*&uy>x+v#WLH-91P+*s$J*}X=Tywr-Tg9P4k_A$Alz(sbHP)r`%
zeX%^?`H(V%sw$0HsQ;<F-9Q9mr)f~P|7*GiPhXo8?oz~UXQCHt;icaRRV~aK4Km4k
z*mRk%m+p6!1_?BA!ySjg(X$2tpK>}yEBpfZOx?BHB@ubFNQH<6AdsQt$=@EmPM)XT
zWlgHrxSaPUVdUey@SBAY>ESwxZUxo}jK5Ehpm-t4(ZmU(spYdF9%iy5&V|Ou#8Yd2
z``E(y3d>6Pu_NvL(mVNkWJwD#W&&cZ5qis4&w)cw+M0M*tKvLDLYQIW|7(q3BX&UR
z>{B3&QcJF7qBSz^uHCXHP;$LkU-;n_6pdu=i4J36{(jd&=aWNV2bxl4=WOH;*G=GD
z-6p|Ooy$oy5|oB8sr|tL2HCac*xBawh`4e)*4>9wm>$CZNeOHMlg`BXBoKUzrYvpx
zz8@Ve#p&%lRoq2C{cAiQzt(?rCrB|&w2uKe6W^oyZz5nJ8Q*s<{cG1kHt5IU|JXA8
z!DB4s{c>}7>xC=Qy=BQ**pBuxpMR#LUxM2thdLx|)tZKV>z>dNXZ+{v<U8xH+=Y58
zspuL1@2>l)^f_zJG_54-Sa?}-B2a`G{x*jtoi+i#-FQ7M)Y!HrLUmx9V`QXd^FI?w
zx(d+{RbBV$XS~Vg=l>odlS8r2ZHt(-80m2>SX<h^>}DsS8<MVErhSHUdp<22|3fHK
zC7SO~;B_tGv6k^zq9_S5r}U*K60*xgilvT>CW^CZf=eeGatvrR@Hw;GGp100@)-cS
zp9lvO9B<(z#A2jsdhf$pT3V8fU2oIEpV{B|<Dc|Piti`GSc#>*b0-iMDM5@x)mc8W
z_CE*?5b6)Tp1eKSX}hop#P=1`#gIE0pHRRyeFqCXgB~28EuT3iGR%Z3@Nqi}mWQJu
zV#{y!bA7oQ(is$0@$||SM3Z9mW4fA)jkXm8EpInan0LKTgKLx7PMu#aj4}JMi|oX1
z?OM*8IQI%?8GXI%hXx5@Q)1}4dumhChNGMbuj*U`3|orqBc)*MbYX25ySvJ}PycCG
zh_q9&h{9$~-TVJhBPln6D$k!qECAZgJnF8Z?o+ZRT|%u9REd7|e61@omIEC|bLj9B
zfdxAKPsR;Sl1W@4<w~xuQY5^8h}GdA9<_t5zrX^E?(_2JQR7l_Od-+Q6}@ox#Rgop
zmiG4zUzTWFJS9L7+-sC;(XovyPLft<7X<9=75AEvtey^D$#j{~^js~UO(CTQgGs`?
zY(g-sC*3RZz*^~;Y#~{Y{mW0y>=%7mYukItd^{1*-a95`nRJH>rj4He`@TJrjO8fU
z9Bs6#G8v^}ObHj1!g+NgsQ;5(a+U9&nPc1e;)(5rA;?cDU<tu$iS+bqVRx=J#zua5
zZm-2VVde|&TNZq~E>zm7Ag{N}#0&dnu~U7}E=frse^v6*?aixy%jAsbmrPeht`uh6
zb_uA@*e_|1*;m$!i0h<qiteINU19-ol~Ba9-8BkZ?<@pIB*Grz^75y~;7veT`Jrv`
zTX4x6YOyzQ>QKMN4}jeCt!$9F-UCARU2@;TVfG?Osv!1WV>L$`bJ@NXv%IFgJ7ohR
zWfHG7rZwo7!K#(|OuTyJKjHz}mx;;W$3tK)`9i=Bev&RO@z47~+JFfLIc|Ut4UzHb
z(YGev+i3UfJJaG%IqG*}rCq5moq#y2EBBx;jFzKe7;mzpPXi8$gY42nT2+13zP6G;
z2G6uFn|T48v{x2Lyp`RY)wdt{?3;Q%wt`BzVgYuu-Rtn)2vrhAtyF(oA6>VJ!5T6d
zG4*Yq@=k%)6C-{qo`RRSLDQA)Q{VTy?|8_QYc-hh>me#k$s<&=Q=pmM!lFq40PIL#
zX)-g?AYxo1?@(o;m%l?qFE)=VmmUuq4%D>o%(U+`v>Vjf6g6fo`98g^BDVTY?)*V>
z1bWV~AvkN`P(SSe6t87Xb_6Flr@;Hd0V*2AZ<q_3DT5y+OVY%_E5Il;qf~j)uD%wZ
zirhATk5&PG7ln5=_!o;M4Uo@IMyyhg4@_7m<p&kdT}!p*--*o9Vid?{t8PEyj~_S=
z#`EvYmG>+S8!9YAp;C!|BO12;vUB_8y?V|`)N=ihDloo#gYCIyjMU7@x!j)=**L&#
z(HdCqg#kjcu=mgmNaKF{b|6&3bz#W$!M>~Z11ZX82W_3O{fmWgQ*9yLW`nP_p-P*H
z?{{xTL2gV$7jepgKe4osU(~*gD`VO2(IlZ&Dkt_imBB8J5CVu=o)>C5p97|T2-SFA
zdW}?LjI94B{to|_(~y`rIqQ4<5akk_P3eyFQC-NkIAK0cRkrm=b%l!bwhQm=o_!;X
z_rwn0t_Q_m)jh^S@FY&CW7ozFJ5s$oDJTPUqur+zmxbmpNWNsP<95V32-gznS~->1
zF#}Juff4!ZN=DgR4s%3Z(H0bt1CCpRB2$NMY@w@IYQEiURdHDQ3UA79p$p=_R$`8!
z@=9N%Y=}Nc5L2X;HMM%=h2EJuP&0Ob(mq)A@Z<7NUXdg>qiD1Ts8B8r+5u_<tt_qv
zJgx+{caM;ja`$V2@Iw>HZf;H+RlUId^>vWj;n7;c)Yf1}Ac6zlaIN0OeSKpY>rm@+
z`6fMYBth{3)``pXQ=&HxFiV2(nQKU#MEyRPq5_>@8a29K&xf7la@AovE5<Fu>4Mg6
z&3$2E>0BfEV%gwh4QfJ%v%QvEJ+|p$<^Bo^1(@8m`||Q~dfEMfB+QS~`P15#jfuA^
z6ouLzYWhxzAhyj}QRSmSv2(54Jv!rnLv(+IoL6VUx|T<6OmFH`uO6y$!^*(`3sJL*
zR&=hoHJEmbv@Xe1JN=?tz~kdLA~0f_cKOT5Z0hL~*R;MXa3ZZ&dLm}+Uws%uKn@qb
z5<^fc{0|(o39K@&ZMwJ2i>rxkwNTv?y-(JEsELV@tSvpqXOz9r<_;wLCUlhA0<Mdh
z;Tpu)w|q5HShcZ7%e}1ydmG>q)s-*OTf-N|G5d5t@C{fu#=E@@7?kOGp*kvOK1!@G
z%KDrx@XkFW`@G=LAZqcfGC|{{>x<d2$RmLJETmMkEjLdXIl^LfQ<@#eE5vs_Oxm3#
z|D%L~D?6HVJpt>nXl{&HqGlko;=Nt)*eu}`5_`IRAo>7XJ#(2uwAxDNoL`f4pwg#>
zj^A{ngaA2oy9!$CJF8d69_=JKr;c?w4ndE!y`k!ha`8_FiI;ipP9KAnos;`!>lNvb
zAB3jo8wgG=$)o1PumU1#8p{$xH9x%rKVm)7{KSU<qenFg@;#RdTfzjs5bi<=vog`Q
z!OZadNW=Eio;#HvDPZb1)nFyP2(GFjzM+#=%oB%y-`jJzRi0_qsM>$^-i9Y3BKMwZ
zDTHgVem-g9>NjZks2g*?f|&=}sqol!F%~MZ#&Zh-G%rq407*Arx0fB5*0Yg-G-doD
ztW-QjW6gFoAfS}%_ZrSQSy^%7jrjGe(Ua#KZKKZrViN7`eD|S+qWo)~@-cIHff&fm
zN1ceGA4z)IzU)dl$szyOdf9ceve}3tejwfWufaDOnaEv&+B(GSAD5^H_ZaZKiZE?u
zHithgSd)J$9GlGUr}<hyOx)`1kRNv9%NA|Y=cxMvP8oHqsC^aj-k6?%@ZQFRPw_Cm
zF<&m#wy?J}C%08Pn_oKq?|u7eQuF6g`-udw`i47DPet)vi~XQ=WAcL%yyi4NY^VF}
ziv@ZAJH68J!RQXcx>VU>%T2lkVOlRMFBIZ^p<=CdoAUsB9>(Q2DBWP6q1R6oE*P5R
zDwtz!9|DM^t_lRWabBUL@oP&DytL<l<;FN?QGorkn#b$)@14L^;OxIz=fRE8YEQqF
zEALahqwq`IU6P%<kK5Xh4it<+(s#v~Q{fDxYmw)Lv_+gxe2n(dx3r$Zx-P-O9lPnW
z?Jo8|#Oz<PR`NGTMVT?Mym=SC!rL6t9&wGSi_q@QSheEssXpU#Wp}DGxZEGlsp<Lf
zEV?LMsChNnZ9)-WQ>Qf_cBlcLriD@Xp5mYPxZ%D#^bxwPL0A@~Qp|*w_6$J(HAB`5
z3iWu#!#mh-Wb@Xf)>BNmEO2y^lwXzn>$TsiYDFvP2NieH3VrWY)}D<v--HI^Mr*&U
znz+}>QJHtI)2}$iyz^jgZN;vw2bB6tVos&U3F>LpKU~$2p(ZflwrUK+6}=)r;qX=w
z7VoDB>rSwpmzMx)3V)3G*l%_QmE}5miL0(%x6OIUcF8Uh<DiVxUGIF@hhb)edcsV&
zu3S;}&ow4jGbYvZu1nPAVi)S^%C?Brw))1R53$Q^pOTpZQ0W|eKZ>Kti6#$pk_u5+
z0V7OI#RfRR+w2e5^q|Fj3cQ+G@j2`v51Ldu<@JhT0F>F}A2UX~$m*1wdOQR1NId+}
zhh3fP;eg|FuLBBfrq=s=Y{r~#zup0*Sgx{m>5S5%Hfnmtx<+xj{F%q2SO*-@$QhY<
zZ7eC>0q@1mZfx9q99)R;7uMwNTj|yPZU-t)9b!F~@qunoJ869mAVZAH+dFS_wvQa`
z0_K#HzoB^9?G%RMK5I;l7+Kx2h*(d@DmfThIzUy|;DJKP9j}Ur+xkhcYQ@QcmwAr&
zeJ8qF4i{59nDhj93UPv-PTE`6^ZVC<0POp{#6mF!2@&D!^(3S1nT*-Xv>GGl`5vb$
zjo`kPmfuPZixV0(93m%I^CDTGR?#gD$&+O1hw(gpKGNK-z}_m527~xx(1`VZ53C=p
zr|;{JW!iPAhmy(t+e1af&jy|3i(9<qvY;1YMhABl4o!4qnE{WZ5{}=h^Sk&dtgQIu
zgU!{+MH@Iv)W_BtX4T!r5^S{`4uJG_E`4SH#KO|N=;`gzq*oNemCwDT^sGY(hhxID
zvt^%=^-bcb;@e&A1Fwj&aOkRN@$G*JJHch^e^Ryf3>@ZqO`Zn7YL}b5($u=-Gn~w`
z9PK_V+kPCCeWPr6odpbiyh}I0k>+JbB6!LOb4%8E0j@uYshkQ=X?(-Ag{7w=sf|Rw
zFJ)pIJB6-#qQHUv{&#HX+VaK*PT8Jgih$9;(b0n!8cA491uYg#1SeM;y9HE=)#cVj
zz{5kpdeENkalvftwYQa2USZ#X;iypyV0ts3Gt{(tO=P2Ra7+nacaZI=O)x;s<T<TB
z`^W0nV4q>^29I&~rij}uaNFjDe4ipDXYRneKC6p}K!<V^NGMhjEE6kPWw@(9d3;(|
z+!PaIBUN?sK4Mg*dptQ&Y6<|D$=QT%-0+<j)eYCE&0InMr9@tu&1q}Td~j7<-d|#G
zXiges=vo|RaPW}(!dd{{82eA>kRZL`L2q=)0v)|T5X7ZdbtS$q|3vuMJs`hp*SR`{
z{oQ20Mq%`^Mycze!;SV{+We$dkeczbd8U5S9QdB;%JMXX`Scsc_do51q}w(t=1Y2F
z0NI9lNXPG`I7V%yy|}ulMr<WiTLI}~-6K#NH=UfxnfLeNR}F$t<6q~&VppQ4t_dGF
zcddntBjBTDExlGMoz?YEtvU|7I=ZnsI<1X*u87^Qb@B0yByc1K2<UZ~@X(n;BVR*P
zi=i`<i;EHZ6p4g|EizKS^d+ODox4cjrgLV?mMVol4%1ST-@JZ0P3jcUjz}16cm@AC
zlI%R>KYTmfoa`EAfKKpVZKVa;RpwLf$d)2Cd61rFAr4{><eK`mSX18;aJ|pm$Xnh7
z)<B{_z)2y_;o0x2%gpr`UhUZ@wz8SzWX0p=MT6>HhLmE1`;OhdlKD+pjp<IP=5P>G
z!oK$7%jzvq`o6GrTCg|J&T8gG)h?`Q=J99*0J6Y+sYa;9vY*oM`p=!JNtecT{1aQc
z)$ZR~x$|EiT`d_k+&~l$yy#wV^w3s4-S5av%w!Ifj8Ibps?}sKW~nHJ{1S=jiPLhR
ze#Vvh_5PH1H<qLhkg};<qL$tnDJSl*j8Dh1@b(!=uvPHIE-y;k!LPL;{e|&J%}~;(
z9|gFuHV+NqhS-$~vmB|^Xo@rm$vy`b6<qNrNqvbGsq7K9X8L=J<MaD!x*pHU2aAWi
zGgyar9^f=?t%97xoyG3jDev<h*sg~RBS<T6Q4eQt(Z^@CvrWH&=XKh_!hO0pC!`oC
z7Kq+zvHV`yInqNFR62~Q3~<tc2U)@XGFM+eMbeqW@B*51E@Z96CdW*d%c-*0p5BZC
zQ_eb0T<A!n23}2%qx`Z+bW-mWT`96J2Rz*dq0hJ<ig&Hd-&Cs*suP>?1@I^)r2s{A
zUAs?A`*(92Y1fpxzQ&nqvNjwnv2gU3YVQ1gi9^hySY7u#3**z?)vnn;dEiN7=Nd`7
z^;3NH1a;?+=cui+x1ICckHyL0{Pr#(%M7>c+3Oxg?WAQE#`*f=x1L+hnIwPLR3ID7
zm_QQfYp_(j-6&RoCw>%6_0))GeiV(OM5;D7gFj#E9cDl}L90P%Dbc{rY%hhRmIJvN
zSgB+6wr99NMD4Ttu9yGQ%+1d3&MU6j#ibF*fOmlOc$Y;!r_Lyjt#f|yP>n6y6sDd3
zK#4=xi7QJSp3a|CGPCDG8g)-nDjk9bDc*efg)=Ftp0QL^nBY6iDZ)iQmo>D3J9;LP
z$*T*J&Vz?9Tu5Njo~U_F0yJV%jA6rwXv^2O_aG^>sFP>ZB-BFT2Z;OmYPnaQk*yWu
z2N2own-nH{&WZb>joMmn44>1}73*Fn@Yp44&bHqq)eXX-6%;E0alixj`!&RF>zD8a
zQP5<E%HdrJ0{4s!;JArg`I2yLKA!9Jhw_6N_aL~1Pd~ROy8@3WM4+V^nMbSWHyeCe
zZ@ak|jX?>{oV2+Nis5D>7f@Bz>QOUH5yAETTr(c8&hxy{#OQe7&x>r1cl{(Ou5|WP
z2h=BU$$~{Zr;SrC2gBGWc`&OH4qraPh!(*3=tSbC#~nHz$v#HiuDU_WZNaJqt(rNQ
zrU|RWpt3Dp#?F+|A=wENerY$jnxrNx8)H`8MFZsF^D`e^SZr_^38g=F_dJ>%m7yQ=
z1X$0ww}Tsb2%;L;TnDgYC}y$Bt?9U1zq~d;$zIOfw?k&QR?G6ZiR8r|Swf>loEE0g
z$u5D!4H~L}AtuVMMH91I{ExNu=EEOXH!jBa$8rBi5vO&VU$l+weZJL+V5X#1ct}GJ
zcD6z2#@R!l_slDB|DkGc(^t3A=kduiWfBt{?mLdB9Yp(PB>sCCiODtRIwFMei>84N
zVL$CCm*LlVE4`O39bvuSJYbvZ3c`1vM%~lSAOW}y-&bNC_C&!72R<~nHa&OZS4fPw
zcBuC<Ox22I{;0v;;}dDi7{W$wZXhjnxBz?+NmG90-@u5y(@^XCSa=r3cg=<YJZIQl
z_7U<`%8TB&B5cCv%uUS^dxF6wA&(1ks*1gvwbmiEG;d{Ei<$T<bUNO(?rh?(wl((C
zhDI0wI<l=M*wlYmaQCsqWOMH;!NT+}h|iZrNcuV*z}N2@^}ttpqzldXIn}cJo~iGu
zIxb$As#x%o_W7XQFWGLXu@K>>v1|^-bIX3Eb6zinkDIwEyJPXbJQ>y~l*#8(BSmeQ
zy~W0g^}i_AMXROEHl>V=0Tk(?FB9HZwLaLJQvC$FPs`HV8i-C>qT9H$eT7M-4OeA5
zNvJkb()#f6C%N&&pTQ&IGa3t=iy}Un<oO7E%|UdQ*f5Yay22Nn^mIxs#EG!&3>nmx
zQVYf2^U!TbJ3)dmorMSzvO7UVnI3ZTXuwR5@Z@f*uP{%9Z%E!*3WmE_vzv{*Ng;P`
z?{BfcI>^s6^8%FxvebtMxZ*+Vxi%bg@?Y7(spuHhrNY<lWs8Rm>Y&c-#17hNg4<>5
zPIafeF^2nNB7y_{st|sJ2j}m#g+nuZHD)eT?XIx#1)&KDX|m@y7Upno&r?qpL-Vr;
z*GjGr{K>iZ&c^kGz;Mzee{@$KGnp@AVZX)B3aug+dnR?1cnD_xO=lUYYi)%r0{8_v
z9nt;&m^UMYll>u)shG7A>>d7v8j{=du+)1@Z@x)>CfGJi7w+*Q1&1n}?eliZ!d=S3
zQwO2c>rZ}ASBD54fkb7jJkxN1s7a2EH#jp47dlP7B^czz)8XjiX}p$YeP1-^_W+$H
z1N;!LXzpD;^WMG_i{?b?Q|Fkj)fyU&%pnBM-pgv?<P_I)GRx9@-l9;(^>$f9g#J!y
zj}7~RFutofy5x5J{IYgiCMBchZn!@R<my?t`AYk#pV+5?0`0(+A1H2y=RzLOGQi=F
zdOm<n`kHb+P))F`4qL_XQ@li)>L1_^zm@fu6(pC1D%$uz>W)yTwq$gM(k0V#_N&}l
zm+}1xIpv1<-6Q1%WZT0Iw^UjV`eh_cp-vYDl}yrM%a^~ku)vrqZ5RV~j^DAljn1mj
zlzjH7k}f4F5(HOByTZtew6J%hf?nh_7Z{C^>IJ9h1()_OEN6?tdVRL%KmN>ivo`=q
zYe*Kum7CDqH=BMura85#n@>ePh@W>4g}Ldypa$P>ZV0LqkA@`8!fsg*2O@5efx_Ig
z|4<gO#+4r+YAk(i!6eoSMf)YXYP+MqUraEz{007PZ@$9B2f?b%l9OS%)$Vd2in9%1
ziCvsp!z21Oh`DM=c+t-l+8p=j4nSX3ogel(B0+W2`oqjotyEq&mw}F78nwJ+Vmspz
zSC)~k92yJO$kJ;Hp!6}`POEH3uoPmm3L&efkSP(H92+(x&EUXZPLbB7Ytw`!wGp<F
zr3B$=;#aD2{p%!{=$MWbBo%CDN4QYSy$^@X&0)vRy8%`XDJNyd?cAt1nUK4!hWQm%
zE`%Ai{_GrDl%8nKXw26-S0!#MeiL9tLwo5wNN-;aEBLm04sIAJF1KJl#HiOrnjNO5
zAs(aFfE_XRtnAeQ@^pA!*UU9nS6ezhCd9Z=&K~!#UxO_z$t6G_?e?=fv6_v`hxjEP
zfKPUIAYPU~-#i<KVE<o&nbF5Sj`Fo=a@#tLID-)h>E7%JlbZn$g|>GLuHiTUv`ln2
zHqx1+o&CN&clvJz;IW68wXiT_cjtN|3BYNyvJ7Bc&(2BnLEqM$^68b%+u6kfAm-AR
zAvL$q7m1I^B3D@>a7*(gy$iyu@cMjD>K?zq2oojCklnuJ`mD6%%<{{oFPe5A7R0Hb
z|LXosM;j4bU7@<oQfU;ZVr)A?o!lVUnDMtICS^Zvv_9i|1;aC#RR?HnFXQK0nY;F`
zO(%WgIme39;qbC0<+et`&UZDDdr$K*?z|@Cye9J0t|`KO<K=GXTFYCxh$%gfNzJ%j
zg5M1bIRV8EpS3LI4mX8!&S8);^6Locq`I=ADMd?ukLKPt)2c1hOo#-MdP{$dG!Qz^
z;~rRY+E)98c=X-s5!tS$0$lrt-U_xkm*4*5p{EfmeYCCOyw7V1IWLI8;+rQDOwf`T
zXqh;{!1HZGj2ODUt4EKCn;g19;R_*2<O3}2lmO+PYmIw{{n|Lw?RQ75Xl~a!d8Y$(
zqTaiZhc#tbCry>gpBB2UfN9*dWL?LP);Vlz5J<Pp+<gV}UZ=(QbwCPsNbMO$O;wf*
z%IA7uH>YQ0!vj}sQ-tnFrl_e>-B~uI)_TBsDX}Q+oVkFQWc`-WR&R#8Jhe>Fv*aP=
zmIX8Pn!k8igExn;;}qsQGIqS#WgE$A`4cA}S?B;&V@>pkSHTtU&jQNT3DI*}8h0Ks
zBOAdM-XN)JpAf?`sX=+{0r`YU4Q_8Vb}Vd3nFjS?dInUHfm;@_PCm!51rxWHs<bJ5
zhu5lHVBketxb5Zqk`zpN?en^BTpAZu?=~K)c<LgqW&|Hj>r2R_g}T|#>AIg`G8AH1
zCVinyNwhafMb?2N{4d0dZSi1=H3FR5ufUYOw`ljH4I4Mzw!bP$3#XY~yG>P$xw`%j
zP!+kPz=#UIvO6d6$rL#sO0LF5q>DYzyI$jAW9(YmNN4$lhlA%~{*^PBSrH@0|3plv
z>$3QYyv#c;*cU0*#kP}UqYZh%h1?W8&+~93JKoZYlZ5VE;xE~XdP@$T7mPi#^l>|D
z;$!wazw$ESXkKDo@JcwE@O^zb9X^@kTwGK$iYW_S_QMS8UxH{A{t>wzY+MkA{R{qj
zi>5s^fmEg6p30(FOgH<o%zG|dsm2a!b)+7ozeHoZ>Ntj~wN_QDdmV!?^M`4ZsmCy$
zB-I`oY<W#aJL%s?Q#-*HDf1&&G~C4*=m}FvJVC`t#@AKf)a2yrEzpUZRv^`cglw&c
zma<qC^m+~zw7TWn(IR1eto1zyTX%Ztr=_jK$MnXg*kO&He%*sCUXYcbMO8R`-q-U{
z6<k_JYsh3lt-n_jqns0GFu1MIIRBPM&O{!|A2#8)45wE}dX%4<og$Bh|56m{$buwU
zgLDN|+hO^w^gcW;xx&Jg5)tPpX_LJSm;=&zPeAStf}4Sz;hg6;aMzqLstRw^SX@fw
zKIfgW^KX4#eN%!pu*3hB6Gw@kuy2PW0PRj0)~8w~ufSjC2m6)EQ<}?&@YgUzgz-bg
z(Wp*$(4qmoJ~|b1F8N8fB_-Dqg#cxnX3mkNM*(<0qB=C9T9s1M3Se)l1rz(PrRXc8
zW<RLV)YNSuol<r9n-C+lN;Jg(aCFhI(^`kAbN1t|@Y4Z7a`E#CO>$AJU$|pA_b_~s
z4TUJz&y?}$4tEv7`A8#hsN$tryGyYjF&-q(!mp!i2IxXDClW`p?zQ!P9uqDs6wVsn
z2!Pbj;|Atak-@L#G1_jVIUO{hVBC!JT0J$HIHshvqf+#GZwYe569@IeD_>|V;^Lxp
zcB!lTHpq*g*_)7LRtf|X6ddm)1Pa(|xPKq3E3C>fY))8_<^E_HYIfK?I~$DYB*Lm+
zW8HXGAV*@azPwskb9V8mK?+I&oUW-)v3i+3$c7G4eHO&D=h+AhKjKm;_;si&n%d%1
z9WLN!*yIV-A}LcfK^Ax6xQ0DDNiM@wU*Oq*$+KW#IMB^j%^R!;*x#?K2RRzx8rW~F
zOm6S)wd&>ad7GXy%S|_1IrY4+Ck|gcb$sx#d=0ik`O+MPhrU?tyk{)=gI4*b<>jep
z&58AAg*$!n0#{(qw0+9<B!f2i^ijlH9|9I<s#08H@4*~3ufhkkY8vm=h^@EGBGW^j
zW+x}nFnx<oi@IOyTv@X2+xmrXXM65=7u<^EZ2;aM`8VyE{EWJB2)TY`yfp_?9)zS@
zn{*{vFKqe@sI}1_)xG8SE-jeBJ+<}gX@L>n<#<!-2}8CskM83o6`RBQri=@xVOE63
z*~G46Uffx-kF_QQDkm~2c8Iu3(Hb->hXduV5AfJjZ1ajcL@O(ye%5}A3>F9b)>qNY
zlq7J;tM)@nPmoRev)VuA0FTP9>vM}YV?G!@l<#ynIMk*dqJDOtqjy&R`gthYnrz>-
zi?5pbb%ujnAqBRY&a0F1m)CZHuKS*@yLsd+4gsk>4#zaISDCKz;Q8u;Gg2Q-F$aRj
zDqN8Zd3k5r;8=XfT+ZNjM2RsNAk)Ur)KUtgR-Kc%<WTCPd!RHHN(16M#1<8PYhY}#
zJl)ULB&F&k7CrUB08`%lJ0?wxrw`P1C*z0Z3f@=i5lMcEtTS{Hs%Ncpdv@yiR#l~5
z(tPT|;M;Q5e4-Bb%g$`cjv}Q}_kp6A6!X;m#)Rl=JWWDsW|FZEiL=6COU~g|-@msg
zDTfQ%Vz!y<q}5*u;PSJFAc|NEjteIv!S#=~e^=3DF}NujwIvW5I~=RO8emabgxb6U
zTEZegP)ood12Qzu->6yI@kFebVY2!=E#Io1G<)@6Oql#$cXWJ2c}+dhB9bT(JjZ#T
zqN7Ro{AL_X*l9;)M||15%7|qsUvQLfEhWsh#gwX^W|W8ay2Y!AWg00+-2h8nJb6=t
zR~au0{yU=4YA9H4Rm<%Z7Swtwy?;!12<FV7WhNb?et(s5ZoTPZ=qgfeV<bw<Mh`#N
zdw#!663ZQ2+u3osemb7|4cZy!+$oLxl*XM3lE_uc>CVw=n3{>p<mbFToZk%n$T_S$
z^yO0P#<P!1FF%7g38f39lhd4;c|;>8Yrx52tlFwv)Ndn%PFt6}5k7G<TH@Y|Z;ly%
z{dJm<BvJ4v6U#rzx+W)VozM2^nrdHBj8QXdBX>03V2R6Un!-+{YfWb;rqzm@p)`j`
zJ3rvPe8NRhf9Xvb?@-G?5%o;?erwMITw8Ehy+TrTYg||H5adxgZPZ(tx#xRu-Rx}E
zb6gbiv&u#p(%CTMr<WJO=-1`(TPh^gxp<rC;IACJFP*?Ex)|}ChDxmG<&Nrvu9jAD
zy4uUT3AOI&1Ui03*gO-ihVoI`9X_1+P{{3CklcYG%UIWq^yf21;`4Ww2l1~2g$4V_
z!Nlb9TR|-~BAeL8t673ag7LO?sk990$f6a<GaJlOd2OrZFUBFsK2ydnJ~9J;cBLkP
zZX=yjE-zI*D=?1y7XiqFR*6agthI+(ZY8C|_@!pwYslOMwA0AYLLExB54?q5K7tkQ
zw@2q;ZMruQK>nMyetG1`YBN?=W;F)Z3)}!LjUYlPl+4O-I}eBPW@)Fwk?G6FtFvwB
z^V<Bkm|mUX%65W%1t?p(&aTR!Hjx;+HPo{0wqlv<9V>svWg2V!xkY~fqw|WoS>@Vt
z=DDO*J5-{7;0YZPWv)yew<m{0)66j{(q+5w;lZAShpMq&yaLUqZYagM_OsgsktOsF
zFD_(htE5tmkn&To#wsM@C)D4tIlE3j-gk=ZNOz8kYo#$TCWLe3YfaCqm4PKk{@W^<
zl^E!G{!k1S`(Q35QY<vW%XqkR)DcE4hM-iY1O56oxU&eoerT7uk>Yhelie82X;=vt
zIWA}7ls@%7OAeRGuRZU+1HF0FS6UhQp5>2#t?y)gys#A}$|(%rSKVZ?WJkz0#T+W0
zuN=B%sDc>~ntu%0G!#CTyfAT-7Y?Ii0zr9ClLiFnW3U!GEaoG;$!yKEUb;EAum4_l
z-FVyA9JQGhb^K}M3@v;qY_lRKvBM)Zt`u;;*kH872Aa!WvOb12Kkif8cF&!Yzt{5r
z0c}8%zY+5Hr$2g814p&+SIFpd>DAmWQ953$%`qz}-hCJ)YnowKt(U10_zVY-DuKjy
z(@h-88*fZUqcWt?yK&4WfwqnVtajWZ$lt%vaS)wDd`T%@LRpr269V55jP%Z=MERQd
z4g#^sRc28ZuyeWH9B@QHS_gLNMj5?o>}%3j$A9#L`S(b7kNxoI>g2C}MU1nrzn*>L
z8HN+grw(PoMutQ0F`QTp{bMq~A+;)u+&`TMm`;z;cBdM*K)6Uw-Rro47iNZ9dqs|Q
z)zsI{#Jj`<Io7#)+PP#Q|2vfDW;f)z;H^#O`rt9AGSz>uUbs~+m+Gd7mq{oj*RZr6
z^_-E=uGpX}HrapBM*Rl`{1xpJ%Wq|vVfn30Cz!jlonIZi&MHucq})9>P-#@;nS-G{
z74BMfUOJWO>K;$0|6J!cuUmut>+O~9y3^PiLN7zy1El_EIG4U0dcgADqwS|oU~z?e
zsPC~wSpD&QqgBSwR1lTz(_I;=MD9uxTl(wCZ!k~J(;po^OkrZR)P4^zRck)$>)c^o
ztRK_g*(mJow64#;fF^eg=P=igg3D#ze95guT_5f?p%`UoWkV(b_~5{t3=K9_xwP}N
z{HR>A#aB&dkjW{o85C<-F`&FMK>8AgIpu&fMGo^y0%>X;4)a=Jyso2YB5aH<kw0V2
zLq6XYI^Q3tzDI@f^wQ@KDbL6FbDl31-hUPSOnrVZo>gS|(=yGBMzH70OWF#H>@gR$
z6&@xF`l(yc9WR|e2yOI38(tbqxl1oA=^{n7?vljDcAn70C=;McBtWC^fNnL}=#88^
zz{vfEP=|Kdxh;^++)?yYdgYbOh2%EU+IupVy_Z}NgLb#fUP$*md`M3!RNc@Q%e4s>
z%Kt&PhJAEv*ry~?wQ&Ujv8b9GYp6X?HvNf6H;=;2BkY4jjf1o^*U%f5U<GlHngOXB
zxfwU@VQjT#z~BD2#(<-3w5xr}o59{mjZ@wGw=bW*ZJT}3<VCnoS#8b!5R*o}S83b3
zcjsl>XXnbNSu9`YL(20po=<ZU@Y5mh%T(v>%jo<pv@wAOCEsil9i{+0VZ(&cfkC!c
z<Ww`t23pj-vD_SZqh=pwqT{2FIwrKcG-R?~&J)|$>LC|tQo`PJwXK)+i~4e5f5ncR
z^~9b(q&(+-Gqa~>cZ$13%<=R-+5~gsZgpGEe8)nUOQ3|~&zv^F8g3J$4mD=xS<_5!
zR4XCdeM2iD%OmfZ!EF(V>*KhZf{J%sgzeo%<h|R7ymuRs_iiJ?WpLo$t)KdZ*-QD3
zF?-_Rz~1Ep%eCHP=M0>qc#o}^-$#AN=I|f;?MrMfsP7n^BbNJa-~)zxxX+C~WnE}r
zw8FybK3ku%X0GIy@+o5!u9HibT}&%DQy(*GVH77X@Rjf~+qJR3!Zg}l%FS%`#bq}O
zdA>qzx%n!&e^+k4%D+{f=jNZ>syy%4Jde=d^L%l9jL(HJ#(dlL!qe#VUC{ogjxkqf
zOYBzW+W%BfW{Pnp*-y+w_e0puZ4G>~D0u5({s&Z$q<C8|<88f+xAijK)&t(wbG+q#
zWO`I3C+*CeOzUAb-OpUf@;%I6VmcO#RS)GA-zaKFJe#bn#q5ZMygOL}k~6U^R%k4X
z4&HSXTMGEsIRh^*$Yd78`;HFZQsu@GxM<$6yQS)2Uz{dkg;6+mOy_BL495ZuK5Hm`
z++NhaSi;xj4D5>)dG<v*Djs=M*v^@OCu+~NII2D!1$_!1e)xC0^E<pdgYF##y$VHG
zCCu;F3)Ju+UKsVqO$i{QGex<c&bv@syg=_=$Tm45v`bF7Iuz4NxUlbGL5Gjg>nxxz
zIl@h~3Ak4LZ$J81tXqe^AAa%Bp*;^=HqYs(>a88V=;+=j^Br1LpRj7-F+|E-{e5ni
zp*eM8W-pT1Y)bNMN_-zwn;qlqJ&gR+u~F+zwmSPy8Rk$X<1-mX7b8*V<8z9(%6su5
z+bU~vY?X66Z0T0DYP{5~>JfQWH|MPC8YLrT1y&NXv+SzOdjqGPtzmn?>F)Z$z~D0j
z37q6hF2PAos6ExWD{)zT36;e^tCz)fz11lIXKc5W-Tm17S``0Z*^()H1lj#eK(`0_
zGp`frh0N>Ry|?%DLFz?bEEg%pYVo|{XJgMSEEiPrMVD*OtWfNk6~*kCl4{Sy?38$%
zzcTy4%9Uh2rl_pM%#?Wa-gI?6NTp<@4HHyqeSO8d1XjagC#~TSVD`*how(S%DE<`H
zipP&jscz!uLlB@lqj@s4reFiG!q&Gmg%*5jX$DJkM>^PFQA1Yc$k1{b2GDxz4TqT=
zL4AR#e9@mHU$EzUmFELF&mUHv59U07NO>;nvouc=m4DK7y!dD6`3@v3xpZpoJiWPP
z8HgriOf_~s#S5~tyymbFwU)(XAW5u$nm=QBb>dI=+>^P7)RSY`d&t)8=GgClA7g8+
z5$1Fq^^blGmv%w>qK;dfF)g)FeK_B~sMhvY?TZzPeUZ{qyG%>%GA*^swA2o?)UHVU
zAno%mJCK?N(m#@382`6l&p$4GfBe@!p6(}!?91c=a$feAB%J+r_LY~=K8?P*biOle
zzPT-O5XJ&?J=WpQkUcLCmX3|+LS?jFw=1RqdrVGRhefdMim<2(L3OOlb#j;M<Sy6A
zU9N*J*U9XofG05aykd2(Gcd4|Ja1s&DXO2cNB9iNM7gmQx;4yoVx8kMPfuXr3LmB?
zV3q}R?Wj1xu9Y(^u)SPSROHej9xNHwMN|`HAN6Op2dGM7-nA|>vLrs$m&r&yOXI|3
zx^nl6FRoi<e&kWH9=)1Li`#fB>R#yAO6PGjXWh~qpEM>QLFaYd>`oe^V8K2MpMVcJ
z;a;<6>qqXF89z+|^JIp!XP%H9ix=EcT8GNZ`A>7V)8DnzK3Zlk6q~91cSFuzlUKG8
z>C_+9hDB{eUTP!qDxQi?G*@v(13!2a9?*`B(i;}3fl-9AAY`LrzLPF<bPuyLos$~2
z{CVa`V*2&ac=MPu^AI`wkik6|Ubi;V{^B+=HNC)U$$k~Y+}`&r)!Q$~z7|G5KJ>G0
zmT}gRhqJo*)43*V_m@?7YvT;kJ5kOK%KBI#_pyTZ(ZT&QD=1ZzvvetbZh@rS$M32g
z$*`U&efaHfS*OsB$9<Qb=&xeWixDX1b6t2_9Jl72s(Sh;YNZJIwerm<y^&o^WzwUm
zN=d;-yB&bPC8N-e@h+epAl<yBp&i+KgxuI4awnB-j*$uSl=v*}W`HU{WgA)RICI&?
z&SV?$q@2%LqzQ6OTuQ5tq4zZA_aynUjgnU^h2)Tag+@oxxoBcy5gt8lZ9RBcD=MsT
zkcllD)^FLee#4gV#6W6bf}XCeEGsXoq$j$LzD5?2A5;JGb;qfTe`P6tsO<xxeK1pB
zDUo{dOLBxPwLXiSa8SE1D1PQh3Mu!KSMIO*Tw*6#6e7VP-~8+oEROR5fa7r(+dA}N
zl=66y&heU)C_Ae}gDr_O_-A&G-gcW@XU@$T6P7wjreq&Zjcur!B*^30KOPDM!r|8T
ziq^`}fy_uedg(97qK=Eb$;Ps#aL8*7_6>ElPXx}RupR|G_$I*Gj<m;`SF@A5MX0dy
zy%Ia3aVvg9A#6qS^C&!64u^OJVDJpwP*+>TDjR(&=c~3-@m!=-3ch7d71X_uB1&<K
zS56{jP*F)^v^5g+cUXP)!C3!9hcDQ*Xe2RI(_(86cXY1q78_c;#zvpVRp)8xi3Z!v
zk|`RA4~FXMT#aGBX$<E|6mC~=#)`WKxUtOh$>+q@7n6|bNZ@Xi!7fK5RMF0X#>SuF
zKC;M}4q}sqbrb}=6x-~P0_Fr+_xuyXnM|kjm3zseWsyH;gUiUm*I|CX3G?$fwF#Z4
z&oQgBMRKvzjt`n`GG13IYZJrS9g1gMJ%c1*P87;iz-J+qAjkgBB2~BhQez9p5}6u5
zjBjUWTc<Z@jaYvr-hNxHXDFGPpyS#eC=0YZtj3?xu|?RQq%<>t&<jeYKUeWuL)um8
zqA=T(G%RRM_0nfmaCNbjzzZrSti<fjsl=qFF6<g!*w~(MkBm**Bx`q|Z(`AKI35p&
zySvH6!ily;L$T`iHN(W+1pOEfhFi_8VS2*#@om75?}`}iG=!&e%2$g<nyb*WSaFe#
zzX3$12J}S(4kZ2Oo!@OGWnY5VPA$D6l6?{|1=ryBD7}aXN%U(Fggzm~NGPYpt;&e$
zWC87nwQ>qv#TOP*;23>ohNm$q#XI;N69BaIIhC7{ldihlp>*dCZ0G#Ru3$@GVrXb0
z&=QJH&TnsRZO0?p-5rhM@D6r`qh*GQuBOgJqO+;1!cZ0scMZPj_m+8mJ~;UwZ}C}t
zE%Zcj<~|shM_?=vXEq_QGR`<`TD&P%p2semzGOTxaZg|LKx6n+0aq`>9a@Ken4}_d
zl4<C=Ik$_CG_)r?Bcq5pF~ppakp-XU$RqYCsN?Y>>MXoA94U{4>51~#*$>H6WZzlw
zSa%WoK|YTaxFtd5u{!&~+45M^oGcc3ES!E!c9-O_h5K+eJT{@@u_}ip!h-<A<1nt2
zzvk<DoWHsYssLxoU)x9t{`$e^0>MZ)(oqqq92v|EBYy?%+SicBU4gg4{C^YRY)7kZ
z5#H*~=dD$r0B_ZD)X*o#QD2>tp#ncu@S`L@<&?7!rZ9eLEWl5VI(}+Y_^H1rKUKLY
z;_o+MP96vNC_k;s<EOmhgS&v(j)jBeVMc>`4TZ`Gzh-K*F@{?$G?)Dx?9}<Y;@p%>
z738KNf$>v-{Yij*Zhl&L1<L%?#cjZa_^I(@@l#W2c8dHIPTz$-l;Wp_pwH*02}d43
zotoN>+!X09uDkD1`f%p_wD5@4)6r>#@9X($WYiY<X)BHUL>AS`BS4X##x%5eZho4Y
zgP(Ff8s&V|RG6IxRUZ1d%2^rC+y~?R2#m#<@l$s`Kds8=r(Vu?3h~p>$K$7eD9KJ?
zUcW=Ok?$xv>veyb&gz&;XQkeG^^C;+C*0r;u*iH7>*+4AxcltYTAWfmIzypOJY=<Y
z+uBtt*REZ;YHet+H_<ytPrED2%*b=iGSB@s*+m|f^;(k0bBpbnIhePa+)t5HQ8A~U
zo_O*p#q-vou-w2i7hYJir)T|p<W@p@eM8wdnO^%S!0~k${|2N!L};YbUR1B;{LG>9
zU10|QOv`u6INvR!+zw#K;kuH-b)A7H_j#KE#{-s#(b3e_(3g&dH~jEDa{C3Ic%8}9
z;%~Isn%vgD_V9>r3Cak63%GI=U~R+tmfpg-mJ8#CS;2I<A{FJeRhX8R@>(11jF)<!
ziqyf##uG1-Jcg<eetqKQaCg70#Tsr2My6XskwA+t+2FIM+7p?Gx21KAYzx%`1MqoG
z{b8@$<o34Jd75IazKBUOwFaVnY;8XX^YOT<yGjbLm2BFY(Yj12Du%EfL|!kEx+{v*
z;$-D^e&dWRlQE2`;u@pOY6C-{J(6=DzIP~*h)RdZV;yh(Cb(edGt8IvDVT$AQQdWg
zJ{DFd!RzOs0izO`fq=utXpu`^d<B@RMwuni?!IcjHS7zwhJ*EfOKa6XJXm3{hP?5?
zaTs9WunnzE#^yjxrRjM1v$d|icw(52>yrTYxBoJ|)$vK{t=;J{cPQ!}%`CPWD?9z&
zgX8Iz_I5uUWZQTqG~U-y-8S7%+>QO+iJ@S?U+xdk6OR4&05`rt_0|#Q-lgTU#de(?
z@_rx6>T*_ks<<XVsv-yPYJ2yFnE0iIm$YVo0pC7M$LVA0t&aZ}dh765hu_yS)ZahU
z;`4Wm4Tplk5FU~C_DBSWx34`ADKnI}Hb&y{NMmccp)3+;@B6X0-dyMP)|u<QN1GeU
z8=IRO%Nv?0&O8j`eV?kgdKH}U=B&jMqrm#*<-L}vk6Gcf28GW$@>nc&+sG9q{#k=F
zh(%p4ql`McRGb;lgvS7hj%iT``7Fp!-yy5cN^f<ZvEJ(Vi}ltbOw@EnJQPk3lVv4&
zs575JR-G*mP3d`PNW(&rhr;RWFt(+5sI!1hc;-BGgq$4@Z7<A1k%s~v90gcQ@zBEa
z{xNu{@sr}Ara4%s$n6Nr=Hj26S{A|%oC`xi{%O$hPlJuE2MpvW!ao;i*eCK&82@hp
ze5Lqjsqt@6qWKH)PvghopQfVh6Zt2={cV7IZvI&q`cFzvEzUnprP(L)PdI%8`at#6
zBK*^t&p)fQ{1edVgr0knHpV@X&OS`XY1LWr&%z@n^Uqegh|ZjUo~fRCmi*Inmh2Px
zCye)fFcxRVKMRdQF8>s`-M0|`One;vS&Dnk&SGX=^g|IN;&ZdR0E#eX2N2BSD+2Qo
zaLs;L@cWsZ@1KS5e^U5+c>iR9zyDdz_xShE&t6pM`y%bXI!F8Z?{nWTe{WOX|3Zns
z*SG(g?)w)De_v#LU!UXsZ_EMz8*{X;|Nf=f9fiIx0^ghXJ<tWLZ?{x@4hTv+3g=23
zEy0A8vcHEmctb+fJk?ARDTkW`diqu@GoJTEBKy`TdE54jpZ}aZTyuJw<_WZme({?q
z*r@S(Sols#H4wMtFk49Kwb!Em@IsWHyGWcwz1eq2vp4%n_BAhQ&b~w8a|m~cgJKXZ
z(-;)RIHK@0fJy|TVCVo~+||UL@;m(zb2BkVei_)*3wpq=?Csk~@20>DVlc2dDjIqW
zVsvxh1^UiyaC%>S9z!Y64$j|MY>!!u0R!r3jsbbcoV1Z@5=oKwweKdgwCU3yh~=zl
z+ScqzXzO_4wsf`<HSQnfre+`uc97uQBgAdqrfuqJ+LBm)`UANoY|9H{HD4YpqK=gr
zKA7lOg(4#)V)iHf?o^DV+~hO#J=xpIF7}=~bYnFrTr4KV?Yeie{LuJL;tWL0ZeoTP
zyPW|$S#{$)vY1@ZlD%iXXzUTU_ZY?b*?U^Z1&hhN8%e!@b+|4Tz4ALXb6Q37c+hu(
zL=u5F!T^{9Nx$gLdkLAsMzZ&|kgeDVg>xIgxrx2AN`0pVzOynsNjS~!6ah%xAXoSY
z@s(t2OZHxTW%ggmH92szdSTlDPX10oeJ7#u0CYT)33m#h1O$vE6LuoyfVC~J34CW8
z&v9!LS`i;|>k+AJ`)NRK0_#SD;OPCtQ2afH;_jZ`a1=jIr|tAf@=HE*bF`!Sp6J`5
zKFZp5e$$iv9czaiq3w`go<6Cx6A})K!{Q!j$Blv}Z|7<B7(%DM4tdQuF`AFBn(wmL
zq_em4h4V0N=ubVQgQGJ6+2F94kl$OSz8BZMS@T{yAajzvS4`0N0yd{}-kW`+=TFLe
zLolbq;!@pvRd|uO!Mej>O1APs4BTDW7x-&C+4y%Tuch#A19+F}#zBRby%(VM^A-Id
z1NQ~8E9bp`>dC&LzIPqa=a@W?qy)Z{D~p`aMUFn^@-pASCz81NM0T>K!(yef)aPnT
zi!*9lenwExZ#1W^?C)gcq%Db+*)VSjc#=aPK&Ox={Q-E?F5pI3lb7+UobTWn{f=@j
z#tV?X)!`Vs<$czt@o#c?0SCp*#zD&#o?<Y)p#8uZ1Ox8{&Ap(~@8_Y-81B#I?(OPA
zp412@g?=;KbTeetl5%~6jH;70S|GCzskmSe+QcdG5Qw5~tY64uP#KN$(EhVFS+%An
zQE|(|w0f**Lts;vXlU%4>W(hyYf8A=`u*X7h|?f;K^ttnf4_U>P@p+A9W$9iL+$Q{
z?up3LDhyQs!zvjDQH8-njTl-O<c0$(I1GjtC=AZXV94L!=1w&AEs1td^)+f?cserC
z-QaE?3Yko?=~Q!IXl3{JRTvz?a=?R~LJPv+!^#8<hQ@q6z{0->N0KN0>>yfXc&w>6
z;&MfLo7yImO-;$kw#WGDc#F=7-5>YE|MnhFY+gs_M9ibmyqxxl+CI{d1`VH8DrY${
zpuV<A93(k>_c6I)84wP@a@xKpHqqHJFXriy0a4ob%$^iSVV(A3*4k=N)%r9zsq_8J
zNr^;tK|)giU~DV_JtP57f>bw2#AI&H(sj%5W7vdKuK5k)N$GjOF7{qOPU7Blmp44Q
zePnq1qSpGZp<tu0FXHlLRwjB^4ES7;L{kC;3h{W4QR=8mPp;2QT(PpdYsKz)i8YId
z>jIwS=+fB0rxrxoCb#s*r$_qgqh$03;-+gar=P*RekQP62~9tP_#nvonSkE)(0&GK
zv<UVSW!-9S8IdK7+<swZTy4MU947r_5CkIVN5s`UxS}_)G6UV}^EC#Cy6RgOZ66-l
zJ{k6QrM+=7emzJR_unu|qV;_v)A9Z-lWma&pBjiQ9Zh-yb;FC-B<Ag2(bc{3iiynn
z$#h+ZgklE3b+dRWc^Nd4I^j}{f9BOVm!2_+CC6uR^Yq3njKa-wI5wl&z8kZ1pa(tM
zGQ1kab&s6($H2p69B_w;Xsf6J<|>TX_OLArn(v3(vfR}CLbFcpaR+PN%~pG(t<Kiz
z)qW5!wOgtz)mCq9qjJOe$Ct$2<T=nR|9|SorB1)2wjMyQuB+~7)P4|mTP+p}euHwu
z`tg8x6L~_!E)3$yKpi~<9>sSn2Jf+S`TXXX9M?C><N820<P~EcAhzs#;_*0p?}y?Q
zWFw4E6_Y1<EE>w^MlMg1JPUJ!+uZ<%c!k?uWp~pP?xBl&$&=zw(QYq<FchJY8y7H_
ziG4<3_6Df~cZ{rX1z2+4o5=YzQ|ENn!Euyd#l22jrOlalLYv4l;xu_zl5igey>mq+
ztQp;3j5ESjY^EEqoI=jrUJl2#aNG*VF*qKD<5>aIK?}m$aA3?#)ovO(0ONE8U<RTH
zT4E4qBpl?oW#pkLRd!=$8t7I8WGY~;ZwEH@^uQrb*WB;97Y_O!VgPvWl+drMAIDpn
zM_T{j8ur3++5}d$Uj_%(gf^%sin0_?>$cFQUU>BbsJr|<ymO`e&V1ashM@$$lSO$h
zgJTo^jJ;E$?{t7P@b|r&!s5TzKTv~s@kMbp`H6T65dZ7P>ENQ75l7QN*Ud3#g)WyF
zxOp3n_s~~u!<dEk$tGZ;M+1zx0mdBQ*-e`=u+h3qzUgK+7MKqM%!dKy!vOPPpl>&#
zZ-M-frs!Bsf3B7`#g!?c%*QqhLAEl3Air#{|G<9VUh&k;?gIzhH?zL1hrZ0hPan+f
z3u1;*!365SAftk)AqsqgS$9?JfbO~}cHp-@IOakz%Wm2WLGHc4x^sf<mvBG788*Q%
zAEbP5%l-q+d&TvK-3Jf44|9697U<b2SVe7!KWdCcObhSj^ac&^2D$@?e1sApWFWvf
zLD-JBcq&33SIB4cI2{&GsKOKI7T4CbCK_z5;g+(R21l!{A<<d~Z5|QNBQJ>`g78h-
zRAcB0wpq^G45L&*i5?1AK=wLggKlr4qcUK(2P!)fUiYAQUPo8jeu1UD+;V|E-PHl3
z@FQ^%xkGBC`)c#BwR-O4$Nb`A1!%TJ2N;LaaRN;jBB#rhbNHW-&aO<R3lDEgix&=Y
z(O|4=aIh;j81gmRz0J*Dd!rBL=11ZL!1T9t^TUhvFAPeItAS>aXIEN}-4K&eJqPDw
z9CzI(m?I_p)d^+>w=pPTex%}%CYqOaeku=Isnc;;L-x@^AX7TCl<bqR5=$V5&PZ$o
zC3ytd@+S#%fGmzirX=H2j4#ZHog^Us3r6_n&=kV{p2$m?-z<}FZjo=S^af*!Qi+c`
zEt6vBQ~2Ir%_hj*n5I_fMO|0XU2L(ZgxV97%GLsvt%WvfY1SwVp){XhTa4NbnPe{_
zSi<b)BSuoeu8PJApUd6Y5cbuE>{5;08yR@nZEdc!IcjY#qqJd}X!JLS&<r7b9^e=T
zRCWtN<nOFve4e~YlUdnmyKc4#5Mg4coQLq=+WJ$&rkJ@UT3^vvTia09RMGx&DBLpQ
zt1{KNTD+c0kE7bsRE??kXJ=<Oi<bhQd{r<@hU^;xdv-v)Nw^8?3rQyB*(>4|LIQZQ
zSu!in_KABzCi^KoD^s4`CQb`K2Ht0u%9UrofM*v2FEC3L%CkG*SsQ(3QJ(D(R|{VO
zI%1Y8HP6<=vsrjnr98VxTq}H;{?4jAyHi{w?1D9FmTc^qKxQ8mCxvz5A7R}l1vCOs
zEGd(*jNd#W-z>+Q-56XbOD4)5Y$O#}zQ57!aJ2-+uMz*~u~)ZwyY0JK-<Ar1q0nxP
z3=3%Hox)1MQ?ulljdNIjKz=QZi+i9AU!K<_O<*RSKu1Z$D!@*fe>9OyjMsWsoRbi(
zu7=3v*w&-737}|I@ZMUbkFU;d5xQpo4rA{Hwre5?Mu6R!BfD^s<*QDc1?3bA$|+>_
z_^eTAoqd+FA4IV3xEa=5aB94B^l5BwJ+$Wld^Af=-kvzSO<XSAEge9gLs00Rl4~U8
zP(McFaqYkus&u}Ol6k7v1>%9aiWZN%-REm}ds-}Y!5&`_<dWs?P`utA2{ko^BKC%O
z$hD{D{<^bn*IlSx3AtB+c5ms~6x6p%Y<2h7sO????KTV1IofT6No+cS0fZELbP#CJ
zsSqSrT-NBRb+xpN?JWS0*KM~kq!VGyE*FkTTLGphNKK_-!i0p@8Jr8ukc0b&s2E_9
znp&njcu{$%r3psxn|A&UNFwd1GU$4dDy~Bm=u-qWA|GlJo@%BZ$shE6eB5raf_hPp
z`VzlEoRem+5Vs3Qq^qd!e=|U+<!Bqw&)axX<#;tbS79BKd)?*|B7H@0Q{v}*ZG1x(
z>z@Y!_M^h0J;1U9V7W`W5?Fez&_ZD`Gg!<N7Au401j`2LoRM=J!$492s+yDl$G5+E
z{{i3rYx5y`?XYYAe%Ikk3Ilb9<0BOqA88JN_9r$4SLZ<^Z2@U~bCK~8VSJVfrzk8H
zAa`=A`-)6s^T;n7sWP6lwT#BGqf%BzckTK7F&X*`ptihmrSHK0<}2w8{ncUD!Go^D
zoHi_mHF^ciNxd*G_x_b!F-IAXd|BPV9ix$-+nGX2TO1+znD0MWf8BMSgZ`H8mb!|7
z+u7dK)b4Z#D(YIg#aDLKZ{A$LtG?FX;jxEWT&|Xo-P7T(rSO%(nA~;dV<HtClL}r0
z1nF3cjmfE=O@W8mn9x~i(~Jqu!>^>vVe)H)bEzytTd=XFUXgEN^ailv1nz3q(4CvB
z5JHd22w5gm>kQh^*e`ICNTp?bhPwAL@-==<6^H3^kKNV^H0$z}GGH>U_HbP7E`qB{
z%Q}qpJN@nw7<)Z2Wj_q$qij$7?jpK6(9)$qOD_f4&nN5=FqDK2Pm?@6P4e6{v3!m&
zH&rK?RL>x+qKVNBCsm8gRn#Jb))xoe&ApmhUoa2_DFG7SCR!Z@X)p9hOF{!$;))hn
z7}v?7B=kl;&X?MNUfQe;wGGP6=MK5{>~S4pa`VlShrA5<SBD-N+PG%)?z#nja#U6>
zW&Q8}s-aBs6f~59UMV$@=YZ$eu^M_>xFrfR(ZkH*>q^7SrsZQ9#iWLUMlm8hASK8X
zz?&suZNA;ZNNr+h1ryPZfOavogozupIj?4z8f2tjEoZ;wrn3z?i<z`FPVHq<BiRV!
zR3!{(>}7m@pB0$>NMxxB1on2>EP$Et%~W74Vp5YDH6}H*sYyQaq_mgX)GjSxQ{%IP
zrU$|x=hL(~cfRB&tj~g+Nr8zjX2wNrOmtzK^z4jtlCR)`N88*f@~-qSm3jLM*ycF=
zvoc7EVn4&tte`#ZRGvNU6uFb?=EHh>nqnY$MAyV@Zd+Qb+9|ZE3DmB3ima61oo`pG
zmUZYy<Qqk+o|SKA@~mspm#B4(+N9u{pGfz?x?iKQNpT9Tu+iCEehSDW+uOtqiWTk*
zEn?DrvPF!wZ-Vx}2|rHev`@!PK@w#E^H57`qhc>NX>C-J1hnKnkj)46Hmc3KsW_H9
z6=!X=lB5DwD}>{SWF#+14*~b<(c7(NigfL)3|Nve&wz!?T=<drKDk4Bp4za^FKolI
z$?Nf`OeJK?vbaIFN_T&J%hvk^%|z6;^#^i?VI{Y1t(n`lbyUX0X1IyTn7HZV*|(B`
zsh+L*_N|{+Us9WWxUhX|_EGVF3G1X|Fvi`oly_7X?l!R-1hI^YeBy?=O%2j9#niy$
z+z*Aj4S`%8I93Q#neJ^bq<aVAL9LQjQ5TH2pz8ghRvoL4humECKFfC9h1x}#wLrUn
zgQ~Z#U9Ng7?Ydz+neJW5+O3q^Wf{I*v|Y@AT6j3!`OuuB=-KBK0LtK&HSPbI4;hme
zKNOA`u$XI9xIo}>hcP+X><qoZ?Nx*cg~iqxhIVHzo#aYXj-t|uIlo`U1;uRz`YC-!
zuWmoC?=h9xC7W=>5au#_MqYI@G9x%9-)#Ob(a5FCi)!Scj{|IX8Jf91o{{zO8LE#{
z*vv9)pNNKz61Pr6m#!|Up?{1q{G_yYLuFB|9d*SIgi{o@3L&Mll+Kh?)n&D`u&Pd{
zPF2<Enm!3@x{>PYdf8&8gz)e&Rdne<aTT5F=pO(c%{2ypiE2)VzD+v@INrZ9)KlGj
zA=S;Frn!o7Z(GsHX{$^*NA+8)p#P-`xuHH^A?J80o#1$x*5T#m|5C;LL=nXt>FftU
zXFXIe-!ALrdFnWY^Gx(|Ms`c<<)4HmZt#@Q#9M^v*`Qb<RRTZ`ni{)BXcLmcm@r>h
zBrFxq5mpIng$=@|gl)n_!X?6P;Y#6vaJBGR;d8=G!e0x2Bit_BB|IQJDtupfNvu0A
z91;Fe_$ft)BhZ5*Gpq{D5h@w6tCL^r__dK=1N<80*Aaf5&#w#kbt%72^Xs|%x|Uzp
z^Xn#l-O8^Q^6Mr1dKtg&<JYVB^#*>uiC=H#*IW7Zc7DB!U+?4BhxqjgetiagIi7>j
zL0=97T%V(nGdAU}S-JBmcVXo&qTJ0_?iMI_OO?B6<?dYNZY{q9AOVmqGwiw@*5*#-
zH@lR(L(1LfmAjjjyCVheZdLAXQ|`W~+&!z@y~yvn;~`WM+(-=kDBO+^P!q5OGR89j
z>y0F=1Th`+J!phqO;C+VO42G*&Je$vQ}mDB<|HdIXXMvN!W^JiIL!Pi->2X^tSuW)
zC8}{J*PSp+nzkZzzmny0<g1N}HxXEHNhatxy9jx;)@uwOXz*w6b^99*gpJ-4UxNP@
z*9OY3Ef3VP%dpkqu)?v0U-$1Z_n7x&f4;*sW!mw1)B3m8n?Aqc(hb?4!|8{sO^es}
z<9~0_vuD`cU1#bxo4ZXDL6gT+X$m=QNQ7+8uI@mf8;{@CEbr-AUPI5<wNKVL7e%9s
zoOP4O+_gu}b$H!ApWEv=_eiaKedw~#WwmyDEs^kCQDe2%Xpih)JK(<%o)+WjOUlRV
z*v(>k4)L$)X}zPyA%z)4P8$QLI}lB$qi|dp^frd0(Qu<TC<y-t{Q-#$c${NkWME)m
zVqjRjq-SS5zs*+$PUaUt5r%6?Z$lvTec%5a|5=#|SwMVFkXi-?0C{8#;Q#;tc${Nk
zWME)n{P&iDfhpoY$A1o{LI$7+GGGn?0D5-?B>(^bc$~G?%S%*I90&066j50uiDFu*
zNor<drekGklI1k13o{=mqa!IsjD%F;!cChBB`A9=dr*t$fvc#XsJ1PNprFU!5Y#e=
z>HG71PyXgM>NMEEhjZtg-@W(T-+A0?I(>OQ{7mN`jnDumAjIQlYT|z>BQVUEGtdUL
ztUt{2CANj}g)qprd<Z~2^gt2IARL1*)j<ax;l0lC8Xde}Sd~?f%eJjh3I&h><&1Oo
z^M1Q{e|f$`aFFeCk37cQq&>`$ls{k}%a2w{8p>Dj0lvXWxWM*txCd{TlK{^Zz0eMy
z;WT`QKISsQW06S)_e!^Xo#3~1GvgqAxCB#h9=7qh#9GGsJ-O&r$FA+DhxGHM%3~JC
zbR}eSe0DHzA)B}8SCqS9C*$21S3?flB*!vOS~+&|zr@(Nv5E0AmV044pGU4!OpS{u
zPaQWbqGENthC*@FFh+9NoFu=AU9Vo=z9H998QbB1AblP3^j+rOsImKiSHxqPncXKA
zQca1(R;pu}>n%#WT*z9Jy{OPb_O8K2lJ8MJj~=DyKgpqB9J2RjQ#eUA>R9%PI*(Om
zc27@KKXe<r$GG>vn6d8yr)-XMYZFjIiNsc_<1){)SH&UgFxMlTL=NUs>X9*zaz*BN
zg|Y9QdX9~(8MpUIzfZx0vGYOx7qA>6S?}vqoqO-QZ-*eSEqMyCu1n=_1<O%YFLCaB
zKuj9DUrNs89arsn{v{^%d!U=KLFQWCNjdC$7yIM(lvI`jaDd19ouc>b36j2zo7g=!
z#OsTVs=iI-%H?Z>`H^+t<GC&$dM^u^zxM~%7T~kAtGqSwx_XZ;p$Oac9xzGrZtgd+
z^Ekj<L-d!*SG@Vx%0A&)qEYt8&+F=S6XaNEZeFhE^VO*Fsp44b@4r;)=QZV<>=xXC
z8*tUwZ!B@2b+=XA-38(v@c%SCV*C?$3{T-9B^sBxfV2c+Dwq%X)*GM=%vFIZx6o!|
z*Q+bLDP56kEQQsoE{kN2UtvyvGB3~Ztts*HZ97F0a}i!!Gwtu-3!L-$qJ1z1Z($TJ
z!zZ`~Lv$Rj!?bPlrJ#Jl|L@@$^uT~3*B#;cu9W)as;)2D=it4Qxu2}%mft6PU=aWS
zc$_W5e@s$|0|0RIJdQIXGR{LJ&ht1W&pf^!U-R&ZqJ%F*eEjf%C=VhfGbBPZe;|5F
zL}X@0Pc$PSG9qHkF~^#7%r(cFbFMk(oNMmZoNKN*=bCfQegFCH^GQfZDEPnkCzvO~
zCyo*giJRXQB_$@=zlXjLU=lGf#*111LH)xl7K1fm_kQI6xRgvu?o1AyOg^bP=|8!h
zQkXK8a(Jrjl;u<m$H3We^QV(ei%y$Pf5emUjri~x{2Be3&#7st>Qray{#pLnk+aJL
zCLu_O5DtEVep*T+rCHJrey0B%BQl6z(k<ze>Av*Y^yTzzQYy(zvXi`|5NVOLM*5aP
z$jHpd%iv@{8QKh6h9kqBu}&tFJINd8QqBSA)aPuJM2d-mQo@w4zszJ#W=1kMezpA?
z%WC)y_{~ktqUx!u)T7^7zfWZ2vJ0{YvIE)k*~dB5oQ@n<P9SHGhNtDz5ZW+pnznkL
zc%E_Ics`g*&-ML5{9`(gn&-TbazS-rIlnu9_0J(XhraY5O#z`G{1@@BD1*fC{Wnwy
zUc_9qU;I$iP_+Ly^7lSd%Uoe@GIy8<#Yx5ZVsbI9m{I(-_z=JVsQ?AY2bcgC5CI6F
z1G)h#FbcSU05Aj0153a<um$Xu43|ulOqEQR%#|#ate0$+?3Em|a4Zrln^jPXDNQY<
zl;)O(OJ_^tWumfeb~D?^?q?6NoopXF%${Y(*(>bL^6qkL`DnSfJXHR!e4&C~0aSn$
zk_uIYwxYAbQZd3w=A>~1963kBF>uTrI|t=>IiX8MmpGR`aOqqP*T6M%?c7!Fzud3f
z!%9r0q0(GwuS6@oAP&@nCeQ{tKsOi!qu_gR8Qi$6zihf}yZnup$iwkSylh?pkHr)4
z<U9?}P?c4cU&XBAR*9+*K9Nu5)A_z?ZgseNwmM$DBH#&Nfm)ywbPKG4QGrVk5X{st
zYRYQ(HPM>)HOn;{HQP1&Py&<!5g{r>hX4qKB#;WyLY<HW8i6JuKNNvt&?2-3eTH_S
zBcWbs6550gp<5Uh&I;qg72&3ES9m1CiU=aJ$Sy)fUNKz^h(WPL91+io<Kh+Zrg%qu
zAW4$oC1eRr!jMc!rX_QdCCR#EOR`rB*Q#rEwcWK=m;h(Od2kWTfg!j7ZibC;KRg8Q
zOB1A!R3+6)JEa!sh;&lwmquhn8C6D?0WwhLmqlbT*`jPs_F1+oJF3Ih5$gPPk-C^X
zU(S?s<svyEZ;^M%t@2U1OCG4l)Th={>T~N0>&xr6>i6o88#E2m4RZ}28dep#ib6%X
zLZFZ<Gzx>ltgtIkBn`<z@{u9LiTIE(GK<8K6=W0HK@OBjO1zS+q$wH7G9_OrQ#LB~
z$~6^1#aB61$5+5B&=uL0xSF9>sb|!4SGiZiSC1O8joQX&6RrtvLYmMfSJPC}YSSl8
zSQFJ8T!XJoUi;b%HqTtgURPh=YJppjmg$yvEn7DPH{ct)T9r1ajkeNSJ6g@H*4E+H
zh1R9k)z(kCT-}Ur?q=#u)y*k=ie9B(xP`x^zxB<aF&GSHgWZ4{T!w&Qqphed(iUr5
zY+ExH7+FT1@ssgOyQ+P%-QON*kF_s$6m$e{6K{{)-s)`V40O(P5xU4-#;*RZp)O~a
zuWP%T)$Q#L^$>e(J&vBqo=-hnJ$pTeCX>lx8ZtRdZd1?{F~v-Ky=A@8-uJ!By&Go0
z44Nf<xIR*!qi?d$-xukN^)2?p{Tu&?{)rC&1JS|MLCT<M&@%XWaCh*?g0&DVWQ)$S
zZP~x0xwCj@?arpP(8{v%tRiddF7>YE?$F(DHl}Ul9{!%{-kd$tF0!xPXWjST|1!iH
ziax+T&^(AgI2`UD-g>BhIQ8&oL^m?`i1f(xC^X6#op_w|7<}w<;2enK-4pT?!IKrV
z2nA6Y+KhIhHq?pw(I^^6*U)YBU<^A(9HWgDje%pvG3VI)*nu<ODRT}ugU;3QgmK1r
z^SEO?Hoh}KoDfV5OpHvpCf1%(o~oXDpZ+_UHHkh;ex`f&@!7E}%Z0dRT$|66pX;Bm
zy76wVd&K?4lk6$<bbBJ6IZxd4(X-*%^6YvJy-8l2m*~y%O1y}-*=zSYywl!!@8S#W
z3(kvSAI8V`p}vimX)iS|-7j~h7*kWP5??jELSHTV^ZbZ^$p6lN_?q@w``Z8dI3N$$
z0}F2w-|*io1PQ_B;7oAmZP8oZ+ej!O)Dmh5Py7#ecHBGw0096115^OZ03rZr00sa7
z00000015yA0ssOT00RI4c$}S7%Wfk@6usSc?8#&j<sl&vkf;fxkwzpPJB&1t2tg7l
zLxiKeMk4{Mvb$o}q#vfcoE~S#AMgz<VabwDV1dMr1!9F*BC%nG#Hrim#K|NY3AeiY
zblr1br*4^u-t+I0N9At?cZoe(z*mWVs?#S9SLkhe>Tq?8GeOhTakxfv)OWZ}hxE0>
zcW8=!ad?v6r{5f&qBp%+hwst6_sHRC`o`OEc!n0epB$d0Y3~n*=V-$FyTo|Rd;d7>
z(Y!zDuulvATMk!f&JP@}(xU${I`XQM`252E++mL%`ad}A(~SSC!xj3#|I^{>7-xbO
zEAKd5qovA+4%g}1%9jq`p*Jf3Iy^}y)dh#AXstSQ_#Qo|e(&(KC#pX?JVTEs{&IMh
z7HjW1JV$qHA33~0ruMYNXzoGnM~6M&-yQbpVZGvTg_i1%9In#i`e)l|f0*fR&xo}~
z^NHBq+7sJp)(^yH6bUxuB2&4_&Qus!>p`SmioG;RMax=`RCf@`Y%ORs8XMMbAC_Y&
zD@=K=vRtQ$V9qPnQOxgz&V>{vlc9=bb}G`7i)xlpLRaTTWhxXp5wT2$;>94-d8j)E
zv2#(RSrisJRKZpng|@&@4)^q_5}WO|PEMD_fgD8QOeP{!nP^FwU`Z>)<i(>jk7cBi
zo-zIP)zzXXf;Ixh94F|c@t8rkW4rRKhl+R7#E9ec#1t}9wuMM{RFb34L1Gh`o)X8q
zEwSHMNl9*%gk>?(r5Q8>!NHs_D{&@uB-@b^g*H71v$J_9q*)iPj=YoUzR81JM?so(
zSNC^Xw_C348>^n{xR2OTTBxieky5_0(#<28_s~c(h}-DoTxDrLl8}FFK}HZTk!IXc
z<>$&6+NKnL2Sdt8Q<r*VNYENJs7X(Nc4>?D@V2ZBya0HUB3xq3Bj+!Lx5E1j&k)R!
z?*Z6~UIOhQN`hNZ>%#X4wk{38k?>uEUxTlW3;z)s#Ti9e8H@6qGw1lWHZd0G^}g5Y
zyuSpjxhIL+Ag)9|iej$t6l*D++@Q^}Q=!Y=hAXm4*l4yU>z_brK`+p6#;g#}4!C9A
zHfO=nB3KHn+Pnesq`WdxFhrywutqhCcchKWW;x|54qzYf9-VPc!ErLyUGhrF(pWXl
zGGZPfL(a8YH4;|Hpu2Ti#Z~YXAX`~cJXTMDPExGg?4jWkUvQp*RcgG$5$x0*v#thg
z$;yp(A!3ab>r?7E;GAXM29Xqd@1i?F`^c=glgr4hD-mq(*XN~)%9=dgmz7=3dpQ$U
zDb4fRW{m{zl&zzjm>ulLA!7p##5Fx~#I$?g=W`Qq4H4dU*SoOq!@u>vd~QE)m$UvU
zz6zdGN$2d}2I^pLR(L97<gjzp1?L3T6v%enanoTV+w*N_%T{5tSdXvZTkcI4lMsU|
z`3&TApLsTJsa3=kSoJ&1QLNPKIkt|7zj`iH!u7V@9Gvqz<rvbPxSJVvvdgJYc#m$&
zX#da7`GlDN0LS~?`Tzg`c${TeWndde6P=lL<e0)`NST}JPLd^+b-GjHB#ql7ZkiMo
zS+SMaQYuTKGBYzXGcz+YGc&_i?kC;8vrex5yq(?Ic>_Cprz0Wv_iu(at^m`JpPE^n
zWg$g2D{?4B6KN84&}44lVKjxN(sneBI%#{_fu_?8nn_)>Bh8|nXg1BEooN@^mDZtk
zX+2t>HlPh@qYBR^v?)!X%_v9)nG~WhMW~&k6r(t8PFv8Hv=xoz05{S&T8@^d6=+3T
ziB_gnXjN*WwP_R&ryJ=evS>72MXS+BTAi9{7~M*@(am%VJxq_#y>uU4L08g6)Ib-~
zTGT?Vv^8x@yU`vrhPI)*X*fMfyVK`%C>>5m(UEjCokQo+LDWVE(t{MB1f^*NeMcwJ
z_jEh0L6^~SG@jO^lj#(Chz_GW=wiBr&ZZN&iJQ5FPN##pl`f@2sF6<PHXgwvc@&T4
zF+7&X@p8O8ufQwvO1v_!!mILXygHAk`{@D7@fx&%`gl#6&uj78ybk4gUFzrcczxc0
zH{^|YW8Q=}<q5nQ2iahgLmcJ^w{w(Z9Oun>3*M5q;;nfb-j)+=aguFz=zLDmKb3+O
zcp^{Y4rcn27V>1CLPZ+jsk|NS$<w%#x91&rI?v#l+{HWcEc%^y;@LchcjjGqSKf_x
z=RN2(Dsh??(PExUWzKLnXL%m?a4+Y0J}=-t?&mxgcp>jeA90ZfxI|01%!_C#FQ&a{
zZ(c&ncquRAz32kooA%*-cwgR+_vZunKt6~M=0o^UK8$YQ!}$n4l8>TK=r2B+kD(*@
zSU!%A=M(ru`j+<PlW2cBfKR6V_!K^sPvg`13_g?3;<Nc2K9|qq^Z5e4kT2qk>1)1(
z?&M4PGQOOz;4A4fzKWjUtLZVC%h%8~d@Wzc*HeaXplkU?zKL(<TliMKjc?~W_)fly
zj_14SZ~B+I`5wNP@8kRF3(E2X{2)C}Px3?jFg?YO@T2?~Kh96^ll&Av&Cl?&^cz3N
z&+`lXBELjW^UL%MJxhP^EA$+{%CGV3{06_tZ}Hpw4!_Iq@%#J%e@GwlNBl8=Li6}j
zdY(U{7pR9nr<eE({*u3<6ZmVoj=$k=`8)oef8ZbaC;pj#;a~ZGbUpvZzw;mbCw)wR
z@?ZQn|HJ<Z35(D-^s*>=g?godUX@`IkVY9UP0}na(kg8-LPp9c87*UEtc;_tWI0)$
z?vfQ`MOjH!rcY%RI*l%;GwBREkIs@+Wi?q{#>*PArmQ7v%Q~{ItS9Ts2C|`SBpb^n
zvZ+jv%_Jy>m=cn(M5JA!5|g-WE?dZ!vXyKt+sL+(5KEF`iz6wSD3hc^Cd(9=D%;64
z>6Gnd2bnH2WTte<jxtMjlG!pxc9vaaSJ_Q=mpvpcb0s6)l9hSVBfXN7`LaOzq+jw<
zkcG0R6lFk4QkF%sSeD3AStfhQ-m;JEEBnd*a)2Bt2g$*5h#V@1$>DN@94SZ1(Q=F&
zE62(4a)O*FC&|fjikvE^$?0;2oGE9?*>aAYE9c4ia)DeZ7s<tPiCij|$>nl|Tq#${
z)pCtoE7!^Oa)aC`H_6R%i`*)=$?bB7+$nd--ExoIEBDF$@_;-j56Q#wh&(Ef(GT<^
z{Unde6Y``yB~QyU@~k{3&&vz)qP!$8%PaD#ye6;98}g>SC2z|+@~*rm@5=}Bp?oAC
z%O~=wd?ugE7xJZiML)~e@(uk$zsk4roqR7p$dB@q{4Br7ukt_nO@5a@<WKoa{+55_
zUqy<QD5V<IFcnaZYPf1r&8kJUsx~!3jZ~x5Xf;NSRpZoh^oClV-lVtbSbC3+p(MRd
zAJAKB1$sxVs8&)d)4TM(T1Bm@R#U61@oEjVrdms_t=3WNs`b?RY6G>Q+DL7zHc^|Z
z32HMHRE9EDNQG5IwX3L#skqu)ZK1YQTdA$pHfmdyP?kz6TRAGFCaOuQLrqpw)Ks;d
znx;C{_G$+;UCmH4RhQaP%~Crxmh-t_FlaZXC#3sIL8Gy!4$bCe*<wLY)rO|tm9VR5
z8m6Ps2m#w0oN}??DmkH)I~hzEz%Z~K7z4(E37`eEftn;}!H#7DVb6*HVb6j+3-&D7
zu~LEb++uc7wkjHqcB_J+#&mx=Q!L~g(}kWwKD(efUCiZs(wTC}Q%z~KqCjl5Lcj?0
zJhXw(*AZ6GYE8Rlby{1sJ6af>&g6=ja{s)(>=HdQGToia7PAAn0WG(|ZEtf=G3{-}
zQV0^_@gyCUmRj7{^mtS)PS0j3*VkQdGv$}8Xq{x#NufF^QfH-Jvg@Qen~8XxWZ_Jv
zv#y&hnOc&XGu?$!I+MxfOM&i8x)N_7>*h0%b*1L);5w>=Dp*!y76s4h)L<S}vrz3M
zs+UBylc;vmt#Z(e#9Ml5?zHw~3jO_QP1xKs$lcoGFLKY&B9HNJLs@E@^+ZGO+;p*}
zx3)Eq^PS_Q1&lZ55LQly709{el-&Hr-2B1g95h4GfEHWk`}|+vbEI`aPcfU#_oegQ
zxlEwXRSxvIX>INE={59KTn*&)dS0*Red{gx^uoeGsaROpn^oC-PoUs;9QjVB+6ukp
z%0Voa`}@-6Qe#1f(NY-bOAqwA#)lPp3&p(Fw-K;|df2dIqizoBWqXLa+nzqU)Q$#<
z+Vi6Jyy){hFrYIu;LFf}Z+%oxw%A`u&)hzAXG5t{q(DhqD)}^Z&J8P2_QmeN2PfKA
zu8X~l5|(udhn1@>8YsJk99FI+!9h?d1nGE5K{A|FV6l#1aV>(T#omn)SnS>&ElX>Q
zuonp|b(bKb2U79CGS^)5vRc<A{I_WkXf`GgLISsGV(4w^pwHTY*sAaK3WlNQVG;;^
z{oJTvqA}s<7?ZwhAz5?K*4KhvaVK30IQn*Q^zGmby&Yh+^3-q;fKzunzzYlBC*e&J
z-Y4OG()BuM#8NF2eP?H)|8|%-$lWlh+S-$Rt=-}8kOVN+*a4q9(ApikwL2y^c1+fX
zSVPNXAL$hTfu5?>rs@+q)!)Tat6e-zuTRtK(|qg7sT{#hzhB5j(rN3gJH4Ig+)iKT
zcKSLO@i@qp4Lde+>L8c4hp36|>ElG$?Sbjq+v(cd>Ha`x=tRu$C1S>qel(o8(=fBz
zursxRng4CrM4-#xu%LF@+q&vP?m}U@(6C)@!*=P0?eaA&Vo4!3$IAtB;5dO<+VNSw
zhMhIou(Nc-&i+rsh6A(x4U1!$iUsEAhMiM)(1YPnunmZ3tCl=8fgxZR7y-5eqrez2
z4om<oU=nBp9bighI0$2@m-v(Tq5Ysd8xDr_1M>s<4fh%LGuAQxVf(@R0YhN+1M`FV
zP4`*$bJhv}QTyTh0Yl*SgN?mrC$@fH|7y+uo*N$&n`{fu9Jjq0?d~I(FJIHEASS&}
zMKD|DwOLJ<@-&r{<D0u{`F<#M==McCj__|VXR_w>m*44mU>@^5jcc<0^50}WEKksX
z3LLo+j@Z5-FVU9ux7eHL4YQft_Ec=JZ<^({$7bO&!6)iW^*RfHF>o#S$!)i~<~Ume
z`R||JtWB=lu<ikW)mO*3$#JzC^OWCN{P8~2_xR@5SIqksgv&p%<+FVFvHIt?FMEGM
zZTz6N;Q4|1B5#)3SjW{ih+WfNvpnH`Q!RH{@BxKa!Y;K<adiRn7c77NvGQwUyxG4F
zw`PYwfxVCVvqb6}76&d^-fzFo@J;ta{(PSLieyEFH%Aln*{Th1bRUe<x%cE>;rZ#m
zyANygtJ`YmUNb(s^xi4!ebza%!ZpMVPw6(@bI9wwCcVOj<#W}HvtPeV+md9*-aKou
zb^MER%R7!!7h5s7m#(n<vf|XkSD}_Mwr3ymnYOQfQ552yXXC~aFZsnb=pYBM3}!vn
zP?*3Z!4h>qMIhc->Bp>v?{$s4)=OP9`10t+`uBUe4$0q*7j2O*{2|C*FY_<rMnAvA
zw7O?ArrD*7T5?F%J!o!EY*gp`?RiB0yFlfJ4S9|7@?ktR5r#VaYsK#W|C9Fjzscq;
zH?D_#UnXZ#_;ALryXF#yjtU)e6?AfLxgelu;>02D%^~i~(JRDNRdx$lA?)LDs`%0N
zVFT|&MX5uF^;+7KIQS1vki2~EwYY5gi+x<CI}hJ;N^G20vN-p1+|ro0jwfsXOHaA;
zRyq59#JY9<WoLh;gx9<Ht}E*M(DTyVdH%N7?wilms@YDEeX8ATeX)t({OZS!Tj33>
zEM6?q?)|pEd6)6oU)FurSK`}NIm_>iD>{7D`O4l^q1H^h!&AS@WT<DwcdSa%KK1l=
zXtYeEcEdgE2-f>Wx<3p{zpj|Ss?u`VI-_@2b|^lTw2JJ`zB9pT=j7zqtCv21<!x3{
zd^Yp@9oDbRYd(b={Hp)#Xr5m&$F~2p*WmyW*YHP+H|~q-TB4OC8?j;`^G8OfgF#Un
zn2*~D6t3dX%5D40Sbyl3xZ>m<ncv=b%`M+%+}*wDyR9Xs%aV#IAC>h}W$()x$<Euk
zbn*IX2Z7X_4+kb_N4<I0=Cf6Q-j0WJoVkD2oW5J^d-;rztzJl0fa&Hl9{v2O{Abnx
z*FZ`A?mnZH+sAqH;*CR7Dz5Kle)?=}gR!5*qI1(`n*RHddPsSC+NnK-?)4Wht()C7
zbrFkH^U3VJ^Mfvi<j;9EpS!Gg{;`#Y=a1@zZ;m-z#9H4I{CNLI=V*(48S?p0-X+%E
MFY8G^#bq@g0FloIga7~l

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Light.woff2 b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Light.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..955a6eab5bb8ecd74832dacec514b6b28fd51fdd
GIT binary patch
literal 35916
zcmV)3K+C^(Pew8T0RR910E|ok5dZ)H0lC-!0E_AX0ssI200000000000000000000
z0000QfdU(fBpj1E24Db`W(YnBf^!iN3WdTng!V=OHUcCAnotXkDgXo^1(OK}xDO12
z0$UJ_kph;7?Rf#REz^11W=hFl43(=9kL{pz-uja&Gccrp-OA>6Ko|YAL(cyH|Ns9b
z$wbC1xxyu(Z9$+iw>h0|^IwR}4AM+I#3#v<1nE+whD4pRM6gAT5LTMZQcRYUa_W_`
z?b>ZuU3&OgACVA|wkTi4!MvzVXG3ppse2f_5WNdioKn=s=LW7{kmLKPg}m8$vimnx
zzh#`0Nl5${V4bUU{T$ya3z+t^_-3kA3vumesRo^pOf5^P@}^pZ1oaQ}0SGzZ+2M0^
z@3+%$H#Os<?)*~+#Z=M11>7a8nwVitXmS85M=i$E<||E8j4!|oT~kr0OdIt-iVbsV
zhp>0)t8BnZK_&#$8({h~@zo}6iJu8&u$=Z2=wwe%x~4y1w)Dp&g{s|l)%Pa<A=5cP
z8{p9)AvwM~ei<^3?@s72Xd_t#lw)&gBr8wOcjcjXzbEfZ^4}tpMN)tk9ke}uE9=Nr
zlO<5d79mIE=KpK~8w!>NL;)2W){ZH)@(gL%G5APA($&OtG@TGr8Y%tKOzFHY-;~-7
zn#6<t<KXvackfG^g0xr?4u}$LCixS7u;4btG<@|^-wmpoZMIcD7a>G=jG$1U6iy;6
zxH+QuC8*reQ-6T&yZaNu5F|6<ju*WqB`DVH254@@M)b%5V{BsuTi=4+b}S2)#KuO9
zPQ@Tp#6sVSw;RlOKNKt9Pv7|0_wD!ZYqQ>;@%O^4AsTU-m^9YJr6pD&&Q@{UTUa|M
zR~>UCA~Gt3FddA&Sh67i{p(4W)S%K-zS)uk*<f5CTPE^o+rYG~xxCt*pe=Sd6v4QL
zA9YpVlh~4FnaZRx6}hOps+$L^7Z9QYnmvU0IJ$J_f(QWOox|u;kQ_44m;3sisl4*}
zdciJR3y0BbjL5*(9x&>bZzH3D=gsr|OK&tPl}1w;kR`N%bAr(};}H!cI-a>|Yl_<h
z1*iSKw6P?CT5U6|C*r?qGuV=$Ho#*#v-F5~I}Jd5!;WEn(Td36ypZ##Lpyz`EB}L7
zH2Y~EkbHm%Pb6zk)>)P;FgA6H9*Y*cG9b9EH0Q*zpFq=DHG^3rtyyfHNUm8pSiff9
zhI$B1TPiXzpg)Cl-@|#?HshBYmXPSx&gBOP$edY6XsjdrdH8jGCqyaPp%KX?UFdi9
zV~GK@>w(|AN%`(M_rASM2|xl<sjk;h*!Gpu1;L;Q=41>+Kp6O+soMVlXwSS4jJEay
z5R_<(q3ML#aB|x1y2w}Rnxd4}x1Va4p``;yJ~c;$P!o1W{pz5vsRyJNzyL(GZ1!iH
z{BdiDLQ*FYhU_Vv3Gsd)tCSRvd-c4+y^sW@kRb($kSK#h7|YMwG;8Nu-mKOkE9gBS
zY_$R6`iko7{Xf#oe2ry|WLeT^c4f)gSk{7IL6#u`Cns>S90HUSMrQ+dcEi*i0ro&e
zb=Ovb@_@=Kx(1=>j)#Bu*+y>fe$}Y94yUk+lb@vR%slq6F2=1E6c~yB|Eg7{yzlV?
zHM5q><}e)|$hjb7v&^JDYtu<W*2g3MqCP*rwwhaOYIsV<_ZV4I-~-q!fL-fMZz$tq
zr7K0w#0qtC1wl5-)}&>M@3iC8MoS}f<Od`MECf2Y`jgAPTZ5({WH(j!=o1zifa3rE
znr8pa<c5ENZBJDl3lkaV>IPoIFH=GVK+7Zlsj74<Nv(1V|2#Ke+pj59mnfC7tSF2+
z9<9GT(9gJdzJ9N>Igq7W@e`Uus#5Pnf$$hf-+7@3ArR8w$9`IQ1A+x&85P+C6;StT
z|GLe$*&~}{58QJ}+BJrnAY9?7j+*$f9wxxF|IdHcR`YbCPqB#&BNmF=b>zdkjA^~S
zF?O&0t1S*HC?i0k#NPW;zh9g!ujPI|>}~D|#()t>AoUp)n$N<&^XoO|>O0-CsGIu`
zLKtVCY=arx{FT-dz(DcP<cZ=#+@(XZbwo{2o|@#T2PqV|+j>6Y`!K?MB$2*veZ4;Y
z&>4i(L2<qgig$M^l(*+KTzm=|DIq0|l=j;Pfq?$nDo1wybng#d7rKikgXBs{i3SuX
z_~(>rjgc24AJ+k1N+*#ONYc<V-Ia3YK}cq-lG@bqNG~#8QHDZg;Tj=4!f?6O_k2*2
zB2!H_N98Sd38S^lwyDsFe2aq^^|pasG)GxAy5o@--nkoJH2KrGZx|jqXqyh#67{nP
za5BFY60%4{L=>@fr!xw8L^73d&XYhg-Y&kvjA%sTLtEFa!nD;qIE*;n1A6U5&ctNG
zL`#%TsiW?q`A0L5k8Vyh1Kj(mj8kw8fm=|Unwb;vKUjo;7`tr$$p@#KWr1odthRur
zO}6-O7ag_VQKy`D)#`b89v5w4bISwIyz$8vbUFV+OFygys566*IUhXa5j=l2{I)X)
zWL-sGs0%+S5;BQ4MKhJA0E-3E39v~(<Vd(r80!RhlRz>_E^Uelh$YOr2ir1b6|h~5
zELXCJq6B>MWitZ#EtJtZ8e9CFv4^6^_hYCCHLn2xrceq%z!D-Op2A+>)NR8JBMke(
z#EAjLh_n222pUU>;OmUXa(Z(s7dH$T8CgvLd0Q7up~|GdLgX}n=so5=h0p?2tRy+N
zg+-D*C!BJ|IXw7Wac#Fa$#j;IrD<ea9y|WrNT5=m*uNk+kI#yY6r4tHX0XQe(*C}n
z7@<U_G}fA!TiQSdV`O4qiKKFsT4!ouW$Q@8!IHW*f5eKUAhxWmzR5oWR#4z;Xsq{_
zqAoQ4Up=D!R`P?MhW7dve;+l$rp6XugMSAT(|x0~sQ(P+1ixvmZm95I#+u;9)>anV
ze>;2Sumb+aIV+cy<Np&k<-T(Lf94N)sXYIG-Nyz2ey#0It!UZBo!;Y*s2v`AvvF?#
zWIDsh?uby$v*vm8wE3`!S3dqr_4pU}cjebY{i~yrrJ8HPSUL@^ic|`<SM2<9(7bG_
zzB>*ExoF9<^)%2(c{H@9o2Q!5dRtf=hJdJUKwzXiDO-^4aXl^ZPyn&C#SEU&{tSKF
z(Ve-scGeBA*rTPCSedf_paizEs=lUHB6y+=Mmd+LoRT<Pe2bdAsF<#eM9&6{Q_iye
z1PCrkqaBNFCSKyE14|HvQW{fW2q!`!*QFzn@O4!_SOmj#V@U|gynI`f_!T+^2(NVT
zPr-E{tdth2{c(9X=uwD?XE*_w2=UUj)m2|ZjWtn`a<f!gX7$`gOpudZ8EO?sB~c|*
z>uL-|$80Di0C4V?QQL&kuk=Q_ockuIZZ-EN=E$z*4f>#cIZ7{~jT|XET}os*Q`9Yn
zZk6;<-DX%5&;Lv&0pB<~XKrC>WzFUB1#zdse<HC&Dw6{WFixfH!C-NC0+B?fgw>@S
zqtO{m7Mo*YT3~f3Ads;N>!RLZG?^_{o894bxjj6+ynTHA{I?f(!T|%UBu>2k_^TBg
z0Fn&_93Y8szymTg6gD&|7O$*WX5vO=QIBp|$5OEHztC2G+`4_|?!Ehm$4kYNShU&Y
z(Qmv$JeVXF_AcG^xXya&Z>Z7oXyACcM$2?>X2LixiVNbnD2YqbxGak+^0*4bHAPs#
z#6b#&B$qN$l*=c%a4<XqBR&bL{mQr<C#3k-rofSb7nOjOgNxOLs(iSh$ibr*#`hq>
zDf39BgSeK4a0KEOg`uw_jwN;4L`|K5*y<2CGSNxg(r7>(S-f@#8l|vc@&%<y5o*II
zZ(lpv+5$U;fdiuq2G^4e!Lu`9^cBJLE<j0J!8bch7WrK^4i|RKTKtBfaO43dv9+2e
zWDJLSi2pcYC~pT=Q|KBQGd{G?6HH0uIL1e#2PX#F5msbKhLx-{LU}{G<GzC%GL01R
zD(bOFFgEIHg@C-OMuvDYh8ob&XvX$Jb8rgxnvhS-VBo-mC{Bxvq(Kg#P8>`C7ZZH_
z7f5dV0kH3@hXf4ZpCLRpcu)g*egJwBkQFn)R0KRX=E&=+@I(Z14cxGY`}u`<k<I*Y
zNs7_>``0Ryb<)o`s_Was+ea5YO%UQft+&4BnQgwYI+KzK5-fyVs4(FoL{f+nEk>+3
z@e(9Tk}O54H0k~^Q69Y>>jD&)14mArxs2?Q;CIm;JbCfv!<Qd_LLy>q1PCNCo`%{q
z5EskO(7X<Jyr=d$WWNKBJLZX3K6>k25aftBh>OGo5{wiX5p=9uo$ErU%u2NiR4FWH
z#fp@yGjW8u={sWyg9R$Au5~L0VZa3zE2RZKjj)X|$|BWj%!!3hBlX(8O{%bSHI%$=
zx*KG$0S207VI%vsNsNXW&SH;pm4+B<O6n@gtFlF{&(y~1IvNAw<sS8<=V8JllyHd~
zaUg=caw6Wj0`1!uc!L2~S74A>N@I*S-KM6uxMgi@Z!8v>;*dEpm*(M8!XqAeCn~70
z>K}i)Iuu^CV&x4FKmPQa;w~QJEq>wyC98jI#eYN=B8kkwMWKmqhUhY21O<nPhKcPF
zCzILTSG$FJTq-87tnOU2&gsaWWY4l!nNSuWi<EUvAD#XW@hW)??M`)n|NHOy8<u4l
zL?TYEvDC2Y7^d;!9ccdlNqYSwF8|#53vop18(auBS}^ZAhaGj&IcUHGJi{m0Z|HmV
z4o8?-L3b|L#*{Y<fki-h*B@;<bU9B{MoHY77{3fsAveS2cLZra^)P>{({uH09C*$D
z|F@oCkdD-$+^#V9IQ;(&(1)LY1AMby@a0+od^xXVm-IrOCB?vi{^Vxc1z-H?@BZq~
z{`%lvTg!cK1z+8NI{)bcCmyllUJ7O1xYhV@{%P)?%k&lhu;roeA@q>=kZ||(y&Lyd
zvNix10$lM|yb1~s+tXA6DANm0UhKb<v5QMh>CB3qpzL1N);*v}#V5b8i@(fMzW#6M
zQ$H<u=I$N7-r`66BY8ij-J9)^C)eEgsrBaji%rLRhClrJp8~>?w>g%%;Py4DNL~EZ
z@)fKj6e%^rB3%_P7s?n$iuQFjGgGK?z1&M{<lLLvbKjUIm6y80QYq@SxJZw#sIt03
z(>Axmoh&x5(7BGA*%?xDX69)>cv#L3*=xV!4mjq4Cq8;rR_Aj>#11dD7_4ihL=4@c
z+$T&@FMi?Wrs7fG5#!rO(^1oVWNJ_K^~~fR>+iXNUYOfw(|c{W_vZD*jNVvUvnAOr
z@3&>OSP2;bf@gC!QHP5KyIei)J9xzH<h7g6F5Y|i?&Y_S{{f=?gy=&8PLfASPLTh|
z|0(MlcvUbiyqSFix-Pn>a98e*%%2v&nLX0@BiFAczvuWzfwzjh^GTU6%6;Z@mio-9
zDaN!OI-Pk-e1CM5Qtla1{reuXtFyc<%D$_bVZPk?_4RL?G`Ml|maPjGZl61Q)g~f|
zoRLYP(&!8(iw$tNJbsn{6pEm1u|z79E0l1FR3=v_Rceh^7ZV#7pOA=5N=`wgqSG)~
zTsod$&gs-!TH4w>yd_<(f|}MPzR89u#;K+m_L<Ik?vbMTo^Og5l`gATUAw04+xm5l
z>zj6T4^Ow{O;k@%_38$4clJ!P4msu&&dOh0wzPbC<%+76)oU9zG>3fO8P*pD_l==V
zbg;<|SLRIUoQrbj<6jqIs<Tn#L?upD>U3K5QS4+3eTZtGVyUk!^Q{$rw9?Ovnp*CA
zn{z^&azG~v;#47AEe>l5{8u9XD~W5xb0Gzniso`LTq%}|Me(ezy{LQd>(|Ey^r?Y;
zZcrcUpPkV)F{)<9)WTQ;3^c?bgAJuquF3``8<`nxQa8`$^}L(+^Km}RAOHGkKF?Rb
z_)WXtZ5F|d`Q=ZW{87J+DrPkDOkZ)ro;YHSP$}MIVu><j0|VQ^7f{rN&?G{=2o(~h
zFfoIVU(ln#Ogju-?`#i3)pHoJgTj-SkAeq`>eCMcQedgo%H3%6?S}y?4g-q7Z3Bn`
zd+#P8YHL2e06SCCcJa-WynwAL@$IG|wQ09Z`OG0qY=<|pYBTYh8E&_-0QEPiyr4ix
zA3!N|6raS@h^-0M)tIFC<$#+h40IC^d2YTXf48AMYPV8ZfWk@LUEkl`U0)rp70?d@
zXK8X;Z*>x1B<wswS|x#CQ-V^qz<X__<HzH=L5&o(L0z+51O9Nfj0grP9fyG|o`k3l
zBg6H!V1N`fp0VEg|8Ya%nv#APJcYrQuTr9LHahNr5RRM)aBd3!?<)rZ-LoLy28)02
zgAM%vM*J)ov~lqC9~gjyVQ2*si6f7UXCjB{LSXPL)FYp4EC5Tqg>l{`LKM$dZL#&J
z|2)RUO{9&Y40cg(Gugw9j?q311KDs8zdT!Ymx{H-%eM9)TlJwi?%lMIfNB`7G>edG
zKz{MOhY;Ieza=3~b$Cow^kXsC@eZMQpRRPN&eh=<>LwTQ>fyT;Yo_m*dF2!kIJXfk
zZl8h`Qg_nIf=Lex8zpp`#0LRYXeDS1(YCVT7>QqrU7=|lV~*w^3mq1|j>_9p+iu*S
z4IgV`wii;zLCYw;Z(Nb}<k4XJTA#jdh^j<z%bI{qjzL`FX$;jbZpqbfL1v04Whd(n
z9BYo*_N%^G1d$CVD*LLDrc<Stz)ntW&0<ljWC_vLdyj)vx`AWp+Fuh2u?qgCF)Yv&
zle4w&H(H4_f@)Uxh`5g-rfIl1w6<GQWsIFhRa+HPiAcOJLg1IO#u#C3+cmf8kNqlv
zbxw9C9q(HkD+F>N&GigCxEbg<1;pi+fm!wTanr$WtY=vl89Csdd*PPudJLGtKCgWB
zYqCx_5)SYT8S~P)M7rI7i03gJt*)ZSQ_Ibzi`b0dm_!l>`DqcYTy9hY%l%<yFJd~k
zdDxf5pw@=NpjrD?yui!al(Mv9x@P;dFWH8ws%cl%PG2lH)^<;H+L&hPN1-~mDt(wH
z-QM@??4QXwR)uZ((O!$LmZYSwM;;rAu9Arx%XvQ3M7xJULqe@PJ6l^fHx52iIuE6w
z#E^u#R7b?AqH(9t5CXQLh(?>2Y1QT+Nzo?RiwKGcZhKYnp#Q6PB7M`D{|0Tn%xWnk
z!-D_=+45g6$h-hCpBzU@Y5@_N8?E+eCbR&s7zWriZZ<J7NEaI~P`hvTXz_}YP{}<z
z<RBo9N=AA*^KAl<+<84nuyH=$f(Hq7Vz+Yb3^>A~U#6kb#R7LeVBev_+0fIj_W^M5
z0#rnUg%@lxfrOxh?8+gre^$yqOk6DOlAy1`Q_FBv)IJtuK9y%b(mqqb!yXlF3lVbC
zIEJZ^ds5*AA!L5egBF<TPIXVQjCb~9@8&}SM5O{|I9C2>J3wgNmf?Z!-k6|iG4@g7
zivl8jodAiy*})~2of#U)U^h^dkCA*Adma!aIgHHtNFpQ~Y^rIYhj9#U3hpN_`5p=b
z`GkbOk&FD8LC0Nco@vs*?shZDiL!ugP(oRH1&^5xxJkyso9XiI<%iYm29g%HYYRKH
zvuJ1|P1?11FL*!|i@nq65zZt-W&R2qy(22SATkWbWY?Fj<T6uK^lvMA$e@BL1?j@}
zDSSiA#|mnD?By@5aBbPqyd#%dk3wcwU?xQec#Gh5HS#!$w-Z9Be3W~XHJ9z~CpPu@
z#89`M+WqjKrK`V3o9f5Tr|5gwunRk3BynWbbBZ0W!kE~8*QJzfvg2?bIO3UxNiu4y
zkce`LD#Am-;|RK~F+=dJI|A~H_tVae8OyZ?+zU#~DO-wqF&!FYJv2x<Ml~p`;XO%u
zI3!$jKEiONWznfbYcWl>|8D)CKo&ByeTe9y_u*majrwqS%`?e?C@80I;Hjc>Mv04J
z`9e~7U7W0x@EauH4i#!Ae#NEwYh{lsAgX9{EXJm;os~iqNBru7b@cAYd(G4`r%&5d
z$8yu#)4AL!o8Nl_OygnwiQ%rnH~GpP>={S?*TPW$M3gZd%kj^Af_V)aW>;+KyVxde
zUR9USmgT71AEVLdQ(*SrrNVRffQ24})-jy+kbHMtHg;^e79|9la~h4!&OahEJ9lh^
z?V!3QSI!!{H}*HmZ*}>JA_Y3M75G^~mzbpVW3~;2e^7aQtLT4${9Z=-ODFz1{K?bq
zloV}%phB!aQ$fC1&MN(<t-towxftUr9et#_JEVbcMZon;EZkDX6as7oW(g^dHtWlk
ze+txe;8P;a1Oa1gc#p?4&|NxpM;1N9U3YRivY8gFc6-^;Q%G}%g3B+5J74;-SgHLm
zWvI2NhTy5H`iMwbyxj*yk#ht2l0^yB-{COHOvrm1qnZkqG9rmd<lK+?jn1d&mctdc
zs75uhVT*auo|#G!R3`Q6W`vXW72tQ{F6<=V5E*zp0uh+%1llPsiZtyAQ$o#@Wtj$-
zEtMLleK(?u)NPQs>kHPn8SKsj#^{(li-yhHw>2RWOseSn6eKdIDs%UMre;I}Hs*rr
zU?n9oa_s1;bsD{WA5Jl(00Y#zb4MGd*ygqT!`afd+4|jwOyDq}Ps(xn^kd54Igq#B
zp>~LwNC%zO_SZQ;Ow6S*^_5zDWp1xOx3!FBOzBc|qT@}(%Ce;U#~9a25kD0ks>fk)
zEtJbw{Rj+B%aj#}Gt5e~K&S_EHKt}YsjX^_Jc)O~_pbfkxtdl7$7-#kFADikbi*q%
zBki^mJBIp~r-z!o*{}~CCr@D-Q>Y9dhO((;bewAOveJpsp@pw$-Nr#z-DIx>z7~EV
zI23{eaJM?rjDLYU{6H1g`wSp^@ID!tF<~jC`k?*MQSy`zK{<nkK%yfnIA5zzk?QId
zF(boqfXN((?QQBEu1-+i*XXoW099(tRD&;$YTh9i5FjK1JkB;@Qja`j+Y0Yul;AzE
z8QM`CF--_86fgIIPs^)dAqa{o+h)^0Ma9pi2$olu88qUBU`H*^;gMXr)r42F>0yj4
zn(cRsH{5ArJJ(KtVXlM(jI^I<N!i*9G4+J#_Q93Q<b(M=KVSjn<=7U^@$G1^KF(Eo
zDl1lOj!Rl1>?m2nWqtK~-{i<vm0>~|m!tIT64f`SZdQ=@FzXJpG>11%a8mRInnvBy
z6w<7Z%HM`0i<IvfGX6`@3S-b%>Q!zN1M<!b<ro=>_<CluBMxPRV0BlFT9iBpWZkt2
zQRj?x+vNdfBA~rsUTCj4`)k#K*lV6;1;LR-X30pn4^xjU5i4>U6q!(ZNpG-RwvK|a
zScB{ezuRHE9D`h@Rbiq3i&(xDm0WsjDBU9UGW(SDe*uCcQbRv@h?uPZhh8*33sDMO
zE6Ns|d=*5K)V0w-gue=fW3|X+bCh$HKZIaj$U=bC)pOxNMdbumJo!ygt#6<@R337t
z)u`)<&Se(~{b3=UucM7FK=}f7K@kQEa-pN8I9q~|SY9TAIw3s`C4x@b%nu7$OFI2X
zM4aq=qlu`sf2l*wQt^wzQn#K;@B1;&t?!YUFA_0m|ET-Ec<5a->iYk)_Z@)K@>gX7
zLzeX|<4fi}9nv(gb%jD(_g1Pu>_{1Qp`IcKZ|UZHs7ld5m338N!4zy6U%M05a@N$0
z&qTClZH+JnNLsCEXT5KjO$9RkHc4oJd6XxnKq;5<z7(3zvFo^EK}JS$WQiRy2zD|g
zlcA|%#xf~j(SY<7{Fr9YW~|7L1R7MFqI!c6+xUvWp5xOif;$9?<Y_rt9dWP(!f^Bd
z|3T3_1x>azqTS{UDcMqxqY~mEPUL`LIi;VA?7Gd<^e&;{$kS9euKCjrr9KR43;gO|
z5U#XMKmwlbT-&yN)g{<Lom){FY8^qyaE=K~(Z%C-fHK+rEP#FA$cU1@UDl58(xtlS
zaqMR=>yTF+v37vO^p2q!ww;*12!VD@EYg2izdl#PvxOpgCXtERZzm(#L%p6<Y(hfm
z*$4-+ZYK-rl=-`{?}`9>3!5pQoAU|#=?t`;Xs7*`^lfY-2A$6hS|XtN_m{EGx}jhq
zwt3?Pc)`BNb99-GdW*RVW^~rdH&jWaVlHbh1cQ1<o!;9!>a);Vs<dvf<PtanOurjG
z0>Y%e^pN>cCcamYt*Ad|bN1p@=Z_E91OtqwwQUZk&@GFVq}y15K{f~4Tew2PJ?0k=
z7eOZT#*)X9#UMyy`IZsHJK}!mpy#SpnYrK<?^re1B6}8TAB}|nlxUgt?nqcA`5kX5
z4*oQ?S*P9}KZCL62sOycir8W{#w>@+vgHi@TWFl2@fw791?e=y>fWk03Oo{DmYz~F
zKXklq&iCLEJcu-HGSH^tMK-4yeeoi7DJLldBDLq^;-HPTDi8ZZF<#}wLg~JnzHcLD
z@~x3?e!~z!$g*Xix8otPAu6Ec*vy~O7-rLQVj#{W8UzQ`b8Ke5Q{5i4z|B1$)Kr8P
z?V{uF4)W$kCMf)vS&deCJRhl^d~yzqZx*Kw-35~2B2#9p;;oYOk+(}FWx7A=LMOuk
z+oxrp=u_rw>Y`ohNl?g|yOKWE`m`0+?;lGevtk1E$dQe)5}Puvf=A&{E_vFoenz@s
zqip+X%L)eRnXKI2z--wo8Z>hX_NoWa2`ryaW~YXI?K|d%MT8HIOgJp+hxIOmh&pPE
z`K%s^ZSpIYtK~Mi^ZqPG6gpNr3OV(W6GjUV?&)84j1@3t`$2R#LH+yVma*Yo-x!Qd
z_k#CE_PK*l=y&^{Qa(Acs@7U`tx=j3MaaCjj0&bJHBh&B9r>tdF;81*(IX9NMJiaw
z79r1TiK8|?&BB-+j^oUrvK`bZA8rdwX$MOvNXxZN84|vgx79D1FBs_Ef@aAUm~n+O
zNUXt1%QY`!M|LUAyLW~EAo$1DL_**__Ju|R=LZ)diCa5Yg!OeP98ic*BKYq>XA?&%
zg`&AA%C)it4O5#%F^c5K`#2!WZN|N_bV+gz#Ihw+8lsftlOEJ5Zoo6xi}Ny71<p8r
z;{|w*8YC>c%d-9A83-7~(%NCVNH43a0$x#`dF^bVBYcL1pwJjm!*rntO_D)e;9j}Q
z>p*n|GK5`I0+PX|y9pWWL9Gnr7-W?Te8Xx_fuQ5f*&ql9GOvUR3#ftz4UD#`wxl(k
zYGxD5?Q^!GuADV2E4tcp;7e!9A~x0I;{4_E%b*#x!d@2uGmz3gH{eVV_?Uij7`rJF
zQjp%`EiqmFMT-@zt|r?`S1uaW@f7*XPA33%M>&n@4U8T`I^zh~2|9#=?+RS;y1>n7
ziW_iUK9kt&+4(Vo)Mx|1kZcnL3G67)g$=Hiv!UTlMH(D?%UqpZmbP3fnDLq)5h!EC
zR3yu&KvXchzU5|cS%_y=v)rUrMhitVigYwEeQ#z^5Z8F1G##_o`J=>sBLy+tXU3Ih
z&yeZ|IyI)2g}$d|@l}N4J8-O^_>=Q$`O#V1yr;36KD8Ui6A7WlP79EVRPA_q`BF8K
z9MQt^!vs_mQ>x6mf;Ee5D`8rrof;cKo+M#P_Oxw6&#RxpuK%wIFBX%bNRzPG<$UDB
zo?J1~*Q8Z5&@IomqE;@H!D3#GY&J-400b5a3qVM*DPj^5IlUqWuB9xtb4XMrjzC*z
zGdU5j95}VRI2;~+nF!Yfmr~T3Ouk+u#L6q{0!e!UF(Izm<F%{j;6vP(m*XmMSaSX;
z@I-v9JXMbJKx)r&cfP%OkGE{tlL0|R(pZ%md{TCR{^dNAgV81NW9w%7egmNd%eTCz
ztfH?x4*A)xJlBC?Y!Tkx3W@;vqO6{-&XiX@=S{Cw@wGCE$Z@uq&CeR-`aF-)%dohB
zjOu4V>Lt8}tdcz~`60tDoK?JT>VV1N8gRsXzH*pzEjKF{PTCWo3Hh4mvJeC<JiK{u
zB3`||=|InD3d~4z2WjYdtQ}GOGN8<6s;BmWS)K~B&m!d1Sv6v|8D<Re2u4&yhdtU#
zdGZ^$XKC@ubd3F^Odd2ju+m;=k5iTe+nkl9q%Ug}mjjc-JA+H{5IgYMC2-{)rQHgK
zZc{gtHhP(8e@nFMacax)tJ$1az|fbp`C%XzswfQ^f{+;0b9gK;Y^1`5nPlT4&RJFI
z&rXNnn!(Y|?LEB8#L@<aB7HMjy<1V2)Ktd6+1Cu3Z!~Wvw>bYf#Q-#Wdy&A7vYU`y
z$~LX@v<=loHd#8@SeC@oZO_>MKREuBieIH@E6Q{ey(=#6V^CF7bTW5F)eQG6`SDju
zWs%S+N~fH=isGJic=3?0I;(|cC~UpQ0pu~l%`@sUxQ$1)j`R*6N2uG#*iEah5qKx9
zQEqUD({e)H*?g*!z-F?zt*Q!I*c;M?Y4wiGCr+!jVJ~~K!U8XPInpF`O7E$aAFEz1
zah9XX`D^Yf;_qWnS5wMK<JpumhEG+Vb};dseR*2)-gtX!B{X`Etp5|=%zK>O+@5v$
zUgmZC*EG?QP0xEhxYF8{KBEo>Nxa}GFU05xPwUb|j6eU`==PNm-_&}`E?oFV>In@X
zJufVEEekkB67>7vnDNXDHOhirF67UTwR`?1$(K^m5-v=ad6q(BQ8A%hQ^WENaxs(E
z4!B`%)rP&lA*q`Q6j;uToaL#ZTI9y6fYq&qvl(xg)8t}_0_DK3S^d5=Hcgw4s|!3K
z^_rguX;po<_-i?k;RYM@<L1`g{Elxh1bpewHu-^NtO6x>4m`!|^B@Qb%T#6MW4Lds
z`%=6&{--k=NsgmlXoCKORq<Z@!L^)s^$yfE_}U9zU7JR(`Vfb$wLdF0t(N{`gc+zg
z77nUu&xasJLq&-WBN>U{ycK95*q+96$X-NN^mD7XQQ{j}G4)(*WM_z0--M*S#?qUi
zs>pF9V#<A3v6pk(KM~~=ybRMKRn{pl=N%0EZ8mhKOoUN#00UfnIJ<I~$)iBTk6hwG
zAz1=zrt;t^a7^m2d!0n4OF;qH5ZyGrS>ZDB=vSU|0+;dT<~}un1o?Q;m(Q(x&#Mxj
zT29t$y;C{}TDG@h%X|A}8+!%r@or(<_L{#V@2^Oi{benxeLUJ~hl}kax1fQ*8h7Om
z65$Uzz@MqehWKeiD-Q>qEYbBW`Q8PFC<+WaSRw)_P(7#~Ec##x3#`VH_;T*JT)j!M
zX|Y^Q3tP0%L)nUo$jzdc<Z+|GPv&4ERTuVULuiLeTr{@Vl?rZ6h{WgaTg^ywO3$Qf
zz`J_w1V(9+xoEp~F}YYw0&}(Kfgt;DSmh<^NvT`}@eKlhZVKY9XfIzP1~!S_Lojge
zXK|F0|G&bgXL}}nS6Gv?6qZ|%$zFe)&4w^cjMLlt`{Fs}^ug1@NBFC$v4J5bj>6(L
zYtt);Iq||t5}8^esmZk0LLSJFUsh-VOaK;-o4les#XHHOW>He;a{B6h7FdCqBPyw#
z+23C~vqWU(6u_(?9Gpu+OL}WjS5URR613S~;?O$mB~xpjf&kWak+eI<e8^YHsgm%z
z_SsXWbj+^Nx}?C(k19Gh$L_OfZFXO7mW~ZAvlht}M_(-f_{$@h6JFOdYl;-(6Dn;4
z4%b2uD9}Y<a(}X|z-Iw-Io7RJmCD?|NSAjZ6`4Wm5WBsk*2y}!_9Xk{Y*HQNIQ!<V
zo9vrwPq25kWJ2c2+KFA&ifr2J?I8gilAr)S4&(^PDbgra5>-5!kW`ma4s<Z5?aCnU
zN?Sj~qMEP0`|9ZS2_Cz;h*+cELaD)L%DGB~DId6yE0an997igaNu^>A4v<Mkp9fYp
zv{XS(!XxBzRT^vMhCEn-84#7!4o#j?J0zQQ{X@>1dgh3e+U=#&>Eo=92Q_>bqrPn-
zp+%6))O+`|VroIDEuBrHbx%PS{HZ!iW_SBMCQy;9)X7ok+8%T*SZkP#s)&h<sX$fN
z=Dqv$+SBY$;|JKg%c>WJ*C5%s-K09oo7jFQ208C~XI}eL@DI4pbz+<5{-ysw%{^pA
z;W4nx#yZ+}l>K(!5%zA47|w%W$Nr?!J~{%fWiud#+0enOm*7bR1kh^^z@~aQnA@^Z
zwbbyMj%?!Vgle`i&+C@D$9m)M{ZZ7wKx;njZVGtu<pAx$;ODL^>PxqYwYIxAu<v!i
z1FaHefuKfhx2glKv3Ioa!_6(+)g`VXicD$e@ROp(ds=eB#(#)>LMGj^X=5nbQbw$8
z+r@r2Uan=bn(9h(!>?b7pUC7tD@}2=3`!ko$|3ehVoc(ktb_lt|69vD#J4A>hwjO6
zusY-@Ot|4HeEn7MC`^fkakN*1E3e8cuLf(b0uGp|>|#*)MN^OV6#Ur#s^#ObW#h8r
zU(N)o*4%}yyxaKz21d;Np)S+OZda`T;jL)rZg{>sD{r}(J<C1{l(KEFW!FRG*ToM;
zbMBA&cLz5&{pE9w{_gM=Ib>P-{{yb5ty)U5`i(p_GvGBxu8{FF`3jl*Q1S*n^cN?C
z^Bme3T(1cKSY7>*DnlZP$4E=I<@jSgt{J_aUe^o{)<0+KQqnOZQifENjUJ!ewR`)H
zu3eLlAK`cl2RaKo<vr9p#cC)S+rX_3b93GC)!e+2vF+TZFdNsVHGN+UA6!&aLTqW>
zvwyT_Zx2<0L=t!K)71(|QNfH}*F^j5BI4o+g#_Db)-98u<E#eILSv8^5aZQ4^H_%p
z53#mJ8E%gTFj7+&IfZ8O86#D*c0L!&)h6TF5UwPwvT5pXhi{R=fcmKq1GJxs{tMJ)
zePhrxCzp4_8?<iHx=BIBS!3pr{J?qn#<8sC`E`NyS!0f*s=(?wqJ$dW-L}8OEO*x>
zyEuim8NHr~wxPm8xF>pggG5Ei*vBGomCJ=|QFI)rI3|%OxGb(jAvxywWyt_#A_@8_
z-g>xTeeAYp#o@WeSF#eCy$cR!xg8#Tl&3fZ^c@9g{%bI{i5kw-#T@@Db6-gjS8FfO
zYh?OtWtb^Kpv2D$$9j6)RgOA-^fz~$bhX(jmuG9l>M&!3Rz(;I#}!YgEYl56gQmUW
zrf{?8k8a2g&Hs3W!#VOXG(UR-v@{pd#L!uBa=8`9&>@;~m#Sg6Mx`evz)w!dzBQ_>
z5{>dX6N2kUmGz@R+zDLKXhG%9pvs+{eAOu)>^x8Yd+B|rYLo~&LX_XPcSfbWU#<GF
z41Tpt@qTqmVa9v@+MIdZYy5RNhiqbYE(l6-AP@*gN>#**H>`Boj}0wTiCr0Wjn&S>
zx3iG@Q(@qVtqJFMyQZDty5i1f|Mw>|yEH3KrZsBSMrE>mb+SR1%mR$5iWDtfZ|M4X
z@5*#r%yhj)4Msv{sYI*;3VuAGjh4n_()37onluxy(NNCAf9V<SCDSqbeYR9CMS<`v
zeRt}L1(2KajB<qBezpJexh2!*7xR>_?eFT@_47+lQ_jM?&j+CdLhgDsLBKGkI?-l;
zlV;s&!*Da@2(zxedd2)eGTZaC2%XhoPF>=+`M<2Bp87mmxWG;T<*6RfLYt{=F3OKG
zZz{}D<kQNpuvR(vQl=5zfi|-_80#_{7BD-Vx>%jlY0i1+pUS<dgSL76v*&q90~%^L
zn2$b_rlso4`SPgXxpIdjx7kxs!Pk4PMDCayERQnh>!@kkGx_K~i^m{E%8K+>Ymr`t
zlzI%gUY$iQj1^ktTCEju@R2bw`kaL}Q`hXyk1{WI<tXy$zVD?i;<i@c2<!xbJk|5$
z7_+Jv#hj;wJp8BA2jO&eYCw`QGxhMV%(Bx&GlRI|^6vyJdw74#`=4gfxZ645={OnX
zz+HG-P;quI+f*1~d*41&nO_Cx6^N}IJzab3@m#8n5i2z54qmCBrZ3@wm5GbWT2L#q
zitSXPFqg!ZClqj6I3RrMdzwUB*4U^mlhEF6gQMsOo&6}4DI;W>;<_Ehis&Y))kM#p
zR&bQ^mA;VvmBQpm&47s`owhnwlq5om5_pe!$V6T~FH*FU`<RQQ24R;bM8z+|Vwc57
zB`hVh!qAm539*%EbY*NpOyzRt;`+jxihZH1gy=hAg0M93DT^CgoR<D{TTXkswOjJj
zZg6mCY+5h9370D?ma!F?5OLsH?B^umdzNO8YQ=K!0GMk#qBt>S_#IXQry!pysx5QS
z)6`Z?+F&=ht`Jsk)F2?eTrRVy!mMizH5z0>F=p$*<}DcPmTUg4(hnDoZ~?9AILII?
zBF&2T))BA(9q69gG6y3~MdD2UVNWaeunS|k!#7|D`7L_)YG}7HUr>YXl7f813x0Zg
z-VLqRyZJ5H!GzvDPV?`fzF_*zjGOe#ig&LM6e4#B^8}Tfy5(hJch-X#ijkz#)ATwc
zi4yZ5t2^=KmegF-6U;&S>7t~P!%wTgVH9<o04I!7QNv)>)59f6MW^YS^mpOQOsZZ1
zs!37ap+s9Bhno`4s-(zK?p7-I+Y(d4wzU?ctVxj4Xf&oqiNSE~DR6%$SAZ0-f}Bw$
z%_MR+Qn_17^zJPQ$JqB@c_H^*!szt5kWQbQYg7o**JaKU|6ygY{vi^d$cQJ{))V`l
z4$bQ`AZ-RWogw_HC|J}Sg6M!&5n)ag{8^6w#R!$`=TJC{$FW0wKwm=ayh@;wouaNk
z;=KUEVUjYA!m_143vT^Dm8iHZ`r|ZdoGH<u7ji{91zb<J5)L(w=X-~2`-3*u%VPC<
z!~I35wZokRY0P8b=IhtrlKSm`QGnn7`DZ?<-*>Mk`7PUq`wnMivvJn%EI8JU93Hj`
z%i~gG#C7ys2ff>5r1#`H>gZzkSv^J*y}Q&czB@y%Ar=b&fxruXD#+jHS3#%3P++kb
z3l#L}KiM|9(6kT^l^a6C_Fm+@y3|Q6AjwhDC^?yol1KY|(%qZn!6?hiFtM1jawX>P
z!!IRm)QozLN~nor8q=v-{30wm7>%XKzIxVf{KuXCDBb<fNE}<YAV9x>5l46!U|(lj
z2Yc9ep_ZJ)iH(45v@!Ypw9+|NGUYHuJ?-~<e{iyu5O_3mYa{DQN;tYOo$?!=HB2qj
z<#=qCT#r`EdW&r~Xh<AD#$q!$G63Mnm~0jxGg5o;o4*^z<A=Z7j88wT&*d;WdF0W@
z<m8W!Mv_yeP6lGC<6F+`8Aq-wU)@mr%KP=Y;?;HK(2<ox$6{hHth&s&{1|v_8&(wt
zVB~pQyh)g}=J8zCl9a{co3ohwnAbDPV~p}hn<ZDirkyW<ohvwZqeE)m{j`;D&@C8D
zd@{D&e{YnW&vH8I5aPSz?D4z&RT*-268k<IiI{N%jvIWihs;<<D5#NJwxkc~3zmTF
z{Nhe(MGPXQBGpWZ>R<OLvb*y3d&kFjD{~|Vsvco)&z=2+og}K}u9dLZ;{`iDi8}5C
znieR*ZhOHSQ<ae5VDDrjJNDHj_KJ4=B;IkAX5U;^1?jJ5M@x3HchaaI{+`0WRa%`2
zl6p=QNKPRKiYESq=!foJ68>>7JW7HT1bx@s$~(1;edRSWeBed{W9gKi;3uXDu8mrW
z3m=e4b(C{}{tTW-IV)MxtI`!Y9L6FI-=vOH!y}a%EknI|?kh;yYG$k~TTY^6@EQCJ
zHPlnyx>C)p^ZMK>dB!h<zDR$}?J<24L+s;gyrh1!nX$u<da;{$xe>E#fm<$<iTme*
zU5(k{CXLsG{&f9)hw%zaRIVuyOQbobs)n&5J3~vCvVbe|HjVkM&1GN4Z!@JTRhH#B
z2!5~k=VCtKa1{b}J|D0bx&X(o#bd64j=)ZK^_r7r;5%4yRtNs(mUdAtQFccJsW26w
z(t_jgY<N(P@j_6VS0FdynP3iQF*{OZF{z3~<l<eo&SuFej^vuh>AJ$gd?;Vl>&&->
z7t>tj4)e|I=qR7*3aesN$^vW>=FQ(<Tss^qXJ;;x<jIr)m3wmxieg#lerc$(*wA_z
zGjoYtja+JX)rOS)&3Y4Qjef)WJG9hQW0Pv++{DY7i`<J^csmbqsQGZ((ef>nvr}&q
z?jC7AVtxJexOT>C(l4T#`LQ{YV!(PfNl#~D3>$1{j=oBi@bH0!A=Kv?E%YSqzH7i)
zI)upP6lB8Yf@SD12dqXS+n4R&z9H-vDDfi^IA?>q7*;{A(85aiX(tcpsRMD>wJO5Y
zNL*oquPm^FzHpLQ1IfboL~7tBscXJm4=KZ@rfT6i+I6|?<M)pPt8aBHe(Zw(*d@Ew
zy&Bjmaj-+s#Ax0%wySxP5SMzdtrTj%5~Ivj-=gWX-jKGCHQ|!KI26ubl5naer7dMm
zyiy_MQ~3&+GJc#Ewl*rVlO8>{HunYZ5G}lx9#)Qxgh!tqe|dmJ_SgXVfAHlmyb3e<
zWmMp{fb&b`VdabfPGCqxiNXdytt#KQ;7a*WmRyWKD4gUwlyFgRr>ZgV8dkOFWv7`X
zw#`bG>s1X^5b5C~D?`-xq#oO!th+N_P*1Q<z~u~cFl=t$2bzMV_)<{)|HiMJFAz)E
z`gt9Fe)PWPo9hh9W4;W^Uvxn@*5Wc1=c}0ob1EK^-)4}0At5?{wST`A(UU>j$xnGD
zop$$2JlYRszImn7?^yO)WR_Z~<?E>(buW|2YM@ipokeeI<wUK0N0YMTp`Sgjh2idn
zp3u*7mK5fLSi;bE{NRt`;E&wUcS08AaM`@I;a)<xx7Ox8bA^%RSD7^DOByaJ3+{3k
znpO>stuna_Ts>L+GP#IVIo!VS96Vh|%PJ^qtf|9&yC6$PPlqpCMPDW7%hh!CTmAqZ
z^*`iH>Q8b)|D!Y}gG?+p8D^|wZ)YFc2HA3fTpq+z{J~W|<NgLbd27|au_~wPe)}Ez
zZbZcyQPzwoff414m)s=o%d1wbiM6}58ma+VU<v&!Gxw>%X^4``St3|ubQBC06&(Q+
zvE)&5r@^qvoIZ4&zk%?ooLIhoCja_Sy7z0v-2gq>ymjQz#63N+KgU)w<D2{;Z7TfQ
z;x7Whshq+3ftqk!b8Hne<P1y}8RRJ%n&k0vCQ@`U275^@Z2|S|-qq}IhBjG@l4W2%
zt?{vD?y80fS*<_LM+#LENu~iyD%;^Bb14dB*bWkE{i<@-(5`A2$ZCEPk}5T*lr$@8
zyK-JV^Y+7_Z^X#zk8lJwwt^AYkRvhB;Wut)Hx#v}$g~Q4;OSbCnzkXXh7ocGE)6Z?
zaaM;*1^txiKmhRbvKP>cr){<Lw`I$3j6vXlL@gHui2x9xkq8d}e*x#<M-~=#1MU6+
zT5GW(IP_2U=`|6JTwvg8UXjztc?6v~QONPsoN!Hj{8z(J6WH=(w?HJtG-O-djO^*-
zXTMnpEp!9!8Z{t!zZDf%yIsYVMO$nB0-k-pRU(mb1RRN6vUP$qx0&8tIaUemuMMMJ
z7N#lGNwai+2fR>Pv>AztwGXW=j7_gL=3geW9%5x;R(4#=VdrE(FV+zI+g&0!I36$#
zI%lVp$416fq~!5N#Fo$gyLAzJ=dPWM&1b*xDfkOG2a|AW#iod)z47roXFgc90Y~wX
z(u<A<#GN{c;+&T*t#-jqam`i%_k1V^Ul2|YFA>JIZU2L-R$jT-{Vd#fc2>*}7wU{!
zvrCh&R%kORi4lwF#g(n=p})}#DeVyq_(0W~8r%3miC)md&LH1g+d10?Ph)E!6zqr!
zT7hcW^B^T(DwETA2>L?d84j46mv)noFVNi(M>Ey^ZE?vo{s(aJG3a6paR~)~bH@+E
zUlWmg;}IL|ucK71u=1H5Q#xk1_=kI{X7=}Uj7MPPfbdNA#j(OcBc}z$te=3of|^jz
zL>bIoz~kX9KB^@hqAklnzAcQdq^~rBMkQTYn+gAj*n9FvuQDn-hgqE3s#IWnpu&<K
z69<_-N>UZ7(oN->HD382GAhSKU19NQx-Xx7Y;|ABY$M@}$Nkq`kK@<~A|<9wD&g7V
zt`6()v8e;d^>`QNyyV(tL*{+keLqAn^2Lp;y>z<toW#5D1M159S_NHU1dXfcs}8ZA
zBUaS2xM)pt=ZX=!CbCyhBjrw`Rl8mo4s&I)FVWmq7zhWm7JXd~c9YzcKm5-)QO0jT
zrF1m6r=_$9a)&DvTu`#an?#vs9e8QE#XIFq$b?5pQo?Jz+YeSm!m3M;?$bQD`i`yU
zz5SopaM;89mg6Y<ZO3XZdo?HW2z&SPI#_o5O$O~xwZLKWz_CT2?ykv&m#uu7R7ZK*
zT&=W~Cc8iu{ID`~sbdmz$dY5+FB^VHvKl|(J>8+NH+gk<t1m%>CeEWZ8nft!y@o<9
zZrW~Q(i2;@W9U3biL+$%(&_wnjPJpp|34w0uVVoG+`P033QfUHga!Re*Bgfh`i(R7
zOqt*vELF*2sP&C){Qpyn1cj`OKS$SAPUQT0KyZa{bLH{nUlS8g$0IiQevVQq;AMb|
zGw(*a<kZEV^xxo(Kjap*kC5spo7pFKl?YCMXomF^`Ko8sOpvi<7Z#%o)spIX;LJzr
z`0VCu7Pcjjo9v)!uuO&;`zgu61@LlXCYOAA#fLwGUq@nfPEC%$O8jiuAWx&zx=G@d
zA2-Jr)uyXF3afuuzN}EGQWfS$f~&$GQ0ho~+v)A!=Itl^_5l6B^|vz)ltn<E`svfA
zvC{+1LL^*zX>fX+MrWfL&^g8S=S=+z4tr!IC(sl{WTGBTeqcpy-X_juc;<e)2&ZOM
zI4Wk<q~cS#eHP5>e`ZdAZfq3a$i6-b0S6*C1DlZtfLDast<#oHpQRYd*H1)vCq>Q+
zP`HgUlsp7(`xY{X?=RKTnCE0VIeRB}?(8tfOy0YT<8kHpaq}j4+P<z~E+;e=kLLK(
zeb*9YW_5E9HN2~BuiuK>3c^IkOfL$(TQEvQ<~8Rzds`3eJ<ztln^R~Xnh@d%ncx^I
z<NDi085o91&?JCO{+ObmiSvQ^*PKk=KbUZ8#bT`xn2^)!>F<u3H>01Ksd}3p#mK2N
zMQM(!1(E(ZrOurD+77LT%<=P4wiLDvRS&uEsexhR_s)^LjQ>?(lIkd7EW(nsq`JWR
ziPZ3te1>UrmS=l%pELWM;d?R(^ev6LRCqL#wRY?%%VN`CC056uL+a5~GGiH~oF4aX
zs;IJw6SW{9-|cE4D%KX#dnI15ok?ETKzCvHxLe*z-jzN&iYslfYANeLzyzMQ@``|i
zeSfR;8As2Gc07aE`88LS`|7GZl04d<^GD<awpg!}lVm2XS(&x>Z{swmNmuBCT>F{(
zKhGgFy!z}bm6VJ@3y&9hs|pJaeAKSFetd|WvpaIh@v=FF3&;v`#K{#mMt)?7W00o@
z>?5$8h}Hc7+lZ4{(&<`6bI#Jn?T)**iBplNM-y6RDbB-PbKwdo%TX9*D0FQN$oA{M
z3ckdR*1+MV-}WV3>(9B`&;R#xPw=R^N9P#bJ_b$WPJ{G_3qSD_X26FR<p0E-7v4>}
z)#te$&x_h?SUu_6#`W#(jqAUiWLUj7Y99Nnt?LR-7H?%0wwOBeoaWA!LRP#LC%e*R
zJL~DVR?fz;eQSMe<J{k!`rA0@riE8_&0jbHy+vGI2B=}*gb0>~0xBF`kyIOK3`~^S
zg`lr}XflK;0TosPPFLV56crYDbnk!gmAAs)Nwf+vK)wlU2c=w@$)fr6lFrJeASv1O
z(F=RtWZoer67OX0-CsWwR>$UygWx>y6Lq9zooKw^7;o0DS-fdS*IsS|LknYcUB7t|
zZ)$aKK<SOHtN{4Ad(fFI)#KVH{3o@l$63&sd*|Q*SzXc_pzclWzx+*@@DZ73gllfs
z8aPx0UB-g?*%QG4FhI}0FB#D!8AFwAAX$bVl(7s&C0Z$GMi^{5B;{%AtwohEP!`k(
z6Ba&k@Zo1ScTooneB876QTEW~8?v9_uPtXQcM912__H@V&iA|lyQgIT^G&Qiv|O=7
z2?xi=tLM<xARdU21iKNIJ0~MhBpTpZPXF@2^?C9^8t7l{%&M=1fsYd(gK+Ro;v2BM
z2G$1$#}VV8KI+;z_8IoMaWH!fPxPk-Q2mMVuXSMxzzKvYbg!%9VF6e1m}mb-SD1{^
zmH$7(;I9V2KJh=TdTyoB6kV|^fBx`aTUz6VC6D{BgM;Id<Dh<Lfpz^yc%umq+8t)&
zUT{~|Q#o4KHYwo(2nXA;VZ7_BuRl!$f8Zkh0snsH<KknS+HKd?vF~48XK>`2Ot?IT
z``1~VUOIu^%UN-^E*l13Y+O6f<~8lB?$%rb;oz)})Ev9!`Qz~=+I+Y4EcSh%&F;(L
zs~IUbfg+g!5%}9@*XbYM;w9TfS|BUe?z6?d&zllXK0KUI;-{_smj~xPsq>d76n1PP
zXI`2UUtu=KSIoII7uB*QQpvG$B9&WOP$?(>2$`y3fzk;hil1Tsm^50w!>oS*2d~X@
ztUZ4V$uGLLoHI<r(uO(9dt9TvT-m=zT;6K0*IVs+_m%@m6d((fD4;QlcX)IXS3V5_
zpF+3UKW@GunwEcAu;|1hLGr%Tv+K7oezGM`kJ1#hYsBR#XJ*C0VRgIm<3;7!g<_$&
zusmCoAHS;(29KL{raVR5t|`z8N7KUXO>)9so`(e*3Uu-$kws%J<9f6v1(e7)%5`EB
z!*=C$1>aCnTVm50bGHP_g6&o_1EXJWOD%GaLMKBP!)5*N$HB2vSoiP}&)JYIS(tXG
zAON0T`>^qak}E+5-=7jYM+*&~8hpC4JaSP|JwfM)abxgYSvy{dtPCIVeP1{R*$YS4
zjzQBX(;)q9PA$3WMXGGX&g%=HO`Yrzzv^uL&~P@+m>(&7?V<_wD4$4(DtGvSD<|Yy
zes;M~gz}YkLw+8ACGJ3+iliis`*lEx^B3MG3`Oi4gQnr8@hq}l7W9nbU`LHnjPCN*
zu8Yt`AXYU&8pm{e=Yz%|JAYI-ZPm1Y5o6C>2@l6=GzSX7c&*!lvBBNk=NWT*Z7JVf
z;^vsFE5p0~sY{n|vfp;Hy_EJr;h1oOZxqTH@D@v4I8B*8ab`=m=EPt<MKgL+zH+}J
zdY5rL*qEGcOX3K=ooJt{t8Ej*8UJUP;`;oCi?zZrp}5%38$T(-@IRd)!(>`wXV%rl
z&djvPFypC%8zSABnVLxVhQZW7UFE5*(y|{9&mh2EFngJ$z+2^dF$VFLP4K@Tm<TNz
z+*7G4!goUU^vr~1&Eg&7jDxaeFWTsBf7Y=%hp*DFz658e7nz$-3aBLKz9LQ6)|jvW
zgtOT1zQ7v|c#+*<_cfIm@|4u2=~4R%e9wL#n<Ay=;VArdi~Y-q-{G;>bB=LF;b9X)
zg!W%MDZIIPfm3<)@jA`B*+=^HaKqz3Zd({Gh~uGSyzVEj#iy^bxRcRI9eE_~zjVFq
zXLx{Zf9ZCRofUsBqW%PR0%PG$ku@EU%P$J$;<o8$^PYSYV+gHO&wqJYZm{GUjh0+P
z=k-C~0_wWx2f}#NS_Y?=PNMg6ME9o`!NA9nk3m)CE_xGr)!ms5E4F#^3GZwB`1j<$
zJm398mSiX1eEfzzHV(eIse03xM&t?ChU<}e@3e3m=mhnqhZcoY%<70@`&5dplDAWN
zHpE$&aKlOX`jg-p*hR1?nl?lq`g?~N^cl;+)<{|WO};!T=lHL>?mWgD`daj=1k%V;
zoVNZWJfzs^V&prVmnv|}_AX&8{#QmON)2^h4NQ_9Hu%O_yjh!^&SN&0nmUmKzGiq-
zy-09bJ`Iws8p)W%mH+!A^T#r%tdd##oA1Exbgi};h*b^0VNeWF-^-w}>}5j0ju@g@
zUFGc^HH|st=K1rcYf=UwP3Q5*kyL1(sd5W|AFUkXEo>7B+cmjbp*EL0Bg-(Bcf8i$
zRG0a0{6jT?rIN3|o{qDowkUX5Q2%d?xGg-!jpNOi1xK!E)TblWYuIb}tFqH-+NJ7C
zrK(;ixnd;pN_Q-uPJm6@2X}D)6gPif`(`b)P%_vztu1U_d_$T-neI-D=W-M;;;TYe
zbsT(~f#YbCxe5hdJdV;6;nd<vcHqQf_k8xyHOIBWF_HL~SeU{2!cyw%K9+DOE3-x+
z@pUnbdtn-+y{?6(ro;GC+$p{DNSn=#a1xGiFFEu<j3X86P2Boq;}B<-z!sUDeiN~R
zx#BPAujx2c;lWb?>^zXFb3&YEnJ)|WggMg{26?4+OM(Jx0t?%kqm^7@yFO?ad$+;A
zK$)CMIxNu$pU0jQb5Pb@h_U^P1d{@N4~%9=CC5l3^`(hq?;~V0cNTZ^-|(89Bv*cU
z8J$!|xj3+QaJr4(q2JNFJ*i|AYRi1GI7UJ@Rj!kl6uDW{Ila3mK}!TO<iafBSP%T}
z%$^*zJR~$^dDQ%|!h1kVY$tR=m;oinSVMH}(S}h6YiWekkTeG=<C_(|``!wn=L`x<
ze!AXyCl8fEZ$<~EHWF}~ZoCivmMM_4>4Yu+!Fcz{Y^_kjIh&O9b~%Z902d^4vWNqP
z+MM&`eILdDLj)*?s)j8cUs=2Qn?qqEu&0Z$SyG6XrIyP$`;$U03>0k?AA%pEp@9d!
z$nWmu(dA;bT9Jv;u3$7sq?x9?FW8%3@O-%}tCfwD0yBa$0!gH|)UR*zUTPN^_L(wT
zLdXkr*;KhoBV>^7XTg){X*4;Ioh=7wY4l{c<_d`sEvLLeX+npfCsQbsaVkxjd+7gQ
z!?FL+CzV=v@-kDjD;Sd`AEM23>v<)KM5UfKnG{G4B_Gp7d+}Q@HL$Dm4^I7iL#YE&
z&qKW_<s|a@74`dm6+!nWOW4(E^th%o`iEgE5U~UWim$MMF{62L1m@#qoszakbWBUs
z^9Y{;%6HgVR5-oe(OFKQ>A4)Po;Ho5HbpHD3H}mOc(m?2I=9llf`e&=0E(OVx9h++
z9i<l@CkG|{lb(*<)|f-TIUB_vq>x^E-TD*6zw?MbR@;Z98?JOhBRWW6&L9!VG-MDm
z70cWFKDeI3Ii7*7I0sAU#M7m`8q4k-)D;i1f8`K3YI@Bv*x=DxnvZ4wBjiRQmeiRR
zh_Mn<{I|77j9wyb2%Q2SSE1P(*c@55jp+lWGKiQcQ(bP59@d#|Ne>>uu{N<dGN?a&
zJ}wl;6mIss?gdLUne{I-`bkU))hCwnDEN)R@b}16@=1Z<BsmrN9v-|APm%JnPZOb(
z;ADtMgpz|(Ai^s0DpY8y#5gRep{P}@Pil@Kfk!`9QleCK*8^XDyq(TV4HS13yhjO(
zCvyo4Tol#f4_smb2WPBF+0?)-=jZ8lVn6a)Z<}ugI&baUj>GMdQtGd<uF<YsOgo=;
z<#pg9?%>lsUdZ1Es|9L(Fn`G7O+SAg`TGBKujdQ`<_iTvVLrqeuP3sqOyAfvdlv>@
zczZc8Ljm{KDND;@%1V_w1Y9u#NI|9=8Z6uhrS2}oMB8*qe+67GyKvI91AOn$_-ZuT
zSY!ln_t%A%@oKhjBfs7SB9M)Hw{sr*0R(EP3l|bl8==+$;D_ZvprZ9zDW;>68nYKw
zZM1Z^?eH8U^QH|cv@L&*VbPYZ*TY`fJ>yLukFroUw;`QH*QYtrCT0*Ph*&q&fg>$z
zjF*#-9IhGn%!52%_6Rv%-nf;B>lmsdmhTD^ZF1_<=qy9J8)ai1g~mG%T6SUgq_K)~
zak$)KRvM70L7LGDDh(|+`^hr*wAt00gCx73C8tr*iX7m;RQ1OYXB6m{!2`lB&Y){j
zU66v<tg)3dCE{|r)=F+ta+34ty6**h-`N`zj7%Y>SfsL%I9zK6sKJy7*a@Zv;vE`7
zG5{bg!nMGF1VOt)Y)CM%1!*NhjWvVIv5`b7OtFxeU~II%+bj5<u1n72D9KG)YdKvk
zDQ9YI#AXGQ>Y{6ATnr14!TSNErL`W(64pJ7iaGve)|(^fwRR2lO$m_{;WE9i7qgyj
zO@(7`ST~AbBSh5ZFUU>>`D1&s5)pw=CRh2Zon+!86V9_1$zJixJCI@eHq0+L`6npu
zffj-JUHi<SuCC6m|JrA63~{v&Cw6vsb^q2ryn&jJ4?(sp%jtB;u!v3>Wo>89vKI3D
zOsbQ)wf-WOQaY9q6O%DUDaBr_-<nC)GRByUF~(4aId&b5NsU3Lp>J<{F{v139EbL^
zv0<z@XZN|=<JSe!Nr#O1PgGjiA+4By<J<Bz?U*<>5n;Z0t#E(;zB4Yp;gS2-k}m@+
z-(>cgkn^@HwK66;!;QhpnkMsPk%>aIjcjS+U{@t85#e7AT75N+lo(}Wv|zs+d=O#E
zQbdPuB^JBV*MZnNg`0qBWqZ~p;r+F9qLUT_`@P+q+L+oAgAlke2x(IXy*ZY&2gwrF
zf9yv`_tQ_@vUcp=&~t3hy3AXjKDYc$ze`at*E)xP?d<I8eCwuravkb~4(>0_^sHk$
zynDEZ&f#RK<m)YM!*9E~JG*}s2~zA2-8D<6dV8T87d|xo4PxeEtWUNuEp4I9hh01q
z@kSoqUz&VjrMT@G?-~r&)S22DgUE7Y5VFR}5jOt`YYpK5LF}OPWgOyI@css*q#F+9
z#EYJ%W9YRK(+AoH-*&Zjw*Kn&SQ}E2bi(X0Z{kzbtl1T&@D;l*h&_`!+oQrr7<RUj
zhDYz_6J)##h$G^@ne9Rx|7PvXpO07&R7gPy3;$vFd9i0;GeNXlEb#Yg1E{ax#tP$X
zw;6`IxH9WxSNVakHG_U%tdIC0Zr*WAok6&q&ib}>-mh&1#v_OCv~vI6Z;#R8{o8*}
z%eM!;=gR!n6=by^3V`(%!dr#y7Wq}JKPY%0VmG0XbxpH9krfcvt=18Y9*9^^Sul7C
zr&(P0*FOj!NV-rJS69cMR}g1+S=OSF(PhYJBzo-whwT*S@43Dqe$&XUx`ufCl+uTf
z3#doliNZyh+){hxJ6qz~N`cb0I5Huw%4)P$#ZA*$-Ej}&+}3Fe;)>!L8hD$&v>u=8
zwCb%#5Z5N`Svs2H0l~R8E#gvElFrI1^($pr<t~CX$r51=u#pf-#M9Fr;lJV|e@m%3
z0ss(lYF%<_1x16Om8#Ox=ybSMsk4H1|6>Az{FvZpk*-$mXRZNZM_VzhMQ_`K&(zp*
z702>rjiX-Y)H!PkHq{h#T@2R-KM>&Z8)%>azxLt(AJlyK0|ONqe%Sh23Z6dH-*N}S
z0sA-Pf5cd%u#DQP`;LzgL(Ng7fxOyYd=wtX8?iPs-8gzX-P6qS^{GC?MPbr}f3+O2
z4{0D>VzpQI;2CPo5vYN5iB11CjzR0_joz;94ZcOAPxTot%9o&rA}ob$*cTz*3x$CL
zHte&6Fy7ty=w<zJpW}{?DpJJS-lSN9=?vnrqtDXCQ3Y_!Ea49zI>sPk-_NHJ*ymWp
zZV>6E|NoPNC&A)%-CS|NelJg1k@^?!1{=bl^@3-}de_<)>g(wJV^QGskY_Oxe^3DM
z>u(Z&0}RF6R|5X}-;$CNcXV!H47&a36}mslyis$&KBT1{x0<!?;bj?mSiX$4OMTdw
zb$a)*)_r7=*Y=iuU*|bM<LEP7l!8Y%EMZJQ@ugf*{|{{78(MaYst+IiiWF7%eP4Bo
zgYbV#4%mmZB;zVDYOn62XCk*8_Vi9(hjMz?taabB6SWSY*=H$QT2|~s8i)9&8rmRk
zKxphhx@eXnq>n~}EK6Aop=XwRIa@2EK(U+^kso+NHDs}t3T~=)v_@;EMw(e-ZPq!-
za^gQxn!c#uEumP#VgZG4Q}XPJ>pwtr*N+&2fu$^AF$)1&A`u`&w)KE;60tUsk61IB
zILm_RGDMD}b6_dt6hi0{VmqxP10i6_=>1}5?s4Yw%q0r}dVjh`c9Vz5tf1%L^561h
zxD0*GLq380whCnh3bP_!P*V|kq2cd*Z5P@}oIx3e4X=Xz`dc^bGZGyYBz|Udukd%7
ziXoP=7{W#lW`fx$h+LXK`=5$sGA+*f%sRb$Ia@0iS<sx0vzVHxnTmPW4+wAzi3uox
za8{r%z!HcJE!4>EjT_u=IFV<V$2{g~9`iJhdEiwV@(TEwe#X!AGk&(4Wf@{I3ng1X
zh9LI1?uD=kbPleBzb}N*vZT}k&gP8lfUe;5j;*4)PV(6;@t*C#N8vyq0Z5TqNElHS
zxric+7-9(`QBpU*dM&%K%y-vt4#2<e;6WUA{6c5%>kIvgg!pS+e0yM-zvKbgm##3y
zPP;Id`Eag`W52$`sxiVoS4}$_r(2QxV*0WtA$kn{EAT5<3-b*GpY0H+cY#~iuW^Ly
zfyYQ6fy2Hq^Zksbp93lP%19z-IhPfe>3=aP{C!3hFrCGzi!wX)=hMZa+AC90PEV(s
z442HiAZ?kc=yy*^cLND{*_HEA#0#kA@0sKIn<OwB{8fFjfpKYUGcM@cUo-#zhKn}=
z9n;u-0oPrn`w|bIH`iM(qu07`(0TLr9o?7hRaZ}~k9?cvt*wtB^4p1bz@tI%C}O|R
z*6%5%T&Dva=|pF`kVjX#Q9zGD&;LWc1n=6HNtNWe7Ycp-W1EqG{@SuDxBIJo?n{%i
znhl%Xn}6%kBlbxBJ^o++46--VRoNHT;rZV$-g86EnEXe5KJx>jv1}<Q?cfoB%`FJM
z$do5X>*?DyN8a$@RvRCah6iWjUbXW=f#WtR@TEB|`}uQ)dQy%%_jLv?e@j+7GOp@$
zAuxJYNjt32C8)4TnJ6u`S=gw$UfNZMPI0PYeZ`Yd&w@UU_6PBCDLzTS3S9!&q)e0+
z+qJM!b-lE!4xQpD)iHkQD_)qVAO2lIzXt!b-BK8-E-Q;!+P50?5O@(5+g)r+D;l6n
z(8DG&UgMrTeRa^jWYwbuzOY93>qclFQp6dvL3`PXA1E!y&;EMf1o3kg8jzEtRMKv&
z`P22j!X`3r8;!i;Zxp{UQv0+APV<HS<SK~>*&rfquAn{TK6I$(Wb=m#^m(Ih7hDOY
z61Kw%U4n`>iR3vWrCrkOhmr6dCA~%5uos-ZGH73->d^w9tI-qW;%H7be{50c&YBa!
zVitCACYT)bu}{8{Wg2-L#QUZABmpaQ(TdkJ{9#FL1jVVw71~!D+WM5Vf}N(h4WuB~
zM6LGN8eQdH>5J7)YhuQx&l?Q~@oy9>JHJ6F$Yoi|nT<Br{79xk6IGpytX+T7`jrdn
zl~K})sV{VM$)TLWe!q#nU6@7J3m(qxG;issXpWDXLXD4;F=n_U7Mfk#TWXCicP97H
zM3txCS#a`dC?l0or1HMOR2bR0-*Q^A@7%D4FY30=Adn#uE#=JA8eQ)E2s2fhsJg8;
z>_V;@{nQ(Ge`PlEz{xUnAV<=gg`6*f2wu`AjyBta45$XOArvoXhPLgIn^2*NyPvCM
zw7X6A!$|m^h+a>(7kqLXFiQ;C$SFMc+ZbSi*`7Q0Wh5O*YJuBfHmrE@<5qKB)DE1b
zF=Nd^m9R@h<;>6;lSLDEs|sH9^zA{_ro<7gK+~R%Vh;#3$Jz+q@eYhR*9=VL(q?L+
zUoRGO#ksw@a@1CX+7TMqV>O5$ZJE`#1)FrG7ji_$f?0Ih8RRe_G`tVN@b1XHRevjO
zZPShJpmrTID_c~-*l2IpUXnlv!s^<pB_6k|G(x9krI<2;Re|^;X$7FQ-7;!#l~!xD
z_y_S#>tn2`6#Yg<8OYH`7Ol0%>|kbcDe_BOHUz}yW0ehZg85IEBPj@u^6*|4R(yhU
zt&cd0?PImL<%LTzn?-|dM3aRw7RLTFtH5aI72xu43yi~)lorTh%eT#%yOx;}=j_3#
zeQA)enU~6q*6bc^<87~dlUxwLBc+^1zSq8}7jIK$)a{oZKG^>CKJ9T|A7QW1TAw+h
zz^c?5S+DZvJfZBMf8>hMFi+#@s8G@V=YK4ckPTOcR@GQ<89kjVguTmT@>14_@b9l?
zpw(l?@L$QYA49C-SJu$Qq!m~HPxa7!Ie=2J{`2DYCT9g)+wh#e4UmZcq9pHOzoN6j
zBfklr`jTMxH~{AJmrx7B->86oLi9sj*~;DtR%u9X;FR(Q$+2Vz`tZ-mIVQwO$eE1#
z6=9up{NB9~UbAYH{~c1R-}xn4NA4O%JwHDAC{GC+w6<+oVs3#XrDUfqseN@xbW9nF
z`|%SsM^j@-y-!*-K^n;%6C=?mwWmIzF;Wws=HyTN6!jx;u2O$QRIsL+7|!@0njqK%
zk!%3Zk}8nGk42~k<pjz!qeL@mBx!}pNM}#Sm4#SHs+>Y>&(HZN2xYx0Z9a6}{PU5K
zOjYE4Q%}7=X-VlaZH@udlMi_C#ivR`o_L6O(Pfmzo~*sjCezV+;>NzkCy%I@E~wk@
zmWX_Suw@5svo)gVtv`bw!+#U)lTT56h61nLLahJp-|F4^)oaeE)6VO0)ADld`AHPv
z1vSOS?a`!)#?xcUL%vk8Vomz%qH2kecOmB$46nw{Dw}wT8I$!%-r{p@p^+Ks$|P8V
z?oe=<O#qDglVpN)PR}0a2b2Ee{5Wqg;S-Gb8epEESh-?Ha6qH!5veKVt<*!z{+d7G
zI`!DDirx3-H<I9rN<WeLG4UJ#el`N@1q8CzM;}ju*-{hEJf#ZWbAsS2Re3vY0qS*q
zh^@$yV&5u=U>b-`GxDZAGGwe;t{EdCu-8&g5Jhj3w!N-QDa41F2L({u`Z6s^ObZGQ
z2w=GCMya7>nJjq$SrYVL??7+&h#io5l=JUWa^s~kgAiI=<2LFNxuoP;V8S$%dL{rB
zCn>L4>?vR}@RU$Hb`pG80HrhQPAjp#NU;guf#~5NoAy$e4x)%g^?hj<1=Jrl)Cp($
zs>dtEpP#yP)fiQWRL1aTHD^lNhEq_&mc<3Ni+q138$xAc#%5}>XRifFjYtnQ*PN&@
z3p}x=Ou)(BL}TnKVsQRcd(bF*DPw{es(~SGi)<}{An1BtMeHPOP14dwLlJ=irSFw}
zaaL)JQz%Rh{g*f?!9vyoljM!IX^<rxa8b7(^h9I_eTV@xlm%I)GIe-SCi-<MmR#S<
zd$zE!Cq9K&37@?KFrXh_J}hl32NHQX99bCZM)K1J<TqiVPY-mqLqK4I@qnTB_<CTH
z5Lug_j4cd*l;iCh<*hXid0h@iT^Hxnn>h%$SM#F1Lt!pnF`Q9<1we12pKdp}o??rj
z=*u9Ztk7X1^+J9{Ms4gTo*5J|53!iwsdkI&)R6=T)DFl43XW==bkP&}F;}#};&^FQ
zq!T4nrG|$70N;-)&Zb8AdRigOkJo({z|{Em5JjvNy>N<cb5vUQe{sP+v558k2dLZ&
zx3=wiozBMfxE+uYi|(vvi-)t<W<oqAvl9`+1_@#pmTK=kdB=G=0*t&UIT-yd5S%a(
zE40wgqgF`5h}x+vH5Bv%uXz`_K0*y~m+0_2Mq5FV>5*E$5)r(15|=7gnQtV657|tc
zcCP|*HE4L^h#?i2#X3FWRFRri#E1zAI0`Ewfc2Ihh*1d=LMhf+j9p-ibuln?gxV1i
zmVL+y7(x*O{C!;retV3`i?I9Xu^ko17lkfRptU&5v_5R91^fpE?|c&$@z2YfdcI4T
zaK&pqY#9<GC<-b~S%E!mTq5vM7Z1&JCv_&!ch?=i{%FO6L(R>QLAH6x$v}W{5J<Rb
z?$=eb)6B$*8VJlrqm71;#l9DU*aRA@RNJfXiB-a7gMwg(NJ&;Cl~&t2(+L@>Lirq-
zK*{gMB;g?zN}v#}-g^}=Y<BNK0)DoppY}&^_(JrlGZ8A{0h~<|Ty~T6Bdhsf(g5()
zlt=R9(nv`w2Fb*v61xA|iAI|O2{BsWff11)fwH-;k%A_nYqz##WK&1;Fb{_Lqznp|
zelGIpvclB>46j@s&FAo#$|x)J)G?Z30rZ9edg1!o#WC?QHm~ihtZCq(x;JFj!_>Cj
zCWI(?-BqZOj-iyy<aq0-9ioLU@o-LAg^As8Q$|tWoFdwvr=$xRmYdQe7hZv_J|6(!
zWm);~wmYUtw`7mEg>0iOd$A>RgFm#jJuW16ur=Q#`w+H!ct*uiODIj_g1s8r>}Dd`
zD)~W){d~pO5_~k#ChzgVP1||?v3B65U9#K!ah{vq#Lla-;~ofeZ6ogW#Pvs*ma*CJ
zQ4Hp%d6^xV+Gi&mm|WSL3fdYxP9gN@;vn&tdZNJJKY`b*m5j9~uJTBl)pGi%F20i`
z&fVE`?sNkhgAkwLDKP<{*stPuAnN;J?^kX6`yjNPo>N_#5w({MJ_)RCZV;?>-rCyf
zol0Z%++{vr_%j*>)qdeL?YE*VQFSZythoJS7A8pGY{1oZ!8$8NNp=`!Am)LOVI*3A
z6C6es7p3OFE!qAO{FF|C@IiR=lWz|uPBUIvvn$MxCd)~!?4Z70Fmj4=S0gK|!R`9V
zk(0R>N&j$uA{LdP`$aZQaFdVKvZq%dV!p?)pGR6<Os?HLh+%tlBZw<pTqQ){8+@1N
zhAi|Ip%1eId!recZ#kaq#cbZpd)(VAMBso44tzh0-8VrDD~E8xRM^0Z?P8CZ5fDQc
ztZ*8Tgm)m}M@Z}mstB;AHxCRgQbo4&B9q^tfALk!O>X99IPf{eC)9KMZ?qb(7+TZ1
z*FOU*sx^B%q&HH6cXMt+W$W__7xI0ao2N3*n;gNbG`!$))0jI4zNYI@gQnQg$?Bwz
zuX9a$Ko7Qt6gHV%>-{vt!RkIk-vN!kKF$UQ*X7Lo_$DU3C{Iq{Y0El)9KDpz*Agq_
z!X>SHuNfkY<573w=;EmxR&0n<zwBQ;Fn_X8J`bDNU|D2GdVV|{XO$y|_}rD7bI&pH
zcST|eqHR=CjDSzTmr~APn*cmTy-}*I(E3UhT^Q`!!$o@W+xOuH1zKuXv>&{JH9v!^
zf?_<Tr0Xm9s~H|UNSUv`Aq$uP>$+xd`{ZBrPByemX`#3Iork>*v}k4b$tV4P7zmUh
z{|da%Jw*A7z#o2`tR9Z?5q>{iNp0Lz%o<?Boz%ng%{JH12JTS4!uX^l2l#$=<-Set
z>IojYB8xv<$j%&GJ=siMbU}~;*TK>DxmDDswu?+=Y5uk*vwJ-;PlKnc10VtGFnt)J
zff+`Q`CwyP(G1#m#^bO1@CXoh^>EKO3llr<dX$v7;Fr}XY!?CtW;wl?1u!tnMFyUR
zTe&I_KaRU6vv2qor}l387*a+NV6Q$cFx^+!KQvhkpu&Y}>;`nTcBBbGgz(|Jpg@re
zAXWjH9BK?p{kWN>g%DC%JU&WW7BXs_M4oIc$1#sJKyEx&VqE}<G_(?pAV`vE&>#S*
zW~xcJ3L#>0BVg9U5G0n%bW1^EDo;4UEX=lxFDb4S50|p%l*uGxB)zYfxqg*GfQdMW
z37A#Yd}kr2Hlc`(&-X!q=N<;Wy*$LsmxSFzUFjl4f{;Uxc6!M_KTg>~rqAkLt-|Tj
z^8lr$$SPl(d$2(3A1wQwCn1=F081?cAl%-|g{BNQgmG1u^uJwzsLVU)grR}Iw$r(F
z)JF7?i0%VQHR-;FbT(7boL1>b$@2}Qx?Ai0y0sfHHUeNBz$<Z^0fY=&kUSzCB!=Ty
zCj#E3uDr-Pcn#R5Z?)C3q7t<O9z?tRQs&%6Q(us+>*N<Fw1JK}0M&J?OjG;Bo6azt
zZUylHNXpJ#9;fJUtQ-%n81QEw*pl^7bPD>xHR0hEqXL<4d6-DVAgzE#(Drhg5k|f_
zavVR^I)P@6tL<$?N^3SU0TGhR-d?au2(PfT5rQBL)4XkAXDZ&`y5KOS-6<6QYkv#a
z&>LYGi%4N%)>+pz7ShqQp$lC|@3b+(>xhxWp!M0RFb=TT8L;RCx*Jc;4ALT#2r;S(
z5h(i~w4ww8in0+^02IpHSFFDesawqh;l>mw?I>n0;<)7>{JGdKh+`Bp_tQCYv`*xf
zNHTItrrUE>t#n{M@EE}he$v&A52-}ZN}6c}-)joENfl%HTD&UdCa!$p-YXt0Bu@fh
z&>L-sZr26Oph8V~If%^B@Cf)w>DXLMeH_7gQLv4a7^W@&m;-)k=xm9-uyRJud|zrr
z`Y0iFhze%`Bt*QQkQoXRbl+E401h`>mnM7IvKws|xoO&#(UQ9VR(!&sfqzK3midB6
z)(so1@xj=}vf=mNsU6hkPxyo%NZDom3aq6?M^*!ZlE&u_n=i5rRBg7WzWkX)S{(A~
zdl`Tg@8HE9kG<5b05%z&vHiVS0ng2F$6^sz)%8mYy*M8(?-h<B^64&Sr51P*t~ekz
z&Ox+|?@gp$-WDRKvG9IA&+r`9=YmFj*rRchjm^w44!mU#oC4vLb!WV^DkP-7u<*t2
zzJ^EfCQHp0F2pbGk5GmbLKMLcEyTY0o{Vp>&|MS6Hl8EmI6v&Jf8PNzdZ1<W7=Qnm
z{|O?K9+MSMcw6QmkHLv7MT(QTpDa1Pw6eKL1q5n{GnzP(ZX>$v;GfjOqG{OD@X(^c
zB(^P_VT~5JJIC*I!Rmr#dOTo^1Vw;>=<#35{2!nt^Mezod917CUu++5Wm?k_9i1m~
zQC)yV7Yf=L{vqWTe=m>Vj}PC5_PJj`{Q={PUrSa17IAR}n0OI1(V^a9M6k;uaMs45
zp<;ofcPi8gb)%Uz!r`aUNdyQd0Xw_(bTq<C%s7x>bB_d>=s;3R%p$iZ08~B~Dakk{
zv)0Elk;i6A#MC0;Wax=u;9FI+Np#K*pP`h5(c=Vy9G^K4y~4#1jslX+T2lazupG<7
zC3GQG&_D2s5T6KvH_VnuNXT0$!$EwdM-hO}84vKICH?<fzAt}<@Hu{APLI#&sQfD7
z?{kpu7cHs<b~wx-euexr2nYy>m@**{6d~$u<9=h4oFbL#o&FyDC;(cG0?-F8zBnT<
zN2kCZ3Fnd$%+zem)}PQ)EQTNo^eC{>S=;H{F7W~Vn*k^s{lN9&SQv9djnDlk)^Cn+
zC~<vUaT!wT<+w%cW`$Eos8xSa6{@Fef+a9I)H0BwgB_Jm>4RD+7Fa3*4FHtT4!raA
z>whmezuN85+8YoEmx*<H$EK8ddxF&l5tK^@g*7}J1@L}b?O^0Ni#3pm`k<LU<{Inl
zfG}(8VQmHqiJH?)q0+Y_Gj$1<U3;YmDOww~Ak>Y#g`)HTw#}}{3EFu|M{P}nFNG%w
zIdp<MJm{-@zxvUv<53SvW=*?%8~{czuR&%1w#0)RgbA?p5fUim@UE*gCL`9_1h1-x
zAX2+*cfep3^&BxsTBzZd1ArD{b%<F|_#%jz!)KidoQ~>H8n7RO2V@Z>^lb)}6=HH-
zR5*9so3s^pQyrNB)+Qyz7r-OlA0#2cI2D3=?Ouyf6nl&YJpg_ChOg+xW<Uk6Nwa2<
zc~Ev>@KY<pIz4C1t;3ntw&|Ri>0<N*&W{9tDo6s)WUjkWt5BEI=JZGe*#RnK0)Yk~
zwAXzy?aWlsM6trOruai!-9gIS#`fA6)9{!IZ9+iVW2sq0sfd$nPt(i^*Kw!`oj&JO
zIT953ys8TdjxdlgT0UfhQ-|d+_ZkBaeEnTx*A6tsDQ_|07`eI32DjYCU@C3?D$V!j
z8CBg?jdhT%Xhy7(XH56mWKGHk=49s7Ab?0QX%r>ypiG~oY&xSFb@Mo-m^lLhTEhzV
z2ioPbOtDxVUG1ihj2chcf%;(K$@$qAtg<tgG&tWwkLLtwkfB{~@(>*_kE;h@f#VPy
zM9;?YMQNb_qE%`cyE*<2ZCi^*0b`O{f@BSk^46s|Ypc6ipX1svsYo(372nHHY3!#Q
z#F7~wIqnQC7=6N#3tD!W={#bY{Ej04l#tO`k_6RvzcB#LLLrP6+)zobqH+a8)XoTf
zF~H*zC&Du5%*AMQ)gdy0m02H+3cK^1GrlK<ouw?)Fuw1QUQs%ok~&7gE`0hd3YUH4
zv};#ZtOt<*92}J@Qz_G$G_*R&6*^MahU*M~s@Je{u>qKh={#t4`b<sTvq8`e!a46}
zSXfwV(|j||LZJo)>h7a>D_Y?%?CH8wjnA}n;5$^?)H~30(Ty%^_g181FyM=e(^&*9
z0zU6MNoj_t1pqWKbSUy?YUchbYhI+5a38~G3h<_Uq_#TZq{S4Cj-l!f;RnK?^LvWt
z5xb{r^>bCR!ey@~!Vffm*Aw6cq}Jy;%`9J`GNOnd)wrrsM8#}^*czFZ>MLD%U4{|<
z5t&SH7^*#|w74gNk&{`@81{aOK5|)CJ!!WpV${VF9?t%9W~|0?6wy?7sTR>*D#02g
z%)E6O^3wV?CZRthD3G-E8mG=3ZAY@$OfL(ELLbcmnMj`~I8$t{3|fW#NQbb@WN43P
z=51L}M<a<&Y(!jdn22|S<x+6jTppQxj_<jDNs?EdO0(%1ZOleDd8xV6(6gBduFrAI
z<IB_i(Y`o?<JlVnp}CSvQd{B4saIKANTf_ldWxQUJw0z&;dwmI&%;0f&W|pHZd~cH
zlh&!H?D?WTS)!F~Zd)DRK62{z(m7kKXQBtwNFa(3sTM_~A|M-++N*H^wnJ?!Jsne_
zfks>yF+b`hB@v#+bbj!;b3^x9?>SO9M<_pdvB{k4Wt*+A-oZ=s6rSkDypEJOEkzGz
zj06o4xe*rBLHy287(s+A+9i~-^&Z@s7vFjHPt!I`VXzj)O|Mkw`YBs$VcIcqx}r}^
zXGv8GQp(z@2<pstzM-lZG3k$x0y--@cfD7>!Z`(Rdfvoz>JUK}pnx^a1d966wSqB`
zF+ja>l<5#!**8v&4vgnoQP3TfYh8^I6I5nqM-KeT6=;?*h-Zl0r}5ggJMPjRdsRor
zegvR`IJ~Z0#34t{Io6nQ9TUHD)RG^%-N3IJO^YR9{K_&ye)Rl6#+U*4vGT@BAJ^+!
z>bbAlg=}+kP3A`45dEm%P|79`@*wb-FD1-fyzqh;H_c))=7hph@i<n5c#(r*UY;}=
zwev_R!NaC8m_$HSL!->W<Z)iH1V94$`1sRfw37{xFV4_stNXOQNWa^oPXIvugAhf^
z&X)8|(2yCdTe^zqnNajwt*Jec;Ox3aSuLwIvy+W$vWx+&vcic5uoCJ(fya4}Dky7G
zKdctM5@%eq7v&-9yMer48R7GvgLG1Mqcvs{3mUZr3FO1i2zqW9w|sjCe(PmM9cMMf
zK+;FjvG1;AhStUKkQF*QD~&<A2ty`Tb)n}PrRV|Gk*XsmKKy{ns9Fe$5GgBGu@ZNv
z?g{2MbMm&An90*HQ6+AnDqO0tov+aX2G{L+b9vy{IN;EA&`g7^zRZTj)&|eQ)(7(T
zskv<MggY^4TOb*Nb|9XvuS<J{XqL_pAX-<&VyL(bKq12Yo)?M|>U3rKHsQfR2|y$P
zpy=`O;A?Ba05IcRaSy>d7lss$^B4e$^xY?vG2u>~-6fu;Hs_`dZ>q>4*|<^vNO<EX
zS|{UB$)qhHy;hc%$wMU_<}zsWZR4vZ!{p7dOkR6&EZOPnSn0UcwuR2X?RbA3;ogZ}
zXbs8{k_L{u6(~p}-bW^)bs>ZTq!39sZ#Iullc&O-hTR8|3uG@e&fI$+v^@;vDKl0U
zOfh~KCr8&(`EFq%J%APmMTwa0z8qMu^+jUB-5A5H3eT-J+C|W=xu;TkZ>xMztVB%^
z!rND4U;K=lF^~2N`vQ>}$BN6NTBGD`p_#8UVv)REPDOw<oC-jt6p>gcqcC}yKx6oo
zf)!@Vzhp>^wGhOHx_eddsaMXdk#Q&{Rb)$c0yuB$jVaRYy1^Q{qcgRxfjJ9FaH<<Z
zuL==Nf{-znQB2EWtl38vkx;`k;)g5nH3Z^dA=o!@BGPM{fT4oGWJGh2u}arEmkLBa
zu>ZF<QmR(-bEvBp*)wlOnZjpF$4R>E4VJ+z2r9m76+Nu-O<nQcO+C`Ve_YVxy?H~b
znZ^}G6N2_w!xswO1IcO`d3Pm|K3?Z|yb$d@yt-WctjanMA$%;Z^1e082p_*JCy)f(
zFYkqzY1AFuL7#NzTfD;p;dIfzKo-Gg&LT?yh%i0OGq{L%(Na8PVcP4vp1CJnA7n&H
z*X?LANkW7w5tklO$gWrMtTxwvb0uRrtj+rUEu&@f8OVs-1TE(&&z8?!QJ`L<>Ljv&
zW5*rO*<I_9>Q=;3V^j&{L^GM~<B9|!d<?F#rxjlt9}X_(F<cn+M!m&!-v{kuu!<Uj
z(!;@l567)&J-gJo-zcG$Nj;Eo!Whb=QTzY8^wN|{c{_giKC@?Z{nhDHPGbN>-)|lL
zIko$MkjHKGC$<Opp%y&|NB~emzYFmsyE~Pb0w;>YEMf#r`A7omuciwWvc@|4LCe`y
zTQ=N~YvqS2qG3~*b`85~nM2_m9g5XG%7Ea2b+xdQx}b<L%H&#hyn7X*Y((XQ{}=qw
zPegej>h8~^w(c!$Z4gpGoQA2v+KDTsV-%v|Ya@F!#Z(Edh_+^YTQ^xMbc6!}@mB-g
znpx-j#HJYg4`Z;}Vde34vLMztd?2)|T6q?&z@UHqxscW0>dlCJ`q?a}$6{B<tg_2n
zGs!q(TzbJi*63L6WBI2ekEEfMOfY$%1t#~-a%1n}h=TgdPv{8-K_v}vbcIVvhB^Lm
ztvP-#`QBo+kv!tQr?6I^?~^^lMoQP*SM!UJ144<%p^^q+Zc7lG@XI`jIiu5<o;|Gc
zA>Y@l|KFos@CJk<-0p%EJh~gNBat+sk^SaMhLglMQQNGU>1Mi<xKtV0xRt8eN!+nY
zh?m&JCKl0%ZfgFV&}k~=A-pr4XO^qYF-VcxlyqQ>QXtT8Qv>JzTw&@oXzCB?`84Y+
zqV<S4M2}3-W&jO@!Lzy4wZ-mHk4#mnPOZ+!eJB9M>d{Dhugbm@F$7$3Dxb-Z1tZgH
z5;NY!4AHOoM2=nnZv^xL{F(p-3=9l>GbsMtNY>pjjifeIOC36gN5wGZpe$Ul^C9vY
zAt^~hfdU1Jyz)r0Y#{DPw5#$5qk-@G{Z^rX1aAtQaYOnM=tBIu=B+y|R32P{MD-U4
z-Uy!BgRmM%6m1oQ)U7|7Cp97u=*3Bl=BMVpPQ?z9ved%P_V;a!URbM3zbUC3tw*k>
zDcAPBifkP^W7V>09@$E$8*zrk0uwhHZ#Cq0XKH<P={64$Ce?0t`m&({Xm6P&1GH`u
z5zb90+gZ;0u2n-$i00Tp)eob`W&j5?>PA>qL}o&x5DhVg4&ZoUIB=ebNUqyyyAu8F
z;Uk-gh~><##*An17Od{XcAVSryLux^eb4j)1sVa^!kvc(U#QPz%ky!q`HIFVyZKBc
z>(Gp25O7Rk2$*H*S_O_B^APwnlvR2f*M-cqA+PU^G-Gt3K%}eyaaeU%_&!1DfFr}j
z>31n;m7<@@=7dt4;A-`$x|ccM)mwNRDY|WmffW3rNaU2b6Y!770VFBT%}W!XGx4n`
zxO1tLp{ECI8m>b>V#HWr(^y1f4EPa8vn;7Dr403FzG*2)z2t!F!bFFqcDcF@wqSXn
znAd=-X{Nt8<y|>fl%Y4$s=Q=`2=nRyAxp187YRv*9`UAdk$-LzFNG65Icrt^cZY#I
z>V!fOx3%NXY1*HzV8#+096Pu|`Hv&+ZZI_$!JAo9q^#G~Y0t*Z@(6VzTc!F|p6D|W
z`GXq~=L?5L{6t`ITadluRQng{%5Xi-<#+*iq0{yYqm|VoR)SssFp$sj3roDTC>K_%
z&FcjvW5`)o6&0=3!&$yDS&9DbDJt<ehgorfdioo$Zc-f(pDU-Snm9+UAkG(VZ|Sb1
zl;RWx+O&u>cyl>8^U075$pEq;|7`|#_GqH5TC=QJeqVBbJQNQ|%+i_e^sb7+=P$S7
z03dVLS&Y*_C}e)wvLaDZB~?-(1u1n6+nMfcoel<r@4(bOwVLV9%*{|z*`(kCR55;8
z#d?K7)MjW5x>9|jS{$qEfijN9gQNf}A~G==qh-w{KKMCF3nYZeT^T#Hb8s!jx-!R4
zYlAA`Tso~Khlkr3r-g2mvBL-T97jnSwL%1d-I~h+*cOCj4O|$|XQEExj#H^JN(_ak
z@g6_fnchrmZ@T2Ndg=Hns{F8dVZH|?nJb3v(|*g+^E8wlM-nAeo<vEMM2Iml7m>nL
z3}8LjretY!6;&E0Fh_S0S6r5LO3I{6N+g@)ki*Zl`roa~r;o9>Jye_;vjI?W(pq)C
z_LB?}J?UsNE<U#q7LnJzZ!a`bbs)d2x|A`gsEj+Vxt8%YGSqo;@QX&deu)I}aVcwQ
z4Mv#g2SR%N<#lvlI-{bh#B6;l3Qj7PdL#;Ag%LXBR3$Xs#M@8@4*}=Vl0U1@S3kDr
z8{s>STA$7v!1_^&89&S!|8+6V=MB~37(p-@QbfdE$xzvNHj+WChaDVqQZy*wOPsIV
zL<+yGSEYn}L|4|szYFlAyk(^qeWoq%>}dHf-pKH(4ETO{WC_=ou=wyy5qXcmi>FRT
z0G7ozEnWDHUGVFx1gS@c0nieYH$`6)=2&nEV0f00hv;~5YL96tx@u<!!ABw~AThyc
zzp;XqnX(2@oqk7Vjr*~b`TBTh)c5tgtt*Z4ng%M~inTdHMd#^~%t=gGQuo+c`rYwH
zsFpPi$wBx_LHjXj1E28QM%$-`$kcSr4jj{mk@hK-xu<&I&dg6>J{kaAD{jZx38I<#
zD{p4zue518Tpe@j<dWTSY&M%_R4OxjyK_gTY={Ax<BVy<V5cng-genXHzU>JcAwR$
z7gyyHTw!VY(NZr5$<AOmGoNWbhBZFZJxii$1&i&aU|;{e3Qt&KU=~%c+S?=ApeA;y
zJTV2v-$>x;i@*+9ca!fMy`d~7Bl-pKFfwu^x$b)IKphfA(FF$b5e&nIL)6GWo9Wrb
zQzF@cq=ra0A&<DXM$L(X3=0r2AV?5n3~XVo`VGpQgb1z_VY>UyB<>j{Ml^y0C@Cp`
zBs~07K3~Bktx-RtDa&T5iGp&X$j*?S&B!Fp2ItHX%qwr^=oL3w3%oNkqFt0ck{wTH
zQ_xfRdV^f{F%XjbAIaDH+00RD^6X5=aU=IX`}wr9`ekj+K3sr$PkN6%dXuFV9Jx!v
ze)f;xGt~vldoyxdeksPJ>Lqix)|RaD-D0<gN`zouJq0j$%7-!jlz}$E6aX2B!3?mi
zY%xCr_K?z?_N^IVDFfVi#w^;>hy%=>C%}iIdcrgOgs6JYUjA;d{iJ_Gd8%GQA(g9g
z7y8iOg4Cwi!>3Rf{au}@8ODZ9A#mgVy4HnD45$h?F#vh%h4W~W!67p9XH&M@W`)t&
z^cl_iGo0}$W^6O$lx5@~m-2^;d^5(Ec7Whlx*LA*D&y0?ifr+%*CZ0e(B~xRCT#!T
zp7a5VEkZ2#*+n9HpM-}&m-o1N4glR)EM-AUTS*J}RFn<IEbOxh5T+@7iA%f{)AENJ
zI(XSvnpS>h@MHMd@*(3F@rQVXui+mqOx4~G0RFxL1ONaykJ8ud9-;N~cF^}YTaUNg
z;^Rx@f$<7H*`s9N+FtUV1O@^ESOQDQK#=w|L2bJ(X<z>0Bh=Tp_YbzZe$=ZB2_5#h
zdCYcEpm}2HV&C5nhEL1S1bz%ZyL@<m)QJmxE5BOIUlqCm0o__dfc+}`2%_3QVsf}v
zl|}#n1b6^|XM=8itashEy_eGC{|+#?)otTUe!CL<+07{S{FY<MP2{N?aVldj9~n}g
zH<KQ&W!}=A0>$&UY*g3_ZC@AtR_#FE+2y7L){zZPKtuSrHlQ@xZDlEUxsUp2^<ba8
zlKcz@BAc8Ncq1&OYp5Va+E3GnXL8${LkO0Tvg4!cKf`a5_2|o$-6y|*SM5Obx=>Jc
znSf7_4j2D$vn9IE27uidcK^?(()3~9_P%XOv&*N<f7!UzbbwK_8^PB>R?PW$b>RiQ
zq%-(bFu$lq>W8zm&c*SI))hvv4;*Oh*0z@-QXiMOt1npegWi`QFhW3M{c$HCOW80;
zrT65~kl>JO)$etI%D>yE2{lybIs=LeZ4Dx{;nR;NFDa732UfWHVvn88%J}RXQY<8R
zPu^lcNs;rxrebon9^u^G8u|K7(j*Nx=t{#O`M#L>1CACdoWGZoNrY)A^G$zRW*Rg)
z`}0=MkaaEJ`WXNqKP99&_UY>+3&y7nO0ge4=-jK+Ge0<RW!j7bv8=2lfRN3fg0DgH
z`xbgi`zhnZwFl~x_t=m>8~~7iG?=czEMAx-QVu|QkFp#U6$H2sd2q36Z~@mrH6r91
zA{M8T7`9c(i<;Ql_iPo}oplN$J0>I{2{9=|#vbS83@zUK{Kz3TON=F(QxDgj#KwF@
z?J5;$AKmq7Q+;yKVXLDlP9Onp_BTWh`J_y2#;|MjkPA*IXrDgAC<NiaG}(Yr3lVUH
z<JK75zQo}4c&h*36-!=`@R-Lwr6a2K^@WL$g$Nd?EW-%kAZC}rtI!{8i=j1iH&GV0
z#csx-aola*<ZPl5s9;NeCx`9sc#zHCl(?%l)jZi+R$xFuzI7sC!50w>jk&dBY_ut7
zhpD!@9uiJ(`^Z3NZ<<-PYpkfI>g9XC#*_l*&7WA1C^<UlsE3~V8e+InRK}YkUm3kQ
zDpXlwrCMg&9K1gHihg@u?TM1-|99m$|IG@NL~@0ygF32*daAEAq|!hDwdUB$U-SNr
z!Nh!=WSXb>s>-X@5-rnm)w`GfjEw^XB>3PDf1CK3{~9Oq2Jrs*k6j<zl#S)!!q*?@
z2LlAC(fR|BM-#*G9rER_9Gt-<Aho~9{#8;MO11I)iV4OjU40sI9qbWD45_WfTt!Mn
z3`a}G9kSE&rR0L7^4NIv#c)d8ON19pjEV*?0jrCRf*y#>L>G0&5Nw_`0>#h0F6sSE
zcKi{$ey7~<Qd5Sqn_g`39W!mc=H+x(yMGi{eHB;YEa1RDZr3j#Uw{q>BNmu}UOKVG
zpNhNpSMf&<yzVErP!f_bDQq|lJTv|7*9U|IJ?Z6@6;upM$r?<HLYDL@s%&g@?qpod
zHK*>^vHTY$J%dDZm?t6_e#1oPv(Cse<&3iP_1jtbHSnmKdUWVK%)YFC6*ME>NoBpe
z*15mLcC!vMYKL1!QhUl38-B}hN0m+8|B-xFsKfMpm2w&-T?Fw+f{2sMQ_58~9^KQv
zuQ2+)E2c#6u63?PjK{r>XH3g5DdE#uMmuW4l%*EHNk-CR=N^+b!%1m__f(nARr2Z8
zk$<%|(34?At6s2m3H9C?(RT74FiW%&0i&$PK7y?Ran<NyaG>B7blE&fxGRt9*Sd`P
zFFL~)$a<$Hp2GLEh<#MGRa(@Fh@V|-#q&9~#xonPVp;{=xn+!_+W8#Hh5C_k|7R8i
zN~a~X^MJ>%f4Ax_Z#`Cq8^0P>0scs^XXr<qi%O76)<rK3-pIB3C0t60t?{k(R$EwG
zvhB35tsG}Vwzqa?XVqZ`jb_KihuC)#`7R;{$P$-wpou!(lL=x^AugtI%E3ieTqPL>
zy%p-TFwagtMAzCZ#|qV(XU{V>_n+p2HX1q&$}Dr2vD7B0x!)V_ikxb18o?Oe|1qN#
zt<-+5?s)nuqaVZh?SA}i0q6@C^nt(jgg`my8W}Y%1kfKOJZ_}r<}b+ZBbd3U`9<&)
zMRK5|5bJcUIX$k%(Q#KmAAQWEFBgsUli(o(_=kXjJR+PymXyI@!nnpzuJJM*h7;ea
z-8?*g$%rJKNbpP%1>7@wsUXaMtbD0X0FKWPK;(G<%q{Oiq~_}@&EXK1K)7Tj4F~2O
zh9jTy7bkho6BqGS2TyR_iEBu%0XK=g&j<*+Alx~3S(&Gjk4_r|Fta<3iQLCvkoO@F
zko5^)DI`9-^<Frgi!3UN$ScBKi}hsFe^XkyJGh}T*h_?zN^OZHG_=wXjCd=5NlSK;
zh8gT29mcv@p6YLc%qeDhQ~tn7(!|97V7~`*O;W6g!yBqPo!IOYGgO0(G@x=`Mk*81
zyALTnuEaBLPCDzHC}3zIAVkeTTc98j-mZ2=otBkZwuPbtTT8P<29O(_8tD}@2@xWu
zfmV6*04Fh1$(=&7w6}K|xEq*e<Xw$x?vl{r+~OR(A<J=jV1IRvG#qv&&2K>+_<^(_
zxD$i`>K)iR3u4pzjEs3hf)g`L18LfYNyyEeoP7~8bIF5&hY6+CX#oqB)2T5(XqKX@
z(ff&GTB>3RT*L!Q7A#r^E*)tAglP4fItLt*BtSFgi0BZCxf;RDg|%;!PzgjxQ#1w#
zo-XvEj6@J#GH2n@wjIiAGlgYtJd1{5%Mtz@P@6F<BS-(gngAl{gb3i3hhSg{J29s1
z(?+t$GXgLkD0+ay7!gtsYEY{1SjHkmHfbTx8RD^mNR&n`h?wp<M}RGhcp-+U9$_Jd
zy`fSxf*PAif)O!QkjG;{!K~zrY*B<JAjcF`xJz?9K?tVAiS|-k=6hYC&7o{^lR1u{
zur~=>6xx<FYkiTt>?{~EPV9L!Vui#AX(N;JFqDQ$i86CCsu$rxJ(mKlwxp0onjZwC
znT)xD7U5Y2R#{?N@_~rNM6NO_4})F;4m4ZBXAX5N-lDAxQm9VY?<_rKFfm8N=6YP@
zxr4mrOn~p<sg^nyFGwvwKC?Phq!xv2HY3$UnnNO2;z(ej<8r8xgi{;x3K&2HKx;<U
zFW5Y@l2FVst(qc~R3;{)BLq4s>J}x<h%@QPQ52@Q$}|U9T{9id2>%{6B0@7B81~Lb
zLkM{21^I?E2(UG5O{c#pPG{c(X(C~OQT7*9&YZh&=_()w27!jAmbQ+rp1y&h*WP&R
zo%cTYNX6`1Yuk=pFXKp%FZ=&|xDW4H6873h?zx}Po=WpTpo0$C?|`!|IOT*bwps^n
zt#l7X5IgY6XD{sgCB%87{5Q`G!)BUgrB(Vx>T?>G4DrM(%l%DQU(JOv<%@;Rm~Yg2
z8>}!thZNGV#6Sug4GOlX76r(nP$ZT}jbw6#QpFjoHQN3;V?yi2WE+j!G&M`L)2c~5
z5;-H2LZ#6eOcop9aC!VJ0Votf*<y*5jGTg!DwB@2@@r2+OUJ+8LC;W8#f(f9GRu+6
z!pg?Z!6~n1`3e+raVz5CRov}PC21>DTB$NV<tkLFQmsa<I`tYfYSP?uEn2lTRlAOK
zI{P=huF7@mF`<q0P19?V$)=cTmf7Z*Yo7V-vp|JPRlV#p%c3e&TWpD?mi4J+L55~p
z(F_Y2)u^?~YHO^u&Uzb|Y-F}cT_0?SvBg%~Y`3G=wspn!E~OpqX0gZK#@Um_eg_<M
z$YDnubqwQpOS}uAEwLB|M<9CyQ0T;D3^oOaClE<w3YA8$XE0f8PAWGoy^)u}7YIdS
ziBu-Hvv+Xp=;Z7Yc+GUn3cBZbczOlDLq5KK{tBf^t<mb*+Z&9gUNo64R$BxZa9|7{
z-ADpAsfWU?<d8_K?ETEO<Foh5UZrR?wpneJ4L;yVEA1PX*4lGcsWn=i-e5ElRV(JI
z@>ZMO;mpfdcSNDfUF7i=mz0+I$}1|Xs%vWN>KhuHnp;|lt`$2~x$d3`6MHA!uwsC)
zTM?`BlyM=IR(FJTJ|t<D7iCp9vF-X{oaSZS_Ji<S@lo6R{p3Jrr3E^Lc<4i@P^##M
z&Wh_v0}x?C85dGFOv`p$k5bwp)!|>GonaV?@|kth#`nWGkxu)k<yC`(0SjWq*$8$9
zlY^6s`!F2G3phhW^+=%r63^*kS(z9^!?<qRuAfY2^Tn`StvB1F<K6z`^z8iN@`@zr
z9d!dlL4VrET=~4;&+~4kmXuOM$lP#qp^VH=6z;!NAm>KR%8%frNyPaMXB-PFKGY6o
zSTk&~uy3hko>HyVkrWeH6vLG&t&U{AS|D`jz0}QDhLYiHL59hHTk7mun-Big+9$-l
z-&;SsUb2Oqs_`&IREuOZy4NZsdA2sa4Xe^F{hJLP+&TslNDKX+4B2w(knT3ksLDfd
z&rib<9Ysc}^i@}=;U-Y4t#8tcBW-2=;g_3<JlJ{TTShZosA?Uljc#<8R0EX^v5J!g
zVGFhTA=_UfK=!-~*zgHacYs}qXTmi3m=@T3D7%xlilDXwyU+A3P19Fe?;%iLb-}zU
zu8_OFmCl-cUM`NMX?PLmdASI$lN3RULvj{$cnYHR<RGrYb7}Qd-H~Orz(<HK{V!rn
z2Fq1>6y-$&mnRR#bvKT&Ar#H?15ipGgUpC$(x#J#l4o2;vi${ATD3A(La+};p2vIs
z!(#dxZfg#@(u0iOMmn;>{FR-q+s*KLA-V~UK);zmau#}7-~P1Y4M@}2i3OYHByAiv
zpi`*WeGQyyU2To66I|yQJ7X70=hM@+)JB`LhS0?cUW&1&-MveuAX9rTtk#ZAbd@vx
zRcA^syLsYkM2q+-r3L1~uF*2q+#<4p&q7gqqr&AC_a-<2_Z%2TPP%znY6W}EUzVE_
zwSzye$^8Eomk0NxVd$O{0CxqbMscN*R%^dV0;t~Q#B!HFMe65yQ*I4|g7*vw*G06?
zBtebPK-8e%Qhv(v1*GT(9XWinW>y0EtR(_T@<x=17C5X1THvHznZ-8AuAOX0j@^Qj
zCucu+#rijhIzj)9QCxEuajWIvfCPsIf`Xz2NBOjdk%k&+^5kja*Bv_R#+ufQwV@$V
zZmgq3A|+FDq6LM7eu`jzZ~S)SExoP(v7!8o#jAJi=l0nf^;HG`lXj=oR||$Pp^OWu
zw6TsGuP6{ElyM=IHr8?D6#~M9GA^Xj#yW1i!a$f%#)VYcSjUZ5j%?_ESm2r(J@2=l
zR-Z*qMwVdoh}1UDTqYyRrzm^>j?4qet;S?&HW9USwxubegbd{<-hAX^gc~m>Gn;2+
zB8zUt%;Q4h7*kGR2)&P&A^{#*;P8^Bw=_0aYIAO>4CEEK_wmwb8(1)^1In?~E(9-d
z0aGp`0RR9N=>Ave3^-+642zlx5GIsyA)nM2^B3)voc=HFUQJK8mtue8+jC)^U3+d`
zZR06!JiD`atWIrc-fE?Ww%YB)uF0<?`B${lii(DH(@>WU6&35Up=}x}8tQUY<t>@4
zDdR#a4aPd|743o~m1h=!2<AdXmcbS8hG^okkk{*Fn7QtUE)`bBBT85FGJj9-8vpk9
z+O%3{;CJs?`d7*`Cs(@{2o=&Gz?SQ%Ld;?c2jRC_@wK|`e0>^p<KfF<zh6g)%Zlzr
zPopxYGx@onBa4RQ`PIPJutQ!AGB3qLd3$kA_N=g8<vJ9<+lNoxW2%E3C_(mou^&HY
z%a^@+c^R|kpO&`kSJTdi8f7%TMdzkwSi2ZUrT?8j<m-S{z^@G%i%?hf5%UCRt8(|_
z9;lP5hEMXo_*&3v*rJlTWT2a`N4QE2DpBfty_WR#^;N-YS;$+TJz9AoP+`-;3pRcm
zqTu&g=F>3xgPHl531p)JAm9Id@6iFRl%ezR2as>)rp47}@L)e~N9U7Rcp>6(ybD?<
zLNxWAJaiGc#-^FOPAs_ec<#tt{zvB&n-;!BJ{9&6s&w}AeytGTv^;<2=l!rk{L#xR
z-z1OnRQ)!?Bi=VKCirQ!Pz3$T;Uv(F&pf;guU~j1;u?JJp)X3(2NF`BLZ*MRAe~LW
E0I?7Y`Tzg`

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff
new file mode 100644
index 0000000000000000000000000000000000000000..fc65a679c2263361a257094d35efba5431eac4da
GIT binary patch
literal 51872
zcmY&;Q;=p&u<h5jZB1j^wr$(C&1u`7wykN~wr$(??f=AmIHzK-+&fogMOE&Is)v>3
zCMPNi00R89v0MPs&kLx#^2h%l=0Em-H!)!m(VqanAI|Cr0XU{O5@PalN<Z8M000~U
z0D#fUaIn#fDX9no0HDr4J{JH0)y81!Zz-ol$N2O8?`NFm5A-9MESrsN4D0~_XjlLM
z;-?SVv>+Av)X3GD5CA~p|A`U)5Ac^D#HRLUHa}d@&pg0Ca01uJWHU2x`WcHH_7j8n
zKfnYKnOS?7{&0T*fb@F+kZIhBPQbUhiGeWyP!;tP!}0@GI61^S^B?ku8~E{we((#-
z4kW<b#@YRcI{^SdQ~>}udEL{-Eo(cYpSW6n006x72jvSP?$R~}?mzQ3|HJ_Z{{vVA
zh@7o~jR^qI0{N4($4`#Q!cYt9_I6G``FH%R84e8qz(oLsmXfk}H2E1jn)(0p*unaJ
z!0et$I>rDUD32&I`)>LND-O{fgosdmUA++b{X<8*Y3B&^IgZSyN9DlZX}2d(peOh~
z04EsS!K6XX^m-6&tNO-8k;0n9w_M50>#?bPO{HS9VO6J{$2s?9Aa;cH+L+hzpHkNd
z-^DHb3cx+HGpSc<K>=~N`~FXQx|UVGVNr2PYehwMMnx-nQW`m(#3vB4^JqY!P`HS&
zy!hYHI7kE{Bz*M16GVJL0d8Rks08>qBJQBj6GS}Xf7kC<H4k$|La_H9_gi*b9jmJz
z2;lnUo(vJWHNHF4V^ewrSfZCVGmGQmgt2<Htx(sEChM&0L@mtBjz}hkRevEnA0KO6
z@uz79c^5*h8e3->ciGc$0$~=!dSQ8pAD$RQzs<2pT0$~f0v)6ki|M1WCKQWHo!rAc
zsLgpJO>Swnc0qUz@o>_82i+v6jPUp-gP-iuijRn2uwOdh!(0riv#SgnSBRRzLG~ON
zh(rha_s|$<yD+>4lGJI*RMOt-G6orMlU7JsJ=?&VE1{Dqrzw}3B9f!?(C|ifKrjES
z?b_LoO`}<UgRHhvl(%rvpr4-yU1-@#&K_eTslMhk9OW<vuD66jaSs^9$lbtdbo5Z2
zSyAu8ss58(ABSd%Wby7Z^X_G{Ot}k$>k>*CAXaT-nU^eDTp<!sZ!ES?PBwYuKK6-q
zUq2hLtyPH%m$}1DH6t$6vqNDrY;>2kM%vGdedha}5-Cxu#7*|(zE^gUv_i~E)#t}f
zwl(9Kyl8ixg}l{{tX%4=4&5w!`Pgv0x5<wuyMAlR3fkp_`#!mCrq7m7(Sx5<0GTZA
zn7zk0XCA{wWiffN=Ze!HP-s1xVn4uY9f&vP6o7eJaNuNy`mPWPJ`jJz+l4M5B98!i
zOBk&qpl2UK=dV=!%3bfb*HRHyBg7y#u}`-eDL6DTV-}sT@Op{fNFf)a5k>y4&6y;%
zHe?N()np!dqNppDtk!(TS^8WpZ>O?{F^-x2v%B0glT);;lT;9gznSiBXfB`nG|ltY
zFTo0vGK#c|ifD>-5RsA*1;dDzREepn3&D_~7=j>#+zAIT#og(7G2Iy=!oo@+HEj>)
z6tCO_XhR_IVG-1oGwq)IU|7hE-F)wTjwdG4oQ^U%V1VO7qB}4Y39@)$IApFHsRbAF
zhfExp`{#AJy~C){0#F|W<#U8i0=W-J0t{h=e~0`!jpyYn!Dot#KAy$+L=+{Gya5I`
zGEcxK@l!4&LFtSS)P4{Hil7Gs%7++WZmqPKW0~OGM~TSDM5fN^8?`!kJ9&>9ma~XP
z&mT84v(8HI<L-C7+zrw!SZkYWw-pv~$WldBL|JM2d|O4IM-w#ChE11D4#`DnX=;s&
z&6y2d4(W_RuNhf0HBpHvl-x$$CO9(duzi_!RNx9dM{V+ARgI@MIYuy;!&Z5hm_Qw4
zkqr5_nOE|?$&IF`WhXGh$T7_X{^mJ()OSUAk}P(nnm^3Gt(LQkV?05_ddGpFV(LIP
zS~+<()wC;4HJVT|^&m2uDS4H<>}?>&oWo-RHd2Vj`Wx->H<n#>$H&|6aV0Fa7A(i8
zqzcs|Ej;;FwJoixkpp5zMj%3-W0CZsly_Ecc7V&y>5iOSj>BVo^j+4KyfBGN<IuP<
zu~Dwe>-x+tYY24#YhspR-r47!n7W`hfvP|G^)8>0nj{G+nz{6Ta^8tee#l%(4{vn2
zj6N%VOc1MxTK0}*;$ckMF6<PJ1#Y(1<&i}#+-Kq4K}ta}lE6eZpQlWO>9B}Q!=kB!
zBerTKqsq&^)zrU3qpgqVq6ZCH39Pe>A=$jKv<Gy-0Nd1%c56tmL7#L<ACGsyl(zyO
zBR6C^t@03tjQo}~(d8p@bkuv(nCIKF&fhL0D&1#dbvX0?8pOtA7z-B4-Fw=AaodCY
zsr|zs@+^(wYWrWc-!J)*_Yw15yL)XXL*08r-SL=mYT}o#<?~%<-d|jAENs&mQ`0OM
z&y(aAcCA}lv6`=Lg}2iiL+`Huf&(G%zw57%1dG{DYfe$IP7rQ)P*;12-Vt*5SeL5>
zJbE{*UkT0kY8`>yM_C_m*K^r&jmKJBliRn?A0XGe1RiwP_tQXjPxV!LN72}J00?`;
zP<zZv_J9KRko^;Wa@kxMatG)V2XHe7_+tkk42~Wp_j2E!1<$n0Grn7luOXpt#M9*|
zh|{UP**{q%aydtR%h?rdF*SU*8be+OS(ck46ADpYO@DKqi1rUj(+HClNl}V4EemDI
zq&k_!m5WD|Jv53(sTg&1ld<m7J811rU<EzN(NArEFI%b<wPu&2S*2wAHCLL}PEiWo
z>jj6WcN<|F;Si`@9m5|2T`A)qpo|G^6md_GCtqqts^5EQdMG>nMvkjxlr^a`9{ojs
zDH`^-&e%^>?aueMOos1woM6R!Srbg?DU-z}QFr;eQyI^DbF-exgKh$2m9WNAx{VXX
zQaxOR67Hu3T9qJgRq-O^JouU)@S1Q2VqR4PT5v<`Q(o9wDcs<K03Jo&x-$;Q3H;`)
zFWML6pk@g&$;+3<{+ZFUR%123d`*sDr`j3B8a4~h9Gl*9J!Kc%yUQidQ`|UQ`ZIU?
z`<w%TZ!Qjkw@<mkr*cV-EmIXUL047j5y(RU_97)cAJV;2>8Hi0(O=!_)58qil?Uiw
zB8>7Bt8+L-tObl2vxwj53$>IiSJ(@N2~!qzQ>8!(x4-7U2uxm|tAv?B!W$8PFT%Q;
z61+g*>4i}LoSjDc1`+xneLZ76Jqv@Vy*)iUaCun!nBG`2{`ni$nU&HLu+kGSn1Am;
zFj32~a(m`S)%^4P{dd_=n5KQvz`@Z)MGe5fhfWDO$k2;*_$B@Qu>guPFiZcP%acnk
z&S+}D;5XOrrpN(8kcfhiObEp4kXN*Wq%B(Z2v@y2lK#Ln_!KiKd-8?izc)>ataK<6
zQu$#9H?`(6#KulfcP~~y?aZcl?ytO<?yuOMx3;wb=-wg%gmX>uN(D7n(`LolNOC^M
zJC7dUgQj_0nnu8?28de%R1bo<Lxzl+`9$e?eC4^Embqk>d8}u-v}bv|nuO@}^mz5N
zC~!M{fr1oK_QHh73KRkP6d`*Q{^m)-u#$PqW4Y90dEAt_#FTlAo<@K-ed2V@G=qKu
zEJXdaY^bz&xYc+N4x}(`NI}~J0bNiM0>QZLLK;7M)S$pF%z!_CLI3fEGWyrI<b<XL
zWUYj5t_1eK;EuLLthdDAG+^@cy)h+$Y5b+-p&om)o&dKFbHEHdR1Y`!8@pTSj}$g6
zPL+t}mzFt#q8<P`(W_l~L`97PrlB6CfN7JxDI%H&(Plt|zUwogPuZ4Bm9ABbJF@$c
zu>Fr4y8vn}k7q5Hrq}?e{q8roAsA;Klj{ys)G)U*zZk43-;@c#)V^T#p{K1rF~VJ(
zgnN>N^by+FnBi(lck<+z{^7rX$|zM^Dyj=V_4x#*;^Z%LV_p*$xsGBB6}|b;4ED6h
z)`bl#a;0QFv;_gNheq18cjH>%$;a0!LDeLejVasVmCXyJCv30!&)UvmUhABO?)uIr
zm-pIFZmsBlX$DsP*<1p$VJJuLnULf|Wfu!BAsqXetyAFSv#Wa-_O4v-nL<+v$A2@%
zC-*%!AlcSzPMci0+H^Kmu3X(>y?S^qvK{5T4sM@~pFuqZxec-nv%#~kvbT24h@C`O
zf2Z6ZHIg6eQnD<vvDT+&)0&9t%FQpI*RH0T@bK-!fW6n@E?H#N>t~PFXb?-IuW(!_
zx^{F9=-SyeyRE!m?mw|TiM-eO4F55$`}LRlJWE^lAIcef7j?3?hA1c0sYVlLduHcU
z@3x7>Zj<ABOV$SI6{}lG`!>%*1~rFoA}KA86l>z)g;YAVw9kiUT@hLFbeo@Yw;uUU
zg;UK^I+7w^XL|pHUzCN@vr5)Cw*!GQc(pQk0##QvIfg`a%7sKshj#x(atB@Kbf+lj
zHOaNmwa>NAHQqJq^)EZ?I}tlNLKD-mizJStm8m3f)0#L%R*SOMJ>EEG#^1cZY0E4X
z-4%E5$?3ek3??(Wy=|r}w2N#g8##hVe1hC7`;JD3JBUcv%FP$Ol`5<xeW<1+nR!UK
z7GnMtT+F$uv=?u#U70`EPfFh<kNW8~=vj2%*-4#a+%DPm&&3iIqkeAMWjALMoM2c#
zA%6I_Jo9Oo4Psl&6>3K3VVZhJutB>pJTdn{j&D<><m#;1m@vx14Rce8b44rYxagX>
z0P#eL`$W<A`{>Ft8`3hHPO+^0Dz;*phxBBY!QEdnkUoSWkxl3N?L8nhH0U$5v^liM
zBf1nhw6rd|%&Y|xRXwEoq3=v%?d=j>g)eQCbldnR{oloer)Xy{dr#NOAQAh>sQ8|8
zp*=RO6Ld#r3{|H8bt|SD2hK7(0b48RGdl#IBU-i-Y4$Bh$AM+X4t4?o^2QkTMxV5{
zSOx(D6Px13P~}Gu$Lnv9v`8*Y<Bf5nj*&^7iSbR?BmAR%-lJpPkbeLK`lw;7(dpA0
zOp>{dqiuEnZ++6QEVNfS)uV2W=<Pbn??K9r9!_q%p-;EU!E@`;3Tsii?cG@GUV@dR
z<(16mRsY!4V7b?X&KD-Hd-9=M)W}<KBM#V5cBoYM1QGX?MUQxwd7fz-=_&JI>_KRC
zkID{J0jiZTv(^-&V08o4LNr?ui}d1iBbg;>n^{$xdDqU8!#Cl`MtQ5p_&&5eagnmH
zi1NP?6+!tbiVT!RbCd;jlqGjVlM&^GW99kf<;7>^dFkavYvl!Pb%!;u(DI@til!zC
zswPUVhw`#ul&JwpS{^@3&5$EHvOz8fhE@{ba6D8XLQ`d8S;k~hlp3@u6t!q*D+?Ve
zPptLy6E!c|V-|@{&LvAPFHSF5N-t^B%%Ad9(`iuFX_Ve<6y#~t^l6X<Y19pD5D09P
z5K1mJcvIeI)~jEXyku5*WD$5|mVjdsqG=G?XOXH+uVB?IjM2<}^3>~WknDW)e>)R;
zoX37#7KNKNg*z5<-Q#IG25~(ATpSW!9AjqguzQc*vJXPNL>M_F+_D=Mz6TW8M-@0^
zlGrD~*hfa%hhFdTrtSfy?joh`GpOxAZTAb=nu^*Qg5w(y{pq#OHu(LOX7YlOa;T>f
z<7;T~XK?Q$ef9%H>Mk~F*CJ)dgmv%Px-Y|eaNW9}%X+-sdRWVP((Gc2{9@?%q8H=h
zR7=VErizv`-DSePK0w24D-q}4Y$Y8lkwU7A44bo3vx^kpsoCbN&gL=?&kPgKf)kR`
zgL3SnaBQWjDpZB>TEqc{zZ@sWj{<-J5Cup9)B(HzWq=|e5}*sv02Bf!etd0!4d6!u
z0Kk5;17Lw90N_AYKal%@0{|Qt6aWj{_Jacetb3+T9FOb(J!0@MJM{s$IC=sN6)IKD
zNtq&hynH%81vebuT($5MytswAwtY$-lMZc&enNE9$`I4_<jH+Kik}tofz1p09!%oP
zyYx<tFCoP1U55l==90>_L`NOFlXWGW+uUiou>v1>B<GoXcXZp*wVl>jMC+YC#;q!f
zbKdd&Z%j%Bymh=iJg2(w#L3Js=@q)5Pig%gI=ZZO(Zmjmexcl6DoP$oBWDpK9P@`U
z!n>ujze(p}zKzCRI};TwY^e~E7^TXfnEL`6LPAlLcU<UM`rHhlnb?PhH*(U#8p;x|
zm)zKu3W)~MESV@g+ga_!Flgi~Ew2#2&a4i8{t^%X1ON$u{!w$lzP|wo?wJjpP~Jvj
z@$_GBr*$-|Db%A0bQ;anqfN%63Eq-jv9}ydO~XzrPR0qN(u`Pg;)2ygSjc2hID&qp
zP@t@<h}Dx4$SdIxDO~=fi2V@M-}L<Y_1f(kBhQm{-azBu<yU{?<Pvy-K0m(`xjT;)
zEA-r4TwGLKR2m7PWd#tql)(i^PlBpd1Za{Vi;Byq>I2r?RS}m!UodrpV#PP_0|+ta
z+oz!qxIf^w%XejWs<{YkA~Uqfa>frFBc7}SI!NASlXs^viU%ro+45)V9R#AnM0<%q
zgSpheY7mk89`JNuTv|Wixw`sk7xuX*_(K7BUKR>e&O^t+qInN}<?`?*7Y$u^Blq%L
zt$v^8s2!CANTpa~>ABQR?;d&h+4(lK&3rp$g!s*Tv)LD9;jXb-6)=So+^RM#gH>JX
zbr@aNgejJH1o2^_N*wcC_1=3nshMF#RiBHg*X#{*IiJ4BLJ&2PA|UZkS^FqVGGZW8
z!liB=ZmvEC6lXb@jqcnH-wr3fzaP4xvE4jX=wWz!<<?C_smDHR2DVfv-}FNK7{U0X
z0gb>!Dv+gyevBvSfy6=lSP%pQ$Ga}D5+(sy`m?M#8&)UfG}&Pp-2UpC7o?X}#5;aZ
zFC?1l#>ZP!H5=<37W4Rgf2w#6m@~47ygcpg84_CVC(^i3;s+o3<ghOA-}mjY$e5hc
zBreWj9~Q~GnmqL29QB}BGa+SL$u&XE(R^44dvZ@Ipn%B0=|?^ldv&Dra>3MTf>ru9
z`QI&LGWM*+X*3bwTNV@h{B54SHqeNj7m22Dr_aynaw9FabZ#3j*bcU2XKBaq&~qQx
zb8C{5sv*?!Orir&bSbS_+$HyD#!1AvNX4VuRiqnAL~zk`sCI>rVo8;npamIqyqIX!
zOzn;@1m3+x%%nWcOX=3MShSqddifmLvv|4p2qc?4g-}j|w?<8UwH(#uXSUGy;|q95
zSf@%cW4}2rIY3>IF4KbRCe5cG_=<SuYIXI*y`!cV`rQ|x{_Nsa&`*O1qS#3CJTzKg
z(!@&h*fc`%KRz56i+79u1?u_JGLYz3#lmW@syPEMPelbu&vrv_sL$y)p{?pcsFHN0
z9q!5};X?P^%9F-Zc3`-c+-c@W&#K9z>@{d8lrb!*+)rPf+S3yA<|8zoC7^Vzys|!@
zK(^iRVhFFJam^k>hZLe8=g_4QR8c#snx`ocShP~BI9PfoFyXQp0Dy7DWAFnF?=Q}m
zD-zB0gAMV1tITDT^a<^%7Fk!M{}axBn4=oYV*3c#cb4(}C@Xmj$^Lj-?*1}vvWuUf
zp>>I-+iLym<E`Kq(ax4WNb4YQf*2B<)i=-Xe;niy!Bgqu*KHz<U&U!{kd;4*59*Ez
zKm%45L<8n_2DMk#1SpH0+w|^o0sPQf21vbo@B0X8l0Zj(Oi)pQ7)AF?O@N3UKcRdj
zSVZ*9*K+SXYDVFRn7w&onJ`tn9s3L`6au_qqcVU4f^sLXEn}eE<?@SuwvI1>G8{<3
z5u<%-gQcapo(#F##u`C%4X>=soblp`TJQR4uYO(O(8fj2>9uQIu*~e!CTo5EGnU-2
z7E~LrK9S7KN{Aw#j_RKE1;ix%Y&B*_zi`Rek(p-2rU2=sK(wM(>oO#5c5SG<I3zB1
zVo=Z}5A#$1BQLF522N18#R!eePT9Nw(xHc<^gy9h40o9-&Q{1z21vf{IKV&vTbOwb
z5G^o;6c8Yh%^2mh(7QR*0&2AMJCo@jBnGuwIdBM3%9i)$jsMWr6P^&rYo&HChc&q^
zJg1maL7>#^0Nypb3k3bc1u|nCM4Z3$!=o9QJ+H)r1?rw)H$9EFBbpymAGM7oSII|s
z6=@2w;bCR)oN3mCrZ>P4IWl&(7{PpA%mQU~>=(@fv>U;XA&x@WxkQbmm~fsiQFX@#
zPfuQ%Bx8nhs{4y;z%W5BSf9Wv(GrKR^?Ce+IeuSyJF%Gw=}Vh<_{>Hw+TYOHs45WV
zQ=fPcj0<O~cN7HGVGR$7yiA#4KS)U}>Y69oJ09^X1Q|4m!d@rZ&k_vmNH|O-_rWp`
zw4BRD6_LdCr4ebFD|1d+R9>NW+Q?F@6@Y&fV<l<Xp>X`*zptjAk$JRF0i*8iGBlo2
zKOLW)@vwMxc0nS4UNm#$_rfXPAkzXWg0N&6cO=i4@;jI=MdUt|wP`GjS<w+FSO+hM
z9L+hX!k8luxV#=PPHM{si#r8|$xb5xej!*bMJ7b6hFE=Zo%1K1*_%599ii2#lEwC|
z_6uUCf3;PF4CP||gpf`!V?4LEUDd#G39_+avgjrOx^+?{5=@W1f>n$9Mndjg9?U1K
zX;pokdX0Y6?k;cnkCf{pLZ|SVd!X9D+u)<`1-4qS<E4oe4zn6AuqTDsL^%*?GPLIo
z6O|YOHmEu3<Y9$g4E`9BJnd|tC`CyYdfxmSizoyZ@^TsvP}X9-yLSDjPf7Rt`Fb`^
zf9>_Q$8&QX-Iv7;brXzwusYBsC7P(h&To{52v>(n4mJx>hNZYH^P=FSnEe5xSfqEO
z)h&@)ABWF$k%&-JcRft^1dLY&tkfOoBa!%}BoJl7TtgTSvC&N@02A;8tP&+fxe=2X
zTCEYtSupgI8`e}IGzl`}6SxW^d6tr&Bo#`BjrP}2Yw(5W?Mz1@Lqkviji!qT^&E;u
zxlQRK^&<0vhwx)=MI-5sMaFveC!vNUI8XLPdr|uYc~q3#U+5Q5F}Lbw5lG52UNOXL
zn=-<YP7}d6iP;ihQH2q5;YUM;)E<sh;MD~%8&(+F1$59jMed{zYC92H5iB@{s5uKA
z|E3+p_-=?hOiqidjSPo&0iC!jo`M1%`_ZugXFW`U41Amt0^s>=PX27;wzIOTN4Bj7
zZLP=$wyE24*4?25T|(RZb}Y2E3>G~%CXWf<;H%=4<lu{n0nOZwyp8$W?;2U%<Qfd9
zj1v(wty7WGM2rXyI|Xt(@+IKTFv!xsVj?V+eriT?Hp)*c3FG$;Yt}k6#`+cXd%$?r
zO5#AS#edZT#QW4UG?tL;k8<s!e1K@#JeGBQXU;4$+JH~TJ;|m`%;e=-Drk@Db$HkA
z4@JzK*d{PYqU4yS9DoKMR7El(4`>gQn#M!&BwJumk1146DWu`#4@}KN3{6RtQPXN6
z&CtTK;0fD|f$0f|swjDXDHA~emBq6w^T!sNT!C-rdUa9F+LhYUU)dL~W81YDUzC<C
z*Ht~c540B4VgN#nrDExJ9&swDBnKQ_k}EnbOvP5V&0t{LEf)^4NdU!e^^Mg~Ln9P>
zH@KqNrg_*!3wk`Ra&+uX**21(JH~S}4s(v3GDZ`pm1LuD$vqCeP-w^LIBA$?UMI$Z
zpLO@g)3I*xJNc6iZ%)#7Bn#CA;MjFgV)6F(9=5ZEJC-I$Ui7xa-#a%c4!nfI>i1OU
zBnGOb_NssP$fR(KL3rIqKEV=J9i|*cji0P~9CTRgI=;*z<|%d>pw_f0VRw6uYvx79
z1>?EP2Fk!GII1JqkWrD8yFI=RPI6uCA71YB-;z`H{sebV3m&2s{=-oiWKcvilZw$`
zzKy`{Nt1)O$b_c=97;LB6=|Rn{lddrEq=%u3;fndF)UIVCX?E4AD)6ujmbn>T(!7^
z)jjtmdp1?e-K3ehGBt0OaOaA#L1RIcyhOjv+S3|Tir=X6--dQhQR#_=eFz6M1jMlA
zg<$blar8LA?5PF0Q%a1dPHizJEs%v&dwo$g%aS%+CJJZpF!T7FOkNPkO3P;Rdqenz
zHY<43q3?TDEH;FZm9bYOMi)jQj^`N^QP>Wvp#73YzD33RKz%(^vD19>a@)kzX!jJ$
zX_8Z};X{n1>B_47K4Hq~Wt9P<eM~xX-2-kevu;TXi=db^s{$pYjG&|8m!$BGAP=M3
ztwT<AiEuXQmg3Fw@rT7DOpJu6o2Hq2#Idl)d<la;0cD5E&5Ud%6)iL@Bh^-SweE~M
z6h#x^h12U8BNm4p*=E5Ns`e8*26809SIP7yznvdgr1Lo`y0DW(&(_jGQ(G1rxV>=8
z{p8Bi(M8hG!lP~+jz4`1_Os3+vAWPzp)fem{L4h23#&Y}c*0pNDs>kKpWDbd{j7Mo
zG7$usWt`^VfMfqxee!O8itUNuGW^-L!j8ahO83P^OipfiQjJWpY#@2>Ebn!zR#Ikg
zLfxHu_TLQl^N^OW1ZRMP9rKG2W(lq3{U0FVHh~6k&l5x91En-0DOH=janfpY=~OdE
zKcOHMG13hGdBZW)JYt_~PM<fO+wj6^dyDWvBxh?J>@{#UsnH^EJ;%PASMux^Z=GNJ
zF6%-u<%tm0T=sQiNOkR<Y2tYEpRPS(^1^+k$Yh8*roDo*MFsn<q))GvW)DQo*o`NY
zq%r-%Lsb^TYiaW()0I9>mWGCwu`O>CMzK8g*CYN`@VFY&bu4!NQ^_j1%H_d~k~BNh
z8T-+g@#^v5k~>pNFA0!}4MF3IYvnDYCdpq~ggKMhBw4E%P!q9-b~NAm8kT`F1Iu!|
z!^KtudfVWG88R>j<Etg&)bDt2OBc-=+a&x<=(%aR_M4OsnU`Tuudof~ozs(3gunr<
zlZZGuh^$lH-bmW2jT?T2hY?szB(L}A%YuqdVH0N|j1(S<YMT32ZUo(Dc-K*Kq@A9*
zts(mT?!LiPyVsZ9`;X)B>L4-Yi@xqh_mA;W!!|IS{wK1+W?WhOS!t~oft58z+D)@C
z&qQ()1ZGq25fn5KJ0Sy;DFRK)MTS-qg?gmc06*$pAyVGVpwQISK+zQh1o9o^+jGcZ
zVp)fOkSFNH?4Q#@_+O7>&>wPpt9FqUtG-0Nk4O~UFg_G{K7T2$gpk-&ENz2`b%n~n
zawL~l+CnE1CEt2x6h6{@9uyp(X2p(0_CG$}u8-4TrFBaf4ZxR+z0=|<PBLa8u{rGV
z{~oSNNH2g=z%<_|-gz1wTd#NR4;dBJ3TvHx^Xo~BH$A=itXK5wY<3=ejyqoz=WQ$P
zQP%e+p|0ph48_!#VPJ6;!HL3%sATdxXko3D)zA6!6OCUGsW?$lrN0i#e&4w@Cwvv~
z1(5MNO;W9@)u}PI1!SmG<wwQy59&yPy*Jal)rwfh>``|*1{WISc_uwJgPb`>l!j&x
zaLNd()Y|katYpxv6dD>u1CO<igh<t%aWgeryo5k#=8BIMkmKXb5s%sNN9xHZZ}_-{
zpCz<c&**r(Av1n<UWAUsUz_9rNq=zF0vn)D)C9iVe*fs!8vualWLnQLPo)RDN;VCT
z^L6tsvel9=80H7shM0Y{1^CtyY{w<tIwvau9!!z0={bcjOjC@1AXX@elRbZ%T;V*5
zJ=WS&md4+1BR#G4F0LbtsS7JDPLG>45c1P<@nyjVl2;K_yYe89gT6UUbNW4IADE<|
z6SB<ba}w)C11h292Iw1;TJR%0W=`p&u%nyVf>1v9_rug}+Ujat&W40f8jp4Vn?Fsw
zY<AWqEfq+>?CFiUnBW$VjWfk#g6+c9GMQiuaPDv_2e`*THdfINxth>rLF5?yl@K=d
zrH7I$V84Is`fan>;hjlH4__ijAP~HdxUom%U|-&)x?Xg8NDAJEs>yH>Obam$J0;yA
zR{RTauP(;nft}{?Vp?W)^w|sB&b-yiqsVJiCt-(Bc%cF-HC`|=1{K=bH3Fmuk0lxg
z>VKX=;f9>vc6YzprA}%8dBwsFUmb^Ay)Sf~{<`_vhLrAZ0WhA!2x5BV$eN>TqTWYn
ziLjRF;}r#3nhS&I{H5BNDlH5sx2PcerK;o_td(L0cCFcU*{f&nIm5UPlB^f*dp#**
zB;&;iA-c|#SeqC3x|E!?>?wwA+BWxj`(-l@Q~mWB(lub;P*hS-Bx6T$JIItR74s{f
zB-@@C-*!GVBgQ&%%9UA=z)IhmQCaU~&?VaA+q$*-mdKEjA=NjIZwGomj|z*U<Id=>
z^7|42+A|Fu(-$+WJ99smWe}3~gk{+bZD#>^MWsSVf%wZPRzcC@nsvS*bUoxuG8|^m
zX+5?VGit;d3sv#T*6nR41IoYAtljvgwY8>rdJkeR<Gq8RZG@j9X@id9Hk~QK2utAl
zA{z$y;smdL)1XqW`}e~++r|0f6Rpm@i+OK<O}8w(JeHs*!D4+hn5XOPdqfm3My(c;
zQ-&wpvT`_q1xb`wK}TK@IAK|MOG{ykk3O+$si+>KlVP#2UOR5PU4t=dy7z^~o&8f!
z{Xy3w(p_`Q8gfrJVT`zq8d^iAxem3VoJh5HYwM)7V~}MVcxkZ4vAl!IXap4}PRn0|
z9;1{=`n?evruBwIuC1?ryDdh}VyyM2C*MEMm(AT7+M-9YS-d`!@9Z5tfXnPct9^V_
z)Vt*&;uxg{R!WUd<FOu(US`#3<1YG98%CX%$FfbMwcAa$uk@3VdbvM0fpVrLb>C_R
zv6?1LKq^3;Xp@b&A^#)}F&vrT%_`D0L06I%71sIV#D{=te`^WK8WRv7*I$#A&=P&S
zy(Uz`>;BbB^%2x=!BsnQCF)f(6&3vY9)<3NrWC;wizfK%5Sc5)2Tj5qmx!U!)r!WK
z++C#6kSX`{EccdpH74mccU@L--=osdaLwhh^9DkWqo_1{UZ7MKa=oZkPV@LcQnL;B
zue=NIou;Z>u^Z2>XuEP4U1QjJrfb>If)m~4E)nCB{sGDc72Q^AN-QvfwuW_1a>dje
zu;S%#Tmqq7IIx1^@AVWrl${?Fxtzo8Ts0}J?LYHIw^O^8@ToxvO1HTZ_eNtWu!+=?
zMJI6Fw`GWTL|hnaOkwdJ_0=HmR%baHHR-__O&izuSvOdd%+2^Qfs&(({3uu0-SJgR
zx)$@Rp&x92HtfC!HoffJ&@<W*pFXN!Iw#O>X-A!?66i`e9iohR7dVR6kINNyH<_ME
zTVvR2^eTI*T6!>#(f9dD*s*I|VLsu>CmpwQZ}3D=Ip<7P%%^(G+{ef<pOkrRd%p!@
z{Q^Q;RVYJ=A{k3S=0v$64d%6#^&N4?h}npl(F4#K9i>eqW><h(lARQ1Ov#mGWwMq|
z?fW~Urhab;MC+m@w+EI6N|9qFM`A$iGI2=Wv9Q&K_Oj{C@G-`C<O)yw$}-~w8pmxZ
zrWitWD3p98l&inZyrZ#f8{Lw5og|P*jZwW~(fqt`^J^*GIB;>^@EV5HGm&(3%7n6;
zV~0fK@BrF?&uCdR@qk3Twoyi5lzaAw%AQBZk!pS{o!+R(ybByYg;1OmzXE|J*J$-P
z>=`5!-h}V5?P&_+z)RzMFRLb@S%Fi~mk*(#0_sxCMNl5{_s}mtIs^*Mf`tKKQPKH|
z+;q%VtT{Y^Ti|t*Bo>6dxmg}@R$$!w@)kqy3EH~&PfL|>pVD@d^7+Fnt*h`Tw~#2V
zF`N!0vF{msy<Zj@<)qgbZr=<mM<*-zcH7Cf^0->ujc5g(q@6;xRvaD}I~+it0zQRE
zrj$FYPzQ*4GEPUYhI&a%TxEo~e%Dw{AV^7yX?z8VoDlh>EW^=?x6sK`DFt=FtKE%W
zdO7Vb>)A)|p72qk<1U_JO6?UUbp^BY-8J=g#)WiDjD|O{_*2r^8mo(l*;!g=s8wB_
z1qDv5OI=i#t4x`{$6@CY$Y5MSXcm>;_bL7)&B4ty^9xu%qMZIMi>Jf_5|rB;FcBG5
zdbG$N`}!h$ea#sxn&;zb`1*Y|9qigfII#{Mx5pJ68u#o5wK>rregAb-)kt~xu3!~0
z&f5-uHG+1W8OG`O2@B{J5~F42_Cdlg1v^vq<}aA14(3LP5dPhW&8lSk4|=~p8~U^5
zZgT0B#E5pre}3Gj`c&F+Cj^Du7E|1`SjT7l)PHsssKb*n4Zj68deml;#HkqhJ|_Dz
z6yD}~)qs{^GMd>=iuLMBQXNI-Aux(O|8N%c6X0Etm>YU`hQ1JRXHi!@Bjv>8gX*hH
zZoU#roWO^|*$+m64+}70ngJ9@#Lauml`8`#6I;=gQm<J~R|Ckn+EU&&d_nUk01gS$
zKq|CL7)qpw7l9r@zoF4{6BSO(+T)OTL(epFX6P!+@Phx$VIYkaK!zQ`9zv~27&A%F
z2_7j58Dy0r+8Zp(ETf>}s<)~Ts*&XM?S>t&u9AJ{qe(W+mcW0Z+SEJj#b<t(s-%jm
zZi*tYYe2tPg3YdSW4-}E!PLpZrU@Ei8W&w*xwzqKcl}B#1{LZy5$|6}NLApLMx#Va
z$O!P3RnuiR=}Q)`A6k`)AwD0)<qCSI#(n&4h<uA1hxTAKg`)k#<kU2-VHaqwAMhY2
zLq&z2@nW8@s;yoOb-HVqw}M-2w_DUIfwaYf8|<kfNU-u^)8qlmJYnZuBV>~D1T4tJ
zB=h%o#gOT&*$^Fst`qjn3YpB1tdv?gXxPN+RJNeWj{VX>yevIR5Ke^(ifL%hE5K?|
zt{SqJWMC7w70VOa94k9x)Q^R3Zi_VLk5#=_@$pys6}s<V`fa0%i`=sglVS7MZk``s
z$uIG-+FwnvxR7rUPfcdmSGKUY91eTpi3LXl*x)g;nOB_b$nIO`Us}!0H1nwz1s!*R
zeHjN82axs*ICnHGS;BoB+O0R!_6;;(-q(SJkCXeu3S0Y!0$Vs{=OjdiJgdI(+Znsh
zJO?U*{3;NgNMm$!cZVKD((=$<gbBRc^$es-cHBc}q#27=MvuJh($DID{@p@2X51J&
z#1FXI<J@t?MU1(7$oNAe!hz=7C$q3Durkjg92|Z{^My0)!k0o|y!V@{#uJBAmO>?O
z;;=<)QZD^PiaUN4HA<U;USes&InwCoD6&@nAoj(3WM%c_kv;CW_ds=vE?CpLT)6`o
zj@}ciMEo?jfvD^uZw0@3lEWb%@d}2>JyuOx=E$B_q=Qlc)v;2}hC$fk{^ozuEA5aQ
zpZ9D!Uc4vMARXVcl8+S<fOy?&Wa`AZq;#0i2+eb*co!<Y26BI^d%<34zQ49-vN$HS
z{@{bTVbWSmy&o@ZKgLJSkm9B2O8F#uzkJ?lCP@54^;NW;!tHJu^NqfiNLG;_(q#5U
zYd!rz1KWkMV+*a9dlRg+(*C9Sak;@xv;>X7&(qEq_Oed%0ata*KqQDySR*v4+5mh~
zdPbs;r>@p8i6LHS&57gjE=AXR)x~x9ViFs>?MCo%C3VwIUbr3qv2C&IhJ`G?TG>Xo
z>C=2(lv>?=-R1B?+GHdi1{n6hTdRE*T8ZiGYd3%Q1lWSP^RF~JFl}zFyK&sj6WCaj
zG)JJ#FxMjr9ZvM^D>OLw3-r-$SC*Q&Hq^nGKhO!NZ3c1JHkIF%zCH{1{`=W4cERUP
z+r);f?=-X5cK_J^A($?IBF$V@Ie!mExqf{-Lu;+O+`E<kBf?%o%%MbTE_Z(I$R%_4
znVHbK<3#Fytg?RUzQ-o+YbhF}Jn`lE7idv9@^5CM<r|y8=1r5XYbo**LW4SjByw)G
z^W|Sqh^BQI5O&{^jnvkX5VYn{iX{4Q!I&6E_&}2z3+Flnk61IETXW+x)>G=6q7v=a
zRn5ZiaE6wsq3q>eVEL7_2=&_Lo|`Mna(Jy^W{mX*kogDLTIyxXl7tTfR?!zlCXpk%
z4L?7Kkuj!HFSRI~^J#y~q7rNga)#v)rxov<Z{Ep(vGkc{)bo*9D=)Jv+w$StQ%1)W
z;L}c9Y-x}$z}rfMLGFo5_+A5TraSGNz}cnILsLc1(M2{Srg`GYy_2V&W>_dOi#7qf
zdq17D#A)}=>iy|a$Llqq-r+T!Yvp4L=8=zZv1WSFZtq?yfwT<Agz!o}9-_&;*Rfy^
zY>M;^n79M>u^Z~~Jctmv)f$H!NmNC6YaXoy4U{Y~{PL(vx^MgQs$JzeQ@2Khm6}>e
z$8EOFyOlx<2(Ex#;FqucL+;X?XV|F7bI7R9DuMTN`yRpfM~ua7>U`yPXkImpyc*q^
zJUC)LV`a7IR<$SxWI6Da_qM;(76oT|%X4xMiA3QehA0VK9Z4J4AoArGKb2w3AoDID
z54*1ssR~|<#+FnlyooWfXhHdp<?)o#VX<PQ!gh-L+r6fbN%-h&mb5FEG-<Nu<?(iw
zx#S`%%i3cKxY4(BcWrNejRW&yaXRw6PYQT@(FC?j$R1}EExw~@l?l36V3D2X>ay%b
zMcTRT%<>eZC5ruV0IqB@Jhor>KHZ=k`>8C(pLJM76tqOBB8ObqsL(v#GF7R@Nq^<i
z!Rq;J6cJ`PSu9LQseBnbW!FXJ5S>Pxq8RMU9E@-A_tVnR$9a7_*nfXmgK0cE8-g;_
z3EXV8A)LNqt+NylmwxvxW8j1*W*NWtUMTU)`~C@)f56FmVZ(lb^I(2o5U9)g77zN}
z_jJdw<Yg9<pLQ$K*mx~B<(?A>-lGBwr7?U8!{VS_=h9;R?eMjKPY@`jJ^ANb{^`9%
z@l>EY9k{C)0<9Qms1S_e`n;6?8M1TZV-Qw=w{3d#3hrZT<T}d{C$gj55=p)S(1N(5
z80KjU+CmzulzQPnsa0UyeL+?_BGYXHY`t`E?*_54xgZ)^l+j`dp)IY5uCIwo0f|eH
zP9x>*=zct9(4nnc14HZRJuT+O*Z5B2H9`*Ez1rN+(h{^R{|gmP<OHIu=!M1l4sA~x
zkV!YjE;k!Cu2&7bR}?mG8*~5PQ$Anc9=P3X7wrTAj0=t0Oq2@;bPNc#bT8xdh7WRK
zq2@cIzZZ9;n6ZtUwQccht9hux=$e#Q2V7Cd5IBJaEz{E?k#qk^^>cHZxn46bBwI@H
zS5&gXz(LKLE@n7>Xzzdwi~xqC%YI(=3MC^8x%nxh!nYtFh)6Zv6M&PEfoDP`E-hS5
z@Nl}P`5wDDUSA9hQ0t!R8<jGM^V#|o-p5}|H`!(Ut*TwMENsu`zY3gsb=e8_qGLR^
zA?&K5lj0UsPMX$oL2bpmNwqwqEJji`C<c_8(~OTF_#B8o37vn3<VdrwHr2{8gYdXO
zE1m3WgFz3$R@FeC;ppsBwL8{v{OZTcFtp_WG(cZrmsg18VmiTGr5ZE>E=%N4?taAG
zYR1}|8E!qYHjQk8C-nvk+0-fk)#x4}lC2cVT}XiuB{C1<N%`o@{$}mT$cDOE-KZMu
zr8vv$8*x^D(NMw~_-XZBNrug1wt<$+$}x13mi79&=5YDAWk}$c^Geq$2H%5dv`W6J
ziG|aM+>f9hO(hJnsb@2c;NiRY%YfPJQ3Qn)jdO~r3~};^cQkCb&&NG6w3K9h9$(JW
zD=6tAWAhopZF;JyDU0Wd&sX%L`<!!Zu)>ec8@gB*@gaI&9Js$vd@EZWQmlcxkE}vO
z{vNMe^v$>z_*h|yol1Cm-xeIm9I{Jp9u)TyATfWoM)Gr^K<^Vt#K<J09``I^9eLvZ
zkG!d%`4sLpez-MXcKp??4pt3YJ=N5;&V2i`EE$9}h^Jar$?qMj8_YMEew8X1Hd&di
zhMBcC`}u=~Xqc+s-#gjva^BOqdVf}&PCq`1YRO_dy9QHlo7k;&eYs@otw!<hYMj66
zAwNeT?FgHGz2^p1{pxnJHY^b<Jf+K(sfo7*--;6D@$xOZq4&!UIhV4p>MhdEe=j!h
z*WRode^+%$prkP{mH6WEnvvIm7>^`ZGYIS3pfwdRaz+LC*oTuluq-QKmBw?ivxwPe
zBBuCrG36PbWTW=&0c05FNt`JJ3S__c!VKu4Skwv}6ddb`Cck>seSBc!H<o5*SWoQC
z?=`D*y^GyyU1m?2Gx{9tY>aGdNLG~y@L)l~3m)fq%w5Is(8U&PQV7<v-4->@oC~nr
zTqIn$Zi*!D5Kq2asBFc$lxc?RRgYggocNj}-Ki)yUuS+JnEj2g<vMwFP42;?1jSbf
zr1f63qMjSg?BOT>DYi$f?=fHeKAZ*PW<9@~s9>T>>^yrgPu1jMDo{-f{j<3BMWyIA
zFSF{~?z#N=d~N9AxFv!<Ld9<eKl8~Om3@$KWtLj02_KGg14#j~XtKP6kwnc}7#dm>
z%KO%%ItkupS-vyHTK4&IYP!+bTo7cI%Yf*b*E>*mDxT(KZ$fjdloLW|m&^g-65wZe
z`Ad#{GhZ{uQo?mw7B;3eW4vIfwmPq@r<`r4^nPHZuAq7=HG&MK>DJTgA20LD^KI9A
zg$;N{V=Yk_dn=s%9SChd^?paiA?<ev_a-lS27ifN9wpCUDB*4(U*#-B@!P>TL4%$a
zl;b3m*VDuf8f<8weSEMG3{2?F*npr8=Ga#7DQa#fTY}I*Ns=0KzFc09Tl03rwGj=$
zL<cL$F^m0_)_8`-fo!}!QsaueuWv&&?b`M8<38@$bMX3lzkR$AWK!3_yRsVlMli<w
z?@vMHsBqqe1bG`{t`6Y&0wtiAY4C<<Ra%`LC#B__3O$~Gn!Z<WW#{?fVvI0pcHSVo
z_!8WGXV@zzh!um&f2X7Wwbnd{I3`dyk(EX#;1r~D<{k-#STjFDM(Swsx!JyMwAzKS
zUAaFl;+>l%+0Yx2VJhx$=_Fy8RQxTbJjzxkrs7bvk~D4(MYi==n(4G(nv&d_qO>*(
zXqy|i5rGG~ETRcv?QRQ<GN`CGbL)tBP=Y5wA!ai3c0#kk=Fah>q2nzW(JC0Rl06-e
z0vZeBLi<piZe&-d#0TzCvR7DsL38JCLvimYR93{xQ<9oQ&-U8gHTbz;fzF+L-VyG_
ztmeRKF-RzX!q+L!6?G1Pc(~he*w;xpzrMNbw4jr?e7xh+&2BfnI($TONah9f`3Tbu
z{q}YH_pfx)2i3r`l)D@-MEv<8v6;xF*LFpkPd>kDmOg^Fic|VMK0Qd!0heivtI+d_
zIIY0<;3zNOvU|tvX>L(Oaia)pg=W0@54#?Vp;!a2wtzZtDF7FuK_M>*ni7GT46iZ%
zj8pB^dQdp^E;dJta=ePy%;@;<haYnW`kW!ngg_^v{6vNmKU$GFIwVua-Y>-0U*e?3
zeRW@_Fna6{!E}Xs8=u4<a>oYg<^!5_9iv}zWrpA6GE*p%SDQB}%#S5}P}eoaf7V5_
z3&GqkIk2(l6oacX!O!V~We?z+7mEMHFInu!R0gl+B;LnXU+D&a69g~Mtwea#3R8zZ
zi#8o#WE`>qF3u^e^1Z{E`BuK;V;}g$`L^TmcM{)#3WE_JsuJ5C*bJ{17|l|&Hx{YC
zs6z=vyWj3dvYj{j%g(rMw+%yJ26VbsnYYPc-9M+vQY$1Jh_Zbz>eer4-vp)(vcV$0
zAq+aY1Rfh#$lf%oO^)MU`GH(8kBmvO5-SC;S#ve$V9&xg+q=8Ho`=9^J|Fp?ziHos
zmHHTRFO>Q|qj-a4dn6z!GE8ZAxsjg_<GyZb=Kjt%hdCdjo}%Uf(7<MTYVR+qkGqP|
zG9i>5E%`<N5=B9^C?bQ{pH@rhKF;jWg)y{B6V!Ffx#F++%wGvT1U5P7(uYwx0IeCT
ztmwyMgee=zkb0}aeXz-8Py9)duGIS)s`aIQ-9Q`cCLP-@U$ygr(gXg(El0M~ENAMb
z9m?mkf5(=4<$;Nq+ye4eijnSMClpaIvAs?K`ylYlmbw^36EIe*Fr7Gi0e^R}NBp{J
zv3c5`(|X3_yd?g3uKp=pog>c-Eux(X7IuUkU2AeL|9rn<-|-c#Mxa^fRoy2yYIBe!
z;#G}lRaM}$-Taz;E|=_8E&S4Q1L~2&9a%Q`55GFKT9H$Ev{5{K>R+}^C)$*2PR*vq
z?E3)!gaV3XWw|BnBD*2TiXEMI!#ocOkEZjZWA`J*Hj|?5VZ?^Dh#C2#C3toXu<Xw4
zYF$l7-F32BAHiqoQvm^~l=1KP+Kb^Qvt5FsXQEozm%NbOpkbK9Qe4kOds1z(JLRX&
zgQ7gT>!~lrxV<*9lk^S}F(@!i5e4b%BN?-`pv*TgCbM#VO?~rPj_SAAj1*opoP`*C
zju)G|l`At=){#TSs92N(EQrbAM1h0jVFQQLWsTAi3dTos4{*@*v{0-^3rGB8w)YU;
zz~zXS&Fb61NAYF$!&J{*yggJ8IQJ}qhu*}N8;P#I#SO;DXVwEr*)+YL_L&<D(_Kn`
z7VGjv&u4L}vZMsRqELNX;VLV29~zlk!m5h0$g~s1I)$EOr$MK<P|*@sYpa}6)pLmc
zR<TqrMTONf+Z)et2wv=yoNM*$96ebCgeQsTG~t#;D%2YY!3U7<{QCyi{eZv4?+(1X
zpjS=e3gejVdkLqF6xhOPm%<{AXPz(F9u3PlvmDC4{vxhS1oU{t8#^Z=?Tz>)ZNH=r
zJOnj#x<C`zD`VA%HOZd1w8}NT$Vg75hJpQSrRLqI@yXEF#iho>E)%e?mEEE%xU@4k
z?cX0BE+G`%?Ot`O<~81VPV|u9%#eNQK!Y&faMShwtsj04759CL9v2Hozemd`bC&mV
zqu*^Qf^2&*C<G^5GpZmaB`K1bmKwFkw!WC2p-w(gjE+2+_yl$>;@7Waw0A14pC_&-
z3dFY7RyP?|Z!9mRmOd7G;=oDGqmin#g+r@>F_4$!FTNkGqaPn@3YMt~*(m8$P@y%1
z`z~*I@OlvI%wC|K^wEXY@2UB*+R(!bObDuzdL<}hX~gNGJpv5ezL$8bMHH|_@C~&$
z*Qs0QIQZMKDWJ)v{nl|Nyk8fb#(`mA)tsJvB|uV9RSYARk+7pm$njA|o1FtsM09l|
zTwIHnmU$7aDbo57?wT|UYSF7N`F^eE=xx=Ty}n*3?O8s}ukwYzFNuxN=4<#Sb<k}v
z&!QF+__EOhoaY4YF&Nm^3o4wwIIR=%pINk~L5c|UNxl0&08}KET#p-=+H$L_zc<jE
znB=4o*1?hkyjnCc$at^z(F^6BJs8`Sblw?l#VqR@&B{g3y&n$9g}lhtz9P+dh>LHj
zL^r$g0(o@CLAH#=8;`H1(rd@BCcdnVTYTMv{>Z;S@fa4l<6V70+VyLP;dedzwIYUZ
z4CjMtjj^*7F)KHvNGVM`5clzjrMU)4`M`BA_bKMa&pi)T!3U&F`$3sPmm+gkj49eM
zsUT97sdrt}9kx!mp2Pdp&!bdimS+>VQ_syIx4x+#dR4cb@gxt^(w#=KvN505J)a9s
zi8pi~I5OfBFvazU)V3US4|*GR_A#}<A04B#i+2S$(d#jrL&4WSf=<pbr>yl+qw6K%
zKS?26C*y;@`I#F1^6C3ynkczmxV>JOvubXEQ{<D7+RQ=618frLkm99mnqqwM#C6@#
zX+ynRv0`#60cSn`{gK-J(fVhdzkIl^qAwMbDbwq2)ai-8gLNr09WO0A13yiMW!Q2b
zh-+|q$jzXtQ+WVUXbefCVK1dtmU~&|N1N4>%#4sM_J{agPMlD~PaAL9AcMi?rFg5H
z?u>#bKER1E$-4J(v@nk*w>}mUGNt;b{%bYjSoGE**xT%Os@i{W7^5CW2ZTMwre|1?
zK4y}ldmYB{7n(;Hs-tx~?lAHmgNW-He=I1i35i1`gQV+h3%oNtq+dhbD616hY#}`8
ze64)qS)D}SV3vlojNWNlHC~h(yvphKse&M-4IJ9RiA(X<{{wSCjK3C>X~EeaXr1bC
zbjSO>p`LzEcf*3tplfLL)|KcE1Mb2b?-e7eFfhp$5I#PMO-%4$2_4@J#CvROubASP
zOze3x&)U$;*D74<%zQHi&H2rI8<bU!%+4aj_^NhR@NwzIe;y6`CLXWWhIvicDhKlx
zR&HA&Ktnhcv)O8+$)j{u4^HQV7FbvSv^8Rwy~;D@o{ZXJBh8j$2j&J2c*gCSP+e@K
z#d5{rxdV~prKF~<vpuQTrlZ5b7R|q@OTFzG@BF1@M~>*)TLv2an(qTB-sfGo%5YS?
zCKNGz!;n4N64C##q=U_Q$*yF^d=n)F_OPNkUu2?G#$XDZw6#<hU@~nkwE^m?28H@3
za8ggj(Hfv)nTgnxbqzlE+^J{Z{t4YDu6g#<bI)a4cYSylxjK6fUhqhFC8vP~0zd<G
zz(sMbjtHMw1%j2b)Q0#_;k};Uo&26#wToill_Bp^jHHTgP9kNXLe0X%8Y7ZqgI0qk
z-d*-eHID#edtmrd2d9pp5jW4!N8qdw!N^`u!egR_5Gll(O_eZFB|O#?4gSKXI12;Y
zx$~K4c8bko@w(nb*H}aQSaapf^oREDotXa6v12DsimNA2)<%bdnK7f<I5mPPX~=ti
z^dnKT)4Nl%{=a|yYt_EcJ{8LFT^x5_Kv#!w4z<iqp-a{>oA}a`cx&+wn)7UBZsHMO
z0>=k(kQNFvz;9bjQu5RC^SH|4wiJW1V0PxdUI)LZq)%?S9yFkO*Kp^siI-16sU|k3
zJnn$bLp|G*m?SD@OT~3KtNZ~KTOu(qG(>bN2e49eE5Qk7#`Q9$s<)T*8wUE850>?p
zdM%~>`cQ{?m(d<^2P$^e%#W1z+teCOR9{n}jp?nQFjj?{y!w)>?hJhV#t%McKYpqD
z=*^aqmZs*~tVy5nG<DDEkBYC7_1|iAppU7t&qk)|$#-iGV|;I%^W!$*o#=xe6XJ?F
zRwNnaq=s`q&>Em_YG<j0KXC2Ct<*4G`?Q>%(X{5{v&&~>CmpcmL<%%#0!Ea2E(Qtk
z>lEeI7v<SD77}1}iOy`3)2lmhSY^;n$JLrV=B-wJ`&peq_0f-&1*?I5m$x)bSo~F!
zMz>kPz>Rg~lU07pgtx6+hx8}KJLBB-2wz26!^;EU>N#id$h22vz%CM`i}F<Ppc9J0
zQox{AmP|AhH5j5>N+Sdbw@;UGyL73Mxp`~3kgeg87%FuR4o9b>IzF>Yjm#a~=dpUI
zQIi@rZIW>lFDigz)?4DfVOJ;oyMKz`5A7?-mK?rt;dfI5Q@`Ukk|fUTtKTC|DPP8n
zi1eNj9zpIb?JeWpLeA&uqS=TjAcBAh6SgyQAUJO9w)2pq4A<@=vWQX(<uVjj8<Cvl
zFmXqO?-8<eoQ$&4QrZ;0(~BY(Pn1YDPm~Bcdj#;3OS9Gd?mY`FZIcyYZ$_&(>8qUn
zk=~kEwY$8|*kZL;8tcsZnfbm&+e~916oCK5)q{(@dn{k~2U3PAQ<dK2Ob0uLD~v;>
z`W9<rO{LM{^xAqS5`#a?{uB8?Fw_zX{x5+Bs56@lqrG#eno6`+!|fM%cmEYTDmij1
z+EGD!tcKc<Sn3HvDNaMZIq|7XVHZVOGs$*Q#)7R18q{za)X?T(#+5`IXK)~H%b89+
zEV*CG>@mCGnoakpI#arXgj9*kx>TpCCtZ4f%o4l*K&W3ujOu|9A(J~MU;e=lUWQwC
zgM2)@cL&*-eF86P=V+rjhH;(q2-mJsLmk7GXjH~|Z6z4w0Lo(_ltLK=n2qd!1j%^?
zjbZp#-FcROIhWEQESh*gjtK}XOiW*#o3jQQt9I4LhJ#Pv^O@jqTm4*BL%_NqHfEa*
zbzvu2&l+OeVs>%W4!7(l$)U#PI>Y+=1ZvO1JVtHeYjh227Cy?BpjO($IwRZ5$9quj
zDeqX$;-O5rg-SZ^1XQxoJ<t4Yapgy88e!`5Eb$Gd&0{%xaa*;FZ(djNF~@b%zJ`;q
zjwjfwjTC2t&21if3Vhix-2WbpWiUCjfB#H!(4x8Lp38cBFFVrG)zy;h=y-_y)4%$o
zM~)o1JnFx5wxeS<+cnne-s^7dXpOXX{B&-P$%dfzhyzqMJdMiuUn?6HrD@5HI1>{K
zHK5JJM3s>_QN%Hg2u;Ps#6(Qm^5RM+vl9LCjW4V|^PSV*Y$IKJh(G%ZUVE}HkS;-)
z%Y*KkiUs*ei_$dZ@|c{1IgfJ}t(RuFSJOGnGA$6E!^K8Dn$U8@!pl_hlNNu$?RO<y
z94?{R=ZKtt3x8V%kSgX>K^ZSD)y6fPZ48c568k-Bovcov16XWyi4?tPmN09|s)LJM
za@8I1@4V<g54UfA^Zwo)y^C<Yw0G}^*`q*PTL(Dlqj^`52L{Q#g=a5eq98A4H{QI}
z_@aZ%yub)uP#XE@mf-I2?*Kq|hkphS_=m{iUrXqQvJUlr+-%kfPvQQ}67wG>((bZO
zco_gcgX?#=L4!Hl@L_}QGQ%I0^i(G`R;m$Bz{4%<4;8FxYV?>Ym5W5+LZEsHB?kUl
zd5>F1_pTxW{!k^Tme@jD!8Jb9nzR+jsL~4K6vSF+JA<s9L1<?Xniph_S@=LL7q`*}
z+#)30H=veR>*m#Zc(qM3v9tLdqw<)QN@Kw8Kwt}l4(YMlV#$=pdhUt@{Ci_&xvDRb
z=&M?uIrC~swB@GPUc0F!TJkEHm>QgV|F$vs9NR{&^mu+aKK>q$XN1N-zgM(?4&Vg7
z7sdEzH?Vr9%({(Cwscc9Wy|@lT=uk-f0xtQP{)0waa}7r&mww*To2Gv{RugWn{qsi
zI?!_7*BX&>XduIFJdgg?LF$bSN+OdZx`-s-gYXKg9a;fH@C4j|GQLJ?==A8Szdbws
zj(4iH8smaF-cz-!tlr-A#rn><4o#WHG=Fo+w&DI=>E5GGwZ{3>U5|f(Tz0KxD(0zg
zt-Nq;$G!nSxlH`o)rYV9n^?_`f%prA@{u}_#d<-8PY7SZH6lqJ;kJx&R@%leNooa?
z;Ug$=vWV|8DF(cZO7&%&InY8Ctk4WToMegg*@A+ov{F>s0AiI@Dx{$bu4JMp8{nd8
zXpO015&D}Lydh#zhLhvQEADJDM<M2^fSM>3CU7HBn>(P0#A4IQ`V#HVcfCLR8*<2y
zY@0}>1DAERZj0CJ7vJ@MV#|KS)Rpv)H}(2Y8XDq5O;%UM*n)b@Tpx|52g0G+mW<C^
zgAltfo^pi-TjRiLQ5W?A9r|F|y$x3xDULkfzFtIqP6Rk})aU2a=j7?yk@`x`vq%qy
zvNDXfqcmrN(kvnbcbUBB@lYUZ9;J1Y9G$7EW6~}J?lb|hv4Ahz|C*jSeY?guZ|>}`
z+O2oi?E>m@G&fpH_G8M`cuBd=G<VbZn4_`O{rH(D&s}=8c}JV8Ay%>epQ70rwmUKT
zDj`S2S2b5&c3ozFEtR)gxV(isZQHlrUoMjQm6<?MBnO(kfFz~4fYK|KJl42x0?)|R
zLb5NeyhyU*Wpi_%%=Qxyq@c!jcz>&iYU&GU;Bp9!3OfIbc@S81P3if5WaXv87vXvt
zom80rP$^}BtTn3h)Lm62nSoU}A9OM|{vu_-z@BH};Q(XL8sTYp_!9ncQra!8P$+Jw
zu$53@8C$NRsUV_lE)`8@o&qvQ)=8I3S-UlpcYNpO`w~N9_IIRGGuU<?dGke!uk7}l
zHKXbC#JJELn+Y~8ENFCI*Gy)?l>LPu*;m^sj=o-uqauAF7lC(<o-V><3WGowt~|W}
zF=*s61)+|1NCzUfh^R456}N~;`U9qT)ze?-FqVdvE0dqf`7cnylxi{Ynpq4W%SM=z
zgo!v8By?7EppR2i)SbKF-xW{5zqZ-=E1Z3o41Drmn|20go93E+_QdSm?Dp+*v;TPX
zSl@}A5u{(tS052^-_{}gJzAd|`BbwTm06sNw#Sqi<n(e5CBP#c#+wnZcAx+!fTAcs
z;A%Y!4+r2{!Hn2TsNEyAGx<zdlO!(58PCjhCkrPR8yX*TYP(5FR18aPfahfV`N}x_
zdjs<_o_^((lShx9Jay{S!o&g@IdS0Nl@km9z1)UgR6Bu(c~$7d9TeiiU8|raOMBuj
zBWnuY!&`GRWtK+F04i}W!x|fHW{G#gZ<r+>^JT=T?qC6ET(&<pt}$*Hq~q$MRtaWd
zq}FDIjS-E_a;6}Y{6u>C#||7A>mTbM?Mh_EY@u%N*i9_m^{I~T?v7K;&Uo^A<KE$(
zQIp0vH5M3%xwO}iY&hzPhS&cy68FU;EH1<YA9nF|pfacM9^CSGanXRy)`A(6)2NDE
zCyI1OWxT%1DETtv#+zkMm5j`+jG3N7RLgv{6vxv^kr8=rFU#HCeM84ilg{HW9VaV6
zmn#Std7PbDxBh4Hcy>;Cj#93_YS1N(Fe>-3w3gLexL5l5@_j;QA?+8PnU*V|%u%G=
zT;gso(gu@`8ny6A<Om>CfX(kPRT%MvDB8)BciwsDap>o9sdtOewb$0`F|HiXXI0rm
zKh3P8Rh3;3SiCufYh`%F7I1?SE}oa+B{F;z&wu4`oyt!fih6SI)#ue;DwlUD%Zo3_
z<+T^sc=T*U#BtFGGFzYUL5vjY7Y68fB&n<<jR()nl0$zZ0vZ+A=3ezV0AC2JnERE1
zX8eiLWdJk&K$Se8i8;!*%PF>#ls91<ZVx+dqe7LFu%rV718AAu@&sC+TlS(CBmNef
z&Tcf-mbK~=A$!nS)ze7=&FW1LrG4YNSaV{??yjvWu^1~2EtZ;6OJlXayH2HQ^%tE~
z`(8YyHtL_q;XNkf*O)gRIlP<l#``BST%A|`qw@F#<?$O;(2bwc6Q_nve;3N5lY3=&
zI1hfC3~yGzld6Ygcyx>R8w%d9rS;{07t4Fk2V{6W5B?q*p5SnL<|?IY4VzDY5EO8&
z46oP%ZcxJMd}8nt89s_<7jZcDz`5yr^?CJ|%H>_k@^n72^4dSJ`4orw<bnAV0_yz;
zZgTeW;4KUX%K3WmklWmRvLe8$02G}oR&=6?(%0WgCk`%l=zyH$S`Y4mFxMh0w_LD;
z9Sn+N4+BM6d*UYLj4L`z>3Kef9C>FdDQDT|6~`-mSDHY|vyYV%j5zYoT#|7flW}Xz
z8;cyyO?hMe6B(|~EB{e>?1J*xjVfryPicu$!lu6q<<ZE!GCZ6IzfFcWE8rb;PH6aC
z+Wh^7g7=ru`f|UEDeAM+xuoH9N%{Ww$nXTMNhRoLUM?fZ;=RHMM&4Bmqr#Z%hnQGn
z_kpcXTI30aFLGW@7y)4hsG^d#$UP%s@Px{lo|{ax>?B~Vupd)ko<w7vmK*kJ=>CXE
zQ&EQbd}(0C8=!atJvcZ_?I$8#M;c`N35%Y@P}mrcyF;0MkRBv#lUE17BFG@lqpdtL
zjjNOa%o3T?)3)H3f73C!Z!zifxV(lD(_~Le+N8E*0@HJEHiz6k{X|J^pxqUpv8eR}
zwIP2v8gC%ZVCdr^!_>Cf*;FFZ<aWjT0=}T<#+4rH!1gw0U2Viy-PPHX+Sb)F><^hd
zP2}CZoq<4SFC9xdX2^$BnBPe&oaZz_&&yU-A-w!g=zg0AuK;*?xgfy>JYPFP&)4q2
zbIXE!@4CSAIybDj;PbQcd0#9vyn;V5n>*fXQFFy(!W(I<aV<3hMnDh`Qd2(lvD+;8
zv=U>6x!Ynv2r^H*M$!cbblVS3m#OtueM8XV3fo%}`e13J&0JI0TvM@pxhu0*H0Gb%
zJt(%SRce>V=Cqd8S48xtib{h`qZ+EX=PEp}FNZIfdn!W?UlLsWSC~UO%0qORjtf0Z
z0e%}j1KbIW^TUOw-J#%&cW<HAmXk|)GKGsA*ZvohQ%m79Wt$$?F3@;QiP+_%!4pLR
z`6RCM{L9?(@?3ntqCIv@YZ=godV53d$s`%MIzDK)>s{}<t38URgD*zgcSWO5Mx#Uh
zhx>=b552(8LLa7kK+TxruW$!w%blO1qog=X$#ph&dQv~4z6@meGC+12rK>U{J97oq
zeZS%mKxtE`j^@BbmA#B_Az%U%BXb($L~Eqh`oFL>bg=^i|MqW>_w;OjXyB)Iq|-ar
zAJ3hqKGDa{Q)g#Zz99ravc3b^Q<Af%h*24%z+?S$_{vYy@8<Di)(CDVySZ%+kAGrM
z7<*VxcgZ3GJ5D{qziH&(FiSxb^9a@DvfwJcLUjS=8%k`ceD6^5gB{M!<{EcqPpTVT
zL^~!UJ=1orr7v_Kyvsci`c$RMwC7aM=vWV2;y)#w)5}+7#;zD~`vwp8bS+VL(bQ5?
zCf->etuCvs=w9DFGMquDQTWKot@VuG5rIaNDdD(AmNI#dDB+afZC+kwD2AgPs4Ce6
z{;zWVI=TK)fEzg+SLJwqJ9&A1UVBRA@-EV%sQ(4Iyq4tjEvTO=S|7xGA_M#o_3k`<
zpL@I%oh-Sz^jFQ35#jo2;2Bzr)=z_+&&P{dWMpwLN@HYk)$FIMrYsmr+Ljcc;>rny
zI;DhYPDlyK$x$jBc8I0*rqtd(uLlSxxtMlk@0lEojaCmCQ$D-fT3S=d<iy(H&W1LR
zg^7r%8erYNkxMe^!=rxR&~j(c)H@aS_SD82EcMlvGD%#_mi3NXd@+YkIgUS(@vK&i
z%dqMpAplZWi@---8ef@C@<y3{Z>8gjzl#<8?t<u}vR5gWy%gh2Wv@~$d$sU!zCg#h
z4&HGTr*eS1YFUlSEQ@@)7wG{;miT}%K0Vi(1)h*o%};9+I71X@OazWyo~V;cyxdD{
zEKH!6xX6=iENZWq?Oy74OG9*mjYW2yG%&okr2Qj@d()SWu)&y>unv8u=e*OoS<dDr
z$DIsU87St;^WZqA^t61{pA~RA2N_(qF3I1`Ts}4w=z1JZC1x%CHU~G#<*WWn0jKpd
zxb6>3HnQ=v(i>4jt0~B#&h1J?M;&e+zM-6~(h6q6*iWd{1S=StXSF;Mv?L;#-UzqC
z<DnBUxb*u|NDOx7*ipj|P0odY79XD6+k5;2A1F9Lc@9^jPZb}cys@!y9{a5NJm8{&
z(y3u%zwkN01v`UXebMu{H$QXnFQ3PKDL+kpQ93{!#)Q6)@q18Gv_#}%$lx%x6ar)J
zEP$&F#pM}a&xgbO4v3ilv69d1$^7^7`t^DBmvT7rZ$)|J-@Lrm$@urBz`y&&*U`yK
zcrR|(O5U`FJP(+y&OY9pe;2q3GqFc6pEty0a6R8&7pVAS;NQ8a#WqMSwle2t>F;n1
z@-+GUpv8aTdVKahbeNP1{Ac;`@;T+5ji6F*SH*LtT^V0Lk@2Ob_)-gF8y&EkzD~XQ
z(SPUzNbx~#qQ@>_ypW#zQ|OJ#(<HW%lZzI`91fK`B!;{UgXI)Y78;9OZe(7_7RgHt
zutlY2+}g;;k}L18eaDX<f8luF&{Ii{AX3I3MD_ZgSHx@9|FLUa70Mw-Y5i6UdjJ<B
zDjUTG+P`l?`CYPqOJ~wna;tPD9quhPwj$&xyIz2t93DUzK96Sh!ak>Lx(u=D<o=!5
zyL{d6_by+n>{j@>)=xO$Vmx%;MhAIT#DjlmOr@S~qHW1HIQ3T*^9>qPiL{FCIP}_Y
z$*%sPmsZ63MSLgF6~1Sk&f7{#S7ozo-U3_<Q+~nV1`c1$u8Tv!+XJcq>7lETiMDq)
z^=VXJ^enz2K9>XhHD~dmszJ}<v+_fEN!8SqS8C_(y>;zZxBZ{7$3*q-mfrtz_FJ^Q
z6VR>-)wh9PRY5a#RO|m7pcB<S7HKWW<znJ%z^lweOZ!#D;LzU+-rr^g91oM*M)Ues
z{R#P+PWXNn^Xg(f<xt)vJc#^0|16@ST_)wZvAO*t9S>rV@)u$@L%sZ$J|eltib&~s
zVGbC(td{p+)$$&!R8HVqF{wE2wNkpEJFQmN#AEM$?_<P%&prPD>&dyud@J!SW<Lh^
zAG393j;?QXT@eMEPFGOCslJd82fX1K_K3Gufqx^Fr}CtQ@iCF{eSzc62E6xnJjJA*
zXZkVH-SjQzX;F3*zE6_Fan2KV6Vehvt|(#Bfr!#UiIHiyIDw0W0*?xG{3FLSC3a7w
zaboYW%Pawh`PlODt{zcMPx?NzoPC<~oYLFew(KL~YhPn?c@E|h;-Vg7mDN16QGsvy
zMb0NL%9g@ieH2k$7J2;9kJG|BRDrR|7*Wj|Z_4!unZcKT%+Dy7gq*&iudR^BxlF@5
z1ai&^$|tD4t+1O}`8YE;%97SRSyhH}Sv9XblPe0~qX0MZdQh%tR@7sY>(}SiUn-Y(
z<(0QW{iU0{7vv3<HV3Enr!_GywgZ8<gl6<Xv&fbg3tu(aGA3+AaJg+zmdL?E?FDXb
zQYxXARGfp-Q#CKch|NyRHOk*Q0b-#~LHQ}fzlG6E`I)|mX+^lDFQ79sYXW@9%?kmY
zgZPGG?#$lKxz2Qt{PeA!QI%%RnV>ggST#3v9PM^ZlP&TS`R(}FHj~Sct;^oapDM}$
ze-ux2ooHg~L>^v$B9D(+VDlN_t>U%xtjMd8eMzTU>8anLgw3LD89rM#r+%xfVyodt
zJVQ!^>3lWknetZ4Ofl%+U!8A^vBeVw!MUx@$N-|Ib8p6t_ZI(-kMpox@!Cs0;o+MQ
zDWIpN@oqS|{)6$c>9KM8!Ta8tn-e6gpk?Sg;3`~1%d|X&=hJd4m65|mwA?EG80y#j
zcYggB#38PGH%rtk%^`Wt<;^7ey@k#VrhDiA?t<u5eE{I`BDjm{-mIns{Z=KE^L>wU
zpye)M6-{T|LJu7czY(+bvU|?j;=wWWC<c}Ls}~J>j>qLZmXP1$o`;97j3W1BucA*j
zc^{+|ATjR4*{=~E2EfyBVG5{fx`BabjC)M%*fSju`cLJNgD=4!&A45ml}wqNZD>5}
z2?RWFkwZ;>f78xjzfIjVJGEoS^rBZ)-QP5^{f%(Y=`Nkn)%aqW-l5xvdg;f11Uh?r
zI|JEo2K-ZK-<)r0y5VGxbznz~ZCe1&C-FQ|MpHo?2|FJ<2B+(6J{;EBD9By7&QdKO
zzoEMS1<|PB*EPI6gVTC)aC)9IdO;Y4G0ZqnEsP7>6l)YeuPQfOJqYMw!`09BEBb+n
z^<!>SYFEYpJKcF4I?N8x!yW1IXscGv@LDOwP#apKupd+SEMY<EL@3Nr$Gq3$@aAe(
zv=9E6g$v!Y!%#z<dy*DzcF2Fy@E_IY;Xlg#5|U_17(#UOqdr_yK61o0=$`LN^c$)h
zj5{3Zw#FMS89VCfwQTD>9&zqy>}_kj>R_o_Z!L4+&dGNbdm#gw_NE?R!0R_wSCok5
z>MB=z)69;rzt0vJcGalMHAVnT?9${mQk3J8TQBADg7GzSIVvoF!(@tHke8@eQyuqe
zDv-A?l0%LQo;l4<>C38?WzYtcR=K%T%7~x`D6@Q{Da9g7t&QBMBDpkKb=295W^9*E
zgG+f#bA7|_f1J$y-SV+}t{+HGPtNW#Kky&o!LIRzcklUO*@3-BuUhEnwfDpqzLVX7
zeu99PG4c06Z1<xc#CH3)%#~N<#iTrXponsFPD|`8f>N4aw3H`Si1UpOoDOjc!<E>`
z#nMu`Gg>P9aY;tpyv)9ZnSfc$HA{$S^2}s1j?NwJ@f{tGc&y9otQ+dxzmQsJ?5|EZ
zJuNl<dPkebymMQ!w>NnOd3;gbzR>G(^(?ec)CU~(`ZD`K_spEv-&1dI4Al6`%gXGo
z7U$qLZRyD9=*as2hMqO?v$yMnJ28xhNz_}O7FA>_RUVOtsi2tpp^He4pL*HTOV2%E
z%Z#}bYO2?mx>h5bRH(BNJ=B@IJ0e-+f)wMT@#{l<hYvq=vcZ1Y`STz9*!l6UOJ}cH
zv^h?HZt<Ggd)a)TeNv%K$t?T~x}=4UQ8=9^t8|{6r1Rt=a3GNxm_I(c@8}gy74AA3
z3iXYPY%cOW{oG7aX4f%uJ#Dj%MHalsV$1T<2$?+c0ct#f?ac8!Uyxs<2M5!9w@iF3
z`weNcEc+_4QXda_w~XOz2Ii2Otpyv1fjU76N7=oO$?h8kaFwCBJd@oE;G+On^LkL?
z2<E?+*RRj3zf>;o$}4Y`%a<z4)4d)B*KV*iYX;V=1`#7_Ze67zBy9@%&Yh^j0j&sV
zz1X&}k_T?#Vpj`QGKN?iVUGPx7jdG)U1D^8pmBpdfKXKXTkN_Tv;?;3+a-rNN4}-`
z_+~D0Tp=Tq*{^|;=CYAx8D}vV6XPah^%EIqY8hu2#95Hb8-!a>*U9%A=R6^Yq9Qjc
zrIT<0wX?L92H9eP`v*qunZm4vY^}iTTlMELe~#8;+6+U9xH#L`ay%a+dzZyZUntvG
zjD(YNe`j+vPIf40bC$~DJBnzNmDd-PM|dgl3rdyj8wt*2@i&#_G5^|-Le6T)gXh-s
zJh(#tZ78gVLUOu60QgiJg*P(2izwk*8E#O*sb3z0m*m0K9FBPS!KU@=^Xe~^%S-#N
zte&69<)!_=DQLf5fd{9cgy-bVJUHDua?*Nq3Oo#}z6@{|h4<vapO)c$TfozK@G~;J
zOV)u##bI%k!*jY>;d^`Y>UmZ!Ka^K~MK0e#<!C&sF(U4#=cIbj{_K=vmGjR@Ejq`f
z$ltdpqB|$D;j(ME!17bDol86)wt?r&VHt7Ear5RZDd5q0h({HF^+PWed+v6SK;ox~
zt4VFi!gJ#4`p9N%XmwQ_&weL|gYz;D29!8cUn#=}^Kf~G3{U2jAC+-DEa4c(N*-H3
z9rHYWw%~Uo%CS**$?(xUc#8}l%Y%>0@X0OSpWgKSx!+Ca)z>S-XY$}F8NQv?B;rYf
z^z7?OkQ46CWjkei6uEenqWo?}CkE#BD9~d^!;exjMWjQrS<p^772>u0TYW@2kgI}j
z<CI@ua+{fY^yoZhmn5HAU3wBXk<jg}SFMthkKBL%Bia9|n_oG1PF&sf!(BhyC0?2R
z1la@qr2WCUL~dqtnC~5Q(LH{gV;%;#D&c2kc*e@ssxFvIKJf)m`>sX>pLxzp*hpeu
zlf}8NC^nKb9;2LFMoak<8i}Z`ib2OUAU4!cU9}Z-)u!`kS&|Nqh|*&P*x95~+}&St
zuanY#sYqtqkjb6&If4#L$)yHQ`)D+$D%xAy6?*20W2fI0pGf))mzG!@LGkFx1J^~T
zl8!1z$nN%}TkAJ>w;MG6<3mq;zAlpSy6quHl@f<f%Q#Hu;r2bie><(ne?6RbVf_+=
z6z5)s^M7aj%Y<-omF_0Kh;&jXyE$P#mo2jvDg9^Ea=fDecfRuwp&_UIV7oQ;P)?aY
z9(RXR8^1jO*9tc1YAE%pAV^0_@ZUWxJs72jTj_(eM_$e2ZCJh(6kW=>h9-wrrQ~!U
zNYsm?qyK#7^yxGI{MD(6iK)3;ZzUu1r16=@9(!irkptHpIP${me=vM;yog$IJN<is
zu4_okK1R!GK>>eShWF*c&&u#jVR=<rjpO4Jl<+fh`L4Y3bUgFR1FzE4b<M}QR5ZVx
zs@_fNe^xF(lvjR5es71G&BX;e7xD1n8!`JJTh+E)&A7oY$C<cSVwL1^sp4tfRMaNN
zv`RIdrfB9X;ZH0%Newmkp-(;VBI&<P7dO`e`E2$a>CC=JhBsT>vJZ-@%imi5!SZJ7
zni4N}$o)?$@IFR)y^pUE(H!38@yo+L{oaoCsmDQ{oTe~30=&K^buYkYir{*>M)>#|
z5f!97Jk9j1@DOJH%bg>c+t23=aPu?n>HeE6jay_^FUXm~wA9N<I#{xoxDw1?Q|au)
z{5V~B<Q{1rIGAyeyDL2%!M1Vhe%lQPhYw{Oj?BTqK*C*_{Z*waX=vMbX2ETbOs5=O
z+X8#{w=JESTR5}Sw$G8Aj_P#L=_Ea47SBPq&^Z%ebB2$F3|FO9h@k+-X#{k73*aJW
z6ALKH)47o=54bC#IaN`~w+XN3GH|d6HQQg2Wv6JDI_Fcu_I}tdH9-sXegK?+i=JD6
zCk0S;wSf;yG?f;IM5G6Se#KI&t4#8#2A8u_4e+=WHBVJhdO^zHnSZi@nwwn+%?pAf
zNoZJMLW;hmzPP4@$4Val^-5pg3i}}O?6*ZSfB49e!}IGK<b7{@8{l$`+F5W+n)2jo
zVA5g<<$FOvDW5a=(=xno3wVA#XJmL+9{fEryi<b1JS(C6Ehs4E_QLk`=GF78Tz)98
z{EA$j$?XK{kI-{^Ji@VJ{#LZwU6j9hjK$7dI=(QK<nc4+%0*onHs@!vk3LGyKKke;
zA$DT+fqBt5KfnI_`OTS`+}Atgz9#v67=?MF;qze{<xOcmROap46!DXyd`im0aw{3Z
znPl+@Lwge@7j#k7IJ<VnmsgBesf~4k`u48u^D8UjYE~DSOf)Rb)LMNFrrv(?+w2M(
zJ8a(+l^;J1fPkuJV;>#E9z1u3*~-c5bj(AH291IO{<I8FbNDpyiz#jc9HjgrOyO4o
zGHRh7e8Y_t-kgIoUl8Gp5)Sp$@_Gh}$|G$y3(ObfZ?gY@XM->}aN75wK(%RJ^v$x^
z<Yy*g(e&)EvIz6~D9iKukWzO@{vV~(R$U@L>E&M#p5X^m@R`s9Bu@Q5Tw-l@*W~2N
zH@-nSiMnQ24cVXl%E02l3k-jZ7oZHmJbqdQB^+g%RwmO(aD~maS^9-s9_W!$BCiL!
z%zW4|@y5+e39=~gRFq|+NDV5o;}u2nanp7cOA&#7q0D;CPW~fCmIajNb>bddXrNYk
zOk?+cyYx`s(9btQ-!27x_*Y`hI_Vl;xEk)iVtu9SvH~{|K?z4%jO6h{8BRH2Vfiz1
z`7Rkgit9dy=lEXky}g^%^DN+gh~qw@fPWFncW+XDMSd^ihai@1d<%5a-vLka2rcLr
zA;?)#oSc{CVseUdSSg1Dvr0LsxQAXoPD+s){&bcHOqNrNcsbu?sgj=>@H_hCKvjW0
zh2^Ja{w+!@EXX3~9KkXS&{UgryM>$71FCA$yeru@k*Ti*QDq?a_$`&@C+@C^^j#Ba
zd@=-ggfDNsb^H7s``X~~XzMN87tZW!BdzwZh1}pfc_8~dXlw9C_6{H7k>ZQ;tS{%S
ziwwVa$oQ3bFv3$bcWe-~TZ5eY3jJK`x6WCfJ6vrO#S7yUJQSH~_>o8UkEo_$WlSZi
zXM7rooZ%?XK_Z@?rd>=mHdll_h$)oz@ZRd*vy}a>iFA4bF7X9VbLhPX5i=<2A+ii7
z*Zn6CUVGpqpy$9z83(D{93fv*&^F)zaQ3w#+D_$YTTve2&&uV8^2)yp%BS8eEI&kL
z@5zF>Fshh?c{5A)lW68Ld(j-?GJ7#RCC^dGu2Tyi-6A@()e1+@B2y=CoGNXB7s+Za
z9RY{(dFdLiBxWwfmq*YFC`s_O%zy(-%*DJGaz-1UNeUAr^|#$}%Wb5QS#Bz(;2$cw
z)ONE=G#op;a`+hS&m8qN!1Xo8Xjv;L;AdrcMuJlx1jg?fXOD4PG5VEYJ@LFb!Qf|Q
zc!sY70g%DGAlDCLj@oKrL>N`913pBGPk|*RmrpahPFiS$D*{)s4?-n%Ehy(n5LlY+
z2Dr17G%=_;=O^5`c2Tbaf1y)>V#_zrtDrs4+B0*NcdTho=fG%tW->MuGT*wg^IGpn
z!%SDYKi<D>dEwft$~JW==+|VT!_k(Wj+QpPp~M%@crH1Th>m!|iFmk8Uv4_GZRu)V
zq`;qm;rMA8$CBI~fU)v&-N{eKDo$~aXDFbI<Puge0M1Rh$Cmnik>yc&qICt|>RF{*
z85Zt7z*gM`BXc8QWZTHhGcFK8q{ulydbPq|z-97ukl)<bG}_)+pPX%v?Q-_C_C?#;
z+xC7a`v;IAFZ;ZP)Wjy@k(rK0r+=)yyVcv8T>i@X-z)PjRtg7k@nI*1ZJ8@yX*{K;
z?wF*EHUiJyqLnk~GXsTTO*%6M1z#y&9-DgP@XEsv|5)_@_+e^}=$!w~^741)*%;7v
zQTlJslbdQ_Y%+OrQ~TC%Xe!P}FSXTK(CfNF_I0nImXo)Tn-_aZ^1O@%E@`4!m{|x^
zMwSzg3P@HtLs5d~Cz@ON0Olt8z->p5-X^Zzv2yUh%KLw}9LBxLVW259o&Slv`On}_
z%kaKJIG2qQdGIrG`7RkgDw?U>oyc21<o9yfD8HU(<?=&$<#V!8e)$#oy^^jt415aS
zE41OMJKBO=+C0?zB31-tNWXI5nTn@j*+g$fJ%w)_p>9bHU?y&n=9y@9^t|XowNXo-
zOrO=*qoGjLu36DmHwR+rLJiVvAK%kv0zGVJc%-eiwrwO#p2!|a{J*_@34B}CmG`^v
zY4I+}vaHS8EZg!TTef9c-j~?%lEhh@%}H!0glr_3I1otKp%iG@O4)|e7HCU5?SwWd
z1%?UmgXsWe3Z-W0ZlL{!ZcLf!*QOnYA@X<5z4tvm=}C4{n2+%Ep5Bw+-Ojydx&L#{
z`oZ?HiYAwpDU^9upRF?H(&<90W>=8Usrm)?FqdWMuN#&AYE}B{lsG|SQ%!39!T844
zBK?*0Xl7l*WTc{b?4km$xmqKsofC^9n!+NBH_5)3m@2EWG;bRn-PWAe^h106^@(q+
zJhZZ^3O=x@a+@Ypvdg<F+MA%V%exxdn^4Iv?+WVosN`~dqTEl)F7of$8j;j1;wkr&
zb0qaX&XIV^97(>XJLUHUaDNq@o9!uclifd|+?V-A0kn4&oonqW?Pb5eLHT`0miKq3
z-2aa9p23v&+^GCs-WzO0J1KonM@l`}{U?<By(#yp4Uo3CgVP}5<#Zy(o&5x=AE{U<
zzCvn1$7R_Wp2odQ?O3Yk5FZ;|Os_y$#evS_db+QN`C*f22L%Fw*_b!X?6XySDJI|M
zuXX#vX1(P_@#VyeS2L``=?^DLD$2~IuJf*g^}kkNaS}1#;wY{}dEA|yxydZ0LGw6`
z%s8dzT&=MwZjD-P0TPq~wU6KX_~{RcA@^qDJxAPf`tQ#XS0=2iKhgiCnC3lQfhDKr
zJ=MqyJlt5a`%)~aW~>d91XvujA@cI<A7*g_TAV!n5G}78#?MA=Dcg`~otZ6#rpV35
zPzQBv154r{H64#pm#ZWuU{(=*AZCq8z`_Gon-majiRuqNAU{of={LV2+h5#CrgUU#
z=XBz`D-z$uI@LiR)PpQlgV~?GT!YIhU#MerYc?!v6q_NM<}He#;blV-ja4PnDx<$n
zFxDd_^)#<&w*~8NzRue{-PbqO<Gt?Yw@ND<g<V5KU4@RI{jCjo!@nOrytb}x?cvei
z59e*TKCim9*S~M$^z_Dk{@&K=yz2!ymtZ^05i_pS1&B3JwMPT3O!}&(qKDF;l{BpU
z9BYk1223)WC&m51aPM3#R<2&%y1ef=27sGLJUp?Yozx^Y;CyCryv(8j<5(@ov54@y
zUXcC3z2d(E2dPItaDUQDk)mU1qK2q-tfVPOptcHz`SkP2R2o6-Gs7f@sMSLD1nm)f
ziOpCf8KdeQl9T`s_3<se<^G|Wu`hl;(pzoQ8LT7iD65wDZfR>D9_;GwUtUnP>Y}lW
zzj|JaK6g*9<ZRv8+uT~-zoWbB+~rl9Mwbl@4KHtBzkEc%WK@D!LdOSTXqO<cQ${Vi
z&CKiBy|`Cna)@LcVg+ic^c58mx}92~#E8auUBUUQV3yOKSpz5m${`$sfl6gRHMz68
zo!ACeoY&W7tM05!{EiTRMbLBaO>V#6eUqpg?%ZM4>wBZV)*8pS&*Sj;Kl^j9tJLMC
zxv{T>j}JwRGfKma@;sQ?+LHoDQ#>AVo6r)xmRz3r(~PK_074Q4%fex?2i{x6GG`U@
z%vmhPuPz;9ktr^OJVA|fvH-$%Vf4y`RWeFPF-su22UAK14IG1BS~@-DWXJlQiND)P
z4t~b%nrOIg>z5K6-nr$LcgTjZiL2RuB9ChzvPc@tm%bnB9S|Nx<t$Y%)37OxTQoP>
zFXCva*4)w{)3~Q4@mWCkv~WM`YPzR|$^pyTQOFer%N$08E~dV1jIH=+a&2QU-b3Os
z(3yEol6cfeBY^ryFch=}6DvA*Hs|`>_2#DS9mA8u9ow7C^=@Bo^Ulr{lW)KMbf7}7
zFKmD0b@=a*_CmeBBJlLvZ!;d$4t*twW>|;!AmA!YQ}d+Bm6_n4GP>OiXNI3LGuk&(
z+BYcy5}9*WCEPcqc^Yv2G@4BD8Im!)Y-$*Ycf)L8$hES)tE0N+x}&wB=8o1O*FfcY
z8}^+ai$pt*7PK_<R8&^`^Zey&1KvDOMKI9X7#r(t=*ZXQ^hQTE(>AV$@z+zo+9|DH
zZMONKagWU8{K&YW-pumUTB9~2n!kgwcu)$iKR-|Z&R+T<>fXAA>_7eCTel?c0s5r%
zLENFpVpEv7d^HL8{ZLPba05o;aMGwrS?j5B2v#TJP}?~v6_(;II$w(^6_x@OmQpGd
z=?Gzp+)~9Qp-A!H1X9EFK}c(q+e8Qu+~myE6&GJ*@!K4JOMhEmAX2hv3YLkvxu-l1
zy<YG1UOeFo-Z#&3Pmlz(sYKY2jHH@1)1sQ4(;MoL1SKi)Q43hM<S6oRD)J!NU}$~E
zK%uSBNM>g4s|wh=0u>XM`o0ZqU&dgstFF7N++m`94YIE2q<mlo#hX+<y>wq|0=cTC
zGjzAjpJ%S}Iu5`PxQ)XKGZ?olEpbdJWVZx+ubwr|9^S4q-h8uBw_Viz`Oi?(;I_o;
zq<&lCO(u5>0+qY-(aN2dmY;a(x@vr1lXYco@r*8V8C~Q8UF0fs!DxVnYJg$iF!SRX
z_#X%Q^oR5FxYjb?MO<rb0?Wybd9QVg%CwqX?~BQ_xhk1<Bl*_M8#CzekAZvXTF-nC
zu>4Z`oyADQ??zgmJ^^DFV|}`{_0h;<>iQ(bFa4Y%a0UPJ82RbrOvKv2Y=jIGY@k?P
zNvMGDEg;Y4gZ|zxypA?AIeL<4X67zD_swP1-8Fb|ZnaI`h~bTxN^jCoV=_?`Qlrdu
zU1nLW!ga^YcZXa&XDga?yXn(%D6QMV0)fdPJDcGg;SPVWTyxpX9#jWlDrL+`9oHw-
z6$Ad{`Sx~yATXgf*L%G6KC{_Z@AcH1_0yUg_rTBm?W3c=9Bo^@XLXw?<a71z8ynl#
z>+*#@{80TZksucO8~81y0gSyLreoq2*xu-2v5(H5aa8nD=Z}U4G~Ox6O=We08=A*`
zo^Aq1**{y5yp`BL^U}=vouV$W0ZH^7S<fTyL-|(WFf%mT7Fz=~4g_j|Oac6q#gHqA
z(9{wIEFp^tE_`s{yk$=8m~mc52aLs(LpLEFZy1x9-K=*Fw|-+L(iZKRnR&+_IQIDQ
z`Wm12#nT@$UKAEEe*yN{Hf>JyrPmydvz$87Jq!77U=cp#e}VASVfdFI(06Oe2|VPw
z)WY~pKwV720p#DQHt40tPZQoR^MJ}3i*R5~DFe#}S0ss=sSDs==jLP;6YG8@P#p^M
zZ$_bx1+WgCj}y!;I4p9^@T6?4sjYhJQC@M7TZ<g?Nx8Lvk()TTW)&_lGwG25GZl++
zToJ48y7AN}*6!G`_CIG1R@K*69fTF@AGWSvzjb3pusn!xRamRSY@T4W1K`uQp!J){
z9L)NiJIBtB?7EP~^;{D|NG*d{5!K91jkHxj)NshWhCwh&6t1HIn<C+c&mCQJ&N*uy
z{%TcSUDYzuaqhXIjy7yOSXEIKe4IRjR3i$zxSoe?+ld@rw(IB9mSt^GvH^cveS?8y
zeRs{=i9@g;yoW<@4X-F0ffF#&qrj_*K=v*dp2BsBWuwj-(;^wcOU}G5U?@Eq8H^c_
z1mQRm;~|=KS-_M@$k7Gh=z^SLNuYd+{sezd7Z2QV?PWi~R7V^Ye2`t}g&)Gj+w?J`
zJH4Xjm(zf5)@Y`PZb>OW__G&=dYii}<F4J~?OS`x1HIci*6ntU756qRTmAwGi@mFA
zE5ogWa*Tk%c%-U+RWGwm;J$E^h}wE1#^Mhv*^R;}K8BizXqxCa8iY$ytWOFCJOc1R
zxN$W`0G^UVT*=;-OggR-ayi5kwJ$j)h#KBWVV~Udkla|^U1yiZw{KrBIcvK@wS9HA
z(%Qb-Vz(tX*Wxbb=K_0t>)<a%-5Y(IT1s6#`&O^s*W+4W;EmPps*QOIlnb`rba0;H
zfI+wnNmb5eZl%sEa=&n!t0uHdretm`$UR$t0`Ny{{R%{67fb`(wer<LjuAnVTgPJ@
zH4;EEkqY5*T-{F6(OFsGxA(1e)c1ynFmuj8Nn_OjS$B0+O-<F+6N7e}sMC)ORd&_c
zPk;FAWmSH1Jm_=zf{L#P;>D@2he-XM%IRKRDcT=W*Vz2*enBa5DLHVOk#ea5^bk1E
zgQ4y@Oj<jo$H?arKe=zmfde}de?8sM($a9NcyjuQ3wG_gU^3F!&=~oH!1cQdTJLho
zzb+LpF3-tD?J=1x%W`dHd(5TeuEhK-6nXj~<T(rb#Q}<quw(pItAata(lB>v5=N$-
zXW6MmUh2*B6kls*>%w-pI0y$_Ek48UO6XukQ(YCzF-5GH+kX-Em{iEP-G)%LELs>Z
z@!Jc?)XKey&+Q{uek$bu;k0|?+^g1o@AN~b?zu<&(<S#k8BJohY2Q`Sz7yg5CW<M2
z6hBfirAz0O)!;8Rwi^u$HR(mhY&Qj&@W|MIEU`@yx82C8J_B>OKUYlm=a4%~K`Waf
z5yGHx@X0l5J3h4p|GQ)Sz+hp8$5o!+vu8!eo{klJdh&x#PetM2f$@$#B$jxI2+g$y
zgP}6~KZzfaIz0X-TnRU7n->Ip&g3M-D-?t7(TH}ALcUVUedWbwd9&nBH*oZrft#L8
z5!OJ=dChRp7@)*N-JQ4zr*u&Z4ko5ya0KGqbV=37@b0eVoxSHwb;LWmV#BVH>hsqf
zyx3M5_4wm8j$q$ZL0{9*mX4?;5}R&mHrGe11KrX1?oE~PkVCJRN<;0T$XMKm`=CG%
zR=S6g-IlWm&QYUqX1C>7m@UVGww&b`6d6zNryt_U#EZ|9)2D|>+w%#A5vFy*{XUUK
zW0m^3JqNLx&EiRB&yh0QbCOll0HhaLsx|hU?Qr9uV$Uh0_MB1<zhL$px`?o}o9Xr(
zOK@cN{Od2e#1bfV`Yfxv2P>OPb{s%-@sCMu;{7VW&Y<_#JU(4j|90XPiqRM7K2j;#
zbVR-ir$yJ+#6-vxij<|$WJ^%KvVnwa2HloN(&FYF^3?|m9KljoSJ*pP;<e_H%Vx#N
zD|X(#cc9d1n=a{^yngxzr-xuH+Ncc;*N^%nUY35ZPU6hO7et2MB*F@KUi81U2e*Pe
zH~Wri502x%s%U*0lfGrNKEdQ(GkxAd>*t5>T{Lg+#7P=BPaO-7m+noMdb}A(hB@;w
z=rfD;Dn{gBsBtN3wMXMf8d4p(>R{RLk6vCQR-Je7tpay>*gG~oEZP!3IdUlpoqoh$
z>&Qc5UicN%^C!S9J+M-L9t{aqLWtp%cmxZN94QmMWvK7k4g_H{Uk-lJC|?Nlf^EAR
z=PAE2DmQYu0nN(Kws~r5w(%dxpfnr+=D~{zPZ6+LIVWcV#H46!K{DEu%zt!`p`|Kx
z>frh3ADnF)Z*{v{$J=HNy=`s1yYQne_ls9u^~K!HoAu#U`+9r#t*X~;7FQV84eT3O
zXXGcgXKsaY8lvOYM#pU#@aBFI6_3lZ+8i`a0GWX7vQJmi=0H0qpLN@UeBomkxDw>U
zN{|msJIyxnv&4C`Z_KXT2jut=lH(~_&t}GNVG>Uvk)`I<)ntj)u+wDBDHh+}&b?)+
zdSRhk0XjaZnWc!DArNAMxcMOYyxz5}wrfMHW46%OT-nz;J4=lD1)lEd{#RcMcUG8e
zWb^4E%G(6Y?|hQkBc#msh(*>?jbBo#J)$s!JtDN{eE4_zuh3r)iaY2rB=gz*;yQ|n
zyECgjVv(_3B=@d*uv6_3bkI%mpqu1DH_3x;f<ZT>4LTqA|4KBTyB+=wY*jVo!9OxR
z0@w19q)c6^$ke5o>=v12YEfClQ)Ox~U8c_LLwe6`i6bac|Mur7Q12HfrxT}U_wG3a
z5_MZ+<gd(DA&)<^Ra~BKt1xf@)|Ooh(7?cxXcLX@&(M0fSjrxMVuWFcc^>ohHt3<R
zLwh=>_UxJZ(!JP>X410<80e`(d-fdK+R)V1)bIgzKSDmJ-*AYxW!;jt2@T<#Y5mlC
z%lK@X(vsT6gkl1`AUjqG>#;^G61-d!zQ37ScFd<WvA{WCNhY^IY6x&@;HolnXJV0y
zSYRFy7bR9#;z;^%dshs;_S)bI?)!}?Sg*hK+5*K_`_K-{v~kBF>fnv3SD*UkH=km(
zjPh>`+JLr#YtwB7NgP8qCaGC1G)N^Sl5P&S6|mOJ=7B=RJfO50#Q=;=w3vC|y88Y)
zyEwRK&!A)v_k|+^;gXW@K&04RoSR#W<^g&x@QhrsJ@GP(+>a)Y4ww1Y-E#i<x2*GT
zEv#4;dnLB4qENY@eNh1)c_PDLT!zCV=WMi?)v^6G8kAZ%qc_wL@lvxaMWC&j(X20m
zeOE~B0~Wp&TtN4I3sda_tx+l!rI`mRx`#{w$Hc_`bBHiIy|1(>I7BYFE)tDKuA9Eq
z>juel*T(5tkhZ<Ah6Cbsq%qVONn%<u<swYYU1;L>v*%h%&CO=w_jz6_k!Cl7?-Wyk
zu~|eKb4JSJ7%qmjxE+L*{<Gchf`Jg{pGeu}0wIN>E}T0l?BbSp5*v7)S~K0hKBD!3
z)0p1X8ci|^9a^Z3e5C+}aU5<u1Bb}$f1=P!zsNgYqFE(yPzVCdS!xln1#Q8*ckM!4
z$noPdb8Fw3xMl4e{MGh_Hqiog^kSSefoF@&YK2nQQLUY-f^QdbNC=X!h38H#WmPYQ
zs+Y=DFNLZzy(fnQj94^cn1EcoqqG*ZP)s=wLy3zwsoz#;inK?&XUI9H^seP?8#j!v
z9-KAB>Nj;qeBK*G-Nb8wify~j+p}dMLv9sOP+?8xmHRpB`>Nb9CXvw#-=AbQUrU<J
zSF=;0Hd^MXFfwu+K^vTzSE7wsfNjt{HrWDflP$nD;cC3gEWkDx9QqTq?AnC4<tGui
z=w)Bf3T%_D!2IF?DsJPvP9|Vsn|^W@Eu@WPn(^L_ZM}g&@3!_eJ00Vu_DJv03ok5R
z*3?@(=GtALn0i~om9?vS#dZ4LRrOVo_#o{&hSz~U6<N*=oN>y1ikrw!!A%q{92U0$
z?Nwr~YaQ1^vRZ~Udls4mavD^gCSsS`&lKhtSE!84c5H;+Hd1T3#7$&Y9#&f3yR*-s
zz9eyJnV4CqW0p_1avmf6g=os(MQ}g3A!TUlTrrsi`s420`1acwp+Eb6g?-6zKiJ1C
z??}%3G9KtK#RD}UHqHk`(Rg-vAd7+rvM@XlQv!-KcpwXporB9@nCo@*(004xB(yX>
z2voU$@7-Ls2tsIjVrQ<gxoR<xP~ZC8>sZ_6xbykKN|ftUbqq~BP_+h;-BmrsK|QeY
zAO3-S_t>#S&xF`AF>(6k2|>0O>*-h&rqDaX3E7qV`Re=V&&ux`_<f2WqQB>P2zb1@
z^)RL(8CR5UhtljnYXEWTl=K>D2$2;AF6UF~*8%nG=z5&X0b;qT6hJl>>5rj^EE;$b
zdA`}dX=+A|E*jZs?7VM21zYqE$M3X4o63b_xLc9Bj%s+KW;0D=22MIiFsrXFGh&aq
zFXiym-T-MpmcMzNMf&HM_RHLHojd?9dsaSA<4%;rG(bMeS{>5@t1!$iJpD=Yo;}TD
zH5bjCXAasO<wc$Co#rxUY5uCpv3cx;i!Yvg+GB@)aF<;?VKBQ(@1CV&oI~wI&}QLw
zv_GfD$6tDnT$9OFo#`dHFYUs&SXk&r+A}Wb8J7xmkPl}U--A%H00gr)_x4ZyW&EN;
zN?ZF_TgMP!Gnd{_Ob;ZGh123{+RiH6^ROs6t5B}3O5gKXq=D*vuQCpiVe1kTlPcMk
zF9H`U6>?54=jx3>zUQIrSX-C@6RwiLUeZ)xEi@2w!<wn7H8*d~D|T6`dd=jERkgKM
zwHquA15?XBL$`;?iJq}TYg)_|>nlU$p~{mg{E7*)I^?I>8<xgmWOL=Gw2mu?iEW~>
z5iNufEmZAs=POQoR6Ctl5Mj8(Ls4NP6!J3t62*lf7<Yc2>0^j5z_BWZj{hq1KM&tt
zC`$3qUb+Q~<c-G#rkLx%X4V2Tw7{%FHoD=LY&yHyn7Rebd6t~cMO>beo2CpTwk|n@
z;i;RZnsN9(bJG;L5C^V7)V=l^9ksXKTFZxvV`uuQynY4R;g`UhE-JvM$s;)Yg-Q++
z+S+_*ZN6%}WRagYhJx0x9pSzAlG5XQ@7*ivUVU{_V*JfFg(QA&UdcT`c@M0A9M^}q
zXINbKp-eAFVJzXy%`<lfa;xmd>W4FvRQ*gY@GBU3Cfa(GJ@urQ26qL(q<&Ul|A7WV
z!g8Jg`P@64I5phcGfYSU8CC>K`MiE)r6jE!A(sq~jt-~b6~$d#9zRUwaUa?}Olf{~
z8WfXS9>;jyCPf}ERnY4m-aS5;DtbAbLWW$kAiNOM^j;K^(^AUX#=3Uq@O-c-$z$vL
z@;<3x4qs5>hZ@c#ZYhlk%<kuAiGOM$OkQY&`_&ZtMMEj2%0U`GuWSNDamHTL@!rYB
zU?EZzm_aXzO19u%o<B4+Xc!!tc>$J@7vaCtAHJxri$%ByMeWr6k8C-d7UA>ipzV@^
zjFh+sm&9$ZSVI%k&<?CisU+2X|8nDvN=-lYshPEF<=U=2O8c%J=&4u4)Eb{gS&Ml{
zq*<O8@iowFFsU^Sl(kqingV!>(ZV9s6!66@$IK#F5xOTYiyt<sP5`epk)Soin_{%1
zto3HoSj^C}>d3QCTs_!ohz(xyPoG%7^Wdsg2Z<rEbM@+-A3X97qZ5W-MESHVX=_1j
zE$2&ilq-wU_6{^{)g<m#H?0VM%J$KV7_PBM#x)iJaTY0yH4E!gr07j0N<;0mp)oDi
zk>WC^LBVhR+Dv`Cxf^(u3hyXmG?#Io85hb=dLCsWtB;5#!wLuOa)y-8WgyCsfb&w!
z1l6n8aQ_G+2XL9Ww4s}0p~wvL3_fn*XDiPM42wk!qrPLi?7*SWvVr?2pV6DjOUAZu
zo+%HFl~>Qqyi;73Gr4Z#rpJ$0)#z`#P4BZKq7(N$Db^3i(4V4X(3}a>&c$*fY^3@h
z&izdv4Qrm1AK27NN)L2ng6g*$XWp3k+Rm>rwCw2*Z@GoliRI~g59(x6eV<qI_F*<f
zs_!hck?+yp(NLCxo#u#o2N>0ZI9ZrqS6W<df4W4t2u_}dKUa)WG;?Md&8y{y1LI?q
znUhJQ%v>}siZE;@lXDOkQb>0uaFdOt6H84grOlwJPA!#^Lo+qKR7xt%gB_!?jRg8Y
zlzGAi`8(#+O!1BSFG0arWs31Eq{f8q=l6M>U3mW$^gG2ip>MIsV%&Av+1kMFuU*tG
znq^--t6lU`65dDQPxqWU{SdCTP==RcK`zqDSsIUBH8ho&*-jg-q<MVdp@8aV7@*oH
zY=?`3a9j<CG$qGz&~v{v)a$JlZV;tIcoQPCo_>0EirhxR)30o#@hYJFRSRozzaR9!
zi*Y_FdTMzjl`b>KWGPo-@S1rH8$^?sxEaY2fr>r=H;%*c3>*@Ris+zAOdd7B^F3}B
zh0+pab!IqCW(Hljt+=$%+f-5B<(|bh=v~XgT^nMK>5YfSEz9ED>pCk;`FV+7izlal
zlV6BCv(uN&2uZA#Dt>EsN_^F9aaWUZUHf@lRE~jQdr!Kwy(ez2%hW&e7*=2^D+}|#
zl>5uZBWA%k*;rm-3_vXBBVtj8xz+O9>3dDrr|w0;LYubTq}q!zPP4XAoEo-StV@|k
z?EVSmKF0<vyuYwtJc98^g#W-cv&>#XDyv*%?~?Ny4f5kGY?j4Io8@-6aWKth$u=0#
zX33(@IOfnEDDmB4!b6TR-Wy`u(;<0#IwWsThve<)5WGJ`x2FRX^JZBL7ssrdp1$$6
zJ#A`e9DG%vapB>u6d1QCIPPnIRO0v))3FJ&vH26DJKV=dyj&CdfEA|2AUPwvoR{0a
z7Q@T26)nQ0$|l*wd1m-IG;#jAf6Bi&fbP7>eK|RuwM!uA#t#<WmW*jhu_`qx{Mro)
zeocuDL+}5ya=$y}{;YDpjo+tuH~M=Pr>*M)Sx($fWA8YH$5DZG32v>9wMgs|jZCbz
zKkDHL+l|#TN=BVnPvyltzA>%mil@MbhDlmZMzc<tW}Pz4I%S%5!ke6oW|2zyR&FYO
zj{L+Iw(K}CtH#fb@6%P)yq$O@9Y1#p`=SB*Xjnvj_{%t-GUl8*o9MZV*e%s)RTlFl
z-EQgSc1y3yZYj@cKgw#n%_yE^zzy1hFEt-H(7e9((%JKi0(OVbyu7QgDpovRyPn3w
zJ(>8?-FL%`{A+nYZ_t-lKQ>+4__I41P2jq?Sv<+)ZF`!$y)=d<DQ{=BS2A8Ha#7sD
zdc*-e;^6kmLT;~gsO*(Vm`X;R<Y#U@z3I}6Xk-1R<y`~V*iEyDSizBI?F`glJAayZ
z1#Msprx#9(8%5mi3E@7xN6F!c=(vz#L0K9#sm{WsMNP7(U?JKj8=S3tQ&0vYWWZEr
z@UQV~&b4|<?D1fJo!MbB5|4S?-o4u{IT(&c!@1&)P1k(=Kv#*~I%R3zao=U%KnTg+
zSq^SdW8O@{!$^#&cFLt;P>XUb37)XER$?QGTayys_f-cVIeWlVfbt_XQf~blvlJ@F
zsrV<PhQZ|!V*(!=X`IzE;a)t+HqGJ`o95C0G|fy-8adBkCKvN4lqge{@dYq1!ayiI
zWHoP(c9aXQ-m~ZGJpmlr8?V8kbs-nU*$Q(teM~V|*a|9RXRu{zTB#B2&&HOiAQz`(
zNFK9g%Ah7ofXUjHVD*9fiSt_r?mxgPz4i2i3<RGo1}yrz@!0QZ&#Yw8-(|e2Y|kuD
zw`XSN{Uto?p={5@tbR^f-8{OHTez4RQws3Zbu9F|Iw<H^U*b1?eZ<u_Hr7`lM*?k{
zym2E&2HkRScX9FVgJd9g_rZg^)i`GHUM@rLr!sUU%FvFqJ@6%?V_FmxlS9dX6Xe}e
zDcdnq<>sVNjfTwp{d-knbFF8^;NS{REeh57-rA1JeMZYaenBNSU%35}?H5v!+83|)
z8*XFrgCF;>s9e#;<O)fB{|4oLN6P(Yl>412_fIJI`}utuAB)O8?X=JMcc6)VC`hFG
z<!X44W(Fs{u20gXfhFn^B1}ybD0GH(s)a}K0XmK)ZJyMg7z@Tu{qD*wBZko}m;LV6
zTW4=22135`t-J63R$`27{PwpuQhdXHaZ<#E{o5yL!q^NpMU8&6i2RglQ#4Wi_e^Yx
zg^Ep)F0M9taka^dt4&^9Z4@`c7FTY2RN8IDCH7Qw%?iV+@gqO}*Q<v|^dl>deE$m@
z_TBW!Pu@iKkPTN}d+n8phZFCgf-!5BbxE;{jXk$L)^J|q7QS?nPHJ^5vTUff%z?r&
z0v<B~SQ&2FO7&+0M_3v-GbPIGz&>!ZK5)x@;FkNqE%$*N`oOI)m<;G=#%lD=zM778
zQO9hdClc(5BG*w6(Dk5umy*pgRvTqU6ZI##5oM;-eSFPYtf2!ngUN-TG6gV~wK<oz
znd#05L&=+G=Yl!xpj^IA3+7~g7{@%t)+cR}trXI238K#O{>;>Db@TeVn%P->FueYc
zh9g(J`r3-+IbZsU&Ww<5Vk|OTPjxfI_0sX}OVK@O>{aR)5#{?d`>CHszHk-JwvUdd
zC)}^#Y%%pwnX|&=eg9jt@63KnM$`}8lR>x2mZO2j*{8p&rR~dESSz+rIq`CoDwLdp
z9@$Q$i4U%}XL%H0rVW0hAi`uP35q~q@GU)Ue4!#NnW+k3qcI>5GG}3WD79%N5p&EW
zNeZ3_H#wS$XYV5S-lZ>V@$T5zKlqh#66=rW=Oms0F?jl9LH>c6#Gg7wWlTZ~rPr&_
z<d90QnlaP3JE;v}Nr<_A4rLyZOYM2Ak$G|>^JpWj9FLkur>-2lLqY!qV{#+EkZj2;
zYnsxCU;P_v6%;$he0E^UeOl`yu{5k)z1^-;u--Z}-dzJXZinMRI3zSXVuF&jaO_|=
zFi8{3SYqKD20AKKqd+7Z@q#SJ3=UcJX{|fOcT(Ib65oJ!r1(LRzoYWe-DH6Li})<m
z6SK`z`AF7O<P9;`lF_zHaC#L81*hbg+brFufv+B}2ul!#iNTm14rSBSXm4@c<)Dwy
zJW0Z<M@LuVQCU`2iHD`2zyb#uICpZ}x#w=1Jh!j0Zn&<Iew<U3m!DTeKj@sgpNx{{
zDb9WyF68;tl{2eXthx<^w!uj8Tm{sdL6SX$sa3jwFCf$o9E-+`p*XJRkL{WE9o?{v
zjP|@2Y3ok>poj6QFF-w>guZnPn~{D}d6foy&X=kZ7GDla66a9=v2aZmaVIQyp_ALL
zp+5t$xB_-n8IepRN}f#o{?dTY=k$7WmKP27^bMKIs%-22nT!tYa8)~X4u{w7s*3kC
z*{jQargp6F9Y70@Lw#!z16atjBxTK0z*X@QyirI6_5vPjCO~T`z?}dWjrwzy5Cn42
z0o{d||5bMiGpP#H@(az8NkW783#La8Bkppa*VW>U+IpkCE9?Ei#&rXo%dDM+p@2V7
z83>6XXR)rt5e}D?$LfQ1ddU#3YVNEp3)?+zht65r2ZV7ujL&nFPKq%8A&Nw1KG^CQ
z1KG5l5}eq$Go`c0o*0F+<r19PJ#<eIIm6MJS>PA~qF<ct`$$1!lng$4|IU_{Rr;9=
z$>{L1-~C6|Fe#@Y8ie~`Jf5UBqr*6zQf)@rV*k?d@R?;=E>Py90q$58{{<IxJ>_{W
znm7poA1;d=7>KQ&Bxp=YuAJ<DN^J2c`Rb#2`htqe=<w>1w*KPwqDp@t5CDqH?Kl0W
zc=V{t-xF^ig+4F!2Fi3UpCvcvRTP-N0PXrNm1~2-ZWKFZgh`HBokK?>c&q!p5V;Ke
z(UnM<jwd)4qksYrbNEtgrkOpJbqBPTXb~~WG3L`2AQBlGbJjGwW=1v?>ho)=qRZ9{
zdV@hP9Ase4KxlbqgSl~iE3uXL#M(zXy`CJ8mwvFh{t(d8QzAB|8$WU>GtQcQGT9Oq
z15u=7r3314qj?Rumh^8oT|Y=1*TZX@2R^YZ@hVW&=b+!7f%%CmbqsASg3v0&*%Hgw
z>gpx)iqPJ3aSi+moW>5|KhMIgIXHd^hja*T+zuBH!f^s`Acxuk7qf6&Lw|u$rgSt5
zor{KtLSfOxqlXsqC4db|H4VPa7Jk_wL_M=pN5P&tCgI^#bwN+%P=EhWr6*XoY8d+*
z4<FDO9Au!oAy}WQ%d7R*wYJvzYx8uu^}&Yjw;i@zn}dEl<h1A8o%91~@9&^PpN2l<
zv{$OoUMbym&uW+BtdI6y9@h=7Ksm(0%yJ$|dxe;ln4{{D9x3gi0WR1>;vv)pIR7)I
zw)1IcP1LETyrHo>Iq``NX{5Jie=6+(4`29z{0lj9Ry^F5(Z-?W;R3h5C3(31qw;XW
zk}Mo~ID9-$4lT;VGq>Svcz9I9!;>5w_&0n!3H`bl|8`|mHO`cO*O0U3-@v_V99r&8
z`8UwP<51sP;pMFS+m+70%}ep`E3kXn8X?f+qrXI=JJfH2HA29wVl=a;`XRcQ8ZVS{
zk1{WB9#aUFJ<L@qUPzY|maz01_CjOI3vu(kni5>)$0rrMwUkgm4LVQ_@*JhTMdbp%
zs%N4?CKu>4$OU?hT%hN2K_IJKkdz8QCV(-166)p`?m$9Hl?nLXr7MF9!geSHS`Wm+
z=F59$Mi3&<2}lo8nVw=>^35#A9%D^TSx|7w$rs_`5E}c7sXl}mPn^`I03z`c0)y@U
zW|tR8B%{2r4CMuA^LL@mXC^OX9*@Q3g;H)vPs$7Wk0dV`78MsjRsi~X3fi-ntdOY{
z>9Ru9nkFl7NdbAn=b+D?fwh3i3Xh$UtdP0qs*|!pHD3p+VX9ZdRIkQqE5UrJMvpTi
zy?79g6ZnJ)4b?Eit6_#$Q=uWig@$UnvY^mVE$awodDX<VC_;q>{TT`k@g;-?E-d)C
z@L<R+Gjt{eg@=;j1C<$Iyq<=>JTsZWl`b=wmy#JgoW~|*hRTm7GyHlnkwFr^LUxnK
zKB~@N%H#!|A}{DZvd%xeY}qg#<!*O59<t8Ad)xZ)ZQI7zZ)>Zs?ys(=A6ts@a#3K&
z<vRc0lLO?qtn-sJc_G^#WGO@V$!!p68zjss&vqOsT1?0HMT|5~ge)!upW3>?y?^!E
zKa;Qiq^>!h`1cORgTD^-co_QD!(;}<o|jqY=bO8wOlF`Z&0+P`GD8lnl@rb$w#cSP
z3{p~Jumw+C;B~r6ou1qtLy0f!>FhDNDoO^v`5}3DlOtr;6_=LT9TuO%)Lv@~2E3-`
z4q12EO^#8${|M${NR=D1;bBUd@1Kr>g}i*!i;$l*!i_#S*1{n%F`$U%8Mn|DdZ|5K
zNAX)09$AWH4Gyl24IXOkHn-+ic-)?#*B9`Xx$Tj%hLX01_Q4uYpl*chu60-1${Zy|
zTWP%1oMW-oRJ(nRHU5xZrw;||TN#i43XIcZN!?#kcsxSEleyY8k#>@l!**+VZcvH?
z($Rf`BA$U*KPStX{H3V-D=a?P4eB-NzyH(Q8XAUld&sf=SAHJryZGLF;jItDcsxS&
z{=L|_sXShznP|tu4bK_8*satHG2}&b&9xx7OMyO=%5yXWw^r%@q&C)JYAFc1Jzk&F
zR%Yxe{P|aN_4(z&hTb9G<9>(EURIow^GxTg-5ZNW2WdZlgSQ3O&dq9X63>!)K5bDc
zt-V^OXIo)7SZAZ%*Y_^BSJd1046MsH<W^VKcMNxxc|0CC$nMduU~jC}R6o{4ii2&9
zEz4Rxjy#8le&GB#4zzNN>h`@1LZ{}}*>?O(sn-Lw!{|=Q*%hdA9|if2>^$82;U#0@
zHT{?OB>oegVm}>Mbo<Z$ZMuD_EXxP@swJD)(0Ej>*7r?t%ZC~}YV*jtQO}o+04REA
zQ}6$FRlawy#^<i+>FVmKaQkWodtGH^E<D^`uNw}sJ6>BJ%F*Rid8?vP_<+k$d2Re(
z?baNtoqpWuEXgTx(ho{oU!!Y2)8%PgNSU?@Q{=)Wd&|Sy>fx)HF@p2OLuZLRL#%1c
zn8IpO8~ptwZS^o}Gsq-%AW2)C*p8`540$rhYmbqQXT_5pS#<bxo-A;?-`}FcXW_|)
zC0R1^WcWBv)-1}C9qB8p`D}S|T+5R?WR?v480zpa^y^~$*pbeU&1b`pd&ybz<Nj>y
z7<e(z!7-@sV!Sx>xPM%{ID4Xwv*yKyrC2fWVHlssD2*)2hxsa-iCS=cbQ$@uPQ!<(
z0zZ$gHCgy@pTdTb55rhI0(D!A4=*+rZYA);net)%N8!VUtZW$hFmKD!d^mH9WImkK
zm9>2MOtj|geAuu!8%911AIG2#RBz70haKsB*sSKmKvwfwE=+8U3*$PB{d64WALGMs
zoDm<+++#B5t)`F#&Uy6;qZJR)>FI*$>4NF$V$%~X?1arz7h25nQy4Q^=kbA_+0<OJ
zO^v-kJOit|?kv>mvt;0gv*+KGKWsch{++3pQ~9^RZSR@*cl4w1@5Q+H0>VZF%>OHf
z#OD?c2naC5Powxl;U2o@p1<%x#^-Y>&!2_o&u4y~<^8XwJjcI(A@lPr<-fK>`P%0%
zEb<)R|Gg!ir<VV^=J|`6pJ!?R8%w<ZM@!WIM@y8ieg6H-&$HC;$M_u99T&w8&%uIK
zi%JB`G|A(m5*g>CcI1)J)DZWPXxwTiLD!zae#7XOb|n6Ag#5bfyWjgB`E20yI(Tnj
zVL^<GPoiLz^x4EDlv3PH?2^@(PXe#Jvi_g`aXtNvk}wIy5`QImEr~Z0KZ}vP#9wLs
zOu{u{m)H*c$c}J8#({+=p{j_!*405}>?UH52W)|m(MOD-KlH4MkS&Ok-a;a)dfpV<
zdsbBI43h?3^~#=k`pz}f|L_HPX9RgCe<yD%595#tstBvIF#_xZ12Bh7Bor5OdsZs%
zev|wji{6@uk(aj8qG?$)#@orvWi1i~9c4EkEDM&X#c1y(cH<g#QLoaH$V-WsToRU5
z27NUu_Z3O@m619qV_#K<dP&F_hz0EN2#MRt`LweVKOj-oS=T6qtmt8Ro?3)Sv0pro
zy;Gljr$FUid?&F5Lq<C>!i(*;poN^jzegM(UHysYyCnT)@x0A?sWb6hKj|6}d-jty
z^ofOcR><$nNxrj?22rK&1Q{p<Z-fpo24exS0$$RUcwTu4X~IJ2TRWkU^m?k^nNO37
zQaXv+jP^Kzn%Y5b3AErVNoRlJIecZ}MRM2v9<q#d^(CHzS564&^;ESTZ7TwOXFL|Q
z$Dx)$fT38_@&~zAo!Ior31~Te;WC0H(EcnJSbkNrP>uAEn&T@e7Cedj9WZZ_1WsX$
zjKmW?D`C|BI(Z)q8ql<OvS&q&q(A)^a$GMVbu3v<lI}8<6HP7WzRAC`a>z-b4oOdr
zpZ<$ps#(Fxu?k1TZt*C_`)AP>i&-BTXs&fSmUPb8FjoMZHRK2EJ-)>C>^<ZQw4h^?
z#K=m5_qyS|%jNgxCf|$m-l%%71&BGuXXA3(4~grEkM#q8Z{oivk16l93KLM@E$qE{
znd*y+8_YX&N@X+8N?%JZPyC9%wuQcy7?s!3`c6Q7`Fr)5>dW2>wGHqYeKYNb#IMNZ
zyc=40HzfW$`Q9NKQ;LrxS%hA4l}S$MB&Y5ce41}f8%TkbyiUyL1Gz*wYI;eQ6X4}U
z)5=i@llGgKpI9f8CGaED!Vh4LdJp&!X2gCJIMI7>e=P0Ud;HnE?3sX$<2Q*DVk5}D
zb_6|gf4wJRJX84quZoe?jnVPzJ^hVj>&~sDvA^f_lrLgqPd_Ypu->sJ%C(SOyI+7Z
zBdAoT`UozP&Rkx*IbGV>1W+yGdOcmm-=wRpdPSdS%lVt?FKlhs3)SLE@h+IxcJ$HX
zKHF@4D^=4M_wbMrr5Mnam8_@3;fk(mu3y>eo$P7$)Vd2fMeFFm-5we53HsVrM~aKv
zmzRcO-dmD%SP12>m&+HE<(Fb6(UkJ3TQ@CUt_dsubCl=`-L;<Po=I=(%KGM2Q5UCt
zR{kyCSg3S)dvS4Ob(=5PGak7;S$>YNTwE(&%*yvC%P&hGf6Ts(b%=2{mpa0T;MYAX
zVcZXw*7~hhe{E@aO=mFJxh8xV2myOwfsRFiGvs!MoCTX5jU$bXBTWuAIO*k+Wcj#m
zxK+GW$vkJnND3uhv6Jx0aS^M4T(G?cD^g?U6<AKocQj$KjgHN7MU?WLK#S<FT9470
zO*|WN*4PCp;kBrKDS>`zG9xNaI)!0~36F>vj>sIGMWSc*Z;jgx%qakSvx?Gh&?0v1
z?i9W5=1P0@$nNg$-J>;Q%i9l@H4RkP4n$nBCZEo*Sv<KpN9VA|mTzt!IXoV&TYKTq
zwmX_T*5A=G-r);$ZERV6Uf3;m>?KW%u2b6ONo!Xan-SLTNzrzNp*3~1U5GR<r0v4o
z-G$uRB1;iPB-s))U?f|Kd7i0!LDp<G=zL8vS7e~JdZ4N7VEgj1n$g|eJ-bIj_R3~A
zjN)$6L_WQ@Lv)AFTivp;E8y!GZ@FWANAn%qhAv!N7al)6(!P0l%<j<Tu>7u<i<9I{
z@l99|4xr_TuN_x1yb8}~#UKm(VvT%3b3kAYfq7(n1|<_FvpE(D25vpK!WC}#7IHgI
zX@u>S(3no#9A>BX8XEib5@&_Q9yD8ArW#YUQvF4oH09(M6c)DF%ase-hFipG^0N2~
zSZ-+>IL7>J+mNHQA&0l2KxsoUZ9@s%VU8aqsy3*bKwEHrslTY$V{*G{Op$WcwK!dv
zlb`2owg=@i)cy0~Ve&Mr-ICCWb=OgFi$Fa!nU^MEA!Q{w=L5euK`(GN8F+@H&y(`R
zTjC4jj9&kbxS3oc{u}hVAMGzm-@}N~5y&cyZ0je*TwN0g)W9KbE_aqX%jpNRe_bzb
zC$Ee1@J=_2bY8}d1^H$<>vu%m%@4{KYgOgJS3gfrfrg+W9=r$hR$Egl5>5H`Vx`wx
z3CCCXS={cmTkYPo4_43n#Q{+-*{DPlK`BR0P=+ueNvUTfl`b%l5Y!<A>;ZV;e9yKm
z-EfElp2xypg9G~X6zB`zl<e@{Aaf_stNJwUZ7ANrVj3UVHZ2bLA8o)fIR<~fOaA-x
zv5v?iYoZs(r|iaK)7u8b-#0w!r(=4hI7EIS8Gue#o}`V-VSle<y{?mKT1Q(WXV693
zrlJu17#?@Z<*eZ40Lu#*LE1g{XvZZ-;zuRJJ++4q*WN?RTMgwE!QT$2l$W{p#Cf*I
zAh2?l%jLAt3#E7M(5$~4?KpCz^{BX-7e&h&rERm}JP=fVdMRaTr}FKbv<m1vAT(CK
zod(MXUS1@TaBXLeGpFBHV71!KzM$RKpW~|T6i0oLHis`z?6KyBDg!=8Tg1oa&1K>U
zc}qeCp&9FujG309&8p&adGS@qLg`|Lq691k;f@izb6Z6&CZWr%*ygm4h$GQxn}5Gf
zl63q1ZP6&B*<)f1Xsa1kOOG&yC3`vJE78s?r5egQc94&ObDrTN*{W5-VZ-VEv@{1T
zOezv~H<p)_lsCHNb37gwW7a^#UDi-;wU#%Oxg!B<SIixQ{{?2-6`RO$3D+yTP|0H4
z$-clyYyu}bn)gx{QOJ?jQ73nvPOd(qRNp)r6KJjHnVs3eY7RpwW%34A0mvG5F1ys-
zSW!|^fvw<YW=xgpLWkptFI#+dE_a>J0%x}i&K*vd#pR?QFy8ONcwZ-(HFG-ioFq|W
zkaUbXu9eI$GJ5&CSVqF)JMg?bWloa_-600?1&#;`HaMk&@#v{lhO5Lf8lUf_g$VgJ
zw6#oVVOys>?yNaN{X~sQz;O&Oj;O*yTrB6=)}DH!gNi=!sIp;DB$D>}SV3*B&u(>=
z)j3PrExLSLzNe-nQ1@zCQ6%3|WGb-hrCsAXNf!v({kA-;&v{UvA)p$kP@#=A!euua
z(924P$IeBt5IQm^qnFP+YW&4r?S@r3RrPkG1DJPSLw@Yl+!9}DRj1pOQ{pIh73Y+g
zO$GI6!(LceSSwBnkAMtflyr$7VYkA)S@E!N2k=FsWKiz?qqtdE2|Udx8I^lm#qBT#
z^KdUmxp$E`Ap9KGf1{ME-1}F!_X${gjZ&U+?;5z*PVeO__r}E`;RMWoqg0@}w;Jvh
z!M#G&y;1(VBGvn1P)0M{GqZaFS@@#ZA?%khQ}`AEB@4>Bb68k-6JD^1kR#7*{wN~f
znUnAAkT2*RgqY6D+@I$GK^-o5?&S>?9=od}dQ{4-bU8ab{<3qiZqJA#!UWK{Q7V?}
zD~Mv5a0Ae;QL-*<<@J4lJS&_7`a=$I=~Of&4fZC@GcIbx2q!emYcv);#}_%gdrCA_
z)(#%RvL0NB2sZ#-8>JGZAD&(q76ulc1KGhR(8x(>nlf?`hDrr8-?XLhea`OxDV4{Q
zg>NKI3(FRsr)&v{v$4qtbF=r_wWFOcVjX&*4sM`jqh#ZC5EoX8eZphX7lC~3Ag%LN
zHBA;&Nk@59{40T$S~y!FGF!n_zP78*o!e=1)%pE(E?Z}=yROTN0z;pxCh81$t2~}6
zZ@?L?aovp4!{1!KJyZD-FwqR<KfGmIkG6c$6n@N;EdO>W-zT)5Y58R^;@)}m%0yDr
z)<i{n8ESOLY(s_5;f%K&%uuhjl}=}e*XQ1?)a^Q`TfH!Hrgg)Jk0?Nvz}YpAdvkQe
zXf&6MwT)-hJ2KSWrzjsGwI%{zv~79bOf`}S{6xN>{BmKVI3^sE9-ux9J|U{%iYD%(
znh%p8$v%sD&_^einWRM|<NWTeTYk@{$uU_v{(<+=aN_3~MPzAVJ=E!&P$vV-k8)Zk
zBde29u9J}`9x!sA%p`O<_%y6IRE{%|MXFVQP3zI4E!W(gUbA0)w*KJ3`p@2#rEX`}
zUs8ttl9s60wOh7!Po>pLx+wgp_nOQag$VuCO&qjNdBV87if388r&xy}svaE_a%eaJ
z<nB3ir=1GmMp*ziGC?0rZ%hdPGAtm!y#hP#3Oa<Bel|Qi8~!ZRVI$0sW74hYbtJT>
z_DZ^_ucJ_CMZ#wygN5mdw3uF^S<f!7zv{~H<&o-+YG+PciM`I_sk4`~<v6Q5#QB-X
z#*L8~yS<{W#P9aoZ17Q1S3zS{iizKe-ISKjuy3S{eUnEui%flkCek?To5b&6b$gw4
z4jr$0-Z=svBjJ7NPK=!_?4mMMzZ@}^R>&;ZgC*cCo@d?{X1QFGT&{`wa^}D@6N4P&
zAk<uDwk)zNjYgtc_|R?#HH;kLi`^w6ujZY|g1x@7-7AxIm1%q}r}6fsXj~~;L+POP
z+ZUnq&!R=3CY{vRu0N5H&_$5Lx<RJ7734p^aH)V$DOwq?JeFR$552Sx{qqROeL}D*
z6)dFj?hU6>4MJ%Au_{@J5HN*{2wKs+JnMr83t9y?aL6U~vK&+|*u0xZv#!ez@NU0V
zyBgnrLwMi5@D0r7B^k(@(v8s0T`HRw=Q+<XR->nh-#;Vp%<d2V|2B3>hD^pTBHSY7
zke8+FL4w$(vUNSdve;*>GR?cSgl(&sXKP-<w&gQf%VgVD@><F%%*o4Vwk>u>+m_@d
zPfN3zZCi61lh(6%afkwVB#N&~vu#PUvTaL(o&!29WIB4PZEHazGwe`VpR!uJB%-!<
ziNf_#E_q$Lf!exOXSa3jkf{ZG`vmR!kJZd2<z}9%!u{fVqF#E2%D-LNZCnq?ov>Z5
zs2o2)%x3U<Pd0ebzCI<sN4_cjGnCV)wRtV}cC`WQJv4x!{p&rlOMX{6Jv<<j!FG0`
zrx8;lG)`U87$)6GjbUgn`Xl*;bONU0GOaD_qnf}@sO?hkLpksY#LH#1fjuK%`0?Tp
z8pE<$zOV%GeYpfy#-C|hY|K`h$@X>D2ChG6G;pDP>rdn@D9ggi%4FX<J8RaT(ydu&
z%X%AVW}e!z)@8C~sp%w1Jt}(oq44q9v)*39p7jfITz5XVXI-$AJqv5Dqx2=Rdg^4N
z)iI*|3|u@QKkmi5hdz>ltEp+n$Q9WPTt5z8)R=f+F#{L3LA)dEm%amwSa;fvyuW7)
z?e-kG6*=<M%b_BhcqVp+?<jT#s>i)6Jf<6_I&R~ctP_8;p8f73diLK~hqKnTHFZ$w
z+E|AU=yRXYCh!R953mU75{;0~>rjA>nzRn3v<?_&HXlQ{&!={{HN9T@6xDiPhU)1$
zWaT=MNUvnFe)S*&)e9F0G$$3-&n4H-CD)Jnw_yD$pnez_9BWxU$7q?!L$YvjX7Z3V
zt2|^(Z)0{*?}7|%mX2%XzUMWKNBR4|gk!o<E`KkVxyA#jJ1EInEzuZLAHS9^9m%ex
z3sg`4FX5X|Po}3Ym-X}qWF<*a*tg3U5B@J|?$XVRY3{<u(61ktHm~z!*XEh7_*>$j
z^~@7CF`ZSt<Y`yLY&v~*y*0B=PX}DB)4vP-6{kA=8clC0EG5ZtvPv{(r|Dl+mr>J~
zZpf<Xqn`g;N{eUFXCITUPe(IN*Z-~XzOI|<`g;VPDhAWUMC4R4+hu+K0Y%?WGO{z%
z{F$nBHrl?fTchpsIU#+(=ftvXqbyI3|1G8egDgrv?a#H)pC^D0QSbk$toLtc>OOX?
zm#X}Hkkz_BP3lsJp}#R-1cobc3Vm>~7LFZo%))UE9Jj;qARH&)covR1IDQGoA2F!M
zRDXWTTN>fOm*@{=Gc6?9%~GZXI<6m=SwR<H%&Y(_^y-BQF;Bu+uvQwDy<DgfVnRQx
zgrmZkuuj-0Y!Rk}bA;W(dBTivNVr(IRJc;OR=8gHv~a8NS>bNsUg06=q^E@Mi_Vk6
zZNf{!%d}~?L9=d~W0{Wd+RV>3es=P689xX4xt^bU`FS}%5ApLDKd<KJaem&y&y)PT
zou7B{^Im?Q;pc<=e1xB`;O9^B^QZawW`4e%pYP`9d-?gx{QPx(ew3e|o)b<A&q41<
zbts_UD?CRXbIi(>Pr33dS5?YYNV!_BTn#B#W6ITP<!W5H+QP4(l2DPIbL_kaXl%do
zn*++#4a(Jx%GJ%v)ve0aZOYZ{%GDjp)t$=Kv&z+r{Hn3364f$0u0{cL#OWoVm|>1_
z^v5y+T}O)u(mF=IP<g1ToOTRSENYL&qqa(ZHpb~6+l?`n!cou9p{OxP&u|#|S-y_L
zGfb`A6pxy4SKJ;oN~*F#bUT>E3gz>S%6cJ~axo_L*%}CWt;D0Ry3`#=e9<0oUs|R2
z%-;q7ttbiRUXvRvVW*xVtF;J@o&0?Cu(8>AIPuechE;}rHyS42nl#)vb-`5Pr||L2
zCc}!!HvI1``swH~HaZQBMq{I4pu*rV6c{RPX0r{ChQ?s95s%*%uWfE#TTDMcQ9I(a
zjn>zX+MFW~+DmR5w|eY;zujXUzpcbRS$VMXV2Q<2LL~gmD=sQ3Rv(F5t?*wZeyqTc
zyOgh=U>7UsXC;43KPIikR;h~B$Yy30Y7Ex5x7Wk5x586aRbO9K=BW^b{|D;D`waki
zoMT{QU|?VZqB5(U@%%Pl8913=07V$CCA|%U(D!}+z4-T+shK4Q$mIm7WncgRWws5%
z0001ZoMT{QU|?bV_m+WyDdOLYe=nGt8Gs_lfH?#Jq8|qJ0001ZoUPYONK{c62k_g0
zQc|(UgqjgkHfiPLt1@TIByzH=qk}WbWKx@21TBJSp%E0(u0;gV`vE<rpjEW$v5gi{
z5VUC3Dq6RQ!0i8V|EKxpBsSUzKhAxeIgjt0``xR_dSMTLCi_Pxw80?wJeF&_A<lCt
zGjJHjAPara$m{L=z0~bz{RUoZh7d%d9+EsyLkxysgqoliQfy-Yj>2|cxBb~t+N*&c
zmSQ=Sv)@KV`riiYVKwKtmHkM+tsK`DxI~>QMj!cGDPPi0xB}1M19ZWD-jn$~hS!Xx
zwDF1LUV`_e{yR7ThbT>DBx9H&S))2$I{}q&4$i^^$GD3sAi}wKDJgYq?+rU-p7&uL
z?@MgL9M4L|GR(E{gZWLrN7(|K*!KX}xQ%Ng>#&S<<X$ze*K!SNSwG5kka`V_MICh(
z(IR!+Rz#)hxP<%}L|CW&KjyH)IgdNsu~lK6KAYfKE|ffbqb7LvEy)4<Opphx1j~%W
zGr2tRFA*xFZb%)=e(P7<IM^Jv!~}av8!V!bqka}i#r<8eET-5DB=;_*k$I|C$MSqP
zIpkTeIlWMO$gZ1x4`>m#>uM~3%_W7@4XI<vdwO5bw5E~|YaCLyr%2qk%V*K7$QU-7
z;2GUvf*dV#v$;>ogK)+K`$5i;E4&VqdmoK)y?2`2^DOTQ`|Z%p7|1&#%JoPvcG7nR
zkL4Y=m&Y-VBgsC-Sx@$pVS9G&(q5^_{~qu;mKfs^Ho5Nrc?Y(twJBve#5vosy6=Dz
z#zC|&E;`5hj7=5C+ChCPRyJN~#!LEMVDk5<&Sk#&doS6wi!zSgoRj=E&e+==J*LK#
z;W%{;7$<3~-{e2TF~%CEf2jUbCV%hsatxbUcZ%bYx(z1(xruQtG`lY=xGo#mmhI0s
zxf~E@%tVrH$~W09xB(a80$k(yJ-7;&=TPr1JcLKQF6U2R5}vYrFo$|&MIbq5jS0T%
z>^o(Z3BHG_RJoR9pVz7~L6zjOjfIT6jj7EaC*eF?hc71hmXz3iH^KMoDP@A!WWL`R
zkGF8d^QKd944%U%9EUe>8io`(zX`A8-w>3y;SRijN$7$;$1tSx%U@sBwbXp)!LiJz
zKZ)zWZ(G|m%m4s*oGrqCNSX-)0AS}iuWQYDuIrqd*Sf9D>$<M%b6vezsi$V$+pC%R
zqtrR)S~=IsInR}qXXaY-nrki1TB&obTI;%qh=_=gh>(bo5fULsM2L(CiHLFkd^`^f
z2Fv-snRBFb!{<K1Rq(wp@(?hD^*r<Z$oW$wh;$;iz7&5s8y6SX5_j}f`B$s)iSe3v
z-v#&u>4mWi2MM_e?u3(z)QjyG7g0H=K2+cm`jYgL^->s3MjO#<U*~*n!@w{(7$aux
za^mI2%X61wiA{;o#Mn3a-}EJ6lB7xgZ}Yxw#G<k0EBRLfSN5+&aY!5vmxB}FR&W8_
zKJEyQ!sGCHcmZC4*WfL9JAMkkjt}C)1R_B}SWC`ImL}VhXOnk`5OIYVBF4VU{Vtjk
zOo?7iy1IJxEY<$K;(Px!=-SwI#C7s@_4TzhUYah=pB7C=rqj~3>HhSi^v@)Sq#^Z?
z7D<r|bVgo=F2kO&O->;9{s8;IkeQfi$vpkB`^ViZFzfgy=%;A5@8_(aS8~+9Ab!#2
z!gH0m(f@s-Sa0OsSiEtTXZw}#Ykz)fzOw*cfGWThq!r{8PzyK(f&vAA1TX**AOjSD
z3Ge_ZP!H$;3t$I^fpK6CSOEgS4zLd#0iO%w3nvTb3s(y_3qysG!V@Zjil!2%nbbTg
zg9;Uqizr33BCu$uXus&FSXY87ahFV$ER^_5wn}zO4ohM*I1NR^(N<^y+79iI7Nf)I
zD0&OsO7EkO&?o8h^i}#MJ;V?&6bvWB$CzO(G1eJDMwk(0oc?C{&CbL!jm#P55_6py
zWTmnwEE)@BiCKQu7HgMv$cnL*Y(IO8y~{pi$4cR)s8U>MT4_$HzjUi~mjiKXICUH?
zr=8Qo8Q_d@W;x58jWR=-rOaM7TsB@dSFS5JmD_%Y|2_i7gGpd2m<4*lX>bu-1Gm9F
zFan-bAS%!m_KM+(aVQB&g|eW0hzao^DO3;XAQNPRoRANiftH|kC<uk2D0Iq2axq*Y
zm&~PbX<U#i=5BIB+ym}055`0D2s{!mmsiY#cs0B_o|YHl9q^7TVU_)r?#ijk1wNAx
z^2K}=zlm?;ck}!CZvGU1f$y)%uVPm5suWe4Dnpf}>R131BnYsAYl0=gx*#YB3!;M4
zYGgI0npjP)rc{qqPgb7^6NFgdHDR_85VD1Qp<K8s+!Tg{2f||!Pb3x9i*zEB$R=`%
ze4-iAl2{;Ch&AGW#h=7yk~m4CBt?=b0VHe*Um}+*O4cOXl08X8a#Ax~GhQ=Sn^LQ;
z)z_M9du#V<Bef?|gcL0$NJ-LMX|c3ix*`22{Z~elfikg7C2NuyW!<uVnOim`Tafu>
zTe4l*p)4kc%TaQiJWbv!58OoFgl^6&;EH$!MqyQ)-b%Y=xMfnJl_ur7GI+cAcF*mw
zDqEGOYE<b}?W$?j{2jv`)18$%LY<;+_Acsf%iTb|ygt$(ZBR8VG^{j4)k3vg9lKX|
zZ|2^ThOE(R+BH^9uVziNsrjhcyPtV~_Woicx>4CU{s8?z`Cz>%u}Rl-)~s&UH=CP#
zn_bP`=IQ3JHdpJ_PHPvnYY)W_RS%nV2|BE<U$>>()g9_$kKm8gkB;?q`tTprKNbvV
z1HoW5^coHgvB&VosK>a+^2hTnsV!M8eXWRAbSt)1+^T5RwCY+<j4)%I5p5(GNycm=
zVAL3w+p^m7+n8;<w%NAjwvBdEyR9R>1Jgn5Aa_tYXdP>&a?|pklt0JJDQ3OdY>t?Z
zI|ZGJPEDtw)6!|{3|r(DwPn8x>Z<8dcFlJ!cWrcSTd7v}lY}RUPmE7ix^dn1?(kFa
z>HO2fo}QlAGx;<3vs2r!E%KM@uRt%X*Y_OtT>pI0o@DQ}@4jHX@V)rh*VOm%CGDm8
zWvHLu?|v2cs{WO0fHyEV;2#JM><`2oNJo+*%|UT69Ri2Sp>tRq{SKdF&f#|)Ig^~l
zPMveo861QSk_Y*N#=)7veHX?>b=AB2UF)vTLkUB;A^wnaXlEEVEFN~gM!s%(y*-jM
zA|0`gght~=Nu%u1_R+b~Q#Zxk?LP8gJlP(-XV$aiS@UdpLY{rkC(o%D;YE3|-c+y3
z+vqiTU0$Cz;NA5eydk}*do$-t^{IUeZxL^qZ|!e`W0_<6vB*2lyODPr?>>)%<EHVk
z@xTOpB6p&G!ZmR;sho69hTaq3Yu-nuil+vqA|J9pSUzk_r%r4C#{P}{hy5Sb;oABD
z0RR94Q~=8WA^>Rs1^@v7000003IG5C00J2R0{{VdoSjrnZzDwztr<J^h9t`-K}bYe
z(YuNOiDY7%MJr_`BqvhXNVf8|S|lV6ZBNJU&3u@dc4m_+`~ZFfNA8H9z>OORPH;p*
z962B^ysl{{PBsxMJerx;-LGD~uC6f=edxbI9+kg!+$HwtEsRR+Q=QrlSLi-Hb+|gk
znV|(z4%cXjGKcGQKwmq2hvw*KhiB;{`pw}vddFLKc%GKLHHR1I8}Ac`@6w9*lf(CD
z!Ta6eMXGs!l^BmD?;nRfTJjeh_G#IF*Wn5+`Yng6wBm1LM_zRn!_WOM9QNn~{|ASC
zy6gYyaD_JgKOL@4ab{?x^1j10TCKDluG6;_<?tPPyIOa6mX53U9iF3B^(%+xX|?*j
z!wa6M{^IamdNlL5!}n;Vw(jsE&DTD4c$rM?sKnUZYVAjdJ>Wka_UXO)n+{j#Vf|x=
ztMsV;*>*ZQ%k*GqM61z!EOxiLVmr-7foMmO;DuadDp%R53In^kAE}q3n<lB~s!)&P
zLp2yjGHV5mM&pSk*oP%q&U(ptuCiRGiD1&J%M;n}gwBN&CX=CxWp*Oc;|oJ;a52!i
zQJD&bPDCt|Gx1`a={(eZ1MyrGX%>Zr4pp#~Mxpf>%CoLMQKH@J>EvWh?8|W^PGurO
zm5Gj&38u7yCQlxwc`PH942>CWZfq1q5%eGwvrf=Y<EaJRi|x#_Asp|di4jNXu_<Jx
z>=q*3S4j?^<HRO1LnV%OJ7RC7l9Jph32S0fm}bxn1dF*?uEeR-k?ciE6xs|?n4R{4
zkY-ajA9+91Ba;WYj)F8BZ0zlHZeO`_bY<Cd9giR$rG?7+5;^57GlM*m`4Brv#&Hi@
zIagUaiX`ekH6bHZFp*|DQ03>!7}}<kMs$Y11x*7Qk|9AYYEY9N1MSilb@8^W47>oi
zO%X0J6_N8u;jQpK#WMtR61xw!qL)Bjs3f>8vK&ogis=w|02Gl#bPJIUzMfntOwdGW
zqO*QQ=Q(H3F|w;-O6hf?uep8j5LolhByIy`39bUux;(*LO2;=;XwB7Pz|n?1*(_Gg
zy0Y;J>MiI6@@CAk9{S*x&TY<u)gr7En6-HW<VksEQpXTVXTTb66rW2Em(6m*F7~lL
z=94<*oPu>SygL}%nUbY3Yuw6&d58=-d$rX__>>KHw@Dkg3cdnlD;>pCegf<y#mvnS
z8Zq$&=NVXKjdxhV_TCZK)qpLza+6btxJHWWQ`U3LIZNLLN{XX*u{%L~$gFrLm&%<h
z3U=?;JEjR|P44kaXIJxH_JpmJ=AP|wjRc>R^-*@r4$k9%u|W;QwR+^xv}Zr!uP0y+
z5#ILH2Uy=jeCL0~-2T-~cl%R}3+_}&=d92J>Z5bkxi4d6v7Iyk=NPLgkllCR^@r8A
z-?{BB>%-=;5nm#<Je>i0AqH3SSCG>K=2_jcS`j<2_3txBajpKFV|_$C8oEpg`|Y_-
zIOliD8q$4pH#2N+mwlh`DczRQ{;!?$H)8$;jNIW-0001ZoMl*LU>iphotbr{n8Ia9
znVafPk|mXOx>MpLjoTz{niLgVv6a|TDodd<Gcz+YGcz+YGs9QzC*8iYPOkpEo!!}a
z13P=CBO&+q?;*5Z1(=5X)XeHE3n{W$kwa;kNRz0YCUXN1qbW4CvTqu7(Dt+gO{W<&
zlR9ZfnngR&Y??zm(=N0ttwZb5dbB=mKpWCVv@vZ$o6-c@j6!6PNnwgml-ejpaZ1qU
zv;}QRThUk!aFE8)a<n|HKr7Npv@)$it5PGaO`~`?-AFf)MWg8|T8&21>eNKT=vKOo
zZl+u4VS0q_rTgd#x{@xU2D*^eqGoELt!Z1@jrO21v<=-&!|74loj#{S>2NxVj-;dM
z96FZ{qE<SP9;5&zDMKUZJ35KJr`u@_x{Qva@w6tLOsCL8bQs-17t<wlHl4_g+{Ddv
zIvva{bSWJ|K{}ONc?6H-Q9PQ*@K_$l%klEO0<Xv`@yfgkuga_O>O7wArw1s<YtRDf
z<27kMuf=QgI+W*ish`*5^?3u{kT>Fuc@y50C-7z*VuMW%bA+SZ#xag_f;Z<acuU@j
zx8`klTTZgYDYn_6^EplbR0>+)i9CthndwVf$dh>r6={H{@^-W*PvZ{Wo_FBsJcDO)
zC-2C!=y%?UXY(B1nRnq`c{kpj_n_CP#2H#di+L`Uxr@6w%k#L0dpXDRc>(uvKj*o?
z3wclah>JYHC0fE|UPMcIG3`Zr^AcLdOL-aZMHleiv=8sY`|^IgKOevc@<DtsAHs+7
zVRQo@&PVW(d=z~`fAP_L3?0G8@^O4TpTH;5x3n*xMElbLd@}9Fr|_wK8lTQ*@R@uT
zpUvm+xqKd<&lm88d=XzvU-Kn&Ctu2!@#TC4UrC?wRrCa3O^?xBzJ{*hYxz39p1Sx3
zx|VO`oA_qFg>U8C_;$X7@8rAac)pwdrhloM@8Ns-KE9v6pe#SY57OiGBtOIt(^LEi
zKgy5s<NO3a$xrdq{0u)!zwvYYJiov%@=NqIzf8~2v-AhQLeKH5{2IT`Z}6M^7QfB!
z@VopTzt11=hx8$T#2@n~G><=}=lL^wfqM9JdWpZ_FZnAvfxo8f_#6I~zvJ)u2mX<N
z;-C2!{+0hn*Yj`uJO9Cd(#P~C|HXgvKm4zdun2uaFN>mAs8<^3RT(A$3CeJ3lqPAG
z7HO3cGEzp#Xc;48WgLAa%gOR|m#iQw%1W{_eJZQaX>>WANoUY`be60ttI6szUe=H`
zWi44-){%8(Jy~BikPT%c*;qD_O=W^?CLuAzl(0l3Ds2*zxFlqA*+RCItz>K2Mz)ot
zSdtQ397)SWnI!EpS*FNT*-oZOhior9$aI+@Go@2@lv%Qq%$7N_v+N?f%5Ji|>>(MM
zD_znpS(zt2(knTcFAJnk`Xw&~Stxr-Q3j+WWmzPPWr-}6WwMv-E&Is6vY+fP2greP
zkQ^+B$f0tW94<%5k#dwAEyu{Qa-1A5C&-C%lAJ83$f<IgoGxd`nR1q#E$7I&a-N(o
z7s!Qjkz6d7$fa_bTrOA0m2#C_E!W7ka-Cc+H^_~0liVz~$gOgl+%9*>opP7lE%(U1
za-ZBU56FY^kUT7p$fNQY{Xjp`Px81tAy3Lv^0Yi7&&qT1yu2VU%1iRHydtm4Yx26h
zA#ciC^0vGq@5+1fzI-4790}w0AIeAav3w$*%4hPqd?8=TSM;-dE#J^D^s9U;-^us#
zgZwBz$<OkO{3`#G-{g1sL;jS%<Zt;${#B${iBhUT4O0OXRKr!HYEsRrMYXCCYNQ&a
zMyoMutQx15qc_y@^d`Mc$I^Rr45jFG`heb2E6_V?MYWPznck)M)hcRLwVGO8jaO@^
zHPu>bZMBYCSFNYkR~x7e)kbP#wTaqPO;DSukTR61!YZPos!hdITqV@zY74cc+DdJ$
zwo%)vq_R{>*~(FAHBn7c?P{``qNb|t)HKzhwpTl->1u|WsXEn;YL?n5SkC7{p^zQS
zOvv<=LPoHs4$Y=z*<wLY)rO|tm9(p98m6Ps2m{+1oN}??Dmme_I~htEzzDDn7zZYR
zNuUL^ftn;_!H#7DVb6*JVb6j+3-&D7vC@Ie++uc7wkjHqcB_KnV5UFQRV?I#nL<w?
zpIy+DDdzG$nXYolQ%!5MVnA%Q!oVo>JhXw(*AZ6GYRR}}by{1rJ6af>>B<$m%Kh{D
zvP<;L$V_)GTg(pR2DID?x4q3h#f-NZOJPWe$CGqaT55J<)8kRKI6a%CTwiy+&9q;#
zVs(;HCxz>zXq}aM$*z;?Y$g+Rl7%yw$+~Vfch!>A)YV-mWxBeu`BI>}D^rO#kahDJ
z$huNfc5oe4LKQ44m_@;}IyIO_)htvyh3chH?G&n=a;qFNqlxC8nma8$U4{Psj3#XA
z8RTy1@fW#gXpzVGx1lVx&3dAtcW$QG+*{ik$obB3$^s^WIfRwdVFhw-IVCqgn43R%
zoI_?f7SLkze4qade2%m%=qYBi`MylPJJ%KHbCm;qZdzOVe0mLi6;}g!y`I<WdEa_-
zKC`eeP%0J{_GVQ!-xDbK9Y?-X>DEGTxpEMT<^H}*xfCquFq#VkeVKt?*Z8nPZ=smi
z`ZfY~P!Ah+Y}Cy`y=)IrciYoPm)fyFQF~t0o)>+d2L^P827DPB@U4&P$rk%7>6zPy
z?rbPkiWDemOC_JC&beU)%D&hg_~67^%XP7rQNpq=;jnVGMFVBGki*KgBsd5vjUXK_
zDM*Hs4lLFYEUrb+xY)Z<0*l?-qj_m<5%!{irS1|$^guchSmv5*T2|}2r2jSz0nK0%
zAtZ5|CWqdp4*IMOh^_iwuV4gv9;SfM*Uya#CWA>w$C&b63#poew!RkZiaVK7z|psZ
zqi+Xi=<NWjm8XV-0Gztp0bW?}J_T=5@ID3aQ?Az`Bc5)a=sP<T{kOx!LGFf0)z+Tm
zYwdP_ha`dVU^{$jM{Bq1)^491Y@e(Tv4-Z!KGG@v13gu%P1PrKs=te;R=aqbUZ1Ad
zr}@^EQ#pbie!q~5l+)T#cX~U}xgEaF?eKLj;&G5G8+L5u)Ilz74^b1_)5nRh+XB<I
zx6`$^)BS<Y(21DgOT>&J{b)D|r(tHbVP|RsGymJL$v~&SVL|P*wRYBp+=;?;qG3DT
zhV9f1+v#gq#F9pAj+YDMz;ObzwBxgU4LfVFVQ1-vo&BGNjRa=<8y3eh9S_XW4Lhgq
zpob#iP%9A6RxNpG0>i)vFbZq~#(;5P0+<9^z!cC1I>5BXNC*i22>2u5kAOb{{s{OZ
z;E#Ym0{#g2BjAsKKLY*;_#@yqz;A%x0KWl#1N;W~4e%S_H^6Uz-vGY>egpgl_zmzI
z;5Wf<g5Lzc34Rm&CiqS8o8ULWZ-U<hzX^U5{3iHK@SEVb!DAb(kZbQ>)mi@;vEf4$
zdlHyWc(S*Sn4;c3_!>nlHhf7T)&$mV_>@2_Daejz7yBIWMG;#9u_myd1aA~}lAb<t
zlY~77yfLgho<8h3;7`L|+OvZkCb4c|-NJer_AS`AU^fkndhuX?6g7={c|uL2$WIjh
zSU?+gY^>YJw~hRUyn0|h;pN#|$NXSj!5{MKgS@7^`eNNi+zI%Xf*r@R>(vK*A+J8j
zdjfLO<M;CI>3j8oJqNrotUKUMLeGIeY1m79cD(ukEv#EuPs6?i`xfk`fg#i<g!+WM
z{GdJ|PtW7?>I1)Q{1J_XfcT>t!ynZc{s6~fKu;fkT;tdm#~<D}{_w`bK-i1JUL5w~
zuos8Dc%sq$<DM_~7l&6KPI^#UtY?}^y_I)-ci1?uP%dJWTLis<+!C!fP<aUE^)Q?3
z=`Cr+d=3URM|Zx^?{c_<Do1qyj_Ocz=uwr!)vI#2dYZ!>Y7SS?<*1I_c#J5%GU$7x
zf>B@_Fb0eR6Tl=e1hjxDAiicq@ijx=TV@PjHDVFyC9q%LdliH~y1yz2KN9dG0saL1
zv9KQ>Ff9GdF)e-0Dsl)i{J;kf!vaqV_EU&I;UV&1Vch~hKCBq{uwqyqTFt)y`Az!&
zY1h(XUt`^yddS3uW9oO`P#70W7#Bqt4G~5|gwY6LTr6Q+G+{JKION%B115p`A_(K*
z8^*&njE86#54$iPieWqi!)9DNuU~ZBq5CS-=ni`eg#~)jjkeW$m2$NQCHMVrP;$4A
zti5Box<1)8=aXIi5jDBWIl?oh-wb`)t{+W0_g$HR>?mIu+=*s<xT%gu)V=~##X%Qc
zX|(?x)SYS4-*Mcb{xVb@`pET?8?GLr?p3YY5!k&dyXQbVQTswv6<bg)->%x?$XWsP
z-oYyPmTSR}sjEl*RL#pVb(L|a#tkXY(8*d{<AxN<og7<VEj?FXvZ0%6PP^wWq#t0F
zzk;#<17!_2vj6}9c$`g;%Syvg6h&9!qqViRK8PjMCt3>;Q53`nR-_94fzF&dQqc~@
zxj)f<M#-d~;4it8?4qL;*4_KulRJWhlz<vx5~M5;W*=jH9CN@rt1JQKWK5ky_Shg&
zHZzP>yW^aF$w?@*S!2;=iY_N~WG0zu9=YdCF1DHBhD(j#(RGEHrN^mO`<q-+ev1~%
zRBg&!sMA)bVKc{-*6#efy~4ISkrEY~D&M^Epsm_A^WJAyxx&=6&5?e#<vDtxhK84t
z5p3RmJ~w-rSRwdGxi(+Tt~oUGT;I9A&vQoZNA4FKm_zf&e4&~1E2H{PdcEL1^Rdt4
P+}F)P&R-4TL^c2bDYBv@

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2 b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..c352e40e34a3eef6e345efbae3ded3998cdcfbb3
GIT binary patch
literal 36648
zcmV)0K+eB+Pew8T0RR910FNjD5dZ)H0l%yO0FJ}}0ssI200000000000000000000
z0000QfdU(fFdUOQ24Db`W(Y(Hf^!iN3Wd!Og@Y*pHUcCAnotXkDgXo^1(OvAxDO12
zU|a8!kpboI@5=^lL~mB5>;UwCwVgmu+gJ|QdGox2#?l;K!FKfkx&yLuB`-Mp|NsC0
zCCMU=H9ItW=Qtutl~VG(5pj1Sgb)y`RHu}hrcOvHqwOlxRYueiF-xYaj}g5M7*3;^
zaBiteY<XSy5{G#TCJugeo6DO;bMf1BA>r<i1rN;WNu|<Xyy^3$6iH%QosGfS`|LId
z)$HgS^G?xz!gY?Qgb|_eNu~Lt&_=$qvv7FJG`5>*Z#g8PL!^VKxH1jmPR}xn_13t3
zgc2Mw8&Sbi@h4XDb==uL9m)ky{T+=X=xd!j1X-;{vJzBIe6iUbXGw5~ecg?|f=38T
zS5<rRF7~fXj;WTgX&om16P);BcjJzSS*r-KPgTIp&FDNK*-tmog$Of^qMK(Hda=d)
z+m!IL*|hD|^k%xnn{4<b=HnhnG&_mq{hFqJ-A#+VB@GX`^uID1zHU@Hyn%{Jhq4;T
z+(znqM8(9d_UnY_dAs_~MM2Xw{Y^zHIY`K89l_4Nu<wAj^*_4{hysFyBG$6Rt_XtA
z%aVtPQHWSv>t&=DUEZtGm+kxBi~L@J|IgrWd%ItC2uVw_P$V)z$QjYcer)f(FCinA
zDj}0qDoHcb#WSfA3lV;t^Xm>!o|`;i$wDD1bw>TK2S9Dd<#hN{SW<XRPK!Cek-fPI
z@8?Wm2<2qRx)1^Za5&Yhbf6ZHqHA&Du7%ccIu@6#i>z3~T3+}WdPJX)-<qyQa@Uny
z-S+7JGFQS}!XwN(6<qKFzifWlErG*pX}@#0rF6|%LV#w5WpEWPQ+sksaE0rpkbKK=
z0NiFJ_%qFef*?mwK@bEKl=H4U4<a5EPioq^njJfKkJU;$cazTh%f7vDceMs@mdG~>
zw69XGd!2uv|2lDRZZbE~Pz0;NFIe(aecumMsn%X|NV7P24tSN1OrR@#(bnEb&ukc(
zX-{l|N{TECFQSmrx$N!Q2athID4G<209tsxyvaapAl-_Y$*JBE<PN1ReXxr|iY;!{
zmcdi>{BZ2ucT-T8VGsm^B4`+cIT_=on#K3`v#)Fp#1QX{hz<K|rTqNv;M@q6)~K_a
z3rs+^U92jAfWQCW%zIluCscumJIShFA-v2IXJ@EssC{pLmyP(O(1*no;z|-rhZ%`-
zN}@-6wlr95GxUG%r~Pnm{4c|d>=c6{c7VcAQK(+?Rauje8Rs&L$<&1E`|8(L$lV6?
zmaez!Am@fCEgHcSB<S1y|842leUIxw)hp}*>Q<w=)zw1VU;dK(2{khdEQOoNbMg=I
zaFE$SHjWm^&KK;ZY$(VU!3_u?;Gd`eE<W$y57NsEI8a~_j<i5e680oQb^HmA1rq)6
z)Q`2i|DA#F{8BIp1@8>4ocP{Pm#`!nNopZ7v}TQD60g(pg;L@9|KF!WverMl^HWV6
z4rP9VP!SFzleB$|m-#Uc%2QV){{Nd=mg2sTU!Z)~f>^9^JSCec2+<&iN?W5jt#nP2
zkiQ=43;6Z$XGo}ZS`?Ntp2M}4g;&Kc0$|Iy9vjOC$$Ecs8${WJ$_*CTC>pr`sY-SJ
zx{R%4LqoFwT2W-3&vJgw7iH}K)WMpk?qzqY-Dm+e2cHYj0#v|?I4JJ_Z%)5<9#|*+
z$VPcXitP_67R{DhJ?<M|<_!a4Mo2ISD<J8}U~3UbHquA`1B1K_!g~}P^+Qge;E=wY
zGTxZ;=|d`-(3UZmTXfz2re64e{z^Fs*k&H(d(Y3l;}E<zG#Rii=GA(|wEovLrMLU`
zyU2Ggod$-mzdArdgP3OZrM0ssr^!x{Q&FqKrPRGUq8KHtp)eB=9;&@H`(LqLeONb}
z9TF!l<i21eS%Zzkkg{_aOPz2A34!f&L6}fIh;)et+yw)l-}e6JJ!i^(-Sm|Wb`c4K
z3_-*OE0RiZAIq9ilYn8CVvCBbB7$X}Tie|+rcw{rXV|gK!o(OP1WrIDhPHO)*KGfg
zsrIC9(&hed$6cP-Od~*JopdL`FA#C9BXkc)8iY){_1I7AAMX3S@|SX@XE$**Hc-F;
zhtjkY0+hmYl{Q#35L(d?rVC?n(u=Y<MOl2`jtH7sREP+blF|YTG-_05IZN+(PapcQ
z*o{90(GTxOP)kn*W}0kZ7OEBu0di)W0h{|?d}q>w>0V{{<-lk%Jr?|TI_^ixoML}(
zHIouN=#81k`Q)IJ)1*Kwm7cCe=*YS!YzT;LYd;7josY}K+4XLl38tB2krmdxblhT>
z{f?&I1|D_VMc3WUz&f4}y}^5Y#20)=cR2eS|B$1}y*?8?fzmGF?mPk|S>A>cnnH50
zVx$5OdiZD=Pd=TvF(Y%PMs97A0;Kf8;m9SudmYu$Cgh`CrO*|DQpk$tM+hq{zfp}%
z%5`fin!s2>vGM5EEa|-4lK6jk8U>`|I`!S}qAjq@8k_8JJKG(0(s@_0^R|b)#yfQR
z^t1F1llsoj_=cZ&!e1zzl&xMt;<pTz!$-s@?6&0Qsy@8fB|2pb0rjo?QhziaD+62<
zLYdCQ%|Q=gVT2}!dJG{yF4rvpWSFqSICqr?ff<7p9JUHF%Vy(dwnX1-ud%45_s(tv
z(eFSsC{1!kmj)%HuRV&GCK5o&adM!A1_`ZXlIh4KmC<NqPy~gM8pBIbWM<EVrEHav
zFe)S9_YoNlx=_-{XA267S$YIugR&v;T3ZSzHRCKKohsRDjDi-Nrc7MApou&tCPpzD
z3s!8{ao{u;CQ;2hsh<jZUpf8$x(gF;mFIdGF&QNdnq=g%<Wf>2=~&WY>qtUQMNJ}I
zrfhi#8k7RqQX*2#e$z@SgF+6RidH^7!`?s@iRb}&cl>LxZU}{`yYqXD)x=Nudd1({
zqlrXm?B4GuX-`@i(S6*n=Y~vSR(ZjIt%)pSP4`WICz%D=L3{VZf9LEKAtyO6SG^gz
z%|m(3Tab6?lFz&q`G%k5iMJttx!z$YoO9KL2?u^zJQQh~$jMz*1H7O<BbiCD+l2(i
zj>eA1_Oa_xS%dZZ!v1nk8eHp3$59C7D<{Yn+_$&_smOi9#g5+qW2d9@S~qDGvh%7E
z*Uu1ROfkn2YizN{VW(Y&0OF}&5N2p)H|qOX#8D}cTula9ghN8l)=$Iq(6gEQKJbJ8
z(%a00;gwwDhaDgGulNcuN``C1BKJ|htF54xkzq`;G9D!;Jj;dc&wjmJE_K@mKEYIO
z%qV)8mqoOdVuGH<<Qp9dKCc8rEdQgf1<s*07{#YhkQP-4<ZUho_`OEKM?DV8d8R--
z28X7LzL0FqCJ-U$yyp)p6V+^x5hj>nffY8`Q6GW<^ghR#uPHt!ZpnM<0luVpZ36W0
zoG^e}m`FE;2pV~g05aC~7G@Zc<|tT%kY`qpccB<n;!cpPI~~nQkS=KC)pN<zx!}La
z{^|JvJ|17hoo=rW{&Ss@bTAx^!z2S1kjkfL=NFe(*A+?|sI_{d*^1gXw;V;BEhFnX
zz(QR;eFH-yV-r&|a|=r=Ya5YRB9+zEOYb&o6W||fRaSL_b_CHI0tTWr6dZ`nA*jtM
z*_2F^6$@VYN`2W-!#UXe|7iQUbML{UC(mBILok9uW3V_pfk+}#s3D3I&{J!&Gr>qz
zta-qYF(@S!&Q!57drx4WfI|w718^|}cm$j=a0Borz|Xuwz>JOAbAXvYAr=6Y>xhFQ
zLQGXC0$X(bPMPJDH!xTg%YB%mSQ^qbP>-VO!HWnBpniQxFEo(T9fM*3vYeH9-6+{v
zH%d;nj*^=rbL0(748E+)+b{}|ALmdFQqmvP!UxKfK>lgPX$M7_E(n;5DI3L<0B<b=
z0@)NO{_gOsM<B}jZ$pPFzA$nK5e0|&KfOE!LBt12$~r9|h_Hb2J;XSCA<c|0<cKic
zC2x4>#pf(0!(^GRv(3;s&A|lFqeZ1bYNO9by4)27Q+WCK1r+KNUM$=!EOltqxj_xA
zd%1Q9$Bc}G#oLmNYomYpCq!?)Mt~)+MY5gB*Z?g!$!rt@wRX@mJ~I3o000(k&H?6u
zV`}`DeZd8hyaDn60P5&<fSxe8b86eSSb?rEgf?*Gzd*(zutG$aub0QzDQ2eaMk!He
zu-3P_SGHEED#I=DDy>G;5(_OgNQ4qWMNLDCM9HU9K+m9%kx3CV3o9Et2PYR=n-PYX
zrLuY8#3ZC-<P-yYRR1ShOhQT~T?V;K3R$w{$dv~-)EL9e26B>Po#zrSd&93e>yq;>
zxZ#?w_<<gNrdu$Ci?9(6l5L)9l@jU<TW-WsrkiTBZN^z?e0)qa!K6;XxNu!&d5?AG
zS!TH%d10FhK|u}fFxeEq_d+pHf2*yr)uO!QQ$N)mb)@5c;M>|MSD`_pdT|S^N<aU#
zk+x`&aM~F&tk9&{=nQkYLtNo%_t@?}4|>(>fV}D#e&x5K7$0L{94r+}!bC|+#$*Y=
z(<s2Z%)?<ZpkgGyfrgsnK<2pC4es--FSW}#uIX;x<NG`o#>d3T%1RYgNFN4_01yE#
zKs{guU_GD*unVvca1?L{@Br`<@CEQAiVd3zTM7FCm%_nt7#s=5z=`lYxB%{lkA`=^
z*TB2s8{k{vJK=la0r&y<Aw(=90f9td5Cnu4VL%ihoQNfe&4^uyzY+To#}R5vNPu`E
zo{3jNLF5tzM1&YXv}s-=+hj#S+Alx;6W<5{prWTD0fP+F?vNf2-|iWE(ecS|T*rMB
zSd6rz0IyfH30t@J9pC9)UfTZceYS7*Wd8x0_xN?VgUj)39M`+<JPDuol=Wc>3sBzw
z8fBfbUfH5thXa&xh1$gqUkX}Lsv)`^pW(pu9Ygzpq#@!7|Nr{yGzjz5Qr}>nb^f0N
z!|dG(zfCsyc_N+T=RsS~me`VepV=}lJ=Lkh;HUkXHLCye$(M6AThZ-jUk*0-<+?}h
zkEVHkgX0(Cc*oAl4VRg5HGVvo6WB_I8miO#pLTw_`*4Chtp1z)AS^&;JH|->Snt$n
zGo4PJHaus3>RB-tu=%xiaNx?yo_5sCW9DCM%^S8Uz#Vh>m5YWy%nv(1^dI#v5W8ng
z=f0xB#ohI0bvGaK?pJ`q=S@LB=j$l_z0|VrFxI%{!r3P4Yqc^Hn*$e^H&AqxV%KN9
z6)nB9Owq4CxexDC24>AFd3e?ON`8#uSF4H3t#HuO_!qnEmbZBf5A18=d&K9g&m)8G
zVYkP2JMFVepZ*^9wr76z4xtuD-z!6^6b`G#HYr6#$=|%}gq(c+eP}z5>7R4DZ(@&3
z>ap=XFu5nDG+=h`wfEeNUYpYg(|Tb^KP>LMW&LMqKdpec0P>CREya!HgXe+z68K58
zg;=buq_#`9jm!=ic9PpA(;iuNQ-JzguEWyw!5yM?Nsh~MT}5{V%{6&$V!DChx_o!>
z{fp-gu9vi)QF%b@1&ybao>TkC>I;W&{!`+&Qopzzhx3Y#zrfkub5w>~So#gR@#XFU
zssCon*6NiiqgrD&i3U5i%=*&5`YF`!&)0i8{|@{62RBaAEHBEcZrZLN#%W&GZ9mTI
zeqJ75Amj+Cw6Sc*^?U#kCX{g@l{VIKJwFIPO;WCl#Nr8~IYsC_(Uq;JC@IUT7SusT
z<d~WmW~PN@<2ZPazLWnFyoP@1ZWBav;;bkyt7-q42GE8Fcn@1z*LK>ub?w~ez%5;l
z_nFNH7M~QTX!lX=J+&i7cT7viwRS>tN7c3eqz<0kp;J1nz5`eFMrZHL?W1{pGQU9!
z`eI>Ub@kbL$zGEDsT`zn7Rh;J7f@V8bxyVuGMz$jT9z}2PQrUkW`M$LdT$xMWAdKa
z8wTGM`pV^pB0m*pTv(YxkSl%c6O&a|TVt(-*6FUjT&31(JYJpNXfQd<PISL^tIcwX
zu^BRQ+8G(b$~E$eiCU?jCI?Zx5=NT@D8od)l){mbaHJZJjEAEvDD^sg{SN*PpdLA|
zWMle@bh1jQ>&*>+WS+ymP$UPiFC((EP&8L0#X_+hYO#LxloAHl(=EEXo;5sASMh$C
zaW}`!ZK^<6LoYk|T0)u0{-VQ;xmhSKLhtw@Mkl<v^>3r(|0i5g-oxjC^&iuO<Vxeq
zOv64RSb1Yi@`mw|xlm%1M(_Q+8hBCSh*N1z1#mOsX3Batxtpn)1m#L<^j#lL2U6aM
z_EIRADJ#Olt)Er7a=e+l5~v^G39F4}AK4U2miK04jgQEML;%4^$*Hm58-8A3YXOKK
z1@r@8#5TaxRe(0XKs6P3VjuKVK~}pPkq1~biiy#&;+Zx|aUG6J!$KfXmN!X#VJQs7
zog*uj^+bQO5)*5(&Y7%wsaKJ-wA3SNJ0ENrVa0ss$W6Casnd1gn%C+>bGA1&u>sYP
zU&9gvjU#^z-4ce?^Oto9raCxAD%!D#b#~976h2zp_0H2e>|<T+qJ=vD^;&D}y=LM{
zA7D6kQIe~Dup-p=T29CuFdBt-m5sL&s<2kB6@pgy4Uec=Qem!$X)J?-X26LKlcEjz
zw@0?#xIG)*)Mi=jNj(P5q4c)#UN+c`RX49MzAg{7FA{p64v$8cd$H_e?mJi9r5OAr
z(i-9b*{TD!tVMMcYFYZFFnN)UL{#=w15DOzDaR4uLyqFTO7I9GI^8<jn~)5^ge!Lj
zY7ul)W37QeeMnU?Ki>zEA~h-@?l7wR;6vGl63p4Cil7Tczj9QHp}aOQlbIj9F-8&7
zv`==OWS71}%Uk8quCbvbLuO0)yd%06*UtME6zW*N(a+4++e|zR6;C2LGP-sL{#=v0
z9Mt+<+_=t7vJOfBN{AeMO?@pc0?Fo|x7_9abeL#KZ}x5uX`szO$fHPyOui>1<@2dC
z#MsX!WFlr^*ALMYnKG;SHdmLaI3Xt+ZP`dFCbOGoXe#z4v)0>~oi(e3F>Y+1XtUAx
zvdje`v6K3;Zgp$F)5gy3d7RrJ9DTAK%C6C&sTFYDfup%Z5`j@Yx622w=;FE{GsL)2
zcYGY{CS~Imr4K@371~ElK;g!%Ar^V7iUFLsty<Y%w3Vnb8)XIoBoVd~Ri*2qfBIsA
zvkUXzswp>IqZI?a0AbV|hlgA7)5IvpCPal6Cr*V?s6TAR{#~GW-q}bcK{l;Dqf~?F
zo}p+U^cQAPQjuP{qdx}-LiMtNsVocCOX_{D>Bfp)CiE(4$Y=eeB91PI=JGK($j-kT
z&yQOV!CW#N-(0w9#D1A%d0<jBSnA~_5Hm_P%W$>Z8%Y_JQfI?Zy09m7%)iZ$9qAeT
z{DGR^!~Yaw8fQw_paCm%xKM7IVyYp61cd(dTXDV&d3xJgiOkeZjy<`9U*VTXu%;r{
zxJe*QQ>O^n+_z3zoNcly=)uzChi<gak40Eh^1reKwe@#<NyFtPJpi9>hrJ9J0=fb+
zM-X=DC}boRl1;!xPqz3FoLHqAGuS~@g#--+^}wwtx`f$fc7hK29Yi(f+v$?DHmIP?
zm-nxrz^J+UsTgFyIk1f`I-PpqbdM&4jHddjn?VHg$=q=`92W#NCrjx8H;kynVL}_M
z9?wldp-)tQTPAc%zbuIN<_epuu#k8cG=~y7W-;3gmTbzvIjerW!vL^^GM8{`*wR~;
zY$+O*%)bM<R3=bv(l?K{brOUP6QvpmKyhxXjc|g))o9i&x)+Y1&bntblK0vN!M)hw
z_=_Tu)RLh)%TiE!p&fN$<*U2yJS2-k3x!maXf3c~I0#Da6w0755&A-;vFRqXjv(zy
zOpdP=NfO!y$V>2hFccEq6_Uj;FKHLM2u9J?r&ej1Dh#w%xlet|H4r0-M50!#fD$1p
zqS|)7OH2~o&bHWEEVB`=hAU6GS0jrxK?<Hxr6_z{#p;t+ZeZLO05yLQ6-;F3kcuI)
z&YIr~vFs4lx#-0>H8b5Hc>_4{N^7{~I81Kj2s<G4D!cHq&Vg`CKdI;RrCa(=tu@wV
z65GrShu^p_@kfb)-EknCR%~-p#x(7;<zXO8^q_roIKr5JCy6+A15=`N;Akl=ea0is
zwR&vXa4Sj(v?v;#adJqg?S#3pYcfBbz>`g?qH|3p9t59l;&eUI;&vsKPSEZ*1hzD1
zb!W;0{_&A@cF$e<K~ka$TE%zL9|hKw!SPkwHc{U>p4))^m@KGed?MfCkOxOPw^Oh)
zKM7~u<Omt-*5&f0bB>#MEE24TEDjkqgP6QbF$a?RYDD`?qF}5x`kQQqKkdpmIK}N3
z9w(a+)nGwvLcc26Qe;b-XXM#ZeZtmqN5~YZ_EK_0EvoZyheGxyqro#SrI4J!<<ckx
zEr`DS=8_8KZryD3u=A@@iiG){aEOg#d8d64tg1-<&x``;WCcBJ<MdHP1X`8GM{}$Z
zYRX}OwN^_zp+#s9jP44YFF5&yo=3jG_#t;aMx~B5k0;bqqQ2$5+ofF0csuOw&k6BU
zdRgDq&Qs@_5GOb3n~+@TO?#7nEC)%bn?ge0+o@{RJn_6?yx1b$K5+GsN`My;>n29H
zMe6tte|ttPW%fu^VE-w8RJY^D)<mvH9No~;hRARRETy}wPQZsGMkG4Q`b}g_enZ`J
zvy+Y<&sJWpn*k~=D5u5m>iC6n!Hn~B>~4Dbw+e8CMJ|V-^`t~xDvx-UufgyYxn0KJ
z`mlKQ@6wqIF<7^=w^_X^6M<Xap3!;c%>r=a@Lqrv2^uq0`FxtRO31X@r5*ROZf)Q=
ziq<2!7c3(?f<i>+kk3BDzuxi#l`ZX|jQtO#1a=cAc!UX|CUJNI_v&<CWU#j<X|b`Z
zrN99$15fR!ML?$yiG1C=D`AEn1}=+ANxkdz9#uODXeVwZ(U_%6FLe2Z23~l-5@WrB
zr6e;sQmBwjwPu394;Ba^CB9R3{XNl<e9+|g$Co*D;)JjOv?b)3f<^P%O>r_u2@}K9
z?n-v3a&i>O2SIWz#4yBtkCw`f>jilSmlCL%*7O}bL3uHBg|Vh?EUR&WHnfy#3JqQ4
znwAP%(m&v*t@#<e2yIgv7KqWBhzgGUtauQ=h(#`>0Oy;jVXLGBi+7iJw5AZ9_@oxW
zn1#!KUZEoJ_e6MO`?uU;*XYQys6I*eSRr3wx!30JcoD2C8{R8;LK<j_)bhSo{7+Ng
zZNz02!maS|N$mMf($34;rN218Raizo2z-}HQ2>?YEK5k*sYW$aA!SnE$5_ar#J5|P
z9l8iYS80vT#M8+CZt7U*8~*oODeufu&UY~?4^CJ7?i9c~C7g}8VUdL}VgFoU<6JX4
zA0`n&!ZwcM;CP|=W&niGaq2WsS*w~*gY6oeS&s2=Bv_FldcCTaXsG_zGxikI6z_-a
z5dFshgpv}HyqT;z`sHPWh2Ffyf42_gHhmXMVX%}C&YE{A%+z3{!Ur-N@4<#pGAKN}
z<L_M1OLz_ALk58f0FfQjgW;KYCdp>~PmQy?SUv{KK{cLf>+Kz!YGv2hC+W%@kcQ!1
zC~{!5I=sDJS(UqVF%%54<9a)ZXc`bY%gTr7WrX|=)7Vyd(U256czmB}H^REz`V9AW
z{QZYy_O|N`Jg#)zTUHctRjwlcfzB~Xn*EL;Ms{T6k-+#6p|ApiD^uWLRZJ;I>Iyvy
z9mDiV66C~~3MVq}pM@`>j4cVYIdO`b1MqFb8v;8Hzdts3Mxa2BEyvacO2s}r_ctsW
zEgLPsGssUU7RY?mV$XHd`G+K*b&DwsJ-!KNdIl8_%iPBWPPYOINVQPi0ctI*>!SQU
zGmsEc(Xg=q9ii+Ngvbpw+z(CI=J~v{lzRJl_=16E+hOg6lx{X(RQ3P9mTZYIo)|*y
zM#%Y41lkdZZr%)c3pw!n3k7oEYYNl%9t1uRwO)Y_VirzRTaXBQS;+d4k2Na?MWnJx
z1ZGe6sS#tiK$}1#?v6Vv2|b5Mg4t;UW^j(<59?lNbti_b+-~al!BsPajIc<>5Z~Ee
zmrf6eLz<*wl`vW4Th{Y%M~O+}KV{=6{V0P#lG~BeIrhheRRu8U0x=<<@wsinYyEDm
z2#%lKpzG*8wnUAXn_PX3eIVQRsD~vfQZs5f!b3W5=56-rtXaiD4O9sskJY*0hKl+9
zU`#LryDk@41=`hQS#N|b3_NLX+CV9tuQMo96=n*`-2`VsEksG|^u)I=$=`cf-twHv
z2nbN8NbG@{iRn&72tATuD$})NHcA3-lYoyC9M4p9Rc+u*ax4)fGy%Jg7>5UtAksGZ
zbP~u^<j!SbK^(k%A5Q{7r4<iLmNtF~kJ#vVzB*<6>-%)lZ5!VORXwvWf2H0Fb2z|D
zq_d`#kj&Qng%;$IMPS4~q*4N5PTMGsQ;Hgj=WH`znLw7@KV~F_Lt$@#QhM<`<f#KA
zuBamU0_h?koYBL@k`$-AB)VSq@f62w0A@{Pp`zH|!H#qdF-uOr^beHGx+ov^V|Y(r
zG?q%{>jM0W@1C^5v23=3mqlhFN8RB_7PCGzMfxG0h5`nr)l?Q+eM<FnQvI@q5DQdz
zu(Y-Zq0(g3io;tMg6SQ>w^$Xm7#XTzi-qV)vWb*mF)&|O#WnvDYD8gXv<eCG?6iV*
zv3wHASU^&bU)}YT5&5X%%=b3L)K|ilB~GwzQVk`-7?_{_`KstGZ@;x_#6M;xDi%`{
zt;+pfC&A)BU#8iTqZ7oAD@U#Zigi)5Nzf{fLzuM=2(iWDiu|lXr=96a*+DSACgYsu
zl$KbGMH~&++ou>Ua5GYuLyG`Tp~87u(jm8%O(ND0e3{|MC0jBD<u(Y1uVC~p1H21O
z4zo3UXgrWiAP$I8L{54_(U<RTG7xthuLR?8lFAg!8_G}!Ak{))n@Oia^(6|D?iq$4
z+O<X@Sb@5L6$TqZqd|QzOM*k#B&E`syTYRdKPqJN3*zua(@!!1^+rA`^-yINUsWKs
zO1JDISDqBL9tfI+93k6^19DUqb=cicA`VwNAD*P;ROb(}P8#lroS+A_Rq7rtB5zF0
z<SyJQ@ThaNu>KIR3RxC=H5u$uj`fP9O^jIBq>fdiAAxTALdO|K$Ij81JI#jPD9bgA
zLBm+M8|+4!ntve>wgaVXx-d~oVrJ#cWx<RTl1+T~)ar<Nb6RR8Il6-_k3w7a$`Vrx
z5UE`HU>S!>an~-HRhkM3T@QkVRRHPV5mV2@Nv@{bsg$-l_>#lrkP_C7Qt8(J!9eDM
zUdB!0_A5Y(jR>RZEIqMg0^nF6L6ChYi_<0oLc5KUhf3^Gwhdt$rxhcx2o!4EdLzjd
z1>&tG*E+8U3KvCY2?id~ugBBA6HbC=QJkXJZ0sTJR|<VLB39KrZk03@S6IGp-Zv3a
z2T`j9F(>w>Cm|#@jcDK|q54`{Jh$XK)uFfN{=SK%k{DWShtLClb`AM9mGnt(;3E6G
zxTLJ5{KN_@Y7BRX>Jh#^A9HkvJHp70)?#GW#5JEDVL75fQN3YgIy@6|Ad0LB_g|~o
z*9CHK^o5RyfZ!dT;42uNFabUx&G<}*eBGhUq7k4qA`Fz|yh>KJkdOc-Al3LRG$0S7
zo^mC3Spd86slLdVHOe&w3I1D_^K3~VN`x1${kA|F66qB&oG>*+SBubA^WimJr#nKi
zGOK>ZY>sqS7WEZ0*SW^-`=z~6SXLtu4N9~U`@N++!-=kBO%;4fCMm=E3{A=i%$722
z%xP1v0LLT>&qMIwZ^_)*Y<j35<J<@*(z9t}4IYd^J@Cj*^2KA%Q;V8UNTxB)BUcYS
zV92Z}H6)1o9dXxnUiNYGR5Y^X*eBg3XKcG5EZmePW*d`8A9{pl`XzI%gNRU}f2YJJ
z_GgdXY3Z2ssp%0TQUwYN6@moTby=@uI{wu*TTWO7V`J8-c?GY9mLO^`WTKMtD^&LZ
zQX%=S<TTmg0h=awhWCV>vVW(DK;*NCRV4H-m!%^Wlw0M+=`^<QPn2V(;rO%uyDnYa
zaw4VlT~xBo@fQRHK9l$tuO7Btf%IsjOSy2N`20A)Y$>tJh;SzQ-b|UbXFA++%tm8#
z*vx|8C@kzAkf?-O!{OYN8P@m1!~kd8UhiBI!|nxBAkQCrXi+PF5}6K=ADPkkq)Wx2
zRV!{89QuKy#7jmlU#BZ!Vbi0D>1%~V>03We#q*LP!fO&m&KdaUl?bH0knadTt%I?&
z9D_}uqy0}Aw7e>{ezYVf!MRAgaU5w``7@%?Sh$VgLG2+mW9)5W__b0E2_CSn?tD|4
z?zYY3dYsFLpFs?s@U!=OMxI%;q3hpgBZ<(|jLu=P-+LQ8UbI+ihL1<9T(hB#aC_3}
z5au|~=j+Sn$_s8j*LP;!(<g;;xR*)o#3tO&QKjaJp%C%O^aCretYE$t_i9OV9XuRT
zJ*N37)laoCu^fK=RPj^^pJ1)_zu>Rn$!L;w3OCj<lhOQ)Rq3RLOm;;ze=2@_Dhyiw
z4s^JxianR|1+NYj?}~C)*2rh9`MIpovk<oaYJ+&etpv|p!CP6I(8d$c?}BgFvld~I
z`kTNu3f-cU7T&^icaThAdHQP6CR=%E0!1a$zMFKV<VISOr9};WU{cb8t1@gk=_wux
zkdm8Ila=+e9qCGw8>OPaP*#?#q{v3^a+oETn+2Q5R2gvkqj!*;mkq%dHc)c<hE^5&
zLk&-j5tWRBWDDE8sq3cXHFGRH014<a5xYx&9shyST_K(vDqjx@S6-Y#6mDJl%pTt{
zJ-Sv(J2<UdMdy9}ee*t?kux)AZ&|u=mEv4M+S=<q6uV4zENKqcwN5`<+SNO|(&OvA
zT+`7xI+}YXsS)3$>5$xHS9y##zaTm1BCFuca=n?2KNRe-G>a!(K4R9_?$RtmW_Jx-
z4lK&e^&@a5j1?3}ZYNx@56K$?>!_iTlC&I&7X^IvUjJM{;BJ&nwnrp4O1f}Sx2T&(
zb8%`R*(6hxKA=rK(NRctj_&j&!%=Q>zU5e2+I<Pte|oX-H6BRMz*BKZwNaj=`<||p
z+{{N)PMq(S7jY)p{auJXjB1%(w%3jfOf^_KQ)5e5Uz$8qww-Ht({q<KrpOxbp?Q^H
z23hGBBOau?)KfAu@4crD0<4lD)uxHn=Pz=*hK!dVr;i3S6}h<N9Cnx3kZs4G?iRLR
zNZZA(ekHEUO;NXs<J-9+Gn)%_%NV;7?9S~wu>Eq{F1Tj%9E=d$C726|M*U2(&b|Z-
zfyIql1a>N>Nezb{{mwWPiDfD*I~!T9#YiNJ8Sel=WFO;;OUd0bxd!-G5Q<YX;O~U?
z;yMhckr>8+3+L2!Ike>Uw-xQAv!wz*D_N2=8W1C-(fQA#*~EPrA4|X^6C=+m#?_6>
zdz1BE1bn9=Gt~~AR9>yBFG(-2$-b%PP^@JPU2%=W8pJhW_*8xtMqbn?jzf){w#*z{
zEvL|JmF;b9mF+gVoKhWZ@d11dio|8#kn-L{<hNysES{WgVDsb?RZo00)`~}p?J2A(
zIVZ9r@(R!_A-sgeVotbxNzyfi{80w{wGKRsFJqaQd^v}BE?!NXS}NcxaY$&sftR10
zh<pc{EkqQvSjZBV$coO*F{d*5a0fl3E`}*r@Uul&JwJbW1=D;Q8k9<cg#?8&1(I<}
z^LoQJ%Qi<ITO!zMIWg@7at8k*I9x~1)l8Ll@C%}6&w^s#Z4O`Z1d7S&06rzFX&F#n
zwIYd_vpNO_LOD&%i3uoC)`n+OA`y8bE;l~$F0#REQ<aLqD|2^g)1hH>EsK-dqo>AT
zA=nt|csaL+Ce*aHHVR$o?-`;Toq#LiJA!L!>uMV}mMvBk*4DLP@%Azf_bkiWvDehP
zQ3TsQRl%nVV1+K1qnRq7aRNo}heV9eo{jn^J{quoO5Pv1%2n!A_!zB~@hx8v=}upz
zya+oqDL!dv#^V=sMA&x2b_=w3z2#_0QO!W|D~nm0$B_um$LzxH^a3}VpZE>|HM%)7
zCSE@KoHlvaPciQwUy01zL+P<3&g`)q6K4V(eBkyA*o^Zgz}4t120^c_bU1PrL%~Q?
zqcco{LN0ux0pPPJR#%COpnq5hF$R%{j-VMCB8fyL8ElN%h458DK1&xnGq~?joa<&V
zwqG#kcQ)?jMsDsfbXdN21Xs*$SeVB0dKhvM<H$bCj=IbULo-#u%a1c)@~9jXCVG$K
z)W_P$YgdLmq14UV5{Va!zLX=yFnvPAi_g2o1DC!eRLz!3xqvOn%UETpg*L(_diM5P
z`VIXS_d>HZ<_XVm`T)N2m;VISitg|(2KRV+h4%duB!~@ikamPAc2vlBgk|rrii6q&
zv%`e5+gk94fPeoU&o81XgDFaWQR6>9j)YB6R8!=^WVyV$>1a^(k1`U6#57x42#uG1
zP0-$AZx3c~w<rtg@+`|H87#2qFH-vMV`<&Fz9oNxBuGMyq}^^^-$$3X$jHnW6Mz?_
zDhUoErbv=ABVQV62gs@91GF*EDs!n`Vk*%~t-9%BM@*ftZ_Abm`=*j*V5e%A8_`6+
zHxVY|`wJTj`F<H};yrqk8?j5}l!5u$zRPVBcJ%a2*wJ?Rv`s1V6gCvfJj#M$a#LWB
zQ+ST<q~8!a&8EN+r}kX7Gn=g3;tHA4rX(SKzP`;n$L^a%RzRGrwQfWs?b1Z3l;ba~
zAr4oEsxeHCgwAd^ye;)KuF8Hv_6PhKFEZfq7HW*$(cQ*>)?WUm?@s}fPF6!m_A1*5
zO?mwEBr2(bIq$QR9mCOkZ=ZhgsDOt0&37g~MZfEsP*N4}+gVQPe&w($p))))JfXv`
zs7eAI6>%YzaT5b2wL25Tl8OU!&QIo1gp1%65?MCrRFe_fR^i3Uqg`!AQwm`O*Dh!J
zi$)a=ynHS$Y>ea^0zkL7rnp#tQMwl-)KBpi;#}S-#vfZJ2=YaF7>P1By8zUx+}l&X
zw4|>*TEF~4PgT*en&|adqb!p|tVWILN^(rKdR;h|x9z1L9bs!_H;4t}t(kP5D4&KW
zzLH0P_^I;<sFLv>zrKOO77CB5uz4&Fg^Yu9!ljXpXro6)gsVBZPFP8a&F;6~bz+>%
z-F;hd0HOZ}R`<gX5aPD@m&l|2Bv}DkDn;z^l2yJ$bpLqYXq$`iCb)OmT@5C?Ku+o_
zPFxVaGgP&z2Ds2UGHHR6)VEO76$*K75{*}NNGAu&Cu=*JZj_$Zj~9&*PXLN1NXImt
z(Kii-Jl{FivPC>USTf(TrA6cjDM<fbQvJQi+22g(SiRwcZ3>RRu(?n_rUGM6h*A)n
ziNht>(N|DCaS3rfsAMX=A{Nr5;`)n6fsi5auYCmQN=X9hZ^b5y<9I~qnqC2O$KfrG
zO%5C4apjYft~?_#A~!(Kd;6d$b&h?T213m-z!pGdcx<95lXF-3*Q|1V)Dk|2h6^D+
zrjkYYCOPNp5~pxvCP@#vkKkbuLP^pd;EjT%Ef@ky#-lot^GLF|dA;9s>?I-!izj5{
zgyu+SLb<c3Vw$25b*1ayVoUKf^Mkkb!LI);%$RSX&3s}$UIA@~YseIY>3;{3ktK<n
zwEUZvee5XxEsV<-F+-UGJ|{~{u}3CNnP<jF<lE3}5u${xz)zLdSWWvE23%^9KFfZ(
zsQY>=8-bTYn_=o~3ev=3+K@CWM<`&2vxNc<dzKg1reXHziADDx75`@z<04oN*cW(d
zrbwxO7^>rH8QSKeni|G{msQn;K2+*OnQ-YB2aIXqiRjdLs)){mXs(1YKpajSV8m6c
zkbBhG7v_C(o?ZEhZ%fBEHojlqCD$VeVi+(<ltIre6QtfmO0k%WP(flEs_WA222IAn
z1Qg~SI%Gs^%1N2r!O{WUL9VbBMyWMQ6gFnTXnjfWS{)<Rf~<&KdawY(L5GoF?mb%5
zpsvj3d6JiT>*32-E_*tIWkJ)q-~vj6DL;7Y&5V4Zr=~{e$<Mg8B{*sGyo~-dZmd^6
zF)yYNCzN4m4Z?rmdgOeh9*)wmIc&{@`9FLFA1HrFzptF2p-zi)c>c~&G%k!-1_eXQ
zh>tlzUO?*Ggw%w$si@zNt$(lX7In2z?$3*swWb^6fAnG9SQzPvfts-l7V>1-m<$7K
zb5kSXT#u)(flFpAkV^carXF6A^%aSKzvo=~P)z#JbNOh35t;}!5{NrH@r~h|3%y5v
z-S~~mCmhMvK<2?DnfMeTiI;-5lrPf^bL6q3vbT{aKleM)v!#^^Fr36Okm}KWhAY%5
ztCWMY_tTs7f{apoC`Sr!qt(mpt9+N##{4o=1INx8@Qha@UGlAzJ81Rrw)jOm?44uf
z9aqWiBVO#7gi~F4c14vj?c4lS0nP8A0@8EAi=uL;2c3`Rj@7I5W4Y1igQnAdZv^M2
z1E||o`#Ev;jZa1c9>B3{O~IzMSoj0L=qDTPah&}saqKtVgV^|+cJCRvQ442JZ}}W6
zpIg$i7P^l3*hxiP8NY7c;ai`v8)sZw)JEtx*n~Q}#e$+RjTZ0`EG!YrLWCayRY?WQ
zan`Lx)@8VY^$~0lCpu>fWNNuwh=L%R&xCzX(WQKc1x=K>C)d#W7MtKanMPKs;lU-r
zHXUtMdyifFdOV+Jzm`vOPKy6Er_5RA01tUXe;9_j*Q!X!BDx^ObV1t-40$sek?>0x
zt4|PXNCcYL7~Y&J$T)F`S`?r=<cdq*KK6~&_p=2t6Tj4T9fmEQoh?CMvT6`_kvb9y
zsk^H|WQD5jKQuq=6kH6u3a>E|Mk^JB7LBnA&!$~13S|mmwr6A$n#Ds|$qe#<XS_;F
z&gGqMRax9dV}V<d!zG1qksBkxs4-}EY84ua-^)P(1!N!2is3<0AiT6R2rnh@n7VjG
z^7hHR5pS0TI*g=-#lh61VbI0vUo@f50{;M!VuP_FG7J0&oDiA-PN95stbDGfJfx_f
z`ymQO3xL@ohUC=+nE<MVs)c|cQ(a510}cfPdffP&4a6KZMa~CFCCGipzdsdc<Ovi!
z&iXV?-@?>VP$cL&1O<a;EX5hs)*_RMMfGL0ij}bjn}=PK5oUAZiZC4TA20`l!SR1M
z3Jf|z5FMmBRe_`9#`8Ds$x=|7)&y7(V;}2u;px5^ukZ%%<Q%mo=GUo~eY$-Qoe#&d
zmVyXcQL<)AM=`GLVvC_8w!YcWo_f*N(r3%*HK=^hnSaB#C_~^ry0mj`jh_UtMGsT-
z_m>YOKmeE2u-h_m-s*9sKbemYSPz^kzRHot1X%m6pr7jkS`HM01Ga7(!N+}Ax{SkF
zrVUFu9LrM3Vvc_MZ~729C~*k&*qfZX>h9yIK6c!_HZ*0AqEU;YpjEt<$JXeeagG*Q
zV)*6V@}WYuH&nK#GTtz1j1%LV$oP88R?B<Qgpj_}y=8m#teJ)Zio+JpU~4As-YI{S
zxLz59n+;34^M8>BP5S>%5-gj+jrW|rqxMb>*17yQ(BN>L#asDDYOUwNpf@d>9oD}{
zJ9z&Gi~jeN6`k+XrPQ3c@*C~kVY&JgVXqu3Ryd4CtwTnUG2;b*a2}71%P$J9LJs`<
z_6`zCPs60qQDXr*S(Dl|#S$?JFLx?_o{s)EAWzC-xpTxW%w&~{vB!&zMZa|{kIlSV
zE=<AWg(>A%YlIP6hg}u5@yG_eY-}_$HV33Qv_4g-u^CAaGu5Ec+p)Q`P_eH!E&5lF
zKP(FUO4$v@(fdmo)vTz@C$0w?f1fkQN~h;ooqbuFKv%jc84zk$&Z6t)fPbc!r~d@U
z&ri(?-c8<ic(fE*!4g>zxp}r!79UZ_$b&j^xN;PMp`do?QmpG{_-Pv#Tg9?cV*4PR
z5ebD{A=wxtOZo8H3P)7w%<5{5^_yIU!TVgpR%>y+e}65<B)wM<Tm}v|#JgESCWz0u
z&k@ArdU3>*_Q2My?SUyIF-XL_V<P46&E-ke;lf9%XNG60N8v)LnwPsb-$c5@6Ny1P
zQ)<pAE^Hre-k#9DgwGZH6J<UF>De{hyj!%1euZ~Q8t0^O@G)?YNpUEalSV)&o|~b$
zs>(d1lGDY@Ak+EzTokzXps71JA_V%j<m|1Ga(78YxEknVW{Cxf^31`fC5b}RE>^0s
z-0cc0Gn8?IUF1v&cihwY*VkpA*+<!_rV3w4M7hCLk9}u5?T$!Xh>x0nGuOP4XWG4V
zw|VT&+NSljA?s@!cGiwHw-Hwxb$)&A&H&=^cI;u9;-oAvKZDH>s{yOT3K1?@Oc5t%
zj!@0Kt-8H<XaPW<sxJV@hQCm~SW-UBj<uH$+d&EsN^T1CaU<Hf7<PBd0>`dGP*@c=
z)SXHW^>V{oyK;q@rnLV?6W)Q{S50Zt4SHdCsF$1Y0087<#do(w1(Tp_c<U8eBNUK~
z%gv|~(LVU?#$Sw4g>pA)PmxMNW^-fhv{uo1nHLnhiuRyZ$k%mEu!odSLVO-QXns#u
z(s_1(_lp#(5k*v6N+|sI708$SSY}{)fw9oS%(Mo{kKfpk%Fd0{t{%EM`i~=@CKND}
zpe_XeG^`~aAFpbD_h(YqftV_pkz=ML@KiTac$j%uFBT8K!PBm-RS#h)7{j!5MQ-rL
zq9NDCL$5o7FXk%f!?X+qD|ARMzdMf9y~w4_&~b<9qJwn24BftiFvE32eB~=J{I_{9
zJ4+fP0A+0N4D84dfQCY(*;Yew3gm2Mys2i?tj9nqFDpmw^pur3%hWl%EGlqXr>2wU
zp)obglOB&4zXEAPuE6Is5wZwPEAJT&8djY0=dlae7(a%17~OIL9Vpm&fj;pxoiwrd
z^qJ~KZ46x-!;(yXzTbwi^)Fp(cV=IuoT{Yq@i_B7svXi8F^A$fGuFQ&)NNG{>ptOU
zKDi+skhaXkf2w;Z`qOm)>J;VHj^45RCjo*vBT7vp8&LU*7!z{YfpKY{4~?QhY&5PO
zO#_B=qDzgf_d{iZT6g0N4#||rVrY5f@{|+ARiub)WwH*!PmlLI^pdG_jR26OcxN)v
zMUwlN+=|kSJc2YS^B_9)FH6ab$Q^<vFU-M6;G-Elhd=x1)lelu*3-6iLSPDTgxff>
zb`nP$ZGrcG++vF1h67<c-^-euDIaaDEjjdXVYJ84@<0L*S9%)RQ0L3ZnWcvR4JS7o
z@*g@S@~a(pVZ<2ILMk5@6Qdp>;iZulntqy20dx3srl`lLM8?Z^p(ooPo6_{Abj`>a
zAm~BmmR_}?cWZk!SoAk(DZLeyoRn&|u9(2vc;exhvHr1AlT=P?E&Ay8Ru>j|E8P?2
z|C;W69bfJwIm>I;Fu1jn+QL8~7{iO8nKMFEAY{-*9N>GgLXwuiox7UaAZhtsMXeyB
zh!xbTkz;9F$!PLc+Sm?9PrfC8y3?$SsIW|&MhEzM&-eOjiZW<IeMZWxD_pgfX)R`4
z{tk%t#5dsU%rYM-FA9wtA04xDS^qi-HgVc#SUM&!eOU1UD#0yBbz<8>CeB^G;GfI)
z^$)X9-3glMZyyxh`$5g}=@B%!LZ-2)EjhVV6$+8NLkJ%xXEX3O0RY<LG?jj|pYbz3
zh|^$^oNY4j=#_3Nf~=y`06pBCBq`R47vhL$Yz*pDN=L?WdfNIl`tpxVAs+sm8db<s
z4qZWB%a4FgK!MnRO^JB<md@!9L9w$&L~htUX*l08ylhrAsvc2yGs57QHgzh0t2B_g
zWOgvEVxYnj%;u8AP1sAst;9=M3YE`wbE$mc;=1e;{H3_&JM*vky7??0i=kr}2=@{h
zXViPIAJ9_&(3ChwqK_>>RIr3rL~gz<^@zB<VtekoWHFS~(YhSAeUXc`)<lEIvKIXH
z5;)wNH)eyoYi0|@LdL7J-8(0ddz)h??TX1QNX?vdR56>(AMQtyq8=ig93D4&#5_q!
zzfpZF9pN8f!Ios={{ztE+tC~$qJ+spx>>>sdsM4$`49({8w^YxljUP|>AL=+nyZ()
z&B%8T@l8>xsJ9$Tz(2D`=*f8aptEg3K7g;P=469UD&_-h3o4<@wc+2T?Yefb;Lo_`
zT@L)H_3u-5vp1}7CEs5s*)g9K;+`r%)%#$*efHc5mo7e|OfHF#nS1ti5{#XAbdM#`
zF^gcFg}=Mma&(y|sKNrQZ@|7!krxy>H5H7pp16ve2#~YF@x(_z#+Y)r=5B8eAbv~d
zcI9}-Yk7P<iSNVRFWAt|l<z-~&c6I~U9uRs{L=4U%Ght-ZJzh8!Z(zner5#(t;GEm
zr{it9+HNd<KTO?D7OF}xMNozM>m^<EI*mRCgBe3l)8%AvFr|Prg+{4uZ)?-Gt1Ku(
zke!p3F4r{IGCsx!vDyrH%xoFByi*V#pj^hw53H1?_mX?NCq;6F01r8YpdU`|GZg}6
zk=OFWVKVy0yWLQf$Ce1^nR=%i3Rc{3gq6E&<+FUvJQGVL&t)bu)Y;}()1nJ+w$4)Z
zuq|u=1!|nBg;U5{cq`OMk(0BRrNQ!lZsLO&+<diDZ^|{Io^tYSVmRtzTGCW|-<=S6
zCY-O6N`YY_s))&?ibPSBi=t1Y3C>sQ*0>u|1(P?d)2y4*KP_d~^^`{kfcePiqR_b6
zLVViu>u6ZBnMBgzCYYd(Lchcx_MP+FYMtXBWCxPa<y9@i(lQ7_Qu1K^S~CJu@PX>S
zzFPCzn~Thg9;`J-fGZ7Ff5A+aHzm-97_EqRpkFR$C!Q&Sd}RLYiG=(qy@I^z5#@VV
zWfzWDXBLW$;*_(wggu-+dR>N-6EqXvoHF+16O!t7kJ1*ZceKcviCIxf%-9T)gdQh#
zViF3PokhZ;0%NBJ<d?B8oM%f@#z!fMAxR;`C?(M0$~5E}$}w`>{<0Tkes{R8q5iq`
zFY4c_4Oxd|tYF?odMf~ZKeF=V^L5E&XN`S?%9lhL>@7^DEa$!w`{(#o;7tH|qPc$v
zm!EZtYiT%1pDRf=&_62de`NYor~7tEV{j2>3sEb>vdRu5jn~rFrbEW-9H*tOJD$ZS
zV)^~eae3JSd73C^z#fdWD(3|1b!OBiWM@&iv^*O8iW6)FwT|#R8HeDy<jLc9^z?A$
z<diAT5^jo&QYy;bez!RzWf}SMoL+-DDu9p0PD5O#qoLLpypmg9NRx8=UV0Yn1NudY
zB$$kx1yPyRFiN&5<x76aH4ZVHkV>GEY|h0;!d-rU361WjuQsl}n7}Bzdog_K6d3~k
z_Aa-#rl`<c<L;XGesZ}J=PaLGlH7QEs6bsj|DVd(%i!2){G{yo1e2k6cf_I+rB)14
z9$h0_Qgho4w$wU2@ckL)?8wM={i*Xl+IVk#%MCclMCX_P-zrGTo*%$y@8n4tg?lWE
z_gIPmKv~%IH!!KN?|i^-e`EK{;!fQz{$U674_SQM>3YhPt%1vi)z1i@Bzh@@)-%)&
zEQe#qGS$#}>C5!*@9-Ih`%|sCqb|a|#(alJl<zQl;TK1xl~29zEb38N&=!?JWyuk;
zrSyz#y*_(pM6LzRk|XMrqf)ws-YT*JusqbWH1zR{QE>)VFhWeFXQ>dm9GHvAKoqkC
z7IdzWE|Ty>BEE>h<Ot|`p+H>j42tJVMfru^5x~KN<P>4^Ms(tkTAT$<gl0h(&+WJa
z`3+77|Awq+07^i$ziXTj)M|<9_0glv`4<qA3f2_%>IaHvmd`A%eqxxdw1CI0a{az9
z%e!Ap+CSk*H}L#-(>WLYs`#4ynwWmoMcncIJh1!81k(QEy!)~T_TA73Lk)Sjkl7_m
zr85@6!{N0VlQqrQLRdI7HZMnroWrB4y%dADQo$xjUjfp#k|j+hk1nX{e%K7)lkzga
z>+P5hTafRvqxI6Lx_yz~1i*&x!#NE!3kUHo+qlHh?`s2@`G*ykg7vhuK7g-xsbVs^
zYiQX)M5=)mG`{9jNX@4PfD08=5HdY;dS*ya0m|jm2l+1FyzB$``fv98>I^|!0lwbr
zy*}As&mPNd%idle^SFdAeJG{FLsTABh&c-Q^&%8@=G}-grE)~syE8E8#josN&R;**
z4|6VvzJ<bX?|AM0LV&OLT(6Ij%cPtB1h`5B&)L+o*S8J-f=0d%JH13C3>W0s-mv}m
zAi$r9+ycC1s-f6o+wvcoE4zByErXjUX<Syl-euLiuQJqNj-imKW0;yMTu{kosk$(W
z;jz@c-CE3AOzNC@d+9q$03|a^f_$61Zt@O!6Wgrhgzd$*aVLkdg>rUwA3%#+;O6V-
zR@k72f4}8v-TZ|LLi`&NPatOlPwsDvNI$tD!8(kud*}kVuIsZBja_@vus^oMm8n$m
zWh;JUL`U`vpMrH`rws2I`Q`7SPP#cimA`ke2r#G?i_ooSpQR09`3@KHMpOS__?oTb
zF_TijDU&ecTNHguoEiTtEv)dl-4(vV@Lx6pi%1}{2(#X}n)~KEQ|*3wq$4*kMVwF^
zF;;uZVV4h<Ei7B;9rJNo<5EoE)1~(B6k1EE<o`KRCxREm^)P>Wq}V*12bU+sdYC_Z
zQl|!=pdaD?sjQ{sQ9Zj8BTS>%8nIAhDP;)g<e02(c9GMOSCSo#pU0qz(&mU6b(J~V
zng+KCEfS?22YxDTYfj5TD!$rb4nbcvZTRt!U%x5>&#zAoai1SsH6^2dR0wc{K6uLe
zP~{$`Q-F?Vxb9bG``fy0tH%^a%m#UoJmccsnJ}k-e0QWRY>M+jOTcfC_8kxS?E`kd
zY^<XOB6<<m+>)N&I1JtF#njwW9@+%zgDEYabH{{ME+03JvCeAtn;Wz^svKO#W&?)o
z+RS>hA1Hm6tKob3L*+B~_F2E|&Z2<de%<c3%LlzJOM`(+xe-`~KG_a(aIg5UW}bZl
zLvsvff57j+^mz)a3f~M92)BK^94IMt75=nr{0{|P3v8Xv|0A0UD{9I-75>8KcjgDt
z+!bG4_ktty9i@Jc2Upp#W|5V^Br!-#LiSH)Q@|5w)%N)#mwsX=R!2MIuUb&VqJ;~L
zx$3+cvT#8puW`tWeo_RUlZ+K7v<`I&D4u`^Cm3c+Xu^s9%QW&*aGKZ|RaJ~VUa}CZ
zBYR^*xzvl)VYb-bksp6<n22&Pg)`rV*9V&c<`Sc`yryVK;5(SJ?6PeLN;JwbIg3C{
zBNDPEw@e61npr#$)$8NVzvR}q2U;yrGq-8BUG_C)*4Qo%;7!bTbn-=5n-5TA8N3Rd
zBS6uN`T}>kM{7|DrY+toQWyEPxPgu=)myG%C$%SmY=N9ddK0f8EOhht2a~kt)IHF%
z@72I%`HzezUxhw1k@mb>AjhFIhfXeTl$keEHd~Hb783=I%1<QxG^ObQn)dl387OYM
zc{jV?1MpnV9;F<WQFqDb-Tw>I3Ty!1&1;}r@}u|6E?6(0kM=8jX7^09e7USE@3^*r
z^P*Jt`{oSM7j4H!@_=^U?C8w|wcV=M+pX#u+Y4CAFn^4yYlNO1y<CTxpN34EkHH4o
z#X-Ky=Pvtf(UP3~=lXq9a6yi%)2})<XW!U%_WkMi9kuH6p9u+p^|+w2D~qorn6_;#
zySMmW*_c&MUukq~kysuS86+2pS)*7Y(GUrL$osI!8x`Uh?wM7+@@}3A1-oxbvHr{P
zpjzQ6{;A*|Pwy<>iJ;Ry$Dq1h(|+tarfH@%`|9tJ)(ItEMYOPqJM;s^9rv9Bz*Pcx
z&IV^udOA1zB4NUO+B%;AA3<{MO0COeFcmst65XB3&bV<PXX2S-2y$kZ57;oXc;*IE
z5UL)av%%;VUftZ6xzIb9n(%WwW@XNIrd~W?JV5u;9me|qwVl!N+*}v#?6s&=WMC7-
zA2|9{e+FF2jm%`|lNBHb*V*EzGB%9bo*kpI>T8qU`5{-GdMHOrp;0|ln$bqI9ns#q
z;?|2l6ucV<;e^=pemoDU^H$p;JH5VJPxgeG;`9Mc{gTlQdiS2q&L&p|DskBh^b4@f
zSbt^Ml_73U!=rDZ-3%uOBTdoX#&+q~cG_k+e`Dj#8TDsNKD^NJUxc?z+5S1nF0*Mu
z=*E#1=g6L7RD=Wfp*G4Dyrg9lG&0)esVOV-)YuxUt(kZdgM`lv``L0wPrkllin$IZ
z+TR;N8Bf)0&Es*NNW}HqwvYcz2>!V^t*;Wd5|>o4wXl-9l3F=t(M+U8$*wLP2_9KG
zjb-z|VN`i=gIwXk<<XwK%O|L4F0`iB)#w^LGU(j7d}nRFxqOwd)2sgs9J&rDc<3zS
ztR4Ox<4N@nk{1vydRd5J(7WS~J_I=ZzYJQFoN0d^5`XsCI}{5IgKoFx+xse@y>U#g
zqbyATTBBdLZnkBL4guo#Ox3r4-|8aQJ+O75cU*+_A1Ze*DKQ)*@!RJ+7dz(%fcx@b
zg@y`W6rpnl>spuDj*WZde{%In9Jha`9gOI}@l=vxyN!8O314KhM=ROrLpU++5ITIq
z)WVnp;Su2nV%Eg9Tf*|Zbr7C6F$`oI%NCG;M9r?S=ShG>{Ot+5*C<H3!OvlFgiI3X
z+YYBnLpx#yV~{UK0^)9gDR*)5M`X5)f=8Vl#L)j?%$huYJ~so`bFvBn4^IRn2@z?W
zM!rP6o44Y5@jD8GB9<e7RV(YtyPI~0H39qBXtIczOIJ!HG)@L&#tkJLF(exRj!H?o
zJdG0)b@_b*fhSU^_-ToP>Cja^ah6_pg3~<_L*_E6bd{K&2g5@{p?Daqap_}82Wcu_
zFe#BYg}}t|#Bv4!)te4DBZcFHR4x}Had7DwKzc8Vkc}4Ihu(@0i@yaWPnf(eB@`3^
z3QbXZ_v7X5HGcLL5Qr42dF3)^AP#76?VeEB7w*#+iM!pse$ETR>Hk0o&nezK0uVG$
zU{?UOdB*JxTW9)HY|s7yQsWkL3(jEX#lA?Vj~3-3d(*k;y~wZ?!V+^JJj~W8?6>;9
z?t)u-?3(f%RlFwevu#_C&v>g11Q@v>A?aM(Ee-4WL1_MMZ2aRnOcsx{CGVFR<IXwr
zGuE9Cg`A2DiATdyauKjS(V;Qjpsasx<&7iI`B@q8w&MW61{7XQZQ^WR$(?q=_Ah}-
z;D|;o29z!;&u~+hyrS&JL7)#30if4O@jJ?HMIQV&e|=~Ta03)Y7$B2{{IX0pTOeVe
zswroxJ-kefVzp*s9U8xgND*=OB%e<VPdpEWyqSx}H*ndlzgUx_m28}w&CkIi+nxh9
zQel{@48~OqjJgr<ybX!XMYB$W_r`|C?ghgWCe4G~jth;u4H;<JnF<+pmh|4h&4)WB
zT{xhHcY?m_IJ+TmXV|+QxZ3~=;|?Dt4me}3pAA26Ndu)2{t?c}V9CkVb6oJ)txecr
z7jPFaI?6v`d&6|(W&R>x)9%To&x0&LvYl)J20fR{?)BdVg}%5alaC_;t>^6LfL7u-
znLKl}{aLEZuFCQEZp3l(Qp^pi^qMQKkL2a$=W{qiBoCpiFw;ONBHUMr$miwp`A!;n
z9g4|a26#TVfX!IOy|!7n5qm2xe3^gu-IGaQ46^{fM%cG4wE%`am&xwk?N3jOWzX-s
zbelcm>m6Rmq{Zi#MUI05EIldq))3)=cS^P61Za@TI`13OYES9Gj=(+`NoJ<2i<k!M
zADI~xtbmGHu4diiaJ~gufyI4`pl=+`y{nWsFM_Px2(b18Vhkq3r}W28K#-!JmpNHU
zTD%;YjhDg9P-WIg`$#%dkPTh1;W%t2{%sZq1<!;X->?9hUFPYR-k&U}(kz4Fv*k#9
zk~V8jy^06^hl?6jL&CD}m@Ek9`?3UcrV6wTDJBw`TOFj#hRcNmH}-?-1dPV=nAK=o
zl^HanV$T^M^72maGna0Ry|p1ORghq!OVaAaS~rf#bYmsz^jeWkF;_t9sMgJ+K(t{z
zSUQ#mf1PBYs3_iuS0rnhnWR-sGnZLOuOcX36(z~=ng_?G^I%~REoEM{u0ud-*2*SI
zua&6XSSHJj6KgT`5_*CuRS>t~R_w;o#|JXXmc$CFz(vO_e>3?R0=@G7R#rL^e%%6}
z839>5@!4Z!IDF;O6G$C0A7{!Knv`(1nMoK1e#(iy2Z{yVLqBTQQA6nO^7;#QTX2l^
z;LFP4@y@onP@7*^JmXbXR027;X~B|F^^NuA+ASYc0%~Nec@^U%*sY!IO&l0!M<%j7
zItAnAa9~Y~#%;~ViSxwS#q9+tsPf5IBO|X)2CV1ZUM$Ye#!(=jqP>u0X3QrD4Ibl3
zhOnYP_2g0^)ZEUpDMx<BbrscfbDABH4=frC$zy$hgvAhW)F#Mna18i1<ne>&h+P*g
z1)Z9R`V6T6o*IXwo!>fdYzaMDdp^06a`#irGjPg7>OW)n!lnk3!&(UxaMGEoRm-xA
zo2`XW=$Bayg3?+Cyw6<gm-1?&`jj+;gYElk?X=a(1-)>1?*irOX>0%LgHE?#x0%Id
zJ;WzA-S?9Z<y^I&KuJn^IkvNc?5@e#-T*tE#O?s6tA2(=ue^hj@bed}yDYdDXpGvP
zNd!6b;rKcBrs|Gk_-UTs#q@3*8e4C{S$AEqk7t<2n)CMxc3E)Gnp@}CnrZ}>FS6+f
zcU!71jx~%i9DL08S^C2I>Gk9DYPzA&?wY*utKfz7n921P+?`qPUCj(@hkfp9`k=$D
zGnKP64wV((wC=E24N{Lx--hC>(E(kv>=L|kMe9^sQ(4F5`?EG0R1$nlb#^Bl3r%?Q
zJR=arF6(SzyZrU+BaaP@&T4k#<gn;Ks=IIQ#K?J5PYmDH6*0seY@dZU_-*B+Fma#Y
zkic;t4>=rH?!S0XtCdK)dakoej57yZoNgbJUpN+~(bL`4_o1o#bR#*7Xt&<Gp~zWt
zvNoPc`WyX_`D~7X3m3TiIe~iZmgrbw8rSWNF-Khx`Aq2NB@tE`oQ9)Tr!9gWBNIC4
zVLZF)=&=_*-!ZD&pKOwL$bPGvYxhGlgo8T>GrH|xaNLgHaW~$J-c-X)(lJuiyl0h4
zf@X~$dx(}ab@y`3FPI>iG{`(?l7v)ChD9r)!z4?j(S+oSI1ydaCFv^}V_-2mYq~u#
z9xZcjr(}|FI7+eC=kHXO3En{_d1=%%2=tPo!4`E?Udz|1zP~6Wy1mzzsF$eVXMT3I
z(`>H9>{x+W=X-P|s#VFw5xv0xS8?{Lpxui+vvFVqSkR*F`%_gy;*fK(<JzLSFKj7U
zb0F0}JWJNy>*M<MQq!aPzMHHBQp>9twpbi8fW-f8D%pDKo^6B}F0*xe7Tnu9AN8-?
zGvYosi}D=to{tuZwQ7$QFUcJbV|tLU4(CPqJU%+TAJ}VqGrdbQ&e%uxF`Z6WZK8ko
z+~<a*0KH*t)gBYi(H0&O6`(iF{5yJLoyA_;n}wa=Q)%oY``Db`m<qL@_+|S)5PQ}F
zGaCraTUhHdg5YWJ5+9Lu<Yu;{5)M*!e5`VP>3AF{UUEPm!{WV*b+dKSxhOmG3mr>9
zrxSYy4T*tZHRxXS_lfwwiLv6}^}oXyqkqQee-D19XXbVFd(~5Y^1{P`bsrh6H>E!&
zI8pqGS!qY7+GE8_fFjsa*W57Iwb@#XhX@DJ0KK6A;P*cCt$6S{V#3Bm>;K@HfWQ08
zo<!kV(mUb{JGMwBNhuNTJe{6RTJ?}(TE^w<@x~l$)r|YoJ4)P%S&B#ri;%q5Zfm6<
zjkQ-jYW5#?={{^ss)174lu;Kf9{aLK@6$xBkXQC86@-rSD~lqwBqQdd_2OE%&(GB(
zrTX@#;GkdYs-|iPUW+E^#s%@ptaPc`W95}z^RuVR1*%o@`lj28^iEySH(`-j0>=^i
zZ(koYvGGjyY5FI0*i#6&At@!qe{Ux3$CO_bEd_mKeN8_SG=wpDO#;LH+)H%l5<nSy
zR67(q$%DL4Y(L10f9P-iJU%)d4(t*;GF_2ooY$_ht4(b!vudMPPE?+?<38=(Fju<`
zRMe>Bga4WsS+s+u$2hrP`!>w$4qgJS*)BPWGG$A;M~D1T<t)9jm!Jye?WTLg0IOPL
zaRF6#(o-$SWxs8h*KR=$0AXIZHjpvZSICbUPyUa_IZZC#PQQ?+(}$Gbg0|n3V<k3B
zup74++M(+)wI6u%v!IVEa0fz50R4`HC1OHQGAq5ir#4ppe>`E(g*Al@@7`^?uy1w`
zCKv@q_-A{yt2BkqKpvN#LNLyD9}P~yy_nI09boWq&cpOwl|n<uQAetP5nkI!RWMR@
zQC#Oj4lzfuANX>CJ(vx~(kAy+UP2ozoM&yuENkE_u`TIrTRULmwse~}_*!{-Rp6ef
z!7k99)-B>&E`gW0yRTwL*>I4<Y1{)>%y{zpyj~@`bs?F{1NN(Q%F<7K14E<eYy%5y
zm0b7Nd07dY7-T>?l8}d7C_x0xF&Gel0ThfdK?SRFc0bG(9zy>i&O-^C7(UAbZa+ox
z0V6`mxmviQyge=egy1LK|BQeE=Ya!ZFXiRLsklGJXzP@qI3+oklJm?OG1>aBMO-Gl
zn=uqO6!Z$dQ=+>u0^16t&ii+g@C~?$%qkaD!gPS<dyn6O-s`}L79{dO0QwvB-kdsr
zq=pu_2TmBzV2+N{CPueK;Ose#oaXKu0m})OVf6mUpY8!zD9deL)4FA3>+=EnFjS66
zJ@Rk*Hd~(bPRgTC+vwdH9r!fpz4?zYz9Z%Fr&H4VGhV@`W&Q`C&hZ`NUrBlG(?VTt
zU>?30Ug!aWl6rJ^<}NbUi%y^f97(v6rQqQcNF^dZmxO|nqH`uP@Xrzle{2JEr2JwS
zAAW~@Pve5?@An^vniKz2zWB1ecflctUMw^G{iomlq+cb`z8y$hfG2(X;%~BV_W!&4
zBd-46bO0oTk${C40RS*G460r`*i5enQbuQAWg(ghw}X}0GJcy}NABqW-gcA?EkI1v
z#Xh27wh>~~9O;1p$i8M^TAcg>6YiWQ<LDy<+C`8@EZRWTp+AA^)`LPK6$278CceSG
zlErW=VLnJ&(qxOe5NJ0M<PrD{3#Jx-0<CV-gMv!MV9^Xre1mf(7->pF^5uhM(pt$V
zWk`iqn?uKkInzIr%@!sCIzDuZ;DbDJL>q{jYp%t$gnmdd8#R#qWbTTEf%D0zJ{GP;
zBYaYhjfKFyw@ung83Ju$d-7@)zoA>bqer6HpAbJ{a1i*l4_T!gF4g+Qq&bW<NQJO1
zgTzWu>O3ckxnz?FwH6#!Np>O7E`mH_!3ONySEK2jWHqHJoFI$F!YcfRPAfrZC6Tj`
zRJb*%m&eh`3V>-;Jg4n9X$Q_@RzxN`9z-k^NuU51fEOl`w%U@iEQ%0lHxcC7zy|D8
z9pZ%Xq{ccy7LBnAzhTbGmwRbw6h178+IF%q4j;P3JUfOBJ9|_Dcx3v{oav`bi>DCA
zJ=T`%DJ}Sj!8&5$uHn0Lw{etHamb<zz~Q_t;1(_iv!rCUDd?*hI~2%Ql2O`xZ8D8x
z%J>k^Qpoc}h|Ej2rTGzj#Gr9q<hHDoq9E_IxB94CuE?uqF_xs`uI(4paLGsyuJ9TT
zaV-_bk$K6sT+ftNe8k|2Tf&LT81?+pdfaAS_1ZY)D1d^|W;FsT2wajXc!IA+HLSaL
z>cW(SabV)kZDPSk{K|r;+AU=mR~(_`l0h)=(IkS<ZHb%(DX~IacC*Arf+~2TG8zI&
zTD8VR;qiz?wVKSLZz2<XF>EnX$rQpkFi&#}KH`@KfoNQ5v|308Lk2!-q9j#7o-hsg
zNd-LK_R@k9c`T-n2wW3TEGRTNlT>Ge_zsO^dmVs%7nBmJ3j=&dCu9_@9RTzK35^h_
zn^Iq`h)_(mvrbImy;VkSW|P>5j=H?saZ94mN(kY2*5X;mhh7sw9A6?~ez0eEKhgwe
zhT;eisJMvmz+z$LjIk_Olz}Dzc=ax1*3O)xM{-$eqq_JWu>V9~lRDl>3zEA$sonh%
zX7&Tfu*uMvi)&vS%{;1kG>%hF2MC&tE@S~mZ9D`N;?(m~;?YuTg))aK2og!mS2SU*
z0X$_EBr^Ytdan}z%)kS3vIu3ND^|-{9dSIOPHa8!pz=oE%kPqJjd6kRSR+E^DWyC$
zOTO!WtMKK$WsU!j9iDTZS1`&3x>t`V*YEU-qp-*dpXh8o_7)s%!1b<uowEnnhl8yt
z#HQ2y&rLMlq;Uik?PrN<<@!}e_@naw_kXdH{G}xC^SGo-c;)qV?|#2CeC3}4_w6kj
z!TqKH`ly%}8F?w2+PB8EXDLr}HRnhS1xL_@e@U!#yr(T87xk=7lTM+G>P$aJV}1?c
zQ2?NJ7_!*R9C43QFdC3*s~piG<tXY#7iy(dDnqDthc1Z~O_f+Vc+U_TB0cd@PG;0l
z*bQM!oDne&Zpya>N@g$A>9?Ym1YNkP4ER)Xa_?`afBF_3#8g&WSfaM7sUE0Z{ZoM2
zneSh(?Q@|UA$)YLXGnK?Yc7-?r=OfakM-_NW#ql6r`NDoQ$opEz?7hY%;6amXs>uE
zv2cKsu+sL3U-HXbjJH7Idpc7!TG3v6FGu8uh$!so;dILm5Y_C!2=9FBz+(rUJtZg1
z+u4}|Z~v2L$wT%jue=2*9N@i&R$i+hpK?dm+BrnqBUHERF40IvMMa#oPP_4SjbJ1z
zeps%AQ&aE_DjD%e2JpFSIbr3-kF}0Ie!idxm8l0aP|nzenVA91C)w5~X3uSN>NtlJ
zHb>?h+y1-ko|7x;=>d|Xk}_n9S@<U#%gE2;`iOJBc&x=uN(h>91%jTSXrGAOyG1Ji
z@>vRk`SAH9>xK(|nTToTDU|TU|D;L8q}Ui0pg}i>$W=L|&<+?xFbc#*8F`}~*~FeP
zwuit$_j(Ue^s3&j?`u;E7OcRl5M>xwOr>Jqt`y{N;p?ob)bt$1Wju`M2>>zK9pW1Q
zNX`F#sy_O!^D$!ADLo%hs2DKRTA%eEaZz+&=mFMg2w4UD6MMH#^z>Pho<gCjZ8Hg;
z5kT$vdQ+q(g<aW%?WJ|EL62cfv<Y0D^|^TQkz6PhSfCgOg-?(4TR8wj9A+7?9BksZ
z0UX#?bfwM-G55)`KnHff^#v~th-FVbwq@6T8gt>jiIntEYl-jAdA7jZo^5B3T@k+M
zbGkZ7-Z?8Gw8<GmH!`BQWdTF;;x`hh`OqXtH{%3d2yEPcw&jEb>p`GuS4e>gstkq8
z_5O3X?D{<~<X{%v8;J?u8uwgxLVFAbuF^Bf?Dr7H=fI$)9#)^3CJNq(3IU6DI(F;|
zUY%QDZN7hfw{)FMCE`>iC8XOfCF3$6UxXFMvt_IF1SJ$uhF+)ys~wo}ez`Wc0JaO%
z-P*|9ZE8Ixr!B<-6k8R>Y${AF)l#OON0!?&t&Zw;t)!iYY$yQhD(%OTAo^dYk6v($
z^VLAwNIcfAkexa`0RnY1vP;2E;(HgpD&NnF=B=18wkXJ0qw5V?wN^ozjOJ%B!pmM0
z=G_l^_8Ku~F@Ac(DvVv%f@WR)IYc$&uR^K$8bN!e0H^yh_2%<;5PZ%IZeSa2kGu@%
z4&5NIK=%t-4Pd<$Hh|*sF;|mXwjB$NNN8AZY{=Ws(~BkN6;lqcNAo=roDd)`g!CyR
zLXqAD-=eM{)sUK;ShMX6&w>oR78VOKfc=5dTvDJs$7vobGy!r4X?xb#Xp996WV6w9
z4<$$uq2WC#n*EE?((2N;Etdg6v#sY)awmCV%@wiyq@#+bDW^e*;~oz)qn=NA6cgG8
zDP5Mr&_$Y9y_|Vc7T6}4uyIg<b9ESO1Tr*7l=&I}|M%PWo{Tuao=lzN%a%OVB}}+b
zTG#+L{_I$(qQVvxqo_qJSM<lNs97eljnLGA_Cg$FsdqnV=%#VZ1iPG71MDJDMhPIn
zjT6$Os@6K+d16!J6ZVn~$_$TiP{Wdf2wKOKoA7}#!5Z+$+(a|502B2vFI&;PKuD0c
z#6F96ln{_CP$J%B04Gi*p<f>R++ZA0HdldPJ5LS>4C*&H2heHby_Z%`x#ilq9e`<P
zWy$<8Vztkv65u>5B&zizc`ykZ<!JRed?kXZSi@f^)*I^+ftG1s(;^P)@Uo$xB6IG0
zP<1M?Eoj!A!JP?~{2MwmxyPRLo-+n0aVyE^0hb(r-wYCPIM9W8Xwb-E0gI-wKfAKB
zPYWzmJGoe_(Z1{n=S^Qs5?*=R*`lnsHK2bCAex~!+S!5vJBvt$Ef0Db`z5b}M|nuB
zN(Nt2063Mw!rgzvxoy{P+?!q?7#{N&!@TEMi=Wd!IKOZD9B1g2zQ)G2^ZCN&zCCGU
zw(M-LFmPQ(OY64R-LP9>5*#t7S<NCdLW`++=biH!UU{_ce$Jt`VW(ZPyZm#?6A7HM
zAcW#@`oIh5-Q~msx$irt%QZLESesQmES+sl9m@T^>v5)Cx`L*MU+F3?DO`D(j&}yy
zmGec;a@a-|J2bjzjOdGKcJ#uOflwIsXySe1Ev48Ae=G~`+wHDnt1fSaZrk+*=Yo2Q
zetr-BLU;Gdy5eqij5E(09*Nhb?JU{zRt6NC4SuVC3>`gJFAyYfIM_G!Rz{FhvSVNj
zGF*5(!ydzn-~#u{8Up<Z)DuEJrF%elI~)nQ+bzog);6KJRfdA5vuT6Bq1XZSPJaZ>
z(%cX8q~L+Zu%^>+)mEh&u6BkRYTR{&Z-6FDtChNI(a@N=_Yoo{we%5$hPRk!8qgTC
za2pD+tVOSa_+^|Bs@w9n8W%35t0{I#u*^3B;+r6+$Z5#ucI<hG4RBuI!L(<2AZd<K
zyZ80;<VVnl@rAt&U+A_i!c{;9Ho*};0_+0U5k*p9Wi=h(Hg0U56u^mOGY`EfGy95}
z1UuW<?yGxHT8PSa`2&rIV~!@}q(d3N7;k}#_f+@AQfczEv&>FCfNAT)#Jg}Mv6RzE
zi0AoGoWsMp1Up;VcFS)(0JX!GqeD9UOP&(dpqE31+WmYl*Y*rdoL;~bmkuDDUz)Cb
zxLDU#Twt<Ot^46Hj|y|=a=A_UL0f{sX9C5L&KP2dv&d&YM>Ko?3b4wINb7A(<^dD4
z+Ka5P%N|Tn+xX}(D-9mSk1tHmmKdS=J|z0`Vf?6%B#_?ePf8xteJjAb-t*mQf1$$Q
zG(n+4x1S|vXQRAFdS^H3JqY{ZVi`zw;+~6VPZISW$s-k;KnUNEob5K<hkdTS6aBdT
zw3E#4x4-+W{i(HQj2C!yN_RBUS-<eEE}W91Z#!%3@*iJQ!Ik|SKAnj2sz1&fd)J-B
zmE<hYwB3B%O&$?CyL4eay-dx$?AWDy7whHe$i^%mi_9`l`k-w}OfATou|WJ&(EuF)
z30Sz2QOG!l6m#2woVs}{4wO@#FD5{HJSVS{r{;>)#WYKziO9lypHE0<espBSg_Ubs
zjCK_e3~awc1&E6xlelIu4cwz2{t6_}h5jo5sH#8L%z;~TgiOkZuh#Cw-5Mz?6hE?>
z;~X~w_NuKy(HAs;0=d0I_A2QZK@db+-%>vd<leA$>Vl`$q6}{#MaOjF=bj{J*+g-)
zL;)k~`3CEfOQg5aDZ7l2^o*$2Z+wFIF+ytJd5n}hQOcdSI_CkRf0aK%Jw@S55LccM
z8p3iQDGO^+YwQ6`(6jKzRP#?*^m-P4Lhv}5{(3KhlYBm7#p2${ya;e5z=jB!P>Waj
z`_^;EmQJVy$NM0_M*$Ybo$q7zE5lH$&V~AV2ew}K_Qt&X;;5bC5j@euyCNu&X%wN*
z++XFkqW}i_iJINCxzZBw^YJAJ!$A@Z%mX@wpZs75#ECaM7Fo+JF}Mm=9T6tj3g19S
zER8&e2yQ{imAG;}1k}@{yV}TVSO&0S4))lIltBAEy;;W@dMGW}(~6L6m_tg$k7RXl
ztK7`^T5p2gi(gbjBi^u9fw*<OrK+qZ0*7D=I5=_202kTpP-xH_4nWBXgp);Ee4@8d
zLBGTRfY=coC)-y2!qQeDqsigIe0Aee_++{!5DT8wGUVBMrDnF|2<CDr81*2WuPr)C
z8y-6uyX^G@BM_Db<;lWw*@iUH${^IlD6&Tcj7au&8|FBlPP^+^4XO2NWfOyMEHGcX
zfF2U79yBt}LkmTs3)dd&zM@{wMqa0)27zo5q$;Vs;Ug#m4nUOSX>o6%<8aT^qKdv#
zQbJX80#&SPW-Bs9k!ek|wAh7G2?Q1+HBH$6=4XUS><W9{c@nowfX4j!YjTtP(7;2J
zJG(K|9nEiw<hl7FuMB*#99nZNVcx7>nuL-vQCoxHiCNCvH_#3VXl0SC+s1sUQdQg!
z*2bLdH~SYtZohg6Bhf;b@t$z%j1OgfdfkzX)z8c!G@F66L3k3+6GtN0#HwP)!~Mh&
zU_Q&A%qA9*FhO*FsKQ`HI=Fg#L24V7kP{f|T`sOB>OtGa>H&<4`-^>Lzwg+WdT0B!
z^M#;#{@Ewl0*OUo_d3+d;WmY-%0iu`kmdqa7+C4H(y|My>AuG@44uu82i^d}66i6U
zqUj4Z$}h_djzA}-0O8>9myN)|8ru^)Qhb!L3FoN=l1u#qL7dbPSZ>xuu(B+BdFsG&
zcE9V!k>ivwp%O_x_RATxNdx;FA!E&<8}=bEg9$UVPLOgHxn$NyvZDhtfT@|A-2gT~
zUkRfddLpBQs&18e7|!W1Lg;a16ns%$w(-~<BLhPLBkd`4sL)4`LXI;07m05*TouFx
z4K%pedx$`K!h<v?anN2_3AscmV=TWvHpqOLDKpq!^a?3_6%>za5(iN=zG5bW84L`>
zt}gI>Tlu(PHY{vWJ7u(GU@}4Ii*84p2CA)h*0+BbME|D%694myf;Kve3=Wk$luG~e
zOv9mikyUADbH5@>;>e;523`&&?6gKF`Qd`jsuxf<qD&a~l}e)VKyZ?GvIiCFo*r*x
zkO^%@qrlp^iw+zR6Gy!Gwa$to!X*|9NU)=T7#SCahHUj3&=P=Vqat}T<E1A6wJ*hx
z!MhK!i7pgRbOAOs?6a;p8#+VTmtsllQ~nm&BJa+&Y=W9GlTpNg6t&+$Ga2Nk|3K!+
zP{^(fU!OQS50O&_5D<*3TlA}Hy+rSU-l0hF_4BVOT;Vi;S@#VN?z&&y;nER?ai2dM
z&?6UghcsEiDyAKSfP9=Fo4qWlR#<AtWv`Gvc2`(oMXSK;DcZv{D->L%E$zhag(X)l
zCk~+rCg7N?AFy`aAO)_>&RS*wl?gahtQ>#e;Lcq3C#r0Vt9G_Q!m57KO%@?t;cO7n
zGxu)NavhKi1wZ$4yJ~j*${Ya2VgoO~qU)!h^Ifi!^#}Ldq|1>Yn<oJS;bR6R3x_28
zF-#cB|MC&f5OWQ;hArnzrGBKh^I>leiQc48CjI5mT%Wi<na<w}OH;fyQbA}NwCXjW
ze=r2}(8I=3>d?vR=Csva&(a0sf7qK+x`&?UjF%-L?cf!apOPWhiGUfIxtQd9_&IXN
zL$3k1xm3REGsFv|ARO90xIhLO3ST@WusQ&C#(FyR{K4iF#st6yA+T<MxIPwu!c?5|
z)IiXsGlA1l?Sm~`5&asn5aQ6pXzjOjvyOaSge=ur1qdx|5c821COkUwZ>g;(?MMUl
zxogrb<M9!kg$aS7AB0y{!PCJuMglL4#3YU}8;k)MZ@E?hVqNw{WJErgs*7N6(BtTF
z?3~Sc?>L-52Od`o$kVY&!4a#b9KA+z>41j(mR|rWBLM;pfOD5aeD6jqoGrzspK>tp
z+5~Yxuf=cZSP(flB~i7YPk0bPT?MxcDEQNpoq6OfoW)s7XxHA=Ox>{KC^vbP50nw2
z8H}IOV5d8CHyK4Fr8&VYoKcj^=HsF3P`37_3#@yL%F&Y2cKQR=8)~xI*-((;%%Q1i
z;Eb2DtTDA=YQ3B?X2{|DKCq`-zLkne#nwgC4Xs=2WX+|9jT`9^oN?Mw$s?Y@!5y>`
zu7;rq7yp_y)-JjY8Z?fT&S3)&XBwA#1WE@-1&`^!4TaTKTS;fqEbdp*#@HS%{*{$$
z9L`D?@VP*#V+;=b><klDpnVff7F@WV*G{?Hx;|_{Prh&Nz2wRMDg`LDO-jwp38@W=
z0XDe=qu*5lD2~y)(Fh6v7_U7~EwmM?Tq?NU`hXSVB<yRrY5wi)=G2}|2e8_Q5p2_Y
zx(_Dwk-^-g%i8=3gnF4F-d={^+Ky#ST>5+r-}3gl4Aa$eCFsi)nK}P18pb(`M-#L&
zoq`zy!p<s-rvC_vX(<Av>pJ#Q_)KnQGhGItbpm00pw6M61JL?pR!c0xfh(L`iHcK^
z{%S9IdO?T~p|j;F7rIR$i3C9((C4*IXaXmV5JlbC>BFp}!4Y9@wTiUWMprXxas^#c
zC0+8shRcfieV9mDyUMhceQ3cp2jrr|%j<Irg&H?U%G6Ec)fR01NS0c|W=yb^7_9pe
zGzWp`^DV=)X!YHzgd)%h{Af*Nf2x^lj_9u)p*6NX{JPW6@JEz;RglHA21u*oDHZr7
z!u)+MK=52oKhQz+GK?^hEZB2~s^JOsw?qI-nNw@vdIiCj5N+$p+8)u<2d91+2%&y6
zH<|X@>CmLtYol$&VuK`7WUNf;?f9;-n5=n7V$RN+>1Gv#vI8AS5v+NsNRy#0uE>yZ
z7ejtaugiM5B(5si{|hhj;nM1*%K{DkVAmOvm`aQRl`d$c!lD@AF6H1~%$<AqG0gos
z`Rvy-rGN|Ll}a<#JAISs8QAQ(Jys@)BZY9V714J3baVF1Hnb&VQ0}q<uov?(5SnkF
z`OrsE7*{KL*nL<&f@*k#N9GYM1mO5=j#hHIhHR%S-Fld>$LKm;XXByCmaz_HeE!%%
z_HXZc<=C}j2mQ~tj^P$>+qAqMI@@`%usOn3Kt(`@fz%h4;-%Jr7KAb3D(IMlvQ|<r
ztctM3l?he)wyK|@!PK!sJa@Dl!#Ou!%VSf9Po0KsDsrz$Qi`Km2^C(E(Ia>Uf`&*r
z<cg?PSwx~Br3g&L@K;I5^YMe#|2G=QQt>EWXgWHj?dEL{RH>xCKAV-MdX()R+Q!73
zR2>0ZYa3T$+pyvbk=T_MBqqGB>m(Y=oRxXnaw2$!Q?nir0CX*0w>0tW;mWAN1U#6;
zrWnDs9)qL@1lvWSBwpjv6c`uB3>iP7_(a2(8vhHlk?@1PM(!w)ttq1*bcQn(AHi7)
zV3CM-b%s9Yc_>hr0$VK27>Vo>w$CfI<k`|%$;nvbjkrsIbc#=5K&-n^zPhBRpY9k}
zxW+9WKgaJ2_yq?g7jKzAyClwh%`lFatp%8V$x6#Q^;m}X%wAQWHBk-|*%xC->@6uP
z#YA4&IT;FRw$3EHpsh@hjwy4JO&7|H+fADQ%ECcrPZe=6D)3>Ah-^dhHcb&2La171
z!MxNnGt7qzflER0BbcP~cPD<+pe(5hFQXOVs-VaMa_H5VK%$32vA0sUi^5CXPncGU
zY|CuY>KB5DL3F<A9k|;4r?^j#qMm|se*N8Q;_~qT-*{18qrMqPOk*S{aZtlrQa=u~
zQ4yac0*yL-BO-sC7NlZMdVbwB^-QmmU9spL3sQjIJ+9kV%Eq_{6RS$Z=w;TN&Pmql
zL2oUakG_D4*YUTkee)>aCKrCl%-%0knLR}t8OmV%&#W!@D?$rRi`|%b;kEo^go2|{
zi3(GSpJwsod)}XJnqj6+W_htb-~tz$E?{v8lwDVMocWHSVT;a_M`mUKdEi*PcVdrV
z6Re(k%$E3&z_qXhNYq+pF*K$QpvoHi3t&z-w)G|AB6hC;KuYe~^TJ(>S~$r`<<SoK
zk3r=Mv82%6hX9C$e4E5LM($Bn!J=!6bXsp%t9%s2HQYdK)}c3(Z8j!6hEiHx8t%&v
zon#=HP3?W7kT_b9Rd_xBm`<1#G$WdBvIJuUszg@5PT(k8!?-7(RJ2K^5$_`>$O}yv
z1@K+iZ@yW&qqfpAds_B5h+Lp(rqT1xC3-^+6(}=SO3qj#tq`k4``*h)R~*gCdNEuA
zv0cvwp-A5ra$Q=kplZcIT%C%y(CU^v@OBZ6RMwdQQiAnL=|}rPmdO2S>|?Kc7y2Bk
zQ`AZ`QflolG&D>FVk&be8mB%gPDEr%|84NfUPntc5o`r|(^ZTzMp=FkS?X`yTyHHS
zfE=Ec47G%fXeA;jjzzP!6m^`N6`$IeC<|{51!4vB9zx^Jd{vYl$|MV&gk7!DkF-Fc
zYj5nz5LLO=HZ^e-W{%L6ZnBI+aBMs}w>RSi-Zb3sMKO)46))S!QdGWwsY01kkzaI2
zwG;P24)PvFW-^1{>cKL0HdL+lwGsq8XV6_7HxpaDn8%4056wGOB{~-kF6xS%3ZPIC
zB=JmXz3Z!q^#0n7P+x=6>qB@wukH$PSaqF;5Z)J8#U|(EjM(WY*vAt1D_wQ{)^QpS
z1A8@{zRXk6l+a)YFDFofIbK=<IM|(*7^{f#W_FYQ%$pd@+1`;eL(L*mwVg$T2(2vO
z_j*0n)S_*VX6d?cW(aY|xFvNR!(Njx!7v0Ur^DT#zG;uHRBAPj4!E#HMhR=1_l{qv
z`e3O+x`fK%3iHun233hecwg+<ljZLdnBheL>p)o#>RVp)Q^Dbj7mm1Pz816Ur#xc~
zzBq`fGHwTA6B`U=lc<6HCS5pGsqdN}zK-QLt)D4b8i)L#VRFA)KnY>G=?%!m6abLZ
zCt&htdV-Cid%5Hv%xG`8J1*g+`8iJw*ptFB199V9i_JQdXjI#2N&ryg<b(@y#R3=H
zaOf*&JWDP+k8zwE;@+5i_zELvax~@?n~nMM2jdv$Gj1%lzOpdZYgAe#-jv8a8#5Du
z1vZ@4g^34k#SX?WS6J>vp1R$hVA6ihow_ma!R7cLT*u**Km5%Z-G%vuh`>$e9IwwR
zd&auv2{^iO&99fW5I!&6Je(XcxUqMApS6CGE^;z%%xfa|B$loVLhbu*wX8`aFzq5p
zEys6!Hr?-GF5oMX<61?JY)lS7#<sU=mx4}k)4MbF-V8QY@aL_-M@d5~BF^zDv*1LV
zdu?gk!aXmV<{8lgF~kN}nD7D4Y8;!0YH8ev^Ckr6gFLp6JGVIF=aWmaGr*|CshIf{
z#Sd*SrW-Iz&knwp7A80V8jd+ZcI#P5&ibb4!ouMdDQ*|EGYrKA-GoO#L#xL)klI=;
zbQ;l!L;vh3ms|+BG0f=q&9L4gQU_Y?XPXa}CpG&|CESlmkOWD9zy>jsG8;1pcEDC6
zcVSKGvV3fcRyT!v7`C5t)5}X{kO0Q6n}V#z-QJA@9&ciPsKdRD&O+K4p>Y@^Q=kr@
zKBv$NX6tUgfk$#P3gsrT-a;@Oo7k?HXk|)ND-GKu47i9TW=YE-{CUQ=uE-e$`898F
z-J-uWquJidc-E0!YlBxZ%n~!~c;5NPu;(yCS76}tps;@{jTVR=QPe=C&We<;ga^Wx
zP=^_WY~B)NNrud1BF&3~jFeC<z3wB;*k>n8)l`|P`$&^CNs%TgQlz;Wk96`|47i-A
z{5CBv^klZ@-5{Y7O;_yotrr#B78tNsR&~o~CV0Y9sQwTT40BrI02;p;KN;2X_KA_8
zyonqMvzU!}@IDT9y|5daaJ-*}{37QqpTogBZkDC)^-%P>E@@q*<Hc6nX+nHF=PIKF
z4(g*lT_K)vwC#m=_fwvXskre@KRRspjNz5^EhzE}&S_D?F+QM0UpQcTDZK6Y_otKb
zD8PhmSg%r}#YXDyVpTO{pfq(^HW+0riE}$b*U)zC&l<xJe1vxHM^R~lzBNT1f#r-S
zXBY!!W(LpCl2tnL<I=nL)OYan<Yg8>(6z9}8Z6qKGg;Nk0U5(a`!&^VNApBAF0vn>
z>N>cdb7`JkALaf+Wh%qrqplWnYThtilCa3*YM|M!hVE&<;;qVHVUx$y>`dnJX1fW+
zLd*1~fRg4t)&+Ln`uit+NUq$Ne#vpO_#C=v70U0LA1+t*7}1H)JHOtm2(7J&f4|0i
z^sreU+vp3%2~o=lSmf%Nf$CYq8XlqWz65WS^6F(GAd}w8KMs!~MCyA!@AIJf#)j*}
zhi`ntg@hn-3wnte3G-69$;j3AKBsbCF3x+j5`pRR=D-z_Hv@1)Nxj1E@F6!je_>;1
zkLOSM$X7lHbqCkmIhMt9Z5I+vTT>93M>C;Djhe-{p)>!~WMZ8Ds3ds9hSceqQo61x
zemLzokI+W6#j5T&^-YF?+#lWNX*al=e#H;S4qp-GY8E+39yWZ1NsNBVIuj>ht8gK%
zJ<A2+PVWBEJaXie2g3EH2%&fsy$)DWQ`Opv?p;n%sU^Rt9ck?EH(XlzIahN-?2Ax8
zDphzS!p*q5LBg5@Il5jRr-*F1z#Ec9vPc$KV8JNpeLZsg%v;<;<u?uHP;|M`X5Onc
z#VkP=z=9-Q1_s8g!QR+L3ABy*Oj-0#QY1xEps>Z}E_Fj|@%5|2xar;PXKg#$S{wQy
zjU$fT9LMscJ`9FMZpt!mCvwS&wWGPQu}Ne?V;7AL#MB(9admZa9608+8@Ga=M5OpO
zGT$|yml?tEE&&Oox+%l3uHDcVBL&$+k7&m<51E^@mP8?`yi+fv+b|lXtQ%Te2msd(
zigDAs>sf2xdu}mya*!6*Hk=F<ihD!ZiOnCyZj(i+IX=q_EWLV(bT=$=T?a4@00wC3
z!(+W)uUHlJHGI^UP7=$souo;cq)2TVV?WUP2234GXRlI=a>NR8vHC<@9x|dy0V=D;
zB~?--6)IcF$k<Yg#r<IFb1d^vQRy|#cmOs@E!3P(Ob2rA0f}fclYe2sX~R$c_&vu>
zD@ynJ(k64c4p2m6AWK<_()rh{V6NB^rxaI9Q;an|KlIvD*DcLbl~7J$oDmyfdESmh
zK^2N+pHqs>s;-+u$3%p>GA4P~@HKu!Po1|sy#id9w?Cvf*9i$XK1(Vj5hP`ymENw5
z4d^7w&L1e-kmXURllH|mV^XYhKZ76)=~IL`bYOppNZF6!;?`Xvo=Pc6OJ;$4AZaX}
z$wAf9E_R8_thJ3!iypX85^N<VBrP2?mjJ?;qvkp7002M-I``QUEw?04)O6*+aWD;5
zzFAUa7ytzpdo+aDpTCGVUpY}4DWPN|-+M|PbZbOlO#I6n5?hl#QKHWhN9m$(#FKD5
zPw7#bnl{PYQFWf|7cIF1Kw{I3Svr#OgNvkOCcBn>=6Xr~R6+${QQSq!1z?wc^lP-R
zU$ULMQ~s`6FO*&}e(@YR8~g}Q=@~X|@tD5;#K}MjOQ9@>ir5Td*@>S#F0&^Mr)FPE
z+yjHTjc--BxT9FHm1eh)9(SY51KE@(j4cn~?gs#`bH7JV#+L7SH?};u<G~kT5&(bz
z4*<9cf0+H_ee*j#6}t=3bk<XRQxyP^-|g-eWb>TboQCI%J^@_q$HucBqe9W`H<z2Q
zX^Q}kw}W=Q$+LZeff*(`fs*I26vC&_p(&UhieMQjiY>hgg@_O7&t0F~kBn6BeEp<O
z3tLNKmsny+OYn&$bW;VESk9N?2um!@(vAm9W4Ct&z?86uFpMpqlDqE>KDO;<J-Qu*
zDHb)+-1&YzF!+T8j+q6Wg0O-iZj-|U`M#!{g18&?B|4BA+cq+sQI-l6{v1&HbnT{B
zjO&u{mySO>9TRF&3;A2lr$X<-NO%WG&l1^8dW%g0RvlsLqZ#HSU_g){##(;`Bh~L(
zR*4Cz>R^kA@%tg4+03()r`+uuSmNuUtO`L`JBGU=0IX<)s!%qiFh#f6G5<*5NGB^4
zh8|^RKOb8{l$H4dKXAkAp5cXgRZk(|b<p=#v>dc~lcc$0Y)YjZMW7JZx5UTh2XOZV
zARe2;z30c~?{TO0z-yln-9&qf$^LXa1wF9QyXMLJS%|G(k0<vzBgf`<xsKPe((CQ=
zOZuKA8@D-piqC|jL}|)g=P1$Gqx@jatzKWre_1kkY0Js0Z;SPmql2crVQfCuJyU>2
zWPJ6M3fUB!*G=NMbzj9$@{Ir9-w=PKpyI!$VwlRm)=hI@(n2*tSl(j40$4mEVfQi6
z)*oR&7NQ>wxsvjZu7KU8v`(xf#JLV|>sfEomsV_{Zae_n7PS+e<p)I7TUw(X1>5oT
zO7SZub+ZdFk<N)c8ETug@UM_rJ<_ALYy^SZN`jbk#i_2D7Y%@fAm2nFdy)wX7(Cge
zvT+S<uu<(<*n~AfV{C@3S%c=Jr@|Lo)FN#Qdmdbp&&}|pe8)z&;ni|!aRBmUjPKm&
zHeAnr&lNpsWzU!>OBj(&{4v5aLeB2qA@hH@Bs=rW&Bz+Qr=aVgZ<Z6iqCo1f)_f@J
zEvxrTZC}6XcrAVG1ps~&3G(WZ*Kz2+e{8XzK!hjL|3&ROhRL6MWLJ<xcLOvGmm(dS
z2Ok9VxHLb%&-jn(wK?F|?_!X=i<3&1o<1|uX~g`9S%6>nw2@f9x#vkct2QL#w|<~S
zd-u!5MB#HbnRqm>+TpX4Plh!a%w1)m?dK=mxDYXi{eVh-=Uc-x`agG2ScS9BI_UpE
z{@N10tEVn*HCF;DQlt*-UCG3wgiciQq!c^<3dhLMvv8Ut@|lXueE4-T%wPRPxAwT<
zL-QO|b5JbEER9n!f18PI3Ls|LwIAxwM{{ygIEw!#H;v$5r(`}K)@WMmopz>abWLvn
zccx*Q_)wP4GklrFBRoYFuQfE)L>u<Qfj%wz?_H4yBf9?lCxWQ(pKbOI2OyBOD4^V*
zA;BRRYK~2sE54ElBYXpNezZ$*^AS^S8sZI`u|Ii+iY(p4b=^F;n8E#zT8u!>-J5NY
zzZ~TuV{mlH6-MXK!`u49?f+RD^lqE9No!j4Q_B*+oxyj?@yaUh9QGh)c~8Mm$3pa5
z{j6yu{X|ox06I(ryZb!#tLYCN;U9pfUW41zuJ(J80JxX%Od}?FmF(yR4n02LS^rw4
zAIGiV(7lZ!_g^p!7?LORn<~-Kp!g3>%PTB$WfPJR5D>+)c<sQ6`SXH$I?X{ueP5xG
zXHaE^XHkV|UznHP0apwPnpZQYjP252`XTDcIlA2PfCc5hf6x2amEq<!Zw?Z&Vs)Ex
zgK9IIm+g+aCNYVDF~(#IIsY<G3HNl%q)bYbro{AUMDHjScUPfr?iT<rc-B0r7-QTp
zz;AAGxq?I_!ir`#xOp#{TjlC>E5h*V3C>XKmbs}2va0O$>~-@F9x^Fijd0==_f4F)
z+gHT@8Z5Zmc&n|k!B5q7I&X86e@1KDssqp0{eO;S*{AAr#kyv<9{bvlJoU^Aul&M4
z9-A9)P7NnJaP>{>cnEg`fbWjgk$3Rm2mah&L}<DG!CI%HyS|)zWLngLLgCp$ky53z
z+FkB4Urp1PDIGKBELgnuuia$sPX|th1jun%OoDI^fV1PDs9r_^tqxN0FPW0?#AI3u
zQ<Tb7rS6t>_a9)f;$3;-c_%QWB7RvZ*_Q1&BWGo2&dGV%m51Z>O^GA}0Ra@o|EvsD
z4dy@1u%7_<a`!84{A*R#*RQ%}qX04#00Ea5FA8*HWV-$i!p~=H?KOgf=+UauW>{2L
zna2omm&J)`lhCM~UW`LEOI28l+B1<dAgeM|hZe2)$|$0STa{6qT_o3$L&}WeU6SB4
zwR#42TDcfDZboKO*4T_GfX>j?ZRoYlCGYW8i>}TIcZD!MijN|!#oPK-E21uVAf79u
z0wxn$Y@r#n5hS2tXuVyHKjNsKIvpx?j1-PX!4BtCHZixu6-5dNRH0cavjxOeBqi5m
zmC>QgAc|1xO0~|It<!Vlf3cwFqLQ$zjE03{8_Y<CksFIFQ^ro{zf-h{Xl@f4HMQjx
zcB2>y@V*X&=A~`D)D!d-T;>)dT13sNx6UHzs$^E~^PfYcETcsuZ5#1k4*HQ1xFVaX
ziIQ-uS~QAgS!7*(s9qYq)DILr{rq-vjxtxVtrtHp?Yr{Pn#AaYc@*^0v4_#${7MUD
z^sc)|g1q-#<-UGy;73w3hzQe{j`Uiq)T2pNsxzHH9NLrT78!A80cB!nnLR!O2=?Z-
z6xWGfTFQLR_?J}kS4H}_jIzie2$%A<t`fqBJ3{)icTm|PGE5A@{1pUuUq7BN6Xoo4
z#9aTHzA~t2<LDXsY51OdR3jq%RcP}&f+On6phx%`d;(^1Fasgd$U-AOLMdfhV5Q~e
znNv$c-Ijc)MHcI9^jCF5uT8bheW(rm*byr*gb|^ZjM9TxNQmtrC=p_WA=H3|M`$Ez
zHPbSWO|{aXo_1`t>}~SOU(Vethao+M@60m!(XM~m=l;<*Eng!H7!){hfnm`DeDK13
z`d5CGr)StN<JIRf<ff`OTnHdr|9sLq1OR{n7`nhpxU;Uqk6;e89uJY4!IA_JIUXyG
zV+7i8VxCaz4~eIf$KpRNGZKb4de%C<e2Toq+05TFw4MX|^+tKh$S8qAQhA!RGXDdL
z>a+4EX8`<0IRHV`6{ytu35Kqo%8FxCG>Jpq0ZI1PhAVwYl5EHT#bX{8;!_Wo2-0&O
zOJyDZPh@K{NV735SQ2s4Cjq9@V>Ic%P_(m<LCm)2Sm%aMRc2(o0VW=3ck$Ng8w0-?
zcWuE#6JbsC`+z;nXJd`qWzv?79;ovTC3Ii|NXE=e#Ox$gS!F(C@4w)jZ=x*&=*D5b
zm>BhnOhDrCf||$hhl4gJaD57=gLVO~tBjqO>ZyCJn%dL=M;8?7Y}|^i2e6P_`J4kL
zxM(=gxvm9`zXg2^5Y;<&p&~OqM+5)|IseVZS}{wQ9wY9z^2HWzZe}+$xwzQ;Py+k;
zehvpR?WY0V?-^PKK1T0x0oPzMa0C}FC|Y>%Z@y8*()`*;ih%8w|65$=?R4%Yb~U{!
z<TCwBOlV_@r<o}tc&v7%4*FYhA2G)4hSeB&l2D1jfj#?9fhX7IfWdn8w>fY{QGi(j
zWzhrLmNsxX-@)$`)B}-OhMiOS&QhX%IxFyX3gsQh?cF4o+*J|_sm1Hg;5YLES#l1L
z)8=f7bY^2miPXwc7$12QhZTgjPH$%|tfNyqh~Sinha`|7FvJ@eL$;K729lqaB5IN+
zogk~M>qwbd?wVbTZMeo!gUX~Dwuxhy1gR0YchsT|wDK$<A<0020yCr%VmyH<`(y>2
zWC<}XymQpJu-lm~%`r6R>bRVN@?tczl`7HY>9@+ha?ed}XTJh0q_Z&~mx{6v;m}}9
zNU=69s@T@PJSmiW(?EmJ^9yd6Hgy)*r#l5yCy`lmhl-U%x~nkz2(%JmU}wqhUHMp+
zQ@Chi3m25_E=|uVqXaRMx40EY5B$x&g8dYGTw`n>W7h$PmXX^s2c?@?mP+G9R*+o?
zX9=CP3_wvIC&wZTB!ie!C$&u376+P<fYof1NVb$H>3}4T?Ru%S9E2!JD=A{IDsuuE
zGqL<a@t;yDP<DAjcyUo0AhA<B=meHsI6!axQR5$S&ei=0<j0gCf<%ZyW3ad^BRPRc
zB2%a|I)lk#bGSUdKnMW*u5x2|NSe&#zyHsNWGJF+=j3%?C!z_#qyXAGJ2ky=u$Eo!
z;rFafkXs@GpO=%f!bY<93)PXU7qyvYhUsQnVWr82Ti|HH0u>eI*SdU2fNwNg3Szzm
zj@7PDxAm5rH`!`yn<@DnHdTl%1$Kv1jFDo+iI)H*Q4%nSWKb|E9qvfw;8j)yQ3WKF
z)T*KL!AO%X0~SsuJc2AlB-zL)sB+|@q05txfr*8UgNuhxK-lWx{x<|HF$t-mb>}A|
zudaFu%4(^oX=v%_85o(E`@zDhfQ_AllS^T@YEV>TO^O?(gj*?(GG65>RI1`rt%hH1
zPt>W`P_t1Q8?C9<#%Qj^Xf0Y>b5>)GHO_bw%(uWoi!8Rp(so>CxfNFSv|nagUB@-n
zT4%ity=$DVCYX3r<1MhoR@-d1!%n;Gw#Q!W_Sx@%gT1l7z{8F>>X_p_*V7rjovPTD
zwmj{Ovz6G<*5_St(IuB%an&`~-Du96{qCPT;I=#N_Oen(9PM;x9do>MPB`h5)6O{S
zobxWY*msv)cEwfKI`4WHy4X)Qy5y!?ZoA`OclEpHz6Tz9<Z+KZG2m$)&${e+S1R+O
zYhHTgb#wo(ZF{`)-UlCj^4Xv-zWNsbpC5kuuWJAK<+uO+K{HcN*C(jXyaT9ffONr9
zuk(ub<sXus4Wvc>veXi*V3fuTe)?hxFI*&+NM&+`QpF?8YbiGXLNJ12^pd~{k}?=g
zW{cHkcQ{>ckJsl91ViCSlvkKfQnqv^o68p-MiR<Q#C(l%H=3<>r`w~K#9%lYPo}f^
zV!2vx=<RNQIG)az>+SwvCgwkMqQ5^$w<<<=40rcs5U+8X(YsZ{XA%HHFv8*T_yVCw
zERo9O3MFK4aZ8|W-EzZV#BkrmY70M{E;n`X|CT?$W5R5)+UyRe%kA;{`~V2SNC~Di
zC)<N8eE%0luoXChj0HrRVL4nLUmz5TB~qDOp;W0gTAkitG?^_l-6!iM@pON|$rLJ$
z&R~8va_(1*;e67M(f=z;KXqTqe!m&WsBE+JX|~7h-ibR|%D0hx_b+mTiw(|E^cU6?
zQ7)A>PHr2*lu4x#R{haHnfKT1+dnDtH2jm3k+xRWt;bfr{*N1r((!M%Z*SMlqKBv>
zsXXNYr_|aD(g=qhQXXk=E~S|GJ{)PRTl%tz%>C<+_-dw!x@47m$t4R_+!5zdx!L5E
zt4lE{G!h85UNQ~Ava8G=;%Yk;N4ltem$Fkkx5^Z5E4#K^BN{S76?=ARpeI(z57&(Q
zx&cmr{|`QC_JFp`rkuKEVOh~aRh!)t1RL965s)e_Ns=T9c@E3U)?TP`;)pL_SCU(n
zkelT?4V=1|imDV%(yAzM-eAc_XS<sALAvm<6xqF&tiZ_ZOx~0advZzo+;*b$X<ox>
z-(F(pb!PJ~3845tkkw2;?U+ruN&Ajs`Ea24;R>eA7;X;iefKQ`WP1;f%UM7EX#rMg
zQV+FjEBC*}=zQC4N4G1nb+~EUcQx|h;O+4AaNe_4s<y|N)0N69T|{XK)KX66X$@!8
z#&cgounrGAJPj7J^X_?XyuCD#y4Dbd@F@2FH5HKb9%6Yp*6yuJ&1ce$w{(1F_wq)}
zRxJYzu|3x5RcVk)!KsE0J?l=)&YNTf=Q%Q9Lm{du;WaH*t2%31P;3AD>RJCDHw8Q?
zcgB_(0l-=SRhUfC15%~=%?UvL{zj*S8Jh5J+RBHc=qd3GA!~_d$q88{BvfItsklGQ
z_eZbLhxDrfb6n4v6&7DMgvExv%l;Ae?<qXqw7COb<2&;tZ1K`L=fbOj(1RmInoBSG
zME?UaCh~uDWydYK*Usg+jLS%fVH%lYQGph1MT<Vy=lW&!pM_t>(Hi3(#zudeip|P3
znas(IIX5*JPA(AaxqWEIUhi%c;Hnl4c)Y7x{V4?hva9FSOR5u0DC0sZZLH(wH3h<i
zGA^Xj#yW0ZBOpvD<3cKJtmEc22Ev3gE~L`NI&NM&az^gu0IrizK3DR~*hf)DdRYNF
zk0_;c#xhhC)gvkU?24=dq?R>PMMe{(4!SxhBZ(H$Ne#&xkFqYp&8sOhXV2P16<h_)
z%;Q4hjA^GZM4zE83V?fA@Ssi0X_~DiYMcv|5Inmgd)tH$plxt~&f0-;EVT<=L<biz
z<uVBX0B`_(^(u4(GRnA+sg^*PP{xH^-LZaNwd*TUsEm0(n+=Y~VQGDzxJnl07mZ7J
zUfo%S`|9bhRRiL0NrTC7TI~_9uVz8OOdtT90vZDXfyV$TAOJMBE)4jRnoKF<LMjc$
zI_^Y^nxvFv6aW#-g-W);#0x>`fTpoRawWrz^&omsWuzPutt)*nH^+X9|Md5*>HZsm
zxNZ*8|FCQ(n|7H)sE`H$wp^zgG0QX#!kMy4#jP$2zYJQMP%RIK%?4|_abHH6aY}jU
zHw9O1_`@r)C0hGRVAe5D@;2U{7!<~dF{~`R>+E|oXaeU}_U>M8y&;>grs{dz42PeT
zEiE-!2$3NLY<67@m<5K={EL6E3!uQ%84PQusUk<&GNpsut=9r^Td4V4)E0Gl5M~G>
zTau=!-!RoFL8v+wPjR)x%~e}5j928n&n|v}0_!#`to;$h!q=GlKHPc!mt$M4kqCgi
zzJZ;EPR6eN$aBbh-muZ_&=$oo)Cfz7ODC2X?M(f;4?R#?!rPd)JhU@4x_A`#j(gTZ
z`L~~(QIl%2BAs7{xjM)^7mnp%xS~vOYsOLVHtjpU?blqIU#P({G?Q6NTMsSNBI^7R
zbu;23{ts`b+{Og@>G295<0U>VIPVu;wO(q*ix~qx!Ke5#sDCHNR}SvnDrY~6P%?2K
I0hlKM0JgZhVgLXD

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Regular.woff b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Regular.woff
new file mode 100644
index 0000000000000000000000000000000000000000..7d63d89f24bcc0c0f5dc3cda086fee67da266ebf
GIT binary patch
literal 50664
zcmY&fb8sh3wEe{!+jcg#v2n7oZQHhO+qR93ZQHh;m+$@cURRx&xu?&axjof&tLJu&
zi;Rc}00{8Y22TKpKPRB}ksteing7`T+eC$gMScXnKb*x6{IQL(#YJUh6n?l1007t-
z004u*NLSAhRZtQD06-0YY%Tx*vYo-&&s;`<j`8O@_s38D2fE?YW=95AdbU5@2LJ%^
z(+6c-kb-n-;Os~M03g!;X#V^MI9A|%V_OreAMQsB1pEUBaJ5V}6Fr9?8_DrUgYdt=
z01%p3x*7j)fdD{q5&%fQ6e3pN&(uioCyp|l9}Vk&fHg(9GW{WcIK>~E@CPJdp&-eo
zR*tSe+z&uh0RUK8?bA6jOB;h9UF8e_0ABioa`Er0Y%4w2pXWCG=l}%&0W2Iu##+zH
z=!Z-F88asU@awXILj}Ut#^EQnmW7{tp#T6l@F1@eoVNBxKfVJeKYm|7<7m7f7a4ZV
zBpqXb3Xnw*o_#m=#*9U<1tBC*T~{qcdjHT8Yu-5meU2sb>Qy@MbJ*<-;O`B(_s0$b
zw=-&#F}@x|-KxHElBcjF_9<5|@px=5UsI~sY+Tjq;Bm}-8HgESy*A{vuT|(C;k&qn
zTLHLcb|u!y%*!DR_dLJ*p$JC=B2h;v!psSPplH;xkbpySI>x=gBZ$WN#g+!rn<Cc*
z*wMoU`q5+PV}bYUa$<;w2I$el4E+$`^Z!OMoc>+EH~U6J1hjQdceq?{b(}wL!GrIx
zi_k;+I8wVHPqtji$PjUIFfavynzu%Lda>6KZ+^}2ytt4R+Z3ophKH56IDeNf(_16=
zos_QSC#{xSWYkQ2{ThzslZgR$@8B|q7sZ-y4RJe=>s8=B?hHh&l)uBVP!)x=Y0@6#
zvWw7Oe+k?ycf$6dP<hn4!ga+TK@uH2GT*r~M_wP)tCN9=Fk>J+8mQ10utcB>bs21J
zCDld~)R$)&mB2@=;c!3K7Edyj=S5ti?}Hkei9o4u^=CLJRdINvjPnkUF+7d^(sEaZ
z2~A2zFO+d}8&uuzs$KeKp8A>}*#knm0iCWLal@pOCDF0#9KyET<GCCUkspBY9Dw89
zhwipl8ekivGecs1t}93~T*aysFF{W;860}6*N-M$MwD)GW*~kFw__lj_UA^{&$;N=
zSItD|7^HpGME{8p2L)mvvRzl${qF^uL%SZoC$&W2^=YXN)@w1=Y<{-g>h<k5o_%A=
zp5S$ehu;EEce`{?gbDXksz*idZ&10A`CoZ>RVT;pzK_?8)7%dlF^>dh`>fGG>y1*H
zjrB>ZNz`Td#G5ot+?YbxQ+ddCg>KP0z&(Q|DsOgtQ3+qAJ^?iw3dK#4YNEHU<=J%K
zpu|fzDVhAfE{3NKf~P7Ug<v!zBMcD?_2Z;Pnscls4j7HF>%Fme?S-9;W9OaSMnGO6
zNIgKzq&Yvc-!(HSS;I0>31Rq~>E4Rw@~KbVG;jS9tS~X7NVBMjrbr6`F$qB+lyFIz
zh?=?(3<;7U5JJ$EU;snRm7W*Fl_5MdwDgy{_2HcSl?y*@Fa#b9ysBcR&66)Q3z?yd
z&%M|2#6+6IQ6>j8aBOf?C%Qa-7B4i1)O8cJz+(Q8ksWjYyf(LIC^f1-%7cJxj^IfE
z_W`lLK8(=sV3N}~UcM4Mrr4<CS@cf?5yHPWz~Ba^@p#0(iiO0;U2y@L528Tf^nd`_
zU_Fejl~z+sBkcP~VJWGIlsR34HakxT&r$ty7Llm=;}&MtS;>8z{mz%WL7D|iO;gSG
z!Xgf7s>q5+3w5t=i>UJ`{3hDa>5|DInMe(Fjd9UAlcCEYtud%I152i6Dp9$T+sNB^
zduA=RFXPS%9Kq+vO<v5Z@swu!a0XMDD$f!l$YV_6A-{IhO1?Ll(e$+JcxGrBrkQ}>
zJSUI3&hSrviyf)v53_HpWo%*@Pf#)6u_36KI*|-kPM(d`ZHiM2CKQa_2n}XRUS%%-
zHIie@;W7am$VFkMM!BWNuq*F)dHO!CgvQi?<rtJyp}3`mCEY5wr&Tp^K+H(-hs$y-
zl0KC3&dSUVaM?KAk(0}CxQ&m#OFNSnCUU7A8a5>~$#i>MpV?#$p)6od%reY7dc6}-
z7yOH->Q8#T%V(q}PDG4iF1??ecVLqpGL_K59bGP?&x#upz$~JczGInq7?ZRKJ%wd~
zovm|vWKjw8T6njUkdu$VH<HfhDHCQoEFx1gYwqNTsanaX@~~|)_Ulw@?<2hEMTJrT
z>ndYNGHojD1zpg?GS;Wv8WL#KC0)|R<sC5Qt-wRi4W3S`JcK4Aza>p@`iK}E_1rY%
z`8Kcjv&o1|_nKH8&iwxegr+2DGZxC-ds_c->x28L{lh@=EVbflTavo(m;8wP@cHiD
zz4nu#p1q--I1CvTvCG%;`R+5%FRnKhw&{$iX%_V7N%9Muwk?eq^;eg|+v$y=_g4V^
zfuLvV`s*+J#cYQ)hsYQQ2$ws^tGxuzaG86|%hduNog3D#_?CN>&VZhytPj}hxonxH
zV~wrJ?c3)Mkn3H1H@fTlX&{@YhAN$-C@dQQge^jdEyg9ge*t^&{>dM5>0D@XJE#&n
za1%SYV>=*pj$Q@Va-ZG>_q5A1zFYLKA;E8i)8#3M)2Y2#?<``OoTI+w><YH%8opb#
zA&-MB^UaY7xk!)Z)SM^6{X^0;f+TrT<RW$RLTNIIE@m;s;t@qRwc=4KMlJ0m%)9hX
zTALFX0e5n=Q|sT$=1N6v+2yDf$=SXwmBw{bl!Es<L1F1V23Q8z_$pV&aECxwig*Xe
zV}ct++!N$Um+BF!_a5qQiVnY#V(S>Cjj9YsN$4*{LR0GveMMC6d~VC6_<qL<RJ@lp
zLx-F)nQapGl&?FK@w~UR=qNpC$1_$5sV${jIZ!M$z=kW}e43$F3Gh}GFaDYbU-JcC
z6Usoyt7=3IYK(cx3tcOP9bDkYrN~=%#0EKm+nn`5{rWYiUV=pY@};(aX7H@hRE;NF
zljGZ^at5)6#lkbkrnB5Y*$w;dbjkA+I}V%v%-!)mXNT{Tiw*DTRWA3bSdwGSRK<+n
zT~&Gn@=$=aNJ-C!c&||UX*Oz*s$G40n4!J$07W9qC`++Ahh4;4z?d<M@Qt=mN6B)9
zwV)qAWmZ2`3M6++GXI5d^!i*S#0(PFgz$S2#?=`A1p-$mnEEGo8tCeU>t^=$j`j8~
z45Iw&?cIUR!`#QP#E|mK->A&2l$?N(oPfqiy#v8ODaXw1ogY>4+XaC1+mIP&e^bK1
zP)A1gLBT}M^V>^OOLq7s0)sFC7-^w%|I5phOHPicD!|}3*YCzi{(`>{1b#8W6RG~X
zq8%h{)v$%X`lt2R510mzVkUV{wovT%rcsfF7Dar@ALzkNjkyfbvD4Gti`7pXlPRA2
zD-WjoE4JsYZA}20r!YUkT(hh~LCw{)NpUu!jMwqbquckOaUPet0kE<j!WKW(g8<Hu
zKBIa*VR{~4c`m1UE}3~A>sc=CSst%C0a^n+Zo@1x>`q^R07ay&5CM`Lg?~Op@E(Pq
zX`&F!-#q5AT<WnrZpvID$~;DQ1HhXuQM!7XUOzr2g6>*2WLg~TY8(g$VyNaX0qX;P
zZBSx-f!OUr8ee*pz<_QHfA3V#TD}lQzlPSFkTn0Sm5|Mqfc_Vp(bn+w)@bZTj6ZyD
zOo?D>skA)QV{eud;Fh6w7y*Z>VS1@CyOn;3p~GTS38=nlnIp)m{-6{8G%Js&sFA_c
zR3qgutg<(SMe-op^$5{+y(V-i+jFVXHR^Cic0b~`Yq_xsAlLGE)^cfz^?*9=euL|S
zapo~O?=VFUb36VKg)!!vGQywQ7pOjTx7H<szl#-jO%#_rLLD2^Urp{wnjF(Tto5&q
zRJNv~y6{z<k7p`Q`Z6`-HDZzJEH+cpnGeZePm5?<*svg1NYX)F;1_*pqD^}@tOK5W
ze612tPITIsvL0U9yg+=y@@V+1>l)^@%xUat=z4N`ulwZIh^kG~v*^#};+GCZK61_c
zMLtw^vEUTUv7gyC1x`M@x_4pg%=MlrI3;(SnlV1P@4oSiZO!Vm*_o?dYg6gU*(JuK
zm**neUdH3#_Sx_m)LnpEFIztwJo_qpYuALxL74S-^8Ha0`LQ-7%OV?VLwYu?k%+d;
z{PKC-YKjpL-##?hdp*vQSyqE?_Gpb7ktEs*$A!FeXV-wXjZKTo%KPR16Wf#Ud%f4N
zw{bm5D)o7mrgSax8GAQ%lBb#oC*-MGGiOI;*VMmlBeUIR`}NkW4bm%Cmy(Wco`(!-
z4xa>4S{@13gu@GobZSYj5BK^avf}A>U&S6B@|_BYnx%9^dA_dn{t4emGlyrTtZ!~R
zd`IwVMeum4?rL&$@#^FY@#s#?{)?nex~}Oi5zuSmYr$);YprYCYm{pe8_PRk8#)3b
z<FSiGj-!>SL~!GpSb0{nvbH_mSZ2oGyuWG7%;jC>ckjvRJU#R#GkQF&rpz>ptSFl}
z0*So>T`T*JMu$5IN!Q9v7d;g!EdF{?P5ou&A?8|$t}VEjb5`mo-dwvfeQubPyh|GO
z)oIi*>$$U$I7h!-vgx0TAuLAu+_cGV$;3ZFw|qkQ@M(SKQ!^RFGMg(@kIKU^_6%o(
za$<O5?)x>qO_7|dwPt0+C=ENzO(n(^C8y=2ZQ=yP6Dj5uN#F0KEzPV?%WOQwvPM#D
z!8HHNommQJe@RdB5RzCro$I&ffW*+C*U-}D&?1k>QpC{Gy2vuK21sP}kn)GFBaNk}
zQ&bh6q(S0slXrUU#e};^*FW~&?v+78_K{JsJ;g#>ELsPs&dg}4Ouy?k3>OaUWp;eF
zHqd8w2tIq%YzNZpTaeBJ^UfWtczmRdG3t#zNlnoVd<G^q`Hi8<k3f#s-ymrbTo{HM
z;|850lUftwn=nUsNBg`-$J)WQ0C@VyVa(C#(;E!pxz3|)Rljdt(yuJkR~hA_9<`|L
zddlxX%FbR+ZkwS`m&(C&%h3u;5xVW&7|Vb7D~Zc1na`_!F{?o`ukl?kOdj{-L$@dq
zx8MdGFd^)aDX#ItuE~pTaZd9*(^isGra@SPP^xZ~oyz=FD`O^Y$p%5Hddh{U*1~4#
z#pec6OOjTz%2xBvT_uNaLJ>`}7LRd#sCi<-Wuf8aso@oY`AYH(ltput1@)9AcSDom
z<%MJA`Q_!sXXSb6<wa}d1?}~RH84=JB1ZDYMsmtV3eJbJ(qNP+{)rlHKcQyG5FS|}
zl><X5@N?K7$`PU}GqEgVu*gdcS`><yHMW<943#I;x%-Nk7VR+$M<wNwrI#0{7c8Zh
zG^^)NxvOY3Drz-JZZ-+<G^u+vN`o|M2Q=~rG>Ho)mFm4I?lbE&EdIS@mV0F3e`FSi
zWf7!l6x?T#s7tS4RWFQI&wX;&>1zDj_2~C@Cipmy^|&koJ8KMkEa<$)(|io#d;qvO
zB)B-n$lPJ~9KB^9gnS7%u#3NC*DrkcFR+a)u*)R2O@y|Mh_DU0-s4T#14`LNOxb5p
z*@4{d7qm7OvDOF2(<Ai$XPd3}`zy`p1wHvtM=jb%-^_b(?<0No14QC3CUVy-dB=!#
z@7c01!*X!lvY*Rxyu)%>!*bH(Vv77?==kCv`o*b+g5ymUEoZvZglmJpn#ooIcI|8>
z9V?++ijx$Zqe6?51mCI2=B(D{G7rxT6VHMJqQZk>%%f0DrLr<)h2dKG0lJ?IC&$m1
zJ34?HAOv9fsbUBKoB{R#VL&*53?L4W2Ppsa!hEv>V1UE{;6N5Xkokcf02~+;00Z3q
zg98AJYo=B#kMsaNLeMZf^#QmTT09LE3RTTXnLK-(Z2BJxZdks#YN01MF*8$5+vGeZ
zE!tq+_^9TUA*Soellul_UkjuIs~5CA=!BPd$(<S>0*KeUPH}?FC8cZe&U$tS%Su?6
zxzlt*IX>_R&NJ7ZsP?668;!B>wmV()TV-U&yyN@d7?g6j>$rQk4)tLPlbNBCD|CUM
zlDfTgbXgrD37uyBg1P^wD0wIi9EA<AO&`Vx?v~C{6VF9`nhd*lCMsCiQXu}Kmnwo{
z?DMM$3Pw`iaiL}Dax;KtVjb$=$VdvQDT>2fa${A>CFn)5WFqryXLS@qqmr|<zCw_k
zSseVlB_IF@;1>Yur*Z@K{SAP3&1~#~^wb!(qyKsWBKQSN0_4yB%bz_o0wCr~K%2b-
zwnZc?UU;=$Z#Y_lGc4^%5q`ebhQp;Z$7;RDSvAjYr>`M}BbuKvZ+6irJ~3on-u{Nv
zp_IWn_4&IjD}9Hpq2qdYH+Rpo=X;y+1`p!>>2m6)O-xNd<?kSZ)~OgFDEcd(=pZf#
zROK?UJZ7OT3^n-IQuZznNWX-)WFv#-5w@IH{-N#NzNx^f2CZI$-V7@owSaP!(vC5u
zdl@B%f=~)(qk|2{_cIaFv;2(5K#B5K35CAaK<pD}iOs$K)j$7u$EALYTjdY3%@3pw
zYPu>xrJl!<jl34LCP5{@lW11-xHvq`fB8$H-$u`O6K*ncJ<pnA6SPNnd>Yca^<3I@
zJ%CUL<hBW7zH9;@UgF>Vgk?oh^7l!Vrnr!FCILBQ=Qy2p&XTkz)Wwv!xNVmPBBtlF
zJ<e}I)uFh$%Y>xy`0TW=r#*px+-csxYxAr6U~b|;puUFg;@lLjIqk`sMl*AQwfQAI
zh08xVntso~_7lks<VG)0q}xN79+LN;%(@M*XQ`p*f`QFB>dOq_pAtYRLasc)NGO09
z<RrI1dd&!RBaK4+*B{B7a#VhJ1o90ha^P;PV1L6Hnl$NjPr|ipMajhWFHEQDCWz|p
zx85e6v%$d*^^Q-ph@K87dvEX49SE76ubqEIO>7WY>VnazqNPoGZuZ9q^?!yx0g)XF
zovK|q?rk}s*ixV=8_=-8%;DZ=h`A~c4TwNdFu8tPt7d$U_IyeJz0XRB(5$EVOlDFh
z_QuJX?bzwP@h8LF4EhsTBFA}BJXOVqd*PEUlIwEUx*OcBd7w3(wuTiMB5HoVMWQGB
z&J~F=7pIj4PGOToA<bcpiIF++QTcg1<oUA9Us6qG213@__2danEC})En+zvUFxJyk
zmo3rEk<Rb+{0$r{=Q4*G@0KAjd85RFf}+L#2B%Na(i(+5UM3zs153D&MR&q$>$;NR
zjmy4Kma>tf_#z&ZZ8}J(M3)TVf@=nfsv;i6-{@<^Wh^G7w_$zkkzrs%g;JrvRBC?F
z5>h(OdAH~0I91r_IZi>cu7ThXqhI<#63_-$vOl;yR_^r5;KaE|-=sukLkCY<{D#X6
zhbv@*%_2VC((1qL-iUTIcbidYoyWqvDPd)SU#Z?YyM$dA3n7hV1udz@82k4JeE8Wo
zz?n4^x6JM?o@Z{kda>m>xXO1P^Q01Ds6nk25!)285^8NfKp97v1$ZI+ij_Dw%}FJ9
zQwiHnwJ49t7URwgM|sKW;H!A?N(fDaI;(JFey@RqhIc1@%PJ;^Cj-~@;`KO(Pa}0Q
zVf~(5-rzOn7}^EsvrJc$b1;yd>H#vLFaCxmvwXbIIb5-lXL_5FG)5M>xmI}AkPf&r
zDbZBO7k%v_F8~!x7P^WJ3qME)=#Mqd;xCBZBk-5dcjZ7r3gjZA;Jur9L4YeT@1D4;
zE~@oNw}XK4i~|~{v>jql$8Yy#U>9i1nG6n0lX(Uro!DQNNcNfoa~l9fTo{&8>d7Ub
zIfKPwe>wU7uaQy`W1&%P$qlz-++}BH)yvc9t41cC^~LQ|OFr9#`Nib)^YRvK7&r?-
zYTNy|O$VMfx<bZ&=mcmTyo+;eoQq3b)l}?pVSV?2?#<a*xsqnf$>5<qxQMAmhSP?a
zpj&u36sM@k$L66nJD1*gQyHPr<}FnX>WrWo?ICi`Q>;TYq&^-h-+4SjafZX+X?(N;
zT@nl!aG>lc08$MKY}!n>P%&bBW>>7Md{4ROnLIT=OdM41NDW!$kb2tq3%?-Duh=6j
zTFr9%rQ4y&Ww>j*Vcgg1Z!k#`-V}RI1qZva4OGto$Hyu{ndYX;=@xvL19(_i+%X6^
zT2!*wBG2f$M78E@@kR?2`M%UzwU;3%0D2G&3Y8fz)sfMJYF&mh7!l>Kv<9rdEn*UI
zYj*g)m5tK)1sXX*ExfrJ_(kjD6#E56WSE=Mz?;&49l69$S*1^xARdc<n^)PgQP8p6
zXR0ZYtEY}CouHyk$y2gWL8dw;Az3aw4}KaeaI|f?v6alR8Mb@XVKrfiq|aLwC31Gs
zi+0PC%=Pg(j2T(n=)JfjzvSY!bwS=%?gXGBiO>L>qZ;T5`bzN^frDaT<SHjq32B1l
z?r}{)>2FLKLebK_NlGUAcJxP_sdv<@HyqyRThHL#YoN>d)4$^k8N5U5kO(RugL~8#
z-5j$|OZTdDyeQ<lWY`WujW9b;HDO6C?UPwb&x6BSGZNvN(dE@@f`C{u?X5)#Zj#jW
z!^&zg)mqNs1tJP@{1T}RCIERRS|-czlrDki4f<A7BXo*EFCNTTuN1Ms{#1X*ySeQX
z%N})lU-B4jm^tD!y@*Ct)x~JG8}vmVwW?ONe9^b4Y+l*P%6`g&e4EJv*!4owSX76_
z&ynGOYUxE;j>Bb=Uv>_<guQ;O6_hb3Rj=s&MIO$p<H!wjAE?BVoh!|S<b9_lcDh?^
z5JT1^fB?w*i<FjRp_S2(o9d;KG#TgowYbIAQM~@z*v0?Y3%9xMo4r>5is+&9jKFB-
zhP_k|bY2wbaO^_~7k2X3o~4a$f*P$jLyDp<5m664WvAtRC6VJ(OPz}FxVM0~M}f!u
zYakLUi>Kz-0v4!yxKTRO9uA;|R!-6V-5Uld0oE1qZ7VHxYgL3>Bg^2SR2;*KlpN|4
zSuM46g+L0!T^ubWO(UiNZiry+F-KubQ)|j&T3alWV=2_{pIue%DKThXJcf)o<5wvh
zzmB-vP=9f#x1h$1M)^{0AM5NJaTQi@(ze5_s>pxG;jdp>9Pn5#JgHhJu{W0cM<w(q
z&mv8DI;(8jf;o_5r|Gr!N;&hfbg*iymTgrPs}#2$^0$@ibm9Wz&NbNCu<-n4scQd*
zf0Q|dQpEObMD{AYY-eC^?qZJUbI)Ce?!*iGq~*;%!{1;ZbHoV3KuNeZeT#g|T_-@`
zx`E=6H(~lC`H`1Jg%JuptX$w$5VP$0TZd<>HI9~aq&+dOoi@nbGw9F5`{N6WoYILr
zXdsyCrI3>ot^qOa_mU$F$&h=jqH-Ws>)ok*z(k(n@E>bmd*&z5){T#x&(#=~AuZ(j
z5VGVPDocKv{i(dL!NFPl!@RE2cT8Q)c=jGIg6bmF%PIufq<PFk8T>)|F)d(RNtBev
zjZ#CDMwD|o!`O!@u)}<mZ!B@<$=q;sat0;22o=J9V0i~*bTI{%LX+|J{<PkyjUCT|
z{`tltC-y^!XX~?cZZxq|b-}WyyjhWu<;KA@iC1d|LxWpc5-baAmNP|grvNTg$j6|G
zth~T2V^$$fl$%#lPHjSr8gM(XmjOha8smTYCww^q(rw=?ze-mDge)SNaqRLenh*;k
zoCL%Y=Ash&7`u&s?P+2qVGQ|P*7VYRE}NEz<PBME>^AOE=7GX$T|SXsX=W0%*rSnT
zet%aH;6npbLji1T?V48*b?I1J<EG<8@}i@3aZE%5ph`l1o0&J0=^5{2#dwx$%4-~K
zq)eb@^Iu`x#uixEFk(7@-k66tp9gXy37)oV)mbZ!r2g$R_pMJTFU-gt$#vh~NM03w
z+`Bk($bb8&sL3v&19d~)T86*2K(Q7M8WAIZ3*KnZCsJnqW5!aUTS8$IUO};(FGhI!
zF~VEqLy7<8AO(Bke6>-Aox0+=xU^POA>^bN=?TVXVH3|3By9SO(6|nH$qLy*Kx~aP
z5$+uEB7Ix6S0;(31Eh~~+Bpzz0S3}V46w>2mMXP}DrE^_x8}!*K-M9~d|R7`K({~p
z_u8fNHS`Y`<gA-iUZK_b{mEh(%BIude948Vd%o8C)wIKfNGrKHgDpQl4HY+UuPQoT
zmv~La&<rm~D4qnioI;$u!SSZyAZXBu-^59HjkT-5$LT!U?Xj~|+oNaEV)mF%%-iKF
zV>{3p2pvr+@v3<jNkvJ!lB`+ZBJl!^L5ZrAV#*vr;8mGX>d#y=LXA<{1bVjs9)+jh
zMJl<?_%!;}<h;|A)Tu3)Yc?JSRBl#8caun=VHv6Ct8=n~A<H{YxF;6BltoztF@J7z
zc&9Lg>Rqs>I({t9ZCU(5lHWGW#mm1(K8g=zmzR!dUaIO!sqhdenLqC?0v#Jf^{vVX
z<~3S=Mf4PbrdiUZVU_h33T*>&mU)msWWg26?J!_FF3o8Op%RpBXt{81G&h1AnP`gn
zGk=3|=~~D{!v?9hFF#hV3IsPnGBQWQqWf&TH~CnXzR{mcARy+SBd9`J;AGg!v5riL
z8Ze(FDUqd1mPb@OxfV7*bLgcLYJ)ijY#1`?WsMb%aAw*F$64hZp%ew?#>08X+&mcs
z+yHI5OG~*#`z6<{`;O0sG9I6eQ&*RFmh1IK871Qu%TmiN&yrZm*{=G7vuRH5R(dY;
zW&3P+e2WKg>z!6<(hx283fLq_A_?>%CydaOC|c^A+1UCyG8iKG_7xUopjEwgi6gBn
zxUNSJWeO!ZuV-_2cXNw-V`;;9Tucu0VHVOGjn3J(C!2~b0@zH8l9)xww<l+=#3Cfh
zg?{R-^-Vpj(I_;#@}e4G;bp>aEg9U}FDPmhp4k^@D#qFwZOp_?dEx`h)B7rmPJ8e+
z0eh3hn+KM=`-b(k0^91Y!H-?&rp0ikWYimxXdS;$CC6S=2HSNqv-Cg%F?;_gF|iU_
z3c%(dM7DCH(Du^dfkUXqvdXF(j*CaF^@fDP7s=nXU}f>d-fg>nR$c#<Bwv4==geR`
z(q6NKff2gT;Qbiu>ps%em|vadDI;mOnBx-h)DSHhhK{N2rXAb&#B$R=Wp@}-;5QTg
z6EvWu2^A+%{ySyhAC?m2Gmvj)qV?0iNtIJreRvP4&fMu;aEhm^(LTDA<WNLj!61`?
zq&^ohb1S6h(A+NbwXL?XCuLjdq9qY_3&TVpFEyic9LB$t#~}A2aTcey!=q`#`1t4w
z`}-E#m?6i;{8R3jTV{yCb~N?!^GOfVw8}~a5GY7DQfS0umyHo*)-V(8iWED25~JhQ
z7sb_s6_F8G3L1q~HZM{;6W{Y3oAd2fQ-+X@Sfb=32275Xwx3wYDyg_4I>sSo7*yTf
zTE>-$&Z$^#mF5)VA<gE9xC>|njPGKfaYdwyTi05z8Eo(gmuw)MQ21@9k9`tB78tl)
zAZy0Ra@Ar5W7>jXA1(AMb;6d?dzGHUNNE9A+rw`W*jAH7X9(nV(g`z)ZRGpZ*OTCu
zbsK@&(}T>K?n>yXqf-5BQl-9ePp4t<G%$+BAmN_PPyP7SSv^P1>Fd?$<nP}uyFm6A
zKImV6$tL659DntE@nuTHV*{O?)Z*l_pBhN@kKWZf&(rXQ#I(9y+5kLskW6ZMhz<?#
zb_|SnL>wO@ZnV;&xI`c0%!(H@MA_aRl`e7W^(UkNg${<-x$_*YE<)`hZvkWqwS0Xw
z!2??zEp*!rY;h|d(|1oD&t%4(0pq9G3hNMETmLT0u;<4EwA<0Hju^#p8}%_49iUr|
zup`->wt<ay5fG-{AMKjYYLoKR@_*{Z&9Q}z^jEYvobC@Z1v%32Fok?H)tT(JJ3_nM
z9BI;pNM?S!j&y@&5eUO_&Lg}SaHc8+8?@C#DC>i2$hZ>klxOO~(T>WsRg6;yxv?Ux
z(Dae-GsaYAy`fIAiSbG|Uu)|S^{20R9f5EC<X;yRMy1~es+75tS8mJxfnN|g)+@KV
zN)HYn5EM0akZc5Bq#o^g1FVxxvHE<kzKS{sma>U6>Xvqy6%;6E07^_>yMhq7sF)a7
zEI%*dvcp?ajO;9lobB(E$3aAOUUiKZ-Z05o2K8aL6)?wZe}-_PoCT;XffRkbyU)Ti
zl$xTKVVZ2#c6PSjBsp0SrTrNg{h~_PmcS{fuSIze;n*}P#`<ZXt242o@R2W<z~DZQ
zU^bd389EYjBB_#UZTAeDT&wewh2rkjwgbb)Q?1}INtt>Rd-he6(&g6#!L5K0kAzX(
z=N(<oTs#7vwkN@5J$wya&)_mzhQM=zd)75u+=G6S_4gGh3W6LER(~q|)$WIK3>2B_
zP4eu_wB3539mH3)2>LJ)1PIiV@u!4{>r#Zs@F5GcAhwLsU;UBbj-;T$`fr9&#2I3S
zxGWM;x3~%lTzgTyiqE9z@0!)!gCuUPw)VTlL9)NPN4y%fyiYvB$AYle;ln3)oL|oT
z^V3)1)gixX-uR}k+x4ZHTn3+x^T6QIwaJ?d@$2PFry;H-6>v|{G-Lldt%bWNV}#IJ
zoEJ;$R<wdke3X+-vGa~muq?<5WtcV34pNL;{px6Nu!?H(c)qT(vdkFmc4%|PO!w-C
zKB((!Mj!dKNI0N_hN!)Slcs9CJgaY3sVqRPtmyNqWFM|;7vU_56FzBGf=AV8s$9vx
zogI&a;BBxhM+||tT(97*b+{h%;jJ|aaetrNy1)KD5#5%Fj=g-|@VMp)nwIB7=g-VP
zKea}o{<}O6HKkCOO_>5by5Z*#VAhCby?2v-z?v5NQV}4UyvP-O@r`Oml;eGt7m`hR
zLH&Vaw*qT*n#<e*YZND+UXf}e45OEjXiq++%F={UYgSRz96%oz3862Vm)zN7%l68@
zOLn)DiIJJ6;Q^^#syW`!2c$Xa5yab_&V&^H(NlRwMS#-EVNTSWi0rt0^IY?=I`=s}
z=Hx>^vw(sTDW;-hQ320uWAynAZ_?1TD&CXY^QZ0_FAv&G6+vI1v+#PWWzJ>k+*l3%
zaqg>y-;Az>U%~~Eg47ne=}*$l-%NVGLgR|4fnVE1zk)*}6o-Q-{N0(d^;?4Q(}t+%
z3A<DDdk?Hg0<^-KYUBobb%Jc1Nplg;&yny3uwr>ivtUgbC69-*|7x8u&hy7}n;WGI
zo#xn{>M$y%oQSN+qCIaW=gen|XTBFSFzMeVoDP$4c%iJ#0`Kn|JM;_2Q8WeA052W~
zM|5j}^)UI+Hy{p6W1$jFMp;JRz_J`EkAdeR8cRqJKAGEFS~?o`F{>PpPmn8U)LS*%
z=3HaiuRmQ6!x_Q!BNt$ncn};p8X0qOk%<Ot#>~CKv{w4DKIfsBpu7bE31_tB%UU`B
zfPh2<MTirSm34u!f&f-7bQ77pP!7FKD$2(lX-nkS5R@4;O}VPQX#Caij11&bSjNW1
zZ|>xk{g3|+TpKO&wk;qZ*w-23(6(=?B`qnjS>f1P1$eerC-3jqN4~M7%so7iaW|7|
zOIDXy(9EVC*_+PNmri6$Eod*ztz%Z&X)bACUw79uz0=2BE^ihE56{65$VP^u>U-?_
zhnmGm6@)$NdLKPB-{OkR4Xe`YRGGa%OeBy=@N-B3oV;yKH)cp3kL49m@CojZ9wwtP
z(38m5pLnm-(c>_WY?_FiGgk{x6SVmlDG1%%g{D#R1<;t7S~Y*RBR4!(@l5g|u$z>v
zjb&%MY+0d<R!R&EpW?SsvMrW4McfVUyYOVJO>B;Z+Cgl$&@P@Nr|Bjh46PJs4xDcd
zciFmaHLi~g?cGYXMDF{9E(ZiWByc$<4NnnI60{8QREJ|H>BACVWua_>9P(aEqxZ2U
zMg8T{xyW9MP9!j6o3O()j1pietEyJE>$^^?FQy1>JdIr*(z~@@RSSxZ&P^>9YC613
zMUCC>kIY^$fRRP*p|5$`3RpU8Jx#?TpW58fkyp0Mg38v)z}`qlJbNWCQxsl`3f4tz
z{Nh*6DCWLO5e*^{_Io`-mj#OEheOM5x;XAW`$%txxW2ueFTFL<>wQ(|CiDg@@YiZU
zd1HVYJhe_XREZV3@mZ=1J84^6|Iv;zVE;K^90&Z-P$Flea-J*e0X_qCB*@aX`6aD2
zNz7W)Iv`-IaUE!5oZ@!94uoZhdWfQJwDAZwNUR?nHPcKUl^72V85hG`O;|d*k1ASd
z<7sl+>e|_^YFenLM^71VC>R7R{bP6&ZhkhwV*GrFn8(|+cvp5>v-tfCS5;lpZ0m)B
zZ-dBuam)VK6X_CjPnEF2<sXdT)zNraJz*piMRod`{I;Zrr*n#sOEgVE-D~A~#(JDc
z_{VQ@-g<cvw+xTiJ6{q$4FS*m8Q@0xU!NDKLqR3BH+OxYCjBNs-S*!#<G+r6gBFGl
zJ@A$nz3mc`S+Ur)z(2TK9SR8Ge((xi`*&_Y{i2le>~kI@HJNDq5Q0yXobTZGUe2l&
zS5UsgFD%v}O7q?%s;#1?m&=^tjcf(v3!o4)ev`5$*<xsh<uc7AV?et$#IatgdQmPU
zRkJjuyG)xln%{15C`nmmT`hrdF~hF!&obQa*X=a^OUk~meJDmY{w}W8tsb{_2AjOR
zvSn4baC8fJI(G?)c8pmk#?@Q@9G34_@Sc8NH_nerL+o|4{`9WHv&}{b4(gDD!9)~E
zYXh_84P?iUQjrT4neBB2QcKBB5nVFU;DJyrE0XAm`nzlr)Z7~A!}Kw9vj!K7_l2ej
zW8NlLlF_zDaXJZ{UWJn?Fq<YRad3<TA%;o!v1FAI4cpQV;Q|XoYocDv>{K_EcX~=#
zv249yT8!p0iCy3j&|>u@TJR;t>@Ik>lB44^9{|z+ua2jz58@4H)?iu3Cn;}L+eU}?
zV~fwj<vYn?wuH+c9+c~CC}ik8md`tKA(k#-v$xH1q+-F567poUc$lvsmjI;Z9m9;A
zMM8q$`u2|Cnt{V2_C%Om?Tcp`K0YPP+ZoMCoT4WoxrXkyk-z*62%7$5^)*7)xzFes
z>Ec;aojYHE48rdA(s#xNbtks@qXAd>8;HF_c(he|QYn;P4I2!;R!70;taQ)LWY)P&
zcvfFWuC(SuOzx?0y)vt80(D1wg^0Y8`NA)X+v0*-B@?i5%5|yxsy$DIW-7O|p6VmK
zcpo`ur53gI89Dm^y2RDr)w;=`U<V57A7uCd9B-3Ps{`p>o*4(|u`8V86&hwx7jod-
zJ><KqmZ+YR8>}V4@p<=VigO?+w`@yPIsqP9Fsa1wwdxI8<_8j~LStl<mOoL7Ij%Tg
zFO{x{XB$suKnI0Z9T+^^AU7`7iYBw$$S%8iacQ>?9fJ{#(ymXUHy1f#)8ReWtHvh;
zK4@zVsM()3YuShIwLIe`7$2N>+#{&JwxxI1W$Ii$^VmDBKdtiBT;UfQ&T9}tIs)Nm
z-SJS}M?*%0OIQ7qfoYVc!7BHu@7*Aa8d$5MI`=)cEnY`N-SKriJFovi?6#Tgwhb2@
zi=8^K+-%T88d;%THsHUCjX!t0-2H=i-x0SQuAHG#hkP=bCa=Zy2;?jN#DQ=n?iK9Y
zdRzy)INcI#ZT?(;qn<-xhFu|`w3t8+OA8*L{Ia8XIHTP)_D4I`S<7VBP%LVYwRO*&
zlk_~p7n9m``6hV!ZA{`pH+Kc)b?y4<5fo+BJBC~9zJ>s6CH5ytT3vB%E|@)wD@$Z6
zRz-Z0??Z6crH9sH`06pUG~|oRj}(ST_%DN)RfT8csm14uI^VeuwpSI9_PFvbV;;uZ
zza=R&g+fX}=|&IUee-F_7gifS?kzs{O?s8%>8Yk$W$ywO2Fxg#$5xAKPg$O9o0Cgi
z3C=E%EI2!XRa?1Kh!0d<uK{P9e~;4UY;P>hqc8hY$ig2@gt<`+w2`6uXoak2On#g8
z@|MQ)N=^DzMJWGzWS#Rg&Qf!p4|60uWG`07Ne%u4URQ5Rdee5p?Nx(;@nJhbp{8<F
zL7Hw(IKj3zu6Z&nx^#JE5NU88L;S4$kGNWdSRdWQosY6Jp4>XQqBj}qy$+HMxH_*s
z**5!+?jTFQKjlB)^gqdXQ%KU{)$;_zQ^GrQx(V8S=~P3}bZWP&T>VV&y?_w4{ob~H
z-<kw4Ak~{w64(m}ju;4noJ>pU^_c2#cU3|^gV481+L@k@uI*KqflANO)KzjZPN{P9
zP{;!g;={hySQ_#4@8&*@>?L-4v|af<KPMG_vR(A!JNv>Tbg2cvsaQJ}JVaGXdyAGX
zCbU1wgb$qpdq3XB5vDq-UQ{XvB@$T7e|AKVg7!^pHY-dP#%|cZG`)&dQKHpZzF@E|
zwlQ{BOvuJ^Dea8>9vb>})IFhkwKPU(%5%}0xeYvi6ewlX^!x06uX}e1P9~zeZ4%?*
z(39UlqoI2c*BbX}(BXbpI)ddYtgzACigGVZgC=YDRnv_RWY8OKKu#irs<9{MsSL9E
zQgR`h315hU@I7H@CNDC4JaCRpRv08Nj|r`LUEhHac1gF#tR{?z5rP~BMsQr?tmK8$
znfZeDytHKB9@nUwX=WMtA*ouTe6L;T#a*9RyguV1i`Ek<^lVZgY5G^vwCN7!o#Jl7
zydj;Bm=Ks?CW?yb#A`E=5>tPKxq(k!K7S6+c<EnMQT^mEflKTZQ=00$pi>!1A2{B<
zzq^C|LFjfy=D=PSNUj-p%zdo?pwFg!ZL8VqC9egDWqq$A>MgK)+EvCM>!D<*ue3Em
zZ)^wiU6-SMotL12y!ED71T>+Vg(Qm8<|H8$2Pn#kr%Zo~NI2f)((Y($3jay!lBC*U
zpRrWkV=`E8j@ljGUOWtecOxH0bF2MGA@2AoEC%IB%CLPD=nM$jnU~%n@2LT}uM=-0
z5gj7>#z%fP|I$c1joWWa*~hG@`Z(KL4vM;+0D+G%jUpzX&1(v#d}$%pzDx>zyx+d=
z9bBH;+nWJ3M*we&@*~7}+~Y4l6yP_NaYbd(g+xj*5P;ZiSzR254Pn)tPc^#s6lNC?
zOk+)<em(#ESxp<fn8Aat?aqsL&uWhFFmf|7apTQ_-nk~gFEAHvcJ`8blRi(=)%x&g
zeOcp(zO+9|PN((3qaOjS2=-jU4b}|Or7jCW+`Hc&9<hWUwu_rdQ;nrm(uQ**y%cgr
zCz*jNB|<(?DvuwKjMNQ7!B>7lcq^=$jlU%kaOGurz;)$h1WNUBw-{uilBv8xRcFdd
zj}n-~s@bG69no8TLdA-5xnSbO>B0F%&vanEF2igv8RN!&W`|=R7X6Da!akZiL+OkE
zhhTJTt>1L`>(x%fy_u?c9KGe(--~$lV5E8VOOiqs7wd8wHbc^yuc!3m6S1OyaosW^
zBtQ#Rs+nf=N-bh2`+DUh1bQ}cv~Wqip7%8P_NDr<V)Wcmh9yYnF1uX3&v-VSiblT=
zjO;Z}oV%BF7O4H(`YtZ`B{_3CWgY2z^x^7syBXeEf}|O?x#yLwe2C2_s9Bod6qxtM
zto=Jh6kp3Vf?lodTLs;OyH-thw8PVNIT9=v$hy^8r`OgpP-N#BXQK*WG780^t_5Xk
zq4;INtZjK=<N_kmi5H&FLYURRY#lb-qRB<AmB1DJ`L&BH?tPiHTJO~EF_O5{!X-E>
zwsL55!Vtwe*u2fzbJ~1axAL^kI=S7>lMEYuEe%KJ5(9CV(I3)m$+L25)4-Lm+b#A}
zZ+sfG4e!ELUiLZmD8fA1o<`8<PvnePi6@F`P`bDdZREsyzbek~X)(Cz@KPlA1pZ^P
zbI<Zp*(cgngMm~-^QsgS9liMijrK~~F)O<XHXcE)q8pmuM{j|Ql`0DOBfa-Gs^;pT
zhL_5*&lFBh6|Q=MBpyUHxX&Gso>xoZNG43I;InNSL_Y$GLC!*K?6EUUQ5QS4;~F!+
z8t@|3*hD%MU<C<d+2`R-iaRdtLa>Bte$R#DUBAA{g9jDm!c}Ckvaxk(?(Ec(ahNf{
z(L$`)V^%*|-DHx+vnGazrnO)*82wSI9PLK)=y%+0n{sQc*#4e3k2gh&x+zk+zb*^x
zl8fM42;2Umwrf6wN*47`mRDjfv60lwRB@>U_SG`EU|na++uc!-CH<W2Y%*gpko&`&
z`iZh`)5~Rd(Z%ZhWw@m);%VQ@5Z?{YbOI#c<@kh*banB(Xwy`tQy|e#?hBNgUtC|Q
zo#8JVUMELb8qi9i@a}|^SAa?&LQe-S+rJBnt};{G9zolNmLav7q%i4o_x84`cA@@6
z35R<Wh5OtQ{CtI^>ZAjP8U@AkhFi()qti4u9Zme%=0+DzQDc7-qAWq&fP=p#5j3f%
zV-M`;YnGRfG?T$~d7>WQ1J)H1FPeJ2qYwb{?cMq@#wiX6cp&=3lT%?#P|mLCjl|u6
zZBPF#`lz8l9sQtiTUk*DjZ;z*;|>nv;|^MNs_330-2vkg5b(4#_>v|CAhXn;`phrZ
z_8_k&WB3~rbn%y}65er1I5_G~va{txHcmInN&`NOTn%9Bhtq?&(XLZ;S`1fsya(W-
zFt0ijgj%BZT3Q?*J8wv+#e=kU==*fDXk*SMV8zfTf;ZFP4-v}0txm$a*K?Y<xuA2_
z->*`mxj9)`?Y}-oR#pn)rt6x_?YmSgk?0utu=3icAB^lXDDJ6BPsu70$Avo7VNmC0
zP{^pQqfWBjjPKG+vXS}d@9xHlE!cjq>7q&JRaG)4R}=M&iPTAo9N8;UQT>*;FgI!m
zQn#jAn(?w+niAQXBC|9}Z=W8w5~lUNETRc!?co-v{-=!2%Xr6lM@5io7MDhXCj>7t
z3O}celpwcjCASN!66aBt>INWpr`f)f9aOU=KQGq4Do!@S-@#?HlQ|@t+M>qTf%MA9
zUs#D(KSBR|;k94>Z02jHV|Q4G5uXBm#Xlat!sF@zA0zJKcMEPEq&OU>McY!`%s~j%
zI#dXvFFskMtLVvm|GK(HW?R*EU68V-+W><v4$KMjzH{CIFs;yBF}*OW3;1f46>PY=
z0`ZiY7yePgg;q>jfTBudW->Mz488utb#@!+FstfJDI~wUS8$~rLcBS5RD*AxdZF}n
z%jDqI#dBbOikEC-OM4KQd4ljsk3=}A>6-qgUuOEQm{}_<z2es>NsT6dzu}OP?Lqq-
z^Xe1$ZRS-d=NN0H^kis`>cPli@5g*}-)aK>aDX7w`;Luytv4H*D9S5!`X{g?TjW^v
zAwQy8Hd<-gX??s!I)Ll2GAJk~9va_^^_%+RdzO3X?h558%)iv{d5G-*Xwe>B7YAU(
zmbe@{6@m;e@r}%X3o%y8K?vp*A*bRsrnbmvlC-hBNF7}niZ9ah20dbL-nuL|54I5@
z!pA(IRmRb<AP?>O$?;5`z#Dq(L*pJB*O9no82UZ^&o!&)O{q29Tgh^DejztMd04i8
zsBR~uuJd<}58O<9H{8RQckZ(fVn82mA7btaS6~kUMwS!^ZRVd@jTy(`*UfqMnAvqo
zbsP)UWQy(?dOGh#hv{i>-g)fTRAOjsGRtz0hB@YE^X-hT`?#_kWC!E)EfnwMG~W0g
zDxW~_*%icsQlyr?cN(a7@+p@%^SC3&-h4lw@q^pGfCH#}4Nrsh7K@%P>O?+?#Od<W
zhSfiuvB4oJqo_9*`$1JII|(+_-9SzaspB*lhlGxMc>XB&s_h}NXPLodX(YMpfi(VO
zZUs;pqx-L&V!Y*_KR=>uKC*B!6}~l9dJ`?nM2_J=3SW);b3{Z9vD3HhzPz&5@Jbm}
zeoMJVZNPj~W>4C5FNf`y?%#9CdU@Ql6ai)Z%0X8iDH$or#f!+t8_4@(r>Z^f%E|?o
zlk`v8<e1=7zg{Bn2%(p7p@3A0?pej>l?U_CxokvA#PKN4CmHI6gWhT9!Z_diw1|eT
z3+3g)q$%f<dc@%!3LYIVn^>x_?8f0C(5SJCU<qh?o<8$NHHS!;Sf#I!>EIgbnq*t;
z#l8E6?}3A>(JQvPNFyRqmh4t#U|abbIN{-NfhLth0W6`<xY&Ec;+pf#cDd7d92I(-
zbQ|aOyia(K#6vYQ8vY<wIepw1BPDmP%YrtoeJlw^xHt<dd(hW-N$<!pU32(Z-+AZ$
zrA^|)$=p)Eih)JMOXacew_Ua-s5cr)JjSr$p&JVQgs+;=ZTg?7{W!M(-#++5SPW$(
z^}%BIQkjfK^}lAIweLqt$<ox#;fq{>srr}H=11H>Fyo(tU806Yn;MQj4sj6VY*txQ
z6Kac(2D77X-dtN-*{oRemtUrdjs@_bpZ5!2hlij22?qOfENt<UFoT>-SqyJMK633&
zl{!cc1P&d4lG^Y{*Wl7s5--o*4ciZY|H@QyevIQhD}|5G5-o4rI|0axZq_NHa9N6_
z$;4+}VJ_?{in{p?o+P)1xPq9GDXa1tZf8OrA>j_$Dp=YoP`eyDQZRcN!PNv=k?gCA
zk)h)(x3{N%^3Z0q)HINK+*Ahzb-)?luH9<6uwQ)9X5=KCr-<=?kaltednb_*(mIZN
zcz@mT@Tn;6v5tE1vbs%+#%@M)(o1J>Wdco`5>TKTNlk=`Od5w<NnvO<ONfb}x3D&{
zx>qqokS0D#OX@f@Qq83~Jz;m={=;)Iwd}KDuN|#sJ7t8GMm`;x7^GQm(Y#%`s`hMV
zJGGDiuiR^UI9Pkp+AlT~%FUNX;pC?C23ZC>u@Per$Zni##fqZtof3Q0bEbLveio&=
z^Xrjh*I_vMFe3zj?vx;t=~An&!Eo2syaEg6CQ(|OV{P@Lv^Y|(8kmuh8DeAnEg{eC
zN>tC}ZzZbHbeLqZ;PX^rR81JBvoB~U8(lSKoUDr1(($|-r9;EZ10uHA``09{n_k`|
z6Z@_|+h!W+y}7(hj&cvjz&8VO*zKAgb^K=*J%*HZex$<}Xe@!*-guJiDsK2m!Ydr6
zbTt*qMQVJGSe@r*n`}!<5Km@XfV!nk*a~aSp8ioZj|ciR=Z0E%fZgHIu!GM>S9fQp
zwTF(x*(L+evMx&J*00YY1C0!wb(U_er|1zr-2B#f!wPY0NWT5$@C&ic2Q{oNk-^|E
zW~p`3Wq!LEom7Qo%F-ZXy<JjukPa@oY|}pLVd0?g3=$^ezX<>fzl9u&hE*}RK4J5x
z3f^W4_U0P5d=8W_grryU<?Hm5F=H1SUd{IoESqQT*9*g>sY?f4`R|R~FfhSQgLDH!
zcioKaPMIF7KXWB&(tdCc?#N@FiQm4Nz!TF8P%6d1stk~NIDq~e#AiL#x<tELVOgD@
zonJnGY;zK`LN@9;fuu&zHIi{+bwkha1JDG{4z{Z_?FF5W+Zeht^b#Js8Z9=p`21c2
z_>PPxo3s_KIwPM1F3aZ|5-`Kh9NkZDNrOE29V~4uXf1{?w=tpuk|orHynK7p9R`Tf
z`}(eDefx#0KvuISqXe}Hl|nE~m2(A_SPHwI+RIGu!D?)rm-t=bX^zh}Dcko2t%H_V
z_>y3<?sPgs1wr6NK`@421`A;fzBB^pS}5D`=%&cREzBZ_S6Ax-5NSq;GqcTmWMo*_
zYi5z4o0g*e=1HG06Ch?r-I&_rnk2bM@H8hhsgRbptu8AZ%WSMROFr+vl?uA5{4#m1
z$A-@@gTtOYNt$QQ@uZ!DI);w=4YI?G8@Mo?PnjcpJF*|&2jG-YTCF9;40t)KViD`f
zerpINzPFLGJu#L=dJH&~oNQ5#j4LCXl$df}AFM}I+DoF&O(3V|lOXLNiJt&tO-pg(
zRYKf{*P=onkyH+`Duf1$QyP~KiUg1}q2Ws}GM7I2O3$=}TCF>M5q|5hC8tL+Yk3f`
z*%{fkI^_=s?Zz+TO>Qo+@v;km2a|=|yHC2x$f(2wn=Z8fv2XQyjr||S-Z40{sOuK(
z*hwep*tTsa9ox2TCmmZ)Y^!72wr!(hJn_x@-tV4s>-;)ZyK3#Kz5ndJYK=M9Tw{#+
zbgC$;Uq7)^{t_d8A|DHM%PHfd?q-?WxLL>u?9Wcu8@O&Xwvp5JwCTcf=3h)%E2Tbm
z390O6URK-{qGMYOCrW4mIQ@8(2@)<x<!m!gDjTcR2qk3q?Q%Yu-g=AiHD5A+_Bq&g
zK7nnENQQB8-tz?pYPW;6qMr7GxXi-pFoSc-!|D%Cfk_SnLg8xU{2p(tM}{lwn1A6d
zELN7s>qbkejJ=p{Q9{kCw;lYo;>c#>6p81@7P2#a$lsS$oZ#hc9CxGRo})dHev?^@
zCwiG$*iEBUECxaISyD2b<msVEYGW~XYU+rKCB?0CV<`!(L&5`U?Ccus>;}dm#Kj`!
z@={f2@p^h(00D}s7tqB%X+)u&*GuXLTYFc7>rtXV9{Cp1<qEU-$8J{xV~6`w0d?Tn
zJ^*HWx|AOcFs$R5C9OxODP*mjA^HEpLTU`T3t!<V`$!~A6|$5KugJF{<ykY*jN4S=
zA`$Vplcal0Jv@7j7>gbizi#1=Z3^K0l_pBlBI!e^;Rm9BR*hS+jD%OLnGKuP4J;Uq
zvOSr({bM?OaD8;^;#Vh-OS|LFEbpN)o5ayc=#J;dSUhC~qwCJ+o>0zL`72#BzXi-6
zrPzK-U})5FFQ&xr8Q?~CS?iCn7MMp0*^JF`$UII1TGu~ii=)ZVtM2U`h#aQIUU1L=
zorPObJI3mX?J)~=0Y7sf*Oh_*g|YLhjKVQ@$a1kZI~4u0CLsI^pgLJ*6H&i5={#c!
z^$pT~d`&AI`(mLZT8LhI$CGy!6vp2`uu4=boHfDuW@Mn~RV#koTsJ}9M#(P?|LdK`
zVg6LKmbkZ29xd2Cnetn(a1L{2O<}Q$|I`8@5ZQ;#9^v1^DiRPtqIY(BrLmGk8$OB*
zXo}Q9c~^jr6vA6yZ>3X<ebP3ycFwjjuReG(%}!{OEgZHy3UvjvGa?5>4LH)-9Q_qD
zH@R8@$sLl2Nm>-NOo|2o^NpGzldMFyS-4=ow(2HpMrxK4z5dD8^!Qo2I6k2C#M>p(
z_#6<A1xBa;ShL2zS)_=D^ABOdRLUAr7=!P!aU+ENO-wizeD26~<Av+t%8*^ubQ&vd
zD5+^;IaCz>YZ*ILZptFmdVcVaxOVvtSmEEgL{e0(Oy)u&Drsaw3RdtXUc@pgvu>$Q
z?`tF4hJ{762X_mcjEkw9(HlYOmMP_H<|a<7^GqC^7+ds#_mofDC*SM@6WWyMaGEf;
z6ZZ@?5x<(2J|p%45s`#JjY2AAbBm^nDKi@WLzqVR=|RZP!W*=^oGT65<quS=8i!e(
z9$0RhgtOX^4H~Vo+52U8{>UQry8A$OwI&;gWD`+@=PVQK022tYc9t<U1@cF2&ds4S
zq_bJE%n@EZA0K^ib5}}B8f+O~t&Cs>>!<Xtn>9AxqVwye>GI!UR^l5QHQ3nIUCxV~
zUMEIo-JzUT$1F8*b_5?|41Tw$t(|iGO08xolju49NR9Z5pmQ|!gT?j|@-8sO#JYFv
z(btvGF-K08Y-p)N!_73flB8f77Z52dZE0#m)EOSYD40_^DVd$kMw7Ffw3BSUq&@l0
z=bC<0m=p4bj)8jGP*^%h^Hg;e40yf<J~zi9#MFL~!=jc6ON|cgJk~n2*A*%~@by#h
z&=)e&Sn#+$p`uEzyN<)0yI8a;99`(s#MeIB_erXfE6|(U8pGOa{z<=(Jo+^ud0#^n
z8eWiJcd9_%xaDKtQOqFLjsO!R%YZO#(^T>zXdTsmm8Xe|<M%%~HIAjYenA609e0ZE
z8sX{eihN{jv(Gj8JsjH1$1D7H17%TBGq>&HJSG_-{<Zmh%8w;g_QhV*aQpd6-0+D6
z1&Kh&H*as-wGGYN)eua}n1uEu%&@G@j~UxerEM+WKux5eQK?3%^bJlr(N$!ml`}Js
zgproK2`WQjurT$FuN9x26;BG_ar%D}=yQ4Hr^?92YwW!N(<!q^l>ab?1<L~2rI9uZ
zPDRAdi;cPyO3&~kL$%nZ9L$TIbxFLnwU0KJEEMY%#VZo8t#&18x(0kvS=@gmurfAe
zk+dTc%5HuLM~bpn??;<Bm@Hg+0va^31(9n7Ef+Lj^VfQ4HKWk^O-AmZc6+{1z`h_-
z{a;dhu(C0?z;gYw(_cV4OrIDiPLk=<Zt#Sod4tWFC7-$(Pf`rWLsy%OjrH<y54qPq
z)#23pAEy-&%D$TV8D>Yj?Ny+h3(${6tKd~~gU6h(?XJRRC}v69<M7=TXf^}lvP|4A
z6d$PO`fZG-yMyf+7hYF!-;jO$0^jr~KvtnSDECM9SfYD<Z&1nyxw8|#hhN>`**oP}
zA`nPrdF3+FcDvk#5ep?T3q{6gzU``=L`=&Qm?{JVdKJN74kB{%$3uCZOfTwe+#1=#
z`p=TDqICKoH+)2&If$P*^yyD)gV%;yo()5ijVcvj>}N539Cr6>Ud);`M>Q>Gdp-Je
zm#i_*$nzX_>uonb=PsCUp(Ks>)RBf++WYNCnV~5O^wKb^SS}uq+vg%6oWvOK#^>j|
z5_X9vbRBaoesmS!X=R|{3u~&6Z_Ev&nRO0xJk6?#2Q7{!w1NhJ-id2%L7)?$l=fil
zzHpX4{^*eg`5W{0T*oSXIkXKI{>Ay26WWv*LH-e3fk0itPFjpg+sNMl__uLhqNYS}
zioTPZ+jF41{+dy;XC@IgQ9ZHZy@8_dXwv+2F4H7C;9hxAK!PkyTF)lnOkGsI;(_nK
zlkR+Yx^b=(SUYqi$A<oSF~0rrLP#_+sE=%K=Y2f<a-n7G-BRQy3Vp;^MR<>(NU(f@
z>|fl5&FNJk=_Nk^KX3qBR_%bjl9nW>N8-=`O>j$^Y01q_Qyun)m1hstqalvJkErrj
z|J`MSYGq96H<NpGoP-hk#ev<SLga<3|7fpTwd9NJj0MZXe7*=CYve;`lX(J7UY<m$
zo8gz5LsH^vuaqCsmIr)Px7%0P6~lYDj(+9TBV_63*=e1OOT~z$w(}{;Gphfdh}XcX
zfd_vg{EmSG`7n=`$nYFqg`$YHGWo?FnGxPowa(~ywS+<8&a%LM6nO{gJU`?@SfF39
zfbbL55toR>{O<5#h08+C9t>T&zn?<M@;V%Cb%<dt;#4?5d`cms4bCGU@YN5?7D`R!
z96)(GueG@-G!K}s%gW8}K4?TPo5Q+oUOd79&7p}wx9`NGYUnbr?`T@%^>JU(SVVze
zYVv=&qEM^u6gq(wucEhaHvGrsnnYpbT-c`UtMkrQJbZ>~3s0eYgI0Wt?zOqJW5LiN
z{H$OGEw=OnLG-u#xwMi^s#o$G_l=E?Zqbdr^p_~?{ssj(KDFp55;-{|168Mh@G}yB
zs5mIb<A?Xe3&N@-IX9riE+93@*QIT?&neKaid)E4Y25j5yCB}SO7v?_D90PAWrI*Y
zS^>L;bd!)xY~-<R@0Ll%0PYyp3ze?A>C&1E|N8(Be`+>m76r(Z$G-F7ioWgBg<O|N
z1Lw>2AV<PkcK9#_Icq!O*Wl`BXul`-SFvA52!=x)-4ugzzJDw7uGzglz!V8g)={a#
zoKP)V=ZSE!4{=28Pz}u|vD{hA7kr`Z+>`zQ`#gK-G<9p~o<e~<*^22lBVG4zgi<G7
zF?zE^#I{c&bmHND*s`$c3n`nw?<FkI{)`I9fHpRw)$ApK7_<4dQx~y{4}8Coo<N=O
z-o2OqCO4iXT(k^HpDx0dwIgtZbSep=p+!&l5LZasVxMCxQ&+-NN}W|MkGkb3epwy=
z3ao6;DQynPK|}cAR4w+iRm7<z6C<^`&mRK`kI(J2+yCqNjj4F&!+pp1c`yg2{v{&_
zs%D+#mZ1Df#~s#L1x&!Ha+Zh(Ugx0??KH{c-~n56;RbsvjfILu>o-jAtQv(QX?v_w
zB>UfbKbDIgsB^<*e|cO_PUW{`H`!<E12@BUdxj<-Mr-mW2O5;Xyg==%9?5Q!0n|1U
z2`ALHgtNe$d`)kc-rTpFu%AQ(US<9+jbB7&lf0hN133|!0Li6VM!fOT4K{lcd!ver
z!F8nhw3m$AazdY_D1nf`XXxkslvdVHoB5O}T=llO(YiYRis(9ZEVA5iV%gfua&yyE
zYqZC+frW6~$y{m2$?Q}KegJ2;4hvU09_NeUd=D^5%qcBwEvcri;d*(SRG%iAaz;7v
zP1;X-#>fVpKYT%998Jq!<(T{EEw<Xp%N;<YHiUu}R*g2ZtH6v>Wy6We1qzMkDa|9J
zD$7e1>2}u)Md0Ir{MttONROo3(R!?-Z+_BQ@n95hHc+m(^r4h@y9ZA{dIL$Oqm#53
ziUN#c?o<n9*yF0ltyYDQtStE4_kL8pSgGh7zuR9qx?}741>|=Z#{C&1tc8j3i86#Y
zCZ0-FhwDm0-+^{o!}%G0qE?+JK;nF$G^ZoK^#G2MazP~m)n_>v9-&2@ZEC@}|1WKC
z7_kPR9BLqU{ENpe2!Ar=Qm8F5mmgD@=Vq8<9Wv5$CoCKj-gV6iSG=XTu$l%F=jxgy
z5K;9uc%!fAE_h={Htz9pm#W3h-@0Mh$R$|zDHicdp;8L%UX|zmspibG<zZ8M2ou6I
z3rE`Gz}~dA+~jh4!ogMYMIdE_l{W4y@z3xdz%XZsiZ8vAFi*?)jRnkm5&Si6V2^sJ
zJIxNmf>!)|{oO}%{fU><IS<)D64}j$sYk}Z*-I5*naw6v7v-0OP$&oGUZXC&9c5q-
z<$$#*5GBWQaBmgNed+nUayiafMlolaNv-=BCDS;FaRl2S7DBM{ozo6UQp<F<Ec&C0
zldo?72k%ingVdYtkR__Pr86&4l(0}vABRwgw+cs)j`YuOAuDxG4CWuSx(6Myp~gy1
z!!!F+-U{`shN3i+T=lHh2i{EHAus^iedJ<tZXfA^0Oi4g2J@9a&44V~fs<ZnrS$k$
z9j~6DXJ^W<2h^g60g0Y-=LD4YR@G1m>0Js_4|8Pt*Hhlvo0me59vKNCsnYT1migDn
znMca=h547ly?cq5&;NY+auLe8#XHBrZD#OwfdUT$t;0x1Lw_%Ihuj2URDGIwPZd{L
zuTD%EUQE0R)l{{=i<#LW3^pxiq<d*_j+yEMt@Me7HfkjK#8Q((@XjaK3&NxoKCY@P
z-x?xC?8OK2%EBm5s(a>x;!JZ8b}FeuiCUhs)ZDtS;rVImYK;jO0kRO`Y<*d<e<_)z
zJVQbRE-FNr*v@T3hV7%ND!+(_>$`#z?fuWMgD4sAE?)HSi8WuSYy({8!oitbZVT6`
z)j71dZ%ZfN0&`PG-H@?Wlj}=WP<o)v^dfpKt?MB60Mp%A#M#r+84UZhiw|XPTQ3lY
zz~5__J;WNTQx9fk^ybyetV0i!Kgiq(X*b=JZ>SSt9GM2u(DRNJy(y7dDr~qExJ{(D
zG26NWJ^nY-&bbAq!q+?m|K}hNLtaAE>xrsBdq={m2E2xt43m7ecxXlTS(kgYoT;51
z99tXD-@R7;GH1;xy4dgr!|UXnzJCRE`$gA>t!WBNjo!35$L-6ZwcCppHnodo_Xq@?
z_;LSe_G6|wLOFs^a`&Z~b+l)mlo&Z)yn$_dLeaRafr9fy9qW1c#?*DS4$_W_<c~zX
zr9X6@cA1KO_1MQbk3=cRr~9(+9R25+Jem{l2LnCTaj9|^U;Ua-#%Z2$V=6H)mTqO~
zm~w~qqne^)ViLMWP(&bK3~q~ql$mL^*}I&*GJ)v~4J?a8eqPHqqCj3)YbC#|Z|*;o
zWr&KN<CXqUdBe~@`-h{7F4oAPJg}ldoTgZ%3_13CxYiL%BdWh2F8W|wYA%~pNL#cx
zt!)8Rod_yX)B=<@%o~#9U|<9u297R`>-mA^PDx2^P4X%{LD?@cpOI!oyaCo-MKAkW
z`q4u72Bmze^1QnFZAqGWQCzXwj)@F<@e(?&b7Z)MzCiKDc~?MiBhkBDdpMo;xv<oy
zVVHMb<>OPOx4RO<JaD!^cZ(UnjktA%2anejmi5}7;+Yh!hwD&p)!*p>?7DEf2<q;0
z8<EMq*0|n!5V_%0JG#(%K`c&bGAyLcfU0JmNsp|)ZD{uoygg;0wTVX;irfih5RTL?
z!}!fQ3h;YgXa-V$*d03Ps4hf4qi>Ygyy^kYj9`KPK~mzRv*ExM;g4GOm_qevx#dg+
zdjjem-L6TH4yg!Pa{wX)$Anb^7@>%dG2ovZj7qigZQ;O3lwR=>7w_3Cg`-ij8qr|-
zf>SgX7d{R@)pA>=@&;3R-~XYbw$RqmE1cT+djP82LKgqz;XS|}e|%?|tTliro(&K2
zUrF_F3{SLOOT+f2nf0>s?G=KrxBaSB+(VOfGLWC)K(z`p44?Y`wDs>V<>*pYejzCg
zAC<FfWbW#r6R%$f=Ee1iu@h2}j{on-#9y6y+S&|~%M<!1a`41PyjWd+=%qxeS&d<X
z(AJ5b6u!$4YQ@MTFBZu-?I_ErFn?9j;B!}GaCpW5lUVQ0)$+HE`i!|dQp=kYk20s)
zMlmu~7KNJS52XV}dJ!((TJ#aqig4fcXE_a8N4ez&F3os{IP<QWF^3=&L>Btr&N9no
z8B*gtqytS5l6F^zPEGkT3>SxY)!)OwT@!O>dC~j#ym&{9aq0P8wLvveLA9hMR|%p?
zn5!Y}C{XMD$g_@vIpEEALUlFlWh78e=V96_k$9wbzH)F-7Yh4DT4^3efHA;ViTej$
zdIAgyKoIQnQ$2-c2~k8(U;hi{N&69^R|2e9bKn(Zk7{y6e5PGZ|H<axxjEVAXG4?x
zP14-`yx}P${e$nQNW`c|mxMwu1lD$5@Ik9{Yd-jA0ensJF5}9^P_D254%hXz%464S
z#UshJ8bD%uYGTC$^O_%|lZWOL0O`6JmI*p}kYx+jY|lGFQ=kRt?C6Iov#_j6MSVDp
zP9$L(=T>@!;%I7ze6d87fiTxRb$|x6yIbz=Nvqt`cjh7{2%q^SZWy)oPld_l+g6fC
znrULmpP2HNe?vk4@}piI<W2DQ7h=8$ka6B4Epvq_R8hxNou{XLobEN1iwlrRAK9v$
z<84eicB3wTyC7p6dl56q(zZ6*`EuLJK>f5IDn{kkN#>8IucPxmGj!R@!9yzDl)e*w
z-o$Z@veWQnVI`AHLo9CDrUobR1EY(YZwO+&+Fqnhxn5WLWR%@0%8A4^XI0_8LNK=e
z!gQ2sfA6nBX8`%#9lyEl`VXJOPdbK~EJ9)*sy8&h#pifDet`d#_BENgC~tJwu<H4l
z_-g3;9~!n{DL<hDY9^0g8;%d|A#C+@JPx1qOt~ZO^=%Q?1?;TzqaH(gSoEQ1S*8wP
zVFdas;anL4%c3Vv;n551xq19wR&F_ZwvYD@pYaRB4E{NKc^)}A%a<foVS<+2H>dv<
zCO2jK(^bHIHC!+3F!&&CNY5lF=liUJPc15KeNYKf@m!)6hG_PnTiQKG{l$HlpnV7Q
z<@YqfdW5@-A4#A&RNVyihaEXz(1W`1hjw;9_~b?E5+7;S41EyL#hoddp)OANpB*H^
zRfR83KTl!eRnK;bTQaE9T^CCofp;e-%I?!snHspnE$-|3+yz4Akmf1z!B~%l<lN#j
z0~p?^biR<Zl@S-35RUH5#3`tSK)z#|7fg3G2v+atUU-W~9d9y?re{;+txw<tyqiMC
zwXkfrEO%reH@hvTI!%4%&E=aIt~#b^GPY0E-<BwcCYCcn(X-yHzr??QDtToDdb9$0
z-f!J1S}F{xkr4Z0o6Fqr&X=;(4LO4#)Q@as8bj36=rxSQ_g#2mq-h~<IV20lZ^!O=
z&9>$VOG*c_L-a%v!1wE-`=R1bLU%&_R2+uqQ`lYf%}@JcRs*WKAf{Hr0UUq>wLJ6x
zw@UV;e{eW)j-zD@H~lx|atf*U$=1<RSNgnQ?u6W9<k-JB30!f7t@=YC8WW~a89yy5
zTm@<jX|qTqxf<?EbwBfyH#F{b5%M&AS5psU?PvcLZYR7lRd%CJ$9p9b;iNeW^^WVE
zzM*KtP#Ua;I#6@wRLQKw3_O!-bP|>*nT_Aj-GEo@#AU-Fb>Gj`$NIoJ5cyTN;cz2K
z2q$ZReJA$0jpmf^MIcJLflqJWjzLG_8n&j9g;&XEj;*jT5%F}>45$icV@I+{wJq8(
zw-bK(dt6*!zK#Y!TxUehQ!%|+a9$D^mUbR`hT&MG#lYA%HMYocPD`{WsnlaT=A!xI
z`=HVSArx6G9Zv+mQ~aYtfCNj&j(dLpIeV1EDNTwGjdip3$vq&wSnO4GFfeJKTq`t_
zY!D<Yb-ua2GqE|_<s(LIwmE71@NcKMVt3~G3VL;_JM}SuRpO5e;$6$a4$r9$75|Dd
zbC}%NjS0-V8&*yUTi^K9Gvm3CA`Tk9$UNS`EuLa~)Nr*}9E{(mf}?{mbhz#`isr-i
zAlr{;EMb?X!4Ax(DAei&^D}-i6hhi)7Re7*7z7?<54Rv_-nHTPJ*iVd!fb29VT1^j
zi|j_*J!Kp1th3Ye^8lo_Z<Pqv4hB|3Id`+`L)hf8g1^LxMbdai+~ibOPB<GE9Rl0O
zI=X*>x7Az3Kjnvuo{drp@{$d>oqzBQ(k5terNpg#OkL93IR?gTaW7Jo1dS}~QZ#$m
zSb7Y7LT(XJR)`~PtArbB@Qpd>YBTM50pCIVgL~m*F2yXMPWR{MuV*@TxTwg-sN7jy
z+-67c=mx_~qc;@gJYiWFM_^tPX3#?;h6A)>ANC2g`YL6%=6)zmVV2|h>gY7Bqw~nH
zp9faQ#UH1S&Bq3x6XDOKVd(O2%hZZ_LZYE3s4+?o=JR`r_x~#md-X<5(vxtGx{YnL
z_57X(B}xoi3$`nDP<vl?Wbmu-@w9@lJqr#o9xU)OQED4cF&ZMf#fBFC4%5<smJ@e`
z_4mjr<>*Bap+Hk%suIo&x&pbsJ~;($OAvZfmyx(diM++j?4r{%g}cspQXZ7W?I2er
z?U0w|IZ@JKfytgU+C;&0ocV;8tV9FY^#|?R?P^^M)Lj_l$gtb>s{5ix$&K}G`{L^0
z2zyTfp9f)6n&YuViPrGj9R?gpV~?Dd_#8tB{~iOQPs3td8!?n#h`SOfNW?haB5}i+
zh@gI(yOOZmrwto=;WElED$Kgs-T`w5&6Bg*@)Z=ap62<i&Lb`#&0ieM1o2C;)6>l*
z9Igbs*aWyRZ~v4Jyqirpyc3)I1l2p)Huq)R8!z=v3A%k*UbOUrB=~c<`F~yBb-*_@
z&+WZ%`AbY5K#ON>JvOzQxw{)Usr}umKkIAKC%zKyWx%%3S@%}><*3Boy65S;*;Kcl
zg1Nz-604cqO!-(3m72rTTCgR~O*h;=zSxRxXJ%;{EArrAcC{P6s0i;M6B%j_RllP5
z*0z8+@orti_q1ART<(6do!i2B<w^bZxrwpbY2fRHcN`LQVee1DN`iX~_n;aYHcI!`
zhmA9S6eLYTuDcXyhaP&I`uNkg4PU9KHnPzkH*VO8cig;j&D2B^Hay!NciCArc&LC1
z#W~aG&eGfm<GHt>zvf_Q?BzwW8;4wv)+fw16pbDQZR;!1eW3QsdxarkXPzxDxe8>H
zeeyrKszN`NfRTd;%)1zt-hUF*>SG50XaE18=w0mp4@FCQmn*tz)y~A&Xc0>t-Q&h?
z=gS<!UiS%IiSzEAHoC^c6DsVSX#<ExmEk4jo|`>f%{h5R;rQ?@T=J7guo8FFTDz7~
zE7Q>U?Q}RuoI}hN4B<FO=ZL%iFzZE=0l66lmYBQgBb*5{J3`!<D9nomJVL5+taH#x
z>d)P9qR%T%gcM~6Tb%suQ8^~}s0id(W%PP!c?gR$D~Ktxf1L<&GOTnw%SRx@3KA(o
zpmE&rM>(rBZ)=GkNUxeBGUQJs-odU1-&s3(*3|PR4{3TLFkFL*&&ykOR*6F8IY=GY
zlD#qp{b&Cka{l<cT?pF)5ZAv&_NdHRwWF&f@ZTrWyH<!2ygsE?AK!TxhARrB07I(!
z{T*vAVbzH8yd30z0x4P66$Mjk%D~nyBz^GK)P5-G-MI0a1QZ|%Gp{<R0NKDB8Rx8s
zfG%9;A0-YN|7(S%*QN>|$(PEbyDklUU^|Y-yczNVw*|aV=B3EC8p?%|WApD2B>NDm
z%GS}{#L5Dcj9y<4MPcWgO&*Kak)=VM7_yLW`+;xsfu6QcC^Vl@?wia&yul)#cU?zQ
z+xkB+d~bg}%5FsKt}C5(MgXl!1o3+ZWSpJKq2^NKBc_{jzgvfx5=Xdyk7$*TqzG`7
z<Dxd{hsr1SYMTOGV0ftpIFdWRkY7I9Dy|v$TNIS^!{B#xK#|68=o-n9s|t)-!kdX3
zZUT*(m7C4y8gDiG(I}qKo@o^#=b#Z}A|s@s)EWeF`OtQ)fi)IdW<n|mF4|%E!Zs#-
z(;#^$4eE(wjQ>REMZE;pJ0AGzZ_eWU-xq2+;<i5i*+<c!_GAFJf;}o61oC;)o}f;C
zwq;Lk`NHe7USniy^Fz8*@5zte7*c4Lk{co#=F}235{iZ3PCYRg2A$-V;lJk|DjH4h
zsZCr9{xoebM`M%5#tmz?{N=LH3!!4G(y4*L2Px>LWUi@Cf5AdQfkbk!(QLDJe0I0)
zy70I?=pm!NWM2JwUg^PJXPLdVLALF94vZO*XKNc^uVF#8B~$%4#?F0}PSKmRhU>@#
zw0esU>sEu8x{uxPBMSU5KFDu<V$t~uzNMMmIYd@Jt=@n9O;D9poX4x*<5*Hyke?xz
zch3j)@{3`SgzncDbu~1!IMo--fvSvdB3M?v(-q<a1W$V_u%%Q!oBN~1JT)aXHFndR
zK@*oKfhCU5&Fr`YbtS9yEIFYcr%}K2!w8`*>a#2J3WTT%0<+2z9O7LOi(d)UUpN#g
zwU;1auMGuAg|+={qAnf!Ea|XI*5uL!<3OKSwmzL-aXs`XuK8)z=2H_omHaJQSr?6N
zYDY`%RQigFR1NKDSN+Q@K6a}j))tH|Uw9&%2?&CdS1&tqr`?^9n{qg{hwT)8RX3fx
zy&{jp;1@%O&86s|LC@~b4X0tLnbOH9*RTfwTa~Ze#RE%OkNQBSJ5a@ZCV+7Bjt3g0
zJ;u@cRr_ra5PsT}*N6_=RlPh%TmN~*W&5F@_kQ+1pX5|qgI9Y_6<<W*L&r%5gpz83
zHdDk`(r>_Mk~_)`pIC29UIzJal9#;P9P12sjpvl~IbE29N!omMhtaaba|RgLx7?B&
z3lE+FJHrM62_MnrZr|diyy3jv-Yk>qK9ZB;>)-j%Rx{}VmAc^SdUhnl@PzfICUGJa
zK9!6e$^+>y8hT>VD*hL>XJ~K8S6S_ozRgD;;ed+}ClZdeCQKG6%tQkj1={cdg5^Z*
zmKb$*s325%#52a8LNHGbKM-an0Yu(3u%X{GFO?>XqJUg<L=C+={K4jKK*kCl_loLC
zwxi06lg6tFJeof|L}A)@-%B+qyNT9daYB9PsDI*ddOlnD)0sPI(O4*IrVHAdaTn|m
zaXIn7u%)vxN13^^2iz|e{yJ%w(R>jg5h87}fA<8h>T9n^Zg%sHFEZq2-8XaqiR`jN
zMKqcTQ%}H%#u;VtQIxbTjJ%PZ@<U#v6|d$07)R6`N)X`xk9kB*fBYf;Hq|rJz|p5^
zMo-$@1E@5|2JwW`@|{KO#38d1KIlYT2*+4IRA_IBtls1gwVMYzH~X&X)*|*MS1;$6
z{`6up$Kt$*nd0^4Y0V5UTiK+B!_tt-)ij~EK^W>xU&tRsOS`Z1+-^brrDxX-v!AJF
z8WNcQ9Q}R%`|Hn(Us=<hk#`mOPev%;%>+YZmKVm(_08E^-3hZ^6<p@MCM7iZt^>Sm
z8gu0IW&i~~$kk|9RJ@>#i+#JjbOOP`(o4LDOZ5W)vv<R}lgpZR&im>zn%ip1<5mMd
zbZ)Dv%H->ZJJewF={^?^@`uLXN&;{-7Bsk8Iyzo^5X=(Tc&La@jW;h55fKE`o<H`f
z)R<Q=Og)49!pJxuWi|J1ly-hK^AG=e_?_p&kP<fLz@yBVf#y*$+Nh9kq_R5fkr7NB
zkjWWCD+_wxvp6yBIyhSIJntj9$wU25_DzL<VV_Q5#cmw2Q)qK8*IH=dJ(2VNxOh?H
zliu-ueP{F{G;E(-?+*^7ze;Ev;D0Y>hJjRvf5$h(->lmI1d^!t*uGrA1QGzyqZP`+
zAKnec5NUWT;3@YOreGCse9MIdm+Z}}Mu6g`kes8H2w^pBjK$LB^wTi>a~+zk7H^jP
zr5eGnvecb^QQ}Ou#V2*Qr(m(K00W5Le(4wBX*NaMI%;HwjMEoCF82E*$HvEDc5WYc
zlc&Bzz5Qb)V|``HQol&R;uX_xP5@ym>a(zJK|AGoEpxYJ<9a)jaO)o!-62Dj+x6~~
z${sR6$`1VZUcgeUM=`LF_wqvZbYBi@2Sr}oLl7<Ba>Y*6Zf-ys2~)}knfkCXQgmOc
z2DdU=hfSxdehgKf#6id4JSOS(AA$~YVnt-q(V(F8EY^kKWd5(P6}hjya*m&dB?kV%
z`a||ypxORTxQZc(xJEM$0~zW2FwY@BZ8?;v<c6gK_AfF`fm*_y4jxn=G!ryu0s#**
z6DQK+xrr0zXh$98(RdgDH?(g)nWu`Sy3iMR+&4}g@fOy5>L{$MV)9ZBkab7}Y^X-(
znd2LuU8c!hP}cQ*Y)2oMQHNcqIW{plTXylsE^|mKnX_{|)Zf>KmlVvAnj}s;lWs4x
z?D%D{DosRqdcHlGE*h8rGZDV*e#x+|Zqtl+MTqn#RN>ZZk^-VK`)K?b$oQFOEsUg(
zsw6W4(%EXkC;45Xk~h1dFaa{|++TyfOhU9zl8Vo0gisl3)v1C1qym9R0nqd@HdMmH
z*#QTc95sh>{0kMOGDBWWG&3&QTkBaF_!L~{?A;9>>)+nyhb>?7OTRrK&rqT!mMcF*
z_V@-e(celj5>XRMl+_u!rZ{GsU0tk{cP?>D2D;Bgb(>^An5H!9Db`rHvMyN`FW$$9
zuKNbM#!PCx9T@o)km7T6-ny+>`G%%^utwrWIj6^3R|sRda7U<R90lNy?pZOm5irL_
z<Y+b({7*i6hEz2?;T^L(yTUiJ$hqWT)Xl;YARd_NP2XV}-TdM`=jv5AE1nUO{9h&$
zk|>ZK|2;lzhzF%7BwT)uz*Z8P*!c?4;rHy{H0@0XZ}22OnoAtd)NGAdl!u(~*!)7Q
znl;{)N%nqRVm`_9DJe{k2i5_2;iO`49Y>;(LRHV;F@8lC54zZ_C;XF+<9n*W^m$~e
z*WeUEm(WX3_cRUO%WaOD_p1=CwhDB_DsiS%C017r5m9!eZ!hnfjNG1a^{CIrt(mQL
zq^Bvqs!3cpX13Q4lrFGHtU6(5VYe*%VVdCl11+1%je*bW^mI-njz6;84j)aO-{o<;
zOWps{r~>EH(eoB1m0_Xyjx0ogdO$*IH*WMs2j(3Bf1SFQs7s?;9}4JkFSGg>VhjG?
zt6aMMkoblgV2x@~^_ObGe~OQf-k2$zn6M&E&?hr3vQBYL#x9h`S^Gh$1;Xl(`Su56
z4xO5|Sk)nue~J7uXMKd|vSNdZam=f1zMFPbd6`Khr1!v}(_iGpCf~6k{?tl)CP{qy
ztji6qwG}H1+k4DRnKHM~pl-b|EiyOSO;-6U<$Y7t;Ry|x?km8a7~gJUE2;@$Tx|<8
ze^ljqY9EF`P_k7an++&&w!WY|CYV5AGnZaumP?{irDURa<mxKN5(+EZu)B=xzwp$@
z&(OkH!98U-4!U6B8;GtW-*w5Ogj~Sf4a!r*fV8E&rk<oXPx<>@672eh<WXuaV3dKs
zkxmJFI;AtU?|-E@Jwkj+-ZDRO3vRj-U$NeivFMlV|9r~LwJod9>#KbNbh`b6Uc)Mu
z^#{R%cBoQQpHA9AwuLUv7%VA*1c(O0q<8<5!pQ}lqJC*BBDlCvR8Yxa(;NU|ordb6
z%T}){pdYD^d8pfmQN1>2{w=cM#{^=xw=8GZ?p-FAFg*+Qf<+@SY#UwQX`*8+ac8zn
zu$`z_=c4u~dmMvJjv4%$DmKe4WNLbF46FuY%F&hwt6YY6D)??g6nZzA6T@novk;F}
z!|LODA!jlr*)SySLv4S#cG07%fU5W`S<gwK*YlLmb$hcOhsBSvyY<!5$^c?h*^c7-
z%tzO3YP~}%onkUp-#Ok)J0Z7Emyo?0i#k%56wgUY#1ssU0t(|2D3dQJrJ1-B!mL0}
z)i?{Ol_s1nqokE5BX#ztR7Crg!ry>j28Wvw4^j%FmiKw{=uA>R5!K+6;`jCWnAf+-
z%M9NJpEOoE7q?U5FK9YCa+cZ$-IqV_<*a$8^FNMDfK9Nz?#v&<q_`kaZvmA-(eq^M
z;Zvrlj5W>;ZSjc>keuaqMPZUC&WH=OS@y+FOdwW#bC9T%AeKl$>|wut*`AS`{Qd&z
zO6z{<dY+}o!4Z37v-mo7c>SDy_;A7Jmi)VE`E9QOen4CWtIDWc8~I=^EK7BUaX2{6
zzV>4f<!%U72dSh@OlVmCK?+NL-)im%+I5?GR(q+duEpKhUG=(`uHnk7#&ZpL4p`H^
z&XG+P>vkI&4U8x$i2x%}_n}itS=U@&<Uv{MI+~Ymxr~i|U{Agxh45pV>h&Iie`v?G
zm#s9Cp<WSJ1@B<K)f!GpC<@FiHUh+YQ!8zJ)2aK67SR@r9K5{d_7CF->$IqSjjb%9
zuqB@dx=Y$>Cy58m`ph#np!luF^e{>5e%KrAHEE`SR6@p;K_4E=OcDSynQ&lePjlKS
z<`9me=QfRdlP|(OChvQmB7H{#*J}B5BEFe}c3YcFOUefXX;`nBP}tt*TijK~rtTYq
z1qQO$DkHt?Y+y8DD*y>KT^7fqy6k`J^-GXr3w(U5Iq(b{+btd!3MLLy72RWFdspeI
z3K7m6_<sZc#z;WJr3TS;q!gQ1F)#c+tk|CmO+VTT95+NS^AUy~DGNu`=AMim0fqAk
z_$Ns{(_j;xZ*bhoy09Me;7encLfb6S-hbx4Q)O=jFW{rkhyt@sVU@zLrdB<N*!bFk
zP5o9#q+WKl=U;Qn^O4_E;(hkh@2vs4jYkqzUM;hJmJ^^N6HoGU4|5iUJGffrYofJG
z((n)&;^f_riAtE=Q0ZlSoLYbG-c7bTPNZ9qvYY9#K?j&u%A!?rl{Kp36e}|IO^qO1
z)wPD{({rVVT`Z_8-+SmbboocJv{VkP!ti6;67b%zZ8%-W%d|fy<(zUtFvy}OyBGRs
z%=>v=D5zb3L@utaLR;~>opDJq`8<u}00i7otlpK2oe(B&AIE|K31V>qqL)y$18)B$
zjVPo2GvHago+Z8aoSB9oG5R9^9d<cc#Fa<_`iuiU+C8%n+1SZ+Co}=!qfnQO-n+2Q
zTJ|e6VMYJTQK^*5*8gq=*h9gmS0$eMetS_fkoB<Z`xR?}jvYL;Yc4F$`;x^d%bM#=
z_Dsnkf-VW`5p5n^&3@1OWcc1iH8PZ2UU=)KXVMUhabS``;}oNht#hfj)Kws`?RD{p
zL8dGns;iri_bvCaw7)v-tmrIkVpG~dVP9Tiz@bN4mKaFzSQ!<xP8Cv~Utw?0`TK|-
z`i9nT=E`eLw8?-HSq-n7Xl6OA0v3a|JVNws<Ax=iPj^6uO&YWR{IFS8Ci(Ru?sBpD
z_S@h1^~d8%o+plI##bM!#52`wo)ml~jtLc%s0><=Dq3_UeB_KH3s}a5S>vc=-Lzz_
zoJm`(s9AX(V~+<m$CyJU_6zbJIa4r`%`-4Zdj-GI<F8wdrPCa-{9v;$5;+V**TdhD
z{H5QBGbc*h-vvQSU+@P44_-Gz=A^(zgCP-9Y{g`u!JK6E#w@{b$v=xCGu9xMgt}d&
zJBQ54d5;JF`ufJ!FhJ$zSTyZdek4hyl})$CVKd_UPSBZ#^x4-5q3E&W0l1k4YLE&g
z*r3FmP!?7h-GMABaI=Nl1Yt99F?h<u`e3!oHgWf<A>+g=IQ>~k;{b}^afZ6=2xR0X
z3-Z)AD}<LIQKD|XyCH_E+?6N2QX=zIasZ_)%4Eks)dFKN0jb|xUj5<Qui8=P>u;I<
z5hO)^IU;lE#?1zSn5<%(s;YEX+NM_jk=naGPP_B_mScEBm`TgxNd)j&<V((N8sYnV
zxgn>QC3t03YWFNZgjp(TOiRk+7n^%8qF6}z&v_rDk3(-2zw_J~;O04a##;@Hzg_=b
zg_s6yw2+(~4f>J##OBZ65IPd(aO-+l3!^`Vg4ZWwh8&FRggKO?+787*?MRu!umBy9
zhphxS-3KOh<sih63b!6@f|aN2uoGil;N303n-~2nqHZpDJSz?%0L}O;01g%Eg=hc)
zne&|=A1ZuUkXlXC-^p`b2F(oLrMiGM+;G*m47&O7^I^lMQnmL66Vk+3(dfOL)nN^J
z%x4ANwGQXU^X#tVrF?-;#TQFEy-%7uoK8Ir`C%aFg({swPsoQTB%y0d6W5zm2tkDm
zO8S7gX&{!8{s%l4ZR~=78c|R3Hvi``aFKC)`y-&cVbLvU$4|oIz>m#VCB(HH7sxV*
z;-@Pt7ZAv`XWNOrBiINk2_ZfWtHhQ}jwT393Z*5dc?Y}YDhd!rs-<^nS?=KFYV+%C
zFdsVw0yF)-&TlF3Ep2b1{Metn|0eZi*%lN_&BURXix<-h%CnqZP*NOy{mCW?yXk4f
zo6%+Pg4DzOSwXP>xIrlV`MOo>x7K_bvLa7*bfmAK#{`YO3}~u#aTK%WmJz(f9S``O
ziuoK@Ubz}=2=0An+WV)CZ!MjSpfXPtU((4698`#_A)lxBM6#D-5;&xunj^=3q`?-p
z{&Gj-{(0<NS;A56_ozN~>v2ycGa+Y-sog}67RxRw;&2&>kzC`5HB8UpL64J;EA0H9
zT(9GQ)$bKgeI0miAVQV8+G`vwjyJEkc2j0!qEhg;e8V)h2pe_R_5|pKxxCT?KcE$%
z7Hh1QkBj96TPyzSX}?8@y*Vo2{=SoWF<8+-G#$jp!{a%Y08C@&l_TIWYRy6E+qEL?
zC-Sn$r$z=;l1Pb?7&D6`jd0MPD|5Ns_^(OI6C5-R2h@OM4g_q}3Ca28+%8ZUVbqUh
zy_FE_#EVpLQ0Zkdb=^ZP8~t@GspWNjp#+rQ0m~W$m%o<~{+$L|ysCKniqI6nRt2<O
z4IWth6w0(i<A<|E@^Eb=uP=rbCoflGF6&ffCiWDc4ZxEl#l%Z6fWM2c{5?U8gdYvp
zo8keBV5c$4*C@dumcrUPSVcmzLCRb5yRo$I%;q$LJ4h}Fq^Hihjik*LJn7aSe{<E)
zovzKrXn&rpx_DSY{<2>~ew31flJ7TD#&r>X7dn?8xDm=nTQV(Q$Xe%J+A<X?w7^en
zE|bCjNd!MPEsbemG$A{=rvj5dXmKDgaii0`>eog9two)yrst~d*4Im*>qOOKAgPu+
z&hCY6IaR9m{O>{<@si{1uRYm5LN99EM~MMpa&*x5CP~`h-}?cZm%O%cH&k+8)OSM$
z$L?dhu@hOHiYM?_`f&W&h<>|RCL6Mbf^{<aGbOqH!2a}fxRVf<w*s+$BST383_9#A
z&T9?@9fdpzpYrg}kX~0|d(}jpN8(0s+?tDei96>L`A1KQICJ?A@)VA#&S86$E`F)2
z3^}Avi6XV1n|elL<_u#T1iYst#teNL0`gHTmUYzzq-2?*l#0%2sWx{FQnH)XMrO^L
z83)uO)qldN(KzTW?Sw>>jnA@;Pg9Iz*z6IXzfN;<r}6w=rg_Ld%;We|CxP-wo|R<x
z<oh!hH^B4Yp0E~5s_9IT4wX8*Z5>vIeb7lKSrfb3&YpVRp|rY6T*waC=&!dTp%DaX
zKsuJjWpvHT^q;z!>H7M-_+RypQnh^Nq~|-3svr^Av~79JgegfyeNij2DQR@<K_m+N
z%yo_qMh*)ZbCFhIeRgcge{}^|I6;;)70Y=HXi%HhGsH=Y{r_-}(rl>|E=8KU_3(U1
z0BNXa0{msDmI*TY<P;N7ueFMDFa(xZ$C@TBob*K0#HxQPL`8pmd(FIrET)SmUm%Kl
ziMt`l!N=&Z`dX~bi;-M)-xOCqUFxp=yN+%T>cRjZk*%AjJaO)W4XAT#VXy^Xpz89F
zhpDGpu}z=ux2M#ZG-XL&m?+}NAp46Y*L&Epr`<{xqQC6&eVn&?F#BG5)*0GC+xq%l
zrPkXYIWiHmqvA~d5Kta^z7RF&YC&)8GZ9qg;;bZo1GrK-+c1XpL5eUfj#Wrvg)<G0
z9r9X56D^7chQJy}Lm=HH>9_8ZOE)cP+HP^qs&5&!T_){P!-qeKmr2cAeLT(}g`IEj
zd6bYb|LkRk;l~9KpP0<49>I_LR~xZ7JDJ2Ic?J0O-g5nF+dsy7$Tp|hG)yfdxzs;~
z<sP4{mUP_do4DFecH*)bnC|{3;l|I$5xfF^FEJ-~D(O7@mFyF1=F>m80CbLyHl_R2
zdlPU~Yex6=X8y1{{u@2VGftt!a;pKJqcDsV;<F>PYmH(Nqa0NWvO6(NG~)GqEXBXg
zTP6awQ%1vbbZit&@l)1_`Y4hDgs12`L7AStX5ZFg5a+MEVhh^Pzt&~$RCFnp*Qvum
znBh~va})%|nSZL*slP*f^EembS_HSsAlJzLKsrH8(TbRB!j~j2^3xQqdI31JJVgi*
zyg;FwYnWN+Xz0@{A8AnV+(ILV@z;9je`~QdEV|a<M{lt9!Kr){{Q7+jy%rs=QcGe}
zL825kxc#_=FaP|yJ(hzoyE@B4RV`Pn7B`?BN6yaPU@<qTA=BVkSU^`zwN`IrG?aqk
zmVXy1p;tqHEMFJ3NY57FqSi~E=X5-C5<h)V!ORX)zlGq^e(X6QkJH<VkmiWPc%Zh1
z*euh{!Nna~cf)9~Hvv9PMyP{~EoSq$#wz8?KK@YW!wT!CJ8Z%Xhn7bdjAFTJR)UUi
zM3(nbwWou}_qR%Gd$yx`fXxz~yl@IEa7@yK1)%<Hcg5q|?yA{U)ro7HgF{^V-7}TB
zG{hEC?6jO(j&j4y;M)t>BD)p;OFH$2YoLZm>P_qRwVLeHIv0u`uwtfQ;v2PCs!H$C
zJQYCO8Nu*B7YsuumHA$}Z9Xa9cWb+^m{{siVTFJp&0^;IN6_Yw_xdJE)Ek>Mt;yu7
z`(Y=oSOV{p1wzfQa|}WaeTFL>xm;DL<-wfZmE3hy{&Z{kS+shJEY#162WN#Ys{BS4
zIhx_J8iq=!93IGcbS;Wd15EcVhyhIDBvjP2K(m<Oym(5PgF2Oly1!#h1ORmWzOVp$
zuREfG&PfoPozW1unr$PeW>4(O7HgrGnPv<|sMdF<!azZzzLmU?Fo5z_zzUJ+IWvN6
zfu?WSM{4Rv{Y8ACt!EZE?@A+hDY@U<Mnt)RG{T_d%3x~CmcxxfTWR@%gi$7vo)(wQ
z@$Kvw`OKIEK9{6p+jmf~IWeGV<zG=FJT$XbVfu~Mq~M9lq%A~M$p}Lg3(rhYK}=^w
z96tg-g6P2)nE{PB{Ce?kzak<+wV=x{uh`TwXluK+aR|!=6CRNfIo`U>9jMGRFYqJX
z=AwzFbNhCqX4QS;TXMVF>b0v4R*4hi6f_BOD1f4wJsx5UPJL8f8xtV)vyB}TtPc5G
z`mSt8!g%<7=VS}+`hi*!;O%uV8RDPD+d8s+dLa@P3z~?J=SQhfykyl<v(<<bwITbZ
zrcZYz?E=+dsT<z^_*{PhOke%zZyKz(PE23*RSSZ-e_%Oex%Ca$Br}3}>0s`dr`U%^
zhlZMd?u2Cs_Ba+@W<nE<`r|J8!RQ#*YRkSgb0`4D<=w4_H&enKN7IxcQ^Jm1@5CY%
zh&6nD8{@hA*~|B|KdgU?#I$)zLIAu?wov;`NJ!HUNp6NBF^c?pF~I<39$T7=@89;^
zQvT+xUvgg`HkcA<KWJ_dVN`4%JpNf0tmzqE{M*PDz_*eO)aqLz2-uFucXv9?#`K~H
z><-3L<r;Gv!bu$S1Anfyk(+;C0FNa_z|&&4tWP9?V|}3K>vIrKVx!~GddqAnM)~Z`
zE+Bub@%vg7TWLgK=L(LN|K0aW3UQuGd0v@jQE_17T}2l5Ss{l77?!!3_ZQy7I_g(H
z$JvimWGILhP|?u@!lW=Nzr#GbRq5|TkyI57FVNJCs^&B&2PL38DUCm7Nu{XPM#|w)
zVdiFa{4Qc6uXo0Px${L`Na}kwhdq$g@PG`4Jf}%c@|JVeDQ4EexOJOgt~8a_&bv*n
zhwb;GZa#--jKc<dFw@e2n?&$(sz+jK|Cvh6G#;j*B%(!5=mYuPf+FFrF~rwe5TB*#
z!5(w$T3304>*4j{rPCU5s(+2-wkoi#_>oQOGius!=G|}Kc>&%JIt4d#h6}i<VUmjz
zr4uiC=4`x20b)i8DpMot(Mr#CWhO*``!}$QueMKp70&!3bdTBA+Zb6QYtdv30c!*H
zP=y1!@Y}a?8A7FvE;-R}ulDAz!Obr4Ag&NKPUPI#kvlo#%_{OlBA++-Wdkq?ll~UW
zG5ZF2<U1-ka&@ulyPqranRB#i?#HuntLL!Sr2yl{vkR-MdP%Epx~+~y4p4$SW{m!{
zrYf->gOR`z31Hrbk(GqzdxL8x+%GX!e7VlcZL^}?c#fv|aDNRWtGV85(7fVlU@UpO
zyk!32$@TGa@ER2@Y1(eem1inlCH@aYVjvJ;T3Wt>gGV9uAjQQ-<%7WqrV|s(rZ}p9
z_T>^A=l&#aul2^nY<6%i7RBrOBs;g=z<60%<97z7TEC+NynqrU={J@~$$i=gAT@t$
zWbkRuil`iAuRO{+Y2TwVQLy8Nvtx$CPKj0>c;L!RM>jbsZTB1jCiD0=JiazT$t<aL
z>~KaxVY<rML6|3}>Uj+x^cOmNBbHX+#AcH*ldqmh+yDu+(y~Mpr1_GWSQN*pk?2$r
z#fnGL<JvL>o5kTo)}Z-_W1xEp0-I9D>2=sO&#RR=olWIr=&(eLs(d`Co3V6D5+4ed
zl9DRzJ0}gEGi>SbQC}}Kw6*`5meVnHlDJNkyH=`INzA~<>9jRZ*s##Wh+Y&X_>bPb
zFbc^@io|~Q)P4!)GkP8e6aBo6<uc@jmBakjQ`QP0x{+k;W36>i%T@FpKxC296ED~!
z6VNyq@Oh`4i)Y(Z{r!V+E{=2L1v~tg^CTpnL8%3|lcP$8fKB;y3ii5}`d!_nLv8Ny
zZ@<2J!sFr?agO>PD>(ykG;M=dHpz3`Y;lJ__*6U_DqGJdn`Z5bekw-fYx)Cf+l#RQ
zqJFP=Lm6Q<iFR+j`sgFa=x>Bb*L@=n9B9Y)KTIyM{!o@ZET4Gw+~qC;w9`I;uTW{t
zOgo&lMQ}tJLX<MGsi*m<jP&|;bQAAB+j8iH0W~Vr@Pi&@-<5!F+!uvCG9(C1TQ+Wq
zT9Q8*9@sv9y6g#`WTXQU2&hNAnw}<Hd_x4dxr=AyBXcLi6ci)Let!Fio=Ab6IH2A^
z5iym7HOk{1@zUQ85FPky9hs#P(0jJ|QvlIX<%3^vB%=*SiMU*;(rr&mT?3{^64j$9
z&Yo-YDSkhNO?vo7Ii#I+&eO8ASzE}*w56|8&9*QO-*o7&MxM6@C9GIRLD!gLsqv-6
zjFQtiR9MOkSyW+}&XjOj0{DNW-V|Bn=&~?ZV-aY6&<@X@=zAW$v_0o4_ZlHrgC=mj
zz2_@@<P-B|C3u5ok$PiD(auX6xyc92g7<Bw>5d7aB%@j);w2LCxma|zq>603X_6yF
z`C{rpWl8?Nbm=sxLi~T|`o`eSo}k^>8{4++WRs0;Yh&BCZQHhO+jg>Xelc(U@2&UK
zt=ly-=bWyo`7l$}r>AH7c|>C(#}p)^fq8$MyI_<Z=C{s8F&szLCD+l)k4;tH2e|DM
zRsfsSk)t*&*XN(<5K4A;ntlG2c|LkjDs8Fxw@izb>2p7}bUR8x&L87+KDk=m>2(d+
z2qG{j)JxJ($|3bcVB9rEw*~i3S^G;bd4(^x6Z@kh3eP;V>^nF7{X0FZI32|zgM5_q
zLJ;0P&xU#K4}Pwo0~AyBBfPVHYhSrRu^u77dGdv&BbbD+iJlQZIX3NHdVn%l0@Kh4
zQ}kd^Fu}8$d<IS%c;G!i0y{$^V?oL_#6c9LQ`4lpmV_KH9GjUL<3vthF7GR5C1qmA
zHd=3ObB>AsD`c=TQ`hw&T#Vf;7I$7%?KA2H`+%*8-@Y!ypLRpg@jbA*JYlMzy=wL|
z{_@kAU`CCHgk!EcIUaj#A820X2lbWbtM4{&snY5CbJM1faUR>zRzFQTV)oyLuROzL
z=+Vs-*b`y7^#iu*QEdO4h`3OCz;Z}Bsd`ZhlA~?Puir%ng>rre62Pjf8>TaVoHQG<
zFqV>oe`}Y3$2&+8l}ibjB>$P9{z_9qr>cY?O%e@oH(+BB!7BhbA$`L&6E^-0|4O@a
zpGb{VADc1f?YnO5s(X1|O_&{8g*vbk$zZUNK--pk_;OQtH}Y@13V0gFnJ-{Eh;cfd
zblk&78TEb%(;Lxk&MJ<uQR%QJX5qU7v6u*#<FZ;Qd>J9xD}%{=to^|jW102LGv=4d
zYuu!rf|pRemAy^i_V#e#;SRtK`EJfVa_#fBP0K~^CRSE{W4)wkM;Dc$-JD5c@P+0|
zsUaF&U~@Q7XHnt>%9IYmVLpac>>z?^$V8sSw0ZE4X-Pi`0&XHP5&y>x%@4u|^6&RK
z$+_Rxg->cKLD)dp7}dG*fOG$mb7mC{1p8d#{DVLQMd<e*IVldWOXl_13jLlrp;vL1
zS6N8)9{E0*TsKWKpmkb~kVuvN@f%&i!Ec=+R0wG~z8+zn#ea7pR(YpN@F@^Lil_I)
zyR5(ITP1wfb$3@WqrlNl1vqX2L|&cswL4#0OG?lhQDaPTwT<JB|Fm|w$xeHNNySbK
zFiGyn0_!8AHQpn31n75ver^EYj9oHa8X?L3R{wR=QP@gmKzV~;JrG|q$B6qH$rYC8
z(Q`k3dphptKWG?-K~laSl2<N^h=-{FK=9BdP+es~gR3qHACJwE>1g!FWm|kM9tVu)
z9qdd@;1I*K;*@Lj2IJ<NX^IT(<dg0o2kMTcl&Dzv%?8&?E#G4U;i(HXnWYHeo6aNh
zz!0?-`lYGa&$|)CwtNNfXbynx0QcTV(@B)bq1)<lr{}lP-|;eijnPM<Xw|`pNlvDX
zMgurxZWy-FB?=_GBreh1KjS5NPaS<wd-XQH^JzT0PCj>{K3|OOHfFioF23scpK_23
zS>!7r+9s`DtCN0@g1CtB@z5<R>ao#1t1KdC=#I$7HcPXhg>9$5lpxy@+-FM0bSC4v
z(r{g?{@toBYMazvr==#~c<>FJ-kU9Xu-9TikLgPEAzvN7GQP%WOuLPQ>^Dhoby;{d
z%9>t6R$Ql5*(pv6IKgHBopO^vp;xOvRC&|+1?6P5ML1w|qinL~S!3zYU}=6C1N$|B
zT8CXl?s$%Rci+j`g6cvkJ-F?981BZRPG>%>Sh@O&>-N4)XXm^1NL&>1`u8`F36gKB
zW1S)PlFQU!R<chnYg<;F<=oUu^3kJZ*UcIaOK$t#A$soMd80Jk!|UG_I+nz}zaW!e
zs%xdgllY%^cGdu#y>-f&()1gqHC6#&44cIV<8@YyCp%?qu4Vj~B%pa*e|WFvdOzH2
z-nh|krr1S^O|)Xw-05X}_8}|$MtqIwy(9a15j!6n^k?g~b1hE~EBeX0YQs@bj_tzx
z*F7H#@_k!irpBu@*iJgMV<Y6RkOl4Vnvx+kCE`krgyi+ZkM691AUP_XE%KdAQhpoK
zLgw|ArEv+;9^nh|9Hm-22JN=1#SG(j?KN}h=0=tg`Q@1}{@wM^Mwgc4VpY)JOOup_
z8c`mxeGR4~MY=Y-d?+Z-(m@8%wLRD?oLN|v6UH#l8=_V{eay9LlgAbF6-oYpdfy4z
z@Lh^u*Gi=ZXy|DZSVaLHWJ~O?V9ePWUIFcx2(x%0@u*Xy5vD1pime9Cwdor|?hr^e
zu+b+QLV!uIB}Zz?O`$-h>rSPSqY|$cd9Kh7(o=QUIj4UUL0NHuET3Iu4Q2%Yj<~oc
z-LYp{&V`lYQWA|6Ye^Squ-O4wS(3d*cl7=%ZEGSH+DnSLRaQD4{ZmQk>n!7RYTnTG
z<rT;5ios{uDhnXeMLUxWDl7R=MiFXB@7Nv0hQ%YjN-IA{QsuWz%kNFCO6h?w^e7rd
zG*amK&j=IG7{=Sa+C9CT7hm?T{HuWyy41up&WT)xxp&SKV>jz9lF9TFe^R}Sm1Gk}
zjfM7@MU}YAm%w!3^S9Sm#{W?I54E;jr!7?N+V5-WeH<KN#G`qCh8K$Eirb!jzs|m&
z?=&4YZ_|R@eQzcfiCf^~6iX7&5jz@hQp)^EooQB(bhk=^SQnLWsSjM#g)Hrcm-WO;
zd*a5tGH#J}-(#PHp2|*K%%Hm{)q7t~VY^_vATQN>d&;Qayy>`bX#%b0aeBWBT|g+-
z^ThJ1@uN&^))QLI8PA6vi5-JZj#fAi$g!XnQRV_?=lTwh#-Cj?!iMXS=N_>`l5ihC
z?UOp~oBZDrdD=$<$o0~lPC_%^E<eeUI9SUnv%Go))>8CjCot(z+(OudEiF^wI#j)K
zROwP8Z^WwJJ{;pp*_fI}*Z-ms5Uw53uq!%n#{sJ6gX`}$4TI!k+f}5YuwQMF2T&k$
ziy1zRxAYW$Jxz>DRFhn5aZ`YMGN?PeJ#?_W)%xnNCi$<jHzX%Two*Iv^m{y;UO6W~
zCJ*rqjr0C6Z`YFX??4(nd%N|z4?SJHzRf`TLqnDaqn_hu{xMG6J?Q}myPho1!z?Qi
zw%?8Kvc)l_Ov|ANm=G)(H=9`WzPHgyCNJ_BGLW=T5-@?i7|BUPejPHC^Ch5}^0el1
z@QYxx(gvG=mm;U8pP|AZyb$3Chl@lUOBQp}*9VsPuJ2r3xJcs*>v5JiJG+~#P+;V)
z1wlZhVoNR}{I&OZF(nI`Y%SeAQ-?7$>e;aAGxd)7rPmxo&0%%?>Ne7>%>zrEIL1;w
zGB!5xqONhM@B_QRCEjLAQrJipdL*0=MsRUA6-IuJc-;SS!r|oP6l+81PM|Nn3hb;w
zm?OnTZjAMZliG;|@^6$IqSXhgPq<UPlb1O6usG*%fdeWBdI)j@n*AVYEAm7=s(O<L
zOXg8rmYXCOsCW37>T{m)nZzAc)JL*98c_q<s>jRkEYtp|HPtnjW72Gv=}5YG(m`%f
z170S(ltA3_u=J0tBKBxz6fI=<=rJ{YC=r=LVd27C3%FqHePb$y);$c(7H^Nmr>o48
z)=`Y+Urltxww_VOA<QK1Bp1#c*rK~6{7;#(38*l~5s!-X-p81TSb;e&VG^VXzYi1!
zhLGu*281qW)$f*8$^0(`JYH)i0?Q-k?0BeU*nbt}+ECbefe+q+<!n;lUNjvqt*DsC
zM+Iq^Uvr#EWP$5UjIbKYlbY-FN}4c)O0t9(U9M%^m2u>AdJuo+a=Zle3sNdClskfH
zRd}Y7W4C;w{&LleU{XH4TWz~q5iBk*Z8j%ll-&D79tpNJaKj;#i(?vns*LD1PWeTX
z=d4Wcf(zbsJ<2h1C)Jf;`dT6ZMk>?@cCb59NsiJKr||ojH!ax%n`USXrTmx5d%=jP
zl$__Yom5_lx0QLrDmsg(aC}F-N&U#_g0gBm9$WJM(}`W6?gE?0Cb}3Oo2+eLsniIQ
zkQaXKWFweBRNO1*A95BgBV%2EAx&f~rmP@yTHd#U063Yy<qBrv5o4H$`Z50SD3PHO
z9&@C@5B98IM{1=V9VuouZCJl*&t3*|wbb%zN^iMw2QO|h%R_~4$v16h5CmVA*(qtZ
zZZo;yQzzdM2cE4RrUF8RkQYW@|CmxdlUv3`6&A)~v5wfujLw-+&l_5^iwGfo<c`{&
zG4_!AoRFKU$eFGprZ3wG;y`Krddo%fey72FQX)=^umJX`X{Ri7RIX@tfb4*tQvKi~
zOI*Q>QYB8&7N`u8E`wT8IZ#)#zcV-@b+OW5KQIqdlMUf6RPm4zF(l2wzCc)B4ZVe>
z!_h*=)5K+I$IjSbOab>m)(Na;BeE_F>FU>gp&p4J6^YCC+^bHyT@RRlOh=k~af`3=
zhmcZI(f1#O6MtJ}*jfyq61z2gS5$M=4^~6&i8}q)>DH?PLsV6(2i`bLBX8$V+oj))
zz-(zCYEPr?Z;{s)eUyX0YfY-@7u{TjyQem1h+Y`qeDkkw2d|>v@q8j@tt3DGMBdrD
zqLx8L3(p1{3w!8W<ljHym*AAgrl3qIHsfgjkr?<7)e<Ue_|}K4bIPl}Zvkx`6<rmX
z9jKbE$up9@3G_TTTs2&cdF_flx{;7EE1oH=kGXn8)Vc>HUg}XbfnhAh@?a0LjKEBL
zBJZcUoZBVCsB$7qr5((3^%e*H^RX*9^C4C62~N;%OyiN73g~knQzt)`wA7hzaK7*w
zE(YCyGnS?D07ZX|CFU;WcLVowo8PK1U8ud@2v;-{Qh?ZfFE?}TVJVg@RDc59Ac~uA
z<eh@6uF9VFs~I1+9)Co;S*6&p2ou6E=PezaWnf()A%gSiz5Jp3ojloT)#n-KX)VUz
zETdT;{MzA0FoKACSy3W@0mO0k25*;A<w3S)x+?_i{S4eGm=|Bj={FM`!G87%?#}oD
z5wkoM-Se^u|2zZ>9Wo?^`ZDc@uv8bhM9w$3dg)>GKo-j5@p0CblTB}cfk3c<UYhI4
zu6>qkN4hl^!OfD7_)fW2m@~-DkZtUtJ?dfht7zOL&tVJh&U))5kezH^v)+i~qfXbX
z-vatx_Pq8^y6+0FQAojyC&H$8(g4U!RKtb_`$|DqqsDJv0APmJg4R)y?Z(K9B|xK$
zUf{xN`gSUm3ZuBqfuS(`bcAQSy=yUTlIJr*>0WzB<7XmIZ87w--S~QAOy&w3@I3}0
z7lp52&+~`gXG!wBN#Pbq1iv8S-?%((QTGtuXY^F?ei)mB)T(#&U>RyEW~?-M>dhC2
zHe{}@qWPj9tRw5IP+u&mPHRW!Z60K5DhqJGG|zBnWcdzWI7LK%4_9FaBR9TZNZ15|
z8P?`gFm&~KYgk<UB+mIRZY?GIG-nUo$-s?r+<tWW-(xyj;ZkqMu5&`+!f!0n`q>&m
zj74>rLcaQGG9*~28V4M1j6ON5y7PWD1m#oAZ)ca`d4A|TG)G4>RaZ}(+OG_C7XJF|
z8HSsIWXjo+b6Sf1tx(j(RRTB!4<3b@58Gl&dJPb&sI)iJmS9d01dWj=MTZe<1w{xZ
zNsGsM8xbv$70}tNAH$_V0Qc&*V?>2{spPAlfUq=GGWG3dm}3ny!S4R8=A0~b^8-pq
z=h4~OGlCK(66mZM&5mtxP3rks>1K^Tc~+S)<w(}p0MSZ~BMqU|z!Ea3DLK?&WONF$
zp+O(1a9OUeNo?Alk6*=9aQwW>gcSKGu2v9=-uvQU`HeOjo5=RcuA3Md-qV3j&Y5{y
zaf$nnt$bit31f>I0w;MP%Zud1paBLPovq0gE0ZRG-<R^z*<A`5t#MScIYcn8m?PK(
zyTT%CnN203bVF0~-sr<`ad91APx9n36Z6{Ehf!zwOW|L|w3K3vU818mA^zQ&Q^=b<
z;GloF$8IfD&&*^v!Opj!|13&0lRBPnZW&u|<6kYa^6NNG_V@PL?}ExF1rBDiUOJ}+
zYl?ZGDcwGn2NYwr7&6+Xi?<^J_@_vY=bIbpZ**T+$4A6@m5U&-<JN4U2F~7T0ZRo9
zwde_K8R-=j!-g8{#MkNsLWU_@$H(Ozf3rTE>#OVQU!I`z{<^ulH-V|WB_pwVX5{Pg
zD0K5lqF(V&E+48DRN*VEq`8$&XL!j+S2fgX<&_a#NL}twvm8^wfubjz$78srD?YT$
zuuj_i>=p!t;_}kl<4$B03J#3Rmm(Tgscj)f2rp{hl@nYm^3n;M_m?)fkt|sTW)FVI
z&VDz4x&A4zI_q**0<xS{Hz8LvU>Q}mx$cxdo@AOjDCL|_6eSK{@0W@)N?ROLB3`LN
z_Hq8RJxZqZ0&F=50$0MPGjbTm2Bm4K2hF_%gX&SKCwdpLKP}bq?e8BeX7HaCtDQ6i
z<A}Id6lDV#KvzYQGcYWdxs{JdKZAqCq3#ub!l=SD;nt21V)p#)YFf)s-mk1CWLWmW
z)UBWylk`vh+h#CJHzfrS{#H1z8FnkTAyS0_#O~q`9a83VeFN?lP3alfsNJbVi&-pW
zBPu8<z%%|^iXGR}K`Fp2_Gz&~3hJ@8`+57nn$vM9=hN|@F%Mot+BrFNGi=ct+{^m{
z{O<1-UPAYahdIwbc!&5Sg%2d^WxZT{IDY%Otwu6HEt&Rlf2Vfs&YF0Cqker6-22An
z3S^x=f2Te2eSA56-M~J41Hj+Ze0{lZp2=Td1!(U*oxi_}-oDyDb}`a+F)-Y|7|{K`
zF<>8RzrJ37bbdFfw|phA@(6Z9UhpVM5Qt0WY=Xd?P+_DL4%xIsTcFA&KXDRkdQ=My
zF6IiI_vPdB#)~=28G`Ny<Tu6RE$ojimwr;wp-{oG@gZ2Dd_gLWVC@7MTfr-VSBYI$
zxu^JN(ue{()=zY1kwu2$*+21ApXmOx^jk#qNSTbIcK+v~4kG)g5Gtrm0VPPFhs;E3
z%x6$5=bz$w&lVRQE=_1K=Qbi&7n*vAnVPF?;vGq{7Ja#6EF6ce&p&6$=}ddZm+8S)
zFj%eZ(tM9hMXtY*ceqh1qzaysW<`Ciw02`!imlJ4t;@W<H(`rvUtR6=C;-pJ4TVP9
zpMo*>h7J1RS`Rh+M+>&9E{~uKC>|`<hkU>F_<7X=y=vr5Y95BvKu;=KY7mys1H<!Z
zu!JQrIx?iuhl=LMJA~QPB7?DbzcDIvBpj{#SmH^POI?ZheCj5V8uF@hF$HPn+$`Vb
zeLg7f{(dxlJrq`(d4XfUfK`C&iG4^D_?G;Yp}9qwJZGs_Rt(JzqYxyJ1QXGl5x&K_
z4%+gQb=a>O<6RUa%M(p_61<TlhAW4iMDHKI6$c^bW&WH!Yt*oVZMTC-HzoF%h{;Wt
z9@3G3BNOfG__K9iN?T^);6Vt`h^3Gfv*;27CTO@|F<yJ3irR{UQu5#10~1mBkV74A
zE-%H3$R(X*e`bF}(rO2P8-K|tdne+bpYX|%E#58s%<Cok%p;iYrh7g_;ir3wc()gz
zM`YI!DD88FQi5wpF`DHhH09l7D?KyRTx_k62CxOpFqFh)=CdrVzBRR_3A7RewAY?J
zh8^iJ(@`GU5B}b7Jy@5qU9_J!mSY^BzTIX#MO-M)PWQnM{kq|S@-O0<73<&5%`uL>
z%Iln^6G-Xf3EbBy%?O42$=$--Al#gPR<tvn|EfCeNt4W-=5+<BHH@kyp$=duz9$zm
zN(_|a**(!(7b2$95@)#Lrcxc#^3Lx`;!HfJ5r>B1OgDK<_K78-IG3mrr_hi?q~QIA
zlrEJu%mI6hycyX#lbNcdM|yFky5G)xLGgO#dCLa+J?R>T!{}WZ#N~3AR;MIiKpKg`
zDF5JuPf6b^c=80yU9Ula8JM@%oo5D%_2!NJj1?HTe#yz2{B28)k|f<rWz#{~KrPYU
zAuPHKDoMClYh?GEWGS-n08bs&`F9=Aq?DL>R%TT0(|946ku7hSezA_FxT^}D;nj3u
z_A-@N+qPOJREF8NFEQWFF7q<M9fevmHeCOo>_H4I<TeJ-K%_y&VFN@K#<p>yE@X%u
zdrDCF;QbpYxDx$>l~sI~q6!|d%*|2miX_G@40U7x*R2T?rluO%XZ%-r$~z0=S};=-
ziDwYDA5ov-zN|ln6Fa_OvymT3zYsY#QX-N#MXQ_UpsZ=z@*?8`9-K=o>y}Wb;I}Lg
zt4HSNmq{(tam26YAqYeI^b_O9SmsaAU*(Fo8tnmTpu{Yf%Ljjl*vElkfNs)22Y@h1
zK0EDl>4cfoB0!0sZ_<53Uy53VN_0^sDYfQqSZJD2=D|f00)XzBBh(SX_jy=D7|;*5
zvPvjRZxbI_dHJzR65{}IBA0CykRp}Hk2xL3EZtHrQA@Vi;k0;FvNcxW)zU9j{v{Qn
z5}C+yB~6NR>+N11cY7+9TG*?XT0K`eCoLgzHCH=JdMFas|EzTD%cjvpx8D`~7V@j^
zz#T>~0{4fOnvcg`{<<mC8F>LC!xJS8=+p5siAPDHi&C)={&gRN!-xPN=YtkA?nSG}
z09P{9Gp{r-Yw)$PGB=d_7>y?Eq#rxyy!}x@ebf#HIlKPcro6P!YAR@VRH9Z~>@D}a
zqQKe6_=){^?PiRmG^_oh&(YmEhdUV0JM2{0^4CSx8wfh^sC!_{aKNj|&RnYm`)7-D
z7>(gJID~UJw<ad70V07=&I7uy{P5?FJ8o@zNmrf7Ok*heSF2!rHK9wIC6Q8jX3k$h
zc4C_<M<g0Ji7xj(6sFEVH1t?1Gb?B+fiRvbGBgN7ktO^j95?M;6xV`9bwD{-XSRnn
zL)pmVw-aZl&hNr$YgRR^z>yQb`@puvA_f|g!Z@5EQ{gW7Y{c9VVTpb>)*+-n`tyNm
zf^&QP%(UUFaQe}uFKQ+WephA2#*Q&Wu5VQ@P7!<knV1MJk>5BGkn>&)<n*-{X<p$D
zRkK9396@>q@VSeZ5Aht*j@PKJ`KGsBYWn{c;qqa0O{^H>>ahJc2=a2;^-DyWiDIy*
z5jGog2xi|N0(KAY4p;dusju(XqFv->aL#~Z!%5C$%Ha@H8Jzm1?*>%#JC~5PB&x<5
zgmF<Iis2{4(iD0&xY^di^$x-SZnu;^^L#7}@P(dapgwL;Cz866UUCpfP$=teqf2$1
zm;CqvySW3sy?)>Pu5^Z8$I?Z>4H;6t7H$>7yg1y~<e$6LHKGD&r?(Tdd=e0vLcX>6
zRP3z0HNOY%)P+Syq>4fKg+qun^ATD4x%tq87`8UO_5=@qFtX~t`+&a<HfB8@lq4d*
zo!nIgK!)45uvh`n0MvHSb%nEYAb?pR%EiFo>*P)fC?xieWlZ+29`=S-PpSz0LAE^D
z?qX<%-Jlxd(sbAW*I3kvW5=T`o}|n#F>6>UkL8)DW=|cyY|)s+v+r(Xr>d#8k!{Oj
zW3O#)TKd<#%&8M@QkL$Ba{Zj!cP<Mh*qiRwCf%!VleaD6XON%+*8sXRDs$)4>MF8?
zNXdYJSwPcu?PG84XxVC+DvHY1sl8o2?UzU~<l`hAfyi;pbXF`1rD0}NVi9B!fTJI#
z;T-aIq>x}}dg|@&us~%Me=XX({Lj`2K(lsmD|OlWf<sM%^MUST4RFuGCvZmz`+N*q
z3pe<pxAAX&`^$*rV9B-@*Jl^5UekDI!(cn)X2@pSC5TM{_KjuZf-xgsSeFS8SzDs)
zB|iQ(PnL>~nX+`Su9wN*(YF?N+rk-9lPszH&hh;vZ0`;A_~^!vDw$K%`<<g6nlbc4
z3EZvmjxqleJ3<IU5TIGW;)p5&HsBb4@#K4x8tS0fZ|_aB6-GEnu14k-BuMh+oMwum
z{-JomB<<WAt;MAMHqrbL@eRQrr$V5wBlKrtt8ul5rT<qEi$z?#EUQ&q+HYyeA;sQV
zJM%BLN8Y%KLtWaWP`GrxMRZ{q3>VlFE94j(ZW2fW9v|X5Ri`&>+?n(|Dfr>4VC0@V
zESgMX+HbmhWudi9izBbr(&e1jq?bf<<wJ-S`?{R@-1^!sS?7k)^B<kfT=NM7qAS#H
zpF^1LLXj`<V$=;DVl9fzD6d@Xvart$)`#n7n%p^Ax0h{eo74;P371FtpPpwd`#)|F
z0N`$A>RnF-T=beDCpCuT@#mO5Pr9R=-#x7=+H_l?=astDYFk<iTcPIpx)t(UT9G%<
zFF6W5uAOTPxP6{F4C&{5z|#V>@K9Q}0Js5mU{vEM;KnqDdKpEb!;6@M5t6)u#RqY+
zq@;|OUA}=EsJS-eZy@|Tm=Apdup9dkde7q(GU3EjF%bOJPUHFI1MR-!%UhLB7m%pg
z@WKJ?sBe$)iGB1qb&AVj&gxVvz$E5YY(7SvOFh9e(%79eDfD;|OA~1tKC)FDYth{w
zuPWvS^DBq>vvL`2U~jx!Ta=n$6peBE*-?a3is0+l8w1o@#51ZO!ZXIp7}}g!_0K`c
zub2~p?A3!^e&IDF!oQZ8_&2jLlWI>7{Lr7(L!VszYN8HxUqi3aZqWpBHUq9YLMWwe
z!K8wO4TP)fcV(m_`k2EUl79>$T$T;cHuh@}ATG9J)Fq5{QhTi}#BpmXY#lp5_W%8}
z>pQlIRrOhL?m7W|z6sEnzS0|*LRofRR!K2>TkZVUagCp0ff7nncxAKG+gRNL<8Bet
zdGsaV|Nab{eY(+$QET&v#c4rtUjTy)o^u>{wigWdOg|2)mKhao<W1jq%SU-Nc*_RM
zGmFNb;BSY%avrqD1sdu?82C<T51hQ{O<1}qfsXW2;_n?;41kE6Q6Z5;M_&a*%`#jN
z`V<QV2$YR_22hh5)Z_>KF2X2!#XCs4H9kkDWDij*A<9e)m~Fn!hAVAu_y|gtj!K9K
zn>@zE^PPU04b(RBfj2aK6S$w8MlLZ{8elp{OnJOi=QT%E51~Vmr~=>89BpGmTXokp
z>JkeRHx&};GvC;?+Y?;;8f-`Wd=r0EQBxskuSb+VWff4&d5uR0I<x4~SveIX(Fx#~
zRC)c~&)K1Z`v)6DDn0H<nRld|g>2C&fY2A{<#e$567GzL?8q;RJ~Jm`)AbPT?KOv-
z7W{??w((p1k!@9(AphMUvrUMTS_gW323(^iwU`P8rJ!Nh{TDz6!oE+amG-<G;rPIP
z5{+$D2cxZkY(XP(JZ*dD48XQ=g;{RNGwj<{Wb*lzA+<H5-@SZX*`itLO>gNvC$T5l
z%K-luSHldxsLN4)BqGX>j#c7^5r|(mBzI1NuyL1C4kX=nzsd=bp+N-`UA^ZL^Jucf
zJ6%F=g+<?~Az=`aler1G`B4*SI(t7NE)FpaJgd{F3|%WcXo{2!V&b=Kw(9t;o%W?k
z+nOd2#xcAKX85DE8d6l(FQ>aV?W9XtJes`IzZqO%<|oZUWGB;|;C3R!0Hx{E(u^rN
zpjr`dpLL@?YVioCD;})Gk>;Ty`vkjb7)kDIM$cuVKv0Ql6Lv0wla_=8#*idZl&Ew!
zel=9msON^)ZdDn00bj?T4Ao_hKt4(40*qpu+cp>KOZIi%-;reIz_8LFXemM81SS;Y
zX``25E)?StvW8useYcB*EeE%Y!Ef>|<l~xsGo;J;C<``lZw=a;U}0Iv=v-tl>?};?
z;o;CdNY8hj7A%QPnd>;aLl>P|9Y_cjN(QEX0VH;?1GKqi)ou3R>6Dze^W92waAa6$
zCX{u;5-tyLx}h-GNTm(J@%YKJquuz22_bNhY%udR0$Ib74H)`#Lne>#9!efrV@>sr
zbsE#m?~=Mo%RF1k;4T%d^W6)s=ygi-rBqkSwv+!Uz(1bBuaq0PT&T|-v3Qnq=jDna
z?@}&RH_MHD*fUde(HP|<5F8Sbc=d!{USs!quM{>*TPlu5HFh}3?cSARR3cpbkLbrR
zPwWUr>yLWWty8)4{~fBkm=X*rvZbynk|Am><GxYAMFww-HCU_EZ{^I6nz5xCHi2X=
z-^?+Yr>s@AoVIFsN8BXPbzKyCZ3g7(N7+{Te7R=Lc_*b*?V>e5`QcrlIY~DLuh+NG
zS&eQaSXU7T&Ldi(og)e~L0c+VxtYlkp{xjOKB<)4(6dw4QG9gPf#1*s&Luj&>B=PR
z;T{vOquXfT9*Ra`VxYI6i6fXavCQG^qT|mEg<WG)hglLR8~gC|PI%>f=JZy4xcI1O
zTysR7uEaJgBv|9T(Vc!SNN>t6Fc=X{3uM-4o0S+jyXoMku^R9n-yDI68h5t_L$ax;
zmzbfqz|x8`Krnfrk|X$2N$Z1zGjL1Q-*16OSox2s%ID3U##GidwbIOT7Kbhs>_R_=
zm3Q$mcxj-R-airUW;aqF-ZSm2y$KYsOR1qNWQOnXGT<&YF?GAgT}(IfvN%U^#N3Un
z#EKR}XlNEIO6Ag5DEfi{-W19=v89llGtYc`#jrt4`E?vy%6g`qr$64<HRf@>-+O}!
zv2B69JJY=Gi5{)b6JgIbdq>-v=Wx9Uy)&hKTY5e{_p8VyAFhu7>_X__i@VjDD_iU|
zN&_IGv72s?OS~QNcnV{4Gh3>|Pb@a!#ER9hV(97Wh%J^r!kTZ?V9j@v$iMLykD%sz
z8R+wTQRHVukV;8A+K9gt%62lu4VH0~YE=LR_0j=ZY3vCj9in9z$OAwn$b$oEb)sp=
zKYM4`%^<9O#eD~ZTsy&+HoI)QZo5RgNW0G4-XI1TS7<#ne6$`KUIx`1)N4c^+K-^T
z5j;0A*Qo{E{5kf<%0L~OAG*^TA{T3uRel#dqg5goJi}F$OYZ&}jJgZfP8)+&yh{)0
zPTiRr&-F)am+llDTccI2wMSJKzJVG&qt(u}M}C)2?oR%h8o7Yo!A3WiB2h%a8{{0a
zC+fzn`^jp*t%$K|?shiH){vu>pHwxnR`R12@Ty?{AwO+asalLf?+&T2%x#AYlXhou
zn{Db=n^Y~=v1$^B3rB|wfWrk<`y;MpbIWpo=mE$8Vvz74e~>dWWjb8FphwiQJqd_3
z!ec=ZF|s03TtOnTy&5l;>rz7{zl6`_0)?3dFJ0%#@FS7KFauxby$iB0<fG|kf;~#s
zZO4I!#Jn0_`P839!iH_pKl$8U`^d_a4o!p)j-7HiuWl-g-nBI9k9T1!t|zcBy-1#A
z4@$*EJlzl2*6bucIzu%Co3*y9H*&pAO09~mfH<2sy%Oh8_M7!AALw3h-)tMB4%n}E
zjQi{B<yO^ynvZX)+YNOy)ikTi+*2lyU7DK0`KoT8E8IIf{5E7a9IRDkPH-F?2t3qk
zxQ^>LX?3sK{(h6DnQepD88zElovSwWYw=(=#;S$$6<)=*t^3~=`_^DPJ%qkp{_LV-
zo^<aST{4?h>{LQT4rx?^)%hpa*Crnwq?Z8Ej7$<;J$1w{ob#R(V4xrQuK@u9ntilE
z(L>;U1cC-V`%k^{_e$;mK6holKEA<0!oc)?zM=o?`=7XF8R{G9>l=JLr}+C*9qi`%
z`lfMX`Vo)$ae)C9UqcfAPoL&hDRE^5Ovu{{F>os72=#JvlVc;<@?z*1V-~iw<TM#p
z_z-jCG<}*%Ykf>ULKqki!Qb+0fpap5O=ZD!7HSt`#&Z=ik#=vsTi@n0+P2*A<R6@;
zJC4`sjy-@jwQ!ybybp<wTE?I}jK5Q2dq7H#+X$w6U~obdtoJ@?vzFU~ReIojb1=qp
zAiwP3CWzn__*3)9TPi=zOt)d{VxKdsJTf2*=qA~P$WVp(ml|^Nd%IJgIqbZuyILdN
ztt~mW3K0#9&RLfjKhGvKLLSJvLTl&12WeNGa7#E-pmMWKcG+ZeF&Ay41_{&MP!Zqa
z8n=#d_uHEFNa|o}`q<i_w6InwC!4NiV<yA2+?8CWgx|S<mud9|Urv}t6uwl`0KpeW
z7%gfDA}hET-N<E!jnh#oRJBrEVq{eWO(%3?F6>HfNYyR(`69}Hj~%H_NXEF|-s5Ri
zW|gk5w7^f!JKHYkmAKVF53aop<BYTU-}rDPvQYr*vgu2|)L+pKuBiqY&`XDPELHG3
zo;}bd(!~aGFF#eka#0nRLSs6^kTqZ=-(IhZEqAld)0vsfQ<psVvf6GyNC>|yXKN93
zO`?qzIrNV?UZb`xW7}?!v<?Wor$vj!zHlw*I=)+eL%x^Y*yv!KTy7!=Gm>9>%zggd
z+3YSl_Rwv;)6Bg}Txt!hq0(7p^P(p&zX3}geW0CWt-X12n5oYJ6dqg~N2k9`0Td;^
z@i`|t6h=BiylG9c<In=NOFHB-(`ysfZ~U}J45F@nNxg9)CoX-_AE;}+cy=e?*DBd%
z8Ddy&MJM>A3>!VY&PkSB1*-CrM0Il&K1$c*`J`-`_EBExXI(kt8tf9jMlvC&157Oq
zqnfVaZp?8nNWFLeVDV&Xqy}3R_Gjc0o-SkoUnO|Bj;xpXByI<b*+JU$3+oDmyx-~r
zKjuWdM?Zn&)L8|rl6$xKd4L1d(|_Y_7d}SCyRX)uxOS&#o}(``{H3`fcD&}XG7N8f
zpV;E?{IfbFGja;K6(S)%fYWBt4pw_k(FOi_!gIqpZiDIX$_SRa&$8v0m~wH8%fGi9
zhg+1$2(NSt1OI!V;0pd$Bk_rAX3F+L*R|zPU$5bnFWPdHVEtVd!!3($z*js28Wd=6
zW!1T+eT8e4yZWTd<iuU~RN{0wlg8P?Hm5ryqj%S0C5!EJ0x+Qg$kMQ^Zk{RDFrCp>
zt8w>6LShpA^Cv1SEG+J)hJ`_ehlSnb#-9G56Xk#HK9Efm_Z2U$ZLC#)&mG3u^XZxQ
zW`vIp;T)76#`d;&b%ziVd8I0(KBhjSIG}hSr_{vNaMo54Dv1AbxTF56xVAX^XH!6J
zPkeWa{g`p3NlC>HEsZT3;Zyi-f-fR=u8>$(coA@se<5~e<@5saD0&@zf^#w6aI`l)
zl=+moNcEAFZcB%W8WNQMS1<~~aO9<4+zN&r2UZ_#&7ukvr!S;_aSiQ~{E3#03oTbt
zV!1SA5qTlX{Hy7>De+<Hp+@Vj)vndSmThmWHNKU_24I_e(0B*fZSUckt~0ILCfY&T
zI`424q@>`4stf{1JCP@u&uUNeQV@qB&I?`=`ViwAkr~f7OlfRuRFaaGU%a>7&>5kw
z%ksbWKMaBd%dWp6z1Y)i0#L3?p(|26?i&45#90<*9OAh4^_+Tf7t0@RYqRu59k|+a
z0Z{HsyrTGN6w{ScEJqsmSwO!ifnNOJ=0M2%Ae?~k38AEcSo(<T;4gv1`rP@639@4+
zDNYvwYJk>zgnKzGdpK6mQ9_n8ICkMFO4B7~?CbmMh?|t%>TXY6KT1IiH(09R{DJX-
z^?~96@qL?p`+aa0|MaOL7SnX8DJaKW_kjpwvIeUv_>@pIqv(46dUW;5D!P`4YQy*S
z!b_?rT+j6G{vAWHo8c|X&eH>ZPWBm6)0D=ca69NNZTe7}1B*7jHp?yRYoh0<(nGrY
z6?dL4k{x%X0w|pz8e)w3$dpL-h$q2cWsuW(c$Z+^dx)ISGXYZiMQw#^O1QN6G+)YH
zeXRhV*PUJY-e_n>kN`hweW(%eTyQ-w%K+>C{JkpD9x#;z-2)ovWN9HRd3G>CUU9f&
z;bwWzBfpM3A2ZAhKa)HWGrSG|m^?=!p)mA=AR{83F?1oJWO<4xDA(Z80a%3qC=%kZ
zGU5;@RH{GA1>6daW@=8f9h5SZR}6Uc*_7ExheLmuWV8uSB8~r6uTi!{Y!1=a>m`s%
zBqNeJL2H3XC?_x^u#LAI$vcoYi)Sy`l*?f!$V!?POD%*ssd;FV(=4XQBuyvNrCKLl
zC5sp{F&EK_m?SrkYcuy+Dd!N%r5qlky4PfpV2ralbZ=FOtL~wpil{SHr>Rv@vZK96
zpcr={7fY)pB*_}LE|kJiQ6x7~XHOJ0ls_%8I{RnTrj&~<EB)hoZ1527E{(DH%PKAt
zF;WO^F|zju6BI?Cddly&I*i%?#2N_CzRenB=$?cdx^4i;Exng^4fS(Mt#XlClS-BH
zAL<sG;$+8UvFG*Y*XPwHY3tAxB{ZesS{P+K)of*@3dahp<q!)1=nQCo;0&ZEcf?Fd
zB<~U51Ao_C>$#leUyC8Oij0<w!i>d@95MS?uvpaC9!y0zPSE){>p85b5QxC(x))3j
zTF;;4PqzE0eTsTHC%KV}Ip>|XptmCrFTP^_)*KCnvesoWX^_MLW$Ox->hu<w&|V&&
z8)lLzD;6WwbZ3)9cL2gNLFBKl5J5wW!ut7*w(T}sU^+#q;$i)Ib(5SD-e!R-FOR>=
zn$ar|7fsG99Y#8qimWQ^=b8U9GBP(ZJ2JgwCa2p>)Eax;o813tys<1aW|=BzMOF8k
zG&-;Gw&Yn~%Dv^XJmJ2dIc@>IPAOv4L&puAt+=1?DCp|x!s*88-cq_#$`;SI9Pl{{
ztMx0(R2DV2J9b8_aBOfq{oV83nbu`3TAONn!E~{8@m<7uOmv5P61p|dq|aGhMRiH^
zT)SdiusX?i2zS<YtDKZNIy+ClK)$gZ4YaA%Ug++u2RE1TJz{%K$LL%vx5--vuN)w^
zQLee3D?CT>9*5lLa^MiA2cN?DH04rF2O8HSufaWKx_fqlcMNSSFz&g#40odW{}5Em
z?VeuXTxDKIKAk>ky$d|+eN4abTuMDRzsEq{f@T4q_AT|j^_Ayd);_#{Ox}p!@ZOZ{
zT!D%EPY7D~;nsm~fY$giFydk&c)?6Tra&q|+yx%}?$5KD<2}Q@#MXu<2cHbYxw*Ts
z+>Pzs#VPo89y$^_^Hu(3>XXJui)TZygZ(=uPV9wbCH_{N^$(15h^&A-|Cs=sHz)RT
ztj|~4hr;*#_Z?vr!o|Six5-E5o!;FyPyp?E!+*Xds4~zb5HXNB5XAp@nm|A>|GAqW
zSU{jaz;0RD$}+Y%BA7m_E?FV0@n%JY6ULe<ghft<p#_rXRN<&d;|e_$#P&FzqZj5T
zuG2G|n@0$CK{&4TP}flQf(JV!9GrH79}u#HIqt4mZA56u8t$j*&wzB(onJ`Z7@tB&
z%f1MAmG*z6)|T6T1Ddex5>?G<_Ew~)aBoL}Tg~GW{7%oqBK&Oozm)o5AAq~z+~Rji
zdO>?+Tnk!3bPE^?o<Vg>I_DACMlNI@9Q%-N34l4w$lh4?$h+6wNcPaTt-C-^6fA)^
z<**4LaQna9`+jNKc%j%Ou3+y%Ze@h0gMRVtu~)Ny!MNr0Sk!=dMXF}*f_DiHo=o~-
zm=`YpdST}?`vsyK@#mt^FZFzWYZbUQF7v&4gmWnH>31Je-LJOaH!ZzzZ4%F$0uWmT
zVpqE^5qIw8i}5Hp+1G~kt=$YR5e~SKbcSF&Rg}pp$LW%PZ}?dM1+OvOAE+0jkW5Cm
zEuQn<;eX%j-guQrRz_;qxsLEx-&W75itkINmr19T7GUZ==?V`jYn4hSLuE%wpPU}N
z7A>*QgN9b9DvOvWNK#Pm&`Tejliqvj<x<x#-&}k+$>2_eBbhQ|gj0*iP2iGO%O^&U
z22U3$ypFZsoxFLdsMr$(GuNOXL|BdFzo)nI;`!g%#PirI!s(Dn9mncF2p*&JQncG1
ziArW&J`VM(A3NLC!mFbwiFDX=plj0}H6WWR;;eH>DsMy+PF=6W$Jr=>;k(HHt8c7d
zCa0fp`qXcI_K00x$4r)1D0d*77yOz`&?LEnY63{CK;FQrpc|49Jw^S-9!r+u$Sk`I
zc8GDgt&cg+lKkQS3d4oo$b)x2XPU#eUgTFLP$8y=*dY!Sn=(@0p{jzpz}kT(E^QAD
zQ+)t2BT#Yph^dz|2kHFt0r8jTgSUMuy}<mM6@cYhcXs*wm;CwU#S-^=8**hz?ch{i
zDxk6;Dm`mJb?!FAu{Ic`B<m_ISfIG}#Qo(1mm0<+#xWT;d5=%fYI^y(sx8C6G+_BT
zdcIVDQaKwedB&e@jZ(EUcGQ*27(Mh>PW_wZnyO7Uyg9VHbx8qR6@zzi%+RTBYPm8V
zu+5caleqG8fo14Sm1V-@Wr$M(cybyGn`_;8O6kus_KgF$fC5ETrErtcaY?JbI?3?c
zl80=Y>GWoP!oek9mh2obHx2)|%JB{j+Q)kBtT@J02XDLDl<t)$8;UN~wBx#IG}vy^
zse;@T{q<p3eIUd^20l95=XUICW>|V)|36#4;BTPU?(bv((i7TjUJ>lmqd!Dicqp$u
z*8OKxQ5e>M7SJoouYu|J#+RFg+31_SUeVk;DhJDLr7edL^>KjpIsWKtM>Bi~mYXpo
zZlQYccg!6x+>L0y(u~*pehh;fU0Vp8-r_WpkHtm!jK=q>^TT&2z1!Qiz%|q`<A3{4
zV|8PF95IyabmyY90b}v>%zv^8WXaAQN(ZFTo1``wsAyXC$!m(#Qu)j*|0{!P*A&}5
zjj}agUD?~~?w~xLr6fep&$i($RX~ezxfvSS#zZ3QG!B$9DCWq`i$*qk2CTs)3ofkD
z+-9B|{w0fW84}q|B+~`iX6)1X+2$A<$t(HJ=6ssD?0LV;NjB2h^JUCItVM~K$BGz5
zC?%~?3pvTm95)IM##LEUKyi!gn{esmXyziCkgKID=PH)OEf*>AWjPjPjONWcB2)|G
zo4DEL89S0~3Nz2qJ<{;=J<s_&0&0pP%lzq!n#<BP%`7)kTnhQj;H-sR7wtTfTuNup
z<+-zB3P{e)xI<@(1Iv=(iy)~4B_`&e-eP16Zq8ja1L+G7&3LONC>E(65b;D^=Z2l2
zvt=k8NI2|SvSlG1VRa-!<>O?JRq^{t+)-msn0Ql0?;N^#Yx`*KCA*kw2P*F^y0~LY
z-_L<3?y#Hu-rQs|W-aeFy5Nu9b^e_`x_R&TaU4av8SVse97A_ug|6u*4&v@pw%SJ>
zX6;S8!|o)s9#p&I?4%@G*+*EroO#JE3M2=hyAjJBxR|^ppx+zi2#H|tm3QG?jW4zZ
zww~g-anJ6hZ;kWZ)4pZ99<1|*x}NB`sqI9&p7NcsQQL>0&;QMlGQK{rnYrh^-q)SE
z^V03}-kNo%zZKd^<Ub%runSMIOUIa3)Rp#}-{}fUxl_~~@tg<m;<y@ozwPNFxS55#
z_4h&~m<qkk_rl!n3k4u~krIgH&gprv5(tak)_dU+%u?KjKhRxGQp|rj0p&`+-%)u{
z=Z=&Du)N4~hfM)oUi7)6rvPv-BE2Du#Z@op?eWxmV=pefezgTtFEqY>wM9eL)Y^Q@
z7tNlS+EV%+|C(c*SMr{pP`=hHdr$Vw9?vU1cg{^o@44Q;6VDgjp4gjG2Fgi<+rS4-
zzHx-R!3R`<f!KS)4^Dx8xqGD#R)N9U1=|l?fmw#z$PZ+`X@*6X2inj6+BwG$>fVuB
z0L}+_@31X^=Yzg?^acR=LF7F~a5wq^J$)#5?*0{}n}`3ErF+2eBJ~v^yH|V8{T0%C
zeDmV@<?nlx`(pYP?0ft@|M-FPJ>z?u`@r-$?t8cUf$&X3N-MNGH(79o_AeR2LnTB)
zFo0PbyI3NlSVYq}C@M*TS#xAe9ZO7a+@XPYn~Y{Gq9LJ5I(LHF30p_jbOFl=Kbx#_
z!lI!gDwWMhMnh^P#o0(lLvAJYmyz(gID3kWky1Kgri3^g6B#X1xMb!KDLwSMc1s+c
zG0nPWOT0B1RZ7tiy4V7<0=Yz|bP5%-N=u^o0@3CqJhP6;nDluPo^i@0X`^w>C2i+e
zO+)4-y{EY7iDasil<J8@?1QB2$t1{mqR+V0Z}f>A#Ux-V2@Uk|zZ%3+#_5-2o|F6y
zwU-p{lfVvnxTKJiL=I)R<e`&D4n??RVv|e`Rk#$Rlh6(YS)>$`R1Otc<fW5X4kcM+
zrjuL_HCYs=li&{d+N2nhWDe!p<f)Tr4#nDJYLjdZ)!G!RlfN7aw@5iA=^QGz$Xh4z
z97?yyTqpS)YPTq!CqeJ?cu5f^N$$&dMFr+!9};=#<;D{4lX&T7#~Samco`VRBJUHr
zXlutR?=!mSZO1b2Q@ZGG#yanFx)=z?Lhs|fXmiI(@6)~L^~O@~(LQPQCOIBpKdE;o
zT<^O-2n8nO?i)Xddndk+I9_$W;slNvUbVjBebJ~?3z9742bBfEm+h9zGL#!F6sj$P
zu**fu=PjbJD@V$Y&we%Yc1zeU)<59m2~&Aw^XIiL!92uhDpVJ#mbjjKH%DlyRF|l#
z9WG;A$TX*DYE>7ms9cmkE<;!dH^*<NTNkjape&hC7r_}<)J2q)Stv1~sX;flZwOhJ
zw<u~=S}(Owm(wOIwm2wj$|xr)cV<+Tl$BfXu*o==T`j{~bY$e4RLC|LW)!=?qyXCG
zDUBdmP?)`o<B%@Pg&WN^TGZ#1n)AzD$*!m~Le9-t2l`qdO0{X*BxO0TWJ3}l7(uK+
zNBp7nk<Eau^T?q!>I~Bb*BT3cs{U*MAFNB26Cq1$nod%6L`dOwg+6E6`K%Kf?5lps
zoYI&n2CH#IyO&%^rS`q@B6c+<qE<B{BMcvcHQ|GE5L%NabB#v!3i?$Hx~OS~!*Gqm
zcUR9PG9L@`V)i6@BWOD?UF(3gt)m>%q4kI&Ci1}@DPD3)#`fC7&z&FHSre+&RUKQF
ztJEs>Y@;QKeQqjE(_-C*iz4pK@jux{`_d*2HJQ^o&7^G3&KQ#EH8oZ}sAty`$_KxU
zwtw&yYL=4lw8fI;!6#u%6mF|zf1N0>!7j_@dr&F!1Y4ADQHM^M6||S0$V$nn<@X8{
zLT7qFv}<{Mh+Jc79x;7cQ%YGk=9ryjXHv00ue3lo^f+cU_=W4k8yq{S`)6s%j!z?;
zoFL$z#$m;PH?8sRo_>SmlF-3=M66r&FeKj0ID@}w9}DJYuB7ni>I&-(FAM6{o<4f)
zwC-fh($c~uMFa<TYa~xUgTeTW#@{G+G|y?(;12z5?`2S;2Z!a3IvEhuGs#<J<FA_2
zqC9EtiT1-=8Vte1adwnv*c4vcetMWMcA=GD%E4$p70EsI2hPq1)(e5~x+wH$82*y<
zIC;wF!I4ZR2AAAW94{?*v{CnOU~qxt=d53G(77~Ds;iRXtkSlg7hI=1s&k-&h7XV=
z5g(O;N!Tv}jdTW!v_xX1f3ivH`}_Q4rmLwH)NQ6;-U)`p7bgpr(teVWvfM#`-YX+4
z$ddL4N@A;IJioO(klzyYZ)MLa4W@riC>7zuqhn#fxc<UCQ;FBAP{m1JbuD=3fout?
z-NV)1*=9eM2Tz~Harr(WQ0cXc9|n4tTz@9&PxGF1;XD>rno03CGxO*BiqSr<(?GGg
znc`Kmb9<LaLZ5dHK6hh&<Eo>1lLB&+;zGD8-@PH$DdJmDZrMWV5jy+d%|ZDxeNM@0
z>G`91kMpr|jDF}Y!x8aulgm=}4!o01XssiUS9YWu?$N~0jB+{0N&U>|(9?rG@nuV&
z)78~~#W#<|iP+=?naNOG09$szJWa!TEq$YJ>YLq)dG6d+U7p9fI@_7ZwSY0(Y)!V+
zfX8{z<qaFXz6dGv@EC;4E~}6B>g6YRJ%eqvrqN+w`dw;3;=leCGF&e`3hL-II8)AZ
zmM_k^1T)mAB88vP$H)j01+wYKq>q~hMDCBy2;Srm&sfcb3=$v!@&)xmvKQ<3|9Sa!
zi?hekXY02M)C=qb`T_Cs|0rZ`Gxs?9Z3A|}dZB$#KFDAGABFYpnkdcD#=DyHim#~E
z3*vAN*a=F`Wi?4jSt~-1C{pAvJ;g9Ke`g$iGDInuhQsw?en>Bcp%lR?=qieSoZ#%D
z(%3ChgIQo(xQm!l1W)*v#k$m|OTtizDlPTax||Si*eAL*8PGyq-nAe2#X{X2`9;cm
zlwbk$xt|rdV86E}=_RVgPm^<Ve_CM@P*=MR?Rq1urK;^wZ$)wi`6~WD0Z;?4{9woN
z?0WS9U&yNu@}7X4^!UAed-`5|V9x<>4C@YflhAYEPa5{po*l0~Knv>@*3+<W!M+8%
zX<!KT386kAFF&YH$kX%qy!yZ|8-GM&At3&!#_&fqhCjfu7|_$lAJ;hc#qozXjz7Hd
zFc9|Quos8DIPAq?FP><0|G4MN{l(#xhm#(Z7VDX&Qg7uQ-yJs2E0l{E<rYD2Ah$&8
z4OAY2c|FYLdU{J*F`t7$&C#7N^t&AHpvqAlfTKFp9C}pcaP_JjuAb&_hnmAxbUCUc
zHy$I3uMGMgsbCb?28;pYzyvS}3;``*3W%>6QGCtN_m&yMSB+Q%dI{{;_g)3zkM6Gu
z!jA;}NPs^9e=O|B2MkL;b4*L0vx*#o3_tL}!?3`Ug8dZYPk4wtSXj5fj}I#bKCBp)
zhgP%ie}0qxf7-RQ*w<M1rXDhJ;h6f}Hx$Oj62?UlMni<r5MeYz7#B+z7fl$A5)OHG
z+JH%*z6ip2_=fSY4dWpi#=|a*hhi8H!LS+E&g&N)cj&$fHM+yzLSccPbfazcUZq^^
zLCJmp8<gDbBWv$iuC7mZ&G}?ke?(2La*ptf={G~4w(Cce&V5&AAUn!e26v(vA8xAS
z5w))XRdLWoR~qeq2X$wf^miP0sJ{$Vhdy$><c6z<sC!kbb_8~>%I-PPPSm~-RmB#R
z%eSkxII>m%y?3w*zU5l*W9sTrKUMQ`OkHK%sc}QfGjy^R*SH~tawo^uS4+>;mu%?f
zn$zyN3+V?~<*#7u|EwoBk$9XvkULJpKp2ESn@0lX5hXk<3W#`vBLxD80x7W|R7eO4
zI*=$yMI^AOI0Ppk$B4G^5x5ew@h?~3{<Ax?vpa?aM?gfF1datlzsR-dbE&;88bCG4
zStR9xBQ~i|6mr>~x#Nms*6dKoBR8D0Pi)E*Z#+qB$yg}Ubm_>`WW|&j9=PU=y)tK(
zHV1O7t4N_#I92KntJF-Xa<57qc1zcsQn%Sq<;awIe)!@=>4&C7YTV_dbm{d=GFFl?
zG;hBLao;JmPLMg>Bz`)2#?&}8j*K-s-!rCqS>Y@AYaAK>j6>tQx_G|o*Mbi}$CG*4
QGqXAKIiCL_%0K=90Nq5|6#xJL

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Regular.woff2 b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-Regular.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..d0d7ded90791221663847f6258089ab9bb76c23a
GIT binary patch
literal 35536
zcmV)AK*YayPew8T0RR910E*B65dZ)H0lLTl0E%q@0ssI200000000000000000000
z0000QfdU(f3>=d>24Db`W(YtDf^!iN3Wdceg!xSYHUcCAnotXkDgXo^1(9tBxDO12
zG+P&qkphLs?jAx-S-K4YtY*4Qh5qW)Fb`|s*PA7S(!zC?3bsuv-rY)+4@7Eb|NsC0
z|BhrKhyGu<O9D`e($cNlW)yvlijJ0YQmkq<RTW$2j=Aftsj@Yuo;Hkabg2@!O;S7g
znRhMGy2Z!UcTTcwo3%9|Q`p1B-5+4<z;oV^%BSUJ&3sxV$bcOaPQ+sYb&u|L+U4?E
z{{8(L(N|gO%Id|(y%#-;n3CK9`oI(K%qBN{Guq8Z*nYx)3eDiRtI<)u<fbS$+mINP
z@j*<O6r2M#_(%L^SuXo$@-*=c1Z6bJtv(X6m^>cj>8^<4Q*dBtT?t9@k#dwZ?k~T7
zE0eW!J#GGPkR-US_(ZZ8@Nv7lz80i<tAQ*bwgU9ef)vnfHs4+9PDKtb1;25RS;*14
z?0NktB;&fcTfpaL+2=S8_Ixy!Aqk0DhJ;s0mSq)Cj?Jl&tehmh_2v2J$35TL=brbz
zI1nNe#Rl~WNs(!sgfs~ZNlEv3{#x^}YJbGlZ}SPro2(PHbr1rv1H9Y%0l08Y5Sb#1
zECmE)zxNpP0S`nRs5n<?!<MCKHn(M6)5g~7%DSr?t$E$Ge`EdYR^6z;g4=C-)xZNu
zsQ=UHe^)bt(TspB%i|FlNRFl0QGo0R^oc6m<?f|hpe)tET~coMUbR4ZOWmFmS6h9m
zmnp=5EZx5U@6IMd22iFcglxrPqo@{zunY>Ei$n=z%6x16_7<^tB$p%>@dU|UClng>
zXQ=npSH#ZOWM=jwiAZRev}uU22x(9XhXSL<>K?gt4p-Gr0L<IwTc+>6!zI(xccV@*
z`(lBlkTxVNQrq#i{1*T~CR~zeA2@w*-j+%-Zic{*`?Y;Ld&mh4G3(}`&@N+=?1hh5
z3d3w=m{l8^G?MW>X8zl~SJS2mP(ZhU;<Eo<lb>?#dt)BaNn;gDp@=T<SLJqH-&X6h
z6cWOlCd1I;U(Bw1T%v4Gzq%|ry*LsQEFmXaIE==u5&ONwsE!{aAG~i9eRoztX;GvS
z5<2(^9O4P(wg(3BRPmvJoF$m^rsfuMp7@FW#1%=I3?2K-_4b}*2fC%2EZ`LY@5Hj@
zvF4Az$EkH{T{$h0bM;)q3&<zNT(fW}009c~qBHAw2o@L%CCJNb&)#`~oqQ}BWqBxe
z=hQW(XOqp@dKY83+XfA|y&m0;_iK1R;Sr`z1sA+vaBd;b<Z8QWNC#piLU_KL|G8yZ
zk0?qsj4>T-gmR4l1W%BlZ<oSt&sbFK+1Bpw|2OU4T|XzRg0~(KC2m&;FB6ubEuG<W
z9X|fHzfG{#7tx3T5v$Ap?@z7vKfQoq#>X0T1_)aKs%D|xn%=+vNgv+-maT8g0xMSR
zB*(HQW5o){j-1t(oU9}}bL==nz?L}~3d1s2L{Jm9yIr?M$5d|7Ra9@$t^fa~R;IY`
z;iG!i<A7MK<atWg1PL0=rc5;B(@8?`<Dve5zaIXI&Fww2T!?FY4-aJx3$MUNqd<-|
zv3!syawb%$qewnf5M&cB4}HrTe(IB(I5>YnsCAClTibi7$W07NoWy~!IQ-tV%BDK`
z!&%CqTzSVr-fOF<1~N6Z#C^b|2&@INYt;GwKdo7km%Ft)suK6j2!q98WSIQqFDJ>C
z(rA|BC&awWnydqd??Q9n`TmD4o6BLK8-a2^-5N@J^p#@!(2cc$=i^k(`u7NuhjpKe
zO9fo(yccV=Y!ga1y~XO@FHHsmW2VFx#b&XVmFh-t40vHgM+tZGn*Xlf_SO6MR3-a#
zwq`IFF+(AR6k<d~I?~VE{J&*8#Is${Z7<ZR0}a6anD#xk-~Q~i{I(@olc>=S0y4;;
zgCtk)eVyO0$nNX;rMyOR35jt53pbS+OQT}R<}AoaS6@|zF~D_80<1-*wJ1ZAhbWTp
zH=hROE6tR@nfjZHf_Y<>NAhOVFwA4ZG@Hk6PRLYqaf+p<$b-`GT!rdK)xtw<P@#s=
z5I~Kk!-hb>YKz4o$DimV9hHYzK|AcDt74t8(BPL-wBI9gW8~-6J1T@(ZLCs=G>q$0
zGL+y@HkEZNN7N_je#T;GM2u+=hE_Uyl<t$8rcI0_8S-qgi~I*1b4JS<H0C8Y+#_)Y
z>^aMC8#eyzr~mSqYxOD(W&iC`TEEH%2unN&l*F{}SP##`6R;MZhNt4ScnKbh8ElNB
zu^E1g=U`RSq%0ZPS!`NSQ8yfGf?0|G?Cva-4hjY~J_)7e)*4@Cb{XS-Q+9TVOg(zd
zCd_9Pro6lgs@a^M440Vl%l5bZe%Evov*akU!#?Lf;*<+4U!$Jxc;u<q_Brw<{RJO=
z_g95FEgk{tT4TV4!SO>&hzJ}>7*a~~?n}~R0D<Z|_s)Q57aHr~S$I4)V`Eh~kBtSJ
z06Y(CVOP`W?3mz{Oci(%o{lG*X6cRrPsO9z+GLhPEElKO31w7V8=);r7#Ap<``unM
z?%6qKPWPiMx>h`fjJ5h20)S953czC1xEgekie+x!@P#A<)da$uY(91HV$3ZPX^oH!
z7F?mTwdMN8>K6h8uTUKT|51%OkpdjzC|v;BVn*~C!5E9o(?YcEbEYA#@Ft$)1>V9n
z-o`t4m+h5=x*i#sLE3Z<9{KAP;KTm!Md3q<sdNULry<go$n^|OG+_I#*H9*hFVrHW
zI{HQm1P}$OY(ZRsCN7rg8W@|}AncthRYueuCvBZ{oqN+749wlS==vbu7)q>3H2sx4
zvJE?SZ@uJB-3X%s*KNJ!E`gXa5l>ZSEo0fnT)bfG%YI)rHP+yvt$+20?@f|+;{*C}
ztT$Qun!mvhy(!WazJ;HAQ>DlJ0`Gd$q(A7NpN9Oxm#zE|$YF;uYsLpZBynn_4ddH`
zY>$PJw+<eS=8Wc!rlS>Z5<D~pdcx5~Ef2MEk7<^*VA_*y5$k~M$+Z@%K%*z$O2j+_
zM)Pl$`HxmY$0j6aVa<t`AW@QJDYDBUr(ALi0@89opm9~joO=JZg@G_gH{7T-iZe@3
z;)~_stSe55uWIzgKpPXyHQ$7HEJUMye$aU%7~{E7LFkNnWNm>~QyZ_zw#qD^*=B+=
z&+d#QdaGV)q>9@)ZLX(U@SC}fOf`0FCsbU{!KB9T{Wbtj0JwZ?7C}lNy+RbdNfOAY
zqhP!JfVPH3ih|k_my$#NM<>}SqJScEdTw8BHFI{{_z4pyO`dJ`)op~Y1r;B=@HY*X
z2d_hK!{|axGE)Hn0Kj-@Ia^M$TOl5T@suRn=q+a@LRv2&0T>I8QkR!1M=J7iW;Hd2
z0S6W#W=O>~ESw<x&tyBK<>Ars$?0%3o=o8noSe-U%hh@lgb}<4`9iT&u2gIF26G3R
zt#+r|>ksx1bS_bZ5I*NbDNThUu|z79E0ij=1^^+g13&%pyN=**qoe}~;EKS(7&2^x
z6qW*1^ZA1eKs8kWNCPT62rfJNFc|Z4S&fU)D(h_x*K!@9rT>?<Gkf+OICSLL(sez3
z14AQYmI<4~<?#hrv_YjL9E*-iNJ?#`^(@(O;Ux&hkVVZUYtX|P{Adn8UcgV5@Y5B1
zxQ355@Nod2gz#ykK=4xVwsyd4r5U~kdVN28A<4#F-PC?*cM4uD#Yz_gx3rSGj(I%v
z*w94Lddmy51>|%b2Mm1@ez`MEv=$Jjc4jPj)5)T|jmpZr>1LZgy~PBxPW@a)9#Nj_
z43-pTXui7sAS3@Oj8c_qnNJNI&t?vc<}kl&BSYADjsSFr;C0WSqKgdb1(k*de;h)6
z`VAO_pezu8ua(z;Au|?`<`Rwlv%%cHmMpfwiI%R1qP?KOHEv_kalBM)Lb@Ee46DIo
zG)Zq_4jfxrxu(#K&|7Y=0b8q4eVYUDFC}dV#cQY`-3pyLJO(Y>EONVqdR{kh+DlQc
zG&cf%wT2_$U;=m?8NTc64*LWEoO0{}I1T<O!blDMI!JkU@PBDQHNH2nI~3T9YwNWt
zurm}Cngi1pSfv*ck#YVqBZCV|@%9R^f(3UT*Vvo7$tvslHsVFvM&`1l%jRY!eH2tQ
zY3R~1WXQyng@uiSizizSJ^`UzB4QGi{CG*QuCYLI@$d=A5H9ve;UCd4V#SG<AW@QJ
zICv=tQW23<3*sXY<d!sN2E82NIFDR$(>2%KcUPGj&FVFTAU@(CF474xw2(tdW;8KI
z7rj`~cG@FEzR)PN3lq_ygb$gSnsHmC$dPMbKEsak$AAZH7AXq&KBmezD^z5+jC_HQ
zlMQW8eOhxiA2OZMS~hH1v6iMFCHVmUcCqK+lFQ=c5wjDRn)FkZ0p_!TAvUpvV~h>*
zKKE0W%3#6}O1MOh@QA{Q8zDq>gach^4ty>J*J@y$iIiM<OL9EZvXCWgVjo{{yK<|b
zPI$il4N4fIG~x$-PzPh2Zh1Pfz1`mHZ|m*4gLnMS^sg@M`tGLvatn6-{MPoLRmpgk
zWSwk~RoQ-alMP{G*<71;?BPgM7MFhwcRlL)UhEm~hj>a>Q?hQ4a8!zzvZQRFZ0_(*
z(kVK3MJ^&wqt_<>@QE>szp3q6+$@z2>#%rK-1s=+WF1={VlV$-jolyjT>TtCmWVzT
zKsPq8+j`Lr-Ok%{M=q@Wq5pP8w{#!S>CQKFvnUf;hWA#FU{fA5f=GePc+<ZQXW=4J
z(5$CDroI&jr*h``?GdYY!@E`I3EI@k$$?$||M!JtV>!=85f{vh5C5NpPX4$Iemjc%
zem){h^YZ~ZPep37U+`QuvoMGg2f<H&VaABQw52o|-LUJ=rS1g3JXM}up5x?;jC?Zr
zb9Yl^TQfJ#&X4y?n^cHYjKpwN^5@&1gYY3M=KNzdp#V8HP3Zs(A=IG<1E_wTROhe@
zt6)VK1tna8{aenD7S(BbGdjDat!$({UDFf6&OLe47Kbjf7ed_nVqZbq)p0vcY;mC6
z*q&O~=0(!-V*oev#^ao;XglXAHU0Yp3+Z486UkAb)xtXvqNOZ~g@x(`36<AzwQo_J
z>Zkfqf4I(^<Ijta&{-)bD7jD#%a!NEq^Zhu%fo5215b{Zs#usg3-aiXUlie8>00-@
z-rfGDYm0Mrm0(qvpv>1?l<uejv6(8%-~P$V!7@$)yUn|+g1RHA|7KQhRu!gKs;5e`
ztFo|Gebt&*qeZpNt<LKDSJk(+!8Hx6hcti$#wKH+4xtIVWIc*lO1d~I1ZsGyMCv3O
zWSUgk6k5*H=?c=BSUzKgP?@V1t3(<JBm<>Dp5K7qYQI$KZE7q^wd!EiX0=xB1-gs%
z{_8VbX{f?erhj%J-ILfqXEL;MfBJ-(<*o)<{j-|T77I0{A}zTLml*0Bt^EEyyQa<g
zYWt3zOPB9nIDf;ov^Yt#yeKPEH*MDs<1{bpwjbwpKPoi<qJ>o2ShnMOK7a@l%D9k9
zTejnReh|LaPJ^wjW5PslSyy2}UHhtv>BbqRndZ63fNN3F;^HMG%Su<2uc_KtKh&_P
zakzO)%gCg8eUAL8nki7Xem1;ka<9GLxxh2%Ug=w1xwd*;&HCESO<P-M5S@kE+&Vn`
zx*?!jqAND3CnopQl%DCTL|YF;^hjinMJ3VRLj^TiP>aR2FR5c$UCZlUQP0vkw-Q)*
zJ_8;jF;AK!<I4q-Y~w6bwvZ#@ny~yes4{7EXm)CGY0cM!=quIZ)n8_yJS*13Y}m3B
zD^7lgyJ5aS2#CZ#(`voleBN$<9HwJFVe-CiOEihaU=5BvE>kB3n#N<XC}Eb#V@%JS
zHL!Nh49hCq_~T{+ZbEU30#o<KR262NEDF+|fMD#tNa9G))r8g3=trO&_L)Gb*7ju>
zy`Bl23KU}|NUBRF{c`FWr2WF5UOijt=jqC{?|ALeG&e6UG_!CoJ-VesKBOS6b0cbI
zg6hDOG!iP-Z)yDIh5WB@a?&16We>(~;yA?wd}t^@)RraRH%{V0q6Qn+OWt`uuZq8j
zSICWp%GV&xq_j{BCiZ!ZuYvq7hX($D-p@=(=s^bhu5_U-l=Q8j({Z(Q3MIf~Cu)%?
zKr)#Sq`jH5CP1VN6Yxjy$xwi~Ec~=E&Ic@C8`M|e@W~I3`3P+J2UvFuzn7PQZ}wT!
ze=CB12T3FlVmuNBRTF~4qfm={YD%Gao0BbwB~ldZX^F3A{YNn_ZY7(j%HfdpVj*id
z8VXvcX&?&@BFpaSLE5Ym-!{!J@=zbE>ps-91=Nb^Rw)r!36q!4Ye=#D`5g(BRHuxg
ziar)^Y<*A|;{9!1W(aNEitubVa!b=Mw^<{7-@K_D0l|Su=<UTLSTPNUR^AfvkYbF~
zWi8$cRI!nyB}Pl;MkuHnsi-S9E9Vei4f58;RBC9x-M7b;`?=!R+M4B)G?ZWsl--*!
zWh41|A-Dpzubc8)!Bb7q(h-vKGsn7i|GFvez9JTVMQp(V;L|oY{4efaa?6z}J5uIj
z&%#$C1tv)@fx7~}aUH7)=YkxgEMI{xCwR2*_0RAGk^s@?tTpH`{))PG?&k*|3EmG>
ztULT8U#hwExkTsE8u)WrHp1T!*9ir_{d3tlS`U8zYGaI`i*fcR!6pCd3B9{4w-?st
zftbTo`;zU`K=j4i&pFjzDL1nrb1eSrjLA()%?tAuiQaIgZ}2}!sns*Q$0@EngFHZ<
zC_r192JZ*JU;jV;8j{{EBUSVlu}(45nhD!3#-SwscU9!4wKwE&lNKaqv$5+n+FcUo
z^h#wXRph`F_`EMQ$B&GQyE{=k?VOV~m>Y-n{Pa$`wV$}!GP1WPqNQ!+EI!Kd_04^E
zOlrq?rez#u`e_pg!*!swf$xaCjD?lJ{o8{<@v>~b`lhh1J3T!<etdK=OKIS5G=o+@
z4Yl|aUhI+^Wvem)e<myTx3$g+DZ<T*CS<TP^PQy#ePtZVGXTMEjE`O`pF2N!XCMi{
zh<1z8JvA7VPXztv&oPh;WwoB}@D#C8NMq(8Y9}A5%uufNronr-+*Y$6+GGe9B<hC$
z1VLQ4W=jaHWRf-6?52G}pA;ycd0aJPx6<ZtSMZ5@PqFcWRpTqb$lhC{1~Q+4<b+kL
zI?0<nY5$q@KxxPBxddW4Q|fk)#aH@GG|Ww0q2!7}ucc>C#-~$W>w^43P2-;S`p*A8
zLPf11>m2w4rn&`MYZ@xD!jc5cQD8x7??)Aw4muz<5QNoe)rI)=-^52q@S0dPAI7LK
zX|63G`idv`oo(4UV7cxe3=`EpfX0w^;wnEBFM?OQVD8jsBLF^p97<^`MhtVP13|1o
zcxZ}aPT><i#yyR<i#j+d(fLVJ#A`#0A%l*PSrh7E)t)`cQ5&S#I=I(0mA6Ou!aaQE
zNoxBonkV-lrVU>ThYLk`^6Z#Ra!uGVwM69}rR&9FXtRKa#!rt3plZ*fW9Sf?f{Ih4
z!$Bsq@fb3u86nmTDq6)Qi$Y)_iw5UR+sFlKRA4HTftx{P2_d+K4h}NW4WKgCkLjqr
zQ!hKc_-rH^ZrlYG8!@=+$;0nM-+X{!JozbUT}&v~T2a)8FOLi?z~yoxAVGb5XW7Jk
z!HW-tlmO2iF=Ot<$yR)4rfcX){L<V<FY(-tp)}~7K~CfQ%~U-K1xg>XR@N~R&we(x
znVK8B`#_$616)Or!K`MlsA`ZRfKvpn2H9oaVBP>l(TkJEg5y`pksdZDf7&sxu`~=j
z+vZOc-<^MUSn*!tJ?Ogw@eAl{Y0v$U%J^#W^ymw9O(>r3o3zGOs`%r>0mf=Ct_gDJ
zOd2!7)#%|))>{RER#PBoZ_Fj<b4&>rudL7zE#9j$weUi!6{W25_N~gyoDRxyD(&)I
zNtWgu_Z)2@<w&a<XBy4l2G+EfQc#)$!tD}InmLWXkI{;15geL}^7eU@!#B-L)!$Lq
zA?OUB=|b3jl21a>o=B9z_iF#lr1EATVpcM3=+$Z8wLn(hVsI7TDatiw@3(n*r^Q0)
zzpo8nfWGXUX1O`3-q@2W{Br`K@oakM9>q`J=}@2LQ2m%AS=mv37n83UUbnOTqSjr~
zLe%#{OlH|seq^NW_4jRQWLrjzvaEpxa(ms|Vk%X_IX=^#MS_DXa}7O*63<p%S|&1+
zLMqD1IncU%#~`cKm|ugn@uhH7x%Mgcc&ctd8v6>oZ2eYeQaD#Tc?E!M9xWZN5>3UO
z+74FeEzRIyYwFr8X}o}ODMcM8xIvmYmu1;E?5HWF3ZHFuehq7)DE~myp!OoaA!Fnn
zDD{6)de4d5dPe0lk$O_1sz@L0QZD#9ActnvIaKHX3L)e&V>a`EeC`EXD8m+~+zJ)Q
z2<yRwF0rhNbYcjUE>VG7HRMV#gRCLO5@la~TM<mLk0DWp%4@9nqUz{EH;N<^M^(up
z>A%@51|ElxGsZcd-ewTr)wKDtAWs1;9>xb+H$Es#6q=<;6@4N@dea6n)SM^xS!*~?
zls|Gta#>gN+uVtP&-wL)dc_xbHLOjQj63D_t=ux)^1_yPqwqq}J8$icMVNwiK5+PC
zJm_;eZTece;<4n?!CC+QqcOc}K7*p`kwf@KG$5+sedu~(=rET`ZNXO9#yVS6G81`~
zfF-KQE)2^mh5}ZniO25Mw@t%I;52%LnJD1!V}<!za)1Wx`_DW}#Bi+p0#J>%duV8d
zS6wTC`}8Q(C$w9--&r!yh)_Zy{X`kTmx620n1~Y%z1d46ns%a;lR+9!-8Qk8T{_46
z5^ee5Ip%+$0Egb5G}w)ZaCwnJa!T1E86Ok&uc1U4_N;A{&F%{DD91tA?wkt@3>5T+
zzJ}_01?Sy=abY;7az{UM+v4<C5Q;FqY4`?kpcDZV47;Yy_QH%5vs_LyK6NTNKtQr}
z2W}v{7MsF(OEU^wNON6=wC#3urLAmIz2tkO&+mQ%jFH@pVZx~98dS}Esd`H|uY_WS
zYVM#;y<rVy;WA4L%>8(lc`{bMd%6A>KA?{J{0H=;xp=Qu=zNuWSC!}PRuY|uCa+IS
zEy+N&jT%^fI(=R5Sg}VHj=hAsCt{a<If4AB_W+7lr>PtF1-s#bj0J72^qf<txg7H~
zpWc-ml?H0a`t>7$pn)<Q1vg@>(N&>Kdu+r$C0lNFJ2F&CQ2y=StH*BTjbCgRRpUv(
zsvcqg_#5E&X>6X8^so0f8K#8D_(FpU2)0i3Yzd@(0e1l{P(D321zgkjG~>{K1cxvi
z$*7vBr2P-|-0#8u@u?Wd+9*it9jH?fP_L}894n4*%FA7g`o2PuqOls^sa2~&Tnk06
ze1bTK_SH)!MyO(~5R(D}8Tw&ioPNmf%QqNALkBOKn8IDnzN|ij6)eCN&TkLTa-XpK
zq@F=DeOJ$i9@&=mEaUT(0pvN&g#hK@HRE&aPU2O8)qOI^C#nvD7Sl1J72`zM!jPV1
zT3TRE&(LGaen#|CAg(Qq52ojtD5+lB0~5reKc>+vlChh5IjmNINvcSV8I_M(vhWcW
z4E~s4?kz0CBG!6hnNDehBj8l^?XqfgF1`iKx<x$bQZuAB0~XFW(Jx+;!3$w+OEJxW
z0OAep1R6AMcMP2gBub#!Xn9~{X$1Rbh#55!Ia5IEpU#1;Iw{u0**p^eYlHG*XxuNE
zDcp17pwNP%^rP)*88rJnRK11ICcI(aEIzQW>mUV8zWBg0A)(J4D0lfPW}!kQSdXH;
zE8fG9&gorH8`A^Y<8C#dU(|b|l^ZIdf$9{SbyG}_D!ll5%7?nx)81%^KwffB_iG?~
zm7>^~5+(4J(urc~NHazi$+t5Qm>9&DG7nMnBIO)T9r`}X;fs0#wU#@WJv3(`E3VXM
zu$a9Yj@l^?HJD-cse?n_Zgw$uJA4K>r`?RMw+J3}MeL|Z^JwiE69f8UEH}<<+s4c0
z1{uz&On%3cRiH=iAvrnK-A^P;!#-yB$rz{ucR)|<#O=@$O6{z(81QJD<RXR2*+vs_
z7PEwiZh$Rq=OT&Or%g2$N=ur=Nh9b%oZl|QrGW53)P&m@(&`T}WD*mu-`q81W@ec~
ztr#*3<UC{xJiNM}kG;MFA*#yjL^PvbqeaRTvdYqD)_)KrrLCVCHE9%~DB}8(1DV1R
zGc+LNVvA&tb#NVdbl)h|RBi&uouP1IpwTgso2p7RD01eo(P5>8ZrRTTvVk!PRSTQ5
zBd!S*T|br#D7a?~DnYSjm&i1Ct4m}DOYz84$`nbIo>%rK)E+Ka_P)u;0KYmd4$PT6
zLOOiOawp&MB+c8EO1)9@v)F4(tF9|}U-Sv4v}@=1rJjt+ZpA!9#Y@nGX4FNrH#!3e
z+WV$}KR(cGC{SE`<s^dV{7|m3FdXN&sT;affh}f43FI{pU33dm&$!TWjjLd15`YHD
zl>5csBy-|wC>w5+1tBqc=#?%smhank;|0qK?GWsmfO<<|^p`+rDQ$3!+SP>SxA-k8
z%Cj4z=82bTC3AR6%xA2m40E9vAy`JuG~Zhr*`{rgIBQ7xEnJ39!`YgS1~X^J5#vLV
zCvB!|>A<U`Ua>^hPVd9S_!#&uj^B-F{ogbw@z@1xKj?EaZW%4H=x_6Aor@5Ys<NX|
z!{IG+JjF49@e)WgYJRoNOAsyg&c@*6Dk%NX486f8Y*se}pT()zrVnvKazTg9{~MVo
zl1XlhRy)Nwv#zuq1~N}65SbYM6&3ok#eLus8{7~#VBs5|A$J*eJMHyO_2TAI`A8)u
zZ$O~0`0I8jWYsMokb1=#!Bm>V1V&#M<ueH%hKC=RKA_Z<)o-OA<lBH81dv7VQ+x$2
zFE@y+S@G4hr%upW$Ox*(+sN^SCv@wI@Pz9|hJ!>t*}OcL)&+rel%nBifDrg8xGvL4
z1yR-qPaY81WCD9;un{}al!IO^50M7mJKZXc8<q13`ExV$4D~DRRJjmTr}?A`dSUW1
z<+@kUASKq1B!@w_ro%hQJEU<l72<f)09$K93t6`@W<@`gcixm$t|=?WSY_18y}x|l
z7#!*F@=+P4FJ_5KPls3011JLxh68bJIaY0%xmik@8o;H$h@^d^O#<*GeqwQ=Kd{#+
zs};L|1!|z3;^R-k%gZTLPQniC2yG5|BBJ+*+tL()>~vEyK0z1JM-0U{LU_{w_NdIA
zp|rx<mV|*8honG;$gd@2Q5NOlyD4_5M}_pnaw4LmVeZV#&$7&@`y`W37I$NQ!(u}>
zpw@*uYw%6@NmWLi8TO3?wdpgUT|#N>O@=EnyvhoL0U~i1N5h-qH-;)?8H2C@7KnF8
zOc5Ac7b9SU2;IWf6Z;CyKj>B!C8Y=r*f3+xY?U4_s_z~d^y&1XFJI_!Eob5m57g|z
zh+>SJz=oT?CdP`ySg($a($8kIu^2&MLySyPPtC&vsd2}jK5;|I-M5is11%63Wdajf
zzfZ_=1~;J6ROw0!0ksJlp~%Sh@CkVXmPSYJ0@xe%^WDYS4S^9~IMaac*^e;NRJAj=
z4ipssp~?qw-O?_2Eb=c~mDD-hXz@T}3Ph&L;)LYAIw;zxXwzm^_4Jj~T4gkIra-{d
zQqf&82dA-8QUHTZ!(yh2i&EqnkUIhPHFYJa$TlCH98J(d`B9KyzK=ciE#0BO1F040
ztoAR&$S0uFBM_L)KS^F#CPgE1k3UC&ep8Jak|EhAVJi0&d_y~h1(2Nosc=MlWi3G9
zn3mQfMZ^~;W=TC14`l{56J{aFqFxhj-q0_ahn&>8Ez;5fk5VcSA)@V{<Pcj>AHq5-
zEL)p}hpj`cO)7TkfXtyC4L*mhRaLzDLhB0YiGJ0JF$?u}N}yk{Jf=`+MpQG-wW`5_
zCrxpKF?zp4$4P;HP+qx*)b>&SS+I7@a{aPAkeIL$V24yjX_ax`8#J4sprRTjLc<zJ
z<h!KjyaDr=*(vuFr48pwG?d<Hd;@1Ni*{$h6#_Unq?zHcwi6+8peKq=z{w|@nHRf3
zUfSt5k<E+3UMD&>3Tu4DTm1<;vzX3`W}Jk2V4$t+=Hzt|X3||julX%`B9`M(5=1^7
zRezz0=52&VsD>NGO31mf#*Z!i25B^1;i-QhA9jiZ*`)OuX0?H>7F`CnhH$TbpSltL
zD(UXkOUpXA$F|C&BzbrBFE=+kA-5z5BSbEpSE1gyk>aEvQJN#z?HCCW_GUKD%ck7O
z58Z-=JI&<2XHL*>LW4~qo0C%tLapLiCf~{*&^L}omC(dU`Q+dxM$=tJP`EPc2Rw6B
zp<1{IS|>0ZvNlI1rWE9Th4OoQQAoIkxruc6O$^bV9ClIdLGGitAGLeo*jMeoxA$}I
z_wBw#yAlER1vBxha7VIE&MD)TxiPE7*1^U!QYEfDdYrclXP4+v^DkiIv0bc3%3j{B
z6JOGVV|rybbC~6%mgT80j++pj+(l3x%`dlKd8H+=cSGZ`?2_5}NDJx{B^)E29hV+K
zC@w0$7$mNAUWS*_)CqQ*zA$AErYU1HAiQFEVZZNf@`Y2cs?svPsLBhy72<-FF=u1J
z<yl^s>Q-2QQKs9*tAq~~lblVYg}RHzM89dhD@jgEqz8dc6P<)e&vF|vQrNa=+eBox
zHkQ^5Ap%M)+YqVNpb4s@JyEf2R+tGwYZ^BB&iqYq_0b)%>LW4ZmfH_;Iuc_TIVU_F
zYlBj=`aA*KLr>DSM}w79rjMrg6BE0$No+dGI<<YEewZw6SOf2<%~$xfWokkyW?$Fg
zb0$vfzsIS4cVP~gVh9B{^l}ov?MlhH&2WuKHZjE^7jG7<RoX-wmo3b-T0Ww2w8Bru
zLaPN*RmTY=6_QuhO1@ZMrR3GgR4UeG6?Q(cY=?_8F33bPwMKIdJy*(cONJZGfCiDI
zYK9^3q;X29r8H1;%0hSY2*t&AXh<Oe+ZuZsm}0=PKkl}Y6y&(8%3AuOh@QvmuyvAE
zG(H3dGx6GNa-Qv}lB>A7w(BLg%nNm4qBoT_OKvoI=mXSiY!+^rw+mJ&+p0KYTS?44
zRK_u&L6sbdo-wfqU2V5YTw&iZK1nd98rRpbrFNfrkY~>i?-w5A(L6?f_+Vf11ospi
zzsr|FZtsewd6uO{8Rxj*{n@<-&yS|3Cpn(;X%sw$G4gEDzAKrJkWdPgVX};PUo^-x
zK+qm+qB?nz2w8oL@&q&?Sw(UHg2wI*Yny1TlE;I5GeLb~3G!K}ElxxOHKCqCGT217
z9oG0X7bT3}_GOiKU?rjTVwybvRggIH7Y?Mbq_BH=L(7IF|L05xFkyylh0W{G>uSss
znsUS+^Y|)8cK*`%lwZ>e>k92DIt@)oWV_XI&L*8IogUAaJFwk>=6M^vb@OM;uw)?D
zw4xCh4fja!pClUG7)~Skgx8o5jeEO_nI%tPKBwclq%lDw@L6Z33@dWc^Pp6pOh)wO
zX`E@f#$pxn{<f0IrB)SWP~4P^f{8Rfz!CTQN&&!yW5S1@J|FS1B6pfxLoS!eDL$FT
zl_uaLX?bd5VJ>}I5~ZO3D;nR(5pVO=scm=3o&7bGAO@DA)j{=kV9(TJ;ITTsP?a|V
zUS4<E^8O3_kv6WxGSe`tyFn$-JUKbc(~4tgC!zn~+j7(KDvLguB{>w5dX(bTbWo2H
zR2k_hD!Cz<72djk|EiMf;feRt+A0bR1~oBGevr?jzzU1ToO!}7?PC+zy;AAKX}UVk
zeN?EQFsV`Dq5uEc8X$$WWQ}WF!#M_84uCI}$(T5DX`G44#>Tk+Ze%!{eOcF?yBnCy
z!^#pWn&8$ADl9V%b58ow@*~4j`<b%2y6|mn8DB`YvZ@jzEYr$BY0|x1XL1!>Z=66q
zkr0(|f*MX2nm0Ce3=BQJ2R!Dh?D+>ofezQyyx^S}s=U)PJ-_A&d>vhY>{?%b^ZR#V
z-*;Fyd>r;!fkQ=yK=jZd@R(MJHb~JYN13b>0P32It;o!`R~ZV_n`xx3%qej$l{EM2
zY%52@h-<ynGqFNfnp8t9m&=I0fn2JP-N+}C6U)N{AYij124u4V@r9P4U}Ryt$Q1Z`
zz)~D{?cgr#ZNLkR30OIGCHQ<6=07(G>EJR5v8o<?2kvd-VQn44`6ZhC47t>l$&P!q
z@nZ~(9eyjPW~RTGXpCWLs+bz?z*psY6qz!EQOXDo$UOMV)S4$c>Y+TLif`Br&cK1V
zKJ$^+;OnXOEGxD;CP=Rct_nuOwSON&+4Q%qFVF+Ow|NW&5>Q&n$&mb$Zb%{K=?_^@
zHMYIMj=eRLWv4J>pGwUQGABCEX=*4s@!xo0sjEw3LyK&VuFh*H_)Ekb_{JK73G5Qj
zzx=nk>$d4E8nxU`v#CuDu6_mQgN{d%bAKW)|K@1(?<66mA!YEziu}>Hp2{-2LaP0Y
z8ZnymnRtB~n99Ju*GW`4!VH0mFWK;Ep-S*IH=TP*u;^K%8ard^=-$0kM`v)MWPHMp
zT2gOU#5l<Tch|Pn!rczictlsPmUJN=4<!R7qi1Jy?;RQG-aF&$*E$4guWP77+L3yO
z(KEJ3?|E&}TmJCq^F3q7^cAl)dX3C{ya>}X!NjArH67i3Xwv9(o;I18uuDtot%<lw
zL@eF44T4CQbu$w*+Eluwz@q}o#Dmf(88U|Oqxjg_3<TK2-vjJ>`WqjNVzLO@G9!)!
zSKA{E)+iAD!$yKwu7pPZ(<{Q9z6lSJF{A-i*naVY=O1bRE-m+UjUlZ@z2IE&FTqQy
zhVbBDO3Y&olXC)Tay4V-*3Fh6t7)wBtu9Ev_fp2hDhs)Jh4?uxC%(5Mm#So)y-rsM
zFeQa9b5UUwB531BISUtM4z}Rrw+W(<aKJP6xva9*>$OuZ+sSC%F4)53^R@`KJ6{b?
zSd8v$n#LsLc@n2t)IRpsmyOki=db*FU1`}Kgm&-DuXF_MXO-;8Xx^q#{_^T~Vumg0
z1S`j9d3Xww6;kA^(qheF43N4cg<RTQR*$shQZ|2U<EXU8JSk^DA7%(+K-5hHO4+nV
zpR=VEX?>}u>t(~E!#No1szx5KaTUgzGmI>>M6|QbF1l9hqO;8r?Usd_-!F|8louI-
zFovLAa@~3bJ;G6A%EE|i!G*g^3U>z^f~SjKI64U#3Py%H6<+&5;YxJL%KU+%C%#)w
zFWl=IL^}r^y+yZ?&VZO-x0U)1BOB1J4J?)5?K#H7_-GtRo2<WOoR_xVF(X#lXN=B^
zb{hBb03ZRO$c(4dU2zF<yQtv=di7{?MLUU(N+somsku1{#bV*lzfPzAyPTlC!{}qX
zds>g_3zM(8UR-6VN`+wx6{u2zN{j)&A+y+o&*8&_$`pMTBLP}C$U3uQiX(inUZ>0n
z5@~Y9B9NQ^|8LwJbu5!*PFP1$Wl|t5{s{V8&ziw>7RB(?O%Sp*v1R>#{r|?*8Uop1
z0TjV<6JU1e$|m=_4XkIA-_$@g{Y4kUccHz1w1{FugO+?0u}vwRL`zN%5&1cvEXo37
zRg?bu|BZ3?Cz8%&hJqopS)W~2qX|ixJ~clJo10IQf+U~Z!k(*ZDRAv<nCmtJ2&?ps
zYQQE^v5kbzR4Xrsj9hS$1sa{jtPR#$%({4~RF6cpX3sMIaLs^i)-tiO!qjh5^-P`F
z@oR9K5|P3aO6uzD{J~Ns(iZ%y*`Z^m>Th~dr5b}nB~6gpRYs#7h}XW}32MZ`2%%1_
zR_lP^j%Fl|TG(^-EpCT0*T88^gz8jFb`A;o`C)OsmDf8R`?-Dx%hnLk$q)l~^oCz_
zon~awFNUcS)5zuPzb#jhFNxAw8fI`*x8K(Ef+k%7rV8aF&IgR?1)ZyUCoS~4sxa4`
za<+!4qH;#A{+y;`MhdlxBi}YpF;!*Dt8lApI#So=dEJaWDUv2oC3*NAS}|tzPgyc;
zRdchpT9$xq|6sx-IW=W)fobl_nUctgOoJg)I>j<Z+eTka-$u*a4#<G*y=yQC1xd^c
zH9RyVVhwIXc;5Du3?pXv<6?0B8DS``4n;kYK-jqhG<5?ry%m0aGuz^Tc+91EQ+%r6
z3BZq=nHu|~tff|0cII9jfmi3;rR#({{#;AQ=9YrvzZ`y!$B_7pKQcweUD@~-5!sz-
z8Tr2Q(;3Jd=bCUZ0b+8<8XK6JR%u=DM3t$*2$4EhBGsy{8SslF^@ijWQ08n%-`(O#
z#Z}dk`T$L(?SfEG>J_yci#FQ!nOu@8@<gRexsik1y4q*kU@R83lX~L^Iy_qO2HTLC
zATVUxXe@#*ANJ2ml3x$JF3*_wA$d_qllNzCpepd&zRVmzU5KJX6wo5KfDnZqbpbUX
z`n(;lpl-<ACqHG*nDPhaJ}GmUj;0T1lI~;vn37>WCD$f>kUjYApAXfSZNnAx$Df_0
zvnBC|QJqhtx{XZTuCn(a$9A9E?w$#2>L^krppA30x#Kj_tc%$36s*INu0{tIs`Ms{
zS!Y(`NEMG_qH$>^3a(^M8sE!XonMg^aoL<3MFEsyRABA~K~-R){!!=TKVB#EFt~q~
zW>$RgqH0Ywgxfw(OnVkU`(Ce)XR+PMtmo~w=;F^A^aj>@dx9R$;fnPLrkVyYn!C%q
z$6R{lAu8~xkQ%o&Is8u!Ouk34^LTb4@4;lwE2}Qj6RG1fqZJJ-*vXz`HnAte&IXo(
zbxt;!Lx8rP<+#%1YHC4_G-o3Rxuwi0EXCY1JzQ$D+q^pNk0{eNd`kM}3~}a?j8xg3
z)Bq)`ky$`j<B526db%A%RO9JmW@JikiaPRDauWi0*3`IED=&aB_@&^-h%@CKF5Sd3
zvgi;&ic1Pg!qIZyR&IXjVn1ZNUd~8VDd(V+bL5HK>3()<S!-E~_OmBX-d#aHo`n9Y
zk|UhM5?XjVeTv0Qee>1ghPkoFQ<OTDY$si`cUk7q#894$aUeBg4YSDP@H!F1YcwKW
z#EEzfeAX>y9`hC}?Agg|rrmwJfkxZ#_AV_6Rj?abtVZ@rZ@?aH|CAl)RB@_viaUlo
zf}bSSTl)T~yP|{751v(?eW-cJ1%sXu-dRIQe4mPGOK;CsFG`rxqh6SHTR(YJpS>of
z6@%8lNIl_1Bk%szDV{e!b>$mk+q!kyWP<SGcMl-Bh#6ClkzM@`{O+phGflcI0{j_F
z#H>6I{k#bZV^_Wh0n9I(@A<99J}Ws}zWwQNpe3az?*e4)<-&XCw>K+eu2$%q1#Z*y
zyaQteK3%R28xjnUd~}q57pw<6<B{^11pLptG~NmB0sYHY29>X{k9=J)aw-e?j*$eJ
z3+~BczS^91lE@qd>a$X^>VeP~kG@;9D7HTz&9E~XN)XJq_8aqfA9w+i_-WeND&=t2
z784!=<I6pk3!`>Q&yC+s0xJ$cH{O4~$0*iyYmCKCr>$5gG|1y2beKw|W@=2u2UM#G
z6zgAqnkcl9EMe9N_l413=c;v4L$@W1Prr_8{nQ(E=eRM6Ew2!2z4RW7#S%xZd^sCF
zB1pfu!OY0X;g`A%7n_w>*j8dn`f2C|rud$cA&fILRgfXA)8?z+U0J{a&t+y+A6RlC
z;OSxBF=B=0lS&Qt0MR*O??ZRP;X@4#hYpk7W7D&X(gFfY!8ojSK?;FsWD<Hgyn8x_
z8y1MKi1)=u>WOuE5JDD++*Bz*EXm6wdPQn8l`Xf3dU=sjJFNBQiF&F$YI8Y%-pgFT
zX0&K5G9SQ8*PGB<TO3v~zon&eeo<JhyjqF2DHZR}NGcX>8Nu{gohkrby;GZ)YxfLz
zFwwmBQ>0vo*w@P=lFnKj3O|LO7d?lb61k0`BRz`f>Vt@<O4HVx+H>9h=V&M}mbj$n
zMBY&>c!T)0F^?L*`zdq8Yph*F132O7_%g_{DGtbF6V2ycsY19$9`NRvAxo?_smvKM
zs=ePsN2syH$6XiiU}nI+git%WQOz5dT6rMGCxU28QYp@c(vrX`c@>DR5ZT`hgUq-;
zZ!jS0W+K(yP+J*PC7;~I>r;onIsA{LTX-JA{__k_Z8LzQW24ag@d?F8C!klU@UYK^
zN^P8(2(`#t_l@mqeJQ~wl3f4_O(bgN-2^I?e#e_6?SJWgxmNr&H-o!RoX7Ng4c01@
zIl?S~lCL<N9{Sr{Y^ywC(Immf_GRt^o+NCeJhUP-ymeD(jXZ4X94?3f2)#8CV$u&k
z|FtA_^y;d?iqI-~+IvihQj)WPB2_I`A1<nG$oc=>tS0MG+~ryp1XQSeY1@1jm00D9
zpGZ?52XODJe()WQKdd^^{dPKL#;lAFb3bbSP|X>z>P-IVM)EJ$HJk=nkUZ>3ki6q8
zZ)|3fwbTQ$^5Oi;foJKo7IH+zwd@O&FY8Kqa#ddP=h8pFzZadnCJAdlBY5}jOFDTd
zS!&YRBzD&1qNiEvjJ}))Y&QErPUfr4?hgu;EWjyIi2`351Ku$L-suBhid2#Sr%a^?
z`Y=nmG|KF7mQ^p%n=yLaFBY>vfA%ch^c;zw@^2OyZT=%0uaAZjGlu%dHfMl|{&;<M
znW>eW^?FLYbMk~Wm(fm5dbZQsR8#A1a%OANo#?r%WvitoDI|j)nmU9@>pqa<JLAb8
z)+NaSxeyWp^5x^;A@JxiM5}{zY=kYm!Sdc_-GQz?)KuNmh-SL(+;#4sr-bGy3x}iM
zd!bL$zc^T`_5D`6Ggp-h>z|oR-&^v{Q6Qiahs0vBu_0m=5C!I&EgOB<IR%{c5f70_
zWZfXAU=9}ka_lYIL^O>I?3;RQdQdnVU8#s|Tit!5Yc@c*gu28{{JK(<Tr2==X`&Fn
z7R4&XoL{-2%wN~XIfZTJjK(##1W;FK<4>dCSDxO;!Ey|EE?&*Re>J^CwRoWmMOHOW
zsf!Wb645n;$I6kabRkodm~t{bVOmj{YFU<}P4AwAi4v7Unqi`>p)73XUcMcoCu?{1
zuE7fu6<wi-Z9cXhGq4;WT+w{0$o)N@{D?TI7f4-i+vVbOVrmp~-m_Gs5pO1|ORH_@
z$?AYULpL_uLT=qWma1*(FO+U*cP=OpF`zVqn$6n{{RW-Fh;n<TL0xE-wHOM2oW6F=
zU*#-lWrhDi_mQKA5eTdh=Itgw6CmA7atiE9Qj@}7%2v+5etk|a((8a6g>o5KpO#hD
zmzGvlmn|%O0zH1ZNHENmbB6_srYN9Rd27X31$cU&&x+1ZS`Q?bXrfRT^=4gqBCur>
zzg@oM-^GGG+;r}cU~%R~iIOkO5GeVQjqB@AX+RAb`x<Elm#&kb;nS~9!)0V+!u%AK
zQ^57>)?1=K(#22Ur|<%L_1QW{Q2fRyqP}a<c{=+!gSKuUK;2CVf*+JLrx>#(_^oD>
z;oeRAF6C`49Qk6bq_zDaVZX2HtsBIO^+tomW64(tfGu?Tu4~Hnt!DpJ)@=+*qNJo-
ztP>{)1Cq7e0#59^nn-xf#NEC$Vvar!Lj9x`h9_~S9*Rf!x#!uD8K1brY#ykioTlh_
z;6-U9ZT=K0FH{@+7g5sU2xX*r;*&`U8Lvj%16LAa7eqxYTYfMcC=9Qd-yD4RwY|X>
zdId9P*wVr(Ho0DappSr2ll)6&Pr@^QNmi@uI=61SDTUjsxhG_6!k(`S`&P?WYhaB^
zt}3PGXXn5VXI;{F5_48Du0=ICrN3U3KPBdth=7j<j?9|d$(yhCt}KU&3Kbz*7BVeV
zrzlo-9sNbA8Ibo8U<*HH>C-sAGfk$Ul*?rlpA2%*3IA@I7u?7GR~{zp!48zp`AGb$
zq$t3d%sL&vK5%+-Zb2Mj&HOZebEoE<w-H8ld8taS(!iS4^40so&}Hjt0b1Oly{k)(
z31EUzfzF5pAPw(mVGRmK!`7Z>b73m|D832kNsiq~xioEO7)J)OLr1CI;Gg&-c?vN<
zoC5b>Y5vyw1;U_9igs)(zj^f?&*#HhfkQv9;sL98vO^%c?v?xf;vHLCh6=X7%CPfE
zan3Qp?EO9ej2|%4^K5;6W3~PdbhIUVgrxk4S=!`ta?^MM4a{4ae)+e}^<_VNm8w!K
ze(!c8zP%#3hn`?()Pk8~hWc+bGtbkqT|)4Ox9#OHvNWzb*Ot%t88+oFijECtWes0R
zbd;j(Ock>Rm5w+V%jue!a=9$VN7pQ8WGUIHC<l;{X*5inApDrD<GR@~U+W&gYpMow
zFe8FX=5^~2Qv+xpUqtI)kD<{~957?-x>r;+aqX3^AKJAx-@5x@RbYqL%cofZK`CCU
z2Gt&yLYoW}iK?Q&y-y-e&#ydX01X&E&S^A~xTTfdwK9(MPZn){UZ*BOtyGvj23xL$
z`hnFvn#THHX3E~dzGJ|&g=v}sz1ct5CN?Uju8*>A2s<APzQW8#dE>m*g^}lt$VU#o
zoEz^O_b5r@e?jpig(DG)yK9Ca95<Vua2Y;XRnnrJ{>N%U{6f`)XakE{Y4A<S{chFz
zo{C%?wLsh7uY=b8vmN@F+FjM$+cK!0x_WT){fG%w<F3rGnKw2JsEepRylE9;thP!k
ztb{X9tT`Ou#L?TUcr;%*Kd8B2blo9-gPYY)rhb=8!a_1X%GvlUJAQOf@H0e8a82+O
z0wuh>BHV0bcgM-;y38TU*BvSfC6V<ZZzQ5tphl_bKsur<f8Jgo(mNw+#zBWVDiOcV
zrPf&YY4l$X(a>W};o2K*kp@Rx4*J4=vK)X6Z(w8<VoBT>X-$UKEmR?vYale_zHu=e
ziBQ5ca(4d)iL;q+EkV|)Ol)B@+j=msa(&WtTll)fGkxFkTPHtpD4wnlLup!S?4Wp!
zdqu&L)YV4Xx47B#lzZvRYBQ%)lHxaCR$h6F_T!tCh~PHzj469ZMr_H9$RQ8=0CHoU
z&It`8+0N8i%^$0`VwDYDr^=2^Tu*U(eJy(Nug`>8^3l8Y?bpyg|CSnGc}qi?xRCMh
zq9!GYF8q>9%l(pDV%~k5+lsU}tuO+QC`TTxY81SDbbd6HdB%{%JIgBTii#@h%66`O
zx>GRB<#UGxJAG>&4<^Rs$3xpC*t(SFsF)ei@_bKxL8q?lk3G;D0U_xub40s&VRODa
zU)-Xt!p){fdsa<4FDg5K0hSGRCjFQW;y`=p_LPTQN;Mut2RhcxVj0dgHGkV={LjBE
zWH<@dn@3NKA+2$(h#}_u4P$6KdflS&hJNxtMZy=I$%O)-KUA3}+tOmdb%<;QOu4*(
zDYD@@3@x%P%Cyja@JdbBypxf7dc9X&16zR_Z#_Lye{x<|&6SeoQ$-*dESg#bN@AOT
z4t{?cF3Q64ku<59Jkj2kx=~qO!GqB2L#$gvH-A4ju@D1dR7RngXQPVMq;f?b*(-yP
zv>Z@nhb;;OcBo7`ow(8h!<BWYAca;VEvvwy4@s0<Vs1efE%Z-C#!+e_^()%$-J4#K
zZNvnkoxFK<V=t-=q@0Gu@qpwv+Ca$`^J(W5{ouku{glg7_f`}WX5aJ%zHlpFz@i`i
zibT7w3Qw5icieXkcioCR#@rK=>iOucw{m~5DPFi=8ule%;hLPty%Wp~vF;NsBAvmV
z2h1$jhmmVQsV?0&f#S=xRafX$+{7Qyxg|Vfo2$441qTK7paTYs=%4@jEUrVrcbW!n
zJG#$?EQ>>DAA0FnECD6BLQVilK(@d6uBI38tYkp>*cYJ<Dp4%WN5iK_ZG(&r(Ukin
zBBDh#O$-d27+8fi0|AABOEpV1fdPepxv>NV-}&b!jD}zQQwKMdqE?{c(|<`|1M1Y*
z;05sO)39t@NEXmP(*@+1ld<bK9?rVhlP-T$z(qppi%;^^R%>MSrx&S&ivg$)h7U>S
z`(#+~IiDI+P~GeA52E4I?<KH67u>%34d!c2YA)ikz(?S{++vy3kV?7)qu~w#%5bLd
z%yu;No5Gl%(I1~mr-fHE^+TstfX|LEu@vT;Ed@^NlUo3b%E)9;SbgIS0u+2a|6tDc
zwY5#-##1mF9&F0>8_IeAG<5lR`Ngz{!AQEU;Gd)`zGA?Zd%X(1Y(QW;w3Gp5+r=u5
z@JmB4ri4GtpN*v-T9M%O*O$IBpbhuR0>FfVrAz6Ib8F*k?DqJYwdc}R3E}qLoD$A%
zdw9azpFaZtAjv|G9L+{QwitAV0>t_P4WIAU%#NBt%5|O^WKkJJ29-5985teW3oq|3
zsi?24tf(*X`!vp8CgzKmW$%pg_Kf!Fb6OE`xK4VR_1)TwnpWon^NLj~%t=dwh6Xlf
zo=<q&7a)TwWiCf>=NCNQgaO}==>zNqIS}8@hdenpga40LLChv`?7L2TaDIghlG{g<
zu&xdW8P`|TK{k~vJ_k}7%JoHkZSh>SSS`_Kw7=lj<XW1Ve2Cs?nTvVi?3|p=CYf%z
z3B}e?q%{0{OiBhk_5h#1N%N;H%_*228}L&M2DzV9@V!c}NWB(V|N9g<GVf5!H`5xz
zRwvd{fhsVYM$`k95pm@q3v0&f#t_6ldSDD`4Q)jX1G*YY!OH~CfR@o5kk&Lw=Xz2w
zGT~`&iq!2dz3yU*Ovx1zNpgkA<YHg<AfB`Nce(lOv*G@*B?u_d*>xLX!~3(O|NW*h
zr1jrcq}BMqJb70PW~T<5DR&ZN2}O!?$T@DfI=Gy{QeyutM#hk0&Zxb0X6rU+>|<f@
zow0#y6&U$+Y~#`$EkpW0l6qv7bNdE2lmE<-q&b+gYGwDDC`XHZ%s!=X6iM2($@wzW
z-i|0+9G0g^X2xIeIoeLWN_N`1Ym=YrpGVam6;ATiRR5)o1g0jLH>-Pe?_T@m<LJh6
z{$=Q5r~CUXGW(dFZ}IoWPGM-3cs5&w&#=Z0wzS61%do2OEL}plA=IMUqYAYc!V^Bc
zXw9VSl(wD!|0e=dh%P916js!g{5ghXO>2BA8X+ipTQ+7d6O+ZuvNuldMP&@?uTW0I
zs#W(J<&DqVK+5pAeEb1CTit^^BC^UrWK?r2hG@XQ9*y_cC#1ZS7Y|OUps?p&8Ft0s
zhrq+|tHV6`k!;&KK`%l04G*nXctvyOH-QqJ&->va0`xlagJ$U)FHF*pf!_kgOXc`5
zTbmjU#G%S7`dL3e7bXV@C(0)>X6IJzJQJ$lwb<z~>T^8X7!%|hV9R-@bVhbS$N0-z
zYdybfSnw?GE^urteE|>pt9aKcB6LAMu>bRMiPhn>SR5|v>@y8|LuhNP{pc6beO%-!
z6$it}U|=}xZ~Gi5_|E&Eu#0-ti(b61gJ%d(hP%Ue4QF2N<?u^GMfS{1g7Cw4FyP_S
z@QcIp<mkF)$`zOysF->_CTBkHF>i863ASos%N2dS$}Ec8v-A%1S(Huxz4ZZl<#%$s
zQ{TbGC|RGpFNqJT2jKR<B$pheF1jG;Ex7QE^NPOzebHOY*HEg+d4L$w_O5=7iMuza
zNbcIRF<sYPj9w&Pv`mI~EuKZCsHyJ8p}!i`4{meKT;nh6=?wN$VW?+Rvcmizrxj5x
zUzpja&v|rCc@9QkpR)PAXSfWnRk^`}<M2iGBbb;Gf_WC)UG93MyJxL5SrCiNd&nd|
zx_*g{UFX=IJyw}z=#~XW_#-m&;k;NUKgzt896V)(43OL9fHg<JUI1?x^G+Eb7=Qmg
zL=)swMLcxqtZgg~9W*ySCZEOF=+t8X9%_4FK7L*%N)8gOQLo|5&*e>vdH89>O-Yab
z%|y;ycxUDB9BtpW$s#`2_?3M*1zi!S3m^u$6lD8BrB@K1rj)#{lq1WzEiLXj!6VU)
z7bsOxG_0OOSsLjJ+&X7%Oj6WSPI*~qRMh16cG<s|VR8S0&92y=B3%n}C#~h%lW6I4
zCQkT3h8DG+JBsj+@FVrdTM^k288Y)~Do?U7d2Vrm7M(C^#&M5%@xcw@HAgmGo%R|U
zJ-uowGT-d*kG`ZRXt{FvHRReWP!lC|3|@+KFs{KjYd_xhz`XyxJr*+=z_Mfxrxc=2
z=id9Vxcg}~eHn@jq_?qdX}TR37k;)~z5Nlc-(iujBDBo$dFsBvgH!2p7(1}SJnfU`
zmI~sj#DoD|0n$38U#1_rgsGm4t8_K^;uPI}Y`!BiB$Yiwo@K7?%O6Ez)jE!w4C`j9
zjZk?}p_PrKbQy(3k%nw$22aJyV3J*CHb$Qe#fF}Y{u%GxtIKW-sTI!aR)UK1?83vk
zDDlji^AY$!Tm*?q?sV9!!S_?;Dk(RMuiRK_xR{?zrkfJc$(gBy&KsM8&t{8tJQnTl
zUnCyS;F%=q+}jzCZcd@nn+burW>!v<Ri``2AO6Gnw}>m!<>jJhp6adN(04F&Hu|Q7
z!iS_nkpTh)=aa|(S4gEhNEniXj(fCI7=H0-K1&VgO#mYqTEkv|&<myvHFCe%OV0tM
z0+9*)Vlb16Aqvlw{D-=w^0lCFVibR&RK`?m^inqM0t0>5Kw<(p3I&kEBpL3aMJH%%
zy%f3!@k97f5DD(0H{!$ajdbX!{V<1-3=OasAPJ_p!?}F5td<c}sr9D8woVT0C_Wdx
zOw43X-bdIGatI@F``MjVDqwO-U(1P<jr|BCKf-E&aKgU8By*-~N!qR{PG~=S7E}Sr
zJr29gfS?S!BNF7LJz~n3kJ7?IS_z>L3csvpomf}$%$+r*J{Kf7vVSc+u%)HI!@<Pv
z#NpnohBloMi03jgttOM?Q1)N>v<m*3e;*!UkqDuQL5V^#HJeI$9vu{S8^@WuIN6&e
z{LCQteuMtwD3t{Yx->h7^X6Z-Kj3gVN?FTJRO^oVjEd|@KaGDIP9T!;D4Z-MaetRK
z?8KFvq2Lb8*1t*Yo7p*l%8^we1+-FHGk>%=|IhUNwQhKLA(gd_n**x$66AO+UQXyO
zwfx#SGnKg=e+s*klnQ2rTq(??bbde|6A<WE#p0`Ug5VhXLnkFO6)8JUco82I|AGL!
zOj$|9C59y8h}YMCtL~NAbijly&v?s#t=7&xFB3CqkHUL`w~xj99cFJ7`yR93VEW^Z
zCm9oSvfL(1$a-Ro*MDcYjj`VzUZF`!+~dkj_%Z!Q>ht3fqY=-Or5N^@`Nio>4)9`N
z5cXmMlAX>k?_VYU^VhFV*J2W;^ey-nQ%YLwbp4t*T29fKRwvAto_cR3I9Z7a&?>Cf
zFB_~1EeWHX45lTfD4!c+lq7|ABdsDpi8**le=8p<J^{2*Yt@)=@cZ$;m+@hJ-J^_u
z4~Aok$F}%feS-^jB=sGM67(oZ9U%9{AxOzV>*DH|1TWrZ6G~kiaETe|8dQq;sx-`X
zIU|XRJ+LKCX&WgZh>g7*)Cam`!)zR)Fr{U)45==HnTrLnWKbAKNHayeCT-@;Nj&1)
zOnMz5{De3mNbbY91hdPK%E}Hdr@$fXVO;$9m0MHWkNcEYL?jmFeZqm6`h*1}h|M8^
z3(jVl{SG@iOq@dVMj(exf)YT%X!Ng3&M4UaO+d?5h@nX?%!BA{8hZsxF7;*VVMdF#
z%(F(H1*<Z^$qq5W(`+QOImR?Tkm3=tqP2hT`)Zh^>i|hMU^UpGBMpDG@9$_WOPJ!}
z1I9EC+emIEc*G8HvMK}CXL;6?X&Ehg*q14lSFkj8dK;9NT*%NsTU!FYX$MKR4j<0;
zVQOf2^!hhJ*RN1>ZO*K#PK>yfF)1Qy<Y(35GzluZHl9ubmzGT5t+`GTFVzxP$3?`g
zCWgstnGRTw^?xEeK$2hsl#8-;x%caw)UoH99JO+(#s4)exYRLAdj_ZaZx<`~bKHid
zsb$(3jV7cjO?3cIu8EH7BI6IJ(y9<sV+PZn^WXGi+JyAl;rC%-?}v#DX^u_*kHl0Y
z4ZTaA%iawoy+yR(-{2B(Z}5ZHGjnV^yWh}$sHU{rTdpeh_h1?RL?iVdx-9TSwLJII
z`sHFuvm@aP<i-3acR(*#q6{p*CoyH3<G)7wm&Amn4c)iI)f3B1w}EcBP~d3Ml6Ft0
zxeb>!8c)cA{B)+1Yf3AhS=;B6$@AN-&pKyTcco4a{Bc6$tjwdBbXCvn467`ynY=m~
zB6m#)gwu6%ODwluEo?7IJv#rnCrQ+trv`i8YRfy(ekN~Y!ZMk6+83_xXj`O{Zju-D
zaCw4!)s9ZGtBpflZ7}$1d*i~B9TOL6y$j1_U1_e0p6Y`19y5EJ?o`&4J(J2d5eS=P
z(r0BgN~bsqP}}^+Oo=m~9Mh(JvjB7b-K!5iD^3EF7#qE~cwBVP)aequZb{>ulO0Wq
zwBB93`b0;>8g=n%5)Q`DK{cVM`tE{2%-INfafM;O$I&C@KmL6+PWEr5*n>;x>EfSk
zqvsPx4*N4_N5=f`Rmw}ohhvw)eEIt3iyglASl0q2HCECKlwp8v^W<#G{QkVeA496&
zpD#egAhe~sHRtc-;<3;@yZWuw9_7A-KEJbKq4z-nXL~x+$$hi_1VTX-(H8-Q47bs9
z6!2sUJvcD>5FrdsyRV*{a5Z{1y|=2X8{^IaY4f{mn+b`LzSWsX*nAPc7I5xf;j*-4
z&9+kC6i1!Wmr`Q`qM>&q$Pw>}<=NbKw#9aoXP0-xrqg0;>?V6nY_Hy46nj6m$lkj&
zwm7!QLN$#mzn<x`8;na7+rfpodTQ|tjcKDzUd~O@+qva_!$aBCg*1DTEzBO^phtmG
zKkGf9exU|GyO3fPWoLI<N)}FS&aBGRQU_BYT^5T4CzLYi)%r{f2qF+B@H1#t?!GKp
zRutw%?Aey#_#rJ>7H4+n=Pk?X6}zytz2IBDq1aYnFZ^P&z04Z>DcX&YD$C){)y;#=
zi4Nr&M+W-@9xwVw*kWUt{6$-X)rKt-bsnf0vBR#F6M6i0@F#p8A8k$oUFd1Evgu~%
zxn7_yl-8KPdbZdAaga1p`bC{XP0-KmLUxceQks9w<4_HBp{GNYL!FJD>jmoK4TMl=
zo=i5`5KZ)DihwOP*<b?Al^xWX$n52O%>p$eD=UU7*NOqi&cV(Cy+B=<HJJb3#1O`w
zXzRu}WnhCn(Q+%(6XUA+_V_~QGy`n05t3c(CWTK-EaVgpRS{}}gmc?3)0YX2ZQR7G
z;kv=L;sNOe1%Piqg6BTJ1x{tBc&fp#wsH{|2ROaKbO_2!6nVWdcp0<B28gTOPMMj`
zp(kw!VelwdE=wq!_x4_7rgN)g9;)2%eW6T;YM>XWi>M4!-(b21iXp8Qy#kvsoLUAY
z@8Rq(1mMAkz8Dm3nE&6{VgtlgX~&6{i8{CXW>wElD37n_K)t<ZW;zd*9bz`XY%mcf
zDN{B;TnYA<4CV*JLu8w=WY)ZiCYvv}QqM}3L*fU^i!9Ti6%nf$u)@S)s5it~DLFKf
zL+jAg1x=GpH@UWGYh>H;pW2$RxafLHVFk-YEWHuMtE1#U+i1eCZ!lmbD_G7Fh+QR?
zXJ}=j2CZnbLdB5DHK&QObEpQ|3R^L(!35qQ_tV;oBDSNklChOCD_PDGh<!gjvrK~!
z`EeWd-|^{O(k}(dd8yyGWYUs`dy$t#Eut=_`tdKFGTn)-Ag5vN(;&aT&)e2#m1YD*
zrA>BrB-TrH#;jyHB$ak>sY@ws9&!-|kLL=7JO1BnDSLY_vP^?zE-J^dTd7N3s&c|_
zZh}1&u7SdrMN9extbp99CBtrYV|c43u=U0@*IetGYh81Vx{smm1HaO*_?3Reukszw
zQkJtsWU|Ln$bDWfLQ+ZeC3q6y`aUwxia>mf{Tt-FD9LIyzRfNBe-C*>%0m`SQb{D2
zPTo(Gjr<zPMm~a+Qb|{~g5AnhHuGZpE1XAQAHU&^JVGAQ;CfMiAqnm6&3Ksrh+P;`
z_~*g|{g0WbVh6Oc#(w=q?_CkBN=%^lWoBVAqity21H13C(0d4+qqhV!4%U7DSR;G~
zwxy3}kkP}#_lZ`228P%6c46ebT$Ab6{$I{;JiRwE@R97MbCy2&ezU@F?(JQbXWqww
zuVx(fibD!NW@UE{3uX20UJ56p(l`F`d;k0&zyN-vHQ~bC4P}RMy$;oY3F?9R{xP~w
zI!x;=(F_Pt&-ViWcAs<)JH+)o>j5gJW8CLYhq4(=@9=g?lZMRgGw8@S0N{cLJ_L|~
z5Da7?2L&hv<+(&vAgbUm`!x3VWhMRNOg1I`a)%bY{y)}w{ylR>O8-X_{n&5hu}E*b
ze_#EtLyTzK^HLsSc~9PMi?0ZBbDWWP-wFUSr8}J8m^T6dN9?*VTj%UgAbN;L#q32<
z6%KD_xHrpixRlE|?^EWEdQDKPlpDK&hrJy)&CuHvjsPh9XZ+@#@JMaV44xeEDw0IK
zu{6*lNi%_6eE?06SrRgI2<W33VX0AI04GNFd5)jJIpS3$iF#vcphuEs0=xPEnjo_z
zWato*zP+Y3BP^4+64zFzfmk1z#-;jzFjqI7<#jdr1lSoFfH=#mAnA#ySCa-fk_S|!
z&1i#KuAQl9m`Wp8Pf3LG77pQ>cp~k#LQe||g6ro`yNVzGZL2P@6F5h_%E%`A5laI-
z!sQ3Wre_7_bMU=3mHQ43B*qbjQJAOTiq|0JB`7#!HMFFcx+7T8fE@8E9&J!iqRj9-
zc!cB4i7|!v;6%KGR64J+Qj3WwWrKneo&GnlWEiWVMa^DSAqK!%C1N<*yR3+o{@qkv
z99m@g3>@()l0?l|9XNysN61Vb7H!cE@Rtq}ttyCI@V52?c-2eN^#+p$uY&v#GxqEQ
zXCO8J&N`z%AzCU27QqZQSXOWv*ABSL<qJaXP>NDB)EC(y9dTNw0Ye$GCo>oEVing|
zV}UiW^>Hs^mKr9|UWI2TF*p{n1_m_a4PHrkmC+$KHtQ?X7O{lYHp{kJmyJa;^X#cS
z#-bx;TGr3QY8<vEq>>FzSOf!_@dn@0g%%7Q>$a0(U1?z8ah@Yf-CDI`BF83%UEJ_J
z2SstHq9o<4YqO&#jqEdZInG!FBNS(DU-N2AxA9=duI?OITQb7P&E|puW^uCk%7kUY
zD5xZ4OJ)y>R3@|5&?yBARbm`iBW7Ck{2CCKshI4as8Dc(TZ)!2VG)ebcs+q(TRVVD
zYfGy(!fbBH0292#RdpO-FG&<Vdk2O!SXSWO7-LM@TpBTBZ8W|@boBPl(45wS0rk#=
zB0Oc{iO=X+?<~615m7igTg(u0&!k7IKT7M1AT$UIskgGJM_V54UQcpgajv&&@v7s<
zn37kDGkP~w>bqbCaPnJ`8Jn?r2B?v&vFAF$9*@+o+B(xzqF3`fj%KpYb1NZlz^_v(
zJmdLH8hPRU5P7D@3tz}*ardmwxV}<D>!-6h%xuK%R*Kn)%`C?g&OQXpUh;wQl0PzY
z_iPEh*n~$vcDOV=q6vG0ti`+za_F|Zi_w1cAO*-sm3Y+ddY-eGaWbPW_8u{~{l*x-
zi-!{K2{s<a0#YPbQOmNP=2yr0tn-{&wf_egO)*zETX$w8cPp&%S>j<~{WSf#&hEM~
zbJ(C{$Vk7g64qGhmIucFU>CMJR-an=4>bmk|CfH!LRN}w`h2;@rVa*FCYtN%JZCtS
z%5JTdB>2q%|35|O-(O#*+pzXsaO>Z9q5K&D>+iSG7z#5ApwEdp@vHo#vWZjXsOxsP
z{6JzLSOjhOm&DE*dFp7|R-bg4__RxPjGwvqV>HLCYeY8I0F1tfJk`}hnh|otj?}vP
zfyjWcC}}76)Xr*SR|JA|EO!yX9iO{~OBj`BhR^^>hz&C`!-hgTfN_(VJ`ur<ii%1}
zzrDu!4V_XB-p-2rQ{g2boI<%(-zEBjn-2<83{KbY$4Tm9p?R34L7DWRm2V_<+_o}H
z%W2!sZbP_%?qmtkby^=4?ScU`kxr@o>_>s`u+)14oTIe0?Gi6~Ne<%|ko0>8sU2=<
zZkPp3&TQs5^)+8$t$*v15l2D#8pZ1hCkp8F3s#(Nef>Kg&U|^`3%Y}?9g$96F&)Rx
z&w4receMD=vwSkQz?5kk1_Msf#fmyuT{(%a1WBA;5p}T#F>Cgi&U%GROLBl`Jgd(f
zPX9K8Dp%P;B`AT5XgTH^0AYL6>7vWs{5JsI_HX#D^BdphD>nR$ia!DF-`!LTuLlw&
zhb7g+HQBaJ5*l_fSMyig#yP541d(<;7)+Y7ZL-pTLbd~-J)alUCw(x$%SKH3p5HDM
zJ6Bpb)I0LUfJi^4=72_19Wr0!jAGs95W&`9Nl6VWa^yT;x(f3^OaR*dwu?|~S!*{Z
zr5Q|=F7hJ&WUX&7l*aT|uwYSba27_Z;T7fWJc-f&9Fc*+ZV@k$agyur(V9f5)u1`k
zxZ19fAcN56z%*;XuqWfQ*U`O(0>rZdUF`{#;Bf))05pw`gSBrm?EL!>Q*7lp!wj4?
zrl@+RbQui#4?65AXAV`1SJ>{xmU=5}I01n`G;ekjT_6Cb0=IP?ns-kp8vc4d=qfrk
z2sDCxl3>KQ;q`)p`B!nb857=$-K!|OB`PY{OGbQ{Q`2t1IqU-1sBe(K)H|k_QIXt>
zj7;oLcN{K(17ql|0t0filwP3!2-hWe$SRPEul4Dk$0vM6+;ZCy{rWNp7}5!0yxv0;
z)&?l*Dzi{I->6|}8^m=$`P)+iaclzUd%(BmH%s5hE`i55<ex-QeHW%~1M)3cfgLA8
zJ?1jUTjQ=m_IIAU$LvwrsG59lWng$>whE6?e{9TbUM46l(=NnrR4;a5{)8RJh0vdv
zBY3~e63iB=I^Qf9bH}vFe^2lQ3}Q>wMv<5llL4g5RPlDn*s0SJAW$!n9U0t6LhF($
z@+qg7_`Pp4QE2Lx0ygN>KEY}&pMH}ne2vqg?*Zt`4npijk00G&8k}tqZpScxuh}en
zG5o=iFF>UG6NfK+AMCn=vfJ)kkv0n#TP{1+of?Fi!S0R}zL*aNlv=bxjVD-72&8VM
zY?e5iaWMLG5-LqsgehrLPPja?386;I6zrLiIt%S>%rO~nTG?DtpzWK+s6#>d9E6_K
zp(<#(<+trcIMy4n+-eA6ovbfR-zk+$4hHItM$%7_Y(jJ)IJ`^{YqO)K?)!pXe#v$9
zljAI$Y1iUGq?LKX!<f)E$h`6zBHgPK2f(?~WCFV?7p68U%hoy^W`PR*Sms^~qMyV)
zaytoWpRm!b8+^xvZn}gCSBy47&Zp<NHnpy;*gtD|asfWBL@h9hZ3HAo+6ql@LcR0J
zz^uM#IRd9p4*)w7C}RYW;G8I}(ll#Z?LwH-QWIa)!X|n6U7aRgg_8>pd=oQn!h6O9
z8vuLxffgSkF}#(C986WHHq{qifpV{^T^SLLN*WQ<+qLJ4Yd>3zvCPj>aCT1a34lKi
zFjE$N+p`oZ#RGUg>rwe4q<GEG0<gOP2uznpa_`bfaVt)eIZDL^=0qm1m`k1eX!(!9
zr&piQJkNLZV?dIz@$h{Ilnf`dD|fAe1maisjiX=WjkoaoEV{G{LEN3!+T?f$oUacB
z1n;-ui-~Qa#NmMSNp?@~j7{i(gQ*>l51ZR(U31X#@j9|b9t@q3$%CnlE5nfSzS)3+
z2&6KztS>F&ioXC!NAZ!2CC0lH0P@aT26=Dafg2hw>dj;b*`p2!v%ir1s*!i}A)Q{p
zZFj1_Xmv6;<=TqGw^XaPs?+D)X;EhR=N32xi@}Anottj;uD9|#6<J$)(^KckD=SF8
z<=I0PyYt1D_H@3d^&Fkz{u<A4188|3D|33jI)5-w>*vn5I5$?V!O+*s{TRJ7`>}9@
zMpFjNwHtNgLlyUWP3*I~eH!1(&Al@B_RMdX-8~>cy*p$VK&Jc(egI(~xBGmw?D96~
zdjV0KXd~(sCXWT92cYuy?#7~KZ5ngpJp7W&&Kh-tQTj6Fu=G~LBiRJGd4Zq<`-4MS
z-@|Cgj6@hlAm++xE5Qj(m=yF-^q0_|5ZZHf286f4Goa|c?^pugSnR*6C8KS*Es=ZJ
zHyXr~rau!qrt5?KAPGkO$9k4eSLlEAaa$}_0+T=66txK$XKUsIq|DbC`izL9(|LCA
zhd8!IzXWj?$7%_o{0ran=FpMOH^Qu2OVCaa%ww0vEo`aAy|3+xzz5d&;P3Lxz7Db&
zs>fP4g>`M7zgZ)F2#BExJ_sBTjkk$R8yh_OqE&m8bCS~k`7}-QgAhGdJ-Fa{j6ukn
zcAjA2-6;HynMLbzU85Wzr$*4|ARgl&A5(tBb&ikWSiHT6V|wnslV?5mu5Yz=Z+O%0
z;D)XKW5MN^S#+IWoA_B(5;o8R<6XCxS~qR+>9J4GaEKv3pXYmr>e*I*@>HEYq^q}d
z_L+VqfA#5GF&6TRBbuxy3M`55WuEZl7m(3!U={w)1%OC)Nb+enGp3yNszbP@AI|fr
zv`T+oIDK(%5U-_1&P%6(NW}#B1$ZMD5Zj*uaLaAfANAvUyOS`s_pgV=(Pu{dfN}e7
z_!Vb-d$vzOJTB^!Pw}!J^~VYI=A~6h`g~}5#dN;DNL}M&E`uJ>T7KI+om)J#c!c!}
z<l`}aP`hx5Lt;CXqWt5rs{o(%>*%)Na@ib@us*gDv~CxG>ZRPz`I3;o@^6k5_{J~!
zgJawGmhC65Ch1ItbSzkLAXc?B9h2Qzk-$x*$Qu9&I8JI^LZ!I_q5yz!G{5u|8S!fd
zJ2!D<zCLyw%4YYY>2voW)=g!L9!`fB!fzc>yglb|X@#xCNDl+U^=5bmWp1lgeN~Ad
z4f^;M047ZYh~$s9K=3T@W%8Q~(oUz>zhCLm8kzce(Hv#h4E9nhu*8l=00r`|o{XzF
zOB4YJD`cg<kH+7M^l8I9w@=bEg;X;mQ|}qlP%&qzCdN=Z6-um02!)*$X5jTg2^t0g
zNIf$~BGd>e6B)i*wo5IFYoepQ-U#sB<p^^M!VcdE%gt3!D660zo~xMrp4zHHw@M+v
zjFmhkhnInIM|{%4;r8EyI0(@AgoELa_Ym{tp}eWxv+O7X(Krv!<Z^gn)>;V%kIbl}
zs#c0w=ApHeR^@x+6HMg?(=ad(vJj+($CKuN3f^!OO%BrA9Cswc0X2HsNs)Sd0Da(O
z<s1pf+P2UL@<`}9pv;hI8^~Z775B4IpD{`G4l>hh(7kQaDh59Zi2XVE9|GU154TJU
zjOKP3BK<?W3XQ5eSca@NC2FJ+r~|H|T^5xs_@folM0I$t8kkrGfD2iFbZCvInc+yg
z;JF7NpH1%Qwne|0n~l43Z)9wMrVVeYL1P%6UO183CHfJ3lEe9lgli$KAa;t_0DPs!
z(x6n0+-3=ZgD?-t>I8(u`H`zl))fze<QVthW+`h`4_CNX%e7?rh=29Z4%X`o><TQW
z*2#2SOjBc};Xu(FmZ1j5d4oyVSyD_f><&sHOtY8iOVNX_o<kK$W_VqyQUqE>II!pk
zMWu+A0Mhml<Iq>28!y}YE%vp%l5ktqvk&FlA6^QflQQh@Z)wTT@U|8diT2TZ4onhv
zh}zFKMDvit2uf|p)u};Jo1m4@Q(tu6+Av$y!WOfp7I(<!nM9iS!J{f3BP6fSg>5?_
z0J;!5n;u6?eTNp(KuS&-QqB6N_x5o~UBUiJiC$m<zz%S97=M4^((HJtO>C;vO!b~M
z>20(h3*bflaFq`R3A%387l0GJNXXE1|Kr0eH*I+w>qOQ_2tFuy;ipO@Otv?YzvMB+
z&c-y((pvo>13{PQ<iG*TOI&J&c9+|J5npu*v(a+~U@ZRldT!XtP@N+;fWK(@iI=F3
zZLoQ-WT*WtBH`0%pFh`^YkiZo4@|D-NhMSw8Mu2U>rcibCSrufr7ghq47;$4_TXV%
zCRcF52HUJ--JZU5ZUPl$17b?HPCWF~m0eSzpjO>U=z+%?UNt6WC7Kl#7+P2zKEsF(
zQ&JRc`Xc7>mAw$!GonX&Jnhdu)q6iDZ_hoKZrDwa!Yl7cZlH_n^Rq(7!YkKpGo*#x
zPg$!i=s<g2>$wPZqdRG3Fh;-$>8r?3>l&<rxwKF(D3!Ia>o8!QjS1HuSUXlv{Dl7t
zvfI$>o*`dbQ#1kZ#ZK4lb~eqWR-owK;-*xSC3Mri(R@Zw1`hn{86Kfp4Eux1K=S<+
zbY^&fy0EzX)%X-(;NdP{P|u?aXB<QL)lD0<ukuubSeY+XW)o^g{1NdzEbWDWaISdH
zyA`24giAbdAfb#;<Oqc^O;ud>E`Y2#ri|pABJ^{o(pR{ho$^5InGV#m&lE1zv!qQw
z^W&#S_*&jhQnd!9>Q3Lohc}{sR%^|q(JdrPhZ6hz;I|QfKt4aZi{tbX`QQi1cQmK?
zBF^vNv#bK7MdF=Pcps^o(uC%A(&Sd4kRd}ph`>gQl9s1uw2LQKw1jYlXCM=l1`O=E
zD<jznP}Cqtv4MWYwJ2{N5Hp%Ri@kQ-k#5~^n`RfrH~^jM(XNv!+~^MO7;lDH>SHLh
z2Q%m!uEe+Lrb*ldn*+r9ha3~P27SDKerBo4A))=#*%MVvxJp!#lzKEz$4^VDW?d3q
zK>ct}Py&N)Tc0_lIrO(GR;XuC=nf#L19Nhp6@Hj|z?+|*@miKMAjo}iwP=Md9SxjP
z&rN7O>$BO7Q}B#?q|nzbQV~anV#_^;X&##0evlhLV&)2(!$odDykI`s4lUmeskJQ@
zgqo2%P_!Vxi|(%c_~w)b4|fyk<pd0r-jwi`yn8~QOi(DOmYUz#=tFY)W3R*Uihah6
z*FE4s1sPj(eu^Q}PAI5)3Yx}p95@LJP*}Es86mi&FL8Q^4GN>8uxA$0T%8Llmh>q)
z=2`ML$Wsz0jv^N<Z+OUY6F<!$*V)uiNOg1r*e0S8UIc3yuhI$vrUge)s|gbJ)u!2?
zzzo2=<<f(RwV9%V?~>`cvQz!sqP<j@e*KAxNgvT!#fZQNuBU$`#S`fb%bqt9%YJ18
z8}e(O0V;d~fdL@#r(?c+%*EtlP~kQflmZfKxr0sdP=drrv#IpWB*3`hh5<!CI@1fV
z#!I%QyLWhpr+rb`P_|3y5S0-JLJcY+zLY;(A%1n&tPRQ|Dy`L?`YyHvrF%+y{e%5B
zWI?65aFCj&W-*BE;6Rq{rzP)mWz%3$8p!1cJ)Uvic+yj7@PWqt5uQb;kCB2-uo+y%
z!caVGNDSz{kj^+}J2Z6cxmp~)-7RU>o%qGiz56{k$DBXf#v*TjpA94a^c~R$zrC5}
zW5+o`93&1-mwAAde;xMT_SbE()S?{Yclcp^8IM3Ql{pw19mD#tagX@2gf{PLBT2=w
zW9wTE3+FS)l=_p*<_3l7mY!rKwY5rzB|_vgX?=@dJ^<((i_D88De!s-^T=r=Zm*|x
z9_U=M5?;WC%a5y$k-?TaQ#3VCiK89~^fLN+p4`~+c)wfOFc9fExaroWFXPOh`)f=u
zD4vk>*IAS~npp|#k1z>=5%IIQ`38ll+k~fEgZs%TThzP(q^>JVRcU%EL!+8l*4f~e
zKf>;BR8ckQX1H+S#=0Ojxt2?ZA|3q20;yj#xc{$Z>Lvy6FF)BZ+d*%+=AtmG8_jus
z(0{^lwn-6>Uhx%w9!Fj1_PH@Ofw_Zw_;B-XhBD1Ja@pSOEFxg*86djyQTXNv?U)Mc
zI0f~r;SE8}A*GDKf!VxJxy3?VyqB;xDeGOvxa?GQu_xQfNF({r2#+im3{KAwom0h9
z*3hV+B#5O^WSO~ZVu#By%(y`|(^E!<%9&hTlK|7?saS&4#d@imMrMXB6AU5TF>@T;
z93?jMsr<xV=4+#qfVe7d7L?3lFc2oCwL_{GC1`EWNBu5$MN|2K?CP505tCF9?x%3l
zc)5}yp4Jz-8^Ygai(Z(Z-oZrSR-pmL+87qMD^k4AG{Fw*G2EJUmpvw5Tz7*PIlNTW
zfc}irZSyU6suqT#KSSb!f5k0lsPXDCSI@H(A1xXF^P&Sz-rX}zEa2<mPkV+=!zcQ=
z{d4&{xWXOW;X7~;fP>np;KfNFd@(w2i}<r$K%%@gWsC2OPv<x0_AM;p0=05c*rO-l
zfhH$RQv{WYnk<M8GA(?sJtGxxkSb}Rb62KkvZvCUYnw}S-*6AVFEDx^`Hp+`FX*-l
zptq@$pIF*Z=@~b?V<BGdLZ;b3RYCO*3SbSx8U`735u&aY>5ulBV1<frzT~%Ir+XPB
zrCM2o=+b;?h3XiqYgiuS14>P`MJv`6NUH2QCV&ef)lk|sG+<5%<_((U#!XLfB{2MO
z<*CFOFTvJ;z5#-&ZFAaIGx9WK$GSfPjNOPu07)U(DWI0^vIS9r!mv<^SDqb(^+1+S
zgO3Y+Lxo`pAKfW&4@JH>wyb#K>UfRfBaH%vOvP!8vXXHK5pswHJ04O-Rp&;VfN5~<
zYeUD1&m$Ru<4izfS-p}e<^Z-AG3~mzUf0rei-{eN*(R--3s?rh6cbBKQQH+fr#9h~
z3k-%|*%B;SC=Q<oUD(J4TYc~?P2Pd8qxNCiNRsk)-mj6ej!Z!l%}FE*nv<R;ftEiX
z3=L5UxQw)8%x)mMFBH?#+0vQ_N)CKCGFNS?EIo#ov}Ps&@gqKNi%qBjc?ri}rlCDj
zzL`$_838i0p5!T+tAW(kMnnFL8g~Es%LrG>_tJ4dlbIm`#pHbrEXYOR%re8YPMOIq
zrR`(TV?cU14!&b*g07UE^4^e{4|0&*>HP)irXtTA>~^b37oyK>OrNTTEU2OG07Z2i
zN4mIGioy)h!Ky)FoFA#W<WnEyuS|Nvb3FFdm4iF(hGn_UqBG0yr7@ORHZ4)--LPGD
zX%|9;UhVLw^}b=P9|2+SIF*S}($MB2o20QwF9IQwKoQs5RVA4z4tX}>+bmUhPUw(h
z0gwo#l$4BCi^SAfSO!heHTab#94Y3*0{}^6^*YJZpu8_W2+XZ5nQ6(E0LsWoi*iTK
zMVt|3GbY7o+IlLBB1O6Bbp49M@uIs1?k%Pm$uh+4b*gOCOuos}QkHMx!fd%KQTXx?
zr;3Kxb3m~@B4QmTDneNgkU}I7e6>V6OJB<HbnH5aS|I258x;Q7n7Gi94#rB(5@$C}
zD}0dNRu7V3(JWYk$dQz*nKbFMCAnt^Z;dU)lx#}OFNtQF29tp=S$AR6<l40Du@xPY
zAUh!{FyAE)#30C#36NOy4PugdMkeKFHN7Bi-PZptzOmpo=XzQI-8_ua$oiGm5oGHk
z&8Jz6uAtvd6zAK7QY!)1$TSp8rA&BMmae2}s#k`GG5!qin?0zj0L(Ie7DZTOngxm<
z7BrxWm|}@3n$@a|zgEn9LMpM5{nZZe^#y)ULX(7qS`1w0TMj!=sw+JqgHkkiGC31V
zt@w*8weS`8<jtNndd=R6fbftj#<_xonvxq8f3z9jUUe$ly7<Au%dGZxY9zqw4^(Bg
zn2bp>kPz{p(CeUT-#dd@5}8{YFQ6U+rE0%S)s^r2#+hsqHz9m$T#fuJJNgGtdHR_2
z@N$0_!d1uKMlRwrP9Jak!IAm$fArjbPL&eCq(X)&5Qw`Q8Le2gSyD|-3U|)&%%A#}
zv|KaCP;KuDAwre7$%qDu*{@|v8q>%a=APva5s>@X(@Y89gV{3Y$0>1!nB)g=<UZxh
ztS<Xf9Q`e=^W-^P(pS>l3X!ISCR|Gud_@Inmlt$Pb$EJ<)-@pSg-sSHqmdfil37l#
zLXuA`hF9g*-m=lQg6xcYtRW8jIa=n;fX_Z-XjdfT`GeV#R4?US_RX~kdT2(AR-3)Z
z=&>H8tmH#40N6`X$6(WZlZ{wdc|B!Ia|KD~aj^(O<}ep1XVE2YpHlJ@sT+-4s)lHx
zFccK|2?mCcQ*d@+Q=`R-9^2?>?<T>59WD$<nVI)PAeO+oKy9yrS(e^1y|s8G!X}C2
z<nJk7V+E-a>U@FHwFBt72?j)k*SVKA{;NU9ZW5hS+rCE!9F~RX1VHC3Goz;_v8ksH
zX%z*aqg^rhTHrSn%)fG+Z!w2OSdW8d9%PZahZ_xB0ojcb#tiO$u~EXX9hr#&)t$6v
zr{aq7OnK`F;xSV^GPZ)8Bm-kk&6vM|v3Xyay2>L7R$W!X4lPg?E&#X4CoN&AgNeem
zX}u<h=i$!l==|*1sTda4;B#Fs4Mq)!?PtLiGbJJ=y#^7ddcbC+*o(NCO`d$)%&(3}
zr=kH2OeBw;aeSm`7Mc?&M`bGfV>-w{1Jb#y-W}LpM8=%FByI_-*?uOIcI<>s=-9Af
zuZoEFC@kf26S6?iUiiwLW<iuiiV4yWfY8JS<{7E3R}_{2;#7#nI`W!Lo?>#OoVeM5
zHx6owg)50bwqn=U5vfOkYBF})Y@ftq0Wf9{ccs_LW}G6BfGcL^m26yC&|lRUjT|#+
zhuEwA!GC~U;e#9QcRq!Zm#0CY=1#IvXgdS-8r4_K)k>@zyrvXh-(^l3g0K@J<WQwi
z+6|Eqc6`|zQw94WYNn`Dr19RmZxH=qn3YRWeqCBLph?>sm<u|m6k@q0%8NddNX+Xy
z+ELY2m{Tlt`K38V^LRV0@*d8V1X`9N9qrWxe+p^%fOsis3zRdwDbWeFeY3*#cgVHp
z_+fOezWtl5@ghCGWIP3vI2_T1tgOTQwR!YDTb0*D(bTTBpt(UNLcuGGV+!(98pMff
z<?}eTv)0~~OQH-YHbysob&j7m5MDM-<VI;M*FYAf#4sF;jBzjt#{lqc5G*iWAroHD
zrImZL8v`D&f=Ia-PX`nKk~TJ}@6;~rL^}Ysv;w<?qou&7dbJ!iKZqr-YSkvJo{%JK
zl1JFJ6PHX9+rOPfmUcp$+sxN}<mC3s;$2^Rn283tVS>`sDl<XXEL*R@71z{n*5p?$
zG!V;}QcJwJUPI9ZtN#+2Bo}!CG@dR5C&bNjR4T+)-)O5E4q-E;1*=-g<h}Fw+bt1k
zAXA1MM=s>?0#REYhh~!h#h3(Pw^D}rC!*0s6>$cHa@7|lQI6cKauHCQLK%qLjt!e_
z=I4aUR)|~zadWNgbxqZ|Q|4RVR}2f*q&~0;Et&&ZG%tM>YpTF&I6S|prz1$C$&;`A
zOW8(0^%gpbD6So<PP_JQ4m-q%IY!S+g9ESTqNYWg`1ZQscYnkvMW_+YYO1f@Np=90
zc6&wGmc{9Pios49NvE=dLe%t&*8zieRL#WBz@4-XghYbe3*ikmU&oRyl_Q#fT|^vh
z((5ZPiL$Rlo`6f#dUuWGC#oh+=}dlxuz?Di9VLti*XK*X+o1p#le>;zI8JeHFsNq<
zI7Gyp@pQC=mdkv<vvihDfx;Dkp-!oy9iLcCE%NEhA>K~FEiw>0QtXR(n1<jBQ(GZ3
zXbJXZIs>2<(NnfO>a?A<(?ZL&n#W}VM!i>CTgXJe_N$;Q80;p_(DfT^&+Cz5-^2(e
zwe2+AfEsw1S=PILb^B}>h&Bs-jt{fasqc-uKxQNafmgDptv*(BLZOzV1v-SugEA25
z(#f}!VP?+m*M?I9qV2xcxw!cYF~i2<ujmRyUYdu_&@Wbv2mqIkm+^%uC)qFtg{ZMH
z&FvloiHDgfW}&{I-BAG`+>Z_Vrhs|cs88?aWkxG+#AMHPP5~sRXazOPJAy~HBEkjG
zzMo&3)Pz~aCQoo)_VP~IX`ta6E;s4Lsw%gemhxz+GM{Uuybmvma|TN7vbMeItU9ZX
zfN;Wfc2BT9i2{;~Vf`>wgC7>4M@{}unCLUFo-&{=UF6mC4L%h`J?4`ADr;GS=ZpJH
zXG4A#5RYTmd#l_xQO}2>SIk#OJ{3QikzuRbpop0$MvC1Iwr4<kd_|23iuK^uN1{pj
zb6ul0jSr6nIPG#&>M$xN7!CCjj&}eekG6mKL);_kSDfMvSXWCPU}^$2TFhv#7K7$n
z-qLm0&l;)kh(*Ttgp;!?0-{cFj3AiwC@k_ivI_|r3z{z;Txkm~@?!kvG43{8!yI_H
zP2K>o9sU7vk(Z9k8t)vF9}A<_X#N%04<E(tcS``HB0aJRzmf~Kz7~`U&oKbmd8BT2
zel^eU^5&!#p1}yOme^ZqOd|`7yu0Im7O==AenkC2Cfuy(x8QTDyU`kXu_-9e5D~IC
zp7(<Ey3W^eW!|KrO6A)$?Ig&(?YGNfLbG(+j|Zk+T-dI_6<&mLDsnSO8)RiH-H$76
zklONmx^UlKuwniq?eiC8Mba@kGTMXnWB{mjio5@8Wb`v{jg0osT6JWGVQ6zkcDGVS
zE5nR3!;91$88Neg7+^TdnS=*+o2}igfjzkqVivbL8+EMHERM4ynB(Z?{zxt-Vf%11
zG8$<;g(-EUxJsZu#s1pj@w$+~5hwOg@D^XYQ`ezY=A=?#28`cKpoNgYGhh~*JTf6a
z+Doxq&VX{n(zu}hP77N9X&x3I*gPt}yYLduurG2<(lfKKgRA1TGZ@em9#yF_k?9PQ
zIU>!3Th)aY2gE+dEEo(362v+3hSaNle|g8JX;(^-+2N8E?LGxgtYjIWU`7FSA+6@m
z4zO)!GS_KXI;z@ilwGqjE8UzbkC#@@aJ91J+u9esh=Gk%I4j{r*A!3Y>j_n@^H&?I
zBg3Cr%Z&_w#x1)A-v5kf6Xlk$Tg#(m(A%)~{RQ?EX#DFBRd0PZGR&+A16loFcC!L|
zsz=QFWi4fo7a(4ftGq?@!b;0)@*r^o`ggU1>X3i%n^AZXm@#!s8&~*hZKW$eEp}_D
zL<m-^ADsva4VEioa7<YPKn3h4`J}R?!(!Y5b`!OX^reijlmYH^cA+huwuQTL4|rSD
zOem-B5mm2QtG5hW&%0OKkLuY8shzjGILD|3Ub|=we}u-d@0oP<uyV=}0(Wf<;h6^Q
zW06e^Krw{X0p(Yc?9Mf1FDs0$+g$T{%&y>DOdN7##%44pEp_;bmc<`ZyahP(^4PiT
z*z^o8S8r#&NgMcA9Oc=r?G}n%dUp9fuCgmr6vqsm;PRy}zc|hxeH<O;rPrXmdMr~H
zhLC2_klx`HWT%F;?2Bc^O1$`@{+MlB-g&W-c!H+e*P8c-a{B8G?`R0o9h-c3`V*I{
zBF$8|4FY-Dj|n!PG?^VQ!pWJx4nZIeLAaNPM_L`5Pfla`9lRsqOa7Eo)j!ov^|C#z
zHEYp3-Pj<)E14N5s@*HE=4&-e+iHIQ*eN?t^W|4O4t_X>H*~f(6<is*DEzvO=2EF`
zY5i&coUmveMMMV+&2vdytXP5oK|9DQj=N>4F$5)qvc&XqWLt@@tf!9+G6@NR?8uTl
z`?-D1{J|N)?5i4Udgt&)8;a3oEZ~=%=wNrmTsjg-=~ONEbp?o*FqGG*nCtedK3dJc
zdN%e@UeXW1k9{|HdnSJQ&@Ahwm4Q&B9z{HS%4ZK{A2?TrqA#&)s25%(DN^W8+pF4w
zwYHzXA&!dR%UZYF1A&}&eduH<5)vHpSWkjC=!L?$43#VuGhLY~J+qrq>Q!}X@^Ol^
zla2$&|F1cydK(pVC2zvlmb@?&Wy$@Po-5%Ri#P3|lW>A*vcXpPo@8hqpBFS5U*36j
zeJ0<sWl>y(ulEw!gS|3e*6+HCFU-yiq30EJ=jB)N1OQZdz`8ox!vt^(UsTVm9Ig+z
zYpLi^PXeE6zHTmF=M(wUbeS4leptCKY+W2~7c{gKarwHQ*Q=@c+g$H$-6`XmU6N+!
z*NrB4K7?H@eZKj--|U=kz1Hy6vQ|~~bc)+~t$o&2q`n+N^VK!a9{_;PYq*?N0J$w+
z<xKY**`j@pLagDn>Fq#3D(cbTjiur0|GHhbZO34PtCx~fUORR;b8JPP9Ups#VZvXg
zWGT41d5x$#=U0DS%Si!F`v}d6l@#FWjksjpn5}E;s>>Ux^>_QU_A}tF0xDY4t&fwf
zDV;3zVbGm+odcKdo34~%R#%!&3?i`fqdG70##<CoQTc7+d+}0mRB6&-V4MY(S#F*6
z%!xU0=D}N_V9^p}$dzx4o%T59j2mYUuc!B)+Vf0Jq;&H1w?PYq4Lw+lF<}|YS;u<J
zLp=O0RdXY5tF_l^4E*8?e<F))@+qW<Eo@^ul6d)NwF&+}fj;@88P4+3zxe`E2l(#V
zS3bQBvm26HDCfMuDi|Q3Gtvy8E-o}x6R$(Q+>G<fjpY{|>^HlelA4q^hzCUPN~&C)
zVzuB1CF*!}%mC?%D5a7Sa3XR?lM>HP1ct;wx&lfO;oM;*Sw=&J6OG$NJ3&+u3VoE1
z$@P-SvzAHY=RB^g+Y8w958U_L<E#hlnsNblI?n*eWAHG$sLmtasx42&O#F`gLw@+#
zYUXExC<6w~#Eb@s(&2YGy4x53Ehjv7Aek%)6&8*Lllm43e|-67CA%l|7atxEP9A$w
zNJ>O;ny?TNbBDfftJm(=6KPYV=y)U)^N_IO)ZYQ%!<qP9SrhX9mxiyZBX$*Iacb#3
z7_-Xn{bs~pqBzT{JA*}ZGv)!Rq|8d^U31Zv->lR{1*3BliF?J()9_ZpIy{|4?};E3
zyeg!eW#BY%?82A_FDoj}U+X%9-b}0ecT6#ze9D^uV-|JM=#i@&n}vd~$IU&EdVLZ)
zaAwC#>fxdTbuU{RXb6&(5`-e$gLRsNcpxGtx7i|i-V5mQTbNZK4jqkTS;6gK(&8r6
zDN3GyFMa+CO&`nBGy;%Hl~6M9=vB`rA2tDq5ThD0OHj&l+k0axPq~-PXT>4!DW{)~
ze`b>$?$HC-4%qwBckIW2%!276)n()n_buL$p?NEdD_R_B7xV^;BM2BOl$*#}YizR8
z(hP;RSoc-yHMZMnB-`X1pmX<CyZAFP`)4EL*s^=vIg22#r?_M=LdPL?R9v_Yq@!I<
ziWbSR%@NIZxlZc-m@l*L==8|9lV~u-B9-u$<Et&Bi$WJBFCPW2JLiQfS8Fla5%cIS
z&Y;Bg`TzCfL&P<7%ea586n<R~fpNY!L~3y%fYqSj9iYP1e);JNmLAgehOBWGp4kp+
zRP~OJ76==cggjhflVDd8lDNtgKe~pd&#tAEMAzApjB-7#j4lOc*O4avXMu7)I6$#N
zw;Qwy57i$F09;xEj8G~TipkYkiLrIGPU9mE#S^*I>XSUjI#xclKCFOO|ESR9fhWUo
z@F^R%`;;?v+*T>;81FS8fW0ScilSW%#?wj(h^CY2v6{lv>A}NP#sFk<po}H&#b5nI
z>l@YhXufndLKsm$pm+sb&bU2d%7MYukb*J<A#4E>7Xs0PtTVU#PgQEc5CJed!n)ch
z%XKWWUK=5}qeQy~Lpfkw+z74(w|AOFQE#7b>B5wD0U1D800%<$t`UE0BE;Mgml9ZO
zOvG+!IfF6PtXTwLP)GN5lcMPn;HrV4-s%u0R(kI!lxlZuIr<FKF)iBcS`S8Ge`kLO
z%*ira5-_Wsg9T>z@~R842^Pm=kYs}ZW_>pOp)=4TvbZ2oHIQ<rhCR+o(!I3C0Za<B
z^+-y35XD#kzczG<(ACjXtxzmCSgB4A*IE;ZN=*R{?Adn;uynZqVHL7auj&rKL{UJ^
zWFTXQu)L6E$WB`A0EMjpo^g)hB3MPEAmPA{y|1=Qz2(`OCcZ&gK#)k72q?ZbX8yvx
zABvGu{Scs}3qoK`5%QeQiksG~ZB@srL|Fj>hKvw{R5P;<%xFR(qnEd&%7lDuX(Uta
zSt2=|JGy`eu`$CV(3)F9b7-`uP-I~90_1?1ZarJ%lL12^0l)!46<z|D7^MnCI9Vn%
zg0a30ue2`gE_mtPQ52KdydeoxHl<O^MA@3SH&M5j4U=J*<oHBWwrnxNG|FgupbSm6
z5@~BS9;Md#V-`nwd@T&7$xbrbia8!=A9f8C6(^@S4nmU{8A(>Ohk-6Z33ONjL#EJ^
z5M9J*g-0s;&f4QZh)El^({Z^fUHiTvM=y_k&k`6_wvpw+pptD5o2Su^hg?b5lptJb
zB_Rq&5=I8h;UI$A`<HGx(HpKc7?VZd?_fz=z-V@a!Hza;LSY<Lj0Rerg|Sk#qQ7k8
z7*A)g{02o4q{9paHce$i81&F!?E`taZ_Vn;_~)6%z7Dbz1u}^>G%{wH<TC1UxIDf<
zC=yGgGP!#X9zA*X0#Ezex^(5*x1_+!qo)0TK7=q3m0-B$y2TXjGD;*}p*U{2?S`A4
zdghVG&N}Bf>KG9$#(YXnTqsdEf$uqIb+Y8hm#4r%hoqC0rUcg_uH>}ZuFV~+Z8p<G
zw#-gXWV`ExlMdLH5F~?yut!G1;FL@}yUumP&3OQXU<Ac*f~05`o0G)2S|I-de9zAh
zfv_OI3-m`KmB|%K6{FT@b$Wx*WVTpsc8AmD_Aqtz^bI<qX+!~TWXxhzX)$3JRyc=S
zD331?io_DBOkTh9;L($3FJ8TQ?}rYssG=1`Dk0cZrAExmQ4NM`{jD;Bq>7;}inT1h
z1gqk#ZT#!#)J*qIOiE4xfgw;B9MM%I3XQ3{AqCiSad-leR7*+QODUC7@=kz8XE0f8
z4wuIl2u0$WMBS6h<O*fA;$BI5k=w;CYIXYDuXf33GFz-RyTj>fi@V{Z*N2{7-qq(>
zEbV2lW%AxAlq$7GOIt@*uRb3P42_I?Ytp;kH{e?z%*-t;t*mWq2~a!xDjXc0@~Y{h
zOP}(t>x-MaM_YZn?C0&{t6hgqUApz??TWvC0|u*V(2!vxq*=g#gE2JwS_HoFDKva9
z4w<dp>zKpw*jvT3(zenW`|Netvfqzr`=mv@y?;of)#(jJlbPjKhGlxY197_Y-SRo%
zDJ&{3@s^gA`zk7{s%vWN>KhuHnp;}i+F5VqPML1f<SA3Tr#*;dGkJ35*)n}^-`xIz
zd4uvfu|P+>AWE{L0!=qe%XVDP55g!;(u_A(uH;!ZEroPk1nDSd>Fe>nqZ3;?ZeTe@
zagt_vQC4-+cKt9;^8#p_>5o9`&z_ih7zs99)-CinubbHR-^<@m1v3__&9FP1F1N?)
z^8+9RBN0`!zZI2MxCB?;A9X@Qkjct)G>gsQ^7sOwNGy@c<O-#VQERk1y}@WQlSrSV
zhsh%SVHb%dQkk4Re*aaoI{+99+&O#v1YD|`oi^|ILT(SJ(hQvOybdA7%9wp}GSKMY
z?HCUu&re_&REVlEZbURkq+nHJ#;=)R;ZQI=^9#c;@GCE{@vl=?&bTD<?;kdhBmVLG
z>#xu8Voui}l!LFDLA4LQgY(RNda+xVZ2SB7nQpJcJwmeQ|DO=uUFqzd(b})`m~yS&
z&-VQz*M3(<NA6Gt&bPaj>E+kuwfTp-d%RqnxlHNSS35N7thj^ep;foYj~9s7v9&Gt
z(ECk^UOp2&!uq|0oI>RIIpISY?@7o029HWlP}ypyzTl%B;WUJiX3^2nrT+zp+qPPl
zy<Mf~tBdV#s@2A+$`S5()A%z@dQAGJbb-H})WzBRwobPbxrBGS*G}6)0uhO8l4E&s
zL}tsH96GnwCs|W_dpp~7p2s?Yr%&B#>i8njG2=bir|lh`<nt4XUOp2W6)|S3+jvj7
z`Z#v+dV6*3e{+-gT4P+yJ+ICGZqDAno*vJ>J{P_X+gtt@A_=XY7EfN@pC>KH*6v2`
zEJv-JgMe884H<l*No;f^SnMQua9FbZ@ayL4{i#1afBZN?f!axGA*|ft{vJK)-JS@3
zHFU2_ef%Q5b*it=__O3F#<#vbFcEEcVlLYw<p`Y2wA^@F?yZ*(lQ`3RaKxZ>x~OAL
z>vY;)YVEkC&AeKh%?;NuAEbaqT1=Q7O#}o0Ldint5Y!u6%)##2-aUbO(toqZV)e``
z1`md4&~bpeXv;tk&=C*-NXqM2=XRt<!=tT}@pa8(Wx*EfV!?{6=hkGgCPk{Yx0A^g
zHf&`UCnl$z>};Jl1{`^j$KJ;AV0e)HHlcqTdtY-)-)PxqpV1i71pq2iR!LDrhoXoN
z#s}lf|GmXYGFl@!CfTr*Dx0!TrP3vh^kAw2rmvCmgZZ^Q@}nO%N_rq`O^?gtzEi(S
zhJDy>cdlkI2+0a2RMi;Q4H2^yfJjy_p{mBXZitu-0U}w!gsK|jx*=jV0*GV<6RK*A
z>xPKgM!t>v{Q{T5rSZ5uxL7|MI(ftjFj8ohFvjeJQmH0I)1QQ84an8C;nipwBHC3e
z<6;VJng^9r%sJ5w3lXzC#&1t<uGO&P#%vO*qHx|uLm<i%=tTk0hz0C>S`sSJtCKcI
zpeh2Mgk=pvqc=<sxC%o%pb(@o0fxnZ07xNfQ~&?~;3}})i_ody6ilcZtyO?XRxqLJ
z%eH>e-v257f37&Y`u<QYU(PrGN3H$s+4?`PHD68gtyv}czx}UzNZS6-JLJcek>|AU
z|0Mt&`MskfvE35=L!zT&`yr8Aq9f5iY-{<1oNo#yRMi;Zx*--^6>}BMYyc=DLba&P
zVWHI#C(IC3&@Km`T{z0sR!k)bZPvoB`k%hN;$Lj8ru~1S;P>^C{tL+6E&Wl3tWcFh
zfONv5H6$F~NM-VqM+M$+nyWW(P+zn6_T}Xvfu7Dijv{03;QJ*b?@^Ah{j11UcB81m
z$l6#@-?R#nSLk18!=Pf1X!+?30egd`=Y3f|8^2}qcs^bKQ>Qv@rrx4OuyU(px(X9t
zl);AoevP36YCviO>@kMQ&$y1{uGPCN3HUMV_{WpQZeJ~|%*y-;hGw6TeJafIHh#A2
z#~-e3PJ132^%=l)M&7!9YsXOdcc`4`vyN3Pgf153rX7Wp4}g4rot4^H!u#|fTtHs-
zxC0YmimtF{-LPdKLJL3fK`$!rR02PBY~Jq81bAZR8P_fJE&N<**z9dG5GrgM=>C4l
z6n^O`D`o2?bfrvma6b0bmG&k7d&PC5r|lb$J63hd_jx+)WB8X859<#+XAo&;az<+u
omRR8g=jv#6iE5Qf3Ik_2$0bg}{nRF2Ps2#MUtnm%Xxz7F03s32@c;k-

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff
new file mode 100644
index 0000000000000000000000000000000000000000..1da7753cf283671f5d127c6949074d212a843c58
GIT binary patch
literal 52936
zcmY&eQ<P>su>IP$ZQHhO+qP}n)3&F3rfr+kwr!iY|NC@r*4jC#eRd_4JY?lmj)#J{
zH~<LnQ>I4&$UhIDDa9ZAf0_T-|63$P#l(LEsXv^}4}x*baik;^6;uEK;2#q>8~^~L
zrW#f&lTc9;1^_@Ke{3EA0IHqIKFC@@g@O5}J@_Nk{DEOihE2b^sN7E*8W#Wn`RNm;
z9jf4g!o<$V5deVu3jjd;=%LMvGmuYB++7I)0MvsYKhghyfCge@?r32L0H9I+#Pc&a
zkPEn04u^%2%a1QQ?T-fWe}M@gvat0s|KWH6fNUQCkfpAcow~E7nUN^~P}lmSVgC<s
z0*G&xKjaU0^J5eJfE3IC#Ld#q)e``yNBxOU{U;_x{qtflTL+ULU6U;U0ABKgsy}Zk
z?si6=KjU`(HzwhK0E-4uus5<Z`{5XW*1_Xv-Zf928sd%)F0KGTZ^chvXaE2%0w|h{
zx}&q%kMC6P&m6#iumV70ukg$v8)t$JQA8A*eK+^VN<?%7AtKaV*DOJP|Im|c-Z=t&
zP9*p1RXYfB+3gJx>J7UO#t8#=GHX;YzaB!{s=aYjrnDsqs8X@;d2Fs)Q>)%=T-EE~
zb1iroOc-UqHsyEztI|CxaB&O20`Sb~N}ZBf5W*bkVQJ_ttD&P-Pi8Rc$d0+HI-ht7
z4u$M<mwW?Bj<<qfqD+FHEqp>mi;1}T3n>;c6yh4m7aHsu2|r5|Tq5Eb$tN0dgBav@
z+#%2)lL61fo9|`r|8ITL!Q2qAu#bzrLr&h1G2?r9v;oU)QFi=?C<w8~*aKtB+JGbA
z1l``<`-(biNS+Bv^ZeZ45%FH!5p=Up*P@gov%sM0;fnb%NvBK<q{c5cY&&@v*M0~6
ziZB8~v!7Z02DaLEBK4Tk2=5Q8{t(VB9{&zXv(hj}oX5@lN^g<uwsOIB6Ph$x@HU8n
zM#c(Ve6*i&1SchY)TgmJE((fqh^KH4+WwxzZR#&%oxzf)<H861I;|YU5k9-ITAK*s
zs)k&j_37NWLIjafL)?^6WrQWed1KKpuTzR4aV0bzHLG*8;tlt$qg0bLC!@8m6x?@;
zu^slSJ^nk-$?m{%N@F+%N`|_q<pt;kzX_FS&zR?xehc3|jri!7c!G`Jd-^PkeIH-4
zAyI$%2?~3C)3n-G5;gBAhR(}$a8cd|D=;(6<kO=9{ksit(Rc=7_6lsLBk7Elanaw0
z8uqLdc$4yweb9(#rel`bwx?Y>1FW+=_p@fIQudesP6haH&Hs*i)@9=o-n^54YNhm}
zRmkcY>lF%B3K6js!dpu`@C*zfvZlS(n0vT8D4R^p><?UJ&BkOkQ#o2au-P*%Nl+I1
z#~5D(!xu>+C(w;jyLemM)+$pxE7uz13-X(~#&{Z=UFHvWHw?rtrol8^aXJ%I%Z8zJ
z)vr{jY(|z5+1N}!I*}?eJM<~y@Tu_C-pDg-H9eE6^H%+K?Jngg0|l1+pB=<ACoNsa
zHdPIA<eTN*j_&fQU)!>9{SvGsHM>-|w3Mz?4-qL1Q8<!lNu8LMwge0riYXLA#FKCk
zQ__=>AJda5Ix?~XQrrG;PWj41h&~(w9~MDVHOJv85QdH1)Fa^D?|5=D)8#0K3kEnb
zJgyT%nIM-RhD+|ciB@>AXxPk&bzokf*Ef<DEg1DdSTSGZB!u^XB-j{Mlq#I`G>KoJ
z9G@jI?syjC6H%N<`UV)>#4;J5Bv7@41f?q}ME5}gD4G!vq8M(3xwX=2iDia!A1fv&
z7n3n(Xwv57>*70RT*W3HH-Fs1%04T*kGtRbayLY`V5@7X+g?)2B~KGu9c!cQ_iYn*
z9!JnbANj9*YFHswM_Xq?V$Ndta#(L1dd<X^rI|)Tsr)wfHrbg~kK@a{vl>_AId+pD
zt7amj**TiY61K*-+zjd%i)1*c-SW4<o5EODW?nKYi~`F{2o>MSqoF&(lXRIY&HQ2B
zZLNYsBJ&9v);kUa4NE7o$;!#Ixwb=DhRLLgxfhYiO!=$AWnUu&<{Taiu!&L})~`6P
zUkRM*JAS@_k1LT0bzu1><u#~YnNexC>g|~|O<WK&azfFHT#IB675uXbvx7ViE_W0Z
z3S3?jWAF0r6eX!VT8E}hDNPF9KG$asxx=UnSd+6%^R9mH#I(hI$ut9LuXjbvv?QrW
zajX^hQ}Zqyio=#N26$u3m5jMb<HA^_wDNaslMmyv4w0vDY;d!6ZjWplQGN^WPBKc$
zF$8AvMSPWFEQh7!T2{@STnRNR*)=|nZRSCpTJ8Nr7rkiEDqvleOlg))6}_MfM%d=Y
z^jpKijfP}PhIssg=KR(87zN?~GJhY!kW<`}rMP{>jE(tjn(}>H*9SRd$7cCWu8!pV
z{|d1w4aSO%diS0_c*6eRetQ2flp<HFtk#jV?)#-E=01A9dv~w>WVmN<xF-oyK|}KL
zwQ9cm%=e4ujg8}9_VhnCjOQte3x~EXodoSykCNMe8^iCf0D=P%-(TymkOYf)E^98a
z2`&&GcTiV*DZbGP_gI&!#e4=g>|e<(_ZpocJx948aMyEr3Qfm4TT|P&&mSPyy98bg
z*Z2Q`9G)6#436Tk9RLuHh!KvMmz=@HoZ<T?f)w%vFcePE<xb!hPVmQ0Kp0%TDxOsV
zy$jx%muCXE7+=F8--xHn(-5cAd$azzBntUQ{mXgP9PxhyZncJe4sxwGM<<nHeVTve
zKN0O8l4TO6DU+d;YFn4clgo6mN~)HPs(NXajnOdc>8D}cWp&a!oWKfuQ=p&PQ!QJo
zmA2(oq1mM81-AS)ubZY8xi<)l%IYz}Ho+m#xH^VE1iDhiKR_85*(l|mq)5Bej?ujL
z(e_exp+ZiqW0p6oF&!giycCc8Rc{(7u5lM|TPY_%l_*^OUfB#2amr$~Nz_xd?o!G3
z-qK>A_Mo54TqCNrlx62axzqp`t%Cb$g;pcXUsJXSIS;-T2)rhmjaXRIh!)nE@KhMN
zRslD(AcRL*xbBJrast0O8-Vr&Iiy{VO!D%jwSQ*vtkYDBulOfFuuJ0%Vhx*(Z;r!Y
zxq-SH?%nN@?<sKtF6)`M<9*JFAfNyT!Pl=!=~J~l-=3w0m7u$(;t1rS7<-YLQ2^;)
zrQ*|S%;cAT?df5*{>lS1sTi{&<?0+xDSI(<_AKHz`a&Hw+ZFbLaq_fP{d5J8(k<!y
z7lGO9bB!n~NK_Ld)gr8?Il&7Ao<TV6&)#WbXcTQ2ZD?p_Xm|!Z2MfD}n8g%N6g-Pc
zIX*Q#Ha<AnJJ=Z*L?sX?RG0+>&713shKY@d>CeJ+|Dj=IXlP`Ls%~hQ1D|YSY_tYe
zf{OVI2;Uk;!ZYV`;WYr#6?f!z+2b%t?04I+Wqgw4D4LjkQ6deJGifUBA*HCOa~cVe
z1TnD%Qr)~Cjti{GG+efeOG1Ljwm|WSz5wrcC`LR0cq5=NxPi=tp(80*&_|Y#>k{c-
zvdj@-%On+xS3F^6jA7<Fj;Q`C@9PCoA8Ml4D8oH6@x7Ad5yb+&VqiaWC~<v33!)Ko
zYLR<FKWc_pCxfDJI1+t~^4yhjta3ePsTq_5r<tkL_!^$Hl6?0Got)PzGMXtC^_tzd
zw?SF5nPU9-9a@zp_hBV9jiQ`-NxSc&P87C#%=kj8S_N6PceRpEw>Vyhl-t=j_Cm@Z
z6@zXe5UrGO_YiH?`twdH6Rm9W`uG-lxl8OvOY-x1UG^LHk}$neBswz_?Xeb1y6hQo
z_R^Zt+vaDywK4O?jHy$@%Em+u&y4tbvRqRd?ujMWB)b#q3@aIq*Sr?I&BOPH#LH8w
zjFX+uK>d1>E;9-Ti#JVZf;kQM!hEv%#r3!^_oOS>L;@zoryKuC!qt>M_aObPE!d=d
zMpgfD##M>DW0vrV?%2dnOsaPlX?VlN+(Vb)QHD+sEE)(Vd3f{AT%*Wm8S>FTzbGS^
zrQl8K$(VAe?rh4&k+za#8|HlnT)CpLoEx8Y%-5)f87GjjCTw4GJ#bwPo+s1xGc71G
zdwsc*z?_j`4o$BSXzNivMvb|`$egP@uw@Qi(^sARlc6RJ+deYUoXKepY1LPm%p0`7
z`mwnpnw)ES$9b2=UVmrwcw_t(`WhDJ=@(`n6B8ZXj!KP*N|uRi_WU9grzcaVH;MU3
zsd467zd)o}q1UKaslBT;Z_ypV=NUD)NDJ^;t~aP9k^dSHiZSW(0Esy08C(ZqcV2mF
zh%#Xg2E`h4!a9sQ3cVx=7sZlE-niK8$7)M!XwI<Py39%8X{Gc?ru7)(yC5HOZ-|(1
zW~g)~uyPjAu)~08CExFaXro7PQW;LM=i$2G6AtGcxAThSX$}Fs8sFc@pzx$r@J1PH
zw9DB`;*A?4jo{tS;XMbor8(YEx$E@N)^}Vz5POc5P1lihT0JDgpk(O4y^EfpQqiMd
zm(7}>`_oPsbC~tq4wKHIiyyOFHbLXi=F7Z!n1NLrDCV*rHp)|5(0ML=$fu#X<sn8J
z4M8u<?<!_i_MAjRPSAmj9@XX;GwSWVG~bYy_xSL;{5v4;3v`1kD9kfwYRSzNO#>MG
z=K9?nIamY|Q5cd1fmjprihhW!RmTzGs!vZk2$&9^awdIGu|$$;)2!4+k1{z!5N2po
zXD(Y}{PcA9V)fI(Vw&&%%7^9tisN}}TNi-tD<(uZ*Q}^g{O9VQMOhw_g5UAZqu2M4
zc_ELs39z~m;+7E2gD~!}F|&3NQC6WqRROnk0l9S{`&j|~Ss}kRA$kKNUc)R3+)jUp
zFlDTxC?T>EWpEK?_#S1DWvVEwbRp|_0quApFLePibs@923E<6;I7>UzXn+6<(Qqvf
zDl-XgH3@_ZDN+|w*#1CBAC!bZIB~m#E|3v5G^86d*#8&kUx5hbpoZ4`h|J*Jm59xi
zkbxK6vDWDI)_9yoOhJJ+mQ*mUU-W#m<8QW;;I@%Ym?4LnQAWQKc7F#UMUF_)q@V?6
z=8U3f27^xa>Ha>Vp+y1H(u`HYw9DHR6EB2lHzGpc^_w)LZZDw8(y7B8-Tg@3{>zJ9
z47FCsw^l$`W(3r6M+I&S#$CwbzQYnb!s{w10c$QWZALJ?FI;=*ZEr}7aF;0MnJOiF
zgf>2IyqeyVHZ^W|_&4}>thzl7%|)Q*d@@T}+Lxs%zZsiCXPK3n!F)tEXJ$;>!iEin
zN}2)Mf{?^R6Mg2pX&vy?<7<tudaB#TwEf7+<^|Fdwok)nUDpV|ZGK}<L)Vkrd)+6m
zPTb#2Bb$Ld9wGTilq1g^NQ&Xgiv_oEuKk?0X>f|!)x8Txcb@kgk!hvlU)d8=``#On
z9BX!`&F(zydYfuj?j8v~y?hsW&I&#Ux6h`}px(l~MtR10;CWYhTe}v-E@JFd>GwxX
z6vz71Y>OQ14Ow~gX5#t^^ULRTs~Ki|0{bvv@AbG#R=Ev^d1HUHh-J}NxGt34JG%z;
z9UNLbR^Bi7pE#bx-s}BF{LSl0f6<=j>dOB`Ipgf6P4m?f=Y~4fYUb|9>6-4_HnZAo
zc3yAI-5|SS_bBh!=6lGd<qAk4qvw-hPdU7h$)c6@`|z$WB`^Ee9;n)5K(SNp@@FXv
zNm-yPYhW@k*2?8sE%%$(iNF=SRuw#%rn{B`L#j6YLMpyfci<welcDQhmpJG($+gI}
z-?iQ~-Zknqse|pEm;(c$nfdrdD%a7<bSk*{pG0MLtID=L{zO)0Dt;>ZN^50L<=uM<
z245eesq7wKyJ;)kQakD<u22%cP|x4}M`I(ML}Y7KmW#eB)i%<8G}F?od?Y*z@qddi
z=G@gf$~M=oET0>uWbe|(0u34sta|PoWX>^emmCJ>5{SxBKQ|rnT5<?ZFl?U?KLT2x
z1+*-Nu&w4wwBrgf&3&Uepxv0BSo<L-wkgvK^w#XmnC0O{cxfbg;*|8<^ex<g_+lmf
zVi^bg^yOKN=~>OE+15zQY*^+Yy;<dO_m_-h51~lpvv{a{2W5tb{Dzk{hZp(8mtuyO
z*2R}ubwFZkht)p}UFmFn-QsHSWld6VoBXr>UQBw6clB}hcCQQ(agL5j?x~hIV$-`o
zcjm;?<OE%}VR~@kEOQcYw1GZzLI^mc<++gM-GX!;Sa<GVClerVjMHxP%j!yG6ELxG
zC~pk^{s`rIr2@%};lVWBm@w%aozk0}*n~a8KicO%I@S;W3qW9u9l;v=cY1?KGS_*u
ztr_%fNcNSB_Nt(M)T0%*T~GZzMBUlT&Fe7y>G6B$+;*(mR-9pbH^H`#U?p{VCFglH
zC}A~B;WfGIg~jKdV)zy{<`&$93pRoiD#J5b%rkw_E6Hu1@1LFQv}G9f5VWS(?@o0g
znw4>jwsey)O(XRZG<z|ttg>?xxg}Y<S#`U4_pb88H_@0TMVrT@ezZbKvC7Eks$bF7
zp+#!SOw^@w)W!AG<#)qV(N!hmRYg@*WoK1|SyiQLRmJV~hksz96~)bz&CQh5%~aeE
z73IOGGlElfynaH>Rv<dEL#_gbRuSTIK2#z?Q)gjY#$;2L8L}x6w`y#!j2Ny;sq+pL
zw=CUb6^lzNAkV5Q%PL;VDsR><n)cSvYgE;1lHF_)=4;aSYm^6R(hq4A3TcuONvkk=
zQ{87ZXjqiKWL0`(6MAHof@2e*YZTdMlc~$9X4fu>*DiSSHt1@U?s^P*I}>@F$9`NE
zhnqErI~H-@<7++!aX$cD91>m}W9IB|`i|Xl4ne&{n>Z!kavGPs2NyfW7CYsTIHtll
z#>6;AT<`H`>;Yx$B4z9|Y3x944~W>Ci`yH6;~NqA_c`VnQGI2ay<nst8fe7_7+d)d
z?R{j;et^i_CB*JprSF)r?>*c0XWI^~+Ya#9PITCg=-5tKTuf733?E<gVO*T*sJPzL
z&~s<GO?oy2Ygufi;QXEa&A?8il;I}F;i}T&CL?fau{o=^xy;8m!@{@Vf~4}Gn(!!^
z@LOFSs@ild`T!$Hft%|mBZmQy2B-nd0c-#%fG!{lpa?JpGy^DqvWeD!KtSk^0`{8|
z01G4q00*-9fx-`*0N}u&09fGm9~=N+J#+LD`Q!%~5yM6}X%E09(Ua+DP-*_0R4Q{O
zDP{>$^1=zs)rvmBOIlg#I;I!0=+TE8CdW0e46|HMo!mE|1lk}U*u9|d!KA#r%kKOM
zAcT0m>y#qQSyH=}>a6E<vHcC_F?X6}sw4m&!+qx26W6|U?VvLr-F9b)ajTBvT6la<
zg-NZ1w~n`m=TaY)GL;i4yTTCqDQnovz>wP^p3-SGAX3mrL(NBR;wol>WBD*nc(-)+
zEA?C=pvkm*XR?}&BLhMjqe2xFb6-eHL?o8_jt4#0ke3NG2m8?YMnP6oOH~T?k{A28
zQi@R=TMi1}c5X)*3>pPn>njB5na#n^DggmN0FVIapX@5w_cs8+GpDf&%FtveiSgr2
zs@YbinM|hMR%*#lGlFic9)6<P*MQ~8b2#t)CMPc>6AIRY5Htdc5foVx7>bnWfFIJ7
zC=rM^wAXx02p(7XX@cpUu7l`30O!sk<sxy4JoZD)?b*k>-j;CtI{-xCx^3ZXS?{N)
z>8a602FoB0{+6bw0F6u3L0u3|Q?#=@@m5wm&qoa<1L7BN9*~%PN+pDR`-rCibH@J)
zcUW|!a9qwwphyO9>APlc)iYKp%*UnMsGjR`GU{l>@h5>^J%llqUVwooBF)i4^%d5P
zobh<sS(dxw$-2Rl3&z_P0>TEhT#=d3E}Ru0;|x24i3x@uXPtd>tt&5ZcLsXjMfLqn
zz*K=Xj$J_8{7%&euW%c>jsod@9=VJn>E@oIYSDXMxd3`kx;nceoNOK-(T>q=OH?L)
zm#tY}RAgUfF>qyY9bJyXcNA5&uLEx|tMj7ebNys;S?9>h#&Qd#P(-pbD$)Kt{dwFy
z78zhVa{zi^{JmcAEq>A?1qperg*qv$Xjy8<LV|wyh4a?~T<qHoa)t_E4O)Q)$wbsU
z2;e@+iXjda#KIsLEZ=d1td5AD1jJB$G%jBnSFynPHIb@Fg>bASVfaW{bU;cL=zaYi
z=AA{!J!#BenPo9rqNT-kfNjQvrRy8Eb-1q8T;g-}e3_(MU8#+6xLC;AjLWxj&2Adl
zWZ4{;*+x2X5cO5F4DAd2(gy88ESRbj40v=7fMU;sq;KO$gRD?}&`1Du74?{d{=yPO
z@N1aquH4a=GEjtKgb2ghtu6m#BZrg1RgbV_Ulf8cTAA8{g=fHXxwsk)UeNP77EN`3
zVLP#4Ut6Z=*7nSE^LV*Q#NqBtd16NXlw~CrJdY^&gD$-;62s^zAMIHg!M3u2Ef$dq
zs;eTLDwj%0bJBVshKni3vpXf@Uc!>jE&SGT4BojF#J9;3@f+eidA4?I>*^`iDF5*n
zJ)<{ez5pvW=5*O+VP9>{YT@4&8kVAx9*ynjFYLcgaM8Cf0_6f31@y%N4Fx)BU$EZ>
zCkd_wjekf5*N$NyR^j%IO;|Na<_+N<R=XeA#Uq<?;!<*-`T&SLHG_yaiclHAffra6
zh;O?_B)WT4qz<N5qmu=wLy%Nh8WAhP8|GfJYAf5Mjs(e5*Elzx?;dVu9#3A-1X!0b
z37X^=BOav?0rM~eK8)KC1Lp)=88FM{62Y}mit)TGIg6?m5MU((Kqsy=xh@$bw>p#(
zsos@C@>Xah6A<+)m*SwT_4Y80>zuCCZzW;n-0(ii6%o^wJ6nGT)q(bvd6R;OWXbA6
zHU}@*D5ms9<}UQV#~A#^b|zkW-liP<?`6U^c*mw3+{r(WUeAwVw(}y!RVI|`402G~
z!<dX>ZnSbnb;9gW`(DD4zRH|%=^SEP7Q*S8sIqE>1Ennm12MAB3QZQl32|zw;6|(o
z0HgT7uc?eu(MJKMKh>ZiY(7{ovhH49c_@D&I`=xb*oak&-3{y1QF&!@iaq(F`~eR9
zZ?LbTE)T@0e+zvoCLBRb%dX(WCCQ(q)JA;9-!&fb({5~Tja;2$jy~lLHYP<ZxDQut
z^oY*xcGAB-!ntia&Z{0wJ8!iEMwW5yPFGd(<r`Zl_AW0^<y62mvias}YpQPBfP=E4
zhvW2;3N>Byii_Ry`<RnrVOuz=sj0DXvKo&JxlQY^HrNT;xY7TD8$X2HMo{YIYlr@F
z+$8>6-<nwP%L(wL7U>rf>c>O8UfbIZ29&NAu}@W)RrC+|1V$-mxqq{lJaMwZ&as7Y
zIB*mWl!cSS@bb{6p;59@{V(LK=RO<?L=H*r{HK>GI*+`g`fXOQ%8Gm!&qbCL!V_+!
zTi5e&*nDpf3`LUOM*M|wMLEt&m}rDI5gqIRufHjhc`vz9_$+w&9h{8}l^N)m79^Qu
z_9X8Iz*sptAE_iCxun<12FX!??!_tylVlVV(G(UE*&is388w4)?BG--Hrb+PQdhlD
z@>fJPQt_&)mUM_Xm|OA<wDm<KL{D3dd`mLA!zW^SAAyYM>m9UmqpXVZf*nU_)&QC5
z$r`JezanX0|1+kRGJi8ZWIfQ!FD=e)NZs<Jx#9H;%v9ju<USBXB+5<M!%&<&)Im*@
z6!f!`&UDjI@rFK|lLzlJQBTbtMI+4f?2V4U+XpsGv<0MySGoB#yyEQ))l58-VGP1?
z5CvUEw;qCP=(~jUE^@*wc!O`qubF>p%<jY48-)@cl-+>B^x*OtV|Up{nT7FGEr8W!
zc!4n6Sksi{(m=8P;AGNe+TOpf?@)idhae>1P{3$8x%4iX!ptND1O0L^`sUTa@u0I9
z^C*~Nn_01I)#ZRuUo%fVIzM2~phL#ZfUBlMekm#Svkc@PL_uF-Rxt}rW%s&P2#|M=
zt@X)cy#pse9eR!Vcf?U~G4uqTie6Sww=W3O6sg5lP@*pc<C~6VRR|1(7JZ~nQlKpa
z(*#+OW?3j)QJRZhIKM_F4&p~z$>be^-QZze==>S?{Bfi4?AMJU;Xg3h{|vMJ>UTxk
zY=gUI7U-@9V-negEmUoZF_LAIW}a#SrzKO`7&zyI`@L}-d3_%?*=d>E^OENk4Pl69
z3~@t|e=}4>+F&f7Y9%KohzESOlqj60+Xcx;+^ebDsH;2e4u#DgpvoC=W(w5_e&u$N
zou!FV6(Yrwt%yF<DePbDd1IhlX#)ec&W6@Brl~PfcVr%OpImlCUS5HJPr4?&buc;H
zA2GVco<E@F+2fkw6`FS&`G&?@^!wcmlKPxK!-Yz(0H2ofEwbxRId9Pl1y~B@AySd|
zvNU|qJso_#B~9ZpWvh}5qM+!l@*GPSAJ+ouL}4?UHt@VvFxf_qXvZ@n8<E6fSetHH
zj{r!<Z&Jx;5D^d@i0~5#yW>%3^apmn#dUj49fy6nhCfqbZLF(*qlqb$Vq2r9&6Q7V
zkJ7la`)qvIxwE}9l+9>9P;Mzx*;760zdasr!<GFcHN^<#{!}6q@^>q*yy86E%I{Gv
zN37&XMV>7_hA_(;EX_APhKON$1-T!BmxPip-Fj0CXPiLTV=ALbkE$L5JsKJXL8@gP
z8M6QCBIS6|G=gReeO^!QE3=7EG03&Jly_N|(rk4_Dl-?2q+XlSX?k0*4>fO`0blCV
zD|nHxk`Xa#%^_sa2_36I!ai-uTPZROtH<Z2N28444uqU9$QzREh(<tMoDkA<Mvj0M
zr9$4iqrdiEwk6MlX*bvFVtVIc&<dDjS>R0Kn>ovJ0@?A_tFl|cacsLRi(Z2+ra)V1
zWW&{5KDchmrYd0h>xA*ZCc2D<n<r9Xn^n<g9tMA`>9a(|laNvs9IrwaD?}$*hvr?D
z1Fc(9fc_Wt%#O1rS=WkW{#;zZdBv*hs5C==Es<jQ)qEMwqOu+Q5+|&#0t}lOW@P3}
zecYq^=29*CC^d@Ltf7?h_bzVbo{MNC)85>y_*)ykp(M7f)GxG`9kT0;*4kNw<W=Cq
zcGIdG-k*A8h=cMH^rXEqe8>Y6qSG~(UI%@gyI{})Owma>%@r9HN%i|<fsTYr&i;T;
zA?nP4|9+C;_5Eze?|jGau|g~;N60Ogt<8M3gvLgY5BLlXa-UTYh|(ZhhH|V3vd{Hi
zn|fICyR!4eVDr}U@kEVe(Jlf)U`u85-APGrMH+wbd=KZ~t$%999-)y3?vfpHdB@ZW
zF*e#g=0*PYa<9^suO@IC+BI*nFTwR83eYHz&S`_#G)y=`7iqw>S3{9^Jnhd(usnEg
zG0pGstSOv=Ns<<(x4+-<rQ7Q0YP?6e*8%q8$uac&<m~&WsVJRJM>S{B@v||ULAq(?
zuQh5*N)<OUQKe}(yni^a+k8N##QSkA0)1}gv?X)V+g`bH=4-L(t402M4k^wV)zs7;
zaZ8nLYgS}5C7oGyFD@9`kOrAx6{Ui7StJf7l~eYJRO5}n*nU`s)EnvPF9EHWSzf&X
zn_UxSQT0kOdSJ?9WBWAxN=<7hjC4F3h3UPDB2SGz3JW>dC?GE%u#rTblSR*<jYy3&
z(tl+7u|3ZO)BFZ68C8}4c;6~oJ)Kl9s6U|BdXm=46Khrle{;4$s-hzXH%^0BMA6oD
z_?4zA*%BGVQ`OqefR-t*w@QDaI@*U?0j$%!x)P{I#Zy;h)3C=__m7sgA(?Ov%j3Si
zZah8O9g4`B_|;o)tiiwVvgF88Dz%($chd8m{KP&VJY9d!cXKT4%7=a>kpeS$(BAlA
zmc#GFDBZ3zCJ46MAa}YR3sIIbk7U`d^=3#nQZu^cY4PC_)3Q;og63fSoCDFuw%y+z
z>~iPZ0-v<yLTVQIH#Yff9h%rIYZt+t=iAE({7-P61|>)5QQfiQuR7h2l|ooH4xr*l
zFJzyy0~3BlhRzcHmT<)aLlMgek3Y^@Ic~_Cjo3J&=BxJvizg8st})s>wy^Jho&1+E
zEFnHzJ~TAa^RZ(xa^&9aGNM};Rdd-egO1lI7Nc#+R=$At;pNte#D`21&+!}E*X~a=
zk(f;D@3I<c><#nN9<4joT`**-kJOT(AruP*ULRrW5V^r+h24=Vo5AfZ@S$wDhJ%hU
zmqCLUGC!MLv!!RmnOHPSa#G#6WcGC=6lpBWRc(WuTq4lW=x`61e8gnM%Y@xz?J+p%
zG*wiwg#t1vK^#b!>hTWfN-j}ze-Wls-|nh4Y{_;8t~N)PW7iA<%V|TH6^Qz#_*RDO
zj@y2<_o<rra2JEGwfU#`By|Oo6)WEM`}*d`)AYjW+O9B!hVEc@qO4QS?GWBN*Bq{W
zs^UiCKpj*K9q?8CjOZFNOSW~GK$%_SXGIV`q*1EC;hn++A(~K(-QE0Cf=FdiFrdT!
z#FB5Xu?T<1H`Wh?i$^mjB+cK)dz4Q`AP<G%Pey)T1mmrh%047Hdd${AUvzbD2$>XU
zPeFynCr!Iw?_=}N+W~<er@cYKQ$R0z#;?GyC}|Myj!2Yg5N*-j*l`6(Als&?78Fd5
zY|$`{H_CSu8qc<OJ5-;Ki*=$pA1yPh(JOD$=^_nFmKs$tv^fPP5g*14t)sS#YBaWt
zW+o%c4IEfJrEq1)(N&y5d&q#xg<q$V1&L-4$oGS%^J&@qzRwH+ejKMO&&aH|>l*$Y
zH!=NC;u;YJcJ4@Kv8dEnX{rSQI4&4EgGZFpb0XzZ(CAfv>b{0wxl1wRjdH;Yg_JC`
zo;@9w3{z-Y7;V}{!)5BucvU4bOq<7o&?%E|t)rHM2P5xr5cgNd&vYDWYCAtr-q|4d
zUAeHl3mpl#?u2)70;9}?6Paa>m;FCpzMcG~>;bJ%pfaG*8Tw1SIW*f9Z5Q(jy<++d
zmW5*4cihdAsJ^}H+(9jiF!JWtj!^d8lW%EH@rx1?3Y8d_cvH|I9GhriWZ~Cn=u>zD
z5#Tw~K_UYd=;+g)`*cm2v$L||87qDhOLQ$)^1SzHe|TLsmY8PUr>2Wtg-p*0$4QWo
z#Bfsnf;C~WXaKu;=M*`FL?bSS-uv0b)^OUYF-p+@43M*>l(kVglFtSx<Z-(XJ*&F-
z+;nw2?`8^VU;>CW#EgS9LwD_i_oiVdam5>y(vz_(eHng>dFFu4GO^GVk^!&lO@XoP
zBkv2wHO4h9xCI+8#WZVK4Fp0-jiEuJFSgX43;0JGz0BM~_>=FyL_NM1+>X6treFKa
zwVp;T-sTj33=r#vd!}FPw<GJYEHg)c?nWBqp6V1(8l_ianz*;R>zIbnpHspHWh4Mu
zqES02oxP7);fB}s`CPSowR0k?jGRG#OP?Xk*S&ysZpd7^Z=-7`J+~RQ!O<L{aXkK(
z@57F^NYPJLJ>Qw8ERBaJQMV|%IFsrFLwr_l>k3L;$EkRc_T_P~&4s@@yEskos96_q
znMY$;ekr@aQO+1~LFs7r0e?Qf8_R5#<i?6uJ%=5kfXBPG&31w~)*fJM3Kzja&TiL>
zbzmJ>c~H_T#8_rMdTlt$8jgsp9JsS4C)@^353I)I@|u@17Tj>Wwb=_t$&om&%xQqU
z#vTfrlm>G@#hcPJYuLjx3s#s#q@!CB2SHQ%MHX0V-HQwRPwf{;kl>)MQ+uJyzqdyp
zZk#|OIk5YrvNB12Nqm|MgXefRe*xb_2+I!GnP(>I=Nda4O7;+(Gz|>U#YuL<MpBjn
zR|dcPT01GZ<EQm*C_2xb53hQHU&KLfnurpWUoLR;^rhVHtMB8nd}iG29A_?GxHv)Q
zm4gY?Cb2HXzm=rm1e8&njK&<-dj&64P23(&+=>?$<0woFrZo-=$F5(1UoCf6Fh(Qj
z<CE86G|(=wJ1?uy>=<-V+7^poz24N0*o(UJ<jqXs7B`3^q^q^nwc#pEnTrFLDYFd+
z-8>YY)F16Uw-NHY%6l|Y2)wslS`0ikFApd)+um}oJKJv%7xTfKNc1o~iV_vu>C>HZ
z>fMya&d6B6r(w!!)FsPeQa8x&ZF=5j%J}8JC*K2_WT*X3r+Io*F}2i3Q5VR6&iyuK
z_rD|YY~z*P2Q#b6(gEyxYzyRw#K}s6wsfjVdSMb{0vO$b@w!6(UVG8O1p1n~2|A&X
zju^G0N`-NMMq_<(1J1`MQQ;vru(WYoEBW96)Q%k$C_y3PEB-<*K;1q`_MQVFgnTEr
zyQ;5uwT*Wc(k(nPOK-PVS5BRs?AaK#$!OmzNqX07ZlV6X4jgP*7Z1gt7;?clr){k`
zB)dhgmpj2mLJa5GQ>X;yKi_O4<x2j}f>h=}E_(`Xc3P`uuE5A~m!KYpVre0%?k;-Q
z0n28^4-F+Tdt);01|}qrTfIX9`zpH#`Wqa2`sr_!>cM*<_bq5Ae<G$s^bXr6vJ<;X
zI#kCVG@@K7s#7O>b&Dn}gRwR!)*o#jyI2qcu2V+Ft_cWUXNM#b=G_}bcz^?aD5gP3
zGN6|iEK=`#g|=(^uTR&N8Y&C_qPL=0;hWxt&c~Dg(vVu0cw}S&2Aii@b!{d|sqWIn
zIor&cwm)}T+XD&m8OuBG#>%&^K0UdJ66f{1MYk}or(@=y*>A;PzW~*3FPOsGWj|Qg
zDwLTur{BO>zmN#*Imxdmw38*NoOdtJjL!mJWSw$dF!KzgDn|BCK(O^E1jXaOX;E)2
z&y+@ZY+|vK&m$hUx1MYz=tCE!mt=IJ5PR_3)Wf@VI!_O8uAOaOi05}G7BKQC)5!)V
zdH3OFe-MRx&zA7^1{Z&q7%S%n&XVL^<6cE|VrLzR=)ZEUZmKVuL%(`zQ7F@y?AE-0
zbL38`h^A0b#`*_Oz(p;y4CR9*vpm5+=w`Eyn+A(+*AKkBuLLUCUY*KjHf%rJ3711J
zmN?jHE?_O`1G7y?m!F$W<9leK)}k-ottN@z{rF9$lDd`5=x#St%ea=|OY;=p3Rb#;
z-EAN=LSJIhm>^(-DBP&m{cA7*=VTfg52}S+3l}Kf>*s=mU%W+A%X3MwwBg-k`XoGO
zou{#ffhjHK?D<UF*N%?Q@3YNv*l|OepDzBpY}TR1#nghErv>4KGEYw5=2K8Zjr^K@
zDTO8Q-2DV)?K(#oXRjSxl9>FRt$hn0!(%oNY}Y6J)-V=ovbDdX727lttxc6&t9(u~
z`ct=V8~1#Z=fp05W%a44I|KzvPc4q`*Mlep3;p)Y)>R(wjDCPV25$DXPQZQXx=v9x
z`?Yioc|c6xY*QUcQ`fdn2}AkB;r-3*i0LrwOiIKW<;YF3c7>NuGkUXfbR&gziL7`i
z(}uxKmXffV-)+zLJM-Nd!Qg4wGJOwZ)gMG~-{yf2caT39X6HisCnoDbzI>5?Hjwl`
zH^IADEkss8w8bD9agN#hIZH`$W}W1ffFIRqU3Qb%pUgtssJi25+~4ot!TD7DPJuN3
zJ=NA;^3?{pZl_u}8gzFsY=^A;J8;Hob)9}_KN&R_NpmuBI9J@TjzegAvS(GtVowlV
zF3pWlVN&5ui_SKtO1_R<5<`xX)N2fxzD9g>dSfqb3!#LOV%1O*cG*C(XUj!>#LHIf
zd&vD|$|ru@N;nh`PbfRJ*k_4!V7}E!jUtL-HoII{wBW+5EAOI3F{2BA>ZApz5;-GD
zpc$h#%ou~GybDP^p?ozGfD?q7pZV9wNSHGew<}~6sRaDM($Yx2%z!e@gE*tanr?9t
z9uMK4YLpC)hKaZoCyj~W%_q#^cX#W2GFWvSx&SU2Gh;w{XfRl4n#_OrqQYym8>(7T
zTtv_&;u_ssquk4L9yz=>bOS+1*UK#A+l`8IZaMwfu|YRBoA`&~sA{cLuC;TQx_E&3
zSB+|=+Chv-8u&o}B}>+s+;qC<CxmXCRxS-6{u$Q+TtXP|=cRfIk8xDThmnR^*W#k}
zc4r~c20GK#a{{P<JL4+Lw&iD5)4JYl(w_gRN0fc9hwupsx+3o>2KX;7u$#@87yQHz
z`84QR05;4&l7_jvyOTU?E#Xv_<!|_Qmbm_)u<<noXga2&k&9zu7<n0A>!5P_Bh}&(
zR+f_n8eGMst25E)Vf*@E5Ppx-vI+5=yr8bv?oB?rch)t{0w2#-9>jU;B?Jf`?Wy5r
zw^NySJp6=#LwAF8-nUy54PRpiO2*de21kj6k${6anFP>{CAyU6kKM=Xu_?IMyc^DI
zXYaL+_l|Weg^z+8i90-Js$p)z1m*cL(wqS`=<nD;6s#Qw0@71~Ek}^?z)s%LK$M4P
zzJ3ES9<C4co@=5%ACE6iv<!l;myN!5QkD99>@n+ghzo7J<ZBr{!)K(`i&jQ%{6l9S
zeRt;v@VhjYF#YU~I(J-&$=05)(P(@A7OZPf8SH1ytwj$afh}I?-^68M&mw&6J8}^-
zoy20Oi%Kfg9XuLw`qayLNQ!$v5++%BsFO0b8j~HN_tL9>53)bp7OvgjyYT;&`{5@b
z)GG8fU|gk54=WbCOd;U1L=)7>!|vFF7=rPS3H7e;Pf(0@^;;Ako2PDb<>i;;Ay;9x
zFIRN-_?<j_{X_4TI1JFvTWY8*$rDJ97&W?8s~9OP2vT-hPumV+PcPX`&=5Yg-R)4^
ze&0D2`+4i|q`#q>#W!quM-;f-rtT;9E!!S*MONpR88N5<vu)P%iU09LblS3~^QrE6
z^1%P4W2pe_ucffwaA<z9jcydw&jH|--qoS<rRo1hxZ(KgE9?rDALi}eNpQYOlw<UX
zMttCZmmdhkQX7n21r4PMv4p1Z5>go}W<sRVb0d7Sq(g$B|GIP3aTUCs8?wzu`Wc`4
zRg3JuED?+qqvdk7q!Yr>>A0`wVHa*N=MIT9EAoZ1R(FnXJ3_x}zSa%fn+?E6b!X^_
zvA@6M5Bhi*nK&P<G3H~1(=>D%iOdWfOlvmbX{cP(72FR%YriOCmWt6*6Q24`holha
z5iW#po%Rp#dVM!`SuyemJ%#W&KEIh3dz&syyGC7VzD~bVHzJ|7BmqQ7xqL!1Sjg<f
zhi10yJx^?Zf880N#UX~lO?cHIz1cyq&xE9olKSmj$@r$8()hhd6P$a%_?5!zkEvU#
zjEReYul1sfp_dg?i$R}C+Fv>GB!_gLvtLL@m~dT2SsR#B_v*Yn(^W<_pGeNBwcVOj
zsVeB{kfOzNyV$YM8bo*W9WD=HYkvRT%y#5US+kpwu`^IRvlSAQQ93i#mV$Zo=(S&^
zB4?>oYch({Qkl!lr|C#c<xb!C^J;2@7|Q#MsHYKJP@pxSvmO)(!8a0jl7C_lb<qNu
zX-zsuas_JgTvt<Q?I0h3v19S-5fGSfM$Ef#PMI96(YSJIYsPP@axmCdlW1<wanaZs
ze;Fz}tQHd4d$fsp#CgF@HWfc(!;<AzJ(#!YG@P#V46;G+148Nsd;A7FTZSYeHD5)S
zH;xTnvF|~x(GU^?j|sZm@Q*Ro_r3M2%5FBW-Hvyg64HB~Y5y3C)CPhp<U}Ifc6gwz
zEqO!?$-V+f@qym<xfQ$p=g{fX@iOPWbnU0OoL^W=zYyk1c<R>Bp3|u(&$U+x|Jk}i
zjA=seA0v!<<yK#tw7WIH`N4?om54OkR0(35ZOCIB55|~@O;_RJ_*kqQ!a`(4Zo+3(
zv!!ZR(dE3}Ok)GLNZ!T1d4JMI-qn|Mem2ep`&Q<1Jt5BXTSy{hX`$$Pb*tEc$toLH
z$VP&-gAWHbd(ugRv$Ar5bfC6%6&2pRCF6Nk^TI>}wh63NWW9kGm|=oM2abtKydP}4
z-OJ&;0tQoCGO;;`nDH#^EoV-j=uyo$!zv+)*b=-P8a8m(?m_9}etSy_nD@;3*Li@s
zO=49mI8wK}rev}$A&LfS9zfW8e^YqXA@purwbx-(0!Besi2Nl%)O8rmLiA{aU#G;h
zth#dP>6!}r^tJ_fJ6m}uF_0~%RxxGxP%}j4F075=pC2&2Q{BxS(1Ud)Dr<p&9k72H
z7+NXVa0%Q(_VFxDsl>@4Ij#BR?LDWkz~Rq@<iRt2Tbx;*iaMw!8h`_M1Da_%1dwo`
zbaph&C>-mR!dzXNG$&1?dQiCC49tGXTe9hdpT^hNV05bPDP8wMD1}7HqA2rHq1iBq
zF>GNzeCyoQ*t$a;aV+3|ViR$dO<|z*I`!nY-h4S=Q$;s}Z&^$*D0^G2b4WAT2H@PL
z-{&QXg>2~WDb#P%S`qG8o^slna*FrB>EU_tn_uJfAnYv^bOHu~0?h$It=cU;8|6YA
zT>N`cK{9zhS}52dFWwQKHDBT|o3kWgB-GlV={E_1sMX&pl5_K-2fdYBx=m(zK)$pD
zQCgP1Wvj^QFY}Yk?Wik&L)RO1)fZbe9Wx-n&CJ2=n^>AN4xy3IXGEJbz{<+ObtZdM
zdn>7aIoUDupSnJ%UA>LU%#6BSoh_g<gqcg2e?GdV`?$Y&d9igdF==qsK&AAb;riyj
z=?`&WKBgng>iL3?l8B7HBdd-h4gmR5mfP&322~4%dZyN%;Mo%5myAcf>UgpFFnxb4
zp3oU0FDTca!K)Qvn8GqzBJ{GVHQ1wRSI%>m4Ci1U^7S55o&`KZS4B`1V7h$bB^kK~
z#?O>ny{_Z(a$suGT1NFGU;O-;l_&*tC#K-5`zP4;JV%UYFTRaTVKiS)Z~iu3WVi3=
zOOSUSUkBs#>6|4qaeK8XslUcg%#$1E&cRYHLF2rdR&GuY?ZUyv9h9AQUAqhZ&-8B>
zuytBHlXUHK$=0Txnz7yd@ISop24flwT<hnu#--y6<|+A(v6mAgr+KYHCxZxH0&4N~
zblMJDH%Nj1SSk_acCf}2cZdBvLY|FnRkqHj9HNubWytAYQ_3yalYU*^cHUOJWXHq-
zA0EN=l5~TrFoLR_h?D+YzHamqCt~iy?2j%Du9UDrS$9tymne|aI{FKeZ9oN6gV%5a
z%>)I46@maPg^3jDVS7%>PdVlQr^h__G5v`j%ZS!eX0tNRzhPSbm^z--rhyWzc{r=k
zHlLd2yDHe9-@m>!lf8EN%<H=R7Fc03*ZkwArRSetK2)q@f$3nm?Y+5h@TjxC-l8wS
z<+i!>Bx`7CQBTq4LP;oK%hgWDyksO35hC}L=(B=nT?ZPX=ZdW_=ALZO*C~iw7S?$K
zF=$M1ogO-b+GC+!H$cMN6qp6MF}wey*L!9x2eQ~-4tRjn{ny_Fnvr!GCl2cT8d!Fq
zgLZ{7beqlRIzZV!hfvTRayX=dDU7zeWP@VQHUu<-kl_if;X{{}l(VtR0Zp<gW8~G!
zaxaB%Xa#)^pj`Ke7L4GwIylvd#{o|P6qhy+$wK&I5t$>y!Md$n;zb^b%E%Z~&LrPt
zPVxJedAN<pC#>U)duCmzSLa?djzk3a6}IZNN5tl}4je@K#NLVQQ6<!<32a<)Xk_wW
zZdZ}yBYrMqluB{Nt^ZM$wK0j>=6QDWK}irRwLs%Of|5M;N#e*iJ_Ql&MOe>9ZjA`*
z6|tGg|Kk)D#*3w*&)S%k5?(?kwV%(dFE?Yu<Y4eT-u}9J6@3K<l6H1#<wo{8*kji4
zEsm?9tMsxg<G$tcZ(8QBz%^0!Xb+&vXgUX>NTbbk*E*S22R=&YCeqcCX;Z<K@lG=t
zpI?k%rD#nULk;N)e%$f%<5aKJZ8emp=L*<m&U$yrlgbO8zLR?zDG~i(gX>f+`52V%
zZd@8a+<wVbY_2Pt{Xmajp|y7YZ>@!`$<3-##3B|MkGI~JtpvH<owse-Q~J!!j)nMD
z%u?1Q9eWkYwxIbaLz^#4l*{zE-j^F$yPuMMAt$2dvO@`i>?>88s3qZ}HnE<o`M*4@
zPT)dC06nvd<9c(5XIyiHbD{eikX=W#?J;lP*c$mOSC;ge`T>xuCa9HWv>nTAGtofP
zU2__kCo^t#om-z!jzWmgDZV^uzhlzn6X^7>P`!4Csn7wePRm#YRoNK-g+xU)Q=UfP
z`CJyEmw&(q(MvRPj!wsF32g*j7fR128ke2CYwuB8BCig<Bmt24{C%!-Ux&1#kbjta
zu#MadPB_NciY5~ZmSpUst<Ma#VWH}0Qi*BlFCYH|eBv2=zj+L$_!xMZT@NA9tQ(?k
z#j!UjZ(H(L#@K1nm{$h15vq+r$1h}2uXYti>!i}0=({sPKsPr4t!Eb1-Z|?y&58I}
z#S~8cxgfh-IqR44F%a@z>K{oRmBoB6rYJr@u%e3Up}ZfbypL7kK2;WEDh60Ps`>7V
zpqT(=|IXHK%f?Pu3-lQMtD3bi6l)Cr5T##nNyawA;O({TH#8MKVSk6;^+h>`_jo7J
z-Ew`>eXWO!hH(lq1#nfGRQA6xER?N&M-S#KxT^}r6nL0Bt`cbZ?y&gDBFUoWMJ6Gj
z(w93S_cIJ)a)C4i0Cx<1Li|(OzSUObY%RRy_0~|6PFi2Y45uD#*C$D3Wu{+_oIUkN
zu>KZHH!Bn-PouxGId#?=Ft09l_xMzme;r|nebY|)cR1#dNKg?_oCu*cjX;8=IvM7|
zIJa1FB!Q`)^jA!NB{Nl)_6Ww}pT7H(?bj;Fzy9w@17*rcb>^ONez_*&d~$_pjCfoj
z@BTQMYEQJzp1fD>bFXl;orL3=xlnKi#Tjd>Ksc2d(JM-s%Wa3bc4$u&8>3x}gAa|L
z+&JXN_hPYx7VoWiow{w{bC!%Jz#0NO^^P^QQZZh*h(POmHlg+#^7A901mY$1=@5*d
zu_vK@O~t{fX;O@A*@cZv77QLU&iWsT&a5LsN%(zhX;oYAjf(XiZPOO;l|RPgYf&_e
z<10-5l=0(x4SIIM1~A8)?IR5Ah)4ggkI*wsyy_Fpq4ZplG9a)8q@A0;s=yHvO1*MQ
zQ6Wbp$74|>5?`mINrz?ymnnkl0mvZzaQ!@i-VTWTlU4<Ij&o28GAz-KTjB4flKU^{
zR=TG;qV4gK$-nnP9gqY;+NT>j(%*xYZoTyXN#J5>5*VRu<|dK1>A27~U7LXPkUmzX
zJQUtUsJ?%ZXb`L_2w38@KrR+ZZvfdb(_EBtZi`i!nJfFdfPJ)t`8M6yCHs7?ZR<Z9
zE`}_peWz0fRGaramVZ#a+@dFkZ2LDcx?l&qt=j~^xH$v?DMhi&NYo$>eS%YU*SYo-
zE=0wZIRqjx8paej+b$qTS{Q}<KHeDJF_g1r^z6qhH|WACe`2c(B%M>B#tCCRn+0VL
ziJMF=qgT8`tT=Xl^;bXRm-!U*D~#bCJk|3l0+E8k<!*m1J_`m^ZGe#VQ-uCYDIeo1
zK;r)9jN54^R>Unz7n|y&J;YRB$huuv_F=B^BzWnZ-86U8THg?VOT<R=q?c^ZC>E3X
zt$x+<H1+J9x?WAp!7+;^-4!3yQXPol$(G*pv4;`ie;c#1>7lXZ$FVB%gvzh;axu0&
zARv9;Wej)I1s#roKcQ0MdZIL*>x?s64xLXbaN>6*xoeS{j(zaxS|$lcQ}S8wxxSb^
z9qPkH2V>r(+PBdA*T6*ik>PsdElo9{>^s@xBhvRG+2v1{xV+RY2YjNKE@8}qNlv=E
z7%sEHSWG?SbAr0*Y~@^osursayh4#U%Q0u(z=u@=FGurLQpg@x7SH7#WG#>6=QVoy
zXm#&BurRVo8g`gG5t$9IrP`1Wa9fJZ5zY-i>lzS5Q-4MunJ$~&!o9dQTd6A$ts6*;
zyrn_^S&eXqG2B#VrWYNF&E-`QO-_ifwXe0gpR-?ayE&(>uWS2yqnnXO!@QL3&8+OV
z@AF1Fol6&qh~J!_MF&SJ+*{2T;b(8piEk6V0ph)?0<>{>BzIvNkH+F`?mL=R2X3_|
zfH2ImlEDQfLEJ^Lhm`JdO<C!JFz>8&7x|Hh!54;8uq|llFr71@^?!&u=ip4DHj77-
zOgyn|+nP+wiEZ09Cbn(cwrx9aY`-x!yZdcz)vd1nqpPdBy6g76=Q-#1%t;1mvkO>Z
zTL7O(!Cu{jDovS_a|jS2M~?5^7_>AP5u2M4(p@v~$aO;oj;^9LSl+ff^anhjjBoJ2
z#2ThdKE+N9Azn{@`6Aj{i{%Ie{zBQi<Il4_<j=EI-eaHL5=;zE2k1-&&-GUUAj@xW
zxAM3H;^l`4-6{90_07lsG&hM4x8RmpR8OlZ!Y`}k__}29v1Ht~bzahi;E6r5w(ntO
z*gOq~OL`AIdf3v*n-(KSy)_TDOiv3BTvpXpFTPgo;rTz`TfaHB`(=dlU?woreDW<l
z1(?14&ImrJvdQ<vytq1kT5hslF&0$V9BPm**P$aD;!?|A^25xeCE{!7O>4`K@%NIE
zg7a6gAg(CPkUK;3AY&r6{uexM9=7sZ(t455C{EQ*6}mcNBIi!c!sI4UYe`FW1#aa?
zH+D?>?eVg|;-<A#&8_tY!+vB)o>_{T!6(j(6ICtbmCDP@+$wT92qH>Vqfm$GKEg>e
zZ4K_i;~_>Pj?LIvfKR$)@crS;Q2cl@_Pl6f5k9n!+~mSo!YK^ZZ(z!q@z(y@StNWw
zWvBPbYG%`~g{-NYk7G&c@OhGLM$|RBrKgVt1!-XLmMO;f`M2##wxBL`{<?!_j`c^J
zCTtJE#rY*pA;D}<-XUTr`mM!g8&Y*I^!ec%WNJz5=!Yf^IO>dva$>kea=7`wGz4Kc
zdg5YeK>W_+wTc=7bs?OmL_=V%IkTCcI+)@gLzVA+Y%sS0xw4~J^d#OzhE3)I<i)i3
z7X8GLe@$aEq)-aM%8vcr0p9EWNIB%**pZ$xV6I$6lbT=7Ia0b2xDxN&?$6$wUHBdN
z?_DrMC)XG0u4$R1kYgAjT`jXMg1!QZB<R6`c$*<_2A&}S^excyzyev%0HBv`GQ5^$
z#JrizjieJ-RHstZ%<kvfu3PI&aR|!<j6hP4VW*n+*Oi)+;oUi>T>jMTS-y{(fxjHD
zhEp$gZ)Q73@ELES8v`##sXg+96H>J{_()6^0S-;`Ud+Z)&P*-wg@vm3GEhoz&b)-b
zfZEM}btjCqBfaNIqqe1)6DYQX!I7(nr5Ze1voqXgZ*tXn>aM5;9G80*1h^q${^FCy
z%=AzQFn{%W0pj07QBfja@Q5+{P7XIutluCw#`cl+vU_E`!paTQjVdU|+3K76>IO#_
zWCE#Y?@%C^jFP+&rEf;t8a8;@rAFtVv*tFwu@Qh)Ur)!sl?^Mo-h2FY+XgixJT?zk
zrHjhAW~Z=?e~odq$KbD)tx{;L`Q((XBizzil~W21D4$cVQ4`94I<Y`}d=if<LFWl2
z*9L0EnfdZ;*U}orv~@J!Jr}8TKc19L4_{H5Hr}<Vj2*(${4(XyE&7Faam5w05#-Tc
z6LD2HrKAiY8Pt|)-dGsml@Ka3`+Jtwo!QMkpqSd9>W3gA2^tPPP!`dqiF9~<a^#T!
zb9F!6hQOA;_G7gVe4*-6GmIM7GQxtU!!Tocsk5~D(fRSM1$PObPhn}UovOf7bj!}@
zBUurc5}NX&*%WgEv(gf5kzIE7tJE=t&?cYhO?6)`WxZ(p$R;hLn|WJ)w@_Om3PYOT
zxKMkj$sp@b77AyXX?$s4g&HIwyYAMD7=GLPM2mOjnIVVo%bgfL)iXjCzc4+W*DnUQ
zLY_cF#w#XFR~$Pbfk^K^&ybL^PI5g4o9p@=!)`729T7X=G%TXk_!`mX`%6l-^uy%k
z2kQRg%o%9Ysls$-NM?uHrZuYP>a@Q$0{S62TPgjdoKhF$!3xZ~sA~MNnGd=IGpUgB
zCEcg6w6X&p!l0Q2>)(0}WAF&A4e^$<ovziSj*UE*s`Fy1HCx<I3asnYHp&$D_R16o
z@d+%ZV$PRdPvxb52wlAc5S5ouw!_Z;49)n64$3vMVI!aNo`HA`gXVzhwPFD_iFs>R
ziFz?unHmSLvY4;N`^lyDu%`|CDWFwT?W!;ZXNUPt&f8-Dh2>ey>{tGo9?zwO``5bR
z2K%1B-HSU_<UxJHNFM-0<edUIEHiYH+b_aDG?$+JH{qo)fbknm6NHI%p+KI^1;r4=
zG#x3Wzp}io&7#%B!0F2iD3`QdUV&X6b|4eN3TfSYwuD!2Lr{rbU8NK<nVcSa2<r@)
zGgDT@>67p!#j$<wJTc}j&-+NfAWm%JbPAKzo=UUaQul~_YwS=bN2a2Lo*gEuHnt$+
z`FfhZ*5~dd-3qM^F`?IKQt2X3%n8+~0=gN!u5boGWX)<+P}*Nt>yv<K@$B4;?T?q<
zZ)q<wh~+@RSMyT|Eu6<1Z-{e*QZe52`+XE@>k`o&z+su1p?X@*K3E}%*(&LE#e$u6
z7VX&xigIHoWifohGRe6KgRf(`JdN2|j3rjlEScN50?~R>;*g8t@kp0Q<i!gzHO(a=
z&qB>Yo5J)^|Ne`2Hqf93yH<7VvX)YqSc8$bIbHMw$;Pl6(0pjY=U5?1AH!pemU({u
zTpwcR1W|vxrd?%vwc-P}BRD|HCaXaTPNt*OIBeB0=yU2%`u=vu*>k<7^x*-V7I+6V
z(Y4Xl+~SNOSp$H_VadZyewl1E281HA&u|XsHhCk2L3kLZM;Rud`F@OG{nCEZV}h$(
zdPYp$2gh!M(rW1HzXy>Wr1K&rpcw@tbGabp#?A8iks{063!TVl(?FY4Ys8U%$sPRs
zEyD|`Mn&ENv(;v3TIgL`y_CV84oAP!-hE_LP_u^4o_?<V8MP7OsxFG{ebRP?g7IX1
zy>9*@n#*S5kor))EqAS^{gKFIRnM{9;X(2;w)u<*GZP#07GN#7(*l7_L>PK*?7=gl
zBrR$?wTbTUI<r&TAx~OYzXjh|ry`f}OA7dVRk&z7@mZ;$)|uEF%Se+V;b01;y<*o%
zRWKuDO-4Z?v%ase$P58kFYJL;uyPV?scqXrow2MX85*8BTi8&3wS6`df-RII02o73
zAUWa{ZgBUwU)XtFUa+Hi=Ws8qwZ~bLe<s#Z`U|_z`Hyo`2SB|Uom5LUZ)`J#*<Jrj
z-FP%#0ka6CV6r&%{wH?J4a0_e@_B3OA0lFGx7moi2E1hPfdwA+j6gbg$8XQo>|V&U
z{$nj(aI&9!#j4iH{93>4-Ely*p|)5;>ncP`|A0{ji`9a5XUc7k^K4JzGxexl6(Q|T
z%)hzepzlauuC)`q7Sfw$qZZ|YCKhnAU3W(GiTB>sm?>Q!Mlr`!e>0WE?{HK^NDh97
z(q`9C=WmirBCR!Vhaa2DBm-5@3o6H5Cym!Q<KXAfc(DJ)!RDrsOK#p>vS#&^NljJ+
zX+oSE@2C6S`*mPh;C%>UF_@`K$wCKgLPK2gZL&D?fsM?k{<Ti=2AhLt5faM``}@?L
zzPd)@UMP#aeGz(NdTHrA3qcTR@nx=iO=@YVgo2M3cLpDx_r-TE#_nrp=ff$HG#W`2
zfYup}rb7xXn8q<k?1fmCa;aCHmR6j~aezDNp4nj*#@gQ%H?5z%1u{|>N1cBFw()F0
zq(%KwhU_bt!zvdO$6~RF6qb1&KC893v=|n7*rp#h_)3;(4Hu2g^u{+XRw<hcuBOLY
z2G?aKO7A#a=^ugX`{W7OnK0NjZsoh5q}%FfZ|MhK%oVgaUyoZ)T_T%VR{14>rXNGN
z>;2qcP|0>2yh8AKs^*(X3Y_e${i!XS>E%?m-TE6SDtA%>&}Yyh_J9HsJvf}<DGezw
z=@}^V`FL|h(bdFfhLjhY-BV**SNpsEK84x^T-%|TB09d4c@4Pz2A4|1E>|6RTMRh6
z^S4NiwV-a6(k*=g`PH)pyHaGg722yfFN+E5pt{gfwz^>e)~q&kSI-B6=qvW62Wy9?
z8VD~nk4{*ib&o8%TJE;+he^X){7goC{24tUU1;?j29N^)2;9UdqG}VF>4oN53ceCm
z0Z?s-kgJ0<U(Wg7fjXbBD(-uXI&*?WTr5e4e$D}t7)0o^w6u7HVClMUsub^U`w#qz
z-327Jn*!dzmh${Kp`DlMJhQ?xO(Qp{r)0$F>W~?r;elkzzF1)3wQKt?GNM_6%3Og}
z^W9f)%4R8*S#<iCJ-%;SVFJYppnG;|vj9qUxct7A(2&uHE{lOyhHyF-J-dMz?lA4z
zi8J^fh6|b=(0s4ZZG(<;T~hS>v@k%NgNhom)<(fu3S29QS&o9u*Mn_mgVlTAlMQ$e
zALcdCS8md9oH2b>wO0Nrq#733m9^oxw$%)?va|I#yAV>tZaoT733$DSP1_3WcE24N
zIjZie+hR|<<eYFTgFk4|iG5KtBU^?s$7Z7e6iljjQZ_M_H6@R|GZ_(jQqn0Y0(#O#
z7h_=`Iy7MZh$@It4U8KG*$E8}3elW#CM`NBH>Ph$l($NTOJXO0^X%#vQ<8(^U7@vB
z4Q+a+@_@m9h|YnxAk#zyyT)H^rq)?f*tT}u5Ez*{^KvI+KjK}5{dHMi`gVuy$eh9E
zRLM|y8T#vgQDu>E&7YZPuPZPW*@g={DCUJ$@m;fUHvY75Y9!mLQnb!e!WYzLgK0bh
zN9KvLAD_3l%;VEY=+FFVhfWt0nGMlMm^hoRD2hFw9=4FKYsc(OAo(&~V47d{FP&K;
zD1A+B(+pa#TOB>i*xi1A(Q}(-nZRNc9y@w*HNAK-=`nVzlq+aVj%Nwm`J<2K-dkqb
zVn(RN>oC*e6K3J4aAsLYwUnG(wQwvb9-fQ#v(Vwo-_lVtoryf{IA{Q`S;q33A>|Ze
z94?Ep9^6}}x2bFotbjb397Y?Il~+>*hl<y6fu)e}Dm%lov^I{HnInXYz&zU}^9qr$
z9T!!?c0yFHCWY9bd<Hz{jlA<+WRrJ`gzdL8erxNXaSwC~%$N^}Ge+)1O@6SrJio2*
z{3<pV69+>co}kD*^&pB3+7w>)cZDEZY(-kVj4fWO$N3}eAXJU=mjY8FCkc!I;}47J
zO*N)jpg`!^?_6uC3fr&GaF`fNsTom4fG@x+`p-^>`A&{0R4dh+PB9B=!S3|j0&l8T
zUZCd(%vNr;$(g>xkGyDG&Tirsr&1<1)Rq8oZU_%<i&F+xkhysWwmU_-uOE_W3UrV<
zED-~S>KrWbzvB-2{|!e@Ur3B&*5m~Hu6RR#4pDG)PH@@TJ(uOY9(keIb=?h>@hP6~
zzFu4zexyv#v!r`znVO+$^>;<Ey2p(Wz67#YK1iwYS!Sp(9Dep{>&9Iksf?sUHabOc
zBB%;z;Q-b{3Ohnx*mE^vj3)oF`lAHKRmhOblA<^Ap*J}NCL&_0h$dx*A90j~?C$6y
z5C|~&PE9ae%MKZoZLR2dUVeUl6n@ozyfFlIV1_jrPzz6D!F!dVJu5`~`sZ$sL;n`f
z7k@Qs-;m>IjZichUdpX|Al^>KzFx%}p-k4o!fk-Lp)fI%FFx8ll1w%X2yoKcTEy<F
zj96T({Fw}HEq_%d=tI2qq3GzS_@NkR*f!a_nhaDNQMWmLk><`yGhH51yk5M<KE+~D
zIw=_`Mp)6I3b5M`Z_=R^M$U%F)&xLe+lBWGu)O|z+9#A`?Ik}=$A>bu`DN@;z>Kd`
zq<fk|XiS+DIawJQm;Gp1#IGvq@s0yaa3Bv|MAWIt%_Fvg*PU*Jk*!6za+lyc3Vxq)
zk8@Ph^9<>pTdV7?gLHp7{J9)}A4cKq1yYigSkRzS!^;zN{nVy1-|1ZxuS6^FQd@XO
zC?x+Vu_Q-=&L4&gdhxGt9zoF<>90Y4R_8(cd9|jW550I-Mo4v+$6R;k<g}nN*AU2*
ztYTUD3xof#Ba2%UTn&x?xlyYuG{%=mu<F<iZ+62t09)Cw6%~lOQ$bg@Lzv=9GZDcs
zdPwsX5hdo*rk#ldB(wz@lAB5=FpH0nRH>SL3yk#pFb_0`5JeoKNVyMTWqbYx_VfhL
z269E&ctj(e4BUsm3s{XMD0LrdN$Ad$MBF8~%NRmUR@we~05QCmh7x<KBn;<pD%by2
zS5#5$u^%;%E>`ORZ&p^Ko1hH4&|Mvcn#!tmFgP?TqUfb8Q)G(5f2d~ve8cC3XWt``
z-{qs&SJh?47f1vIvkhx%D}3s^zyG;=K16#!&3zaw(_h*Nk~%p6#lolZ-=P{PfO!r*
zt6=8Ux%0_-_t8FmOZk#Kr}?ZdIN-}k28<y;do%M60UopzC0EeM?!)>ljoh@MwI~Lh
zB{_#|MaH}=0(~$~y~=m*$GXluo5+AT2X{8c?%X-ll1GX-lO>#!C8e<#{Qeth3cT_e
zI#>;N%ck_nKg^S}aPuG9T1EkjC#;!=8KSLmgVTnMc6=znWRNJa*m=bt?kbt_1L28^
z)fb!jrs?@k$W7IpJ(M6QRb}0M8Vh;nugQCq1u2DGN*=?t4#{z5YcHbASB}ilKr+KS
zAuJm;!^tJGLxSWHM$#C49YMt*u0wQ{FYVTUD)_O0IOOLnX5MnYnE5vOzT?EZ5AK(b
z_W37<9>M>9bRqeinCUVY{*uJ&p1;8*W-`=y$h(HW6PWTeF^4HR8W{U$Bt))D4v;&Z
zYP=;?u$+^vQz)xCAlnDSH-(51n14`fvu*ZOF$c%1!MmCx%D*TdP>??MqQqB3*siY{
zJW%T3jMlK5oV}p(@&@=`^ueRy%;3sN6tqM0k+4r|lQW^LfdJj>Mww*7pZ!Qt3_F?g
zDiZSsk*lP{Ub-20+Dtt?dlt&JN^_>0j^WQ{hs&%nGy-K^fy>mEwF?VehDw%iQ$8ge
zxD7}RJH9Qt99(Vo<Q;UoF2}C{+s-RFxMrh&n|Z98qkkJVmr(E|ggL)~!_yPEo3i7>
z3^gzYJ5|OamNEJ&6xn4j5c4HyECt-GgSF2m_sW3ORH2b_9>qZM|1(uHy|t>1ofUa1
zQgKtj2x-6s`7HFxMu(lL-v2wf9qmd}o(Uu*<pzZJ6|LVF{sIT(6`p$5NS|~_qMY%F
z+&HWbq3QQ?w_urXx%+h?gj2Dk5wVQK?zxe%r><Q%Y|O;Iu|eznMt_TNfdnKoH7p4k
zVsDbXd}Mc@TzY&hoR-rpX;3N+KD%y3of)RtgFVWWR2@u#!1%&DiVnU*fG_UoUqAxw
z2DzgiKjsG*Mz9W8|8{W|WW&DhPFmNHr$2;--#+kLXc32?3+i0#?Q-kPP9o@|n@trD
z8$DcTQ8X63B`v%4arGF9KAqMHJbFE63)Y!Sk$xq<yd-^_p0?kjGo9Xy4~yFrd7D-{
z`rUX5Ydxnm6PBI%>`!oWKTH9N=#qzbf6_GiSx!gRPW{y6m2f{i;Q|^_f(lP~=Ip1B
zN2n=wDk+A>?Xs&=vDe#$?z5`{tGXVQdYR*6nHZHJTHaSbMSMO_9deg+0o(Ng@N(U1
znq6vI2KCO_N}2ku-MigR-sR!HOgivVFq3|QHclqm(WGn4^Xlem*E~2LO3;V9DHacN
z>^B-*a0km$;IAuJ()Q-uXtDBaQANnpuR^#!MkV+T(Eg~J98gr8q^@|lTZBlfp{O&G
z?BW}HF;`V>UE^rw*=TU>-H6Y&nz1$mgMpbDYIZ9jnOSLCwACDigEQJi#^BLf@RU7R
z$-)Fahde#GT;e|z6Q3PK%ej&-#n&ojIHyJ22YPRvK)?`Kea<y(VeOEBML{U`+ecNK
zI1tG9Q|)LL@n-ED+9kp(DOdD<X;<r#tmKR|9C6{Q9!Af*+}*9mKVp7QEAz#rUq9Ns
z$n>4!KtC^vj}@QtE=%O66GHJnw`e{*CS>ikv|$!+8@a4DsxqdLq}S)COsE}Ytl|(_
zZT*#-g7EFGHRF>l@cwcym2mE5n<P9`JFj<%Qqf>5W}1wWm6`)hhNv?@fA$_<?kpv3
ztsPT&88tHS8>BIk27E_xnn>xFAByJ#m1N<J(iijBIUGuR5ILKD-E`@-GVq|=$!YaV
z#<tBW0mud7<qd%Y>M~=i4!9L)hRg-8<-Z`pDr9yG=%cEnNytc~dus9x!+OruZaZKN
zGRL)-w7w~9Oi3N(ZJSHOBhO%XhB)`HY!fsgZG-(oC9h{tI+n&zf-FxKCOl=*G3W+P
zPg1#k7fw%>IlX^k*}TNNoCu{@4-5?UbqoL&^(kKI@R1+S(mp$WV%Rl5a8q}Fao_~X
zeFyOwXCO^%UEN@DK*Pa3wz;!Pnyj+X6Yav!!3^qE<oc)NbfhLYI`jM%UPgD*BjRcC
z59`})f<7^fnOT#vX-->2TGi=ya%umbo7zqJGPz64%uw12R!!BnM|EGM_AL!Ct6=7w
zltX&pXwkPSdD?BX#CO=L4v{+I0@&4fA3!^%v+kQtGIeWA#V+!D<d_m0-goz|u7Y^3
zX}LPxkW-3hs(E`cogz}|ddhn+0;#v|_yM%6rO|u-S>Dw?3y!s_0_QhXGkyHMf0=Sp
z5a>g0bFU4(c7E&&3iJEGA?(W#F>ce7Wk$QYyASV30T@>X_OA6h&tGVBXuTO~bTt)v
z{0HJ<XvRqhjt^<HBV!0a`1#2{qmR<NQMB3jHwP<eZlkFSa`&(vD{gMlYz8I`e@oSA
zY$DCyc{}-Y)0{oK;#>5=%fnVrcB@l!;|3xRb+c0560~XPwk9AX^jjr+2jI9G%Vfvo
zDg>$i2rt$-bhGm+qv~Gl?{z>wxY&`u_IV(CWrU%*K*evq_75><v!Ad^&~XnR(r|EG
z+fy6Nb0`s3jlZl(-WwA(jIsnyqO6g+M}jt&qDFb*6{$I-yt8%G*4{J))j8)}rBv>q
zy$dSodhO@B9qcu{z*Bo7F-cmsYNY<qvQpxrqMyb0;Tyk?=hNbbZO-1$&qjDk$JWAf
zNYvK#&m_kG?amcr@|5H~t;Xs-m>>7nQu{R6;R~NquQ}&<3iV;f^x0;~C54Z1JDbIS
zfTKa3QZvTE3FJt;hsS0JYYe|!eOt&$z$-Y<7N88jf0oUq@tR@IY5^uzTMVyt{%FM%
z>Fxz^BjixOtK&j-<177;NUlDFT;Hb3oq;yH&dBmt9X<~C>$!#F72pG(nM52QYwr#U
z8ZUF*xexU1x^C(6r6|{3b9@yG!Z7o^tyB*@E9p&<{}Pa*X2M3F4f9Or#{NM&dKf&U
znwJ^JI_Ja2$oa6+`-Y?3`(_3Qel%yDt3n=ilhEH4;M9Cs;g=M{Fh8q#$o1Q}Lw5XL
zBjQa08-J1T5OokPP~(NcTG}B7EpA?_AD@w4o$h+5{IE-{EVH-8gjr(KvB*i#Rcq?%
zb|#DKo@*}<QSC4PPP2gI<b6=c?x%Nib!Evpj<%|>e{n4us$DrpVn3)YJYfaV_WF;{
zD5dQ2Wb)w=;o+y-0<0So^z@*oz2dsWfnwR3$O+GL_i-|>mGXJrG-}FzjwEU3U+QKI
znh17aDX<VaXeaL@H5H49z-ZepAF;rd;-0JBr57LhE}G2_*XkR_({2Sz7dG^6%I;Th
z_L!9c-aou*OrL7-uo2pVh{oeyYwo+oSkkQ{@x=WVg7|OTlxw}*iXpcnm{e0ZHk6N$
zSba}DLPKdmKgc(~^Ywd#FQh3ZW~RTp+k)?)7F#sY9$8ojbRorF<|YrqkU5hm#wS?Z
z4Sw(<+VRyu9fp<W<CkdlG#B~=YHe#PY$E_wYwmPnS(K7QI89nKKD$#<I!bX|;6B7B
z>kIXdvaOGhXR_-bIL8y#m;?jU)?weh7IMSiB9_Mf%5=hI>`!hUB=-P8{;6Z;-znmJ
zp^Gw;K#BJ%5NEuzikkfWZ3*7O2*HN1hbiF!_2|?Rz4|(QV^1FvvorP(QSRvcDe6Pf
zAVe`2zM_|Uk%`COM=(|s47qHcQLejW0jnENfJIL=eWEcpz8s!pI58qgSIZT3gBvAc
zEmBb?n=Iu-#=mLI&9(EYr7dB*C^hD_bVGvp)Mez+4dg|UvYMm~?1p^jv{c_{CRL;%
zwso9-M28|S^PKU+xJ8GR^$CXB3Rfkti3+RM<{FgMdyvDwZl7T2W@fFm*9-Nz`8*DH
z{3nds2>$`)B{FW70bo)P-Qc385gRw-q38w!s=lk5e_#yCXc5FnF49eXdxerGX<xg7
zr_qXIQa-N&FcrPAcgL9BXqHCQWvE`OBufw92%8vxZS}KYDv}!4!Jnz4dejb{=cj7F
zkHzQ4a;z26471i|f`e^s+{`H}9K4|79DGERZe53h{%h+UG;oR_A?F3d@Nj2SaLk4;
z*B{%O<-IgSa3)g%q`)zVHir?l_1sG%kiFsRx_WfUwvLiByk}V8HI_<i1Av<Bf32ZQ
zaXd9wY(K4MclIjR%grAtDCV_YU5idfYJ_gB*zu{oHqM`3ZQJ)B7jOpr<N4styQCS{
z9ri<6sg}1OG+m_twO8c75eA%7EZ(2tdl%+?Xmxus+}$A=s&`Nk-f7`IQjvyTP(MW3
zcrl7uqu+nl?A0FriZv`)9C>7Ug`m)4deGh;2)JZDwtr?s`7O(E$nFEPiq}vsavBZ8
z*3<aYb1t?S3U)8H4qQs3)zlMgdvi4Waex2^&_7l2YcY1$r+|h-Xk4MNn)6#iAV}Jv
zDrCv7ijwpsY1$XdI@fMUbE7{=cLok$Tz49MFwMyA3qSs@upEZ>O9%<ypk|<AmxQE_
zuNv6)o3Q3kUR1&4E9)(;0`#_%`_^npgJ$^yd1xX$Y46--+cddl8LLI<|H@xU^?<sg
z97O2EX7HX|E{+y_sG7M|CwT;T7b$2h-h{8^r7<eW)Ls80U{X0x1h2CI)!;R+jvFUt
z+tdpWM$0Z7x?oo%mv#)F7^yPu47Q)#aJUvSI=Z@q;<vlhn^~P*O?>hG655RB(n9Eb
z>y2@GYa#F)@i%Ix!V|ce>c=gRY9zBk^+g<fl%MpP%%5AHG*>-3EaXMcjdX~Rm7#MW
zkA+wqL8c@%I6)Y4C?cbT^X>(1ns#b19QPI5{EatOo$j76oF0#^nYyVAg0%HzKR;Vb
zm$kjLFCsQCX)vy}{HlJSAK_LMlaeZ2+n7(;VNAX_y=wL6Wvsc{Tt9btvbx;dJZaKs
zT0hFWD8|d~4PCMVYHBrUNa?QZ3d<;e&F=ON!Lo|so(QsjUcd!Et(rHiME+%9$)k?R
z8)*D1jW#z(YARQC@uT=qwvO#}@02l0O4`{nd5Y#+XU4Cb^d^d94hM~pn7es87d<}W
z{|`yws^_Z+{v~be{(h<&XL@?(+J_<oWtF0%9L$*=@RT}+OCtSC*11`PTUB<x?T*}z
zFa{}A=zBy>R$Ikgj-z+o@m}qP(b(pBqM9nsg+ycF$^&XC+41l)TezOnQZOBhS*Zrr
z4R4b1P1x%RFALXrDDoHO51OoOZU2w1beRuNfbU>x801TgfA)zK=0lFH&xsTCqByba
zZ#?KHBiApV-NNMVqYhp2UfRi<aID&U52AYSA$RsOsXJw*=0C#2c!pf|@nJ>M;pr!0
zpej8}HzUkB+PeD2O=2a8fVcIKjh=mO{?E|vwd7lkt3NPLJkoXjn7M4j{}OsLI9JC)
zjPqndq0qydKAVfn$%|ODZ84xF&`k>nlp2>Zk4pnYi(zIIlcFwSZ&L$rXKVOnadj=Q
zbTL;v*?!v&Vs<F_^nNE3go#zT;cIGp=;u$)u-6>?v#Fgo6@#jN!EZ{%8AlQk7U8H2
z$Wl{HuMfaVT3v8bEkms@1%)MHwrZW7Tg5exMFx<Ymf`X%iY8Qcx`?>{!}0qoXJND>
zZbu%XtCU<~DnR3Okc{jmk3sKG;{-@f0?LWYGnj#XF3JF8i(Qe({#SUwbA+s+_OIZ=
zy}g-GK3_>*w2hsR@~ovfPkDqYyOi>WO3D4Fh}D2A`xLF!h(G2EVOT_=yaV2v(eEj_
z8#8h9iTtdNmCfM4LqFh%jC-Q5pmlJQuhV?j)kOx(Oqf7SN$gRe&nc>J_z7%@7&&<<
z&}kIxI9N4S4P#ew=@w5zv`e#vYQaz-R}xB#g|g7|Ov%RNu?I?I!$Yc~?4MOqeuEvt
zRFmFs*N&59gRN!cQ|dUR;=?fZugR*v@eZL87-Fi1zhn5RUC?}#2m30R5g?N|6?I7l
zLY6TToFrbXnWM!>arf264DoXfs~3$`KW&t_*w3J`_*KBJ;{p3H&*VTCEPnO7&xX6V
zdz$T<JRq8HOWpzBx$(0Tw3m3WZ9ZEbS;}EjC6xJYw1KE%^I!P~C~BUyri;YxJ^mk%
zn@V=emZ1_*$_*gd-_Y^$gB^=QL(lUQ6$CytwHdJ74@}9U<OAjgO{sXuy>9Z>8%D1C
zv5xCcBfw4u1%lV5<=$SOFX!_>;NEus%9zIq#U<i2*&RK5$jhf7)XgkqmZ1*h3JqXp
z^oE~LvxHlZUVl2CS0ff|^G&s1l;W|;F-q@C)`Bi}nt1t!;@4TKkA;7l3Fyb-AK(Jz
zNs0gcygq_jWmfV1F@h^?^;H$`RUxY$^*qb}vm{;@P@scWw22F5Y5!(%dTSm#(dTxl
zEa+LHx#DfPUq%~3nzoKDaIenQw?A#=evbX+e}MNiAU!aL;eltqnjUYRNZEZD=ymOQ
z7<KK+G@;P5GC*D74?bVP-025G&BzRA4UWX}Th&*f6vJq;1jfzthP=s|;P0CyyWq{r
zL#p#e*?uNI+<2Y>(EN`*eN_Tz#SsG@YUB9DJpoa?)4_)#f|eF^E1d_a<&l}I;Lxrf
zdb<?=gRS_hgGqpemD75JJ?|RJ%J+E)MZm+}a-YR>gun4|zlROS`HN=E^K_@-rX_m5
zB4@5bQm&UE1@J+3V`}843ynK0o}>y;d;UT8?$YK}UC4{qF@N_x-8)_0Im)8aSeKN$
zCcVy#9fnAa>r(Y_R2bY_qaSIC3+wKlGoD0LX<1gQ=|H%tZe{DGXNQ-tDnje18!9c7
zoh(W5LODX}e}na?vGW=DZg5U}lrb2eXx+<iz`{^{B<IJyf$ZiJ9|?PVVGR`7e0A&V
zP2AfHFwZ9AGap`fp=h`CF}l!?r%!BZIUYQtkvNo*(I7Q7LS6W~3&wVM+EDM`7tL{<
z1afUazaw~OKBBYW7^vj4v9M7kIkwM`OKp&+Wl6Om6tQN`%SU$&W*OL|hmTsEotzaz
zQNa;l+r>XEaq)nc`3xA#TywK>3+!sOS`({2Tgv_q8KIVld)xdPC-RCbcGdAdIV79Q
z#Y$CVsM@g9R!FzfDH>u87YVBxGjlI~y)^(gq%=n}@47T6<a>p#l|3C~;NGju6q709
zR-bAdD3Y)K87$3h+%Y4fi^K}0bh~9L6`7r)A9qJt%NzSIr-(jD!LZCHeMQURy&B;1
z>rfudf*Q6TTZjZyQ6ZNUn9Ww$X>ylGka=~=HN`L9c*9JEyUE1Svacsd`h+hK?FO1A
z$*S1Bqe1*+1L6jMsGj$Y+3Ne>SKPqd*_gdy>{bo&sgJcy)~6r6D7TG@oag`?e|$(@
zbd;qR&{Fu6Y~xL+xUKAQVG?&=AUo_Vvr_g@+&)MO0%to`h>swUf>JyXncNC*$Elo-
zVBho=aDV-LQw0ba08pSta}aPT7jf0Hkg5XqSxYW?XT8`;F5xK=B^$MM^Mw*`1M1(!
z#)XW|Pa7#9)qVOnHxv8dwNWdyw=}tr8CG^mmpNW4t?`0P|K^}$z$({`IZeD5qdWFi
z&m4^QU{o6zrj&H&Qto(Tfe0$k+~ML4-~;v_-Ujb}+);2J>5B%!2Y~sg1RryGfgOFC
z*ks`3I|}>(gp@{IdlGj8JO~3MBmGJ~apu*n|J32$@BF$d=`n3|`gzuL@9~ysr-A?7
zD}wPW>InlM4Evs$({2hepHS&+L;;;gd-9KBx_YQce?yOj!K7kA@t|U$`k(T)8BsRp
zA+ZsxQwf@{fEzeIR%3h*YtpirV9iDh_9{)+*ELg#hp1v986#6bh<C_qcI*q-I2d2!
z$fZ{|6X!Vjr8Ilot{(}@h`iEkkQTmKf&y7h$Z)@r`3odKTUwuq%1$X6m?v#a)zFkA
zvS)wq2rjke$DcXbU%5}Q&_J8bN`dlJ`-h}{+!FKk0fj9ks`~BLw*N24ckplQ9K-Ss
z2LYG72ICo@U0WBeHE!7!QfwQO1`q`&yHc(^zC)if^Ie6g-oe#5@c+gdRV&PJ^?opp
zcuRwZcqnaEeE+6=81|#6`<mZ7S82KtHv6X>06!h*aiNYaMXtx{-xbG+%90#cb{(dg
zC1{>e5=eL-H1)`x{{B4u?NJYK<xTGIb2JY=C}I3UCqqQX?Rz?4L2nlMR}ClHj?-_w
zEr=FM*pPN`<d2@=xI23zoy<><`bk}`W9JlYVuLxgn=s9RBe2|EYd%rcf%1M!GZV`m
zSdo3<g1e>u%ucCtBL;9p7$h1(JJg;@s*Ir-QQ8~&NuVO4fB6&oy0?t(u(CWao7nOQ
zj=^wRl0q59ncOr#FidLXgmKMDhx;)6&cb65bc08sm5~%Wq9!g2GkD8hX0*)?a)?b>
ziMJLg4DfaTKR(L43nhNH^^9t^U>JN{VaL0{{C@~3e(70aho6WT$B-{jKV*cyKy|<5
zL>C4`(p3`0p1IoI?Id2F1Ye}}jbCJxT(Y&8pzgxkRGABm1cGJh9aE<8z7GQ-2W?bq
zIyYXCf=B~7mHkMK>Ea$Wd89d0nkHx*3QYMP(5i?eDgBm)oO|&QHpZgpdO9{oa&oMd
z&+rOv1m@i@#Tho_L?+s0X_$CfZWnzF0<xDhP-8*ho(wyleveR2224ULs|cbaXaQiK
zk#N@uyS>LOnKLXqGU$Q(MW1+Tf8&=14xJ!<MT4KhBELKZ5iEajd1Ky>lDl1;!w?{j
zly7#Wtm%0C_}&_|j(dZjwSfw9lcAZ0;#~nDPZJmOwP`yVUu}H%&oVy3vsUA`K8?a;
zOwO2h9v17eKiyW^pX~;n{zF1hJhh_F&RA{wI8&^+20p1i46<8GkqyLQYbP;q?Fu+`
zHdrs?MK-mFT37%>il4N!m*_VfQJ?u}_Bj{;c*A8$=YmQ+MLQ@&VazF^I)*7kvb>3w
zm1P+z>X1nkf-7UE#dZ~agZpW#k{iSo;b8D_WyaWT!?Yr8KQMzMG+(w!UuHIV!6i!A
zx~Rajrb8O6`^?@RpT3ZUM``e@e`V~Qb8nP%Y_UHzn~`6@l1B5n-0BOaQPk<I#=y#+
zwlNyNTqX_6c$`HEj`Ne;MA9TVEPI-0&C*F6H|6fq92OcbL}zxMZhPt?gEMSN^#}Qq
z0s?2UIBz!v*&!RX6CZh`jEYP!GrE!0%N(saS<#M(&uE<qdkKWmNwV6)<g43T5(F?j
zB~i0`TSNXBZTHJE_hxfQK-*1eMa2AaoZC0pKV2Aur`ot3D2XXr-%uPG9QtAEVz}3v
z<oZs9d_O`QFKt`oMW|d)e`L(ZuS|?{7~#QtX<rWE)1k#rF>s69DuTdJO>*biW~HPt
z^S|F`{`cvp<V#olcB-yTi77P2?L&s74`l4W7gA#G7n5awc8JELoLt$V1@^T}n}0kg
zt>1Meh?zKcE@2UL^{H}dQq2wc|EJ5;*HtYc)t#oFNx1}zbU7Qi)`LQ9l<D!$FQUHe
zbag>{X)YxRe4<-78%fnyde<4PCA+T0S-a~RQrJJnHmPUEra~D0T>A+j4sT)!EtOZ~
z#N89@qR~DZ#2ouakFqhVQ2>(P&CGv(tkGDqWFZ-0mibaTWc15(y6fpw;%_|4?h~T;
z^{J%K)28Q4*YT!f&vE8w>hn{Xb{N~h7N}}a$>v4vJiLi}P(|@n#mZ9TwJvqG5^xHq
za3&ZwS~+Ci^jb2*c=TTfJebS8mt*J5s-UtzzZ8_p0W<>0y}C!T2fAmPsl5qQ69q`E
z`LW%Uldo0kqe@DXwQjeUr>4q2;#Cy;xuPyXk3na9dS}^syL!>cyX=r}hJ4IRR`3^l
z#Y5^&+3R;R4vrAp#JfiP0bfLq!Bh*|xJk;tGYOLpCvQf>yb=N>4Es;y3f+*$rmDm5
zg@+}H4mLeZagWt~FUv`DI<Ifvsd(38cOPBRmxOUg^ka7ffr&sZLZzIO>D+{5fvQAA
zqr?Jl#Ep$9w_mmQWXOI;w*TnF$wmFEQn5?D2zx*lTmKaiZL%g%0WkuGfZVg1yddk#
z1>J(*cQXR(M>F$+ZE<{=L=aO3seKUsv+VBO<kIyfHh*7s9HU^?-}cm;yaI7?xLp7h
z#v2OF>Oyq!k|J;SPyKMcewO~bPh8<JHD7`QU6q7bKOnL<&gR0hq}%0c#z20q?x{L)
zcqXe|;Y_9}mI6M`V=L9t3D;t3lC69WPp!4qX~kGg-1;~45{wyc1JP{~j&3^=LPU}B
zF@9g&Iho<{AD(fQUwX%;Vtq;tCi7$t;CxQ;Sh3O^BXgm=VIWPRCnT5AE@`3np#xN6
zq^jjd0$+;n>SuGRx@fBJC%LT(6xF<(SsuoQXK4OwF6&HUQwW?%Gn4WKVT_v~Kj6#x
z_1<`B&E&v(+thfW+bv-i0OF1JU35`OFx)<#J(=5eL8L7#*j<Kh_HK47v$NHGH+uk{
z&jBo#^aOBl36{2(7`&mRoK}mqJ~klG5`n(s3?*rZ=ZYb^$?5f}dcDznNypbzs`A-K
z`iw@r%P4wPFRK}Z{r3_%{5#X7ZaHXa86XA|-6y(S9umQ#kX>%@B$jB0nR+QYx;cvB
z1sB3TLpm!>Yl(qdp+k><K~0N&45Y5~$Q_sHxl%0aCu6nz>635(s~`fse_j(|mf&_N
z!C9BeIx4ANilVehstl<ruE8$_Ep52{HM^N9AM@W%$VgU7`WN*E9<0vRc#**=`0XdQ
z*VcID&ZEKgWD=t!@?5P=G;=!}-+D{KB5jYu`>ZPc^<r8zvt)O3(^kYf3HZj_t82`%
z7VkXE_6o-$8#m{lR3DaC*1lD@(Ds`ju?4(1DJ!CUiCsZmzm}c_5u3}9Z#LnsnMh$N
zh`c^rB89$d)7(AcHa>zn@e1vDhJ30p;A-OoAvmKHEe7*>x!Ph7iVF?AjC0~_6Mu8~
zrM75;MN@Q_+0M$(t0X=YSY2Fv-aQyzT)PT2m%9e?7F=CIcr3OyHa4vZvhF*yZN8Y=
zlL;tnsi5~l4LIeKJD7DlSayxWDM)Z(*9%65(%kh$=Vg#h68_D<3>aNklfc56j8Yq7
z)t_K>bK!8rO%(n6YqXE6s&L{DUs>%kRD;sMM?r{!(E71wMabilx>3aU8Oqu9Xm>U@
zutzr1c{LWik3%4S>2W+5<~n%gA5mRUA3sDNYgPz{bOAn3na2b3{A#wW9z4ttc`MlN
zmzE5}DQ6KaFn>aPqq>LhPYiEBv}dkNRgdXrI}b0qhI>;P_#+4ztCA9nO*G+ynA9#Z
zf_6@vS^pm^9Z}t%$l?6y6J3u4sBd_<zV6_nmwUZsIgPt}1Jxd^Sa>=jsb|Tz!Dp^K
zbxf}u&Uatr?ivRmc}JuIJh7~i-vRHXUH|8ClLr^a)+)cm0hgKv+djreQ57qnfemeT
zUg;IBMwPlt+T}lHK$b8h8CWqgXW1udituT~gJ_OdZ~tzCM#V7Yz|t$x-=c9Et9BY%
zVsO}AFqhP4e5q|6A>dKlUSHJZTxaWU^_pJ>o^>ny)s(!PTxVviJi6FMcEItP54e$Y
zG+5GebdB3)ja6%pdl-Y^WDxyYzts3z_qwOh?w$Va__CrvRnao1)foM6;in0Za&2W|
z4s9g^-cfZwqwwvMZ!g&K+WRYeQRrVz9DXmX6f1vF(y7(CxI0_k0=T?drd3Nuifoi&
zEjL#PwtKkNxxz9(hW4A56X(+aJf0v1mLNv812O^z>98}CX7+l6;!q9c!XB~yx-*L<
z;ugPgjmulUv2~W=lJ#`b`RJef_xr#9Iz0mU<=QNn6&#NzR%M^wJr&|0=*`Xr$LwM8
zm3i3K(3Z*rhmkMFLyIi((&ABa>2f`S#?A?Cx;^WL<PA`3vf73&ZjwEqEG@~afLj&F
zh=NwD#s!|qtm@u}zj4-{B=r+{QUbY?%1~%|A=7(nEUbS`q8~LZCh3g$;ea*NaE=;)
zj1)bRlHxjS6l4+Pu*(qSEN9qsuNCGbJN3N46EzuWo~atxX`cOEjmfN;+68K6CS|zk
zYp{LRnQQ3Zv=#h8I!qzXe$EZ8l|aIX6Ok0R9mm1k^sgNVPct<c{ffWJ`8?h96nKH&
zsK~7$sobB2dxbq+^3UW;6RV+DI9;L{G>rwpe#ba3>4uT(iO8PQ&t!YH680F)mboM0
z^Jaw>mnZU;lliOhu`ojgqx&#r8z%6Du6!%kvX5y1Zs2R27vOb8%<T-lRuuh5`VL=T
zS}h=Jdw`hk!N^UXu`O}Afen@$);pD6T<DsmNVgp)4M~h_>O8U3lL)RyxXW6oA0$1F
z$TJ7&T5l#@FNA^hz*Nf8Iaro5K}T-H(m_5f<*+!eNNgMBXAl^F%m|-xEnDy(8KR1$
z&LC`SNSQv<u^F#}V;uA-Ev|Y>g?lfVf6~9ay@P!9cEZu#Am7n|_xERi@!#P;)9;MP
zu<j9X6Mo(G63&H*+R0W0MwuLC&foYAN051!R>{T&wr$Py6E*S)emRN`VJJeG;HRm3
zs?8UD8#1~BfkIXgaf{b}3#d{mpZHVjrE}JcyEw2fR5FH!VI(lK0SQU39YTBVc;g1x
z;Rw8;YVTyD&Mtx59P_}#v{zxNx5>$#;Bx$z=>Z^H-mXBlXUw}XTOO0YcVkkqXjiSX
z*gkx<tjpg9a7;7QU!S^#9u@>I*P8?qy4EN?KWiMX<<rz9hAm0*hN!h%uvz9-8O%Ca
zWxd;=Vof;%{2NW+NO2~63bBiu6TUns=1Gg2yQqHR;;j2K4^c=jK*&EtPKIpUcP<{S
z>jfy78#a5|;@)4o%`wwkA46=UM)3Y{=6+F99N@O?`P`ga2cYNF7`Hw$^i)O;QM<+)
zpE+h1#{~z+k<Uz*+fix(36Cs)z~NsZ@%{m^!~14jQiU3`r$b;=`@g?2noQH`X&Qir
zVi{B>j^PPM+UH`{{&+D9#5LSFMT!(d;^xBsDXrDz8{2ZLi6~-WLEL5F_2nZ`!s=+*
zNUnt9t@cjMoIq-2dM?liG>{8=aSvnzAE+?bZx>X5IFIweahE6`#NZN-5uGoz&3WOZ
z-~k>B^sVA5%vHj{1A_I5doqPr5Ce7dqeRY+mC%1COsKlvyf)XR!h%iE>j&$6p_7*6
zh9ynLUO_A1rUkW(!tQTtjLH1?07>O%iH;2_aXO+1Pt9Sbnawz9Gt2R<sIrv5o;8dK
z6WXHND&p~aTXwznI|1dPCBegs8=lLi+bV#c6F#Oy#0D``lvz(7R3oJ8A0e6rpTXWV
zQyi&Z7KU9Go|q#vn?Li`fc8%U1LA-;PgP<{zq2N_?A1UHBds>*NB-in#*`;^OO`aN
zLl#h3+1ZU7Vd(MFI<~%uV1sAl)P^5`S7})?BelJ-xFzL)?<7Axf?TH1^7r?iNV&%O
zX~hAhVez=s2r)ShudTYCS_!f_&Xk&#1P2Ss!kihUHYY4MZ9V2d11p2+<Y=L3U@t95
z9IQMx(<!|p>q6Q#M-TQ&P`=4~U=y>ttOzj2(mJ=s4-H#Ra}K-`-cqnXoU>!a=I35p
zBgB3%Bw;K3?^QaYn8m<e3bnV5F4@x)2rt*Gr?@az0B$9!C1YF^TkX*k{J7jW+Gu{1
zK8)<bg$0~1OcXVM@$BKDQl<Df^;z#sS75Eap5)hw+FJZ;6(z^;eiY06=_wRVe1(RF
z3K<Q3m8F?Ysi#OE%Ax@dwf!_cX-<{dz_jhen04DxqNC=t;O$r8r+4zA=M~aCNAe4m
zxl(>eVN1YE^azN9-bn{7Jn@PIhpPisdVznSmFXB<YRDjaEng-f$D95~5SK3oT@bJ@
z_`J?I+NO6rH41Bfu86|`?g#b!rCR^|)z3ECy&|a^S?^<u+ugW)<{IFw`P1n1K8E+#
zCP~?%_mf0Vo7K>qH3@z+aQL%g_%mVjI4ac8o4w|;<af0?eB}xFs*fHE1Wbmw3?a&a
ziGYtE0^Jxhq&>>7(QD>pW!yWW;wV9cWGW_WIIzo?m<Z9DQ}z|%0Og@es9<zF9ag8@
zu2yQyGA37DmiCic%)cLw<xOj)9BiyTS+8ko;5Dx4*wP+`j+QXpWT-=<>!(ATLmT=%
zS!o@q?!e=j8iat^US438{W?30lO3RCDNO|rt(~Zja-$Ty?0^dS&x2KKxe4$@*wi=Q
zb7inPIyzbY>nxaNS;u`+IpJpz&v}<}1O;u^t@HVMhGa&1iK`m+K3m+a2wp-)MFLaH
zHergY>(E{WH_d7q4FHxPJMEz57PVK|f<5KMn%Bi*zb-JU<jbw>)OF-WhCA!E#dJtF
z69+|Trq~S5s}OqnlPq$R8j*<yQBSCYl#|etuqKN_CLJ=fYRo2odM-`>7?dO{V$=rX
zSPo;|leDTUgxuPJX()c()2=LUjHltd3Fyh_#b8s5^af4WcBprwX=itV*FJ<^wlBR(
z@sVYLM;DR8+tTUxim}8ez?weW_b-g(>?Y7pIWO5{+vLb28RsI!d0kw4{zO4UNS=5(
zrkXC)#CVzD*4Bm&sFK{`P1@K>yGw~~o1UbCx8ZK^B<H@76kR<v!qMerBkdd<ZYiyZ
z-4T&vxv<J{8YvuCNc{ErHRZk=`aFMe-*9^Wn_LJxHaoe|7JD8=X5aeejlD3&&aV;k
z{>z3xO7LmX{z2_B&kCL@F))jAI_FQVc{l{sG<X^L;lF__aWKbG_&sGPzAyVX|5V&9
zF3)A;%J2~OzfY3JWi8P<bnDZ6NalB-<QqI+uz!v%57u;u=XwyOh*s1(CmRMP<nwm@
z1G{bYa7EM`P?DG&9LYB2G!y-6YyHOy-9LeZ9~$IY#=G0(kckR*(pJ0?cN3+ks7xf8
zN!pjDb0O2yKE1~2dIjmA@U|+$Q!lh|Vey4hW4NIsc-CxnJJ%WM>#ii-X~<PoVSzWx
zHr?0BnpZ5go1@k9r9P*-%7IeA3+9=RO?_z|dug69uMj4-z~Yn<`aFBB(I3sn{6^*-
z5H6u<E=#;UWw{)+W~pQOpyNKn%^cMHDeVT24;L`?H(8z#YHUe+T#vY|&DsvwP#Rs$
z=F?$ojnhrdeYzM0#+aPQ9M~?B>k^%Lm*QgcV&IUw`rPs2+glzpdO}Ykc-!g`Jyf`o
z9ww&85mI2@C<6i?>e^zhjjiHZF|dB4R=-PN*I8gHN3aRXq&-C?5IjW>`=IZ3+BhV>
z;cwQxE0`^&(x6xWYCpQ%FR^|MxBafx1@@Y*v&4c{X8zxBdrcSuS-iUaW;h(4kNJXk
zD9gB?_?~_a63%vBqStR;gU=n9?4OPsKaF0Z&`?EmiOg|2h3AgnU^qI6EsG<FEo<8T
za)xr%wjZkg^608-YqJk&wBQDA1G+pBQH;A<@~+@(o$dH3$+7c8MCT9C{S<Jf!ZgE2
z(%jvJzztU*W~HUr5r~seYlOm@Mf;aIQC}*|BeI=*4*Qy`-X5*JZ@P=!I?E)jxWtVZ
znoaAg?OW@vrX*qs59BA-vZ2=>ko4^F7gi!Ib+?l{b-9<4BM%!~1!GD+S$=5y(3QxY
z+V4I=Qcr{ycKs|ZkVoKBhI7u)<X|IFc5#SWZOiC}=OXLp46>dnt3Z!RhB%{~Lox8G
zvHA6ZXW3pcrHXUm-#)AMmB#X6nAh&NW!Kc!?kO{}N)|!Vq6S7cz=mW>Yu4i=HJ9i!
zQS_)ciB*PW1KnuV;k?h0{YR&U^zwx0669?Ve(f=g6@fpecBzsa`ydH<YGv>Or4g!C
zJztnfit-ZD8O(qfjah1Qo^$aa?iCULB4z8K+1S3|zW}8`TEElPINEG;+e{|%#;%s&
z(qr3Gb7TO2w&kj^U}W8isY_3+tDH24yPN;5xjSrDZfIYyJ+4OxgS4tp@<>)-T!gGh
zIp?L=(!_jJ8>s*Tot+0%b6}Z)Tr1d2<^CT;N1>Z5ZkVtuh6&R69<j2o*IMb>xG~ti
zvPCq_F5B&Hk1ZFcuH3ry%IT|v0i8j=XA{}Zhlon=IunmbYu7ie7t$CVRer=j3+ZxI
zo<%Dw&!7J-Gmo5=m9rdK5@uW?pAr87v0)^Pc_SB>sc|896J=)TLeJ00KuBjF&p0+C
zN(>@IEQb*ya^juTE>cGu%5!#hp6%18G3~2&O*$nSGm(L4=75!hpOZ;8l?sBu2r@g-
z=HiSxqxVis08-`e{%+>Xwtt(ve%l#r?*_!h5zz~gVIU23SVfB_?vk1o^<kMR(03_=
zseo*<vpmeMc-H2~#PCoqo?4r!c4*#;+9OK#gCIH_i4_y(lKMq0?K9$`Kk2=TyEd#{
zv;LypCGn<JoeiPjb&_uK*--V4OLpzuu3%f#Ib4_@Lv|h!le%1$gT~7IO5iNhg|^@m
zuqbBQf;FccY9nZ|suKjJ=DVmB*u|~DE^Y;O32#BSP{S`n4p_-e1(t%S6BTzIZXe}G
zf6R}B3~)N-_+ZZt8aQ^{9^a1Aj^>`B#~vH%UDRdU>fdMK_TR;wiD><3_m-Z~#z;rU
z2=)WT0FnNYXl@P@!^2`ZVk^uU*b3wPVQDAC_7uJc3Pi0?r`>;QFbC-ZmhHU&8c-9}
zOYPrE<^eOC8mV2G?muK^HY;Z#sEy2<#4`u0oXl1^kp)y_ncvSs^uHLU2rh^vBK<50
zoyp#*yl^7t;Lg`x%L6A`|0eCv4juzdqB%-T?A)z`^k)g?mV5~OU}#47ESpm9vj+pi
z8KrGXg>Xh3O_Yf-QlpRP(F3<g!Ly^JJRP&Wo^9+JNXNO)8+Jsta&H{z%8xtRvFWhM
zx~c&3XkcCOtSY9xSy)aFT}!m<XRb_J+8icn7yAsy`pD8Reo_3<7a{b#<eQv4_wO|3
zJjGVLu3(J#{GiyQ3Z>kvE{E8FtlYrL5tD?!XE_eod|FDzAI=56fK}2lYjfUs-x^S!
z#weoJa`fbHsl3or?1`RZmGqnw^n?@iIQYdl<gi!ND~L=NjU@>4f}PD|&I|GBK)rqC
zx*0WQY4u@Ur1H*}GayUv(6&N?^d8dIu<&SFo5?E!HEq?Hn9~j()S|4*MiXY5mZ1kk
z4B$f@v`j3~crg>7p*h<A5VNH4fU@vi_$-K4+6UV8fnm3P96xd-zA<|F@jdpaH&|J^
zsH3MQ=@_Zn_%d|I(WB3QC*;=~bb;{Ri^Fv%E~o1$+MHnEzW|4xnX&m7K6hv9X|AXt
z>a%VSc?#Wtd1&u=xCy|+O+X&fO&;a+8}<-7wr)Du@g_e9&)N2yu0J<)@aDa?Do-G4
zZB33=CtV{$fY~8z5cCl?2!(b}Mw)JY8MZs~?@4P(n;~EPt<dLGiqP|?SoRr9fxx+H
zy*9|u@DQQP&e*<&9v#;Z@@ueZMGC~tp&UTQ)L=I!-zBOrZ|O5DvBhjN8^mDs>fO6n
z-?gg9?z7tyF}rv&QdJeH*=eim|H$CIpWC)+(>C#o9phK5?y%X*w^zkNv8qSam>8>Y
zAM|ggO=4lZN4^jR$p5g~SeZv24r~dZ>m__5l#mfFQJtWh6<0;v!=6=;XMin6^w}T+
zYqmdv_%nbL0|HG#ECY-mgzWr0$XkwNC@qX&(6p#PetS59D`Wd)8|7J>f?xnA|57&H
z{dA1|f(<js@+Os{18EvgrwlB#D?OssQ`by2!DxQ^nkn&G7{+Ev_reP}hOfT5B|Uc4
z=${Sx`<-CnUjX}BP@&J#dw>;6l)Ng~QZpYiGZ`{7U%Sk>c1emjGp3t|Fga}5kz2*c
z*N@zK1lHe+Q#)RHWnP;b0c99jFBy&y4>=>?-y|q@<Z>n?(-{&9=)PR>@<0yY9IH$P
z+#JMM=^kMXtTG3vy6hKts`!v9Hz9XfLC1(BQF0wbuShhDVk6Ivz;F)TJwxJOhkCn*
zq`yi-ik8c-+eSu5N5nmY!^4AW%%rr3>G_8-mu>~vORQPGg&Mi5*7G5AhE)MFxD<4L
zkabxFnMiu>dXpn(F42fY!;VR&tgBPwI#>1_Hrj5#g6irWk-Cvingq9Lu=r*L_xPL=
z^VGohGMZSc>T--<Ov{Th%8Qk9zW1pi<%<xHww<OXQtMb6UpLM2td5wX=`|heogxUD
zA}5IG7!VV(Gy3M4TSmK$-J>_0A&bnJC!RQS?!6zWX|W53K<&>wd&!r3s(`MLj#^4R
zm8n)JGcPuo*>ULs2GMVcpi_v%o%EYG->gu2^UX8cwsC5=O$!`92Xfpgd<<;gndfpE
z@QfyPoJNmR?|J}9TZBka3wa8~jG7qkkR}+k&(ZBonWC^baeG5VvsH{bVyr1fHz;lY
zRJWwn*t+Cn&z`!szs1<nf7_|6*B#!sY}r21klMO*>DK@L@;j7ID25Vr-vH&S|Dt@-
z#pIYrxF(J6%DYZ5jU*Q|5dOND9qn5wwvu{07+Cfxo>Ec7fDF8C9&8(n)|>&jL_Fwr
zy>D#Ms`WDs%?<5DeEn;<a?@4@y99Ko2)><U9RQtc7L?>nolE1wsR4kcbuJK6X6{Q*
z>9;8%*I+rdF-1f-M}w~!&QkQcon0L)g)vzZ+IdslL|B1k1P)(r(VP9YjhAeg@rO4C
zDrRQ>wbXB#ShZ!#cODMc81J~l5O4vK6vm3;3t_Cn8T*tD#qW*gYFI(Wj$++6elV<i
zT8p4lD6K=#b1l#z0KNX;;Rh))_uP9YPEeX?9>M=4G+9-j(|H9U+ar`iUbr9P?`%lB
z4f9w*o{0xwQe@_tU=WpX_AnK|>8E|@>3V=!J$(p^3bP#OK3h7mm5rZa#6gr(C4)jz
zO7JuT^mm%>EHmwuHoK=f{go=eXJ!g4RhrME{i~oqSUyw9z+zo5=bzz}GsC~@FZelc
z>}er0hV>9DXYqf@^RL6th_M16YKg}7>n*4ApO&vFXfMrkZlBX$YDW}S^ccdh?n|Z6
zeV5czk3J-z+s<fpE{%7x8tWk1=qVknn0$e;fTH+aQt%>qF&u}5C@B_><FJqg&w?r>
z4OgjW&Uj!baM+nm6~}WHAY+d`HoHZ<Lu{IUb~gs@pfr!pFT-=9O5r%nFNH@d<C%P#
zJ2p+F5)#m6<0g3zgg4__l@YgW-y*N?l9#+45B8aw^-;W`p?(zeDh;b?k5c+6f|MCp
zjW<}!_?a#=h#&Os+OU7Qt*>==EuQYC-jPmEzgmPRzPqoT7Us3_X!~@Oe>8DqGs@F(
zYKL^RuNn!zHc4m5-s~cn&oQA^HNKi7?Q}o!v8^Ik8i=PV&il=&I6ym`>8Qy(brKDY
zjQUu@`JKKcsw1isH{X0hb%eE~?~GVqq&lN|>KkAG2h{<Veu}mYadgmTsUc&I(eg)?
za)w!&|L^%J=@7(d5&k!?{j|V2C|lGtsFU|&KZOZkL>nrb07fzajF|H%ZBQ*)(4fkZ
ziZOcr9pmTUF|Pc_`1yB?R2IYY?+9Y@?1gcAV0g6+jkX+aQA790M+LeU9o`Mlyn^7p
z&-^cr2V(bdoEqsk{fY8mjEo0h`dYwGt%RL6UJ#})$a1L^!t~LVubF9>Enq?AF`^d4
z_<;%aU-~vz6a@KQy7ic`C|Oe&<+pBs>0L^^PsGI3sql5TEBHDkRt}c`U!}Y+qx`T^
z-pR@lkB7gfvFmzMw2ym8ACF4_Q<}?b<17%zWdR$fk)7n~DM#zkyeVaxH_^}y=PumV
zNsqksY`ZF;Yc0lKCBx&TfL$-=UoYohFL|_=BX^;c+eYaDu^Qh;bho*1qkZGVtQz09
z{+PbL{>Q1GWaIk)^22Nlt0e%8Kb6-$dLa>n^V&z%Alq#C?Th5*7Yu$&qvo8CowP({
zCv~v?0Hr4f=Fu5_q3POdn|6c`%}iRWe7?%k{+{9b&a%}NJH+PHOQdhke&Q2i)x1zs
zZO|L4Vqcz)EqeJA&_Cp#v`GTTpReV$l@`GCoQJKnB->UBJ>@}52J0LT=^PL5Hnq*P
zB!|s3jibql{hXZ@4a=^$b$3Z6^g?gfvWBj*<x6G(QUmPqzmY1V-Vm*OaJn(^0>VE9
zY5v^&dl;h@Y^Squ;kMIYhV4`nf>)#bs*R}*1sr5%VNK{F7Q<FK$(1S~P6}9YqMA{o
z28@E8lp{v9KpWAvq{tC+x;tu$YpqVJk^Ir-$;r*PURTr5P*W^z*l_jVT{hr!J1(^@
z+V;tVkEGr^v~S-b2Ef7nCdKSog)_j&nRe5Kp<4wdqD;P_>$glt8|+pxA7B|$2lAnp
z$<WK}C|ex@=EC7$5fpxE4B&<*W@n$6U2zP8Y}3^Nfq)bPvR<O;k-nMjH5SO94?(l6
zYF3A|@%4OH2fiSbvEY25;0D<0TTEG_7+J&5SR>s=ElY@(A3S*Z!SF}+?)}I;x5k>A
zV@|P!LSu;h&R_t6V++S|FL0lBFQ93?W)pJ$_E!~BGs^}GJYYr-E{^r)c36(wvPb9!
z!5*zVc9&T3rDJyioc*j=wfEdNU~S76vzPQfkGT)pe1t~b=eAfIQyW6rw%FWyi48{v
zoEfQb^Tv5F%VHK)Hv>+f4z({wSRz0gb031Z-ZnFH+tS|DfA{u^rrxDXdo4VA>(dWC
z^z@!fF1A=MzC;|5FTUiGi!(9a()|L>oic^Ev?%EKo-99;1*4hT6=ar2l~ssdqukD#
zsn;t?KN?!asY!KqlR!ErievrwF+=&O&#2OvU=&)+d@_1*>D^SP@c;%Kb)8PC>loDK
zw=3n{8Rch`@}7+HN0ss+R*vz@P#5dM{$ig=%lCn<lxg?W?0E~gA)Us@G37I^Ho(#j
zr6z0(9D-+BI@_b|zyH-eYX^*j>+bl~ZMWTc+r0SXSMIv&E2+`lfBUz)Y5vP8X)F35
zeET#e-pyg_)a=^|=yRF2PAl%$&&SqTqS!id>2>m@*U6V&CtrG<WI{XH(#ve3N~7Jd
z0KeY4qR%w2>f<jw^Qpl;WB<s<pS^R#vD=RwyItHTuDj}*YpzOt{jaGM885_T<GA<I
zaetStPZ1`gP&ct>=_U)~wbfl<S4<YU@!AFkBQ-FPk*H{|&>24389v?_KHeEV-Wfj9
z89vq-Y9p!w`lb>|@9wR;Vsh5(s|y8N-9$Fc7ndBl89?6=0DW(uNOi8a`cS{W8|HYX
zey>@ZH3ks1uYD2uDb+QKXxoc;+i8Rhy2%IpPVrfKwGeHz96(&kHp|U8-Fp@t3)EKC
zB#oN~TdH@-%1~9i!x6R4&g%WuJN`p=^rJuh+0a1I=T1r{17O<c&4$@@Yy%wg!HoS8
z#?D3mkpw$annHgTGuVEF55c!UWb^ex{R<LY?vm;J*4JmBpZz-bE*QHn$Nn#0jt$I{
zfWND!Yt#HPsRwn>Ye5qUAyQPJL3`CwHn&YKr~priG-9#|70MYtVYz0h$Htc^>Xi*Q
z98LtLfMLr{H_B*=i)5KqMc3ce*1PNMH=Ypx=7iqa5!|(DZ2VKp#m4@YB2(&nM7N%P
z$ZWafs?=ZlSMl|>2QjwC!90?AmZ2Fl4YF2kyh}su^>d&aNo+4rOdDCu8(FO2y=m-4
zhn+exwyPnK(RSX*Po`UPBW;?}h@ZbsTLpGUhEMlEE=|WmzHpj$r1c}9Nnn&4z~Bge
zKx@bg{D#orXrul>M$#sOa5s59fZwDc>}fk=v|TYkj?QWvNu<+?CrND*ZH-^g)FH%A
zqi*yGaYX!~^bBEcFWHKqZp3#z{Ls#~V1)E?m{u0S@-B~)PSd9v*uvq9143R0YAdr9
zDw>fJQ6&irC4L1KRIzdOiWRHj74di?@UmMhcJdNOc5mCdd-v9DyT|Kd%VKr-wyCt(
zTwIE8IF~*zE)$<ae}b*BmYWf1m)E;k-3AfcU_?Bd1-&?A?sbh;bqiZK&}`kFXg9`^
za4`1x#hWS*Piz#Ijs2`{peOY*^ON}uq2p=Nw|-$0OtQ@QrW#N^TcVH-AJ!zz!5+%T
zL=oU=Xii8k_$ne}i;?!Eg7j5cx2$49d^+{(L(x#E+#fKkE?qh>ytFJ7bB#SAE?d3D
zU+t0HWx?`LtaG5%Qx~l$X$E@lB)sqxp|?r+6c{EmbL?u6@GN;{ITVzls&2t^IU{rG
zPJ70IQRln##)!w_pq@i;SQ$9|MRM~C@_GvkAS9;_*D--5Veq3R6rwGQ`Js?e>O>)?
zjecUF!te8V8myh3!M5R5P2uY1^`jN>sIPd$QWFk@qM<6O-euF-oo;7aFw)T!t=H=e
zjn#{KEDoPF;SPDqWw!&Mk-}YM9DjiPR4QCb)hHWd3U!R{d^@FbnD}t|rc?0cCD>Oo
zxzaLl+mV<+(%EPr9+4=ENqV*=M~w-w_Zy#?TC`}RVPaBTwq*G)elxN}bbJi?>^?GX
z&*HhrVTc=_d3LAqPhkUz;})?|o`*l<3T+^cGBCjOAsqw!IPqs?<QhQ5v4Z@9M$Q$Z
z>42N66zIxOB@^OTA1l(Et76`!O2?YzJtJj9B~>Bv&+iTy$E|;nuGk-l+Cq(i?&S@^
zgsY;$t8@G9rs9`Kuia1Dc18k|R24*!Pv=YX<j?=5?sqFlA@T>;W58d&-=R%0v8A{S
zbKWcK6WQS;N@XKYAV3x>b>ye`($(JDgztEDZ$rRfD6WaM4zC^S3i*8%A#r5QNNjO;
zv)$X+*yNR*(f;=CWg~4xKJt&e;aGi<@Dhxb3F3D_I-p|JSmjHlgJT8pDkWSObc8$n
zVC*AOO@3qcs+D5!dh*!%k&mxP{R`o$Pa{7(gDZNw@Kxxw4k1aGVz#E&E|@C@>ny?0
zp>#EmU_V|5%ejYe$XlTF9>6K?qceLaxw(?OX5f4mfRA^Rn+HH0CX5uvS#2br;>bP@
zc1K5~md<aK4%8c_BV&4J$_jU8EUv~StLmeH>LtU&OR59W`c+GOe!mZ2!GPZ%2#O<p
z&C#YJU2$EwA(?Cl*A?rEnxf5pzjQf@94>tOoZDe8ce%>V4mauJPm<339_c?PnNgFV
zlJTgn!3^J^e5tsQ?;t;vgZ5W3@Ew=|gV>WL3aT99`@1OLft@c(bm$n(l8wzb@@8Ty
zzlU5mzQxRVZgkn|`~4Lme|Lq{M3^paYhAxca#jrxmRqw2`K^ue9Fg(!e-(cy9=#wL
z@5^cb(8_p$S@_a2UjIR5ykS8R4>F#-JtrP0DC2Xt;X-75LL=kT5}wF+^7b_8*Fy5$
zmvakpzVf|JykPlGq<f89E8S7P6Fzu~(Ay;3l~=y|vgNz&0_A(AIM)dLv^=LXzX5rs
z`a_oNub+otCvu&T`~&2nf^wa$%DK3S%5{AXxvtm9bv={ok-Ty}E!BxkCu8+2p$%oa
zKTD>Efq<}zBDD!NjC(-~0(k}IFy^N+-OsjwI<^HY$Di5B)sKsE2=qC+V_Hb0JKrrN
z(Z%XK5*=JPg!@U`&XBe(EYoxMc_EqZVK(!$OxJ%PnQmA}q=QT+Z!lJcWO}ZaWXtq~
zBTJ?;i4JS{r;#6yUXV=BT_+5D9DA!4<|UO;wM?eha+zMsWqK`<>9s_r*K(O&OJsU2
z%Jc}6>9q*(1({yU{G6$Cokfub3(NH6LNcAoavziJhTI~3I4#FNmzM6pbDtz*^gYsl
z=O@#B*)rX>kWA-Dg|tks{va~_je;Vb@WTV*r1<y;-M71_OxGzgUH5_a?MoLgUJ9?U
z*Bgcx-?vY0y=dLmt?Mq@+FM^UQd5s_8%m3dK&Th7efyWh%fyHIzFo|c>G{qgT@+!+
z+aO^ZWO7yK<J7V94I{?5;37Tx<mR=(eH&N)m-w|8>N+}7Zx09x9p5B<8>DhwvBl-y
zx3j}u7nSQ+(;Sd*XD8ht*G)*3mt2E%dCidOa$2f8Bj213d%WeIplQTV9&QTt4q5%x
zuHLWA3*xsg@>aQJyTkAD*~1=dcfB(juCO)^AkAMO{O~ZLxgPN4LhhLAxyRp#wAsk_
zMh3QpwL_;wMb{x3%*HZ1id+~d*qZ!N(QA_fwr+DY*j5<`g@b;t%N;9ib#=yj$LfNW
z_2c4Xy)Rbot2LH8yPY;uX}Pn!(O=P87mn$5`kH7{2h7_qkg+(G-h0c6j0RYFI^Lxw
zVo;hp=m{*#wTd_*9h&${Sw3g|9M|{QLt*c&i0Eh^+1`qMpS@*QQ_~9lcJZOn=U(j`
zy!ylm^3<=8K0k$+k}ILlGchHa-GR2x{p5RNT0VoL4S1nY0s)v_ip#1UKw=8=!Nn(J
z4w=k0;xGnceMg_Q%UtCP_yg|ppnkOE^{)_#9j$I2TuM6LA9MObE<NESlj)hkX-`Fa
zqHQts_m^3loMeIA1g)Z{ir|8=(0&_R=3?ArVj(<$n}S#dYov|y5R=HFS|VpA>KGhz
zRW*CA9NkcCFxA#H_Kf%Xz1{#3yW-@s-snJky|sB&T(nnpwzLg*`(4Fuzu#T#^21#C
zCeB|F#RC+Ur<TY0PV-&pNrCnd-Mc&^1>a;x#qHB0qHtiNblvdGNa|g>20uvp>Jeq1
zedPV^v$5A|*~(jjFipD0F0^N+`*Pg}*fS3-sju)?4fgg9R{1OHmkfA(5T@Ga@%l+O
z`~Bi%vc9s$q%+loVu?g7gtwLT$>&^UCcDdJH<h{Wc9)qPZnwi!=4SKs+qh?rf;1{o
z_y+LXuJh>y+NvOHbCAg(W1RC8MAR|T)iX<|B1$H+xzDEgW`J^%$p?Ca*DGF@<{QRH
z-$c$q9-^|9=vt498!kw;dh+a-vt_Hmj^N(UemRe9H7qDvL8g(nhsD(eWveH9^|W26
zY)xuq>i`$6l!kATek~+dJ=t>Ab|G?gK)i6dI+kCof?OrM@GzmdkX+3@_7^5s^&g&G
zH7qDrVZM^aokD&nC|B8PnTs=EEOa^Ls!k(UaYKC;{BiQg)j>t9f?Oqi4zv}LtA+a9
zuLPC9V7aRQAad1^PppDmWo=qmuI6qTm#Z$ezs)6A^&d#C8VZY5kgGU<SJQoN9=Ynt
zma8_kTqW#vHiuLd?M$i?nM(TU5i;K{M5gBMF)mYUafV%pOg-N{>jlbG!v%>{Hb=ir
z`r`a#YOY?$l&J!<JLi(Ai4P)E3rSRn6-64ePC6~cq#w**CP2&pcsEKf3iq*CQh&|)
z{anWP&yer`G57a*p8wAo-^1UZ$^Ctv`d?b0e(m>X3VaXG|KS4PXV(8?&G$db{e7PH
z|73yZzq|nbFE3EP_WS2@f1iiGSKxa%cf@$vFOvl&2}(Q7n91X-s!Zp(BOA%Er%zUh
ziKN3VMoW(j^%#dgcV+7BJH%f{o_^{n@s8NJbySwS=HHcur8A)Xr2R7?xu%@-i*DIr
zG>fjexm~9o-G%S1VzXG^m--*k*q3@K^-7;;O#Kf&$08h+dZi&kw;P~zRB{SW64C+j
ztZN|jxUHf)8F5Bp#tP9GduPewW^scUPW^I&*t~ejuce_Si(@*&ltC9GpYWMWNFUe<
zZKXmSZ1e1y`Oe<cFh8=I8slWShhWb!s~Ag4cJ;Hx*Rkr<xOj?IjdhWJuw%~Z9Ca0d
zJ%WB!L6!-7!ftdAh;HK&bycs?MxIKI^O~Tp0O_k0ysyM`Um4K@8TzU^HXz1~k@ko?
z85fgoaWAF#mtvSc=cuNy;=(3rv2+=Irat{l0d-S&rs#~ujBe3L9_)5T?c$-Gz4DM4
zU!MBsPMv;_blD!gt~2%0a<O?x?%gRa2G|98W*vW~DgDe647Q5TB)UtCJdt#Ok%%v;
z4j=N*%ONHXJVcB^CHPdRB%4mvGtHPV75OCLG`f=_q0~*(j7S_FiDbefQ%{MXMY@}p
zr(VKGX49!^JJ>h`d}gve;Z71N2?NI36ZT&tRX`>@ViVQ?52Q6<e+C)G&?eNvyvWQa
zmKNhrlKN!wJ$42%ID{CFQWh5_9JW79%`z50jl*{C&*B4Y<QAwW&3C!#No3VCJM{-z
zkN6zcBR+8M&q_Td!j)ux9)qMNfVg7yJPAlD06{kJIpZV~w4Kfx@m2a5XX*wvaXy9h
z+&%Tz^t=g>=Pu{ZElNKZ=DjiFx$XS9(sF!m>ITu7@!Y>o-JO1JiLjQ?yOln-I2XOJ
zzLC)cKT|sMN+`J>P5qudwj=woNbg!gFMF;&7rp$s?fkhnP;!4Sel+8`s%{Wy9Qz(V
zj`{-pQj|5x(P@EA^NpF~_%^M0r&P=*`U3T+`6XRXgw>PCsz>2VtVSwM?c~)E`DB@Y
zOj;r9h<t)f-`^*a=?|p5KkK_cu<w3HzeAs?+ojD?oam}<0H`o*?vq5P%#;I+6eFdL
z^6^VcS_oMkgs7G!FJ*j?;!9d$I=x;Oqu(g^LVE2!L+Xsf#y1lWTp&$5U%MG!x+$Zo
zmvP!5zgFGT75x#~alcl7pliEch!Os}pX}4z6d+QWUf*15{$gQg6y#-4q!sJ2GGAiF
zqD3nbzNyxhmR81g7Od^2_}ZSz%AU3HQd9TXSod)i50;R+FXDCU({-0;&lw&BGff1o
z`(s!)R5`^7$f)}`)MzT@6x^h$TM&lGxF3TAyI}lfXR}X|`Ia^AkWCywn(a#VCp!4F
zFZVgCgLZqc+S#<WzpAQ#ZPPxsbefU0Vt3T*jk?WyyshJ{t;<@y%KDa3znHEc)(gLS
zx^YhwCqzpq4@o^Dn-qsA0lZ>P2qS=MCtgw~soA^7ObIDg>h};H93=CySy&5)v2meA
z#e*6=Y({7;+y~0<Lb@tA!IsG$q=}T!1hh?YF1cy`4N5--_Mr%R=*}8s+^AjZ-_awv
z1Fcog+NG0y{g;f_jtnnZ6!$d`R8{vkdK%+yopFzJdXG_8=4u;R-!Xn<T|?vQ1EX7R
z?-;DRY&fy1BM|Idn;6;K;FNlHh`rhE3T3sc37XNQ&1;D5Y9g&^#CCygrv%#tapp^y
zb%bjV84UqLL%Nxeyc!h?aU-GG9VbobudW(s_Qe-186K$}zofr!a%ruzsx{z7dd1_U
z{mzEHBZ;-0!9d5V#PDTxgB`bT89lJNv0>ej@s9N)ZLTuX2Xw9;kv59Il3pi^!66Dd
z!FA($7%xv6$6>@9j@XcCuuGQPU^U)AoD~p$1|{<*wGEaC28QTxkUNOeTEc8Ft~foe
zP$k-OxzWUe(N&Lns>)ndcDv8kY;Uhoe~>nonu;x!k{(}Fxxu#FC~Xp-lYU8-+(W0)
zsmuF8+G0}LVq$HvC~Yajwv>|+>e*keYKyul*pj`@3R_vo=Jzz)TB53ZX;X=**zE4~
zRq-on!>6Sy#Al`F$WZjac<2x?BA}-v^^!z7-{Lnj{ALZ@9Do^SU|E(vE!L-Ak)GQ|
z$N6W{YVo-AThjZBz;=@M<%8$G$lL!>enT@?fFL!}!-Ncjb5nJ76M0FiYy2L64ZeY0
z;uz`wpA%jxp~xpQLq5?sH8k6KMplfwRh10sA#e|w`bHXhZH9%J1;b7$i{t~jxE&d!
zBr-mIzb+iEBd@QrtF)=o<MLEyy$NV{yIUHNoN@p)o^cwslO|6CIF78Pbv3_<^P7kH
z%^YJI9gr6z(uIi8SC>p&GE835K;)^!*T@UT;7u}>zav8|hej9!Wo)q~Z3VP$();9j
zV%N%vX=$MP$qvA{k>^bF=VbH9YR(aHeq+TO5PWXq=cgxDO26%RvKr_)EDefp$b_rG
zhAlIvet~5suwrHce2$KHfsVIahYjYrfMK?vqDB6keBHw<9%mI3sTwm9UHX#a!&hJ3
ze_XbFw)N1V*3Sx*hNXmta`1P^paJ_Nt1A?ifq*vjz(Jr@j`Awo@J1O6H)--8X|2*y
zRx8vwOzJEq>qHNX16wEZ)~R6G-eUY}!8s-(0L={n`eBWBm&#&eV_$>6XxwRWltWs)
za)-q^UgU4+mxjYF-M*?wS)jbQxI9o6sq%HVgt7gHq#^Owgchq%M}t?{^V0(QGp!kf
zH7=E)v5*OQRik@vrO97X;x|?9bvH^w)1B3q$vT~US#{?$wd+fb;wiZsLd`=IDuG4c
zgQ$ZGaUl9Y2|UCx?SJ1{3QDBQVL5~8Y9a}e$w`fbVq-8-?Q~Wrg8b_9xBL83V|i6;
zD7dJqoL@%*o^Tty$UvTyn#CvN7swrC3EU56N$#HuY$q`GLv&as&+g|pcjf0|8{8at
zTnL^8GX^&0EVW2`XmiNm%9<u&i=aXEtp~<#3MFb>5DtW2J3Bk&y7-aBsV~_p8vXvp
z3Om0Jwv@Jz|8dOUkjBJYWhZG@G-Dl*<GH4-m@r1g<QnIlvhzi{mVZWaijC46ko6;D
z{)=%GZXy|C5nae-nDx<Q;;oH_J0$pPQTWICX7M}VDk-$l<J1uSTU#7`I^pk~=*1Z)
zFMwr2mQ>?edfb<hBNH^PWI{PmD3!#nmQG7!QP}P92AbT?{xY4}V-D6kqfIXbtO<*~
zq}1Zp%a^Rz$+}3?Q|T-MdUg_emSG~XDv%hN`xd4%Vz4j=4vL$peldtJ<W%B2zUpa+
zIQqH_7ZuesdW>$H-DhsKbiGvM2)k>${MI6;JK`-XDlfHKTAV)Ezs-+J8->S-tTM{F
z6l8XXvT5lG;XWdJjj};0`<Ap?xQNJ2qij^lR!f_NzsJ8dDP@z=fbcsaPmFSrQuZt<
zJ3-cIqg<?%T}8@<u*|HKEt3X?C&)T%lr5^VrKGH!l$EH;hDlj5DJxaVj!KQf?POjX
zWg9IM#QA%rcHw{w*+rHJpuOOPFmd)barQFtnPH;KTb!5d#1cTIa(#lHm>OV*7`^=F
zMX`{_KiqtcY_IeC`$LiN1f}H(X-wEg_}VC!ary*Natfa$ylRvk^ZOWmpAb(8`=zT$
zE#WLQBwBLivs4NjA+8CAk4>~E_E)vvd~jN_HZ-p~2zC9<e6w%@PW0q*rQN5<{(F4>
zM@0Xq5LmQ&0o4d-hVfmMNOflD4Rw)79eI5(;&pi=_(q@irPO~5OXq)xVhNbBrPWC0
z;@F)Rt?7CZXlN%igal}>lhGi}4@up^cja#oc6AH=nsr=RVwIKk?My8KR#i|k%Sqeq
zOu(>pzEp-)zj2^BSTy1cG=#$q0q00juz4^rHZdu6`|I1hv0zP*{2%kS)%!nHNtBKc
zjUIO{dSoJ9bI|kn<iwbk9&s@7RHeF)PNFPYdLGgz4^08WC!w?8fdnk7rB21OIjDT#
zx<stf>+5Yfl7rUCTEf^B;m{rxy>}6Mn}m;?hh~_EUI;`9s%-gLDnBE1q%r18&K4x|
z?v8S_r-E#e+)>yfHPZc(@<H3a7fMZLOG!zmw@SGoJE8gIWX*X@{w5^MBO}wxrK%!2
z74bU9vR0YNtV+w_G8U22%XBZffU0gD+cEhkoOj{yH&9)VM5MXslZn&|5sw?(J~G~b
zZ)CD&FBMM7-vU<}q6qZzIUdJ3j`SHheRS5BQu@yF?E{aNO%^oN4Xt7U5}R-9J9ez+
zw$Ef!`S156r>B$md@eto@3SA}9Q`OSK;&(cJH~cqQ7B)VcrtohPVywuhe@#-X)6}C
za_JD)2n9t<yp1=s6Cn}=crKVQ6ac6XCOqm#o!`iHek0v%fGv^g{a+^Q(|_I4f6cL;
zTX2Z4yQgD%y5k<Y7OW(5>+faoX7&lgnx1+)=T1Y1m}+H%LG=?m-S}da1EOZ>oNfE)
z^=-59y54$!(LlMoF&J!gmk$*A>wBeN9q3%YzVm?FQ{CjM@rT{+u)oIDRP6!!QvV_)
z3C~@4f5|!fs~ER8x%vz2z)9L)sed60=&Q8Da4h3Whtctq<<CRRXW?>zg}{yToW`s9
zjg{ZL&9WSqoduIf86FE^{Z?K-+<>*C3{3*pK@&Y`*-vBUmLfhny`F-xl2F=kd{J%G
z<Lhg=O3%psJkbaSLg9+tOBI^=eAvn6Lsy>ppiBd$X3b=XI3w-?li^cf@u<(94Li3#
znSU;jvHc9uJ;TC{0$A>_3qzc$5MS3rye~u8m*KO327X*{D3q39SbxL2nHDgxb~%*P
zOke|q#Rr@WLd>!mgc?_n0$>4Iy~0P;Yd`Dm@w`j?=WpsgcC7m*`7_#u{lQy0_U!4n
z1#Nv~w){%IowWC$%GSr^*}TREq{y@H@|y>Q|DR2LvXIl%N7^zYi{g*uj}xutT9vJj
zwdK2-w#;i9^#H%wscIDFC|khhSH|-56twwG=CS$7VlJDX>=B=p50fFO)7o=zKS)!;
z%1o8bu<*%;w^94v|A@!sdr6nqY3+MTqkk!!k6G`3)Mh=f>yf_xIpLcH?0WgSHiv=l
zgW2_Dv5;Nww~|vngF5@Fe0Dv(V*X`m9vHn7{04apI(-mta@qEF?3I3-VcYwy_#OHG
zLWT#eZ7;h5b^U7l-fy{m@3-PKe^xd>oaFX_33{W);IMIwl9#se$@eNYzTb&&$fw9;
z-IZ<QgTK__DlF0GvE}$AzZql}lJ{%pJC$MQ`y-*_Y4X?Ww06EZ&FmcH17KqVAe|j5
zLhIgQ@>)+`d!T1>Tl)T(QSY0i-bcwgF`%{eDdT-oJ1z<l@5kQvW&wNO@5NsedaOc+
z*4|f$PtM=?_xnuaAK3n05l_i);C}0}+_t{~+TR%+pvSGv1n&*ykYJt@2Ufw8`1OZp
z5PT)SK@fBa;iU9a@d@2gW)r-2VVmHKl$-Tv_?w{2u1CAzN&X>Y8Vn%QV>7Al`C;1!
zf4q==aQ<HD7s3G<&da;F#`z-GFHiED33>z6n@}NzQCQGo^JLm$1N+)HgzxG$p>3e$
z{CMzvS|{GfX`T2$G+eMnLPG<yNN^gKk^Zg_It7ec4>VXf4Hiy=1tZo24KAbsqV$>}
zwEo%54tHeJbwn{X9LYhRZW%W>q~6UYGEH9>(c7AYD+B;6BYi$jpO4e$BlOX*8dZco
zI7kO7Yv(BIQmu_EIOnIeIr3_4#_Trc*G{j@(PsJET2=K0P2*wj@JHb>-Db9T8052B
zkx>(TdZ#Dxhi#XTk1c4I_@i)2w}sgy2Dx40B;Vsr&>Nu9h}H_}|B-D%zWY3E6Ysk}
zKP>x%uDYOo;&rha=_@9*u+=ZeTBltT^Vul!@40i^C~&y5Y!q(@f3F)y8^vZ#kG)u+
z>+<SO7h<_+KYl!s({dr-k<W4gc8k{uPrko?{E%!HIL?{2i`PY2w;XL3`?>96f?6to
zCL0<u7)>@#6SrTarQ7*gGUy)eLaZ0M<=NH?K3|I2d>Ol-`SRf!GIYgx3>my{$#{>T
zE#q}=%b4Jni<9&Qs0^V!gLSgno`IQoO2F|9l6e3y#!J?R5@8K_?F8&B<b)xwyUFVT
z@_LxOo*}O}^7<uty$w-Lwhx&zp3*{I@DTlheCCq0JX*+Hf@Az4m`eJtIo5rXZlr*z
zgskGL=BuP)8RFqOFzkJ$P$#qtL&B)AOjs$b6*dT4gzdsEVN$qMxLi0O91*S&t`}|)
zZWcZ!+#!5I_>^$J@I}%`PYKUS-qXUJ!aoVmW6SO&t-5oL#s-Fa8@oE$)yu8{c8#)Y
z6T1$u>nOX9vFl29UB#~J*mVoLZe!O8cHP6Sd)f7JcHPgeN7?l{cD;#RKgO=Nv+LdL
z`YCq3pIyJeu3ur-N7?ls=7iJ24@mFG4WPW=FZ=-gt!&C&g>n~G?rM~~m~uC&+>I%B
zE0w!d%H2BUZVS63L=r+So}<@G2#ZZAzqw4gyG6OXRk^!exw|vR-QCLFJ<8p^%H1={
z-H+H^OKUY`C2+&S6afdT0nu10LN<9Pc^L^?CrU+Nb%++C@&;BpV;SW3ggcQ;IIG#!
zn8ZJJ8{26LPCdKE62>TA$;-&D{60y(gDl*w$%GZq2JVDWR@D{5V|EIdWz!9H!5Eow
z?NlptHjCm*<w1STHU3EI|F|RmYijhtv!5jY#>=BcGeyyIdg(89I7-RuVs^dyuyK*`
zaO$V~3@Z%#ZZ&NC`8LC?+ppN3`YCz)?ncA-wodr>bA0#o8(X}F7NfDnFj8gk7%YZr
zr_JVsS941=+5)fNl&x8`XiXWu-&nuY>s;2<w9M&U`hdIq&UKEUI~;Zg9qaBacW<lS
zU%kKFZZ8*QcrPw1EiF^OQg=AWziN0Jhqq5EA8({L<M>|9zQ(t0jxvW_Ln(6FC_ydJ
zrmn6g^4eP!4Ae9=)dYf7g7E(V=;}AS00031000620241CyI&7H^#BP2=l}o!0M=sd
zSO5S30M=sPP5#^d_yT|gcK`qY2>=2B000000C=2ZU}RumVf^=&fq^OF-|c_5nHm^?
zBFKO_1OT6^2G{@q0C=3O*GXtoK^O<{*AbIao6y#xU2N@=RJ%k?wToJ-H8fq)#%d`P
zic&;xp6UjIMLl{@i{gghf>jTq7paIR5f6d~1y6eQ=t&R+m-zq4f6SL>sA)~%$IN?~
z_h#mMGv9kP<4%o(KQsPECxoCEBCK~n547=E${6g22^fZc2=e>}ULWFdJ2XNR8o&>A
zP{r$fFaoVG0Hf5%F~lLv-wyEpaz4Q8Y*`QG&<({<!oKo(O+Z-->tHpkf{i?uxi+!y
zV$NZPx)>uV-*666KA|sg79PTDjycZXdU*ah>(b9Vk~TZxDeQ)qun#6FP9E@b3|}eA
z@yofRobPEk$zvJg4q65^oLjdNQ+4~>n2XHw2CwgCY^!<C3TS5R!+bVaBZ;?+dywPq
z<1-dz?90b^55fxGqk_LxvTca>ZBfs<mO67M@m(E<awqXa9eXH}K@HnT%sLh+-*w#P
z;vQcS#`oSN_j0ym{mTBFP=-w^4_Mn-##|he$`k)mP1&^ds=CkBXL|<yj8~(J<gRFr
z_+6^zp=0OEA+M@0q1Frr)Hp{MsX)~O8DcAAu{k|kHaCS$`g=gDH<|ZTAPbe0O<S+3
zOAZ`S>NAk{O1aC3NzcFkqr@eD9(u3Ho>*g&d%25aEj6j!C;MX@rcH7^NG@5<^Fy=;
zR`Qx5-d9Td?Xb-x*R{N3?R!XaLXgLP*7us^`%mhdOmdw@8K)^IG|Ba3IA%L{>7&4;
ze-A`V>RUo$UB=SSakj8;;bAGUXjbRl?*I?uFwOCene-eRVQhrgr016*lO8WQx1R0t
z7?U9B=Myo1=Z6NWV;ncb7G7(gi_OuJlZz<Eaq1i}MNNEOdJnfS){ywdm<zoZZTz>g
zf8_MUPmamPQSTf3&e!zXls;=sYHis4Y;!<Zwe96S<(up_T!V9P9xlUGxC>X{(gNDu
zf?2vl58ysLf_prdZ97j6$hwv9BH06dB;R|ohDudw*NBh5+tNO(LUMSIrEFur`|UH4
zZ^ko>$p<(A7vM8vctknOdESIijPEO$bezN>9D>KN3l72yI0ggCakvQ2@NY;S${Dy0
zb1(~S(Bl&0{cSnyIKMvXbDYPih0cRxSxkQt!}xFK2PN?U0C=1&!hc9wi30%e=C^fT
zYt6aNYieHWnwi&ieY>u8FLi0|<;?79`ZV)LbIx_voLA;`W@fFLYt8FgYiicaUd}7$
zn#2(i5hFrINJM;z7$GAfB94%VF-Cm<IX<6+goMI>h0jpWc+bR9O{m@fFcMLT-Df#x
zy=Olq@sr$18(+x22&0qGX7urw(3h*pxMY2@{~YR^@?7BDeo9e_FXiKT#(CTM#ni&o
z;ne5_%mw8I`-Kk}D#n6|Tr9ZgxOj>!z*?|#m#~-gm%^9gX>Dl-Y4NX$zZ%44;q<tz
zueo2_(y{4%ml>BgFCSh$#;4#(_+q>azlz_$AK*_2SOSSaBghC^LNlSCFhU3uqJ(Y2
z9x;=sCN5=UXNWR-GNv=uNi5PFDN5S^hWt$|Gm;s*g1)kFC7w0>t>)XcY$$s?`}ivU
zs{HCaIgbpHz2pcvmP5=b%Nfd9%6Ug2QUHpSVyDbg-sdLevUBTmeYx+c_0)~;IN$B%
z!Flh#XM8`DkIoPLkoZHeKvQt=Uu$7j;lhv9ALonkMaWNU#mZ}_YxZlA60l_RXa3Ls
z1Ngw-z*`^+Yy*2h95^jaF2$8Dl&+R;l<t%slzyTm(XccUjY<P(AWcY9(sVQvt()ec
zxoLh{n6^aM&<%78-A*5-BlKDNB0WOir0>#?=${#A#t!3vadI7X-ErN0-Css6TV=|a
zT4poT%Is&3Fn!DrbAh?a++ek_dRT)jFDuBJXRVY2<zTt6Tv@IwH<fpnJIdW`EStoJ
z*=n|)ZD#kehuGunY4+bYSU31LHbDT?gJ!S~90GU1ICxr-T!E|Tu5eViEBqB<4vAys
z^m9fyK2C_Uz**&NaCSHcH+yal-t=%WTq2jkE#j7Od0Z*CfotU2xC51pN^T`wsjk#l
znt8=M7LU(ctx{L5Rc%%6Rvq!R{ARwD-_IZ6`}iUL0)LgiQ7x=iR_m%at7FxN)gPfm
z2m=uz3RDD@K|DwbX&?h+f$Y#Q<b{IJJhTF>L+_z|=va^-ND-t990Iq%F9-{k1X01Z
zU{4SioC;Hgcp+I>D0B<`!mx0u2C9+QG}IVtPHIj?$s(L6OO!7HM4(70Qi^mUljuNn
zB1Va^Vv?9DE*5*l0r8x8S-b}GU@5GD4X_2a!^5x_4#M;BiX>kGNctsiiC+?yEJ-#b
zJCXy*iBu}pNDWep)GpnV#-*pV$+fuJtlIoqpcbqZ*6!CH%MxTpnN2nzbIK6etZY%X
zCfkzj%8u%Ib<#RbouSTBXP4*6OXO^MNd8HYq`)dj3ZufN7*IGBh+<Z;sE8;w6)~k+
zsaKknTgqMKk@B+&tx8j6s`6ANDz*wz$yE(1qspclP&rkIYF4#aPp((jyX$vv@ozbA
zd2a>OM72*Hy`6QNdV5~O()4Ra?j+xl-U(@ww5eK9%h$r%KCR;}bXR)UdH0~9prN}V
z_6z(=NSCb(HIf?1jkZR6<6Ki_6Q$|xy`p>8d;R)Ey-+XHtMxj)Pd}{>>lf}P-S^x_
z?te504Nk-9ukf$S&6s9m^QRVFi>al%#nIw!@wJ3nVn(uY(C9G+jB~AdttGAO)>!MI
z$!1zKMNFHfnCb90^>2r5+O`kBgTF7ENoK0qYYw!d+Ed%{?d0~t_O|wQ3t#~){tiqB
zu_L=f+oA6;cXV4()?_Qzino%jg;u}{TFutYPOwwhsqEBsu61s8?%KSzU>CV7zYFLB
zyM$fJu6TF%gTx2c2cLUvJ-a<ey`o-u?_zJHce6Lvd)Ry22llP>tv}>HTzdHK;f9@T
z&$k11mOc0g^T_nb`e?sDufOLp=5hVw=s@v6!xQuq!;@_X(P4Ca{)79+{9xvwa4`O-
z?9Y`U^pJWe`jq&z@9F+9aoG0^{jB-f+(_z(c_i|@@cHobRVU9m>s)rOJ7dlxSAq-U
z%5)XD%3P4E-eqvvT!SvJYu2^wI(DbKS?)G>$h|d+8qFJpM{T2Xqx+sTPnoC5<MgbJ
zC5)wx5y#*$&)D{O`nYu5`6BT}^NTGn;MIDa-hU=26TAt-M0n!RSLoCFmM8Njd6T-y
z@Z>*}9}yIig5Z!$go+d)41|LS5jo;Sd`J+9B3n}lQ<y3IlzJ-YPxQC>BQJ?BbuYsK
zRDc^m0!M!}{1u%pnue!)rnhHsGqf4?jDO}skQ&qk-@T%}GQ0}EIu5ZzZ(paq*1cYs
zO`KKE2Hv3Fz;909oX+L{A3EdKQvd+~00UG2%K#z(X#fTQ0RR91000UA00IC482|$S
z0eGC9RL^f4MHGI!cI|{TQPM)Ch6B?oqDUkg+f<5_N=QjjM8H;lRXvo`*gLVO-XCUn
zoUQXGaN@?DBmV?v?%X(V<AUIhxbV%}apI(jKzQYyZ)d*uy>H&UT_XC}e}_COe;fFh
z*rNw{DzQ&>dhT$AKA<lguFi1o&?0GvYxFLiIb5eB`o`foTA*JXo~KXfH-{JKq4&_?
zyL8WMIlM^UdY?JGM62G<4lmQ9_ou@vH0S+YV%+Ze6^A|G_Z;@=zW>DG3a$9t4p(W_
z{}Ma$s`Ge!<tv9hTK9i+*rz4`cZVzVssES5)fvtmTCHq3T%(UFM-JEN`^v=OIeJii
z-{E;Wtu`HApjP!ehwsu_^(TiHJyHF`;U#)nTXcAtR%<&Buh4SsD~IossSQhv&8^jb
zb=ae|dc|R%KCC}+xI&NX9fzy*w4OdsN0UtV2S&6S%`LIN+Y!&xY!rx{C=$Goi%jJz
zJ5ym`R}UlgMs(676(=gzyJ-|2tNu8WSu1EX8qX}n-Ym_sHm0nXD$8}62&Uacn2P@(
zbS|VYnG97dv!O^&FKn&DOJC<kWhxXp5wT1r;`KPwd8m5^^0_F|ED8%9s=$g`r=gs5
z^iYYNZdWJ6b#W-ikvNly2vsK9QYM(v3Yt85oaV8NR5CDTw7t1m6h+X5P|P|(FO6pw
zbT_s$F9vYDmnKG>q^G8knX+4mbWbHYe2x>F$PAP?*>8)3kxEK(yCkfOX=$24GY~B1
zV!0A$Qb)2IDN$%MKw<WFj)XMZ!uiO1nI4%u$aNH?S%33juYLQ<l>?j>KG*RG@=;o-
ztS6CGzBSX&Bbg7dmt-7wv6*v~rK3pV1ZE~=gc>H&%m%9bTp5GUNJ=A`P==31p9W+|
zP>UMWq%EL*+NBOg%gVqDfOjauC1xUWeo`0}#xvYQFsHGHU@Lk9)PYKZzn~M$#E9R;
zQv~)gp8GTgmQZLRy20183zaFFs!g>v-qL!>E^^FRhL}<NuL{?kKY9$Td6yD@19b_O
zjj?Wrm`myOhH9<5y7W2PurFI1t7e_q_yn~V^cs85m=)sQ1GjW<a~7;-eHECsd4mh5
z)0&1*ngDCKQG7;STsF&)T^wS4%%^t7IR)!v7~6QZGbKx7*0_}^^B5U&_G+t<@QEAj
zZksl76?_HARyvAj`~=uZikX`uG-Bcl&I_>08t<`!;V!XtHDF7w-1HP8u94#Ul=U2Q
z&eFGmlH%xn>`u@DGArK6rE>d<g5CS|u4%$qll%SB+10$4Jz*=Qxp%u<Bf%$SeUu%u
zhx0gMY)}JntsXfv?b(m`O$yjUgwdXQAL|E*Z~w2D+uy+J9sfDTg1c4HIjeMmdgz`F
z?#&pPY&Z46ImK!UWcS{4J!19kdv80;da=1|#Mg)|PpFT6h{2WoCgk*pc~-ZqSi}zO
z8T6Q=xL*Ixu|6Vx8n{de`|Y}JIOj2C4e8#yn;Evh%id4;q;AV-zjNpOo|u0CpDpF!
z0C=2bSY==vM-!cyb)=ZWWk{Ku>Q0g+m36vP;v|jRByO4%6<e{D*itG>p)xZwGcz+Y
zGcz;8SMDd>zOzoQ{=A*t*?9vyd#58I_xJDdv{MC`hWymb>MRQ>vRRQsX_`orsGTNr
z0}rDqG?liaY1Bd6(+)J9X3$LPq#bEi#oBC|Lp#$hv@5Mc>(Y9(K5al7(nho~Z9<#U
z1lo*3WROW=icpl=C`NHg(B`xSZAn|vSPpQI#?f-LJgq=0(n_>4twO6(BdtxNcsSii
zH<3l7=_*={M$+okM8oJ-x{Yq8Tj*hWgzlyL=nA@$E}{mykk+DRYN4%ZTiT8GpfR)!
z-A%*kQQDn8r$gy*I*N{@qv;$vmky#<I*=Zu03|6yBj`IiiN2@XX$`uJj-&CkCY?;D
z&_i?>-9Z=AC3H5O$c@~@&2%~)%q?^&9YR4mm0NiPkK|E2n#b^19>>e^^1K4C$Sd)6
zTz51Zjux&_l;UUas@4{xS~Y^8MOA|sKg8DB>xoqnwTh}yMGz~sqKejtO=7hotx<c`
zXsFSu2yKJldGG!A-g)QTd(S=JKlhw_|GM9I&PVKGeMhX>+tx1mS;Z_n=r=~cO;YpG
zefQpPUn^%LtQtK-`W4Brl&StTJ(6cBM~3Vv>Yf1R2zgq*_~AxINWA)n;$zdL=dIrY
z9y^#dzL#s6E8W)wpINg)S5mZ77EoKJP(pSjE1yqn6(t-iV@~2&Dv4Ox(oXP#$?OuE
zHY1i&y!J(xP!X$6k?Kpue_qYJ9uR`wO^sASwx@P#e%x<Mj?x_5kEHt(4BBY#VT&|B
zdX}o)BL(gAaK39Ad1J>gYM&eW!8B56*HD1Ggyr1rKc+nC0*Uu9Q}huk;(f^!w@?Q0
z5o7X81aGuK%-(0jWwG8tN-#nj-6J#Jid9D6J)WRLsO}6_k*g8H=n*e63c-&a-XOOl
zuAoN+$cYFpmj^EUQlT;~#Js(A#6=eq^q$QDqYL=OXGdJp+f~0qEnG<lm8l~)XswXy
ziAxMxU!aEL5_Z=*Hy`5)Pcbpog}bX!hq^d4?iB#k!#PemkOU2IPFT@cK*Se&BpM@V
z$b4^tb{S-f-wUB><9n1TrU$q14@SqD4w&%<qkA<pVf;uCxsb+>AKoQ*(yrh~#mUJB
z5-|hf`!cs)fGRf*Z^Gl?)DU<Ient6EK`RcA`A!u%&Bc>;51CJ2#(YH|3Y;R(jA!<P
zXuz}HnDM~_nX~sX`ykr8%zBpSJIbkLwy$l4H*JKZ=`Iof;gcinlgsae<WjXX8uAvp
z{?J1-TI%BvB{Z-0ssizTqiVE#&6rQ9X0eIaSY&8!t$fYUwyN`VrkaPMy~Fg&Fb^Gj
zm)n`_9@ppWZodS3#QRH2+!)BIzbh#pz@6PD`6y)0pe3j7k>8wA3$ns4i`zle<l|kG
z<c_L6hG0EshA~O@R|XP7Q4%*Rt-^Ec8?!&<YBo0gxYkJwZ0!6YA%s$J8dtPO$>@wL
zXrh$aD$I|hlrP%LM%{-nRY-c+OQw3H0FWMyc|R173Aje|4`pUT94u25kW46r<%=pA
z6JEk{L=^`K4X`3nWf?*WEUO4mMW}`q6e)QTP_W!0#SKC`tfWX;fRG4#=?KUpRKie>
zO11<f4C$yiLuiH-J1VOa(qY+)fL1~stZ-2&f=~?0TU7i*=!T&el{pCsluT{F3qm<X
z@I((ZZcb^_mM|w;QLx(AM~NYnuCA-fL{my**R^*<KT1cJL=e$|($aN(ml#FCA+CxO
zjVTR?YiObur5z#RLbN?-LMSeeDIe4z6qbknP!edeprKJpDoy?@dwDJ5@Xl!tel6(G
z_|&GtF2(YVKD!kipLM@Yxv%B$8^h)|T)q{;l}0r;d~1cPa?8~xDF@Qs;KA{z(94yZ
zVMnn`mI`A)U&H1sC~B$l&y>8cO>+U%Dc#oZwb~TaH=sEKYFk$|Z7cGceagu9VROp7
zHQ1uCI#V^Lx=+5W><#$6wWHMJH($2q%K7^NbuEyEHx-T6_veTZA2kTra>uu^pbYb7
zXY$rmX^SoJ^_Fi^LABOYhHvh?oZZNVL20Jgrw4=sq=8X=6~{2fW6oXQd@2dEgv8c(
z3uDSZvoI_O7r`1<<TY2ZK%x7>JuLO?g3QvfCa*4H;|e4n^=5d`yNy0S&v+fS-+<YM
z@wiWm?%|LYdEt!kn)p>$3~0gzu$F)_Fqc3(jpfLFqv)XC!iKECepq)s|K`3P*#m8C
z+}I4)-($W*ZJuFJ2P_aOSiD8#>m;7+A2pgDd?GxEP;&UhvBrd5WNqyp)_I@SDrD+*
z@7Z8>s5R6#{NV%%(yXCVSn22z(LIcHX><?Eg2*I(ukbEL!iU~wZ~II0gk>e(L=ANW
z-j0~cnw*?+lJ)55Xgsuh)u~{WqMjMHFGT3Dv<m-+n2iA+Kq23-w@FJa-*^gO4NTmw
zwAkT&Qw*ySRV`2XR#E^|)Nx8YJx!DL-xgX6+}aap_ws%@NNp_zG!MheN+SX>vx^Q3
z!i2+-U?RPAyyjEngxf_}pr({MD8CT&xaf|j2LHJhg+QONAu_OlO*zk|L)-%#aE5O)
z1j!H`2z59@z}chuY#AcEOJ=@DF2h4~o9b08<oWr3d)nMy=8>g}8RM+BrVWDHVSlBr
zpW+_0^8z+i3fLH;QE|>Q)Ov%U()^jD+w|cUoFd`=lf(Mm{+y7Ag$d)l`35~ml=EbX
zKy(Q+Q)m(sLP(fg_hNg9|5#xQwUGSNU)JV1udI)$6dE(q?sk!gn-!z33_w4IHRM71
z8kJV-Va<jN9<7=E#AZni7Wwq6V#;#&lCn7f(_DbJhGT8Boe=ueFAL?7J^@_6%i=13
za^F4B)8exGR^SKy`+G;l*==^ZVOMSm$>2ii<RuKpQ7!wkx+CXzu8i<2nAowI(R5bn
z1QX-Ie2NoWbwNT4j{nDVAj-1mb8A9*5VP0j&V?FeRZ_x*qK*YRXFIa)#URV`vTOBr
zM&f!&V96#B$aB)#KcN3J$z*D}WX*3UJSJH(ZqF5n@HO^p{XQLt7WGHhayBG=8`2$@
zJ7v(ZsR2*PgucP2R{h=ixWE~WnW;LrWVX_b7kh%wOwE~?heW`EW9?$NksWfF(%0Zj
zh4YQBP3kKJa6)(%CI(aL7w9LNTmKa&aU)|X`KzV_X%Z*i8kefvJe%FTd>L+@Ez}R`
zRpH(Acd*Uj(}{~IeW7J;z%NNVlk5vyes%*{M*bOjY->Hi^i5b&6M`M+SMTPvl~jC!
zPD*F=JQm>}R2Ad=^qC!4_48HowN(p+zefL7jP$1O?_p6nr^`OX2D{SdGeUcK$C$Kw
zupjQRu$;0`udlp_1=0VXM!hE}cKHIH?g;~piONnES@~a}Hz<CY5r63k4^5g1CKs-#
z|2NRu`%t7V({I;h%Kl8KVf7Ag6nDP)<Hpu=`-&x9s<0v8Ho`<=nsGXSnxhAhBV>39
z9E*|-xg6*lr0Yu<gLuQUN_1h_Pj#5PZqx!xbbn~BnD;x+HzkNlHw3zU>4OXZR-A2h
zRcB9w@43eCs-?pt4Z!6DbKX-sz4HsyE)ZJNX-QphSKZb1Ifrwm!KqiAvYpU{={j=%
zVzhXPninrN*z|ph86ojbmk>n{8z=pt>}*B-iu=#R6|li*9LQ(ZG%CF7=z?OcgPnOl
z?&2~34@{iIq|DQ!r(LvHx)SE2r2TCMyg!XAJu#MAF5E9ZC3t#a&|S}5JjRhUD=;X&
zq;X3A(62X}+V3p-`)j>Ga`)r1&nKi3<v#++wpYyH7SLL+g8SXb2t>rJ#yblo_Z$bt
z?zlznI(3UkmT$jQ3cqhD`n*SQFsT_w;Kp`gZ3*(mVHYyCEkZUS$o{DsSR1^cU(A=o
z_VfDdjK#-kvN+YvsET-qtB49y8Us&{qzWT8R_-d!^X{J7on;5l%Pwl}T$@t$F3VAT
zPeC1cc@_{i`8wSxBdwucTwDck^67?R^oTr+MU~dW%-K(G&idQx8E934;_CK=9;90<
zNUe9by8<1?ogIRY&U|LZ-CZ7jDTEUS5+80yEL#W)m<kFQ3bJz<oj)rqZjJyq$6Ic<
z?rW^g`7lVw9bID~t%`i;znzxsEy&raa^<<3He<i1Bk-L|=wj)y{NyG_6kFm&_;`BD
zR1^}{%q;ciB*+5R`l7b`VS9(}oZ3dV(+QVaJ5jUmCT96^^!!_Ku+|?ZRNN0)7oo)z
zPefaFhVdkh-uk`1fRhjQOl-2)XM8z$vGi2cTI`7MD@)|<m9n2Ub@N=a-SV9qdO*mr
zfi<6lh0^ld`L6^h-N90?JT=#^_eFJ$@rm{D#~WeD;HnSoa-(BIP)AY{FR@tE!9`KW
z)}^tu(9f)ML#aaC6^V~;IAa(_sSo{Is{?0tn$bge4QJ9guf&&`{q+~0ID|CN_td*I
zMX<7|L8SpZ$(n9)#Bj=pU_vFClk*mDHi2LN*7;u2z1!}^dgBtEO`pa+JE;epH^tmD
zym=sW&-c>QMl>E3f_Xd}ohEEZ93p|N?H})D7Yxzo6@8qw)-&UeB%z|r24IZ?*L8js
zkE_~$OJWPd+N;ZX7df?9=YgLetj~@b5lvDQ@W0OBC>@)ZI4f2A)ns$$w)NSHzHH2v
zJV>?LGe+aps%XG#Oss)+qS@JT_g-fFGuM1snVFJb{aPrz_~DaB%d+cX$J@WxwOY(-
h@zZe_lV3JKRvdRg6LS4>E$I+WWfr@y|LMYoe*#6Z(kTD{

literal 0
HcmV?d00001

diff --git a/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2 b/0.16/_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..79dffdb85f745ed98906f3f28dfe833acd98e287
GIT binary patch
literal 37592
zcmV)5K*_&%Pew8T0RR910Fu}M5&!@I0nJbV0FrV50ssI200000000000000000000
z0000QfdU(fDjbtK24EJKIslYr2t)~ja}f{<h0t7uhd2Q?0we>PPz#MJ00bZflPw3h
z4-A7?Tf(4`1J2D!5chGS+%|wdiu$zqJMjeWusz`1Hdrg0<7gPw;LVej+^q_Z)sZpT
z|NsC0|AypC9Qw(-^OBsS3W^Qgy8ksz%qga*!b@pOTUxduQKPU}=e>*7S{aX0hN7&P
z3_{hbCmsV7hf`R>n0w};R+)yETGf_u$8#3bw&%0#$4h;EZ{4%(s>m4^Y>5f0L&JT_
z(Ftox;42%`W_-`g;y99-7OTGWLgCxwMnC$L&Rs%U<q|L;35mZCE);wJN=Q&|dQ=3g
zV!#w7WT7(PD)3yZ8drHZ=*{rJWG4zS!<5_7wvT%0v$=PyEqFK2)Ze04Deg%nLxshX
zr+-uDqoSWrms_Y8^*CMihYClhm^#3QTlh(s<OUy;3(ifMWaNJW<EG7sC?vSG>5eqZ
zbmQmI*y(+_{)usr`EQp3Nl46u8#BX&NBG6&5$}eKY{)vm7}qrnk&wkALOOBreOg7+
zDyzCjI3pNI21&MOgkaFxGQlE*b;Q11KJPuU_a2aStV1Tx0Ju$3K-4@J6-5w1xgzI1
zm8XKDay<C)V=8IcyLOlMpO#v!vs%Yif1Rz`s;yPqdRtpN4d6CM$aNn-7>}|FvoInb
z(OjPic3>42?mCR5-T@euP#<IbcJJPoCKeeIDKbPtp(M?P7AzDIc}US1nb~_FYf46=
zLWw9VN!U79=gKz_@W=f37NF60%fOAR0Vg<X@zHnrLvvbZSGl=&z;<BN6Dos7mN7^d
zYB>a#%Pac@@E-ufw&~g5@#!>+ZDadSRqG!}5P<{;Fo>ck5~7kNIV>foBwdQ~UGC)j
zDS4;Ap!(;cvy1Mw{MkGGpRuRka7DNNvAS@mF1tzT!*t0E{?;aX-A~pb3#&Mij337F
z2j}HDoB&Vx7rx~1Wbu-wP@n;;@JB9Ug(kkaX5p~R)bfq9on+dj4&NmZ%l@Slzpwed
zRyCk-FRiB#qrFUNso32`m_K|5+31+_S5Ar9!d5W{=*QeMElVXR+{1n-O;!o8>G_5L
z|Ip6=G7GFrlfWi39iZ*Uxt@9<ETov?bbvqMGAen>tTqQ>(;W{i^Q=XJq&95Z*i>1K
z*<*Q!OhC4ItQyc;x*zfq-sOSFIE#nXJw5uoCvvh7QUWayHJ)IQ=wjuUWg$4!jC_l%
zktqLIf9(s7`b~c)+-5?*@xY^6aao}uM(w*Dj=e^Zh#?knC5_e6@@!8LhEW4rlu+Jq
zA;g!<pT9@9g`96P1wZOF;FO6YWCl7R%*~a?LIMf^K)pX~`elP>==KM7d9<ZkoJP3P
ze^8|h1*)ivM=FFOr)W)fr!d0fJpQ9DUG-0Tl@oTtV}Ul_Y3jIHvJMoTq5!up5xC{4
z-8LW`W`gJYwO9M^JGlXV^FpXsMwt8!_)MH3xB;SeU84%9b}FL^PBF+YGnGEl*2&t_
zebWxW%Q(fFnceL^Ell9OPy0i;C}9s+<NmO#KJ7ixm4i&eH@Lcy7I5#eXLe4QIe89N
z_lkXcpfXq=auEGALgvsx7JmOfU#(1W>!Cg<`F}tNGLGl3%@l+jkj>6^eQzoo<Usw$
zssZXh)D9andv@6vta0sREerKO8vH<WjpOXJOeh~D#s-l#$P|Q1GNs{$qyK--FPk@j
zD9qCF!)npF*ZW@5ym$Kw$)_gG!2jG1wjJQ48E86PV(6bmQoxRk0k*8n3gKB;v2ul9
zzjs+lNBXJ@Sr;6jSt6t~1Bj^+(^FSVXLrkK`eN@oMU+@<UDV7HIV6^l*%Dx$e|~9O
zKlg0^Y7?PQW)g&oa2T2TLwoe07UfyDNO-;)w(LHNCL4o9BHV0tIGrw}U+eVA%U#1I
zV<kL%-gF;!H<{4(NK@!mVgYx-K)Y$q`QLu^=D8qr6nhaBc*?T1RryM0=5u_v|Bu|0
zEc-QgU&G)oVy8iZ1hHa8ZiCq8f3Gp^KN!VpUNfcHqAMb9aHB-I(tUsCFJs#CT{qqw
zx4UdzC}MzsK@<IdAO2?le_{WQ{#Bl>7*N^<N+9u>D1qmHwpkxg5y5OokaWfBw>?2V
z)WASAxkV4y7D!?(&}#qzPy<h81W0&6@T((?nFGXv0EnFc@o+necy_@{UNWkYSCcLJ
z@p<oZg;#L3S7WQZh(8L7@5^(bMOzgSB3UDZn5i$pAOJ2*@e7{0y>1)gZ{H8>0^<H{
zaSsS2@Q2gsaZ#FJh~MrY;)x?XL78UGk580vK^GRgszzv_y7<{QiDQ^1e$;eOP!T2o
zeD?X_bO&_mH49a=MCo!AVPGtCGcX}qu=T^e<wt$?(|<jNOr$&J_*l-`p?id!@y+JI
z6fZ{sEjjZ=Z^Rg+L=S!RQ)V4Q6i`XAn=B78OqGzUEa8Vlo**-B;)pLy%W%AoHlkem
z`!N|M{puzt=#bj?JU|KS?k7u}yU<vNYA(1R9`RzR_Wy_fQ3L9D_I~%P>b~lwNbyq5
z?Q{joRI9_Ansnf+Ls$K2G9I5Q^U-&Ibr~?G49N^1il4(+4o-UXErCdPTk#G(dniEx
zEMHR>E{P8o!(fOKY0IiunkhiAlo_Ol_0|VQEd(mmC^P0pmhe|lwS2LyBNpu-o{g%o
zFc*C!eA<>~sQt{-i@f?(LTCB!6gRozym~Z9B&!O*a6!TazymMr@C$LGLxl&u>e4|I
zqvB6gjp9id+--<JA_E!ZuD}T|B1}nH13`l!L|YYr$kqtWuaXoDpK=4Re>}taCjxGG
zAffBzKt!F-E77G&FMaewhk*&EX4aC#w<H`6HB)fOzCW85cm-}bxefsvoKWF~E~G<5
zH&a4dPBUzK4R*Mo!6%?ir>Gt=Nf~)a7z!Mn<>D9AF06|oE+q?9RqZCKa2ow0DT+uE
zou6hyj<UG1`c#A>t@|H)l$;tViaO>(yHhvbXw%}j5daG1?$qN`t%P0@8I8Zcu;(O7
zg&r-qe_ACsmQgKrffAeO5PqP>k9?QEffl<Mkg-6IGj6Fv0yypAzS9^7G$nb2l<+q8
zn9?%79H6T|hIlc-PVq{-Endd^vYbWY!{8QgyH|^3DyY&RESaFm7*R2TM}`SXNAODB
zsYO7%BinT~x|EWdF&kJpRBF(!*Puv=G8JHehvZ3mhG^!qUD>z$eYb)SRjFHD?U=TG
z49I?tY1PC{y;ASiXZ8KO56eVFM|A8d6KSD5<tuUt3{;fN1H$D|w=F|hv{!5Qk?IUY
zqP^9P;%65-bW1FwlgD)P=!@P)PYqbwv@wzfbu>itrilOzzFwxl9^mKB#w3IVrN>A*
z(vAXA3+8iYxq~1a)NY#G1{k_?{7+PU4-hRT=bpJw(J-=c^78TPM-?k|et^M(W%kSn
z;kgK5g`_a;u-u{iwo8>#N-34`$mzDitCN<g%DAT{S6RIkvqdc)D@m!0#qO4_Bv8&^
z_ekZL?vd9j1;~|7`ZCi`_?xUL)ZXFk=;X}d^7w-MYeJD&B9+NO1r)1cusA${NFr0H
znKv3hXE0f88(TY|G5rbw82v0#Tq}r@thiK7H;8H34t0CHKL2=LIaI-cTTG$)`2TJK
zH5P$^>W{&LIxNB+mPw%+trd+o-l~tr&9#D3{-ceBjf0DaPe4dQN=8mWNkvUVOGnQj
zPriaEJ&hryqGrU@stwx?JLaT4Or?=VjCQVyuCAeL>gd`Ax-LQ2H_;6(bYmOc)Im3=
z>=aSMOl(4wn2hhj{DR0U;$$*fdy817H0g)Ik!EtcUn{s{J$>?m9c7|)yu9A3R|sEg
z6{b4S9xst{fb7EZ@YBbxME9{fNqy`|;RdpYiR5<W^8tM*q<;g&Bc&>tcbk<DI$}tK
zq$Z4+0>BId1dic~zf{&~=pV+ABY~z)D3<R=`#8plBm@O+yw3-G$VU)}#$h<Lt8P>>
zt6a8+!TTL_#=3S{sCzyAJY2S)*Qi|)MO&vWoY2s-2qlv4M%l<+uEC9AW5!JoYi}~7
zZ<A9jtnh%LzzZLU?(U)CtPts{B1KOskUXCd19+(mfVJ<TqbQ0IV7J|=umFocfZlr-
zi09$)XIEcdZvegw_$WfPlz=1(+zj+{0DvW52mpfsfx$EjJXJsJW8v5Wj^VV{IH6jr
zT<|E|@P=AeXz<~IE-yYo_Y3tOx^@EE2<jjt>}?|bq+OH_UFsyP$0LmXB`T&ay2a^{
za6_^mnNq}L6el6)N%{UMrKmI+WwI(cr`kW{y`Zlj6!m+_K;J8?QL%!mA@zEndf(|c
z%78J|SYg5kP4=xJQ`Ai}Y$qc*sjDvfbTdc54nP_q$Fv860aBhQr{(V%Mpz~|W_Xq?
zu&yGKB4oBkg?&*=)|8@iU?{^>;TGp)bIHNw##5Pn)nh=YhKN_OC+de(W6m`7xfhzd
zl8;K8Y8vgi)zNma9)L5Zn9wmTX2vk*Snw=G2c-=uf<aVJJ!59!5Ei5j<v8tz31C;^
zoL5_dI-;het7Pq|=v=DqAPrp~!@y0$(=x&~!7-ETp84=C39PhY4Qt%M`?ic7l0Aw8
z+K-sm3CkJBg&5az-bmajdhn$B3wXTgd`yG}@-<SR$qP02iy|!)Yo$axr8>M!XFn^)
zhJmXJD)GI_n|@O5t!iL;Ppyv`_NgDerq1W9>V4@4h_8K*<;OsSpEUZVCU5q;@3i<!
zD}mZ*C(Ip0eXA3;E;xGO>I>efH~9Mdy1hCG-Wuke5r1#AJRgkdBcF`(*<}7=dbz&x
z%?#ho`VVu<@zXrNEb`m3{`g-j+v%TG{#!>7APl)EY<1|0pgWSDDD=i2eWhK091O(C
zVB8Ev_2GD|HY4%o(fIa^F~+0&gn>P6(jZfYn6}MlY~L@<+QFQi%o}0Bs4v>Jr!3jS
zvW8Yz)%@?VHRF5IIx4Xz?D-)8ad!g;Apk&>07bFU`^QeahsX?^`|%|L$1v-*)&)u?
zF7Dr_E?Pj%pyo5opAGqp20ih@F{%opt)r`_Z^gjS$f`B#Hr!Jgo7l8v+m2m(_8mC%
zLqGE)pL*-M0V61eb9sD<76?URHHo@ZL#8QLFj`9Lx|21i)&oEbUiP8Sg=HzCQmtw=
zSn3J&88v7qf*hofjiSn`q`V4h=%l+Yy3(b?kYraZa5dJP#;0|G!hvt(uB}9S=%rUu
zvxc7xiopWkcSAb>9!`d1m02WGlBZb*?s3RlexCQL?cQdMwR8#<v9RhFo<q)Kr&abV
zv0sJT=G)7rK4*O8DXB)7Dcx-ItP=sCOZ@NSJ|zoPNMREmVHdtiS3-n}hjeTKfEzu8
z!x{i+sF+ZpO0O-sZ<MiSTj@GF5|ivrLpXV)Qm8_&Om|kXYHb;>tjyuLJ{glgrBgGk
zX`2V;v3X(Mkso-*Z+t7m%KuWD0u^esm@sES6J5+>!ki^EJQ4sC+bc8HY&r6(S(^^m
ziYut7doZt!vBIIkiNblUP^;AXwN0(O=4oC|2ik{z%>Q$9KmVewbpU8ER84ZKwo$$w
znQnzk+%c6EcJ42PEmR%>=^MSdHqk?$l3@%yar15S@H{!M&ijbR*fCe+D1m0(tpsMa
zhC@XC{Y8@&?JgCQRTj*qpS8Iok{^FjwZT+t)?;)lPf9o}G32=XbZ0if6rB2a5GyAW
zx$ZyAjj0#Vy$uj9Mj)KBl7dvX^}O-SSZ0w0gWzYHNyY_Zf(P8gPk#l+4IuhjYuy0R
zBbgW0CT)oD(eiHO?NV{dBi|$UBi<wQqs#|qAKrPm!EpgtR05X&T78iKO?gQQ06@g5
zRF&$ECyu3>)+Vw!W&nt}o@m|xl4W#MMs4)Ra%?7G_RX0I1rQ=%oI0g+pnqV@RkxK?
z2Df{SguABXvfCAdYU>BvKHmd7Y+h@$bEeCWGAkAIRF^d^aU9U5)FN(L77m{h_>Nat
zP_52>^Otlg3FVmIxi5kJu%l{Ks@7~PnKL9aLfXGVm1k*eo`E%5o2;~(?2%Tg%~_ov
zeDFPCwN+LftD5Tjo3PRP>}oU!=!EPU88uTn11ID5m>Nsr|L#>n&Y`|zPIt(qj=SF@
zGZX2rOJ9!-)NNKhmNafo{T4KAX%pr)U~P-mv|xS9*0p2<6f^-S41ypKT45n%BXmGR
z1Rmmu_K_54KWPVTae$0NwmNK^Beom0(@{I5IcAU3vW%1SB^B2&T*q`1+YKzYaNNar
z2hVL>4@o^Bai7>P3jD<2F{KysJ*W4xJZ}~IL#dDc;qo7kf4O}H?<J$ACfw_LXJLIy
z;5%nx*>dd)$^Wuxy-}?(YjswOY;xk@6nOu@o?d19&Rx4#4DDUAXm|%YNwd5tD_b{h
z*AL?~FYC4+=XF1CJdsSrEZgBUF1N?)^8<)5p^OWuv~gU|55nlHVud;gBj{j+cUk~+
zaC1doC+KyvY4uBj(v@RVzH^=`@3oJ*XZ^eJ)AZ}I=9n6EhDtMeIsflhjfTaw*+qNT
z(Q~f4R9|ae4d3i{)q8et6nV$S+ff%96V@4Bo#XE--JRFjS-wu7bjpps6xwMmofO`P
z#f@3kq~%Rn*^HrPt!mDSrfnkuoi@NEMW-mcEY}pOt7yJKcLl|H1YaY$D90sa7ZCkO
z;~AY-OnznY8>`<Hdd=)F<^ELWZxudqX31J{g%xG1D}M{hie7d(C6?PLmtT+-z3Tf1
zqj_4c^=`S3`>|0jLs$`D{J$M;EGp6As>HfY+{Q9+;1&2fpjs@5Tmh!@0FyVsbQNI6
z0>=nc?vWNySlfbg)BKr4k&WYI%?2~?3*ZEgKC0LP6%1o`VWvsP6kFgz3zRd9Er^0K
z*A&UQE%Wh)v1BZ(-Al&rjZ=<2p(&GVN|NBjrACg?#MZN*8s2uy=0tI1C*x@}>ctkM
zjNlSFfz*vtg@d0;;ml7Ow3NbH+72SMX@2v3^UE`^=7g*fF->E(dHKxQyu4*I8;xQM
zHk1^j|AmDPylQr#Eyp<y35>0RF~QBy_$@_@;%4*9!XmQ7>p)ctmwSaJgq#q!AgykZ
zoNdvUH>ESa3|3?g2gtQ<M9juau?1&g?bi9(nTTmNijDyw2sq$^kMgIEtq-vc0ATwE
z0N(;oz7_x@mjWR02cV6B|K3Ir003Yx7_{U7h#(^TL_CrlXcsZfqo^ZZwbmK9RA&R1
zSrNd7##ytkC;P86Y_Mt4CPA8=(yK_eIM*TCmSPY~1tOb6<6M_yReaZGJ#wNxYYzMV
zo>QROP&|c2Wa|aWAHTK;uzveO0b#0x<5+DsTAqeGKqB+mQrAm;*w-QZx)OYy5#KCX
zBlBIuqana-vIO$44#7yO8?@|`=rE&6h_0h}FImi56<Q-{joomF;+zN^$!V-&jK<8c
zTI@V)NA24)OE&J#hEKI2>n*9nkTR%r*ElQd#gk^UtuMc>$o~*<`L8*eT~#)it8%c-
zr(h~yAT`(le5ex|l>WC|q?$t+*igy5vn+frlAxl5;J6OpGn2V2g#!?*mF3mZ#R9tp
z7QDE75E0?6=#yG2j2L;XMNMPO`+L$71Tl%Zz{~nhWva(i7skZd36glByqGM+A>T}C
zK`TXLrIdtR6tmqs;jer^Zw~e4Ug<QDB6DKR?Lc(L+drGK>_Mr`iqvTI^TCjrriLfx
z4I(|etMXRt+Xgh2r+Bf0op=kf09hi3vZg*e*8}$OFFYC=Tn`Hc`YEs<Y@&T4Y+B|V
zBLD4Z&yUNX$lx+Ah=!Hd`wMhU6k~7>NfIJ^CeLeMw;WkTyT3(Oo3b=UItc|kCdX%e
z-5G;<J-w2=_7jH<BWb-zS$Dnc40b1{cMRI!+)o?#S*ETNe9vyUcC>2?n}NEB4Vx||
zt~HhnkHtR@j{W4kuDjjMO>WOA=g-=PB$SwF%wj?5DJv-#g_gsCzC>E|&zWUiN(wl}
z>qyLEhM6KlABX*sj|fhC^<Oe6xRY9<G;D%c1gTngci5mn_LQ`O31w%E0az-{+FLw&
zFFr}gdHPZU=18j&>jCW2(P1GVD-<<!EKA?x*bHOso0cJittB&&AVG8JmLZS)t`oc^
zDoMQ?j>|<atY_)ExCFf}Tgv-JWwD6EiUl&^t4}e5+U<my4YEMA=A3fhEPdE|cOgCp
zLF++|S44Kq+ZpYG@+JIa^A*T>b!$c<uf<d!-vy{EF&q6@ZNRRDHJHSoCAJr-z3QoQ
z4GexEz>4o<{WnSsHw_fW;tDlZzivvL_-AJh+3sb<aF~s0vEXzvGnr+)BSZQY$VL|K
zcJmpc{(T&(csw(P8tHO@6=^^zSx|V#o|}#%nb%>`WA%Ci1wtASI8%mMDA^+nqgjfa
zGzD$A0WP^=9p34Mw|FcgMrQwb-m5B6(0mBUlR<B~U&i6D@F>DoXt`p@rPtKAO#?!^
z2mlR6uX!GjZthcJG2oczHe+n7lC6$4+i;^TG^k)zN(T~q6GE7#qdf^xN^2TX$5Opq
z?Zub+L&rUTLdA~j={vXOy;XiUiDyG_tKA4C?Kat))n0pt4(_9);}_5-daTDLn!eI2
zN#UipiyR9trG)YQL*fS9NRF<ew5Gl>Q$DQHVT40LN3BuYn9*WS4kcYyOS^5$<-X<C
zjzLiLBdzM61#>T>p=uB-7+<r+wz7eBpoajb;&l1|^XT;_<GG9%fTev+jzZ*ux$s)a
zE8{g9n8yiRebZhmMv?pW-BbnZ4FkJ9^G@uGujTt31#}~wn<t4en%>jd3z}gv5Co~U
zmTw_Y<|Q^=@1bpsVO2%lt&~k^0wLN5cvdaAH=UqiHP<j%T_7**?NMck4y|qr&rVTD
z#+>pk`=TN(D@p9Kb-&Ze{KPgn&-LS)7r+YVOJvO6U}UR=om|aff#JZlbhoFAE+4%&
zy>9hd>PSa{>@dUM%fMThi4nPRc$rmssj(p+v)7kum!8>6wAqx&(7sw^10@=0>JCii
zl>U&OmfR}iJ<^XYB}aLs@V<OhD(Z^h@26_w4f<}*Hum`H{ZT2(fl3_7JbQeXLOtc~
zRKmi2k=@<46G;>b!COY}e!+rUVTr(mJ59FTCc~`o8M!g)wK-^$GA=V1McKKvomrp0
z&reR4y<Mr7uupX((M0=EIx807X*$u4txDS~R%kQ2iA1L$`q)EKiy@jNi$b@n#wgpe
zW7W%3LP#n%9mNMn=YZr}%%4F>%hQEtvzaX0q4bBN9tFlW*w5|R8JuPk6D|`Vqr4SX
zU%+wnPJDf?$OM=grR+r-EA%V&_kx@<?N{n6Q?_5SWyg^}W{UM7mo|I5i*$<`a9>Lf
zb4iHgAIeNwA{3L{miL6sWPyidQKmIzGWR`UL%7Wh6Nj(p6xL(jX4ae_ad>02HA+MB
zd;0!9gemqj0BHO-|Iz=8|LjZ>RBKM|vV_rSLN-%hpIo+nAFo@-hf8W-pIGZ!$nSrl
zxB<_K9w%FrUbKt-xnFOKEu9@^_J0vD@C}Q&T0G4<D~GYgg~-c9U@=hh%wQa0o*~Wo
z`|?sDgJR+OrZG&S;a1l*v>2l-AWPqVbYVG$!CiZ47_Aj+R#a_~3(e>%<v!fXsrNv1
zn~1}&7qsouC7a2sAN`qTrKi`W?M&V|Q)}5KZeTJu0)LH*RtZC+tI!nB$~d&X8V5Cm
zgec?k5Iq4l(<53(-KzO%EXkFL;NBia(*dI)f{vK_y1{V}Go;}=$}N@L$We+*m4_4y
zC=&-Ok1D0Imn>8Yo7j(|a|;s#1+NZ6gDSPGhsDba0xW5kou+Tg4QPC8#wOaJ(n45Z
z2)rbF?I9?^?%`AgKdVvUb(C0kWp&ZKUJp^;Gl%>z>H8gYM|;^z^^(1!kL2(L$U}4}
zjI!8&EAT~YW&<t3iC#aXU(32mmJs5grFzE($`~!_o-C4c%u^LZ|ILB}8>1B#zBPuI
z-m-&tQV;eb!`{4A50Sc(odw&4_y*`fOKdMqAGSX%NSSuUwR3|E#re7~9H{vCW1D?S
z{K)E}tvAa1fF%_FQn0RB(9KO6<l{P$<`AS`7;vRTmSCmnQqzeNucFRD+;;2fyi&@0
z4v@F<QbIQLcjeN|+cUaBnA)*T`;>)uM`e{UdP&?hN_tF>v6<v-VYGYr8V3aZ;?<oH
zMg0NNG$Qb39;$fb`nQ{~nX?eXbrt=1*ueDCe~+%J!_Ru=`#RrWAM#yc{)?P(NS2N$
z*q+TZkbGZsmW}KcfrbrrOH|5~lsyy$`0zP@^{1^xY7}Daw!Tyk_<}Q~&$?yDQ;yaF
zK#sHPLfdzEsW`!XxY5}STJNh4=$7lsK-Jy--{zvwH453L)K5+|${JcXm0M<av`G`R
zZ^s*r9Ht*2P?bA}iVnhlfC0;0WCj+dZ9gxX&8Shpi}IqoZf`&4@@O<zhZSm(Edc4*
z>RRfUZ=p;-PQ&OkHY)1hXDw<^Fp*-DlN{;v(S#y&xDMHf<${zCBSjxdG$SSZjk{qo
zZ@GFQvYMCCbv`7srGr3&+K(ecHv%yUJR2_xT#2{a{1(rzj5h;RR7hNP8?t`LkVpBE
z2|DRKPUKB!<~D1;j-vn-hJX1WSb!tr4dKy!m)@||d>_*`=mBuSxv^p_+<8q*!on9-
zr*+m(ROl?5mEgP1VHMV3mUN2G_u?C5%5;{y9bwa~`^^4be<!?OTbmbBOahS1gw?LH
zzE@cgf%mCq>afZ~ESK3m8}N+xw8t22rSb<bIMdSu6o}7}VSF7pYEf>Bm}B#zK`(w#
z7qpe>iO0dp<55Wd=*VG<f59E*g_WD?GoKm47`Q^-A=6<Y79RH~Y|=iM<ib)a12JVZ
zZQ*#>$z*uflh~HXRRBaijdC_j)H>w%+M!fa;Kyr?t2o?7jt-{hdB$7yn7LwI_}M%a
z29*Fdra}(XzD5G-?(Z2eu@CcsQ1-|BK0HJ+DkAQ8(T}e3oAxrgNcEVeDseZi!;4lo
zNB-z~;oy8&)n>Bh>cse0H&3u?8}Z7^0Km8(#)PDdfAQ6%B^CBmFk-F$L>VVX9DD%6
z753MwKWmM}V4<Jwv4+d1<pV?ubOYyMRkM$lltN7O4wW^}ZDE@Mk)o2H=Ybqs&oE8)
z6e<z);udjNx3BGJszHu|pGVio>!rVCCQnCt?*eWK`YWr}JyScFB;P*DnjdP32dH`3
z&wL_=@*t+R8;qw!nfw#9X<u@(R=kAoP^3zyz@EXmypb60B3K+YW4SRDxGa`<qh1HO
zo`NdXIw!|QeIrE0<tF|EhK~JX#r7?*oEd^(b>n}Z$<rwZZ&=xJPceg|z)*?lN)9NU
z^tHR=Qf^nwChn>J+aaNn%mkXNnMqYV_p0v8KD98BDhOoY1<L!^u?bK99A{9#tAV*P
z(r+WS%8Zc2C9HkSz3OjvtNo@yWQZ*)Wl4uk={CsQW=|fA17ll71J^HaLVxmcyUxG1
z7#>ro=)>I6Tv5}j+8gBf1J20L*+9MKuD$(gyhh5Z^!9$G`<FWz?$?3cXOlb2ZoTn&
zG=C?-&045Gt??3s$W^=J#>G_C152PUbBx`{v3L?(7L>l^_!{&!A(pkxhSe>^W8Sqe
z`-KuEBAM+gV<EqOCyE=%@F*4(p%Cl1M-GDMW#{v8>Hm3OsjW8ZrL3j$Mxl`V)#2T3
z3G@M`r}M6Kl1St-Y+I%|<|C(t<0l_OnS^)YY%+t!0An?vG~{e1H{vHE41;R(4P#Vk
z{+-*!5!hJ=yF&>4uOC_ph72;p5`_VFhn){W%PNU05GNqDPQY}?OsxVryEz8Y19hFl
z;51;SgU!ilz9SE7C_!$8z5@Y><uv?3)zL~t^Q@?>u)OgGrc)81HPE2GxFPHyl*Cxl
z5O3jPRLy&WE(J=oK4f7a4$Zs#6iqm@<*lZX{d$uFauhlmBJX&VeVrzIkF};Dn?cwd
zr$3$<bW%LvL5RJ8fi#gDDplQyu8jFX+2UCtln`N_WKK%I+*?<-7suiPN1`VZPdlla
zLs^DrdAHx3*DeP_GA*w4=09cOkeaw>R4M1~N@kOCUHM0fX<ty<bI;wj-hkqdpSXu*
zVuOnOI&q8esDm2C5OmP<$qZmSMqOZkA&2o1aizoObXH?eP{T2kYw}^;W6vvfrAeyC
z3Oj~djT|;)W{RN1?S^tL`m#-&|HXxECGIy@LkFIL86SP%qn7B~Bwh7bAe=QlBB+*x
zBd#+3m?f9DMg?O6Xb|rfE+YW<4nGKY@~&a6oIgq&2HbmKY?&R%2@bpGeBCF0Tovp1
zix|_oVcx-Qh1;+_qA%{-yXXNgwINM;smN>y!-_IB95>5Wv1-5}LkJ}oNg_XpVay6h
z=b$NR8!#!eTl9SR#4N!Fl4*f9#T?k#$`Aq7F+fQ1`y6i=IeNiif+7-dF^e@PbdD7+
zVxLGleil4Vh6b2>1PS8zU+$B9uFw+Px&D3&PihxoM%jUg9z+X(h{t@F?;m_)5S<2&
zl=P1tqDJ)Q_}qhUSvX|~#b3rR(6rK-JZ!<oeVMDAr!W_;G-shs_u}<s<M6i330_AK
zq0d>bCH+#+;ADm!4xk3O2)g0Ox7oF814BJ{+f*))JV1m$%=<7v%fm(=&|m;jnJaU&
z**#B&Vh$1HbX?^ea#B8N<?sj2YVR)6rgc@DOhsgCnb4(dP^JD1c^I57iT6?2M`;*R
z3?sNT0Y}{8p?4co0^2Ox3Qn@)Ptr$JB~CGDP%ycr8<Hq-%OjihjORe>1e953{pAG_
zjN(M})e28EcvEi?*C##d%B*1}e3pZcZaS5692)sHOWdc!ShMG1d&Y8d#iW}Qe0sx>
z2$E$unPvA=(E!+o`ICiC{My_~fo8k4qpe&H7+u>quQ9HcW~94CYrMSfQfm0M(NV@y
z?Uj!Vw`;3vt~{QMkLGZiR$}i|)%-FRr-0J>*9qh{qw;EW`gnlSQK9Al{S8X0@VHxS
znlzSR|CmM+DI?068>CWQKP>DN(`YamH*2q0?{Eg@2-dE48FEx*B*c3qd`%{@K%HF{
zFZ$n?TKJjS%^X~O42BcU`QY-6dlPQLj?roT*$pH7^R@0bEIrHUEc%zKJAAttGn)|o
z=&zu)N8Hc&iP(s9&=3FN$0X&rMAm1To|Cdg3g7=}%nx7FHN&@~PAI0xm|jOWuaYIs
zKOij3<Wn+f|6_4R%gbJJ?o==_%J1Zq-cjH)t{=*Dnvk(QeR=jy1l~|D+fwaydOM?j
z#M?XS@6?@refIsGl7cIj{bKbs;x+hjTVZt@t2#7Q7sF3ArI#`tens%~=a7DG5Gly)
z$je-sS^Swv>gd7|*OUlDvR@c^s`_};RjJ}6`2FGHefVQKJd<JeXXt51@^@*pBX6Vn
zmU6PDzK&OiBkqxuru^Me@`L>$9+d^;CY3{aol!6zW9^JNn{i`aO7+Koc3P$SQHzZ4
zDHViyZ!Sq=`GTil>|?Eh?$E`b{!ScJu6EVn7Zs?kT#??Po>5fm>+nGNvJemb(fBq)
zrnpgub8bbVaE6gmXnLr(nYTSYi@1m##0eD{BDJUf-t{jkjHFERpQ!jLWOFgMM)Mbr
zAn0mQ-?;ITg2}^|J~bYC>!7%}?_M`8a$Zoz&J8!i?xB<$kR*@w8{r^Zw=@Qqi_dQY
zd4Hh=7SrLY8y;g?_k_iSx|)Rs8!C)P)UqUPsYw=NNK>WbQTX@T6t1*=FR?4aw0P6~
zR#oc7B3#S5Sj<+=i6*&iUw)*-b^J?W1}{!#F_Ta?!lm+!&6{2H_tHv5(Rg<)GofR_
zHga_J<j40DgPr6OQB9Mc&Z{U$N-4qfoRPAy-0b+qY(a@s=;j2buE7p~+08`J7AzEo
z?DAsd>5-O6gl3r){Lq21dZuj(3Lb{$)u%cOT2xhdg598_bE?x-EnaUp4qwFY^F37W
zhIsk;<16tK<h!lU3M0*%S6uwWL$a^)`LXJMYm$iZoBS^?WmF%~CxV*lr5n^V*VFtz
z`v0@||Nra1XHk#as8MbIi}GD{{(k?{x03^#7}Z_=FX&GH|C|4w{oP@~tx`Nxp{5o<
zUMYU#T0tpLh6Py3qgJe%1^xQH1^=Vt08iuoey8#n+Q@d+lL3O6@Rk0|UZ)}PXiQ#8
zL)q!XCeOp|@>sH{jpl#}i%$Pbjh&qYGV-S%6F-nx+2`j(F0cT$HQ%r2QWixSspfL#
zYyZblT~f_?nL|>89v9Gf277_}i|6)?+_#L}QN4=q_i`+iK8NEyzdTDzx{Q_=)kxyg
zAJ1KG3o)peJc0^AmB2i*xwW&4c`BwM#2x?!yv2#$uK>yBM8H?)5?MSc*U05bXUd)i
zxNwF{gMpJv5Yc#9T+E6^cuCl-Ut`S_hI32NB6)^OAjoiu<RxgX(Is8&q+ib_uz6y(
zmBo{ANi{)btX0)AX=BQU8+dl$FvA69XptQ06bLh1Lb(miW))`&q{uQ3&Ktp!%Y<YZ
zrctFj?q%CfGQ?6=$jeX`PdroO9_H-y>@=qb(cI|$o-ez<1SQx37kol`!Jux5qBMHR
zj8uHuWypOL0)t))x<k!l<KdFUc}cmn4T<n7RB5?C#+t37CZb3e<!`2V3Gnt%Y|;Zn
z8$%oar==qss-Yadj)a|Kn3^}X<vzlsU>@a?SL?hbHl?|@r$=3Z{asc-&lB;*JZFeO
z#o`s$4)g_T``cgqaybvVAgYANRqe;#!WHu_)o_+6<k~Nb3a=ke#vz(4qF_+B>`FsA
zzs_hR`GcCL0WK(1!4v8nZiha=ZK{Z8e-)8XO!8V2Q4ULj$%(7Sn)usp&TbEGWY}}j
z<#65!y6-6#3FLas`50Gfp&eeRdm1<l))e&i4NYDCk+pW~@0Z2`E~R^G%2rQi-!{)B
zB^u~ef)Aa>W_?!;I-8Nh&MnAqs3{{3Btw5qZ}BcsTauH;x<d;L*qb$mT6{RS#IzJR
zrUoa5UBMJe#QEZl$J~sgR2Cet_hAH;J7jzITqej=9+3x*#vR}UJb*<IZ+~S+7OT_C
z5liX&k9dywu;G?&8bKp{XGAl}Bgj<O2jlb)Z9g2v6m(j=+G=cqL|v%LWI6e%tse@e
z1G7)a5lp_{)|jV}>jiPEJeBEd5Ir4ZH$68wH$Co!wl)bDr^XKe12vD22M@{~uuq1p
zaBQ%d$Ae`N!L|5Zq4HgAvR$EsU6tbC^;~}_&%fTEaT+xG-wM7hM;QVD-rUiTr^8n&
zs>rGk0LXS65AOf7OeSZD9G-4W)wTcU8op(034v6Rsve)4;W@hhEMj;}XUGSy<s6)K
zZhY*kaX1U2OWo|w8-MTeIxReovJCW_%Ag{oDx?}%&h)1aW;F$wTg`OeHInJ*9Aplf
zZn{#>51-dRxo>a(iFq^wY$;l{vjbU`yuT?`#;LM(+Bj9R)TaH(RUOEkXxUN&OsX2c
zKX=xVUAt!;o_qgVwMtfM>nfI&s`6fFO%vmFj4^y2{xqYm&@^$ePCkaNLlg8{%OSH?
zmm-ngx`|zf`zB}846vD*vK<{rPw0MQs+d`2Ya&k9-1#??#g@?6ePti&+uE1mrw9WR
z0~u2pLv}+3=Y~HwI8Us+_K)#@a?&+hjHy&tlPk*TPhRjMm-vAlcwHxq&sCM(Y~C{*
zaDL13^D7^qpRHKwa!(8x=ymi_3FuPjVC7(CDCm;ZH}OR|AOr!+Cw$E(K7`F)nb<To
z_o4zHE_0DZcKB?o1wJ^!iIUCjJGp0%_Gyk6_36xX6|uv{s8T2xEdw)Fnm-X*S=-`v
z)8nWd3~Q^UKghx4JfPQFg2(1)gc?DXH`aA*trW&uAL5}EjC?yLNw4D8fakGZQ7w*G
znYMIRWS-X}?|^^MgbKIy@_mMaIvbA0`-6)6G*=y79$ny7yJvb^jO|pOuHc{^L-#qd
z|6Z$_Zi_B-8FG>NVorIwt4?p|H{7hl)bUS`AIZ{dc3?0&H2SO~10$N4daAl4Q>)cr
z;QX#pGorYDYJ91+oboV)`mnHk-l=h!m^9v;une*-RKChDUlj`3mawdzG`@>JAHtvC
zp;@V&&+CE+y7cqsPgQ+cFeGf_b%yXeC2jLhn>t>7{t+zC-yv8WB3xXsW6_TjpPTHz
zT5J#U>K?ZgKd3UbHM&~GskG0qZ#>mTUilSjPkdth9u#lSrDCU6r{UF<bin?K{M?2Q
z+5cSZu-Be`6Pk3SK(_blrM%%<JLa|j@u@z#ffz<<p)l89X~u0@e9e@~U|Hc~a4iv^
zqRbJAxTUN8rKYG6;U-K7ehG!F#IH7Mp_doSV~F&kq~9|nIHW}xItqGj>++*XXswK1
zp30-><Byzzlxyb6=scc?kqaTqDI!Jvys|mcrqr`rt~x3mbIh;r=^<qxr6~Fr$mmh^
zujvGR+G<1rjh0a;5tXBaxw`ljr~Yzij+|wGlg1N>*bwG_971`JH!^w7rb1FU-+>V;
zGTcHfah9r~xNz$5D`&pSM7E7KuQ*zV$RKFbs?+l*bX1{4=*Zwy@*Z)fbAIKr|D#WR
zkIHuZW%xVL^y-!GjZ%pZ7nH%pbflUf*O{I`jG{b#cDm%t!elnPF!{@LNj6`UXM{Y_
zndAhdT1+p4^F7>WIq~G%Jl<_`e9kitpSg?>iC@ZOF2zR@mUUo{G$;IMZ9$#zT8)={
zj=tq^(fxG0ZiMG3(i4)DSRUK;?r($wi@8RWCu1{@Jo(ga%Q*l<pw=S5QC&IT){0F`
zJs8O*gvITsP@1D<;J&pB3Y;PR`CPa=?T=p;9s1jb1W{VYPB%6dG*V<$@O3pU$n_$J
zBU>P}qq*W#8?CjR7qaOYhAXXXY?M}VG0!)Kq*ydsS=K~Pyhl5e6I)7<s8NhY;TSw7
zeQA0O91+14aU%jt-}W$ipxjTjPimq6)E3o7aF6fg@nN({1ROC*TM<I5K<LOsY+@uD
z1s|frH)V#%<ZMg{fg<aEwRvwpqaxw$9#$(0PI_9-VD5!~?&kW;a(KE`8}8Ds*3}4O
z8eGD45+-_AaG@p93C~BbkpIW6?p%>4QRcx@;qxeSVdC!z=E6dk@t;y(zS7(?R8#6v
zgo;Q=lVyn0>G}S3`XC#P7fH0?a#rrD#V=;p=~B>{0)6G(dkRoSt)@%%Zw|?VF@_p3
z$ZFA+j0CO~SooEht-^BQ1{N;vT?-6rOIa<5!Gw(kic~rreMDR0?%8?$%%W>IuN7fe
ze%yZ8gjUSs6%r^Z+y7(E51xwVwH6i?w(_E<g6A{;+nz!p6!H!gO|r7O-U9^)-10&2
z;Bp+|G4TDZEY@U^GWw&K6aBpS75yz*Ghvk@^M2g16Yp4G5nxQ)3RMMVg@wdb1@cNA
zVhTgTOd;Isz!UFTi+3NYaIe98c89}o&o5mf*BVu7ZGl|Q1ZvgkL?(w_fMGyiS+V;s
zt>#bKyAjeBN@>cZi0nTgvucM*(n=^kzwzd$5A*sP`V&4Zt!k*MPuVst`eTsl=Ws(3
znc_+t66@!F!jP*_@SCsH@p}#^{)G}5ipq$2nKE4V?+f#w1(L258`ivxodLfUjjm`v
zHT{kBo>FDp%s+gt8L%Cm|KC4y@{tcSjJdg*@o*xUla}Xwseb9D;bVk##C!{>ORppO
z1uPt4RLif^$5e**6V-|qNKnO89(4l?7p2tL&f%B~N~;R-Ns1pnWw90XPz`3sKPd<m
zQifz@q@|9jF<>RD4_}Otz|vuotSp!$T>`^A!bYV=NtHyMQc+q^ZXP@yo|hYxwmS~V
zlyO_eG+1^{wv?HYbTlOal8};0d0V*tCk^WfOY@TlZsPbr9G?L$_JWFmhLwhuKv1!F
zU~iGQ1UNDa&-j|c%%>X!@u;_%8S8HGr>nDh$~>uLbC&$eH3~vR{Oy$V6ENH|e1WmV
zSy))&G+xOU$T5qs9PA>DTdxjZm?efq<7PxV#jJ<}#)1Rmp1stK<BNH%>ksn~u7muO
zO(!K-=HUPmgw5+?K73t2S~mKm;mIzjw|UfnPfE}N61_baTgqogS2dOQXI!lIk5{u-
zg^W0A!;OEps6vsXob>ggxpx43&Hc3g&$q5fNV#>*j{H@p=Nvj$zvrJGs~S7q^gxis
z|F7!1D(L%t2Pg5tCWepRSN&bJ6#vuEDvf3pk2<8$)ZVibz~29QpFo8{Q)mPNEd>gr
z;%6J`A^k>M%CD^pAAYCsDb++;d3n`3QQO0+D5p?=fa5ZM=-l_p&)x-D&HM@G{tZNw
zUtT3xD)(3ZKy3=4HaG0s-pzU^nE#FlQ2R?N2>9Ff-fjHWY-jZA6hzAFXlM2gO&7@%
zKHV?*u-pqHQr(K*!2+|@R*vw_IIE_zp+9u4wR-o^Cb%y{gfDBF8m^8T)0}4}KPC$f
zdhF96Q0O%Wg(E#GU8%)VP^zVBcnLCKs6;Bj#r=X8Bd>10w;GxJ6FLR`6YP$ie7i!;
zd+jtyS~mJfaKV3{qF+vGlXC=ahO!JZU@&MfaPkDv18~`YId^5^l<aJAitKKqGTc&W
z&j+<FnZ+m^mZSoi$+{=c(I#tIAsJ?27>xzBI1&?qdUVl4OVd<ZN){ykK|cR4qntAx
zv$5>4mcz1@moqMB?h!zH;YOq}5foNz^yFKYC*Q+pF!z%Ajx_C1K2RRaidLh%B845v
z;+1BKq!|@_c4`SttU-}DI-1{?X5KWvU2uA$Sfi{Y4*nBmA`}bxLcwD?IG&K^+R~`X
z1@-k4OdGJ)Rr9k!R8~saX+4(r^Nt5LTBAuM3aQzxBKFPD)Mkr_E!yxoOD6a2!>*r0
zQ6-ogUR3beCv$lU`Kc%7=Z5FzC#Mv0UhYYD1^-2ZS(VU}?mVM9vOC(nJ3g>aCDV*Y
zyU!#KABb`vk_<(EK#rFw`-~aI;Ba8IQ+<MjA!X?PEKXw@yevtfK5#6D_B9&;+I7si
z78(kBS#j(xq}bO44AX-eI3z>9SZxWt=7K0t=km~&>Z<avYG;)wq$KyFA^)y2{M0~2
zOkt(gQr+Z?^g0cV=#B>mE#X14ve*1_>gz*0_^yKk2i?A%4fEDEK-M<Qwj!T9_+RY1
zc6*bv`N+hPW@q!X(+XI6vLsu;Rh9v-DGUlis)DM8ks!9kSEW~11*suwlv)ijeYz*T
zw+e7(fRIgiapoj5EI4^qh*uUmQ<W*-(Kl3n)SU>alZLpXmmLbJmWFn3;mgP_k~3-#
z`pSCJmoCz9VV-z;NR4dzEigHIni3j@1A4ZJ2@$GL)FcyLBAU$nzONnVFnThMG#QNK
z9wb3tMP#O&F>WqDj9O?M+E>VtME$Tgjc=W=s1_<mt9Z3bdHCzGgq<H28vdapG~9L$
zJF%p=xU7hcE50i^ai;^t$cZOXqdxPne6>QDOmv`h{V*IZ9xwmxvkouF(-z5W>_nyP
zN}>d{4jr6D%nC+lp4+Sj^Ud5K1Ix?+Pc{P1VgU0Oz>^F!E6Bh#^C94E*#{D{4{WQo
zB8xbHBG<Gc4zg%+GO74$b3*Oq<n+HBA1E+&j0}pMSUxd{l|iS$)D%;3VS4CgdxEvP
zZPxGLL=lapD|LD*OPxBFh?W4JJ6JNv=wz@;SVuccjd)y|Dh-Dxzr!N4-)((UcD`)G
zje%TN4GS@VARJC#G?Bi&X2Tis%1JU|ruXEj-oXKmeSl*P>O6hB8dH6H-d@1AuG02P
zmuJXm78-Z_QCjvcuS_1N5m+c(ibW8sDOcpRx1Q@eiFaSPD*0OCT1eR6cTV_O-~jW}
z4g2Tq?VX$z66x{8v<36^iB5FaJOlFKmjNc+&JkNsEJ&o=-FiJp!mD<4_VEi-I7}^z
zQj#({LJ6nkr8!};?2Mf{ea{LrC`NBvl7#<6P^zUmmIucpYg70qGh_G~qrSVe$D;J5
z=ix+SZl(fJDAnnjs0Y7+<RNu?2KNLe27-bj-R>CQ3b84u618q$kM2)0ICbDlTvb#0
zHxw&L1C|3LJ+79UcLe3mP7c((W`JOK{O>rq=V1TAQ+5LuG@cFWcFv{B(5XR|X-Wwy
zcWwos;${%7U}UbPPb<nBxC<Te-!`BeD_gxpJ@N0@p(_Pj!&YGnS*N$x%n-pFL`Ek`
zHh%t>j?RHjX@yk8>-bdZZLFyH-+&iYbi3l#v=aKuoa~&J^pZ7PQM10;G2sBk?TRjO
zl)@(Gu@&VYQIx`<)SEq;>8*P2XMcKj4k9OuZVK`+ZMhk_Hl}Bud6?#*nQ7(v&Pb)F
zy_*FLj2s^cbg+YMEjq)>>+yH{HK^Ai%%_28fyH#i>CTGK`1<IW;XP4D=Mj^32IWAr
z@A-rJA2FC~aeRQYb|)+6GFm=#@<!6}tGX}S&SA5yC8g056+65>y<2DTEZt`_U(5?<
zX|=lil7b=<t5B5wd}AKsZ-anCx*G({R_E)j{(A9;SV5LHD?MVdCB?tYL(iZUF!8`W
zAL9$qKjm;yxM!GH#FqF@w33Wni6txF3Zmrgtu^G5=1p5j-Rug7FGVGZGw&slyH9Oi
z{{|YjFmn3XSI!7YS!B(EMiCR^`#L<YX!hK>`n6CV?LuS-r~6sAEhO^7w2DkTaXS&8
ziNDXQ<K1U2>Y`pztz#$8<d&u7OIZOHyPU1#&%;T0K^S{DZpi&?YQUBx)e1DITCqI7
z1#_sCC$ZDT+6+X$LarcQ(V9}uPi7jyK;C}--dC@zjJ|bDkyywTZan5+-8nTk`tw@H
zT<C~v-V(#Mqr@_lOT^7L0S;|g0kian^^zzQz|phmV!etr8xbOYdl>nCM+X<zXTIVp
zzU<NJxaC|{fVHH2$rj$&@QVt!fZdt$4wN*SJ^Px4>EK=AEg1WDafvZF01Wgw4_5_5
z>5z&|z+mQE5%faoigVq91QQ$+we+4{&ocigcJt`g!thtG4}e$e30POs(?4|O>JRj_
zu5hJ$>$W+BiaGd%t)5Kt19D|MU0m-C_S?aK+wu4+i^E~+=C&5c;<kQ1ve@r@+Rnc$
znswRv^)Q#;jEv5jS5jlgny<@@6iGdQE|o2d*(lH|s43y4bumiO9MOdzznq399GX4x
z^noU!@P$Vs?6%76-)>w>vvDqHE;B#HIkh_9rgSmh4c}2!I-0$bjZXjl7R!HzC1;^f
zS!Aq^oh{CA2UBL@wC0%$7nu6>7IYr4_<Xk3<gXXMjul|E=(JyZEh*laa`Y+c4RZH4
zhn;E0n=VoUt0+Iq?4xPxid`TF)jg;xBECo58t2|#qjSa8!tjdnqMAkZof0QcDWNcu
z7>XQyf^*)fZ#ORz4zoNgAT8Z;!h&EhEQrZ;3(ZW&TTm&#yu2(<Vo6v9<wbUe4S7c`
zW5=hZ?LtGB5HA1ph%M2p6rm6iMZ)9eipA4=R>z!``bRT5N*o@!zi!Wlk_}xq=gz{l
zzQv0<h)|sFYz~d@LEW)moF9a>FCNTwz!-@+x#C#hgiW!pzVEuo-we7^a}0O<ot{64
zx)8ulp9NUGZp5Ju%%Q;E@wu+yuU5KNULAHplN>6IQ@fCB6ZT^jbE>CcN2a(yl^P(s
z%&j%(6pNdpYL5-i5;W#B2*fUN(yax7f^J=!mHJ3WNyq&o#_>gg=6E)r==+^H{ao&b
z!_AK&L@1wfG4%hV#Q-}%#J`PrXi0a4Tj3^)O5n3Bd;3n#tyZ(E?H~}yez40H35fpZ
zxp>6`jw#1L@pe}K53H%3sy(i+STJar3O8J=+Y1^0D<@sM&74Z}G`pkndsAtDbPH36
ziJV_V0}BQW;i(7yRYTot8nyxImjbu>F~TzGLILYf2e!H4d|8Qm9?)HUxXDE^nsP+d
zS5x{5_~_1zEpJ`i({~a`h)E}1vVTvL`^7WZqCfo&(TMasHi{fF4?edPqI52=J#}Qe
z)mI0dS}^deA$j)9!@GCQJTe<=WSm6jhcH)pMVYT`8-luua>I=4aZLwo<O&S=I%^H@
z0~Pmao;v)aJ!*5D2tAA7g7KgZ!(a`QvOnD*QIpOqZEI@^S!5f#sGl<azGd!SP>cwg
z16R{X$p%|Kf=M){0QlDREMg@t6LXMMRI~8#JJUc@9TwM*TjN@DX8A#I^Nf2!EAl4>
z0s~Y%T$z%W>-4r%4lMk$hW;Q2o%4WR<C>uej0(z25MS!Pvwz2t@VMS6N4gV&Nr{+q
zbK~EqsO@NqI;KH2(s<b%TYPm@s6{gK0*6N>ny;q<v<^oc`zSJRV3+)1AUUN40!nQ6
z^<x@$smFGyo5CR6(5`F$szu{>C;AOH4gH#guOIk+)d2ymY5E@2>F#2!e>&*w8KsND
z#7ODVc6EBQFs~dd7MEj%dFjpScIi?vGHj9K#TxaWc{^ZL28LBAWLOPVupN1Ss@J@5
z6b;Kg*&aDZ?xATEdO_UqNT7re&bFb2nv6zGOZpCFosVe@ULVrF96U{w+!C9Q1;;38
zOp*@KkXFJMAe};)4b7^c@+Dji{{)T65wYbWp2Sla43}$E9EUd+I7`OlAS`pnFbHr7
ziw32!=J(CGmOLMt2%VqY<KNs99Pj`}0_+GkcQ>NPy0LDg;BD(d-NIJS8;)OSO6smH
z`|}=;aj%(pWB~jTmo%-9xVN714f~q+8k_M=J#lZ}wAJ8`xF^IT&5V1z``&}daAP|;
zM<+k4Lekk(ID{qZEB2w&U?C~*=(*DL5sj$6mSXZY=|!Zj7oe<1RF&VdUK`wVE^Hb2
znz<9WIRHuvf&)_pzAFoAjz_56;Fb6oYB|fJK$GRVMaDaUHHmK7DeNV%p1P+S7#JB;
z%fno;iiMBnz0by&nZfgGe}Vwp3#^L(+99)2xepMaJ)*8&8yvWD`bq%!-?FR&v$Vmd
zzkAm_JrdBo8b088<T)@B;4d?$-jFs)Z=|L)VWu!miK$)?`0pk-{PLs7DxEH}>d|F5
z{M6szb?@giNh-7qC#3?}^VgOD10#n=0xi0l#=v)=4H6W4$KhjvgSa6oS)X=301Q-R
z279+wZw*nvNc+b-z#pWP#$HKyughxZ5BGTf8}7&!L;p>d;PJ_lf5p)3kB`eTTnISA
zl~KOEE;G1sSd@3C9N29`8ZmH;5y?A9+Sv$fTvies7@l81BkWxTuRtx^ozQSom1LWZ
z{$>G%Cz5eTi$_C6G7fp&%~HzhDw+A${g>R_x#8Gvw?nNY&HZ{{{dJ8X$#nR97EuJ-
zA77=_CRFVg!*C^0-dBy7YE0uRZ`AG&)O@O&`?LQIc*aLA#VmU8b5;~EaMCDjt;)fw
zbXz+R$P{P_64Bw8k8iI-ZC^>&!L2YonGDlgb#O8?sG9HsPr|<-1n!nKk8h|;F$}Q7
zYw4IYNkUn8kLg^EK^4@wO1-MH`KPv89i%<Ro+SDC_?)*Sij367{%>O`81yw~1MT-_
z5t@Q&#ZW}(X4;!CIA4Rnr5oA*b;(GSoYphYa7!E4VJMIll(B9TBZ)*#MO|5iV>-dX
zAj{Aj4eCwx^n#`?cM;lZ#oPwJyS{R6Rt`!pWjf%c{PA9R#A(PckNgw-Yo}B)K9Eu7
zM-J8kLjKGG%?5<Dc5>@9xFVtDW*K2%<68Y3e|f}$BqvftrP|<4ve@IXg`qu#cV<lt
zm?h&kCk6~th5=1mt*`s%m<3JInD*d|j<TqMhl+z!?3eJOqL;L)u$rO4!J46}Fl)ax
zz=X&$$x2bXXw%H}Z87IAFP|7NOc<sW)CKwdOa2Gl`WMeJ!b*}Hl6wOWu%idABH%Th
z=7EN~E0N<qOPPhaF%fV7^~yxO#cY{*Dg4K^Vo*P`4hZyB)VEYs4c=7z@uZq?56{Ej
zBRHN_e`gG&*{iZZ%WU61<R-o#NC+>8nXhx)6TXRAn(+bPl2Vak|NH9cer8JvXXL>~
zGR50E5N7a+09?VZE)Komy#OlaS1aIw6JT=(Vq2zoBoD?=cygq6#GrNG>`Bw?aDph6
zpoDf!D-n==6F!U}s3@@wFBhc}$x;NqAbx?jCSd`-K!zl`qrIPRk$xm~eLfex`|e;S
zvX;;7cnxa-TYwf<AvsmA(4~<f69b8A=lKRvmbNCp$!;c^?M?ok;OPrn-vUMf@$&O!
zC1pPbJb<1*mu$HZD1|%A_I^V&u+~w?`m`AVV2kJd8PFnWCfkx<T3Jz*S7J7GPt77M
zrQ9qgaYZuDXz}RbDvz?@jg7voO0p2Z8dW>={s?@*iZ_W~m!Ewa(i+LqU&#qe3#_37
zt$Y3;B);fb+G<Z7SvqZ;RSvYln>6oj>zbRlbMCYO+op0ltR1MjE4{9<2;402IoRXe
zN`{k*Y0Q=^$gH;>9T>X~$0>31mD!cY|7Ev=j}m_+JDDCiy1Ya(D83We3@ED{_-h}i
zShON-7@g;|7Zy70c}un$7#c`@v|%U;kewAXT!N%#!Lz7H%+Wc*;J}r`R{}3l0hQi9
zd^@lzGuV5z>#BE{C|LjMZ@a$r`itQAL(<VrnZa#SbEc5ay_?(a&biyRc=FOyX6K3|
z0NICfk%a5wh=kL`pQhQ0g#93JX_R|@!<4#$BrGp=I^nL_t)|!Kg9l~j*~ddxIL_3}
z^T8JbrLXL>OJ=80BFDNw%N6XzALa9=1qGpFyYX%7ylbW3gGS^?ios;H%N9SPZX@Ao
zo#0K^?C(JPc7H6h8Y<02#RXZ!wm%nH7mdD3j4!$a+Ta*BdEoYitqZqXN2M*CtkXxQ
zTVrGMIBJzYD8(=!?q5jUP?vIM)HRA5z!kd!V9m_mQ*&L}IDte_N`~KpIzz}6aCMj0
zS{i%Du%X4gYNf-@!NOUS`}WQ{vCzZAD0v9z44b0GG7>qbHHVDWYUEtV7sQM;l7C$P
zJOE!TWIVrj)mYtNi&)~Xzy4u76oHjb1m~mQ9OS<~MI@5CC%Q9<%N)xPK2*T{x%XxW
zG!%E2znxnr!L`R2Zl^DEoaomN)PXb9^~gC+;OV$kprP1%+->YSHGV;?W=csI*d1k7
zS~?+{W4(`Ll}1chP4;i%NIQ6i?=T9OP~`SiSG#>hbGqG_odgMCCnoImz-0?zyhh^j
zj4!4a09AcBb6St_0TH%J_}a6o^x2jVy0h-x8}DT$yxxx-_h&B2Ot0Ez@{^X3{HvzM
zU|lM%$2lD`-Fd|23LwEORboDt0YEa&J-YUfU}q{;W~xv}EcKgS{y((`HP<!wi~cY6
zey#o?cS?}Yi~hC*`43&Td=`@@unq3k*{?Xxwk=%*PYbQK=qk1Q{m(VQ?)JW<tXR-F
z>=;32J%25d|6F{?2OLX(-}`#F>Jxic`Fj&ITL^+a9CU7opx>~tZe88NJs^!et-|VQ
z4>jQ-40W&2rx`rz|Do_0dS&ASpI`eTWshO{@RB^Ov&60^Eb`1Sd&#Z_HL^2Hoz;nS
zkBP9xoD2&OI~nu)qU@MT;e24thPQCa@K%NWqzabjE5hAQ3WDB<hn+AP((WpmLWxq4
zo0N33p+rs`FN9&!w?zYERwgBFf7UYN4Nan_6H%8w039EqDJG6Q{Z+==VF_id6p{LE
zd@wWvo~at*s+9Y*YaV%jrbsDDODxzi*Hg2zeiYIG+C#}ozLthh<>YC!TnR2!L^&^j
z!yhFBp$}lt>s9K|$QvJ9NfMPIPnwk^?IJGiRZZ?WQB<(fn3XH$(>eJnxr9v%O}P(C
z`6Xp|FDss^`mya&{roiDj7&aWqSW(9s4udC<0=GR!H`H43_L<L4$S@nMau4~c$~5=
zE;Mdi3VRjF8&WRBg~yM=V59NjaTj)-A=GyS(nznGTx8lrOalcaYD3;?7mT;TIJEs5
z{KbO@wR!Zoi>KUUdk4BgpPb1Oo_kKb&vZ>4RuCxij4cN!DyEkrNIu1idb*KYCAFE1
zTByu0qiI(q7cmbb#mqyvnRGK76iPz>Yc{c~=xToU`6{|KS}**=zVBeI@vjzWa7=%E
z(&aWua=!56H*lVajZ66PfWa}jQwS(byP3cIy=n_Kjg}n}AD;^6Akwx*Lm~$fq3&DZ
zw%#0`oQy>beg%vlLJ`&UQr_<M^4XUvz9q7VB3<V~P|1QCtebJ^<=|F!O6tBuP~tc=
z{<KmSGX2XRL^~n1pdBPM=>wIflr|CEBDso(a?tMgHpz%){cwI$3x=?ROjC=;Ve=Ei
zpz~8x)<MG(*A<p@*GZ%SfOKEDRfu;B6%1U)(9ggfba>V~9`9WioW2A2^H2s(!4O<Z
z8j6L)3?(HE#Xw?*pi7ds$3o(^Cns-@gT!ucJMC4G5-M*B*{3D}Q+i33;XxEHL9m?i
z6$o02a0P%?3`o^NM^_*|9@KHpxGNa(tQ8?T$(alhPi|P)h>6;_h?#$lcnw?58wDQ$
z?@jGVOn+meP)MbBgUuj>7hwhk-yI>adG-t%{Ui%rPX?LCyknb$T+hlPM{%z^D%)j{
z_IcgqM^PdsiKx+#h)fYhTke)YvoR`xmckUp6ERUoi%446&1Y6&jcMbmM)@_&yK%rb
z$ewm7v*v+~-mqUa?ykf=%F+aH-zQ%as&Q)m)aOlDT2#~NAp<Rt)LoX}mA99*!C`KY
zn4h^3kYNTU7@s-w#=H49Z9$3WALbO+q7;K1V7P~!m6M#0(;_Jp&A>TqPNWyYA<M~R
zbk6B*Fn=zT*LzF_#2Z<z?OBe-voJl#Rwt-01SKa2UC<=BiTO#_QR>`Wl<K<cf>_-P
z_@X(jfSM|pTF;%JDcglQ!8xM&<8H_HxQfwWlqBB7&}6kL3*0!qz=KyCvYO<@lKEn~
zLYPGIN~p>DJFha)-yMVIQ>FOIm+oAz!_6zO^tc_~T=z?SrIZTIKlU9u^VOYXJyk;T
zCMkqsx@3N_JgZ4<@Zbe}H%?iQ)v95{n^2PA(Tce3x0lmyN#a9`xiK;8|9ucD!5my_
zOQkj8^~dA~(XjS!|9+2hXWz2w60QWt#T$NubAxi$ihY}unDiFAQrs1BtMINsfpAQE
zDB@kWrW%$&_uHHnVdIMpEL?C1iLU8A>%`;^--mIXdM|yejBMC^yqIw9z^-{}l;{wi
zS`!sjL&YBwsZsNG9eB*ei`wzERh5}YuWw64#1`KKD$pu?Ef$YYLrF#OG_e$w#+{`~
zQuZVvX!PasS!1_gBb!>~(g7PB;jqJCb_W8^>Uhi|N0M1L%i9>&l8_(&&0^GV@mA0`
zi!xX@$la$7FZizAm3WUJ#zfuR5}x+0xy41=-5b9FO~zq_%(5Q*U4F}Cf^Wslrc}dV
z-}h`3rxP7r%7#chpFZ2*^VK2FVgbpC$pGx?wL-d_wb=Mgeh7!~^8UHG5{zziF9NZ5
zR42jAJ$L^TY^5COm~>MF$Ko6lx+z5?M_vmf?iti77%Xk^$jKa6o2tO&z^9V<P%l0v
z=8J!iL8O?2MY4rmlV_ApxGC=_SU?gq34q1EVG-W!<6u+pc*N&(mR;zo?W)~AclVkN
zpmP~A(Cwy_PRKLL=3Jq+Uz2AzCV=yyd-k$BowZ%H+Zdjm^kaV$v)44k#2g;^PuOEC
z_s)LKX&g&S8*Ah|pS^eGF-fec+f8vy$TRMe%3NB|kxjjnP8f{TGxAN2vmK~hTlT!I
z4>gtJN^#f;9K*&2ZdpqWdgHc%*`r;q4`aymTYvlKi)>hYiKc0uad&7->sTrl|JSf%
zr&FhSIlaiRbh46SV7)LShsCY%5OU6AC0D$CTXshLt1=7mFUyI6zk}4~^R2<V!M7%d
zM)$Ly_5GqJKO2l3Vpbb@kz2!ujRo9t$nSRJ6nuCa07v}hKO#i@bg5UZBF8NBP(-d4
zzh_hy^gd$LQudfW?UCi<KoyE&&FHEs3p%Q;rp8cnSk)`!R%wLuouq!aF%o}4ieA)N
zusIDH<$^{*(>9;rOGz(0w68?3Y}8ijhImxqp@)hK3KsVxk&##1&bZ<ozPMH=#n~E{
zl^IuU)ETSe2Gz!rxJPj%#=#YFrEv|K`s?%`R;OlKjhaQ971zpyOf_}!eP*&QK7o^*
zq&AYB0B6N0v1b~S3Z@%_Oqgg{^!G%MQJ)pjeGgOfSQHB5rUu-Zl{h78VX9n3AQC|=
zPhF@)neG2uXdKEQgO$~lDK>c6u|UUr*_#WuejDiVTkpVc|EPVt0Kj%YclgF9p<%8+
z{4;;J*|(%QqBAh?N8HE87c?mU1uH5g&?;Q-VP6sVaP_3)7`sR*hsfQoscaG4$NjY|
z<k?qJO~g7WyPHegyj$3<*ldE_Ubl}1YGpKB;R;tsfjn18DMYd@z!k2L0(n<~JSj!u
z=H0@HN(so5Qo!vq18)|@gohuewPo0!#A{%X#SSa%nk`6MI3+l(hd}rM4(%7$Dtmym
zUF(b7{bExsWs9_u?;*W(b7i%{E@YX0x~NYQ8W^=6^s<GPF=!Ipjs7=U`zIJxE#A}b
zDK(PPBPshA(}k=fjCE*_EofL`OrlL-^c<Lp-@w~Z?V(ogKr*C4RJKTv*!-kx6>qAm
zMsr0~Q|0%|vp@jn1^Nr5+y>C8G6|s-tN(=0?6_+uccOSgaCKtr^w36d47O77;_1*t
zsCPglL75lSBz5f;joDVBlyHHKny|ER5|;A#y|6#0ro<AH*n4_eK9EdEgF-}#*^rPy
z(UBRj1<frw9)y<xs*N3|F;nM)2@*Bb#?Bc&&Fh#HRu}J=1ZYb~CL$AoC6b`b3!rWy
zl_k2Sa>l4_KV3pCNCkR^E##-vJv*^+i?nf<2<J~YEmqyNOuP$Sb=UfgJ&RFM<SID?
z^&d64lTW2f$=Cu6;AJnt80>&qk3>Gf$EaT10+{uv=+4xG#HaXlT`T0Xe@HExIw`w5
zm$)n4RlDnKHj&*8?nbkD^6qH#oiQHSDIeEkwq~amaTAIB>dkDUufHMLjFk9<9=9bo
zAn{O_Tr=rJ>G);66=q_o8jMlZB50i=mv-+Vy}Yh)bqRSF_Ou!4VZu`_IY3U|YiJ;y
z&BTB@vsTE~BatTVndfyHO7lz?jTUQ2*#&fMYjKOS<vNsCM));y&B%FNjdi`iUgbl<
zs{2E-q2O=!w*V@@S}LnT*i-hqyUbdGp+qR<aa1I+t#O2<jM<dIy`@~IG?1_s8kuKy
zut#~nMX4TaZJh?nRjma}SMxJ=a4FKvIg|gPvg@J9Dpl+$@<JKjsk!1to};%qh^z7C
zDIv!k`kIjtjdv3@f%5RgtUgMnu}6|l1I@h0J?8=Cx?YnajVe(R1>v#Qt^{6g8hY*J
z%=;u!{c^2pZyQb?O`?QR^CAgS-6RrcaU<DC0OPfXmNAZHH+2jxb?A3sIEqjI<6Iup
z+fXAUplHFMjn2?7!v>zi28O6GMi<*^KmBF4`hSE<e~C*_YX0C+9kA97|AJ9j#F<uS
z5AR6^05u9pe?3Yt;2y|+b#N;7$Itq~k<bu}I3vDj$mIHeRcW+^#;aDRzhXX8`tFB-
z&VqlG5486Vi8O&NeZF0&*^$eDWHUOE*Y)JBXc|Re5~MflUux>?xo9oHM==PUpBVY?
zr?ZWJk%4J5cu2g?ht=eJHVnaNsnTC78@spBr>&N(q;sd~=bxw3vFANf@BBBujl>7t
zzNYj3I=W*I8}|p@d2}Z>v)6F$Z%KFU;o|;$^xdF)HnTT#*YNMpbZ>CXm#^}lc?G~C
z)2M#Yo976*d=G#n8#@Ol7dMs0!$<d2TLW9|gK#03ud_)Q-~VC<@R8Po{jc-4_iowC
z@muSS|K^6*p7xzRKI+HXW&G>C{PBNUxrkl=(h83W>3;H2rfbid#o>7P){g)Hvev19
zu$XTE06?N2rax?AOnIGqnd)rL$R?cA3nsHUmr~}3JdIvWHj@{TX0>fy<oNSk@T1@~
zR?xOLW$hjv=P=k0Gf>$N00mSN5ef1L8bPcNf4Jrd?@A0&`Q-vjUk;zeJSv+l1Fev?
z3<k2}6tuEHWdlF~)kH*sJOTu9IDCe=d}l!AmkTU?IeZrLsBEqbv_jSjXS64p9jAcX
zva+&FBWv&O0BHerXAqNiv$6moEue;oXpjVX1b8A}$=9}SDMMR02RNgx((ETR1)~yT
zQ!oG$GTj+Ot6i!fAR&{~9<W~lthbG0XwOv^pn#%?YAVPZ(+FZy_?!2a<vMA4K}dzL
zelPH<^x%T!{(eKxOsaC^Cu_SI&b<XM8<@1Jt<Xc@BO*e#A0`9C00or73Xn(8fQVfO
z^V!oY)v0L7>lFbTNoZPxAl5|~Z!W_LY8LD%?f5X|X2Yel>Z?7Vx}zE_YOBm^UOF%+
zsp(ZuH`TUi#gH9IK!SB*RwF}j2{nNN3G$?d``d%v<!_Wiy+p%p)$S6+KcT51BGQN1
zg=_+NjR^u8jR?EC%8CU9)DTe~HaCfwu7LG1!$P_{gW%iOl?NtZMvE-+oq}k<0;Sr6
z(FrRtsS7SOmWE7#D>utFPl%=!W&=Xt9@`Uq>R@^3_rnx`CcrN4Kay-Y*RzSbgW%t;
ziW_haF2#8`#C{xAaP|^hghhU6`L+qsYKsC`C+Ao6jys+pr@{8agv$&LUa>43b~+x1
zJ)0aVj92CoH!#k@c{s#=993{9z7V)bE;_Q8jq%W<ro8*8^Kjh@ye$YPR=>(alpY0&
zn$#mLc^<PSRSO}hO53T5OaK$Gh7;`EvuMBq`%nt7o^O;m&CR4%U;+sSU~#Hm8Okge
zD&qZ`;GPiW(+UID6zzp|K~VG#&jXqmtgt*2_Fn_+c+PJ3SB(b=dZZLK0246LnBeRh
zu)y{dpoi1kyw(a3NT~r>3saE<o+}S<ph+YiHi5yBMIH`<JX$WWK)RGMjHYVb2~i(H
zX@@Fu>+XPedHS9NX3L5unH?PnUbH0!TqDx7hmcQi=58q>vL2z=mB^k#rISr-DWBf$
z(o6<}fYzMlWXc0ilPY*poDu?G1fb>J4_PltX9p1go=t;-uX%RbV~K}F563s+mzUgW
zA=7gX-;*1<%YnOhGVv2A+g^C%k(yn8<=9TQ9G?ii`mRo>H2RXqmOYy=lLv!vEqFZ(
zH{eiw4rsf>Cp7KYsoD?I=`H11cv=82rXam+=#G|WL)#K~3WElA%tTzT9+4=Ju7V<`
zII6&|S@NvZqfVba;b_a<IBO|{7Rlf$Ex9wXMC80I&rfqOWYvyGvGjwMTY@+~qP>ty
ztQ!7Sj5nqkT<-*PBzNS=So7kF?ioxzW65qiP0p^44dV+U7m_Bs{}QyqLgLUe-|=yx
zWXV7r;wRIysnzBTJ4;`Z9&s48KgICh`vX0|?VrF)=TA_*0-(QN&;S)nL1?v@6K_r3
zShYqQ*%X?%k(*x>L@elY*uF}fO!(HKQmb(rSq9KH*4ENvVVge66jW3@>ig&3s8Ey!
zhA4YfWydpRx9uxMhm=+Y72{{;<TL&@1S=;mDvuIE&6Hl$`c_c)QiX;{srYfO(&MQr
zyCqVp3!;vI)A;tl&_|h;(w3_9n`^nYuWJq7Nu%<QbgulsB|NARJq?Y%SvxonZjJAA
zLx$WK{yX$=A%N13nZTKVRefWCY9Ko4DC_1?>;HOCmugU%Z)=o|?B%MeDkTU&J$ZnE
zXP0`IG(;f>T!dD3M11HYK8znhX1_H1)pd$j-F8=^B0o%ola9|!mPQv>wx+t~=Qd1H
z<0Ac3yxQ<`=k>ogJ=}A<$I5(7F4s2v@^LyRe?y`ejZH8rGSBLz`s2N-ch_oOMiq-1
z$FV4!ksxIDj84~|0KDVf+Ma7{kH3OTx`PyS0Uh<rEm&9EHT41XcX4a3-@QeJYSKMv
zz$*tJ`U<!GU&tdbQQwf`tMYiMw9=1@)!hBJ{lKxEUc5nyM5=+L4g54=45fcgJ^^66
z6lgEN^Fc;~$No^jvOpR#An_j)_HWhQ8LmOSDG#~Hb4szTG^F4-D0ZAFbzCWzxl4ys
z5Lk@TxyxIpY<pQ)V$Bfy7;h{}m~w4IySi3>50AL9v-GqITVV*K`3X^?W>+NmAppAa
z4|X)eAhNiWLJFN4DXXk0cSr%hmt<Q2baZc^K;~Y7u6C4E@F4-@4rh0TgQnHYE*v3s
z@m)v}MufXyaMNirx^SL}QX6kd61f0jCdJ8qh1|`Nux9<bimC&I$JJ!Wcj}FYMgSn#
zYe9c&7gTcM&n{_H=unV1iPZ+rc4N|X)!<;}VXO~adj=3zKr;BWS+FaqWfB2h&!s}H
z=`YO|P<1RU<J60~Jm+_nrW@i}*Z+J+3ly}2TD5aj2_>;cp-b$)z&$uj_{4asH6DoG
zHxwzn6PMCFB=i}}<a)`7hhIY+etg)e8=DRdX)o`D*2<QS>%))TK^3_NM)Z^Omr7I1
zoUx6+<ThqLEhRB*Kz<w68hrLNwxgg61I>A?BiQ%ANFC)?49xKQSy2CJ?L(e6HVknS
zmK{k~baNb2oC+B--;@pcC>mY8xKsBtj<(9k5dfsgwI7MSP(P(&Y&oP|K9D*WrQ0>G
zQ%4aXP}3ua7#u{LazW+!5qEhUi+J>UUDlomRXwea-6O2j{WGW|RsTsaHw5{){@ChW
zz?~A=RzkCFhC-+PpGT%-&k$XG_XXqvDyKdjgnCjzrU!Dzjeq*gG%Q^s=K<-cE(qe>
zu^p=Re^PCH08n9c;C#q!2hfP5hxK%$)JZD6UUGar<skHrKoAr_CRbLeAOc~s4z9wK
zgQO?=;&9s_&qv7sO4#I&N$gw3*oYN&Cb5xjR2KlHc56!JPBw#-?6+iL)g-Wx%{tRG
zN{}PMK(U)jkz=R=CMt33ly(*CvWudW%#<h`e;1+lNtJgTlT~<_YV^Fq$FV~DAg3#8
z7PP1XVuMYgJ{sXl&8K2EvKmpxIvOT{g3-V5n!x*xFnL|Z=o`$kijFVLaNku};mB(}
zT=?^YOjEVGjZiT)YPo_R&qZ-riG4(v_EZT;DNDWUl{v|*9fOfBNxngVLj<bu0!VO5
zg_K#(uT`RxG@q$Zg{-1aDYMQw%2cC=eO6{0^KOMtjTN*&I%W*b!~#roV_w+hK|vuw
zsZxyPV-y0$-i|S}09dyn>HW!A%#Ds5gC!+Kg4xKxJeV``$(KNrHzMheY|%Aw1I|or
z21<uAMdNF^<CcH(^#G_nD!Ce~R_9`*!O;XjCxxVvr<XwvYM-;N7&?*Al)#G<Rcgfi
z*jPbC41(9sj|MKpOX+?09jNqps<vBOtT`8C@j~lYFZu*_ySw>!FB7ZokRG0m@6l@!
z?b$Vihi-U21npsfZ>D3gb<u^-L@Xf#*E5*b$%T!P8|!Fo#v<ez#~hKxMricfCQAUh
z3|Cxe_f*g~Fw?r9n=STSfITWRHC~nyWSZ5SoO7EJf|n-YrxbwY58*}7KQ+}$?@nc>
z$&m4g;Ye_BjNIwVmt&(^;+Yo<7PASAFq%a;Y+6K5OAneHxiyc;fDU_ul@dsOZ|~}D
zi|4)2nXmoPDV&x7>Onu8qqe4~zP9Jnc@JKEYJfuOIKqbZmRFi6xcc*_dymDPo-O$}
zn@wTNJO`t`{94~ec1^#nT%du(!1LUvsatDPMG<VOKOWS1KNdsW3fQv7<nPWH!aGQp
zW6p*C3UTv%zx&d<_xD0yJM^iQ;MT!A|KJy<kAJ?YlFMl{v~((t+4q#~99b(h1I(FK
z^w!-*8fQ<wqacH`!LF;fF%+AU1i=UdZtWgxX|!UswaRs+e}cOq*nYYP3h#w2rt<5W
z6##42{qR+(Dz&LL9nt`M8_=^u_PuF_JqMnlhGPRHPdUG7R{EX<3viYyiO)&0WnMr=
ze2b~89t|kdGEsnLTl7N^3ml*hBh%n~$-_fOGPQ(B7BEwH4t!8{bP@0^Z{Q}51-xU%
zJG~W|eI+t>oM4t!W4SLdn=LYhfCzTM3p@ahgZ7Am<Df)W6a)ahAkj6qDUb!l*0B@W
z3Q24wIqE@AU*CJdA{Dpe|ESe)MXRM+dQ<?kkp)x`DyujMV=4|Zor6OT`UOl!;};dd
zWW+(9N8$0O_kHkNK5syO;#0b#9{2pWQsZ!>)xEm^ckT!);JcthZoa%PVR`~0Y~H~J
zUz=fmHPs7(cz7;jFTEAdjQxZA!A~FO>yEGI&rXL_GzP!nDX#Qchkaqi;k(|O{eDIp
zoAqK;hV#dO4V{m2Yn-tw?)8Z9<MX07@lyEvo$1FJLWJKLN?$#TGQ_^3$vk~ja=O`5
zfLFZgH{yr>iBt8337i)&w4cti;^=Q(6Mt~>&Q7O5J>*Ld<DUE-nx48Zzk#&X!GYP>
z>wPcZFFKk2wq48fao4LpXB2qoZO>$c(0}LRob9J8yqgEk9^$sb)<|dVjnDL*nN5xl
z$JVO5Q|<th^`TWQST(&KlV1PEvDi$Uz6^A&FQ4R+n7%B&a$*f%(M$2*)b4LP4^M?p
zh7DPdDWXX;1ZzrOGwARiUVy4X36_|rf&(A{i?r|+D)u{tMlqN-3c-AwDymUUX-srR
z<oxzTSG<5|IKm3XAe(t8N2Ig841dpvIo-mA2tmO1C%6E}G_P9`sd(T$=|cVmFb!L~
z{MW!{R3w%_c`q3d08sEWm)MSYvuG}%xEpZTR1iu!(g766JQ$HLa?Jz;%f=7>xk`_a
z*)mV+f&<xtv|%B|7-V1OJLH8BL+uuFL#>tTFab5C>Nu#u3c>;7v5S=Z+4D%)9<^n0
zI8^S3z_U`8TXf9-6-@p~3Ez*xBLT$@^Y2vw1m2U16iw7l<sK_H0!+4$QPOywo=%)!
zT57@couolvjU60>Io`L(uL)JFu5*?ZBBagxmfC#$^1NO1FkP4d-K%uRbmE~+39RUA
zV+RI&;o*s&fbvd~i8$$kAOezEu=u*~!=wKp8{!o;IHxwvH6a|Umxj?2OKM(2FK8^B
zBu)iSBRQ;=1oSzeo>8W2NXAwZK+`$gw@HfyF5ffz8#5rbsRV0A0el8?@NBq&jPGup
z7#N+WyI`*G!ei8kp~Q7dxUED?qqrbQD3(uyZ9A-zqMF@X4Q3*MSq8oDi5L5N+ORaZ
z5-0#bn9g)@)4bo9xwq_o%^Mn-s;%otvR%eRHJ#%<yz($*mp}!)zGR#F<UYevXSwE?
zRl~}7<fd0DI0k;hH*U9j0T?tqJqQ#kSSYYqUl<L1)Z0A|%CK_n4i`)C-bA#7q02a8
zXe2aN@uZafaA-UTHo@Yuzk)-r3!q0G#iP`LiUx@UiL5o(2VH9PRld!_fQwOA4<k^6
zkSOSwmq6QSWEK-s<k3j^lJ@B10{BM*=v#5Q_Y(BRh_`Y|2}Gr)om>XU5A=ze!-nTd
z-B_w1`f`yl@$yYYDOqU^B{s%vB_N>*%8LlZr%$q`ugxHdp}8+Q7E)?d$yi)4D~v@N
zZ$WslevO;h<iN7$>(70I`LPd*LB0Gy_JyH&YlbQFhuH%Tp>5bmn}f(bQx85~EIX-N
zY6JvaHMGM<m~+<~Z>9IqzA^#A9}-b^AVIgSAOleP_+-=8r{!e*;w1f1&uzOf&9rNn
zqxnI>tuHK)F&j+qYg0&BWLkzoy7qKYK+5!6M(%^7oQUl2dtU79JHQS{?B4iPiHTaU
z_jHP`X9ErnK-0Cblno>YZl-}r`O4T8y%8t|8&ue|;n|R_`t-WIe%91m_TkU<Jhuu}
zh%+7?IJ)IB#gGvVf1Jg_$YDlKSi^AmwmAW7UB0(Lj*|kQr*C>%KyW4XWVbxhV;+O^
ze63I*)*Gb|dg76W7p*Fj;#nh=o{tt5qi{?Koy?zQ2=~{dBgd}K=X?qhaQ~xD>a^;I
z5BKRua?4)&hpKSmYDEJ%L?a_`mm8}lpyesS1lI4pR$5gA?M*Efit0vpOk^;!fPq+Q
zd4As3DlC}Jthm`O39T8JEcm$Xx5xAHdBlR!F0q)qD;>hLT)?A0Ifi^E^Ol_rhKUZT
z?-q57F0;Fgo04ETRLPxJ1Jbfv&Yzwy=(u-*O@}gs+jli39}XnG;okiwd=ay`*_=Bo
zv>OcqGwLonb%0P1Z0FXCTT!^ebO8xx>=2M}MRZyLHCu%23<{(!I7f0MFp<@;0}84#
z*FDy_@Ey3y%E?(>qR)j=e0SNhy<~zHANA;XP<>)Ds#o@5CjVSf2Cytqv3H+i7T6@C
zrnASI6$~n{F_Z;hOwFDYOxPa~(3-i#Z~Ky9B`~hhAMjGarW*6#U_Wz;RxS9&S8ygU
z#{w7Xi+NeCW{W#h5s}F)*i##2w;&+K1U#Fy(3FUoaUIxD+2nk5Cm5A~G3t!)rcdqc
z*)tU8v?vLPBY(T>6<Y3EN4@B^KTOatm=74hG3<FZ8&o@Qq5f4vp{E_lnYJq5lA^;O
z4Ycw?!%&yQ1~LmE7ks(vw7UgP<F9S>5g%&4Yi}>PA{1<a5UZrn%`1FjX~}{NvV0wq
zXF%>DiJq{rC@L{uJh|Yo6{U$n%E`K6$+LP;MH7dG8)DgWOF}iNiEo>>QwXp3iHWIK
z=s53dvBGt+S&2B0k=hW3Li?boR*$lRfIh1@OC}qPRcJ=5sg4$sbfW)KjhSo84Yy)Q
zzD!a<%#TOc2-`%{Gpi6SEy%F#6%JHh%083Et!!30h3i`+pp!1k)s0y!EGb|@NOU?t
zbl;*tAtHf2weaTCg+O>gpS`QhI`|ff45BfFJZO8>)<MZzb1W$xzxO=`GlL<p=(_J0
zaxxuj%y5Qt?+m}^9%IwLL<LAHL6wk#)V_Y~;b?5&2*7dAiKZ3nvM(b;!`?P+jflaB
zd~{4q_tOGlVpw1RmpyT`vUHFDo051|TDMU?_-Y*_#a#zua`F2<q_621T(~P-YXL8y
zOr=|-K_8XwlV;5V2jt?0wi#|VAI~UKym-#@c<J4#yEna)^P3p)+O3f~k$%p>SZm2$
z+9o4h;mM!qU#5`>Q}P7@6Rxhd=@ptaP8L<;j<H9(wVNZ<UirdM_I+Xxf)6@%50Uo}
zX1#jJ%Nbyc#~$0hLFQ^-p8`NqsrA`og5@DD$1`KK?;2n5fO+l;SkcXAmahrttseJF
z$M?JBETf6$<bo?aFqCGt&g7y9{WAwTx*FTzXe#*@9{T$o9XQNCihbc%a=qA7E&j}U
zHtuH;y&o!>mPR$kZ{>D5s_zVXhBU)ZQnrn>b;Hxz%8-I|nr!nZBIj$ltwrg76yKfz
z$ie`}V(8_nM47JAcAcV%{sB_}g4QS(c1i^~<O3cb>VXP#D<beG1bqlW>MWr}ENVL3
zwWkj*@M3ovnO>ASyoY7F^H4K6$;ufBxuG#YX!un<v4jZ+tbRE2T-zsjn@C}-v0P!X
zVw)!a#HT}49gPVigeXeNM}fjX$Id4aF-Oa<p~XDxTWOo7Se;z!U;wvJmYyg8u-1L3
z^lLyD@^iqBG!Fnjq20A^l*nyMBK<MLcT@xg1!bm6143E(BuHYLf2#d;1AuU4kf5%c
z1$~xzu<8G>SnAxS(!2B{Pv<&A!$c~h7%RyG1SaBjzsO#FEE*%LXcoAS+Wa}jihm?%
zba$v*1v@za-gS>+iGw~vfxU!89pGe4K5BDAS&Fl}c<TCGUrbW6wK@8^M{GDZzjmV4
z$o#J4GtuTCysj@CikyWGZ;2}io0e|l-FoIV6Nukm!)yXhFRK-Uwek$R&jiq+f)5Pw
zp7(AxD_xhVH64_>L-V&&O?fI}Dy=!yYHXZGt{&t>pLm|I+m2_hIUJiv)Ps6kB_Wep
z(?F=1u`77rV5Ss%mgaqagcQT-oqUm*ZSNx2ikZ%EWOdI@*svO#mOd<EX!$iN(J3S3
zD^-SPIXZ^<ZPX<8AhI_$)<)CDz;tcpRn^_vk590$SqfJ8TF4bZdMs{*wmP>a<-;v*
zO_M#+5CuY^p6*b+p%M{bcU^fMS9_*Dx*yfoS{$aB(i9HDgXcDUM2s$T5z_g(!%{M$
z=oGU&%hU?I2nxyNz5clr;h6Y|&U$pTZo<k$I$2qliOg{;J9dS!r#z3KOzP5ib;1Pw
z-*#GAGND?RLa*f6IZ-bxN@0y7SRp~$*m4VlH@~>Y^$zZX)U+QH#>V<uv8_U2|GrVK
zNZxT8PC9vJAb1cthk~&rVUR=HfP1N&5NK}*igvuU%72ar*=3omQ+j*3y6nCXAiX5h
zmTnPs7(|S+yYY~ZKxBB-W<HAT!`AhAN~ByV#+IS>c{)mNVB6?|FuQgz_bLUu0aHv{
zEUxA|CS5E5T8kw~Pu2tO;<KDqvtORF<!rH(7PD+tB*Ii3WsZZJs3GJX7?SYbodb8L
z$PQ7km3rFe$&uxs*eoRFe11U9@=!;e>bPJ>6LNoZF)_lJ=SjQss;6ZQsFYyt>ovlb
z$cM7@9n7yz*`4|dIe++UzN|8!GgjMpiE4oDM^Z*QsYf4by?d=L@M0GdII*&IxkgS&
ztVN#3fc7n=+HhFQ9lp`NNzMlJ!x*5r=AvR;=n%<(14n_+qE2h8Ds;o3#_rdN_(gqU
z32csFpr255HG@M~Y`D;x*iuhtX>v3aJrjyGfTtA51TO5_5kFNBY|m;0>aWItq*U1l
z0FKz7^)O$Bb|5$D*z0L%L(X67GP4t<ruR7<<T;p611&9imoFBHGsNpBQ8rprCksKn
zxtvYVMH;0*>#?^ece0xloPR4|&CRoOlvHUNb}4n<3Hj5k`&d>*&w$=hPEYKsv94O^
z{bGcjqllDr;+ZIGs$a#r5r&IqpF+$W8#d({n#R9TxKN#v^&@cLBIw1BXbm&D$7|b^
z+me2e)tB5ICpg)4qSQ*lo^3u1B>GHUm3Q(6zrAQjm!&${mrj!k-iMbDfKe`{uP|^r
zKyP|MEr{uhMJq@6I>!kBq~cCE$FId|b)1Y7(&P}NH(*!o=#dK&y*30ulx&lggoMZT
z+Gf?`-j7^bUFjMtc`6K7WdVUU{Gy$b2yNzkqzEOtN_w)n)$_GLs=;J+!xY=bls|=l
zk{WoL790+ivZGgFg-|Qx#Oowx;!F+`^0v!ERT{B9r3FR)3!4DkNY3zPIfg0O$@Hn&
zO%Rcw!rx}N?5jixu{VUe_@LG>cbOS%y}%N6rgO~Oln!T~9;y&TjLa>!*&;p)@4}d9
z+DYfEIu6w8^tMVt4UiDuKKnXS%PEpC0ag)2MczRb+P^&*C_@|JOf4T17-QPpqNZ4+
zV74SP`K4uI)rGGWqR>IdzJf$WQyY)e=Q_6mOS+8~S5mMS7)MvZNXdhlSp{!1#zq?6
z;gM!4=h@5pSYT_K0;mjYO)P35F-UE>j7)={Y2__8!B#~$V};xrKn(XW@(N@oRi;P#
zkuo6LD~u&AA`>=k<XVgr3D;F}F(`XN;qrWZp`a!gR5Z13PY9mPU^dc^HWc5%#tokm
z>kTGrW(a}huShAHny#7{TeWH5ws9s)>D{VRQOy&YMnXxB@j|gA$g~%i#GcG4h<Q+2
zrMh0Z^ToC&h7~ox&$vqSYoDfQu?qqeI)Kcd>F~O_j+8msm=5ZDpi)GIDu)t$&N&A-
z*LB6undQXWL@NrPh-0R`V4Yn)eAz47NczhyX_)lPb58&E98gO6^<o7mVp-wnZpSEJ
z<Kul}T5OHb<qH6Wr3iVbxqej2SCKN*q9|4%Z;&Rja8Gi2$Luj}OBWUuDVv`W+zBVm
z0h^CEMkZm}6A|ay0b6IKNX2<W!=d!vnm}==uR<lELNtC5HBEGv)*cMr_cd$Qviu(Y
z$2TgzprkGn7K_7z*~_gY+Lr6K?tkj8y}~HS`}lx>i8{IFn#E>JXx1PfHe@}gpdBum
z>VoD<t25m-txFp8^_+^su`>uYr8Bdakvr3v2VXJ_f7bfT(X{Zo-lgKbT2}cNJcG}K
z!X+mOmpxgakWwe9lmNWK1dTP8Jdy~^737_TkKRdcg!0Jmnpc;J$y7&w)wM~oS5mzc
zljI=Gb{QuAW-V^#_YwtwWG8??a<V86sO*A9-g`@J@wA3y2wqzR%^lCGk<lp#dzfTO
zS9+QCkF+dw2Hnx9>JLwsw$_Q4zKFn0<^p>+TIEPH(%fycwzw-6`T~B9nrW3w6M90J
z&=y>ZQ{>vJnP+Ec=1(CC?u4-nR3>gUQxr#;S|;e`tC41x(F@~QoTy6;8Ob~$DiS2?
z6-Fh^i>npMYdAqOnQN=`Xw5a{kkD@}W5Np$$Y{DgNIF78WYh%$TZEhq)L}<gXDYtR
z!yTd&^1(TB!u@mY5u`&<^Y7m3&R(~TpLuv<NTd?$#gN$MN%4{#XW7xgQf7J}Dgnvy
z56GW;RhqLt3cj{1a%HXyro0&#D~OO-&;>JEH=QHu7-iCxx$=Q4or3Jp=mz~u8E9So
zYor-zPGe4z&B^AfX2Td)Je*mQC7FWKYFVid$j(YnOMCLTVqDNMwLlkm><N-`K!aC_
z;DyVVou#l$<MQ=4g|Q%Vz>c#DU!E)-f{8}k*R0~c`BvAJF;=V7tYEmbp>ErCSFu8s
z(5wLUnNIEE+6v}S4>-pCunJXAk_=OB0VW-sw$NN=a74U7>h#c;QYg4cOS~4eVKdnL
zgM}?){qAFf_tGLW&_KEZlc>kuc$-+DC7A$QibX?_Za6lNqew$|$q>y!K(3RVBH_(J
zf=M<Wo7&Bc+aCY~rM{n4S%O3o72iTXP^#6bRSCJDERdNGjnZ2zrGwE_+&7}hPLoah
zPlL$Qtb|Pf${LRsZY1!5M=zSfP1}NEjm|UHN)uIvth9A{-D&cSH-iw}X$^cB1B+uu
z76VmSL)ap4eDvpj?nVXg69MjtfaTaN&X0_(f>bDD%;cMIQphU_ff~b=w>)P6V&=dh
z2bzU2IRZECSk31408(TJi4Sl_Kr4oqV{4$iQXn2#PRJ6D3j!_1UMCsd^6P#c6R9n~
z#&}9h|1~zc0v42}Kd@xwl_M>Ci>AB<q)07<v@|fCBZSm89S)7KPhwYY;@4r|G=LX7
z0b1=q3aroyDTok^bG^Swv)xbq-F?`>s#Q54b2s5K3Aru76BzJwxnGSwGAx*HsPC0E
z&z|G8`aRTDI&#n!Juf;wTy?}llbXc0{i;3tDNjN}p%(L3?$yAMT{HIv3f|mX4rB;h
z!5Ej7|AFF0y6w)Go<CU3RV_Ts)Y(?;jmFA@jA+xf#FO1<tEP|1Q*hJGTYwBb_tWDS
z%2k{1W*5i9q${Bff3Ip07G$Fwgfu^k=b#ETxl+6iL~M7I>?UQ860^$HjW%rfa461_
zm2w~|cPEBg0vz9+Uviodp(;VkNf8&4`oo>}N@@aQ$kC%p0Dy^_XYTuWGJQs(0MMxE
zV-ZDykV<y^(gt6WV^-z(M!(u<RJckhjqV}U;@aA2p=DkbGGxeZ6VoDcEi*fVC{gOU
zxIzt5=Z@4>+0Vv|QVQ)vHd(xdRj_o@vF2Z{hn)*p3L+B_B;YtmAy~C&OB46&SprVR
zLMbdkh)59(#toFH%;`sirrR7UY5xmp&^YgKVCZb!<VEdW7aLgZwN0)9CFZ*N`A9R;
z+-*}1JZMHiK_i9z&LpHljA>;1Y@ZD<z<@#9dpfcC;S~r^+xg9Y68o+`J<^<s8Bnae
zDL5S`R;AgLfMfM)l0%0(tz+g<RSV?L+u_O@XrNIJE84nx_(*fcCdS~K?7y}Nef*B)
zNONXpgycIwI!}9z6zO`s{3J=TH|QbiIt3-B74;BsI2c&|oA1e@G!RuKW<@o<3rHt#
zggrK&`vu`3T>?9^VW0m8Zs@7;ktoP*qxfaQ|0{j9`F-J-J8o*qIw}qMx|PtxDm~of
z>h8JJHhu~K8&&AJhL?HpG7lbkz*ez;soeY~R_w3Wt_>lEma%DRoPf__3(hs=CuPlW
z>+gtG-_L2<I!CO0yHeTBA7?X&-MKZ^rl0*5Xq%`q$#WqqvofophMMz7Z_Y)c&@psc
zBpeH05rKg2)!mrIEM^%pok=4zlk*_fywVXh0|i&=Dl<+Bz*yfx&+*UNZh-?Cu4`()
z)Zo?<G=(<o+S<wsCVb_Pd0$8U#UQ}q-$!z3d6n0LCCb~aYMcaFXQ-uax@Jk7NItSI
zZLsfsEGaD=54?i<FI~}vM+F6=p<Zgez@<9eFGuL~OyhiJzb8G(Vu?S})fLB002yq}
z=;w?18l5<?0}3JmqFvuP;FSm>&QPa0LJ&myl+BFVvv&eARzenipLPSesj1>T5)0fV
z`w`exdAwlWgcuWZS*1yDkISB%$0;1SAf{lpnpu6B>R#tmYg%bcXLfHqya9L$JTvRd
zGdz*|0BgQe{&$Ie@?aZ?&RM)JFaQkfT_2Z(eEwy;i^MCcCiE6y+K=Z*fw=V6dDLX+
zg~|OsGhKRCDHzc6<wvHuk-U^lo~x}%5!6kjsHAS{^#t2lAxorbT-aRjd5Skrz>RFE
z*NaO^5y}$nIy(|)$MYRByRukT^Mcy5>*lv_56?2nsMIN^&ZqY^mM_~u?)W3F3|aBF
zRPj{ERONyA%y*3gYtzH7-bo>z(@e&tMz*53&}#%5g7frD9dy-{RgBC|q3;C%snguE
zuSRBfUysbrOt0q*>v_kOpPvErpZZLKr+b=+T%Y8Oc6w}7r{*`GWjgzG20_sQkll_t
zVsKOxnOO5zIBJU1H&ws77b%u(c=J`d5w7U=M-D1|$cEVvLkzP64s_6~!BCseT%AE2
zncb1T=LIs-bdw(559<^fHL=^dxgHRHBVi>R15bfz!CGjOJrB8Aic?@Sge0!&OL)K(
zyCG+%4gnwx7e6qlflk0#H-rb83}nuO^qeBQ2}#(z*Y22d4;Prv1_OcwAr}A`i|D2P
z9ps(J&36Tzf@k*%`TdM<3$e7}68CMxQ=Q96HE2U7!N4>cw7TzRd3~S+$@$h<3%ub3
z8`lTimC@nXB59}#)RygVcDe^ghN+|Ht~xX4*;|LWvg(Ue1D_}8CT8o%^c4D@07N5G
z=(|5Mz59A(diQla2~#BD`eZyBgPzvbzk_LK-fB^uzMQM>ae|FZU!pr~L@JL?qhBBS
zAz8n(wbh3TP;AK}j$aQC*M01}x=tLk=1DC5&Gf$-Alq#`n9ydE=wAHMMxY~CTc|?_
zO}5xe0Bs$Ua3=^D?H^-68iLNGTsuw1Jzy_!Q+KhpQk=U0*PN6irKWr6Po9G9%dQhj
zlT)(zGh6izh1=orUTbm3+N&UNEs=qo4|OkbFQL%<ZL_W(dhHxT;JW$uO;pC%R}8cf
z!1jrk*W2Zd`$&0lO>Uz`b=~K>_3!iwx?m(%AnPQdIU&^SGsXO)i4&N=&M2$wILHo|
z4b5-EcY-uK7HLa|7;U!$vLmF~xmA0fS$93DWzSMx6oe2(yf<}4!jro8{Mjj6;>-*N
z-v)r}W_WJN;)Xr^_C(59r>CD5V~E?*uJ&7*+5u-0nC<@Khd*UT<4K9Kt^r~NY@}iH
z_>;S4$4$=Xr>x2z(5>Cz;2LcBp8gW4#GsjMnVs1^`;cOa4is!p*_<BOjBOrM5a3B^
ze$?=ZuLIn5#w6wT%61MdkiA6PjsQ%7b=z{71)gzgBoP6>YU*qJAuXdVbL$hDPuU;n
z!YsF}q9Kq$K@2;(0)zL}o02@c*z*sgpme7$h8@C0Y<vVNO;>I3@YAF2-tM63!pD)8
zM&<2y3!z{xrEz(g45*=onjwj9#Ufrp5f6{old+o;U)LUc6PwL5rK+ruI5(|@;TfJ>
zapM&-{`vp+ohT@sV3t4IJ44EcdVYw0W%Eq6+$}|8Sded7vEM|Unpn*%-j)98xG4vr
zE<dT?2Ka^VcJP#~R(CbQeBY{~j`}i%TTTI1^y;QTF!$SjuPZW|QKDTxf5a<o%<qOS
zc?!~%nysfXy5ql);E)SFg&O8AFUu%ss!zM`nBgHt=yhHid4_DNXD5>n(Z#xHSI_6C
zOwpbBADW@BhjcSqmGw9eSt-+^Bd^Rjj;GwykB)mJ?!&1EqsGCUu_sjL6&cLs28pgA
zGY!c(66r3}B18#0my8q0Fp7spX{f9$e^U`9PS$6#??!)-U4O~1-n;GS>+HMZmBqb?
zKz%p><U<V!RiS8HlP+b|O?;IkrIk)^Hnj;esZ>^WNsH(|J4&;3qLjEOE;6AblQkom
z4O`oB!c?9B1~%EPN6=+iGvH^IG-bD0HJ`Ow>(Ah^N|mNBYP^ph?19s!Qb7#~eYm_6
zsZ-O=Y(k{e^m?5g{a-AuWvV{^F6ZJ804|?Na;0v#5@6u6NKnwVJ*cKA`mSE0<q$o)
z$u8A`a|Hv7=u2}wI;ye3tlhe6Gn)wr5>zeZ{MVQ&xTYJkm}ST)!)Cl=Ho}#;%3|Za
z0U%{XU7AsiohcfV3|yXHn^92Enna^{)lJDp+R!l|cu0TGL8@;1-l&)iE@>LF**c#w
zsnWbN4V`JN9UZyi6dNEWX6MLAV*_p~G07IVJTCx7jN)-}QvM4Yhe(qA%ni**XWF^@
zmj39<0}tK$%ALT=;*0BE0n;%s`?{PWKlcKF7|l82^Y}mi|EIaGSN<E_DX58-`!t|L
ziymuS@ii3K5l--g7yPgn2XGiCa2gkJ8DY4Kr+5V#qLF}P<e&)5`!)W6=#UJP9|-F@
z{@FYbE-5SJq>5CPj?zhbN-ych_v+sPN~U9)cb)oya4sLtvwSXJNZQgOWweZyak2&6
z-<(VX0bpR|y9ei7_ga2qPVqbdaPO_r5c*+(=h~mWz+av)0C0i<00c-{eg(ku753*|
zQ22VIxGA<J1cZBf`_xcbh~$w*Ok(s9iI$U{o+@YwnLNZ41C|0tq>QAMh>(v$9n%w*
zlp(n?cgRvQ`AvY5Odv+AND$<n`IL<$U23Q)CLF97#4H0|0Bou<aunW~Z?8*oV~``3
zG8r**gBeI1Q{Ws&m6rin2tfToAr(NINHf!u%7_TUAz(*A)8ef>vydrY1d$U9E?f${
zWVxgZ`uXRN$k0M4LxCh%5pXmTB`1Xg(;>?3ggFPvduA2j|If?xLWubP^9lmSI%yDj
z<bT1$41yUQtB8x}v(I1e3g9GW8ttVYRS|}n76IT-$td^_>x2|}7E$jx22p{m!jkV<
zaRgWuxoO&{I*!;opodU`v5rDrk^3<9Fg%wz)<cM&|0FI!<x;P5L#>l;n^~|Q<{lEo
z_Iz50LS0UNMK3{sRk52wMl!7|VmzoG?|T@E^oqnxoD>}~f`*lbOmEHxd}%#>2N4F%
ziHu51f+m7OKA-7Kf_hfpq_eP7pi9pP$w>+#u@I8+3>dTryz~wK2Tm`ev5ZX8ikpRa
z%S<;aDmg+-BMO>EGU=+c!pI=R#_8yp#0mC_rkyeGe|&%=A593yL1&*At^iqY@8K7t
z$Ax|m=(BoYkMN5-^LF2V`5Yk#9k~`Wlt=Vj9l1=)Y_is3ORTcSR;&NkN{x-#_N>iz
z+F`%#e`_=TH5ei#nqEq&a(aA78btI|M1YzOKnWzG0F|UozttY_Tk5q<W2+XMslE}<
ztabjn9KdpOK?QFowF(_wg^9ae4*&sHj8w3EMyqVOcQ-6N(S4R*ydOPm8W?T7Un>BA
z4W<EN8&@b;00sd7TmXYR5oyWK`B$(gj)>5vh+)SxfgWpCnWsR*0vm!(XUAsqIpD@a
zoapl`H#)PO3O)KX{2&fI2<Iywjb3ej4*F<s)#}m*qXcCcjKHW8@Q?XZB0TN{+t~@=
zusI2Y!Zc%e_$2~kv$#rk3c|~T4VmT3jz6ti9B4b($%&H=^tjP`kVHk|fJY-^MtRVG
zgy$1JvgPMH+TQ~fkIqj7B#s0~RF5$jsbf_Fv189$qW>6L7HWN_NuZVis_2@A%`UHn
zeX(THH?d1_d6t=?cNkk^VM$uj7^KsKC~pbY2WMD;7tn^(bfOR#LLH5WPyd0O^%+Gd
zK>v-wl8=Niy`uYrvJvoAV?RU)Tsj;N6m+!J8Edz)s-)!1{jKeRlY<eW36*TfIjrj}
ztgjS)WoR8R%QsN9k2(%0AAP^2Bbhd_vfKyBY#4PU$0&_{A4x0DvOkhoI`^ncE8NJJ
z^iBLhh)53?f##c_VY)uwz3>V==ymbK^c}1W`7h$MGqMc|&(L_O>@y-)fOvu0XlfH)
z9l4aa%r&5V*jcss2MlpKxB`JbRPsj(;iSpR9XOJZ3c#&vH=cmLhB5H*O`Ti})F}%1
zMz0fO2#Ni&iB()I_HoT9<Q*8Z3=1p~O=_?dZh;5|1v$53*H5W83C7ltWbh@|$fn?U
zGPeqx10q}zb>{O=qSR0zL{Ki`a6l?Dx-e@?l&VsKLQn)i1i2()Ky<uuDC`s*K=9p5
z1mvI&OG$vqElHKB7^OkyrW`sL3pYdE<1$Qvo2hDnNURI?>JVg<I$#779N<ZW(FZCl
zSRN-3fu^%J4RWupJ_AcH#?Vx2PBSE_M}viwDV^`Fr1WA(7=wXON48i}Oo->Aq6cs&
zPbDb~ZQLM*^)2;{q&t<ffYj(nm{cno47q_&0jdy{adTiKp)^}$^dNK+VBp45=%}0T
zX-r-#P~mz>w_`qtD4~~wafb_0yTzwEM(}+ce3ki;d#n<0c$6TRD<m>CvvjrGkTw;-
zmcu}y5=ImZc`z=(0crs3{a17jRr3!tgC6g|zaW#7g-SUgkjJS#xtLD`s)$OJpuwt`
zA>eV1RvZ!kKIIbRW)BGK3(5!q9J)fOo}iC%j>ks}2|njMzkvh*B8I?BFqoyFq@tz?
zmK3a|qi2vOUjZW%GYhLiMT*(jl_)h4e_q2zj0yxtb`0_V`QSvHFXA{$4LdRZZ8tBv
zxuD=YuQAo?(5XqYVkL56@h1x0P%&nLL%CSvE+&p^9rnU=FVRqi%aG{_Jie+84wEWG
z;x%z5Xp!@GQDm8ZB0;+s>gCCoE;bV6A~$~FX(0T~GaK-;;K@T?@}&vIh!rPZf<#bB
zlBIx2l?E=omRdUt(fR5iJ0}B5W|yG52qQ~2EF3%nB9a_rxhSY;=opw-*f_X&_ymMR
z#3ZC-<Za~C{{Xm5NkwfMd$B7twE1eGqpz7ko_qz2Ow25-3j14;Vm5XqN;#A%@7pLU
zYOjN{HZE=+m8w*$;Z>_ny#|e%H20KGi`F`|X|GF1-ObRMUl+gbHo04m9=-YmJo4BR
zPd)S8i>kcz3Kgy1RxA|JE0Ll^ixJyc_pRt*kHq)DkVHw6rAU<~U4~3q=(6R=mDh-E
zAqo^KQmmwYd*ay}r(^b2j-f(j#~q5HS`DUJu-EF;L$He0&}tTUQj=ya4R+EQXXD0q
z&R4y0-UVN~=#tB>m~z!OEx6{o8*aMQTeo}H`<C44gS)<U&wUR(^qucL^4JqU_^~ce
zJ@dR%e(Iy2`_ySK`r;Qay(-2?6-NE$cW=D)&U+vH;ZGlHTJX0|{^{wK{p&yflg2`W
zi5ai>Is*4NGg#a^3b@$frL=HSz89XI2tQXORf;ULYm^rsT)FXc%N0tMTBFtJiN+#-
zW24Dzu@)BnG;H>gQb$>Nh12EsR902jcx&tG8ycIMeJzAz(ebgm{M|EpdWTp9Ok{yK
zri=@z^iRWB=R=Zac~Mq%Y}&3L#%W&GZD%Bl`-Q%r96ywe?{ntgM)R^#^F!+c0ze2x
zPz)zXie^}j7evWjJiM*Yw;vfl%@p&u{V2We=coG%G5&XX7Y1gF)n<1%U2c!p=LbLt
zMsju0-y#J?*uzBq(Ue+(tP<*JhQsCY1wxTnB9+M%N)@ZtXmxsn(PXyh@h{fvSAP6o
z-IcBqM2V?zsa)m*0IVa6Uwr-zcr<O%<wCGo1QWE-B4!pL)o5hP?w5qvFy`U=Zi@F8
z2n1oeA*7XRiHK0$5Z0gFGS;kzw&Ooof%wBmrt8hw$zv;<jE|o#87Ua+@a>=9-m<53
z1>)#wSlrkebCUcUMk#qZ3RdLg$H6D$S;t07Bc}f<<dPb@fON8CIR>=CVPS3Vvdl`<
zNrnW2mn|7Vm(2V_kc!7goj&!Y&b@K(oQE~xw%jN8#WZG$#A5R`$)QiaB;1BfAd=&o
z5l_%r6H!t6qto5`LT>#T$}ii)MumiJs8^}gPiWGliFwR5?HqW5m7CuLt=7`L^+hD>
zbA^D{{&*)0(u-sfP;eYh8M?3OsEy=%_{MB(4_^)mBYjIJ%6HD>uCjJTCrlq~?UB>=
z>yCGxdDYRs(Kl|hIf;}<e{^};ep7zFexbMxk#RGIo9kcGL4q7U^-iABx&J2-S*EG=
zPUpK#m;V#!3azjR6`O|bS9_PEQ?FlE53DnO^GfZFy3^;@Hh1jkCW2n*jvfx4n&a9y
z?xx5W4)@%D`s>{rryqw*c>$>~MIja*q`3boelgpX?Q^9sEx*0;B|dY#`QZ9^`tLd+
z`pY5*&egcFaT;?ZFC6UlaL=Bb=jYqR7s5A(2mAWCW^Lc7Kh3_KNBiT+tGWChw}S#H
zJ%c$N2wI911Bmpk$@JfGpmzKICNN<{4CLlm^M^0Ir&M4FwUpB5aZrntsTe?mJcxZ?
zGYiZa^Xfk9)?;>o6Z3v>xN!Hl!NColw%>QJ`4VplU>?LZ?X7bxzPc}ZaOB8i`NchI
z`ll%>ng7?VV!5Sw?N}Kr0+|W`9g&$eJ^F$kV{D9#{pCML-8Y>z=6%eSXck$uTZ_dS
ztXN}92g00HbH>h@6KA=%BF$s7-u%HoHtLTs-2U04*1<WU6T*Ztt{Xy{mdb7d5Me?Y
z*9{>}OJz3%h%ljy>xPh~rLr3VM3_*<bwfzgQrT_sHQpW%xIOK0$RFyL>ZgGL_gn#b
z2`@>rrb#H5YEy>!tSoB**VHC!&=iP_)2QH5iZ|y@m6Sbg4BbLySLWpP)|Ora2X3l1
z<2p*pwgv)`XE2Kbpq>jHXIhq$%$j6Ot56+*XJy%E6FLQ922P=CI|xI%P@r3MD1Zpl
z(*XbgfKw2MPPz)69JYiqu4l9k5Me?Y*VDNg=c{&5UUEGx=FP5gmfuDfS<$dNJX1(5
zZ6p8sX&&-F3A2$x3D{`ZoaI-=59&<-?EpaFSinF401gC=1wg=nS-GI{NC|{8t{VbK
z(^3P@9J$2LGyo7I#*OrrFwiU!C)j#Wr5uuw(=;`5I!cu?#9NHSY2MB2Tl}Y&Thpz~
zS8ks#(q8~I9k#=qGce)0gaDaLWHcn2Y~+|Jd)yf(yE^c1(CCco`)}VqSums1T1ftl
zqR*t`rC$PrX6o_^{+u3BInZy1kL0F)M7k)nL-H_oiybfHb1ecI#-4W`XZ|G{r-?ld
z^5F7QZX=gX9l#OcR9Q2vgz`l`jQEEO@I4>~(z}6}2vOM((IM(hvNQh-SSOS6Qg|+k
zjN;J4#CazcO|eDFl7k7_5hcFvMB*w8Xa-|?@3VGV@*Xce|I!|?<}VP*x)0mpS#<s7
zYkG1q0f60qVpUgDWO{?HAxE~YDTE%DW6z<zLC;R?+R%kMlKP1cykH;=wlHs+^<+aC
zws6NL)hriaPoc-C3-4lI=_NyLb__CQFBxd3|L0^!%*XL0<JbP@V^b@wCt2@!!M2{b
z<sAQ6N7{X?{vyTE3zUq!Hmr^zk1QHk!v<Hbpu5q*B0$KjwS_^V0kS~Hht|uInvY0f
QrxiA{gSdP}mcC~Z0NPn}>;M1&

literal 0
HcmV?d00001

diff --git a/0.16/_static/images/arrow-down-orange.svg b/0.16/_static/images/arrow-down-orange.svg
new file mode 100644
index 00000000000..e9d8e9ecf24
--- /dev/null
+++ b/0.16/_static/images/arrow-down-orange.svg
@@ -0,0 +1,19 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="13px" height="13px" viewBox="0 0 13 13" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <title>Group 5</title>
+    <desc>Created with Sketch.</desc>
+    <defs></defs>
+    <g id="desktop" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Sphinx_Elements" transform="translate(-119.000000, -5164.000000)" fill="#EE4C2C">
+            <g id="Main-Copy-6" transform="translate(98.000000, 4927.000000)">
+                <g id="Group-6" transform="translate(2.000000, 220.000000)">
+                    <g id="Group-5" transform="translate(25.500000, 23.500000) rotate(90.000000) translate(-25.500000, -23.500000) translate(19.000000, 17.000000)">
+                        <polygon id="Fill-1" points="4.87283333 1.2818 8.83956667 5.24766667 8.83956667 2.6832 6.1555 0"></polygon>
+                        <polygon id="Fill-2" points="8.83913333 5.2481 0 5.2481 0 7.06116667 8.83913333 7.06116667 4.8724 11.0279 6.15506667 12.3105667 12.311 6.15463333 11.4036 5.2481"></polygon>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
\ No newline at end of file
diff --git a/0.16/_static/images/arrow-right-with-tail.svg b/0.16/_static/images/arrow-right-with-tail.svg
new file mode 100644
index 00000000000..5843588fca6
--- /dev/null
+++ b/0.16/_static/images/arrow-right-with-tail.svg
@@ -0,0 +1,19 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="13px" height="13px" viewBox="0 0 13 13" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <title>Page 1</title>
+    <desc>Created with Sketch.</desc>
+    <defs></defs>
+    <g id="desktop" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="01_Home" transform="translate(-1202.000000, -1930.000000)" fill="#EE4C2C">
+            <g id="community" transform="translate(-18.000000, 1075.000000)">
+                <g id="newsletter" transform="translate(740.000000, 843.000000)">
+                    <g id="Page-1" transform="translate(480.000000, 12.000000)">
+                        <polygon id="Fill-1" points="4.87266 1.28206 8.83939333 5.24792667 8.83939333 2.68346 6.15532667 0.00026"></polygon>
+                        <polygon id="Fill-2" points="8.83896 5.24836 -0.000173333333 5.24836 -0.000173333333 7.06142667 8.83896 7.06142667 4.87222667 11.02816 6.15489333 12.3108267 12.3108267 6.15489333 11.4034267 5.24836"></polygon>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
\ No newline at end of file
diff --git a/0.16/_static/images/chevron-down-black.svg b/0.16/_static/images/chevron-down-black.svg
new file mode 100644
index 00000000000..097bc076ecf
--- /dev/null
+++ b/0.16/_static/images/chevron-down-black.svg
@@ -0,0 +1,16 @@
+<svg width="13px" height="8px" viewBox="0 0 13 8" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <desc>Created with Sketch.</desc>
+    <defs></defs>
+    <g id="desktop" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Cloud-Button-Sequence" transform="translate(-1311.000000, -3818.000000)" stroke="#262626" stroke-width="2">
+            <g id="GetStarted-Copy-3" transform="translate(98.000000, 3249.000000)">
+                <g id="Via-CLoud" transform="translate(855.000000, 212.000000)">
+                    <g id="Button" transform="translate(0.000000, 328.000000)">
+                        <polyline id="Page-1" transform="translate(364.566667, 32.500000) rotate(90.000000) translate(-364.566667, -32.500000) " points="362 27 367.133333 32.4998402 362 38"></polyline>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
diff --git a/0.16/_static/images/chevron-down-grey.svg b/0.16/_static/images/chevron-down-grey.svg
new file mode 100644
index 00000000000..82d6514f250
--- /dev/null
+++ b/0.16/_static/images/chevron-down-grey.svg
@@ -0,0 +1,18 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0.4 0.2 12.4 7.2" enable-background="new 0.4 0.2 12.4 7.2" xml:space="preserve">
+<desc>Created with Sketch.</desc>
+<g id="desktop">
+	<g id="Cloud-Button-Sequence" transform="translate(-1311.000000, -3818.000000)">
+		<g id="GetStarted-Copy-3" transform="translate(98.000000, 3249.000000)">
+			<g id="Via-CLoud" transform="translate(855.000000, 212.000000)">
+				<g id="Button" transform="translate(0.000000, 328.000000)">
+					<polyline id="Page-1" fill="none" stroke="#C7C7C7" stroke-width="2" points="370.1,29.9 364.6,35.1 359.1,29.9 					"/>
+				</g>
+			</g>
+		</g>
+	</g>
+</g>
+</svg>
diff --git a/0.16/_static/images/chevron-down-orange.svg b/0.16/_static/images/chevron-down-orange.svg
new file mode 100644
index 00000000000..fd79a57854c
--- /dev/null
+++ b/0.16/_static/images/chevron-down-orange.svg
@@ -0,0 +1,16 @@
+<svg width="13px" height="8px" viewBox="0 0 13 8" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <desc>Created with Sketch.</desc>
+    <defs></defs>
+    <g id="desktop" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Cloud-Button-Sequence" transform="translate(-1311.000000, -3818.000000)" stroke="#EE4C2C" stroke-width="2">
+            <g id="GetStarted-Copy-3" transform="translate(98.000000, 3249.000000)">
+                <g id="Via-CLoud" transform="translate(855.000000, 212.000000)">
+                    <g id="Button" transform="translate(0.000000, 328.000000)">
+                        <polyline id="Page-1" transform="translate(364.566667, 32.500000) rotate(90.000000) translate(-364.566667, -32.500000) " points="362 27 367.133333 32.4998402 362 38"></polyline>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
diff --git a/0.16/_static/images/chevron-down-white.svg b/0.16/_static/images/chevron-down-white.svg
new file mode 100644
index 00000000000..e6c94e27b64
--- /dev/null
+++ b/0.16/_static/images/chevron-down-white.svg
@@ -0,0 +1,16 @@
+<svg width="13px" height="8px" viewBox="0 0 13 8" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <desc>Created with Sketch.</desc>
+    <defs></defs>
+    <g id="desktop" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Cloud-Button-Sequence" transform="translate(-1311.000000, -3818.000000)" stroke="#FFFFFF" stroke-width="2">
+            <g id="GetStarted-Copy-3" transform="translate(98.000000, 3249.000000)">
+                <g id="Via-CLoud" transform="translate(855.000000, 212.000000)">
+                    <g id="Button" transform="translate(0.000000, 328.000000)">
+                        <polyline id="Page-1" transform="translate(364.566667, 32.500000) rotate(90.000000) translate(-364.566667, -32.500000) " points="362 27 367.133333 32.4998402 362 38"></polyline>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
diff --git a/0.16/_static/images/chevron-right-orange.svg b/0.16/_static/images/chevron-right-orange.svg
new file mode 100644
index 00000000000..7033fc93bf4
--- /dev/null
+++ b/0.16/_static/images/chevron-right-orange.svg
@@ -0,0 +1,17 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0.3 0.3 8.2 14.4" enable-background="new 0.3 0.3 8.2 14.4" xml:space="preserve">
+<title>Page 1</title>
+<desc>Created with Sketch.</desc>
+<g id="desktop">
+	<g id="_x30_1_x5F_Home" transform="translate(-864.000000, -1683.000000)">
+		<g id="community" transform="translate(-18.000000, 1391.000000)">
+			<g id="cta" transform="translate(741.000000, 277.000000)">
+				<polyline id="Page-1" fill="none" stroke="#EE4C2C" stroke-width="2" points="142,16 148.1,22.5 142,29 				"/>
+			</g>
+		</g>
+	</g>
+</g>
+</svg>
diff --git a/0.16/_static/images/chevron-right-white.svg b/0.16/_static/images/chevron-right-white.svg
new file mode 100644
index 00000000000..dd9e77f2616
--- /dev/null
+++ b/0.16/_static/images/chevron-right-white.svg
@@ -0,0 +1,17 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+     viewBox="0.3 0.3 8.2 14.4" enable-background="new 0.3 0.3 8.2 14.4" xml:space="preserve">
+<title>Page 1</title>
+<desc>Created with Sketch.</desc>
+<g id="desktop">
+    <g id="_x30_1_x5F_Home" transform="translate(-864.000000, -1683.000000)">
+        <g id="community" transform="translate(-18.000000, 1391.000000)">
+            <g id="cta" transform="translate(741.000000, 277.000000)">
+                <polyline id="Page-1" fill="none" stroke="#FFFFFF" stroke-width="2" points="142,16 148.1,22.5 142,29                 "/>
+            </g>
+        </g>
+    </g>
+</g>
+</svg>
\ No newline at end of file
diff --git a/0.16/_static/images/home-footer-background.jpg b/0.16/_static/images/home-footer-background.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b307bb57f48511ac8f1e76ac6bdf6cea97441b21
GIT binary patch
literal 38907
zcmc(I30xCb*Z!RZh#(Rdw4y+$XrV3zmsU_<g0`Z*R#02(R<Xr}wkis)AWW1hYL$p$
zU1~|(iq<OPhFUdPtwB(e7Ah(#C}3DbmV_lCnf%X8LP#JWcKLt(e%~9(I&<$?pYxn^
zwGXrfsO!XW6UHG7LufqwAJRTU+>h3LwGg3?KSl!)LY)v6_Z6~$f5BT&ANYTS7(zzV
zGnPUjv+3X1IqF4i1@a0C4O<Zw5gNA4W%zrZ$ZO1mk1eQgfdA-!51{|qHOFb&Lq0lV
zQOdL4`?6t7j5Zy0>x7-N6f&`Hh|vvWcEhwc5eI(R5~Kg4KL-DUF_;#XR;*6eHl5)O
z=er^X#$+-qn3k5*&tTE;I<n|y+5PRI{#HGve8K9s+<w^lgoB-2$E05GIrWdK{=>fv
z-(YRi>#g1neFnVa=Kk)$5$}I6(#w0)*bhG%7chRp#A%;=I{mX5GiQa&U$Agd=vQAy
ztXR2f^_sPjvEOdo6t{WH)}6Z&cYnWU@4o$q4j(!CQ}VInKc6{!?)-(ci<d6{cH`#n
zx5OE@Gylxa`Rjh}gNKivJufUOejzI<Evv4PD{AZN8<Z;QzA(hp-4^_>?#^1=7c2tC
zu&`iSu&DdO7^|rp?`C27_E4+t{!>_AEbq~8*!oWPV-gOgUbl7~KDDanm*Icd^y)w2
z*??;5*62Iy*o|%YTX$x-vF7{ALt{ah{@H(8_>zD24e)906A;UD+DFKiiGeUNyCEEr
z(@Kz#!Fy-=lUkKbixvdSOZj`T92A6cEs;I-(0x!yY<8*^xvI9sa6By%jb5^`W~sGk
zJsJFcC}W<#14pJLJYt4(J2!n4z5yF}VT~^#C>g4_t#QW`Sh^@bto{e3lo?9>tPOJS
z@I!c)dnYt6rYH|YX_3c0-upj_q6Z3;<u07q(@vtgrK2LYoz`D7vhx;N?@o&*Bxm5Q
zSq(fV-32jXV+C>zu}8-3!Au~DbizYokIw1t$KA_Yi^_?yYSCb+MS9U9xUaX>BA+0J
zRQILA9fYKTQ><Lbd&i<1e0Wbyec_0v4|~rJ%T_mztbq@G%Z$IGMf_g{s!9)XQw}!k
zjqdTatMQ)y0-uJ_tK91Iw`tL3zo~>B{)lrEOR?P6wIJaIUVmUdyMEdZ(Uy=`#J7wJ
zgbMegeKSVpsvA9O)S}HT_H_lT3bq`!_MzxEUVouoz;w9sQv3;Di@5gK4@SGFU!(qx
z8~kfIMFgJ78$2M(&T*;p)xKKfe_g$&ocmVy4`V)&mrD6JwCKLC<Rf(EWYTCY5`4kG
zLnfj!YNM?`J$t?Uffgm?R(q>sf>`v@`p$NA7Gx>BRr6!cVM8d9WA45nAg|$vVS8|Q
zYA@jO36D%_A(>jKUSMs|zS(%N<iD0!=C{PP{K}DXp_N_^f(&}0SC6(Hu!VO=KFLc7
z8Q4QH?V@TGhrE`EeLZd8Db~@OB2_OUMWhI_oc=Q6KoyZ8lXH^qU;}jOhP+<T+Nec-
zS!$8TbF@<Typ6RZze<Zh1mSgCSoT~&i72Z^e5XkW-TQOo@GLne3Iuq=FkDiC?~!dq
zvvewE-<_@0BBoBfsaI>rx$1?yUOMr#^Q$ENPT}R}1S$uijlxL9+q%6DzNN?5u)(r!
zr?_R}jlsvStK<A`@jj?{*)PsE|H`hv;xh3kQ*Xw&^{>l25M-%py3kQU7C&FVc2$e+
zy`_u^zNwQPy{)kB`KzPMhg0LN7MIFV_~kS;F@bzZsF2Y1kvUs}mS+?+23a0^HFFtf
zFn-0ZH^J8^*o%|8psxb@jUSfmcTKdph%YJKk9JZk!T5A>Rb(Arv{F5p;3Q)7La?Wc
zS;;jTyO@2Pljwi6QWP)dgay{HYw8L|VD8=yT7+jSV?IIWFzp`(@pNY_Sgau=zN})&
z{U!rM%MRA>f|u`TP7Quc{7jA}7;9O(Mg}y45r4SMow@mMi)8YVpBi&~4}V5vhiYc=
z9NyNViI+KwhZ52!U$az;zNd7wXCp_V=I7#Z4npfDbHrG%BtKaG62y&kVd*tMnD8ga
zAzMjabWn>%HH+COLLgILtQT6ZXC*7i;3Rxt^J{I3Ad~+q*q#%7RfQY5DU@+lC-eVM
z_$(1xwEnga8ZEUtdVbTgdWR{GhUW0=_+DDHx^k6Q#t7`(&&Li1Sq6DA*K2UesR*uJ
zvrn?`PRZdzm5SIgT)Di`_$3=uTs!uM+f9a2V7!IK_Q8ARj^YW^%J2+Hp-Lg_(mTxl
z*#c!?z35xb+Ws*!DsL_4z7<NjzX8)FtdtNkAo!-NTFK?mUst<$StQ{hm6ax_q!#BW
zZL^#1(zu#a<V>)a4(E2?z;*?1Baq+c;o3i&g$}VCr52eF3q{I0&K`VLlkuPz(qij_
z0>w)rgFMWMGngCt6~+dCkl_MiPw{@Vx9hk7*te|wZPF&+_mAc7*F!(~C`FIA6Vt=y
z=0z>9+>a!Q&+7`G&I}-Z0u}Y-&j^&C@llKz447-J<k?g3MmmMp-HwVRPJu0$p=?ql
z#)bgh?b9-to2_0i$l^yf3mPSa%0l^@31Q(YEh;Q4B!8WtMF;nzl(x1N6tSaD00mhj
z+^jkMp#v^-K2pc5RyikP=fl3<5gxO4bgGkYVg!GZ%<X~GIqYkm^T~L2p!_BAo}b<g
z>VJa~8z0K}w6*@hH8@^N28Xbd_?1KQ+?jsuL>isDKK7XtT|`zUON*kH#T*FF!A>}C
z>Gx`E!sFG>;#2HG2*^s_z(H?Tfh0P{%xMf_X|s)_Ply~NxIG3fDz7Eyc+?c;;hxN6
zMgd-b%9Mhj*Y;NX%`;dR>J9C?CkD@JPh$lsCq|(qZ)zEwChz0yS$H7AbwphC=+Css
zDfm9myK&cdmn27Oh?0doH(JsXW~Y`?k%{;P5uC`FsB<M~`CT;W<gevq^d4hJWZ&Hm
zG=#{?KM-V*na+<XRIgWGz4SYxgBOeoL9wfvj8n$U1o5*V1Y}x`>`{}WwX}$Zz>|tf
zM_`^V+xT*|XphK2Xes>nJ#eof6;D^CKVv-&_j11RQ{ZVonPZUE<0<_I_O$YyP$tev
z(juV~uaw(eN=aJlFw7u-v4TK(C1vAx&vwl(Z;CqTh>lXy;##~xy-NeDr@*)lw99PM
zZYvTBmt-s>#`t!6<<6Yd_uIW8uS6?r>*r1mDDzbg#(eepLKLTF6zPWZGby`(gsc;g
zfhc3*@JampyaDunX>*LFj=GF}vsbqf=Kjs}h1Fe)qD4=`WZW)#W7N@y@Z$?Vb(*<>
zf82|6TxfT1M7KjOC0l>?+lqvC6OH05>yRZG>}RFfVD$X<{f=n;Wqh=hM>=Xz9o{&y
zA;=<8XKLs$zV-dN_jd#V(xN>gFA%j1%7mMQe2(x%wW5Q%^elWh6&%JYW7dGQ(Z&_?
zcL5Jlg=f`(n0Ioe45~+Q<*iWWj(^z~V_!ddn7sJFX7SUo;03k#WWjc{wBO*bdm@`r
zm|7z_?jto=39Jl!6d8O_lI>b6v<@}09eQPi5ZLcK<G8rvMv%Jcpou&DUE=+nK0;sR
zJ>KHUzftG=gHrwU@uFj|&SBy3hTXG$oUihgPDBrZxEMuK)`t04>h)5!XsqC;fO1+}
zINHC+scB0>N_UZwx}=X06mi$YF!dO;RHcc~qBAK<1%&7ZWdUin?g?K<*&59_oV}N!
zw>37Dptorqs7rqyT$$E*S6u>-(i`|6jN7Nr{J9JM6g=DPz|L8Am0#U8YK1zKdCb(}
zFqnT4&Z2BiZWV;IU&61^7kZhg01^6F1gY)sO_6E`lIuZBIhu7^^brW2-#mk`U58|C
zwn~>mvs`hRl3lEPZq6=deI60vUAsR}(@BHxDdzSpF_HrM!f5I3_>+vgSFEPr&e-N0
zn9}GPD@k!>=G$Xw&H&>D9Y%J7??=@)ibZqJVFQeI5cXUgR;+T=41*x*d)<2IewL0%
zu3r<$9zwXJ)`x{}Vgu>T`zmiv<;`%8pkW4h@YJI1`Ehq&hC;%w!%Nz?e{Ai2a!Lu7
znHM9FJrwPcsa<+7lR{0_qg8Hf`IUzYe?uex(7OB-X06&TlyNC%+PS4MM?7kt@N=t#
z=<9dYQ!YB+94XHeWLFD2cYAW(4XYCbD2X{0xGa7-w^u0RBM3=X;YGw0{*4#9bV>i4
zZ5a{_F3N7erG<XK5Q1lF(ZDx#-mHKlH_lik#xGckKc2K+d7<#t&{A%nbC~#viGXx-
zT{17aDn{@p?>*X;fy!EJ)$nBknYcW`=vsBL$x;;<y#-|Oo@z8ox388K@c|74zeX+R
z_OR*#D3C-ET&xUv(Hu@vc}uw#z2{q^gSix(EV;qoJD$%~uM5<o&x5sS%m~b^@9Wj?
zeA3d8?VHJJ<!4nxctn@{xN+#p{z-T1k8FwFG+Sv$3{|a5XO9~3V=~|GTQ5LlW@}MO
z#eLqO2}hsYvhLKMc_m=E_*Ax>C=~J1M_}9yM>bh>yKXTr_7wz;rCjUR#j4nduKRv)
zvX6eDsB@M%$2qOJ)tkBClYtwj@T^#HbNES!UM8MQVh77BR7Ir7or(qxYDR^ouH}5q
zZjt0vOpu9H)@4a8vUTR39EhZA0_#JxNNR}|HSJ1!QM1*L=WnU|$H~a<#OA1)n6O5f
z?^<trY!z(L>r667n+y5?VZxv#DnhW33>H-JCfTiRzJ4$om}Jz4KRhlx%{;;3`UQ)V
zMkm}Vu>I=Qk-a-6=e+z*szp~NIN$k63FExZ2=o$O-sxyt8Te|b54&!%U4HS)2yWjg
z6L-0+@OHXhmMo1I%bmZi6?V3H<CKckHBxX~Xkl`7Mr&=>z;XJFh)+m)y2=Lc(V!?N
zokE#sw)Yw_KKIq3jU1kORIL^{pMy}T@RX?VIiV`kB2_nGCwE3p*pA_fdxRiMGtRfF
z3;MW8NV*TI!v#vtDZ&mC$tGW7k(t-RsJUbQu1C?N-W<7R*>Gjdie{i`1xSp9N{V|h
zK>PqErN9r$E4@}1ipQ<85q|HBAkR~hpJ1tB^AnyOVNCF44@`cr%-Z=#%vgKu*pY7W
zdn-ssJKQHy%DnDUL?H@SmA_L#3dATp02SP^^c~^2L=G6%bDrzPb7@zCpSnpN=EhCS
zu3UGCd8(6m>Co>{amwJMXV;Cq+n2MG>)H3gwjY0CEk00dQLrB+H!X!(ZDq%Qkno$w
zw46HTC7+Rj$=TjY?gxjrhXj2mUHvNXc<F6b%$y*dg3y_}T*H!k*9moSCKw-GIPDY}
z9U0Rb=l9b`tnRu^#>6AAVHaYq35skfq>g#)?<foSV8?D5uIJG=vjbn8x{qI(B554k
zm>!|<;m?WS_Bz%^CPx5zN+}A1=Xf%$l#KpC8AJSrd3wKS_HX0n{X8=sIe^KJwpv><
zbU*ri|MY!X&Oi2y8mJ<sM(&bCqOHyE-fz9%nrRoeDY(+3CgbIr>#7~;PK+~`H%;69
z#g1scl8giK0oyTItmc;uZCICX5H@OwD7=zn>C>e7grHb`wc##W?si<YA5GLHU}lw}
z)sxKG|BHJs*Vc<)JV}ez)w-s9<oDz3Ywpa0#y~wL{+tlsgn0Nb$WvUoU4xiXIX_sY
zbbeY`{X3dzHZlG05eeq^mjU+L0^a3>7L|<SxwJA1?^k?LvuC*$&3HmQ<UF~B<9p>|
zFDstMiAAxH?y$22Cy^JOa~t_JAEN{{eFH9Opd&3uDlY}aEv~%hj{O<nbaRR0cb8_&
zi!Z2iDVP_V;Z>Nw9~CY=`RM+_Eti;yVU=%}#9ZYcW^P3%+g27~cdp~kWvt_LQKfIM
zs?F1qKED*AVV%}Q7Ab<PX<38)Vtl%;DiVHL8e&FK0RERfAm>DfGG-a=x+N87{oHJl
z9Pf1$>oB-28x)>8&NkTslB5K<@lojErYXagI(ECk9;@6Ho96~GEd}RNZf~ww9WUA<
zDXig9Ss0+WIyHfBxR9T$tP3j!AKA$wUADjJ3J*rPb?#j$A4z78QWYtDMFlPnIVNHG
z*QI|J`FyBQ?kb6?Qrod(jx@Sk{fNf5sSVx}ETRq&Ng`SNaM%X^wAv9U(c}yDhNX=M
zVXwEjxBDaC2?gw%Qr0w!F}<h0t#<9NN-gHED^e&B7N3ZDnOC^_;Oc5zvy#wsl3KvG
zTND&ANg}yMeGI^Ga{*uf=`X1;aJf>DTig@|wjEl<Laj_tQ!lf}hMjiJzo<nngvZk5
zO(jAA9;vidpQ8G6WLk7meW{MZLuQqeer{THDh0r}K4?R;(V-2gTiVBe&Q}EgP+d_f
zl{c)0urTvsK`luL#o2BOFvWm@mS2ZR&to?sNv{4oD#9E+{o9WhzcWRi9v;1Fq%9uj
z;y7}$=kmv%ERQzz0Xw*etv<h0l<tc^Qb$GPZ7T}0+Gvzj0G?b>TT!X9i#|h4)Fqet
zn5b2bX>d$Xm{dT{t3_FIq221K-oHkWcc9);2O+4?rVVLm^67?vtNV?2x7@J%v3Df@
zj;(5bb>%K@--*|G^TIo2&|pAI0Gb1ka+3zQn@==uU0RZjr{;md-A4F@GE9x~;D;CT
zHfzzN0fFjHM0^(%CT*vTtSf5peFJl(j8+$JScX=2YW33(R`tVAD?_{jSG(@gB8+uy
zt!wfu^cYO25t_&vv=TC`9)NuS_9xwr{9*Ha?Yux#67IzuuD%N<pp61I)^B7;1ElZ|
z$lwk@F;w{&&xI%n8+L#nqeU{NDbi9z#3^)0i`_T^1XD^CpRuebrcTMQVf1}l<}J0?
zqI1L^S$UZ#-ktG87oy8lww%R~?Qj5TH9;8)THuvc<GrCI4@wZT93r(&aDh5}C?Ke$
z<cX65TW!5u<xFTk!EPqRkB+DH%5)#;zpTIH-&ux#8=fB+cwDU+)d%IUdW(A(I$8O8
zNowk$7}+YBr!6pDO4ATO{ULcyoYa9^`Y;zTp2D;m)v^dhhQ`$B+pV4g?hp+9N?g$g
z9d1+4S{g>n@(C9N@A5O(PK|OdtXvKHwZ<MhN`0KB9~jbU48UrGT$@iAZvLQIEUNuO
z1TjK46KTZW7pWSE@<Ws(DLF>nN?T~Q9k&*r_WG2}a1+eJaS8X|cUa`b$q~%+_WO-D
zYVV;LAv1Ch#b34dy<L7AbYEbP4WR4@Ye;Kpw#P;t05X7}NRRy`KtcUSzLJv@Ql&K1
zQ2=%NufQDdyriY?_?;1v-^K{CY9Q~>Z?pb^@feS&9)T^GEyUw+4WGPAQujjVnjpy0
z!)QY|$(Sq9t{4wuN?zdi@=H|<d0uIam1H9^oNL=EYN7?Vi&sdEl>OW%EjPQ$akXUb
z@U4L4*i)dWpuAN+5o&bRv^mTLW`rZ@rlJcYhM2<fWc+AC5F)q(WQF$)mL0oShF61?
zew9qda_x=3%8Wi4z6)MsE_@pNfoI@TM^Dx2vPjYqQX;^>VV?otHWE~G0lWC}=VwPY
z&Ir2`qKs{ck)_T5;gx+IcMds&bDXmqEd}#S>4N#fv*Y@*n4fsvTR7OZFTBg$QF&h~
zU->QS?O$^3sxs_Gyv$43Wuh)J^8O~Jp+P`i25Hbc1=hRQ>nh1*5vu65LTk{Qw%|_?
zTMiUPTB4$6V0~kcXdP<>F}+L$^p;e4Qw7GcLab7X$`4bjx~au|4IY!uV5bzToJCXO
zue{(54zk$%g#V0O-UP-{V1r1a=&hO$tQ8fIXPigVfoth5RS<JMoxD}1DC+<qBDyjr
zxb1e_$$2&opKHg8Np^w~NfF$k@dh_3!S9#D)(I|OA$>(+a7Elcv^%3KMeyxZL=M1G
zDM|PnQ;ul!(<Uh6c#@x73;=VQvR0DWEQ=AB`RZuA@i-KBq?Ub}!c}*(S7q|#eg~g4
zRzqMHMQx>2E|n!aeZu_0<mTbd*Z_P7rG3HwDUc|>LA)1TUBjDb3beF}my_2z7uk(U
z4z3TJmVZztFTE|ay=-z-A1r?Rp~H^&`L}#`tf*}iCmr4fW$}`xbu))KzL|Z+!tdqZ
z+ZDkqe%lfv65qtgT{#LR@eV4v<vBF_0uG?`&rKk-`IytX+pnA(>QO}SjcrnkL<(Cp
zi}{RCc|fiu*Tq~iscq8CiBkj>hW_L?&y+jSx4D)<XFhhT3N`EUYyZ%qAJJG!VGSMv
zivQDMU#{^fRB=jcU+{*sVB}%_L?KF<vRYN7Gx;>n3P7Qk0r(b@0*3a@_E}95ldb&r
z;YaCM*nVZ-ZR1YI%JDk|&Rf|`Vc8AWc&_gnp)GwOtpzxH%Lgt_k|DMDvp<{Q3;=!2
z0yCjTQZ-=uJ*>Rkl-Qtms>O>}7DW?#GV^7*4cvEhI>Y=ihbS!7q9=hGE1tW~Xc?`6
z>57>C)#rnnrIQxDm^?}zpZ|>e)~9F1giKKY6&ldmV@v?d10_BXbqOGo1Af`&!Zc+a
zzK1gl?RAtJuXOPe5!B%Gv*ct~>ou-9ebDWTyqgalZw0adB3w~5wBmL~y%lee{vP0=
zSH+TTcRDZclSDkipE#29?O4ZML#;{>VxeU$Azi;{zECp;7|U=@E(^svpr7y->weph
zuKcJhaBli)w+HMh#{{mE#gt#hryhG2F!yEPbU&GD-U=W-rh+H(QJCJI*}#EXTL>YY
z3saQ8L6oQ`N}A76X+)h2C>PwgYFY_8TMM~06#FUrYXqoMpt6ic4^xtrOni^Cf<g#c
zqnDG{!c--|?c%-N%1{m%Ct;-p%Gay!MOJS`KbpodI%3iqO+>?LASTh#C2g4HAku!k
zUK9Rw&+?WDr~JiFd!mk9pYM(FJm;@Uc=(_wRgx`b&jU=oR?0NH&Ph*3PT2LwT6I*+
z6%s_^kLK&L@2(G^?@_GJ7K~9^J-eTf#pWu>Sk6Al`IbR$0hPCjYYKHmLre6xO|v^=
zMsP^%0p^Wpc8q3}LS4d~3=~1&4CzTefPV*;P~mJysWRpwh`6j6?Kkp!#u;4q9S-AN
zdb;k2<P^ic4Gyweabf;~L*AUnz5&vTIT}0mLD@iT!tTUp>!+2B>w*S=75e!Xu}x<V
zcUUMxF@64yKM$~9F>kAOT!~@%720EmJTKZX)=M!SZbu0G6o73;g5~kMQ6qD{(p=W1
zY(lGW`3>y2jWE?lxN|acnjq{_K6^PMBGxrp4$hDNg~Gu{zRlfoKS2D(1u<Gf!*g*H
zXhWA>37e7Q7F6E0Ec?j^ii*6dQk5*v@r5Cxrt5&|!0<b*z9dc=(NyB4MFv(jDX}Hs
z(Oy&n(Uu-;EMEd<>MhZEG@1WE@e<O{`<ieXI+a>%eZrsoL}Vv86`gu7lra&7JUL{`
z{!w$8eOwa84b0tjj`_g^=jo4SZl_<egbuhQDcLc|yu`aRUV4ATYJ8mZc-70Oi0r&=
zQcIy(`nD;<z=Eg+th+{C-2~t0!b*za9Tf%Cf&-0uSX>`G>VswFawz2=gHqt%%<go*
za+F!Q&AG4`j~m;-Pxf{?limZOWGOB9y70JGTwK#47y=aqGBHs5&O){gZieL}WehuE
z>uKxTiT3+%e<0;37qX<M$(fp=8h6x7YL~#jnM$q!3BC#h&QGUo!~%r9DN}Y|7bF0w
zv$h-ry!`d-CgzR~LmF0p9ZgB}xy`C%`)kspR&a(^=1Nz*!}HMl8Dp=sUNQpA=peIZ
z)bZrG#?>;;A5aq(Br@jdC<*Nd^~(q~8Lat1?GD03TSgtnLXeWHTqermFIK1P`H-0Z
zb%~M`0peu+=^0-Qp25ii#x4stdlmZ4ryg!Q;Wt1R1hdeNiKJ*ftYaqct-BGA*%as}
zSK6R&##hwu6^Qv6^I~GF%hKCO<Y0dfM_^@yq?G&CvG$71`oINf6b97M(}gc5jj<SA
zc(OYfXdaiyADSYhX2$?zqxk%+((8H#%0GxnEYZvh>%8aQ!KX-Sgq;->IhZoj)J2+v
z$vH+bj@|S;%4^(Ji|}VHgA-tz0UkaF;<^)L^iYGhrEi+?;ZiY#lU_P*XJ$u(iU;S;
z)l~P7m6aVn!=I}~BXiW(0OOW6d%%{3BJbJ!S~-y$K(4^L7zL;;wTx2Pw&DnrbjwKW
zTJ!AmsXC-pHwF8S`kfgm#{(gLvj7dynesPyu&A=#`81XTr51z-e~r$<1)48F=~;YK
z)Mn}_%)5Um!vHhjK|Ikz6RP8j>*GDU>0&35Bo)w#S?C58*NPylwFr8noD5dSa_XnC
zcEVU?kk#q8d9DW?K%G2b4OoECHVKod=QE{(wA?=<<%CF)j@{}$1%caZpw{ZF?VXND
zkR{rQEOstKWgnxjXXkZFU+u6%;j~NP5Vo#5&+9Il+nPAvWYB;=FPDtf3@_(8K4~jg
zH!0_L@}(uERScmMe#jCDyd|NY0!P92m{2`zCqVW+0YIZs#$`<UtBqGiZ%0}`q;>j@
z9U%Zk^A+{b6DR*IDN;AaT*6cIj7R-1+YYL1P!flT5(aPJ8_2qD3lA^NJIuVi6SouP
zI_Gak1Yn;AiO~g%K++vqBg^MNlTCAF3l_s<<xTT@aEkNRRDuxt0|-#6XFt*F=N+4)
z`IwFu+rCooZa;1~)?XmsRpNUu28zSq_%A)ZO|Y-=f|qlalz(bi^x2Il6*N;%M)ip@
z^aSbky}kUwlsso__1lsf;G0Y2g|!(beHc1SUAfN4#xlDTwSPj>#XVlXX3N+id7)Rv
zV?B>0G|+`!-!s=T=n_;J(~W5JLKl)C3~^0)9?_r{vYRi<GeuNbOoEvc+MwgGln?@S
z+*Vp{?rXun8RN93{w~N%j<LLe5B?>9GHB|H(2&q!dC}1w)>M+waVC8*j@MrYI$A$l
zU+iP+vC7vkHg;FTyw$)<o$kJKz@x)ufzw_Ea=ss03cUD<y10er&I4PI41rp{u0@Bw
z>xeW?`3tBa^3^4bKDxIWZ5v%-Ru)tdlEy(`7V4n4McQVaipE#~o>a-IDh;F=CR_s5
z<%8#ThQ_LObhtDWik6?dHCDh+4s7+4&HH3@9mY>a+Fph}hF_aE?)By+feNN1ROP9i
zScif%)%T*08C(>$F&gS(9*RH>3bNEMV;k3G#ys}l_AK{tA!K0S`;tPL01E~(!~JU^
z$@U_X4ZLl=73)x5+JeHUi$s&IwoI>HV`YGkfObY=WMB;}|H}^!yJQjtOyy7s$6qkU
zzcIa!3qnk^43$_mLB*7UP^~sGGc*(jPv2?-&tmr;6DdECmvOnUQ>YQ7Wmuy%;iGE&
zt^jLe2v|c1))Ko$g3Ps9OH2#NPA%2o$F%5qJ1OK%fo6rEG7Wz7n7Q9HvlE8v0gzzL
z;yNIR$QyzzFB?gosa-T`IWV|4<Aw&{i>Lt4Fe+Z^17?aB4Ly1O#!~b}Fd@#;^q|{C
zr$SiU?)A02AxoA8XY76i*xY0`&>a9AgEZ2V_-_!PC93G<%5^8n=!hDgfyPGdU8~f{
z$k;(q>1vEfvjBKji(cqDp;(^nkngMu(^eO$zbgxHE4jp~9uS>2KghCZdkDu}cF<P>
z<)D+|(*>uQCFXhr^>&tS6{HGBJ~6J%vedp6qP0J0Jb?dp3e06$WL3;1j>@L6TmWz!
zr+s;n$dyE0Se2nqPtPqA1s4OH_nz-rISL6RF9+7=DWw*lyon#s1rYtCCCWlDjHODc
z<wmnq5ZLkmEFJqNu@6;1%OyW9X@du+_ld?`+l;S`gtz_p2U@XC5ynAt1gj0on)RPg
z1B$jE{_8Tyj8#@BlxHvVhKL#O*o_={oQxG}0vpHno<{*Al$e_#x#~sP3^k+!5(TPg
zX&Vrjl~#ji0ls6g{yOksiinaye5gh|NBcO??<516&dztGx&aMY&-jxMNRG_3cnprx
zYR==Vw`)VlD5??L)!1WoOy{kqB^Q#=K@R}Y-{X<%O+iOed#Zvr2eRbDN<lf73XT3z
zOoVYc05egtWWG^TA6~U>h+dP@)4iI#)qri-3GL>!W)3hn3|}#o@K+|`lB;OaX=1Gi
zBl~lTD<oFD!Q+lyRC4jBl?|#GCV$<2I(BXf_)M7(Py}k;s8;<{-nT)`|JQ-E2<YA%
zIRF%ZxlcOV?Z$fnshZJ*N0vS2nb7y3n$3~rZf&A9QJxTr0H*caVNkK0p+%K_Q2d*^
z3z$(6cp2x&=$ylRL{Dvz5%ovc1#i;lPDxQv8?2aFMk_>D(8y@f9E#%>jsB6JV?c-Q
zA9?|5SI?nv0}1PUeU%ZFd;@&fWCU8km_}}nV%AaRJstJUF4Rbb7#rP$ZhkqK%NEHz
zKapl$V5xGxlzEf$G%O>e7Mj06mj`qcmauZ<B*CuMGnmceHq;NcW=b~6R8!klVIJMK
z#XRuCz6B6>Z&wX<h63M7^igv_+a#YY{Sh<vbjf&4KoMZCK0y|T!bcs$$MWSWVpw{2
z>m9Y3fIjY`{$SfBALIbfcNfS3T@u12{9M|E4WM!zt2v@_j;4^?4I+8HQ|{=4#qIil
zLTInbp{}frIF5ShzewPEr{!1IXwipszr-hRZeG~{2Ti(iOTiIn(KkX%xvwag%;nnj
zce<`-H;75X1}$4r`Kz3z=!6SEgSvDh`}rN_WIJgHJLuQ7nkJ2HkZZivbg%x7ehCKz
zSx{U7j>DoCD0oP8AZTr#JF3ae0l4@KkKObP!`~)pb&ra4amfue@Kvu(RHePZ8~DZi
z;~9_8v+nWVf}zmtqbuK*O&@lCdzv}`ZnYd5w1b#eHj;8q2JwEFy!@7wc{_Nz8ltD|
zQmiQDM9eo@bgKl*S{m1|mlL>0{TArPEVTOJw8pzy6fb4|e%WrLXsG(a1)QCGi085M
z+DRo)Mf(aZdzDP??5Su>#v7m!LLjt-UV}3!^*=!8uh6n_;Oa$MBo0N5gNdhvdKJ%p
zmiP&*nUj$9ox7?sLP?A-!9F?=^9#<Y_RmcY&g6OBf3`PrC`mx0CDnO%$vL)`j0dI_
zx3i!wpj<OkBuVMZeQT`~Qdd)C47plf!^7`|eWZpT{Eh>BD5n08_#lR|LL()Hs<Q}O
z!VJY^Y4t1_mz9+Ngp-6Hm*nz~BIhV#8T47Kt<YRoO7(BBdux)_2YfLuQ^Wk6n4J74
z!RkmJ`~1lI&!Fo=$gKES^B3IFG1NGXSV>fuA-|Syv1|+>BqP+1cusYt_(efOy%ssb
z=l&#?<JFa<H<y)ngnUjE=BvYbj%mAw{&uss%Kb~iMW$9M=5==WC~CYAM_Smh)zN#@
zs)&n>%)GC=+;^X*k>QQ+$;qfH$beIS#n)DdPx-*)moscBxkhOH(YA{(Lgf{GwP*`D
zw)!?au?ypS5&37ug;$|%?p}fI%oVlr`Bou2E7VNKook8+{HIYtOycZkWE7_nt`NkH
zyc?{1NmS-*(PAO9@qI#;szoA#lE0U5$4O#R18uUI2>(C$3^s26k1xW08>*QP622RG
z)Ga3D55cWH(xPhWilFZ)2?|UxAk6VW%uV0N9l1jkbk?HZf!!##L86EM0|aq4*$_-E
zf8A6qlECDhYj08S#x#i{`Cp>SjnF}rh=>Iq>e{fzXGB9W;&k{^hyTYsa?fuoencRX
z3t3&f_{ufJ@1Ib=4J~f$UX$?RXQJM@khXeAd!f}2@*t`!!pVnRyT-RG-AHe~<~T|8
zL8l&ldA8TE3lrR4c@Pay1M|BYis5H=kx{s&ahghbgX#z(<^!F{H=opa0GI%~^bW3s
zAqw@SM1uzTgqA*9C2stVbc0SGM1**#>)}p95qT?E`6|EuHz}ssCXzpaTZIog5VBv1
z{LZi%!(cU5!D{q^eFh(rQ~?y@A&Bz;ERt^?*)R*?se8z%G)>jH_P^7TyfwMr4zJiR
z028#QSg;p<m`}z8)7J_L(AWo8Dv3sg2;`pqG87oQVC5?Rq<3)2p<6_8vG_g=NP#Vz
zQfrSYU%n5<<a>}n*g7!?=Z#>cGOX#-|M_c-jZYH4-`acO_$>?aw;!D`bdKi@?7)KY
zgV|$ae@8)jbI;mW0XDMkGV+mxf#jW+VlSc~|2@TXe8(kZX-u*f{X!eACc_gY1e3Oo
z>NTH*eA;=K5*`4KNi0&`h5UOrTKy=mF_vu9qKXwL`lRS0UMT`!!!myM7cE*w<P-4I
zZ&eU42}mO0pX~GWtSLTyiGI1Hn8bmQ1?Kh{s~@f=wCFTP0$r7(E5N?{#A)imF#SNT
zWjDU!vC~GXYdG~!5NBzavIwudjQk$w%QJ|4xPXfwF0f6}QikSNavxEsfS!HoQHw}X
zu~Ldsqf_^P`E>61k3~_hey*APR*JG8uRnQDi+&Ou1(nKIWGjT0<Xxv?R}yQMlz(V?
zSld1Dnmn+xkcEh$lLC>M@PncTNVq6lHi+C0+AYwWrUq%f)i{8VT>`-h)?5QY+zlcx
zBQ->w3gKtK8VVX~!5Wqz@-}D<Q3?AR2H#-x5yug{$UrUH0q*BKN38-@q85b^*#s@t
zVX82lQP~b{+_Y#1zx?890l#q(b+1%s1g%)O8}*A|xj?mI0yQ8-6<~WnW7WE!cgMzU
zO(y4#SAjNwozp1<u3D=_8FWDK@^$~F?-Yj$#Q*3#ibjf9tmS|3<(w3b3}#Rqy^#{s
zS0O2sv$N$HkFYP6hTv;@s%mggOr1j}Vx!KS&fF1@rX<&>Y@hR-+Ka*5`$J=(5{v~<
zD$_O5c!mQtjY!vp-h$i-(Q8LWk}dP7&(pjhSD;5m8G=3^HGQzOxwA}|>WduQs_e1I
zAs>`Ya@yb-mGJ<yXVv!_7R&(Zq*x;!mU$<-tEGc#2Na{k#MDqm&5T(A+s6D79sC~g
zPKL(!Vq*@r7+rDi{8H!!@q=$8&?l(LX<)E*RTmxHZ@>yMsAadpcUy+n`ne?Fuk>tF
zj!fN(%F*IOZRfNtV`?BjDk?}Pd=BYU!kB2XO#N(9`nvKm@47E=b%}Ld_7MtKZA!X-
z@<qV+>-ntX+XT<T!d|60l&p1_XPasAL0U*m*SF<g?++IJHBz$zSJ_CdHUI|_TpZQa
zs=(-6z>qqnDj;QbDWt}*P!tfO%%F|I8g!0gE(PE~QN&1u>Fa?m*&XBS9SILPz-*-v
z{WO&aqN)f_c3EG|D|bfboN*5<D*WQqksNWD6e22{2SqEunTlIBK2ivbalv`IMN~bf
zh+SL(-Fl%0@IkK8kvMh3Ld;Do8z*++D;4;iAeL>%=9hF1dN?O=!?90^Yg#mr<UB2v
z#w>Diw>sBnl53Yi=i(Yq)Fex5!*a_k(boa?Q)~nJ2w~_R*R2eYRD6RLb+t$|hz)dI
z6##w-M&dKI3L5BPbRD(@SC!)VVKQi^XuJK%XiH%2^<7~SWt4m?%%ci`eZ^W4qv<^|
z$e{_Ar)R#HeyQOv=zjJc?N~@-r!XIHrx6f=C1|oqwLmrW7&i0MoWs1#^i5NTb@JUT
zw3`-`{phbN88y<O1+_zWsfSc(?ic#9T?b?P-Ym_;jxQjOi^>l|qBouQUh|xSUDUJ@
zlS?sXvoYAlI@MOQ9W9Nm(KRT}v=mt#<d`sEMdOje#^eMbnW??&d*48xZ!NWBu8Yfm
zB@jCVH*}phI_VC!rYqMKF4y-A?g{`DEu{rAFh!6ZLYG~*F9ZSy*oH|S&Ad#SdP`gT
zmfdB*yeRWK#Pf=K?MNrf2ZP;yunqV=uHis<z?oMT<N*INHfx#Hl@C0}#->GS(OTv8
zI){P2zc%himjGT=f{}jR3_XhSx+osPy$2x_b#xTZv|!hE+mde4z-Q>S{`K@{Qw!D}
z6~ol9HTCKYP1B-<<O~D855L{1+lB8H`7k5zsg5Fa0Qk#Y7JN1}o;ir&Fh0Yjt{>0)
z<FSF?OqaW6amBGYw*sru(&}e#Q}}E<ZMA?xNDO*l&>iznwvbc`5?R42@g_sU$h*J*
zG->7bo{^F*Qr?JYXw^^dt{X@Pyd{%)Df(4x(W*oZHZd=%@P6w<N#n1TX1q#HmiJb7
zQb%z6OqnBo2U7+&Cf8oS%~Al&^(BSwEsG$D6sr8UU<u@-DNW<5&QW*Xo$8vV*t(PG
z_BmLA)+i}+OZ(#$EjJH2dJ*rZzEmu+!JTCf*r$fMGnYQS)!Xyt+MPsn90fOAn&)?9
z*Q-%dOMS%fdQj4c&`c?g)d*u4i|zqQ(V{QC)q%C9a|TTJt^>>g)9p9M+cN#iroXU$
zX(0{7ffmDG2C<(*nj5I;attdo9Y=%x@^jgg^ab<AzUyG&d^3pk?y;{bzJ5A?ZD{YR
z$i278UUg*o@}#(g@6nw%K)bs9&!ZN?*w5MI2HmtKV9lDFmqlN+-Sk1=_RUl2=)%3(
znR+dFYt(F!Hbs)s+Dl#X-sv^zj21<mVBa~#RrM%6pib{IF|%OUj`VttEQOl=X=7&>
zjY~w7P|mw9(yS$6fEBO>=rPH(05;D@rlr(%6&|M)*HSZbt=*Yhmd@m#;K&-N%1d7}
z-UeE~Kj}Upow+mcf;)41Uf`Tf7t0}ox?VtB8;83@g$Gpe7W3>agODTgLQ!8fHM>w3
zA2c3dC3cPaEY%)L(Fn4XQPumK=HAfe%l6V*@DKT6(!xBos#e&kL}&l#NLrRxdt1{q
z$G1sPOf1ZQDv%ev@b!R!naokO{6ZaE3Dt~)=fPFA3JTue{rN;M+h00Sq3DCHsQCRZ
zz9(nj;SXUtv!^-Wzdg-=vUIq#W>~sIqRCyDy*2%lU(bLBqAwu3VwasDokhFR9GyIy
z5;6xESzc<qaY(NE8L)Jz$xR6Yn7Km@N7CCwcxczbQzMYul}LBS_~JD`gbR-OvQs&l
zm31Sq6Sn1^g_~YoODd2hkdvP>Q6MBxPLLm67qR8k5Mk1#&HSimqx02x7$KuYX9=Gm
zR&Vq5^8s5O-dS>Nc$xUsv3f6j&$I`9P~L`zE)JaG^%2)q9`ym)ennOGjNjWJg;GZ8
zbwn3zx0~uV1d=eOz(8sTXI<cYO=L2z*Q{G>(P{T@=RPePzl_XcXlpv1)tM8=1F+re
zZI8_F8r6;K_|w7U^C#9O6bnPHmxZ`B%+6~(c`MR++zjS0qtMhe`Y`>AIa#ck4G02w
zo~_N?A!^D`5+UH6YBgq#HE3RA^m0vK!3{aI4tjb!0L0*z4_LKP7^3467+M_-1Nx{Y
zz&~CifYczn<F?iojvGc3IE0B-N3Uk=>ZHa)(z~cG^vx4}8L<(ife6B24hY&NqP?DY
zaA1uU0rR!mNOh)Mry8&XRLse8HYb9er`MhIzMXlCJ>+qt2g@yASedj9OLP3H&$s>?
zIkNt>{Z)dh*VeaQBZ3PikY|8QK~4EWnFe)jXSRu{-S|&!p(rFftzm_}ISROcQK#>)
zKdx+8Cwl<H#ZT%xwp!-%_di|fUNG$R+lkvYu)lO?CgzN&J0CX8=bY^t-${UMca9=f
zUX+Z@$U+M{7B%)+g_q?6Dz8ZlYKnZohv*wyoLhqf6fvxn^cNRH?^KI{U2TaU6AE4J
zTR)v@{E>F~UiWE|s!NqGA6G@&c}6<U>w`oTZe*B&15DNc(Q=YY&V8&OL)bcbG1sDG
znKygv(5K37{JVT%$?~!Y#qVf|yA9(Q!gwsMK3Au2`|na_GIwd_0|^W|s4-(dL<g^F
zq7v^q`Xq%mJ(S5k&h!6wZ5^~i8F!5c^VdRfK#h)J?rt$8)AB!dzZeV>J_$b#V8ch~
zb7RwGc9t*fNDE7>r<Bt#FnBt`VXGd57MP`xfoyiy^pnhSk7Qo)V<jJk<Ufpmb#;B7
zBxWr%eIVf*B+F(xhUQ#I8|GA5&*Bo>$OCq=)n4hhtc6_?T@kWk`AgYhVG>Tfy@Aw8
zM~^L<gN@{*y~?#Y_i7aC1(%+SXwcJX^efi#p(DHs)AehQXJzW)s4!o{!tX4B-4=cC
z&Nv(mP<k{y$;+B+Za}|LA1~E0^>11_EJpf#7?DSTj+V7cZaiKIq*riEL{}ppNp(Ax
zu%!{|TltzFV9IIJICEM|!&8Hk3Hc^j6`&~JHnnN#^D^_(RWUQLqv-+}{^~ONo9zwT
z$*RVanr9_zanFjf(FfM0zZGiK!nHe%O<p4oQ60Y=Pu+}nlW`nH64U8ZmM21OPhPsX
zJrJ+xrglwnK7z&XyDjW8Z)kdf9dbJMCK<lb{<ocbFb4+K?t~;9h8SA4a6B}ENENwU
z1ve==06m}AG^N%5_1|KMJ>#kg5p_}l6|c0SQ?=;tqR8^!47{AU%wa~?%yIs#RC_G*
z>poGl()Kyevhs2MX&ABbs_63LMF)!(@P@ZAD~5eFtenRqEjoc$3N)6Xj12m$kS<jC
zmSsE>nT~;>Uldb&Lj}sf5w%$Osl%7{k*O$Vks69BGgFr2Zg1o_x(IT4gq0ewR(|6c
z_5{2#fpj57KcY!{R#pDVWjD?YeqkRY+I(vUBi*|w-1itX+7cXjiLOhY?ngB2fVsm<
zur>k6R`-e=!@XR01bmHh(*$?tjLe^VtuWOQ+5=;}gq=rFrW8)u0kMT{?A8aMwCE^!
zXH$>ig@@v2US*}RcIEL;4dwP7c<_^f+mI(zhoDsp_O0DCt2K_=bhT|WUfU5|J@#a-
z$;Ys>v_b6v!LjK=bh2U6N@&`bVfxKB>#r@Grl}We>Y_!*4QYGpx$iA8A?CjYg$JM4
zqUl6FZ8-NMI>I|0mO{uk8V}>8uh}$tp87&DZsLbbuYz6mpQ%zSwjHZ4=Q?<}W~?6F
z%PmKtvWu<itk7GEkL_!Iiy7xdmKGE55kq5NMFv^@e)Z2l@sk}Zr$#m&f8~48^;)FV
z^7kdl4?t<WpjDop>t}q#P?N^d{oTg6WxMF4=~6?pOnk=m#DZATA8HNJ0khDftsm`X
zdOY`KFN*|a-3ucDpvn<Ir9NRd)YSAE(Pl~{O$eaO-7_By+_rN$kfq9jQ#49yl>^fr
zdd3`4CCSpH%e<Uar<-aoZ(7IFC&_ID1P+1q`E3<Ep%Tf91Pg3(hLdTcqyOu_%6IJh
zR^pA7^i;rv4h(Dm&*D%lCRZ~P$SKeQG6=?Fw{K69SNf%gjE(z3?pgM86nm9ZLwafM
zU~Eq_oHuaikzqSKd0VWs;*C7HsE2qLnYw1Vdhe?1btU2Mj7JyPX}O#~E~#-JwXkED
z5%D_Y($tKkhK)>AYi>P@tMR=Y9l0oWKY9SciqSNKfv54gaMA^R&dN$;!g+sx$=eTp
zeX#Cswr01YVnd|hYRNKxdu(YldeK%kN9kIJsJxb}_G(3K4&i9hWnFdvhWoV>YfOXM
z4wY=&M?NL$Aq@dkAhC5hp~K<9nNgdjTMRp$;Zkcs2&PE?CSuOeq^Qo4VAY_!I9N*g
zWp{`X2EgflEn|c(t2+H6KD2>&4|+qtIQ!O@cfw~B0rw;liqm$9CR7LOila3PrjtEt
z(i)Sioko#PmyfZ=!f{06>}+)`KTGg+%SLrT*I$oPn(hsA;=RTEcloY~^}&HxB3BhG
z&)bTwQdi&1<}gmzy>}U!O)bA$lXju7{<GSpH%cFqA37FuM*xxf-zxa)4W)hiXD{fs
z-P{NaQ+aC9Xr&*_t3QLSr3-<Nnv66tZR|FDadQezP0yr3g47m`p^<Rl7`~^(Xx?JW
z%&T2U5`RJfMI+{0M!Uc`Func~)Jv$!%tD>!>*A~VQJM&i(cw+l)auk;_?7e+rkg5C
zwK|ixphb+&`e<oh!504KJWjaiUEjc|Y$B8Q4)i2<Q@p}6yr}80aJ1M_BS4+i0YK7t
zak0$bdk;I%H5|VeF!x931kc-kbMvWD-Lj?iPAU!|=?86Bl-fY(tQ+VFgH&`Qny7yF
zvdO@M=w}ghI4wBhF-d5?h8)DS&fs6#Y}M^ihbv7+1=ZI`8;l_PSqXDwC5lJ?2L-`J
zJ4aruyh^3sFY$fqp}Kpe^D;PEti0g>iR%v!XllxBK0>3VIREn+p~o!EEhxRIY|#0p
zMK_y*jA?u&F8!Vy3k6|ri$Try*smc_MY>%}Xu60b)HJvuU^l7aAxtfNOme8h1dO2~
zifZ2x1hUz^cK&{6#g}o1!~G7Y#R!*+gXc)C_8K4k1Pw}510({l&oXO<DkJ7QO`xJ8
zza;%F`a9ZjLaky{BQbALTi3s0^Lj2D%{2peVVEA*d?!X<4^RH<sS1@p@u3R|2BAAr
zXHhiI2ZfOGUlg%h<7@7})MMO9g><!LZJX?~tKZc3)l`<!Q}y-tZzgdZIQP_-T{QPu
zRX)V9(J?cGwm%yR1Ug4oax!;*w|5I(TEsp`3QP`<iupu0vk|f_5|>dUdrXdTVVZI!
zO&8wmUzQddIu*q3h-Q}ptTYha-}-UpsnPM{qUxR4V-Gk+RP0BmVe+s<p+EMsN#u&B
z-mzclW{>^+bI&pHJVSLD9E6&XP)D%p#wHmud`sP#vaJuJqN9FuoH;lEWb~Jxv3@FS
z(4XMdw!hQ!F`FB&^r=Lvhkqa4F)i-=J)$UF=D(XfTmBk8W+#4y*a`FX;f$}&Ji{=5
zqcv<1T%$54hrn?b#QKs*=SYyUZ!<sLngpaMb{2I+X`&Gv+++65g}bLYE*kRkaLV#r
zVZGwsC(J{-P??6xjRe5ePbABuK2Q5<YBa=~nLq|_$Rp7J_Sg(nJ+vA5C1PvLA4lsM
zNE<?I>BC|FON!m7s#nL!g`LkCF4}Z?O|J@8ALE*{{)SJR8HLr<FV=}VebMZBPfZJ=
zS324{njdLt8pjxOInB=w)W??khj!aUtv(Ty_owKbFqWDcl0{SMUe5(>J%Gnd32^>A
zhZ$ca)&R}1lKdEj?|%b!Zs(xr8)P*(hUaDx5B*$CCnT8}WZP<Al2=<jd*M>V-fMRx
zGTz(ufkO7;*(b~onsde=YZo-Z4cQ%$T2-7B?KmKJEl5i{#7`!sRcvkc_EdJy$}8uW
zfTvCZy3G0v2~0&KgLhN$A*<uaqrs#){zA=Ldck|&--0o!TLRb4A{KgM#k$$;A-YJA
zImRPl3`tDO)lGMzCf3lxM?FwS1E0bV0~Kzqk{&qsw=HLrXm*pD=>RwUo{7*kk%_$3
zlZ{4w806n@DgS1P(BVD*oq=CD5{2SYa*fPEe^9nbL4ii;|GxjuD5HS$8!oXq-N@(w
z|2HB`oM%18D>+5((BE%bpQ4WRs*PErKBi+F(CJhwzy3**^n+PGgH{fba*{K-_+D>G
z)}OrL`yA{XCJM%Ce4(v!3AOVU3BUJ`D}t#sHh#&1qG&I$qf<Mtj1yeZ<T?3{QMi_G
z6D)qleb3yfOVpN75)7HN`szhEhmjt)gtte8G>EDp1=I&<(hJeI&()6Qrz^<ARl?5p
z*ba&*!MZuDFi((`T)q`uGFK|o3sCXmXkSBmMa>vA2umx~1y;A=<(4LOK^6(w=hG3G
z6TGSQ8OBD6()<I^Seyb!usnsCMdB9`6<nifMmA{sPh0#Ho0FxdCB?Zg`IO<hb}?Qo
zRQHtBaB|4xP)78HJTj6fjK}u^m|L&}B`kGWYH?gKgfHXE%i(ljmQQ@ZdEEv7ds1)J
z|Hzi3u4E(*Gw(&zIk3!FCLB?0ML=nS_ROuB1m=eav|gcBZlGOuFL9X!&YGr^MeE8D
zBZsq0i>?qm=-HdEjn!M$zkm7phAoplW;hUZj-t=L86Wti=+!Ru>E+6_XI$Gx`{1X$
zNL4WDi3^Ik-G`hTr0&aa{7{e`%8`@|-HJ}bl%08Yyk)dt_WumdivdtcV5T#p`5a*j
zVJt;Jl&NchVDuW!rP+ieb#dN&@xtEfNPZd4If_Dqto*4FQ-islog)v#Fx3+@aC%=P
zU@B!WzT4~xqmI9PR+_JV1|TandyyXJW02R51MF5S1ku;d=?Fq*DuZUP9oWBcKvJDd
zwZ!Pe0oR=~o;xILP}#{!SZ+1y&l-5c=jY5Q`jE%2gnlh$0sHQ?4ZL@K{=8jZw6Jnp
zLCo0o5TIc_b+T^v*$VG*3-Fga@jXD*R}A337eEYLRxsX|oUTq6zc@+OqOZoTw0~fe
zH@k8Bb@eD_!=lG_UC<1OE&e5tifI>C3$)nW0Kg)*BTutOQvH9`mA0Aiy@ZW%nnLwP
z{Rw$g<aR@UP^r1d&3uhsZxX3FVcDtoaBAHbQmxR&8*U<DujO%?BBZ*%J{>*&i=3_m
z<aZWbJ5_E)x2YMkmN~sYb<Pj028egx@46w}ZJ)8i8q`y@8fsrl3HF=ib~Gv_R_<>)
z#nTM!!q}!8`3UONBNY%f7U-p$`N8vs_og}rj*&qt(Te4}Uae6)f%b&i!-o#OdDi*J
zfnAWnN|Glw{g6QkgVV0AsXDVD`zK;cK<@VRNayI@AxqFVhkx1X&}o5hCqO+}54J&<
ze2z&Y)wG1le02eJ+zvH?+~C#>-%4L-E3-?#Y~`3=zJQt^M^~HTO;qD=w!WD2imiSe
zZ2t6d*zQdZfxuISUX>zg2l?|x;SMc#Cc0d-2-fxO7L8P|@$OGBl^wAcIFDxuI4UF2
zX#F^Eh`Q>fnW6Bgc~{ZfuR(3+RflhyEVbb+NQa>nDwsNyd$gIz)wbz`g11<$(x1C*
z+UnJwW=?6@*3r?ZnnPYjcQ5_=6}~F2$Z_~;q8z$%##FX(%u8z6vZ@aP2FBF%xL)tH
z>_)0MHAC3tLD9^hQI4_7*vFhPqsaqb0>J#Q#3H??3i3a$aD@V7uuQ$4P7F;qoG>08
zNeAFfV`6lNRhk*-V#YnT1R(olo;vDi1;wg_9sVz&pR+~ANEZv@61YYrddt0j<<Uek
zFm`xuP5Ja2M<iuNh?;dz>P(J8)wURpr;$`rQ*qxEn1Cpnafh|ZISv;9u_jwCv>KZw
zdmxb1(8s3#rE{m0HE12H=}kxr_1y;5ar&GIBU95>_<(%`aOV{mENax)(VCHA^hHfp
z6%(w+X9R)8AJVCk6iYc*%hr^1Y5SzzUQ15d+Kt-#;8|{p^WlUHO{C-ep6IECr){#!
zwhg1YMoAgce+5#*B$LNqOyny|^6%&%D`eWe*`C-pC9gvj27c?`tY%s0%;);)RP(04
zlNRQXTcYqtQ(&W&m|_d<wnL|NiNWn!nt&Q+c(iTpDrF_xVBfzntV?HmjL^V40mtxc
zb)CjD-7eNqm4v@XNH|HM&{5GO5%f9rxyiZ7&te;D+)d8XY{~aGI`6fu{yfHanmTB&
z1QwVPB-2jpY{*47JP#f?Ea2(Vx5^*aI<Zn^&gZc0#`s7J2W)jpFmlEW)j1;Yd9Y^h
z3esa<dY={lP6x2@kUW>RXMREZb^^&M1h)udZB6mje=&>e1U7T#Sm@E5!2pqD@#Qbw
znY;9S`}4oj?MLa5$}N^l3$`}3#4*;)Myej9vl!nC=j7cr+UXX$N<i7_gj`(Ar8~-W
zKKgaC+2Z`?H(;0=;}2MMRe3EL;tcJy*(Z^&*Azfi1QXGZZA~O-J7#=S{K4`qvy70o
ze?TXA7=NYPzIU%HrxkN@YQ1mQIbv}31!RQ=r8mAi>Pg4=kd-UTj&y5SRgs(Apb*+6
zIF8?1co+o~=80UB8&(LZGo4J$8iX@~av;jc^;0Lzez`Gi&&uig9T=)VjwFu4!B3`N
zX2Rb$p;Z01ki)5i5Pr_ykzR+xoGwH3`OPW=U-gN~(X**DWa5u_A2?(RPV&uCc&l0t
z?QS8h=GysR6NPH{lk=Zt*Gir})ibI82Cxej3j3^*eGE=67#RIW19djsN;K(6g5(rT
z<5B@5G{-2o?f$m!=#>CRkRr(lY=DvGF%x%iUk1Wls-L1Ew>xGeHWYv|*&C_&@=~gs
zZx)0VW^E|-k^W;MSCs>KfB*|Vv4!w`v$I=9S<lU?_1xYgjIKx@YZyOR-A}stn}5kE
zZk+{_e`i;N!f6JhmbzAx1y5>*)LRLnt+yzb>Zk3i3bJTvb?MI;Q>FrSX56AI9D|j~
zrn*6OU4#3qU6aJ%9tCq7f^;XGjsI3rLCu$fNC>QIB6S<=F%j`%cq&QoQ*Uri<33$E
z5MbRec5kBj(7(^Ux#ROd`?U|^#}pN<-&W{5dHDJR2Ufay3vJE;zIr^H8ct}jIdJk5
z%GRaG?>3!5r3;OWGIm3JNhvZ_8uKb*mf@0Qn8IU(^a2_;X1q9Aso>lp?Wx(AOzp$}
E2S0P)JOBUy

literal 0
HcmV?d00001

diff --git a/0.16/_static/images/icon-close.svg b/0.16/_static/images/icon-close.svg
new file mode 100644
index 00000000000..348964e79f7
--- /dev/null
+++ b/0.16/_static/images/icon-close.svg
@@ -0,0 +1,21 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="20px" height="21px" viewBox="0 0 20 21" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <title>Page 1</title>
+    <desc>Created with Sketch.</desc>
+    <defs></defs>
+    <g id="desktop" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Behaviors" transform="translate(-421.000000, -1035.000000)" fill="#BBBBBA">
+            <g id="mobile" transform="translate(-3.000000, -4.000000)">
+                <g id="expanded-menu" transform="translate(92.000000, 949.000000)">
+                    <g id="Group-2" transform="translate(26.000000, 85.000000)">
+                        <g id="Page-1" transform="translate(316.000000, 15.503381) rotate(-21.000000) translate(-316.000000, -15.503381) translate(304.000000, 3.503381)">
+                            <polygon id="Fill-1" points="7.8940033 -8.03165789e-06 5.75107666 0.955277357 10.3761069 11.3312155 8.03165788e-06 15.9562457 0.955213104 18.0992527 11.3313119 13.4743028 15.9562618 23.850241 18.0993491 22.8950359 7.89416393 -8.03165789e-06"></polygon>
+                            <polygon id="Fill-2" points="17.7071311 8.06344719 18.6623362 10.2066148 23.8503053 7.89405952 22.8951002 5.75097225"></polygon>
+                        </g>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
\ No newline at end of file
diff --git a/0.16/_static/images/icon-menu-dots-dark.svg b/0.16/_static/images/icon-menu-dots-dark.svg
new file mode 100644
index 00000000000..fa2ad044b3f
--- /dev/null
+++ b/0.16/_static/images/icon-menu-dots-dark.svg
@@ -0,0 +1,42 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg width="25px" height="6px" viewBox="0 0 25 6" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <title>Page 1</title>
+    <desc>Created with Sketch.</desc>
+    <defs>
+        <polygon id="path-1" points="0 0.4815 4.81659091 0.4815 4.81659091 5.249925 0 5.249925"></polygon>
+        <polygon id="path-3" points="0.683560606 0.4815 5.50022727 0.4815 5.50022727 5.249925 0.683560606 5.249925"></polygon>
+        <polygon id="path-5" points="0.61030303 0.4815 5.42621212 0.4815 5.42621212 5.249925 0.61030303 5.249925"></polygon>
+    </defs>
+    <g id="desktop" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Behaviors" transform="translate(-856.000000, -249.000000)">
+            <g id="mobile" transform="translate(-3.000000, -4.000000)">
+                <g id="condensed-menu-scroll" transform="translate(530.000000, 158.000000)">
+                    <g id="Page-1" transform="translate(329.000000, 95.000000)">
+                        <g id="Group-3">
+                            <mask id="mask-2" fill="white">
+                                <use xlink:href="#path-1"></use>
+                            </mask>
+                            <g id="Clip-2"></g>
+                            <path d="M-7.57575758e-05,2.865675 C-7.57575758e-05,1.548675 1.07871212,0.481425 2.40825758,0.481425 C3.73856061,0.481425 4.81659091,1.548675 4.81659091,2.865675 C4.81659091,4.182675 3.73856061,5.249925 2.40825758,5.249925 C1.07871212,5.249925 -7.57575758e-05,4.182675 -7.57575758e-05,2.865675" id="Fill-1" fill-opacity="0.5" fill="#8C8C8C" mask="url(#mask-2)"></path>
+                        </g>
+                        <g id="Group-6" transform="translate(9.000000, 0.000000)">
+                            <mask id="mask-4" fill="white">
+                                <use xlink:href="#path-3"></use>
+                            </mask>
+                            <g id="Clip-5"></g>
+                            <path d="M0.683560606,2.865675 C0.683560606,1.548675 1.76234848,0.481425 3.09265152,0.481425 C4.42295455,0.481425 5.50022727,1.548675 5.50022727,2.865675 C5.50022727,4.182675 4.42295455,5.249925 3.09265152,5.249925 C1.76234848,5.249925 0.683560606,4.182675 0.683560606,2.865675" id="Fill-4" fill-opacity="0.5" fill="#8C8C8C" mask="url(#mask-4)"></path>
+                        </g>
+                        <g id="Group-9" transform="translate(19.000000, 0.000000)">
+                            <mask id="mask-6" fill="white">
+                                <use xlink:href="#path-5"></use>
+                            </mask>
+                            <g id="Clip-8"></g>
+                            <path d="M0.61030303,2.865675 C0.61030303,1.548675 1.68833333,0.481425 3.01863636,0.481425 C4.34893939,0.481425 5.42621212,1.548675 5.42621212,2.865675 C5.42621212,4.182675 4.34893939,5.249925 3.01863636,5.249925 C1.68833333,5.249925 0.61030303,4.182675 0.61030303,2.865675" id="Fill-7" fill-opacity="0.5" fill="#8C8C8C" mask="url(#mask-6)"></path>
+                        </g>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
\ No newline at end of file
diff --git a/0.16/_static/images/logo-dark.svg b/0.16/_static/images/logo-dark.svg
new file mode 100644
index 00000000000..9b4c1a56ac6
--- /dev/null
+++ b/0.16/_static/images/logo-dark.svg
@@ -0,0 +1,30 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0.6 539.9 487.3 120.2" enable-background="new 0.6 539.9 487.3 120.2" xml:space="preserve">
+<g>
+	<path fill="#EE4C2C" d="M63.1,567.3l-6.6,6.6c10.8,10.8,10.8,28.2,0,38.8c-10.8,10.8-28.2,10.8-38.8,0c-10.8-10.8-10.8-28.2,0-38.8
+		l0,0l17.1-17.1l2.4-2.4l0,0v-12.9l-25.8,25.8c-14.4,14.4-14.4,37.6,0,52s37.6,14.4,51.7,0C77.5,604.8,77.5,581.7,63.1,567.3z"/>
+	<circle fill="#EE4C2C" cx="50.2" cy="560.9" r="4.8"/>
+</g>
+<g>
+	<g>
+		<path d="M129.8,600.3h-11.1v28.5h-8.4v-81.1c0,0,19.2,0,20.4,0c21.3,0,31.5,10.5,31.5,25.2C162.5,591,149.9,600.3,129.8,600.3z
+			 M130.7,555.8c-0.9,0-11.7,0-11.7,0v37.3l11.4-0.3c15.3-0.3,23.7-6.3,23.7-18.9C154.1,562.1,145.7,555.8,130.7,555.8z"/>
+		<path d="M199.8,628.5l-4.8,12.9c-5.4,14.4-11.1,18.6-19.2,18.6c-4.5,0-7.8-1.2-11.4-2.7l2.4-7.5c2.7,1.5,5.7,2.7,9,2.7
+			c4.5,0,7.8-2.4,12.3-13.8l3.9-10.5l-23.1-58.6h8.7l18.6,49l18.3-49h8.4L199.8,628.5z"/>
+		<path d="M250.3,555.8v73.3h-8.4v-73.3h-28.5V548h65.2v7.8C278.5,555.8,250.3,555.8,250.3,555.8z"/>
+		<path d="M302.3,630.6c-16.5,0-28.5-12.3-28.5-31.2c0-18.9,12.6-31.5,29.4-31.5s28.5,12.3,28.5,31.2
+			C331.4,618,318.8,630.6,302.3,630.6z M302.6,575.4c-12.6,0-20.7,9.9-20.7,24c0,14.4,8.4,24.3,21,24.3s20.7-9.9,20.7-24
+			C323.6,585,315.2,575.4,302.6,575.4z"/>
+		<path d="M351.8,629.1h-8.1v-59.5l8.1-1.8v12.6c3.9-7.5,9.6-12.6,17.4-12.6c3.9,0,7.5,1.2,10.5,2.7l-2.1,7.5
+			c-2.7-1.5-5.7-2.7-9-2.7c-6.3,0-12,4.8-16.8,15.3V629.1L351.8,629.1z"/>
+		<path d="M411.3,630.6c-18,0-29.1-12.9-29.1-31.2c0-18.6,12.3-31.5,29.1-31.5c7.2,0,13.5,1.8,18.6,5.1l-2.1,7.2
+			c-4.5-3-10.2-4.8-16.5-4.8c-12.9,0-20.7,9.6-20.7,23.7c0,14.4,8.4,24,21,24c6,0,12-1.8,16.5-4.8l1.8,7.5
+			C424.5,628.8,418.2,630.6,411.3,630.6z"/>
+		<path d="M479.5,629.1v-38.5c0-10.5-4.2-15-12.6-15c-6.9,0-13.5,3.6-18.3,8.4v45.1h-8.1v-87.4l8.1-1.8c0,0,0,37.3,0,37.6
+			c6.3-6.3,14.1-9.3,20.7-9.3c11.4,0,18.6,7.5,18.6,20.4v40.6H479.5z"/>
+	</g>
+</g>
+</svg>
diff --git a/0.16/_static/images/logo-facebook-dark.svg b/0.16/_static/images/logo-facebook-dark.svg
new file mode 100644
index 00000000000..cff17915c4f
--- /dev/null
+++ b/0.16/_static/images/logo-facebook-dark.svg
@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="88.4 13.1 1298 2499.4" enable-background="new 88.4 13.1 1298 2499.4" xml:space="preserve">
+<path fill="#2D2D2D" d="M930.9,2512.5V1372.4h382.7l57.3-444.3h-440V644.4c0-128.6,35.7-216.3,220.2-216.3l235.3-0.1V30.6
+	c-40.7-5.4-180.4-17.5-342.8-17.5c-339.2,0-571.5,207.1-571.5,587.3v327.7H88.4v444.3h383.7v1140.1L930.9,2512.5L930.9,2512.5z"/>
+</svg>
diff --git a/0.16/_static/images/logo-icon.svg b/0.16/_static/images/logo-icon.svg
new file mode 100644
index 00000000000..575f6823e47
--- /dev/null
+++ b/0.16/_static/images/logo-icon.svg
@@ -0,0 +1,12 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0.6 1067.9 90.3 109.1" enable-background="new 0.6 1067.9 90.3 109.1" xml:space="preserve">
+<g>
+	<path fill="#EE4C2C" d="M77.6,1099.6l-8.1,8.1c13.3,13.3,13.3,34.7,0,47.8c-13.3,13.3-34.7,13.3-47.8,0
+		c-13.3-13.3-13.3-34.7,0-47.8l0,0l21.1-21.1l3-3l0,0v-15.9l-31.8,31.8c-17.7,17.7-17.7,46.3,0,64c17.7,17.7,46.3,17.7,63.7,0
+		C95.3,1145.8,95.3,1117.4,77.6,1099.6z"/>
+	<circle fill="#EE4C2C" cx="61.7" cy="1091.8" r="5.9"/>
+</g>
+</svg>
diff --git a/0.16/_static/images/logo-twitter-dark.svg b/0.16/_static/images/logo-twitter-dark.svg
new file mode 100644
index 00000000000..1572570f88c
--- /dev/null
+++ b/0.16/_static/images/logo-twitter-dark.svg
@@ -0,0 +1,16 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1"
+	 id="svg3626" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:cc="http://creativecommons.org/ns#" xmlns:svg="http://www.w3.org/2000/svg"
+	 xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px" viewBox="0 0 300 244.2"
+	 enable-background="new 0 0 300 244.2" xml:space="preserve">
+<g id="layer1" transform="translate(-539.17946,-568.85777)">
+	<path id="path3611" fill="#2D2D2D" d="M633.9,812c112.5,0,174-93.2,174-174c0-2.6-0.1-5.3-0.2-7.9c11.9-8.6,22.3-19.4,30.5-31.7
+		c-11,4.9-22.7,8.1-35.1,9.6c12.6-7.6,22.3-19.5,26.9-33.8c-11.8,7-24.9,12.1-38.8,14.8c-11.2-11.9-27-19.3-44.6-19.3
+		c-33.8,0-61.1,27.4-61.1,61.1c0,4.8,0.5,9.5,1.6,13.9c-50.8-2.6-95.9-26.9-126-63.9c-5.3,9-8.3,19.5-8.3,30.7
+		c0,21.2,10.8,39.9,27.2,50.9c-10-0.3-19.5-3.1-27.7-7.6c0,0.3,0,0.5,0,0.8c0,29.6,21.1,54.3,49.1,59.9c-5.1,1.4-10.5,2.2-16.1,2.2
+		c-3.9,0-7.8-0.4-11.5-1.1c7.8,24.3,30.4,42,57.1,42.5c-20.9,16.4-47.3,26.2-75.9,26.2c-4.9,0-9.8-0.3-14.6-0.8
+		C567.2,801.9,599.4,812,633.9,812"/>
+</g>
+</svg>
diff --git a/0.16/_static/images/logo-youtube-dark.svg b/0.16/_static/images/logo-youtube-dark.svg
new file mode 100644
index 00000000000..e3cfedd79d1
--- /dev/null
+++ b/0.16/_static/images/logo-youtube-dark.svg
@@ -0,0 +1,21 @@
+<svg width="32px" height="32px" viewBox="0 0 32 32" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <defs>
+        <polygon id="path-1" points="0.266666667 0.363636364 19.2666667 0.363636364 19.2666667 13.3636364 0.266666667 13.3636364"></polygon>
+    </defs>
+    <g id="Page-1" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Footer" transform="translate(-1143.000000, -439.000000)">
+            <g id="footer" transform="translate(0.000000, 200.000000)">
+                <g id="Twitter" transform="translate(1143.000000, 239.000000)">
+                    <circle id="Oval" fill="#CCCDD1" cx="16" cy="16" r="16"></circle>
+                    <g id="Group-3" transform="translate(6.000000, 9.000000)">
+                        <mask id="mask-2" fill="white">
+                            <use xlink:href="#path-1"></use>
+                        </mask>
+                        <g id="Clip-2"></g>
+                        <path d="M7.8729924,9.64708726 L7.8729924,4.08018547 L12.8184285,6.86363636 L7.8729924,9.64708726 Z M17.2006266,0.761297078 C15.7138423,0.363636364 9.76670489,0.363636364 9.76670489,0.363636364 C9.76670489,0.363636364 3.81950814,0.363636364 2.3327238,0.746091721 C1.5344866,0.960069399 0.876999507,1.60258279 0.658035103,2.39784619 C0.266666667,3.85077029 0.266666667,6.86363636 0.266666667,6.86363636 C0.266666667,6.86363636 0.266666667,9.89193994 0.658035103,11.3294265 C0.87723706,12.1246899 1.51892689,12.751998 2.3327238,12.9660337 C3.83506786,13.3636364 9.76670489,13.3636364 9.76670489,13.3636364 C9.76670489,13.3636364 15.7138423,13.3636364 17.2006266,12.981181 C18.0144829,12.7672033 18.6561133,12.1401274 18.8753747,11.344864 C19.2665867,9.89193994 19.2665867,6.87884172 19.2665867,6.87884172 C19.2665867,6.87884172 19.2822434,3.85077029 18.8753747,2.39784619 C18.6561133,1.60258279 18.0144829,0.975274756 17.2006266,0.761297078 L17.2006266,0.761297078 Z" id="Fill-1" fill="#000000" mask="url(#mask-2)"></path>
+                    </g>
+                </g>
+            </g>
+        </g>
+    </g>
+</svg>
diff --git a/0.16/_static/images/logo.svg b/0.16/_static/images/logo.svg
new file mode 100644
index 00000000000..f8d44b98425
--- /dev/null
+++ b/0.16/_static/images/logo.svg
@@ -0,0 +1,31 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 17.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0.6 539.9 487.3 120.2" enable-background="new 0.6 539.9 487.3 120.2" xml:space="preserve">
+<g>
+	<path fill="#EE4C2C" d="M63.1,567.3l-6.6,6.6c10.8,10.8,10.8,28.2,0,38.8c-10.8,10.8-28.2,10.8-38.8,0c-10.8-10.8-10.8-28.2,0-38.8
+		l0,0l17.1-17.1l2.4-2.4l0,0v-12.9l-25.8,25.8c-14.4,14.4-14.4,37.6,0,52s37.6,14.4,51.7,0C77.5,604.8,77.5,581.7,63.1,567.3z"/>
+	<circle fill="#EE4C2C" cx="50.2" cy="560.9" r="4.8"/>
+</g>
+<g>
+	<g>
+		<path fill="#FFFFFF" d="M129.8,600.3h-11.1v28.5h-8.4v-81.1c0,0,19.2,0,20.4,0c21.3,0,31.5,10.5,31.5,25.2
+			C162.5,591,149.9,600.3,129.8,600.3z M130.7,555.8c-0.9,0-11.7,0-11.7,0v37.3l11.4-0.3c15.3-0.3,23.7-6.3,23.7-18.9
+			C154.1,562.1,145.7,555.8,130.7,555.8z"/>
+		<path fill="#FFFFFF" d="M199.8,628.5l-4.8,12.9c-5.4,14.4-11.1,18.6-19.2,18.6c-4.5,0-7.8-1.2-11.4-2.7l2.4-7.5
+			c2.7,1.5,5.7,2.7,9,2.7c4.5,0,7.8-2.4,12.3-13.8l3.9-10.5l-23.1-58.6h8.7l18.6,49l18.3-49h8.4L199.8,628.5z"/>
+		<path fill="#FFFFFF" d="M250.3,555.8v73.3h-8.4v-73.3h-28.5V548h65.2v7.8C278.5,555.8,250.3,555.8,250.3,555.8z"/>
+		<path fill="#FFFFFF" d="M302.3,630.6c-16.5,0-28.5-12.3-28.5-31.2c0-18.9,12.6-31.5,29.4-31.5s28.5,12.3,28.5,31.2
+			C331.4,618,318.8,630.6,302.3,630.6z M302.6,575.4c-12.6,0-20.7,9.9-20.7,24c0,14.4,8.4,24.3,21,24.3s20.7-9.9,20.7-24
+			C323.6,585,315.2,575.4,302.6,575.4z"/>
+		<path fill="#FFFFFF" d="M351.8,629.1h-8.1v-59.5l8.1-1.8v12.6c3.9-7.5,9.6-12.6,17.4-12.6c3.9,0,7.5,1.2,10.5,2.7l-2.1,7.5
+			c-2.7-1.5-5.7-2.7-9-2.7c-6.3,0-12,4.8-16.8,15.3V629.1L351.8,629.1z"/>
+		<path fill="#FFFFFF" d="M411.3,630.6c-18,0-29.1-12.9-29.1-31.2c0-18.6,12.3-31.5,29.1-31.5c7.2,0,13.5,1.8,18.6,5.1l-2.1,7.2
+			c-4.5-3-10.2-4.8-16.5-4.8c-12.9,0-20.7,9.6-20.7,23.7c0,14.4,8.4,24,21,24c6,0,12-1.8,16.5-4.8l1.8,7.5
+			C424.5,628.8,418.2,630.6,411.3,630.6z"/>
+		<path fill="#FFFFFF" d="M479.5,629.1v-38.5c0-10.5-4.2-15-12.6-15c-6.9,0-13.5,3.6-18.3,8.4v45.1h-8.1v-87.4l8.1-1.8
+			c0,0,0,37.3,0,37.6c6.3-6.3,14.1-9.3,20.7-9.3c11.4,0,18.6,7.5,18.6,20.4v40.6H479.5z"/>
+	</g>
+</g>
+</svg>
diff --git a/0.16/_static/images/pytorch-colab.svg b/0.16/_static/images/pytorch-colab.svg
new file mode 100644
index 00000000000..2ab15e2f307
--- /dev/null
+++ b/0.16/_static/images/pytorch-colab.svg
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 23.0.1, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 181.13 105" style="enable-background:new 0 0 181.13 105;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#FFA800;}
+	.st1{fill:#FFDD3D;}
+</style>
+<g>
+	<path class="st0" d="M26.84,52.6c0-1.78,0.18-3.51,0.52-5.18c0.34-1.67,0.85-3.29,1.5-4.83c0.65-1.54,1.45-3,2.37-4.37
+		c0.92-1.37,1.98-2.64,3.14-3.81L16.92,16.96C7.8,26.08,2.16,38.68,2.16,52.6S7.8,79.12,16.92,88.24l17.45-17.45
+		C29.72,66.13,26.84,59.7,26.84,52.6z"/>
+	<path class="st1" d="M52.56,26.87c6.84,0,13.03,2.68,17.64,7.03l14.23-20.35C75.74,6.45,64.65,2.19,52.56,2.19
+		c-13.92,0-26.52,5.64-35.64,14.76l17.45,17.45C39.03,29.75,45.46,26.87,52.56,26.87z"/>
+	<path class="st1" d="M70.2,71.29c-4.61,4.35-10.8,7.03-17.64,7.03c-7.1,0-13.53-2.88-18.19-7.53L16.92,88.24
+		C26.04,97.36,38.64,103,52.56,103c12.04,0,23.09-4.23,31.76-11.27L70.2,71.29z"/>
+	<path class="st0" d="M164.44,16.96l-17.45,17.45c4.66,4.66,7.53,11.09,7.53,18.19c0,14.21-11.52,25.72-25.72,25.72
+		c-7.1,0-13.53-2.88-18.19-7.53L93.16,88.24c9.12,9.12,21.72,14.76,35.64,14.76c27.84,0,50.4-22.57,50.4-50.4
+		C179.2,38.68,173.56,26.08,164.44,16.96z"/>
+	<path class="st1" d="M103.07,52.6c0-14.21,11.52-25.72,25.72-25.72c7.1,0,13.53,2.88,18.19,7.53l17.45-17.45
+		c-9.12-9.12-21.72-14.76-35.64-14.76c-27.84,0-50.4,22.57-50.4,50.4c0,13.92,5.64,26.52,14.76,35.64l17.45-17.45
+		C105.95,66.13,103.07,59.7,103.07,52.6z"/>
+</g>
+</svg>
diff --git a/0.16/_static/images/pytorch-download.svg b/0.16/_static/images/pytorch-download.svg
new file mode 100644
index 00000000000..cc37d638e92
--- /dev/null
+++ b/0.16/_static/images/pytorch-download.svg
@@ -0,0 +1,10 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 23.0.1, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 105 105" style="enable-background:new 0 0 105 105;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill-rule:evenodd;clip-rule:evenodd;fill:#333333;}
+</style>
+<polygon class="st0" points="91.74,42.58 59.53,74.8 59.52,74.84 59.52,74.8 59.52,3 44.8,3 44.8,74.8 12.58,42.58 2.16,53
+	52.16,103 102.15,53 "/>
+</svg>
diff --git a/0.16/_static/images/pytorch-github.svg b/0.16/_static/images/pytorch-github.svg
new file mode 100644
index 00000000000..2c2570da1de
--- /dev/null
+++ b/0.16/_static/images/pytorch-github.svg
@@ -0,0 +1,15 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 23.0.1, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 105 105" style="enable-background:new 0 0 105 105;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill-rule:evenodd;clip-rule:evenodd;fill:#333333;}
+</style>
+<path class="st0" d="M52.54,2.44c-27.61,0-50,22.39-50,50c0,22.09,14.33,40.83,34.2,47.44c2.5,0.46,3.41-1.09,3.41-2.41
+	c0-1.19-0.04-4.33-0.07-8.5c-13.91,3.02-16.84-6.7-16.84-6.7c-2.27-5.78-5.55-7.31-5.55-7.31c-4.54-3.1,0.34-3.04,0.34-3.04
+	c5.02,0.35,7.66,5.15,7.66,5.15c4.46,7.64,11.7,5.43,14.55,4.15c0.45-3.23,1.75-5.43,3.17-6.68c-11.1-1.26-22.78-5.55-22.78-24.71
+	c0-5.46,1.95-9.92,5.15-13.42c-0.51-1.26-2.23-6.35,0.49-13.23c0,0,4.2-1.34,13.75,5.13c3.99-1.11,8.27-1.66,12.52-1.68
+	c4.25,0.02,8.52,0.57,12.52,1.68c9.55-6.47,13.74-5.13,13.74-5.13c2.73,6.88,1.01,11.97,0.5,13.23c3.2,3.5,5.14,7.96,5.14,13.42
+	c0,19.21-11.69,23.43-22.83,24.67c1.8,1.54,3.39,4.6,3.39,9.26c0,6.68-0.06,12.08-0.06,13.72c0,1.34,0.9,2.89,3.44,2.4
+	c19.85-6.62,34.16-25.35,34.16-47.44C102.54,24.83,80.15,2.44,52.54,2.44z"/>
+</svg>
diff --git a/0.16/_static/images/pytorch-x.svg b/0.16/_static/images/pytorch-x.svg
new file mode 100644
index 00000000000..74856ea9fda
--- /dev/null
+++ b/0.16/_static/images/pytorch-x.svg
@@ -0,0 +1,10 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 23.0.1, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 200 200" style="enable-background:new 0 0 200 200;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#EE4C2A;}
+</style>
+<rect x="-27.55" y="90.14" transform="matrix(0.7071 0.7071 -0.7071 0.7071 100 -41.4214)" class="st0" width="255.1" height="19.72"/>
+<rect x="-27.55" y="90.14" transform="matrix(-0.7071 0.7071 -0.7071 -0.7071 241.4214 100)" class="st0" width="255.1" height="19.72"/>
+</svg>
diff --git a/0.16/_static/images/search-icon.svg b/0.16/_static/images/search-icon.svg
new file mode 100644
index 00000000000..ebb0df86773
--- /dev/null
+++ b/0.16/_static/images/search-icon.svg
@@ -0,0 +1,19 @@
+
+<svg width="16px" height="16px" viewBox="0 0 16 16" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+    <!-- Generator: Sketch 51.2 (57519) - http://www.bohemiancoding.com/sketch -->
+    <desc>Created with Sketch.</desc>
+    <defs>
+        <polygon id="path-1" points="0 0 16 0 16 16 0 16"></polygon>
+    </defs>
+    <g id="Page-1" stroke="none" stroke-width="1" fill="none" fill-rule="evenodd">
+        <g id="Artboard" transform="translate(-63.000000, -93.000000)">
+            <g id="ic" transform="translate(63.000000, 93.000000)">
+                <mask id="mask-2" fill="white">
+                    <use xlink:href="#path-1"></use>
+                </mask>
+                <g id="Clip-2"></g>
+                <path d="M2.95913767,9.58220606 C1.13316435,7.75622206 1.13316435,4.78513897 2.95913767,2.95915497 C3.87212433,2.04616298 5.07138409,1.5896202 6.27064385,1.5896202 C7.46990361,1.5896202 8.66925692,2.04616298 9.58224358,2.95915497 C11.4082169,4.78513897 11.4082169,7.75622206 9.58224358,9.58220606 C8.69769713,10.4667577 7.52163859,10.9538926 6.27064385,10.9538926 C5.01974267,10.9538926 3.84368413,10.4667577 2.95913767,9.58220606 Z M16,14.8357224 L11.2495505,10.0852451 C13.1351173,7.62730486 12.9554949,4.08320445 10.7067543,1.83454417 C8.26070966,-0.611514724 4.28057804,-0.611514724 1.83453345,1.83454417 C-0.611511149,4.28069663 -0.611511149,8.26075796 1.83453345,10.7068169 C3.01947955,11.8917699 4.59491713,12.544308 6.27064385,12.544308 C7.66870437,12.544308 8.99594505,12.0887007 10.085841,11.2501776 L14.8356356,16 L16,14.8357224 Z" id="Fill-1" fill="#EE4C2C" mask="url(#mask-2)"></path>
+            </g>
+        </g>
+    </g>
+</svg>
diff --git a/0.16/_static/images/view-page-source-icon.svg b/0.16/_static/images/view-page-source-icon.svg
new file mode 100644
index 00000000000..6f5bbe0748f
--- /dev/null
+++ b/0.16/_static/images/view-page-source-icon.svg
@@ -0,0 +1,13 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 22.1.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg width="20px" height="20px" version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 300 225.9" style="enable-background:new 0 0 300 225.9;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#EE4C2C;}
+</style>
+<g>
+	<path class="st0" d="M0,0v225.9h300V0H0z M282.4,208.3H17.6V17.6h264.7V208.3z"/>
+	<polygon class="st0" points="64.4,166.1 76.8,178.5 142.4,112.9 76.8,47.3 64.4,59.8 64.4,59.8 117.5,112.9 	"/>
+	<rect x="158.8" y="156.4" class="st0" width="88.2" height="17.6"/>
+</g>
+</svg>
diff --git a/0.16/_static/img/pytorch-logo-dark.png b/0.16/_static/img/pytorch-logo-dark.png
new file mode 100644
index 0000000000000000000000000000000000000000..0288a564e22755147c1daf1cfec1db59fcc998c9
GIT binary patch
literal 27763
zcmeIbcU+T8_b>X;n-o!yBE5?A-g__7n}84?6e$5h6@efEA~rw)DN>{(MMRJeBA_Bw
z=^YfLBgG&s=Lx#^cC+`(dG9^HbMHUo^N~C=Yu1{X?<zC1CW%pcm(+*}X$S!TAl6V<
zHUI!@@Db?8#|8iDimYXVZv-Cd=5PSuWkLVL@Xwa>1pq=_XC);)Jtvqq4DJN;VAN1j
zV)XEWIXJsQ0l<Ga)zHz%aGp{IIlif=bqn!O3uZup&uE|+jW`>_&c#GPrg@7gV~9e(
z>9mRpHtpN=TUht*A?{G<bDz3RIEeR|Df3>~(_6RS?@k3ib?L7~&UPHU9FU!Rwvqax
z3$GFXRJ^*7z6jz}z9RkU$*{IJty7<Q<wFTsJpeNN7xql>eHIL0J6Kx!B6A~NJ%Hi=
zkq94X)JbmSKOc!AT~K^%iGjF@(dZo`u7!^{3&>x)r<eoCt70Hh6WI-cbZo$;-OhFf
z(BlDYI6`JW0*KV@$AK7t#X}|vjPw|Q@eK5qGT<ZyJnJ&Ktpb?w17{ty-bet0JOICj
zse?LDS_w3Ek`a{w_@@AV{afJ|0PH}(<}C}0KXCIga8`BQRQlV+VzN0d5UIyy(p9Vy
zDxntmJf3)_ro8l=of_vZo)v{)LXzZpUk5y<77P_8o!s~U0O>Il;A}Vc{X0pDJ3B@1
zRghToEHvPvm~Cxm4!(31ddLI7ly`8?0XN?Z8iW)s!tEfFZ3Wxu3Q@WbGTNbxRQ@H9
zK0j=_;ep0RCF4QW(9oC9p9eMHC|W??n+6>?zP4^O-FN*FEWL-^nyjB@4HdKsRm0w%
zeEnflJNtZBEWu6N(b@ZIdzD0}J-Uy~Z!~PH%(%|Z=#zWiQ%<_GA(+h^r5Jmmg{>E5
zKRv;{izPcD0+e2r#R!by$+NP<`mV)6X6D^EjRb)88d%dOPJAqcQ`l6OA8J)@UoD*t
zKsac`dIG=|WmaCJ_b=qy@Bu(MJ(R0dfp)Wrin|f_d=t)S6VaZH=uJh|H?I}R6bbDS
z=e=w$JiDRD8PQQn$7RF!NuEx)QQs*d&Vz*awec$w84uDO2>z2M&e~7{Y=yV@XCO?&
zw=isPo3Y%(KleC%o2f}1n>?D8@!hSn`pjA}BI*JLw~ZK$)#=t{yf7sqEmf;yq}~F8
zH})cG)dU`fmg&oUI`vYiDo>2&rfMcpx4pMO(tXPG<ZiEW8kswy8FSr*PNc*4WtH0a
z8^2QF3B7UU=WT6Pd@jyS5NFr^ocY<Q!d9A6lV?~pQCoIh&;78;Lsc3HM6vx886orv
z>So$zmj>1I7|*i`pB^R<Bfz?;|C)n2E>EkRz54WzH&nKQ0+D9w+${a%_s{T=1>RD6
zBgRS|r6kBv^#D`jd6rQxd9P}(y8+8QcYd<4It@$QyJ35mGCYZg+>96BhJQ}|-27SS
zGwo-lbqm5&TY1l{ZNq6!gZEc8slSkY5&nV;O%zlxewt)3mX~34TIBv^#iEP~gA)D7
zyfeZ+NfI4llp0#uMVXI2(m0)SLOPdx^{zfW_K5ycLbqve`@GuMgs)heey4~dD8kxq
zIdW1bQ{hoLQu|SrBx>hSe~hMXGvXH$SWKLIO!^pRY{pf=b)6#SJ>etGN9>Pix%iCo
z3vvp^3y6$2jN6Uw=U*`lDpoZXGMvdL_!M7gT#%}_Xh>`5m@kzVS>TulGm^^+Hs6oJ
zHE-5#)>X>VPk(MbQ`^c&WJmqNpPR{rxG=&eROAi&^NykGzFakl`*bzYG6ne=5=ugv
zLZ&YaHXb_2;mss^Zl`!(BADaMP1p#&J$`oe!t@2&9AWJ$0jQ-*L3~2IS)qENexcf8
zooMlLZfZ`QN{#t$D2YOwCWj@5n&bUWoYXnPA;Ym;!dyxn8Og0;-U63w+Kd`yzXtc*
zmBc6YR}fFMVY%Z0b#puCgZTE1_T6@Jx71FV;l4P<iaT<ulS;Zp#1ke>$JaLUO<&2X
zSzaj{@(7YJbu*@_NBU`>>RnaRJI<|^GpsXit!}MP*>TyK*tK*PbXJlKbgFVL<Va~t
zX}5KJ=*aBoe=L$LBQPL1mpqp|kUUXobJ_ngcST;sE2u`<=gT?fD;2fw;g==Mv(1bz
zTfI;%N-bjRuPLg@J)K*1Deh9?lYYnM(r*>-D)K9-OTP@_jucn8RRlpCZO=JYwiGw7
z#MQ+GRR`VOm%$^B93m;CyG$1hFZVhtX(BnE)S0p(7u2#DKhNiA)<xDeC^S=Xrv1Vj
z=TyVwk7|zR9IN1A8@HmPv~INXCdl$8)3%Pit^2I$z2#{&b**<RHTgyI>*c(Wk`>Vv
z=atCOv=72*7oM04!MmH^DoyHt>5(H!d&KWD@hn(ZNjKE-<s1v29bb%T(LiZswqF4f
zvU*PTyR28*jO>iG*tp=hx%CIJ#>%&q?YhmdC)VK;N+B#Ld(<^l50Hys!*Ilk3RMWT
zz}(%wcAcr=o}A<e^n2ORvcjG8Dg_FK`!|Yid}%6Ta(cV;poj@Mm)IH8=}Z!?7yPxq
z8xJxIiEa))#StDEZW<wX&*PREO^-l=h+ogB5`02=M>$@3JlRn|SE7VRQ-n*lfcL8O
zuvoQlwOlK&gZYO(vs5z;SFQ6Byj;bi?ymI{d&?V<)2Y)$GlAz4`QBQ21Y~r+-VD#B
z%;zQNmW-FzzI93b;hF*&S03dZw)Hc9Z1)HQUempHWajG;?YT9mWXVN;X<eO5=UVJi
zOsz-&Yl~1e_m`@h9QhAwwTZRIk|m!|CiybD3Cc*;353PQJn&R3VA6@Ja9nVlb?|L}
z1`RJeXmO9O<fDIt_s-?TgJ(GAJcJdBIbV$~a}IFfl{rF99U3~K-b=(6KDpSj{Ncs>
zkq2Tat7)iZ^4H{3u@%`7TBba7X0vtAu1b_idj^aU&QRc7j4;7{CjIJm1(#c&ZsD?N
zLUC{%`!yYx&?`;vZ#2Eca9JcWCj1f}v^YAjY^7_)XFl-~vJ;U?D@ze|yZG$Lxde)k
zLC#VCFSL=c_YIFnAC~7AzNxbBa_DuKpB?9UVnGsLkT9D-;!Zp?TuwY;Qog+PU1fwX
zp@VazCaglK>S=@Yt6FI8yUq?NIioD2O`{zn(Usn@mym&N_nsBnz}oM^OMN32wa9(h
zcl9I27{)q%4t>yFjolYWx#FmT`R({!H>;vcZl23+SiM;Lcq}9p167~;=z3+04DR<|
zTvQ`zW2iq!UG?K379}>mo6<VX;-MVyc^*kHfutvv9QM5v_-qRABZDuUBOTvW#iaEl
z)nvn@T?<=_YTXaAJ^c&g*Y?y#7)VourOg90H;mpn56|ED80Wa&btm;vYEUrmk5QYE
z%!<G{yUvNLq_EX$GViHg6;$+`RIOB}<@zh{^|FyKXG~w2CY4B9yDfMxNFT&~)NQLt
zsYxoXvUI6hxmNcw&;dD5`$VjBY_l%l8zrj0?}>!WSlz}o7_tSqH9<KR?8UL)*<Lf`
z$FWjWt6OWmDZi{Rq%;=UCueamI=DI-a<$PfeWPbo4T&2b)^!lGb)QrT=l<jS?GeNg
zi%-O5B_yWhy!Pk!a+SL5yV#$cJIFj}FFr@f7rfoKIMCUX*prd4o)FjIc0GH4p>dJA
zWTf?VD}DVte<6V@n<l%<O@5<(bgNG%NOeD7XAI#)IsJIPW!Rm&8%G@%hr-56+}N`d
z;u!$|DGp~NGjB5;ZAm+r8!yBjW((!@ck=)@q5vQ*=kEcrbAfs@+Cm+j-DQ|pYZ{mt
zo$Y0qO+|G0bv%@yPR{Dryr72HE*aTfbFq`KXO@#Cl=hbd1-L=IA&mZRuI_M2e;MXO
zxsu>Bx|)xf@vw=viwv^@x<f`Y9X&=Rm=}~$gjbZuPC!6_QA~nYKtx<fP=uRNkY7NA
zk6(gMfS-q7KvIBTQjnkV*Na(}5PXvMvUiX)P*(lb9axfKcJlW2kmTd@^Yi2N6XJz=
zIr0feNJ#MU3-SpH@_;RP-~sO55Pu$bILl8Vf5=gW!tK1AJ-nS^?u=-;5L=j!w+u5g
zx}!f|zvkuU@n=Ww@L$-0DDwG3Jop57`T72h$=>cy9uFTc*TbRN+wnnNp>9xjZ#c*&
z@UMIxe{BB$WdB>2f0F-W$DmGhbpD<BkMX*>{cG27Zxvq<f?tsSt=m61;YI--P(A}F
z9OmO?2UYO}r^51Yv-fs(_)9r|lZ2-H$FW2Go&N(Hn(~wF&=h`ZNE)20q>>jD;tlgM
zg27y6e_GjJ=O^&^gtjV1P92Dyvpf1wbwQf%DAQjW|4SZJ8R89<1$o6lTjb#vHWCz)
z6cmx<7vkXu|MLGP<Y&txGIU_}&JF>8lOZ5#BmlBXNQ!{w{g;g2S{{`FdXPQD8}eUC
z`MvFLa_sFS9bjH=5N}y$H;5yY&%@nOn(wd5-&+1@SCUFFSC|(#EU2uIG~d5f|0@@G
zQq}u;P?d!ns3|gPs3?gFNQjE^2=WU2F6Os(zsWhYSxF6dxHrVz4yvIn3s&$tJKIY_
zA$Il-b^;PSLK0$79$^O&2ObE-7RtkK>mXw101*`z6cRjChK|nf?Eet0472k=`yD9y
zXNa<g*@5E!nx3G5xVWgDh#-%EovkR3fPff;$JW6?80<~VPE0^pTtZX;@<&I%3HwJ!
zx?av;1ckW%)&@Nnd$2PHsDmhk9}1e2gSZ5bu&B5l4@8(B$|E5pC?O_dFDfi9^wY4w
zx%`Wme{`kp3<vE#;5UB&z2Y!N8bUq)Ui$~Z)%h^ic|g42P_!q@F#qAoe+I`tV;`F5
zFd|4o?9fqC)(-7xP<v^<zt#RVtUoz_u{r-!o&OyKKjr_%{-68sbAq}b=cYeJ9k%+<
zLf|k5Z$F3^RKXFnkpC`*{FM85#-klb^P$%jSLgp5s(`usLcRW<C<9bE1YDD$UT|3l
zFPIx6#KXhY*$#qMET6Bt{jWvsPk&(ahB5vYGJa0?HyQsQ&J_;v{Xa9h|1O67zcspl
ziR=G=8r@%BYv%-UcZAx@^8Mkke^mT;y8CIpe~%=81i*hr9%=Nlj*eQO2OS!fEEx6w
zQTJERKiiw3qmQf4VRRAVKYW3^Kk)U7_jmeHUi1GA@9*@ZynnG8yE}W!f-A`3P>)pq
z#)aN}{9G6f!HqH$J-MS*M_cPT`$JvLl%2s<2ab-O0{nu)zqpQ){*~4IFIGW;zj6LX
z`NjEDSHJCO4@1t+T5!J%Zq@nz+^hd18vPlz|0f^*n$rKI)FZM^IynN!@x>Fkj>qQ+
z^#raXfE-^uf$MmDj!;kFIs(Y?#S^%W$L9$31g;~19A7+v>v(*QP*31G0?6^j6S$7Y
z=Lq!#t|NdPUp#^9czljfPvANN$nnJ!xQ@r?2=xT6BY+%VJb~+Ye2!30;5q`x@x>Fk
zj>qQ+^#raXfE-^uf$MmDj!;kFIs(Y?#S^%W$L9$31g;~19A7+v>v(*QP*31G0?6^j
z6S$7Y=Lq!#t|NdPUp#^9czljfPvANN$nnJ!xQ@r?2=xT6BY+%VJb~+Ye2!30;5q`x
z@x>Fkj>qQ+^#raXfE-^uf$MmDj!;kFIs(Y?#S^%W$L9$31g;~19A7*ZF2cXQ{R(vl
zzw+t_ez%pFrhyy$UM!=Xx`7S=Tw?<OL>K`4K!NY`0N}$901Gw%AejgN=U}ncEvf*(
z?xLZrVB|l%G}hrWqTb*7O>OmCF#qXO6bc`o4H4aL^oG3>9wJb`BX5MAdB&+%%l)!9
z_6?+=uoUJC7AkG*>*qLM+FLL@RG*|opjOYPJ^2<1P&O)#MOzrFjVzm>zoZMK4;|cR
zKYT(Q{+bLpK8(X6p#Jj>!C4#dRqXz~drs5<K(mCjb`;W3()c)WKy}yoGyo*I)r=W{
z9P-g&AsUjX{edBgB1}MjgHEQLn2`b$AuPHrm*BnC$WXVxe7`4&5)(*&oCnsYTQ!BK
z^|r8tNIwk+04Gs$u&%sK+hF}|n$KhKH*Vx#2;hLoP;T!Jm|OR535D<^c<+rcJWZj*
z0@BSBF~Jsrmdk5?<Qj9rU_TwoU=1Q-7aV||fn!@k@b}%47dK6?fWv9!rA43H8&fs{
z2imRv3_aYiWpN}*t;&oD<WhwYgPkX>E>WP^V}!FnWjIlyha>!HwV_iow%7DvL+COY
z0HEvA>30J6Mmj-)LDH`(#$bj!D4Acty+8XNRTFrSD9b|t{u&R+uy*r$_u}WUD@ml+
ztezUGldcJo=b#4w4i}K7-y>xUit}cfcai-XB%+SJ=artOgr=aonhc$kdnCL&Raq13
z&UFd%{DK*Wr7>`*7erbI{f~X+ULV=7Y{H-x&^nX98{UiRJQI@5GJiWBl)(y$$g7N|
zJLu%sKEr9z`2{Sa>%_gw_xDCnyS!ZD*>a?t@Olp)jGszhx$fUIp>r`0(~SP>+_S23
z0C)hl2V!3@9Pkl;m3tU)sbOE>vexj@4?9uxlUMjJMncDuZLwx>o)52InUO*5vn5de
zoPxZ;U_Wg8eTXdg#^~#j-5(cG0pCF#N|;jpq}!kv50r4`#DZ1G8RIseXc2*GKkrsK
zk`}V@^;SFvV1xG~9cgUfcc0;#sQI&gXxbbIe5dh}>;vg7Pu8G4(4uZ$L^HWoTA(3Y
zI=$z|7Bsfe_93q2s!^W=>R|C1-@Nh!2Ty-&ed#2_PXrp92@sFJr?sVVMEA)#gTAQe
zi-`~c+5PJdFoIQMNv{PZfzlq{MSDca1JY|sGBRlgY~wwdps}FqJSFYBpY*1H*5HaZ
zrbePKVON;v8}JUEKUzZn7}Idnp|n&8f!qdG2WZUi53N)D#@0gz(5Rud=-+VS0|nD*
zOla<WYkOz=GUE{j(3o*(@(EHGzd0hEuYlt^G{ZM%veLS1e|6ouu26C@-*B2~r(E!-
z-Vt>#%1A+3tl&rvE$)u-=@cQfMWeOXsOc}1X5^74TLjjiUD{^ON~LZ7W)Uvf{IHyE
zJ@womiS}pnblV^4Vn!lgP?h673%|PTE0ko^b@l=P4<|5ee4m?;mdu2L%^#9pi^@XL
zoZk^Pe<oNAiWa}Q?5hRN59s<zODC2+56*}w1}siGuU_`G1YJ*08|-xQbOb2=;4!y}
zgehpIOVJm=CelR{>ocIU;xeJ}oROtsK;6^=>(L|B%LFa!Z7@^4EV?~<Qt3aM{6H;P
z@q(Qlntp(+2mP<n5>tW>7?X18Jq0KUZSS0Fa+&b?h?{7|p&gQhyP|AYHUTW6MNJ}@
zFQOV+d=bCc@%rz7OIcUoo9YECf@(8Bj_pg&3RjS99Xz^>44V<LlCh%iY(?fwFo5(c
z_bNt~yBY$#mD>T}GRH$yNc84r{5#vN@lrR3*a&2>3(q_Vj@Qb00GO^EbAJquzfopC
zv8&@MjlAqvt9`OPtrkgw2~el)Zdic4%;AVSBj0wFz{vKHY^ieVZBbdYvEVolr!C5C
zwU1I;<#LZx?(G93VvMj|jN&hl&h1O4*r^*CeX(rG4C$KJteyfxEdo7hpS2A3xuKG2
zMy4K*?>9nviHUH)agc+QH~H+KDmu9&JHv0UA0$vk6k&p(p+U<0RMJ!B3*x}D0YLN{
z25{F+DRd1%0vgf5p~$3E*pM+`3Wh>uazl{kPz#n<O%6f|%Ugivijesya_k*=b?_D5
zpW`Eody(`6siGl=T+o|qGXkYduSb9jPq}47&Ya8#yt;P#=fme0@Mgotm22E~E2EWL
zj1Y$I0zk=^l*PSnKDSyV?PXm5)Thh7yBjcix*kCQ(Ml9O7F6_2EfMr%&h-laG;R~E
z6;S|rV@J8=Q{l+u!8r<z=8=hz{3THd#(6%C#Rr^bABP)LK`R9T=w85Xe3Z436mn-K
zs1N|B8sq)ct6y=DHM&p0kG6ZkMvi#DU1J+&Fm=J6iD^y6V79KSipvR(IoZKt{mHF;
zgq;Xwwa6R&%H1w<DrHftT2ylHQ=pu?d0;Mt%zdiJIA?SlK7%p7TVa1=mw5YBB<G6;
zg8(8X5LH+#w@F+)o`M^_h1=^A%igr9kdgqYk?n6EY63$WvW52~N#`m^?Dlm>!}#ip
z#fC#-%vzZ$>?WV$T?7vHS3_V`Eez!(4YGwQ)Ni;NmPp07#vOSU%o^rBHelATg@+U|
z5!i@u+=r+Mu}hSN^A|@TdFz$S?Z!r2pWXH^ee?RF|M<4p3#}W^hbLzC$u^P3%d`hU
zuB<$4LH1S{<bWF$7HUooC1HEIF=0N!y4j;Co{75D)nCx&EbcqX)E|$sDtYDscGV`7
zYA>5LtX!`_A?*t@!)BTN^K1HmvHn`g+TJU|t}~PM!h6*oonyFmwU3KT*VZ!}6i`8Q
z6pt|GI;*%@G2<=Gg{en?6*>PDjh)$)WQn<IoxKN4ELSNQeU)rRH}Ra*zPQbArJUZF
z>NZ%V*)%8D5X&fl3`~X?=&yyiysbbC3Q+AXr0mY7NRe)|ogPSJ^Xw#6>YKiM`<$K5
zt4sM=iyo`QRlSTDh&kLV&1;R-1Vi09Y#2G!7PJ{x9hv0a%O3Z5=DH=>VvG_6OO>T1
zTs9$R44Qr6x1q4ADyq>&v?5!}NbQk5&DMTXSzb64rQBP|?>;ZQJ)Rtc54y<`*CX$l
zlp1%NZb<`y`yXxfq+Aw=&(QPctDIuhMD<+Rj|eUvVQpLr76F~z2UI&1CaR>&Vj!Fg
zsGWUM|4IkD`}}MuDm>|11&S>SQKECb3XdG%bo3a?i!@FoT<iI|=4@t}<4I~v0Tbue
z+qo~g+r8E$_Ha{#VYA{#dS%}jZ4DLfb0dt&z9__+NxaRS#s*ERvAR*MS3=<aqKm{^
zakg~i0c)dj%O!kD2jV6FBKkZn$|?m5k@CHUOL>@$vXV?xHJKMQi4S(U3c`oaZ=>=`
zroUWC%K>L6ol8q^(j0M}zR)=3jL6pxeg7~|_tU6bk6tw~l2m}q^_|T`@gzE)-On{R
zT(^9SW;?%4_;FvOqeNYDB6$Eb!s%K+NSN+cEkRu93#(@4h`IJ(TniM)*<|2BhO+Yy
zbm7V1b`6@EwqSm^pyT;q5@lK8a`V$Wy_WVmULDX3{VSTC6#MzLaR$q|J!C&C`7$Sy
zfBq0CH|KJzB6ClUCd!p7z57Gpd_><ck|vI|vucsQUehTi)u*{`Aol_q0Co(;NPXm`
z{(?KTv}K=sg1UWKQ_bJVW^OXrq*Ib|33#Xr#R4TXao)zYTe-07smd)65%45TR$$GQ
ztk5>t*e!g2CRa7GA@5sl!(&&$r345E9wPmWz}L@V^cb}2n1eW|X0km$r@M7lB9J6L
zbs+t^GnQnu){r)ix>0jJkfF^jBJMp~m$v>$DOvx3fsI^{lZD_x@2RePY7Sgi)F(Y6
zSTVqX*7#D!RX@3IrrlZatkya=z!dU6vM(&OHv~aGK$ZO?eIT3K!)AcH^ZT|L>jxx5
zFU=vBFYiK15s4v2JUGIFp38(Lf%67Cq-Xs{&gTjx0eJCwdD!Ysw^zY^xfnCGCOH_o
z$-)<t<nmOxC?6MIRudW2m1OnrBqpgNkrD-pRDXng8YCx!e(i@`w;|j6NE(X`8i6vi
z{@poSYF!-h5{3M<zH2xS53rctW)WltA9x1o^=~t>ZF6jkX=9#p#sxJVhxI8*7su?q
zJC}U<ySUQUuofj`Z(zf$3yyl`UfN<<#nmk%;0B6wr9rj>T!8YRj5REU`M?+z(uUZA
z;_?UB*jx8#RYtCdr~yTq;xisjbVJ%r(ax3{D91GHNg8l9qYu&RtFq+O#(5<yxu=5v
z%Fc@|IUJ=rbs0#nO<BBIA?0F6A)gr5iF|*BnKBz2wBmDb`+t0H4&HdIj~7UxC3Q2F
zFB)%Ec<BR>E*~H<K!9yC$3t`wS%5SNS>%uZU{q^XjdHoR+O?2bxMTVwlWDvgTYo`(
zE@FH);I3S=datQ5>U+&<%3|aeJ~u;clxys1+96{bpeJLm!K510lm=xXu}&y?io{%w
zTi!jQFS9;`qCy(hiro#t^ZhLuJ1;KID;v;wrcJ;ax9M2{67px)1;tn1xuw$uw|rGJ
zaFfDc-EWofb3Ti0EiorJ6S)`OQzue9Kk4vvoBUiQRutC<w*3~C{1l9HI(iqCG|nVt
z`0l@xTfwn%&RUMWx$82#HYpe5P4~ljmW%T40k`w&x6A5OuCKdlb?{Fa8&R}~b~eyG
zb^lt>J&cXmqW_ZdRJ+-xfk9ub$Fnorhxg0(aotongHaHB#na4BFk5e3AE@EgSLMam
z2REkZ`qJf=#`XH-zn}Ris;OiiP*=->p@%wPmB#zHw!f(QN!0P`0cMp@%dQu=@3EsQ
z`3$A|Jf3Jvo@4jq!jm^f<i$M~tSFw92NMIUK`V9E*H+ucckMJ*DY{ZJ*-u#(ea~p%
zxAg4GT;dgTbz#&5jldvb44@IHxE#2eJHA^Pdrzr5Rn6mSswAv(FQt^$K<>3+J8SB-
z)ro7XbH*lI5dsI!phZwm<O-}<;(*Iou+!Kor)X>LN0@wIG1PR}>c?hScZLP_`DZ(%
z!lk5NsxvTV5O<Nm-19|>Q5DZsO>n}?kb{&lS0WoxpQ^c90@K33)%E%I$yM8!f*Y_~
ztD%p=Bl||zgk$xpjgS>7HzVcyxH?K{WUsqe?2z(zw_%O<VBgijuD#C)XweS8F&QC|
zV>Nxw-ZL_oQyGr4eGi&;NAqk?r!lx9)m&mmj^DSh5g*b}N`kw3k8%k;qa>XeEp-cZ
z(m$|vaYmo)l<e9};1tI3CK`x)Q)inDZaNe1t6y^DnRgo`mz@2!zl-0hyT7)M&xk<=
zQ4@$6%M@@{8s-<e+N@0R;T`EC8c?8?Nh}UMy;%>orJS-9yBRG9MsI|OGq=ToPO}9{
zM`xuuzb7cM(7($60~HA?FulUsT6W>wf~eEI?rhh7<F5%-Uv*!}QbH}wKT!KNQb9j&
z`6`|%XC_#Cvg@)c@V08&Y%An!>wBLX(ienr01^vD;35`^EiYr%VX-Ls8IsoSp1orp
z6?HZ`vIJ^@sZaD}<^qqti)Di^D~8~Z5uT{6B}J=9KN0E+7v)a!z_zSqLJBtH8Xf^6
z39poa8ozHHGC>f!lfh7L^)zx-?R_gkZHxAMKtESS$<8;^IG;h1QW0L`{q{$flnO3(
zL2K3~+}5mj5Aw2D8&zA{X)r^SzYy!J*&1p%Quw_>U7c(Wc3SHOw<J7D93h;P*{H=*
zUK~VZ-^)EY($M_`vdxmXmW3e<JK3+<T(Kz}T;#1f1Sz{^79n_jI9ICRD^Ck_bNq~F
zlrO8>q)?}|&WE-c&1wL1qxo4Sc=&sKi<T7NvbFs~;pHg|S7S+H*XEckm#}8*{I_fL
z(^G-G$qT16QK4;NQ_5Qfz+j%4^{io;S>08}RBn?>g%8r@3LNE4czrdWe%!>Cclkg}
z4rp|YxH|1otLhLQgs5ZQ*{lwyAlTC<(#G&l?UK&=zG9P5W^N?l(@9K9w#QL+VZXt3
z8ta^RdNnrcaTfBOtxpQK$;<#Q>OLzr`pBX~Sy&l6KF*GNHLrbnlmwz_(Yn9I91la$
z(tVET`X1X=sUJfo!M+WmKReCjLuC@pb%oAmWkS1;!$A~``(3x|_T=KclekTC!2>Jz
zt_}DX++N0H&Csx8n&G8e)pujp5Ll2}1c{Z!-%)A37Cgs=8S(&diY~p?77;J5E_K<J
zI?m~S)h<M<slF+|aEDZyX?-1a_qNzUE>iP*;o|1)0>p^2yvft{BrnHG)jlLNb-;AT
zrmy3@`(kTiYo5!YI1X|ka^7Z6HLM4M|Jj;mkK@`{ZE28Ba_M_sy)v^fRDM>aaqYJW
zfzuu|xD^4!#q)fl{YCm)`-Ay*EE02)5kX^lIv0l`=L?W+S03LEC!HYQ4#NgY=dfCw
z7<ZE74@PaeOZA__F%h{fCI@MS{)=+sP!ZJaGZ=;xi&)QC8gi`M7yGd-TrNX6Oqx)F
z*l*m<KOM(C$gRDDK>$qhlY*y-%9)hprJaOSxYZE%o|x^b+2mt&srq%)OXh$@lrp)^
z8A`$wwd>A?EAEq0CSOo9qM{YZV5Lb?azI{!Zpld05QA4yVvM)=R41%c32#utwF%{1
zX6Clg?E!H^I6s?901v<JtFuy#)3t6+Pw%-F*zz{F(|{{HBCPB^ubdPGqbpsaiOh<z
zme5M$#Ww~AA+&pQK38k)pVM3%4_DOy%CQE11O$cI&E6JQz(-)qKQc&~EPcHv7o86{
z<F7Fg9h@1e-Ot@*7=(J$=>7=d2}1>HAgZRdC4FDDTQKEs`}>UguDzK|gtsyS1c-hj
zc+*9?i3&~v{+NzTBU9an3|FX6X%ANqlC05Sqaw91wJ9c&0FKe=qDG&4FG_aRUtqBU
zdi$8%i}5c@RH(0J;*GCI7GXwGq-hoVYk7uJnJ1v;xJ-7tlDiWjItcDe!v&A%N=3~x
z(kmhs=NKAet5D$P<m`^HLcpn9%u@{hjh&}DXJWmi(oI5}^LU94>KX%E41-iL!Ayah
zX<rG(gPty##WQ{!@|cK7BLUOeg{HR&l%iWku`P(5h4Gh9s4N$`mu?;An+8m`487|Y
z3vI_2R$nTLH1FOjhQA*hcXFhxH0~JATev_jVR#z!b564*tOjQbmPoNlO6Z!^+$gK-
zOT(>~6+0i~Q_bU1FnhxyBV%$#L+Qd0j?smVvrj6O!aQ569d!DNw*9eD@=A3X1Xng6
zB)!c!>p9l?{rk9MXH|aI#%<CdhqFedUA${vLCtpelLS{$`N0>k9s4o-6A)R?xpRj5
zu3A#WvKoir$-!M-Z`U_Czf&vP`=N<xgG}as)G>mpjWAzOB9+}O5d9YW)`pik8`Vf8
z)t1RU%EjODns(jcz9wa%G{a`As={y(YVscJ^WcJ|D12n```B`~s#C^8c1}D)z&B9d
zht0}@Sa4+YJA|w86%K>X2QgxQf%bJ&zczZ8Q?-<v=~FdED^@|X5IW|^wayXR9oNsT
zdJ1&-4Xgsbc~xK7<NES_nwu(4@x%wW5_n41fdxU%dJ))U0jle};i$W~#~RGs*q~BC
zrH_K2+7H~arx%gtMco!dMNK(#SCPlwu5=5)k{cKfl$%<|<ckCIO-tj1mF_Oj@C$(I
z(XG2wrfi*6E;Ex2_h6NU^(fS5cN}$~d>KQ;yE7=RBL&64U_e`ZYrYtO+VzE`U%{BF
zCWfOJAdk1zUwv7WtBBpysUY&ZJkjR!JO_+<;BmJG&yYj<VW<6zK|<2}xi~;;&N63~
zn@})&lfhr4FgU$qf8FwXl;<aytH{Gmrp)Jku?H(d-ze?QfG)(spkCP!OhD%%*4so`
zFBZFLycI%Qej0u?G$^H;&-9}CGg0u&R&njhY}l3I>4j*?J4MyiBPw&gu6<EU!U;wu
zE5fqiVIW;NAMV}`_4j}9j_{e>#yBhmo=|2Xv8|_SX>G$t8LAS`ZQ_O+4<_IEX6HAP
z@|1%9dkj06ekrGO0p3JcHs9yK$SFV)zBF77YJG%#H38Vxe(^$5d?9eP|J%@FidL03
zm{cpDVS1TC{lsX%3d1XA6yy6YIq@|Utm%Hd1~8^m!tnKVr33vo!r8&5jqI;l&M4l<
zx|e+&TmyC&sc;)gEtqPRKF&N6<XQzYO!CiWYxgH42>A8uDhkZ6ttMDXq{oLcHZ`Zp
zS$L22oU%(oXMo;t&M|G-KBcj;;F3Z#Z$~x>3<&Q|2qpm{uj6aLY&he%P5k;=0s-lk
zRxwXEmsObE!F>l5Z4mWBCa)bfm|yBvguidLPQ^~-Tv5pv1l!z_$F$|a{-nwqyNc$Y
z4rR+m?MA(<l|7e@k~h3T0BoU;nqb$;w2ZMC<GN;Jel%10R%LlChzQvlgqcnq5}q_=
z91N!TKxZDfK)uk6GjiL7OCCJ0d6^@^w45>F(v@U|Rmge`<44B#UC7;;z!S)OA?9tl
z6g`#GhE+1zC_ryQ9)Ii90qxe3NFFnof_(tUM>U~R9K^T71mnu@Od4_S<8|9pDqse2
zU8dNho$p1Ts?M7fXD+2pHb}D!b?RD0rV@ZZhEL84P)H+b>Gw>B=fG8J)BxSC7)y>A
zmaiX+pM$xEiwSBu{d(2#H-qd2*^>4m9N_`lq;x^8=ffe7G!WTv_Y3*f-!k~j2QZwn
zNOZ`Uixv0;B!z=64oeo7sRP_qEjCMKY-t<Tu@S5Qji?Es!qmf@yixU*t5e3rCSteY
z7Zt$;goN(A5igjsOroDMm-;?lc|m@^aC#=yBf06aMb1abx!!^^fHnr_fMu^hQ-gV1
zOyRJYYFgeJz(oWUw2lw@N#eF;!BmQ;!{2+h-w4UySx9&|LhQs#4t#lQ^g-k4X9j>|
zl`qWB*ZF&cIjd5?aG;AD>s{koy~pL{#9R%iRCtiPO#0(QOponGRIJJYkZx4fBp~%8
zJwY#m_ucF=<MMd@>b61e9A*77Dhe4qlTx@-%3(`~&UDfTsds=GtO-9L@69m(n%fpa
zoEsPv5#LN3{a#dhY9{|q&bCI*)i_uX#iMheUC66Q3>);G6Z{BtCt(AyH%qg)B~{B7
z8Cq+oJU)sLlw@bs!$k~XQWs!}-SBz!u6oP)a=X?2MDQA+;N1H)c$gB>IjBxZawQ@M
z7Jd)zt|%n!<mT~#R9LE8x}O55ntr=b!nVH}iAvKN*6Sw)dPpKZ7KFut$!zxMzUr3c
z@k(*0_K{k*&J;Mj)mrY|se2OMop;T<X&5m)x-n9a2M@ulvVTTFI|M~?$;c?|<t5p#
z#B6Ll1lJc_i<&#fp5zY~F`UsbjU9(ySoL3p&sjQjucm4xdEgu-cg^0jT2Y;^Oe4KE
z6TgCMV@$FaKC-W}uGBf=+j*DG%5-1G?%+q5EG;Vl4#2PgK+M{JR|N9QtkI8*Z64%>
zyI_E~R$P>TYj){)ObgOYbrJPuT9TEk;8!x*Ck*l~XT33L$PwHGvQ<`BX@%L9F;0@!
zB3*{KZw1VzXuZM&Z5RN!gG43WIX*<faAqn|)Rh5eLmk<jFHKU1XEzC)p;#0wzX#_9
z6L(e1U?v1W_>$a&S%LQ{7MN$+Z%1~G&i0h%VLG`@{<wKxwT%umK*|bac-CY13T@P3
zQoIq|^SRlJ$w})I&9+2n;<YQFVx?qAAQtr(LLJWzFbPTSJ=39}?yOp=3$BOro#xLF
z#6c&qOTWEa8$1V|*xfwV(%ybwUX)}FAiJ-nQPt#`k9q6q#A5l<<q!1EVMH3NnCO84
zSnx#*+pt_r%fdWwCyNJRYB!4}xi!0MI;-HtjS~Y9-el!FMn?9hlE+&+t1fDcaU5tW
zG9Tv4^UTW3zO-r}ifOi0xfwP)p%y!QU^d8QO(Ld4bMqXw`NR(=|D{4b`iW1I4JBi_
z{)gE`0N`@i&+#A3vcdrG2l#fgT4UEydCwWs>++Wh05BJ-qm6m9OE?gk`~aPd1OQEj
zC|;AtOO?^d;mhj_^m`4yYk^J=QEJ=|(=D#gnUSrn(0nc{!hEHJ=}kkqH{sNfulsXq
zh<NNf13p*I@~7<9hfOG<GvKinlGhSc+p-BKJB@sefy$<p0d^zIoAN=vjR&iur`N3K
zMxSX!k)v-V8ZFDrI-o*)KiV)Rbm-!M(>SQuWec69w}ENzatIMP7^}q!^i4=M{f=C$
z(ZCJsen+Z;JbcE}hemq7Tl@wb9MLI)18YCYvWX##8d_?)1dvG^YJuY+N5KEgV#b(k
zFYebAsY5BDGv=%s2y@x5I#<ci*Qa+`w-d7NLmcrwA&|i=$v1VFb1+A*gjs%Lm2-sb
z)`Ev2r6E?iRSrz@72Cu6l`UQt$%N?Zpyu%EzT*0WvE_5vxV9(vwx7$>X*c!-D0i}0
zcQZRo1^Jq_(bAJuj%|Ka;@nm>&{_3H!fym+{=LgteezJV{&!1R<h2#-t$Qm=UtZyE
z3MrT_Lh+Zi`@I`MUB*qTTAIx;#PM)E-&5sn%uV7?F+#gi8m-K|Kan^>i8exUdBzL7
z63>xrT)z0P-r?2VgG^3<Z!Q#UC8=+Vha|Z_-@2Cc-r;m-`Mr(K_Q^Wyi^YVT8j_~0
zBU9iSF;x`8y>C5RXYDn20~?gKt2)wR@mc?7ZhS!=ZldvHeEd+9cCQ!(1%be{j#*t;
z*C_vXtu6f)Xspc6bQTHEY^7?!<5qT)AO0Rk28vd-m;Jfy(?c&DX2*z+jgNUU&d-Xe
zImQECeC;4HQIIzYE;Kpr>ElgvmZG3lV6YReQkdd=Qi?*j%vm;kZUk4Wy#^Kkkn5+2
zoG{tJ+!=T!Pm`Y*Jsh>UXP^P*G_3N~O|Hk3UtgBqE*-StnpbW~KDb0&+@t#K5szKN
z>ge4j&#qK>nj-b}!hlt=&Z?NK#ap^B`0`g%KGxOqjUaczf9$U^ucFHI@-YsT`5?B1
zs|&9V$whhn_KuNpZ3aWZpsdH!%sa>~GDg^wZ9gjw6d`K4U%<wGszDVe?J=C!Fy>al
z@S`ll;v1jqtR?4ixMnBYiW!10uws{W&+Tt%qjEjc_E|QAoQY*!cr{T9=%{262P|Ii
zroB+Syz=BDLBkGlgLJ~f1F>E+QH{z^5@M$ePpf14%aLp>FZSQh(m8Mkll4@k;8k$?
zJa}wky>a7}J$L}mpr)brFD7n)apP~-#Y|hAEiIa5%{|0>Ar@ev$j&aUG<mK^oz#By
zVy80LsIpY5Ip3Ai@lBm<SD@~^+~IBLbL6yJ;!Dd9yD+86s+djik^sC;;WeGAVU>R-
zg;k#2W#>kCcw2at>8mswfsGW|CCZD#4RkpX3Ud@4C@XRsx*geA05|B0`s+g@=GVq_
zb-MfhF2R8Xh4*aP$g}`YU|K*dFetE{+l7UWEX?YdwG@v4W18?5s-EI>I9H85EfJ<{
z3buI8YGF<n3z#whTTN?X^3x$E8NSfc(p9p|r_7=*5XeuM7E5EJw9rX)@RpFEyCQV$
zLQ}{hR%iB-@8m0wPM=qVavT`7NLb7FyzI>YS0s1?0{&MrDcI_i7ja*I!9h{$U<?)E
zFlq=Tp?Z`X)yYRb(ur<ce>{75<G2@vh*{Enl<dGWpNNC<;OqWipI~pN(3o4&VcC+z
zNqgC@M0S;@=_ha{tc^y`V(Ur*z@!WD$BT$v?m_^Jiu0_ElB^gx<*n_z(iE@*F=6lw
z0bV*cu*&DyK3cld?;m5?VjJ@C%*=B}xE+;^@0%NlE~~WBKsm~htok>DfpUaI7Y}Eb
z=S<G1OIc`0VTegJD8m=G8Q*eH8m85}dX@-tB_98+vPdJmpG)(=ndMMD3Q-7WDtyQC
z)-W%P8T;pwV0wnQBb4q&yJfFhv!RVbBVCDM9&S2ncPJbA<j29`e3jyWWIbYj@;(5`
z1KwGr^_~TPLtP@qDoTO~vwrDvQ5`auD+i}s=JW#%NfIRAp=m{MV9s*&<GykYFtq2Y
zKn8YMge*0AE669ITA~o(HJmDsHnyt~_j;RR$YCGtu>gpWjv}+o_-lnV&WF2>qr`(k
z=_!SYYHk@|Uc-apir?6ZF*f%|_guj0{pRYQq4{zW{@~5Nm5~SdW?=d>QRTqH#;(~e
z?K$J*l_<oAK$PresG*3<07AQWmu&mvTqrtV`SYK)sj3d3j_1<cHy>lj#)&Dr$f5hj
z2%%)z4CZB*>g3`#uc#=N-`k&5RbieHf4IX}Rqa$JCJf#<@3LXC>agm7w`!d8VOixl
zVUxPCvmaK~R8(W@sp-HJz1D?QSu;VKd}dE^LL#1eqJ0~V=>y)l&Bq7^R)8>pBKQYF
zFer92JJU|!mQy@e2}ckF9*Y6j*Z?<Z>I;$f@?z`hAs6Ps6kSo(jci-jEqM1U&(6E2
z!rqWnLi67N;rW#LS!ZT$%F{ntv}T36&X`7f`dGLceS4kv$Hye54Zsgg+|r-35}ah+
z8yH?@^w;PnP1o$KQdTi2zc_C_)sTwCf!ohCi6XyID1srUH|So8_LDqurUwQ7XTG=B
zHNl)CiNdD%Vg*gs@W?#9CWDzLMfFQvN=wMT&XuzsDmd{BW>kF$(RVGXJ<XaTHV-MF
zlWjkM@2fROG({K$`~;FWdiw&tgo#Fg!$&k5`h~Vb^y}ps3JaLu<>6S3k#<5?XsBFL
KE>^U;@xK6=e0Wg+

literal 0
HcmV?d00001

diff --git a/0.16/_static/img/pytorch-logo-dark.svg b/0.16/_static/img/pytorch-logo-dark.svg
new file mode 100644
index 00000000000..717a3ce942f
--- /dev/null
+++ b/0.16/_static/img/pytorch-logo-dark.svg
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 21.0.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 199.7 40.2" style="enable-background:new 0 0 199.7 40.2;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#F05732;}
+	.st1{fill:#9E529F;}
+	.st2{fill:#333333;}
+</style>
+<path class="st0" d="M102.7,12.2c-1.3-1-1.8,3.9-4.4,3.9c-3,0-4-13-6.3-13c-0.7,0-0.8-0.4-7.9,21.3c-2.9,9,4.4,15.8,11.8,15.8
+	c4.6,0,12.3-3,12.3-12.6C108.2,20.5,104.7,13.7,102.7,12.2z M95.8,35.3c-3.7,0-6.7-3.1-6.7-7c0-3.9,3-7,6.7-7s6.7,3.1,6.7,7
+	C102.5,32.1,99.5,35.3,95.8,35.3z"/>
+<path class="st1" d="M99.8,0c-0.5,0-1.8,2.5-1.8,3.6c0,1.5,1,2,1.8,2c0.8,0,1.8-0.5,1.8-2C101.5,2.5,100.2,0,99.8,0z"/>
+<path class="st2" d="M0,39.5V14.9h11.5c5.3,0,8.3,3.6,8.3,7.9c0,4.3-3,7.9-8.3,7.9H5.2v8.8H0z M14.4,22.8c0-2.1-1.6-3.3-3.7-3.3H5.2
+	v6.6h5.5C12.8,26.1,14.4,24.8,14.4,22.8z"/>
+<path class="st2" d="M35.2,39.5V29.4l-9.4-14.5h6l6.1,9.8l6.1-9.8h5.9l-9.4,14.5v10.1H35.2z"/>
+<path class="st2" d="M63.3,39.5v-20h-7.2v-4.6h19.6v4.6h-7.2v20H63.3z"/>
+<path class="st2" d="M131.4,39.5l-4.8-8.7h-3.8v8.7h-5.2V14.9H129c5.1,0,8.3,3.4,8.3,7.9c0,4.3-2.8,6.7-5.4,7.3l5.6,9.4H131.4z
+	 M131.9,22.8c0-2-1.6-3.3-3.7-3.3h-5.5v6.6h5.5C130.3,26.1,131.9,24.9,131.9,22.8z"/>
+<path class="st2" d="M145.6,27.2c0-7.6,5.7-12.7,13.1-12.7c5.4,0,8.5,2.9,10.3,6l-4.5,2.2c-1-2-3.2-3.6-5.8-3.6
+	c-4.5,0-7.7,3.4-7.7,8.1c0,4.6,3.2,8.1,7.7,8.1c2.5,0,4.7-1.6,5.8-3.6l4.5,2.2c-1.7,3.1-4.9,6-10.3,6
+	C151.3,39.9,145.6,34.7,145.6,27.2z"/>
+<path class="st2" d="M194.5,39.5V29.1h-11.6v10.4h-5.2V14.9h5.2v9.7h11.6v-9.7h5.3v24.6H194.5z"/>
+</svg>
diff --git a/0.16/_static/img/pytorch-logo-flame.png b/0.16/_static/img/pytorch-logo-flame.png
new file mode 100644
index 0000000000000000000000000000000000000000..370633f2ec2b7cf35a37a283095229de337f46e4
GIT binary patch
literal 1010
zcmV<O0}cF%P)<h;3K|Lk000e1NJLTq001BW001Be1^@s6b9#F800004b3#c}2nYxW
zd<bNS00009a7bBm000ZY000ZY0qkOip#T5?8FWQhbW?9;ba!ELWdL_~cP?peYja~^
zaAhuUa%Y?FJQ@H118+%0K~z|Uy_Q{Q990y@e`hw$Hfgf63b7QbQWKCig*Lk}QYti1
zwN32?EKD}rP^x(lMEpR(2k{|KA{GQw+qafVq-Kew;0J<GN}-^rTebD6icrLcM4Iew
z)siN;$A`Gd&bYI)vyDAZbN~08|M_utX77bniW3Juc>3glQ%A$r$!ato85yx2xcuN%
zU{ge`&4<q(zFKRo6}H)|)3vMMboZ2N8?S8I2DAf~5t9ky7YaZ@u;jQ}<|FOxZCEi`
zEC7oDs#^Tt0E(H!9)T^uLpRqK_m=eGGsC~4x*O4F=G*4q598SooU#(eaIo~!7)Sx1
zEkm3gz5rYZ;|C!opnV-tf&S&SlG0UpiE9SP+pT*5yR;TzLovO*`>p|qlJ;swtf%e@
zz;DiuCXiGGsD7_D<0S3;U{}qPsR^JZGPf6Kt+3T~BeSj3Wf;+SQMowdBp+KffJ);p
zMZ`p2D?R5k-A|xA3q&nT$F#2_8t5IAa6Of5Zf@3TV4Y8jL3kc0p!y!j`W16|!OrH6
z-w{B5ATfZD4a%z&s!aZ=(f0c%zPja?Q^QXnPCd<G=~Hcy+XK~dssc<qoh^v&s}}%3
zFhg|%M6JgCKy#H8188*u>HVXBXQeMF7(i7jjn9CbP_YuWX>YWrVvU07QQ<=$m^IO+
zTf6*v_#Q?#h8~TvFyH5q7eO-dY;Iy{`CUo>7C1CJm^dfONjxvNDoM%?uJZ7mK*PfA
zeZX3e{Ps%o|HeJzmre<cdUdtf;#cDfps?E`%^4#{{PLF#8yDW<-^)oUMA=XWfVijy
zzPWcIR|o=V9lJaY{OI-~9vT2ok*Z7uH7^Tq?p$a9fxq3E+E#OW;Xe0Ley<1MCzqh~
zOgo(|LA`G}U2Q1a%Qd6sT4(@7KK3Y0QETn1K>)^*H-L!C3O)@D5R368;2NExZx+*u
z{eCrtbYe*1C0C6yX=}%MznZG2;mklHWeJ~n<OP9aM9eY!^bbD+%%r=wTIegF!(LAj
z8M4Q7r|SX$m`x`~1%1(H`!ANJ*yOXHjb|?nR`U-$j^o+f%fLH6+v}<{AGd8e6y#s?
zy*{5#J_$MsJX}3frxad~XTKklpp!BT0DwN1TvIHV$93tF@(8dIhys(KUyE3u#|nir
g^2IN{e>~p!2hn)}dc<Aup#T5?07*qoM6N<$f|bz8KL7v#

literal 0
HcmV?d00001

diff --git a/0.16/_static/img/pytorch-logo-flame.svg b/0.16/_static/img/pytorch-logo-flame.svg
new file mode 100644
index 00000000000..5f2fb76be77
--- /dev/null
+++ b/0.16/_static/img/pytorch-logo-flame.svg
@@ -0,0 +1,33 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   xmlns:dc="http://purl.org/dc/elements/1.1/"
+   xmlns:cc="http://creativecommons.org/ns#"
+   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+   xmlns:svg="http://www.w3.org/2000/svg"
+   xmlns="http://www.w3.org/2000/svg"
+   height="40.200001"
+   width="40.200001"
+   xml:space="preserve"
+   viewBox="0 0 40.200002 40.2"
+   y="0px"
+   x="0px"
+   id="Layer_1"
+   version="1.1"><metadata
+     id="metadata4717"><rdf:RDF><cc:Work
+         rdf:about=""><dc:format>image/svg+xml</dc:format><dc:type
+           rdf:resource="http://purl.org/dc/dcmitype/StillImage" /><dc:title></dc:title></cc:Work></rdf:RDF></metadata><defs
+     id="defs4715" /><style
+     id="style4694"
+     type="text/css">
+	.st0{fill:#F05732;}
+	.st1{fill:#9E529F;}
+	.st2{fill:#333333;}
+</style><path
+     style="fill:#f05732"
+     id="path4696"
+     d="m 26.975479,12.199999 c -1.3,-1 -1.8,3.9 -4.4,3.9 -3,0 -4,-12.9999998 -6.3,-12.9999998 -0.7,0 -0.8,-0.4 -7.9000003,21.2999998 -2.9000001,9 4.4000003,15.8 11.8000003,15.8 4.6,0 12.3,-3 12.3,-12.6 0,-7.1 -3.5,-13.9 -5.5,-15.4 z m -6.9,23.1 c -3.7,0 -6.7,-3.1 -6.7,-7 0,-3.9 3,-7 6.7,-7 3.7,0 6.7,3.1 6.7,7 0,3.8 -3,7 -6.7,7 z"
+     class="st0" /><path
+     style="fill:#9e529f"
+     id="path4698"
+     d="m 24.075479,-7.6293945e-7 c -0.5,0 -1.8,2.49999996293945 -1.8,3.59999996293945 0,1.5 1,2 1.8,2 0.8,0 1.8,-0.5 1.8,-2 -0.1,-1.1 -1.4,-3.59999996293945 -1.8,-3.59999996293945 z"
+     class="st1" /></svg>
diff --git a/0.16/_static/jquery-3.6.0.js b/0.16/_static/jquery-3.6.0.js
new file mode 100644
index 00000000000..fc6c299b73e
--- /dev/null
+++ b/0.16/_static/jquery-3.6.0.js
@@ -0,0 +1,10881 @@
+/*!
+ * jQuery JavaScript Library v3.6.0
+ * https://jquery.com/
+ *
+ * Includes Sizzle.js
+ * https://sizzlejs.com/
+ *
+ * Copyright OpenJS Foundation and other contributors
+ * Released under the MIT license
+ * https://jquery.org/license
+ *
+ * Date: 2021-03-02T17:08Z
+ */
+( function( global, factory ) {
+
+	"use strict";
+
+	if ( typeof module === "object" && typeof module.exports === "object" ) {
+
+		// For CommonJS and CommonJS-like environments where a proper `window`
+		// is present, execute the factory and get jQuery.
+		// For environments that do not have a `window` with a `document`
+		// (such as Node.js), expose a factory as module.exports.
+		// This accentuates the need for the creation of a real `window`.
+		// e.g. var jQuery = require("jquery")(window);
+		// See ticket #14549 for more info.
+		module.exports = global.document ?
+			factory( global, true ) :
+			function( w ) {
+				if ( !w.document ) {
+					throw new Error( "jQuery requires a window with a document" );
+				}
+				return factory( w );
+			};
+	} else {
+		factory( global );
+	}
+
+// Pass this if window is not defined yet
+} )( typeof window !== "undefined" ? window : this, function( window, noGlobal ) {
+
+// Edge <= 12 - 13+, Firefox <=18 - 45+, IE 10 - 11, Safari 5.1 - 9+, iOS 6 - 9.1
+// throw exceptions when non-strict code (e.g., ASP.NET 4.5) accesses strict mode
+// arguments.callee.caller (trac-13335). But as of jQuery 3.0 (2016), strict mode should be common
+// enough that all such attempts are guarded in a try block.
+"use strict";
+
+var arr = [];
+
+var getProto = Object.getPrototypeOf;
+
+var slice = arr.slice;
+
+var flat = arr.flat ? function( array ) {
+	return arr.flat.call( array );
+} : function( array ) {
+	return arr.concat.apply( [], array );
+};
+
+
+var push = arr.push;
+
+var indexOf = arr.indexOf;
+
+var class2type = {};
+
+var toString = class2type.toString;
+
+var hasOwn = class2type.hasOwnProperty;
+
+var fnToString = hasOwn.toString;
+
+var ObjectFunctionString = fnToString.call( Object );
+
+var support = {};
+
+var isFunction = function isFunction( obj ) {
+
+		// Support: Chrome <=57, Firefox <=52
+		// In some browsers, typeof returns "function" for HTML <object> elements
+		// (i.e., `typeof document.createElement( "object" ) === "function"`).
+		// We don't want to classify *any* DOM node as a function.
+		// Support: QtWeb <=3.8.5, WebKit <=534.34, wkhtmltopdf tool <=0.12.5
+		// Plus for old WebKit, typeof returns "function" for HTML collections
+		// (e.g., `typeof document.getElementsByTagName("div") === "function"`). (gh-4756)
+		return typeof obj === "function" && typeof obj.nodeType !== "number" &&
+			typeof obj.item !== "function";
+	};
+
+
+var isWindow = function isWindow( obj ) {
+		return obj != null && obj === obj.window;
+	};
+
+
+var document = window.document;
+
+
+
+	var preservedScriptAttributes = {
+		type: true,
+		src: true,
+		nonce: true,
+		noModule: true
+	};
+
+	function DOMEval( code, node, doc ) {
+		doc = doc || document;
+
+		var i, val,
+			script = doc.createElement( "script" );
+
+		script.text = code;
+		if ( node ) {
+			for ( i in preservedScriptAttributes ) {
+
+				// Support: Firefox 64+, Edge 18+
+				// Some browsers don't support the "nonce" property on scripts.
+				// On the other hand, just using `getAttribute` is not enough as
+				// the `nonce` attribute is reset to an empty string whenever it
+				// becomes browsing-context connected.
+				// See https://github.com/whatwg/html/issues/2369
+				// See https://html.spec.whatwg.org/#nonce-attributes
+				// The `node.getAttribute` check was added for the sake of
+				// `jQuery.globalEval` so that it can fake a nonce-containing node
+				// via an object.
+				val = node[ i ] || node.getAttribute && node.getAttribute( i );
+				if ( val ) {
+					script.setAttribute( i, val );
+				}
+			}
+		}
+		doc.head.appendChild( script ).parentNode.removeChild( script );
+	}
+
+
+function toType( obj ) {
+	if ( obj == null ) {
+		return obj + "";
+	}
+
+	// Support: Android <=2.3 only (functionish RegExp)
+	return typeof obj === "object" || typeof obj === "function" ?
+		class2type[ toString.call( obj ) ] || "object" :
+		typeof obj;
+}
+/* global Symbol */
+// Defining this global in .eslintrc.json would create a danger of using the global
+// unguarded in another place, it seems safer to define global only for this module
+
+
+
+var
+	version = "3.6.0",
+
+	// Define a local copy of jQuery
+	jQuery = function( selector, context ) {
+
+		// The jQuery object is actually just the init constructor 'enhanced'
+		// Need init if jQuery is called (just allow error to be thrown if not included)
+		return new jQuery.fn.init( selector, context );
+	};
+
+jQuery.fn = jQuery.prototype = {
+
+	// The current version of jQuery being used
+	jquery: version,
+
+	constructor: jQuery,
+
+	// The default length of a jQuery object is 0
+	length: 0,
+
+	toArray: function() {
+		return slice.call( this );
+	},
+
+	// Get the Nth element in the matched element set OR
+	// Get the whole matched element set as a clean array
+	get: function( num ) {
+
+		// Return all the elements in a clean array
+		if ( num == null ) {
+			return slice.call( this );
+		}
+
+		// Return just the one element from the set
+		return num < 0 ? this[ num + this.length ] : this[ num ];
+	},
+
+	// Take an array of elements and push it onto the stack
+	// (returning the new matched element set)
+	pushStack: function( elems ) {
+
+		// Build a new jQuery matched element set
+		var ret = jQuery.merge( this.constructor(), elems );
+
+		// Add the old object onto the stack (as a reference)
+		ret.prevObject = this;
+
+		// Return the newly-formed element set
+		return ret;
+	},
+
+	// Execute a callback for every element in the matched set.
+	each: function( callback ) {
+		return jQuery.each( this, callback );
+	},
+
+	map: function( callback ) {
+		return this.pushStack( jQuery.map( this, function( elem, i ) {
+			return callback.call( elem, i, elem );
+		} ) );
+	},
+
+	slice: function() {
+		return this.pushStack( slice.apply( this, arguments ) );
+	},
+
+	first: function() {
+		return this.eq( 0 );
+	},
+
+	last: function() {
+		return this.eq( -1 );
+	},
+
+	even: function() {
+		return this.pushStack( jQuery.grep( this, function( _elem, i ) {
+			return ( i + 1 ) % 2;
+		} ) );
+	},
+
+	odd: function() {
+		return this.pushStack( jQuery.grep( this, function( _elem, i ) {
+			return i % 2;
+		} ) );
+	},
+
+	eq: function( i ) {
+		var len = this.length,
+			j = +i + ( i < 0 ? len : 0 );
+		return this.pushStack( j >= 0 && j < len ? [ this[ j ] ] : [] );
+	},
+
+	end: function() {
+		return this.prevObject || this.constructor();
+	},
+
+	// For internal use only.
+	// Behaves like an Array's method, not like a jQuery method.
+	push: push,
+	sort: arr.sort,
+	splice: arr.splice
+};
+
+jQuery.extend = jQuery.fn.extend = function() {
+	var options, name, src, copy, copyIsArray, clone,
+		target = arguments[ 0 ] || {},
+		i = 1,
+		length = arguments.length,
+		deep = false;
+
+	// Handle a deep copy situation
+	if ( typeof target === "boolean" ) {
+		deep = target;
+
+		// Skip the boolean and the target
+		target = arguments[ i ] || {};
+		i++;
+	}
+
+	// Handle case when target is a string or something (possible in deep copy)
+	if ( typeof target !== "object" && !isFunction( target ) ) {
+		target = {};
+	}
+
+	// Extend jQuery itself if only one argument is passed
+	if ( i === length ) {
+		target = this;
+		i--;
+	}
+
+	for ( ; i < length; i++ ) {
+
+		// Only deal with non-null/undefined values
+		if ( ( options = arguments[ i ] ) != null ) {
+
+			// Extend the base object
+			for ( name in options ) {
+				copy = options[ name ];
+
+				// Prevent Object.prototype pollution
+				// Prevent never-ending loop
+				if ( name === "__proto__" || target === copy ) {
+					continue;
+				}
+
+				// Recurse if we're merging plain objects or arrays
+				if ( deep && copy && ( jQuery.isPlainObject( copy ) ||
+					( copyIsArray = Array.isArray( copy ) ) ) ) {
+					src = target[ name ];
+
+					// Ensure proper type for the source value
+					if ( copyIsArray && !Array.isArray( src ) ) {
+						clone = [];
+					} else if ( !copyIsArray && !jQuery.isPlainObject( src ) ) {
+						clone = {};
+					} else {
+						clone = src;
+					}
+					copyIsArray = false;
+
+					// Never move original objects, clone them
+					target[ name ] = jQuery.extend( deep, clone, copy );
+
+				// Don't bring in undefined values
+				} else if ( copy !== undefined ) {
+					target[ name ] = copy;
+				}
+			}
+		}
+	}
+
+	// Return the modified object
+	return target;
+};
+
+jQuery.extend( {
+
+	// Unique for each copy of jQuery on the page
+	expando: "jQuery" + ( version + Math.random() ).replace( /\D/g, "" ),
+
+	// Assume jQuery is ready without the ready module
+	isReady: true,
+
+	error: function( msg ) {
+		throw new Error( msg );
+	},
+
+	noop: function() {},
+
+	isPlainObject: function( obj ) {
+		var proto, Ctor;
+
+		// Detect obvious negatives
+		// Use toString instead of jQuery.type to catch host objects
+		if ( !obj || toString.call( obj ) !== "[object Object]" ) {
+			return false;
+		}
+
+		proto = getProto( obj );
+
+		// Objects with no prototype (e.g., `Object.create( null )`) are plain
+		if ( !proto ) {
+			return true;
+		}
+
+		// Objects with prototype are plain iff they were constructed by a global Object function
+		Ctor = hasOwn.call( proto, "constructor" ) && proto.constructor;
+		return typeof Ctor === "function" && fnToString.call( Ctor ) === ObjectFunctionString;
+	},
+
+	isEmptyObject: function( obj ) {
+		var name;
+
+		for ( name in obj ) {
+			return false;
+		}
+		return true;
+	},
+
+	// Evaluates a script in a provided context; falls back to the global one
+	// if not specified.
+	globalEval: function( code, options, doc ) {
+		DOMEval( code, { nonce: options && options.nonce }, doc );
+	},
+
+	each: function( obj, callback ) {
+		var length, i = 0;
+
+		if ( isArrayLike( obj ) ) {
+			length = obj.length;
+			for ( ; i < length; i++ ) {
+				if ( callback.call( obj[ i ], i, obj[ i ] ) === false ) {
+					break;
+				}
+			}
+		} else {
+			for ( i in obj ) {
+				if ( callback.call( obj[ i ], i, obj[ i ] ) === false ) {
+					break;
+				}
+			}
+		}
+
+		return obj;
+	},
+
+	// results is for internal usage only
+	makeArray: function( arr, results ) {
+		var ret = results || [];
+
+		if ( arr != null ) {
+			if ( isArrayLike( Object( arr ) ) ) {
+				jQuery.merge( ret,
+					typeof arr === "string" ?
+						[ arr ] : arr
+				);
+			} else {
+				push.call( ret, arr );
+			}
+		}
+
+		return ret;
+	},
+
+	inArray: function( elem, arr, i ) {
+		return arr == null ? -1 : indexOf.call( arr, elem, i );
+	},
+
+	// Support: Android <=4.0 only, PhantomJS 1 only
+	// push.apply(_, arraylike) throws on ancient WebKit
+	merge: function( first, second ) {
+		var len = +second.length,
+			j = 0,
+			i = first.length;
+
+		for ( ; j < len; j++ ) {
+			first[ i++ ] = second[ j ];
+		}
+
+		first.length = i;
+
+		return first;
+	},
+
+	grep: function( elems, callback, invert ) {
+		var callbackInverse,
+			matches = [],
+			i = 0,
+			length = elems.length,
+			callbackExpect = !invert;
+
+		// Go through the array, only saving the items
+		// that pass the validator function
+		for ( ; i < length; i++ ) {
+			callbackInverse = !callback( elems[ i ], i );
+			if ( callbackInverse !== callbackExpect ) {
+				matches.push( elems[ i ] );
+			}
+		}
+
+		return matches;
+	},
+
+	// arg is for internal usage only
+	map: function( elems, callback, arg ) {
+		var length, value,
+			i = 0,
+			ret = [];
+
+		// Go through the array, translating each of the items to their new values
+		if ( isArrayLike( elems ) ) {
+			length = elems.length;
+			for ( ; i < length; i++ ) {
+				value = callback( elems[ i ], i, arg );
+
+				if ( value != null ) {
+					ret.push( value );
+				}
+			}
+
+		// Go through every key on the object,
+		} else {
+			for ( i in elems ) {
+				value = callback( elems[ i ], i, arg );
+
+				if ( value != null ) {
+					ret.push( value );
+				}
+			}
+		}
+
+		// Flatten any nested arrays
+		return flat( ret );
+	},
+
+	// A global GUID counter for objects
+	guid: 1,
+
+	// jQuery.support is not used in Core but other projects attach their
+	// properties to it so it needs to exist.
+	support: support
+} );
+
+if ( typeof Symbol === "function" ) {
+	jQuery.fn[ Symbol.iterator ] = arr[ Symbol.iterator ];
+}
+
+// Populate the class2type map
+jQuery.each( "Boolean Number String Function Array Date RegExp Object Error Symbol".split( " " ),
+	function( _i, name ) {
+		class2type[ "[object " + name + "]" ] = name.toLowerCase();
+	} );
+
+function isArrayLike( obj ) {
+
+	// Support: real iOS 8.2 only (not reproducible in simulator)
+	// `in` check used to prevent JIT error (gh-2145)
+	// hasOwn isn't used here due to false negatives
+	// regarding Nodelist length in IE
+	var length = !!obj && "length" in obj && obj.length,
+		type = toType( obj );
+
+	if ( isFunction( obj ) || isWindow( obj ) ) {
+		return false;
+	}
+
+	return type === "array" || length === 0 ||
+		typeof length === "number" && length > 0 && ( length - 1 ) in obj;
+}
+var Sizzle =
+/*!
+ * Sizzle CSS Selector Engine v2.3.6
+ * https://sizzlejs.com/
+ *
+ * Copyright JS Foundation and other contributors
+ * Released under the MIT license
+ * https://js.foundation/
+ *
+ * Date: 2021-02-16
+ */
+( function( window ) {
+var i,
+	support,
+	Expr,
+	getText,
+	isXML,
+	tokenize,
+	compile,
+	select,
+	outermostContext,
+	sortInput,
+	hasDuplicate,
+
+	// Local document vars
+	setDocument,
+	document,
+	docElem,
+	documentIsHTML,
+	rbuggyQSA,
+	rbuggyMatches,
+	matches,
+	contains,
+
+	// Instance-specific data
+	expando = "sizzle" + 1 * new Date(),
+	preferredDoc = window.document,
+	dirruns = 0,
+	done = 0,
+	classCache = createCache(),
+	tokenCache = createCache(),
+	compilerCache = createCache(),
+	nonnativeSelectorCache = createCache(),
+	sortOrder = function( a, b ) {
+		if ( a === b ) {
+			hasDuplicate = true;
+		}
+		return 0;
+	},
+
+	// Instance methods
+	hasOwn = ( {} ).hasOwnProperty,
+	arr = [],
+	pop = arr.pop,
+	pushNative = arr.push,
+	push = arr.push,
+	slice = arr.slice,
+
+	// Use a stripped-down indexOf as it's faster than native
+	// https://jsperf.com/thor-indexof-vs-for/5
+	indexOf = function( list, elem ) {
+		var i = 0,
+			len = list.length;
+		for ( ; i < len; i++ ) {
+			if ( list[ i ] === elem ) {
+				return i;
+			}
+		}
+		return -1;
+	},
+
+	booleans = "checked|selected|async|autofocus|autoplay|controls|defer|disabled|hidden|" +
+		"ismap|loop|multiple|open|readonly|required|scoped",
+
+	// Regular expressions
+
+	// http://www.w3.org/TR/css3-selectors/#whitespace
+	whitespace = "[\\x20\\t\\r\\n\\f]",
+
+	// https://www.w3.org/TR/css-syntax-3/#ident-token-diagram
+	identifier = "(?:\\\\[\\da-fA-F]{1,6}" + whitespace +
+		"?|\\\\[^\\r\\n\\f]|[\\w-]|[^\0-\\x7f])+",
+
+	// Attribute selectors: http://www.w3.org/TR/selectors/#attribute-selectors
+	attributes = "\\[" + whitespace + "*(" + identifier + ")(?:" + whitespace +
+
+		// Operator (capture 2)
+		"*([*^$|!~]?=)" + whitespace +
+
+		// "Attribute values must be CSS identifiers [capture 5]
+		// or strings [capture 3 or capture 4]"
+		"*(?:'((?:\\\\.|[^\\\\'])*)'|\"((?:\\\\.|[^\\\\\"])*)\"|(" + identifier + "))|)" +
+		whitespace + "*\\]",
+
+	pseudos = ":(" + identifier + ")(?:\\((" +
+
+		// To reduce the number of selectors needing tokenize in the preFilter, prefer arguments:
+		// 1. quoted (capture 3; capture 4 or capture 5)
+		"('((?:\\\\.|[^\\\\'])*)'|\"((?:\\\\.|[^\\\\\"])*)\")|" +
+
+		// 2. simple (capture 6)
+		"((?:\\\\.|[^\\\\()[\\]]|" + attributes + ")*)|" +
+
+		// 3. anything else (capture 2)
+		".*" +
+		")\\)|)",
+
+	// Leading and non-escaped trailing whitespace, capturing some non-whitespace characters preceding the latter
+	rwhitespace = new RegExp( whitespace + "+", "g" ),
+	rtrim = new RegExp( "^" + whitespace + "+|((?:^|[^\\\\])(?:\\\\.)*)" +
+		whitespace + "+$", "g" ),
+
+	rcomma = new RegExp( "^" + whitespace + "*," + whitespace + "*" ),
+	rcombinators = new RegExp( "^" + whitespace + "*([>+~]|" + whitespace + ")" + whitespace +
+		"*" ),
+	rdescend = new RegExp( whitespace + "|>" ),
+
+	rpseudo = new RegExp( pseudos ),
+	ridentifier = new RegExp( "^" + identifier + "$" ),
+
+	matchExpr = {
+		"ID": new RegExp( "^#(" + identifier + ")" ),
+		"CLASS": new RegExp( "^\\.(" + identifier + ")" ),
+		"TAG": new RegExp( "^(" + identifier + "|[*])" ),
+		"ATTR": new RegExp( "^" + attributes ),
+		"PSEUDO": new RegExp( "^" + pseudos ),
+		"CHILD": new RegExp( "^:(only|first|last|nth|nth-last)-(child|of-type)(?:\\(" +
+			whitespace + "*(even|odd|(([+-]|)(\\d*)n|)" + whitespace + "*(?:([+-]|)" +
+			whitespace + "*(\\d+)|))" + whitespace + "*\\)|)", "i" ),
+		"bool": new RegExp( "^(?:" + booleans + ")$", "i" ),
+
+		// For use in libraries implementing .is()
+		// We use this for POS matching in `select`
+		"needsContext": new RegExp( "^" + whitespace +
+			"*[>+~]|:(even|odd|eq|gt|lt|nth|first|last)(?:\\(" + whitespace +
+			"*((?:-\\d)?\\d*)" + whitespace + "*\\)|)(?=[^-]|$)", "i" )
+	},
+
+	rhtml = /HTML$/i,
+	rinputs = /^(?:input|select|textarea|button)$/i,
+	rheader = /^h\d$/i,
+
+	rnative = /^[^{]+\{\s*\[native \w/,
+
+	// Easily-parseable/retrievable ID or TAG or CLASS selectors
+	rquickExpr = /^(?:#([\w-]+)|(\w+)|\.([\w-]+))$/,
+
+	rsibling = /[+~]/,
+
+	// CSS escapes
+	// http://www.w3.org/TR/CSS21/syndata.html#escaped-characters
+	runescape = new RegExp( "\\\\[\\da-fA-F]{1,6}" + whitespace + "?|\\\\([^\\r\\n\\f])", "g" ),
+	funescape = function( escape, nonHex ) {
+		var high = "0x" + escape.slice( 1 ) - 0x10000;
+
+		return nonHex ?
+
+			// Strip the backslash prefix from a non-hex escape sequence
+			nonHex :
+
+			// Replace a hexadecimal escape sequence with the encoded Unicode code point
+			// Support: IE <=11+
+			// For values outside the Basic Multilingual Plane (BMP), manually construct a
+			// surrogate pair
+			high < 0 ?
+				String.fromCharCode( high + 0x10000 ) :
+				String.fromCharCode( high >> 10 | 0xD800, high & 0x3FF | 0xDC00 );
+	},
+
+	// CSS string/identifier serialization
+	// https://drafts.csswg.org/cssom/#common-serializing-idioms
+	rcssescape = /([\0-\x1f\x7f]|^-?\d)|^-$|[^\0-\x1f\x7f-\uFFFF\w-]/g,
+	fcssescape = function( ch, asCodePoint ) {
+		if ( asCodePoint ) {
+
+			// U+0000 NULL becomes U+FFFD REPLACEMENT CHARACTER
+			if ( ch === "\0" ) {
+				return "\uFFFD";
+			}
+
+			// Control characters and (dependent upon position) numbers get escaped as code points
+			return ch.slice( 0, -1 ) + "\\" +
+				ch.charCodeAt( ch.length - 1 ).toString( 16 ) + " ";
+		}
+
+		// Other potentially-special ASCII characters get backslash-escaped
+		return "\\" + ch;
+	},
+
+	// Used for iframes
+	// See setDocument()
+	// Removing the function wrapper causes a "Permission Denied"
+	// error in IE
+	unloadHandler = function() {
+		setDocument();
+	},
+
+	inDisabledFieldset = addCombinator(
+		function( elem ) {
+			return elem.disabled === true && elem.nodeName.toLowerCase() === "fieldset";
+		},
+		{ dir: "parentNode", next: "legend" }
+	);
+
+// Optimize for push.apply( _, NodeList )
+try {
+	push.apply(
+		( arr = slice.call( preferredDoc.childNodes ) ),
+		preferredDoc.childNodes
+	);
+
+	// Support: Android<4.0
+	// Detect silently failing push.apply
+	// eslint-disable-next-line no-unused-expressions
+	arr[ preferredDoc.childNodes.length ].nodeType;
+} catch ( e ) {
+	push = { apply: arr.length ?
+
+		// Leverage slice if possible
+		function( target, els ) {
+			pushNative.apply( target, slice.call( els ) );
+		} :
+
+		// Support: IE<9
+		// Otherwise append directly
+		function( target, els ) {
+			var j = target.length,
+				i = 0;
+
+			// Can't trust NodeList.length
+			while ( ( target[ j++ ] = els[ i++ ] ) ) {}
+			target.length = j - 1;
+		}
+	};
+}
+
+function Sizzle( selector, context, results, seed ) {
+	var m, i, elem, nid, match, groups, newSelector,
+		newContext = context && context.ownerDocument,
+
+		// nodeType defaults to 9, since context defaults to document
+		nodeType = context ? context.nodeType : 9;
+
+	results = results || [];
+
+	// Return early from calls with invalid selector or context
+	if ( typeof selector !== "string" || !selector ||
+		nodeType !== 1 && nodeType !== 9 && nodeType !== 11 ) {
+
+		return results;
+	}
+
+	// Try to shortcut find operations (as opposed to filters) in HTML documents
+	if ( !seed ) {
+		setDocument( context );
+		context = context || document;
+
+		if ( documentIsHTML ) {
+
+			// If the selector is sufficiently simple, try using a "get*By*" DOM method
+			// (excepting DocumentFragment context, where the methods don't exist)
+			if ( nodeType !== 11 && ( match = rquickExpr.exec( selector ) ) ) {
+
+				// ID selector
+				if ( ( m = match[ 1 ] ) ) {
+
+					// Document context
+					if ( nodeType === 9 ) {
+						if ( ( elem = context.getElementById( m ) ) ) {
+
+							// Support: IE, Opera, Webkit
+							// TODO: identify versions
+							// getElementById can match elements by name instead of ID
+							if ( elem.id === m ) {
+								results.push( elem );
+								return results;
+							}
+						} else {
+							return results;
+						}
+
+					// Element context
+					} else {
+
+						// Support: IE, Opera, Webkit
+						// TODO: identify versions
+						// getElementById can match elements by name instead of ID
+						if ( newContext && ( elem = newContext.getElementById( m ) ) &&
+							contains( context, elem ) &&
+							elem.id === m ) {
+
+							results.push( elem );
+							return results;
+						}
+					}
+
+				// Type selector
+				} else if ( match[ 2 ] ) {
+					push.apply( results, context.getElementsByTagName( selector ) );
+					return results;
+
+				// Class selector
+				} else if ( ( m = match[ 3 ] ) && support.getElementsByClassName &&
+					context.getElementsByClassName ) {
+
+					push.apply( results, context.getElementsByClassName( m ) );
+					return results;
+				}
+			}
+
+			// Take advantage of querySelectorAll
+			if ( support.qsa &&
+				!nonnativeSelectorCache[ selector + " " ] &&
+				( !rbuggyQSA || !rbuggyQSA.test( selector ) ) &&
+
+				// Support: IE 8 only
+				// Exclude object elements
+				( nodeType !== 1 || context.nodeName.toLowerCase() !== "object" ) ) {
+
+				newSelector = selector;
+				newContext = context;
+
+				// qSA considers elements outside a scoping root when evaluating child or
+				// descendant combinators, which is not what we want.
+				// In such cases, we work around the behavior by prefixing every selector in the
+				// list with an ID selector referencing the scope context.
+				// The technique has to be used as well when a leading combinator is used
+				// as such selectors are not recognized by querySelectorAll.
+				// Thanks to Andrew Dupont for this technique.
+				if ( nodeType === 1 &&
+					( rdescend.test( selector ) || rcombinators.test( selector ) ) ) {
+
+					// Expand context for sibling selectors
+					newContext = rsibling.test( selector ) && testContext( context.parentNode ) ||
+						context;
+
+					// We can use :scope instead of the ID hack if the browser
+					// supports it & if we're not changing the context.
+					if ( newContext !== context || !support.scope ) {
+
+						// Capture the context ID, setting it first if necessary
+						if ( ( nid = context.getAttribute( "id" ) ) ) {
+							nid = nid.replace( rcssescape, fcssescape );
+						} else {
+							context.setAttribute( "id", ( nid = expando ) );
+						}
+					}
+
+					// Prefix every selector in the list
+					groups = tokenize( selector );
+					i = groups.length;
+					while ( i-- ) {
+						groups[ i ] = ( nid ? "#" + nid : ":scope" ) + " " +
+							toSelector( groups[ i ] );
+					}
+					newSelector = groups.join( "," );
+				}
+
+				try {
+					push.apply( results,
+						newContext.querySelectorAll( newSelector )
+					);
+					return results;
+				} catch ( qsaError ) {
+					nonnativeSelectorCache( selector, true );
+				} finally {
+					if ( nid === expando ) {
+						context.removeAttribute( "id" );
+					}
+				}
+			}
+		}
+	}
+
+	// All others
+	return select( selector.replace( rtrim, "$1" ), context, results, seed );
+}
+
+/**
+ * Create key-value caches of limited size
+ * @returns {function(string, object)} Returns the Object data after storing it on itself with
+ *	property name the (space-suffixed) string and (if the cache is larger than Expr.cacheLength)
+ *	deleting the oldest entry
+ */
+function createCache() {
+	var keys = [];
+
+	function cache( key, value ) {
+
+		// Use (key + " ") to avoid collision with native prototype properties (see Issue #157)
+		if ( keys.push( key + " " ) > Expr.cacheLength ) {
+
+			// Only keep the most recent entries
+			delete cache[ keys.shift() ];
+		}
+		return ( cache[ key + " " ] = value );
+	}
+	return cache;
+}
+
+/**
+ * Mark a function for special use by Sizzle
+ * @param {Function} fn The function to mark
+ */
+function markFunction( fn ) {
+	fn[ expando ] = true;
+	return fn;
+}
+
+/**
+ * Support testing using an element
+ * @param {Function} fn Passed the created element and returns a boolean result
+ */
+function assert( fn ) {
+	var el = document.createElement( "fieldset" );
+
+	try {
+		return !!fn( el );
+	} catch ( e ) {
+		return false;
+	} finally {
+
+		// Remove from its parent by default
+		if ( el.parentNode ) {
+			el.parentNode.removeChild( el );
+		}
+
+		// release memory in IE
+		el = null;
+	}
+}
+
+/**
+ * Adds the same handler for all of the specified attrs
+ * @param {String} attrs Pipe-separated list of attributes
+ * @param {Function} handler The method that will be applied
+ */
+function addHandle( attrs, handler ) {
+	var arr = attrs.split( "|" ),
+		i = arr.length;
+
+	while ( i-- ) {
+		Expr.attrHandle[ arr[ i ] ] = handler;
+	}
+}
+
+/**
+ * Checks document order of two siblings
+ * @param {Element} a
+ * @param {Element} b
+ * @returns {Number} Returns less than 0 if a precedes b, greater than 0 if a follows b
+ */
+function siblingCheck( a, b ) {
+	var cur = b && a,
+		diff = cur && a.nodeType === 1 && b.nodeType === 1 &&
+			a.sourceIndex - b.sourceIndex;
+
+	// Use IE sourceIndex if available on both nodes
+	if ( diff ) {
+		return diff;
+	}
+
+	// Check if b follows a
+	if ( cur ) {
+		while ( ( cur = cur.nextSibling ) ) {
+			if ( cur === b ) {
+				return -1;
+			}
+		}
+	}
+
+	return a ? 1 : -1;
+}
+
+/**
+ * Returns a function to use in pseudos for input types
+ * @param {String} type
+ */
+function createInputPseudo( type ) {
+	return function( elem ) {
+		var name = elem.nodeName.toLowerCase();
+		return name === "input" && elem.type === type;
+	};
+}
+
+/**
+ * Returns a function to use in pseudos for buttons
+ * @param {String} type
+ */
+function createButtonPseudo( type ) {
+	return function( elem ) {
+		var name = elem.nodeName.toLowerCase();
+		return ( name === "input" || name === "button" ) && elem.type === type;
+	};
+}
+
+/**
+ * Returns a function to use in pseudos for :enabled/:disabled
+ * @param {Boolean} disabled true for :disabled; false for :enabled
+ */
+function createDisabledPseudo( disabled ) {
+
+	// Known :disabled false positives: fieldset[disabled] > legend:nth-of-type(n+2) :can-disable
+	return function( elem ) {
+
+		// Only certain elements can match :enabled or :disabled
+		// https://html.spec.whatwg.org/multipage/scripting.html#selector-enabled
+		// https://html.spec.whatwg.org/multipage/scripting.html#selector-disabled
+		if ( "form" in elem ) {
+
+			// Check for inherited disabledness on relevant non-disabled elements:
+			// * listed form-associated elements in a disabled fieldset
+			//   https://html.spec.whatwg.org/multipage/forms.html#category-listed
+			//   https://html.spec.whatwg.org/multipage/forms.html#concept-fe-disabled
+			// * option elements in a disabled optgroup
+			//   https://html.spec.whatwg.org/multipage/forms.html#concept-option-disabled
+			// All such elements have a "form" property.
+			if ( elem.parentNode && elem.disabled === false ) {
+
+				// Option elements defer to a parent optgroup if present
+				if ( "label" in elem ) {
+					if ( "label" in elem.parentNode ) {
+						return elem.parentNode.disabled === disabled;
+					} else {
+						return elem.disabled === disabled;
+					}
+				}
+
+				// Support: IE 6 - 11
+				// Use the isDisabled shortcut property to check for disabled fieldset ancestors
+				return elem.isDisabled === disabled ||
+
+					// Where there is no isDisabled, check manually
+					/* jshint -W018 */
+					elem.isDisabled !== !disabled &&
+					inDisabledFieldset( elem ) === disabled;
+			}
+
+			return elem.disabled === disabled;
+
+		// Try to winnow out elements that can't be disabled before trusting the disabled property.
+		// Some victims get caught in our net (label, legend, menu, track), but it shouldn't
+		// even exist on them, let alone have a boolean value.
+		} else if ( "label" in elem ) {
+			return elem.disabled === disabled;
+		}
+
+		// Remaining elements are neither :enabled nor :disabled
+		return false;
+	};
+}
+
+/**
+ * Returns a function to use in pseudos for positionals
+ * @param {Function} fn
+ */
+function createPositionalPseudo( fn ) {
+	return markFunction( function( argument ) {
+		argument = +argument;
+		return markFunction( function( seed, matches ) {
+			var j,
+				matchIndexes = fn( [], seed.length, argument ),
+				i = matchIndexes.length;
+
+			// Match elements found at the specified indexes
+			while ( i-- ) {
+				if ( seed[ ( j = matchIndexes[ i ] ) ] ) {
+					seed[ j ] = !( matches[ j ] = seed[ j ] );
+				}
+			}
+		} );
+	} );
+}
+
+/**
+ * Checks a node for validity as a Sizzle context
+ * @param {Element|Object=} context
+ * @returns {Element|Object|Boolean} The input node if acceptable, otherwise a falsy value
+ */
+function testContext( context ) {
+	return context && typeof context.getElementsByTagName !== "undefined" && context;
+}
+
+// Expose support vars for convenience
+support = Sizzle.support = {};
+
+/**
+ * Detects XML nodes
+ * @param {Element|Object} elem An element or a document
+ * @returns {Boolean} True iff elem is a non-HTML XML node
+ */
+isXML = Sizzle.isXML = function( elem ) {
+	var namespace = elem && elem.namespaceURI,
+		docElem = elem && ( elem.ownerDocument || elem ).documentElement;
+
+	// Support: IE <=8
+	// Assume HTML when documentElement doesn't yet exist, such as inside loading iframes
+	// https://bugs.jquery.com/ticket/4833
+	return !rhtml.test( namespace || docElem && docElem.nodeName || "HTML" );
+};
+
+/**
+ * Sets document-related variables once based on the current document
+ * @param {Element|Object} [doc] An element or document object to use to set the document
+ * @returns {Object} Returns the current document
+ */
+setDocument = Sizzle.setDocument = function( node ) {
+	var hasCompare, subWindow,
+		doc = node ? node.ownerDocument || node : preferredDoc;
+
+	// Return early if doc is invalid or already selected
+	// Support: IE 11+, Edge 17 - 18+
+	// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+	// two documents; shallow comparisons work.
+	// eslint-disable-next-line eqeqeq
+	if ( doc == document || doc.nodeType !== 9 || !doc.documentElement ) {
+		return document;
+	}
+
+	// Update global variables
+	document = doc;
+	docElem = document.documentElement;
+	documentIsHTML = !isXML( document );
+
+	// Support: IE 9 - 11+, Edge 12 - 18+
+	// Accessing iframe documents after unload throws "permission denied" errors (jQuery #13936)
+	// Support: IE 11+, Edge 17 - 18+
+	// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+	// two documents; shallow comparisons work.
+	// eslint-disable-next-line eqeqeq
+	if ( preferredDoc != document &&
+		( subWindow = document.defaultView ) && subWindow.top !== subWindow ) {
+
+		// Support: IE 11, Edge
+		if ( subWindow.addEventListener ) {
+			subWindow.addEventListener( "unload", unloadHandler, false );
+
+		// Support: IE 9 - 10 only
+		} else if ( subWindow.attachEvent ) {
+			subWindow.attachEvent( "onunload", unloadHandler );
+		}
+	}
+
+	// Support: IE 8 - 11+, Edge 12 - 18+, Chrome <=16 - 25 only, Firefox <=3.6 - 31 only,
+	// Safari 4 - 5 only, Opera <=11.6 - 12.x only
+	// IE/Edge & older browsers don't support the :scope pseudo-class.
+	// Support: Safari 6.0 only
+	// Safari 6.0 supports :scope but it's an alias of :root there.
+	support.scope = assert( function( el ) {
+		docElem.appendChild( el ).appendChild( document.createElement( "div" ) );
+		return typeof el.querySelectorAll !== "undefined" &&
+			!el.querySelectorAll( ":scope fieldset div" ).length;
+	} );
+
+	/* Attributes
+	---------------------------------------------------------------------- */
+
+	// Support: IE<8
+	// Verify that getAttribute really returns attributes and not properties
+	// (excepting IE8 booleans)
+	support.attributes = assert( function( el ) {
+		el.className = "i";
+		return !el.getAttribute( "className" );
+	} );
+
+	/* getElement(s)By*
+	---------------------------------------------------------------------- */
+
+	// Check if getElementsByTagName("*") returns only elements
+	support.getElementsByTagName = assert( function( el ) {
+		el.appendChild( document.createComment( "" ) );
+		return !el.getElementsByTagName( "*" ).length;
+	} );
+
+	// Support: IE<9
+	support.getElementsByClassName = rnative.test( document.getElementsByClassName );
+
+	// Support: IE<10
+	// Check if getElementById returns elements by name
+	// The broken getElementById methods don't pick up programmatically-set names,
+	// so use a roundabout getElementsByName test
+	support.getById = assert( function( el ) {
+		docElem.appendChild( el ).id = expando;
+		return !document.getElementsByName || !document.getElementsByName( expando ).length;
+	} );
+
+	// ID filter and find
+	if ( support.getById ) {
+		Expr.filter[ "ID" ] = function( id ) {
+			var attrId = id.replace( runescape, funescape );
+			return function( elem ) {
+				return elem.getAttribute( "id" ) === attrId;
+			};
+		};
+		Expr.find[ "ID" ] = function( id, context ) {
+			if ( typeof context.getElementById !== "undefined" && documentIsHTML ) {
+				var elem = context.getElementById( id );
+				return elem ? [ elem ] : [];
+			}
+		};
+	} else {
+		Expr.filter[ "ID" ] =  function( id ) {
+			var attrId = id.replace( runescape, funescape );
+			return function( elem ) {
+				var node = typeof elem.getAttributeNode !== "undefined" &&
+					elem.getAttributeNode( "id" );
+				return node && node.value === attrId;
+			};
+		};
+
+		// Support: IE 6 - 7 only
+		// getElementById is not reliable as a find shortcut
+		Expr.find[ "ID" ] = function( id, context ) {
+			if ( typeof context.getElementById !== "undefined" && documentIsHTML ) {
+				var node, i, elems,
+					elem = context.getElementById( id );
+
+				if ( elem ) {
+
+					// Verify the id attribute
+					node = elem.getAttributeNode( "id" );
+					if ( node && node.value === id ) {
+						return [ elem ];
+					}
+
+					// Fall back on getElementsByName
+					elems = context.getElementsByName( id );
+					i = 0;
+					while ( ( elem = elems[ i++ ] ) ) {
+						node = elem.getAttributeNode( "id" );
+						if ( node && node.value === id ) {
+							return [ elem ];
+						}
+					}
+				}
+
+				return [];
+			}
+		};
+	}
+
+	// Tag
+	Expr.find[ "TAG" ] = support.getElementsByTagName ?
+		function( tag, context ) {
+			if ( typeof context.getElementsByTagName !== "undefined" ) {
+				return context.getElementsByTagName( tag );
+
+			// DocumentFragment nodes don't have gEBTN
+			} else if ( support.qsa ) {
+				return context.querySelectorAll( tag );
+			}
+		} :
+
+		function( tag, context ) {
+			var elem,
+				tmp = [],
+				i = 0,
+
+				// By happy coincidence, a (broken) gEBTN appears on DocumentFragment nodes too
+				results = context.getElementsByTagName( tag );
+
+			// Filter out possible comments
+			if ( tag === "*" ) {
+				while ( ( elem = results[ i++ ] ) ) {
+					if ( elem.nodeType === 1 ) {
+						tmp.push( elem );
+					}
+				}
+
+				return tmp;
+			}
+			return results;
+		};
+
+	// Class
+	Expr.find[ "CLASS" ] = support.getElementsByClassName && function( className, context ) {
+		if ( typeof context.getElementsByClassName !== "undefined" && documentIsHTML ) {
+			return context.getElementsByClassName( className );
+		}
+	};
+
+	/* QSA/matchesSelector
+	---------------------------------------------------------------------- */
+
+	// QSA and matchesSelector support
+
+	// matchesSelector(:active) reports false when true (IE9/Opera 11.5)
+	rbuggyMatches = [];
+
+	// qSa(:focus) reports false when true (Chrome 21)
+	// We allow this because of a bug in IE8/9 that throws an error
+	// whenever `document.activeElement` is accessed on an iframe
+	// So, we allow :focus to pass through QSA all the time to avoid the IE error
+	// See https://bugs.jquery.com/ticket/13378
+	rbuggyQSA = [];
+
+	if ( ( support.qsa = rnative.test( document.querySelectorAll ) ) ) {
+
+		// Build QSA regex
+		// Regex strategy adopted from Diego Perini
+		assert( function( el ) {
+
+			var input;
+
+			// Select is set to empty string on purpose
+			// This is to test IE's treatment of not explicitly
+			// setting a boolean content attribute,
+			// since its presence should be enough
+			// https://bugs.jquery.com/ticket/12359
+			docElem.appendChild( el ).innerHTML = "<a id='" + expando + "'></a>" +
+				"<select id='" + expando + "-\r\\' msallowcapture=''>" +
+				"<option selected=''></option></select>";
+
+			// Support: IE8, Opera 11-12.16
+			// Nothing should be selected when empty strings follow ^= or $= or *=
+			// The test attribute must be unknown in Opera but "safe" for WinRT
+			// https://msdn.microsoft.com/en-us/library/ie/hh465388.aspx#attribute_section
+			if ( el.querySelectorAll( "[msallowcapture^='']" ).length ) {
+				rbuggyQSA.push( "[*^$]=" + whitespace + "*(?:''|\"\")" );
+			}
+
+			// Support: IE8
+			// Boolean attributes and "value" are not treated correctly
+			if ( !el.querySelectorAll( "[selected]" ).length ) {
+				rbuggyQSA.push( "\\[" + whitespace + "*(?:value|" + booleans + ")" );
+			}
+
+			// Support: Chrome<29, Android<4.4, Safari<7.0+, iOS<7.0+, PhantomJS<1.9.8+
+			if ( !el.querySelectorAll( "[id~=" + expando + "-]" ).length ) {
+				rbuggyQSA.push( "~=" );
+			}
+
+			// Support: IE 11+, Edge 15 - 18+
+			// IE 11/Edge don't find elements on a `[name='']` query in some cases.
+			// Adding a temporary attribute to the document before the selection works
+			// around the issue.
+			// Interestingly, IE 10 & older don't seem to have the issue.
+			input = document.createElement( "input" );
+			input.setAttribute( "name", "" );
+			el.appendChild( input );
+			if ( !el.querySelectorAll( "[name='']" ).length ) {
+				rbuggyQSA.push( "\\[" + whitespace + "*name" + whitespace + "*=" +
+					whitespace + "*(?:''|\"\")" );
+			}
+
+			// Webkit/Opera - :checked should return selected option elements
+			// http://www.w3.org/TR/2011/REC-css3-selectors-20110929/#checked
+			// IE8 throws error here and will not see later tests
+			if ( !el.querySelectorAll( ":checked" ).length ) {
+				rbuggyQSA.push( ":checked" );
+			}
+
+			// Support: Safari 8+, iOS 8+
+			// https://bugs.webkit.org/show_bug.cgi?id=136851
+			// In-page `selector#id sibling-combinator selector` fails
+			if ( !el.querySelectorAll( "a#" + expando + "+*" ).length ) {
+				rbuggyQSA.push( ".#.+[+~]" );
+			}
+
+			// Support: Firefox <=3.6 - 5 only
+			// Old Firefox doesn't throw on a badly-escaped identifier.
+			el.querySelectorAll( "\\\f" );
+			rbuggyQSA.push( "[\\r\\n\\f]" );
+		} );
+
+		assert( function( el ) {
+			el.innerHTML = "<a href='' disabled='disabled'></a>" +
+				"<select disabled='disabled'><option/></select>";
+
+			// Support: Windows 8 Native Apps
+			// The type and name attributes are restricted during .innerHTML assignment
+			var input = document.createElement( "input" );
+			input.setAttribute( "type", "hidden" );
+			el.appendChild( input ).setAttribute( "name", "D" );
+
+			// Support: IE8
+			// Enforce case-sensitivity of name attribute
+			if ( el.querySelectorAll( "[name=d]" ).length ) {
+				rbuggyQSA.push( "name" + whitespace + "*[*^$|!~]?=" );
+			}
+
+			// FF 3.5 - :enabled/:disabled and hidden elements (hidden elements are still enabled)
+			// IE8 throws error here and will not see later tests
+			if ( el.querySelectorAll( ":enabled" ).length !== 2 ) {
+				rbuggyQSA.push( ":enabled", ":disabled" );
+			}
+
+			// Support: IE9-11+
+			// IE's :disabled selector does not pick up the children of disabled fieldsets
+			docElem.appendChild( el ).disabled = true;
+			if ( el.querySelectorAll( ":disabled" ).length !== 2 ) {
+				rbuggyQSA.push( ":enabled", ":disabled" );
+			}
+
+			// Support: Opera 10 - 11 only
+			// Opera 10-11 does not throw on post-comma invalid pseudos
+			el.querySelectorAll( "*,:x" );
+			rbuggyQSA.push( ",.*:" );
+		} );
+	}
+
+	if ( ( support.matchesSelector = rnative.test( ( matches = docElem.matches ||
+		docElem.webkitMatchesSelector ||
+		docElem.mozMatchesSelector ||
+		docElem.oMatchesSelector ||
+		docElem.msMatchesSelector ) ) ) ) {
+
+		assert( function( el ) {
+
+			// Check to see if it's possible to do matchesSelector
+			// on a disconnected node (IE 9)
+			support.disconnectedMatch = matches.call( el, "*" );
+
+			// This should fail with an exception
+			// Gecko does not error, returns false instead
+			matches.call( el, "[s!='']:x" );
+			rbuggyMatches.push( "!=", pseudos );
+		} );
+	}
+
+	rbuggyQSA = rbuggyQSA.length && new RegExp( rbuggyQSA.join( "|" ) );
+	rbuggyMatches = rbuggyMatches.length && new RegExp( rbuggyMatches.join( "|" ) );
+
+	/* Contains
+	---------------------------------------------------------------------- */
+	hasCompare = rnative.test( docElem.compareDocumentPosition );
+
+	// Element contains another
+	// Purposefully self-exclusive
+	// As in, an element does not contain itself
+	contains = hasCompare || rnative.test( docElem.contains ) ?
+		function( a, b ) {
+			var adown = a.nodeType === 9 ? a.documentElement : a,
+				bup = b && b.parentNode;
+			return a === bup || !!( bup && bup.nodeType === 1 && (
+				adown.contains ?
+					adown.contains( bup ) :
+					a.compareDocumentPosition && a.compareDocumentPosition( bup ) & 16
+			) );
+		} :
+		function( a, b ) {
+			if ( b ) {
+				while ( ( b = b.parentNode ) ) {
+					if ( b === a ) {
+						return true;
+					}
+				}
+			}
+			return false;
+		};
+
+	/* Sorting
+	---------------------------------------------------------------------- */
+
+	// Document order sorting
+	sortOrder = hasCompare ?
+	function( a, b ) {
+
+		// Flag for duplicate removal
+		if ( a === b ) {
+			hasDuplicate = true;
+			return 0;
+		}
+
+		// Sort on method existence if only one input has compareDocumentPosition
+		var compare = !a.compareDocumentPosition - !b.compareDocumentPosition;
+		if ( compare ) {
+			return compare;
+		}
+
+		// Calculate position if both inputs belong to the same document
+		// Support: IE 11+, Edge 17 - 18+
+		// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+		// two documents; shallow comparisons work.
+		// eslint-disable-next-line eqeqeq
+		compare = ( a.ownerDocument || a ) == ( b.ownerDocument || b ) ?
+			a.compareDocumentPosition( b ) :
+
+			// Otherwise we know they are disconnected
+			1;
+
+		// Disconnected nodes
+		if ( compare & 1 ||
+			( !support.sortDetached && b.compareDocumentPosition( a ) === compare ) ) {
+
+			// Choose the first element that is related to our preferred document
+			// Support: IE 11+, Edge 17 - 18+
+			// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+			// two documents; shallow comparisons work.
+			// eslint-disable-next-line eqeqeq
+			if ( a == document || a.ownerDocument == preferredDoc &&
+				contains( preferredDoc, a ) ) {
+				return -1;
+			}
+
+			// Support: IE 11+, Edge 17 - 18+
+			// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+			// two documents; shallow comparisons work.
+			// eslint-disable-next-line eqeqeq
+			if ( b == document || b.ownerDocument == preferredDoc &&
+				contains( preferredDoc, b ) ) {
+				return 1;
+			}
+
+			// Maintain original order
+			return sortInput ?
+				( indexOf( sortInput, a ) - indexOf( sortInput, b ) ) :
+				0;
+		}
+
+		return compare & 4 ? -1 : 1;
+	} :
+	function( a, b ) {
+
+		// Exit early if the nodes are identical
+		if ( a === b ) {
+			hasDuplicate = true;
+			return 0;
+		}
+
+		var cur,
+			i = 0,
+			aup = a.parentNode,
+			bup = b.parentNode,
+			ap = [ a ],
+			bp = [ b ];
+
+		// Parentless nodes are either documents or disconnected
+		if ( !aup || !bup ) {
+
+			// Support: IE 11+, Edge 17 - 18+
+			// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+			// two documents; shallow comparisons work.
+			/* eslint-disable eqeqeq */
+			return a == document ? -1 :
+				b == document ? 1 :
+				/* eslint-enable eqeqeq */
+				aup ? -1 :
+				bup ? 1 :
+				sortInput ?
+				( indexOf( sortInput, a ) - indexOf( sortInput, b ) ) :
+				0;
+
+		// If the nodes are siblings, we can do a quick check
+		} else if ( aup === bup ) {
+			return siblingCheck( a, b );
+		}
+
+		// Otherwise we need full lists of their ancestors for comparison
+		cur = a;
+		while ( ( cur = cur.parentNode ) ) {
+			ap.unshift( cur );
+		}
+		cur = b;
+		while ( ( cur = cur.parentNode ) ) {
+			bp.unshift( cur );
+		}
+
+		// Walk down the tree looking for a discrepancy
+		while ( ap[ i ] === bp[ i ] ) {
+			i++;
+		}
+
+		return i ?
+
+			// Do a sibling check if the nodes have a common ancestor
+			siblingCheck( ap[ i ], bp[ i ] ) :
+
+			// Otherwise nodes in our document sort first
+			// Support: IE 11+, Edge 17 - 18+
+			// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+			// two documents; shallow comparisons work.
+			/* eslint-disable eqeqeq */
+			ap[ i ] == preferredDoc ? -1 :
+			bp[ i ] == preferredDoc ? 1 :
+			/* eslint-enable eqeqeq */
+			0;
+	};
+
+	return document;
+};
+
+Sizzle.matches = function( expr, elements ) {
+	return Sizzle( expr, null, null, elements );
+};
+
+Sizzle.matchesSelector = function( elem, expr ) {
+	setDocument( elem );
+
+	if ( support.matchesSelector && documentIsHTML &&
+		!nonnativeSelectorCache[ expr + " " ] &&
+		( !rbuggyMatches || !rbuggyMatches.test( expr ) ) &&
+		( !rbuggyQSA     || !rbuggyQSA.test( expr ) ) ) {
+
+		try {
+			var ret = matches.call( elem, expr );
+
+			// IE 9's matchesSelector returns false on disconnected nodes
+			if ( ret || support.disconnectedMatch ||
+
+				// As well, disconnected nodes are said to be in a document
+				// fragment in IE 9
+				elem.document && elem.document.nodeType !== 11 ) {
+				return ret;
+			}
+		} catch ( e ) {
+			nonnativeSelectorCache( expr, true );
+		}
+	}
+
+	return Sizzle( expr, document, null, [ elem ] ).length > 0;
+};
+
+Sizzle.contains = function( context, elem ) {
+
+	// Set document vars if needed
+	// Support: IE 11+, Edge 17 - 18+
+	// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+	// two documents; shallow comparisons work.
+	// eslint-disable-next-line eqeqeq
+	if ( ( context.ownerDocument || context ) != document ) {
+		setDocument( context );
+	}
+	return contains( context, elem );
+};
+
+Sizzle.attr = function( elem, name ) {
+
+	// Set document vars if needed
+	// Support: IE 11+, Edge 17 - 18+
+	// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+	// two documents; shallow comparisons work.
+	// eslint-disable-next-line eqeqeq
+	if ( ( elem.ownerDocument || elem ) != document ) {
+		setDocument( elem );
+	}
+
+	var fn = Expr.attrHandle[ name.toLowerCase() ],
+
+		// Don't get fooled by Object.prototype properties (jQuery #13807)
+		val = fn && hasOwn.call( Expr.attrHandle, name.toLowerCase() ) ?
+			fn( elem, name, !documentIsHTML ) :
+			undefined;
+
+	return val !== undefined ?
+		val :
+		support.attributes || !documentIsHTML ?
+			elem.getAttribute( name ) :
+			( val = elem.getAttributeNode( name ) ) && val.specified ?
+				val.value :
+				null;
+};
+
+Sizzle.escape = function( sel ) {
+	return ( sel + "" ).replace( rcssescape, fcssescape );
+};
+
+Sizzle.error = function( msg ) {
+	throw new Error( "Syntax error, unrecognized expression: " + msg );
+};
+
+/**
+ * Document sorting and removing duplicates
+ * @param {ArrayLike} results
+ */
+Sizzle.uniqueSort = function( results ) {
+	var elem,
+		duplicates = [],
+		j = 0,
+		i = 0;
+
+	// Unless we *know* we can detect duplicates, assume their presence
+	hasDuplicate = !support.detectDuplicates;
+	sortInput = !support.sortStable && results.slice( 0 );
+	results.sort( sortOrder );
+
+	if ( hasDuplicate ) {
+		while ( ( elem = results[ i++ ] ) ) {
+			if ( elem === results[ i ] ) {
+				j = duplicates.push( i );
+			}
+		}
+		while ( j-- ) {
+			results.splice( duplicates[ j ], 1 );
+		}
+	}
+
+	// Clear input after sorting to release objects
+	// See https://github.com/jquery/sizzle/pull/225
+	sortInput = null;
+
+	return results;
+};
+
+/**
+ * Utility function for retrieving the text value of an array of DOM nodes
+ * @param {Array|Element} elem
+ */
+getText = Sizzle.getText = function( elem ) {
+	var node,
+		ret = "",
+		i = 0,
+		nodeType = elem.nodeType;
+
+	if ( !nodeType ) {
+
+		// If no nodeType, this is expected to be an array
+		while ( ( node = elem[ i++ ] ) ) {
+
+			// Do not traverse comment nodes
+			ret += getText( node );
+		}
+	} else if ( nodeType === 1 || nodeType === 9 || nodeType === 11 ) {
+
+		// Use textContent for elements
+		// innerText usage removed for consistency of new lines (jQuery #11153)
+		if ( typeof elem.textContent === "string" ) {
+			return elem.textContent;
+		} else {
+
+			// Traverse its children
+			for ( elem = elem.firstChild; elem; elem = elem.nextSibling ) {
+				ret += getText( elem );
+			}
+		}
+	} else if ( nodeType === 3 || nodeType === 4 ) {
+		return elem.nodeValue;
+	}
+
+	// Do not include comment or processing instruction nodes
+
+	return ret;
+};
+
+Expr = Sizzle.selectors = {
+
+	// Can be adjusted by the user
+	cacheLength: 50,
+
+	createPseudo: markFunction,
+
+	match: matchExpr,
+
+	attrHandle: {},
+
+	find: {},
+
+	relative: {
+		">": { dir: "parentNode", first: true },
+		" ": { dir: "parentNode" },
+		"+": { dir: "previousSibling", first: true },
+		"~": { dir: "previousSibling" }
+	},
+
+	preFilter: {
+		"ATTR": function( match ) {
+			match[ 1 ] = match[ 1 ].replace( runescape, funescape );
+
+			// Move the given value to match[3] whether quoted or unquoted
+			match[ 3 ] = ( match[ 3 ] || match[ 4 ] ||
+				match[ 5 ] || "" ).replace( runescape, funescape );
+
+			if ( match[ 2 ] === "~=" ) {
+				match[ 3 ] = " " + match[ 3 ] + " ";
+			}
+
+			return match.slice( 0, 4 );
+		},
+
+		"CHILD": function( match ) {
+
+			/* matches from matchExpr["CHILD"]
+				1 type (only|nth|...)
+				2 what (child|of-type)
+				3 argument (even|odd|\d*|\d*n([+-]\d+)?|...)
+				4 xn-component of xn+y argument ([+-]?\d*n|)
+				5 sign of xn-component
+				6 x of xn-component
+				7 sign of y-component
+				8 y of y-component
+			*/
+			match[ 1 ] = match[ 1 ].toLowerCase();
+
+			if ( match[ 1 ].slice( 0, 3 ) === "nth" ) {
+
+				// nth-* requires argument
+				if ( !match[ 3 ] ) {
+					Sizzle.error( match[ 0 ] );
+				}
+
+				// numeric x and y parameters for Expr.filter.CHILD
+				// remember that false/true cast respectively to 0/1
+				match[ 4 ] = +( match[ 4 ] ?
+					match[ 5 ] + ( match[ 6 ] || 1 ) :
+					2 * ( match[ 3 ] === "even" || match[ 3 ] === "odd" ) );
+				match[ 5 ] = +( ( match[ 7 ] + match[ 8 ] ) || match[ 3 ] === "odd" );
+
+				// other types prohibit arguments
+			} else if ( match[ 3 ] ) {
+				Sizzle.error( match[ 0 ] );
+			}
+
+			return match;
+		},
+
+		"PSEUDO": function( match ) {
+			var excess,
+				unquoted = !match[ 6 ] && match[ 2 ];
+
+			if ( matchExpr[ "CHILD" ].test( match[ 0 ] ) ) {
+				return null;
+			}
+
+			// Accept quoted arguments as-is
+			if ( match[ 3 ] ) {
+				match[ 2 ] = match[ 4 ] || match[ 5 ] || "";
+
+			// Strip excess characters from unquoted arguments
+			} else if ( unquoted && rpseudo.test( unquoted ) &&
+
+				// Get excess from tokenize (recursively)
+				( excess = tokenize( unquoted, true ) ) &&
+
+				// advance to the next closing parenthesis
+				( excess = unquoted.indexOf( ")", unquoted.length - excess ) - unquoted.length ) ) {
+
+				// excess is a negative index
+				match[ 0 ] = match[ 0 ].slice( 0, excess );
+				match[ 2 ] = unquoted.slice( 0, excess );
+			}
+
+			// Return only captures needed by the pseudo filter method (type and argument)
+			return match.slice( 0, 3 );
+		}
+	},
+
+	filter: {
+
+		"TAG": function( nodeNameSelector ) {
+			var nodeName = nodeNameSelector.replace( runescape, funescape ).toLowerCase();
+			return nodeNameSelector === "*" ?
+				function() {
+					return true;
+				} :
+				function( elem ) {
+					return elem.nodeName && elem.nodeName.toLowerCase() === nodeName;
+				};
+		},
+
+		"CLASS": function( className ) {
+			var pattern = classCache[ className + " " ];
+
+			return pattern ||
+				( pattern = new RegExp( "(^|" + whitespace +
+					")" + className + "(" + whitespace + "|$)" ) ) && classCache(
+						className, function( elem ) {
+							return pattern.test(
+								typeof elem.className === "string" && elem.className ||
+								typeof elem.getAttribute !== "undefined" &&
+									elem.getAttribute( "class" ) ||
+								""
+							);
+				} );
+		},
+
+		"ATTR": function( name, operator, check ) {
+			return function( elem ) {
+				var result = Sizzle.attr( elem, name );
+
+				if ( result == null ) {
+					return operator === "!=";
+				}
+				if ( !operator ) {
+					return true;
+				}
+
+				result += "";
+
+				/* eslint-disable max-len */
+
+				return operator === "=" ? result === check :
+					operator === "!=" ? result !== check :
+					operator === "^=" ? check && result.indexOf( check ) === 0 :
+					operator === "*=" ? check && result.indexOf( check ) > -1 :
+					operator === "$=" ? check && result.slice( -check.length ) === check :
+					operator === "~=" ? ( " " + result.replace( rwhitespace, " " ) + " " ).indexOf( check ) > -1 :
+					operator === "|=" ? result === check || result.slice( 0, check.length + 1 ) === check + "-" :
+					false;
+				/* eslint-enable max-len */
+
+			};
+		},
+
+		"CHILD": function( type, what, _argument, first, last ) {
+			var simple = type.slice( 0, 3 ) !== "nth",
+				forward = type.slice( -4 ) !== "last",
+				ofType = what === "of-type";
+
+			return first === 1 && last === 0 ?
+
+				// Shortcut for :nth-*(n)
+				function( elem ) {
+					return !!elem.parentNode;
+				} :
+
+				function( elem, _context, xml ) {
+					var cache, uniqueCache, outerCache, node, nodeIndex, start,
+						dir = simple !== forward ? "nextSibling" : "previousSibling",
+						parent = elem.parentNode,
+						name = ofType && elem.nodeName.toLowerCase(),
+						useCache = !xml && !ofType,
+						diff = false;
+
+					if ( parent ) {
+
+						// :(first|last|only)-(child|of-type)
+						if ( simple ) {
+							while ( dir ) {
+								node = elem;
+								while ( ( node = node[ dir ] ) ) {
+									if ( ofType ?
+										node.nodeName.toLowerCase() === name :
+										node.nodeType === 1 ) {
+
+										return false;
+									}
+								}
+
+								// Reverse direction for :only-* (if we haven't yet done so)
+								start = dir = type === "only" && !start && "nextSibling";
+							}
+							return true;
+						}
+
+						start = [ forward ? parent.firstChild : parent.lastChild ];
+
+						// non-xml :nth-child(...) stores cache data on `parent`
+						if ( forward && useCache ) {
+
+							// Seek `elem` from a previously-cached index
+
+							// ...in a gzip-friendly way
+							node = parent;
+							outerCache = node[ expando ] || ( node[ expando ] = {} );
+
+							// Support: IE <9 only
+							// Defend against cloned attroperties (jQuery gh-1709)
+							uniqueCache = outerCache[ node.uniqueID ] ||
+								( outerCache[ node.uniqueID ] = {} );
+
+							cache = uniqueCache[ type ] || [];
+							nodeIndex = cache[ 0 ] === dirruns && cache[ 1 ];
+							diff = nodeIndex && cache[ 2 ];
+							node = nodeIndex && parent.childNodes[ nodeIndex ];
+
+							while ( ( node = ++nodeIndex && node && node[ dir ] ||
+
+								// Fallback to seeking `elem` from the start
+								( diff = nodeIndex = 0 ) || start.pop() ) ) {
+
+								// When found, cache indexes on `parent` and break
+								if ( node.nodeType === 1 && ++diff && node === elem ) {
+									uniqueCache[ type ] = [ dirruns, nodeIndex, diff ];
+									break;
+								}
+							}
+
+						} else {
+
+							// Use previously-cached element index if available
+							if ( useCache ) {
+
+								// ...in a gzip-friendly way
+								node = elem;
+								outerCache = node[ expando ] || ( node[ expando ] = {} );
+
+								// Support: IE <9 only
+								// Defend against cloned attroperties (jQuery gh-1709)
+								uniqueCache = outerCache[ node.uniqueID ] ||
+									( outerCache[ node.uniqueID ] = {} );
+
+								cache = uniqueCache[ type ] || [];
+								nodeIndex = cache[ 0 ] === dirruns && cache[ 1 ];
+								diff = nodeIndex;
+							}
+
+							// xml :nth-child(...)
+							// or :nth-last-child(...) or :nth(-last)?-of-type(...)
+							if ( diff === false ) {
+
+								// Use the same loop as above to seek `elem` from the start
+								while ( ( node = ++nodeIndex && node && node[ dir ] ||
+									( diff = nodeIndex = 0 ) || start.pop() ) ) {
+
+									if ( ( ofType ?
+										node.nodeName.toLowerCase() === name :
+										node.nodeType === 1 ) &&
+										++diff ) {
+
+										// Cache the index of each encountered element
+										if ( useCache ) {
+											outerCache = node[ expando ] ||
+												( node[ expando ] = {} );
+
+											// Support: IE <9 only
+											// Defend against cloned attroperties (jQuery gh-1709)
+											uniqueCache = outerCache[ node.uniqueID ] ||
+												( outerCache[ node.uniqueID ] = {} );
+
+											uniqueCache[ type ] = [ dirruns, diff ];
+										}
+
+										if ( node === elem ) {
+											break;
+										}
+									}
+								}
+							}
+						}
+
+						// Incorporate the offset, then check against cycle size
+						diff -= last;
+						return diff === first || ( diff % first === 0 && diff / first >= 0 );
+					}
+				};
+		},
+
+		"PSEUDO": function( pseudo, argument ) {
+
+			// pseudo-class names are case-insensitive
+			// http://www.w3.org/TR/selectors/#pseudo-classes
+			// Prioritize by case sensitivity in case custom pseudos are added with uppercase letters
+			// Remember that setFilters inherits from pseudos
+			var args,
+				fn = Expr.pseudos[ pseudo ] || Expr.setFilters[ pseudo.toLowerCase() ] ||
+					Sizzle.error( "unsupported pseudo: " + pseudo );
+
+			// The user may use createPseudo to indicate that
+			// arguments are needed to create the filter function
+			// just as Sizzle does
+			if ( fn[ expando ] ) {
+				return fn( argument );
+			}
+
+			// But maintain support for old signatures
+			if ( fn.length > 1 ) {
+				args = [ pseudo, pseudo, "", argument ];
+				return Expr.setFilters.hasOwnProperty( pseudo.toLowerCase() ) ?
+					markFunction( function( seed, matches ) {
+						var idx,
+							matched = fn( seed, argument ),
+							i = matched.length;
+						while ( i-- ) {
+							idx = indexOf( seed, matched[ i ] );
+							seed[ idx ] = !( matches[ idx ] = matched[ i ] );
+						}
+					} ) :
+					function( elem ) {
+						return fn( elem, 0, args );
+					};
+			}
+
+			return fn;
+		}
+	},
+
+	pseudos: {
+
+		// Potentially complex pseudos
+		"not": markFunction( function( selector ) {
+
+			// Trim the selector passed to compile
+			// to avoid treating leading and trailing
+			// spaces as combinators
+			var input = [],
+				results = [],
+				matcher = compile( selector.replace( rtrim, "$1" ) );
+
+			return matcher[ expando ] ?
+				markFunction( function( seed, matches, _context, xml ) {
+					var elem,
+						unmatched = matcher( seed, null, xml, [] ),
+						i = seed.length;
+
+					// Match elements unmatched by `matcher`
+					while ( i-- ) {
+						if ( ( elem = unmatched[ i ] ) ) {
+							seed[ i ] = !( matches[ i ] = elem );
+						}
+					}
+				} ) :
+				function( elem, _context, xml ) {
+					input[ 0 ] = elem;
+					matcher( input, null, xml, results );
+
+					// Don't keep the element (issue #299)
+					input[ 0 ] = null;
+					return !results.pop();
+				};
+		} ),
+
+		"has": markFunction( function( selector ) {
+			return function( elem ) {
+				return Sizzle( selector, elem ).length > 0;
+			};
+		} ),
+
+		"contains": markFunction( function( text ) {
+			text = text.replace( runescape, funescape );
+			return function( elem ) {
+				return ( elem.textContent || getText( elem ) ).indexOf( text ) > -1;
+			};
+		} ),
+
+		// "Whether an element is represented by a :lang() selector
+		// is based solely on the element's language value
+		// being equal to the identifier C,
+		// or beginning with the identifier C immediately followed by "-".
+		// The matching of C against the element's language value is performed case-insensitively.
+		// The identifier C does not have to be a valid language name."
+		// http://www.w3.org/TR/selectors/#lang-pseudo
+		"lang": markFunction( function( lang ) {
+
+			// lang value must be a valid identifier
+			if ( !ridentifier.test( lang || "" ) ) {
+				Sizzle.error( "unsupported lang: " + lang );
+			}
+			lang = lang.replace( runescape, funescape ).toLowerCase();
+			return function( elem ) {
+				var elemLang;
+				do {
+					if ( ( elemLang = documentIsHTML ?
+						elem.lang :
+						elem.getAttribute( "xml:lang" ) || elem.getAttribute( "lang" ) ) ) {
+
+						elemLang = elemLang.toLowerCase();
+						return elemLang === lang || elemLang.indexOf( lang + "-" ) === 0;
+					}
+				} while ( ( elem = elem.parentNode ) && elem.nodeType === 1 );
+				return false;
+			};
+		} ),
+
+		// Miscellaneous
+		"target": function( elem ) {
+			var hash = window.location && window.location.hash;
+			return hash && hash.slice( 1 ) === elem.id;
+		},
+
+		"root": function( elem ) {
+			return elem === docElem;
+		},
+
+		"focus": function( elem ) {
+			return elem === document.activeElement &&
+				( !document.hasFocus || document.hasFocus() ) &&
+				!!( elem.type || elem.href || ~elem.tabIndex );
+		},
+
+		// Boolean properties
+		"enabled": createDisabledPseudo( false ),
+		"disabled": createDisabledPseudo( true ),
+
+		"checked": function( elem ) {
+
+			// In CSS3, :checked should return both checked and selected elements
+			// http://www.w3.org/TR/2011/REC-css3-selectors-20110929/#checked
+			var nodeName = elem.nodeName.toLowerCase();
+			return ( nodeName === "input" && !!elem.checked ) ||
+				( nodeName === "option" && !!elem.selected );
+		},
+
+		"selected": function( elem ) {
+
+			// Accessing this property makes selected-by-default
+			// options in Safari work properly
+			if ( elem.parentNode ) {
+				// eslint-disable-next-line no-unused-expressions
+				elem.parentNode.selectedIndex;
+			}
+
+			return elem.selected === true;
+		},
+
+		// Contents
+		"empty": function( elem ) {
+
+			// http://www.w3.org/TR/selectors/#empty-pseudo
+			// :empty is negated by element (1) or content nodes (text: 3; cdata: 4; entity ref: 5),
+			//   but not by others (comment: 8; processing instruction: 7; etc.)
+			// nodeType < 6 works because attributes (2) do not appear as children
+			for ( elem = elem.firstChild; elem; elem = elem.nextSibling ) {
+				if ( elem.nodeType < 6 ) {
+					return false;
+				}
+			}
+			return true;
+		},
+
+		"parent": function( elem ) {
+			return !Expr.pseudos[ "empty" ]( elem );
+		},
+
+		// Element/input types
+		"header": function( elem ) {
+			return rheader.test( elem.nodeName );
+		},
+
+		"input": function( elem ) {
+			return rinputs.test( elem.nodeName );
+		},
+
+		"button": function( elem ) {
+			var name = elem.nodeName.toLowerCase();
+			return name === "input" && elem.type === "button" || name === "button";
+		},
+
+		"text": function( elem ) {
+			var attr;
+			return elem.nodeName.toLowerCase() === "input" &&
+				elem.type === "text" &&
+
+				// Support: IE<8
+				// New HTML5 attribute values (e.g., "search") appear with elem.type === "text"
+				( ( attr = elem.getAttribute( "type" ) ) == null ||
+					attr.toLowerCase() === "text" );
+		},
+
+		// Position-in-collection
+		"first": createPositionalPseudo( function() {
+			return [ 0 ];
+		} ),
+
+		"last": createPositionalPseudo( function( _matchIndexes, length ) {
+			return [ length - 1 ];
+		} ),
+
+		"eq": createPositionalPseudo( function( _matchIndexes, length, argument ) {
+			return [ argument < 0 ? argument + length : argument ];
+		} ),
+
+		"even": createPositionalPseudo( function( matchIndexes, length ) {
+			var i = 0;
+			for ( ; i < length; i += 2 ) {
+				matchIndexes.push( i );
+			}
+			return matchIndexes;
+		} ),
+
+		"odd": createPositionalPseudo( function( matchIndexes, length ) {
+			var i = 1;
+			for ( ; i < length; i += 2 ) {
+				matchIndexes.push( i );
+			}
+			return matchIndexes;
+		} ),
+
+		"lt": createPositionalPseudo( function( matchIndexes, length, argument ) {
+			var i = argument < 0 ?
+				argument + length :
+				argument > length ?
+					length :
+					argument;
+			for ( ; --i >= 0; ) {
+				matchIndexes.push( i );
+			}
+			return matchIndexes;
+		} ),
+
+		"gt": createPositionalPseudo( function( matchIndexes, length, argument ) {
+			var i = argument < 0 ? argument + length : argument;
+			for ( ; ++i < length; ) {
+				matchIndexes.push( i );
+			}
+			return matchIndexes;
+		} )
+	}
+};
+
+Expr.pseudos[ "nth" ] = Expr.pseudos[ "eq" ];
+
+// Add button/input type pseudos
+for ( i in { radio: true, checkbox: true, file: true, password: true, image: true } ) {
+	Expr.pseudos[ i ] = createInputPseudo( i );
+}
+for ( i in { submit: true, reset: true } ) {
+	Expr.pseudos[ i ] = createButtonPseudo( i );
+}
+
+// Easy API for creating new setFilters
+function setFilters() {}
+setFilters.prototype = Expr.filters = Expr.pseudos;
+Expr.setFilters = new setFilters();
+
+tokenize = Sizzle.tokenize = function( selector, parseOnly ) {
+	var matched, match, tokens, type,
+		soFar, groups, preFilters,
+		cached = tokenCache[ selector + " " ];
+
+	if ( cached ) {
+		return parseOnly ? 0 : cached.slice( 0 );
+	}
+
+	soFar = selector;
+	groups = [];
+	preFilters = Expr.preFilter;
+
+	while ( soFar ) {
+
+		// Comma and first run
+		if ( !matched || ( match = rcomma.exec( soFar ) ) ) {
+			if ( match ) {
+
+				// Don't consume trailing commas as valid
+				soFar = soFar.slice( match[ 0 ].length ) || soFar;
+			}
+			groups.push( ( tokens = [] ) );
+		}
+
+		matched = false;
+
+		// Combinators
+		if ( ( match = rcombinators.exec( soFar ) ) ) {
+			matched = match.shift();
+			tokens.push( {
+				value: matched,
+
+				// Cast descendant combinators to space
+				type: match[ 0 ].replace( rtrim, " " )
+			} );
+			soFar = soFar.slice( matched.length );
+		}
+
+		// Filters
+		for ( type in Expr.filter ) {
+			if ( ( match = matchExpr[ type ].exec( soFar ) ) && ( !preFilters[ type ] ||
+				( match = preFilters[ type ]( match ) ) ) ) {
+				matched = match.shift();
+				tokens.push( {
+					value: matched,
+					type: type,
+					matches: match
+				} );
+				soFar = soFar.slice( matched.length );
+			}
+		}
+
+		if ( !matched ) {
+			break;
+		}
+	}
+
+	// Return the length of the invalid excess
+	// if we're just parsing
+	// Otherwise, throw an error or return tokens
+	return parseOnly ?
+		soFar.length :
+		soFar ?
+			Sizzle.error( selector ) :
+
+			// Cache the tokens
+			tokenCache( selector, groups ).slice( 0 );
+};
+
+function toSelector( tokens ) {
+	var i = 0,
+		len = tokens.length,
+		selector = "";
+	for ( ; i < len; i++ ) {
+		selector += tokens[ i ].value;
+	}
+	return selector;
+}
+
+function addCombinator( matcher, combinator, base ) {
+	var dir = combinator.dir,
+		skip = combinator.next,
+		key = skip || dir,
+		checkNonElements = base && key === "parentNode",
+		doneName = done++;
+
+	return combinator.first ?
+
+		// Check against closest ancestor/preceding element
+		function( elem, context, xml ) {
+			while ( ( elem = elem[ dir ] ) ) {
+				if ( elem.nodeType === 1 || checkNonElements ) {
+					return matcher( elem, context, xml );
+				}
+			}
+			return false;
+		} :
+
+		// Check against all ancestor/preceding elements
+		function( elem, context, xml ) {
+			var oldCache, uniqueCache, outerCache,
+				newCache = [ dirruns, doneName ];
+
+			// We can't set arbitrary data on XML nodes, so they don't benefit from combinator caching
+			if ( xml ) {
+				while ( ( elem = elem[ dir ] ) ) {
+					if ( elem.nodeType === 1 || checkNonElements ) {
+						if ( matcher( elem, context, xml ) ) {
+							return true;
+						}
+					}
+				}
+			} else {
+				while ( ( elem = elem[ dir ] ) ) {
+					if ( elem.nodeType === 1 || checkNonElements ) {
+						outerCache = elem[ expando ] || ( elem[ expando ] = {} );
+
+						// Support: IE <9 only
+						// Defend against cloned attroperties (jQuery gh-1709)
+						uniqueCache = outerCache[ elem.uniqueID ] ||
+							( outerCache[ elem.uniqueID ] = {} );
+
+						if ( skip && skip === elem.nodeName.toLowerCase() ) {
+							elem = elem[ dir ] || elem;
+						} else if ( ( oldCache = uniqueCache[ key ] ) &&
+							oldCache[ 0 ] === dirruns && oldCache[ 1 ] === doneName ) {
+
+							// Assign to newCache so results back-propagate to previous elements
+							return ( newCache[ 2 ] = oldCache[ 2 ] );
+						} else {
+
+							// Reuse newcache so results back-propagate to previous elements
+							uniqueCache[ key ] = newCache;
+
+							// A match means we're done; a fail means we have to keep checking
+							if ( ( newCache[ 2 ] = matcher( elem, context, xml ) ) ) {
+								return true;
+							}
+						}
+					}
+				}
+			}
+			return false;
+		};
+}
+
+function elementMatcher( matchers ) {
+	return matchers.length > 1 ?
+		function( elem, context, xml ) {
+			var i = matchers.length;
+			while ( i-- ) {
+				if ( !matchers[ i ]( elem, context, xml ) ) {
+					return false;
+				}
+			}
+			return true;
+		} :
+		matchers[ 0 ];
+}
+
+function multipleContexts( selector, contexts, results ) {
+	var i = 0,
+		len = contexts.length;
+	for ( ; i < len; i++ ) {
+		Sizzle( selector, contexts[ i ], results );
+	}
+	return results;
+}
+
+function condense( unmatched, map, filter, context, xml ) {
+	var elem,
+		newUnmatched = [],
+		i = 0,
+		len = unmatched.length,
+		mapped = map != null;
+
+	for ( ; i < len; i++ ) {
+		if ( ( elem = unmatched[ i ] ) ) {
+			if ( !filter || filter( elem, context, xml ) ) {
+				newUnmatched.push( elem );
+				if ( mapped ) {
+					map.push( i );
+				}
+			}
+		}
+	}
+
+	return newUnmatched;
+}
+
+function setMatcher( preFilter, selector, matcher, postFilter, postFinder, postSelector ) {
+	if ( postFilter && !postFilter[ expando ] ) {
+		postFilter = setMatcher( postFilter );
+	}
+	if ( postFinder && !postFinder[ expando ] ) {
+		postFinder = setMatcher( postFinder, postSelector );
+	}
+	return markFunction( function( seed, results, context, xml ) {
+		var temp, i, elem,
+			preMap = [],
+			postMap = [],
+			preexisting = results.length,
+
+			// Get initial elements from seed or context
+			elems = seed || multipleContexts(
+				selector || "*",
+				context.nodeType ? [ context ] : context,
+				[]
+			),
+
+			// Prefilter to get matcher input, preserving a map for seed-results synchronization
+			matcherIn = preFilter && ( seed || !selector ) ?
+				condense( elems, preMap, preFilter, context, xml ) :
+				elems,
+
+			matcherOut = matcher ?
+
+				// If we have a postFinder, or filtered seed, or non-seed postFilter or preexisting results,
+				postFinder || ( seed ? preFilter : preexisting || postFilter ) ?
+
+					// ...intermediate processing is necessary
+					[] :
+
+					// ...otherwise use results directly
+					results :
+				matcherIn;
+
+		// Find primary matches
+		if ( matcher ) {
+			matcher( matcherIn, matcherOut, context, xml );
+		}
+
+		// Apply postFilter
+		if ( postFilter ) {
+			temp = condense( matcherOut, postMap );
+			postFilter( temp, [], context, xml );
+
+			// Un-match failing elements by moving them back to matcherIn
+			i = temp.length;
+			while ( i-- ) {
+				if ( ( elem = temp[ i ] ) ) {
+					matcherOut[ postMap[ i ] ] = !( matcherIn[ postMap[ i ] ] = elem );
+				}
+			}
+		}
+
+		if ( seed ) {
+			if ( postFinder || preFilter ) {
+				if ( postFinder ) {
+
+					// Get the final matcherOut by condensing this intermediate into postFinder contexts
+					temp = [];
+					i = matcherOut.length;
+					while ( i-- ) {
+						if ( ( elem = matcherOut[ i ] ) ) {
+
+							// Restore matcherIn since elem is not yet a final match
+							temp.push( ( matcherIn[ i ] = elem ) );
+						}
+					}
+					postFinder( null, ( matcherOut = [] ), temp, xml );
+				}
+
+				// Move matched elements from seed to results to keep them synchronized
+				i = matcherOut.length;
+				while ( i-- ) {
+					if ( ( elem = matcherOut[ i ] ) &&
+						( temp = postFinder ? indexOf( seed, elem ) : preMap[ i ] ) > -1 ) {
+
+						seed[ temp ] = !( results[ temp ] = elem );
+					}
+				}
+			}
+
+		// Add elements to results, through postFinder if defined
+		} else {
+			matcherOut = condense(
+				matcherOut === results ?
+					matcherOut.splice( preexisting, matcherOut.length ) :
+					matcherOut
+			);
+			if ( postFinder ) {
+				postFinder( null, results, matcherOut, xml );
+			} else {
+				push.apply( results, matcherOut );
+			}
+		}
+	} );
+}
+
+function matcherFromTokens( tokens ) {
+	var checkContext, matcher, j,
+		len = tokens.length,
+		leadingRelative = Expr.relative[ tokens[ 0 ].type ],
+		implicitRelative = leadingRelative || Expr.relative[ " " ],
+		i = leadingRelative ? 1 : 0,
+
+		// The foundational matcher ensures that elements are reachable from top-level context(s)
+		matchContext = addCombinator( function( elem ) {
+			return elem === checkContext;
+		}, implicitRelative, true ),
+		matchAnyContext = addCombinator( function( elem ) {
+			return indexOf( checkContext, elem ) > -1;
+		}, implicitRelative, true ),
+		matchers = [ function( elem, context, xml ) {
+			var ret = ( !leadingRelative && ( xml || context !== outermostContext ) ) || (
+				( checkContext = context ).nodeType ?
+					matchContext( elem, context, xml ) :
+					matchAnyContext( elem, context, xml ) );
+
+			// Avoid hanging onto element (issue #299)
+			checkContext = null;
+			return ret;
+		} ];
+
+	for ( ; i < len; i++ ) {
+		if ( ( matcher = Expr.relative[ tokens[ i ].type ] ) ) {
+			matchers = [ addCombinator( elementMatcher( matchers ), matcher ) ];
+		} else {
+			matcher = Expr.filter[ tokens[ i ].type ].apply( null, tokens[ i ].matches );
+
+			// Return special upon seeing a positional matcher
+			if ( matcher[ expando ] ) {
+
+				// Find the next relative operator (if any) for proper handling
+				j = ++i;
+				for ( ; j < len; j++ ) {
+					if ( Expr.relative[ tokens[ j ].type ] ) {
+						break;
+					}
+				}
+				return setMatcher(
+					i > 1 && elementMatcher( matchers ),
+					i > 1 && toSelector(
+
+					// If the preceding token was a descendant combinator, insert an implicit any-element `*`
+					tokens
+						.slice( 0, i - 1 )
+						.concat( { value: tokens[ i - 2 ].type === " " ? "*" : "" } )
+					).replace( rtrim, "$1" ),
+					matcher,
+					i < j && matcherFromTokens( tokens.slice( i, j ) ),
+					j < len && matcherFromTokens( ( tokens = tokens.slice( j ) ) ),
+					j < len && toSelector( tokens )
+				);
+			}
+			matchers.push( matcher );
+		}
+	}
+
+	return elementMatcher( matchers );
+}
+
+function matcherFromGroupMatchers( elementMatchers, setMatchers ) {
+	var bySet = setMatchers.length > 0,
+		byElement = elementMatchers.length > 0,
+		superMatcher = function( seed, context, xml, results, outermost ) {
+			var elem, j, matcher,
+				matchedCount = 0,
+				i = "0",
+				unmatched = seed && [],
+				setMatched = [],
+				contextBackup = outermostContext,
+
+				// We must always have either seed elements or outermost context
+				elems = seed || byElement && Expr.find[ "TAG" ]( "*", outermost ),
+
+				// Use integer dirruns iff this is the outermost matcher
+				dirrunsUnique = ( dirruns += contextBackup == null ? 1 : Math.random() || 0.1 ),
+				len = elems.length;
+
+			if ( outermost ) {
+
+				// Support: IE 11+, Edge 17 - 18+
+				// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+				// two documents; shallow comparisons work.
+				// eslint-disable-next-line eqeqeq
+				outermostContext = context == document || context || outermost;
+			}
+
+			// Add elements passing elementMatchers directly to results
+			// Support: IE<9, Safari
+			// Tolerate NodeList properties (IE: "length"; Safari: <number>) matching elements by id
+			for ( ; i !== len && ( elem = elems[ i ] ) != null; i++ ) {
+				if ( byElement && elem ) {
+					j = 0;
+
+					// Support: IE 11+, Edge 17 - 18+
+					// IE/Edge sometimes throw a "Permission denied" error when strict-comparing
+					// two documents; shallow comparisons work.
+					// eslint-disable-next-line eqeqeq
+					if ( !context && elem.ownerDocument != document ) {
+						setDocument( elem );
+						xml = !documentIsHTML;
+					}
+					while ( ( matcher = elementMatchers[ j++ ] ) ) {
+						if ( matcher( elem, context || document, xml ) ) {
+							results.push( elem );
+							break;
+						}
+					}
+					if ( outermost ) {
+						dirruns = dirrunsUnique;
+					}
+				}
+
+				// Track unmatched elements for set filters
+				if ( bySet ) {
+
+					// They will have gone through all possible matchers
+					if ( ( elem = !matcher && elem ) ) {
+						matchedCount--;
+					}
+
+					// Lengthen the array for every element, matched or not
+					if ( seed ) {
+						unmatched.push( elem );
+					}
+				}
+			}
+
+			// `i` is now the count of elements visited above, and adding it to `matchedCount`
+			// makes the latter nonnegative.
+			matchedCount += i;
+
+			// Apply set filters to unmatched elements
+			// NOTE: This can be skipped if there are no unmatched elements (i.e., `matchedCount`
+			// equals `i`), unless we didn't visit _any_ elements in the above loop because we have
+			// no element matchers and no seed.
+			// Incrementing an initially-string "0" `i` allows `i` to remain a string only in that
+			// case, which will result in a "00" `matchedCount` that differs from `i` but is also
+			// numerically zero.
+			if ( bySet && i !== matchedCount ) {
+				j = 0;
+				while ( ( matcher = setMatchers[ j++ ] ) ) {
+					matcher( unmatched, setMatched, context, xml );
+				}
+
+				if ( seed ) {
+
+					// Reintegrate element matches to eliminate the need for sorting
+					if ( matchedCount > 0 ) {
+						while ( i-- ) {
+							if ( !( unmatched[ i ] || setMatched[ i ] ) ) {
+								setMatched[ i ] = pop.call( results );
+							}
+						}
+					}
+
+					// Discard index placeholder values to get only actual matches
+					setMatched = condense( setMatched );
+				}
+
+				// Add matches to results
+				push.apply( results, setMatched );
+
+				// Seedless set matches succeeding multiple successful matchers stipulate sorting
+				if ( outermost && !seed && setMatched.length > 0 &&
+					( matchedCount + setMatchers.length ) > 1 ) {
+
+					Sizzle.uniqueSort( results );
+				}
+			}
+
+			// Override manipulation of globals by nested matchers
+			if ( outermost ) {
+				dirruns = dirrunsUnique;
+				outermostContext = contextBackup;
+			}
+
+			return unmatched;
+		};
+
+	return bySet ?
+		markFunction( superMatcher ) :
+		superMatcher;
+}
+
+compile = Sizzle.compile = function( selector, match /* Internal Use Only */ ) {
+	var i,
+		setMatchers = [],
+		elementMatchers = [],
+		cached = compilerCache[ selector + " " ];
+
+	if ( !cached ) {
+
+		// Generate a function of recursive functions that can be used to check each element
+		if ( !match ) {
+			match = tokenize( selector );
+		}
+		i = match.length;
+		while ( i-- ) {
+			cached = matcherFromTokens( match[ i ] );
+			if ( cached[ expando ] ) {
+				setMatchers.push( cached );
+			} else {
+				elementMatchers.push( cached );
+			}
+		}
+
+		// Cache the compiled function
+		cached = compilerCache(
+			selector,
+			matcherFromGroupMatchers( elementMatchers, setMatchers )
+		);
+
+		// Save selector and tokenization
+		cached.selector = selector;
+	}
+	return cached;
+};
+
+/**
+ * A low-level selection function that works with Sizzle's compiled
+ *  selector functions
+ * @param {String|Function} selector A selector or a pre-compiled
+ *  selector function built with Sizzle.compile
+ * @param {Element} context
+ * @param {Array} [results]
+ * @param {Array} [seed] A set of elements to match against
+ */
+select = Sizzle.select = function( selector, context, results, seed ) {
+	var i, tokens, token, type, find,
+		compiled = typeof selector === "function" && selector,
+		match = !seed && tokenize( ( selector = compiled.selector || selector ) );
+
+	results = results || [];
+
+	// Try to minimize operations if there is only one selector in the list and no seed
+	// (the latter of which guarantees us context)
+	if ( match.length === 1 ) {
+
+		// Reduce context if the leading compound selector is an ID
+		tokens = match[ 0 ] = match[ 0 ].slice( 0 );
+		if ( tokens.length > 2 && ( token = tokens[ 0 ] ).type === "ID" &&
+			context.nodeType === 9 && documentIsHTML && Expr.relative[ tokens[ 1 ].type ] ) {
+
+			context = ( Expr.find[ "ID" ]( token.matches[ 0 ]
+				.replace( runescape, funescape ), context ) || [] )[ 0 ];
+			if ( !context ) {
+				return results;
+
+			// Precompiled matchers will still verify ancestry, so step up a level
+			} else if ( compiled ) {
+				context = context.parentNode;
+			}
+
+			selector = selector.slice( tokens.shift().value.length );
+		}
+
+		// Fetch a seed set for right-to-left matching
+		i = matchExpr[ "needsContext" ].test( selector ) ? 0 : tokens.length;
+		while ( i-- ) {
+			token = tokens[ i ];
+
+			// Abort if we hit a combinator
+			if ( Expr.relative[ ( type = token.type ) ] ) {
+				break;
+			}
+			if ( ( find = Expr.find[ type ] ) ) {
+
+				// Search, expanding context for leading sibling combinators
+				if ( ( seed = find(
+					token.matches[ 0 ].replace( runescape, funescape ),
+					rsibling.test( tokens[ 0 ].type ) && testContext( context.parentNode ) ||
+						context
+				) ) ) {
+
+					// If seed is empty or no tokens remain, we can return early
+					tokens.splice( i, 1 );
+					selector = seed.length && toSelector( tokens );
+					if ( !selector ) {
+						push.apply( results, seed );
+						return results;
+					}
+
+					break;
+				}
+			}
+		}
+	}
+
+	// Compile and execute a filtering function if one is not provided
+	// Provide `match` to avoid retokenization if we modified the selector above
+	( compiled || compile( selector, match ) )(
+		seed,
+		context,
+		!documentIsHTML,
+		results,
+		!context || rsibling.test( selector ) && testContext( context.parentNode ) || context
+	);
+	return results;
+};
+
+// One-time assignments
+
+// Sort stability
+support.sortStable = expando.split( "" ).sort( sortOrder ).join( "" ) === expando;
+
+// Support: Chrome 14-35+
+// Always assume duplicates if they aren't passed to the comparison function
+support.detectDuplicates = !!hasDuplicate;
+
+// Initialize against the default document
+setDocument();
+
+// Support: Webkit<537.32 - Safari 6.0.3/Chrome 25 (fixed in Chrome 27)
+// Detached nodes confoundingly follow *each other*
+support.sortDetached = assert( function( el ) {
+
+	// Should return 1, but returns 4 (following)
+	return el.compareDocumentPosition( document.createElement( "fieldset" ) ) & 1;
+} );
+
+// Support: IE<8
+// Prevent attribute/property "interpolation"
+// https://msdn.microsoft.com/en-us/library/ms536429%28VS.85%29.aspx
+if ( !assert( function( el ) {
+	el.innerHTML = "<a href='#'></a>";
+	return el.firstChild.getAttribute( "href" ) === "#";
+} ) ) {
+	addHandle( "type|href|height|width", function( elem, name, isXML ) {
+		if ( !isXML ) {
+			return elem.getAttribute( name, name.toLowerCase() === "type" ? 1 : 2 );
+		}
+	} );
+}
+
+// Support: IE<9
+// Use defaultValue in place of getAttribute("value")
+if ( !support.attributes || !assert( function( el ) {
+	el.innerHTML = "<input/>";
+	el.firstChild.setAttribute( "value", "" );
+	return el.firstChild.getAttribute( "value" ) === "";
+} ) ) {
+	addHandle( "value", function( elem, _name, isXML ) {
+		if ( !isXML && elem.nodeName.toLowerCase() === "input" ) {
+			return elem.defaultValue;
+		}
+	} );
+}
+
+// Support: IE<9
+// Use getAttributeNode to fetch booleans when getAttribute lies
+if ( !assert( function( el ) {
+	return el.getAttribute( "disabled" ) == null;
+} ) ) {
+	addHandle( booleans, function( elem, name, isXML ) {
+		var val;
+		if ( !isXML ) {
+			return elem[ name ] === true ? name.toLowerCase() :
+				( val = elem.getAttributeNode( name ) ) && val.specified ?
+					val.value :
+					null;
+		}
+	} );
+}
+
+return Sizzle;
+
+} )( window );
+
+
+
+jQuery.find = Sizzle;
+jQuery.expr = Sizzle.selectors;
+
+// Deprecated
+jQuery.expr[ ":" ] = jQuery.expr.pseudos;
+jQuery.uniqueSort = jQuery.unique = Sizzle.uniqueSort;
+jQuery.text = Sizzle.getText;
+jQuery.isXMLDoc = Sizzle.isXML;
+jQuery.contains = Sizzle.contains;
+jQuery.escapeSelector = Sizzle.escape;
+
+
+
+
+var dir = function( elem, dir, until ) {
+	var matched = [],
+		truncate = until !== undefined;
+
+	while ( ( elem = elem[ dir ] ) && elem.nodeType !== 9 ) {
+		if ( elem.nodeType === 1 ) {
+			if ( truncate && jQuery( elem ).is( until ) ) {
+				break;
+			}
+			matched.push( elem );
+		}
+	}
+	return matched;
+};
+
+
+var siblings = function( n, elem ) {
+	var matched = [];
+
+	for ( ; n; n = n.nextSibling ) {
+		if ( n.nodeType === 1 && n !== elem ) {
+			matched.push( n );
+		}
+	}
+
+	return matched;
+};
+
+
+var rneedsContext = jQuery.expr.match.needsContext;
+
+
+
+function nodeName( elem, name ) {
+
+	return elem.nodeName && elem.nodeName.toLowerCase() === name.toLowerCase();
+
+}
+var rsingleTag = ( /^<([a-z][^\/\0>:\x20\t\r\n\f]*)[\x20\t\r\n\f]*\/?>(?:<\/\1>|)$/i );
+
+
+
+// Implement the identical functionality for filter and not
+function winnow( elements, qualifier, not ) {
+	if ( isFunction( qualifier ) ) {
+		return jQuery.grep( elements, function( elem, i ) {
+			return !!qualifier.call( elem, i, elem ) !== not;
+		} );
+	}
+
+	// Single element
+	if ( qualifier.nodeType ) {
+		return jQuery.grep( elements, function( elem ) {
+			return ( elem === qualifier ) !== not;
+		} );
+	}
+
+	// Arraylike of elements (jQuery, arguments, Array)
+	if ( typeof qualifier !== "string" ) {
+		return jQuery.grep( elements, function( elem ) {
+			return ( indexOf.call( qualifier, elem ) > -1 ) !== not;
+		} );
+	}
+
+	// Filtered directly for both simple and complex selectors
+	return jQuery.filter( qualifier, elements, not );
+}
+
+jQuery.filter = function( expr, elems, not ) {
+	var elem = elems[ 0 ];
+
+	if ( not ) {
+		expr = ":not(" + expr + ")";
+	}
+
+	if ( elems.length === 1 && elem.nodeType === 1 ) {
+		return jQuery.find.matchesSelector( elem, expr ) ? [ elem ] : [];
+	}
+
+	return jQuery.find.matches( expr, jQuery.grep( elems, function( elem ) {
+		return elem.nodeType === 1;
+	} ) );
+};
+
+jQuery.fn.extend( {
+	find: function( selector ) {
+		var i, ret,
+			len = this.length,
+			self = this;
+
+		if ( typeof selector !== "string" ) {
+			return this.pushStack( jQuery( selector ).filter( function() {
+				for ( i = 0; i < len; i++ ) {
+					if ( jQuery.contains( self[ i ], this ) ) {
+						return true;
+					}
+				}
+			} ) );
+		}
+
+		ret = this.pushStack( [] );
+
+		for ( i = 0; i < len; i++ ) {
+			jQuery.find( selector, self[ i ], ret );
+		}
+
+		return len > 1 ? jQuery.uniqueSort( ret ) : ret;
+	},
+	filter: function( selector ) {
+		return this.pushStack( winnow( this, selector || [], false ) );
+	},
+	not: function( selector ) {
+		return this.pushStack( winnow( this, selector || [], true ) );
+	},
+	is: function( selector ) {
+		return !!winnow(
+			this,
+
+			// If this is a positional/relative selector, check membership in the returned set
+			// so $("p:first").is("p:last") won't return true for a doc with two "p".
+			typeof selector === "string" && rneedsContext.test( selector ) ?
+				jQuery( selector ) :
+				selector || [],
+			false
+		).length;
+	}
+} );
+
+
+// Initialize a jQuery object
+
+
+// A central reference to the root jQuery(document)
+var rootjQuery,
+
+	// A simple way to check for HTML strings
+	// Prioritize #id over <tag> to avoid XSS via location.hash (#9521)
+	// Strict HTML recognition (#11290: must start with <)
+	// Shortcut simple #id case for speed
+	rquickExpr = /^(?:\s*(<[\w\W]+>)[^>]*|#([\w-]+))$/,
+
+	init = jQuery.fn.init = function( selector, context, root ) {
+		var match, elem;
+
+		// HANDLE: $(""), $(null), $(undefined), $(false)
+		if ( !selector ) {
+			return this;
+		}
+
+		// Method init() accepts an alternate rootjQuery
+		// so migrate can support jQuery.sub (gh-2101)
+		root = root || rootjQuery;
+
+		// Handle HTML strings
+		if ( typeof selector === "string" ) {
+			if ( selector[ 0 ] === "<" &&
+				selector[ selector.length - 1 ] === ">" &&
+				selector.length >= 3 ) {
+
+				// Assume that strings that start and end with <> are HTML and skip the regex check
+				match = [ null, selector, null ];
+
+			} else {
+				match = rquickExpr.exec( selector );
+			}
+
+			// Match html or make sure no context is specified for #id
+			if ( match && ( match[ 1 ] || !context ) ) {
+
+				// HANDLE: $(html) -> $(array)
+				if ( match[ 1 ] ) {
+					context = context instanceof jQuery ? context[ 0 ] : context;
+
+					// Option to run scripts is true for back-compat
+					// Intentionally let the error be thrown if parseHTML is not present
+					jQuery.merge( this, jQuery.parseHTML(
+						match[ 1 ],
+						context && context.nodeType ? context.ownerDocument || context : document,
+						true
+					) );
+
+					// HANDLE: $(html, props)
+					if ( rsingleTag.test( match[ 1 ] ) && jQuery.isPlainObject( context ) ) {
+						for ( match in context ) {
+
+							// Properties of context are called as methods if possible
+							if ( isFunction( this[ match ] ) ) {
+								this[ match ]( context[ match ] );
+
+							// ...and otherwise set as attributes
+							} else {
+								this.attr( match, context[ match ] );
+							}
+						}
+					}
+
+					return this;
+
+				// HANDLE: $(#id)
+				} else {
+					elem = document.getElementById( match[ 2 ] );
+
+					if ( elem ) {
+
+						// Inject the element directly into the jQuery object
+						this[ 0 ] = elem;
+						this.length = 1;
+					}
+					return this;
+				}
+
+			// HANDLE: $(expr, $(...))
+			} else if ( !context || context.jquery ) {
+				return ( context || root ).find( selector );
+
+			// HANDLE: $(expr, context)
+			// (which is just equivalent to: $(context).find(expr)
+			} else {
+				return this.constructor( context ).find( selector );
+			}
+
+		// HANDLE: $(DOMElement)
+		} else if ( selector.nodeType ) {
+			this[ 0 ] = selector;
+			this.length = 1;
+			return this;
+
+		// HANDLE: $(function)
+		// Shortcut for document ready
+		} else if ( isFunction( selector ) ) {
+			return root.ready !== undefined ?
+				root.ready( selector ) :
+
+				// Execute immediately if ready is not present
+				selector( jQuery );
+		}
+
+		return jQuery.makeArray( selector, this );
+	};
+
+// Give the init function the jQuery prototype for later instantiation
+init.prototype = jQuery.fn;
+
+// Initialize central reference
+rootjQuery = jQuery( document );
+
+
+var rparentsprev = /^(?:parents|prev(?:Until|All))/,
+
+	// Methods guaranteed to produce a unique set when starting from a unique set
+	guaranteedUnique = {
+		children: true,
+		contents: true,
+		next: true,
+		prev: true
+	};
+
+jQuery.fn.extend( {
+	has: function( target ) {
+		var targets = jQuery( target, this ),
+			l = targets.length;
+
+		return this.filter( function() {
+			var i = 0;
+			for ( ; i < l; i++ ) {
+				if ( jQuery.contains( this, targets[ i ] ) ) {
+					return true;
+				}
+			}
+		} );
+	},
+
+	closest: function( selectors, context ) {
+		var cur,
+			i = 0,
+			l = this.length,
+			matched = [],
+			targets = typeof selectors !== "string" && jQuery( selectors );
+
+		// Positional selectors never match, since there's no _selection_ context
+		if ( !rneedsContext.test( selectors ) ) {
+			for ( ; i < l; i++ ) {
+				for ( cur = this[ i ]; cur && cur !== context; cur = cur.parentNode ) {
+
+					// Always skip document fragments
+					if ( cur.nodeType < 11 && ( targets ?
+						targets.index( cur ) > -1 :
+
+						// Don't pass non-elements to Sizzle
+						cur.nodeType === 1 &&
+							jQuery.find.matchesSelector( cur, selectors ) ) ) {
+
+						matched.push( cur );
+						break;
+					}
+				}
+			}
+		}
+
+		return this.pushStack( matched.length > 1 ? jQuery.uniqueSort( matched ) : matched );
+	},
+
+	// Determine the position of an element within the set
+	index: function( elem ) {
+
+		// No argument, return index in parent
+		if ( !elem ) {
+			return ( this[ 0 ] && this[ 0 ].parentNode ) ? this.first().prevAll().length : -1;
+		}
+
+		// Index in selector
+		if ( typeof elem === "string" ) {
+			return indexOf.call( jQuery( elem ), this[ 0 ] );
+		}
+
+		// Locate the position of the desired element
+		return indexOf.call( this,
+
+			// If it receives a jQuery object, the first element is used
+			elem.jquery ? elem[ 0 ] : elem
+		);
+	},
+
+	add: function( selector, context ) {
+		return this.pushStack(
+			jQuery.uniqueSort(
+				jQuery.merge( this.get(), jQuery( selector, context ) )
+			)
+		);
+	},
+
+	addBack: function( selector ) {
+		return this.add( selector == null ?
+			this.prevObject : this.prevObject.filter( selector )
+		);
+	}
+} );
+
+function sibling( cur, dir ) {
+	while ( ( cur = cur[ dir ] ) && cur.nodeType !== 1 ) {}
+	return cur;
+}
+
+jQuery.each( {
+	parent: function( elem ) {
+		var parent = elem.parentNode;
+		return parent && parent.nodeType !== 11 ? parent : null;
+	},
+	parents: function( elem ) {
+		return dir( elem, "parentNode" );
+	},
+	parentsUntil: function( elem, _i, until ) {
+		return dir( elem, "parentNode", until );
+	},
+	next: function( elem ) {
+		return sibling( elem, "nextSibling" );
+	},
+	prev: function( elem ) {
+		return sibling( elem, "previousSibling" );
+	},
+	nextAll: function( elem ) {
+		return dir( elem, "nextSibling" );
+	},
+	prevAll: function( elem ) {
+		return dir( elem, "previousSibling" );
+	},
+	nextUntil: function( elem, _i, until ) {
+		return dir( elem, "nextSibling", until );
+	},
+	prevUntil: function( elem, _i, until ) {
+		return dir( elem, "previousSibling", until );
+	},
+	siblings: function( elem ) {
+		return siblings( ( elem.parentNode || {} ).firstChild, elem );
+	},
+	children: function( elem ) {
+		return siblings( elem.firstChild );
+	},
+	contents: function( elem ) {
+		if ( elem.contentDocument != null &&
+
+			// Support: IE 11+
+			// <object> elements with no `data` attribute has an object
+			// `contentDocument` with a `null` prototype.
+			getProto( elem.contentDocument ) ) {
+
+			return elem.contentDocument;
+		}
+
+		// Support: IE 9 - 11 only, iOS 7 only, Android Browser <=4.3 only
+		// Treat the template element as a regular one in browsers that
+		// don't support it.
+		if ( nodeName( elem, "template" ) ) {
+			elem = elem.content || elem;
+		}
+
+		return jQuery.merge( [], elem.childNodes );
+	}
+}, function( name, fn ) {
+	jQuery.fn[ name ] = function( until, selector ) {
+		var matched = jQuery.map( this, fn, until );
+
+		if ( name.slice( -5 ) !== "Until" ) {
+			selector = until;
+		}
+
+		if ( selector && typeof selector === "string" ) {
+			matched = jQuery.filter( selector, matched );
+		}
+
+		if ( this.length > 1 ) {
+
+			// Remove duplicates
+			if ( !guaranteedUnique[ name ] ) {
+				jQuery.uniqueSort( matched );
+			}
+
+			// Reverse order for parents* and prev-derivatives
+			if ( rparentsprev.test( name ) ) {
+				matched.reverse();
+			}
+		}
+
+		return this.pushStack( matched );
+	};
+} );
+var rnothtmlwhite = ( /[^\x20\t\r\n\f]+/g );
+
+
+
+// Convert String-formatted options into Object-formatted ones
+function createOptions( options ) {
+	var object = {};
+	jQuery.each( options.match( rnothtmlwhite ) || [], function( _, flag ) {
+		object[ flag ] = true;
+	} );
+	return object;
+}
+
+/*
+ * Create a callback list using the following parameters:
+ *
+ *	options: an optional list of space-separated options that will change how
+ *			the callback list behaves or a more traditional option object
+ *
+ * By default a callback list will act like an event callback list and can be
+ * "fired" multiple times.
+ *
+ * Possible options:
+ *
+ *	once:			will ensure the callback list can only be fired once (like a Deferred)
+ *
+ *	memory:			will keep track of previous values and will call any callback added
+ *					after the list has been fired right away with the latest "memorized"
+ *					values (like a Deferred)
+ *
+ *	unique:			will ensure a callback can only be added once (no duplicate in the list)
+ *
+ *	stopOnFalse:	interrupt callings when a callback returns false
+ *
+ */
+jQuery.Callbacks = function( options ) {
+
+	// Convert options from String-formatted to Object-formatted if needed
+	// (we check in cache first)
+	options = typeof options === "string" ?
+		createOptions( options ) :
+		jQuery.extend( {}, options );
+
+	var // Flag to know if list is currently firing
+		firing,
+
+		// Last fire value for non-forgettable lists
+		memory,
+
+		// Flag to know if list was already fired
+		fired,
+
+		// Flag to prevent firing
+		locked,
+
+		// Actual callback list
+		list = [],
+
+		// Queue of execution data for repeatable lists
+		queue = [],
+
+		// Index of currently firing callback (modified by add/remove as needed)
+		firingIndex = -1,
+
+		// Fire callbacks
+		fire = function() {
+
+			// Enforce single-firing
+			locked = locked || options.once;
+
+			// Execute callbacks for all pending executions,
+			// respecting firingIndex overrides and runtime changes
+			fired = firing = true;
+			for ( ; queue.length; firingIndex = -1 ) {
+				memory = queue.shift();
+				while ( ++firingIndex < list.length ) {
+
+					// Run callback and check for early termination
+					if ( list[ firingIndex ].apply( memory[ 0 ], memory[ 1 ] ) === false &&
+						options.stopOnFalse ) {
+
+						// Jump to end and forget the data so .add doesn't re-fire
+						firingIndex = list.length;
+						memory = false;
+					}
+				}
+			}
+
+			// Forget the data if we're done with it
+			if ( !options.memory ) {
+				memory = false;
+			}
+
+			firing = false;
+
+			// Clean up if we're done firing for good
+			if ( locked ) {
+
+				// Keep an empty list if we have data for future add calls
+				if ( memory ) {
+					list = [];
+
+				// Otherwise, this object is spent
+				} else {
+					list = "";
+				}
+			}
+		},
+
+		// Actual Callbacks object
+		self = {
+
+			// Add a callback or a collection of callbacks to the list
+			add: function() {
+				if ( list ) {
+
+					// If we have memory from a past run, we should fire after adding
+					if ( memory && !firing ) {
+						firingIndex = list.length - 1;
+						queue.push( memory );
+					}
+
+					( function add( args ) {
+						jQuery.each( args, function( _, arg ) {
+							if ( isFunction( arg ) ) {
+								if ( !options.unique || !self.has( arg ) ) {
+									list.push( arg );
+								}
+							} else if ( arg && arg.length && toType( arg ) !== "string" ) {
+
+								// Inspect recursively
+								add( arg );
+							}
+						} );
+					} )( arguments );
+
+					if ( memory && !firing ) {
+						fire();
+					}
+				}
+				return this;
+			},
+
+			// Remove a callback from the list
+			remove: function() {
+				jQuery.each( arguments, function( _, arg ) {
+					var index;
+					while ( ( index = jQuery.inArray( arg, list, index ) ) > -1 ) {
+						list.splice( index, 1 );
+
+						// Handle firing indexes
+						if ( index <= firingIndex ) {
+							firingIndex--;
+						}
+					}
+				} );
+				return this;
+			},
+
+			// Check if a given callback is in the list.
+			// If no argument is given, return whether or not list has callbacks attached.
+			has: function( fn ) {
+				return fn ?
+					jQuery.inArray( fn, list ) > -1 :
+					list.length > 0;
+			},
+
+			// Remove all callbacks from the list
+			empty: function() {
+				if ( list ) {
+					list = [];
+				}
+				return this;
+			},
+
+			// Disable .fire and .add
+			// Abort any current/pending executions
+			// Clear all callbacks and values
+			disable: function() {
+				locked = queue = [];
+				list = memory = "";
+				return this;
+			},
+			disabled: function() {
+				return !list;
+			},
+
+			// Disable .fire
+			// Also disable .add unless we have memory (since it would have no effect)
+			// Abort any pending executions
+			lock: function() {
+				locked = queue = [];
+				if ( !memory && !firing ) {
+					list = memory = "";
+				}
+				return this;
+			},
+			locked: function() {
+				return !!locked;
+			},
+
+			// Call all callbacks with the given context and arguments
+			fireWith: function( context, args ) {
+				if ( !locked ) {
+					args = args || [];
+					args = [ context, args.slice ? args.slice() : args ];
+					queue.push( args );
+					if ( !firing ) {
+						fire();
+					}
+				}
+				return this;
+			},
+
+			// Call all the callbacks with the given arguments
+			fire: function() {
+				self.fireWith( this, arguments );
+				return this;
+			},
+
+			// To know if the callbacks have already been called at least once
+			fired: function() {
+				return !!fired;
+			}
+		};
+
+	return self;
+};
+
+
+function Identity( v ) {
+	return v;
+}
+function Thrower( ex ) {
+	throw ex;
+}
+
+function adoptValue( value, resolve, reject, noValue ) {
+	var method;
+
+	try {
+
+		// Check for promise aspect first to privilege synchronous behavior
+		if ( value && isFunction( ( method = value.promise ) ) ) {
+			method.call( value ).done( resolve ).fail( reject );
+
+		// Other thenables
+		} else if ( value && isFunction( ( method = value.then ) ) ) {
+			method.call( value, resolve, reject );
+
+		// Other non-thenables
+		} else {
+
+			// Control `resolve` arguments by letting Array#slice cast boolean `noValue` to integer:
+			// * false: [ value ].slice( 0 ) => resolve( value )
+			// * true: [ value ].slice( 1 ) => resolve()
+			resolve.apply( undefined, [ value ].slice( noValue ) );
+		}
+
+	// For Promises/A+, convert exceptions into rejections
+	// Since jQuery.when doesn't unwrap thenables, we can skip the extra checks appearing in
+	// Deferred#then to conditionally suppress rejection.
+	} catch ( value ) {
+
+		// Support: Android 4.0 only
+		// Strict mode functions invoked without .call/.apply get global-object context
+		reject.apply( undefined, [ value ] );
+	}
+}
+
+jQuery.extend( {
+
+	Deferred: function( func ) {
+		var tuples = [
+
+				// action, add listener, callbacks,
+				// ... .then handlers, argument index, [final state]
+				[ "notify", "progress", jQuery.Callbacks( "memory" ),
+					jQuery.Callbacks( "memory" ), 2 ],
+				[ "resolve", "done", jQuery.Callbacks( "once memory" ),
+					jQuery.Callbacks( "once memory" ), 0, "resolved" ],
+				[ "reject", "fail", jQuery.Callbacks( "once memory" ),
+					jQuery.Callbacks( "once memory" ), 1, "rejected" ]
+			],
+			state = "pending",
+			promise = {
+				state: function() {
+					return state;
+				},
+				always: function() {
+					deferred.done( arguments ).fail( arguments );
+					return this;
+				},
+				"catch": function( fn ) {
+					return promise.then( null, fn );
+				},
+
+				// Keep pipe for back-compat
+				pipe: function( /* fnDone, fnFail, fnProgress */ ) {
+					var fns = arguments;
+
+					return jQuery.Deferred( function( newDefer ) {
+						jQuery.each( tuples, function( _i, tuple ) {
+
+							// Map tuples (progress, done, fail) to arguments (done, fail, progress)
+							var fn = isFunction( fns[ tuple[ 4 ] ] ) && fns[ tuple[ 4 ] ];
+
+							// deferred.progress(function() { bind to newDefer or newDefer.notify })
+							// deferred.done(function() { bind to newDefer or newDefer.resolve })
+							// deferred.fail(function() { bind to newDefer or newDefer.reject })
+							deferred[ tuple[ 1 ] ]( function() {
+								var returned = fn && fn.apply( this, arguments );
+								if ( returned && isFunction( returned.promise ) ) {
+									returned.promise()
+										.progress( newDefer.notify )
+										.done( newDefer.resolve )
+										.fail( newDefer.reject );
+								} else {
+									newDefer[ tuple[ 0 ] + "With" ](
+										this,
+										fn ? [ returned ] : arguments
+									);
+								}
+							} );
+						} );
+						fns = null;
+					} ).promise();
+				},
+				then: function( onFulfilled, onRejected, onProgress ) {
+					var maxDepth = 0;
+					function resolve( depth, deferred, handler, special ) {
+						return function() {
+							var that = this,
+								args = arguments,
+								mightThrow = function() {
+									var returned, then;
+
+									// Support: Promises/A+ section 2.3.3.3.3
+									// https://promisesaplus.com/#point-59
+									// Ignore double-resolution attempts
+									if ( depth < maxDepth ) {
+										return;
+									}
+
+									returned = handler.apply( that, args );
+
+									// Support: Promises/A+ section 2.3.1
+									// https://promisesaplus.com/#point-48
+									if ( returned === deferred.promise() ) {
+										throw new TypeError( "Thenable self-resolution" );
+									}
+
+									// Support: Promises/A+ sections 2.3.3.1, 3.5
+									// https://promisesaplus.com/#point-54
+									// https://promisesaplus.com/#point-75
+									// Retrieve `then` only once
+									then = returned &&
+
+										// Support: Promises/A+ section 2.3.4
+										// https://promisesaplus.com/#point-64
+										// Only check objects and functions for thenability
+										( typeof returned === "object" ||
+											typeof returned === "function" ) &&
+										returned.then;
+
+									// Handle a returned thenable
+									if ( isFunction( then ) ) {
+
+										// Special processors (notify) just wait for resolution
+										if ( special ) {
+											then.call(
+												returned,
+												resolve( maxDepth, deferred, Identity, special ),
+												resolve( maxDepth, deferred, Thrower, special )
+											);
+
+										// Normal processors (resolve) also hook into progress
+										} else {
+
+											// ...and disregard older resolution values
+											maxDepth++;
+
+											then.call(
+												returned,
+												resolve( maxDepth, deferred, Identity, special ),
+												resolve( maxDepth, deferred, Thrower, special ),
+												resolve( maxDepth, deferred, Identity,
+													deferred.notifyWith )
+											);
+										}
+
+									// Handle all other returned values
+									} else {
+
+										// Only substitute handlers pass on context
+										// and multiple values (non-spec behavior)
+										if ( handler !== Identity ) {
+											that = undefined;
+											args = [ returned ];
+										}
+
+										// Process the value(s)
+										// Default process is resolve
+										( special || deferred.resolveWith )( that, args );
+									}
+								},
+
+								// Only normal processors (resolve) catch and reject exceptions
+								process = special ?
+									mightThrow :
+									function() {
+										try {
+											mightThrow();
+										} catch ( e ) {
+
+											if ( jQuery.Deferred.exceptionHook ) {
+												jQuery.Deferred.exceptionHook( e,
+													process.stackTrace );
+											}
+
+											// Support: Promises/A+ section 2.3.3.3.4.1
+											// https://promisesaplus.com/#point-61
+											// Ignore post-resolution exceptions
+											if ( depth + 1 >= maxDepth ) {
+
+												// Only substitute handlers pass on context
+												// and multiple values (non-spec behavior)
+												if ( handler !== Thrower ) {
+													that = undefined;
+													args = [ e ];
+												}
+
+												deferred.rejectWith( that, args );
+											}
+										}
+									};
+
+							// Support: Promises/A+ section 2.3.3.3.1
+							// https://promisesaplus.com/#point-57
+							// Re-resolve promises immediately to dodge false rejection from
+							// subsequent errors
+							if ( depth ) {
+								process();
+							} else {
+
+								// Call an optional hook to record the stack, in case of exception
+								// since it's otherwise lost when execution goes async
+								if ( jQuery.Deferred.getStackHook ) {
+									process.stackTrace = jQuery.Deferred.getStackHook();
+								}
+								window.setTimeout( process );
+							}
+						};
+					}
+
+					return jQuery.Deferred( function( newDefer ) {
+
+						// progress_handlers.add( ... )
+						tuples[ 0 ][ 3 ].add(
+							resolve(
+								0,
+								newDefer,
+								isFunction( onProgress ) ?
+									onProgress :
+									Identity,
+								newDefer.notifyWith
+							)
+						);
+
+						// fulfilled_handlers.add( ... )
+						tuples[ 1 ][ 3 ].add(
+							resolve(
+								0,
+								newDefer,
+								isFunction( onFulfilled ) ?
+									onFulfilled :
+									Identity
+							)
+						);
+
+						// rejected_handlers.add( ... )
+						tuples[ 2 ][ 3 ].add(
+							resolve(
+								0,
+								newDefer,
+								isFunction( onRejected ) ?
+									onRejected :
+									Thrower
+							)
+						);
+					} ).promise();
+				},
+
+				// Get a promise for this deferred
+				// If obj is provided, the promise aspect is added to the object
+				promise: function( obj ) {
+					return obj != null ? jQuery.extend( obj, promise ) : promise;
+				}
+			},
+			deferred = {};
+
+		// Add list-specific methods
+		jQuery.each( tuples, function( i, tuple ) {
+			var list = tuple[ 2 ],
+				stateString = tuple[ 5 ];
+
+			// promise.progress = list.add
+			// promise.done = list.add
+			// promise.fail = list.add
+			promise[ tuple[ 1 ] ] = list.add;
+
+			// Handle state
+			if ( stateString ) {
+				list.add(
+					function() {
+
+						// state = "resolved" (i.e., fulfilled)
+						// state = "rejected"
+						state = stateString;
+					},
+
+					// rejected_callbacks.disable
+					// fulfilled_callbacks.disable
+					tuples[ 3 - i ][ 2 ].disable,
+
+					// rejected_handlers.disable
+					// fulfilled_handlers.disable
+					tuples[ 3 - i ][ 3 ].disable,
+
+					// progress_callbacks.lock
+					tuples[ 0 ][ 2 ].lock,
+
+					// progress_handlers.lock
+					tuples[ 0 ][ 3 ].lock
+				);
+			}
+
+			// progress_handlers.fire
+			// fulfilled_handlers.fire
+			// rejected_handlers.fire
+			list.add( tuple[ 3 ].fire );
+
+			// deferred.notify = function() { deferred.notifyWith(...) }
+			// deferred.resolve = function() { deferred.resolveWith(...) }
+			// deferred.reject = function() { deferred.rejectWith(...) }
+			deferred[ tuple[ 0 ] ] = function() {
+				deferred[ tuple[ 0 ] + "With" ]( this === deferred ? undefined : this, arguments );
+				return this;
+			};
+
+			// deferred.notifyWith = list.fireWith
+			// deferred.resolveWith = list.fireWith
+			// deferred.rejectWith = list.fireWith
+			deferred[ tuple[ 0 ] + "With" ] = list.fireWith;
+		} );
+
+		// Make the deferred a promise
+		promise.promise( deferred );
+
+		// Call given func if any
+		if ( func ) {
+			func.call( deferred, deferred );
+		}
+
+		// All done!
+		return deferred;
+	},
+
+	// Deferred helper
+	when: function( singleValue ) {
+		var
+
+			// count of uncompleted subordinates
+			remaining = arguments.length,
+
+			// count of unprocessed arguments
+			i = remaining,
+
+			// subordinate fulfillment data
+			resolveContexts = Array( i ),
+			resolveValues = slice.call( arguments ),
+
+			// the primary Deferred
+			primary = jQuery.Deferred(),
+
+			// subordinate callback factory
+			updateFunc = function( i ) {
+				return function( value ) {
+					resolveContexts[ i ] = this;
+					resolveValues[ i ] = arguments.length > 1 ? slice.call( arguments ) : value;
+					if ( !( --remaining ) ) {
+						primary.resolveWith( resolveContexts, resolveValues );
+					}
+				};
+			};
+
+		// Single- and empty arguments are adopted like Promise.resolve
+		if ( remaining <= 1 ) {
+			adoptValue( singleValue, primary.done( updateFunc( i ) ).resolve, primary.reject,
+				!remaining );
+
+			// Use .then() to unwrap secondary thenables (cf. gh-3000)
+			if ( primary.state() === "pending" ||
+				isFunction( resolveValues[ i ] && resolveValues[ i ].then ) ) {
+
+				return primary.then();
+			}
+		}
+
+		// Multiple arguments are aggregated like Promise.all array elements
+		while ( i-- ) {
+			adoptValue( resolveValues[ i ], updateFunc( i ), primary.reject );
+		}
+
+		return primary.promise();
+	}
+} );
+
+
+// These usually indicate a programmer mistake during development,
+// warn about them ASAP rather than swallowing them by default.
+var rerrorNames = /^(Eval|Internal|Range|Reference|Syntax|Type|URI)Error$/;
+
+jQuery.Deferred.exceptionHook = function( error, stack ) {
+
+	// Support: IE 8 - 9 only
+	// Console exists when dev tools are open, which can happen at any time
+	if ( window.console && window.console.warn && error && rerrorNames.test( error.name ) ) {
+		window.console.warn( "jQuery.Deferred exception: " + error.message, error.stack, stack );
+	}
+};
+
+
+
+
+jQuery.readyException = function( error ) {
+	window.setTimeout( function() {
+		throw error;
+	} );
+};
+
+
+
+
+// The deferred used on DOM ready
+var readyList = jQuery.Deferred();
+
+jQuery.fn.ready = function( fn ) {
+
+	readyList
+		.then( fn )
+
+		// Wrap jQuery.readyException in a function so that the lookup
+		// happens at the time of error handling instead of callback
+		// registration.
+		.catch( function( error ) {
+			jQuery.readyException( error );
+		} );
+
+	return this;
+};
+
+jQuery.extend( {
+
+	// Is the DOM ready to be used? Set to true once it occurs.
+	isReady: false,
+
+	// A counter to track how many items to wait for before
+	// the ready event fires. See #6781
+	readyWait: 1,
+
+	// Handle when the DOM is ready
+	ready: function( wait ) {
+
+		// Abort if there are pending holds or we're already ready
+		if ( wait === true ? --jQuery.readyWait : jQuery.isReady ) {
+			return;
+		}
+
+		// Remember that the DOM is ready
+		jQuery.isReady = true;
+
+		// If a normal DOM Ready event fired, decrement, and wait if need be
+		if ( wait !== true && --jQuery.readyWait > 0 ) {
+			return;
+		}
+
+		// If there are functions bound, to execute
+		readyList.resolveWith( document, [ jQuery ] );
+	}
+} );
+
+jQuery.ready.then = readyList.then;
+
+// The ready event handler and self cleanup method
+function completed() {
+	document.removeEventListener( "DOMContentLoaded", completed );
+	window.removeEventListener( "load", completed );
+	jQuery.ready();
+}
+
+// Catch cases where $(document).ready() is called
+// after the browser event has already occurred.
+// Support: IE <=9 - 10 only
+// Older IE sometimes signals "interactive" too soon
+if ( document.readyState === "complete" ||
+	( document.readyState !== "loading" && !document.documentElement.doScroll ) ) {
+
+	// Handle it asynchronously to allow scripts the opportunity to delay ready
+	window.setTimeout( jQuery.ready );
+
+} else {
+
+	// Use the handy event callback
+	document.addEventListener( "DOMContentLoaded", completed );
+
+	// A fallback to window.onload, that will always work
+	window.addEventListener( "load", completed );
+}
+
+
+
+
+// Multifunctional method to get and set values of a collection
+// The value/s can optionally be executed if it's a function
+var access = function( elems, fn, key, value, chainable, emptyGet, raw ) {
+	var i = 0,
+		len = elems.length,
+		bulk = key == null;
+
+	// Sets many values
+	if ( toType( key ) === "object" ) {
+		chainable = true;
+		for ( i in key ) {
+			access( elems, fn, i, key[ i ], true, emptyGet, raw );
+		}
+
+	// Sets one value
+	} else if ( value !== undefined ) {
+		chainable = true;
+
+		if ( !isFunction( value ) ) {
+			raw = true;
+		}
+
+		if ( bulk ) {
+
+			// Bulk operations run against the entire set
+			if ( raw ) {
+				fn.call( elems, value );
+				fn = null;
+
+			// ...except when executing function values
+			} else {
+				bulk = fn;
+				fn = function( elem, _key, value ) {
+					return bulk.call( jQuery( elem ), value );
+				};
+			}
+		}
+
+		if ( fn ) {
+			for ( ; i < len; i++ ) {
+				fn(
+					elems[ i ], key, raw ?
+						value :
+						value.call( elems[ i ], i, fn( elems[ i ], key ) )
+				);
+			}
+		}
+	}
+
+	if ( chainable ) {
+		return elems;
+	}
+
+	// Gets
+	if ( bulk ) {
+		return fn.call( elems );
+	}
+
+	return len ? fn( elems[ 0 ], key ) : emptyGet;
+};
+
+
+// Matches dashed string for camelizing
+var rmsPrefix = /^-ms-/,
+	rdashAlpha = /-([a-z])/g;
+
+// Used by camelCase as callback to replace()
+function fcamelCase( _all, letter ) {
+	return letter.toUpperCase();
+}
+
+// Convert dashed to camelCase; used by the css and data modules
+// Support: IE <=9 - 11, Edge 12 - 15
+// Microsoft forgot to hump their vendor prefix (#9572)
+function camelCase( string ) {
+	return string.replace( rmsPrefix, "ms-" ).replace( rdashAlpha, fcamelCase );
+}
+var acceptData = function( owner ) {
+
+	// Accepts only:
+	//  - Node
+	//    - Node.ELEMENT_NODE
+	//    - Node.DOCUMENT_NODE
+	//  - Object
+	//    - Any
+	return owner.nodeType === 1 || owner.nodeType === 9 || !( +owner.nodeType );
+};
+
+
+
+
+function Data() {
+	this.expando = jQuery.expando + Data.uid++;
+}
+
+Data.uid = 1;
+
+Data.prototype = {
+
+	cache: function( owner ) {
+
+		// Check if the owner object already has a cache
+		var value = owner[ this.expando ];
+
+		// If not, create one
+		if ( !value ) {
+			value = {};
+
+			// We can accept data for non-element nodes in modern browsers,
+			// but we should not, see #8335.
+			// Always return an empty object.
+			if ( acceptData( owner ) ) {
+
+				// If it is a node unlikely to be stringify-ed or looped over
+				// use plain assignment
+				if ( owner.nodeType ) {
+					owner[ this.expando ] = value;
+
+				// Otherwise secure it in a non-enumerable property
+				// configurable must be true to allow the property to be
+				// deleted when data is removed
+				} else {
+					Object.defineProperty( owner, this.expando, {
+						value: value,
+						configurable: true
+					} );
+				}
+			}
+		}
+
+		return value;
+	},
+	set: function( owner, data, value ) {
+		var prop,
+			cache = this.cache( owner );
+
+		// Handle: [ owner, key, value ] args
+		// Always use camelCase key (gh-2257)
+		if ( typeof data === "string" ) {
+			cache[ camelCase( data ) ] = value;
+
+		// Handle: [ owner, { properties } ] args
+		} else {
+
+			// Copy the properties one-by-one to the cache object
+			for ( prop in data ) {
+				cache[ camelCase( prop ) ] = data[ prop ];
+			}
+		}
+		return cache;
+	},
+	get: function( owner, key ) {
+		return key === undefined ?
+			this.cache( owner ) :
+
+			// Always use camelCase key (gh-2257)
+			owner[ this.expando ] && owner[ this.expando ][ camelCase( key ) ];
+	},
+	access: function( owner, key, value ) {
+
+		// In cases where either:
+		//
+		//   1. No key was specified
+		//   2. A string key was specified, but no value provided
+		//
+		// Take the "read" path and allow the get method to determine
+		// which value to return, respectively either:
+		//
+		//   1. The entire cache object
+		//   2. The data stored at the key
+		//
+		if ( key === undefined ||
+				( ( key && typeof key === "string" ) && value === undefined ) ) {
+
+			return this.get( owner, key );
+		}
+
+		// When the key is not a string, or both a key and value
+		// are specified, set or extend (existing objects) with either:
+		//
+		//   1. An object of properties
+		//   2. A key and value
+		//
+		this.set( owner, key, value );
+
+		// Since the "set" path can have two possible entry points
+		// return the expected data based on which path was taken[*]
+		return value !== undefined ? value : key;
+	},
+	remove: function( owner, key ) {
+		var i,
+			cache = owner[ this.expando ];
+
+		if ( cache === undefined ) {
+			return;
+		}
+
+		if ( key !== undefined ) {
+
+			// Support array or space separated string of keys
+			if ( Array.isArray( key ) ) {
+
+				// If key is an array of keys...
+				// We always set camelCase keys, so remove that.
+				key = key.map( camelCase );
+			} else {
+				key = camelCase( key );
+
+				// If a key with the spaces exists, use it.
+				// Otherwise, create an array by matching non-whitespace
+				key = key in cache ?
+					[ key ] :
+					( key.match( rnothtmlwhite ) || [] );
+			}
+
+			i = key.length;
+
+			while ( i-- ) {
+				delete cache[ key[ i ] ];
+			}
+		}
+
+		// Remove the expando if there's no more data
+		if ( key === undefined || jQuery.isEmptyObject( cache ) ) {
+
+			// Support: Chrome <=35 - 45
+			// Webkit & Blink performance suffers when deleting properties
+			// from DOM nodes, so set to undefined instead
+			// https://bugs.chromium.org/p/chromium/issues/detail?id=378607 (bug restricted)
+			if ( owner.nodeType ) {
+				owner[ this.expando ] = undefined;
+			} else {
+				delete owner[ this.expando ];
+			}
+		}
+	},
+	hasData: function( owner ) {
+		var cache = owner[ this.expando ];
+		return cache !== undefined && !jQuery.isEmptyObject( cache );
+	}
+};
+var dataPriv = new Data();
+
+var dataUser = new Data();
+
+
+
+//	Implementation Summary
+//
+//	1. Enforce API surface and semantic compatibility with 1.9.x branch
+//	2. Improve the module's maintainability by reducing the storage
+//		paths to a single mechanism.
+//	3. Use the same single mechanism to support "private" and "user" data.
+//	4. _Never_ expose "private" data to user code (TODO: Drop _data, _removeData)
+//	5. Avoid exposing implementation details on user objects (eg. expando properties)
+//	6. Provide a clear path for implementation upgrade to WeakMap in 2014
+
+var rbrace = /^(?:\{[\w\W]*\}|\[[\w\W]*\])$/,
+	rmultiDash = /[A-Z]/g;
+
+function getData( data ) {
+	if ( data === "true" ) {
+		return true;
+	}
+
+	if ( data === "false" ) {
+		return false;
+	}
+
+	if ( data === "null" ) {
+		return null;
+	}
+
+	// Only convert to a number if it doesn't change the string
+	if ( data === +data + "" ) {
+		return +data;
+	}
+
+	if ( rbrace.test( data ) ) {
+		return JSON.parse( data );
+	}
+
+	return data;
+}
+
+function dataAttr( elem, key, data ) {
+	var name;
+
+	// If nothing was found internally, try to fetch any
+	// data from the HTML5 data-* attribute
+	if ( data === undefined && elem.nodeType === 1 ) {
+		name = "data-" + key.replace( rmultiDash, "-$&" ).toLowerCase();
+		data = elem.getAttribute( name );
+
+		if ( typeof data === "string" ) {
+			try {
+				data = getData( data );
+			} catch ( e ) {}
+
+			// Make sure we set the data so it isn't changed later
+			dataUser.set( elem, key, data );
+		} else {
+			data = undefined;
+		}
+	}
+	return data;
+}
+
+jQuery.extend( {
+	hasData: function( elem ) {
+		return dataUser.hasData( elem ) || dataPriv.hasData( elem );
+	},
+
+	data: function( elem, name, data ) {
+		return dataUser.access( elem, name, data );
+	},
+
+	removeData: function( elem, name ) {
+		dataUser.remove( elem, name );
+	},
+
+	// TODO: Now that all calls to _data and _removeData have been replaced
+	// with direct calls to dataPriv methods, these can be deprecated.
+	_data: function( elem, name, data ) {
+		return dataPriv.access( elem, name, data );
+	},
+
+	_removeData: function( elem, name ) {
+		dataPriv.remove( elem, name );
+	}
+} );
+
+jQuery.fn.extend( {
+	data: function( key, value ) {
+		var i, name, data,
+			elem = this[ 0 ],
+			attrs = elem && elem.attributes;
+
+		// Gets all values
+		if ( key === undefined ) {
+			if ( this.length ) {
+				data = dataUser.get( elem );
+
+				if ( elem.nodeType === 1 && !dataPriv.get( elem, "hasDataAttrs" ) ) {
+					i = attrs.length;
+					while ( i-- ) {
+
+						// Support: IE 11 only
+						// The attrs elements can be null (#14894)
+						if ( attrs[ i ] ) {
+							name = attrs[ i ].name;
+							if ( name.indexOf( "data-" ) === 0 ) {
+								name = camelCase( name.slice( 5 ) );
+								dataAttr( elem, name, data[ name ] );
+							}
+						}
+					}
+					dataPriv.set( elem, "hasDataAttrs", true );
+				}
+			}
+
+			return data;
+		}
+
+		// Sets multiple values
+		if ( typeof key === "object" ) {
+			return this.each( function() {
+				dataUser.set( this, key );
+			} );
+		}
+
+		return access( this, function( value ) {
+			var data;
+
+			// The calling jQuery object (element matches) is not empty
+			// (and therefore has an element appears at this[ 0 ]) and the
+			// `value` parameter was not undefined. An empty jQuery object
+			// will result in `undefined` for elem = this[ 0 ] which will
+			// throw an exception if an attempt to read a data cache is made.
+			if ( elem && value === undefined ) {
+
+				// Attempt to get data from the cache
+				// The key will always be camelCased in Data
+				data = dataUser.get( elem, key );
+				if ( data !== undefined ) {
+					return data;
+				}
+
+				// Attempt to "discover" the data in
+				// HTML5 custom data-* attrs
+				data = dataAttr( elem, key );
+				if ( data !== undefined ) {
+					return data;
+				}
+
+				// We tried really hard, but the data doesn't exist.
+				return;
+			}
+
+			// Set the data...
+			this.each( function() {
+
+				// We always store the camelCased key
+				dataUser.set( this, key, value );
+			} );
+		}, null, value, arguments.length > 1, null, true );
+	},
+
+	removeData: function( key ) {
+		return this.each( function() {
+			dataUser.remove( this, key );
+		} );
+	}
+} );
+
+
+jQuery.extend( {
+	queue: function( elem, type, data ) {
+		var queue;
+
+		if ( elem ) {
+			type = ( type || "fx" ) + "queue";
+			queue = dataPriv.get( elem, type );
+
+			// Speed up dequeue by getting out quickly if this is just a lookup
+			if ( data ) {
+				if ( !queue || Array.isArray( data ) ) {
+					queue = dataPriv.access( elem, type, jQuery.makeArray( data ) );
+				} else {
+					queue.push( data );
+				}
+			}
+			return queue || [];
+		}
+	},
+
+	dequeue: function( elem, type ) {
+		type = type || "fx";
+
+		var queue = jQuery.queue( elem, type ),
+			startLength = queue.length,
+			fn = queue.shift(),
+			hooks = jQuery._queueHooks( elem, type ),
+			next = function() {
+				jQuery.dequeue( elem, type );
+			};
+
+		// If the fx queue is dequeued, always remove the progress sentinel
+		if ( fn === "inprogress" ) {
+			fn = queue.shift();
+			startLength--;
+		}
+
+		if ( fn ) {
+
+			// Add a progress sentinel to prevent the fx queue from being
+			// automatically dequeued
+			if ( type === "fx" ) {
+				queue.unshift( "inprogress" );
+			}
+
+			// Clear up the last queue stop function
+			delete hooks.stop;
+			fn.call( elem, next, hooks );
+		}
+
+		if ( !startLength && hooks ) {
+			hooks.empty.fire();
+		}
+	},
+
+	// Not public - generate a queueHooks object, or return the current one
+	_queueHooks: function( elem, type ) {
+		var key = type + "queueHooks";
+		return dataPriv.get( elem, key ) || dataPriv.access( elem, key, {
+			empty: jQuery.Callbacks( "once memory" ).add( function() {
+				dataPriv.remove( elem, [ type + "queue", key ] );
+			} )
+		} );
+	}
+} );
+
+jQuery.fn.extend( {
+	queue: function( type, data ) {
+		var setter = 2;
+
+		if ( typeof type !== "string" ) {
+			data = type;
+			type = "fx";
+			setter--;
+		}
+
+		if ( arguments.length < setter ) {
+			return jQuery.queue( this[ 0 ], type );
+		}
+
+		return data === undefined ?
+			this :
+			this.each( function() {
+				var queue = jQuery.queue( this, type, data );
+
+				// Ensure a hooks for this queue
+				jQuery._queueHooks( this, type );
+
+				if ( type === "fx" && queue[ 0 ] !== "inprogress" ) {
+					jQuery.dequeue( this, type );
+				}
+			} );
+	},
+	dequeue: function( type ) {
+		return this.each( function() {
+			jQuery.dequeue( this, type );
+		} );
+	},
+	clearQueue: function( type ) {
+		return this.queue( type || "fx", [] );
+	},
+
+	// Get a promise resolved when queues of a certain type
+	// are emptied (fx is the type by default)
+	promise: function( type, obj ) {
+		var tmp,
+			count = 1,
+			defer = jQuery.Deferred(),
+			elements = this,
+			i = this.length,
+			resolve = function() {
+				if ( !( --count ) ) {
+					defer.resolveWith( elements, [ elements ] );
+				}
+			};
+
+		if ( typeof type !== "string" ) {
+			obj = type;
+			type = undefined;
+		}
+		type = type || "fx";
+
+		while ( i-- ) {
+			tmp = dataPriv.get( elements[ i ], type + "queueHooks" );
+			if ( tmp && tmp.empty ) {
+				count++;
+				tmp.empty.add( resolve );
+			}
+		}
+		resolve();
+		return defer.promise( obj );
+	}
+} );
+var pnum = ( /[+-]?(?:\d*\.|)\d+(?:[eE][+-]?\d+|)/ ).source;
+
+var rcssNum = new RegExp( "^(?:([+-])=|)(" + pnum + ")([a-z%]*)$", "i" );
+
+
+var cssExpand = [ "Top", "Right", "Bottom", "Left" ];
+
+var documentElement = document.documentElement;
+
+
+
+	var isAttached = function( elem ) {
+			return jQuery.contains( elem.ownerDocument, elem );
+		},
+		composed = { composed: true };
+
+	// Support: IE 9 - 11+, Edge 12 - 18+, iOS 10.0 - 10.2 only
+	// Check attachment across shadow DOM boundaries when possible (gh-3504)
+	// Support: iOS 10.0-10.2 only
+	// Early iOS 10 versions support `attachShadow` but not `getRootNode`,
+	// leading to errors. We need to check for `getRootNode`.
+	if ( documentElement.getRootNode ) {
+		isAttached = function( elem ) {
+			return jQuery.contains( elem.ownerDocument, elem ) ||
+				elem.getRootNode( composed ) === elem.ownerDocument;
+		};
+	}
+var isHiddenWithinTree = function( elem, el ) {
+
+		// isHiddenWithinTree might be called from jQuery#filter function;
+		// in that case, element will be second argument
+		elem = el || elem;
+
+		// Inline style trumps all
+		return elem.style.display === "none" ||
+			elem.style.display === "" &&
+
+			// Otherwise, check computed style
+			// Support: Firefox <=43 - 45
+			// Disconnected elements can have computed display: none, so first confirm that elem is
+			// in the document.
+			isAttached( elem ) &&
+
+			jQuery.css( elem, "display" ) === "none";
+	};
+
+
+
+function adjustCSS( elem, prop, valueParts, tween ) {
+	var adjusted, scale,
+		maxIterations = 20,
+		currentValue = tween ?
+			function() {
+				return tween.cur();
+			} :
+			function() {
+				return jQuery.css( elem, prop, "" );
+			},
+		initial = currentValue(),
+		unit = valueParts && valueParts[ 3 ] || ( jQuery.cssNumber[ prop ] ? "" : "px" ),
+
+		// Starting value computation is required for potential unit mismatches
+		initialInUnit = elem.nodeType &&
+			( jQuery.cssNumber[ prop ] || unit !== "px" && +initial ) &&
+			rcssNum.exec( jQuery.css( elem, prop ) );
+
+	if ( initialInUnit && initialInUnit[ 3 ] !== unit ) {
+
+		// Support: Firefox <=54
+		// Halve the iteration target value to prevent interference from CSS upper bounds (gh-2144)
+		initial = initial / 2;
+
+		// Trust units reported by jQuery.css
+		unit = unit || initialInUnit[ 3 ];
+
+		// Iteratively approximate from a nonzero starting point
+		initialInUnit = +initial || 1;
+
+		while ( maxIterations-- ) {
+
+			// Evaluate and update our best guess (doubling guesses that zero out).
+			// Finish if the scale equals or crosses 1 (making the old*new product non-positive).
+			jQuery.style( elem, prop, initialInUnit + unit );
+			if ( ( 1 - scale ) * ( 1 - ( scale = currentValue() / initial || 0.5 ) ) <= 0 ) {
+				maxIterations = 0;
+			}
+			initialInUnit = initialInUnit / scale;
+
+		}
+
+		initialInUnit = initialInUnit * 2;
+		jQuery.style( elem, prop, initialInUnit + unit );
+
+		// Make sure we update the tween properties later on
+		valueParts = valueParts || [];
+	}
+
+	if ( valueParts ) {
+		initialInUnit = +initialInUnit || +initial || 0;
+
+		// Apply relative offset (+=/-=) if specified
+		adjusted = valueParts[ 1 ] ?
+			initialInUnit + ( valueParts[ 1 ] + 1 ) * valueParts[ 2 ] :
+			+valueParts[ 2 ];
+		if ( tween ) {
+			tween.unit = unit;
+			tween.start = initialInUnit;
+			tween.end = adjusted;
+		}
+	}
+	return adjusted;
+}
+
+
+var defaultDisplayMap = {};
+
+function getDefaultDisplay( elem ) {
+	var temp,
+		doc = elem.ownerDocument,
+		nodeName = elem.nodeName,
+		display = defaultDisplayMap[ nodeName ];
+
+	if ( display ) {
+		return display;
+	}
+
+	temp = doc.body.appendChild( doc.createElement( nodeName ) );
+	display = jQuery.css( temp, "display" );
+
+	temp.parentNode.removeChild( temp );
+
+	if ( display === "none" ) {
+		display = "block";
+	}
+	defaultDisplayMap[ nodeName ] = display;
+
+	return display;
+}
+
+function showHide( elements, show ) {
+	var display, elem,
+		values = [],
+		index = 0,
+		length = elements.length;
+
+	// Determine new display value for elements that need to change
+	for ( ; index < length; index++ ) {
+		elem = elements[ index ];
+		if ( !elem.style ) {
+			continue;
+		}
+
+		display = elem.style.display;
+		if ( show ) {
+
+			// Since we force visibility upon cascade-hidden elements, an immediate (and slow)
+			// check is required in this first loop unless we have a nonempty display value (either
+			// inline or about-to-be-restored)
+			if ( display === "none" ) {
+				values[ index ] = dataPriv.get( elem, "display" ) || null;
+				if ( !values[ index ] ) {
+					elem.style.display = "";
+				}
+			}
+			if ( elem.style.display === "" && isHiddenWithinTree( elem ) ) {
+				values[ index ] = getDefaultDisplay( elem );
+			}
+		} else {
+			if ( display !== "none" ) {
+				values[ index ] = "none";
+
+				// Remember what we're overwriting
+				dataPriv.set( elem, "display", display );
+			}
+		}
+	}
+
+	// Set the display of the elements in a second loop to avoid constant reflow
+	for ( index = 0; index < length; index++ ) {
+		if ( values[ index ] != null ) {
+			elements[ index ].style.display = values[ index ];
+		}
+	}
+
+	return elements;
+}
+
+jQuery.fn.extend( {
+	show: function() {
+		return showHide( this, true );
+	},
+	hide: function() {
+		return showHide( this );
+	},
+	toggle: function( state ) {
+		if ( typeof state === "boolean" ) {
+			return state ? this.show() : this.hide();
+		}
+
+		return this.each( function() {
+			if ( isHiddenWithinTree( this ) ) {
+				jQuery( this ).show();
+			} else {
+				jQuery( this ).hide();
+			}
+		} );
+	}
+} );
+var rcheckableType = ( /^(?:checkbox|radio)$/i );
+
+var rtagName = ( /<([a-z][^\/\0>\x20\t\r\n\f]*)/i );
+
+var rscriptType = ( /^$|^module$|\/(?:java|ecma)script/i );
+
+
+
+( function() {
+	var fragment = document.createDocumentFragment(),
+		div = fragment.appendChild( document.createElement( "div" ) ),
+		input = document.createElement( "input" );
+
+	// Support: Android 4.0 - 4.3 only
+	// Check state lost if the name is set (#11217)
+	// Support: Windows Web Apps (WWA)
+	// `name` and `type` must use .setAttribute for WWA (#14901)
+	input.setAttribute( "type", "radio" );
+	input.setAttribute( "checked", "checked" );
+	input.setAttribute( "name", "t" );
+
+	div.appendChild( input );
+
+	// Support: Android <=4.1 only
+	// Older WebKit doesn't clone checked state correctly in fragments
+	support.checkClone = div.cloneNode( true ).cloneNode( true ).lastChild.checked;
+
+	// Support: IE <=11 only
+	// Make sure textarea (and checkbox) defaultValue is properly cloned
+	div.innerHTML = "<textarea>x</textarea>";
+	support.noCloneChecked = !!div.cloneNode( true ).lastChild.defaultValue;
+
+	// Support: IE <=9 only
+	// IE <=9 replaces <option> tags with their contents when inserted outside of
+	// the select element.
+	div.innerHTML = "<option></option>";
+	support.option = !!div.lastChild;
+} )();
+
+
+// We have to close these tags to support XHTML (#13200)
+var wrapMap = {
+
+	// XHTML parsers do not magically insert elements in the
+	// same way that tag soup parsers do. So we cannot shorten
+	// this by omitting <tbody> or other required elements.
+	thead: [ 1, "<table>", "</table>" ],
+	col: [ 2, "<table><colgroup>", "</colgroup></table>" ],
+	tr: [ 2, "<table><tbody>", "</tbody></table>" ],
+	td: [ 3, "<table><tbody><tr>", "</tr></tbody></table>" ],
+
+	_default: [ 0, "", "" ]
+};
+
+wrapMap.tbody = wrapMap.tfoot = wrapMap.colgroup = wrapMap.caption = wrapMap.thead;
+wrapMap.th = wrapMap.td;
+
+// Support: IE <=9 only
+if ( !support.option ) {
+	wrapMap.optgroup = wrapMap.option = [ 1, "<select multiple='multiple'>", "</select>" ];
+}
+
+
+function getAll( context, tag ) {
+
+	// Support: IE <=9 - 11 only
+	// Use typeof to avoid zero-argument method invocation on host objects (#15151)
+	var ret;
+
+	if ( typeof context.getElementsByTagName !== "undefined" ) {
+		ret = context.getElementsByTagName( tag || "*" );
+
+	} else if ( typeof context.querySelectorAll !== "undefined" ) {
+		ret = context.querySelectorAll( tag || "*" );
+
+	} else {
+		ret = [];
+	}
+
+	if ( tag === undefined || tag && nodeName( context, tag ) ) {
+		return jQuery.merge( [ context ], ret );
+	}
+
+	return ret;
+}
+
+
+// Mark scripts as having already been evaluated
+function setGlobalEval( elems, refElements ) {
+	var i = 0,
+		l = elems.length;
+
+	for ( ; i < l; i++ ) {
+		dataPriv.set(
+			elems[ i ],
+			"globalEval",
+			!refElements || dataPriv.get( refElements[ i ], "globalEval" )
+		);
+	}
+}
+
+
+var rhtml = /<|&#?\w+;/;
+
+function buildFragment( elems, context, scripts, selection, ignored ) {
+	var elem, tmp, tag, wrap, attached, j,
+		fragment = context.createDocumentFragment(),
+		nodes = [],
+		i = 0,
+		l = elems.length;
+
+	for ( ; i < l; i++ ) {
+		elem = elems[ i ];
+
+		if ( elem || elem === 0 ) {
+
+			// Add nodes directly
+			if ( toType( elem ) === "object" ) {
+
+				// Support: Android <=4.0 only, PhantomJS 1 only
+				// push.apply(_, arraylike) throws on ancient WebKit
+				jQuery.merge( nodes, elem.nodeType ? [ elem ] : elem );
+
+			// Convert non-html into a text node
+			} else if ( !rhtml.test( elem ) ) {
+				nodes.push( context.createTextNode( elem ) );
+
+			// Convert html into DOM nodes
+			} else {
+				tmp = tmp || fragment.appendChild( context.createElement( "div" ) );
+
+				// Deserialize a standard representation
+				tag = ( rtagName.exec( elem ) || [ "", "" ] )[ 1 ].toLowerCase();
+				wrap = wrapMap[ tag ] || wrapMap._default;
+				tmp.innerHTML = wrap[ 1 ] + jQuery.htmlPrefilter( elem ) + wrap[ 2 ];
+
+				// Descend through wrappers to the right content
+				j = wrap[ 0 ];
+				while ( j-- ) {
+					tmp = tmp.lastChild;
+				}
+
+				// Support: Android <=4.0 only, PhantomJS 1 only
+				// push.apply(_, arraylike) throws on ancient WebKit
+				jQuery.merge( nodes, tmp.childNodes );
+
+				// Remember the top-level container
+				tmp = fragment.firstChild;
+
+				// Ensure the created nodes are orphaned (#12392)
+				tmp.textContent = "";
+			}
+		}
+	}
+
+	// Remove wrapper from fragment
+	fragment.textContent = "";
+
+	i = 0;
+	while ( ( elem = nodes[ i++ ] ) ) {
+
+		// Skip elements already in the context collection (trac-4087)
+		if ( selection && jQuery.inArray( elem, selection ) > -1 ) {
+			if ( ignored ) {
+				ignored.push( elem );
+			}
+			continue;
+		}
+
+		attached = isAttached( elem );
+
+		// Append to fragment
+		tmp = getAll( fragment.appendChild( elem ), "script" );
+
+		// Preserve script evaluation history
+		if ( attached ) {
+			setGlobalEval( tmp );
+		}
+
+		// Capture executables
+		if ( scripts ) {
+			j = 0;
+			while ( ( elem = tmp[ j++ ] ) ) {
+				if ( rscriptType.test( elem.type || "" ) ) {
+					scripts.push( elem );
+				}
+			}
+		}
+	}
+
+	return fragment;
+}
+
+
+var rtypenamespace = /^([^.]*)(?:\.(.+)|)/;
+
+function returnTrue() {
+	return true;
+}
+
+function returnFalse() {
+	return false;
+}
+
+// Support: IE <=9 - 11+
+// focus() and blur() are asynchronous, except when they are no-op.
+// So expect focus to be synchronous when the element is already active,
+// and blur to be synchronous when the element is not already active.
+// (focus and blur are always synchronous in other supported browsers,
+// this just defines when we can count on it).
+function expectSync( elem, type ) {
+	return ( elem === safeActiveElement() ) === ( type === "focus" );
+}
+
+// Support: IE <=9 only
+// Accessing document.activeElement can throw unexpectedly
+// https://bugs.jquery.com/ticket/13393
+function safeActiveElement() {
+	try {
+		return document.activeElement;
+	} catch ( err ) { }
+}
+
+function on( elem, types, selector, data, fn, one ) {
+	var origFn, type;
+
+	// Types can be a map of types/handlers
+	if ( typeof types === "object" ) {
+
+		// ( types-Object, selector, data )
+		if ( typeof selector !== "string" ) {
+
+			// ( types-Object, data )
+			data = data || selector;
+			selector = undefined;
+		}
+		for ( type in types ) {
+			on( elem, type, selector, data, types[ type ], one );
+		}
+		return elem;
+	}
+
+	if ( data == null && fn == null ) {
+
+		// ( types, fn )
+		fn = selector;
+		data = selector = undefined;
+	} else if ( fn == null ) {
+		if ( typeof selector === "string" ) {
+
+			// ( types, selector, fn )
+			fn = data;
+			data = undefined;
+		} else {
+
+			// ( types, data, fn )
+			fn = data;
+			data = selector;
+			selector = undefined;
+		}
+	}
+	if ( fn === false ) {
+		fn = returnFalse;
+	} else if ( !fn ) {
+		return elem;
+	}
+
+	if ( one === 1 ) {
+		origFn = fn;
+		fn = function( event ) {
+
+			// Can use an empty set, since event contains the info
+			jQuery().off( event );
+			return origFn.apply( this, arguments );
+		};
+
+		// Use same guid so caller can remove using origFn
+		fn.guid = origFn.guid || ( origFn.guid = jQuery.guid++ );
+	}
+	return elem.each( function() {
+		jQuery.event.add( this, types, fn, data, selector );
+	} );
+}
+
+/*
+ * Helper functions for managing events -- not part of the public interface.
+ * Props to Dean Edwards' addEvent library for many of the ideas.
+ */
+jQuery.event = {
+
+	global: {},
+
+	add: function( elem, types, handler, data, selector ) {
+
+		var handleObjIn, eventHandle, tmp,
+			events, t, handleObj,
+			special, handlers, type, namespaces, origType,
+			elemData = dataPriv.get( elem );
+
+		// Only attach events to objects that accept data
+		if ( !acceptData( elem ) ) {
+			return;
+		}
+
+		// Caller can pass in an object of custom data in lieu of the handler
+		if ( handler.handler ) {
+			handleObjIn = handler;
+			handler = handleObjIn.handler;
+			selector = handleObjIn.selector;
+		}
+
+		// Ensure that invalid selectors throw exceptions at attach time
+		// Evaluate against documentElement in case elem is a non-element node (e.g., document)
+		if ( selector ) {
+			jQuery.find.matchesSelector( documentElement, selector );
+		}
+
+		// Make sure that the handler has a unique ID, used to find/remove it later
+		if ( !handler.guid ) {
+			handler.guid = jQuery.guid++;
+		}
+
+		// Init the element's event structure and main handler, if this is the first
+		if ( !( events = elemData.events ) ) {
+			events = elemData.events = Object.create( null );
+		}
+		if ( !( eventHandle = elemData.handle ) ) {
+			eventHandle = elemData.handle = function( e ) {
+
+				// Discard the second event of a jQuery.event.trigger() and
+				// when an event is called after a page has unloaded
+				return typeof jQuery !== "undefined" && jQuery.event.triggered !== e.type ?
+					jQuery.event.dispatch.apply( elem, arguments ) : undefined;
+			};
+		}
+
+		// Handle multiple events separated by a space
+		types = ( types || "" ).match( rnothtmlwhite ) || [ "" ];
+		t = types.length;
+		while ( t-- ) {
+			tmp = rtypenamespace.exec( types[ t ] ) || [];
+			type = origType = tmp[ 1 ];
+			namespaces = ( tmp[ 2 ] || "" ).split( "." ).sort();
+
+			// There *must* be a type, no attaching namespace-only handlers
+			if ( !type ) {
+				continue;
+			}
+
+			// If event changes its type, use the special event handlers for the changed type
+			special = jQuery.event.special[ type ] || {};
+
+			// If selector defined, determine special event api type, otherwise given type
+			type = ( selector ? special.delegateType : special.bindType ) || type;
+
+			// Update special based on newly reset type
+			special = jQuery.event.special[ type ] || {};
+
+			// handleObj is passed to all event handlers
+			handleObj = jQuery.extend( {
+				type: type,
+				origType: origType,
+				data: data,
+				handler: handler,
+				guid: handler.guid,
+				selector: selector,
+				needsContext: selector && jQuery.expr.match.needsContext.test( selector ),
+				namespace: namespaces.join( "." )
+			}, handleObjIn );
+
+			// Init the event handler queue if we're the first
+			if ( !( handlers = events[ type ] ) ) {
+				handlers = events[ type ] = [];
+				handlers.delegateCount = 0;
+
+				// Only use addEventListener if the special events handler returns false
+				if ( !special.setup ||
+					special.setup.call( elem, data, namespaces, eventHandle ) === false ) {
+
+					if ( elem.addEventListener ) {
+						elem.addEventListener( type, eventHandle );
+					}
+				}
+			}
+
+			if ( special.add ) {
+				special.add.call( elem, handleObj );
+
+				if ( !handleObj.handler.guid ) {
+					handleObj.handler.guid = handler.guid;
+				}
+			}
+
+			// Add to the element's handler list, delegates in front
+			if ( selector ) {
+				handlers.splice( handlers.delegateCount++, 0, handleObj );
+			} else {
+				handlers.push( handleObj );
+			}
+
+			// Keep track of which events have ever been used, for event optimization
+			jQuery.event.global[ type ] = true;
+		}
+
+	},
+
+	// Detach an event or set of events from an element
+	remove: function( elem, types, handler, selector, mappedTypes ) {
+
+		var j, origCount, tmp,
+			events, t, handleObj,
+			special, handlers, type, namespaces, origType,
+			elemData = dataPriv.hasData( elem ) && dataPriv.get( elem );
+
+		if ( !elemData || !( events = elemData.events ) ) {
+			return;
+		}
+
+		// Once for each type.namespace in types; type may be omitted
+		types = ( types || "" ).match( rnothtmlwhite ) || [ "" ];
+		t = types.length;
+		while ( t-- ) {
+			tmp = rtypenamespace.exec( types[ t ] ) || [];
+			type = origType = tmp[ 1 ];
+			namespaces = ( tmp[ 2 ] || "" ).split( "." ).sort();
+
+			// Unbind all events (on this namespace, if provided) for the element
+			if ( !type ) {
+				for ( type in events ) {
+					jQuery.event.remove( elem, type + types[ t ], handler, selector, true );
+				}
+				continue;
+			}
+
+			special = jQuery.event.special[ type ] || {};
+			type = ( selector ? special.delegateType : special.bindType ) || type;
+			handlers = events[ type ] || [];
+			tmp = tmp[ 2 ] &&
+				new RegExp( "(^|\\.)" + namespaces.join( "\\.(?:.*\\.|)" ) + "(\\.|$)" );
+
+			// Remove matching events
+			origCount = j = handlers.length;
+			while ( j-- ) {
+				handleObj = handlers[ j ];
+
+				if ( ( mappedTypes || origType === handleObj.origType ) &&
+					( !handler || handler.guid === handleObj.guid ) &&
+					( !tmp || tmp.test( handleObj.namespace ) ) &&
+					( !selector || selector === handleObj.selector ||
+						selector === "**" && handleObj.selector ) ) {
+					handlers.splice( j, 1 );
+
+					if ( handleObj.selector ) {
+						handlers.delegateCount--;
+					}
+					if ( special.remove ) {
+						special.remove.call( elem, handleObj );
+					}
+				}
+			}
+
+			// Remove generic event handler if we removed something and no more handlers exist
+			// (avoids potential for endless recursion during removal of special event handlers)
+			if ( origCount && !handlers.length ) {
+				if ( !special.teardown ||
+					special.teardown.call( elem, namespaces, elemData.handle ) === false ) {
+
+					jQuery.removeEvent( elem, type, elemData.handle );
+				}
+
+				delete events[ type ];
+			}
+		}
+
+		// Remove data and the expando if it's no longer used
+		if ( jQuery.isEmptyObject( events ) ) {
+			dataPriv.remove( elem, "handle events" );
+		}
+	},
+
+	dispatch: function( nativeEvent ) {
+
+		var i, j, ret, matched, handleObj, handlerQueue,
+			args = new Array( arguments.length ),
+
+			// Make a writable jQuery.Event from the native event object
+			event = jQuery.event.fix( nativeEvent ),
+
+			handlers = (
+				dataPriv.get( this, "events" ) || Object.create( null )
+			)[ event.type ] || [],
+			special = jQuery.event.special[ event.type ] || {};
+
+		// Use the fix-ed jQuery.Event rather than the (read-only) native event
+		args[ 0 ] = event;
+
+		for ( i = 1; i < arguments.length; i++ ) {
+			args[ i ] = arguments[ i ];
+		}
+
+		event.delegateTarget = this;
+
+		// Call the preDispatch hook for the mapped type, and let it bail if desired
+		if ( special.preDispatch && special.preDispatch.call( this, event ) === false ) {
+			return;
+		}
+
+		// Determine handlers
+		handlerQueue = jQuery.event.handlers.call( this, event, handlers );
+
+		// Run delegates first; they may want to stop propagation beneath us
+		i = 0;
+		while ( ( matched = handlerQueue[ i++ ] ) && !event.isPropagationStopped() ) {
+			event.currentTarget = matched.elem;
+
+			j = 0;
+			while ( ( handleObj = matched.handlers[ j++ ] ) &&
+				!event.isImmediatePropagationStopped() ) {
+
+				// If the event is namespaced, then each handler is only invoked if it is
+				// specially universal or its namespaces are a superset of the event's.
+				if ( !event.rnamespace || handleObj.namespace === false ||
+					event.rnamespace.test( handleObj.namespace ) ) {
+
+					event.handleObj = handleObj;
+					event.data = handleObj.data;
+
+					ret = ( ( jQuery.event.special[ handleObj.origType ] || {} ).handle ||
+						handleObj.handler ).apply( matched.elem, args );
+
+					if ( ret !== undefined ) {
+						if ( ( event.result = ret ) === false ) {
+							event.preventDefault();
+							event.stopPropagation();
+						}
+					}
+				}
+			}
+		}
+
+		// Call the postDispatch hook for the mapped type
+		if ( special.postDispatch ) {
+			special.postDispatch.call( this, event );
+		}
+
+		return event.result;
+	},
+
+	handlers: function( event, handlers ) {
+		var i, handleObj, sel, matchedHandlers, matchedSelectors,
+			handlerQueue = [],
+			delegateCount = handlers.delegateCount,
+			cur = event.target;
+
+		// Find delegate handlers
+		if ( delegateCount &&
+
+			// Support: IE <=9
+			// Black-hole SVG <use> instance trees (trac-13180)
+			cur.nodeType &&
+
+			// Support: Firefox <=42
+			// Suppress spec-violating clicks indicating a non-primary pointer button (trac-3861)
+			// https://www.w3.org/TR/DOM-Level-3-Events/#event-type-click
+			// Support: IE 11 only
+			// ...but not arrow key "clicks" of radio inputs, which can have `button` -1 (gh-2343)
+			!( event.type === "click" && event.button >= 1 ) ) {
+
+			for ( ; cur !== this; cur = cur.parentNode || this ) {
+
+				// Don't check non-elements (#13208)
+				// Don't process clicks on disabled elements (#6911, #8165, #11382, #11764)
+				if ( cur.nodeType === 1 && !( event.type === "click" && cur.disabled === true ) ) {
+					matchedHandlers = [];
+					matchedSelectors = {};
+					for ( i = 0; i < delegateCount; i++ ) {
+						handleObj = handlers[ i ];
+
+						// Don't conflict with Object.prototype properties (#13203)
+						sel = handleObj.selector + " ";
+
+						if ( matchedSelectors[ sel ] === undefined ) {
+							matchedSelectors[ sel ] = handleObj.needsContext ?
+								jQuery( sel, this ).index( cur ) > -1 :
+								jQuery.find( sel, this, null, [ cur ] ).length;
+						}
+						if ( matchedSelectors[ sel ] ) {
+							matchedHandlers.push( handleObj );
+						}
+					}
+					if ( matchedHandlers.length ) {
+						handlerQueue.push( { elem: cur, handlers: matchedHandlers } );
+					}
+				}
+			}
+		}
+
+		// Add the remaining (directly-bound) handlers
+		cur = this;
+		if ( delegateCount < handlers.length ) {
+			handlerQueue.push( { elem: cur, handlers: handlers.slice( delegateCount ) } );
+		}
+
+		return handlerQueue;
+	},
+
+	addProp: function( name, hook ) {
+		Object.defineProperty( jQuery.Event.prototype, name, {
+			enumerable: true,
+			configurable: true,
+
+			get: isFunction( hook ) ?
+				function() {
+					if ( this.originalEvent ) {
+						return hook( this.originalEvent );
+					}
+				} :
+				function() {
+					if ( this.originalEvent ) {
+						return this.originalEvent[ name ];
+					}
+				},
+
+			set: function( value ) {
+				Object.defineProperty( this, name, {
+					enumerable: true,
+					configurable: true,
+					writable: true,
+					value: value
+				} );
+			}
+		} );
+	},
+
+	fix: function( originalEvent ) {
+		return originalEvent[ jQuery.expando ] ?
+			originalEvent :
+			new jQuery.Event( originalEvent );
+	},
+
+	special: {
+		load: {
+
+			// Prevent triggered image.load events from bubbling to window.load
+			noBubble: true
+		},
+		click: {
+
+			// Utilize native event to ensure correct state for checkable inputs
+			setup: function( data ) {
+
+				// For mutual compressibility with _default, replace `this` access with a local var.
+				// `|| data` is dead code meant only to preserve the variable through minification.
+				var el = this || data;
+
+				// Claim the first handler
+				if ( rcheckableType.test( el.type ) &&
+					el.click && nodeName( el, "input" ) ) {
+
+					// dataPriv.set( el, "click", ... )
+					leverageNative( el, "click", returnTrue );
+				}
+
+				// Return false to allow normal processing in the caller
+				return false;
+			},
+			trigger: function( data ) {
+
+				// For mutual compressibility with _default, replace `this` access with a local var.
+				// `|| data` is dead code meant only to preserve the variable through minification.
+				var el = this || data;
+
+				// Force setup before triggering a click
+				if ( rcheckableType.test( el.type ) &&
+					el.click && nodeName( el, "input" ) ) {
+
+					leverageNative( el, "click" );
+				}
+
+				// Return non-false to allow normal event-path propagation
+				return true;
+			},
+
+			// For cross-browser consistency, suppress native .click() on links
+			// Also prevent it if we're currently inside a leveraged native-event stack
+			_default: function( event ) {
+				var target = event.target;
+				return rcheckableType.test( target.type ) &&
+					target.click && nodeName( target, "input" ) &&
+					dataPriv.get( target, "click" ) ||
+					nodeName( target, "a" );
+			}
+		},
+
+		beforeunload: {
+			postDispatch: function( event ) {
+
+				// Support: Firefox 20+
+				// Firefox doesn't alert if the returnValue field is not set.
+				if ( event.result !== undefined && event.originalEvent ) {
+					event.originalEvent.returnValue = event.result;
+				}
+			}
+		}
+	}
+};
+
+// Ensure the presence of an event listener that handles manually-triggered
+// synthetic events by interrupting progress until reinvoked in response to
+// *native* events that it fires directly, ensuring that state changes have
+// already occurred before other listeners are invoked.
+function leverageNative( el, type, expectSync ) {
+
+	// Missing expectSync indicates a trigger call, which must force setup through jQuery.event.add
+	if ( !expectSync ) {
+		if ( dataPriv.get( el, type ) === undefined ) {
+			jQuery.event.add( el, type, returnTrue );
+		}
+		return;
+	}
+
+	// Register the controller as a special universal handler for all event namespaces
+	dataPriv.set( el, type, false );
+	jQuery.event.add( el, type, {
+		namespace: false,
+		handler: function( event ) {
+			var notAsync, result,
+				saved = dataPriv.get( this, type );
+
+			if ( ( event.isTrigger & 1 ) && this[ type ] ) {
+
+				// Interrupt processing of the outer synthetic .trigger()ed event
+				// Saved data should be false in such cases, but might be a leftover capture object
+				// from an async native handler (gh-4350)
+				if ( !saved.length ) {
+
+					// Store arguments for use when handling the inner native event
+					// There will always be at least one argument (an event object), so this array
+					// will not be confused with a leftover capture object.
+					saved = slice.call( arguments );
+					dataPriv.set( this, type, saved );
+
+					// Trigger the native event and capture its result
+					// Support: IE <=9 - 11+
+					// focus() and blur() are asynchronous
+					notAsync = expectSync( this, type );
+					this[ type ]();
+					result = dataPriv.get( this, type );
+					if ( saved !== result || notAsync ) {
+						dataPriv.set( this, type, false );
+					} else {
+						result = {};
+					}
+					if ( saved !== result ) {
+
+						// Cancel the outer synthetic event
+						event.stopImmediatePropagation();
+						event.preventDefault();
+
+						// Support: Chrome 86+
+						// In Chrome, if an element having a focusout handler is blurred by
+						// clicking outside of it, it invokes the handler synchronously. If
+						// that handler calls `.remove()` on the element, the data is cleared,
+						// leaving `result` undefined. We need to guard against this.
+						return result && result.value;
+					}
+
+				// If this is an inner synthetic event for an event with a bubbling surrogate
+				// (focus or blur), assume that the surrogate already propagated from triggering the
+				// native event and prevent that from happening again here.
+				// This technically gets the ordering wrong w.r.t. to `.trigger()` (in which the
+				// bubbling surrogate propagates *after* the non-bubbling base), but that seems
+				// less bad than duplication.
+				} else if ( ( jQuery.event.special[ type ] || {} ).delegateType ) {
+					event.stopPropagation();
+				}
+
+			// If this is a native event triggered above, everything is now in order
+			// Fire an inner synthetic event with the original arguments
+			} else if ( saved.length ) {
+
+				// ...and capture the result
+				dataPriv.set( this, type, {
+					value: jQuery.event.trigger(
+
+						// Support: IE <=9 - 11+
+						// Extend with the prototype to reset the above stopImmediatePropagation()
+						jQuery.extend( saved[ 0 ], jQuery.Event.prototype ),
+						saved.slice( 1 ),
+						this
+					)
+				} );
+
+				// Abort handling of the native event
+				event.stopImmediatePropagation();
+			}
+		}
+	} );
+}
+
+jQuery.removeEvent = function( elem, type, handle ) {
+
+	// This "if" is needed for plain objects
+	if ( elem.removeEventListener ) {
+		elem.removeEventListener( type, handle );
+	}
+};
+
+jQuery.Event = function( src, props ) {
+
+	// Allow instantiation without the 'new' keyword
+	if ( !( this instanceof jQuery.Event ) ) {
+		return new jQuery.Event( src, props );
+	}
+
+	// Event object
+	if ( src && src.type ) {
+		this.originalEvent = src;
+		this.type = src.type;
+
+		// Events bubbling up the document may have been marked as prevented
+		// by a handler lower down the tree; reflect the correct value.
+		this.isDefaultPrevented = src.defaultPrevented ||
+				src.defaultPrevented === undefined &&
+
+				// Support: Android <=2.3 only
+				src.returnValue === false ?
+			returnTrue :
+			returnFalse;
+
+		// Create target properties
+		// Support: Safari <=6 - 7 only
+		// Target should not be a text node (#504, #13143)
+		this.target = ( src.target && src.target.nodeType === 3 ) ?
+			src.target.parentNode :
+			src.target;
+
+		this.currentTarget = src.currentTarget;
+		this.relatedTarget = src.relatedTarget;
+
+	// Event type
+	} else {
+		this.type = src;
+	}
+
+	// Put explicitly provided properties onto the event object
+	if ( props ) {
+		jQuery.extend( this, props );
+	}
+
+	// Create a timestamp if incoming event doesn't have one
+	this.timeStamp = src && src.timeStamp || Date.now();
+
+	// Mark it as fixed
+	this[ jQuery.expando ] = true;
+};
+
+// jQuery.Event is based on DOM3 Events as specified by the ECMAScript Language Binding
+// https://www.w3.org/TR/2003/WD-DOM-Level-3-Events-20030331/ecma-script-binding.html
+jQuery.Event.prototype = {
+	constructor: jQuery.Event,
+	isDefaultPrevented: returnFalse,
+	isPropagationStopped: returnFalse,
+	isImmediatePropagationStopped: returnFalse,
+	isSimulated: false,
+
+	preventDefault: function() {
+		var e = this.originalEvent;
+
+		this.isDefaultPrevented = returnTrue;
+
+		if ( e && !this.isSimulated ) {
+			e.preventDefault();
+		}
+	},
+	stopPropagation: function() {
+		var e = this.originalEvent;
+
+		this.isPropagationStopped = returnTrue;
+
+		if ( e && !this.isSimulated ) {
+			e.stopPropagation();
+		}
+	},
+	stopImmediatePropagation: function() {
+		var e = this.originalEvent;
+
+		this.isImmediatePropagationStopped = returnTrue;
+
+		if ( e && !this.isSimulated ) {
+			e.stopImmediatePropagation();
+		}
+
+		this.stopPropagation();
+	}
+};
+
+// Includes all common event props including KeyEvent and MouseEvent specific props
+jQuery.each( {
+	altKey: true,
+	bubbles: true,
+	cancelable: true,
+	changedTouches: true,
+	ctrlKey: true,
+	detail: true,
+	eventPhase: true,
+	metaKey: true,
+	pageX: true,
+	pageY: true,
+	shiftKey: true,
+	view: true,
+	"char": true,
+	code: true,
+	charCode: true,
+	key: true,
+	keyCode: true,
+	button: true,
+	buttons: true,
+	clientX: true,
+	clientY: true,
+	offsetX: true,
+	offsetY: true,
+	pointerId: true,
+	pointerType: true,
+	screenX: true,
+	screenY: true,
+	targetTouches: true,
+	toElement: true,
+	touches: true,
+	which: true
+}, jQuery.event.addProp );
+
+jQuery.each( { focus: "focusin", blur: "focusout" }, function( type, delegateType ) {
+	jQuery.event.special[ type ] = {
+
+		// Utilize native event if possible so blur/focus sequence is correct
+		setup: function() {
+
+			// Claim the first handler
+			// dataPriv.set( this, "focus", ... )
+			// dataPriv.set( this, "blur", ... )
+			leverageNative( this, type, expectSync );
+
+			// Return false to allow normal processing in the caller
+			return false;
+		},
+		trigger: function() {
+
+			// Force setup before trigger
+			leverageNative( this, type );
+
+			// Return non-false to allow normal event-path propagation
+			return true;
+		},
+
+		// Suppress native focus or blur as it's already being fired
+		// in leverageNative.
+		_default: function() {
+			return true;
+		},
+
+		delegateType: delegateType
+	};
+} );
+
+// Create mouseenter/leave events using mouseover/out and event-time checks
+// so that event delegation works in jQuery.
+// Do the same for pointerenter/pointerleave and pointerover/pointerout
+//
+// Support: Safari 7 only
+// Safari sends mouseenter too often; see:
+// https://bugs.chromium.org/p/chromium/issues/detail?id=470258
+// for the description of the bug (it existed in older Chrome versions as well).
+jQuery.each( {
+	mouseenter: "mouseover",
+	mouseleave: "mouseout",
+	pointerenter: "pointerover",
+	pointerleave: "pointerout"
+}, function( orig, fix ) {
+	jQuery.event.special[ orig ] = {
+		delegateType: fix,
+		bindType: fix,
+
+		handle: function( event ) {
+			var ret,
+				target = this,
+				related = event.relatedTarget,
+				handleObj = event.handleObj;
+
+			// For mouseenter/leave call the handler if related is outside the target.
+			// NB: No relatedTarget if the mouse left/entered the browser window
+			if ( !related || ( related !== target && !jQuery.contains( target, related ) ) ) {
+				event.type = handleObj.origType;
+				ret = handleObj.handler.apply( this, arguments );
+				event.type = fix;
+			}
+			return ret;
+		}
+	};
+} );
+
+jQuery.fn.extend( {
+
+	on: function( types, selector, data, fn ) {
+		return on( this, types, selector, data, fn );
+	},
+	one: function( types, selector, data, fn ) {
+		return on( this, types, selector, data, fn, 1 );
+	},
+	off: function( types, selector, fn ) {
+		var handleObj, type;
+		if ( types && types.preventDefault && types.handleObj ) {
+
+			// ( event )  dispatched jQuery.Event
+			handleObj = types.handleObj;
+			jQuery( types.delegateTarget ).off(
+				handleObj.namespace ?
+					handleObj.origType + "." + handleObj.namespace :
+					handleObj.origType,
+				handleObj.selector,
+				handleObj.handler
+			);
+			return this;
+		}
+		if ( typeof types === "object" ) {
+
+			// ( types-object [, selector] )
+			for ( type in types ) {
+				this.off( type, selector, types[ type ] );
+			}
+			return this;
+		}
+		if ( selector === false || typeof selector === "function" ) {
+
+			// ( types [, fn] )
+			fn = selector;
+			selector = undefined;
+		}
+		if ( fn === false ) {
+			fn = returnFalse;
+		}
+		return this.each( function() {
+			jQuery.event.remove( this, types, fn, selector );
+		} );
+	}
+} );
+
+
+var
+
+	// Support: IE <=10 - 11, Edge 12 - 13 only
+	// In IE/Edge using regex groups here causes severe slowdowns.
+	// See https://connect.microsoft.com/IE/feedback/details/1736512/
+	rnoInnerhtml = /<script|<style|<link/i,
+
+	// checked="checked" or checked
+	rchecked = /checked\s*(?:[^=]|=\s*.checked.)/i,
+	rcleanScript = /^\s*<!(?:\[CDATA\[|--)|(?:\]\]|--)>\s*$/g;
+
+// Prefer a tbody over its parent table for containing new rows
+function manipulationTarget( elem, content ) {
+	if ( nodeName( elem, "table" ) &&
+		nodeName( content.nodeType !== 11 ? content : content.firstChild, "tr" ) ) {
+
+		return jQuery( elem ).children( "tbody" )[ 0 ] || elem;
+	}
+
+	return elem;
+}
+
+// Replace/restore the type attribute of script elements for safe DOM manipulation
+function disableScript( elem ) {
+	elem.type = ( elem.getAttribute( "type" ) !== null ) + "/" + elem.type;
+	return elem;
+}
+function restoreScript( elem ) {
+	if ( ( elem.type || "" ).slice( 0, 5 ) === "true/" ) {
+		elem.type = elem.type.slice( 5 );
+	} else {
+		elem.removeAttribute( "type" );
+	}
+
+	return elem;
+}
+
+function cloneCopyEvent( src, dest ) {
+	var i, l, type, pdataOld, udataOld, udataCur, events;
+
+	if ( dest.nodeType !== 1 ) {
+		return;
+	}
+
+	// 1. Copy private data: events, handlers, etc.
+	if ( dataPriv.hasData( src ) ) {
+		pdataOld = dataPriv.get( src );
+		events = pdataOld.events;
+
+		if ( events ) {
+			dataPriv.remove( dest, "handle events" );
+
+			for ( type in events ) {
+				for ( i = 0, l = events[ type ].length; i < l; i++ ) {
+					jQuery.event.add( dest, type, events[ type ][ i ] );
+				}
+			}
+		}
+	}
+
+	// 2. Copy user data
+	if ( dataUser.hasData( src ) ) {
+		udataOld = dataUser.access( src );
+		udataCur = jQuery.extend( {}, udataOld );
+
+		dataUser.set( dest, udataCur );
+	}
+}
+
+// Fix IE bugs, see support tests
+function fixInput( src, dest ) {
+	var nodeName = dest.nodeName.toLowerCase();
+
+	// Fails to persist the checked state of a cloned checkbox or radio button.
+	if ( nodeName === "input" && rcheckableType.test( src.type ) ) {
+		dest.checked = src.checked;
+
+	// Fails to return the selected option to the default selected state when cloning options
+	} else if ( nodeName === "input" || nodeName === "textarea" ) {
+		dest.defaultValue = src.defaultValue;
+	}
+}
+
+function domManip( collection, args, callback, ignored ) {
+
+	// Flatten any nested arrays
+	args = flat( args );
+
+	var fragment, first, scripts, hasScripts, node, doc,
+		i = 0,
+		l = collection.length,
+		iNoClone = l - 1,
+		value = args[ 0 ],
+		valueIsFunction = isFunction( value );
+
+	// We can't cloneNode fragments that contain checked, in WebKit
+	if ( valueIsFunction ||
+			( l > 1 && typeof value === "string" &&
+				!support.checkClone && rchecked.test( value ) ) ) {
+		return collection.each( function( index ) {
+			var self = collection.eq( index );
+			if ( valueIsFunction ) {
+				args[ 0 ] = value.call( this, index, self.html() );
+			}
+			domManip( self, args, callback, ignored );
+		} );
+	}
+
+	if ( l ) {
+		fragment = buildFragment( args, collection[ 0 ].ownerDocument, false, collection, ignored );
+		first = fragment.firstChild;
+
+		if ( fragment.childNodes.length === 1 ) {
+			fragment = first;
+		}
+
+		// Require either new content or an interest in ignored elements to invoke the callback
+		if ( first || ignored ) {
+			scripts = jQuery.map( getAll( fragment, "script" ), disableScript );
+			hasScripts = scripts.length;
+
+			// Use the original fragment for the last item
+			// instead of the first because it can end up
+			// being emptied incorrectly in certain situations (#8070).
+			for ( ; i < l; i++ ) {
+				node = fragment;
+
+				if ( i !== iNoClone ) {
+					node = jQuery.clone( node, true, true );
+
+					// Keep references to cloned scripts for later restoration
+					if ( hasScripts ) {
+
+						// Support: Android <=4.0 only, PhantomJS 1 only
+						// push.apply(_, arraylike) throws on ancient WebKit
+						jQuery.merge( scripts, getAll( node, "script" ) );
+					}
+				}
+
+				callback.call( collection[ i ], node, i );
+			}
+
+			if ( hasScripts ) {
+				doc = scripts[ scripts.length - 1 ].ownerDocument;
+
+				// Reenable scripts
+				jQuery.map( scripts, restoreScript );
+
+				// Evaluate executable scripts on first document insertion
+				for ( i = 0; i < hasScripts; i++ ) {
+					node = scripts[ i ];
+					if ( rscriptType.test( node.type || "" ) &&
+						!dataPriv.access( node, "globalEval" ) &&
+						jQuery.contains( doc, node ) ) {
+
+						if ( node.src && ( node.type || "" ).toLowerCase()  !== "module" ) {
+
+							// Optional AJAX dependency, but won't run scripts if not present
+							if ( jQuery._evalUrl && !node.noModule ) {
+								jQuery._evalUrl( node.src, {
+									nonce: node.nonce || node.getAttribute( "nonce" )
+								}, doc );
+							}
+						} else {
+							DOMEval( node.textContent.replace( rcleanScript, "" ), node, doc );
+						}
+					}
+				}
+			}
+		}
+	}
+
+	return collection;
+}
+
+function remove( elem, selector, keepData ) {
+	var node,
+		nodes = selector ? jQuery.filter( selector, elem ) : elem,
+		i = 0;
+
+	for ( ; ( node = nodes[ i ] ) != null; i++ ) {
+		if ( !keepData && node.nodeType === 1 ) {
+			jQuery.cleanData( getAll( node ) );
+		}
+
+		if ( node.parentNode ) {
+			if ( keepData && isAttached( node ) ) {
+				setGlobalEval( getAll( node, "script" ) );
+			}
+			node.parentNode.removeChild( node );
+		}
+	}
+
+	return elem;
+}
+
+jQuery.extend( {
+	htmlPrefilter: function( html ) {
+		return html;
+	},
+
+	clone: function( elem, dataAndEvents, deepDataAndEvents ) {
+		var i, l, srcElements, destElements,
+			clone = elem.cloneNode( true ),
+			inPage = isAttached( elem );
+
+		// Fix IE cloning issues
+		if ( !support.noCloneChecked && ( elem.nodeType === 1 || elem.nodeType === 11 ) &&
+				!jQuery.isXMLDoc( elem ) ) {
+
+			// We eschew Sizzle here for performance reasons: https://jsperf.com/getall-vs-sizzle/2
+			destElements = getAll( clone );
+			srcElements = getAll( elem );
+
+			for ( i = 0, l = srcElements.length; i < l; i++ ) {
+				fixInput( srcElements[ i ], destElements[ i ] );
+			}
+		}
+
+		// Copy the events from the original to the clone
+		if ( dataAndEvents ) {
+			if ( deepDataAndEvents ) {
+				srcElements = srcElements || getAll( elem );
+				destElements = destElements || getAll( clone );
+
+				for ( i = 0, l = srcElements.length; i < l; i++ ) {
+					cloneCopyEvent( srcElements[ i ], destElements[ i ] );
+				}
+			} else {
+				cloneCopyEvent( elem, clone );
+			}
+		}
+
+		// Preserve script evaluation history
+		destElements = getAll( clone, "script" );
+		if ( destElements.length > 0 ) {
+			setGlobalEval( destElements, !inPage && getAll( elem, "script" ) );
+		}
+
+		// Return the cloned set
+		return clone;
+	},
+
+	cleanData: function( elems ) {
+		var data, elem, type,
+			special = jQuery.event.special,
+			i = 0;
+
+		for ( ; ( elem = elems[ i ] ) !== undefined; i++ ) {
+			if ( acceptData( elem ) ) {
+				if ( ( data = elem[ dataPriv.expando ] ) ) {
+					if ( data.events ) {
+						for ( type in data.events ) {
+							if ( special[ type ] ) {
+								jQuery.event.remove( elem, type );
+
+							// This is a shortcut to avoid jQuery.event.remove's overhead
+							} else {
+								jQuery.removeEvent( elem, type, data.handle );
+							}
+						}
+					}
+
+					// Support: Chrome <=35 - 45+
+					// Assign undefined instead of using delete, see Data#remove
+					elem[ dataPriv.expando ] = undefined;
+				}
+				if ( elem[ dataUser.expando ] ) {
+
+					// Support: Chrome <=35 - 45+
+					// Assign undefined instead of using delete, see Data#remove
+					elem[ dataUser.expando ] = undefined;
+				}
+			}
+		}
+	}
+} );
+
+jQuery.fn.extend( {
+	detach: function( selector ) {
+		return remove( this, selector, true );
+	},
+
+	remove: function( selector ) {
+		return remove( this, selector );
+	},
+
+	text: function( value ) {
+		return access( this, function( value ) {
+			return value === undefined ?
+				jQuery.text( this ) :
+				this.empty().each( function() {
+					if ( this.nodeType === 1 || this.nodeType === 11 || this.nodeType === 9 ) {
+						this.textContent = value;
+					}
+				} );
+		}, null, value, arguments.length );
+	},
+
+	append: function() {
+		return domManip( this, arguments, function( elem ) {
+			if ( this.nodeType === 1 || this.nodeType === 11 || this.nodeType === 9 ) {
+				var target = manipulationTarget( this, elem );
+				target.appendChild( elem );
+			}
+		} );
+	},
+
+	prepend: function() {
+		return domManip( this, arguments, function( elem ) {
+			if ( this.nodeType === 1 || this.nodeType === 11 || this.nodeType === 9 ) {
+				var target = manipulationTarget( this, elem );
+				target.insertBefore( elem, target.firstChild );
+			}
+		} );
+	},
+
+	before: function() {
+		return domManip( this, arguments, function( elem ) {
+			if ( this.parentNode ) {
+				this.parentNode.insertBefore( elem, this );
+			}
+		} );
+	},
+
+	after: function() {
+		return domManip( this, arguments, function( elem ) {
+			if ( this.parentNode ) {
+				this.parentNode.insertBefore( elem, this.nextSibling );
+			}
+		} );
+	},
+
+	empty: function() {
+		var elem,
+			i = 0;
+
+		for ( ; ( elem = this[ i ] ) != null; i++ ) {
+			if ( elem.nodeType === 1 ) {
+
+				// Prevent memory leaks
+				jQuery.cleanData( getAll( elem, false ) );
+
+				// Remove any remaining nodes
+				elem.textContent = "";
+			}
+		}
+
+		return this;
+	},
+
+	clone: function( dataAndEvents, deepDataAndEvents ) {
+		dataAndEvents = dataAndEvents == null ? false : dataAndEvents;
+		deepDataAndEvents = deepDataAndEvents == null ? dataAndEvents : deepDataAndEvents;
+
+		return this.map( function() {
+			return jQuery.clone( this, dataAndEvents, deepDataAndEvents );
+		} );
+	},
+
+	html: function( value ) {
+		return access( this, function( value ) {
+			var elem = this[ 0 ] || {},
+				i = 0,
+				l = this.length;
+
+			if ( value === undefined && elem.nodeType === 1 ) {
+				return elem.innerHTML;
+			}
+
+			// See if we can take a shortcut and just use innerHTML
+			if ( typeof value === "string" && !rnoInnerhtml.test( value ) &&
+				!wrapMap[ ( rtagName.exec( value ) || [ "", "" ] )[ 1 ].toLowerCase() ] ) {
+
+				value = jQuery.htmlPrefilter( value );
+
+				try {
+					for ( ; i < l; i++ ) {
+						elem = this[ i ] || {};
+
+						// Remove element nodes and prevent memory leaks
+						if ( elem.nodeType === 1 ) {
+							jQuery.cleanData( getAll( elem, false ) );
+							elem.innerHTML = value;
+						}
+					}
+
+					elem = 0;
+
+				// If using innerHTML throws an exception, use the fallback method
+				} catch ( e ) {}
+			}
+
+			if ( elem ) {
+				this.empty().append( value );
+			}
+		}, null, value, arguments.length );
+	},
+
+	replaceWith: function() {
+		var ignored = [];
+
+		// Make the changes, replacing each non-ignored context element with the new content
+		return domManip( this, arguments, function( elem ) {
+			var parent = this.parentNode;
+
+			if ( jQuery.inArray( this, ignored ) < 0 ) {
+				jQuery.cleanData( getAll( this ) );
+				if ( parent ) {
+					parent.replaceChild( elem, this );
+				}
+			}
+
+		// Force callback invocation
+		}, ignored );
+	}
+} );
+
+jQuery.each( {
+	appendTo: "append",
+	prependTo: "prepend",
+	insertBefore: "before",
+	insertAfter: "after",
+	replaceAll: "replaceWith"
+}, function( name, original ) {
+	jQuery.fn[ name ] = function( selector ) {
+		var elems,
+			ret = [],
+			insert = jQuery( selector ),
+			last = insert.length - 1,
+			i = 0;
+
+		for ( ; i <= last; i++ ) {
+			elems = i === last ? this : this.clone( true );
+			jQuery( insert[ i ] )[ original ]( elems );
+
+			// Support: Android <=4.0 only, PhantomJS 1 only
+			// .get() because push.apply(_, arraylike) throws on ancient WebKit
+			push.apply( ret, elems.get() );
+		}
+
+		return this.pushStack( ret );
+	};
+} );
+var rnumnonpx = new RegExp( "^(" + pnum + ")(?!px)[a-z%]+$", "i" );
+
+var getStyles = function( elem ) {
+
+		// Support: IE <=11 only, Firefox <=30 (#15098, #14150)
+		// IE throws on elements created in popups
+		// FF meanwhile throws on frame elements through "defaultView.getComputedStyle"
+		var view = elem.ownerDocument.defaultView;
+
+		if ( !view || !view.opener ) {
+			view = window;
+		}
+
+		return view.getComputedStyle( elem );
+	};
+
+var swap = function( elem, options, callback ) {
+	var ret, name,
+		old = {};
+
+	// Remember the old values, and insert the new ones
+	for ( name in options ) {
+		old[ name ] = elem.style[ name ];
+		elem.style[ name ] = options[ name ];
+	}
+
+	ret = callback.call( elem );
+
+	// Revert the old values
+	for ( name in options ) {
+		elem.style[ name ] = old[ name ];
+	}
+
+	return ret;
+};
+
+
+var rboxStyle = new RegExp( cssExpand.join( "|" ), "i" );
+
+
+
+( function() {
+
+	// Executing both pixelPosition & boxSizingReliable tests require only one layout
+	// so they're executed at the same time to save the second computation.
+	function computeStyleTests() {
+
+		// This is a singleton, we need to execute it only once
+		if ( !div ) {
+			return;
+		}
+
+		container.style.cssText = "position:absolute;left:-11111px;width:60px;" +
+			"margin-top:1px;padding:0;border:0";
+		div.style.cssText =
+			"position:relative;display:block;box-sizing:border-box;overflow:scroll;" +
+			"margin:auto;border:1px;padding:1px;" +
+			"width:60%;top:1%";
+		documentElement.appendChild( container ).appendChild( div );
+
+		var divStyle = window.getComputedStyle( div );
+		pixelPositionVal = divStyle.top !== "1%";
+
+		// Support: Android 4.0 - 4.3 only, Firefox <=3 - 44
+		reliableMarginLeftVal = roundPixelMeasures( divStyle.marginLeft ) === 12;
+
+		// Support: Android 4.0 - 4.3 only, Safari <=9.1 - 10.1, iOS <=7.0 - 9.3
+		// Some styles come back with percentage values, even though they shouldn't
+		div.style.right = "60%";
+		pixelBoxStylesVal = roundPixelMeasures( divStyle.right ) === 36;
+
+		// Support: IE 9 - 11 only
+		// Detect misreporting of content dimensions for box-sizing:border-box elements
+		boxSizingReliableVal = roundPixelMeasures( divStyle.width ) === 36;
+
+		// Support: IE 9 only
+		// Detect overflow:scroll screwiness (gh-3699)
+		// Support: Chrome <=64
+		// Don't get tricked when zoom affects offsetWidth (gh-4029)
+		div.style.position = "absolute";
+		scrollboxSizeVal = roundPixelMeasures( div.offsetWidth / 3 ) === 12;
+
+		documentElement.removeChild( container );
+
+		// Nullify the div so it wouldn't be stored in the memory and
+		// it will also be a sign that checks already performed
+		div = null;
+	}
+
+	function roundPixelMeasures( measure ) {
+		return Math.round( parseFloat( measure ) );
+	}
+
+	var pixelPositionVal, boxSizingReliableVal, scrollboxSizeVal, pixelBoxStylesVal,
+		reliableTrDimensionsVal, reliableMarginLeftVal,
+		container = document.createElement( "div" ),
+		div = document.createElement( "div" );
+
+	// Finish early in limited (non-browser) environments
+	if ( !div.style ) {
+		return;
+	}
+
+	// Support: IE <=9 - 11 only
+	// Style of cloned element affects source element cloned (#8908)
+	div.style.backgroundClip = "content-box";
+	div.cloneNode( true ).style.backgroundClip = "";
+	support.clearCloneStyle = div.style.backgroundClip === "content-box";
+
+	jQuery.extend( support, {
+		boxSizingReliable: function() {
+			computeStyleTests();
+			return boxSizingReliableVal;
+		},
+		pixelBoxStyles: function() {
+			computeStyleTests();
+			return pixelBoxStylesVal;
+		},
+		pixelPosition: function() {
+			computeStyleTests();
+			return pixelPositionVal;
+		},
+		reliableMarginLeft: function() {
+			computeStyleTests();
+			return reliableMarginLeftVal;
+		},
+		scrollboxSize: function() {
+			computeStyleTests();
+			return scrollboxSizeVal;
+		},
+
+		// Support: IE 9 - 11+, Edge 15 - 18+
+		// IE/Edge misreport `getComputedStyle` of table rows with width/height
+		// set in CSS while `offset*` properties report correct values.
+		// Behavior in IE 9 is more subtle than in newer versions & it passes
+		// some versions of this test; make sure not to make it pass there!
+		//
+		// Support: Firefox 70+
+		// Only Firefox includes border widths
+		// in computed dimensions. (gh-4529)
+		reliableTrDimensions: function() {
+			var table, tr, trChild, trStyle;
+			if ( reliableTrDimensionsVal == null ) {
+				table = document.createElement( "table" );
+				tr = document.createElement( "tr" );
+				trChild = document.createElement( "div" );
+
+				table.style.cssText = "position:absolute;left:-11111px;border-collapse:separate";
+				tr.style.cssText = "border:1px solid";
+
+				// Support: Chrome 86+
+				// Height set through cssText does not get applied.
+				// Computed height then comes back as 0.
+				tr.style.height = "1px";
+				trChild.style.height = "9px";
+
+				// Support: Android 8 Chrome 86+
+				// In our bodyBackground.html iframe,
+				// display for all div elements is set to "inline",
+				// which causes a problem only in Android 8 Chrome 86.
+				// Ensuring the div is display: block
+				// gets around this issue.
+				trChild.style.display = "block";
+
+				documentElement
+					.appendChild( table )
+					.appendChild( tr )
+					.appendChild( trChild );
+
+				trStyle = window.getComputedStyle( tr );
+				reliableTrDimensionsVal = ( parseInt( trStyle.height, 10 ) +
+					parseInt( trStyle.borderTopWidth, 10 ) +
+					parseInt( trStyle.borderBottomWidth, 10 ) ) === tr.offsetHeight;
+
+				documentElement.removeChild( table );
+			}
+			return reliableTrDimensionsVal;
+		}
+	} );
+} )();
+
+
+function curCSS( elem, name, computed ) {
+	var width, minWidth, maxWidth, ret,
+
+		// Support: Firefox 51+
+		// Retrieving style before computed somehow
+		// fixes an issue with getting wrong values
+		// on detached elements
+		style = elem.style;
+
+	computed = computed || getStyles( elem );
+
+	// getPropertyValue is needed for:
+	//   .css('filter') (IE 9 only, #12537)
+	//   .css('--customProperty) (#3144)
+	if ( computed ) {
+		ret = computed.getPropertyValue( name ) || computed[ name ];
+
+		if ( ret === "" && !isAttached( elem ) ) {
+			ret = jQuery.style( elem, name );
+		}
+
+		// A tribute to the "awesome hack by Dean Edwards"
+		// Android Browser returns percentage for some values,
+		// but width seems to be reliably pixels.
+		// This is against the CSSOM draft spec:
+		// https://drafts.csswg.org/cssom/#resolved-values
+		if ( !support.pixelBoxStyles() && rnumnonpx.test( ret ) && rboxStyle.test( name ) ) {
+
+			// Remember the original values
+			width = style.width;
+			minWidth = style.minWidth;
+			maxWidth = style.maxWidth;
+
+			// Put in the new values to get a computed value out
+			style.minWidth = style.maxWidth = style.width = ret;
+			ret = computed.width;
+
+			// Revert the changed values
+			style.width = width;
+			style.minWidth = minWidth;
+			style.maxWidth = maxWidth;
+		}
+	}
+
+	return ret !== undefined ?
+
+		// Support: IE <=9 - 11 only
+		// IE returns zIndex value as an integer.
+		ret + "" :
+		ret;
+}
+
+
+function addGetHookIf( conditionFn, hookFn ) {
+
+	// Define the hook, we'll check on the first run if it's really needed.
+	return {
+		get: function() {
+			if ( conditionFn() ) {
+
+				// Hook not needed (or it's not possible to use it due
+				// to missing dependency), remove it.
+				delete this.get;
+				return;
+			}
+
+			// Hook needed; redefine it so that the support test is not executed again.
+			return ( this.get = hookFn ).apply( this, arguments );
+		}
+	};
+}
+
+
+var cssPrefixes = [ "Webkit", "Moz", "ms" ],
+	emptyStyle = document.createElement( "div" ).style,
+	vendorProps = {};
+
+// Return a vendor-prefixed property or undefined
+function vendorPropName( name ) {
+
+	// Check for vendor prefixed names
+	var capName = name[ 0 ].toUpperCase() + name.slice( 1 ),
+		i = cssPrefixes.length;
+
+	while ( i-- ) {
+		name = cssPrefixes[ i ] + capName;
+		if ( name in emptyStyle ) {
+			return name;
+		}
+	}
+}
+
+// Return a potentially-mapped jQuery.cssProps or vendor prefixed property
+function finalPropName( name ) {
+	var final = jQuery.cssProps[ name ] || vendorProps[ name ];
+
+	if ( final ) {
+		return final;
+	}
+	if ( name in emptyStyle ) {
+		return name;
+	}
+	return vendorProps[ name ] = vendorPropName( name ) || name;
+}
+
+
+var
+
+	// Swappable if display is none or starts with table
+	// except "table", "table-cell", or "table-caption"
+	// See here for display values: https://developer.mozilla.org/en-US/docs/CSS/display
+	rdisplayswap = /^(none|table(?!-c[ea]).+)/,
+	rcustomProp = /^--/,
+	cssShow = { position: "absolute", visibility: "hidden", display: "block" },
+	cssNormalTransform = {
+		letterSpacing: "0",
+		fontWeight: "400"
+	};
+
+function setPositiveNumber( _elem, value, subtract ) {
+
+	// Any relative (+/-) values have already been
+	// normalized at this point
+	var matches = rcssNum.exec( value );
+	return matches ?
+
+		// Guard against undefined "subtract", e.g., when used as in cssHooks
+		Math.max( 0, matches[ 2 ] - ( subtract || 0 ) ) + ( matches[ 3 ] || "px" ) :
+		value;
+}
+
+function boxModelAdjustment( elem, dimension, box, isBorderBox, styles, computedVal ) {
+	var i = dimension === "width" ? 1 : 0,
+		extra = 0,
+		delta = 0;
+
+	// Adjustment may not be necessary
+	if ( box === ( isBorderBox ? "border" : "content" ) ) {
+		return 0;
+	}
+
+	for ( ; i < 4; i += 2 ) {
+
+		// Both box models exclude margin
+		if ( box === "margin" ) {
+			delta += jQuery.css( elem, box + cssExpand[ i ], true, styles );
+		}
+
+		// If we get here with a content-box, we're seeking "padding" or "border" or "margin"
+		if ( !isBorderBox ) {
+
+			// Add padding
+			delta += jQuery.css( elem, "padding" + cssExpand[ i ], true, styles );
+
+			// For "border" or "margin", add border
+			if ( box !== "padding" ) {
+				delta += jQuery.css( elem, "border" + cssExpand[ i ] + "Width", true, styles );
+
+			// But still keep track of it otherwise
+			} else {
+				extra += jQuery.css( elem, "border" + cssExpand[ i ] + "Width", true, styles );
+			}
+
+		// If we get here with a border-box (content + padding + border), we're seeking "content" or
+		// "padding" or "margin"
+		} else {
+
+			// For "content", subtract padding
+			if ( box === "content" ) {
+				delta -= jQuery.css( elem, "padding" + cssExpand[ i ], true, styles );
+			}
+
+			// For "content" or "padding", subtract border
+			if ( box !== "margin" ) {
+				delta -= jQuery.css( elem, "border" + cssExpand[ i ] + "Width", true, styles );
+			}
+		}
+	}
+
+	// Account for positive content-box scroll gutter when requested by providing computedVal
+	if ( !isBorderBox && computedVal >= 0 ) {
+
+		// offsetWidth/offsetHeight is a rounded sum of content, padding, scroll gutter, and border
+		// Assuming integer scroll gutter, subtract the rest and round down
+		delta += Math.max( 0, Math.ceil(
+			elem[ "offset" + dimension[ 0 ].toUpperCase() + dimension.slice( 1 ) ] -
+			computedVal -
+			delta -
+			extra -
+			0.5
+
+		// If offsetWidth/offsetHeight is unknown, then we can't determine content-box scroll gutter
+		// Use an explicit zero to avoid NaN (gh-3964)
+		) ) || 0;
+	}
+
+	return delta;
+}
+
+function getWidthOrHeight( elem, dimension, extra ) {
+
+	// Start with computed style
+	var styles = getStyles( elem ),
+
+		// To avoid forcing a reflow, only fetch boxSizing if we need it (gh-4322).
+		// Fake content-box until we know it's needed to know the true value.
+		boxSizingNeeded = !support.boxSizingReliable() || extra,
+		isBorderBox = boxSizingNeeded &&
+			jQuery.css( elem, "boxSizing", false, styles ) === "border-box",
+		valueIsBorderBox = isBorderBox,
+
+		val = curCSS( elem, dimension, styles ),
+		offsetProp = "offset" + dimension[ 0 ].toUpperCase() + dimension.slice( 1 );
+
+	// Support: Firefox <=54
+	// Return a confounding non-pixel value or feign ignorance, as appropriate.
+	if ( rnumnonpx.test( val ) ) {
+		if ( !extra ) {
+			return val;
+		}
+		val = "auto";
+	}
+
+
+	// Support: IE 9 - 11 only
+	// Use offsetWidth/offsetHeight for when box sizing is unreliable.
+	// In those cases, the computed value can be trusted to be border-box.
+	if ( ( !support.boxSizingReliable() && isBorderBox ||
+
+		// Support: IE 10 - 11+, Edge 15 - 18+
+		// IE/Edge misreport `getComputedStyle` of table rows with width/height
+		// set in CSS while `offset*` properties report correct values.
+		// Interestingly, in some cases IE 9 doesn't suffer from this issue.
+		!support.reliableTrDimensions() && nodeName( elem, "tr" ) ||
+
+		// Fall back to offsetWidth/offsetHeight when value is "auto"
+		// This happens for inline elements with no explicit setting (gh-3571)
+		val === "auto" ||
+
+		// Support: Android <=4.1 - 4.3 only
+		// Also use offsetWidth/offsetHeight for misreported inline dimensions (gh-3602)
+		!parseFloat( val ) && jQuery.css( elem, "display", false, styles ) === "inline" ) &&
+
+		// Make sure the element is visible & connected
+		elem.getClientRects().length ) {
+
+		isBorderBox = jQuery.css( elem, "boxSizing", false, styles ) === "border-box";
+
+		// Where available, offsetWidth/offsetHeight approximate border box dimensions.
+		// Where not available (e.g., SVG), assume unreliable box-sizing and interpret the
+		// retrieved value as a content box dimension.
+		valueIsBorderBox = offsetProp in elem;
+		if ( valueIsBorderBox ) {
+			val = elem[ offsetProp ];
+		}
+	}
+
+	// Normalize "" and auto
+	val = parseFloat( val ) || 0;
+
+	// Adjust for the element's box model
+	return ( val +
+		boxModelAdjustment(
+			elem,
+			dimension,
+			extra || ( isBorderBox ? "border" : "content" ),
+			valueIsBorderBox,
+			styles,
+
+			// Provide the current computed size to request scroll gutter calculation (gh-3589)
+			val
+		)
+	) + "px";
+}
+
+jQuery.extend( {
+
+	// Add in style property hooks for overriding the default
+	// behavior of getting and setting a style property
+	cssHooks: {
+		opacity: {
+			get: function( elem, computed ) {
+				if ( computed ) {
+
+					// We should always get a number back from opacity
+					var ret = curCSS( elem, "opacity" );
+					return ret === "" ? "1" : ret;
+				}
+			}
+		}
+	},
+
+	// Don't automatically add "px" to these possibly-unitless properties
+	cssNumber: {
+		"animationIterationCount": true,
+		"columnCount": true,
+		"fillOpacity": true,
+		"flexGrow": true,
+		"flexShrink": true,
+		"fontWeight": true,
+		"gridArea": true,
+		"gridColumn": true,
+		"gridColumnEnd": true,
+		"gridColumnStart": true,
+		"gridRow": true,
+		"gridRowEnd": true,
+		"gridRowStart": true,
+		"lineHeight": true,
+		"opacity": true,
+		"order": true,
+		"orphans": true,
+		"widows": true,
+		"zIndex": true,
+		"zoom": true
+	},
+
+	// Add in properties whose names you wish to fix before
+	// setting or getting the value
+	cssProps: {},
+
+	// Get and set the style property on a DOM Node
+	style: function( elem, name, value, extra ) {
+
+		// Don't set styles on text and comment nodes
+		if ( !elem || elem.nodeType === 3 || elem.nodeType === 8 || !elem.style ) {
+			return;
+		}
+
+		// Make sure that we're working with the right name
+		var ret, type, hooks,
+			origName = camelCase( name ),
+			isCustomProp = rcustomProp.test( name ),
+			style = elem.style;
+
+		// Make sure that we're working with the right name. We don't
+		// want to query the value if it is a CSS custom property
+		// since they are user-defined.
+		if ( !isCustomProp ) {
+			name = finalPropName( origName );
+		}
+
+		// Gets hook for the prefixed version, then unprefixed version
+		hooks = jQuery.cssHooks[ name ] || jQuery.cssHooks[ origName ];
+
+		// Check if we're setting a value
+		if ( value !== undefined ) {
+			type = typeof value;
+
+			// Convert "+=" or "-=" to relative numbers (#7345)
+			if ( type === "string" && ( ret = rcssNum.exec( value ) ) && ret[ 1 ] ) {
+				value = adjustCSS( elem, name, ret );
+
+				// Fixes bug #9237
+				type = "number";
+			}
+
+			// Make sure that null and NaN values aren't set (#7116)
+			if ( value == null || value !== value ) {
+				return;
+			}
+
+			// If a number was passed in, add the unit (except for certain CSS properties)
+			// The isCustomProp check can be removed in jQuery 4.0 when we only auto-append
+			// "px" to a few hardcoded values.
+			if ( type === "number" && !isCustomProp ) {
+				value += ret && ret[ 3 ] || ( jQuery.cssNumber[ origName ] ? "" : "px" );
+			}
+
+			// background-* props affect original clone's values
+			if ( !support.clearCloneStyle && value === "" && name.indexOf( "background" ) === 0 ) {
+				style[ name ] = "inherit";
+			}
+
+			// If a hook was provided, use that value, otherwise just set the specified value
+			if ( !hooks || !( "set" in hooks ) ||
+				( value = hooks.set( elem, value, extra ) ) !== undefined ) {
+
+				if ( isCustomProp ) {
+					style.setProperty( name, value );
+				} else {
+					style[ name ] = value;
+				}
+			}
+
+		} else {
+
+			// If a hook was provided get the non-computed value from there
+			if ( hooks && "get" in hooks &&
+				( ret = hooks.get( elem, false, extra ) ) !== undefined ) {
+
+				return ret;
+			}
+
+			// Otherwise just get the value from the style object
+			return style[ name ];
+		}
+	},
+
+	css: function( elem, name, extra, styles ) {
+		var val, num, hooks,
+			origName = camelCase( name ),
+			isCustomProp = rcustomProp.test( name );
+
+		// Make sure that we're working with the right name. We don't
+		// want to modify the value if it is a CSS custom property
+		// since they are user-defined.
+		if ( !isCustomProp ) {
+			name = finalPropName( origName );
+		}
+
+		// Try prefixed name followed by the unprefixed name
+		hooks = jQuery.cssHooks[ name ] || jQuery.cssHooks[ origName ];
+
+		// If a hook was provided get the computed value from there
+		if ( hooks && "get" in hooks ) {
+			val = hooks.get( elem, true, extra );
+		}
+
+		// Otherwise, if a way to get the computed value exists, use that
+		if ( val === undefined ) {
+			val = curCSS( elem, name, styles );
+		}
+
+		// Convert "normal" to computed value
+		if ( val === "normal" && name in cssNormalTransform ) {
+			val = cssNormalTransform[ name ];
+		}
+
+		// Make numeric if forced or a qualifier was provided and val looks numeric
+		if ( extra === "" || extra ) {
+			num = parseFloat( val );
+			return extra === true || isFinite( num ) ? num || 0 : val;
+		}
+
+		return val;
+	}
+} );
+
+jQuery.each( [ "height", "width" ], function( _i, dimension ) {
+	jQuery.cssHooks[ dimension ] = {
+		get: function( elem, computed, extra ) {
+			if ( computed ) {
+
+				// Certain elements can have dimension info if we invisibly show them
+				// but it must have a current display style that would benefit
+				return rdisplayswap.test( jQuery.css( elem, "display" ) ) &&
+
+					// Support: Safari 8+
+					// Table columns in Safari have non-zero offsetWidth & zero
+					// getBoundingClientRect().width unless display is changed.
+					// Support: IE <=11 only
+					// Running getBoundingClientRect on a disconnected node
+					// in IE throws an error.
+					( !elem.getClientRects().length || !elem.getBoundingClientRect().width ) ?
+					swap( elem, cssShow, function() {
+						return getWidthOrHeight( elem, dimension, extra );
+					} ) :
+					getWidthOrHeight( elem, dimension, extra );
+			}
+		},
+
+		set: function( elem, value, extra ) {
+			var matches,
+				styles = getStyles( elem ),
+
+				// Only read styles.position if the test has a chance to fail
+				// to avoid forcing a reflow.
+				scrollboxSizeBuggy = !support.scrollboxSize() &&
+					styles.position === "absolute",
+
+				// To avoid forcing a reflow, only fetch boxSizing if we need it (gh-3991)
+				boxSizingNeeded = scrollboxSizeBuggy || extra,
+				isBorderBox = boxSizingNeeded &&
+					jQuery.css( elem, "boxSizing", false, styles ) === "border-box",
+				subtract = extra ?
+					boxModelAdjustment(
+						elem,
+						dimension,
+						extra,
+						isBorderBox,
+						styles
+					) :
+					0;
+
+			// Account for unreliable border-box dimensions by comparing offset* to computed and
+			// faking a content-box to get border and padding (gh-3699)
+			if ( isBorderBox && scrollboxSizeBuggy ) {
+				subtract -= Math.ceil(
+					elem[ "offset" + dimension[ 0 ].toUpperCase() + dimension.slice( 1 ) ] -
+					parseFloat( styles[ dimension ] ) -
+					boxModelAdjustment( elem, dimension, "border", false, styles ) -
+					0.5
+				);
+			}
+
+			// Convert to pixels if value adjustment is needed
+			if ( subtract && ( matches = rcssNum.exec( value ) ) &&
+				( matches[ 3 ] || "px" ) !== "px" ) {
+
+				elem.style[ dimension ] = value;
+				value = jQuery.css( elem, dimension );
+			}
+
+			return setPositiveNumber( elem, value, subtract );
+		}
+	};
+} );
+
+jQuery.cssHooks.marginLeft = addGetHookIf( support.reliableMarginLeft,
+	function( elem, computed ) {
+		if ( computed ) {
+			return ( parseFloat( curCSS( elem, "marginLeft" ) ) ||
+				elem.getBoundingClientRect().left -
+					swap( elem, { marginLeft: 0 }, function() {
+						return elem.getBoundingClientRect().left;
+					} )
+			) + "px";
+		}
+	}
+);
+
+// These hooks are used by animate to expand properties
+jQuery.each( {
+	margin: "",
+	padding: "",
+	border: "Width"
+}, function( prefix, suffix ) {
+	jQuery.cssHooks[ prefix + suffix ] = {
+		expand: function( value ) {
+			var i = 0,
+				expanded = {},
+
+				// Assumes a single number if not a string
+				parts = typeof value === "string" ? value.split( " " ) : [ value ];
+
+			for ( ; i < 4; i++ ) {
+				expanded[ prefix + cssExpand[ i ] + suffix ] =
+					parts[ i ] || parts[ i - 2 ] || parts[ 0 ];
+			}
+
+			return expanded;
+		}
+	};
+
+	if ( prefix !== "margin" ) {
+		jQuery.cssHooks[ prefix + suffix ].set = setPositiveNumber;
+	}
+} );
+
+jQuery.fn.extend( {
+	css: function( name, value ) {
+		return access( this, function( elem, name, value ) {
+			var styles, len,
+				map = {},
+				i = 0;
+
+			if ( Array.isArray( name ) ) {
+				styles = getStyles( elem );
+				len = name.length;
+
+				for ( ; i < len; i++ ) {
+					map[ name[ i ] ] = jQuery.css( elem, name[ i ], false, styles );
+				}
+
+				return map;
+			}
+
+			return value !== undefined ?
+				jQuery.style( elem, name, value ) :
+				jQuery.css( elem, name );
+		}, name, value, arguments.length > 1 );
+	}
+} );
+
+
+function Tween( elem, options, prop, end, easing ) {
+	return new Tween.prototype.init( elem, options, prop, end, easing );
+}
+jQuery.Tween = Tween;
+
+Tween.prototype = {
+	constructor: Tween,
+	init: function( elem, options, prop, end, easing, unit ) {
+		this.elem = elem;
+		this.prop = prop;
+		this.easing = easing || jQuery.easing._default;
+		this.options = options;
+		this.start = this.now = this.cur();
+		this.end = end;
+		this.unit = unit || ( jQuery.cssNumber[ prop ] ? "" : "px" );
+	},
+	cur: function() {
+		var hooks = Tween.propHooks[ this.prop ];
+
+		return hooks && hooks.get ?
+			hooks.get( this ) :
+			Tween.propHooks._default.get( this );
+	},
+	run: function( percent ) {
+		var eased,
+			hooks = Tween.propHooks[ this.prop ];
+
+		if ( this.options.duration ) {
+			this.pos = eased = jQuery.easing[ this.easing ](
+				percent, this.options.duration * percent, 0, 1, this.options.duration
+			);
+		} else {
+			this.pos = eased = percent;
+		}
+		this.now = ( this.end - this.start ) * eased + this.start;
+
+		if ( this.options.step ) {
+			this.options.step.call( this.elem, this.now, this );
+		}
+
+		if ( hooks && hooks.set ) {
+			hooks.set( this );
+		} else {
+			Tween.propHooks._default.set( this );
+		}
+		return this;
+	}
+};
+
+Tween.prototype.init.prototype = Tween.prototype;
+
+Tween.propHooks = {
+	_default: {
+		get: function( tween ) {
+			var result;
+
+			// Use a property on the element directly when it is not a DOM element,
+			// or when there is no matching style property that exists.
+			if ( tween.elem.nodeType !== 1 ||
+				tween.elem[ tween.prop ] != null && tween.elem.style[ tween.prop ] == null ) {
+				return tween.elem[ tween.prop ];
+			}
+
+			// Passing an empty string as a 3rd parameter to .css will automatically
+			// attempt a parseFloat and fallback to a string if the parse fails.
+			// Simple values such as "10px" are parsed to Float;
+			// complex values such as "rotate(1rad)" are returned as-is.
+			result = jQuery.css( tween.elem, tween.prop, "" );
+
+			// Empty strings, null, undefined and "auto" are converted to 0.
+			return !result || result === "auto" ? 0 : result;
+		},
+		set: function( tween ) {
+
+			// Use step hook for back compat.
+			// Use cssHook if its there.
+			// Use .style if available and use plain properties where available.
+			if ( jQuery.fx.step[ tween.prop ] ) {
+				jQuery.fx.step[ tween.prop ]( tween );
+			} else if ( tween.elem.nodeType === 1 && (
+				jQuery.cssHooks[ tween.prop ] ||
+					tween.elem.style[ finalPropName( tween.prop ) ] != null ) ) {
+				jQuery.style( tween.elem, tween.prop, tween.now + tween.unit );
+			} else {
+				tween.elem[ tween.prop ] = tween.now;
+			}
+		}
+	}
+};
+
+// Support: IE <=9 only
+// Panic based approach to setting things on disconnected nodes
+Tween.propHooks.scrollTop = Tween.propHooks.scrollLeft = {
+	set: function( tween ) {
+		if ( tween.elem.nodeType && tween.elem.parentNode ) {
+			tween.elem[ tween.prop ] = tween.now;
+		}
+	}
+};
+
+jQuery.easing = {
+	linear: function( p ) {
+		return p;
+	},
+	swing: function( p ) {
+		return 0.5 - Math.cos( p * Math.PI ) / 2;
+	},
+	_default: "swing"
+};
+
+jQuery.fx = Tween.prototype.init;
+
+// Back compat <1.8 extension point
+jQuery.fx.step = {};
+
+
+
+
+var
+	fxNow, inProgress,
+	rfxtypes = /^(?:toggle|show|hide)$/,
+	rrun = /queueHooks$/;
+
+function schedule() {
+	if ( inProgress ) {
+		if ( document.hidden === false && window.requestAnimationFrame ) {
+			window.requestAnimationFrame( schedule );
+		} else {
+			window.setTimeout( schedule, jQuery.fx.interval );
+		}
+
+		jQuery.fx.tick();
+	}
+}
+
+// Animations created synchronously will run synchronously
+function createFxNow() {
+	window.setTimeout( function() {
+		fxNow = undefined;
+	} );
+	return ( fxNow = Date.now() );
+}
+
+// Generate parameters to create a standard animation
+function genFx( type, includeWidth ) {
+	var which,
+		i = 0,
+		attrs = { height: type };
+
+	// If we include width, step value is 1 to do all cssExpand values,
+	// otherwise step value is 2 to skip over Left and Right
+	includeWidth = includeWidth ? 1 : 0;
+	for ( ; i < 4; i += 2 - includeWidth ) {
+		which = cssExpand[ i ];
+		attrs[ "margin" + which ] = attrs[ "padding" + which ] = type;
+	}
+
+	if ( includeWidth ) {
+		attrs.opacity = attrs.width = type;
+	}
+
+	return attrs;
+}
+
+function createTween( value, prop, animation ) {
+	var tween,
+		collection = ( Animation.tweeners[ prop ] || [] ).concat( Animation.tweeners[ "*" ] ),
+		index = 0,
+		length = collection.length;
+	for ( ; index < length; index++ ) {
+		if ( ( tween = collection[ index ].call( animation, prop, value ) ) ) {
+
+			// We're done with this property
+			return tween;
+		}
+	}
+}
+
+function defaultPrefilter( elem, props, opts ) {
+	var prop, value, toggle, hooks, oldfire, propTween, restoreDisplay, display,
+		isBox = "width" in props || "height" in props,
+		anim = this,
+		orig = {},
+		style = elem.style,
+		hidden = elem.nodeType && isHiddenWithinTree( elem ),
+		dataShow = dataPriv.get( elem, "fxshow" );
+
+	// Queue-skipping animations hijack the fx hooks
+	if ( !opts.queue ) {
+		hooks = jQuery._queueHooks( elem, "fx" );
+		if ( hooks.unqueued == null ) {
+			hooks.unqueued = 0;
+			oldfire = hooks.empty.fire;
+			hooks.empty.fire = function() {
+				if ( !hooks.unqueued ) {
+					oldfire();
+				}
+			};
+		}
+		hooks.unqueued++;
+
+		anim.always( function() {
+
+			// Ensure the complete handler is called before this completes
+			anim.always( function() {
+				hooks.unqueued--;
+				if ( !jQuery.queue( elem, "fx" ).length ) {
+					hooks.empty.fire();
+				}
+			} );
+		} );
+	}
+
+	// Detect show/hide animations
+	for ( prop in props ) {
+		value = props[ prop ];
+		if ( rfxtypes.test( value ) ) {
+			delete props[ prop ];
+			toggle = toggle || value === "toggle";
+			if ( value === ( hidden ? "hide" : "show" ) ) {
+
+				// Pretend to be hidden if this is a "show" and
+				// there is still data from a stopped show/hide
+				if ( value === "show" && dataShow && dataShow[ prop ] !== undefined ) {
+					hidden = true;
+
+				// Ignore all other no-op show/hide data
+				} else {
+					continue;
+				}
+			}
+			orig[ prop ] = dataShow && dataShow[ prop ] || jQuery.style( elem, prop );
+		}
+	}
+
+	// Bail out if this is a no-op like .hide().hide()
+	propTween = !jQuery.isEmptyObject( props );
+	if ( !propTween && jQuery.isEmptyObject( orig ) ) {
+		return;
+	}
+
+	// Restrict "overflow" and "display" styles during box animations
+	if ( isBox && elem.nodeType === 1 ) {
+
+		// Support: IE <=9 - 11, Edge 12 - 15
+		// Record all 3 overflow attributes because IE does not infer the shorthand
+		// from identically-valued overflowX and overflowY and Edge just mirrors
+		// the overflowX value there.
+		opts.overflow = [ style.overflow, style.overflowX, style.overflowY ];
+
+		// Identify a display type, preferring old show/hide data over the CSS cascade
+		restoreDisplay = dataShow && dataShow.display;
+		if ( restoreDisplay == null ) {
+			restoreDisplay = dataPriv.get( elem, "display" );
+		}
+		display = jQuery.css( elem, "display" );
+		if ( display === "none" ) {
+			if ( restoreDisplay ) {
+				display = restoreDisplay;
+			} else {
+
+				// Get nonempty value(s) by temporarily forcing visibility
+				showHide( [ elem ], true );
+				restoreDisplay = elem.style.display || restoreDisplay;
+				display = jQuery.css( elem, "display" );
+				showHide( [ elem ] );
+			}
+		}
+
+		// Animate inline elements as inline-block
+		if ( display === "inline" || display === "inline-block" && restoreDisplay != null ) {
+			if ( jQuery.css( elem, "float" ) === "none" ) {
+
+				// Restore the original display value at the end of pure show/hide animations
+				if ( !propTween ) {
+					anim.done( function() {
+						style.display = restoreDisplay;
+					} );
+					if ( restoreDisplay == null ) {
+						display = style.display;
+						restoreDisplay = display === "none" ? "" : display;
+					}
+				}
+				style.display = "inline-block";
+			}
+		}
+	}
+
+	if ( opts.overflow ) {
+		style.overflow = "hidden";
+		anim.always( function() {
+			style.overflow = opts.overflow[ 0 ];
+			style.overflowX = opts.overflow[ 1 ];
+			style.overflowY = opts.overflow[ 2 ];
+		} );
+	}
+
+	// Implement show/hide animations
+	propTween = false;
+	for ( prop in orig ) {
+
+		// General show/hide setup for this element animation
+		if ( !propTween ) {
+			if ( dataShow ) {
+				if ( "hidden" in dataShow ) {
+					hidden = dataShow.hidden;
+				}
+			} else {
+				dataShow = dataPriv.access( elem, "fxshow", { display: restoreDisplay } );
+			}
+
+			// Store hidden/visible for toggle so `.stop().toggle()` "reverses"
+			if ( toggle ) {
+				dataShow.hidden = !hidden;
+			}
+
+			// Show elements before animating them
+			if ( hidden ) {
+				showHide( [ elem ], true );
+			}
+
+			/* eslint-disable no-loop-func */
+
+			anim.done( function() {
+
+				/* eslint-enable no-loop-func */
+
+				// The final step of a "hide" animation is actually hiding the element
+				if ( !hidden ) {
+					showHide( [ elem ] );
+				}
+				dataPriv.remove( elem, "fxshow" );
+				for ( prop in orig ) {
+					jQuery.style( elem, prop, orig[ prop ] );
+				}
+			} );
+		}
+
+		// Per-property setup
+		propTween = createTween( hidden ? dataShow[ prop ] : 0, prop, anim );
+		if ( !( prop in dataShow ) ) {
+			dataShow[ prop ] = propTween.start;
+			if ( hidden ) {
+				propTween.end = propTween.start;
+				propTween.start = 0;
+			}
+		}
+	}
+}
+
+function propFilter( props, specialEasing ) {
+	var index, name, easing, value, hooks;
+
+	// camelCase, specialEasing and expand cssHook pass
+	for ( index in props ) {
+		name = camelCase( index );
+		easing = specialEasing[ name ];
+		value = props[ index ];
+		if ( Array.isArray( value ) ) {
+			easing = value[ 1 ];
+			value = props[ index ] = value[ 0 ];
+		}
+
+		if ( index !== name ) {
+			props[ name ] = value;
+			delete props[ index ];
+		}
+
+		hooks = jQuery.cssHooks[ name ];
+		if ( hooks && "expand" in hooks ) {
+			value = hooks.expand( value );
+			delete props[ name ];
+
+			// Not quite $.extend, this won't overwrite existing keys.
+			// Reusing 'index' because we have the correct "name"
+			for ( index in value ) {
+				if ( !( index in props ) ) {
+					props[ index ] = value[ index ];
+					specialEasing[ index ] = easing;
+				}
+			}
+		} else {
+			specialEasing[ name ] = easing;
+		}
+	}
+}
+
+function Animation( elem, properties, options ) {
+	var result,
+		stopped,
+		index = 0,
+		length = Animation.prefilters.length,
+		deferred = jQuery.Deferred().always( function() {
+
+			// Don't match elem in the :animated selector
+			delete tick.elem;
+		} ),
+		tick = function() {
+			if ( stopped ) {
+				return false;
+			}
+			var currentTime = fxNow || createFxNow(),
+				remaining = Math.max( 0, animation.startTime + animation.duration - currentTime ),
+
+				// Support: Android 2.3 only
+				// Archaic crash bug won't allow us to use `1 - ( 0.5 || 0 )` (#12497)
+				temp = remaining / animation.duration || 0,
+				percent = 1 - temp,
+				index = 0,
+				length = animation.tweens.length;
+
+			for ( ; index < length; index++ ) {
+				animation.tweens[ index ].run( percent );
+			}
+
+			deferred.notifyWith( elem, [ animation, percent, remaining ] );
+
+			// If there's more to do, yield
+			if ( percent < 1 && length ) {
+				return remaining;
+			}
+
+			// If this was an empty animation, synthesize a final progress notification
+			if ( !length ) {
+				deferred.notifyWith( elem, [ animation, 1, 0 ] );
+			}
+
+			// Resolve the animation and report its conclusion
+			deferred.resolveWith( elem, [ animation ] );
+			return false;
+		},
+		animation = deferred.promise( {
+			elem: elem,
+			props: jQuery.extend( {}, properties ),
+			opts: jQuery.extend( true, {
+				specialEasing: {},
+				easing: jQuery.easing._default
+			}, options ),
+			originalProperties: properties,
+			originalOptions: options,
+			startTime: fxNow || createFxNow(),
+			duration: options.duration,
+			tweens: [],
+			createTween: function( prop, end ) {
+				var tween = jQuery.Tween( elem, animation.opts, prop, end,
+					animation.opts.specialEasing[ prop ] || animation.opts.easing );
+				animation.tweens.push( tween );
+				return tween;
+			},
+			stop: function( gotoEnd ) {
+				var index = 0,
+
+					// If we are going to the end, we want to run all the tweens
+					// otherwise we skip this part
+					length = gotoEnd ? animation.tweens.length : 0;
+				if ( stopped ) {
+					return this;
+				}
+				stopped = true;
+				for ( ; index < length; index++ ) {
+					animation.tweens[ index ].run( 1 );
+				}
+
+				// Resolve when we played the last frame; otherwise, reject
+				if ( gotoEnd ) {
+					deferred.notifyWith( elem, [ animation, 1, 0 ] );
+					deferred.resolveWith( elem, [ animation, gotoEnd ] );
+				} else {
+					deferred.rejectWith( elem, [ animation, gotoEnd ] );
+				}
+				return this;
+			}
+		} ),
+		props = animation.props;
+
+	propFilter( props, animation.opts.specialEasing );
+
+	for ( ; index < length; index++ ) {
+		result = Animation.prefilters[ index ].call( animation, elem, props, animation.opts );
+		if ( result ) {
+			if ( isFunction( result.stop ) ) {
+				jQuery._queueHooks( animation.elem, animation.opts.queue ).stop =
+					result.stop.bind( result );
+			}
+			return result;
+		}
+	}
+
+	jQuery.map( props, createTween, animation );
+
+	if ( isFunction( animation.opts.start ) ) {
+		animation.opts.start.call( elem, animation );
+	}
+
+	// Attach callbacks from options
+	animation
+		.progress( animation.opts.progress )
+		.done( animation.opts.done, animation.opts.complete )
+		.fail( animation.opts.fail )
+		.always( animation.opts.always );
+
+	jQuery.fx.timer(
+		jQuery.extend( tick, {
+			elem: elem,
+			anim: animation,
+			queue: animation.opts.queue
+		} )
+	);
+
+	return animation;
+}
+
+jQuery.Animation = jQuery.extend( Animation, {
+
+	tweeners: {
+		"*": [ function( prop, value ) {
+			var tween = this.createTween( prop, value );
+			adjustCSS( tween.elem, prop, rcssNum.exec( value ), tween );
+			return tween;
+		} ]
+	},
+
+	tweener: function( props, callback ) {
+		if ( isFunction( props ) ) {
+			callback = props;
+			props = [ "*" ];
+		} else {
+			props = props.match( rnothtmlwhite );
+		}
+
+		var prop,
+			index = 0,
+			length = props.length;
+
+		for ( ; index < length; index++ ) {
+			prop = props[ index ];
+			Animation.tweeners[ prop ] = Animation.tweeners[ prop ] || [];
+			Animation.tweeners[ prop ].unshift( callback );
+		}
+	},
+
+	prefilters: [ defaultPrefilter ],
+
+	prefilter: function( callback, prepend ) {
+		if ( prepend ) {
+			Animation.prefilters.unshift( callback );
+		} else {
+			Animation.prefilters.push( callback );
+		}
+	}
+} );
+
+jQuery.speed = function( speed, easing, fn ) {
+	var opt = speed && typeof speed === "object" ? jQuery.extend( {}, speed ) : {
+		complete: fn || !fn && easing ||
+			isFunction( speed ) && speed,
+		duration: speed,
+		easing: fn && easing || easing && !isFunction( easing ) && easing
+	};
+
+	// Go to the end state if fx are off
+	if ( jQuery.fx.off ) {
+		opt.duration = 0;
+
+	} else {
+		if ( typeof opt.duration !== "number" ) {
+			if ( opt.duration in jQuery.fx.speeds ) {
+				opt.duration = jQuery.fx.speeds[ opt.duration ];
+
+			} else {
+				opt.duration = jQuery.fx.speeds._default;
+			}
+		}
+	}
+
+	// Normalize opt.queue - true/undefined/null -> "fx"
+	if ( opt.queue == null || opt.queue === true ) {
+		opt.queue = "fx";
+	}
+
+	// Queueing
+	opt.old = opt.complete;
+
+	opt.complete = function() {
+		if ( isFunction( opt.old ) ) {
+			opt.old.call( this );
+		}
+
+		if ( opt.queue ) {
+			jQuery.dequeue( this, opt.queue );
+		}
+	};
+
+	return opt;
+};
+
+jQuery.fn.extend( {
+	fadeTo: function( speed, to, easing, callback ) {
+
+		// Show any hidden elements after setting opacity to 0
+		return this.filter( isHiddenWithinTree ).css( "opacity", 0 ).show()
+
+			// Animate to the value specified
+			.end().animate( { opacity: to }, speed, easing, callback );
+	},
+	animate: function( prop, speed, easing, callback ) {
+		var empty = jQuery.isEmptyObject( prop ),
+			optall = jQuery.speed( speed, easing, callback ),
+			doAnimation = function() {
+
+				// Operate on a copy of prop so per-property easing won't be lost
+				var anim = Animation( this, jQuery.extend( {}, prop ), optall );
+
+				// Empty animations, or finishing resolves immediately
+				if ( empty || dataPriv.get( this, "finish" ) ) {
+					anim.stop( true );
+				}
+			};
+
+		doAnimation.finish = doAnimation;
+
+		return empty || optall.queue === false ?
+			this.each( doAnimation ) :
+			this.queue( optall.queue, doAnimation );
+	},
+	stop: function( type, clearQueue, gotoEnd ) {
+		var stopQueue = function( hooks ) {
+			var stop = hooks.stop;
+			delete hooks.stop;
+			stop( gotoEnd );
+		};
+
+		if ( typeof type !== "string" ) {
+			gotoEnd = clearQueue;
+			clearQueue = type;
+			type = undefined;
+		}
+		if ( clearQueue ) {
+			this.queue( type || "fx", [] );
+		}
+
+		return this.each( function() {
+			var dequeue = true,
+				index = type != null && type + "queueHooks",
+				timers = jQuery.timers,
+				data = dataPriv.get( this );
+
+			if ( index ) {
+				if ( data[ index ] && data[ index ].stop ) {
+					stopQueue( data[ index ] );
+				}
+			} else {
+				for ( index in data ) {
+					if ( data[ index ] && data[ index ].stop && rrun.test( index ) ) {
+						stopQueue( data[ index ] );
+					}
+				}
+			}
+
+			for ( index = timers.length; index--; ) {
+				if ( timers[ index ].elem === this &&
+					( type == null || timers[ index ].queue === type ) ) {
+
+					timers[ index ].anim.stop( gotoEnd );
+					dequeue = false;
+					timers.splice( index, 1 );
+				}
+			}
+
+			// Start the next in the queue if the last step wasn't forced.
+			// Timers currently will call their complete callbacks, which
+			// will dequeue but only if they were gotoEnd.
+			if ( dequeue || !gotoEnd ) {
+				jQuery.dequeue( this, type );
+			}
+		} );
+	},
+	finish: function( type ) {
+		if ( type !== false ) {
+			type = type || "fx";
+		}
+		return this.each( function() {
+			var index,
+				data = dataPriv.get( this ),
+				queue = data[ type + "queue" ],
+				hooks = data[ type + "queueHooks" ],
+				timers = jQuery.timers,
+				length = queue ? queue.length : 0;
+
+			// Enable finishing flag on private data
+			data.finish = true;
+
+			// Empty the queue first
+			jQuery.queue( this, type, [] );
+
+			if ( hooks && hooks.stop ) {
+				hooks.stop.call( this, true );
+			}
+
+			// Look for any active animations, and finish them
+			for ( index = timers.length; index--; ) {
+				if ( timers[ index ].elem === this && timers[ index ].queue === type ) {
+					timers[ index ].anim.stop( true );
+					timers.splice( index, 1 );
+				}
+			}
+
+			// Look for any animations in the old queue and finish them
+			for ( index = 0; index < length; index++ ) {
+				if ( queue[ index ] && queue[ index ].finish ) {
+					queue[ index ].finish.call( this );
+				}
+			}
+
+			// Turn off finishing flag
+			delete data.finish;
+		} );
+	}
+} );
+
+jQuery.each( [ "toggle", "show", "hide" ], function( _i, name ) {
+	var cssFn = jQuery.fn[ name ];
+	jQuery.fn[ name ] = function( speed, easing, callback ) {
+		return speed == null || typeof speed === "boolean" ?
+			cssFn.apply( this, arguments ) :
+			this.animate( genFx( name, true ), speed, easing, callback );
+	};
+} );
+
+// Generate shortcuts for custom animations
+jQuery.each( {
+	slideDown: genFx( "show" ),
+	slideUp: genFx( "hide" ),
+	slideToggle: genFx( "toggle" ),
+	fadeIn: { opacity: "show" },
+	fadeOut: { opacity: "hide" },
+	fadeToggle: { opacity: "toggle" }
+}, function( name, props ) {
+	jQuery.fn[ name ] = function( speed, easing, callback ) {
+		return this.animate( props, speed, easing, callback );
+	};
+} );
+
+jQuery.timers = [];
+jQuery.fx.tick = function() {
+	var timer,
+		i = 0,
+		timers = jQuery.timers;
+
+	fxNow = Date.now();
+
+	for ( ; i < timers.length; i++ ) {
+		timer = timers[ i ];
+
+		// Run the timer and safely remove it when done (allowing for external removal)
+		if ( !timer() && timers[ i ] === timer ) {
+			timers.splice( i--, 1 );
+		}
+	}
+
+	if ( !timers.length ) {
+		jQuery.fx.stop();
+	}
+	fxNow = undefined;
+};
+
+jQuery.fx.timer = function( timer ) {
+	jQuery.timers.push( timer );
+	jQuery.fx.start();
+};
+
+jQuery.fx.interval = 13;
+jQuery.fx.start = function() {
+	if ( inProgress ) {
+		return;
+	}
+
+	inProgress = true;
+	schedule();
+};
+
+jQuery.fx.stop = function() {
+	inProgress = null;
+};
+
+jQuery.fx.speeds = {
+	slow: 600,
+	fast: 200,
+
+	// Default speed
+	_default: 400
+};
+
+
+// Based off of the plugin by Clint Helfers, with permission.
+// https://web.archive.org/web/20100324014747/http://blindsignals.com/index.php/2009/07/jquery-delay/
+jQuery.fn.delay = function( time, type ) {
+	time = jQuery.fx ? jQuery.fx.speeds[ time ] || time : time;
+	type = type || "fx";
+
+	return this.queue( type, function( next, hooks ) {
+		var timeout = window.setTimeout( next, time );
+		hooks.stop = function() {
+			window.clearTimeout( timeout );
+		};
+	} );
+};
+
+
+( function() {
+	var input = document.createElement( "input" ),
+		select = document.createElement( "select" ),
+		opt = select.appendChild( document.createElement( "option" ) );
+
+	input.type = "checkbox";
+
+	// Support: Android <=4.3 only
+	// Default value for a checkbox should be "on"
+	support.checkOn = input.value !== "";
+
+	// Support: IE <=11 only
+	// Must access selectedIndex to make default options select
+	support.optSelected = opt.selected;
+
+	// Support: IE <=11 only
+	// An input loses its value after becoming a radio
+	input = document.createElement( "input" );
+	input.value = "t";
+	input.type = "radio";
+	support.radioValue = input.value === "t";
+} )();
+
+
+var boolHook,
+	attrHandle = jQuery.expr.attrHandle;
+
+jQuery.fn.extend( {
+	attr: function( name, value ) {
+		return access( this, jQuery.attr, name, value, arguments.length > 1 );
+	},
+
+	removeAttr: function( name ) {
+		return this.each( function() {
+			jQuery.removeAttr( this, name );
+		} );
+	}
+} );
+
+jQuery.extend( {
+	attr: function( elem, name, value ) {
+		var ret, hooks,
+			nType = elem.nodeType;
+
+		// Don't get/set attributes on text, comment and attribute nodes
+		if ( nType === 3 || nType === 8 || nType === 2 ) {
+			return;
+		}
+
+		// Fallback to prop when attributes are not supported
+		if ( typeof elem.getAttribute === "undefined" ) {
+			return jQuery.prop( elem, name, value );
+		}
+
+		// Attribute hooks are determined by the lowercase version
+		// Grab necessary hook if one is defined
+		if ( nType !== 1 || !jQuery.isXMLDoc( elem ) ) {
+			hooks = jQuery.attrHooks[ name.toLowerCase() ] ||
+				( jQuery.expr.match.bool.test( name ) ? boolHook : undefined );
+		}
+
+		if ( value !== undefined ) {
+			if ( value === null ) {
+				jQuery.removeAttr( elem, name );
+				return;
+			}
+
+			if ( hooks && "set" in hooks &&
+				( ret = hooks.set( elem, value, name ) ) !== undefined ) {
+				return ret;
+			}
+
+			elem.setAttribute( name, value + "" );
+			return value;
+		}
+
+		if ( hooks && "get" in hooks && ( ret = hooks.get( elem, name ) ) !== null ) {
+			return ret;
+		}
+
+		ret = jQuery.find.attr( elem, name );
+
+		// Non-existent attributes return null, we normalize to undefined
+		return ret == null ? undefined : ret;
+	},
+
+	attrHooks: {
+		type: {
+			set: function( elem, value ) {
+				if ( !support.radioValue && value === "radio" &&
+					nodeName( elem, "input" ) ) {
+					var val = elem.value;
+					elem.setAttribute( "type", value );
+					if ( val ) {
+						elem.value = val;
+					}
+					return value;
+				}
+			}
+		}
+	},
+
+	removeAttr: function( elem, value ) {
+		var name,
+			i = 0,
+
+			// Attribute names can contain non-HTML whitespace characters
+			// https://html.spec.whatwg.org/multipage/syntax.html#attributes-2
+			attrNames = value && value.match( rnothtmlwhite );
+
+		if ( attrNames && elem.nodeType === 1 ) {
+			while ( ( name = attrNames[ i++ ] ) ) {
+				elem.removeAttribute( name );
+			}
+		}
+	}
+} );
+
+// Hooks for boolean attributes
+boolHook = {
+	set: function( elem, value, name ) {
+		if ( value === false ) {
+
+			// Remove boolean attributes when set to false
+			jQuery.removeAttr( elem, name );
+		} else {
+			elem.setAttribute( name, name );
+		}
+		return name;
+	}
+};
+
+jQuery.each( jQuery.expr.match.bool.source.match( /\w+/g ), function( _i, name ) {
+	var getter = attrHandle[ name ] || jQuery.find.attr;
+
+	attrHandle[ name ] = function( elem, name, isXML ) {
+		var ret, handle,
+			lowercaseName = name.toLowerCase();
+
+		if ( !isXML ) {
+
+			// Avoid an infinite loop by temporarily removing this function from the getter
+			handle = attrHandle[ lowercaseName ];
+			attrHandle[ lowercaseName ] = ret;
+			ret = getter( elem, name, isXML ) != null ?
+				lowercaseName :
+				null;
+			attrHandle[ lowercaseName ] = handle;
+		}
+		return ret;
+	};
+} );
+
+
+
+
+var rfocusable = /^(?:input|select|textarea|button)$/i,
+	rclickable = /^(?:a|area)$/i;
+
+jQuery.fn.extend( {
+	prop: function( name, value ) {
+		return access( this, jQuery.prop, name, value, arguments.length > 1 );
+	},
+
+	removeProp: function( name ) {
+		return this.each( function() {
+			delete this[ jQuery.propFix[ name ] || name ];
+		} );
+	}
+} );
+
+jQuery.extend( {
+	prop: function( elem, name, value ) {
+		var ret, hooks,
+			nType = elem.nodeType;
+
+		// Don't get/set properties on text, comment and attribute nodes
+		if ( nType === 3 || nType === 8 || nType === 2 ) {
+			return;
+		}
+
+		if ( nType !== 1 || !jQuery.isXMLDoc( elem ) ) {
+
+			// Fix name and attach hooks
+			name = jQuery.propFix[ name ] || name;
+			hooks = jQuery.propHooks[ name ];
+		}
+
+		if ( value !== undefined ) {
+			if ( hooks && "set" in hooks &&
+				( ret = hooks.set( elem, value, name ) ) !== undefined ) {
+				return ret;
+			}
+
+			return ( elem[ name ] = value );
+		}
+
+		if ( hooks && "get" in hooks && ( ret = hooks.get( elem, name ) ) !== null ) {
+			return ret;
+		}
+
+		return elem[ name ];
+	},
+
+	propHooks: {
+		tabIndex: {
+			get: function( elem ) {
+
+				// Support: IE <=9 - 11 only
+				// elem.tabIndex doesn't always return the
+				// correct value when it hasn't been explicitly set
+				// https://web.archive.org/web/20141116233347/http://fluidproject.org/blog/2008/01/09/getting-setting-and-removing-tabindex-values-with-javascript/
+				// Use proper attribute retrieval(#12072)
+				var tabindex = jQuery.find.attr( elem, "tabindex" );
+
+				if ( tabindex ) {
+					return parseInt( tabindex, 10 );
+				}
+
+				if (
+					rfocusable.test( elem.nodeName ) ||
+					rclickable.test( elem.nodeName ) &&
+					elem.href
+				) {
+					return 0;
+				}
+
+				return -1;
+			}
+		}
+	},
+
+	propFix: {
+		"for": "htmlFor",
+		"class": "className"
+	}
+} );
+
+// Support: IE <=11 only
+// Accessing the selectedIndex property
+// forces the browser to respect setting selected
+// on the option
+// The getter ensures a default option is selected
+// when in an optgroup
+// eslint rule "no-unused-expressions" is disabled for this code
+// since it considers such accessions noop
+if ( !support.optSelected ) {
+	jQuery.propHooks.selected = {
+		get: function( elem ) {
+
+			/* eslint no-unused-expressions: "off" */
+
+			var parent = elem.parentNode;
+			if ( parent && parent.parentNode ) {
+				parent.parentNode.selectedIndex;
+			}
+			return null;
+		},
+		set: function( elem ) {
+
+			/* eslint no-unused-expressions: "off" */
+
+			var parent = elem.parentNode;
+			if ( parent ) {
+				parent.selectedIndex;
+
+				if ( parent.parentNode ) {
+					parent.parentNode.selectedIndex;
+				}
+			}
+		}
+	};
+}
+
+jQuery.each( [
+	"tabIndex",
+	"readOnly",
+	"maxLength",
+	"cellSpacing",
+	"cellPadding",
+	"rowSpan",
+	"colSpan",
+	"useMap",
+	"frameBorder",
+	"contentEditable"
+], function() {
+	jQuery.propFix[ this.toLowerCase() ] = this;
+} );
+
+
+
+
+	// Strip and collapse whitespace according to HTML spec
+	// https://infra.spec.whatwg.org/#strip-and-collapse-ascii-whitespace
+	function stripAndCollapse( value ) {
+		var tokens = value.match( rnothtmlwhite ) || [];
+		return tokens.join( " " );
+	}
+
+
+function getClass( elem ) {
+	return elem.getAttribute && elem.getAttribute( "class" ) || "";
+}
+
+function classesToArray( value ) {
+	if ( Array.isArray( value ) ) {
+		return value;
+	}
+	if ( typeof value === "string" ) {
+		return value.match( rnothtmlwhite ) || [];
+	}
+	return [];
+}
+
+jQuery.fn.extend( {
+	addClass: function( value ) {
+		var classes, elem, cur, curValue, clazz, j, finalValue,
+			i = 0;
+
+		if ( isFunction( value ) ) {
+			return this.each( function( j ) {
+				jQuery( this ).addClass( value.call( this, j, getClass( this ) ) );
+			} );
+		}
+
+		classes = classesToArray( value );
+
+		if ( classes.length ) {
+			while ( ( elem = this[ i++ ] ) ) {
+				curValue = getClass( elem );
+				cur = elem.nodeType === 1 && ( " " + stripAndCollapse( curValue ) + " " );
+
+				if ( cur ) {
+					j = 0;
+					while ( ( clazz = classes[ j++ ] ) ) {
+						if ( cur.indexOf( " " + clazz + " " ) < 0 ) {
+							cur += clazz + " ";
+						}
+					}
+
+					// Only assign if different to avoid unneeded rendering.
+					finalValue = stripAndCollapse( cur );
+					if ( curValue !== finalValue ) {
+						elem.setAttribute( "class", finalValue );
+					}
+				}
+			}
+		}
+
+		return this;
+	},
+
+	removeClass: function( value ) {
+		var classes, elem, cur, curValue, clazz, j, finalValue,
+			i = 0;
+
+		if ( isFunction( value ) ) {
+			return this.each( function( j ) {
+				jQuery( this ).removeClass( value.call( this, j, getClass( this ) ) );
+			} );
+		}
+
+		if ( !arguments.length ) {
+			return this.attr( "class", "" );
+		}
+
+		classes = classesToArray( value );
+
+		if ( classes.length ) {
+			while ( ( elem = this[ i++ ] ) ) {
+				curValue = getClass( elem );
+
+				// This expression is here for better compressibility (see addClass)
+				cur = elem.nodeType === 1 && ( " " + stripAndCollapse( curValue ) + " " );
+
+				if ( cur ) {
+					j = 0;
+					while ( ( clazz = classes[ j++ ] ) ) {
+
+						// Remove *all* instances
+						while ( cur.indexOf( " " + clazz + " " ) > -1 ) {
+							cur = cur.replace( " " + clazz + " ", " " );
+						}
+					}
+
+					// Only assign if different to avoid unneeded rendering.
+					finalValue = stripAndCollapse( cur );
+					if ( curValue !== finalValue ) {
+						elem.setAttribute( "class", finalValue );
+					}
+				}
+			}
+		}
+
+		return this;
+	},
+
+	toggleClass: function( value, stateVal ) {
+		var type = typeof value,
+			isValidValue = type === "string" || Array.isArray( value );
+
+		if ( typeof stateVal === "boolean" && isValidValue ) {
+			return stateVal ? this.addClass( value ) : this.removeClass( value );
+		}
+
+		if ( isFunction( value ) ) {
+			return this.each( function( i ) {
+				jQuery( this ).toggleClass(
+					value.call( this, i, getClass( this ), stateVal ),
+					stateVal
+				);
+			} );
+		}
+
+		return this.each( function() {
+			var className, i, self, classNames;
+
+			if ( isValidValue ) {
+
+				// Toggle individual class names
+				i = 0;
+				self = jQuery( this );
+				classNames = classesToArray( value );
+
+				while ( ( className = classNames[ i++ ] ) ) {
+
+					// Check each className given, space separated list
+					if ( self.hasClass( className ) ) {
+						self.removeClass( className );
+					} else {
+						self.addClass( className );
+					}
+				}
+
+			// Toggle whole class name
+			} else if ( value === undefined || type === "boolean" ) {
+				className = getClass( this );
+				if ( className ) {
+
+					// Store className if set
+					dataPriv.set( this, "__className__", className );
+				}
+
+				// If the element has a class name or if we're passed `false`,
+				// then remove the whole classname (if there was one, the above saved it).
+				// Otherwise bring back whatever was previously saved (if anything),
+				// falling back to the empty string if nothing was stored.
+				if ( this.setAttribute ) {
+					this.setAttribute( "class",
+						className || value === false ?
+							"" :
+							dataPriv.get( this, "__className__" ) || ""
+					);
+				}
+			}
+		} );
+	},
+
+	hasClass: function( selector ) {
+		var className, elem,
+			i = 0;
+
+		className = " " + selector + " ";
+		while ( ( elem = this[ i++ ] ) ) {
+			if ( elem.nodeType === 1 &&
+				( " " + stripAndCollapse( getClass( elem ) ) + " " ).indexOf( className ) > -1 ) {
+				return true;
+			}
+		}
+
+		return false;
+	}
+} );
+
+
+
+
+var rreturn = /\r/g;
+
+jQuery.fn.extend( {
+	val: function( value ) {
+		var hooks, ret, valueIsFunction,
+			elem = this[ 0 ];
+
+		if ( !arguments.length ) {
+			if ( elem ) {
+				hooks = jQuery.valHooks[ elem.type ] ||
+					jQuery.valHooks[ elem.nodeName.toLowerCase() ];
+
+				if ( hooks &&
+					"get" in hooks &&
+					( ret = hooks.get( elem, "value" ) ) !== undefined
+				) {
+					return ret;
+				}
+
+				ret = elem.value;
+
+				// Handle most common string cases
+				if ( typeof ret === "string" ) {
+					return ret.replace( rreturn, "" );
+				}
+
+				// Handle cases where value is null/undef or number
+				return ret == null ? "" : ret;
+			}
+
+			return;
+		}
+
+		valueIsFunction = isFunction( value );
+
+		return this.each( function( i ) {
+			var val;
+
+			if ( this.nodeType !== 1 ) {
+				return;
+			}
+
+			if ( valueIsFunction ) {
+				val = value.call( this, i, jQuery( this ).val() );
+			} else {
+				val = value;
+			}
+
+			// Treat null/undefined as ""; convert numbers to string
+			if ( val == null ) {
+				val = "";
+
+			} else if ( typeof val === "number" ) {
+				val += "";
+
+			} else if ( Array.isArray( val ) ) {
+				val = jQuery.map( val, function( value ) {
+					return value == null ? "" : value + "";
+				} );
+			}
+
+			hooks = jQuery.valHooks[ this.type ] || jQuery.valHooks[ this.nodeName.toLowerCase() ];
+
+			// If set returns undefined, fall back to normal setting
+			if ( !hooks || !( "set" in hooks ) || hooks.set( this, val, "value" ) === undefined ) {
+				this.value = val;
+			}
+		} );
+	}
+} );
+
+jQuery.extend( {
+	valHooks: {
+		option: {
+			get: function( elem ) {
+
+				var val = jQuery.find.attr( elem, "value" );
+				return val != null ?
+					val :
+
+					// Support: IE <=10 - 11 only
+					// option.text throws exceptions (#14686, #14858)
+					// Strip and collapse whitespace
+					// https://html.spec.whatwg.org/#strip-and-collapse-whitespace
+					stripAndCollapse( jQuery.text( elem ) );
+			}
+		},
+		select: {
+			get: function( elem ) {
+				var value, option, i,
+					options = elem.options,
+					index = elem.selectedIndex,
+					one = elem.type === "select-one",
+					values = one ? null : [],
+					max = one ? index + 1 : options.length;
+
+				if ( index < 0 ) {
+					i = max;
+
+				} else {
+					i = one ? index : 0;
+				}
+
+				// Loop through all the selected options
+				for ( ; i < max; i++ ) {
+					option = options[ i ];
+
+					// Support: IE <=9 only
+					// IE8-9 doesn't update selected after form reset (#2551)
+					if ( ( option.selected || i === index ) &&
+
+							// Don't return options that are disabled or in a disabled optgroup
+							!option.disabled &&
+							( !option.parentNode.disabled ||
+								!nodeName( option.parentNode, "optgroup" ) ) ) {
+
+						// Get the specific value for the option
+						value = jQuery( option ).val();
+
+						// We don't need an array for one selects
+						if ( one ) {
+							return value;
+						}
+
+						// Multi-Selects return an array
+						values.push( value );
+					}
+				}
+
+				return values;
+			},
+
+			set: function( elem, value ) {
+				var optionSet, option,
+					options = elem.options,
+					values = jQuery.makeArray( value ),
+					i = options.length;
+
+				while ( i-- ) {
+					option = options[ i ];
+
+					/* eslint-disable no-cond-assign */
+
+					if ( option.selected =
+						jQuery.inArray( jQuery.valHooks.option.get( option ), values ) > -1
+					) {
+						optionSet = true;
+					}
+
+					/* eslint-enable no-cond-assign */
+				}
+
+				// Force browsers to behave consistently when non-matching value is set
+				if ( !optionSet ) {
+					elem.selectedIndex = -1;
+				}
+				return values;
+			}
+		}
+	}
+} );
+
+// Radios and checkboxes getter/setter
+jQuery.each( [ "radio", "checkbox" ], function() {
+	jQuery.valHooks[ this ] = {
+		set: function( elem, value ) {
+			if ( Array.isArray( value ) ) {
+				return ( elem.checked = jQuery.inArray( jQuery( elem ).val(), value ) > -1 );
+			}
+		}
+	};
+	if ( !support.checkOn ) {
+		jQuery.valHooks[ this ].get = function( elem ) {
+			return elem.getAttribute( "value" ) === null ? "on" : elem.value;
+		};
+	}
+} );
+
+
+
+
+// Return jQuery for attributes-only inclusion
+
+
+support.focusin = "onfocusin" in window;
+
+
+var rfocusMorph = /^(?:focusinfocus|focusoutblur)$/,
+	stopPropagationCallback = function( e ) {
+		e.stopPropagation();
+	};
+
+jQuery.extend( jQuery.event, {
+
+	trigger: function( event, data, elem, onlyHandlers ) {
+
+		var i, cur, tmp, bubbleType, ontype, handle, special, lastElement,
+			eventPath = [ elem || document ],
+			type = hasOwn.call( event, "type" ) ? event.type : event,
+			namespaces = hasOwn.call( event, "namespace" ) ? event.namespace.split( "." ) : [];
+
+		cur = lastElement = tmp = elem = elem || document;
+
+		// Don't do events on text and comment nodes
+		if ( elem.nodeType === 3 || elem.nodeType === 8 ) {
+			return;
+		}
+
+		// focus/blur morphs to focusin/out; ensure we're not firing them right now
+		if ( rfocusMorph.test( type + jQuery.event.triggered ) ) {
+			return;
+		}
+
+		if ( type.indexOf( "." ) > -1 ) {
+
+			// Namespaced trigger; create a regexp to match event type in handle()
+			namespaces = type.split( "." );
+			type = namespaces.shift();
+			namespaces.sort();
+		}
+		ontype = type.indexOf( ":" ) < 0 && "on" + type;
+
+		// Caller can pass in a jQuery.Event object, Object, or just an event type string
+		event = event[ jQuery.expando ] ?
+			event :
+			new jQuery.Event( type, typeof event === "object" && event );
+
+		// Trigger bitmask: & 1 for native handlers; & 2 for jQuery (always true)
+		event.isTrigger = onlyHandlers ? 2 : 3;
+		event.namespace = namespaces.join( "." );
+		event.rnamespace = event.namespace ?
+			new RegExp( "(^|\\.)" + namespaces.join( "\\.(?:.*\\.|)" ) + "(\\.|$)" ) :
+			null;
+
+		// Clean up the event in case it is being reused
+		event.result = undefined;
+		if ( !event.target ) {
+			event.target = elem;
+		}
+
+		// Clone any incoming data and prepend the event, creating the handler arg list
+		data = data == null ?
+			[ event ] :
+			jQuery.makeArray( data, [ event ] );
+
+		// Allow special events to draw outside the lines
+		special = jQuery.event.special[ type ] || {};
+		if ( !onlyHandlers && special.trigger && special.trigger.apply( elem, data ) === false ) {
+			return;
+		}
+
+		// Determine event propagation path in advance, per W3C events spec (#9951)
+		// Bubble up to document, then to window; watch for a global ownerDocument var (#9724)
+		if ( !onlyHandlers && !special.noBubble && !isWindow( elem ) ) {
+
+			bubbleType = special.delegateType || type;
+			if ( !rfocusMorph.test( bubbleType + type ) ) {
+				cur = cur.parentNode;
+			}
+			for ( ; cur; cur = cur.parentNode ) {
+				eventPath.push( cur );
+				tmp = cur;
+			}
+
+			// Only add window if we got to document (e.g., not plain obj or detached DOM)
+			if ( tmp === ( elem.ownerDocument || document ) ) {
+				eventPath.push( tmp.defaultView || tmp.parentWindow || window );
+			}
+		}
+
+		// Fire handlers on the event path
+		i = 0;
+		while ( ( cur = eventPath[ i++ ] ) && !event.isPropagationStopped() ) {
+			lastElement = cur;
+			event.type = i > 1 ?
+				bubbleType :
+				special.bindType || type;
+
+			// jQuery handler
+			handle = ( dataPriv.get( cur, "events" ) || Object.create( null ) )[ event.type ] &&
+				dataPriv.get( cur, "handle" );
+			if ( handle ) {
+				handle.apply( cur, data );
+			}
+
+			// Native handler
+			handle = ontype && cur[ ontype ];
+			if ( handle && handle.apply && acceptData( cur ) ) {
+				event.result = handle.apply( cur, data );
+				if ( event.result === false ) {
+					event.preventDefault();
+				}
+			}
+		}
+		event.type = type;
+
+		// If nobody prevented the default action, do it now
+		if ( !onlyHandlers && !event.isDefaultPrevented() ) {
+
+			if ( ( !special._default ||
+				special._default.apply( eventPath.pop(), data ) === false ) &&
+				acceptData( elem ) ) {
+
+				// Call a native DOM method on the target with the same name as the event.
+				// Don't do default actions on window, that's where global variables be (#6170)
+				if ( ontype && isFunction( elem[ type ] ) && !isWindow( elem ) ) {
+
+					// Don't re-trigger an onFOO event when we call its FOO() method
+					tmp = elem[ ontype ];
+
+					if ( tmp ) {
+						elem[ ontype ] = null;
+					}
+
+					// Prevent re-triggering of the same event, since we already bubbled it above
+					jQuery.event.triggered = type;
+
+					if ( event.isPropagationStopped() ) {
+						lastElement.addEventListener( type, stopPropagationCallback );
+					}
+
+					elem[ type ]();
+
+					if ( event.isPropagationStopped() ) {
+						lastElement.removeEventListener( type, stopPropagationCallback );
+					}
+
+					jQuery.event.triggered = undefined;
+
+					if ( tmp ) {
+						elem[ ontype ] = tmp;
+					}
+				}
+			}
+		}
+
+		return event.result;
+	},
+
+	// Piggyback on a donor event to simulate a different one
+	// Used only for `focus(in | out)` events
+	simulate: function( type, elem, event ) {
+		var e = jQuery.extend(
+			new jQuery.Event(),
+			event,
+			{
+				type: type,
+				isSimulated: true
+			}
+		);
+
+		jQuery.event.trigger( e, null, elem );
+	}
+
+} );
+
+jQuery.fn.extend( {
+
+	trigger: function( type, data ) {
+		return this.each( function() {
+			jQuery.event.trigger( type, data, this );
+		} );
+	},
+	triggerHandler: function( type, data ) {
+		var elem = this[ 0 ];
+		if ( elem ) {
+			return jQuery.event.trigger( type, data, elem, true );
+		}
+	}
+} );
+
+
+// Support: Firefox <=44
+// Firefox doesn't have focus(in | out) events
+// Related ticket - https://bugzilla.mozilla.org/show_bug.cgi?id=687787
+//
+// Support: Chrome <=48 - 49, Safari <=9.0 - 9.1
+// focus(in | out) events fire after focus & blur events,
+// which is spec violation - http://www.w3.org/TR/DOM-Level-3-Events/#events-focusevent-event-order
+// Related ticket - https://bugs.chromium.org/p/chromium/issues/detail?id=449857
+if ( !support.focusin ) {
+	jQuery.each( { focus: "focusin", blur: "focusout" }, function( orig, fix ) {
+
+		// Attach a single capturing handler on the document while someone wants focusin/focusout
+		var handler = function( event ) {
+			jQuery.event.simulate( fix, event.target, jQuery.event.fix( event ) );
+		};
+
+		jQuery.event.special[ fix ] = {
+			setup: function() {
+
+				// Handle: regular nodes (via `this.ownerDocument`), window
+				// (via `this.document`) & document (via `this`).
+				var doc = this.ownerDocument || this.document || this,
+					attaches = dataPriv.access( doc, fix );
+
+				if ( !attaches ) {
+					doc.addEventListener( orig, handler, true );
+				}
+				dataPriv.access( doc, fix, ( attaches || 0 ) + 1 );
+			},
+			teardown: function() {
+				var doc = this.ownerDocument || this.document || this,
+					attaches = dataPriv.access( doc, fix ) - 1;
+
+				if ( !attaches ) {
+					doc.removeEventListener( orig, handler, true );
+					dataPriv.remove( doc, fix );
+
+				} else {
+					dataPriv.access( doc, fix, attaches );
+				}
+			}
+		};
+	} );
+}
+var location = window.location;
+
+var nonce = { guid: Date.now() };
+
+var rquery = ( /\?/ );
+
+
+
+// Cross-browser xml parsing
+jQuery.parseXML = function( data ) {
+	var xml, parserErrorElem;
+	if ( !data || typeof data !== "string" ) {
+		return null;
+	}
+
+	// Support: IE 9 - 11 only
+	// IE throws on parseFromString with invalid input.
+	try {
+		xml = ( new window.DOMParser() ).parseFromString( data, "text/xml" );
+	} catch ( e ) {}
+
+	parserErrorElem = xml && xml.getElementsByTagName( "parsererror" )[ 0 ];
+	if ( !xml || parserErrorElem ) {
+		jQuery.error( "Invalid XML: " + (
+			parserErrorElem ?
+				jQuery.map( parserErrorElem.childNodes, function( el ) {
+					return el.textContent;
+				} ).join( "\n" ) :
+				data
+		) );
+	}
+	return xml;
+};
+
+
+var
+	rbracket = /\[\]$/,
+	rCRLF = /\r?\n/g,
+	rsubmitterTypes = /^(?:submit|button|image|reset|file)$/i,
+	rsubmittable = /^(?:input|select|textarea|keygen)/i;
+
+function buildParams( prefix, obj, traditional, add ) {
+	var name;
+
+	if ( Array.isArray( obj ) ) {
+
+		// Serialize array item.
+		jQuery.each( obj, function( i, v ) {
+			if ( traditional || rbracket.test( prefix ) ) {
+
+				// Treat each array item as a scalar.
+				add( prefix, v );
+
+			} else {
+
+				// Item is non-scalar (array or object), encode its numeric index.
+				buildParams(
+					prefix + "[" + ( typeof v === "object" && v != null ? i : "" ) + "]",
+					v,
+					traditional,
+					add
+				);
+			}
+		} );
+
+	} else if ( !traditional && toType( obj ) === "object" ) {
+
+		// Serialize object item.
+		for ( name in obj ) {
+			buildParams( prefix + "[" + name + "]", obj[ name ], traditional, add );
+		}
+
+	} else {
+
+		// Serialize scalar item.
+		add( prefix, obj );
+	}
+}
+
+// Serialize an array of form elements or a set of
+// key/values into a query string
+jQuery.param = function( a, traditional ) {
+	var prefix,
+		s = [],
+		add = function( key, valueOrFunction ) {
+
+			// If value is a function, invoke it and use its return value
+			var value = isFunction( valueOrFunction ) ?
+				valueOrFunction() :
+				valueOrFunction;
+
+			s[ s.length ] = encodeURIComponent( key ) + "=" +
+				encodeURIComponent( value == null ? "" : value );
+		};
+
+	if ( a == null ) {
+		return "";
+	}
+
+	// If an array was passed in, assume that it is an array of form elements.
+	if ( Array.isArray( a ) || ( a.jquery && !jQuery.isPlainObject( a ) ) ) {
+
+		// Serialize the form elements
+		jQuery.each( a, function() {
+			add( this.name, this.value );
+		} );
+
+	} else {
+
+		// If traditional, encode the "old" way (the way 1.3.2 or older
+		// did it), otherwise encode params recursively.
+		for ( prefix in a ) {
+			buildParams( prefix, a[ prefix ], traditional, add );
+		}
+	}
+
+	// Return the resulting serialization
+	return s.join( "&" );
+};
+
+jQuery.fn.extend( {
+	serialize: function() {
+		return jQuery.param( this.serializeArray() );
+	},
+	serializeArray: function() {
+		return this.map( function() {
+
+			// Can add propHook for "elements" to filter or add form elements
+			var elements = jQuery.prop( this, "elements" );
+			return elements ? jQuery.makeArray( elements ) : this;
+		} ).filter( function() {
+			var type = this.type;
+
+			// Use .is( ":disabled" ) so that fieldset[disabled] works
+			return this.name && !jQuery( this ).is( ":disabled" ) &&
+				rsubmittable.test( this.nodeName ) && !rsubmitterTypes.test( type ) &&
+				( this.checked || !rcheckableType.test( type ) );
+		} ).map( function( _i, elem ) {
+			var val = jQuery( this ).val();
+
+			if ( val == null ) {
+				return null;
+			}
+
+			if ( Array.isArray( val ) ) {
+				return jQuery.map( val, function( val ) {
+					return { name: elem.name, value: val.replace( rCRLF, "\r\n" ) };
+				} );
+			}
+
+			return { name: elem.name, value: val.replace( rCRLF, "\r\n" ) };
+		} ).get();
+	}
+} );
+
+
+var
+	r20 = /%20/g,
+	rhash = /#.*$/,
+	rantiCache = /([?&])_=[^&]*/,
+	rheaders = /^(.*?):[ \t]*([^\r\n]*)$/mg,
+
+	// #7653, #8125, #8152: local protocol detection
+	rlocalProtocol = /^(?:about|app|app-storage|.+-extension|file|res|widget):$/,
+	rnoContent = /^(?:GET|HEAD)$/,
+	rprotocol = /^\/\//,
+
+	/* Prefilters
+	 * 1) They are useful to introduce custom dataTypes (see ajax/jsonp.js for an example)
+	 * 2) These are called:
+	 *    - BEFORE asking for a transport
+	 *    - AFTER param serialization (s.data is a string if s.processData is true)
+	 * 3) key is the dataType
+	 * 4) the catchall symbol "*" can be used
+	 * 5) execution will start with transport dataType and THEN continue down to "*" if needed
+	 */
+	prefilters = {},
+
+	/* Transports bindings
+	 * 1) key is the dataType
+	 * 2) the catchall symbol "*" can be used
+	 * 3) selection will start with transport dataType and THEN go to "*" if needed
+	 */
+	transports = {},
+
+	// Avoid comment-prolog char sequence (#10098); must appease lint and evade compression
+	allTypes = "*/".concat( "*" ),
+
+	// Anchor tag for parsing the document origin
+	originAnchor = document.createElement( "a" );
+
+originAnchor.href = location.href;
+
+// Base "constructor" for jQuery.ajaxPrefilter and jQuery.ajaxTransport
+function addToPrefiltersOrTransports( structure ) {
+
+	// dataTypeExpression is optional and defaults to "*"
+	return function( dataTypeExpression, func ) {
+
+		if ( typeof dataTypeExpression !== "string" ) {
+			func = dataTypeExpression;
+			dataTypeExpression = "*";
+		}
+
+		var dataType,
+			i = 0,
+			dataTypes = dataTypeExpression.toLowerCase().match( rnothtmlwhite ) || [];
+
+		if ( isFunction( func ) ) {
+
+			// For each dataType in the dataTypeExpression
+			while ( ( dataType = dataTypes[ i++ ] ) ) {
+
+				// Prepend if requested
+				if ( dataType[ 0 ] === "+" ) {
+					dataType = dataType.slice( 1 ) || "*";
+					( structure[ dataType ] = structure[ dataType ] || [] ).unshift( func );
+
+				// Otherwise append
+				} else {
+					( structure[ dataType ] = structure[ dataType ] || [] ).push( func );
+				}
+			}
+		}
+	};
+}
+
+// Base inspection function for prefilters and transports
+function inspectPrefiltersOrTransports( structure, options, originalOptions, jqXHR ) {
+
+	var inspected = {},
+		seekingTransport = ( structure === transports );
+
+	function inspect( dataType ) {
+		var selected;
+		inspected[ dataType ] = true;
+		jQuery.each( structure[ dataType ] || [], function( _, prefilterOrFactory ) {
+			var dataTypeOrTransport = prefilterOrFactory( options, originalOptions, jqXHR );
+			if ( typeof dataTypeOrTransport === "string" &&
+				!seekingTransport && !inspected[ dataTypeOrTransport ] ) {
+
+				options.dataTypes.unshift( dataTypeOrTransport );
+				inspect( dataTypeOrTransport );
+				return false;
+			} else if ( seekingTransport ) {
+				return !( selected = dataTypeOrTransport );
+			}
+		} );
+		return selected;
+	}
+
+	return inspect( options.dataTypes[ 0 ] ) || !inspected[ "*" ] && inspect( "*" );
+}
+
+// A special extend for ajax options
+// that takes "flat" options (not to be deep extended)
+// Fixes #9887
+function ajaxExtend( target, src ) {
+	var key, deep,
+		flatOptions = jQuery.ajaxSettings.flatOptions || {};
+
+	for ( key in src ) {
+		if ( src[ key ] !== undefined ) {
+			( flatOptions[ key ] ? target : ( deep || ( deep = {} ) ) )[ key ] = src[ key ];
+		}
+	}
+	if ( deep ) {
+		jQuery.extend( true, target, deep );
+	}
+
+	return target;
+}
+
+/* Handles responses to an ajax request:
+ * - finds the right dataType (mediates between content-type and expected dataType)
+ * - returns the corresponding response
+ */
+function ajaxHandleResponses( s, jqXHR, responses ) {
+
+	var ct, type, finalDataType, firstDataType,
+		contents = s.contents,
+		dataTypes = s.dataTypes;
+
+	// Remove auto dataType and get content-type in the process
+	while ( dataTypes[ 0 ] === "*" ) {
+		dataTypes.shift();
+		if ( ct === undefined ) {
+			ct = s.mimeType || jqXHR.getResponseHeader( "Content-Type" );
+		}
+	}
+
+	// Check if we're dealing with a known content-type
+	if ( ct ) {
+		for ( type in contents ) {
+			if ( contents[ type ] && contents[ type ].test( ct ) ) {
+				dataTypes.unshift( type );
+				break;
+			}
+		}
+	}
+
+	// Check to see if we have a response for the expected dataType
+	if ( dataTypes[ 0 ] in responses ) {
+		finalDataType = dataTypes[ 0 ];
+	} else {
+
+		// Try convertible dataTypes
+		for ( type in responses ) {
+			if ( !dataTypes[ 0 ] || s.converters[ type + " " + dataTypes[ 0 ] ] ) {
+				finalDataType = type;
+				break;
+			}
+			if ( !firstDataType ) {
+				firstDataType = type;
+			}
+		}
+
+		// Or just use first one
+		finalDataType = finalDataType || firstDataType;
+	}
+
+	// If we found a dataType
+	// We add the dataType to the list if needed
+	// and return the corresponding response
+	if ( finalDataType ) {
+		if ( finalDataType !== dataTypes[ 0 ] ) {
+			dataTypes.unshift( finalDataType );
+		}
+		return responses[ finalDataType ];
+	}
+}
+
+/* Chain conversions given the request and the original response
+ * Also sets the responseXXX fields on the jqXHR instance
+ */
+function ajaxConvert( s, response, jqXHR, isSuccess ) {
+	var conv2, current, conv, tmp, prev,
+		converters = {},
+
+		// Work with a copy of dataTypes in case we need to modify it for conversion
+		dataTypes = s.dataTypes.slice();
+
+	// Create converters map with lowercased keys
+	if ( dataTypes[ 1 ] ) {
+		for ( conv in s.converters ) {
+			converters[ conv.toLowerCase() ] = s.converters[ conv ];
+		}
+	}
+
+	current = dataTypes.shift();
+
+	// Convert to each sequential dataType
+	while ( current ) {
+
+		if ( s.responseFields[ current ] ) {
+			jqXHR[ s.responseFields[ current ] ] = response;
+		}
+
+		// Apply the dataFilter if provided
+		if ( !prev && isSuccess && s.dataFilter ) {
+			response = s.dataFilter( response, s.dataType );
+		}
+
+		prev = current;
+		current = dataTypes.shift();
+
+		if ( current ) {
+
+			// There's only work to do if current dataType is non-auto
+			if ( current === "*" ) {
+
+				current = prev;
+
+			// Convert response if prev dataType is non-auto and differs from current
+			} else if ( prev !== "*" && prev !== current ) {
+
+				// Seek a direct converter
+				conv = converters[ prev + " " + current ] || converters[ "* " + current ];
+
+				// If none found, seek a pair
+				if ( !conv ) {
+					for ( conv2 in converters ) {
+
+						// If conv2 outputs current
+						tmp = conv2.split( " " );
+						if ( tmp[ 1 ] === current ) {
+
+							// If prev can be converted to accepted input
+							conv = converters[ prev + " " + tmp[ 0 ] ] ||
+								converters[ "* " + tmp[ 0 ] ];
+							if ( conv ) {
+
+								// Condense equivalence converters
+								if ( conv === true ) {
+									conv = converters[ conv2 ];
+
+								// Otherwise, insert the intermediate dataType
+								} else if ( converters[ conv2 ] !== true ) {
+									current = tmp[ 0 ];
+									dataTypes.unshift( tmp[ 1 ] );
+								}
+								break;
+							}
+						}
+					}
+				}
+
+				// Apply converter (if not an equivalence)
+				if ( conv !== true ) {
+
+					// Unless errors are allowed to bubble, catch and return them
+					if ( conv && s.throws ) {
+						response = conv( response );
+					} else {
+						try {
+							response = conv( response );
+						} catch ( e ) {
+							return {
+								state: "parsererror",
+								error: conv ? e : "No conversion from " + prev + " to " + current
+							};
+						}
+					}
+				}
+			}
+		}
+	}
+
+	return { state: "success", data: response };
+}
+
+jQuery.extend( {
+
+	// Counter for holding the number of active queries
+	active: 0,
+
+	// Last-Modified header cache for next request
+	lastModified: {},
+	etag: {},
+
+	ajaxSettings: {
+		url: location.href,
+		type: "GET",
+		isLocal: rlocalProtocol.test( location.protocol ),
+		global: true,
+		processData: true,
+		async: true,
+		contentType: "application/x-www-form-urlencoded; charset=UTF-8",
+
+		/*
+		timeout: 0,
+		data: null,
+		dataType: null,
+		username: null,
+		password: null,
+		cache: null,
+		throws: false,
+		traditional: false,
+		headers: {},
+		*/
+
+		accepts: {
+			"*": allTypes,
+			text: "text/plain",
+			html: "text/html",
+			xml: "application/xml, text/xml",
+			json: "application/json, text/javascript"
+		},
+
+		contents: {
+			xml: /\bxml\b/,
+			html: /\bhtml/,
+			json: /\bjson\b/
+		},
+
+		responseFields: {
+			xml: "responseXML",
+			text: "responseText",
+			json: "responseJSON"
+		},
+
+		// Data converters
+		// Keys separate source (or catchall "*") and destination types with a single space
+		converters: {
+
+			// Convert anything to text
+			"* text": String,
+
+			// Text to html (true = no transformation)
+			"text html": true,
+
+			// Evaluate text as a json expression
+			"text json": JSON.parse,
+
+			// Parse text as xml
+			"text xml": jQuery.parseXML
+		},
+
+		// For options that shouldn't be deep extended:
+		// you can add your own custom options here if
+		// and when you create one that shouldn't be
+		// deep extended (see ajaxExtend)
+		flatOptions: {
+			url: true,
+			context: true
+		}
+	},
+
+	// Creates a full fledged settings object into target
+	// with both ajaxSettings and settings fields.
+	// If target is omitted, writes into ajaxSettings.
+	ajaxSetup: function( target, settings ) {
+		return settings ?
+
+			// Building a settings object
+			ajaxExtend( ajaxExtend( target, jQuery.ajaxSettings ), settings ) :
+
+			// Extending ajaxSettings
+			ajaxExtend( jQuery.ajaxSettings, target );
+	},
+
+	ajaxPrefilter: addToPrefiltersOrTransports( prefilters ),
+	ajaxTransport: addToPrefiltersOrTransports( transports ),
+
+	// Main method
+	ajax: function( url, options ) {
+
+		// If url is an object, simulate pre-1.5 signature
+		if ( typeof url === "object" ) {
+			options = url;
+			url = undefined;
+		}
+
+		// Force options to be an object
+		options = options || {};
+
+		var transport,
+
+			// URL without anti-cache param
+			cacheURL,
+
+			// Response headers
+			responseHeadersString,
+			responseHeaders,
+
+			// timeout handle
+			timeoutTimer,
+
+			// Url cleanup var
+			urlAnchor,
+
+			// Request state (becomes false upon send and true upon completion)
+			completed,
+
+			// To know if global events are to be dispatched
+			fireGlobals,
+
+			// Loop variable
+			i,
+
+			// uncached part of the url
+			uncached,
+
+			// Create the final options object
+			s = jQuery.ajaxSetup( {}, options ),
+
+			// Callbacks context
+			callbackContext = s.context || s,
+
+			// Context for global events is callbackContext if it is a DOM node or jQuery collection
+			globalEventContext = s.context &&
+				( callbackContext.nodeType || callbackContext.jquery ) ?
+				jQuery( callbackContext ) :
+				jQuery.event,
+
+			// Deferreds
+			deferred = jQuery.Deferred(),
+			completeDeferred = jQuery.Callbacks( "once memory" ),
+
+			// Status-dependent callbacks
+			statusCode = s.statusCode || {},
+
+			// Headers (they are sent all at once)
+			requestHeaders = {},
+			requestHeadersNames = {},
+
+			// Default abort message
+			strAbort = "canceled",
+
+			// Fake xhr
+			jqXHR = {
+				readyState: 0,
+
+				// Builds headers hashtable if needed
+				getResponseHeader: function( key ) {
+					var match;
+					if ( completed ) {
+						if ( !responseHeaders ) {
+							responseHeaders = {};
+							while ( ( match = rheaders.exec( responseHeadersString ) ) ) {
+								responseHeaders[ match[ 1 ].toLowerCase() + " " ] =
+									( responseHeaders[ match[ 1 ].toLowerCase() + " " ] || [] )
+										.concat( match[ 2 ] );
+							}
+						}
+						match = responseHeaders[ key.toLowerCase() + " " ];
+					}
+					return match == null ? null : match.join( ", " );
+				},
+
+				// Raw string
+				getAllResponseHeaders: function() {
+					return completed ? responseHeadersString : null;
+				},
+
+				// Caches the header
+				setRequestHeader: function( name, value ) {
+					if ( completed == null ) {
+						name = requestHeadersNames[ name.toLowerCase() ] =
+							requestHeadersNames[ name.toLowerCase() ] || name;
+						requestHeaders[ name ] = value;
+					}
+					return this;
+				},
+
+				// Overrides response content-type header
+				overrideMimeType: function( type ) {
+					if ( completed == null ) {
+						s.mimeType = type;
+					}
+					return this;
+				},
+
+				// Status-dependent callbacks
+				statusCode: function( map ) {
+					var code;
+					if ( map ) {
+						if ( completed ) {
+
+							// Execute the appropriate callbacks
+							jqXHR.always( map[ jqXHR.status ] );
+						} else {
+
+							// Lazy-add the new callbacks in a way that preserves old ones
+							for ( code in map ) {
+								statusCode[ code ] = [ statusCode[ code ], map[ code ] ];
+							}
+						}
+					}
+					return this;
+				},
+
+				// Cancel the request
+				abort: function( statusText ) {
+					var finalText = statusText || strAbort;
+					if ( transport ) {
+						transport.abort( finalText );
+					}
+					done( 0, finalText );
+					return this;
+				}
+			};
+
+		// Attach deferreds
+		deferred.promise( jqXHR );
+
+		// Add protocol if not provided (prefilters might expect it)
+		// Handle falsy url in the settings object (#10093: consistency with old signature)
+		// We also use the url parameter if available
+		s.url = ( ( url || s.url || location.href ) + "" )
+			.replace( rprotocol, location.protocol + "//" );
+
+		// Alias method option to type as per ticket #12004
+		s.type = options.method || options.type || s.method || s.type;
+
+		// Extract dataTypes list
+		s.dataTypes = ( s.dataType || "*" ).toLowerCase().match( rnothtmlwhite ) || [ "" ];
+
+		// A cross-domain request is in order when the origin doesn't match the current origin.
+		if ( s.crossDomain == null ) {
+			urlAnchor = document.createElement( "a" );
+
+			// Support: IE <=8 - 11, Edge 12 - 15
+			// IE throws exception on accessing the href property if url is malformed,
+			// e.g. http://example.com:80x/
+			try {
+				urlAnchor.href = s.url;
+
+				// Support: IE <=8 - 11 only
+				// Anchor's host property isn't correctly set when s.url is relative
+				urlAnchor.href = urlAnchor.href;
+				s.crossDomain = originAnchor.protocol + "//" + originAnchor.host !==
+					urlAnchor.protocol + "//" + urlAnchor.host;
+			} catch ( e ) {
+
+				// If there is an error parsing the URL, assume it is crossDomain,
+				// it can be rejected by the transport if it is invalid
+				s.crossDomain = true;
+			}
+		}
+
+		// Convert data if not already a string
+		if ( s.data && s.processData && typeof s.data !== "string" ) {
+			s.data = jQuery.param( s.data, s.traditional );
+		}
+
+		// Apply prefilters
+		inspectPrefiltersOrTransports( prefilters, s, options, jqXHR );
+
+		// If request was aborted inside a prefilter, stop there
+		if ( completed ) {
+			return jqXHR;
+		}
+
+		// We can fire global events as of now if asked to
+		// Don't fire events if jQuery.event is undefined in an AMD-usage scenario (#15118)
+		fireGlobals = jQuery.event && s.global;
+
+		// Watch for a new set of requests
+		if ( fireGlobals && jQuery.active++ === 0 ) {
+			jQuery.event.trigger( "ajaxStart" );
+		}
+
+		// Uppercase the type
+		s.type = s.type.toUpperCase();
+
+		// Determine if request has content
+		s.hasContent = !rnoContent.test( s.type );
+
+		// Save the URL in case we're toying with the If-Modified-Since
+		// and/or If-None-Match header later on
+		// Remove hash to simplify url manipulation
+		cacheURL = s.url.replace( rhash, "" );
+
+		// More options handling for requests with no content
+		if ( !s.hasContent ) {
+
+			// Remember the hash so we can put it back
+			uncached = s.url.slice( cacheURL.length );
+
+			// If data is available and should be processed, append data to url
+			if ( s.data && ( s.processData || typeof s.data === "string" ) ) {
+				cacheURL += ( rquery.test( cacheURL ) ? "&" : "?" ) + s.data;
+
+				// #9682: remove data so that it's not used in an eventual retry
+				delete s.data;
+			}
+
+			// Add or update anti-cache param if needed
+			if ( s.cache === false ) {
+				cacheURL = cacheURL.replace( rantiCache, "$1" );
+				uncached = ( rquery.test( cacheURL ) ? "&" : "?" ) + "_=" + ( nonce.guid++ ) +
+					uncached;
+			}
+
+			// Put hash and anti-cache on the URL that will be requested (gh-1732)
+			s.url = cacheURL + uncached;
+
+		// Change '%20' to '+' if this is encoded form body content (gh-2658)
+		} else if ( s.data && s.processData &&
+			( s.contentType || "" ).indexOf( "application/x-www-form-urlencoded" ) === 0 ) {
+			s.data = s.data.replace( r20, "+" );
+		}
+
+		// Set the If-Modified-Since and/or If-None-Match header, if in ifModified mode.
+		if ( s.ifModified ) {
+			if ( jQuery.lastModified[ cacheURL ] ) {
+				jqXHR.setRequestHeader( "If-Modified-Since", jQuery.lastModified[ cacheURL ] );
+			}
+			if ( jQuery.etag[ cacheURL ] ) {
+				jqXHR.setRequestHeader( "If-None-Match", jQuery.etag[ cacheURL ] );
+			}
+		}
+
+		// Set the correct header, if data is being sent
+		if ( s.data && s.hasContent && s.contentType !== false || options.contentType ) {
+			jqXHR.setRequestHeader( "Content-Type", s.contentType );
+		}
+
+		// Set the Accepts header for the server, depending on the dataType
+		jqXHR.setRequestHeader(
+			"Accept",
+			s.dataTypes[ 0 ] && s.accepts[ s.dataTypes[ 0 ] ] ?
+				s.accepts[ s.dataTypes[ 0 ] ] +
+					( s.dataTypes[ 0 ] !== "*" ? ", " + allTypes + "; q=0.01" : "" ) :
+				s.accepts[ "*" ]
+		);
+
+		// Check for headers option
+		for ( i in s.headers ) {
+			jqXHR.setRequestHeader( i, s.headers[ i ] );
+		}
+
+		// Allow custom headers/mimetypes and early abort
+		if ( s.beforeSend &&
+			( s.beforeSend.call( callbackContext, jqXHR, s ) === false || completed ) ) {
+
+			// Abort if not done already and return
+			return jqXHR.abort();
+		}
+
+		// Aborting is no longer a cancellation
+		strAbort = "abort";
+
+		// Install callbacks on deferreds
+		completeDeferred.add( s.complete );
+		jqXHR.done( s.success );
+		jqXHR.fail( s.error );
+
+		// Get transport
+		transport = inspectPrefiltersOrTransports( transports, s, options, jqXHR );
+
+		// If no transport, we auto-abort
+		if ( !transport ) {
+			done( -1, "No Transport" );
+		} else {
+			jqXHR.readyState = 1;
+
+			// Send global event
+			if ( fireGlobals ) {
+				globalEventContext.trigger( "ajaxSend", [ jqXHR, s ] );
+			}
+
+			// If request was aborted inside ajaxSend, stop there
+			if ( completed ) {
+				return jqXHR;
+			}
+
+			// Timeout
+			if ( s.async && s.timeout > 0 ) {
+				timeoutTimer = window.setTimeout( function() {
+					jqXHR.abort( "timeout" );
+				}, s.timeout );
+			}
+
+			try {
+				completed = false;
+				transport.send( requestHeaders, done );
+			} catch ( e ) {
+
+				// Rethrow post-completion exceptions
+				if ( completed ) {
+					throw e;
+				}
+
+				// Propagate others as results
+				done( -1, e );
+			}
+		}
+
+		// Callback for when everything is done
+		function done( status, nativeStatusText, responses, headers ) {
+			var isSuccess, success, error, response, modified,
+				statusText = nativeStatusText;
+
+			// Ignore repeat invocations
+			if ( completed ) {
+				return;
+			}
+
+			completed = true;
+
+			// Clear timeout if it exists
+			if ( timeoutTimer ) {
+				window.clearTimeout( timeoutTimer );
+			}
+
+			// Dereference transport for early garbage collection
+			// (no matter how long the jqXHR object will be used)
+			transport = undefined;
+
+			// Cache response headers
+			responseHeadersString = headers || "";
+
+			// Set readyState
+			jqXHR.readyState = status > 0 ? 4 : 0;
+
+			// Determine if successful
+			isSuccess = status >= 200 && status < 300 || status === 304;
+
+			// Get response data
+			if ( responses ) {
+				response = ajaxHandleResponses( s, jqXHR, responses );
+			}
+
+			// Use a noop converter for missing script but not if jsonp
+			if ( !isSuccess &&
+				jQuery.inArray( "script", s.dataTypes ) > -1 &&
+				jQuery.inArray( "json", s.dataTypes ) < 0 ) {
+				s.converters[ "text script" ] = function() {};
+			}
+
+			// Convert no matter what (that way responseXXX fields are always set)
+			response = ajaxConvert( s, response, jqXHR, isSuccess );
+
+			// If successful, handle type chaining
+			if ( isSuccess ) {
+
+				// Set the If-Modified-Since and/or If-None-Match header, if in ifModified mode.
+				if ( s.ifModified ) {
+					modified = jqXHR.getResponseHeader( "Last-Modified" );
+					if ( modified ) {
+						jQuery.lastModified[ cacheURL ] = modified;
+					}
+					modified = jqXHR.getResponseHeader( "etag" );
+					if ( modified ) {
+						jQuery.etag[ cacheURL ] = modified;
+					}
+				}
+
+				// if no content
+				if ( status === 204 || s.type === "HEAD" ) {
+					statusText = "nocontent";
+
+				// if not modified
+				} else if ( status === 304 ) {
+					statusText = "notmodified";
+
+				// If we have data, let's convert it
+				} else {
+					statusText = response.state;
+					success = response.data;
+					error = response.error;
+					isSuccess = !error;
+				}
+			} else {
+
+				// Extract error from statusText and normalize for non-aborts
+				error = statusText;
+				if ( status || !statusText ) {
+					statusText = "error";
+					if ( status < 0 ) {
+						status = 0;
+					}
+				}
+			}
+
+			// Set data for the fake xhr object
+			jqXHR.status = status;
+			jqXHR.statusText = ( nativeStatusText || statusText ) + "";
+
+			// Success/Error
+			if ( isSuccess ) {
+				deferred.resolveWith( callbackContext, [ success, statusText, jqXHR ] );
+			} else {
+				deferred.rejectWith( callbackContext, [ jqXHR, statusText, error ] );
+			}
+
+			// Status-dependent callbacks
+			jqXHR.statusCode( statusCode );
+			statusCode = undefined;
+
+			if ( fireGlobals ) {
+				globalEventContext.trigger( isSuccess ? "ajaxSuccess" : "ajaxError",
+					[ jqXHR, s, isSuccess ? success : error ] );
+			}
+
+			// Complete
+			completeDeferred.fireWith( callbackContext, [ jqXHR, statusText ] );
+
+			if ( fireGlobals ) {
+				globalEventContext.trigger( "ajaxComplete", [ jqXHR, s ] );
+
+				// Handle the global AJAX counter
+				if ( !( --jQuery.active ) ) {
+					jQuery.event.trigger( "ajaxStop" );
+				}
+			}
+		}
+
+		return jqXHR;
+	},
+
+	getJSON: function( url, data, callback ) {
+		return jQuery.get( url, data, callback, "json" );
+	},
+
+	getScript: function( url, callback ) {
+		return jQuery.get( url, undefined, callback, "script" );
+	}
+} );
+
+jQuery.each( [ "get", "post" ], function( _i, method ) {
+	jQuery[ method ] = function( url, data, callback, type ) {
+
+		// Shift arguments if data argument was omitted
+		if ( isFunction( data ) ) {
+			type = type || callback;
+			callback = data;
+			data = undefined;
+		}
+
+		// The url can be an options object (which then must have .url)
+		return jQuery.ajax( jQuery.extend( {
+			url: url,
+			type: method,
+			dataType: type,
+			data: data,
+			success: callback
+		}, jQuery.isPlainObject( url ) && url ) );
+	};
+} );
+
+jQuery.ajaxPrefilter( function( s ) {
+	var i;
+	for ( i in s.headers ) {
+		if ( i.toLowerCase() === "content-type" ) {
+			s.contentType = s.headers[ i ] || "";
+		}
+	}
+} );
+
+
+jQuery._evalUrl = function( url, options, doc ) {
+	return jQuery.ajax( {
+		url: url,
+
+		// Make this explicit, since user can override this through ajaxSetup (#11264)
+		type: "GET",
+		dataType: "script",
+		cache: true,
+		async: false,
+		global: false,
+
+		// Only evaluate the response if it is successful (gh-4126)
+		// dataFilter is not invoked for failure responses, so using it instead
+		// of the default converter is kludgy but it works.
+		converters: {
+			"text script": function() {}
+		},
+		dataFilter: function( response ) {
+			jQuery.globalEval( response, options, doc );
+		}
+	} );
+};
+
+
+jQuery.fn.extend( {
+	wrapAll: function( html ) {
+		var wrap;
+
+		if ( this[ 0 ] ) {
+			if ( isFunction( html ) ) {
+				html = html.call( this[ 0 ] );
+			}
+
+			// The elements to wrap the target around
+			wrap = jQuery( html, this[ 0 ].ownerDocument ).eq( 0 ).clone( true );
+
+			if ( this[ 0 ].parentNode ) {
+				wrap.insertBefore( this[ 0 ] );
+			}
+
+			wrap.map( function() {
+				var elem = this;
+
+				while ( elem.firstElementChild ) {
+					elem = elem.firstElementChild;
+				}
+
+				return elem;
+			} ).append( this );
+		}
+
+		return this;
+	},
+
+	wrapInner: function( html ) {
+		if ( isFunction( html ) ) {
+			return this.each( function( i ) {
+				jQuery( this ).wrapInner( html.call( this, i ) );
+			} );
+		}
+
+		return this.each( function() {
+			var self = jQuery( this ),
+				contents = self.contents();
+
+			if ( contents.length ) {
+				contents.wrapAll( html );
+
+			} else {
+				self.append( html );
+			}
+		} );
+	},
+
+	wrap: function( html ) {
+		var htmlIsFunction = isFunction( html );
+
+		return this.each( function( i ) {
+			jQuery( this ).wrapAll( htmlIsFunction ? html.call( this, i ) : html );
+		} );
+	},
+
+	unwrap: function( selector ) {
+		this.parent( selector ).not( "body" ).each( function() {
+			jQuery( this ).replaceWith( this.childNodes );
+		} );
+		return this;
+	}
+} );
+
+
+jQuery.expr.pseudos.hidden = function( elem ) {
+	return !jQuery.expr.pseudos.visible( elem );
+};
+jQuery.expr.pseudos.visible = function( elem ) {
+	return !!( elem.offsetWidth || elem.offsetHeight || elem.getClientRects().length );
+};
+
+
+
+
+jQuery.ajaxSettings.xhr = function() {
+	try {
+		return new window.XMLHttpRequest();
+	} catch ( e ) {}
+};
+
+var xhrSuccessStatus = {
+
+		// File protocol always yields status code 0, assume 200
+		0: 200,
+
+		// Support: IE <=9 only
+		// #1450: sometimes IE returns 1223 when it should be 204
+		1223: 204
+	},
+	xhrSupported = jQuery.ajaxSettings.xhr();
+
+support.cors = !!xhrSupported && ( "withCredentials" in xhrSupported );
+support.ajax = xhrSupported = !!xhrSupported;
+
+jQuery.ajaxTransport( function( options ) {
+	var callback, errorCallback;
+
+	// Cross domain only allowed if supported through XMLHttpRequest
+	if ( support.cors || xhrSupported && !options.crossDomain ) {
+		return {
+			send: function( headers, complete ) {
+				var i,
+					xhr = options.xhr();
+
+				xhr.open(
+					options.type,
+					options.url,
+					options.async,
+					options.username,
+					options.password
+				);
+
+				// Apply custom fields if provided
+				if ( options.xhrFields ) {
+					for ( i in options.xhrFields ) {
+						xhr[ i ] = options.xhrFields[ i ];
+					}
+				}
+
+				// Override mime type if needed
+				if ( options.mimeType && xhr.overrideMimeType ) {
+					xhr.overrideMimeType( options.mimeType );
+				}
+
+				// X-Requested-With header
+				// For cross-domain requests, seeing as conditions for a preflight are
+				// akin to a jigsaw puzzle, we simply never set it to be sure.
+				// (it can always be set on a per-request basis or even using ajaxSetup)
+				// For same-domain requests, won't change header if already provided.
+				if ( !options.crossDomain && !headers[ "X-Requested-With" ] ) {
+					headers[ "X-Requested-With" ] = "XMLHttpRequest";
+				}
+
+				// Set headers
+				for ( i in headers ) {
+					xhr.setRequestHeader( i, headers[ i ] );
+				}
+
+				// Callback
+				callback = function( type ) {
+					return function() {
+						if ( callback ) {
+							callback = errorCallback = xhr.onload =
+								xhr.onerror = xhr.onabort = xhr.ontimeout =
+									xhr.onreadystatechange = null;
+
+							if ( type === "abort" ) {
+								xhr.abort();
+							} else if ( type === "error" ) {
+
+								// Support: IE <=9 only
+								// On a manual native abort, IE9 throws
+								// errors on any property access that is not readyState
+								if ( typeof xhr.status !== "number" ) {
+									complete( 0, "error" );
+								} else {
+									complete(
+
+										// File: protocol always yields status 0; see #8605, #14207
+										xhr.status,
+										xhr.statusText
+									);
+								}
+							} else {
+								complete(
+									xhrSuccessStatus[ xhr.status ] || xhr.status,
+									xhr.statusText,
+
+									// Support: IE <=9 only
+									// IE9 has no XHR2 but throws on binary (trac-11426)
+									// For XHR2 non-text, let the caller handle it (gh-2498)
+									( xhr.responseType || "text" ) !== "text"  ||
+									typeof xhr.responseText !== "string" ?
+										{ binary: xhr.response } :
+										{ text: xhr.responseText },
+									xhr.getAllResponseHeaders()
+								);
+							}
+						}
+					};
+				};
+
+				// Listen to events
+				xhr.onload = callback();
+				errorCallback = xhr.onerror = xhr.ontimeout = callback( "error" );
+
+				// Support: IE 9 only
+				// Use onreadystatechange to replace onabort
+				// to handle uncaught aborts
+				if ( xhr.onabort !== undefined ) {
+					xhr.onabort = errorCallback;
+				} else {
+					xhr.onreadystatechange = function() {
+
+						// Check readyState before timeout as it changes
+						if ( xhr.readyState === 4 ) {
+
+							// Allow onerror to be called first,
+							// but that will not handle a native abort
+							// Also, save errorCallback to a variable
+							// as xhr.onerror cannot be accessed
+							window.setTimeout( function() {
+								if ( callback ) {
+									errorCallback();
+								}
+							} );
+						}
+					};
+				}
+
+				// Create the abort callback
+				callback = callback( "abort" );
+
+				try {
+
+					// Do send the request (this may raise an exception)
+					xhr.send( options.hasContent && options.data || null );
+				} catch ( e ) {
+
+					// #14683: Only rethrow if this hasn't been notified as an error yet
+					if ( callback ) {
+						throw e;
+					}
+				}
+			},
+
+			abort: function() {
+				if ( callback ) {
+					callback();
+				}
+			}
+		};
+	}
+} );
+
+
+
+
+// Prevent auto-execution of scripts when no explicit dataType was provided (See gh-2432)
+jQuery.ajaxPrefilter( function( s ) {
+	if ( s.crossDomain ) {
+		s.contents.script = false;
+	}
+} );
+
+// Install script dataType
+jQuery.ajaxSetup( {
+	accepts: {
+		script: "text/javascript, application/javascript, " +
+			"application/ecmascript, application/x-ecmascript"
+	},
+	contents: {
+		script: /\b(?:java|ecma)script\b/
+	},
+	converters: {
+		"text script": function( text ) {
+			jQuery.globalEval( text );
+			return text;
+		}
+	}
+} );
+
+// Handle cache's special case and crossDomain
+jQuery.ajaxPrefilter( "script", function( s ) {
+	if ( s.cache === undefined ) {
+		s.cache = false;
+	}
+	if ( s.crossDomain ) {
+		s.type = "GET";
+	}
+} );
+
+// Bind script tag hack transport
+jQuery.ajaxTransport( "script", function( s ) {
+
+	// This transport only deals with cross domain or forced-by-attrs requests
+	if ( s.crossDomain || s.scriptAttrs ) {
+		var script, callback;
+		return {
+			send: function( _, complete ) {
+				script = jQuery( "<script>" )
+					.attr( s.scriptAttrs || {} )
+					.prop( { charset: s.scriptCharset, src: s.url } )
+					.on( "load error", callback = function( evt ) {
+						script.remove();
+						callback = null;
+						if ( evt ) {
+							complete( evt.type === "error" ? 404 : 200, evt.type );
+						}
+					} );
+
+				// Use native DOM manipulation to avoid our domManip AJAX trickery
+				document.head.appendChild( script[ 0 ] );
+			},
+			abort: function() {
+				if ( callback ) {
+					callback();
+				}
+			}
+		};
+	}
+} );
+
+
+
+
+var oldCallbacks = [],
+	rjsonp = /(=)\?(?=&|$)|\?\?/;
+
+// Default jsonp settings
+jQuery.ajaxSetup( {
+	jsonp: "callback",
+	jsonpCallback: function() {
+		var callback = oldCallbacks.pop() || ( jQuery.expando + "_" + ( nonce.guid++ ) );
+		this[ callback ] = true;
+		return callback;
+	}
+} );
+
+// Detect, normalize options and install callbacks for jsonp requests
+jQuery.ajaxPrefilter( "json jsonp", function( s, originalSettings, jqXHR ) {
+
+	var callbackName, overwritten, responseContainer,
+		jsonProp = s.jsonp !== false && ( rjsonp.test( s.url ) ?
+			"url" :
+			typeof s.data === "string" &&
+				( s.contentType || "" )
+					.indexOf( "application/x-www-form-urlencoded" ) === 0 &&
+				rjsonp.test( s.data ) && "data"
+		);
+
+	// Handle iff the expected data type is "jsonp" or we have a parameter to set
+	if ( jsonProp || s.dataTypes[ 0 ] === "jsonp" ) {
+
+		// Get callback name, remembering preexisting value associated with it
+		callbackName = s.jsonpCallback = isFunction( s.jsonpCallback ) ?
+			s.jsonpCallback() :
+			s.jsonpCallback;
+
+		// Insert callback into url or form data
+		if ( jsonProp ) {
+			s[ jsonProp ] = s[ jsonProp ].replace( rjsonp, "$1" + callbackName );
+		} else if ( s.jsonp !== false ) {
+			s.url += ( rquery.test( s.url ) ? "&" : "?" ) + s.jsonp + "=" + callbackName;
+		}
+
+		// Use data converter to retrieve json after script execution
+		s.converters[ "script json" ] = function() {
+			if ( !responseContainer ) {
+				jQuery.error( callbackName + " was not called" );
+			}
+			return responseContainer[ 0 ];
+		};
+
+		// Force json dataType
+		s.dataTypes[ 0 ] = "json";
+
+		// Install callback
+		overwritten = window[ callbackName ];
+		window[ callbackName ] = function() {
+			responseContainer = arguments;
+		};
+
+		// Clean-up function (fires after converters)
+		jqXHR.always( function() {
+
+			// If previous value didn't exist - remove it
+			if ( overwritten === undefined ) {
+				jQuery( window ).removeProp( callbackName );
+
+			// Otherwise restore preexisting value
+			} else {
+				window[ callbackName ] = overwritten;
+			}
+
+			// Save back as free
+			if ( s[ callbackName ] ) {
+
+				// Make sure that re-using the options doesn't screw things around
+				s.jsonpCallback = originalSettings.jsonpCallback;
+
+				// Save the callback name for future use
+				oldCallbacks.push( callbackName );
+			}
+
+			// Call if it was a function and we have a response
+			if ( responseContainer && isFunction( overwritten ) ) {
+				overwritten( responseContainer[ 0 ] );
+			}
+
+			responseContainer = overwritten = undefined;
+		} );
+
+		// Delegate to script
+		return "script";
+	}
+} );
+
+
+
+
+// Support: Safari 8 only
+// In Safari 8 documents created via document.implementation.createHTMLDocument
+// collapse sibling forms: the second one becomes a child of the first one.
+// Because of that, this security measure has to be disabled in Safari 8.
+// https://bugs.webkit.org/show_bug.cgi?id=137337
+support.createHTMLDocument = ( function() {
+	var body = document.implementation.createHTMLDocument( "" ).body;
+	body.innerHTML = "<form></form><form></form>";
+	return body.childNodes.length === 2;
+} )();
+
+
+// Argument "data" should be string of html
+// context (optional): If specified, the fragment will be created in this context,
+// defaults to document
+// keepScripts (optional): If true, will include scripts passed in the html string
+jQuery.parseHTML = function( data, context, keepScripts ) {
+	if ( typeof data !== "string" ) {
+		return [];
+	}
+	if ( typeof context === "boolean" ) {
+		keepScripts = context;
+		context = false;
+	}
+
+	var base, parsed, scripts;
+
+	if ( !context ) {
+
+		// Stop scripts or inline event handlers from being executed immediately
+		// by using document.implementation
+		if ( support.createHTMLDocument ) {
+			context = document.implementation.createHTMLDocument( "" );
+
+			// Set the base href for the created document
+			// so any parsed elements with URLs
+			// are based on the document's URL (gh-2965)
+			base = context.createElement( "base" );
+			base.href = document.location.href;
+			context.head.appendChild( base );
+		} else {
+			context = document;
+		}
+	}
+
+	parsed = rsingleTag.exec( data );
+	scripts = !keepScripts && [];
+
+	// Single tag
+	if ( parsed ) {
+		return [ context.createElement( parsed[ 1 ] ) ];
+	}
+
+	parsed = buildFragment( [ data ], context, scripts );
+
+	if ( scripts && scripts.length ) {
+		jQuery( scripts ).remove();
+	}
+
+	return jQuery.merge( [], parsed.childNodes );
+};
+
+
+/**
+ * Load a url into a page
+ */
+jQuery.fn.load = function( url, params, callback ) {
+	var selector, type, response,
+		self = this,
+		off = url.indexOf( " " );
+
+	if ( off > -1 ) {
+		selector = stripAndCollapse( url.slice( off ) );
+		url = url.slice( 0, off );
+	}
+
+	// If it's a function
+	if ( isFunction( params ) ) {
+
+		// We assume that it's the callback
+		callback = params;
+		params = undefined;
+
+	// Otherwise, build a param string
+	} else if ( params && typeof params === "object" ) {
+		type = "POST";
+	}
+
+	// If we have elements to modify, make the request
+	if ( self.length > 0 ) {
+		jQuery.ajax( {
+			url: url,
+
+			// If "type" variable is undefined, then "GET" method will be used.
+			// Make value of this field explicit since
+			// user can override it through ajaxSetup method
+			type: type || "GET",
+			dataType: "html",
+			data: params
+		} ).done( function( responseText ) {
+
+			// Save response for use in complete callback
+			response = arguments;
+
+			self.html( selector ?
+
+				// If a selector was specified, locate the right elements in a dummy div
+				// Exclude scripts to avoid IE 'Permission Denied' errors
+				jQuery( "<div>" ).append( jQuery.parseHTML( responseText ) ).find( selector ) :
+
+				// Otherwise use the full result
+				responseText );
+
+		// If the request succeeds, this function gets "data", "status", "jqXHR"
+		// but they are ignored because response was set above.
+		// If it fails, this function gets "jqXHR", "status", "error"
+		} ).always( callback && function( jqXHR, status ) {
+			self.each( function() {
+				callback.apply( this, response || [ jqXHR.responseText, status, jqXHR ] );
+			} );
+		} );
+	}
+
+	return this;
+};
+
+
+
+
+jQuery.expr.pseudos.animated = function( elem ) {
+	return jQuery.grep( jQuery.timers, function( fn ) {
+		return elem === fn.elem;
+	} ).length;
+};
+
+
+
+
+jQuery.offset = {
+	setOffset: function( elem, options, i ) {
+		var curPosition, curLeft, curCSSTop, curTop, curOffset, curCSSLeft, calculatePosition,
+			position = jQuery.css( elem, "position" ),
+			curElem = jQuery( elem ),
+			props = {};
+
+		// Set position first, in-case top/left are set even on static elem
+		if ( position === "static" ) {
+			elem.style.position = "relative";
+		}
+
+		curOffset = curElem.offset();
+		curCSSTop = jQuery.css( elem, "top" );
+		curCSSLeft = jQuery.css( elem, "left" );
+		calculatePosition = ( position === "absolute" || position === "fixed" ) &&
+			( curCSSTop + curCSSLeft ).indexOf( "auto" ) > -1;
+
+		// Need to be able to calculate position if either
+		// top or left is auto and position is either absolute or fixed
+		if ( calculatePosition ) {
+			curPosition = curElem.position();
+			curTop = curPosition.top;
+			curLeft = curPosition.left;
+
+		} else {
+			curTop = parseFloat( curCSSTop ) || 0;
+			curLeft = parseFloat( curCSSLeft ) || 0;
+		}
+
+		if ( isFunction( options ) ) {
+
+			// Use jQuery.extend here to allow modification of coordinates argument (gh-1848)
+			options = options.call( elem, i, jQuery.extend( {}, curOffset ) );
+		}
+
+		if ( options.top != null ) {
+			props.top = ( options.top - curOffset.top ) + curTop;
+		}
+		if ( options.left != null ) {
+			props.left = ( options.left - curOffset.left ) + curLeft;
+		}
+
+		if ( "using" in options ) {
+			options.using.call( elem, props );
+
+		} else {
+			curElem.css( props );
+		}
+	}
+};
+
+jQuery.fn.extend( {
+
+	// offset() relates an element's border box to the document origin
+	offset: function( options ) {
+
+		// Preserve chaining for setter
+		if ( arguments.length ) {
+			return options === undefined ?
+				this :
+				this.each( function( i ) {
+					jQuery.offset.setOffset( this, options, i );
+				} );
+		}
+
+		var rect, win,
+			elem = this[ 0 ];
+
+		if ( !elem ) {
+			return;
+		}
+
+		// Return zeros for disconnected and hidden (display: none) elements (gh-2310)
+		// Support: IE <=11 only
+		// Running getBoundingClientRect on a
+		// disconnected node in IE throws an error
+		if ( !elem.getClientRects().length ) {
+			return { top: 0, left: 0 };
+		}
+
+		// Get document-relative position by adding viewport scroll to viewport-relative gBCR
+		rect = elem.getBoundingClientRect();
+		win = elem.ownerDocument.defaultView;
+		return {
+			top: rect.top + win.pageYOffset,
+			left: rect.left + win.pageXOffset
+		};
+	},
+
+	// position() relates an element's margin box to its offset parent's padding box
+	// This corresponds to the behavior of CSS absolute positioning
+	position: function() {
+		if ( !this[ 0 ] ) {
+			return;
+		}
+
+		var offsetParent, offset, doc,
+			elem = this[ 0 ],
+			parentOffset = { top: 0, left: 0 };
+
+		// position:fixed elements are offset from the viewport, which itself always has zero offset
+		if ( jQuery.css( elem, "position" ) === "fixed" ) {
+
+			// Assume position:fixed implies availability of getBoundingClientRect
+			offset = elem.getBoundingClientRect();
+
+		} else {
+			offset = this.offset();
+
+			// Account for the *real* offset parent, which can be the document or its root element
+			// when a statically positioned element is identified
+			doc = elem.ownerDocument;
+			offsetParent = elem.offsetParent || doc.documentElement;
+			while ( offsetParent &&
+				( offsetParent === doc.body || offsetParent === doc.documentElement ) &&
+				jQuery.css( offsetParent, "position" ) === "static" ) {
+
+				offsetParent = offsetParent.parentNode;
+			}
+			if ( offsetParent && offsetParent !== elem && offsetParent.nodeType === 1 ) {
+
+				// Incorporate borders into its offset, since they are outside its content origin
+				parentOffset = jQuery( offsetParent ).offset();
+				parentOffset.top += jQuery.css( offsetParent, "borderTopWidth", true );
+				parentOffset.left += jQuery.css( offsetParent, "borderLeftWidth", true );
+			}
+		}
+
+		// Subtract parent offsets and element margins
+		return {
+			top: offset.top - parentOffset.top - jQuery.css( elem, "marginTop", true ),
+			left: offset.left - parentOffset.left - jQuery.css( elem, "marginLeft", true )
+		};
+	},
+
+	// This method will return documentElement in the following cases:
+	// 1) For the element inside the iframe without offsetParent, this method will return
+	//    documentElement of the parent window
+	// 2) For the hidden or detached element
+	// 3) For body or html element, i.e. in case of the html node - it will return itself
+	//
+	// but those exceptions were never presented as a real life use-cases
+	// and might be considered as more preferable results.
+	//
+	// This logic, however, is not guaranteed and can change at any point in the future
+	offsetParent: function() {
+		return this.map( function() {
+			var offsetParent = this.offsetParent;
+
+			while ( offsetParent && jQuery.css( offsetParent, "position" ) === "static" ) {
+				offsetParent = offsetParent.offsetParent;
+			}
+
+			return offsetParent || documentElement;
+		} );
+	}
+} );
+
+// Create scrollLeft and scrollTop methods
+jQuery.each( { scrollLeft: "pageXOffset", scrollTop: "pageYOffset" }, function( method, prop ) {
+	var top = "pageYOffset" === prop;
+
+	jQuery.fn[ method ] = function( val ) {
+		return access( this, function( elem, method, val ) {
+
+			// Coalesce documents and windows
+			var win;
+			if ( isWindow( elem ) ) {
+				win = elem;
+			} else if ( elem.nodeType === 9 ) {
+				win = elem.defaultView;
+			}
+
+			if ( val === undefined ) {
+				return win ? win[ prop ] : elem[ method ];
+			}
+
+			if ( win ) {
+				win.scrollTo(
+					!top ? val : win.pageXOffset,
+					top ? val : win.pageYOffset
+				);
+
+			} else {
+				elem[ method ] = val;
+			}
+		}, method, val, arguments.length );
+	};
+} );
+
+// Support: Safari <=7 - 9.1, Chrome <=37 - 49
+// Add the top/left cssHooks using jQuery.fn.position
+// Webkit bug: https://bugs.webkit.org/show_bug.cgi?id=29084
+// Blink bug: https://bugs.chromium.org/p/chromium/issues/detail?id=589347
+// getComputedStyle returns percent when specified for top/left/bottom/right;
+// rather than make the css module depend on the offset module, just check for it here
+jQuery.each( [ "top", "left" ], function( _i, prop ) {
+	jQuery.cssHooks[ prop ] = addGetHookIf( support.pixelPosition,
+		function( elem, computed ) {
+			if ( computed ) {
+				computed = curCSS( elem, prop );
+
+				// If curCSS returns percentage, fallback to offset
+				return rnumnonpx.test( computed ) ?
+					jQuery( elem ).position()[ prop ] + "px" :
+					computed;
+			}
+		}
+	);
+} );
+
+
+// Create innerHeight, innerWidth, height, width, outerHeight and outerWidth methods
+jQuery.each( { Height: "height", Width: "width" }, function( name, type ) {
+	jQuery.each( {
+		padding: "inner" + name,
+		content: type,
+		"": "outer" + name
+	}, function( defaultExtra, funcName ) {
+
+		// Margin is only for outerHeight, outerWidth
+		jQuery.fn[ funcName ] = function( margin, value ) {
+			var chainable = arguments.length && ( defaultExtra || typeof margin !== "boolean" ),
+				extra = defaultExtra || ( margin === true || value === true ? "margin" : "border" );
+
+			return access( this, function( elem, type, value ) {
+				var doc;
+
+				if ( isWindow( elem ) ) {
+
+					// $( window ).outerWidth/Height return w/h including scrollbars (gh-1729)
+					return funcName.indexOf( "outer" ) === 0 ?
+						elem[ "inner" + name ] :
+						elem.document.documentElement[ "client" + name ];
+				}
+
+				// Get document width or height
+				if ( elem.nodeType === 9 ) {
+					doc = elem.documentElement;
+
+					// Either scroll[Width/Height] or offset[Width/Height] or client[Width/Height],
+					// whichever is greatest
+					return Math.max(
+						elem.body[ "scroll" + name ], doc[ "scroll" + name ],
+						elem.body[ "offset" + name ], doc[ "offset" + name ],
+						doc[ "client" + name ]
+					);
+				}
+
+				return value === undefined ?
+
+					// Get width or height on the element, requesting but not forcing parseFloat
+					jQuery.css( elem, type, extra ) :
+
+					// Set width or height on the element
+					jQuery.style( elem, type, value, extra );
+			}, type, chainable ? margin : undefined, chainable );
+		};
+	} );
+} );
+
+
+jQuery.each( [
+	"ajaxStart",
+	"ajaxStop",
+	"ajaxComplete",
+	"ajaxError",
+	"ajaxSuccess",
+	"ajaxSend"
+], function( _i, type ) {
+	jQuery.fn[ type ] = function( fn ) {
+		return this.on( type, fn );
+	};
+} );
+
+
+
+
+jQuery.fn.extend( {
+
+	bind: function( types, data, fn ) {
+		return this.on( types, null, data, fn );
+	},
+	unbind: function( types, fn ) {
+		return this.off( types, null, fn );
+	},
+
+	delegate: function( selector, types, data, fn ) {
+		return this.on( types, selector, data, fn );
+	},
+	undelegate: function( selector, types, fn ) {
+
+		// ( namespace ) or ( selector, types [, fn] )
+		return arguments.length === 1 ?
+			this.off( selector, "**" ) :
+			this.off( types, selector || "**", fn );
+	},
+
+	hover: function( fnOver, fnOut ) {
+		return this.mouseenter( fnOver ).mouseleave( fnOut || fnOver );
+	}
+} );
+
+jQuery.each(
+	( "blur focus focusin focusout resize scroll click dblclick " +
+	"mousedown mouseup mousemove mouseover mouseout mouseenter mouseleave " +
+	"change select submit keydown keypress keyup contextmenu" ).split( " " ),
+	function( _i, name ) {
+
+		// Handle event binding
+		jQuery.fn[ name ] = function( data, fn ) {
+			return arguments.length > 0 ?
+				this.on( name, null, data, fn ) :
+				this.trigger( name );
+		};
+	}
+);
+
+
+
+
+// Support: Android <=4.0 only
+// Make sure we trim BOM and NBSP
+var rtrim = /^[\s\uFEFF\xA0]+|[\s\uFEFF\xA0]+$/g;
+
+// Bind a function to a context, optionally partially applying any
+// arguments.
+// jQuery.proxy is deprecated to promote standards (specifically Function#bind)
+// However, it is not slated for removal any time soon
+jQuery.proxy = function( fn, context ) {
+	var tmp, args, proxy;
+
+	if ( typeof context === "string" ) {
+		tmp = fn[ context ];
+		context = fn;
+		fn = tmp;
+	}
+
+	// Quick check to determine if target is callable, in the spec
+	// this throws a TypeError, but we will just return undefined.
+	if ( !isFunction( fn ) ) {
+		return undefined;
+	}
+
+	// Simulated bind
+	args = slice.call( arguments, 2 );
+	proxy = function() {
+		return fn.apply( context || this, args.concat( slice.call( arguments ) ) );
+	};
+
+	// Set the guid of unique handler to the same of original handler, so it can be removed
+	proxy.guid = fn.guid = fn.guid || jQuery.guid++;
+
+	return proxy;
+};
+
+jQuery.holdReady = function( hold ) {
+	if ( hold ) {
+		jQuery.readyWait++;
+	} else {
+		jQuery.ready( true );
+	}
+};
+jQuery.isArray = Array.isArray;
+jQuery.parseJSON = JSON.parse;
+jQuery.nodeName = nodeName;
+jQuery.isFunction = isFunction;
+jQuery.isWindow = isWindow;
+jQuery.camelCase = camelCase;
+jQuery.type = toType;
+
+jQuery.now = Date.now;
+
+jQuery.isNumeric = function( obj ) {
+
+	// As of jQuery 3.0, isNumeric is limited to
+	// strings and numbers (primitives or objects)
+	// that can be coerced to finite numbers (gh-2662)
+	var type = jQuery.type( obj );
+	return ( type === "number" || type === "string" ) &&
+
+		// parseFloat NaNs numeric-cast false positives ("")
+		// ...but misinterprets leading-number strings, particularly hex literals ("0x...")
+		// subtraction forces infinities to NaN
+		!isNaN( obj - parseFloat( obj ) );
+};
+
+jQuery.trim = function( text ) {
+	return text == null ?
+		"" :
+		( text + "" ).replace( rtrim, "" );
+};
+
+
+
+// Register as a named AMD module, since jQuery can be concatenated with other
+// files that may use define, but not via a proper concatenation script that
+// understands anonymous AMD modules. A named AMD is safest and most robust
+// way to register. Lowercase jquery is used because AMD module names are
+// derived from file names, and jQuery is normally delivered in a lowercase
+// file name. Do this after creating the global so that if an AMD module wants
+// to call noConflict to hide this version of jQuery, it will work.
+
+// Note that for maximum portability, libraries that are not jQuery should
+// declare themselves as anonymous modules, and avoid setting a global if an
+// AMD loader is present. jQuery is a special case. For more information, see
+// https://github.com/jrburke/requirejs/wiki/Updating-existing-libraries#wiki-anon
+
+if ( typeof define === "function" && define.amd ) {
+	define( "jquery", [], function() {
+		return jQuery;
+	} );
+}
+
+
+
+
+var
+
+	// Map over jQuery in case of overwrite
+	_jQuery = window.jQuery,
+
+	// Map over the $ in case of overwrite
+	_$ = window.$;
+
+jQuery.noConflict = function( deep ) {
+	if ( window.$ === jQuery ) {
+		window.$ = _$;
+	}
+
+	if ( deep && window.jQuery === jQuery ) {
+		window.jQuery = _jQuery;
+	}
+
+	return jQuery;
+};
+
+// Expose jQuery and $ identifiers, even in AMD
+// (#7102#comment:10, https://github.com/jquery/jquery/pull/557)
+// and CommonJS for browser emulators (#13566)
+if ( typeof noGlobal === "undefined" ) {
+	window.jQuery = window.$ = jQuery;
+}
+
+
+
+
+return jQuery;
+} );
diff --git a/0.16/_static/jquery.js b/0.16/_static/jquery.js
new file mode 100644
index 00000000000..c4c6022f298
--- /dev/null
+++ b/0.16/_static/jquery.js
@@ -0,0 +1,2 @@
+/*! jQuery v3.6.0 | (c) OpenJS Foundation and other contributors | jquery.org/license */
+!function(e,t){"use strict";"object"==typeof module&&"object"==typeof module.exports?module.exports=e.document?t(e,!0):function(e){if(!e.document)throw new Error("jQuery requires a window with a document");return t(e)}:t(e)}("undefined"!=typeof window?window:this,function(C,e){"use strict";var t=[],r=Object.getPrototypeOf,s=t.slice,g=t.flat?function(e){return t.flat.call(e)}:function(e){return t.concat.apply([],e)},u=t.push,i=t.indexOf,n={},o=n.toString,v=n.hasOwnProperty,a=v.toString,l=a.call(Object),y={},m=function(e){return"function"==typeof e&&"number"!=typeof e.nodeType&&"function"!=typeof e.item},x=function(e){return null!=e&&e===e.window},E=C.document,c={type:!0,src:!0,nonce:!0,noModule:!0};function b(e,t,n){var r,i,o=(n=n||E).createElement("script");if(o.text=e,t)for(r in c)(i=t[r]||t.getAttribute&&t.getAttribute(r))&&o.setAttribute(r,i);n.head.appendChild(o).parentNode.removeChild(o)}function w(e){return null==e?e+"":"object"==typeof e||"function"==typeof e?n[o.call(e)]||"object":typeof e}var f="3.6.0",S=function(e,t){return new S.fn.init(e,t)};function p(e){var t=!!e&&"length"in e&&e.length,n=w(e);return!m(e)&&!x(e)&&("array"===n||0===t||"number"==typeof t&&0<t&&t-1 in e)}S.fn=S.prototype={jquery:f,constructor:S,length:0,toArray:function(){return s.call(this)},get:function(e){return null==e?s.call(this):e<0?this[e+this.length]:this[e]},pushStack:function(e){var t=S.merge(this.constructor(),e);return t.prevObject=this,t},each:function(e){return S.each(this,e)},map:function(n){return this.pushStack(S.map(this,function(e,t){return n.call(e,t,e)}))},slice:function(){return this.pushStack(s.apply(this,arguments))},first:function(){return this.eq(0)},last:function(){return this.eq(-1)},even:function(){return this.pushStack(S.grep(this,function(e,t){return(t+1)%2}))},odd:function(){return this.pushStack(S.grep(this,function(e,t){return t%2}))},eq:function(e){var t=this.length,n=+e+(e<0?t:0);return this.pushStack(0<=n&&n<t?[this[n]]:[])},end:function(){return this.prevObject||this.constructor()},push:u,sort:t.sort,splice:t.splice},S.extend=S.fn.extend=function(){var e,t,n,r,i,o,a=arguments[0]||{},s=1,u=arguments.length,l=!1;for("boolean"==typeof a&&(l=a,a=arguments[s]||{},s++),"object"==typeof a||m(a)||(a={}),s===u&&(a=this,s--);s<u;s++)if(null!=(e=arguments[s]))for(t in e)r=e[t],"__proto__"!==t&&a!==r&&(l&&r&&(S.isPlainObject(r)||(i=Array.isArray(r)))?(n=a[t],o=i&&!Array.isArray(n)?[]:i||S.isPlainObject(n)?n:{},i=!1,a[t]=S.extend(l,o,r)):void 0!==r&&(a[t]=r));return a},S.extend({expando:"jQuery"+(f+Math.random()).replace(/\D/g,""),isReady:!0,error:function(e){throw new Error(e)},noop:function(){},isPlainObject:function(e){var t,n;return!(!e||"[object Object]"!==o.call(e))&&(!(t=r(e))||"function"==typeof(n=v.call(t,"constructor")&&t.constructor)&&a.call(n)===l)},isEmptyObject:function(e){var t;for(t in e)return!1;return!0},globalEval:function(e,t,n){b(e,{nonce:t&&t.nonce},n)},each:function(e,t){var n,r=0;if(p(e)){for(n=e.length;r<n;r++)if(!1===t.call(e[r],r,e[r]))break}else for(r in e)if(!1===t.call(e[r],r,e[r]))break;return e},makeArray:function(e,t){var n=t||[];return null!=e&&(p(Object(e))?S.merge(n,"string"==typeof e?[e]:e):u.call(n,e)),n},inArray:function(e,t,n){return null==t?-1:i.call(t,e,n)},merge:function(e,t){for(var n=+t.length,r=0,i=e.length;r<n;r++)e[i++]=t[r];return e.length=i,e},grep:function(e,t,n){for(var r=[],i=0,o=e.length,a=!n;i<o;i++)!t(e[i],i)!==a&&r.push(e[i]);return r},map:function(e,t,n){var r,i,o=0,a=[];if(p(e))for(r=e.length;o<r;o++)null!=(i=t(e[o],o,n))&&a.push(i);else for(o in e)null!=(i=t(e[o],o,n))&&a.push(i);return g(a)},guid:1,support:y}),"function"==typeof Symbol&&(S.fn[Symbol.iterator]=t[Symbol.iterator]),S.each("Boolean Number String Function Array Date RegExp Object Error Symbol".split(" "),function(e,t){n["[object "+t+"]"]=t.toLowerCase()});var d=function(n){var e,d,b,o,i,h,f,g,w,u,l,T,C,a,E,v,s,c,y,S="sizzle"+1*new Date,p=n.document,k=0,r=0,m=ue(),x=ue(),A=ue(),N=ue(),j=function(e,t){return e===t&&(l=!0),0},D={}.hasOwnProperty,t=[],q=t.pop,L=t.push,H=t.push,O=t.slice,P=function(e,t){for(var n=0,r=e.length;n<r;n++)if(e[n]===t)return n;return-1},R="checked|selected|async|autofocus|autoplay|controls|defer|disabled|hidden|ismap|loop|multiple|open|readonly|required|scoped",M="[\\x20\\t\\r\\n\\f]",I="(?:\\\\[\\da-fA-F]{1,6}"+M+"?|\\\\[^\\r\\n\\f]|[\\w-]|[^\0-\\x7f])+",W="\\["+M+"*("+I+")(?:"+M+"*([*^$|!~]?=)"+M+"*(?:'((?:\\\\.|[^\\\\'])*)'|\"((?:\\\\.|[^\\\\\"])*)\"|("+I+"))|)"+M+"*\\]",F=":("+I+")(?:\\((('((?:\\\\.|[^\\\\'])*)'|\"((?:\\\\.|[^\\\\\"])*)\")|((?:\\\\.|[^\\\\()[\\]]|"+W+")*)|.*)\\)|)",B=new RegExp(M+"+","g"),$=new RegExp("^"+M+"+|((?:^|[^\\\\])(?:\\\\.)*)"+M+"+$","g"),_=new RegExp("^"+M+"*,"+M+"*"),z=new RegExp("^"+M+"*([>+~]|"+M+")"+M+"*"),U=new RegExp(M+"|>"),X=new RegExp(F),V=new RegExp("^"+I+"$"),G={ID:new RegExp("^#("+I+")"),CLASS:new RegExp("^\\.("+I+")"),TAG:new RegExp("^("+I+"|[*])"),ATTR:new RegExp("^"+W),PSEUDO:new RegExp("^"+F),CHILD:new RegExp("^:(only|first|last|nth|nth-last)-(child|of-type)(?:\\("+M+"*(even|odd|(([+-]|)(\\d*)n|)"+M+"*(?:([+-]|)"+M+"*(\\d+)|))"+M+"*\\)|)","i"),bool:new RegExp("^(?:"+R+")$","i"),needsContext:new RegExp("^"+M+"*[>+~]|:(even|odd|eq|gt|lt|nth|first|last)(?:\\("+M+"*((?:-\\d)?\\d*)"+M+"*\\)|)(?=[^-]|$)","i")},Y=/HTML$/i,Q=/^(?:input|select|textarea|button)$/i,J=/^h\d$/i,K=/^[^{]+\{\s*\[native \w/,Z=/^(?:#([\w-]+)|(\w+)|\.([\w-]+))$/,ee=/[+~]/,te=new RegExp("\\\\[\\da-fA-F]{1,6}"+M+"?|\\\\([^\\r\\n\\f])","g"),ne=function(e,t){var n="0x"+e.slice(1)-65536;return t||(n<0?String.fromCharCode(n+65536):String.fromCharCode(n>>10|55296,1023&n|56320))},re=/([\0-\x1f\x7f]|^-?\d)|^-$|[^\0-\x1f\x7f-\uFFFF\w-]/g,ie=function(e,t){return t?"\0"===e?"\ufffd":e.slice(0,-1)+"\\"+e.charCodeAt(e.length-1).toString(16)+" ":"\\"+e},oe=function(){T()},ae=be(function(e){return!0===e.disabled&&"fieldset"===e.nodeName.toLowerCase()},{dir:"parentNode",next:"legend"});try{H.apply(t=O.call(p.childNodes),p.childNodes),t[p.childNodes.length].nodeType}catch(e){H={apply:t.length?function(e,t){L.apply(e,O.call(t))}:function(e,t){var n=e.length,r=0;while(e[n++]=t[r++]);e.length=n-1}}}function se(t,e,n,r){var i,o,a,s,u,l,c,f=e&&e.ownerDocument,p=e?e.nodeType:9;if(n=n||[],"string"!=typeof t||!t||1!==p&&9!==p&&11!==p)return n;if(!r&&(T(e),e=e||C,E)){if(11!==p&&(u=Z.exec(t)))if(i=u[1]){if(9===p){if(!(a=e.getElementById(i)))return n;if(a.id===i)return n.push(a),n}else if(f&&(a=f.getElementById(i))&&y(e,a)&&a.id===i)return n.push(a),n}else{if(u[2])return H.apply(n,e.getElementsByTagName(t)),n;if((i=u[3])&&d.getElementsByClassName&&e.getElementsByClassName)return H.apply(n,e.getElementsByClassName(i)),n}if(d.qsa&&!N[t+" "]&&(!v||!v.test(t))&&(1!==p||"object"!==e.nodeName.toLowerCase())){if(c=t,f=e,1===p&&(U.test(t)||z.test(t))){(f=ee.test(t)&&ye(e.parentNode)||e)===e&&d.scope||((s=e.getAttribute("id"))?s=s.replace(re,ie):e.setAttribute("id",s=S)),o=(l=h(t)).length;while(o--)l[o]=(s?"#"+s:":scope")+" "+xe(l[o]);c=l.join(",")}try{return H.apply(n,f.querySelectorAll(c)),n}catch(e){N(t,!0)}finally{s===S&&e.removeAttribute("id")}}}return g(t.replace($,"$1"),e,n,r)}function ue(){var r=[];return function e(t,n){return r.push(t+" ")>b.cacheLength&&delete e[r.shift()],e[t+" "]=n}}function le(e){return e[S]=!0,e}function ce(e){var t=C.createElement("fieldset");try{return!!e(t)}catch(e){return!1}finally{t.parentNode&&t.parentNode.removeChild(t),t=null}}function fe(e,t){var n=e.split("|"),r=n.length;while(r--)b.attrHandle[n[r]]=t}function pe(e,t){var n=t&&e,r=n&&1===e.nodeType&&1===t.nodeType&&e.sourceIndex-t.sourceIndex;if(r)return r;if(n)while(n=n.nextSibling)if(n===t)return-1;return e?1:-1}function de(t){return function(e){return"input"===e.nodeName.toLowerCase()&&e.type===t}}function he(n){return function(e){var t=e.nodeName.toLowerCase();return("input"===t||"button"===t)&&e.type===n}}function ge(t){return function(e){return"form"in e?e.parentNode&&!1===e.disabled?"label"in e?"label"in e.parentNode?e.parentNode.disabled===t:e.disabled===t:e.isDisabled===t||e.isDisabled!==!t&&ae(e)===t:e.disabled===t:"label"in e&&e.disabled===t}}function ve(a){return le(function(o){return o=+o,le(function(e,t){var n,r=a([],e.length,o),i=r.length;while(i--)e[n=r[i]]&&(e[n]=!(t[n]=e[n]))})})}function ye(e){return e&&"undefined"!=typeof e.getElementsByTagName&&e}for(e in d=se.support={},i=se.isXML=function(e){var t=e&&e.namespaceURI,n=e&&(e.ownerDocument||e).documentElement;return!Y.test(t||n&&n.nodeName||"HTML")},T=se.setDocument=function(e){var t,n,r=e?e.ownerDocument||e:p;return r!=C&&9===r.nodeType&&r.documentElement&&(a=(C=r).documentElement,E=!i(C),p!=C&&(n=C.defaultView)&&n.top!==n&&(n.addEventListener?n.addEventListener("unload",oe,!1):n.attachEvent&&n.attachEvent("onunload",oe)),d.scope=ce(function(e){return a.appendChild(e).appendChild(C.createElement("div")),"undefined"!=typeof e.querySelectorAll&&!e.querySelectorAll(":scope fieldset div").length}),d.attributes=ce(function(e){return e.className="i",!e.getAttribute("className")}),d.getElementsByTagName=ce(function(e){return e.appendChild(C.createComment("")),!e.getElementsByTagName("*").length}),d.getElementsByClassName=K.test(C.getElementsByClassName),d.getById=ce(function(e){return a.appendChild(e).id=S,!C.getElementsByName||!C.getElementsByName(S).length}),d.getById?(b.filter.ID=function(e){var t=e.replace(te,ne);return function(e){return e.getAttribute("id")===t}},b.find.ID=function(e,t){if("undefined"!=typeof t.getElementById&&E){var n=t.getElementById(e);return n?[n]:[]}}):(b.filter.ID=function(e){var n=e.replace(te,ne);return function(e){var t="undefined"!=typeof e.getAttributeNode&&e.getAttributeNode("id");return t&&t.value===n}},b.find.ID=function(e,t){if("undefined"!=typeof t.getElementById&&E){var n,r,i,o=t.getElementById(e);if(o){if((n=o.getAttributeNode("id"))&&n.value===e)return[o];i=t.getElementsByName(e),r=0;while(o=i[r++])if((n=o.getAttributeNode("id"))&&n.value===e)return[o]}return[]}}),b.find.TAG=d.getElementsByTagName?function(e,t){return"undefined"!=typeof t.getElementsByTagName?t.getElementsByTagName(e):d.qsa?t.querySelectorAll(e):void 0}:function(e,t){var n,r=[],i=0,o=t.getElementsByTagName(e);if("*"===e){while(n=o[i++])1===n.nodeType&&r.push(n);return r}return o},b.find.CLASS=d.getElementsByClassName&&function(e,t){if("undefined"!=typeof t.getElementsByClassName&&E)return t.getElementsByClassName(e)},s=[],v=[],(d.qsa=K.test(C.querySelectorAll))&&(ce(function(e){var t;a.appendChild(e).innerHTML="<a id='"+S+"'></a><select id='"+S+"-\r\\' msallowcapture=''><option selected=''></option></select>",e.querySelectorAll("[msallowcapture^='']").length&&v.push("[*^$]="+M+"*(?:''|\"\")"),e.querySelectorAll("[selected]").length||v.push("\\["+M+"*(?:value|"+R+")"),e.querySelectorAll("[id~="+S+"-]").length||v.push("~="),(t=C.createElement("input")).setAttribute("name",""),e.appendChild(t),e.querySelectorAll("[name='']").length||v.push("\\["+M+"*name"+M+"*="+M+"*(?:''|\"\")"),e.querySelectorAll(":checked").length||v.push(":checked"),e.querySelectorAll("a#"+S+"+*").length||v.push(".#.+[+~]"),e.querySelectorAll("\\\f"),v.push("[\\r\\n\\f]")}),ce(function(e){e.innerHTML="<a href='' disabled='disabled'></a><select disabled='disabled'><option/></select>";var t=C.createElement("input");t.setAttribute("type","hidden"),e.appendChild(t).setAttribute("name","D"),e.querySelectorAll("[name=d]").length&&v.push("name"+M+"*[*^$|!~]?="),2!==e.querySelectorAll(":enabled").length&&v.push(":enabled",":disabled"),a.appendChild(e).disabled=!0,2!==e.querySelectorAll(":disabled").length&&v.push(":enabled",":disabled"),e.querySelectorAll("*,:x"),v.push(",.*:")})),(d.matchesSelector=K.test(c=a.matches||a.webkitMatchesSelector||a.mozMatchesSelector||a.oMatchesSelector||a.msMatchesSelector))&&ce(function(e){d.disconnectedMatch=c.call(e,"*"),c.call(e,"[s!='']:x"),s.push("!=",F)}),v=v.length&&new RegExp(v.join("|")),s=s.length&&new RegExp(s.join("|")),t=K.test(a.compareDocumentPosition),y=t||K.test(a.contains)?function(e,t){var n=9===e.nodeType?e.documentElement:e,r=t&&t.parentNode;return e===r||!(!r||1!==r.nodeType||!(n.contains?n.contains(r):e.compareDocumentPosition&&16&e.compareDocumentPosition(r)))}:function(e,t){if(t)while(t=t.parentNode)if(t===e)return!0;return!1},j=t?function(e,t){if(e===t)return l=!0,0;var n=!e.compareDocumentPosition-!t.compareDocumentPosition;return n||(1&(n=(e.ownerDocument||e)==(t.ownerDocument||t)?e.compareDocumentPosition(t):1)||!d.sortDetached&&t.compareDocumentPosition(e)===n?e==C||e.ownerDocument==p&&y(p,e)?-1:t==C||t.ownerDocument==p&&y(p,t)?1:u?P(u,e)-P(u,t):0:4&n?-1:1)}:function(e,t){if(e===t)return l=!0,0;var n,r=0,i=e.parentNode,o=t.parentNode,a=[e],s=[t];if(!i||!o)return e==C?-1:t==C?1:i?-1:o?1:u?P(u,e)-P(u,t):0;if(i===o)return pe(e,t);n=e;while(n=n.parentNode)a.unshift(n);n=t;while(n=n.parentNode)s.unshift(n);while(a[r]===s[r])r++;return r?pe(a[r],s[r]):a[r]==p?-1:s[r]==p?1:0}),C},se.matches=function(e,t){return se(e,null,null,t)},se.matchesSelector=function(e,t){if(T(e),d.matchesSelector&&E&&!N[t+" "]&&(!s||!s.test(t))&&(!v||!v.test(t)))try{var n=c.call(e,t);if(n||d.disconnectedMatch||e.document&&11!==e.document.nodeType)return n}catch(e){N(t,!0)}return 0<se(t,C,null,[e]).length},se.contains=function(e,t){return(e.ownerDocument||e)!=C&&T(e),y(e,t)},se.attr=function(e,t){(e.ownerDocument||e)!=C&&T(e);var n=b.attrHandle[t.toLowerCase()],r=n&&D.call(b.attrHandle,t.toLowerCase())?n(e,t,!E):void 0;return void 0!==r?r:d.attributes||!E?e.getAttribute(t):(r=e.getAttributeNode(t))&&r.specified?r.value:null},se.escape=function(e){return(e+"").replace(re,ie)},se.error=function(e){throw new Error("Syntax error, unrecognized expression: "+e)},se.uniqueSort=function(e){var t,n=[],r=0,i=0;if(l=!d.detectDuplicates,u=!d.sortStable&&e.slice(0),e.sort(j),l){while(t=e[i++])t===e[i]&&(r=n.push(i));while(r--)e.splice(n[r],1)}return u=null,e},o=se.getText=function(e){var t,n="",r=0,i=e.nodeType;if(i){if(1===i||9===i||11===i){if("string"==typeof e.textContent)return e.textContent;for(e=e.firstChild;e;e=e.nextSibling)n+=o(e)}else if(3===i||4===i)return e.nodeValue}else while(t=e[r++])n+=o(t);return n},(b=se.selectors={cacheLength:50,createPseudo:le,match:G,attrHandle:{},find:{},relative:{">":{dir:"parentNode",first:!0}," ":{dir:"parentNode"},"+":{dir:"previousSibling",first:!0},"~":{dir:"previousSibling"}},preFilter:{ATTR:function(e){return e[1]=e[1].replace(te,ne),e[3]=(e[3]||e[4]||e[5]||"").replace(te,ne),"~="===e[2]&&(e[3]=" "+e[3]+" "),e.slice(0,4)},CHILD:function(e){return e[1]=e[1].toLowerCase(),"nth"===e[1].slice(0,3)?(e[3]||se.error(e[0]),e[4]=+(e[4]?e[5]+(e[6]||1):2*("even"===e[3]||"odd"===e[3])),e[5]=+(e[7]+e[8]||"odd"===e[3])):e[3]&&se.error(e[0]),e},PSEUDO:function(e){var t,n=!e[6]&&e[2];return G.CHILD.test(e[0])?null:(e[3]?e[2]=e[4]||e[5]||"":n&&X.test(n)&&(t=h(n,!0))&&(t=n.indexOf(")",n.length-t)-n.length)&&(e[0]=e[0].slice(0,t),e[2]=n.slice(0,t)),e.slice(0,3))}},filter:{TAG:function(e){var t=e.replace(te,ne).toLowerCase();return"*"===e?function(){return!0}:function(e){return e.nodeName&&e.nodeName.toLowerCase()===t}},CLASS:function(e){var t=m[e+" "];return t||(t=new RegExp("(^|"+M+")"+e+"("+M+"|$)"))&&m(e,function(e){return t.test("string"==typeof e.className&&e.className||"undefined"!=typeof e.getAttribute&&e.getAttribute("class")||"")})},ATTR:function(n,r,i){return function(e){var t=se.attr(e,n);return null==t?"!="===r:!r||(t+="","="===r?t===i:"!="===r?t!==i:"^="===r?i&&0===t.indexOf(i):"*="===r?i&&-1<t.indexOf(i):"$="===r?i&&t.slice(-i.length)===i:"~="===r?-1<(" "+t.replace(B," ")+" ").indexOf(i):"|="===r&&(t===i||t.slice(0,i.length+1)===i+"-"))}},CHILD:function(h,e,t,g,v){var y="nth"!==h.slice(0,3),m="last"!==h.slice(-4),x="of-type"===e;return 1===g&&0===v?function(e){return!!e.parentNode}:function(e,t,n){var r,i,o,a,s,u,l=y!==m?"nextSibling":"previousSibling",c=e.parentNode,f=x&&e.nodeName.toLowerCase(),p=!n&&!x,d=!1;if(c){if(y){while(l){a=e;while(a=a[l])if(x?a.nodeName.toLowerCase()===f:1===a.nodeType)return!1;u=l="only"===h&&!u&&"nextSibling"}return!0}if(u=[m?c.firstChild:c.lastChild],m&&p){d=(s=(r=(i=(o=(a=c)[S]||(a[S]={}))[a.uniqueID]||(o[a.uniqueID]={}))[h]||[])[0]===k&&r[1])&&r[2],a=s&&c.childNodes[s];while(a=++s&&a&&a[l]||(d=s=0)||u.pop())if(1===a.nodeType&&++d&&a===e){i[h]=[k,s,d];break}}else if(p&&(d=s=(r=(i=(o=(a=e)[S]||(a[S]={}))[a.uniqueID]||(o[a.uniqueID]={}))[h]||[])[0]===k&&r[1]),!1===d)while(a=++s&&a&&a[l]||(d=s=0)||u.pop())if((x?a.nodeName.toLowerCase()===f:1===a.nodeType)&&++d&&(p&&((i=(o=a[S]||(a[S]={}))[a.uniqueID]||(o[a.uniqueID]={}))[h]=[k,d]),a===e))break;return(d-=v)===g||d%g==0&&0<=d/g}}},PSEUDO:function(e,o){var t,a=b.pseudos[e]||b.setFilters[e.toLowerCase()]||se.error("unsupported pseudo: "+e);return a[S]?a(o):1<a.length?(t=[e,e,"",o],b.setFilters.hasOwnProperty(e.toLowerCase())?le(function(e,t){var n,r=a(e,o),i=r.length;while(i--)e[n=P(e,r[i])]=!(t[n]=r[i])}):function(e){return a(e,0,t)}):a}},pseudos:{not:le(function(e){var r=[],i=[],s=f(e.replace($,"$1"));return s[S]?le(function(e,t,n,r){var i,o=s(e,null,r,[]),a=e.length;while(a--)(i=o[a])&&(e[a]=!(t[a]=i))}):function(e,t,n){return r[0]=e,s(r,null,n,i),r[0]=null,!i.pop()}}),has:le(function(t){return function(e){return 0<se(t,e).length}}),contains:le(function(t){return t=t.replace(te,ne),function(e){return-1<(e.textContent||o(e)).indexOf(t)}}),lang:le(function(n){return V.test(n||"")||se.error("unsupported lang: "+n),n=n.replace(te,ne).toLowerCase(),function(e){var t;do{if(t=E?e.lang:e.getAttribute("xml:lang")||e.getAttribute("lang"))return(t=t.toLowerCase())===n||0===t.indexOf(n+"-")}while((e=e.parentNode)&&1===e.nodeType);return!1}}),target:function(e){var t=n.location&&n.location.hash;return t&&t.slice(1)===e.id},root:function(e){return e===a},focus:function(e){return e===C.activeElement&&(!C.hasFocus||C.hasFocus())&&!!(e.type||e.href||~e.tabIndex)},enabled:ge(!1),disabled:ge(!0),checked:function(e){var t=e.nodeName.toLowerCase();return"input"===t&&!!e.checked||"option"===t&&!!e.selected},selected:function(e){return e.parentNode&&e.parentNode.selectedIndex,!0===e.selected},empty:function(e){for(e=e.firstChild;e;e=e.nextSibling)if(e.nodeType<6)return!1;return!0},parent:function(e){return!b.pseudos.empty(e)},header:function(e){return J.test(e.nodeName)},input:function(e){return Q.test(e.nodeName)},button:function(e){var t=e.nodeName.toLowerCase();return"input"===t&&"button"===e.type||"button"===t},text:function(e){var t;return"input"===e.nodeName.toLowerCase()&&"text"===e.type&&(null==(t=e.getAttribute("type"))||"text"===t.toLowerCase())},first:ve(function(){return[0]}),last:ve(function(e,t){return[t-1]}),eq:ve(function(e,t,n){return[n<0?n+t:n]}),even:ve(function(e,t){for(var n=0;n<t;n+=2)e.push(n);return e}),odd:ve(function(e,t){for(var n=1;n<t;n+=2)e.push(n);return e}),lt:ve(function(e,t,n){for(var r=n<0?n+t:t<n?t:n;0<=--r;)e.push(r);return e}),gt:ve(function(e,t,n){for(var r=n<0?n+t:n;++r<t;)e.push(r);return e})}}).pseudos.nth=b.pseudos.eq,{radio:!0,checkbox:!0,file:!0,password:!0,image:!0})b.pseudos[e]=de(e);for(e in{submit:!0,reset:!0})b.pseudos[e]=he(e);function me(){}function xe(e){for(var t=0,n=e.length,r="";t<n;t++)r+=e[t].value;return r}function be(s,e,t){var u=e.dir,l=e.next,c=l||u,f=t&&"parentNode"===c,p=r++;return e.first?function(e,t,n){while(e=e[u])if(1===e.nodeType||f)return s(e,t,n);return!1}:function(e,t,n){var r,i,o,a=[k,p];if(n){while(e=e[u])if((1===e.nodeType||f)&&s(e,t,n))return!0}else while(e=e[u])if(1===e.nodeType||f)if(i=(o=e[S]||(e[S]={}))[e.uniqueID]||(o[e.uniqueID]={}),l&&l===e.nodeName.toLowerCase())e=e[u]||e;else{if((r=i[c])&&r[0]===k&&r[1]===p)return a[2]=r[2];if((i[c]=a)[2]=s(e,t,n))return!0}return!1}}function we(i){return 1<i.length?function(e,t,n){var r=i.length;while(r--)if(!i[r](e,t,n))return!1;return!0}:i[0]}function Te(e,t,n,r,i){for(var o,a=[],s=0,u=e.length,l=null!=t;s<u;s++)(o=e[s])&&(n&&!n(o,r,i)||(a.push(o),l&&t.push(s)));return a}function Ce(d,h,g,v,y,e){return v&&!v[S]&&(v=Ce(v)),y&&!y[S]&&(y=Ce(y,e)),le(function(e,t,n,r){var i,o,a,s=[],u=[],l=t.length,c=e||function(e,t,n){for(var r=0,i=t.length;r<i;r++)se(e,t[r],n);return n}(h||"*",n.nodeType?[n]:n,[]),f=!d||!e&&h?c:Te(c,s,d,n,r),p=g?y||(e?d:l||v)?[]:t:f;if(g&&g(f,p,n,r),v){i=Te(p,u),v(i,[],n,r),o=i.length;while(o--)(a=i[o])&&(p[u[o]]=!(f[u[o]]=a))}if(e){if(y||d){if(y){i=[],o=p.length;while(o--)(a=p[o])&&i.push(f[o]=a);y(null,p=[],i,r)}o=p.length;while(o--)(a=p[o])&&-1<(i=y?P(e,a):s[o])&&(e[i]=!(t[i]=a))}}else p=Te(p===t?p.splice(l,p.length):p),y?y(null,t,p,r):H.apply(t,p)})}function Ee(e){for(var i,t,n,r=e.length,o=b.relative[e[0].type],a=o||b.relative[" "],s=o?1:0,u=be(function(e){return e===i},a,!0),l=be(function(e){return-1<P(i,e)},a,!0),c=[function(e,t,n){var r=!o&&(n||t!==w)||((i=t).nodeType?u(e,t,n):l(e,t,n));return i=null,r}];s<r;s++)if(t=b.relative[e[s].type])c=[be(we(c),t)];else{if((t=b.filter[e[s].type].apply(null,e[s].matches))[S]){for(n=++s;n<r;n++)if(b.relative[e[n].type])break;return Ce(1<s&&we(c),1<s&&xe(e.slice(0,s-1).concat({value:" "===e[s-2].type?"*":""})).replace($,"$1"),t,s<n&&Ee(e.slice(s,n)),n<r&&Ee(e=e.slice(n)),n<r&&xe(e))}c.push(t)}return we(c)}return me.prototype=b.filters=b.pseudos,b.setFilters=new me,h=se.tokenize=function(e,t){var n,r,i,o,a,s,u,l=x[e+" "];if(l)return t?0:l.slice(0);a=e,s=[],u=b.preFilter;while(a){for(o in n&&!(r=_.exec(a))||(r&&(a=a.slice(r[0].length)||a),s.push(i=[])),n=!1,(r=z.exec(a))&&(n=r.shift(),i.push({value:n,type:r[0].replace($," ")}),a=a.slice(n.length)),b.filter)!(r=G[o].exec(a))||u[o]&&!(r=u[o](r))||(n=r.shift(),i.push({value:n,type:o,matches:r}),a=a.slice(n.length));if(!n)break}return t?a.length:a?se.error(e):x(e,s).slice(0)},f=se.compile=function(e,t){var n,v,y,m,x,r,i=[],o=[],a=A[e+" "];if(!a){t||(t=h(e)),n=t.length;while(n--)(a=Ee(t[n]))[S]?i.push(a):o.push(a);(a=A(e,(v=o,m=0<(y=i).length,x=0<v.length,r=function(e,t,n,r,i){var o,a,s,u=0,l="0",c=e&&[],f=[],p=w,d=e||x&&b.find.TAG("*",i),h=k+=null==p?1:Math.random()||.1,g=d.length;for(i&&(w=t==C||t||i);l!==g&&null!=(o=d[l]);l++){if(x&&o){a=0,t||o.ownerDocument==C||(T(o),n=!E);while(s=v[a++])if(s(o,t||C,n)){r.push(o);break}i&&(k=h)}m&&((o=!s&&o)&&u--,e&&c.push(o))}if(u+=l,m&&l!==u){a=0;while(s=y[a++])s(c,f,t,n);if(e){if(0<u)while(l--)c[l]||f[l]||(f[l]=q.call(r));f=Te(f)}H.apply(r,f),i&&!e&&0<f.length&&1<u+y.length&&se.uniqueSort(r)}return i&&(k=h,w=p),c},m?le(r):r))).selector=e}return a},g=se.select=function(e,t,n,r){var i,o,a,s,u,l="function"==typeof e&&e,c=!r&&h(e=l.selector||e);if(n=n||[],1===c.length){if(2<(o=c[0]=c[0].slice(0)).length&&"ID"===(a=o[0]).type&&9===t.nodeType&&E&&b.relative[o[1].type]){if(!(t=(b.find.ID(a.matches[0].replace(te,ne),t)||[])[0]))return n;l&&(t=t.parentNode),e=e.slice(o.shift().value.length)}i=G.needsContext.test(e)?0:o.length;while(i--){if(a=o[i],b.relative[s=a.type])break;if((u=b.find[s])&&(r=u(a.matches[0].replace(te,ne),ee.test(o[0].type)&&ye(t.parentNode)||t))){if(o.splice(i,1),!(e=r.length&&xe(o)))return H.apply(n,r),n;break}}}return(l||f(e,c))(r,t,!E,n,!t||ee.test(e)&&ye(t.parentNode)||t),n},d.sortStable=S.split("").sort(j).join("")===S,d.detectDuplicates=!!l,T(),d.sortDetached=ce(function(e){return 1&e.compareDocumentPosition(C.createElement("fieldset"))}),ce(function(e){return e.innerHTML="<a href='#'></a>","#"===e.firstChild.getAttribute("href")})||fe("type|href|height|width",function(e,t,n){if(!n)return e.getAttribute(t,"type"===t.toLowerCase()?1:2)}),d.attributes&&ce(function(e){return e.innerHTML="<input/>",e.firstChild.setAttribute("value",""),""===e.firstChild.getAttribute("value")})||fe("value",function(e,t,n){if(!n&&"input"===e.nodeName.toLowerCase())return e.defaultValue}),ce(function(e){return null==e.getAttribute("disabled")})||fe(R,function(e,t,n){var r;if(!n)return!0===e[t]?t.toLowerCase():(r=e.getAttributeNode(t))&&r.specified?r.value:null}),se}(C);S.find=d,S.expr=d.selectors,S.expr[":"]=S.expr.pseudos,S.uniqueSort=S.unique=d.uniqueSort,S.text=d.getText,S.isXMLDoc=d.isXML,S.contains=d.contains,S.escapeSelector=d.escape;var h=function(e,t,n){var r=[],i=void 0!==n;while((e=e[t])&&9!==e.nodeType)if(1===e.nodeType){if(i&&S(e).is(n))break;r.push(e)}return r},T=function(e,t){for(var n=[];e;e=e.nextSibling)1===e.nodeType&&e!==t&&n.push(e);return n},k=S.expr.match.needsContext;function A(e,t){return e.nodeName&&e.nodeName.toLowerCase()===t.toLowerCase()}var N=/^<([a-z][^\/\0>:\x20\t\r\n\f]*)[\x20\t\r\n\f]*\/?>(?:<\/\1>|)$/i;function j(e,n,r){return m(n)?S.grep(e,function(e,t){return!!n.call(e,t,e)!==r}):n.nodeType?S.grep(e,function(e){return e===n!==r}):"string"!=typeof n?S.grep(e,function(e){return-1<i.call(n,e)!==r}):S.filter(n,e,r)}S.filter=function(e,t,n){var r=t[0];return n&&(e=":not("+e+")"),1===t.length&&1===r.nodeType?S.find.matchesSelector(r,e)?[r]:[]:S.find.matches(e,S.grep(t,function(e){return 1===e.nodeType}))},S.fn.extend({find:function(e){var t,n,r=this.length,i=this;if("string"!=typeof e)return this.pushStack(S(e).filter(function(){for(t=0;t<r;t++)if(S.contains(i[t],this))return!0}));for(n=this.pushStack([]),t=0;t<r;t++)S.find(e,i[t],n);return 1<r?S.uniqueSort(n):n},filter:function(e){return this.pushStack(j(this,e||[],!1))},not:function(e){return this.pushStack(j(this,e||[],!0))},is:function(e){return!!j(this,"string"==typeof e&&k.test(e)?S(e):e||[],!1).length}});var D,q=/^(?:\s*(<[\w\W]+>)[^>]*|#([\w-]+))$/;(S.fn.init=function(e,t,n){var r,i;if(!e)return this;if(n=n||D,"string"==typeof e){if(!(r="<"===e[0]&&">"===e[e.length-1]&&3<=e.length?[null,e,null]:q.exec(e))||!r[1]&&t)return!t||t.jquery?(t||n).find(e):this.constructor(t).find(e);if(r[1]){if(t=t instanceof S?t[0]:t,S.merge(this,S.parseHTML(r[1],t&&t.nodeType?t.ownerDocument||t:E,!0)),N.test(r[1])&&S.isPlainObject(t))for(r in t)m(this[r])?this[r](t[r]):this.attr(r,t[r]);return this}return(i=E.getElementById(r[2]))&&(this[0]=i,this.length=1),this}return e.nodeType?(this[0]=e,this.length=1,this):m(e)?void 0!==n.ready?n.ready(e):e(S):S.makeArray(e,this)}).prototype=S.fn,D=S(E);var L=/^(?:parents|prev(?:Until|All))/,H={children:!0,contents:!0,next:!0,prev:!0};function O(e,t){while((e=e[t])&&1!==e.nodeType);return e}S.fn.extend({has:function(e){var t=S(e,this),n=t.length;return this.filter(function(){for(var e=0;e<n;e++)if(S.contains(this,t[e]))return!0})},closest:function(e,t){var n,r=0,i=this.length,o=[],a="string"!=typeof e&&S(e);if(!k.test(e))for(;r<i;r++)for(n=this[r];n&&n!==t;n=n.parentNode)if(n.nodeType<11&&(a?-1<a.index(n):1===n.nodeType&&S.find.matchesSelector(n,e))){o.push(n);break}return this.pushStack(1<o.length?S.uniqueSort(o):o)},index:function(e){return e?"string"==typeof e?i.call(S(e),this[0]):i.call(this,e.jquery?e[0]:e):this[0]&&this[0].parentNode?this.first().prevAll().length:-1},add:function(e,t){return this.pushStack(S.uniqueSort(S.merge(this.get(),S(e,t))))},addBack:function(e){return this.add(null==e?this.prevObject:this.prevObject.filter(e))}}),S.each({parent:function(e){var t=e.parentNode;return t&&11!==t.nodeType?t:null},parents:function(e){return h(e,"parentNode")},parentsUntil:function(e,t,n){return h(e,"parentNode",n)},next:function(e){return O(e,"nextSibling")},prev:function(e){return O(e,"previousSibling")},nextAll:function(e){return h(e,"nextSibling")},prevAll:function(e){return h(e,"previousSibling")},nextUntil:function(e,t,n){return h(e,"nextSibling",n)},prevUntil:function(e,t,n){return h(e,"previousSibling",n)},siblings:function(e){return T((e.parentNode||{}).firstChild,e)},children:function(e){return T(e.firstChild)},contents:function(e){return null!=e.contentDocument&&r(e.contentDocument)?e.contentDocument:(A(e,"template")&&(e=e.content||e),S.merge([],e.childNodes))}},function(r,i){S.fn[r]=function(e,t){var n=S.map(this,i,e);return"Until"!==r.slice(-5)&&(t=e),t&&"string"==typeof t&&(n=S.filter(t,n)),1<this.length&&(H[r]||S.uniqueSort(n),L.test(r)&&n.reverse()),this.pushStack(n)}});var P=/[^\x20\t\r\n\f]+/g;function R(e){return e}function M(e){throw e}function I(e,t,n,r){var i;try{e&&m(i=e.promise)?i.call(e).done(t).fail(n):e&&m(i=e.then)?i.call(e,t,n):t.apply(void 0,[e].slice(r))}catch(e){n.apply(void 0,[e])}}S.Callbacks=function(r){var e,n;r="string"==typeof r?(e=r,n={},S.each(e.match(P)||[],function(e,t){n[t]=!0}),n):S.extend({},r);var i,t,o,a,s=[],u=[],l=-1,c=function(){for(a=a||r.once,o=i=!0;u.length;l=-1){t=u.shift();while(++l<s.length)!1===s[l].apply(t[0],t[1])&&r.stopOnFalse&&(l=s.length,t=!1)}r.memory||(t=!1),i=!1,a&&(s=t?[]:"")},f={add:function(){return s&&(t&&!i&&(l=s.length-1,u.push(t)),function n(e){S.each(e,function(e,t){m(t)?r.unique&&f.has(t)||s.push(t):t&&t.length&&"string"!==w(t)&&n(t)})}(arguments),t&&!i&&c()),this},remove:function(){return S.each(arguments,function(e,t){var n;while(-1<(n=S.inArray(t,s,n)))s.splice(n,1),n<=l&&l--}),this},has:function(e){return e?-1<S.inArray(e,s):0<s.length},empty:function(){return s&&(s=[]),this},disable:function(){return a=u=[],s=t="",this},disabled:function(){return!s},lock:function(){return a=u=[],t||i||(s=t=""),this},locked:function(){return!!a},fireWith:function(e,t){return a||(t=[e,(t=t||[]).slice?t.slice():t],u.push(t),i||c()),this},fire:function(){return f.fireWith(this,arguments),this},fired:function(){return!!o}};return f},S.extend({Deferred:function(e){var o=[["notify","progress",S.Callbacks("memory"),S.Callbacks("memory"),2],["resolve","done",S.Callbacks("once memory"),S.Callbacks("once memory"),0,"resolved"],["reject","fail",S.Callbacks("once memory"),S.Callbacks("once memory"),1,"rejected"]],i="pending",a={state:function(){return i},always:function(){return s.done(arguments).fail(arguments),this},"catch":function(e){return a.then(null,e)},pipe:function(){var i=arguments;return S.Deferred(function(r){S.each(o,function(e,t){var n=m(i[t[4]])&&i[t[4]];s[t[1]](function(){var e=n&&n.apply(this,arguments);e&&m(e.promise)?e.promise().progress(r.notify).done(r.resolve).fail(r.reject):r[t[0]+"With"](this,n?[e]:arguments)})}),i=null}).promise()},then:function(t,n,r){var u=0;function l(i,o,a,s){return function(){var n=this,r=arguments,e=function(){var e,t;if(!(i<u)){if((e=a.apply(n,r))===o.promise())throw new TypeError("Thenable self-resolution");t=e&&("object"==typeof e||"function"==typeof e)&&e.then,m(t)?s?t.call(e,l(u,o,R,s),l(u,o,M,s)):(u++,t.call(e,l(u,o,R,s),l(u,o,M,s),l(u,o,R,o.notifyWith))):(a!==R&&(n=void 0,r=[e]),(s||o.resolveWith)(n,r))}},t=s?e:function(){try{e()}catch(e){S.Deferred.exceptionHook&&S.Deferred.exceptionHook(e,t.stackTrace),u<=i+1&&(a!==M&&(n=void 0,r=[e]),o.rejectWith(n,r))}};i?t():(S.Deferred.getStackHook&&(t.stackTrace=S.Deferred.getStackHook()),C.setTimeout(t))}}return S.Deferred(function(e){o[0][3].add(l(0,e,m(r)?r:R,e.notifyWith)),o[1][3].add(l(0,e,m(t)?t:R)),o[2][3].add(l(0,e,m(n)?n:M))}).promise()},promise:function(e){return null!=e?S.extend(e,a):a}},s={};return S.each(o,function(e,t){var n=t[2],r=t[5];a[t[1]]=n.add,r&&n.add(function(){i=r},o[3-e][2].disable,o[3-e][3].disable,o[0][2].lock,o[0][3].lock),n.add(t[3].fire),s[t[0]]=function(){return s[t[0]+"With"](this===s?void 0:this,arguments),this},s[t[0]+"With"]=n.fireWith}),a.promise(s),e&&e.call(s,s),s},when:function(e){var n=arguments.length,t=n,r=Array(t),i=s.call(arguments),o=S.Deferred(),a=function(t){return function(e){r[t]=this,i[t]=1<arguments.length?s.call(arguments):e,--n||o.resolveWith(r,i)}};if(n<=1&&(I(e,o.done(a(t)).resolve,o.reject,!n),"pending"===o.state()||m(i[t]&&i[t].then)))return o.then();while(t--)I(i[t],a(t),o.reject);return o.promise()}});var W=/^(Eval|Internal|Range|Reference|Syntax|Type|URI)Error$/;S.Deferred.exceptionHook=function(e,t){C.console&&C.console.warn&&e&&W.test(e.name)&&C.console.warn("jQuery.Deferred exception: "+e.message,e.stack,t)},S.readyException=function(e){C.setTimeout(function(){throw e})};var F=S.Deferred();function B(){E.removeEventListener("DOMContentLoaded",B),C.removeEventListener("load",B),S.ready()}S.fn.ready=function(e){return F.then(e)["catch"](function(e){S.readyException(e)}),this},S.extend({isReady:!1,readyWait:1,ready:function(e){(!0===e?--S.readyWait:S.isReady)||(S.isReady=!0)!==e&&0<--S.readyWait||F.resolveWith(E,[S])}}),S.ready.then=F.then,"complete"===E.readyState||"loading"!==E.readyState&&!E.documentElement.doScroll?C.setTimeout(S.ready):(E.addEventListener("DOMContentLoaded",B),C.addEventListener("load",B));var $=function(e,t,n,r,i,o,a){var s=0,u=e.length,l=null==n;if("object"===w(n))for(s in i=!0,n)$(e,t,s,n[s],!0,o,a);else if(void 0!==r&&(i=!0,m(r)||(a=!0),l&&(a?(t.call(e,r),t=null):(l=t,t=function(e,t,n){return l.call(S(e),n)})),t))for(;s<u;s++)t(e[s],n,a?r:r.call(e[s],s,t(e[s],n)));return i?e:l?t.call(e):u?t(e[0],n):o},_=/^-ms-/,z=/-([a-z])/g;function U(e,t){return t.toUpperCase()}function X(e){return e.replace(_,"ms-").replace(z,U)}var V=function(e){return 1===e.nodeType||9===e.nodeType||!+e.nodeType};function G(){this.expando=S.expando+G.uid++}G.uid=1,G.prototype={cache:function(e){var t=e[this.expando];return t||(t={},V(e)&&(e.nodeType?e[this.expando]=t:Object.defineProperty(e,this.expando,{value:t,configurable:!0}))),t},set:function(e,t,n){var r,i=this.cache(e);if("string"==typeof t)i[X(t)]=n;else for(r in t)i[X(r)]=t[r];return i},get:function(e,t){return void 0===t?this.cache(e):e[this.expando]&&e[this.expando][X(t)]},access:function(e,t,n){return void 0===t||t&&"string"==typeof t&&void 0===n?this.get(e,t):(this.set(e,t,n),void 0!==n?n:t)},remove:function(e,t){var n,r=e[this.expando];if(void 0!==r){if(void 0!==t){n=(t=Array.isArray(t)?t.map(X):(t=X(t))in r?[t]:t.match(P)||[]).length;while(n--)delete r[t[n]]}(void 0===t||S.isEmptyObject(r))&&(e.nodeType?e[this.expando]=void 0:delete e[this.expando])}},hasData:function(e){var t=e[this.expando];return void 0!==t&&!S.isEmptyObject(t)}};var Y=new G,Q=new G,J=/^(?:\{[\w\W]*\}|\[[\w\W]*\])$/,K=/[A-Z]/g;function Z(e,t,n){var r,i;if(void 0===n&&1===e.nodeType)if(r="data-"+t.replace(K,"-$&").toLowerCase(),"string"==typeof(n=e.getAttribute(r))){try{n="true"===(i=n)||"false"!==i&&("null"===i?null:i===+i+""?+i:J.test(i)?JSON.parse(i):i)}catch(e){}Q.set(e,t,n)}else n=void 0;return n}S.extend({hasData:function(e){return Q.hasData(e)||Y.hasData(e)},data:function(e,t,n){return Q.access(e,t,n)},removeData:function(e,t){Q.remove(e,t)},_data:function(e,t,n){return Y.access(e,t,n)},_removeData:function(e,t){Y.remove(e,t)}}),S.fn.extend({data:function(n,e){var t,r,i,o=this[0],a=o&&o.attributes;if(void 0===n){if(this.length&&(i=Q.get(o),1===o.nodeType&&!Y.get(o,"hasDataAttrs"))){t=a.length;while(t--)a[t]&&0===(r=a[t].name).indexOf("data-")&&(r=X(r.slice(5)),Z(o,r,i[r]));Y.set(o,"hasDataAttrs",!0)}return i}return"object"==typeof n?this.each(function(){Q.set(this,n)}):$(this,function(e){var t;if(o&&void 0===e)return void 0!==(t=Q.get(o,n))?t:void 0!==(t=Z(o,n))?t:void 0;this.each(function(){Q.set(this,n,e)})},null,e,1<arguments.length,null,!0)},removeData:function(e){return this.each(function(){Q.remove(this,e)})}}),S.extend({queue:function(e,t,n){var r;if(e)return t=(t||"fx")+"queue",r=Y.get(e,t),n&&(!r||Array.isArray(n)?r=Y.access(e,t,S.makeArray(n)):r.push(n)),r||[]},dequeue:function(e,t){t=t||"fx";var n=S.queue(e,t),r=n.length,i=n.shift(),o=S._queueHooks(e,t);"inprogress"===i&&(i=n.shift(),r--),i&&("fx"===t&&n.unshift("inprogress"),delete o.stop,i.call(e,function(){S.dequeue(e,t)},o)),!r&&o&&o.empty.fire()},_queueHooks:function(e,t){var n=t+"queueHooks";return Y.get(e,n)||Y.access(e,n,{empty:S.Callbacks("once memory").add(function(){Y.remove(e,[t+"queue",n])})})}}),S.fn.extend({queue:function(t,n){var e=2;return"string"!=typeof t&&(n=t,t="fx",e--),arguments.length<e?S.queue(this[0],t):void 0===n?this:this.each(function(){var e=S.queue(this,t,n);S._queueHooks(this,t),"fx"===t&&"inprogress"!==e[0]&&S.dequeue(this,t)})},dequeue:function(e){return this.each(function(){S.dequeue(this,e)})},clearQueue:function(e){return this.queue(e||"fx",[])},promise:function(e,t){var n,r=1,i=S.Deferred(),o=this,a=this.length,s=function(){--r||i.resolveWith(o,[o])};"string"!=typeof e&&(t=e,e=void 0),e=e||"fx";while(a--)(n=Y.get(o[a],e+"queueHooks"))&&n.empty&&(r++,n.empty.add(s));return s(),i.promise(t)}});var ee=/[+-]?(?:\d*\.|)\d+(?:[eE][+-]?\d+|)/.source,te=new RegExp("^(?:([+-])=|)("+ee+")([a-z%]*)$","i"),ne=["Top","Right","Bottom","Left"],re=E.documentElement,ie=function(e){return S.contains(e.ownerDocument,e)},oe={composed:!0};re.getRootNode&&(ie=function(e){return S.contains(e.ownerDocument,e)||e.getRootNode(oe)===e.ownerDocument});var ae=function(e,t){return"none"===(e=t||e).style.display||""===e.style.display&&ie(e)&&"none"===S.css(e,"display")};function se(e,t,n,r){var i,o,a=20,s=r?function(){return r.cur()}:function(){return S.css(e,t,"")},u=s(),l=n&&n[3]||(S.cssNumber[t]?"":"px"),c=e.nodeType&&(S.cssNumber[t]||"px"!==l&&+u)&&te.exec(S.css(e,t));if(c&&c[3]!==l){u/=2,l=l||c[3],c=+u||1;while(a--)S.style(e,t,c+l),(1-o)*(1-(o=s()/u||.5))<=0&&(a=0),c/=o;c*=2,S.style(e,t,c+l),n=n||[]}return n&&(c=+c||+u||0,i=n[1]?c+(n[1]+1)*n[2]:+n[2],r&&(r.unit=l,r.start=c,r.end=i)),i}var ue={};function le(e,t){for(var n,r,i,o,a,s,u,l=[],c=0,f=e.length;c<f;c++)(r=e[c]).style&&(n=r.style.display,t?("none"===n&&(l[c]=Y.get(r,"display")||null,l[c]||(r.style.display="")),""===r.style.display&&ae(r)&&(l[c]=(u=a=o=void 0,a=(i=r).ownerDocument,s=i.nodeName,(u=ue[s])||(o=a.body.appendChild(a.createElement(s)),u=S.css(o,"display"),o.parentNode.removeChild(o),"none"===u&&(u="block"),ue[s]=u)))):"none"!==n&&(l[c]="none",Y.set(r,"display",n)));for(c=0;c<f;c++)null!=l[c]&&(e[c].style.display=l[c]);return e}S.fn.extend({show:function(){return le(this,!0)},hide:function(){return le(this)},toggle:function(e){return"boolean"==typeof e?e?this.show():this.hide():this.each(function(){ae(this)?S(this).show():S(this).hide()})}});var ce,fe,pe=/^(?:checkbox|radio)$/i,de=/<([a-z][^\/\0>\x20\t\r\n\f]*)/i,he=/^$|^module$|\/(?:java|ecma)script/i;ce=E.createDocumentFragment().appendChild(E.createElement("div")),(fe=E.createElement("input")).setAttribute("type","radio"),fe.setAttribute("checked","checked"),fe.setAttribute("name","t"),ce.appendChild(fe),y.checkClone=ce.cloneNode(!0).cloneNode(!0).lastChild.checked,ce.innerHTML="<textarea>x</textarea>",y.noCloneChecked=!!ce.cloneNode(!0).lastChild.defaultValue,ce.innerHTML="<option></option>",y.option=!!ce.lastChild;var ge={thead:[1,"<table>","</table>"],col:[2,"<table><colgroup>","</colgroup></table>"],tr:[2,"<table><tbody>","</tbody></table>"],td:[3,"<table><tbody><tr>","</tr></tbody></table>"],_default:[0,"",""]};function ve(e,t){var n;return n="undefined"!=typeof e.getElementsByTagName?e.getElementsByTagName(t||"*"):"undefined"!=typeof e.querySelectorAll?e.querySelectorAll(t||"*"):[],void 0===t||t&&A(e,t)?S.merge([e],n):n}function ye(e,t){for(var n=0,r=e.length;n<r;n++)Y.set(e[n],"globalEval",!t||Y.get(t[n],"globalEval"))}ge.tbody=ge.tfoot=ge.colgroup=ge.caption=ge.thead,ge.th=ge.td,y.option||(ge.optgroup=ge.option=[1,"<select multiple='multiple'>","</select>"]);var me=/<|&#?\w+;/;function xe(e,t,n,r,i){for(var o,a,s,u,l,c,f=t.createDocumentFragment(),p=[],d=0,h=e.length;d<h;d++)if((o=e[d])||0===o)if("object"===w(o))S.merge(p,o.nodeType?[o]:o);else if(me.test(o)){a=a||f.appendChild(t.createElement("div")),s=(de.exec(o)||["",""])[1].toLowerCase(),u=ge[s]||ge._default,a.innerHTML=u[1]+S.htmlPrefilter(o)+u[2],c=u[0];while(c--)a=a.lastChild;S.merge(p,a.childNodes),(a=f.firstChild).textContent=""}else p.push(t.createTextNode(o));f.textContent="",d=0;while(o=p[d++])if(r&&-1<S.inArray(o,r))i&&i.push(o);else if(l=ie(o),a=ve(f.appendChild(o),"script"),l&&ye(a),n){c=0;while(o=a[c++])he.test(o.type||"")&&n.push(o)}return f}var be=/^([^.]*)(?:\.(.+)|)/;function we(){return!0}function Te(){return!1}function Ce(e,t){return e===function(){try{return E.activeElement}catch(e){}}()==("focus"===t)}function Ee(e,t,n,r,i,o){var a,s;if("object"==typeof t){for(s in"string"!=typeof n&&(r=r||n,n=void 0),t)Ee(e,s,n,r,t[s],o);return e}if(null==r&&null==i?(i=n,r=n=void 0):null==i&&("string"==typeof n?(i=r,r=void 0):(i=r,r=n,n=void 0)),!1===i)i=Te;else if(!i)return e;return 1===o&&(a=i,(i=function(e){return S().off(e),a.apply(this,arguments)}).guid=a.guid||(a.guid=S.guid++)),e.each(function(){S.event.add(this,t,i,r,n)})}function Se(e,i,o){o?(Y.set(e,i,!1),S.event.add(e,i,{namespace:!1,handler:function(e){var t,n,r=Y.get(this,i);if(1&e.isTrigger&&this[i]){if(r.length)(S.event.special[i]||{}).delegateType&&e.stopPropagation();else if(r=s.call(arguments),Y.set(this,i,r),t=o(this,i),this[i](),r!==(n=Y.get(this,i))||t?Y.set(this,i,!1):n={},r!==n)return e.stopImmediatePropagation(),e.preventDefault(),n&&n.value}else r.length&&(Y.set(this,i,{value:S.event.trigger(S.extend(r[0],S.Event.prototype),r.slice(1),this)}),e.stopImmediatePropagation())}})):void 0===Y.get(e,i)&&S.event.add(e,i,we)}S.event={global:{},add:function(t,e,n,r,i){var o,a,s,u,l,c,f,p,d,h,g,v=Y.get(t);if(V(t)){n.handler&&(n=(o=n).handler,i=o.selector),i&&S.find.matchesSelector(re,i),n.guid||(n.guid=S.guid++),(u=v.events)||(u=v.events=Object.create(null)),(a=v.handle)||(a=v.handle=function(e){return"undefined"!=typeof S&&S.event.triggered!==e.type?S.event.dispatch.apply(t,arguments):void 0}),l=(e=(e||"").match(P)||[""]).length;while(l--)d=g=(s=be.exec(e[l])||[])[1],h=(s[2]||"").split(".").sort(),d&&(f=S.event.special[d]||{},d=(i?f.delegateType:f.bindType)||d,f=S.event.special[d]||{},c=S.extend({type:d,origType:g,data:r,handler:n,guid:n.guid,selector:i,needsContext:i&&S.expr.match.needsContext.test(i),namespace:h.join(".")},o),(p=u[d])||((p=u[d]=[]).delegateCount=0,f.setup&&!1!==f.setup.call(t,r,h,a)||t.addEventListener&&t.addEventListener(d,a)),f.add&&(f.add.call(t,c),c.handler.guid||(c.handler.guid=n.guid)),i?p.splice(p.delegateCount++,0,c):p.push(c),S.event.global[d]=!0)}},remove:function(e,t,n,r,i){var o,a,s,u,l,c,f,p,d,h,g,v=Y.hasData(e)&&Y.get(e);if(v&&(u=v.events)){l=(t=(t||"").match(P)||[""]).length;while(l--)if(d=g=(s=be.exec(t[l])||[])[1],h=(s[2]||"").split(".").sort(),d){f=S.event.special[d]||{},p=u[d=(r?f.delegateType:f.bindType)||d]||[],s=s[2]&&new RegExp("(^|\\.)"+h.join("\\.(?:.*\\.|)")+"(\\.|$)"),a=o=p.length;while(o--)c=p[o],!i&&g!==c.origType||n&&n.guid!==c.guid||s&&!s.test(c.namespace)||r&&r!==c.selector&&("**"!==r||!c.selector)||(p.splice(o,1),c.selector&&p.delegateCount--,f.remove&&f.remove.call(e,c));a&&!p.length&&(f.teardown&&!1!==f.teardown.call(e,h,v.handle)||S.removeEvent(e,d,v.handle),delete u[d])}else for(d in u)S.event.remove(e,d+t[l],n,r,!0);S.isEmptyObject(u)&&Y.remove(e,"handle events")}},dispatch:function(e){var t,n,r,i,o,a,s=new Array(arguments.length),u=S.event.fix(e),l=(Y.get(this,"events")||Object.create(null))[u.type]||[],c=S.event.special[u.type]||{};for(s[0]=u,t=1;t<arguments.length;t++)s[t]=arguments[t];if(u.delegateTarget=this,!c.preDispatch||!1!==c.preDispatch.call(this,u)){a=S.event.handlers.call(this,u,l),t=0;while((i=a[t++])&&!u.isPropagationStopped()){u.currentTarget=i.elem,n=0;while((o=i.handlers[n++])&&!u.isImmediatePropagationStopped())u.rnamespace&&!1!==o.namespace&&!u.rnamespace.test(o.namespace)||(u.handleObj=o,u.data=o.data,void 0!==(r=((S.event.special[o.origType]||{}).handle||o.handler).apply(i.elem,s))&&!1===(u.result=r)&&(u.preventDefault(),u.stopPropagation()))}return c.postDispatch&&c.postDispatch.call(this,u),u.result}},handlers:function(e,t){var n,r,i,o,a,s=[],u=t.delegateCount,l=e.target;if(u&&l.nodeType&&!("click"===e.type&&1<=e.button))for(;l!==this;l=l.parentNode||this)if(1===l.nodeType&&("click"!==e.type||!0!==l.disabled)){for(o=[],a={},n=0;n<u;n++)void 0===a[i=(r=t[n]).selector+" "]&&(a[i]=r.needsContext?-1<S(i,this).index(l):S.find(i,this,null,[l]).length),a[i]&&o.push(r);o.length&&s.push({elem:l,handlers:o})}return l=this,u<t.length&&s.push({elem:l,handlers:t.slice(u)}),s},addProp:function(t,e){Object.defineProperty(S.Event.prototype,t,{enumerable:!0,configurable:!0,get:m(e)?function(){if(this.originalEvent)return e(this.originalEvent)}:function(){if(this.originalEvent)return this.originalEvent[t]},set:function(e){Object.defineProperty(this,t,{enumerable:!0,configurable:!0,writable:!0,value:e})}})},fix:function(e){return e[S.expando]?e:new S.Event(e)},special:{load:{noBubble:!0},click:{setup:function(e){var t=this||e;return pe.test(t.type)&&t.click&&A(t,"input")&&Se(t,"click",we),!1},trigger:function(e){var t=this||e;return pe.test(t.type)&&t.click&&A(t,"input")&&Se(t,"click"),!0},_default:function(e){var t=e.target;return pe.test(t.type)&&t.click&&A(t,"input")&&Y.get(t,"click")||A(t,"a")}},beforeunload:{postDispatch:function(e){void 0!==e.result&&e.originalEvent&&(e.originalEvent.returnValue=e.result)}}}},S.removeEvent=function(e,t,n){e.removeEventListener&&e.removeEventListener(t,n)},S.Event=function(e,t){if(!(this instanceof S.Event))return new S.Event(e,t);e&&e.type?(this.originalEvent=e,this.type=e.type,this.isDefaultPrevented=e.defaultPrevented||void 0===e.defaultPrevented&&!1===e.returnValue?we:Te,this.target=e.target&&3===e.target.nodeType?e.target.parentNode:e.target,this.currentTarget=e.currentTarget,this.relatedTarget=e.relatedTarget):this.type=e,t&&S.extend(this,t),this.timeStamp=e&&e.timeStamp||Date.now(),this[S.expando]=!0},S.Event.prototype={constructor:S.Event,isDefaultPrevented:Te,isPropagationStopped:Te,isImmediatePropagationStopped:Te,isSimulated:!1,preventDefault:function(){var e=this.originalEvent;this.isDefaultPrevented=we,e&&!this.isSimulated&&e.preventDefault()},stopPropagation:function(){var e=this.originalEvent;this.isPropagationStopped=we,e&&!this.isSimulated&&e.stopPropagation()},stopImmediatePropagation:function(){var e=this.originalEvent;this.isImmediatePropagationStopped=we,e&&!this.isSimulated&&e.stopImmediatePropagation(),this.stopPropagation()}},S.each({altKey:!0,bubbles:!0,cancelable:!0,changedTouches:!0,ctrlKey:!0,detail:!0,eventPhase:!0,metaKey:!0,pageX:!0,pageY:!0,shiftKey:!0,view:!0,"char":!0,code:!0,charCode:!0,key:!0,keyCode:!0,button:!0,buttons:!0,clientX:!0,clientY:!0,offsetX:!0,offsetY:!0,pointerId:!0,pointerType:!0,screenX:!0,screenY:!0,targetTouches:!0,toElement:!0,touches:!0,which:!0},S.event.addProp),S.each({focus:"focusin",blur:"focusout"},function(e,t){S.event.special[e]={setup:function(){return Se(this,e,Ce),!1},trigger:function(){return Se(this,e),!0},_default:function(){return!0},delegateType:t}}),S.each({mouseenter:"mouseover",mouseleave:"mouseout",pointerenter:"pointerover",pointerleave:"pointerout"},function(e,i){S.event.special[e]={delegateType:i,bindType:i,handle:function(e){var t,n=e.relatedTarget,r=e.handleObj;return n&&(n===this||S.contains(this,n))||(e.type=r.origType,t=r.handler.apply(this,arguments),e.type=i),t}}}),S.fn.extend({on:function(e,t,n,r){return Ee(this,e,t,n,r)},one:function(e,t,n,r){return Ee(this,e,t,n,r,1)},off:function(e,t,n){var r,i;if(e&&e.preventDefault&&e.handleObj)return r=e.handleObj,S(e.delegateTarget).off(r.namespace?r.origType+"."+r.namespace:r.origType,r.selector,r.handler),this;if("object"==typeof e){for(i in e)this.off(i,t,e[i]);return this}return!1!==t&&"function"!=typeof t||(n=t,t=void 0),!1===n&&(n=Te),this.each(function(){S.event.remove(this,e,n,t)})}});var ke=/<script|<style|<link/i,Ae=/checked\s*(?:[^=]|=\s*.checked.)/i,Ne=/^\s*<!(?:\[CDATA\[|--)|(?:\]\]|--)>\s*$/g;function je(e,t){return A(e,"table")&&A(11!==t.nodeType?t:t.firstChild,"tr")&&S(e).children("tbody")[0]||e}function De(e){return e.type=(null!==e.getAttribute("type"))+"/"+e.type,e}function qe(e){return"true/"===(e.type||"").slice(0,5)?e.type=e.type.slice(5):e.removeAttribute("type"),e}function Le(e,t){var n,r,i,o,a,s;if(1===t.nodeType){if(Y.hasData(e)&&(s=Y.get(e).events))for(i in Y.remove(t,"handle events"),s)for(n=0,r=s[i].length;n<r;n++)S.event.add(t,i,s[i][n]);Q.hasData(e)&&(o=Q.access(e),a=S.extend({},o),Q.set(t,a))}}function He(n,r,i,o){r=g(r);var e,t,a,s,u,l,c=0,f=n.length,p=f-1,d=r[0],h=m(d);if(h||1<f&&"string"==typeof d&&!y.checkClone&&Ae.test(d))return n.each(function(e){var t=n.eq(e);h&&(r[0]=d.call(this,e,t.html())),He(t,r,i,o)});if(f&&(t=(e=xe(r,n[0].ownerDocument,!1,n,o)).firstChild,1===e.childNodes.length&&(e=t),t||o)){for(s=(a=S.map(ve(e,"script"),De)).length;c<f;c++)u=e,c!==p&&(u=S.clone(u,!0,!0),s&&S.merge(a,ve(u,"script"))),i.call(n[c],u,c);if(s)for(l=a[a.length-1].ownerDocument,S.map(a,qe),c=0;c<s;c++)u=a[c],he.test(u.type||"")&&!Y.access(u,"globalEval")&&S.contains(l,u)&&(u.src&&"module"!==(u.type||"").toLowerCase()?S._evalUrl&&!u.noModule&&S._evalUrl(u.src,{nonce:u.nonce||u.getAttribute("nonce")},l):b(u.textContent.replace(Ne,""),u,l))}return n}function Oe(e,t,n){for(var r,i=t?S.filter(t,e):e,o=0;null!=(r=i[o]);o++)n||1!==r.nodeType||S.cleanData(ve(r)),r.parentNode&&(n&&ie(r)&&ye(ve(r,"script")),r.parentNode.removeChild(r));return e}S.extend({htmlPrefilter:function(e){return e},clone:function(e,t,n){var r,i,o,a,s,u,l,c=e.cloneNode(!0),f=ie(e);if(!(y.noCloneChecked||1!==e.nodeType&&11!==e.nodeType||S.isXMLDoc(e)))for(a=ve(c),r=0,i=(o=ve(e)).length;r<i;r++)s=o[r],u=a[r],void 0,"input"===(l=u.nodeName.toLowerCase())&&pe.test(s.type)?u.checked=s.checked:"input"!==l&&"textarea"!==l||(u.defaultValue=s.defaultValue);if(t)if(n)for(o=o||ve(e),a=a||ve(c),r=0,i=o.length;r<i;r++)Le(o[r],a[r]);else Le(e,c);return 0<(a=ve(c,"script")).length&&ye(a,!f&&ve(e,"script")),c},cleanData:function(e){for(var t,n,r,i=S.event.special,o=0;void 0!==(n=e[o]);o++)if(V(n)){if(t=n[Y.expando]){if(t.events)for(r in t.events)i[r]?S.event.remove(n,r):S.removeEvent(n,r,t.handle);n[Y.expando]=void 0}n[Q.expando]&&(n[Q.expando]=void 0)}}}),S.fn.extend({detach:function(e){return Oe(this,e,!0)},remove:function(e){return Oe(this,e)},text:function(e){return $(this,function(e){return void 0===e?S.text(this):this.empty().each(function(){1!==this.nodeType&&11!==this.nodeType&&9!==this.nodeType||(this.textContent=e)})},null,e,arguments.length)},append:function(){return He(this,arguments,function(e){1!==this.nodeType&&11!==this.nodeType&&9!==this.nodeType||je(this,e).appendChild(e)})},prepend:function(){return He(this,arguments,function(e){if(1===this.nodeType||11===this.nodeType||9===this.nodeType){var t=je(this,e);t.insertBefore(e,t.firstChild)}})},before:function(){return He(this,arguments,function(e){this.parentNode&&this.parentNode.insertBefore(e,this)})},after:function(){return He(this,arguments,function(e){this.parentNode&&this.parentNode.insertBefore(e,this.nextSibling)})},empty:function(){for(var e,t=0;null!=(e=this[t]);t++)1===e.nodeType&&(S.cleanData(ve(e,!1)),e.textContent="");return this},clone:function(e,t){return e=null!=e&&e,t=null==t?e:t,this.map(function(){return S.clone(this,e,t)})},html:function(e){return $(this,function(e){var t=this[0]||{},n=0,r=this.length;if(void 0===e&&1===t.nodeType)return t.innerHTML;if("string"==typeof e&&!ke.test(e)&&!ge[(de.exec(e)||["",""])[1].toLowerCase()]){e=S.htmlPrefilter(e);try{for(;n<r;n++)1===(t=this[n]||{}).nodeType&&(S.cleanData(ve(t,!1)),t.innerHTML=e);t=0}catch(e){}}t&&this.empty().append(e)},null,e,arguments.length)},replaceWith:function(){var n=[];return He(this,arguments,function(e){var t=this.parentNode;S.inArray(this,n)<0&&(S.cleanData(ve(this)),t&&t.replaceChild(e,this))},n)}}),S.each({appendTo:"append",prependTo:"prepend",insertBefore:"before",insertAfter:"after",replaceAll:"replaceWith"},function(e,a){S.fn[e]=function(e){for(var t,n=[],r=S(e),i=r.length-1,o=0;o<=i;o++)t=o===i?this:this.clone(!0),S(r[o])[a](t),u.apply(n,t.get());return this.pushStack(n)}});var Pe=new RegExp("^("+ee+")(?!px)[a-z%]+$","i"),Re=function(e){var t=e.ownerDocument.defaultView;return t&&t.opener||(t=C),t.getComputedStyle(e)},Me=function(e,t,n){var r,i,o={};for(i in t)o[i]=e.style[i],e.style[i]=t[i];for(i in r=n.call(e),t)e.style[i]=o[i];return r},Ie=new RegExp(ne.join("|"),"i");function We(e,t,n){var r,i,o,a,s=e.style;return(n=n||Re(e))&&(""!==(a=n.getPropertyValue(t)||n[t])||ie(e)||(a=S.style(e,t)),!y.pixelBoxStyles()&&Pe.test(a)&&Ie.test(t)&&(r=s.width,i=s.minWidth,o=s.maxWidth,s.minWidth=s.maxWidth=s.width=a,a=n.width,s.width=r,s.minWidth=i,s.maxWidth=o)),void 0!==a?a+"":a}function Fe(e,t){return{get:function(){if(!e())return(this.get=t).apply(this,arguments);delete this.get}}}!function(){function e(){if(l){u.style.cssText="position:absolute;left:-11111px;width:60px;margin-top:1px;padding:0;border:0",l.style.cssText="position:relative;display:block;box-sizing:border-box;overflow:scroll;margin:auto;border:1px;padding:1px;width:60%;top:1%",re.appendChild(u).appendChild(l);var e=C.getComputedStyle(l);n="1%"!==e.top,s=12===t(e.marginLeft),l.style.right="60%",o=36===t(e.right),r=36===t(e.width),l.style.position="absolute",i=12===t(l.offsetWidth/3),re.removeChild(u),l=null}}function t(e){return Math.round(parseFloat(e))}var n,r,i,o,a,s,u=E.createElement("div"),l=E.createElement("div");l.style&&(l.style.backgroundClip="content-box",l.cloneNode(!0).style.backgroundClip="",y.clearCloneStyle="content-box"===l.style.backgroundClip,S.extend(y,{boxSizingReliable:function(){return e(),r},pixelBoxStyles:function(){return e(),o},pixelPosition:function(){return e(),n},reliableMarginLeft:function(){return e(),s},scrollboxSize:function(){return e(),i},reliableTrDimensions:function(){var e,t,n,r;return null==a&&(e=E.createElement("table"),t=E.createElement("tr"),n=E.createElement("div"),e.style.cssText="position:absolute;left:-11111px;border-collapse:separate",t.style.cssText="border:1px solid",t.style.height="1px",n.style.height="9px",n.style.display="block",re.appendChild(e).appendChild(t).appendChild(n),r=C.getComputedStyle(t),a=parseInt(r.height,10)+parseInt(r.borderTopWidth,10)+parseInt(r.borderBottomWidth,10)===t.offsetHeight,re.removeChild(e)),a}}))}();var Be=["Webkit","Moz","ms"],$e=E.createElement("div").style,_e={};function ze(e){var t=S.cssProps[e]||_e[e];return t||(e in $e?e:_e[e]=function(e){var t=e[0].toUpperCase()+e.slice(1),n=Be.length;while(n--)if((e=Be[n]+t)in $e)return e}(e)||e)}var Ue=/^(none|table(?!-c[ea]).+)/,Xe=/^--/,Ve={position:"absolute",visibility:"hidden",display:"block"},Ge={letterSpacing:"0",fontWeight:"400"};function Ye(e,t,n){var r=te.exec(t);return r?Math.max(0,r[2]-(n||0))+(r[3]||"px"):t}function Qe(e,t,n,r,i,o){var a="width"===t?1:0,s=0,u=0;if(n===(r?"border":"content"))return 0;for(;a<4;a+=2)"margin"===n&&(u+=S.css(e,n+ne[a],!0,i)),r?("content"===n&&(u-=S.css(e,"padding"+ne[a],!0,i)),"margin"!==n&&(u-=S.css(e,"border"+ne[a]+"Width",!0,i))):(u+=S.css(e,"padding"+ne[a],!0,i),"padding"!==n?u+=S.css(e,"border"+ne[a]+"Width",!0,i):s+=S.css(e,"border"+ne[a]+"Width",!0,i));return!r&&0<=o&&(u+=Math.max(0,Math.ceil(e["offset"+t[0].toUpperCase()+t.slice(1)]-o-u-s-.5))||0),u}function Je(e,t,n){var r=Re(e),i=(!y.boxSizingReliable()||n)&&"border-box"===S.css(e,"boxSizing",!1,r),o=i,a=We(e,t,r),s="offset"+t[0].toUpperCase()+t.slice(1);if(Pe.test(a)){if(!n)return a;a="auto"}return(!y.boxSizingReliable()&&i||!y.reliableTrDimensions()&&A(e,"tr")||"auto"===a||!parseFloat(a)&&"inline"===S.css(e,"display",!1,r))&&e.getClientRects().length&&(i="border-box"===S.css(e,"boxSizing",!1,r),(o=s in e)&&(a=e[s])),(a=parseFloat(a)||0)+Qe(e,t,n||(i?"border":"content"),o,r,a)+"px"}function Ke(e,t,n,r,i){return new Ke.prototype.init(e,t,n,r,i)}S.extend({cssHooks:{opacity:{get:function(e,t){if(t){var n=We(e,"opacity");return""===n?"1":n}}}},cssNumber:{animationIterationCount:!0,columnCount:!0,fillOpacity:!0,flexGrow:!0,flexShrink:!0,fontWeight:!0,gridArea:!0,gridColumn:!0,gridColumnEnd:!0,gridColumnStart:!0,gridRow:!0,gridRowEnd:!0,gridRowStart:!0,lineHeight:!0,opacity:!0,order:!0,orphans:!0,widows:!0,zIndex:!0,zoom:!0},cssProps:{},style:function(e,t,n,r){if(e&&3!==e.nodeType&&8!==e.nodeType&&e.style){var i,o,a,s=X(t),u=Xe.test(t),l=e.style;if(u||(t=ze(s)),a=S.cssHooks[t]||S.cssHooks[s],void 0===n)return a&&"get"in a&&void 0!==(i=a.get(e,!1,r))?i:l[t];"string"===(o=typeof n)&&(i=te.exec(n))&&i[1]&&(n=se(e,t,i),o="number"),null!=n&&n==n&&("number"!==o||u||(n+=i&&i[3]||(S.cssNumber[s]?"":"px")),y.clearCloneStyle||""!==n||0!==t.indexOf("background")||(l[t]="inherit"),a&&"set"in a&&void 0===(n=a.set(e,n,r))||(u?l.setProperty(t,n):l[t]=n))}},css:function(e,t,n,r){var i,o,a,s=X(t);return Xe.test(t)||(t=ze(s)),(a=S.cssHooks[t]||S.cssHooks[s])&&"get"in a&&(i=a.get(e,!0,n)),void 0===i&&(i=We(e,t,r)),"normal"===i&&t in Ge&&(i=Ge[t]),""===n||n?(o=parseFloat(i),!0===n||isFinite(o)?o||0:i):i}}),S.each(["height","width"],function(e,u){S.cssHooks[u]={get:function(e,t,n){if(t)return!Ue.test(S.css(e,"display"))||e.getClientRects().length&&e.getBoundingClientRect().width?Je(e,u,n):Me(e,Ve,function(){return Je(e,u,n)})},set:function(e,t,n){var r,i=Re(e),o=!y.scrollboxSize()&&"absolute"===i.position,a=(o||n)&&"border-box"===S.css(e,"boxSizing",!1,i),s=n?Qe(e,u,n,a,i):0;return a&&o&&(s-=Math.ceil(e["offset"+u[0].toUpperCase()+u.slice(1)]-parseFloat(i[u])-Qe(e,u,"border",!1,i)-.5)),s&&(r=te.exec(t))&&"px"!==(r[3]||"px")&&(e.style[u]=t,t=S.css(e,u)),Ye(0,t,s)}}}),S.cssHooks.marginLeft=Fe(y.reliableMarginLeft,function(e,t){if(t)return(parseFloat(We(e,"marginLeft"))||e.getBoundingClientRect().left-Me(e,{marginLeft:0},function(){return e.getBoundingClientRect().left}))+"px"}),S.each({margin:"",padding:"",border:"Width"},function(i,o){S.cssHooks[i+o]={expand:function(e){for(var t=0,n={},r="string"==typeof e?e.split(" "):[e];t<4;t++)n[i+ne[t]+o]=r[t]||r[t-2]||r[0];return n}},"margin"!==i&&(S.cssHooks[i+o].set=Ye)}),S.fn.extend({css:function(e,t){return $(this,function(e,t,n){var r,i,o={},a=0;if(Array.isArray(t)){for(r=Re(e),i=t.length;a<i;a++)o[t[a]]=S.css(e,t[a],!1,r);return o}return void 0!==n?S.style(e,t,n):S.css(e,t)},e,t,1<arguments.length)}}),((S.Tween=Ke).prototype={constructor:Ke,init:function(e,t,n,r,i,o){this.elem=e,this.prop=n,this.easing=i||S.easing._default,this.options=t,this.start=this.now=this.cur(),this.end=r,this.unit=o||(S.cssNumber[n]?"":"px")},cur:function(){var e=Ke.propHooks[this.prop];return e&&e.get?e.get(this):Ke.propHooks._default.get(this)},run:function(e){var t,n=Ke.propHooks[this.prop];return this.options.duration?this.pos=t=S.easing[this.easing](e,this.options.duration*e,0,1,this.options.duration):this.pos=t=e,this.now=(this.end-this.start)*t+this.start,this.options.step&&this.options.step.call(this.elem,this.now,this),n&&n.set?n.set(this):Ke.propHooks._default.set(this),this}}).init.prototype=Ke.prototype,(Ke.propHooks={_default:{get:function(e){var t;return 1!==e.elem.nodeType||null!=e.elem[e.prop]&&null==e.elem.style[e.prop]?e.elem[e.prop]:(t=S.css(e.elem,e.prop,""))&&"auto"!==t?t:0},set:function(e){S.fx.step[e.prop]?S.fx.step[e.prop](e):1!==e.elem.nodeType||!S.cssHooks[e.prop]&&null==e.elem.style[ze(e.prop)]?e.elem[e.prop]=e.now:S.style(e.elem,e.prop,e.now+e.unit)}}}).scrollTop=Ke.propHooks.scrollLeft={set:function(e){e.elem.nodeType&&e.elem.parentNode&&(e.elem[e.prop]=e.now)}},S.easing={linear:function(e){return e},swing:function(e){return.5-Math.cos(e*Math.PI)/2},_default:"swing"},S.fx=Ke.prototype.init,S.fx.step={};var Ze,et,tt,nt,rt=/^(?:toggle|show|hide)$/,it=/queueHooks$/;function ot(){et&&(!1===E.hidden&&C.requestAnimationFrame?C.requestAnimationFrame(ot):C.setTimeout(ot,S.fx.interval),S.fx.tick())}function at(){return C.setTimeout(function(){Ze=void 0}),Ze=Date.now()}function st(e,t){var n,r=0,i={height:e};for(t=t?1:0;r<4;r+=2-t)i["margin"+(n=ne[r])]=i["padding"+n]=e;return t&&(i.opacity=i.width=e),i}function ut(e,t,n){for(var r,i=(lt.tweeners[t]||[]).concat(lt.tweeners["*"]),o=0,a=i.length;o<a;o++)if(r=i[o].call(n,t,e))return r}function lt(o,e,t){var n,a,r=0,i=lt.prefilters.length,s=S.Deferred().always(function(){delete u.elem}),u=function(){if(a)return!1;for(var e=Ze||at(),t=Math.max(0,l.startTime+l.duration-e),n=1-(t/l.duration||0),r=0,i=l.tweens.length;r<i;r++)l.tweens[r].run(n);return s.notifyWith(o,[l,n,t]),n<1&&i?t:(i||s.notifyWith(o,[l,1,0]),s.resolveWith(o,[l]),!1)},l=s.promise({elem:o,props:S.extend({},e),opts:S.extend(!0,{specialEasing:{},easing:S.easing._default},t),originalProperties:e,originalOptions:t,startTime:Ze||at(),duration:t.duration,tweens:[],createTween:function(e,t){var n=S.Tween(o,l.opts,e,t,l.opts.specialEasing[e]||l.opts.easing);return l.tweens.push(n),n},stop:function(e){var t=0,n=e?l.tweens.length:0;if(a)return this;for(a=!0;t<n;t++)l.tweens[t].run(1);return e?(s.notifyWith(o,[l,1,0]),s.resolveWith(o,[l,e])):s.rejectWith(o,[l,e]),this}}),c=l.props;for(!function(e,t){var n,r,i,o,a;for(n in e)if(i=t[r=X(n)],o=e[n],Array.isArray(o)&&(i=o[1],o=e[n]=o[0]),n!==r&&(e[r]=o,delete e[n]),(a=S.cssHooks[r])&&"expand"in a)for(n in o=a.expand(o),delete e[r],o)n in e||(e[n]=o[n],t[n]=i);else t[r]=i}(c,l.opts.specialEasing);r<i;r++)if(n=lt.prefilters[r].call(l,o,c,l.opts))return m(n.stop)&&(S._queueHooks(l.elem,l.opts.queue).stop=n.stop.bind(n)),n;return S.map(c,ut,l),m(l.opts.start)&&l.opts.start.call(o,l),l.progress(l.opts.progress).done(l.opts.done,l.opts.complete).fail(l.opts.fail).always(l.opts.always),S.fx.timer(S.extend(u,{elem:o,anim:l,queue:l.opts.queue})),l}S.Animation=S.extend(lt,{tweeners:{"*":[function(e,t){var n=this.createTween(e,t);return se(n.elem,e,te.exec(t),n),n}]},tweener:function(e,t){m(e)?(t=e,e=["*"]):e=e.match(P);for(var n,r=0,i=e.length;r<i;r++)n=e[r],lt.tweeners[n]=lt.tweeners[n]||[],lt.tweeners[n].unshift(t)},prefilters:[function(e,t,n){var r,i,o,a,s,u,l,c,f="width"in t||"height"in t,p=this,d={},h=e.style,g=e.nodeType&&ae(e),v=Y.get(e,"fxshow");for(r in n.queue||(null==(a=S._queueHooks(e,"fx")).unqueued&&(a.unqueued=0,s=a.empty.fire,a.empty.fire=function(){a.unqueued||s()}),a.unqueued++,p.always(function(){p.always(function(){a.unqueued--,S.queue(e,"fx").length||a.empty.fire()})})),t)if(i=t[r],rt.test(i)){if(delete t[r],o=o||"toggle"===i,i===(g?"hide":"show")){if("show"!==i||!v||void 0===v[r])continue;g=!0}d[r]=v&&v[r]||S.style(e,r)}if((u=!S.isEmptyObject(t))||!S.isEmptyObject(d))for(r in f&&1===e.nodeType&&(n.overflow=[h.overflow,h.overflowX,h.overflowY],null==(l=v&&v.display)&&(l=Y.get(e,"display")),"none"===(c=S.css(e,"display"))&&(l?c=l:(le([e],!0),l=e.style.display||l,c=S.css(e,"display"),le([e]))),("inline"===c||"inline-block"===c&&null!=l)&&"none"===S.css(e,"float")&&(u||(p.done(function(){h.display=l}),null==l&&(c=h.display,l="none"===c?"":c)),h.display="inline-block")),n.overflow&&(h.overflow="hidden",p.always(function(){h.overflow=n.overflow[0],h.overflowX=n.overflow[1],h.overflowY=n.overflow[2]})),u=!1,d)u||(v?"hidden"in v&&(g=v.hidden):v=Y.access(e,"fxshow",{display:l}),o&&(v.hidden=!g),g&&le([e],!0),p.done(function(){for(r in g||le([e]),Y.remove(e,"fxshow"),d)S.style(e,r,d[r])})),u=ut(g?v[r]:0,r,p),r in v||(v[r]=u.start,g&&(u.end=u.start,u.start=0))}],prefilter:function(e,t){t?lt.prefilters.unshift(e):lt.prefilters.push(e)}}),S.speed=function(e,t,n){var r=e&&"object"==typeof e?S.extend({},e):{complete:n||!n&&t||m(e)&&e,duration:e,easing:n&&t||t&&!m(t)&&t};return S.fx.off?r.duration=0:"number"!=typeof r.duration&&(r.duration in S.fx.speeds?r.duration=S.fx.speeds[r.duration]:r.duration=S.fx.speeds._default),null!=r.queue&&!0!==r.queue||(r.queue="fx"),r.old=r.complete,r.complete=function(){m(r.old)&&r.old.call(this),r.queue&&S.dequeue(this,r.queue)},r},S.fn.extend({fadeTo:function(e,t,n,r){return this.filter(ae).css("opacity",0).show().end().animate({opacity:t},e,n,r)},animate:function(t,e,n,r){var i=S.isEmptyObject(t),o=S.speed(e,n,r),a=function(){var e=lt(this,S.extend({},t),o);(i||Y.get(this,"finish"))&&e.stop(!0)};return a.finish=a,i||!1===o.queue?this.each(a):this.queue(o.queue,a)},stop:function(i,e,o){var a=function(e){var t=e.stop;delete e.stop,t(o)};return"string"!=typeof i&&(o=e,e=i,i=void 0),e&&this.queue(i||"fx",[]),this.each(function(){var e=!0,t=null!=i&&i+"queueHooks",n=S.timers,r=Y.get(this);if(t)r[t]&&r[t].stop&&a(r[t]);else for(t in r)r[t]&&r[t].stop&&it.test(t)&&a(r[t]);for(t=n.length;t--;)n[t].elem!==this||null!=i&&n[t].queue!==i||(n[t].anim.stop(o),e=!1,n.splice(t,1));!e&&o||S.dequeue(this,i)})},finish:function(a){return!1!==a&&(a=a||"fx"),this.each(function(){var e,t=Y.get(this),n=t[a+"queue"],r=t[a+"queueHooks"],i=S.timers,o=n?n.length:0;for(t.finish=!0,S.queue(this,a,[]),r&&r.stop&&r.stop.call(this,!0),e=i.length;e--;)i[e].elem===this&&i[e].queue===a&&(i[e].anim.stop(!0),i.splice(e,1));for(e=0;e<o;e++)n[e]&&n[e].finish&&n[e].finish.call(this);delete t.finish})}}),S.each(["toggle","show","hide"],function(e,r){var i=S.fn[r];S.fn[r]=function(e,t,n){return null==e||"boolean"==typeof e?i.apply(this,arguments):this.animate(st(r,!0),e,t,n)}}),S.each({slideDown:st("show"),slideUp:st("hide"),slideToggle:st("toggle"),fadeIn:{opacity:"show"},fadeOut:{opacity:"hide"},fadeToggle:{opacity:"toggle"}},function(e,r){S.fn[e]=function(e,t,n){return this.animate(r,e,t,n)}}),S.timers=[],S.fx.tick=function(){var e,t=0,n=S.timers;for(Ze=Date.now();t<n.length;t++)(e=n[t])()||n[t]!==e||n.splice(t--,1);n.length||S.fx.stop(),Ze=void 0},S.fx.timer=function(e){S.timers.push(e),S.fx.start()},S.fx.interval=13,S.fx.start=function(){et||(et=!0,ot())},S.fx.stop=function(){et=null},S.fx.speeds={slow:600,fast:200,_default:400},S.fn.delay=function(r,e){return r=S.fx&&S.fx.speeds[r]||r,e=e||"fx",this.queue(e,function(e,t){var n=C.setTimeout(e,r);t.stop=function(){C.clearTimeout(n)}})},tt=E.createElement("input"),nt=E.createElement("select").appendChild(E.createElement("option")),tt.type="checkbox",y.checkOn=""!==tt.value,y.optSelected=nt.selected,(tt=E.createElement("input")).value="t",tt.type="radio",y.radioValue="t"===tt.value;var ct,ft=S.expr.attrHandle;S.fn.extend({attr:function(e,t){return $(this,S.attr,e,t,1<arguments.length)},removeAttr:function(e){return this.each(function(){S.removeAttr(this,e)})}}),S.extend({attr:function(e,t,n){var r,i,o=e.nodeType;if(3!==o&&8!==o&&2!==o)return"undefined"==typeof e.getAttribute?S.prop(e,t,n):(1===o&&S.isXMLDoc(e)||(i=S.attrHooks[t.toLowerCase()]||(S.expr.match.bool.test(t)?ct:void 0)),void 0!==n?null===n?void S.removeAttr(e,t):i&&"set"in i&&void 0!==(r=i.set(e,n,t))?r:(e.setAttribute(t,n+""),n):i&&"get"in i&&null!==(r=i.get(e,t))?r:null==(r=S.find.attr(e,t))?void 0:r)},attrHooks:{type:{set:function(e,t){if(!y.radioValue&&"radio"===t&&A(e,"input")){var n=e.value;return e.setAttribute("type",t),n&&(e.value=n),t}}}},removeAttr:function(e,t){var n,r=0,i=t&&t.match(P);if(i&&1===e.nodeType)while(n=i[r++])e.removeAttribute(n)}}),ct={set:function(e,t,n){return!1===t?S.removeAttr(e,n):e.setAttribute(n,n),n}},S.each(S.expr.match.bool.source.match(/\w+/g),function(e,t){var a=ft[t]||S.find.attr;ft[t]=function(e,t,n){var r,i,o=t.toLowerCase();return n||(i=ft[o],ft[o]=r,r=null!=a(e,t,n)?o:null,ft[o]=i),r}});var pt=/^(?:input|select|textarea|button)$/i,dt=/^(?:a|area)$/i;function ht(e){return(e.match(P)||[]).join(" ")}function gt(e){return e.getAttribute&&e.getAttribute("class")||""}function vt(e){return Array.isArray(e)?e:"string"==typeof e&&e.match(P)||[]}S.fn.extend({prop:function(e,t){return $(this,S.prop,e,t,1<arguments.length)},removeProp:function(e){return this.each(function(){delete this[S.propFix[e]||e]})}}),S.extend({prop:function(e,t,n){var r,i,o=e.nodeType;if(3!==o&&8!==o&&2!==o)return 1===o&&S.isXMLDoc(e)||(t=S.propFix[t]||t,i=S.propHooks[t]),void 0!==n?i&&"set"in i&&void 0!==(r=i.set(e,n,t))?r:e[t]=n:i&&"get"in i&&null!==(r=i.get(e,t))?r:e[t]},propHooks:{tabIndex:{get:function(e){var t=S.find.attr(e,"tabindex");return t?parseInt(t,10):pt.test(e.nodeName)||dt.test(e.nodeName)&&e.href?0:-1}}},propFix:{"for":"htmlFor","class":"className"}}),y.optSelected||(S.propHooks.selected={get:function(e){var t=e.parentNode;return t&&t.parentNode&&t.parentNode.selectedIndex,null},set:function(e){var t=e.parentNode;t&&(t.selectedIndex,t.parentNode&&t.parentNode.selectedIndex)}}),S.each(["tabIndex","readOnly","maxLength","cellSpacing","cellPadding","rowSpan","colSpan","useMap","frameBorder","contentEditable"],function(){S.propFix[this.toLowerCase()]=this}),S.fn.extend({addClass:function(t){var e,n,r,i,o,a,s,u=0;if(m(t))return this.each(function(e){S(this).addClass(t.call(this,e,gt(this)))});if((e=vt(t)).length)while(n=this[u++])if(i=gt(n),r=1===n.nodeType&&" "+ht(i)+" "){a=0;while(o=e[a++])r.indexOf(" "+o+" ")<0&&(r+=o+" ");i!==(s=ht(r))&&n.setAttribute("class",s)}return this},removeClass:function(t){var e,n,r,i,o,a,s,u=0;if(m(t))return this.each(function(e){S(this).removeClass(t.call(this,e,gt(this)))});if(!arguments.length)return this.attr("class","");if((e=vt(t)).length)while(n=this[u++])if(i=gt(n),r=1===n.nodeType&&" "+ht(i)+" "){a=0;while(o=e[a++])while(-1<r.indexOf(" "+o+" "))r=r.replace(" "+o+" "," ");i!==(s=ht(r))&&n.setAttribute("class",s)}return this},toggleClass:function(i,t){var o=typeof i,a="string"===o||Array.isArray(i);return"boolean"==typeof t&&a?t?this.addClass(i):this.removeClass(i):m(i)?this.each(function(e){S(this).toggleClass(i.call(this,e,gt(this),t),t)}):this.each(function(){var e,t,n,r;if(a){t=0,n=S(this),r=vt(i);while(e=r[t++])n.hasClass(e)?n.removeClass(e):n.addClass(e)}else void 0!==i&&"boolean"!==o||((e=gt(this))&&Y.set(this,"__className__",e),this.setAttribute&&this.setAttribute("class",e||!1===i?"":Y.get(this,"__className__")||""))})},hasClass:function(e){var t,n,r=0;t=" "+e+" ";while(n=this[r++])if(1===n.nodeType&&-1<(" "+ht(gt(n))+" ").indexOf(t))return!0;return!1}});var yt=/\r/g;S.fn.extend({val:function(n){var r,e,i,t=this[0];return arguments.length?(i=m(n),this.each(function(e){var t;1===this.nodeType&&(null==(t=i?n.call(this,e,S(this).val()):n)?t="":"number"==typeof t?t+="":Array.isArray(t)&&(t=S.map(t,function(e){return null==e?"":e+""})),(r=S.valHooks[this.type]||S.valHooks[this.nodeName.toLowerCase()])&&"set"in r&&void 0!==r.set(this,t,"value")||(this.value=t))})):t?(r=S.valHooks[t.type]||S.valHooks[t.nodeName.toLowerCase()])&&"get"in r&&void 0!==(e=r.get(t,"value"))?e:"string"==typeof(e=t.value)?e.replace(yt,""):null==e?"":e:void 0}}),S.extend({valHooks:{option:{get:function(e){var t=S.find.attr(e,"value");return null!=t?t:ht(S.text(e))}},select:{get:function(e){var t,n,r,i=e.options,o=e.selectedIndex,a="select-one"===e.type,s=a?null:[],u=a?o+1:i.length;for(r=o<0?u:a?o:0;r<u;r++)if(((n=i[r]).selected||r===o)&&!n.disabled&&(!n.parentNode.disabled||!A(n.parentNode,"optgroup"))){if(t=S(n).val(),a)return t;s.push(t)}return s},set:function(e,t){var n,r,i=e.options,o=S.makeArray(t),a=i.length;while(a--)((r=i[a]).selected=-1<S.inArray(S.valHooks.option.get(r),o))&&(n=!0);return n||(e.selectedIndex=-1),o}}}}),S.each(["radio","checkbox"],function(){S.valHooks[this]={set:function(e,t){if(Array.isArray(t))return e.checked=-1<S.inArray(S(e).val(),t)}},y.checkOn||(S.valHooks[this].get=function(e){return null===e.getAttribute("value")?"on":e.value})}),y.focusin="onfocusin"in C;var mt=/^(?:focusinfocus|focusoutblur)$/,xt=function(e){e.stopPropagation()};S.extend(S.event,{trigger:function(e,t,n,r){var i,o,a,s,u,l,c,f,p=[n||E],d=v.call(e,"type")?e.type:e,h=v.call(e,"namespace")?e.namespace.split("."):[];if(o=f=a=n=n||E,3!==n.nodeType&&8!==n.nodeType&&!mt.test(d+S.event.triggered)&&(-1<d.indexOf(".")&&(d=(h=d.split(".")).shift(),h.sort()),u=d.indexOf(":")<0&&"on"+d,(e=e[S.expando]?e:new S.Event(d,"object"==typeof e&&e)).isTrigger=r?2:3,e.namespace=h.join("."),e.rnamespace=e.namespace?new RegExp("(^|\\.)"+h.join("\\.(?:.*\\.|)")+"(\\.|$)"):null,e.result=void 0,e.target||(e.target=n),t=null==t?[e]:S.makeArray(t,[e]),c=S.event.special[d]||{},r||!c.trigger||!1!==c.trigger.apply(n,t))){if(!r&&!c.noBubble&&!x(n)){for(s=c.delegateType||d,mt.test(s+d)||(o=o.parentNode);o;o=o.parentNode)p.push(o),a=o;a===(n.ownerDocument||E)&&p.push(a.defaultView||a.parentWindow||C)}i=0;while((o=p[i++])&&!e.isPropagationStopped())f=o,e.type=1<i?s:c.bindType||d,(l=(Y.get(o,"events")||Object.create(null))[e.type]&&Y.get(o,"handle"))&&l.apply(o,t),(l=u&&o[u])&&l.apply&&V(o)&&(e.result=l.apply(o,t),!1===e.result&&e.preventDefault());return e.type=d,r||e.isDefaultPrevented()||c._default&&!1!==c._default.apply(p.pop(),t)||!V(n)||u&&m(n[d])&&!x(n)&&((a=n[u])&&(n[u]=null),S.event.triggered=d,e.isPropagationStopped()&&f.addEventListener(d,xt),n[d](),e.isPropagationStopped()&&f.removeEventListener(d,xt),S.event.triggered=void 0,a&&(n[u]=a)),e.result}},simulate:function(e,t,n){var r=S.extend(new S.Event,n,{type:e,isSimulated:!0});S.event.trigger(r,null,t)}}),S.fn.extend({trigger:function(e,t){return this.each(function(){S.event.trigger(e,t,this)})},triggerHandler:function(e,t){var n=this[0];if(n)return S.event.trigger(e,t,n,!0)}}),y.focusin||S.each({focus:"focusin",blur:"focusout"},function(n,r){var i=function(e){S.event.simulate(r,e.target,S.event.fix(e))};S.event.special[r]={setup:function(){var e=this.ownerDocument||this.document||this,t=Y.access(e,r);t||e.addEventListener(n,i,!0),Y.access(e,r,(t||0)+1)},teardown:function(){var e=this.ownerDocument||this.document||this,t=Y.access(e,r)-1;t?Y.access(e,r,t):(e.removeEventListener(n,i,!0),Y.remove(e,r))}}});var bt=C.location,wt={guid:Date.now()},Tt=/\?/;S.parseXML=function(e){var t,n;if(!e||"string"!=typeof e)return null;try{t=(new C.DOMParser).parseFromString(e,"text/xml")}catch(e){}return n=t&&t.getElementsByTagName("parsererror")[0],t&&!n||S.error("Invalid XML: "+(n?S.map(n.childNodes,function(e){return e.textContent}).join("\n"):e)),t};var Ct=/\[\]$/,Et=/\r?\n/g,St=/^(?:submit|button|image|reset|file)$/i,kt=/^(?:input|select|textarea|keygen)/i;function At(n,e,r,i){var t;if(Array.isArray(e))S.each(e,function(e,t){r||Ct.test(n)?i(n,t):At(n+"["+("object"==typeof t&&null!=t?e:"")+"]",t,r,i)});else if(r||"object"!==w(e))i(n,e);else for(t in e)At(n+"["+t+"]",e[t],r,i)}S.param=function(e,t){var n,r=[],i=function(e,t){var n=m(t)?t():t;r[r.length]=encodeURIComponent(e)+"="+encodeURIComponent(null==n?"":n)};if(null==e)return"";if(Array.isArray(e)||e.jquery&&!S.isPlainObject(e))S.each(e,function(){i(this.name,this.value)});else for(n in e)At(n,e[n],t,i);return r.join("&")},S.fn.extend({serialize:function(){return S.param(this.serializeArray())},serializeArray:function(){return this.map(function(){var e=S.prop(this,"elements");return e?S.makeArray(e):this}).filter(function(){var e=this.type;return this.name&&!S(this).is(":disabled")&&kt.test(this.nodeName)&&!St.test(e)&&(this.checked||!pe.test(e))}).map(function(e,t){var n=S(this).val();return null==n?null:Array.isArray(n)?S.map(n,function(e){return{name:t.name,value:e.replace(Et,"\r\n")}}):{name:t.name,value:n.replace(Et,"\r\n")}}).get()}});var Nt=/%20/g,jt=/#.*$/,Dt=/([?&])_=[^&]*/,qt=/^(.*?):[ \t]*([^\r\n]*)$/gm,Lt=/^(?:GET|HEAD)$/,Ht=/^\/\//,Ot={},Pt={},Rt="*/".concat("*"),Mt=E.createElement("a");function It(o){return function(e,t){"string"!=typeof e&&(t=e,e="*");var n,r=0,i=e.toLowerCase().match(P)||[];if(m(t))while(n=i[r++])"+"===n[0]?(n=n.slice(1)||"*",(o[n]=o[n]||[]).unshift(t)):(o[n]=o[n]||[]).push(t)}}function Wt(t,i,o,a){var s={},u=t===Pt;function l(e){var r;return s[e]=!0,S.each(t[e]||[],function(e,t){var n=t(i,o,a);return"string"!=typeof n||u||s[n]?u?!(r=n):void 0:(i.dataTypes.unshift(n),l(n),!1)}),r}return l(i.dataTypes[0])||!s["*"]&&l("*")}function Ft(e,t){var n,r,i=S.ajaxSettings.flatOptions||{};for(n in t)void 0!==t[n]&&((i[n]?e:r||(r={}))[n]=t[n]);return r&&S.extend(!0,e,r),e}Mt.href=bt.href,S.extend({active:0,lastModified:{},etag:{},ajaxSettings:{url:bt.href,type:"GET",isLocal:/^(?:about|app|app-storage|.+-extension|file|res|widget):$/.test(bt.protocol),global:!0,processData:!0,async:!0,contentType:"application/x-www-form-urlencoded; charset=UTF-8",accepts:{"*":Rt,text:"text/plain",html:"text/html",xml:"application/xml, text/xml",json:"application/json, text/javascript"},contents:{xml:/\bxml\b/,html:/\bhtml/,json:/\bjson\b/},responseFields:{xml:"responseXML",text:"responseText",json:"responseJSON"},converters:{"* text":String,"text html":!0,"text json":JSON.parse,"text xml":S.parseXML},flatOptions:{url:!0,context:!0}},ajaxSetup:function(e,t){return t?Ft(Ft(e,S.ajaxSettings),t):Ft(S.ajaxSettings,e)},ajaxPrefilter:It(Ot),ajaxTransport:It(Pt),ajax:function(e,t){"object"==typeof e&&(t=e,e=void 0),t=t||{};var c,f,p,n,d,r,h,g,i,o,v=S.ajaxSetup({},t),y=v.context||v,m=v.context&&(y.nodeType||y.jquery)?S(y):S.event,x=S.Deferred(),b=S.Callbacks("once memory"),w=v.statusCode||{},a={},s={},u="canceled",T={readyState:0,getResponseHeader:function(e){var t;if(h){if(!n){n={};while(t=qt.exec(p))n[t[1].toLowerCase()+" "]=(n[t[1].toLowerCase()+" "]||[]).concat(t[2])}t=n[e.toLowerCase()+" "]}return null==t?null:t.join(", ")},getAllResponseHeaders:function(){return h?p:null},setRequestHeader:function(e,t){return null==h&&(e=s[e.toLowerCase()]=s[e.toLowerCase()]||e,a[e]=t),this},overrideMimeType:function(e){return null==h&&(v.mimeType=e),this},statusCode:function(e){var t;if(e)if(h)T.always(e[T.status]);else for(t in e)w[t]=[w[t],e[t]];return this},abort:function(e){var t=e||u;return c&&c.abort(t),l(0,t),this}};if(x.promise(T),v.url=((e||v.url||bt.href)+"").replace(Ht,bt.protocol+"//"),v.type=t.method||t.type||v.method||v.type,v.dataTypes=(v.dataType||"*").toLowerCase().match(P)||[""],null==v.crossDomain){r=E.createElement("a");try{r.href=v.url,r.href=r.href,v.crossDomain=Mt.protocol+"//"+Mt.host!=r.protocol+"//"+r.host}catch(e){v.crossDomain=!0}}if(v.data&&v.processData&&"string"!=typeof v.data&&(v.data=S.param(v.data,v.traditional)),Wt(Ot,v,t,T),h)return T;for(i in(g=S.event&&v.global)&&0==S.active++&&S.event.trigger("ajaxStart"),v.type=v.type.toUpperCase(),v.hasContent=!Lt.test(v.type),f=v.url.replace(jt,""),v.hasContent?v.data&&v.processData&&0===(v.contentType||"").indexOf("application/x-www-form-urlencoded")&&(v.data=v.data.replace(Nt,"+")):(o=v.url.slice(f.length),v.data&&(v.processData||"string"==typeof v.data)&&(f+=(Tt.test(f)?"&":"?")+v.data,delete v.data),!1===v.cache&&(f=f.replace(Dt,"$1"),o=(Tt.test(f)?"&":"?")+"_="+wt.guid+++o),v.url=f+o),v.ifModified&&(S.lastModified[f]&&T.setRequestHeader("If-Modified-Since",S.lastModified[f]),S.etag[f]&&T.setRequestHeader("If-None-Match",S.etag[f])),(v.data&&v.hasContent&&!1!==v.contentType||t.contentType)&&T.setRequestHeader("Content-Type",v.contentType),T.setRequestHeader("Accept",v.dataTypes[0]&&v.accepts[v.dataTypes[0]]?v.accepts[v.dataTypes[0]]+("*"!==v.dataTypes[0]?", "+Rt+"; q=0.01":""):v.accepts["*"]),v.headers)T.setRequestHeader(i,v.headers[i]);if(v.beforeSend&&(!1===v.beforeSend.call(y,T,v)||h))return T.abort();if(u="abort",b.add(v.complete),T.done(v.success),T.fail(v.error),c=Wt(Pt,v,t,T)){if(T.readyState=1,g&&m.trigger("ajaxSend",[T,v]),h)return T;v.async&&0<v.timeout&&(d=C.setTimeout(function(){T.abort("timeout")},v.timeout));try{h=!1,c.send(a,l)}catch(e){if(h)throw e;l(-1,e)}}else l(-1,"No Transport");function l(e,t,n,r){var i,o,a,s,u,l=t;h||(h=!0,d&&C.clearTimeout(d),c=void 0,p=r||"",T.readyState=0<e?4:0,i=200<=e&&e<300||304===e,n&&(s=function(e,t,n){var r,i,o,a,s=e.contents,u=e.dataTypes;while("*"===u[0])u.shift(),void 0===r&&(r=e.mimeType||t.getResponseHeader("Content-Type"));if(r)for(i in s)if(s[i]&&s[i].test(r)){u.unshift(i);break}if(u[0]in n)o=u[0];else{for(i in n){if(!u[0]||e.converters[i+" "+u[0]]){o=i;break}a||(a=i)}o=o||a}if(o)return o!==u[0]&&u.unshift(o),n[o]}(v,T,n)),!i&&-1<S.inArray("script",v.dataTypes)&&S.inArray("json",v.dataTypes)<0&&(v.converters["text script"]=function(){}),s=function(e,t,n,r){var i,o,a,s,u,l={},c=e.dataTypes.slice();if(c[1])for(a in e.converters)l[a.toLowerCase()]=e.converters[a];o=c.shift();while(o)if(e.responseFields[o]&&(n[e.responseFields[o]]=t),!u&&r&&e.dataFilter&&(t=e.dataFilter(t,e.dataType)),u=o,o=c.shift())if("*"===o)o=u;else if("*"!==u&&u!==o){if(!(a=l[u+" "+o]||l["* "+o]))for(i in l)if((s=i.split(" "))[1]===o&&(a=l[u+" "+s[0]]||l["* "+s[0]])){!0===a?a=l[i]:!0!==l[i]&&(o=s[0],c.unshift(s[1]));break}if(!0!==a)if(a&&e["throws"])t=a(t);else try{t=a(t)}catch(e){return{state:"parsererror",error:a?e:"No conversion from "+u+" to "+o}}}return{state:"success",data:t}}(v,s,T,i),i?(v.ifModified&&((u=T.getResponseHeader("Last-Modified"))&&(S.lastModified[f]=u),(u=T.getResponseHeader("etag"))&&(S.etag[f]=u)),204===e||"HEAD"===v.type?l="nocontent":304===e?l="notmodified":(l=s.state,o=s.data,i=!(a=s.error))):(a=l,!e&&l||(l="error",e<0&&(e=0))),T.status=e,T.statusText=(t||l)+"",i?x.resolveWith(y,[o,l,T]):x.rejectWith(y,[T,l,a]),T.statusCode(w),w=void 0,g&&m.trigger(i?"ajaxSuccess":"ajaxError",[T,v,i?o:a]),b.fireWith(y,[T,l]),g&&(m.trigger("ajaxComplete",[T,v]),--S.active||S.event.trigger("ajaxStop")))}return T},getJSON:function(e,t,n){return S.get(e,t,n,"json")},getScript:function(e,t){return S.get(e,void 0,t,"script")}}),S.each(["get","post"],function(e,i){S[i]=function(e,t,n,r){return m(t)&&(r=r||n,n=t,t=void 0),S.ajax(S.extend({url:e,type:i,dataType:r,data:t,success:n},S.isPlainObject(e)&&e))}}),S.ajaxPrefilter(function(e){var t;for(t in e.headers)"content-type"===t.toLowerCase()&&(e.contentType=e.headers[t]||"")}),S._evalUrl=function(e,t,n){return S.ajax({url:e,type:"GET",dataType:"script",cache:!0,async:!1,global:!1,converters:{"text script":function(){}},dataFilter:function(e){S.globalEval(e,t,n)}})},S.fn.extend({wrapAll:function(e){var t;return this[0]&&(m(e)&&(e=e.call(this[0])),t=S(e,this[0].ownerDocument).eq(0).clone(!0),this[0].parentNode&&t.insertBefore(this[0]),t.map(function(){var e=this;while(e.firstElementChild)e=e.firstElementChild;return e}).append(this)),this},wrapInner:function(n){return m(n)?this.each(function(e){S(this).wrapInner(n.call(this,e))}):this.each(function(){var e=S(this),t=e.contents();t.length?t.wrapAll(n):e.append(n)})},wrap:function(t){var n=m(t);return this.each(function(e){S(this).wrapAll(n?t.call(this,e):t)})},unwrap:function(e){return this.parent(e).not("body").each(function(){S(this).replaceWith(this.childNodes)}),this}}),S.expr.pseudos.hidden=function(e){return!S.expr.pseudos.visible(e)},S.expr.pseudos.visible=function(e){return!!(e.offsetWidth||e.offsetHeight||e.getClientRects().length)},S.ajaxSettings.xhr=function(){try{return new C.XMLHttpRequest}catch(e){}};var Bt={0:200,1223:204},$t=S.ajaxSettings.xhr();y.cors=!!$t&&"withCredentials"in $t,y.ajax=$t=!!$t,S.ajaxTransport(function(i){var o,a;if(y.cors||$t&&!i.crossDomain)return{send:function(e,t){var n,r=i.xhr();if(r.open(i.type,i.url,i.async,i.username,i.password),i.xhrFields)for(n in i.xhrFields)r[n]=i.xhrFields[n];for(n in i.mimeType&&r.overrideMimeType&&r.overrideMimeType(i.mimeType),i.crossDomain||e["X-Requested-With"]||(e["X-Requested-With"]="XMLHttpRequest"),e)r.setRequestHeader(n,e[n]);o=function(e){return function(){o&&(o=a=r.onload=r.onerror=r.onabort=r.ontimeout=r.onreadystatechange=null,"abort"===e?r.abort():"error"===e?"number"!=typeof r.status?t(0,"error"):t(r.status,r.statusText):t(Bt[r.status]||r.status,r.statusText,"text"!==(r.responseType||"text")||"string"!=typeof r.responseText?{binary:r.response}:{text:r.responseText},r.getAllResponseHeaders()))}},r.onload=o(),a=r.onerror=r.ontimeout=o("error"),void 0!==r.onabort?r.onabort=a:r.onreadystatechange=function(){4===r.readyState&&C.setTimeout(function(){o&&a()})},o=o("abort");try{r.send(i.hasContent&&i.data||null)}catch(e){if(o)throw e}},abort:function(){o&&o()}}}),S.ajaxPrefilter(function(e){e.crossDomain&&(e.contents.script=!1)}),S.ajaxSetup({accepts:{script:"text/javascript, application/javascript, application/ecmascript, application/x-ecmascript"},contents:{script:/\b(?:java|ecma)script\b/},converters:{"text script":function(e){return S.globalEval(e),e}}}),S.ajaxPrefilter("script",function(e){void 0===e.cache&&(e.cache=!1),e.crossDomain&&(e.type="GET")}),S.ajaxTransport("script",function(n){var r,i;if(n.crossDomain||n.scriptAttrs)return{send:function(e,t){r=S("<script>").attr(n.scriptAttrs||{}).prop({charset:n.scriptCharset,src:n.url}).on("load error",i=function(e){r.remove(),i=null,e&&t("error"===e.type?404:200,e.type)}),E.head.appendChild(r[0])},abort:function(){i&&i()}}});var _t,zt=[],Ut=/(=)\?(?=&|$)|\?\?/;S.ajaxSetup({jsonp:"callback",jsonpCallback:function(){var e=zt.pop()||S.expando+"_"+wt.guid++;return this[e]=!0,e}}),S.ajaxPrefilter("json jsonp",function(e,t,n){var r,i,o,a=!1!==e.jsonp&&(Ut.test(e.url)?"url":"string"==typeof e.data&&0===(e.contentType||"").indexOf("application/x-www-form-urlencoded")&&Ut.test(e.data)&&"data");if(a||"jsonp"===e.dataTypes[0])return r=e.jsonpCallback=m(e.jsonpCallback)?e.jsonpCallback():e.jsonpCallback,a?e[a]=e[a].replace(Ut,"$1"+r):!1!==e.jsonp&&(e.url+=(Tt.test(e.url)?"&":"?")+e.jsonp+"="+r),e.converters["script json"]=function(){return o||S.error(r+" was not called"),o[0]},e.dataTypes[0]="json",i=C[r],C[r]=function(){o=arguments},n.always(function(){void 0===i?S(C).removeProp(r):C[r]=i,e[r]&&(e.jsonpCallback=t.jsonpCallback,zt.push(r)),o&&m(i)&&i(o[0]),o=i=void 0}),"script"}),y.createHTMLDocument=((_t=E.implementation.createHTMLDocument("").body).innerHTML="<form></form><form></form>",2===_t.childNodes.length),S.parseHTML=function(e,t,n){return"string"!=typeof e?[]:("boolean"==typeof t&&(n=t,t=!1),t||(y.createHTMLDocument?((r=(t=E.implementation.createHTMLDocument("")).createElement("base")).href=E.location.href,t.head.appendChild(r)):t=E),o=!n&&[],(i=N.exec(e))?[t.createElement(i[1])]:(i=xe([e],t,o),o&&o.length&&S(o).remove(),S.merge([],i.childNodes)));var r,i,o},S.fn.load=function(e,t,n){var r,i,o,a=this,s=e.indexOf(" ");return-1<s&&(r=ht(e.slice(s)),e=e.slice(0,s)),m(t)?(n=t,t=void 0):t&&"object"==typeof t&&(i="POST"),0<a.length&&S.ajax({url:e,type:i||"GET",dataType:"html",data:t}).done(function(e){o=arguments,a.html(r?S("<div>").append(S.parseHTML(e)).find(r):e)}).always(n&&function(e,t){a.each(function(){n.apply(this,o||[e.responseText,t,e])})}),this},S.expr.pseudos.animated=function(t){return S.grep(S.timers,function(e){return t===e.elem}).length},S.offset={setOffset:function(e,t,n){var r,i,o,a,s,u,l=S.css(e,"position"),c=S(e),f={};"static"===l&&(e.style.position="relative"),s=c.offset(),o=S.css(e,"top"),u=S.css(e,"left"),("absolute"===l||"fixed"===l)&&-1<(o+u).indexOf("auto")?(a=(r=c.position()).top,i=r.left):(a=parseFloat(o)||0,i=parseFloat(u)||0),m(t)&&(t=t.call(e,n,S.extend({},s))),null!=t.top&&(f.top=t.top-s.top+a),null!=t.left&&(f.left=t.left-s.left+i),"using"in t?t.using.call(e,f):c.css(f)}},S.fn.extend({offset:function(t){if(arguments.length)return void 0===t?this:this.each(function(e){S.offset.setOffset(this,t,e)});var e,n,r=this[0];return r?r.getClientRects().length?(e=r.getBoundingClientRect(),n=r.ownerDocument.defaultView,{top:e.top+n.pageYOffset,left:e.left+n.pageXOffset}):{top:0,left:0}:void 0},position:function(){if(this[0]){var e,t,n,r=this[0],i={top:0,left:0};if("fixed"===S.css(r,"position"))t=r.getBoundingClientRect();else{t=this.offset(),n=r.ownerDocument,e=r.offsetParent||n.documentElement;while(e&&(e===n.body||e===n.documentElement)&&"static"===S.css(e,"position"))e=e.parentNode;e&&e!==r&&1===e.nodeType&&((i=S(e).offset()).top+=S.css(e,"borderTopWidth",!0),i.left+=S.css(e,"borderLeftWidth",!0))}return{top:t.top-i.top-S.css(r,"marginTop",!0),left:t.left-i.left-S.css(r,"marginLeft",!0)}}},offsetParent:function(){return this.map(function(){var e=this.offsetParent;while(e&&"static"===S.css(e,"position"))e=e.offsetParent;return e||re})}}),S.each({scrollLeft:"pageXOffset",scrollTop:"pageYOffset"},function(t,i){var o="pageYOffset"===i;S.fn[t]=function(e){return $(this,function(e,t,n){var r;if(x(e)?r=e:9===e.nodeType&&(r=e.defaultView),void 0===n)return r?r[i]:e[t];r?r.scrollTo(o?r.pageXOffset:n,o?n:r.pageYOffset):e[t]=n},t,e,arguments.length)}}),S.each(["top","left"],function(e,n){S.cssHooks[n]=Fe(y.pixelPosition,function(e,t){if(t)return t=We(e,n),Pe.test(t)?S(e).position()[n]+"px":t})}),S.each({Height:"height",Width:"width"},function(a,s){S.each({padding:"inner"+a,content:s,"":"outer"+a},function(r,o){S.fn[o]=function(e,t){var n=arguments.length&&(r||"boolean"!=typeof e),i=r||(!0===e||!0===t?"margin":"border");return $(this,function(e,t,n){var r;return x(e)?0===o.indexOf("outer")?e["inner"+a]:e.document.documentElement["client"+a]:9===e.nodeType?(r=e.documentElement,Math.max(e.body["scroll"+a],r["scroll"+a],e.body["offset"+a],r["offset"+a],r["client"+a])):void 0===n?S.css(e,t,i):S.style(e,t,n,i)},s,n?e:void 0,n)}})}),S.each(["ajaxStart","ajaxStop","ajaxComplete","ajaxError","ajaxSuccess","ajaxSend"],function(e,t){S.fn[t]=function(e){return this.on(t,e)}}),S.fn.extend({bind:function(e,t,n){return this.on(e,null,t,n)},unbind:function(e,t){return this.off(e,null,t)},delegate:function(e,t,n,r){return this.on(t,e,n,r)},undelegate:function(e,t,n){return 1===arguments.length?this.off(e,"**"):this.off(t,e||"**",n)},hover:function(e,t){return this.mouseenter(e).mouseleave(t||e)}}),S.each("blur focus focusin focusout resize scroll click dblclick mousedown mouseup mousemove mouseover mouseout mouseenter mouseleave change select submit keydown keypress keyup contextmenu".split(" "),function(e,n){S.fn[n]=function(e,t){return 0<arguments.length?this.on(n,null,e,t):this.trigger(n)}});var Xt=/^[\s\uFEFF\xA0]+|[\s\uFEFF\xA0]+$/g;S.proxy=function(e,t){var n,r,i;if("string"==typeof t&&(n=e[t],t=e,e=n),m(e))return r=s.call(arguments,2),(i=function(){return e.apply(t||this,r.concat(s.call(arguments)))}).guid=e.guid=e.guid||S.guid++,i},S.holdReady=function(e){e?S.readyWait++:S.ready(!0)},S.isArray=Array.isArray,S.parseJSON=JSON.parse,S.nodeName=A,S.isFunction=m,S.isWindow=x,S.camelCase=X,S.type=w,S.now=Date.now,S.isNumeric=function(e){var t=S.type(e);return("number"===t||"string"===t)&&!isNaN(e-parseFloat(e))},S.trim=function(e){return null==e?"":(e+"").replace(Xt,"")},"function"==typeof define&&define.amd&&define("jquery",[],function(){return S});var Vt=C.jQuery,Gt=C.$;return S.noConflict=function(e){return C.$===S&&(C.$=Gt),e&&C.jQuery===S&&(C.jQuery=Vt),S},"undefined"==typeof e&&(C.jQuery=C.$=S),S});
diff --git a/0.16/_static/js/modernizr.min.js b/0.16/_static/js/modernizr.min.js
new file mode 100644
index 00000000000..f65d4797478
--- /dev/null
+++ b/0.16/_static/js/modernizr.min.js
@@ -0,0 +1,4 @@
+/* Modernizr 2.6.2 (Custom Build) | MIT & BSD
+ * Build: http://modernizr.com/download/#-fontface-backgroundsize-borderimage-borderradius-boxshadow-flexbox-hsla-multiplebgs-opacity-rgba-textshadow-cssanimations-csscolumns-generatedcontent-cssgradients-cssreflections-csstransforms-csstransforms3d-csstransitions-applicationcache-canvas-canvastext-draganddrop-hashchange-history-audio-video-indexeddb-input-inputtypes-localstorage-postmessage-sessionstorage-websockets-websqldatabase-webworkers-geolocation-inlinesvg-smil-svg-svgclippaths-touch-webgl-shiv-mq-cssclasses-addtest-prefixed-teststyles-testprop-testallprops-hasevent-prefixes-domprefixes-load
+ */
+;window.Modernizr=function(a,b,c){function D(a){j.cssText=a}function E(a,b){return D(n.join(a+";")+(b||""))}function F(a,b){return typeof a===b}function G(a,b){return!!~(""+a).indexOf(b)}function H(a,b){for(var d in a){var e=a[d];if(!G(e,"-")&&j[e]!==c)return b=="pfx"?e:!0}return!1}function I(a,b,d){for(var e in a){var f=b[a[e]];if(f!==c)return d===!1?a[e]:F(f,"function")?f.bind(d||b):f}return!1}function J(a,b,c){var d=a.charAt(0).toUpperCase()+a.slice(1),e=(a+" "+p.join(d+" ")+d).split(" ");return F(b,"string")||F(b,"undefined")?H(e,b):(e=(a+" "+q.join(d+" ")+d).split(" "),I(e,b,c))}function K(){e.input=function(c){for(var d=0,e=c.length;d<e;d++)u[c[d]]=c[d]in k;return u.list&&(u.list=!!b.createElement("datalist")&&!!a.HTMLDataListElement),u}("autocomplete autofocus list placeholder max min multiple pattern required step".split(" ")),e.inputtypes=function(a){for(var d=0,e,f,h,i=a.length;d<i;d++)k.setAttribute("type",f=a[d]),e=k.type!=="text",e&&(k.value=l,k.style.cssText="position:absolute;visibility:hidden;",/^range$/.test(f)&&k.style.WebkitAppearance!==c?(g.appendChild(k),h=b.defaultView,e=h.getComputedStyle&&h.getComputedStyle(k,null).WebkitAppearance!=="textfield"&&k.offsetHeight!==0,g.removeChild(k)):/^(search|tel)$/.test(f)||(/^(url|email)$/.test(f)?e=k.checkValidity&&k.checkValidity()===!1:e=k.value!=l)),t[a[d]]=!!e;return t}("search tel url email datetime date month week time datetime-local number range color".split(" "))}var d="2.6.2",e={},f=!0,g=b.documentElement,h="modernizr",i=b.createElement(h),j=i.style,k=b.createElement("input"),l=":)",m={}.toString,n=" -webkit- -moz- -o- -ms- ".split(" "),o="Webkit Moz O ms",p=o.split(" "),q=o.toLowerCase().split(" "),r={svg:"http://www.w3.org/2000/svg"},s={},t={},u={},v=[],w=v.slice,x,y=function(a,c,d,e){var f,i,j,k,l=b.createElement("div"),m=b.body,n=m||b.createElement("body");if(parseInt(d,10))while(d--)j=b.createElement("div"),j.id=e?e[d]:h+(d+1),l.appendChild(j);return f=["&#173;",'<style id="s',h,'">',a,"</style>"].join(""),l.id=h,(m?l:n).innerHTML+=f,n.appendChild(l),m||(n.style.background="",n.style.overflow="hidden",k=g.style.overflow,g.style.overflow="hidden",g.appendChild(n)),i=c(l,a),m?l.parentNode.removeChild(l):(n.parentNode.removeChild(n),g.style.overflow=k),!!i},z=function(b){var c=a.matchMedia||a.msMatchMedia;if(c)return c(b).matches;var d;return y("@media "+b+" { #"+h+" { position: absolute; } }",function(b){d=(a.getComputedStyle?getComputedStyle(b,null):b.currentStyle)["position"]=="absolute"}),d},A=function(){function d(d,e){e=e||b.createElement(a[d]||"div"),d="on"+d;var f=d in e;return f||(e.setAttribute||(e=b.createElement("div")),e.setAttribute&&e.removeAttribute&&(e.setAttribute(d,""),f=F(e[d],"function"),F(e[d],"undefined")||(e[d]=c),e.removeAttribute(d))),e=null,f}var a={select:"input",change:"input",submit:"form",reset:"form",error:"img",load:"img",abort:"img"};return d}(),B={}.hasOwnProperty,C;!F(B,"undefined")&&!F(B.call,"undefined")?C=function(a,b){return B.call(a,b)}:C=function(a,b){return b in a&&F(a.constructor.prototype[b],"undefined")},Function.prototype.bind||(Function.prototype.bind=function(b){var c=this;if(typeof c!="function")throw new TypeError;var d=w.call(arguments,1),e=function(){if(this instanceof e){var a=function(){};a.prototype=c.prototype;var f=new a,g=c.apply(f,d.concat(w.call(arguments)));return Object(g)===g?g:f}return c.apply(b,d.concat(w.call(arguments)))};return e}),s.flexbox=function(){return J("flexWrap")},s.canvas=function(){var a=b.createElement("canvas");return!!a.getContext&&!!a.getContext("2d")},s.canvastext=function(){return!!e.canvas&&!!F(b.createElement("canvas").getContext("2d").fillText,"function")},s.webgl=function(){return!!a.WebGLRenderingContext},s.touch=function(){var c;return"ontouchstart"in a||a.DocumentTouch&&b instanceof DocumentTouch?c=!0:y(["@media (",n.join("touch-enabled),("),h,")","{#modernizr{top:9px;position:absolute}}"].join(""),function(a){c=a.offsetTop===9}),c},s.geolocation=function(){return"geolocation"in navigator},s.postmessage=function(){return!!a.postMessage},s.websqldatabase=function(){return!!a.openDatabase},s.indexedDB=function(){return!!J("indexedDB",a)},s.hashchange=function(){return A("hashchange",a)&&(b.documentMode===c||b.documentMode>7)},s.history=function(){return!!a.history&&!!history.pushState},s.draganddrop=function(){var a=b.createElement("div");return"draggable"in a||"ondragstart"in a&&"ondrop"in a},s.websockets=function(){return"WebSocket"in a||"MozWebSocket"in a},s.rgba=function(){return D("background-color:rgba(150,255,150,.5)"),G(j.backgroundColor,"rgba")},s.hsla=function(){return D("background-color:hsla(120,40%,100%,.5)"),G(j.backgroundColor,"rgba")||G(j.backgroundColor,"hsla")},s.multiplebgs=function(){return D("background:url(https://),url(https://),red url(https://)"),/(url\s*\(.*?){3}/.test(j.background)},s.backgroundsize=function(){return J("backgroundSize")},s.borderimage=function(){return J("borderImage")},s.borderradius=function(){return J("borderRadius")},s.boxshadow=function(){return J("boxShadow")},s.textshadow=function(){return b.createElement("div").style.textShadow===""},s.opacity=function(){return E("opacity:.55"),/^0.55$/.test(j.opacity)},s.cssanimations=function(){return J("animationName")},s.csscolumns=function(){return J("columnCount")},s.cssgradients=function(){var a="background-image:",b="gradient(linear,left top,right bottom,from(#9f9),to(white));",c="linear-gradient(left top,#9f9, white);";return D((a+"-webkit- ".split(" ").join(b+a)+n.join(c+a)).slice(0,-a.length)),G(j.backgroundImage,"gradient")},s.cssreflections=function(){return J("boxReflect")},s.csstransforms=function(){return!!J("transform")},s.csstransforms3d=function(){var a=!!J("perspective");return a&&"webkitPerspective"in g.style&&y("@media (transform-3d),(-webkit-transform-3d){#modernizr{left:9px;position:absolute;height:3px;}}",function(b,c){a=b.offsetLeft===9&&b.offsetHeight===3}),a},s.csstransitions=function(){return J("transition")},s.fontface=function(){var a;return y('@font-face {font-family:"font";src:url("https://")}',function(c,d){var e=b.getElementById("smodernizr"),f=e.sheet||e.styleSheet,g=f?f.cssRules&&f.cssRules[0]?f.cssRules[0].cssText:f.cssText||"":"";a=/src/i.test(g)&&g.indexOf(d.split(" ")[0])===0}),a},s.generatedcontent=function(){var a;return y(["#",h,"{font:0/0 a}#",h,':after{content:"',l,'";visibility:hidden;font:3px/1 a}'].join(""),function(b){a=b.offsetHeight>=3}),a},s.video=function(){var a=b.createElement("video"),c=!1;try{if(c=!!a.canPlayType)c=new Boolean(c),c.ogg=a.canPlayType('video/ogg; codecs="theora"').replace(/^no$/,""),c.h264=a.canPlayType('video/mp4; codecs="avc1.42E01E"').replace(/^no$/,""),c.webm=a.canPlayType('video/webm; codecs="vp8, vorbis"').replace(/^no$/,"")}catch(d){}return c},s.audio=function(){var a=b.createElement("audio"),c=!1;try{if(c=!!a.canPlayType)c=new Boolean(c),c.ogg=a.canPlayType('audio/ogg; codecs="vorbis"').replace(/^no$/,""),c.mp3=a.canPlayType("audio/mpeg;").replace(/^no$/,""),c.wav=a.canPlayType('audio/wav; codecs="1"').replace(/^no$/,""),c.m4a=(a.canPlayType("audio/x-m4a;")||a.canPlayType("audio/aac;")).replace(/^no$/,"")}catch(d){}return c},s.localstorage=function(){try{return localStorage.setItem(h,h),localStorage.removeItem(h),!0}catch(a){return!1}},s.sessionstorage=function(){try{return sessionStorage.setItem(h,h),sessionStorage.removeItem(h),!0}catch(a){return!1}},s.webworkers=function(){return!!a.Worker},s.applicationcache=function(){return!!a.applicationCache},s.svg=function(){return!!b.createElementNS&&!!b.createElementNS(r.svg,"svg").createSVGRect},s.inlinesvg=function(){var a=b.createElement("div");return a.innerHTML="<svg/>",(a.firstChild&&a.firstChild.namespaceURI)==r.svg},s.smil=function(){return!!b.createElementNS&&/SVGAnimate/.test(m.call(b.createElementNS(r.svg,"animate")))},s.svgclippaths=function(){return!!b.createElementNS&&/SVGClipPath/.test(m.call(b.createElementNS(r.svg,"clipPath")))};for(var L in s)C(s,L)&&(x=L.toLowerCase(),e[x]=s[L](),v.push((e[x]?"":"no-")+x));return e.input||K(),e.addTest=function(a,b){if(typeof a=="object")for(var d in a)C(a,d)&&e.addTest(d,a[d]);else{a=a.toLowerCase();if(e[a]!==c)return e;b=typeof b=="function"?b():b,typeof f!="undefined"&&f&&(g.className+=" "+(b?"":"no-")+a),e[a]=b}return e},D(""),i=k=null,function(a,b){function k(a,b){var c=a.createElement("p"),d=a.getElementsByTagName("head")[0]||a.documentElement;return c.innerHTML="x<style>"+b+"</style>",d.insertBefore(c.lastChild,d.firstChild)}function l(){var a=r.elements;return typeof a=="string"?a.split(" "):a}function m(a){var b=i[a[g]];return b||(b={},h++,a[g]=h,i[h]=b),b}function n(a,c,f){c||(c=b);if(j)return c.createElement(a);f||(f=m(c));var g;return f.cache[a]?g=f.cache[a].cloneNode():e.test(a)?g=(f.cache[a]=f.createElem(a)).cloneNode():g=f.createElem(a),g.canHaveChildren&&!d.test(a)?f.frag.appendChild(g):g}function o(a,c){a||(a=b);if(j)return a.createDocumentFragment();c=c||m(a);var d=c.frag.cloneNode(),e=0,f=l(),g=f.length;for(;e<g;e++)d.createElement(f[e]);return d}function p(a,b){b.cache||(b.cache={},b.createElem=a.createElement,b.createFrag=a.createDocumentFragment,b.frag=b.createFrag()),a.createElement=function(c){return r.shivMethods?n(c,a,b):b.createElem(c)},a.createDocumentFragment=Function("h,f","return function(){var n=f.cloneNode(),c=n.createElement;h.shivMethods&&("+l().join().replace(/\w+/g,function(a){return b.createElem(a),b.frag.createElement(a),'c("'+a+'")'})+");return n}")(r,b.frag)}function q(a){a||(a=b);var c=m(a);return r.shivCSS&&!f&&!c.hasCSS&&(c.hasCSS=!!k(a,"article,aside,figcaption,figure,footer,header,hgroup,nav,section{display:block}mark{background:#FF0;color:#000}")),j||p(a,c),a}var c=a.html5||{},d=/^<|^(?:button|map|select|textarea|object|iframe|option|optgroup)$/i,e=/^(?:a|b|code|div|fieldset|h1|h2|h3|h4|h5|h6|i|label|li|ol|p|q|span|strong|style|table|tbody|td|th|tr|ul)$/i,f,g="_html5shiv",h=0,i={},j;(function(){try{var a=b.createElement("a");a.innerHTML="<xyz></xyz>",f="hidden"in a,j=a.childNodes.length==1||function(){b.createElement("a");var a=b.createDocumentFragment();return typeof a.cloneNode=="undefined"||typeof a.createDocumentFragment=="undefined"||typeof a.createElement=="undefined"}()}catch(c){f=!0,j=!0}})();var r={elements:c.elements||"abbr article aside audio bdi canvas data datalist details figcaption figure footer header hgroup mark meter nav output progress section summary time video",shivCSS:c.shivCSS!==!1,supportsUnknownElements:j,shivMethods:c.shivMethods!==!1,type:"default",shivDocument:q,createElement:n,createDocumentFragment:o};a.html5=r,q(b)}(this,b),e._version=d,e._prefixes=n,e._domPrefixes=q,e._cssomPrefixes=p,e.mq=z,e.hasEvent=A,e.testProp=function(a){return H([a])},e.testAllProps=J,e.testStyles=y,e.prefixed=function(a,b,c){return b?J(a,b,c):J(a,"pfx")},g.className=g.className.replace(/(^|\s)no-js(\s|$)/,"$1$2")+(f?" js "+v.join(" "):""),e}(this,this.document),function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i<a.length;i++)j=a[i],e(j)?g(j,0,l,0):w(j)?B(j):Object(j)===j&&h(j,l);else Object(a)===a&&h(a,l)},B.addPrefix=function(a,b){z[a]=b},B.addFilter=function(a){x.push(a)},B.errorTimeout=1e4,null==b.readyState&&b.addEventListener&&(b.readyState="loading",b.addEventListener("DOMContentLoaded",A=function(){b.removeEventListener("DOMContentLoaded",A,0),b.readyState="complete"},0)),a.yepnope=k(),a.yepnope.executeStack=h,a.yepnope.injectJs=function(a,c,d,e,i,j){var k=b.createElement("script"),l,o,e=e||B.errorTimeout;k.src=a;for(o in d)k.setAttribute(o,d[o]);c=j?h:c||f,k.onreadystatechange=k.onload=function(){!l&&g(k.readyState)&&(l=1,c(),k.onload=k.onreadystatechange=null)},m(function(){l||(l=1,c(1))},e),i?k.onload():n.parentNode.insertBefore(k,n)},a.yepnope.injectCss=function(a,c,d,e,g,i){var e=b.createElement("link"),j,c=i?h:c||f;e.href=a,e.rel="stylesheet",e.type="text/css";for(j in d)e.setAttribute(j,d[j]);g||(n.parentNode.insertBefore(e,n),m(c,0))}}(this,document),Modernizr.load=function(){yepnope.apply(window,[].slice.call(arguments,0))};
diff --git a/0.16/_static/js/theme.js b/0.16/_static/js/theme.js
new file mode 100644
index 00000000000..ceee8f062c3
--- /dev/null
+++ b/0.16/_static/js/theme.js
@@ -0,0 +1,1128 @@
+require=(function(){function r(e,n,t){function o(i,f){if(!n[i]){if(!e[i]){var c="function"==typeof require&&require;if(!f&&c)return c(i,!0);if(u)return u(i,!0);var a=new Error("Cannot find module '"+i+"'");throw a.code="MODULE_NOT_FOUND",a}var p=n[i]={exports:{}};e[i][0].call(p.exports,function(r){var n=e[i][1][r];return o(n||r)},p,p.exports,r,e,n,t)}return n[i].exports}for(var u="function"==typeof require&&require,i=0;i<t.length;i++)o(t[i]);return o}return r})()({1:[function(require,module,exports){
+window.utilities = {
+  scrollTop: function() {
+    var supportPageOffset = window.pageXOffset !== undefined;
+    var isCSS1Compat = ((document.compatMode || "") === "CSS1Compat");
+    var scrollLeft = supportPageOffset ? window.pageXOffset : isCSS1Compat ? document.documentElement.scrollLeft : document.body.scrollLeft;
+    return supportPageOffset ? window.pageYOffset : isCSS1Compat ? document.documentElement.scrollTop : document.body.scrollTop;
+  },
+
+  // Modified from https://stackoverflow.com/a/27078401
+  throttle: function(func, wait, options) {
+    var context, args, result;
+    var timeout = null;
+    var previous = 0;
+    if (!options) options = {};
+    var later = function() {
+      previous = options.leading === false ? 0 : Date.now();
+      timeout = null;
+      result = func.apply(context, args);
+      if (!timeout) context = args = null;
+    };
+    return function() {
+      var now = Date.now();
+      if (!previous && options.leading === false) previous = now;
+      var remaining = wait - (now - previous);
+      context = this;
+      args = arguments;
+      if (remaining <= 0 || remaining > wait) {
+        if (timeout) {
+          clearTimeout(timeout);
+          timeout = null;
+        }
+        previous = now;
+        result = func.apply(context, args);
+        if (!timeout) context = args = null;
+      } else if (!timeout && options.trailing !== false) {
+        timeout = setTimeout(later, remaining);
+      }
+      return result;
+    };
+  },
+
+  closest: function (el, selector) {
+    var matchesFn;
+
+    // find vendor prefix
+    ['matches','webkitMatchesSelector','mozMatchesSelector','msMatchesSelector','oMatchesSelector'].some(function(fn) {
+      if (typeof document.body[fn] == 'function') {
+        matchesFn = fn;
+        return true;
+      }
+      return false;
+    });
+
+    var parent;
+
+    // traverse parents
+    while (el) {
+      parent = el.parentElement;
+      if (parent && parent[matchesFn](selector)) {
+        return parent;
+      }
+      el = parent;
+    }
+
+    return null;
+  },
+
+  // Modified from https://stackoverflow.com/a/18953277
+  offset: function(elem) {
+    if (!elem) {
+      return;
+    }
+
+    rect = elem.getBoundingClientRect();
+
+    // Make sure element is not hidden (display: none) or disconnected
+    if (rect.width || rect.height || elem.getClientRects().length) {
+      var doc = elem.ownerDocument;
+      var docElem = doc.documentElement;
+
+      return {
+        top: rect.top + window.pageYOffset - docElem.clientTop,
+        left: rect.left + window.pageXOffset - docElem.clientLeft
+      };
+    }
+  },
+
+  headersHeight: function() {
+    if (document.getElementById("pytorch-left-menu").classList.contains("make-fixed")) {
+      return document.getElementById("pytorch-page-level-bar").offsetHeight;
+    } else {
+      return document.getElementById("header-holder").offsetHeight +
+             document.getElementById("pytorch-page-level-bar").offsetHeight;
+    }
+  },
+
+  windowHeight: function() {
+    return window.innerHeight ||
+           document.documentElement.clientHeight ||
+           document.body.clientHeight;
+  }
+}
+
+},{}],2:[function(require,module,exports){
+var cookieBanner = {
+  init: function() {
+    cookieBanner.bind();
+
+    var cookieExists = cookieBanner.cookieExists();
+
+    if (!cookieExists) {
+      cookieBanner.setCookie();
+      cookieBanner.showCookieNotice();
+    }
+  },
+
+  bind: function() {
+    $(".close-button").on("click", cookieBanner.hideCookieNotice);
+  },
+
+  cookieExists: function() {
+    var cookie = localStorage.getItem("returningPytorchUser");
+
+    if (cookie) {
+      return true;
+    } else {
+      return false;
+    }
+  },
+
+  setCookie: function() {
+    localStorage.setItem("returningPytorchUser", true);
+  },
+
+  showCookieNotice: function() {
+    $(".cookie-banner-wrapper").addClass("is-visible");
+  },
+
+  hideCookieNotice: function() {
+    $(".cookie-banner-wrapper").removeClass("is-visible");
+  }
+};
+
+$(function() {
+  cookieBanner.init();
+});
+
+},{}],3:[function(require,module,exports){
+window.filterTags = {
+  bind: function() {
+    var options = {
+      valueNames: [{ data: ["tags"] }],
+      page: "6",
+      pagination: true
+    };
+
+    var tutorialList = new List("tutorial-cards", options);
+
+    function filterSelectedTags(cardTags, selectedTags) {
+      return cardTags.some(function(tag) {
+        return selectedTags.some(function(selectedTag) {
+          return selectedTag == tag;
+        });
+      });
+    }
+
+    function updateList() {
+      var selectedTags = [];
+
+      $(".selected").each(function() {
+        selectedTags.push($(this).data("tag"));
+      });
+
+      tutorialList.filter(function(item) {
+        var cardTags;
+
+        if (item.values().tags == null) {
+          cardTags = [""];
+        } else {
+          cardTags = item.values().tags.split(",");
+        }
+
+        if (selectedTags.length == 0) {
+          return true;
+        } else {
+          return filterSelectedTags(cardTags, selectedTags);
+        }
+      });
+    }
+
+    $(".filter-btn").on("click", function() {
+      if ($(this).data("tag") == "all") {
+        $(this).addClass("all-tag-selected");
+        $(".filter").removeClass("selected");
+      } else {
+        $(this).toggleClass("selected");
+        $("[data-tag='all']").removeClass("all-tag-selected");
+      }
+
+      // If no tags are selected then highlight the 'All' tag
+
+      if (!$(".selected")[0]) {
+        $("[data-tag='all']").addClass("all-tag-selected");
+      }
+
+      updateList();
+    });
+  }
+};
+
+},{}],4:[function(require,module,exports){
+// Modified from https://stackoverflow.com/a/32396543
+window.highlightNavigation = {
+  navigationListItems: document.querySelectorAll("#pytorch-right-menu li"),
+  sections: document.querySelectorAll(".pytorch-article .section"),
+  sectionIdTonavigationLink: {},
+
+  bind: function() {
+    if (!sideMenus.displayRightMenu) {
+      return;
+    };
+
+    for (var i = 0; i < highlightNavigation.sections.length; i++) {
+      var id = highlightNavigation.sections[i].id;
+      highlightNavigation.sectionIdTonavigationLink[id] =
+        document.querySelectorAll('#pytorch-right-menu li a[href="#' + id + '"]')[0];
+    }
+
+    $(window).scroll(utilities.throttle(highlightNavigation.highlight, 100));
+  },
+
+  highlight: function() {
+    var rightMenu = document.getElementById("pytorch-right-menu");
+
+    // If right menu is not on the screen don't bother
+    if (rightMenu.offsetWidth === 0 && rightMenu.offsetHeight === 0) {
+      return;
+    }
+
+    var scrollPosition = utilities.scrollTop();
+    var OFFSET_TOP_PADDING = 25;
+    var offset = document.getElementById("header-holder").offsetHeight +
+                 document.getElementById("pytorch-page-level-bar").offsetHeight +
+                 OFFSET_TOP_PADDING;
+
+    var sections = highlightNavigation.sections;
+
+    for (var i = (sections.length - 1); i >= 0; i--) {
+      var currentSection = sections[i];
+      var sectionTop = utilities.offset(currentSection).top;
+
+      if (scrollPosition >= sectionTop - offset) {
+        var navigationLink = highlightNavigation.sectionIdTonavigationLink[currentSection.id];
+        var navigationListItem = utilities.closest(navigationLink, "li");
+
+        if (navigationListItem && !navigationListItem.classList.contains("active")) {
+          for (var i = 0; i < highlightNavigation.navigationListItems.length; i++) {
+            var el = highlightNavigation.navigationListItems[i];
+            if (el.classList.contains("active")) {
+              el.classList.remove("active");
+            }
+          }
+
+          navigationListItem.classList.add("active");
+
+          // Scroll to active item. Not a requested feature but we could revive it. Needs work.
+
+          // var menuTop = $("#pytorch-right-menu").position().top;
+          // var itemTop = navigationListItem.getBoundingClientRect().top;
+          // var TOP_PADDING = 20
+          // var newActiveTop = $("#pytorch-side-scroll-right").scrollTop() + itemTop - menuTop - TOP_PADDING;
+
+          // $("#pytorch-side-scroll-right").animate({
+          //   scrollTop: newActiveTop
+          // }, 100);
+        }
+
+        break;
+      }
+    }
+  }
+};
+
+},{}],5:[function(require,module,exports){
+window.mainMenuDropdown = {
+  bind: function() {
+    $("[data-toggle='ecosystem-dropdown']").on("click", function() {
+      toggleDropdown($(this).attr("data-toggle"));
+    });
+
+    $("[data-toggle='resources-dropdown']").on("click", function() {
+      toggleDropdown($(this).attr("data-toggle"));
+    });
+
+    function toggleDropdown(menuToggle) {
+      var showMenuClass = "show-menu";
+      var menuClass = "." + menuToggle + "-menu";
+
+      if ($(menuClass).hasClass(showMenuClass)) {
+        $(menuClass).removeClass(showMenuClass);
+      } else {
+        $("[data-toggle=" + menuToggle + "].show-menu").removeClass(
+          showMenuClass
+        );
+        $(menuClass).addClass(showMenuClass);
+      }
+    }
+  }
+};
+
+},{}],6:[function(require,module,exports){
+window.mobileMenu = {
+  bind: function() {
+    $("[data-behavior='open-mobile-menu']").on('click', function(e) {
+      e.preventDefault();
+      $(".mobile-main-menu").addClass("open");
+      $("body").addClass('no-scroll');
+
+      mobileMenu.listenForResize();
+    });
+
+    $("[data-behavior='close-mobile-menu']").on('click', function(e) {
+      e.preventDefault();
+      mobileMenu.close();
+    });
+  },
+
+  listenForResize: function() {
+    $(window).on('resize.ForMobileMenu', function() {
+      if ($(this).width() > 768) {
+        mobileMenu.close();
+      }
+    });
+  },
+
+  close: function() {
+    $(".mobile-main-menu").removeClass("open");
+    $("body").removeClass('no-scroll');
+    $(window).off('resize.ForMobileMenu');
+  }
+};
+
+},{}],7:[function(require,module,exports){
+window.mobileTOC = {
+  bind: function() {
+    $("[data-behavior='toggle-table-of-contents']").on("click", function(e) {
+      e.preventDefault();
+
+      var $parent = $(this).parent();
+
+      if ($parent.hasClass("is-open")) {
+        $parent.removeClass("is-open");
+        $(".pytorch-left-menu").slideUp(200, function() {
+          $(this).css({display: ""});
+        });
+      } else {
+        $parent.addClass("is-open");
+        $(".pytorch-left-menu").slideDown(200);
+      }
+    });
+  }
+}
+
+},{}],8:[function(require,module,exports){
+window.pytorchAnchors = {
+  bind: function() {
+    // Replace Sphinx-generated anchors with anchorjs ones
+    $(".headerlink").text("");
+
+    window.anchors.add(".pytorch-article .headerlink");
+
+    $(".anchorjs-link").each(function() {
+      var $headerLink = $(this).closest(".headerlink");
+      var href = $headerLink.attr("href");
+      var clone = this.outerHTML;
+
+      $clone = $(clone).attr("href", href);
+      $headerLink.before($clone);
+      $headerLink.remove();
+    });
+  }
+};
+
+},{}],9:[function(require,module,exports){
+// Modified from https://stackoverflow.com/a/13067009
+// Going for a JS solution to scrolling to an anchor so we can benefit from
+// less hacky css and smooth scrolling.
+
+window.scrollToAnchor = {
+  bind: function() {
+    var document = window.document;
+    var history = window.history;
+    var location = window.location
+    var HISTORY_SUPPORT = !!(history && history.pushState);
+
+    var anchorScrolls = {
+      ANCHOR_REGEX: /^#[^ ]+$/,
+      offsetHeightPx: function() {
+        var OFFSET_HEIGHT_PADDING = 20;
+        // TODO: this is a little janky. We should try to not rely on JS for this
+        return utilities.headersHeight() + OFFSET_HEIGHT_PADDING;
+      },
+
+      /**
+       * Establish events, and fix initial scroll position if a hash is provided.
+       */
+      init: function() {
+        this.scrollToCurrent();
+        // This interferes with clicks below it, causing a double fire
+        // $(window).on('hashchange', $.proxy(this, 'scrollToCurrent'));
+        $('body').on('click', 'a', $.proxy(this, 'delegateAnchors'));
+        $('body').on('click', '#pytorch-right-menu li span', $.proxy(this, 'delegateSpans'));
+      },
+
+      /**
+       * Return the offset amount to deduct from the normal scroll position.
+       * Modify as appropriate to allow for dynamic calculations
+       */
+      getFixedOffset: function() {
+        return this.offsetHeightPx();
+      },
+
+      /**
+       * If the provided href is an anchor which resolves to an element on the
+       * page, scroll to it.
+       * @param  {String} href
+       * @return {Boolean} - Was the href an anchor.
+       */
+      scrollIfAnchor: function(href, pushToHistory) {
+        var match, anchorOffset;
+
+        if(!this.ANCHOR_REGEX.test(href)) {
+          return false;
+        }
+
+        match = document.getElementById(href.slice(1));
+
+        if(match) {
+          var anchorOffset = $(match).offset().top - this.getFixedOffset();
+
+          $('html, body').scrollTop(anchorOffset);
+
+          // Add the state to history as-per normal anchor links
+          if(HISTORY_SUPPORT && pushToHistory) {
+            history.pushState({}, document.title, location.pathname + href);
+          }
+        }
+
+        return !!match;
+      },
+
+      /**
+       * Attempt to scroll to the current location's hash.
+       */
+      scrollToCurrent: function(e) {
+        if(this.scrollIfAnchor(window.location.hash) && e) {
+          e.preventDefault();
+        }
+      },
+
+      delegateSpans: function(e) {
+        var elem = utilities.closest(e.target, "a");
+
+        if(this.scrollIfAnchor(elem.getAttribute('href'), true)) {
+          e.preventDefault();
+        }
+      },
+
+      /**
+       * If the click event's target was an anchor, fix the scroll position.
+       */
+      delegateAnchors: function(e) {
+        var elem = e.target;
+
+        if(this.scrollIfAnchor(elem.getAttribute('href'), true)) {
+          e.preventDefault();
+        }
+      }
+    };
+
+    $(document).ready($.proxy(anchorScrolls, 'init'));
+  }
+};
+
+},{}],10:[function(require,module,exports){
+window.sideMenus = {
+  rightMenuIsOnScreen: function() {
+    return document.getElementById("pytorch-content-right").offsetParent !== null;
+  },
+
+  isFixedToBottom: false,
+
+  bind: function() {
+    sideMenus.handleLeftMenu();
+
+    var rightMenuLinks = document.querySelectorAll("#pytorch-right-menu li");
+    var rightMenuHasLinks = rightMenuLinks.length > 1;
+
+    if (!rightMenuHasLinks) {
+      for (var i = 0; i < rightMenuLinks.length; i++) {
+        rightMenuLinks[i].style.display = "none";
+      }
+    }
+
+    if (rightMenuHasLinks) {
+      // Don't show the Shortcuts menu title text unless there are menu items
+      document.getElementById("pytorch-shortcuts-wrapper").style.display = "block";
+
+      // We are hiding the titles of the pages in the right side menu but there are a few
+      // pages that include other pages in the right side menu (see 'torch.nn' in the docs)
+      // so if we exclude those it looks confusing. Here we add a 'title-link' class to these
+      // links so we can exclude them from normal right side menu link operations
+      var titleLinks = document.querySelectorAll(
+        "#pytorch-right-menu #pytorch-side-scroll-right \
+         > ul > li > a.reference.internal"
+      );
+
+      for (var i = 0; i < titleLinks.length; i++) {
+        var link = titleLinks[i];
+
+        link.classList.add("title-link");
+
+        if (
+          link.nextElementSibling &&
+          link.nextElementSibling.tagName === "UL" &&
+          link.nextElementSibling.children.length > 0
+        ) {
+          link.classList.add("has-children");
+        }
+      }
+
+      // Add + expansion signifiers to normal right menu links that have sub menus
+      var menuLinks = document.querySelectorAll(
+        "#pytorch-right-menu ul li ul li a.reference.internal"
+      );
+
+      for (var i = 0; i < menuLinks.length; i++) {
+        if (
+          menuLinks[i].nextElementSibling &&
+          menuLinks[i].nextElementSibling.tagName === "UL"
+        ) {
+          menuLinks[i].classList.add("not-expanded");
+        }
+      }
+
+      // If a hash is present on page load recursively expand menu items leading to selected item
+      var linkWithHash =
+        document.querySelector(
+          "#pytorch-right-menu a[href=\"" + window.location.hash + "\"]"
+        );
+
+      if (linkWithHash) {
+        // Expand immediate sibling list if present
+        if (
+          linkWithHash.nextElementSibling &&
+          linkWithHash.nextElementSibling.tagName === "UL" &&
+          linkWithHash.nextElementSibling.children.length > 0
+        ) {
+          linkWithHash.nextElementSibling.style.display = "block";
+          linkWithHash.classList.add("expanded");
+        }
+
+        // Expand ancestor lists if any
+        sideMenus.expandClosestUnexpandedParentList(linkWithHash);
+      }
+
+      // Bind click events on right menu links
+      $("#pytorch-right-menu a.reference.internal").on("click", function() {
+        if (this.classList.contains("expanded")) {
+          this.nextElementSibling.style.display = "none";
+          this.classList.remove("expanded");
+          this.classList.add("not-expanded");
+        } else if (this.classList.contains("not-expanded")) {
+          this.nextElementSibling.style.display = "block";
+          this.classList.remove("not-expanded");
+          this.classList.add("expanded");
+        }
+      });
+
+      sideMenus.handleRightMenu();
+    }
+
+    $(window).on('resize scroll', function(e) {
+      sideMenus.handleNavBar();
+
+      sideMenus.handleLeftMenu();
+
+      if (sideMenus.rightMenuIsOnScreen()) {
+        sideMenus.handleRightMenu();
+      }
+    });
+  },
+
+  leftMenuIsFixed: function() {
+    return document.getElementById("pytorch-left-menu").classList.contains("make-fixed");
+  },
+
+  handleNavBar: function() {
+    var mainHeaderHeight = document.getElementById('header-holder').offsetHeight;
+
+    // If we are scrolled past the main navigation header fix the sub menu bar to top of page
+    if (utilities.scrollTop() >= mainHeaderHeight) {
+      document.getElementById("pytorch-left-menu").classList.add("make-fixed");
+      document.getElementById("pytorch-page-level-bar").classList.add("left-menu-is-fixed");
+    } else {
+      document.getElementById("pytorch-left-menu").classList.remove("make-fixed");
+      document.getElementById("pytorch-page-level-bar").classList.remove("left-menu-is-fixed");
+    }
+  },
+
+  expandClosestUnexpandedParentList: function (el) {
+    var closestParentList = utilities.closest(el, "ul");
+
+    if (closestParentList) {
+      var closestParentLink = closestParentList.previousElementSibling;
+      var closestParentLinkExists = closestParentLink &&
+                                    closestParentLink.tagName === "A" &&
+                                    closestParentLink.classList.contains("reference");
+
+      if (closestParentLinkExists) {
+        // Don't add expansion class to any title links
+         if (closestParentLink.classList.contains("title-link")) {
+           return;
+         }
+
+        closestParentList.style.display = "block";
+        closestParentLink.classList.remove("not-expanded");
+        closestParentLink.classList.add("expanded");
+        sideMenus.expandClosestUnexpandedParentList(closestParentLink);
+      }
+    }
+  },
+
+  handleLeftMenu: function () {
+    var windowHeight = utilities.windowHeight();
+    var topOfFooterRelativeToWindow = document.getElementById("docs-tutorials-resources").getBoundingClientRect().top;
+
+    if (topOfFooterRelativeToWindow >= windowHeight) {
+      document.getElementById("pytorch-left-menu").style.height = "100%";
+    } else {
+      var howManyPixelsOfTheFooterAreInTheWindow = windowHeight - topOfFooterRelativeToWindow;
+      var leftMenuDifference = howManyPixelsOfTheFooterAreInTheWindow;
+      document.getElementById("pytorch-left-menu").style.height = (windowHeight - leftMenuDifference) + "px";
+    }
+  },
+
+  handleRightMenu: function() {
+    var rightMenuWrapper = document.getElementById("pytorch-content-right");
+    var rightMenu = document.getElementById("pytorch-right-menu");
+    var rightMenuList = rightMenu.getElementsByTagName("ul")[0];
+    var article = document.getElementById("pytorch-article");
+    var articleHeight = article.offsetHeight;
+    var articleBottom = utilities.offset(article).top + articleHeight;
+    var mainHeaderHeight = document.getElementById('header-holder').offsetHeight;
+
+    if (utilities.scrollTop() < mainHeaderHeight) {
+      rightMenuWrapper.style.height = "100%";
+      rightMenu.style.top = 0;
+      rightMenu.classList.remove("scrolling-fixed");
+      rightMenu.classList.remove("scrolling-absolute");
+    } else {
+      if (rightMenu.classList.contains("scrolling-fixed")) {
+        var rightMenuBottom =
+          utilities.offset(rightMenuList).top + rightMenuList.offsetHeight;
+
+        if (rightMenuBottom >= articleBottom) {
+          rightMenuWrapper.style.height = articleHeight + mainHeaderHeight + "px";
+          rightMenu.style.top = utilities.scrollTop() - mainHeaderHeight + "px";
+          rightMenu.classList.add("scrolling-absolute");
+          rightMenu.classList.remove("scrolling-fixed");
+        }
+      } else {
+        rightMenuWrapper.style.height = articleHeight + mainHeaderHeight + "px";
+        rightMenu.style.top =
+          articleBottom - mainHeaderHeight - rightMenuList.offsetHeight + "px";
+        rightMenu.classList.add("scrolling-absolute");
+      }
+
+      if (utilities.scrollTop() < articleBottom - rightMenuList.offsetHeight) {
+        rightMenuWrapper.style.height = "100%";
+        rightMenu.style.top = "";
+        rightMenu.classList.remove("scrolling-absolute");
+        rightMenu.classList.add("scrolling-fixed");
+      }
+    }
+
+    var rightMenuSideScroll = document.getElementById("pytorch-side-scroll-right");
+    var sideScrollFromWindowTop = rightMenuSideScroll.getBoundingClientRect().top;
+
+    rightMenuSideScroll.style.height = utilities.windowHeight() - sideScrollFromWindowTop + "px";
+  }
+};
+
+},{}],"pytorch-sphinx-theme":[function(require,module,exports){
+var jQuery = (typeof(window) != 'undefined') ? window.jQuery : require('jquery');
+
+// Sphinx theme nav state
+function ThemeNav () {
+
+    var nav = {
+        navBar: null,
+        win: null,
+        winScroll: false,
+        winResize: false,
+        linkScroll: false,
+        winPosition: 0,
+        winHeight: null,
+        docHeight: null,
+        isRunning: false
+    };
+
+    nav.enable = function (withStickyNav) {
+        var self = this;
+
+        // TODO this can likely be removed once the theme javascript is broken
+        // out from the RTD assets. This just ensures old projects that are
+        // calling `enable()` get the sticky menu on by default. All other cals
+        // to `enable` should include an argument for enabling the sticky menu.
+        if (typeof(withStickyNav) == 'undefined') {
+            withStickyNav = true;
+        }
+
+        if (self.isRunning) {
+            // Only allow enabling nav logic once
+            return;
+        }
+
+        self.isRunning = true;
+        jQuery(function ($) {
+            self.init($);
+
+            self.reset();
+            self.win.on('hashchange', self.reset);
+
+            if (withStickyNav) {
+                // Set scroll monitor
+                self.win.on('scroll', function () {
+                    if (!self.linkScroll) {
+                        if (!self.winScroll) {
+                            self.winScroll = true;
+                            requestAnimationFrame(function() { self.onScroll(); });
+                        }
+                    }
+                });
+            }
+
+            // Set resize monitor
+            self.win.on('resize', function () {
+                if (!self.winResize) {
+                    self.winResize = true;
+                    requestAnimationFrame(function() { self.onResize(); });
+                }
+            });
+
+            self.onResize();
+        });
+
+    };
+
+    // TODO remove this with a split in theme and Read the Docs JS logic as
+    // well, it's only here to support 0.3.0 installs of our theme.
+    nav.enableSticky = function() {
+        this.enable(true);
+    };
+
+    nav.init = function ($) {
+        var doc = $(document),
+            self = this;
+
+        this.navBar = $('div.pytorch-side-scroll:first');
+        this.win = $(window);
+
+        // Set up javascript UX bits
+        $(document)
+            // Shift nav in mobile when clicking the menu.
+            .on('click', "[data-toggle='pytorch-left-menu-nav-top']", function() {
+                $("[data-toggle='wy-nav-shift']").toggleClass("shift");
+                $("[data-toggle='rst-versions']").toggleClass("shift");
+            })
+
+            // Nav menu link click operations
+            .on('click', ".pytorch-menu-vertical .current ul li a", function() {
+                var target = $(this);
+                // Close menu when you click a link.
+                $("[data-toggle='wy-nav-shift']").removeClass("shift");
+                $("[data-toggle='rst-versions']").toggleClass("shift");
+                // Handle dynamic display of l3 and l4 nav lists
+                self.toggleCurrent(target);
+                self.hashChange();
+            })
+            .on('click', "[data-toggle='rst-current-version']", function() {
+                $("[data-toggle='rst-versions']").toggleClass("shift-up");
+            })
+
+        // Make tables responsive
+        $("table.docutils:not(.field-list,.footnote,.citation)")
+            .wrap("<div class='wy-table-responsive'></div>");
+
+        // Add extra class to responsive tables that contain
+        // footnotes or citations so that we can target them for styling
+        $("table.docutils.footnote")
+            .wrap("<div class='wy-table-responsive footnote'></div>");
+        $("table.docutils.citation")
+            .wrap("<div class='wy-table-responsive citation'></div>");
+
+        // Add expand links to all parents of nested ul
+        $('.pytorch-menu-vertical ul').not('.simple').siblings('a').each(function () {
+            var link = $(this);
+                expand = $('<span class="toctree-expand"></span>');
+            expand.on('click', function (ev) {
+                self.toggleCurrent(link);
+                ev.stopPropagation();
+                return false;
+            });
+            link.prepend(expand);
+        });
+    };
+
+    nav.reset = function () {
+        // Get anchor from URL and open up nested nav
+        var anchor = encodeURI(window.location.hash) || '#';
+
+        try {
+            var vmenu = $('.pytorch-menu-vertical');
+            var link = vmenu.find('[href="' + anchor + '"]');
+            if (link.length === 0) {
+                // this link was not found in the sidebar.
+                // Find associated id element, then its closest section
+                // in the document and try with that one.
+                var id_elt = $('.document [id="' + anchor.substring(1) + '"]');
+                var closest_section = id_elt.closest('div.section');
+                link = vmenu.find('[href="#' + closest_section.attr("id") + '"]');
+                if (link.length === 0) {
+                    // still not found in the sidebar. fall back to main section
+                    link = vmenu.find('[href="#"]');
+                }
+            }
+            // If we found a matching link then reset current and re-apply
+            // otherwise retain the existing match
+            if (link.length > 0) {
+                $('.pytorch-menu-vertical .current').removeClass('current');
+                link.addClass('current');
+                link.closest('li.toctree-l1').addClass('current');
+                link.closest('li.toctree-l1').parent().addClass('current');
+                link.closest('li.toctree-l1').addClass('current');
+                link.closest('li.toctree-l2').addClass('current');
+                link.closest('li.toctree-l3').addClass('current');
+                link.closest('li.toctree-l4').addClass('current');
+            }
+        }
+        catch (err) {
+            console.log("Error expanding nav for anchor", err);
+        }
+
+    };
+
+    nav.onScroll = function () {
+        this.winScroll = false;
+        var newWinPosition = this.win.scrollTop(),
+            winBottom = newWinPosition + this.winHeight,
+            navPosition = this.navBar.scrollTop(),
+            newNavPosition = navPosition + (newWinPosition - this.winPosition);
+        if (newWinPosition < 0 || winBottom > this.docHeight) {
+            return;
+        }
+        this.navBar.scrollTop(newNavPosition);
+        this.winPosition = newWinPosition;
+    };
+
+    nav.onResize = function () {
+        this.winResize = false;
+        this.winHeight = this.win.height();
+        this.docHeight = $(document).height();
+    };
+
+    nav.hashChange = function () {
+        this.linkScroll = true;
+        this.win.one('hashchange', function () {
+            this.linkScroll = false;
+        });
+    };
+
+    nav.toggleCurrent = function (elem) {
+        var parent_li = elem.closest('li');
+        parent_li.siblings('li.current').removeClass('current');
+        parent_li.siblings().find('li.current').removeClass('current');
+        parent_li.find('> ul li.current').removeClass('current');
+        parent_li.toggleClass('current');
+    }
+
+    return nav;
+};
+
+module.exports.ThemeNav = ThemeNav();
+
+if (typeof(window) != 'undefined') {
+    window.SphinxRtdTheme = {
+        Navigation: module.exports.ThemeNav,
+        // TODO remove this once static assets are split up between the theme
+        // and Read the Docs. For now, this patches 0.3.0 to be backwards
+        // compatible with a pre-0.3.0 layout.html
+        StickyNav: module.exports.ThemeNav,
+    };
+}
+
+
+// requestAnimationFrame polyfill by Erik Möller. fixes from Paul Irish and Tino Zijdel
+// https://gist.github.com/paulirish/1579671
+// MIT license
+
+(function() {
+    var lastTime = 0;
+    var vendors = ['ms', 'moz', 'webkit', 'o'];
+    for(var x = 0; x < vendors.length && !window.requestAnimationFrame; ++x) {
+        window.requestAnimationFrame = window[vendors[x]+'RequestAnimationFrame'];
+        window.cancelAnimationFrame = window[vendors[x]+'CancelAnimationFrame']
+                                   || window[vendors[x]+'CancelRequestAnimationFrame'];
+    }
+
+    if (!window.requestAnimationFrame)
+        window.requestAnimationFrame = function(callback, element) {
+            var currTime = new Date().getTime();
+            var timeToCall = Math.max(0, 16 - (currTime - lastTime));
+            var id = window.setTimeout(function() { callback(currTime + timeToCall); },
+              timeToCall);
+            lastTime = currTime + timeToCall;
+            return id;
+        };
+
+    if (!window.cancelAnimationFrame)
+        window.cancelAnimationFrame = function(id) {
+            clearTimeout(id);
+        };
+}());
+
+$(".sphx-glr-thumbcontainer").removeAttr("tooltip");
+$("table").removeAttr("border");
+
+// This code replaces the default sphinx gallery download buttons
+// with the 3 download buttons at the top of the page
+
+var downloadNote = $(".sphx-glr-download-link-note.admonition.note");
+if (downloadNote.length >= 1) {
+    var tutorialUrlArray = $("#tutorial-type").text().split('/');
+        tutorialUrlArray[0] = tutorialUrlArray[0] + "_source"
+
+    var githubLink = "https://github.com/pytorch/tutorials/blob/master/" + tutorialUrlArray.join("/") + ".py",
+        notebookLink = $(".reference.download")[1].href,
+        notebookDownloadPath = notebookLink.split('_downloads')[1],
+        colabLink = "https://colab.research.google.com/github/pytorch/tutorials/blob/gh-pages/_downloads" + notebookDownloadPath;
+
+    $("#google-colab-link").wrap("<a href=" + colabLink + " data-behavior='call-to-action-event' data-response='Run in Google Colab' target='_blank'/>");
+    $("#download-notebook-link").wrap("<a href=" + notebookLink + " data-behavior='call-to-action-event' data-response='Download Notebook'/>");
+    $("#github-view-link").wrap("<a href=" + githubLink + " data-behavior='call-to-action-event' data-response='View on Github' target='_blank'/>");
+} else {
+    $(".pytorch-call-to-action-links").hide();
+}
+
+//This code handles the Expand/Hide toggle for the Docs/Tutorials left nav items
+
+$(document).ready(function() {
+  var caption = "#pytorch-left-menu p.caption";
+  var collapseAdded = $(this).not("checked");
+  $(caption).each(function () {
+    var menuName = this.innerText.replace(/[^\w\s]/gi, "").trim();
+    $(this).find("span").addClass("checked");
+    if (collapsedSections.includes(menuName) == true && collapseAdded && sessionStorage.getItem(menuName) !== "expand" || sessionStorage.getItem(menuName) == "collapse") {
+      $(this.firstChild).after("<span class='expand-menu'>[ + ]</span>");
+      $(this.firstChild).after("<span class='hide-menu collapse'>[ - ]</span>");
+      $(this).next("ul").hide();
+    } else if (collapsedSections.includes(menuName) == false && collapseAdded || sessionStorage.getItem(menuName) == "expand") {
+      $(this.firstChild).after("<span class='expand-menu collapse'>[ + ]</span>");
+      $(this.firstChild).after("<span class='hide-menu'>[ - ]</span>");
+    }
+  });
+
+  $(".expand-menu").on("click", function () {
+    $(this).prev(".hide-menu").toggle();
+    $(this).parent().next("ul").toggle();
+    var menuName = $(this).parent().text().replace(/[^\w\s]/gi, "").trim();
+    if (sessionStorage.getItem(menuName) == "collapse") {
+      sessionStorage.removeItem(menuName);
+    }
+    sessionStorage.setItem(menuName, "expand");
+    toggleList(this);
+  });
+
+  $(".hide-menu").on("click", function () {
+    $(this).next(".expand-menu").toggle();
+    $(this).parent().next("ul").toggle();
+    var menuName = $(this).parent().text().replace(/[^\w\s]/gi, "").trim();
+    if (sessionStorage.getItem(menuName) == "expand") {
+      sessionStorage.removeItem(menuName);
+    }
+    sessionStorage.setItem(menuName, "collapse");
+    toggleList(this);
+  });
+
+  function toggleList(menuCommand) {
+    $(menuCommand).toggle();
+  }
+});
+
+// Build an array from each tag that's present
+
+var tagList = $(".tutorials-card-container").map(function() {
+    return $(this).data("tags").split(",").map(function(item) {
+        return item.trim();
+      });
+}).get();
+
+function unique(value, index, self) {
+      return self.indexOf(value) == index && value != ""
+    }
+
+// Only return unique tags
+
+var tags = tagList.sort().filter(unique);
+
+// Add filter buttons to the top of the page for each tag
+
+function createTagMenu() {
+    tags.forEach(function(item){
+    $(".tutorial-filter-menu").append(" <div class='tutorial-filter filter-btn filter' data-tag='" + item + "'>" + item + "</div>")
+  })
+};
+
+createTagMenu();
+
+// Remove hyphens if they are present in the filter buttons
+
+$(".tags").each(function(){
+    var tags = $(this).text().split(",");
+    tags.forEach(function(tag, i ) {
+       tags[i] = tags[i].replace(/-/, ' ')
+    })
+    $(this).html(tags.join(", "));
+});
+
+// Remove hyphens if they are present in the card body
+
+$(".tutorial-filter").each(function(){
+    var tag = $(this).text();
+    $(this).html(tag.replace(/-/, ' '))
+})
+
+// Remove any empty p tags that Sphinx adds
+
+$("#tutorial-cards p").each(function(index, item) {
+    if(!$(item).text().trim()) {
+        $(item).remove();
+    }
+});
+
+// Jump back to top on pagination click
+
+$(document).on("click", ".page", function() {
+    $('html, body').animate(
+      {scrollTop: $("#dropdown-filter-tags").position().top},
+      'slow'
+    );
+});
+
+var link = $("a[href='intermediate/speech_command_recognition_with_torchaudio.html']");
+
+if (link.text() == "SyntaxError") {
+    console.log("There is an issue with the intermediate/speech_command_recognition_with_torchaudio.html menu item.");
+    link.text("Speech Command Recognition with torchaudio");
+}
+
+$(".stars-outer > i").hover(function() {
+    $(this).prevAll().addBack().toggleClass("fas star-fill");
+});
+
+$(".stars-outer > i").on("click", function() {
+    $(this).prevAll().each(function() {
+        $(this).addBack().addClass("fas star-fill");
+    });
+
+    $(".stars-outer > i").each(function() {
+        $(this).unbind("mouseenter mouseleave").css({
+            "pointer-events": "none"
+        });
+    });
+})
+
+$("#pytorch-side-scroll-right li a").on("click", function (e) {
+  var href = $(this).attr("href");
+  $('html, body').stop().animate({
+    scrollTop: $(href).offset().top - 100
+  }, 850);
+  e.preventDefault;
+});
+
+var lastId,
+  topMenu = $("#pytorch-side-scroll-right"),
+  topMenuHeight = topMenu.outerHeight() + 1,
+  // All sidenav items
+  menuItems = topMenu.find("a"),
+  // Anchors for menu items
+  scrollItems = menuItems.map(function () {
+    var item = $(this).attr("href");
+    if (item.length) {
+      return item;
+    }
+  });
+
+$(window).scroll(function () {
+  var fromTop = $(this).scrollTop() + topMenuHeight;
+  var article = ".section";
+
+  $(article).each(function (i) {
+    var offsetScroll = $(this).offset().top - $(window).scrollTop();
+    if (
+      offsetScroll <= topMenuHeight + 200 &&
+      offsetScroll >= topMenuHeight - 200 &&
+      scrollItems[i] == "#" + $(this).attr("id") &&
+      $(".hidden:visible")
+    ) {
+      $(menuItems).removeClass("side-scroll-highlight");
+      $(menuItems[i]).addClass("side-scroll-highlight");
+    }
+  });
+});
+
+
+},{"jquery":"jquery"}]},{},[1,2,3,4,5,6,7,8,9,10,"pytorch-sphinx-theme"]);
diff --git a/0.16/_static/js/vendor/anchor.min.js b/0.16/_static/js/vendor/anchor.min.js
new file mode 100644
index 00000000000..1216eeac267
--- /dev/null
+++ b/0.16/_static/js/vendor/anchor.min.js
@@ -0,0 +1,9 @@
+// @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&dn=expat.txt Expat
+//
+// AnchorJS - v4.3.0 - 2020-10-21
+// https://www.bryanbraun.com/anchorjs/
+// Copyright (c) 2020 Bryan Braun; Licensed MIT
+//
+// @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&dn=expat.txt Expat
+!function(A,e){"use strict";"function"==typeof define&&define.amd?define([],e):"object"==typeof module&&module.exports?module.exports=e():(A.AnchorJS=e(),A.anchors=new A.AnchorJS)}(this,function(){"use strict";return function(A){function d(A){A.icon=Object.prototype.hasOwnProperty.call(A,"icon")?A.icon:"",A.visible=Object.prototype.hasOwnProperty.call(A,"visible")?A.visible:"hover",A.placement=Object.prototype.hasOwnProperty.call(A,"placement")?A.placement:"right",A.ariaLabel=Object.prototype.hasOwnProperty.call(A,"ariaLabel")?A.ariaLabel:"Anchor",A.class=Object.prototype.hasOwnProperty.call(A,"class")?A.class:"",A.base=Object.prototype.hasOwnProperty.call(A,"base")?A.base:"",A.truncate=Object.prototype.hasOwnProperty.call(A,"truncate")?Math.floor(A.truncate):64,A.titleText=Object.prototype.hasOwnProperty.call(A,"titleText")?A.titleText:""}function f(A){var e;if("string"==typeof A||A instanceof String)e=[].slice.call(document.querySelectorAll(A));else{if(!(Array.isArray(A)||A instanceof NodeList))throw new TypeError("The selector provided to AnchorJS was invalid.");e=[].slice.call(A)}return e}this.options=A||{},this.elements=[],d(this.options),this.isTouchDevice=function(){return Boolean("ontouchstart"in window||window.TouchEvent||window.DocumentTouch&&document instanceof DocumentTouch)},this.add=function(A){var e,t,o,n,i,s,a,r,c,l,h,u,p=[];if(d(this.options),"touch"===(h=this.options.visible)&&(h=this.isTouchDevice()?"always":"hover"),0===(e=f(A=A||"h2, h3, h4, h5, h6")).length)return this;for(!function(){if(null!==document.head.querySelector("style.anchorjs"))return;var A,e=document.createElement("style");e.className="anchorjs",e.appendChild(document.createTextNode("")),void 0===(A=document.head.querySelector('[rel="stylesheet"],style'))?document.head.appendChild(e):document.head.insertBefore(e,A);e.sheet.insertRule(".anchorjs-link{opacity:0;text-decoration:none;-webkit-font-smoothing:antialiased;-moz-osx-font-smoothing:grayscale}",e.sheet.cssRules.length),e.sheet.insertRule(":hover>.anchorjs-link,.anchorjs-link:focus{opacity:1}",e.sheet.cssRules.length),e.sheet.insertRule("[data-anchorjs-icon]::after{content:attr(data-anchorjs-icon)}",e.sheet.cssRules.length),e.sheet.insertRule('@font-face{font-family:anchorjs-icons;src:url(data:n/a;base64,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) format("truetype")}',e.sheet.cssRules.length)}(),t=document.querySelectorAll("[id]"),o=[].map.call(t,function(A){return A.id}),i=0;i<e.length;i++)if(this.hasAnchorJSLink(e[i]))p.push(i);else{if(e[i].hasAttribute("id"))n=e[i].getAttribute("id");else if(e[i].hasAttribute("data-anchor-id"))n=e[i].getAttribute("data-anchor-id");else{for(c=r=this.urlify(e[i].textContent),a=0;void 0!==s&&(c=r+"-"+a),a+=1,-1!==(s=o.indexOf(c)););s=void 0,o.push(c),e[i].setAttribute("id",c),n=c}(l=document.createElement("a")).className="anchorjs-link "+this.options.class,l.setAttribute("aria-label",this.options.ariaLabel),l.setAttribute("data-anchorjs-icon",this.options.icon),this.options.titleText&&(l.title=this.options.titleText),u=document.querySelector("base")?window.location.pathname+window.location.search:"",u=this.options.base||u,l.href=u+"#"+n,"always"===h&&(l.style.opacity="1"),""===this.options.icon&&(l.style.font="1em/1 anchorjs-icons","left"===this.options.placement&&(l.style.lineHeight="inherit")),"left"===this.options.placement?(l.style.position="absolute",l.style.marginLeft="-1em",l.style.paddingRight=".5em",e[i].insertBefore(l,e[i].firstChild)):(l.style.paddingLeft=".375em",e[i].appendChild(l))}for(i=0;i<p.length;i++)e.splice(p[i]-i,1);return this.elements=this.elements.concat(e),this},this.remove=function(A){for(var e,t,o=f(A),n=0;n<o.length;n++)(t=o[n].querySelector(".anchorjs-link"))&&(-1!==(e=this.elements.indexOf(o[n]))&&this.elements.splice(e,1),o[n].removeChild(t));return this},this.removeAll=function(){this.remove(this.elements)},this.urlify=function(A){var e=document.createElement("textarea");e.innerHTML=A,A=e.value;return this.options.truncate||d(this.options),A.trim().replace(/'/gi,"").replace(/[& +$,:;=?@"#{}|^~[`%!'<>\]./()*\\\n\t\b\v\u00A0]/g,"-").replace(/-{2,}/g,"-").substring(0,this.options.truncate).replace(/^-+|-+$/gm,"").toLowerCase()},this.hasAnchorJSLink=function(A){var e=A.firstChild&&-1<(" "+A.firstChild.className+" ").indexOf(" anchorjs-link "),t=A.lastChild&&-1<(" "+A.lastChild.className+" ").indexOf(" anchorjs-link ");return e||t||!1}}});
+// @license-end
\ No newline at end of file
diff --git a/0.16/_static/js/vendor/bootstrap.min.js b/0.16/_static/js/vendor/bootstrap.min.js
new file mode 100644
index 00000000000..c4c0d1f95cd
--- /dev/null
+++ b/0.16/_static/js/vendor/bootstrap.min.js
@@ -0,0 +1,7 @@
+/*!
+  * Bootstrap v4.3.1 (https://getbootstrap.com/)
+  * Copyright 2011-2019 The Bootstrap Authors (https://github.com/twbs/bootstrap/graphs/contributors)
+  * Licensed under MIT (https://github.com/twbs/bootstrap/blob/master/LICENSE)
+  */
+!function(t,e){"object"==typeof exports&&"undefined"!=typeof module?e(exports,require("jquery"),require("popper.js")):"function"==typeof define&&define.amd?define(["exports","jquery","popper.js"],e):e((t=t||self).bootstrap={},t.jQuery,t.Popper)}(this,function(t,g,u){"use strict";function i(t,e){for(var n=0;n<e.length;n++){var i=e[n];i.enumerable=i.enumerable||!1,i.configurable=!0,"value"in i&&(i.writable=!0),Object.defineProperty(t,i.key,i)}}function s(t,e,n){return e&&i(t.prototype,e),n&&i(t,n),t}function l(o){for(var t=1;t<arguments.length;t++){var r=null!=arguments[t]?arguments[t]:{},e=Object.keys(r);"function"==typeof Object.getOwnPropertySymbols&&(e=e.concat(Object.getOwnPropertySymbols(r).filter(function(t){return Object.getOwnPropertyDescriptor(r,t).enumerable}))),e.forEach(function(t){var e,n,i;e=o,i=r[n=t],n in e?Object.defineProperty(e,n,{value:i,enumerable:!0,configurable:!0,writable:!0}):e[n]=i})}return o}g=g&&g.hasOwnProperty("default")?g.default:g,u=u&&u.hasOwnProperty("default")?u.default:u;var e="transitionend";function n(t){var e=this,n=!1;return g(this).one(_.TRANSITION_END,function(){n=!0}),setTimeout(function(){n||_.triggerTransitionEnd(e)},t),this}var _={TRANSITION_END:"bsTransitionEnd",getUID:function(t){for(;t+=~~(1e6*Math.random()),document.getElementById(t););return t},getSelectorFromElement:function(t){var e=t.getAttribute("data-target");if(!e||"#"===e){var n=t.getAttribute("href");e=n&&"#"!==n?n.trim():""}try{return document.querySelector(e)?e:null}catch(t){return null}},getTransitionDurationFromElement:function(t){if(!t)return 0;var e=g(t).css("transition-duration"),n=g(t).css("transition-delay"),i=parseFloat(e),o=parseFloat(n);return i||o?(e=e.split(",")[0],n=n.split(",")[0],1e3*(parseFloat(e)+parseFloat(n))):0},reflow:function(t){return t.offsetHeight},triggerTransitionEnd:function(t){g(t).trigger(e)},supportsTransitionEnd:function(){return Boolean(e)},isElement:function(t){return(t[0]||t).nodeType},typeCheckConfig:function(t,e,n){for(var i in n)if(Object.prototype.hasOwnProperty.call(n,i)){var o=n[i],r=e[i],s=r&&_.isElement(r)?"element":(a=r,{}.toString.call(a).match(/\s([a-z]+)/i)[1].toLowerCase());if(!new RegExp(o).test(s))throw new Error(t.toUpperCase()+': Option "'+i+'" provided type "'+s+'" but expected type "'+o+'".')}var a},findShadowRoot:function(t){if(!document.documentElement.attachShadow)return null;if("function"!=typeof t.getRootNode)return t instanceof ShadowRoot?t:t.parentNode?_.findShadowRoot(t.parentNode):null;var e=t.getRootNode();return e instanceof ShadowRoot?e:null}};g.fn.emulateTransitionEnd=n,g.event.special[_.TRANSITION_END]={bindType:e,delegateType:e,handle:function(t){if(g(t.target).is(this))return t.handleObj.handler.apply(this,arguments)}};var o="alert",r="bs.alert",a="."+r,c=g.fn[o],h={CLOSE:"close"+a,CLOSED:"closed"+a,CLICK_DATA_API:"click"+a+".data-api"},f="alert",d="fade",m="show",p=function(){function i(t){this._element=t}var t=i.prototype;return t.close=function(t){var e=this._element;t&&(e=this._getRootElement(t)),this._triggerCloseEvent(e).isDefaultPrevented()||this._removeElement(e)},t.dispose=function(){g.removeData(this._element,r),this._element=null},t._getRootElement=function(t){var e=_.getSelectorFromElement(t),n=!1;return e&&(n=document.querySelector(e)),n||(n=g(t).closest("."+f)[0]),n},t._triggerCloseEvent=function(t){var e=g.Event(h.CLOSE);return g(t).trigger(e),e},t._removeElement=function(e){var n=this;if(g(e).removeClass(m),g(e).hasClass(d)){var t=_.getTransitionDurationFromElement(e);g(e).one(_.TRANSITION_END,function(t){return n._destroyElement(e,t)}).emulateTransitionEnd(t)}else this._destroyElement(e)},t._destroyElement=function(t){g(t).detach().trigger(h.CLOSED).remove()},i._jQueryInterface=function(n){return this.each(function(){var t=g(this),e=t.data(r);e||(e=new i(this),t.data(r,e)),"close"===n&&e[n](this)})},i._handleDismiss=function(e){return function(t){t&&t.preventDefault(),e.close(this)}},s(i,null,[{key:"VERSION",get:function(){return"4.3.1"}}]),i}();g(document).on(h.CLICK_DATA_API,'[data-dismiss="alert"]',p._handleDismiss(new p)),g.fn[o]=p._jQueryInterface,g.fn[o].Constructor=p,g.fn[o].noConflict=function(){return g.fn[o]=c,p._jQueryInterface};var v="button",y="bs.button",E="."+y,C=".data-api",T=g.fn[v],S="active",b="btn",I="focus",D='[data-toggle^="button"]',w='[data-toggle="buttons"]',A='input:not([type="hidden"])',N=".active",O=".btn",k={CLICK_DATA_API:"click"+E+C,FOCUS_BLUR_DATA_API:"focus"+E+C+" blur"+E+C},P=function(){function n(t){this._element=t}var t=n.prototype;return t.toggle=function(){var t=!0,e=!0,n=g(this._element).closest(w)[0];if(n){var i=this._element.querySelector(A);if(i){if("radio"===i.type)if(i.checked&&this._element.classList.contains(S))t=!1;else{var o=n.querySelector(N);o&&g(o).removeClass(S)}if(t){if(i.hasAttribute("disabled")||n.hasAttribute("disabled")||i.classList.contains("disabled")||n.classList.contains("disabled"))return;i.checked=!this._element.classList.contains(S),g(i).trigger("change")}i.focus(),e=!1}}e&&this._element.setAttribute("aria-pressed",!this._element.classList.contains(S)),t&&g(this._element).toggleClass(S)},t.dispose=function(){g.removeData(this._element,y),this._element=null},n._jQueryInterface=function(e){return this.each(function(){var t=g(this).data(y);t||(t=new n(this),g(this).data(y,t)),"toggle"===e&&t[e]()})},s(n,null,[{key:"VERSION",get:function(){return"4.3.1"}}]),n}();g(document).on(k.CLICK_DATA_API,D,function(t){t.preventDefault();var e=t.target;g(e).hasClass(b)||(e=g(e).closest(O)),P._jQueryInterface.call(g(e),"toggle")}).on(k.FOCUS_BLUR_DATA_API,D,function(t){var e=g(t.target).closest(O)[0];g(e).toggleClass(I,/^focus(in)?$/.test(t.type))}),g.fn[v]=P._jQueryInterface,g.fn[v].Constructor=P,g.fn[v].noConflict=function(){return g.fn[v]=T,P._jQueryInterface};var L="carousel",j="bs.carousel",H="."+j,R=".data-api",x=g.fn[L],F={interval:5e3,keyboard:!0,slide:!1,pause:"hover",wrap:!0,touch:!0},U={interval:"(number|boolean)",keyboard:"boolean",slide:"(boolean|string)",pause:"(string|boolean)",wrap:"boolean",touch:"boolean"},W="next",q="prev",M="left",K="right",Q={SLIDE:"slide"+H,SLID:"slid"+H,KEYDOWN:"keydown"+H,MOUSEENTER:"mouseenter"+H,MOUSELEAVE:"mouseleave"+H,TOUCHSTART:"touchstart"+H,TOUCHMOVE:"touchmove"+H,TOUCHEND:"touchend"+H,POINTERDOWN:"pointerdown"+H,POINTERUP:"pointerup"+H,DRAG_START:"dragstart"+H,LOAD_DATA_API:"load"+H+R,CLICK_DATA_API:"click"+H+R},B="carousel",V="active",Y="slide",z="carousel-item-right",X="carousel-item-left",$="carousel-item-next",G="carousel-item-prev",J="pointer-event",Z=".active",tt=".active.carousel-item",et=".carousel-item",nt=".carousel-item img",it=".carousel-item-next, .carousel-item-prev",ot=".carousel-indicators",rt="[data-slide], [data-slide-to]",st='[data-ride="carousel"]',at={TOUCH:"touch",PEN:"pen"},lt=function(){function r(t,e){this._items=null,this._interval=null,this._activeElement=null,this._isPaused=!1,this._isSliding=!1,this.touchTimeout=null,this.touchStartX=0,this.touchDeltaX=0,this._config=this._getConfig(e),this._element=t,this._indicatorsElement=this._element.querySelector(ot),this._touchSupported="ontouchstart"in document.documentElement||0<navigator.maxTouchPoints,this._pointerEvent=Boolean(window.PointerEvent||window.MSPointerEvent),this._addEventListeners()}var t=r.prototype;return t.next=function(){this._isSliding||this._slide(W)},t.nextWhenVisible=function(){!document.hidden&&g(this._element).is(":visible")&&"hidden"!==g(this._element).css("visibility")&&this.next()},t.prev=function(){this._isSliding||this._slide(q)},t.pause=function(t){t||(this._isPaused=!0),this._element.querySelector(it)&&(_.triggerTransitionEnd(this._element),this.cycle(!0)),clearInterval(this._interval),this._interval=null},t.cycle=function(t){t||(this._isPaused=!1),this._interval&&(clearInterval(this._interval),this._interval=null),this._config.interval&&!this._isPaused&&(this._interval=setInterval((document.visibilityState?this.nextWhenVisible:this.next).bind(this),this._config.interval))},t.to=function(t){var e=this;this._activeElement=this._element.querySelector(tt);var n=this._getItemIndex(this._activeElement);if(!(t>this._items.length-1||t<0))if(this._isSliding)g(this._element).one(Q.SLID,function(){return e.to(t)});else{if(n===t)return this.pause(),void this.cycle();var i=n<t?W:q;this._slide(i,this._items[t])}},t.dispose=function(){g(this._element).off(H),g.removeData(this._element,j),this._items=null,this._config=null,this._element=null,this._interval=null,this._isPaused=null,this._isSliding=null,this._activeElement=null,this._indicatorsElement=null},t._getConfig=function(t){return t=l({},F,t),_.typeCheckConfig(L,t,U),t},t._handleSwipe=function(){var t=Math.abs(this.touchDeltaX);if(!(t<=40)){var e=t/this.touchDeltaX;0<e&&this.prev(),e<0&&this.next()}},t._addEventListeners=function(){var e=this;this._config.keyboard&&g(this._element).on(Q.KEYDOWN,function(t){return e._keydown(t)}),"hover"===this._config.pause&&g(this._element).on(Q.MOUSEENTER,function(t){return e.pause(t)}).on(Q.MOUSELEAVE,function(t){return e.cycle(t)}),this._config.touch&&this._addTouchEventListeners()},t._addTouchEventListeners=function(){var n=this;if(this._touchSupported){var e=function(t){n._pointerEvent&&at[t.originalEvent.pointerType.toUpperCase()]?n.touchStartX=t.originalEvent.clientX:n._pointerEvent||(n.touchStartX=t.originalEvent.touches[0].clientX)},i=function(t){n._pointerEvent&&at[t.originalEvent.pointerType.toUpperCase()]&&(n.touchDeltaX=t.originalEvent.clientX-n.touchStartX),n._handleSwipe(),"hover"===n._config.pause&&(n.pause(),n.touchTimeout&&clearTimeout(n.touchTimeout),n.touchTimeout=setTimeout(function(t){return n.cycle(t)},500+n._config.interval))};g(this._element.querySelectorAll(nt)).on(Q.DRAG_START,function(t){return t.preventDefault()}),this._pointerEvent?(g(this._element).on(Q.POINTERDOWN,function(t){return e(t)}),g(this._element).on(Q.POINTERUP,function(t){return i(t)}),this._element.classList.add(J)):(g(this._element).on(Q.TOUCHSTART,function(t){return e(t)}),g(this._element).on(Q.TOUCHMOVE,function(t){var e;(e=t).originalEvent.touches&&1<e.originalEvent.touches.length?n.touchDeltaX=0:n.touchDeltaX=e.originalEvent.touches[0].clientX-n.touchStartX}),g(this._element).on(Q.TOUCHEND,function(t){return i(t)}))}},t._keydown=function(t){if(!/input|textarea/i.test(t.target.tagName))switch(t.which){case 37:t.preventDefault(),this.prev();break;case 39:t.preventDefault(),this.next()}},t._getItemIndex=function(t){return this._items=t&&t.parentNode?[].slice.call(t.parentNode.querySelectorAll(et)):[],this._items.indexOf(t)},t._getItemByDirection=function(t,e){var n=t===W,i=t===q,o=this._getItemIndex(e),r=this._items.length-1;if((i&&0===o||n&&o===r)&&!this._config.wrap)return e;var s=(o+(t===q?-1:1))%this._items.length;return-1===s?this._items[this._items.length-1]:this._items[s]},t._triggerSlideEvent=function(t,e){var n=this._getItemIndex(t),i=this._getItemIndex(this._element.querySelector(tt)),o=g.Event(Q.SLIDE,{relatedTarget:t,direction:e,from:i,to:n});return g(this._element).trigger(o),o},t._setActiveIndicatorElement=function(t){if(this._indicatorsElement){var e=[].slice.call(this._indicatorsElement.querySelectorAll(Z));g(e).removeClass(V);var n=this._indicatorsElement.children[this._getItemIndex(t)];n&&g(n).addClass(V)}},t._slide=function(t,e){var n,i,o,r=this,s=this._element.querySelector(tt),a=this._getItemIndex(s),l=e||s&&this._getItemByDirection(t,s),c=this._getItemIndex(l),h=Boolean(this._interval);if(o=t===W?(n=X,i=$,M):(n=z,i=G,K),l&&g(l).hasClass(V))this._isSliding=!1;else if(!this._triggerSlideEvent(l,o).isDefaultPrevented()&&s&&l){this._isSliding=!0,h&&this.pause(),this._setActiveIndicatorElement(l);var u=g.Event(Q.SLID,{relatedTarget:l,direction:o,from:a,to:c});if(g(this._element).hasClass(Y)){g(l).addClass(i),_.reflow(l),g(s).addClass(n),g(l).addClass(n);var f=parseInt(l.getAttribute("data-interval"),10);this._config.interval=f?(this._config.defaultInterval=this._config.defaultInterval||this._config.interval,f):this._config.defaultInterval||this._config.interval;var d=_.getTransitionDurationFromElement(s);g(s).one(_.TRANSITION_END,function(){g(l).removeClass(n+" "+i).addClass(V),g(s).removeClass(V+" "+i+" "+n),r._isSliding=!1,setTimeout(function(){return g(r._element).trigger(u)},0)}).emulateTransitionEnd(d)}else g(s).removeClass(V),g(l).addClass(V),this._isSliding=!1,g(this._element).trigger(u);h&&this.cycle()}},r._jQueryInterface=function(i){return this.each(function(){var t=g(this).data(j),e=l({},F,g(this).data());"object"==typeof i&&(e=l({},e,i));var n="string"==typeof i?i:e.slide;if(t||(t=new r(this,e),g(this).data(j,t)),"number"==typeof i)t.to(i);else if("string"==typeof n){if("undefined"==typeof t[n])throw new TypeError('No method named "'+n+'"');t[n]()}else e.interval&&e.ride&&(t.pause(),t.cycle())})},r._dataApiClickHandler=function(t){var e=_.getSelectorFromElement(this);if(e){var n=g(e)[0];if(n&&g(n).hasClass(B)){var i=l({},g(n).data(),g(this).data()),o=this.getAttribute("data-slide-to");o&&(i.interval=!1),r._jQueryInterface.call(g(n),i),o&&g(n).data(j).to(o),t.preventDefault()}}},s(r,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"Default",get:function(){return F}}]),r}();g(document).on(Q.CLICK_DATA_API,rt,lt._dataApiClickHandler),g(window).on(Q.LOAD_DATA_API,function(){for(var t=[].slice.call(document.querySelectorAll(st)),e=0,n=t.length;e<n;e++){var i=g(t[e]);lt._jQueryInterface.call(i,i.data())}}),g.fn[L]=lt._jQueryInterface,g.fn[L].Constructor=lt,g.fn[L].noConflict=function(){return g.fn[L]=x,lt._jQueryInterface};var ct="collapse",ht="bs.collapse",ut="."+ht,ft=g.fn[ct],dt={toggle:!0,parent:""},gt={toggle:"boolean",parent:"(string|element)"},_t={SHOW:"show"+ut,SHOWN:"shown"+ut,HIDE:"hide"+ut,HIDDEN:"hidden"+ut,CLICK_DATA_API:"click"+ut+".data-api"},mt="show",pt="collapse",vt="collapsing",yt="collapsed",Et="width",Ct="height",Tt=".show, .collapsing",St='[data-toggle="collapse"]',bt=function(){function a(e,t){this._isTransitioning=!1,this._element=e,this._config=this._getConfig(t),this._triggerArray=[].slice.call(document.querySelectorAll('[data-toggle="collapse"][href="#'+e.id+'"],[data-toggle="collapse"][data-target="#'+e.id+'"]'));for(var n=[].slice.call(document.querySelectorAll(St)),i=0,o=n.length;i<o;i++){var r=n[i],s=_.getSelectorFromElement(r),a=[].slice.call(document.querySelectorAll(s)).filter(function(t){return t===e});null!==s&&0<a.length&&(this._selector=s,this._triggerArray.push(r))}this._parent=this._config.parent?this._getParent():null,this._config.parent||this._addAriaAndCollapsedClass(this._element,this._triggerArray),this._config.toggle&&this.toggle()}var t=a.prototype;return t.toggle=function(){g(this._element).hasClass(mt)?this.hide():this.show()},t.show=function(){var t,e,n=this;if(!this._isTransitioning&&!g(this._element).hasClass(mt)&&(this._parent&&0===(t=[].slice.call(this._parent.querySelectorAll(Tt)).filter(function(t){return"string"==typeof n._config.parent?t.getAttribute("data-parent")===n._config.parent:t.classList.contains(pt)})).length&&(t=null),!(t&&(e=g(t).not(this._selector).data(ht))&&e._isTransitioning))){var i=g.Event(_t.SHOW);if(g(this._element).trigger(i),!i.isDefaultPrevented()){t&&(a._jQueryInterface.call(g(t).not(this._selector),"hide"),e||g(t).data(ht,null));var o=this._getDimension();g(this._element).removeClass(pt).addClass(vt),this._element.style[o]=0,this._triggerArray.length&&g(this._triggerArray).removeClass(yt).attr("aria-expanded",!0),this.setTransitioning(!0);var r="scroll"+(o[0].toUpperCase()+o.slice(1)),s=_.getTransitionDurationFromElement(this._element);g(this._element).one(_.TRANSITION_END,function(){g(n._element).removeClass(vt).addClass(pt).addClass(mt),n._element.style[o]="",n.setTransitioning(!1),g(n._element).trigger(_t.SHOWN)}).emulateTransitionEnd(s),this._element.style[o]=this._element[r]+"px"}}},t.hide=function(){var t=this;if(!this._isTransitioning&&g(this._element).hasClass(mt)){var e=g.Event(_t.HIDE);if(g(this._element).trigger(e),!e.isDefaultPrevented()){var n=this._getDimension();this._element.style[n]=this._element.getBoundingClientRect()[n]+"px",_.reflow(this._element),g(this._element).addClass(vt).removeClass(pt).removeClass(mt);var i=this._triggerArray.length;if(0<i)for(var o=0;o<i;o++){var r=this._triggerArray[o],s=_.getSelectorFromElement(r);if(null!==s)g([].slice.call(document.querySelectorAll(s))).hasClass(mt)||g(r).addClass(yt).attr("aria-expanded",!1)}this.setTransitioning(!0);this._element.style[n]="";var a=_.getTransitionDurationFromElement(this._element);g(this._element).one(_.TRANSITION_END,function(){t.setTransitioning(!1),g(t._element).removeClass(vt).addClass(pt).trigger(_t.HIDDEN)}).emulateTransitionEnd(a)}}},t.setTransitioning=function(t){this._isTransitioning=t},t.dispose=function(){g.removeData(this._element,ht),this._config=null,this._parent=null,this._element=null,this._triggerArray=null,this._isTransitioning=null},t._getConfig=function(t){return(t=l({},dt,t)).toggle=Boolean(t.toggle),_.typeCheckConfig(ct,t,gt),t},t._getDimension=function(){return g(this._element).hasClass(Et)?Et:Ct},t._getParent=function(){var t,n=this;_.isElement(this._config.parent)?(t=this._config.parent,"undefined"!=typeof this._config.parent.jquery&&(t=this._config.parent[0])):t=document.querySelector(this._config.parent);var e='[data-toggle="collapse"][data-parent="'+this._config.parent+'"]',i=[].slice.call(t.querySelectorAll(e));return g(i).each(function(t,e){n._addAriaAndCollapsedClass(a._getTargetFromElement(e),[e])}),t},t._addAriaAndCollapsedClass=function(t,e){var n=g(t).hasClass(mt);e.length&&g(e).toggleClass(yt,!n).attr("aria-expanded",n)},a._getTargetFromElement=function(t){var e=_.getSelectorFromElement(t);return e?document.querySelector(e):null},a._jQueryInterface=function(i){return this.each(function(){var t=g(this),e=t.data(ht),n=l({},dt,t.data(),"object"==typeof i&&i?i:{});if(!e&&n.toggle&&/show|hide/.test(i)&&(n.toggle=!1),e||(e=new a(this,n),t.data(ht,e)),"string"==typeof i){if("undefined"==typeof e[i])throw new TypeError('No method named "'+i+'"');e[i]()}})},s(a,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"Default",get:function(){return dt}}]),a}();g(document).on(_t.CLICK_DATA_API,St,function(t){"A"===t.currentTarget.tagName&&t.preventDefault();var n=g(this),e=_.getSelectorFromElement(this),i=[].slice.call(document.querySelectorAll(e));g(i).each(function(){var t=g(this),e=t.data(ht)?"toggle":n.data();bt._jQueryInterface.call(t,e)})}),g.fn[ct]=bt._jQueryInterface,g.fn[ct].Constructor=bt,g.fn[ct].noConflict=function(){return g.fn[ct]=ft,bt._jQueryInterface};var It="dropdown",Dt="bs.dropdown",wt="."+Dt,At=".data-api",Nt=g.fn[It],Ot=new RegExp("38|40|27"),kt={HIDE:"hide"+wt,HIDDEN:"hidden"+wt,SHOW:"show"+wt,SHOWN:"shown"+wt,CLICK:"click"+wt,CLICK_DATA_API:"click"+wt+At,KEYDOWN_DATA_API:"keydown"+wt+At,KEYUP_DATA_API:"keyup"+wt+At},Pt="disabled",Lt="show",jt="dropup",Ht="dropright",Rt="dropleft",xt="dropdown-menu-right",Ft="position-static",Ut='[data-toggle="dropdown"]',Wt=".dropdown form",qt=".dropdown-menu",Mt=".navbar-nav",Kt=".dropdown-menu .dropdown-item:not(.disabled):not(:disabled)",Qt="top-start",Bt="top-end",Vt="bottom-start",Yt="bottom-end",zt="right-start",Xt="left-start",$t={offset:0,flip:!0,boundary:"scrollParent",reference:"toggle",display:"dynamic"},Gt={offset:"(number|string|function)",flip:"boolean",boundary:"(string|element)",reference:"(string|element)",display:"string"},Jt=function(){function c(t,e){this._element=t,this._popper=null,this._config=this._getConfig(e),this._menu=this._getMenuElement(),this._inNavbar=this._detectNavbar(),this._addEventListeners()}var t=c.prototype;return t.toggle=function(){if(!this._element.disabled&&!g(this._element).hasClass(Pt)){var t=c._getParentFromElement(this._element),e=g(this._menu).hasClass(Lt);if(c._clearMenus(),!e){var n={relatedTarget:this._element},i=g.Event(kt.SHOW,n);if(g(t).trigger(i),!i.isDefaultPrevented()){if(!this._inNavbar){if("undefined"==typeof u)throw new TypeError("Bootstrap's dropdowns require Popper.js (https://popper.js.org/)");var o=this._element;"parent"===this._config.reference?o=t:_.isElement(this._config.reference)&&(o=this._config.reference,"undefined"!=typeof this._config.reference.jquery&&(o=this._config.reference[0])),"scrollParent"!==this._config.boundary&&g(t).addClass(Ft),this._popper=new u(o,this._menu,this._getPopperConfig())}"ontouchstart"in document.documentElement&&0===g(t).closest(Mt).length&&g(document.body).children().on("mouseover",null,g.noop),this._element.focus(),this._element.setAttribute("aria-expanded",!0),g(this._menu).toggleClass(Lt),g(t).toggleClass(Lt).trigger(g.Event(kt.SHOWN,n))}}}},t.show=function(){if(!(this._element.disabled||g(this._element).hasClass(Pt)||g(this._menu).hasClass(Lt))){var t={relatedTarget:this._element},e=g.Event(kt.SHOW,t),n=c._getParentFromElement(this._element);g(n).trigger(e),e.isDefaultPrevented()||(g(this._menu).toggleClass(Lt),g(n).toggleClass(Lt).trigger(g.Event(kt.SHOWN,t)))}},t.hide=function(){if(!this._element.disabled&&!g(this._element).hasClass(Pt)&&g(this._menu).hasClass(Lt)){var t={relatedTarget:this._element},e=g.Event(kt.HIDE,t),n=c._getParentFromElement(this._element);g(n).trigger(e),e.isDefaultPrevented()||(g(this._menu).toggleClass(Lt),g(n).toggleClass(Lt).trigger(g.Event(kt.HIDDEN,t)))}},t.dispose=function(){g.removeData(this._element,Dt),g(this._element).off(wt),this._element=null,(this._menu=null)!==this._popper&&(this._popper.destroy(),this._popper=null)},t.update=function(){this._inNavbar=this._detectNavbar(),null!==this._popper&&this._popper.scheduleUpdate()},t._addEventListeners=function(){var e=this;g(this._element).on(kt.CLICK,function(t){t.preventDefault(),t.stopPropagation(),e.toggle()})},t._getConfig=function(t){return t=l({},this.constructor.Default,g(this._element).data(),t),_.typeCheckConfig(It,t,this.constructor.DefaultType),t},t._getMenuElement=function(){if(!this._menu){var t=c._getParentFromElement(this._element);t&&(this._menu=t.querySelector(qt))}return this._menu},t._getPlacement=function(){var t=g(this._element.parentNode),e=Vt;return t.hasClass(jt)?(e=Qt,g(this._menu).hasClass(xt)&&(e=Bt)):t.hasClass(Ht)?e=zt:t.hasClass(Rt)?e=Xt:g(this._menu).hasClass(xt)&&(e=Yt),e},t._detectNavbar=function(){return 0<g(this._element).closest(".navbar").length},t._getOffset=function(){var e=this,t={};return"function"==typeof this._config.offset?t.fn=function(t){return t.offsets=l({},t.offsets,e._config.offset(t.offsets,e._element)||{}),t}:t.offset=this._config.offset,t},t._getPopperConfig=function(){var t={placement:this._getPlacement(),modifiers:{offset:this._getOffset(),flip:{enabled:this._config.flip},preventOverflow:{boundariesElement:this._config.boundary}}};return"static"===this._config.display&&(t.modifiers.applyStyle={enabled:!1}),t},c._jQueryInterface=function(e){return this.each(function(){var t=g(this).data(Dt);if(t||(t=new c(this,"object"==typeof e?e:null),g(this).data(Dt,t)),"string"==typeof e){if("undefined"==typeof t[e])throw new TypeError('No method named "'+e+'"');t[e]()}})},c._clearMenus=function(t){if(!t||3!==t.which&&("keyup"!==t.type||9===t.which))for(var e=[].slice.call(document.querySelectorAll(Ut)),n=0,i=e.length;n<i;n++){var o=c._getParentFromElement(e[n]),r=g(e[n]).data(Dt),s={relatedTarget:e[n]};if(t&&"click"===t.type&&(s.clickEvent=t),r){var a=r._menu;if(g(o).hasClass(Lt)&&!(t&&("click"===t.type&&/input|textarea/i.test(t.target.tagName)||"keyup"===t.type&&9===t.which)&&g.contains(o,t.target))){var l=g.Event(kt.HIDE,s);g(o).trigger(l),l.isDefaultPrevented()||("ontouchstart"in document.documentElement&&g(document.body).children().off("mouseover",null,g.noop),e[n].setAttribute("aria-expanded","false"),g(a).removeClass(Lt),g(o).removeClass(Lt).trigger(g.Event(kt.HIDDEN,s)))}}}},c._getParentFromElement=function(t){var e,n=_.getSelectorFromElement(t);return n&&(e=document.querySelector(n)),e||t.parentNode},c._dataApiKeydownHandler=function(t){if((/input|textarea/i.test(t.target.tagName)?!(32===t.which||27!==t.which&&(40!==t.which&&38!==t.which||g(t.target).closest(qt).length)):Ot.test(t.which))&&(t.preventDefault(),t.stopPropagation(),!this.disabled&&!g(this).hasClass(Pt))){var e=c._getParentFromElement(this),n=g(e).hasClass(Lt);if(n&&(!n||27!==t.which&&32!==t.which)){var i=[].slice.call(e.querySelectorAll(Kt));if(0!==i.length){var o=i.indexOf(t.target);38===t.which&&0<o&&o--,40===t.which&&o<i.length-1&&o++,o<0&&(o=0),i[o].focus()}}else{if(27===t.which){var r=e.querySelector(Ut);g(r).trigger("focus")}g(this).trigger("click")}}},s(c,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"Default",get:function(){return $t}},{key:"DefaultType",get:function(){return Gt}}]),c}();g(document).on(kt.KEYDOWN_DATA_API,Ut,Jt._dataApiKeydownHandler).on(kt.KEYDOWN_DATA_API,qt,Jt._dataApiKeydownHandler).on(kt.CLICK_DATA_API+" "+kt.KEYUP_DATA_API,Jt._clearMenus).on(kt.CLICK_DATA_API,Ut,function(t){t.preventDefault(),t.stopPropagation(),Jt._jQueryInterface.call(g(this),"toggle")}).on(kt.CLICK_DATA_API,Wt,function(t){t.stopPropagation()}),g.fn[It]=Jt._jQueryInterface,g.fn[It].Constructor=Jt,g.fn[It].noConflict=function(){return g.fn[It]=Nt,Jt._jQueryInterface};var Zt="modal",te="bs.modal",ee="."+te,ne=g.fn[Zt],ie={backdrop:!0,keyboard:!0,focus:!0,show:!0},oe={backdrop:"(boolean|string)",keyboard:"boolean",focus:"boolean",show:"boolean"},re={HIDE:"hide"+ee,HIDDEN:"hidden"+ee,SHOW:"show"+ee,SHOWN:"shown"+ee,FOCUSIN:"focusin"+ee,RESIZE:"resize"+ee,CLICK_DISMISS:"click.dismiss"+ee,KEYDOWN_DISMISS:"keydown.dismiss"+ee,MOUSEUP_DISMISS:"mouseup.dismiss"+ee,MOUSEDOWN_DISMISS:"mousedown.dismiss"+ee,CLICK_DATA_API:"click"+ee+".data-api"},se="modal-dialog-scrollable",ae="modal-scrollbar-measure",le="modal-backdrop",ce="modal-open",he="fade",ue="show",fe=".modal-dialog",de=".modal-body",ge='[data-toggle="modal"]',_e='[data-dismiss="modal"]',me=".fixed-top, .fixed-bottom, .is-fixed, .sticky-top",pe=".sticky-top",ve=function(){function o(t,e){this._config=this._getConfig(e),this._element=t,this._dialog=t.querySelector(fe),this._backdrop=null,this._isShown=!1,this._isBodyOverflowing=!1,this._ignoreBackdropClick=!1,this._isTransitioning=!1,this._scrollbarWidth=0}var t=o.prototype;return t.toggle=function(t){return this._isShown?this.hide():this.show(t)},t.show=function(t){var e=this;if(!this._isShown&&!this._isTransitioning){g(this._element).hasClass(he)&&(this._isTransitioning=!0);var n=g.Event(re.SHOW,{relatedTarget:t});g(this._element).trigger(n),this._isShown||n.isDefaultPrevented()||(this._isShown=!0,this._checkScrollbar(),this._setScrollbar(),this._adjustDialog(),this._setEscapeEvent(),this._setResizeEvent(),g(this._element).on(re.CLICK_DISMISS,_e,function(t){return e.hide(t)}),g(this._dialog).on(re.MOUSEDOWN_DISMISS,function(){g(e._element).one(re.MOUSEUP_DISMISS,function(t){g(t.target).is(e._element)&&(e._ignoreBackdropClick=!0)})}),this._showBackdrop(function(){return e._showElement(t)}))}},t.hide=function(t){var e=this;if(t&&t.preventDefault(),this._isShown&&!this._isTransitioning){var n=g.Event(re.HIDE);if(g(this._element).trigger(n),this._isShown&&!n.isDefaultPrevented()){this._isShown=!1;var i=g(this._element).hasClass(he);if(i&&(this._isTransitioning=!0),this._setEscapeEvent(),this._setResizeEvent(),g(document).off(re.FOCUSIN),g(this._element).removeClass(ue),g(this._element).off(re.CLICK_DISMISS),g(this._dialog).off(re.MOUSEDOWN_DISMISS),i){var o=_.getTransitionDurationFromElement(this._element);g(this._element).one(_.TRANSITION_END,function(t){return e._hideModal(t)}).emulateTransitionEnd(o)}else this._hideModal()}}},t.dispose=function(){[window,this._element,this._dialog].forEach(function(t){return g(t).off(ee)}),g(document).off(re.FOCUSIN),g.removeData(this._element,te),this._config=null,this._element=null,this._dialog=null,this._backdrop=null,this._isShown=null,this._isBodyOverflowing=null,this._ignoreBackdropClick=null,this._isTransitioning=null,this._scrollbarWidth=null},t.handleUpdate=function(){this._adjustDialog()},t._getConfig=function(t){return t=l({},ie,t),_.typeCheckConfig(Zt,t,oe),t},t._showElement=function(t){var e=this,n=g(this._element).hasClass(he);this._element.parentNode&&this._element.parentNode.nodeType===Node.ELEMENT_NODE||document.body.appendChild(this._element),this._element.style.display="block",this._element.removeAttribute("aria-hidden"),this._element.setAttribute("aria-modal",!0),g(this._dialog).hasClass(se)?this._dialog.querySelector(de).scrollTop=0:this._element.scrollTop=0,n&&_.reflow(this._element),g(this._element).addClass(ue),this._config.focus&&this._enforceFocus();var i=g.Event(re.SHOWN,{relatedTarget:t}),o=function(){e._config.focus&&e._element.focus(),e._isTransitioning=!1,g(e._element).trigger(i)};if(n){var r=_.getTransitionDurationFromElement(this._dialog);g(this._dialog).one(_.TRANSITION_END,o).emulateTransitionEnd(r)}else o()},t._enforceFocus=function(){var e=this;g(document).off(re.FOCUSIN).on(re.FOCUSIN,function(t){document!==t.target&&e._element!==t.target&&0===g(e._element).has(t.target).length&&e._element.focus()})},t._setEscapeEvent=function(){var e=this;this._isShown&&this._config.keyboard?g(this._element).on(re.KEYDOWN_DISMISS,function(t){27===t.which&&(t.preventDefault(),e.hide())}):this._isShown||g(this._element).off(re.KEYDOWN_DISMISS)},t._setResizeEvent=function(){var e=this;this._isShown?g(window).on(re.RESIZE,function(t){return e.handleUpdate(t)}):g(window).off(re.RESIZE)},t._hideModal=function(){var t=this;this._element.style.display="none",this._element.setAttribute("aria-hidden",!0),this._element.removeAttribute("aria-modal"),this._isTransitioning=!1,this._showBackdrop(function(){g(document.body).removeClass(ce),t._resetAdjustments(),t._resetScrollbar(),g(t._element).trigger(re.HIDDEN)})},t._removeBackdrop=function(){this._backdrop&&(g(this._backdrop).remove(),this._backdrop=null)},t._showBackdrop=function(t){var e=this,n=g(this._element).hasClass(he)?he:"";if(this._isShown&&this._config.backdrop){if(this._backdrop=document.createElement("div"),this._backdrop.className=le,n&&this._backdrop.classList.add(n),g(this._backdrop).appendTo(document.body),g(this._element).on(re.CLICK_DISMISS,function(t){e._ignoreBackdropClick?e._ignoreBackdropClick=!1:t.target===t.currentTarget&&("static"===e._config.backdrop?e._element.focus():e.hide())}),n&&_.reflow(this._backdrop),g(this._backdrop).addClass(ue),!t)return;if(!n)return void t();var i=_.getTransitionDurationFromElement(this._backdrop);g(this._backdrop).one(_.TRANSITION_END,t).emulateTransitionEnd(i)}else if(!this._isShown&&this._backdrop){g(this._backdrop).removeClass(ue);var o=function(){e._removeBackdrop(),t&&t()};if(g(this._element).hasClass(he)){var r=_.getTransitionDurationFromElement(this._backdrop);g(this._backdrop).one(_.TRANSITION_END,o).emulateTransitionEnd(r)}else o()}else t&&t()},t._adjustDialog=function(){var t=this._element.scrollHeight>document.documentElement.clientHeight;!this._isBodyOverflowing&&t&&(this._element.style.paddingLeft=this._scrollbarWidth+"px"),this._isBodyOverflowing&&!t&&(this._element.style.paddingRight=this._scrollbarWidth+"px")},t._resetAdjustments=function(){this._element.style.paddingLeft="",this._element.style.paddingRight=""},t._checkScrollbar=function(){var t=document.body.getBoundingClientRect();this._isBodyOverflowing=t.left+t.right<window.innerWidth,this._scrollbarWidth=this._getScrollbarWidth()},t._setScrollbar=function(){var o=this;if(this._isBodyOverflowing){var t=[].slice.call(document.querySelectorAll(me)),e=[].slice.call(document.querySelectorAll(pe));g(t).each(function(t,e){var n=e.style.paddingRight,i=g(e).css("padding-right");g(e).data("padding-right",n).css("padding-right",parseFloat(i)+o._scrollbarWidth+"px")}),g(e).each(function(t,e){var n=e.style.marginRight,i=g(e).css("margin-right");g(e).data("margin-right",n).css("margin-right",parseFloat(i)-o._scrollbarWidth+"px")});var n=document.body.style.paddingRight,i=g(document.body).css("padding-right");g(document.body).data("padding-right",n).css("padding-right",parseFloat(i)+this._scrollbarWidth+"px")}g(document.body).addClass(ce)},t._resetScrollbar=function(){var t=[].slice.call(document.querySelectorAll(me));g(t).each(function(t,e){var n=g(e).data("padding-right");g(e).removeData("padding-right"),e.style.paddingRight=n||""});var e=[].slice.call(document.querySelectorAll(""+pe));g(e).each(function(t,e){var n=g(e).data("margin-right");"undefined"!=typeof n&&g(e).css("margin-right",n).removeData("margin-right")});var n=g(document.body).data("padding-right");g(document.body).removeData("padding-right"),document.body.style.paddingRight=n||""},t._getScrollbarWidth=function(){var t=document.createElement("div");t.className=ae,document.body.appendChild(t);var e=t.getBoundingClientRect().width-t.clientWidth;return document.body.removeChild(t),e},o._jQueryInterface=function(n,i){return this.each(function(){var t=g(this).data(te),e=l({},ie,g(this).data(),"object"==typeof n&&n?n:{});if(t||(t=new o(this,e),g(this).data(te,t)),"string"==typeof n){if("undefined"==typeof t[n])throw new TypeError('No method named "'+n+'"');t[n](i)}else e.show&&t.show(i)})},s(o,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"Default",get:function(){return ie}}]),o}();g(document).on(re.CLICK_DATA_API,ge,function(t){var e,n=this,i=_.getSelectorFromElement(this);i&&(e=document.querySelector(i));var o=g(e).data(te)?"toggle":l({},g(e).data(),g(this).data());"A"!==this.tagName&&"AREA"!==this.tagName||t.preventDefault();var r=g(e).one(re.SHOW,function(t){t.isDefaultPrevented()||r.one(re.HIDDEN,function(){g(n).is(":visible")&&n.focus()})});ve._jQueryInterface.call(g(e),o,this)}),g.fn[Zt]=ve._jQueryInterface,g.fn[Zt].Constructor=ve,g.fn[Zt].noConflict=function(){return g.fn[Zt]=ne,ve._jQueryInterface};var ye=["background","cite","href","itemtype","longdesc","poster","src","xlink:href"],Ee={"*":["class","dir","id","lang","role",/^aria-[\w-]*$/i],a:["target","href","title","rel"],area:[],b:[],br:[],col:[],code:[],div:[],em:[],hr:[],h1:[],h2:[],h3:[],h4:[],h5:[],h6:[],i:[],img:["src","alt","title","width","height"],li:[],ol:[],p:[],pre:[],s:[],small:[],span:[],sub:[],sup:[],strong:[],u:[],ul:[]},Ce=/^(?:(?:https?|mailto|ftp|tel|file):|[^&:/?#]*(?:[/?#]|$))/gi,Te=/^data:(?:image\/(?:bmp|gif|jpeg|jpg|png|tiff|webp)|video\/(?:mpeg|mp4|ogg|webm)|audio\/(?:mp3|oga|ogg|opus));base64,[a-z0-9+/]+=*$/i;function Se(t,s,e){if(0===t.length)return t;if(e&&"function"==typeof e)return e(t);for(var n=(new window.DOMParser).parseFromString(t,"text/html"),a=Object.keys(s),l=[].slice.call(n.body.querySelectorAll("*")),i=function(t,e){var n=l[t],i=n.nodeName.toLowerCase();if(-1===a.indexOf(n.nodeName.toLowerCase()))return n.parentNode.removeChild(n),"continue";var o=[].slice.call(n.attributes),r=[].concat(s["*"]||[],s[i]||[]);o.forEach(function(t){(function(t,e){var n=t.nodeName.toLowerCase();if(-1!==e.indexOf(n))return-1===ye.indexOf(n)||Boolean(t.nodeValue.match(Ce)||t.nodeValue.match(Te));for(var i=e.filter(function(t){return t instanceof RegExp}),o=0,r=i.length;o<r;o++)if(n.match(i[o]))return!0;return!1})(t,r)||n.removeAttribute(t.nodeName)})},o=0,r=l.length;o<r;o++)i(o);return n.body.innerHTML}var be="tooltip",Ie="bs.tooltip",De="."+Ie,we=g.fn[be],Ae="bs-tooltip",Ne=new RegExp("(^|\\s)"+Ae+"\\S+","g"),Oe=["sanitize","whiteList","sanitizeFn"],ke={animation:"boolean",template:"string",title:"(string|element|function)",trigger:"string",delay:"(number|object)",html:"boolean",selector:"(string|boolean)",placement:"(string|function)",offset:"(number|string|function)",container:"(string|element|boolean)",fallbackPlacement:"(string|array)",boundary:"(string|element)",sanitize:"boolean",sanitizeFn:"(null|function)",whiteList:"object"},Pe={AUTO:"auto",TOP:"top",RIGHT:"right",BOTTOM:"bottom",LEFT:"left"},Le={animation:!0,template:'<div class="tooltip" role="tooltip"><div class="arrow"></div><div class="tooltip-inner"></div></div>',trigger:"hover focus",title:"",delay:0,html:!1,selector:!1,placement:"top",offset:0,container:!1,fallbackPlacement:"flip",boundary:"scrollParent",sanitize:!0,sanitizeFn:null,whiteList:Ee},je="show",He="out",Re={HIDE:"hide"+De,HIDDEN:"hidden"+De,SHOW:"show"+De,SHOWN:"shown"+De,INSERTED:"inserted"+De,CLICK:"click"+De,FOCUSIN:"focusin"+De,FOCUSOUT:"focusout"+De,MOUSEENTER:"mouseenter"+De,MOUSELEAVE:"mouseleave"+De},xe="fade",Fe="show",Ue=".tooltip-inner",We=".arrow",qe="hover",Me="focus",Ke="click",Qe="manual",Be=function(){function i(t,e){if("undefined"==typeof u)throw new TypeError("Bootstrap's tooltips require Popper.js (https://popper.js.org/)");this._isEnabled=!0,this._timeout=0,this._hoverState="",this._activeTrigger={},this._popper=null,this.element=t,this.config=this._getConfig(e),this.tip=null,this._setListeners()}var t=i.prototype;return t.enable=function(){this._isEnabled=!0},t.disable=function(){this._isEnabled=!1},t.toggleEnabled=function(){this._isEnabled=!this._isEnabled},t.toggle=function(t){if(this._isEnabled)if(t){var e=this.constructor.DATA_KEY,n=g(t.currentTarget).data(e);n||(n=new this.constructor(t.currentTarget,this._getDelegateConfig()),g(t.currentTarget).data(e,n)),n._activeTrigger.click=!n._activeTrigger.click,n._isWithActiveTrigger()?n._enter(null,n):n._leave(null,n)}else{if(g(this.getTipElement()).hasClass(Fe))return void this._leave(null,this);this._enter(null,this)}},t.dispose=function(){clearTimeout(this._timeout),g.removeData(this.element,this.constructor.DATA_KEY),g(this.element).off(this.constructor.EVENT_KEY),g(this.element).closest(".modal").off("hide.bs.modal"),this.tip&&g(this.tip).remove(),this._isEnabled=null,this._timeout=null,this._hoverState=null,(this._activeTrigger=null)!==this._popper&&this._popper.destroy(),this._popper=null,this.element=null,this.config=null,this.tip=null},t.show=function(){var e=this;if("none"===g(this.element).css("display"))throw new Error("Please use show on visible elements");var t=g.Event(this.constructor.Event.SHOW);if(this.isWithContent()&&this._isEnabled){g(this.element).trigger(t);var n=_.findShadowRoot(this.element),i=g.contains(null!==n?n:this.element.ownerDocument.documentElement,this.element);if(t.isDefaultPrevented()||!i)return;var o=this.getTipElement(),r=_.getUID(this.constructor.NAME);o.setAttribute("id",r),this.element.setAttribute("aria-describedby",r),this.setContent(),this.config.animation&&g(o).addClass(xe);var s="function"==typeof this.config.placement?this.config.placement.call(this,o,this.element):this.config.placement,a=this._getAttachment(s);this.addAttachmentClass(a);var l=this._getContainer();g(o).data(this.constructor.DATA_KEY,this),g.contains(this.element.ownerDocument.documentElement,this.tip)||g(o).appendTo(l),g(this.element).trigger(this.constructor.Event.INSERTED),this._popper=new u(this.element,o,{placement:a,modifiers:{offset:this._getOffset(),flip:{behavior:this.config.fallbackPlacement},arrow:{element:We},preventOverflow:{boundariesElement:this.config.boundary}},onCreate:function(t){t.originalPlacement!==t.placement&&e._handlePopperPlacementChange(t)},onUpdate:function(t){return e._handlePopperPlacementChange(t)}}),g(o).addClass(Fe),"ontouchstart"in document.documentElement&&g(document.body).children().on("mouseover",null,g.noop);var c=function(){e.config.animation&&e._fixTransition();var t=e._hoverState;e._hoverState=null,g(e.element).trigger(e.constructor.Event.SHOWN),t===He&&e._leave(null,e)};if(g(this.tip).hasClass(xe)){var h=_.getTransitionDurationFromElement(this.tip);g(this.tip).one(_.TRANSITION_END,c).emulateTransitionEnd(h)}else c()}},t.hide=function(t){var e=this,n=this.getTipElement(),i=g.Event(this.constructor.Event.HIDE),o=function(){e._hoverState!==je&&n.parentNode&&n.parentNode.removeChild(n),e._cleanTipClass(),e.element.removeAttribute("aria-describedby"),g(e.element).trigger(e.constructor.Event.HIDDEN),null!==e._popper&&e._popper.destroy(),t&&t()};if(g(this.element).trigger(i),!i.isDefaultPrevented()){if(g(n).removeClass(Fe),"ontouchstart"in document.documentElement&&g(document.body).children().off("mouseover",null,g.noop),this._activeTrigger[Ke]=!1,this._activeTrigger[Me]=!1,this._activeTrigger[qe]=!1,g(this.tip).hasClass(xe)){var r=_.getTransitionDurationFromElement(n);g(n).one(_.TRANSITION_END,o).emulateTransitionEnd(r)}else o();this._hoverState=""}},t.update=function(){null!==this._popper&&this._popper.scheduleUpdate()},t.isWithContent=function(){return Boolean(this.getTitle())},t.addAttachmentClass=function(t){g(this.getTipElement()).addClass(Ae+"-"+t)},t.getTipElement=function(){return this.tip=this.tip||g(this.config.template)[0],this.tip},t.setContent=function(){var t=this.getTipElement();this.setElementContent(g(t.querySelectorAll(Ue)),this.getTitle()),g(t).removeClass(xe+" "+Fe)},t.setElementContent=function(t,e){"object"!=typeof e||!e.nodeType&&!e.jquery?this.config.html?(this.config.sanitize&&(e=Se(e,this.config.whiteList,this.config.sanitizeFn)),t.html(e)):t.text(e):this.config.html?g(e).parent().is(t)||t.empty().append(e):t.text(g(e).text())},t.getTitle=function(){var t=this.element.getAttribute("data-original-title");return t||(t="function"==typeof this.config.title?this.config.title.call(this.element):this.config.title),t},t._getOffset=function(){var e=this,t={};return"function"==typeof this.config.offset?t.fn=function(t){return t.offsets=l({},t.offsets,e.config.offset(t.offsets,e.element)||{}),t}:t.offset=this.config.offset,t},t._getContainer=function(){return!1===this.config.container?document.body:_.isElement(this.config.container)?g(this.config.container):g(document).find(this.config.container)},t._getAttachment=function(t){return Pe[t.toUpperCase()]},t._setListeners=function(){var i=this;this.config.trigger.split(" ").forEach(function(t){if("click"===t)g(i.element).on(i.constructor.Event.CLICK,i.config.selector,function(t){return i.toggle(t)});else if(t!==Qe){var e=t===qe?i.constructor.Event.MOUSEENTER:i.constructor.Event.FOCUSIN,n=t===qe?i.constructor.Event.MOUSELEAVE:i.constructor.Event.FOCUSOUT;g(i.element).on(e,i.config.selector,function(t){return i._enter(t)}).on(n,i.config.selector,function(t){return i._leave(t)})}}),g(this.element).closest(".modal").on("hide.bs.modal",function(){i.element&&i.hide()}),this.config.selector?this.config=l({},this.config,{trigger:"manual",selector:""}):this._fixTitle()},t._fixTitle=function(){var t=typeof this.element.getAttribute("data-original-title");(this.element.getAttribute("title")||"string"!==t)&&(this.element.setAttribute("data-original-title",this.element.getAttribute("title")||""),this.element.setAttribute("title",""))},t._enter=function(t,e){var n=this.constructor.DATA_KEY;(e=e||g(t.currentTarget).data(n))||(e=new this.constructor(t.currentTarget,this._getDelegateConfig()),g(t.currentTarget).data(n,e)),t&&(e._activeTrigger["focusin"===t.type?Me:qe]=!0),g(e.getTipElement()).hasClass(Fe)||e._hoverState===je?e._hoverState=je:(clearTimeout(e._timeout),e._hoverState=je,e.config.delay&&e.config.delay.show?e._timeout=setTimeout(function(){e._hoverState===je&&e.show()},e.config.delay.show):e.show())},t._leave=function(t,e){var n=this.constructor.DATA_KEY;(e=e||g(t.currentTarget).data(n))||(e=new this.constructor(t.currentTarget,this._getDelegateConfig()),g(t.currentTarget).data(n,e)),t&&(e._activeTrigger["focusout"===t.type?Me:qe]=!1),e._isWithActiveTrigger()||(clearTimeout(e._timeout),e._hoverState=He,e.config.delay&&e.config.delay.hide?e._timeout=setTimeout(function(){e._hoverState===He&&e.hide()},e.config.delay.hide):e.hide())},t._isWithActiveTrigger=function(){for(var t in this._activeTrigger)if(this._activeTrigger[t])return!0;return!1},t._getConfig=function(t){var e=g(this.element).data();return Object.keys(e).forEach(function(t){-1!==Oe.indexOf(t)&&delete e[t]}),"number"==typeof(t=l({},this.constructor.Default,e,"object"==typeof t&&t?t:{})).delay&&(t.delay={show:t.delay,hide:t.delay}),"number"==typeof t.title&&(t.title=t.title.toString()),"number"==typeof t.content&&(t.content=t.content.toString()),_.typeCheckConfig(be,t,this.constructor.DefaultType),t.sanitize&&(t.template=Se(t.template,t.whiteList,t.sanitizeFn)),t},t._getDelegateConfig=function(){var t={};if(this.config)for(var e in this.config)this.constructor.Default[e]!==this.config[e]&&(t[e]=this.config[e]);return t},t._cleanTipClass=function(){var t=g(this.getTipElement()),e=t.attr("class").match(Ne);null!==e&&e.length&&t.removeClass(e.join(""))},t._handlePopperPlacementChange=function(t){var e=t.instance;this.tip=e.popper,this._cleanTipClass(),this.addAttachmentClass(this._getAttachment(t.placement))},t._fixTransition=function(){var t=this.getTipElement(),e=this.config.animation;null===t.getAttribute("x-placement")&&(g(t).removeClass(xe),this.config.animation=!1,this.hide(),this.show(),this.config.animation=e)},i._jQueryInterface=function(n){return this.each(function(){var t=g(this).data(Ie),e="object"==typeof n&&n;if((t||!/dispose|hide/.test(n))&&(t||(t=new i(this,e),g(this).data(Ie,t)),"string"==typeof n)){if("undefined"==typeof t[n])throw new TypeError('No method named "'+n+'"');t[n]()}})},s(i,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"Default",get:function(){return Le}},{key:"NAME",get:function(){return be}},{key:"DATA_KEY",get:function(){return Ie}},{key:"Event",get:function(){return Re}},{key:"EVENT_KEY",get:function(){return De}},{key:"DefaultType",get:function(){return ke}}]),i}();g.fn[be]=Be._jQueryInterface,g.fn[be].Constructor=Be,g.fn[be].noConflict=function(){return g.fn[be]=we,Be._jQueryInterface};var Ve="popover",Ye="bs.popover",ze="."+Ye,Xe=g.fn[Ve],$e="bs-popover",Ge=new RegExp("(^|\\s)"+$e+"\\S+","g"),Je=l({},Be.Default,{placement:"right",trigger:"click",content:"",template:'<div class="popover" role="tooltip"><div class="arrow"></div><h3 class="popover-header"></h3><div class="popover-body"></div></div>'}),Ze=l({},Be.DefaultType,{content:"(string|element|function)"}),tn="fade",en="show",nn=".popover-header",on=".popover-body",rn={HIDE:"hide"+ze,HIDDEN:"hidden"+ze,SHOW:"show"+ze,SHOWN:"shown"+ze,INSERTED:"inserted"+ze,CLICK:"click"+ze,FOCUSIN:"focusin"+ze,FOCUSOUT:"focusout"+ze,MOUSEENTER:"mouseenter"+ze,MOUSELEAVE:"mouseleave"+ze},sn=function(t){var e,n;function i(){return t.apply(this,arguments)||this}n=t,(e=i).prototype=Object.create(n.prototype),(e.prototype.constructor=e).__proto__=n;var o=i.prototype;return o.isWithContent=function(){return this.getTitle()||this._getContent()},o.addAttachmentClass=function(t){g(this.getTipElement()).addClass($e+"-"+t)},o.getTipElement=function(){return this.tip=this.tip||g(this.config.template)[0],this.tip},o.setContent=function(){var t=g(this.getTipElement());this.setElementContent(t.find(nn),this.getTitle());var e=this._getContent();"function"==typeof e&&(e=e.call(this.element)),this.setElementContent(t.find(on),e),t.removeClass(tn+" "+en)},o._getContent=function(){return this.element.getAttribute("data-content")||this.config.content},o._cleanTipClass=function(){var t=g(this.getTipElement()),e=t.attr("class").match(Ge);null!==e&&0<e.length&&t.removeClass(e.join(""))},i._jQueryInterface=function(n){return this.each(function(){var t=g(this).data(Ye),e="object"==typeof n?n:null;if((t||!/dispose|hide/.test(n))&&(t||(t=new i(this,e),g(this).data(Ye,t)),"string"==typeof n)){if("undefined"==typeof t[n])throw new TypeError('No method named "'+n+'"');t[n]()}})},s(i,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"Default",get:function(){return Je}},{key:"NAME",get:function(){return Ve}},{key:"DATA_KEY",get:function(){return Ye}},{key:"Event",get:function(){return rn}},{key:"EVENT_KEY",get:function(){return ze}},{key:"DefaultType",get:function(){return Ze}}]),i}(Be);g.fn[Ve]=sn._jQueryInterface,g.fn[Ve].Constructor=sn,g.fn[Ve].noConflict=function(){return g.fn[Ve]=Xe,sn._jQueryInterface};var an="scrollspy",ln="bs.scrollspy",cn="."+ln,hn=g.fn[an],un={offset:10,method:"auto",target:""},fn={offset:"number",method:"string",target:"(string|element)"},dn={ACTIVATE:"activate"+cn,SCROLL:"scroll"+cn,LOAD_DATA_API:"load"+cn+".data-api"},gn="dropdown-item",_n="active",mn='[data-spy="scroll"]',pn=".nav, .list-group",vn=".nav-link",yn=".nav-item",En=".list-group-item",Cn=".dropdown",Tn=".dropdown-item",Sn=".dropdown-toggle",bn="offset",In="position",Dn=function(){function n(t,e){var n=this;this._element=t,this._scrollElement="BODY"===t.tagName?window:t,this._config=this._getConfig(e),this._selector=this._config.target+" "+vn+","+this._config.target+" "+En+","+this._config.target+" "+Tn,this._offsets=[],this._targets=[],this._activeTarget=null,this._scrollHeight=0,g(this._scrollElement).on(dn.SCROLL,function(t){return n._process(t)}),this.refresh(),this._process()}var t=n.prototype;return t.refresh=function(){var e=this,t=this._scrollElement===this._scrollElement.window?bn:In,o="auto"===this._config.method?t:this._config.method,r=o===In?this._getScrollTop():0;this._offsets=[],this._targets=[],this._scrollHeight=this._getScrollHeight(),[].slice.call(document.querySelectorAll(this._selector)).map(function(t){var e,n=_.getSelectorFromElement(t);if(n&&(e=document.querySelector(n)),e){var i=e.getBoundingClientRect();if(i.width||i.height)return[g(e)[o]().top+r,n]}return null}).filter(function(t){return t}).sort(function(t,e){return t[0]-e[0]}).forEach(function(t){e._offsets.push(t[0]),e._targets.push(t[1])})},t.dispose=function(){g.removeData(this._element,ln),g(this._scrollElement).off(cn),this._element=null,this._scrollElement=null,this._config=null,this._selector=null,this._offsets=null,this._targets=null,this._activeTarget=null,this._scrollHeight=null},t._getConfig=function(t){if("string"!=typeof(t=l({},un,"object"==typeof t&&t?t:{})).target){var e=g(t.target).attr("id");e||(e=_.getUID(an),g(t.target).attr("id",e)),t.target="#"+e}return _.typeCheckConfig(an,t,fn),t},t._getScrollTop=function(){return this._scrollElement===window?this._scrollElement.pageYOffset:this._scrollElement.scrollTop},t._getScrollHeight=function(){return this._scrollElement.scrollHeight||Math.max(document.body.scrollHeight,document.documentElement.scrollHeight)},t._getOffsetHeight=function(){return this._scrollElement===window?window.innerHeight:this._scrollElement.getBoundingClientRect().height},t._process=function(){var t=this._getScrollTop()+this._config.offset,e=this._getScrollHeight(),n=this._config.offset+e-this._getOffsetHeight();if(this._scrollHeight!==e&&this.refresh(),n<=t){var i=this._targets[this._targets.length-1];this._activeTarget!==i&&this._activate(i)}else{if(this._activeTarget&&t<this._offsets[0]&&0<this._offsets[0])return this._activeTarget=null,void this._clear();for(var o=this._offsets.length;o--;){this._activeTarget!==this._targets[o]&&t>=this._offsets[o]&&("undefined"==typeof this._offsets[o+1]||t<this._offsets[o+1])&&this._activate(this._targets[o])}}},t._activate=function(e){this._activeTarget=e,this._clear();var t=this._selector.split(",").map(function(t){return t+'[data-target="'+e+'"],'+t+'[href="'+e+'"]'}),n=g([].slice.call(document.querySelectorAll(t.join(","))));n.hasClass(gn)?(n.closest(Cn).find(Sn).addClass(_n),n.addClass(_n)):(n.addClass(_n),n.parents(pn).prev(vn+", "+En).addClass(_n),n.parents(pn).prev(yn).children(vn).addClass(_n)),g(this._scrollElement).trigger(dn.ACTIVATE,{relatedTarget:e})},t._clear=function(){[].slice.call(document.querySelectorAll(this._selector)).filter(function(t){return t.classList.contains(_n)}).forEach(function(t){return t.classList.remove(_n)})},n._jQueryInterface=function(e){return this.each(function(){var t=g(this).data(ln);if(t||(t=new n(this,"object"==typeof e&&e),g(this).data(ln,t)),"string"==typeof e){if("undefined"==typeof t[e])throw new TypeError('No method named "'+e+'"');t[e]()}})},s(n,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"Default",get:function(){return un}}]),n}();g(window).on(dn.LOAD_DATA_API,function(){for(var t=[].slice.call(document.querySelectorAll(mn)),e=t.length;e--;){var n=g(t[e]);Dn._jQueryInterface.call(n,n.data())}}),g.fn[an]=Dn._jQueryInterface,g.fn[an].Constructor=Dn,g.fn[an].noConflict=function(){return g.fn[an]=hn,Dn._jQueryInterface};var wn="bs.tab",An="."+wn,Nn=g.fn.tab,On={HIDE:"hide"+An,HIDDEN:"hidden"+An,SHOW:"show"+An,SHOWN:"shown"+An,CLICK_DATA_API:"click"+An+".data-api"},kn="dropdown-menu",Pn="active",Ln="disabled",jn="fade",Hn="show",Rn=".dropdown",xn=".nav, .list-group",Fn=".active",Un="> li > .active",Wn='[data-toggle="tab"], [data-toggle="pill"], [data-toggle="list"]',qn=".dropdown-toggle",Mn="> .dropdown-menu .active",Kn=function(){function i(t){this._element=t}var t=i.prototype;return t.show=function(){var n=this;if(!(this._element.parentNode&&this._element.parentNode.nodeType===Node.ELEMENT_NODE&&g(this._element).hasClass(Pn)||g(this._element).hasClass(Ln))){var t,i,e=g(this._element).closest(xn)[0],o=_.getSelectorFromElement(this._element);if(e){var r="UL"===e.nodeName||"OL"===e.nodeName?Un:Fn;i=(i=g.makeArray(g(e).find(r)))[i.length-1]}var s=g.Event(On.HIDE,{relatedTarget:this._element}),a=g.Event(On.SHOW,{relatedTarget:i});if(i&&g(i).trigger(s),g(this._element).trigger(a),!a.isDefaultPrevented()&&!s.isDefaultPrevented()){o&&(t=document.querySelector(o)),this._activate(this._element,e);var l=function(){var t=g.Event(On.HIDDEN,{relatedTarget:n._element}),e=g.Event(On.SHOWN,{relatedTarget:i});g(i).trigger(t),g(n._element).trigger(e)};t?this._activate(t,t.parentNode,l):l()}}},t.dispose=function(){g.removeData(this._element,wn),this._element=null},t._activate=function(t,e,n){var i=this,o=(!e||"UL"!==e.nodeName&&"OL"!==e.nodeName?g(e).children(Fn):g(e).find(Un))[0],r=n&&o&&g(o).hasClass(jn),s=function(){return i._transitionComplete(t,o,n)};if(o&&r){var a=_.getTransitionDurationFromElement(o);g(o).removeClass(Hn).one(_.TRANSITION_END,s).emulateTransitionEnd(a)}else s()},t._transitionComplete=function(t,e,n){if(e){g(e).removeClass(Pn);var i=g(e.parentNode).find(Mn)[0];i&&g(i).removeClass(Pn),"tab"===e.getAttribute("role")&&e.setAttribute("aria-selected",!1)}if(g(t).addClass(Pn),"tab"===t.getAttribute("role")&&t.setAttribute("aria-selected",!0),_.reflow(t),t.classList.contains(jn)&&t.classList.add(Hn),t.parentNode&&g(t.parentNode).hasClass(kn)){var o=g(t).closest(Rn)[0];if(o){var r=[].slice.call(o.querySelectorAll(qn));g(r).addClass(Pn)}t.setAttribute("aria-expanded",!0)}n&&n()},i._jQueryInterface=function(n){return this.each(function(){var t=g(this),e=t.data(wn);if(e||(e=new i(this),t.data(wn,e)),"string"==typeof n){if("undefined"==typeof e[n])throw new TypeError('No method named "'+n+'"');e[n]()}})},s(i,null,[{key:"VERSION",get:function(){return"4.3.1"}}]),i}();g(document).on(On.CLICK_DATA_API,Wn,function(t){t.preventDefault(),Kn._jQueryInterface.call(g(this),"show")}),g.fn.tab=Kn._jQueryInterface,g.fn.tab.Constructor=Kn,g.fn.tab.noConflict=function(){return g.fn.tab=Nn,Kn._jQueryInterface};var Qn="toast",Bn="bs.toast",Vn="."+Bn,Yn=g.fn[Qn],zn={CLICK_DISMISS:"click.dismiss"+Vn,HIDE:"hide"+Vn,HIDDEN:"hidden"+Vn,SHOW:"show"+Vn,SHOWN:"shown"+Vn},Xn="fade",$n="hide",Gn="show",Jn="showing",Zn={animation:"boolean",autohide:"boolean",delay:"number"},ti={animation:!0,autohide:!0,delay:500},ei='[data-dismiss="toast"]',ni=function(){function i(t,e){this._element=t,this._config=this._getConfig(e),this._timeout=null,this._setListeners()}var t=i.prototype;return t.show=function(){var t=this;g(this._element).trigger(zn.SHOW),this._config.animation&&this._element.classList.add(Xn);var e=function(){t._element.classList.remove(Jn),t._element.classList.add(Gn),g(t._element).trigger(zn.SHOWN),t._config.autohide&&t.hide()};if(this._element.classList.remove($n),this._element.classList.add(Jn),this._config.animation){var n=_.getTransitionDurationFromElement(this._element);g(this._element).one(_.TRANSITION_END,e).emulateTransitionEnd(n)}else e()},t.hide=function(t){var e=this;this._element.classList.contains(Gn)&&(g(this._element).trigger(zn.HIDE),t?this._close():this._timeout=setTimeout(function(){e._close()},this._config.delay))},t.dispose=function(){clearTimeout(this._timeout),this._timeout=null,this._element.classList.contains(Gn)&&this._element.classList.remove(Gn),g(this._element).off(zn.CLICK_DISMISS),g.removeData(this._element,Bn),this._element=null,this._config=null},t._getConfig=function(t){return t=l({},ti,g(this._element).data(),"object"==typeof t&&t?t:{}),_.typeCheckConfig(Qn,t,this.constructor.DefaultType),t},t._setListeners=function(){var t=this;g(this._element).on(zn.CLICK_DISMISS,ei,function(){return t.hide(!0)})},t._close=function(){var t=this,e=function(){t._element.classList.add($n),g(t._element).trigger(zn.HIDDEN)};if(this._element.classList.remove(Gn),this._config.animation){var n=_.getTransitionDurationFromElement(this._element);g(this._element).one(_.TRANSITION_END,e).emulateTransitionEnd(n)}else e()},i._jQueryInterface=function(n){return this.each(function(){var t=g(this),e=t.data(Bn);if(e||(e=new i(this,"object"==typeof n&&n),t.data(Bn,e)),"string"==typeof n){if("undefined"==typeof e[n])throw new TypeError('No method named "'+n+'"');e[n](this)}})},s(i,null,[{key:"VERSION",get:function(){return"4.3.1"}},{key:"DefaultType",get:function(){return Zn}},{key:"Default",get:function(){return ti}}]),i}();g.fn[Qn]=ni._jQueryInterface,g.fn[Qn].Constructor=ni,g.fn[Qn].noConflict=function(){return g.fn[Qn]=Yn,ni._jQueryInterface},function(){if("undefined"==typeof g)throw new TypeError("Bootstrap's JavaScript requires jQuery. jQuery must be included before Bootstrap's JavaScript.");var t=g.fn.jquery.split(" ")[0].split(".");if(t[0]<2&&t[1]<9||1===t[0]&&9===t[1]&&t[2]<1||4<=t[0])throw new Error("Bootstrap's JavaScript requires at least jQuery v1.9.1 but less than v4.0.0")}(),t.Util=_,t.Alert=p,t.Button=P,t.Carousel=lt,t.Collapse=bt,t.Dropdown=Jt,t.Modal=ve,t.Popover=sn,t.Scrollspy=Dn,t.Tab=Kn,t.Toast=ni,t.Tooltip=Be,Object.defineProperty(t,"__esModule",{value:!0})});
+//# sourceMappingURL=bootstrap.min.js.map
\ No newline at end of file
diff --git a/0.16/_static/js/vendor/popper.min.js b/0.16/_static/js/vendor/popper.min.js
new file mode 100644
index 00000000000..bb1aaae3e86
--- /dev/null
+++ b/0.16/_static/js/vendor/popper.min.js
@@ -0,0 +1,5 @@
+/*
+ Copyright (C) Federico Zivolo 2020
+ Distributed under the MIT License (license terms are at http://opensource.org/licenses/MIT).
+ */(function(e,t){'object'==typeof exports&&'undefined'!=typeof module?module.exports=t():'function'==typeof define&&define.amd?define(t):e.Popper=t()})(this,function(){'use strict';function e(e){return e&&'[object Function]'==={}.toString.call(e)}function t(e,t){if(1!==e.nodeType)return[];var o=e.ownerDocument.defaultView,n=o.getComputedStyle(e,null);return t?n[t]:n}function o(e){return'HTML'===e.nodeName?e:e.parentNode||e.host}function n(e){if(!e)return document.body;switch(e.nodeName){case'HTML':case'BODY':return e.ownerDocument.body;case'#document':return e.body;}var i=t(e),r=i.overflow,p=i.overflowX,s=i.overflowY;return /(auto|scroll|overlay)/.test(r+s+p)?e:n(o(e))}function i(e){return e&&e.referenceNode?e.referenceNode:e}function r(e){return 11===e?re:10===e?pe:re||pe}function p(e){if(!e)return document.documentElement;for(var o=r(10)?document.body:null,n=e.offsetParent||null;n===o&&e.nextElementSibling;)n=(e=e.nextElementSibling).offsetParent;var i=n&&n.nodeName;return i&&'BODY'!==i&&'HTML'!==i?-1!==['TH','TD','TABLE'].indexOf(n.nodeName)&&'static'===t(n,'position')?p(n):n:e?e.ownerDocument.documentElement:document.documentElement}function s(e){var t=e.nodeName;return'BODY'!==t&&('HTML'===t||p(e.firstElementChild)===e)}function d(e){return null===e.parentNode?e:d(e.parentNode)}function a(e,t){if(!e||!e.nodeType||!t||!t.nodeType)return document.documentElement;var o=e.compareDocumentPosition(t)&Node.DOCUMENT_POSITION_FOLLOWING,n=o?e:t,i=o?t:e,r=document.createRange();r.setStart(n,0),r.setEnd(i,0);var l=r.commonAncestorContainer;if(e!==l&&t!==l||n.contains(i))return s(l)?l:p(l);var f=d(e);return f.host?a(f.host,t):a(e,d(t).host)}function l(e){var t=1<arguments.length&&void 0!==arguments[1]?arguments[1]:'top',o='top'===t?'scrollTop':'scrollLeft',n=e.nodeName;if('BODY'===n||'HTML'===n){var i=e.ownerDocument.documentElement,r=e.ownerDocument.scrollingElement||i;return r[o]}return e[o]}function f(e,t){var o=2<arguments.length&&void 0!==arguments[2]&&arguments[2],n=l(t,'top'),i=l(t,'left'),r=o?-1:1;return e.top+=n*r,e.bottom+=n*r,e.left+=i*r,e.right+=i*r,e}function m(e,t){var o='x'===t?'Left':'Top',n='Left'==o?'Right':'Bottom';return parseFloat(e['border'+o+'Width'])+parseFloat(e['border'+n+'Width'])}function h(e,t,o,n){return ee(t['offset'+e],t['scroll'+e],o['client'+e],o['offset'+e],o['scroll'+e],r(10)?parseInt(o['offset'+e])+parseInt(n['margin'+('Height'===e?'Top':'Left')])+parseInt(n['margin'+('Height'===e?'Bottom':'Right')]):0)}function c(e){var t=e.body,o=e.documentElement,n=r(10)&&getComputedStyle(o);return{height:h('Height',t,o,n),width:h('Width',t,o,n)}}function g(e){return le({},e,{right:e.left+e.width,bottom:e.top+e.height})}function u(e){var o={};try{if(r(10)){o=e.getBoundingClientRect();var n=l(e,'top'),i=l(e,'left');o.top+=n,o.left+=i,o.bottom+=n,o.right+=i}else o=e.getBoundingClientRect()}catch(t){}var p={left:o.left,top:o.top,width:o.right-o.left,height:o.bottom-o.top},s='HTML'===e.nodeName?c(e.ownerDocument):{},d=s.width||e.clientWidth||p.width,a=s.height||e.clientHeight||p.height,f=e.offsetWidth-d,h=e.offsetHeight-a;if(f||h){var u=t(e);f-=m(u,'x'),h-=m(u,'y'),p.width-=f,p.height-=h}return g(p)}function b(e,o){var i=2<arguments.length&&void 0!==arguments[2]&&arguments[2],p=r(10),s='HTML'===o.nodeName,d=u(e),a=u(o),l=n(e),m=t(o),h=parseFloat(m.borderTopWidth),c=parseFloat(m.borderLeftWidth);i&&s&&(a.top=ee(a.top,0),a.left=ee(a.left,0));var b=g({top:d.top-a.top-h,left:d.left-a.left-c,width:d.width,height:d.height});if(b.marginTop=0,b.marginLeft=0,!p&&s){var w=parseFloat(m.marginTop),y=parseFloat(m.marginLeft);b.top-=h-w,b.bottom-=h-w,b.left-=c-y,b.right-=c-y,b.marginTop=w,b.marginLeft=y}return(p&&!i?o.contains(l):o===l&&'BODY'!==l.nodeName)&&(b=f(b,o)),b}function w(e){var t=1<arguments.length&&void 0!==arguments[1]&&arguments[1],o=e.ownerDocument.documentElement,n=b(e,o),i=ee(o.clientWidth,window.innerWidth||0),r=ee(o.clientHeight,window.innerHeight||0),p=t?0:l(o),s=t?0:l(o,'left'),d={top:p-n.top+n.marginTop,left:s-n.left+n.marginLeft,width:i,height:r};return g(d)}function y(e){var n=e.nodeName;if('BODY'===n||'HTML'===n)return!1;if('fixed'===t(e,'position'))return!0;var i=o(e);return!!i&&y(i)}function E(e){if(!e||!e.parentElement||r())return document.documentElement;for(var o=e.parentElement;o&&'none'===t(o,'transform');)o=o.parentElement;return o||document.documentElement}function v(e,t,r,p){var s=4<arguments.length&&void 0!==arguments[4]&&arguments[4],d={top:0,left:0},l=s?E(e):a(e,i(t));if('viewport'===p)d=w(l,s);else{var f;'scrollParent'===p?(f=n(o(t)),'BODY'===f.nodeName&&(f=e.ownerDocument.documentElement)):'window'===p?f=e.ownerDocument.documentElement:f=p;var m=b(f,l,s);if('HTML'===f.nodeName&&!y(l)){var h=c(e.ownerDocument),g=h.height,u=h.width;d.top+=m.top-m.marginTop,d.bottom=g+m.top,d.left+=m.left-m.marginLeft,d.right=u+m.left}else d=m}r=r||0;var v='number'==typeof r;return d.left+=v?r:r.left||0,d.top+=v?r:r.top||0,d.right-=v?r:r.right||0,d.bottom-=v?r:r.bottom||0,d}function x(e){var t=e.width,o=e.height;return t*o}function O(e,t,o,n,i){var r=5<arguments.length&&void 0!==arguments[5]?arguments[5]:0;if(-1===e.indexOf('auto'))return e;var p=v(o,n,r,i),s={top:{width:p.width,height:t.top-p.top},right:{width:p.right-t.right,height:p.height},bottom:{width:p.width,height:p.bottom-t.bottom},left:{width:t.left-p.left,height:p.height}},d=Object.keys(s).map(function(e){return le({key:e},s[e],{area:x(s[e])})}).sort(function(e,t){return t.area-e.area}),a=d.filter(function(e){var t=e.width,n=e.height;return t>=o.clientWidth&&n>=o.clientHeight}),l=0<a.length?a[0].key:d[0].key,f=e.split('-')[1];return l+(f?'-'+f:'')}function L(e,t,o){var n=3<arguments.length&&void 0!==arguments[3]?arguments[3]:null,r=n?E(t):a(t,i(o));return b(o,r,n)}function S(e){var t=e.ownerDocument.defaultView,o=t.getComputedStyle(e),n=parseFloat(o.marginTop||0)+parseFloat(o.marginBottom||0),i=parseFloat(o.marginLeft||0)+parseFloat(o.marginRight||0),r={width:e.offsetWidth+i,height:e.offsetHeight+n};return r}function T(e){var t={left:'right',right:'left',bottom:'top',top:'bottom'};return e.replace(/left|right|bottom|top/g,function(e){return t[e]})}function C(e,t,o){o=o.split('-')[0];var n=S(e),i={width:n.width,height:n.height},r=-1!==['right','left'].indexOf(o),p=r?'top':'left',s=r?'left':'top',d=r?'height':'width',a=r?'width':'height';return i[p]=t[p]+t[d]/2-n[d]/2,i[s]=o===s?t[s]-n[a]:t[T(s)],i}function D(e,t){return Array.prototype.find?e.find(t):e.filter(t)[0]}function N(e,t,o){if(Array.prototype.findIndex)return e.findIndex(function(e){return e[t]===o});var n=D(e,function(e){return e[t]===o});return e.indexOf(n)}function P(t,o,n){var i=void 0===n?t:t.slice(0,N(t,'name',n));return i.forEach(function(t){t['function']&&console.warn('`modifier.function` is deprecated, use `modifier.fn`!');var n=t['function']||t.fn;t.enabled&&e(n)&&(o.offsets.popper=g(o.offsets.popper),o.offsets.reference=g(o.offsets.reference),o=n(o,t))}),o}function k(){if(!this.state.isDestroyed){var e={instance:this,styles:{},arrowStyles:{},attributes:{},flipped:!1,offsets:{}};e.offsets.reference=L(this.state,this.popper,this.reference,this.options.positionFixed),e.placement=O(this.options.placement,e.offsets.reference,this.popper,this.reference,this.options.modifiers.flip.boundariesElement,this.options.modifiers.flip.padding),e.originalPlacement=e.placement,e.positionFixed=this.options.positionFixed,e.offsets.popper=C(this.popper,e.offsets.reference,e.placement),e.offsets.popper.position=this.options.positionFixed?'fixed':'absolute',e=P(this.modifiers,e),this.state.isCreated?this.options.onUpdate(e):(this.state.isCreated=!0,this.options.onCreate(e))}}function W(e,t){return e.some(function(e){var o=e.name,n=e.enabled;return n&&o===t})}function B(e){for(var t=[!1,'ms','Webkit','Moz','O'],o=e.charAt(0).toUpperCase()+e.slice(1),n=0;n<t.length;n++){var i=t[n],r=i?''+i+o:e;if('undefined'!=typeof document.body.style[r])return r}return null}function H(){return this.state.isDestroyed=!0,W(this.modifiers,'applyStyle')&&(this.popper.removeAttribute('x-placement'),this.popper.style.position='',this.popper.style.top='',this.popper.style.left='',this.popper.style.right='',this.popper.style.bottom='',this.popper.style.willChange='',this.popper.style[B('transform')]=''),this.disableEventListeners(),this.options.removeOnDestroy&&this.popper.parentNode.removeChild(this.popper),this}function A(e){var t=e.ownerDocument;return t?t.defaultView:window}function M(e,t,o,i){var r='BODY'===e.nodeName,p=r?e.ownerDocument.defaultView:e;p.addEventListener(t,o,{passive:!0}),r||M(n(p.parentNode),t,o,i),i.push(p)}function F(e,t,o,i){o.updateBound=i,A(e).addEventListener('resize',o.updateBound,{passive:!0});var r=n(e);return M(r,'scroll',o.updateBound,o.scrollParents),o.scrollElement=r,o.eventsEnabled=!0,o}function I(){this.state.eventsEnabled||(this.state=F(this.reference,this.options,this.state,this.scheduleUpdate))}function R(e,t){return A(e).removeEventListener('resize',t.updateBound),t.scrollParents.forEach(function(e){e.removeEventListener('scroll',t.updateBound)}),t.updateBound=null,t.scrollParents=[],t.scrollElement=null,t.eventsEnabled=!1,t}function U(){this.state.eventsEnabled&&(cancelAnimationFrame(this.scheduleUpdate),this.state=R(this.reference,this.state))}function Y(e){return''!==e&&!isNaN(parseFloat(e))&&isFinite(e)}function V(e,t){Object.keys(t).forEach(function(o){var n='';-1!==['width','height','top','right','bottom','left'].indexOf(o)&&Y(t[o])&&(n='px'),e.style[o]=t[o]+n})}function j(e,t){Object.keys(t).forEach(function(o){var n=t[o];!1===n?e.removeAttribute(o):e.setAttribute(o,t[o])})}function q(e,t){var o=e.offsets,n=o.popper,i=o.reference,r=$,p=function(e){return e},s=r(i.width),d=r(n.width),a=-1!==['left','right'].indexOf(e.placement),l=-1!==e.placement.indexOf('-'),f=t?a||l||s%2==d%2?r:Z:p,m=t?r:p;return{left:f(1==s%2&&1==d%2&&!l&&t?n.left-1:n.left),top:m(n.top),bottom:m(n.bottom),right:f(n.right)}}function K(e,t,o){var n=D(e,function(e){var o=e.name;return o===t}),i=!!n&&e.some(function(e){return e.name===o&&e.enabled&&e.order<n.order});if(!i){var r='`'+t+'`';console.warn('`'+o+'`'+' modifier is required by '+r+' modifier in order to work, be sure to include it before '+r+'!')}return i}function z(e){return'end'===e?'start':'start'===e?'end':e}function G(e){var t=1<arguments.length&&void 0!==arguments[1]&&arguments[1],o=he.indexOf(e),n=he.slice(o+1).concat(he.slice(0,o));return t?n.reverse():n}function _(e,t,o,n){var i=e.match(/((?:\-|\+)?\d*\.?\d*)(.*)/),r=+i[1],p=i[2];if(!r)return e;if(0===p.indexOf('%')){var s;switch(p){case'%p':s=o;break;case'%':case'%r':default:s=n;}var d=g(s);return d[t]/100*r}if('vh'===p||'vw'===p){var a;return a='vh'===p?ee(document.documentElement.clientHeight,window.innerHeight||0):ee(document.documentElement.clientWidth,window.innerWidth||0),a/100*r}return r}function X(e,t,o,n){var i=[0,0],r=-1!==['right','left'].indexOf(n),p=e.split(/(\+|\-)/).map(function(e){return e.trim()}),s=p.indexOf(D(p,function(e){return-1!==e.search(/,|\s/)}));p[s]&&-1===p[s].indexOf(',')&&console.warn('Offsets separated by white space(s) are deprecated, use a comma (,) instead.');var d=/\s*,\s*|\s+/,a=-1===s?[p]:[p.slice(0,s).concat([p[s].split(d)[0]]),[p[s].split(d)[1]].concat(p.slice(s+1))];return a=a.map(function(e,n){var i=(1===n?!r:r)?'height':'width',p=!1;return e.reduce(function(e,t){return''===e[e.length-1]&&-1!==['+','-'].indexOf(t)?(e[e.length-1]=t,p=!0,e):p?(e[e.length-1]+=t,p=!1,e):e.concat(t)},[]).map(function(e){return _(e,i,t,o)})}),a.forEach(function(e,t){e.forEach(function(o,n){Y(o)&&(i[t]+=o*('-'===e[n-1]?-1:1))})}),i}function J(e,t){var o,n=t.offset,i=e.placement,r=e.offsets,p=r.popper,s=r.reference,d=i.split('-')[0];return o=Y(+n)?[+n,0]:X(n,p,s,d),'left'===d?(p.top+=o[0],p.left-=o[1]):'right'===d?(p.top+=o[0],p.left+=o[1]):'top'===d?(p.left+=o[0],p.top-=o[1]):'bottom'===d&&(p.left+=o[0],p.top+=o[1]),e.popper=p,e}var Q=Math.min,Z=Math.floor,$=Math.round,ee=Math.max,te='undefined'!=typeof window&&'undefined'!=typeof document&&'undefined'!=typeof navigator,oe=function(){for(var e=['Edge','Trident','Firefox'],t=0;t<e.length;t+=1)if(te&&0<=navigator.userAgent.indexOf(e[t]))return 1;return 0}(),ne=te&&window.Promise,ie=ne?function(e){var t=!1;return function(){t||(t=!0,window.Promise.resolve().then(function(){t=!1,e()}))}}:function(e){var t=!1;return function(){t||(t=!0,setTimeout(function(){t=!1,e()},oe))}},re=te&&!!(window.MSInputMethodContext&&document.documentMode),pe=te&&/MSIE 10/.test(navigator.userAgent),se=function(e,t){if(!(e instanceof t))throw new TypeError('Cannot call a class as a function')},de=function(){function e(e,t){for(var o,n=0;n<t.length;n++)o=t[n],o.enumerable=o.enumerable||!1,o.configurable=!0,'value'in o&&(o.writable=!0),Object.defineProperty(e,o.key,o)}return function(t,o,n){return o&&e(t.prototype,o),n&&e(t,n),t}}(),ae=function(e,t,o){return t in e?Object.defineProperty(e,t,{value:o,enumerable:!0,configurable:!0,writable:!0}):e[t]=o,e},le=Object.assign||function(e){for(var t,o=1;o<arguments.length;o++)for(var n in t=arguments[o],t)Object.prototype.hasOwnProperty.call(t,n)&&(e[n]=t[n]);return e},fe=te&&/Firefox/i.test(navigator.userAgent),me=['auto-start','auto','auto-end','top-start','top','top-end','right-start','right','right-end','bottom-end','bottom','bottom-start','left-end','left','left-start'],he=me.slice(3),ce={FLIP:'flip',CLOCKWISE:'clockwise',COUNTERCLOCKWISE:'counterclockwise'},ge=function(){function t(o,n){var i=this,r=2<arguments.length&&void 0!==arguments[2]?arguments[2]:{};se(this,t),this.scheduleUpdate=function(){return requestAnimationFrame(i.update)},this.update=ie(this.update.bind(this)),this.options=le({},t.Defaults,r),this.state={isDestroyed:!1,isCreated:!1,scrollParents:[]},this.reference=o&&o.jquery?o[0]:o,this.popper=n&&n.jquery?n[0]:n,this.options.modifiers={},Object.keys(le({},t.Defaults.modifiers,r.modifiers)).forEach(function(e){i.options.modifiers[e]=le({},t.Defaults.modifiers[e]||{},r.modifiers?r.modifiers[e]:{})}),this.modifiers=Object.keys(this.options.modifiers).map(function(e){return le({name:e},i.options.modifiers[e])}).sort(function(e,t){return e.order-t.order}),this.modifiers.forEach(function(t){t.enabled&&e(t.onLoad)&&t.onLoad(i.reference,i.popper,i.options,t,i.state)}),this.update();var p=this.options.eventsEnabled;p&&this.enableEventListeners(),this.state.eventsEnabled=p}return de(t,[{key:'update',value:function(){return k.call(this)}},{key:'destroy',value:function(){return H.call(this)}},{key:'enableEventListeners',value:function(){return I.call(this)}},{key:'disableEventListeners',value:function(){return U.call(this)}}]),t}();return ge.Utils=('undefined'==typeof window?global:window).PopperUtils,ge.placements=me,ge.Defaults={placement:'bottom',positionFixed:!1,eventsEnabled:!0,removeOnDestroy:!1,onCreate:function(){},onUpdate:function(){},modifiers:{shift:{order:100,enabled:!0,fn:function(e){var t=e.placement,o=t.split('-')[0],n=t.split('-')[1];if(n){var i=e.offsets,r=i.reference,p=i.popper,s=-1!==['bottom','top'].indexOf(o),d=s?'left':'top',a=s?'width':'height',l={start:ae({},d,r[d]),end:ae({},d,r[d]+r[a]-p[a])};e.offsets.popper=le({},p,l[n])}return e}},offset:{order:200,enabled:!0,fn:J,offset:0},preventOverflow:{order:300,enabled:!0,fn:function(e,t){var o=t.boundariesElement||p(e.instance.popper);e.instance.reference===o&&(o=p(o));var n=B('transform'),i=e.instance.popper.style,r=i.top,s=i.left,d=i[n];i.top='',i.left='',i[n]='';var a=v(e.instance.popper,e.instance.reference,t.padding,o,e.positionFixed);i.top=r,i.left=s,i[n]=d,t.boundaries=a;var l=t.priority,f=e.offsets.popper,m={primary:function(e){var o=f[e];return f[e]<a[e]&&!t.escapeWithReference&&(o=ee(f[e],a[e])),ae({},e,o)},secondary:function(e){var o='right'===e?'left':'top',n=f[o];return f[e]>a[e]&&!t.escapeWithReference&&(n=Q(f[o],a[e]-('right'===e?f.width:f.height))),ae({},o,n)}};return l.forEach(function(e){var t=-1===['left','top'].indexOf(e)?'secondary':'primary';f=le({},f,m[t](e))}),e.offsets.popper=f,e},priority:['left','right','top','bottom'],padding:5,boundariesElement:'scrollParent'},keepTogether:{order:400,enabled:!0,fn:function(e){var t=e.offsets,o=t.popper,n=t.reference,i=e.placement.split('-')[0],r=Z,p=-1!==['top','bottom'].indexOf(i),s=p?'right':'bottom',d=p?'left':'top',a=p?'width':'height';return o[s]<r(n[d])&&(e.offsets.popper[d]=r(n[d])-o[a]),o[d]>r(n[s])&&(e.offsets.popper[d]=r(n[s])),e}},arrow:{order:500,enabled:!0,fn:function(e,o){var n;if(!K(e.instance.modifiers,'arrow','keepTogether'))return e;var i=o.element;if('string'==typeof i){if(i=e.instance.popper.querySelector(i),!i)return e;}else if(!e.instance.popper.contains(i))return console.warn('WARNING: `arrow.element` must be child of its popper element!'),e;var r=e.placement.split('-')[0],p=e.offsets,s=p.popper,d=p.reference,a=-1!==['left','right'].indexOf(r),l=a?'height':'width',f=a?'Top':'Left',m=f.toLowerCase(),h=a?'left':'top',c=a?'bottom':'right',u=S(i)[l];d[c]-u<s[m]&&(e.offsets.popper[m]-=s[m]-(d[c]-u)),d[m]+u>s[c]&&(e.offsets.popper[m]+=d[m]+u-s[c]),e.offsets.popper=g(e.offsets.popper);var b=d[m]+d[l]/2-u/2,w=t(e.instance.popper),y=parseFloat(w['margin'+f]),E=parseFloat(w['border'+f+'Width']),v=b-e.offsets.popper[m]-y-E;return v=ee(Q(s[l]-u,v),0),e.arrowElement=i,e.offsets.arrow=(n={},ae(n,m,$(v)),ae(n,h,''),n),e},element:'[x-arrow]'},flip:{order:600,enabled:!0,fn:function(e,t){if(W(e.instance.modifiers,'inner'))return e;if(e.flipped&&e.placement===e.originalPlacement)return e;var o=v(e.instance.popper,e.instance.reference,t.padding,t.boundariesElement,e.positionFixed),n=e.placement.split('-')[0],i=T(n),r=e.placement.split('-')[1]||'',p=[];switch(t.behavior){case ce.FLIP:p=[n,i];break;case ce.CLOCKWISE:p=G(n);break;case ce.COUNTERCLOCKWISE:p=G(n,!0);break;default:p=t.behavior;}return p.forEach(function(s,d){if(n!==s||p.length===d+1)return e;n=e.placement.split('-')[0],i=T(n);var a=e.offsets.popper,l=e.offsets.reference,f=Z,m='left'===n&&f(a.right)>f(l.left)||'right'===n&&f(a.left)<f(l.right)||'top'===n&&f(a.bottom)>f(l.top)||'bottom'===n&&f(a.top)<f(l.bottom),h=f(a.left)<f(o.left),c=f(a.right)>f(o.right),g=f(a.top)<f(o.top),u=f(a.bottom)>f(o.bottom),b='left'===n&&h||'right'===n&&c||'top'===n&&g||'bottom'===n&&u,w=-1!==['top','bottom'].indexOf(n),y=!!t.flipVariations&&(w&&'start'===r&&h||w&&'end'===r&&c||!w&&'start'===r&&g||!w&&'end'===r&&u),E=!!t.flipVariationsByContent&&(w&&'start'===r&&c||w&&'end'===r&&h||!w&&'start'===r&&u||!w&&'end'===r&&g),v=y||E;(m||b||v)&&(e.flipped=!0,(m||b)&&(n=p[d+1]),v&&(r=z(r)),e.placement=n+(r?'-'+r:''),e.offsets.popper=le({},e.offsets.popper,C(e.instance.popper,e.offsets.reference,e.placement)),e=P(e.instance.modifiers,e,'flip'))}),e},behavior:'flip',padding:5,boundariesElement:'viewport',flipVariations:!1,flipVariationsByContent:!1},inner:{order:700,enabled:!1,fn:function(e){var t=e.placement,o=t.split('-')[0],n=e.offsets,i=n.popper,r=n.reference,p=-1!==['left','right'].indexOf(o),s=-1===['top','left'].indexOf(o);return i[p?'left':'top']=r[o]-(s?i[p?'width':'height']:0),e.placement=T(t),e.offsets.popper=g(i),e}},hide:{order:800,enabled:!0,fn:function(e){if(!K(e.instance.modifiers,'hide','preventOverflow'))return e;var t=e.offsets.reference,o=D(e.instance.modifiers,function(e){return'preventOverflow'===e.name}).boundaries;if(t.bottom<o.top||t.left>o.right||t.top>o.bottom||t.right<o.left){if(!0===e.hide)return e;e.hide=!0,e.attributes['x-out-of-boundaries']=''}else{if(!1===e.hide)return e;e.hide=!1,e.attributes['x-out-of-boundaries']=!1}return e}},computeStyle:{order:850,enabled:!0,fn:function(e,t){var o=t.x,n=t.y,i=e.offsets.popper,r=D(e.instance.modifiers,function(e){return'applyStyle'===e.name}).gpuAcceleration;void 0!==r&&console.warn('WARNING: `gpuAcceleration` option moved to `computeStyle` modifier and will not be supported in future versions of Popper.js!');var s,d,a=void 0===r?t.gpuAcceleration:r,l=p(e.instance.popper),f=u(l),m={position:i.position},h=q(e,2>window.devicePixelRatio||!fe),c='bottom'===o?'top':'bottom',g='right'===n?'left':'right',b=B('transform');if(d='bottom'==c?'HTML'===l.nodeName?-l.clientHeight+h.bottom:-f.height+h.bottom:h.top,s='right'==g?'HTML'===l.nodeName?-l.clientWidth+h.right:-f.width+h.right:h.left,a&&b)m[b]='translate3d('+s+'px, '+d+'px, 0)',m[c]=0,m[g]=0,m.willChange='transform';else{var w='bottom'==c?-1:1,y='right'==g?-1:1;m[c]=d*w,m[g]=s*y,m.willChange=c+', '+g}var E={"x-placement":e.placement};return e.attributes=le({},E,e.attributes),e.styles=le({},m,e.styles),e.arrowStyles=le({},e.offsets.arrow,e.arrowStyles),e},gpuAcceleration:!0,x:'bottom',y:'right'},applyStyle:{order:900,enabled:!0,fn:function(e){return V(e.instance.popper,e.styles),j(e.instance.popper,e.attributes),e.arrowElement&&Object.keys(e.arrowStyles).length&&V(e.arrowElement,e.arrowStyles),e},onLoad:function(e,t,o,n,i){var r=L(i,t,e,o.positionFixed),p=O(o.placement,r,t,e,o.modifiers.flip.boundariesElement,o.modifiers.flip.padding);return t.setAttribute('x-placement',p),V(t,{position:o.positionFixed?'fixed':'absolute'}),o},gpuAcceleration:void 0}}},ge});
+//# sourceMappingURL=popper.min.js.map
diff --git a/0.16/_static/jupyterlite_badge_logo.svg b/0.16/_static/jupyterlite_badge_logo.svg
new file mode 100644
index 00000000000..5de36d7fd5b
--- /dev/null
+++ b/0.16/_static/jupyterlite_badge_logo.svg
@@ -0,0 +1,3 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg width="91px" height="20px" version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" xml:space="preserve" xmlns:serif="http://www.serif.com/" style="fill-rule:evenodd;clip-rule:evenodd;stroke-linejoin:round;stroke-miterlimit:2;"><g transform="matrix(1,0,0,1,-17102,-9779)"><g id="lite-badge-launch" transform="matrix(0.0930728,0,0,0.0635416,15477.2,9173.48)"><rect x="17457.1" y="9529.53" width="977.729" height="314.755" style="fill:none;"/><clipPath id="_clip1"><rect x="17457.1" y="9529.53" width="977.729" height="314.755"/></clipPath><g clip-path="url(#_clip1)"><g transform="matrix(1.03361,0,0,1,-586.737,-1.01863e-10)"><g transform="matrix(2.43224,0,0,2.42119,-20084.1,-14912.4)"><path d="M15802.4,10095L15447.7,10095C15440.6,10095 15434.8,10103.8 15434.8,10114.5L15434.8,10205.5C15434.8,10216.3 15440.6,10225 15447.7,10225L15802.4,10225L15802.4,10095Z" style="fill:rgb(85,85,85);"/></g><g transform="matrix(-0.76361,0,0,2.42119,30189.2,-14912.4)"><path d="M15802.4,10095L15475.7,10095C15453.1,10095 15434.8,10103.8 15434.8,10114.5L15434.8,10205.5C15434.8,10216.3 15453.1,10225 15475.7,10225L15802.4,10225L15802.4,10095Z" style="fill:rgb(247,220,30);"/></g><g transform="matrix(2.74885,0,0,12.7513,10841.5,-10210.3)"><g transform="matrix(0.845162,0,0,0.273799,2185.28,1469.95)"><circle cx="316.432" cy="323.632" r="24.936" style="fill:rgb(247,220,30);"/></g><g transform="matrix(0.845162,0,0,0.273799,2185.28,1469.95)"><path d="M326.851,350.903C326.851,349.85 326.002,348.995 324.957,348.995L307.381,348.995C306.336,348.995 305.487,349.85 305.487,350.903C305.487,351.957 306.336,352.812 307.381,352.812L324.957,352.812C326.002,352.812 326.851,351.957 326.851,350.903Z" style="fill:rgb(249,249,249);"/></g><g transform="matrix(0.845162,0,0,0.273799,2185.28,1469.95)"><path d="M326.828,355.958C326.828,354.883 325.962,354.01 324.895,354.01L307.443,354.01C306.376,354.01 305.509,354.883 305.509,355.958C305.509,357.033 306.376,357.906 307.443,357.906L324.895,357.906C325.962,357.906 326.828,357.033 326.828,355.958Z" style="fill:rgb(249,249,249);"/></g><g transform="matrix(0.845162,0,0,0.273799,2185.28,1469.95)"><path d="M322.293,361.032C322.293,359.968 321.436,359.104 320.38,359.104L311.962,359.104C310.906,359.104 310.048,359.968 310.048,361.032C310.048,362.096 310.906,362.959 311.962,362.959L320.38,362.959C321.436,362.959 322.293,362.096 322.293,361.032Z" style="fill:rgb(249,249,249);"/></g><g transform="matrix(0.845162,0,0,0.273799,2425.45,1551.72)"><g transform="matrix(1,0,0,1,-288.772,-300.351)"><path d="M329.047,341.228C326.537,341.125 324.679,339.593 323.472,336.633C322.575,334.405 322.127,331.677 322.127,328.447C322.127,328.28 322.137,328.066 322.157,327.804C322.303,325.607 322.71,323.592 323.378,321.757C324.461,318.802 325.992,317.324 327.97,317.324C328.076,317.324 328.185,317.329 328.299,317.339C330.1,317.453 331.001,318.498 331.001,320.476C331.001,322.635 330.074,325.031 328.219,327.664C327.043,329.342 325.808,330.703 324.515,331.746C324.871,333.724 325.261,335.208 325.687,336.198C326.481,338.03 327.659,338.946 329.221,338.946C329.324,338.946 329.434,338.936 329.55,338.915C329.892,338.852 330.371,338.603 330.986,338.167C331.58,337.759 332.044,337.555 332.376,337.555C332.523,337.555 332.664,337.603 332.8,337.698C333.102,337.905 333.253,338.186 333.253,338.541C333.231,339.083 332.845,339.64 332.097,340.212C331.159,340.889 330.143,341.228 329.047,341.228ZM327.989,319.683C327.371,319.693 326.83,319.957 326.364,320.476C325.174,321.822 324.523,324.631 324.41,328.904C325.035,328.267 325.659,327.516 326.284,326.652C328.013,324.286 328.877,322.29 328.877,320.665C328.877,320.582 328.855,320.484 328.813,320.37C328.646,319.912 328.37,319.683 327.985,319.683L327.989,319.683Z" style="fill:rgb(85,85,85);fill-rule:nonzero;"/></g><g transform="matrix(1,0,0,1,-288.772,-300.351)"><path d="M321.235,316.187C321.235,316.188 321.235,316.189 321.235,316.19C321.235,317.629 320.051,318.813 318.612,318.813L318.612,318.813C317.173,318.813 315.989,317.629 315.989,316.19C315.989,316.189 315.989,316.188 315.989,316.187C315.989,314.748 317.173,313.564 318.612,313.564C320.051,313.564 321.235,314.748 321.235,316.187Z" style="fill:rgb(85,85,85);fill-rule:nonzero;"/></g><g transform="matrix(1,0,0,1,-288.772,-300.351)"><path d="M318.049,319.981C318.727,319.991 319.076,320.432 319.096,321.304L317.361,333.711C319.966,331.721 321.43,330.596 321.753,330.336C322.138,330.004 322.534,329.832 322.94,329.822C323.086,329.822 323.227,329.869 323.363,329.962C323.665,330.171 323.816,330.437 323.816,330.76C323.804,331.261 323.522,331.746 322.97,332.215C321.44,333.505 319.415,335.092 316.893,336.977L316.175,342.15C316.144,342.39 316.098,342.624 316.035,342.853C315.556,344.685 314.353,345.601 312.426,345.601C311.468,345.601 310.624,345.242 309.893,344.524C309.175,343.806 308.816,342.999 308.816,342.105C308.816,341.165 309.041,340.394 309.489,339.792C309.738,339.48 310.227,339.051 310.955,338.507L314.704,335.695C316.027,326.958 316.705,322.365 316.738,321.916C316.738,321.853 316.743,321.796 316.753,321.746C316.866,320.569 317.298,319.981 318.049,319.981ZM313.941,340.899L314.206,338.995L312.471,340.321C311.72,340.883 311.34,341.415 311.33,341.916C311.33,342.092 311.361,342.259 311.424,342.415C311.6,342.874 311.934,343.103 312.426,343.103C312.622,343.103 312.798,343.071 312.955,343.008C313.456,342.789 313.784,342.086 313.937,340.899L313.941,340.899ZM317.331,316.995C317.331,316.809 317.378,316.633 317.471,316.466C317.69,316.058 318.029,315.854 318.488,315.854C318.674,315.854 318.855,315.901 319.032,315.994C319.43,316.213 319.629,316.547 319.629,316.995C319.629,317.182 319.581,317.365 319.485,317.543C319.269,317.939 318.936,318.137 318.488,318.137C318.299,318.137 318.116,318.089 317.94,317.993C317.534,317.787 317.331,317.454 317.331,316.995Z" style="fill:rgb(85,85,85);fill-rule:nonzero;"/></g></g></g><g transform="matrix(3.03205,0,0,14.065,10159.7,-12659.1)"><g transform="matrix(0.295204,0,0,0.0956347,2538.31,1594.22)"><text x="-181.194px" y="0px" style="font-family:'Verdana', sans-serif;font-size:130.5px;fill:rgb(1,1,1);fill-opacity:0.3;">launch</text></g><g transform="matrix(0.295204,0,0,0.0956347,2538.31,1593.26)"><text x="-181.194px" y="0px" style="font-family:'Verdana', sans-serif;font-size:130.5px;fill:white;">launch</text></g></g><g transform="matrix(3.5905,0,0,16.6556,9244.95,-16787)"><g transform="matrix(0.295204,0,0,0.0956347,2538.31,1594.22)"><text x="-181.194px" y="0px" style="font-family:'Verdana', sans-serif;font-size:110px;fill:rgb(1,1,1);fill-opacity:0.1;">lite</text></g><g transform="matrix(0.295204,0,0,0.0956347,2538.31,1593.26)"><text x="-181.194px" y="0px" style="font-family:'Verdana', sans-serif;font-size:110px;fill:rgb(56,56,56);">lite</text></g></g></g></g></g></g></svg>
\ No newline at end of file
diff --git a/0.16/_static/language_data.js b/0.16/_static/language_data.js
new file mode 100644
index 00000000000..2e22b06ab13
--- /dev/null
+++ b/0.16/_static/language_data.js
@@ -0,0 +1,199 @@
+/*
+ * language_data.js
+ * ~~~~~~~~~~~~~~~~
+ *
+ * This script contains the language-specific data used by searchtools.js,
+ * namely the list of stopwords, stemmer, scorer and splitter.
+ *
+ * :copyright: Copyright 2007-2022 by the Sphinx team, see AUTHORS.
+ * :license: BSD, see LICENSE for details.
+ *
+ */
+
+var stopwords = ["a", "and", "are", "as", "at", "be", "but", "by", "for", "if", "in", "into", "is", "it", "near", "no", "not", "of", "on", "or", "such", "that", "the", "their", "then", "there", "these", "they", "this", "to", "was", "will", "with"];
+
+
+/* Non-minified version is copied as a separate JS file, is available */
+
+/**
+ * Porter Stemmer
+ */
+var Stemmer = function() {
+
+  var step2list = {
+    ational: 'ate',
+    tional: 'tion',
+    enci: 'ence',
+    anci: 'ance',
+    izer: 'ize',
+    bli: 'ble',
+    alli: 'al',
+    entli: 'ent',
+    eli: 'e',
+    ousli: 'ous',
+    ization: 'ize',
+    ation: 'ate',
+    ator: 'ate',
+    alism: 'al',
+    iveness: 'ive',
+    fulness: 'ful',
+    ousness: 'ous',
+    aliti: 'al',
+    iviti: 'ive',
+    biliti: 'ble',
+    logi: 'log'
+  };
+
+  var step3list = {
+    icate: 'ic',
+    ative: '',
+    alize: 'al',
+    iciti: 'ic',
+    ical: 'ic',
+    ful: '',
+    ness: ''
+  };
+
+  var c = "[^aeiou]";          // consonant
+  var v = "[aeiouy]";          // vowel
+  var C = c + "[^aeiouy]*";    // consonant sequence
+  var V = v + "[aeiou]*";      // vowel sequence
+
+  var mgr0 = "^(" + C + ")?" + V + C;                      // [C]VC... is m>0
+  var meq1 = "^(" + C + ")?" + V + C + "(" + V + ")?$";    // [C]VC[V] is m=1
+  var mgr1 = "^(" + C + ")?" + V + C + V + C;              // [C]VCVC... is m>1
+  var s_v   = "^(" + C + ")?" + v;                         // vowel in stem
+
+  this.stemWord = function (w) {
+    var stem;
+    var suffix;
+    var firstch;
+    var origword = w;
+
+    if (w.length < 3)
+      return w;
+
+    var re;
+    var re2;
+    var re3;
+    var re4;
+
+    firstch = w.substr(0,1);
+    if (firstch == "y")
+      w = firstch.toUpperCase() + w.substr(1);
+
+    // Step 1a
+    re = /^(.+?)(ss|i)es$/;
+    re2 = /^(.+?)([^s])s$/;
+
+    if (re.test(w))
+      w = w.replace(re,"$1$2");
+    else if (re2.test(w))
+      w = w.replace(re2,"$1$2");
+
+    // Step 1b
+    re = /^(.+?)eed$/;
+    re2 = /^(.+?)(ed|ing)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      re = new RegExp(mgr0);
+      if (re.test(fp[1])) {
+        re = /.$/;
+        w = w.replace(re,"");
+      }
+    }
+    else if (re2.test(w)) {
+      var fp = re2.exec(w);
+      stem = fp[1];
+      re2 = new RegExp(s_v);
+      if (re2.test(stem)) {
+        w = stem;
+        re2 = /(at|bl|iz)$/;
+        re3 = new RegExp("([^aeiouylsz])\\1$");
+        re4 = new RegExp("^" + C + v + "[^aeiouwxy]$");
+        if (re2.test(w))
+          w = w + "e";
+        else if (re3.test(w)) {
+          re = /.$/;
+          w = w.replace(re,"");
+        }
+        else if (re4.test(w))
+          w = w + "e";
+      }
+    }
+
+    // Step 1c
+    re = /^(.+?)y$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      re = new RegExp(s_v);
+      if (re.test(stem))
+        w = stem + "i";
+    }
+
+    // Step 2
+    re = /^(.+?)(ational|tional|enci|anci|izer|bli|alli|entli|eli|ousli|ization|ation|ator|alism|iveness|fulness|ousness|aliti|iviti|biliti|logi)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      suffix = fp[2];
+      re = new RegExp(mgr0);
+      if (re.test(stem))
+        w = stem + step2list[suffix];
+    }
+
+    // Step 3
+    re = /^(.+?)(icate|ative|alize|iciti|ical|ful|ness)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      suffix = fp[2];
+      re = new RegExp(mgr0);
+      if (re.test(stem))
+        w = stem + step3list[suffix];
+    }
+
+    // Step 4
+    re = /^(.+?)(al|ance|ence|er|ic|able|ible|ant|ement|ment|ent|ou|ism|ate|iti|ous|ive|ize)$/;
+    re2 = /^(.+?)(s|t)(ion)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      re = new RegExp(mgr1);
+      if (re.test(stem))
+        w = stem;
+    }
+    else if (re2.test(w)) {
+      var fp = re2.exec(w);
+      stem = fp[1] + fp[2];
+      re2 = new RegExp(mgr1);
+      if (re2.test(stem))
+        w = stem;
+    }
+
+    // Step 5
+    re = /^(.+?)e$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      re = new RegExp(mgr1);
+      re2 = new RegExp(meq1);
+      re3 = new RegExp("^" + C + v + "[^aeiouwxy]$");
+      if (re.test(stem) || (re2.test(stem) && !(re3.test(stem))))
+        w = stem;
+    }
+    re = /ll$/;
+    re2 = new RegExp(mgr1);
+    if (re.test(w) && re2.test(w)) {
+      re = /.$/;
+      w = w.replace(re,"");
+    }
+
+    // and turn initial Y back to y
+    if (firstch == "y")
+      w = firstch.toLowerCase() + w.substr(1);
+    return w;
+  }
+}
+
diff --git a/0.16/_static/minus.png b/0.16/_static/minus.png
new file mode 100644
index 0000000000000000000000000000000000000000..d96755fdaf8bb2214971e0db9c1fd3077d7c419d
GIT binary patch
literal 90
zcmeAS@N?(olHy`uVBq!ia0vp^+#t*WBp7;*Yy1LIik>cxAr*|t7R?Mi>2?kWtu=nj
kDsEF_5m^0CR;1wuP-*O&G^0G}KYk!hp00i_>zopr08q^qX#fBK

literal 0
HcmV?d00001

diff --git a/0.16/_static/no_image.png b/0.16/_static/no_image.png
new file mode 100644
index 0000000000000000000000000000000000000000..8c2d48d5d3f0044cf0399b3608d8eb5db6f9d12d
GIT binary patch
literal 4315
zcmZu#cQl+^wEv7Yqs$n@O(aT$L<tcjL>;}C=%WOY=zT`-Ekq0V5`x4aMu{NNf+*n<
z66KpH;|{}!gwZ2<f3n{D^PP3}+3W0m&R%DovVOmPo|+hGvodir0RX_NtD|9ht{cuJ
z6v}YkX0|a~pDP-SskRzWGkj&~yaV&oL1WGb6aNT}ns7nLxseg8YpBUcrhy~GZn&;(
zubrD19_njroa^&)pxzb@0Bp9p8mi_Yq_vl+zMMN;pEsgo5xjBP7g;NWt=>&nysHrG
z0Is>2pg8&bU~%jmP#w09%E+E{cKXfUYnz`+$>gQMZA;Ck3`Hh2nh%QykGcLe4RW<{
zQ;-htbPfJBvF^iHUpwo2@bbr_W|JoA?RwXd4aKwPnF3{LKYV|8qEhH=QaE7^oFeRQ
zBBp}ipZZz{SEiGN($5lzeJ-F05!Zy+^Oi_@>IjmQ?>?L~yEgcsSTdZsfxDGCdJo3i
zOq2?zJ@qMmXl-%TCEa5sfRF36uZ_fOBpI0PSz*CD%GGKxOOf(!`@F3DX<p*l()fP8
zc!=E`H)%}6oTtDsP+btblCl!I<K!}X(P@Gi2ts75ic`@Qqv~zAmtc^a*fu{_7I5IX
zTm6)^d`kOi!0hl615j;9SMi2s(<dgJZN#>C4K&If51b<`6}uL5b>z~lD}_YRggJL(
zR)oMSQ2tje{IY5Yj=H*zr~J$DKB!<6AMJFmKlA(I2MjTMw3!M(sVb$L{YG2`;=t+S
zuom?Z3|ncJ>NZt|9JQg}&rm1h`1fK+rNrNF_nJYWIbjt?1%hC|K0s-XL0Enm?#0{^
z9#rb;^K}1gZ)Te)JE#N0-hTW3QYUYn_qApLKK?z(2UZJ#o2MoNPIt$XfO^QU)d--F
zXH`BxyQvKy)=t+7g*b$!Y}{x|OlV0OKjl&04HNP(6h1uYGPwcN?EnXR5$O3Sl;Lx}
z(?<F813|%Vdq&)L8m2f$|0Ok#;2IGnDnsi!zV>Pl(|eqOat|Q((7g0X9j2BX$O5mU
zwDn5atHnDPR$VqpfV+sJOE1GcvZ?2k=4NFggq#1ZXv2si+h268TDMo7LJ>DEf00Br
z;PWK3_>wf4JB~zQSY<@4Rzy{IHAmT#7RE`=#i%6Zd%rJcT!!V%>Z+RCWQL<61zx0O
zp%&*%euVDC9FR@hxY4-vuEutY(P?=Z*RRvn@W8xXF3J)Gw%~KMJvh|314<+F7!Yyq
zzC1|A;l1Ybr%NyC%9iHH=C!Ft%DuRsjap=CXuO`w;+Pe&Y=(59uVLKGJC$rmWGrgg
zODk#8g1j761-aFn<R3V#)9mCjd3J_%$qsD4q1nn+?3rd=R)XPKvQUi5EH=(GDFX_i
z@|gLr?dceMR=kpSqQ=LbRa)=w6nRSskD|G3Vv-3Nz@5J9t2^{A<B6}X-z|ko7BSCl
z7AkSlP=A_3MA<+d@G))O@K-xa;#D@#b(6tPA{Gwc=6Z!lKu0Xy@<<EOoSi^d#%vrX
zLOU?e;^(+vSU=pG&@`1jn!uZb)!IjCUK%^vfQ|uY-bin0<(4Ld=L*$c08RC=4D<AD
z<b6<O>HJ1#qNN~lM(9uE@)DXg^^^LC4X(ioEbS1-<D2Y*<W^hd926sCe$`4V9bN04
z?&_(qdn_}37|Xic<u-|`j(^Ubhu*5O%CwPp!(B|mx%>)|pc)70tB;GPZO@~yvAHcv
z%x00*%rI}(p-`Eb1kyto76p4T@pgMtoFBwWdQE!wJt-%T`R}mD5co;4AY<$JopD<s
zkVeR>u!F_%Sb+O{E42l*RyUR!FO%eA{553q%ut@4Ae8<*lKYZh6{kxg9ld;tXRTha
z?3gIPbAiK6DVyr{_HWc8KhmXL*o|6~C_g5>KVs&`?}e0pRNme%EfxoO8|$GJrZ`yK
zE)$i-l)8&YM7O=tAziD4a;BJ9g0aK*tI$*QKZ2l8>|e_ET(8k?>b9u4TZEY1ezM~-
zJv;Sd!bWv;3RlOLc8d6yOmlR0e(2Tgul@`jXONkDNCs-Yo0kk_Usn>RAb`K4LjGA%
zaKm(<!nZhDPh*F`PCY*SCbQCF&ySm$3sC&FrF;c%nq%LFl5Czgdz=VOgVJN=Jf6j_
z+`q*J`cu=exSdNef)Z@tPDZ7jNUTI!&MSSMi8vrq)T?h}h@x&xFIY|&dx=;%av>2h
z7o)uLLF{C1Ap`m}Co@arT?tu0?P2dT1>fz$j2w2VnwBB}bNufX0H)8mg@@(K-HUYa
z<piiF4YitrcmL%Hc@>~!6OwLNR4_{T8CEbVF@srA!uu;E^%$2lB>(V$BCy{~3nsQc
z9*$O`zfFYySHr{ib0g>B0!D>7)2Ic%`!U%7%8Kn-sxJkGSHQax>OuHMNih)LnmG#E
zU!HNRRVcr#t&7wDZrp6u#U9mVYo#Ah@H{IrGO^R8n^M=lOuHP+nUnP1dc|_|N*e@d
zM`V6K2D~kr3zr7NZ~-F?<^(w)<9I=CIt@|ia>A5_7B1*5NI@o*H4zU7SZc>aHf9cC
zXGIX>-3vi}CRom-#}D-5eKi(w)t1+SKg@iA)-TfHef56E?Gw`5ua0)$YgYmdPq=}`
zW^)2O`s=D}i1U!~hwuO9k6gx$==LW&<W5I-uby-?Zl^#o`uMolt0taXWP#zt-Do3i
zsXl{nJC?Q9fJ1Vw5EvcDDyBPC1)$qi9?y#L_nQx%jiVe-#l}ch=GzX!btjVnV$-RI
zKZ3;Tck=^|U*_I9pE9(H1YaOsym!e+o3iPJ5zH>6YwG6+Iaw7!37oYu2!CKW27mD2
z(zrJ5YG~ztLr{#%Ov?cGqx|^WlKqh(3dP>QzFf<Nkf9+zahL0P@+F1Q#6@~AT1k=A
z1MKbBT`HixIQv?+7}&ve?w?l7F7G7y5{wTW_{{RgH7dOhqo{2u?&6<{$oG@!_r@&N
zq;}v$^IWimH(Jg8S_3R>u$J~7jAFZa{A`SPNEm}>x5l6DbuQ(R3h74sf-eR34Cl^n
zgfe1{i^sOtOLu>9_wX9pdK6pKeF&x{g>N2901CGUNut;8dwLBKLg>AJh#qQ&`Ltuq
zEB#8AtpS@Q?nVxpV|y5WmYK3)6qrTKFxJQ+WK^;ke;8ia+Sv=8p*<%C;+p@<AFzZ-
zi&_x^!E`*-pk3IhDqf*9i`iW-^K+K5p0bMJKPei{e^t3e_zk_A)Z7mA?HLyWA0s5-
z=qK_|n(l=fg!er@{3>VX9TSyM8~lAmUh(Xkb1M@$v9?)NR+adDqo}X*Wkx*uu|O4E
zd`xIU<885Arlnf>e9U1((ZnnrdBOLNiCH-Et-RuTLKGdD&1v^5=<4>+Y%=+o$iP)g
zh<%8E#<=}5G8x|l&s+u_PFP-YIL4j0A0N5ceAe=JRxF}PjS8(HUc9^ZmF7E3RVjYQ
zuW`_)Z+<DZso<bAqVizQs-h-uS3b<yoG;a0G?r^R5hrB`G)v{!;RFgFk%(U%cE!;M
z<eu=QhaR<pd0}g)TO3?Fq-6JYHC^t`4^?6gXhRj`*j(qj>^@qovbxdEKKngA>9$1*
z!6$)NbmIAhIb+Icqjb!{KwUov4T;z*iZ59e=0(hzIXdGrurL`{$n37X^%q%SbHLP~
zzZ~h%K(10bsyk6yu+@LpG#spNq-Ai#D!TADYK!V5xu$AU^NJK?Lqc=GM_qT;Jgn>W
z$Q-)1Va4GTR~6ydg#jt)9e?Y-+5M&bi9?p1PrfsQ`b7DfdD(s9Gksi34BD1SY1pd1
zGN@}w=r__Q!ycXq_nTnUwMf<%0)>>0-7$_xR=BIJbw=@UKT`v;h$iGFylp(a&V48g
zmD?v}yqh=EwN&@q$Zodv0X$*lIGHv{TC+O-*TxNfTq+DBbB}?7i7H1BP#D7jc*EOl
zfmXuVWaCcgWZH>3#GXa=8X71vxZhS->o-{>qGaUm-8oB`k2vZYR+Ev`!Hsrf4w;Y(
zV)NVu_V}61Pvj4=-$(+3&G@TZ?9}eVp=L_yaUImwy3$hcQ$t($46S2vMjn@H(;P2c
z`sf^(X<m;)5fG3s<-KVhx4n{Co63+{!Nkd!qZpX7i0FBuqyu?*W<W7O1*qXH;7vUZ
z|6Q`n_)lgRg&XI-fs+;j4;qHG$3E2!J7sZom@)AZk2P#q^3ld7U(o!x0lg&8j6&|4
zvBS2HN`q`XOj4}hBy3jxeb-dByE?JtZJ1F;pk%fCIpVZj_T9RIU!kX#OZ|&0VrO%c
z?OqKQ1P_wEE;h~seX3d0>6oJvKk_pd;P8Idu6Phe0sbUyQVxb)FK+EElYp^8x7`6H
zaawS+qESx}*Orqn@f@`X-?k(;?#3#|dKP7`t55wS9Kg>FzW#OAtc$}S)!XSDS2;MT
z&Oob__!lSZ9GP3j8aRLQkq65WI}B^`w2GL@8NObeBo!EM7A)H(CnyZwaAUyH$@_gU
zDtO7wELIqZ|1PLBO3~ov`sd26E`NX1Tc~D{CwFWiJd@G*If*i|I(uCNC$YLSf31U}
z-KjTEPe+&U;!*pDxr8y-GK(H9TW{utIW!$HNv|mLDp(#in~co}#KH?r2r3f2a&d|P
z`G<^{l?y$o?lX{fohCMj(<SoeA5sLr9XpYofaD!x2IB#34FlX74?C5DvPxydzAy3%
zb=YH&e7;4pnq&mW&+WQ7BN`Br!{nBOU2svcI8*e&k-g%lm{WuEm8yBxwm>xRRm3n!
zp0#!EG0IJBk#0_XeLhVT1V6>TV#Q6-{eFqYV}ozaDd@#Z7k>u_pZim<*k$P1*k8{K
zjEctX<xOLRlXLJn!Hv~#O(8sTW)&H)?{vgjFZR?_&PapdECmxvV`-vL@JL<D>ouLc
ztXTVtJsUSs-#~*k2h|N^&1dG1&5sFlIZd0t2UFNkW`?)Z=_qnThBbXbH<B8Au5w+b
zactUJ4SPLd_lx{QNQ;@v>`910?N6fKiUYWuYUf+umu&dU>dqGFovFS$>HJ#yDX;|6
zR68m}C!r!|Q6=1wt-M-N%kQ4a>KlqX`^j?s8h$G0%#zmM8)cvLIkiBw=i>SMd9@Kv
zuh_gLGe~G2=F?7_>}97$*eq29VZI1F=koA~Dtr0FCnnwML`AT#W+xcYgaNiX{Kh)a
zbXs(U-sK}@iKwzG8}xY&dK8XQaIckv>-J!EuHL=7H(RWe|HEj`srN6H0v?yT9t!2f
zZ7X5#ckde$7Cs(Wgr?$V|5B*dzGOg^%9>;gqO|-q{%SjzElgARuKd%Y#ttz=RvHD#
z=BXxVvFHCD>^0~p@E2xpEaFUOCRmwHzhGMY*%=~o!Pcn82!78XIwm*#3G?rm?&{3W
z!%1<I&HLgrioq%U3H!End0{!98^(d!tHj37WJ?}CL-F@ku0Q|2DUzL#?|ki;CkgAO
zn7<<W(`V(M%fN4skGmEn-%XUmnbJxA=H?r>W)#O~l{IF|>Wc4MAGQ1Z9x9#@{1Qo&
zJh~poCb^!d<!;~`J-0gxE&n@J_C>qmqsjSmYI7VE$ww3e7O;HlhIKRYuVnGnH8Vf(
z_%qq>;Za>%<S45H#oS#?!56>#zP7j5=B|#mUT_>Pt?{DzP!XV(XhdAWw;#4!TG0vJ
z*W)fq3}SzVglo1OAASuAZO`5^`+u_a|4Z2afO59lmos30Xp|#TIWqpwcZ05`kw%T0
HW8D7$6^1w`

literal 0
HcmV?d00001

diff --git a/0.16/_static/plus.png b/0.16/_static/plus.png
new file mode 100644
index 0000000000000000000000000000000000000000..7107cec93a979b9a5f64843235a16651d563ce2d
GIT binary patch
literal 90
zcmeAS@N?(olHy`uVBq!ia0vp^+#t*WBp7;*Yy1LIik>cxAr*|t7R?Mi>2?kWtu>-2
m3q%Vub%g%s<8sJhVPMczOq}xhg9DJoz~JfX=d#Wzp$Pyb1r*Kz

literal 0
HcmV?d00001

diff --git a/0.16/_static/pygments.css b/0.16/_static/pygments.css
new file mode 100644
index 00000000000..0d49244edab
--- /dev/null
+++ b/0.16/_static/pygments.css
@@ -0,0 +1,75 @@
+pre { line-height: 125%; }
+td.linenos .normal { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+span.linenos { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+.highlight .hll { background-color: #ffffcc }
+.highlight { background: #eeffcc; }
+.highlight .c { color: #408090; font-style: italic } /* Comment */
+.highlight .err { border: 1px solid #FF0000 } /* Error */
+.highlight .k { color: #007020; font-weight: bold } /* Keyword */
+.highlight .o { color: #666666 } /* Operator */
+.highlight .ch { color: #408090; font-style: italic } /* Comment.Hashbang */
+.highlight .cm { color: #408090; font-style: italic } /* Comment.Multiline */
+.highlight .cp { color: #007020 } /* Comment.Preproc */
+.highlight .cpf { color: #408090; font-style: italic } /* Comment.PreprocFile */
+.highlight .c1 { color: #408090; font-style: italic } /* Comment.Single */
+.highlight .cs { color: #408090; background-color: #fff0f0 } /* Comment.Special */
+.highlight .gd { color: #A00000 } /* Generic.Deleted */
+.highlight .ge { font-style: italic } /* Generic.Emph */
+.highlight .ges { font-weight: bold; font-style: italic } /* Generic.EmphStrong */
+.highlight .gr { color: #FF0000 } /* Generic.Error */
+.highlight .gh { color: #000080; font-weight: bold } /* Generic.Heading */
+.highlight .gi { color: #00A000 } /* Generic.Inserted */
+.highlight .go { color: #333333 } /* Generic.Output */
+.highlight .gp { color: #c65d09; font-weight: bold } /* Generic.Prompt */
+.highlight .gs { font-weight: bold } /* Generic.Strong */
+.highlight .gu { color: #800080; font-weight: bold } /* Generic.Subheading */
+.highlight .gt { color: #0044DD } /* Generic.Traceback */
+.highlight .kc { color: #007020; font-weight: bold } /* Keyword.Constant */
+.highlight .kd { color: #007020; font-weight: bold } /* Keyword.Declaration */
+.highlight .kn { color: #007020; font-weight: bold } /* Keyword.Namespace */
+.highlight .kp { color: #007020 } /* Keyword.Pseudo */
+.highlight .kr { color: #007020; font-weight: bold } /* Keyword.Reserved */
+.highlight .kt { color: #902000 } /* Keyword.Type */
+.highlight .m { color: #208050 } /* Literal.Number */
+.highlight .s { color: #4070a0 } /* Literal.String */
+.highlight .na { color: #4070a0 } /* Name.Attribute */
+.highlight .nb { color: #007020 } /* Name.Builtin */
+.highlight .nc { color: #0e84b5; font-weight: bold } /* Name.Class */
+.highlight .no { color: #60add5 } /* Name.Constant */
+.highlight .nd { color: #555555; font-weight: bold } /* Name.Decorator */
+.highlight .ni { color: #d55537; font-weight: bold } /* Name.Entity */
+.highlight .ne { color: #007020 } /* Name.Exception */
+.highlight .nf { color: #06287e } /* Name.Function */
+.highlight .nl { color: #002070; font-weight: bold } /* Name.Label */
+.highlight .nn { color: #0e84b5; font-weight: bold } /* Name.Namespace */
+.highlight .nt { color: #062873; font-weight: bold } /* Name.Tag */
+.highlight .nv { color: #bb60d5 } /* Name.Variable */
+.highlight .ow { color: #007020; font-weight: bold } /* Operator.Word */
+.highlight .w { color: #bbbbbb } /* Text.Whitespace */
+.highlight .mb { color: #208050 } /* Literal.Number.Bin */
+.highlight .mf { color: #208050 } /* Literal.Number.Float */
+.highlight .mh { color: #208050 } /* Literal.Number.Hex */
+.highlight .mi { color: #208050 } /* Literal.Number.Integer */
+.highlight .mo { color: #208050 } /* Literal.Number.Oct */
+.highlight .sa { color: #4070a0 } /* Literal.String.Affix */
+.highlight .sb { color: #4070a0 } /* Literal.String.Backtick */
+.highlight .sc { color: #4070a0 } /* Literal.String.Char */
+.highlight .dl { color: #4070a0 } /* Literal.String.Delimiter */
+.highlight .sd { color: #4070a0; font-style: italic } /* Literal.String.Doc */
+.highlight .s2 { color: #4070a0 } /* Literal.String.Double */
+.highlight .se { color: #4070a0; font-weight: bold } /* Literal.String.Escape */
+.highlight .sh { color: #4070a0 } /* Literal.String.Heredoc */
+.highlight .si { color: #70a0d0; font-style: italic } /* Literal.String.Interpol */
+.highlight .sx { color: #c65d09 } /* Literal.String.Other */
+.highlight .sr { color: #235388 } /* Literal.String.Regex */
+.highlight .s1 { color: #4070a0 } /* Literal.String.Single */
+.highlight .ss { color: #517918 } /* Literal.String.Symbol */
+.highlight .bp { color: #007020 } /* Name.Builtin.Pseudo */
+.highlight .fm { color: #06287e } /* Name.Function.Magic */
+.highlight .vc { color: #bb60d5 } /* Name.Variable.Class */
+.highlight .vg { color: #bb60d5 } /* Name.Variable.Global */
+.highlight .vi { color: #bb60d5 } /* Name.Variable.Instance */
+.highlight .vm { color: #bb60d5 } /* Name.Variable.Magic */
+.highlight .il { color: #208050 } /* Literal.Number.Integer.Long */
\ No newline at end of file
diff --git a/0.16/_static/pytorch-logo-dark.svg b/0.16/_static/pytorch-logo-dark.svg
new file mode 100644
index 00000000000..717a3ce942f
--- /dev/null
+++ b/0.16/_static/pytorch-logo-dark.svg
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!-- Generator: Adobe Illustrator 21.0.0, SVG Export Plug-In . SVG Version: 6.00 Build 0)  -->
+<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
+	 viewBox="0 0 199.7 40.2" style="enable-background:new 0 0 199.7 40.2;" xml:space="preserve">
+<style type="text/css">
+	.st0{fill:#F05732;}
+	.st1{fill:#9E529F;}
+	.st2{fill:#333333;}
+</style>
+<path class="st0" d="M102.7,12.2c-1.3-1-1.8,3.9-4.4,3.9c-3,0-4-13-6.3-13c-0.7,0-0.8-0.4-7.9,21.3c-2.9,9,4.4,15.8,11.8,15.8
+	c4.6,0,12.3-3,12.3-12.6C108.2,20.5,104.7,13.7,102.7,12.2z M95.8,35.3c-3.7,0-6.7-3.1-6.7-7c0-3.9,3-7,6.7-7s6.7,3.1,6.7,7
+	C102.5,32.1,99.5,35.3,95.8,35.3z"/>
+<path class="st1" d="M99.8,0c-0.5,0-1.8,2.5-1.8,3.6c0,1.5,1,2,1.8,2c0.8,0,1.8-0.5,1.8-2C101.5,2.5,100.2,0,99.8,0z"/>
+<path class="st2" d="M0,39.5V14.9h11.5c5.3,0,8.3,3.6,8.3,7.9c0,4.3-3,7.9-8.3,7.9H5.2v8.8H0z M14.4,22.8c0-2.1-1.6-3.3-3.7-3.3H5.2
+	v6.6h5.5C12.8,26.1,14.4,24.8,14.4,22.8z"/>
+<path class="st2" d="M35.2,39.5V29.4l-9.4-14.5h6l6.1,9.8l6.1-9.8h5.9l-9.4,14.5v10.1H35.2z"/>
+<path class="st2" d="M63.3,39.5v-20h-7.2v-4.6h19.6v4.6h-7.2v20H63.3z"/>
+<path class="st2" d="M131.4,39.5l-4.8-8.7h-3.8v8.7h-5.2V14.9H129c5.1,0,8.3,3.4,8.3,7.9c0,4.3-2.8,6.7-5.4,7.3l5.6,9.4H131.4z
+	 M131.9,22.8c0-2-1.6-3.3-3.7-3.3h-5.5v6.6h5.5C130.3,26.1,131.9,24.9,131.9,22.8z"/>
+<path class="st2" d="M145.6,27.2c0-7.6,5.7-12.7,13.1-12.7c5.4,0,8.5,2.9,10.3,6l-4.5,2.2c-1-2-3.2-3.6-5.8-3.6
+	c-4.5,0-7.7,3.4-7.7,8.1c0,4.6,3.2,8.1,7.7,8.1c2.5,0,4.7-1.6,5.8-3.6l4.5,2.2c-1.7,3.1-4.9,6-10.3,6
+	C151.3,39.9,145.6,34.7,145.6,27.2z"/>
+<path class="st2" d="M194.5,39.5V29.1h-11.6v10.4h-5.2V14.9h5.2v9.7h11.6v-9.7h5.3v24.6H194.5z"/>
+</svg>
diff --git a/0.16/_static/searchtools.js b/0.16/_static/searchtools.js
new file mode 100644
index 00000000000..ac4d5861f95
--- /dev/null
+++ b/0.16/_static/searchtools.js
@@ -0,0 +1,531 @@
+/*
+ * searchtools.js
+ * ~~~~~~~~~~~~~~~~
+ *
+ * Sphinx JavaScript utilities for the full-text search.
+ *
+ * :copyright: Copyright 2007-2022 by the Sphinx team, see AUTHORS.
+ * :license: BSD, see LICENSE for details.
+ *
+ */
+"use strict";
+
+/**
+ * Simple result scoring code.
+ */
+if (typeof Scorer === "undefined") {
+  var Scorer = {
+    // Implement the following function to further tweak the score for each result
+    // The function takes a result array [docname, title, anchor, descr, score, filename]
+    // and returns the new score.
+    /*
+    score: result => {
+      const [docname, title, anchor, descr, score, filename] = result
+      return score
+    },
+    */
+
+    // query matches the full name of an object
+    objNameMatch: 11,
+    // or matches in the last dotted part of the object name
+    objPartialMatch: 6,
+    // Additive scores depending on the priority of the object
+    objPrio: {
+      0: 15, // used to be importantResults
+      1: 5, // used to be objectResults
+      2: -5, // used to be unimportantResults
+    },
+    //  Used when the priority is not in the mapping.
+    objPrioDefault: 0,
+
+    // query found in title
+    title: 15,
+    partialTitle: 7,
+    // query found in terms
+    term: 5,
+    partialTerm: 2,
+  };
+}
+
+const _removeChildren = (element) => {
+  while (element && element.lastChild) element.removeChild(element.lastChild);
+};
+
+/**
+ * See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_Expressions#escaping
+ */
+const _escapeRegExp = (string) =>
+  string.replace(/[.*+\-?^${}()|[\]\\]/g, "\\$&"); // $& means the whole matched string
+
+const _displayItem = (item, highlightTerms, searchTerms) => {
+  const docBuilder = DOCUMENTATION_OPTIONS.BUILDER;
+  const docUrlRoot = DOCUMENTATION_OPTIONS.URL_ROOT;
+  const docFileSuffix = DOCUMENTATION_OPTIONS.FILE_SUFFIX;
+  const docLinkSuffix = DOCUMENTATION_OPTIONS.LINK_SUFFIX;
+  const showSearchSummary = DOCUMENTATION_OPTIONS.SHOW_SEARCH_SUMMARY;
+
+  const [docName, title, anchor, descr] = item;
+
+  let listItem = document.createElement("li");
+  let requestUrl;
+  let linkUrl;
+  if (docBuilder === "dirhtml") {
+    // dirhtml builder
+    let dirname = docName + "/";
+    if (dirname.match(/\/index\/$/))
+      dirname = dirname.substring(0, dirname.length - 6);
+    else if (dirname === "index/") dirname = "";
+    requestUrl = docUrlRoot + dirname;
+    linkUrl = requestUrl;
+  } else {
+    // normal html builders
+    requestUrl = docUrlRoot + docName + docFileSuffix;
+    linkUrl = docName + docLinkSuffix;
+  }
+  const params = new URLSearchParams();
+  params.set("highlight", [...highlightTerms].join(" "));
+  let linkEl = listItem.appendChild(document.createElement("a"));
+  linkEl.href = linkUrl + "?" + params.toString() + anchor;
+  linkEl.innerHTML = title;
+  if (descr)
+    listItem.appendChild(document.createElement("span")).innerText =
+      " (" + descr + ")";
+  else if (showSearchSummary)
+    fetch(requestUrl)
+      .then((responseData) => responseData.text())
+      .then((data) => {
+        if (data)
+          listItem.appendChild(
+            Search.makeSearchSummary(data, searchTerms, highlightTerms)
+          );
+      });
+  Search.output.appendChild(listItem);
+};
+const _finishSearch = (resultCount) => {
+  Search.stopPulse();
+  Search.title.innerText = _("Search Results");
+  if (!resultCount)
+    Search.status.innerText = Documentation.gettext(
+      "Your search did not match any documents. Please make sure that all words are spelled correctly and that you've selected enough categories."
+    );
+  else
+    Search.status.innerText = _(
+      `Search finished, found ${resultCount} page(s) matching the search query.`
+    );
+};
+const _displayNextItem = (
+  results,
+  resultCount,
+  highlightTerms,
+  searchTerms
+) => {
+  // results left, load the summary and display it
+  // this is intended to be dynamic (don't sub resultsCount)
+  if (results.length) {
+    _displayItem(results.pop(), highlightTerms, searchTerms);
+    setTimeout(
+      () => _displayNextItem(results, resultCount, highlightTerms, searchTerms),
+      5
+    );
+  }
+  // search finished, update title and status message
+  else _finishSearch(resultCount);
+};
+
+/**
+ * Default splitQuery function. Can be overridden in ``sphinx.search`` with a
+ * custom function per language.
+ *
+ * The regular expression works by splitting the string on consecutive characters
+ * that are not Unicode letters, numbers, underscores, or emoji characters.
+ * This is the same as ``\W+`` in Python, preserving the surrogate pair area.
+ */
+if (typeof splitQuery === "undefined") {
+  var splitQuery = (query) => query
+      .split(/[^\p{Letter}\p{Number}_\p{Emoji_Presentation}]+/gu)
+      .filter(term => term)  // remove remaining empty strings
+}
+
+/**
+ * Search Module
+ */
+const Search = {
+  _index: null,
+  _queued_query: null,
+  _pulse_status: -1,
+
+  htmlToText: (htmlString) => {
+    const htmlElement = document
+      .createRange()
+      .createContextualFragment(htmlString);
+    _removeChildren(htmlElement.querySelectorAll(".headerlink"));
+    const docContent = htmlElement.querySelector('[role="main"]');
+    if (docContent !== undefined) return docContent.textContent;
+    console.warn(
+      "Content block not found. Sphinx search tries to obtain it via '[role=main]'. Could you check your theme or template."
+    );
+    return "";
+  },
+
+  init: () => {
+    const query = new URLSearchParams(window.location.search).get("q");
+    document
+      .querySelectorAll('input[name="q"]')
+      .forEach((el) => (el.value = query));
+    if (query) Search.performSearch(query);
+  },
+
+  loadIndex: (url) =>
+    (document.body.appendChild(document.createElement("script")).src = url),
+
+  setIndex: (index) => {
+    Search._index = index;
+    if (Search._queued_query !== null) {
+      const query = Search._queued_query;
+      Search._queued_query = null;
+      Search.query(query);
+    }
+  },
+
+  hasIndex: () => Search._index !== null,
+
+  deferQuery: (query) => (Search._queued_query = query),
+
+  stopPulse: () => (Search._pulse_status = -1),
+
+  startPulse: () => {
+    if (Search._pulse_status >= 0) return;
+
+    const pulse = () => {
+      Search._pulse_status = (Search._pulse_status + 1) % 4;
+      Search.dots.innerText = ".".repeat(Search._pulse_status);
+      if (Search._pulse_status >= 0) window.setTimeout(pulse, 500);
+    };
+    pulse();
+  },
+
+  /**
+   * perform a search for something (or wait until index is loaded)
+   */
+  performSearch: (query) => {
+    // create the required interface elements
+    const searchText = document.createElement("h2");
+    searchText.textContent = _("Searching");
+    const searchSummary = document.createElement("p");
+    searchSummary.classList.add("search-summary");
+    searchSummary.innerText = "";
+    const searchList = document.createElement("ul");
+    searchList.classList.add("search");
+
+    const out = document.getElementById("search-results");
+    Search.title = out.appendChild(searchText);
+    Search.dots = Search.title.appendChild(document.createElement("span"));
+    Search.status = out.appendChild(searchSummary);
+    Search.output = out.appendChild(searchList);
+
+    const searchProgress = document.getElementById("search-progress");
+    // Some themes don't use the search progress node
+    if (searchProgress) {
+      searchProgress.innerText = _("Preparing search...");
+    }
+    Search.startPulse();
+
+    // index already loaded, the browser was quick!
+    if (Search.hasIndex()) Search.query(query);
+    else Search.deferQuery(query);
+  },
+
+  /**
+   * execute search (requires search index to be loaded)
+   */
+  query: (query) => {
+    // stem the search terms and add them to the correct list
+    const stemmer = new Stemmer();
+    const searchTerms = new Set();
+    const excludedTerms = new Set();
+    const highlightTerms = new Set();
+    const objectTerms = new Set(splitQuery(query.toLowerCase().trim()));
+    splitQuery(query.trim()).forEach((queryTerm) => {
+      const queryTermLower = queryTerm.toLowerCase();
+
+      // maybe skip this "word"
+      // stopwords array is from language_data.js
+      if (
+        stopwords.indexOf(queryTermLower) !== -1 ||
+        queryTerm.match(/^\d+$/)
+      )
+        return;
+
+      // stem the word
+      let word = stemmer.stemWord(queryTermLower);
+      // select the correct list
+      if (word[0] === "-") excludedTerms.add(word.substr(1));
+      else {
+        searchTerms.add(word);
+        highlightTerms.add(queryTermLower);
+      }
+    });
+
+    // console.debug("SEARCH: searching for:");
+    // console.info("required: ", [...searchTerms]);
+    // console.info("excluded: ", [...excludedTerms]);
+
+    // array of [docname, title, anchor, descr, score, filename]
+    let results = [];
+    _removeChildren(document.getElementById("search-progress"));
+
+    // lookup as object
+    objectTerms.forEach((term) =>
+      results.push(...Search.performObjectSearch(term, objectTerms))
+    );
+
+    // lookup as search terms in fulltext
+    results.push(...Search.performTermsSearch(searchTerms, excludedTerms));
+
+    // let the scorer override scores with a custom scoring function
+    if (Scorer.score) results.forEach((item) => (item[4] = Scorer.score(item)));
+
+    // now sort the results by score (in opposite order of appearance, since the
+    // display function below uses pop() to retrieve items) and then
+    // alphabetically
+    results.sort((a, b) => {
+      const leftScore = a[4];
+      const rightScore = b[4];
+      if (leftScore === rightScore) {
+        // same score: sort alphabetically
+        const leftTitle = a[1].toLowerCase();
+        const rightTitle = b[1].toLowerCase();
+        if (leftTitle === rightTitle) return 0;
+        return leftTitle > rightTitle ? -1 : 1; // inverted is intentional
+      }
+      return leftScore > rightScore ? 1 : -1;
+    });
+
+    // remove duplicate search results
+    // note the reversing of results, so that in the case of duplicates, the highest-scoring entry is kept
+    let seen = new Set();
+    results = results.reverse().reduce((acc, result) => {
+      let resultStr = result.slice(0, 4).concat([result[5]]).map(v => String(v)).join(',');
+      if (!seen.has(resultStr)) {
+        acc.push(result);
+        seen.add(resultStr);
+      }
+      return acc;
+    }, []);
+
+    results = results.reverse();
+
+    // for debugging
+    //Search.lastresults = results.slice();  // a copy
+    // console.info("search results:", Search.lastresults);
+
+    // print the results
+    _displayNextItem(results, results.length, highlightTerms, searchTerms);
+  },
+
+  /**
+   * search for object names
+   */
+  performObjectSearch: (object, objectTerms) => {
+    const filenames = Search._index.filenames;
+    const docNames = Search._index.docnames;
+    const objects = Search._index.objects;
+    const objNames = Search._index.objnames;
+    const titles = Search._index.titles;
+
+    const results = [];
+
+    const objectSearchCallback = (prefix, match) => {
+      const name = match[4]
+      const fullname = (prefix ? prefix + "." : "") + name;
+      const fullnameLower = fullname.toLowerCase();
+      if (fullnameLower.indexOf(object) < 0) return;
+
+      let score = 0;
+      const parts = fullnameLower.split(".");
+
+      // check for different match types: exact matches of full name or
+      // "last name" (i.e. last dotted part)
+      if (fullnameLower === object || parts.slice(-1)[0] === object)
+        score += Scorer.objNameMatch;
+      else if (parts.slice(-1)[0].indexOf(object) > -1)
+        score += Scorer.objPartialMatch; // matches in last name
+
+      const objName = objNames[match[1]][2];
+      const title = titles[match[0]];
+
+      // If more than one term searched for, we require other words to be
+      // found in the name/title/description
+      const otherTerms = new Set(objectTerms);
+      otherTerms.delete(object);
+      if (otherTerms.size > 0) {
+        const haystack = `${prefix} ${name} ${objName} ${title}`.toLowerCase();
+        if (
+          [...otherTerms].some((otherTerm) => haystack.indexOf(otherTerm) < 0)
+        )
+          return;
+      }
+
+      let anchor = match[3];
+      if (anchor === "") anchor = fullname;
+      else if (anchor === "-") anchor = objNames[match[1]][1] + "-" + fullname;
+
+      const descr = objName + _(", in ") + title;
+
+      // add custom score for some objects according to scorer
+      if (Scorer.objPrio.hasOwnProperty(match[2]))
+        score += Scorer.objPrio[match[2]];
+      else score += Scorer.objPrioDefault;
+
+      results.push([
+        docNames[match[0]],
+        fullname,
+        "#" + anchor,
+        descr,
+        score,
+        filenames[match[0]],
+      ]);
+    };
+    Object.keys(objects).forEach((prefix) =>
+      objects[prefix].forEach((array) =>
+        objectSearchCallback(prefix, array)
+      )
+    );
+    return results;
+  },
+
+  /**
+   * search for full-text terms in the index
+   */
+  performTermsSearch: (searchTerms, excludedTerms) => {
+    // prepare search
+    const terms = Search._index.terms;
+    const titleTerms = Search._index.titleterms;
+    const docNames = Search._index.docnames;
+    const filenames = Search._index.filenames;
+    const titles = Search._index.titles;
+
+    const scoreMap = new Map();
+    const fileMap = new Map();
+
+    // perform the search on the required terms
+    searchTerms.forEach((word) => {
+      const files = [];
+      const arr = [
+        { files: terms[word], score: Scorer.term },
+        { files: titleTerms[word], score: Scorer.title },
+      ];
+      // add support for partial matches
+      if (word.length > 2) {
+        const escapedWord = _escapeRegExp(word);
+        Object.keys(terms).forEach((term) => {
+          if (term.match(escapedWord) && !terms[word])
+            arr.push({ files: terms[term], score: Scorer.partialTerm });
+        });
+        Object.keys(titleTerms).forEach((term) => {
+          if (term.match(escapedWord) && !titleTerms[word])
+            arr.push({ files: titleTerms[word], score: Scorer.partialTitle });
+        });
+      }
+
+      // no match but word was a required one
+      if (arr.every((record) => record.files === undefined)) return;
+
+      // found search word in contents
+      arr.forEach((record) => {
+        if (record.files === undefined) return;
+
+        let recordFiles = record.files;
+        if (recordFiles.length === undefined) recordFiles = [recordFiles];
+        files.push(...recordFiles);
+
+        // set score for the word in each file
+        recordFiles.forEach((file) => {
+          if (!scoreMap.has(file)) scoreMap.set(file, {});
+          scoreMap.get(file)[word] = record.score;
+        });
+      });
+
+      // create the mapping
+      files.forEach((file) => {
+        if (fileMap.has(file) && fileMap.get(file).indexOf(word) === -1)
+          fileMap.get(file).push(word);
+        else fileMap.set(file, [word]);
+      });
+    });
+
+    // now check if the files don't contain excluded terms
+    const results = [];
+    for (const [file, wordList] of fileMap) {
+      // check if all requirements are matched
+
+      // as search terms with length < 3 are discarded
+      const filteredTermCount = [...searchTerms].filter(
+        (term) => term.length > 2
+      ).length;
+      if (
+        wordList.length !== searchTerms.size &&
+        wordList.length !== filteredTermCount
+      )
+        continue;
+
+      // ensure that none of the excluded terms is in the search result
+      if (
+        [...excludedTerms].some(
+          (term) =>
+            terms[term] === file ||
+            titleTerms[term] === file ||
+            (terms[term] || []).includes(file) ||
+            (titleTerms[term] || []).includes(file)
+        )
+      )
+        break;
+
+      // select one (max) score for the file.
+      const score = Math.max(...wordList.map((w) => scoreMap.get(file)[w]));
+      // add result to the result list
+      results.push([
+        docNames[file],
+        titles[file],
+        "",
+        null,
+        score,
+        filenames[file],
+      ]);
+    }
+    return results;
+  },
+
+  /**
+   * helper function to return a node containing the
+   * search summary for a given text. keywords is a list
+   * of stemmed words, highlightWords is the list of normal, unstemmed
+   * words. the first one is used to find the occurrence, the
+   * latter for highlighting it.
+   */
+  makeSearchSummary: (htmlText, keywords, highlightWords) => {
+    const text = Search.htmlToText(htmlText).toLowerCase();
+    if (text === "") return null;
+
+    const actualStartPosition = [...keywords]
+      .map((k) => text.indexOf(k.toLowerCase()))
+      .filter((i) => i > -1)
+      .slice(-1)[0];
+    const startWithContext = Math.max(actualStartPosition - 120, 0);
+
+    const top = startWithContext === 0 ? "" : "...";
+    const tail = startWithContext + 240 < text.length ? "..." : "";
+
+    let summary = document.createElement("div");
+    summary.classList.add("context");
+    summary.innerText = top + text.substr(startWithContext, 240).trim() + tail;
+
+    highlightWords.forEach((highlightWord) =>
+      _highlightText(summary, highlightWord, "highlighted")
+    );
+
+    return summary;
+  },
+};
+
+_ready(Search.init);
diff --git a/0.16/_static/sg_gallery-binder.css b/0.16/_static/sg_gallery-binder.css
new file mode 100644
index 00000000000..420005d2270
--- /dev/null
+++ b/0.16/_static/sg_gallery-binder.css
@@ -0,0 +1,11 @@
+/* CSS for binder integration */
+
+div.binder-badge {
+  margin: 1em auto;
+  vertical-align: middle;
+}
+
+div.lite-badge {
+  margin: 1em auto;
+  vertical-align: middle;
+}
diff --git a/0.16/_static/sg_gallery-dataframe.css b/0.16/_static/sg_gallery-dataframe.css
new file mode 100644
index 00000000000..fac74c43b7c
--- /dev/null
+++ b/0.16/_static/sg_gallery-dataframe.css
@@ -0,0 +1,47 @@
+/* Pandas dataframe css */
+/* Taken from: https://github.com/spatialaudio/nbsphinx/blob/fb3ba670fc1ba5f54d4c487573dbc1b4ecf7e9ff/src/nbsphinx.py#L587-L619 */
+html[data-theme="light"] {
+  --sg-text-color: #000;
+  --sg-tr-odd-color: #f5f5f5;
+  --sg-tr-hover-color: rgba(66, 165, 245, 0.2);
+}
+html[data-theme="dark"] {
+  --sg-text-color: #fff;
+  --sg-tr-odd-color: #373737;
+  --sg-tr-hover-color: rgba(30, 81, 122, 0.2);
+}
+
+table.dataframe {
+  border: none !important;
+  border-collapse: collapse;
+  border-spacing: 0;
+  border-color: transparent;
+  color: var(--sg-text-color);
+  font-size: 12px;
+  table-layout: fixed;
+  width: auto;
+}
+table.dataframe thead {
+  border-bottom: 1px solid var(--sg-text-color);
+  vertical-align: bottom;
+}
+table.dataframe tr,
+table.dataframe th,
+table.dataframe td {
+  text-align: right;
+  vertical-align: middle;
+  padding: 0.5em 0.5em;
+  line-height: normal;
+  white-space: normal;
+  max-width: none;
+  border: none;
+}
+table.dataframe th {
+  font-weight: bold;
+}
+table.dataframe tbody tr:nth-child(odd) {
+  background: var(--sg-tr-odd-color);
+}
+table.dataframe tbody tr:hover {
+  background: var(--sg-tr-hover-color);
+}
diff --git a/0.16/_static/sg_gallery-rendered-html.css b/0.16/_static/sg_gallery-rendered-html.css
new file mode 100644
index 00000000000..93dc2ffb024
--- /dev/null
+++ b/0.16/_static/sg_gallery-rendered-html.css
@@ -0,0 +1,224 @@
+/* Adapted from notebook/static/style/style.min.css */
+html[data-theme="light"] {
+  --sg-text-color: #000;
+  --sg-background-color: #ffffff;
+  --sg-code-background-color: #eff0f1;
+  --sg-tr-hover-color: rgba(66, 165, 245, 0.2);
+  --sg-tr-odd-color: #f5f5f5;
+}
+html[data-theme="dark"] {
+  --sg-text-color: #fff;
+  --sg-background-color: #121212;
+  --sg-code-background-color: #2f2f30;
+  --sg-tr-hover-color: rgba(66, 165, 245, 0.2);
+  --sg-tr-odd-color: #1f1f1f;
+}
+
+.rendered_html {
+  color: var(--sg-text-color);
+  /* any extras will just be numbers: */
+}
+.rendered_html em {
+  font-style: italic;
+}
+.rendered_html strong {
+  font-weight: bold;
+}
+.rendered_html u {
+  text-decoration: underline;
+}
+.rendered_html :link {
+  text-decoration: underline;
+}
+.rendered_html :visited {
+  text-decoration: underline;
+}
+.rendered_html h1 {
+  font-size: 185.7%;
+  margin: 1.08em 0 0 0;
+  font-weight: bold;
+  line-height: 1.0;
+}
+.rendered_html h2 {
+  font-size: 157.1%;
+  margin: 1.27em 0 0 0;
+  font-weight: bold;
+  line-height: 1.0;
+}
+.rendered_html h3 {
+  font-size: 128.6%;
+  margin: 1.55em 0 0 0;
+  font-weight: bold;
+  line-height: 1.0;
+}
+.rendered_html h4 {
+  font-size: 100%;
+  margin: 2em 0 0 0;
+  font-weight: bold;
+  line-height: 1.0;
+}
+.rendered_html h5 {
+  font-size: 100%;
+  margin: 2em 0 0 0;
+  font-weight: bold;
+  line-height: 1.0;
+  font-style: italic;
+}
+.rendered_html h6 {
+  font-size: 100%;
+  margin: 2em 0 0 0;
+  font-weight: bold;
+  line-height: 1.0;
+  font-style: italic;
+}
+.rendered_html h1:first-child {
+  margin-top: 0.538em;
+}
+.rendered_html h2:first-child {
+  margin-top: 0.636em;
+}
+.rendered_html h3:first-child {
+  margin-top: 0.777em;
+}
+.rendered_html h4:first-child {
+  margin-top: 1em;
+}
+.rendered_html h5:first-child {
+  margin-top: 1em;
+}
+.rendered_html h6:first-child {
+  margin-top: 1em;
+}
+.rendered_html ul:not(.list-inline),
+.rendered_html ol:not(.list-inline) {
+  padding-left: 2em;
+}
+.rendered_html ul {
+  list-style: disc;
+}
+.rendered_html ul ul {
+  list-style: square;
+  margin-top: 0;
+}
+.rendered_html ul ul ul {
+  list-style: circle;
+}
+.rendered_html ol {
+  list-style: decimal;
+}
+.rendered_html ol ol {
+  list-style: upper-alpha;
+  margin-top: 0;
+}
+.rendered_html ol ol ol {
+  list-style: lower-alpha;
+}
+.rendered_html ol ol ol ol {
+  list-style: lower-roman;
+}
+.rendered_html ol ol ol ol ol {
+  list-style: decimal;
+}
+.rendered_html * + ul {
+  margin-top: 1em;
+}
+.rendered_html * + ol {
+  margin-top: 1em;
+}
+.rendered_html hr {
+  color: var(--sg-text-color);
+  background-color: var(--sg-text-color);
+}
+.rendered_html pre {
+  margin: 1em 2em;
+  padding: 0px;
+  background-color: var(--sg-background-color);
+}
+.rendered_html code {
+  background-color: var(--sg-code-background-color);
+}
+.rendered_html p code {
+  padding: 1px 5px;
+}
+.rendered_html pre code {
+  background-color: var(--sg-background-color);
+}
+.rendered_html pre,
+.rendered_html code {
+  border: 0;
+  color: var(--sg-text-color);
+  font-size: 100%;
+}
+.rendered_html blockquote {
+  margin: 1em 2em;
+}
+.rendered_html table {
+  margin-left: auto;
+  margin-right: auto;
+  border: none;
+  border-collapse: collapse;
+  border-spacing: 0;
+  color: var(--sg-text-color);
+  font-size: 12px;
+  table-layout: fixed;
+}
+.rendered_html thead {
+  border-bottom: 1px solid var(--sg-text-color);
+  vertical-align: bottom;
+}
+.rendered_html tr,
+.rendered_html th,
+.rendered_html td {
+  text-align: right;
+  vertical-align: middle;
+  padding: 0.5em 0.5em;
+  line-height: normal;
+  white-space: normal;
+  max-width: none;
+  border: none;
+}
+.rendered_html th {
+  font-weight: bold;
+}
+.rendered_html tbody tr:nth-child(odd) {
+  background: var(--sg-tr-odd-color);
+}
+.rendered_html tbody tr:hover {
+  color: var(--sg-text-color);
+  background: var(--sg-tr-hover-color);
+}
+.rendered_html * + table {
+  margin-top: 1em;
+}
+.rendered_html p {
+  text-align: left;
+}
+.rendered_html * + p {
+  margin-top: 1em;
+}
+.rendered_html img {
+  display: block;
+  margin-left: auto;
+  margin-right: auto;
+}
+.rendered_html * + img {
+  margin-top: 1em;
+}
+.rendered_html img,
+.rendered_html svg {
+  max-width: 100%;
+  height: auto;
+}
+.rendered_html img.unconfined,
+.rendered_html svg.unconfined {
+  max-width: none;
+}
+.rendered_html .alert {
+  margin-bottom: initial;
+}
+.rendered_html * + .alert {
+  margin-top: 1em;
+}
+[dir="rtl"] .rendered_html p {
+  text-align: right;
+}
diff --git a/0.16/_static/sg_gallery.css b/0.16/_static/sg_gallery.css
new file mode 100644
index 00000000000..72227837d38
--- /dev/null
+++ b/0.16/_static/sg_gallery.css
@@ -0,0 +1,342 @@
+/*
+Sphinx-Gallery has compatible CSS to fix default sphinx themes
+Tested for Sphinx 1.3.1 for all themes: default, alabaster, sphinxdoc,
+scrolls, agogo, traditional, nature, haiku, pyramid
+Tested for Read the Docs theme 0.1.7 */
+
+/* Define light colors */
+:root, html[data-theme="light"], body[data-theme="light"]{
+  --sg-tooltip-foreground: black;
+  --sg-tooltip-background: rgba(250, 250, 250, 0.9);
+  --sg-tooltip-border: #ccc transparent;
+  --sg-thumb-box-shadow-color: #6c757d40;
+  --sg-thumb-hover-border: #0069d9;
+  --sg-script-out: #888;
+  --sg-script-pre: #fafae2;
+  --sg-pytb-foreground: #000;
+  --sg-pytb-background: #ffe4e4;
+  --sg-pytb-border-color: #f66;
+  --sg-download-a-background-color: #ffc;
+  --sg-download-a-background-image: linear-gradient(to bottom, #ffc, #d5d57e);
+  --sg-download-a-border-color: 1px solid #c2c22d;
+  --sg-download-a-color: #000;
+  --sg-download-a-hover-background-color: #d5d57e;
+  --sg-download-a-hover-box-shadow-1: rgba(255, 255, 255, 0.1);
+  --sg-download-a-hover-box-shadow-2: rgba(0, 0, 0, 0.25);
+}
+@media(prefers-color-scheme: light) {
+  :root[data-theme="auto"], html[data-theme="auto"], body[data-theme="auto"] {
+    --sg-tooltip-foreground: black;
+    --sg-tooltip-background: rgba(250, 250, 250, 0.9);
+    --sg-tooltip-border: #ccc transparent;
+    --sg-thumb-box-shadow-color: #6c757d40;
+    --sg-thumb-hover-border: #0069d9;
+    --sg-script-out: #888;
+    --sg-script-pre: #fafae2;
+    --sg-pytb-foreground: #000;
+    --sg-pytb-background: #ffe4e4;
+    --sg-pytb-border-color: #f66;
+    --sg-download-a-background-color: #ffc;
+    --sg-download-a-background-image: linear-gradient(to bottom, #ffc, #d5d57e);
+    --sg-download-a-border-color: 1px solid #c2c22d;
+    --sg-download-a-color: #000;
+    --sg-download-a-hover-background-color: #d5d57e;
+    --sg-download-a-hover-box-shadow-1: rgba(255, 255, 255, 0.1);
+    --sg-download-a-hover-box-shadow-2: rgba(0, 0, 0, 0.25);
+  }
+}
+
+html[data-theme="dark"], body[data-theme="dark"] {
+  --sg-tooltip-foreground: white;
+  --sg-tooltip-background: rgba(10, 10, 10, 0.9);
+  --sg-tooltip-border: #333 transparent;
+  --sg-thumb-box-shadow-color: #79848d40;
+  --sg-thumb-hover-border: #003975;
+  --sg-script-out: rgb(179, 179, 179);
+  --sg-script-pre: #2e2e22;
+  --sg-pytb-foreground: #fff;
+  --sg-pytb-background: #1b1717;
+  --sg-pytb-border-color: #622;
+  --sg-download-a-background-color: #443;
+  --sg-download-a-background-image: linear-gradient(to bottom, #443, #221);
+  --sg-download-a-border-color: 1px solid #3a3a0d;
+  --sg-download-a-color: #fff;
+  --sg-download-a-hover-background-color: #616135;
+  --sg-download-a-hover-box-shadow-1: rgba(0, 0, 0, 0.1);
+  --sg-download-a-hover-box-shadow-2: rgba(255, 255, 255, 0.25);
+}
+@media(prefers-color-scheme: dark){
+  html[data-theme="auto"], body[data-theme="auto"] {
+    --sg-tooltip-foreground: white;
+    --sg-tooltip-background: rgba(10, 10, 10, 0.9);
+    --sg-tooltip-border: #333 transparent;
+    --sg-thumb-box-shadow-color: #79848d40;
+    --sg-thumb-hover-border: #003975;
+    --sg-script-out: rgb(179, 179, 179);
+    --sg-script-pre: #2e2e22;
+    --sg-pytb-foreground: #fff;
+    --sg-pytb-background: #1b1717;
+    --sg-pytb-border-color: #622;
+    --sg-download-a-background-color: #443;
+    --sg-download-a-background-image: linear-gradient(to bottom, #443, #221);
+    --sg-download-a-border-color: 1px solid #3a3a0d;
+    --sg-download-a-color: #fff;
+    --sg-download-a-hover-background-color: #616135;
+    --sg-download-a-hover-box-shadow-1: rgba(0, 0, 0, 0.1);
+    --sg-download-a-hover-box-shadow-2: rgba(255, 255, 255, 0.25);
+  }
+}
+
+.sphx-glr-thumbnails {
+  width: 100%;
+  margin: 0px 0px 20px 0px;
+
+  /* align thumbnails on a grid */
+  justify-content: space-between;
+  display: grid;
+  /* each grid column should be at least 160px (this will determine
+  the actual number of columns) and then take as much of the
+  remaining width as possible */
+  grid-template-columns: repeat(auto-fill, minmax(160px, 1fr));
+  gap: 15px;
+}
+.sphx-glr-thumbnails .toctree-wrapper {
+  /* hide empty toctree divs added to the DOM
+  by sphinx even though the toctree is hidden
+  (they would fill grid places with empty divs) */
+  display: none;
+}
+.sphx-glr-thumbcontainer {
+  background: transparent;
+  -moz-border-radius: 5px;
+  -webkit-border-radius: 5px;
+  border-radius: 5px;
+  box-shadow: 0 0 10px var(--sg-thumb-box-shadow-color);
+
+  /* useful to absolutely position link in div */
+  position: relative;
+
+  /* thumbnail width should include padding and borders
+  and take all available space */
+  box-sizing: border-box;
+  width: 100%;
+  padding: 10px;
+  border: 1px solid transparent;
+
+  /* align content in thumbnail */
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  gap: 7px;
+}
+.sphx-glr-thumbcontainer p {
+  position: absolute;
+  top: 0;
+  left: 0;
+}
+.sphx-glr-thumbcontainer p,
+.sphx-glr-thumbcontainer p a {
+  /* link should cover the whole thumbnail div */
+  width: 100%;
+  height: 100%;
+}
+.sphx-glr-thumbcontainer p a span {
+  /* text within link should be masked
+  (we are just interested in the href) */
+  display: none;
+}
+.sphx-glr-thumbcontainer:hover {
+  border: 1px solid;
+  border-color: var(--sg-thumb-hover-border);
+  cursor: pointer;
+}
+.sphx-glr-thumbcontainer a.internal {
+  bottom: 0;
+  display: block;
+  left: 0;
+  box-sizing: border-box;
+  padding: 150px 10px 0;
+  position: absolute;
+  right: 0;
+  top: 0;
+}
+/* Next one is to avoid Sphinx traditional theme to cover all the
+thumbnail with its default link Background color */
+.sphx-glr-thumbcontainer a.internal:hover {
+  background-color: transparent;
+}
+
+.sphx-glr-thumbcontainer p {
+  margin: 0 0 0.1em 0;
+}
+.sphx-glr-thumbcontainer .figure {
+  margin: 10px;
+  width: 160px;
+}
+.sphx-glr-thumbcontainer img {
+  display: inline;
+  max-height: 112px;
+  max-width: 160px;
+}
+.sphx-glr-thumbcontainer[tooltip]:hover:after {
+  background: var(--sg-tooltip-background);
+  -webkit-border-radius: 4px;
+  -moz-border-radius: 4px;
+  border-radius: 4px;
+  color: var(--sg-tooltip-foreground);
+  content: attr(tooltip);
+  padding: 10px;
+  z-index: 98;
+  width: 100%;
+  height: 100%;
+  position: absolute;
+  pointer-events: none;
+  top: 0;
+  box-sizing: border-box;
+  overflow: hidden;
+  backdrop-filter: blur(3px);
+}
+
+.sphx-glr-script-out {
+  color: var(--sg-script-out);
+  display: flex;
+  gap: 0.5em;
+}
+.sphx-glr-script-out::before {
+  content: "Out:";
+  /* These numbers come from the pre style in the pydata sphinx theme. This
+   * turns out to match perfectly on the rtd theme, but be a bit too low for
+   * the pydata sphinx theme. As I could not find a dimension to use that was
+   * scaled the same way, I just picked one option that worked pretty close for
+   * both. */
+  line-height: 1.4;
+  padding-top: 10px;
+}
+.sphx-glr-script-out .highlight {
+  background-color: transparent;
+  /* These options make the div expand... */
+  flex-grow: 1;
+  /* ... but also keep it from overflowing its flex container. */
+  overflow: auto;
+}
+.sphx-glr-script-out .highlight pre {
+  background-color: var(--sg-script-pre);
+  border: 0;
+  max-height: 30em;
+  overflow: auto;
+  padding-left: 1ex;
+  /* This margin is necessary in the pydata sphinx theme because pre has a box
+   * shadow which would be clipped by the overflow:auto in the parent div
+   * above. */
+  margin: 2px;
+  word-break: break-word;
+}
+.sphx-glr-script-out + p {
+  margin-top: 1.8em;
+}
+blockquote.sphx-glr-script-out {
+  margin-left: 0pt;
+}
+.sphx-glr-script-out.highlight-pytb .highlight pre {
+  color: var(--sg-pytb-foreground);
+  background-color: var(--sg-pytb-background);
+  border: 1px solid var(--sg-pytb-border-color);
+  margin-top: 10px;
+  padding: 7px;
+}
+
+div.sphx-glr-footer {
+  text-align: center;
+}
+
+div.sphx-glr-download {
+  margin: 1em auto;
+  vertical-align: middle;
+}
+
+div.sphx-glr-download a {
+  background-color: var(--sg-download-a-background-color);
+  background-image: var(--sg-download-a-background-image);
+  border-radius: 4px;
+  border: 1px solid var(--sg-download-a-border-color);
+  color: var(--sg-download-a-color);
+  display: inline-block;
+  font-weight: bold;
+  padding: 1ex;
+  text-align: center;
+}
+
+div.sphx-glr-download code.download {
+  display: inline-block;
+  white-space: normal;
+  word-break: normal;
+  overflow-wrap: break-word;
+  /* border and background are given by the enclosing 'a' */
+  border: none;
+  background: none;
+}
+
+div.sphx-glr-download a:hover {
+  box-shadow: inset 0 1px 0 var(--sg-download-a-hover-box-shadow-1), 0 1px 5px var(--sg-download-a-hover-box-shadow-2);
+  text-decoration: none;
+  background-image: none;
+  background-color: var(--sg-download-a-hover-background-color);
+}
+
+.sphx-glr-example-title:target::before {
+  display: block;
+  content: "";
+  margin-top: -50px;
+  height: 50px;
+  visibility: hidden;
+}
+
+ul.sphx-glr-horizontal {
+  list-style: none;
+  padding: 0;
+}
+ul.sphx-glr-horizontal li {
+  display: inline;
+}
+ul.sphx-glr-horizontal img {
+  height: auto !important;
+}
+
+.sphx-glr-single-img {
+  margin: auto;
+  display: block;
+  max-width: 100%;
+}
+
+.sphx-glr-multi-img {
+  max-width: 42%;
+  height: auto;
+}
+
+div.sphx-glr-animation {
+  margin: auto;
+  display: block;
+  max-width: 100%;
+}
+div.sphx-glr-animation .animation {
+  display: block;
+}
+
+p.sphx-glr-signature a.reference.external {
+  -moz-border-radius: 5px;
+  -webkit-border-radius: 5px;
+  border-radius: 5px;
+  padding: 3px;
+  font-size: 75%;
+  text-align: right;
+  margin-left: auto;
+  display: table;
+}
+
+.sphx-glr-clear {
+  clear: both;
+}
+
+a.sphx-glr-backref-instance {
+  text-decoration: none;
+}
diff --git a/0.16/_static/underscore-1.13.1.js b/0.16/_static/underscore-1.13.1.js
new file mode 100644
index 00000000000..ffd77af9648
--- /dev/null
+++ b/0.16/_static/underscore-1.13.1.js
@@ -0,0 +1,2042 @@
+(function (global, factory) {
+  typeof exports === 'object' && typeof module !== 'undefined' ? module.exports = factory() :
+  typeof define === 'function' && define.amd ? define('underscore', factory) :
+  (global = typeof globalThis !== 'undefined' ? globalThis : global || self, (function () {
+    var current = global._;
+    var exports = global._ = factory();
+    exports.noConflict = function () { global._ = current; return exports; };
+  }()));
+}(this, (function () {
+  //     Underscore.js 1.13.1
+  //     https://underscorejs.org
+  //     (c) 2009-2021 Jeremy Ashkenas, Julian Gonggrijp, and DocumentCloud and Investigative Reporters & Editors
+  //     Underscore may be freely distributed under the MIT license.
+
+  // Current version.
+  var VERSION = '1.13.1';
+
+  // Establish the root object, `window` (`self`) in the browser, `global`
+  // on the server, or `this` in some virtual machines. We use `self`
+  // instead of `window` for `WebWorker` support.
+  var root = typeof self == 'object' && self.self === self && self ||
+            typeof global == 'object' && global.global === global && global ||
+            Function('return this')() ||
+            {};
+
+  // Save bytes in the minified (but not gzipped) version:
+  var ArrayProto = Array.prototype, ObjProto = Object.prototype;
+  var SymbolProto = typeof Symbol !== 'undefined' ? Symbol.prototype : null;
+
+  // Create quick reference variables for speed access to core prototypes.
+  var push = ArrayProto.push,
+      slice = ArrayProto.slice,
+      toString = ObjProto.toString,
+      hasOwnProperty = ObjProto.hasOwnProperty;
+
+  // Modern feature detection.
+  var supportsArrayBuffer = typeof ArrayBuffer !== 'undefined',
+      supportsDataView = typeof DataView !== 'undefined';
+
+  // All **ECMAScript 5+** native function implementations that we hope to use
+  // are declared here.
+  var nativeIsArray = Array.isArray,
+      nativeKeys = Object.keys,
+      nativeCreate = Object.create,
+      nativeIsView = supportsArrayBuffer && ArrayBuffer.isView;
+
+  // Create references to these builtin functions because we override them.
+  var _isNaN = isNaN,
+      _isFinite = isFinite;
+
+  // Keys in IE < 9 that won't be iterated by `for key in ...` and thus missed.
+  var hasEnumBug = !{toString: null}.propertyIsEnumerable('toString');
+  var nonEnumerableProps = ['valueOf', 'isPrototypeOf', 'toString',
+    'propertyIsEnumerable', 'hasOwnProperty', 'toLocaleString'];
+
+  // The largest integer that can be represented exactly.
+  var MAX_ARRAY_INDEX = Math.pow(2, 53) - 1;
+
+  // Some functions take a variable number of arguments, or a few expected
+  // arguments at the beginning and then a variable number of values to operate
+  // on. This helper accumulates all remaining arguments past the function’s
+  // argument length (or an explicit `startIndex`), into an array that becomes
+  // the last argument. Similar to ES6’s "rest parameter".
+  function restArguments(func, startIndex) {
+    startIndex = startIndex == null ? func.length - 1 : +startIndex;
+    return function() {
+      var length = Math.max(arguments.length - startIndex, 0),
+          rest = Array(length),
+          index = 0;
+      for (; index < length; index++) {
+        rest[index] = arguments[index + startIndex];
+      }
+      switch (startIndex) {
+        case 0: return func.call(this, rest);
+        case 1: return func.call(this, arguments[0], rest);
+        case 2: return func.call(this, arguments[0], arguments[1], rest);
+      }
+      var args = Array(startIndex + 1);
+      for (index = 0; index < startIndex; index++) {
+        args[index] = arguments[index];
+      }
+      args[startIndex] = rest;
+      return func.apply(this, args);
+    };
+  }
+
+  // Is a given variable an object?
+  function isObject(obj) {
+    var type = typeof obj;
+    return type === 'function' || type === 'object' && !!obj;
+  }
+
+  // Is a given value equal to null?
+  function isNull(obj) {
+    return obj === null;
+  }
+
+  // Is a given variable undefined?
+  function isUndefined(obj) {
+    return obj === void 0;
+  }
+
+  // Is a given value a boolean?
+  function isBoolean(obj) {
+    return obj === true || obj === false || toString.call(obj) === '[object Boolean]';
+  }
+
+  // Is a given value a DOM element?
+  function isElement(obj) {
+    return !!(obj && obj.nodeType === 1);
+  }
+
+  // Internal function for creating a `toString`-based type tester.
+  function tagTester(name) {
+    var tag = '[object ' + name + ']';
+    return function(obj) {
+      return toString.call(obj) === tag;
+    };
+  }
+
+  var isString = tagTester('String');
+
+  var isNumber = tagTester('Number');
+
+  var isDate = tagTester('Date');
+
+  var isRegExp = tagTester('RegExp');
+
+  var isError = tagTester('Error');
+
+  var isSymbol = tagTester('Symbol');
+
+  var isArrayBuffer = tagTester('ArrayBuffer');
+
+  var isFunction = tagTester('Function');
+
+  // Optimize `isFunction` if appropriate. Work around some `typeof` bugs in old
+  // v8, IE 11 (#1621), Safari 8 (#1929), and PhantomJS (#2236).
+  var nodelist = root.document && root.document.childNodes;
+  if (typeof /./ != 'function' && typeof Int8Array != 'object' && typeof nodelist != 'function') {
+    isFunction = function(obj) {
+      return typeof obj == 'function' || false;
+    };
+  }
+
+  var isFunction$1 = isFunction;
+
+  var hasObjectTag = tagTester('Object');
+
+  // In IE 10 - Edge 13, `DataView` has string tag `'[object Object]'`.
+  // In IE 11, the most common among them, this problem also applies to
+  // `Map`, `WeakMap` and `Set`.
+  var hasStringTagBug = (
+        supportsDataView && hasObjectTag(new DataView(new ArrayBuffer(8)))
+      ),
+      isIE11 = (typeof Map !== 'undefined' && hasObjectTag(new Map));
+
+  var isDataView = tagTester('DataView');
+
+  // In IE 10 - Edge 13, we need a different heuristic
+  // to determine whether an object is a `DataView`.
+  function ie10IsDataView(obj) {
+    return obj != null && isFunction$1(obj.getInt8) && isArrayBuffer(obj.buffer);
+  }
+
+  var isDataView$1 = (hasStringTagBug ? ie10IsDataView : isDataView);
+
+  // Is a given value an array?
+  // Delegates to ECMA5's native `Array.isArray`.
+  var isArray = nativeIsArray || tagTester('Array');
+
+  // Internal function to check whether `key` is an own property name of `obj`.
+  function has$1(obj, key) {
+    return obj != null && hasOwnProperty.call(obj, key);
+  }
+
+  var isArguments = tagTester('Arguments');
+
+  // Define a fallback version of the method in browsers (ahem, IE < 9), where
+  // there isn't any inspectable "Arguments" type.
+  (function() {
+    if (!isArguments(arguments)) {
+      isArguments = function(obj) {
+        return has$1(obj, 'callee');
+      };
+    }
+  }());
+
+  var isArguments$1 = isArguments;
+
+  // Is a given object a finite number?
+  function isFinite$1(obj) {
+    return !isSymbol(obj) && _isFinite(obj) && !isNaN(parseFloat(obj));
+  }
+
+  // Is the given value `NaN`?
+  function isNaN$1(obj) {
+    return isNumber(obj) && _isNaN(obj);
+  }
+
+  // Predicate-generating function. Often useful outside of Underscore.
+  function constant(value) {
+    return function() {
+      return value;
+    };
+  }
+
+  // Common internal logic for `isArrayLike` and `isBufferLike`.
+  function createSizePropertyCheck(getSizeProperty) {
+    return function(collection) {
+      var sizeProperty = getSizeProperty(collection);
+      return typeof sizeProperty == 'number' && sizeProperty >= 0 && sizeProperty <= MAX_ARRAY_INDEX;
+    }
+  }
+
+  // Internal helper to generate a function to obtain property `key` from `obj`.
+  function shallowProperty(key) {
+    return function(obj) {
+      return obj == null ? void 0 : obj[key];
+    };
+  }
+
+  // Internal helper to obtain the `byteLength` property of an object.
+  var getByteLength = shallowProperty('byteLength');
+
+  // Internal helper to determine whether we should spend extensive checks against
+  // `ArrayBuffer` et al.
+  var isBufferLike = createSizePropertyCheck(getByteLength);
+
+  // Is a given value a typed array?
+  var typedArrayPattern = /\[object ((I|Ui)nt(8|16|32)|Float(32|64)|Uint8Clamped|Big(I|Ui)nt64)Array\]/;
+  function isTypedArray(obj) {
+    // `ArrayBuffer.isView` is the most future-proof, so use it when available.
+    // Otherwise, fall back on the above regular expression.
+    return nativeIsView ? (nativeIsView(obj) && !isDataView$1(obj)) :
+                  isBufferLike(obj) && typedArrayPattern.test(toString.call(obj));
+  }
+
+  var isTypedArray$1 = supportsArrayBuffer ? isTypedArray : constant(false);
+
+  // Internal helper to obtain the `length` property of an object.
+  var getLength = shallowProperty('length');
+
+  // Internal helper to create a simple lookup structure.
+  // `collectNonEnumProps` used to depend on `_.contains`, but this led to
+  // circular imports. `emulatedSet` is a one-off solution that only works for
+  // arrays of strings.
+  function emulatedSet(keys) {
+    var hash = {};
+    for (var l = keys.length, i = 0; i < l; ++i) hash[keys[i]] = true;
+    return {
+      contains: function(key) { return hash[key]; },
+      push: function(key) {
+        hash[key] = true;
+        return keys.push(key);
+      }
+    };
+  }
+
+  // Internal helper. Checks `keys` for the presence of keys in IE < 9 that won't
+  // be iterated by `for key in ...` and thus missed. Extends `keys` in place if
+  // needed.
+  function collectNonEnumProps(obj, keys) {
+    keys = emulatedSet(keys);
+    var nonEnumIdx = nonEnumerableProps.length;
+    var constructor = obj.constructor;
+    var proto = isFunction$1(constructor) && constructor.prototype || ObjProto;
+
+    // Constructor is a special case.
+    var prop = 'constructor';
+    if (has$1(obj, prop) && !keys.contains(prop)) keys.push(prop);
+
+    while (nonEnumIdx--) {
+      prop = nonEnumerableProps[nonEnumIdx];
+      if (prop in obj && obj[prop] !== proto[prop] && !keys.contains(prop)) {
+        keys.push(prop);
+      }
+    }
+  }
+
+  // Retrieve the names of an object's own properties.
+  // Delegates to **ECMAScript 5**'s native `Object.keys`.
+  function keys(obj) {
+    if (!isObject(obj)) return [];
+    if (nativeKeys) return nativeKeys(obj);
+    var keys = [];
+    for (var key in obj) if (has$1(obj, key)) keys.push(key);
+    // Ahem, IE < 9.
+    if (hasEnumBug) collectNonEnumProps(obj, keys);
+    return keys;
+  }
+
+  // Is a given array, string, or object empty?
+  // An "empty" object has no enumerable own-properties.
+  function isEmpty(obj) {
+    if (obj == null) return true;
+    // Skip the more expensive `toString`-based type checks if `obj` has no
+    // `.length`.
+    var length = getLength(obj);
+    if (typeof length == 'number' && (
+      isArray(obj) || isString(obj) || isArguments$1(obj)
+    )) return length === 0;
+    return getLength(keys(obj)) === 0;
+  }
+
+  // Returns whether an object has a given set of `key:value` pairs.
+  function isMatch(object, attrs) {
+    var _keys = keys(attrs), length = _keys.length;
+    if (object == null) return !length;
+    var obj = Object(object);
+    for (var i = 0; i < length; i++) {
+      var key = _keys[i];
+      if (attrs[key] !== obj[key] || !(key in obj)) return false;
+    }
+    return true;
+  }
+
+  // If Underscore is called as a function, it returns a wrapped object that can
+  // be used OO-style. This wrapper holds altered versions of all functions added
+  // through `_.mixin`. Wrapped objects may be chained.
+  function _$1(obj) {
+    if (obj instanceof _$1) return obj;
+    if (!(this instanceof _$1)) return new _$1(obj);
+    this._wrapped = obj;
+  }
+
+  _$1.VERSION = VERSION;
+
+  // Extracts the result from a wrapped and chained object.
+  _$1.prototype.value = function() {
+    return this._wrapped;
+  };
+
+  // Provide unwrapping proxies for some methods used in engine operations
+  // such as arithmetic and JSON stringification.
+  _$1.prototype.valueOf = _$1.prototype.toJSON = _$1.prototype.value;
+
+  _$1.prototype.toString = function() {
+    return String(this._wrapped);
+  };
+
+  // Internal function to wrap or shallow-copy an ArrayBuffer,
+  // typed array or DataView to a new view, reusing the buffer.
+  function toBufferView(bufferSource) {
+    return new Uint8Array(
+      bufferSource.buffer || bufferSource,
+      bufferSource.byteOffset || 0,
+      getByteLength(bufferSource)
+    );
+  }
+
+  // We use this string twice, so give it a name for minification.
+  var tagDataView = '[object DataView]';
+
+  // Internal recursive comparison function for `_.isEqual`.
+  function eq(a, b, aStack, bStack) {
+    // Identical objects are equal. `0 === -0`, but they aren't identical.
+    // See the [Harmony `egal` proposal](https://wiki.ecmascript.org/doku.php?id=harmony:egal).
+    if (a === b) return a !== 0 || 1 / a === 1 / b;
+    // `null` or `undefined` only equal to itself (strict comparison).
+    if (a == null || b == null) return false;
+    // `NaN`s are equivalent, but non-reflexive.
+    if (a !== a) return b !== b;
+    // Exhaust primitive checks
+    var type = typeof a;
+    if (type !== 'function' && type !== 'object' && typeof b != 'object') return false;
+    return deepEq(a, b, aStack, bStack);
+  }
+
+  // Internal recursive comparison function for `_.isEqual`.
+  function deepEq(a, b, aStack, bStack) {
+    // Unwrap any wrapped objects.
+    if (a instanceof _$1) a = a._wrapped;
+    if (b instanceof _$1) b = b._wrapped;
+    // Compare `[[Class]]` names.
+    var className = toString.call(a);
+    if (className !== toString.call(b)) return false;
+    // Work around a bug in IE 10 - Edge 13.
+    if (hasStringTagBug && className == '[object Object]' && isDataView$1(a)) {
+      if (!isDataView$1(b)) return false;
+      className = tagDataView;
+    }
+    switch (className) {
+      // These types are compared by value.
+      case '[object RegExp]':
+        // RegExps are coerced to strings for comparison (Note: '' + /a/i === '/a/i')
+      case '[object String]':
+        // Primitives and their corresponding object wrappers are equivalent; thus, `"5"` is
+        // equivalent to `new String("5")`.
+        return '' + a === '' + b;
+      case '[object Number]':
+        // `NaN`s are equivalent, but non-reflexive.
+        // Object(NaN) is equivalent to NaN.
+        if (+a !== +a) return +b !== +b;
+        // An `egal` comparison is performed for other numeric values.
+        return +a === 0 ? 1 / +a === 1 / b : +a === +b;
+      case '[object Date]':
+      case '[object Boolean]':
+        // Coerce dates and booleans to numeric primitive values. Dates are compared by their
+        // millisecond representations. Note that invalid dates with millisecond representations
+        // of `NaN` are not equivalent.
+        return +a === +b;
+      case '[object Symbol]':
+        return SymbolProto.valueOf.call(a) === SymbolProto.valueOf.call(b);
+      case '[object ArrayBuffer]':
+      case tagDataView:
+        // Coerce to typed array so we can fall through.
+        return deepEq(toBufferView(a), toBufferView(b), aStack, bStack);
+    }
+
+    var areArrays = className === '[object Array]';
+    if (!areArrays && isTypedArray$1(a)) {
+        var byteLength = getByteLength(a);
+        if (byteLength !== getByteLength(b)) return false;
+        if (a.buffer === b.buffer && a.byteOffset === b.byteOffset) return true;
+        areArrays = true;
+    }
+    if (!areArrays) {
+      if (typeof a != 'object' || typeof b != 'object') return false;
+
+      // Objects with different constructors are not equivalent, but `Object`s or `Array`s
+      // from different frames are.
+      var aCtor = a.constructor, bCtor = b.constructor;
+      if (aCtor !== bCtor && !(isFunction$1(aCtor) && aCtor instanceof aCtor &&
+                               isFunction$1(bCtor) && bCtor instanceof bCtor)
+                          && ('constructor' in a && 'constructor' in b)) {
+        return false;
+      }
+    }
+    // Assume equality for cyclic structures. The algorithm for detecting cyclic
+    // structures is adapted from ES 5.1 section 15.12.3, abstract operation `JO`.
+
+    // Initializing stack of traversed objects.
+    // It's done here since we only need them for objects and arrays comparison.
+    aStack = aStack || [];
+    bStack = bStack || [];
+    var length = aStack.length;
+    while (length--) {
+      // Linear search. Performance is inversely proportional to the number of
+      // unique nested structures.
+      if (aStack[length] === a) return bStack[length] === b;
+    }
+
+    // Add the first object to the stack of traversed objects.
+    aStack.push(a);
+    bStack.push(b);
+
+    // Recursively compare objects and arrays.
+    if (areArrays) {
+      // Compare array lengths to determine if a deep comparison is necessary.
+      length = a.length;
+      if (length !== b.length) return false;
+      // Deep compare the contents, ignoring non-numeric properties.
+      while (length--) {
+        if (!eq(a[length], b[length], aStack, bStack)) return false;
+      }
+    } else {
+      // Deep compare objects.
+      var _keys = keys(a), key;
+      length = _keys.length;
+      // Ensure that both objects contain the same number of properties before comparing deep equality.
+      if (keys(b).length !== length) return false;
+      while (length--) {
+        // Deep compare each member
+        key = _keys[length];
+        if (!(has$1(b, key) && eq(a[key], b[key], aStack, bStack))) return false;
+      }
+    }
+    // Remove the first object from the stack of traversed objects.
+    aStack.pop();
+    bStack.pop();
+    return true;
+  }
+
+  // Perform a deep comparison to check if two objects are equal.
+  function isEqual(a, b) {
+    return eq(a, b);
+  }
+
+  // Retrieve all the enumerable property names of an object.
+  function allKeys(obj) {
+    if (!isObject(obj)) return [];
+    var keys = [];
+    for (var key in obj) keys.push(key);
+    // Ahem, IE < 9.
+    if (hasEnumBug) collectNonEnumProps(obj, keys);
+    return keys;
+  }
+
+  // Since the regular `Object.prototype.toString` type tests don't work for
+  // some types in IE 11, we use a fingerprinting heuristic instead, based
+  // on the methods. It's not great, but it's the best we got.
+  // The fingerprint method lists are defined below.
+  function ie11fingerprint(methods) {
+    var length = getLength(methods);
+    return function(obj) {
+      if (obj == null) return false;
+      // `Map`, `WeakMap` and `Set` have no enumerable keys.
+      var keys = allKeys(obj);
+      if (getLength(keys)) return false;
+      for (var i = 0; i < length; i++) {
+        if (!isFunction$1(obj[methods[i]])) return false;
+      }
+      // If we are testing against `WeakMap`, we need to ensure that
+      // `obj` doesn't have a `forEach` method in order to distinguish
+      // it from a regular `Map`.
+      return methods !== weakMapMethods || !isFunction$1(obj[forEachName]);
+    };
+  }
+
+  // In the interest of compact minification, we write
+  // each string in the fingerprints only once.
+  var forEachName = 'forEach',
+      hasName = 'has',
+      commonInit = ['clear', 'delete'],
+      mapTail = ['get', hasName, 'set'];
+
+  // `Map`, `WeakMap` and `Set` each have slightly different
+  // combinations of the above sublists.
+  var mapMethods = commonInit.concat(forEachName, mapTail),
+      weakMapMethods = commonInit.concat(mapTail),
+      setMethods = ['add'].concat(commonInit, forEachName, hasName);
+
+  var isMap = isIE11 ? ie11fingerprint(mapMethods) : tagTester('Map');
+
+  var isWeakMap = isIE11 ? ie11fingerprint(weakMapMethods) : tagTester('WeakMap');
+
+  var isSet = isIE11 ? ie11fingerprint(setMethods) : tagTester('Set');
+
+  var isWeakSet = tagTester('WeakSet');
+
+  // Retrieve the values of an object's properties.
+  function values(obj) {
+    var _keys = keys(obj);
+    var length = _keys.length;
+    var values = Array(length);
+    for (var i = 0; i < length; i++) {
+      values[i] = obj[_keys[i]];
+    }
+    return values;
+  }
+
+  // Convert an object into a list of `[key, value]` pairs.
+  // The opposite of `_.object` with one argument.
+  function pairs(obj) {
+    var _keys = keys(obj);
+    var length = _keys.length;
+    var pairs = Array(length);
+    for (var i = 0; i < length; i++) {
+      pairs[i] = [_keys[i], obj[_keys[i]]];
+    }
+    return pairs;
+  }
+
+  // Invert the keys and values of an object. The values must be serializable.
+  function invert(obj) {
+    var result = {};
+    var _keys = keys(obj);
+    for (var i = 0, length = _keys.length; i < length; i++) {
+      result[obj[_keys[i]]] = _keys[i];
+    }
+    return result;
+  }
+
+  // Return a sorted list of the function names available on the object.
+  function functions(obj) {
+    var names = [];
+    for (var key in obj) {
+      if (isFunction$1(obj[key])) names.push(key);
+    }
+    return names.sort();
+  }
+
+  // An internal function for creating assigner functions.
+  function createAssigner(keysFunc, defaults) {
+    return function(obj) {
+      var length = arguments.length;
+      if (defaults) obj = Object(obj);
+      if (length < 2 || obj == null) return obj;
+      for (var index = 1; index < length; index++) {
+        var source = arguments[index],
+            keys = keysFunc(source),
+            l = keys.length;
+        for (var i = 0; i < l; i++) {
+          var key = keys[i];
+          if (!defaults || obj[key] === void 0) obj[key] = source[key];
+        }
+      }
+      return obj;
+    };
+  }
+
+  // Extend a given object with all the properties in passed-in object(s).
+  var extend = createAssigner(allKeys);
+
+  // Assigns a given object with all the own properties in the passed-in
+  // object(s).
+  // (https://developer.mozilla.org/docs/Web/JavaScript/Reference/Global_Objects/Object/assign)
+  var extendOwn = createAssigner(keys);
+
+  // Fill in a given object with default properties.
+  var defaults = createAssigner(allKeys, true);
+
+  // Create a naked function reference for surrogate-prototype-swapping.
+  function ctor() {
+    return function(){};
+  }
+
+  // An internal function for creating a new object that inherits from another.
+  function baseCreate(prototype) {
+    if (!isObject(prototype)) return {};
+    if (nativeCreate) return nativeCreate(prototype);
+    var Ctor = ctor();
+    Ctor.prototype = prototype;
+    var result = new Ctor;
+    Ctor.prototype = null;
+    return result;
+  }
+
+  // Creates an object that inherits from the given prototype object.
+  // If additional properties are provided then they will be added to the
+  // created object.
+  function create(prototype, props) {
+    var result = baseCreate(prototype);
+    if (props) extendOwn(result, props);
+    return result;
+  }
+
+  // Create a (shallow-cloned) duplicate of an object.
+  function clone(obj) {
+    if (!isObject(obj)) return obj;
+    return isArray(obj) ? obj.slice() : extend({}, obj);
+  }
+
+  // Invokes `interceptor` with the `obj` and then returns `obj`.
+  // The primary purpose of this method is to "tap into" a method chain, in
+  // order to perform operations on intermediate results within the chain.
+  function tap(obj, interceptor) {
+    interceptor(obj);
+    return obj;
+  }
+
+  // Normalize a (deep) property `path` to array.
+  // Like `_.iteratee`, this function can be customized.
+  function toPath$1(path) {
+    return isArray(path) ? path : [path];
+  }
+  _$1.toPath = toPath$1;
+
+  // Internal wrapper for `_.toPath` to enable minification.
+  // Similar to `cb` for `_.iteratee`.
+  function toPath(path) {
+    return _$1.toPath(path);
+  }
+
+  // Internal function to obtain a nested property in `obj` along `path`.
+  function deepGet(obj, path) {
+    var length = path.length;
+    for (var i = 0; i < length; i++) {
+      if (obj == null) return void 0;
+      obj = obj[path[i]];
+    }
+    return length ? obj : void 0;
+  }
+
+  // Get the value of the (deep) property on `path` from `object`.
+  // If any property in `path` does not exist or if the value is
+  // `undefined`, return `defaultValue` instead.
+  // The `path` is normalized through `_.toPath`.
+  function get(object, path, defaultValue) {
+    var value = deepGet(object, toPath(path));
+    return isUndefined(value) ? defaultValue : value;
+  }
+
+  // Shortcut function for checking if an object has a given property directly on
+  // itself (in other words, not on a prototype). Unlike the internal `has`
+  // function, this public version can also traverse nested properties.
+  function has(obj, path) {
+    path = toPath(path);
+    var length = path.length;
+    for (var i = 0; i < length; i++) {
+      var key = path[i];
+      if (!has$1(obj, key)) return false;
+      obj = obj[key];
+    }
+    return !!length;
+  }
+
+  // Keep the identity function around for default iteratees.
+  function identity(value) {
+    return value;
+  }
+
+  // Returns a predicate for checking whether an object has a given set of
+  // `key:value` pairs.
+  function matcher(attrs) {
+    attrs = extendOwn({}, attrs);
+    return function(obj) {
+      return isMatch(obj, attrs);
+    };
+  }
+
+  // Creates a function that, when passed an object, will traverse that object’s
+  // properties down the given `path`, specified as an array of keys or indices.
+  function property(path) {
+    path = toPath(path);
+    return function(obj) {
+      return deepGet(obj, path);
+    };
+  }
+
+  // Internal function that returns an efficient (for current engines) version
+  // of the passed-in callback, to be repeatedly applied in other Underscore
+  // functions.
+  function optimizeCb(func, context, argCount) {
+    if (context === void 0) return func;
+    switch (argCount == null ? 3 : argCount) {
+      case 1: return function(value) {
+        return func.call(context, value);
+      };
+      // The 2-argument case is omitted because we’re not using it.
+      case 3: return function(value, index, collection) {
+        return func.call(context, value, index, collection);
+      };
+      case 4: return function(accumulator, value, index, collection) {
+        return func.call(context, accumulator, value, index, collection);
+      };
+    }
+    return function() {
+      return func.apply(context, arguments);
+    };
+  }
+
+  // An internal function to generate callbacks that can be applied to each
+  // element in a collection, returning the desired result — either `_.identity`,
+  // an arbitrary callback, a property matcher, or a property accessor.
+  function baseIteratee(value, context, argCount) {
+    if (value == null) return identity;
+    if (isFunction$1(value)) return optimizeCb(value, context, argCount);
+    if (isObject(value) && !isArray(value)) return matcher(value);
+    return property(value);
+  }
+
+  // External wrapper for our callback generator. Users may customize
+  // `_.iteratee` if they want additional predicate/iteratee shorthand styles.
+  // This abstraction hides the internal-only `argCount` argument.
+  function iteratee(value, context) {
+    return baseIteratee(value, context, Infinity);
+  }
+  _$1.iteratee = iteratee;
+
+  // The function we call internally to generate a callback. It invokes
+  // `_.iteratee` if overridden, otherwise `baseIteratee`.
+  function cb(value, context, argCount) {
+    if (_$1.iteratee !== iteratee) return _$1.iteratee(value, context);
+    return baseIteratee(value, context, argCount);
+  }
+
+  // Returns the results of applying the `iteratee` to each element of `obj`.
+  // In contrast to `_.map` it returns an object.
+  function mapObject(obj, iteratee, context) {
+    iteratee = cb(iteratee, context);
+    var _keys = keys(obj),
+        length = _keys.length,
+        results = {};
+    for (var index = 0; index < length; index++) {
+      var currentKey = _keys[index];
+      results[currentKey] = iteratee(obj[currentKey], currentKey, obj);
+    }
+    return results;
+  }
+
+  // Predicate-generating function. Often useful outside of Underscore.
+  function noop(){}
+
+  // Generates a function for a given object that returns a given property.
+  function propertyOf(obj) {
+    if (obj == null) return noop;
+    return function(path) {
+      return get(obj, path);
+    };
+  }
+
+  // Run a function **n** times.
+  function times(n, iteratee, context) {
+    var accum = Array(Math.max(0, n));
+    iteratee = optimizeCb(iteratee, context, 1);
+    for (var i = 0; i < n; i++) accum[i] = iteratee(i);
+    return accum;
+  }
+
+  // Return a random integer between `min` and `max` (inclusive).
+  function random(min, max) {
+    if (max == null) {
+      max = min;
+      min = 0;
+    }
+    return min + Math.floor(Math.random() * (max - min + 1));
+  }
+
+  // A (possibly faster) way to get the current timestamp as an integer.
+  var now = Date.now || function() {
+    return new Date().getTime();
+  };
+
+  // Internal helper to generate functions for escaping and unescaping strings
+  // to/from HTML interpolation.
+  function createEscaper(map) {
+    var escaper = function(match) {
+      return map[match];
+    };
+    // Regexes for identifying a key that needs to be escaped.
+    var source = '(?:' + keys(map).join('|') + ')';
+    var testRegexp = RegExp(source);
+    var replaceRegexp = RegExp(source, 'g');
+    return function(string) {
+      string = string == null ? '' : '' + string;
+      return testRegexp.test(string) ? string.replace(replaceRegexp, escaper) : string;
+    };
+  }
+
+  // Internal list of HTML entities for escaping.
+  var escapeMap = {
+    '&': '&amp;',
+    '<': '&lt;',
+    '>': '&gt;',
+    '"': '&quot;',
+    "'": '&#x27;',
+    '`': '&#x60;'
+  };
+
+  // Function for escaping strings to HTML interpolation.
+  var _escape = createEscaper(escapeMap);
+
+  // Internal list of HTML entities for unescaping.
+  var unescapeMap = invert(escapeMap);
+
+  // Function for unescaping strings from HTML interpolation.
+  var _unescape = createEscaper(unescapeMap);
+
+  // By default, Underscore uses ERB-style template delimiters. Change the
+  // following template settings to use alternative delimiters.
+  var templateSettings = _$1.templateSettings = {
+    evaluate: /<%([\s\S]+?)%>/g,
+    interpolate: /<%=([\s\S]+?)%>/g,
+    escape: /<%-([\s\S]+?)%>/g
+  };
+
+  // When customizing `_.templateSettings`, if you don't want to define an
+  // interpolation, evaluation or escaping regex, we need one that is
+  // guaranteed not to match.
+  var noMatch = /(.)^/;
+
+  // Certain characters need to be escaped so that they can be put into a
+  // string literal.
+  var escapes = {
+    "'": "'",
+    '\\': '\\',
+    '\r': 'r',
+    '\n': 'n',
+    '\u2028': 'u2028',
+    '\u2029': 'u2029'
+  };
+
+  var escapeRegExp = /\\|'|\r|\n|\u2028|\u2029/g;
+
+  function escapeChar(match) {
+    return '\\' + escapes[match];
+  }
+
+  // In order to prevent third-party code injection through
+  // `_.templateSettings.variable`, we test it against the following regular
+  // expression. It is intentionally a bit more liberal than just matching valid
+  // identifiers, but still prevents possible loopholes through defaults or
+  // destructuring assignment.
+  var bareIdentifier = /^\s*(\w|\$)+\s*$/;
+
+  // JavaScript micro-templating, similar to John Resig's implementation.
+  // Underscore templating handles arbitrary delimiters, preserves whitespace,
+  // and correctly escapes quotes within interpolated code.
+  // NB: `oldSettings` only exists for backwards compatibility.
+  function template(text, settings, oldSettings) {
+    if (!settings && oldSettings) settings = oldSettings;
+    settings = defaults({}, settings, _$1.templateSettings);
+
+    // Combine delimiters into one regular expression via alternation.
+    var matcher = RegExp([
+      (settings.escape || noMatch).source,
+      (settings.interpolate || noMatch).source,
+      (settings.evaluate || noMatch).source
+    ].join('|') + '|$', 'g');
+
+    // Compile the template source, escaping string literals appropriately.
+    var index = 0;
+    var source = "__p+='";
+    text.replace(matcher, function(match, escape, interpolate, evaluate, offset) {
+      source += text.slice(index, offset).replace(escapeRegExp, escapeChar);
+      index = offset + match.length;
+
+      if (escape) {
+        source += "'+\n((__t=(" + escape + "))==null?'':_.escape(__t))+\n'";
+      } else if (interpolate) {
+        source += "'+\n((__t=(" + interpolate + "))==null?'':__t)+\n'";
+      } else if (evaluate) {
+        source += "';\n" + evaluate + "\n__p+='";
+      }
+
+      // Adobe VMs need the match returned to produce the correct offset.
+      return match;
+    });
+    source += "';\n";
+
+    var argument = settings.variable;
+    if (argument) {
+      // Insure against third-party code injection. (CVE-2021-23358)
+      if (!bareIdentifier.test(argument)) throw new Error(
+        'variable is not a bare identifier: ' + argument
+      );
+    } else {
+      // If a variable is not specified, place data values in local scope.
+      source = 'with(obj||{}){\n' + source + '}\n';
+      argument = 'obj';
+    }
+
+    source = "var __t,__p='',__j=Array.prototype.join," +
+      "print=function(){__p+=__j.call(arguments,'');};\n" +
+      source + 'return __p;\n';
+
+    var render;
+    try {
+      render = new Function(argument, '_', source);
+    } catch (e) {
+      e.source = source;
+      throw e;
+    }
+
+    var template = function(data) {
+      return render.call(this, data, _$1);
+    };
+
+    // Provide the compiled source as a convenience for precompilation.
+    template.source = 'function(' + argument + '){\n' + source + '}';
+
+    return template;
+  }
+
+  // Traverses the children of `obj` along `path`. If a child is a function, it
+  // is invoked with its parent as context. Returns the value of the final
+  // child, or `fallback` if any child is undefined.
+  function result(obj, path, fallback) {
+    path = toPath(path);
+    var length = path.length;
+    if (!length) {
+      return isFunction$1(fallback) ? fallback.call(obj) : fallback;
+    }
+    for (var i = 0; i < length; i++) {
+      var prop = obj == null ? void 0 : obj[path[i]];
+      if (prop === void 0) {
+        prop = fallback;
+        i = length; // Ensure we don't continue iterating.
+      }
+      obj = isFunction$1(prop) ? prop.call(obj) : prop;
+    }
+    return obj;
+  }
+
+  // Generate a unique integer id (unique within the entire client session).
+  // Useful for temporary DOM ids.
+  var idCounter = 0;
+  function uniqueId(prefix) {
+    var id = ++idCounter + '';
+    return prefix ? prefix + id : id;
+  }
+
+  // Start chaining a wrapped Underscore object.
+  function chain(obj) {
+    var instance = _$1(obj);
+    instance._chain = true;
+    return instance;
+  }
+
+  // Internal function to execute `sourceFunc` bound to `context` with optional
+  // `args`. Determines whether to execute a function as a constructor or as a
+  // normal function.
+  function executeBound(sourceFunc, boundFunc, context, callingContext, args) {
+    if (!(callingContext instanceof boundFunc)) return sourceFunc.apply(context, args);
+    var self = baseCreate(sourceFunc.prototype);
+    var result = sourceFunc.apply(self, args);
+    if (isObject(result)) return result;
+    return self;
+  }
+
+  // Partially apply a function by creating a version that has had some of its
+  // arguments pre-filled, without changing its dynamic `this` context. `_` acts
+  // as a placeholder by default, allowing any combination of arguments to be
+  // pre-filled. Set `_.partial.placeholder` for a custom placeholder argument.
+  var partial = restArguments(function(func, boundArgs) {
+    var placeholder = partial.placeholder;
+    var bound = function() {
+      var position = 0, length = boundArgs.length;
+      var args = Array(length);
+      for (var i = 0; i < length; i++) {
+        args[i] = boundArgs[i] === placeholder ? arguments[position++] : boundArgs[i];
+      }
+      while (position < arguments.length) args.push(arguments[position++]);
+      return executeBound(func, bound, this, this, args);
+    };
+    return bound;
+  });
+
+  partial.placeholder = _$1;
+
+  // Create a function bound to a given object (assigning `this`, and arguments,
+  // optionally).
+  var bind = restArguments(function(func, context, args) {
+    if (!isFunction$1(func)) throw new TypeError('Bind must be called on a function');
+    var bound = restArguments(function(callArgs) {
+      return executeBound(func, bound, context, this, args.concat(callArgs));
+    });
+    return bound;
+  });
+
+  // Internal helper for collection methods to determine whether a collection
+  // should be iterated as an array or as an object.
+  // Related: https://people.mozilla.org/~jorendorff/es6-draft.html#sec-tolength
+  // Avoids a very nasty iOS 8 JIT bug on ARM-64. #2094
+  var isArrayLike = createSizePropertyCheck(getLength);
+
+  // Internal implementation of a recursive `flatten` function.
+  function flatten$1(input, depth, strict, output) {
+    output = output || [];
+    if (!depth && depth !== 0) {
+      depth = Infinity;
+    } else if (depth <= 0) {
+      return output.concat(input);
+    }
+    var idx = output.length;
+    for (var i = 0, length = getLength(input); i < length; i++) {
+      var value = input[i];
+      if (isArrayLike(value) && (isArray(value) || isArguments$1(value))) {
+        // Flatten current level of array or arguments object.
+        if (depth > 1) {
+          flatten$1(value, depth - 1, strict, output);
+          idx = output.length;
+        } else {
+          var j = 0, len = value.length;
+          while (j < len) output[idx++] = value[j++];
+        }
+      } else if (!strict) {
+        output[idx++] = value;
+      }
+    }
+    return output;
+  }
+
+  // Bind a number of an object's methods to that object. Remaining arguments
+  // are the method names to be bound. Useful for ensuring that all callbacks
+  // defined on an object belong to it.
+  var bindAll = restArguments(function(obj, keys) {
+    keys = flatten$1(keys, false, false);
+    var index = keys.length;
+    if (index < 1) throw new Error('bindAll must be passed function names');
+    while (index--) {
+      var key = keys[index];
+      obj[key] = bind(obj[key], obj);
+    }
+    return obj;
+  });
+
+  // Memoize an expensive function by storing its results.
+  function memoize(func, hasher) {
+    var memoize = function(key) {
+      var cache = memoize.cache;
+      var address = '' + (hasher ? hasher.apply(this, arguments) : key);
+      if (!has$1(cache, address)) cache[address] = func.apply(this, arguments);
+      return cache[address];
+    };
+    memoize.cache = {};
+    return memoize;
+  }
+
+  // Delays a function for the given number of milliseconds, and then calls
+  // it with the arguments supplied.
+  var delay = restArguments(function(func, wait, args) {
+    return setTimeout(function() {
+      return func.apply(null, args);
+    }, wait);
+  });
+
+  // Defers a function, scheduling it to run after the current call stack has
+  // cleared.
+  var defer = partial(delay, _$1, 1);
+
+  // Returns a function, that, when invoked, will only be triggered at most once
+  // during a given window of time. Normally, the throttled function will run
+  // as much as it can, without ever going more than once per `wait` duration;
+  // but if you'd like to disable the execution on the leading edge, pass
+  // `{leading: false}`. To disable execution on the trailing edge, ditto.
+  function throttle(func, wait, options) {
+    var timeout, context, args, result;
+    var previous = 0;
+    if (!options) options = {};
+
+    var later = function() {
+      previous = options.leading === false ? 0 : now();
+      timeout = null;
+      result = func.apply(context, args);
+      if (!timeout) context = args = null;
+    };
+
+    var throttled = function() {
+      var _now = now();
+      if (!previous && options.leading === false) previous = _now;
+      var remaining = wait - (_now - previous);
+      context = this;
+      args = arguments;
+      if (remaining <= 0 || remaining > wait) {
+        if (timeout) {
+          clearTimeout(timeout);
+          timeout = null;
+        }
+        previous = _now;
+        result = func.apply(context, args);
+        if (!timeout) context = args = null;
+      } else if (!timeout && options.trailing !== false) {
+        timeout = setTimeout(later, remaining);
+      }
+      return result;
+    };
+
+    throttled.cancel = function() {
+      clearTimeout(timeout);
+      previous = 0;
+      timeout = context = args = null;
+    };
+
+    return throttled;
+  }
+
+  // When a sequence of calls of the returned function ends, the argument
+  // function is triggered. The end of a sequence is defined by the `wait`
+  // parameter. If `immediate` is passed, the argument function will be
+  // triggered at the beginning of the sequence instead of at the end.
+  function debounce(func, wait, immediate) {
+    var timeout, previous, args, result, context;
+
+    var later = function() {
+      var passed = now() - previous;
+      if (wait > passed) {
+        timeout = setTimeout(later, wait - passed);
+      } else {
+        timeout = null;
+        if (!immediate) result = func.apply(context, args);
+        // This check is needed because `func` can recursively invoke `debounced`.
+        if (!timeout) args = context = null;
+      }
+    };
+
+    var debounced = restArguments(function(_args) {
+      context = this;
+      args = _args;
+      previous = now();
+      if (!timeout) {
+        timeout = setTimeout(later, wait);
+        if (immediate) result = func.apply(context, args);
+      }
+      return result;
+    });
+
+    debounced.cancel = function() {
+      clearTimeout(timeout);
+      timeout = args = context = null;
+    };
+
+    return debounced;
+  }
+
+  // Returns the first function passed as an argument to the second,
+  // allowing you to adjust arguments, run code before and after, and
+  // conditionally execute the original function.
+  function wrap(func, wrapper) {
+    return partial(wrapper, func);
+  }
+
+  // Returns a negated version of the passed-in predicate.
+  function negate(predicate) {
+    return function() {
+      return !predicate.apply(this, arguments);
+    };
+  }
+
+  // Returns a function that is the composition of a list of functions, each
+  // consuming the return value of the function that follows.
+  function compose() {
+    var args = arguments;
+    var start = args.length - 1;
+    return function() {
+      var i = start;
+      var result = args[start].apply(this, arguments);
+      while (i--) result = args[i].call(this, result);
+      return result;
+    };
+  }
+
+  // Returns a function that will only be executed on and after the Nth call.
+  function after(times, func) {
+    return function() {
+      if (--times < 1) {
+        return func.apply(this, arguments);
+      }
+    };
+  }
+
+  // Returns a function that will only be executed up to (but not including) the
+  // Nth call.
+  function before(times, func) {
+    var memo;
+    return function() {
+      if (--times > 0) {
+        memo = func.apply(this, arguments);
+      }
+      if (times <= 1) func = null;
+      return memo;
+    };
+  }
+
+  // Returns a function that will be executed at most one time, no matter how
+  // often you call it. Useful for lazy initialization.
+  var once = partial(before, 2);
+
+  // Returns the first key on an object that passes a truth test.
+  function findKey(obj, predicate, context) {
+    predicate = cb(predicate, context);
+    var _keys = keys(obj), key;
+    for (var i = 0, length = _keys.length; i < length; i++) {
+      key = _keys[i];
+      if (predicate(obj[key], key, obj)) return key;
+    }
+  }
+
+  // Internal function to generate `_.findIndex` and `_.findLastIndex`.
+  function createPredicateIndexFinder(dir) {
+    return function(array, predicate, context) {
+      predicate = cb(predicate, context);
+      var length = getLength(array);
+      var index = dir > 0 ? 0 : length - 1;
+      for (; index >= 0 && index < length; index += dir) {
+        if (predicate(array[index], index, array)) return index;
+      }
+      return -1;
+    };
+  }
+
+  // Returns the first index on an array-like that passes a truth test.
+  var findIndex = createPredicateIndexFinder(1);
+
+  // Returns the last index on an array-like that passes a truth test.
+  var findLastIndex = createPredicateIndexFinder(-1);
+
+  // Use a comparator function to figure out the smallest index at which
+  // an object should be inserted so as to maintain order. Uses binary search.
+  function sortedIndex(array, obj, iteratee, context) {
+    iteratee = cb(iteratee, context, 1);
+    var value = iteratee(obj);
+    var low = 0, high = getLength(array);
+    while (low < high) {
+      var mid = Math.floor((low + high) / 2);
+      if (iteratee(array[mid]) < value) low = mid + 1; else high = mid;
+    }
+    return low;
+  }
+
+  // Internal function to generate the `_.indexOf` and `_.lastIndexOf` functions.
+  function createIndexFinder(dir, predicateFind, sortedIndex) {
+    return function(array, item, idx) {
+      var i = 0, length = getLength(array);
+      if (typeof idx == 'number') {
+        if (dir > 0) {
+          i = idx >= 0 ? idx : Math.max(idx + length, i);
+        } else {
+          length = idx >= 0 ? Math.min(idx + 1, length) : idx + length + 1;
+        }
+      } else if (sortedIndex && idx && length) {
+        idx = sortedIndex(array, item);
+        return array[idx] === item ? idx : -1;
+      }
+      if (item !== item) {
+        idx = predicateFind(slice.call(array, i, length), isNaN$1);
+        return idx >= 0 ? idx + i : -1;
+      }
+      for (idx = dir > 0 ? i : length - 1; idx >= 0 && idx < length; idx += dir) {
+        if (array[idx] === item) return idx;
+      }
+      return -1;
+    };
+  }
+
+  // Return the position of the first occurrence of an item in an array,
+  // or -1 if the item is not included in the array.
+  // If the array is large and already in sort order, pass `true`
+  // for **isSorted** to use binary search.
+  var indexOf = createIndexFinder(1, findIndex, sortedIndex);
+
+  // Return the position of the last occurrence of an item in an array,
+  // or -1 if the item is not included in the array.
+  var lastIndexOf = createIndexFinder(-1, findLastIndex);
+
+  // Return the first value which passes a truth test.
+  function find(obj, predicate, context) {
+    var keyFinder = isArrayLike(obj) ? findIndex : findKey;
+    var key = keyFinder(obj, predicate, context);
+    if (key !== void 0 && key !== -1) return obj[key];
+  }
+
+  // Convenience version of a common use case of `_.find`: getting the first
+  // object containing specific `key:value` pairs.
+  function findWhere(obj, attrs) {
+    return find(obj, matcher(attrs));
+  }
+
+  // The cornerstone for collection functions, an `each`
+  // implementation, aka `forEach`.
+  // Handles raw objects in addition to array-likes. Treats all
+  // sparse array-likes as if they were dense.
+  function each(obj, iteratee, context) {
+    iteratee = optimizeCb(iteratee, context);
+    var i, length;
+    if (isArrayLike(obj)) {
+      for (i = 0, length = obj.length; i < length; i++) {
+        iteratee(obj[i], i, obj);
+      }
+    } else {
+      var _keys = keys(obj);
+      for (i = 0, length = _keys.length; i < length; i++) {
+        iteratee(obj[_keys[i]], _keys[i], obj);
+      }
+    }
+    return obj;
+  }
+
+  // Return the results of applying the iteratee to each element.
+  function map(obj, iteratee, context) {
+    iteratee = cb(iteratee, context);
+    var _keys = !isArrayLike(obj) && keys(obj),
+        length = (_keys || obj).length,
+        results = Array(length);
+    for (var index = 0; index < length; index++) {
+      var currentKey = _keys ? _keys[index] : index;
+      results[index] = iteratee(obj[currentKey], currentKey, obj);
+    }
+    return results;
+  }
+
+  // Internal helper to create a reducing function, iterating left or right.
+  function createReduce(dir) {
+    // Wrap code that reassigns argument variables in a separate function than
+    // the one that accesses `arguments.length` to avoid a perf hit. (#1991)
+    var reducer = function(obj, iteratee, memo, initial) {
+      var _keys = !isArrayLike(obj) && keys(obj),
+          length = (_keys || obj).length,
+          index = dir > 0 ? 0 : length - 1;
+      if (!initial) {
+        memo = obj[_keys ? _keys[index] : index];
+        index += dir;
+      }
+      for (; index >= 0 && index < length; index += dir) {
+        var currentKey = _keys ? _keys[index] : index;
+        memo = iteratee(memo, obj[currentKey], currentKey, obj);
+      }
+      return memo;
+    };
+
+    return function(obj, iteratee, memo, context) {
+      var initial = arguments.length >= 3;
+      return reducer(obj, optimizeCb(iteratee, context, 4), memo, initial);
+    };
+  }
+
+  // **Reduce** builds up a single result from a list of values, aka `inject`,
+  // or `foldl`.
+  var reduce = createReduce(1);
+
+  // The right-associative version of reduce, also known as `foldr`.
+  var reduceRight = createReduce(-1);
+
+  // Return all the elements that pass a truth test.
+  function filter(obj, predicate, context) {
+    var results = [];
+    predicate = cb(predicate, context);
+    each(obj, function(value, index, list) {
+      if (predicate(value, index, list)) results.push(value);
+    });
+    return results;
+  }
+
+  // Return all the elements for which a truth test fails.
+  function reject(obj, predicate, context) {
+    return filter(obj, negate(cb(predicate)), context);
+  }
+
+  // Determine whether all of the elements pass a truth test.
+  function every(obj, predicate, context) {
+    predicate = cb(predicate, context);
+    var _keys = !isArrayLike(obj) && keys(obj),
+        length = (_keys || obj).length;
+    for (var index = 0; index < length; index++) {
+      var currentKey = _keys ? _keys[index] : index;
+      if (!predicate(obj[currentKey], currentKey, obj)) return false;
+    }
+    return true;
+  }
+
+  // Determine if at least one element in the object passes a truth test.
+  function some(obj, predicate, context) {
+    predicate = cb(predicate, context);
+    var _keys = !isArrayLike(obj) && keys(obj),
+        length = (_keys || obj).length;
+    for (var index = 0; index < length; index++) {
+      var currentKey = _keys ? _keys[index] : index;
+      if (predicate(obj[currentKey], currentKey, obj)) return true;
+    }
+    return false;
+  }
+
+  // Determine if the array or object contains a given item (using `===`).
+  function contains(obj, item, fromIndex, guard) {
+    if (!isArrayLike(obj)) obj = values(obj);
+    if (typeof fromIndex != 'number' || guard) fromIndex = 0;
+    return indexOf(obj, item, fromIndex) >= 0;
+  }
+
+  // Invoke a method (with arguments) on every item in a collection.
+  var invoke = restArguments(function(obj, path, args) {
+    var contextPath, func;
+    if (isFunction$1(path)) {
+      func = path;
+    } else {
+      path = toPath(path);
+      contextPath = path.slice(0, -1);
+      path = path[path.length - 1];
+    }
+    return map(obj, function(context) {
+      var method = func;
+      if (!method) {
+        if (contextPath && contextPath.length) {
+          context = deepGet(context, contextPath);
+        }
+        if (context == null) return void 0;
+        method = context[path];
+      }
+      return method == null ? method : method.apply(context, args);
+    });
+  });
+
+  // Convenience version of a common use case of `_.map`: fetching a property.
+  function pluck(obj, key) {
+    return map(obj, property(key));
+  }
+
+  // Convenience version of a common use case of `_.filter`: selecting only
+  // objects containing specific `key:value` pairs.
+  function where(obj, attrs) {
+    return filter(obj, matcher(attrs));
+  }
+
+  // Return the maximum element (or element-based computation).
+  function max(obj, iteratee, context) {
+    var result = -Infinity, lastComputed = -Infinity,
+        value, computed;
+    if (iteratee == null || typeof iteratee == 'number' && typeof obj[0] != 'object' && obj != null) {
+      obj = isArrayLike(obj) ? obj : values(obj);
+      for (var i = 0, length = obj.length; i < length; i++) {
+        value = obj[i];
+        if (value != null && value > result) {
+          result = value;
+        }
+      }
+    } else {
+      iteratee = cb(iteratee, context);
+      each(obj, function(v, index, list) {
+        computed = iteratee(v, index, list);
+        if (computed > lastComputed || computed === -Infinity && result === -Infinity) {
+          result = v;
+          lastComputed = computed;
+        }
+      });
+    }
+    return result;
+  }
+
+  // Return the minimum element (or element-based computation).
+  function min(obj, iteratee, context) {
+    var result = Infinity, lastComputed = Infinity,
+        value, computed;
+    if (iteratee == null || typeof iteratee == 'number' && typeof obj[0] != 'object' && obj != null) {
+      obj = isArrayLike(obj) ? obj : values(obj);
+      for (var i = 0, length = obj.length; i < length; i++) {
+        value = obj[i];
+        if (value != null && value < result) {
+          result = value;
+        }
+      }
+    } else {
+      iteratee = cb(iteratee, context);
+      each(obj, function(v, index, list) {
+        computed = iteratee(v, index, list);
+        if (computed < lastComputed || computed === Infinity && result === Infinity) {
+          result = v;
+          lastComputed = computed;
+        }
+      });
+    }
+    return result;
+  }
+
+  // Sample **n** random values from a collection using the modern version of the
+  // [Fisher-Yates shuffle](https://en.wikipedia.org/wiki/Fisher–Yates_shuffle).
+  // If **n** is not specified, returns a single random element.
+  // The internal `guard` argument allows it to work with `_.map`.
+  function sample(obj, n, guard) {
+    if (n == null || guard) {
+      if (!isArrayLike(obj)) obj = values(obj);
+      return obj[random(obj.length - 1)];
+    }
+    var sample = isArrayLike(obj) ? clone(obj) : values(obj);
+    var length = getLength(sample);
+    n = Math.max(Math.min(n, length), 0);
+    var last = length - 1;
+    for (var index = 0; index < n; index++) {
+      var rand = random(index, last);
+      var temp = sample[index];
+      sample[index] = sample[rand];
+      sample[rand] = temp;
+    }
+    return sample.slice(0, n);
+  }
+
+  // Shuffle a collection.
+  function shuffle(obj) {
+    return sample(obj, Infinity);
+  }
+
+  // Sort the object's values by a criterion produced by an iteratee.
+  function sortBy(obj, iteratee, context) {
+    var index = 0;
+    iteratee = cb(iteratee, context);
+    return pluck(map(obj, function(value, key, list) {
+      return {
+        value: value,
+        index: index++,
+        criteria: iteratee(value, key, list)
+      };
+    }).sort(function(left, right) {
+      var a = left.criteria;
+      var b = right.criteria;
+      if (a !== b) {
+        if (a > b || a === void 0) return 1;
+        if (a < b || b === void 0) return -1;
+      }
+      return left.index - right.index;
+    }), 'value');
+  }
+
+  // An internal function used for aggregate "group by" operations.
+  function group(behavior, partition) {
+    return function(obj, iteratee, context) {
+      var result = partition ? [[], []] : {};
+      iteratee = cb(iteratee, context);
+      each(obj, function(value, index) {
+        var key = iteratee(value, index, obj);
+        behavior(result, value, key);
+      });
+      return result;
+    };
+  }
+
+  // Groups the object's values by a criterion. Pass either a string attribute
+  // to group by, or a function that returns the criterion.
+  var groupBy = group(function(result, value, key) {
+    if (has$1(result, key)) result[key].push(value); else result[key] = [value];
+  });
+
+  // Indexes the object's values by a criterion, similar to `_.groupBy`, but for
+  // when you know that your index values will be unique.
+  var indexBy = group(function(result, value, key) {
+    result[key] = value;
+  });
+
+  // Counts instances of an object that group by a certain criterion. Pass
+  // either a string attribute to count by, or a function that returns the
+  // criterion.
+  var countBy = group(function(result, value, key) {
+    if (has$1(result, key)) result[key]++; else result[key] = 1;
+  });
+
+  // Split a collection into two arrays: one whose elements all pass the given
+  // truth test, and one whose elements all do not pass the truth test.
+  var partition = group(function(result, value, pass) {
+    result[pass ? 0 : 1].push(value);
+  }, true);
+
+  // Safely create a real, live array from anything iterable.
+  var reStrSymbol = /[^\ud800-\udfff]|[\ud800-\udbff][\udc00-\udfff]|[\ud800-\udfff]/g;
+  function toArray(obj) {
+    if (!obj) return [];
+    if (isArray(obj)) return slice.call(obj);
+    if (isString(obj)) {
+      // Keep surrogate pair characters together.
+      return obj.match(reStrSymbol);
+    }
+    if (isArrayLike(obj)) return map(obj, identity);
+    return values(obj);
+  }
+
+  // Return the number of elements in a collection.
+  function size(obj) {
+    if (obj == null) return 0;
+    return isArrayLike(obj) ? obj.length : keys(obj).length;
+  }
+
+  // Internal `_.pick` helper function to determine whether `key` is an enumerable
+  // property name of `obj`.
+  function keyInObj(value, key, obj) {
+    return key in obj;
+  }
+
+  // Return a copy of the object only containing the allowed properties.
+  var pick = restArguments(function(obj, keys) {
+    var result = {}, iteratee = keys[0];
+    if (obj == null) return result;
+    if (isFunction$1(iteratee)) {
+      if (keys.length > 1) iteratee = optimizeCb(iteratee, keys[1]);
+      keys = allKeys(obj);
+    } else {
+      iteratee = keyInObj;
+      keys = flatten$1(keys, false, false);
+      obj = Object(obj);
+    }
+    for (var i = 0, length = keys.length; i < length; i++) {
+      var key = keys[i];
+      var value = obj[key];
+      if (iteratee(value, key, obj)) result[key] = value;
+    }
+    return result;
+  });
+
+  // Return a copy of the object without the disallowed properties.
+  var omit = restArguments(function(obj, keys) {
+    var iteratee = keys[0], context;
+    if (isFunction$1(iteratee)) {
+      iteratee = negate(iteratee);
+      if (keys.length > 1) context = keys[1];
+    } else {
+      keys = map(flatten$1(keys, false, false), String);
+      iteratee = function(value, key) {
+        return !contains(keys, key);
+      };
+    }
+    return pick(obj, iteratee, context);
+  });
+
+  // Returns everything but the last entry of the array. Especially useful on
+  // the arguments object. Passing **n** will return all the values in
+  // the array, excluding the last N.
+  function initial(array, n, guard) {
+    return slice.call(array, 0, Math.max(0, array.length - (n == null || guard ? 1 : n)));
+  }
+
+  // Get the first element of an array. Passing **n** will return the first N
+  // values in the array. The **guard** check allows it to work with `_.map`.
+  function first(array, n, guard) {
+    if (array == null || array.length < 1) return n == null || guard ? void 0 : [];
+    if (n == null || guard) return array[0];
+    return initial(array, array.length - n);
+  }
+
+  // Returns everything but the first entry of the `array`. Especially useful on
+  // the `arguments` object. Passing an **n** will return the rest N values in the
+  // `array`.
+  function rest(array, n, guard) {
+    return slice.call(array, n == null || guard ? 1 : n);
+  }
+
+  // Get the last element of an array. Passing **n** will return the last N
+  // values in the array.
+  function last(array, n, guard) {
+    if (array == null || array.length < 1) return n == null || guard ? void 0 : [];
+    if (n == null || guard) return array[array.length - 1];
+    return rest(array, Math.max(0, array.length - n));
+  }
+
+  // Trim out all falsy values from an array.
+  function compact(array) {
+    return filter(array, Boolean);
+  }
+
+  // Flatten out an array, either recursively (by default), or up to `depth`.
+  // Passing `true` or `false` as `depth` means `1` or `Infinity`, respectively.
+  function flatten(array, depth) {
+    return flatten$1(array, depth, false);
+  }
+
+  // Take the difference between one array and a number of other arrays.
+  // Only the elements present in just the first array will remain.
+  var difference = restArguments(function(array, rest) {
+    rest = flatten$1(rest, true, true);
+    return filter(array, function(value){
+      return !contains(rest, value);
+    });
+  });
+
+  // Return a version of the array that does not contain the specified value(s).
+  var without = restArguments(function(array, otherArrays) {
+    return difference(array, otherArrays);
+  });
+
+  // Produce a duplicate-free version of the array. If the array has already
+  // been sorted, you have the option of using a faster algorithm.
+  // The faster algorithm will not work with an iteratee if the iteratee
+  // is not a one-to-one function, so providing an iteratee will disable
+  // the faster algorithm.
+  function uniq(array, isSorted, iteratee, context) {
+    if (!isBoolean(isSorted)) {
+      context = iteratee;
+      iteratee = isSorted;
+      isSorted = false;
+    }
+    if (iteratee != null) iteratee = cb(iteratee, context);
+    var result = [];
+    var seen = [];
+    for (var i = 0, length = getLength(array); i < length; i++) {
+      var value = array[i],
+          computed = iteratee ? iteratee(value, i, array) : value;
+      if (isSorted && !iteratee) {
+        if (!i || seen !== computed) result.push(value);
+        seen = computed;
+      } else if (iteratee) {
+        if (!contains(seen, computed)) {
+          seen.push(computed);
+          result.push(value);
+        }
+      } else if (!contains(result, value)) {
+        result.push(value);
+      }
+    }
+    return result;
+  }
+
+  // Produce an array that contains the union: each distinct element from all of
+  // the passed-in arrays.
+  var union = restArguments(function(arrays) {
+    return uniq(flatten$1(arrays, true, true));
+  });
+
+  // Produce an array that contains every item shared between all the
+  // passed-in arrays.
+  function intersection(array) {
+    var result = [];
+    var argsLength = arguments.length;
+    for (var i = 0, length = getLength(array); i < length; i++) {
+      var item = array[i];
+      if (contains(result, item)) continue;
+      var j;
+      for (j = 1; j < argsLength; j++) {
+        if (!contains(arguments[j], item)) break;
+      }
+      if (j === argsLength) result.push(item);
+    }
+    return result;
+  }
+
+  // Complement of zip. Unzip accepts an array of arrays and groups
+  // each array's elements on shared indices.
+  function unzip(array) {
+    var length = array && max(array, getLength).length || 0;
+    var result = Array(length);
+
+    for (var index = 0; index < length; index++) {
+      result[index] = pluck(array, index);
+    }
+    return result;
+  }
+
+  // Zip together multiple lists into a single array -- elements that share
+  // an index go together.
+  var zip = restArguments(unzip);
+
+  // Converts lists into objects. Pass either a single array of `[key, value]`
+  // pairs, or two parallel arrays of the same length -- one of keys, and one of
+  // the corresponding values. Passing by pairs is the reverse of `_.pairs`.
+  function object(list, values) {
+    var result = {};
+    for (var i = 0, length = getLength(list); i < length; i++) {
+      if (values) {
+        result[list[i]] = values[i];
+      } else {
+        result[list[i][0]] = list[i][1];
+      }
+    }
+    return result;
+  }
+
+  // Generate an integer Array containing an arithmetic progression. A port of
+  // the native Python `range()` function. See
+  // [the Python documentation](https://docs.python.org/library/functions.html#range).
+  function range(start, stop, step) {
+    if (stop == null) {
+      stop = start || 0;
+      start = 0;
+    }
+    if (!step) {
+      step = stop < start ? -1 : 1;
+    }
+
+    var length = Math.max(Math.ceil((stop - start) / step), 0);
+    var range = Array(length);
+
+    for (var idx = 0; idx < length; idx++, start += step) {
+      range[idx] = start;
+    }
+
+    return range;
+  }
+
+  // Chunk a single array into multiple arrays, each containing `count` or fewer
+  // items.
+  function chunk(array, count) {
+    if (count == null || count < 1) return [];
+    var result = [];
+    var i = 0, length = array.length;
+    while (i < length) {
+      result.push(slice.call(array, i, i += count));
+    }
+    return result;
+  }
+
+  // Helper function to continue chaining intermediate results.
+  function chainResult(instance, obj) {
+    return instance._chain ? _$1(obj).chain() : obj;
+  }
+
+  // Add your own custom functions to the Underscore object.
+  function mixin(obj) {
+    each(functions(obj), function(name) {
+      var func = _$1[name] = obj[name];
+      _$1.prototype[name] = function() {
+        var args = [this._wrapped];
+        push.apply(args, arguments);
+        return chainResult(this, func.apply(_$1, args));
+      };
+    });
+    return _$1;
+  }
+
+  // Add all mutator `Array` functions to the wrapper.
+  each(['pop', 'push', 'reverse', 'shift', 'sort', 'splice', 'unshift'], function(name) {
+    var method = ArrayProto[name];
+    _$1.prototype[name] = function() {
+      var obj = this._wrapped;
+      if (obj != null) {
+        method.apply(obj, arguments);
+        if ((name === 'shift' || name === 'splice') && obj.length === 0) {
+          delete obj[0];
+        }
+      }
+      return chainResult(this, obj);
+    };
+  });
+
+  // Add all accessor `Array` functions to the wrapper.
+  each(['concat', 'join', 'slice'], function(name) {
+    var method = ArrayProto[name];
+    _$1.prototype[name] = function() {
+      var obj = this._wrapped;
+      if (obj != null) obj = method.apply(obj, arguments);
+      return chainResult(this, obj);
+    };
+  });
+
+  // Named Exports
+
+  var allExports = {
+    __proto__: null,
+    VERSION: VERSION,
+    restArguments: restArguments,
+    isObject: isObject,
+    isNull: isNull,
+    isUndefined: isUndefined,
+    isBoolean: isBoolean,
+    isElement: isElement,
+    isString: isString,
+    isNumber: isNumber,
+    isDate: isDate,
+    isRegExp: isRegExp,
+    isError: isError,
+    isSymbol: isSymbol,
+    isArrayBuffer: isArrayBuffer,
+    isDataView: isDataView$1,
+    isArray: isArray,
+    isFunction: isFunction$1,
+    isArguments: isArguments$1,
+    isFinite: isFinite$1,
+    isNaN: isNaN$1,
+    isTypedArray: isTypedArray$1,
+    isEmpty: isEmpty,
+    isMatch: isMatch,
+    isEqual: isEqual,
+    isMap: isMap,
+    isWeakMap: isWeakMap,
+    isSet: isSet,
+    isWeakSet: isWeakSet,
+    keys: keys,
+    allKeys: allKeys,
+    values: values,
+    pairs: pairs,
+    invert: invert,
+    functions: functions,
+    methods: functions,
+    extend: extend,
+    extendOwn: extendOwn,
+    assign: extendOwn,
+    defaults: defaults,
+    create: create,
+    clone: clone,
+    tap: tap,
+    get: get,
+    has: has,
+    mapObject: mapObject,
+    identity: identity,
+    constant: constant,
+    noop: noop,
+    toPath: toPath$1,
+    property: property,
+    propertyOf: propertyOf,
+    matcher: matcher,
+    matches: matcher,
+    times: times,
+    random: random,
+    now: now,
+    escape: _escape,
+    unescape: _unescape,
+    templateSettings: templateSettings,
+    template: template,
+    result: result,
+    uniqueId: uniqueId,
+    chain: chain,
+    iteratee: iteratee,
+    partial: partial,
+    bind: bind,
+    bindAll: bindAll,
+    memoize: memoize,
+    delay: delay,
+    defer: defer,
+    throttle: throttle,
+    debounce: debounce,
+    wrap: wrap,
+    negate: negate,
+    compose: compose,
+    after: after,
+    before: before,
+    once: once,
+    findKey: findKey,
+    findIndex: findIndex,
+    findLastIndex: findLastIndex,
+    sortedIndex: sortedIndex,
+    indexOf: indexOf,
+    lastIndexOf: lastIndexOf,
+    find: find,
+    detect: find,
+    findWhere: findWhere,
+    each: each,
+    forEach: each,
+    map: map,
+    collect: map,
+    reduce: reduce,
+    foldl: reduce,
+    inject: reduce,
+    reduceRight: reduceRight,
+    foldr: reduceRight,
+    filter: filter,
+    select: filter,
+    reject: reject,
+    every: every,
+    all: every,
+    some: some,
+    any: some,
+    contains: contains,
+    includes: contains,
+    include: contains,
+    invoke: invoke,
+    pluck: pluck,
+    where: where,
+    max: max,
+    min: min,
+    shuffle: shuffle,
+    sample: sample,
+    sortBy: sortBy,
+    groupBy: groupBy,
+    indexBy: indexBy,
+    countBy: countBy,
+    partition: partition,
+    toArray: toArray,
+    size: size,
+    pick: pick,
+    omit: omit,
+    first: first,
+    head: first,
+    take: first,
+    initial: initial,
+    last: last,
+    rest: rest,
+    tail: rest,
+    drop: rest,
+    compact: compact,
+    flatten: flatten,
+    without: without,
+    uniq: uniq,
+    unique: uniq,
+    union: union,
+    intersection: intersection,
+    difference: difference,
+    unzip: unzip,
+    transpose: unzip,
+    zip: zip,
+    object: object,
+    range: range,
+    chunk: chunk,
+    mixin: mixin,
+    'default': _$1
+  };
+
+  // Default Export
+
+  // Add all of the Underscore functions to the wrapper object.
+  var _ = mixin(allExports);
+  // Legacy Node.js API.
+  _._ = _;
+
+  return _;
+
+})));
+//# sourceMappingURL=underscore-umd.js.map
diff --git a/0.16/_static/underscore.js b/0.16/_static/underscore.js
new file mode 100644
index 00000000000..cf177d4285a
--- /dev/null
+++ b/0.16/_static/underscore.js
@@ -0,0 +1,6 @@
+!function(n,r){"object"==typeof exports&&"undefined"!=typeof module?module.exports=r():"function"==typeof define&&define.amd?define("underscore",r):(n="undefined"!=typeof globalThis?globalThis:n||self,function(){var t=n._,e=n._=r();e.noConflict=function(){return n._=t,e}}())}(this,(function(){
+//     Underscore.js 1.13.1
+//     https://underscorejs.org
+//     (c) 2009-2021 Jeremy Ashkenas, Julian Gonggrijp, and DocumentCloud and Investigative Reporters & Editors
+//     Underscore may be freely distributed under the MIT license.
+var n="1.13.1",r="object"==typeof self&&self.self===self&&self||"object"==typeof global&&global.global===global&&global||Function("return this")()||{},t=Array.prototype,e=Object.prototype,u="undefined"!=typeof Symbol?Symbol.prototype:null,o=t.push,i=t.slice,a=e.toString,f=e.hasOwnProperty,c="undefined"!=typeof ArrayBuffer,l="undefined"!=typeof DataView,s=Array.isArray,p=Object.keys,v=Object.create,h=c&&ArrayBuffer.isView,y=isNaN,d=isFinite,g=!{toString:null}.propertyIsEnumerable("toString"),b=["valueOf","isPrototypeOf","toString","propertyIsEnumerable","hasOwnProperty","toLocaleString"],m=Math.pow(2,53)-1;function j(n,r){return r=null==r?n.length-1:+r,function(){for(var t=Math.max(arguments.length-r,0),e=Array(t),u=0;u<t;u++)e[u]=arguments[u+r];switch(r){case 0:return n.call(this,e);case 1:return n.call(this,arguments[0],e);case 2:return n.call(this,arguments[0],arguments[1],e)}var o=Array(r+1);for(u=0;u<r;u++)o[u]=arguments[u];return o[r]=e,n.apply(this,o)}}function _(n){var r=typeof n;return"function"===r||"object"===r&&!!n}function w(n){return void 0===n}function A(n){return!0===n||!1===n||"[object Boolean]"===a.call(n)}function x(n){var r="[object "+n+"]";return function(n){return a.call(n)===r}}var S=x("String"),O=x("Number"),M=x("Date"),E=x("RegExp"),B=x("Error"),N=x("Symbol"),I=x("ArrayBuffer"),T=x("Function"),k=r.document&&r.document.childNodes;"function"!=typeof/./&&"object"!=typeof Int8Array&&"function"!=typeof k&&(T=function(n){return"function"==typeof n||!1});var D=T,R=x("Object"),F=l&&R(new DataView(new ArrayBuffer(8))),V="undefined"!=typeof Map&&R(new Map),P=x("DataView");var q=F?function(n){return null!=n&&D(n.getInt8)&&I(n.buffer)}:P,U=s||x("Array");function W(n,r){return null!=n&&f.call(n,r)}var z=x("Arguments");!function(){z(arguments)||(z=function(n){return W(n,"callee")})}();var L=z;function $(n){return O(n)&&y(n)}function C(n){return function(){return n}}function K(n){return function(r){var t=n(r);return"number"==typeof t&&t>=0&&t<=m}}function J(n){return function(r){return null==r?void 0:r[n]}}var G=J("byteLength"),H=K(G),Q=/\[object ((I|Ui)nt(8|16|32)|Float(32|64)|Uint8Clamped|Big(I|Ui)nt64)Array\]/;var X=c?function(n){return h?h(n)&&!q(n):H(n)&&Q.test(a.call(n))}:C(!1),Y=J("length");function Z(n,r){r=function(n){for(var r={},t=n.length,e=0;e<t;++e)r[n[e]]=!0;return{contains:function(n){return r[n]},push:function(t){return r[t]=!0,n.push(t)}}}(r);var t=b.length,u=n.constructor,o=D(u)&&u.prototype||e,i="constructor";for(W(n,i)&&!r.contains(i)&&r.push(i);t--;)(i=b[t])in n&&n[i]!==o[i]&&!r.contains(i)&&r.push(i)}function nn(n){if(!_(n))return[];if(p)return p(n);var r=[];for(var t in n)W(n,t)&&r.push(t);return g&&Z(n,r),r}function rn(n,r){var t=nn(r),e=t.length;if(null==n)return!e;for(var u=Object(n),o=0;o<e;o++){var i=t[o];if(r[i]!==u[i]||!(i in u))return!1}return!0}function tn(n){return n instanceof tn?n:this instanceof tn?void(this._wrapped=n):new tn(n)}function en(n){return new Uint8Array(n.buffer||n,n.byteOffset||0,G(n))}tn.VERSION=n,tn.prototype.value=function(){return this._wrapped},tn.prototype.valueOf=tn.prototype.toJSON=tn.prototype.value,tn.prototype.toString=function(){return String(this._wrapped)};var un="[object DataView]";function on(n,r,t,e){if(n===r)return 0!==n||1/n==1/r;if(null==n||null==r)return!1;if(n!=n)return r!=r;var o=typeof n;return("function"===o||"object"===o||"object"==typeof r)&&function n(r,t,e,o){r instanceof tn&&(r=r._wrapped);t instanceof tn&&(t=t._wrapped);var i=a.call(r);if(i!==a.call(t))return!1;if(F&&"[object Object]"==i&&q(r)){if(!q(t))return!1;i=un}switch(i){case"[object RegExp]":case"[object String]":return""+r==""+t;case"[object Number]":return+r!=+r?+t!=+t:0==+r?1/+r==1/t:+r==+t;case"[object Date]":case"[object Boolean]":return+r==+t;case"[object Symbol]":return u.valueOf.call(r)===u.valueOf.call(t);case"[object ArrayBuffer]":case un:return n(en(r),en(t),e,o)}var f="[object Array]"===i;if(!f&&X(r)){if(G(r)!==G(t))return!1;if(r.buffer===t.buffer&&r.byteOffset===t.byteOffset)return!0;f=!0}if(!f){if("object"!=typeof r||"object"!=typeof t)return!1;var c=r.constructor,l=t.constructor;if(c!==l&&!(D(c)&&c instanceof c&&D(l)&&l instanceof l)&&"constructor"in r&&"constructor"in t)return!1}o=o||[];var s=(e=e||[]).length;for(;s--;)if(e[s]===r)return o[s]===t;if(e.push(r),o.push(t),f){if((s=r.length)!==t.length)return!1;for(;s--;)if(!on(r[s],t[s],e,o))return!1}else{var p,v=nn(r);if(s=v.length,nn(t).length!==s)return!1;for(;s--;)if(p=v[s],!W(t,p)||!on(r[p],t[p],e,o))return!1}return e.pop(),o.pop(),!0}(n,r,t,e)}function an(n){if(!_(n))return[];var r=[];for(var t in n)r.push(t);return g&&Z(n,r),r}function fn(n){var r=Y(n);return function(t){if(null==t)return!1;var e=an(t);if(Y(e))return!1;for(var u=0;u<r;u++)if(!D(t[n[u]]))return!1;return n!==hn||!D(t[cn])}}var cn="forEach",ln="has",sn=["clear","delete"],pn=["get",ln,"set"],vn=sn.concat(cn,pn),hn=sn.concat(pn),yn=["add"].concat(sn,cn,ln),dn=V?fn(vn):x("Map"),gn=V?fn(hn):x("WeakMap"),bn=V?fn(yn):x("Set"),mn=x("WeakSet");function jn(n){for(var r=nn(n),t=r.length,e=Array(t),u=0;u<t;u++)e[u]=n[r[u]];return e}function _n(n){for(var r={},t=nn(n),e=0,u=t.length;e<u;e++)r[n[t[e]]]=t[e];return r}function wn(n){var r=[];for(var t in n)D(n[t])&&r.push(t);return r.sort()}function An(n,r){return function(t){var e=arguments.length;if(r&&(t=Object(t)),e<2||null==t)return t;for(var u=1;u<e;u++)for(var o=arguments[u],i=n(o),a=i.length,f=0;f<a;f++){var c=i[f];r&&void 0!==t[c]||(t[c]=o[c])}return t}}var xn=An(an),Sn=An(nn),On=An(an,!0);function Mn(n){if(!_(n))return{};if(v)return v(n);var r=function(){};r.prototype=n;var t=new r;return r.prototype=null,t}function En(n){return _(n)?U(n)?n.slice():xn({},n):n}function Bn(n){return U(n)?n:[n]}function Nn(n){return tn.toPath(n)}function In(n,r){for(var t=r.length,e=0;e<t;e++){if(null==n)return;n=n[r[e]]}return t?n:void 0}function Tn(n,r,t){var e=In(n,Nn(r));return w(e)?t:e}function kn(n){return n}function Dn(n){return n=Sn({},n),function(r){return rn(r,n)}}function Rn(n){return n=Nn(n),function(r){return In(r,n)}}function Fn(n,r,t){if(void 0===r)return n;switch(null==t?3:t){case 1:return function(t){return n.call(r,t)};case 3:return function(t,e,u){return n.call(r,t,e,u)};case 4:return function(t,e,u,o){return n.call(r,t,e,u,o)}}return function(){return n.apply(r,arguments)}}function Vn(n,r,t){return null==n?kn:D(n)?Fn(n,r,t):_(n)&&!U(n)?Dn(n):Rn(n)}function Pn(n,r){return Vn(n,r,1/0)}function qn(n,r,t){return tn.iteratee!==Pn?tn.iteratee(n,r):Vn(n,r,t)}function Un(){}function Wn(n,r){return null==r&&(r=n,n=0),n+Math.floor(Math.random()*(r-n+1))}tn.toPath=Bn,tn.iteratee=Pn;var zn=Date.now||function(){return(new Date).getTime()};function Ln(n){var r=function(r){return n[r]},t="(?:"+nn(n).join("|")+")",e=RegExp(t),u=RegExp(t,"g");return function(n){return n=null==n?"":""+n,e.test(n)?n.replace(u,r):n}}var $n={"&":"&amp;","<":"&lt;",">":"&gt;",'"':"&quot;","'":"&#x27;","`":"&#x60;"},Cn=Ln($n),Kn=Ln(_n($n)),Jn=tn.templateSettings={evaluate:/<%([\s\S]+?)%>/g,interpolate:/<%=([\s\S]+?)%>/g,escape:/<%-([\s\S]+?)%>/g},Gn=/(.)^/,Hn={"'":"'","\\":"\\","\r":"r","\n":"n","\u2028":"u2028","\u2029":"u2029"},Qn=/\\|'|\r|\n|\u2028|\u2029/g;function Xn(n){return"\\"+Hn[n]}var Yn=/^\s*(\w|\$)+\s*$/;var Zn=0;function nr(n,r,t,e,u){if(!(e instanceof r))return n.apply(t,u);var o=Mn(n.prototype),i=n.apply(o,u);return _(i)?i:o}var rr=j((function(n,r){var t=rr.placeholder,e=function(){for(var u=0,o=r.length,i=Array(o),a=0;a<o;a++)i[a]=r[a]===t?arguments[u++]:r[a];for(;u<arguments.length;)i.push(arguments[u++]);return nr(n,e,this,this,i)};return e}));rr.placeholder=tn;var tr=j((function(n,r,t){if(!D(n))throw new TypeError("Bind must be called on a function");var e=j((function(u){return nr(n,e,r,this,t.concat(u))}));return e})),er=K(Y);function ur(n,r,t,e){if(e=e||[],r||0===r){if(r<=0)return e.concat(n)}else r=1/0;for(var u=e.length,o=0,i=Y(n);o<i;o++){var a=n[o];if(er(a)&&(U(a)||L(a)))if(r>1)ur(a,r-1,t,e),u=e.length;else for(var f=0,c=a.length;f<c;)e[u++]=a[f++];else t||(e[u++]=a)}return e}var or=j((function(n,r){var t=(r=ur(r,!1,!1)).length;if(t<1)throw new Error("bindAll must be passed function names");for(;t--;){var e=r[t];n[e]=tr(n[e],n)}return n}));var ir=j((function(n,r,t){return setTimeout((function(){return n.apply(null,t)}),r)})),ar=rr(ir,tn,1);function fr(n){return function(){return!n.apply(this,arguments)}}function cr(n,r){var t;return function(){return--n>0&&(t=r.apply(this,arguments)),n<=1&&(r=null),t}}var lr=rr(cr,2);function sr(n,r,t){r=qn(r,t);for(var e,u=nn(n),o=0,i=u.length;o<i;o++)if(r(n[e=u[o]],e,n))return e}function pr(n){return function(r,t,e){t=qn(t,e);for(var u=Y(r),o=n>0?0:u-1;o>=0&&o<u;o+=n)if(t(r[o],o,r))return o;return-1}}var vr=pr(1),hr=pr(-1);function yr(n,r,t,e){for(var u=(t=qn(t,e,1))(r),o=0,i=Y(n);o<i;){var a=Math.floor((o+i)/2);t(n[a])<u?o=a+1:i=a}return o}function dr(n,r,t){return function(e,u,o){var a=0,f=Y(e);if("number"==typeof o)n>0?a=o>=0?o:Math.max(o+f,a):f=o>=0?Math.min(o+1,f):o+f+1;else if(t&&o&&f)return e[o=t(e,u)]===u?o:-1;if(u!=u)return(o=r(i.call(e,a,f),$))>=0?o+a:-1;for(o=n>0?a:f-1;o>=0&&o<f;o+=n)if(e[o]===u)return o;return-1}}var gr=dr(1,vr,yr),br=dr(-1,hr);function mr(n,r,t){var e=(er(n)?vr:sr)(n,r,t);if(void 0!==e&&-1!==e)return n[e]}function jr(n,r,t){var e,u;if(r=Fn(r,t),er(n))for(e=0,u=n.length;e<u;e++)r(n[e],e,n);else{var o=nn(n);for(e=0,u=o.length;e<u;e++)r(n[o[e]],o[e],n)}return n}function _r(n,r,t){r=qn(r,t);for(var e=!er(n)&&nn(n),u=(e||n).length,o=Array(u),i=0;i<u;i++){var a=e?e[i]:i;o[i]=r(n[a],a,n)}return o}function wr(n){var r=function(r,t,e,u){var o=!er(r)&&nn(r),i=(o||r).length,a=n>0?0:i-1;for(u||(e=r[o?o[a]:a],a+=n);a>=0&&a<i;a+=n){var f=o?o[a]:a;e=t(e,r[f],f,r)}return e};return function(n,t,e,u){var o=arguments.length>=3;return r(n,Fn(t,u,4),e,o)}}var Ar=wr(1),xr=wr(-1);function Sr(n,r,t){var e=[];return r=qn(r,t),jr(n,(function(n,t,u){r(n,t,u)&&e.push(n)})),e}function Or(n,r,t){r=qn(r,t);for(var e=!er(n)&&nn(n),u=(e||n).length,o=0;o<u;o++){var i=e?e[o]:o;if(!r(n[i],i,n))return!1}return!0}function Mr(n,r,t){r=qn(r,t);for(var e=!er(n)&&nn(n),u=(e||n).length,o=0;o<u;o++){var i=e?e[o]:o;if(r(n[i],i,n))return!0}return!1}function Er(n,r,t,e){return er(n)||(n=jn(n)),("number"!=typeof t||e)&&(t=0),gr(n,r,t)>=0}var Br=j((function(n,r,t){var e,u;return D(r)?u=r:(r=Nn(r),e=r.slice(0,-1),r=r[r.length-1]),_r(n,(function(n){var o=u;if(!o){if(e&&e.length&&(n=In(n,e)),null==n)return;o=n[r]}return null==o?o:o.apply(n,t)}))}));function Nr(n,r){return _r(n,Rn(r))}function Ir(n,r,t){var e,u,o=-1/0,i=-1/0;if(null==r||"number"==typeof r&&"object"!=typeof n[0]&&null!=n)for(var a=0,f=(n=er(n)?n:jn(n)).length;a<f;a++)null!=(e=n[a])&&e>o&&(o=e);else r=qn(r,t),jr(n,(function(n,t,e){((u=r(n,t,e))>i||u===-1/0&&o===-1/0)&&(o=n,i=u)}));return o}function Tr(n,r,t){if(null==r||t)return er(n)||(n=jn(n)),n[Wn(n.length-1)];var e=er(n)?En(n):jn(n),u=Y(e);r=Math.max(Math.min(r,u),0);for(var o=u-1,i=0;i<r;i++){var a=Wn(i,o),f=e[i];e[i]=e[a],e[a]=f}return e.slice(0,r)}function kr(n,r){return function(t,e,u){var o=r?[[],[]]:{};return e=qn(e,u),jr(t,(function(r,u){var i=e(r,u,t);n(o,r,i)})),o}}var Dr=kr((function(n,r,t){W(n,t)?n[t].push(r):n[t]=[r]})),Rr=kr((function(n,r,t){n[t]=r})),Fr=kr((function(n,r,t){W(n,t)?n[t]++:n[t]=1})),Vr=kr((function(n,r,t){n[t?0:1].push(r)}),!0),Pr=/[^\ud800-\udfff]|[\ud800-\udbff][\udc00-\udfff]|[\ud800-\udfff]/g;function qr(n,r,t){return r in t}var Ur=j((function(n,r){var t={},e=r[0];if(null==n)return t;D(e)?(r.length>1&&(e=Fn(e,r[1])),r=an(n)):(e=qr,r=ur(r,!1,!1),n=Object(n));for(var u=0,o=r.length;u<o;u++){var i=r[u],a=n[i];e(a,i,n)&&(t[i]=a)}return t})),Wr=j((function(n,r){var t,e=r[0];return D(e)?(e=fr(e),r.length>1&&(t=r[1])):(r=_r(ur(r,!1,!1),String),e=function(n,t){return!Er(r,t)}),Ur(n,e,t)}));function zr(n,r,t){return i.call(n,0,Math.max(0,n.length-(null==r||t?1:r)))}function Lr(n,r,t){return null==n||n.length<1?null==r||t?void 0:[]:null==r||t?n[0]:zr(n,n.length-r)}function $r(n,r,t){return i.call(n,null==r||t?1:r)}var Cr=j((function(n,r){return r=ur(r,!0,!0),Sr(n,(function(n){return!Er(r,n)}))})),Kr=j((function(n,r){return Cr(n,r)}));function Jr(n,r,t,e){A(r)||(e=t,t=r,r=!1),null!=t&&(t=qn(t,e));for(var u=[],o=[],i=0,a=Y(n);i<a;i++){var f=n[i],c=t?t(f,i,n):f;r&&!t?(i&&o===c||u.push(f),o=c):t?Er(o,c)||(o.push(c),u.push(f)):Er(u,f)||u.push(f)}return u}var Gr=j((function(n){return Jr(ur(n,!0,!0))}));function Hr(n){for(var r=n&&Ir(n,Y).length||0,t=Array(r),e=0;e<r;e++)t[e]=Nr(n,e);return t}var Qr=j(Hr);function Xr(n,r){return n._chain?tn(r).chain():r}function Yr(n){return jr(wn(n),(function(r){var t=tn[r]=n[r];tn.prototype[r]=function(){var n=[this._wrapped];return o.apply(n,arguments),Xr(this,t.apply(tn,n))}})),tn}jr(["pop","push","reverse","shift","sort","splice","unshift"],(function(n){var r=t[n];tn.prototype[n]=function(){var t=this._wrapped;return null!=t&&(r.apply(t,arguments),"shift"!==n&&"splice"!==n||0!==t.length||delete t[0]),Xr(this,t)}})),jr(["concat","join","slice"],(function(n){var r=t[n];tn.prototype[n]=function(){var n=this._wrapped;return null!=n&&(n=r.apply(n,arguments)),Xr(this,n)}}));var Zr=Yr({__proto__:null,VERSION:n,restArguments:j,isObject:_,isNull:function(n){return null===n},isUndefined:w,isBoolean:A,isElement:function(n){return!(!n||1!==n.nodeType)},isString:S,isNumber:O,isDate:M,isRegExp:E,isError:B,isSymbol:N,isArrayBuffer:I,isDataView:q,isArray:U,isFunction:D,isArguments:L,isFinite:function(n){return!N(n)&&d(n)&&!isNaN(parseFloat(n))},isNaN:$,isTypedArray:X,isEmpty:function(n){if(null==n)return!0;var r=Y(n);return"number"==typeof r&&(U(n)||S(n)||L(n))?0===r:0===Y(nn(n))},isMatch:rn,isEqual:function(n,r){return on(n,r)},isMap:dn,isWeakMap:gn,isSet:bn,isWeakSet:mn,keys:nn,allKeys:an,values:jn,pairs:function(n){for(var r=nn(n),t=r.length,e=Array(t),u=0;u<t;u++)e[u]=[r[u],n[r[u]]];return e},invert:_n,functions:wn,methods:wn,extend:xn,extendOwn:Sn,assign:Sn,defaults:On,create:function(n,r){var t=Mn(n);return r&&Sn(t,r),t},clone:En,tap:function(n,r){return r(n),n},get:Tn,has:function(n,r){for(var t=(r=Nn(r)).length,e=0;e<t;e++){var u=r[e];if(!W(n,u))return!1;n=n[u]}return!!t},mapObject:function(n,r,t){r=qn(r,t);for(var e=nn(n),u=e.length,o={},i=0;i<u;i++){var a=e[i];o[a]=r(n[a],a,n)}return o},identity:kn,constant:C,noop:Un,toPath:Bn,property:Rn,propertyOf:function(n){return null==n?Un:function(r){return Tn(n,r)}},matcher:Dn,matches:Dn,times:function(n,r,t){var e=Array(Math.max(0,n));r=Fn(r,t,1);for(var u=0;u<n;u++)e[u]=r(u);return e},random:Wn,now:zn,escape:Cn,unescape:Kn,templateSettings:Jn,template:function(n,r,t){!r&&t&&(r=t),r=On({},r,tn.templateSettings);var e=RegExp([(r.escape||Gn).source,(r.interpolate||Gn).source,(r.evaluate||Gn).source].join("|")+"|$","g"),u=0,o="__p+='";n.replace(e,(function(r,t,e,i,a){return o+=n.slice(u,a).replace(Qn,Xn),u=a+r.length,t?o+="'+\n((__t=("+t+"))==null?'':_.escape(__t))+\n'":e?o+="'+\n((__t=("+e+"))==null?'':__t)+\n'":i&&(o+="';\n"+i+"\n__p+='"),r})),o+="';\n";var i,a=r.variable;if(a){if(!Yn.test(a))throw new Error("variable is not a bare identifier: "+a)}else o="with(obj||{}){\n"+o+"}\n",a="obj";o="var __t,__p='',__j=Array.prototype.join,"+"print=function(){__p+=__j.call(arguments,'');};\n"+o+"return __p;\n";try{i=new Function(a,"_",o)}catch(n){throw n.source=o,n}var f=function(n){return i.call(this,n,tn)};return f.source="function("+a+"){\n"+o+"}",f},result:function(n,r,t){var e=(r=Nn(r)).length;if(!e)return D(t)?t.call(n):t;for(var u=0;u<e;u++){var o=null==n?void 0:n[r[u]];void 0===o&&(o=t,u=e),n=D(o)?o.call(n):o}return n},uniqueId:function(n){var r=++Zn+"";return n?n+r:r},chain:function(n){var r=tn(n);return r._chain=!0,r},iteratee:Pn,partial:rr,bind:tr,bindAll:or,memoize:function(n,r){var t=function(e){var u=t.cache,o=""+(r?r.apply(this,arguments):e);return W(u,o)||(u[o]=n.apply(this,arguments)),u[o]};return t.cache={},t},delay:ir,defer:ar,throttle:function(n,r,t){var e,u,o,i,a=0;t||(t={});var f=function(){a=!1===t.leading?0:zn(),e=null,i=n.apply(u,o),e||(u=o=null)},c=function(){var c=zn();a||!1!==t.leading||(a=c);var l=r-(c-a);return u=this,o=arguments,l<=0||l>r?(e&&(clearTimeout(e),e=null),a=c,i=n.apply(u,o),e||(u=o=null)):e||!1===t.trailing||(e=setTimeout(f,l)),i};return c.cancel=function(){clearTimeout(e),a=0,e=u=o=null},c},debounce:function(n,r,t){var e,u,o,i,a,f=function(){var c=zn()-u;r>c?e=setTimeout(f,r-c):(e=null,t||(i=n.apply(a,o)),e||(o=a=null))},c=j((function(c){return a=this,o=c,u=zn(),e||(e=setTimeout(f,r),t&&(i=n.apply(a,o))),i}));return c.cancel=function(){clearTimeout(e),e=o=a=null},c},wrap:function(n,r){return rr(r,n)},negate:fr,compose:function(){var n=arguments,r=n.length-1;return function(){for(var t=r,e=n[r].apply(this,arguments);t--;)e=n[t].call(this,e);return e}},after:function(n,r){return function(){if(--n<1)return r.apply(this,arguments)}},before:cr,once:lr,findKey:sr,findIndex:vr,findLastIndex:hr,sortedIndex:yr,indexOf:gr,lastIndexOf:br,find:mr,detect:mr,findWhere:function(n,r){return mr(n,Dn(r))},each:jr,forEach:jr,map:_r,collect:_r,reduce:Ar,foldl:Ar,inject:Ar,reduceRight:xr,foldr:xr,filter:Sr,select:Sr,reject:function(n,r,t){return Sr(n,fr(qn(r)),t)},every:Or,all:Or,some:Mr,any:Mr,contains:Er,includes:Er,include:Er,invoke:Br,pluck:Nr,where:function(n,r){return Sr(n,Dn(r))},max:Ir,min:function(n,r,t){var e,u,o=1/0,i=1/0;if(null==r||"number"==typeof r&&"object"!=typeof n[0]&&null!=n)for(var a=0,f=(n=er(n)?n:jn(n)).length;a<f;a++)null!=(e=n[a])&&e<o&&(o=e);else r=qn(r,t),jr(n,(function(n,t,e){((u=r(n,t,e))<i||u===1/0&&o===1/0)&&(o=n,i=u)}));return o},shuffle:function(n){return Tr(n,1/0)},sample:Tr,sortBy:function(n,r,t){var e=0;return r=qn(r,t),Nr(_r(n,(function(n,t,u){return{value:n,index:e++,criteria:r(n,t,u)}})).sort((function(n,r){var t=n.criteria,e=r.criteria;if(t!==e){if(t>e||void 0===t)return 1;if(t<e||void 0===e)return-1}return n.index-r.index})),"value")},groupBy:Dr,indexBy:Rr,countBy:Fr,partition:Vr,toArray:function(n){return n?U(n)?i.call(n):S(n)?n.match(Pr):er(n)?_r(n,kn):jn(n):[]},size:function(n){return null==n?0:er(n)?n.length:nn(n).length},pick:Ur,omit:Wr,first:Lr,head:Lr,take:Lr,initial:zr,last:function(n,r,t){return null==n||n.length<1?null==r||t?void 0:[]:null==r||t?n[n.length-1]:$r(n,Math.max(0,n.length-r))},rest:$r,tail:$r,drop:$r,compact:function(n){return Sr(n,Boolean)},flatten:function(n,r){return ur(n,r,!1)},without:Kr,uniq:Jr,unique:Jr,union:Gr,intersection:function(n){for(var r=[],t=arguments.length,e=0,u=Y(n);e<u;e++){var o=n[e];if(!Er(r,o)){var i;for(i=1;i<t&&Er(arguments[i],o);i++);i===t&&r.push(o)}}return r},difference:Cr,unzip:Hr,transpose:Hr,zip:Qr,object:function(n,r){for(var t={},e=0,u=Y(n);e<u;e++)r?t[n[e]]=r[e]:t[n[e][0]]=n[e][1];return t},range:function(n,r,t){null==r&&(r=n||0,n=0),t||(t=r<n?-1:1);for(var e=Math.max(Math.ceil((r-n)/t),0),u=Array(e),o=0;o<e;o++,n+=t)u[o]=n;return u},chunk:function(n,r){if(null==r||r<1)return[];for(var t=[],e=0,u=n.length;e<u;)t.push(i.call(n,e,e+=r));return t},mixin:Yr,default:tn});return Zr._=Zr,Zr}));
\ No newline at end of file
diff --git a/0.16/auto_examples/index.html b/0.16/auto_examples/index.html
new file mode 100644
index 00000000000..0b44a607aa9
--- /dev/null
+++ b/0.16/auto_examples/index.html
@@ -0,0 +1,743 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Examples and tutorials &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Transforms" href="transforms/index.html" />
+    <link rel="prev" title="get_graph_node_names" href="../generated/torchvision.models.feature_extraction.get_graph_node_names.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Examples and tutorials</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/auto_examples/index.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="examples-and-tutorials">
+<span id="gallery"></span><h1>Examples and tutorials<a class="headerlink" href="#examples-and-tutorials" title="Permalink to this heading">¶</a></h1>
+<div class="sphx-glr-thumbnails"></div><section id="transforms">
+<h2>Transforms<a class="headerlink" href="#transforms" title="Permalink to this heading">¶</a></h2>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div></div></section>
+<section id="others">
+<h2>Others<a class="headerlink" href="#others" title="Permalink to this heading">¶</a></h2>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images."><img alt="" src="../_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png" />
+<p><a class="reference internal" href="others/plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the..."><img alt="" src="../_images/sphx_glr_plot_video_api_thumb.png" />
+<p><a class="reference internal" href="others/plot_video_api.html#sphx-glr-auto-examples-others-plot-video-api-py"><span class="std std-ref">Video API</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Video API</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div><div class="toctree-wrapper compound">
+</div>
+<div class="sphx-glr-footer sphx-glr-footer-gallery docutils container">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">all</span> <span class="pre">examples</span> <span class="pre">in</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">auto_examples_python.zip</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">all</span> <span class="pre">examples</span> <span class="pre">in</span> <span class="pre">Jupyter</span> <span class="pre">notebooks:</span> <span class="pre">auto_examples_jupyter.zip</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="transforms/index.html" class="btn btn-neutral float-right" title="Transforms" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../generated/torchvision.models.feature_extraction.get_graph_node_names.html" class="btn btn-neutral" title="get_graph_node_names" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Examples and tutorials</a><ul>
+<li><a class="reference internal" href="#transforms">Transforms</a></li>
+<li><a class="reference internal" href="#others">Others</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/others/index.html b/0.16/auto_examples/others/index.html
new file mode 100644
index 00000000000..f62686b0620
--- /dev/null
+++ b/0.16/auto_examples/others/index.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Others &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Optical Flow: Predicting movement with the RAFT model" href="plot_optical_flow.html" />
+    <link rel="prev" title="How to write your own TVTensor class" href="../transforms/plot_custom_tv_tensors.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+      <li>Others</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/others/index.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="others">
+<span id="sphx-glr-auto-examples-others"></span><h1>Others<a class="headerlink" href="#others" title="Permalink to this heading">¶</a></h1>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images."><img alt="" src="../../_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png" />
+<p><a class="reference internal" href="plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the..."><img alt="" src="../../_images/sphx_glr_plot_video_api_thumb.png" />
+<p><a class="reference internal" href="plot_video_api.html#sphx-glr-auto-examples-others-plot-video-api-py"><span class="std std-ref">Video API</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Video API</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div><div class="toctree-wrapper compound">
+</div>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_optical_flow.html" class="btn btn-neutral float-right" title="Optical Flow: Predicting movement with the RAFT model" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../transforms/plot_custom_tv_tensors.html" class="btn btn-neutral" title="How to write your own TVTensor class" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Others</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/others/plot_optical_flow.html b/0.16/auto_examples/others/plot_optical_flow.html
new file mode 100644
index 00000000000..202bd31696c
--- /dev/null
+++ b/0.16/auto_examples/others/plot_optical_flow.html
@@ -0,0 +1,904 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Optical Flow: Predicting movement with the RAFT model &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Repurposing masks into bounding boxes" href="plot_repurposing_annotations.html" />
+    <link rel="prev" title="Others" href="index.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Others</a> &gt;</li>
+        
+      <li>Optical Flow: Predicting movement with the RAFT model</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/others/plot_optical_flow.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="optical-flow-predicting-movement-with-the-raft-model">
+<span id="sphx-glr-auto-examples-others-plot-optical-flow-py"></span><h1>Optical Flow: Predicting movement with the RAFT model<a class="headerlink" href="#optical-flow-predicting-movement-with-the-raft-model" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_optical_flow.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>Optical flow is the task of predicting movement between two images, usually two
+consecutive frames of a video. Optical flow models take two images as input, and
+predict a flow: the flow indicates the displacement of every single pixel in the
+first image, and maps it to its corresponding pixel in the second image. Flows
+are (2, H, W)-dimensional tensors, where the first axis corresponds to the
+predicted horizontal and vertical displacements.</p>
+<p>The following example illustrates how torchvision can be used to predict flows
+using our implementation of the RAFT model. We will also see how to convert the
+predicted flows to RGB images for visualization.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
+<span class="kn">import</span> <span class="nn">torchvision.transforms.functional</span> <span class="k">as</span> <span class="nn">F</span>
+
+
+<a href="https://matplotlib.org/stable/api/matplotlib_configuration_api.html#matplotlib.RcParams" title="matplotlib.RcParams" class="sphx-glr-backref-module-matplotlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">plt</span><span class="o">.</span><span class="n">rcParams</span></a><span class="p">[</span><span class="s2">&quot;savefig.bbox&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;tight&quot;</span>
+
+
+<span class="k">def</span> <span class="nf">plot</span><span class="p">(</span><span class="n">imgs</span><span class="p">,</span> <span class="o">**</span><span class="n">imshow_kwargs</span><span class="p">):</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">imgs</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="nb">list</span><span class="p">):</span>
+        <span class="c1"># Make a 2d grid even if there&#39;s just 1 row</span>
+        <span class="n">imgs</span> <span class="o">=</span> <span class="p">[</span><span class="n">imgs</span><span class="p">]</span>
+
+    <span class="n">num_rows</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">imgs</span><span class="p">)</span>
+    <span class="n">num_cols</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">imgs</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">_</span></a><span class="p">,</span> <span class="n">axs</span> <span class="o">=</span> <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.subplots.html#matplotlib.pyplot.subplots" title="matplotlib.pyplot.subplots" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">subplots</span></a><span class="p">(</span><span class="n">nrows</span><span class="o">=</span><span class="n">num_rows</span><span class="p">,</span> <span class="n">ncols</span><span class="o">=</span><span class="n">num_cols</span><span class="p">,</span> <span class="n">squeeze</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">row_idx</span><span class="p">,</span> <span class="n">row</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">imgs</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">col_idx</span><span class="p">,</span> <span class="n">img</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">row</span><span class="p">):</span>
+            <span class="n">ax</span> <span class="o">=</span> <span class="n">axs</span><span class="p">[</span><span class="n">row_idx</span><span class="p">,</span> <span class="n">col_idx</span><span class="p">]</span>
+            <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">to_pil_image</span><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span><span class="p">))</span>
+            <span class="n">ax</span><span class="o">.</span><span class="n">imshow</span><span class="p">(</span><a href="https://numpy.org/doc/stable/reference/generated/numpy.asarray.html#numpy.asarray" title="numpy.asarray" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">asarray</span></a><span class="p">(</span><span class="n">img</span><span class="p">),</span> <span class="o">**</span><span class="n">imshow_kwargs</span><span class="p">)</span>
+            <span class="n">ax</span><span class="o">.</span><span class="n">set</span><span class="p">(</span><span class="n">xticklabels</span><span class="o">=</span><span class="p">[],</span> <span class="n">yticklabels</span><span class="o">=</span><span class="p">[],</span> <span class="n">xticks</span><span class="o">=</span><span class="p">[],</span> <span class="n">yticks</span><span class="o">=</span><span class="p">[])</span>
+
+    <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.tight_layout.html#matplotlib.pyplot.tight_layout" title="matplotlib.pyplot.tight_layout" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">tight_layout</span></a><span class="p">()</span>
+</pre></div>
+</div>
+<section id="reading-videos-using-torchvision">
+<h2>Reading Videos Using Torchvision<a class="headerlink" href="#reading-videos-using-torchvision" title="Permalink to this heading">¶</a></h2>
+<p>We will first read a video using <a class="reference internal" href="../../generated/torchvision.io.read_video.html#torchvision.io.read_video" title="torchvision.io.read_video"><code class="xref py py-func docutils literal notranslate"><span class="pre">read_video()</span></code></a>.
+Alternatively one can use the new <a class="reference internal" href="../../generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader" title="torchvision.io.VideoReader"><code class="xref py py-class docutils literal notranslate"><span class="pre">VideoReader</span></code></a> API (if
+torchvision is built from source).
+The video we will use here is free of use from <a class="reference external" href="https://www.pexels.com/video/a-man-playing-a-game-of-basketball-5192157/">pexels.com</a>,
+credits go to <a class="reference external" href="https://www.pexels.com/&#64;pavel-danilyuk">Pavel Danilyuk</a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">tempfile</span>
+<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a>
+<span class="kn">from</span> <span class="nn">urllib.request</span> <span class="kn">import</span> <a href="https://docs.python.org/3/library/urllib.request.html#urllib.request.urlretrieve" title="urllib.request.urlretrieve" class="sphx-glr-backref-module-urllib-request sphx-glr-backref-type-py-function"><span class="n">urlretrieve</span></a>
+
+
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">video_url</span></a> <span class="o">=</span> <span class="s2">&quot;https://download.pytorch.org/tutorial/pexelscom_pavel_danilyuk_basketball_hd.mp4&quot;</span>
+<a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">video_path</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/tempfile.html#tempfile.mkdtemp" title="tempfile.mkdtemp" class="sphx-glr-backref-module-tempfile sphx-glr-backref-type-py-function"><span class="n">tempfile</span><span class="o">.</span><span class="n">mkdtemp</span></a><span class="p">())</span> <span class="o">/</span> <span class="s2">&quot;basketball.mp4&quot;</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">_</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/urllib.request.html#urllib.request.urlretrieve" title="urllib.request.urlretrieve" class="sphx-glr-backref-module-urllib-request sphx-glr-backref-type-py-function"><span class="n">urlretrieve</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">video_url</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">video_path</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<p><a class="reference internal" href="../../generated/torchvision.io.read_video.html#torchvision.io.read_video" title="torchvision.io.read_video"><code class="xref py py-func docutils literal notranslate"><span class="pre">read_video()</span></code></a> returns the video frames, audio frames and
+the metadata associated with the video. In our case, we only need the video
+frames.</p>
+<p>Here we will just make 2 predictions between 2 pre-selected pairs of frames,
+namely frames (100, 101) and (150, 151). Each of these pairs corresponds to a
+single model input.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_video</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">_</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">_</span></a> <span class="o">=</span> <span class="n">read_video</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">video_path</span></a><span class="p">),</span> <span class="n">output_format</span><span class="o">=</span><span class="s2">&quot;TCHW&quot;</span><span class="p">)</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.stack.html#torch.stack" title="torch.stack" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">stack</span></a><span class="p">([</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">[</span><span class="mi">100</span><span class="p">],</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">[</span><span class="mi">150</span><span class="p">]])</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.stack.html#torch.stack" title="torch.stack" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">stack</span></a><span class="p">([</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">[</span><span class="mi">101</span><span class="p">],</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">[</span><span class="mi">151</span><span class="p">]])</span>
+
+<span class="n">plot</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_optical_flow_001.png" srcset="../../_images/sphx_glr_plot_optical_flow_001.png" alt="plot optical flow" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/pytorch/vision/torchvision/io/video.py:161: UserWarning: The pts_unit &#39;pts&#39; gives wrong results. Please use pts_unit &#39;sec&#39;.
+  warnings.warn(&quot;The pts_unit &#39;pts&#39; gives wrong results. Please use pts_unit &#39;sec&#39;.&quot;)
+</pre></div>
+</div>
+<p>The RAFT model accepts RGB images. We first get the frames from
+<a class="reference internal" href="../../generated/torchvision.io.read_video.html#torchvision.io.read_video" title="torchvision.io.read_video"><code class="xref py py-func docutils literal notranslate"><span class="pre">read_video()</span></code></a> and resize them to ensure their dimensions
+are divisible by 8. Note that we explicitly use <code class="docutils literal notranslate"><span class="pre">antialias=False</span></code>, because
+this is how those models were trained. Then we use the transforms bundled into
+the weights in order to preprocess the input and rescale its values to the
+required <code class="docutils literal notranslate"><span class="pre">[-1,</span> <span class="pre">1]</span></code> interval.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models.optical_flow</span> <span class="kn">import</span> <span class="n">Raft_Large_Weights</span>
+
+<span class="n">weights</span> <span class="o">=</span> <span class="n">Raft_Large_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">transforms</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+
+<span class="k">def</span> <span class="nf">preprocess</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a><span class="p">):</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">resize</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">[</span><span class="mi">520</span><span class="p">,</span> <span class="mi">960</span><span class="p">],</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">resize</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">[</span><span class="mi">520</span><span class="p">,</span> <span class="mi">960</span><span class="p">],</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">transforms</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a><span class="p">)</span>
+
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a> <span class="o">=</span> <span class="n">preprocess</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a><span class="p">)</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;shape = </span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span><span class="o">.</span><span class="n">shape</span></a><span class="si">}</span><span class="s2">, dtype = </span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>shape = torch.Size([2, 3, 520, 960]), dtype = torch.float32
+</pre></div>
+</div>
+</section>
+<section id="estimating-optical-flow-using-raft">
+<h2>Estimating Optical flow using RAFT<a class="headerlink" href="#estimating-optical-flow-using-raft" title="Permalink to this heading">¶</a></h2>
+<p>We will use our RAFT implementation from
+<a class="reference internal" href="../../models/generated/torchvision.models.optical_flow.raft_large.html#torchvision.models.optical_flow.raft_large" title="torchvision.models.optical_flow.raft_large"><code class="xref py py-func docutils literal notranslate"><span class="pre">raft_large()</span></code></a>, which follows the same
+architecture as the one described in the <a class="reference external" href="https://arxiv.org/abs/2003.12039">original paper</a>.
+We also provide the <a class="reference internal" href="../../models/generated/torchvision.models.optical_flow.raft_small.html#torchvision.models.optical_flow.raft_small" title="torchvision.models.optical_flow.raft_small"><code class="xref py py-func docutils literal notranslate"><span class="pre">raft_small()</span></code></a> model
+builder, which is smaller and faster to run, sacrificing a bit of accuracy.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models.optical_flow</span> <span class="kn">import</span> <span class="n">raft_large</span>
+
+<span class="c1"># If you can, run this example on a GPU, it will be a lot faster.</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a> <span class="o">=</span> <span class="s2">&quot;cuda&quot;</span> <span class="k">if</span> <a href="https://pytorch.org/docs/stable/generated/torch.cuda.is_available.html#torch.cuda.is_available" title="torch.cuda.is_available" class="sphx-glr-backref-module-torch-cuda sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">is_available</span></a><span class="p">()</span> <span class="k">else</span> <span class="s2">&quot;cpu&quot;</span>
+
+<span class="n">model</span> <span class="o">=</span> <span class="n">raft_large</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">Raft_Large_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a><span class="p">)</span>
+<span class="n">model</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.eval" title="torch.nn.Module.eval" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-method"><span class="n">model</span><span class="o">.</span><span class="n">eval</span></a><span class="p">()</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">list_of_flows</span></a> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span><span class="o">.</span><span class="n">to</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a><span class="p">),</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img2_batch</span></a><span class="o">.</span><span class="n">to</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a><span class="p">))</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;type = </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">list_of_flows</span></a><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;length = </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">list_of_flows</span></a><span class="p">)</span><span class="si">}</span><span class="s2"> = number of iterations of the model&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/raft_large_C_T_SKHT_V2-ff5fadd5.pth&quot; to /root/.cache/torch/hub/checkpoints/raft_large_C_T_SKHT_V2-ff5fadd5.pth
+type = &lt;class &#39;list&#39;&gt;
+length = 12 = number of iterations of the model
+</pre></div>
+</div>
+<p>The RAFT model outputs lists of predicted flows where each entry is a
+(N, 2, H, W) batch of predicted flows that corresponds to a given “iteration”
+in the model. For more details on the iterative nature of the model, please
+refer to the <a class="reference external" href="https://arxiv.org/abs/2003.12039">original paper</a>. Here, we
+are only interested in the final predicted flows (they are the most accurate
+ones), so we will just retrieve the last item in the list.</p>
+<p>As described above, a flow is a tensor with dimensions (2, H, W) (or (N, 2, H,
+W) for batches of flows) where each entry corresponds to the horizontal and
+vertical displacement of each pixel from the first image to the second image.
+Note that the predicted flows are in “pixel” unit, they are not normalized
+w.r.t. the dimensions of the images.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">predicted_flows</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">list_of_flows</span></a><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;dtype = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">predicted_flows</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;shape = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">predicted_flows</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2"> = (N, 2, H, W)&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;min = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">predicted_flows</span></a><span class="o">.</span><span class="n">min</span><span class="p">()</span><span class="si">}</span><span class="s2">, max = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">predicted_flows</span></a><span class="o">.</span><span class="n">max</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>dtype = torch.float32
+shape = torch.Size([2, 2, 520, 960]) = (N, 2, H, W)
+min = -3.8997180461883545, max = 6.400400161743164
+</pre></div>
+</div>
+</section>
+<section id="visualizing-predicted-flows">
+<h2>Visualizing predicted flows<a class="headerlink" href="#visualizing-predicted-flows" title="Permalink to this heading">¶</a></h2>
+<p>Torchvision provides the <a class="reference internal" href="../../generated/torchvision.utils.flow_to_image.html#torchvision.utils.flow_to_image" title="torchvision.utils.flow_to_image"><code class="xref py py-func docutils literal notranslate"><span class="pre">flow_to_image()</span></code></a> utility to
+convert a flow into an RGB image. It also supports batches of flows.
+each “direction” in the flow will be mapped to a given RGB color. In the
+images below, pixels with similar colors are assumed by the model to be moving
+in similar directions. The model is properly able to predict the movement of
+the ball and the player. Note in particular the different predicted direction
+of the ball in the first image (going to the left) and in the second image
+(going up).</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">flow_to_image</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">flow_imgs</span></a> <span class="o">=</span> <span class="n">flow_to_image</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">predicted_flows</span></a><span class="p">)</span>
+
+<span class="c1"># The images have been mapped into [-1, 1] but for plotting we want them in [0, 1]</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a> <span class="o">=</span> <span class="p">[(</span><span class="n">img1</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="mi">2</span> <span class="k">for</span> <span class="n">img1</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">]</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">grid</span></a> <span class="o">=</span> <span class="p">[[</span><span class="n">img1</span><span class="p">,</span> <span class="n">flow_img</span><span class="p">]</span> <span class="k">for</span> <span class="p">(</span><span class="n">img1</span><span class="p">,</span> <span class="n">flow_img</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img1_batch</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">flow_imgs</span></a><span class="p">)]</span>
+<span class="n">plot</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">grid</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_optical_flow_002.png" srcset="../../_images/sphx_glr_plot_optical_flow_002.png" alt="plot optical flow" class = "sphx-glr-single-img"/></section>
+<section id="bonus-creating-gifs-of-predicted-flows">
+<h2>Bonus: Creating GIFs of predicted flows<a class="headerlink" href="#bonus-creating-gifs-of-predicted-flows" title="Permalink to this heading">¶</a></h2>
+<p>In the example above we have only shown the predicted flows of 2 pairs of
+frames. A fun way to apply the Optical Flow models is to run the model on an
+entire video, and create a new video from all the predicted flows. Below is a
+snippet that can get you started with this. We comment out the code, because
+this example is being rendered on a machine without a GPU, and it would take
+too long to run it.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># from torchvision.io import write_jpeg</span>
+<span class="c1"># for i, (img1, img2) in enumerate(zip(frames, frames[1:])):</span>
+<span class="c1">#     # Note: it would be faster to predict batches of flows instead of individual flows</span>
+<span class="c1">#     img1, img2 = preprocess(img1, img2)</span>
+
+<span class="c1">#     list_of_flows = model(img1.to(device), img2.to(device))</span>
+<span class="c1">#     predicted_flow = list_of_flows[-1][0]</span>
+<span class="c1">#     flow_img = flow_to_image(predicted_flow).to(&quot;cpu&quot;)</span>
+<span class="c1">#     output_folder = &quot;/tmp/&quot;  # Update this to the folder of your choice</span>
+<span class="c1">#     write_jpeg(flow_img, output_folder + f&quot;predicted_flow_{i}.jpg&quot;)</span>
+</pre></div>
+</div>
+<p>Once the .jpg flow images are saved, you can convert them into a video or a
+GIF using ffmpeg with e.g.:</p>
+<p>ffmpeg -f image2 -framerate 30 -i predicted_flow_%d.jpg -loop -1 flow.gif</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 10.096 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-others-plot-optical-flow-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/6acd2f50f8be085d0a8ac5874560f268/plot_optical_flow.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_optical_flow.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/8c66a0169c2c41937d8db4238842fe5b/plot_optical_flow.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_optical_flow.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_repurposing_annotations.html" class="btn btn-neutral float-right" title="Repurposing masks into bounding boxes" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="index.html" class="btn btn-neutral" title="Others" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Optical Flow: Predicting movement with the RAFT model</a><ul>
+<li><a class="reference internal" href="#reading-videos-using-torchvision">Reading Videos Using Torchvision</a></li>
+<li><a class="reference internal" href="#estimating-optical-flow-using-raft">Estimating Optical flow using RAFT</a></li>
+<li><a class="reference internal" href="#visualizing-predicted-flows">Visualizing predicted flows</a></li>
+<li><a class="reference internal" href="#bonus-creating-gifs-of-predicted-flows">Bonus: Creating GIFs of predicted flows</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/others/plot_repurposing_annotations.html b/0.16/auto_examples/others/plot_repurposing_annotations.html
new file mode 100644
index 00000000000..be6e095fe8d
--- /dev/null
+++ b/0.16/auto_examples/others/plot_repurposing_annotations.html
@@ -0,0 +1,944 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Repurposing masks into bounding boxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Torchscript support" href="plot_scripted_tensor_transforms.html" />
+    <link rel="prev" title="Optical Flow: Predicting movement with the RAFT model" href="plot_optical_flow.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Others</a> &gt;</li>
+        
+      <li>Repurposing masks into bounding boxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/others/plot_repurposing_annotations.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="repurposing-masks-into-bounding-boxes">
+<span id="sphx-glr-auto-examples-others-plot-repurposing-annotations-py"></span><h1>Repurposing masks into bounding boxes<a class="headerlink" href="#repurposing-masks-into-bounding-boxes" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_repurposing_annotations.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>The following example illustrates the operations available
+the <a class="reference internal" href="../../ops.html#ops"><span class="std std-ref">torchvision.ops</span></a> module for repurposing
+segmentation masks into object localization annotations for different tasks
+(e.g. transforming masks used by instance and panoptic segmentation
+methods into bounding boxes used by object detection methods).</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
+
+<span class="kn">import</span> <span class="nn">torchvision.transforms.functional</span> <span class="k">as</span> <span class="nn">F</span>
+
+
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ASSETS_DIRECTORY</span></a> <span class="o">=</span> <span class="s2">&quot;../assets&quot;</span>
+
+<a href="https://matplotlib.org/stable/api/matplotlib_configuration_api.html#matplotlib.RcParams" title="matplotlib.RcParams" class="sphx-glr-backref-module-matplotlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">plt</span><span class="o">.</span><span class="n">rcParams</span></a><span class="p">[</span><span class="s2">&quot;savefig.bbox&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;tight&quot;</span>
+
+
+<span class="k">def</span> <span class="nf">show</span><span class="p">(</span><span class="n">imgs</span><span class="p">):</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">imgs</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+        <span class="n">imgs</span> <span class="o">=</span> <span class="p">[</span><span class="n">imgs</span><span class="p">]</span>
+    <span class="n">fix</span><span class="p">,</span> <span class="n">axs</span> <span class="o">=</span> <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.subplots.html#matplotlib.pyplot.subplots" title="matplotlib.pyplot.subplots" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">subplots</span></a><span class="p">(</span><span class="n">ncols</span><span class="o">=</span><span class="nb">len</span><span class="p">(</span><span class="n">imgs</span><span class="p">),</span> <span class="n">squeeze</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">imgs</span><span class="p">):</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">to_pil_image</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span>
+        <span class="n">axs</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">imshow</span><span class="p">(</span><a href="https://numpy.org/doc/stable/reference/generated/numpy.asarray.html#numpy.asarray" title="numpy.asarray" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">asarray</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">))</span>
+        <span class="n">axs</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">set</span><span class="p">(</span><span class="n">xticklabels</span><span class="o">=</span><span class="p">[],</span> <span class="n">yticklabels</span><span class="o">=</span><span class="p">[],</span> <span class="n">xticks</span><span class="o">=</span><span class="p">[],</span> <span class="n">yticks</span><span class="o">=</span><span class="p">[])</span>
+</pre></div>
+</div>
+<section id="masks">
+<h2>Masks<a class="headerlink" href="#masks" title="Permalink to this heading">¶</a></h2>
+<p>In tasks like instance and panoptic segmentation, masks are commonly defined, and are defined by this package,
+as a multi-dimensional array (e.g. a NumPy array or a PyTorch tensor) with the following shape:</p>
+<blockquote>
+<div><p>(num_objects, height, width)</p>
+</div></blockquote>
+<p>Where num_objects is the number of annotated objects in the image. Each (height, width) object corresponds to exactly
+one object. For example, if your input image has the dimensions 224 x 224 and has four annotated objects the shape
+of your masks annotation has the following shape:</p>
+<blockquote>
+<div><p>(4, 224, 224).</p>
+</div></blockquote>
+<p>A nice property of masks is that they can be easily repurposed to be used in methods to solve a variety of object
+localization tasks.</p>
+</section>
+<section id="converting-masks-to-bounding-boxes">
+<h2>Converting Masks to Bounding Boxes<a class="headerlink" href="#converting-masks-to-bounding-boxes" title="Permalink to this heading">¶</a></h2>
+<p>For example, the <a class="reference internal" href="../../generated/torchvision.ops.masks_to_boxes.html#torchvision.ops.masks_to_boxes" title="torchvision.ops.masks_to_boxes"><code class="xref py py-func docutils literal notranslate"><span class="pre">masks_to_boxes()</span></code></a> operation can be used to
+transform masks into bounding boxes that can be
+used as input to detection models such as FasterRCNN and RetinaNet.
+We will take images and masks from the <a class="reference external" href="https://www.cis.upenn.edu/~jshi/ped_html/">PenFudan Dataset</a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_image</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img_path</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/os.path.html#os.path.join" title="os.path.join" class="sphx-glr-backref-module-os-path sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ASSETS_DIRECTORY</span></a><span class="p">,</span> <span class="s2">&quot;FudanPed00054.png&quot;</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask_path</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/os.path.html#os.path.join" title="os.path.join" class="sphx-glr-backref-module-os-path sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ASSETS_DIRECTORY</span></a><span class="p">,</span> <span class="s2">&quot;FudanPed00054_mask.png&quot;</span><span class="p">)</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img_path</span></a><span class="p">)</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask_path</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<p>Here the masks are represented as a PNG Image, with floating point values.
+Each pixel is encoded as different colors, with 0 being background.
+Notice that the spatial dimensions of image and mask match.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>torch.Size([1, 498, 533])
+torch.Size([3, 498, 533])
+tensor([[[0, 0, 0,  ..., 0, 0, 0],
+         [0, 0, 0,  ..., 0, 0, 0],
+         [0, 0, 0,  ..., 0, 0, 0],
+         ...,
+         [0, 0, 0,  ..., 0, 0, 0],
+         [0, 0, 0,  ..., 0, 0, 0],
+         [0, 0, 0,  ..., 0, 0, 0]]], dtype=torch.uint8)
+</pre></div>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># We get the unique colors, as these would be the object ids.</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.unique.html#torch.unique" title="torch.unique" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">unique</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a><span class="p">)</span>
+
+<span class="c1"># first id is the background, so remove it.</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a><span class="p">[</span><span class="mi">1</span><span class="p">:]</span>
+
+<span class="c1"># split the color-encoded mask into a set of boolean masks.</span>
+<span class="c1"># Note that this snippet would work as well if the masks were float values instead of ints.</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a> <span class="o">==</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+</pre></div>
+</div>
+<p>Now the masks are a boolean tensor.
+The first dimension in this case 3 and denotes the number of instances: there are 3 people in the image.
+The other two dimensions are height and width, which are equal to the dimensions of the image.
+For each instance, the boolean tensors represent if the particular pixel
+belongs to the segmentation mask of the image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>torch.Size([3, 498, 533])
+tensor([[[False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         ...,
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False]],
+
+        [[False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         ...,
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False]],
+
+        [[False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         ...,
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False],
+         [False, False, False,  ..., False, False, False]]])
+</pre></div>
+</div>
+<p>Let us visualize an image and plot its corresponding segmentation masks.
+We will use the <a class="reference internal" href="../../generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks" title="torchvision.utils.draw_segmentation_masks"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_segmentation_masks()</span></code></a> to draw the segmentation masks.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">draw_segmentation_masks</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">drawn_masks</span></a> <span class="o">=</span> <span class="p">[]</span>
+<span class="k">for</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a> <span class="ow">in</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="p">:</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">drawn_masks</span></a><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">draw_segmentation_masks</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.8</span><span class="p">,</span> <span class="n">colors</span><span class="o">=</span><span class="s2">&quot;blue&quot;</span><span class="p">))</span>
+
+<span class="n">show</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">drawn_masks</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_repurposing_annotations_001.png" srcset="../../_images/sphx_glr_plot_repurposing_annotations_001.png" alt="plot repurposing annotations" class = "sphx-glr-single-img"/><p>To convert the boolean masks into bounding boxes.
+We will use the <a class="reference internal" href="../../generated/torchvision.ops.masks_to_boxes.html#torchvision.ops.masks_to_boxes" title="torchvision.ops.masks_to_boxes"><code class="xref py py-func docutils literal notranslate"><span class="pre">masks_to_boxes()</span></code></a> from the torchvision.ops module
+It returns the boxes in <code class="docutils literal notranslate"><span class="pre">(xmin,</span> <span class="pre">ymin,</span> <span class="pre">xmax,</span> <span class="pre">ymax)</span></code> format.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.ops</span> <span class="kn">import</span> <span class="n">masks_to_boxes</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a> <span class="o">=</span> <span class="n">masks_to_boxes</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>torch.Size([3, 4])
+tensor([[ 96., 134., 181., 417.],
+        [286., 113., 357., 331.],
+        [363., 120., 436., 328.]])
+</pre></div>
+</div>
+<p>As the shape denotes, there are 3 boxes and in <code class="docutils literal notranslate"><span class="pre">(xmin,</span> <span class="pre">ymin,</span> <span class="pre">xmax,</span> <span class="pre">ymax)</span></code> format.
+These can be visualized very easily with <a class="reference internal" href="../../generated/torchvision.utils.draw_bounding_boxes.html#torchvision.utils.draw_bounding_boxes" title="torchvision.utils.draw_bounding_boxes"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_bounding_boxes()</span></code></a> utility
+provided in <a class="reference internal" href="../../utils.html#utils"><span class="std std-ref">torchvision.utils</span></a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">draw_bounding_boxes</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">drawn_boxes</span></a> <span class="o">=</span> <span class="n">draw_bounding_boxes</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a><span class="p">,</span> <span class="n">colors</span><span class="o">=</span><span class="s2">&quot;red&quot;</span><span class="p">)</span>
+<span class="n">show</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">drawn_boxes</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_repurposing_annotations_002.png" srcset="../../_images/sphx_glr_plot_repurposing_annotations_002.png" alt="plot repurposing annotations" class = "sphx-glr-single-img"/><p>These boxes can now directly be used by detection models in torchvision.
+Here is demo with a Faster R-CNN model loaded from
+<a class="reference internal" href="../../models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="torchvision.models.detection.fasterrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn()</span></code></a></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models.detection</span> <span class="kn">import</span> <span class="n">fasterrcnn_resnet50_fpn</span><span class="p">,</span> <span class="n">FasterRCNN_ResNet50_FPN_Weights</span>
+
+<span class="n">weights</span> <span class="o">=</span> <span class="n">FasterRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">fasterrcnn_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+
+<span class="n">tranforms</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <span class="n">tranforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a> <span class="o">=</span> <span class="p">{}</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a>
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.ones.html#torch.ones" title="torch.ones" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">ones</span></a><span class="p">((</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),),</span> <span class="n">dtype</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">int64</span></a><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">detection_outputs</span></a> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="p">[</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">])</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth&quot; to /root/.cache/torch/hub/checkpoints/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth
+torch.Size([3, 498, 533])
+</pre></div>
+</div>
+</section>
+<section id="converting-segmentation-dataset-to-detection-dataset">
+<h2>Converting Segmentation Dataset to Detection Dataset<a class="headerlink" href="#converting-segmentation-dataset-to-detection-dataset" title="Permalink to this heading">¶</a></h2>
+<p>With this utility it becomes very simple to convert a segmentation dataset to a detection dataset.
+With this we can now use a segmentation dataset to train a detection model.
+One can similarly convert panoptic dataset to detection dataset.
+Here is an example where we re-purpose the dataset from the
+<a class="reference external" href="https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html">PenFudan Detection Tutorial</a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">SegmentationToDetectionDataset</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.Dataset" title="torch.utils.data.Dataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">Dataset</span></a><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">,</span> <span class="n">transforms</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">root</span> <span class="o">=</span> <span class="n">root</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span>
+        <span class="c1"># load all image files, sorting them to</span>
+        <span class="c1"># ensure that they are aligned</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">imgs</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><a href="https://docs.python.org/3/library/os.html#os.listdir" title="os.listdir" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">listdir</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/os.path.html#os.path.join" title="os.path.join" class="sphx-glr-backref-module-os-path sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span></a><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;PNGImages&quot;</span><span class="p">))))</span>
+        <span class="bp">self</span><span class="o">.</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><a href="https://docs.python.org/3/library/os.html#os.listdir" title="os.listdir" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">listdir</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/os.path.html#os.path.join" title="os.path.join" class="sphx-glr-backref-module-os-path sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span></a><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;PedMasks&quot;</span><span class="p">))))</span>
+
+    <span class="k">def</span> <span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">idx</span><span class="p">):</span>
+        <span class="c1"># load images and masks</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img_path</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/os.path.html#os.path.join" title="os.path.join" class="sphx-glr-backref-module-os-path sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span></a><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;PNGImages&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">imgs</span><span class="p">[</span><span class="n">idx</span><span class="p">])</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask_path</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/os.path.html#os.path.join" title="os.path.join" class="sphx-glr-backref-module-os-path sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span></a><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">root</span><span class="p">,</span> <span class="s2">&quot;PedMasks&quot;</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="p">[</span><span class="n">idx</span><span class="p">])</span>
+
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img_path</span></a><span class="p">)</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask_path</span></a><span class="p">)</span>
+
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_image_dtype</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">float</span></a><span class="p">)</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">convert_image_dtype</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">float</span></a><span class="p">)</span>
+
+        <span class="c1"># We get the unique colors, as these would be the object ids.</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.unique.html#torch.unique" title="torch.unique" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">unique</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a><span class="p">)</span>
+
+        <span class="c1"># first id is the background, so remove it.</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a><span class="p">[</span><span class="mi">1</span><span class="p">:]</span>
+
+        <span class="c1"># split the color-encoded mask into a set of boolean masks.</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">mask</span></a> <span class="o">==</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">obj_ids</span></a><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a> <span class="o">=</span> <span class="n">masks_to_boxes</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="p">)</span>
+
+        <span class="c1"># there is only one class</span>
+        <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.ones.html#torch.ones" title="torch.ones" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">ones</span></a><span class="p">((</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">masks</span></a><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],),</span> <span class="n">dtype</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">int64</span></a><span class="p">)</span>
+
+        <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a> <span class="o">=</span> <span class="p">{}</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a>
+        <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">)</span>
+
+        <span class="k">return</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a>
+</pre></div>
+</div>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 2.507 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-others-plot-repurposing-annotations-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/6159c523793cb1afbb503259bd141f91/plot_repurposing_annotations.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_repurposing_annotations.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/2753c5fd8349a28b023c919d8ddf3573/plot_repurposing_annotations.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_repurposing_annotations.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_scripted_tensor_transforms.html" class="btn btn-neutral float-right" title="Torchscript support" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_optical_flow.html" class="btn btn-neutral" title="Optical Flow: Predicting movement with the RAFT model" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Repurposing masks into bounding boxes</a><ul>
+<li><a class="reference internal" href="#masks">Masks</a></li>
+<li><a class="reference internal" href="#converting-masks-to-bounding-boxes">Converting Masks to Bounding Boxes</a></li>
+<li><a class="reference internal" href="#converting-segmentation-dataset-to-detection-dataset">Converting Segmentation Dataset to Detection Dataset</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/others/plot_scripted_tensor_transforms.html b/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
new file mode 100644
index 00000000000..fe3a5825cfc
--- /dev/null
+++ b/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
@@ -0,0 +1,823 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Torchscript support &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Video API" href="plot_video_api.html" />
+    <link rel="prev" title="Repurposing masks into bounding boxes" href="plot_repurposing_annotations.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Others</a> &gt;</li>
+        
+      <li>Torchscript support</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/others/plot_scripted_tensor_transforms.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="torchscript-support">
+<span id="sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"></span><h1>Torchscript support<a class="headerlink" href="#torchscript-support" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_scripted_tensor_transforms.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>This example illustrates <a class="reference external" href="https://pytorch.org/docs/stable/jit.html">torchscript</a> support of the torchvision
+<a class="reference internal" href="../../transforms.html#transforms"><span class="std std-ref">transforms</span></a> on Tensor images.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a>
+
+<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>
+
+<span class="kn">import</span> <span class="nn">torchvision.transforms</span> <span class="k">as</span> <span class="nn">v1</span>
+<span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_image</span>
+
+<a href="https://matplotlib.org/stable/api/matplotlib_configuration_api.html#matplotlib.RcParams" title="matplotlib.RcParams" class="sphx-glr-backref-module-matplotlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">plt</span><span class="o">.</span><span class="n">rcParams</span></a><span class="p">[</span><span class="s2">&quot;savefig.bbox&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;tight&#39;</span>
+<a href="https://pytorch.org/docs/stable/generated/torch.manual_seed.html#torch.manual_seed" title="torch.manual_seed" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">manual_seed</span></a><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+<span class="c1"># If you&#39;re trying to run that on collab, you can download the assets and the</span>
+<span class="c1"># helpers from https://github.com/pytorch/vision/tree/main/gallery/</span>
+<span class="kn">import</span> <span class="nn">sys</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sys</span><span class="o">.</span><span class="n">path</span></a> <span class="o">+=</span> <span class="p">[</span><span class="s2">&quot;../transforms&quot;</span><span class="p">]</span>
+<span class="kn">from</span> <span class="nn">helpers</span> <span class="kn">import</span> <span class="n">plot</span>
+<a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ASSETS_PATH</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><span class="s1">&#39;../assets&#39;</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Most transforms support torchscript. For composing transforms, we use
+<a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.Sequential</span></code></a> instead of
+<a class="reference internal" href="../../generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose" title="torchvision.transforms.v2.Compose"><code class="xref py py-class docutils literal notranslate"><span class="pre">Compose</span></code></a>:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ASSETS_PATH</span></a> <span class="o">/</span> <span class="s1">&#39;dog1.jpg&#39;</span><span class="p">))</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog2</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ASSETS_PATH</span></a> <span class="o">/</span> <span class="s1">&#39;dog2.jpg&#39;</span><span class="p">))</span>
+
+<a href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="torch.nn.Sequential" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="torch.nn.Sequential" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span></a><span class="p">(</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v1</span><span class="o">.</span><span class="n">RandomCrop</span></a><span class="p">(</span><span class="mi">224</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v1</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mf">0.3</span><span class="p">),</span>
+<span class="p">)</span>
+
+<span class="n">scripted_transforms</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.jit.script.html#torch.jit.script" title="torch.jit.script" class="sphx-glr-backref-module-torch-jit sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">script</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="torch.nn.Sequential" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a><span class="p">)</span>
+
+<span class="n">plot</span><span class="p">([</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1</span></a><span class="p">,</span> <span class="n">scripted_transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1</span></a><span class="p">),</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog2</span></a><span class="p">,</span> <span class="n">scripted_transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog2</span></a><span class="p">)])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_scripted_tensor_transforms_001.png" srcset="../../_images/sphx_glr_plot_scripted_tensor_transforms_001.png" alt="plot scripted tensor transforms" class = "sphx-glr-single-img"/><div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>Above we have used transforms from the <code class="docutils literal notranslate"><span class="pre">torchvision.transforms</span></code>
+namespace, i.e. the “v1” transforms. The v2 transforms from the
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> namespace are the <a class="reference internal" href="../../transforms.html#v1-or-v2"><span class="std std-ref">recommended</span></a> way to use transforms in your code.</p>
+<p>The v2 transforms also support torchscript, but if you call
+<code class="docutils literal notranslate"><span class="pre">torch.jit.script()</span></code> on a v2 <strong>class</strong> transform, you’ll actually end up
+with its (scripted) v1 equivalent.  This may lead to slightly different
+results between the scripted and eager executions due to implementation
+differences between v1 and v2.</p>
+<p>If you really need torchscript support for the v2 transforms, <strong>we
+recommend scripting the functionals</strong> from the
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2.functional</span></code> namespace to avoid surprises.</p>
+</div>
+<p>Below we now show how to combine image transformations and a model forward
+pass, while using <code class="docutils literal notranslate"><span class="pre">torch.jit.script</span></code> to obtain a single scripted module.</p>
+<p>Let’s define a <code class="docutils literal notranslate"><span class="pre">Predictor</span></code> module that transforms the input tensor and then
+applies an ImageNet model on it.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models</span> <span class="kn">import</span> <span class="n">resnet18</span><span class="p">,</span> <span class="n">ResNet18_Weights</span>
+
+
+<span class="k">class</span> <span class="nc">Predictor</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">nn</span><span class="o">.</span><span class="n">Module</span></a><span class="p">):</span>
+
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet18_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">resnet18</span> <span class="o">=</span> <span class="n">resnet18</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="torch.nn.Sequential" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="torch.nn.Sequential" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a><span class="p">(</span><span class="n">antialias</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span></a><span class="p">)</span> <span class="o">-&gt;</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span></a><span class="p">:</span>
+        <span class="k">with</span> <a href="https://pytorch.org/docs/stable/generated/torch.no_grad.html#torch.no_grad" title="torch.no_grad" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span></a><span class="p">():</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="torch.nn.Sequential" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+            <span class="n">y_pred</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">resnet18</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">y_pred</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Now, let’s define scripted and non-scripted instances of <code class="docutils literal notranslate"><span class="pre">Predictor</span></code> and
+apply it on multiple tensor images of the same size</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a> <span class="o">=</span> <span class="s2">&quot;cuda&quot;</span> <span class="k">if</span> <a href="https://pytorch.org/docs/stable/generated/torch.cuda.is_available.html#torch.cuda.is_available" title="torch.cuda.is_available" class="sphx-glr-backref-module-torch-cuda sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">is_available</span></a><span class="p">()</span> <span class="k">else</span> <span class="s2">&quot;cpu&quot;</span>
+
+<span class="n">predictor</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">Predictor</span></a><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a><span class="p">)</span>
+<span class="n">scripted_predictor</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.jit.script.html#torch.jit.script" title="torch.jit.script" class="sphx-glr-backref-module-torch-jit sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">script</span></a><span class="p">(</span><span class="n">predictor</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a><span class="p">)</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.stack.html#torch.stack" title="torch.stack" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">stack</span></a><span class="p">([</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog2</span></a><span class="p">])</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">device</span></a><span class="p">)</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res</span></a> <span class="o">=</span> <span class="n">predictor</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">)</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res_scripted</span></a> <span class="o">=</span> <span class="n">scripted_predictor</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/resnet18-f37072fd.pth&quot; to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
+</pre></div>
+</div>
+<p>We can verify that the prediction of the scripted and non-scripted models are
+the same:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">json</span>
+
+<span class="k">with</span> <span class="nb">open</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><span class="s1">&#39;../assets&#39;</span><span class="p">)</span> <span class="o">/</span> <span class="s1">&#39;imagenet_class_index.json&#39;</span><span class="p">)</span> <span class="k">as</span> <a href="https://docs.python.org/3/library/io.html#io.TextIOWrapper" title="io.TextIOWrapper" class="sphx-glr-backref-module-io sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels_file</span></a><span class="p">:</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/json.html#json.load" title="json.load" class="sphx-glr-backref-module-json sphx-glr-backref-type-py-function"><span class="n">json</span><span class="o">.</span><span class="n">load</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/io.html#io.TextIOWrapper" title="io.TextIOWrapper" class="sphx-glr-backref-module-io sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels_file</span></a><span class="p">)</span>
+
+<span class="k">for</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="p">,</span> <span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">pred</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">pred_scripted</span></a><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res_scripted</span></a><span class="p">)):</span>
+    <span class="k">assert</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">pred</span></a> <span class="o">==</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">pred_scripted</span></a>
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Prediction for Dog </span><span class="si">{</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="w"> </span><span class="o">+</span><span class="w"> </span><span class="mi">1</span><span class="si">}</span><span class="s2">: </span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a><span class="p">[</span><span class="nb">str</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">pred</span></a><span class="o">.</span><span class="n">item</span><span class="p">())]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Prediction for Dog 1: [&#39;n02113023&#39;, &#39;Pembroke&#39;]
+Prediction for Dog 2: [&#39;n02106662&#39;, &#39;German_shepherd&#39;]
+</pre></div>
+</div>
+<p>Since the model is scripted, it can be easily dumped on disk and re-used</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">tempfile</span>
+
+<span class="k">with</span> <a href="https://docs.python.org/3/library/tempfile.html#tempfile.NamedTemporaryFile" title="tempfile.NamedTemporaryFile" class="sphx-glr-backref-module-tempfile sphx-glr-backref-type-py-function"><span class="n">tempfile</span><span class="o">.</span><span class="n">NamedTemporaryFile</span></a><span class="p">()</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.jit.ScriptModule.html#torch.jit.ScriptModule.save" title="torch.jit.ScriptModule.save" class="sphx-glr-backref-module-torch-jit sphx-glr-backref-type-py-method"><span class="n">scripted_predictor</span><span class="o">.</span><span class="n">save</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">f</span><span class="o">.</span><span class="n">name</span></a><span class="p">)</span>
+
+    <span class="n">dumped_scripted_predictor</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.jit.load.html#torch.jit.load" title="torch.jit.load" class="sphx-glr-backref-module-torch-jit sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">load</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">f</span><span class="o">.</span><span class="n">name</span></a><span class="p">)</span>
+    <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res_scripted_dumped</span></a> <span class="o">=</span> <span class="n">dumped_scripted_predictor</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">)</span>
+<span class="k">assert</span> <span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res_scripted_dumped</span></a> <span class="o">==</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res_scripted</span></a><span class="p">)</span><span class="o">.</span><span class="n">all</span><span class="p">()</span>
+</pre></div>
+</div>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 1.683 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-others-plot-scripted-tensor-transforms-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/1c7d3f45724a1171e328ce5cdd7bc79a/plot_scripted_tensor_transforms.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_scripted_tensor_transforms.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/4333051365715be174a783856419631f/plot_scripted_tensor_transforms.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_scripted_tensor_transforms.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_video_api.html" class="btn btn-neutral float-right" title="Video API" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_repurposing_annotations.html" class="btn btn-neutral" title="Repurposing masks into bounding boxes" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Torchscript support</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/others/plot_video_api.html b/0.16/auto_examples/others/plot_video_api.html
new file mode 100644
index 00000000000..72e59903f59
--- /dev/null
+++ b/0.16/auto_examples/others/plot_video_api.html
@@ -0,0 +1,1227 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Video API &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Visualization utilities" href="plot_visualization_utils.html" />
+    <link rel="prev" title="Torchscript support" href="plot_scripted_tensor_transforms.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Others</a> &gt;</li>
+        
+      <li>Video API</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/others/plot_video_api.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="video-api">
+<span id="sphx-glr-auto-examples-others-plot-video-api-py"></span><h1>Video API<a class="headerlink" href="#video-api" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_video_api.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-others-plot-video-api-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>This example illustrates some of the APIs that torchvision offers for
+videos, together with the examples on how to build datasets and more.</p>
+<section id="introduction-building-a-new-video-object-and-examining-the-properties">
+<h2>1. Introduction: building a new video object and examining the properties<a class="headerlink" href="#introduction-building-a-new-video-object-and-examining-the-properties" title="Permalink to this heading">¶</a></h2>
+<p>First we select a video to test the object out. For the sake of argument
+we’re using one from kinetics400 dataset.
+To create it, we need to define the path and the stream we want to use.</p>
+<p>Chosen video statistics:</p>
+<ul class="simple">
+<li><dl class="simple">
+<dt>WUzgd7C1pWA.mp4</dt><dd><ul>
+<li><dl class="simple">
+<dt>source:</dt><dd><ul>
+<li><p>kinetics-400</p></li>
+</ul>
+</dd>
+</dl>
+</li>
+<li><dl class="simple">
+<dt>video:</dt><dd><ul>
+<li><p>H-264</p></li>
+<li><p>MPEG-4 AVC (part 10) (avc1)</p></li>
+<li><p>fps: 29.97</p></li>
+</ul>
+</dd>
+</dl>
+</li>
+<li><dl class="simple">
+<dt>audio:</dt><dd><ul>
+<li><p>MPEG AAC audio (mp4a)</p></li>
+<li><p>sample rate: 48K Hz</p></li>
+</ul>
+</dd>
+</dl>
+</li>
+</ul>
+</dd>
+</dl>
+</li>
+</ul>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torchvision</span>
+<span class="kn">from</span> <span class="nn">torchvision.datasets.utils</span> <span class="kn">import</span> <span class="n">download_url</span>
+<span class="n">torchvision</span><span class="o">.</span><span class="n">set_video_backend</span><span class="p">(</span><span class="s2">&quot;video_reader&quot;</span><span class="p">)</span>
+
+<span class="c1"># Download the sample video</span>
+<span class="n">download_url</span><span class="p">(</span>
+    <span class="s2">&quot;https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;.&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;WUzgd7C1pWA.mp4&quot;</span>
+<span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">video_path</span></a> <span class="o">=</span> <span class="s2">&quot;./WUzgd7C1pWA.mp4&quot;</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/WUzgd7C1pWA.mp4 to ./WUzgd7C1pWA.mp4
+
+3.7%
+7.4%
+11.1%
+14.7%
+18.4%
+22.1%
+25.8%
+29.5%
+33.2%
+36.8%
+40.5%
+44.2%
+47.9%
+51.6%
+55.3%
+58.9%
+62.6%
+66.3%
+70.0%
+73.7%
+77.4%
+81.0%
+84.7%
+88.4%
+92.1%
+95.8%
+99.5%
+100.0%
+</pre></div>
+</div>
+<p>Streams are defined in a similar fashion as torch devices. We encode them as strings in a form
+of <code class="docutils literal notranslate"><span class="pre">stream_type:stream_id</span></code> where <code class="docutils literal notranslate"><span class="pre">stream_type</span></code> is a string and <code class="docutils literal notranslate"><span class="pre">stream_id</span></code> a long int.
+The constructor accepts passing a <code class="docutils literal notranslate"><span class="pre">stream_type</span></code> only, in which case the stream is auto-discovered.
+Firstly, let’s get the metadata for our particular video:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">stream</span></a> <span class="o">=</span> <span class="s2">&quot;video&quot;</span>
+<span class="n">video</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">io</span><span class="o">.</span><span class="n">VideoReader</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">video_path</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">stream</span></a><span class="p">)</span>
+<span class="n">video</span><span class="o">.</span><span class="n">get_metadata</span><span class="p">()</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;video&#39;: {&#39;duration&#39;: [10.9109], &#39;fps&#39;: [29.97002997002997]}, &#39;audio&#39;: {&#39;duration&#39;: [10.9], &#39;framerate&#39;: [48000.0]}, &#39;subtitles&#39;: {&#39;duration&#39;: []}, &#39;cc&#39;: {&#39;duration&#39;: []}}
+</pre></div>
+</div>
+<p>Here we can see that video has two streams - a video and an audio stream.
+Currently available stream types include [‘video’, ‘audio’].
+Each descriptor consists of two parts: stream type (e.g. ‘video’) and a unique stream id
+(which are determined by video encoding).
+In this way, if the video container contains multiple streams of the same type,
+users can access the one they want.
+If only stream type is passed, the decoder auto-detects first stream of that type and returns it.</p>
+<p>Let’s read all the frames from the video stream. By default, the return value of
+<code class="docutils literal notranslate"><span class="pre">next(video_reader)</span></code> is a dict containing the following fields.</p>
+<p>The return fields are:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">data</span></code>: containing a torch.tensor</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">pts</span></code>: containing a float timestamp of this particular frame</p></li>
+</ul>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">metadata</span></a> <span class="o">=</span> <span class="n">video</span><span class="o">.</span><span class="n">get_metadata</span><span class="p">()</span>
+<span class="n">video</span><span class="o">.</span><span class="n">set_current_stream</span><span class="p">(</span><span class="s2">&quot;audio&quot;</span><span class="p">)</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># we are going to save the frames here.</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ptss</span></a> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># pts is a presentation timestamp in seconds (float) of each frame</span>
+<span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a> <span class="ow">in</span> <span class="n">video</span><span class="p">:</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ptss</span></a><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">])</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;PTS for first five frames &quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ptss</span></a><span class="p">[:</span><span class="mi">5</span><span class="p">])</span>
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Total number of frames: &quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">))</span>
+<a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">approx_nf</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">metadata</span></a><span class="p">[</span><span class="s1">&#39;audio&#39;</span><span class="p">][</span><span class="s1">&#39;duration&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">*</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">metadata</span></a><span class="p">[</span><span class="s1">&#39;audio&#39;</span><span class="p">][</span><span class="s1">&#39;framerate&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Approx total number of datapoints we can expect: &quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">approx_nf</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Read data size: &quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">))</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>PTS for first five frames  [0.0, 0.021332999999999998, 0.042667, 0.064, 0.08533299999999999]
+Total number of frames:  511
+Approx total number of datapoints we can expect:  523200.0
+Read data size:  523264
+</pre></div>
+</div>
+<p>But what if we only want to read certain time segment of the video?
+That can be done easily using the combination of our <code class="docutils literal notranslate"><span class="pre">seek</span></code> function, and the fact that each call
+to next returns the presentation timestamp of the returned frame in seconds.</p>
+<p>Given that our implementation relies on python iterators,
+we can leverage itertools to simplify the process and make it more pythonic.</p>
+<p>For example, if we wanted to read ten frames from second second:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">itertools</span>
+<span class="n">video</span><span class="o">.</span><span class="n">set_current_stream</span><span class="p">(</span><span class="s2">&quot;video&quot;</span><span class="p">)</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># we are going to save the frames here.</span>
+
+<span class="c1"># We seek into a second second of the video and use islice to get 10 frames since</span>
+<span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">pts</span></a> <span class="ow">in</span> <a href="https://docs.python.org/3/library/itertools.html#itertools.islice" title="itertools.islice" class="sphx-glr-backref-module-itertools sphx-glr-backref-type-py-function"><span class="n">itertools</span><span class="o">.</span><span class="n">islice</span></a><span class="p">(</span><span class="n">video</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="mi">10</span><span class="p">):</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">)</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Total number of frames: &quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">))</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Total number of frames:  10
+</pre></div>
+</div>
+<p>Or if we wanted to read from 2nd to 5th second,
+We seek into a second second of the video,
+then we utilize the itertools takewhile to get the
+correct number of frames:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">video</span><span class="o">.</span><span class="n">set_current_stream</span><span class="p">(</span><span class="s2">&quot;video&quot;</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># we are going to save the frames here.</span>
+<span class="n">video</span> <span class="o">=</span> <span class="n">video</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+
+<span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a> <span class="ow">in</span> <a href="https://docs.python.org/3/library/itertools.html#itertools.takewhile" title="itertools.takewhile" class="sphx-glr-backref-module-itertools sphx-glr-backref-type-py-function"><span class="n">itertools</span><span class="o">.</span><span class="n">takewhile</span></a><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="mi">5</span><span class="p">,</span> <span class="n">video</span><span class="p">):</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Total number of frames: &quot;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">))</span>
+<a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">approx_nf</span></a> <span class="o">=</span> <span class="p">(</span><span class="mi">5</span> <span class="o">-</span> <span class="mi">2</span><span class="p">)</span> <span class="o">*</span> <span class="n">video</span><span class="o">.</span><span class="n">get_metadata</span><span class="p">()[</span><span class="s1">&#39;video&#39;</span><span class="p">][</span><span class="s1">&#39;fps&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;We can expect approx: &quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">approx_nf</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Tensor size: &quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Total number of frames:  90
+We can expect approx:  89.91008991008991
+Tensor size:  torch.Size([3, 256, 340])
+</pre></div>
+</div>
+</section>
+<section id="building-a-sample-read-video-function">
+<h2>2. Building a sample read_video function<a class="headerlink" href="#building-a-sample-read-video-function" title="Permalink to this heading">¶</a></h2>
+<p>We can utilize the methods above to build the read video function that follows
+the same API to the existing <code class="docutils literal notranslate"><span class="pre">read_video</span></code> function.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">example_read_video</span><span class="p">(</span><span class="n">video_object</span><span class="p">,</span> <span class="n">start</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">end</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">read_video</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">read_audio</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
+    <span class="k">if</span> <span class="n">end</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">end</span> <span class="o">=</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">end</span> <span class="o">&lt;</span> <span class="n">start</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="s2">&quot;end time should be larger than start time, got &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;start time=</span><span class="si">{</span><span class="n">start</span><span class="si">}</span><span class="s2"> and end time=</span><span class="si">{</span><span class="n">end</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+    <span class="n">video_frames</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.empty.html#torch.empty" title="torch.empty" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">empty</span></a><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">video_pts</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">if</span> <span class="n">read_video</span><span class="p">:</span>
+        <span class="n">video_object</span><span class="o">.</span><span class="n">set_current_stream</span><span class="p">(</span><span class="s2">&quot;video&quot;</span><span class="p">)</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a> <span class="ow">in</span> <a href="https://docs.python.org/3/library/itertools.html#itertools.takewhile" title="itertools.takewhile" class="sphx-glr-backref-module-itertools sphx-glr-backref-type-py-function"><span class="n">itertools</span><span class="o">.</span><span class="n">takewhile</span></a><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">end</span><span class="p">,</span> <span class="n">video_object</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="n">start</span><span class="p">)):</span>
+            <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+            <span class="n">video_pts</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">video_frames</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.stack.html#torch.stack" title="torch.stack" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">stack</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="n">audio_frames</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.empty.html#torch.empty" title="torch.empty" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">empty</span></a><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">audio_pts</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">if</span> <span class="n">read_audio</span><span class="p">:</span>
+        <span class="n">video_object</span><span class="o">.</span><span class="n">set_current_stream</span><span class="p">(</span><span class="s2">&quot;audio&quot;</span><span class="p">)</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a> <span class="ow">in</span> <a href="https://docs.python.org/3/library/itertools.html#itertools.takewhile" title="itertools.takewhile" class="sphx-glr-backref-module-itertools sphx-glr-backref-type-py-function"><span class="n">itertools</span><span class="o">.</span><span class="n">takewhile</span></a><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="n">end</span><span class="p">,</span> <span class="n">video_object</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="n">start</span><span class="p">)):</span>
+            <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+            <span class="n">audio_pts</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">audio_frames</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.cat.html#torch.cat" title="torch.cat" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">cat</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frames</span></a><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">video_frames</span><span class="p">,</span> <span class="n">audio_frames</span><span class="p">,</span> <span class="p">(</span><span class="n">video_pts</span><span class="p">,</span> <span class="n">audio_pts</span><span class="p">),</span> <span class="n">video_object</span><span class="o">.</span><span class="n">get_metadata</span><span class="p">()</span>
+
+
+<span class="c1"># Total number of frames should be 327 for video and 523264 datapoints for audio</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">vf</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">af</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">info</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">meta</span></a> <span class="o">=</span> <span class="n">example_read_video</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">vf</span></a><span class="o">.</span><span class="n">size</span><span class="p">(),</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">af</span></a><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>torch.Size([327, 3, 256, 340]) torch.Size([523264, 1])
+</pre></div>
+</div>
+</section>
+<section id="building-an-example-randomly-sampled-dataset-can-be-applied-to-training-dataset-of-kinetics400">
+<h2>3. Building an example randomly sampled dataset (can be applied to training dataset of kinetics400)<a class="headerlink" href="#building-an-example-randomly-sampled-dataset-can-be-applied-to-training-dataset-of-kinetics400" title="Permalink to this heading">¶</a></h2>
+<p>Cool, so now we can use the same principle to make the sample dataset.
+We suggest trying out iterable dataset for this purpose.
+Here, we are going to build an example dataset that reads randomly selected 10 frames of video.</p>
+<p>Make sample dataset</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">os</span>
+<a href="https://docs.python.org/3/library/os.html#os.makedirs" title="os.makedirs" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">makedirs</span></a><span class="p">(</span><span class="s2">&quot;./dataset&quot;</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/os.html#os.makedirs" title="os.makedirs" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">makedirs</span></a><span class="p">(</span><span class="s2">&quot;./dataset/1&quot;</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/os.html#os.makedirs" title="os.makedirs" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">makedirs</span></a><span class="p">(</span><span class="s2">&quot;./dataset/2&quot;</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Download the videos</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.datasets.utils</span> <span class="kn">import</span> <span class="n">download_url</span>
+<span class="n">download_url</span><span class="p">(</span>
+    <span class="s2">&quot;https://github.com/pytorch/vision/blob/main/test/assets/videos/WUzgd7C1pWA.mp4?raw=true&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;./dataset/1&quot;</span><span class="p">,</span> <span class="s2">&quot;WUzgd7C1pWA.mp4&quot;</span>
+<span class="p">)</span>
+<span class="n">download_url</span><span class="p">(</span>
+    <span class="s2">&quot;https://github.com/pytorch/vision/blob/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi?raw=true&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;./dataset/1&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RATRACE_wave_f_nm_np1_fr_goo_37.avi&quot;</span>
+<span class="p">)</span>
+<span class="n">download_url</span><span class="p">(</span>
+    <span class="s2">&quot;https://github.com/pytorch/vision/blob/main/test/assets/videos/SOX5yA1l24A.mp4?raw=true&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;./dataset/2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;SOX5yA1l24A.mp4&quot;</span>
+<span class="p">)</span>
+<span class="n">download_url</span><span class="p">(</span>
+    <span class="s2">&quot;https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi?raw=true&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;./dataset/2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;v_SoccerJuggling_g23_c01.avi&quot;</span>
+<span class="p">)</span>
+<span class="n">download_url</span><span class="p">(</span>
+    <span class="s2">&quot;https://github.com/pytorch/vision/blob/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi?raw=true&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;./dataset/2&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;v_SoccerJuggling_g24_c01.avi&quot;</span>
+<span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/WUzgd7C1pWA.mp4 to ./dataset/1/WUzgd7C1pWA.mp4
+
+3.7%
+7.4%
+11.1%
+14.7%
+18.4%
+22.1%
+25.8%
+29.5%
+33.2%
+36.8%
+40.5%
+44.2%
+47.9%
+51.6%
+55.3%
+58.9%
+62.6%
+66.3%
+70.0%
+73.7%
+77.4%
+81.0%
+84.7%
+88.4%
+92.1%
+95.8%
+99.5%
+100.0%
+Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/RATRACE_wave_f_nm_np1_fr_goo_37.avi to ./dataset/1/RATRACE_wave_f_nm_np1_fr_goo_37.avi
+
+12.4%
+24.9%
+37.3%
+49.7%
+62.1%
+74.6%
+87.0%
+99.4%
+100.0%
+Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/SOX5yA1l24A.mp4 to ./dataset/2/SOX5yA1l24A.mp4
+
+5.8%
+11.7%
+17.5%
+23.4%
+29.2%
+35.1%
+40.9%
+46.8%
+52.6%
+58.5%
+64.3%
+70.2%
+76.0%
+81.9%
+87.7%
+93.6%
+99.4%
+100.0%
+Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/v_SoccerJuggling_g23_c01.avi to ./dataset/2/v_SoccerJuggling_g23_c01.avi
+
+6.4%
+12.9%
+19.3%
+25.8%
+32.2%
+38.7%
+45.1%
+51.6%
+58.0%
+64.5%
+70.9%
+77.3%
+83.8%
+90.2%
+96.7%
+100.0%
+Downloading https://raw.githubusercontent.com/pytorch/vision/main/test/assets/videos/v_SoccerJuggling_g24_c01.avi to ./dataset/2/v_SoccerJuggling_g24_c01.avi
+
+5.3%
+10.5%
+15.8%
+21.0%
+26.3%
+31.6%
+36.8%
+42.1%
+47.3%
+52.6%
+57.9%
+63.1%
+68.4%
+73.6%
+78.9%
+84.2%
+89.4%
+94.7%
+99.9%
+100.0%
+</pre></div>
+</div>
+<p>Housekeeping and utilities</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">random</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.datasets.folder</span> <span class="kn">import</span> <span class="n">make_dataset</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a> <span class="k">as</span> <span class="n">t</span>
+
+
+<span class="k">def</span> <span class="nf">_find_classes</span><span class="p">(</span><span class="nb">dir</span><span class="p">):</span>
+    <span class="n">classes</span> <span class="o">=</span> <span class="p">[</span><span class="n">d</span><span class="o">.</span><span class="n">name</span> <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/os.html#os.scandir" title="os.scandir" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">scandir</span></a><span class="p">(</span><span class="nb">dir</span><span class="p">)</span> <span class="k">if</span> <span class="n">d</span><span class="o">.</span><span class="n">is_dir</span><span class="p">()]</span>
+    <span class="n">classes</span><span class="o">.</span><span class="n">sort</span><span class="p">()</span>
+    <span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="n">cls_name</span><span class="p">:</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a> <span class="k">for</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="p">,</span> <span class="n">cls_name</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">classes</span><span class="p">)}</span>
+    <span class="k">return</span> <span class="n">classes</span><span class="p">,</span> <span class="n">class_to_idx</span>
+
+
+<span class="k">def</span> <span class="nf">get_samples</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">extensions</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;.mp4&quot;</span><span class="p">,</span> <span class="s2">&quot;.avi&quot;</span><span class="p">)):</span>
+    <span class="n">_</span><span class="p">,</span> <span class="n">class_to_idx</span> <span class="o">=</span> <span class="n">_find_classes</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">make_dataset</span><span class="p">(</span><span class="n">root</span><span class="p">,</span> <span class="n">class_to_idx</span><span class="p">,</span> <span class="n">extensions</span><span class="o">=</span><span class="n">extensions</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>We are going to define the dataset and some basic arguments.
+We assume the structure of the FolderDataset, and add the following parameters:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">clip_len</span></code>: length of a clip in frames</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">frame_transform</span></code>: transform for every frame individually</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">video_transform</span></code>: transform on a video sequence</p></li>
+</ul>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>We actually add epoch size as using <code class="xref py py-func docutils literal notranslate"><span class="pre">IterableDataset()</span></code>
+class allows us to naturally oversample clips or images from each video if needed.</p>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">RandomDataset</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.IterableDataset" title="torch.utils.data.IterableDataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">IterableDataset</span></a><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">root</span><span class="p">,</span> <span class="n">epoch_size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">frame_transform</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">video_transform</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">clip_len</span><span class="o">=</span><span class="mi">16</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.IterableDataset" title="torch.utils.data.IterableDataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">RandomDataset</span></a><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">samples</span> <span class="o">=</span> <span class="n">get_samples</span><span class="p">(</span><span class="n">root</span><span class="p">)</span>
+
+        <span class="c1"># Allow for temporal jittering</span>
+        <span class="k">if</span> <span class="n">epoch_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">epoch_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">epoch_size</span> <span class="o">=</span> <span class="n">epoch_size</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">clip_len</span> <span class="o">=</span> <span class="n">clip_len</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">frame_transform</span> <span class="o">=</span> <span class="n">frame_transform</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">video_transform</span> <span class="o">=</span> <span class="n">video_transform</span>
+
+    <span class="k">def</span> <span class="fm">__iter__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">epoch_size</span><span class="p">):</span>
+            <span class="c1"># Get random sample</span>
+            <span class="n">path</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <a href="https://docs.python.org/3/library/random.html#random.choice" title="random.choice" class="sphx-glr-backref-module-random sphx-glr-backref-type-py-function"><span class="n">random</span><span class="o">.</span><span class="n">choice</span></a><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">samples</span><span class="p">)</span>
+            <span class="c1"># Get video object</span>
+            <span class="n">vid</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">io</span><span class="o">.</span><span class="n">VideoReader</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="s2">&quot;video&quot;</span><span class="p">)</span>
+            <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">metadata</span></a> <span class="o">=</span> <span class="n">vid</span><span class="o">.</span><span class="n">get_metadata</span><span class="p">()</span>
+            <span class="n">video_frames</span> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># video frame buffer</span>
+
+            <span class="c1"># Seek and return frames</span>
+            <span class="n">max_seek</span> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">metadata</span></a><span class="p">[</span><span class="s2">&quot;video&quot;</span><span class="p">][</span><span class="s1">&#39;duration&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">clip_len</span> <span class="o">/</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">metadata</span></a><span class="p">[</span><span class="s2">&quot;video&quot;</span><span class="p">][</span><span class="s1">&#39;fps&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+            <span class="n">start</span> <span class="o">=</span> <a href="https://docs.python.org/3/library/random.html#random.uniform" title="random.uniform" class="sphx-glr-backref-module-random sphx-glr-backref-type-py-function"><span class="n">random</span><span class="o">.</span><span class="n">uniform</span></a><span class="p">(</span><span class="mf">0.</span><span class="p">,</span> <span class="n">max_seek</span><span class="p">)</span>
+            <span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a> <span class="ow">in</span> <a href="https://docs.python.org/3/library/itertools.html#itertools.islice" title="itertools.islice" class="sphx-glr-backref-module-itertools sphx-glr-backref-type-py-function"><span class="n">itertools</span><span class="o">.</span><span class="n">islice</span></a><span class="p">(</span><span class="n">vid</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="n">start</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">clip_len</span><span class="p">):</span>
+                <span class="n">video_frames</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">frame_transform</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">]))</span>
+                <span class="n">current_pts</span> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">frame</span></a><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">]</span>
+            <span class="c1"># Stack it into a tensor</span>
+            <span class="n">video</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.stack.html#torch.stack" title="torch.stack" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">stack</span></a><span class="p">(</span><span class="n">video_frames</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_transform</span><span class="p">:</span>
+                <span class="n">video</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">video_transform</span><span class="p">(</span><span class="n">video</span><span class="p">)</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s1">&#39;path&#39;</span><span class="p">:</span> <span class="n">path</span><span class="p">,</span>
+                <span class="s1">&#39;video&#39;</span><span class="p">:</span> <span class="n">video</span><span class="p">,</span>
+                <span class="s1">&#39;target&#39;</span><span class="p">:</span> <span class="n">target</span><span class="p">,</span>
+                <span class="s1">&#39;start&#39;</span><span class="p">:</span> <span class="n">start</span><span class="p">,</span>
+                <span class="s1">&#39;end&#39;</span><span class="p">:</span> <span class="n">current_pts</span><span class="p">}</span>
+            <span class="k">yield</span> <span class="n">output</span>
+</pre></div>
+</div>
+<p>Given a path of videos in a folder structure, i.e:</p>
+<ul class="simple">
+<li><dl class="simple">
+<dt>dataset</dt><dd><ul>
+<li><dl class="simple">
+<dt>class 1</dt><dd><ul>
+<li><p>file 0</p></li>
+<li><p>file 1</p></li>
+<li><p>…</p></li>
+</ul>
+</dd>
+</dl>
+</li>
+<li><dl class="simple">
+<dt>class 2</dt><dd><ul>
+<li><p>file 0</p></li>
+<li><p>file 1</p></li>
+<li><p>…</p></li>
+</ul>
+</dd>
+</dl>
+</li>
+<li><p>…</p></li>
+</ul>
+</dd>
+</dl>
+</li>
+</ul>
+<p>We can generate a dataloader and test the dataset.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a> <span class="o">=</span> <span class="p">[</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">t</span><span class="o">.</span><span class="n">Resize</span></a><span class="p">((</span><span class="mi">112</span><span class="p">,</span> <span class="mi">112</span><span class="p">))]</span>
+<span class="n">frame_transform</span> <span class="o">=</span> <span class="n">t</span><span class="o">.</span><span class="n">Compose</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transforms</span></a><span class="p">)</span>
+
+<span class="n">dataset</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.IterableDataset" title="torch.utils.data.IterableDataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">RandomDataset</span></a><span class="p">(</span><span class="s2">&quot;./dataset&quot;</span><span class="p">,</span> <span class="n">epoch_size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">frame_transform</span><span class="o">=</span><span class="n">frame_transform</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torch.utils.data</span> <span class="kn">import</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">DataLoader</span></a>
+<a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">loader</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">DataLoader</span></a><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">12</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data</span></a> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;video&quot;</span><span class="p">:</span> <span class="p">[],</span> <span class="s1">&#39;start&#39;</span><span class="p">:</span> <span class="p">[],</span> <span class="s1">&#39;end&#39;</span><span class="p">:</span> <span class="p">[],</span> <span class="s1">&#39;tensorsize&#39;</span><span class="p">:</span> <span class="p">[]}</span>
+<span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a> <span class="ow">in</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">loader</span></a><span class="p">:</span>
+    <span class="k">for</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">[</span><span class="s1">&#39;path&#39;</span><span class="p">])):</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data</span></a><span class="p">[</span><span class="s1">&#39;video&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">[</span><span class="s1">&#39;path&#39;</span><span class="p">][</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="p">])</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data</span></a><span class="p">[</span><span class="s1">&#39;start&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">[</span><span class="s1">&#39;start&#39;</span><span class="p">][</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="p">]</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data</span></a><span class="p">[</span><span class="s1">&#39;end&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">[</span><span class="s1">&#39;end&#39;</span><span class="p">][</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="p">]</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+        <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data</span></a><span class="p">[</span><span class="s1">&#39;tensorsize&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">[</span><span class="s1">&#39;video&#39;</span><span class="p">][</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/pytorch/vision/torchvision/transforms/functional.py:1603: UserWarning: The default value of the antialias parameter of all the resizing transforms (Resize(), RandomResizedCrop(), etc.) will change from None to True in v0.17, in order to be consistent across the PIL and Tensor backends. To suppress this warning, directly pass antialias=True (recommended, future default), antialias=None (current default, which means False for Tensors and True for PIL), or antialias=False (only works on Tensors - PIL will still use antialiasing). This also applies if you are using the inference transforms from the models weights: update the call to weights.transforms(antialias=True).
+  warnings.warn(
+{&#39;video&#39;: [&#39;./dataset/1/RATRACE_wave_f_nm_np1_fr_goo_37.avi&#39;, &#39;./dataset/2/SOX5yA1l24A.mp4&#39;, &#39;./dataset/2/SOX5yA1l24A.mp4&#39;, &#39;./dataset/2/v_SoccerJuggling_g24_c01.avi&#39;, &#39;./dataset/2/v_SoccerJuggling_g23_c01.avi&#39;], &#39;start&#39;: [1.2933978369482158, 1.7918684185786267, 4.373791147055652, 3.9307513195312067, 3.6426948895065263], &#39;end&#39;: [1.7999999999999998, 2.3023, 4.9049, 4.437767, 4.170833], &#39;tensorsize&#39;: [torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112]), torch.Size([16, 3, 112, 112])]}
+</pre></div>
+</div>
+</section>
+<section id="data-visualization">
+<h2>4. Data Visualization<a class="headerlink" href="#data-visualization" title="Permalink to this heading">¶</a></h2>
+<p>Example of visualized video</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
+
+<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.figure.html#matplotlib.pyplot.figure" title="matplotlib.pyplot.figure" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">figure</span></a><span class="p">(</span><span class="n">figsize</span><span class="o">=</span><span class="p">(</span><span class="mi">12</span><span class="p">,</span> <span class="mi">12</span><span class="p">))</span>
+<span class="k">for</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">16</span><span class="p">):</span>
+    <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.subplot.html#matplotlib.pyplot.subplot" title="matplotlib.pyplot.subplot" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">subplot</span></a><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+    <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.imshow.html#matplotlib.pyplot.imshow" title="matplotlib.pyplot.imshow" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">imshow</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">[</span><span class="s2">&quot;video&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">i</span></a><span class="p">,</span> <span class="o">...</span><span class="p">]</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+    <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.axis.html#matplotlib.pyplot.axis" title="matplotlib.pyplot.axis" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">axis</span></a><span class="p">(</span><span class="s2">&quot;off&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_video_api_001.png" srcset="../../_images/sphx_glr_plot_video_api_001.png" alt="plot video api" class = "sphx-glr-single-img"/><p>Cleanup the video and dataset:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">os</span>
+<span class="kn">import</span> <span class="nn">shutil</span>
+<a href="https://docs.python.org/3/library/os.html#os.remove" title="os.remove" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">remove</span></a><span class="p">(</span><span class="s2">&quot;./WUzgd7C1pWA.mp4&quot;</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/shutil.html#shutil.rmtree" title="shutil.rmtree" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span></a><span class="p">(</span><span class="s2">&quot;./dataset&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 5.277 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-others-plot-video-api-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/175e3bb1b03a25b015d65842a3b8a883/plot_video_api.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_video_api.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/4b3151295a1f6b75ffd88837fb31f719/plot_video_api.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_video_api.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_visualization_utils.html" class="btn btn-neutral float-right" title="Visualization utilities" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_scripted_tensor_transforms.html" class="btn btn-neutral" title="Torchscript support" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Video API</a><ul>
+<li><a class="reference internal" href="#introduction-building-a-new-video-object-and-examining-the-properties">1. Introduction: building a new video object and examining the properties</a></li>
+<li><a class="reference internal" href="#building-a-sample-read-video-function">2. Building a sample read_video function</a></li>
+<li><a class="reference internal" href="#building-an-example-randomly-sampled-dataset-can-be-applied-to-training-dataset-of-kinetics400">3. Building an example randomly sampled dataset (can be applied to training dataset of kinetics400)</a></li>
+<li><a class="reference internal" href="#data-visualization">4. Data Visualization</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/others/plot_visualization_utils.html b/0.16/auto_examples/others/plot_visualization_utils.html
new file mode 100644
index 00000000000..e2508627d97
--- /dev/null
+++ b/0.16/auto_examples/others/plot_visualization_utils.html
@@ -0,0 +1,1299 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Visualization utilities &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Training references" href="../../training_references.html" />
+    <link rel="prev" title="Video API" href="plot_video_api.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Others</a> &gt;</li>
+        
+      <li>Visualization utilities</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/others/plot_visualization_utils.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="visualization-utilities">
+<span id="sphx-glr-auto-examples-others-plot-visualization-utils-py"></span><h1>Visualization utilities<a class="headerlink" href="#visualization-utilities" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_visualization_utils.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>This example illustrates some of the utilities that torchvision offers for
+visualizing images, bounding boxes, segmentation masks and keypoints.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
+
+<span class="kn">import</span> <span class="nn">torchvision.transforms.functional</span> <span class="k">as</span> <span class="nn">F</span>
+
+
+<a href="https://matplotlib.org/stable/api/matplotlib_configuration_api.html#matplotlib.RcParams" title="matplotlib.RcParams" class="sphx-glr-backref-module-matplotlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">plt</span><span class="o">.</span><span class="n">rcParams</span></a><span class="p">[</span><span class="s2">&quot;savefig.bbox&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;tight&#39;</span>
+
+
+<span class="k">def</span> <span class="nf">show</span><span class="p">(</span><span class="n">imgs</span><span class="p">):</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">imgs</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+        <span class="n">imgs</span> <span class="o">=</span> <span class="p">[</span><span class="n">imgs</span><span class="p">]</span>
+    <span class="n">fig</span><span class="p">,</span> <span class="n">axs</span> <span class="o">=</span> <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.subplots.html#matplotlib.pyplot.subplots" title="matplotlib.pyplot.subplots" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">subplots</span></a><span class="p">(</span><span class="n">ncols</span><span class="o">=</span><span class="nb">len</span><span class="p">(</span><span class="n">imgs</span><span class="p">),</span> <span class="n">squeeze</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">img</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">imgs</span><span class="p">):</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">img</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
+        <span class="n">img</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">to_pil_image</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+        <span class="n">axs</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">imshow</span><span class="p">(</span><a href="https://numpy.org/doc/stable/reference/generated/numpy.asarray.html#numpy.asarray" title="numpy.asarray" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">asarray</span></a><span class="p">(</span><span class="n">img</span><span class="p">))</span>
+        <span class="n">axs</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">set</span><span class="p">(</span><span class="n">xticklabels</span><span class="o">=</span><span class="p">[],</span> <span class="n">yticklabels</span><span class="o">=</span><span class="p">[],</span> <span class="n">xticks</span><span class="o">=</span><span class="p">[],</span> <span class="n">yticks</span><span class="o">=</span><span class="p">[])</span>
+</pre></div>
+</div>
+<section id="visualizing-a-grid-of-images">
+<h2>Visualizing a grid of images<a class="headerlink" href="#visualizing-a-grid-of-images" title="Permalink to this heading">¶</a></h2>
+<p>The <a class="reference internal" href="../../generated/torchvision.utils.make_grid.html#torchvision.utils.make_grid" title="torchvision.utils.make_grid"><code class="xref py py-func docutils literal notranslate"><span class="pre">make_grid()</span></code></a> function can be used to create a
+tensor that represents multiple images in a grid.  This util requires a single
+image of dtype <code class="docutils literal notranslate"><span class="pre">uint8</span></code> as input.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">make_grid</span>
+<span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_image</span>
+<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_int</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><span class="s1">&#39;../assets&#39;</span><span class="p">)</span> <span class="o">/</span> <span class="s1">&#39;dog1.jpg&#39;</span><span class="p">))</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog2_int</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><span class="s1">&#39;../assets&#39;</span><span class="p">)</span> <span class="o">/</span> <span class="s1">&#39;dog2.jpg&#39;</span><span class="p">))</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a> <span class="o">=</span> <span class="p">[</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_int</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog2_int</span></a><span class="p">]</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">grid</span></a> <span class="o">=</span> <span class="n">make_grid</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">)</span>
+<span class="n">show</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">grid</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_001.png" srcset="../../_images/sphx_glr_plot_visualization_utils_001.png" alt="plot visualization utils" class = "sphx-glr-single-img"/></section>
+<section id="visualizing-bounding-boxes">
+<h2>Visualizing bounding boxes<a class="headerlink" href="#visualizing-bounding-boxes" title="Permalink to this heading">¶</a></h2>
+<p>We can use <a class="reference internal" href="../../generated/torchvision.utils.draw_bounding_boxes.html#torchvision.utils.draw_bounding_boxes" title="torchvision.utils.draw_bounding_boxes"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_bounding_boxes()</span></code></a> to draw boxes on an
+image. We can set the colors, labels, width as well as font and font size.
+The boxes are in <code class="docutils literal notranslate"><span class="pre">(xmin,</span> <span class="pre">ymin,</span> <span class="pre">xmax,</span> <span class="pre">ymax)</span></code> format.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">draw_bounding_boxes</span>
+
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.tensor.html#torch.tensor" title="torch.tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span></a><span class="p">([[</span><span class="mi">50</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">100</span><span class="p">,</span> <span class="mi">200</span><span class="p">],</span> <span class="p">[</span><span class="mi">210</span><span class="p">,</span> <span class="mi">150</span><span class="p">,</span> <span class="mi">350</span><span class="p">,</span> <span class="mi">430</span><span class="p">]],</span> <span class="n">dtype</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">float</span></a><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">colors</span></a> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;blue&quot;</span><span class="p">,</span> <span class="s2">&quot;yellow&quot;</span><span class="p">]</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">result</span></a> <span class="o">=</span> <span class="n">draw_bounding_boxes</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_int</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">colors</span></a><span class="o">=</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">colors</span></a><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="mi">5</span><span class="p">)</span>
+<span class="n">show</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">result</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_002.png" srcset="../../_images/sphx_glr_plot_visualization_utils_002.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><p>Naturally, we can also plot bounding boxes produced by torchvision detection
+models.  Here is a demo with a Faster R-CNN model loaded from
+<a class="reference internal" href="../../models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="torchvision.models.detection.fasterrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn()</span></code></a>
+model. For more details on the output of such models, you may
+refer to <a class="reference internal" href="#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models.detection</span> <span class="kn">import</span> <span class="n">fasterrcnn_resnet50_fpn</span><span class="p">,</span> <span class="n">FasterRCNN_ResNet50_FPN_Weights</span>
+
+
+<span class="n">weights</span> <span class="o">=</span> <span class="n">FasterRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">transforms</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">transforms</span><span class="p">(</span><span class="n">d</span><span class="p">)</span> <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">]</span>
+
+<span class="n">model</span> <span class="o">=</span> <span class="n">fasterrcnn_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="n">model</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.eval" title="torch.nn.Module.eval" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-method"><span class="n">model</span><span class="o">.</span><span class="n">eval</span></a><span class="p">()</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">outputs</span></a> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">outputs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[{&#39;boxes&#39;: tensor([[215.9767, 171.1661, 402.0078, 378.7391],
+        [344.6341, 172.6735, 357.6114, 220.1435],
+        [153.1306, 185.5567, 172.9223, 254.7014]], grad_fn=&lt;StackBackward0&gt;), &#39;labels&#39;: tensor([18,  1,  1]), &#39;scores&#39;: tensor([0.9989, 0.0701, 0.0611], grad_fn=&lt;IndexBackward0&gt;)}, {&#39;boxes&#39;: tensor([[ 23.5964, 132.4331, 449.9359, 493.0222],
+        [225.8182, 124.6292, 467.2861, 492.2621],
+        [ 18.5248, 135.4171, 420.9786, 479.2225]], grad_fn=&lt;StackBackward0&gt;), &#39;labels&#39;: tensor([18, 18, 17]), &#39;scores&#39;: tensor([0.9980, 0.0879, 0.0671], grad_fn=&lt;IndexBackward0&gt;)}]
+</pre></div>
+</div>
+<p>Let’s plot the boxes detected by our model. We will only plot the boxes with a
+score greater than a given threshold.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">score_threshold</span></a> <span class="o">=</span> <span class="mf">.8</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_boxes</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="n">draw_bounding_boxes</span><span class="p">(</span><span class="n">dog_int</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boxes</span></a><span class="o">=</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a><span class="p">[</span><span class="s1">&#39;boxes&#39;</span><span class="p">][</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a><span class="p">[</span><span class="s1">&#39;scores&#39;</span><span class="p">]</span> <span class="o">&gt;</span> <a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">score_threshold</span></a><span class="p">],</span> <span class="n">width</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">dog_int</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">outputs</span></a><span class="p">)</span>
+<span class="p">]</span>
+<span class="n">show</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_boxes</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_003.png" srcset="../../_images/sphx_glr_plot_visualization_utils_003.png" alt="plot visualization utils" class = "sphx-glr-single-img"/></section>
+<section id="visualizing-segmentation-masks">
+<h2>Visualizing segmentation masks<a class="headerlink" href="#visualizing-segmentation-masks" title="Permalink to this heading">¶</a></h2>
+<p>The <a class="reference internal" href="../../generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks" title="torchvision.utils.draw_segmentation_masks"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_segmentation_masks()</span></code></a> function can be used to
+draw segmentation masks on images. Semantic segmentation and instance
+segmentation models have different outputs, so we will treat each
+independently.</p>
+<section id="semantic-segmentation-models">
+<span id="semantic-seg-output"></span><h3>Semantic segmentation models<a class="headerlink" href="#semantic-segmentation-models" title="Permalink to this heading">¶</a></h3>
+<p>We will see how to use it with torchvision’s FCN Resnet-50, loaded with
+<a class="reference internal" href="../../models/generated/torchvision.models.segmentation.fcn_resnet50.html#torchvision.models.segmentation.fcn_resnet50" title="torchvision.models.segmentation.fcn_resnet50"><code class="xref py py-func docutils literal notranslate"><span class="pre">fcn_resnet50()</span></code></a>. Let’s start by looking
+at the output of the model.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models.segmentation</span> <span class="kn">import</span> <span class="n">fcn_resnet50</span><span class="p">,</span> <span class="n">FCN_ResNet50_Weights</span>
+
+<span class="n">weights</span> <span class="o">=</span> <span class="n">FCN_ResNet50_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">transforms</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">(</span><span class="n">resize_size</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+
+<span class="n">model</span> <span class="o">=</span> <span class="n">fcn_resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="n">model</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.eval" title="torch.nn.Module.eval" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-method"><span class="n">model</span><span class="o">.</span><span class="n">eval</span></a><span class="p">()</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.stack.html#torch.stack" title="torch.stack" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">stack</span></a><span class="p">([</span><span class="n">transforms</span><span class="p">(</span><span class="n">d</span><span class="p">)</span> <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">])</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">)[</span><span class="s1">&#39;out&#39;</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span><span class="o">.</span><span class="n">shape</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span><span class="o">.</span><span class="n">min</span></a><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span><span class="o">.</span><span class="n">max</span></a><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/fcn_resnet50_coco-1167a1af.pth&quot; to /root/.cache/torch/hub/checkpoints/fcn_resnet50_coco-1167a1af.pth
+torch.Size([2, 21, 500, 500]) -7.089669704437256 14.858257293701172
+</pre></div>
+</div>
+<p>As we can see above, the output of the segmentation model is a tensor of shape
+<code class="docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_classes,</span> <span class="pre">H,</span> <span class="pre">W)</span></code>. Each value is a non-normalized score, and
+we can normalize them into <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">1]</span></code> by using a softmax. After the softmax,
+we can interpret each value as a probability indicating how likely a given
+pixel is to belong to a given class.</p>
+<p>Let’s plot the masks that have been detected for the dog class and for the
+boat class:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sem_class_to_idx</span></a> <span class="o">=</span> <span class="p">{</span><span class="bp">cls</span><span class="p">:</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">idx</span></a> <span class="k">for</span> <span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">idx</span></a><span class="p">,</span> <span class="bp">cls</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">])}</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">normalized_masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.functional.softmax.html#torch.nn.functional.softmax" title="torch.nn.functional.softmax" class="sphx-glr-backref-module-torch-nn-functional sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">softmax</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_and_boat_masks</span></a> <span class="o">=</span> <span class="p">[</span>
+    <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">normalized_masks</span></a><span class="p">[</span><span class="n">img_idx</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sem_class_to_idx</span></a><span class="p">[</span><span class="bp">cls</span><span class="p">]]</span>
+    <span class="k">for</span> <span class="n">img_idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">))</span>
+    <span class="k">for</span> <span class="bp">cls</span> <span class="ow">in</span> <span class="p">(</span><span class="s1">&#39;dog&#39;</span><span class="p">,</span> <span class="s1">&#39;boat&#39;</span><span class="p">)</span>
+<span class="p">]</span>
+
+<span class="n">show</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_and_boat_masks</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_004.png" srcset="../../_images/sphx_glr_plot_visualization_utils_004.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><p>As expected, the model is confident about the dog class, but not so much for
+the boat class.</p>
+<p>The <a class="reference internal" href="../../generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks" title="torchvision.utils.draw_segmentation_masks"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_segmentation_masks()</span></code></a> function can be used to
+plots those masks on top of the original image. This function expects the
+masks to be boolean masks, but our masks above contain probabilities in <code class="docutils literal notranslate"><span class="pre">[0,</span>
+<span class="pre">1]</span></code>. To get boolean masks, we can do the following:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">class_dim</span></a> <span class="o">=</span> <span class="mi">1</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boolean_dog_masks</span></a> <span class="o">=</span> <span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">normalized_masks</span></a><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">class_dim</span></a><span class="p">)</span> <span class="o">==</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sem_class_to_idx</span></a><span class="p">[</span><span class="s1">&#39;dog&#39;</span><span class="p">])</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;shape = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boolean_dog_masks</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">, dtype = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boolean_dog_masks</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+<span class="n">show</span><span class="p">([</span><span class="n">m</span><span class="o">.</span><span class="n">float</span><span class="p">()</span> <span class="k">for</span> <span class="n">m</span> <span class="ow">in</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boolean_dog_masks</span></a><span class="p">])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_005.png" srcset="../../_images/sphx_glr_plot_visualization_utils_005.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>shape = torch.Size([2, 500, 500]), dtype = torch.bool
+</pre></div>
+</div>
+<p>The line above where we define <code class="docutils literal notranslate"><span class="pre">boolean_dog_masks</span></code> is a bit cryptic, but you
+can read it as the following query: “For which pixels is ‘dog’ the most likely
+class?”</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>While we’re using the <code class="docutils literal notranslate"><span class="pre">normalized_masks</span></code> here, we would have
+gotten the same result by using the non-normalized scores of the model
+directly (as the softmax operation preserves the order).</p>
+</div>
+<p>Now that we have boolean masks, we can use them with
+<a class="reference internal" href="../../generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks" title="torchvision.utils.draw_segmentation_masks"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_segmentation_masks()</span></code></a> to plot them on top of the
+original images:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">draw_segmentation_masks</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_masks</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="n">draw_segmentation_masks</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">masks</span><span class="o">=</span><span class="n">mask</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.7</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">img</span><span class="p">,</span> <span class="n">mask</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boolean_dog_masks</span></a><span class="p">)</span>
+<span class="p">]</span>
+<span class="n">show</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_masks</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_006.png" srcset="../../_images/sphx_glr_plot_visualization_utils_006.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><p>We can plot more than one mask per image! Remember that the model returned as
+many masks as there are classes. Let’s ask the same query as above, but this
+time for <em>all</em> classes, not just the dog class: “For each pixel and each class
+C, is class C the most likely class?”</p>
+<p>This one is a bit more involved, so we’ll first show how to do it with a
+single image, and then we’ll generalize to the batch</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">num_classes</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">normalized_masks</span></a><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">normalized_masks</span></a><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+<a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">class_dim</span></a> <span class="o">=</span> <span class="mi">0</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_all_classes_masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span></a><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">class_dim</span></a><span class="p">)</span> <span class="o">==</span> <a href="https://pytorch.org/docs/stable/generated/torch.arange.html#torch.arange" title="torch.arange" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">arange</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">num_classes</span></a><span class="p">)[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;dog1_masks shape = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">, dtype = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;dog1_all_classes_masks = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_all_classes_masks</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">, dtype = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_all_classes_masks</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_with_all_masks</span></a> <span class="o">=</span> <span class="n">draw_segmentation_masks</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_int</span></a><span class="p">,</span> <span class="n">masks</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_all_classes_masks</span></a><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">.6</span><span class="p">)</span>
+<span class="n">show</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_with_all_masks</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_007.png" srcset="../../_images/sphx_glr_plot_visualization_utils_007.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>dog1_masks shape = torch.Size([21, 500, 500]), dtype = torch.float32
+dog1_all_classes_masks = torch.Size([21, 500, 500]), dtype = torch.bool
+</pre></div>
+</div>
+<p>We can see in the image above that only 2 masks were drawn: the mask for the
+background and the mask for the dog. This is because the model thinks that
+only these 2 classes are the most likely ones across all the pixels. If the
+model had detected another class as the most likely among other pixels, we
+would have seen its mask above.</p>
+<p>Removing the background mask is as simple as passing
+<code class="docutils literal notranslate"><span class="pre">masks=dog1_all_classes_masks[1:]</span></code>, because the background class is the
+class with index 0.</p>
+<p>Let’s now do the same but for an entire batch of images. The code is similar
+but involves a bit more juggling with the dimensions.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">class_dim</span></a> <span class="o">=</span> <span class="mi">1</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">all_classes_masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">normalized_masks</span></a><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">class_dim</span></a><span class="p">)</span> <span class="o">==</span> <a href="https://pytorch.org/docs/stable/generated/torch.arange.html#torch.arange" title="torch.arange" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">arange</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">num_classes</span></a><span class="p">)[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;shape = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">all_classes_masks</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">, dtype = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">all_classes_masks</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+<span class="c1"># The first dimension is the classes now, so we need to swap it</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">all_classes_masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">all_classes_masks</span></a><span class="o">.</span><span class="n">swapaxes</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_masks</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="n">draw_segmentation_masks</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">masks</span><span class="o">=</span><span class="n">mask</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">.6</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">img</span><span class="p">,</span> <span class="n">mask</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">all_classes_masks</span></a><span class="p">)</span>
+<span class="p">]</span>
+<span class="n">show</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_masks</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_008.png" srcset="../../_images/sphx_glr_plot_visualization_utils_008.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>shape = torch.Size([21, 2, 500, 500]), dtype = torch.bool
+</pre></div>
+</div>
+</section>
+<section id="instance-segmentation-models">
+<span id="instance-seg-output"></span><h3>Instance segmentation models<a class="headerlink" href="#instance-segmentation-models" title="Permalink to this heading">¶</a></h3>
+<p>Instance segmentation models have a significantly different output from the
+semantic segmentation models. We will see here how to plot the masks for such
+models. Let’s start by analyzing the output of a Mask-RCNN model. Note that
+these models don’t require the images to be normalized, so we don’t need to
+use the normalized batch.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>We will here describe the output of a Mask-RCNN model. The models in
+<a class="reference internal" href="../../models.html#object-det-inst-seg-pers-keypoint-det"><span class="std std-ref">Object Detection, Instance Segmentation and Person Keypoint Detection</span></a> all have a similar output
+format, but some of them may have extra info like keypoints for
+<a class="reference internal" href="../../models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.keypointrcnn_resnet50_fpn" title="torchvision.models.detection.keypointrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">keypointrcnn_resnet50_fpn()</span></code></a>, and some
+of them may not have masks, like
+<a class="reference internal" href="../../models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="torchvision.models.detection.fasterrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn()</span></code></a>.</p>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models.detection</span> <span class="kn">import</span> <span class="n">maskrcnn_resnet50_fpn</span><span class="p">,</span> <span class="n">MaskRCNN_ResNet50_FPN_Weights</span>
+
+<span class="n">weights</span> <span class="o">=</span> <span class="n">MaskRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">transforms</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">transforms</span><span class="p">(</span><span class="n">d</span><span class="p">)</span> <span class="k">for</span> <span class="n">d</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">]</span>
+
+<span class="n">model</span> <span class="o">=</span> <span class="n">maskrcnn_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="n">model</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.eval" title="torch.nn.Module.eval" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-method"><span class="n">model</span><span class="o">.</span><span class="n">eval</span></a><span class="p">()</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth&quot; to /root/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
+[{&#39;boxes&#39;: tensor([[219.7444, 168.1722, 400.7378, 384.0263],
+        [343.9716, 171.2287, 358.3447, 222.6263],
+        [301.0303, 192.6917, 313.8879, 232.3154]], grad_fn=&lt;StackBackward0&gt;), &#39;labels&#39;: tensor([18,  1,  1]), &#39;scores&#39;: tensor([0.9987, 0.7187, 0.6525], grad_fn=&lt;IndexBackward0&gt;), &#39;masks&#39;: tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          ...,
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.]]],
+
+
+        [[[0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          ...,
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.]]],
+
+
+        [[[0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          ...,
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.]]]], grad_fn=&lt;UnsqueezeBackward0&gt;)}, {&#39;boxes&#39;: tensor([[ 44.6767, 137.9018, 446.5324, 487.3429],
+        [  0.0000, 288.0053, 489.9292, 490.2352]], grad_fn=&lt;StackBackward0&gt;), &#39;labels&#39;: tensor([18, 15]), &#39;scores&#39;: tensor([0.9978, 0.0697], grad_fn=&lt;IndexBackward0&gt;), &#39;masks&#39;: tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          ...,
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.]]],
+
+
+        [[[0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          ...,
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.],
+          [0., 0., 0.,  ..., 0., 0., 0.]]]], grad_fn=&lt;UnsqueezeBackward0&gt;)}]
+</pre></div>
+</div>
+<p>Let’s break this down. For each image in the batch, the model outputs some
+detections (or instances). The number of detections varies for each input
+image. Each instance is described by its bounding box, its label, its score
+and its mask.</p>
+<p>The way the output is organized is as follows: the output is a list of length
+<code class="docutils literal notranslate"><span class="pre">batch_size</span></code>. Each entry in the list corresponds to an input image, and it
+is a dict with keys ‘boxes’, ‘labels’, ‘scores’, and ‘masks’. Each value
+associated to those keys has <code class="docutils literal notranslate"><span class="pre">num_instances</span></code> elements in it.  In our case
+above there are 3 instances detected in the first image, and 2 instances in
+the second one.</p>
+<p>The boxes can be plotted with <a class="reference internal" href="../../generated/torchvision.utils.draw_bounding_boxes.html#torchvision.utils.draw_bounding_boxes" title="torchvision.utils.draw_bounding_boxes"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_bounding_boxes()</span></code></a>
+as above, but here we’re more interested in the masks. These masks are quite
+different from the masks that we saw above for the semantic segmentation
+models.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_output</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_output</span></a><span class="p">[</span><span class="s1">&#39;masks&#39;</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;shape = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">, dtype = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">, &quot;</span>
+      <span class="sa">f</span><span class="s2">&quot;min = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span></a><span class="o">.</span><span class="n">min</span><span class="p">()</span><span class="si">}</span><span class="s2">, max = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_masks</span></a><span class="o">.</span><span class="n">max</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>shape = torch.Size([3, 1, 500, 500]), dtype = torch.float32, min = 0.0, max = 0.9999862909317017
+</pre></div>
+</div>
+<p>Here the masks correspond to probabilities indicating, for each pixel, how
+likely it is to belong to the predicted label of that instance. Those
+predicted labels correspond to the ‘labels’ element in the same output dict.
+Let’s see which labels were predicted for the instances of the first image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="s2">&quot;For the first dog, the following instances were detected:&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">([</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">][</span><span class="n">label</span><span class="p">]</span> <span class="k">for</span> <span class="n">label</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_output</span></a><span class="p">[</span><span class="s1">&#39;labels&#39;</span><span class="p">]])</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>For the first dog, the following instances were detected:
+[&#39;dog&#39;, &#39;person&#39;, &#39;person&#39;]
+</pre></div>
+</div>
+<p>Interestingly, the model detects two persons in the image. Let’s go ahead and
+plot those masks. Since <a class="reference internal" href="../../generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks" title="torchvision.utils.draw_segmentation_masks"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_segmentation_masks()</span></code></a>
+expects boolean masks, we need to convert those probabilities into boolean
+values. Remember that the semantic of those masks is “How likely is this pixel
+to belong to the predicted class?”. As a result, a natural way of converting
+those masks into boolean values is to threshold them with the 0.5 probability
+(one could also choose a different threshold).</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">proba_threshold</span></a> <span class="o">=</span> <span class="mf">0.5</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_bool_masks</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_output</span></a><span class="p">[</span><span class="s1">&#39;masks&#39;</span><span class="p">]</span> <span class="o">&gt;</span> <a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">proba_threshold</span></a>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;shape = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_bool_masks</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2">, dtype = </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_bool_masks</span><span class="o">.</span><span class="n">dtype</span></a><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+<span class="c1"># There&#39;s an extra dimension (1) to the masks. We need to remove it</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_bool_masks</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_bool_masks</span></a><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+<span class="n">show</span><span class="p">(</span><span class="n">draw_segmentation_masks</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_int</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_bool_masks</span></a><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mf">0.9</span><span class="p">))</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_009.png" srcset="../../_images/sphx_glr_plot_visualization_utils_009.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>shape = torch.Size([3, 1, 500, 500]), dtype = torch.bool
+</pre></div>
+</div>
+<p>The model seems to have properly detected the dog, but it also confused trees
+with people. Looking more closely at the scores will help us plot more
+relevant masks:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog1_output</span></a><span class="p">[</span><span class="s1">&#39;scores&#39;</span><span class="p">])</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>tensor([0.9987, 0.7187, 0.6525], grad_fn=&lt;IndexBackward0&gt;)
+</pre></div>
+</div>
+<p>Clearly the model is more confident about the dog detection than it is about
+the people detections. That’s good news. When plotting the masks, we can ask
+for only those that have a good score. Let’s use a score threshold of .75
+here, and also plot the masks of the second dog.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">score_threshold</span></a> <span class="o">=</span> <span class="mf">.75</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boolean_masks</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="n">out</span><span class="p">[</span><span class="s1">&#39;masks&#39;</span><span class="p">][</span><span class="n">out</span><span class="p">[</span><span class="s1">&#39;scores&#39;</span><span class="p">]</span> <span class="o">&gt;</span> <a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">score_threshold</span></a><span class="p">]</span> <span class="o">&gt;</span> <a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">proba_threshold</span></a>
+    <span class="k">for</span> <span class="n">out</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">output</span></a>
+<span class="p">]</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_masks</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="n">draw_segmentation_masks</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">mask</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">1</span><span class="p">))</span>
+    <span class="k">for</span> <span class="n">img</span><span class="p">,</span> <span class="n">mask</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dog_list</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">boolean_masks</span></a><span class="p">)</span>
+<span class="p">]</span>
+<span class="n">show</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dogs_with_masks</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_010.png" srcset="../../_images/sphx_glr_plot_visualization_utils_010.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><p>The two ‘people’ masks in the first image where not selected because they have
+a lower score than the score threshold. Similarly, in the second image, the
+instance with class 15 (which corresponds to ‘bench’) was not selected.</p>
+</section>
+</section>
+<section id="visualizing-keypoints">
+<span id="keypoint-output"></span><h2>Visualizing keypoints<a class="headerlink" href="#visualizing-keypoints" title="Permalink to this heading">¶</a></h2>
+<p>The <a class="reference internal" href="../../generated/torchvision.utils.draw_keypoints.html#torchvision.utils.draw_keypoints" title="torchvision.utils.draw_keypoints"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_keypoints()</span></code></a> function can be used to
+draw keypoints on images. We will see how to use it with
+torchvision’s KeypointRCNN loaded with <a class="reference internal" href="../../models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.keypointrcnn_resnet50_fpn" title="torchvision.models.detection.keypointrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">keypointrcnn_resnet50_fpn()</span></code></a>.
+We will first have a look at output of the model.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models.detection</span> <span class="kn">import</span> <span class="n">keypointrcnn_resnet50_fpn</span><span class="p">,</span> <span class="n">KeypointRCNN_ResNet50_FPN_Weights</span>
+<span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_image</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">person_int</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><span class="s2">&quot;../assets&quot;</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;person1.jpg&quot;</span><span class="p">))</span>
+
+<span class="n">weights</span> <span class="o">=</span> <span class="n">KeypointRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">transforms</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">person_float</span></a> <span class="o">=</span> <span class="n">transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">person_int</span></a><span class="p">)</span>
+
+<span class="n">model</span> <span class="o">=</span> <span class="n">keypointrcnn_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">progress</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+<span class="n">model</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.eval" title="torch.nn.Module.eval" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-method"><span class="n">model</span><span class="o">.</span><span class="n">eval</span></a><span class="p">()</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">outputs</span></a> <span class="o">=</span> <span class="n">model</span><span class="p">([</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">person_float</span></a><span class="p">])</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">outputs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://download.pytorch.org/models/keypointrcnn_resnet50_fpn_coco-fc266e95.pth&quot; to /root/.cache/torch/hub/checkpoints/keypointrcnn_resnet50_fpn_coco-fc266e95.pth
+[{&#39;boxes&#39;: tensor([[124.3751, 177.9242, 327.6354, 574.7064],
+        [124.3625, 180.7574, 290.1061, 390.7958]], grad_fn=&lt;StackBackward0&gt;), &#39;labels&#39;: tensor([1, 1]), &#39;scores&#39;: tensor([0.9998, 0.1070], grad_fn=&lt;IndexBackward0&gt;), &#39;keypoints&#39;: tensor([[[208.0176, 214.2408,   1.0000],
+         [208.0176, 207.0375,   1.0000],
+         [197.8246, 210.6392,   1.0000],
+         [208.0176, 211.8398,   1.0000],
+         [178.6378, 217.8425,   1.0000],
+         [221.2086, 253.8590,   1.0000],
+         [160.6502, 269.4662,   1.0000],
+         [243.9929, 304.2822,   1.0000],
+         [138.4654, 328.8935,   1.0000],
+         [277.5698, 340.8990,   1.0000],
+         [153.4551, 374.5144,   1.0000],
+         [226.0053, 375.7150,   1.0000],
+         [226.0053, 370.3125,   1.0000],
+         [221.8082, 455.5516,   1.0000],
+         [273.9723, 448.9486,   1.0000],
+         [193.6275, 546.1932,   1.0000],
+         [273.3727, 545.5930,   1.0000]],
+
+        [[207.8327, 214.6636,   1.0000],
+         [207.2343, 207.4622,   1.0000],
+         [198.2590, 209.8627,   1.0000],
+         [208.4310, 210.4628,   1.0000],
+         [178.5134, 218.2642,   1.0000],
+         [219.7997, 251.8704,   1.0000],
+         [162.3579, 269.2736,   1.0000],
+         [245.5289, 304.6800,   1.0000],
+         [138.4238, 330.4848,   1.0000],
+         [278.4382, 346.0876,   1.0000],
+         [153.3826, 374.8929,   1.0000],
+         [233.5618, 368.2917,   1.0000],
+         [225.7832, 367.6916,   1.0000],
+         [289.8069, 357.4897,   1.0000],
+         [245.5289, 389.8956,   1.0000],
+         [281.4300, 349.0882,   1.0000],
+         [209.0294, 389.8956,   1.0000]]], grad_fn=&lt;CopySlices&gt;), &#39;keypoints_scores&#39;: tensor([[16.0164, 16.6672, 15.8312,  4.6510, 14.2053,  8.8280,  9.1136, 12.2084,
+         12.1901, 13.8453, 10.7090,  5.5852,  7.5005, 11.3378,  9.3700,  8.2987,
+          8.4479],
+        [12.9326, 13.8158, 14.9053,  3.9368, 12.9585,  6.4240,  6.8328, 10.4227,
+          9.2907, 10.1066, 10.1019,  0.1822,  4.3057, -4.9904, -2.7409, -2.7874,
+         -3.9329]], grad_fn=&lt;CopySlices&gt;)}]
+</pre></div>
+</div>
+<p>As we see the output contains a list of dictionaries.
+The output list is of length batch_size.
+We currently have just a single image so length of list is 1.
+Each entry in the list corresponds to an input image,
+and it is a dict with keys <cite>boxes</cite>, <cite>labels</cite>, <cite>scores</cite>, <cite>keypoints</cite> and <cite>keypoint_scores</cite>.
+Each value associated to those keys has <cite>num_instances</cite> elements in it.
+In our case above there are 2 instances detected in the image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">kpts</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">outputs</span></a><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;keypoints&#39;</span><span class="p">]</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">scores</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">outputs</span></a><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;scores&#39;</span><span class="p">]</span>
+
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">kpts</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">scores</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>tensor([[[208.0176, 214.2408,   1.0000],
+         [208.0176, 207.0375,   1.0000],
+         [197.8246, 210.6392,   1.0000],
+         [208.0176, 211.8398,   1.0000],
+         [178.6378, 217.8425,   1.0000],
+         [221.2086, 253.8590,   1.0000],
+         [160.6502, 269.4662,   1.0000],
+         [243.9929, 304.2822,   1.0000],
+         [138.4654, 328.8935,   1.0000],
+         [277.5698, 340.8990,   1.0000],
+         [153.4551, 374.5144,   1.0000],
+         [226.0053, 375.7150,   1.0000],
+         [226.0053, 370.3125,   1.0000],
+         [221.8082, 455.5516,   1.0000],
+         [273.9723, 448.9486,   1.0000],
+         [193.6275, 546.1932,   1.0000],
+         [273.3727, 545.5930,   1.0000]],
+
+        [[207.8327, 214.6636,   1.0000],
+         [207.2343, 207.4622,   1.0000],
+         [198.2590, 209.8627,   1.0000],
+         [208.4310, 210.4628,   1.0000],
+         [178.5134, 218.2642,   1.0000],
+         [219.7997, 251.8704,   1.0000],
+         [162.3579, 269.2736,   1.0000],
+         [245.5289, 304.6800,   1.0000],
+         [138.4238, 330.4848,   1.0000],
+         [278.4382, 346.0876,   1.0000],
+         [153.3826, 374.8929,   1.0000],
+         [233.5618, 368.2917,   1.0000],
+         [225.7832, 367.6916,   1.0000],
+         [289.8069, 357.4897,   1.0000],
+         [245.5289, 389.8956,   1.0000],
+         [281.4300, 349.0882,   1.0000],
+         [209.0294, 389.8956,   1.0000]]], grad_fn=&lt;CopySlices&gt;)
+tensor([0.9998, 0.1070], grad_fn=&lt;IndexBackward0&gt;)
+</pre></div>
+</div>
+<p>The KeypointRCNN model detects there are two instances in the image.
+If you plot the boxes by using <code class="xref py py-func docutils literal notranslate"><span class="pre">draw_bounding_boxes()</span></code>
+you would recognize they are the person and the surfboard.
+If we look at the scores, we will realize that the model is much more confident about the person than surfboard.
+We could now set a threshold confidence and plot instances which we are confident enough.
+Let us set a threshold of 0.75 and filter out the keypoints corresponding to the person.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">detect_threshold</span></a> <span class="o">=</span> <span class="mf">0.75</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">idx</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.where.html#torch.where" title="torch.where" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">where</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">scores</span></a> <span class="o">&gt;</span> <a href="https://docs.python.org/3/library/functions.html#float" title="builtins.float" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">detect_threshold</span></a><span class="p">)</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">keypoints</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">kpts</span></a><span class="p">[</span><a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">idx</span></a><span class="p">]</span>
+
+<span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">keypoints</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>tensor([[[208.0176, 214.2408,   1.0000],
+         [208.0176, 207.0375,   1.0000],
+         [197.8246, 210.6392,   1.0000],
+         [208.0176, 211.8398,   1.0000],
+         [178.6378, 217.8425,   1.0000],
+         [221.2086, 253.8590,   1.0000],
+         [160.6502, 269.4662,   1.0000],
+         [243.9929, 304.2822,   1.0000],
+         [138.4654, 328.8935,   1.0000],
+         [277.5698, 340.8990,   1.0000],
+         [153.4551, 374.5144,   1.0000],
+         [226.0053, 375.7150,   1.0000],
+         [226.0053, 370.3125,   1.0000],
+         [221.8082, 455.5516,   1.0000],
+         [273.9723, 448.9486,   1.0000],
+         [193.6275, 546.1932,   1.0000],
+         [273.3727, 545.5930,   1.0000]]], grad_fn=&lt;IndexBackward0&gt;)
+</pre></div>
+</div>
+<p>Great, now we have the keypoints corresponding to the person.
+Each keypoint is represented by x, y coordinates and the visibility.
+We can now use the <a class="reference internal" href="../../generated/torchvision.utils.draw_keypoints.html#torchvision.utils.draw_keypoints" title="torchvision.utils.draw_keypoints"><code class="xref py py-func docutils literal notranslate"><span class="pre">draw_keypoints()</span></code></a> function to draw keypoints.
+Note that the utility expects uint8 images.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">draw_keypoints</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res</span></a> <span class="o">=</span> <span class="n">draw_keypoints</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">person_int</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">keypoints</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">colors</span></a><span class="o">=</span><span class="s2">&quot;blue&quot;</span><span class="p">,</span> <span class="n">radius</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+<span class="n">show</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_011.png" srcset="../../_images/sphx_glr_plot_visualization_utils_011.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><p>As we see the keypoints appear as colored circles over the image.
+The coco keypoints for a person are ordered and represent the following list.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">coco_keypoints</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;nose&quot;</span><span class="p">,</span> <span class="s2">&quot;left_eye&quot;</span><span class="p">,</span> <span class="s2">&quot;right_eye&quot;</span><span class="p">,</span> <span class="s2">&quot;left_ear&quot;</span><span class="p">,</span> <span class="s2">&quot;right_ear&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;left_shoulder&quot;</span><span class="p">,</span> <span class="s2">&quot;right_shoulder&quot;</span><span class="p">,</span> <span class="s2">&quot;left_elbow&quot;</span><span class="p">,</span> <span class="s2">&quot;right_elbow&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;left_wrist&quot;</span><span class="p">,</span> <span class="s2">&quot;right_wrist&quot;</span><span class="p">,</span> <span class="s2">&quot;left_hip&quot;</span><span class="p">,</span> <span class="s2">&quot;right_hip&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;left_knee&quot;</span><span class="p">,</span> <span class="s2">&quot;right_knee&quot;</span><span class="p">,</span> <span class="s2">&quot;left_ankle&quot;</span><span class="p">,</span> <span class="s2">&quot;right_ankle&quot;</span><span class="p">,</span>
+<span class="p">]</span>
+</pre></div>
+</div>
+<p>What if we are interested in joining the keypoints?
+This is especially useful in creating pose detection or action recognition.
+We can join the keypoints easily using the <cite>connectivity</cite> parameter.
+A close observation would reveal that we would need to join the points in below
+order to construct human skeleton.</p>
+<p>nose -&gt; left_eye -&gt; left_ear.                              (0, 1), (1, 3)</p>
+<p>nose -&gt; right_eye -&gt; right_ear.                            (0, 2), (2, 4)</p>
+<p>nose -&gt; left_shoulder -&gt; left_elbow -&gt; left_wrist.         (0, 5), (5, 7), (7, 9)</p>
+<p>nose -&gt; right_shoulder -&gt; right_elbow -&gt; right_wrist.      (0, 6), (6, 8), (8, 10)</p>
+<p>left_shoulder -&gt; left_hip -&gt; left_knee -&gt; left_ankle.      (5, 11), (11, 13), (13, 15)</p>
+<p>right_shoulder -&gt; right_hip -&gt; right_knee -&gt; right_ankle.  (6, 12), (12, 14), (14, 16)</p>
+<p>We will create a list containing these keypoint ids to be connected.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">connect_skeleton</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">2</span><span class="p">),</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">),</span> <span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">6</span><span class="p">),</span> <span class="p">(</span><span class="mi">5</span><span class="p">,</span> <span class="mi">7</span><span class="p">),</span> <span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">8</span><span class="p">),</span>
+    <span class="p">(</span><span class="mi">7</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="p">(</span><span class="mi">8</span><span class="p">,</span> <span class="mi">10</span><span class="p">),</span> <span class="p">(</span><span class="mi">5</span><span class="p">,</span> <span class="mi">11</span><span class="p">),</span> <span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">12</span><span class="p">),</span> <span class="p">(</span><span class="mi">11</span><span class="p">,</span> <span class="mi">13</span><span class="p">),</span> <span class="p">(</span><span class="mi">12</span><span class="p">,</span> <span class="mi">14</span><span class="p">),</span> <span class="p">(</span><span class="mi">13</span><span class="p">,</span> <span class="mi">15</span><span class="p">),</span> <span class="p">(</span><span class="mi">14</span><span class="p">,</span> <span class="mi">16</span><span class="p">)</span>
+<span class="p">]</span>
+</pre></div>
+</div>
+<p>We pass the above list to the connectivity parameter to connect the keypoints.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res</span></a> <span class="o">=</span> <span class="n">draw_keypoints</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">person_int</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">keypoints</span></a><span class="p">,</span> <span class="n">connectivity</span><span class="o">=</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">connect_skeleton</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">colors</span></a><span class="o">=</span><span class="s2">&quot;blue&quot;</span><span class="p">,</span> <span class="n">radius</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">width</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
+<span class="n">show</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">res</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_visualization_utils_012.png" srcset="../../_images/sphx_glr_plot_visualization_utils_012.png" alt="plot visualization utils" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 14.555 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-others-plot-visualization-utils-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/27d6228101da5d268ad41f8f06010df8/plot_visualization_utils.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_visualization_utils.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/95e7320166df5d4ddcbdd5ea64a5c98b/plot_visualization_utils.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_visualization_utils.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../../training_references.html" class="btn btn-neutral float-right" title="Training references" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_video_api.html" class="btn btn-neutral" title="Video API" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Visualization utilities</a><ul>
+<li><a class="reference internal" href="#visualizing-a-grid-of-images">Visualizing a grid of images</a></li>
+<li><a class="reference internal" href="#visualizing-bounding-boxes">Visualizing bounding boxes</a></li>
+<li><a class="reference internal" href="#visualizing-segmentation-masks">Visualizing segmentation masks</a><ul>
+<li><a class="reference internal" href="#semantic-segmentation-models">Semantic segmentation models</a></li>
+<li><a class="reference internal" href="#instance-segmentation-models">Instance segmentation models</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#visualizing-keypoints">Visualizing keypoints</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/others/sg_execution_times.html b/0.16/auto_examples/others/sg_execution_times.html
new file mode 100644
index 00000000000..e8ca1632633
--- /dev/null
+++ b/0.16/auto_examples/others/sg_execution_times.html
@@ -0,0 +1,700 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Computation times &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Computation times</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/others/sg_execution_times.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="computation-times">
+<span id="sphx-glr-auto-examples-others-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this heading">¶</a></h1>
+<p><strong>00:34.118</strong> total execution time for <strong>auto_examples_others</strong> files:</p>
+<table class="docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_visualization_utils.py</span></code>)</p></td>
+<td><p>00:14.555</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_optical_flow.py</span></code>)</p></td>
+<td><p>00:10.096</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="plot_video_api.html#sphx-glr-auto-examples-others-plot-video-api-py"><span class="std std-ref">Video API</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_video_api.py</span></code>)</p></td>
+<td><p>00:05.277</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_repurposing_annotations.py</span></code>)</p></td>
+<td><p>00:02.507</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_scripted_tensor_transforms.py</span></code>)</p></td>
+<td><p>00:01.683</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Computation times</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/index.html b/0.16/auto_examples/transforms/index.html
new file mode 100644
index 00000000000..b664ea1e085
--- /dev/null
+++ b/0.16/auto_examples/transforms/index.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Transforms &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Getting started with transforms v2" href="plot_transforms_getting_started.html" />
+    <link rel="prev" title="Examples and tutorials" href="../index.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+      <li>Transforms</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/index.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="transforms">
+<span id="transforms-gallery"></span><span id="sphx-glr-auto-examples-transforms"></span><h1>Transforms<a class="headerlink" href="#transforms" title="Permalink to this heading">¶</a></h1>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div></div><div class="toctree-wrapper compound">
+</div>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_transforms_getting_started.html" class="btn btn-neutral float-right" title="Getting started with transforms v2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../index.html" class="btn btn-neutral" title="Examples and tutorials" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Transforms</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/plot_custom_transforms.html b/0.16/auto_examples/transforms/plot_custom_transforms.html
new file mode 100644
index 00000000000..4e0b9b6287d
--- /dev/null
+++ b/0.16/auto_examples/transforms/plot_custom_transforms.html
@@ -0,0 +1,827 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>How to write your own v2 transforms &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="TVTensors FAQ" href="plot_tv_tensors.html" />
+    <link rel="prev" title="How to use CutMix and MixUp" href="plot_cutmix_mixup.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Transforms</a> &gt;</li>
+        
+      <li>How to write your own v2 transforms</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/plot_custom_transforms.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="how-to-write-your-own-v2-transforms">
+<span id="sphx-glr-auto-examples-transforms-plot-custom-transforms-py"></span><h1>How to write your own v2 transforms<a class="headerlink" href="#how-to-write-your-own-v2-transforms" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_transforms.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>This guide explains how to write transforms that are compatible with the
+torchvision transforms V2 API.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+</pre></div>
+</div>
+<section id="just-create-a-nn-module-and-override-the-forward-method">
+<h2>Just create a <code class="docutils literal notranslate"><span class="pre">nn.Module</span></code> and override the <code class="docutils literal notranslate"><span class="pre">forward</span></code> method<a class="headerlink" href="#just-create-a-nn-module-and-override-the-forward-method" title="Permalink to this heading">¶</a></h2>
+<p>In most cases, this is all you’re going to need, as long as you already know
+the structure of the input that your transform will expect. For example if
+you’re just doing image classification, your transform will typically accept a
+single image as input, or a <code class="docutils literal notranslate"><span class="pre">(img,</span> <span class="pre">label)</span></code> input. So you can just hard-code
+your <code class="docutils literal notranslate"><span class="pre">forward</span></code> method to accept just that, e.g.</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">MyCustomTransform</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span></a><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a><span class="p">):</span>
+        <span class="c1"># Do some transformations</span>
+        <span class="k">return</span> <span class="n">new_img</span><span class="p">,</span> <span class="n">new_label</span>
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This means that if you have a custom transform that is already compatible
+with the V1 transforms (those in <code class="docutils literal notranslate"><span class="pre">torchvision.transforms</span></code>), it will
+still work with the V2 transforms without any change!</p>
+</div>
+<p>We will illustrate this more completely below with a typical detection case,
+where our samples are just images, bounding boxes and labels:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">MyCustomTransform</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span></a><span class="p">):</span>
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <span class="n">bboxes</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a><span class="p">):</span>  <span class="c1"># we assume inputs are always structured like this</span>
+        <span class="nb">print</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;I&#39;m transforming an image of shape </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s2"> &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;with bboxes = </span><span class="si">{</span><span class="n">bboxes</span><span class="si">}</span><span class="se">\n</span><span class="si">{</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+        <span class="c1"># Do some transformations. Here, we&#39;re just passing though the input</span>
+        <span class="k">return</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <span class="n">bboxes</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a>
+
+
+<span class="n">transforms</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Compose</span></a><span class="p">([</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">MyCustomTransform</span></a><span class="p">(),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomResizedCrop</span></a><span class="p">((</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Normalize</span></a><span class="p">(</span><span class="n">mean</span><span class="o">=</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">std</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">])</span>
+<span class="p">])</span>
+
+<a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">H</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">W</span></a> <span class="o">=</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.rand.html#torch.rand" title="torch.rand" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">rand</span></a><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">H</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">W</span></a><span class="p">)</span>
+<span class="n">bboxes</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">(</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.tensor.html#torch.tensor" title="torch.tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span></a><span class="p">([[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">20</span><span class="p">],</span> <span class="p">[</span><span class="mi">50</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">70</span><span class="p">,</span> <span class="mi">70</span><span class="p">]]),</span>
+    <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;XYXY&quot;</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="o">=</span><span class="p">(</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">H</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">W</span></a><span class="p">)</span>
+<span class="p">)</span>
+<a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a> <span class="o">=</span> <span class="mi">3</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_img</span></a><span class="p">,</span> <span class="n">out_bboxes</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_label</span></a> <span class="o">=</span> <span class="n">transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <span class="n">bboxes</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>I&#39;m transforming an image of shape torch.Size([3, 256, 256]) with bboxes = BoundingBoxes([[ 0, 10, 10, 20],
+               [50, 50, 70, 70]], format=BoundingBoxFormat.XYXY, canvas_size=(256, 256))
+label = 3
+</pre></div>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Output image shape: </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_img</span></a><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="se">\n</span><span class="s2">out_bboxes = </span><span class="si">{</span><span class="n">out_bboxes</span><span class="si">}</span><span class="se">\n</span><span class="si">{</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_label</span></a><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Output image shape: torch.Size([3, 224, 224])
+out_bboxes = BoundingBoxes([[224,   0, 224,   0],
+               [136,   0, 173,   0]], format=BoundingBoxFormat.XYXY, canvas_size=(224, 224))
+out_label = 3
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>While working with TVTensor classes in your code, make sure to
+familiarize yourself with this section:
+<a class="reference internal" href="plot_tv_tensors.html#tv-tensor-unwrapping-behaviour"><span class="std std-ref">I had a TVTensor but now I have a Tensor. Help!</span></a></p>
+</div>
+</section>
+<section id="supporting-arbitrary-input-structures">
+<h2>Supporting arbitrary input structures<a class="headerlink" href="#supporting-arbitrary-input-structures" title="Permalink to this heading">¶</a></h2>
+<p>In the section above, we have assumed that you already know the structure of
+your inputs and that you’re OK with hard-coding this expected structure in
+your code. If you want your custom transforms to be as flexible as possible,
+this can be a bit limiting.</p>
+<p>A key feature of the builtin Torchvision V2 transforms is that they can accept
+arbitrary input structure and return the same structure as output (with
+transformed entries). For example, transforms can accept a single image, or a
+tuple of <code class="docutils literal notranslate"><span class="pre">(img,</span> <span class="pre">label)</span></code>, or an arbitrary nested dictionary as input:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">structured_input</span></a> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;img&quot;</span><span class="p">:</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span>
+    <span class="s2">&quot;annotations&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">bboxes</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a><span class="p">),</span>
+    <span class="s2">&quot;something_that_will_be_ignored&quot;</span><span class="p">:</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="s2">&quot;hello&quot;</span><span class="p">)</span>
+<span class="p">}</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">structured_output</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mi">1</span><span class="p">)(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">structured_input</span></a><span class="p">)</span>
+
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">structured_output</span></a><span class="p">,</span> <span class="nb">dict</span><span class="p">)</span>
+<span class="k">assert</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">structured_output</span></a><span class="p">[</span><span class="s2">&quot;something_that_will_be_ignored&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="s2">&quot;hello&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The transformed bboxes are:</span><span class="se">\n</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">structured_output</span></a><span class="p">[</span><span class="s1">&#39;annotations&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>The transformed bboxes are:
+BoundingBoxes([[246,  10, 256,  20],
+               [186,  50, 206,  70]], format=BoundingBoxFormat.XYXY, canvas_size=(256, 256))
+</pre></div>
+</div>
+<p>If you want to reproduce this behavior in your own transform, we invite you to
+look at our <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/_transform.py">code</a>
+and adapt it to your needs.</p>
+<p>In brief, the core logic is to unpack the input into a flat list using <a class="reference external" href="https://github.com/pytorch/pytorch/blob/main/torch/utils/_pytree.py">pytree</a>, and
+then transform only the entries that can be transformed (the decision is made
+based on the <strong>class</strong> of the entries, as all TVTensors are
+tensor-subclasses) plus some custom logic that is out of score here - check the
+code for details. The (potentially transformed) entries are then repacked and
+returned, in the same structure as the input.</p>
+<p>We do not provide public dev-facing tools to achieve that at this time, but if
+this is something that would be valuable to you, please let us know by opening
+an issue on our <a class="reference external" href="https://github.com/pytorch/vision/issues">GitHub repo</a>.</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 0.006 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-transforms-plot-custom-transforms-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/163bc3a8b6281b990b458e3eef431d01/plot_custom_transforms.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_custom_transforms.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/3f3cbe5cc8b5758610d0ab95995b0b8c/plot_custom_transforms.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_custom_transforms.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_tv_tensors.html" class="btn btn-neutral float-right" title="TVTensors FAQ" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_cutmix_mixup.html" class="btn btn-neutral" title="How to use CutMix and MixUp" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">How to write your own v2 transforms</a><ul>
+<li><a class="reference internal" href="#just-create-a-nn-module-and-override-the-forward-method">Just create a <code class="docutils literal notranslate"><span class="pre">nn.Module</span></code> and override the <code class="docutils literal notranslate"><span class="pre">forward</span></code> method</a></li>
+<li><a class="reference internal" href="#supporting-arbitrary-input-structures">Supporting arbitrary input structures</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/plot_custom_tv_tensors.html b/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
new file mode 100644
index 00000000000..22cc922c0e2
--- /dev/null
+++ b/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
@@ -0,0 +1,810 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>How to write your own TVTensor class &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Others" href="../others/index.html" />
+    <link rel="prev" title="TVTensors FAQ" href="plot_tv_tensors.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Transforms</a> &gt;</li>
+        
+      <li>How to write your own TVTensor class</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/plot_custom_tv_tensors.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="how-to-write-your-own-tvtensor-class">
+<span id="sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"></span><h1>How to write your own TVTensor class<a class="headerlink" href="#how-to-write-your-own-tvtensor-class" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_custom_tv_tensors.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>This guide is intended for advanced users and downstream library maintainers. We explain how to
+write your own TVTensor class, and how to make it compatible with the built-in
+Torchvision v2 transforms. Before continuing, make sure you have read
+<a class="reference internal" href="plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+</pre></div>
+</div>
+<p>We will create a very simple class that just inherits from the base
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a> class. It will be enough to cover
+what you need to know to implement your more elaborate uses-cases. If you need
+to create a class that carries meta-data, take a look at how the
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> class is <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/tv_tensors/_bounding_box.py">implemented</a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">MyTVTensor</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">TVTensor</span></a><span class="p">):</span>
+    <span class="k">pass</span>
+
+
+<span class="n">my_dp</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">MyTVTensor</span></a><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">])</span>
+<span class="n">my_dp</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>MyTVTensor([1., 2., 3.])
+</pre></div>
+</div>
+<p>Now that we have defined our custom TVTensor class, we want it to be
+compatible with the built-in torchvision transforms, and the functional API.
+For that, we need to implement a kernel which performs the core of the
+transformation, and then “hook” it to the functional that we want to support
+via <a class="reference internal" href="../../generated/torchvision.transforms.v2.functional.register_kernel.html#torchvision.transforms.v2.functional.register_kernel" title="torchvision.transforms.v2.functional.register_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">register_kernel()</span></code></a>.</p>
+<p>We illustrate this process below: we create a kernel for the “horizontal flip”
+operation of our MyTVTensor class, and register it to the functional API.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.transforms.v2</span> <span class="kn">import</span> <span class="n">functional</span> <span class="k">as</span> <span class="n">F</span>
+
+
+<span class="nd">@F</span><span class="o">.</span><span class="n">register_kernel</span><span class="p">(</span><span class="n">functional</span><span class="o">=</span><span class="s2">&quot;hflip&quot;</span><span class="p">,</span> <span class="n">tv_tensor_cls</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">MyTVTensor</span></a><span class="p">)</span>
+<span class="k">def</span> <span class="nf">hflip_my_tv_tensor</span><span class="p">(</span><span class="n">my_dp</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+    <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Flipping!&quot;</span><span class="p">)</span>
+    <span class="n">out</span> <span class="o">=</span> <span class="n">my_dp</span><span class="o">.</span><span class="n">flip</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">my_dp</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>To understand why <a class="reference internal" href="../../generated/torchvision.tv_tensors.wrap.html#torchvision.tv_tensors.wrap" title="torchvision.tv_tensors.wrap"><code class="xref py py-func docutils literal notranslate"><span class="pre">wrap()</span></code></a> is used, see
+<a class="reference internal" href="plot_tv_tensors.html#tv-tensor-unwrapping-behaviour"><span class="std std-ref">I had a TVTensor but now I have a Tensor. Help!</span></a>. Ignore the <code class="docutils literal notranslate"><span class="pre">*args,</span> <span class="pre">**kwargs</span></code> for now,
+we will explain it below in <a class="reference internal" href="#param-forwarding"><span class="std std-ref">Parameter forwarding, and ensuring future compatibility of your kernels</span></a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In our call to <code class="docutils literal notranslate"><span class="pre">register_kernel</span></code> above we used a string
+<code class="docutils literal notranslate"><span class="pre">functional=&quot;hflip&quot;</span></code> to refer to the functional we want to hook into. We
+could also have used the  functional <em>itself</em>, i.e.
+<code class="docutils literal notranslate"><span class="pre">&#64;register_kernel(functional=F.hflip,</span> <span class="pre">...)</span></code>.</p>
+</div>
+<p>Now that we have registered our kernel, we can call the functional API on a
+<code class="docutils literal notranslate"><span class="pre">MyTVTensor</span></code> instance:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">my_dp</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">MyTVTensor</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/generated/torch.rand.html#torch.rand" title="torch.rand" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">rand</span></a><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">))</span>
+<span class="n">_</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">hflip</span><span class="p">(</span><span class="n">my_dp</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Flipping!
+</pre></div>
+</div>
+<p>And we can also use the
+<a class="reference internal" href="../../generated/torchvision.transforms.v2.RandomHorizontalFlip.html#torchvision.transforms.v2.RandomHorizontalFlip" title="torchvision.transforms.v2.RandomHorizontalFlip"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomHorizontalFlip</span></code></a> transform, since it relies on <code class="xref py py-func docutils literal notranslate"><span class="pre">hflip()</span></code> internally:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">t</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+<span class="n">_</span> <span class="o">=</span> <span class="n">t</span><span class="p">(</span><span class="n">my_dp</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Flipping!
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>We cannot register a kernel for a transform class, we can only register a
+kernel for a <strong>functional</strong>. The reason we can’t register a transform
+class is because one transform may internally rely on more than one
+functional, so in general we can’t register a single kernel for a given
+class.</p>
+</div>
+<section id="parameter-forwarding-and-ensuring-future-compatibility-of-your-kernels">
+<span id="param-forwarding"></span><h2>Parameter forwarding, and ensuring future compatibility of your kernels<a class="headerlink" href="#parameter-forwarding-and-ensuring-future-compatibility-of-your-kernels" title="Permalink to this heading">¶</a></h2>
+<p>The functional API that you’re hooking into is public and therefore
+<strong>backward</strong> compatible: we guarantee that the parameters of these functionals
+won’t be removed or renamed without a proper deprecation cycle. However, we
+don’t guarantee <strong>forward</strong> compatibility, and we may add new parameters in
+the future.</p>
+<p>Imagine that in a future version, Torchvision adds a new <code class="docutils literal notranslate"><span class="pre">inplace</span></code> parameter
+to its <code class="xref py py-func docutils literal notranslate"><span class="pre">hflip()</span></code> functional. If you
+already defined and registered your own kernel as</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">hflip_my_tv_tensor</span><span class="p">(</span><span class="n">my_dp</span><span class="p">):</span>  <span class="c1"># noqa</span>
+    <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Flipping!&quot;</span><span class="p">)</span>
+    <span class="n">out</span> <span class="o">=</span> <span class="n">my_dp</span><span class="o">.</span><span class="n">flip</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">my_dp</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>then calling <code class="docutils literal notranslate"><span class="pre">F.hflip(my_dp)</span></code> will <strong>fail</strong>, because <code class="docutils literal notranslate"><span class="pre">hflip</span></code> will try to
+pass the new <code class="docutils literal notranslate"><span class="pre">inplace</span></code> parameter to your kernel, but your kernel doesn’t
+accept it.</p>
+<p>For this reason, we recommend to always define your kernels with
+<code class="docutils literal notranslate"><span class="pre">*args,</span> <span class="pre">**kwargs</span></code> in their signature, as done above. This way, your kernel
+will be able to accept any new parameter that we may add in the future.
+(Technically, adding <cite>**kwargs</cite> only should be enough).</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 0.004 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-transforms-plot-custom-tv-tensors-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/4890c38ddf15e56f86ebbdd480cb7ade/plot_custom_tv_tensors.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_custom_tv_tensors.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/8b062cf0fa8a991b8cff18fa361c8ffe/plot_custom_tv_tensors.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_custom_tv_tensors.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../others/index.html" class="btn btn-neutral float-right" title="Others" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_tv_tensors.html" class="btn btn-neutral" title="TVTensors FAQ" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">How to write your own TVTensor class</a><ul>
+<li><a class="reference internal" href="#parameter-forwarding-and-ensuring-future-compatibility-of-your-kernels">Parameter forwarding, and ensuring future compatibility of your kernels</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/plot_cutmix_mixup.html b/0.16/auto_examples/transforms/plot_cutmix_mixup.html
new file mode 100644
index 00000000000..c9df80ab8db
--- /dev/null
+++ b/0.16/auto_examples/transforms/plot_cutmix_mixup.html
@@ -0,0 +1,859 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>How to use CutMix and MixUp &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="How to write your own v2 transforms" href="plot_custom_transforms.html" />
+    <link rel="prev" title="Transforms v2: End-to-end object detection/segmentation example" href="plot_transforms_e2e.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Transforms</a> &gt;</li>
+        
+      <li>How to use CutMix and MixUp</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/plot_cutmix_mixup.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="how-to-use-cutmix-and-mixup">
+<span id="sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"></span><h1>How to use CutMix and MixUp<a class="headerlink" href="#how-to-use-cutmix-and-mixup" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_cutmix_mixup.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p><a class="reference internal" href="../../generated/torchvision.transforms.v2.CutMix.html#torchvision.transforms.v2.CutMix" title="torchvision.transforms.v2.CutMix"><code class="xref py py-class docutils literal notranslate"><span class="pre">CutMix</span></code></a> and
+<a class="reference internal" href="../../generated/torchvision.transforms.v2.MixUp.html#torchvision.transforms.v2.MixUp" title="torchvision.transforms.v2.MixUp"><code class="xref py py-class docutils literal notranslate"><span class="pre">MixUp</span></code></a> are popular augmentation strategies
+that can improve classification accuracy.</p>
+<p>These transforms are slightly different from the rest of the Torchvision
+transforms, because they expect
+<strong>batches</strong> of samples as input, not individual images. In this example we’ll
+explain how to use them: after the <code class="docutils literal notranslate"><span class="pre">DataLoader</span></code>, or as part of a collation
+function.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision.datasets</span> <span class="kn">import</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.Dataset" title="torch.utils.data.Dataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">FakeData</span></a>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+
+
+<a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NUM_CLASSES</span></a> <span class="o">=</span> <span class="mi">100</span>
+</pre></div>
+</div>
+<section id="pre-processing-pipeline">
+<h2>Pre-processing pipeline<a class="headerlink" href="#pre-processing-pipeline" title="Permalink to this heading">¶</a></h2>
+<p>We’ll use a simple but typical image classification pipeline:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">preproc</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Compose</span></a><span class="p">([</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">PILToTensor</span></a><span class="p">(),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomResizedCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">ToDtype</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">float32</span></a><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>  <span class="c1"># to float32 in [0, 1]</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Normalize</span></a><span class="p">(</span><span class="n">mean</span><span class="o">=</span><span class="p">(</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">),</span> <span class="n">std</span><span class="o">=</span><span class="p">(</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">)),</span>  <span class="c1"># typically from ImageNet</span>
+<span class="p">])</span>
+
+<span class="n">dataset</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.Dataset" title="torch.utils.data.Dataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">FakeData</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span> <span class="n">num_classes</span><span class="o">=</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NUM_CLASSES</span></a><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">preproc</span><span class="p">)</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a> <span class="o">=</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span><span class="o">.</span><span class="n">dtype</span></a><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">label</span></a><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>type(img) = &lt;class &#39;torch.Tensor&#39;&gt;, img.dtype = torch.float32, img.shape = torch.Size([3, 224, 224]), label = 67
+</pre></div>
+</div>
+<p>One important thing to note is that neither CutMix nor MixUp are part of this
+pre-processing pipeline. We’ll add them a bit later once we define the
+DataLoader. Just as a refresher, this is what the DataLoader and training loop
+would look like if we weren’t using CutMix or MixUp:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torch.utils.data</span> <span class="kn">import</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">DataLoader</span></a>
+
+<a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dataloader</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">DataLoader</span></a><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+<span class="k">for</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a> <span class="ow">in</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dataloader</span></a><span class="p">:</span>
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="nb">print</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span><span class="o">.</span><span class="n">dtype</span></a><span class="p">)</span>
+    <span class="c1"># &lt;rest of the training loop here&gt;</span>
+    <span class="k">break</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4])
+torch.int64
+</pre></div>
+</div>
+</section>
+<section id="where-to-use-mixup-and-cutmix">
+<h2>Where to use MixUp and CutMix<a class="headerlink" href="#where-to-use-mixup-and-cutmix" title="Permalink to this heading">¶</a></h2>
+<section id="after-the-dataloader">
+<h3>After the DataLoader<a class="headerlink" href="#after-the-dataloader" title="Permalink to this heading">¶</a></h3>
+<p>Now let’s add CutMix and MixUp. The simplest way to do this right after the
+DataLoader: the Dataloader has already batched the images and labels for us,
+and this is exactly what these transforms expect as input:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dataloader</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">DataLoader</span></a><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+<span class="n">cutmix</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">CutMix</span></a><span class="p">(</span><span class="n">num_classes</span><span class="o">=</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NUM_CLASSES</span></a><span class="p">)</span>
+<span class="n">mixup</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">MixUp</span></a><span class="p">(</span><span class="n">num_classes</span><span class="o">=</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NUM_CLASSES</span></a><span class="p">)</span>
+<span class="n">cutmix_or_mixup</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomChoice</span></a><span class="p">([</span><span class="n">cutmix</span><span class="p">,</span> <span class="n">mixup</span><span class="p">])</span>
+
+<span class="k">for</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a> <span class="ow">in</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dataloader</span></a><span class="p">:</span>
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Before CutMix/MixUp: </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+    <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a> <span class="o">=</span> <span class="n">cutmix_or_mixup</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a><span class="p">)</span>
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;After CutMix/MixUp: </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="c1"># &lt;rest of the training loop here&gt;</span>
+    <span class="k">break</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Before CutMix/MixUp: images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4])
+After CutMix/MixUp: images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4, 100])
+</pre></div>
+</div>
+<p>Note how the labels were also transformed: we went from a batched label of
+shape (batch_size,) to a tensor of shape (batch_size, num_classes). The
+transformed labels can still be passed as-is to a loss function like
+<a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.functional.cross_entropy.html#torch.nn.functional.cross_entropy" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.nn.functional.cross_entropy()</span></code></a>.</p>
+</section>
+<section id="as-part-of-the-collation-function">
+<h3>As part of the collation function<a class="headerlink" href="#as-part-of-the-collation-function" title="Permalink to this heading">¶</a></h3>
+<p>Passing the transforms after the DataLoader is the simplest way to use CutMix
+and MixUp, but one disadvantage is that it does not take advantage of the
+DataLoader multi-processing. For that, we can pass those transforms as part of
+the collation function (refer to the <a class="reference external" href="https://pytorch.org/docs/stable/data.html#dataloader-collate-fn">PyTorch docs</a> to learn
+more about collation).</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torch.utils.data</span> <span class="kn">import</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.default_collate" title="torch.utils.data.default_collate" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-function"><span class="n">default_collate</span></a>
+
+
+<span class="k">def</span> <span class="nf">collate_fn</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">):</span>
+    <span class="k">return</span> <span class="n">cutmix_or_mixup</span><span class="p">(</span><span class="o">*</span><a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.default_collate" title="torch.utils.data.default_collate" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-function"><span class="n">default_collate</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">))</span>
+
+
+<a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dataloader</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">DataLoader</span></a><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">num_workers</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">collate_fn</span><span class="o">=</span><span class="n">collate_fn</span><span class="p">)</span>
+
+<span class="k">for</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a> <span class="ow">in</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">dataloader</span></a><span class="p">:</span>
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">images</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="c1"># No need to call cutmix_or_mixup, it&#39;s already been called as part of the DataLoader!</span>
+    <span class="c1"># &lt;rest of the training loop here&gt;</span>
+    <span class="k">break</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>images.shape = torch.Size([4, 3, 224, 224]), labels.shape = torch.Size([4, 100])
+</pre></div>
+</div>
+</section>
+</section>
+<section id="non-standard-input-format">
+<h2>Non-standard input format<a class="headerlink" href="#non-standard-input-format" title="Permalink to this heading">¶</a></h2>
+<p>So far we’ve used a typical sample structure where we pass <code class="docutils literal notranslate"><span class="pre">(images,</span>
+<span class="pre">labels)</span></code> as inputs. MixUp and CutMix will magically work by default with most
+common sample structures: tuples where the second parameter is a tensor label,
+or dict with a “label[s]” key. Look at the documentation of the
+<code class="docutils literal notranslate"><span class="pre">labels_getter</span></code> parameter for more details.</p>
+<p>If your samples have a different structure, you can still use CutMix and MixUp
+by passing a callable to the <code class="docutils literal notranslate"><span class="pre">labels_getter</span></code> parameter. For example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;imgs&quot;</span><span class="p">:</span> <a href="https://pytorch.org/docs/stable/generated/torch.rand.html#torch.rand" title="torch.rand" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">rand</span></a><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span>
+    <span class="s2">&quot;target&quot;</span><span class="p">:</span> <span class="p">{</span>
+        <span class="s2">&quot;classes&quot;</span><span class="p">:</span> <a href="https://pytorch.org/docs/stable/generated/torch.randint.html#torch.randint" title="torch.randint" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">randint</span></a><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NUM_CLASSES</span></a><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">4</span><span class="p">,)),</span>
+        <span class="s2">&quot;some_other_key&quot;</span><span class="p">:</span> <span class="s2">&quot;this is going to be passed-through&quot;</span>
+    <span class="p">}</span>
+<span class="p">}</span>
+
+
+<span class="k">def</span> <span class="nf">labels_getter</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">):</span>
+    <span class="k">return</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">[</span><span class="s2">&quot;target&quot;</span><span class="p">][</span><span class="s2">&quot;classes&quot;</span><span class="p">]</span>
+
+
+<a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">CutMix</span></a><span class="p">(</span><span class="n">num_classes</span><span class="o">=</span><a href="https://docs.python.org/3/library/functions.html#int" title="builtins.int" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NUM_CLASSES</span></a><span class="p">,</span> <span class="n">labels_getter</span><span class="o">=</span><span class="n">labels_getter</span><span class="p">)(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">batch</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out</span></a><span class="p">[</span><span class="s1">&#39;imgs&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out</span></a><span class="p">[</span><span class="s1">&#39;target&#39;</span><span class="p">][</span><span class="s1">&#39;classes&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>out[&#39;imgs&#39;].shape = torch.Size([4, 3, 224, 224]), out[&#39;target&#39;][&#39;classes&#39;].shape = torch.Size([4, 100])
+</pre></div>
+</div>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 0.155 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-transforms-plot-cutmix-mixup-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/5a9ea474844f53cd409e782bd87efb13/plot_cutmix_mixup.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_cutmix_mixup.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/66c906e53334aa968837db20de95f3b2/plot_cutmix_mixup.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_cutmix_mixup.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_custom_transforms.html" class="btn btn-neutral float-right" title="How to write your own v2 transforms" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_transforms_e2e.html" class="btn btn-neutral" title="Transforms v2: End-to-end object detection/segmentation example" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">How to use CutMix and MixUp</a><ul>
+<li><a class="reference internal" href="#pre-processing-pipeline">Pre-processing pipeline</a></li>
+<li><a class="reference internal" href="#where-to-use-mixup-and-cutmix">Where to use MixUp and CutMix</a><ul>
+<li><a class="reference internal" href="#after-the-dataloader">After the DataLoader</a></li>
+<li><a class="reference internal" href="#as-part-of-the-collation-function">As part of the collation function</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#non-standard-input-format">Non-standard input format</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/plot_transforms_e2e.html b/0.16/auto_examples/transforms/plot_transforms_e2e.html
new file mode 100644
index 00000000000..1e839ea3279
--- /dev/null
+++ b/0.16/auto_examples/transforms/plot_transforms_e2e.html
@@ -0,0 +1,895 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Transforms v2: End-to-end object detection/segmentation example &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="How to use CutMix and MixUp" href="plot_cutmix_mixup.html" />
+    <link rel="prev" title="Illustration of transforms" href="plot_transforms_illustrations.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Transforms</a> &gt;</li>
+        
+      <li>Transforms v2: End-to-end object detection/segmentation example</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/plot_transforms_e2e.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="transforms-v2-end-to-end-object-detection-segmentation-example">
+<span id="sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"></span><h1>Transforms v2: End-to-end object detection/segmentation example<a class="headerlink" href="#transforms-v2-end-to-end-object-detection-segmentation-example" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_e2e.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>Object detection and segmentation tasks are natively supported:
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> enables jointly transforming images, videos,
+bounding boxes, and masks.</p>
+<p>This example showcases an end-to-end instance segmentation training case using
+Torchvision utils from <code class="docutils literal notranslate"><span class="pre">torchvision.datasets</span></code>, <code class="docutils literal notranslate"><span class="pre">torchvision.models</span></code> and
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>. Everything covered here can be applied similarly
+to object detection or semantic segmentation tasks.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">pathlib</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">torch.utils.data</span>
+
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">models</span><span class="p">,</span> <span class="n">datasets</span><span class="p">,</span> <span class="n">tv_tensors</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+
+<a href="https://pytorch.org/docs/stable/generated/torch.manual_seed.html#torch.manual_seed" title="torch.manual_seed" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">manual_seed</span></a><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+<span class="c1"># This loads fake data for illustration purposes of this example. In practice, you&#39;ll have</span>
+<span class="c1"># to replace this with the proper data.</span>
+<span class="c1"># If you&#39;re trying to run that on collab, you can download the assets and the</span>
+<span class="c1"># helpers from https://github.com/pytorch/vision/tree/main/gallery/</span>
+<a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ROOT</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">pathlib</span><span class="o">.</span><span class="n">Path</span></a><span class="p">(</span><span class="s2">&quot;../assets&quot;</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;coco&quot;</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">IMAGES_PATH</span></a> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ROOT</span></a> <span class="o">/</span> <span class="s2">&quot;images&quot;</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ANNOTATIONS_PATH</span></a> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.PosixPath" title="pathlib.PosixPath" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ROOT</span></a> <span class="o">/</span> <span class="s2">&quot;instances.json&quot;</span><span class="p">)</span>
+<span class="kn">from</span> <span class="nn">helpers</span> <span class="kn">import</span> <span class="n">plot</span>
+</pre></div>
+</div>
+<section id="dataset-preparation">
+<h2>Dataset preparation<a class="headerlink" href="#dataset-preparation" title="Permalink to this heading">¶</a></h2>
+<p>We start off by loading the <a class="reference internal" href="../../generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection" title="torchvision.datasets.CocoDetection"><code class="xref py py-class docutils literal notranslate"><span class="pre">CocoDetection</span></code></a> dataset to have a look at what it currently
+returns.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">dataset</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.Dataset" title="torch.utils.data.Dataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">datasets</span><span class="o">.</span><span class="n">CocoDetection</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">IMAGES_PATH</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ANNOTATIONS_PATH</span></a><span class="p">)</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sample</span></a> <span class="o">=</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+<a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sample</span></a>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span><span class="w"> </span><span class="si">= }</span><span class="se">\n</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">)</span><span class="w"> </span><span class="si">= }</span><span class="se">\n</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="w"> </span><span class="si">= }</span><span class="se">\n</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loading annotations into memory...
+Done (t=0.00s)
+creating index...
+index created!
+type(img) = &lt;class &#39;PIL.Image.Image&#39;&gt;
+type(target) = &lt;class &#39;list&#39;&gt;
+type(target[0]) = &lt;class &#39;dict&#39;&gt;
+target[0].keys() = dict_keys([&#39;segmentation&#39;, &#39;iscrowd&#39;, &#39;image_id&#39;, &#39;bbox&#39;, &#39;category_id&#39;, &#39;id&#39;])
+</pre></div>
+</div>
+<p>Torchvision datasets preserve the data structure and types as it was intended
+by the datasets authors. So by default, the output structure may not always be
+compatible with the models or the transforms.</p>
+<p>To overcome that, we can use the
+<a class="reference internal" href="../../generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html#torchvision.datasets.wrap_dataset_for_transforms_v2" title="torchvision.datasets.wrap_dataset_for_transforms_v2"><code class="xref py py-func docutils literal notranslate"><span class="pre">wrap_dataset_for_transforms_v2()</span></code></a> function. For
+<a class="reference internal" href="../../generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection" title="torchvision.datasets.CocoDetection"><code class="xref py py-class docutils literal notranslate"><span class="pre">CocoDetection</span></code></a>, this changes the target
+structure to a single dictionary of lists:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">dataset</span> <span class="o">=</span> <span class="n">datasets</span><span class="o">.</span><span class="n">wrap_dataset_for_transforms_v2</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;boxes&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">,</span> <span class="s2">&quot;masks&quot;</span><span class="p">))</span>
+
+<a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sample</span></a> <span class="o">=</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+<a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a> <span class="o">=</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sample</span></a>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span><span class="w"> </span><span class="si">= }</span><span class="se">\n</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">)</span><span class="w"> </span><span class="si">= }</span><span class="se">\n</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="o">.</span><span class="n">keys</span><span class="p">()</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s1">&#39;boxes&#39;</span><span class="p">])</span><span class="w"> </span><span class="si">= }</span><span class="se">\n</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s1">&#39;labels&#39;</span><span class="p">])</span><span class="w"> </span><span class="si">= }</span><span class="se">\n</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s1">&#39;masks&#39;</span><span class="p">])</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>type(img) = &lt;class &#39;PIL.Image.Image&#39;&gt;
+type(target) = &lt;class &#39;dict&#39;&gt;
+target.keys() = dict_keys([&#39;boxes&#39;, &#39;masks&#39;, &#39;labels&#39;])
+type(target[&#39;boxes&#39;]) = &lt;class &#39;torchvision.tv_tensors._bounding_boxes.BoundingBoxes&#39;&gt;
+type(target[&#39;labels&#39;]) = &lt;class &#39;torch.Tensor&#39;&gt;
+type(target[&#39;masks&#39;]) = &lt;class &#39;torchvision.tv_tensors._mask.Mask&#39;&gt;
+</pre></div>
+</div>
+<p>We used the <code class="docutils literal notranslate"><span class="pre">target_keys</span></code> parameter to specify the kind of output we’re
+interested in. Our dataset now returns a target which is dict where the values
+are <a class="reference internal" href="plot_transforms_getting_started.html#what-are-tv-tensors"><span class="std std-ref">TVTensors</span></a> (all are <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>
+subclasses). We’re dropped all unncessary keys from the previous output, but
+if you need any of the original keys e.g. “image_id”, you can still ask for
+it.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>If you just want to do detection, you don’t need and shouldn’t pass
+“masks” in <code class="docutils literal notranslate"><span class="pre">target_keys</span></code>: if masks are present in the sample, they will
+be transformed, slowing down your transformations unnecessarily.</p>
+</div>
+<p>As baseline, let’s have a look at a sample without transformations:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">plot</span><span class="p">([</span><span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">1</span><span class="p">]])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_e2e_001.png" srcset="../../_images/sphx_glr_plot_transforms_e2e_001.png" alt="plot transforms e2e" class = "sphx-glr-single-img"/></section>
+<section id="transforms">
+<h2>Transforms<a class="headerlink" href="#transforms" title="Permalink to this heading">¶</a></h2>
+<p>Let’s now define our pre-processing transforms. All the transforms know how
+to handle images, bouding boxes and masks when relevant.</p>
+<p>Transforms are typically passed as the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter of the
+dataset so that they can leverage multi-processing from the
+<a class="reference external" href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.utils.data.DataLoader</span></code></a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">transforms</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Compose</span></a><span class="p">(</span>
+    <span class="p">[</span>
+        <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">ToImage</span></a><span class="p">(),</span>
+        <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomPhotometricDistort</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+        <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomZoomOut</span></a><span class="p">(</span><span class="n">fill</span><span class="o">=</span><span class="p">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">:</span> <span class="p">(</span><span class="mi">123</span><span class="p">,</span> <span class="mi">117</span><span class="p">,</span> <span class="mi">104</span><span class="p">),</span> <span class="s2">&quot;others&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">}),</span>
+        <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomIoUCrop</span></a><span class="p">(),</span>
+        <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+        <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">SanitizeBoundingBoxes</span></a><span class="p">(),</span>
+        <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">ToDtype</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">float32</span></a><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <span class="p">]</span>
+<span class="p">)</span>
+
+<span class="n">dataset</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.Dataset" title="torch.utils.data.Dataset" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">datasets</span><span class="o">.</span><span class="n">CocoDetection</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">IMAGES_PATH</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">ANNOTATIONS_PATH</span></a><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">transforms</span><span class="p">)</span>
+<span class="n">dataset</span> <span class="o">=</span> <span class="n">datasets</span><span class="o">.</span><span class="n">wrap_dataset_for_transforms_v2</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">target_keys</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">,</span> <span class="s2">&quot;masks&quot;</span><span class="p">])</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loading annotations into memory...
+Done (t=0.00s)
+creating index...
+index created!
+</pre></div>
+</div>
+<p>A few things are worth noting here:</p>
+<ul class="simple">
+<li><p>We’re converting the PIL image into a
+<code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code> object. This isn’t strictly
+necessary, but relying on Tensors (here: a Tensor subclass) will
+<a class="reference internal" href="../../transforms.html#transforms-perf"><span class="std std-ref">generally be faster</span></a>.</p></li>
+<li><p>We are calling <a class="reference internal" href="../../generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes" title="torchvision.transforms.v2.SanitizeBoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">SanitizeBoundingBoxes</span></code></a> to
+make sure we remove degenerate bounding boxes, as well as their
+corresponding labels and masks.
+<a class="reference internal" href="../../generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes" title="torchvision.transforms.v2.SanitizeBoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">SanitizeBoundingBoxes</span></code></a> should be placed
+at least once at the end of a detection pipeline; it is particularly
+critical if <a class="reference internal" href="../../generated/torchvision.transforms.v2.RandomIoUCrop.html#torchvision.transforms.v2.RandomIoUCrop" title="torchvision.transforms.v2.RandomIoUCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomIoUCrop</span></code></a> was used.</p></li>
+</ul>
+<p>Let’s look how the sample looks like with our augmentation pipeline in place:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">plot</span><span class="p">([</span><span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">1</span><span class="p">]])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_e2e_002.png" srcset="../../_images/sphx_glr_plot_transforms_e2e_002.png" alt="plot transforms e2e" class = "sphx-glr-single-img"/><p>We can see that the color of the images were distorted, zoomed in or out, and flipped.
+The bounding boxes and the masks were transformed accordingly. And without any further ado, we can start training.</p>
+</section>
+<section id="data-loading-and-training-loop">
+<h2>Data loading and training loop<a class="headerlink" href="#data-loading-and-training-loop" title="Permalink to this heading">¶</a></h2>
+<p>Below we’re using Mask-RCNN which is an instance segmentation model, but
+everything we’ve covered in this tutorial also applies to object detection and
+semantic segmentation tasks.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data_loader</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">DataLoader</span></a><span class="p">(</span>
+    <span class="n">dataset</span><span class="p">,</span>
+    <span class="n">batch_size</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+    <span class="c1"># We need a custom collation function here, since the object detection</span>
+    <span class="c1"># models expect a sequence of images and target dictionaries. The default</span>
+    <span class="c1"># collation function tries to torch.stack() the individual elements,</span>
+    <span class="c1"># which fails in general for object detection, because the number of bouding</span>
+    <span class="c1"># boxes varies between the images of a same batch.</span>
+    <span class="n">collate_fn</span><span class="o">=</span><span class="k">lambda</span> <span class="n">batch</span><span class="p">:</span> <span class="nb">tuple</span><span class="p">(</span><span class="nb">zip</span><span class="p">(</span><span class="o">*</span><span class="n">batch</span><span class="p">)),</span>
+<span class="p">)</span>
+
+<span class="n">model</span> <span class="o">=</span> <span class="n">models</span><span class="o">.</span><span class="n">get_model</span><span class="p">(</span><span class="s2">&quot;maskrcnn_resnet50_fpn_v2&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">weights_backbone</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
+
+<span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">targets</span></a> <span class="ow">in</span> <a href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="torch.utils.data.DataLoader" class="sphx-glr-backref-module-torch-utils-data sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">data_loader</span></a><span class="p">:</span>
+    <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">loss_dict</span></a> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">targets</span></a><span class="p">)</span>
+    <span class="c1"># Put your training logic here</span>
+
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="p">[</span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span><span class="o">.</span><span class="n">shape</span></a><span class="w"> </span><span class="k">for</span><span class="w"> </span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="w"> </span><span class="ow">in</span><span class="w"> </span><a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a><span class="p">]</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="p">[</span><span class="nb">type</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">)</span><span class="w"> </span><span class="k">for</span><span class="w"> </span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="w"> </span><span class="ow">in</span><span class="w"> </span><a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">targets</span></a><span class="p">]</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="k">for</span> <a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">name</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">loss_val</span></a> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">loss_dict</span></a><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+        <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">name</span></a><span class="si">:</span><span class="s2">&lt;20</span><span class="si">}{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">loss_val</span></a><span class="si">:</span><span class="s2">.3f</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[img.shape for img in imgs] = [torch.Size([3, 512, 512]), torch.Size([3, 409, 493])]
+[type(target) for target in targets] = [&lt;class &#39;dict&#39;&gt;, &lt;class &#39;dict&#39;&gt;]
+loss_classifier     4.721
+loss_box_reg        0.006
+loss_mask           0.734
+loss_objectness     0.691
+loss_rpn_box_reg    0.036
+</pre></div>
+</div>
+</section>
+<section id="training-references">
+<h2>Training References<a class="headerlink" href="#training-references" title="Permalink to this heading">¶</a></h2>
+<p>From there, you can check out the <a class="reference external" href="https://github.com/pytorch/vision/tree/main/references">torchvision references</a> where you’ll find
+the actual training scripts we use to train our models.</p>
+<p><strong>Disclaimer</strong> The code in our references is more complex than what you’ll
+need for your own use-cases: this is because we’re supporting different
+backends (PIL, tensors, TVTensors) and different transforms namespaces (v1 and
+v2). So don’t be afraid to simplify and only keep what you need.</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 4.813 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-transforms-plot-transforms-e2e-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/40e082c3b2415f11f987eaa06c4d7513/plot_transforms_e2e.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_transforms_e2e.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/fd385b39c48cf3b5eb72e901830b8621/plot_transforms_e2e.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_transforms_e2e.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_cutmix_mixup.html" class="btn btn-neutral float-right" title="How to use CutMix and MixUp" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_transforms_illustrations.html" class="btn btn-neutral" title="Illustration of transforms" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Transforms v2: End-to-end object detection/segmentation example</a><ul>
+<li><a class="reference internal" href="#dataset-preparation">Dataset preparation</a></li>
+<li><a class="reference internal" href="#transforms">Transforms</a></li>
+<li><a class="reference internal" href="#data-loading-and-training-loop">Data loading and training loop</a></li>
+<li><a class="reference internal" href="#training-references">Training References</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/plot_transforms_getting_started.html b/0.16/auto_examples/transforms/plot_transforms_getting_started.html
new file mode 100644
index 00000000000..e365e69810f
--- /dev/null
+++ b/0.16/auto_examples/transforms/plot_transforms_getting_started.html
@@ -0,0 +1,949 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Getting started with transforms v2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Illustration of transforms" href="plot_transforms_illustrations.html" />
+    <link rel="prev" title="Transforms" href="index.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Transforms</a> &gt;</li>
+        
+      <li>Getting started with transforms v2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/plot_transforms_getting_started.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="getting-started-with-transforms-v2">
+<span id="sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"></span><h1>Getting started with transforms v2<a class="headerlink" href="#getting-started-with-transforms-v2" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_getting_started.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>This example illustrates all of what you need to know to get started with the
+new <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> API. We’ll cover simple tasks like
+image classification, and more advanced ones like object detection /
+segmentation.</p>
+<p>First, a bit of setup</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
+<a href="https://matplotlib.org/stable/api/matplotlib_configuration_api.html#matplotlib.RcParams" title="matplotlib.RcParams" class="sphx-glr-backref-module-matplotlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">plt</span><span class="o">.</span><span class="n">rcParams</span></a><span class="p">[</span><span class="s2">&quot;savefig.bbox&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;tight&#39;</span>
+
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+<span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_image</span>
+
+<a href="https://pytorch.org/docs/stable/generated/torch.manual_seed.html#torch.manual_seed" title="torch.manual_seed" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">manual_seed</span></a><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+<span class="c1"># If you&#39;re trying to run that on collab, you can download the assets and the</span>
+<span class="c1"># helpers from https://github.com/pytorch/vision/tree/main/gallery/</span>
+<span class="kn">from</span> <span class="nn">helpers</span> <span class="kn">import</span> <span class="n">plot</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><span class="s1">&#39;../assets&#39;</span><span class="p">)</span> <span class="o">/</span> <span class="s1">&#39;astronaut.jpg&#39;</span><span class="p">))</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span><span class="o">.</span><span class="n">dtype</span></a><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>type(img) = &lt;class &#39;torch.Tensor&#39;&gt;, img.dtype = torch.uint8, img.shape = torch.Size([3, 512, 512])
+</pre></div>
+</div>
+<section id="the-basics">
+<h2>The basics<a class="headerlink" href="#the-basics" title="Permalink to this heading">¶</a></h2>
+<p>The Torchvision transforms behave like a regular <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.Module</span></code></a> (in
+fact, most of them are): instantiate a transform, pass an input, get a
+transformed output:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">transform</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">))</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out</span></a> <span class="o">=</span> <span class="n">transform</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span>
+
+<span class="n">plot</span><span class="p">([</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out</span></a><span class="p">])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_getting_started_001.png" srcset="../../_images/sphx_glr_plot_transforms_getting_started_001.png" alt="plot transforms getting started" class = "sphx-glr-single-img"/></section>
+<section id="i-just-want-to-do-image-classification">
+<h2>I just want to do image classification<a class="headerlink" href="#i-just-want-to-do-image-classification" title="Permalink to this heading">¶</a></h2>
+<p>If you just care about image classification, things are very simple. A basic
+classification pipeline may look like this:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">transforms</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Compose</span></a><span class="p">([</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomResizedCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">ToDtype</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">float32</span></a><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Normalize</span></a><span class="p">(</span><span class="n">mean</span><span class="o">=</span><span class="p">[</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">],</span> <span class="n">std</span><span class="o">=</span><span class="p">[</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">]),</span>
+<span class="p">])</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out</span></a> <span class="o">=</span> <span class="n">transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">)</span>
+
+<span class="n">plot</span><span class="p">([</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out</span></a><span class="p">])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_getting_started_002.png" srcset="../../_images/sphx_glr_plot_transforms_getting_started_002.png" alt="plot transforms getting started" class = "sphx-glr-single-img"/><p>Such transformation pipeline is typically passed as the <code class="docutils literal notranslate"><span class="pre">transform</span></code> argument
+to the <a class="reference internal" href="../../datasets.html#datasets"><span class="std std-ref">Datasets</span></a>, e.g. <code class="docutils literal notranslate"><span class="pre">ImageNet(...,</span>
+<span class="pre">transform=transforms)</span></code>.</p>
+<p>That’s pretty much all there is. From there, read through our <a class="reference internal" href="../../transforms.html#transforms"><span class="std std-ref">main docs</span></a> to learn more about recommended practices and conventions, or
+explore more <a class="reference internal" href="index.html#transforms-gallery"><span class="std std-ref">examples</span></a> e.g. how to use augmentation
+transforms like <a class="reference internal" href="plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">CutMix and MixUp</span></a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>If you’re already relying on the <code class="docutils literal notranslate"><span class="pre">torchvision.transforms</span></code> v1 API,
+we recommend to <a class="reference internal" href="../../transforms.html#v1-or-v2"><span class="std std-ref">switch to the new v2 transforms</span></a>. It’s
+very easy: the v2 transforms are fully compatible with the v1 API, so you
+only need to change the import!</p>
+</div>
+</section>
+<section id="detection-segmentation-videos">
+<h2>Detection, Segmentation, Videos<a class="headerlink" href="#detection-segmentation-videos" title="Permalink to this heading">¶</a></h2>
+<p>The new Torchvision transforms in the <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> namespace
+support tasks beyond image classification: they can also transform bounding
+boxes, segmentation / detection masks, or videos.</p>
+<p>Let’s briefly look at a detection example with bounding boxes.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>  <span class="c1"># we&#39;ll describe this a bit later, bare with us</span>
+
+<span class="n">boxes</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">(</span>
+    <span class="p">[</span>
+        <span class="p">[</span><span class="mi">15</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">370</span><span class="p">,</span> <span class="mi">510</span><span class="p">],</span>
+        <span class="p">[</span><span class="mi">275</span><span class="p">,</span> <span class="mi">340</span><span class="p">,</span> <span class="mi">510</span><span class="p">,</span> <span class="mi">510</span><span class="p">],</span>
+        <span class="p">[</span><span class="mi">130</span><span class="p">,</span> <span class="mi">345</span><span class="p">,</span> <span class="mi">210</span><span class="p">,</span> <span class="mi">425</span><span class="p">]</span>
+    <span class="p">],</span>
+    <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;XYXY&quot;</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:])</span>
+
+<span class="n">transforms</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Compose</span></a><span class="p">([</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomResizedCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomPhotometricDistort</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+    <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span>
+<span class="p">])</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_img</span></a><span class="p">,</span> <span class="n">out_boxes</span> <span class="o">=</span> <span class="n">transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <span class="n">boxes</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">boxes</span><span class="p">),</span> <span class="nb">type</span><span class="p">(</span><span class="n">out_boxes</span><span class="p">))</span>
+
+<span class="n">plot</span><span class="p">([(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <span class="n">boxes</span><span class="p">),</span> <span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_img</span></a><span class="p">,</span> <span class="n">out_boxes</span><span class="p">)])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_getting_started_003.png" srcset="../../_images/sphx_glr_plot_transforms_getting_started_003.png" alt="plot transforms getting started" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&lt;class &#39;torchvision.tv_tensors._bounding_boxes.BoundingBoxes&#39;&gt; &lt;class &#39;torchvision.tv_tensors._bounding_boxes.BoundingBoxes&#39;&gt;
+</pre></div>
+</div>
+<p>The example above focuses on object detection. But if we had masks
+(<a class="reference internal" href="../../generated/torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.tv_tensors.Mask</span></code></a>) for object segmentation or semantic
+segmentation, or videos (<a class="reference internal" href="../../generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.tv_tensors.Video</span></code></a>), we could have
+passed them to the transforms in exactly the same way.</p>
+<p>By now you likely have a few questions: what are these TVTensors, how do we
+use them, and what is the expected input/output of those transforms? We’ll
+answer these in the next sections.</p>
+</section>
+<section id="what-are-tvtensors">
+<span id="what-are-tv-tensors"></span><h2>What are TVTensors?<a class="headerlink" href="#what-are-tvtensors" title="Permalink to this heading">¶</a></h2>
+<p>TVTensors are <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclasses. The available TVTensors are
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a>,
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">Mask</span></code></a>, and
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>.</p>
+<p>TVTensors look and feel just like regular tensors - they <strong>are</strong> tensors.
+Everything that is supported on a plain <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> like <code class="docutils literal notranslate"><span class="pre">.sum()</span></code>
+or any <code class="docutils literal notranslate"><span class="pre">torch.*</span></code> operator will also work on a TVTensor:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">img_dp</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/generated/torch.randint.html#torch.randint" title="torch.randint" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">randint</span></a><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span></a><span class="p">))</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">img_dp</span><span class="p">,</span><span class="w"> </span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span></a><span class="p">)</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">img_dp</span><span class="o">.</span><span class="n">dtype</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><span class="n">img_dp</span><span class="o">.</span><span class="n">shape</span><span class="w"> </span><span class="si">= }</span><span class="s2">, </span><span class="si">{</span><span class="n">img_dp</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span><span class="w"> </span><span class="si">= }</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>isinstance(img_dp, torch.Tensor) = True
+img_dp.dtype = torch.uint8, img_dp.shape = torch.Size([3, 256, 256]), img_dp.sum() = tensor(25087958)
+</pre></div>
+</div>
+<p>These TVTensor classes are at the core of the transforms: in order to
+transform a given input, the transforms first look at the <strong>class</strong> of the
+object, and dispatch to the appropriate implementation accordingly.</p>
+<p>You don’t need to know much more about TVTensors at this point, but advanced
+users who want to learn more can refer to
+<a class="reference internal" href="plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a>.</p>
+</section>
+<section id="what-do-i-pass-as-input">
+<h2>What do I pass as input?<a class="headerlink" href="#what-do-i-pass-as-input" title="Permalink to this heading">¶</a></h2>
+<p>Above, we’ve seen two examples: one where we passed a single image as input
+i.e. <code class="docutils literal notranslate"><span class="pre">out</span> <span class="pre">=</span> <span class="pre">transforms(img)</span></code>, and one where we passed both an image and
+bounding boxes, i.e. <code class="docutils literal notranslate"><span class="pre">out_img,</span> <span class="pre">out_boxes</span> <span class="pre">=</span> <span class="pre">transforms(img,</span> <span class="pre">boxes)</span></code>.</p>
+<p>In fact, transforms support <strong>arbitrary input structures</strong>. The input can be a
+single image, a tuple, an arbitrarily nested dictionary… pretty much
+anything. The same structure will be returned as output. Below, we use the
+same detection transforms, but pass a tuple (image, target_dict) as input and
+we’re getting the same structure as output:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;boxes&quot;</span><span class="p">:</span> <span class="n">boxes</span><span class="p">,</span>
+    <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <a href="https://pytorch.org/docs/stable/generated/torch.arange.html#torch.arange" title="torch.arange" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">arange</span></a><span class="p">(</span><span class="n">boxes</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]),</span>
+    <span class="s2">&quot;this_is_ignored&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;arbitrary&quot;</span><span class="p">,</span> <span class="p">{</span><span class="s2">&quot;structure&quot;</span><span class="p">:</span> <span class="s2">&quot;!&quot;</span><span class="p">})</span>
+<span class="p">}</span>
+
+<span class="c1"># Re-using the transforms and definitions from above.</span>
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_target</span></a> <span class="o">=</span> <span class="n">transforms</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">)</span>
+
+<span class="n">plot</span><span class="p">([(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">target</span></a><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">]),</span> <span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_img</span></a><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_target</span></a><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">])])</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">out_target</span></a><span class="p">[</span><span class="s1">&#39;this_is_ignored&#39;</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_getting_started_004.png" srcset="../../_images/sphx_glr_plot_transforms_getting_started_004.png" alt="plot transforms getting started" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>(&#39;arbitrary&#39;, {&#39;structure&#39;: &#39;!&#39;})
+</pre></div>
+</div>
+<p>We passed a tuple so we get a tuple back, and the second element is the
+tranformed target dict. Transforms don’t really care about the structure of
+the input; as mentioned above, they only care about the <strong>type</strong> of the
+objects and transforms them accordingly.</p>
+<p><em>Foreign</em> objects like strings or ints are simply passed-through. This can be
+useful e.g. if you want to associate a path with every single sample when
+debugging!</p>
+<div class="admonition note" id="passthrough-heuristic">
+<p class="admonition-title">Note</p>
+<p><strong>Disclaimer</strong> This note is slightly advanced and can be safely skipped on
+a first read.</p>
+<p>Pure <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> objects are, in general, treated as images (or
+as videos for video-specific transforms). Indeed, you may have noticed
+that in the code above we haven’t used the
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a> class at all, and yet our images
+got transformed properly. Transforms follow the following logic to
+determine whether a pure Tensor should be treated as an image (or video),
+or just ignored:</p>
+<ul class="simple">
+<li><p>If there is an <a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>,
+or <a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">PIL.Image.Image</span></code></a> instance in the input, all other pure
+tensors are passed-through.</p></li>
+<li><p>If there is no <a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a> or
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a> instance, only the first pure
+<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> will be transformed as image or video, while all
+others will be passed-through. Here “first” means “first in a depth-wise
+traversal”.</p></li>
+</ul>
+<p>This is what happened in the detection example above: the first pure
+tensor was the image so it got transformed properly, and all other pure
+tensor instances like the <code class="docutils literal notranslate"><span class="pre">labels</span></code> were passed-through (although labels
+can still be transformed by some transforms like
+<a class="reference internal" href="../../generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes" title="torchvision.transforms.v2.SanitizeBoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">SanitizeBoundingBoxes</span></code></a>!).</p>
+</div>
+</section>
+<section id="transforms-and-datasets-intercompatibility">
+<span id="transforms-datasets-intercompatibility"></span><h2>Transforms and Datasets intercompatibility<a class="headerlink" href="#transforms-and-datasets-intercompatibility" title="Permalink to this heading">¶</a></h2>
+<p>Roughly speaking, the output of the datasets must correspond to the input of
+the transforms. How to do that depends on whether you’re using the torchvision
+<a class="reference internal" href="../../datasets.html#datasets"><span class="std std-ref">built-in datatsets</span></a>, or your own custom datasets.</p>
+<section id="using-built-in-datasets">
+<h3>Using built-in datasets<a class="headerlink" href="#using-built-in-datasets" title="Permalink to this heading">¶</a></h3>
+<p>If you’re just doing image classification, you don’t need to do anything. Just
+use <code class="docutils literal notranslate"><span class="pre">transform</span></code> argument of the dataset e.g. <code class="docutils literal notranslate"><span class="pre">ImageNet(...,</span>
+<span class="pre">transform=transforms)</span></code> and you’re good to go.</p>
+<p>Torchvision also supports datasets for object detection or segmentation like
+<a class="reference internal" href="../../generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection" title="torchvision.datasets.CocoDetection"><code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.datasets.CocoDetection</span></code></a>. Those datasets predate
+the existence of the <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> module and of the
+TVTensors, so they don’t return TVTensors out of the box.</p>
+<p>An easy way to force those datasets to return TVTensors and to make them
+compatible with v2 transforms is to use the
+<a class="reference internal" href="../../generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html#torchvision.datasets.wrap_dataset_for_transforms_v2" title="torchvision.datasets.wrap_dataset_for_transforms_v2"><code class="xref py py-func docutils literal notranslate"><span class="pre">torchvision.datasets.wrap_dataset_for_transforms_v2()</span></code></a> function:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.datasets</span> <span class="kn">import</span> <span class="n">CocoDetection</span><span class="p">,</span> <span class="n">wrap_dataset_for_transforms_v2</span>
+
+<span class="n">dataset</span> <span class="o">=</span> <span class="n">CocoDetection</span><span class="p">(</span><span class="o">...</span><span class="p">,</span> <span class="n">transforms</span><span class="o">=</span><span class="n">my_transforms</span><span class="p">)</span>
+<span class="n">dataset</span> <span class="o">=</span> <span class="n">wrap_dataset_for_transforms_v2</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+<span class="c1"># Now the dataset returns TVTensors!</span>
+</pre></div>
+</div>
+</section>
+<section id="using-your-own-datasets">
+<h3>Using your own datasets<a class="headerlink" href="#using-your-own-datasets" title="Permalink to this heading">¶</a></h3>
+<p>If you have a custom dataset, then you’ll need to convert your objects into
+the appropriate TVTensor classes. Creating TVTensor instances is very easy,
+refer to <a class="reference internal" href="plot_tv_tensors.html#tv-tensor-creation"><span class="std std-ref">How do I construct a TVTensor?</span></a> for more details.</p>
+<p>There are two main places where you can implement that conversion logic:</p>
+<ul class="simple">
+<li><p>At the end of the datasets’s <code class="docutils literal notranslate"><span class="pre">__getitem__</span></code> method, before returning the
+sample (or by sub-classing the dataset).</p></li>
+<li><p>As the very first step of your transforms pipeline</p></li>
+</ul>
+<p>Either way, the logic will depend on your specific dataset.</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 0.838 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-transforms-plot-transforms-getting-started-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/8faa1256268baabef0c4b947064f34f4/plot_transforms_getting_started.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_transforms_getting_started.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/d1e0b3022ec2c789e633f49bb9dd0cf4/plot_transforms_getting_started.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_transforms_getting_started.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_transforms_illustrations.html" class="btn btn-neutral float-right" title="Illustration of transforms" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="index.html" class="btn btn-neutral" title="Transforms" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Getting started with transforms v2</a><ul>
+<li><a class="reference internal" href="#the-basics">The basics</a></li>
+<li><a class="reference internal" href="#i-just-want-to-do-image-classification">I just want to do image classification</a></li>
+<li><a class="reference internal" href="#detection-segmentation-videos">Detection, Segmentation, Videos</a></li>
+<li><a class="reference internal" href="#what-are-tvtensors">What are TVTensors?</a></li>
+<li><a class="reference internal" href="#what-do-i-pass-as-input">What do I pass as input?</a></li>
+<li><a class="reference internal" href="#transforms-and-datasets-intercompatibility">Transforms and Datasets intercompatibility</a><ul>
+<li><a class="reference internal" href="#using-built-in-datasets">Using built-in datasets</a></li>
+<li><a class="reference internal" href="#using-your-own-datasets">Using your own datasets</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/plot_transforms_illustrations.html b/0.16/auto_examples/transforms/plot_transforms_illustrations.html
new file mode 100644
index 00000000000..6a48616c0f5
--- /dev/null
+++ b/0.16/auto_examples/transforms/plot_transforms_illustrations.html
@@ -0,0 +1,1077 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Illustration of transforms &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Transforms v2: End-to-end object detection/segmentation example" href="plot_transforms_e2e.html" />
+    <link rel="prev" title="Getting started with transforms v2" href="plot_transforms_getting_started.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Transforms</a> &gt;</li>
+        
+      <li>Illustration of transforms</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/plot_transforms_illustrations.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="illustration-of-transforms">
+<span id="sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"></span><h1>Illustration of transforms<a class="headerlink" href="#illustration-of-transforms" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_transforms_illustrations.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>This example illustrates some of the various transforms available in <a class="reference internal" href="../../transforms.html#transforms"><span class="std std-ref">the
+torchvision.transforms.v2 module</span></a>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
+<span class="kn">from</span> <span class="nn">pathlib</span> <span class="kn">import</span> <a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a>
+<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+
+<a href="https://matplotlib.org/stable/api/matplotlib_configuration_api.html#matplotlib.RcParams" title="matplotlib.RcParams" class="sphx-glr-backref-module-matplotlib sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">plt</span><span class="o">.</span><span class="n">rcParams</span></a><span class="p">[</span><span class="s2">&quot;savefig.bbox&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;tight&#39;</span>
+
+<span class="c1"># if you change the seed, make sure that the randomly-applied transforms</span>
+<span class="c1"># properly show that the image can be both transformed and *not* transformed!</span>
+<a href="https://pytorch.org/docs/stable/generated/torch.manual_seed.html#torch.manual_seed" title="torch.manual_seed" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">manual_seed</span></a><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+<span class="c1"># If you&#39;re trying to run that on collab, you can download the assets and the</span>
+<span class="c1"># helpers from https://github.com/pytorch/vision/tree/main/gallery/</span>
+<span class="kn">from</span> <span class="nn">helpers</span> <span class="kn">import</span> <span class="n">plot</span>
+<a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a> <span class="o">=</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.open" title="PIL.Image.open" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-function"><span class="n">Image</span><span class="o">.</span><span class="n">open</span></a><span class="p">(</span><a href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="pathlib.Path" class="sphx-glr-backref-module-pathlib sphx-glr-backref-type-py-class"><span class="n">Path</span></a><span class="p">(</span><span class="s1">&#39;../assets&#39;</span><span class="p">)</span> <span class="o">/</span> <span class="s1">&#39;astronaut.jpg&#39;</span><span class="p">)</span>
+</pre></div>
+</div>
+<section id="geometric-transforms">
+<h2>Geometric Transforms<a class="headerlink" href="#geometric-transforms" title="Permalink to this heading">¶</a></h2>
+<p>Geometric image transformation refers to the process of altering the geometric properties of an image,
+such as its shape, size, orientation, or position.
+It involves applying mathematical operations to the image pixels or coordinates to achieve the desired transformation.</p>
+<section id="pad">
+<h3>Pad<a class="headerlink" href="#pad" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.Pad.html#torchvision.transforms.Pad" title="torchvision.transforms.Pad"><code class="xref py py-class docutils literal notranslate"><span class="pre">Pad</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.pad.html#torchvision.transforms.functional.pad" title="torchvision.transforms.functional.pad"><code class="xref py py-func docutils literal notranslate"><span class="pre">pad()</span></code></a>)
+pads all image borders with some pixel values.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">padded_imgs</span></a> <span class="o">=</span> <span class="p">[</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Pad</span></a><span class="p">(</span><span class="n">padding</span><span class="o">=</span><span class="n">padding</span><span class="p">)(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">padding</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">30</span><span class="p">,</span> <span class="mi">50</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">padded_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_001.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_001.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="resize">
+<h3>Resize<a class="headerlink" href="#resize" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.Resize.html#torchvision.transforms.Resize" title="torchvision.transforms.Resize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Resize</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.resize.html#torchvision.transforms.functional.resize" title="torchvision.transforms.functional.resize"><code class="xref py py-func docutils literal notranslate"><span class="pre">resize()</span></code></a>)
+resizes an image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">resized_imgs</span></a> <span class="o">=</span> <span class="p">[</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Resize</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">)(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">size</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">30</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">100</span><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image.size" title="PIL.Image.Image.size" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-attribute"><span class="n">orig_img</span><span class="o">.</span><span class="n">size</span></a><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">resized_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_002.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_002.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="centercrop">
+<h3>CenterCrop<a class="headerlink" href="#centercrop" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.CenterCrop.html#torchvision.transforms.CenterCrop" title="torchvision.transforms.CenterCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">CenterCrop</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.center_crop.html#torchvision.transforms.functional.center_crop" title="torchvision.transforms.functional.center_crop"><code class="xref py py-func docutils literal notranslate"><span class="pre">center_crop()</span></code></a>)
+crops the given image at the center.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">center_crops</span></a> <span class="o">=</span> <span class="p">[</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">CenterCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="n">size</span><span class="p">)(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">size</span> <span class="ow">in</span> <span class="p">(</span><span class="mi">30</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">100</span><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image.size" title="PIL.Image.Image.size" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-attribute"><span class="n">orig_img</span><span class="o">.</span><span class="n">size</span></a><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">center_crops</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_003.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_003.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="fivecrop">
+<h3>FiveCrop<a class="headerlink" href="#fivecrop" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.FiveCrop.html#torchvision.transforms.FiveCrop" title="torchvision.transforms.FiveCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">FiveCrop</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.five_crop.html#torchvision.transforms.functional.five_crop" title="torchvision.transforms.functional.five_crop"><code class="xref py py-func docutils literal notranslate"><span class="pre">five_crop()</span></code></a>)
+crops the given image into four corners and the central crop.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">top_left</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">top_right</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">bottom_left</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">bottom_right</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">center</span></a><span class="p">)</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">FiveCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">100</span><span class="p">,</span> <span class="mi">100</span><span class="p">))(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">top_left</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">top_right</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">bottom_left</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">bottom_right</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">center</span></a><span class="p">])</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_004.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_004.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomperspective">
+<h3>RandomPerspective<a class="headerlink" href="#randomperspective" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomPerspective.html#torchvision.transforms.RandomPerspective" title="torchvision.transforms.RandomPerspective"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomPerspective</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.perspective.html#torchvision.transforms.functional.perspective" title="torchvision.transforms.functional.perspective"><code class="xref py py-func docutils literal notranslate"><span class="pre">perspective()</span></code></a>)
+performs random perspective transform on an image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">perspective_transformer</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomPerspective</span></a><span class="p">(</span><span class="n">distortion_scale</span><span class="o">=</span><span class="mf">0.6</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">1.0</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">perspective_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">perspective_transformer</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">perspective_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_005.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_005.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomrotation">
+<h3>RandomRotation<a class="headerlink" href="#randomrotation" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomRotation.html#torchvision.transforms.RandomRotation" title="torchvision.transforms.RandomRotation"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomRotation</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.rotate.html#torchvision.transforms.functional.rotate" title="torchvision.transforms.functional.rotate"><code class="xref py py-func docutils literal notranslate"><span class="pre">rotate()</span></code></a>)
+rotates an image with random angle.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">rotater</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomRotation</span></a><span class="p">(</span><span class="n">degrees</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">180</span><span class="p">))</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">rotated_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">rotater</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">rotated_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_006.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_006.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomaffine">
+<h3>RandomAffine<a class="headerlink" href="#randomaffine" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomAffine.html#torchvision.transforms.RandomAffine" title="torchvision.transforms.RandomAffine"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAffine</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.affine.html#torchvision.transforms.functional.affine" title="torchvision.transforms.functional.affine"><code class="xref py py-func docutils literal notranslate"><span class="pre">affine()</span></code></a>)
+performs random affine transform on an image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">affine_transfomer</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomAffine</span></a><span class="p">(</span><span class="n">degrees</span><span class="o">=</span><span class="p">(</span><span class="mi">30</span><span class="p">,</span> <span class="mi">70</span><span class="p">),</span> <span class="n">translate</span><span class="o">=</span><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.3</span><span class="p">),</span> <span class="n">scale</span><span class="o">=</span><span class="p">(</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.75</span><span class="p">))</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">affine_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">affine_transfomer</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">affine_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_007.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_007.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="elastictransform">
+<h3>ElasticTransform<a class="headerlink" href="#elastictransform" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.ElasticTransform.html#torchvision.transforms.ElasticTransform" title="torchvision.transforms.ElasticTransform"><code class="xref py py-class docutils literal notranslate"><span class="pre">ElasticTransform</span></code></a> transform
+(see also <code class="xref py py-func docutils literal notranslate"><span class="pre">elastic_transform()</span></code>)
+Randomly transforms the morphology of objects in images and produces a
+see-through-water-like effect.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">elastic_transformer</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">ElasticTransform</span></a><span class="p">(</span><span class="n">alpha</span><span class="o">=</span><span class="mf">250.0</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">elastic_transformer</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">2</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_008.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_008.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomcrop">
+<h3>RandomCrop<a class="headerlink" href="#randomcrop" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomCrop.html#torchvision.transforms.RandomCrop" title="torchvision.transforms.RandomCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomCrop</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.crop.html#torchvision.transforms.functional.crop" title="torchvision.transforms.functional.crop"><code class="xref py py-func docutils literal notranslate"><span class="pre">crop()</span></code></a>)
+crops an image at a random location.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">cropper</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">128</span><span class="p">,</span> <span class="mi">128</span><span class="p">))</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">crops</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">cropper</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">crops</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_009.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_009.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomresizedcrop">
+<h3>RandomResizedCrop<a class="headerlink" href="#randomresizedcrop" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop" title="torchvision.transforms.RandomResizedCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.resized_crop.html#torchvision.transforms.functional.resized_crop" title="torchvision.transforms.functional.resized_crop"><code class="xref py py-func docutils literal notranslate"><span class="pre">resized_crop()</span></code></a>)
+crops an image at a random location, and then resizes the crop to a given
+size.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">resize_cropper</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomResizedCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">))</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">resized_crops</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">resize_cropper</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">resized_crops</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_010.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_010.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+</section>
+<section id="photometric-transforms">
+<h2>Photometric Transforms<a class="headerlink" href="#photometric-transforms" title="Permalink to this heading">¶</a></h2>
+<p>Photometric image transformation refers to the process of modifying the photometric properties of an image,
+such as its brightness, contrast, color, or tone.
+These transformations are applied to change the visual appearance of an image
+while preserving its geometric structure.</p>
+<p>Except <a class="reference internal" href="../../generated/torchvision.transforms.Grayscale.html#torchvision.transforms.Grayscale" title="torchvision.transforms.Grayscale"><code class="xref py py-class docutils literal notranslate"><span class="pre">Grayscale</span></code></a>, the following transforms are random,
+which means that the same transform
+instance will produce different result each time it transforms a given image.</p>
+<section id="grayscale">
+<h3>Grayscale<a class="headerlink" href="#grayscale" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.Grayscale.html#torchvision.transforms.Grayscale" title="torchvision.transforms.Grayscale"><code class="xref py py-class docutils literal notranslate"><span class="pre">Grayscale</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.to_grayscale.html#torchvision.transforms.functional.to_grayscale" title="torchvision.transforms.functional.to_grayscale"><code class="xref py py-func docutils literal notranslate"><span class="pre">to_grayscale()</span></code></a>)
+converts an image to grayscale</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">gray_img</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">Grayscale</span></a><span class="p">()(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">,</span> <a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="PIL.Image.Image" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">gray_img</span></a><span class="p">],</span> <span class="n">cmap</span><span class="o">=</span><span class="s1">&#39;gray&#39;</span><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_011.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_011.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="colorjitter">
+<h3>ColorJitter<a class="headerlink" href="#colorjitter" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.ColorJitter.html#torchvision.transforms.ColorJitter" title="torchvision.transforms.ColorJitter"><code class="xref py py-class docutils literal notranslate"><span class="pre">ColorJitter</span></code></a> transform
+randomly changes the brightness, contrast, saturation, hue, and other properties of an image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">jitter</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">ColorJitter</span></a><span class="p">(</span><span class="n">brightness</span><span class="o">=</span><span class="mf">.5</span><span class="p">,</span> <span class="n">hue</span><span class="o">=</span><span class="mf">.3</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">jittered_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">jitter</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">jittered_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_012.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_012.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="gaussianblur">
+<h3>GaussianBlur<a class="headerlink" href="#gaussianblur" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.GaussianBlur.html#torchvision.transforms.GaussianBlur" title="torchvision.transforms.GaussianBlur"><code class="xref py py-class docutils literal notranslate"><span class="pre">GaussianBlur</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.gaussian_blur.html#torchvision.transforms.functional.gaussian_blur" title="torchvision.transforms.functional.gaussian_blur"><code class="xref py py-func docutils literal notranslate"><span class="pre">gaussian_blur()</span></code></a>)
+performs gaussian blur transform on an image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">blurrer</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">GaussianBlur</span></a><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="p">(</span><span class="mi">5</span><span class="p">,</span> <span class="mi">9</span><span class="p">),</span> <span class="n">sigma</span><span class="o">=</span><span class="p">(</span><span class="mf">0.1</span><span class="p">,</span> <span class="mf">5.</span><span class="p">))</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">blurred_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">blurrer</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">blurred_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_013.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_013.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randominvert">
+<h3>RandomInvert<a class="headerlink" href="#randominvert" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomInvert.html#torchvision.transforms.RandomInvert" title="torchvision.transforms.RandomInvert"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomInvert</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.invert.html#torchvision.transforms.functional.invert" title="torchvision.transforms.functional.invert"><code class="xref py py-func docutils literal notranslate"><span class="pre">invert()</span></code></a>)
+randomly inverts the colors of the given image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">inverter</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomInvert</span></a><span class="p">()</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">invertered_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">inverter</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">invertered_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_014.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_014.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomposterize">
+<h3>RandomPosterize<a class="headerlink" href="#randomposterize" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomPosterize.html#torchvision.transforms.RandomPosterize" title="torchvision.transforms.RandomPosterize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomPosterize</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.posterize.html#torchvision.transforms.functional.posterize" title="torchvision.transforms.functional.posterize"><code class="xref py py-func docutils literal notranslate"><span class="pre">posterize()</span></code></a>)
+randomly posterizes the image by reducing the number of bits
+of each color channel.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">posterizer</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomPosterize</span></a><span class="p">(</span><span class="n">bits</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">posterized_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">posterizer</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">posterized_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_015.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_015.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomsolarize">
+<h3>RandomSolarize<a class="headerlink" href="#randomsolarize" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomSolarize.html#torchvision.transforms.RandomSolarize" title="torchvision.transforms.RandomSolarize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomSolarize</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.solarize.html#torchvision.transforms.functional.solarize" title="torchvision.transforms.functional.solarize"><code class="xref py py-func docutils literal notranslate"><span class="pre">solarize()</span></code></a>)
+randomly solarizes the image by inverting all pixel values above
+the threshold.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">solarizer</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomSolarize</span></a><span class="p">(</span><span class="n">threshold</span><span class="o">=</span><span class="mf">192.0</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">solarized_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">solarizer</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">solarized_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_016.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_016.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomadjustsharpness">
+<h3>RandomAdjustSharpness<a class="headerlink" href="#randomadjustsharpness" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness" title="torchvision.transforms.RandomAdjustSharpness"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAdjustSharpness</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.adjust_sharpness.html#torchvision.transforms.functional.adjust_sharpness" title="torchvision.transforms.functional.adjust_sharpness"><code class="xref py py-func docutils literal notranslate"><span class="pre">adjust_sharpness()</span></code></a>)
+randomly adjusts the sharpness of the given image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">sharpness_adjuster</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomAdjustSharpness</span></a><span class="p">(</span><span class="n">sharpness_factor</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sharpened_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">sharpness_adjuster</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sharpened_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_017.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_017.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomautocontrast">
+<h3>RandomAutocontrast<a class="headerlink" href="#randomautocontrast" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast" title="torchvision.transforms.RandomAutocontrast"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.autocontrast.html#torchvision.transforms.functional.autocontrast" title="torchvision.transforms.functional.autocontrast"><code class="xref py py-func docutils literal notranslate"><span class="pre">autocontrast()</span></code></a>)
+randomly applies autocontrast to the given image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">autocontraster</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomAutocontrast</span></a><span class="p">()</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">autocontrasted_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">autocontraster</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">autocontrasted_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_018.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_018.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomequalize">
+<h3>RandomEqualize<a class="headerlink" href="#randomequalize" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomEqualize.html#torchvision.transforms.RandomEqualize" title="torchvision.transforms.RandomEqualize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomEqualize</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.equalize.html#torchvision.transforms.functional.equalize" title="torchvision.transforms.functional.equalize"><code class="xref py py-func docutils literal notranslate"><span class="pre">equalize()</span></code></a>)
+randomly equalizes the histogram of the given image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">equalizer</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomEqualize</span></a><span class="p">()</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">equalized_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">equalizer</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">equalized_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_019.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_019.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+</section>
+<section id="augmentation-transforms">
+<h2>Augmentation Transforms<a class="headerlink" href="#augmentation-transforms" title="Permalink to this heading">¶</a></h2>
+<p>The following transforms are combinations of multiple transforms,
+either geometric or photometric, or both.</p>
+<section id="autoaugment">
+<h3>AutoAugment<a class="headerlink" href="#autoaugment" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.AutoAugment.html#torchvision.transforms.AutoAugment" title="torchvision.transforms.AutoAugment"><code class="xref py py-class docutils literal notranslate"><span class="pre">AutoAugment</span></code></a> transform
+automatically augments data based on a given auto-augmentation policy.
+See <a class="reference internal" href="../../generated/torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.AutoAugmentPolicy"><code class="xref py py-class docutils literal notranslate"><span class="pre">AutoAugmentPolicy</span></code></a> for the available policies.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">policies</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">v2</span><span class="o">.</span><span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">CIFAR10</span><span class="p">,</span> <span class="n">v2</span><span class="o">.</span><span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">IMAGENET</span><span class="p">,</span> <span class="n">v2</span><span class="o">.</span><span class="n">AutoAugmentPolicy</span><span class="o">.</span><span class="n">SVHN</span><span class="p">]</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">augmenters</span></a> <span class="o">=</span> <span class="p">[</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">AutoAugment</span></a><span class="p">(</span><span class="n">policy</span><span class="p">)</span> <span class="k">for</span> <span class="n">policy</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">policies</span></a><span class="p">]</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a> <span class="o">=</span> <span class="p">[</span>
+    <span class="p">[</span><span class="n">augmenter</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+    <span class="k">for</span> <span class="n">augmenter</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">augmenters</span></a>
+<span class="p">]</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">row_title</span></a> <span class="o">=</span> <span class="p">[</span><span class="nb">str</span><span class="p">(</span><span class="n">policy</span><span class="p">)</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39;.&#39;</span><span class="p">)[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="k">for</span> <span class="n">policy</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">policies</span></a><span class="p">]</span>
+<span class="n">plot</span><span class="p">([[</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <span class="n">row</span> <span class="k">for</span> <span class="n">row</span> <span class="ow">in</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a><span class="p">],</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">row_title</span></a><span class="o">=</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">row_title</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_020.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_020.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randaugment">
+<h3>RandAugment<a class="headerlink" href="#randaugment" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandAugment.html#torchvision.transforms.RandAugment" title="torchvision.transforms.RandAugment"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandAugment</span></code></a> is an alternate version of AutoAugment.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">augmenter</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandAugment</span></a><span class="p">()</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">augmenter</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_021.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_021.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="trivialaugmentwide">
+<h3>TrivialAugmentWide<a class="headerlink" href="#trivialaugmentwide" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.TrivialAugmentWide.html#torchvision.transforms.TrivialAugmentWide" title="torchvision.transforms.TrivialAugmentWide"><code class="xref py py-class docutils literal notranslate"><span class="pre">TrivialAugmentWide</span></code></a> is an alternate implementation of AutoAugment.
+However, instead of transforming an image multiple times, it transforms an image only once
+using a random transform from a given list with a random strength number.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">augmenter</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">TrivialAugmentWide</span></a><span class="p">()</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">augmenter</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_022.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_022.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="augmix">
+<h3>AugMix<a class="headerlink" href="#augmix" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.AugMix.html#torchvision.transforms.AugMix" title="torchvision.transforms.AugMix"><code class="xref py py-class docutils literal notranslate"><span class="pre">AugMix</span></code></a> transform interpolates between augmented versions of an image.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">augmenter</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">AugMix</span></a><span class="p">()</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">augmenter</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_023.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_023.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+</section>
+<section id="randomly-applied-transforms">
+<h2>Randomly-applied Transforms<a class="headerlink" href="#randomly-applied-transforms" title="Permalink to this heading">¶</a></h2>
+<p>The following transforms are randomly-applied given a probability <code class="docutils literal notranslate"><span class="pre">p</span></code>.  That is, given <code class="docutils literal notranslate"><span class="pre">p</span> <span class="pre">=</span> <span class="pre">0.5</span></code>,
+there is a 50% chance to return the original image, and a 50% chance to return the transformed image,
+even when called with the same transform instance!</p>
+<section id="randomhorizontalflip">
+<h3>RandomHorizontalFlip<a class="headerlink" href="#randomhorizontalflip" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomHorizontalFlip.html#torchvision.transforms.RandomHorizontalFlip" title="torchvision.transforms.RandomHorizontalFlip"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomHorizontalFlip</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.hflip.html#torchvision.transforms.functional.hflip" title="torchvision.transforms.functional.hflip"><code class="xref py py-func docutils literal notranslate"><span class="pre">hflip()</span></code></a>)
+performs horizontal flip of an image, with a given probability.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">hflipper</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">hflipper</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_024.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_024.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomverticalflip">
+<h3>RandomVerticalFlip<a class="headerlink" href="#randomverticalflip" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomVerticalFlip.html#torchvision.transforms.RandomVerticalFlip" title="torchvision.transforms.RandomVerticalFlip"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomVerticalFlip</span></code></a> transform
+(see also <a class="reference internal" href="../../generated/torchvision.transforms.functional.vflip.html#torchvision.transforms.functional.vflip" title="torchvision.transforms.functional.vflip"><code class="xref py py-func docutils literal notranslate"><span class="pre">vflip()</span></code></a>)
+performs vertical flip of an image, with a given probability.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">vflipper</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomVerticalFlip</span></a><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">vflipper</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_025.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_025.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/></section>
+<section id="randomapply">
+<h3>RandomApply<a class="headerlink" href="#randomapply" title="Permalink to this heading">¶</a></h3>
+<p>The <a class="reference internal" href="../../generated/torchvision.transforms.RandomApply.html#torchvision.transforms.RandomApply" title="torchvision.transforms.RandomApply"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomApply</span></code></a> transform
+randomly applies a list of transforms, with a given probability.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">applier</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomApply</span></a><span class="p">(</span><span class="n">transforms</span><span class="o">=</span><span class="p">[</span><a href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module" class="sphx-glr-backref-module-torch-nn sphx-glr-backref-type-py-class"><span class="n">v2</span><span class="o">.</span><span class="n">RandomCrop</span></a><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">))],</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
+<a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a> <span class="o">=</span> <span class="p">[</span><span class="n">applier</span><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">4</span><span class="p">)]</span>
+<span class="n">plot</span><span class="p">([</span><a href="https://pillow.readthedocs.io/en/stable/reference/plugins.html#PIL.JpegImagePlugin.JpegImageFile" title="PIL.JpegImagePlugin.JpegImageFile" class="sphx-glr-backref-module-PIL-JpegImagePlugin sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">orig_img</span></a><span class="p">]</span> <span class="o">+</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">transformed_imgs</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<img src="../../_images/sphx_glr_plot_transforms_illustrations_026.png" srcset="../../_images/sphx_glr_plot_transforms_illustrations_026.png" alt="plot transforms illustrations" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 8.772 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-transforms-plot-transforms-illustrations-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/f9ef41d32e46834ba2f93a1d3c4b4568/plot_transforms_illustrations.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_transforms_illustrations.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/a8ef8e39164e980a89a9afdfc51d961e/plot_transforms_illustrations.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_transforms_illustrations.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_transforms_e2e.html" class="btn btn-neutral float-right" title="Transforms v2: End-to-end object detection/segmentation example" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_transforms_getting_started.html" class="btn btn-neutral" title="Getting started with transforms v2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Illustration of transforms</a><ul>
+<li><a class="reference internal" href="#geometric-transforms">Geometric Transforms</a><ul>
+<li><a class="reference internal" href="#pad">Pad</a></li>
+<li><a class="reference internal" href="#resize">Resize</a></li>
+<li><a class="reference internal" href="#centercrop">CenterCrop</a></li>
+<li><a class="reference internal" href="#fivecrop">FiveCrop</a></li>
+<li><a class="reference internal" href="#randomperspective">RandomPerspective</a></li>
+<li><a class="reference internal" href="#randomrotation">RandomRotation</a></li>
+<li><a class="reference internal" href="#randomaffine">RandomAffine</a></li>
+<li><a class="reference internal" href="#elastictransform">ElasticTransform</a></li>
+<li><a class="reference internal" href="#randomcrop">RandomCrop</a></li>
+<li><a class="reference internal" href="#randomresizedcrop">RandomResizedCrop</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#photometric-transforms">Photometric Transforms</a><ul>
+<li><a class="reference internal" href="#grayscale">Grayscale</a></li>
+<li><a class="reference internal" href="#colorjitter">ColorJitter</a></li>
+<li><a class="reference internal" href="#gaussianblur">GaussianBlur</a></li>
+<li><a class="reference internal" href="#randominvert">RandomInvert</a></li>
+<li><a class="reference internal" href="#randomposterize">RandomPosterize</a></li>
+<li><a class="reference internal" href="#randomsolarize">RandomSolarize</a></li>
+<li><a class="reference internal" href="#randomadjustsharpness">RandomAdjustSharpness</a></li>
+<li><a class="reference internal" href="#randomautocontrast">RandomAutocontrast</a></li>
+<li><a class="reference internal" href="#randomequalize">RandomEqualize</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#augmentation-transforms">Augmentation Transforms</a><ul>
+<li><a class="reference internal" href="#autoaugment">AutoAugment</a></li>
+<li><a class="reference internal" href="#randaugment">RandAugment</a></li>
+<li><a class="reference internal" href="#trivialaugmentwide">TrivialAugmentWide</a></li>
+<li><a class="reference internal" href="#augmix">AugMix</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#randomly-applied-transforms">Randomly-applied Transforms</a><ul>
+<li><a class="reference internal" href="#randomhorizontalflip">RandomHorizontalFlip</a></li>
+<li><a class="reference internal" href="#randomverticalflip">RandomVerticalFlip</a></li>
+<li><a class="reference internal" href="#randomapply">RandomApply</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/plot_tv_tensors.html b/0.16/auto_examples/transforms/plot_tv_tensors.html
new file mode 100644
index 00000000000..a7505414c78
--- /dev/null
+++ b/0.16/auto_examples/transforms/plot_tv_tensors.html
@@ -0,0 +1,925 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>TVTensors FAQ &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="How to write your own TVTensor class" href="plot_custom_tv_tensors.html" />
+    <link rel="prev" title="How to write your own v2 transforms" href="plot_custom_transforms.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../index.html">Examples and tutorials</a> &gt;</li>
+        
+          <li><a href="index.html">Transforms</a> &gt;</li>
+        
+      <li>TVTensors FAQ</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/plot_tv_tensors.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section class="sphx-glr-example-title" id="tvtensors-faq">
+<span id="sphx-glr-auto-examples-transforms-plot-tv-tensors-py"></span><h1>TVTensors FAQ<a class="headerlink" href="#tvtensors-faq" title="Permalink to this heading">¶</a></h1>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Try on <a class="reference external" href="https://colab.research.google.com/github/pytorch/vision/blob/gh-pages/main/_generated_ipynb_notebooks/plot_tv_tensors.ipynb">collab</a>
+or <a class="reference internal" href="#sphx-glr-download-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">go to the end</span></a> to download the full example code.</p>
+</div>
+<p>TVTensors are Tensor subclasses introduced together with
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>. This example showcases what these TVTensors are
+and how they behave.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p><strong>Intended Audience</strong> Unless you’re writing your own transforms or your own TVTensors, you
+probably do not need to read this guide. This is a fairly low-level topic
+that most users will not need to worry about: you do not need to understand
+the internals of TVTensors to efficiently rely on
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>. It may however be useful for advanced users
+trying to implement their own datasets, transforms, or work directly with
+the TVTensors.</p>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">PIL.Image</span>
+
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+</pre></div>
+</div>
+<section id="what-are-tvtensors">
+<h2>What are TVTensors?<a class="headerlink" href="#what-are-tvtensors" title="Permalink to this heading">¶</a></h2>
+<p>TVTensors are zero-copy tensor subclasses:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">tensor</span></a> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.rand.html#torch.rand" title="torch.rand" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">rand</span></a><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">)</span>
+<span class="n">image</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">tensor</span></a><span class="p">)</span>
+
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span></a><span class="p">)</span>
+<span class="k">assert</span> <span class="n">image</span><span class="o">.</span><span class="n">data_ptr</span><span class="p">()</span> <span class="o">==</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">tensor</span></a><span class="o">.</span><span class="n">data_ptr</span><span class="p">()</span>
+</pre></div>
+</div>
+<p>Under the hood, they are needed in <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> to correctly dispatch to the appropriate function
+for the input data.</p>
+<p><code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.tv_tensors</span></code> supports four types of TVTensors:</p>
+<ul class="simple">
+<li><p><a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a></p></li>
+<li><p><a class="reference internal" href="../../generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a></p></li>
+<li><p><a class="reference internal" href="../../generated/torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a></p></li>
+<li><p><a class="reference internal" href="../../generated/torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">Mask</span></code></a></p></li>
+</ul>
+</section>
+<section id="what-can-i-do-with-a-tvtensor">
+<h2>What can I do with a TVTensor?<a class="headerlink" href="#what-can-i-do-with-a-tvtensor" title="Permalink to this heading">¶</a></h2>
+<p>TVTensors look and feel just like regular tensors - they <strong>are</strong> tensors.
+Everything that is supported on a plain <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> like <code class="docutils literal notranslate"><span class="pre">.sum()</span></code> or
+any <code class="docutils literal notranslate"><span class="pre">torch.*</span></code> operator will also work on TVTensors. See
+<a class="reference internal" href="#tv-tensor-unwrapping-behaviour"><span class="std std-ref">I had a TVTensor but now I have a Tensor. Help!</span></a> for a few gotchas.</p>
+</section>
+<section id="how-do-i-construct-a-tvtensor">
+<span id="tv-tensor-creation"></span><h2>How do I construct a TVTensor?<a class="headerlink" href="#how-do-i-construct-a-tvtensor" title="Permalink to this heading">¶</a></h2>
+<section id="using-the-constructor">
+<h3>Using the constructor<a class="headerlink" href="#using-the-constructor" title="Permalink to this heading">¶</a></h3>
+<p>Each TVTensor class takes any tensor-like data that can be turned into a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">image</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">([[[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">]]]])</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Image([[[[0, 1],
+         [1, 0]]]], )
+</pre></div>
+</div>
+<p>Similar to other PyTorch creations ops, the constructor also takes the <code class="docutils literal notranslate"><span class="pre">dtype</span></code>, <code class="docutils literal notranslate"><span class="pre">device</span></code>, and <code class="docutils literal notranslate"><span class="pre">requires_grad</span></code>
+parameters.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">float_image</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">([[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">]]],</span> <span class="n">dtype</span><span class="o">=</span><a href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="torch.dtype" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">torch</span><span class="o">.</span><span class="n">float32</span></a><span class="p">,</span> <span class="n">requires_grad</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">float_image</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Image([[[0., 1.],
+        [1., 0.]]], grad_fn=&lt;AliasBackward0&gt;, )
+</pre></div>
+</div>
+<p>In addition, <a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a> and <a class="reference internal" href="../../generated/torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">Mask</span></code></a> can also take a
+<a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">PIL.Image.Image</span></code></a> directly:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">image</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">(</span><a href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.open" title="PIL.Image.open" class="sphx-glr-backref-module-PIL-Image sphx-glr-backref-type-py-function"><span class="n">PIL</span><span class="o">.</span><span class="n">Image</span><span class="o">.</span><span class="n">open</span></a><span class="p">(</span><span class="s2">&quot;../assets/astronaut.jpg&quot;</span><span class="p">))</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">,</span> <span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>torch.Size([3, 512, 512]) torch.uint8
+</pre></div>
+</div>
+<p>Some TVTensors require additional metadata to be passed in ordered to be constructed. For example,
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> requires the coordinate format as well as the size of the
+corresponding image (<code class="docutils literal notranslate"><span class="pre">canvas_size</span></code>) alongside the actual values. These
+metadata are required to properly transform the bounding boxes.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">bboxes</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">(</span>
+    <span class="p">[[</span><span class="mi">17</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">344</span><span class="p">,</span> <span class="mi">495</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">10</span><span class="p">]],</span>
+    <span class="nb">format</span><span class="o">=</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxFormat</span><span class="o">.</span><span class="n">XYXY</span><span class="p">,</span>
+    <span class="n">canvas_size</span><span class="o">=</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">:]</span>
+<span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">bboxes</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>BoundingBoxes([[ 17,  16, 344, 495],
+               [  0,  10,   0,  10]], format=BoundingBoxFormat.XYXY, canvas_size=torch.Size([512, 512]))
+</pre></div>
+</div>
+</section>
+<section id="using-tv-tensors-wrap">
+<h3>Using <code class="docutils literal notranslate"><span class="pre">tv_tensors.wrap()</span></code><a class="headerlink" href="#using-tv-tensors-wrap" title="Permalink to this heading">¶</a></h3>
+<p>You can also use the <a class="reference internal" href="../../generated/torchvision.tv_tensors.wrap.html#torchvision.tv_tensors.wrap" title="torchvision.tv_tensors.wrap"><code class="xref py py-func docutils literal notranslate"><span class="pre">wrap()</span></code></a> function to wrap a tensor object
+into a TVTensor. This is useful when you already have an object of the
+desired type, which typically happens when writing transforms: you just want
+to wrap the output like the input.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">new_bboxes</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/generated/torch.tensor.html#torch.tensor" title="torch.tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-function"><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span></a><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">20</span><span class="p">,</span> <span class="mi">30</span><span class="p">,</span> <span class="mi">40</span><span class="p">])</span>
+<span class="n">new_bboxes</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">new_bboxes</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">bboxes</span><span class="p">)</span>
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">new_bboxes</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">)</span>
+<span class="k">assert</span> <span class="n">new_bboxes</span><span class="o">.</span><span class="n">canvas_size</span> <span class="o">==</span> <span class="n">bboxes</span><span class="o">.</span><span class="n">canvas_size</span>
+</pre></div>
+</div>
+<p>The metadata of <code class="docutils literal notranslate"><span class="pre">new_bboxes</span></code> is the same as <code class="docutils literal notranslate"><span class="pre">bboxes</span></code>, but you could pass
+it as a parameter to override it.</p>
+</section>
+</section>
+<section id="i-had-a-tvtensor-but-now-i-have-a-tensor-help">
+<span id="tv-tensor-unwrapping-behaviour"></span><h2>I had a TVTensor but now I have a Tensor. Help!<a class="headerlink" href="#i-had-a-tvtensor-but-now-i-have-a-tensor-help" title="Permalink to this heading">¶</a></h2>
+<p>By default, operations on <a class="reference internal" href="../../generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a> objects
+will return a pure Tensor:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">bboxes</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">)</span>
+
+<span class="c1"># Shift bboxes by 3 pixels in both H and W</span>
+<span class="n">new_bboxes</span> <span class="o">=</span> <span class="n">bboxes</span> <span class="o">+</span> <span class="mi">3</span>
+
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">new_bboxes</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span></a><span class="p">)</span>
+<span class="k">assert</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">new_bboxes</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This behavior only affects native <code class="docutils literal notranslate"><span class="pre">torch</span></code> operations. If you are using
+the built-in <code class="docutils literal notranslate"><span class="pre">torchvision</span></code> transforms or functionals, you will always get
+as output the same type that you passed as input (pure <code class="docutils literal notranslate"><span class="pre">Tensor</span></code> or
+<code class="docutils literal notranslate"><span class="pre">TVTensor</span></code>).</p>
+</div>
+<section id="but-i-want-a-tvtensor-back">
+<h3>But I want a TVTensor back!<a class="headerlink" href="#but-i-want-a-tvtensor-back" title="Permalink to this heading">¶</a></h3>
+<p>You can re-wrap a pure tensor into a TVTensor by just calling the TVTensor
+constructor, or by using the <a class="reference internal" href="../../generated/torchvision.tv_tensors.wrap.html#torchvision.tv_tensors.wrap" title="torchvision.tv_tensors.wrap"><code class="xref py py-func docutils literal notranslate"><span class="pre">wrap()</span></code></a> function
+(see more details above in <a class="reference internal" href="#tv-tensor-creation"><span class="std std-ref">How do I construct a TVTensor?</span></a>):</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">new_bboxes</span> <span class="o">=</span> <span class="n">bboxes</span> <span class="o">+</span> <span class="mi">3</span>
+<span class="n">new_bboxes</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">new_bboxes</span><span class="p">,</span> <span class="n">like</span><span class="o">=</span><span class="n">bboxes</span><span class="p">)</span>
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">new_bboxes</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">)</span>
+</pre></div>
+</div>
+<p>Alternatively, you can use the <a class="reference internal" href="../../generated/torchvision.tv_tensors.set_return_type.html#torchvision.tv_tensors.set_return_type" title="torchvision.tv_tensors.set_return_type"><code class="xref py py-func docutils literal notranslate"><span class="pre">set_return_type()</span></code></a>
+as a global config setting for the whole program, or as a context manager
+(read its docs to learn more about caveats):</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">with</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">set_return_type</span><span class="p">(</span><span class="s2">&quot;TVTensor&quot;</span><span class="p">):</span>
+    <span class="n">new_bboxes</span> <span class="o">=</span> <span class="n">bboxes</span> <span class="o">+</span> <span class="mi">3</span>
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">new_bboxes</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span></a><span class="p">)</span>
+</pre></div>
+</div>
+</section>
+<section id="why-is-this-happening">
+<h3>Why is this happening?<a class="headerlink" href="#why-is-this-happening" title="Permalink to this heading">¶</a></h3>
+<p><strong>For performance reasons</strong>. <a class="reference internal" href="../../generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a>
+classes are Tensor subclasses, so any operation involving a
+<a class="reference internal" href="../../generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a> object will go through the
+<a class="reference external" href="https://pytorch.org/docs/stable/notes/extending.html#extending-torch">__torch_function__</a>
+protocol. This induces a small overhead, which we want to avoid when possible.
+This doesn’t matter for built-in <code class="docutils literal notranslate"><span class="pre">torchvision</span></code> transforms because we can
+avoid the overhead there, but it could be a problem in your model’s
+<code class="docutils literal notranslate"><span class="pre">forward</span></code>.</p>
+<p><strong>The alternative isn’t much better anyway.</strong> For every operation where
+preserving the <a class="reference internal" href="../../generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a> type makes
+sense, there are just as many operations where returning a pure Tensor is
+preferable: for example, is <code class="docutils literal notranslate"><span class="pre">img.sum()</span></code> still an <a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>?
+If we were to preserve <a class="reference internal" href="../../generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a> types all
+the way, even model’s logits or the output of the loss function would end up
+being of type <a class="reference internal" href="../../generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>, and surely that’s not
+desirable.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This behaviour is something we’re actively seeking feedback on. If you find this surprising or if you
+have any suggestions on how to better support your use-cases, please reach out to us via this issue:
+<a class="reference external" href="https://github.com/pytorch/vision/issues/7319">https://github.com/pytorch/vision/issues/7319</a></p>
+</div>
+</section>
+<section id="exceptions">
+<h3>Exceptions<a class="headerlink" href="#exceptions" title="Permalink to this heading">¶</a></h3>
+<p>There are a few exceptions to this “unwrapping” rule:
+<a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.Tensor.clone.html#torch.Tensor.clone" title="(in PyTorch v2.0)"><code class="xref py py-meth docutils literal notranslate"><span class="pre">clone()</span></code></a>, <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.Tensor.to.html#torch.Tensor.to" title="(in PyTorch v2.0)"><code class="xref py py-meth docutils literal notranslate"><span class="pre">to()</span></code></a>,
+<a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.Tensor.detach.html#torch.Tensor.detach" title="(in PyTorch v2.0)"><code class="xref py py-meth docutils literal notranslate"><span class="pre">torch.Tensor.detach()</span></code></a>, and <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.Tensor.requires_grad_.html#torch.Tensor.requires_grad_" title="(in PyTorch v2.0)"><code class="xref py py-meth docutils literal notranslate"><span class="pre">requires_grad_()</span></code></a> retain
+the TVTensor type.</p>
+<p>Inplace operations on TVTensors like <code class="docutils literal notranslate"><span class="pre">obj.add_()</span></code> will preserve the type of
+<code class="docutils literal notranslate"><span class="pre">obj</span></code>. However, the <strong>returned</strong> value of inplace operations will be a pure
+tensor:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">image</span> <span class="o">=</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">([[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">]]])</span>
+
+<a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">new_image</span></a> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+
+<span class="c1"># image got transformed in-place and is still a TVTensor Image, but new_image</span>
+<span class="c1"># is a Tensor. They share the same underlying data and they&#39;re equal, just</span>
+<span class="c1"># different classes.</span>
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">)</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+
+<span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">new_image</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span></a><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">new_image</span></a><span class="p">,</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class"><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span></a><span class="p">)</span>
+<span class="k">assert</span> <span class="p">(</span><a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">new_image</span></a> <span class="o">==</span> <span class="n">image</span><span class="p">)</span><span class="o">.</span><span class="n">all</span><span class="p">()</span>
+<span class="k">assert</span> <a href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="torch.Tensor" class="sphx-glr-backref-module-torch sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">new_image</span></a><span class="o">.</span><span class="n">data_ptr</span><span class="p">()</span> <span class="o">==</span> <span class="n">image</span><span class="o">.</span><span class="n">data_ptr</span><span class="p">()</span>
+</pre></div>
+</div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Image([[[2, 4],
+        [4, 2]]], )
+</pre></div>
+</div>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> (0 minutes 0.008 seconds)</p>
+<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-transforms-plot-tv-tensors-py">
+<div class="sphx-glr-download sphx-glr-download-python docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/92763696a071eab76a5ea62a1d9f8c5c/plot_tv_tensors.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_tv_tensors.py</span></code></a></p>
+</div>
+<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
+<p><a class="reference download internal" download="" href="../../_downloads/bf35a573bbfe2773c19cd458bf8508c4/plot_tv_tensors.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_tv_tensors.ipynb</span></code></a></p>
+</div>
+</div>
+<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
+</section>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="plot_custom_tv_tensors.html" class="btn btn-neutral float-right" title="How to write your own TVTensor class" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="plot_custom_transforms.html" class="btn btn-neutral" title="How to write your own v2 transforms" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">TVTensors FAQ</a><ul>
+<li><a class="reference internal" href="#what-are-tvtensors">What are TVTensors?</a></li>
+<li><a class="reference internal" href="#what-can-i-do-with-a-tvtensor">What can I do with a TVTensor?</a></li>
+<li><a class="reference internal" href="#how-do-i-construct-a-tvtensor">How do I construct a TVTensor?</a><ul>
+<li><a class="reference internal" href="#using-the-constructor">Using the constructor</a></li>
+<li><a class="reference internal" href="#using-tv-tensors-wrap">Using <code class="docutils literal notranslate"><span class="pre">tv_tensors.wrap()</span></code></a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#i-had-a-tvtensor-but-now-i-have-a-tensor-help">I had a TVTensor but now I have a Tensor. Help!</a><ul>
+<li><a class="reference internal" href="#but-i-want-a-tvtensor-back">But I want a TVTensor back!</a></li>
+<li><a class="reference internal" href="#why-is-this-happening">Why is this happening?</a></li>
+<li><a class="reference internal" href="#exceptions">Exceptions</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/auto_examples/transforms/sg_execution_times.html b/0.16/auto_examples/transforms/sg_execution_times.html
new file mode 100644
index 00000000000..214f282b485
--- /dev/null
+++ b/0.16/auto_examples/transforms/sg_execution_times.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Computation times &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Computation times</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/auto_examples/transforms/sg_execution_times.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="computation-times">
+<span id="sphx-glr-auto-examples-transforms-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this heading">¶</a></h1>
+<p><strong>00:14.597</strong> total execution time for <strong>auto_examples_transforms</strong> files:</p>
+<table class="docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_transforms_illustrations.py</span></code>)</p></td>
+<td><p>00:08.772</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_transforms_e2e.py</span></code>)</p></td>
+<td><p>00:04.813</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_transforms_getting_started.py</span></code>)</p></td>
+<td><p>00:00.838</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_cutmix_mixup.py</span></code>)</p></td>
+<td><p>00:00.155</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_tv_tensors.py</span></code>)</p></td>
+<td><p>00:00.008</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_custom_transforms.py</span></code>)</p></td>
+<td><p>00:00.006</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a> (<code class="docutils literal notranslate"><span class="pre">plot_custom_tv_tensors.py</span></code>)</p></td>
+<td><p>00:00.004</p></td>
+<td><p>0.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Computation times</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/datasets.html b/0.16/datasets.html
new file mode 100644
index 00000000000..32f76faaeb9
--- /dev/null
+++ b/0.16/datasets.html
@@ -0,0 +1,1011 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Datasets &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="Caltech101" href="generated/torchvision.datasets.Caltech101.html" />
+    <link rel="prev" title="raft_small" href="models/generated/torchvision.models.optical_flow.raft_small.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Datasets</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/datasets.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="datasets">
+<span id="id1"></span><h1>Datasets<a class="headerlink" href="#datasets" title="Permalink to this heading">¶</a></h1>
+<p>Torchvision provides many built-in datasets in the <code class="docutils literal notranslate"><span class="pre">torchvision.datasets</span></code>
+module, as well as utility classes for building your own datasets.</p>
+<section id="built-in-datasets">
+<h2>Built-in datasets<a class="headerlink" href="#built-in-datasets" title="Permalink to this heading">¶</a></h2>
+<p>All datasets are subclasses of <a class="reference external" href="https://pytorch.org/docs/stable/data.html#torch.utils.data.Dataset" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.utils.data.Dataset</span></code></a>
+i.e, they have <code class="docutils literal notranslate"><span class="pre">__getitem__</span></code> and <code class="docutils literal notranslate"><span class="pre">__len__</span></code> methods implemented.
+Hence, they can all be passed to a <a class="reference external" href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.utils.data.DataLoader</span></code></a>
+which can load multiple samples in parallel using <code class="docutils literal notranslate"><span class="pre">torch.multiprocessing</span></code> workers.
+For example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">imagenet_data</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">ImageNet</span><span class="p">(</span><span class="s1">&#39;path/to/imagenet_root/&#39;</span><span class="p">)</span>
+<span class="n">data_loader</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">DataLoader</span><span class="p">(</span><span class="n">imagenet_data</span><span class="p">,</span>
+                                          <span class="n">batch_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+                                          <span class="n">shuffle</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                                          <span class="n">num_workers</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">nThreads</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>All the datasets have almost similar API. They all have two common arguments:
+<code class="docutils literal notranslate"><span class="pre">transform</span></code> and  <code class="docutils literal notranslate"><span class="pre">target_transform</span></code> to transform the input and target respectively.
+You can also create your own datasets using the provided <a class="reference internal" href="#base-classes-datasets"><span class="std std-ref">base classes</span></a>.</p>
+<section id="image-classification">
+<h3>Image classification<a class="headerlink" href="#image-classification" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Caltech101.html#torchvision.datasets.Caltech101" title="torchvision.datasets.Caltech101"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Caltech101</span></code></a>(root[, target_type, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://data.caltech.edu/records/20086">Caltech 101</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Caltech256.html#torchvision.datasets.Caltech256" title="torchvision.datasets.Caltech256"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Caltech256</span></code></a>(root[, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://data.caltech.edu/records/20087">Caltech 256</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.CelebA.html#torchvision.datasets.CelebA" title="torchvision.datasets.CelebA"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CelebA</span></code></a>(root[, split, target_type, ...])</p></td>
+<td><p><a class="reference external" href="http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html">Large-scale CelebFaces Attributes (CelebA) Dataset</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.CIFAR10.html#torchvision.datasets.CIFAR10" title="torchvision.datasets.CIFAR10"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CIFAR10</span></code></a>(root[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://www.cs.toronto.edu/~kriz/cifar.html">CIFAR10</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.CIFAR100.html#torchvision.datasets.CIFAR100" title="torchvision.datasets.CIFAR100"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CIFAR100</span></code></a>(root[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://www.cs.toronto.edu/~kriz/cifar.html">CIFAR100</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Country211.html#torchvision.datasets.Country211" title="torchvision.datasets.Country211"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Country211</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://github.com/openai/CLIP/blob/main/data/country211.md">The Country211 Data Set</a> from OpenAI.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.DTD.html#torchvision.datasets.DTD" title="torchvision.datasets.DTD"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DTD</span></code></a>(root[, split, partition, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/dtd/">Describable Textures Dataset (DTD)</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.EMNIST.html#torchvision.datasets.EMNIST" title="torchvision.datasets.EMNIST"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EMNIST</span></code></a>(root, split, **kwargs)</p></td>
+<td><p><a class="reference external" href="https://www.westernsydney.edu.au/bens/home/reproducible_research/emnist">EMNIST</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.EuroSAT.html#torchvision.datasets.EuroSAT" title="torchvision.datasets.EuroSAT"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EuroSAT</span></code></a>(root[, transform, target_transform, ...])</p></td>
+<td><p>RGB version of the <a class="reference external" href="https://github.com/phelber/eurosat">EuroSAT</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.FakeData.html#torchvision.datasets.FakeData" title="torchvision.datasets.FakeData"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FakeData</span></code></a>([size, image_size, num_classes, ...])</p></td>
+<td><p>A fake dataset that returns randomly generated images and returns them as PIL images</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.FashionMNIST.html#torchvision.datasets.FashionMNIST" title="torchvision.datasets.FashionMNIST"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FashionMNIST</span></code></a>(root[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://github.com/zalandoresearch/fashion-mnist">Fashion-MNIST</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.FER2013.html#torchvision.datasets.FER2013" title="torchvision.datasets.FER2013"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FER2013</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge">FER2013</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.FGVCAircraft.html#torchvision.datasets.FGVCAircraft" title="torchvision.datasets.FGVCAircraft"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FGVCAircraft</span></code></a>(root[, split, ...])</p></td>
+<td><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/">FGVC Aircraft</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Flickr8k.html#torchvision.datasets.Flickr8k" title="torchvision.datasets.Flickr8k"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Flickr8k</span></code></a>(root, ann_file[, transform, ...])</p></td>
+<td><p><a class="reference external" href="http://hockenmaier.cs.illinois.edu/8k-pictures.html">Flickr8k Entities</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Flickr30k.html#torchvision.datasets.Flickr30k" title="torchvision.datasets.Flickr30k"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Flickr30k</span></code></a>(root, ann_file[, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://bryanplummer.com/Flickr30kEntities/">Flickr30k Entities</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Flowers102.html#torchvision.datasets.Flowers102" title="torchvision.datasets.Flowers102"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Flowers102</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/flowers/102/">Oxford 102 Flower</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Food101.html#torchvision.datasets.Food101" title="torchvision.datasets.Food101"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Food101</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/">The Food-101 Data Set</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.GTSRB.html#torchvision.datasets.GTSRB" title="torchvision.datasets.GTSRB"><code class="xref py py-obj docutils literal notranslate"><span class="pre">GTSRB</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://benchmark.ini.rub.de/">German Traffic Sign Recognition Benchmark (GTSRB)</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.INaturalist.html#torchvision.datasets.INaturalist" title="torchvision.datasets.INaturalist"><code class="xref py py-obj docutils literal notranslate"><span class="pre">INaturalist</span></code></a>(root[, version, target_type, ...])</p></td>
+<td><p><a class="reference external" href="https://github.com/visipedia/inat_comp">iNaturalist</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.ImageNet.html#torchvision.datasets.ImageNet" title="torchvision.datasets.ImageNet"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ImageNet</span></code></a>(root[, split])</p></td>
+<td><p><a class="reference external" href="http://image-net.org/">ImageNet</a> 2012 Classification Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.KMNIST.html#torchvision.datasets.KMNIST" title="torchvision.datasets.KMNIST"><code class="xref py py-obj docutils literal notranslate"><span class="pre">KMNIST</span></code></a>(root[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://github.com/rois-codh/kmnist">Kuzushiji-MNIST</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.LFWPeople.html#torchvision.datasets.LFWPeople" title="torchvision.datasets.LFWPeople"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LFWPeople</span></code></a>(root[, split, image_set, ...])</p></td>
+<td><p><a class="reference external" href="http://vis-www.cs.umass.edu/lfw/">LFW</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.LSUN.html#torchvision.datasets.LSUN" title="torchvision.datasets.LSUN"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LSUN</span></code></a>(root[, classes, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://www.yf.io/p/lsun">LSUN</a> dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.MNIST.html#torchvision.datasets.MNIST" title="torchvision.datasets.MNIST"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MNIST</span></code></a>(root[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="http://yann.lecun.com/exdb/mnist/">MNIST</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Omniglot.html#torchvision.datasets.Omniglot" title="torchvision.datasets.Omniglot"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Omniglot</span></code></a>(root[, background, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://github.com/brendenlake/omniglot">Omniglot</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.OxfordIIITPet.html#torchvision.datasets.OxfordIIITPet" title="torchvision.datasets.OxfordIIITPet"><code class="xref py py-obj docutils literal notranslate"><span class="pre">OxfordIIITPet</span></code></a>(root[, split, target_types, ...])</p></td>
+<td><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/pets/">Oxford-IIIT Pet Dataset</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Places365.html#torchvision.datasets.Places365" title="torchvision.datasets.Places365"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Places365</span></code></a>(root, split, small, download, ...)</p></td>
+<td><p><a class="reference external" href="http://places2.csail.mit.edu/index.html">Places365</a> classification dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.PCAM.html#torchvision.datasets.PCAM" title="torchvision.datasets.PCAM"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PCAM</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://github.com/basveeling/pcam">PCAM Dataset</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.QMNIST.html#torchvision.datasets.QMNIST" title="torchvision.datasets.QMNIST"><code class="xref py py-obj docutils literal notranslate"><span class="pre">QMNIST</span></code></a>(root[, what, compat, train])</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/qmnist">QMNIST</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.RenderedSST2.html#torchvision.datasets.RenderedSST2" title="torchvision.datasets.RenderedSST2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RenderedSST2</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://github.com/openai/CLIP/blob/main/data/rendered-sst2.md">The Rendered SST2 Dataset</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.SEMEION.html#torchvision.datasets.SEMEION" title="torchvision.datasets.SEMEION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SEMEION</span></code></a>(root[, transform, target_transform, ...])</p></td>
+<td><p><a class="reference external" href="http://archive.ics.uci.edu/ml/datasets/semeion+handwritten+digit">SEMEION</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.SBU.html#torchvision.datasets.SBU" title="torchvision.datasets.SBU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SBU</span></code></a>(root[, transform, target_transform, ...])</p></td>
+<td><p><a class="reference external" href="http://www.cs.virginia.edu/~vicente/sbucaptions/">SBU Captioned Photo</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.StanfordCars.html#torchvision.datasets.StanfordCars" title="torchvision.datasets.StanfordCars"><code class="xref py py-obj docutils literal notranslate"><span class="pre">StanfordCars</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://ai.stanford.edu/~jkrause/cars/car_dataset.html">Stanford Cars</a> Dataset</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.STL10.html#torchvision.datasets.STL10" title="torchvision.datasets.STL10"><code class="xref py py-obj docutils literal notranslate"><span class="pre">STL10</span></code></a>(root[, split, folds, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://cs.stanford.edu/~acoates/stl10/">STL10</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.SUN397.html#torchvision.datasets.SUN397" title="torchvision.datasets.SUN397"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SUN397</span></code></a>(root[, transform, target_transform, ...])</p></td>
+<td><p><a class="reference external" href="https://vision.princeton.edu/projects/2010/SUN/">The SUN397 Data Set</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.SVHN.html#torchvision.datasets.SVHN" title="torchvision.datasets.SVHN"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SVHN</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="http://ufldl.stanford.edu/housenumbers/">SVHN</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.USPS.html#torchvision.datasets.USPS" title="torchvision.datasets.USPS"><code class="xref py py-obj docutils literal notranslate"><span class="pre">USPS</span></code></a>(root[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#usps">USPS</a> Dataset.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="image-detection-or-segmentation">
+<h3>Image detection or segmentation<a class="headerlink" href="#image-detection-or-segmentation" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection" title="torchvision.datasets.CocoDetection"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CocoDetection</span></code></a>(root, annFile[, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://cocodataset.org/#detection-2016">MS Coco Detection</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.CelebA.html#torchvision.datasets.CelebA" title="torchvision.datasets.CelebA"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CelebA</span></code></a>(root[, split, target_type, ...])</p></td>
+<td><p><p><a class="reference external" href="http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html">Large-scale CelebFaces Attributes (CelebA) Dataset</a> Dataset.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Cityscapes.html#torchvision.datasets.Cityscapes" title="torchvision.datasets.Cityscapes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Cityscapes</span></code></a>(root[, split, mode, target_type, ...])</p></td>
+<td><p><a class="reference external" href="http://www.cityscapes-dataset.com/">Cityscapes</a> Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Kitti.html#torchvision.datasets.Kitti" title="torchvision.datasets.Kitti"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Kitti</span></code></a>(root[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark">KITTI</a> Dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.OxfordIIITPet.html#torchvision.datasets.OxfordIIITPet" title="torchvision.datasets.OxfordIIITPet"><code class="xref py py-obj docutils literal notranslate"><span class="pre">OxfordIIITPet</span></code></a>(root[, split, target_types, ...])</p></td>
+<td><p><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/pets/">Oxford-IIIT Pet Dataset</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.SBDataset.html#torchvision.datasets.SBDataset" title="torchvision.datasets.SBDataset"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SBDataset</span></code></a>(root[, image_set, mode, download, ...])</p></td>
+<td><p><a class="reference external" href="http://home.bharathh.info/pubs/codes/SBD/download.html">Semantic Boundaries Dataset</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.VOCSegmentation.html#torchvision.datasets.VOCSegmentation" title="torchvision.datasets.VOCSegmentation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VOCSegmentation</span></code></a>(root[, year, image_set, ...])</p></td>
+<td><p><a class="reference external" href="http://host.robots.ox.ac.uk/pascal/VOC/">Pascal VOC</a> Segmentation Dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.VOCDetection.html#torchvision.datasets.VOCDetection" title="torchvision.datasets.VOCDetection"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VOCDetection</span></code></a>(root[, year, image_set, ...])</p></td>
+<td><p><p><a class="reference external" href="http://host.robots.ox.ac.uk/pascal/VOC/">Pascal VOC</a> Detection Dataset.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.WIDERFace.html#torchvision.datasets.WIDERFace" title="torchvision.datasets.WIDERFace"><code class="xref py py-obj docutils literal notranslate"><span class="pre">WIDERFace</span></code></a>(root[, split, transform, ...])</p></td>
+<td><p><a class="reference external" href="http://shuoyang1213.me/WIDERFACE/">WIDERFace</a> Dataset.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="optical-flow">
+<h3>Optical Flow<a class="headerlink" href="#optical-flow" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a>(root[, split, transforms])</p></td>
+<td><p><a class="reference external" href="https://lmb.informatik.uni-freiburg.de/resources/datasets/FlyingChairs.en.html#flyingchairs">FlyingChairs</a> Dataset for optical flow.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a>(root[, split, pass_name, ...])</p></td>
+<td><p><a class="reference external" href="https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html">FlyingThings3D</a> dataset for optical flow.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.HD1K.html#torchvision.datasets.HD1K" title="torchvision.datasets.HD1K"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HD1K</span></code></a>(root[, split, transforms])</p></td>
+<td><p><a class="reference external" href="http://hci-benchmark.iwr.uni-heidelberg.de/">HD1K</a> dataset for optical flow.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-obj docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>(root[, split, transforms])</p></td>
+<td><p><a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_scene_flow.php?benchmark=flow">KITTI</a> dataset for optical flow (2015).</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Sintel.html#torchvision.datasets.Sintel" title="torchvision.datasets.Sintel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Sintel</span></code></a>(root[, split, pass_name, transforms])</p></td>
+<td><p><a class="reference external" href="http://sintel.is.tue.mpg.de/">Sintel</a> Dataset for optical flow.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="stereo-matching">
+<h3>Stereo Matching<a class="headerlink" href="#stereo-matching" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.CarlaStereo.html#torchvision.datasets.CarlaStereo" title="torchvision.datasets.CarlaStereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CarlaStereo</span></code></a>(root[, transforms])</p></td>
+<td><p>Carla simulator data linked in the <a class="reference external" href="https://github.com/megvii-research/CREStereo">CREStereo github repo</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Kitti2012Stereo.html#torchvision.datasets.Kitti2012Stereo" title="torchvision.datasets.Kitti2012Stereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Kitti2012Stereo</span></code></a>(root[, split, transforms])</p></td>
+<td><p>KITTI dataset from the <a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_stereo_flow.php">2012 stereo evaluation benchmark</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.Kitti2015Stereo.html#torchvision.datasets.Kitti2015Stereo" title="torchvision.datasets.Kitti2015Stereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Kitti2015Stereo</span></code></a>(root[, split, transforms])</p></td>
+<td><p>KITTI dataset from the <a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_scene_flow.php">2015 stereo evaluation benchmark</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.CREStereo.html#torchvision.datasets.CREStereo" title="torchvision.datasets.CREStereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CREStereo</span></code></a>(root[, transforms])</p></td>
+<td><p>Synthetic dataset used in training the <a class="reference external" href="https://arxiv.org/pdf/2203.11483.pdf">CREStereo</a> architecture.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.FallingThingsStereo.html#torchvision.datasets.FallingThingsStereo" title="torchvision.datasets.FallingThingsStereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FallingThingsStereo</span></code></a>(root[, variant, transforms])</p></td>
+<td><p><a class="reference external" href="https://research.nvidia.com/publication/2018-06_falling-things-synthetic-dataset-3d-object-detection-and-pose-estimation">FallingThings</a> dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.SceneFlowStereo.html#torchvision.datasets.SceneFlowStereo" title="torchvision.datasets.SceneFlowStereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SceneFlowStereo</span></code></a>(root[, variant, pass_name, ...])</p></td>
+<td><p>Dataset interface for <a class="reference external" href="https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html">Scene Flow</a> datasets.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.SintelStereo.html#torchvision.datasets.SintelStereo" title="torchvision.datasets.SintelStereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SintelStereo</span></code></a>(root[, pass_name, transforms])</p></td>
+<td><p>Sintel <a class="reference external" href="http://sintel.is.tue.mpg.de/stereo">Stereo Dataset</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.InStereo2k.html#torchvision.datasets.InStereo2k" title="torchvision.datasets.InStereo2k"><code class="xref py py-obj docutils literal notranslate"><span class="pre">InStereo2k</span></code></a>(root[, split, transforms])</p></td>
+<td><p><a class="reference external" href="https://github.com/YuhuaXu/StereoDataset">InStereo2k</a> dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.ETH3DStereo.html#torchvision.datasets.ETH3DStereo" title="torchvision.datasets.ETH3DStereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ETH3DStereo</span></code></a>(root[, split, transforms])</p></td>
+<td><p>ETH3D <a class="reference external" href="https://www.eth3d.net/datasets">Low-Res Two-View</a> dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Middlebury2014Stereo.html#torchvision.datasets.Middlebury2014Stereo" title="torchvision.datasets.Middlebury2014Stereo"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Middlebury2014Stereo</span></code></a>(root[, split, ...])</p></td>
+<td><p>Publicly available scenes from the Middlebury dataset <cite>2014 version &lt;https://vision.middlebury.edu/stereo/data/scenes2014/&gt;</cite>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="image-pairs">
+<h3>Image pairs<a class="headerlink" href="#image-pairs" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.LFWPairs.html#torchvision.datasets.LFWPairs" title="torchvision.datasets.LFWPairs"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LFWPairs</span></code></a>(root[, split, image_set, ...])</p></td>
+<td><p><p><a class="reference external" href="http://vis-www.cs.umass.edu/lfw/">LFW</a> Dataset.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.PhotoTour.html#torchvision.datasets.PhotoTour" title="torchvision.datasets.PhotoTour"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PhotoTour</span></code></a>(root, name[, train, transform, ...])</p></td>
+<td><p><a class="reference external" href="http://matthewalunbrown.com/patchdata/patchdata.html">Multi-view Stereo Correspondence</a> Dataset.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="image-captioning">
+<h3>Image captioning<a class="headerlink" href="#image-captioning" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.CocoCaptions.html#torchvision.datasets.CocoCaptions" title="torchvision.datasets.CocoCaptions"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CocoCaptions</span></code></a>(root, annFile[, transform, ...])</p></td>
+<td><p><a class="reference external" href="https://cocodataset.org/#captions-2015">MS Coco Captions</a> Dataset.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="video-classification">
+<h3>Video classification<a class="headerlink" href="#video-classification" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.HMDB51.html#torchvision.datasets.HMDB51" title="torchvision.datasets.HMDB51"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HMDB51</span></code></a>(root, annotation_path, frames_per_clip)</p></td>
+<td><p><a class="reference external" href="https://serre-lab.clps.brown.edu/resource/hmdb-a-large-human-motion-database/">HMDB51</a> dataset.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.Kinetics.html#torchvision.datasets.Kinetics" title="torchvision.datasets.Kinetics"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Kinetics</span></code></a>(root, frames_per_clip[, ...])</p></td>
+<td><p><a class="reference external" href="https://www.deepmind.com/open-source/kinetics">Generic Kinetics</a> dataset.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.UCF101.html#torchvision.datasets.UCF101" title="torchvision.datasets.UCF101"><code class="xref py py-obj docutils literal notranslate"><span class="pre">UCF101</span></code></a>(root, annotation_path, frames_per_clip)</p></td>
+<td><p><a class="reference external" href="https://www.crcv.ucf.edu/data/UCF101.php">UCF101</a> dataset.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="video-prediction">
+<h3>Video prediction<a class="headerlink" href="#video-prediction" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.MovingMNIST.html#torchvision.datasets.MovingMNIST" title="torchvision.datasets.MovingMNIST"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MovingMNIST</span></code></a>(root[, split, split_ratio, ...])</p></td>
+<td><p><a class="reference external" href="http://www.cs.toronto.edu/~nitish/unsupervised_video/">MovingMNIST</a> Dataset.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+<section id="base-classes-for-custom-datasets">
+<span id="base-classes-datasets"></span><h2>Base classes for custom datasets<a class="headerlink" href="#base-classes-for-custom-datasets" title="Permalink to this heading">¶</a></h2>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder" title="torchvision.datasets.DatasetFolder"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DatasetFolder</span></code></a>(root, loader[, extensions, ...])</p></td>
+<td><p>A generic data loader.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.datasets.ImageFolder.html#torchvision.datasets.ImageFolder" title="torchvision.datasets.ImageFolder"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ImageFolder</span></code></a>(root, transform, ...)</p></td>
+<td><p>A generic data loader where the images are arranged in this way by default: .</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.VisionDataset.html#torchvision.datasets.VisionDataset" title="torchvision.datasets.VisionDataset"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VisionDataset</span></code></a>(root[, transforms, transform, ...])</p></td>
+<td><p>Base Class For making datasets which are compatible with torchvision.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="transforms-v2">
+<h2>Transforms v2<a class="headerlink" href="#transforms-v2" title="Permalink to this heading">¶</a></h2>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html#torchvision.datasets.wrap_dataset_for_transforms_v2" title="torchvision.datasets.wrap_dataset_for_transforms_v2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">wrap_dataset_for_transforms_v2</span></code></a>(dataset[, ...])</p></td>
+<td><p>[BETA] Wrap a <code class="docutils literal notranslate"><span class="pre">torchvision.dataset</span></code> for usage with <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.datasets.Caltech101.html" class="btn btn-neutral float-right" title="Caltech101" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="models/generated/torchvision.models.optical_flow.raft_small.html" class="btn btn-neutral" title="raft_small" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Datasets</a><ul>
+<li><a class="reference internal" href="#built-in-datasets">Built-in datasets</a><ul>
+<li><a class="reference internal" href="#image-classification">Image classification</a></li>
+<li><a class="reference internal" href="#image-detection-or-segmentation">Image detection or segmentation</a></li>
+<li><a class="reference internal" href="#optical-flow">Optical Flow</a></li>
+<li><a class="reference internal" href="#stereo-matching">Stereo Matching</a></li>
+<li><a class="reference internal" href="#image-pairs">Image pairs</a></li>
+<li><a class="reference internal" href="#image-captioning">Image captioning</a></li>
+<li><a class="reference internal" href="#video-classification">Video classification</a></li>
+<li><a class="reference internal" href="#video-prediction">Video prediction</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#base-classes-for-custom-datasets">Base classes for custom datasets</a></li>
+<li><a class="reference internal" href="#transforms-v2">Transforms v2</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/feature_extraction.html b/0.16/feature_extraction.html
new file mode 100644
index 00000000000..be882368842
--- /dev/null
+++ b/0.16/feature_extraction.html
@@ -0,0 +1,846 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Feature extraction for model inspection &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="create_feature_extractor" href="generated/torchvision.models.feature_extraction.create_feature_extractor.html" />
+    <link rel="prev" title="VideoReader" href="generated/torchvision.io.VideoReader.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Feature extraction for model inspection</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/feature_extraction.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="feature-extraction-for-model-inspection">
+<h1>Feature extraction for model inspection<a class="headerlink" href="#feature-extraction-for-model-inspection" title="Permalink to this heading">¶</a></h1>
+<p>The <code class="docutils literal notranslate"><span class="pre">torchvision.models.feature_extraction</span></code> package contains
+feature extraction utilities that let us tap into our models to access intermediate
+transformations of our inputs. This could be useful for a variety of
+applications in computer vision. Just a few examples are:</p>
+<ul class="simple">
+<li><p>Visualizing feature maps.</p></li>
+<li><p>Extracting features to compute image descriptors for tasks like facial
+recognition, copy-detection, or image retrieval.</p></li>
+<li><p>Passing selected features to downstream sub-networks for end-to-end training
+with a specific task in mind. For example, passing a hierarchy of features
+to a Feature Pyramid Network with object detection heads.</p></li>
+</ul>
+<p>Torchvision provides <a class="reference internal" href="generated/torchvision.models.feature_extraction.create_feature_extractor.html#torchvision.models.feature_extraction.create_feature_extractor" title="torchvision.models.feature_extraction.create_feature_extractor"><code class="xref py py-func docutils literal notranslate"><span class="pre">create_feature_extractor()</span></code></a> for this purpose.
+It works by following roughly these steps:</p>
+<ol class="arabic simple">
+<li><p>Symbolically tracing the model to get a graphical representation of
+how it transforms the input, step by step.</p></li>
+<li><p>Setting the user-selected graph nodes as outputs.</p></li>
+<li><p>Removing all redundant nodes (anything downstream of the output nodes).</p></li>
+<li><p>Generating python code from the resulting graph and bundling that into a
+PyTorch module together with the graph itself.</p></li>
+</ol>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>The <a class="reference external" href="https://pytorch.org/docs/stable/fx.html">torch.fx documentation</a>
+provides a more general and detailed explanation of the above procedure and
+the inner workings of the symbolic tracing.</p>
+<p id="about-node-names"><strong>About Node Names</strong></p>
+<p>In order to specify which nodes should be output nodes for extracted
+features, one should be familiar with the node naming convention used here
+(which differs slightly from that used in <code class="docutils literal notranslate"><span class="pre">torch.fx</span></code>). A node name is
+specified as a <code class="docutils literal notranslate"><span class="pre">.</span></code> separated path walking the module hierarchy from top level
+module down to leaf operation or leaf module. For instance <code class="docutils literal notranslate"><span class="pre">&quot;layer4.2.relu&quot;</span></code>
+in ResNet-50 represents the output of the ReLU of the 2nd block of the 4th
+layer of the <code class="docutils literal notranslate"><span class="pre">ResNet</span></code> module. Here are some finer points to keep in mind:</p>
+<ul class="simple">
+<li><p>When specifying node names for <a class="reference internal" href="generated/torchvision.models.feature_extraction.create_feature_extractor.html#torchvision.models.feature_extraction.create_feature_extractor" title="torchvision.models.feature_extraction.create_feature_extractor"><code class="xref py py-func docutils literal notranslate"><span class="pre">create_feature_extractor()</span></code></a>, you may
+provide a truncated version of a node name as a shortcut. To see how this
+works, try creating a ResNet-50 model and printing the node names with
+<code class="docutils literal notranslate"><span class="pre">train_nodes,</span> <span class="pre">_</span> <span class="pre">=</span> <span class="pre">get_graph_node_names(model)</span> <span class="pre">print(train_nodes)</span></code> and
+observe that the last node pertaining to <code class="docutils literal notranslate"><span class="pre">layer4</span></code> is
+<code class="docutils literal notranslate"><span class="pre">&quot;layer4.2.relu_2&quot;</span></code>. One may specify <code class="docutils literal notranslate"><span class="pre">&quot;layer4.2.relu_2&quot;</span></code> as the return
+node, or just <code class="docutils literal notranslate"><span class="pre">&quot;layer4&quot;</span></code> as this, by convention, refers to the last node
+(in order of execution) of <code class="docutils literal notranslate"><span class="pre">layer4</span></code>.</p></li>
+<li><p>If a certain module or operation is repeated more than once, node names get
+an additional <code class="docutils literal notranslate"><span class="pre">_{int}</span></code> postfix to disambiguate. For instance, maybe the
+addition (<code class="docutils literal notranslate"><span class="pre">+</span></code>) operation is used three times in the same <code class="docutils literal notranslate"><span class="pre">forward</span></code>
+method. Then there would be <code class="docutils literal notranslate"><span class="pre">&quot;path.to.module.add&quot;</span></code>,
+<code class="docutils literal notranslate"><span class="pre">&quot;path.to.module.add_1&quot;</span></code>, <code class="docutils literal notranslate"><span class="pre">&quot;path.to.module.add_2&quot;</span></code>. The counter is
+maintained within the scope of the direct parent. So in ResNet-50 there is
+a <code class="docutils literal notranslate"><span class="pre">&quot;layer4.1.add&quot;</span></code> and a <code class="docutils literal notranslate"><span class="pre">&quot;layer4.2.add&quot;</span></code>. Because the addition
+operations reside in different blocks, there is no need for a postfix to
+disambiguate.</p></li>
+</ul>
+<p><strong>An Example</strong></p>
+<p>Here is an example of how we might extract features for MaskRCNN:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision.models</span> <span class="kn">import</span> <span class="n">resnet50</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.feature_extraction</span> <span class="kn">import</span> <span class="n">get_graph_node_names</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.feature_extraction</span> <span class="kn">import</span> <span class="n">create_feature_extractor</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.detection.mask_rcnn</span> <span class="kn">import</span> <span class="n">MaskRCNN</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.detection.backbone_utils</span> <span class="kn">import</span> <span class="n">LastLevelMaxPool</span>
+<span class="kn">from</span> <span class="nn">torchvision.ops.feature_pyramid_network</span> <span class="kn">import</span> <span class="n">FeaturePyramidNetwork</span>
+
+
+<span class="c1"># To assist you in designing the feature extractor you may want to print out</span>
+<span class="c1"># the available nodes for resnet50.</span>
+<span class="n">m</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">()</span>
+<span class="n">train_nodes</span><span class="p">,</span> <span class="n">eval_nodes</span> <span class="o">=</span> <span class="n">get_graph_node_names</span><span class="p">(</span><span class="n">resnet50</span><span class="p">())</span>
+
+<span class="c1"># The lists returned, are the names of all the graph nodes (in order of</span>
+<span class="c1"># execution) for the input model traced in train mode and in eval mode</span>
+<span class="c1"># respectively. You&#39;ll find that `train_nodes` and `eval_nodes` are the same</span>
+<span class="c1"># for this example. But if the model contains control flow that&#39;s dependent</span>
+<span class="c1"># on the training mode, they may be different.</span>
+
+<span class="c1"># To specify the nodes you want to extract, you could select the final node</span>
+<span class="c1"># that appears in each of the main layers:</span>
+<span class="n">return_nodes</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="c1"># node_name: user-specified key for output dict</span>
+    <span class="s1">&#39;layer1.2.relu_2&#39;</span><span class="p">:</span> <span class="s1">&#39;layer1&#39;</span><span class="p">,</span>
+    <span class="s1">&#39;layer2.3.relu_2&#39;</span><span class="p">:</span> <span class="s1">&#39;layer2&#39;</span><span class="p">,</span>
+    <span class="s1">&#39;layer3.5.relu_2&#39;</span><span class="p">:</span> <span class="s1">&#39;layer3&#39;</span><span class="p">,</span>
+    <span class="s1">&#39;layer4.2.relu_2&#39;</span><span class="p">:</span> <span class="s1">&#39;layer4&#39;</span><span class="p">,</span>
+<span class="p">}</span>
+
+<span class="c1"># But `create_feature_extractor` can also accept truncated node specifications</span>
+<span class="c1"># like &quot;layer1&quot;, as it will just pick the last node that&#39;s a descendent of</span>
+<span class="c1"># of the specification. (Tip: be careful with this, especially when a layer</span>
+<span class="c1"># has multiple outputs. It&#39;s not always guaranteed that the last operation</span>
+<span class="c1"># performed is the one that corresponds to the output you desire. You should</span>
+<span class="c1"># consult the source code for the input model to confirm.)</span>
+<span class="n">return_nodes</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s1">&#39;layer1&#39;</span><span class="p">:</span> <span class="s1">&#39;layer1&#39;</span><span class="p">,</span>
+    <span class="s1">&#39;layer2&#39;</span><span class="p">:</span> <span class="s1">&#39;layer2&#39;</span><span class="p">,</span>
+    <span class="s1">&#39;layer3&#39;</span><span class="p">:</span> <span class="s1">&#39;layer3&#39;</span><span class="p">,</span>
+    <span class="s1">&#39;layer4&#39;</span><span class="p">:</span> <span class="s1">&#39;layer4&#39;</span><span class="p">,</span>
+<span class="p">}</span>
+
+<span class="c1"># Now you can build the feature extractor. This returns a module whose forward</span>
+<span class="c1"># method returns a dictionary like:</span>
+<span class="c1"># {</span>
+<span class="c1">#     &#39;layer1&#39;: output of layer 1,</span>
+<span class="c1">#     &#39;layer2&#39;: output of layer 2,</span>
+<span class="c1">#     &#39;layer3&#39;: output of layer 3,</span>
+<span class="c1">#     &#39;layer4&#39;: output of layer 4,</span>
+<span class="c1"># }</span>
+<span class="n">create_feature_extractor</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">return_nodes</span><span class="o">=</span><span class="n">return_nodes</span><span class="p">)</span>
+
+<span class="c1"># Let&#39;s put all that together to wrap resnet50 with MaskRCNN</span>
+
+<span class="c1"># MaskRCNN requires a backbone with an attached FPN</span>
+<span class="k">class</span> <span class="nc">Resnet50WithFPN</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Resnet50WithFPN</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="c1"># Get a resnet50 backbone</span>
+        <span class="n">m</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">()</span>
+        <span class="c1"># Extract 4 main layers (note: MaskRCNN needs this particular name</span>
+        <span class="c1"># mapping for return nodes)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">body</span> <span class="o">=</span> <span class="n">create_feature_extractor</span><span class="p">(</span>
+            <span class="n">m</span><span class="p">,</span> <span class="n">return_nodes</span><span class="o">=</span><span class="p">{</span><span class="sa">f</span><span class="s1">&#39;layer</span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">:</span> <span class="nb">str</span><span class="p">(</span><span class="n">v</span><span class="p">)</span>
+                             <span class="k">for</span> <span class="n">v</span><span class="p">,</span> <span class="n">k</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">])})</span>
+        <span class="c1"># Dry run to get number of channels for FPN</span>
+        <span class="n">inp</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randn</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">)</span>
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+            <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">body</span><span class="p">(</span><span class="n">inp</span><span class="p">)</span>
+        <span class="n">in_channels_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">o</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="k">for</span> <span class="n">o</span> <span class="ow">in</span> <span class="n">out</span><span class="o">.</span><span class="n">values</span><span class="p">()]</span>
+        <span class="c1"># Build FPN</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="mi">256</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fpn</span> <span class="o">=</span> <span class="n">FeaturePyramidNetwork</span><span class="p">(</span>
+            <span class="n">in_channels_list</span><span class="p">,</span> <span class="n">out_channels</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">,</span>
+            <span class="n">extra_blocks</span><span class="o">=</span><span class="n">LastLevelMaxPool</span><span class="p">())</span>
+
+    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">body</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fpn</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">x</span>
+
+
+<span class="c1"># Now we can build our model!</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">MaskRCNN</span><span class="p">(</span><span class="n">Resnet50WithFPN</span><span class="p">(),</span> <span class="n">num_classes</span><span class="o">=</span><span class="mi">91</span><span class="p">)</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+</pre></div>
+</div>
+<section id="api-reference">
+<h2>API Reference<a class="headerlink" href="#api-reference" title="Permalink to this heading">¶</a></h2>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.feature_extraction.create_feature_extractor.html#torchvision.models.feature_extraction.create_feature_extractor" title="torchvision.models.feature_extraction.create_feature_extractor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">create_feature_extractor</span></code></a>(model[, ...])</p></td>
+<td><p>Creates a new graph module that returns intermediate nodes from a given model as dictionary with user specified keys as strings, and the requested outputs as values.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.feature_extraction.get_graph_node_names.html#torchvision.models.feature_extraction.get_graph_node_names" title="torchvision.models.feature_extraction.get_graph_node_names"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_graph_node_names</span></code></a>(model[, tracer_kwargs, ...])</p></td>
+<td><p>Dev utility to return node names in order of execution.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.feature_extraction.create_feature_extractor.html" class="btn btn-neutral float-right" title="create_feature_extractor" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.io.VideoReader.html" class="btn btn-neutral" title="VideoReader" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Feature extraction for model inspection</a><ul>
+<li><a class="reference internal" href="#api-reference">API Reference</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/classification_quant_table.html b/0.16/generated/classification_quant_table.html
new file mode 100644
index 00000000000..e7ef8185ceb
--- /dev/null
+++ b/0.16/generated/classification_quant_table.html
@@ -0,0 +1,790 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>&lt;no title&gt;</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/classification_quant_table.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Acc&#64;1</strong></p></th>
+<th class="head"><p><strong>Acc&#64;5</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GIPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>69.826</p></td>
+<td><p>89.404</p></td>
+<td><p>6.6M</p></td>
+<td><p>1.5</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>77.176</p></td>
+<td><p>93.354</p></td>
+<td><p>27.2M</p></td>
+<td><p>5.71</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights.IMAGENET1K_QNNPACK_V1</span></code></p></td>
+<td><p>71.658</p></td>
+<td><p>90.15</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights.IMAGENET1K_QNNPACK_V1</span></code></p></td>
+<td><p>73.004</p></td>
+<td><p>90.858</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.22</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv3">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>78.986</p></td>
+<td><p>94.48</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V2</span></code></p></td>
+<td><p>82.574</p></td>
+<td><p>96.132</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>82.898</p></td>
+<td><p>96.326</p></td>
+<td><p>83.5M</p></td>
+<td><p>15.46</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5935">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>69.494</p></td>
+<td><p>88.882</p></td>
+<td><p>11.7M</p></td>
+<td><p>1.81</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>75.92</p></td>
+<td><p>92.814</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V2</span></code></p></td>
+<td><p>80.282</p></td>
+<td><p>94.976</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>57.972</p></td>
+<td><p>79.78</p></td>
+<td><p>1.4M</p></td>
+<td><p>0.04</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>68.36</p></td>
+<td><p>87.582</p></td>
+<td><p>2.3M</p></td>
+<td><p>0.14</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>72.052</p></td>
+<td><p>90.7</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></p></td>
+<td><p>75.354</p></td>
+<td><p>92.488</p></td>
+<td><p>7.4M</p></td>
+<td><p>0.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+</tbody>
+</table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul class="simple">
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/classification_table.html b/0.16/generated/classification_table.html
new file mode 100644
index 00000000000..b37d604106d
--- /dev/null
+++ b/0.16/generated/classification_table.html
@@ -0,0 +1,1497 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>&lt;no title&gt;</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/classification_table.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Acc&#64;1</strong></p></th>
+<th class="head"><p><strong>Acc&#64;5</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">AlexNet_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>56.522</p></td>
+<td><p>79.066</p></td>
+<td><p>61.1M</p></td>
+<td><p>0.71</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>84.062</p></td>
+<td><p>96.87</p></td>
+<td><p>88.6M</p></td>
+<td><p>15.36</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>84.414</p></td>
+<td><p>96.976</p></td>
+<td><p>197.8M</p></td>
+<td><p>34.36</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.616</p></td>
+<td><p>96.65</p></td>
+<td><p>50.2M</p></td>
+<td><p>8.68</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>82.52</p></td>
+<td><p>96.146</p></td>
+<td><p>28.6M</p></td>
+<td><p>4.46</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet121_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>74.434</p></td>
+<td><p>91.972</p></td>
+<td><p>8.0M</p></td>
+<td><p>2.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet161_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>77.138</p></td>
+<td><p>93.56</p></td>
+<td><p>28.7M</p></td>
+<td><p>7.73</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet169_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>75.6</p></td>
+<td><p>92.806</p></td>
+<td><p>14.1M</p></td>
+<td><p>3.36</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet201_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>76.896</p></td>
+<td><p>93.37</p></td>
+<td><p>20.0M</p></td>
+<td><p>4.29</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>77.692</p></td>
+<td><p>93.532</p></td>
+<td><p>5.3M</p></td>
+<td><p>0.39</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>78.642</p></td>
+<td><p>94.186</p></td>
+<td><p>7.8M</p></td>
+<td><p>0.69</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>79.838</p></td>
+<td><p>94.934</p></td>
+<td><p>7.8M</p></td>
+<td><p>0.69</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-lr-wd-crop-tuning">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>80.608</p></td>
+<td><p>95.31</p></td>
+<td><p>9.1M</p></td>
+<td><p>1.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>82.008</p></td>
+<td><p>96.054</p></td>
+<td><p>12.2M</p></td>
+<td><p>1.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.384</p></td>
+<td><p>96.594</p></td>
+<td><p>19.3M</p></td>
+<td><p>4.39</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.444</p></td>
+<td><p>96.628</p></td>
+<td><p>30.4M</p></td>
+<td><p>10.27</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>84.008</p></td>
+<td><p>96.916</p></td>
+<td><p>43.0M</p></td>
+<td><p>19.07</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>84.122</p></td>
+<td><p>96.908</p></td>
+<td><p>66.3M</p></td>
+<td><p>37.75</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>85.808</p></td>
+<td><p>97.788</p></td>
+<td><p>118.5M</p></td>
+<td><p>56.08</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>85.112</p></td>
+<td><p>97.156</p></td>
+<td><p>54.1M</p></td>
+<td><p>24.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>84.228</p></td>
+<td><p>96.878</p></td>
+<td><p>21.5M</p></td>
+<td><p>8.37</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>69.778</p></td>
+<td><p>89.53</p></td>
+<td><p>6.6M</p></td>
+<td><p>1.5</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#googlenet">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>77.294</p></td>
+<td><p>93.45</p></td>
+<td><p>27.2M</p></td>
+<td><p>5.71</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#inception-v3">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_5_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>67.734</p></td>
+<td><p>87.49</p></td>
+<td><p>2.2M</p></td>
+<td><p>0.1</p></td>
+<td><p><a class="reference external" href="https://github.com/1e100/mnasnet_trainer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_75_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>71.18</p></td>
+<td><p>90.496</p></td>
+<td><p>3.2M</p></td>
+<td><p>0.21</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/6019">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_0_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>73.456</p></td>
+<td><p>91.51</p></td>
+<td><p>4.4M</p></td>
+<td><p>0.31</p></td>
+<td><p><a class="reference external" href="https://github.com/1e100/mnasnet_trainer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_3_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>76.506</p></td>
+<td><p>93.522</p></td>
+<td><p>6.3M</p></td>
+<td><p>0.53</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/6019">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MaxVit_T_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.7</p></td>
+<td><p>96.722</p></td>
+<td><p>30.9M</p></td>
+<td><p>5.56</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#maxvit">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>71.878</p></td>
+<td><p>90.286</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>72.154</p></td>
+<td><p>90.822</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>74.042</p></td>
+<td><p>91.34</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.22</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>75.274</p></td>
+<td><p>92.566</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.22</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>67.668</p></td>
+<td><p>87.402</p></td>
+<td><p>2.5M</p></td>
+<td><p>0.06</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>80.058</p></td>
+<td><p>94.944</p></td>
+<td><p>54.3M</p></td>
+<td><p>15.94</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>82.716</p></td>
+<td><p>96.196</p></td>
+<td><p>54.3M</p></td>
+<td><p>15.94</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>77.04</p></td>
+<td><p>93.44</p></td>
+<td><p>9.2M</p></td>
+<td><p>1.6</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>79.668</p></td>
+<td><p>94.922</p></td>
+<td><p>9.2M</p></td>
+<td><p>1.6</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>80.622</p></td>
+<td><p>95.248</p></td>
+<td><p>107.8M</p></td>
+<td><p>31.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>83.014</p></td>
+<td><p>96.288</p></td>
+<td><p>107.8M</p></td>
+<td><p>31.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>78.364</p></td>
+<td><p>93.992</p></td>
+<td><p>15.3M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>81.196</p></td>
+<td><p>95.43</p></td>
+<td><p>15.3M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>72.834</p></td>
+<td><p>90.95</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>74.864</p></td>
+<td><p>92.322</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>75.212</p></td>
+<td><p>92.348</p></td>
+<td><p>7.3M</p></td>
+<td><p>0.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>77.522</p></td>
+<td><p>93.826</p></td>
+<td><p>7.3M</p></td>
+<td><p>0.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>79.344</p></td>
+<td><p>94.686</p></td>
+<td><p>39.6M</p></td>
+<td><p>8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>81.682</p></td>
+<td><p>95.678</p></td>
+<td><p>39.6M</p></td>
+<td><p>8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></p></td>
+<td><p>88.228</p></td>
+<td><p>98.682</p></td>
+<td><p>644.8M</p></td>
+<td><p>374.57</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></p></td>
+<td><p>86.068</p></td>
+<td><p>97.844</p></td>
+<td><p>644.8M</p></td>
+<td><p>127.52</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>80.424</p></td>
+<td><p>95.24</p></td>
+<td><p>83.6M</p></td>
+<td><p>15.91</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>82.886</p></td>
+<td><p>96.328</p></td>
+<td><p>83.6M</p></td>
+<td><p>15.91</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></p></td>
+<td><p>86.012</p></td>
+<td><p>98.054</p></td>
+<td><p>83.6M</p></td>
+<td><p>46.73</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></p></td>
+<td><p>83.976</p></td>
+<td><p>97.244</p></td>
+<td><p>83.6M</p></td>
+<td><p>15.91</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>77.95</p></td>
+<td><p>93.966</p></td>
+<td><p>11.2M</p></td>
+<td><p>1.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>80.876</p></td>
+<td><p>95.444</p></td>
+<td><p>11.2M</p></td>
+<td><p>1.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>80.878</p></td>
+<td><p>95.34</p></td>
+<td><p>145.0M</p></td>
+<td><p>32.28</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>83.368</p></td>
+<td><p>96.498</p></td>
+<td><p>145.0M</p></td>
+<td><p>32.28</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></p></td>
+<td><p>86.838</p></td>
+<td><p>98.362</p></td>
+<td><p>145.0M</p></td>
+<td><p>94.83</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></p></td>
+<td><p>84.622</p></td>
+<td><p>97.48</p></td>
+<td><p>145.0M</p></td>
+<td><p>32.28</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>78.948</p></td>
+<td><p>94.576</p></td>
+<td><p>19.4M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>81.982</p></td>
+<td><p>95.972</p></td>
+<td><p>19.4M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>74.046</p></td>
+<td><p>91.716</p></td>
+<td><p>4.3M</p></td>
+<td><p>0.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>75.804</p></td>
+<td><p>92.742</p></td>
+<td><p>4.3M</p></td>
+<td><p>0.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>76.42</p></td>
+<td><p>93.136</p></td>
+<td><p>6.4M</p></td>
+<td><p>0.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>78.828</p></td>
+<td><p>94.502</p></td>
+<td><p>6.4M</p></td>
+<td><p>0.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>80.032</p></td>
+<td><p>95.048</p></td>
+<td><p>39.4M</p></td>
+<td><p>8.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>82.828</p></td>
+<td><p>96.33</p></td>
+<td><p>39.4M</p></td>
+<td><p>8.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>79.312</p></td>
+<td><p>94.526</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>82.834</p></td>
+<td><p>96.228</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.246</p></td>
+<td><p>96.454</p></td>
+<td><p>83.5M</p></td>
+<td><p>15.46</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5935">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>77.618</p></td>
+<td><p>93.698</p></td>
+<td><p>25.0M</p></td>
+<td><p>4.23</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>81.198</p></td>
+<td><p>95.34</p></td>
+<td><p>25.0M</p></td>
+<td><p>4.23</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>77.374</p></td>
+<td><p>93.546</p></td>
+<td><p>44.5M</p></td>
+<td><p>7.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>81.886</p></td>
+<td><p>95.78</p></td>
+<td><p>44.5M</p></td>
+<td><p>7.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet152_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>78.312</p></td>
+<td><p>94.046</p></td>
+<td><p>60.2M</p></td>
+<td><p>11.51</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet152_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>82.284</p></td>
+<td><p>96.002</p></td>
+<td><p>60.2M</p></td>
+<td><p>11.51</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>69.758</p></td>
+<td><p>89.078</p></td>
+<td><p>11.7M</p></td>
+<td><p>1.81</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet34_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>73.314</p></td>
+<td><p>91.42</p></td>
+<td><p>21.8M</p></td>
+<td><p>3.66</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>76.13</p></td>
+<td><p>92.862</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>80.858</p></td>
+<td><p>95.434</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#issuecomment-1013906621">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>60.552</p></td>
+<td><p>81.746</p></td>
+<td><p>1.4M</p></td>
+<td><p>0.04</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>69.362</p></td>
+<td><p>88.316</p></td>
+<td><p>2.3M</p></td>
+<td><p>0.14</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>72.996</p></td>
+<td><p>91.086</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>76.23</p></td>
+<td><p>93.006</p></td>
+<td><p>7.4M</p></td>
+<td><p>0.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>58.092</p></td>
+<td><p>80.42</p></td>
+<td><p>1.2M</p></td>
+<td><p>0.82</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/49#issuecomment-277560717">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>58.178</p></td>
+<td><p>80.624</p></td>
+<td><p>1.2M</p></td>
+<td><p>0.35</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/49#issuecomment-277560717">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_B_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.582</p></td>
+<td><p>96.64</p></td>
+<td><p>87.8M</p></td>
+<td><p>15.43</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_S_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.196</p></td>
+<td><p>96.36</p></td>
+<td><p>49.6M</p></td>
+<td><p>8.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_T_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>81.474</p></td>
+<td><p>95.776</p></td>
+<td><p>28.3M</p></td>
+<td><p>4.49</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_B_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>84.112</p></td>
+<td><p>96.864</p></td>
+<td><p>87.9M</p></td>
+<td><p>20.32</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_S_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>83.712</p></td>
+<td><p>96.816</p></td>
+<td><p>49.7M</p></td>
+<td><p>11.55</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_T_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>82.072</p></td>
+<td><p>96.132</p></td>
+<td><p>28.4M</p></td>
+<td><p>5.94</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_BN_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>70.37</p></td>
+<td><p>89.81</p></td>
+<td><p>132.9M</p></td>
+<td><p>7.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>69.02</p></td>
+<td><p>88.628</p></td>
+<td><p>132.9M</p></td>
+<td><p>7.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_BN_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>71.586</p></td>
+<td><p>90.374</p></td>
+<td><p>133.1M</p></td>
+<td><p>11.31</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>69.928</p></td>
+<td><p>89.246</p></td>
+<td><p>133.0M</p></td>
+<td><p>11.31</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_BN_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>73.36</p></td>
+<td><p>91.516</p></td>
+<td><p>138.4M</p></td>
+<td><p>15.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>71.592</p></td>
+<td><p>90.382</p></td>
+<td><p>138.4M</p></td>
+<td><p>15.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_Weights.IMAGENET1K_FEATURES</span></code></p></td>
+<td><p>nan</p></td>
+<td><p>nan</p></td>
+<td><p>138.4M</p></td>
+<td><p>15.47</p></td>
+<td><p><a class="reference external" href="https://github.com/amdegroot/ssd.pytorch#training-ssd">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_BN_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>74.218</p></td>
+<td><p>91.842</p></td>
+<td><p>143.7M</p></td>
+<td><p>19.63</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>72.376</p></td>
+<td><p>90.876</p></td>
+<td><p>143.7M</p></td>
+<td><p>19.63</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>81.072</p></td>
+<td><p>95.318</p></td>
+<td><p>86.6M</p></td>
+<td><p>17.56</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_b_16">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></p></td>
+<td><p>85.304</p></td>
+<td><p>97.65</p></td>
+<td><p>86.9M</p></td>
+<td><p>55.48</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></p></td>
+<td><p>81.886</p></td>
+<td><p>96.18</p></td>
+<td><p>86.6M</p></td>
+<td><p>17.56</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_32_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>75.912</p></td>
+<td><p>92.466</p></td>
+<td><p>88.2M</p></td>
+<td><p>4.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_b_32">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_H_14_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></p></td>
+<td><p>88.552</p></td>
+<td><p>98.694</p></td>
+<td><p>633.5M</p></td>
+<td><p>1016.72</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_H_14_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></p></td>
+<td><p>85.708</p></td>
+<td><p>97.73</p></td>
+<td><p>632.0M</p></td>
+<td><p>167.29</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>79.662</p></td>
+<td><p>94.638</p></td>
+<td><p>304.3M</p></td>
+<td><p>61.55</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_l_16">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></p></td>
+<td><p>88.064</p></td>
+<td><p>98.512</p></td>
+<td><p>305.2M</p></td>
+<td><p>361.99</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></p></td>
+<td><p>85.146</p></td>
+<td><p>97.422</p></td>
+<td><p>304.3M</p></td>
+<td><p>61.55</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_32_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>76.972</p></td>
+<td><p>93.07</p></td>
+<td><p>306.5M</p></td>
+<td><p>15.38</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_l_32">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>78.848</p></td>
+<td><p>94.284</p></td>
+<td><p>126.9M</p></td>
+<td><p>22.75</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/912#issue-445437439">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>82.51</p></td>
+<td><p>96.02</p></td>
+<td><p>126.9M</p></td>
+<td><p>22.75</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.IMAGENET1K_V1</span></code></p></td>
+<td><p>78.468</p></td>
+<td><p>94.086</p></td>
+<td><p>68.9M</p></td>
+<td><p>11.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/912#issue-445437439">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.IMAGENET1K_V2</span></code></p></td>
+<td><p>81.602</p></td>
+<td><p>95.758</p></td>
+<td><p>68.9M</p></td>
+<td><p>11.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+</tbody>
+</table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul class="simple">
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/detection_keypoint_table.html b/0.16/generated/detection_keypoint_table.html
new file mode 100644
index 00000000000..2e4bd1b4bac
--- /dev/null
+++ b/0.16/generated/detection_keypoint_table.html
@@ -0,0 +1,706 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>&lt;no title&gt;</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/detection_keypoint_table.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Box MAP</strong></p></th>
+<th class="head"><p><strong>Keypoint MAP</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.COCO_LEGACY</span></code></p></td>
+<td><p>50.6</p></td>
+<td><p>61.1</p></td>
+<td><p>59.1M</p></td>
+<td><p>133.92</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/1606">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.COCO_V1</span></code></p></td>
+<td><p>54.6</p></td>
+<td><p>65</p></td>
+<td><p>59.1M</p></td>
+<td><p>137.42</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#keypoint-r-cnn">link</a></p></td>
+</tr>
+</tbody>
+</table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul class="simple">
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/detection_table.html b/0.16/generated/detection_table.html
new file mode 100644
index 00000000000..a02f48d55f6
--- /dev/null
+++ b/0.16/generated/detection_table.html
@@ -0,0 +1,744 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>&lt;no title&gt;</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/detection_table.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 63%" />
+<col style="width: 10%" />
+<col style="width: 10%" />
+<col style="width: 10%" />
+<col style="width: 6%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Box MAP</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights.COCO_V1</span></code></p></td>
+<td><p>39.2</p></td>
+<td><p>32.3M</p></td>
+<td><p>128.21</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#fcos-resnet-50-fpn">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.COCO_V1</span></code></p></td>
+<td><p>22.8</p></td>
+<td><p>19.4M</p></td>
+<td><p>0.72</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-320-fpn">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights.COCO_V1</span></code></p></td>
+<td><p>32.8</p></td>
+<td><p>19.4M</p></td>
+<td><p>4.49</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-fpn">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights.COCO_V1</span></code></p></td>
+<td><p>46.7</p></td>
+<td><p>43.7M</p></td>
+<td><p>280.37</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5763">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights.COCO_V1</span></code></p></td>
+<td><p>37</p></td>
+<td><p>41.8M</p></td>
+<td><p>134.38</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-resnet-50-fpn">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights.COCO_V1</span></code></p></td>
+<td><p>41.5</p></td>
+<td><p>38.2M</p></td>
+<td><p>152.24</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5756">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights.COCO_V1</span></code></p></td>
+<td><p>36.4</p></td>
+<td><p>34.0M</p></td>
+<td><p>151.54</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#retinanet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights.COCO_V1</span></code></p></td>
+<td><p>25.1</p></td>
+<td><p>35.6M</p></td>
+<td><p>34.86</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#ssd300-vgg16">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights.COCO_V1</span></code></p></td>
+<td><p>21.3</p></td>
+<td><p>3.4M</p></td>
+<td><p>0.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#ssdlite320-mobilenetv3-large">link</a></p></td>
+</tr>
+</tbody>
+</table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul class="simple">
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/instance_segmentation_table.html b/0.16/generated/instance_segmentation_table.html
new file mode 100644
index 00000000000..dcba209ec91
--- /dev/null
+++ b/0.16/generated/instance_segmentation_table.html
@@ -0,0 +1,706 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>&lt;no title&gt;</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/instance_segmentation_table.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Box MAP</strong></p></th>
+<th class="head"><p><strong>Mask MAP</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights.COCO_V1</span></code></p></td>
+<td><p>47.4</p></td>
+<td><p>41.8</p></td>
+<td><p>46.4M</p></td>
+<td><p>333.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5773">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights.COCO_V1</span></code></p></td>
+<td><p>37.9</p></td>
+<td><p>34.6</p></td>
+<td><p>44.4M</p></td>
+<td><p>134.38</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#mask-r-cnn">link</a></p></td>
+</tr>
+</tbody>
+</table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul class="simple">
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/segmentation_table.html b/0.16/generated/segmentation_table.html
new file mode 100644
index 00000000000..3fc2327d25d
--- /dev/null
+++ b/0.16/generated/segmentation_table.html
@@ -0,0 +1,734 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>&lt;no title&gt;</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/segmentation_table.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Mean IoU</strong></p></th>
+<th class="head"><p><strong>pixelwise Acc</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</span></code></p></td>
+<td><p>60.3</p></td>
+<td><p>91.2</p></td>
+<td><p>11.0M</p></td>
+<td><p>10.45</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_mobilenet_v3_large">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</span></code></p></td>
+<td><p>67.4</p></td>
+<td><p>92.4</p></td>
+<td><p>61.0M</p></td>
+<td><p>258.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet101">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</span></code></p></td>
+<td><p>66.4</p></td>
+<td><p>92.4</p></td>
+<td><p>42.0M</p></td>
+<td><p>178.72</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet50">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</span></code></p></td>
+<td><p>63.7</p></td>
+<td><p>91.9</p></td>
+<td><p>54.3M</p></td>
+<td><p>232.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet101">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</span></code></p></td>
+<td><p>60.5</p></td>
+<td><p>91.4</p></td>
+<td><p>35.3M</p></td>
+<td><p>152.72</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet50">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</span></code></p></td>
+<td><p>57.9</p></td>
+<td><p>91.2</p></td>
+<td><p>3.2M</p></td>
+<td><p>2.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#lraspp_mobilenet_v3_large">link</a></p></td>
+</tr>
+</tbody>
+</table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul class="simple">
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.CIFAR10.html b/0.16/generated/torchvision.datasets.CIFAR10.html
new file mode 100644
index 00000000000..6ac4a17e911
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.CIFAR10.html
@@ -0,0 +1,731 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CIFAR10 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CIFAR100" href="torchvision.datasets.CIFAR100.html" />
+    <link rel="prev" title="CelebA" href="torchvision.datasets.CelebA.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>CIFAR10</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.CIFAR10.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="cifar10">
+<h1>CIFAR10<a class="headerlink" href="#cifar10" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.CIFAR10">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">CIFAR10</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/cifar.html#CIFAR10"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CIFAR10" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.cs.toronto.edu/~kriz/cifar.html">CIFAR10</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">cifar-10-batches-py</span></code> exists or will be saved to if download is set to True.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, creates dataset from training set, otherwise
+creates from test set.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.CIFAR10.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/cifar.html#CIFAR10.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CIFAR10.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.CIFAR100.html" class="btn btn-neutral float-right" title="CIFAR100" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.CelebA.html" class="btn btn-neutral" title="CelebA" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CIFAR10</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.CIFAR100.html b/0.16/generated/torchvision.datasets.CIFAR100.html
new file mode 100644
index 00000000000..73b3f353699
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.CIFAR100.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CIFAR100 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Country211" href="torchvision.datasets.Country211.html" />
+    <link rel="prev" title="CIFAR10" href="torchvision.datasets.CIFAR10.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>CIFAR100</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.CIFAR100.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="cifar100">
+<h1>CIFAR100<a class="headerlink" href="#cifar100" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.CIFAR100">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">CIFAR100</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/cifar.html#CIFAR100"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CIFAR100" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.cs.toronto.edu/~kriz/cifar.html">CIFAR100</a> Dataset.</p>
+<p>This is a subclass of the <cite>CIFAR10</cite> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.CIFAR100.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.CIFAR100.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Country211.html" class="btn btn-neutral float-right" title="Country211" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.CIFAR10.html" class="btn btn-neutral" title="CIFAR10" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CIFAR100</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.CREStereo.html b/0.16/generated/torchvision.datasets.CREStereo.html
new file mode 100644
index 00000000000..fa97ff81b12
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.CREStereo.html
@@ -0,0 +1,760 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CREStereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FallingThingsStereo" href="torchvision.datasets.FallingThingsStereo.html" />
+    <link rel="prev" title="Kitti2015Stereo" href="torchvision.datasets.Kitti2015Stereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>CREStereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.CREStereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="crestereo">
+<h1>CREStereo<a class="headerlink" href="#crestereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.CREStereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">CREStereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#CREStereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CREStereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>Synthetic dataset used in training the <a class="reference external" href="https://arxiv.org/pdf/2203.11483.pdf">CREStereo</a> architecture.
+Dataset details on the official paper <a class="reference external" href="https://github.com/megvii-research/CREStereo">repo</a>.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">CREStereo</span>
+        <span class="n">tree</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img2_left</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img2_right</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img2_left</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img2_right</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="o">...</span>
+        <span class="n">shapenet</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="o">...</span>
+        <span class="n">reflective</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="o">...</span>
+        <span class="n">hole</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_left</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="n">img1_right</span><span class="o">.</span><span class="n">disp</span><span class="o">.</span><span class="n">jpg</span>
+            <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Root directory of the dataset.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.CREStereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#CREStereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CREStereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is implicitly <code class="docutils literal notranslate"><span class="pre">None</span></code> if the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter does not
+generate a valid mask.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.FallingThingsStereo.html" class="btn btn-neutral float-right" title="FallingThingsStereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Kitti2015Stereo.html" class="btn btn-neutral" title="Kitti2015Stereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CREStereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Caltech101.html b/0.16/generated/torchvision.datasets.Caltech101.html
new file mode 100644
index 00000000000..b3d5b5fcdf6
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Caltech101.html
@@ -0,0 +1,738 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Caltech101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Caltech256" href="torchvision.datasets.Caltech256.html" />
+    <link rel="prev" title="Datasets" href="../datasets.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Caltech101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Caltech101.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="caltech101">
+<h1>Caltech101<a class="headerlink" href="#caltech101" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Caltech101">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Caltech101</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'category'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/caltech.html#Caltech101"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Caltech101" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://data.caltech.edu/records/20086">Caltech 101</a> Dataset.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>This class needs <a class="reference external" href="https://docs.scipy.org/doc/">scipy</a> to load target files from <cite>.mat</cite> format.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">caltech101</span></code> exists or will be saved to if download is set to True.</p></li>
+<li><p><strong>target_type</strong> (<em>string</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em>, </em><em>optional</em>) – Type of target to use, <code class="docutils literal notranslate"><span class="pre">category</span></code> or
+<code class="docutils literal notranslate"><span class="pre">annotation</span></code>. Can also be a list to output a tuple with all specified
+target types.  <code class="docutils literal notranslate"><span class="pre">category</span></code> represents the target class, and
+<code class="docutils literal notranslate"><span class="pre">annotation</span></code> is a list of points from a hand-generated outline.
+Defaults to <code class="docutils literal notranslate"><span class="pre">category</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Caltech101.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/caltech.html#Caltech101.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Caltech101.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where the type of target specified by target_type.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Caltech256.html" class="btn btn-neutral float-right" title="Caltech256" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../datasets.html" class="btn btn-neutral" title="Datasets" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Caltech101</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Caltech256.html b/0.16/generated/torchvision.datasets.Caltech256.html
new file mode 100644
index 00000000000..8169c77606d
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Caltech256.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Caltech256 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CelebA" href="torchvision.datasets.CelebA.html" />
+    <link rel="prev" title="Caltech101" href="torchvision.datasets.Caltech101.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Caltech256</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Caltech256.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="caltech256">
+<h1>Caltech256<a class="headerlink" href="#caltech256" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Caltech256">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Caltech256</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/caltech.html#Caltech256"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Caltech256" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://data.caltech.edu/records/20087">Caltech 256</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">caltech256</span></code> exists or will be saved to if download is set to True.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Caltech256.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/caltech.html#Caltech256.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Caltech256.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.CelebA.html" class="btn btn-neutral float-right" title="CelebA" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Caltech101.html" class="btn btn-neutral" title="Caltech101" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Caltech256</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.CarlaStereo.html b/0.16/generated/torchvision.datasets.CarlaStereo.html
new file mode 100644
index 00000000000..b3b01e604c9
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.CarlaStereo.html
@@ -0,0 +1,745 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CarlaStereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Kitti2012Stereo" href="torchvision.datasets.Kitti2012Stereo.html" />
+    <link rel="prev" title="Sintel" href="torchvision.datasets.Sintel.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>CarlaStereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.CarlaStereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="carlastereo">
+<h1>CarlaStereo<a class="headerlink" href="#carlastereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.CarlaStereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">CarlaStereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#CarlaStereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CarlaStereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>Carla simulator data linked in the <a class="reference external" href="https://github.com/megvii-research/CREStereo">CREStereo github repo</a>.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">carla</span><span class="o">-</span><span class="n">highres</span>
+        <span class="n">trainingF</span>
+            <span class="n">scene1</span>
+                <span class="n">img0</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp0GT</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp1GT</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+            <span class="n">scene2</span>
+                <span class="n">img0</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp0GT</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp1GT</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+            <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where <cite>carla-highres</cite> is located.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.CarlaStereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#CarlaStereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CarlaStereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 3-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+If a <code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is generated within the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter,
+a 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code> is returned.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Kitti2012Stereo.html" class="btn btn-neutral float-right" title="Kitti2012Stereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Sintel.html" class="btn btn-neutral" title="Sintel" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CarlaStereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.CelebA.html b/0.16/generated/torchvision.datasets.CelebA.html
new file mode 100644
index 00000000000..60bf1f1492b
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.CelebA.html
@@ -0,0 +1,744 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CelebA &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CIFAR10" href="torchvision.datasets.CIFAR10.html" />
+    <link rel="prev" title="Caltech256" href="torchvision.datasets.Caltech256.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>CelebA</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.CelebA.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="celeba">
+<h1>CelebA<a class="headerlink" href="#celeba" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.CelebA">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">CelebA</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'attr'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/celeba.html#CelebA"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CelebA" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html">Large-scale CelebFaces Attributes (CelebA) Dataset</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where images are downloaded to.</p></li>
+<li><p><strong>split</strong> (<em>string</em>) – One of {‘train’, ‘valid’, ‘test’, ‘all’}.
+Accordingly dataset is selected.</p></li>
+<li><p><strong>target_type</strong> (<em>string</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em>, </em><em>optional</em>) – <p>Type of target to use, <code class="docutils literal notranslate"><span class="pre">attr</span></code>, <code class="docutils literal notranslate"><span class="pre">identity</span></code>, <code class="docutils literal notranslate"><span class="pre">bbox</span></code>,
+or <code class="docutils literal notranslate"><span class="pre">landmarks</span></code>. Can also be a list to output a tuple with all specified target types.
+The targets represent:</p>
+<blockquote>
+<div><ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">attr</span></code> (Tensor shape=(40,) dtype=int): binary (0, 1) labels for attributes</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">identity</span></code> (int): label for each person (data points with the same identity are the same person)</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">bbox</span></code> (Tensor shape=(4,) dtype=int): bounding box (x, y, width, height)</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">landmarks</span></code> (Tensor shape=(10,) dtype=int): landmark points (lefteye_x, lefteye_y, righteye_x,
+righteye_y, nose_x, nose_y, leftmouth_x, leftmouth_y, rightmouth_x, rightmouth_y)</p></li>
+</ul>
+</div></blockquote>
+<p>Defaults to <code class="docutils literal notranslate"><span class="pre">attr</span></code>. If empty, <code class="docutils literal notranslate"><span class="pre">None</span></code> will be returned as target.</p>
+</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.PILToTensor</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.CelebA.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/celeba.html#CelebA.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CelebA.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.CIFAR10.html" class="btn btn-neutral float-right" title="CIFAR10" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Caltech256.html" class="btn btn-neutral" title="Caltech256" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CelebA</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Cityscapes.html b/0.16/generated/torchvision.datasets.Cityscapes.html
new file mode 100644
index 00000000000..939bdb2a1b4
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Cityscapes.html
@@ -0,0 +1,756 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Cityscapes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Kitti" href="torchvision.datasets.Kitti.html" />
+    <link rel="prev" title="CocoDetection" href="torchvision.datasets.CocoDetection.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Cityscapes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Cityscapes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="cityscapes">
+<h1>Cityscapes<a class="headerlink" href="#cityscapes" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Cityscapes">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Cityscapes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'fine'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'instance'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/cityscapes.html#Cityscapes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Cityscapes" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://www.cityscapes-dataset.com/">Cityscapes</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory <code class="docutils literal notranslate"><span class="pre">leftImg8bit</span></code>
+and <code class="docutils literal notranslate"><span class="pre">gtFine</span></code> or <code class="docutils literal notranslate"><span class="pre">gtCoarse</span></code> are located.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The image split to use, <code class="docutils literal notranslate"><span class="pre">train</span></code>, <code class="docutils literal notranslate"><span class="pre">test</span></code> or <code class="docutils literal notranslate"><span class="pre">val</span></code> if mode=”fine”
+otherwise <code class="docutils literal notranslate"><span class="pre">train</span></code>, <code class="docutils literal notranslate"><span class="pre">train_extra</span></code> or <code class="docutils literal notranslate"><span class="pre">val</span></code></p></li>
+<li><p><strong>mode</strong> (<em>string</em><em>, </em><em>optional</em>) – The quality mode to use, <code class="docutils literal notranslate"><span class="pre">fine</span></code> or <code class="docutils literal notranslate"><span class="pre">coarse</span></code></p></li>
+<li><p><strong>target_type</strong> (<em>string</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em>, </em><em>optional</em>) – Type of target to use, <code class="docutils literal notranslate"><span class="pre">instance</span></code>, <code class="docutils literal notranslate"><span class="pre">semantic</span></code>, <code class="docutils literal notranslate"><span class="pre">polygon</span></code>
+or <code class="docutils literal notranslate"><span class="pre">color</span></code>. Can also be a list to output a tuple with all specified target types.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes input sample and its target as entry
+and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<p class="rubric">Examples</p>
+<p>Get semantic segmentation target</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">dataset</span> <span class="o">=</span> <span class="n">Cityscapes</span><span class="p">(</span><span class="s1">&#39;./data/cityscapes&#39;</span><span class="p">,</span> <span class="n">split</span><span class="o">=</span><span class="s1">&#39;train&#39;</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s1">&#39;fine&#39;</span><span class="p">,</span>
+                     <span class="n">target_type</span><span class="o">=</span><span class="s1">&#39;semantic&#39;</span><span class="p">)</span>
+
+<span class="n">img</span><span class="p">,</span> <span class="n">smnt</span> <span class="o">=</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+</pre></div>
+</div>
+<p>Get multiple targets</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">dataset</span> <span class="o">=</span> <span class="n">Cityscapes</span><span class="p">(</span><span class="s1">&#39;./data/cityscapes&#39;</span><span class="p">,</span> <span class="n">split</span><span class="o">=</span><span class="s1">&#39;train&#39;</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s1">&#39;fine&#39;</span><span class="p">,</span>
+                     <span class="n">target_type</span><span class="o">=</span><span class="p">[</span><span class="s1">&#39;instance&#39;</span><span class="p">,</span> <span class="s1">&#39;color&#39;</span><span class="p">,</span> <span class="s1">&#39;polygon&#39;</span><span class="p">])</span>
+
+<span class="n">img</span><span class="p">,</span> <span class="p">(</span><span class="n">inst</span><span class="p">,</span> <span class="n">col</span><span class="p">,</span> <span class="n">poly</span><span class="p">)</span> <span class="o">=</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+</pre></div>
+</div>
+<p>Validate on the “coarse” set</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">dataset</span> <span class="o">=</span> <span class="n">Cityscapes</span><span class="p">(</span><span class="s1">&#39;./data/cityscapes&#39;</span><span class="p">,</span> <span class="n">split</span><span class="o">=</span><span class="s1">&#39;val&#39;</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s1">&#39;coarse&#39;</span><span class="p">,</span>
+                     <span class="n">target_type</span><span class="o">=</span><span class="s1">&#39;semantic&#39;</span><span class="p">)</span>
+
+<span class="n">img</span><span class="p">,</span> <span class="n">smnt</span> <span class="o">=</span> <span class="n">dataset</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Cityscapes.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/cityscapes.html#Cityscapes.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Cityscapes.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is a tuple of all target types if target_type is a list with more
+than one item. Otherwise, target is a json object if target_type=”polygon”, else the image segmentation.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Kitti.html" class="btn btn-neutral float-right" title="Kitti" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.CocoDetection.html" class="btn btn-neutral" title="CocoDetection" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Cityscapes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.CocoCaptions.html b/0.16/generated/torchvision.datasets.CocoCaptions.html
new file mode 100644
index 00000000000..4de2ad4753a
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.CocoCaptions.html
@@ -0,0 +1,753 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CocoCaptions &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="HMDB51" href="torchvision.datasets.HMDB51.html" />
+    <link rel="prev" title="PhotoTour" href="torchvision.datasets.PhotoTour.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>CocoCaptions</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.CocoCaptions.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="cococaptions">
+<h1>CocoCaptions<a class="headerlink" href="#cococaptions" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.CocoCaptions">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">CocoCaptions</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">annFile</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/coco.html#CocoCaptions"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CocoCaptions" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://cocodataset.org/#captions-2015">MS Coco Captions</a> Dataset.</p>
+<p>It requires the <a class="reference external" href="https://github.com/pdollar/coco/tree/master/PythonAPI">COCO API to be installed</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where images are downloaded to.</p></li>
+<li><p><strong>annFile</strong> (<em>string</em>) – Path to json annotation file.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.PILToTensor</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes input sample and its target as entry
+and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torchvision.datasets</span> <span class="k">as</span> <span class="nn">dset</span>
+<span class="kn">import</span> <span class="nn">torchvision.transforms</span> <span class="k">as</span> <span class="nn">transforms</span>
+<span class="n">cap</span> <span class="o">=</span> <span class="n">dset</span><span class="o">.</span><span class="n">CocoCaptions</span><span class="p">(</span><span class="n">root</span> <span class="o">=</span> <span class="s1">&#39;dir where images are&#39;</span><span class="p">,</span>
+                        <span class="n">annFile</span> <span class="o">=</span> <span class="s1">&#39;json annotation file&#39;</span><span class="p">,</span>
+                        <span class="n">transform</span><span class="o">=</span><span class="n">transforms</span><span class="o">.</span><span class="n">PILToTensor</span><span class="p">())</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="s1">&#39;Number of samples: &#39;</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">cap</span><span class="p">))</span>
+<span class="n">img</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">cap</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span> <span class="c1"># load 4th sample</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Image Size: &quot;</span><span class="p">,</span> <span class="n">img</span><span class="o">.</span><span class="n">size</span><span class="p">())</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">target</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Output:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">Number</span> <span class="n">of</span> <span class="n">samples</span><span class="p">:</span> <span class="mi">82783</span>
+<span class="n">Image</span> <span class="n">Size</span><span class="p">:</span> <span class="p">(</span><span class="mi">3</span><span class="n">L</span><span class="p">,</span> <span class="mi">427</span><span class="n">L</span><span class="p">,</span> <span class="mi">640</span><span class="n">L</span><span class="p">)</span>
+<span class="p">[</span><span class="sa">u</span><span class="s1">&#39;A plane emitting smoke stream flying over a mountain.&#39;</span><span class="p">,</span>
+<span class="sa">u</span><span class="s1">&#39;A plane darts across a bright blue sky behind a mountain covered in snow&#39;</span><span class="p">,</span>
+<span class="sa">u</span><span class="s1">&#39;A plane leaves a contrail above the snowy mountain top.&#39;</span><span class="p">,</span>
+<span class="sa">u</span><span class="s1">&#39;A mountain that has a plane flying overheard in the distance.&#39;</span><span class="p">,</span>
+<span class="sa">u</span><span class="s1">&#39;A mountain view with a plume of smoke in the background&#39;</span><span class="p">]</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.CocoCaptions.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.CocoCaptions.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.HMDB51.html" class="btn btn-neutral float-right" title="HMDB51" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.PhotoTour.html" class="btn btn-neutral" title="PhotoTour" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CocoCaptions</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.CocoDetection.html b/0.16/generated/torchvision.datasets.CocoDetection.html
new file mode 100644
index 00000000000..a256636df2c
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.CocoDetection.html
@@ -0,0 +1,736 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CocoDetection &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Cityscapes" href="torchvision.datasets.Cityscapes.html" />
+    <link rel="prev" title="USPS" href="torchvision.datasets.USPS.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>CocoDetection</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.CocoDetection.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="cocodetection">
+<h1>CocoDetection<a class="headerlink" href="#cocodetection" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.CocoDetection">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">CocoDetection</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">annFile</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/coco.html#CocoDetection"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CocoDetection" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://cocodataset.org/#detection-2016">MS Coco Detection</a> Dataset.</p>
+<p>It requires the <a class="reference external" href="https://github.com/pdollar/coco/tree/master/PythonAPI">COCO API to be installed</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where images are downloaded to.</p></li>
+<li><p><strong>annFile</strong> (<em>string</em>) – Path to json annotation file.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.PILToTensor</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes input sample and its target as entry
+and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">CocoDetection</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div><dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.CocoDetection.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/coco.html#CocoDetection.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.CocoDetection.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Cityscapes.html" class="btn btn-neutral float-right" title="Cityscapes" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.USPS.html" class="btn btn-neutral" title="USPS" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CocoDetection</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Country211.html b/0.16/generated/torchvision.datasets.Country211.html
new file mode 100644
index 00000000000..ac47a052194
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Country211.html
@@ -0,0 +1,731 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Country211 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="DTD" href="torchvision.datasets.DTD.html" />
+    <link rel="prev" title="CIFAR100" href="torchvision.datasets.CIFAR100.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Country211</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Country211.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="country211">
+<h1>Country211<a class="headerlink" href="#country211" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Country211">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Country211</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/country211.html#Country211"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Country211" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/openai/CLIP/blob/main/data/country211.md">The Country211 Data Set</a> from OpenAI.</p>
+<p>This dataset was built by filtering the images from the YFCC100m dataset
+that have GPS coordinate corresponding to a ISO-3166 country code. The
+dataset is balanced by sampling 150 train images, 50 validation images, and
+100 test images for each country.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default), <code class="docutils literal notranslate"><span class="pre">&quot;valid&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and puts it into
+<code class="docutils literal notranslate"><span class="pre">root/country211/</span></code>. If dataset is already downloaded, it is not downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Country211.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.Country211.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(sample, target) where target is class_index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.DTD.html" class="btn btn-neutral float-right" title="DTD" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.CIFAR100.html" class="btn btn-neutral" title="CIFAR100" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Country211</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.DTD.html b/0.16/generated/torchvision.datasets.DTD.html
new file mode 100644
index 00000000000..4a912a1011f
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.DTD.html
@@ -0,0 +1,735 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>DTD &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="EMNIST" href="torchvision.datasets.EMNIST.html" />
+    <link rel="prev" title="Country211" href="torchvision.datasets.Country211.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>DTD</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.DTD.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="dtd">
+<h1>DTD<a class="headerlink" href="#dtd" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.DTD">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">DTD</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">partition</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/dtd.html#DTD"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.DTD" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/dtd/">Describable Textures Dataset (DTD)</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default), <code class="docutils literal notranslate"><span class="pre">&quot;val&quot;</span></code>, or <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>partition</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – <p>The dataset partition. Should be <code class="docutils literal notranslate"><span class="pre">1</span> <span class="pre">&lt;=</span> <span class="pre">partition</span> <span class="pre">&lt;=</span> <span class="pre">10</span></code>. Defaults to <code class="docutils literal notranslate"><span class="pre">1</span></code>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The partition only changes which split each image belongs to. Thus, regardless of the selected
+partition, combining all splits will result in all images.</p>
+</div>
+</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in a PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again. Default is False.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.DTD.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/dtd.html#DTD.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.DTD.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.EMNIST.html" class="btn btn-neutral float-right" title="EMNIST" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Country211.html" class="btn btn-neutral" title="Country211" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">DTD</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.DatasetFolder.html b/0.16/generated/torchvision.datasets.DatasetFolder.html
new file mode 100644
index 00000000000..335e9935651
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.DatasetFolder.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>DatasetFolder &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ImageFolder" href="torchvision.datasets.ImageFolder.html" />
+    <link rel="prev" title="MovingMNIST" href="torchvision.datasets.MovingMNIST.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>DatasetFolder</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.DatasetFolder.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="datasetfolder">
+<h1>DatasetFolder<a class="headerlink" href="#datasetfolder" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.DatasetFolder">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">DatasetFolder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">loader</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">extensions</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">is_valid_file</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/folder.html#DatasetFolder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.DatasetFolder" title="Permalink to this definition">¶</a></dt>
+<dd><p>A generic data loader.</p>
+<p>This default directory structure can be customized by overriding the
+<a class="reference internal" href="#torchvision.datasets.DatasetFolder.find_classes" title="torchvision.datasets.DatasetFolder.find_classes"><code class="xref py py-meth docutils literal notranslate"><span class="pre">find_classes()</span></code></a> method.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory path.</p></li>
+<li><p><strong>loader</strong> (<em>callable</em>) – A function to load a sample given its path.</p></li>
+<li><p><strong>extensions</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em>[</em><em>string</em><em>]</em>) – A list of allowed extensions.
+both extensions and is_valid_file should not be passed.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in
+a sample and returns a transformed version.
+E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code> for images.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes
+in the target and transforms it.</p></li>
+<li><p><strong>is_valid_file</strong> – A function that takes path of a file
+and check if the file is a valid file (used to check of corrupt files)
+both extensions and is_valid_file should not be passed.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.DatasetFolder.find_classes">
+<span class="sig-name descname"><span class="pre">find_classes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">directory</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/folder.html#DatasetFolder.find_classes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.DatasetFolder.find_classes" title="Permalink to this definition">¶</a></dt>
+<dd><p>Find the class folders in a dataset structured as follows:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>directory/
+├── class_x
+│   ├── xxx.ext
+│   ├── xxy.ext
+│   └── ...
+│       └── xxz.ext
+└── class_y
+    ├── 123.ext
+    ├── nsdf3.ext
+    └── ...
+    └── asd932_.ext
+</pre></div>
+</div>
+<p>This method can be overridden to only consider
+a subset of classes, or to adapt to a different dataset directory structure.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>directory</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Root directory path, corresponding to <code class="docutils literal notranslate"><span class="pre">self.root</span></code></p>
+</dd>
+<dt class="field-even">Raises<span class="colon">:</span></dt>
+<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#FileNotFoundError" title="(in Python v3.12)"><strong>FileNotFoundError</strong></a> – If <code class="docutils literal notranslate"><span class="pre">dir</span></code> has no class folders.</p>
+</dd>
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>List of all classes and dictionary mapping each class to an index.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>(Tuple[List[<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)">str</a>], Dict[<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)">str</a>, <a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)">int</a>]])</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.DatasetFolder.make_dataset">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">make_dataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">directory</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">class_to_idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">extensions</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">is_valid_file</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/folder.html#DatasetFolder.make_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.DatasetFolder.make_dataset" title="Permalink to this definition">¶</a></dt>
+<dd><p>Generates a list of samples of a form (path_to_sample, class).</p>
+<p>This can be overridden to e.g. read files from a compressed zip file instead of from the disk.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>directory</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – root dataset directory, corresponding to <code class="docutils literal notranslate"><span class="pre">self.root</span></code>.</p></li>
+<li><p><strong>class_to_idx</strong> (<em>Dict</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – Dictionary mapping class name to class index.</p></li>
+<li><p><strong>extensions</strong> (<em>optional</em>) – A list of allowed extensions.
+Either extensions or is_valid_file should be passed. Defaults to None.</p></li>
+<li><p><strong>is_valid_file</strong> (<em>optional</em>) – A function that takes path of a file
+and checks if the file is a valid file
+(used to check of corrupt files) both extensions and
+is_valid_file should not be passed. Defaults to None.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Raises<span class="colon">:</span></dt>
+<dd class="field-even"><ul class="simple">
+<li><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#ValueError" title="(in Python v3.12)"><strong>ValueError</strong></a> – In case <code class="docutils literal notranslate"><span class="pre">class_to_idx</span></code> is empty.</p></li>
+<li><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#ValueError" title="(in Python v3.12)"><strong>ValueError</strong></a> – In case <code class="docutils literal notranslate"><span class="pre">extensions</span></code> and <code class="docutils literal notranslate"><span class="pre">is_valid_file</span></code> are None or both are not None.</p></li>
+<li><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#FileNotFoundError" title="(in Python v3.12)"><strong>FileNotFoundError</strong></a> – In case no valid file was found for any class.</p></li>
+</ul>
+</dd>
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>samples of a form (path_to_sample, class)</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>List[Tuple[<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)">str</a>, <a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)">int</a>]]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.ImageFolder.html" class="btn btn-neutral float-right" title="ImageFolder" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.MovingMNIST.html" class="btn btn-neutral" title="MovingMNIST" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">DatasetFolder</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.EMNIST.html b/0.16/generated/torchvision.datasets.EMNIST.html
new file mode 100644
index 00000000000..f85c2edd715
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.EMNIST.html
@@ -0,0 +1,734 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>EMNIST &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="EuroSAT" href="torchvision.datasets.EuroSAT.html" />
+    <link rel="prev" title="DTD" href="torchvision.datasets.DTD.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>EMNIST</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.EMNIST.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="emnist">
+<h1>EMNIST<a class="headerlink" href="#emnist" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.EMNIST">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">EMNIST</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/mnist.html#EMNIST"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.EMNIST" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.westernsydney.edu.au/bens/home/reproducible_research/emnist">EMNIST</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where <code class="docutils literal notranslate"><span class="pre">EMNIST/raw/train-images-idx3-ubyte</span></code>
+and  <code class="docutils literal notranslate"><span class="pre">EMNIST/raw/t10k-images-idx3-ubyte</span></code> exist.</p></li>
+<li><p><strong>split</strong> (<em>string</em>) – The dataset has 6 different splits: <code class="docutils literal notranslate"><span class="pre">byclass</span></code>, <code class="docutils literal notranslate"><span class="pre">bymerge</span></code>,
+<code class="docutils literal notranslate"><span class="pre">balanced</span></code>, <code class="docutils literal notranslate"><span class="pre">letters</span></code>, <code class="docutils literal notranslate"><span class="pre">digits</span></code> and <code class="docutils literal notranslate"><span class="pre">mnist</span></code>. This argument specifies
+which one to use.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, creates dataset from <code class="docutils literal notranslate"><span class="pre">training.pt</span></code>,
+otherwise from <code class="docutils literal notranslate"><span class="pre">test.pt</span></code>.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.EMNIST.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.EMNIST.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.EuroSAT.html" class="btn btn-neutral float-right" title="EuroSAT" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.DTD.html" class="btn btn-neutral" title="DTD" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">EMNIST</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.ETH3DStereo.html b/0.16/generated/torchvision.datasets.ETH3DStereo.html
new file mode 100644
index 00000000000..43fe81c0f78
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.ETH3DStereo.html
@@ -0,0 +1,769 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ETH3DStereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Middlebury2014Stereo" href="torchvision.datasets.Middlebury2014Stereo.html" />
+    <link rel="prev" title="InStereo2k" href="torchvision.datasets.InStereo2k.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>ETH3DStereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.ETH3DStereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="eth3dstereo">
+<h1>ETH3DStereo<a class="headerlink" href="#eth3dstereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.ETH3DStereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">ETH3DStereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#ETH3DStereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.ETH3DStereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>ETH3D <a class="reference external" href="https://www.eth3d.net/datasets">Low-Res Two-View</a> dataset.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">ETH3D</span>
+        <span class="n">two_view_training</span>
+            <span class="n">scene1</span>
+                <span class="n">im1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im0</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">images</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">cameras</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+            <span class="n">scene2</span>
+                <span class="n">im1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im0</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">images</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">cameras</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+            <span class="o">...</span>
+        <span class="n">two_view_training_gt</span>
+            <span class="n">scene1</span>
+                <span class="n">disp0GT</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">mask0nocc</span><span class="o">.</span><span class="n">png</span>
+            <span class="n">scene2</span>
+                <span class="n">disp0GT</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">mask0nocc</span><span class="o">.</span><span class="n">png</span>
+            <span class="o">...</span>
+        <span class="n">two_view_testing</span>
+            <span class="n">scene1</span>
+                <span class="n">im1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im0</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">images</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">cameras</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+            <span class="n">scene2</span>
+                <span class="n">im1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im0</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">images</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">cameras</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+            <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the ETH3D Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split of scenes, either “train” (default) or “test”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.ETH3DStereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#ETH3DStereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.ETH3DStereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is implicitly <code class="docutils literal notranslate"><span class="pre">None</span></code> if the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter does not
+generate a valid mask.
+Both <code class="docutils literal notranslate"><span class="pre">disparity</span></code> and <code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> are <code class="docutils literal notranslate"><span class="pre">None</span></code> if the dataset split is test.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Middlebury2014Stereo.html" class="btn btn-neutral float-right" title="Middlebury2014Stereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.InStereo2k.html" class="btn btn-neutral" title="InStereo2k" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ETH3DStereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.EuroSAT.html b/0.16/generated/torchvision.datasets.EuroSAT.html
new file mode 100644
index 00000000000..37e8c632a52
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.EuroSAT.html
@@ -0,0 +1,728 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>EuroSAT &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FakeData" href="torchvision.datasets.FakeData.html" />
+    <link rel="prev" title="EMNIST" href="torchvision.datasets.EMNIST.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>EuroSAT</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.EuroSAT.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="eurosat">
+<h1>EuroSAT<a class="headerlink" href="#eurosat" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.EuroSAT">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">EuroSAT</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/eurosat.html#EuroSAT"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.EuroSAT" title="Permalink to this definition">¶</a></dt>
+<dd><p>RGB version of the <a class="reference external" href="https://github.com/phelber/eurosat">EuroSAT</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where <code class="docutils literal notranslate"><span class="pre">root/eurosat</span></code> exists.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again. Default is False.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.EuroSAT.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.EuroSAT.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(sample, target) where target is class_index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.FakeData.html" class="btn btn-neutral float-right" title="FakeData" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.EMNIST.html" class="btn btn-neutral" title="EMNIST" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">EuroSAT</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.FER2013.html b/0.16/generated/torchvision.datasets.FER2013.html
new file mode 100644
index 00000000000..9d2f7ae8b46
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.FER2013.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FER2013 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FGVCAircraft" href="torchvision.datasets.FGVCAircraft.html" />
+    <link rel="prev" title="FashionMNIST" href="torchvision.datasets.FashionMNIST.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>FER2013</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.FER2013.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fer2013">
+<h1>FER2013<a class="headerlink" href="#fer2013" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.FER2013">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">FER2013</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/fer2013.html#FER2013"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FER2013" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge">FER2013</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">root/fer2013</span></code> exists.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default), or <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in an PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.FER2013.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/fer2013.html#FER2013.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FER2013.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.FGVCAircraft.html" class="btn btn-neutral float-right" title="FGVCAircraft" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.FashionMNIST.html" class="btn btn-neutral" title="FashionMNIST" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FER2013</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.FGVCAircraft.html b/0.16/generated/torchvision.datasets.FGVCAircraft.html
new file mode 100644
index 00000000000..df685fe873a
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.FGVCAircraft.html
@@ -0,0 +1,745 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FGVCAircraft &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Flickr8k" href="torchvision.datasets.Flickr8k.html" />
+    <link rel="prev" title="FER2013" href="torchvision.datasets.FER2013.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>FGVCAircraft</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.FGVCAircraft.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fgvcaircraft">
+<h1>FGVCAircraft<a class="headerlink" href="#fgvcaircraft" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.FGVCAircraft">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">FGVCAircraft</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'trainval'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">annotation_level</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'variant'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/fgvc_aircraft.html#FGVCAircraft"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FGVCAircraft" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/">FGVC Aircraft</a> Dataset.</p>
+<p>The dataset contains 10,000 images of aircraft, with 100 images for each of 100
+different aircraft model variants, most of which are airplanes.
+Aircraft models are organized in a three-levels hierarchy. The three levels, from
+finer to coarser, are:</p>
+<ul class="simple">
+<li><dl class="simple">
+<dt><code class="docutils literal notranslate"><span class="pre">variant</span></code>, e.g. Boeing 737-700. A variant collapses all the models that are visually</dt><dd><p>indistinguishable into one class. The dataset comprises 100 different variants.</p>
+</dd>
+</dl>
+</li>
+<li><p><code class="docutils literal notranslate"><span class="pre">family</span></code>, e.g. Boeing 737. The dataset comprises 70 different families.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">manufacturer</span></code>, e.g. Boeing. The dataset comprises 30 different manufacturers.</p></li>
+</ul>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the FGVC Aircraft dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">train</span></code>, <code class="docutils literal notranslate"><span class="pre">val</span></code>,
+<code class="docutils literal notranslate"><span class="pre">trainval</span></code> and <code class="docutils literal notranslate"><span class="pre">test</span></code>.</p></li>
+<li><p><strong>annotation_level</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – The annotation level, supports <code class="docutils literal notranslate"><span class="pre">variant</span></code>,
+<code class="docutils literal notranslate"><span class="pre">family</span></code> and <code class="docutils literal notranslate"><span class="pre">manufacturer</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.FGVCAircraft.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/fgvc_aircraft.html#FGVCAircraft.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FGVCAircraft.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Flickr8k.html" class="btn btn-neutral float-right" title="Flickr8k" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.FER2013.html" class="btn btn-neutral" title="FER2013" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FGVCAircraft</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.FakeData.html b/0.16/generated/torchvision.datasets.FakeData.html
new file mode 100644
index 00000000000..878828b8cf7
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.FakeData.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FakeData &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FashionMNIST" href="torchvision.datasets.FashionMNIST.html" />
+    <link rel="prev" title="EuroSAT" href="torchvision.datasets.EuroSAT.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>FakeData</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.FakeData.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fakedata">
+<h1>FakeData<a class="headerlink" href="#fakedata" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.FakeData">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">FakeData</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1000</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(3,</span> <span class="pre">224,</span> <span class="pre">224)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">10</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">random_offset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/fakedata.html#FakeData"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FakeData" title="Permalink to this definition">¶</a></dt>
+<dd><p>A fake dataset that returns randomly generated images and returns them as PIL images</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Size of the dataset. Default: 1000 images</p></li>
+<li><p><strong>image_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em>, </em><em>optional</em>) – Size if the returned images. Default: (3, 224, 224)</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Number of classes in the dataset. Default: 10</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>random_offset</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Offsets the index-based random seed used to
+generate each image. Default: 0</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">FakeData</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div><dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.FakeData.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/fakedata.html#FakeData.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FakeData.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is class_index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.FashionMNIST.html" class="btn btn-neutral float-right" title="FashionMNIST" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.EuroSAT.html" class="btn btn-neutral" title="EuroSAT" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FakeData</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.FallingThingsStereo.html b/0.16/generated/torchvision.datasets.FallingThingsStereo.html
new file mode 100644
index 00000000000..8f4278df4db
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.FallingThingsStereo.html
@@ -0,0 +1,763 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FallingThingsStereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SceneFlowStereo" href="torchvision.datasets.SceneFlowStereo.html" />
+    <link rel="prev" title="CREStereo" href="torchvision.datasets.CREStereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>FallingThingsStereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.FallingThingsStereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fallingthingsstereo">
+<h1>FallingThingsStereo<a class="headerlink" href="#fallingthingsstereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.FallingThingsStereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">FallingThingsStereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">variant</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'single'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#FallingThingsStereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FallingThingsStereo" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://research.nvidia.com/publication/2018-06_falling-things-synthetic-dataset-3d-object-detection-and-pose-estimation">FallingThings</a> dataset.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">FallingThings</span>
+        <span class="n">single</span>
+            <span class="n">dir1</span>
+                <span class="n">scene1</span>
+                    <span class="n">_object_settings</span><span class="o">.</span><span class="n">json</span>
+                    <span class="n">_camera_settings</span><span class="o">.</span><span class="n">json</span>
+                    <span class="n">image1</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">image1</span><span class="o">.</span><span class="n">right</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">image1</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">jpg</span>
+                    <span class="n">image1</span><span class="o">.</span><span class="n">right</span><span class="o">.</span><span class="n">jpg</span>
+                    <span class="n">image2</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">image2</span><span class="o">.</span><span class="n">right</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">image2</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">jpg</span>
+                    <span class="n">image2</span><span class="o">.</span><span class="n">right</span>
+                    <span class="o">...</span>
+                <span class="n">scene2</span>
+            <span class="o">...</span>
+        <span class="n">mixed</span>
+            <span class="n">scene1</span>
+                <span class="n">_object_settings</span><span class="o">.</span><span class="n">json</span>
+                <span class="n">_camera_settings</span><span class="o">.</span><span class="n">json</span>
+                <span class="n">image1</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">image1</span><span class="o">.</span><span class="n">right</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">image1</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">jpg</span>
+                <span class="n">image1</span><span class="o">.</span><span class="n">right</span><span class="o">.</span><span class="n">jpg</span>
+                <span class="n">image2</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">image2</span><span class="o">.</span><span class="n">right</span><span class="o">.</span><span class="n">depth</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">image2</span><span class="o">.</span><span class="n">left</span><span class="o">.</span><span class="n">jpg</span>
+                <span class="n">image2</span><span class="o">.</span><span class="n">right</span>
+                <span class="o">...</span>
+            <span class="n">scene2</span>
+            <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where FallingThings is located.</p></li>
+<li><p><strong>variant</strong> (<em>string</em>) – Which variant to use. Either “single”, “mixed”, or “both”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.FallingThingsStereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#FallingThingsStereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FallingThingsStereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 3-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+If a <code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is generated within the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter,
+a 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code> is returned.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.SceneFlowStereo.html" class="btn btn-neutral float-right" title="SceneFlowStereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.CREStereo.html" class="btn btn-neutral" title="CREStereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FallingThingsStereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.FashionMNIST.html b/0.16/generated/torchvision.datasets.FashionMNIST.html
new file mode 100644
index 00000000000..b52756054bd
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.FashionMNIST.html
@@ -0,0 +1,731 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FashionMNIST &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FER2013" href="torchvision.datasets.FER2013.html" />
+    <link rel="prev" title="FakeData" href="torchvision.datasets.FakeData.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>FashionMNIST</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.FashionMNIST.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fashionmnist">
+<h1>FashionMNIST<a class="headerlink" href="#fashionmnist" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.FashionMNIST">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">FashionMNIST</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/mnist.html#FashionMNIST"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FashionMNIST" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/zalandoresearch/fashion-mnist">Fashion-MNIST</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where <code class="docutils literal notranslate"><span class="pre">FashionMNIST/raw/train-images-idx3-ubyte</span></code>
+and  <code class="docutils literal notranslate"><span class="pre">FashionMNIST/raw/t10k-images-idx3-ubyte</span></code> exist.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, creates dataset from <code class="docutils literal notranslate"><span class="pre">train-images-idx3-ubyte</span></code>,
+otherwise from <code class="docutils literal notranslate"><span class="pre">t10k-images-idx3-ubyte</span></code>.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.FashionMNIST.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.FashionMNIST.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.FER2013.html" class="btn btn-neutral float-right" title="FER2013" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.FakeData.html" class="btn btn-neutral" title="FakeData" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FashionMNIST</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Flickr30k.html b/0.16/generated/torchvision.datasets.Flickr30k.html
new file mode 100644
index 00000000000..75f02fdc47d
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Flickr30k.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Flickr30k &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Flowers102" href="torchvision.datasets.Flowers102.html" />
+    <link rel="prev" title="Flickr8k" href="torchvision.datasets.Flickr8k.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Flickr30k</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Flickr30k.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="flickr30k">
+<h1>Flickr30k<a class="headerlink" href="#flickr30k" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Flickr30k">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Flickr30k</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ann_file</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/flickr.html#Flickr30k"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Flickr30k" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://bryanplummer.com/Flickr30kEntities/">Flickr30k Entities</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where images are downloaded to.</p></li>
+<li><p><strong>ann_file</strong> (<em>string</em>) – Path to annotation file.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.PILToTensor</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Flickr30k.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/flickr.html#Flickr30k.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Flickr30k.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Tuple (image, target). target is a list of captions for the image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Flowers102.html" class="btn btn-neutral float-right" title="Flowers102" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Flickr8k.html" class="btn btn-neutral" title="Flickr8k" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Flickr30k</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Flickr8k.html b/0.16/generated/torchvision.datasets.Flickr8k.html
new file mode 100644
index 00000000000..faa3085a6d0
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Flickr8k.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Flickr8k &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Flickr30k" href="torchvision.datasets.Flickr30k.html" />
+    <link rel="prev" title="FGVCAircraft" href="torchvision.datasets.FGVCAircraft.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Flickr8k</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Flickr8k.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="flickr8k">
+<h1>Flickr8k<a class="headerlink" href="#flickr8k" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Flickr8k">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Flickr8k</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ann_file</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/flickr.html#Flickr8k"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Flickr8k" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://hockenmaier.cs.illinois.edu/8k-pictures.html">Flickr8k Entities</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where images are downloaded to.</p></li>
+<li><p><strong>ann_file</strong> (<em>string</em>) – Path to annotation file.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.PILToTensor</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Flickr8k.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/flickr.html#Flickr8k.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Flickr8k.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Tuple (image, target). target is a list of captions for the image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Flickr30k.html" class="btn btn-neutral float-right" title="Flickr30k" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.FGVCAircraft.html" class="btn btn-neutral" title="FGVCAircraft" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Flickr8k</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Flowers102.html b/0.16/generated/torchvision.datasets.Flowers102.html
new file mode 100644
index 00000000000..c4848dbc28d
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Flowers102.html
@@ -0,0 +1,737 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Flowers102 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Food101" href="torchvision.datasets.Food101.html" />
+    <link rel="prev" title="Flickr30k" href="torchvision.datasets.Flickr30k.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Flowers102</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Flowers102.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="flowers102">
+<h1>Flowers102<a class="headerlink" href="#flowers102" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Flowers102">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Flowers102</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/flowers102.html#Flowers102"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Flowers102" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/flowers/102/">Oxford 102 Flower</a> Dataset.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>This class needs <a class="reference external" href="https://docs.scipy.org/doc/">scipy</a> to load target files from <cite>.mat</cite> format.</p>
+</div>
+<p>Oxford 102 Flower is an image classification dataset consisting of 102 flower categories. The
+flowers were chosen to be flowers commonly occurring in the United Kingdom. Each class consists of
+between 40 and 258 images.</p>
+<p>The images have large scale, pose and light variations. In addition, there are categories that
+have large variations within the category, and several very similar categories.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default), <code class="docutils literal notranslate"><span class="pre">&quot;val&quot;</span></code>, or <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in an PIL image and returns a
+transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Flowers102.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/flowers102.html#Flowers102.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Flowers102.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Food101.html" class="btn btn-neutral float-right" title="Food101" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Flickr30k.html" class="btn btn-neutral" title="Flickr30k" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Flowers102</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.FlyingChairs.html b/0.16/generated/torchvision.datasets.FlyingChairs.html
new file mode 100644
index 00000000000..234a1bdd1fc
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.FlyingChairs.html
@@ -0,0 +1,743 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FlyingChairs &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FlyingThings3D" href="torchvision.datasets.FlyingThings3D.html" />
+    <link rel="prev" title="WIDERFace" href="torchvision.datasets.WIDERFace.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>FlyingChairs</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.FlyingChairs.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="flyingchairs">
+<h1>FlyingChairs<a class="headerlink" href="#flyingchairs" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.FlyingChairs">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">FlyingChairs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#FlyingChairs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FlyingChairs" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://lmb.informatik.uni-freiburg.de/resources/datasets/FlyingChairs.en.html#flyingchairs">FlyingChairs</a> Dataset for optical flow.</p>
+<p>You will also need to download the FlyingChairs_train_val.txt file from the dataset page.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">FlyingChairs</span>
+        <span class="n">data</span>
+            <span class="mi">00001</span><span class="n">_flow</span><span class="o">.</span><span class="n">flo</span>
+            <span class="mi">00001</span><span class="n">_img1</span><span class="o">.</span><span class="n">ppm</span>
+            <span class="mi">00001</span><span class="n">_img2</span><span class="o">.</span><span class="n">ppm</span>
+            <span class="o">...</span>
+        <span class="n">FlyingChairs_train_val</span><span class="o">.</span><span class="n">txt</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the FlyingChairs Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, either “train” (default) or “val”</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in
+<code class="docutils literal notranslate"><span class="pre">img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask</span></code> and returns a transformed version.
+<code class="docutils literal notranslate"><span class="pre">valid_flow_mask</span></code> is expected for consistency with other datasets which
+return a built-in valid mask, such as <a class="reference internal" href="torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-class docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.FlyingChairs.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#FlyingChairs.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FlyingChairs.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 3-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow)</span></code>.
+The flow is a numpy array of shape (2, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">flow</span></code> is None if <code class="docutils literal notranslate"><span class="pre">split=&quot;val&quot;</span></code>.
+If a valid flow mask is generated within the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter,
+a 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask)</span></code> is returned.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.FlyingThings3D.html" class="btn btn-neutral float-right" title="FlyingThings3D" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.WIDERFace.html" class="btn btn-neutral" title="WIDERFace" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FlyingChairs</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.FlyingThings3D.html b/0.16/generated/torchvision.datasets.FlyingThings3D.html
new file mode 100644
index 00000000000..84a9b6e20a7
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.FlyingThings3D.html
@@ -0,0 +1,748 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FlyingThings3D &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="HD1K" href="torchvision.datasets.HD1K.html" />
+    <link rel="prev" title="FlyingChairs" href="torchvision.datasets.FlyingChairs.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>FlyingThings3D</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.FlyingThings3D.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="flyingthings3d">
+<h1>FlyingThings3D<a class="headerlink" href="#flyingthings3d" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.FlyingThings3D">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">FlyingThings3D</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pass_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'clean'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">camera</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'left'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#FlyingThings3D"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FlyingThings3D" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html">FlyingThings3D</a> dataset for optical flow.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">FlyingThings3D</span>
+        <span class="n">frames_cleanpass</span>
+            <span class="n">TEST</span>
+            <span class="n">TRAIN</span>
+        <span class="n">frames_finalpass</span>
+            <span class="n">TEST</span>
+            <span class="n">TRAIN</span>
+        <span class="n">optical_flow</span>
+            <span class="n">TEST</span>
+            <span class="n">TRAIN</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the intel FlyingThings3D Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, either “train” (default) or “test”</p></li>
+<li><p><strong>pass_name</strong> (<em>string</em><em>, </em><em>optional</em>) – The pass to use, either “clean” (default) or “final” or “both”. See link above for
+details on the different passes.</p></li>
+<li><p><strong>camera</strong> (<em>string</em><em>, </em><em>optional</em>) – Which camera to return images from. Can be either “left” (default) or “right” or “both”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in
+<code class="docutils literal notranslate"><span class="pre">img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask</span></code> and returns a transformed version.
+<code class="docutils literal notranslate"><span class="pre">valid_flow_mask</span></code> is expected for consistency with other datasets which
+return a built-in valid mask, such as <a class="reference internal" href="torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-class docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.FlyingThings3D.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#FlyingThings3D.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.FlyingThings3D.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 3-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow)</span></code>.
+The flow is a numpy array of shape (2, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">flow</span></code> is None if <code class="docutils literal notranslate"><span class="pre">split=&quot;test&quot;</span></code>.
+If a valid flow mask is generated within the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter,
+a 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask)</span></code> is returned.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.HD1K.html" class="btn btn-neutral float-right" title="HD1K" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.FlyingChairs.html" class="btn btn-neutral" title="FlyingChairs" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FlyingThings3D</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Food101.html b/0.16/generated/torchvision.datasets.Food101.html
new file mode 100644
index 00000000000..d76519566cd
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Food101.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Food101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="GTSRB" href="torchvision.datasets.GTSRB.html" />
+    <link rel="prev" title="Flowers102" href="torchvision.datasets.Flowers102.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Food101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Food101.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="food101">
+<h1>Food101<a class="headerlink" href="#food101" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Food101">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Food101</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/food101.html#Food101"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Food101" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/">The Food-101 Data Set</a>.</p>
+<p>The Food-101 is a challenging data set of 101 food categories with 101,000 images.
+For each class, 250 manually reviewed test images are provided as well as 750 training images.
+On purpose, the training images were not cleaned, and thus still contain some amount of noise.
+This comes mostly in the form of intense colors and sometimes wrong labels. All images were
+rescaled to have a maximum side length of 512 pixels.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default) and <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again. Default is False.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Food101.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/food101.html#Food101.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Food101.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.GTSRB.html" class="btn btn-neutral float-right" title="GTSRB" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Flowers102.html" class="btn btn-neutral" title="Flowers102" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Food101</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.GTSRB.html b/0.16/generated/torchvision.datasets.GTSRB.html
new file mode 100644
index 00000000000..e0949ced80b
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.GTSRB.html
@@ -0,0 +1,728 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>GTSRB &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="INaturalist" href="torchvision.datasets.INaturalist.html" />
+    <link rel="prev" title="Food101" href="torchvision.datasets.Food101.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>GTSRB</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.GTSRB.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="gtsrb">
+<h1>GTSRB<a class="headerlink" href="#gtsrb" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.GTSRB">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">GTSRB</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/gtsrb.html#GTSRB"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.GTSRB" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://benchmark.ini.rub.de/">German Traffic Sign Recognition Benchmark (GTSRB)</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default), or <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.GTSRB.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/gtsrb.html#GTSRB.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.GTSRB.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.INaturalist.html" class="btn btn-neutral float-right" title="INaturalist" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Food101.html" class="btn btn-neutral" title="Food101" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">GTSRB</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.HD1K.html b/0.16/generated/torchvision.datasets.HD1K.html
new file mode 100644
index 00000000000..f90a0317b64
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.HD1K.html
@@ -0,0 +1,740 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>HD1K &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="KittiFlow" href="torchvision.datasets.KittiFlow.html" />
+    <link rel="prev" title="FlyingThings3D" href="torchvision.datasets.FlyingThings3D.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>HD1K</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.HD1K.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="hd1k">
+<h1>HD1K<a class="headerlink" href="#hd1k" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.HD1K">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">HD1K</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#HD1K"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.HD1K" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://hci-benchmark.iwr.uni-heidelberg.de/">HD1K</a> dataset for optical flow.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">hd1k</span>
+        <span class="n">hd1k_challenge</span>
+            <span class="n">image_2</span>
+        <span class="n">hd1k_flow_gt</span>
+            <span class="n">flow_occ</span>
+        <span class="n">hd1k_input</span>
+            <span class="n">image_2</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the HD1K Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, either “train” (default) or “test”</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in
+<code class="docutils literal notranslate"><span class="pre">img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask</span></code> and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.HD1K.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#HD1K.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.HD1K.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask)</span></code> where <code class="docutils literal notranslate"><span class="pre">valid_flow_mask</span></code>
+is a numpy boolean mask of shape (H, W)
+indicating which flow values are valid. The flow is a numpy array of
+shape (2, H, W) and the images are PIL images. <code class="docutils literal notranslate"><span class="pre">flow</span></code> and <code class="docutils literal notranslate"><span class="pre">valid_flow_mask</span></code> are None if
+<code class="docutils literal notranslate"><span class="pre">split=&quot;test&quot;</span></code>.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.KittiFlow.html" class="btn btn-neutral float-right" title="KittiFlow" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.FlyingThings3D.html" class="btn btn-neutral" title="FlyingThings3D" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">HD1K</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.HMDB51.html b/0.16/generated/torchvision.datasets.HMDB51.html
new file mode 100644
index 00000000000..bf78d8c65ff
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.HMDB51.html
@@ -0,0 +1,757 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>HMDB51 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Kinetics" href="torchvision.datasets.Kinetics.html" />
+    <link rel="prev" title="CocoCaptions" href="torchvision.datasets.CocoCaptions.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>HMDB51</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.HMDB51.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="hmdb51">
+<h1>HMDB51<a class="headerlink" href="#hmdb51" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.HMDB51">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">HMDB51</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">annotation_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">frames_per_clip</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">step_between_clips</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">frame_rate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fold</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_precomputed_metadata</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_workers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_min_dimension</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_audio_samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'THWC'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/hmdb51.html#HMDB51"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.HMDB51" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://serre-lab.clps.brown.edu/resource/hmdb-a-large-human-motion-database/">HMDB51</a>
+dataset.</p>
+<p>HMDB51 is an action recognition video dataset.
+This dataset consider every video as a collection of video clips of fixed size, specified
+by <code class="docutils literal notranslate"><span class="pre">frames_per_clip</span></code>, where the step in frames between each clip is given by
+<code class="docutils literal notranslate"><span class="pre">step_between_clips</span></code>.</p>
+<p>To give an example, for 2 videos with 10 and 15 frames respectively, if <code class="docutils literal notranslate"><span class="pre">frames_per_clip=5</span></code>
+and <code class="docutils literal notranslate"><span class="pre">step_between_clips=5</span></code>, the dataset size will be (2 + 3) = 5, where the first two
+elements will come from video 1, and the next three elements from video 2.
+Note that we drop clips which do not have exactly <code class="docutils literal notranslate"><span class="pre">frames_per_clip</span></code> elements, so not all
+frames in a video might be present.</p>
+<p>Internally, it uses a VideoClips object to handle clip creation.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the HMDB51 Dataset.</p></li>
+<li><p><strong>annotation_path</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Path to the folder containing the split files.</p></li>
+<li><p><strong>frames_per_clip</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of frames in a clip.</p></li>
+<li><p><strong>step_between_clips</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of frames between each clip.</p></li>
+<li><p><strong>fold</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Which fold to use. Should be between 1 and 3.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If <code class="docutils literal notranslate"><span class="pre">True</span></code>, creates a dataset from the train split,
+otherwise from the <code class="docutils literal notranslate"><span class="pre">test</span></code> split.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a TxHxWxC video
+and returns a transformed version.</p></li>
+<li><p><strong>output_format</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – The format of the output video tensors (before transforms).
+Can be either “THWC” (default) or “TCHW”.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><p>A 3-tuple with the following entries:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>video (Tensor[T, H, W, C] or Tensor[T, C, H, W]): The <cite>T</cite> video frames</p></li>
+<li><p>audio(Tensor[K, L]): the audio frames, where <cite>K</cite> is the number of channels
+and <cite>L</cite> is the number of points</p></li>
+<li><p>label (int): class of the video clip</p></li>
+</ul>
+</div></blockquote>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.HMDB51.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/hmdb51.html#HMDB51.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.HMDB51.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Kinetics.html" class="btn btn-neutral float-right" title="Kinetics" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.CocoCaptions.html" class="btn btn-neutral" title="CocoCaptions" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">HMDB51</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.INaturalist.html b/0.16/generated/torchvision.datasets.INaturalist.html
new file mode 100644
index 00000000000..c00c70a592d
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.INaturalist.html
@@ -0,0 +1,762 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>INaturalist &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ImageNet" href="torchvision.datasets.ImageNet.html" />
+    <link rel="prev" title="GTSRB" href="torchvision.datasets.GTSRB.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>INaturalist</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.INaturalist.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="inaturalist">
+<h1>INaturalist<a class="headerlink" href="#inaturalist" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.INaturalist">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">INaturalist</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">version</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'2021_train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'full'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/inaturalist.html#INaturalist"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.INaturalist" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/visipedia/inat_comp">iNaturalist</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where the image files are stored.
+This class does not require/use annotation files.</p></li>
+<li><p><strong>version</strong> (<em>string</em><em>, </em><em>optional</em>) – Which version of the dataset to download/use. One of
+‘2017’, ‘2018’, ‘2019’, ‘2021_train’, ‘2021_train_mini’, ‘2021_valid’.
+Default: <cite>2021_train</cite>.</p></li>
+<li><p><strong>target_type</strong> (<em>string</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em>, </em><em>optional</em>) – <p>Type of target to use, for 2021 versions, one of:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">full</span></code>: the full category (species)</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">kingdom</span></code>: e.g. “Animalia”</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">phylum</span></code>: e.g. “Arthropoda”</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">class</span></code>: e.g. “Insecta”</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">order</span></code>: e.g. “Coleoptera”</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">family</span></code>: e.g. “Cleridae”</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">genus</span></code>: e.g. “Trichodes”</p></li>
+</ul>
+<p>for 2017-2019 versions, one of:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">full</span></code>: the full (numeric) category</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">super</span></code>: the super category, e.g. “Amphibians”</p></li>
+</ul>
+<p>Can also be a list to output a tuple with all specified target types.
+Defaults to <code class="docutils literal notranslate"><span class="pre">full</span></code>.</p>
+</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.INaturalist.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/inaturalist.html#INaturalist.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.INaturalist.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where the type of target specified by target_type.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.INaturalist.category_name">
+<span class="sig-name descname"><span class="pre">category_name</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">category_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">category_id</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></span><a class="reference internal" href="../_modules/torchvision/datasets/inaturalist.html#INaturalist.category_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.INaturalist.category_name" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>category_type</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – one of “full”, “kingdom”, “phylum”, “class”, “order”, “family”, “genus” or “super”</p></li>
+<li><p><strong>category_id</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – an index (class id) from this category</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>the name of the category</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.ImageNet.html" class="btn btn-neutral float-right" title="ImageNet" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.GTSRB.html" class="btn btn-neutral" title="GTSRB" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">INaturalist</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.ImageFolder.html b/0.16/generated/torchvision.datasets.ImageFolder.html
new file mode 100644
index 00000000000..020b0a1366a
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.ImageFolder.html
@@ -0,0 +1,719 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ImageFolder &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="VisionDataset" href="torchvision.datasets.VisionDataset.html" />
+    <link rel="prev" title="DatasetFolder" href="torchvision.datasets.DatasetFolder.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>ImageFolder</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.ImageFolder.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="imagefolder">
+<h1>ImageFolder<a class="headerlink" href="#imagefolder" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.ImageFolder">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">ImageFolder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">root:</span> <span class="pre">str,</span> <span class="pre">transform:</span> <span class="pre">~typing.Optional[~typing.Callable]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">target_transform:</span> <span class="pre">~typing.Optional[~typing.Callable]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">loader:</span> <span class="pre">~typing.Callable[[str],</span> <span class="pre">~typing.Any]</span> <span class="pre">=</span> <span class="pre">&lt;function</span> <span class="pre">default_loader&gt;,</span> <span class="pre">is_valid_file:</span> <span class="pre">~typing.Optional[~typing.Callable[[str],</span> <span class="pre">bool]]</span> <span class="pre">=</span> <span class="pre">None</span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/folder.html#ImageFolder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.ImageFolder" title="Permalink to this definition">¶</a></dt>
+<dd><p>A generic data loader where the images are arranged in this way by default:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span><span class="o">/</span><span class="n">dog</span><span class="o">/</span><span class="n">xxx</span><span class="o">.</span><span class="n">png</span>
+<span class="n">root</span><span class="o">/</span><span class="n">dog</span><span class="o">/</span><span class="n">xxy</span><span class="o">.</span><span class="n">png</span>
+<span class="n">root</span><span class="o">/</span><span class="n">dog</span><span class="o">/</span><span class="p">[</span><span class="o">...</span><span class="p">]</span><span class="o">/</span><span class="n">xxz</span><span class="o">.</span><span class="n">png</span>
+
+<span class="n">root</span><span class="o">/</span><span class="n">cat</span><span class="o">/</span><span class="mf">123.</span><span class="n">png</span>
+<span class="n">root</span><span class="o">/</span><span class="n">cat</span><span class="o">/</span><span class="n">nsdf3</span><span class="o">.</span><span class="n">png</span>
+<span class="n">root</span><span class="o">/</span><span class="n">cat</span><span class="o">/</span><span class="p">[</span><span class="o">...</span><span class="p">]</span><span class="o">/</span><span class="n">asd932_</span><span class="o">.</span><span class="n">png</span>
+</pre></div>
+</div>
+<p>This class inherits from <a class="reference internal" href="torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder" title="torchvision.datasets.DatasetFolder"><code class="xref py py-class docutils literal notranslate"><span class="pre">DatasetFolder</span></code></a> so
+the same methods can be overridden to customize the dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory path.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>loader</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function to load an image given its path.</p></li>
+<li><p><strong>is_valid_file</strong> – A function that takes path of an Image file
+and check if the file is a valid file (used to check of corrupt files)</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.VisionDataset.html" class="btn btn-neutral float-right" title="VisionDataset" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.DatasetFolder.html" class="btn btn-neutral" title="DatasetFolder" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ImageFolder</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.ImageNet.html b/0.16/generated/torchvision.datasets.ImageNet.html
new file mode 100644
index 00000000000..ea02e4954b5
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.ImageNet.html
@@ -0,0 +1,734 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ImageNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="KMNIST" href="torchvision.datasets.KMNIST.html" />
+    <link rel="prev" title="INaturalist" href="torchvision.datasets.INaturalist.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>ImageNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.ImageNet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="imagenet">
+<h1>ImageNet<a class="headerlink" href="#imagenet" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.ImageNet">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">ImageNet</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/imagenet.html#ImageNet"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.ImageNet" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://image-net.org/">ImageNet</a> 2012 Classification Dataset.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Before using this class, it is required to download ImageNet 2012 dataset from
+<a class="reference external" href="https://image-net.org/challenges/LSVRC/2012/2012-downloads.php">here</a> and
+place the files <code class="docutils literal notranslate"><span class="pre">ILSVRC2012_devkit_t12.tar.gz</span></code> and <code class="docutils literal notranslate"><span class="pre">ILSVRC2012_img_train.tar</span></code>
+or <code class="docutils literal notranslate"><span class="pre">ILSVRC2012_img_val.tar</span></code> based on <code class="docutils literal notranslate"><span class="pre">split</span></code> in the root directory.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the ImageNet Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">train</span></code>, or <code class="docutils literal notranslate"><span class="pre">val</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>loader</strong> – A function to load an image given its path.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.ImageNet.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.ImageNet.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(sample, target) where target is class_index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.KMNIST.html" class="btn btn-neutral float-right" title="KMNIST" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.INaturalist.html" class="btn btn-neutral" title="INaturalist" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ImageNet</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.InStereo2k.html b/0.16/generated/torchvision.datasets.InStereo2k.html
new file mode 100644
index 00000000000..cc55fb535f3
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.InStereo2k.html
@@ -0,0 +1,750 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>InStereo2k &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ETH3DStereo" href="torchvision.datasets.ETH3DStereo.html" />
+    <link rel="prev" title="SintelStereo" href="torchvision.datasets.SintelStereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>InStereo2k</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.InStereo2k.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="instereo2k">
+<h1>InStereo2k<a class="headerlink" href="#instereo2k" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.InStereo2k">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">InStereo2k</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#InStereo2k"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.InStereo2k" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/YuhuaXu/StereoDataset">InStereo2k</a> dataset.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">InStereo2k</span>
+        <span class="n">train</span>
+            <span class="n">scene1</span>
+                <span class="n">left</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">right</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">left_disp</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">right_disp</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">scene2</span>
+            <span class="o">...</span>
+        <span class="n">test</span>
+            <span class="n">scene1</span>
+                <span class="n">left</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">right</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">left_disp</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">right_disp</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">scene2</span>
+            <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where InStereo2k is located.</p></li>
+<li><p><strong>split</strong> (<em>string</em>) – Either “train” or “test”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.InStereo2k.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#InStereo2k.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.InStereo2k.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 3-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+If a <code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is generated within the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter,
+a 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code> is returned.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.ETH3DStereo.html" class="btn btn-neutral float-right" title="ETH3DStereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.SintelStereo.html" class="btn btn-neutral" title="SintelStereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">InStereo2k</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.KMNIST.html b/0.16/generated/torchvision.datasets.KMNIST.html
new file mode 100644
index 00000000000..824b1e71a55
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.KMNIST.html
@@ -0,0 +1,731 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>KMNIST &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="LFWPeople" href="torchvision.datasets.LFWPeople.html" />
+    <link rel="prev" title="ImageNet" href="torchvision.datasets.ImageNet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>KMNIST</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.KMNIST.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="kmnist">
+<h1>KMNIST<a class="headerlink" href="#kmnist" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.KMNIST">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">KMNIST</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/mnist.html#KMNIST"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.KMNIST" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/rois-codh/kmnist">Kuzushiji-MNIST</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where <code class="docutils literal notranslate"><span class="pre">KMNIST/raw/train-images-idx3-ubyte</span></code>
+and  <code class="docutils literal notranslate"><span class="pre">KMNIST/raw/t10k-images-idx3-ubyte</span></code> exist.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, creates dataset from <code class="docutils literal notranslate"><span class="pre">train-images-idx3-ubyte</span></code>,
+otherwise from <code class="docutils literal notranslate"><span class="pre">t10k-images-idx3-ubyte</span></code>.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.KMNIST.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#torchvision.datasets.KMNIST.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.LFWPeople.html" class="btn btn-neutral float-right" title="LFWPeople" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.ImageNet.html" class="btn btn-neutral" title="ImageNet" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">KMNIST</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Kinetics.html b/0.16/generated/torchvision.datasets.Kinetics.html
new file mode 100644
index 00000000000..16ad9323afb
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Kinetics.html
@@ -0,0 +1,778 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Kinetics &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="UCF101" href="torchvision.datasets.UCF101.html" />
+    <link rel="prev" title="HMDB51" href="torchvision.datasets.HMDB51.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Kinetics</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Kinetics.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="kinetics">
+<h1>Kinetics<a class="headerlink" href="#kinetics" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Kinetics">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Kinetics</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">frames_per_clip</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'400'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">frame_rate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">step_between_clips</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">extensions</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">('avi',</span> <span class="pre">'mp4')</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_download_workers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_workers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_precomputed_metadata</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_min_dimension</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_audio_samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_audio_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_legacy</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'TCHW'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/kinetics.html#Kinetics"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kinetics" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.deepmind.com/open-source/kinetics">Generic Kinetics</a>
+dataset.</p>
+<p>Kinetics-400/600/700 are action recognition video datasets.
+This dataset consider every video as a collection of video clips of fixed size, specified
+by <code class="docutils literal notranslate"><span class="pre">frames_per_clip</span></code>, where the step in frames between each clip is given by
+<code class="docutils literal notranslate"><span class="pre">step_between_clips</span></code>.</p>
+<p>To give an example, for 2 videos with 10 and 15 frames respectively, if <code class="docutils literal notranslate"><span class="pre">frames_per_clip=5</span></code>
+and <code class="docutils literal notranslate"><span class="pre">step_between_clips=5</span></code>, the dataset size will be (2 + 3) = 5, where the first two
+elements will come from video 1, and the next three elements from video 2.
+Note that we drop clips which do not have exactly <code class="docutils literal notranslate"><span class="pre">frames_per_clip</span></code> elements, so not all
+frames in a video might be present.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – <p>Root directory of the Kinetics Dataset.
+Directory should be structured as follows:
+.. code:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>root/
+├── split
+│   ├──  class1
+│   │   ├──  clip1.mp4
+│   │   ├──  clip2.mp4
+│   │   ├──  clip3.mp4
+│   │   ├──  ...
+│   ├──  class2
+│   │   ├──   clipx.mp4
+│   │    └── ...
+</pre></div>
+</div>
+<p>Note: split is appended automatically using the split argument.</p>
+</p></li>
+<li><p><strong>frames_per_clip</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of frames in a clip</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – select between Kinetics-400 (default), Kinetics-600, and Kinetics-700</p></li>
+<li><p><strong>split</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – split of the dataset to consider; supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default) <code class="docutils literal notranslate"><span class="pre">&quot;val&quot;</span></code> <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code></p></li>
+<li><p><strong>frame_rate</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – If omitted, interpolate different frame rate for each clip.</p></li>
+<li><p><strong>step_between_clips</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of frames between each clip</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in a TxHxWxC video
+and returns a transformed version.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – Download the official version of the dataset to root folder.</p></li>
+<li><p><strong>num_workers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Use multiple workers for VideoClips creation</p></li>
+<li><p><strong>num_download_workers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Use multiprocessing in order to speed up download.</p></li>
+<li><p><strong>output_format</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – The format of the output video tensors (before transforms).
+Can be either “THWC” or “TCHW” (default).
+Note that in most other utils and datasets, the default is actually “THWC”.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><p>A 3-tuple with the following entries:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>video (Tensor[T, C, H, W] or Tensor[T, H, W, C]): the <cite>T</cite> video frames in torch.uint8 tensor</p></li>
+<li><p>audio(Tensor[K, L]): the audio frames, where <cite>K</cite> is the number of channels
+and <cite>L</cite> is the number of points in torch.float tensor</p></li>
+<li><p>label (int): class of the video clip</p></li>
+</ul>
+</div></blockquote>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+<dt class="field-even">Raises<span class="colon">:</span></dt>
+<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#RuntimeError" title="(in Python v3.12)"><strong>RuntimeError</strong></a> – If <code class="docutils literal notranslate"><span class="pre">download</span> <span class="pre">is</span> <span class="pre">True</span></code> and the video archives are already extracted.</p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Kinetics.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/kinetics.html#Kinetics.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kinetics.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.UCF101.html" class="btn btn-neutral float-right" title="UCF101" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.HMDB51.html" class="btn btn-neutral" title="HMDB51" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Kinetics</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Kitti.html b/0.16/generated/torchvision.datasets.Kitti.html
new file mode 100644
index 00000000000..a141be050c0
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Kitti.html
@@ -0,0 +1,758 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Kitti &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SBDataset" href="torchvision.datasets.SBDataset.html" />
+    <link rel="prev" title="Cityscapes" href="torchvision.datasets.Cityscapes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Kitti</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Kitti.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="kitti">
+<h1>Kitti<a class="headerlink" href="#kitti" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Kitti">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Kitti</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/kitti.html#Kitti"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kitti" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark">KITTI</a> Dataset.</p>
+<p>It corresponds to the “left color images of object” dataset, for object detection.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – <p>Root directory where images are downloaded to.
+Expects the following folder structure if download=False:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;root&gt;
+    └── Kitti
+        └─ raw
+            ├── training
+            |   ├── image_2
+            |   └── label_2
+            └── testing
+                └── image_2
+</pre></div>
+</div>
+</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Use <code class="docutils literal notranslate"><span class="pre">train</span></code> split if true, else <code class="docutils literal notranslate"><span class="pre">test</span></code> split.
+Defaults to <code class="docutils literal notranslate"><span class="pre">train</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.PILToTensor</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes input sample
+and its target as entry and returns a transformed version.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Kitti.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/kitti.html#Kitti.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kitti.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get item at a given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><p>(image, target), where
+target is a list of dictionaries with the following keys:</p>
+<ul class="simple">
+<li><p>type: str</p></li>
+<li><p>truncated: float</p></li>
+<li><p>occluded: int</p></li>
+<li><p>alpha: float</p></li>
+<li><p>bbox: float[4]</p></li>
+<li><p>dimensions: float[3]</p></li>
+<li><p>locations: float[3]</p></li>
+<li><p>rotation_y: float</p></li>
+</ul>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.SBDataset.html" class="btn btn-neutral float-right" title="SBDataset" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Cityscapes.html" class="btn btn-neutral" title="Cityscapes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Kitti</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Kitti2012Stereo.html b/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
new file mode 100644
index 00000000000..db8cb559a03
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
@@ -0,0 +1,757 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Kitti2012Stereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Kitti2015Stereo" href="torchvision.datasets.Kitti2015Stereo.html" />
+    <link rel="prev" title="CarlaStereo" href="torchvision.datasets.CarlaStereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Kitti2012Stereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Kitti2012Stereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="kitti2012stereo">
+<h1>Kitti2012Stereo<a class="headerlink" href="#kitti2012stereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Kitti2012Stereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Kitti2012Stereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#Kitti2012Stereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kitti2012Stereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>KITTI dataset from the <a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_stereo_flow.php">2012 stereo evaluation benchmark</a>.
+Uses the RGB images for consistency with KITTI 2015.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">Kitti2012</span>
+        <span class="n">testing</span>
+            <span class="n">colored_0</span>
+                <span class="mf">1_10.</span><span class="n">png</span>
+                <span class="mf">2_10.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">colored_1</span>
+                <span class="mf">1_10.</span><span class="n">png</span>
+                <span class="mf">2_10.</span><span class="n">png</span>
+                <span class="o">...</span>
+        <span class="n">training</span>
+            <span class="n">colored_0</span>
+                <span class="mf">1_10.</span><span class="n">png</span>
+                <span class="mf">2_10.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">colored_1</span>
+                <span class="mf">1_10.</span><span class="n">png</span>
+                <span class="mf">2_10.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">disp_noc</span>
+                <span class="mf">1.</span><span class="n">png</span>
+                <span class="mf">2.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">calib</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where <cite>Kitti2012</cite> is located.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split of scenes, either “train” (default) or “test”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Kitti2012Stereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#Kitti2012Stereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kitti2012Stereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is implicitly <code class="docutils literal notranslate"><span class="pre">None</span></code> if the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter does not
+generate a valid mask.
+Both <code class="docutils literal notranslate"><span class="pre">disparity</span></code> and <code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> are <code class="docutils literal notranslate"><span class="pre">None</span></code> if the dataset split is test.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Kitti2015Stereo.html" class="btn btn-neutral float-right" title="Kitti2015Stereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.CarlaStereo.html" class="btn btn-neutral" title="CarlaStereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Kitti2012Stereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Kitti2015Stereo.html b/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
new file mode 100644
index 00000000000..0240118af6f
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
@@ -0,0 +1,760 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Kitti2015Stereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CREStereo" href="torchvision.datasets.CREStereo.html" />
+    <link rel="prev" title="Kitti2012Stereo" href="torchvision.datasets.Kitti2012Stereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Kitti2015Stereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Kitti2015Stereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="kitti2015stereo">
+<h1>Kitti2015Stereo<a class="headerlink" href="#kitti2015stereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Kitti2015Stereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Kitti2015Stereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#Kitti2015Stereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kitti2015Stereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>KITTI dataset from the <a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_scene_flow.php">2015 stereo evaluation benchmark</a>.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">Kitti2015</span>
+        <span class="n">testing</span>
+            <span class="n">image_2</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">image_3</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+        <span class="n">training</span>
+            <span class="n">image_2</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">image_3</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">disp_occ_0</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">disp_occ_1</span>
+                <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+            <span class="n">calib</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where <cite>Kitti2015</cite> is located.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split of scenes, either “train” (default) or “test”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Kitti2015Stereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#Kitti2015Stereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Kitti2015Stereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is implicitly <code class="docutils literal notranslate"><span class="pre">None</span></code> if the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter does not
+generate a valid mask.
+Both <code class="docutils literal notranslate"><span class="pre">disparity</span></code> and <code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> are <code class="docutils literal notranslate"><span class="pre">None</span></code> if the dataset split is test.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.CREStereo.html" class="btn btn-neutral float-right" title="CREStereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Kitti2012Stereo.html" class="btn btn-neutral" title="Kitti2012Stereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Kitti2015Stereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.KittiFlow.html b/0.16/generated/torchvision.datasets.KittiFlow.html
new file mode 100644
index 00000000000..ffff8d80473
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.KittiFlow.html
@@ -0,0 +1,739 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>KittiFlow &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Sintel" href="torchvision.datasets.Sintel.html" />
+    <link rel="prev" title="HD1K" href="torchvision.datasets.HD1K.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>KittiFlow</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.KittiFlow.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="kittiflow">
+<h1>KittiFlow<a class="headerlink" href="#kittiflow" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.KittiFlow">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">KittiFlow</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#KittiFlow"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.KittiFlow" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://www.cvlibs.net/datasets/kitti/eval_scene_flow.php?benchmark=flow">KITTI</a> dataset for optical flow (2015).</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">KittiFlow</span>
+        <span class="n">testing</span>
+            <span class="n">image_2</span>
+        <span class="n">training</span>
+            <span class="n">image_2</span>
+            <span class="n">flow_occ</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the KittiFlow Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, either “train” (default) or “test”</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in
+<code class="docutils literal notranslate"><span class="pre">img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask</span></code> and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.KittiFlow.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#KittiFlow.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.KittiFlow.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask)</span></code>
+where <code class="docutils literal notranslate"><span class="pre">valid_flow_mask</span></code> is a numpy boolean mask of shape (H, W)
+indicating which flow values are valid. The flow is a numpy array of
+shape (2, H, W) and the images are PIL images. <code class="docutils literal notranslate"><span class="pre">flow</span></code> and <code class="docutils literal notranslate"><span class="pre">valid_flow_mask</span></code> are None if
+<code class="docutils literal notranslate"><span class="pre">split=&quot;test&quot;</span></code>.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Sintel.html" class="btn btn-neutral float-right" title="Sintel" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.HD1K.html" class="btn btn-neutral" title="HD1K" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">KittiFlow</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.LFWPairs.html b/0.16/generated/torchvision.datasets.LFWPairs.html
new file mode 100644
index 00000000000..77953b0ad7d
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.LFWPairs.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>LFWPairs &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="PhotoTour" href="torchvision.datasets.PhotoTour.html" />
+    <link rel="prev" title="Middlebury2014Stereo" href="torchvision.datasets.Middlebury2014Stereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>LFWPairs</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.LFWPairs.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lfwpairs">
+<h1>LFWPairs<a class="headerlink" href="#lfwpairs" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.LFWPairs">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">LFWPairs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'10fold'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_set</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'funneled'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/lfw.html#LFWPairs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.LFWPairs" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://vis-www.cs.umass.edu/lfw/">LFW</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">lfw-py</span></code> exists or will be saved to if download is set to True.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The image split to use. Can be one of <code class="docutils literal notranslate"><span class="pre">train</span></code>, <code class="docutils literal notranslate"><span class="pre">test</span></code>,
+<code class="docutils literal notranslate"><span class="pre">10fold</span></code>. Defaults to <code class="docutils literal notranslate"><span class="pre">10fold</span></code>.</p></li>
+<li><p><strong>image_set</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – Type of image funneling to use, <code class="docutils literal notranslate"><span class="pre">original</span></code>, <code class="docutils literal notranslate"><span class="pre">funneled</span></code> or
+<code class="docutils literal notranslate"><span class="pre">deepfunneled</span></code>. Defaults to <code class="docutils literal notranslate"><span class="pre">funneled</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomRotation</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.LFWPairs.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/lfw.html#LFWPairs.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.LFWPairs.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image1, image2, target) where target is <cite>0</cite> for different indentities and <cite>1</cite> for same identities.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.PhotoTour.html" class="btn btn-neutral float-right" title="PhotoTour" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Middlebury2014Stereo.html" class="btn btn-neutral" title="Middlebury2014Stereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">LFWPairs</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.LFWPeople.html b/0.16/generated/torchvision.datasets.LFWPeople.html
new file mode 100644
index 00000000000..b7c4930198f
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.LFWPeople.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>LFWPeople &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="LSUN" href="torchvision.datasets.LSUN.html" />
+    <link rel="prev" title="KMNIST" href="torchvision.datasets.KMNIST.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>LFWPeople</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.LFWPeople.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lfwpeople">
+<h1>LFWPeople<a class="headerlink" href="#lfwpeople" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.LFWPeople">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">LFWPeople</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'10fold'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_set</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'funneled'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/lfw.html#LFWPeople"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.LFWPeople" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://vis-www.cs.umass.edu/lfw/">LFW</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">lfw-py</span></code> exists or will be saved to if download is set to True.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The image split to use. Can be one of <code class="docutils literal notranslate"><span class="pre">train</span></code>, <code class="docutils literal notranslate"><span class="pre">test</span></code>,
+<code class="docutils literal notranslate"><span class="pre">10fold</span></code> (default).</p></li>
+<li><p><strong>image_set</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – Type of image funneling to use, <code class="docutils literal notranslate"><span class="pre">original</span></code>, <code class="docutils literal notranslate"><span class="pre">funneled</span></code> or
+<code class="docutils literal notranslate"><span class="pre">deepfunneled</span></code>. Defaults to <code class="docutils literal notranslate"><span class="pre">funneled</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomRotation</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.LFWPeople.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/lfw.html#LFWPeople.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.LFWPeople.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Tuple (image, target) where target is the identity of the person.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.LSUN.html" class="btn btn-neutral float-right" title="LSUN" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.KMNIST.html" class="btn btn-neutral" title="KMNIST" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">LFWPeople</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.LSUN.html b/0.16/generated/torchvision.datasets.LSUN.html
new file mode 100644
index 00000000000..3a3a10d6a67
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.LSUN.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>LSUN &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="MNIST" href="torchvision.datasets.MNIST.html" />
+    <link rel="prev" title="LFWPeople" href="torchvision.datasets.LFWPeople.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>LSUN</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.LSUN.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lsun">
+<h1>LSUN<a class="headerlink" href="#lsun" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.LSUN">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">LSUN</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/lsun.html#LSUN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.LSUN" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.yf.io/p/lsun">LSUN</a> dataset.</p>
+<p>You will need to install the <code class="docutils literal notranslate"><span class="pre">lmdb</span></code> package to use this dataset: run
+<code class="docutils literal notranslate"><span class="pre">pip</span> <span class="pre">install</span> <span class="pre">lmdb</span></code></p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory for the database files.</p></li>
+<li><p><strong>classes</strong> (<em>string</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a>) – One of {‘train’, ‘val’, ‘test’} or a list of
+categories to load. e,g. [‘bedroom_train’, ‘church_outdoor_train’].</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.LSUN.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/lsun.html#LSUN.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.LSUN.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Tuple (image, target) where target is the index of the target category.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.MNIST.html" class="btn btn-neutral float-right" title="MNIST" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.LFWPeople.html" class="btn btn-neutral" title="LFWPeople" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">LSUN</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.MNIST.html b/0.16/generated/torchvision.datasets.MNIST.html
new file mode 100644
index 00000000000..2614af13161
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.MNIST.html
@@ -0,0 +1,731 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MNIST &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Omniglot" href="torchvision.datasets.Omniglot.html" />
+    <link rel="prev" title="LSUN" href="torchvision.datasets.LSUN.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>MNIST</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.MNIST.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mnist">
+<h1>MNIST<a class="headerlink" href="#mnist" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.MNIST">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">MNIST</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/mnist.html#MNIST"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.MNIST" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://yann.lecun.com/exdb/mnist/">MNIST</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where <code class="docutils literal notranslate"><span class="pre">MNIST/raw/train-images-idx3-ubyte</span></code>
+and  <code class="docutils literal notranslate"><span class="pre">MNIST/raw/t10k-images-idx3-ubyte</span></code> exist.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, creates dataset from <code class="docutils literal notranslate"><span class="pre">train-images-idx3-ubyte</span></code>,
+otherwise from <code class="docutils literal notranslate"><span class="pre">t10k-images-idx3-ubyte</span></code>.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.MNIST.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/mnist.html#MNIST.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.MNIST.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Omniglot.html" class="btn btn-neutral float-right" title="Omniglot" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.LSUN.html" class="btn btn-neutral" title="LSUN" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MNIST</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html b/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
new file mode 100644
index 00000000000..8d93d3c043e
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
@@ -0,0 +1,774 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Middlebury2014Stereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="LFWPairs" href="torchvision.datasets.LFWPairs.html" />
+    <link rel="prev" title="ETH3DStereo" href="torchvision.datasets.ETH3DStereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Middlebury2014Stereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Middlebury2014Stereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="middlebury2014stereo">
+<h1>Middlebury2014Stereo<a class="headerlink" href="#middlebury2014stereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Middlebury2014Stereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Middlebury2014Stereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">calibration</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'perfect'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_ambient_views</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#Middlebury2014Stereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Middlebury2014Stereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>Publicly available scenes from the Middlebury dataset <cite>2014 version &lt;https://vision.middlebury.edu/stereo/data/scenes2014/&gt;</cite>.</p>
+<p>The dataset mostly follows the original format, without containing the ambient subdirectories.  :</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">Middlebury2014</span>
+        <span class="n">train</span>
+            <span class="n">scene1</span><span class="o">-</span><span class="p">{</span><span class="n">perfect</span><span class="p">,</span><span class="n">imperfect</span><span class="p">}</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">im</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im1E</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im1L</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">-</span><span class="n">n</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">-</span><span class="n">sd</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="n">y</span><span class="o">.</span><span class="n">pfm</span>
+            <span class="n">scene2</span><span class="o">-</span><span class="p">{</span><span class="n">perfect</span><span class="p">,</span><span class="n">imperfect</span><span class="p">}</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">im</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im1E</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im1L</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">-</span><span class="n">n</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">-</span><span class="n">sd</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="n">y</span><span class="o">.</span><span class="n">pfm</span>
+            <span class="o">...</span>
+        <span class="n">additional</span>
+            <span class="n">scene1</span><span class="o">-</span><span class="p">{</span><span class="n">perfect</span><span class="p">,</span><span class="n">imperfect</span><span class="p">}</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">im</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im1E</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">im1L</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">-</span><span class="n">n</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">-</span><span class="n">sd</span><span class="o">.</span><span class="n">pfm</span>
+                <span class="n">disp</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="n">y</span><span class="o">.</span><span class="n">pfm</span>
+            <span class="o">...</span>
+        <span class="n">test</span>
+            <span class="n">scene1</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">im</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">png</span>
+            <span class="n">scene2</span>
+                <span class="n">calib</span><span class="o">.</span><span class="n">txt</span>
+                <span class="n">im</span><span class="p">{</span><span class="mi">0</span><span class="p">,</span><span class="mi">1</span><span class="p">}</span><span class="o">.</span><span class="n">png</span>
+            <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the Middleburry 2014 Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split of scenes, either “train” (default), “test”, or “additional”</p></li>
+<li><p><strong>use_ambient_views</strong> (<em>boolean</em><em>, </em><em>optional</em>) – Whether to use different expose or lightning views when possible.
+The dataset samples with equal probability between <code class="docutils literal notranslate"><span class="pre">[im1.png,</span> <span class="pre">im1E.png,</span> <span class="pre">im1L.png]</span></code>.</p></li>
+<li><p><strong>calibration</strong> (<em>string</em><em>, </em><em>optional</em>) – Whether or not to use the calibrated (default) or uncalibrated scenes.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+<li><p><strong>download</strong> (<em>boolean</em><em>, </em><em>optional</em>) – Whether or not to download the dataset in the <code class="docutils literal notranslate"><span class="pre">root</span></code> directory.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Middlebury2014Stereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#Middlebury2014Stereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Middlebury2014Stereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is implicitly <code class="docutils literal notranslate"><span class="pre">None</span></code> for <cite>split=test</cite>.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.LFWPairs.html" class="btn btn-neutral float-right" title="LFWPairs" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.ETH3DStereo.html" class="btn btn-neutral" title="ETH3DStereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Middlebury2014Stereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.MovingMNIST.html b/0.16/generated/torchvision.datasets.MovingMNIST.html
new file mode 100644
index 00000000000..cc6651d1e26
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.MovingMNIST.html
@@ -0,0 +1,731 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MovingMNIST &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="DatasetFolder" href="torchvision.datasets.DatasetFolder.html" />
+    <link rel="prev" title="UCF101" href="torchvision.datasets.UCF101.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>MovingMNIST</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.MovingMNIST.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="movingmnist">
+<h1>MovingMNIST<a class="headerlink" href="#movingmnist" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.MovingMNIST">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">MovingMNIST</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">split_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">10</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/moving_mnist.html#MovingMNIST"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.MovingMNIST" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://www.cs.toronto.edu/~nitish/unsupervised_video/">MovingMNIST</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where <code class="docutils literal notranslate"><span class="pre">MovingMNIST/mnist_test_seq.npy</span></code> exists.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">None</span></code> (default), <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.
+If <code class="docutils literal notranslate"><span class="pre">split=None</span></code>, the full data is returned.</p></li>
+<li><p><strong>split_ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The split ratio of number of frames. If <code class="docutils literal notranslate"><span class="pre">split=&quot;train&quot;</span></code>, the first split
+frames <code class="docutils literal notranslate"><span class="pre">data[:,</span> <span class="pre">:split_ratio]</span></code> is returned. If <code class="docutils literal notranslate"><span class="pre">split=&quot;test&quot;</span></code>, the last split frames <code class="docutils literal notranslate"><span class="pre">data[:,</span> <span class="pre">split_ratio:]</span></code>
+is returned. If <code class="docutils literal notranslate"><span class="pre">split=None</span></code>, this parameter is ignored and the all frames data is returned.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in an torch Tensor
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.MovingMNIST.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/datasets/moving_mnist.html#MovingMNIST.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.MovingMNIST.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Video frames (torch Tensor[T, C, H, W]). The <cite>T</cite> is the number of frames.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)">torch.Tensor</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.DatasetFolder.html" class="btn btn-neutral float-right" title="DatasetFolder" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.UCF101.html" class="btn btn-neutral" title="UCF101" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MovingMNIST</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Omniglot.html b/0.16/generated/torchvision.datasets.Omniglot.html
new file mode 100644
index 00000000000..dfcd7049ff0
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Omniglot.html
@@ -0,0 +1,731 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Omniglot &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="OxfordIIITPet" href="torchvision.datasets.OxfordIIITPet.html" />
+    <link rel="prev" title="MNIST" href="torchvision.datasets.MNIST.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Omniglot</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Omniglot.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="omniglot">
+<h1>Omniglot<a class="headerlink" href="#omniglot" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Omniglot">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Omniglot</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">background</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/omniglot.html#Omniglot"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Omniglot" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/brendenlake/omniglot">Omniglot</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">omniglot-py</span></code> exists.</p></li>
+<li><p><strong>background</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, creates dataset from the “background” set, otherwise
+creates from the “evaluation” set. This terminology is defined by the authors.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset zip files from the internet and
+puts it in root directory. If the zip files are already downloaded, they are not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Omniglot.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/omniglot.html#Omniglot.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Omniglot.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target character class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.OxfordIIITPet.html" class="btn btn-neutral float-right" title="OxfordIIITPet" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.MNIST.html" class="btn btn-neutral" title="MNIST" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Omniglot</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.OxfordIIITPet.html b/0.16/generated/torchvision.datasets.OxfordIIITPet.html
new file mode 100644
index 00000000000..6c570848ba4
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.OxfordIIITPet.html
@@ -0,0 +1,737 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>OxfordIIITPet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Places365" href="torchvision.datasets.Places365.html" />
+    <link rel="prev" title="Omniglot" href="torchvision.datasets.Omniglot.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>OxfordIIITPet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.OxfordIIITPet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="oxfordiiitpet">
+<h1>OxfordIIITPet<a class="headerlink" href="#oxfordiiitpet" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.OxfordIIITPet">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">OxfordIIITPet</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'trainval'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_types</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'category'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/oxford_iiit_pet.html#OxfordIIITPet"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.OxfordIIITPet" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.robots.ox.ac.uk/~vgg/data/pets/">Oxford-IIIT Pet Dataset</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;trainval&quot;</span></code> (default) or <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>target_types</strong> (<em>string</em><em>, </em><em>sequence of strings</em><em>, </em><em>optional</em>) – <p>Types of target to use. Can be <code class="docutils literal notranslate"><span class="pre">category</span></code> (default) or
+<code class="docutils literal notranslate"><span class="pre">segmentation</span></code>. Can also be a list to output a tuple with all specified target types. The types represent:</p>
+<blockquote>
+<div><ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">category</span></code> (int): Label for one of the 37 pet categories.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">segmentation</span></code> (PIL image): Segmentation trimap of the image.</p></li>
+</ul>
+</div></blockquote>
+<p>If empty, <code class="docutils literal notranslate"><span class="pre">None</span></code> will be returned as target.</p>
+</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in a PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and puts it into
+<code class="docutils literal notranslate"><span class="pre">root/oxford-iiit-pet</span></code>. If dataset is already downloaded, it is not downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.OxfordIIITPet.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/oxford_iiit_pet.html#OxfordIIITPet.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.OxfordIIITPet.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.Places365.html" class="btn btn-neutral float-right" title="Places365" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Omniglot.html" class="btn btn-neutral" title="Omniglot" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">OxfordIIITPet</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.PCAM.html b/0.16/generated/torchvision.datasets.PCAM.html
new file mode 100644
index 00000000000..964a4600ca4
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.PCAM.html
@@ -0,0 +1,732 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>PCAM &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="QMNIST" href="torchvision.datasets.QMNIST.html" />
+    <link rel="prev" title="Places365" href="torchvision.datasets.Places365.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>PCAM</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.PCAM.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="pcam">
+<h1>PCAM<a class="headerlink" href="#pcam" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.PCAM">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">PCAM</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/pcam.html#PCAM"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.PCAM" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/basveeling/pcam">PCAM Dataset</a>.</p>
+<p>The PatchCamelyon dataset is a binary classification dataset with 327,680
+color images (96px x 96px), extracted from histopathologic scans of lymph node
+sections. Each image is annotated with a binary label indicating presence of
+metastatic tissue.</p>
+<p>This dataset requires the <code class="docutils literal notranslate"><span class="pre">h5py</span></code> package which you can install with <code class="docutils literal notranslate"><span class="pre">pip</span> <span class="pre">install</span> <span class="pre">h5py</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default), <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;val&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in a PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and puts it into <code class="docutils literal notranslate"><span class="pre">root/pcam</span></code>. If
+dataset is already downloaded, it is not downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.PCAM.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/pcam.html#PCAM.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.PCAM.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.QMNIST.html" class="btn btn-neutral float-right" title="QMNIST" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Places365.html" class="btn btn-neutral" title="Places365" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">PCAM</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.PhotoTour.html b/0.16/generated/torchvision.datasets.PhotoTour.html
new file mode 100644
index 00000000000..3851a5e8d11
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.PhotoTour.html
@@ -0,0 +1,737 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>PhotoTour &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CocoCaptions" href="torchvision.datasets.CocoCaptions.html" />
+    <link rel="prev" title="LFWPairs" href="torchvision.datasets.LFWPairs.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>PhotoTour</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.PhotoTour.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="phototour">
+<h1>PhotoTour<a class="headerlink" href="#phototour" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.PhotoTour">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">PhotoTour</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/phototour.html#PhotoTour"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.PhotoTour" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://matthewalunbrown.com/patchdata/patchdata.html">Multi-view Stereo Correspondence</a> Dataset.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>We only provide the newer version of the dataset, since the authors state that it</p>
+<blockquote>
+<div><p>is more suitable for training descriptors based on difference of Gaussian, or Harris corners, as the
+patches are centred on real interest point detections, rather than being projections of 3D points as is the
+case in the old dataset.</p>
+</div></blockquote>
+<p>The original dataset is available under <a class="reference external" href="http://phototour.cs.washington.edu/patches/default.htm">http://phototour.cs.washington.edu/patches/default.htm</a>.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where images are.</p></li>
+<li><p><strong>name</strong> (<em>string</em>) – Name of the dataset to load.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.PhotoTour.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/phototour.html#PhotoTour.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.PhotoTour.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(data1, data2, matches)</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.CocoCaptions.html" class="btn btn-neutral float-right" title="CocoCaptions" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.LFWPairs.html" class="btn btn-neutral" title="LFWPairs" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">PhotoTour</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Places365.html b/0.16/generated/torchvision.datasets.Places365.html
new file mode 100644
index 00000000000..8d44bf41456
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Places365.html
@@ -0,0 +1,738 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Places365 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="PCAM" href="torchvision.datasets.PCAM.html" />
+    <link rel="prev" title="OxfordIIITPet" href="torchvision.datasets.OxfordIIITPet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Places365</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Places365.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="places365">
+<h1>Places365<a class="headerlink" href="#places365" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Places365">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Places365</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">root:</span> <span class="pre">str,</span> <span class="pre">split:</span> <span class="pre">str</span> <span class="pre">=</span> <span class="pre">'train-standard',</span> <span class="pre">small:</span> <span class="pre">bool</span> <span class="pre">=</span> <span class="pre">False,</span> <span class="pre">download:</span> <span class="pre">bool</span> <span class="pre">=</span> <span class="pre">False,</span> <span class="pre">transform:</span> <span class="pre">~typing.Optional[~typing.Callable]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">target_transform:</span> <span class="pre">~typing.Optional[~typing.Callable]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">loader:</span> <span class="pre">~typing.Callable[[str],</span> <span class="pre">~typing.Any]</span> <span class="pre">=</span> <span class="pre">&lt;function</span> <span class="pre">default_loader&gt;</span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/places365.html#Places365"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Places365" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://places2.csail.mit.edu/index.html">Places365</a> classification dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the Places365 dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split. Can be one of <code class="docutils literal notranslate"><span class="pre">train-standard</span></code> (default), <code class="docutils literal notranslate"><span class="pre">train-challenge</span></code>,
+<code class="docutils literal notranslate"><span class="pre">val</span></code>.</p></li>
+<li><p><strong>small</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If <code class="docutils literal notranslate"><span class="pre">True</span></code>, uses the small images, i.e. resized to 256 x 256 pixels, instead of the
+high resolution ones.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If <code class="docutils literal notranslate"><span class="pre">True</span></code>, downloads the dataset components and places them in <code class="docutils literal notranslate"><span class="pre">root</span></code>. Already
+downloaded archives are not downloaded again.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>loader</strong> – A function to load an image given its path.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Raises<span class="colon">:</span></dt>
+<dd class="field-even"><ul class="simple">
+<li><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#RuntimeError" title="(in Python v3.12)"><strong>RuntimeError</strong></a> – If <code class="docutils literal notranslate"><span class="pre">download</span> <span class="pre">is</span> <span class="pre">False</span></code> and the meta files, i.e. the devkit, are not present or corrupted.</p></li>
+<li><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#RuntimeError" title="(in Python v3.12)"><strong>RuntimeError</strong></a> – If <code class="docutils literal notranslate"><span class="pre">download</span> <span class="pre">is</span> <span class="pre">True</span></code> and the image archive is already extracted.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Places365.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/places365.html#Places365.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Places365.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.PCAM.html" class="btn btn-neutral float-right" title="PCAM" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.OxfordIIITPet.html" class="btn btn-neutral" title="OxfordIIITPet" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Places365</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.QMNIST.html b/0.16/generated/torchvision.datasets.QMNIST.html
new file mode 100644
index 00000000000..66ac98e9170
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.QMNIST.html
@@ -0,0 +1,744 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>QMNIST &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RenderedSST2" href="torchvision.datasets.RenderedSST2.html" />
+    <link rel="prev" title="PCAM" href="torchvision.datasets.PCAM.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>QMNIST</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.QMNIST.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="qmnist">
+<h1>QMNIST<a class="headerlink" href="#qmnist" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.QMNIST">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">QMNIST</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">what</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">compat</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/mnist.html#QMNIST"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.QMNIST" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/facebookresearch/qmnist">QMNIST</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset whose <code class="docutils literal notranslate"><span class="pre">raw</span></code>
+subdir contains binary files of the datasets.</p></li>
+<li><p><strong>what</strong> (<em>string</em><em>,</em><em>optional</em>) – Can be ‘train’, ‘test’, ‘test10k’,
+‘test50k’, or ‘nist’ for respectively the mnist compatible
+training set, the 60k qmnist testing set, the 10k qmnist
+examples that match the mnist testing set, the 50k
+remaining qmnist testing examples, or all the nist
+digits. The default is to select ‘train’ or ‘test’
+according to the compatibility argument ‘train’.</p></li>
+<li><p><strong>compat</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>,</em><em>optional</em>) – A boolean that says whether the target
+for each example is class number (for compatibility with
+the MNIST dataloader) or a torch vector containing the
+full qmnist information. Default=True.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from
+the internet and puts it in root directory. If dataset is
+already downloaded, it is not downloaded again.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that
+takes in an PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform
+that takes in the target and transforms it.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>,</em><em>optional</em><em>,</em><em>compatibility</em>) – When argument ‘what’ is
+not specified, this boolean decides whether to load the
+training set or the testing set.  Default: True.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.QMNIST.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/mnist.html#QMNIST.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.QMNIST.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.RenderedSST2.html" class="btn btn-neutral float-right" title="RenderedSST2" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.PCAM.html" class="btn btn-neutral" title="PCAM" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">QMNIST</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.RenderedSST2.html b/0.16/generated/torchvision.datasets.RenderedSST2.html
new file mode 100644
index 00000000000..2d73bfb0ecb
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.RenderedSST2.html
@@ -0,0 +1,734 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RenderedSST2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SEMEION" href="torchvision.datasets.SEMEION.html" />
+    <link rel="prev" title="QMNIST" href="torchvision.datasets.QMNIST.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>RenderedSST2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.RenderedSST2.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="renderedsst2">
+<h1>RenderedSST2<a class="headerlink" href="#renderedsst2" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.RenderedSST2">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">RenderedSST2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/rendered_sst2.html#RenderedSST2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.RenderedSST2" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://github.com/openai/CLIP/blob/main/data/rendered-sst2.md">The Rendered SST2 Dataset</a>.</p>
+<p>Rendered SST2 is an image classification dataset used to evaluate the models capability on optical
+character recognition. This dataset was generated by rendering sentences in the Standford Sentiment
+Treebank v2 dataset.</p>
+<p>This dataset contains two classes (positive and negative) and is divided in three splits: a  train
+split containing 6920 images (3610 positive and 3310 negative), a validation split containing 872 images
+(444 positive and 428 negative), and a test split containing 1821 images (909 positive and 912 negative).</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default), <cite>“val”</cite> and <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again. Default is False.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.RenderedSST2.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/rendered_sst2.html#RenderedSST2.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.RenderedSST2.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.SEMEION.html" class="btn btn-neutral float-right" title="SEMEION" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.QMNIST.html" class="btn btn-neutral" title="QMNIST" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RenderedSST2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.SBDataset.html b/0.16/generated/torchvision.datasets.SBDataset.html
new file mode 100644
index 00000000000..25ebb5b28d3
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.SBDataset.html
@@ -0,0 +1,745 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SBDataset &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="VOCSegmentation" href="torchvision.datasets.VOCSegmentation.html" />
+    <link rel="prev" title="Kitti" href="torchvision.datasets.Kitti.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>SBDataset</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.SBDataset.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sbdataset">
+<h1>SBDataset<a class="headerlink" href="#sbdataset" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.SBDataset">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">SBDataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_set</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'boundaries'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/sbd.html#SBDataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SBDataset" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://home.bharathh.info/pubs/codes/SBD/download.html">Semantic Boundaries Dataset</a></p>
+<p>The SBD currently contains annotations from 11355 images taken from the PASCAL VOC 2011 dataset.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Please note that the train and val splits included with this dataset are different from
+the splits in the PASCAL VOC dataset. In particular some “train” images might be part of
+VOC2012 val.
+If you are interested in testing on VOC 2012 val, then use <cite>image_set=’train_noval’</cite>,
+which excludes all val images.</p>
+</div>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>This class needs <a class="reference external" href="https://docs.scipy.org/doc/">scipy</a> to load target files from <cite>.mat</cite> format.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the Semantic Boundaries Dataset</p></li>
+<li><p><strong>image_set</strong> (<em>string</em><em>, </em><em>optional</em>) – Select the image_set to use, <code class="docutils literal notranslate"><span class="pre">train</span></code>, <code class="docutils literal notranslate"><span class="pre">val</span></code> or <code class="docutils literal notranslate"><span class="pre">train_noval</span></code>.
+Image set <code class="docutils literal notranslate"><span class="pre">train_noval</span></code> excludes VOC 2012 val images.</p></li>
+<li><p><strong>mode</strong> (<em>string</em><em>, </em><em>optional</em>) – Select target type. Possible values ‘boundaries’ or ‘segmentation’.
+In case of ‘boundaries’, the target is an array of shape <cite>[num_classes, H, W]</cite>,
+where <cite>num_classes=20</cite>.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes input sample and its target as entry
+and returns a transformed version. Input sample is PIL image and target is a numpy array
+if <cite>mode=’boundaries’</cite> or PIL image if <cite>mode=’segmentation’</cite>.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.SBDataset.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/sbd.html#SBDataset.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SBDataset.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.VOCSegmentation.html" class="btn btn-neutral float-right" title="VOCSegmentation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Kitti.html" class="btn btn-neutral" title="Kitti" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SBDataset</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.SBU.html b/0.16/generated/torchvision.datasets.SBU.html
new file mode 100644
index 00000000000..82ed892acfd
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.SBU.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SBU &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="StanfordCars" href="torchvision.datasets.StanfordCars.html" />
+    <link rel="prev" title="SEMEION" href="torchvision.datasets.SEMEION.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>SBU</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.SBU.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sbu">
+<h1>SBU<a class="headerlink" href="#sbu" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.SBU">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">SBU</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/sbu.html#SBU"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SBU" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://www.cs.virginia.edu/~vicente/sbucaptions/">SBU Captioned Photo</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where tarball
+<code class="docutils literal notranslate"><span class="pre">SBUCaptionedPhotoDataset.tar.gz</span></code> exists.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.SBU.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/sbu.html#SBU.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SBU.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is a caption for the photo.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.StanfordCars.html" class="btn btn-neutral float-right" title="StanfordCars" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.SEMEION.html" class="btn btn-neutral" title="SEMEION" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SBU</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.SEMEION.html b/0.16/generated/torchvision.datasets.SEMEION.html
new file mode 100644
index 00000000000..1da59be1584
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.SEMEION.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SEMEION &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SBU" href="torchvision.datasets.SBU.html" />
+    <link rel="prev" title="RenderedSST2" href="torchvision.datasets.RenderedSST2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>SEMEION</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.SEMEION.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="semeion">
+<h1>SEMEION<a class="headerlink" href="#semeion" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.SEMEION">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">SEMEION</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/semeion.html#SEMEION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SEMEION" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://archive.ics.uci.edu/ml/datasets/semeion+handwritten+digit">SEMEION</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">semeion.py</span></code> exists.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.SEMEION.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/semeion.html#SEMEION.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SEMEION.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.SBU.html" class="btn btn-neutral float-right" title="SBU" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.RenderedSST2.html" class="btn btn-neutral" title="RenderedSST2" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SEMEION</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.STL10.html b/0.16/generated/torchvision.datasets.STL10.html
new file mode 100644
index 00000000000..2e353a32219
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.STL10.html
@@ -0,0 +1,734 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>STL10 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SUN397" href="torchvision.datasets.SUN397.html" />
+    <link rel="prev" title="StanfordCars" href="torchvision.datasets.StanfordCars.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>STL10</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.STL10.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="stl10">
+<h1>STL10<a class="headerlink" href="#stl10" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.STL10">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">STL10</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">folds</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/stl10.html#STL10"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.STL10" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://cs.stanford.edu/~acoates/stl10/">STL10</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset where directory
+<code class="docutils literal notranslate"><span class="pre">stl10_binary</span></code> exists.</p></li>
+<li><p><strong>split</strong> (<em>string</em>) – One of {‘train’, ‘test’, ‘unlabeled’, ‘train+unlabeled’}.
+Accordingly, dataset is selected.</p></li>
+<li><p><strong>folds</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – One of {0-9} or None.
+For training, loads one of the 10 pre-defined folds of 1k samples for the
+standard evaluation procedure. If no value is passed, loads the 5k samples.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.STL10.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/stl10.html#STL10.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.STL10.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.SUN397.html" class="btn btn-neutral float-right" title="SUN397" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.StanfordCars.html" class="btn btn-neutral" title="StanfordCars" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">STL10</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.SUN397.html b/0.16/generated/torchvision.datasets.SUN397.html
new file mode 100644
index 00000000000..6a3f04f5141
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.SUN397.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SUN397 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SVHN" href="torchvision.datasets.SVHN.html" />
+    <link rel="prev" title="STL10" href="torchvision.datasets.STL10.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>SUN397</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.SUN397.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sun397">
+<h1>SUN397<a class="headerlink" href="#sun397" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.SUN397">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">SUN397</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/sun397.html#SUN397"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SUN397" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://vision.princeton.edu/projects/2010/SUN/">The SUN397 Data Set</a>.</p>
+<p>The SUN397 or Scene UNderstanding (SUN) is a dataset for scene recognition consisting of
+397 categories with 108’754 images.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image and returns a transformed
+version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code>.</p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.SUN397.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/sun397.html#SUN397.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SUN397.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.SVHN.html" class="btn btn-neutral float-right" title="SVHN" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.STL10.html" class="btn btn-neutral" title="STL10" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SUN397</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.SVHN.html b/0.16/generated/torchvision.datasets.SVHN.html
new file mode 100644
index 00000000000..e00f97b40c3
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.SVHN.html
@@ -0,0 +1,737 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SVHN &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="USPS" href="torchvision.datasets.USPS.html" />
+    <link rel="prev" title="SUN397" href="torchvision.datasets.SUN397.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>SVHN</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.SVHN.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="svhn">
+<h1>SVHN<a class="headerlink" href="#svhn" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.SVHN">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">SVHN</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/svhn.html#SVHN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SVHN" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://ufldl.stanford.edu/housenumbers/">SVHN</a> Dataset.
+Note: The SVHN dataset assigns the label <cite>10</cite> to the digit <cite>0</cite>. However, in this Dataset,
+we assign the label <cite>0</cite> to the digit <cite>0</cite> to be compatible with PyTorch loss functions which
+expect the class labels to be in the range <cite>[0, C-1]</cite></p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>This class needs <a class="reference external" href="https://docs.scipy.org/doc/">scipy</a> to load data from <cite>.mat</cite> format.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the dataset where the data is stored.</p></li>
+<li><p><strong>split</strong> (<em>string</em>) – One of {‘train’, ‘test’, ‘extra’}.
+Accordingly dataset is selected. ‘extra’ is Extra training set.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.SVHN.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/svhn.html#SVHN.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SVHN.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.USPS.html" class="btn btn-neutral float-right" title="USPS" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.SUN397.html" class="btn btn-neutral" title="SUN397" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SVHN</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.SceneFlowStereo.html b/0.16/generated/torchvision.datasets.SceneFlowStereo.html
new file mode 100644
index 00000000000..91414dacec5
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.SceneFlowStereo.html
@@ -0,0 +1,771 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SceneFlowStereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SintelStereo" href="torchvision.datasets.SintelStereo.html" />
+    <link rel="prev" title="FallingThingsStereo" href="torchvision.datasets.FallingThingsStereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>SceneFlowStereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.SceneFlowStereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sceneflowstereo">
+<h1>SceneFlowStereo<a class="headerlink" href="#sceneflowstereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.SceneFlowStereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">SceneFlowStereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">variant</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'FlyingThings3D'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pass_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'clean'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#SceneFlowStereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SceneFlowStereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>Dataset interface for <a class="reference external" href="https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html">Scene Flow</a> datasets.
+This interface provides access to the <cite>FlyingThings3D, `Monkaa</cite> and <cite>Driving</cite> datasets.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">SceneFlow</span>
+        <span class="n">Monkaa</span>
+            <span class="n">frames_cleanpass</span>
+                <span class="n">scene1</span>
+                    <span class="n">left</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">right</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="n">scene2</span>
+                    <span class="n">left</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">right</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+            <span class="n">frames_finalpass</span>
+                <span class="n">scene1</span>
+                    <span class="n">left</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">right</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                <span class="o">...</span>
+                <span class="o">...</span>
+            <span class="n">disparity</span>
+                <span class="n">scene1</span>
+                    <span class="n">left</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">pfm</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">pfm</span>
+                    <span class="n">right</span>
+                        <span class="n">img1</span><span class="o">.</span><span class="n">pfm</span>
+                        <span class="n">img2</span><span class="o">.</span><span class="n">pfm</span>
+        <span class="n">FlyingThings3D</span>
+            <span class="o">...</span>
+            <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where SceneFlow is located.</p></li>
+<li><p><strong>variant</strong> (<em>string</em>) – Which dataset variant to user, “FlyingThings3D” (default), “Monkaa” or “Driving”.</p></li>
+<li><p><strong>pass_name</strong> (<em>string</em>) – Which pass to use, “clean” (default), “final” or “both”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.SceneFlowStereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#SceneFlowStereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SceneFlowStereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 3-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity)</span></code>.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images.
+If a <code class="docutils literal notranslate"><span class="pre">valid_mask</span></code> is generated within the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter,
+a 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code> is returned.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.SintelStereo.html" class="btn btn-neutral float-right" title="SintelStereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.FallingThingsStereo.html" class="btn btn-neutral" title="FallingThingsStereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SceneFlowStereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.Sintel.html b/0.16/generated/torchvision.datasets.Sintel.html
new file mode 100644
index 00000000000..ddf4d2be19c
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.Sintel.html
@@ -0,0 +1,760 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Sintel &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CarlaStereo" href="torchvision.datasets.CarlaStereo.html" />
+    <link rel="prev" title="KittiFlow" href="torchvision.datasets.KittiFlow.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>Sintel</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.Sintel.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sintel">
+<h1>Sintel<a class="headerlink" href="#sintel" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.Sintel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">Sintel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pass_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'clean'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#Sintel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Sintel" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://sintel.is.tue.mpg.de/">Sintel</a> Dataset for optical flow.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">Sintel</span>
+        <span class="n">testing</span>
+            <span class="n">clean</span>
+                <span class="n">scene_1</span>
+                <span class="n">scene_2</span>
+                <span class="o">...</span>
+            <span class="n">final</span>
+                <span class="n">scene_1</span>
+                <span class="n">scene_2</span>
+                <span class="o">...</span>
+        <span class="n">training</span>
+            <span class="n">clean</span>
+                <span class="n">scene_1</span>
+                <span class="n">scene_2</span>
+                <span class="o">...</span>
+            <span class="n">final</span>
+                <span class="n">scene_1</span>
+                <span class="n">scene_2</span>
+                <span class="o">...</span>
+            <span class="n">flow</span>
+                <span class="n">scene_1</span>
+                <span class="n">scene_2</span>
+                <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the Sintel Dataset.</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, either “train” (default) or “test”</p></li>
+<li><p><strong>pass_name</strong> (<em>string</em><em>, </em><em>optional</em>) – The pass to use, either “clean” (default), “final”, or “both”. See link above for
+details on the different passes.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in
+<code class="docutils literal notranslate"><span class="pre">img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask</span></code> and returns a transformed version.
+<code class="docutils literal notranslate"><span class="pre">valid_flow_mask</span></code> is expected for consistency with other datasets which
+return a built-in valid mask, such as <a class="reference internal" href="torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-class docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.Sintel.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_optical_flow.html#Sintel.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.Sintel.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 3-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow)</span></code>.
+The flow is a numpy array of shape (2, H, W) and the images are PIL images.
+<code class="docutils literal notranslate"><span class="pre">flow</span></code> is None if <code class="docutils literal notranslate"><span class="pre">split=&quot;test&quot;</span></code>.
+If a valid flow mask is generated within the <code class="docutils literal notranslate"><span class="pre">transforms</span></code> parameter,
+a 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img1,</span> <span class="pre">img2,</span> <span class="pre">flow,</span> <span class="pre">valid_flow_mask)</span></code> is returned.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.CarlaStereo.html" class="btn btn-neutral float-right" title="CarlaStereo" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.KittiFlow.html" class="btn btn-neutral" title="KittiFlow" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Sintel</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.SintelStereo.html b/0.16/generated/torchvision.datasets.SintelStereo.html
new file mode 100644
index 00000000000..fdcff746521
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.SintelStereo.html
@@ -0,0 +1,762 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SintelStereo &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="InStereo2k" href="torchvision.datasets.InStereo2k.html" />
+    <link rel="prev" title="SceneFlowStereo" href="torchvision.datasets.SceneFlowStereo.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>SintelStereo</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.SintelStereo.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sintelstereo">
+<h1>SintelStereo<a class="headerlink" href="#sintelstereo" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.SintelStereo">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">SintelStereo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">pass_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'final'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#SintelStereo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SintelStereo" title="Permalink to this definition">¶</a></dt>
+<dd><p>Sintel <a class="reference external" href="http://sintel.is.tue.mpg.de/stereo">Stereo Dataset</a>.</p>
+<p>The dataset is expected to have the following structure:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">root</span>
+    <span class="n">Sintel</span>
+        <span class="n">training</span>
+            <span class="n">final_left</span>
+                <span class="n">scene1</span>
+                    <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="o">...</span>
+                <span class="o">...</span>
+            <span class="n">final_right</span>
+                <span class="n">scene2</span>
+                    <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="o">...</span>
+                <span class="o">...</span>
+            <span class="n">disparities</span>
+                <span class="n">scene1</span>
+                    <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="o">...</span>
+                <span class="o">...</span>
+            <span class="n">occlusions</span>
+                <span class="n">scene1</span>
+                    <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="o">...</span>
+                <span class="o">...</span>
+            <span class="n">outofframe</span>
+                <span class="n">scene1</span>
+                    <span class="n">img1</span><span class="o">.</span><span class="n">png</span>
+                    <span class="n">img2</span><span class="o">.</span><span class="n">png</span>
+                    <span class="o">...</span>
+                <span class="o">...</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory where Sintel Stereo is located.</p></li>
+<li><p><strong>pass_name</strong> (<em>string</em>) – The name of the pass to use, either “final”, “clean” or “both”.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in a sample and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.SintelStereo.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/_stereo_matching.html#SintelStereo.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.SintelStereo.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return example at given index.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The index of the example to retrieve</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A 4-tuple with <code class="docutils literal notranslate"><span class="pre">(img_left,</span> <span class="pre">img_right,</span> <span class="pre">disparity,</span> <span class="pre">valid_mask)</span></code> is returned.
+The disparity is a numpy array of shape (1, H, W) and the images are PIL images whilst
+the valid_mask is a numpy array of shape (H, W).</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.InStereo2k.html" class="btn btn-neutral float-right" title="InStereo2k" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.SceneFlowStereo.html" class="btn btn-neutral" title="SceneFlowStereo" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SintelStereo</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.StanfordCars.html b/0.16/generated/torchvision.datasets.StanfordCars.html
new file mode 100644
index 00000000000..f049b422ab9
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.StanfordCars.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>StanfordCars &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="STL10" href="torchvision.datasets.STL10.html" />
+    <link rel="prev" title="SBU" href="torchvision.datasets.SBU.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>StanfordCars</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.StanfordCars.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="stanfordcars">
+<h1>StanfordCars<a class="headerlink" href="#stanfordcars" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.StanfordCars">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">StanfordCars</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/stanford_cars.html#StanfordCars"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.StanfordCars" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://ai.stanford.edu/~jkrause/cars/car_dataset.html">Stanford Cars</a> Dataset</p>
+<p>The Cars dataset contains 16,185 images of 196 classes of cars. The data is
+split into 8,144 training images and 8,041 testing images, where each class
+has been split roughly in a 50-50 split</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This class needs <a class="reference external" href="https://docs.scipy.org/doc/">scipy</a> to load target files from <cite>.mat</cite> format.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset</p></li>
+<li><p><strong>split</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset split, supports <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code> (default) or <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.StanfordCars.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/stanford_cars.html#StanfordCars.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.StanfordCars.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns pil_image and class_id for given index</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.STL10.html" class="btn btn-neutral float-right" title="STL10" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.SBU.html" class="btn btn-neutral" title="SBU" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">StanfordCars</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.UCF101.html b/0.16/generated/torchvision.datasets.UCF101.html
new file mode 100644
index 00000000000..7c93a606e06
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.UCF101.html
@@ -0,0 +1,758 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>UCF101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="MovingMNIST" href="torchvision.datasets.MovingMNIST.html" />
+    <link rel="prev" title="Kinetics" href="torchvision.datasets.Kinetics.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>UCF101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.UCF101.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ucf101">
+<h1>UCF101<a class="headerlink" href="#ucf101" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.UCF101">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">UCF101</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">annotation_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">frames_per_clip</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">step_between_clips</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">frame_rate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fold</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_precomputed_metadata</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_workers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_video_min_dimension</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_audio_samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'THWC'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/ucf101.html#UCF101"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.UCF101" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.crcv.ucf.edu/data/UCF101.php">UCF101</a> dataset.</p>
+<p>UCF101 is an action recognition video dataset.
+This dataset consider every video as a collection of video clips of fixed size, specified
+by <code class="docutils literal notranslate"><span class="pre">frames_per_clip</span></code>, where the step in frames between each clip is given by
+<code class="docutils literal notranslate"><span class="pre">step_between_clips</span></code>. The dataset itself can be downloaded from the dataset website;
+annotations that <code class="docutils literal notranslate"><span class="pre">annotation_path</span></code> should be pointing to can be downloaded from <a class="reference external" href="https://www.crcv.ucf.edu/data/UCF101/UCF101TrainTestSplits-RecognitionTask.zip">here</a>.</p>
+<p>To give an example, for 2 videos with 10 and 15 frames respectively, if <code class="docutils literal notranslate"><span class="pre">frames_per_clip=5</span></code>
+and <code class="docutils literal notranslate"><span class="pre">step_between_clips=5</span></code>, the dataset size will be (2 + 3) = 5, where the first two
+elements will come from video 1, and the next three elements from video 2.
+Note that we drop clips which do not have exactly <code class="docutils literal notranslate"><span class="pre">frames_per_clip</span></code> elements, so not all
+frames in a video might be present.</p>
+<p>Internally, it uses a VideoClips object to handle clip creation.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the UCF101 Dataset.</p></li>
+<li><p><strong>annotation_path</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – path to the folder containing the split files;
+see docstring above for download instructions of these files</p></li>
+<li><p><strong>frames_per_clip</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of frames in a clip.</p></li>
+<li><p><strong>step_between_clips</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of frames between each clip.</p></li>
+<li><p><strong>fold</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – which fold to use. Should be between 1 and 3.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – if <code class="docutils literal notranslate"><span class="pre">True</span></code>, creates a dataset from the train split,
+otherwise from the <code class="docutils literal notranslate"><span class="pre">test</span></code> split.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in a TxHxWxC video
+and returns a transformed version.</p></li>
+<li><p><strong>output_format</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – The format of the output video tensors (before transforms).
+Can be either “THWC” (default) or “TCHW”.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><p>A 3-tuple with the following entries:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>video (Tensor[T, H, W, C] or Tensor[T, C, H, W]): The <cite>T</cite> video frames</p></li>
+<li><p>audio(Tensor[K, L]): the audio frames, where <cite>K</cite> is the number of channels
+and <cite>L</cite> is the number of points</p></li>
+<li><p>label (int): class of the video clip</p></li>
+</ul>
+</div></blockquote>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.UCF101.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/ucf101.html#UCF101.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.UCF101.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sample and meta data, optionally transformed by the respective transforms.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(Any)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.MovingMNIST.html" class="btn btn-neutral float-right" title="MovingMNIST" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.Kinetics.html" class="btn btn-neutral" title="Kinetics" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">UCF101</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.USPS.html b/0.16/generated/torchvision.datasets.USPS.html
new file mode 100644
index 00000000000..b7832ac8e73
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.USPS.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>USPS &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CocoDetection" href="torchvision.datasets.CocoDetection.html" />
+    <link rel="prev" title="SVHN" href="torchvision.datasets.SVHN.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>USPS</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.USPS.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="usps">
+<h1>USPS<a class="headerlink" href="#usps" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.USPS">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">USPS</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/usps.html#USPS"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.USPS" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html#usps">USPS</a> Dataset.
+The data-format is : [label [index:value ]*256 n] * num_lines, where <code class="docutils literal notranslate"><span class="pre">label</span></code> lies in <code class="docutils literal notranslate"><span class="pre">[1,</span> <span class="pre">10]</span></code>.
+The value for each pixel lies in <code class="docutils literal notranslate"><span class="pre">[-1,</span> <span class="pre">1]</span></code>. Here we transform the <code class="docutils literal notranslate"><span class="pre">label</span></code> into <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">9]</span></code>
+and make pixel values in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">255]</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset to store``USPS`` data files.</p></li>
+<li><p><strong>train</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, creates dataset from <code class="docutils literal notranslate"><span class="pre">usps.bz2</span></code>,
+otherwise from <code class="docutils literal notranslate"><span class="pre">usps.t.bz2</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.USPS.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/usps.html#USPS.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.USPS.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is index of the target class.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.CocoDetection.html" class="btn btn-neutral float-right" title="CocoDetection" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.SVHN.html" class="btn btn-neutral" title="SVHN" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">USPS</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.VOCDetection.html b/0.16/generated/torchvision.datasets.VOCDetection.html
new file mode 100644
index 00000000000..2bb5787e40f
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.VOCDetection.html
@@ -0,0 +1,734 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>VOCDetection &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="WIDERFace" href="torchvision.datasets.WIDERFace.html" />
+    <link rel="prev" title="VOCSegmentation" href="torchvision.datasets.VOCSegmentation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>VOCDetection</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.VOCDetection.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vocdetection">
+<h1>VOCDetection<a class="headerlink" href="#vocdetection" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.VOCDetection">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">VOCDetection</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">year</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'2012'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_set</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/voc.html#VOCDetection"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.VOCDetection" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://host.robots.ox.ac.uk/pascal/VOC/">Pascal VOC</a> Detection Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the VOC Dataset.</p></li>
+<li><p><strong>year</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset year, supports years <code class="docutils literal notranslate"><span class="pre">&quot;2007&quot;</span></code> to <code class="docutils literal notranslate"><span class="pre">&quot;2012&quot;</span></code>.</p></li>
+<li><p><strong>image_set</strong> (<em>string</em><em>, </em><em>optional</em>) – Select the image_set to use, <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code>, <code class="docutils literal notranslate"><span class="pre">&quot;trainval&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;val&quot;</span></code>. If
+<code class="docutils literal notranslate"><span class="pre">year==&quot;2007&quot;</span></code>, can also be <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.
+(default: alphabetic indexing of VOC’s 20 classes).</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>required</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes input sample and its target as entry
+and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.VOCDetection.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/voc.html#VOCDetection.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.VOCDetection.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is a dictionary of the XML tree.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.WIDERFace.html" class="btn btn-neutral float-right" title="WIDERFace" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.VOCSegmentation.html" class="btn btn-neutral" title="VOCSegmentation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">VOCDetection</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.VOCSegmentation.html b/0.16/generated/torchvision.datasets.VOCSegmentation.html
new file mode 100644
index 00000000000..13d51c6d93a
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.VOCSegmentation.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>VOCSegmentation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="VOCDetection" href="torchvision.datasets.VOCDetection.html" />
+    <link rel="prev" title="SBDataset" href="torchvision.datasets.SBDataset.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>VOCSegmentation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.VOCSegmentation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vocsegmentation">
+<h1>VOCSegmentation<a class="headerlink" href="#vocsegmentation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.VOCSegmentation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">VOCSegmentation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">year</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'2012'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_set</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/voc.html#VOCSegmentation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.VOCSegmentation" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://host.robots.ox.ac.uk/pascal/VOC/">Pascal VOC</a> Segmentation Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of the VOC Dataset.</p></li>
+<li><p><strong>year</strong> (<em>string</em><em>, </em><em>optional</em>) – The dataset year, supports years <code class="docutils literal notranslate"><span class="pre">&quot;2007&quot;</span></code> to <code class="docutils literal notranslate"><span class="pre">&quot;2012&quot;</span></code>.</p></li>
+<li><p><strong>image_set</strong> (<em>string</em><em>, </em><em>optional</em>) – Select the image_set to use, <code class="docutils literal notranslate"><span class="pre">&quot;train&quot;</span></code>, <code class="docutils literal notranslate"><span class="pre">&quot;trainval&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;val&quot;</span></code>. If
+<code class="docutils literal notranslate"><span class="pre">year==&quot;2007&quot;</span></code>, can also be <code class="docutils literal notranslate"><span class="pre">&quot;test&quot;</span></code>.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes input sample and its target as entry
+and returns a transformed version.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.VOCSegmentation.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/voc.html#VOCSegmentation.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.VOCSegmentation.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is the image segmentation.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.VOCDetection.html" class="btn btn-neutral float-right" title="VOCDetection" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.SBDataset.html" class="btn btn-neutral" title="SBDataset" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">VOCSegmentation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.VisionDataset.html b/0.16/generated/torchvision.datasets.VisionDataset.html
new file mode 100644
index 00000000000..7ee924321fd
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.VisionDataset.html
@@ -0,0 +1,719 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>VisionDataset &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="wrap_dataset_for_transforms_v2" href="torchvision.datasets.wrap_dataset_for_transforms_v2.html" />
+    <link rel="prev" title="ImageFolder" href="torchvision.datasets.ImageFolder.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>VisionDataset</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.VisionDataset.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="visiondataset">
+<h1>VisionDataset<a class="headerlink" href="#visiondataset" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.VisionDataset">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">VisionDataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/vision.html#VisionDataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.VisionDataset" title="Permalink to this definition">¶</a></dt>
+<dd><p>Base Class For making datasets which are compatible with torchvision.
+It is necessary to override the <code class="docutils literal notranslate"><span class="pre">__getitem__</span></code> and <code class="docutils literal notranslate"><span class="pre">__len__</span></code> method.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – Root directory of dataset.</p></li>
+<li><p><strong>transforms</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transforms that takes in
+an image and a label and returns the transformed versions of both.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in an PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+</ul>
+</dd>
+</dl>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p><code class="xref py py-attr docutils literal notranslate"><span class="pre">transforms</span></code> and the combination of <code class="xref py py-attr docutils literal notranslate"><span class="pre">transform</span></code> and <code class="xref py py-attr docutils literal notranslate"><span class="pre">target_transform</span></code> are mutually exclusive.</p>
+</div>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">VisionDataset</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.wrap_dataset_for_transforms_v2.html" class="btn btn-neutral float-right" title="wrap_dataset_for_transforms_v2" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.ImageFolder.html" class="btn btn-neutral" title="ImageFolder" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">VisionDataset</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.WIDERFace.html b/0.16/generated/torchvision.datasets.WIDERFace.html
new file mode 100644
index 00000000000..153640a2767
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.WIDERFace.html
@@ -0,0 +1,741 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>WIDERFace &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FlyingChairs" href="torchvision.datasets.FlyingChairs.html" />
+    <link rel="prev" title="VOCDetection" href="torchvision.datasets.VOCDetection.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>WIDERFace</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.WIDERFace.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="widerface">
+<h1>WIDERFace<a class="headerlink" href="#widerface" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.datasets.WIDERFace">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">WIDERFace</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">split</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_transform</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">download</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/datasets/widerface.html#WIDERFace"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.WIDERFace" title="Permalink to this definition">¶</a></dt>
+<dd><p><a class="reference external" href="http://shuoyang1213.me/WIDERFACE/">WIDERFace</a> Dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>root</strong> (<em>string</em>) – <p>Root directory where images and annotations are downloaded to.
+Expects the following folder structure if download=False:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;root&gt;
+    └── widerface
+        ├── wider_face_split (&#39;wider_face_split.zip&#39; if compressed)
+        ├── WIDER_train (&#39;WIDER_train.zip&#39; if compressed)
+        ├── WIDER_val (&#39;WIDER_val.zip&#39; if compressed)
+        └── WIDER_test (&#39;WIDER_test.zip&#39; if compressed)
+</pre></div>
+</div>
+</p></li>
+<li><p><strong>split</strong> (<em>string</em>) – The dataset split to use. One of {<code class="docutils literal notranslate"><span class="pre">train</span></code>, <code class="docutils literal notranslate"><span class="pre">val</span></code>, <code class="docutils literal notranslate"><span class="pre">test</span></code>}.
+Defaults to <code class="docutils literal notranslate"><span class="pre">train</span></code>.</p></li>
+<li><p><strong>transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that  takes in a PIL image
+and returns a transformed version. E.g, <code class="docutils literal notranslate"><span class="pre">transforms.RandomCrop</span></code></p></li>
+<li><p><strong>target_transform</strong> (<em>callable</em><em>, </em><em>optional</em>) – A function/transform that takes in the
+target and transforms it.</p></li>
+<li><p><strong>download</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If true, downloads the dataset from the internet and
+puts it in root directory. If dataset is already downloaded, it is not
+downloaded again.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Special-members<span class="colon">:</span></dt>
+<dd class="field-odd"><p></p></dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.datasets.WIDERFace.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">index</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/datasets/widerface.html#WIDERFace.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.WIDERFace.__getitem__" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>index</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Index</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>(image, target) where target is a dict of annotations for all faces in the image.
+target=None for the test split.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.datasets.FlyingChairs.html" class="btn btn-neutral float-right" title="FlyingChairs" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.VOCDetection.html" class="btn btn-neutral" title="VOCDetection" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">WIDERFace</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html b/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
new file mode 100644
index 00000000000..3e42acc468b
--- /dev/null
+++ b/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
@@ -0,0 +1,778 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>wrap_dataset_for_transforms_v2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Utils" href="../utils.html" />
+    <link rel="prev" title="VisionDataset" href="torchvision.datasets.VisionDataset.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../datasets.html">Datasets</a> &gt;</li>
+        
+      <li>wrap_dataset_for_transforms_v2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="wrap-dataset-for-transforms-v2">
+<h1>wrap_dataset_for_transforms_v2<a class="headerlink" href="#wrap-dataset-for-transforms-v2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.datasets.wrap_dataset_for_transforms_v2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.datasets.</span></span><span class="sig-name descname"><span class="pre">wrap_dataset_for_transforms_v2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_keys</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_dataset_wrapper.html#wrap_dataset_for_transforms_v2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.datasets.wrap_dataset_for_transforms_v2" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Wrap a <code class="docutils literal notranslate"><span class="pre">torchvision.dataset</span></code> for usage with <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The wrap_dataset_for_transforms_v2 function is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">dataset</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">CocoDetection</span><span class="p">(</span><span class="o">...</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">dataset</span> <span class="o">=</span> <span class="n">wrap_dataset_for_transforms_v2</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>For now, only the most popular datasets are supported. Furthermore, the wrapper only supports dataset
+configurations that are fully supported by <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>. If you encounter an error prompting you
+to raise an issue to <code class="docutils literal notranslate"><span class="pre">torchvision</span></code> for a dataset or configuration that you need, please do so.</p>
+</div>
+<p>The dataset samples are wrapped according to the description below.</p>
+<p>Special cases:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p><a class="reference internal" href="torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection" title="torchvision.datasets.CocoDetection"><code class="xref py py-class docutils literal notranslate"><span class="pre">CocoDetection</span></code></a>: Instead of returning the target as list of dicts, the wrapper
+returns a dict of lists. In addition, the key-value-pairs <code class="docutils literal notranslate"><span class="pre">&quot;boxes&quot;</span></code> (in <code class="docutils literal notranslate"><span class="pre">XYXY</span></code> coordinate format),
+<code class="docutils literal notranslate"><span class="pre">&quot;masks&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;labels&quot;</span></code> are added and wrap the data in the corresponding <code class="docutils literal notranslate"><span class="pre">torchvision.tv_tensors</span></code>.
+The original keys are preserved. If <code class="docutils literal notranslate"><span class="pre">target_keys</span></code> is omitted, returns only the values for the
+<code class="docutils literal notranslate"><span class="pre">&quot;image_id&quot;</span></code>, <code class="docutils literal notranslate"><span class="pre">&quot;boxes&quot;</span></code>, and <code class="docutils literal notranslate"><span class="pre">&quot;labels&quot;</span></code>.</p></li>
+<li><p><a class="reference internal" href="torchvision.datasets.VOCDetection.html#torchvision.datasets.VOCDetection" title="torchvision.datasets.VOCDetection"><code class="xref py py-class docutils literal notranslate"><span class="pre">VOCDetection</span></code></a>: The key-value-pairs <code class="docutils literal notranslate"><span class="pre">&quot;boxes&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;labels&quot;</span></code> are added to
+the target and wrap the data in the corresponding <code class="docutils literal notranslate"><span class="pre">torchvision.tv_tensors</span></code>. The original keys are
+preserved. If <code class="docutils literal notranslate"><span class="pre">target_keys</span></code> is omitted, returns only the values for the <code class="docutils literal notranslate"><span class="pre">&quot;boxes&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;labels&quot;</span></code>.</p></li>
+<li><p><a class="reference internal" href="torchvision.datasets.CelebA.html#torchvision.datasets.CelebA" title="torchvision.datasets.CelebA"><code class="xref py py-class docutils literal notranslate"><span class="pre">CelebA</span></code></a>: The target for <code class="docutils literal notranslate"><span class="pre">target_type=&quot;bbox&quot;</span></code> is converted to the <code class="docutils literal notranslate"><span class="pre">XYXY</span></code>
+coordinate format and wrapped into a <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> tv_tensor.</p></li>
+<li><p><a class="reference internal" href="torchvision.datasets.Kitti.html#torchvision.datasets.Kitti" title="torchvision.datasets.Kitti"><code class="xref py py-class docutils literal notranslate"><span class="pre">Kitti</span></code></a>: Instead returning the target as list of dicts, the wrapper returns a
+dict of lists. In addition, the key-value-pairs <code class="docutils literal notranslate"><span class="pre">&quot;boxes&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;labels&quot;</span></code> are added and wrap the data
+in the corresponding <code class="docutils literal notranslate"><span class="pre">torchvision.tv_tensors</span></code>. The original keys are preserved. If <code class="docutils literal notranslate"><span class="pre">target_keys</span></code> is
+omitted, returns only the values for the <code class="docutils literal notranslate"><span class="pre">&quot;boxes&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;labels&quot;</span></code>.</p></li>
+<li><p><a class="reference internal" href="torchvision.datasets.OxfordIIITPet.html#torchvision.datasets.OxfordIIITPet" title="torchvision.datasets.OxfordIIITPet"><code class="xref py py-class docutils literal notranslate"><span class="pre">OxfordIIITPet</span></code></a>: The target for <code class="docutils literal notranslate"><span class="pre">target_type=&quot;segmentation&quot;</span></code> is wrapped into a
+<a class="reference internal" href="torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">Mask</span></code></a> tv_tensor.</p></li>
+<li><p><a class="reference internal" href="torchvision.datasets.Cityscapes.html#torchvision.datasets.Cityscapes" title="torchvision.datasets.Cityscapes"><code class="xref py py-class docutils literal notranslate"><span class="pre">Cityscapes</span></code></a>: The target for <code class="docutils literal notranslate"><span class="pre">target_type=&quot;semantic&quot;</span></code> is wrapped into a
+<a class="reference internal" href="torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">Mask</span></code></a> tv_tensor. The target for <code class="docutils literal notranslate"><span class="pre">target_type=&quot;instance&quot;</span></code> is <em>replaced</em> by
+a dictionary with the key-value-pairs <code class="docutils literal notranslate"><span class="pre">&quot;masks&quot;</span></code> (as <a class="reference internal" href="torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">Mask</span></code></a> tv_tensor) and
+<code class="docutils literal notranslate"><span class="pre">&quot;labels&quot;</span></code>.</p></li>
+<li><p><a class="reference internal" href="torchvision.datasets.WIDERFace.html#torchvision.datasets.WIDERFace" title="torchvision.datasets.WIDERFace"><code class="xref py py-class docutils literal notranslate"><span class="pre">WIDERFace</span></code></a>: The value for key <code class="docutils literal notranslate"><span class="pre">&quot;bbox&quot;</span></code> in the target is converted to <code class="docutils literal notranslate"><span class="pre">XYXY</span></code>
+coordinate format and wrapped into a <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> tv_tensor.</p></li>
+</ul>
+</div></blockquote>
+<p>Image classification datasets</p>
+<blockquote>
+<div><p>This wrapper is a no-op for image classification datasets, since they were already fully supported by
+<code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms</span></code> and thus no change is needed for <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>.</p>
+</div></blockquote>
+<p>Segmentation datasets</p>
+<blockquote>
+<div><p>Segmentation datasets, e.g. <a class="reference internal" href="torchvision.datasets.VOCSegmentation.html#torchvision.datasets.VOCSegmentation" title="torchvision.datasets.VOCSegmentation"><code class="xref py py-class docutils literal notranslate"><span class="pre">VOCSegmentation</span></code></a>, return a two-tuple of
+<a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">PIL.Image.Image</span></code></a>’s. This wrapper leaves the image as is (first item), while wrapping the
+segmentation mask into a <a class="reference internal" href="torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-class docutils literal notranslate"><span class="pre">Mask</span></code></a> (second item).</p>
+</div></blockquote>
+<p>Video classification datasets</p>
+<blockquote>
+<div><p>Video classification datasets, e.g. <a class="reference internal" href="torchvision.datasets.Kinetics.html#torchvision.datasets.Kinetics" title="torchvision.datasets.Kinetics"><code class="xref py py-class docutils literal notranslate"><span class="pre">Kinetics</span></code></a>, return a three-tuple containing a
+<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> for the video and audio and a <a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><code class="xref py py-class docutils literal notranslate"><span class="pre">int</span></code></a> as label. This wrapper wraps the video into a
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a> while leaving the other items as is.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Only datasets constructed with <code class="docutils literal notranslate"><span class="pre">output_format=&quot;TCHW&quot;</span></code> are supported, since the alternative
+<code class="docutils literal notranslate"><span class="pre">output_format=&quot;THWC&quot;</span></code> is not supported by <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>.</p>
+</div>
+</div></blockquote>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>dataset</strong> – the dataset instance to wrap for compatibility with transforms v2.</p></li>
+<li><p><strong>target_keys</strong> – Target keys to return in case the target is a dictionary. If <code class="docutils literal notranslate"><span class="pre">None</span></code> (default), selected keys are
+specific to the dataset. If <code class="docutils literal notranslate"><span class="pre">&quot;all&quot;</span></code>, returns the full target. Can also be a collection of strings for
+fine grained access. Currently only supported for <a class="reference internal" href="torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection" title="torchvision.datasets.CocoDetection"><code class="xref py py-class docutils literal notranslate"><span class="pre">CocoDetection</span></code></a>,
+<a class="reference internal" href="torchvision.datasets.VOCDetection.html#torchvision.datasets.VOCDetection" title="torchvision.datasets.VOCDetection"><code class="xref py py-class docutils literal notranslate"><span class="pre">VOCDetection</span></code></a>, <a class="reference internal" href="torchvision.datasets.Kitti.html#torchvision.datasets.Kitti" title="torchvision.datasets.Kitti"><code class="xref py py-class docutils literal notranslate"><span class="pre">Kitti</span></code></a>, and
+<a class="reference internal" href="torchvision.datasets.WIDERFace.html#torchvision.datasets.WIDERFace" title="torchvision.datasets.WIDERFace"><code class="xref py py-class docutils literal notranslate"><span class="pre">WIDERFace</span></code></a>. See above for details.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">wrap_dataset_for_transforms_v2</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../utils.html" class="btn btn-neutral float-right" title="Utils" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.datasets.VisionDataset.html" class="btn btn-neutral" title="VisionDataset" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">wrap_dataset_for_transforms_v2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.ImageReadMode.html b/0.16/generated/torchvision.io.ImageReadMode.html
new file mode 100644
index 00000000000..f05acafc5fd
--- /dev/null
+++ b/0.16/generated/torchvision.io.ImageReadMode.html
@@ -0,0 +1,699 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ImageReadMode &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="read_video" href="torchvision.io.read_video.html" />
+    <link rel="prev" title="write_file" href="torchvision.io.write_file.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>ImageReadMode</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.ImageReadMode.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="imagereadmode">
+<h1>ImageReadMode<a class="headerlink" href="#imagereadmode" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.io.ImageReadMode">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">ImageReadMode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/io/image.html#ImageReadMode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.ImageReadMode" title="Permalink to this definition">¶</a></dt>
+<dd><p>Support for various modes while reading images.</p>
+<p>Use <code class="docutils literal notranslate"><span class="pre">ImageReadMode.UNCHANGED</span></code> for loading the image as-is,
+<code class="docutils literal notranslate"><span class="pre">ImageReadMode.GRAY</span></code> for converting to grayscale,
+<code class="docutils literal notranslate"><span class="pre">ImageReadMode.GRAY_ALPHA</span></code> for grayscale with transparency,
+<code class="docutils literal notranslate"><span class="pre">ImageReadMode.RGB</span></code> for RGB and <code class="docutils literal notranslate"><span class="pre">ImageReadMode.RGB_ALPHA</span></code> for
+RGB with transparency.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.read_video.html" class="btn btn-neutral float-right" title="read_video" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.write_file.html" class="btn btn-neutral" title="write_file" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ImageReadMode</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.VideoReader.html b/0.16/generated/torchvision.io.VideoReader.html
new file mode 100644
index 00000000000..bf3349b7027
--- /dev/null
+++ b/0.16/generated/torchvision.io.VideoReader.html
@@ -0,0 +1,828 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>VideoReader &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Feature extraction for model inspection" href="../feature_extraction.html" />
+    <link rel="prev" title="write_video" href="torchvision.io.write_video.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>VideoReader</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.VideoReader.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="videoreader">
+<h1>VideoReader<a class="headerlink" href="#videoreader" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.io.VideoReader">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">VideoReader</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">src</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">''</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stream</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'video'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_threads</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/io/video_reader.html#VideoReader"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.VideoReader" title="Permalink to this definition">¶</a></dt>
+<dd><p>Fine-grained video-reading API.
+Supports frame-by-frame reading of various streams from a single video
+container. Much like previous video_reader API it supports the following
+backends: video_reader, pyav, and cuda.
+Backends can be set via <cite>torchvision.set_video_backend</cite> function.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The VideoReader class is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p class="rubric">Example</p>
+<p>The following examples creates a <code class="xref py py-mod docutils literal notranslate"><span class="pre">VideoReader</span></code> object, seeks into 2s
+point, and returns a single frame:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torchvision</span>
+<span class="n">video_path</span> <span class="o">=</span> <span class="s2">&quot;path_to_a_test_video&quot;</span>
+<span class="n">reader</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">io</span><span class="o">.</span><span class="n">VideoReader</span><span class="p">(</span><span class="n">video_path</span><span class="p">,</span> <span class="s2">&quot;video&quot;</span><span class="p">)</span>
+<span class="n">reader</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mf">2.0</span><span class="p">)</span>
+<span class="n">frame</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">reader</span><span class="p">)</span>
+</pre></div>
+</div>
+<p><code class="xref py py-mod docutils literal notranslate"><span class="pre">VideoReader</span></code> implements the iterable API, which makes it suitable to
+using it in conjunction with <a class="reference external" href="https://docs.python.org/3/library/itertools.html#module-itertools" title="(in Python v3.12)"><code class="xref py py-mod docutils literal notranslate"><span class="pre">itertools</span></code></a> for more advanced reading.
+As such, we can use a <code class="xref py py-mod docutils literal notranslate"><span class="pre">VideoReader</span></code> instance inside for loops:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">reader</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span>
+<span class="k">for</span> <span class="n">frame</span> <span class="ow">in</span> <span class="n">reader</span><span class="p">:</span>
+    <span class="n">frames</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">frame</span><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+<span class="c1"># additionally, `seek` implements a fluent API, so we can do</span>
+<span class="k">for</span> <span class="n">frame</span> <span class="ow">in</span> <span class="n">reader</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mi">2</span><span class="p">):</span>
+    <span class="n">frames</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">frame</span><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+</pre></div>
+</div>
+<p>With <a class="reference external" href="https://docs.python.org/3/library/itertools.html#module-itertools" title="(in Python v3.12)"><code class="xref py py-mod docutils literal notranslate"><span class="pre">itertools</span></code></a>, we can read all frames between 2 and 5 seconds with the
+following code:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">for</span> <span class="n">frame</span> <span class="ow">in</span> <span class="n">itertools</span><span class="o">.</span><span class="n">takewhile</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="s1">&#39;pts&#39;</span><span class="p">]</span> <span class="o">&lt;=</span> <span class="mi">5</span><span class="p">,</span> <span class="n">reader</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mi">2</span><span class="p">)):</span>
+    <span class="n">frames</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">frame</span><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+</pre></div>
+</div>
+<p>and similarly, reading 10 frames after the 2s timestamp can be achieved
+as follows:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">for</span> <span class="n">frame</span> <span class="ow">in</span> <span class="n">itertools</span><span class="o">.</span><span class="n">islice</span><span class="p">(</span><span class="n">reader</span><span class="o">.</span><span class="n">seek</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="mi">10</span><span class="p">):</span>
+    <span class="n">frames</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">frame</span><span class="p">[</span><span class="s1">&#39;data&#39;</span><span class="p">])</span>
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Each stream descriptor consists of two parts: stream type (e.g. ‘video’) and
+a unique stream id (which are determined by the video encoding).
+In this way, if the video container contains multiple
+streams of the same type, users can access the one they want.
+If only stream type is passed, the decoder auto-detects first stream of that type.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>src</strong> (<em>string</em><em>, </em><em>bytes object</em><em>, or </em><em>tensor</em>) – The media source.
+If string-type, it must be a file path supported by FFMPEG.
+If bytes, should be an in-memory representation of a file supported by FFMPEG.
+If Tensor, it is interpreted internally as byte buffer.
+It must be one-dimensional, of type <code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code>.</p></li>
+<li><p><strong>stream</strong> (<em>string</em><em>, </em><em>optional</em>) – descriptor of the required stream, followed by the stream id,
+in the format <code class="docutils literal notranslate"><span class="pre">{stream_type}:{stream_id}</span></code>. Defaults to <code class="docutils literal notranslate"><span class="pre">&quot;video:0&quot;</span></code>.
+Currently available options include <code class="docutils literal notranslate"><span class="pre">['video',</span> <span class="pre">'audio']</span></code></p></li>
+<li><p><strong>num_threads</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of threads used by the codec to decode video.
+Default value (0) enables multithreading with codec-dependent heuristic. The performance
+will depend on the version of FFMPEG codecs supported.</p></li>
+<li><p><strong>path</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – </p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">VideoReader</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the..."><img alt="" src="../_images/sphx_glr_plot_video_api_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_video_api.html#sphx-glr-auto-examples-others-plot-video-api-py"><span class="std std-ref">Video API</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Video API</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.io.VideoReader.get_metadata">
+<span class="sig-name descname"><span class="pre">get_metadata</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/io/video_reader.html#VideoReader.get_metadata"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.VideoReader.get_metadata" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns video metadata</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>dictionary containing duration and frame rate for every stream</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>(<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)">dict</a>)</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.io.VideoReader.seek">
+<span class="sig-name descname"><span class="pre">seek</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">time_s</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">keyframes_only</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="#torchvision.io.VideoReader" title="torchvision.io.video_reader.VideoReader"><span class="pre">VideoReader</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/video_reader.html#VideoReader.seek"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.VideoReader.seek" title="Permalink to this definition">¶</a></dt>
+<dd><p>Seek within current stream.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>time_s</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – seek time in seconds</p></li>
+<li><p><strong>keyframes_only</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – allow to seek only to keyframes</p></li>
+</ul>
+</dd>
+</dl>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Current implementation is the so-called precise seek. This
+means following seek, call to <code class="xref py py-mod docutils literal notranslate"><span class="pre">next()</span></code> will return the
+frame with the exact timestamp if it exists or
+the first frame with timestamp larger than <code class="docutils literal notranslate"><span class="pre">time_s</span></code>.</p>
+</div>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.io.VideoReader.set_current_stream">
+<span class="sig-name descname"><span class="pre">set_current_stream</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">stream</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/video_reader.html#VideoReader.set_current_stream"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.VideoReader.set_current_stream" title="Permalink to this definition">¶</a></dt>
+<dd><p>Set current stream.
+Explicitly define the stream we are operating on.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>stream</strong> (<em>string</em>) – descriptor of the required stream. Defaults to <code class="docutils literal notranslate"><span class="pre">&quot;video:0&quot;</span></code>
+Currently available stream types include <code class="docutils literal notranslate"><span class="pre">['video',</span> <span class="pre">'audio']</span></code>.
+Each descriptor consists of two parts: stream type (e.g. ‘video’) and
+a unique stream id (which are determined by video encoding).
+In this way, if the video container contains multiple
+streams of the same type, users can access the one they want.
+If only stream type is passed, the decoder auto-detects first stream
+of that type and returns it.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>True on success, False otherwise</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>(<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)">bool</a>)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../feature_extraction.html" class="btn btn-neutral float-right" title="Feature extraction for model inspection" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.write_video.html" class="btn btn-neutral" title="write_video" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">VideoReader</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.decode_image.html b/0.16/generated/torchvision.io.decode_image.html
new file mode 100644
index 00000000000..0c2118019d0
--- /dev/null
+++ b/0.16/generated/torchvision.io.decode_image.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>decode_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="encode_jpeg" href="torchvision.io.encode_jpeg.html" />
+    <link rel="prev" title="read_image" href="torchvision.io.read_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>decode_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.decode_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="decode-image">
+<h1>decode_image<a class="headerlink" href="#decode-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.decode_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">decode_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.image.ImageReadMode"><span class="pre">ImageReadMode</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ImageReadMode.UNCHANGED</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#decode_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.decode_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>Detects whether an image is a JPEG or PNG and performs the appropriate
+operation to decode the image into a 3 dimensional RGB or grayscale Tensor.</p>
+<p>Optionally converts the image to the desired format.
+The values of the output tensor are uint8 in [0, 255].</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em>) – a one dimensional uint8 tensor containing the raw bytes of the
+PNG or JPEG image.</p></li>
+<li><p><strong>mode</strong> (<a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.ImageReadMode"><em>ImageReadMode</em></a>) – the read mode used for optionally converting the image.
+Default: <code class="docutils literal notranslate"><span class="pre">ImageReadMode.UNCHANGED</span></code>.
+See <code class="docutils literal notranslate"><span class="pre">ImageReadMode</span></code> class for more information on various
+available modes.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>output (Tensor[image_channels, image_height, image_width])</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.encode_jpeg.html" class="btn btn-neutral float-right" title="encode_jpeg" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.read_image.html" class="btn btn-neutral" title="read_image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">decode_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.decode_jpeg.html b/0.16/generated/torchvision.io.decode_jpeg.html
new file mode 100644
index 00000000000..b8c30f547b6
--- /dev/null
+++ b/0.16/generated/torchvision.io.decode_jpeg.html
@@ -0,0 +1,728 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>decode_jpeg &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="write_jpeg" href="torchvision.io.write_jpeg.html" />
+    <link rel="prev" title="encode_jpeg" href="torchvision.io.encode_jpeg.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>decode_jpeg</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.decode_jpeg.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="decode-jpeg">
+<h1>decode_jpeg<a class="headerlink" href="#decode-jpeg" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.decode_jpeg">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">decode_jpeg</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.image.ImageReadMode"><span class="pre">ImageReadMode</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ImageReadMode.UNCHANGED</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'cpu'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#decode_jpeg"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.decode_jpeg" title="Permalink to this definition">¶</a></dt>
+<dd><p>Decodes a JPEG image into a 3 dimensional RGB or grayscale Tensor.
+Optionally converts the image to the desired format.
+The values of the output tensor are uint8 between 0 and 255.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>1</em><em>]</em>) – a one dimensional uint8 tensor containing
+the raw bytes of the JPEG image. This tensor must be on CPU,
+regardless of the <code class="docutils literal notranslate"><span class="pre">device</span></code> parameter.</p></li>
+<li><p><strong>mode</strong> (<a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.ImageReadMode"><em>ImageReadMode</em></a>) – the read mode used for optionally
+converting the image. The supported modes are: <code class="docutils literal notranslate"><span class="pre">ImageReadMode.UNCHANGED</span></code>,
+<code class="docutils literal notranslate"><span class="pre">ImageReadMode.GRAY</span></code> and <code class="docutils literal notranslate"><span class="pre">ImageReadMode.RGB</span></code>
+Default: <code class="docutils literal notranslate"><span class="pre">ImageReadMode.UNCHANGED</span></code>.
+See <code class="docutils literal notranslate"><span class="pre">ImageReadMode</span></code> class for more information on various
+available modes.</p></li>
+<li><p><strong>device</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em> or </em><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><em>torch.device</em></a>) – <p>The device on which the decoded image will
+be stored. If a cuda device is specified, the image will be decoded
+with <a class="reference external" href="https://developer.nvidia.com/nvjpeg">nvjpeg</a>. This is only
+supported for CUDA version &gt;= 10.1</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The device parameter is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>There is a memory leak in the nvjpeg library for CUDA versions &lt; 11.6.
+Make sure to rely on CUDA 11.6 or above before using <code class="docutils literal notranslate"><span class="pre">device=&quot;cuda&quot;</span></code>.</p>
+</div>
+</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>output (Tensor[image_channels, image_height, image_width])</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.write_jpeg.html" class="btn btn-neutral float-right" title="write_jpeg" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.encode_jpeg.html" class="btn btn-neutral" title="encode_jpeg" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">decode_jpeg</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.decode_png.html b/0.16/generated/torchvision.io.decode_png.html
new file mode 100644
index 00000000000..3c5e422660b
--- /dev/null
+++ b/0.16/generated/torchvision.io.decode_png.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>decode_png &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="write_png" href="torchvision.io.write_png.html" />
+    <link rel="prev" title="encode_png" href="torchvision.io.encode_png.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>decode_png</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.decode_png.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="decode-png">
+<h1>decode_png<a class="headerlink" href="#decode-png" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.decode_png">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">decode_png</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.image.ImageReadMode"><span class="pre">ImageReadMode</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ImageReadMode.UNCHANGED</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#decode_png"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.decode_png" title="Permalink to this definition">¶</a></dt>
+<dd><p>Decodes a PNG image into a 3 dimensional RGB or grayscale Tensor.
+Optionally converts the image to the desired format.
+The values of the output tensor are uint8 in [0, 255].</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>1</em><em>]</em>) – a one dimensional uint8 tensor containing
+the raw bytes of the PNG image.</p></li>
+<li><p><strong>mode</strong> (<a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.ImageReadMode"><em>ImageReadMode</em></a>) – the read mode used for optionally
+converting the image. Default: <code class="docutils literal notranslate"><span class="pre">ImageReadMode.UNCHANGED</span></code>.
+See <cite>ImageReadMode</cite> class for more information on various
+available modes.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>output (Tensor[image_channels, image_height, image_width])</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.write_png.html" class="btn btn-neutral float-right" title="write_png" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.encode_png.html" class="btn btn-neutral" title="encode_png" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">decode_png</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.encode_jpeg.html b/0.16/generated/torchvision.io.encode_jpeg.html
new file mode 100644
index 00000000000..734a4dc2d86
--- /dev/null
+++ b/0.16/generated/torchvision.io.encode_jpeg.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>encode_jpeg &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="decode_jpeg" href="torchvision.io.decode_jpeg.html" />
+    <link rel="prev" title="decode_image" href="torchvision.io.decode_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>encode_jpeg</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.encode_jpeg.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="encode-jpeg">
+<h1>encode_jpeg<a class="headerlink" href="#encode-jpeg" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.encode_jpeg">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">encode_jpeg</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">quality</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">75</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#encode_jpeg"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.encode_jpeg" title="Permalink to this definition">¶</a></dt>
+<dd><p>Takes an input tensor in CHW layout and returns a buffer with the contents
+of its corresponding JPEG file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>channels</em><em>, </em><em>image_height</em><em>, </em><em>image_width</em><em>]</em><em>)</em>) – int8 image tensor of
+<code class="docutils literal notranslate"><span class="pre">c</span></code> channels, where <code class="docutils literal notranslate"><span class="pre">c</span></code> must be 1 or 3.</p></li>
+<li><p><strong>quality</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Quality of the resulting JPEG file, it must be a number between
+1 and 100. Default: 75</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><dl class="simple">
+<dt>A one dimensional int8 tensor that contains the raw bytes of the</dt><dd><p>JPEG file.</p>
+</dd>
+</dl>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>output (Tensor[1])</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.decode_jpeg.html" class="btn btn-neutral float-right" title="decode_jpeg" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.decode_image.html" class="btn btn-neutral" title="decode_image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">encode_jpeg</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.encode_png.html b/0.16/generated/torchvision.io.encode_png.html
new file mode 100644
index 00000000000..091ba3bc79d
--- /dev/null
+++ b/0.16/generated/torchvision.io.encode_png.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>encode_png &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="decode_png" href="torchvision.io.decode_png.html" />
+    <link rel="prev" title="write_jpeg" href="torchvision.io.write_jpeg.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>encode_png</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.encode_png.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="encode-png">
+<h1>encode_png<a class="headerlink" href="#encode-png" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.encode_png">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">encode_png</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">compression_level</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">6</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#encode_png"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.encode_png" title="Permalink to this definition">¶</a></dt>
+<dd><p>Takes an input tensor in CHW layout and returns a buffer with the contents
+of its corresponding PNG file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>channels</em><em>, </em><em>image_height</em><em>, </em><em>image_width</em><em>]</em>) – int8 image tensor of
+<code class="docutils literal notranslate"><span class="pre">c</span></code> channels, where <code class="docutils literal notranslate"><span class="pre">c</span></code> must 3 or 1.</p></li>
+<li><p><strong>compression_level</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Compression factor for the resulting file, it must be a number
+between 0 and 9. Default: 6</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><dl class="simple">
+<dt>A one dimensional int8 tensor that contains the raw bytes of the</dt><dd><p>PNG file.</p>
+</dd>
+</dl>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[1]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.decode_png.html" class="btn btn-neutral float-right" title="decode_png" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.write_jpeg.html" class="btn btn-neutral" title="write_jpeg" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">encode_png</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.read_file.html b/0.16/generated/torchvision.io.read_file.html
new file mode 100644
index 00000000000..5e27d0bac92
--- /dev/null
+++ b/0.16/generated/torchvision.io.read_file.html
@@ -0,0 +1,703 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>read_file &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="write_file" href="torchvision.io.write_file.html" />
+    <link rel="prev" title="write_png" href="torchvision.io.write_png.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>read_file</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.read_file.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="read-file">
+<h1>read_file<a class="headerlink" href="#read-file" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.read_file">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">read_file</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#read_file"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.read_file" title="Permalink to this definition">¶</a></dt>
+<dd><p>Reads and outputs the bytes contents of a file as a uint8 Tensor
+with one dimension.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>path</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – the path to the file to be read</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>data (Tensor)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.write_file.html" class="btn btn-neutral float-right" title="write_file" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.write_png.html" class="btn btn-neutral" title="write_png" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">read_file</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.read_image.html b/0.16/generated/torchvision.io.read_image.html
new file mode 100644
index 00000000000..2d1a503d055
--- /dev/null
+++ b/0.16/generated/torchvision.io.read_image.html
@@ -0,0 +1,723 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>read_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="decode_image" href="torchvision.io.decode_image.html" />
+    <link rel="prev" title="Decoding / Encoding images and videos" href="../io.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>read_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.read_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="read-image">
+<h1>read_image<a class="headerlink" href="#read-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.read_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">read_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.image.ImageReadMode"><span class="pre">ImageReadMode</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ImageReadMode.UNCHANGED</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#read_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.read_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>Reads a JPEG or PNG image into a 3 dimensional RGB or grayscale Tensor.
+Optionally converts the image to the desired format.
+The values of the output tensor are uint8 in [0, 255].</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>path</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – path of the JPEG or PNG image.</p></li>
+<li><p><strong>mode</strong> (<a class="reference internal" href="torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.ImageReadMode"><em>ImageReadMode</em></a>) – the read mode used for optionally converting the image.
+Default: <code class="docutils literal notranslate"><span class="pre">ImageReadMode.UNCHANGED</span></code>.
+See <code class="docutils literal notranslate"><span class="pre">ImageReadMode</span></code> class for more information on various
+available modes.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>output (Tensor[image_channels, image_height, image_width])</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">read_image</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images."><img alt="" src="../_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.decode_image.html" class="btn btn-neutral float-right" title="decode_image" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../io.html" class="btn btn-neutral" title="Decoding / Encoding images and videos" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">read_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.read_video.html b/0.16/generated/torchvision.io.read_video.html
new file mode 100644
index 00000000000..6bb5875d884
--- /dev/null
+++ b/0.16/generated/torchvision.io.read_video.html
@@ -0,0 +1,718 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>read_video &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="read_video_timestamps" href="torchvision.io.read_video_timestamps.html" />
+    <link rel="prev" title="ImageReadMode" href="torchvision.io.ImageReadMode.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>read_video</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.read_video.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="read-video">
+<h1>read_video<a class="headerlink" href="#read-video" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.read_video">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">read_video</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">start_pts</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/fractions.html#fractions.Fraction" title="(in Python v3.12)"><span class="pre">Fraction</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">end_pts</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/fractions.html#fractions.Fraction" title="(in Python v3.12)"><span class="pre">Fraction</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pts_unit</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'pts'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'THWC'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/io/video.html#read_video"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.read_video" title="Permalink to this definition">¶</a></dt>
+<dd><p>Reads a video from a file, returning both the video frames and the audio frames</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>filename</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – path to the video file</p></li>
+<li><p><strong>start_pts</strong> (<em>python:int if pts_unit = 'pts'</em><em>, </em><em>python:float / Fraction if pts_unit = 'sec'</em><em>, </em><em>optional</em>) – The start presentation time of the video</p></li>
+<li><p><strong>end_pts</strong> (<em>python:int if pts_unit = 'pts'</em><em>, </em><em>python:float / Fraction if pts_unit = 'sec'</em><em>, </em><em>optional</em>) – The end presentation time</p></li>
+<li><p><strong>pts_unit</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – unit in which start_pts and end_pts values will be interpreted,
+either ‘pts’ or ‘sec’. Defaults to ‘pts’.</p></li>
+<li><p><strong>output_format</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – The format of the output video tensors. Can be either “THWC” (default) or “TCHW”.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>the <cite>T</cite> video frames
+aframes (Tensor[K, L]): the audio frames, where <cite>K</cite> is the number of channels and <cite>L</cite> is the number of points
+info (Dict): metadata for the video and audio. Can contain the fields video_fps (float) and audio_fps (int)</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>vframes (Tensor[T, H, W, C] or Tensor[T, C, H, W])</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">read_video</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.read_video_timestamps.html" class="btn btn-neutral float-right" title="read_video_timestamps" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.ImageReadMode.html" class="btn btn-neutral" title="ImageReadMode" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">read_video</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.read_video_timestamps.html b/0.16/generated/torchvision.io.read_video_timestamps.html
new file mode 100644
index 00000000000..132ae9a6082
--- /dev/null
+++ b/0.16/generated/torchvision.io.read_video_timestamps.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>read_video_timestamps &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="write_video" href="torchvision.io.write_video.html" />
+    <link rel="prev" title="read_video" href="torchvision.io.read_video.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>read_video_timestamps</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.read_video_timestamps.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="read-video-timestamps">
+<h1>read_video_timestamps<a class="headerlink" href="#read-video-timestamps" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.read_video_timestamps">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">read_video_timestamps</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">pts_unit</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'pts'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/io/video.html#read_video_timestamps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.read_video_timestamps" title="Permalink to this definition">¶</a></dt>
+<dd><p>List the video frames timestamps.</p>
+<p>Note that the function decodes the whole video frame-by-frame.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>filename</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – path to the video file</p></li>
+<li><p><strong>pts_unit</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – unit in which timestamp values will be returned
+either ‘pts’ or ‘sec’. Defaults to ‘pts’.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>presentation timestamps for each one of the frames in the video.
+video_fps (float, optional): the frame rate for the video</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>pts (List[<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)">int</a>] if pts_unit = ‘pts’, List[Fraction] if pts_unit = ‘sec’)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.write_video.html" class="btn btn-neutral float-right" title="write_video" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.read_video.html" class="btn btn-neutral" title="read_video" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">read_video_timestamps</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.write_file.html b/0.16/generated/torchvision.io.write_file.html
new file mode 100644
index 00000000000..abae8b2a7c6
--- /dev/null
+++ b/0.16/generated/torchvision.io.write_file.html
@@ -0,0 +1,703 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>write_file &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ImageReadMode" href="torchvision.io.ImageReadMode.html" />
+    <link rel="prev" title="read_file" href="torchvision.io.read_file.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>write_file</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.write_file.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="write-file">
+<h1>write_file<a class="headerlink" href="#write-file" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.write_file">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">write_file</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/image.html#write_file"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.write_file" title="Permalink to this definition">¶</a></dt>
+<dd><p>Writes the contents of an uint8 tensor with one dimension to a
+file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>filename</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – the path to the file to be written</p></li>
+<li><p><strong>data</strong> (<em>Tensor</em>) – the contents to be written to the output file</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.ImageReadMode.html" class="btn btn-neutral float-right" title="ImageReadMode" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.read_file.html" class="btn btn-neutral" title="read_file" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">write_file</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.write_jpeg.html b/0.16/generated/torchvision.io.write_jpeg.html
new file mode 100644
index 00000000000..9d90a8147fe
--- /dev/null
+++ b/0.16/generated/torchvision.io.write_jpeg.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>write_jpeg &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="encode_png" href="torchvision.io.encode_png.html" />
+    <link rel="prev" title="decode_jpeg" href="torchvision.io.decode_jpeg.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>write_jpeg</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.write_jpeg.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="write-jpeg">
+<h1>write_jpeg<a class="headerlink" href="#write-jpeg" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.write_jpeg">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">write_jpeg</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">quality</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">75</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/io/image.html#write_jpeg"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.write_jpeg" title="Permalink to this definition">¶</a></dt>
+<dd><p>Takes an input tensor in CHW layout and saves it in a JPEG file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>channels</em><em>, </em><em>image_height</em><em>, </em><em>image_width</em><em>]</em>) – int8 image tensor of <code class="docutils literal notranslate"><span class="pre">c</span></code>
+channels, where <code class="docutils literal notranslate"><span class="pre">c</span></code> must be 1 or 3.</p></li>
+<li><p><strong>filename</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Path to save the image.</p></li>
+<li><p><strong>quality</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Quality of the resulting JPEG file, it must be a number
+between 1 and 100. Default: 75</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.encode_png.html" class="btn btn-neutral float-right" title="encode_png" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.decode_jpeg.html" class="btn btn-neutral" title="decode_jpeg" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">write_jpeg</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.write_png.html b/0.16/generated/torchvision.io.write_png.html
new file mode 100644
index 00000000000..7c9ead9d9d0
--- /dev/null
+++ b/0.16/generated/torchvision.io.write_png.html
@@ -0,0 +1,706 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>write_png &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="read_file" href="torchvision.io.read_file.html" />
+    <link rel="prev" title="decode_png" href="torchvision.io.decode_png.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>write_png</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.write_png.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="write-png">
+<h1>write_png<a class="headerlink" href="#write-png" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.write_png">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">write_png</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">compression_level</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">6</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/io/image.html#write_png"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.write_png" title="Permalink to this definition">¶</a></dt>
+<dd><p>Takes an input tensor in CHW layout (or HW in the case of grayscale images)
+and saves it in a PNG file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>channels</em><em>, </em><em>image_height</em><em>, </em><em>image_width</em><em>]</em>) – int8 image tensor of
+<code class="docutils literal notranslate"><span class="pre">c</span></code> channels, where <code class="docutils literal notranslate"><span class="pre">c</span></code> must be 1 or 3.</p></li>
+<li><p><strong>filename</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Path to save the image.</p></li>
+<li><p><strong>compression_level</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Compression factor for the resulting file, it must be a number
+between 0 and 9. Default: 6</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.read_file.html" class="btn btn-neutral float-right" title="read_file" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.decode_png.html" class="btn btn-neutral" title="decode_png" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">write_png</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.io.write_video.html b/0.16/generated/torchvision.io.write_video.html
new file mode 100644
index 00000000000..cb28b16ce4d
--- /dev/null
+++ b/0.16/generated/torchvision.io.write_video.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>write_video &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="VideoReader" href="torchvision.io.VideoReader.html" />
+    <link rel="prev" title="read_video_timestamps" href="torchvision.io.read_video_timestamps.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../io.html">Decoding / Encoding images and videos</a> &gt;</li>
+        
+      <li>write_video</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.io.write_video.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="write-video">
+<h1>write_video<a class="headerlink" href="#write-video" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.io.write_video">
+<span class="sig-prename descclassname"><span class="pre">torchvision.io.</span></span><span class="sig-name descname"><span class="pre">write_video</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">video_array</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">fps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">video_codec</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'libx264'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">options</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">audio_array</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">audio_fps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">audio_codec</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">audio_options</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a></span></span><a class="reference internal" href="../_modules/torchvision/io/video.html#write_video"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.io.write_video" title="Permalink to this definition">¶</a></dt>
+<dd><p>Writes a 4d tensor in [T, H, W, C] format in a video file</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>filename</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – path where the video will be saved</p></li>
+<li><p><strong>video_array</strong> (<em>Tensor</em><em>[</em><em>T</em><em>, </em><em>H</em><em>, </em><em>W</em><em>, </em><em>C</em><em>]</em>) – tensor containing the individual frames,
+as a uint8 tensor in [T, H, W, C] format</p></li>
+<li><p><strong>fps</strong> (<em>Number</em>) – video frames per second</p></li>
+<li><p><strong>video_codec</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – the name of the video codec, i.e. “libx264”, “h264”, etc.</p></li>
+<li><p><strong>options</strong> (<em>Dict</em>) – dictionary containing options to be passed into the PyAV video stream</p></li>
+<li><p><strong>audio_array</strong> (<em>Tensor</em><em>[</em><em>C</em><em>, </em><em>N</em><em>]</em>) – tensor containing the audio, where C is the number of channels
+and N is the number of samples</p></li>
+<li><p><strong>audio_fps</strong> (<em>Number</em>) – audio sample rate, typically 44100 or 48000</p></li>
+<li><p><strong>audio_codec</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – the name of the audio codec, i.e. “mp3”, “aac”, etc.</p></li>
+<li><p><strong>audio_options</strong> (<em>Dict</em>) – dictionary containing options to be passed into the PyAV audio stream</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.io.VideoReader.html" class="btn btn-neutral float-right" title="VideoReader" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.io.read_video_timestamps.html" class="btn btn-neutral" title="read_video_timestamps" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">write_video</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html b/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
new file mode 100644
index 00000000000..3e9cd57dcba
--- /dev/null
+++ b/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
@@ -0,0 +1,798 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>create_feature_extractor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="get_graph_node_names" href="torchvision.models.feature_extraction.get_graph_node_names.html" />
+    <link rel="prev" title="Feature extraction for model inspection" href="../feature_extraction.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../feature_extraction.html">Feature extraction for model inspection</a> &gt;</li>
+        
+      <li>create_feature_extractor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.models.feature_extraction.create_feature_extractor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="create-feature-extractor">
+<h1>create_feature_extractor<a class="headerlink" href="#create-feature-extractor" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.feature_extraction.create_feature_extractor">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.feature_extraction.</span></span><span class="sig-name descname"><span class="pre">create_feature_extractor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><span class="pre">Module</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_nodes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train_return_nodes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eval_return_nodes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tracer_kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">suppress_diff_warning</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/fx.html#torch.fx.GraphModule" title="(in PyTorch v2.0)"><span class="pre">GraphModule</span></a></span></span><a class="reference internal" href="../_modules/torchvision/models/feature_extraction.html#create_feature_extractor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.feature_extraction.create_feature_extractor" title="Permalink to this definition">¶</a></dt>
+<dd><p>Creates a new graph module that returns intermediate nodes from a given
+model as dictionary with user specified keys as strings, and the requested
+outputs as values. This is achieved by re-writing the computation graph of
+the model via FX to return the desired nodes as outputs. All unused nodes
+are removed, together with their corresponding parameters.</p>
+<p>Desired output nodes must be specified as a <code class="docutils literal notranslate"><span class="pre">.</span></code> separated
+path walking the module hierarchy from top level module down to leaf
+operation or leaf module. For more details on the node naming conventions
+used here, please see the <a class="reference internal" href="../feature_extraction.html#about-node-names"><span class="std std-ref">relevant subheading</span></a>
+in the <a class="reference external" href="https://pytorch.org/vision/stable/feature_extraction.html">documentation</a>.</p>
+<p>Not all models will be FX traceable, although with some massaging they can
+be made to cooperate. Here’s a (not exhaustive) list of tips:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>If you don’t need to trace through a particular, problematic
+sub-module, turn it into a “leaf module” by passing a list of
+<code class="docutils literal notranslate"><span class="pre">leaf_modules</span></code> as one of the <code class="docutils literal notranslate"><span class="pre">tracer_kwargs</span></code> (see example below).
+It will not be traced through, but rather, the resulting graph will
+hold a reference to that module’s forward method.</p></li>
+<li><p>Likewise, you may turn functions into leaf functions by passing a
+list of <code class="docutils literal notranslate"><span class="pre">autowrap_functions</span></code> as one of the <code class="docutils literal notranslate"><span class="pre">tracer_kwargs</span></code> (see
+example below).</p></li>
+<li><p>Some inbuilt Python functions can be problematic. For instance,
+<code class="docutils literal notranslate"><span class="pre">int</span></code> will raise an error during tracing. You may wrap them in your
+own function and then pass that in <code class="docutils literal notranslate"><span class="pre">autowrap_functions</span></code> as one of
+the <code class="docutils literal notranslate"><span class="pre">tracer_kwargs</span></code>.</p></li>
+</ul>
+</div></blockquote>
+<p>For further information on FX see the
+<a class="reference external" href="https://pytorch.org/docs/stable/fx.html">torch.fx documentation</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>model</strong> (<em>nn.Module</em>) – model on which we will extract the features</p></li>
+<li><p><strong>return_nodes</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – either a <code class="docutils literal notranslate"><span class="pre">List</span></code> or a <code class="docutils literal notranslate"><span class="pre">Dict</span></code>
+containing the names (or partial names - see note above)
+of the nodes for which the activations will be returned. If it is
+a <code class="docutils literal notranslate"><span class="pre">Dict</span></code>, the keys are the node names, and the values
+are the user-specified keys for the graph module’s returned
+dictionary. If it is a <code class="docutils literal notranslate"><span class="pre">List</span></code>, it is treated as a <code class="docutils literal notranslate"><span class="pre">Dict</span></code> mapping
+node specification strings directly to output names. In the case
+that <code class="docutils literal notranslate"><span class="pre">train_return_nodes</span></code> and <code class="docutils literal notranslate"><span class="pre">eval_return_nodes</span></code> are specified,
+this should not be specified.</p></li>
+<li><p><strong>train_return_nodes</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – similar to
+<code class="docutils literal notranslate"><span class="pre">return_nodes</span></code>. This can be used if the return nodes
+for train mode are different than those from eval mode.
+If this is specified, <code class="docutils literal notranslate"><span class="pre">eval_return_nodes</span></code> must also be specified,
+and <code class="docutils literal notranslate"><span class="pre">return_nodes</span></code> should not be specified.</p></li>
+<li><p><strong>eval_return_nodes</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – similar to
+<code class="docutils literal notranslate"><span class="pre">return_nodes</span></code>. This can be used if the return nodes
+for train mode are different than those from eval mode.
+If this is specified, <code class="docutils literal notranslate"><span class="pre">train_return_nodes</span></code> must also be specified,
+and <cite>return_nodes</cite> should not be specified.</p></li>
+<li><p><strong>tracer_kwargs</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – a dictionary of keyword arguments for
+<code class="docutils literal notranslate"><span class="pre">NodePathTracer</span></code> (which passes them onto it’s parent class
+<a class="reference external" href="https://pytorch.org/docs/stable/fx.html#torch.fx.Tracer">torch.fx.Tracer</a>).
+By default, it will be set to wrap and make leaf nodes all torchvision ops:
+{“autowrap_modules”: (math, torchvision.ops,),”leaf_modules”: _get_leaf_modules_for_ops(),}
+WARNING: In case the user provides tracer_kwargs, above default arguments will be appended to the user
+provided dictionary.</p></li>
+<li><p><strong>suppress_diff_warning</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – whether to suppress a warning
+when there are discrepancies between the train and eval version of
+the graph. Defaults to False.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="c1"># Feature extraction with resnet</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">resnet18</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># extract layer1 and layer3, giving as names `feat1` and feat2`</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">create_feature_extractor</span><span class="p">(</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">model</span><span class="p">,</span> <span class="p">{</span><span class="s1">&#39;layer1&#39;</span><span class="p">:</span> <span class="s1">&#39;feat1&#39;</span><span class="p">,</span> <span class="s1">&#39;layer3&#39;</span><span class="p">:</span> <span class="s1">&#39;feat2&#39;</span><span class="p">})</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">out</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">))</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="nb">print</span><span class="p">([(</span><span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">out</span><span class="o">.</span><span class="n">items</span><span class="p">()])</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="p">[(</span><span class="s1">&#39;feat1&#39;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Size</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">56</span><span class="p">,</span> <span class="mi">56</span><span class="p">])),</span>
+<span class="gp">&gt;&gt;&gt; </span>     <span class="p">(</span><span class="s1">&#39;feat2&#39;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Size</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">14</span><span class="p">,</span> <span class="mi">14</span><span class="p">]))]</span>
+
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># Specifying leaf modules and leaf functions</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="k">def</span> <span class="nf">leaf_function</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="c1"># This would raise a TypeError if traced through</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="k">return</span> <span class="nb">int</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="k">class</span> <span class="nc">LeafModule</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="c1"># This would raise a TypeError if traced through</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="nb">int</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="n">x</span> <span class="o">+</span> <span class="mi">4</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="k">class</span> <span class="nc">MyModule</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="bp">self</span><span class="o">.</span><span class="n">leaf_module</span> <span class="o">=</span> <span class="n">LeafModule</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="n">leaf_function</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span>        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">leaf_module</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">create_feature_extractor</span><span class="p">(</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">MyModule</span><span class="p">(),</span> <span class="n">return_nodes</span><span class="o">=</span><span class="p">[</span><span class="s1">&#39;leaf_module&#39;</span><span class="p">],</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">tracer_kwargs</span><span class="o">=</span><span class="p">{</span><span class="s1">&#39;leaf_modules&#39;</span><span class="p">:</span> <span class="p">[</span><span class="n">LeafModule</span><span class="p">],</span>
+<span class="gp">&gt;&gt;&gt; </span>                   <span class="s1">&#39;autowrap_functions&#39;</span><span class="p">:</span> <span class="p">[</span><span class="n">leaf_function</span><span class="p">]})</span>
+</pre></div>
+</div>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.feature_extraction.get_graph_node_names.html" class="btn btn-neutral float-right" title="get_graph_node_names" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../feature_extraction.html" class="btn btn-neutral" title="Feature extraction for model inspection" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">create_feature_extractor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html b/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
new file mode 100644
index 00000000000..c6e07f8ae2c
--- /dev/null
+++ b/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
@@ -0,0 +1,739 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>get_graph_node_names &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Examples and tutorials" href="../auto_examples/index.html" />
+    <link rel="prev" title="create_feature_extractor" href="torchvision.models.feature_extraction.create_feature_extractor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../feature_extraction.html">Feature extraction for model inspection</a> &gt;</li>
+        
+      <li>get_graph_node_names</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.models.feature_extraction.get_graph_node_names.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="get-graph-node-names">
+<h1>get_graph_node_names<a class="headerlink" href="#get-graph-node-names" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.feature_extraction.get_graph_node_names">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.feature_extraction.</span></span><span class="sig-name descname"><span class="pre">get_graph_node_names</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><span class="pre">Module</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tracer_kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">suppress_diff_warning</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/models/feature_extraction.html#get_graph_node_names"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.feature_extraction.get_graph_node_names" title="Permalink to this definition">¶</a></dt>
+<dd><p>Dev utility to return node names in order of execution. See note on node
+names under <a class="reference internal" href="torchvision.models.feature_extraction.create_feature_extractor.html#torchvision.models.feature_extraction.create_feature_extractor" title="torchvision.models.feature_extraction.create_feature_extractor"><code class="xref py py-func docutils literal notranslate"><span class="pre">create_feature_extractor()</span></code></a>. Useful for seeing which node
+names are available for feature extraction. There are two reasons that
+node names can’t easily be read directly from the code for a model:</p>
+<blockquote>
+<div><ol class="arabic simple">
+<li><p>Not all submodules are traced through. Modules from <code class="docutils literal notranslate"><span class="pre">torch.nn</span></code> all
+fall within this category.</p></li>
+<li><p>Nodes representing the repeated application of the same operation
+or leaf module get a <code class="docutils literal notranslate"><span class="pre">_{counter}</span></code> postfix.</p></li>
+</ol>
+</div></blockquote>
+<p>The model is traced twice: once in train mode, and once in eval mode. Both
+sets of node names are returned.</p>
+<p>For more details on the node naming conventions used here, please see the
+<a class="reference internal" href="../feature_extraction.html#about-node-names"><span class="std std-ref">relevant subheading</span></a> in the
+<a class="reference external" href="https://pytorch.org/vision/stable/feature_extraction.html">documentation</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>model</strong> (<em>nn.Module</em>) – model for which we’d like to print node names</p></li>
+<li><p><strong>tracer_kwargs</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – a dictionary of keyword arguments for
+<code class="docutils literal notranslate"><span class="pre">NodePathTracer</span></code> (they are eventually passed onto
+<a class="reference external" href="https://pytorch.org/docs/stable/fx.html#torch.fx.Tracer">torch.fx.Tracer</a>).
+By default, it will be set to wrap and make leaf nodes all torchvision ops:
+{“autowrap_modules”: (math, torchvision.ops,),”leaf_modules”: _get_leaf_modules_for_ops(),}
+WARNING: In case the user provides tracer_kwargs, above default arguments will be appended to the user
+provided dictionary.</p></li>
+<li><p><strong>suppress_diff_warning</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – whether to suppress a warning
+when there are discrepancies between the train and eval version of
+the graph. Defaults to False.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>a list of node names from tracing the model in
+train mode, and another from tracing the model in eval mode.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a>(<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)">list</a>, <a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)">list</a>)</p>
+</dd>
+</dl>
+<p>Examples:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">resnet18</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">train_nodes</span><span class="p">,</span> <span class="n">eval_nodes</span> <span class="o">=</span> <span class="n">get_graph_node_names</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+</pre></div>
+</div>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../auto_examples/index.html" class="btn btn-neutral float-right" title="Examples and tutorials" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.feature_extraction.create_feature_extractor.html" class="btn btn-neutral" title="create_feature_extractor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">get_graph_node_names</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.models.get_model.html b/0.16/generated/torchvision.models.get_model.html
new file mode 100644
index 00000000000..6e3b44d88a7
--- /dev/null
+++ b/0.16/generated/torchvision.models.get_model.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>get_model &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="get_model_weights" href="torchvision.models.get_model_weights.html" />
+    <link rel="prev" title="Models and pre-trained weights" href="../models.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>get_model</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.models.get_model.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="get-model">
+<h1>get_model<a class="headerlink" href="#get-model" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.get_model">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">get_model</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><span class="pre">Module</span></a></span></span><a class="reference internal" href="../_modules/torchvision/models/_api.html#get_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.get_model" title="Permalink to this definition">¶</a></dt>
+<dd><p>Gets the model name and configuration and returns an instantiated model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>name</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – The name under which the model is registered.</p></li>
+<li><p><strong>**config</strong> (<em>Any</em>) – parameters passed to the model builder method.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The initialized model.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>model (nn.Module)</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">get_model</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.get_model_weights.html" class="btn btn-neutral float-right" title="get_model_weights" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../models.html" class="btn btn-neutral" title="Models and pre-trained weights" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">get_model</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.models.get_model_weights.html b/0.16/generated/torchvision.models.get_model_weights.html
new file mode 100644
index 00000000000..cd1fea6ff4e
--- /dev/null
+++ b/0.16/generated/torchvision.models.get_model_weights.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>get_model_weights &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="get_weight" href="torchvision.models.get_weight.html" />
+    <link rel="prev" title="get_model" href="torchvision.models.get_model.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>get_model_weights</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.models.get_model_weights.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="get-model-weights">
+<h1>get_model_weights<a class="headerlink" href="#get-model-weights" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.get_model_weights">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">get_model_weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">[</span></span><span class="pre">WeightsEnum</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/models/_api.html#get_model_weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.get_model_weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns the weights enum class associated to the given model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>name</strong> (<em>callable</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – The model builder function or the name under which it is registered.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The weights enum class associated with the model.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>weights_enum (WeightsEnum)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.get_weight.html" class="btn btn-neutral float-right" title="get_weight" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.get_model.html" class="btn btn-neutral" title="get_model" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">get_model_weights</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.models.get_weight.html b/0.16/generated/torchvision.models.get_weight.html
new file mode 100644
index 00000000000..0ed2839d831
--- /dev/null
+++ b/0.16/generated/torchvision.models.get_weight.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>get_weight &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="list_models" href="torchvision.models.list_models.html" />
+    <link rel="prev" title="get_model_weights" href="torchvision.models.get_model_weights.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>get_weight</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.models.get_weight.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="get-weight">
+<h1>get_weight<a class="headerlink" href="#get-weight" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.get_weight">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">get_weight</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">WeightsEnum</span></span></span><a class="reference internal" href="../_modules/torchvision/models/_api.html#get_weight"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.get_weight" title="Permalink to this definition">¶</a></dt>
+<dd><p>Gets the weights enum value by its full name. Example: “ResNet50_Weights.IMAGENET1K_V1”</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>name</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – The name of the weight enum entry.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The requested weight enum.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>WeightsEnum</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.list_models.html" class="btn btn-neutral float-right" title="list_models" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.get_model_weights.html" class="btn btn-neutral" title="get_model_weights" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">get_weight</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.models.list_models.html b/0.16/generated/torchvision.models.list_models.html
new file mode 100644
index 00000000000..efe02d02c29
--- /dev/null
+++ b/0.16/generated/torchvision.models.list_models.html
@@ -0,0 +1,713 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>list_models &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="AlexNet" href="../models/alexnet.html" />
+    <link rel="prev" title="get_weight" href="torchvision.models.get_weight.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>list_models</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.models.list_models.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="list-models">
+<h1>list_models<a class="headerlink" href="#list-models" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.list_models">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">list_models</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">module</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><span class="pre">module</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">include</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Iterable" title="(in Python v3.12)"><span class="pre">Iterable</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exclude</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Iterable" title="(in Python v3.12)"><span class="pre">Iterable</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/models/_api.html#list_models"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.list_models" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns a list with the names of registered models.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>module</strong> (<em>ModuleType</em><em>, </em><em>optional</em>) – The module from which we want to extract the available models.</p></li>
+<li><p><strong>include</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em> or </em><em>Iterable</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>]</em><em>, </em><em>optional</em>) – Filter(s) for including the models from the set of all models.
+Filters are passed to <a class="reference external" href="https://docs.python.org/3/library/fnmatch.html">fnmatch</a> to match Unix shell-style
+wildcards. In case of many filters, the results is the union of individual filters.</p></li>
+<li><p><strong>exclude</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em> or </em><em>Iterable</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>]</em><em>, </em><em>optional</em>) – Filter(s) applied after include_filters to remove models.
+Filter are passed to <a class="reference external" href="https://docs.python.org/3/library/fnmatch.html">fnmatch</a> to match Unix shell-style
+wildcards. In case of many filters, the results is removal of all the models that match any individual filter.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>A list with the names of available models.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>models (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)">list</a>)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../models/alexnet.html" class="btn btn-neutral float-right" title="AlexNet" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.get_weight.html" class="btn btn-neutral" title="get_weight" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">list_models</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.Conv2dNormActivation.html b/0.16/generated/torchvision.ops.Conv2dNormActivation.html
new file mode 100644
index 00000000000..d15274935a4
--- /dev/null
+++ b/0.16/generated/torchvision.ops.Conv2dNormActivation.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Conv2dNormActivation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Conv3dNormActivation" href="torchvision.ops.Conv3dNormActivation.html" />
+    <link rel="prev" title="sigmoid_focal_loss" href="torchvision.ops.sigmoid_focal_loss.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>Conv2dNormActivation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.Conv2dNormActivation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="conv2dnormactivation">
+<h1>Conv2dNormActivation<a class="headerlink" href="#conv2dnormactivation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.Conv2dNormActivation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">Conv2dNormActivation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">in_channels:</span> <span class="pre">int,</span> <span class="pre">out_channels:</span> <span class="pre">int,</span> <span class="pre">kernel_size:</span> <span class="pre">~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int]]</span> <span class="pre">=</span> <span class="pre">3,</span> <span class="pre">stride:</span> <span class="pre">~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int]]</span> <span class="pre">=</span> <span class="pre">1,</span> <span class="pre">padding:</span> <span class="pre">~typing.Optional[~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int],</span> <span class="pre">str]]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">groups:</span> <span class="pre">int</span> <span class="pre">=</span> <span class="pre">1,</span> <span class="pre">norm_layer:</span> <span class="pre">~typing.Optional[~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]]</span> <span class="pre">=</span> <span class="pre">&lt;class</span> <span class="pre">'torch.nn.modules.batchnorm.BatchNorm2d'&gt;,</span> <span class="pre">activation_layer:</span> <span class="pre">~typing.Optional[~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]]</span> <span class="pre">=</span> <span class="pre">&lt;class</span> <span class="pre">'torch.nn.modules.activation.ReLU'&gt;,</span> <span class="pre">dilation:</span> <span class="pre">~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int]]</span> <span class="pre">=</span> <span class="pre">1,</span> <span class="pre">inplace:</span> <span class="pre">~typing.Optional[bool]</span> <span class="pre">=</span> <span class="pre">True,</span> <span class="pre">bias:</span> <span class="pre">~typing.Optional[bool]</span> <span class="pre">=</span> <span class="pre">None</span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#Conv2dNormActivation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.Conv2dNormActivation" title="Permalink to this definition">¶</a></dt>
+<dd><p>Configurable block used for Convolution2d-Normalization-Activation blocks.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>in_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of channels in the input image</p></li>
+<li><p><strong>out_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of channels produced by the Convolution-Normalization-Activation block</p></li>
+<li><p><strong>kernel_size</strong> – (int, optional): Size of the convolving kernel. Default: 3</p></li>
+<li><p><strong>stride</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Stride of the convolution. Default: 1</p></li>
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – Padding added to all four sides of the input. Default: None, in which case it will be calculated as <code class="docutils literal notranslate"><span class="pre">padding</span> <span class="pre">=</span> <span class="pre">(kernel_size</span> <span class="pre">-</span> <span class="pre">1)</span> <span class="pre">//</span> <span class="pre">2</span> <span class="pre">*</span> <span class="pre">dilation</span></code></p></li>
+<li><p><strong>groups</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Number of blocked connections from input channels to output channels. Default: 1</p></li>
+<li><p><strong>norm_layer</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em><em>, </em><em>optional</em>) – Norm layer that will be stacked on top of the convolution layer. If <code class="docutils literal notranslate"><span class="pre">None</span></code> this layer won’t be used. Default: <code class="docutils literal notranslate"><span class="pre">torch.nn.BatchNorm2d</span></code></p></li>
+<li><p><strong>activation_layer</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em><em>, </em><em>optional</em>) – Activation function which will be stacked on top of the normalization layer (if not None), otherwise on top of the conv layer. If <code class="docutils literal notranslate"><span class="pre">None</span></code> this layer won’t be used. Default: <code class="docutils literal notranslate"><span class="pre">torch.nn.ReLU</span></code></p></li>
+<li><p><strong>dilation</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Spacing between kernel elements. Default: 1</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – Parameter for the activation layer, which can optionally do the operation in-place. Default <code class="docutils literal notranslate"><span class="pre">True</span></code></p></li>
+<li><p><strong>bias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Whether to use bias in the convolution layer. By default, biases are included if <code class="docutils literal notranslate"><span class="pre">norm_layer</span> <span class="pre">is</span> <span class="pre">None</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.Conv3dNormActivation.html" class="btn btn-neutral float-right" title="Conv3dNormActivation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.sigmoid_focal_loss.html" class="btn btn-neutral" title="sigmoid_focal_loss" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Conv2dNormActivation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.Conv3dNormActivation.html b/0.16/generated/torchvision.ops.Conv3dNormActivation.html
new file mode 100644
index 00000000000..4f77e20bd81
--- /dev/null
+++ b/0.16/generated/torchvision.ops.Conv3dNormActivation.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Conv3dNormActivation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="DeformConv2d" href="torchvision.ops.DeformConv2d.html" />
+    <link rel="prev" title="Conv2dNormActivation" href="torchvision.ops.Conv2dNormActivation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>Conv3dNormActivation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.Conv3dNormActivation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="conv3dnormactivation">
+<h1>Conv3dNormActivation<a class="headerlink" href="#conv3dnormactivation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.Conv3dNormActivation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">Conv3dNormActivation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">in_channels:</span> <span class="pre">int,</span> <span class="pre">out_channels:</span> <span class="pre">int,</span> <span class="pre">kernel_size:</span> <span class="pre">~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int,</span> <span class="pre">int]]</span> <span class="pre">=</span> <span class="pre">3,</span> <span class="pre">stride:</span> <span class="pre">~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int,</span> <span class="pre">int]]</span> <span class="pre">=</span> <span class="pre">1,</span> <span class="pre">padding:</span> <span class="pre">~typing.Optional[~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int,</span> <span class="pre">int],</span> <span class="pre">str]]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">groups:</span> <span class="pre">int</span> <span class="pre">=</span> <span class="pre">1,</span> <span class="pre">norm_layer:</span> <span class="pre">~typing.Optional[~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]]</span> <span class="pre">=</span> <span class="pre">&lt;class</span> <span class="pre">'torch.nn.modules.batchnorm.BatchNorm3d'&gt;,</span> <span class="pre">activation_layer:</span> <span class="pre">~typing.Optional[~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]]</span> <span class="pre">=</span> <span class="pre">&lt;class</span> <span class="pre">'torch.nn.modules.activation.ReLU'&gt;,</span> <span class="pre">dilation:</span> <span class="pre">~typing.Union[int,</span> <span class="pre">~typing.Tuple[int,</span> <span class="pre">int,</span> <span class="pre">int]]</span> <span class="pre">=</span> <span class="pre">1,</span> <span class="pre">inplace:</span> <span class="pre">~typing.Optional[bool]</span> <span class="pre">=</span> <span class="pre">True,</span> <span class="pre">bias:</span> <span class="pre">~typing.Optional[bool]</span> <span class="pre">=</span> <span class="pre">None</span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#Conv3dNormActivation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.Conv3dNormActivation" title="Permalink to this definition">¶</a></dt>
+<dd><p>Configurable block used for Convolution3d-Normalization-Activation blocks.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>in_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of channels in the input video.</p></li>
+<li><p><strong>out_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of channels produced by the Convolution-Normalization-Activation block</p></li>
+<li><p><strong>kernel_size</strong> – (int, optional): Size of the convolving kernel. Default: 3</p></li>
+<li><p><strong>stride</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Stride of the convolution. Default: 1</p></li>
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – Padding added to all four sides of the input. Default: None, in which case it will be calculated as <code class="docutils literal notranslate"><span class="pre">padding</span> <span class="pre">=</span> <span class="pre">(kernel_size</span> <span class="pre">-</span> <span class="pre">1)</span> <span class="pre">//</span> <span class="pre">2</span> <span class="pre">*</span> <span class="pre">dilation</span></code></p></li>
+<li><p><strong>groups</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Number of blocked connections from input channels to output channels. Default: 1</p></li>
+<li><p><strong>norm_layer</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em><em>, </em><em>optional</em>) – Norm layer that will be stacked on top of the convolution layer. If <code class="docutils literal notranslate"><span class="pre">None</span></code> this layer won’t be used. Default: <code class="docutils literal notranslate"><span class="pre">torch.nn.BatchNorm3d</span></code></p></li>
+<li><p><strong>activation_layer</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em><em>, </em><em>optional</em>) – Activation function which will be stacked on top of the normalization layer (if not None), otherwise on top of the conv layer. If <code class="docutils literal notranslate"><span class="pre">None</span></code> this layer won’t be used. Default: <code class="docutils literal notranslate"><span class="pre">torch.nn.ReLU</span></code></p></li>
+<li><p><strong>dilation</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Spacing between kernel elements. Default: 1</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – Parameter for the activation layer, which can optionally do the operation in-place. Default <code class="docutils literal notranslate"><span class="pre">True</span></code></p></li>
+<li><p><strong>bias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Whether to use bias in the convolution layer. By default, biases are included if <code class="docutils literal notranslate"><span class="pre">norm_layer</span> <span class="pre">is</span> <span class="pre">None</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.DeformConv2d.html" class="btn btn-neutral float-right" title="DeformConv2d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.Conv2dNormActivation.html" class="btn btn-neutral" title="Conv2dNormActivation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Conv3dNormActivation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.DeformConv2d.html b/0.16/generated/torchvision.ops.DeformConv2d.html
new file mode 100644
index 00000000000..dcf27776e33
--- /dev/null
+++ b/0.16/generated/torchvision.ops.DeformConv2d.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>DeformConv2d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="DropBlock2d" href="torchvision.ops.DropBlock2d.html" />
+    <link rel="prev" title="Conv3dNormActivation" href="torchvision.ops.Conv3dNormActivation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>DeformConv2d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.DeformConv2d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="deformconv2d">
+<h1>DeformConv2d<a class="headerlink" href="#deformconv2d" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.DeformConv2d">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">DeformConv2d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">in_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">out_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">kernel_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dilation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">groups</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/deform_conv.html#DeformConv2d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.DeformConv2d" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.deform_conv2d.html#torchvision.ops.deform_conv2d" title="torchvision.ops.deform_conv2d"><code class="xref py py-func docutils literal notranslate"><span class="pre">deform_conv2d()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.DeformConv2d.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">offset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/deform_conv.html#DeformConv2d.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.DeformConv2d.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>batch_size</em><em>, </em><em>in_channels</em><em>, </em><em>in_height</em><em>, </em><em>in_width</em><em>]</em>) – input tensor</p></li>
+<li><p><strong>offset</strong> (<em>Tensor</em><em>[</em><em>batch_size</em><em>, </em><em>2 * offset_groups * kernel_height * kernel_width</em><em>, </em><em>out_height</em><em>, </em><em>out_width</em><em>]</em>) – offsets to be applied for each position in the convolution kernel.</p></li>
+<li><p><strong>mask</strong> (<em>Tensor</em><em>[</em><em>batch_size</em><em>, </em><em>offset_groups * kernel_height * kernel_width</em><em>, </em><em>out_height</em><em>, </em><em>out_width</em><em>]</em>) – masks to be applied for each position in the convolution kernel.</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.DropBlock2d.html" class="btn btn-neutral float-right" title="DropBlock2d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.Conv3dNormActivation.html" class="btn btn-neutral" title="Conv3dNormActivation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">DeformConv2d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.DropBlock2d.html b/0.16/generated/torchvision.ops.DropBlock2d.html
new file mode 100644
index 00000000000..91a06b0fd94
--- /dev/null
+++ b/0.16/generated/torchvision.ops.DropBlock2d.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>DropBlock2d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="DropBlock3d" href="torchvision.ops.DropBlock3d.html" />
+    <link rel="prev" title="DeformConv2d" href="torchvision.ops.DeformConv2d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>DropBlock2d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.DropBlock2d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="dropblock2d">
+<h1>DropBlock2d<a class="headerlink" href="#dropblock2d" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.DropBlock2d">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">DropBlock2d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-06</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/drop_block.html#DropBlock2d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.DropBlock2d" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.drop_block2d.html#torchvision.ops.drop_block2d" title="torchvision.ops.drop_block2d"><code class="xref py py-func docutils literal notranslate"><span class="pre">drop_block2d()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.DropBlock2d.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/drop_block.html#DropBlock2d.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.DropBlock2d.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>input</strong> (<em>Tensor</em>) – Input feature map on which some areas will be randomly
+dropped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The tensor after DropBlock layer.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.DropBlock3d.html" class="btn btn-neutral float-right" title="DropBlock3d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.DeformConv2d.html" class="btn btn-neutral" title="DeformConv2d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">DropBlock2d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.DropBlock3d.html b/0.16/generated/torchvision.ops.DropBlock3d.html
new file mode 100644
index 00000000000..cd44591a1ec
--- /dev/null
+++ b/0.16/generated/torchvision.ops.DropBlock3d.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>DropBlock3d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FrozenBatchNorm2d" href="torchvision.ops.FrozenBatchNorm2d.html" />
+    <link rel="prev" title="DropBlock2d" href="torchvision.ops.DropBlock2d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>DropBlock3d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.DropBlock3d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="dropblock3d">
+<h1>DropBlock3d<a class="headerlink" href="#dropblock3d" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.DropBlock3d">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">DropBlock3d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-06</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/drop_block.html#DropBlock3d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.DropBlock3d" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.drop_block3d.html#torchvision.ops.drop_block3d" title="torchvision.ops.drop_block3d"><code class="xref py py-func docutils literal notranslate"><span class="pre">drop_block3d()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.DropBlock3d.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/drop_block.html#DropBlock3d.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.DropBlock3d.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>input</strong> (<em>Tensor</em>) – Input feature map on which some areas will be randomly
+dropped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The tensor after DropBlock layer.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.FrozenBatchNorm2d.html" class="btn btn-neutral float-right" title="FrozenBatchNorm2d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.DropBlock2d.html" class="btn btn-neutral" title="DropBlock2d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">DropBlock3d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html b/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
new file mode 100644
index 00000000000..fc019a3a548
--- /dev/null
+++ b/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
@@ -0,0 +1,764 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FeaturePyramidNetwork &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="MultiScaleRoIAlign" href="torchvision.ops.MultiScaleRoIAlign.html" />
+    <link rel="prev" title="ps_roi_pool" href="torchvision.ops.ps_roi_pool.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>FeaturePyramidNetwork</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.FeaturePyramidNetwork.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="featurepyramidnetwork">
+<h1>FeaturePyramidNetwork<a class="headerlink" href="#featurepyramidnetwork" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.FeaturePyramidNetwork">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">FeaturePyramidNetwork</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">in_channels_list</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">out_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">extra_blocks</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><span class="pre">ExtraFPNBlock</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">norm_layer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><span class="pre">Module</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/feature_pyramid_network.html#FeaturePyramidNetwork"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.FeaturePyramidNetwork" title="Permalink to this definition">¶</a></dt>
+<dd><p>Module that adds a FPN from on top of a set of feature maps. This is based on
+<a class="reference external" href="https://arxiv.org/abs/1612.03144">“Feature Pyramid Network for Object Detection”</a>.</p>
+<p>The feature maps are currently supposed to be in increasing depth
+order.</p>
+<p>The input to the model is expected to be an OrderedDict[Tensor], containing
+the feature maps on top of which the FPN will be added.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>in_channels_list</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – number of channels for each feature map that
+is passed to the module</p></li>
+<li><p><strong>out_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of channels of the FPN representation</p></li>
+<li><p><strong>extra_blocks</strong> (<em>ExtraFPNBlock</em><em> or </em><em>None</em>) – if provided, extra operations will
+be performed. It is expected to take the fpn features, the original
+features and the names of the original features as input, and returns
+a new list of feature maps and their corresponding names</p></li>
+<li><p><strong>norm_layer</strong> (<em>callable</em><em>, </em><em>optional</em>) – Module specifying the normalization layer to use. Default: None</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">m</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">FeaturePyramidNetwork</span><span class="p">([</span><span class="mi">10</span><span class="p">,</span> <span class="mi">20</span><span class="p">,</span> <span class="mi">30</span><span class="p">],</span> <span class="mi">5</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># get some dummy data</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span><span class="p">[</span><span class="s1">&#39;feat0&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span><span class="p">[</span><span class="s1">&#39;feat2&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">20</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">16</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span><span class="p">[</span><span class="s1">&#39;feat3&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">30</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># compute the FPN on top of x</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">output</span> <span class="o">=</span> <span class="n">m</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="nb">print</span><span class="p">([(</span><span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">output</span><span class="o">.</span><span class="n">items</span><span class="p">()])</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># returns</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="p">[(</span><span class="s1">&#39;feat0&#39;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Size</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">])),</span>
+<span class="gp">&gt;&gt;&gt; </span>   <span class="p">(</span><span class="s1">&#39;feat2&#39;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Size</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">16</span><span class="p">])),</span>
+<span class="gp">&gt;&gt;&gt; </span>   <span class="p">(</span><span class="s1">&#39;feat3&#39;</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Size</span><span class="p">([</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">]))]</span>
+</pre></div>
+</div>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.FeaturePyramidNetwork.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/ops/feature_pyramid_network.html#FeaturePyramidNetwork.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.FeaturePyramidNetwork.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Computes the FPN for a set of feature maps.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>x</strong> (<em>OrderedDict</em><em>[</em><em>Tensor</em><em>]</em>) – feature maps for each feature level.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><dl class="simple">
+<dt>feature maps after FPN layers.</dt><dd><p>They are ordered from the highest resolution first.</p>
+</dd>
+</dl>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>results (OrderedDict[Tensor])</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.FeaturePyramidNetwork.get_result_from_inner_blocks">
+<span class="sig-name descname"><span class="pre">get_result_from_inner_blocks</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/feature_pyramid_network.html#FeaturePyramidNetwork.get_result_from_inner_blocks"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.FeaturePyramidNetwork.get_result_from_inner_blocks" title="Permalink to this definition">¶</a></dt>
+<dd><p>This is equivalent to self.inner_blocks[idx](x),
+but torchscript doesn’t support this yet</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.FeaturePyramidNetwork.get_result_from_layer_blocks">
+<span class="sig-name descname"><span class="pre">get_result_from_layer_blocks</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">idx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/feature_pyramid_network.html#FeaturePyramidNetwork.get_result_from_layer_blocks"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.FeaturePyramidNetwork.get_result_from_layer_blocks" title="Permalink to this definition">¶</a></dt>
+<dd><p>This is equivalent to self.layer_blocks[idx](x),
+but torchscript doesn’t support this yet</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.MultiScaleRoIAlign.html" class="btn btn-neutral float-right" title="MultiScaleRoIAlign" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.ps_roi_pool.html" class="btn btn-neutral" title="ps_roi_pool" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FeaturePyramidNetwork</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html b/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
new file mode 100644
index 00000000000..95632534a2d
--- /dev/null
+++ b/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FrozenBatchNorm2d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="MLP" href="torchvision.ops.MLP.html" />
+    <link rel="prev" title="DropBlock3d" href="torchvision.ops.DropBlock3d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>FrozenBatchNorm2d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.FrozenBatchNorm2d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="frozenbatchnorm2d">
+<h1>FrozenBatchNorm2d<a class="headerlink" href="#frozenbatchnorm2d" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.FrozenBatchNorm2d">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">FrozenBatchNorm2d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_features</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-05</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#FrozenBatchNorm2d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.FrozenBatchNorm2d" title="Permalink to this definition">¶</a></dt>
+<dd><p>BatchNorm2d where the batch statistics and the affine parameters are fixed</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>num_features</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of features <code class="docutils literal notranslate"><span class="pre">C</span></code> from an expected input of size <code class="docutils literal notranslate"><span class="pre">(N,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code></p></li>
+<li><p><strong>eps</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – a value added to the denominator for numerical stability. Default: 1e-5</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.FrozenBatchNorm2d.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#FrozenBatchNorm2d.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.FrozenBatchNorm2d.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.MLP.html" class="btn btn-neutral float-right" title="MLP" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.DropBlock3d.html" class="btn btn-neutral" title="DropBlock3d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FrozenBatchNorm2d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.MLP.html b/0.16/generated/torchvision.ops.MLP.html
new file mode 100644
index 00000000000..507415ffebf
--- /dev/null
+++ b/0.16/generated/torchvision.ops.MLP.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MLP &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Permute" href="torchvision.ops.Permute.html" />
+    <link rel="prev" title="FrozenBatchNorm2d" href="torchvision.ops.FrozenBatchNorm2d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>MLP</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.MLP.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mlp">
+<h1>MLP<a class="headerlink" href="#mlp" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.MLP">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">MLP</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">in_channels:</span> <span class="pre">int,</span> <span class="pre">hidden_channels:</span> <span class="pre">~typing.List[int],</span> <span class="pre">norm_layer:</span> <span class="pre">~typing.Optional[~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">activation_layer:</span> <span class="pre">~typing.Optional[~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]]</span> <span class="pre">=</span> <span class="pre">&lt;class</span> <span class="pre">'torch.nn.modules.activation.ReLU'&gt;,</span> <span class="pre">inplace:</span> <span class="pre">~typing.Optional[bool]</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">bias:</span> <span class="pre">bool</span> <span class="pre">=</span> <span class="pre">True,</span> <span class="pre">dropout:</span> <span class="pre">float</span> <span class="pre">=</span> <span class="pre">0.0</span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#MLP"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.MLP" title="Permalink to this definition">¶</a></dt>
+<dd><p>This block implements the multi-layer perceptron (MLP) module.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>in_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of channels of the input</p></li>
+<li><p><strong>hidden_channels</strong> (<em>List</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – List of the hidden channel dimensions</p></li>
+<li><p><strong>norm_layer</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em><em>, </em><em>optional</em>) – Norm layer that will be stacked on top of the linear layer. If <code class="docutils literal notranslate"><span class="pre">None</span></code> this layer won’t be used. Default: <code class="docutils literal notranslate"><span class="pre">None</span></code></p></li>
+<li><p><strong>activation_layer</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em><em>, </em><em>optional</em>) – Activation function which will be stacked on top of the normalization layer (if not None), otherwise on top of the linear layer. If <code class="docutils literal notranslate"><span class="pre">None</span></code> this layer won’t be used. Default: <code class="docutils literal notranslate"><span class="pre">torch.nn.ReLU</span></code></p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Parameter for the activation layer, which can optionally do the operation in-place.
+Default is <code class="docutils literal notranslate"><span class="pre">None</span></code>, which uses the respective default values of the <code class="docutils literal notranslate"><span class="pre">activation_layer</span></code> and Dropout layer.</p></li>
+<li><p><strong>bias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – Whether to use bias in the linear layer. Default <code class="docutils literal notranslate"><span class="pre">True</span></code></p></li>
+<li><p><strong>dropout</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – The probability for the dropout layer. Default: 0.0</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.Permute.html" class="btn btn-neutral float-right" title="Permute" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.FrozenBatchNorm2d.html" class="btn btn-neutral" title="FrozenBatchNorm2d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MLP</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html b/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
new file mode 100644
index 00000000000..2b11819319e
--- /dev/null
+++ b/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
@@ -0,0 +1,749 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MultiScaleRoIAlign &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RoIAlign" href="torchvision.ops.RoIAlign.html" />
+    <link rel="prev" title="FeaturePyramidNetwork" href="torchvision.ops.FeaturePyramidNetwork.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>MultiScaleRoIAlign</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.MultiScaleRoIAlign.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="multiscaleroialign">
+<h1>MultiScaleRoIAlign<a class="headerlink" href="#multiscaleroialign" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.MultiScaleRoIAlign">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">MultiScaleRoIAlign</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">featmap_names</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">canonical_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">224</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">canonical_level</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">4</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/poolers.html#MultiScaleRoIAlign"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.MultiScaleRoIAlign" title="Permalink to this definition">¶</a></dt>
+<dd><p>Multi-scale RoIAlign pooling, which is useful for detection with or without FPN.</p>
+<p>It infers the scale of the pooling via the heuristics specified in eq. 1
+of the <a class="reference external" href="https://arxiv.org/abs/1612.03144">Feature Pyramid Network paper</a>.
+They keyword-only parameters <code class="docutils literal notranslate"><span class="pre">canonical_scale</span></code> and <code class="docutils literal notranslate"><span class="pre">canonical_level</span></code>
+correspond respectively to <code class="docutils literal notranslate"><span class="pre">224</span></code> and <code class="docutils literal notranslate"><span class="pre">k0=4</span></code> in eq. 1, and
+have the following meaning: <code class="docutils literal notranslate"><span class="pre">canonical_level</span></code> is the target level of the pyramid from
+which to pool a region of interest with <code class="docutils literal notranslate"><span class="pre">w</span> <span class="pre">x</span> <span class="pre">h</span> <span class="pre">=</span> <span class="pre">canonical_scale</span> <span class="pre">x</span> <span class="pre">canonical_scale</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>featmap_names</strong> (<em>List</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>]</em>) – the names of the feature maps that will be used
+for the pooling.</p></li>
+<li><p><strong>output_size</strong> (<em>List</em><em>[</em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em><em>] or </em><em>List</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – output size for the pooled region</p></li>
+<li><p><strong>sampling_ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – sampling ratio for ROIAlign</p></li>
+<li><p><strong>canonical_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – canonical_scale for LevelMapper</p></li>
+<li><p><strong>canonical_level</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – canonical_level for LevelMapper</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">m</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">ops</span><span class="o">.</span><span class="n">MultiScaleRoIAlign</span><span class="p">([</span><span class="s1">&#39;feat1&#39;</span><span class="p">,</span> <span class="s1">&#39;feat3&#39;</span><span class="p">],</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">i</span> <span class="o">=</span> <span class="n">OrderedDict</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;feat1&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">64</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;feat2&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span><span class="p">)</span>  <span class="c1"># this feature won&#39;t be used in the pooling</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;feat3&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">16</span><span class="p">,</span> <span class="mi">16</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># create some random bounding boxes</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">6</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span> <span class="o">*</span> <span class="mi">256</span><span class="p">;</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:]</span> <span class="o">+=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># original image size, before computing the feature maps</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">image_sizes</span> <span class="o">=</span> <span class="p">[(</span><span class="mi">512</span><span class="p">,</span> <span class="mi">512</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">output</span> <span class="o">=</span> <span class="n">m</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="p">[</span><span class="n">boxes</span><span class="p">],</span> <span class="n">image_sizes</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="nb">print</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">torch</span><span class="o">.</span><span class="n">Size</span><span class="p">([</span><span class="mi">6</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span><span class="p">])</span>
+</pre></div>
+</div>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.MultiScaleRoIAlign.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_shapes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/poolers.html#MultiScaleRoIAlign.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.MultiScaleRoIAlign.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>x</strong> (<em>OrderedDict</em><em>[</em><em>Tensor</em><em>]</em>) – feature maps for each level. They are assumed to have
+all the same number of channels, but they can have different sizes.</p></li>
+<li><p><strong>boxes</strong> (<em>List</em><em>[</em><em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em><em>]</em>) – boxes to be used to perform the pooling operation, in
+(x1, y1, x2, y2) format and in the image reference size, not the feature map
+reference. The coordinate must satisfy <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p></li>
+<li><p><strong>image_shapes</strong> (<em>List</em><em>[</em><em>Tuple</em><em>[</em><em>height</em><em>, </em><em>width</em><em>]</em><em>]</em>) – the sizes of each image before they
+have been fed to a CNN to obtain feature maps. This allows us to infer the
+scale factor for each one of the levels to be pooled.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>result (Tensor)</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.RoIAlign.html" class="btn btn-neutral float-right" title="RoIAlign" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.FeaturePyramidNetwork.html" class="btn btn-neutral" title="FeaturePyramidNetwork" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MultiScaleRoIAlign</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.PSRoIAlign.html b/0.16/generated/torchvision.ops.PSRoIAlign.html
new file mode 100644
index 00000000000..c6150accef0
--- /dev/null
+++ b/0.16/generated/torchvision.ops.PSRoIAlign.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>PSRoIAlign &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="PSRoIPool" href="torchvision.ops.PSRoIPool.html" />
+    <link rel="prev" title="RoIPool" href="torchvision.ops.RoIPool.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>PSRoIAlign</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.PSRoIAlign.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="psroialign">
+<h1>PSRoIAlign<a class="headerlink" href="#psroialign" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.PSRoIAlign">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">PSRoIAlign</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/ps_roi_align.html#PSRoIAlign"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.PSRoIAlign" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.ps_roi_align.html#torchvision.ops.ps_roi_align" title="torchvision.ops.ps_roi_align"><code class="xref py py-func docutils literal notranslate"><span class="pre">ps_roi_align()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.PSRoIAlign.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">rois</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/ps_roi_align.html#PSRoIAlign.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.PSRoIAlign.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.PSRoIPool.html" class="btn btn-neutral float-right" title="PSRoIPool" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.RoIPool.html" class="btn btn-neutral" title="RoIPool" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">PSRoIAlign</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.PSRoIPool.html b/0.16/generated/torchvision.ops.PSRoIPool.html
new file mode 100644
index 00000000000..93103ae7e3d
--- /dev/null
+++ b/0.16/generated/torchvision.ops.PSRoIPool.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>PSRoIPool &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="box_area" href="torchvision.ops.box_area.html" />
+    <link rel="prev" title="PSRoIAlign" href="torchvision.ops.PSRoIAlign.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>PSRoIPool</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.PSRoIPool.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="psroipool">
+<h1>PSRoIPool<a class="headerlink" href="#psroipool" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.PSRoIPool">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">PSRoIPool</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/ps_roi_pool.html#PSRoIPool"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.PSRoIPool" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.ps_roi_pool.html#torchvision.ops.ps_roi_pool" title="torchvision.ops.ps_roi_pool"><code class="xref py py-func docutils literal notranslate"><span class="pre">ps_roi_pool()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.PSRoIPool.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">rois</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/ps_roi_pool.html#PSRoIPool.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.PSRoIPool.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.box_area.html" class="btn btn-neutral float-right" title="box_area" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.PSRoIAlign.html" class="btn btn-neutral" title="PSRoIAlign" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">PSRoIPool</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.Permute.html b/0.16/generated/torchvision.ops.Permute.html
new file mode 100644
index 00000000000..0e9601a1835
--- /dev/null
+++ b/0.16/generated/torchvision.ops.Permute.html
@@ -0,0 +1,713 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Permute &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SqueezeExcitation" href="torchvision.ops.SqueezeExcitation.html" />
+    <link rel="prev" title="MLP" href="torchvision.ops.MLP.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>Permute</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.Permute.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="permute">
+<h1>Permute<a class="headerlink" href="#permute" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.Permute">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">Permute</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dims</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#Permute"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.Permute" title="Permalink to this definition">¶</a></dt>
+<dd><p>This module returns a view of the tensor input with its dimensions permuted.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>dims</strong> (<em>List</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – The desired ordering of dimensions</p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.Permute.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#Permute.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.Permute.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.SqueezeExcitation.html" class="btn btn-neutral float-right" title="SqueezeExcitation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.MLP.html" class="btn btn-neutral" title="MLP" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Permute</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.RoIAlign.html b/0.16/generated/torchvision.ops.RoIAlign.html
new file mode 100644
index 00000000000..4986169ed94
--- /dev/null
+++ b/0.16/generated/torchvision.ops.RoIAlign.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RoIAlign &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RoIPool" href="torchvision.ops.RoIPool.html" />
+    <link rel="prev" title="MultiScaleRoIAlign" href="torchvision.ops.MultiScaleRoIAlign.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>RoIAlign</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.RoIAlign.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="roialign">
+<h1>RoIAlign<a class="headerlink" href="#roialign" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.RoIAlign">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">RoIAlign</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">aligned</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/roi_align.html#RoIAlign"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.RoIAlign" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.roi_align.html#torchvision.ops.roi_align" title="torchvision.ops.roi_align"><code class="xref py py-func docutils literal notranslate"><span class="pre">roi_align()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.RoIAlign.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">rois</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/roi_align.html#RoIAlign.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.RoIAlign.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.RoIPool.html" class="btn btn-neutral float-right" title="RoIPool" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.MultiScaleRoIAlign.html" class="btn btn-neutral" title="MultiScaleRoIAlign" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RoIAlign</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.RoIPool.html b/0.16/generated/torchvision.ops.RoIPool.html
new file mode 100644
index 00000000000..e7d4df78a54
--- /dev/null
+++ b/0.16/generated/torchvision.ops.RoIPool.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RoIPool &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="PSRoIAlign" href="torchvision.ops.PSRoIAlign.html" />
+    <link rel="prev" title="RoIAlign" href="torchvision.ops.RoIAlign.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>RoIPool</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.RoIPool.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="roipool">
+<h1>RoIPool<a class="headerlink" href="#roipool" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.RoIPool">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">RoIPool</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/roi_pool.html#RoIPool"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.RoIPool" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.roi_pool.html#torchvision.ops.roi_pool" title="torchvision.ops.roi_pool"><code class="xref py py-func docutils literal notranslate"><span class="pre">roi_pool()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.RoIPool.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">rois</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/roi_pool.html#RoIPool.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.RoIPool.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.PSRoIAlign.html" class="btn btn-neutral float-right" title="PSRoIAlign" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.RoIAlign.html" class="btn btn-neutral" title="RoIAlign" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RoIPool</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.SqueezeExcitation.html b/0.16/generated/torchvision.ops.SqueezeExcitation.html
new file mode 100644
index 00000000000..371445f8893
--- /dev/null
+++ b/0.16/generated/torchvision.ops.SqueezeExcitation.html
@@ -0,0 +1,719 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SqueezeExcitation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="StochasticDepth" href="torchvision.ops.StochasticDepth.html" />
+    <link rel="prev" title="Permute" href="torchvision.ops.Permute.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>SqueezeExcitation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.SqueezeExcitation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="squeezeexcitation">
+<h1>SqueezeExcitation<a class="headerlink" href="#squeezeexcitation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.SqueezeExcitation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">SqueezeExcitation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">input_channels:</span> <span class="pre">int,</span> <span class="pre">squeeze_channels:</span> <span class="pre">int,</span> <span class="pre">activation:</span> <span class="pre">~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]</span> <span class="pre">=</span> <span class="pre">&lt;class</span> <span class="pre">'torch.nn.modules.activation.ReLU'&gt;,</span> <span class="pre">scale_activation:</span> <span class="pre">~typing.Callable[[...],</span> <span class="pre">~torch.nn.modules.module.Module]</span> <span class="pre">=</span> <span class="pre">&lt;class</span> <span class="pre">'torch.nn.modules.activation.Sigmoid'&gt;</span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#SqueezeExcitation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.SqueezeExcitation" title="Permalink to this definition">¶</a></dt>
+<dd><p>This block implements the Squeeze-and-Excitation block from <a class="reference external" href="https://arxiv.org/abs/1709.01507">https://arxiv.org/abs/1709.01507</a> (see Fig. 1).
+Parameters <code class="docutils literal notranslate"><span class="pre">activation</span></code>, and <code class="docutils literal notranslate"><span class="pre">scale_activation</span></code> correspond to <code class="docutils literal notranslate"><span class="pre">delta</span></code> and <code class="docutils literal notranslate"><span class="pre">sigma</span></code> in eq. 3.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of channels in the input image</p></li>
+<li><p><strong>squeeze_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of squeeze channels</p></li>
+<li><p><strong>activation</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em><em>, </em><em>optional</em>) – <code class="docutils literal notranslate"><span class="pre">delta</span></code> activation. Default: <code class="docutils literal notranslate"><span class="pre">torch.nn.ReLU</span></code></p></li>
+<li><p><strong>scale_activation</strong> (<em>Callable</em><em>[</em><em>...</em><em>, </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a><em>]</em>) – <code class="docutils literal notranslate"><span class="pre">sigma</span></code> activation. Default: <code class="docutils literal notranslate"><span class="pre">torch.nn.Sigmoid</span></code></p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.SqueezeExcitation.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/misc.html#SqueezeExcitation.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.SqueezeExcitation.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.StochasticDepth.html" class="btn btn-neutral float-right" title="StochasticDepth" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.Permute.html" class="btn btn-neutral" title="Permute" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SqueezeExcitation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.StochasticDepth.html b/0.16/generated/torchvision.ops.StochasticDepth.html
new file mode 100644
index 00000000000..368046d3ac7
--- /dev/null
+++ b/0.16/generated/torchvision.ops.StochasticDepth.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>StochasticDepth &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="deform_conv2d" href="torchvision.ops.deform_conv2d.html" />
+    <link rel="prev" title="SqueezeExcitation" href="torchvision.ops.SqueezeExcitation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>StochasticDepth</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.StochasticDepth.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="stochasticdepth">
+<h1>StochasticDepth<a class="headerlink" href="#stochasticdepth" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.ops.StochasticDepth">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">StochasticDepth</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/ops/stochastic_depth.html#StochasticDepth"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.StochasticDepth" title="Permalink to this definition">¶</a></dt>
+<dd><p>See <a class="reference internal" href="torchvision.ops.stochastic_depth.html#torchvision.ops.stochastic_depth" title="torchvision.ops.stochastic_depth"><code class="xref py py-func docutils literal notranslate"><span class="pre">stochastic_depth()</span></code></a>.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.ops.StochasticDepth.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/stochastic_depth.html#StochasticDepth.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.StochasticDepth.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.deform_conv2d.html" class="btn btn-neutral float-right" title="deform_conv2d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.SqueezeExcitation.html" class="btn btn-neutral" title="SqueezeExcitation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">StochasticDepth</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.batched_nms.html b/0.16/generated/torchvision.ops.batched_nms.html
new file mode 100644
index 00000000000..041c14fc230
--- /dev/null
+++ b/0.16/generated/torchvision.ops.batched_nms.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>batched_nms &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="masks_to_boxes" href="torchvision.ops.masks_to_boxes.html" />
+    <link rel="prev" title="Operators" href="../ops.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>batched_nms</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.batched_nms.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="batched-nms">
+<h1>batched_nms<a class="headerlink" href="#batched-nms" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.batched_nms">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">batched_nms</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">scores</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">idxs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">iou_threshold</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#batched_nms"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.batched_nms" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs non-maximum suppression in a batched fashion.</p>
+<p>Each index value correspond to a category, and NMS
+will not be applied between elements of different categories.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – boxes where NMS will be performed. They
+are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p></li>
+<li><p><strong>scores</strong> (<em>Tensor</em><em>[</em><em>N</em><em>]</em>) – scores for each one of the boxes</p></li>
+<li><p><strong>idxs</strong> (<em>Tensor</em><em>[</em><em>N</em><em>]</em>) – indices of the categories for each one of the boxes.</p></li>
+<li><p><strong>iou_threshold</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – discards all overlapping boxes with IoU &gt; iou_threshold</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>int64 tensor with the indices of the elements that have been kept by NMS, sorted
+in decreasing order of scores</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.masks_to_boxes.html" class="btn btn-neutral float-right" title="masks_to_boxes" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../ops.html" class="btn btn-neutral" title="Operators" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">batched_nms</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.box_area.html b/0.16/generated/torchvision.ops.box_area.html
new file mode 100644
index 00000000000..c1ff6c7d1a8
--- /dev/null
+++ b/0.16/generated/torchvision.ops.box_area.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>box_area &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="box_convert" href="torchvision.ops.box_convert.html" />
+    <link rel="prev" title="PSRoIPool" href="torchvision.ops.PSRoIPool.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>box_area</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.box_area.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="box-area">
+<h1>box_area<a class="headerlink" href="#box-area" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.box_area">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">box_area</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#box_area"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.box_area" title="Permalink to this definition">¶</a></dt>
+<dd><p>Computes the area of a set of bounding boxes, which are specified by their
+(x1, y1, x2, y2) coordinates.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – boxes for which the area will be computed. They
+are expected to be in (x1, y1, x2, y2) format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>the area for each box</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.box_convert.html" class="btn btn-neutral float-right" title="box_convert" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.PSRoIPool.html" class="btn btn-neutral" title="PSRoIPool" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">box_area</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.box_convert.html b/0.16/generated/torchvision.ops.box_convert.html
new file mode 100644
index 00000000000..c394b1aba92
--- /dev/null
+++ b/0.16/generated/torchvision.ops.box_convert.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>box_convert &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="box_iou" href="torchvision.ops.box_iou.html" />
+    <link rel="prev" title="box_area" href="torchvision.ops.box_area.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>box_convert</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.box_convert.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="box-convert">
+<h1>box_convert<a class="headerlink" href="#box-convert" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.box_convert">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">box_convert</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">in_fmt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">out_fmt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#box_convert"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.box_convert" title="Permalink to this definition">¶</a></dt>
+<dd><p>Converts boxes from given in_fmt to out_fmt.
+Supported in_fmt and out_fmt are:</p>
+<p>‘xyxy’: boxes are represented via corners, x1, y1 being top left and x2, y2 being bottom right.
+This is the format that torchvision utilities expect.</p>
+<p>‘xywh’ : boxes are represented via corner, width and height, x1, y2 being top left, w, h being width and height.</p>
+<p>‘cxcywh’ : boxes are represented via centre, width and height, cx, cy being center of box, w, h
+being width and height.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – boxes which will be converted.</p></li>
+<li><p><strong>in_fmt</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Input format of given boxes. Supported formats are [‘xyxy’, ‘xywh’, ‘cxcywh’].</p></li>
+<li><p><strong>out_fmt</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Output format of given boxes. Supported formats are [‘xyxy’, ‘xywh’, ‘cxcywh’]</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Boxes into converted format.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, 4]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.box_iou.html" class="btn btn-neutral float-right" title="box_iou" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.box_area.html" class="btn btn-neutral" title="box_area" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">box_convert</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.box_iou.html b/0.16/generated/torchvision.ops.box_iou.html
new file mode 100644
index 00000000000..1fc77fe5be5
--- /dev/null
+++ b/0.16/generated/torchvision.ops.box_iou.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>box_iou &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="clip_boxes_to_image" href="torchvision.ops.clip_boxes_to_image.html" />
+    <link rel="prev" title="box_convert" href="torchvision.ops.box_convert.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>box_iou</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.box_iou.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="box-iou">
+<h1>box_iou<a class="headerlink" href="#box-iou" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.box_iou">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">box_iou</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#box_iou"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.box_iou" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return intersection-over-union (Jaccard index) between two sets of boxes.</p>
+<p>Both sets of boxes are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes1</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – first set of boxes</p></li>
+<li><p><strong>boxes2</strong> (<em>Tensor</em><em>[</em><em>M</em><em>, </em><em>4</em><em>]</em>) – second set of boxes</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>the NxM matrix containing the pairwise IoU values for every element in boxes1 and boxes2</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, M]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.clip_boxes_to_image.html" class="btn btn-neutral float-right" title="clip_boxes_to_image" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.box_convert.html" class="btn btn-neutral" title="box_convert" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">box_iou</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.clip_boxes_to_image.html b/0.16/generated/torchvision.ops.clip_boxes_to_image.html
new file mode 100644
index 00000000000..91b0fac2476
--- /dev/null
+++ b/0.16/generated/torchvision.ops.clip_boxes_to_image.html
@@ -0,0 +1,709 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>clip_boxes_to_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="complete_box_iou" href="torchvision.ops.complete_box_iou.html" />
+    <link rel="prev" title="box_iou" href="torchvision.ops.box_iou.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>clip_boxes_to_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.clip_boxes_to_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="clip-boxes-to-image">
+<h1>clip_boxes_to_image<a class="headerlink" href="#clip-boxes-to-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.clip_boxes_to_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">clip_boxes_to_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#clip_boxes_to_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.clip_boxes_to_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>Clip boxes so that they lie inside an image of size <cite>size</cite>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – boxes in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format
+with <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p></li>
+<li><p><strong>size</strong> (<em>Tuple</em><em>[</em><em>height</em><em>, </em><em>width</em><em>]</em>) – size of the image</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>clipped boxes</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, 4]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.complete_box_iou.html" class="btn btn-neutral float-right" title="complete_box_iou" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.box_iou.html" class="btn btn-neutral" title="box_iou" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">clip_boxes_to_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.complete_box_iou.html b/0.16/generated/torchvision.ops.complete_box_iou.html
new file mode 100644
index 00000000000..d32103c55ad
--- /dev/null
+++ b/0.16/generated/torchvision.ops.complete_box_iou.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>complete_box_iou &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="distance_box_iou" href="torchvision.ops.distance_box_iou.html" />
+    <link rel="prev" title="clip_boxes_to_image" href="torchvision.ops.clip_boxes_to_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>complete_box_iou</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.complete_box_iou.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="complete-box-iou">
+<h1>complete_box_iou<a class="headerlink" href="#complete-box-iou" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.complete_box_iou">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">complete_box_iou</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-07</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#complete_box_iou"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.complete_box_iou" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return complete intersection-over-union (Jaccard index) between two sets of boxes.
+Both sets of boxes are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.
+:param boxes1: first set of boxes
+:type boxes1: Tensor[N, 4]
+:param boxes2: second set of boxes
+:type boxes2: Tensor[M, 4]
+:param eps: small number to prevent division by zero. Default: 1e-7
+:type eps: float, optional</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>the NxM matrix containing the pairwise complete IoU values
+for every element in boxes1 and boxes2</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>Tensor[N, M]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.distance_box_iou.html" class="btn btn-neutral float-right" title="distance_box_iou" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.clip_boxes_to_image.html" class="btn btn-neutral" title="clip_boxes_to_image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">complete_box_iou</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.complete_box_iou_loss.html b/0.16/generated/torchvision.ops.complete_box_iou_loss.html
new file mode 100644
index 00000000000..d1936b44d11
--- /dev/null
+++ b/0.16/generated/torchvision.ops.complete_box_iou_loss.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>complete_box_iou_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="distance_box_iou_loss" href="torchvision.ops.distance_box_iou_loss.html" />
+    <link rel="prev" title="remove_small_boxes" href="torchvision.ops.remove_small_boxes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>complete_box_iou_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.complete_box_iou_loss.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="complete-box-iou-loss">
+<h1>complete_box_iou_loss<a class="headerlink" href="#complete-box-iou-loss" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.complete_box_iou_loss">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">complete_box_iou_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">reduction</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'none'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-07</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/ciou_loss.html#complete_box_iou_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.complete_box_iou_loss" title="Permalink to this definition">¶</a></dt>
+<dd><p>Gradient-friendly IoU loss with an additional penalty that is non-zero when the
+boxes do not overlap. This loss function considers important geometrical
+factors such as overlap area, normalized central point distance and aspect ratio.
+This loss is symmetric, so the boxes1 and boxes2 arguments are interchangeable.</p>
+<p>Both sets of boxes are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>, and The two boxes should have the
+same dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes1</strong> – (Tensor[N, 4] or Tensor[4]) first set of boxes</p></li>
+<li><p><strong>boxes2</strong> – (Tensor[N, 4] or Tensor[4]) second set of boxes</p></li>
+<li><p><strong>reduction</strong> – (string, optional) Specifies the reduction to apply to the output:
+<code class="docutils literal notranslate"><span class="pre">'none'</span></code> | <code class="docutils literal notranslate"><span class="pre">'mean'</span></code> | <code class="docutils literal notranslate"><span class="pre">'sum'</span></code>. <code class="docutils literal notranslate"><span class="pre">'none'</span></code>: No reduction will be
+applied to the output. <code class="docutils literal notranslate"><span class="pre">'mean'</span></code>: The output will be averaged.
+<code class="docutils literal notranslate"><span class="pre">'sum'</span></code>: The output will be summed. Default: <code class="docutils literal notranslate"><span class="pre">'none'</span></code></p></li>
+<li><p><strong>eps</strong> – (float): small number to prevent division by zero. Default: 1e-7</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Loss tensor with the reduction option applied.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+<dl class="simple">
+<dt>Reference:</dt><dd><p>Zhaohui Zheng et al.: Complete Intersection over Union Loss:
+<a class="reference external" href="https://arxiv.org/abs/1911.08287">https://arxiv.org/abs/1911.08287</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.distance_box_iou_loss.html" class="btn btn-neutral float-right" title="distance_box_iou_loss" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.remove_small_boxes.html" class="btn btn-neutral" title="remove_small_boxes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">complete_box_iou_loss</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.deform_conv2d.html b/0.16/generated/torchvision.ops.deform_conv2d.html
new file mode 100644
index 00000000000..48ad785c7e4
--- /dev/null
+++ b/0.16/generated/torchvision.ops.deform_conv2d.html
@@ -0,0 +1,736 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>deform_conv2d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="drop_block2d" href="torchvision.ops.drop_block2d.html" />
+    <link rel="prev" title="StochasticDepth" href="torchvision.ops.StochasticDepth.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>deform_conv2d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.deform_conv2d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="deform-conv2d">
+<h1>deform_conv2d<a class="headerlink" href="#deform-conv2d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.deform_conv2d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">deform_conv2d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">offset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(1,</span> <span class="pre">1)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0,</span> <span class="pre">0)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dilation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(1,</span> <span class="pre">1)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/deform_conv.html#deform_conv2d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.deform_conv2d" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs Deformable Convolution v2, described in
+<a class="reference external" href="https://arxiv.org/abs/1811.11168">Deformable ConvNets v2: More Deformable, Better Results</a> if <code class="xref py py-attr docutils literal notranslate"><span class="pre">mask</span></code> is not <code class="docutils literal notranslate"><span class="pre">None</span></code> and
+Performs Deformable Convolution, described in
+<a class="reference external" href="https://arxiv.org/abs/1703.06211">Deformable Convolutional Networks</a> if <code class="xref py py-attr docutils literal notranslate"><span class="pre">mask</span></code> is <code class="docutils literal notranslate"><span class="pre">None</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>batch_size</em><em>, </em><em>in_channels</em><em>, </em><em>in_height</em><em>, </em><em>in_width</em><em>]</em>) – input tensor</p></li>
+<li><p><strong>offset</strong> (<em>Tensor</em><em>[</em><em>batch_size</em><em>, </em><em>2 * offset_groups * kernel_height * kernel_width</em><em>, </em><em>out_height</em><em>, </em><em>out_width</em><em>]</em>) – offsets to be applied for each position in the convolution kernel.</p></li>
+<li><p><strong>weight</strong> (<em>Tensor</em><em>[</em><em>out_channels</em><em>, </em><em>in_channels // groups</em><em>, </em><em>kernel_height</em><em>, </em><em>kernel_width</em><em>]</em>) – convolution weights,
+split into groups of size (in_channels // groups)</p></li>
+<li><p><strong>bias</strong> (<em>Tensor</em><em>[</em><em>out_channels</em><em>]</em>) – optional bias of shape (out_channels,). Default: None</p></li>
+<li><p><strong>stride</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – distance between convolution centers. Default: 1</p></li>
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – height/width of padding of zeroes around
+each image. Default: 0</p></li>
+<li><p><strong>dilation</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – the spacing between kernel elements. Default: 1</p></li>
+<li><p><strong>mask</strong> (<em>Tensor</em><em>[</em><em>batch_size</em><em>, </em><em>offset_groups * kernel_height * kernel_width</em><em>, </em><em>out_height</em><em>, </em><em>out_width</em><em>]</em>) – masks to be applied for each position in the convolution kernel. Default: None</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>result of convolution</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[batch_sz, out_channels, out_h, out_w]</p>
+</dd>
+</dl>
+<dl>
+<dt>Examples::</dt><dd><div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="nb">input</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">10</span><span class="p">,</span> <span class="mi">10</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">kh</span><span class="p">,</span> <span class="n">kw</span> <span class="o">=</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">3</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">weight</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">5</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="n">kh</span><span class="p">,</span> <span class="n">kw</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># offset and mask should have the same spatial size as the output</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># of the convolution. In this case, for an input of 10, stride of 1</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># and kernel size of 3, without padding, the output size is 8</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">offset</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">kh</span> <span class="o">*</span> <span class="n">kw</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="n">kh</span> <span class="o">*</span> <span class="n">kw</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">out</span> <span class="o">=</span> <span class="n">deform_conv2d</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">offset</span><span class="p">,</span> <span class="n">weight</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">mask</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="nb">print</span><span class="p">(</span><span class="n">out</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># returns</span>
+<span class="gp">&gt;&gt;&gt; </span> <span class="n">torch</span><span class="o">.</span><span class="n">Size</span><span class="p">([</span><span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">8</span><span class="p">])</span>
+</pre></div>
+</div>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.drop_block2d.html" class="btn btn-neutral float-right" title="drop_block2d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.StochasticDepth.html" class="btn btn-neutral" title="StochasticDepth" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">deform_conv2d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.distance_box_iou.html b/0.16/generated/torchvision.ops.distance_box_iou.html
new file mode 100644
index 00000000000..ae30fbcec40
--- /dev/null
+++ b/0.16/generated/torchvision.ops.distance_box_iou.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>distance_box_iou &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="generalized_box_iou" href="torchvision.ops.generalized_box_iou.html" />
+    <link rel="prev" title="complete_box_iou" href="torchvision.ops.complete_box_iou.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>distance_box_iou</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.distance_box_iou.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="distance-box-iou">
+<h1>distance_box_iou<a class="headerlink" href="#distance-box-iou" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.distance_box_iou">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">distance_box_iou</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-07</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#distance_box_iou"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.distance_box_iou" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return distance intersection-over-union (Jaccard index) between two sets of boxes.</p>
+<p>Both sets of boxes are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes1</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – first set of boxes</p></li>
+<li><p><strong>boxes2</strong> (<em>Tensor</em><em>[</em><em>M</em><em>, </em><em>4</em><em>]</em>) – second set of boxes</p></li>
+<li><p><strong>eps</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – small number to prevent division by zero. Default: 1e-7</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>the NxM matrix containing the pairwise distance IoU values
+for every element in boxes1 and boxes2</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, M]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.generalized_box_iou.html" class="btn btn-neutral float-right" title="generalized_box_iou" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.complete_box_iou.html" class="btn btn-neutral" title="complete_box_iou" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">distance_box_iou</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.distance_box_iou_loss.html b/0.16/generated/torchvision.ops.distance_box_iou_loss.html
new file mode 100644
index 00000000000..7e826e79d77
--- /dev/null
+++ b/0.16/generated/torchvision.ops.distance_box_iou_loss.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>distance_box_iou_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="generalized_box_iou_loss" href="torchvision.ops.generalized_box_iou_loss.html" />
+    <link rel="prev" title="complete_box_iou_loss" href="torchvision.ops.complete_box_iou_loss.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>distance_box_iou_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.distance_box_iou_loss.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="distance-box-iou-loss">
+<h1>distance_box_iou_loss<a class="headerlink" href="#distance-box-iou-loss" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.distance_box_iou_loss">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">distance_box_iou_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">reduction</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'none'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-07</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/diou_loss.html#distance_box_iou_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.distance_box_iou_loss" title="Permalink to this definition">¶</a></dt>
+<dd><p>Gradient-friendly IoU loss with an additional penalty that is non-zero when the
+distance between boxes’ centers isn’t zero. Indeed, for two exactly overlapping
+boxes, the distance IoU is the same as the IoU loss.
+This loss is symmetric, so the boxes1 and boxes2 arguments are interchangeable.</p>
+<p>Both sets of boxes are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>, and The two boxes should have the
+same dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes1</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – first set of boxes</p></li>
+<li><p><strong>boxes2</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – second set of boxes</p></li>
+<li><p><strong>reduction</strong> (<em>string</em><em>, </em><em>optional</em>) – Specifies the reduction to apply to the output:
+<code class="docutils literal notranslate"><span class="pre">'none'</span></code> | <code class="docutils literal notranslate"><span class="pre">'mean'</span></code> | <code class="docutils literal notranslate"><span class="pre">'sum'</span></code>. <code class="docutils literal notranslate"><span class="pre">'none'</span></code>: No reduction will be
+applied to the output. <code class="docutils literal notranslate"><span class="pre">'mean'</span></code>: The output will be averaged.
+<code class="docutils literal notranslate"><span class="pre">'sum'</span></code>: The output will be summed. Default: <code class="docutils literal notranslate"><span class="pre">'none'</span></code></p></li>
+<li><p><strong>eps</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – small number to prevent division by zero. Default: 1e-7</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Loss tensor with the reduction option applied.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+<dl class="simple">
+<dt>Reference:</dt><dd><p>Zhaohui Zheng et al.: Distance Intersection over Union Loss:
+<a class="reference external" href="https://arxiv.org/abs/1911.08287">https://arxiv.org/abs/1911.08287</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.generalized_box_iou_loss.html" class="btn btn-neutral float-right" title="generalized_box_iou_loss" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.complete_box_iou_loss.html" class="btn btn-neutral" title="complete_box_iou_loss" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">distance_box_iou_loss</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.drop_block2d.html b/0.16/generated/torchvision.ops.drop_block2d.html
new file mode 100644
index 00000000000..b486fb5e5f2
--- /dev/null
+++ b/0.16/generated/torchvision.ops.drop_block2d.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>drop_block2d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="drop_block3d" href="torchvision.ops.drop_block3d.html" />
+    <link rel="prev" title="deform_conv2d" href="torchvision.ops.deform_conv2d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>drop_block2d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.drop_block2d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="drop-block2d">
+<h1>drop_block2d<a class="headerlink" href="#drop-block2d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.drop_block2d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">drop_block2d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-06</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/drop_block.html#drop_block2d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.drop_block2d" title="Permalink to this definition">¶</a></dt>
+<dd><p>Implements DropBlock2d from <cite>“DropBlock: A regularization method for convolutional networks”
+&lt;https://arxiv.org/abs/1810.12890&gt;</cite>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>C</em><em>, </em><em>H</em><em>, </em><em>W</em><em>]</em>) – The input tensor or 4-dimensions with the first one
+being its batch i.e. a batch with <code class="docutils literal notranslate"><span class="pre">N</span></code> rows.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Probability of an element to be dropped.</p></li>
+<li><p><strong>block_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Size of the block to drop.</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If set to <code class="docutils literal notranslate"><span class="pre">True</span></code>, will do this operation in-place. Default: <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+<li><p><strong>eps</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – A value added to the denominator for numerical stability. Default: 1e-6.</p></li>
+<li><p><strong>training</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – apply dropblock if is <code class="docutils literal notranslate"><span class="pre">True</span></code>. Default: <code class="docutils literal notranslate"><span class="pre">True</span></code>.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The randomly zeroed tensor after dropblock.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, C, H, W]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.drop_block3d.html" class="btn btn-neutral float-right" title="drop_block3d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.deform_conv2d.html" class="btn btn-neutral" title="deform_conv2d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">drop_block2d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.drop_block3d.html b/0.16/generated/torchvision.ops.drop_block3d.html
new file mode 100644
index 00000000000..d27de613a1d
--- /dev/null
+++ b/0.16/generated/torchvision.ops.drop_block3d.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>drop_block3d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="stochastic_depth" href="torchvision.ops.stochastic_depth.html" />
+    <link rel="prev" title="drop_block2d" href="torchvision.ops.drop_block2d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>drop_block3d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.drop_block3d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="drop-block3d">
+<h1>drop_block3d<a class="headerlink" href="#drop-block3d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.drop_block3d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">drop_block3d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-06</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/drop_block.html#drop_block3d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.drop_block3d" title="Permalink to this definition">¶</a></dt>
+<dd><p>Implements DropBlock3d from <cite>“DropBlock: A regularization method for convolutional networks”
+&lt;https://arxiv.org/abs/1810.12890&gt;</cite>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>C</em><em>, </em><em>D</em><em>, </em><em>H</em><em>, </em><em>W</em><em>]</em>) – The input tensor or 5-dimensions with the first one
+being its batch i.e. a batch with <code class="docutils literal notranslate"><span class="pre">N</span></code> rows.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Probability of an element to be dropped.</p></li>
+<li><p><strong>block_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Size of the block to drop.</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If set to <code class="docutils literal notranslate"><span class="pre">True</span></code>, will do this operation in-place. Default: <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+<li><p><strong>eps</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – A value added to the denominator for numerical stability. Default: 1e-6.</p></li>
+<li><p><strong>training</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – apply dropblock if is <code class="docutils literal notranslate"><span class="pre">True</span></code>. Default: <code class="docutils literal notranslate"><span class="pre">True</span></code>.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The randomly zeroed tensor after dropblock.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, C, D, H, W]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.stochastic_depth.html" class="btn btn-neutral float-right" title="stochastic_depth" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.drop_block2d.html" class="btn btn-neutral" title="drop_block2d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">drop_block3d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.generalized_box_iou.html b/0.16/generated/torchvision.ops.generalized_box_iou.html
new file mode 100644
index 00000000000..c95c66f15cd
--- /dev/null
+++ b/0.16/generated/torchvision.ops.generalized_box_iou.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>generalized_box_iou &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="remove_small_boxes" href="torchvision.ops.remove_small_boxes.html" />
+    <link rel="prev" title="distance_box_iou" href="torchvision.ops.distance_box_iou.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>generalized_box_iou</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.generalized_box_iou.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="generalized-box-iou">
+<h1>generalized_box_iou<a class="headerlink" href="#generalized-box-iou" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.generalized_box_iou">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">generalized_box_iou</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#generalized_box_iou"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.generalized_box_iou" title="Permalink to this definition">¶</a></dt>
+<dd><p>Return generalized intersection-over-union (Jaccard index) between two sets of boxes.</p>
+<p>Both sets of boxes are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes1</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – first set of boxes</p></li>
+<li><p><strong>boxes2</strong> (<em>Tensor</em><em>[</em><em>M</em><em>, </em><em>4</em><em>]</em>) – second set of boxes</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>the NxM matrix containing the pairwise generalized IoU values
+for every element in boxes1 and boxes2</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, M]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.remove_small_boxes.html" class="btn btn-neutral float-right" title="remove_small_boxes" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.distance_box_iou.html" class="btn btn-neutral" title="distance_box_iou" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">generalized_box_iou</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.generalized_box_iou_loss.html b/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
new file mode 100644
index 00000000000..06e90c953b4
--- /dev/null
+++ b/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>generalized_box_iou_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="sigmoid_focal_loss" href="torchvision.ops.sigmoid_focal_loss.html" />
+    <link rel="prev" title="distance_box_iou_loss" href="torchvision.ops.distance_box_iou_loss.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>generalized_box_iou_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.generalized_box_iou_loss.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="generalized-box-iou-loss">
+<h1>generalized_box_iou_loss<a class="headerlink" href="#generalized-box-iou-loss" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.generalized_box_iou_loss">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">generalized_box_iou_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">reduction</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'none'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-07</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/giou_loss.html#generalized_box_iou_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.generalized_box_iou_loss" title="Permalink to this definition">¶</a></dt>
+<dd><p>Gradient-friendly IoU loss with an additional penalty that is non-zero when the
+boxes do not overlap and scales with the size of their smallest enclosing box.
+This loss is symmetric, so the boxes1 and boxes2 arguments are interchangeable.</p>
+<p>Both sets of boxes are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>, and The two boxes should have the
+same dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes1</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>] or </em><em>Tensor</em><em>[</em><em>4</em><em>]</em>) – first set of boxes</p></li>
+<li><p><strong>boxes2</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>] or </em><em>Tensor</em><em>[</em><em>4</em><em>]</em>) – second set of boxes</p></li>
+<li><p><strong>reduction</strong> (<em>string</em><em>, </em><em>optional</em>) – Specifies the reduction to apply to the output:
+<code class="docutils literal notranslate"><span class="pre">'none'</span></code> | <code class="docutils literal notranslate"><span class="pre">'mean'</span></code> | <code class="docutils literal notranslate"><span class="pre">'sum'</span></code>. <code class="docutils literal notranslate"><span class="pre">'none'</span></code>: No reduction will be
+applied to the output. <code class="docutils literal notranslate"><span class="pre">'mean'</span></code>: The output will be averaged.
+<code class="docutils literal notranslate"><span class="pre">'sum'</span></code>: The output will be summed. Default: <code class="docutils literal notranslate"><span class="pre">'none'</span></code></p></li>
+<li><p><strong>eps</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – small number to prevent division by zero. Default: 1e-7</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Loss tensor with the reduction option applied.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+<dl class="simple">
+<dt>Reference:</dt><dd><p>Hamid Rezatofighi et al.: Generalized Intersection over Union:
+A Metric and A Loss for Bounding Box Regression:
+<a class="reference external" href="https://arxiv.org/abs/1902.09630">https://arxiv.org/abs/1902.09630</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.sigmoid_focal_loss.html" class="btn btn-neutral float-right" title="sigmoid_focal_loss" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.distance_box_iou_loss.html" class="btn btn-neutral" title="distance_box_iou_loss" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">generalized_box_iou_loss</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.masks_to_boxes.html b/0.16/generated/torchvision.ops.masks_to_boxes.html
new file mode 100644
index 00000000000..75f8c3b8fec
--- /dev/null
+++ b/0.16/generated/torchvision.ops.masks_to_boxes.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>masks_to_boxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="nms" href="torchvision.ops.nms.html" />
+    <link rel="prev" title="batched_nms" href="torchvision.ops.batched_nms.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>masks_to_boxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.masks_to_boxes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="masks-to-boxes">
+<h1>masks_to_boxes<a class="headerlink" href="#masks-to-boxes" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.masks_to_boxes">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">masks_to_boxes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">masks</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#masks_to_boxes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.masks_to_boxes" title="Permalink to this definition">¶</a></dt>
+<dd><p>Compute the bounding boxes around the provided masks.</p>
+<p>Returns a [N, 4] tensor containing bounding boxes. The boxes are in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>masks</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>H</em><em>, </em><em>W</em><em>]</em>) – masks to transform where N is the number of masks
+and (H, W) are the spatial dimensions.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>bounding boxes</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, 4]</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">masks_to_boxes</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.nms.html" class="btn btn-neutral float-right" title="nms" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.batched_nms.html" class="btn btn-neutral" title="batched_nms" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">masks_to_boxes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.nms.html b/0.16/generated/torchvision.ops.nms.html
new file mode 100644
index 00000000000..2c7e30ce6d2
--- /dev/null
+++ b/0.16/generated/torchvision.ops.nms.html
@@ -0,0 +1,720 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>nms &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="roi_align" href="torchvision.ops.roi_align.html" />
+    <link rel="prev" title="masks_to_boxes" href="torchvision.ops.masks_to_boxes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>nms</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.nms.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="nms">
+<h1>nms<a class="headerlink" href="#nms" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.nms">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">nms</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">scores</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">iou_threshold</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#nms"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.nms" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs non-maximum suppression (NMS) on the boxes according
+to their intersection-over-union (IoU).</p>
+<p>NMS iteratively removes lower scoring boxes which have an
+IoU greater than iou_threshold with another (higher scoring)
+box.</p>
+<p>If multiple boxes have the exact same score and satisfy the IoU
+criterion with respect to a reference box, the selected box is
+not guaranteed to be the same between CPU and GPU. This is similar
+to the behavior of argsort in PyTorch when repeated values are present.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em><em>)</em>) – boxes to perform NMS on. They
+are expected to be in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format with <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p></li>
+<li><p><strong>scores</strong> (<em>Tensor</em><em>[</em><em>N</em><em>]</em>) – scores for each one of the boxes</p></li>
+<li><p><strong>iou_threshold</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – discards all overlapping boxes with IoU &gt; iou_threshold</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>int64 tensor with the indices of the elements that have been kept
+by NMS, sorted in decreasing order of scores</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.roi_align.html" class="btn btn-neutral float-right" title="roi_align" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.masks_to_boxes.html" class="btn btn-neutral" title="masks_to_boxes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">nms</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.ps_roi_align.html b/0.16/generated/torchvision.ops.ps_roi_align.html
new file mode 100644
index 00000000000..17d7f5a049b
--- /dev/null
+++ b/0.16/generated/torchvision.ops.ps_roi_align.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ps_roi_align &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ps_roi_pool" href="torchvision.ops.ps_roi_pool.html" />
+    <link rel="prev" title="roi_pool" href="torchvision.ops.roi_pool.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>ps_roi_align</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.ps_roi_align.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ps-roi-align">
+<h1>ps_roi_align<a class="headerlink" href="#ps-roi-align" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.ps_roi_align">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">ps_roi_align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">-</span> <span class="pre">1</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/ps_roi_align.html#ps_roi_align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.ps_roi_align" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs Position-Sensitive Region of Interest (RoI) Align operator
+mentioned in Light-Head R-CNN.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>C</em><em>, </em><em>H</em><em>, </em><em>W</em><em>]</em>) – The input tensor, i.e. a batch with <code class="docutils literal notranslate"><span class="pre">N</span></code> elements. Each element
+contains <code class="docutils literal notranslate"><span class="pre">C</span></code> feature maps of dimensions <code class="docutils literal notranslate"><span class="pre">H</span> <span class="pre">x</span> <span class="pre">W</span></code>.</p></li>
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>K</em><em>, </em><em>5</em><em>] or </em><em>List</em><em>[</em><em>Tensor</em><em>[</em><em>L</em><em>, </em><em>4</em><em>]</em><em>]</em>) – the box coordinates in (x1, y1, x2, y2)
+format where the regions will be taken from.
+The coordinate must satisfy <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.
+If a single Tensor is passed, then the first column should
+contain the index of the corresponding element in the batch, i.e. a number in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">N</span> <span class="pre">-</span> <span class="pre">1]</span></code>.
+If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i
+in the batch.</p></li>
+<li><p><strong>output_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – the size of the output (in bins or pixels) after the pooling
+is performed, as (height, width).</p></li>
+<li><p><strong>spatial_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – a scaling factor that maps the box coordinates to
+the input coordinates. For example, if your boxes are defined on the scale
+of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+the original image), you’ll want to set this to 0.5. Default: 1.0</p></li>
+<li><p><strong>sampling_ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of sampling points in the interpolation grid
+used to compute the output value of each pooled output bin. If &gt; 0,
+then exactly <code class="docutils literal notranslate"><span class="pre">sampling_ratio</span> <span class="pre">x</span> <span class="pre">sampling_ratio</span></code> sampling points per bin are used. If
+&lt;= 0, then an adaptive number of grid points are used (computed as
+<code class="docutils literal notranslate"><span class="pre">ceil(roi_width</span> <span class="pre">/</span> <span class="pre">output_width)</span></code>, and likewise for height). Default: -1</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The pooled RoIs</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[K, C / (output_size[0] * output_size[1]), output_size[0], output_size[1]]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.ps_roi_pool.html" class="btn btn-neutral float-right" title="ps_roi_pool" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.roi_pool.html" class="btn btn-neutral" title="roi_pool" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ps_roi_align</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.ps_roi_pool.html b/0.16/generated/torchvision.ops.ps_roi_pool.html
new file mode 100644
index 00000000000..ad06dc9ea4a
--- /dev/null
+++ b/0.16/generated/torchvision.ops.ps_roi_pool.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ps_roi_pool &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FeaturePyramidNetwork" href="torchvision.ops.FeaturePyramidNetwork.html" />
+    <link rel="prev" title="ps_roi_align" href="torchvision.ops.ps_roi_align.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>ps_roi_pool</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.ps_roi_pool.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ps-roi-pool">
+<h1>ps_roi_pool<a class="headerlink" href="#ps-roi-pool" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.ps_roi_pool">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">ps_roi_pool</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/ps_roi_pool.html#ps_roi_pool"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.ps_roi_pool" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs Position-Sensitive Region of Interest (RoI) Pool operator
+described in R-FCN</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>C</em><em>, </em><em>H</em><em>, </em><em>W</em><em>]</em>) – The input tensor, i.e. a batch with <code class="docutils literal notranslate"><span class="pre">N</span></code> elements. Each element
+contains <code class="docutils literal notranslate"><span class="pre">C</span></code> feature maps of dimensions <code class="docutils literal notranslate"><span class="pre">H</span> <span class="pre">x</span> <span class="pre">W</span></code>.</p></li>
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>K</em><em>, </em><em>5</em><em>] or </em><em>List</em><em>[</em><em>Tensor</em><em>[</em><em>L</em><em>, </em><em>4</em><em>]</em><em>]</em>) – the box coordinates in (x1, y1, x2, y2)
+format where the regions will be taken from.
+The coordinate must satisfy <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.
+If a single Tensor is passed, then the first column should
+contain the index of the corresponding element in the batch, i.e. a number in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">N</span> <span class="pre">-</span> <span class="pre">1]</span></code>.
+If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i
+in the batch.</p></li>
+<li><p><strong>output_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – the size of the output (in bins or pixels) after the pooling
+is performed, as (height, width).</p></li>
+<li><p><strong>spatial_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – a scaling factor that maps the box coordinates to
+the input coordinates. For example, if your boxes are defined on the scale
+of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+the original image), you’ll want to set this to 0.5. Default: 1.0</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The pooled RoIs.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[K, C / (output_size[0] * output_size[1]), output_size[0], output_size[1]]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.FeaturePyramidNetwork.html" class="btn btn-neutral float-right" title="FeaturePyramidNetwork" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.ps_roi_align.html" class="btn btn-neutral" title="ps_roi_align" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ps_roi_pool</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.remove_small_boxes.html b/0.16/generated/torchvision.ops.remove_small_boxes.html
new file mode 100644
index 00000000000..10993509bd6
--- /dev/null
+++ b/0.16/generated/torchvision.ops.remove_small_boxes.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>remove_small_boxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="complete_box_iou_loss" href="torchvision.ops.complete_box_iou_loss.html" />
+    <link rel="prev" title="generalized_box_iou" href="torchvision.ops.generalized_box_iou.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>remove_small_boxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.remove_small_boxes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="remove-small-boxes">
+<h1>remove_small_boxes<a class="headerlink" href="#remove-small-boxes" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.remove_small_boxes">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">remove_small_boxes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">min_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/boxes.html#remove_small_boxes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.remove_small_boxes" title="Permalink to this definition">¶</a></dt>
+<dd><p>Remove boxes which contains at least one side smaller than min_size.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>4</em><em>]</em>) – boxes in <code class="docutils literal notranslate"><span class="pre">(x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2)</span></code> format
+with <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.</p></li>
+<li><p><strong>min_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – minimum size</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>indices of the boxes that have both sides
+larger than min_size</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[K]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.complete_box_iou_loss.html" class="btn btn-neutral float-right" title="complete_box_iou_loss" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.generalized_box_iou.html" class="btn btn-neutral" title="generalized_box_iou" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">remove_small_boxes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.roi_align.html b/0.16/generated/torchvision.ops.roi_align.html
new file mode 100644
index 00000000000..ad60cd32e40
--- /dev/null
+++ b/0.16/generated/torchvision.ops.roi_align.html
@@ -0,0 +1,730 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>roi_align &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="roi_pool" href="torchvision.ops.roi_pool.html" />
+    <link rel="prev" title="nms" href="torchvision.ops.nms.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>roi_align</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.roi_align.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="roi-align">
+<h1>roi_align<a class="headerlink" href="#roi-align" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.roi_align">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">roi_align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">-</span> <span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aligned</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/roi_align.html#roi_align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.roi_align" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs Region of Interest (RoI) Align operator with average pooling, as described in Mask R-CNN.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>C</em><em>, </em><em>H</em><em>, </em><em>W</em><em>]</em>) – The input tensor, i.e. a batch with <code class="docutils literal notranslate"><span class="pre">N</span></code> elements. Each element
+contains <code class="docutils literal notranslate"><span class="pre">C</span></code> feature maps of dimensions <code class="docutils literal notranslate"><span class="pre">H</span> <span class="pre">x</span> <span class="pre">W</span></code>.
+If the tensor is quantized, we expect a batch size of <code class="docutils literal notranslate"><span class="pre">N</span> <span class="pre">==</span> <span class="pre">1</span></code>.</p></li>
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>K</em><em>, </em><em>5</em><em>] or </em><em>List</em><em>[</em><em>Tensor</em><em>[</em><em>L</em><em>, </em><em>4</em><em>]</em><em>]</em>) – the box coordinates in (x1, y1, x2, y2)
+format where the regions will be taken from.
+The coordinate must satisfy <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.
+If a single Tensor is passed, then the first column should
+contain the index of the corresponding element in the batch, i.e. a number in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">N</span> <span class="pre">-</span> <span class="pre">1]</span></code>.
+If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i
+in the batch.</p></li>
+<li><p><strong>output_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – the size of the output (in bins or pixels) after the pooling
+is performed, as (height, width).</p></li>
+<li><p><strong>spatial_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – a scaling factor that maps the box coordinates to
+the input coordinates. For example, if your boxes are defined on the scale
+of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+the original image), you’ll want to set this to 0.5. Default: 1.0</p></li>
+<li><p><strong>sampling_ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of sampling points in the interpolation grid
+used to compute the output value of each pooled output bin. If &gt; 0,
+then exactly <code class="docutils literal notranslate"><span class="pre">sampling_ratio</span> <span class="pre">x</span> <span class="pre">sampling_ratio</span></code> sampling points per bin are used. If
+&lt;= 0, then an adaptive number of grid points are used (computed as
+<code class="docutils literal notranslate"><span class="pre">ceil(roi_width</span> <span class="pre">/</span> <span class="pre">output_width)</span></code>, and likewise for height). Default: -1</p></li>
+<li><p><strong>aligned</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If False, use the legacy implementation.
+If True, pixel shift the box coordinates it by -0.5 for a better alignment with the two
+neighboring pixel indices. This version is used in Detectron2</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The pooled RoIs.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[K, C, output_size[0], output_size[1]]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.roi_pool.html" class="btn btn-neutral float-right" title="roi_pool" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.nms.html" class="btn btn-neutral" title="nms" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">roi_align</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.roi_pool.html b/0.16/generated/torchvision.ops.roi_pool.html
new file mode 100644
index 00000000000..5c6a79656f2
--- /dev/null
+++ b/0.16/generated/torchvision.ops.roi_pool.html
@@ -0,0 +1,721 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>roi_pool &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ps_roi_align" href="torchvision.ops.ps_roi_align.html" />
+    <link rel="prev" title="roi_align" href="torchvision.ops.roi_align.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>roi_pool</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.roi_pool.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="roi-pool">
+<h1>roi_pool<a class="headerlink" href="#roi-pool" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.roi_pool">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">roi_pool</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">spatial_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/roi_pool.html#roi_pool"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.roi_pool" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs Region of Interest (RoI) Pool operator described in Fast R-CNN</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>C</em><em>, </em><em>H</em><em>, </em><em>W</em><em>]</em>) – The input tensor, i.e. a batch with <code class="docutils literal notranslate"><span class="pre">N</span></code> elements. Each element
+contains <code class="docutils literal notranslate"><span class="pre">C</span></code> feature maps of dimensions <code class="docutils literal notranslate"><span class="pre">H</span> <span class="pre">x</span> <span class="pre">W</span></code>.</p></li>
+<li><p><strong>boxes</strong> (<em>Tensor</em><em>[</em><em>K</em><em>, </em><em>5</em><em>] or </em><em>List</em><em>[</em><em>Tensor</em><em>[</em><em>L</em><em>, </em><em>4</em><em>]</em><em>]</em>) – the box coordinates in (x1, y1, x2, y2)
+format where the regions will be taken from.
+The coordinate must satisfy <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span></code>.
+If a single Tensor is passed, then the first column should
+contain the index of the corresponding element in the batch, i.e. a number in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">N</span> <span class="pre">-</span> <span class="pre">1]</span></code>.
+If a list of Tensors is passed, then each Tensor will correspond to the boxes for an element i
+in the batch.</p></li>
+<li><p><strong>output_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – the size of the output after the cropping
+is performed, as (height, width)</p></li>
+<li><p><strong>spatial_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – a scaling factor that maps the box coordinates to
+the input coordinates. For example, if your boxes are defined on the scale
+of a 224x224 image and your input is a 112x112 feature map (resulting from a 0.5x scaling of
+the original image), you’ll want to set this to 0.5. Default: 1.0</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The pooled RoIs.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[K, C, output_size[0], output_size[1]]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.ps_roi_align.html" class="btn btn-neutral float-right" title="ps_roi_align" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.roi_align.html" class="btn btn-neutral" title="roi_align" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">roi_pool</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.sigmoid_focal_loss.html b/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
new file mode 100644
index 00000000000..1e14abf0771
--- /dev/null
+++ b/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>sigmoid_focal_loss &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Conv2dNormActivation" href="torchvision.ops.Conv2dNormActivation.html" />
+    <link rel="prev" title="generalized_box_iou_loss" href="torchvision.ops.generalized_box_iou_loss.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>sigmoid_focal_loss</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.sigmoid_focal_loss.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sigmoid-focal-loss">
+<h1>sigmoid_focal_loss<a class="headerlink" href="#sigmoid-focal-loss" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.sigmoid_focal_loss">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">sigmoid_focal_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">targets</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.25</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">gamma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reduction</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'none'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/focal_loss.html#sigmoid_focal_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.sigmoid_focal_loss" title="Permalink to this definition">¶</a></dt>
+<dd><p>Loss used in RetinaNet for dense detection: <a class="reference external" href="https://arxiv.org/abs/1708.02002">https://arxiv.org/abs/1708.02002</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>inputs</strong> (<em>Tensor</em>) – A float tensor of arbitrary shape.
+The predictions for each example.</p></li>
+<li><p><strong>targets</strong> (<em>Tensor</em>) – A float tensor with the same shape as inputs. Stores the binary
+classification label for each element in inputs
+(0 for the negative class and 1 for the positive class).</p></li>
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Weighting factor in range (0,1) to balance
+positive vs negative examples or -1 for ignore. Default: <code class="docutils literal notranslate"><span class="pre">0.25</span></code>.</p></li>
+<li><p><strong>gamma</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Exponent of the modulating factor (1 - p_t) to
+balance easy vs hard examples. Default: <code class="docutils literal notranslate"><span class="pre">2</span></code>.</p></li>
+<li><p><strong>reduction</strong> (<em>string</em>) – <code class="docutils literal notranslate"><span class="pre">'none'</span></code> | <code class="docutils literal notranslate"><span class="pre">'mean'</span></code> | <code class="docutils literal notranslate"><span class="pre">'sum'</span></code>
+<code class="docutils literal notranslate"><span class="pre">'none'</span></code>: No reduction will be applied to the output.
+<code class="docutils literal notranslate"><span class="pre">'mean'</span></code>: The output will be averaged.
+<code class="docutils literal notranslate"><span class="pre">'sum'</span></code>: The output will be summed. Default: <code class="docutils literal notranslate"><span class="pre">'none'</span></code>.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Loss tensor with the reduction option applied.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.ops.Conv2dNormActivation.html" class="btn btn-neutral float-right" title="Conv2dNormActivation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.generalized_box_iou_loss.html" class="btn btn-neutral" title="generalized_box_iou_loss" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">sigmoid_focal_loss</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.ops.stochastic_depth.html b/0.16/generated/torchvision.ops.stochastic_depth.html
new file mode 100644
index 00000000000..957599ccf9c
--- /dev/null
+++ b/0.16/generated/torchvision.ops.stochastic_depth.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>stochastic_depth &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Decoding / Encoding images and videos" href="../io.html" />
+    <link rel="prev" title="drop_block3d" href="torchvision.ops.drop_block3d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../ops.html">Operators</a> &gt;</li>
+        
+      <li>stochastic_depth</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.ops.stochastic_depth.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="stochastic-depth">
+<h1>stochastic_depth<a class="headerlink" href="#stochastic-depth" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.ops.stochastic_depth">
+<span class="sig-prename descclassname"><span class="pre">torchvision.ops.</span></span><span class="sig-name descname"><span class="pre">stochastic_depth</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">training</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/ops/stochastic_depth.html#stochastic_depth"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.ops.stochastic_depth" title="Permalink to this definition">¶</a></dt>
+<dd><p>Implements the Stochastic Depth from <a class="reference external" href="https://arxiv.org/abs/1603.09382">“Deep Networks with Stochastic Depth”</a> used for randomly dropping residual
+branches of residual architectures.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>input</strong> (<em>Tensor</em><em>[</em><em>N</em><em>, </em><em>...</em><em>]</em>) – The input tensor or arbitrary dimensions with the first one
+being its batch i.e. a batch with <code class="docutils literal notranslate"><span class="pre">N</span></code> rows.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the input to be zeroed.</p></li>
+<li><p><strong>mode</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – <code class="docutils literal notranslate"><span class="pre">&quot;batch&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;row&quot;</span></code>.
+<code class="docutils literal notranslate"><span class="pre">&quot;batch&quot;</span></code> randomly zeroes the entire input, <code class="docutils literal notranslate"><span class="pre">&quot;row&quot;</span></code> zeroes
+randomly selected rows from the batch.</p></li>
+<li><p><strong>training</strong> – apply stochastic depth if is <code class="docutils literal notranslate"><span class="pre">True</span></code>. Default: <code class="docutils literal notranslate"><span class="pre">True</span></code></p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The randomly zeroed tensor.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor[N, …]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../io.html" class="btn btn-neutral float-right" title="Decoding / Encoding images and videos" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.ops.drop_block3d.html" class="btn btn-neutral" title="drop_block3d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">stochastic_depth</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.AugMix.html b/0.16/generated/torchvision.transforms.AugMix.html
new file mode 100644
index 00000000000..b6954e152b6
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.AugMix.html
@@ -0,0 +1,735 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>AugMix &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_brightness" href="torchvision.transforms.functional.adjust_brightness.html" />
+    <link rel="prev" title="TrivialAugmentWide" href="torchvision.transforms.TrivialAugmentWide.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>AugMix</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.AugMix.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="augmix">
+<h1>AugMix<a class="headerlink" href="#augmix" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.AugMix">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">AugMix</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">severity</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">3</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mixture_width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">3</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">chain_depth</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">-</span> <span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">all_ops</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#AugMix"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.AugMix" title="Permalink to this definition">¶</a></dt>
+<dd><p>AugMix data augmentation method based on
+<a class="reference external" href="https://arxiv.org/abs/1912.02781">“AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty”</a>.
+If the image is torch Tensor, it should be of type torch.uint8, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>severity</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The severity of base augmentation operators. Default is <code class="docutils literal notranslate"><span class="pre">3</span></code>.</p></li>
+<li><p><strong>mixture_width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The number of augmentation chains. Default is <code class="docutils literal notranslate"><span class="pre">3</span></code>.</p></li>
+<li><p><strong>chain_depth</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The depth of augmentation chains. A negative value denotes stochastic depth sampled from the interval [1, 3].
+Default is <code class="docutils literal notranslate"><span class="pre">-1</span></code>.</p></li>
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – The hyperparameter for the probability distributions. Default is <code class="docutils literal notranslate"><span class="pre">1.0</span></code>.</p></li>
+<li><p><strong>all_ops</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – Use all operations (including brightness, contrast, color and sharpness). Default is <code class="docutils literal notranslate"><span class="pre">True</span></code>.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">AugMix</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.AugMix.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">orig_img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#AugMix.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.AugMix.forward" title="Permalink to this definition">¶</a></dt>
+<dd><blockquote>
+<div><p>img (PIL Image or Tensor): Image to be transformed.</p>
+</div></blockquote>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Transformed image.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.adjust_brightness.html" class="btn btn-neutral float-right" title="adjust_brightness" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.TrivialAugmentWide.html" class="btn btn-neutral" title="TrivialAugmentWide" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">AugMix</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.AutoAugment.html b/0.16/generated/torchvision.transforms.AutoAugment.html
new file mode 100644
index 00000000000..e10d5f37cc8
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.AutoAugment.html
@@ -0,0 +1,742 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>AutoAugment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandAugment" href="torchvision.transforms.RandAugment.html" />
+    <link rel="prev" title="AutoAugmentPolicy" href="torchvision.transforms.AutoAugmentPolicy.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>AutoAugment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.AutoAugment.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="autoaugment">
+<h1>AutoAugment<a class="headerlink" href="#autoaugment" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.AutoAugment">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">AutoAugment</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">policy</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.autoaugment.AutoAugmentPolicy"><span class="pre">AutoAugmentPolicy</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">AutoAugmentPolicy.IMAGENET</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#AutoAugment"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.AutoAugment" title="Permalink to this definition">¶</a></dt>
+<dd><p>AutoAugment data augmentation method based on
+<a class="reference external" href="https://arxiv.org/pdf/1805.09501.pdf">“AutoAugment: Learning Augmentation Strategies from Data”</a>.
+If the image is torch Tensor, it should be of type torch.uint8, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>policy</strong> (<a class="reference internal" href="torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.AutoAugmentPolicy"><em>AutoAugmentPolicy</em></a>) – Desired policy enum defined by
+<a class="reference internal" href="torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.autoaugment.AutoAugmentPolicy"><code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.autoaugment.AutoAugmentPolicy</span></code></a>. Default is <code class="docutils literal notranslate"><span class="pre">AutoAugmentPolicy.IMAGENET</span></code>.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">AutoAugment</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.AutoAugment.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#AutoAugment.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.AutoAugment.forward" title="Permalink to this definition">¶</a></dt>
+<dd><blockquote>
+<div><p>img (PIL Image or Tensor): Image to be transformed.</p>
+</div></blockquote>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>AutoAugmented image.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.AutoAugment.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transform_num</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#AutoAugment.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.AutoAugment.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for autoaugment transformation</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>params required by the autoaugment transformation</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandAugment.html" class="btn btn-neutral float-right" title="RandAugment" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.AutoAugmentPolicy.html" class="btn btn-neutral" title="AutoAugmentPolicy" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">AutoAugment</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html b/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
new file mode 100644
index 00000000000..6bf3d128367
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
@@ -0,0 +1,699 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>AutoAugmentPolicy &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="AutoAugment" href="torchvision.transforms.AutoAugment.html" />
+    <link rel="prev" title="ConvertImageDtype" href="torchvision.transforms.ConvertImageDtype.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>AutoAugmentPolicy</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.AutoAugmentPolicy.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="autoaugmentpolicy">
+<h1>AutoAugmentPolicy<a class="headerlink" href="#autoaugmentpolicy" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.AutoAugmentPolicy">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">AutoAugmentPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#AutoAugmentPolicy"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.AutoAugmentPolicy" title="Permalink to this definition">¶</a></dt>
+<dd><p>AutoAugment policies learned on different datasets.
+Available policies are IMAGENET, CIFAR10 and SVHN.</p>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">AutoAugmentPolicy</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.AutoAugment.html" class="btn btn-neutral float-right" title="AutoAugment" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.ConvertImageDtype.html" class="btn btn-neutral" title="ConvertImageDtype" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">AutoAugmentPolicy</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.CenterCrop.html b/0.16/generated/torchvision.transforms.CenterCrop.html
new file mode 100644
index 00000000000..1ff70a8709c
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.CenterCrop.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CenterCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FiveCrop" href="torchvision.transforms.FiveCrop.html" />
+    <link rel="prev" title="RandomResizedCrop" href="torchvision.transforms.RandomResizedCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>CenterCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.CenterCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="centercrop">
+<h1>CenterCrop<a class="headerlink" href="#centercrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.CenterCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">CenterCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#CenterCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.CenterCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crops the given image at the center.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.
+If image size is smaller than output size along any edge, image is padded with 0 and then center cropped.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">CenterCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.CenterCrop.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#CenterCrop.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.CenterCrop.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Cropped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.FiveCrop.html" class="btn btn-neutral float-right" title="FiveCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomResizedCrop.html" class="btn btn-neutral" title="RandomResizedCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CenterCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.ColorJitter.html b/0.16/generated/torchvision.transforms.ColorJitter.html
new file mode 100644
index 00000000000..a168182fc9a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.ColorJitter.html
@@ -0,0 +1,765 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ColorJitter &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Grayscale" href="torchvision.transforms.Grayscale.html" />
+    <link rel="prev" title="RandomVerticalFlip" href="torchvision.transforms.RandomVerticalFlip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ColorJitter</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.ColorJitter.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="colorjitter">
+<h1>ColorJitter<a class="headerlink" href="#colorjitter" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.ColorJitter">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">ColorJitter</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">brightness</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">contrast</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">saturation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hue</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ColorJitter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ColorJitter" title="Permalink to this definition">¶</a></dt>
+<dd><p>Randomly change the brightness, contrast, saturation and hue of an image.
+If the image is torch Tensor, it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, mode “1”, “I”, “F” and modes with transparency (alpha channel) are not supported.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>brightness</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter brightness.
+brightness_factor is chosen uniformly from [max(0, 1 - brightness), 1 + brightness]
+or the given [min, max]. Should be non negative numbers.</p></li>
+<li><p><strong>contrast</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter contrast.
+contrast_factor is chosen uniformly from [max(0, 1 - contrast), 1 + contrast]
+or the given [min, max]. Should be non-negative numbers.</p></li>
+<li><p><strong>saturation</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter saturation.
+saturation_factor is chosen uniformly from [max(0, 1 - saturation), 1 + saturation]
+or the given [min, max]. Should be non negative numbers.</p></li>
+<li><p><strong>hue</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter hue.
+hue_factor is chosen uniformly from [-hue, hue] or the given [min, max].
+Should have 0&lt;= hue &lt;= 0.5 or -0.5 &lt;= min &lt;= max &lt;= 0.5.
+To jitter hue, the pixel values of the input image has to be non-negative for conversion to HSV space;
+thus it does not work if you normalize your image to an interval with negative values,
+or use an interpolation that generates negative values before using this function.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">ColorJitter</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.ColorJitter.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ColorJitter.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ColorJitter.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Input image.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Color jittered image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.ColorJitter.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">brightness</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">contrast</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">saturation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hue</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ColorJitter.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ColorJitter.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get the parameters for the randomized transform to be applied on image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>brightness</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the brightness_factor is chosen
+uniformly. Pass None to turn off the transformation.</p></li>
+<li><p><strong>contrast</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the contrast_factor is chosen
+uniformly. Pass None to turn off the transformation.</p></li>
+<li><p><strong>saturation</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the saturation_factor is chosen
+uniformly. Pass None to turn off the transformation.</p></li>
+<li><p><strong>hue</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the hue_factor is chosen uniformly.
+Pass None to turn off the transformation.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The parameters used to apply the randomized transform
+along with their random order.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.Grayscale.html" class="btn btn-neutral float-right" title="Grayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomVerticalFlip.html" class="btn btn-neutral" title="RandomVerticalFlip" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ColorJitter</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.Compose.html b/0.16/generated/torchvision.transforms.Compose.html
new file mode 100644
index 00000000000..ff9dc550ba2
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.Compose.html
@@ -0,0 +1,737 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Compose &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomApply" href="torchvision.transforms.RandomApply.html" />
+    <link rel="prev" title="RandomEqualize" href="torchvision.transforms.RandomEqualize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Compose</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.Compose.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="compose">
+<h1>Compose<a class="headerlink" href="#compose" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.Compose">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">Compose</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Compose"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Compose" title="Permalink to this definition">¶</a></dt>
+<dd><p>Composes several transforms together. This transform does not support torchscript.
+Please, see the note below.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>transforms</strong> (list of <code class="docutils literal notranslate"><span class="pre">Transform</span></code> objects) – list of transforms to compose.</p>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transforms</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">CenterCrop</span><span class="p">(</span><span class="mi">10</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">PILToTensor</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">ConvertImageDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">])</span>
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In order to script the transformations, please use <code class="docutils literal notranslate"><span class="pre">torch.nn.Sequential</span></code> as below.</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">CenterCrop</span><span class="p">(</span><span class="mi">10</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">Normalize</span><span class="p">((</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">),</span> <span class="p">(</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">)),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">scripted_transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">script</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Make sure to use only scriptable transformations, i.e. that work with <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code>, does not require
+<cite>lambda</cite> functions or <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>.</p>
+</div>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Compose</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the..."><img alt="" src="../_images/sphx_glr_plot_video_api_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_video_api.html#sphx-glr-auto-examples-others-plot-video-api-py"><span class="std std-ref">Video API</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Video API</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomApply.html" class="btn btn-neutral float-right" title="RandomApply" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomEqualize.html" class="btn btn-neutral" title="RandomEqualize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Compose</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.ConvertImageDtype.html b/0.16/generated/torchvision.transforms.ConvertImageDtype.html
new file mode 100644
index 00000000000..61a87942dc7
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.ConvertImageDtype.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ConvertImageDtype &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="AutoAugmentPolicy" href="torchvision.transforms.AutoAugmentPolicy.html" />
+    <link rel="prev" title="PILToTensor" href="torchvision.transforms.PILToTensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ConvertImageDtype</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.ConvertImageDtype.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convertimagedtype">
+<h1>ConvertImageDtype<a class="headerlink" href="#convertimagedtype" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.ConvertImageDtype">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">ConvertImageDtype</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ConvertImageDtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ConvertImageDtype" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a tensor image to the given <code class="docutils literal notranslate"><span class="pre">dtype</span></code> and scale the values accordingly.</p>
+<p>This function does not support PIL Image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>dtype</strong> (<em>torch.dpython:type</em>) – Desired data type of the output</p>
+</dd>
+</dl>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>When converting from a smaller to a larger integer <code class="docutils literal notranslate"><span class="pre">dtype</span></code> the maximum values are <strong>not</strong> mapped exactly.
+If converted back and forth, this mismatch has no effect.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Raises<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#RuntimeError" title="(in Python v3.12)"><strong>RuntimeError</strong></a> – When trying to cast <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.float32</span></code> to <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int32</span></code> or <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int64</span></code> as
+    well as for trying to cast <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.float64</span></code> to <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int64</span></code>. These conversions might lead to
+    overflow errors since the floating point <code class="docutils literal notranslate"><span class="pre">dtype</span></code> cannot store consecutive integers over the whole range
+    of the integer <code class="docutils literal notranslate"><span class="pre">dtype</span></code>.</p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.ConvertImageDtype.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ConvertImageDtype.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ConvertImageDtype.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.AutoAugmentPolicy.html" class="btn btn-neutral float-right" title="AutoAugmentPolicy" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.PILToTensor.html" class="btn btn-neutral" title="PILToTensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ConvertImageDtype</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.ElasticTransform.html b/0.16/generated/torchvision.transforms.ElasticTransform.html
new file mode 100644
index 00000000000..d142ca2d181
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.ElasticTransform.html
@@ -0,0 +1,738 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ElasticTransform &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomHorizontalFlip" href="torchvision.transforms.RandomHorizontalFlip.html" />
+    <link rel="prev" title="RandomPerspective" href="torchvision.transforms.RandomPerspective.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ElasticTransform</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.ElasticTransform.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="elastictransform">
+<h1>ElasticTransform<a class="headerlink" href="#elastictransform" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.ElasticTransform">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">ElasticTransform</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">5.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ElasticTransform"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ElasticTransform" title="Permalink to this definition">¶</a></dt>
+<dd><p>Transform a tensor image with elastic transformations.
+Given alpha and sigma, it will generate displacement
+vectors for all pixels based on random offsets. Alpha controls the strength
+and sigma controls the smoothness of the displacements.
+The displacements are added to an identity grid and the resulting grid is
+used to grid_sample from the image.</p>
+<dl class="simple">
+<dt>Applications:</dt><dd><p>Randomly transforms the morphology of objects in images and produces a
+see-through-water-like effect.</p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>sequence of python:floats</em>) – Magnitude of displacements. Default is 50.0.</p></li>
+<li><p><strong>sigma</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>sequence of python:floats</em>) – Smoothness of displacements. Default is 5.0.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em>) – Pixel fill value for the area outside the transformed
+image. Default is <code class="docutils literal notranslate"><span class="pre">0</span></code>. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">ElasticTransform</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.ElasticTransform.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ElasticTransform.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ElasticTransform.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>tensor</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be transformed.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Transformed image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomHorizontalFlip.html" class="btn btn-neutral float-right" title="RandomHorizontalFlip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomPerspective.html" class="btn btn-neutral" title="RandomPerspective" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ElasticTransform</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.FiveCrop.html b/0.16/generated/torchvision.transforms.FiveCrop.html
new file mode 100644
index 00000000000..2b601c8b8dd
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.FiveCrop.html
@@ -0,0 +1,739 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FiveCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="TenCrop" href="torchvision.transforms.TenCrop.html" />
+    <link rel="prev" title="CenterCrop" href="torchvision.transforms.CenterCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>FiveCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.FiveCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fivecrop">
+<h1>FiveCrop<a class="headerlink" href="#fivecrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.FiveCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">FiveCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#FiveCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.FiveCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crop the given image into four corners and the central crop.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading
+dimensions</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform returns a tuple of images and there may be a mismatch in the number of
+inputs and targets your Dataset returns. See below for an example of how to deal with
+this.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an <code class="docutils literal notranslate"><span class="pre">int</span></code>
+instead of sequence like (h, w), a square crop of size (size, size) is made.
+If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transform</span> <span class="o">=</span> <span class="n">Compose</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>   <span class="n">FiveCrop</span><span class="p">(</span><span class="n">size</span><span class="p">),</span> <span class="c1"># this is a list of PIL Images</span>
+<span class="gp">&gt;&gt;&gt; </span>   <span class="n">Lambda</span><span class="p">(</span><span class="k">lambda</span> <span class="n">crops</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">([</span><span class="n">PILToTensor</span><span class="p">()(</span><span class="n">crop</span><span class="p">)</span> <span class="k">for</span> <span class="n">crop</span> <span class="ow">in</span> <span class="n">crops</span><span class="p">]))</span> <span class="c1"># returns a 4D tensor</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">])</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1">#In your test loop you can do the following:</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="nb">input</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">batch</span> <span class="c1"># input is a 5d tensor, target is 2d</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">bs</span><span class="p">,</span> <span class="n">ncrops</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">result</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span> <span class="c1"># fuse batch size and ncrops</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">result_avg</span> <span class="o">=</span> <span class="n">result</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">bs</span><span class="p">,</span> <span class="n">ncrops</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="c1"># avg over crops</span>
+</pre></div>
+</div>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">FiveCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.FiveCrop.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#FiveCrop.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.FiveCrop.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>tuple of 5 images. Image can be PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.TenCrop.html" class="btn btn-neutral float-right" title="TenCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.CenterCrop.html" class="btn btn-neutral" title="CenterCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FiveCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.GaussianBlur.html b/0.16/generated/torchvision.transforms.GaussianBlur.html
new file mode 100644
index 00000000000..2ce48578078
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.GaussianBlur.html
@@ -0,0 +1,753 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>GaussianBlur &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomInvert" href="torchvision.transforms.RandomInvert.html" />
+    <link rel="prev" title="RandomGrayscale" href="torchvision.transforms.RandomGrayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>GaussianBlur</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.GaussianBlur.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="gaussianblur">
+<h1>GaussianBlur<a class="headerlink" href="#gaussianblur" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.GaussianBlur">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">GaussianBlur</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">kernel_size</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.1,</span> <span class="pre">2.0)</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#GaussianBlur"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.GaussianBlur" title="Permalink to this definition">¶</a></dt>
+<dd><p>Blurs image with randomly chosen Gaussian blur.
+If the image is torch Tensor, it is expected
+to have […, C, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>kernel_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em>) – Size of the Gaussian kernel.</p></li>
+<li><p><strong>sigma</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – Standard deviation to be used for
+creating kernel to perform blurring. If float, sigma is fixed. If it is tuple
+of float (min, max), sigma is chosen uniformly at random to lie in the
+given range.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Gaussian blurred version of the input image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">GaussianBlur</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.GaussianBlur.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#GaussianBlur.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.GaussianBlur.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – image to be blurred.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Gaussian blurred image</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.GaussianBlur.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sigma_min</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma_max</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#GaussianBlur.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.GaussianBlur.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Choose sigma for random gaussian blurring.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>sigma_min</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Minimum standard deviation that can be chosen for blurring kernel.</p></li>
+<li><p><strong>sigma_max</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Maximum standard deviation that can be chosen for blurring kernel.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Standard deviation to be passed to calculate kernel for gaussian blurring.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)">float</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomInvert.html" class="btn btn-neutral float-right" title="RandomInvert" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomGrayscale.html" class="btn btn-neutral" title="RandomGrayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">GaussianBlur</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.Grayscale.html b/0.16/generated/torchvision.transforms.Grayscale.html
new file mode 100644
index 00000000000..cbfa68d1ebe
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.Grayscale.html
@@ -0,0 +1,732 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Grayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomGrayscale" href="torchvision.transforms.RandomGrayscale.html" />
+    <link rel="prev" title="ColorJitter" href="torchvision.transforms.ColorJitter.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Grayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.Grayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="grayscale">
+<h1>Grayscale<a class="headerlink" href="#grayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.Grayscale">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">Grayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_output_channels</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Grayscale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Grayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert image to grayscale.
+If the image is torch Tensor, it is expected
+to have […, 3, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>num_output_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – (1 or 3) number of channels desired for output image</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><p>Grayscale version of the input.</p>
+<ul class="simple">
+<li><p>If <code class="docutils literal notranslate"><span class="pre">num_output_channels</span> <span class="pre">==</span> <span class="pre">1</span></code> : returned image is single channel</p></li>
+<li><p>If <code class="docutils literal notranslate"><span class="pre">num_output_channels</span> <span class="pre">==</span> <span class="pre">3</span></code> : returned image is 3 channel with r == g == b</p></li>
+</ul>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Grayscale</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.Grayscale.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Grayscale.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Grayscale.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be converted to grayscale.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Grayscaled image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomGrayscale.html" class="btn btn-neutral float-right" title="RandomGrayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.ColorJitter.html" class="btn btn-neutral" title="ColorJitter" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Grayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.Lambda.html b/0.16/generated/torchvision.transforms.Lambda.html
new file mode 100644
index 00000000000..552e35e0271
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.Lambda.html
@@ -0,0 +1,699 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Lambda &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ToPILImage" href="torchvision.transforms.ToPILImage.html" />
+    <link rel="prev" title="RandomErasing" href="torchvision.transforms.RandomErasing.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Lambda</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.Lambda.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lambda">
+<h1>Lambda<a class="headerlink" href="#lambda" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.Lambda">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">Lambda</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">lambd</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Lambda"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Lambda" title="Permalink to this definition">¶</a></dt>
+<dd><p>Apply a user-defined lambda as a transform. This transform does not support torchscript.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>lambd</strong> (<em>function</em>) – Lambda/function to be used for transform.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.ToPILImage.html" class="btn btn-neutral float-right" title="ToPILImage" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomErasing.html" class="btn btn-neutral" title="RandomErasing" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Lambda</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.LinearTransformation.html b/0.16/generated/torchvision.transforms.LinearTransformation.html
new file mode 100644
index 00000000000..670bb6a4a05
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.LinearTransformation.html
@@ -0,0 +1,730 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>LinearTransformation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Normalize" href="torchvision.transforms.Normalize.html" />
+    <link rel="prev" title="RandomOrder" href="torchvision.transforms.RandomOrder.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>LinearTransformation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.LinearTransformation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lineartransformation">
+<h1>LinearTransformation<a class="headerlink" href="#lineartransformation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.LinearTransformation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">LinearTransformation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transformation_matrix</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mean_vector</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#LinearTransformation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.LinearTransformation" title="Permalink to this definition">¶</a></dt>
+<dd><p>Transform a tensor image with a square transformation matrix and a mean_vector computed
+offline.
+This transform does not support PIL Image.
+Given transformation_matrix and mean_vector, will flatten the torch.*Tensor and
+subtract mean_vector from it which is then followed by computing the dot
+product with the transformation matrix and then reshaping the tensor to its
+original shape.</p>
+<dl class="simple">
+<dt>Applications:</dt><dd><p>whitening transformation: Suppose X is a column vector zero-centered data.
+Then compute the data covariance matrix [D x D] with torch.mm(X.t(), X),
+perform SVD on this matrix and pass it as transformation_matrix.</p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>transformation_matrix</strong> (<em>Tensor</em>) – tensor [D x D], D = C x H x W</p></li>
+<li><p><strong>mean_vector</strong> (<em>Tensor</em>) – tensor [D], D = C x H x W</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.LinearTransformation.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#LinearTransformation.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.LinearTransformation.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>tensor</strong> (<em>Tensor</em>) – Tensor image to be whitened.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Transformed image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.Normalize.html" class="btn btn-neutral float-right" title="Normalize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomOrder.html" class="btn btn-neutral" title="RandomOrder" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">LinearTransformation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.Normalize.html b/0.16/generated/torchvision.transforms.Normalize.html
new file mode 100644
index 00000000000..d16cc57d9da
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.Normalize.html
@@ -0,0 +1,738 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Normalize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomErasing" href="torchvision.transforms.RandomErasing.html" />
+    <link rel="prev" title="LinearTransformation" href="torchvision.transforms.LinearTransformation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Normalize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.Normalize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="normalize">
+<h1>Normalize<a class="headerlink" href="#normalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.Normalize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">Normalize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">mean</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">std</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Normalize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Normalize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Normalize a tensor image with mean and standard deviation.
+This transform does not support PIL Image.
+Given mean: <code class="docutils literal notranslate"><span class="pre">(mean[1],...,mean[n])</span></code> and std: <code class="docutils literal notranslate"><span class="pre">(std[1],..,std[n])</span></code> for <code class="docutils literal notranslate"><span class="pre">n</span></code>
+channels, this transform will normalize each channel of the input
+<code class="docutils literal notranslate"><span class="pre">torch.*Tensor</span></code> i.e.,
+<code class="docutils literal notranslate"><span class="pre">output[channel]</span> <span class="pre">=</span> <span class="pre">(input[channel]</span> <span class="pre">-</span> <span class="pre">mean[channel])</span> <span class="pre">/</span> <span class="pre">std[channel]</span></code></p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform acts out of place, i.e., it does not mutate the input tensor.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>mean</strong> (<em>sequence</em>) – Sequence of means for each channel.</p></li>
+<li><p><strong>std</strong> (<em>sequence</em>) – Sequence of standard deviations for each channel.</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>,</em><em>optional</em>) – Bool to make this operation in-place.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Normalize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.Normalize.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Normalize.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Normalize.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>tensor</strong> (<em>Tensor</em>) – Tensor image to be normalized.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Normalized Tensor image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomErasing.html" class="btn btn-neutral float-right" title="RandomErasing" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.LinearTransformation.html" class="btn btn-neutral" title="LinearTransformation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Normalize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.PILToTensor.html b/0.16/generated/torchvision.transforms.PILToTensor.html
new file mode 100644
index 00000000000..58d5e490713
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.PILToTensor.html
@@ -0,0 +1,700 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>PILToTensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ConvertImageDtype" href="torchvision.transforms.ConvertImageDtype.html" />
+    <link rel="prev" title="ToTensor" href="torchvision.transforms.ToTensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>PILToTensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.PILToTensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="piltotensor">
+<h1>PILToTensor<a class="headerlink" href="#piltotensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.PILToTensor">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">PILToTensor</span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#PILToTensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.PILToTensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a PIL Image to a tensor of the same type - this does not scale values.</p>
+<p>This transform does not support torchscript.</p>
+<p>Converts a PIL Image (H x W x C) to a Tensor of shape (C x H x W).</p>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">PILToTensor</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.ConvertImageDtype.html" class="btn btn-neutral float-right" title="ConvertImageDtype" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.ToTensor.html" class="btn btn-neutral" title="ToTensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">PILToTensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.Pad.html b/0.16/generated/torchvision.transforms.Pad.html
new file mode 100644
index 00000000000..e6b941168d7
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.Pad.html
@@ -0,0 +1,753 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Pad &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomRotation" href="torchvision.transforms.RandomRotation.html" />
+    <link rel="prev" title="TenCrop" href="torchvision.transforms.TenCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Pad</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.Pad.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="pad">
+<h1>Pad<a class="headerlink" href="#pad" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.Pad">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">Pad</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">padding</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_mode</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'constant'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Pad"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Pad" title="Permalink to this definition">¶</a></dt>
+<dd><p>Pad the given image on all sides with the given “pad” value.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means at most 2 leading dimensions for mode reflect and symmetric,
+at most 3 leading dimensions for mode edge,
+and an arbitrary number of leading dimensions for mode constant</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em>) – <p>Padding on each border. If a single int is provided this
+is used to pad all borders. If sequence of length 2 is provided this is the padding
+on left/right and top/bottom respectively. If a sequence of length 4 is provided
+this is the padding for the left, top, right and bottom borders respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode padding as single int is not supported, use a sequence of
+length 1: <code class="docutils literal notranslate"><span class="pre">[padding,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a>) – Pixel fill value for constant fill. Default is 0. If a tuple of
+length 3, it is used to fill R, G, B channels respectively.
+This value is only used when the padding_mode is constant.
+Only number is supported for torch Tensor.
+Only int or tuple value is supported for PIL Image.</p></li>
+<li><p><strong>padding_mode</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – <p>Type of padding. Should be: constant, edge, reflect or symmetric.
+Default is constant.</p>
+<ul>
+<li><p>constant: pads with a constant value, this value is specified with fill</p></li>
+<li><p>edge: pads with the last value at the edge of the image.
+If input a 5D torch Tensor, the last 3 dimensions will be padded instead of the last 2</p></li>
+<li><p>reflect: pads with reflection of image without repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode
+will result in [3, 2, 1, 2, 3, 4, 3, 2]</p></li>
+<li><p>symmetric: pads with reflection of image repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode
+will result in [2, 1, 1, 2, 3, 4, 4, 3]</p></li>
+</ul>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Pad</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.Pad.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Pad.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Pad.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be padded.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Padded image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomRotation.html" class="btn btn-neutral float-right" title="RandomRotation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.TenCrop.html" class="btn btn-neutral" title="TenCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Pad</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandAugment.html b/0.16/generated/torchvision.transforms.RandAugment.html
new file mode 100644
index 00000000000..d075f588062
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandAugment.html
@@ -0,0 +1,732 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandAugment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="TrivialAugmentWide" href="torchvision.transforms.TrivialAugmentWide.html" />
+    <link rel="prev" title="AutoAugment" href="torchvision.transforms.AutoAugment.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandAugment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandAugment.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randaugment">
+<h1>RandAugment<a class="headerlink" href="#randaugment" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandAugment">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandAugment</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_ops</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">magnitude</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">9</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_magnitude_bins</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">31</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#RandAugment"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandAugment" title="Permalink to this definition">¶</a></dt>
+<dd><p>RandAugment data augmentation method based on
+<a class="reference external" href="https://arxiv.org/abs/1909.13719">“RandAugment: Practical automated data augmentation with a reduced search space”</a>.
+If the image is torch Tensor, it should be of type torch.uint8, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>num_ops</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Number of augmentation transformations to apply sequentially.</p></li>
+<li><p><strong>magnitude</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Magnitude for all the transformations.</p></li>
+<li><p><strong>num_magnitude_bins</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The number of different magnitude values.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandAugment</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandAugment.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#RandAugment.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandAugment.forward" title="Permalink to this definition">¶</a></dt>
+<dd><blockquote>
+<div><p>img (PIL Image or Tensor): Image to be transformed.</p>
+</div></blockquote>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Transformed image.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.TrivialAugmentWide.html" class="btn btn-neutral float-right" title="TrivialAugmentWide" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.AutoAugment.html" class="btn btn-neutral" title="AutoAugment" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandAugment</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html b/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
new file mode 100644
index 00000000000..fb2aead4b3a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
@@ -0,0 +1,725 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomAdjustSharpness &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomAutocontrast" href="torchvision.transforms.RandomAutocontrast.html" />
+    <link rel="prev" title="RandomSolarize" href="torchvision.transforms.RandomSolarize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomAdjustSharpness</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomAdjustSharpness.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomadjustsharpness">
+<h1>RandomAdjustSharpness<a class="headerlink" href="#randomadjustsharpness" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomAdjustSharpness">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomAdjustSharpness</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sharpness_factor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAdjustSharpness"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomAdjustSharpness" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust the sharpness of the image randomly with a given probability. If the image is torch Tensor,
+it is expected to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>sharpness_factor</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – How much to adjust the sharpness. Can be
+any non-negative number. 0 gives a blurred image, 1 gives the
+original image while 2 increases the sharpness by a factor of 2.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being sharpened. Default value is 0.5</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomAdjustSharpness</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomAdjustSharpness.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAdjustSharpness.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomAdjustSharpness.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be sharpened.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly sharpened image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomAutocontrast.html" class="btn btn-neutral float-right" title="RandomAutocontrast" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomSolarize.html" class="btn btn-neutral" title="RandomSolarize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomAdjustSharpness</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomAffine.html b/0.16/generated/torchvision.transforms.RandomAffine.html
new file mode 100644
index 00000000000..2b4bd491f16
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomAffine.html
@@ -0,0 +1,756 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomAffine &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomPerspective" href="torchvision.transforms.RandomPerspective.html" />
+    <link rel="prev" title="RandomRotation" href="torchvision.transforms.RandomRotation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomAffine</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomAffine.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomaffine">
+<h1>RandomAffine<a class="headerlink" href="#randomaffine" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomAffine">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomAffine</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">translate</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shear</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAffine"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomAffine" title="Permalink to this definition">¶</a></dt>
+<dd><p>Random affine transformation of the image keeping center invariant.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>degrees</strong> (<em>sequence</em><em> or </em><em>number</em>) – Range of degrees to select from.
+If degrees is a number instead of sequence like (min, max), the range of degrees
+will be (-degrees, +degrees). Set to 0 to deactivate rotations.</p></li>
+<li><p><strong>translate</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em>, </em><em>optional</em>) – tuple of maximum absolute fraction for horizontal
+and vertical translations. For example translate=(a, b), then horizontal shift
+is randomly sampled in the range -img_width * a &lt; dx &lt; img_width * a and vertical shift is
+randomly sampled in the range -img_height * b &lt; dy &lt; img_height * b. Will not translate by default.</p></li>
+<li><p><strong>scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em>, </em><em>optional</em>) – scaling factor interval, e.g (a, b), then scale is
+randomly sampled from the range a &lt;= scale &lt;= b. Will keep original scale by default.</p></li>
+<li><p><strong>shear</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Range of degrees to select from.
+If shear is a number, a shear parallel to the x-axis in the range (-shear, +shear)
+will be applied. Else if shear is a sequence of 2 values a shear parallel to the x-axis in the
+range (shear[0], shear[1]) will be applied. Else if shear is a sequence of 4 values,
+an x-axis shear in (shear[0], shear[1]) and y-axis shear in (shear[2], shear[3]) will be applied.
+Will not apply shear by default.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em>) – Pixel fill value for the area outside the transformed
+image. Default is <code class="docutils literal notranslate"><span class="pre">0</span></code>. If given a number, the value is used for all bands respectively.</p></li>
+<li><p><strong>center</strong> (<em>sequence</em><em>, </em><em>optional</em>) – Optional center of rotation, (x, y). Origin is the upper left corner.
+Default is the center of the image.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomAffine</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomAffine.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAffine.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomAffine.forward" title="Permalink to this definition">¶</a></dt>
+<dd><blockquote>
+<div><p>img (PIL Image or Tensor): Image to be transformed.</p>
+</div></blockquote>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Affine transformed image.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomAffine.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">translate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale_ranges</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shears</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">img_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAffine.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomAffine.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for affine transformation</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>params to be passed to the affine transformation</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomPerspective.html" class="btn btn-neutral float-right" title="RandomPerspective" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomRotation.html" class="btn btn-neutral" title="RandomRotation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomAffine</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomApply.html b/0.16/generated/torchvision.transforms.RandomApply.html
new file mode 100644
index 00000000000..e6d65d8c51b
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomApply.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomApply &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomChoice" href="torchvision.transforms.RandomChoice.html" />
+    <link rel="prev" title="Compose" href="torchvision.transforms.Compose.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomApply</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomApply.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomapply">
+<h1>RandomApply<a class="headerlink" href="#randomapply" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomApply">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomApply</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomApply"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomApply" title="Permalink to this definition">¶</a></dt>
+<dd><p>Apply randomly a list of transformations with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In order to script the transformation, please use <code class="docutils literal notranslate"><span class="pre">torch.nn.ModuleList</span></code> as input instead of list/tuple of
+transforms as shown below:</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">RandomApply</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">ColorJitter</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">]),</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.3</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">scripted_transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">script</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Make sure to use only scriptable transformations, i.e. that work with <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code>, does not require
+<cite>lambda</cite> functions or <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>transforms</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a>) – list of transformations</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomApply</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomApply.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomApply.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomApply.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomChoice.html" class="btn btn-neutral float-right" title="RandomChoice" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.Compose.html" class="btn btn-neutral" title="Compose" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomApply</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomAutocontrast.html b/0.16/generated/torchvision.transforms.RandomAutocontrast.html
new file mode 100644
index 00000000000..c0acd76745e
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomAutocontrast.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomAutocontrast &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomEqualize" href="torchvision.transforms.RandomEqualize.html" />
+    <link rel="prev" title="RandomAdjustSharpness" href="torchvision.transforms.RandomAdjustSharpness.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomAutocontrast</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomAutocontrast.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomautocontrast">
+<h1>RandomAutocontrast<a class="headerlink" href="#randomautocontrast" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomAutocontrast">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomAutocontrast</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAutocontrast"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomAutocontrast" title="Permalink to this definition">¶</a></dt>
+<dd><p>Autocontrast the pixels of the given image randomly with a given probability.
+If the image is torch Tensor, it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being autocontrasted. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomAutocontrast.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAutocontrast.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomAutocontrast.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be autocontrasted.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly autocontrasted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomEqualize.html" class="btn btn-neutral float-right" title="RandomEqualize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomAdjustSharpness.html" class="btn btn-neutral" title="RandomAdjustSharpness" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomAutocontrast</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomChoice.html b/0.16/generated/torchvision.transforms.RandomChoice.html
new file mode 100644
index 00000000000..63e6cd25f9b
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomChoice.html
@@ -0,0 +1,698 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomChoice &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomOrder" href="torchvision.transforms.RandomOrder.html" />
+    <link rel="prev" title="RandomApply" href="torchvision.transforms.RandomApply.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomChoice</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomChoice.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomchoice">
+<h1>RandomChoice<a class="headerlink" href="#randomchoice" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomChoice">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomChoice</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomChoice"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomChoice" title="Permalink to this definition">¶</a></dt>
+<dd><p>Apply single transformation randomly picked from a list. This transform does not support torchscript.</p>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomChoice</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomOrder.html" class="btn btn-neutral float-right" title="RandomOrder" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomApply.html" class="btn btn-neutral" title="RandomApply" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomChoice</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomCrop.html b/0.16/generated/torchvision.transforms.RandomCrop.html
new file mode 100644
index 00000000000..4fad394dcb2
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomCrop.html
@@ -0,0 +1,785 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomResizedCrop" href="torchvision.transforms.RandomResizedCrop.html" />
+    <link rel="prev" title="Resize" href="torchvision.transforms.Resize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomcrop">
+<h1>RandomCrop<a class="headerlink" href="#randomcrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_if_needed</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_mode</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'constant'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crop the given image at a random location.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions,
+but if non-constant padding is used, the input is expected to have at most 2 leading dimensions</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p></li>
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em><em>, </em><em>optional</em>) – <p>Optional padding on each border
+of the image. Default is None. If a single int is provided this
+is used to pad all borders. If sequence of length 2 is provided this is the padding
+on left/right and top/bottom respectively. If a sequence of length 4 is provided
+this is the padding for the left, top, right and bottom borders respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode padding as single int is not supported, use a sequence of
+length 1: <code class="docutils literal notranslate"><span class="pre">[padding,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>pad_if_needed</strong> (<em>boolean</em>) – It will pad the image if smaller than the
+desired size to avoid raising an exception. Since cropping is done
+after padding, the padding seems to be done at a random offset.</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a>) – Pixel fill value for constant fill. Default is 0. If a tuple of
+length 3, it is used to fill R, G, B channels respectively.
+This value is only used when the padding_mode is constant.
+Only number is supported for torch Tensor.
+Only int or tuple value is supported for PIL Image.</p></li>
+<li><p><strong>padding_mode</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – <p>Type of padding. Should be: constant, edge, reflect or symmetric.
+Default is constant.</p>
+<ul>
+<li><p>constant: pads with a constant value, this value is specified with fill</p></li>
+<li><p>edge: pads with the last value at the edge of the image.
+If input a 5D torch Tensor, the last 3 dimensions will be padded instead of the last 2</p></li>
+<li><p>reflect: pads with reflection of image without repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode
+will result in [3, 2, 1, 2, 3, 4, 3, 2]</p></li>
+<li><p>symmetric: pads with reflection of image repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode
+will result in [2, 1, 1, 2, 3, 4, 4, 3]</p></li>
+</ul>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images."><img alt="" src="../_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomCrop.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomCrop.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomCrop.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Cropped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomCrop.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomCrop.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomCrop.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">crop</span></code> for a random crop.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p></li>
+<li><p><strong>output_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a>) – Expected output size of the crop.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>params (i, j, h, w) to be passed to <code class="docutils literal notranslate"><span class="pre">crop</span></code> for random crop.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomResizedCrop.html" class="btn btn-neutral float-right" title="RandomResizedCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.Resize.html" class="btn btn-neutral" title="Resize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomEqualize.html b/0.16/generated/torchvision.transforms.RandomEqualize.html
new file mode 100644
index 00000000000..dd76d823d6d
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomEqualize.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomEqualize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Compose" href="torchvision.transforms.Compose.html" />
+    <link rel="prev" title="RandomAutocontrast" href="torchvision.transforms.RandomAutocontrast.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomEqualize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomEqualize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomequalize">
+<h1>RandomEqualize<a class="headerlink" href="#randomequalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomEqualize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomEqualize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomEqualize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomEqualize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Equalize the histogram of the given image randomly with a given probability.
+If the image is torch Tensor, it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “P”, “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being equalized. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomEqualize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomEqualize.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomEqualize.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomEqualize.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be equalized.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly equalized image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.Compose.html" class="btn btn-neutral float-right" title="Compose" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomAutocontrast.html" class="btn btn-neutral" title="RandomAutocontrast" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomEqualize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomErasing.html b/0.16/generated/torchvision.transforms.RandomErasing.html
new file mode 100644
index 00000000000..8bafeaf7e02
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomErasing.html
@@ -0,0 +1,763 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomErasing &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Lambda" href="torchvision.transforms.Lambda.html" />
+    <link rel="prev" title="Normalize" href="torchvision.transforms.Normalize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomErasing</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomErasing.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomerasing">
+<h1>RandomErasing<a class="headerlink" href="#randomerasing" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomErasing">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomErasing</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.02,</span> <span class="pre">0.33)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.3,</span> <span class="pre">3.3)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">value</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomErasing"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomErasing" title="Permalink to this definition">¶</a></dt>
+<dd><p>Randomly selects a rectangle region in a torch.Tensor image and erases its pixels.
+This transform does not support PIL Image.
+‘Random Erasing Data Augmentation’ by Zhong et al. See <a class="reference external" href="https://arxiv.org/abs/1708.04896">https://arxiv.org/abs/1708.04896</a></p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>p</strong> – probability that the random erasing operation will be performed.</p></li>
+<li><p><strong>scale</strong> – range of proportion of erased area against input image.</p></li>
+<li><p><strong>ratio</strong> – range of aspect ratio of erased area.</p></li>
+<li><p><strong>value</strong> – erasing value. Default is 0. If a single int, it is used to
+erase all pixels. If a tuple of length 3, it is used to erase
+R, G, B channels respectively.
+If a str of ‘random’, erasing each pixel with random values.</p></li>
+<li><p><strong>inplace</strong> – boolean to make this transform inplace. Default set to False.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Erased Image.</p>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transform</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">PILToTensor</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">ConvertImageDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">Normalize</span><span class="p">((</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">),</span> <span class="p">(</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">)),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">RandomErasing</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">])</span>
+</pre></div>
+</div>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomErasing.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomErasing.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomErasing.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>Tensor</em>) – Tensor image to be erased.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Erased Tensor image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>img (Tensor)</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomErasing.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomErasing.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomErasing.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">erase</span></code> for a random erasing.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>Tensor</em>) – Tensor image to be erased.</p></li>
+<li><p><strong>scale</strong> (<em>sequence</em>) – range of proportion of erased area against input image.</p></li>
+<li><p><strong>ratio</strong> (<em>sequence</em>) – range of aspect ratio of erased area.</p></li>
+<li><p><strong>value</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em>, </em><em>optional</em>) – erasing value. If None, it is interpreted as “random”
+(erasing each pixel with random values). If <code class="docutils literal notranslate"><span class="pre">len(value)</span></code> is 1, it is interpreted as a number,
+i.e. <code class="docutils literal notranslate"><span class="pre">value[0]</span></code>.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>params (i, j, h, w, v) to be passed to <code class="docutils literal notranslate"><span class="pre">erase</span></code> for random erasing.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.Lambda.html" class="btn btn-neutral float-right" title="Lambda" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.Normalize.html" class="btn btn-neutral" title="Normalize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomErasing</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomGrayscale.html b/0.16/generated/torchvision.transforms.RandomGrayscale.html
new file mode 100644
index 00000000000..d7e3726d972
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomGrayscale.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomGrayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="GaussianBlur" href="torchvision.transforms.GaussianBlur.html" />
+    <link rel="prev" title="Grayscale" href="torchvision.transforms.Grayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomGrayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomGrayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomgrayscale">
+<h1>RandomGrayscale<a class="headerlink" href="#randomgrayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomGrayscale">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomGrayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomGrayscale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomGrayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>Randomly convert image to grayscale with a probability of p (default 0.1).
+If the image is torch Tensor, it is expected
+to have […, 3, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability that image should be converted to grayscale.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Grayscale version of the input image with probability p and unchanged
+with probability (1-p).
+- If input image is 1 channel: grayscale version is 1 channel
+- If input image is 3 channel: grayscale version is 3 channel with r == g == b</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomGrayscale.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomGrayscale.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomGrayscale.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be converted to grayscale.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly grayscaled image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.GaussianBlur.html" class="btn btn-neutral float-right" title="GaussianBlur" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.Grayscale.html" class="btn btn-neutral" title="Grayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomGrayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html b/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
new file mode 100644
index 00000000000..9ad1632b198
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
@@ -0,0 +1,740 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomHorizontalFlip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomVerticalFlip" href="torchvision.transforms.RandomVerticalFlip.html" />
+    <link rel="prev" title="ElasticTransform" href="torchvision.transforms.ElasticTransform.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomHorizontalFlip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomHorizontalFlip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomhorizontalflip">
+<h1>RandomHorizontalFlip<a class="headerlink" href="#randomhorizontalflip" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomHorizontalFlip">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomHorizontalFlip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomHorizontalFlip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomHorizontalFlip" title="Permalink to this definition">¶</a></dt>
+<dd><p>Horizontally flip the given image randomly with a given probability.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading
+dimensions</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being flipped. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomHorizontalFlip</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images."><img alt="" src="../_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomHorizontalFlip.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomHorizontalFlip.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomHorizontalFlip.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be flipped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly flipped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomVerticalFlip.html" class="btn btn-neutral float-right" title="RandomVerticalFlip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.ElasticTransform.html" class="btn btn-neutral" title="ElasticTransform" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomHorizontalFlip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomInvert.html b/0.16/generated/torchvision.transforms.RandomInvert.html
new file mode 100644
index 00000000000..21886fd0ca1
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomInvert.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomInvert &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomPosterize" href="torchvision.transforms.RandomPosterize.html" />
+    <link rel="prev" title="GaussianBlur" href="torchvision.transforms.GaussianBlur.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomInvert</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomInvert.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randominvert">
+<h1>RandomInvert<a class="headerlink" href="#randominvert" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomInvert">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomInvert</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomInvert"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomInvert" title="Permalink to this definition">¶</a></dt>
+<dd><p>Inverts the colors of the given image randomly with a given probability.
+If img is a Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being color inverted. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomInvert</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomInvert.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomInvert.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomInvert.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be inverted.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly color inverted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomPosterize.html" class="btn btn-neutral float-right" title="RandomPosterize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.GaussianBlur.html" class="btn btn-neutral" title="GaussianBlur" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomInvert</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomOrder.html b/0.16/generated/torchvision.transforms.RandomOrder.html
new file mode 100644
index 00000000000..9d847a6075b
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomOrder.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomOrder &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="LinearTransformation" href="torchvision.transforms.LinearTransformation.html" />
+    <link rel="prev" title="RandomChoice" href="torchvision.transforms.RandomChoice.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomOrder</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomOrder.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomorder">
+<h1>RandomOrder<a class="headerlink" href="#randomorder" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomOrder">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomOrder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomOrder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomOrder" title="Permalink to this definition">¶</a></dt>
+<dd><p>Apply a list of transformations in a random order. This transform does not support torchscript.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.LinearTransformation.html" class="btn btn-neutral float-right" title="LinearTransformation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomChoice.html" class="btn btn-neutral" title="RandomChoice" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomOrder</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomPerspective.html b/0.16/generated/torchvision.transforms.RandomPerspective.html
new file mode 100644
index 00000000000..d3123acc6a2
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomPerspective.html
@@ -0,0 +1,750 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomPerspective &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ElasticTransform" href="torchvision.transforms.ElasticTransform.html" />
+    <link rel="prev" title="RandomAffine" href="torchvision.transforms.RandomAffine.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomPerspective</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomPerspective.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomperspective">
+<h1>RandomPerspective<a class="headerlink" href="#randomperspective" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomPerspective">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomPerspective</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">distortion_scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomPerspective"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomPerspective" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs a random perspective transformation of the given image with a given probability.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>distortion_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – argument to control the degree of distortion and ranges from 0 to 1.
+Default is 0.5.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being transformed. Default is 0.5.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em>) – Pixel fill value for the area outside the transformed
+image. Default is <code class="docutils literal notranslate"><span class="pre">0</span></code>. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomPerspective</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomPerspective.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomPerspective.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomPerspective.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be Perspectively transformed.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly transformed image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomPerspective.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">distortion_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomPerspective.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomPerspective.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">perspective</span></code> for a random perspective transform.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – width of the image.</p></li>
+<li><p><strong>height</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – height of the image.</p></li>
+<li><p><strong>distortion_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – argument to control the degree of distortion and ranges from 0 to 1.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>List containing [top-left, top-right, bottom-right, bottom-left] of the original image,
+List containing [top-left, top-right, bottom-right, bottom-left] of the transformed image.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.ElasticTransform.html" class="btn btn-neutral float-right" title="ElasticTransform" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomAffine.html" class="btn btn-neutral" title="RandomAffine" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomPerspective</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomPosterize.html b/0.16/generated/torchvision.transforms.RandomPosterize.html
new file mode 100644
index 00000000000..dd82bbdc747
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomPosterize.html
@@ -0,0 +1,725 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomPosterize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomSolarize" href="torchvision.transforms.RandomSolarize.html" />
+    <link rel="prev" title="RandomInvert" href="torchvision.transforms.RandomInvert.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomPosterize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomPosterize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomposterize">
+<h1>RandomPosterize<a class="headerlink" href="#randomposterize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomPosterize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomPosterize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">bits</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomPosterize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomPosterize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Posterize the image randomly with a given probability by reducing the
+number of bits for each color channel. If the image is torch Tensor, it should be of type torch.uint8,
+and it is expected to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>bits</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of bits to keep for each channel (0-8)</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being posterized. Default value is 0.5</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomPosterize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomPosterize.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomPosterize.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomPosterize.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be posterized.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly posterized image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomSolarize.html" class="btn btn-neutral float-right" title="RandomSolarize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomInvert.html" class="btn btn-neutral" title="RandomInvert" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomPosterize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomResizedCrop.html b/0.16/generated/torchvision.transforms.RandomResizedCrop.html
new file mode 100644
index 00000000000..4ba310c3730
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomResizedCrop.html
@@ -0,0 +1,792 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomResizedCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CenterCrop" href="torchvision.transforms.CenterCrop.html" />
+    <link rel="prev" title="RandomCrop" href="torchvision.transforms.RandomCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomResizedCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomResizedCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomresizedcrop">
+<h1>RandomResizedCrop<a class="headerlink" href="#randomresizedcrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomResizedCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomResizedCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.08,</span> <span class="pre">1.0)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.75,</span> <span class="pre">1.3333333333333333)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomResizedCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomResizedCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crop a random portion of image and resize it to a given size.</p>
+<p>If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<p>A crop of the original image is made: the crop has a random area (H * W)
+and a random aspect ratio. This crop is finally resized to the given
+size. This is popularly used to train the Inception networks.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em>) – <p>expected output size of the crop, for each edge. If size is an
+int instead of sequence like (h, w), a square output size <code class="docutils literal notranslate"><span class="pre">(size,</span> <span class="pre">size)</span></code> is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode size as single int is not supported, use a sequence of length 1: <code class="docutils literal notranslate"><span class="pre">[size,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>scale</strong> (<em>tuple of python:float</em>) – Specifies the lower and upper bounds for the random area of the crop,
+before resizing. The scale is defined with respect to the area of the original image.</p></li>
+<li><p><strong>ratio</strong> (<em>tuple of python:float</em>) – lower and upper bounds for the random aspect ratio of the crop, before
+resizing.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomResizedCrop.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomResizedCrop.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomResizedCrop.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped and resized.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly cropped and resized image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomResizedCrop.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomResizedCrop.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomResizedCrop.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">crop</span></code> for a random sized crop.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Input image.</p></li>
+<li><p><strong>scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a>) – range of scale of the origin size cropped</p></li>
+<li><p><strong>ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a>) – range of aspect ratio of the origin aspect ratio cropped</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>params (i, j, h, w) to be passed to <code class="docutils literal notranslate"><span class="pre">crop</span></code> for a random
+sized crop.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.CenterCrop.html" class="btn btn-neutral float-right" title="CenterCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomCrop.html" class="btn btn-neutral" title="RandomCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomResizedCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomRotation.html b/0.16/generated/torchvision.transforms.RandomRotation.html
new file mode 100644
index 00000000000..7225ddde6a4
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomRotation.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomRotation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomAffine" href="torchvision.transforms.RandomAffine.html" />
+    <link rel="prev" title="Pad" href="torchvision.transforms.Pad.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomRotation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomRotation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomrotation">
+<h1>RandomRotation<a class="headerlink" href="#randomrotation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomRotation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomRotation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">expand</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomRotation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomRotation" title="Permalink to this definition">¶</a></dt>
+<dd><p>Rotate the image by angle.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>degrees</strong> (<em>sequence</em><em> or </em><em>number</em>) – Range of degrees to select from.
+If degrees is a number instead of sequence like (min, max), the range of degrees
+will be (-degrees, +degrees).</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>expand</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Optional expansion flag.
+If true, expands the output to make it large enough to hold the entire rotated image.
+If false or omitted, make the output image the same size as the input image.
+Note that the expand flag assumes rotation around the center and no translation.</p></li>
+<li><p><strong>center</strong> (<em>sequence</em><em>, </em><em>optional</em>) – Optional center of rotation, (x, y). Origin is the upper left corner.
+Default is the center of the image.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em>) – Pixel fill value for the area outside the rotated
+image. Default is <code class="docutils literal notranslate"><span class="pre">0</span></code>. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomRotation</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomRotation.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomRotation.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomRotation.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be rotated.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Rotated image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomRotation.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomRotation.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomRotation.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">rotate</span></code> for a random rotation.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>angle parameter to be passed to <code class="docutils literal notranslate"><span class="pre">rotate</span></code> for random rotation.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)">float</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomAffine.html" class="btn btn-neutral float-right" title="RandomAffine" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.Pad.html" class="btn btn-neutral" title="Pad" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomRotation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomSolarize.html b/0.16/generated/torchvision.transforms.RandomSolarize.html
new file mode 100644
index 00000000000..874660927c1
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomSolarize.html
@@ -0,0 +1,725 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomSolarize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomAdjustSharpness" href="torchvision.transforms.RandomAdjustSharpness.html" />
+    <link rel="prev" title="RandomPosterize" href="torchvision.transforms.RandomPosterize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomSolarize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomSolarize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomsolarize">
+<h1>RandomSolarize<a class="headerlink" href="#randomsolarize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomSolarize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomSolarize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">threshold</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomSolarize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomSolarize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Solarize the image randomly with a given probability by inverting all pixel
+values above a threshold. If img is a Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>threshold</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – all pixels equal or above this value are inverted.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being solarized. Default value is 0.5</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomSolarize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomSolarize.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomSolarize.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomSolarize.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be solarized.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly solarized image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomAdjustSharpness.html" class="btn btn-neutral float-right" title="RandomAdjustSharpness" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomPosterize.html" class="btn btn-neutral" title="RandomPosterize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomSolarize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.RandomVerticalFlip.html b/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
new file mode 100644
index 00000000000..e3f4b128f5b
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomVerticalFlip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ColorJitter" href="torchvision.transforms.ColorJitter.html" />
+    <link rel="prev" title="RandomHorizontalFlip" href="torchvision.transforms.RandomHorizontalFlip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomVerticalFlip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.RandomVerticalFlip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomverticalflip">
+<h1>RandomVerticalFlip<a class="headerlink" href="#randomverticalflip" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomVerticalFlip">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">RandomVerticalFlip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomVerticalFlip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomVerticalFlip" title="Permalink to this definition">¶</a></dt>
+<dd><p>Vertically flip the given image randomly with a given probability.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading
+dimensions</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being flipped. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomVerticalFlip</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.RandomVerticalFlip.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomVerticalFlip.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.RandomVerticalFlip.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be flipped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Randomly flipped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.ColorJitter.html" class="btn btn-neutral float-right" title="ColorJitter" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandomHorizontalFlip.html" class="btn btn-neutral" title="RandomHorizontalFlip" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomVerticalFlip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.Resize.html b/0.16/generated/torchvision.transforms.Resize.html
new file mode 100644
index 00000000000..3ecd987856f
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.Resize.html
@@ -0,0 +1,775 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Resize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomCrop" href="torchvision.transforms.RandomCrop.html" />
+    <link rel="prev" title="register_kernel" href="torchvision.transforms.v2.functional.register_kernel.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Resize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.Resize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resize">
+<h1>Resize<a class="headerlink" href="#resize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.Resize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">Resize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_size</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Resize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Resize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Resize the input image to the given size.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means a maximum of two leading dimensions</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The output image might be different depending on its type: when downsampling, the interpolation of PIL images
+and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences
+in the performance of a network. Therefore, it is preferable to train and serve a model with the same input
+types. See also below the <code class="docutils literal notranslate"><span class="pre">antialias</span></code> parameter, which can help making the output of PIL images and tensors
+closer.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – <p>Desired output size. If size is a sequence like
+(h, w), output size will be matched to this. If size is an int,
+smaller edge of the image will be matched to this number.
+i.e, if height &gt; width, then image will be rescaled to
+(size * height / width, size).</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode size as single int is not supported, use a sequence of length 1: <code class="docutils literal notranslate"><span class="pre">[size,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>max_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The maximum allowed for the longer edge of
+the resized image. If the longer edge of the image is greater
+than <code class="docutils literal notranslate"><span class="pre">max_size</span></code> after being resized according to <code class="docutils literal notranslate"><span class="pre">size</span></code>,
+<code class="docutils literal notranslate"><span class="pre">size</span></code> will be overruled so that the longer edge is equal to
+<code class="docutils literal notranslate"><span class="pre">max_size</span></code>.
+As a result, the smaller edge may be shorter than <code class="docutils literal notranslate"><span class="pre">size</span></code>. This
+is only supported if <code class="docutils literal notranslate"><span class="pre">size</span></code> is an int (or a sequence of length
+1 in torchscript mode).</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Resize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the..."><img alt="" src="../_images/sphx_glr_plot_video_api_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_video_api.html#sphx-glr-auto-examples-others-plot-video-api-py"><span class="std std-ref">Video API</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Video API</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.Resize.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#Resize.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.Resize.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be scaled.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Rescaled image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.RandomCrop.html" class="btn btn-neutral float-right" title="RandomCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.register_kernel.html" class="btn btn-neutral" title="register_kernel" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Resize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.TenCrop.html b/0.16/generated/torchvision.transforms.TenCrop.html
new file mode 100644
index 00000000000..8b8cf350dcd
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.TenCrop.html
@@ -0,0 +1,739 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>TenCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Pad" href="torchvision.transforms.Pad.html" />
+    <link rel="prev" title="FiveCrop" href="torchvision.transforms.FiveCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>TenCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.TenCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="tencrop">
+<h1>TenCrop<a class="headerlink" href="#tencrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.TenCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">TenCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">vertical_flip</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#TenCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.TenCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crop the given image into four corners and the central crop plus the flipped version of
+these (horizontal flipping is used by default).
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading
+dimensions</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform returns a tuple of images and there may be a mismatch in the number of
+inputs and targets your Dataset returns. See below for an example of how to deal with
+this.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p></li>
+<li><p><strong>vertical_flip</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – Use vertical flipping instead of horizontal</p></li>
+</ul>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transform</span> <span class="o">=</span> <span class="n">Compose</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>   <span class="n">TenCrop</span><span class="p">(</span><span class="n">size</span><span class="p">),</span> <span class="c1"># this is a tuple of PIL Images</span>
+<span class="gp">&gt;&gt;&gt; </span>   <span class="n">Lambda</span><span class="p">(</span><span class="k">lambda</span> <span class="n">crops</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">([</span><span class="n">PILToTensor</span><span class="p">()(</span><span class="n">crop</span><span class="p">)</span> <span class="k">for</span> <span class="n">crop</span> <span class="ow">in</span> <span class="n">crops</span><span class="p">]))</span> <span class="c1"># returns a 4D tensor</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">])</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1">#In your test loop you can do the following:</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="nb">input</span><span class="p">,</span> <span class="n">target</span> <span class="o">=</span> <span class="n">batch</span> <span class="c1"># input is a 5d tensor, target is 2d</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">bs</span><span class="p">,</span> <span class="n">ncrops</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span> <span class="o">=</span> <span class="nb">input</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">result</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="nb">input</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">c</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span> <span class="c1"># fuse batch size and ncrops</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">result_avg</span> <span class="o">=</span> <span class="n">result</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">bs</span><span class="p">,</span> <span class="n">ncrops</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span> <span class="c1"># avg over crops</span>
+</pre></div>
+</div>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.TenCrop.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#TenCrop.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.TenCrop.forward" title="Permalink to this definition">¶</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>tuple of 10 images. Image can be PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.Pad.html" class="btn btn-neutral float-right" title="Pad" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.FiveCrop.html" class="btn btn-neutral" title="FiveCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">TenCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.ToPILImage.html b/0.16/generated/torchvision.transforms.ToPILImage.html
new file mode 100644
index 00000000000..db9c487adcd
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.ToPILImage.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ToPILImage &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ToTensor" href="torchvision.transforms.ToTensor.html" />
+    <link rel="prev" title="Lambda" href="torchvision.transforms.Lambda.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ToPILImage</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.ToPILImage.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="topilimage">
+<h1>ToPILImage<a class="headerlink" href="#topilimage" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.ToPILImage">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">ToPILImage</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ToPILImage"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ToPILImage" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a tensor or an ndarray to PIL Image</p>
+<p>This transform does not support torchscript.</p>
+<p>Converts a torch.*Tensor of shape C x H x W or a numpy ndarray of shape
+H x W x C to a PIL Image while adjusting the value range depending on the <code class="docutils literal notranslate"><span class="pre">mode</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>mode</strong> (<a class="reference external" href="https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes">PIL.Image mode</a>) – <p>color space and pixel depth of input data (optional).
+If <code class="docutils literal notranslate"><span class="pre">mode</span></code> is <code class="docutils literal notranslate"><span class="pre">None</span></code> (default) there are some assumptions made about the input data:</p>
+<ul class="simple">
+<li><p>If the input has 4 channels, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is assumed to be <code class="docutils literal notranslate"><span class="pre">RGBA</span></code>.</p></li>
+<li><p>If the input has 3 channels, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is assumed to be <code class="docutils literal notranslate"><span class="pre">RGB</span></code>.</p></li>
+<li><p>If the input has 2 channels, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is assumed to be <code class="docutils literal notranslate"><span class="pre">LA</span></code>.</p></li>
+<li><p>If the input has 1 channel, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is determined by the data type (i.e <code class="docutils literal notranslate"><span class="pre">int</span></code>, <code class="docutils literal notranslate"><span class="pre">float</span></code>, <code class="docutils literal notranslate"><span class="pre">short</span></code>).</p></li>
+</ul>
+</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.ToTensor.html" class="btn btn-neutral float-right" title="ToTensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.Lambda.html" class="btn btn-neutral" title="Lambda" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ToPILImage</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.ToTensor.html b/0.16/generated/torchvision.transforms.ToTensor.html
new file mode 100644
index 00000000000..43c4007a488
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.ToTensor.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ToTensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="PILToTensor" href="torchvision.transforms.PILToTensor.html" />
+    <link rel="prev" title="ToPILImage" href="torchvision.transforms.ToPILImage.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ToTensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.ToTensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="totensor">
+<h1>ToTensor<a class="headerlink" href="#totensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.ToTensor">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">ToTensor</span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ToTensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.ToTensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a PIL Image or ndarray to tensor and scale the values accordingly.</p>
+<p>This transform does not support torchscript.</p>
+<p>Converts a PIL Image or numpy.ndarray (H x W x C) in the range
+[0, 255] to a torch.FloatTensor of shape (C x H x W) in the range [0.0, 1.0]
+if the PIL Image belongs to one of the modes (L, LA, P, I, F, RGB, YCbCr, RGBA, CMYK, 1)
+or if the numpy.ndarray has dtype = np.uint8</p>
+<p>In the other cases, tensors are returned without scaling.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Because the input image is scaled to [0.0, 1.0], this transformation should not be used when
+transforming target image masks. See the <a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation">references</a> for implementing the transforms for image masks.</p>
+</div>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.PILToTensor.html" class="btn btn-neutral float-right" title="PILToTensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.ToPILImage.html" class="btn btn-neutral" title="ToPILImage" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ToTensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.TrivialAugmentWide.html b/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
new file mode 100644
index 00000000000..373249d0eb5
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
@@ -0,0 +1,730 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>TrivialAugmentWide &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="AugMix" href="torchvision.transforms.AugMix.html" />
+    <link rel="prev" title="RandAugment" href="torchvision.transforms.RandAugment.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>TrivialAugmentWide</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.TrivialAugmentWide.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="trivialaugmentwide">
+<h1>TrivialAugmentWide<a class="headerlink" href="#trivialaugmentwide" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.TrivialAugmentWide">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.</span></span><span class="sig-name descname"><span class="pre">TrivialAugmentWide</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_magnitude_bins</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">31</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#TrivialAugmentWide"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.TrivialAugmentWide" title="Permalink to this definition">¶</a></dt>
+<dd><p>Dataset-independent data-augmentation with TrivialAugment Wide, as described in
+<a class="reference external" href="https://arxiv.org/abs/2103.10158">“TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation”</a>.
+If the image is torch Tensor, it should be of type torch.uint8, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>num_magnitude_bins</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The number of different magnitude values.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">TrivialAugmentWide</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.TrivialAugmentWide.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#TrivialAugmentWide.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.TrivialAugmentWide.forward" title="Permalink to this definition">¶</a></dt>
+<dd><blockquote>
+<div><p>img (PIL Image or Tensor): Image to be transformed.</p>
+</div></blockquote>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Transformed image.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.AugMix.html" class="btn btn-neutral float-right" title="AugMix" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.RandAugment.html" class="btn btn-neutral" title="RandAugment" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">TrivialAugmentWide</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_brightness.html b/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
new file mode 100644
index 00000000000..c28975ab15e
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_brightness &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_contrast" href="torchvision.transforms.functional.adjust_contrast.html" />
+    <link rel="prev" title="AugMix" href="torchvision.transforms.AugMix.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_brightness</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.adjust_brightness.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-brightness">
+<h1>adjust_brightness<a class="headerlink" href="#adjust-brightness" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.adjust_brightness">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_brightness</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">brightness_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#adjust_brightness"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.adjust_brightness" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust brightness of an image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be adjusted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.</p></li>
+<li><p><strong>brightness_factor</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – How much to adjust the brightness. Can be
+any non-negative number. 0 gives a black image, 1 gives the
+original image while 2 increases the brightness by a factor of 2.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Brightness adjusted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.adjust_contrast.html" class="btn btn-neutral float-right" title="adjust_contrast" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.AugMix.html" class="btn btn-neutral" title="AugMix" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_brightness</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_contrast.html b/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
new file mode 100644
index 00000000000..6a9eb7c86a4
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_contrast &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_gamma" href="torchvision.transforms.functional.adjust_gamma.html" />
+    <link rel="prev" title="adjust_brightness" href="torchvision.transforms.functional.adjust_brightness.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_contrast</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.adjust_contrast.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-contrast">
+<h1>adjust_contrast<a class="headerlink" href="#adjust-contrast" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.adjust_contrast">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_contrast</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">contrast_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#adjust_contrast"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.adjust_contrast" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust contrast of an image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be adjusted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.</p></li>
+<li><p><strong>contrast_factor</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – How much to adjust the contrast. Can be any
+non-negative number. 0 gives a solid gray image, 1 gives the
+original image while 2 increases the contrast by a factor of 2.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Contrast adjusted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.adjust_gamma.html" class="btn btn-neutral float-right" title="adjust_gamma" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.adjust_brightness.html" class="btn btn-neutral" title="adjust_brightness" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_contrast</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_gamma.html b/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
new file mode 100644
index 00000000000..192eca1875a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
@@ -0,0 +1,720 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_gamma &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_hue" href="torchvision.transforms.functional.adjust_hue.html" />
+    <link rel="prev" title="adjust_contrast" href="torchvision.transforms.functional.adjust_contrast.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_gamma</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.adjust_gamma.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-gamma">
+<h1>adjust_gamma<a class="headerlink" href="#adjust-gamma" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.adjust_gamma">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_gamma</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">gamma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">gain</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#adjust_gamma"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.adjust_gamma" title="Permalink to this definition">¶</a></dt>
+<dd><p>Perform gamma correction on an image.</p>
+<p>Also known as Power Law Transform. Intensities in RGB mode are adjusted
+based on the following equation:</p>
+<div class="math notranslate nohighlight">
+\[I_{\text{out}} = 255 \times \text{gain} \times \left(\frac{I_{\text{in}}}{255}\right)^{\gamma}\]</div>
+<p>See <a class="reference external" href="https://en.wikipedia.org/wiki/Gamma_correction">Gamma Correction</a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – PIL Image to be adjusted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, modes with transparency (alpha channel) are not supported.</p></li>
+<li><p><strong>gamma</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Non negative real number, same as <span class="math notranslate nohighlight">\(\gamma\)</span> in the equation.
+gamma larger than 1 make the shadows darker,
+while gamma smaller than 1 make dark regions lighter.</p></li>
+<li><p><strong>gain</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – The constant multiplier.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Gamma correction adjusted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.adjust_hue.html" class="btn btn-neutral float-right" title="adjust_hue" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.adjust_contrast.html" class="btn btn-neutral" title="adjust_contrast" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_gamma</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+         <script async="async" src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_hue.html b/0.16/generated/torchvision.transforms.functional.adjust_hue.html
new file mode 100644
index 00000000000..f85476af428
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.adjust_hue.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_hue &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_saturation" href="torchvision.transforms.functional.adjust_saturation.html" />
+    <link rel="prev" title="adjust_gamma" href="torchvision.transforms.functional.adjust_gamma.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_hue</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.adjust_hue.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-hue">
+<h1>adjust_hue<a class="headerlink" href="#adjust-hue" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.adjust_hue">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_hue</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">hue_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#adjust_hue"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.adjust_hue" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust hue of an image.</p>
+<p>The image hue is adjusted by converting the image to HSV and
+cyclically shifting the intensities in the hue channel (H).
+The image is then converted back to original image mode.</p>
+<p><cite>hue_factor</cite> is the amount of shift in H channel and must be in the
+interval <cite>[-0.5, 0.5]</cite>.</p>
+<p>See <a class="reference external" href="https://en.wikipedia.org/wiki/Hue">Hue</a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be adjusted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image mode “1”, “I”, “F” and modes with transparency (alpha channel) are not supported.
+Note: the pixel values of the input image has to be non-negative for conversion to HSV space;
+thus it does not work if you normalize your image to an interval with negative values,
+or use an interpolation that generates negative values before using this function.</p></li>
+<li><p><strong>hue_factor</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – How much to shift the hue channel. Should be in
+[-0.5, 0.5]. 0.5 and -0.5 give complete reversal of hue channel in
+HSV space in positive and negative direction respectively.
+0 means no shift. Therefore, both -0.5 and 0.5 will give an image
+with complementary colors while 0 gives the original image.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Hue adjusted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.adjust_saturation.html" class="btn btn-neutral float-right" title="adjust_saturation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.adjust_gamma.html" class="btn btn-neutral" title="adjust_gamma" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_hue</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_saturation.html b/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
new file mode 100644
index 00000000000..4eb7681d79d
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_saturation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_sharpness" href="torchvision.transforms.functional.adjust_sharpness.html" />
+    <link rel="prev" title="adjust_hue" href="torchvision.transforms.functional.adjust_hue.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_saturation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.adjust_saturation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-saturation">
+<h1>adjust_saturation<a class="headerlink" href="#adjust-saturation" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.adjust_saturation">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_saturation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">saturation_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#adjust_saturation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.adjust_saturation" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust color saturation of an image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be adjusted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.</p></li>
+<li><p><strong>saturation_factor</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – How much to adjust the saturation. 0 will
+give a black and white image, 1 will give the original image while
+2 will enhance the saturation by a factor of 2.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Saturation adjusted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.adjust_sharpness.html" class="btn btn-neutral float-right" title="adjust_sharpness" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.adjust_hue.html" class="btn btn-neutral" title="adjust_hue" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_saturation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html b/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
new file mode 100644
index 00000000000..1077e33364e
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_sharpness &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="affine" href="torchvision.transforms.functional.affine.html" />
+    <link rel="prev" title="adjust_saturation" href="torchvision.transforms.functional.adjust_saturation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_sharpness</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.adjust_sharpness.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-sharpness">
+<h1>adjust_sharpness<a class="headerlink" href="#adjust-sharpness" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.adjust_sharpness">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_sharpness</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">sharpness_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#adjust_sharpness"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.adjust_sharpness" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust the sharpness of an image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be adjusted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.</p></li>
+<li><p><strong>sharpness_factor</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – How much to adjust the sharpness. Can be
+any non-negative number. 0 gives a blurred image, 1 gives the
+original image while 2 increases the sharpness by a factor of 2.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Sharpness adjusted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">adjust_sharpness</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.affine.html" class="btn btn-neutral float-right" title="affine" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.adjust_saturation.html" class="btn btn-neutral" title="adjust_saturation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_sharpness</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.affine.html b/0.16/generated/torchvision.transforms.functional.affine.html
new file mode 100644
index 00000000000..60b11012ac6
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.affine.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>affine &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="autocontrast" href="torchvision.transforms.functional.autocontrast.html" />
+    <link rel="prev" title="adjust_sharpness" href="torchvision.transforms.functional.adjust_sharpness.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>affine</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.affine.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="affine">
+<h1>affine<a class="headerlink" href="#affine" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.affine">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">affine</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">angle</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">translate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">shear</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#affine"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.affine" title="Permalink to this definition">¶</a></dt>
+<dd><p>Apply affine transformation on the image keeping image center invariant.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – image to transform.</p></li>
+<li><p><strong>angle</strong> (<em>number</em>) – rotation angle in degrees between -180 and 180, clockwise direction.</p></li>
+<li><p><strong>translate</strong> (<em>sequence of python:integers</em>) – horizontal and vertical translations (post-rotation translation)</p></li>
+<li><p><strong>scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – overall scale</p></li>
+<li><p><strong>shear</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>sequence</em>) – shear angle value in degrees between -180 to 180, clockwise direction.
+If a sequence is specified, the first value corresponds to a shear parallel to the x-axis, while
+the second value corresponds to a shear parallel to the y-axis.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – <p>Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode single int/float value is not supported, please use a sequence
+of length 1: <code class="docutils literal notranslate"><span class="pre">[value,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>center</strong> (<em>sequence</em><em>, </em><em>optional</em>) – Optional center of rotation. Origin is the upper left corner.
+Default is the center of the image.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Transformed image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">affine</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.autocontrast.html" class="btn btn-neutral float-right" title="autocontrast" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.adjust_sharpness.html" class="btn btn-neutral" title="adjust_sharpness" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">affine</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.autocontrast.html b/0.16/generated/torchvision.transforms.functional.autocontrast.html
new file mode 100644
index 00000000000..3664f1a7244
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.autocontrast.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>autocontrast &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="center_crop" href="torchvision.transforms.functional.center_crop.html" />
+    <link rel="prev" title="affine" href="torchvision.transforms.functional.affine.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>autocontrast</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.autocontrast.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="autocontrast">
+<h1>autocontrast<a class="headerlink" href="#autocontrast" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.autocontrast">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">autocontrast</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#autocontrast"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.autocontrast" title="Permalink to this definition">¶</a></dt>
+<dd><p>Maximize contrast of an image by remapping its
+pixels per channel so that the lowest becomes black and the lightest
+becomes white.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image on which autocontrast is applied.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>An image that was autocontrasted.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">autocontrast</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.center_crop.html" class="btn btn-neutral float-right" title="center_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.affine.html" class="btn btn-neutral" title="affine" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">autocontrast</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.center_crop.html b/0.16/generated/torchvision.transforms.functional.center_crop.html
new file mode 100644
index 00000000000..42b25606f2c
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.center_crop.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>center_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="convert_image_dtype" href="torchvision.transforms.functional.convert_image_dtype.html" />
+    <link rel="prev" title="autocontrast" href="torchvision.transforms.functional.autocontrast.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>center_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.center_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="center-crop">
+<h1>center_crop<a class="headerlink" href="#center-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.center_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">center_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#center_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.center_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crops the given image at the center.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.
+If image size is smaller than output size along any edge, image is padded with 0 and then center cropped.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p></li>
+<li><p><strong>output_size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – (height, width) of the crop box. If int or sequence with single int,
+it is used for both directions.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Cropped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">center_crop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.convert_image_dtype.html" class="btn btn-neutral float-right" title="convert_image_dtype" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.autocontrast.html" class="btn btn-neutral" title="autocontrast" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">center_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html b/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
new file mode 100644
index 00000000000..ac273d202c4
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>convert_image_dtype &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="crop" href="torchvision.transforms.functional.crop.html" />
+    <link rel="prev" title="center_crop" href="torchvision.transforms.functional.center_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>convert_image_dtype</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.convert_image_dtype.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convert-image-dtype">
+<h1>convert_image_dtype<a class="headerlink" href="#convert-image-dtype" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.convert_image_dtype">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">convert_image_dtype</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">torch.float32</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#convert_image_dtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.convert_image_dtype" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a tensor image to the given <code class="docutils literal notranslate"><span class="pre">dtype</span></code> and scale the values accordingly
+This function does not support PIL Image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>image</strong> (<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><em>torch.Tensor</em></a>) – Image to be converted</p></li>
+<li><p><strong>dtype</strong> (<em>torch.dpython:type</em>) – Desired data type of the output</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Converted image</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>When converting from a smaller to a larger integer <code class="docutils literal notranslate"><span class="pre">dtype</span></code> the maximum values are <strong>not</strong> mapped exactly.
+If converted back and forth, this mismatch has no effect.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Raises<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#RuntimeError" title="(in Python v3.12)"><strong>RuntimeError</strong></a> – When trying to cast <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.float32</span></code> to <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int32</span></code> or <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int64</span></code> as
+    well as for trying to cast <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.float64</span></code> to <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int64</span></code>. These conversions might lead to
+    overflow errors since the floating point <code class="docutils literal notranslate"><span class="pre">dtype</span></code> cannot store consecutive integers over the whole range
+    of the integer <code class="docutils literal notranslate"><span class="pre">dtype</span></code>.</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">convert_image_dtype</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.crop.html" class="btn btn-neutral float-right" title="crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.center_crop.html" class="btn btn-neutral" title="center_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">convert_image_dtype</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.crop.html b/0.16/generated/torchvision.transforms.functional.crop.html
new file mode 100644
index 00000000000..f60da230509
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.crop.html
@@ -0,0 +1,718 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="equalize" href="torchvision.transforms.functional.equalize.html" />
+    <link rel="prev" title="convert_image_dtype" href="torchvision.transforms.functional.convert_image_dtype.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="crop">
+<h1>crop<a class="headerlink" href="#crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">top</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">left</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crop the given image at specified location and output size.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.
+If image size is smaller than output size along any edge, image is padded with 0 and then cropped.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped. (0,0) denotes the top left corner of the image.</p></li>
+<li><p><strong>top</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Vertical component of the top left corner of the crop box.</p></li>
+<li><p><strong>left</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Horizontal component of the top left corner of the crop box.</p></li>
+<li><p><strong>height</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Height of the crop box.</p></li>
+<li><p><strong>width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Width of the crop box.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Cropped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">crop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.equalize.html" class="btn btn-neutral float-right" title="equalize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.convert_image_dtype.html" class="btn btn-neutral" title="convert_image_dtype" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.equalize.html b/0.16/generated/torchvision.transforms.functional.equalize.html
new file mode 100644
index 00000000000..0ca3f3979ae
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.equalize.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>equalize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="erase" href="torchvision.transforms.functional.erase.html" />
+    <link rel="prev" title="crop" href="torchvision.transforms.functional.crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>equalize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.equalize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="equalize">
+<h1>equalize<a class="headerlink" href="#equalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.equalize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">equalize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#equalize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.equalize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Equalize the histogram of an image by applying
+a non-linear mapping to the input in order to create a uniform
+distribution of grayscale values in the output.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image on which equalize is applied.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+The tensor dtype must be <code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code> and values are expected to be in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">255]</span></code>.
+If img is PIL Image, it is expected to be in mode “P”, “L” or “RGB”.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>An image that was equalized.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">equalize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.erase.html" class="btn btn-neutral float-right" title="erase" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.crop.html" class="btn btn-neutral" title="crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">equalize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.erase.html b/0.16/generated/torchvision.transforms.functional.erase.html
new file mode 100644
index 00000000000..72b95c39e06
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.erase.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>erase &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="five_crop" href="torchvision.transforms.functional.five_crop.html" />
+    <link rel="prev" title="equalize" href="torchvision.transforms.functional.equalize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>erase</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.erase.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="erase">
+<h1>erase<a class="headerlink" href="#erase" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.erase">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">erase</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">i</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">j</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">h</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">w</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">v</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#erase"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.erase" title="Permalink to this definition">¶</a></dt>
+<dd><p>Erase the input Tensor Image with given value.
+This transform does not support PIL Image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>Tensor Image</em>) – Tensor image of size (C, H, W) to be erased</p></li>
+<li><p><strong>i</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – i in (i,j) i.e coordinates of the upper left corner.</p></li>
+<li><p><strong>j</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – j in (i,j) i.e coordinates of the upper left corner.</p></li>
+<li><p><strong>h</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Height of the erased region.</p></li>
+<li><p><strong>w</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Width of the erased region.</p></li>
+<li><p><strong>v</strong> – Erasing value.</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – For in-place operations. By default, is set False.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Erased image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor Image</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.five_crop.html" class="btn btn-neutral float-right" title="five_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.equalize.html" class="btn btn-neutral" title="equalize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">erase</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.five_crop.html b/0.16/generated/torchvision.transforms.functional.five_crop.html
new file mode 100644
index 00000000000..1e596a638bc
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.five_crop.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>five_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="gaussian_blur" href="torchvision.transforms.functional.gaussian_blur.html" />
+    <link rel="prev" title="erase" href="torchvision.transforms.functional.erase.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>five_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.five_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="five-crop">
+<h1>five_crop<a class="headerlink" href="#five-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.five_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">five_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#five_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.five_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crop the given image into four corners and the central crop.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform returns a tuple of images and there may be a
+mismatch in the number of inputs and targets your <code class="docutils literal notranslate"><span class="pre">Dataset</span></code> returns.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p></li>
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>tuple (tl, tr, bl, br, center)
+Corresponding top left, top right, bottom left, bottom right and center crop.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">five_crop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.gaussian_blur.html" class="btn btn-neutral float-right" title="gaussian_blur" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.erase.html" class="btn btn-neutral" title="erase" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">five_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.gaussian_blur.html b/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
new file mode 100644
index 00000000000..477eac916f2
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
@@ -0,0 +1,732 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>gaussian_blur &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="get_dimensions" href="torchvision.transforms.functional.get_dimensions.html" />
+    <link rel="prev" title="five_crop" href="torchvision.transforms.functional.five_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>gaussian_blur</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.gaussian_blur.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="gaussian-blur">
+<h1>gaussian_blur<a class="headerlink" href="#gaussian-blur" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.gaussian_blur">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">gaussian_blur</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">kernel_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#gaussian_blur"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.gaussian_blur" title="Permalink to this definition">¶</a></dt>
+<dd><p>Performs Gaussian blurring on the image by given kernel.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be blurred</p></li>
+<li><p><strong>kernel_size</strong> (<em>sequence of python:ints</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – <p>Gaussian kernel size. Can be a sequence of integers
+like <code class="docutils literal notranslate"><span class="pre">(kx,</span> <span class="pre">ky)</span></code> or a single integer for square kernels.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode kernel_size as single int is not supported, use a sequence of
+length 1: <code class="docutils literal notranslate"><span class="pre">[ksize,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>sigma</strong> (<em>sequence of python:floats</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – <p>Gaussian kernel standard deviation. Can be a
+sequence of floats like <code class="docutils literal notranslate"><span class="pre">(sigma_x,</span> <span class="pre">sigma_y)</span></code> or a single float to define the
+same sigma in both X/Y directions. If None, then it is computed using
+<code class="docutils literal notranslate"><span class="pre">kernel_size</span></code> as <code class="docutils literal notranslate"><span class="pre">sigma</span> <span class="pre">=</span> <span class="pre">0.3</span> <span class="pre">*</span> <span class="pre">((kernel_size</span> <span class="pre">-</span> <span class="pre">1)</span> <span class="pre">*</span> <span class="pre">0.5</span> <span class="pre">-</span> <span class="pre">1)</span> <span class="pre">+</span> <span class="pre">0.8</span></code>.
+Default, None.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode sigma as single float is
+not supported, use a sequence of length 1: <code class="docutils literal notranslate"><span class="pre">[sigma,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Gaussian Blurred version of the image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">gaussian_blur</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.get_dimensions.html" class="btn btn-neutral float-right" title="get_dimensions" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.five_crop.html" class="btn btn-neutral" title="five_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">gaussian_blur</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.get_dimensions.html b/0.16/generated/torchvision.transforms.functional.get_dimensions.html
new file mode 100644
index 00000000000..1282b5bf1b3
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.get_dimensions.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>get_dimensions &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="get_image_num_channels" href="torchvision.transforms.functional.get_image_num_channels.html" />
+    <link rel="prev" title="gaussian_blur" href="torchvision.transforms.functional.gaussian_blur.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>get_dimensions</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.get_dimensions.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="get-dimensions">
+<h1>get_dimensions<a class="headerlink" href="#get-dimensions" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.get_dimensions">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">get_dimensions</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#get_dimensions"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.get_dimensions" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns the dimensions of an image as [channels, height, width].</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – The image to be checked.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The image dimensions.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>List[<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)">int</a>]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.get_image_num_channels.html" class="btn btn-neutral float-right" title="get_image_num_channels" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.gaussian_blur.html" class="btn btn-neutral" title="gaussian_blur" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">get_dimensions</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html b/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
new file mode 100644
index 00000000000..0ea57f22264
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>get_image_num_channels &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="get_image_size" href="torchvision.transforms.functional.get_image_size.html" />
+    <link rel="prev" title="get_dimensions" href="torchvision.transforms.functional.get_dimensions.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>get_image_num_channels</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.get_image_num_channels.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="get-image-num-channels">
+<h1>get_image_num_channels<a class="headerlink" href="#get-image-num-channels" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.get_image_num_channels">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">get_image_num_channels</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#get_image_num_channels"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.get_image_num_channels" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns the number of channels of an image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – The image to be checked.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The number of channels.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)">int</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.get_image_size.html" class="btn btn-neutral float-right" title="get_image_size" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.get_dimensions.html" class="btn btn-neutral" title="get_dimensions" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">get_image_num_channels</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.get_image_size.html b/0.16/generated/torchvision.transforms.functional.get_image_size.html
new file mode 100644
index 00000000000..238f78b825c
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.get_image_size.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>get_image_size &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="hflip" href="torchvision.transforms.functional.hflip.html" />
+    <link rel="prev" title="get_image_num_channels" href="torchvision.transforms.functional.get_image_num_channels.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>get_image_size</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.get_image_size.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="get-image-size">
+<h1>get_image_size<a class="headerlink" href="#get-image-size" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.get_image_size">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">get_image_size</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#get_image_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.get_image_size" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns the size of an image as [width, height].</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – The image to be checked.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The image size.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>List[<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)">int</a>]</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.hflip.html" class="btn btn-neutral float-right" title="hflip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.get_image_num_channels.html" class="btn btn-neutral" title="get_image_num_channels" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">get_image_size</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.hflip.html b/0.16/generated/torchvision.transforms.functional.hflip.html
new file mode 100644
index 00000000000..b3b3151dea1
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.hflip.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>hflip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="invert" href="torchvision.transforms.functional.invert.html" />
+    <link rel="prev" title="get_image_size" href="torchvision.transforms.functional.get_image_size.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>hflip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.hflip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="hflip">
+<h1>hflip<a class="headerlink" href="#hflip" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.hflip">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">hflip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#hflip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.hflip" title="Permalink to this definition">¶</a></dt>
+<dd><p>Horizontally flip the given image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be flipped. If img
+is a Tensor, it is expected to be in […, H, W] format,
+where … means it can have an arbitrary number of leading
+dimensions.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Horizontally flipped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">hflip</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.invert.html" class="btn btn-neutral float-right" title="invert" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.get_image_size.html" class="btn btn-neutral" title="get_image_size" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">hflip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.invert.html b/0.16/generated/torchvision.transforms.functional.invert.html
new file mode 100644
index 00000000000..a06fd54b78e
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.invert.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>invert &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="normalize" href="torchvision.transforms.functional.normalize.html" />
+    <link rel="prev" title="hflip" href="torchvision.transforms.functional.hflip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>invert</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.invert.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="invert">
+<h1>invert<a class="headerlink" href="#invert" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.invert">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">invert</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#invert"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.invert" title="Permalink to this definition">¶</a></dt>
+<dd><p>Invert the colors of an RGB/grayscale image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to have its colors inverted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Color inverted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">invert</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.normalize.html" class="btn btn-neutral float-right" title="normalize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.hflip.html" class="btn btn-neutral" title="hflip" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">invert</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.normalize.html b/0.16/generated/torchvision.transforms.functional.normalize.html
new file mode 100644
index 00000000000..8f3de5bd2c9
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.normalize.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>normalize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="pad" href="torchvision.transforms.functional.pad.html" />
+    <link rel="prev" title="invert" href="torchvision.transforms.functional.invert.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>normalize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.normalize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="normalize">
+<h1>normalize<a class="headerlink" href="#normalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.normalize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">normalize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mean</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#normalize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.normalize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Normalize a float tensor image with mean and standard deviation.
+This transform does not support PIL Image.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform acts out of place by default, i.e., it does not mutates the input tensor.</p>
+</div>
+<p>See <a class="reference internal" href="torchvision.transforms.Normalize.html#torchvision.transforms.Normalize" title="torchvision.transforms.Normalize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Normalize</span></code></a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>tensor</strong> (<em>Tensor</em>) – Float tensor image of size (C, H, W) or (B, C, H, W) to be normalized.</p></li>
+<li><p><strong>mean</strong> (<em>sequence</em>) – Sequence of means for each channel.</p></li>
+<li><p><strong>std</strong> (<em>sequence</em>) – Sequence of standard deviations for each channel.</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>,</em><em>optional</em>) – Bool to make this operation inplace.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Normalized Tensor image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.pad.html" class="btn btn-neutral float-right" title="pad" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.invert.html" class="btn btn-neutral" title="invert" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">normalize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.pad.html b/0.16/generated/torchvision.transforms.functional.pad.html
new file mode 100644
index 00000000000..b194313f5a2
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.pad.html
@@ -0,0 +1,744 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>pad &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="perspective" href="torchvision.transforms.functional.perspective.html" />
+    <link rel="prev" title="normalize" href="torchvision.transforms.functional.normalize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>pad</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.pad.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="pad">
+<h1>pad<a class="headerlink" href="#pad" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.pad">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">pad</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'constant'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#pad"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.pad" title="Permalink to this definition">¶</a></dt>
+<dd><p>Pad the given image on all sides with the given “pad” value.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means at most 2 leading dimensions for mode reflect and symmetric,
+at most 3 leading dimensions for mode edge,
+and an arbitrary number of leading dimensions for mode constant</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be padded.</p></li>
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em>) – <p>Padding on each border. If a single int is provided this
+is used to pad all borders. If sequence of length 2 is provided this is the padding
+on left/right and top/bottom respectively. If a sequence of length 4 is provided
+this is the padding for the left, top, right and bottom borders respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode padding as single int is not supported, use a sequence of
+length 1: <code class="docutils literal notranslate"><span class="pre">[padding,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a>) – Pixel fill value for constant fill. Default is 0.
+If a tuple of length 3, it is used to fill R, G, B channels respectively.
+This value is only used when the padding_mode is constant.
+Only number is supported for torch Tensor.
+Only int or tuple value is supported for PIL Image.</p></li>
+<li><p><strong>padding_mode</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – <p>Type of padding. Should be: constant, edge, reflect or symmetric.
+Default is constant.</p>
+<ul>
+<li><p>constant: pads with a constant value, this value is specified with fill</p></li>
+<li><p>edge: pads with the last value at the edge of the image.
+If input a 5D torch Tensor, the last 3 dimensions will be padded instead of the last 2</p></li>
+<li><p>reflect: pads with reflection of image without repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode
+will result in [3, 2, 1, 2, 3, 4, 3, 2]</p></li>
+<li><p>symmetric: pads with reflection of image repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode
+will result in [2, 1, 1, 2, 3, 4, 4, 3]</p></li>
+</ul>
+</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Padded image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">pad</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.perspective.html" class="btn btn-neutral float-right" title="perspective" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.normalize.html" class="btn btn-neutral" title="normalize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">pad</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.perspective.html b/0.16/generated/torchvision.transforms.functional.perspective.html
new file mode 100644
index 00000000000..56d51646bad
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.perspective.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>perspective &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="pil_to_tensor" href="torchvision.transforms.functional.pil_to_tensor.html" />
+    <link rel="prev" title="pad" href="torchvision.transforms.functional.pad.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>perspective</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.perspective.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="perspective">
+<h1>perspective<a class="headerlink" href="#perspective" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.perspective">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">perspective</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">startpoints</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">endpoints</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#perspective"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.perspective" title="Permalink to this definition">¶</a></dt>
+<dd><p>Perform perspective transform of the given image.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be transformed.</p></li>
+<li><p><strong>startpoints</strong> (<em>list of list of python:ints</em>) – List containing four lists of two integers corresponding to four corners
+<code class="docutils literal notranslate"><span class="pre">[top-left,</span> <span class="pre">top-right,</span> <span class="pre">bottom-right,</span> <span class="pre">bottom-left]</span></code> of the original image.</p></li>
+<li><p><strong>endpoints</strong> (<em>list of list of python:ints</em>) – List containing four lists of two integers corresponding to four corners
+<code class="docutils literal notranslate"><span class="pre">[top-left,</span> <span class="pre">top-right,</span> <span class="pre">bottom-right,</span> <span class="pre">bottom-left]</span></code> of the transformed image.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – <p>Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode single int/float value is not supported, please use a sequence
+of length 1: <code class="docutils literal notranslate"><span class="pre">[value,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>transformed Image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">perspective</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.pil_to_tensor.html" class="btn btn-neutral float-right" title="pil_to_tensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.pad.html" class="btn btn-neutral" title="pad" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">perspective</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html b/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
new file mode 100644
index 00000000000..44594c0faa3
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>pil_to_tensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="posterize" href="torchvision.transforms.functional.posterize.html" />
+    <link rel="prev" title="perspective" href="torchvision.transforms.functional.perspective.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>pil_to_tensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.pil_to_tensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="pil-to-tensor">
+<h1>pil_to_tensor<a class="headerlink" href="#pil-to-tensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.pil_to_tensor">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">pil_to_tensor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pic</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#pil_to_tensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.pil_to_tensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a <code class="docutils literal notranslate"><span class="pre">PIL</span> <span class="pre">Image</span></code> to a tensor of the same type.
+This function does not support torchscript.</p>
+<p>See <a class="reference internal" href="torchvision.transforms.PILToTensor.html#torchvision.transforms.PILToTensor" title="torchvision.transforms.PILToTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">PILToTensor</span></code></a> for more details.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>A deep copy of the underlying array is performed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>pic</strong> (<em>PIL Image</em>) – Image to be converted to tensor.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Converted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.posterize.html" class="btn btn-neutral float-right" title="posterize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.perspective.html" class="btn btn-neutral" title="perspective" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">pil_to_tensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.posterize.html b/0.16/generated/torchvision.transforms.functional.posterize.html
new file mode 100644
index 00000000000..780a01da39d
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.posterize.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>posterize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resize" href="torchvision.transforms.functional.resize.html" />
+    <link rel="prev" title="pil_to_tensor" href="torchvision.transforms.functional.pil_to_tensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>posterize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.posterize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="posterize">
+<h1>posterize<a class="headerlink" href="#posterize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.posterize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">posterize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">bits</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#posterize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.posterize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Posterize an image by reducing the number of bits for each color channel.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to have its colors posterized.
+If img is torch Tensor, it should be of type torch.uint8, and
+it is expected to be in […, 1 or 3, H, W] format, where … means
+it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p></li>
+<li><p><strong>bits</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The number of bits to keep for each channel (0-8).</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Posterized image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">posterize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.resize.html" class="btn btn-neutral float-right" title="resize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.pil_to_tensor.html" class="btn btn-neutral" title="pil_to_tensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">posterize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.resize.html b/0.16/generated/torchvision.transforms.functional.resize.html
new file mode 100644
index 00000000000..a60352465a9
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.resize.html
@@ -0,0 +1,768 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resized_crop" href="torchvision.transforms.functional.resized_crop.html" />
+    <link rel="prev" title="posterize" href="torchvision.transforms.functional.posterize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>resize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.resize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resize">
+<h1>resize<a class="headerlink" href="#resize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.resize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">resize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#resize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.resize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Resize the input image to the given size.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The output image might be different depending on its type: when downsampling, the interpolation of PIL images
+and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences
+in the performance of a network. Therefore, it is preferable to train and serve a model with the same input
+types. See also below the <code class="docutils literal notranslate"><span class="pre">antialias</span></code> parameter, which can help making the output of PIL images and tensors
+closer.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be resized.</p></li>
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – <p>Desired output size. If size is a sequence like
+(h, w), the output size will be matched to this. If size is an int,
+the smaller edge of the image will be matched to this number maintaining
+the aspect ratio. i.e, if height &gt; width, then image will be rescaled to
+<span class="math notranslate nohighlight">\(\left(\text{size} \times \frac{\text{height}}{\text{width}}, \text{size}\right)\)</span>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode size as single int is not supported, use a sequence of length 1: <code class="docutils literal notranslate"><span class="pre">[size,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>.
+Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>. If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are
+supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>max_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The maximum allowed for the longer edge of
+the resized image. If the longer edge of the image is greater
+than <code class="docutils literal notranslate"><span class="pre">max_size</span></code> after being resized according to <code class="docutils literal notranslate"><span class="pre">size</span></code>,
+<code class="docutils literal notranslate"><span class="pre">size</span></code> will be overruled so that the longer edge is equal to
+<code class="docutils literal notranslate"><span class="pre">max_size</span></code>.
+As a result, the smaller edge may be shorter than <code class="docutils literal notranslate"><span class="pre">size</span></code>. This
+is only supported if <code class="docutils literal notranslate"><span class="pre">size</span></code> is an int (or a sequence of length
+1 in torchscript mode).</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Resized image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">resize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.resized_crop.html" class="btn btn-neutral float-right" title="resized_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.posterize.html" class="btn btn-neutral" title="posterize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+         <script async="async" src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.resized_crop.html b/0.16/generated/torchvision.transforms.functional.resized_crop.html
new file mode 100644
index 00000000000..1a016c01a67
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.resized_crop.html
@@ -0,0 +1,744 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resized_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="rgb_to_grayscale" href="torchvision.transforms.functional.rgb_to_grayscale.html" />
+    <link rel="prev" title="resize" href="torchvision.transforms.functional.resize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>resized_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.resized_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resized-crop">
+<h1>resized_crop<a class="headerlink" href="#resized-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.resized_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">resized_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">top</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">left</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#resized_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.resized_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Crop the given image and resize it to desired size.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<p>Notably used in <a class="reference internal" href="torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop" title="torchvision.transforms.RandomResizedCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code></a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped. (0,0) denotes the top left corner of the image.</p></li>
+<li><p><strong>top</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Vertical component of the top left corner of the crop box.</p></li>
+<li><p><strong>left</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Horizontal component of the top left corner of the crop box.</p></li>
+<li><p><strong>height</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Height of the crop box.</p></li>
+<li><p><strong>width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Width of the crop box.</p></li>
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size. Same semantics as <code class="docutils literal notranslate"><span class="pre">resize</span></code>.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>.
+Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>. If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are
+supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Cropped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">resized_crop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.rgb_to_grayscale.html" class="btn btn-neutral float-right" title="rgb_to_grayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.resize.html" class="btn btn-neutral" title="resize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resized_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html b/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
new file mode 100644
index 00000000000..6574ce0b24f
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
@@ -0,0 +1,720 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>rgb_to_grayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="rotate" href="torchvision.transforms.functional.rotate.html" />
+    <link rel="prev" title="resized_crop" href="torchvision.transforms.functional.resized_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>rgb_to_grayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.rgb_to_grayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="rgb-to-grayscale">
+<h1>rgb_to_grayscale<a class="headerlink" href="#rgb-to-grayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.rgb_to_grayscale">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">rgb_to_grayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_output_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#rgb_to_grayscale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.rgb_to_grayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert RGB image to grayscale version of image.
+If the image is torch Tensor, it is expected
+to have […, 3, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Please, note that this method supports only RGB images as input. For inputs in other color spaces,
+please, consider using meth:<cite>~torchvision.transforms.functional.to_grayscale</cite> with PIL Image.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – RGB Image to be converted to grayscale.</p></li>
+<li><p><strong>num_output_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of channels of the output image. Value can be 1 or 3. Default, 1.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><p>Grayscale version of the image.</p>
+<ul class="simple">
+<li><p>if num_output_channels = 1 : returned image is single channel</p></li>
+<li><p>if num_output_channels = 3 : returned image is 3 channel with r = g = b</p></li>
+</ul>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.rotate.html" class="btn btn-neutral float-right" title="rotate" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.resized_crop.html" class="btn btn-neutral" title="resized_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">rgb_to_grayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.rotate.html b/0.16/generated/torchvision.transforms.functional.rotate.html
new file mode 100644
index 00000000000..5129213add5
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.rotate.html
@@ -0,0 +1,732 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>rotate &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="solarize" href="torchvision.transforms.functional.solarize.html" />
+    <link rel="prev" title="rgb_to_grayscale" href="torchvision.transforms.functional.rgb_to_grayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>rotate</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.rotate.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="rotate">
+<h1>rotate<a class="headerlink" href="#rotate" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.rotate">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">rotate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">angle</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">InterpolationMode</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">expand</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#rotate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.rotate" title="Permalink to this definition">¶</a></dt>
+<dd><p>Rotate the image by angle.
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – image to be rotated.</p></li>
+<li><p><strong>angle</strong> (<em>number</em>) – rotation angle value in degrees, counter-clockwise.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>expand</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Optional expansion flag.
+If true, expands the output image to make it large enough to hold the entire rotated image.
+If false or omitted, make the output image the same size as the input image.
+Note that the expand flag assumes rotation around the center and no translation.</p></li>
+<li><p><strong>center</strong> (<em>sequence</em><em>, </em><em>optional</em>) – Optional center of rotation. Origin is the upper left corner.
+Default is the center of the image.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – <p>Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode single int/float value is not supported, please use a sequence
+of length 1: <code class="docutils literal notranslate"><span class="pre">[value,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Rotated image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">rotate</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.solarize.html" class="btn btn-neutral float-right" title="solarize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.rgb_to_grayscale.html" class="btn btn-neutral" title="rgb_to_grayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">rotate</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.solarize.html b/0.16/generated/torchvision.transforms.functional.solarize.html
new file mode 100644
index 00000000000..1edebfb2672
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.solarize.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>solarize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ten_crop" href="torchvision.transforms.functional.ten_crop.html" />
+    <link rel="prev" title="rotate" href="torchvision.transforms.functional.rotate.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>solarize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.solarize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="solarize">
+<h1>solarize<a class="headerlink" href="#solarize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.solarize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">solarize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">threshold</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#solarize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.solarize" title="Permalink to this definition">¶</a></dt>
+<dd><p>Solarize an RGB/grayscale image by inverting all pixel values above a threshold.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to have its colors inverted.
+If img is torch Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p></li>
+<li><p><strong>threshold</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – All pixels equal or above this value are inverted.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Solarized image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">solarize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.ten_crop.html" class="btn btn-neutral float-right" title="ten_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.rotate.html" class="btn btn-neutral" title="rotate" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">solarize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.ten_crop.html b/0.16/generated/torchvision.transforms.functional.ten_crop.html
new file mode 100644
index 00000000000..1ecfe453f73
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.ten_crop.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ten_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="to_grayscale" href="torchvision.transforms.functional.to_grayscale.html" />
+    <link rel="prev" title="solarize" href="torchvision.transforms.functional.solarize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ten_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.ten_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ten-crop">
+<h1>ten_crop<a class="headerlink" href="#ten-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.ten_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">ten_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">vertical_flip</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#ten_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.ten_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>Generate ten cropped images from the given image.
+Crop the given image into four corners and the central crop plus the
+flipped version of these (horizontal flipping is used by default).
+If the image is torch Tensor, it is expected
+to have […, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform returns a tuple of images and there may be a
+mismatch in the number of inputs and targets your <code class="docutils literal notranslate"><span class="pre">Dataset</span></code> returns.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p></li>
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p></li>
+<li><p><strong>vertical_flip</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – Use vertical flipping instead of horizontal</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>tuple (tl, tr, bl, br, center, tl_flip, tr_flip, bl_flip, br_flip, center_flip)
+Corresponding top left, top right, bottom left, bottom right and
+center crop and same for the flipped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.to_grayscale.html" class="btn btn-neutral float-right" title="to_grayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.solarize.html" class="btn btn-neutral" title="solarize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ten_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.to_grayscale.html b/0.16/generated/torchvision.transforms.functional.to_grayscale.html
new file mode 100644
index 00000000000..0830b755b47
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.to_grayscale.html
@@ -0,0 +1,718 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>to_grayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="to_pil_image" href="torchvision.transforms.functional.to_pil_image.html" />
+    <link rel="prev" title="ten_crop" href="torchvision.transforms.functional.ten_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>to_grayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.to_grayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="to-grayscale">
+<h1>to_grayscale<a class="headerlink" href="#to-grayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.to_grayscale">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">to_grayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_output_channels</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#to_grayscale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.to_grayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert PIL image of any mode (RGB, HSV, LAB, etc) to grayscale version of image.
+This transform does not support torch Tensor.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em>) – PIL Image to be converted to grayscale.</p></li>
+<li><p><strong>num_output_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of channels of the output image. Value can be 1 or 3. Default is 1.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><p>Grayscale version of the image.</p>
+<ul class="simple">
+<li><p>if num_output_channels = 1 : returned image is single channel</p></li>
+<li><p>if num_output_channels = 3 : returned image is 3 channel with r = g = b</p></li>
+</ul>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">to_grayscale</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.to_pil_image.html" class="btn btn-neutral float-right" title="to_pil_image" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.ten_crop.html" class="btn btn-neutral" title="ten_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">to_grayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.to_pil_image.html b/0.16/generated/torchvision.transforms.functional.to_pil_image.html
new file mode 100644
index 00000000000..2d896b3aa00
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.to_pil_image.html
@@ -0,0 +1,721 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>to_pil_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="to_tensor" href="torchvision.transforms.functional.to_tensor.html" />
+    <link rel="prev" title="to_grayscale" href="torchvision.transforms.functional.to_grayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>to_pil_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.to_pil_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="to-pil-image">
+<h1>to_pil_image<a class="headerlink" href="#to-pil-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.to_pil_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">to_pil_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pic</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#to_pil_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.to_pil_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a tensor or an ndarray to PIL Image. This function does not support torchscript.</p>
+<p>See <a class="reference internal" href="torchvision.transforms.ToPILImage.html#torchvision.transforms.ToPILImage" title="torchvision.transforms.ToPILImage"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToPILImage</span></code></a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>pic</strong> (<em>Tensor</em><em> or </em><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><em>numpy.ndarray</em></a>) – Image to be converted to PIL Image.</p></li>
+<li><p><strong>mode</strong> (<a class="reference external" href="https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes">PIL.Image mode</a>) – color space and pixel depth of input data (optional).</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Image converted to PIL Image.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>PIL Image</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">to_pil_image</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.to_tensor.html" class="btn btn-neutral float-right" title="to_tensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.to_grayscale.html" class="btn btn-neutral" title="to_grayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">to_pil_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.to_tensor.html b/0.16/generated/torchvision.transforms.functional.to_tensor.html
new file mode 100644
index 00000000000..dac36254aed
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.to_tensor.html
@@ -0,0 +1,707 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>to_tensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="vflip" href="torchvision.transforms.functional.vflip.html" />
+    <link rel="prev" title="to_pil_image" href="torchvision.transforms.functional.to_pil_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>to_tensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.to_tensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="to-tensor">
+<h1>to_tensor<a class="headerlink" href="#to-tensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.to_tensor">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">to_tensor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pic</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#to_tensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.to_tensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a <code class="docutils literal notranslate"><span class="pre">PIL</span> <span class="pre">Image</span></code> or <code class="docutils literal notranslate"><span class="pre">numpy.ndarray</span></code> to tensor.
+This function does not support torchscript.</p>
+<p>See <a class="reference internal" href="torchvision.transforms.ToTensor.html#torchvision.transforms.ToTensor" title="torchvision.transforms.ToTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToTensor</span></code></a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>pic</strong> (<em>PIL Image</em><em> or </em><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><em>numpy.ndarray</em></a>) – Image to be converted to tensor.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Converted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.functional.vflip.html" class="btn btn-neutral float-right" title="vflip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.to_pil_image.html" class="btn btn-neutral" title="to_pil_image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">to_tensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.functional.vflip.html b/0.16/generated/torchvision.transforms.functional.vflip.html
new file mode 100644
index 00000000000..7a89fed2977
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.functional.vflip.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vflip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="TVTensors" href="../tv_tensors.html" />
+    <link rel="prev" title="to_tensor" href="torchvision.transforms.functional.to_tensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>vflip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.functional.vflip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vflip">
+<h1>vflip<a class="headerlink" href="#vflip" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.functional.vflip">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.functional.</span></span><span class="sig-name descname"><span class="pre">vflip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#vflip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.functional.vflip" title="Permalink to this definition">¶</a></dt>
+<dd><p>Vertically flip the given image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be flipped. If img
+is a Tensor, it is expected to be in […, H, W] format,
+where … means it can have an arbitrary number of leading
+dimensions.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Vertically flipped image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>PIL Image or Tensor</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">vflip</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../tv_tensors.html" class="btn btn-neutral float-right" title="TVTensors" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.functional.to_tensor.html" class="btn btn-neutral" title="to_tensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vflip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.AugMix.html b/0.16/generated/torchvision.transforms.v2.AugMix.html
new file mode 100644
index 00000000000..a0125765ae9
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.AugMix.html
@@ -0,0 +1,738 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>AugMix &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CutMix" href="torchvision.transforms.v2.CutMix.html" />
+    <link rel="prev" title="TrivialAugmentWide" href="torchvision.transforms.v2.TrivialAugmentWide.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>AugMix</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.AugMix.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="augmix">
+<h1>AugMix<a class="headerlink" href="#augmix" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.AugMix">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">AugMix</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">severity</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">3</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mixture_width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">3</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">chain_depth</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">-</span> <span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">all_ops</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#AugMix"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.AugMix" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] AugMix data augmentation method based on
+<a class="reference external" href="https://arxiv.org/abs/1912.02781">“AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty”</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The AugMix transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transformation works on images and videos only.</p>
+<p>If the input is <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it should be of type <code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code>, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>severity</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The severity of base augmentation operators. Default is <code class="docutils literal notranslate"><span class="pre">3</span></code>.</p></li>
+<li><p><strong>mixture_width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The number of augmentation chains. Default is <code class="docutils literal notranslate"><span class="pre">3</span></code>.</p></li>
+<li><p><strong>chain_depth</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The depth of augmentation chains. A negative value denotes stochastic depth sampled from the interval [1, 3].
+Default is <code class="docutils literal notranslate"><span class="pre">-1</span></code>.</p></li>
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – The hyperparameter for the probability distributions. Default is <code class="docutils literal notranslate"><span class="pre">1.0</span></code>.</p></li>
+<li><p><strong>all_ops</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Use all operations (including brightness, contrast, color and sharpness). Default is <code class="docutils literal notranslate"><span class="pre">True</span></code>.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">AugMix</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.AugMix.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#AugMix.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.AugMix.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.CutMix.html" class="btn btn-neutral float-right" title="CutMix" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.TrivialAugmentWide.html" class="btn btn-neutral" title="TrivialAugmentWide" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">AugMix</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.AutoAugment.html b/0.16/generated/torchvision.transforms.v2.AutoAugment.html
new file mode 100644
index 00000000000..11a38ca11fc
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.AutoAugment.html
@@ -0,0 +1,745 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>AutoAugment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandAugment" href="torchvision.transforms.v2.RandAugment.html" />
+    <link rel="prev" title="convert_image_dtype" href="torchvision.transforms.v2.functional.convert_image_dtype.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>AutoAugment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.AutoAugment.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="autoaugment">
+<h1>AutoAugment<a class="headerlink" href="#autoaugment" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.AutoAugment">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">AutoAugment</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">policy</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.autoaugment.AutoAugmentPolicy"><span class="pre">AutoAugmentPolicy</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">AutoAugmentPolicy.IMAGENET</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#AutoAugment"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.AutoAugment" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] AutoAugment data augmentation method based on
+<a class="reference external" href="https://arxiv.org/pdf/1805.09501.pdf">“AutoAugment: Learning Augmentation Strategies from Data”</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The AutoAugment transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transformation works on images and videos only.</p>
+<p>If the input is <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it should be of type <code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code>, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>policy</strong> (<a class="reference internal" href="torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.AutoAugmentPolicy"><em>AutoAugmentPolicy</em></a><em>, </em><em>optional</em>) – Desired policy enum defined by
+<a class="reference internal" href="torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.autoaugment.AutoAugmentPolicy"><code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.autoaugment.AutoAugmentPolicy</span></code></a>. Default is <code class="docutils literal notranslate"><span class="pre">AutoAugmentPolicy.IMAGENET</span></code>.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">AutoAugment</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.AutoAugment.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#AutoAugment.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.AutoAugment.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.AutoAugment.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transform_num</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/autoaugment.html#AutoAugment.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.AutoAugment.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for autoaugment transformation</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>params required by the autoaugment transformation</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandAugment.html" class="btn btn-neutral float-right" title="RandAugment" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.convert_image_dtype.html" class="btn btn-neutral" title="convert_image_dtype" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">AutoAugment</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.CenterCrop.html b/0.16/generated/torchvision.transforms.v2.CenterCrop.html
new file mode 100644
index 00000000000..1865db23374
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.CenterCrop.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CenterCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="FiveCrop" href="torchvision.transforms.v2.FiveCrop.html" />
+    <link rel="prev" title="RandomIoUCrop" href="torchvision.transforms.v2.RandomIoUCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>CenterCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.CenterCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="centercrop">
+<h1>CenterCrop<a class="headerlink" href="#centercrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.CenterCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">CenterCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#CenterCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.CenterCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Crop the input at the center.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The CenterCrop transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<p>If image size is smaller than output size along any edge, image is padded with 0 and then center cropped.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">CenterCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.FiveCrop.html" class="btn btn-neutral float-right" title="FiveCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomIoUCrop.html" class="btn btn-neutral" title="RandomIoUCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CenterCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html b/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
new file mode 100644
index 00000000000..08ad02a21ba
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
@@ -0,0 +1,699 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ClampBoundingBoxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="UniformTemporalSubsample" href="torchvision.transforms.v2.UniformTemporalSubsample.html" />
+    <link rel="prev" title="SanitizeBoundingBoxes" href="torchvision.transforms.v2.SanitizeBoundingBoxes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ClampBoundingBoxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ClampBoundingBoxes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="clampboundingboxes">
+<h1>ClampBoundingBoxes<a class="headerlink" href="#clampboundingboxes" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ClampBoundingBoxes">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ClampBoundingBoxes</span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_meta.html#ClampBoundingBoxes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ClampBoundingBoxes" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Clamp bounding boxes to their corresponding image dimensions.</p>
+<p>The clamping is done according to the bounding boxes’ <code class="docutils literal notranslate"><span class="pre">canvas_size</span></code> meta-data.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ClampBoundingBoxes transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.UniformTemporalSubsample.html" class="btn btn-neutral float-right" title="UniformTemporalSubsample" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.SanitizeBoundingBoxes.html" class="btn btn-neutral" title="SanitizeBoundingBoxes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ClampBoundingBoxes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ColorJitter.html b/0.16/generated/torchvision.transforms.v2.ColorJitter.html
new file mode 100644
index 00000000000..b8fecd96f9f
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ColorJitter.html
@@ -0,0 +1,753 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ColorJitter &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomChannelPermutation" href="torchvision.transforms.v2.RandomChannelPermutation.html" />
+    <link rel="prev" title="elastic" href="torchvision.transforms.v2.functional.elastic.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ColorJitter</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ColorJitter.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="colorjitter">
+<h1>ColorJitter<a class="headerlink" href="#colorjitter" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ColorJitter">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ColorJitter</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">brightness</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">contrast</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">saturation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hue</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#ColorJitter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ColorJitter" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Randomly change the brightness, contrast, saturation and hue of an image or video.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ColorJitter transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, mode “1”, “I”, “F” and modes with transparency (alpha channel) are not supported.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>brightness</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter brightness.
+brightness_factor is chosen uniformly from [max(0, 1 - brightness), 1 + brightness]
+or the given [min, max]. Should be non negative numbers.</p></li>
+<li><p><strong>contrast</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter contrast.
+contrast_factor is chosen uniformly from [max(0, 1 - contrast), 1 + contrast]
+or the given [min, max]. Should be non-negative numbers.</p></li>
+<li><p><strong>saturation</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter saturation.
+saturation_factor is chosen uniformly from [max(0, 1 - saturation), 1 + saturation]
+or the given [min, max]. Should be non negative numbers.</p></li>
+<li><p><strong>hue</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – How much to jitter hue.
+hue_factor is chosen uniformly from [-hue, hue] or the given [min, max].
+Should have 0&lt;= hue &lt;= 0.5 or -0.5 &lt;= min &lt;= max &lt;= 0.5.
+To jitter hue, the pixel values of the input image has to be non-negative for conversion to HSV space;
+thus it does not work if you normalize your image to an interval with negative values,
+or use an interpolation that generates negative values before using this function.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">ColorJitter</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ColorJitter.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">brightness</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">contrast</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">saturation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hue</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#ColorJitter.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ColorJitter.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get the parameters for the randomized transform to be applied on image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>brightness</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the brightness_factor is chosen
+uniformly. Pass None to turn off the transformation.</p></li>
+<li><p><strong>contrast</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the contrast_factor is chosen
+uniformly. Pass None to turn off the transformation.</p></li>
+<li><p><strong>saturation</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the saturation_factor is chosen
+uniformly. Pass None to turn off the transformation.</p></li>
+<li><p><strong>hue</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – The range from which the hue_factor is chosen uniformly.
+Pass None to turn off the transformation.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>The parameters used to apply the randomized transform
+along with their random order.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomChannelPermutation.html" class="btn btn-neutral float-right" title="RandomChannelPermutation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.elastic.html" class="btn btn-neutral" title="elastic" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ColorJitter</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.Compose.html b/0.16/generated/torchvision.transforms.v2.Compose.html
new file mode 100644
index 00000000000..7b5af6e35da
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.Compose.html
@@ -0,0 +1,765 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Compose &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomApply" href="torchvision.transforms.v2.RandomApply.html" />
+    <link rel="prev" title="adjust_gamma" href="torchvision.transforms.v2.functional.adjust_gamma.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Compose</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.Compose.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="compose">
+<h1>Compose<a class="headerlink" href="#compose" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Compose">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">Compose</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#Compose"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Compose" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Composes several transforms together.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The Compose transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support torchscript.
+Please, see the note below.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>transforms</strong> (list of <code class="docutils literal notranslate"><span class="pre">Transform</span></code> objects) – list of transforms to compose.</p>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transforms</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">CenterCrop</span><span class="p">(</span><span class="mi">10</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">PILToTensor</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">ConvertImageDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">])</span>
+</pre></div>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In order to script the transformations, please use <code class="docutils literal notranslate"><span class="pre">torch.nn.Sequential</span></code> as below.</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">CenterCrop</span><span class="p">(</span><span class="mi">10</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">Normalize</span><span class="p">((</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">),</span> <span class="p">(</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">)),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">scripted_transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">script</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Make sure to use only scriptable transformations, i.e. that work with <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code>, does not require
+<cite>lambda</cite> functions or <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>.</p>
+</div>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Compose</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images."><img alt="" src="../_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Compose.extra_repr">
+<span class="sig-name descname"><span class="pre">extra_repr</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#Compose.extra_repr"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Compose.extra_repr" title="Permalink to this definition">¶</a></dt>
+<dd><p>Set the extra representation of the module</p>
+<p>To print customized extra information, you should re-implement
+this method in your own modules. Both single-line and multi-line
+strings are acceptable.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Compose.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#Compose.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Compose.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomApply.html" class="btn btn-neutral float-right" title="RandomApply" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.adjust_gamma.html" class="btn btn-neutral" title="adjust_gamma" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Compose</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html b/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
new file mode 100644
index 00000000000..47071cbb7f9
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
@@ -0,0 +1,705 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ConvertBoundingBoxFormat &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="torchvision.transforms.v2.functional.to_image" href="torchvision.transforms.v2.functional.to_image.html" />
+    <link rel="prev" title="ToDtype" href="torchvision.transforms.v2.ToDtype.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ConvertBoundingBoxFormat</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convertboundingboxformat">
+<h1>ConvertBoundingBoxFormat<a class="headerlink" href="#convertboundingboxformat" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ConvertBoundingBoxFormat">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ConvertBoundingBoxFormat</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors._bounding_boxes.BoundingBoxFormat"><span class="pre">BoundingBoxFormat</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_meta.html#ConvertBoundingBoxFormat"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ConvertBoundingBoxFormat" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Convert bounding box coordinates to the given <code class="docutils literal notranslate"><span class="pre">format</span></code>, eg from “CXCYWH” to “XYXY”.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ConvertBoundingBoxFormat transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>format</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em> or </em><a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors.BoundingBoxFormat"><em>tv_tensors.BoundingBoxFormat</em></a>) – output bounding box format.
+Possible values are defined by <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors.BoundingBoxFormat"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxFormat</span></code></a> and
+string values match the enums, e.g. “XYXY” or “XYWH” etc.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.to_image.html" class="btn btn-neutral float-right" title="torchvision.transforms.v2.functional.to_image" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ToDtype.html" class="btn btn-neutral" title="ToDtype" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ConvertBoundingBoxFormat</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html b/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
new file mode 100644
index 00000000000..814faddc525
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ConvertImageDtype &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="convert_image_dtype" href="torchvision.transforms.v2.functional.convert_image_dtype.html" />
+    <link rel="prev" title="to_tensor" href="torchvision.transforms.v2.functional.to_tensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ConvertImageDtype</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ConvertImageDtype.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convertimagedtype">
+<h1>ConvertImageDtype<a class="headerlink" href="#convertimagedtype" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ConvertImageDtype">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ConvertImageDtype</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">torch.float32</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#ConvertImageDtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ConvertImageDtype" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] [DEPRECATED] Use <code class="docutils literal notranslate"><span class="pre">v2.ToDtype(dtype,</span> <span class="pre">scale=True)</span></code> instead.</p>
+<p>Convert input image to the given <code class="docutils literal notranslate"><span class="pre">dtype</span></code> and scale the values accordingly.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ConvertImageDtype transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>Consider using <code class="docutils literal notranslate"><span class="pre">ToDtype(dtype,</span> <span class="pre">scale=True)</span></code> instead. See <a class="reference internal" href="torchvision.transforms.v2.ToDtype.html#torchvision.transforms.v2.ToDtype" title="torchvision.transforms.v2.ToDtype"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToDtype</span></code></a>.</p>
+</div>
+<p>This function does not support PIL Image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>dtype</strong> (<em>torch.dpython:type</em>) – Desired data type of the output</p>
+</dd>
+</dl>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>When converting from a smaller to a larger integer <code class="docutils literal notranslate"><span class="pre">dtype</span></code> the maximum values are <strong>not</strong> mapped exactly.
+If converted back and forth, this mismatch has no effect.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Raises<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#RuntimeError" title="(in Python v3.12)"><strong>RuntimeError</strong></a> – When trying to cast <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.float32</span></code> to <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int32</span></code> or <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int64</span></code> as
+    well as for trying to cast <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.float64</span></code> to <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.int64</span></code>. These conversions might lead to
+    overflow errors since the floating point <code class="docutils literal notranslate"><span class="pre">dtype</span></code> cannot store consecutive integers over the whole range
+    of the integer <code class="docutils literal notranslate"><span class="pre">dtype</span></code>.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.convert_image_dtype.html" class="btn btn-neutral float-right" title="convert_image_dtype" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.to_tensor.html" class="btn btn-neutral" title="to_tensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ConvertImageDtype</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.CutMix.html b/0.16/generated/torchvision.transforms.v2.CutMix.html
new file mode 100644
index 00000000000..a5cd872ad42
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.CutMix.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>CutMix &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="MixUp" href="torchvision.transforms.v2.MixUp.html" />
+    <link rel="prev" title="AugMix" href="torchvision.transforms.v2.AugMix.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>CutMix</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.CutMix.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="cutmix">
+<h1>CutMix<a class="headerlink" href="#cutmix" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.CutMix">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">CutMix</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels_getter</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'default'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_augment.html#CutMix"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.CutMix" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Apply CutMix to the provided batch of images and labels.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The CutMix transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>Paper: <a class="reference external" href="https://arxiv.org/abs/1905.04899">CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform is meant to be used on <strong>batches</strong> of samples, not
+individual images. See
+<a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a> for detailed usage
+examples.
+The sample pairing is deterministic and done by matching consecutive
+samples in the batch, so the batch needs to be shuffled (this is an
+implementation detail, not a guaranteed convention.)</p>
+</div>
+<p>In the input, the labels are expected to be a tensor of shape <code class="docutils literal notranslate"><span class="pre">(batch_size,)</span></code>. They will be transformed
+into a tensor of shape <code class="docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_classes)</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – hyperparameter of the Beta distribution used for mixup. Default is 1.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of classes in the batch. Used for one-hot-encoding.</p></li>
+<li><p><strong>labels_getter</strong> (<em>callable</em><em> or </em><em>&quot;default&quot;</em><em>, </em><em>optional</em>) – indicates how to identify the labels in the input.
+By default, this will pick the second parameter as the labels if it’s a tensor. This covers the most
+common scenario where this transform is called as <code class="docutils literal notranslate"><span class="pre">CutMix()(imgs_batch,</span> <span class="pre">labels_batch)</span></code>.
+It can also be a callable that takes the same input as the transform, and returns the labels.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">CutMix</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.MixUp.html" class="btn btn-neutral float-right" title="MixUp" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.AugMix.html" class="btn btn-neutral" title="AugMix" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">CutMix</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ElasticTransform.html b/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
new file mode 100644
index 00000000000..a1955501e34
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
@@ -0,0 +1,741 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ElasticTransform &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="horizontal_flip" href="torchvision.transforms.v2.functional.horizontal_flip.html" />
+    <link rel="prev" title="RandomPerspective" href="torchvision.transforms.v2.RandomPerspective.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ElasticTransform</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ElasticTransform.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="elastictransform">
+<h1>ElasticTransform<a class="headerlink" href="#elastictransform" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ElasticTransform">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ElasticTransform</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">50.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">5.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#ElasticTransform"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ElasticTransform" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Transform the input with elastic transformations.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomPerspective transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<p>Given alpha and sigma, it will generate displacement
+vectors for all pixels based on random offsets. Alpha controls the strength
+and sigma controls the smoothness of the displacements.
+The displacements are added to an identity grid and the resulting grid is
+used to transform the input.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Implementation to transform bounding boxes is approximative (not exact).
+We construct an approximation of the inverse grid as <code class="docutils literal notranslate"><span class="pre">inverse_grid</span> <span class="pre">=</span> <span class="pre">identity</span> <span class="pre">-</span> <span class="pre">displacement</span></code>.
+This is not an exact inverse of the grid used to transform images, i.e. <code class="docutils literal notranslate"><span class="pre">grid</span> <span class="pre">=</span> <span class="pre">identity</span> <span class="pre">+</span> <span class="pre">displacement</span></code>.
+Our assumption is that <code class="docutils literal notranslate"><span class="pre">displacement</span> <span class="pre">*</span> <span class="pre">displacement</span></code> is small and can be ignored.
+Large displacements would lead to large errors in the approximation.</p>
+</div>
+<dl class="simple">
+<dt>Applications:</dt><dd><p>Randomly transforms the morphology of objects in images and produces a
+see-through-water-like effect.</p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>sequence of python:floats</em><em>, </em><em>optional</em>) – Magnitude of displacements. Default is 50.0.</p></li>
+<li><p><strong>sigma</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>sequence of python:floats</em><em>, </em><em>optional</em>) – Smoothness of displacements. Default is 5.0.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – Pixel fill value used when the  <code class="docutils literal notranslate"><span class="pre">padding_mode</span></code> is constant.
+Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.
+Fill value can be also a dictionary mapping data type to the fill value, e.g.
+<code class="docutils literal notranslate"><span class="pre">fill={tv_tensors.Image:</span> <span class="pre">127,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">0}</span></code> where <code class="docutils literal notranslate"><span class="pre">Image</span></code> will be filled with 127 and
+<code class="docutils literal notranslate"><span class="pre">Mask</span></code> will be filled with 0.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">ElasticTransform</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.horizontal_flip.html" class="btn btn-neutral float-right" title="horizontal_flip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomPerspective.html" class="btn btn-neutral" title="RandomPerspective" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ElasticTransform</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.FiveCrop.html b/0.16/generated/torchvision.transforms.v2.FiveCrop.html
new file mode 100644
index 00000000000..7df16e39c8a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.FiveCrop.html
@@ -0,0 +1,738 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FiveCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="TenCrop" href="torchvision.transforms.v2.TenCrop.html" />
+    <link rel="prev" title="CenterCrop" href="torchvision.transforms.v2.CenterCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>FiveCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.FiveCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fivecrop">
+<h1>FiveCrop<a class="headerlink" href="#fivecrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.FiveCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">FiveCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#FiveCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.FiveCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Crop the image or video into four corners and the central crop.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The FiveCrop transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a> or a
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a> it can have arbitrary number of leading batch dimensions.
+For example, the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform returns a tuple of images and there may be a mismatch in the number of
+inputs and targets your Dataset returns. See below for an example of how to deal with
+this.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an <code class="docutils literal notranslate"><span class="pre">int</span></code>
+instead of sequence like (h, w), a square crop of size (size, size) is made.
+If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="k">class</span> <span class="nc">BatchMultiCrop</span><span class="p">(</span><span class="n">transforms</span><span class="o">.</span><span class="n">Transform</span><span class="p">):</span>
+<span class="gp">... </span>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">sample</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">,</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Video</span><span class="p">],</span> <span class="o">...</span><span class="p">],</span> <span class="nb">int</span><span class="p">]):</span>
+<span class="gp">... </span>        <span class="n">images_or_videos</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="n">sample</span>
+<span class="gp">... </span>        <span class="n">batch_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">images_or_videos</span><span class="p">)</span>
+<span class="gp">... </span>        <span class="n">image_or_video</span> <span class="o">=</span> <span class="n">images_or_videos</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+<span class="gp">... </span>        <span class="n">images_or_videos</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">wrap</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">images_or_videos</span><span class="p">),</span> <span class="n">like</span><span class="o">=</span><span class="n">image_or_video</span><span class="p">)</span>
+<span class="gp">... </span>        <span class="n">labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">batch_size</span><span class="p">,),</span> <span class="n">label</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">images_or_videos</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+<span class="gp">... </span>        <span class="k">return</span> <span class="n">images_or_videos</span><span class="p">,</span> <span class="n">labels</span>
+<span class="gp">...</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">image</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">))</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">label</span> <span class="o">=</span> <span class="mi">3</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">transform</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span><span class="n">transforms</span><span class="o">.</span><span class="n">FiveCrop</span><span class="p">(</span><span class="mi">224</span><span class="p">),</span> <span class="n">BatchMultiCrop</span><span class="p">()])</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">images</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="n">transform</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">label</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">images</span><span class="o">.</span><span class="n">shape</span>
+<span class="go">torch.Size([5, 3, 224, 224])</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">labels</span>
+<span class="go">tensor([3, 3, 3, 3, 3])</span>
+</pre></div>
+</div>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">FiveCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.TenCrop.html" class="btn btn-neutral float-right" title="TenCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.CenterCrop.html" class="btn btn-neutral" title="CenterCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FiveCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.GaussianBlur.html b/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
new file mode 100644
index 00000000000..2ff8e460d54
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
@@ -0,0 +1,735 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>GaussianBlur &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomInvert" href="torchvision.transforms.v2.RandomInvert.html" />
+    <link rel="prev" title="RandomGrayscale" href="torchvision.transforms.v2.RandomGrayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>GaussianBlur</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.GaussianBlur.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="gaussianblur">
+<h1>GaussianBlur<a class="headerlink" href="#gaussianblur" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.GaussianBlur">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">GaussianBlur</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">kernel_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.1,</span> <span class="pre">2.0)</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#GaussianBlur"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.GaussianBlur" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Blurs image with randomly chosen Gaussian blur.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The GausssianBlur transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a Tensor, it is expected
+to have […, C, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>kernel_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em>) – Size of the Gaussian kernel.</p></li>
+<li><p><strong>sigma</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em> or </em><em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em>) – Standard deviation to be used for
+creating kernel to perform blurring. If float, sigma is fixed. If it is tuple
+of float (min, max), sigma is chosen uniformly at random to lie in the
+given range.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">GaussianBlur</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.GaussianBlur.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sigma_min</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma_max</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#GaussianBlur.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.GaussianBlur.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Choose sigma for random gaussian blurring.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>sigma_min</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Minimum standard deviation that can be chosen for blurring kernel.</p></li>
+<li><p><strong>sigma_max</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Maximum standard deviation that can be chosen for blurring kernel.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Standard deviation to be passed to calculate kernel for gaussian blurring.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)">float</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomInvert.html" class="btn btn-neutral float-right" title="RandomInvert" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomGrayscale.html" class="btn btn-neutral" title="RandomGrayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">GaussianBlur</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.Grayscale.html b/0.16/generated/torchvision.transforms.v2.Grayscale.html
new file mode 100644
index 00000000000..7ac1315ba89
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.Grayscale.html
@@ -0,0 +1,709 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Grayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomGrayscale" href="torchvision.transforms.v2.RandomGrayscale.html" />
+    <link rel="prev" title="RandomPhotometricDistort" href="torchvision.transforms.v2.RandomPhotometricDistort.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Grayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.Grayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="grayscale">
+<h1>Grayscale<a class="headerlink" href="#grayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Grayscale">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">Grayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_output_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#Grayscale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Grayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Convert images or videos to grayscale.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The Grayscale transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it is expected
+to have […, 3 or 1, H, W] shape, where … means an arbitrary number of leading dimensions</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>num_output_channels</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – (1 or 3) number of channels desired for output image</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Grayscale</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomGrayscale.html" class="btn btn-neutral float-right" title="RandomGrayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomPhotometricDistort.html" class="btn btn-neutral" title="RandomPhotometricDistort" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Grayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.Lambda.html b/0.16/generated/torchvision.transforms.v2.Lambda.html
new file mode 100644
index 00000000000..dc5014a37a0
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.Lambda.html
@@ -0,0 +1,713 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Lambda &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="SanitizeBoundingBoxes" href="torchvision.transforms.v2.SanitizeBoundingBoxes.html" />
+    <link rel="prev" title="RandomErasing" href="torchvision.transforms.v2.RandomErasing.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Lambda</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.Lambda.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lambda">
+<h1>Lambda<a class="headerlink" href="#lambda" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Lambda">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">Lambda</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">lambd</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">types</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#Lambda"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Lambda" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Apply a user-defined function as a transform.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The Lambda transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support torchscript.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>lambd</strong> (<em>function</em>) – Lambda/function to be used for transform.</p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Lambda.extra_repr">
+<span class="sig-name descname"><span class="pre">extra_repr</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#Lambda.extra_repr"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Lambda.extra_repr" title="Permalink to this definition">¶</a></dt>
+<dd><p>Set the extra representation of the module</p>
+<p>To print customized extra information, you should re-implement
+this method in your own modules. Both single-line and multi-line
+strings are acceptable.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.SanitizeBoundingBoxes.html" class="btn btn-neutral float-right" title="SanitizeBoundingBoxes" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomErasing.html" class="btn btn-neutral" title="RandomErasing" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Lambda</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.LinearTransformation.html b/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
new file mode 100644
index 00000000000..492171e78f0
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
@@ -0,0 +1,717 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>LinearTransformation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Normalize" href="torchvision.transforms.v2.Normalize.html" />
+    <link rel="prev" title="RandomOrder" href="torchvision.transforms.v2.RandomOrder.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>LinearTransformation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.LinearTransformation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lineartransformation">
+<h1>LinearTransformation<a class="headerlink" href="#lineartransformation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.LinearTransformation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">LinearTransformation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transformation_matrix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mean_vector</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#LinearTransformation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.LinearTransformation" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Transform a tensor image or video with a square transformation matrix and a mean_vector computed offline.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The LinearTransformation transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support PIL Image.
+Given transformation_matrix and mean_vector, will flatten the torch.*Tensor and
+subtract mean_vector from it which is then followed by computing the dot
+product with the transformation matrix and then reshaping the tensor to its
+original shape.</p>
+<dl class="simple">
+<dt>Applications:</dt><dd><p>whitening transformation: Suppose X is a column vector zero-centered data.
+Then compute the data covariance matrix [D x D] with torch.mm(X.t(), X),
+perform SVD on this matrix and pass it as transformation_matrix.</p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>transformation_matrix</strong> (<em>Tensor</em>) – tensor [D x D], D = C x H x W</p></li>
+<li><p><strong>mean_vector</strong> (<em>Tensor</em>) – tensor [D], D = C x H x W</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.Normalize.html" class="btn btn-neutral float-right" title="Normalize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomOrder.html" class="btn btn-neutral" title="RandomOrder" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">LinearTransformation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.MixUp.html b/0.16/generated/torchvision.transforms.v2.MixUp.html
new file mode 100644
index 00000000000..4fc305e0fce
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.MixUp.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MixUp &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="register_kernel" href="torchvision.transforms.v2.functional.register_kernel.html" />
+    <link rel="prev" title="CutMix" href="torchvision.transforms.v2.CutMix.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>MixUp</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.MixUp.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mixup">
+<h1>MixUp<a class="headerlink" href="#mixup" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.MixUp">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">MixUp</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels_getter</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'default'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_augment.html#MixUp"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.MixUp" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Apply MixUp to the provided batch of images and labels.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The MixUp transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>Paper: <a class="reference external" href="https://arxiv.org/abs/1710.09412">mixup: Beyond Empirical Risk Minimization</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform is meant to be used on <strong>batches</strong> of samples, not
+individual images. See
+<a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a> for detailed usage
+examples.
+The sample pairing is deterministic and done by matching consecutive
+samples in the batch, so the batch needs to be shuffled (this is an
+implementation detail, not a guaranteed convention.)</p>
+</div>
+<p>In the input, the labels are expected to be a tensor of shape <code class="docutils literal notranslate"><span class="pre">(batch_size,)</span></code>. They will be transformed
+into a tensor of shape <code class="docutils literal notranslate"><span class="pre">(batch_size,</span> <span class="pre">num_classes)</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – hyperparameter of the Beta distribution used for mixup. Default is 1.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of classes in the batch. Used for one-hot-encoding.</p></li>
+<li><p><strong>labels_getter</strong> (<em>callable</em><em> or </em><em>&quot;default&quot;</em><em>, </em><em>optional</em>) – indicates how to identify the labels in the input.
+By default, this will pick the second parameter as the labels if it’s a tensor. This covers the most
+common scenario where this transform is called as <code class="docutils literal notranslate"><span class="pre">MixUp()(imgs_batch,</span> <span class="pre">labels_batch)</span></code>.
+It can also be a callable that takes the same input as the transform, and returns the labels.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">MixUp</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.register_kernel.html" class="btn btn-neutral float-right" title="register_kernel" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.CutMix.html" class="btn btn-neutral" title="CutMix" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MixUp</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.Normalize.html b/0.16/generated/torchvision.transforms.v2.Normalize.html
new file mode 100644
index 00000000000..05f2c819ba3
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.Normalize.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Normalize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomErasing" href="torchvision.transforms.v2.RandomErasing.html" />
+    <link rel="prev" title="LinearTransformation" href="torchvision.transforms.v2.LinearTransformation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Normalize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.Normalize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="normalize">
+<h1>Normalize<a class="headerlink" href="#normalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Normalize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">Normalize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">mean</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#Normalize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Normalize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Normalize a tensor image or video with mean and standard deviation.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The Normalize transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support PIL Image.
+Given mean: <code class="docutils literal notranslate"><span class="pre">(mean[1],...,mean[n])</span></code> and std: <code class="docutils literal notranslate"><span class="pre">(std[1],..,std[n])</span></code> for <code class="docutils literal notranslate"><span class="pre">n</span></code>
+channels, this transform will normalize each channel of the input
+<code class="docutils literal notranslate"><span class="pre">torch.*Tensor</span></code> i.e.,
+<code class="docutils literal notranslate"><span class="pre">output[channel]</span> <span class="pre">=</span> <span class="pre">(input[channel]</span> <span class="pre">-</span> <span class="pre">mean[channel])</span> <span class="pre">/</span> <span class="pre">std[channel]</span></code></p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform acts out of place, i.e., it does not mutate the input tensor.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>mean</strong> (<em>sequence</em>) – Sequence of means for each channel.</p></li>
+<li><p><strong>std</strong> (<em>sequence</em>) – Sequence of standard deviations for each channel.</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>,</em><em>optional</em>) – Bool to make this operation in-place.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Normalize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomErasing.html" class="btn btn-neutral float-right" title="RandomErasing" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.LinearTransformation.html" class="btn btn-neutral" title="LinearTransformation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Normalize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.PILToTensor.html b/0.16/generated/torchvision.transforms.v2.PILToTensor.html
new file mode 100644
index 00000000000..3ade4353499
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.PILToTensor.html
@@ -0,0 +1,704 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>PILToTensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ToPILImage" href="torchvision.transforms.v2.ToPILImage.html" />
+    <link rel="prev" title="ToPureTensor" href="torchvision.transforms.v2.ToPureTensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>PILToTensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.PILToTensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="piltotensor">
+<h1>PILToTensor<a class="headerlink" href="#piltotensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.PILToTensor">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">PILToTensor</span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_type_conversion.html#PILToTensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.PILToTensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Convert a PIL Image to a tensor of the same type - this does not scale values.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The PILToTensor transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support torchscript.</p>
+<p>Converts a PIL Image (H x W x C) to a Tensor of shape (C x H x W).</p>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">PILToTensor</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ToPILImage.html" class="btn btn-neutral float-right" title="ToPILImage" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ToPureTensor.html" class="btn btn-neutral" title="ToPureTensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">PILToTensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.Pad.html b/0.16/generated/torchvision.transforms.v2.Pad.html
new file mode 100644
index 00000000000..7349e7c3565
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.Pad.html
@@ -0,0 +1,740 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Pad &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomZoomOut" href="torchvision.transforms.v2.RandomZoomOut.html" />
+    <link rel="prev" title="RandomVerticalFlip" href="torchvision.transforms.v2.RandomVerticalFlip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Pad</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.Pad.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="pad">
+<h1>Pad<a class="headerlink" href="#pad" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Pad">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">Pad</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Literal" title="(in Python v3.12)"><span class="pre">Literal</span></a><span class="p"><span class="pre">[</span></span><span class="s"><span class="pre">'constant'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'edge'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'reflect'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'symmetric'</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'constant'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#Pad"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Pad" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Pad the input on all sides with the given “pad” value.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The Pad transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em>) – <p>Padding on each border. If a single int is provided this
+is used to pad all borders. If sequence of length 2 is provided this is the padding
+on left/right and top/bottom respectively. If a sequence of length 4 is provided
+this is the padding for the left, top, right and bottom borders respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode padding as single int is not supported, use a sequence of
+length 1: <code class="docutils literal notranslate"><span class="pre">[padding,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – Pixel fill value used when the  <code class="docutils literal notranslate"><span class="pre">padding_mode</span></code> is constant.
+Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.
+Fill value can be also a dictionary mapping data type to the fill value, e.g.
+<code class="docutils literal notranslate"><span class="pre">fill={tv_tensors.Image:</span> <span class="pre">127,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">0}</span></code> where <code class="docutils literal notranslate"><span class="pre">Image</span></code> will be filled with 127 and
+<code class="docutils literal notranslate"><span class="pre">Mask</span></code> will be filled with 0.</p></li>
+<li><p><strong>padding_mode</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – <p>Type of padding. Should be: constant, edge, reflect or symmetric.
+Default is “constant”.</p>
+<ul>
+<li><p>constant: pads with a constant value, this value is specified with fill</p></li>
+<li><p>edge: pads with the last value at the edge of the image.</p></li>
+<li><p>reflect: pads with reflection of image without repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode
+will result in [3, 2, 1, 2, 3, 4, 3, 2]</p></li>
+<li><p>symmetric: pads with reflection of image repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode
+will result in [2, 1, 1, 2, 3, 4, 4, 3]</p></li>
+</ul>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Pad</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomZoomOut.html" class="btn btn-neutral float-right" title="RandomZoomOut" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomVerticalFlip.html" class="btn btn-neutral" title="RandomVerticalFlip" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Pad</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandAugment.html b/0.16/generated/torchvision.transforms.v2.RandAugment.html
new file mode 100644
index 00000000000..03c4d840cfe
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandAugment.html
@@ -0,0 +1,735 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandAugment &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="TrivialAugmentWide" href="torchvision.transforms.v2.TrivialAugmentWide.html" />
+    <link rel="prev" title="AutoAugment" href="torchvision.transforms.v2.AutoAugment.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandAugment</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandAugment.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randaugment">
+<h1>RandAugment<a class="headerlink" href="#randaugment" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandAugment">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandAugment</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_ops</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">magnitude</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">9</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_magnitude_bins</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">31</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#RandAugment"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandAugment" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] RandAugment data augmentation method based on
+<a class="reference external" href="https://arxiv.org/abs/1909.13719">“RandAugment: Practical automated data augmentation with a reduced search space”</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandAugment transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transformation works on images and videos only.</p>
+<p>If the input is <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it should be of type <code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code>, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>num_ops</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Number of augmentation transformations to apply sequentially.</p></li>
+<li><p><strong>magnitude</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Magnitude for all the transformations.</p></li>
+<li><p><strong>num_magnitude_bins</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The number of different magnitude values.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandAugment</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandAugment.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#RandAugment.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandAugment.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.TrivialAugmentWide.html" class="btn btn-neutral float-right" title="TrivialAugmentWide" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.AutoAugment.html" class="btn btn-neutral" title="AutoAugment" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandAugment</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html b/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
new file mode 100644
index 00000000000..3a8d1f471d6
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomAdjustSharpness &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomAutocontrast" href="torchvision.transforms.v2.RandomAutocontrast.html" />
+    <link rel="prev" title="RandomSolarize" href="torchvision.transforms.v2.RandomSolarize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomAdjustSharpness</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomAdjustSharpness.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomadjustsharpness">
+<h1>RandomAdjustSharpness<a class="headerlink" href="#randomadjustsharpness" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomAdjustSharpness">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomAdjustSharpness</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sharpness_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomAdjustSharpness"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomAdjustSharpness" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Adjust the sharpness of the image or video with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomAdjustSharpness transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>,
+it is expected to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>sharpness_factor</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – How much to adjust the sharpness. Can be
+any non-negative number. 0 gives a blurred image, 1 gives the
+original image while 2 increases the sharpness by a factor of 2.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being sharpened. Default value is 0.5</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomAdjustSharpness</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomAutocontrast.html" class="btn btn-neutral float-right" title="RandomAutocontrast" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomSolarize.html" class="btn btn-neutral" title="RandomSolarize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomAdjustSharpness</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAffine.html b/0.16/generated/torchvision.transforms.v2.RandomAffine.html
new file mode 100644
index 00000000000..b351530718a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomAffine.html
@@ -0,0 +1,749 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomAffine &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomPerspective" href="torchvision.transforms.v2.RandomPerspective.html" />
+    <link rel="prev" title="RandomRotation" href="torchvision.transforms.v2.RandomRotation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomAffine</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomAffine.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomaffine">
+<h1>RandomAffine<a class="headerlink" href="#randomaffine" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomAffine">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomAffine</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/numbers.html#numbers.Number" title="(in Python v3.12)"><span class="pre">Number</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">translate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shear</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomAffine"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomAffine" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Random affine transformation the input keeping center invariant.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomAffine transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>degrees</strong> (<em>sequence</em><em> or </em><em>number</em>) – Range of degrees to select from.
+If degrees is a number instead of sequence like (min, max), the range of degrees
+will be (-degrees, +degrees). Set to 0 to deactivate rotations.</p></li>
+<li><p><strong>translate</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em>, </em><em>optional</em>) – tuple of maximum absolute fraction for horizontal
+and vertical translations. For example translate=(a, b), then horizontal shift
+is randomly sampled in the range -img_width * a &lt; dx &lt; img_width * a and vertical shift is
+randomly sampled in the range -img_height * b &lt; dy &lt; img_height * b. Will not translate by default.</p></li>
+<li><p><strong>scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em>, </em><em>optional</em>) – scaling factor interval, e.g (a, b), then scale is
+randomly sampled from the range a &lt;= scale &lt;= b. Will keep original scale by default.</p></li>
+<li><p><strong>shear</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Range of degrees to select from.
+If shear is a number, a shear parallel to the x-axis in the range (-shear, +shear)
+will be applied. Else if shear is a sequence of 2 values a shear parallel to the x-axis in the
+range (shear[0], shear[1]) will be applied. Else if shear is a sequence of 4 values,
+an x-axis shear in (shear[0], shear[1]) and y-axis shear in (shear[2], shear[3]) will be applied.
+Will not apply shear by default.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – Pixel fill value used when the  <code class="docutils literal notranslate"><span class="pre">padding_mode</span></code> is constant.
+Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.
+Fill value can be also a dictionary mapping data type to the fill value, e.g.
+<code class="docutils literal notranslate"><span class="pre">fill={tv_tensors.Image:</span> <span class="pre">127,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">0}</span></code> where <code class="docutils literal notranslate"><span class="pre">Image</span></code> will be filled with 127 and
+<code class="docutils literal notranslate"><span class="pre">Mask</span></code> will be filled with 0.</p></li>
+<li><p><strong>center</strong> (<em>sequence</em><em>, </em><em>optional</em>) – Optional center of rotation, (x, y). Origin is the upper left corner.
+Default is the center of the image.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomAffine</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomAffine.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">translate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale_ranges</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shears</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">img_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomAffine.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomAffine.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for affine transformation</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>params to be passed to the affine transformation</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomPerspective.html" class="btn btn-neutral float-right" title="RandomPerspective" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomRotation.html" class="btn btn-neutral" title="RandomRotation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomAffine</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomApply.html b/0.16/generated/torchvision.transforms.v2.RandomApply.html
new file mode 100644
index 00000000000..870ddc483ce
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomApply.html
@@ -0,0 +1,746 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomApply &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomChoice" href="torchvision.transforms.v2.RandomChoice.html" />
+    <link rel="prev" title="Compose" href="torchvision.transforms.v2.Compose.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomApply</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomApply.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomapply">
+<h1>RandomApply<a class="headerlink" href="#randomapply" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomApply">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomApply</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.ModuleList.html#torch.nn.ModuleList" title="(in PyTorch v2.0)"><span class="pre">ModuleList</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#RandomApply"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomApply" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Apply randomly a list of transformations with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomApply transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In order to script the transformation, please use <code class="docutils literal notranslate"><span class="pre">torch.nn.ModuleList</span></code> as input instead of list/tuple of
+transforms as shown below:</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">transforms</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">RandomApply</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">transforms</span><span class="o">.</span><span class="n">ColorJitter</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">]),</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.3</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">scripted_transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">script</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Make sure to use only scriptable transformations, i.e. that work with <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code>, does not require
+<cite>lambda</cite> functions or <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>transforms</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a>) – list of transformations</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of applying the list of transforms</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomApply</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomApply.extra_repr">
+<span class="sig-name descname"><span class="pre">extra_repr</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#RandomApply.extra_repr"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomApply.extra_repr" title="Permalink to this definition">¶</a></dt>
+<dd><p>Set the extra representation of the module</p>
+<p>To print customized extra information, you should re-implement
+this method in your own modules. Both single-line and multi-line
+strings are acceptable.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomApply.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#RandomApply.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomApply.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomChoice.html" class="btn btn-neutral float-right" title="RandomChoice" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.Compose.html" class="btn btn-neutral" title="Compose" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomApply</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html b/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
new file mode 100644
index 00000000000..bfb126d1278
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomAutocontrast &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomEqualize" href="torchvision.transforms.v2.RandomEqualize.html" />
+    <link rel="prev" title="RandomAdjustSharpness" href="torchvision.transforms.v2.RandomAdjustSharpness.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomAutocontrast</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomAutocontrast.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomautocontrast">
+<h1>RandomAutocontrast<a class="headerlink" href="#randomautocontrast" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomAutocontrast">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomAutocontrast</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomAutocontrast"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomAutocontrast" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Autocontrast the pixels of the given image or video with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomAutocontrast transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being autocontrasted. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomEqualize.html" class="btn btn-neutral float-right" title="RandomEqualize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomAdjustSharpness.html" class="btn btn-neutral" title="RandomAdjustSharpness" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomAutocontrast</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html b/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
new file mode 100644
index 00000000000..3fc07dc4e6c
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
@@ -0,0 +1,698 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomChannelPermutation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomPhotometricDistort" href="torchvision.transforms.v2.RandomPhotometricDistort.html" />
+    <link rel="prev" title="ColorJitter" href="torchvision.transforms.v2.ColorJitter.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomChannelPermutation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomChannelPermutation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomchannelpermutation">
+<h1>RandomChannelPermutation<a class="headerlink" href="#randomchannelpermutation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomChannelPermutation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomChannelPermutation</span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomChannelPermutation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomChannelPermutation" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Randomly permute the channels of an image or video</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomChannelPermutation transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomPhotometricDistort.html" class="btn btn-neutral float-right" title="RandomPhotometricDistort" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ColorJitter.html" class="btn btn-neutral" title="ColorJitter" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomChannelPermutation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomChoice.html b/0.16/generated/torchvision.transforms.v2.RandomChoice.html
new file mode 100644
index 00000000000..d18eda41254
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomChoice.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomChoice &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomOrder" href="torchvision.transforms.v2.RandomOrder.html" />
+    <link rel="prev" title="RandomApply" href="torchvision.transforms.v2.RandomApply.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomChoice</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomChoice.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomchoice">
+<h1>RandomChoice<a class="headerlink" href="#randomchoice" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomChoice">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomChoice</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#RandomChoice"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomChoice" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Apply single transformation randomly picked from a list.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomChoice transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support torchscript.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>transforms</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a>) – list of transformations</p></li>
+<li><p><strong>p</strong> (<em>list of python:floats</em><em> or </em><em>None</em><em>, </em><em>optional</em>) – probability of each transform being picked.
+If <code class="docutils literal notranslate"><span class="pre">p</span></code> doesn’t sum to 1, it is automatically normalized. If <code class="docutils literal notranslate"><span class="pre">None</span></code>
+(default), all transforms have the same probability.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomChoice</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomChoice.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#RandomChoice.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomChoice.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomOrder.html" class="btn btn-neutral float-right" title="RandomOrder" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomApply.html" class="btn btn-neutral" title="RandomApply" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomChoice</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomCrop.html b/0.16/generated/torchvision.transforms.v2.RandomCrop.html
new file mode 100644
index 00000000000..ac34512925e
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomCrop.html
@@ -0,0 +1,770 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomResizedCrop" href="torchvision.transforms.v2.RandomResizedCrop.html" />
+    <link rel="prev" title="resize" href="torchvision.transforms.v2.functional.resize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomcrop">
+<h1>RandomCrop<a class="headerlink" href="#randomcrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_if_needed</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Literal" title="(in Python v3.12)"><span class="pre">Literal</span></a><span class="p"><span class="pre">[</span></span><span class="s"><span class="pre">'constant'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'edge'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'reflect'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'symmetric'</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'constant'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Crop the input at a random location.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomCrop transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p></li>
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em><em>, </em><em>optional</em>) – <p>Optional padding on each border
+of the image. Default is None. If a single int is provided this
+is used to pad all borders. If sequence of length 2 is provided this is the padding
+on left/right and top/bottom respectively. If a sequence of length 4 is provided
+this is the padding for the left, top, right and bottom borders respectively.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode padding as single int is not supported, use a sequence of
+length 1: <code class="docutils literal notranslate"><span class="pre">[padding,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>pad_if_needed</strong> (<em>boolean</em><em>, </em><em>optional</em>) – It will pad the image if smaller than the
+desired size to avoid raising an exception. Since cropping is done
+after padding, the padding seems to be done at a random offset.</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – Pixel fill value used when the  <code class="docutils literal notranslate"><span class="pre">padding_mode</span></code> is constant.
+Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.
+Fill value can be also a dictionary mapping data type to the fill value, e.g.
+<code class="docutils literal notranslate"><span class="pre">fill={tv_tensors.Image:</span> <span class="pre">127,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">0}</span></code> where <code class="docutils literal notranslate"><span class="pre">Image</span></code> will be filled with 127 and
+<code class="docutils literal notranslate"><span class="pre">Mask</span></code> will be filled with 0.</p></li>
+<li><p><strong>padding_mode</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>optional</em>) – <p>Type of padding. Should be: constant, edge, reflect or symmetric.
+Default is constant.</p>
+<ul>
+<li><p>constant: pads with a constant value, this value is specified with fill</p></li>
+<li><p>edge: pads with the last value at the edge of the image.</p></li>
+<li><p>reflect: pads with reflection of image without repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in reflect mode
+will result in [3, 2, 1, 2, 3, 4, 3, 2]</p></li>
+<li><p>symmetric: pads with reflection of image repeating the last value on the edge.
+For example, padding [1, 2, 3, 4] with 2 elements on both sides in symmetric mode
+will result in [2, 1, 1, 2, 3, 4, 4, 3]</p></li>
+</ul>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomCrop.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomCrop.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomCrop.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">crop</span></code> for a random crop.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Image to be cropped.</p></li>
+<li><p><strong>output_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a>) – Expected output size of the crop.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>params (i, j, h, w) to be passed to <code class="docutils literal notranslate"><span class="pre">crop</span></code> for random crop.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomResizedCrop.html" class="btn btn-neutral float-right" title="RandomResizedCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.resize.html" class="btn btn-neutral" title="resize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomEqualize.html b/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
new file mode 100644
index 00000000000..5f7bb8d2c04
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomEqualize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="permute_channels" href="torchvision.transforms.v2.functional.permute_channels.html" />
+    <link rel="prev" title="RandomAutocontrast" href="torchvision.transforms.v2.RandomAutocontrast.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomEqualize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomEqualize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomequalize">
+<h1>RandomEqualize<a class="headerlink" href="#randomequalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomEqualize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomEqualize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomEqualize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomEqualize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Equalize the histogram of the given image or video with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomEqualize transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “P”, “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being equalized. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomEqualize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.permute_channels.html" class="btn btn-neutral float-right" title="permute_channels" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomAutocontrast.html" class="btn btn-neutral" title="RandomAutocontrast" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomEqualize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomErasing.html b/0.16/generated/torchvision.transforms.v2.RandomErasing.html
new file mode 100644
index 00000000000..bcf717eadce
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomErasing.html
@@ -0,0 +1,753 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomErasing &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Lambda" href="torchvision.transforms.v2.Lambda.html" />
+    <link rel="prev" title="Normalize" href="torchvision.transforms.v2.Normalize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomErasing</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomErasing.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomerasing">
+<h1>RandomErasing<a class="headerlink" href="#randomerasing" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomErasing">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomErasing</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.02,</span> <span class="pre">0.33)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.3,</span> <span class="pre">3.3)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_augment.html#RandomErasing"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomErasing" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Randomly select a rectangle region in the input image or video and erase its pixels.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomErasing transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support PIL Image.
+‘Random Erasing Data Augmentation’ by Zhong et al. See <a class="reference external" href="https://arxiv.org/abs/1708.04896">https://arxiv.org/abs/1708.04896</a></p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – probability that the random erasing operation will be performed.</p></li>
+<li><p><strong>scale</strong> (<em>tuple of python:float</em><em>, </em><em>optional</em>) – range of proportion of erased area against input image.</p></li>
+<li><p><strong>ratio</strong> (<em>tuple of python:float</em><em>, </em><em>optional</em>) – range of aspect ratio of erased area.</p></li>
+<li><p><strong>value</strong> (<em>number</em><em> or </em><em>tuple of numbers</em>) – erasing value. Default is 0. If a single int, it is used to
+erase all pixels. If a tuple of length 3, it is used to erase
+R, G, B channels respectively.
+If a str of ‘random’, erasing each pixel with random values.</p></li>
+<li><p><strong>inplace</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – boolean to make this transform inplace. Default set to False.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Erased input.</p>
+</dd>
+</dl>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span> <span class="k">as</span> <span class="n">transforms</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">transform</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">PILToTensor</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">ConvertImageDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">Normalize</span><span class="p">((</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">),</span> <span class="p">(</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">)),</span>
+<span class="gp">&gt;&gt;&gt; </span>  <span class="n">transforms</span><span class="o">.</span><span class="n">RandomErasing</span><span class="p">(),</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="p">])</span>
+</pre></div>
+</div>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomErasing.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomErasing.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomErasing.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">erase</span></code> for a random erasing.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>Tensor</em>) – Tensor image to be erased.</p></li>
+<li><p><strong>scale</strong> (<em>sequence</em>) – range of proportion of erased area against input image.</p></li>
+<li><p><strong>ratio</strong> (<em>sequence</em>) – range of aspect ratio of erased area.</p></li>
+<li><p><strong>value</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a><em>, </em><em>optional</em>) – erasing value. If None, it is interpreted as “random”
+(erasing each pixel with random values). If <code class="docutils literal notranslate"><span class="pre">len(value)</span></code> is 1, it is interpreted as a number,
+i.e. <code class="docutils literal notranslate"><span class="pre">value[0]</span></code>.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>params (i, j, h, w, v) to be passed to <code class="docutils literal notranslate"><span class="pre">erase</span></code> for random erasing.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.Lambda.html" class="btn btn-neutral float-right" title="Lambda" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.Normalize.html" class="btn btn-neutral" title="Normalize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomErasing</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html b/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
new file mode 100644
index 00000000000..9cabe2183fa
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
@@ -0,0 +1,706 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomGrayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="GaussianBlur" href="torchvision.transforms.v2.GaussianBlur.html" />
+    <link rel="prev" title="Grayscale" href="torchvision.transforms.v2.Grayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomGrayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomGrayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomgrayscale">
+<h1>RandomGrayscale<a class="headerlink" href="#randomgrayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomGrayscale">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomGrayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomGrayscale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomGrayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Randomly convert image or videos to grayscale with a probability of p (default 0.1).</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomGrayscale transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it is expected to have […, 3 or 1, H, W] shape,
+where … means an arbitrary number of leading dimensions</p>
+<p>The output has the same number of channels as the input.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability that image should be converted to grayscale.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.GaussianBlur.html" class="btn btn-neutral float-right" title="GaussianBlur" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.Grayscale.html" class="btn btn-neutral" title="Grayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomGrayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html b/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
new file mode 100644
index 00000000000..5e348c1f69d
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomHorizontalFlip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomVerticalFlip" href="torchvision.transforms.v2.RandomVerticalFlip.html" />
+    <link rel="prev" title="five_crop" href="torchvision.transforms.v2.functional.five_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomHorizontalFlip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomHorizontalFlip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomhorizontalflip">
+<h1>RandomHorizontalFlip<a class="headerlink" href="#randomhorizontalflip" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomHorizontalFlip">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomHorizontalFlip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomHorizontalFlip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomHorizontalFlip" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Horizontally flip the input with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomHorizontalFlip transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – probability of the input being flipped. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomHorizontalFlip</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomVerticalFlip.html" class="btn btn-neutral float-right" title="RandomVerticalFlip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.five_crop.html" class="btn btn-neutral" title="five_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomHorizontalFlip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomInvert.html b/0.16/generated/torchvision.transforms.v2.RandomInvert.html
new file mode 100644
index 00000000000..2b66b1ea44a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomInvert.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomInvert &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomPosterize" href="torchvision.transforms.v2.RandomPosterize.html" />
+    <link rel="prev" title="GaussianBlur" href="torchvision.transforms.v2.GaussianBlur.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomInvert</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomInvert.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randominvert">
+<h1>RandomInvert<a class="headerlink" href="#randominvert" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomInvert">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomInvert</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomInvert"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomInvert" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Inverts the colors of the given image or video with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomInvert transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If img is a Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being color inverted. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomInvert</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomPosterize.html" class="btn btn-neutral float-right" title="RandomPosterize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.GaussianBlur.html" class="btn btn-neutral" title="GaussianBlur" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomInvert</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html b/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
new file mode 100644
index 00000000000..aa64bfc0a2a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
@@ -0,0 +1,728 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomIoUCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="CenterCrop" href="torchvision.transforms.v2.CenterCrop.html" />
+    <link rel="prev" title="RandomResizedCrop" href="torchvision.transforms.v2.RandomResizedCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomIoUCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomIoUCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomioucrop">
+<h1>RandomIoUCrop<a class="headerlink" href="#randomioucrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomIoUCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomIoUCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">min_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.3</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">min_aspect_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_aspect_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">2.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampler_options</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trials</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">40</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomIoUCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomIoUCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Random IoU crop transformation from
+<a class="reference external" href="https://arxiv.org/abs/1512.02325">“SSD: Single Shot MultiBox Detector”</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomIoUCrop transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transformation requires an image or video data and <code class="docutils literal notranslate"><span class="pre">tv_tensors.BoundingBoxes</span></code> in the input.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>In order to properly remove the bounding boxes below the IoU threshold, <cite>RandomIoUCrop</cite>
+must be followed by <a class="reference internal" href="torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes" title="torchvision.transforms.v2.SanitizeBoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">SanitizeBoundingBoxes</span></code></a>, either immediately
+after or later in the transforms pipeline.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>min_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – Minimum factors to scale the input size.</p></li>
+<li><p><strong>max_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – Maximum factors to scale the input size.</p></li>
+<li><p><strong>min_aspect_ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – Minimum aspect ratio for the cropped image or video.</p></li>
+<li><p><strong>max_aspect_ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – Maximum aspect ratio for the cropped image or video.</p></li>
+<li><p><strong>sampler_options</strong> (<em>list of python:float</em><em>, </em><em>optional</em>) – List of minimal IoU (Jaccard) overlap between all the boxes and
+a cropped image or video. Default, <code class="docutils literal notranslate"><span class="pre">None</span></code> which corresponds to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">0.1,</span> <span class="pre">0.3,</span> <span class="pre">0.5,</span> <span class="pre">0.7,</span> <span class="pre">0.9,</span> <span class="pre">1.0]</span></code></p></li>
+<li><p><strong>trials</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Number of trials to find a crop for a given value of minimal IoU (Jaccard) overlap.
+Default, 40.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomIoUCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.CenterCrop.html" class="btn btn-neutral float-right" title="CenterCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomResizedCrop.html" class="btn btn-neutral" title="RandomResizedCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomIoUCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomOrder.html b/0.16/generated/torchvision.transforms.v2.RandomOrder.html
new file mode 100644
index 00000000000..0b1c0130f01
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomOrder.html
@@ -0,0 +1,718 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomOrder &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="LinearTransformation" href="torchvision.transforms.v2.LinearTransformation.html" />
+    <link rel="prev" title="RandomChoice" href="torchvision.transforms.v2.RandomChoice.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomOrder</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomOrder.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomorder">
+<h1>RandomOrder<a class="headerlink" href="#randomorder" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomOrder">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomOrder</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">transforms</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#RandomOrder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomOrder" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Apply a list of transformations in a random order.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomOrder transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support torchscript.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>transforms</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><em>torch.nn.Module</em></a>) – list of transformations</p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomOrder.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_container.html#RandomOrder.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomOrder.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.LinearTransformation.html" class="btn btn-neutral float-right" title="LinearTransformation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomChoice.html" class="btn btn-neutral" title="RandomChoice" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomOrder</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPerspective.html b/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
new file mode 100644
index 00000000000..c37caf83bf6
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
@@ -0,0 +1,743 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomPerspective &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ElasticTransform" href="torchvision.transforms.v2.ElasticTransform.html" />
+    <link rel="prev" title="RandomAffine" href="torchvision.transforms.v2.RandomAffine.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomPerspective</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomPerspective.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomperspective">
+<h1>RandomPerspective<a class="headerlink" href="#randomperspective" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomPerspective">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomPerspective</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">distortion_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomPerspective"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomPerspective" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Perform a random perspective transformation of the input with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomPerspective transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>distortion_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – argument to control the degree of distortion and ranges from 0 to 1.
+Default is 0.5.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – probability of the input being transformed. Default is 0.5.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – Pixel fill value used when the  <code class="docutils literal notranslate"><span class="pre">padding_mode</span></code> is constant.
+Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.
+Fill value can be also a dictionary mapping data type to the fill value, e.g.
+<code class="docutils literal notranslate"><span class="pre">fill={tv_tensors.Image:</span> <span class="pre">127,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">0}</span></code> where <code class="docutils literal notranslate"><span class="pre">Image</span></code> will be filled with 127 and
+<code class="docutils literal notranslate"><span class="pre">Mask</span></code> will be filled with 0.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomPerspective</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomPerspective.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">distortion_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomPerspective.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomPerspective.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">perspective</span></code> for a random perspective transform.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – width of the image.</p></li>
+<li><p><strong>height</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – height of the image.</p></li>
+<li><p><strong>distortion_scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – argument to control the degree of distortion and ranges from 0 to 1.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>List containing [top-left, top-right, bottom-right, bottom-left] of the original image,
+List containing [top-left, top-right, bottom-right, bottom-left] of the transformed image.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ElasticTransform.html" class="btn btn-neutral float-right" title="ElasticTransform" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomAffine.html" class="btn btn-neutral" title="RandomAffine" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomPerspective</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html b/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
new file mode 100644
index 00000000000..d2a9401ea50
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
@@ -0,0 +1,727 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomPhotometricDistort &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Grayscale" href="torchvision.transforms.v2.Grayscale.html" />
+    <link rel="prev" title="RandomChannelPermutation" href="torchvision.transforms.v2.RandomChannelPermutation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomPhotometricDistort</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomPhotometricDistort.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomphotometricdistort">
+<h1>RandomPhotometricDistort<a class="headerlink" href="#randomphotometricdistort" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomPhotometricDistort">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomPhotometricDistort</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">brightness</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.875,</span> <span class="pre">1.125)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">contrast</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.5,</span> <span class="pre">1.5)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">saturation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.5,</span> <span class="pre">1.5)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hue</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(-</span> <span class="pre">0.05,</span> <span class="pre">0.05)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomPhotometricDistort"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomPhotometricDistort" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Randomly distorts the image or video as used in <a class="reference external" href="https://arxiv.org/abs/1512.02325">SSD: Single Shot
+MultiBox Detector</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomPhotometricDistort transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform relies on <a class="reference internal" href="torchvision.transforms.v2.ColorJitter.html#torchvision.transforms.v2.ColorJitter" title="torchvision.transforms.v2.ColorJitter"><code class="xref py py-class docutils literal notranslate"><span class="pre">ColorJitter</span></code></a>
+under the hood to adjust the contrast, saturation, hue, brightness, and also
+randomly permutes channels.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>brightness</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – How much to jitter brightness.
+brightness_factor is chosen uniformly from [min, max]. Should be non negative numbers.</p></li>
+<li><p><strong>float</strong> (<em>contrast tuple of</em>) – How much to jitter contrast.
+contrast_factor is chosen uniformly from [min, max]. Should be non-negative numbers.</p></li>
+<li><p><strong>saturation</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – How much to jitter saturation.
+saturation_factor is chosen uniformly from [min, max]. Should be non negative numbers.</p></li>
+<li><p><strong>hue</strong> (<em>tuple of python:float</em><em> (</em><em>min</em><em>, </em><em>max</em><em>)</em><em>, </em><em>optional</em>) – How much to jitter hue.
+hue_factor is chosen uniformly from [min, max].  Should have -0.5 &lt;= min &lt;= max &lt;= 0.5.
+To jitter hue, the pixel values of the input image has to be non-negative for conversion to HSV space;
+thus it does not work if you normalize your image to an interval with negative values,
+or use an interpolation that generates negative values before using this function.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em><em>) </em><em>probability each distortion operation</em><em> (</em><em>contrast</em><em>, </em><em>saturation</em><em>, </em><em>...</em>) – Default is 0.5.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomPhotometricDistort</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.Grayscale.html" class="btn btn-neutral float-right" title="Grayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomChannelPermutation.html" class="btn btn-neutral" title="RandomChannelPermutation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomPhotometricDistort</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPosterize.html b/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
new file mode 100644
index 00000000000..0d7a853fe0e
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomPosterize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomSolarize" href="torchvision.transforms.v2.RandomSolarize.html" />
+    <link rel="prev" title="RandomInvert" href="torchvision.transforms.v2.RandomInvert.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomPosterize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomPosterize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomposterize">
+<h1>RandomPosterize<a class="headerlink" href="#randomposterize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomPosterize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomPosterize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">bits</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomPosterize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomPosterize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Posterize the image or video with a given probability by reducing the
+number of bits for each color channel.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomPosterize transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it should be of type torch.uint8,
+and it is expected to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>bits</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – number of bits to keep for each channel (0-8)</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being posterized. Default value is 0.5</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomPosterize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomSolarize.html" class="btn btn-neutral float-right" title="RandomSolarize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomInvert.html" class="btn btn-neutral" title="RandomInvert" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomPosterize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomResize.html b/0.16/generated/torchvision.transforms.v2.RandomResize.html
new file mode 100644
index 00000000000..ee97e377a4f
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomResize.html
@@ -0,0 +1,742 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomResize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resize" href="torchvision.transforms.v2.functional.resize.html" />
+    <link rel="prev" title="RandomShortestSize" href="torchvision.transforms.v2.RandomShortestSize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomResize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomResize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomresize">
+<h1>RandomResize<a class="headerlink" href="#randomresize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomResize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomResize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">min_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomResize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomResize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Randomly resize the input.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomResize transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transformation can be used together with <code class="docutils literal notranslate"><span class="pre">RandomCrop</span></code> as data augmentations to train
+models on image segmentation task.</p>
+<p>Output spatial size is randomly sampled from the interval <code class="docutils literal notranslate"><span class="pre">[min_size,</span> <span class="pre">max_size]</span></code>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">size</span> <span class="o">=</span> <span class="n">uniform_sample</span><span class="p">(</span><span class="n">min_size</span><span class="p">,</span> <span class="n">max_size</span><span class="p">)</span>
+<span class="n">output_width</span> <span class="o">=</span> <span class="n">size</span>
+<span class="n">output_height</span> <span class="o">=</span> <span class="n">size</span>
+</pre></div>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>min_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Minimum output size for random sampling</p></li>
+<li><p><strong>max_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Maximum output size for random sampling</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.resize.html" class="btn btn-neutral float-right" title="resize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomShortestSize.html" class="btn btn-neutral" title="RandomShortestSize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomResize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html b/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
new file mode 100644
index 00000000000..8c939d1d5f6
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
@@ -0,0 +1,782 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomResizedCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomIoUCrop" href="torchvision.transforms.v2.RandomIoUCrop.html" />
+    <link rel="prev" title="RandomCrop" href="torchvision.transforms.v2.RandomCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomResizedCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomResizedCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomresizedcrop">
+<h1>RandomResizedCrop<a class="headerlink" href="#randomresizedcrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomResizedCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomResizedCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.08,</span> <span class="pre">1.0)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.75,</span> <span class="pre">1.3333333333333333)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomResizedCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomResizedCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Crop a random portion of the input and resize it to a given size.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomResizedCrop transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<p>A crop of the original input is made: the crop has a random area (H * W)
+and a random aspect ratio. This crop is finally resized to the given
+size. This is popularly used to train the Inception networks.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence</em>) – <p>expected output size of the crop, for each edge. If size is an
+int instead of sequence like (h, w), a square output size <code class="docutils literal notranslate"><span class="pre">(size,</span> <span class="pre">size)</span></code> is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode size as single int is not supported, use a sequence of length 1: <code class="docutils literal notranslate"><span class="pre">[size,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>scale</strong> (<em>tuple of python:float</em><em>, </em><em>optional</em>) – Specifies the lower and upper bounds for the random area of the crop,
+before resizing. The scale is defined with respect to the area of the original image.</p></li>
+<li><p><strong>ratio</strong> (<em>tuple of python:float</em><em>, </em><em>optional</em>) – lower and upper bounds for the random aspect ratio of the crop, before
+resizing.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomResizedCrop.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">img</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomResizedCrop.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomResizedCrop.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">crop</span></code> for a random sized crop.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>img</strong> (<em>PIL Image</em><em> or </em><em>Tensor</em>) – Input image.</p></li>
+<li><p><strong>scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a>) – range of scale of the origin size cropped</p></li>
+<li><p><strong>ratio</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a>) – range of aspect ratio of the origin aspect ratio cropped</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>params (i, j, h, w) to be passed to <code class="docutils literal notranslate"><span class="pre">crop</span></code> for a random
+sized crop.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)">tuple</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomIoUCrop.html" class="btn btn-neutral float-right" title="RandomIoUCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomCrop.html" class="btn btn-neutral" title="RandomCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomResizedCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomRotation.html b/0.16/generated/torchvision.transforms.v2.RandomRotation.html
new file mode 100644
index 00000000000..727d781763b
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomRotation.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomRotation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomAffine" href="torchvision.transforms.v2.RandomAffine.html" />
+    <link rel="prev" title="RandomZoomOut" href="torchvision.transforms.v2.RandomZoomOut.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomRotation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomRotation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomrotation">
+<h1>RandomRotation<a class="headerlink" href="#randomrotation" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomRotation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomRotation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/numbers.html#numbers.Number" title="(in Python v3.12)"><span class="pre">Number</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">expand</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomRotation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomRotation" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Rotate the input by angle.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomRotation transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>degrees</strong> (<em>sequence</em><em> or </em><em>number</em>) – Range of degrees to select from.
+If degrees is a number instead of sequence like (min, max), the range of degrees
+will be (-degrees, +degrees).</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>expand</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Optional expansion flag.
+If true, expands the output to make it large enough to hold the entire rotated image.
+If false or omitted, make the output image the same size as the input image.
+Note that the expand flag assumes rotation around the center (see note below) and no translation.</p></li>
+<li><p><strong>center</strong> (<em>sequence</em><em>, </em><em>optional</em>) – <p>Optional center of rotation, (x, y). Origin is the upper left corner.
+Default is the center of the image.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In theory, setting <code class="docutils literal notranslate"><span class="pre">center</span></code> has no effect if <code class="docutils literal notranslate"><span class="pre">expand=True</span></code>, since the image center will become the
+center of rotation. In practice however, due to numerical precision, this can lead to off-by-one
+differences of the resulting image size compared to using the image center in the first place. Thus, when
+setting <code class="docutils literal notranslate"><span class="pre">expand=True</span></code>, it’s best to leave <code class="docutils literal notranslate"><span class="pre">center=None</span></code> (default).</p>
+</div>
+</p></li>
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – Pixel fill value used when the  <code class="docutils literal notranslate"><span class="pre">padding_mode</span></code> is constant.
+Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.
+Fill value can be also a dictionary mapping data type to the fill value, e.g.
+<code class="docutils literal notranslate"><span class="pre">fill={tv_tensors.Image:</span> <span class="pre">127,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">0}</span></code> where <code class="docutils literal notranslate"><span class="pre">Image</span></code> will be filled with 127 and
+<code class="docutils literal notranslate"><span class="pre">Mask</span></code> will be filled with 0.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomRotation</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomRotation.get_params">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">degrees</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/transforms.html#RandomRotation.get_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomRotation.get_params" title="Permalink to this definition">¶</a></dt>
+<dd><p>Get parameters for <code class="docutils literal notranslate"><span class="pre">rotate</span></code> for a random rotation.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>angle parameter to be passed to <code class="docutils literal notranslate"><span class="pre">rotate</span></code> for random rotation.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)">float</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomAffine.html" class="btn btn-neutral float-right" title="RandomAffine" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomZoomOut.html" class="btn btn-neutral" title="RandomZoomOut" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomRotation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html b/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
new file mode 100644
index 00000000000..f02e32122b3
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
@@ -0,0 +1,734 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomShortestSize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomResize" href="torchvision.transforms.v2.RandomResize.html" />
+    <link rel="prev" title="ScaleJitter" href="torchvision.transforms.v2.ScaleJitter.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomShortestSize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomShortestSize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomshortestsize">
+<h1>RandomShortestSize<a class="headerlink" href="#randomshortestsize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomShortestSize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomShortestSize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">min_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomShortestSize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomShortestSize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Randomly resize the input.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomShortestSize transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>min_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em> or </em><em>sequence of python:int</em>) – Minimum spatial size. Single integer value or a sequence of integer values.</p></li>
+<li><p><strong>max_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Maximum spatial size. Default, None.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomResize.html" class="btn btn-neutral float-right" title="RandomResize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ScaleJitter.html" class="btn btn-neutral" title="ScaleJitter" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomShortestSize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomSolarize.html b/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
new file mode 100644
index 00000000000..38733e5d058
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomSolarize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomAdjustSharpness" href="torchvision.transforms.v2.RandomAdjustSharpness.html" />
+    <link rel="prev" title="RandomPosterize" href="torchvision.transforms.v2.RandomPosterize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomSolarize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomSolarize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomsolarize">
+<h1>RandomSolarize<a class="headerlink" href="#randomsolarize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomSolarize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomSolarize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">threshold</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_color.html#RandomSolarize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomSolarize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Solarize the image or video with a given probability by inverting all pixel
+values above a threshold.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomSolarize transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If img is a Tensor, it is expected to be in […, 1 or 3, H, W] format,
+where … means it can have an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>threshold</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – all pixels equal or above this value are inverted.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – probability of the image being solarized. Default value is 0.5</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomSolarize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomAdjustSharpness.html" class="btn btn-neutral float-right" title="RandomAdjustSharpness" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomPosterize.html" class="btn btn-neutral" title="RandomPosterize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomSolarize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html b/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
new file mode 100644
index 00000000000..bc5c42a8983
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomVerticalFlip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Pad" href="torchvision.transforms.v2.Pad.html" />
+    <link rel="prev" title="RandomHorizontalFlip" href="torchvision.transforms.v2.RandomHorizontalFlip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomVerticalFlip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomVerticalFlip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomverticalflip">
+<h1>RandomVerticalFlip<a class="headerlink" href="#randomverticalflip" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomVerticalFlip">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomVerticalFlip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomVerticalFlip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomVerticalFlip" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Vertically flip the input with a given probability.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomVerticalFlip transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – probability of the input being flipped. Default value is 0.5</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomVerticalFlip</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.Pad.html" class="btn btn-neutral float-right" title="Pad" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomHorizontalFlip.html" class="btn btn-neutral" title="RandomHorizontalFlip" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomVerticalFlip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html b/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
new file mode 100644
index 00000000000..0b25d42e5a0
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RandomZoomOut &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomRotation" href="torchvision.transforms.v2.RandomRotation.html" />
+    <link rel="prev" title="Pad" href="torchvision.transforms.v2.Pad.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>RandomZoomOut</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.RandomZoomOut.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="randomzoomout">
+<h1>RandomZoomOut<a class="headerlink" href="#randomzoomout" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.RandomZoomOut">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">RandomZoomOut</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">side_range</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(1.0,</span> <span class="pre">4.0)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.5</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#RandomZoomOut"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.RandomZoomOut" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] “Zoom out” transformation from
+<a class="reference external" href="https://arxiv.org/abs/1512.02325">“SSD: Single Shot MultiBox Detector”</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The RandomZoomOut transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transformation randomly pads images, videos, bounding boxes and masks creating a zoom out effect.
+Output spatial size is randomly sampled from original size up to a maximum size configured
+with <code class="docutils literal notranslate"><span class="pre">side_range</span></code> parameter:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">r</span> <span class="o">=</span> <span class="n">uniform_sample</span><span class="p">(</span><span class="n">side_range</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">side_range</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+<span class="n">output_width</span> <span class="o">=</span> <span class="n">input_width</span> <span class="o">*</span> <span class="n">r</span>
+<span class="n">output_height</span> <span class="o">=</span> <span class="n">input_height</span> <span class="o">*</span> <span class="n">r</span>
+</pre></div>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>fill</strong> (<em>number</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#dict" title="(in Python v3.12)"><em>dict</em></a><em>, </em><em>optional</em>) – Pixel fill value used when the  <code class="docutils literal notranslate"><span class="pre">padding_mode</span></code> is constant.
+Default is 0. If a tuple of length 3, it is used to fill R, G, B channels respectively.
+Fill value can be also a dictionary mapping data type to the fill value, e.g.
+<code class="docutils literal notranslate"><span class="pre">fill={tv_tensors.Image:</span> <span class="pre">127,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">0}</span></code> where <code class="docutils literal notranslate"><span class="pre">Image</span></code> will be filled with 127 and
+<code class="docutils literal notranslate"><span class="pre">Mask</span></code> will be filled with 0.</p></li>
+<li><p><strong>side_range</strong> (<em>sequence of python:floats</em><em>, </em><em>optional</em>) – tuple of two floats defines minimum and maximum factors to
+scale the input size.</p></li>
+<li><p><strong>p</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – probability that the zoom operation will be performed.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">RandomZoomOut</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomRotation.html" class="btn btn-neutral float-right" title="RandomRotation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.Pad.html" class="btn btn-neutral" title="Pad" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RandomZoomOut</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.Resize.html b/0.16/generated/torchvision.transforms.v2.Resize.html
new file mode 100644
index 00000000000..079a2853e80
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.Resize.html
@@ -0,0 +1,762 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Resize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ScaleJitter" href="torchvision.transforms.v2.ScaleJitter.html" />
+    <link rel="prev" title="Transforming and augmenting images" href="../transforms.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>Resize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.Resize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resize">
+<h1>Resize<a class="headerlink" href="#resize" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.Resize">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">Resize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#Resize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.Resize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Resize the input to the given size.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The Resize transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The output image might be different depending on its type: when downsampling, the interpolation of PIL images
+and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences
+in the performance of a network. Therefore, it is preferable to train and serve a model with the same input
+types. See also below the <code class="docutils literal notranslate"><span class="pre">antialias</span></code> parameter, which can help making the output of PIL images and tensors
+closer.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – <p>Desired output size. If size is a sequence like
+(h, w), output size will be matched to this. If size is an int,
+smaller edge of the image will be matched to this number.
+i.e, if height &gt; width, then image will be rescaled to
+(size * height / width, size).</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In torchscript mode size as single int is not supported, use a sequence of length 1: <code class="docutils literal notranslate"><span class="pre">[size,</span> <span class="pre">]</span></code>.</p>
+</div>
+</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>max_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The maximum allowed for the longer edge of
+the resized image. If the longer edge of the image is greater
+than <code class="docutils literal notranslate"><span class="pre">max_size</span></code> after being resized according to <code class="docutils literal notranslate"><span class="pre">size</span></code>,
+<code class="docutils literal notranslate"><span class="pre">size</span></code> will be overruled so that the longer edge is equal to
+<code class="docutils literal notranslate"><span class="pre">max_size</span></code>.
+As a result, the smaller edge may be shorter than <code class="docutils literal notranslate"><span class="pre">size</span></code>. This
+is only supported if <code class="docutils literal notranslate"><span class="pre">size</span></code> is an int (or a sequence of length
+1 in torchscript mode).</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Resize</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ScaleJitter.html" class="btn btn-neutral float-right" title="ScaleJitter" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../transforms.html" class="btn btn-neutral" title="Transforming and augmenting images" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Resize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html b/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
new file mode 100644
index 00000000000..5531bbe3ea7
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
@@ -0,0 +1,744 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SanitizeBoundingBoxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ClampBoundingBoxes" href="torchvision.transforms.v2.ClampBoundingBoxes.html" />
+    <link rel="prev" title="Lambda" href="torchvision.transforms.v2.Lambda.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>SanitizeBoundingBoxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="sanitizeboundingboxes">
+<h1>SanitizeBoundingBoxes<a class="headerlink" href="#sanitizeboundingboxes" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.SanitizeBoundingBoxes">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">SanitizeBoundingBoxes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">min_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels_getter</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'default'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#SanitizeBoundingBoxes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.SanitizeBoundingBoxes" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Remove degenerate/invalid bounding boxes and their corresponding labels and masks.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The SanitizeBoundingBoxes transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform removes bounding boxes and their associated labels/masks that:</p>
+<ul class="simple">
+<li><p>are below a given <code class="docutils literal notranslate"><span class="pre">min_size</span></code>: by default this also removes degenerate boxes that have e.g. X2 &lt;= X1.</p></li>
+<li><p>have any coordinate outside of their corresponding image. You may want to
+call <a class="reference internal" href="torchvision.transforms.v2.ClampBoundingBoxes.html#torchvision.transforms.v2.ClampBoundingBoxes" title="torchvision.transforms.v2.ClampBoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">ClampBoundingBoxes</span></code></a> first to avoid undesired removals.</p></li>
+</ul>
+<p>It is recommended to call it at the end of a pipeline, before passing the
+input to the models. It is critical to call this transform if
+<a class="reference internal" href="torchvision.transforms.v2.RandomIoUCrop.html#torchvision.transforms.v2.RandomIoUCrop" title="torchvision.transforms.v2.RandomIoUCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomIoUCrop</span></code></a> was called.
+If you want to be extra careful, you may call it after all transforms that
+may modify bounding boxes but once at the end should be enough in most
+cases.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>min_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – </p></li>
+<li><p><strong>labels_getter</strong> (<em>callable</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em> or </em><em>None</em><em>, </em><em>optional</em>) – indicates how to identify the labels in the input.
+By default, this will try to find a “labels” key in the input (case-insensitive), if
+the input is a dict or it is a tuple whose second element is a dict.
+This heuristic should work well with a lot of datasets, including the built-in torchvision datasets.
+It can also be a callable that takes the same input
+as the transform, and returns the labels.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">SanitizeBoundingBoxes</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.SanitizeBoundingBoxes.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#SanitizeBoundingBoxes.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.SanitizeBoundingBoxes.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ClampBoundingBoxes.html" class="btn btn-neutral float-right" title="ClampBoundingBoxes" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.Lambda.html" class="btn btn-neutral" title="Lambda" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SanitizeBoundingBoxes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ScaleJitter.html b/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
new file mode 100644
index 00000000000..290b4b6f567
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
@@ -0,0 +1,736 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ScaleJitter &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomShortestSize" href="torchvision.transforms.v2.RandomShortestSize.html" />
+    <link rel="prev" title="Resize" href="torchvision.transforms.v2.Resize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ScaleJitter</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ScaleJitter.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="scalejitter">
+<h1>ScaleJitter<a class="headerlink" href="#scalejitter" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ScaleJitter">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ScaleJitter</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">target_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale_range</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.1,</span> <span class="pre">2.0)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#ScaleJitter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ScaleJitter" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Perform Large Scale Jitter on the input according to
+<a class="reference external" href="https://arxiv.org/abs/2012.07177">“Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation”</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ScaleJitter transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> (e.g. <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>,
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a>, <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> etc.)
+it can have arbitrary number of leading batch dimensions. For example,
+the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape. A bounding box can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">4]</span></code> shape.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>target_size</strong> (<em>tuple of python:int</em>) – Target size. This parameter defines base scale for jittering,
+e.g. <code class="docutils literal notranslate"><span class="pre">min(target_size[0]</span> <span class="pre">/</span> <span class="pre">width,</span> <span class="pre">target_size[1]</span> <span class="pre">/</span> <span class="pre">height)</span></code>.</p></li>
+<li><p><strong>scale_range</strong> (<em>tuple of python:float</em><em>, </em><em>optional</em>) – Minimum and maximum of the scale range. Default, <code class="docutils literal notranslate"><span class="pre">(0.1,</span> <span class="pre">2.0)</span></code>.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST_EXACT</span></code>,
+<code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> and <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BICUBIC</span></code> are supported.
+The corresponding Pillow integer constants, e.g. <code class="docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code> are accepted as well.</p></li>
+<li><p><strong>antialias</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – <p>Whether to apply antialiasing.
+It only affects <strong>tensors</strong> with bilinear or bicubic modes and it is
+ignored otherwise: on PIL images, antialiasing is always applied on
+bilinear or bicubic modes; on other modes (for PIL images and
+tensors), antialiasing makes no sense and this parameter is ignored.
+Possible values are:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">True</span></code>: will apply antialiasing for bilinear or bicubic modes.
+Other mode aren’t affected. This is probably what you want to use.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">False</span></code>: will not apply antialiasing for tensors on any mode. PIL
+images are still antialiased on bilinear or bicubic modes, because
+PIL doesn’t support no antialias.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">None</span></code>: equivalent to <code class="docutils literal notranslate"><span class="pre">False</span></code> for tensors and <code class="docutils literal notranslate"><span class="pre">True</span></code> for
+PIL images. This value exists for legacy reasons and you probably
+don’t want to use it unless you really know what you are doing.</p></li>
+</ul>
+<p>The current default is <code class="docutils literal notranslate"><span class="pre">None</span></code> <strong>but will change to</strong> <code class="docutils literal notranslate"><span class="pre">True</span></code> <strong>in
+v0.17</strong> for the PIL and Tensor backends to be consistent.</p>
+</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomShortestSize.html" class="btn btn-neutral float-right" title="RandomShortestSize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.Resize.html" class="btn btn-neutral" title="Resize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ScaleJitter</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.TenCrop.html b/0.16/generated/torchvision.transforms.v2.TenCrop.html
new file mode 100644
index 00000000000..fd21fa0bda4
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.TenCrop.html
@@ -0,0 +1,719 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>TenCrop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="crop" href="torchvision.transforms.v2.functional.crop.html" />
+    <link rel="prev" title="FiveCrop" href="torchvision.transforms.v2.FiveCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>TenCrop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.TenCrop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="tencrop">
+<h1>TenCrop<a class="headerlink" href="#tencrop" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.TenCrop">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">TenCrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">vertical_flip</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_geometry.html#TenCrop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.TenCrop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Crop the image or video into four corners and the central crop plus the flipped version of
+these (horizontal flipping is used by default).</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The TenCrop transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>If the input is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> or a <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a> or a
+<a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">Video</span></code></a> it can have arbitrary number of leading batch dimensions.
+For example, the image can have <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> shape.</p>
+<p>See <a class="reference internal" href="torchvision.transforms.v2.FiveCrop.html#torchvision.transforms.v2.FiveCrop" title="torchvision.transforms.v2.FiveCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">FiveCrop</span></code></a> for an example.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>This transform returns a tuple of images and there may be a mismatch in the number of
+inputs and targets your Dataset returns. See below for an example of how to deal with
+this.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>size</strong> (<em>sequence</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Desired output size of the crop. If size is an
+int instead of sequence like (h, w), a square crop (size, size) is
+made. If provided a sequence of length 1, it will be interpreted as (size[0], size[0]).</p></li>
+<li><p><strong>vertical_flip</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Use vertical flipping instead of horizontal</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.crop.html" class="btn btn-neutral float-right" title="crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.FiveCrop.html" class="btn btn-neutral" title="FiveCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">TenCrop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ToDtype.html b/0.16/generated/torchvision.transforms.v2.ToDtype.html
new file mode 100644
index 00000000000..911bdf01e7f
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ToDtype.html
@@ -0,0 +1,726 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ToDtype &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ConvertBoundingBoxFormat" href="torchvision.transforms.v2.ConvertBoundingBoxFormat.html" />
+    <link rel="prev" title="ToPILImage" href="torchvision.transforms.v2.ToPILImage.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ToDtype</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ToDtype.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="todtype">
+<h1>ToDtype<a class="headerlink" href="#todtype" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ToDtype">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ToDtype</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_misc.html#ToDtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ToDtype" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Converts the input to a specific dtype, optionally scaling the values for images or videos.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ToDtype transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p><code class="docutils literal notranslate"><span class="pre">ToDtype(dtype,</span> <span class="pre">scale=True)</span></code> is the recommended replacement for <code class="docutils literal notranslate"><span class="pre">ConvertImageDtype(dtype)</span></code>.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>dtype</strong> (<code class="docutils literal notranslate"><span class="pre">torch.dtype</span></code> or dict of <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code> -&gt; <code class="docutils literal notranslate"><span class="pre">torch.dtype</span></code>) – The dtype to convert to.
+If a <code class="docutils literal notranslate"><span class="pre">torch.dtype</span></code> is passed, e.g. <code class="docutils literal notranslate"><span class="pre">torch.float32</span></code>, only images and videos will be converted
+to that dtype: this is for compatibility with <a class="reference internal" href="torchvision.transforms.v2.ConvertImageDtype.html#torchvision.transforms.v2.ConvertImageDtype" title="torchvision.transforms.v2.ConvertImageDtype"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvertImageDtype</span></code></a>.
+A dict can be passed to specify per-tv_tensor conversions, e.g.
+<code class="docutils literal notranslate"><span class="pre">dtype={tv_tensors.Image:</span> <span class="pre">torch.float32,</span> <span class="pre">tv_tensors.Mask:</span> <span class="pre">torch.int64,</span> <span class="pre">&quot;others&quot;:None}</span></code>. The “others”
+key can be used as a catch-all for any other tv_tensor type, and <code class="docutils literal notranslate"><span class="pre">None</span></code> means no conversion.</p></li>
+<li><p><strong>scale</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Whether to scale the values for images or videos. See <a class="reference internal" href="../transforms.html#range-and-dtype"><span class="std std-ref">Dtype and expected value range</span></a>.
+Default: <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">ToDtype</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="CutMix and MixUp are popular augmentation strategies that can improve classification accuracy."><img alt="" src="../_images/sphx_glr_plot_cutmix_mixup_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to use CutMix and MixUp</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ConvertBoundingBoxFormat.html" class="btn btn-neutral float-right" title="ConvertBoundingBoxFormat" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ToPILImage.html" class="btn btn-neutral" title="ToPILImage" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ToDtype</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ToImage.html b/0.16/generated/torchvision.transforms.v2.ToImage.html
new file mode 100644
index 00000000000..728483fb4eb
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ToImage.html
@@ -0,0 +1,704 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ToImage &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ToPureTensor" href="torchvision.transforms.v2.ToPureTensor.html" />
+    <link rel="prev" title="uniform_temporal_subsample" href="torchvision.transforms.v2.functional.uniform_temporal_subsample.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ToImage</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ToImage.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="toimage">
+<h1>ToImage<a class="headerlink" href="#toimage" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ToImage">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ToImage</span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_type_conversion.html#ToImage"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ToImage" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Convert a tensor, ndarray, or PIL Image to <a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a>
+; this does not scale values.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ToImage transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support torchscript.</p>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">ToImage</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ToPureTensor.html" class="btn btn-neutral float-right" title="ToPureTensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.uniform_temporal_subsample.html" class="btn btn-neutral" title="uniform_temporal_subsample" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ToImage</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ToPILImage.html b/0.16/generated/torchvision.transforms.v2.ToPILImage.html
new file mode 100644
index 00000000000..a6716625373
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ToPILImage.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ToPILImage &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ToDtype" href="torchvision.transforms.v2.ToDtype.html" />
+    <link rel="prev" title="PILToTensor" href="torchvision.transforms.v2.PILToTensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ToPILImage</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ToPILImage.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="topilimage">
+<h1>ToPILImage<a class="headerlink" href="#topilimage" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ToPILImage">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ToPILImage</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_type_conversion.html#ToPILImage"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ToPILImage" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Convert a tensor or an ndarray to PIL Image</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ToPILImage transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transform does not support torchscript.</p>
+<p>Converts a torch.*Tensor of shape C x H x W or a numpy ndarray of shape
+H x W x C to a PIL Image while adjusting the value range depending on the <code class="docutils literal notranslate"><span class="pre">mode</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>mode</strong> (<a class="reference external" href="https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes">PIL.Image mode</a>) – <p>color space and pixel depth of input data (optional).
+If <code class="docutils literal notranslate"><span class="pre">mode</span></code> is <code class="docutils literal notranslate"><span class="pre">None</span></code> (default) there are some assumptions made about the input data:</p>
+<ul class="simple">
+<li><p>If the input has 4 channels, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is assumed to be <code class="docutils literal notranslate"><span class="pre">RGBA</span></code>.</p></li>
+<li><p>If the input has 3 channels, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is assumed to be <code class="docutils literal notranslate"><span class="pre">RGB</span></code>.</p></li>
+<li><p>If the input has 2 channels, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is assumed to be <code class="docutils literal notranslate"><span class="pre">LA</span></code>.</p></li>
+<li><p>If the input has 1 channel, the <code class="docutils literal notranslate"><span class="pre">mode</span></code> is determined by the data type (i.e <code class="docutils literal notranslate"><span class="pre">int</span></code>, <code class="docutils literal notranslate"><span class="pre">float</span></code>,
+<code class="docutils literal notranslate"><span class="pre">short</span></code>).</p></li>
+</ul>
+</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ToDtype.html" class="btn btn-neutral float-right" title="ToDtype" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.PILToTensor.html" class="btn btn-neutral" title="PILToTensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ToPILImage</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ToPureTensor.html b/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
new file mode 100644
index 00000000000..292449f67af
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
@@ -0,0 +1,699 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ToPureTensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="PILToTensor" href="torchvision.transforms.v2.PILToTensor.html" />
+    <link rel="prev" title="ToImage" href="torchvision.transforms.v2.ToImage.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ToPureTensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ToPureTensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="topuretensor">
+<h1>ToPureTensor<a class="headerlink" href="#topuretensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ToPureTensor">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ToPureTensor</span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_type_conversion.html#ToPureTensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ToPureTensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Convert all tv_tensors to pure tensors, removing associated metadata (if any).</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ToPureTensor transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This doesn’t scale or change the values, only the type.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.PILToTensor.html" class="btn btn-neutral float-right" title="PILToTensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ToImage.html" class="btn btn-neutral" title="ToImage" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ToPureTensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.ToTensor.html b/0.16/generated/torchvision.transforms.v2.ToTensor.html
new file mode 100644
index 00000000000..08f955f02c0
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.ToTensor.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ToTensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="to_tensor" href="torchvision.transforms.v2.functional.to_tensor.html" />
+    <link rel="prev" title="torchvision.transforms.v2.functional.convert_bounding_box_format" href="torchvision.transforms.v2.functional.convert_bounding_box_format.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ToTensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.ToTensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="totensor">
+<h1>ToTensor<a class="headerlink" href="#totensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.ToTensor">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">ToTensor</span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_deprecated.html#ToTensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.ToTensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] [DEPRECATED] Use <code class="docutils literal notranslate"><span class="pre">v2.Compose([v2.ToImage(),</span> <span class="pre">v2.ToDtype(torch.float32,</span> <span class="pre">scale=True)])</span></code> instead.</p>
+<p>Convert a PIL Image or ndarray to tensor and scale the values accordingly.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The ToTensor transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p><code class="xref py py-class docutils literal notranslate"><span class="pre">v2.ToTensor</span></code> is deprecated and will be removed in a future release.
+Please use instead <code class="docutils literal notranslate"><span class="pre">v2.Compose([v2.ToImage(),</span> <span class="pre">v2.ToDtype(torch.float32,</span> <span class="pre">scale=True)])</span></code>.</p>
+</div>
+<p>This transform does not support torchscript.</p>
+<p>Converts a PIL Image or numpy.ndarray (H x W x C) in the range
+[0, 255] to a torch.FloatTensor of shape (C x H x W) in the range [0.0, 1.0]
+if the PIL Image belongs to one of the modes (L, LA, P, I, F, RGB, YCbCr, RGBA, CMYK, 1)
+or if the numpy.ndarray has dtype = np.uint8</p>
+<p>In the other cases, tensors are returned without scaling.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Because the input image is scaled to [0.0, 1.0], this transformation should not be used when
+transforming target image masks. See the <a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation">references</a> for implementing the transforms for image masks.</p>
+</div>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.to_tensor.html" class="btn btn-neutral float-right" title="to_tensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.convert_bounding_box_format.html" class="btn btn-neutral" title="torchvision.transforms.v2.functional.convert_bounding_box_format" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ToTensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html b/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
new file mode 100644
index 00000000000..71f44bda931
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
@@ -0,0 +1,733 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>TrivialAugmentWide &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="AugMix" href="torchvision.transforms.v2.AugMix.html" />
+    <link rel="prev" title="RandAugment" href="torchvision.transforms.v2.RandAugment.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>TrivialAugmentWide</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.TrivialAugmentWide.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="trivialaugmentwide">
+<h1>TrivialAugmentWide<a class="headerlink" href="#trivialaugmentwide" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.TrivialAugmentWide">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">TrivialAugmentWide</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_magnitude_bins</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">31</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><span class="pre">Dict</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Type" title="(in Python v3.12)"><span class="pre">Type</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Sequence" title="(in Python v3.12)"><span class="pre">Sequence</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#TrivialAugmentWide"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.TrivialAugmentWide" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Dataset-independent data-augmentation with TrivialAugment Wide, as described in
+<a class="reference external" href="https://arxiv.org/abs/2103.10158">“TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation”</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The TrivialAugmentWide transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>This transformation works on images and videos only.</p>
+<p>If the input is <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, it should be of type <code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code>, and it is expected
+to have […, 1 or 3, H, W] shape, where … means an arbitrary number of leading dimensions.
+If img is PIL Image, it is expected to be in mode “L” or “RGB”.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>num_magnitude_bins</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – The number of different magnitude values.</p></li>
+<li><p><strong>interpolation</strong> (<em>InterpolationMode</em><em>, </em><em>optional</em>) – Desired interpolation enum defined by
+<code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.transforms.InterpolationMode</span></code>. Default is <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>.
+If input is Tensor, only <code class="docutils literal notranslate"><span class="pre">InterpolationMode.NEAREST</span></code>, <code class="docutils literal notranslate"><span class="pre">InterpolationMode.BILINEAR</span></code> are supported.</p></li>
+<li><p><strong>fill</strong> (<em>sequence</em><em> or </em><em>number</em><em>, </em><em>optional</em>) – Pixel fill value for the area outside the transformed
+image. If given a number, the value is used for all bands respectively.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">TrivialAugmentWide</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the various transforms available in the torchvision.transforms..."><img alt="" src="../_images/sphx_glr_plot_transforms_illustrations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_illustrations.html#sphx-glr-auto-examples-transforms-plot-transforms-illustrations-py"><span class="std std-ref">Illustration of transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Illustration of transforms</div>
+</div></div><dl class="py method">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.TrivialAugmentWide.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_auto_augment.html#TrivialAugmentWide.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.TrivialAugmentWide.forward" title="Permalink to this definition">¶</a></dt>
+<dd><p>Defines the computation performed at every call.</p>
+<p>Should be overridden by all subclasses.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Although the recipe for forward pass needs to be defined within
+this function, one should call the <code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code> instance afterwards
+instead of this since the former takes care of running the
+registered hooks while the latter silently ignores them.</p>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.AugMix.html" class="btn btn-neutral float-right" title="AugMix" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandAugment.html" class="btn btn-neutral" title="RandAugment" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">TrivialAugmentWide</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html b/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
new file mode 100644
index 00000000000..ecdbf967a48
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
@@ -0,0 +1,706 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>UniformTemporalSubsample &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="normalize" href="torchvision.transforms.v2.functional.normalize.html" />
+    <link rel="prev" title="ClampBoundingBoxes" href="torchvision.transforms.v2.ClampBoundingBoxes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>UniformTemporalSubsample</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.UniformTemporalSubsample.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="uniformtemporalsubsample">
+<h1>UniformTemporalSubsample<a class="headerlink" href="#uniformtemporalsubsample" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.UniformTemporalSubsample">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.</span></span><span class="sig-name descname"><span class="pre">UniformTemporalSubsample</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/_temporal.html#UniformTemporalSubsample"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.UniformTemporalSubsample" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Uniformly subsample <code class="docutils literal notranslate"><span class="pre">num_samples</span></code> indices from the temporal dimension of the video.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The UniformTemporalSubsample transform is in Beta stage, and while we do not expect disruptive breaking changes, some APIs may slightly change according to user feedback. Please submit any feedback you may have in this issue: https://github.com/pytorch/vision/issues/6753.</p>
+</div>
+<p>Videos are expected to be of shape <code class="docutils literal notranslate"><span class="pre">[...,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code> where <code class="docutils literal notranslate"><span class="pre">T</span></code> denotes the temporal dimension.</p>
+<p>When <code class="docutils literal notranslate"><span class="pre">num_samples</span></code> is larger than the size of temporal dimension of the video, it
+will sample frames based on nearest neighbor interpolation.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>num_samples</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The number of equispaced samples to be selected</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.normalize.html" class="btn btn-neutral float-right" title="normalize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ClampBoundingBoxes.html" class="btn btn-neutral" title="ClampBoundingBoxes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">UniformTemporalSubsample</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
new file mode 100644
index 00000000000..7859ca63cc7
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_brightness &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_saturation" href="torchvision.transforms.v2.functional.adjust_saturation.html" />
+    <link rel="prev" title="equalize" href="torchvision.transforms.v2.functional.equalize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_brightness</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.adjust_brightness.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-brightness">
+<h1>adjust_brightness<a class="headerlink" href="#adjust-brightness" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.adjust_brightness">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_brightness</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">brightness_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#adjust_brightness"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.adjust_brightness" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust brightness.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.adjust_saturation.html" class="btn btn-neutral float-right" title="adjust_saturation" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.equalize.html" class="btn btn-neutral" title="equalize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_brightness</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
new file mode 100644
index 00000000000..a2cb45ac8a4
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_contrast &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="equalize" href="torchvision.transforms.v2.functional.equalize.html" />
+    <link rel="prev" title="autocontrast" href="torchvision.transforms.v2.functional.autocontrast.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_contrast</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.adjust_contrast.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-contrast">
+<h1>adjust_contrast<a class="headerlink" href="#adjust-contrast" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.adjust_contrast">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_contrast</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">contrast_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#adjust_contrast"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.adjust_contrast" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast" title="torchvision.transforms.RandomAutocontrast"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code></a></p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.equalize.html" class="btn btn-neutral float-right" title="equalize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.autocontrast.html" class="btn btn-neutral" title="autocontrast" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_contrast</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
new file mode 100644
index 00000000000..170ad700e7a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_gamma &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Compose" href="torchvision.transforms.v2.Compose.html" />
+    <link rel="prev" title="adjust_hue" href="torchvision.transforms.v2.functional.adjust_hue.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_gamma</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.adjust_gamma.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-gamma">
+<h1>adjust_gamma<a class="headerlink" href="#adjust-gamma" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.adjust_gamma">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_gamma</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">gamma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">gain</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#adjust_gamma"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.adjust_gamma" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust gamma.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.Compose.html" class="btn btn-neutral float-right" title="Compose" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.adjust_hue.html" class="btn btn-neutral" title="adjust_hue" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_gamma</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
new file mode 100644
index 00000000000..f234ebdf1fb
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_hue &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_gamma" href="torchvision.transforms.v2.functional.adjust_gamma.html" />
+    <link rel="prev" title="adjust_saturation" href="torchvision.transforms.v2.functional.adjust_saturation.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_hue</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.adjust_hue.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-hue">
+<h1>adjust_hue<a class="headerlink" href="#adjust-hue" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.adjust_hue">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_hue</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">hue_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#adjust_hue"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.adjust_hue" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust hue</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.adjust_gamma.html" class="btn btn-neutral float-right" title="adjust_gamma" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.adjust_saturation.html" class="btn btn-neutral" title="adjust_saturation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_hue</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
new file mode 100644
index 00000000000..44ab51d5d5d
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_saturation &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_hue" href="torchvision.transforms.v2.functional.adjust_hue.html" />
+    <link rel="prev" title="adjust_brightness" href="torchvision.transforms.v2.functional.adjust_brightness.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_saturation</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.adjust_saturation.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-saturation">
+<h1>adjust_saturation<a class="headerlink" href="#adjust-saturation" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.adjust_saturation">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_saturation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">saturation_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#adjust_saturation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.adjust_saturation" title="Permalink to this definition">¶</a></dt>
+<dd><p>Adjust saturation.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.adjust_hue.html" class="btn btn-neutral float-right" title="adjust_hue" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.adjust_brightness.html" class="btn btn-neutral" title="adjust_brightness" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_saturation</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
new file mode 100644
index 00000000000..586bfa373ad
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>adjust_sharpness &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="autocontrast" href="torchvision.transforms.v2.functional.autocontrast.html" />
+    <link rel="prev" title="solarize" href="torchvision.transforms.v2.functional.solarize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>adjust_sharpness</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.adjust_sharpness.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="adjust-sharpness">
+<h1>adjust_sharpness<a class="headerlink" href="#adjust-sharpness" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.adjust_sharpness">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">adjust_sharpness</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">sharpness_factor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#adjust_sharpness"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.adjust_sharpness" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness" title="torchvision.transforms.RandomAdjustSharpness"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAdjustSharpness</span></code></a></p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.autocontrast.html" class="btn btn-neutral float-right" title="autocontrast" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.solarize.html" class="btn btn-neutral" title="solarize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">adjust_sharpness</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.affine.html b/0.16/generated/torchvision.transforms.v2.functional.affine.html
new file mode 100644
index 00000000000..3b6b663cbfc
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.affine.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>affine &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="perspective" href="torchvision.transforms.v2.functional.perspective.html" />
+    <link rel="prev" title="rotate" href="torchvision.transforms.v2.functional.rotate.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>affine</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.affine.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="affine">
+<h1>affine<a class="headerlink" href="#affine" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.affine">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">affine</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">angle</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">translate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">shear</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#affine"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.affine" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomAffine.html#torchvision.transforms.v2.RandomAffine" title="torchvision.transforms.v2.RandomAffine"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAffine</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.perspective.html" class="btn btn-neutral float-right" title="perspective" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.rotate.html" class="btn btn-neutral" title="rotate" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">affine</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html b/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
new file mode 100644
index 00000000000..990ee955795
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>autocontrast &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_contrast" href="torchvision.transforms.v2.functional.adjust_contrast.html" />
+    <link rel="prev" title="adjust_sharpness" href="torchvision.transforms.v2.functional.adjust_sharpness.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>autocontrast</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.autocontrast.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="autocontrast">
+<h1>autocontrast<a class="headerlink" href="#autocontrast" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.autocontrast">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">autocontrast</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#autocontrast"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.autocontrast" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomAutocontrast.html#torchvision.transforms.v2.RandomAutocontrast" title="torchvision.transforms.v2.RandomAutocontrast"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.adjust_contrast.html" class="btn btn-neutral float-right" title="adjust_contrast" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.adjust_sharpness.html" class="btn btn-neutral" title="adjust_sharpness" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">autocontrast</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.center_crop.html b/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
new file mode 100644
index 00000000000..b2eefc80731
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>center_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="five_crop" href="torchvision.transforms.v2.functional.five_crop.html" />
+    <link rel="prev" title="ten_crop" href="torchvision.transforms.v2.functional.ten_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>center_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.center_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="center-crop">
+<h1>center_crop<a class="headerlink" href="#center-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.center_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">center_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#center_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.center_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop" title="torchvision.transforms.v2.RandomCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomCrop</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.five_crop.html" class="btn btn-neutral float-right" title="five_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.ten_crop.html" class="btn btn-neutral" title="ten_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">center_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html b/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
new file mode 100644
index 00000000000..b42462901c4
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>clamp_bounding_boxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="uniform_temporal_subsample" href="torchvision.transforms.v2.functional.uniform_temporal_subsample.html" />
+    <link rel="prev" title="erase" href="torchvision.transforms.v2.functional.erase.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>clamp_bounding_boxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="clamp-bounding-boxes">
+<h1>clamp_bounding_boxes<a class="headerlink" href="#clamp-bounding-boxes" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.clamp_bounding_boxes">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">clamp_bounding_boxes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors._bounding_boxes.BoundingBoxFormat"><span class="pre">BoundingBoxFormat</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">canvas_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_meta.html#clamp_bounding_boxes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.clamp_bounding_boxes" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.ClampBoundingBoxes.html#torchvision.transforms.v2.ClampBoundingBoxes" title="torchvision.transforms.v2.ClampBoundingBoxes"><code class="xref py py-func docutils literal notranslate"><span class="pre">ClampBoundingBoxes()</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.uniform_temporal_subsample.html" class="btn btn-neutral float-right" title="uniform_temporal_subsample" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.erase.html" class="btn btn-neutral" title="erase" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">clamp_bounding_boxes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html b/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
new file mode 100644
index 00000000000..189c66b1c64
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional.convert_bounding_box_format &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ToTensor" href="torchvision.transforms.v2.ToTensor.html" />
+    <link rel="prev" title="torchvision.transforms.v2.functional.to_dtype" href="torchvision.transforms.v2.functional.to_dtype.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional.convert_bounding_box_format</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="torchvision-transforms-v2-functional-convert-bounding-box-format">
+<h1>torchvision.transforms.v2.functional.convert_bounding_box_format<a class="headerlink" href="#torchvision-transforms-v2-functional-convert-bounding-box-format" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.convert_bounding_box_format">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">convert_bounding_box_format</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">old_format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors._bounding_boxes.BoundingBoxFormat"><span class="pre">BoundingBoxFormat</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">new_format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors._bounding_boxes.BoundingBoxFormat"><span class="pre">BoundingBoxFormat</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_meta.html#convert_bounding_box_format"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.convert_bounding_box_format" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.ConvertBoundingBoxFormat.html#torchvision.transforms.v2.ConvertBoundingBoxFormat" title="torchvision.transforms.v2.ConvertBoundingBoxFormat"><code class="xref py py-func docutils literal notranslate"><span class="pre">ConvertBoundingBoxFormat()</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ToTensor.html" class="btn btn-neutral float-right" title="ToTensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.to_dtype.html" class="btn btn-neutral" title="torchvision.transforms.v2.functional.to_dtype" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">torchvision.transforms.v2.functional.convert_bounding_box_format</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html b/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
new file mode 100644
index 00000000000..b6cd1a5c82a
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>convert_image_dtype &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="AutoAugment" href="torchvision.transforms.v2.AutoAugment.html" />
+    <link rel="prev" title="ConvertImageDtype" href="torchvision.transforms.v2.ConvertImageDtype.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>convert_image_dtype</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.convert_image_dtype.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convert-image-dtype">
+<h1>convert_image_dtype<a class="headerlink" href="#convert-image-dtype" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.convert_image_dtype">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">convert_image_dtype</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">torch.float32</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_misc.html#convert_image_dtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.convert_image_dtype" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] [DEPRECATED] Use to_dtype() instead.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.AutoAugment.html" class="btn btn-neutral float-right" title="AutoAugment" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ConvertImageDtype.html" class="btn btn-neutral" title="ConvertImageDtype" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">convert_image_dtype</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.crop.html b/0.16/generated/torchvision.transforms.v2.functional.crop.html
new file mode 100644
index 00000000000..2940942dbdc
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.crop.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resized_crop" href="torchvision.transforms.v2.functional.resized_crop.html" />
+    <link rel="prev" title="TenCrop" href="torchvision.transforms.v2.TenCrop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="crop">
+<h1>crop<a class="headerlink" href="#crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">top</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">left</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop" title="torchvision.transforms.v2.RandomCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomCrop</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.resized_crop.html" class="btn btn-neutral float-right" title="resized_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.TenCrop.html" class="btn btn-neutral" title="TenCrop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.elastic.html b/0.16/generated/torchvision.transforms.v2.functional.elastic.html
new file mode 100644
index 00000000000..b9f52f1ff56
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.elastic.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>elastic &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ColorJitter" href="torchvision.transforms.v2.ColorJitter.html" />
+    <link rel="prev" title="perspective" href="torchvision.transforms.v2.functional.perspective.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>elastic</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.elastic.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="elastic">
+<h1>elastic<a class="headerlink" href="#elastic" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.elastic">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">elastic</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">displacement</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#elastic"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.elastic" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.ElasticTransform.html#torchvision.transforms.v2.ElasticTransform" title="torchvision.transforms.v2.ElasticTransform"><code class="xref py py-class docutils literal notranslate"><span class="pre">ElasticTransform</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ColorJitter.html" class="btn btn-neutral float-right" title="ColorJitter" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.perspective.html" class="btn btn-neutral" title="perspective" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">elastic</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.equalize.html b/0.16/generated/torchvision.transforms.v2.functional.equalize.html
new file mode 100644
index 00000000000..41fec5b6988
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.equalize.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>equalize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_brightness" href="torchvision.transforms.v2.functional.adjust_brightness.html" />
+    <link rel="prev" title="adjust_contrast" href="torchvision.transforms.v2.functional.adjust_contrast.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>equalize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.equalize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="equalize">
+<h1>equalize<a class="headerlink" href="#equalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.equalize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">equalize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#equalize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.equalize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomEqualize.html#torchvision.transforms.v2.RandomEqualize" title="torchvision.transforms.v2.RandomEqualize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomEqualize</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.adjust_brightness.html" class="btn btn-neutral float-right" title="adjust_brightness" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.adjust_contrast.html" class="btn btn-neutral" title="adjust_contrast" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">equalize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.erase.html b/0.16/generated/torchvision.transforms.v2.functional.erase.html
new file mode 100644
index 00000000000..2f7b0e04709
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.erase.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>erase &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="clamp_bounding_boxes" href="torchvision.transforms.v2.functional.clamp_bounding_boxes.html" />
+    <link rel="prev" title="normalize" href="torchvision.transforms.v2.functional.normalize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>erase</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.erase.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="erase">
+<h1>erase<a class="headerlink" href="#erase" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.erase">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">erase</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">i</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">j</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">h</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">w</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">v</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_augment.html#erase"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.erase" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <code class="xref py py-class docutils literal notranslate"><span class="pre">RandomErase</span></code> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.clamp_bounding_boxes.html" class="btn btn-neutral float-right" title="clamp_bounding_boxes" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.normalize.html" class="btn btn-neutral" title="normalize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">erase</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.five_crop.html b/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
new file mode 100644
index 00000000000..697024aa2ae
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>five_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomHorizontalFlip" href="torchvision.transforms.v2.RandomHorizontalFlip.html" />
+    <link rel="prev" title="center_crop" href="torchvision.transforms.v2.functional.center_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>five_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.five_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="five-crop">
+<h1>five_crop<a class="headerlink" href="#five-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.five_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">five_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#five_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.five_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.FiveCrop.html#torchvision.transforms.v2.FiveCrop" title="torchvision.transforms.v2.FiveCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">FiveCrop</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomHorizontalFlip.html" class="btn btn-neutral float-right" title="RandomHorizontalFlip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.center_crop.html" class="btn btn-neutral" title="center_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">five_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html b/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
new file mode 100644
index 00000000000..ce2ba4b8b12
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>gaussian_blur &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="invert" href="torchvision.transforms.v2.functional.invert.html" />
+    <link rel="prev" title="to_grayscale" href="torchvision.transforms.v2.functional.to_grayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>gaussian_blur</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.gaussian_blur.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="gaussian-blur">
+<h1>gaussian_blur<a class="headerlink" href="#gaussian-blur" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.gaussian_blur">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">gaussian_blur</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">kernel_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sigma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_misc.html#gaussian_blur"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.gaussian_blur" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.GaussianBlur.html#torchvision.transforms.v2.GaussianBlur" title="torchvision.transforms.v2.GaussianBlur"><code class="xref py py-class docutils literal notranslate"><span class="pre">GaussianBlur</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.invert.html" class="btn btn-neutral float-right" title="invert" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.to_grayscale.html" class="btn btn-neutral" title="to_grayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">gaussian_blur</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html b/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
new file mode 100644
index 00000000000..01d6ff4a5ed
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>horizontal_flip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="vertical_flip" href="torchvision.transforms.v2.functional.vertical_flip.html" />
+    <link rel="prev" title="ElasticTransform" href="torchvision.transforms.v2.ElasticTransform.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>horizontal_flip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.horizontal_flip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="horizontal-flip">
+<h1>horizontal_flip<a class="headerlink" href="#horizontal-flip" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.horizontal_flip">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">horizontal_flip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#horizontal_flip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.horizontal_flip" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomHorizontalFlip.html#torchvision.transforms.v2.RandomHorizontalFlip" title="torchvision.transforms.v2.RandomHorizontalFlip"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomHorizontalFlip</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.vertical_flip.html" class="btn btn-neutral float-right" title="vertical_flip" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ElasticTransform.html" class="btn btn-neutral" title="ElasticTransform" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">horizontal_flip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.invert.html b/0.16/generated/torchvision.transforms.v2.functional.invert.html
new file mode 100644
index 00000000000..932eaba8d19
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.invert.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>invert &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="posterize" href="torchvision.transforms.v2.functional.posterize.html" />
+    <link rel="prev" title="gaussian_blur" href="torchvision.transforms.v2.functional.gaussian_blur.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>invert</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.invert.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="invert">
+<h1>invert<a class="headerlink" href="#invert" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.invert">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">invert</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#invert"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.invert" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomInvert.html#torchvision.transforms.v2.RandomInvert" title="torchvision.transforms.v2.RandomInvert"><code class="xref py py-func docutils literal notranslate"><span class="pre">RandomInvert()</span></code></a>.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.posterize.html" class="btn btn-neutral float-right" title="posterize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.gaussian_blur.html" class="btn btn-neutral" title="gaussian_blur" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">invert</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.normalize.html b/0.16/generated/torchvision.transforms.v2.functional.normalize.html
new file mode 100644
index 00000000000..f14bab49ef6
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.normalize.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>normalize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="erase" href="torchvision.transforms.v2.functional.erase.html" />
+    <link rel="prev" title="UniformTemporalSubsample" href="torchvision.transforms.v2.UniformTemporalSubsample.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>normalize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.normalize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="normalize">
+<h1>normalize<a class="headerlink" href="#normalize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.normalize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">normalize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">mean</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inplace</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_misc.html#normalize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.normalize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.Normalize.html#torchvision.transforms.v2.Normalize" title="torchvision.transforms.v2.Normalize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Normalize</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.erase.html" class="btn btn-neutral float-right" title="erase" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.UniformTemporalSubsample.html" class="btn btn-neutral" title="UniformTemporalSubsample" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">normalize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.pad.html b/0.16/generated/torchvision.transforms.v2.functional.pad.html
new file mode 100644
index 00000000000..cab9a1996d7
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.pad.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>pad &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="rotate" href="torchvision.transforms.v2.functional.rotate.html" />
+    <link rel="prev" title="vertical_flip" href="torchvision.transforms.v2.functional.vertical_flip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>pad</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.pad.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="pad">
+<h1>pad<a class="headerlink" href="#pad" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.pad">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">pad</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'constant'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#pad"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.pad" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.Pad.html#torchvision.transforms.v2.Pad" title="torchvision.transforms.v2.Pad"><code class="xref py py-class docutils literal notranslate"><span class="pre">Pad</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.rotate.html" class="btn btn-neutral float-right" title="rotate" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.vertical_flip.html" class="btn btn-neutral" title="vertical_flip" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">pad</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html b/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
new file mode 100644
index 00000000000..2e9ac522184
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
@@ -0,0 +1,717 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>permute_channels &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="rgb_to_grayscale" href="torchvision.transforms.v2.functional.rgb_to_grayscale.html" />
+    <link rel="prev" title="RandomEqualize" href="torchvision.transforms.v2.RandomEqualize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>permute_channels</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.permute_channels.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="permute-channels">
+<h1>permute_channels<a class="headerlink" href="#permute-channels" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.permute_channels">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">permute_channels</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">permutation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#permute_channels"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.permute_channels" title="Permalink to this definition">¶</a></dt>
+<dd><p>Permute the channels of the input according to the given permutation.</p>
+<p>This function supports plain <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>’s, <a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">PIL.Image.Image</span></code></a>’s, and
+<a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.tv_tensors.Image</span></code></a> and <a class="reference internal" href="torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-class docutils literal notranslate"><span class="pre">torchvision.tv_tensors.Video</span></code></a>.</p>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">rgb_image</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="mi">256</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">bgr_image</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">permutate_channels</span><span class="p">(</span><span class="n">rgb_image</span><span class="p">,</span> <span class="n">permutation</span><span class="o">=</span><span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">])</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>permutation</strong> (<em>List</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em>) – <p>Valid permutation of the input channel indices. The index of the element determines the
+channel index in the input and the value determines the channel index in the output. For example,
+<code class="docutils literal notranslate"><span class="pre">permutation=[2,</span> <span class="pre">0</span> <span class="pre">,</span> <span class="pre">1]</span></code></p>
+<ul class="simple">
+<li><p>takes <code class="docutils literal notranslate"><span class="pre">ìnpt[...,</span> <span class="pre">0,</span> <span class="pre">:,</span> <span class="pre">:]</span></code> and puts it at <code class="docutils literal notranslate"><span class="pre">output[...,</span> <span class="pre">2,</span> <span class="pre">:,</span> <span class="pre">:]</span></code>,</p></li>
+<li><p>takes <code class="docutils literal notranslate"><span class="pre">ìnpt[...,</span> <span class="pre">1,</span> <span class="pre">:,</span> <span class="pre">:]</span></code> and puts it at <code class="docutils literal notranslate"><span class="pre">output[...,</span> <span class="pre">0,</span> <span class="pre">:,</span> <span class="pre">:]</span></code>, and</p></li>
+<li><p>takes <code class="docutils literal notranslate"><span class="pre">ìnpt[...,</span> <span class="pre">2,</span> <span class="pre">:,</span> <span class="pre">:]</span></code> and puts it at <code class="docutils literal notranslate"><span class="pre">output[...,</span> <span class="pre">1,</span> <span class="pre">:,</span> <span class="pre">:]</span></code>.</p></li>
+</ul>
+</p>
+</dd>
+<dt class="field-even">Raises<span class="colon">:</span></dt>
+<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/exceptions.html#ValueError" title="(in Python v3.12)"><strong>ValueError</strong></a> – If <code class="docutils literal notranslate"><span class="pre">len(permutation)</span></code> doesn’t match the number of channels in the input.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.rgb_to_grayscale.html" class="btn btn-neutral float-right" title="rgb_to_grayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomEqualize.html" class="btn btn-neutral" title="RandomEqualize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">permute_channels</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.perspective.html b/0.16/generated/torchvision.transforms.v2.functional.perspective.html
new file mode 100644
index 00000000000..1d533b775df
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.perspective.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>perspective &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="elastic" href="torchvision.transforms.v2.functional.elastic.html" />
+    <link rel="prev" title="affine" href="torchvision.transforms.v2.functional.affine.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>perspective</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.perspective.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="perspective">
+<h1>perspective<a class="headerlink" href="#perspective" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.perspective">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">perspective</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">startpoints</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">endpoints</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">coefficients</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#perspective"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.perspective" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomPerspective.html#torchvision.transforms.v2.RandomPerspective" title="torchvision.transforms.v2.RandomPerspective"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomPerspective</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.elastic.html" class="btn btn-neutral float-right" title="elastic" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.affine.html" class="btn btn-neutral" title="affine" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">perspective</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html b/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
new file mode 100644
index 00000000000..05e05aafad1
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional.pil_to_tensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="torchvision.transforms.v2.functional.to_pil_image" href="torchvision.transforms.v2.functional.to_pil_image.html" />
+    <link rel="prev" title="torchvision.transforms.v2.functional.to_image" href="torchvision.transforms.v2.functional.to_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional.pil_to_tensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.pil_to_tensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="torchvision-transforms-v2-functional-pil-to-tensor">
+<h1>torchvision.transforms.v2.functional.pil_to_tensor<a class="headerlink" href="#torchvision-transforms-v2-functional-pil-to-tensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.pil_to_tensor">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">pil_to_tensor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pic</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#pil_to_tensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.pil_to_tensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a <code class="docutils literal notranslate"><span class="pre">PIL</span> <span class="pre">Image</span></code> to a tensor of the same type.
+This function does not support torchscript.</p>
+<p>See <a class="reference internal" href="torchvision.transforms.PILToTensor.html#torchvision.transforms.PILToTensor" title="torchvision.transforms.PILToTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">PILToTensor</span></code></a> for more details.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>A deep copy of the underlying array is performed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>pic</strong> (<em>PIL Image</em>) – Image to be converted to tensor.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Converted image.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Tensor</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.to_pil_image.html" class="btn btn-neutral float-right" title="torchvision.transforms.v2.functional.to_pil_image" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.to_image.html" class="btn btn-neutral" title="torchvision.transforms.v2.functional.to_image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">torchvision.transforms.v2.functional.pil_to_tensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.posterize.html b/0.16/generated/torchvision.transforms.v2.functional.posterize.html
new file mode 100644
index 00000000000..2243f1a40bd
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.posterize.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>posterize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="solarize" href="torchvision.transforms.v2.functional.solarize.html" />
+    <link rel="prev" title="invert" href="torchvision.transforms.v2.functional.invert.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>posterize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.posterize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="posterize">
+<h1>posterize<a class="headerlink" href="#posterize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.posterize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">posterize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">bits</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#posterize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.posterize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomPosterize.html#torchvision.transforms.v2.RandomPosterize" title="torchvision.transforms.v2.RandomPosterize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomPosterize</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.solarize.html" class="btn btn-neutral float-right" title="solarize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.invert.html" class="btn btn-neutral" title="invert" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">posterize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html b/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
new file mode 100644
index 00000000000..85871e7d60e
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
@@ -0,0 +1,700 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>register_kernel &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Resize" href="torchvision.transforms.Resize.html" />
+    <link rel="prev" title="MixUp" href="torchvision.transforms.v2.MixUp.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>register_kernel</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.register_kernel.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="register-kernel">
+<h1>register_kernel<a class="headerlink" href="#register-kernel" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.register_kernel">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">register_kernel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">functional</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tv_tensor_cls</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_utils.html#register_kernel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.register_kernel" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Decorate a kernel to register it for a functional and a (custom) tv_tensor type.</p>
+<p>See <a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a> for usage
+details.</p>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">register_kernel</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.Resize.html" class="btn btn-neutral float-right" title="Resize" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.MixUp.html" class="btn btn-neutral" title="MixUp" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">register_kernel</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.resize.html b/0.16/generated/torchvision.transforms.v2.functional.resize.html
new file mode 100644
index 00000000000..29f03b5cc46
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.resize.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RandomCrop" href="torchvision.transforms.v2.RandomCrop.html" />
+    <link rel="prev" title="RandomResize" href="torchvision.transforms.v2.RandomResize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>resize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.resize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resize">
+<h1>resize<a class="headerlink" href="#resize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.resize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">resize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#resize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.resize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.Resize.html#torchvision.transforms.v2.Resize" title="torchvision.transforms.v2.Resize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Resize</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.RandomCrop.html" class="btn btn-neutral float-right" title="RandomCrop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.RandomResize.html" class="btn btn-neutral" title="RandomResize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html b/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
new file mode 100644
index 00000000000..cea28709089
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resized_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ten_crop" href="torchvision.transforms.v2.functional.ten_crop.html" />
+    <link rel="prev" title="crop" href="torchvision.transforms.v2.functional.crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>resized_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.resized_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resized-crop">
+<h1>resized_crop<a class="headerlink" href="#resized-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.resized_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">resized_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">top</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">left</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">height</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.BILINEAR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">antialias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'warn'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#resized_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.resized_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.v2.RandomResizedCrop" title="torchvision.transforms.v2.RandomResizedCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.ten_crop.html" class="btn btn-neutral float-right" title="ten_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.crop.html" class="btn btn-neutral" title="crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resized_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html b/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
new file mode 100644
index 00000000000..cef77c76cbd
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>rgb_to_grayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="to_grayscale" href="torchvision.transforms.v2.functional.to_grayscale.html" />
+    <link rel="prev" title="permute_channels" href="torchvision.transforms.v2.functional.permute_channels.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>rgb_to_grayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="rgb-to-grayscale">
+<h1>rgb_to_grayscale<a class="headerlink" href="#rgb-to-grayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.rgb_to_grayscale">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">rgb_to_grayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_output_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#rgb_to_grayscale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.rgb_to_grayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.Grayscale.html#torchvision.transforms.v2.Grayscale" title="torchvision.transforms.v2.Grayscale"><code class="xref py py-class docutils literal notranslate"><span class="pre">Grayscale</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.to_grayscale.html" class="btn btn-neutral float-right" title="to_grayscale" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.permute_channels.html" class="btn btn-neutral" title="permute_channels" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">rgb_to_grayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.rotate.html b/0.16/generated/torchvision.transforms.v2.functional.rotate.html
new file mode 100644
index 00000000000..b5c02f231c0
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.rotate.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>rotate &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="affine" href="torchvision.transforms.v2.functional.affine.html" />
+    <link rel="prev" title="pad" href="torchvision.transforms.v2.functional.pad.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>rotate</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.rotate.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="rotate">
+<h1>rotate<a class="headerlink" href="#rotate" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.rotate">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">rotate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">angle</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">interpolation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">InterpolationMode</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">InterpolationMode.NEAREST</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">expand</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">center</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#rotate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.rotate" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomRotation.html#torchvision.transforms.v2.RandomRotation" title="torchvision.transforms.v2.RandomRotation"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomRotation</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.affine.html" class="btn btn-neutral float-right" title="affine" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.pad.html" class="btn btn-neutral" title="pad" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">rotate</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.solarize.html b/0.16/generated/torchvision.transforms.v2.functional.solarize.html
new file mode 100644
index 00000000000..2fd0c34a963
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.solarize.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>solarize &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="adjust_sharpness" href="torchvision.transforms.v2.functional.adjust_sharpness.html" />
+    <link rel="prev" title="posterize" href="torchvision.transforms.v2.functional.posterize.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>solarize</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.solarize.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="solarize">
+<h1>solarize<a class="headerlink" href="#solarize" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.solarize">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">solarize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">threshold</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_color.html#solarize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.solarize" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomSolarize.html#torchvision.transforms.v2.RandomSolarize" title="torchvision.transforms.v2.RandomSolarize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomSolarize</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.adjust_sharpness.html" class="btn btn-neutral float-right" title="adjust_sharpness" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.posterize.html" class="btn btn-neutral" title="posterize" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">solarize</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html b/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
new file mode 100644
index 00000000000..dcab89504de
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ten_crop &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="center_crop" href="torchvision.transforms.v2.functional.center_crop.html" />
+    <link rel="prev" title="resized_crop" href="torchvision.transforms.v2.functional.resized_crop.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>ten_crop</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.ten_crop.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ten-crop">
+<h1>ten_crop<a class="headerlink" href="#ten-crop" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.ten_crop">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">ten_crop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">vertical_flip</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#ten_crop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.ten_crop" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.TenCrop.html#torchvision.transforms.v2.TenCrop" title="torchvision.transforms.v2.TenCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">TenCrop</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.center_crop.html" class="btn btn-neutral float-right" title="center_crop" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.resized_crop.html" class="btn btn-neutral" title="resized_crop" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ten_crop</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html b/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
new file mode 100644
index 00000000000..54e59455168
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional.to_dtype &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="torchvision.transforms.v2.functional.convert_bounding_box_format" href="torchvision.transforms.v2.functional.convert_bounding_box_format.html" />
+    <link rel="prev" title="torchvision.transforms.v2.functional.to_pil_image" href="torchvision.transforms.v2.functional.to_pil_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional.to_dtype</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.to_dtype.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="torchvision-transforms-v2-functional-to-dtype">
+<h1>torchvision.transforms.v2.functional.to_dtype<a class="headerlink" href="#torchvision-transforms-v2-functional-to-dtype" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.to_dtype">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">to_dtype</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">torch.float32</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_misc.html#to_dtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.to_dtype" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.ToDtype.html#torchvision.transforms.v2.ToDtype" title="torchvision.transforms.v2.ToDtype"><code class="xref py py-func docutils literal notranslate"><span class="pre">ToDtype()</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.convert_bounding_box_format.html" class="btn btn-neutral float-right" title="torchvision.transforms.v2.functional.convert_bounding_box_format" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.to_pil_image.html" class="btn btn-neutral" title="torchvision.transforms.v2.functional.to_pil_image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">torchvision.transforms.v2.functional.to_dtype</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html b/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
new file mode 100644
index 00000000000..81f198bda7d
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>to_grayscale &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="gaussian_blur" href="torchvision.transforms.v2.functional.gaussian_blur.html" />
+    <link rel="prev" title="rgb_to_grayscale" href="torchvision.transforms.v2.functional.rgb_to_grayscale.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>to_grayscale</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.to_grayscale.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="to-grayscale">
+<h1>to_grayscale<a class="headerlink" href="#to-grayscale" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.to_grayscale">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">to_grayscale</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_output_channels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="headerlink" href="#torchvision.transforms.v2.functional.to_grayscale" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.Grayscale.html#torchvision.transforms.v2.Grayscale" title="torchvision.transforms.v2.Grayscale"><code class="xref py py-class docutils literal notranslate"><span class="pre">Grayscale</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.gaussian_blur.html" class="btn btn-neutral float-right" title="gaussian_blur" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.rgb_to_grayscale.html" class="btn btn-neutral" title="rgb_to_grayscale" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">to_grayscale</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_image.html b/0.16/generated/torchvision.transforms.v2.functional.to_image.html
new file mode 100644
index 00000000000..217985c24ea
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_image.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional.to_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="torchvision.transforms.v2.functional.pil_to_tensor" href="torchvision.transforms.v2.functional.pil_to_tensor.html" />
+    <link rel="prev" title="ConvertBoundingBoxFormat" href="torchvision.transforms.v2.ConvertBoundingBoxFormat.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional.to_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.to_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="torchvision-transforms-v2-functional-to-image">
+<h1>torchvision.transforms.v2.functional.to_image<a class="headerlink" href="#torchvision-transforms-v2-functional-to-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.to_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">to_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><span class="pre">Image</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><span class="pre">ndarray</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors._image.Image"><span class="pre">Image</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_type_conversion.html#to_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.to_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.ToImage.html#torchvision.transforms.v2.ToImage" title="torchvision.transforms.v2.ToImage"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToImage</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.pil_to_tensor.html" class="btn btn-neutral float-right" title="torchvision.transforms.v2.functional.pil_to_tensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ConvertBoundingBoxFormat.html" class="btn btn-neutral" title="ConvertBoundingBoxFormat" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">torchvision.transforms.v2.functional.to_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html b/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
new file mode 100644
index 00000000000..369f49d57ef
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision.transforms.v2.functional.to_pil_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="torchvision.transforms.v2.functional.to_dtype" href="torchvision.transforms.v2.functional.to_dtype.html" />
+    <link rel="prev" title="torchvision.transforms.v2.functional.pil_to_tensor" href="torchvision.transforms.v2.functional.pil_to_tensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>torchvision.transforms.v2.functional.to_pil_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.to_pil_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="torchvision-transforms-v2-functional-to-pil-image">
+<h1>torchvision.transforms.v2.functional.to_pil_image<a class="headerlink" href="#torchvision-transforms-v2-functional-to-pil-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.to_pil_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">to_pil_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pic</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mode</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/functional.html#to_pil_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.to_pil_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>Convert a tensor or an ndarray to PIL Image. This function does not support torchscript.</p>
+<p>See <a class="reference internal" href="torchvision.transforms.ToPILImage.html#torchvision.transforms.ToPILImage" title="torchvision.transforms.ToPILImage"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToPILImage</span></code></a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>pic</strong> (<em>Tensor</em><em> or </em><a class="reference external" href="https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html#numpy.ndarray" title="(in NumPy v1.26)"><em>numpy.ndarray</em></a>) – Image to be converted to PIL Image.</p></li>
+<li><p><strong>mode</strong> (<a class="reference external" href="https://pillow.readthedocs.io/en/latest/handbook/concepts.html#concept-modes">PIL.Image mode</a>) – color space and pixel depth of input data (optional).</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Image converted to PIL Image.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p>PIL Image</p>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.to_dtype.html" class="btn btn-neutral float-right" title="torchvision.transforms.v2.functional.to_dtype" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.pil_to_tensor.html" class="btn btn-neutral" title="torchvision.transforms.v2.functional.pil_to_tensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">torchvision.transforms.v2.functional.to_pil_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html b/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
new file mode 100644
index 00000000000..6a2705e0309
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>to_tensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ConvertImageDtype" href="torchvision.transforms.v2.ConvertImageDtype.html" />
+    <link rel="prev" title="ToTensor" href="torchvision.transforms.v2.ToTensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>to_tensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.to_tensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="to-tensor">
+<h1>to_tensor<a class="headerlink" href="#to-tensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.to_tensor">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">to_tensor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_deprecated.html#to_tensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.to_tensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] [DEPREACTED] Use to_image() and to_dtype() instead.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ConvertImageDtype.html" class="btn btn-neutral float-right" title="ConvertImageDtype" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.ToTensor.html" class="btn btn-neutral" title="ToTensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">to_tensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html b/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
new file mode 100644
index 00000000000..17ffddc5938
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>uniform_temporal_subsample &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ToImage" href="torchvision.transforms.v2.ToImage.html" />
+    <link rel="prev" title="clamp_bounding_boxes" href="torchvision.transforms.v2.functional.clamp_bounding_boxes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>uniform_temporal_subsample</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="uniform-temporal-subsample">
+<h1>uniform_temporal_subsample<a class="headerlink" href="#uniform-temporal-subsample" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.uniform_temporal_subsample">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">uniform_temporal_subsample</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_temporal.html#uniform_temporal_subsample"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.uniform_temporal_subsample" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.UniformTemporalSubsample.html#torchvision.transforms.v2.UniformTemporalSubsample" title="torchvision.transforms.v2.UniformTemporalSubsample"><code class="xref py py-class docutils literal notranslate"><span class="pre">UniformTemporalSubsample</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.ToImage.html" class="btn btn-neutral float-right" title="ToImage" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.clamp_bounding_boxes.html" class="btn btn-neutral" title="clamp_bounding_boxes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">uniform_temporal_subsample</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html b/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
new file mode 100644
index 00000000000..d170a1cfb16
--- /dev/null
+++ b/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
@@ -0,0 +1,694 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vertical_flip &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="pad" href="torchvision.transforms.v2.functional.pad.html" />
+    <link rel="prev" title="horizontal_flip" href="torchvision.transforms.v2.functional.horizontal_flip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../transforms.html">Transforming and augmenting images</a> &gt;</li>
+        
+      <li>vertical_flip</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.transforms.v2.functional.vertical_flip.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vertical-flip">
+<h1>vertical_flip<a class="headerlink" href="#vertical-flip" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.transforms.v2.functional.vertical_flip">
+<span class="sig-prename descclassname"><span class="pre">torchvision.transforms.v2.functional.</span></span><span class="sig-name descname"><span class="pre">vertical_flip</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inpt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/transforms/v2/functional/_geometry.html#vertical_flip"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.transforms.v2.functional.vertical_flip" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] See <a class="reference internal" href="torchvision.transforms.v2.RandomVerticalFlip.html#torchvision.transforms.v2.RandomVerticalFlip" title="torchvision.transforms.v2.RandomVerticalFlip"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomVerticalFlip</span></code></a> for details.</p>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.transforms.v2.functional.pad.html" class="btn btn-neutral float-right" title="pad" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.transforms.v2.functional.horizontal_flip.html" class="btn btn-neutral" title="horizontal_flip" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vertical_flip</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html b/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
new file mode 100644
index 00000000000..efb56f0bc27
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
@@ -0,0 +1,704 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>BoundingBoxFormat &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="BoundingBoxes" href="torchvision.tv_tensors.BoundingBoxes.html" />
+    <link rel="prev" title="Video" href="torchvision.tv_tensors.Video.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>BoundingBoxFormat</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.BoundingBoxFormat.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="boundingboxformat">
+<h1>BoundingBoxFormat<a class="headerlink" href="#boundingboxformat" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.BoundingBoxFormat">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">BoundingBoxFormat</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_bounding_boxes.html#BoundingBoxFormat"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.BoundingBoxFormat" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Coordinate format of a bounding box.</p>
+<p>Available formats are</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">XYXY</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">XYWH</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">CXCYWH</span></code></p></li>
+</ul>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">BoundingBoxFormat</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.tv_tensors.BoundingBoxes.html" class="btn btn-neutral float-right" title="BoundingBoxes" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.tv_tensors.Video.html" class="btn btn-neutral" title="Video" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">BoundingBoxFormat</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html b/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
new file mode 100644
index 00000000000..7992cdd5052
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
@@ -0,0 +1,729 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>BoundingBoxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Mask" href="torchvision.tv_tensors.Mask.html" />
+    <link rel="prev" title="BoundingBoxFormat" href="torchvision.tv_tensors.BoundingBoxFormat.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>BoundingBoxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.BoundingBoxes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="boundingboxes">
+<h1>BoundingBoxes<a class="headerlink" href="#boundingboxes" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.BoundingBoxes">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">BoundingBoxes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors._bounding_boxes.BoundingBoxFormat"><span class="pre">BoundingBoxFormat</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">canvas_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><span class="pre">device</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">requires_grad</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_bounding_boxes.html#BoundingBoxes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.BoundingBoxes" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for bounding boxes.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>There should be only one <a class="reference internal" href="#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a>
+instance per sample e.g. <code class="docutils literal notranslate"><span class="pre">{&quot;img&quot;:</span> <span class="pre">img,</span> <span class="pre">&quot;bbox&quot;:</span> <span class="pre">BoundingBoxes(...)}</span></code>,
+although one <a class="reference internal" href="#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a> object can
+contain multiple bounding boxes.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>data</strong> – Any data that can be turned into a tensor with <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.as_tensor.html#torch.as_tensor" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.as_tensor()</span></code></a>.</p></li>
+<li><p><strong>format</strong> (<a class="reference internal" href="torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors.BoundingBoxFormat"><em>BoundingBoxFormat</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Format of the bounding box.</p></li>
+<li><p><strong>canvas_size</strong> (<em>two-tuple of python:ints</em>) – Height and width of the corresponding image or video.</p></li>
+<li><p><strong>dtype</strong> (<em>torch.dpython:type</em><em>, </em><em>optional</em>) – Desired data type of the bounding box. If omitted, will be inferred from
+<code class="docutils literal notranslate"><span class="pre">data</span></code>.</p></li>
+<li><p><strong>device</strong> (<a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><em>torch.device</em></a><em>, </em><em>optional</em>) – Desired device of the bounding box. If omitted and <code class="docutils literal notranslate"><span class="pre">data</span></code> is a
+<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the device is taken from it. Otherwise, the bounding box is constructed on the CPU.</p></li>
+<li><p><strong>requires_grad</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Whether autograd should record operations on the bounding box. If omitted and
+<code class="docutils literal notranslate"><span class="pre">data</span></code> is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the value is taken from it. Otherwise, defaults to <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">BoundingBoxes</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.tv_tensors.Mask.html" class="btn btn-neutral float-right" title="Mask" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.tv_tensors.BoundingBoxFormat.html" class="btn btn-neutral" title="BoundingBoxFormat" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">BoundingBoxes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.Image.html b/0.16/generated/torchvision.tv_tensors.Image.html
new file mode 100644
index 00000000000..279dbef4116
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.Image.html
@@ -0,0 +1,724 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Video" href="torchvision.tv_tensors.Video.html" />
+    <link rel="prev" title="TVTensors" href="../tv_tensors.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>Image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.Image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="image">
+<h1>Image<a class="headerlink" href="#image" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.Image">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">Image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><span class="pre">device</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">requires_grad</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_image.html#Image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.Image" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for images.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>In the <a class="reference internal" href="../transforms.html#transforms"><span class="std std-ref">transforms</span></a>, <code class="docutils literal notranslate"><span class="pre">Image</span></code> instances are largely
+interchangeable with pure <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>. See
+<a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#passthrough-heuristic"><span class="std std-ref">this note</span></a> for more details.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>data</strong> (<em>tensor-like</em><em>, </em><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><em>PIL.Image.Image</em></a>) – Any data that can be turned into a tensor with <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.as_tensor.html#torch.as_tensor" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.as_tensor()</span></code></a> as
+well as PIL images.</p></li>
+<li><p><strong>dtype</strong> (<em>torch.dpython:type</em><em>, </em><em>optional</em>) – Desired data type. If omitted, will be inferred from
+<code class="docutils literal notranslate"><span class="pre">data</span></code>.</p></li>
+<li><p><strong>device</strong> (<a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><em>torch.device</em></a><em>, </em><em>optional</em>) – Desired device. If omitted and <code class="docutils literal notranslate"><span class="pre">data</span></code> is a
+<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the device is taken from it. Otherwise, the image is constructed on the CPU.</p></li>
+<li><p><strong>requires_grad</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Whether autograd should record operations. If omitted and
+<code class="docutils literal notranslate"><span class="pre">data</span></code> is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the value is taken from it. Otherwise, defaults to <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Image</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.tv_tensors.Video.html" class="btn btn-neutral float-right" title="Video" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../tv_tensors.html" class="btn btn-neutral" title="TVTensors" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.Mask.html b/0.16/generated/torchvision.tv_tensors.Mask.html
new file mode 100644
index 00000000000..b13e11e12d4
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.Mask.html
@@ -0,0 +1,715 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Mask &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="TVTensor" href="torchvision.tv_tensors.TVTensor.html" />
+    <link rel="prev" title="BoundingBoxes" href="torchvision.tv_tensors.BoundingBoxes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>Mask</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.Mask.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mask">
+<h1>Mask<a class="headerlink" href="#mask" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.Mask">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">Mask</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><span class="pre">device</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">requires_grad</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_mask.html#Mask"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.Mask" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for segmentation and detection masks.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>data</strong> (<em>tensor-like</em><em>, </em><a class="reference external" href="https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image" title="(in Pillow (PIL Fork) v10.0.1)"><em>PIL.Image.Image</em></a>) – Any data that can be turned into a tensor with <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.as_tensor.html#torch.as_tensor" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.as_tensor()</span></code></a> as
+well as PIL images.</p></li>
+<li><p><strong>dtype</strong> (<em>torch.dpython:type</em><em>, </em><em>optional</em>) – Desired data type. If omitted, will be inferred from
+<code class="docutils literal notranslate"><span class="pre">data</span></code>.</p></li>
+<li><p><strong>device</strong> (<a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><em>torch.device</em></a><em>, </em><em>optional</em>) – Desired device. If omitted and <code class="docutils literal notranslate"><span class="pre">data</span></code> is a
+<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the device is taken from it. Otherwise, the mask is constructed on the CPU.</p></li>
+<li><p><strong>requires_grad</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Whether autograd should record operations. If omitted and
+<code class="docutils literal notranslate"><span class="pre">data</span></code> is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the value is taken from it. Otherwise, defaults to <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Mask</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.tv_tensors.TVTensor.html" class="btn btn-neutral float-right" title="TVTensor" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.tv_tensors.BoundingBoxes.html" class="btn btn-neutral" title="BoundingBoxes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Mask</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.TVTensor.html b/0.16/generated/torchvision.tv_tensors.TVTensor.html
new file mode 100644
index 00000000000..f465e89d2f1
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.TVTensor.html
@@ -0,0 +1,740 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>TVTensor &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="set_return_type" href="torchvision.tv_tensors.set_return_type.html" />
+    <link rel="prev" title="Mask" href="torchvision.tv_tensors.Mask.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>TVTensor</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.TVTensor.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="tvtensor">
+<h1>TVTensor<a class="headerlink" href="#tvtensor" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.TVTensor">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">TVTensor</span></span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_tv_tensor.html#TVTensor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.TVTensor" title="Permalink to this definition">¶</a></dt>
+<dd><p>[Beta] Base class for all TVTensors.</p>
+<p>You probably don’t want to use this class unless you’re defining your own
+custom TVTensors. See
+<a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a> for details.</p>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">TVTensor</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="Object detection and segmentation tasks are natively supported: torchvision.transforms.v2 enabl..."><img alt="" src="../_images/sphx_glr_plot_transforms_e2e_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Transforms v2: End-to-end object detection/segmentation example</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide explains how to write transforms that are compatible with the torchvision transforms..."><img alt="" src="../_images/sphx_glr_plot_custom_transforms_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own v2 transforms</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div></div><dl class="py property">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.TVTensor.device">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">device</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><span class="pre">device</span></a></em><a class="headerlink" href="#torchvision.tv_tensors.TVTensor.device" title="Permalink to this definition">¶</a></dt>
+<dd><p>Is the <a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.device</span></code></a> where this Tensor is.</p>
+</dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.TVTensor.ndim">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">ndim</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></em><a class="headerlink" href="#torchvision.tv_tensors.TVTensor.ndim" title="Permalink to this definition">¶</a></dt>
+<dd><p>Alias for <code class="xref py py-meth docutils literal notranslate"><span class="pre">dim()</span></code></p>
+</dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.TVTensor.shape">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">shape</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><span class="pre">Size</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></em><a class="headerlink" href="#torchvision.tv_tensors.TVTensor.shape" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns the size of the <code class="xref py py-attr docutils literal notranslate"><span class="pre">self</span></code> tensor. Alias for <code class="xref py py-attr docutils literal notranslate"><span class="pre">size</span></code>.</p>
+<p>See also <code class="xref py py-meth docutils literal notranslate"><span class="pre">Tensor.size()</span></code>.</p>
+<p>Example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">t</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="mi">5</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">t</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+<span class="go">torch.Size([3, 4, 5])</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">t</span><span class="o">.</span><span class="n">shape</span>
+<span class="go">torch.Size([3, 4, 5])</span>
+</pre></div>
+</div>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.tv_tensors.set_return_type.html" class="btn btn-neutral float-right" title="set_return_type" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.tv_tensors.Mask.html" class="btn btn-neutral" title="Mask" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">TVTensor</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.Video.html b/0.16/generated/torchvision.tv_tensors.Video.html
new file mode 100644
index 00000000000..ae00fafdf10
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.Video.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Video &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="BoundingBoxFormat" href="torchvision.tv_tensors.BoundingBoxFormat.html" />
+    <link rel="prev" title="Image" href="torchvision.tv_tensors.Image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>Video</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.Video.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="video">
+<h1>Video<a class="headerlink" href="#video" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.Video">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">Video</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.dtype" title="(in PyTorch v2.0)"><span class="pre">dtype</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><span class="pre">device</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">requires_grad</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_video.html#Video"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.Video" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for videos.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>data</strong> (<em>tensor-like</em>) – Any data that can be turned into a tensor with <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.as_tensor.html#torch.as_tensor" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.as_tensor()</span></code></a>.</p></li>
+<li><p><strong>dtype</strong> (<em>torch.dpython:type</em><em>, </em><em>optional</em>) – Desired data type. If omitted, will be inferred from
+<code class="docutils literal notranslate"><span class="pre">data</span></code>.</p></li>
+<li><p><strong>device</strong> (<a class="reference external" href="https://pytorch.org/docs/stable/tensor_attributes.html#torch.device" title="(in PyTorch v2.0)"><em>torch.device</em></a><em>, </em><em>optional</em>) – Desired device. If omitted and <code class="docutils literal notranslate"><span class="pre">data</span></code> is a
+<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the device is taken from it. Otherwise, the video is constructed on the CPU.</p></li>
+<li><p><strong>requires_grad</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – Whether autograd should record operations. If omitted and
+<code class="docutils literal notranslate"><span class="pre">data</span></code> is a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>, the value is taken from it. Otherwise, defaults to <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">Video</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates all of what you need to know to get started with the new torchvision.t..."><img alt="" src="../_images/sphx_glr_plot_transforms_getting_started_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Getting started with transforms v2</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.tv_tensors.BoundingBoxFormat.html" class="btn btn-neutral float-right" title="BoundingBoxFormat" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.tv_tensors.Image.html" class="btn btn-neutral" title="Image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Video</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.set_return_type.html b/0.16/generated/torchvision.tv_tensors.set_return_type.html
new file mode 100644
index 00000000000..b78e233b3cc
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.set_return_type.html
@@ -0,0 +1,730 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>set_return_type &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="wrap" href="torchvision.tv_tensors.wrap.html" />
+    <link rel="prev" title="TVTensor" href="torchvision.tv_tensors.TVTensor.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>set_return_type</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.set_return_type.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="set-return-type">
+<h1>set_return_type<a class="headerlink" href="#set-return-type" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.set_return_type">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">set_return_type</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">return_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors/_torch_function_helpers.html#set_return_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.set_return_type" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Set the return type of torch operations on <a class="reference internal" href="torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a>.</p>
+<p>This only affects the behaviour of torch operations. It has no effect on
+<code class="docutils literal notranslate"><span class="pre">torchvision</span></code> transforms or functionals, which will always return as
+output the same type that was passed as input.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>We recommend using <a class="reference internal" href="torchvision.transforms.v2.ToPureTensor.html#torchvision.transforms.v2.ToPureTensor" title="torchvision.transforms.v2.ToPureTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToPureTensor</span></code></a> at
+the end of your transform pipelines if you use
+<code class="docutils literal notranslate"><span class="pre">set_return_type(&quot;TVTensor&quot;)</span></code>. This will avoid the
+<code class="docutils literal notranslate"><span class="pre">__torch_function__</span></code> overhead in the models <code class="docutils literal notranslate"><span class="pre">forward()</span></code>.</p>
+</div>
+<p>Can be used as a global flag for the entire program:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">img</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">5</span><span class="p">))</span>
+<span class="n">img</span> <span class="o">+</span> <span class="mi">2</span>  <span class="c1"># This is a pure Tensor (default behaviour)</span>
+
+<span class="n">set_return_type</span><span class="p">(</span><span class="s2">&quot;TVTensor&quot;</span><span class="p">)</span>
+<span class="n">img</span> <span class="o">+</span> <span class="mi">2</span>  <span class="c1"># This is an Image</span>
+</pre></div>
+</div>
+<p>or as a context manager to restrict the scope:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">img</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">Image</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">,</span> <span class="mi">5</span><span class="p">))</span>
+<span class="n">img</span> <span class="o">+</span> <span class="mi">2</span>  <span class="c1"># This is a pure Tensor</span>
+<span class="k">with</span> <span class="n">set_return_type</span><span class="p">(</span><span class="s2">&quot;TVTensor&quot;</span><span class="p">):</span>
+    <span class="n">img</span> <span class="o">+</span> <span class="mi">2</span>  <span class="c1"># This is an Image</span>
+<span class="n">img</span> <span class="o">+</span> <span class="mi">2</span>  <span class="c1"># This is a pure Tensor</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>return_type</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – Can be “TVTensor” or “Tensor” (case-insensitive).
+Default is “Tensor” (i.e. pure <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a>).</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">set_return_type</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.tv_tensors.wrap.html" class="btn btn-neutral float-right" title="wrap" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.tv_tensors.TVTensor.html" class="btn btn-neutral" title="TVTensor" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">set_return_type</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.tv_tensors.wrap.html b/0.16/generated/torchvision.tv_tensors.wrap.html
new file mode 100644
index 00000000000..e6975091ef7
--- /dev/null
+++ b/0.16/generated/torchvision.tv_tensors.wrap.html
@@ -0,0 +1,714 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>wrap &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Models and pre-trained weights" href="../models.html" />
+    <link rel="prev" title="set_return_type" href="torchvision.tv_tensors.set_return_type.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../tv_tensors.html">TVTensors</a> &gt;</li>
+        
+      <li>wrap</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.tv_tensors.wrap.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="wrap">
+<h1>wrap<a class="headerlink" href="#wrap" title="Permalink to this heading">¶</a></h1>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.tv_tensors.wrap">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.tv_tensors.</span></span><span class="sig-name descname"><span class="pre">wrap</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">wrappee</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">like</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../_modules/torchvision/tv_tensors.html#wrap"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.tv_tensors.wrap" title="Permalink to this definition">¶</a></dt>
+<dd><p>[BETA] Convert a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> (<code class="docutils literal notranslate"><span class="pre">wrappee</span></code>) into the same <a class="reference internal" href="torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a> subclass as <code class="docutils literal notranslate"><span class="pre">like</span></code>.</p>
+<p>If <code class="docutils literal notranslate"><span class="pre">like</span></code> is a <a class="reference internal" href="torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a>, the <code class="docutils literal notranslate"><span class="pre">format</span></code> and <code class="docutils literal notranslate"><span class="pre">canvas_size</span></code> of
+<code class="docutils literal notranslate"><span class="pre">like</span></code> are assigned to <code class="docutils literal notranslate"><span class="pre">wrappee</span></code>, unless they are passed as <code class="docutils literal notranslate"><span class="pre">kwargs</span></code>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>wrappee</strong> (<em>Tensor</em>) – The tensor to convert.</p></li>
+<li><p><strong>like</strong> (<a class="reference internal" href="torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a>) – The reference.
+<code class="docutils literal notranslate"><span class="pre">wrappee</span></code> will be converted into the same subclass as <code class="docutils literal notranslate"><span class="pre">like</span></code>.</p></li>
+<li><p><strong>kwargs</strong> – Can contain “format” and “canvas_size” if <code class="docutils literal notranslate"><span class="pre">like</span></code> is a <code class="xref py py-class docutils literal notranslate"><span class="pre">BoundingBoxes</span></code>.
+Ignored otherwise.</p></li>
+</ul>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">wrap</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip=" TVTensors are Tensor subclasses introduced together with torchvision.transforms.v2. This examp..."><img alt="" src="../_images/sphx_glr_plot_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a></p>
+  <div class="sphx-glr-thumbnail-title">TVTensors FAQ</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This guide is intended for advanced users and downstream library maintainers. We explain how to..."><img alt="" src="../_images/sphx_glr_plot_custom_tv_tensors_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/transforms/plot_custom_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-custom-tv-tensors-py"><span class="std std-ref">How to write your own TVTensor class</span></a></p>
+  <div class="sphx-glr-thumbnail-title">How to write your own TVTensor class</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../models.html" class="btn btn-neutral float-right" title="Models and pre-trained weights" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.tv_tensors.set_return_type.html" class="btn btn-neutral" title="set_return_type" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">wrap</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.utils.draw_bounding_boxes.html b/0.16/generated/torchvision.utils.draw_bounding_boxes.html
new file mode 100644
index 00000000000..dc764271ab8
--- /dev/null
+++ b/0.16/generated/torchvision.utils.draw_bounding_boxes.html
@@ -0,0 +1,730 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>draw_bounding_boxes &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="draw_segmentation_masks" href="torchvision.utils.draw_segmentation_masks.html" />
+    <link rel="prev" title="Utils" href="../utils.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../utils.html">Utils</a> &gt;</li>
+        
+      <li>draw_bounding_boxes</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.utils.draw_bounding_boxes.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="draw-bounding-boxes">
+<h1>draw_bounding_boxes<a class="headerlink" href="#draw-bounding-boxes" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.utils.draw_bounding_boxes">
+<span class="sig-prename descclassname"><span class="pre">torchvision.utils.</span></span><span class="sig-name descname"><span class="pre">draw_bounding_boxes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">boxes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">colors</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fill</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">font</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">font_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/utils.html#draw_bounding_boxes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.utils.draw_bounding_boxes" title="Permalink to this definition">¶</a></dt>
+<dd><p>Draws bounding boxes on given image.
+The values of the input image should be uint8 between 0 and 255.
+If fill is True, Resulting Tensor should be saved as PNG image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>image</strong> (<em>Tensor</em>) – Tensor of shape (C x H x W) and dtype uint8.</p></li>
+<li><p><strong>boxes</strong> (<em>Tensor</em>) – Tensor of size (N, 4) containing bounding boxes in (xmin, ymin, xmax, ymax) format. Note that
+the boxes are absolute coordinates with respect to the image. In other words: <cite>0 &lt;= xmin &lt; xmax &lt; W</cite> and
+<cite>0 &lt;= ymin &lt; ymax &lt; H</cite>.</p></li>
+<li><p><strong>labels</strong> (<em>List</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>]</em>) – List containing the labels of bounding boxes.</p></li>
+<li><p><strong>colors</strong> (<em>color</em><em> or </em><em>list of colors</em><em>, </em><em>optional</em>) – List containing the colors
+of the boxes or single color for all boxes. The color can be represented as
+PIL strings e.g. “red” or “#FF00FF”, or as RGB tuples e.g. <code class="docutils literal notranslate"><span class="pre">(240,</span> <span class="pre">10,</span> <span class="pre">157)</span></code>.
+By default, random colors are generated for boxes.</p></li>
+<li><p><strong>fill</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If <cite>True</cite> fills the bounding box with specified color.</p></li>
+<li><p><strong>width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Width of bounding box.</p></li>
+<li><p><strong>font</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a>) – A filename containing a TrueType font. If the file is not found in this filename, the loader may
+also search in other directories, such as the <cite>fonts/</cite> directory on Windows or <cite>/Library/Fonts/</cite>,
+<cite>/System/Library/Fonts/</cite> and <cite>~/Library/Fonts/</cite> on macOS.</p></li>
+<li><p><strong>font_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – The requested font size in points.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Image Tensor of dtype uint8 with bounding boxes plotted.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>img (Tensor[C, H, W])</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">draw_bounding_boxes</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.utils.draw_segmentation_masks.html" class="btn btn-neutral float-right" title="draw_segmentation_masks" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../utils.html" class="btn btn-neutral" title="Utils" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">draw_bounding_boxes</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.utils.draw_keypoints.html b/0.16/generated/torchvision.utils.draw_keypoints.html
new file mode 100644
index 00000000000..b448e653428
--- /dev/null
+++ b/0.16/generated/torchvision.utils.draw_keypoints.html
@@ -0,0 +1,720 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>draw_keypoints &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="flow_to_image" href="torchvision.utils.flow_to_image.html" />
+    <link rel="prev" title="draw_segmentation_masks" href="torchvision.utils.draw_segmentation_masks.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../utils.html">Utils</a> &gt;</li>
+        
+      <li>draw_keypoints</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.utils.draw_keypoints.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="draw-keypoints">
+<h1>draw_keypoints<a class="headerlink" href="#draw-keypoints" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.utils.draw_keypoints">
+<span class="sig-prename descclassname"><span class="pre">torchvision.utils.</span></span><span class="sig-name descname"><span class="pre">draw_keypoints</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">keypoints</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">connectivity</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">colors</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">radius</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">width</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">3</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/utils.html#draw_keypoints"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.utils.draw_keypoints" title="Permalink to this definition">¶</a></dt>
+<dd><p>Draws Keypoints on given RGB image.
+The values of the input image should be uint8 between 0 and 255.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>image</strong> (<em>Tensor</em>) – Tensor of shape (3, H, W) and dtype uint8.</p></li>
+<li><p><strong>keypoints</strong> (<em>Tensor</em>) – Tensor of shape (num_instances, K, 2) the K keypoints location for each of the N instances,
+in the format [x, y].</p></li>
+<li><p><strong>connectivity</strong> (<em>List</em><em>[</em><em>Tuple</em><em>[</em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>]</em><em>]</em><em>]</em>) – A List of tuple where,
+each tuple contains pair of keypoints to be connected.</p></li>
+<li><p><strong>colors</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><em>str</em></a><em>, </em><em>Tuple</em>) – The color can be represented as
+PIL strings e.g. “red” or “#FF00FF”, or as RGB tuples e.g. <code class="docutils literal notranslate"><span class="pre">(240,</span> <span class="pre">10,</span> <span class="pre">157)</span></code>.</p></li>
+<li><p><strong>radius</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Integer denoting radius of keypoint.</p></li>
+<li><p><strong>width</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a>) – Integer denoting width of line connecting keypoints.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Image Tensor of dtype uint8 with keypoints drawn.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>img (Tensor[C, H, W])</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">draw_keypoints</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.utils.flow_to_image.html" class="btn btn-neutral float-right" title="flow_to_image" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.utils.draw_segmentation_masks.html" class="btn btn-neutral" title="draw_segmentation_masks" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">draw_keypoints</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.utils.draw_segmentation_masks.html b/0.16/generated/torchvision.utils.draw_segmentation_masks.html
new file mode 100644
index 00000000000..d48f6c49d7d
--- /dev/null
+++ b/0.16/generated/torchvision.utils.draw_segmentation_masks.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>draw_segmentation_masks &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="draw_keypoints" href="torchvision.utils.draw_keypoints.html" />
+    <link rel="prev" title="draw_bounding_boxes" href="torchvision.utils.draw_bounding_boxes.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../utils.html">Utils</a> &gt;</li>
+        
+      <li>draw_segmentation_masks</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.utils.draw_segmentation_masks.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="draw-segmentation-masks">
+<h1>draw_segmentation_masks<a class="headerlink" href="#draw-segmentation-masks" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.utils.draw_segmentation_masks">
+<span class="sig-prename descclassname"><span class="pre">torchvision.utils.</span></span><span class="sig-name descname"><span class="pre">draw_segmentation_masks</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">masks</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.8</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">colors</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/utils.html#draw_segmentation_masks"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.utils.draw_segmentation_masks" title="Permalink to this definition">¶</a></dt>
+<dd><p>Draws segmentation masks on given RGB image.
+The values of the input image should be uint8 between 0 and 255.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>image</strong> (<em>Tensor</em>) – Tensor of shape (3, H, W) and dtype uint8.</p></li>
+<li><p><strong>masks</strong> (<em>Tensor</em>) – Tensor of shape (num_masks, H, W) or (H, W) and dtype bool.</p></li>
+<li><p><strong>alpha</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a>) – Float number between 0 and 1 denoting the transparency of the masks.
+0 means full transparency, 1 means no transparency.</p></li>
+<li><p><strong>colors</strong> (<em>color</em><em> or </em><em>list of colors</em><em>, </em><em>optional</em>) – List containing the colors
+of the masks or single color for all masks. The color can be represented as
+PIL strings e.g. “red” or “#FF00FF”, or as RGB tuples e.g. <code class="docutils literal notranslate"><span class="pre">(240,</span> <span class="pre">10,</span> <span class="pre">157)</span></code>.
+By default, random colors are generated for each mask.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>Image Tensor, with segmentation masks drawn on top.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>img (Tensor[C, H, W])</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">draw_segmentation_masks</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.utils.draw_keypoints.html" class="btn btn-neutral float-right" title="draw_keypoints" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.utils.draw_bounding_boxes.html" class="btn btn-neutral" title="draw_bounding_boxes" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">draw_segmentation_masks</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.utils.flow_to_image.html b/0.16/generated/torchvision.utils.flow_to_image.html
new file mode 100644
index 00000000000..1bb6cdb719e
--- /dev/null
+++ b/0.16/generated/torchvision.utils.flow_to_image.html
@@ -0,0 +1,713 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>flow_to_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="make_grid" href="torchvision.utils.make_grid.html" />
+    <link rel="prev" title="draw_keypoints" href="torchvision.utils.draw_keypoints.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../utils.html">Utils</a> &gt;</li>
+        
+      <li>flow_to_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.utils.flow_to_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="flow-to-image">
+<h1>flow_to_image<a class="headerlink" href="#flow-to-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.utils.flow_to_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.utils.</span></span><span class="sig-name descname"><span class="pre">flow_to_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">flow</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/utils.html#flow_to_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.utils.flow_to_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>Converts a flow to an RGB image.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>flow</strong> (<em>Tensor</em>) – Flow of shape (N, 2, H, W) or (2, H, W) and dtype torch.float.</p>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p><dl class="simple">
+<dt>Image Tensor of dtype uint8 where each color corresponds</dt><dd><p>to a given flow direction. Shape is (N, 3, H, W) or (3, H, W) depending on the input.</p>
+</dd>
+</dl>
+</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>img (Tensor)</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">flow_to_image</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.utils.make_grid.html" class="btn btn-neutral float-right" title="make_grid" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.utils.draw_keypoints.html" class="btn btn-neutral" title="draw_keypoints" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">flow_to_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.utils.make_grid.html b/0.16/generated/torchvision.utils.make_grid.html
new file mode 100644
index 00000000000..57af3cee14a
--- /dev/null
+++ b/0.16/generated/torchvision.utils.make_grid.html
@@ -0,0 +1,723 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>make_grid &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="save_image" href="torchvision.utils.save_image.html" />
+    <link rel="prev" title="flow_to_image" href="torchvision.utils.flow_to_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../utils.html">Utils</a> &gt;</li>
+        
+      <li>make_grid</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.utils.make_grid.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="make-grid">
+<h1>make_grid<a class="headerlink" href="#make-grid" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.utils.make_grid">
+<span class="sig-prename descclassname"><span class="pre">torchvision.utils.</span></span><span class="sig-name descname"><span class="pre">make_grid</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">nrow</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">8</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">normalize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">value_range</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Tuple" title="(in Python v3.12)"><span class="pre">Tuple</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scale_each</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_value</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><span class="pre">float</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a></span></span><a class="reference internal" href="../_modules/torchvision/utils.html#make_grid"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.utils.make_grid" title="Permalink to this definition">¶</a></dt>
+<dd><p>Make a grid of images.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>tensor</strong> (<em>Tensor</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a>) – 4D mini-batch Tensor of shape (B x C x H x W)
+or a list of images all of the same size.</p></li>
+<li><p><strong>nrow</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – Number of images displayed in each row of the grid.
+The final grid size is <code class="docutils literal notranslate"><span class="pre">(B</span> <span class="pre">/</span> <span class="pre">nrow,</span> <span class="pre">nrow)</span></code>. Default: <code class="docutils literal notranslate"><span class="pre">8</span></code>.</p></li>
+<li><p><strong>padding</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – amount of padding. Default: <code class="docutils literal notranslate"><span class="pre">2</span></code>.</p></li>
+<li><p><strong>normalize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, shift the image to the range (0, 1),
+by the min and max values specified by <code class="docutils literal notranslate"><span class="pre">value_range</span></code>. Default: <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+<li><p><strong>value_range</strong> (<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#tuple" title="(in Python v3.12)"><em>tuple</em></a><em>, </em><em>optional</em>) – tuple (min, max) where min and max are numbers,
+then these numbers are used to normalize the image. By default, min and max
+are computed from the tensor.</p></li>
+<li><p><strong>scale_each</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If <code class="docutils literal notranslate"><span class="pre">True</span></code>, scale each image in the batch of
+images separately rather than the (min, max) over all images. Default: <code class="docutils literal notranslate"><span class="pre">False</span></code>.</p></li>
+<li><p><strong>pad_value</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#float" title="(in Python v3.12)"><em>float</em></a><em>, </em><em>optional</em>) – Value for the padded pixels. Default: <code class="docutils literal notranslate"><span class="pre">0</span></code>.</p></li>
+</ul>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><p>the tensor containing grid of images.</p>
+</dd>
+<dt class="field-odd">Return type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>grid (Tensor)</p>
+</dd>
+</dl>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">make_grid</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.utils.save_image.html" class="btn btn-neutral float-right" title="save_image" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.utils.flow_to_image.html" class="btn btn-neutral" title="flow_to_image" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">make_grid</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/torchvision.utils.save_image.html b/0.16/generated/torchvision.utils.save_image.html
new file mode 100644
index 00000000000..644b9b4d920
--- /dev/null
+++ b/0.16/generated/torchvision.utils.save_image.html
@@ -0,0 +1,706 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>save_image &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Operators" href="../ops.html" />
+    <link rel="prev" title="make_grid" href="torchvision.utils.make_grid.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../utils.html">Utils</a> &gt;</li>
+        
+      <li>save_image</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/torchvision.utils.save_image.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="save-image">
+<h1>save_image<a class="headerlink" href="#save-image" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.utils.save_image">
+<span class="sig-prename descclassname"><span class="pre">torchvision.utils.</span></span><span class="sig-name descname"><span class="pre">save_image</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensor</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.List" title="(in Python v3.12)"><span class="pre">List</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><span class="pre">Tensor</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fp</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/pathlib.html#pathlib.Path" title="(in Python v3.12)"><span class="pre">Path</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.BinaryIO" title="(in Python v3.12)"><span class="pre">BinaryIO</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><span class="pre">str</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference external" href="https://docs.python.org/3/library/constants.html#None" title="(in Python v3.12)"><span class="pre">None</span></a></span></span><a class="reference internal" href="../_modules/torchvision/utils.html#save_image"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.utils.save_image" title="Permalink to this definition">¶</a></dt>
+<dd><p>Save a given Tensor into an image file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>tensor</strong> (<em>Tensor</em><em> or </em><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#list" title="(in Python v3.12)"><em>list</em></a>) – Image to be saved. If given a mini-batch tensor,
+saves the tensor as a grid of images by calling <code class="docutils literal notranslate"><span class="pre">make_grid</span></code>.</p></li>
+<li><p><strong>fp</strong> (<em>string</em><em> or </em><em>file object</em>) – A filename or a file object</p></li>
+<li><p><strong>format</strong> (<em>Optional</em>) – If omitted, the format to use is determined from the filename extension.
+If a file object was used instead of a filename, this parameter should always be used.</p></li>
+<li><p><strong>**kwargs</strong> – Other arguments are documented in <code class="docutils literal notranslate"><span class="pre">make_grid</span></code>.</p></li>
+</ul>
+</dd>
+</dl>
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../ops.html" class="btn btn-neutral float-right" title="Operators" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.utils.make_grid.html" class="btn btn-neutral" title="make_grid" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">save_image</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/generated/video_table.html b/0.16/generated/video_table.html
new file mode 100644
index 00000000000..e7b9a8ddb0c
--- /dev/null
+++ b/0.16/generated/video_table.html
@@ -0,0 +1,762 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>&lt;no title&gt;</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/generated/video_table.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Acc&#64;1</strong></p></th>
+<th class="head"><p><strong>Acc&#64;5</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MC3_18_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>63.96</p></td>
+<td><p>84.13</p></td>
+<td><p>11.7M</p></td>
+<td><p>43.34</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V1_B_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>78.477</p></td>
+<td><p>93.582</p></td>
+<td><p>36.6M</p></td>
+<td><p>70.6</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/pytorchvideo/blob/main/docs/source/model_zoo.md">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V2_S_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>80.757</p></td>
+<td><p>94.665</p></td>
+<td><p>34.5M</p></td>
+<td><p>64.22</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SlowFast/blob/main/MODEL_ZOO.md">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>67.463</p></td>
+<td><p>86.175</p></td>
+<td><p>31.5M</p></td>
+<td><p>40.52</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">R3D_18_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>63.2</p></td>
+<td><p>83.479</p></td>
+<td><p>33.4M</p></td>
+<td><p>40.7</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">S3D_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>68.368</p></td>
+<td><p>88.05</p></td>
+<td><p>8.3M</p></td>
+<td><p>17.98</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification#s3d">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_B_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>79.427</p></td>
+<td><p>94.386</p></td>
+<td><p>88.0M</p></td>
+<td><p>140.67</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_B_Weights.KINETICS400_IMAGENET22K_V1</span></code></p></td>
+<td><p>81.643</p></td>
+<td><p>95.574</p></td>
+<td><p>88.0M</p></td>
+<td><p>140.67</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_S_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>79.521</p></td>
+<td><p>94.158</p></td>
+<td><p>49.8M</p></td>
+<td><p>82.84</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_T_Weights.KINETICS400_V1</span></code></p></td>
+<td><p>77.715</p></td>
+<td><p>93.519</p></td>
+<td><p>28.2M</p></td>
+<td><p>43.88</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+</tbody>
+</table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul class="simple">
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/genindex.html b/0.16/genindex.html
new file mode 100644
index 00000000000..35a609885af
--- /dev/null
+++ b/0.16/genindex.html
@@ -0,0 +1,2479 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Index &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="#" />
+    <link rel="search" title="Search" href="search.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Index</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+
+<h1 id="index">Index</h1>
+
+<div class="genindex-jumpbox">
+ <a href="#_"><strong>_</strong></a>
+ | <a href="#A"><strong>A</strong></a>
+ | <a href="#B"><strong>B</strong></a>
+ | <a href="#C"><strong>C</strong></a>
+ | <a href="#D"><strong>D</strong></a>
+ | <a href="#E"><strong>E</strong></a>
+ | <a href="#F"><strong>F</strong></a>
+ | <a href="#G"><strong>G</strong></a>
+ | <a href="#H"><strong>H</strong></a>
+ | <a href="#I"><strong>I</strong></a>
+ | <a href="#K"><strong>K</strong></a>
+ | <a href="#L"><strong>L</strong></a>
+ | <a href="#M"><strong>M</strong></a>
+ | <a href="#N"><strong>N</strong></a>
+ | <a href="#O"><strong>O</strong></a>
+ | <a href="#P"><strong>P</strong></a>
+ | <a href="#Q"><strong>Q</strong></a>
+ | <a href="#R"><strong>R</strong></a>
+ | <a href="#S"><strong>S</strong></a>
+ | <a href="#T"><strong>T</strong></a>
+ | <a href="#U"><strong>U</strong></a>
+ | <a href="#V"><strong>V</strong></a>
+ | <a href="#W"><strong>W</strong></a>
+ 
+</div>
+<h2 id="_">_</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.Caltech101.html#torchvision.datasets.Caltech101.__getitem__">__getitem__() (torchvision.datasets.Caltech101 method)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.datasets.Caltech256.html#torchvision.datasets.Caltech256.__getitem__">(torchvision.datasets.Caltech256 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.CarlaStereo.html#torchvision.datasets.CarlaStereo.__getitem__">(torchvision.datasets.CarlaStereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.CelebA.html#torchvision.datasets.CelebA.__getitem__">(torchvision.datasets.CelebA method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.CIFAR10.html#torchvision.datasets.CIFAR10.__getitem__">(torchvision.datasets.CIFAR10 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.CIFAR100.html#torchvision.datasets.CIFAR100.__getitem__">(torchvision.datasets.CIFAR100 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Cityscapes.html#torchvision.datasets.Cityscapes.__getitem__">(torchvision.datasets.Cityscapes method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.CocoCaptions.html#torchvision.datasets.CocoCaptions.__getitem__">(torchvision.datasets.CocoCaptions method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection.__getitem__">(torchvision.datasets.CocoDetection method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Country211.html#torchvision.datasets.Country211.__getitem__">(torchvision.datasets.Country211 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.CREStereo.html#torchvision.datasets.CREStereo.__getitem__">(torchvision.datasets.CREStereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.DTD.html#torchvision.datasets.DTD.__getitem__">(torchvision.datasets.DTD method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.EMNIST.html#torchvision.datasets.EMNIST.__getitem__">(torchvision.datasets.EMNIST method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.ETH3DStereo.html#torchvision.datasets.ETH3DStereo.__getitem__">(torchvision.datasets.ETH3DStereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.EuroSAT.html#torchvision.datasets.EuroSAT.__getitem__">(torchvision.datasets.EuroSAT method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.FakeData.html#torchvision.datasets.FakeData.__getitem__">(torchvision.datasets.FakeData method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.FallingThingsStereo.html#torchvision.datasets.FallingThingsStereo.__getitem__">(torchvision.datasets.FallingThingsStereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.FashionMNIST.html#torchvision.datasets.FashionMNIST.__getitem__">(torchvision.datasets.FashionMNIST method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.FER2013.html#torchvision.datasets.FER2013.__getitem__">(torchvision.datasets.FER2013 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.FGVCAircraft.html#torchvision.datasets.FGVCAircraft.__getitem__">(torchvision.datasets.FGVCAircraft method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Flickr30k.html#torchvision.datasets.Flickr30k.__getitem__">(torchvision.datasets.Flickr30k method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Flickr8k.html#torchvision.datasets.Flickr8k.__getitem__">(torchvision.datasets.Flickr8k method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Flowers102.html#torchvision.datasets.Flowers102.__getitem__">(torchvision.datasets.Flowers102 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs.__getitem__">(torchvision.datasets.FlyingChairs method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D.__getitem__">(torchvision.datasets.FlyingThings3D method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Food101.html#torchvision.datasets.Food101.__getitem__">(torchvision.datasets.Food101 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.GTSRB.html#torchvision.datasets.GTSRB.__getitem__">(torchvision.datasets.GTSRB method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.HD1K.html#torchvision.datasets.HD1K.__getitem__">(torchvision.datasets.HD1K method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.HMDB51.html#torchvision.datasets.HMDB51.__getitem__">(torchvision.datasets.HMDB51 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.ImageNet.html#torchvision.datasets.ImageNet.__getitem__">(torchvision.datasets.ImageNet method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.INaturalist.html#torchvision.datasets.INaturalist.__getitem__">(torchvision.datasets.INaturalist method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.InStereo2k.html#torchvision.datasets.InStereo2k.__getitem__">(torchvision.datasets.InStereo2k method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Kinetics.html#torchvision.datasets.Kinetics.__getitem__">(torchvision.datasets.Kinetics method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Kitti.html#torchvision.datasets.Kitti.__getitem__">(torchvision.datasets.Kitti method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Kitti2012Stereo.html#torchvision.datasets.Kitti2012Stereo.__getitem__">(torchvision.datasets.Kitti2012Stereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Kitti2015Stereo.html#torchvision.datasets.Kitti2015Stereo.__getitem__">(torchvision.datasets.Kitti2015Stereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow.__getitem__">(torchvision.datasets.KittiFlow method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.KMNIST.html#torchvision.datasets.KMNIST.__getitem__">(torchvision.datasets.KMNIST method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.LFWPairs.html#torchvision.datasets.LFWPairs.__getitem__">(torchvision.datasets.LFWPairs method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.LFWPeople.html#torchvision.datasets.LFWPeople.__getitem__">(torchvision.datasets.LFWPeople method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.LSUN.html#torchvision.datasets.LSUN.__getitem__">(torchvision.datasets.LSUN method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Middlebury2014Stereo.html#torchvision.datasets.Middlebury2014Stereo.__getitem__">(torchvision.datasets.Middlebury2014Stereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.MNIST.html#torchvision.datasets.MNIST.__getitem__">(torchvision.datasets.MNIST method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.MovingMNIST.html#torchvision.datasets.MovingMNIST.__getitem__">(torchvision.datasets.MovingMNIST method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Omniglot.html#torchvision.datasets.Omniglot.__getitem__">(torchvision.datasets.Omniglot method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.OxfordIIITPet.html#torchvision.datasets.OxfordIIITPet.__getitem__">(torchvision.datasets.OxfordIIITPet method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.PCAM.html#torchvision.datasets.PCAM.__getitem__">(torchvision.datasets.PCAM method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.PhotoTour.html#torchvision.datasets.PhotoTour.__getitem__">(torchvision.datasets.PhotoTour method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Places365.html#torchvision.datasets.Places365.__getitem__">(torchvision.datasets.Places365 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.QMNIST.html#torchvision.datasets.QMNIST.__getitem__">(torchvision.datasets.QMNIST method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.RenderedSST2.html#torchvision.datasets.RenderedSST2.__getitem__">(torchvision.datasets.RenderedSST2 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.SBDataset.html#torchvision.datasets.SBDataset.__getitem__">(torchvision.datasets.SBDataset method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.SBU.html#torchvision.datasets.SBU.__getitem__">(torchvision.datasets.SBU method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.SceneFlowStereo.html#torchvision.datasets.SceneFlowStereo.__getitem__">(torchvision.datasets.SceneFlowStereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.SEMEION.html#torchvision.datasets.SEMEION.__getitem__">(torchvision.datasets.SEMEION method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.Sintel.html#torchvision.datasets.Sintel.__getitem__">(torchvision.datasets.Sintel method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.SintelStereo.html#torchvision.datasets.SintelStereo.__getitem__">(torchvision.datasets.SintelStereo method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.StanfordCars.html#torchvision.datasets.StanfordCars.__getitem__">(torchvision.datasets.StanfordCars method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.STL10.html#torchvision.datasets.STL10.__getitem__">(torchvision.datasets.STL10 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.SUN397.html#torchvision.datasets.SUN397.__getitem__">(torchvision.datasets.SUN397 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.SVHN.html#torchvision.datasets.SVHN.__getitem__">(torchvision.datasets.SVHN method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.UCF101.html#torchvision.datasets.UCF101.__getitem__">(torchvision.datasets.UCF101 method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.USPS.html#torchvision.datasets.USPS.__getitem__">(torchvision.datasets.USPS method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.VOCDetection.html#torchvision.datasets.VOCDetection.__getitem__">(torchvision.datasets.VOCDetection method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.VOCSegmentation.html#torchvision.datasets.VOCSegmentation.__getitem__">(torchvision.datasets.VOCSegmentation method)</a>
+</li>
+        <li><a href="generated/torchvision.datasets.WIDERFace.html#torchvision.datasets.WIDERFace.__getitem__">(torchvision.datasets.WIDERFace method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="A">A</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.functional.adjust_brightness.html#torchvision.transforms.functional.adjust_brightness">adjust_brightness() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.adjust_brightness.html#torchvision.transforms.v2.functional.adjust_brightness">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.adjust_contrast.html#torchvision.transforms.functional.adjust_contrast">adjust_contrast() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.adjust_contrast.html#torchvision.transforms.v2.functional.adjust_contrast">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.adjust_gamma.html#torchvision.transforms.functional.adjust_gamma">adjust_gamma() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.adjust_gamma.html#torchvision.transforms.v2.functional.adjust_gamma">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.adjust_hue.html#torchvision.transforms.functional.adjust_hue">adjust_hue() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.adjust_hue.html#torchvision.transforms.v2.functional.adjust_hue">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.adjust_saturation.html#torchvision.transforms.functional.adjust_saturation">adjust_saturation() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.adjust_saturation.html#torchvision.transforms.v2.functional.adjust_saturation">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.adjust_sharpness.html#torchvision.transforms.functional.adjust_sharpness">adjust_sharpness() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.adjust_sharpness.html#torchvision.transforms.v2.functional.adjust_sharpness">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.functional.affine.html#torchvision.transforms.functional.affine">affine() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.affine.html#torchvision.transforms.v2.functional.affine">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.alexnet.html#torchvision.models.alexnet">alexnet() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.alexnet.html#torchvision.models.AlexNet_Weights">AlexNet_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.AugMix.html#torchvision.transforms.AugMix">AugMix (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.AugMix.html#torchvision.transforms.v2.AugMix">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.AutoAugment.html#torchvision.transforms.AutoAugment">AutoAugment (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.AutoAugment.html#torchvision.transforms.v2.AutoAugment">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy">AutoAugmentPolicy (class in torchvision.transforms)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.autocontrast.html#torchvision.transforms.functional.autocontrast">autocontrast() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.autocontrast.html#torchvision.transforms.v2.functional.autocontrast">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="B">B</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.ops.batched_nms.html#torchvision.ops.batched_nms">batched_nms() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes">BoundingBoxes (class in torchvision.tv_tensors)</a>
+</li>
+      <li><a href="generated/torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat">BoundingBoxFormat (class in torchvision.tv_tensors)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.ops.box_area.html#torchvision.ops.box_area">box_area() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.box_convert.html#torchvision.ops.box_convert">box_convert() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.box_iou.html#torchvision.ops.box_iou">box_iou() (in module torchvision.ops)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="C">C</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.Caltech101.html#torchvision.datasets.Caltech101">Caltech101 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Caltech256.html#torchvision.datasets.Caltech256">Caltech256 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.CarlaStereo.html#torchvision.datasets.CarlaStereo">CarlaStereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.INaturalist.html#torchvision.datasets.INaturalist.category_name">category_name() (torchvision.datasets.INaturalist method)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.CelebA.html#torchvision.datasets.CelebA">CelebA (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.center_crop.html#torchvision.transforms.functional.center_crop">center_crop() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.center_crop.html#torchvision.transforms.v2.functional.center_crop">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.CenterCrop.html#torchvision.transforms.CenterCrop">CenterCrop (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.CenterCrop.html#torchvision.transforms.v2.CenterCrop">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.CIFAR10.html#torchvision.datasets.CIFAR10">CIFAR10 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.CIFAR100.html#torchvision.datasets.CIFAR100">CIFAR100 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Cityscapes.html#torchvision.datasets.Cityscapes">Cityscapes (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html#torchvision.transforms.v2.functional.clamp_bounding_boxes">clamp_bounding_boxes() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.ClampBoundingBoxes.html#torchvision.transforms.v2.ClampBoundingBoxes">ClampBoundingBoxes (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.ops.clip_boxes_to_image.html#torchvision.ops.clip_boxes_to_image">clip_boxes_to_image() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.CocoCaptions.html#torchvision.datasets.CocoCaptions">CocoCaptions (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.CocoDetection.html#torchvision.datasets.CocoDetection">CocoDetection (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.ColorJitter.html#torchvision.transforms.ColorJitter">ColorJitter (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.ColorJitter.html#torchvision.transforms.v2.ColorJitter">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.ops.complete_box_iou.html#torchvision.ops.complete_box_iou">complete_box_iou() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.complete_box_iou_loss.html#torchvision.ops.complete_box_iou_loss">complete_box_iou_loss() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.Compose.html#torchvision.transforms.Compose">Compose (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.ops.Conv2dNormActivation.html#torchvision.ops.Conv2dNormActivation">Conv2dNormActivation (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.Conv3dNormActivation.html#torchvision.ops.Conv3dNormActivation">Conv3dNormActivation (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html#torchvision.transforms.v2.functional.convert_bounding_box_format">convert_bounding_box_format() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.convert_image_dtype.html#torchvision.transforms.v2.functional.convert_image_dtype">convert_image_dtype (class in torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.convert_image_dtype.html#torchvision.transforms.functional.convert_image_dtype">convert_image_dtype() (in module torchvision.transforms.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html#torchvision.transforms.v2.ConvertBoundingBoxFormat">ConvertBoundingBoxFormat (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.ConvertImageDtype.html#torchvision.transforms.ConvertImageDtype">ConvertImageDtype (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.ConvertImageDtype.html#torchvision.transforms.v2.ConvertImageDtype">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.convnext_base.html#torchvision.models.convnext_base">convnext_base() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.convnext_base.html#torchvision.models.ConvNeXt_Base_Weights">ConvNeXt_Base_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.convnext_large.html#torchvision.models.convnext_large">convnext_large() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.convnext_large.html#torchvision.models.ConvNeXt_Large_Weights">ConvNeXt_Large_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.convnext_small.html#torchvision.models.convnext_small">convnext_small() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.convnext_small.html#torchvision.models.ConvNeXt_Small_Weights">ConvNeXt_Small_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.convnext_tiny.html#torchvision.models.convnext_tiny">convnext_tiny() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.convnext_tiny.html#torchvision.models.ConvNeXt_Tiny_Weights">ConvNeXt_Tiny_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Country211.html#torchvision.datasets.Country211">Country211 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.models.feature_extraction.create_feature_extractor.html#torchvision.models.feature_extraction.create_feature_extractor">create_feature_extractor() (in module torchvision.models.feature_extraction)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.CREStereo.html#torchvision.datasets.CREStereo">CREStereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.crop.html#torchvision.transforms.functional.crop">crop() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.crop.html#torchvision.transforms.v2.functional.crop">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.CutMix.html#torchvision.transforms.v2.CutMix">CutMix (class in torchvision.transforms.v2)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="D">D</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder">DatasetFolder (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.io.decode_image.html#torchvision.io.decode_image">decode_image() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.decode_jpeg.html#torchvision.io.decode_jpeg">decode_jpeg() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.decode_png.html#torchvision.io.decode_png">decode_png() (in module torchvision.io)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html#torchvision.models.segmentation.deeplabv3_mobilenet_v3_large">deeplabv3_mobilenet_v3_large() (in module torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html#torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights">DeepLabV3_MobileNet_V3_Large_Weights (class in torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html#torchvision.models.segmentation.deeplabv3_resnet101">deeplabv3_resnet101() (in module torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html#torchvision.models.segmentation.DeepLabV3_ResNet101_Weights">DeepLabV3_ResNet101_Weights (class in torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html#torchvision.models.segmentation.deeplabv3_resnet50">deeplabv3_resnet50() (in module torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html#torchvision.models.segmentation.DeepLabV3_ResNet50_Weights">DeepLabV3_ResNet50_Weights (class in torchvision.models.segmentation)</a>
+</li>
+      <li><a href="generated/torchvision.ops.deform_conv2d.html#torchvision.ops.deform_conv2d">deform_conv2d() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.DeformConv2d.html#torchvision.ops.DeformConv2d">DeformConv2d (class in torchvision.ops)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.densenet121.html#torchvision.models.densenet121">densenet121() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.densenet121.html#torchvision.models.DenseNet121_Weights">DenseNet121_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.densenet161.html#torchvision.models.densenet161">densenet161() (in module torchvision.models)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.densenet161.html#torchvision.models.DenseNet161_Weights">DenseNet161_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.densenet169.html#torchvision.models.densenet169">densenet169() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.densenet169.html#torchvision.models.DenseNet169_Weights">DenseNet169_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.densenet201.html#torchvision.models.densenet201">densenet201() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.densenet201.html#torchvision.models.DenseNet201_Weights">DenseNet201_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor.device">device (torchvision.tv_tensors.TVTensor property)</a>
+</li>
+      <li><a href="generated/torchvision.ops.distance_box_iou.html#torchvision.ops.distance_box_iou">distance_box_iou() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.distance_box_iou_loss.html#torchvision.ops.distance_box_iou_loss">distance_box_iou_loss() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.utils.draw_bounding_boxes.html#torchvision.utils.draw_bounding_boxes">draw_bounding_boxes() (in module torchvision.utils)</a>
+</li>
+      <li><a href="generated/torchvision.utils.draw_keypoints.html#torchvision.utils.draw_keypoints">draw_keypoints() (in module torchvision.utils)</a>
+</li>
+      <li><a href="generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks">draw_segmentation_masks() (in module torchvision.utils)</a>
+</li>
+      <li><a href="generated/torchvision.ops.drop_block2d.html#torchvision.ops.drop_block2d">drop_block2d() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.drop_block3d.html#torchvision.ops.drop_block3d">drop_block3d() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.DropBlock2d.html#torchvision.ops.DropBlock2d">DropBlock2d (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.DropBlock3d.html#torchvision.ops.DropBlock3d">DropBlock3d (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.DTD.html#torchvision.datasets.DTD">DTD (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="E">E</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.efficientnet_b0.html#torchvision.models.efficientnet_b0">efficientnet_b0() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b0.html#torchvision.models.EfficientNet_B0_Weights">EfficientNet_B0_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b1.html#torchvision.models.efficientnet_b1">efficientnet_b1() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b1.html#torchvision.models.EfficientNet_B1_Weights">EfficientNet_B1_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b2.html#torchvision.models.efficientnet_b2">efficientnet_b2() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b2.html#torchvision.models.EfficientNet_B2_Weights">EfficientNet_B2_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b3.html#torchvision.models.efficientnet_b3">efficientnet_b3() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b3.html#torchvision.models.EfficientNet_B3_Weights">EfficientNet_B3_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b4.html#torchvision.models.efficientnet_b4">efficientnet_b4() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b4.html#torchvision.models.EfficientNet_B4_Weights">EfficientNet_B4_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b5.html#torchvision.models.efficientnet_b5">efficientnet_b5() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b5.html#torchvision.models.EfficientNet_B5_Weights">EfficientNet_B5_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b6.html#torchvision.models.efficientnet_b6">efficientnet_b6() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b6.html#torchvision.models.EfficientNet_B6_Weights">EfficientNet_B6_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b7.html#torchvision.models.efficientnet_b7">efficientnet_b7() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_b7.html#torchvision.models.EfficientNet_B7_Weights">EfficientNet_B7_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_v2_l.html#torchvision.models.efficientnet_v2_l">efficientnet_v2_l() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_v2_l.html#torchvision.models.EfficientNet_V2_L_Weights">EfficientNet_V2_L_Weights (class in torchvision.models)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.efficientnet_v2_m.html#torchvision.models.efficientnet_v2_m">efficientnet_v2_m() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_v2_m.html#torchvision.models.EfficientNet_V2_M_Weights">EfficientNet_V2_M_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_v2_s.html#torchvision.models.efficientnet_v2_s">efficientnet_v2_s() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.efficientnet_v2_s.html#torchvision.models.EfficientNet_V2_S_Weights">EfficientNet_V2_S_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.elastic.html#torchvision.transforms.v2.functional.elastic">elastic() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.ElasticTransform.html#torchvision.transforms.ElasticTransform">ElasticTransform (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.ElasticTransform.html#torchvision.transforms.v2.ElasticTransform">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.EMNIST.html#torchvision.datasets.EMNIST">EMNIST (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.io.encode_jpeg.html#torchvision.io.encode_jpeg">encode_jpeg() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.encode_png.html#torchvision.io.encode_png">encode_png() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.equalize.html#torchvision.transforms.functional.equalize">equalize() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.equalize.html#torchvision.transforms.v2.functional.equalize">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.erase.html#torchvision.transforms.functional.erase">erase() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.erase.html#torchvision.transforms.v2.functional.erase">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.ETH3DStereo.html#torchvision.datasets.ETH3DStereo">ETH3DStereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.EuroSAT.html#torchvision.datasets.EuroSAT">EuroSAT (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose.extra_repr">extra_repr() (torchvision.transforms.v2.Compose method)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.Lambda.html#torchvision.transforms.v2.Lambda.extra_repr">(torchvision.transforms.v2.Lambda method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomApply.html#torchvision.transforms.v2.RandomApply.extra_repr">(torchvision.transforms.v2.RandomApply method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="F">F</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.FakeData.html#torchvision.datasets.FakeData">FakeData (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.FallingThingsStereo.html#torchvision.datasets.FallingThingsStereo">FallingThingsStereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.FashionMNIST.html#torchvision.datasets.FashionMNIST">FashionMNIST (class in torchvision.datasets)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn">fasterrcnn_mobilenet_v3_large_320_fpn() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn">fasterrcnn_mobilenet_v3_large_fpn() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights">FasterRCNN_MobileNet_V3_Large_FPN_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn">fasterrcnn_resnet50_fpn() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html#torchvision.models.detection.fasterrcnn_resnet50_fpn_v2">fasterrcnn_resnet50_fpn_v2() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html#torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights">FasterRCNN_ResNet50_FPN_V2_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights">FasterRCNN_ResNet50_FPN_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.fcn_resnet101.html#torchvision.models.segmentation.fcn_resnet101">fcn_resnet101() (in module torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.fcn_resnet101.html#torchvision.models.segmentation.FCN_ResNet101_Weights">FCN_ResNet101_Weights (class in torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.fcn_resnet50.html#torchvision.models.segmentation.fcn_resnet50">fcn_resnet50() (in module torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.fcn_resnet50.html#torchvision.models.segmentation.FCN_ResNet50_Weights">FCN_ResNet50_Weights (class in torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fcos_resnet50_fpn.html#torchvision.models.detection.fcos_resnet50_fpn">fcos_resnet50_fpn() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.fcos_resnet50_fpn.html#torchvision.models.detection.FCOS_ResNet50_FPN_Weights">FCOS_ResNet50_FPN_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork">FeaturePyramidNetwork (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.FER2013.html#torchvision.datasets.FER2013">FER2013 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.FGVCAircraft.html#torchvision.datasets.FGVCAircraft">FGVCAircraft (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder.find_classes">find_classes() (torchvision.datasets.DatasetFolder method)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.five_crop.html#torchvision.transforms.functional.five_crop">five_crop() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.five_crop.html#torchvision.transforms.v2.functional.five_crop">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.FiveCrop.html#torchvision.transforms.FiveCrop">FiveCrop (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.FiveCrop.html#torchvision.transforms.v2.FiveCrop">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.Flickr30k.html#torchvision.datasets.Flickr30k">Flickr30k (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Flickr8k.html#torchvision.datasets.Flickr8k">Flickr8k (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.utils.flow_to_image.html#torchvision.utils.flow_to_image">flow_to_image() (in module torchvision.utils)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Flowers102.html#torchvision.datasets.Flowers102">Flowers102 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs">FlyingChairs (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D">FlyingThings3D (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Food101.html#torchvision.datasets.Food101">Food101 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.ops.DeformConv2d.html#torchvision.ops.DeformConv2d.forward">forward() (torchvision.ops.DeformConv2d method)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.ops.DropBlock2d.html#torchvision.ops.DropBlock2d.forward">(torchvision.ops.DropBlock2d method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.DropBlock3d.html#torchvision.ops.DropBlock3d.forward">(torchvision.ops.DropBlock3d method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork.forward">(torchvision.ops.FeaturePyramidNetwork method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.FrozenBatchNorm2d.html#torchvision.ops.FrozenBatchNorm2d.forward">(torchvision.ops.FrozenBatchNorm2d method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.MultiScaleRoIAlign.html#torchvision.ops.MultiScaleRoIAlign.forward">(torchvision.ops.MultiScaleRoIAlign method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.Permute.html#torchvision.ops.Permute.forward">(torchvision.ops.Permute method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.PSRoIAlign.html#torchvision.ops.PSRoIAlign.forward">(torchvision.ops.PSRoIAlign method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.PSRoIPool.html#torchvision.ops.PSRoIPool.forward">(torchvision.ops.PSRoIPool method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.RoIAlign.html#torchvision.ops.RoIAlign.forward">(torchvision.ops.RoIAlign method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.RoIPool.html#torchvision.ops.RoIPool.forward">(torchvision.ops.RoIPool method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.SqueezeExcitation.html#torchvision.ops.SqueezeExcitation.forward">(torchvision.ops.SqueezeExcitation method)</a>
+</li>
+        <li><a href="generated/torchvision.ops.StochasticDepth.html#torchvision.ops.StochasticDepth.forward">(torchvision.ops.StochasticDepth method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.AugMix.html#torchvision.transforms.AugMix.forward">(torchvision.transforms.AugMix method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.AutoAugment.html#torchvision.transforms.AutoAugment.forward">(torchvision.transforms.AutoAugment method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.CenterCrop.html#torchvision.transforms.CenterCrop.forward">(torchvision.transforms.CenterCrop method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.ColorJitter.html#torchvision.transforms.ColorJitter.forward">(torchvision.transforms.ColorJitter method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.ConvertImageDtype.html#torchvision.transforms.ConvertImageDtype.forward">(torchvision.transforms.ConvertImageDtype method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.ElasticTransform.html#torchvision.transforms.ElasticTransform.forward">(torchvision.transforms.ElasticTransform method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.FiveCrop.html#torchvision.transforms.FiveCrop.forward">(torchvision.transforms.FiveCrop method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.GaussianBlur.html#torchvision.transforms.GaussianBlur.forward">(torchvision.transforms.GaussianBlur method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.Grayscale.html#torchvision.transforms.Grayscale.forward">(torchvision.transforms.Grayscale method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.LinearTransformation.html#torchvision.transforms.LinearTransformation.forward">(torchvision.transforms.LinearTransformation method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.Normalize.html#torchvision.transforms.Normalize.forward">(torchvision.transforms.Normalize method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.Pad.html#torchvision.transforms.Pad.forward">(torchvision.transforms.Pad method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandAugment.html#torchvision.transforms.RandAugment.forward">(torchvision.transforms.RandAugment method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness.forward">(torchvision.transforms.RandomAdjustSharpness method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomAffine.html#torchvision.transforms.RandomAffine.forward">(torchvision.transforms.RandomAffine method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomApply.html#torchvision.transforms.RandomApply.forward">(torchvision.transforms.RandomApply method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast.forward">(torchvision.transforms.RandomAutocontrast method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomCrop.html#torchvision.transforms.RandomCrop.forward">(torchvision.transforms.RandomCrop method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomEqualize.html#torchvision.transforms.RandomEqualize.forward">(torchvision.transforms.RandomEqualize method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomErasing.html#torchvision.transforms.RandomErasing.forward">(torchvision.transforms.RandomErasing method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomGrayscale.html#torchvision.transforms.RandomGrayscale.forward">(torchvision.transforms.RandomGrayscale method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomHorizontalFlip.html#torchvision.transforms.RandomHorizontalFlip.forward">(torchvision.transforms.RandomHorizontalFlip method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomInvert.html#torchvision.transforms.RandomInvert.forward">(torchvision.transforms.RandomInvert method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomPerspective.html#torchvision.transforms.RandomPerspective.forward">(torchvision.transforms.RandomPerspective method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomPosterize.html#torchvision.transforms.RandomPosterize.forward">(torchvision.transforms.RandomPosterize method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop.forward">(torchvision.transforms.RandomResizedCrop method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomRotation.html#torchvision.transforms.RandomRotation.forward">(torchvision.transforms.RandomRotation method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomSolarize.html#torchvision.transforms.RandomSolarize.forward">(torchvision.transforms.RandomSolarize method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomVerticalFlip.html#torchvision.transforms.RandomVerticalFlip.forward">(torchvision.transforms.RandomVerticalFlip method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.Resize.html#torchvision.transforms.Resize.forward">(torchvision.transforms.Resize method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.TenCrop.html#torchvision.transforms.TenCrop.forward">(torchvision.transforms.TenCrop method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.TrivialAugmentWide.html#torchvision.transforms.TrivialAugmentWide.forward">(torchvision.transforms.TrivialAugmentWide method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.AugMix.html#torchvision.transforms.v2.AugMix.forward">(torchvision.transforms.v2.AugMix method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.AutoAugment.html#torchvision.transforms.v2.AutoAugment.forward">(torchvision.transforms.v2.AutoAugment method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose.forward">(torchvision.transforms.v2.Compose method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandAugment.html#torchvision.transforms.v2.RandAugment.forward">(torchvision.transforms.v2.RandAugment method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomApply.html#torchvision.transforms.v2.RandomApply.forward">(torchvision.transforms.v2.RandomApply method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomChoice.html#torchvision.transforms.v2.RandomChoice.forward">(torchvision.transforms.v2.RandomChoice method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomOrder.html#torchvision.transforms.v2.RandomOrder.forward">(torchvision.transforms.v2.RandomOrder method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes.forward">(torchvision.transforms.v2.SanitizeBoundingBoxes method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.TrivialAugmentWide.html#torchvision.transforms.v2.TrivialAugmentWide.forward">(torchvision.transforms.v2.TrivialAugmentWide method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.ops.FrozenBatchNorm2d.html#torchvision.ops.FrozenBatchNorm2d">FrozenBatchNorm2d (class in torchvision.ops)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="G">G</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.functional.gaussian_blur.html#torchvision.transforms.functional.gaussian_blur">gaussian_blur() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.gaussian_blur.html#torchvision.transforms.v2.functional.gaussian_blur">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.GaussianBlur.html#torchvision.transforms.GaussianBlur">GaussianBlur (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.GaussianBlur.html#torchvision.transforms.v2.GaussianBlur">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.ops.generalized_box_iou.html#torchvision.ops.generalized_box_iou">generalized_box_iou() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.generalized_box_iou_loss.html#torchvision.ops.generalized_box_iou_loss">generalized_box_iou_loss() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.get_dimensions.html#torchvision.transforms.functional.get_dimensions">get_dimensions() (in module torchvision.transforms.functional)</a>
+</li>
+      <li><a href="generated/torchvision.models.feature_extraction.get_graph_node_names.html#torchvision.models.feature_extraction.get_graph_node_names">get_graph_node_names() (in module torchvision.models.feature_extraction)</a>
+</li>
+      <li><a href="index.html#torchvision.get_image_backend">get_image_backend() (in module torchvision)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.get_image_num_channels.html#torchvision.transforms.functional.get_image_num_channels">get_image_num_channels() (in module torchvision.transforms.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.get_image_size.html#torchvision.transforms.functional.get_image_size">get_image_size() (in module torchvision.transforms.functional)</a>
+</li>
+      <li><a href="generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader.get_metadata">get_metadata() (torchvision.io.VideoReader method)</a>
+</li>
+      <li><a href="generated/torchvision.models.get_model.html#torchvision.models.get_model">get_model() (in module torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.models.get_model_weights.html#torchvision.models.get_model_weights">get_model_weights() (in module torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.AutoAugment.html#torchvision.transforms.AutoAugment.get_params">get_params() (torchvision.transforms.AutoAugment static method)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.ColorJitter.html#torchvision.transforms.ColorJitter.get_params">(torchvision.transforms.ColorJitter static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.GaussianBlur.html#torchvision.transforms.GaussianBlur.get_params">(torchvision.transforms.GaussianBlur static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomAffine.html#torchvision.transforms.RandomAffine.get_params">(torchvision.transforms.RandomAffine static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomCrop.html#torchvision.transforms.RandomCrop.get_params">(torchvision.transforms.RandomCrop static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomErasing.html#torchvision.transforms.RandomErasing.get_params">(torchvision.transforms.RandomErasing static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomPerspective.html#torchvision.transforms.RandomPerspective.get_params">(torchvision.transforms.RandomPerspective static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop.get_params">(torchvision.transforms.RandomResizedCrop static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.RandomRotation.html#torchvision.transforms.RandomRotation.get_params">(torchvision.transforms.RandomRotation static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.AutoAugment.html#torchvision.transforms.v2.AutoAugment.get_params">(torchvision.transforms.v2.AutoAugment static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.ColorJitter.html#torchvision.transforms.v2.ColorJitter.get_params">(torchvision.transforms.v2.ColorJitter static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.GaussianBlur.html#torchvision.transforms.v2.GaussianBlur.get_params">(torchvision.transforms.v2.GaussianBlur static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomAffine.html#torchvision.transforms.v2.RandomAffine.get_params">(torchvision.transforms.v2.RandomAffine static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop.get_params">(torchvision.transforms.v2.RandomCrop static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomErasing.html#torchvision.transforms.v2.RandomErasing.get_params">(torchvision.transforms.v2.RandomErasing static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomPerspective.html#torchvision.transforms.v2.RandomPerspective.get_params">(torchvision.transforms.v2.RandomPerspective static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.v2.RandomResizedCrop.get_params">(torchvision.transforms.v2.RandomResizedCrop static method)</a>
+</li>
+        <li><a href="generated/torchvision.transforms.v2.RandomRotation.html#torchvision.transforms.v2.RandomRotation.get_params">(torchvision.transforms.v2.RandomRotation static method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork.get_result_from_inner_blocks">get_result_from_inner_blocks() (torchvision.ops.FeaturePyramidNetwork method)</a>
+</li>
+      <li><a href="generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork.get_result_from_layer_blocks">get_result_from_layer_blocks() (torchvision.ops.FeaturePyramidNetwork method)</a>
+</li>
+      <li><a href="index.html#torchvision.get_video_backend">get_video_backend() (in module torchvision)</a>
+</li>
+      <li><a href="generated/torchvision.models.get_weight.html#torchvision.models.get_weight">get_weight() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.googlenet.html#torchvision.models.googlenet">googlenet() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.googlenet.html#torchvision.models.quantization.googlenet">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.googlenet.html#torchvision.models.quantization.GoogLeNet_QuantizedWeights">GoogLeNet_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.googlenet.html#torchvision.models.GoogLeNet_Weights">GoogLeNet_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.Grayscale.html#torchvision.transforms.Grayscale">Grayscale (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.Grayscale.html#torchvision.transforms.v2.Grayscale">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.GTSRB.html#torchvision.datasets.GTSRB">GTSRB (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="H">H</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.HD1K.html#torchvision.datasets.HD1K">HD1K (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.hflip.html#torchvision.transforms.functional.hflip">hflip() (in module torchvision.transforms.functional)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.HMDB51.html#torchvision.datasets.HMDB51">HMDB51 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.horizontal_flip.html#torchvision.transforms.v2.functional.horizontal_flip">horizontal_flip() (in module torchvision.transforms.v2.functional)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="I">I</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image">Image (class in torchvision.tv_tensors)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.ImageFolder.html#torchvision.datasets.ImageFolder">ImageFolder (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.ImageNet.html#torchvision.datasets.ImageNet">ImageNet (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode">ImageReadMode (class in torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.INaturalist.html#torchvision.datasets.INaturalist">INaturalist (class in torchvision.datasets)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.inception_v3.html#torchvision.models.inception_v3">inception_v3() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.inception_v3">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.Inception_V3_QuantizedWeights">Inception_V3_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.inception_v3.html#torchvision.models.Inception_V3_Weights">Inception_V3_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.InStereo2k.html#torchvision.datasets.InStereo2k">InStereo2k (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.invert.html#torchvision.transforms.functional.invert">invert() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.invert.html#torchvision.transforms.v2.functional.invert">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="K">K</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.keypointrcnn_resnet50_fpn">keypointrcnn_resnet50_fpn() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights">KeypointRCNN_ResNet50_FPN_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Kinetics.html#torchvision.datasets.Kinetics">Kinetics (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Kitti.html#torchvision.datasets.Kitti">Kitti (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.Kitti2012Stereo.html#torchvision.datasets.Kitti2012Stereo">Kitti2012Stereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Kitti2015Stereo.html#torchvision.datasets.Kitti2015Stereo">Kitti2015Stereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow">KittiFlow (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.KMNIST.html#torchvision.datasets.KMNIST">KMNIST (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="L">L</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.Lambda.html#torchvision.transforms.Lambda">Lambda (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.Lambda.html#torchvision.transforms.v2.Lambda">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.LFWPairs.html#torchvision.datasets.LFWPairs">LFWPairs (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.LFWPeople.html#torchvision.datasets.LFWPeople">LFWPeople (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.LinearTransformation.html#torchvision.transforms.LinearTransformation">LinearTransformation (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.LinearTransformation.html#torchvision.transforms.v2.LinearTransformation">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.models.list_models.html#torchvision.models.list_models">list_models() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html#torchvision.models.segmentation.lraspp_mobilenet_v3_large">lraspp_mobilenet_v3_large() (in module torchvision.models.segmentation)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html#torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights">LRASPP_MobileNet_V3_Large_Weights (class in torchvision.models.segmentation)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.LSUN.html#torchvision.datasets.LSUN">LSUN (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="M">M</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.DatasetFolder.html#torchvision.datasets.DatasetFolder.make_dataset">make_dataset() (torchvision.datasets.DatasetFolder static method)</a>
+</li>
+      <li><a href="generated/torchvision.utils.make_grid.html#torchvision.utils.make_grid">make_grid() (in module torchvision.utils)</a>
+</li>
+      <li><a href="generated/torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask">Mask (class in torchvision.tv_tensors)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html#torchvision.models.detection.maskrcnn_resnet50_fpn">maskrcnn_resnet50_fpn() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html#torchvision.models.detection.maskrcnn_resnet50_fpn_v2">maskrcnn_resnet50_fpn_v2() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html#torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights">MaskRCNN_ResNet50_FPN_V2_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html#torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights">MaskRCNN_ResNet50_FPN_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="generated/torchvision.ops.masks_to_boxes.html#torchvision.ops.masks_to_boxes">masks_to_boxes() (in module torchvision.ops)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.maxvit_t.html#torchvision.models.maxvit_t">maxvit_t() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.maxvit_t.html#torchvision.models.MaxVit_T_Weights">MaxVit_T_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.mc3_18.html#torchvision.models.video.mc3_18">mc3_18() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.mc3_18.html#torchvision.models.video.MC3_18_Weights">MC3_18_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Middlebury2014Stereo.html#torchvision.datasets.Middlebury2014Stereo">Middlebury2014Stereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.MixUp.html#torchvision.transforms.v2.MixUp">MixUp (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.ops.MLP.html#torchvision.ops.MLP">MLP (class in torchvision.ops)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mnasnet0_5.html#torchvision.models.mnasnet0_5">mnasnet0_5() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mnasnet0_5.html#torchvision.models.MNASNet0_5_Weights">MNASNet0_5_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mnasnet0_75.html#torchvision.models.mnasnet0_75">mnasnet0_75() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mnasnet0_75.html#torchvision.models.MNASNet0_75_Weights">MNASNet0_75_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mnasnet1_0.html#torchvision.models.mnasnet1_0">mnasnet1_0() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mnasnet1_0.html#torchvision.models.MNASNet1_0_Weights">MNASNet1_0_Weights (class in torchvision.models)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.mnasnet1_3.html#torchvision.models.mnasnet1_3">mnasnet1_3() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mnasnet1_3.html#torchvision.models.MNASNet1_3_Weights">MNASNet1_3_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.MNIST.html#torchvision.datasets.MNIST">MNIST (class in torchvision.datasets)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mobilenet_v2.html#torchvision.models.mobilenet_v2">mobilenet_v2() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.mobilenet_v2.html#torchvision.models.quantization.mobilenet_v2">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.mobilenet_v2.html#torchvision.models.quantization.MobileNet_V2_QuantizedWeights">MobileNet_V2_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mobilenet_v2.html#torchvision.models.MobileNet_V2_Weights">MobileNet_V2_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mobilenet_v3_large.html#torchvision.models.mobilenet_v3_large">mobilenet_v3_large() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.mobilenet_v3_large.html#torchvision.models.quantization.mobilenet_v3_large">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.mobilenet_v3_large.html#torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights">MobileNet_V3_Large_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights">MobileNet_V3_Large_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mobilenet_v3_small.html#torchvision.models.mobilenet_v3_small">mobilenet_v3_small() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.mobilenet_v3_small.html#torchvision.models.MobileNet_V3_Small_Weights">MobileNet_V3_Small_Weights (class in torchvision.models)</a>
+</li>
+      <li>
+    module
+
+      <ul>
+        <li><a href="index.html#module-torchvision">torchvision</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.MovingMNIST.html#torchvision.datasets.MovingMNIST">MovingMNIST (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.ops.MultiScaleRoIAlign.html#torchvision.ops.MultiScaleRoIAlign">MultiScaleRoIAlign (class in torchvision.ops)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.mvit_v1_b.html#torchvision.models.video.mvit_v1_b">mvit_v1_b() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.mvit_v1_b.html#torchvision.models.video.MViT_V1_B_Weights">MViT_V1_B_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.mvit_v2_s.html#torchvision.models.video.mvit_v2_s">mvit_v2_s() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.mvit_v2_s.html#torchvision.models.video.MViT_V2_S_Weights">MViT_V2_S_Weights (class in torchvision.models.video)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="N">N</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor.ndim">ndim (torchvision.tv_tensors.TVTensor property)</a>
+</li>
+      <li><a href="generated/torchvision.ops.nms.html#torchvision.ops.nms">nms() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.Normalize.html#torchvision.transforms.Normalize">Normalize (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.Normalize.html#torchvision.transforms.v2.Normalize">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.functional.normalize.html#torchvision.transforms.functional.normalize">normalize() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.normalize.html#torchvision.transforms.v2.functional.normalize">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="O">O</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.Omniglot.html#torchvision.datasets.Omniglot">Omniglot (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.OxfordIIITPet.html#torchvision.datasets.OxfordIIITPet">OxfordIIITPet (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="P">P</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.Pad.html#torchvision.transforms.Pad">Pad (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.Pad.html#torchvision.transforms.v2.Pad">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.pad.html#torchvision.transforms.functional.pad">pad() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.pad.html#torchvision.transforms.v2.functional.pad">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.PCAM.html#torchvision.datasets.PCAM">PCAM (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.ops.Permute.html#torchvision.ops.Permute">Permute (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.permute_channels.html#torchvision.transforms.v2.functional.permute_channels">permute_channels() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.perspective.html#torchvision.transforms.functional.perspective">perspective() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.perspective.html#torchvision.transforms.v2.functional.perspective">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.PhotoTour.html#torchvision.datasets.PhotoTour">PhotoTour (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.functional.pil_to_tensor.html#torchvision.transforms.functional.pil_to_tensor">pil_to_tensor() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.pil_to_tensor.html#torchvision.transforms.v2.functional.pil_to_tensor">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.PILToTensor.html#torchvision.transforms.PILToTensor">PILToTensor (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.PILToTensor.html#torchvision.transforms.v2.PILToTensor">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.datasets.Places365.html#torchvision.datasets.Places365">Places365 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.posterize.html#torchvision.transforms.functional.posterize">posterize() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.posterize.html#torchvision.transforms.v2.functional.posterize">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.ops.ps_roi_align.html#torchvision.ops.ps_roi_align">ps_roi_align() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.ps_roi_pool.html#torchvision.ops.ps_roi_pool">ps_roi_pool() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.PSRoIAlign.html#torchvision.ops.PSRoIAlign">PSRoIAlign (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.PSRoIPool.html#torchvision.ops.PSRoIPool">PSRoIPool (class in torchvision.ops)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="Q">Q</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.QMNIST.html#torchvision.datasets.QMNIST">QMNIST (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="R">R</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.video.r2plus1d_18.html#torchvision.models.video.r2plus1d_18">r2plus1d_18() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.r2plus1d_18.html#torchvision.models.video.R2Plus1D_18_Weights">R2Plus1D_18_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.r3d_18.html#torchvision.models.video.r3d_18">r3d_18() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.r3d_18.html#torchvision.models.video.R3D_18_Weights">R3D_18_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.optical_flow.raft_large.html#torchvision.models.optical_flow.raft_large">raft_large() (in module torchvision.models.optical_flow)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.optical_flow.raft_large.html#torchvision.models.optical_flow.Raft_Large_Weights">Raft_Large_Weights (class in torchvision.models.optical_flow)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.optical_flow.raft_small.html#torchvision.models.optical_flow.raft_small">raft_small() (in module torchvision.models.optical_flow)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.optical_flow.raft_small.html#torchvision.models.optical_flow.Raft_Small_Weights">Raft_Small_Weights (class in torchvision.models.optical_flow)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.RandAugment.html#torchvision.transforms.RandAugment">RandAugment (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandAugment.html#torchvision.transforms.v2.RandAugment">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness">RandomAdjustSharpness (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomAdjustSharpness.html#torchvision.transforms.v2.RandomAdjustSharpness">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomAffine.html#torchvision.transforms.RandomAffine">RandomAffine (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomAffine.html#torchvision.transforms.v2.RandomAffine">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomApply.html#torchvision.transforms.RandomApply">RandomApply (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomApply.html#torchvision.transforms.v2.RandomApply">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast">RandomAutocontrast (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomAutocontrast.html#torchvision.transforms.v2.RandomAutocontrast">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.RandomChannelPermutation.html#torchvision.transforms.v2.RandomChannelPermutation">RandomChannelPermutation (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.RandomChoice.html#torchvision.transforms.RandomChoice">RandomChoice (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomChoice.html#torchvision.transforms.v2.RandomChoice">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomCrop.html#torchvision.transforms.RandomCrop">RandomCrop (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomEqualize.html#torchvision.transforms.RandomEqualize">RandomEqualize (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomEqualize.html#torchvision.transforms.v2.RandomEqualize">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomErasing.html#torchvision.transforms.RandomErasing">RandomErasing (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomErasing.html#torchvision.transforms.v2.RandomErasing">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomGrayscale.html#torchvision.transforms.RandomGrayscale">RandomGrayscale (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomGrayscale.html#torchvision.transforms.v2.RandomGrayscale">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomHorizontalFlip.html#torchvision.transforms.RandomHorizontalFlip">RandomHorizontalFlip (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomHorizontalFlip.html#torchvision.transforms.v2.RandomHorizontalFlip">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomInvert.html#torchvision.transforms.RandomInvert">RandomInvert (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomInvert.html#torchvision.transforms.v2.RandomInvert">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.RandomIoUCrop.html#torchvision.transforms.v2.RandomIoUCrop">RandomIoUCrop (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.RandomOrder.html#torchvision.transforms.RandomOrder">RandomOrder (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomOrder.html#torchvision.transforms.v2.RandomOrder">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomPerspective.html#torchvision.transforms.RandomPerspective">RandomPerspective (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomPerspective.html#torchvision.transforms.v2.RandomPerspective">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.RandomPhotometricDistort.html#torchvision.transforms.v2.RandomPhotometricDistort">RandomPhotometricDistort (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.RandomPosterize.html#torchvision.transforms.RandomPosterize">RandomPosterize (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomPosterize.html#torchvision.transforms.v2.RandomPosterize">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.RandomResize.html#torchvision.transforms.v2.RandomResize">RandomResize (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop">RandomResizedCrop (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.v2.RandomResizedCrop">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomRotation.html#torchvision.transforms.RandomRotation">RandomRotation (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomRotation.html#torchvision.transforms.v2.RandomRotation">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.RandomShortestSize.html#torchvision.transforms.v2.RandomShortestSize">RandomShortestSize (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.RandomSolarize.html#torchvision.transforms.RandomSolarize">RandomSolarize (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomSolarize.html#torchvision.transforms.v2.RandomSolarize">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.RandomVerticalFlip.html#torchvision.transforms.RandomVerticalFlip">RandomVerticalFlip (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.RandomVerticalFlip.html#torchvision.transforms.v2.RandomVerticalFlip">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.RandomZoomOut.html#torchvision.transforms.v2.RandomZoomOut">RandomZoomOut (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.io.read_file.html#torchvision.io.read_file">read_file() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.read_image.html#torchvision.io.read_image">read_image() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.read_video.html#torchvision.io.read_video">read_video() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.read_video_timestamps.html#torchvision.io.read_video_timestamps">read_video_timestamps() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.register_kernel.html#torchvision.transforms.v2.functional.register_kernel">register_kernel() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_16gf.html#torchvision.models.regnet_x_16gf">regnet_x_16gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_16gf.html#torchvision.models.RegNet_X_16GF_Weights">RegNet_X_16GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_1_6gf.html#torchvision.models.regnet_x_1_6gf">regnet_x_1_6gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_1_6gf.html#torchvision.models.RegNet_X_1_6GF_Weights">RegNet_X_1_6GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_32gf.html#torchvision.models.regnet_x_32gf">regnet_x_32gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_32gf.html#torchvision.models.RegNet_X_32GF_Weights">RegNet_X_32GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_3_2gf.html#torchvision.models.regnet_x_3_2gf">regnet_x_3_2gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_3_2gf.html#torchvision.models.RegNet_X_3_2GF_Weights">RegNet_X_3_2GF_Weights (class in torchvision.models)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.regnet_x_400mf.html#torchvision.models.regnet_x_400mf">regnet_x_400mf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_400mf.html#torchvision.models.RegNet_X_400MF_Weights">RegNet_X_400MF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_800mf.html#torchvision.models.regnet_x_800mf">regnet_x_800mf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_800mf.html#torchvision.models.RegNet_X_800MF_Weights">RegNet_X_800MF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_8gf.html#torchvision.models.regnet_x_8gf">regnet_x_8gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_x_8gf.html#torchvision.models.RegNet_X_8GF_Weights">RegNet_X_8GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_128gf.html#torchvision.models.regnet_y_128gf">regnet_y_128gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_128gf.html#torchvision.models.RegNet_Y_128GF_Weights">RegNet_Y_128GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.regnet_y_16gf">regnet_y_16gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.RegNet_Y_16GF_Weights">RegNet_Y_16GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_1_6gf.html#torchvision.models.regnet_y_1_6gf">regnet_y_1_6gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_1_6gf.html#torchvision.models.RegNet_Y_1_6GF_Weights">RegNet_Y_1_6GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.regnet_y_32gf">regnet_y_32gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.RegNet_Y_32GF_Weights">RegNet_Y_32GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_3_2gf.html#torchvision.models.regnet_y_3_2gf">regnet_y_3_2gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_3_2gf.html#torchvision.models.RegNet_Y_3_2GF_Weights">RegNet_Y_3_2GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_400mf.html#torchvision.models.regnet_y_400mf">regnet_y_400mf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_400mf.html#torchvision.models.RegNet_Y_400MF_Weights">RegNet_Y_400MF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_800mf.html#torchvision.models.regnet_y_800mf">regnet_y_800mf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_800mf.html#torchvision.models.RegNet_Y_800MF_Weights">RegNet_Y_800MF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_8gf.html#torchvision.models.regnet_y_8gf">regnet_y_8gf() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.regnet_y_8gf.html#torchvision.models.RegNet_Y_8GF_Weights">RegNet_Y_8GF_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.ops.remove_small_boxes.html#torchvision.ops.remove_small_boxes">remove_small_boxes() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.RenderedSST2.html#torchvision.datasets.RenderedSST2">RenderedSST2 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.Resize.html#torchvision.transforms.Resize">Resize (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.Resize.html#torchvision.transforms.v2.Resize">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.resize.html#torchvision.transforms.functional.resize">resize() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.resize.html#torchvision.transforms.v2.functional.resize">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.functional.resized_crop.html#torchvision.transforms.functional.resized_crop">resized_crop() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.resized_crop.html#torchvision.transforms.v2.functional.resized_crop">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.resnet101.html#torchvision.models.resnet101">resnet101() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights">ResNet101_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet152.html#torchvision.models.resnet152">resnet152() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet152.html#torchvision.models.ResNet152_Weights">ResNet152_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet18.html#torchvision.models.resnet18">resnet18() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.resnet18.html#torchvision.models.quantization.resnet18">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.resnet18.html#torchvision.models.quantization.ResNet18_QuantizedWeights">ResNet18_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet18.html#torchvision.models.ResNet18_Weights">ResNet18_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet34.html#torchvision.models.resnet34">resnet34() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet34.html#torchvision.models.ResNet34_Weights">ResNet34_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet50.html#torchvision.models.resnet50">resnet50() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.resnet50.html#torchvision.models.quantization.resnet50">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.resnet50.html#torchvision.models.quantization.ResNet50_QuantizedWeights">ResNet50_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights">ResNet50_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnext101_32x8d.html#torchvision.models.resnext101_32x8d">resnext101_32x8d() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.resnext101_32x8d.html#torchvision.models.quantization.resnext101_32x8d">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.resnext101_32x8d.html#torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights">ResNeXt101_32X8D_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnext101_32x8d.html#torchvision.models.ResNeXt101_32X8D_Weights">ResNeXt101_32X8D_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnext101_64x4d.html#torchvision.models.resnext101_64x4d">resnext101_64x4d() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.resnext101_64x4d.html#torchvision.models.quantization.resnext101_64x4d">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.resnext101_64x4d.html#torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights">ResNeXt101_64X4D_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnext101_64x4d.html#torchvision.models.ResNeXt101_64X4D_Weights">ResNeXt101_64X4D_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnext50_32x4d.html#torchvision.models.resnext50_32x4d">resnext50_32x4d() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.resnext50_32x4d.html#torchvision.models.ResNeXt50_32X4D_Weights">ResNeXt50_32X4D_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html#torchvision.models.detection.retinanet_resnet50_fpn">retinanet_resnet50_fpn() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html#torchvision.models.detection.retinanet_resnet50_fpn_v2">retinanet_resnet50_fpn_v2() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html#torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights">RetinaNet_ResNet50_FPN_V2_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html#torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights">RetinaNet_ResNet50_FPN_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.rgb_to_grayscale.html#torchvision.transforms.functional.rgb_to_grayscale">rgb_to_grayscale() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html#torchvision.transforms.v2.functional.rgb_to_grayscale">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.ops.roi_align.html#torchvision.ops.roi_align">roi_align() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.roi_pool.html#torchvision.ops.roi_pool">roi_pool() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.RoIAlign.html#torchvision.ops.RoIAlign">RoIAlign (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.RoIPool.html#torchvision.ops.RoIPool">RoIPool (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.rotate.html#torchvision.transforms.functional.rotate">rotate() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.rotate.html#torchvision.transforms.v2.functional.rotate">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="S">S</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.video.s3d.html#torchvision.models.video.s3d">s3d() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.s3d.html#torchvision.models.video.S3D_Weights">S3D_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes">SanitizeBoundingBoxes (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.utils.save_image.html#torchvision.utils.save_image">save_image() (in module torchvision.utils)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.SBDataset.html#torchvision.datasets.SBDataset">SBDataset (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.SBU.html#torchvision.datasets.SBU">SBU (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.ScaleJitter.html#torchvision.transforms.v2.ScaleJitter">ScaleJitter (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.SceneFlowStereo.html#torchvision.datasets.SceneFlowStereo">SceneFlowStereo (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader.seek">seek() (torchvision.io.VideoReader method)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.SEMEION.html#torchvision.datasets.SEMEION">SEMEION (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader.set_current_stream">set_current_stream() (torchvision.io.VideoReader method)</a>
+</li>
+      <li><a href="index.html#torchvision.set_image_backend">set_image_backend() (in module torchvision)</a>
+</li>
+      <li><a href="generated/torchvision.tv_tensors.set_return_type.html#torchvision.tv_tensors.set_return_type">set_return_type (class in torchvision.tv_tensors)</a>
+</li>
+      <li><a href="index.html#torchvision.set_video_backend">set_video_backend() (in module torchvision)</a>
+</li>
+      <li><a href="generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor.shape">shape (torchvision.tv_tensors.TVTensor property)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.shufflenet_v2_x0_5">shufflenet_v2_x0_5() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html#torchvision.models.quantization.shufflenet_v2_x0_5">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html#torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights">ShuffleNet_V2_X0_5_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.ShuffleNet_V2_X0_5_Weights">ShuffleNet_V2_X0_5_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.shufflenet_v2_x1_0">shufflenet_v2_x1_0() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html#torchvision.models.quantization.shufflenet_v2_x1_0">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html#torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights">ShuffleNet_V2_X1_0_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.ShuffleNet_V2_X1_0_Weights">ShuffleNet_V2_X1_0_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.shufflenet_v2_x1_5">shufflenet_v2_x1_5() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html#torchvision.models.quantization.shufflenet_v2_x1_5">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html#torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights">ShuffleNet_V2_X1_5_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.ShuffleNet_V2_X1_5_Weights">ShuffleNet_V2_X1_5_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.shufflenet_v2_x2_0">shufflenet_v2_x2_0() (in module torchvision.models)</a>
+
+      <ul>
+        <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html#torchvision.models.quantization.shufflenet_v2_x2_0">(in module torchvision.models.quantization)</a>
+</li>
+      </ul></li>
+      <li><a href="models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html#torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights">ShuffleNet_V2_X2_0_QuantizedWeights (class in torchvision.models.quantization)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.ShuffleNet_V2_X2_0_Weights">ShuffleNet_V2_X2_0_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.ops.sigmoid_focal_loss.html#torchvision.ops.sigmoid_focal_loss">sigmoid_focal_loss() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.Sintel.html#torchvision.datasets.Sintel">Sintel (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.SintelStereo.html#torchvision.datasets.SintelStereo">SintelStereo (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.functional.solarize.html#torchvision.transforms.functional.solarize">solarize() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.solarize.html#torchvision.transforms.v2.functional.solarize">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.ops.SqueezeExcitation.html#torchvision.ops.SqueezeExcitation">SqueezeExcitation (class in torchvision.ops)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.squeezenet1_0.html#torchvision.models.squeezenet1_0">squeezenet1_0() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.squeezenet1_0.html#torchvision.models.SqueezeNet1_0_Weights">SqueezeNet1_0_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.squeezenet1_1.html#torchvision.models.squeezenet1_1">squeezenet1_1() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.squeezenet1_1.html#torchvision.models.SqueezeNet1_1_Weights">SqueezeNet1_1_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.ssd300_vgg16.html#torchvision.models.detection.ssd300_vgg16">ssd300_vgg16() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.ssd300_vgg16.html#torchvision.models.detection.SSD300_VGG16_Weights">SSD300_VGG16_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html#torchvision.models.detection.ssdlite320_mobilenet_v3_large">ssdlite320_mobilenet_v3_large() (in module torchvision.models.detection)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html#torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights">SSDLite320_MobileNet_V3_Large_Weights (class in torchvision.models.detection)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.StanfordCars.html#torchvision.datasets.StanfordCars">StanfordCars (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.STL10.html#torchvision.datasets.STL10">STL10 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.ops.stochastic_depth.html#torchvision.ops.stochastic_depth">stochastic_depth() (in module torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.ops.StochasticDepth.html#torchvision.ops.StochasticDepth">StochasticDepth (class in torchvision.ops)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.SUN397.html#torchvision.datasets.SUN397">SUN397 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.SVHN.html#torchvision.datasets.SVHN">SVHN (class in torchvision.datasets)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.swin3d_b.html#torchvision.models.video.swin3d_b">swin3d_b() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.swin3d_b.html#torchvision.models.video.Swin3D_B_Weights">Swin3D_B_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.swin3d_s.html#torchvision.models.video.swin3d_s">swin3d_s() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.swin3d_s.html#torchvision.models.video.Swin3D_S_Weights">Swin3D_S_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.swin3d_t.html#torchvision.models.video.swin3d_t">swin3d_t() (in module torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.video.swin3d_t.html#torchvision.models.video.Swin3D_T_Weights">Swin3D_T_Weights (class in torchvision.models.video)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_b.html#torchvision.models.swin_b">swin_b() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_b.html#torchvision.models.Swin_B_Weights">Swin_B_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_s.html#torchvision.models.swin_s">swin_s() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_s.html#torchvision.models.Swin_S_Weights">Swin_S_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_t.html#torchvision.models.swin_t">swin_t() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_t.html#torchvision.models.Swin_T_Weights">Swin_T_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_v2_b.html#torchvision.models.swin_v2_b">swin_v2_b() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_v2_b.html#torchvision.models.Swin_V2_B_Weights">Swin_V2_B_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_v2_s.html#torchvision.models.swin_v2_s">swin_v2_s() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_v2_s.html#torchvision.models.Swin_V2_S_Weights">Swin_V2_S_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_v2_t.html#torchvision.models.swin_v2_t">swin_v2_t() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.swin_v2_t.html#torchvision.models.Swin_V2_T_Weights">Swin_V2_T_Weights (class in torchvision.models)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="T">T</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.functional.ten_crop.html#torchvision.transforms.functional.ten_crop">ten_crop() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.ten_crop.html#torchvision.transforms.v2.functional.ten_crop">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.TenCrop.html#torchvision.transforms.TenCrop">TenCrop (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.TenCrop.html#torchvision.transforms.v2.TenCrop">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.functional.to_dtype.html#torchvision.transforms.v2.functional.to_dtype">to_dtype() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.to_grayscale.html#torchvision.transforms.functional.to_grayscale">to_grayscale() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.to_grayscale.html#torchvision.transforms.v2.functional.to_grayscale">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.functional.to_image.html#torchvision.transforms.v2.functional.to_image">to_image() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.to_pil_image.html#torchvision.transforms.functional.to_pil_image">to_pil_image() (in module torchvision.transforms.functional)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.functional.to_pil_image.html#torchvision.transforms.v2.functional.to_pil_image">(in module torchvision.transforms.v2.functional)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.functional.to_tensor.html#torchvision.transforms.v2.functional.to_tensor">to_tensor (class in torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.to_tensor.html#torchvision.transforms.functional.to_tensor">to_tensor() (in module torchvision.transforms.functional)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.v2.ToDtype.html#torchvision.transforms.v2.ToDtype">ToDtype (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.ToImage.html#torchvision.transforms.v2.ToImage">ToImage (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.ToPILImage.html#torchvision.transforms.ToPILImage">ToPILImage (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.ToPILImage.html#torchvision.transforms.v2.ToPILImage">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.v2.ToPureTensor.html#torchvision.transforms.v2.ToPureTensor">ToPureTensor (class in torchvision.transforms.v2)</a>
+</li>
+      <li>
+    torchvision
+
+      <ul>
+        <li><a href="index.html#module-torchvision">module</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.ToTensor.html#torchvision.transforms.ToTensor">ToTensor (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.ToTensor.html#torchvision.transforms.v2.ToTensor">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.transforms.TrivialAugmentWide.html#torchvision.transforms.TrivialAugmentWide">TrivialAugmentWide (class in torchvision.transforms)</a>
+
+      <ul>
+        <li><a href="generated/torchvision.transforms.v2.TrivialAugmentWide.html#torchvision.transforms.v2.TrivialAugmentWide">(class in torchvision.transforms.v2)</a>
+</li>
+      </ul></li>
+      <li><a href="generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor">TVTensor (class in torchvision.tv_tensors)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="U">U</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.datasets.UCF101.html#torchvision.datasets.UCF101">UCF101 (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html#torchvision.transforms.v2.functional.uniform_temporal_subsample">uniform_temporal_subsample() (in module torchvision.transforms.v2.functional)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.v2.UniformTemporalSubsample.html#torchvision.transforms.v2.UniformTemporalSubsample">UniformTemporalSubsample (class in torchvision.transforms.v2)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.USPS.html#torchvision.datasets.USPS">USPS (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="V">V</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.transforms.v2.functional.vertical_flip.html#torchvision.transforms.v2.functional.vertical_flip">vertical_flip() (in module torchvision.transforms.v2.functional)</a>
+</li>
+      <li><a href="generated/torchvision.transforms.functional.vflip.html#torchvision.transforms.functional.vflip">vflip() (in module torchvision.transforms.functional)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg11.html#torchvision.models.vgg11">vgg11() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg11_bn.html#torchvision.models.vgg11_bn">vgg11_bn() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg11_bn.html#torchvision.models.VGG11_BN_Weights">VGG11_BN_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg11.html#torchvision.models.VGG11_Weights">VGG11_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg13.html#torchvision.models.vgg13">vgg13() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg13_bn.html#torchvision.models.vgg13_bn">vgg13_bn() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg13_bn.html#torchvision.models.VGG13_BN_Weights">VGG13_BN_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg13.html#torchvision.models.VGG13_Weights">VGG13_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg16.html#torchvision.models.vgg16">vgg16() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg16_bn.html#torchvision.models.vgg16_bn">vgg16_bn() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg16_bn.html#torchvision.models.VGG16_BN_Weights">VGG16_BN_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg16.html#torchvision.models.VGG16_Weights">VGG16_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg19.html#torchvision.models.vgg19">vgg19() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg19_bn.html#torchvision.models.vgg19_bn">vgg19_bn() (in module torchvision.models)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.vgg19_bn.html#torchvision.models.VGG19_BN_Weights">VGG19_BN_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vgg19.html#torchvision.models.VGG19_Weights">VGG19_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video">Video (class in torchvision.tv_tensors)</a>
+</li>
+      <li><a href="generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader">VideoReader (class in torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.VisionDataset.html#torchvision.datasets.VisionDataset">VisionDataset (class in torchvision.datasets)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_b_16.html#torchvision.models.vit_b_16">vit_b_16() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_b_16.html#torchvision.models.ViT_B_16_Weights">ViT_B_16_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_b_32.html#torchvision.models.vit_b_32">vit_b_32() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_b_32.html#torchvision.models.ViT_B_32_Weights">ViT_B_32_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_h_14.html#torchvision.models.vit_h_14">vit_h_14() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_h_14.html#torchvision.models.ViT_H_14_Weights">ViT_H_14_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_l_16.html#torchvision.models.vit_l_16">vit_l_16() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_l_16.html#torchvision.models.ViT_L_16_Weights">ViT_L_16_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_l_32.html#torchvision.models.vit_l_32">vit_l_32() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.vit_l_32.html#torchvision.models.ViT_L_32_Weights">ViT_L_32_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.VOCDetection.html#torchvision.datasets.VOCDetection">VOCDetection (class in torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.VOCSegmentation.html#torchvision.datasets.VOCSegmentation">VOCSegmentation (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="W">W</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="models/generated/torchvision.models.wide_resnet101_2.html#torchvision.models.wide_resnet101_2">wide_resnet101_2() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.wide_resnet101_2.html#torchvision.models.Wide_ResNet101_2_Weights">Wide_ResNet101_2_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.wide_resnet50_2.html#torchvision.models.wide_resnet50_2">wide_resnet50_2() (in module torchvision.models)</a>
+</li>
+      <li><a href="models/generated/torchvision.models.wide_resnet50_2.html#torchvision.models.Wide_ResNet50_2_Weights">Wide_ResNet50_2_Weights (class in torchvision.models)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.WIDERFace.html#torchvision.datasets.WIDERFace">WIDERFace (class in torchvision.datasets)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="generated/torchvision.tv_tensors.wrap.html#torchvision.tv_tensors.wrap">wrap (class in torchvision.tv_tensors)</a>
+</li>
+      <li><a href="generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html#torchvision.datasets.wrap_dataset_for_transforms_v2">wrap_dataset_for_transforms_v2() (in module torchvision.datasets)</a>
+</li>
+      <li><a href="generated/torchvision.io.write_file.html#torchvision.io.write_file">write_file() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.write_jpeg.html#torchvision.io.write_jpeg">write_jpeg() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.write_png.html#torchvision.io.write_png">write_png() (in module torchvision.io)</a>
+</li>
+      <li><a href="generated/torchvision.io.write_video.html#torchvision.io.write_video">write_video() (in module torchvision.io)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/index.html b/0.16/index.html
new file mode 100644
index 00000000000..6258bb7771e
--- /dev/null
+++ b/0.16/index.html
@@ -0,0 +1,856 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>torchvision &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="Transforming and augmenting images" href="transforms.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="#">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>torchvision</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/index.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="torchvision">
+<h1>torchvision<a class="headerlink" href="#torchvision" title="Permalink to this heading">¶</a></h1>
+<p>This library is part of the <a class="reference external" href="http://pytorch.org/">PyTorch</a> project. PyTorch is an open source
+machine learning framework.</p>
+<p>Features described in this documentation are classified by release status:</p>
+<blockquote>
+<div><p><em>Stable:</em>  These features will be maintained long-term and there should generally
+be no major performance limitations or gaps in documentation.
+We also expect to maintain backwards compatibility (although
+breaking changes can happen and notice will be given one release ahead
+of time).</p>
+<p><em>Beta:</em>  Features are tagged as Beta because the API may change based on
+user feedback, because the performance needs to improve, or because
+coverage across operators is not yet complete. For Beta features, we are
+committing to seeing the feature through to the Stable classification.
+We are not, however, committing to backwards compatibility.</p>
+<p><em>Prototype:</em>  These features are typically not available as part of
+binary distributions like PyPI or Conda, except sometimes behind run-time
+flags, and are at an early stage for feedback and testing.</p>
+</div></blockquote>
+<p>The <a class="reference internal" href="#module-torchvision" title="torchvision"><code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision</span></code></a> package consists of popular datasets, model
+architectures, and common image transformations for computer vision.</p>
+<div class="toctree-wrapper compound">
+<p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#start-here">Start here</a></li>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#supported-input-types-and-conventions">Supported input types and conventions</a></li>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#v1-or-v2-which-one-should-i-use">V1 or V2? Which one should I use?</a></li>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#performance-considerations">Performance considerations</a></li>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#transform-classes-functionals-and-kernels">Transform classes, functionals, and kernels</a></li>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#torchscript-support">Torchscript support</a></li>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#v2-api-reference-recommended">V2 API reference - Recommended</a></li>
+<li class="toctree-l2"><a class="reference internal" href="transforms.html#v1-api-reference">V1 API Reference</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.Image.html">Image</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.Video.html">Video</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.BoundingBoxFormat.html">BoundingBoxFormat</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.BoundingBoxes.html">BoundingBoxes</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.Mask.html">Mask</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.TVTensor.html">TVTensor</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.set_return_type.html">set_return_type</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.tv_tensors.wrap.html">wrap</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="models.html#general-information-on-pre-trained-weights">General information on pre-trained weights</a></li>
+<li class="toctree-l2"><a class="reference internal" href="models.html#classification">Classification</a></li>
+<li class="toctree-l2"><a class="reference internal" href="models.html#semantic-segmentation">Semantic Segmentation</a></li>
+<li class="toctree-l2"><a class="reference internal" href="models.html#object-detection-instance-segmentation-and-person-keypoint-detection">Object Detection, Instance Segmentation and Person Keypoint Detection</a></li>
+<li class="toctree-l2"><a class="reference internal" href="models.html#video-classification">Video Classification</a></li>
+<li class="toctree-l2"><a class="reference internal" href="models.html#optical-flow">Optical Flow</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="datasets.html#built-in-datasets">Built-in datasets</a></li>
+<li class="toctree-l2"><a class="reference internal" href="datasets.html#base-classes-for-custom-datasets">Base classes for custom datasets</a></li>
+<li class="toctree-l2"><a class="reference internal" href="datasets.html#transforms-v2">Transforms v2</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.utils.draw_bounding_boxes.html">draw_bounding_boxes</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.utils.draw_segmentation_masks.html">draw_segmentation_masks</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.utils.draw_keypoints.html">draw_keypoints</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.utils.flow_to_image.html">flow_to_image</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.utils.make_grid.html">make_grid</a></li>
+<li class="toctree-l2"><a class="reference internal" href="generated/torchvision.utils.save_image.html">save_image</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="ops.html#detection-and-segmentation-operators">Detection and Segmentation Operators</a></li>
+<li class="toctree-l2"><a class="reference internal" href="ops.html#box-operators">Box Operators</a></li>
+<li class="toctree-l2"><a class="reference internal" href="ops.html#losses">Losses</a></li>
+<li class="toctree-l2"><a class="reference internal" href="ops.html#layers">Layers</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="io.html#images">Images</a></li>
+<li class="toctree-l2"><a class="reference internal" href="io.html#video">Video</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="feature_extraction.html#api-reference">API Reference</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="toctree-wrapper compound">
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+</div>
+<span class="target" id="module-torchvision"></span><dl class="py function">
+<dt class="sig sig-object py" id="torchvision.get_image_backend">
+<span class="sig-prename descclassname"><span class="pre">torchvision.</span></span><span class="sig-name descname"><span class="pre">get_image_backend</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/torchvision.html#get_image_backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.get_image_backend" title="Permalink to this definition">¶</a></dt>
+<dd><p>Gets the name of the package used to load images</p>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.get_video_backend">
+<span class="sig-prename descclassname"><span class="pre">torchvision.</span></span><span class="sig-name descname"><span class="pre">get_video_backend</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="_modules/torchvision.html#get_video_backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.get_video_backend" title="Permalink to this definition">¶</a></dt>
+<dd><p>Returns the currently active video backend used to decode videos.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Name of the video backend. one of {‘pyav’, ‘video_reader’}.</p>
+</dd>
+<dt class="field-even">Return type<span class="colon">:</span></dt>
+<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)">str</a></p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.set_image_backend">
+<span class="sig-prename descclassname"><span class="pre">torchvision.</span></span><span class="sig-name descname"><span class="pre">set_image_backend</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">backend</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/torchvision.html#set_image_backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.set_image_backend" title="Permalink to this definition">¶</a></dt>
+<dd><p>Specifies the package used to load images.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>backend</strong> (<em>string</em>) – Name of the image backend. one of {‘PIL’, ‘accimage’}.
+The <code class="xref py py-mod docutils literal notranslate"><span class="pre">accimage</span></code> package uses the Intel IPP library. It is
+generally faster than PIL, but does not support as many operations.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.set_video_backend">
+<span class="sig-prename descclassname"><span class="pre">torchvision.</span></span><span class="sig-name descname"><span class="pre">set_video_backend</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">backend</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/torchvision.html#set_video_backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.set_video_backend" title="Permalink to this definition">¶</a></dt>
+<dd><p>Specifies the package used to decode videos.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>backend</strong> (<em>string</em>) – Name of the video backend. one of {‘pyav’, ‘video_reader’}.
+The <code class="xref py py-mod docutils literal notranslate"><span class="pre">pyav</span></code> package uses the 3rd party PyAv library. It is a Pythonic
+binding for the FFmpeg libraries.
+The <code class="xref py py-mod docutils literal notranslate"><span class="pre">video_reader</span></code> package includes a native C++ implementation on
+top of FFMPEG libraries, and a python API of TorchScript custom operator.
+It generally decodes faster than <code class="xref py py-mod docutils literal notranslate"><span class="pre">pyav</span></code>, but is perhaps less robust.</p>
+</dd>
+</dl>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Building with FFMPEG is disabled by default in the latest <cite>main</cite>. If you want to use the ‘video_reader’
+backend, please compile torchvision from source.</p>
+</div>
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">set_video_backend</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the APIs that torchvision offers for videos, together with the..."><img alt="" src="_images/sphx_glr_plot_video_api_thumb.png" />
+<p><a class="reference internal" href="auto_examples/others/plot_video_api.html#sphx-glr-auto-examples-others-plot-video-api-py"><span class="std std-ref">Video API</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Video API</div>
+</div></div></dd></dl>
+
+<div class="toctree-wrapper compound">
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+</div>
+<section id="indices">
+<h2>Indices<a class="headerlink" href="#indices" title="Permalink to this heading">¶</a></h2>
+<ul class="simple">
+<li><p><a class="reference internal" href="genindex.html"><span class="std std-ref">Index</span></a></p></li>
+</ul>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="transforms.html" class="btn btn-neutral float-right" title="Transforming and augmenting images" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">torchvision</a><ul>
+<li><a class="reference internal" href="#indices">Indices</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/io.html b/0.16/io.html
new file mode 100644
index 00000000000..28c27092f28
--- /dev/null
+++ b/0.16/io.html
@@ -0,0 +1,800 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Decoding / Encoding images and videos &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="read_image" href="generated/torchvision.io.read_image.html" />
+    <link rel="prev" title="stochastic_depth" href="generated/torchvision.ops.stochastic_depth.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Decoding / Encoding images and videos</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/io.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="decoding-encoding-images-and-videos">
+<h1>Decoding / Encoding images and videos<a class="headerlink" href="#decoding-encoding-images-and-videos" title="Permalink to this heading">¶</a></h1>
+<p>The <code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.io</span></code> package provides functions for performing IO
+operations. They are currently specific to reading and writing images and
+videos.</p>
+<section id="images">
+<h2>Images<a class="headerlink" href="#images" title="Permalink to this heading">¶</a></h2>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.read_image.html#torchvision.io.read_image" title="torchvision.io.read_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">read_image</span></code></a>(path[, mode])</p></td>
+<td><p>Reads a JPEG or PNG image into a 3 dimensional RGB or grayscale Tensor.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.io.decode_image.html#torchvision.io.decode_image" title="torchvision.io.decode_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">decode_image</span></code></a>(input[, mode])</p></td>
+<td><p>Detects whether an image is a JPEG or PNG and performs the appropriate operation to decode the image into a 3 dimensional RGB or grayscale Tensor.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.encode_jpeg.html#torchvision.io.encode_jpeg" title="torchvision.io.encode_jpeg"><code class="xref py py-obj docutils literal notranslate"><span class="pre">encode_jpeg</span></code></a>(input[, quality])</p></td>
+<td><p>Takes an input tensor in CHW layout and returns a buffer with the contents of its corresponding JPEG file.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.io.decode_jpeg.html#torchvision.io.decode_jpeg" title="torchvision.io.decode_jpeg"><code class="xref py py-obj docutils literal notranslate"><span class="pre">decode_jpeg</span></code></a>(input[, mode, device])</p></td>
+<td><p>Decodes a JPEG image into a 3 dimensional RGB or grayscale Tensor.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.write_jpeg.html#torchvision.io.write_jpeg" title="torchvision.io.write_jpeg"><code class="xref py py-obj docutils literal notranslate"><span class="pre">write_jpeg</span></code></a>(input, filename[, quality])</p></td>
+<td><p>Takes an input tensor in CHW layout and saves it in a JPEG file.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.io.encode_png.html#torchvision.io.encode_png" title="torchvision.io.encode_png"><code class="xref py py-obj docutils literal notranslate"><span class="pre">encode_png</span></code></a>(input[, compression_level])</p></td>
+<td><p>Takes an input tensor in CHW layout and returns a buffer with the contents of its corresponding PNG file.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.decode_png.html#torchvision.io.decode_png" title="torchvision.io.decode_png"><code class="xref py py-obj docutils literal notranslate"><span class="pre">decode_png</span></code></a>(input[, mode])</p></td>
+<td><p>Decodes a PNG image into a 3 dimensional RGB or grayscale Tensor.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.io.write_png.html#torchvision.io.write_png" title="torchvision.io.write_png"><code class="xref py py-obj docutils literal notranslate"><span class="pre">write_png</span></code></a>(input, filename[, compression_level])</p></td>
+<td><p>Takes an input tensor in CHW layout (or HW in the case of grayscale images) and saves it in a PNG file.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.read_file.html#torchvision.io.read_file" title="torchvision.io.read_file"><code class="xref py py-obj docutils literal notranslate"><span class="pre">read_file</span></code></a>(path)</p></td>
+<td><p>Reads and outputs the bytes contents of a file as a uint8 Tensor with one dimension.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.io.write_file.html#torchvision.io.write_file" title="torchvision.io.write_file"><code class="xref py py-obj docutils literal notranslate"><span class="pre">write_file</span></code></a>(filename, data)</p></td>
+<td><p>Writes the contents of an uint8 tensor with one dimension to a file.</p></td>
+</tr>
+</tbody>
+</table>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.ImageReadMode.html#torchvision.io.ImageReadMode" title="torchvision.io.ImageReadMode"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ImageReadMode</span></code></a>(value)</p></td>
+<td><p>Support for various modes while reading images.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="video">
+<h2>Video<a class="headerlink" href="#video" title="Permalink to this heading">¶</a></h2>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.read_video.html#torchvision.io.read_video" title="torchvision.io.read_video"><code class="xref py py-obj docutils literal notranslate"><span class="pre">read_video</span></code></a>(filename[, start_pts, end_pts, ...])</p></td>
+<td><p>Reads a video from a file, returning both the video frames and the audio frames</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.io.read_video_timestamps.html#torchvision.io.read_video_timestamps" title="torchvision.io.read_video_timestamps"><code class="xref py py-obj docutils literal notranslate"><span class="pre">read_video_timestamps</span></code></a>(filename[, pts_unit])</p></td>
+<td><p>List the video frames timestamps.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.write_video.html#torchvision.io.write_video" title="torchvision.io.write_video"><code class="xref py py-obj docutils literal notranslate"><span class="pre">write_video</span></code></a>(filename, video_array, fps[, ...])</p></td>
+<td><p>Writes a 4d tensor in [T, H, W, C] format in a video file</p></td>
+</tr>
+</tbody>
+</table>
+<section id="fine-grained-video-api">
+<h3>Fine-grained video API<a class="headerlink" href="#fine-grained-video-api" title="Permalink to this heading">¶</a></h3>
+<p>In addition to the <code class="xref py py-mod docutils literal notranslate"><span class="pre">read_video</span></code> function, we provide a high-performance
+lower-level API for more fine-grained control compared to the <code class="xref py py-mod docutils literal notranslate"><span class="pre">read_video</span></code> function.
+It does all this whilst fully supporting torchscript.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The fine-grained video API is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.io.VideoReader.html#torchvision.io.VideoReader" title="torchvision.io.VideoReader"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VideoReader</span></code></a>([src, stream, num_threads, path])</p></td>
+<td><p>Fine-grained video-reading API.</p></td>
+</tr>
+</tbody>
+</table>
+<p>Example of inspecting a video:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torchvision</span>
+<span class="n">video_path</span> <span class="o">=</span> <span class="s2">&quot;path to a test video&quot;</span>
+<span class="c1"># Constructor allocates memory and a threaded decoder</span>
+<span class="c1"># instance per video. At the moment it takes two arguments:</span>
+<span class="c1"># path to the video file, and a wanted stream.</span>
+<span class="n">reader</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">io</span><span class="o">.</span><span class="n">VideoReader</span><span class="p">(</span><span class="n">video_path</span><span class="p">,</span> <span class="s2">&quot;video&quot;</span><span class="p">)</span>
+
+<span class="c1"># The information about the video can be retrieved using the</span>
+<span class="c1"># `get_metadata()` method. It returns a dictionary for every stream, with</span>
+<span class="c1"># duration and other relevant metadata (often frame rate)</span>
+<span class="n">reader_md</span> <span class="o">=</span> <span class="n">reader</span><span class="o">.</span><span class="n">get_metadata</span><span class="p">()</span>
+
+<span class="c1"># metadata is structured as a dict of dicts with following structure</span>
+<span class="c1"># {&quot;stream_type&quot;: {&quot;attribute&quot;: [attribute per stream]}}</span>
+<span class="c1">#</span>
+<span class="c1"># following would print out the list of frame rates for every present video stream</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">reader_md</span><span class="p">[</span><span class="s2">&quot;video&quot;</span><span class="p">][</span><span class="s2">&quot;fps&quot;</span><span class="p">])</span>
+
+<span class="c1"># we explicitly select the stream we would like to operate on. In</span>
+<span class="c1"># the constructor we select a default video stream, but</span>
+<span class="c1"># in practice, we can set whichever stream we would like</span>
+<span class="n">video</span><span class="o">.</span><span class="n">set_current_stream</span><span class="p">(</span><span class="s2">&quot;video:0&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+</section>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.io.read_image.html" class="btn btn-neutral float-right" title="read_image" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.ops.stochastic_depth.html" class="btn btn-neutral" title="stochastic_depth" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Decoding / Encoding images and videos</a><ul>
+<li><a class="reference internal" href="#images">Images</a><ul>
+</ul>
+</li>
+<li><a class="reference internal" href="#video">Video</a><ul>
+<li><a class="reference internal" href="#fine-grained-video-api">Fine-grained video API</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models.html b/0.16/models.html
new file mode 100644
index 00000000000..0b3afffef3e
--- /dev/null
+++ b/0.16/models.html
@@ -0,0 +1,2459 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Models and pre-trained weights &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="get_model" href="generated/torchvision.models.get_model.html" />
+    <link rel="prev" title="wrap" href="generated/torchvision.tv_tensors.wrap.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Models and pre-trained weights</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/models.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="models-and-pre-trained-weights">
+<span id="models"></span><h1>Models and pre-trained weights<a class="headerlink" href="#models-and-pre-trained-weights" title="Permalink to this heading">¶</a></h1>
+<p>The <code class="docutils literal notranslate"><span class="pre">torchvision.models</span></code> subpackage contains definitions of models for addressing
+different tasks, including: image classification, pixelwise semantic
+segmentation, object detection, instance segmentation, person
+keypoint detection, video classification, and optical flow.</p>
+<section id="general-information-on-pre-trained-weights">
+<h2>General information on pre-trained weights<a class="headerlink" href="#general-information-on-pre-trained-weights" title="Permalink to this heading">¶</a></h2>
+<p>TorchVision offers pre-trained weights for every provided architecture, using
+the PyTorch <a class="reference external" href="https://pytorch.org/docs/stable/hub.html#module-torch.hub" title="(in PyTorch v2.0)"><code class="xref py py-mod docutils literal notranslate"><span class="pre">torch.hub</span></code></a>. Instancing a pre-trained model will download its
+weights to a cache directory. This directory can be set using the <cite>TORCH_HOME</cite>
+environment variable. See <a class="reference external" href="https://pytorch.org/docs/stable/hub.html#torch.hub.load_state_dict_from_url" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.hub.load_state_dict_from_url()</span></code></a> for details.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The pre-trained models provided in this library may have their own licenses or
+terms and conditions derived from the dataset used for training. It is your
+responsibility to determine whether you have permission to use the models for
+your use case.</p>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Backward compatibility is guaranteed for loading a serialized
+<code class="docutils literal notranslate"><span class="pre">state_dict</span></code> to the model created using old PyTorch version.
+On the contrary, loading entire saved models or serialized
+<code class="docutils literal notranslate"><span class="pre">ScriptModules</span></code> (serialized using older versions of PyTorch)
+may not preserve the historic behaviour. Refer to the following
+<a class="reference external" href="https://pytorch.org/docs/stable/notes/serialization.html#id6">documentation</a></p>
+</div>
+<section id="initializing-pre-trained-models">
+<h3>Initializing pre-trained models<a class="headerlink" href="#initializing-pre-trained-models" title="Permalink to this heading">¶</a></h3>
+<p>As of v0.13, TorchVision offers a new <a class="reference external" href="https://pytorch.org/blog/introducing-torchvision-new-multi-weight-support-api/">Multi-weight support API</a>
+for loading different weights to the existing model builder methods:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+
+<span class="c1"># Old weights with accuracy 76.130%</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">)</span>
+
+<span class="c1"># New weights with accuracy 80.858%</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V2</span><span class="p">)</span>
+
+<span class="c1"># Best available weights (currently alias for IMAGENET1K_V2)</span>
+<span class="c1"># Note that these weights may change across versions</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+
+<span class="c1"># Strings are also supported</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="s2">&quot;IMAGENET1K_V2&quot;</span><span class="p">)</span>
+
+<span class="c1"># No weights - random initialization</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Migrating to the new API is very straightforward. The following method calls between the 2 APIs are all equivalent:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.models</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+
+<span class="c1"># Using pretrained weights:</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">IMAGENET1K_V1</span><span class="p">)</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="s2">&quot;IMAGENET1K_V1&quot;</span><span class="p">)</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">pretrained</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>  <span class="c1"># deprecated</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="kc">True</span><span class="p">)</span>  <span class="c1"># deprecated</span>
+
+<span class="c1"># Using no weights:</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+<span class="n">resnet50</span><span class="p">()</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="n">pretrained</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>  <span class="c1"># deprecated</span>
+<span class="n">resnet50</span><span class="p">(</span><span class="kc">False</span><span class="p">)</span>  <span class="c1"># deprecated</span>
+</pre></div>
+</div>
+<p>Note that the <code class="docutils literal notranslate"><span class="pre">pretrained</span></code> parameter is now deprecated, using it will emit warnings and will be removed on v0.15.</p>
+</section>
+<section id="using-the-pre-trained-models">
+<h3>Using the pre-trained models<a class="headerlink" href="#using-the-pre-trained-models" title="Permalink to this heading">¶</a></h3>
+<p>Before using the pre-trained models, one must preprocess the image
+(resize with right resolution/interpolation, apply inference transforms,
+rescale the values etc). There is no standard way to do this as it depends on
+how a given model was trained. It can vary across model families, variants or
+even weight versions. Using the correct preprocessing method is critical and
+failing to do so may lead to decreased accuracy or incorrect outputs.</p>
+<p>All the necessary information for the inference transforms of each pre-trained
+model is provided on its weights documentation. To simplify inference, TorchVision
+bundles the necessary preprocessing transforms into each model weight. These are
+accessible via the <code class="docutils literal notranslate"><span class="pre">weight.transforms</span></code> attribute:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Initialize the Weight Transforms</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">preprocess</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<span class="c1"># Apply it to the input image</span>
+<span class="n">img_transformed</span> <span class="o">=</span> <span class="n">preprocess</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Some models use modules which have different training and evaluation
+behavior, such as batch normalization. To switch between these modes, use
+<code class="docutils literal notranslate"><span class="pre">model.train()</span></code> or <code class="docutils literal notranslate"><span class="pre">model.eval()</span></code> as appropriate. See
+<a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.train" title="(in PyTorch v2.0)"><code class="xref py py-meth docutils literal notranslate"><span class="pre">train()</span></code></a> or <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module.eval" title="(in PyTorch v2.0)"><code class="xref py py-meth docutils literal notranslate"><span class="pre">eval()</span></code></a> for details.</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Initialize model</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">)</span>
+
+<span class="c1"># Set model to eval mode</span>
+<span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+</pre></div>
+</div>
+</section>
+<section id="listing-and-retrieving-available-models">
+<h3>Listing and retrieving available models<a class="headerlink" href="#listing-and-retrieving-available-models" title="Permalink to this heading">¶</a></h3>
+<p>As of v0.14, TorchVision offers a new mechanism which allows listing and
+retrieving models and weights by their names. Here are a few examples on how to
+use them:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># List available models</span>
+<span class="n">all_models</span> <span class="o">=</span> <span class="n">list_models</span><span class="p">()</span>
+<span class="n">classification_models</span> <span class="o">=</span> <span class="n">list_models</span><span class="p">(</span><span class="n">module</span><span class="o">=</span><span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="p">)</span>
+
+<span class="c1"># Initialize models</span>
+<span class="n">m1</span> <span class="o">=</span> <span class="n">get_model</span><span class="p">(</span><span class="s2">&quot;mobilenet_v3_large&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+<span class="n">m2</span> <span class="o">=</span> <span class="n">get_model</span><span class="p">(</span><span class="s2">&quot;quantized_mobilenet_v3_large&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="s2">&quot;DEFAULT&quot;</span><span class="p">)</span>
+
+<span class="c1"># Fetch weights</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">get_weight</span><span class="p">(</span><span class="s2">&quot;MobileNet_V3_Large_QuantizedWeights.DEFAULT&quot;</span><span class="p">)</span>
+<span class="k">assert</span> <span class="n">weights</span> <span class="o">==</span> <span class="n">MobileNet_V3_Large_QuantizedWeights</span><span class="o">.</span><span class="n">DEFAULT</span>
+
+<span class="n">weights_enum</span> <span class="o">=</span> <span class="n">get_model_weights</span><span class="p">(</span><span class="s2">&quot;quantized_mobilenet_v3_large&quot;</span><span class="p">)</span>
+<span class="k">assert</span> <span class="n">weights_enum</span> <span class="o">==</span> <span class="n">MobileNet_V3_Large_QuantizedWeights</span>
+
+<span class="n">weights_enum2</span> <span class="o">=</span> <span class="n">get_model_weights</span><span class="p">(</span><span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">quantization</span><span class="o">.</span><span class="n">mobilenet_v3_large</span><span class="p">)</span>
+<span class="k">assert</span> <span class="n">weights_enum</span> <span class="o">==</span> <span class="n">weights_enum2</span>
+</pre></div>
+</div>
+<p>Here are the available public functions to retrieve models and their corresponding weights:</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.get_model.html#torchvision.models.get_model" title="torchvision.models.get_model"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_model</span></code></a>(name, **config)</p></td>
+<td><p>Gets the model name and configuration and returns an instantiated model.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.get_model_weights.html#torchvision.models.get_model_weights" title="torchvision.models.get_model_weights"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_model_weights</span></code></a>(name)</p></td>
+<td><p>Returns the weights enum class associated to the given model.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.get_weight.html#torchvision.models.get_weight" title="torchvision.models.get_weight"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_weight</span></code></a>(name)</p></td>
+<td><p>Gets the weights enum value by its full name.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.list_models.html#torchvision.models.list_models" title="torchvision.models.list_models"><code class="xref py py-obj docutils literal notranslate"><span class="pre">list_models</span></code></a>([module, include, exclude])</p></td>
+<td><p>Returns a list with the names of registered models.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="using-models-from-hub">
+<h3>Using models from Hub<a class="headerlink" href="#using-models-from-hub" title="Permalink to this heading">¶</a></h3>
+<p>Most pre-trained models can be accessed directly via PyTorch Hub without having TorchVision installed:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="c1"># Option 1: passing weights param as string</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">hub</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="s2">&quot;pytorch/vision&quot;</span><span class="p">,</span> <span class="s2">&quot;resnet50&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="s2">&quot;IMAGENET1K_V2&quot;</span><span class="p">)</span>
+
+<span class="c1"># Option 2: passing weights param as enum</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">hub</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="s2">&quot;pytorch/vision&quot;</span><span class="p">,</span> <span class="s2">&quot;get_weight&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="s2">&quot;ResNet50_Weights.IMAGENET1K_V2&quot;</span><span class="p">)</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">hub</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="s2">&quot;pytorch/vision&quot;</span><span class="p">,</span> <span class="s2">&quot;resnet50&quot;</span><span class="p">,</span> <span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>You can also retrieve all the available weights of a specific model via PyTorch Hub by doing:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">torch</span>
+
+<span class="n">weight_enum</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">hub</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="s2">&quot;pytorch/vision&quot;</span><span class="p">,</span> <span class="s2">&quot;get_model_weights&quot;</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="s2">&quot;resnet50&quot;</span><span class="p">)</span>
+<span class="nb">print</span><span class="p">([</span><span class="n">weight</span> <span class="k">for</span> <span class="n">weight</span> <span class="ow">in</span> <span class="n">weight_enum</span><span class="p">])</span>
+</pre></div>
+</div>
+<p>The only exception to the above are the detection models included on
+<code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.models.detection</span></code>. These models require TorchVision
+to be installed because they depend on custom C++ operators.</p>
+</section>
+</section>
+<section id="classification">
+<h2>Classification<a class="headerlink" href="#classification" title="Permalink to this heading">¶</a></h2>
+<p>The following classification models are available, with or without pre-trained
+weights:</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/alexnet.html">AlexNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/convnext.html">ConvNeXt</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/densenet.html">DenseNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/efficientnet.html">EfficientNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/efficientnetv2.html">EfficientNetV2</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/googlenet.html">GoogLeNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/inception.html">Inception V3</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/maxvit.html">MaxVit</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/mnasnet.html">MNASNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/mobilenetv2.html">MobileNet V2</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/mobilenetv3.html">MobileNet V3</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/regnet.html">RegNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/resnet.html">ResNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/resnext.html">ResNeXt</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/shufflenetv2.html">ShuffleNet V2</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/squeezenet.html">SqueezeNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/swin_transformer.html">SwinTransformer</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/vgg.html">VGG</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/vision_transformer.html">VisionTransformer</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/wide_resnet.html">Wide ResNet</a></li>
+</ul>
+</div>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>Here is an example of how to use the pre-trained image classification models:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_image</span>
+<span class="kn">from</span> <span class="nn">torchvision.models</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_Weights</span>
+
+<span class="n">img</span> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="s2">&quot;test/assets/encode_jpeg/grace_hopper_517x606.jpg&quot;</span><span class="p">)</span>
+
+<span class="c1"># Step 1: Initialize model with the best available weights</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet50_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">)</span>
+<span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+<span class="c1"># Step 2: Initialize the inference transforms</span>
+<span class="n">preprocess</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<span class="c1"># Step 3: Apply inference preprocessing transforms</span>
+<span class="n">batch</span> <span class="o">=</span> <span class="n">preprocess</span><span class="p">(</span><span class="n">img</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+<span class="c1"># Step 4: Use the model and print the predicted category</span>
+<span class="n">prediction</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">batch</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+<span class="n">class_id</span> <span class="o">=</span> <span class="n">prediction</span><span class="o">.</span><span class="n">argmax</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+<span class="n">score</span> <span class="o">=</span> <span class="n">prediction</span><span class="p">[</span><span class="n">class_id</span><span class="p">]</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+<span class="n">category_name</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">][</span><span class="n">class_id</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">category_name</span><span class="si">}</span><span class="s2">: </span><span class="si">{</span><span class="mi">100</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="n">score</span><span class="si">:</span><span class="s2">.1f</span><span class="si">}</span><span class="s2">%&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>The classes of the pre-trained model outputs can be found at <code class="docutils literal notranslate"><span class="pre">weights.meta[&quot;categories&quot;]</span></code>.</p>
+<section id="table-of-all-available-classification-weights">
+<h3>Table of all available classification weights<a class="headerlink" href="#table-of-all-available-classification-weights" title="Permalink to this heading">¶</a></h3>
+<p>Accuracies are reported on ImageNet-1K using single crops:</p>
+<table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Acc&#64;1</strong></p></th>
+<th class="head"><p><strong>Acc&#64;5</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.alexnet.html#torchvision.models.AlexNet_Weights" title="torchvision.models.AlexNet_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">AlexNet_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>56.522</p></td>
+<td><p>79.066</p></td>
+<td><p>61.1M</p></td>
+<td><p>0.71</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.convnext_base.html#torchvision.models.ConvNeXt_Base_Weights" title="torchvision.models.ConvNeXt_Base_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>84.062</p></td>
+<td><p>96.87</p></td>
+<td><p>88.6M</p></td>
+<td><p>15.36</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.convnext_large.html#torchvision.models.ConvNeXt_Large_Weights" title="torchvision.models.ConvNeXt_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>84.414</p></td>
+<td><p>96.976</p></td>
+<td><p>197.8M</p></td>
+<td><p>34.36</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.convnext_small.html#torchvision.models.ConvNeXt_Small_Weights" title="torchvision.models.ConvNeXt_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.616</p></td>
+<td><p>96.65</p></td>
+<td><p>50.2M</p></td>
+<td><p>8.68</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.convnext_tiny.html#torchvision.models.ConvNeXt_Tiny_Weights" title="torchvision.models.ConvNeXt_Tiny_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>82.52</p></td>
+<td><p>96.146</p></td>
+<td><p>28.6M</p></td>
+<td><p>4.46</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.densenet121.html#torchvision.models.DenseNet121_Weights" title="torchvision.models.DenseNet121_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet121_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>74.434</p></td>
+<td><p>91.972</p></td>
+<td><p>8.0M</p></td>
+<td><p>2.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.densenet161.html#torchvision.models.DenseNet161_Weights" title="torchvision.models.DenseNet161_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet161_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>77.138</p></td>
+<td><p>93.56</p></td>
+<td><p>28.7M</p></td>
+<td><p>7.73</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.densenet169.html#torchvision.models.DenseNet169_Weights" title="torchvision.models.DenseNet169_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet169_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>75.6</p></td>
+<td><p>92.806</p></td>
+<td><p>14.1M</p></td>
+<td><p>3.36</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.densenet201.html#torchvision.models.DenseNet201_Weights" title="torchvision.models.DenseNet201_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet201_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>76.896</p></td>
+<td><p>93.37</p></td>
+<td><p>20.0M</p></td>
+<td><p>4.29</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b0.html#torchvision.models.EfficientNet_B0_Weights" title="torchvision.models.EfficientNet_B0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>77.692</p></td>
+<td><p>93.532</p></td>
+<td><p>5.3M</p></td>
+<td><p>0.39</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b1.html#torchvision.models.EfficientNet_B1_Weights" title="torchvision.models.EfficientNet_B1_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>78.642</p></td>
+<td><p>94.186</p></td>
+<td><p>7.8M</p></td>
+<td><p>0.69</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b1.html#torchvision.models.EfficientNet_B1_Weights" title="torchvision.models.EfficientNet_B1_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>79.838</p></td>
+<td><p>94.934</p></td>
+<td><p>7.8M</p></td>
+<td><p>0.69</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-lr-wd-crop-tuning">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b2.html#torchvision.models.EfficientNet_B2_Weights" title="torchvision.models.EfficientNet_B2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>80.608</p></td>
+<td><p>95.31</p></td>
+<td><p>9.1M</p></td>
+<td><p>1.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b3.html#torchvision.models.EfficientNet_B3_Weights" title="torchvision.models.EfficientNet_B3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>82.008</p></td>
+<td><p>96.054</p></td>
+<td><p>12.2M</p></td>
+<td><p>1.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b4.html#torchvision.models.EfficientNet_B4_Weights" title="torchvision.models.EfficientNet_B4_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.384</p></td>
+<td><p>96.594</p></td>
+<td><p>19.3M</p></td>
+<td><p>4.39</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b5.html#torchvision.models.EfficientNet_B5_Weights" title="torchvision.models.EfficientNet_B5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.444</p></td>
+<td><p>96.628</p></td>
+<td><p>30.4M</p></td>
+<td><p>10.27</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b6.html#torchvision.models.EfficientNet_B6_Weights" title="torchvision.models.EfficientNet_B6_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>84.008</p></td>
+<td><p>96.916</p></td>
+<td><p>43.0M</p></td>
+<td><p>19.07</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_b7.html#torchvision.models.EfficientNet_B7_Weights" title="torchvision.models.EfficientNet_B7_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>84.122</p></td>
+<td><p>96.908</p></td>
+<td><p>66.3M</p></td>
+<td><p>37.75</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_v2_l.html#torchvision.models.EfficientNet_V2_L_Weights" title="torchvision.models.EfficientNet_V2_L_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>85.808</p></td>
+<td><p>97.788</p></td>
+<td><p>118.5M</p></td>
+<td><p>56.08</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_v2_m.html#torchvision.models.EfficientNet_V2_M_Weights" title="torchvision.models.EfficientNet_V2_M_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>85.112</p></td>
+<td><p>97.156</p></td>
+<td><p>54.1M</p></td>
+<td><p>24.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.efficientnet_v2_s.html#torchvision.models.EfficientNet_V2_S_Weights" title="torchvision.models.EfficientNet_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>84.228</p></td>
+<td><p>96.878</p></td>
+<td><p>21.5M</p></td>
+<td><p>8.37</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.googlenet.html#torchvision.models.GoogLeNet_Weights" title="torchvision.models.GoogLeNet_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>69.778</p></td>
+<td><p>89.53</p></td>
+<td><p>6.6M</p></td>
+<td><p>1.5</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#googlenet">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.inception_v3.html#torchvision.models.Inception_V3_Weights" title="torchvision.models.Inception_V3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>77.294</p></td>
+<td><p>93.45</p></td>
+<td><p>27.2M</p></td>
+<td><p>5.71</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#inception-v3">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.mnasnet0_5.html#torchvision.models.MNASNet0_5_Weights" title="torchvision.models.MNASNet0_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_5_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>67.734</p></td>
+<td><p>87.49</p></td>
+<td><p>2.2M</p></td>
+<td><p>0.1</p></td>
+<td><p><a class="reference external" href="https://github.com/1e100/mnasnet_trainer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.mnasnet0_75.html#torchvision.models.MNASNet0_75_Weights" title="torchvision.models.MNASNet0_75_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_75_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>71.18</p></td>
+<td><p>90.496</p></td>
+<td><p>3.2M</p></td>
+<td><p>0.21</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/6019">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.mnasnet1_0.html#torchvision.models.MNASNet1_0_Weights" title="torchvision.models.MNASNet1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_0_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>73.456</p></td>
+<td><p>91.51</p></td>
+<td><p>4.4M</p></td>
+<td><p>0.31</p></td>
+<td><p><a class="reference external" href="https://github.com/1e100/mnasnet_trainer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.mnasnet1_3.html#torchvision.models.MNASNet1_3_Weights" title="torchvision.models.MNASNet1_3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_3_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>76.506</p></td>
+<td><p>93.522</p></td>
+<td><p>6.3M</p></td>
+<td><p>0.53</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/6019">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.maxvit_t.html#torchvision.models.MaxVit_T_Weights" title="torchvision.models.MaxVit_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaxVit_T_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.7</p></td>
+<td><p>96.722</p></td>
+<td><p>30.9M</p></td>
+<td><p>5.56</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#maxvit">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.mobilenet_v2.html#torchvision.models.MobileNet_V2_Weights" title="torchvision.models.MobileNet_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>71.878</p></td>
+<td><p>90.286</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.mobilenet_v2.html#torchvision.models.MobileNet_V2_Weights" title="torchvision.models.MobileNet_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>72.154</p></td>
+<td><p>90.822</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>74.042</p></td>
+<td><p>91.34</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.22</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>75.274</p></td>
+<td><p>92.566</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.22</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.mobilenet_v3_small.html#torchvision.models.MobileNet_V3_Small_Weights" title="torchvision.models.MobileNet_V3_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>67.668</p></td>
+<td><p>87.402</p></td>
+<td><p>2.5M</p></td>
+<td><p>0.06</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_16gf.html#torchvision.models.RegNet_X_16GF_Weights" title="torchvision.models.RegNet_X_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>80.058</p></td>
+<td><p>94.944</p></td>
+<td><p>54.3M</p></td>
+<td><p>15.94</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_16gf.html#torchvision.models.RegNet_X_16GF_Weights" title="torchvision.models.RegNet_X_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>82.716</p></td>
+<td><p>96.196</p></td>
+<td><p>54.3M</p></td>
+<td><p>15.94</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_1_6gf.html#torchvision.models.RegNet_X_1_6GF_Weights" title="torchvision.models.RegNet_X_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>77.04</p></td>
+<td><p>93.44</p></td>
+<td><p>9.2M</p></td>
+<td><p>1.6</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_1_6gf.html#torchvision.models.RegNet_X_1_6GF_Weights" title="torchvision.models.RegNet_X_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>79.668</p></td>
+<td><p>94.922</p></td>
+<td><p>9.2M</p></td>
+<td><p>1.6</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_32gf.html#torchvision.models.RegNet_X_32GF_Weights" title="torchvision.models.RegNet_X_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>80.622</p></td>
+<td><p>95.248</p></td>
+<td><p>107.8M</p></td>
+<td><p>31.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_32gf.html#torchvision.models.RegNet_X_32GF_Weights" title="torchvision.models.RegNet_X_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>83.014</p></td>
+<td><p>96.288</p></td>
+<td><p>107.8M</p></td>
+<td><p>31.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_3_2gf.html#torchvision.models.RegNet_X_3_2GF_Weights" title="torchvision.models.RegNet_X_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>78.364</p></td>
+<td><p>93.992</p></td>
+<td><p>15.3M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_3_2gf.html#torchvision.models.RegNet_X_3_2GF_Weights" title="torchvision.models.RegNet_X_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>81.196</p></td>
+<td><p>95.43</p></td>
+<td><p>15.3M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_400mf.html#torchvision.models.RegNet_X_400MF_Weights" title="torchvision.models.RegNet_X_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>72.834</p></td>
+<td><p>90.95</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_400mf.html#torchvision.models.RegNet_X_400MF_Weights" title="torchvision.models.RegNet_X_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>74.864</p></td>
+<td><p>92.322</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_800mf.html#torchvision.models.RegNet_X_800MF_Weights" title="torchvision.models.RegNet_X_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>75.212</p></td>
+<td><p>92.348</p></td>
+<td><p>7.3M</p></td>
+<td><p>0.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_800mf.html#torchvision.models.RegNet_X_800MF_Weights" title="torchvision.models.RegNet_X_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>77.522</p></td>
+<td><p>93.826</p></td>
+<td><p>7.3M</p></td>
+<td><p>0.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_8gf.html#torchvision.models.RegNet_X_8GF_Weights" title="torchvision.models.RegNet_X_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>79.344</p></td>
+<td><p>94.686</p></td>
+<td><p>39.6M</p></td>
+<td><p>8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_x_8gf.html#torchvision.models.RegNet_X_8GF_Weights" title="torchvision.models.RegNet_X_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>81.682</p></td>
+<td><p>95.678</p></td>
+<td><p>39.6M</p></td>
+<td><p>8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_128gf.html#torchvision.models.RegNet_Y_128GF_Weights" title="torchvision.models.RegNet_Y_128GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></a></p></td>
+<td><p>88.228</p></td>
+<td><p>98.682</p></td>
+<td><p>644.8M</p></td>
+<td><p>374.57</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_128gf.html#torchvision.models.RegNet_Y_128GF_Weights" title="torchvision.models.RegNet_Y_128GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></a></p></td>
+<td><p>86.068</p></td>
+<td><p>97.844</p></td>
+<td><p>644.8M</p></td>
+<td><p>127.52</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.RegNet_Y_16GF_Weights" title="torchvision.models.RegNet_Y_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>80.424</p></td>
+<td><p>95.24</p></td>
+<td><p>83.6M</p></td>
+<td><p>15.91</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.RegNet_Y_16GF_Weights" title="torchvision.models.RegNet_Y_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>82.886</p></td>
+<td><p>96.328</p></td>
+<td><p>83.6M</p></td>
+<td><p>15.91</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.RegNet_Y_16GF_Weights" title="torchvision.models.RegNet_Y_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></a></p></td>
+<td><p>86.012</p></td>
+<td><p>98.054</p></td>
+<td><p>83.6M</p></td>
+<td><p>46.73</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_16gf.html#torchvision.models.RegNet_Y_16GF_Weights" title="torchvision.models.RegNet_Y_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></a></p></td>
+<td><p>83.976</p></td>
+<td><p>97.244</p></td>
+<td><p>83.6M</p></td>
+<td><p>15.91</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_1_6gf.html#torchvision.models.RegNet_Y_1_6GF_Weights" title="torchvision.models.RegNet_Y_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>77.95</p></td>
+<td><p>93.966</p></td>
+<td><p>11.2M</p></td>
+<td><p>1.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_1_6gf.html#torchvision.models.RegNet_Y_1_6GF_Weights" title="torchvision.models.RegNet_Y_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>80.876</p></td>
+<td><p>95.444</p></td>
+<td><p>11.2M</p></td>
+<td><p>1.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.RegNet_Y_32GF_Weights" title="torchvision.models.RegNet_Y_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>80.878</p></td>
+<td><p>95.34</p></td>
+<td><p>145.0M</p></td>
+<td><p>32.28</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.RegNet_Y_32GF_Weights" title="torchvision.models.RegNet_Y_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>83.368</p></td>
+<td><p>96.498</p></td>
+<td><p>145.0M</p></td>
+<td><p>32.28</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.RegNet_Y_32GF_Weights" title="torchvision.models.RegNet_Y_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></a></p></td>
+<td><p>86.838</p></td>
+<td><p>98.362</p></td>
+<td><p>145.0M</p></td>
+<td><p>94.83</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_32gf.html#torchvision.models.RegNet_Y_32GF_Weights" title="torchvision.models.RegNet_Y_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></a></p></td>
+<td><p>84.622</p></td>
+<td><p>97.48</p></td>
+<td><p>145.0M</p></td>
+<td><p>32.28</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_3_2gf.html#torchvision.models.RegNet_Y_3_2GF_Weights" title="torchvision.models.RegNet_Y_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>78.948</p></td>
+<td><p>94.576</p></td>
+<td><p>19.4M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_3_2gf.html#torchvision.models.RegNet_Y_3_2GF_Weights" title="torchvision.models.RegNet_Y_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>81.982</p></td>
+<td><p>95.972</p></td>
+<td><p>19.4M</p></td>
+<td><p>3.18</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_400mf.html#torchvision.models.RegNet_Y_400MF_Weights" title="torchvision.models.RegNet_Y_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>74.046</p></td>
+<td><p>91.716</p></td>
+<td><p>4.3M</p></td>
+<td><p>0.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_400mf.html#torchvision.models.RegNet_Y_400MF_Weights" title="torchvision.models.RegNet_Y_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>75.804</p></td>
+<td><p>92.742</p></td>
+<td><p>4.3M</p></td>
+<td><p>0.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_800mf.html#torchvision.models.RegNet_Y_800MF_Weights" title="torchvision.models.RegNet_Y_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>76.42</p></td>
+<td><p>93.136</p></td>
+<td><p>6.4M</p></td>
+<td><p>0.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_800mf.html#torchvision.models.RegNet_Y_800MF_Weights" title="torchvision.models.RegNet_Y_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>78.828</p></td>
+<td><p>94.502</p></td>
+<td><p>6.4M</p></td>
+<td><p>0.83</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_8gf.html#torchvision.models.RegNet_Y_8GF_Weights" title="torchvision.models.RegNet_Y_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>80.032</p></td>
+<td><p>95.048</p></td>
+<td><p>39.4M</p></td>
+<td><p>8.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.regnet_y_8gf.html#torchvision.models.RegNet_Y_8GF_Weights" title="torchvision.models.RegNet_Y_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>82.828</p></td>
+<td><p>96.33</p></td>
+<td><p>39.4M</p></td>
+<td><p>8.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnext101_32x8d.html#torchvision.models.ResNeXt101_32X8D_Weights" title="torchvision.models.ResNeXt101_32X8D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>79.312</p></td>
+<td><p>94.526</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnext101_32x8d.html#torchvision.models.ResNeXt101_32X8D_Weights" title="torchvision.models.ResNeXt101_32X8D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>82.834</p></td>
+<td><p>96.228</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnext101_64x4d.html#torchvision.models.ResNeXt101_64X4D_Weights" title="torchvision.models.ResNeXt101_64X4D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.246</p></td>
+<td><p>96.454</p></td>
+<td><p>83.5M</p></td>
+<td><p>15.46</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5935">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnext50_32x4d.html#torchvision.models.ResNeXt50_32X4D_Weights" title="torchvision.models.ResNeXt50_32X4D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>77.618</p></td>
+<td><p>93.698</p></td>
+<td><p>25.0M</p></td>
+<td><p>4.23</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnext50_32x4d.html#torchvision.models.ResNeXt50_32X4D_Weights" title="torchvision.models.ResNeXt50_32X4D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>81.198</p></td>
+<td><p>95.34</p></td>
+<td><p>25.0M</p></td>
+<td><p>4.23</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights" title="torchvision.models.ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>77.374</p></td>
+<td><p>93.546</p></td>
+<td><p>44.5M</p></td>
+<td><p>7.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights" title="torchvision.models.ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>81.886</p></td>
+<td><p>95.78</p></td>
+<td><p>44.5M</p></td>
+<td><p>7.8</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet152.html#torchvision.models.ResNet152_Weights" title="torchvision.models.ResNet152_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet152_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>78.312</p></td>
+<td><p>94.046</p></td>
+<td><p>60.2M</p></td>
+<td><p>11.51</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet152.html#torchvision.models.ResNet152_Weights" title="torchvision.models.ResNet152_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet152_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>82.284</p></td>
+<td><p>96.002</p></td>
+<td><p>60.2M</p></td>
+<td><p>11.51</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet18.html#torchvision.models.ResNet18_Weights" title="torchvision.models.ResNet18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>69.758</p></td>
+<td><p>89.078</p></td>
+<td><p>11.7M</p></td>
+<td><p>1.81</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet34.html#torchvision.models.ResNet34_Weights" title="torchvision.models.ResNet34_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet34_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>73.314</p></td>
+<td><p>91.42</p></td>
+<td><p>21.8M</p></td>
+<td><p>3.66</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>76.13</p></td>
+<td><p>92.862</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>80.858</p></td>
+<td><p>95.434</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#issuecomment-1013906621">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.ShuffleNet_V2_X0_5_Weights" title="torchvision.models.ShuffleNet_V2_X0_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>60.552</p></td>
+<td><p>81.746</p></td>
+<td><p>1.4M</p></td>
+<td><p>0.04</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.ShuffleNet_V2_X1_0_Weights" title="torchvision.models.ShuffleNet_V2_X1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>69.362</p></td>
+<td><p>88.316</p></td>
+<td><p>2.3M</p></td>
+<td><p>0.14</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.ShuffleNet_V2_X1_5_Weights" title="torchvision.models.ShuffleNet_V2_X1_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>72.996</p></td>
+<td><p>91.086</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.ShuffleNet_V2_X2_0_Weights" title="torchvision.models.ShuffleNet_V2_X2_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>76.23</p></td>
+<td><p>93.006</p></td>
+<td><p>7.4M</p></td>
+<td><p>0.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.squeezenet1_0.html#torchvision.models.SqueezeNet1_0_Weights" title="torchvision.models.SqueezeNet1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>58.092</p></td>
+<td><p>80.42</p></td>
+<td><p>1.2M</p></td>
+<td><p>0.82</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/49#issuecomment-277560717">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.squeezenet1_1.html#torchvision.models.SqueezeNet1_1_Weights" title="torchvision.models.SqueezeNet1_1_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>58.178</p></td>
+<td><p>80.624</p></td>
+<td><p>1.2M</p></td>
+<td><p>0.35</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/49#issuecomment-277560717">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.swin_b.html#torchvision.models.Swin_B_Weights" title="torchvision.models.Swin_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_B_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.582</p></td>
+<td><p>96.64</p></td>
+<td><p>87.8M</p></td>
+<td><p>15.43</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.swin_s.html#torchvision.models.Swin_S_Weights" title="torchvision.models.Swin_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_S_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.196</p></td>
+<td><p>96.36</p></td>
+<td><p>49.6M</p></td>
+<td><p>8.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.swin_t.html#torchvision.models.Swin_T_Weights" title="torchvision.models.Swin_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_T_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>81.474</p></td>
+<td><p>95.776</p></td>
+<td><p>28.3M</p></td>
+<td><p>4.49</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.swin_v2_b.html#torchvision.models.Swin_V2_B_Weights" title="torchvision.models.Swin_V2_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_B_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>84.112</p></td>
+<td><p>96.864</p></td>
+<td><p>87.9M</p></td>
+<td><p>20.32</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.swin_v2_s.html#torchvision.models.Swin_V2_S_Weights" title="torchvision.models.Swin_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_S_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>83.712</p></td>
+<td><p>96.816</p></td>
+<td><p>49.7M</p></td>
+<td><p>11.55</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.swin_v2_t.html#torchvision.models.Swin_V2_T_Weights" title="torchvision.models.Swin_V2_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_T_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>82.072</p></td>
+<td><p>96.132</p></td>
+<td><p>28.4M</p></td>
+<td><p>5.94</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg11_bn.html#torchvision.models.VGG11_BN_Weights" title="torchvision.models.VGG11_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_BN_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>70.37</p></td>
+<td><p>89.81</p></td>
+<td><p>132.9M</p></td>
+<td><p>7.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg11.html#torchvision.models.VGG11_Weights" title="torchvision.models.VGG11_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>69.02</p></td>
+<td><p>88.628</p></td>
+<td><p>132.9M</p></td>
+<td><p>7.61</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg13_bn.html#torchvision.models.VGG13_BN_Weights" title="torchvision.models.VGG13_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_BN_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>71.586</p></td>
+<td><p>90.374</p></td>
+<td><p>133.1M</p></td>
+<td><p>11.31</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg13.html#torchvision.models.VGG13_Weights" title="torchvision.models.VGG13_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>69.928</p></td>
+<td><p>89.246</p></td>
+<td><p>133.0M</p></td>
+<td><p>11.31</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg16_bn.html#torchvision.models.VGG16_BN_Weights" title="torchvision.models.VGG16_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_BN_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>73.36</p></td>
+<td><p>91.516</p></td>
+<td><p>138.4M</p></td>
+<td><p>15.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg16.html#torchvision.models.VGG16_Weights" title="torchvision.models.VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>71.592</p></td>
+<td><p>90.382</p></td>
+<td><p>138.4M</p></td>
+<td><p>15.47</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg16.html#torchvision.models.VGG16_Weights" title="torchvision.models.VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_Weights.IMAGENET1K_FEATURES</span></code></a></p></td>
+<td><p>nan</p></td>
+<td><p>nan</p></td>
+<td><p>138.4M</p></td>
+<td><p>15.47</p></td>
+<td><p><a class="reference external" href="https://github.com/amdegroot/ssd.pytorch#training-ssd">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg19_bn.html#torchvision.models.VGG19_BN_Weights" title="torchvision.models.VGG19_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_BN_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>74.218</p></td>
+<td><p>91.842</p></td>
+<td><p>143.7M</p></td>
+<td><p>19.63</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vgg19.html#torchvision.models.VGG19_Weights" title="torchvision.models.VGG19_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>72.376</p></td>
+<td><p>90.876</p></td>
+<td><p>143.7M</p></td>
+<td><p>19.63</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_b_16.html#torchvision.models.ViT_B_16_Weights" title="torchvision.models.ViT_B_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>81.072</p></td>
+<td><p>95.318</p></td>
+<td><p>86.6M</p></td>
+<td><p>17.56</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_b_16">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_b_16.html#torchvision.models.ViT_B_16_Weights" title="torchvision.models.ViT_B_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></a></p></td>
+<td><p>85.304</p></td>
+<td><p>97.65</p></td>
+<td><p>86.9M</p></td>
+<td><p>55.48</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_b_16.html#torchvision.models.ViT_B_16_Weights" title="torchvision.models.ViT_B_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></a></p></td>
+<td><p>81.886</p></td>
+<td><p>96.18</p></td>
+<td><p>86.6M</p></td>
+<td><p>17.56</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_b_32.html#torchvision.models.ViT_B_32_Weights" title="torchvision.models.ViT_B_32_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_32_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>75.912</p></td>
+<td><p>92.466</p></td>
+<td><p>88.2M</p></td>
+<td><p>4.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_b_32">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_h_14.html#torchvision.models.ViT_H_14_Weights" title="torchvision.models.ViT_H_14_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_H_14_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></a></p></td>
+<td><p>88.552</p></td>
+<td><p>98.694</p></td>
+<td><p>633.5M</p></td>
+<td><p>1016.72</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_h_14.html#torchvision.models.ViT_H_14_Weights" title="torchvision.models.ViT_H_14_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_H_14_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></a></p></td>
+<td><p>85.708</p></td>
+<td><p>97.73</p></td>
+<td><p>632.0M</p></td>
+<td><p>167.29</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_l_16.html#torchvision.models.ViT_L_16_Weights" title="torchvision.models.ViT_L_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>79.662</p></td>
+<td><p>94.638</p></td>
+<td><p>304.3M</p></td>
+<td><p>61.55</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_l_16">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_l_16.html#torchvision.models.ViT_L_16_Weights" title="torchvision.models.ViT_L_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_SWAG_E2E_V1</span></code></a></p></td>
+<td><p>88.064</p></td>
+<td><p>98.512</p></td>
+<td><p>305.2M</p></td>
+<td><p>361.99</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_l_16.html#torchvision.models.ViT_L_16_Weights" title="torchvision.models.ViT_L_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_SWAG_LINEAR_V1</span></code></a></p></td>
+<td><p>85.146</p></td>
+<td><p>97.422</p></td>
+<td><p>304.3M</p></td>
+<td><p>61.55</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.vit_l_32.html#torchvision.models.ViT_L_32_Weights" title="torchvision.models.ViT_L_32_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_32_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>76.972</p></td>
+<td><p>93.07</p></td>
+<td><p>306.5M</p></td>
+<td><p>15.38</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_l_32">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.wide_resnet101_2.html#torchvision.models.Wide_ResNet101_2_Weights" title="torchvision.models.Wide_ResNet101_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>78.848</p></td>
+<td><p>94.284</p></td>
+<td><p>126.9M</p></td>
+<td><p>22.75</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/912#issue-445437439">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.wide_resnet101_2.html#torchvision.models.Wide_ResNet101_2_Weights" title="torchvision.models.Wide_ResNet101_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>82.51</p></td>
+<td><p>96.02</p></td>
+<td><p>126.9M</p></td>
+<td><p>22.75</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.wide_resnet50_2.html#torchvision.models.Wide_ResNet50_2_Weights" title="torchvision.models.Wide_ResNet50_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.IMAGENET1K_V1</span></code></a></p></td>
+<td><p>78.468</p></td>
+<td><p>94.086</p></td>
+<td><p>68.9M</p></td>
+<td><p>11.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/912#issue-445437439">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.wide_resnet50_2.html#torchvision.models.Wide_ResNet50_2_Weights" title="torchvision.models.Wide_ResNet50_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.IMAGENET1K_V2</span></code></a></p></td>
+<td><p>81.602</p></td>
+<td><p>95.758</p></td>
+<td><p>68.9M</p></td>
+<td><p>11.4</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="quantized-models">
+<h3>Quantized models<a class="headerlink" href="#quantized-models" title="Permalink to this heading">¶</a></h3>
+<p>The following architectures provide support for INT8 quantized models, with or without
+pre-trained weights:</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/googlenet_quant.html">Quantized GoogLeNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/inception_quant.html">Quantized InceptionV3</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/mobilenetv2_quant.html">Quantized MobileNet V2</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/mobilenetv3_quant.html">Quantized MobileNet V3</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/resnet_quant.html">Quantized ResNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/resnext_quant.html">Quantized ResNeXt</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/shufflenetv2_quant.html">Quantized ShuffleNet V2</a></li>
+</ul>
+</div>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>Here is an example of how to use the pre-trained quantized image classification models:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.io</span> <span class="kn">import</span> <span class="n">read_image</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.quantization</span> <span class="kn">import</span> <span class="n">resnet50</span><span class="p">,</span> <span class="n">ResNet50_QuantizedWeights</span>
+
+<span class="n">img</span> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="s2">&quot;test/assets/encode_jpeg/grace_hopper_517x606.jpg&quot;</span><span class="p">)</span>
+
+<span class="c1"># Step 1: Initialize model with the best available weights</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">ResNet50_QuantizedWeights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">quantize</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+<span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+<span class="c1"># Step 2: Initialize the inference transforms</span>
+<span class="n">preprocess</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<span class="c1"># Step 3: Apply inference preprocessing transforms</span>
+<span class="n">batch</span> <span class="o">=</span> <span class="n">preprocess</span><span class="p">(</span><span class="n">img</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+<span class="c1"># Step 4: Use the model and print the predicted category</span>
+<span class="n">prediction</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">batch</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+<span class="n">class_id</span> <span class="o">=</span> <span class="n">prediction</span><span class="o">.</span><span class="n">argmax</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+<span class="n">score</span> <span class="o">=</span> <span class="n">prediction</span><span class="p">[</span><span class="n">class_id</span><span class="p">]</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+<span class="n">category_name</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">][</span><span class="n">class_id</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">category_name</span><span class="si">}</span><span class="s2">: </span><span class="si">{</span><span class="mi">100</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="n">score</span><span class="si">}</span><span class="s2">%&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>The classes of the pre-trained model outputs can be found at <code class="docutils literal notranslate"><span class="pre">weights.meta[&quot;categories&quot;]</span></code>.</p>
+<section id="table-of-all-available-quantized-classification-weights">
+<h4>Table of all available quantized classification weights<a class="headerlink" href="#table-of-all-available-quantized-classification-weights" title="Permalink to this heading">¶</a></h4>
+<p>Accuracies are reported on ImageNet-1K using single crops:</p>
+<table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Acc&#64;1</strong></p></th>
+<th class="head"><p><strong>Acc&#64;5</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GIPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.googlenet.html#torchvision.models.quantization.GoogLeNet_QuantizedWeights" title="torchvision.models.quantization.GoogLeNet_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>69.826</p></td>
+<td><p>89.404</p></td>
+<td><p>6.6M</p></td>
+<td><p>1.5</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.Inception_V3_QuantizedWeights" title="torchvision.models.quantization.Inception_V3_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>77.176</p></td>
+<td><p>93.354</p></td>
+<td><p>27.2M</p></td>
+<td><p>5.71</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.mobilenet_v2.html#torchvision.models.quantization.MobileNet_V2_QuantizedWeights" title="torchvision.models.quantization.MobileNet_V2_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights.IMAGENET1K_QNNPACK_V1</span></code></a></p></td>
+<td><p>71.658</p></td>
+<td><p>90.15</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.mobilenet_v3_large.html#torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights" title="torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights.IMAGENET1K_QNNPACK_V1</span></code></a></p></td>
+<td><p>73.004</p></td>
+<td><p>90.858</p></td>
+<td><p>5.5M</p></td>
+<td><p>0.22</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv3">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.resnext101_32x8d.html#torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights" title="torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>78.986</p></td>
+<td><p>94.48</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.resnext101_32x8d.html#torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights" title="torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V2</span></code></a></p></td>
+<td><p>82.574</p></td>
+<td><p>96.132</p></td>
+<td><p>88.8M</p></td>
+<td><p>16.41</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.resnext101_64x4d.html#torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights" title="torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>82.898</p></td>
+<td><p>96.326</p></td>
+<td><p>83.5M</p></td>
+<td><p>15.46</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5935">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.resnet18.html#torchvision.models.quantization.ResNet18_QuantizedWeights" title="torchvision.models.quantization.ResNet18_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>69.494</p></td>
+<td><p>88.882</p></td>
+<td><p>11.7M</p></td>
+<td><p>1.81</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.resnet50.html#torchvision.models.quantization.ResNet50_QuantizedWeights" title="torchvision.models.quantization.ResNet50_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>75.92</p></td>
+<td><p>92.814</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.resnet50.html#torchvision.models.quantization.ResNet50_QuantizedWeights" title="torchvision.models.quantization.ResNet50_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V2</span></code></a></p></td>
+<td><p>80.282</p></td>
+<td><p>94.976</p></td>
+<td><p>25.6M</p></td>
+<td><p>4.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html#torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>57.972</p></td>
+<td><p>79.78</p></td>
+<td><p>1.4M</p></td>
+<td><p>0.04</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html#torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>68.36</p></td>
+<td><p>87.582</p></td>
+<td><p>2.3M</p></td>
+<td><p>0.14</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html#torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>72.052</p></td>
+<td><p>90.7</p></td>
+<td><p>3.5M</p></td>
+<td><p>0.3</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html#torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code></a></p></td>
+<td><p>75.354</p></td>
+<td><p>92.488</p></td>
+<td><p>7.4M</p></td>
+<td><p>0.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+</section>
+<section id="semantic-segmentation">
+<h2>Semantic Segmentation<a class="headerlink" href="#semantic-segmentation" title="Permalink to this heading">¶</a></h2>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>The following semantic segmentation models are available, with or without
+pre-trained weights:</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/deeplabv3.html">DeepLabV3</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/fcn.html">FCN</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/lraspp.html">LRASPP</a></li>
+</ul>
+</div>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>Here is an example of how to use the pre-trained semantic segmentation models:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.io.image</span> <span class="kn">import</span> <span class="n">read_image</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.segmentation</span> <span class="kn">import</span> <span class="n">fcn_resnet50</span><span class="p">,</span> <span class="n">FCN_ResNet50_Weights</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.functional</span> <span class="kn">import</span> <span class="n">to_pil_image</span>
+
+<span class="n">img</span> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="s2">&quot;gallery/assets/dog1.jpg&quot;</span><span class="p">)</span>
+
+<span class="c1"># Step 1: Initialize model with the best available weights</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">FCN_ResNet50_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">fcn_resnet50</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">)</span>
+<span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+<span class="c1"># Step 2: Initialize the inference transforms</span>
+<span class="n">preprocess</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<span class="c1"># Step 3: Apply inference preprocessing transforms</span>
+<span class="n">batch</span> <span class="o">=</span> <span class="n">preprocess</span><span class="p">(</span><span class="n">img</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+<span class="c1"># Step 4: Use the model and visualize the prediction</span>
+<span class="n">prediction</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">batch</span><span class="p">)[</span><span class="s2">&quot;out&quot;</span><span class="p">]</span>
+<span class="n">normalized_masks</span> <span class="o">=</span> <span class="n">prediction</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+<span class="n">class_to_idx</span> <span class="o">=</span> <span class="p">{</span><span class="bp">cls</span><span class="p">:</span> <span class="n">idx</span> <span class="k">for</span> <span class="p">(</span><span class="n">idx</span><span class="p">,</span> <span class="bp">cls</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">])}</span>
+<span class="n">mask</span> <span class="o">=</span> <span class="n">normalized_masks</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="n">class_to_idx</span><span class="p">[</span><span class="s2">&quot;dog&quot;</span><span class="p">]]</span>
+<span class="n">to_pil_image</span><span class="p">(</span><span class="n">mask</span><span class="p">)</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
+</pre></div>
+</div>
+<p>The classes of the pre-trained model outputs can be found at <code class="docutils literal notranslate"><span class="pre">weights.meta[&quot;categories&quot;]</span></code>.
+The output format of the models is illustrated in <a class="reference internal" href="auto_examples/others/plot_visualization_utils.html#semantic-seg-output"><span class="std std-ref">Semantic segmentation models</span></a>.</p>
+<section id="table-of-all-available-semantic-segmentation-weights">
+<h3>Table of all available semantic segmentation weights<a class="headerlink" href="#table-of-all-available-semantic-segmentation-weights" title="Permalink to this heading">¶</a></h3>
+<p>All models are evaluated a subset of COCO val2017, on the 20 categories that are present in the Pascal VOC dataset:</p>
+<table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Mean IoU</strong></p></th>
+<th class="head"><p><strong>pixelwise Acc</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html#torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</span></code></a></p></td>
+<td><p>60.3</p></td>
+<td><p>91.2</p></td>
+<td><p>11.0M</p></td>
+<td><p>10.45</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_mobilenet_v3_large">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html#torchvision.models.segmentation.DeepLabV3_ResNet101_Weights" title="torchvision.models.segmentation.DeepLabV3_ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</span></code></a></p></td>
+<td><p>67.4</p></td>
+<td><p>92.4</p></td>
+<td><p>61.0M</p></td>
+<td><p>258.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet101">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html#torchvision.models.segmentation.DeepLabV3_ResNet50_Weights" title="torchvision.models.segmentation.DeepLabV3_ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</span></code></a></p></td>
+<td><p>66.4</p></td>
+<td><p>92.4</p></td>
+<td><p>42.0M</p></td>
+<td><p>178.72</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet50">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.segmentation.fcn_resnet101.html#torchvision.models.segmentation.FCN_ResNet101_Weights" title="torchvision.models.segmentation.FCN_ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</span></code></a></p></td>
+<td><p>63.7</p></td>
+<td><p>91.9</p></td>
+<td><p>54.3M</p></td>
+<td><p>232.74</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet101">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.segmentation.fcn_resnet50.html#torchvision.models.segmentation.FCN_ResNet50_Weights" title="torchvision.models.segmentation.FCN_ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</span></code></a></p></td>
+<td><p>60.5</p></td>
+<td><p>91.4</p></td>
+<td><p>35.3M</p></td>
+<td><p>152.72</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet50">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html#torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</span></code></a></p></td>
+<td><p>57.9</p></td>
+<td><p>91.2</p></td>
+<td><p>3.2M</p></td>
+<td><p>2.09</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#lraspp_mobilenet_v3_large">link</a></p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+<section id="object-detection-instance-segmentation-and-person-keypoint-detection">
+<span id="object-det-inst-seg-pers-keypoint-det"></span><h2>Object Detection, Instance Segmentation and Person Keypoint Detection<a class="headerlink" href="#object-detection-instance-segmentation-and-person-keypoint-detection" title="Permalink to this heading">¶</a></h2>
+<p>The pre-trained models for detection, instance segmentation and
+keypoint detection are initialized with the classification models
+in torchvision. The models expect a list of <code class="docutils literal notranslate"><span class="pre">Tensor[C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>.
+Check the constructor of the models for more information.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="object-detection">
+<h3>Object Detection<a class="headerlink" href="#object-detection" title="Permalink to this heading">¶</a></h3>
+<p>The following object detection models are available, with or without pre-trained
+weights:</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/faster_rcnn.html">Faster R-CNN</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/fcos.html">FCOS</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/retinanet.html">RetinaNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/ssd.html">SSD</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/ssdlite.html">SSDlite</a></li>
+</ul>
+</div>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>Here is an example of how to use the pre-trained object detection models:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.io.image</span> <span class="kn">import</span> <span class="n">read_image</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.detection</span> <span class="kn">import</span> <span class="n">fasterrcnn_resnet50_fpn_v2</span><span class="p">,</span> <span class="n">FasterRCNN_ResNet50_FPN_V2_Weights</span>
+<span class="kn">from</span> <span class="nn">torchvision.utils</span> <span class="kn">import</span> <span class="n">draw_bounding_boxes</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms.functional</span> <span class="kn">import</span> <span class="n">to_pil_image</span>
+
+<span class="n">img</span> <span class="o">=</span> <span class="n">read_image</span><span class="p">(</span><span class="s2">&quot;test/assets/encode_jpeg/grace_hopper_517x606.jpg&quot;</span><span class="p">)</span>
+
+<span class="c1"># Step 1: Initialize model with the best available weights</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">FasterRCNN_ResNet50_FPN_V2_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">fasterrcnn_resnet50_fpn_v2</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">,</span> <span class="n">box_score_thresh</span><span class="o">=</span><span class="mf">0.9</span><span class="p">)</span>
+<span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+<span class="c1"># Step 2: Initialize the inference transforms</span>
+<span class="n">preprocess</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<span class="c1"># Step 3: Apply inference preprocessing transforms</span>
+<span class="n">batch</span> <span class="o">=</span> <span class="p">[</span><span class="n">preprocess</span><span class="p">(</span><span class="n">img</span><span class="p">)]</span>
+
+<span class="c1"># Step 4: Use the model and visualize the prediction</span>
+<span class="n">prediction</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">batch</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+<span class="n">labels</span> <span class="o">=</span> <span class="p">[</span><span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">prediction</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]]</span>
+<span class="n">box</span> <span class="o">=</span> <span class="n">draw_bounding_boxes</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">boxes</span><span class="o">=</span><span class="n">prediction</span><span class="p">[</span><span class="s2">&quot;boxes&quot;</span><span class="p">],</span>
+                          <span class="n">labels</span><span class="o">=</span><span class="n">labels</span><span class="p">,</span>
+                          <span class="n">colors</span><span class="o">=</span><span class="s2">&quot;red&quot;</span><span class="p">,</span>
+                          <span class="n">width</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">font_size</span><span class="o">=</span><span class="mi">30</span><span class="p">)</span>
+<span class="n">im</span> <span class="o">=</span> <span class="n">to_pil_image</span><span class="p">(</span><span class="n">box</span><span class="o">.</span><span class="n">detach</span><span class="p">())</span>
+<span class="n">im</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
+</pre></div>
+</div>
+<p>The classes of the pre-trained model outputs can be found at <code class="docutils literal notranslate"><span class="pre">weights.meta[&quot;categories&quot;]</span></code>.
+For details on how to plot the bounding boxes of the models, you may refer to <a class="reference internal" href="auto_examples/others/plot_visualization_utils.html#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<section id="table-of-all-available-object-detection-weights">
+<h4>Table of all available Object detection weights<a class="headerlink" href="#table-of-all-available-object-detection-weights" title="Permalink to this heading">¶</a></h4>
+<p>Box MAPs are reported on COCO val2017:</p>
+<table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 63%" />
+<col style="width: 10%" />
+<col style="width: 10%" />
+<col style="width: 10%" />
+<col style="width: 6%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Box MAP</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.fcos_resnet50_fpn.html#torchvision.models.detection.FCOS_ResNet50_FPN_Weights" title="torchvision.models.detection.FCOS_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights.COCO_V1</span></code></a></p></td>
+<td><p>39.2</p></td>
+<td><p>32.3M</p></td>
+<td><p>128.21</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#fcos-resnet-50-fpn">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights" title="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.COCO_V1</span></code></a></p></td>
+<td><p>22.8</p></td>
+<td><p>19.4M</p></td>
+<td><p>0.72</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-320-fpn">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights" title="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights.COCO_V1</span></code></a></p></td>
+<td><p>32.8</p></td>
+<td><p>19.4M</p></td>
+<td><p>4.49</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-fpn">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html#torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights.COCO_V1</span></code></a></p></td>
+<td><p>46.7</p></td>
+<td><p>43.7M</p></td>
+<td><p>280.37</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5763">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights.COCO_V1</span></code></a></p></td>
+<td><p>37</p></td>
+<td><p>41.8M</p></td>
+<td><p>134.38</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-resnet-50-fpn">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html#torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights.COCO_V1</span></code></a></p></td>
+<td><p>41.5</p></td>
+<td><p>38.2M</p></td>
+<td><p>152.24</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5756">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html#torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights" title="torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights.COCO_V1</span></code></a></p></td>
+<td><p>36.4</p></td>
+<td><p>34.0M</p></td>
+<td><p>151.54</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#retinanet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.ssd300_vgg16.html#torchvision.models.detection.SSD300_VGG16_Weights" title="torchvision.models.detection.SSD300_VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights.COCO_V1</span></code></a></p></td>
+<td><p>25.1</p></td>
+<td><p>35.6M</p></td>
+<td><p>34.86</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#ssd300-vgg16">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html#torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights" title="torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights.COCO_V1</span></code></a></p></td>
+<td><p>21.3</p></td>
+<td><p>3.4M</p></td>
+<td><p>0.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#ssdlite320-mobilenetv3-large">link</a></p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+<section id="instance-segmentation">
+<h3>Instance Segmentation<a class="headerlink" href="#instance-segmentation" title="Permalink to this heading">¶</a></h3>
+<p>The following instance segmentation models are available, with or without pre-trained
+weights:</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/mask_rcnn.html">Mask R-CNN</a></li>
+</ul>
+</div>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>For details on how to plot the masks of the models, you may refer to <a class="reference internal" href="auto_examples/others/plot_visualization_utils.html#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<section id="table-of-all-available-instance-segmentation-weights">
+<h4>Table of all available Instance segmentation weights<a class="headerlink" href="#table-of-all-available-instance-segmentation-weights" title="Permalink to this heading">¶</a></h4>
+<p>Box and Mask MAPs are reported on COCO val2017:</p>
+<table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Box MAP</strong></p></th>
+<th class="head"><p><strong>Mask MAP</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html#torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights.COCO_V1</span></code></a></p></td>
+<td><p>47.4</p></td>
+<td><p>41.8</p></td>
+<td><p>46.4M</p></td>
+<td><p>333.58</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5773">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html#torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights.COCO_V1</span></code></a></p></td>
+<td><p>37.9</p></td>
+<td><p>34.6</p></td>
+<td><p>44.4M</p></td>
+<td><p>134.38</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#mask-r-cnn">link</a></p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+<section id="keypoint-detection">
+<h3>Keypoint Detection<a class="headerlink" href="#keypoint-detection" title="Permalink to this heading">¶</a></h3>
+<p>The following person keypoint detection models are available, with or without
+pre-trained weights:</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/keypoint_rcnn.html">Keypoint R-CNN</a></li>
+</ul>
+</div>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>The classes of the pre-trained model outputs can be found at <code class="docutils literal notranslate"><span class="pre">weights.meta[&quot;keypoint_names&quot;]</span></code>.
+For details on how to plot the bounding boxes of the models, you may refer to <a class="reference internal" href="auto_examples/others/plot_visualization_utils.html#keypoint-output"><span class="std std-ref">Visualizing keypoints</span></a>.</p>
+<section id="table-of-all-available-keypoint-detection-weights">
+<h4>Table of all available Keypoint detection weights<a class="headerlink" href="#table-of-all-available-keypoint-detection-weights" title="Permalink to this heading">¶</a></h4>
+<p>Box and Keypoint MAPs are reported on COCO val2017:</p>
+<table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Box MAP</strong></p></th>
+<th class="head"><p><strong>Keypoint MAP</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.COCO_LEGACY</span></code></a></p></td>
+<td><p>50.6</p></td>
+<td><p>61.1</p></td>
+<td><p>59.1M</p></td>
+<td><p>133.92</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/1606">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.COCO_V1</span></code></a></p></td>
+<td><p>54.6</p></td>
+<td><p>65</p></td>
+<td><p>59.1M</p></td>
+<td><p>137.42</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#keypoint-r-cnn">link</a></p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+</section>
+<section id="video-classification">
+<h2>Video Classification<a class="headerlink" href="#video-classification" title="Permalink to this heading">¶</a></h2>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>The following video classification models are available, with or without
+pre-trained weights:</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/video_mvit.html">Video MViT</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/video_resnet.html">Video ResNet</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/video_s3d.html">Video S3D</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/video_swin_transformer.html">Video SwinTransformer</a></li>
+</ul>
+</div>
+<div class="line-block">
+<div class="line"><br /></div>
+</div>
+<p>Here is an example of how to use the pre-trained video classification models:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.io.video</span> <span class="kn">import</span> <span class="n">read_video</span>
+<span class="kn">from</span> <span class="nn">torchvision.models.video</span> <span class="kn">import</span> <span class="n">r3d_18</span><span class="p">,</span> <span class="n">R3D_18_Weights</span>
+
+<span class="n">vid</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">read_video</span><span class="p">(</span><span class="s2">&quot;test/assets/videos/v_SoccerJuggling_g23_c01.avi&quot;</span><span class="p">,</span> <span class="n">output_format</span><span class="o">=</span><span class="s2">&quot;TCHW&quot;</span><span class="p">)</span>
+<span class="n">vid</span> <span class="o">=</span> <span class="n">vid</span><span class="p">[:</span><span class="mi">32</span><span class="p">]</span>  <span class="c1"># optionally shorten duration</span>
+
+<span class="c1"># Step 1: Initialize model with the best available weights</span>
+<span class="n">weights</span> <span class="o">=</span> <span class="n">R3D_18_Weights</span><span class="o">.</span><span class="n">DEFAULT</span>
+<span class="n">model</span> <span class="o">=</span> <span class="n">r3d_18</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">weights</span><span class="p">)</span>
+<span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+<span class="c1"># Step 2: Initialize the inference transforms</span>
+<span class="n">preprocess</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">transforms</span><span class="p">()</span>
+
+<span class="c1"># Step 3: Apply inference preprocessing transforms</span>
+<span class="n">batch</span> <span class="o">=</span> <span class="n">preprocess</span><span class="p">(</span><span class="n">vid</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+
+<span class="c1"># Step 4: Use the model and print the predicted category</span>
+<span class="n">prediction</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">batch</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+<span class="n">label</span> <span class="o">=</span> <span class="n">prediction</span><span class="o">.</span><span class="n">argmax</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+<span class="n">score</span> <span class="o">=</span> <span class="n">prediction</span><span class="p">[</span><span class="n">label</span><span class="p">]</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+<span class="n">category_name</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">meta</span><span class="p">[</span><span class="s2">&quot;categories&quot;</span><span class="p">][</span><span class="n">label</span><span class="p">]</span>
+<span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">category_name</span><span class="si">}</span><span class="s2">: </span><span class="si">{</span><span class="mi">100</span><span class="w"> </span><span class="o">*</span><span class="w"> </span><span class="n">score</span><span class="si">}</span><span class="s2">%&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>The classes of the pre-trained model outputs can be found at <code class="docutils literal notranslate"><span class="pre">weights.meta[&quot;categories&quot;]</span></code>.</p>
+<section id="table-of-all-available-video-classification-weights">
+<h3>Table of all available video classification weights<a class="headerlink" href="#table-of-all-available-video-classification-weights" title="Permalink to this heading">¶</a></h3>
+<p>Accuracies are reported on Kinetics-400 using single crops for clip length 16:</p>
+<table class="table-weights docutils align-default">
+<colgroup>
+<col style="width: 57%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 9%" />
+<col style="width: 5%" />
+</colgroup>
+<thead>
+<tr class="row-odd"><th class="head"><p><strong>Weight</strong></p></th>
+<th class="head"><p><strong>Acc&#64;1</strong></p></th>
+<th class="head"><p><strong>Acc&#64;5</strong></p></th>
+<th class="head"><p><strong>Params</strong></p></th>
+<th class="head"><p><strong>GFLOPS</strong></p></th>
+<th class="head"><p><strong>Recipe</strong></p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.mc3_18.html#torchvision.models.video.MC3_18_Weights" title="torchvision.models.video.MC3_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MC3_18_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>63.96</p></td>
+<td><p>84.13</p></td>
+<td><p>11.7M</p></td>
+<td><p>43.34</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.mvit_v1_b.html#torchvision.models.video.MViT_V1_B_Weights" title="torchvision.models.video.MViT_V1_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V1_B_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>78.477</p></td>
+<td><p>93.582</p></td>
+<td><p>36.6M</p></td>
+<td><p>70.6</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/pytorchvideo/blob/main/docs/source/model_zoo.md">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.mvit_v2_s.html#torchvision.models.video.MViT_V2_S_Weights" title="torchvision.models.video.MViT_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V2_S_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>80.757</p></td>
+<td><p>94.665</p></td>
+<td><p>34.5M</p></td>
+<td><p>64.22</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SlowFast/blob/main/MODEL_ZOO.md">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.r2plus1d_18.html#torchvision.models.video.R2Plus1D_18_Weights" title="torchvision.models.video.R2Plus1D_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>67.463</p></td>
+<td><p>86.175</p></td>
+<td><p>31.5M</p></td>
+<td><p>40.52</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.r3d_18.html#torchvision.models.video.R3D_18_Weights" title="torchvision.models.video.R3D_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">R3D_18_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>63.2</p></td>
+<td><p>83.479</p></td>
+<td><p>33.4M</p></td>
+<td><p>40.7</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.s3d.html#torchvision.models.video.S3D_Weights" title="torchvision.models.video.S3D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">S3D_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>68.368</p></td>
+<td><p>88.05</p></td>
+<td><p>8.3M</p></td>
+<td><p>17.98</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification#s3d">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.swin3d_b.html#torchvision.models.video.Swin3D_B_Weights" title="torchvision.models.video.Swin3D_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_B_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>79.427</p></td>
+<td><p>94.386</p></td>
+<td><p>88.0M</p></td>
+<td><p>140.67</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.swin3d_b.html#torchvision.models.video.Swin3D_B_Weights" title="torchvision.models.video.Swin3D_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_B_Weights.KINETICS400_IMAGENET22K_V1</span></code></a></p></td>
+<td><p>81.643</p></td>
+<td><p>95.574</p></td>
+<td><p>88.0M</p></td>
+<td><p>140.67</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.swin3d_s.html#torchvision.models.video.Swin3D_S_Weights" title="torchvision.models.video.Swin3D_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_S_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>79.521</p></td>
+<td><p>94.158</p></td>
+<td><p>49.8M</p></td>
+<td><p>82.84</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="models/generated/torchvision.models.video.swin3d_t.html#torchvision.models.video.Swin3D_T_Weights" title="torchvision.models.video.Swin3D_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_T_Weights.KINETICS400_V1</span></code></a></p></td>
+<td><p>77.715</p></td>
+<td><p>93.519</p></td>
+<td><p>28.2M</p></td>
+<td><p>43.88</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+<section id="optical-flow">
+<h2>Optical Flow<a class="headerlink" href="#optical-flow" title="Permalink to this heading">¶</a></h2>
+<p>The following Optical Flow models are available, with or without pre-trained</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="models/raft.html">RAFT</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.get_model.html" class="btn btn-neutral float-right" title="get_model" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.tv_tensors.wrap.html" class="btn btn-neutral" title="wrap" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Models and pre-trained weights</a><ul>
+<li><a class="reference internal" href="#general-information-on-pre-trained-weights">General information on pre-trained weights</a><ul>
+<li><a class="reference internal" href="#initializing-pre-trained-models">Initializing pre-trained models</a></li>
+<li><a class="reference internal" href="#using-the-pre-trained-models">Using the pre-trained models</a></li>
+<li><a class="reference internal" href="#listing-and-retrieving-available-models">Listing and retrieving available models</a></li>
+<li><a class="reference internal" href="#using-models-from-hub">Using models from Hub</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#classification">Classification</a><ul>
+<li><a class="reference internal" href="#table-of-all-available-classification-weights">Table of all available classification weights</a></li>
+<li><a class="reference internal" href="#quantized-models">Quantized models</a><ul>
+<li><a class="reference internal" href="#table-of-all-available-quantized-classification-weights">Table of all available quantized classification weights</a></li>
+</ul>
+</li>
+</ul>
+</li>
+<li><a class="reference internal" href="#semantic-segmentation">Semantic Segmentation</a><ul>
+<li><a class="reference internal" href="#table-of-all-available-semantic-segmentation-weights">Table of all available semantic segmentation weights</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#object-detection-instance-segmentation-and-person-keypoint-detection">Object Detection, Instance Segmentation and Person Keypoint Detection</a><ul>
+<li><a class="reference internal" href="#object-detection">Object Detection</a><ul>
+<li><a class="reference internal" href="#table-of-all-available-object-detection-weights">Table of all available Object detection weights</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#instance-segmentation">Instance Segmentation</a><ul>
+<li><a class="reference internal" href="#table-of-all-available-instance-segmentation-weights">Table of all available Instance segmentation weights</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#keypoint-detection">Keypoint Detection</a><ul>
+<li><a class="reference internal" href="#table-of-all-available-keypoint-detection-weights">Table of all available Keypoint detection weights</a></li>
+</ul>
+</li>
+</ul>
+</li>
+<li><a class="reference internal" href="#video-classification">Video Classification</a><ul>
+<li><a class="reference internal" href="#table-of-all-available-video-classification-weights">Table of all available video classification weights</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#optical-flow">Optical Flow</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/alexnet.html b/0.16/models/alexnet.html
new file mode 100644
index 00000000000..5d602c7e172
--- /dev/null
+++ b/0.16/models/alexnet.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>AlexNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="alexnet" href="generated/torchvision.models.alexnet.html" />
+    <link rel="prev" title="list_models" href="../generated/torchvision.models.list_models.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>AlexNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/alexnet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="alexnet">
+<h1>AlexNet<a class="headerlink" href="#alexnet" title="Permalink to this heading">¶</a></h1>
+<p>The AlexNet model was originally introduced in the
+<a class="reference external" href="https://papers.nips.cc/paper/2012/hash/c399862d3b9d6b76c8436e924a68c45b-Abstract.html">ImageNet Classification with Deep Convolutional Neural Networks</a>
+paper. The implemented architecture is slightly different from the original one,
+and is based on <a class="reference external" href="https://arxiv.org/abs/1404.5997">One weird trick for parallelizing convolutional neural networks</a>.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an AlexNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.alexnet.AlexNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/alexnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.alexnet.html#torchvision.models.alexnet" title="torchvision.models.alexnet"><code class="xref py py-obj docutils literal notranslate"><span class="pre">alexnet</span></code></a>(*[, weights, progress])</p></td>
+<td><p>AlexNet model architecture from <a class="reference external" href="https://arxiv.org/abs/1404.5997">One weird trick for parallelizing convolutional neural networks</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.alexnet.html" class="btn btn-neutral float-right" title="alexnet" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../generated/torchvision.models.list_models.html" class="btn btn-neutral" title="list_models" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">AlexNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/convnext.html b/0.16/models/convnext.html
new file mode 100644
index 00000000000..d238ff0ed70
--- /dev/null
+++ b/0.16/models/convnext.html
@@ -0,0 +1,719 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ConvNeXt &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="convnext_tiny" href="generated/torchvision.models.convnext_tiny.html" />
+    <link rel="prev" title="alexnet" href="generated/torchvision.models.alexnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>ConvNeXt</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/convnext.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convnext">
+<h1>ConvNeXt<a class="headerlink" href="#convnext" title="Permalink to this heading">¶</a></h1>
+<p>The ConvNeXt model is based on the <a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a ConvNeXt model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.convnext.ConvNeXt</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.convnext_tiny.html#torchvision.models.convnext_tiny" title="torchvision.models.convnext_tiny"><code class="xref py py-obj docutils literal notranslate"><span class="pre">convnext_tiny</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>ConvNeXt Tiny model architecture from the <a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.convnext_small.html#torchvision.models.convnext_small" title="torchvision.models.convnext_small"><code class="xref py py-obj docutils literal notranslate"><span class="pre">convnext_small</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>ConvNeXt Small model architecture from the <a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.convnext_base.html#torchvision.models.convnext_base" title="torchvision.models.convnext_base"><code class="xref py py-obj docutils literal notranslate"><span class="pre">convnext_base</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>ConvNeXt Base model architecture from the <a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.convnext_large.html#torchvision.models.convnext_large" title="torchvision.models.convnext_large"><code class="xref py py-obj docutils literal notranslate"><span class="pre">convnext_large</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>ConvNeXt Large model architecture from the <a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.convnext_tiny.html" class="btn btn-neutral float-right" title="convnext_tiny" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.alexnet.html" class="btn btn-neutral" title="alexnet" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ConvNeXt</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/deeplabv3.html b/0.16/models/deeplabv3.html
new file mode 100644
index 00000000000..d53b85c6014
--- /dev/null
+++ b/0.16/models/deeplabv3.html
@@ -0,0 +1,718 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>DeepLabV3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="deeplabv3_mobilenet_v3_large" href="generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html" />
+    <link rel="prev" title="shufflenet_v2_x2_0" href="generated/torchvision.models.quantization.shufflenet_v2_x2_0.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>DeepLabV3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/deeplabv3.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="deeplabv3">
+<h1>DeepLabV3<a class="headerlink" href="#deeplabv3" title="Permalink to this heading">¶</a></h1>
+<p>The DeepLabV3 model is based on the <a class="reference external" href="https://arxiv.org/abs/1706.05587">Rethinking Atrous Convolution for Semantic
+Image Segmentation</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a DeepLabV3 model with
+different backbones, with or without pre-trained weights. All the model builders
+internally rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.segmentation.deeplabv3.DeepLabV3</span></code> base class. Please
+refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/deeplabv3.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html#torchvision.models.segmentation.deeplabv3_mobilenet_v3_large" title="torchvision.models.segmentation.deeplabv3_mobilenet_v3_large"><code class="xref py py-obj docutils literal notranslate"><span class="pre">deeplabv3_mobilenet_v3_large</span></code></a>(*[, weights, ...])</p></td>
+<td><p>Constructs a DeepLabV3 model with a MobileNetV3-Large backbone.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.segmentation.deeplabv3_resnet50.html#torchvision.models.segmentation.deeplabv3_resnet50" title="torchvision.models.segmentation.deeplabv3_resnet50"><code class="xref py py-obj docutils literal notranslate"><span class="pre">deeplabv3_resnet50</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Constructs a DeepLabV3 model with a ResNet-50 backbone.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.segmentation.deeplabv3_resnet101.html#torchvision.models.segmentation.deeplabv3_resnet101" title="torchvision.models.segmentation.deeplabv3_resnet101"><code class="xref py py-obj docutils literal notranslate"><span class="pre">deeplabv3_resnet101</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Constructs a DeepLabV3 model with a ResNet-101 backbone.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html" class="btn btn-neutral float-right" title="deeplabv3_mobilenet_v3_large" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.quantization.shufflenet_v2_x2_0.html" class="btn btn-neutral" title="shufflenet_v2_x2_0" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">DeepLabV3</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/densenet.html b/0.16/models/densenet.html
new file mode 100644
index 00000000000..70322ce77cd
--- /dev/null
+++ b/0.16/models/densenet.html
@@ -0,0 +1,720 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>DenseNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="densenet121" href="generated/torchvision.models.densenet121.html" />
+    <link rel="prev" title="convnext_large" href="generated/torchvision.models.convnext_large.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>DenseNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/densenet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="densenet">
+<h1>DenseNet<a class="headerlink" href="#densenet" title="Permalink to this heading">¶</a></h1>
+<p>The DenseNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a DenseNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.densenet.DenseNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.densenet121.html#torchvision.models.densenet121" title="torchvision.models.densenet121"><code class="xref py py-obj docutils literal notranslate"><span class="pre">densenet121</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Densenet-121 model from <a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.densenet161.html#torchvision.models.densenet161" title="torchvision.models.densenet161"><code class="xref py py-obj docutils literal notranslate"><span class="pre">densenet161</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Densenet-161 model from <a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.densenet169.html#torchvision.models.densenet169" title="torchvision.models.densenet169"><code class="xref py py-obj docutils literal notranslate"><span class="pre">densenet169</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Densenet-169 model from <a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.densenet201.html#torchvision.models.densenet201" title="torchvision.models.densenet201"><code class="xref py py-obj docutils literal notranslate"><span class="pre">densenet201</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Densenet-201 model from <a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.densenet121.html" class="btn btn-neutral float-right" title="densenet121" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.convnext_large.html" class="btn btn-neutral" title="convnext_large" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">DenseNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/efficientnet.html b/0.16/models/efficientnet.html
new file mode 100644
index 00000000000..f23391b0fa2
--- /dev/null
+++ b/0.16/models/efficientnet.html
@@ -0,0 +1,736 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>EfficientNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="efficientnet_b0" href="generated/torchvision.models.efficientnet_b0.html" />
+    <link rel="prev" title="densenet201" href="generated/torchvision.models.densenet201.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>EfficientNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/efficientnet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet">
+<h1>EfficientNet<a class="headerlink" href="#efficientnet" title="Permalink to this heading">¶</a></h1>
+<p>The EfficientNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an EfficientNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b0.html#torchvision.models.efficientnet_b0" title="torchvision.models.efficientnet_b0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b0</span></code></a>(*[, weights, progress])</p></td>
+<td><p>EfficientNet B0 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b1.html#torchvision.models.efficientnet_b1" title="torchvision.models.efficientnet_b1"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b1</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>EfficientNet B1 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b2.html#torchvision.models.efficientnet_b2" title="torchvision.models.efficientnet_b2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b2</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>EfficientNet B2 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b3.html#torchvision.models.efficientnet_b3" title="torchvision.models.efficientnet_b3"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b3</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>EfficientNet B3 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b4.html#torchvision.models.efficientnet_b4" title="torchvision.models.efficientnet_b4"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b4</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>EfficientNet B4 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b5.html#torchvision.models.efficientnet_b5" title="torchvision.models.efficientnet_b5"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b5</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>EfficientNet B5 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b6.html#torchvision.models.efficientnet_b6" title="torchvision.models.efficientnet_b6"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b6</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>EfficientNet B6 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_b7.html#torchvision.models.efficientnet_b7" title="torchvision.models.efficientnet_b7"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_b7</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>EfficientNet B7 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks</a> paper.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.efficientnet_b0.html" class="btn btn-neutral float-right" title="efficientnet_b0" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.densenet201.html" class="btn btn-neutral" title="densenet201" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">EfficientNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/efficientnetv2.html b/0.16/models/efficientnetv2.html
new file mode 100644
index 00000000000..9965b412435
--- /dev/null
+++ b/0.16/models/efficientnetv2.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>EfficientNetV2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="efficientnet_v2_s" href="generated/torchvision.models.efficientnet_v2_s.html" />
+    <link rel="prev" title="efficientnet_b7" href="generated/torchvision.models.efficientnet_b7.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>EfficientNetV2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/efficientnetv2.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnetv2">
+<h1>EfficientNetV2<a class="headerlink" href="#efficientnetv2" title="Permalink to this heading">¶</a></h1>
+<p>The EfficientNetV2 model is based on the <a class="reference external" href="https://arxiv.org/abs/2104.00298">EfficientNetV2: Smaller Models and Faster Training</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an EfficientNetV2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_v2_s.html#torchvision.models.efficientnet_v2_s" title="torchvision.models.efficientnet_v2_s"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_v2_s</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs an EfficientNetV2-S architecture from <a class="reference external" href="https://arxiv.org/abs/2104.00298">EfficientNetV2: Smaller Models and Faster Training</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_v2_m.html#torchvision.models.efficientnet_v2_m" title="torchvision.models.efficientnet_v2_m"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_v2_m</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs an EfficientNetV2-M architecture from <a class="reference external" href="https://arxiv.org/abs/2104.00298">EfficientNetV2: Smaller Models and Faster Training</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.efficientnet_v2_l.html#torchvision.models.efficientnet_v2_l" title="torchvision.models.efficientnet_v2_l"><code class="xref py py-obj docutils literal notranslate"><span class="pre">efficientnet_v2_l</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs an EfficientNetV2-L architecture from <a class="reference external" href="https://arxiv.org/abs/2104.00298">EfficientNetV2: Smaller Models and Faster Training</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.efficientnet_v2_s.html" class="btn btn-neutral float-right" title="efficientnet_v2_s" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.efficientnet_b7.html" class="btn btn-neutral" title="efficientnet_b7" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">EfficientNetV2</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/faster_rcnn.html b/0.16/models/faster_rcnn.html
new file mode 100644
index 00000000000..a88b8386bd9
--- /dev/null
+++ b/0.16/models/faster_rcnn.html
@@ -0,0 +1,722 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Faster R-CNN &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="fasterrcnn_resnet50_fpn" href="generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html" />
+    <link rel="prev" title="lraspp_mobilenet_v3_large" href="generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Faster R-CNN</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/faster_rcnn.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="faster-r-cnn">
+<h1>Faster R-CNN<a class="headerlink" href="#faster-r-cnn" title="Permalink to this heading">¶</a></h1>
+<p>The Faster R-CNN model is based on the <a class="reference external" href="https://arxiv.org/abs/1506.01497">Faster R-CNN: Towards Real-Time Object Detection
+with Region Proposal Networks</a>
+paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a Faster R-CNN model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.faster_rcnn.FasterRCNN</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="torchvision.models.detection.fasterrcnn_resnet50_fpn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn</span></code></a>(*[, weights, ...])</p></td>
+<td><p>Faster R-CNN model with a ResNet-50-FPN backbone from the <a class="reference external" href="https://arxiv.org/abs/1506.01497">Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks</a> paper.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html#torchvision.models.detection.fasterrcnn_resnet50_fpn_v2" title="torchvision.models.detection.fasterrcnn_resnet50_fpn_v2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn_v2</span></code></a>(*[, weights, ...])</p></td>
+<td><p>Constructs an improved Faster R-CNN model with a ResNet-50-FPN backbone from <a class="reference external" href="https://arxiv.org/abs/2111.11429">Benchmarking Detection Transfer Learning with Vision Transformers</a> paper.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn" title="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">fasterrcnn_mobilenet_v3_large_fpn</span></code></a>(*[, ...])</p></td>
+<td><p>Constructs a high resolution Faster R-CNN model with a MobileNetV3-Large FPN backbone.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn" title="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">fasterrcnn_mobilenet_v3_large_320_fpn</span></code></a>(*[, ...])</p></td>
+<td><p>Low resolution Faster R-CNN model with a MobileNetV3-Large backbone tuned for mobile use cases.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html" class="btn btn-neutral float-right" title="fasterrcnn_resnet50_fpn" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html" class="btn btn-neutral" title="lraspp_mobilenet_v3_large" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Faster R-CNN</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/fcn.html b/0.16/models/fcn.html
new file mode 100644
index 00000000000..ad6f73656a8
--- /dev/null
+++ b/0.16/models/fcn.html
@@ -0,0 +1,717 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FCN &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="fcn_resnet50" href="generated/torchvision.models.segmentation.fcn_resnet50.html" />
+    <link rel="prev" title="deeplabv3_resnet101" href="generated/torchvision.models.segmentation.deeplabv3_resnet101.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>FCN</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/fcn.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fcn">
+<h1>FCN<a class="headerlink" href="#fcn" title="Permalink to this heading">¶</a></h1>
+<p>The FCN model is based on the <a class="reference external" href="https://arxiv.org/abs/1411.4038">Fully Convolutional Networks for Semantic
+Segmentation</a>
+paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a FCN model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.segmentation.FCN</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.segmentation.fcn_resnet50.html#torchvision.models.segmentation.fcn_resnet50" title="torchvision.models.segmentation.fcn_resnet50"><code class="xref py py-obj docutils literal notranslate"><span class="pre">fcn_resnet50</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Fully-Convolutional Network model with a ResNet-50 backbone from the <a class="reference external" href="https://arxiv.org/abs/1411.4038">Fully Convolutional Networks for Semantic Segmentation</a> paper.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.segmentation.fcn_resnet101.html#torchvision.models.segmentation.fcn_resnet101" title="torchvision.models.segmentation.fcn_resnet101"><code class="xref py py-obj docutils literal notranslate"><span class="pre">fcn_resnet101</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p><p>Fully-Convolutional Network model with a ResNet-101 backbone from the <a class="reference external" href="https://arxiv.org/abs/1411.4038">Fully Convolutional Networks for Semantic Segmentation</a> paper.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.segmentation.fcn_resnet50.html" class="btn btn-neutral float-right" title="fcn_resnet50" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.segmentation.deeplabv3_resnet101.html" class="btn btn-neutral" title="deeplabv3_resnet101" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FCN</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/fcos.html b/0.16/models/fcos.html
new file mode 100644
index 00000000000..f4aa37a20db
--- /dev/null
+++ b/0.16/models/fcos.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>FCOS &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="fcos_resnet50_fpn" href="generated/torchvision.models.detection.fcos_resnet50_fpn.html" />
+    <link rel="prev" title="fasterrcnn_mobilenet_v3_large_320_fpn" href="generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>FCOS</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/fcos.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fcos">
+<h1>FCOS<a class="headerlink" href="#fcos" title="Permalink to this heading">¶</a></h1>
+<p>The FCOS model is based on the <a class="reference external" href="https://arxiv.org/abs/1904.01355">FCOS: Fully Convolutional One-Stage Object Detection</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a FCOS model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.fcos.FCOS</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/fcos.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.fcos_resnet50_fpn.html#torchvision.models.detection.fcos_resnet50_fpn" title="torchvision.models.detection.fcos_resnet50_fpn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">fcos_resnet50_fpn</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Constructs a FCOS model with a ResNet-50-FPN backbone.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.detection.fcos_resnet50_fpn.html" class="btn btn-neutral float-right" title="fcos_resnet50_fpn" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html" class="btn btn-neutral" title="fasterrcnn_mobilenet_v3_large_320_fpn" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">FCOS</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.alexnet.html b/0.16/models/generated/torchvision.models.alexnet.html
new file mode 100644
index 00000000000..76a8c2df1ae
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.alexnet.html
@@ -0,0 +1,760 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>alexnet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="ConvNeXt" href="../convnext.html" />
+    <link rel="prev" title="AlexNet" href="../alexnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../alexnet.html">AlexNet</a> &gt;</li>
+        
+      <li>alexnet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.alexnet.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="alexnet">
+<h1>alexnet<a class="headerlink" href="#alexnet" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.alexnet">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">alexnet</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.AlexNet_Weights" title="torchvision.models.alexnet.AlexNet_Weights"><span class="pre">AlexNet_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">AlexNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/alexnet.html#alexnet"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.alexnet" title="Permalink to this definition">¶</a></dt>
+<dd><p>AlexNet model architecture from <a class="reference external" href="https://arxiv.org/abs/1404.5997">One weird trick for parallelizing convolutional neural networks</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>AlexNet was originally introduced in the <a class="reference external" href="https://papers.nips.cc/paper/2012/hash/c399862d3b9d6b76c8436e924a68c45b-Abstract.html">ImageNet Classification with
+Deep Convolutional Neural Networks</a>
+paper. Our implementation is based instead on the “One weird trick”
+paper above.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.AlexNet_Weights" title="torchvision.models.AlexNet_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">AlexNet_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.AlexNet_Weights" title="torchvision.models.AlexNet_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">AlexNet_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.squeezenet.AlexNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/alexnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.AlexNet_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">AlexNet_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/alexnet.html#AlexNet_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.AlexNet_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">AlexNet_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">AlexNet_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>AlexNet_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights reproduce closely the results of the paper using a simplified training recipe.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">AlexNet_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>56.522</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>79.066</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>61100840</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=63, width=63</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.71</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>233.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">AlexNet_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../convnext.html" class="btn btn-neutral float-right" title="ConvNeXt" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../alexnet.html" class="btn btn-neutral" title="AlexNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">alexnet</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.convnext_base.html b/0.16/models/generated/torchvision.models.convnext_base.html
new file mode 100644
index 00000000000..3c975ed1841
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.convnext_base.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>convnext_base &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="convnext_large" href="torchvision.models.convnext_large.html" />
+    <link rel="prev" title="convnext_small" href="torchvision.models.convnext_small.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../convnext.html">ConvNeXt</a> &gt;</li>
+        
+      <li>convnext_base</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.convnext_base.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convnext-base">
+<h1>convnext_base<a class="headerlink" href="#convnext-base" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.convnext_base">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">convnext_base</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ConvNeXt_Base_Weights" title="torchvision.models.convnext.ConvNeXt_Base_Weights"><span class="pre">ConvNeXt_Base_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ConvNeXt</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#convnext_base"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.convnext_base" title="Permalink to this definition">¶</a></dt>
+<dd><p>ConvNeXt Base model architecture from the
+<a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ConvNeXt_Base_Weights" title="torchvision.models.convnext.ConvNeXt_Base_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ConvNeXt_Base_Weights" title="torchvision.models.convnext.ConvNeXt_Base_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.convnext.ConvNext</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ConvNeXt_Base_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ConvNeXt_Base_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#ConvNeXt_Base_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ConvNeXt_Base_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ConvNeXt_Base_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+        These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+        <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>84.062</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.87</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>88591464</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.36</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>338.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Base_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.convnext_large.html" class="btn btn-neutral float-right" title="convnext_large" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.convnext_small.html" class="btn btn-neutral" title="convnext_small" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">convnext_base</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.convnext_large.html b/0.16/models/generated/torchvision.models.convnext_large.html
new file mode 100644
index 00000000000..938d36696b5
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.convnext_large.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>convnext_large &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="DenseNet" href="../densenet.html" />
+    <link rel="prev" title="convnext_base" href="torchvision.models.convnext_base.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../convnext.html">ConvNeXt</a> &gt;</li>
+        
+      <li>convnext_large</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.convnext_large.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convnext-large">
+<h1>convnext_large<a class="headerlink" href="#convnext-large" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.convnext_large">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">convnext_large</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ConvNeXt_Large_Weights" title="torchvision.models.convnext.ConvNeXt_Large_Weights"><span class="pre">ConvNeXt_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ConvNeXt</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#convnext_large"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.convnext_large" title="Permalink to this definition">¶</a></dt>
+<dd><p>ConvNeXt Large model architecture from the
+<a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ConvNeXt_Large_Weights" title="torchvision.models.convnext.ConvNeXt_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ConvNeXt_Large_Weights" title="torchvision.models.convnext.ConvNeXt_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.convnext.ConvNext</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ConvNeXt_Large_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ConvNeXt_Large_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#ConvNeXt_Large_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ConvNeXt_Large_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ConvNeXt_Large_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+        These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+        <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>84.414</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.976</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>197767336</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>34.36</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>754.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Large_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../densenet.html" class="btn btn-neutral float-right" title="DenseNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.convnext_base.html" class="btn btn-neutral" title="convnext_base" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">convnext_large</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.convnext_small.html b/0.16/models/generated/torchvision.models.convnext_small.html
new file mode 100644
index 00000000000..ae0add8b00a
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.convnext_small.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>convnext_small &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="convnext_base" href="torchvision.models.convnext_base.html" />
+    <link rel="prev" title="convnext_tiny" href="torchvision.models.convnext_tiny.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../convnext.html">ConvNeXt</a> &gt;</li>
+        
+      <li>convnext_small</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.convnext_small.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convnext-small">
+<h1>convnext_small<a class="headerlink" href="#convnext-small" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.convnext_small">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">convnext_small</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ConvNeXt_Small_Weights" title="torchvision.models.convnext.ConvNeXt_Small_Weights"><span class="pre">ConvNeXt_Small_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ConvNeXt</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#convnext_small"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.convnext_small" title="Permalink to this definition">¶</a></dt>
+<dd><p>ConvNeXt Small model architecture from the
+<a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ConvNeXt_Small_Weights" title="torchvision.models.convnext.ConvNeXt_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ConvNeXt_Small_Weights" title="torchvision.models.convnext.ConvNeXt_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.convnext.ConvNext</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ConvNeXt_Small_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ConvNeXt_Small_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#ConvNeXt_Small_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ConvNeXt_Small_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ConvNeXt_Small_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+        These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+        <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.616</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.65</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>50223688</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>8.68</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>191.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Small_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[230]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.convnext_base.html" class="btn btn-neutral float-right" title="convnext_base" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.convnext_tiny.html" class="btn btn-neutral" title="convnext_tiny" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">convnext_small</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.convnext_tiny.html b/0.16/models/generated/torchvision.models.convnext_tiny.html
new file mode 100644
index 00000000000..4bc21bca1e6
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.convnext_tiny.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>convnext_tiny &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="convnext_small" href="torchvision.models.convnext_small.html" />
+    <link rel="prev" title="ConvNeXt" href="../convnext.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../convnext.html">ConvNeXt</a> &gt;</li>
+        
+      <li>convnext_tiny</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.convnext_tiny.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="convnext-tiny">
+<h1>convnext_tiny<a class="headerlink" href="#convnext-tiny" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.convnext_tiny">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">convnext_tiny</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ConvNeXt_Tiny_Weights" title="torchvision.models.convnext.ConvNeXt_Tiny_Weights"><span class="pre">ConvNeXt_Tiny_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ConvNeXt</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#convnext_tiny"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.convnext_tiny" title="Permalink to this definition">¶</a></dt>
+<dd><p>ConvNeXt Tiny model architecture from the
+<a class="reference external" href="https://arxiv.org/abs/2201.03545">A ConvNet for the 2020s</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ConvNeXt_Tiny_Weights" title="torchvision.models.convnext.ConvNeXt_Tiny_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ConvNeXt_Tiny_Weights" title="torchvision.models.convnext.ConvNeXt_Tiny_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.convnext.ConvNext</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ConvNeXt_Tiny_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ConvNeXt_Tiny_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/convnext.html#ConvNeXt_Tiny_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ConvNeXt_Tiny_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ConvNeXt_Tiny_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+        These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+        <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.52</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.146</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#convnext">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>28589128</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.46</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>109.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ConvNeXt_Tiny_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[236]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.convnext_small.html" class="btn btn-neutral float-right" title="convnext_small" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../convnext.html" class="btn btn-neutral" title="ConvNeXt" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">convnext_tiny</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.densenet121.html b/0.16/models/generated/torchvision.models.densenet121.html
new file mode 100644
index 00000000000..98d04556aab
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.densenet121.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>densenet121 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="densenet161" href="torchvision.models.densenet161.html" />
+    <link rel="prev" title="DenseNet" href="../densenet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../densenet.html">DenseNet</a> &gt;</li>
+        
+      <li>densenet121</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.densenet121.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="densenet121">
+<h1>densenet121<a class="headerlink" href="#densenet121" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.densenet121">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">densenet121</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.DenseNet121_Weights" title="torchvision.models.densenet.DenseNet121_Weights"><span class="pre">DenseNet121_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">DenseNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#densenet121"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.densenet121" title="Permalink to this definition">¶</a></dt>
+<dd><p>Densenet-121 model from
+<a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.DenseNet121_Weights" title="torchvision.models.DenseNet121_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet121_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.DenseNet121_Weights" title="torchvision.models.DenseNet121_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet121_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.densenet.DenseNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.DenseNet121_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">DenseNet121_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#DenseNet121_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.DenseNet121_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">DenseNet121_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">DenseNet121_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>DenseNet121_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from LuaTorch.
+Also available as <code class="docutils literal notranslate"><span class="pre">DenseNet121_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>74.434</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.972</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=29, width=29</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>7978856</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>2.83</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>30.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">DenseNet121_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.densenet161.html" class="btn btn-neutral float-right" title="densenet161" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../densenet.html" class="btn btn-neutral" title="DenseNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">densenet121</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.densenet161.html b/0.16/models/generated/torchvision.models.densenet161.html
new file mode 100644
index 00000000000..9c77e1e6447
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.densenet161.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>densenet161 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="densenet169" href="torchvision.models.densenet169.html" />
+    <link rel="prev" title="densenet121" href="torchvision.models.densenet121.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../densenet.html">DenseNet</a> &gt;</li>
+        
+      <li>densenet161</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.densenet161.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="densenet161">
+<h1>densenet161<a class="headerlink" href="#densenet161" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.densenet161">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">densenet161</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.DenseNet161_Weights" title="torchvision.models.densenet.DenseNet161_Weights"><span class="pre">DenseNet161_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">DenseNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#densenet161"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.densenet161" title="Permalink to this definition">¶</a></dt>
+<dd><p>Densenet-161 model from
+<a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.DenseNet161_Weights" title="torchvision.models.DenseNet161_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet161_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.DenseNet161_Weights" title="torchvision.models.DenseNet161_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet161_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.densenet.DenseNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.DenseNet161_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">DenseNet161_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#DenseNet161_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.DenseNet161_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">DenseNet161_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">DenseNet161_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>DenseNet161_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from LuaTorch.
+Also available as <code class="docutils literal notranslate"><span class="pre">DenseNet161_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.138</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.56</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=29, width=29</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>28681000</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>7.73</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>110.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">DenseNet161_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.densenet169.html" class="btn btn-neutral float-right" title="densenet169" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.densenet121.html" class="btn btn-neutral" title="densenet121" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">densenet161</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.densenet169.html b/0.16/models/generated/torchvision.models.densenet169.html
new file mode 100644
index 00000000000..5e0dab7049f
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.densenet169.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>densenet169 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="densenet201" href="torchvision.models.densenet201.html" />
+    <link rel="prev" title="densenet161" href="torchvision.models.densenet161.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../densenet.html">DenseNet</a> &gt;</li>
+        
+      <li>densenet169</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.densenet169.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="densenet169">
+<h1>densenet169<a class="headerlink" href="#densenet169" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.densenet169">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">densenet169</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.DenseNet169_Weights" title="torchvision.models.densenet.DenseNet169_Weights"><span class="pre">DenseNet169_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">DenseNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#densenet169"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.densenet169" title="Permalink to this definition">¶</a></dt>
+<dd><p>Densenet-169 model from
+<a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.DenseNet169_Weights" title="torchvision.models.DenseNet169_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet169_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.DenseNet169_Weights" title="torchvision.models.DenseNet169_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet169_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.densenet.DenseNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.DenseNet169_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">DenseNet169_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#DenseNet169_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.DenseNet169_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">DenseNet169_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">DenseNet169_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>DenseNet169_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from LuaTorch.
+Also available as <code class="docutils literal notranslate"><span class="pre">DenseNet169_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.6</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.806</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=29, width=29</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>14149480</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>3.36</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>54.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">DenseNet169_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.densenet201.html" class="btn btn-neutral float-right" title="densenet201" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.densenet161.html" class="btn btn-neutral" title="densenet161" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">densenet169</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.densenet201.html b/0.16/models/generated/torchvision.models.densenet201.html
new file mode 100644
index 00000000000..53bbc9b508c
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.densenet201.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>densenet201 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="EfficientNet" href="../efficientnet.html" />
+    <link rel="prev" title="densenet169" href="torchvision.models.densenet169.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../densenet.html">DenseNet</a> &gt;</li>
+        
+      <li>densenet201</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.densenet201.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="densenet201">
+<h1>densenet201<a class="headerlink" href="#densenet201" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.densenet201">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">densenet201</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.DenseNet201_Weights" title="torchvision.models.densenet.DenseNet201_Weights"><span class="pre">DenseNet201_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">DenseNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#densenet201"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.densenet201" title="Permalink to this definition">¶</a></dt>
+<dd><p>Densenet-201 model from
+<a class="reference external" href="https://arxiv.org/abs/1608.06993">Densely Connected Convolutional Networks</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.DenseNet201_Weights" title="torchvision.models.DenseNet201_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet201_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.DenseNet201_Weights" title="torchvision.models.DenseNet201_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DenseNet201_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.densenet.DenseNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/densenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.DenseNet201_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">DenseNet201_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/densenet.html#DenseNet201_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.DenseNet201_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">DenseNet201_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">DenseNet201_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>DenseNet201_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from LuaTorch.
+Also available as <code class="docutils literal notranslate"><span class="pre">DenseNet201_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.896</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.37</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=29, width=29</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/116">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>20013928</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.29</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>77.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">DenseNet201_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../efficientnet.html" class="btn btn-neutral float-right" title="EfficientNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.densenet169.html" class="btn btn-neutral" title="densenet169" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">densenet201</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
new file mode 100644
index 00000000000..068b5d0724d
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
@@ -0,0 +1,768 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>fasterrcnn_mobilenet_v3_large_320_fpn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="FCOS" href="../fcos.html" />
+    <link rel="prev" title="fasterrcnn_mobilenet_v3_large_fpn" href="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../faster_rcnn.html">Faster R-CNN</a> &gt;</li>
+        
+      <li>fasterrcnn_mobilenet_v3_large_320_fpn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fasterrcnn-mobilenet-v3-large-320-fpn">
+<h1>fasterrcnn_mobilenet_v3_large_320_fpn<a class="headerlink" href="#fasterrcnn-mobilenet-v3-large-320-fpn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">fasterrcnn_mobilenet_v3_large_320_fpn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights" title="torchvision.models.detection.faster_rcnn.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Large_Weights"><span class="pre">MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">FasterRCNN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#fasterrcnn_mobilenet_v3_large_320_fpn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn" title="Permalink to this definition">¶</a></dt>
+<dd><p>Low resolution Faster R-CNN model with a MobileNetV3-Large backbone tuned for mobile use cases.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>It works similarly to Faster R-CNN with ResNet-50 FPN backbone. See
+<a class="reference internal" href="torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="torchvision.models.detection.fasterrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn()</span></code></a> for more
+details.</p>
+<p>Example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">fasterrcnn_mobilenet_v3_large_320_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">400</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights" title="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights" title="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a>, optional) – The
+pretrained weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from
+final block. Valid values are between 0 and 6, with 6 meaning all backbone layers are
+trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.faster_rcnn.FasterRCNN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#FasterRCNN_MobileNet_V3_Large_320_FPN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_320_FPN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>22.8</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>19386354</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-320-fpn">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>0.72</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>74.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_320_FPN_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../fcos.html" class="btn btn-neutral float-right" title="FCOS" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html" class="btn btn-neutral" title="fasterrcnn_mobilenet_v3_large_fpn" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">fasterrcnn_mobilenet_v3_large_320_fpn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
new file mode 100644
index 00000000000..5ddc71f1fa3
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
@@ -0,0 +1,768 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>fasterrcnn_mobilenet_v3_large_fpn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="fasterrcnn_mobilenet_v3_large_320_fpn" href="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html" />
+    <link rel="prev" title="fasterrcnn_resnet50_fpn_v2" href="torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../faster_rcnn.html">Faster R-CNN</a> &gt;</li>
+        
+      <li>fasterrcnn_mobilenet_v3_large_fpn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fasterrcnn-mobilenet-v3-large-fpn">
+<h1>fasterrcnn_mobilenet_v3_large_fpn<a class="headerlink" href="#fasterrcnn-mobilenet-v3-large-fpn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">fasterrcnn_mobilenet_v3_large_fpn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights" title="torchvision.models.detection.faster_rcnn.FasterRCNN_MobileNet_V3_Large_FPN_Weights"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Large_Weights"><span class="pre">MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">FasterRCNN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#fasterrcnn_mobilenet_v3_large_fpn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a high resolution Faster R-CNN model with a MobileNetV3-Large FPN backbone.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>It works similarly to Faster R-CNN with ResNet-50 FPN backbone. See
+<a class="reference internal" href="torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="torchvision.models.detection.fasterrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn()</span></code></a> for more
+details.</p>
+<p>Example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">fasterrcnn_mobilenet_v3_large_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">400</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights" title="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights" title="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a>, optional) – The
+pretrained weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from
+final block. Valid values are between 0 and 6, with 6 meaning all backbone layers are
+trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.faster_rcnn.FasterRCNN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#FasterRCNN_MobileNet_V3_Large_FPN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.FasterRCNN_MobileNet_V3_Large_FPN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>FasterRCNN_MobileNet_V3_Large_FPN_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>32.8</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>19386354</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-mobilenetv3-large-fpn">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>4.49</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>74.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">FasterRCNN_MobileNet_V3_Large_FPN_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html" class="btn btn-neutral float-right" title="fasterrcnn_mobilenet_v3_large_320_fpn" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html" class="btn btn-neutral" title="fasterrcnn_resnet50_fpn_v2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">fasterrcnn_mobilenet_v3_large_fpn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
new file mode 100644
index 00000000000..a373428e761
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
@@ -0,0 +1,817 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>fasterrcnn_resnet50_fpn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="fasterrcnn_resnet50_fpn_v2" href="torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html" />
+    <link rel="prev" title="Faster R-CNN" href="../faster_rcnn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../faster_rcnn.html">Faster R-CNN</a> &gt;</li>
+        
+      <li>fasterrcnn_resnet50_fpn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fasterrcnn-resnet50-fpn">
+<h1>fasterrcnn_resnet50_fpn<a class="headerlink" href="#fasterrcnn-resnet50-fpn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.fasterrcnn_resnet50_fpn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">fasterrcnn_resnet50_fpn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.faster_rcnn.FasterRCNN_ResNet50_FPN_Weights"><span class="pre">FasterRCNN_ResNet50_FPN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">FasterRCNN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#fasterrcnn_resnet50_fpn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="Permalink to this definition">¶</a></dt>
+<dd><p>Faster R-CNN model with a ResNet-50-FPN backbone from the <a class="reference external" href="https://arxiv.org/abs/1506.01497">Faster R-CNN: Towards Real-Time Object
+Detection with Region Proposal Networks</a>
+paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>The input to the model is expected to be a list of tensors, each of shape <code class="docutils literal notranslate"><span class="pre">[C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>, one for each
+image, and should be in <code class="docutils literal notranslate"><span class="pre">0-1</span></code> range. Different images can have different sizes.</p>
+<p>The behavior of the model changes depending on if it is in training or evaluation mode.</p>
+<p>During training, the model expects both the input tensors and a targets (list of dictionary),
+containing:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the ground-truth boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the class label for each ground-truth box</p></li>
+</ul>
+</div></blockquote>
+<p>The model returns a <code class="docutils literal notranslate"><span class="pre">Dict[Tensor]</span></code> during training, containing the classification and regression
+losses for both the RPN and the R-CNN.</p>
+<p>During inference, the model requires only the input tensors, and returns the post-processed
+predictions as a <code class="docutils literal notranslate"><span class="pre">List[Dict[Tensor]]</span></code>, one for each input image. The fields of the <code class="docutils literal notranslate"><span class="pre">Dict</span></code> are as
+follows, where <code class="docutils literal notranslate"><span class="pre">N</span></code> is the number of detections:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the predicted boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the predicted labels for each detection</p></li>
+<li><p>scores (<code class="docutils literal notranslate"><span class="pre">Tensor[N]</span></code>): the scores of each detection</p></li>
+</ul>
+</div></blockquote>
+<p>For more details on the output, you may refer to <a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<p>Faster R-CNN is exportable to ONNX for a fixed batch size with inputs images of fixed size.</p>
+<p>Example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">fasterrcnn_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">FasterRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># For training</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">images</span><span class="p">,</span> <span class="n">boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">600</span><span class="p">,</span> <span class="mi">1200</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">11</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">boxes</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">2</span><span class="p">:</span><span class="mi">4</span><span class="p">]</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">:</span><span class="mi">2</span><span class="p">]</span> <span class="o">+</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">2</span><span class="p">:</span><span class="mi">4</span><span class="p">]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">91</span><span class="p">,</span> <span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="mi">11</span><span class="p">))</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">images</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">image</span> <span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">images</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">targets</span> <span class="o">=</span> <span class="p">[]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">images</span><span class="p">)):</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">d</span> <span class="o">=</span> <span class="p">{}</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">d</span><span class="p">[</span><span class="s1">&#39;boxes&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">boxes</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">d</span><span class="p">[</span><span class="s1">&#39;labels&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+<span class="gp">&gt;&gt;&gt; </span>    <span class="n">targets</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">d</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">output</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">targets</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># For inference</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">400</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># optionally, if you want to export the model to ONNX:</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">torch</span><span class="o">.</span><span class="n">onnx</span><span class="o">.</span><span class="n">export</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="s2">&quot;faster_rcnn.onnx&quot;</span><span class="p">,</span> <span class="n">opset_version</span> <span class="o">=</span> <span class="mi">11</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from
+final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are
+trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.faster_rcnn.FasterRCNN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">FasterRCNN_ResNet50_FPN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#FasterRCNN_ResNet50_FPN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>FasterRCNN_ResNet50_FPN_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>37.0</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>41755286</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#faster-r-cnn-resnet-50-fpn">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>134.38</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>159.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="The following example illustrates the operations available the ops module for repurposing segme..."><img alt="" src="../../_images/sphx_glr_plot_repurposing_annotations_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_repurposing_annotations.html#sphx-glr-auto-examples-others-plot-repurposing-annotations-py"><span class="std std-ref">Repurposing masks into bounding boxes</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Repurposing masks into bounding boxes</div>
+</div><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html" class="btn btn-neutral float-right" title="fasterrcnn_resnet50_fpn_v2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../faster_rcnn.html" class="btn btn-neutral" title="Faster R-CNN" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">fasterrcnn_resnet50_fpn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
new file mode 100644
index 00000000000..fcbe7a58b4e
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
@@ -0,0 +1,762 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>fasterrcnn_resnet50_fpn_v2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="fasterrcnn_mobilenet_v3_large_fpn" href="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html" />
+    <link rel="prev" title="fasterrcnn_resnet50_fpn" href="torchvision.models.detection.fasterrcnn_resnet50_fpn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../faster_rcnn.html">Faster R-CNN</a> &gt;</li>
+        
+      <li>fasterrcnn_resnet50_fpn_v2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fasterrcnn-resnet50-fpn-v2">
+<h1>fasterrcnn_resnet50_fpn_v2<a class="headerlink" href="#fasterrcnn-resnet50-fpn-v2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.fasterrcnn_resnet50_fpn_v2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">fasterrcnn_resnet50_fpn_v2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.faster_rcnn.FasterRCNN_ResNet50_FPN_V2_Weights"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">FasterRCNN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#fasterrcnn_resnet50_fpn_v2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.fasterrcnn_resnet50_fpn_v2" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs an improved Faster R-CNN model with a ResNet-50-FPN backbone from <a class="reference external" href="https://arxiv.org/abs/2111.11429">Benchmarking Detection
+Transfer Learning with Vision Transformers</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>It works similarly to Faster R-CNN with ResNet-50 FPN backbone. See
+<a class="reference internal" href="torchvision.models.detection.fasterrcnn_resnet50_fpn.html#torchvision.models.detection.fasterrcnn_resnet50_fpn" title="torchvision.models.detection.fasterrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">fasterrcnn_resnet50_fpn()</span></code></a> for more
+details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from
+final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are
+trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.faster_rcnn.FasterRCNN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/faster_rcnn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/faster_rcnn.html#FasterRCNN_ResNet50_FPN_V2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.FasterRCNN_ResNet50_FPN_V2_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>FasterRCNN_ResNet50_FPN_V2_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced using an enhanced training recipe to boost the model accuracy.
+Also available as <code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>46.7</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>43712278</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5763">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>280.37</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>167.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">FasterRCNN_ResNet50_FPN_V2_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html" class="btn btn-neutral float-right" title="fasterrcnn_mobilenet_v3_large_fpn" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.detection.fasterrcnn_resnet50_fpn.html" class="btn btn-neutral" title="fasterrcnn_resnet50_fpn" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">fasterrcnn_resnet50_fpn_v2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
new file mode 100644
index 00000000000..3a6f140e51c
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
@@ -0,0 +1,794 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>fcos_resnet50_fpn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="RetinaNet" href="../retinanet.html" />
+    <link rel="prev" title="FCOS" href="../fcos.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../fcos.html">FCOS</a> &gt;</li>
+        
+      <li>fcos_resnet50_fpn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.fcos_resnet50_fpn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fcos-resnet50-fpn">
+<h1>fcos_resnet50_fpn<a class="headerlink" href="#fcos-resnet50-fpn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.fcos_resnet50_fpn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">fcos_resnet50_fpn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.FCOS_ResNet50_FPN_Weights" title="torchvision.models.detection.fcos.FCOS_ResNet50_FPN_Weights"><span class="pre">FCOS_ResNet50_FPN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">FCOS</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/fcos.html#fcos_resnet50_fpn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.fcos_resnet50_fpn" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a FCOS model with a ResNet-50-FPN backbone.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<dl class="simple">
+<dt>Reference: <a class="reference external" href="https://arxiv.org/abs/1904.01355">FCOS: Fully Convolutional One-Stage Object Detection</a>.</dt><dd><p><a class="reference external" href="https://arxiv.org/abs/2006.09214">FCOS: A simple and strong anchor-free object detector</a>.</p>
+</dd>
+</dl>
+<p>The input to the model is expected to be a list of tensors, each of shape <code class="docutils literal notranslate"><span class="pre">[C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>, one for each
+image, and should be in <code class="docutils literal notranslate"><span class="pre">0-1</span></code> range. Different images can have different sizes.</p>
+<p>The behavior of the model changes depending on if it is in training or evaluation mode.</p>
+<p>During training, the model expects both the input tensors and targets (list of dictionary),
+containing:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the ground-truth boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the class label for each ground-truth box</p></li>
+</ul>
+</div></blockquote>
+<p>The model returns a <code class="docutils literal notranslate"><span class="pre">Dict[Tensor]</span></code> during training, containing the classification and regression
+losses.</p>
+<p>During inference, the model requires only the input tensors, and returns the post-processed
+predictions as a <code class="docutils literal notranslate"><span class="pre">List[Dict[Tensor]]</span></code>, one for each input image. The fields of the <code class="docutils literal notranslate"><span class="pre">Dict</span></code> are as
+follows, where <code class="docutils literal notranslate"><span class="pre">N</span></code> is the number of detections:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the predicted boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the predicted labels for each detection</p></li>
+<li><p>scores (<code class="docutils literal notranslate"><span class="pre">Tensor[N]</span></code>): the scores of each detection</p></li>
+</ul>
+</div></blockquote>
+<p>For more details on the output, you may refer to <a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">fcos_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">FCOS_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">400</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.FCOS_ResNet50_FPN_Weights" title="torchvision.models.detection.FCOS_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.FCOS_ResNet50_FPN_Weights" title="torchvision.models.detection.FCOS_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The pretrained weights for
+the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) resnet layers starting
+from final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are
+trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is set to 3. Default: None</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.FCOS</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/fcos.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.FCOS_ResNet50_FPN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">FCOS_ResNet50_FPN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/fcos.html#FCOS_ResNet50_FPN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.FCOS_ResNet50_FPN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>FCOS_ResNet50_FPN_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>39.2</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>32269600</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#fcos-resnet-50-fpn">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>128.21</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>123.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">FCOS_ResNet50_FPN_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../retinanet.html" class="btn btn-neutral float-right" title="RetinaNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../fcos.html" class="btn btn-neutral" title="FCOS" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">fcos_resnet50_fpn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
new file mode 100644
index 00000000000..8f2e1c0c9cb
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
@@ -0,0 +1,842 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>keypointrcnn_resnet50_fpn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Video MViT" href="../video_mvit.html" />
+    <link rel="prev" title="Keypoint R-CNN" href="../keypoint_rcnn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../keypoint_rcnn.html">Keypoint R-CNN</a> &gt;</li>
+        
+      <li>keypointrcnn_resnet50_fpn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="keypointrcnn-resnet50-fpn">
+<h1>keypointrcnn_resnet50_fpn<a class="headerlink" href="#keypointrcnn-resnet50-fpn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.keypointrcnn_resnet50_fpn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">keypointrcnn_resnet50_fpn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.keypoint_rcnn.KeypointRCNN_ResNet50_FPN_Weights"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_keypoints</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">KeypointRCNN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/keypoint_rcnn.html#keypointrcnn_resnet50_fpn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.keypointrcnn_resnet50_fpn" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a Keypoint R-CNN model with a ResNet-50-FPN backbone.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1703.06870">Mask R-CNN</a>.</p>
+<p>The input to the model is expected to be a list of tensors, each of shape <code class="docutils literal notranslate"><span class="pre">[C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>, one for each
+image, and should be in <code class="docutils literal notranslate"><span class="pre">0-1</span></code> range. Different images can have different sizes.</p>
+<p>The behavior of the model changes depending on if it is in training or evaluation mode.</p>
+<p>During training, the model expects both the input tensors and targets (list of dictionary),
+containing:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the ground-truth boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the class label for each ground-truth box</p></li>
+<li><p>keypoints (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">K,</span> <span class="pre">3]</span></code>): the <code class="docutils literal notranslate"><span class="pre">K</span></code> keypoints location for each of the <code class="docutils literal notranslate"><span class="pre">N</span></code> instances, in the
+format <code class="docutils literal notranslate"><span class="pre">[x,</span> <span class="pre">y,</span> <span class="pre">visibility]</span></code>, where <code class="docutils literal notranslate"><span class="pre">visibility=0</span></code> means that the keypoint is not visible.</p></li>
+</ul>
+</div></blockquote>
+<p>The model returns a <code class="docutils literal notranslate"><span class="pre">Dict[Tensor]</span></code> during training, containing the classification and regression
+losses for both the RPN and the R-CNN, and the keypoint loss.</p>
+<p>During inference, the model requires only the input tensors, and returns the post-processed
+predictions as a <code class="docutils literal notranslate"><span class="pre">List[Dict[Tensor]]</span></code>, one for each input image. The fields of the <code class="docutils literal notranslate"><span class="pre">Dict</span></code> are as
+follows, where <code class="docutils literal notranslate"><span class="pre">N</span></code> is the number of detected instances:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the predicted boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the predicted labels for each instance</p></li>
+<li><p>scores (<code class="docutils literal notranslate"><span class="pre">Tensor[N]</span></code>): the scores or each instance</p></li>
+<li><p>keypoints (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">K,</span> <span class="pre">3]</span></code>): the locations of the predicted keypoints, in <code class="docutils literal notranslate"><span class="pre">[x,</span> <span class="pre">y,</span> <span class="pre">v]</span></code> format.</p></li>
+</ul>
+</div></blockquote>
+<p>For more details on the output, you may refer to <a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<p>Keypoint R-CNN is exportable to ONNX for a fixed batch size with inputs images of fixed size.</p>
+<p>Example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">keypointrcnn_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">KeypointRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">400</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># optionally, if you want to export the model to ONNX:</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">torch</span><span class="o">.</span><span class="n">onnx</span><span class="o">.</span><span class="n">export</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="s2">&quot;keypoint_rcnn.onnx&quot;</span><span class="p">,</span> <span class="n">opset_version</span> <span class="o">=</span> <span class="mi">11</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>num_keypoints</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of keypoints</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from final block.
+Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is
+passed (the default) this value is set to 3.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/keypoint_rcnn.html#KeypointRCNN_ResNet50_FPN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.KeypointRCNN_ResNet50_FPN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_LEGACY'</span></code>.</p>
+<p><strong>KeypointRCNN_ResNet50_FPN_Weights.COCO_LEGACY</strong>:</p>
+<p>
+                These weights were produced by following a similar training recipe as on the paper but use a checkpoint
+                from an early epoch.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>50.6</p></td>
+</tr>
+<tr class="row-even"><td><p>kp_map (on COCO-val2017)</p></td>
+<td><p>61.1</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>no person, person</p></td>
+</tr>
+<tr class="row-even"><td><p>keypoint_names</p></td>
+<td><p>nose, left_eye, right_eye, … (14 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>59137258</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/1606">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>133.92</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>226.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.COCO_LEGACY.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+<p><strong>KeypointRCNN_ResNet50_FPN_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>54.6</p></td>
+</tr>
+<tr class="row-even"><td><p>kp_map (on COCO-val2017)</p></td>
+<td><p>65.0</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>no person, person</p></td>
+</tr>
+<tr class="row-even"><td><p>keypoint_names</p></td>
+<td><p>nose, left_eye, right_eye, … (14 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>59137258</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#keypoint-r-cnn">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>137.42</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>226.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">KeypointRCNN_ResNet50_FPN_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">keypointrcnn_resnet50_fpn</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../video_mvit.html" class="btn btn-neutral float-right" title="Video MViT" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../keypoint_rcnn.html" class="btn btn-neutral" title="Keypoint R-CNN" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">keypointrcnn_resnet50_fpn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
new file mode 100644
index 00000000000..3e5638191e2
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
@@ -0,0 +1,806 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>maskrcnn_resnet50_fpn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="maskrcnn_resnet50_fpn_v2" href="torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html" />
+    <link rel="prev" title="Mask R-CNN" href="../mask_rcnn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mask_rcnn.html">Mask R-CNN</a> &gt;</li>
+        
+      <li>maskrcnn_resnet50_fpn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="maskrcnn-resnet50-fpn">
+<h1>maskrcnn_resnet50_fpn<a class="headerlink" href="#maskrcnn-resnet50-fpn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.maskrcnn_resnet50_fpn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">maskrcnn_resnet50_fpn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.mask_rcnn.MaskRCNN_ResNet50_FPN_Weights"><span class="pre">MaskRCNN_ResNet50_FPN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MaskRCNN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/mask_rcnn.html#maskrcnn_resnet50_fpn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.maskrcnn_resnet50_fpn" title="Permalink to this definition">¶</a></dt>
+<dd><p>Mask R-CNN model with a ResNet-50-FPN backbone from the <a class="reference external" href="https://arxiv.org/abs/1703.06870">Mask R-CNN</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>The input to the model is expected to be a list of tensors, each of shape <code class="docutils literal notranslate"><span class="pre">[C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>, one for each
+image, and should be in <code class="docutils literal notranslate"><span class="pre">0-1</span></code> range. Different images can have different sizes.</p>
+<p>The behavior of the model changes depending on if it is in training or evaluation mode.</p>
+<p>During training, the model expects both the input tensors and targets (list of dictionary),
+containing:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the ground-truth boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the class label for each ground-truth box</p></li>
+<li><p>masks (<code class="docutils literal notranslate"><span class="pre">UInt8Tensor[N,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>): the segmentation binary masks for each instance</p></li>
+</ul>
+</div></blockquote>
+<p>The model returns a <code class="docutils literal notranslate"><span class="pre">Dict[Tensor]</span></code> during training, containing the classification and regression
+losses for both the RPN and the R-CNN, and the mask loss.</p>
+<p>During inference, the model requires only the input tensors, and returns the post-processed
+predictions as a <code class="docutils literal notranslate"><span class="pre">List[Dict[Tensor]]</span></code>, one for each input image. The fields of the <code class="docutils literal notranslate"><span class="pre">Dict</span></code> are as
+follows, where <code class="docutils literal notranslate"><span class="pre">N</span></code> is the number of detected instances:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the predicted boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the predicted labels for each instance</p></li>
+<li><p>scores (<code class="docutils literal notranslate"><span class="pre">Tensor[N]</span></code>): the scores or each instance</p></li>
+<li><p>masks (<code class="docutils literal notranslate"><span class="pre">UInt8Tensor[N,</span> <span class="pre">1,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>): the predicted masks for each instance, in <code class="docutils literal notranslate"><span class="pre">0-1</span></code> range. In order to
+obtain the final segmentation masks, the soft masks can be thresholded, generally
+with a value of 0.5 (<code class="docutils literal notranslate"><span class="pre">mask</span> <span class="pre">&gt;=</span> <span class="pre">0.5</span></code>)</p></li>
+</ul>
+</div></blockquote>
+<p>For more details on the output and on how to plot the masks, you may refer to <a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<p>Mask R-CNN is exportable to ONNX for a fixed batch size with inputs images of fixed size.</p>
+<p>Example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">maskrcnn_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">MaskRCNN_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">400</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt;</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="c1"># optionally, if you want to export the model to ONNX:</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">torch</span><span class="o">.</span><span class="n">onnx</span><span class="o">.</span><span class="n">export</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="s2">&quot;mask_rcnn.onnx&quot;</span><span class="p">,</span> <span class="n">opset_version</span> <span class="o">=</span> <span class="mi">11</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights" title="torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from
+final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are
+trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.mask_rcnn.MaskRCNN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">MaskRCNN_ResNet50_FPN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/mask_rcnn.html#MaskRCNN_ResNet50_FPN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>MaskRCNN_ResNet50_FPN_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>37.9</p></td>
+</tr>
+<tr class="row-even"><td><p>mask_map (on COCO-val2017)</p></td>
+<td><p>34.6</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>44401393</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#mask-r-cnn">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>134.38</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>169.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">maskrcnn_resnet50_fpn</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html" class="btn btn-neutral float-right" title="maskrcnn_resnet50_fpn_v2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../mask_rcnn.html" class="btn btn-neutral" title="Mask R-CNN" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">maskrcnn_resnet50_fpn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
new file mode 100644
index 00000000000..6d72b182f1d
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
@@ -0,0 +1,763 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>maskrcnn_resnet50_fpn_v2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Keypoint R-CNN" href="../keypoint_rcnn.html" />
+    <link rel="prev" title="maskrcnn_resnet50_fpn" href="torchvision.models.detection.maskrcnn_resnet50_fpn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mask_rcnn.html">Mask R-CNN</a> &gt;</li>
+        
+      <li>maskrcnn_resnet50_fpn_v2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="maskrcnn-resnet50-fpn-v2">
+<h1>maskrcnn_resnet50_fpn_v2<a class="headerlink" href="#maskrcnn-resnet50-fpn-v2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.maskrcnn_resnet50_fpn_v2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">maskrcnn_resnet50_fpn_v2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.mask_rcnn.MaskRCNN_ResNet50_FPN_V2_Weights"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MaskRCNN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/mask_rcnn.html#maskrcnn_resnet50_fpn_v2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.maskrcnn_resnet50_fpn_v2" title="Permalink to this definition">¶</a></dt>
+<dd><p>Improved Mask R-CNN model with a ResNet-50-FPN backbone from the <a class="reference external" href="https://arxiv.org/abs/2111.11429">Benchmarking Detection Transfer
+Learning with Vision Transformers</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p><a class="reference internal" href="torchvision.models.detection.maskrcnn_resnet50_fpn.html#torchvision.models.detection.maskrcnn_resnet50_fpn" title="torchvision.models.detection.maskrcnn_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">maskrcnn_resnet50_fpn()</span></code></a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from
+final block. Valid values are between 0 and 5, with 5 meaning all backbone layers are
+trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.mask_rcnn.MaskRCNN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/mask_rcnn.html#MaskRCNN_ResNet50_FPN_V2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.MaskRCNN_ResNet50_FPN_V2_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>MaskRCNN_ResNet50_FPN_V2_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced using an enhanced training recipe to boost the model accuracy.
+Also available as <code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>47.4</p></td>
+</tr>
+<tr class="row-even"><td><p>mask_map (on COCO-val2017)</p></td>
+<td><p>41.8</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>46359409</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5773">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>333.58</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>177.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MaskRCNN_ResNet50_FPN_V2_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../keypoint_rcnn.html" class="btn btn-neutral float-right" title="Keypoint R-CNN" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.detection.maskrcnn_resnet50_fpn.html" class="btn btn-neutral" title="maskrcnn_resnet50_fpn" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">maskrcnn_resnet50_fpn_v2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
new file mode 100644
index 00000000000..60d1c1c6e58
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
@@ -0,0 +1,791 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>retinanet_resnet50_fpn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="retinanet_resnet50_fpn_v2" href="torchvision.models.detection.retinanet_resnet50_fpn_v2.html" />
+    <link rel="prev" title="RetinaNet" href="../retinanet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../retinanet.html">RetinaNet</a> &gt;</li>
+        
+      <li>retinanet_resnet50_fpn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="retinanet-resnet50-fpn">
+<h1>retinanet_resnet50_fpn<a class="headerlink" href="#retinanet-resnet50-fpn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.retinanet_resnet50_fpn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">retinanet_resnet50_fpn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights" title="torchvision.models.detection.retinanet.RetinaNet_ResNet50_FPN_Weights"><span class="pre">RetinaNet_ResNet50_FPN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RetinaNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/retinanet.html#retinanet_resnet50_fpn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.retinanet_resnet50_fpn" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RetinaNet model with a ResNet-50-FPN backbone.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1708.02002">Focal Loss for Dense Object Detection</a>.</p>
+<p>The input to the model is expected to be a list of tensors, each of shape <code class="docutils literal notranslate"><span class="pre">[C,</span> <span class="pre">H,</span> <span class="pre">W]</span></code>, one for each
+image, and should be in <code class="docutils literal notranslate"><span class="pre">0-1</span></code> range. Different images can have different sizes.</p>
+<p>The behavior of the model changes depending on if it is in training or evaluation mode.</p>
+<p>During training, the model expects both the input tensors and targets (list of dictionary),
+containing:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the ground-truth boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the class label for each ground-truth box</p></li>
+</ul>
+</div></blockquote>
+<p>The model returns a <code class="docutils literal notranslate"><span class="pre">Dict[Tensor]</span></code> during training, containing the classification and regression
+losses.</p>
+<p>During inference, the model requires only the input tensors, and returns the post-processed
+predictions as a <code class="docutils literal notranslate"><span class="pre">List[Dict[Tensor]]</span></code>, one for each input image. The fields of the <code class="docutils literal notranslate"><span class="pre">Dict</span></code> are as
+follows, where <code class="docutils literal notranslate"><span class="pre">N</span></code> is the number of detections:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the predicted boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (<code class="docutils literal notranslate"><span class="pre">Int64Tensor[N]</span></code>): the predicted labels for each detection</p></li>
+<li><p>scores (<code class="docutils literal notranslate"><span class="pre">Tensor[N]</span></code>): the scores of each detection</p></li>
+</ul>
+</div></blockquote>
+<p>For more details on the output, you may refer to <a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#instance-seg-output"><span class="std std-ref">Instance segmentation models</span></a>.</p>
+<p>Example:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">retinanet_resnet50_fpn</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">RetinaNet_ResNet50_FPN_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">400</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights" title="torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights" title="torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The pretrained weights for
+the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from final block.
+Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is
+passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.RetinaNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/retinanet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">RetinaNet_ResNet50_FPN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/retinanet.html#RetinaNet_ResNet50_FPN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>RetinaNet_ResNet50_FPN_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>36.4</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>34014999</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#retinanet">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>151.54</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>130.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.detection.retinanet_resnet50_fpn_v2.html" class="btn btn-neutral float-right" title="retinanet_resnet50_fpn_v2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../retinanet.html" class="btn btn-neutral" title="RetinaNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">retinanet_resnet50_fpn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
new file mode 100644
index 00000000000..bbaba0410c1
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
@@ -0,0 +1,759 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>retinanet_resnet50_fpn_v2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="SSD" href="../ssd.html" />
+    <link rel="prev" title="retinanet_resnet50_fpn" href="torchvision.models.detection.retinanet_resnet50_fpn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../retinanet.html">RetinaNet</a> &gt;</li>
+        
+      <li>retinanet_resnet50_fpn_v2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="retinanet-resnet50-fpn-v2">
+<h1>retinanet_resnet50_fpn_v2<a class="headerlink" href="#retinanet-resnet50-fpn-v2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.retinanet_resnet50_fpn_v2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">retinanet_resnet50_fpn_v2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.retinanet.RetinaNet_ResNet50_FPN_V2_Weights"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RetinaNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/retinanet.html#retinanet_resnet50_fpn_v2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.retinanet_resnet50_fpn_v2" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs an improved RetinaNet model with a ResNet-50-FPN backbone.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1912.02424">Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection</a>.</p>
+<p><a class="reference internal" href="torchvision.models.detection.retinanet_resnet50_fpn.html#torchvision.models.detection.retinanet_resnet50_fpn" title="torchvision.models.detection.retinanet_resnet50_fpn"><code class="xref py py-func docutils literal notranslate"><span class="pre">retinanet_resnet50_fpn()</span></code></a> for more details.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights" title="torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The pretrained weights for
+the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from final block.
+Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is
+passed (the default) this value is set to 3.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.RetinaNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/retinanet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/retinanet.html#RetinaNet_ResNet50_FPN_V2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.RetinaNet_ResNet50_FPN_V2_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>RetinaNet_ResNet50_FPN_V2_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced using an enhanced training recipe to boost the model accuracy.
+Also available as <code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>41.5</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>38198935</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5756">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>152.24</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>146.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RetinaNet_ResNet50_FPN_V2_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../ssd.html" class="btn btn-neutral float-right" title="SSD" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.detection.retinanet_resnet50_fpn.html" class="btn btn-neutral" title="retinanet_resnet50_fpn" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">retinanet_resnet50_fpn_v2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html b/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
new file mode 100644
index 00000000000..92c7ca87fe2
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
@@ -0,0 +1,791 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ssd300_vgg16 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="SSDlite" href="../ssdlite.html" />
+    <link rel="prev" title="SSD" href="../ssd.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../ssd.html">SSD</a> &gt;</li>
+        
+      <li>ssd300_vgg16</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.ssd300_vgg16.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ssd300-vgg16">
+<h1>ssd300_vgg16<a class="headerlink" href="#ssd300-vgg16" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.ssd300_vgg16">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">ssd300_vgg16</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.SSD300_VGG16_Weights" title="torchvision.models.detection.ssd.SSD300_VGG16_Weights"><span class="pre">SSD300_VGG16_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.vgg16.html#torchvision.models.VGG16_Weights" title="torchvision.models.vgg.VGG16_Weights"><span class="pre">VGG16_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">VGG16_Weights.IMAGENET1K_FEATURES</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SSD</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/ssd.html#ssd300_vgg16"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.ssd300_vgg16" title="Permalink to this definition">¶</a></dt>
+<dd><p>The SSD300 model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.02325">SSD: Single Shot MultiBox Detector</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each
+image, and should be in 0-1 range. Different images can have different sizes, but they will be resized
+to a fixed size before passing it to the backbone.</p>
+<p>The behavior of the model changes depending on if it is in training or evaluation mode.</p>
+<p>During training, the model expects both the input tensors and targets (list of dictionary),
+containing:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the ground-truth boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (Int64Tensor[N]): the class label for each ground-truth box</p></li>
+</ul>
+</div></blockquote>
+<p>The model returns a Dict[Tensor] during training, containing the classification and regression
+losses.</p>
+<p>During inference, the model requires only the input tensors, and returns the post-processed
+predictions as a List[Dict[Tensor]], one for each input image. The fields of the Dict are as
+follows, where <code class="docutils literal notranslate"><span class="pre">N</span></code> is the number of detections:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>boxes (<code class="docutils literal notranslate"><span class="pre">FloatTensor[N,</span> <span class="pre">4]</span></code>): the predicted boxes in <code class="docutils literal notranslate"><span class="pre">[x1,</span> <span class="pre">y1,</span> <span class="pre">x2,</span> <span class="pre">y2]</span></code> format, with
+<code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">x1</span> <span class="pre">&lt;</span> <span class="pre">x2</span> <span class="pre">&lt;=</span> <span class="pre">W</span></code> and <code class="docutils literal notranslate"><span class="pre">0</span> <span class="pre">&lt;=</span> <span class="pre">y1</span> <span class="pre">&lt;</span> <span class="pre">y2</span> <span class="pre">&lt;=</span> <span class="pre">H</span></code>.</p></li>
+<li><p>labels (Int64Tensor[N]): the predicted labels for each detection</p></li>
+<li><p>scores (Tensor[N]): the scores for each detection</p></li>
+</ul>
+</div></blockquote>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">ssd300_vgg16</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">SSD300_VGG16_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">300</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.SSD300_VGG16_Weights" title="torchvision.models.detection.SSD300_VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights</span></code></a>, optional) – The pretrained
+weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.SSD300_VGG16_Weights" title="torchvision.models.detection.SSD300_VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr
+Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.vgg16.html#torchvision.models.VGG16_Weights" title="torchvision.models.VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_Weights</span></code></a>, optional) – The pretrained weights for the
+backbone</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers starting from final block.
+Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is
+passed (the default) this value is set to 4.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.SSD</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.SSD300_VGG16_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">SSD300_VGG16_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/ssd.html#SSD300_VGG16_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.SSD300_VGG16_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>SSD300_VGG16_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>25.1</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>35641826</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#ssd300-vgg16">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>34.86</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>136.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">SSD300_VGG16_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../ssdlite.html" class="btn btn-neutral float-right" title="SSDlite" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../ssd.html" class="btn btn-neutral" title="SSD" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ssd300_vgg16</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
new file mode 100644
index 00000000000..825217fb5a3
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
@@ -0,0 +1,771 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ssdlite320_mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Mask R-CNN" href="../mask_rcnn.html" />
+    <link rel="prev" title="SSDlite" href="../ssdlite.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../ssdlite.html">SSDlite</a> &gt;</li>
+        
+      <li>ssdlite320_mobilenet_v3_large</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ssdlite320-mobilenet-v3-large">
+<h1>ssdlite320_mobilenet_v3_large<a class="headerlink" href="#ssdlite320-mobilenet-v3-large" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.detection.ssdlite320_mobilenet_v3_large">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">ssdlite320_mobilenet_v3_large</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights" title="torchvision.models.detection.ssdlite.SSDLite320_MobileNet_V3_Large_Weights"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Large_Weights"><span class="pre">MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trainable_backbone_layers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">norm_layer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Callable" title="(in Python v3.12)"><span class="pre">Callable</span></a><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.0)"><span class="pre">Module</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SSD</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/detection/ssdlite.html#ssdlite320_mobilenet_v3_large"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.ssdlite320_mobilenet_v3_large" title="Permalink to this definition">¶</a></dt>
+<dd><p>SSDlite model architecture with input size 320x320 and a MobileNetV3 Large backbone, as
+described at <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> and
+<a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear Bottlenecks</a>.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>See <a class="reference internal" href="torchvision.models.detection.ssd300_vgg16.html#torchvision.models.detection.ssd300_vgg16" title="torchvision.models.detection.ssd300_vgg16"><code class="xref py py-func docutils literal notranslate"><span class="pre">ssd300_vgg16()</span></code></a> for more details.</p>
+<p class="rubric">Example</p>
+<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">model</span> <span class="o">=</span> <span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">detection</span><span class="o">.</span><span class="n">ssdlite320_mobilenet_v3_large</span><span class="p">(</span><span class="n">weights</span><span class="o">=</span><span class="n">SSDLite320_MobileNet_V3_Large_Weights</span><span class="o">.</span><span class="n">DEFAULT</span><span class="p">)</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">x</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">320</span><span class="p">,</span> <span class="mi">320</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="mi">400</span><span class="p">)]</span>
+<span class="gp">&gt;&gt;&gt; </span><span class="n">predictions</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+</pre></div>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights" title="torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights" title="torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model
+(including the background).</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a>, optional) – The pretrained
+weights for the backbone.</p></li>
+<li><p><strong>trainable_backbone_layers</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of trainable (not frozen) layers
+starting from final block. Valid values are between 0 and 6, with 6 meaning all
+backbone layers are trainable. If <code class="docutils literal notranslate"><span class="pre">None</span></code> is passed (the default) this value is
+set to 6.</p></li>
+<li><p><strong>norm_layer</strong> (<em>callable</em><em>, </em><em>optional</em>) – Module specifying the normalization layer to use.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.ssd.SSD</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.detection.</span></span><span class="sig-name descname"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/detection/ssdlite.html#SSDLite320_MobileNet_V3_Large_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.detection.SSDLite320_MobileNet_V3_Large_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights.COCO_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_V1'</span></code>.</p>
+<p><strong>SSDLite320_MobileNet_V3_Large_Weights.COCO_V1</strong>:</p>
+<p>These weights were produced by following a similar training recipe as on the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>box_map (on COCO-val2017)</p></td>
+<td><p>21.3</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>3440060</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, person, bicycle, … (88 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection#ssdlite320-mobilenetv3-large">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>0.58</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>13.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">SSDLite320_MobileNet_V3_Large_Weights.COCO_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../mask_rcnn.html" class="btn btn-neutral float-right" title="Mask R-CNN" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../ssdlite.html" class="btn btn-neutral" title="SSDlite" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ssdlite320_mobilenet_v3_large</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b0.html b/0.16/models/generated/torchvision.models.efficientnet_b0.html
new file mode 100644
index 00000000000..23592fe4489
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b0.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b0 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_b1" href="torchvision.models.efficientnet_b1.html" />
+    <link rel="prev" title="EfficientNet" href="../efficientnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b0</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b0.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b0">
+<h1>efficientnet_b0<a class="headerlink" href="#efficientnet-b0" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b0">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b0</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B0_Weights" title="torchvision.models.efficientnet.EfficientNet_B0_Weights"><span class="pre">EfficientNet_B0_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b0" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B0 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B0_Weights" title="torchvision.models.EfficientNet_B0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B0_Weights" title="torchvision.models.EfficientNet_B0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B0_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B0_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B0_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B0_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B0_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.692</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.532</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>5288548</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.39</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>20.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B0_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_b1.html" class="btn btn-neutral float-right" title="efficientnet_b1" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../efficientnet.html" class="btn btn-neutral" title="EfficientNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b0</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b1.html b/0.16/models/generated/torchvision.models.efficientnet_b1.html
new file mode 100644
index 00000000000..972db29db91
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b1.html
@@ -0,0 +1,786 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b1 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_b2" href="torchvision.models.efficientnet_b2.html" />
+    <link rel="prev" title="efficientnet_b0" href="torchvision.models.efficientnet_b0.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b1</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b1.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b1">
+<h1>efficientnet_b1<a class="headerlink" href="#efficientnet-b1" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b1">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b1</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B1_Weights" title="torchvision.models.efficientnet.EfficientNet_B1_Weights"><span class="pre">EfficientNet_B1_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b1" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B1 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B1_Weights" title="torchvision.models.EfficientNet_B1_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B1_Weights" title="torchvision.models.EfficientNet_B1_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B1_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B1_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B1_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B1_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B1_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.642</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.186</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>7794184</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.69</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>30.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[240]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>EfficientNet_B1_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>79.838</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.934</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-lr-wd-crop-tuning">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>7794184</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.69</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>30.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B1_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[255]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[240]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_b2.html" class="btn btn-neutral float-right" title="efficientnet_b2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_b0.html" class="btn btn-neutral" title="efficientnet_b0" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b1</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b2.html b/0.16/models/generated/torchvision.models.efficientnet_b2.html
new file mode 100644
index 00000000000..cb35698d450
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b2.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_b3" href="torchvision.models.efficientnet_b3.html" />
+    <link rel="prev" title="efficientnet_b1" href="torchvision.models.efficientnet_b1.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b2">
+<h1>efficientnet_b2<a class="headerlink" href="#efficientnet-b2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B2_Weights" title="torchvision.models.efficientnet.EfficientNet_B2_Weights"><span class="pre">EfficientNet_B2_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b2" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B2 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B2_Weights" title="torchvision.models.EfficientNet_B2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B2_Weights" title="torchvision.models.EfficientNet_B2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B2_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B2_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B2_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.608</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.31</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>9109994</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>35.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B2_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[288]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[288]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_b3.html" class="btn btn-neutral float-right" title="efficientnet_b3" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_b1.html" class="btn btn-neutral" title="efficientnet_b1" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b3.html b/0.16/models/generated/torchvision.models.efficientnet_b3.html
new file mode 100644
index 00000000000..9ac0385fcfd
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b3.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_b4" href="torchvision.models.efficientnet_b4.html" />
+    <link rel="prev" title="efficientnet_b2" href="torchvision.models.efficientnet_b2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b3.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b3">
+<h1>efficientnet_b3<a class="headerlink" href="#efficientnet-b3" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b3">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b3</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B3_Weights" title="torchvision.models.efficientnet.EfficientNet_B3_Weights"><span class="pre">EfficientNet_B3_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b3"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b3" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B3 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B3_Weights" title="torchvision.models.EfficientNet_B3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B3_Weights" title="torchvision.models.EfficientNet_B3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B3_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B3_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B3_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B3_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B3_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.008</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.054</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>12233232</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.83</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>47.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B3_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[320]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[300]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_b4.html" class="btn btn-neutral float-right" title="efficientnet_b4" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_b2.html" class="btn btn-neutral" title="efficientnet_b2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b3</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b4.html b/0.16/models/generated/torchvision.models.efficientnet_b4.html
new file mode 100644
index 00000000000..3128de93ce4
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b4.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b4 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_b5" href="torchvision.models.efficientnet_b5.html" />
+    <link rel="prev" title="efficientnet_b3" href="torchvision.models.efficientnet_b3.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b4</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b4.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b4">
+<h1>efficientnet_b4<a class="headerlink" href="#efficientnet-b4" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b4">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b4</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B4_Weights" title="torchvision.models.efficientnet.EfficientNet_B4_Weights"><span class="pre">EfficientNet_B4_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b4"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b4" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B4 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B4_Weights" title="torchvision.models.EfficientNet_B4_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B4_Weights" title="torchvision.models.EfficientNet_B4_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B4_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B4_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B4_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B4_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B4_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.384</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.594</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>19341616</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.39</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>74.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B4_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[384]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[380]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_b5.html" class="btn btn-neutral float-right" title="efficientnet_b5" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_b3.html" class="btn btn-neutral" title="efficientnet_b3" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b4</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b5.html b/0.16/models/generated/torchvision.models.efficientnet_b5.html
new file mode 100644
index 00000000000..8107162669e
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b5.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b5 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_b6" href="torchvision.models.efficientnet_b6.html" />
+    <link rel="prev" title="efficientnet_b4" href="torchvision.models.efficientnet_b4.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b5</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b5.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b5">
+<h1>efficientnet_b5<a class="headerlink" href="#efficientnet-b5" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b5">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b5</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B5_Weights" title="torchvision.models.efficientnet.EfficientNet_B5_Weights"><span class="pre">EfficientNet_B5_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b5"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b5" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B5 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B5_Weights" title="torchvision.models.EfficientNet_B5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B5_Weights" title="torchvision.models.EfficientNet_B5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B5_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B5_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B5_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B5_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B5_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.444</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.628</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>30389784</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>10.27</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>116.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B5_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[456]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[456]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_b6.html" class="btn btn-neutral float-right" title="efficientnet_b6" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_b4.html" class="btn btn-neutral" title="efficientnet_b4" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b5</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b6.html b/0.16/models/generated/torchvision.models.efficientnet_b6.html
new file mode 100644
index 00000000000..5d7d2cbd7c0
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b6.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b6 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_b7" href="torchvision.models.efficientnet_b7.html" />
+    <link rel="prev" title="efficientnet_b5" href="torchvision.models.efficientnet_b5.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b6</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b6.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b6">
+<h1>efficientnet_b6<a class="headerlink" href="#efficientnet-b6" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b6">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b6</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B6_Weights" title="torchvision.models.efficientnet.EfficientNet_B6_Weights"><span class="pre">EfficientNet_B6_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b6"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b6" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B6 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B6_Weights" title="torchvision.models.EfficientNet_B6_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B6_Weights" title="torchvision.models.EfficientNet_B6_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B6_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B6_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B6_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B6_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B6_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>84.008</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.916</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>43040704</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>19.07</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>165.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B6_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[528]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[528]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_b7.html" class="btn btn-neutral float-right" title="efficientnet_b7" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_b5.html" class="btn btn-neutral" title="efficientnet_b5" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b6</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b7.html b/0.16/models/generated/torchvision.models.efficientnet_b7.html
new file mode 100644
index 00000000000..ac224dd95fb
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_b7.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_b7 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="EfficientNetV2" href="../efficientnetv2.html" />
+    <link rel="prev" title="efficientnet_b6" href="torchvision.models.efficientnet_b6.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnet.html">EfficientNet</a> &gt;</li>
+        
+      <li>efficientnet_b7</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_b7.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-b7">
+<h1>efficientnet_b7<a class="headerlink" href="#efficientnet-b7" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_b7">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_b7</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_B7_Weights" title="torchvision.models.efficientnet.EfficientNet_B7_Weights"><span class="pre">EfficientNet_B7_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_b7"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_b7" title="Permalink to this definition">¶</a></dt>
+<dd><p>EfficientNet B7 model architecture from the <a class="reference external" href="https://arxiv.org/abs/1905.11946">EfficientNet: Rethinking Model Scaling for Convolutional
+Neural Networks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_B7_Weights" title="torchvision.models.EfficientNet_B7_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_B7_Weights" title="torchvision.models.EfficientNet_B7_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_B7_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_B7_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_B7_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_B7_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_B7_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>84.122</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.908</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>66347960</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>37.75</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>254.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_B7_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[600]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[600]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../efficientnetv2.html" class="btn btn-neutral float-right" title="EfficientNetV2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_b6.html" class="btn btn-neutral" title="efficientnet_b6" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_b7</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_l.html b/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
new file mode 100644
index 00000000000..57376ca9987
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_v2_l &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="GoogLeNet" href="../googlenet.html" />
+    <link rel="prev" title="efficientnet_v2_m" href="torchvision.models.efficientnet_v2_m.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnetv2.html">EfficientNetV2</a> &gt;</li>
+        
+      <li>efficientnet_v2_l</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_v2_l.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-v2-l">
+<h1>efficientnet_v2_l<a class="headerlink" href="#efficientnet-v2-l" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_v2_l">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_v2_l</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_V2_L_Weights" title="torchvision.models.efficientnet.EfficientNet_V2_L_Weights"><span class="pre">EfficientNet_V2_L_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_v2_l"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_v2_l" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs an EfficientNetV2-L architecture from
+<a class="reference external" href="https://arxiv.org/abs/2104.00298">EfficientNetV2: Smaller Models and Faster Training</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_V2_L_Weights" title="torchvision.models.EfficientNet_V2_L_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_V2_L_Weights" title="torchvision.models.EfficientNet_V2_L_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_V2_L_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_V2_L_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_V2_L_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_V2_L_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_V2_L_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>85.808</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.788</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=33, width=33</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>118515272</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>56.08</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>454.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_L_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[480]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[480]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.5,</span> <span class="pre">0.5,</span> <span class="pre">0.5]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.5,</span> <span class="pre">0.5,</span> <span class="pre">0.5]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../googlenet.html" class="btn btn-neutral float-right" title="GoogLeNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_v2_m.html" class="btn btn-neutral" title="efficientnet_v2_m" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_v2_l</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_m.html b/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
new file mode 100644
index 00000000000..4483f561da9
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_v2_m &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_v2_l" href="torchvision.models.efficientnet_v2_l.html" />
+    <link rel="prev" title="efficientnet_v2_s" href="torchvision.models.efficientnet_v2_s.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnetv2.html">EfficientNetV2</a> &gt;</li>
+        
+      <li>efficientnet_v2_m</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_v2_m.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-v2-m">
+<h1>efficientnet_v2_m<a class="headerlink" href="#efficientnet-v2-m" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_v2_m">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_v2_m</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_V2_M_Weights" title="torchvision.models.efficientnet.EfficientNet_V2_M_Weights"><span class="pre">EfficientNet_V2_M_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_v2_m"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_v2_m" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs an EfficientNetV2-M architecture from
+<a class="reference external" href="https://arxiv.org/abs/2104.00298">EfficientNetV2: Smaller Models and Faster Training</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_V2_M_Weights" title="torchvision.models.EfficientNet_V2_M_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_V2_M_Weights" title="torchvision.models.EfficientNet_V2_M_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_V2_M_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_V2_M_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_V2_M_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_V2_M_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_V2_M_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>85.112</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.156</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=33, width=33</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>54139356</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>24.58</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>208.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_M_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[480]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[480]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_v2_l.html" class="btn btn-neutral float-right" title="efficientnet_v2_l" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.efficientnet_v2_s.html" class="btn btn-neutral" title="efficientnet_v2_s" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_v2_m</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_s.html b/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
new file mode 100644
index 00000000000..07cc266ad41
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>efficientnet_v2_s &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="efficientnet_v2_m" href="torchvision.models.efficientnet_v2_m.html" />
+    <link rel="prev" title="EfficientNetV2" href="../efficientnetv2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../efficientnetv2.html">EfficientNetV2</a> &gt;</li>
+        
+      <li>efficientnet_v2_s</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.efficientnet_v2_s.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="efficientnet-v2-s">
+<h1>efficientnet_v2_s<a class="headerlink" href="#efficientnet-v2-s" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.efficientnet_v2_s">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">efficientnet_v2_s</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.EfficientNet_V2_S_Weights" title="torchvision.models.efficientnet.EfficientNet_V2_S_Weights"><span class="pre">EfficientNet_V2_S_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">EfficientNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#efficientnet_v2_s"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.efficientnet_v2_s" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs an EfficientNetV2-S architecture from
+<a class="reference external" href="https://arxiv.org/abs/2104.00298">EfficientNetV2: Smaller Models and Faster Training</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.EfficientNet_V2_S_Weights" title="torchvision.models.EfficientNet_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.EfficientNet_V2_S_Weights" title="torchvision.models.EfficientNet_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.efficientnet.EfficientNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/efficientnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.EfficientNet_V2_S_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">EfficientNet_V2_S_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/efficientnet.html#EfficientNet_V2_S_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.EfficientNet_V2_S_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>EfficientNet_V2_S_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>84.228</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.878</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=33, width=33</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>21458488</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>8.37</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>82.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">EfficientNet_V2_S_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[384]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[384]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.efficientnet_v2_m.html" class="btn btn-neutral float-right" title="efficientnet_v2_m" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../efficientnetv2.html" class="btn btn-neutral" title="EfficientNetV2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">efficientnet_v2_s</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.googlenet.html b/0.16/models/generated/torchvision.models.googlenet.html
new file mode 100644
index 00000000000..2c715dcf2c7
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.googlenet.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>googlenet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Inception V3" href="../inception.html" />
+    <link rel="prev" title="GoogLeNet" href="../googlenet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../googlenet.html">GoogLeNet</a> &gt;</li>
+        
+      <li>googlenet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.googlenet.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="googlenet">
+<h1>googlenet<a class="headerlink" href="#googlenet" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.googlenet">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">googlenet</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.GoogLeNet_Weights" title="torchvision.models.googlenet.GoogLeNet_Weights"><span class="pre">GoogLeNet_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">GoogLeNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/googlenet.html#googlenet"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.googlenet" title="Permalink to this definition">¶</a></dt>
+<dd><p>GoogLeNet (Inception v1) model architecture from
+<a class="reference external" href="http://arxiv.org/abs/1409.4842">Going Deeper with Convolutions</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.GoogLeNet_Weights" title="torchvision.models.GoogLeNet_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.GoogLeNet_Weights" title="torchvision.models.GoogLeNet_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.GoogLeNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/googlenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.GoogLeNet_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">GoogLeNet_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/googlenet.html#GoogLeNet_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.GoogLeNet_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>GoogLeNet_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.778</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>89.53</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>6624904</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=15, width=15</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#googlenet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.50</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>49.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../inception.html" class="btn btn-neutral float-right" title="Inception V3" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../googlenet.html" class="btn btn-neutral" title="GoogLeNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">googlenet</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.inception_v3.html b/0.16/models/generated/torchvision.models.inception_v3.html
new file mode 100644
index 00000000000..bb8ff078c27
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.inception_v3.html
@@ -0,0 +1,757 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>inception_v3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="MaxVit" href="../maxvit.html" />
+    <link rel="prev" title="Inception V3" href="../inception.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../inception.html">Inception V3</a> &gt;</li>
+        
+      <li>inception_v3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.inception_v3.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="inception-v3">
+<h1>inception_v3<a class="headerlink" href="#inception-v3" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.inception_v3">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">inception_v3</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Inception_V3_Weights" title="torchvision.models.inception.Inception_V3_Weights"><span class="pre">Inception_V3_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Inception3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/inception.html#inception_v3"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.inception_v3" title="Permalink to this definition">¶</a></dt>
+<dd><p>Inception v3 model architecture from
+<a class="reference external" href="http://arxiv.org/abs/1512.00567">Rethinking the Inception Architecture for Computer Vision</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p><strong>Important</strong>: In contrast to the other models the inception_v3 expects tensors with a size of
+N x 3 x 299 x 299, so ensure your images are sized accordingly.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Inception_V3_Weights" title="torchvision.models.Inception_V3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.Inception_V3_Weights" title="torchvision.models.Inception_V3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.Inception3</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/inception.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Inception_V3_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Inception_V3_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/inception.html#Inception_V3_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Inception_V3_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Inception_V3_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.294</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.45</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>27161264</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=75, width=75</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#inception-v3">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>5.71</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>103.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[342]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[299]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../maxvit.html" class="btn btn-neutral float-right" title="MaxVit" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../inception.html" class="btn btn-neutral" title="Inception V3" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">inception_v3</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.maxvit_t.html b/0.16/models/generated/torchvision.models.maxvit_t.html
new file mode 100644
index 00000000000..2a4c2b7f870
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.maxvit_t.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>maxvit_t &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="MNASNet" href="../mnasnet.html" />
+    <link rel="prev" title="MaxVit" href="../maxvit.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../maxvit.html">MaxVit</a> &gt;</li>
+        
+      <li>maxvit_t</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.maxvit_t.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="maxvit-t">
+<h1>maxvit_t<a class="headerlink" href="#maxvit-t" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.maxvit_t">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">maxvit_t</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MaxVit_T_Weights" title="torchvision.models.maxvit.MaxVit_T_Weights"><span class="pre">MaxVit_T_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MaxVit</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/maxvit.html#maxvit_t"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.maxvit_t" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a maxvit_t architecture from
+<a class="reference external" href="https://arxiv.org/abs/2204.01697">MaxViT: Multi-Axis Vision Transformer</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MaxVit_T_Weights" title="torchvision.models.MaxVit_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaxVit_T_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MaxVit_T_Weights" title="torchvision.models.MaxVit_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaxVit_T_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.maxvit.MaxVit</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/maxvit.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MaxVit_T_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MaxVit_T_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/maxvit.html#MaxVit_T_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MaxVit_T_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MaxVit_T_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MaxVit_T_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MaxVit_T_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a similar training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">MaxVit_T_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.7</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.722</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>30919624</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#maxvit">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>5.56</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>118.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MaxVit_T_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[224]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../mnasnet.html" class="btn btn-neutral float-right" title="MNASNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../maxvit.html" class="btn btn-neutral" title="MaxVit" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">maxvit_t</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.mnasnet0_5.html b/0.16/models/generated/torchvision.models.mnasnet0_5.html
new file mode 100644
index 00000000000..a4a411891b1
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.mnasnet0_5.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mnasnet0_5 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="mnasnet0_75" href="torchvision.models.mnasnet0_75.html" />
+    <link rel="prev" title="MNASNet" href="../mnasnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mnasnet.html">MNASNet</a> &gt;</li>
+        
+      <li>mnasnet0_5</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.mnasnet0_5.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mnasnet0-5">
+<h1>mnasnet0_5<a class="headerlink" href="#mnasnet0-5" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.mnasnet0_5">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">mnasnet0_5</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MNASNet0_5_Weights" title="torchvision.models.mnasnet.MNASNet0_5_Weights"><span class="pre">MNASNet0_5_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MNASNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#mnasnet0_5"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.mnasnet0_5" title="Permalink to this definition">¶</a></dt>
+<dd><p>MNASNet with depth multiplier of 0.5 from
+<a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MNASNet0_5_Weights" title="torchvision.models.MNASNet0_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_5_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MNASNet0_5_Weights" title="torchvision.models.MNASNet0_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_5_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.mnasnet.MNASNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MNASNet0_5_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MNASNet0_5_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#MNASNet0_5_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MNASNet0_5_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MNASNet0_5_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MNASNet0_5_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MNASNet0_5_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">MNASNet0_5_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>67.734</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>87.49</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/1e100/mnasnet_trainer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>2218512</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.10</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>8.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MNASNet0_5_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.mnasnet0_75.html" class="btn btn-neutral float-right" title="mnasnet0_75" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../mnasnet.html" class="btn btn-neutral" title="MNASNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mnasnet0_5</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.mnasnet0_75.html b/0.16/models/generated/torchvision.models.mnasnet0_75.html
new file mode 100644
index 00000000000..fc1a3e7ee27
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.mnasnet0_75.html
@@ -0,0 +1,754 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mnasnet0_75 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="mnasnet1_0" href="torchvision.models.mnasnet1_0.html" />
+    <link rel="prev" title="mnasnet0_5" href="torchvision.models.mnasnet0_5.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mnasnet.html">MNASNet</a> &gt;</li>
+        
+      <li>mnasnet0_75</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.mnasnet0_75.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mnasnet0-75">
+<h1>mnasnet0_75<a class="headerlink" href="#mnasnet0-75" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.mnasnet0_75">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">mnasnet0_75</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MNASNet0_75_Weights" title="torchvision.models.mnasnet.MNASNet0_75_Weights"><span class="pre">MNASNet0_75_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MNASNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#mnasnet0_75"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.mnasnet0_75" title="Permalink to this definition">¶</a></dt>
+<dd><p>MNASNet with depth multiplier of 0.75 from
+<a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MNASNet0_75_Weights" title="torchvision.models.MNASNet0_75_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_75_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MNASNet0_75_Weights" title="torchvision.models.MNASNet0_75_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet0_75_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.mnasnet.MNASNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MNASNet0_75_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MNASNet0_75_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#MNASNet0_75_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MNASNet0_75_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MNASNet0_75_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MNASNet0_75_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MNASNet0_75_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MNASNet0_75_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>71.18</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.496</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/6019">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>3170208</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.21</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>12.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MNASNet0_75_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.mnasnet1_0.html" class="btn btn-neutral float-right" title="mnasnet1_0" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.mnasnet0_5.html" class="btn btn-neutral" title="mnasnet0_5" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mnasnet0_75</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.mnasnet1_0.html b/0.16/models/generated/torchvision.models.mnasnet1_0.html
new file mode 100644
index 00000000000..c5807c9abd4
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.mnasnet1_0.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mnasnet1_0 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="mnasnet1_3" href="torchvision.models.mnasnet1_3.html" />
+    <link rel="prev" title="mnasnet0_75" href="torchvision.models.mnasnet0_75.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mnasnet.html">MNASNet</a> &gt;</li>
+        
+      <li>mnasnet1_0</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.mnasnet1_0.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mnasnet1-0">
+<h1>mnasnet1_0<a class="headerlink" href="#mnasnet1-0" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.mnasnet1_0">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">mnasnet1_0</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MNASNet1_0_Weights" title="torchvision.models.mnasnet.MNASNet1_0_Weights"><span class="pre">MNASNet1_0_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MNASNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#mnasnet1_0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.mnasnet1_0" title="Permalink to this definition">¶</a></dt>
+<dd><p>MNASNet with depth multiplier of 1.0 from
+<a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MNASNet1_0_Weights" title="torchvision.models.MNASNet1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_0_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MNASNet1_0_Weights" title="torchvision.models.MNASNet1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_0_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.mnasnet.MNASNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MNASNet1_0_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MNASNet1_0_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#MNASNet1_0_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MNASNet1_0_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MNASNet1_0_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MNASNet1_0_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MNASNet1_0_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">MNASNet1_0_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>73.456</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.51</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/1e100/mnasnet_trainer">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>4383312</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.31</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>16.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MNASNet1_0_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.mnasnet1_3.html" class="btn btn-neutral float-right" title="mnasnet1_3" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.mnasnet0_75.html" class="btn btn-neutral" title="mnasnet0_75" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mnasnet1_0</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.mnasnet1_3.html b/0.16/models/generated/torchvision.models.mnasnet1_3.html
new file mode 100644
index 00000000000..ced32261751
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.mnasnet1_3.html
@@ -0,0 +1,754 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mnasnet1_3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="MobileNet V2" href="../mobilenetv2.html" />
+    <link rel="prev" title="mnasnet1_0" href="torchvision.models.mnasnet1_0.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mnasnet.html">MNASNet</a> &gt;</li>
+        
+      <li>mnasnet1_3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.mnasnet1_3.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mnasnet1-3">
+<h1>mnasnet1_3<a class="headerlink" href="#mnasnet1-3" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.mnasnet1_3">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">mnasnet1_3</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MNASNet1_3_Weights" title="torchvision.models.mnasnet.MNASNet1_3_Weights"><span class="pre">MNASNet1_3_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MNASNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#mnasnet1_3"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.mnasnet1_3" title="Permalink to this definition">¶</a></dt>
+<dd><p>MNASNet with depth multiplier of 1.3 from
+<a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MNASNet1_3_Weights" title="torchvision.models.MNASNet1_3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_3_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MNASNet1_3_Weights" title="torchvision.models.MNASNet1_3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MNASNet1_3_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.mnasnet.MNASNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MNASNet1_3_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MNASNet1_3_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mnasnet.html#MNASNet1_3_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MNASNet1_3_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MNASNet1_3_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MNASNet1_3_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MNASNet1_3_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MNASNet1_3_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.506</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.522</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/6019">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>6282256</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.53</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>24.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MNASNet1_3_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../mobilenetv2.html" class="btn btn-neutral float-right" title="MobileNet V2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.mnasnet1_0.html" class="btn btn-neutral" title="mnasnet1_0" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mnasnet1_3</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v2.html b/0.16/models/generated/torchvision.models.mobilenet_v2.html
new file mode 100644
index 00000000000..f187953bdc8
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.mobilenet_v2.html
@@ -0,0 +1,786 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mobilenet_v2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="MobileNet V3" href="../mobilenetv3.html" />
+    <link rel="prev" title="MobileNet V2" href="../mobilenetv2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mobilenetv2.html">MobileNet V2</a> &gt;</li>
+        
+      <li>mobilenet_v2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.mobilenet_v2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mobilenet-v2">
+<h1>mobilenet_v2<a class="headerlink" href="#mobilenet-v2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.mobilenet_v2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">mobilenet_v2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MobileNet_V2_Weights" title="torchvision.models.mobilenetv2.MobileNet_V2_Weights"><span class="pre">MobileNet_V2_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MobileNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv2.html#mobilenet_v2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.mobilenet_v2" title="Permalink to this definition">¶</a></dt>
+<dd><p>MobileNetV2 architecture from the <a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear
+Bottlenecks</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MobileNet_V2_Weights" title="torchvision.models.MobileNet_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MobileNet_V2_Weights" title="torchvision.models.MobileNet_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.mobilenetv2.MobileNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MobileNet_V2_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MobileNet_V2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv2.html#MobileNet_V2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MobileNet_V2_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MobileNet_V2_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>71.878</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.286</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>3504872</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>13.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>MobileNet_V2_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>72.154</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.822</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>3504872</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>13.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../mobilenetv3.html" class="btn btn-neutral float-right" title="MobileNet V3" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../mobilenetv2.html" class="btn btn-neutral" title="MobileNet V2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mobilenet_v2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
new file mode 100644
index 00000000000..b0fdaa37a7f
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
@@ -0,0 +1,786 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="mobilenet_v3_small" href="torchvision.models.mobilenet_v3_small.html" />
+    <link rel="prev" title="MobileNet V3" href="../mobilenetv3.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mobilenetv3.html">MobileNet V3</a> &gt;</li>
+        
+      <li>mobilenet_v3_large</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.mobilenet_v3_large.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mobilenet-v3-large">
+<h1>mobilenet_v3_large<a class="headerlink" href="#mobilenet-v3-large" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.mobilenet_v3_large">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">mobilenet_v3_large</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Large_Weights"><span class="pre">MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MobileNetV3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv3.html#mobilenet_v3_large"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.mobilenet_v3_large" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a large MobileNetV3 architecture from
+<a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.mobilenet.MobileNetV3</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MobileNet_V3_Large_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MobileNet_V3_Large_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv3.html#MobileNet_V3_Large_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MobileNet_V3_Large_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MobileNet_V3_Large_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>74.042</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.34</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5483032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.22</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>21.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>MobileNet_V3_Large_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve marginally upon the results of the original paper by using a modified version of
+                TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.274</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.566</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5483032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.22</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>21.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.mobilenet_v3_small.html" class="btn btn-neutral float-right" title="mobilenet_v3_small" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../mobilenetv3.html" class="btn btn-neutral" title="MobileNet V3" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mobilenet_v3_large</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v3_small.html b/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
new file mode 100644
index 00000000000..99e8584f00d
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
@@ -0,0 +1,754 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mobilenet_v3_small &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="RegNet" href="../regnet.html" />
+    <link rel="prev" title="mobilenet_v3_large" href="torchvision.models.mobilenet_v3_large.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mobilenetv3.html">MobileNet V3</a> &gt;</li>
+        
+      <li>mobilenet_v3_small</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.mobilenet_v3_small.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mobilenet-v3-small">
+<h1>mobilenet_v3_small<a class="headerlink" href="#mobilenet-v3-small" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.mobilenet_v3_small">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">mobilenet_v3_small</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.MobileNet_V3_Small_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Small_Weights"><span class="pre">MobileNet_V3_Small_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MobileNetV3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv3.html#mobilenet_v3_small"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.mobilenet_v3_small" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a small MobileNetV3 architecture from
+<a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.MobileNet_V3_Small_Weights" title="torchvision.models.MobileNet_V3_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.MobileNet_V3_Small_Weights" title="torchvision.models.MobileNet_V3_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.mobilenet.MobileNetV3</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.MobileNet_V3_Small_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MobileNet_V3_Small_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv3.html#MobileNet_V3_Small_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.MobileNet_V3_Small_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MobileNet_V3_Small_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a simple training recipe.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>67.668</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>87.402</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>2542856</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.06</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>9.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Small_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../regnet.html" class="btn btn-neutral float-right" title="RegNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.mobilenet_v3_large.html" class="btn btn-neutral" title="mobilenet_v3_large" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mobilenet_v3_small</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.optical_flow.raft_large.html b/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
new file mode 100644
index 00000000000..e3d39568419
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
@@ -0,0 +1,943 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>raft_large &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="raft_small" href="torchvision.models.optical_flow.raft_small.html" />
+    <link rel="prev" title="RAFT" href="../raft.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../raft.html">RAFT</a> &gt;</li>
+        
+      <li>raft_large</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.optical_flow.raft_large.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="raft-large">
+<h1>raft_large<a class="headerlink" href="#raft-large" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.optical_flow.raft_large">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.optical_flow.</span></span><span class="sig-name descname"><span class="pre">raft_large</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.optical_flow.Raft_Large_Weights" title="torchvision.models.optical_flow.raft.Raft_Large_Weights"><span class="pre">Raft_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RAFT</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/optical_flow/raft.html#raft_large"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.optical_flow.raft_large" title="Permalink to this definition">¶</a></dt>
+<dd><p>RAFT model from
+<a class="reference external" href="https://arxiv.org/abs/2003.12039">RAFT: Recurrent All Pairs Field Transforms for Optical Flow</a>.</p>
+<p>Please see the example below for a tutorial on how to use this model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.optical_flow.Raft_Large_Weights" title="torchvision.models.optical_flow.Raft_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Raft_Large_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.optical_flow.Raft_Large_Weights" title="torchvision.models.optical_flow.Raft_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Raft_Large_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.optical_flow.RAFT</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/optical_flow/raft.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.optical_flow.Raft_Large_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.optical_flow.</span></span><span class="sig-name descname"><span class="pre">Raft_Large_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/optical_flow/raft.html#Raft_Large_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.optical_flow.Raft_Large_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.C_T_SKHT_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='C_T_V1'</span></code>.</p>
+<p>The metrics reported here are as follows.
+
+    <code class="docutils literal notranslate"><span class="pre">epe</span></code> is the “end-point-error” and indicates how far (in pixels) the
+    predicted flow is from its true value. This is averaged over all pixels
+    of all images. <code class="docutils literal notranslate"><span class="pre">per_image_epe</span></code> is similar, but the average is different:
+    the epe is first computed on each image independently, and then averaged
+    over all images. This corresponds to “Fl-epe” (sometimes written “F1-epe”)
+    in the original paper, and it’s only used on Kitti. <code class="docutils literal notranslate"><span class="pre">fl-all</span></code> is also a
+    Kitti-specific metric, defined by the author of the dataset and used for the
+    Kitti leaderboard. It corresponds to the average of pixels whose epe is
+    either &lt;3px, or &lt;5% of flow’s 2-norm.</p>
+<p><strong>Raft_Large_Weights.C_T_V1</strong>:</p>
+<p>These weights were ported from the original paper. They
+            are trained on <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+            <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>epe (on Sintel-Train-Cleanpass)</p></td>
+<td><p>1.4411</p></td>
+</tr>
+<tr class="row-even"><td><p>epe (on Sintel-Train-Finalpass)</p></td>
+<td><p>2.7894</p></td>
+</tr>
+<tr class="row-odd"><td><p>per_image_epe (on Kitti-Train)</p></td>
+<td><p>5.0172</p></td>
+</tr>
+<tr class="row-even"><td><p>fl_all (on Kitti-Train)</p></td>
+<td><p>17.4506</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>5257536</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/princeton-vl/RAFT">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>211.01</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>20.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.C_T_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+<p><strong>Raft_Large_Weights.C_T_V2</strong>:</p>
+<p>These weights were trained from scratch on
+            <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+            <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>epe (on Sintel-Train-Cleanpass)</p></td>
+<td><p>1.3822</p></td>
+</tr>
+<tr class="row-even"><td><p>epe (on Sintel-Train-Finalpass)</p></td>
+<td><p>2.7161</p></td>
+</tr>
+<tr class="row-odd"><td><p>per_image_epe (on Kitti-Train)</p></td>
+<td><p>4.5118</p></td>
+</tr>
+<tr class="row-even"><td><p>fl_all (on Kitti-Train)</p></td>
+<td><p>16.0679</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>5257536</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/optical_flow">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>211.01</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>20.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.C_T_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+<p><strong>Raft_Large_Weights.C_T_SKHT_V1</strong>:</p>
+<p>
+                These weights were ported from the original paper. They are
+                trained on <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+                <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a> and fine-tuned on
+                Sintel. The Sintel fine-tuning step is a combination of
+                <a class="reference internal" href="../../generated/torchvision.datasets.Sintel.html#torchvision.datasets.Sintel" title="torchvision.datasets.Sintel"><code class="xref py py-class docutils literal notranslate"><span class="pre">Sintel</span></code></a>,
+                <a class="reference internal" href="../../generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-class docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>,
+                <a class="reference internal" href="../../generated/torchvision.datasets.HD1K.html#torchvision.datasets.HD1K" title="torchvision.datasets.HD1K"><code class="xref py py-class docutils literal notranslate"><span class="pre">HD1K</span></code></a>, and
+                <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a> (clean pass).</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>epe (on Sintel-Test-Cleanpass)</p></td>
+<td><p>1.94</p></td>
+</tr>
+<tr class="row-even"><td><p>epe (on Sintel-Test-Finalpass)</p></td>
+<td><p>3.18</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>5257536</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/princeton-vl/RAFT">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>211.01</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>20.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.C_T_SKHT_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+<p><strong>Raft_Large_Weights.C_T_SKHT_V2</strong>:</p>
+<p>
+                These weights were trained from scratch. They are
+                pre-trained on <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+                <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a> and then
+                fine-tuned on Sintel. The Sintel fine-tuning step is a
+                combination of <a class="reference internal" href="../../generated/torchvision.datasets.Sintel.html#torchvision.datasets.Sintel" title="torchvision.datasets.Sintel"><code class="xref py py-class docutils literal notranslate"><span class="pre">Sintel</span></code></a>,
+                <a class="reference internal" href="../../generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-class docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>,
+                <a class="reference internal" href="../../generated/torchvision.datasets.HD1K.html#torchvision.datasets.HD1K" title="torchvision.datasets.HD1K"><code class="xref py py-class docutils literal notranslate"><span class="pre">HD1K</span></code></a>, and
+                <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a> (clean pass).
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>epe (on Sintel-Test-Cleanpass)</p></td>
+<td><p>1.819</p></td>
+</tr>
+<tr class="row-even"><td><p>epe (on Sintel-Test-Finalpass)</p></td>
+<td><p>3.067</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>5257536</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/optical_flow">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>211.01</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>20.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.C_T_SKHT_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+<p><strong>Raft_Large_Weights.C_T_SKHT_K_V1</strong>:</p>
+<p>
+                These weights were ported from the original paper. They are
+                pre-trained on <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+                <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a>,
+                fine-tuned on Sintel, and then fine-tuned on
+                <a class="reference internal" href="../../generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-class docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>. The Sintel fine-tuning
+                step was described above.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>fl_all (on Kitti-Test)</p></td>
+<td><p>5.1</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5257536</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/princeton-vl/RAFT">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>211.01</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>20.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.C_T_SKHT_K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+<p><strong>Raft_Large_Weights.C_T_SKHT_K_V2</strong>:</p>
+<p>
+                These weights were trained from scratch. They are
+                pre-trained on <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+                <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a>,
+                fine-tuned on Sintel, and then fine-tuned on
+                <a class="reference internal" href="../../generated/torchvision.datasets.KittiFlow.html#torchvision.datasets.KittiFlow" title="torchvision.datasets.KittiFlow"><code class="xref py py-class docutils literal notranslate"><span class="pre">KittiFlow</span></code></a>. The Sintel fine-tuning
+                step was described above.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>fl_all (on Kitti-Test)</p></td>
+<td><p>5.19</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5257536</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/optical_flow">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>211.01</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>20.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Large_Weights.C_T_SKHT_K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">raft_large</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.optical_flow.raft_small.html" class="btn btn-neutral float-right" title="raft_small" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../raft.html" class="btn btn-neutral" title="RAFT" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">raft_large</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.optical_flow.raft_small.html b/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
new file mode 100644
index 00000000000..6ffd1339bf3
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
@@ -0,0 +1,808 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>raft_small &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Datasets" href="../../datasets.html" />
+    <link rel="prev" title="raft_large" href="torchvision.models.optical_flow.raft_large.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../raft.html">RAFT</a> &gt;</li>
+        
+      <li>raft_small</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.optical_flow.raft_small.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="raft-small">
+<h1>raft_small<a class="headerlink" href="#raft-small" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.optical_flow.raft_small">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.optical_flow.</span></span><span class="sig-name descname"><span class="pre">raft_small</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.optical_flow.Raft_Small_Weights" title="torchvision.models.optical_flow.raft.Raft_Small_Weights"><span class="pre">Raft_Small_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RAFT</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/optical_flow/raft.html#raft_small"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.optical_flow.raft_small" title="Permalink to this definition">¶</a></dt>
+<dd><p>RAFT “small” model from
+<a class="reference external" href="https://arxiv.org/abs/2003.12039">RAFT: Recurrent All Pairs Field Transforms for Optical Flow</a>.</p>
+<p>Please see the example below for a tutorial on how to use this model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.optical_flow.Raft_Small_Weights" title="torchvision.models.optical_flow.Raft_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Raft_Small_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.optical_flow.Raft_Small_Weights" title="torchvision.models.optical_flow.Raft_Small_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Raft_Small_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.optical_flow.RAFT</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/optical_flow/raft.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.optical_flow.Raft_Small_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.optical_flow.</span></span><span class="sig-name descname"><span class="pre">Raft_Small_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/optical_flow/raft.html#Raft_Small_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.optical_flow.Raft_Small_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Raft_Small_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Raft_Small_Weights.C_T_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='C_T_V1'</span></code>.</p>
+<p>The metrics reported here are as follows.
+
+    <code class="docutils literal notranslate"><span class="pre">epe</span></code> is the “end-point-error” and indicates how far (in pixels) the
+    predicted flow is from its true value. This is averaged over all pixels
+    of all images. <code class="docutils literal notranslate"><span class="pre">per_image_epe</span></code> is similar, but the average is different:
+    the epe is first computed on each image independently, and then averaged
+    over all images. This corresponds to “Fl-epe” (sometimes written “F1-epe”)
+    in the original paper, and it’s only used on Kitti. <code class="docutils literal notranslate"><span class="pre">fl-all</span></code> is also a
+    Kitti-specific metric, defined by the author of the dataset and used for the
+    Kitti leaderboard. It corresponds to the average of pixels whose epe is
+    either &lt;3px, or &lt;5% of flow’s 2-norm.</p>
+<p><strong>Raft_Small_Weights.C_T_V1</strong>:</p>
+<p>These weights were ported from the original paper. They
+            are trained on <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+            <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>epe (on Sintel-Train-Cleanpass)</p></td>
+<td><p>2.1231</p></td>
+</tr>
+<tr class="row-even"><td><p>epe (on Sintel-Train-Finalpass)</p></td>
+<td><p>3.279</p></td>
+</tr>
+<tr class="row-odd"><td><p>per_image_epe (on Kitti-Train)</p></td>
+<td><p>7.6557</p></td>
+</tr>
+<tr class="row-even"><td><p>fl_all (on Kitti-Train)</p></td>
+<td><p>25.2801</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>990162</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/princeton-vl/RAFT">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>47.66</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>3.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Small_Weights.C_T_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+<p><strong>Raft_Small_Weights.C_T_V2</strong>:</p>
+<p>These weights were trained from scratch on
+            <a class="reference internal" href="../../generated/torchvision.datasets.FlyingChairs.html#torchvision.datasets.FlyingChairs" title="torchvision.datasets.FlyingChairs"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingChairs</span></code></a> +
+            <a class="reference internal" href="../../generated/torchvision.datasets.FlyingThings3D.html#torchvision.datasets.FlyingThings3D" title="torchvision.datasets.FlyingThings3D"><code class="xref py py-class docutils literal notranslate"><span class="pre">FlyingThings3D</span></code></a>.
+Also available as <code class="docutils literal notranslate"><span class="pre">Raft_Small_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>epe (on Sintel-Train-Cleanpass)</p></td>
+<td><p>1.9901</p></td>
+</tr>
+<tr class="row-even"><td><p>epe (on Sintel-Train-Finalpass)</p></td>
+<td><p>3.2831</p></td>
+</tr>
+<tr class="row-odd"><td><p>per_image_epe (on Kitti-Train)</p></td>
+<td><p>7.5978</p></td>
+</tr>
+<tr class="row-even"><td><p>fl_all (on Kitti-Train)</p></td>
+<td><p>25.2369</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=128, width=128</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>990162</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/optical_flow">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>47.66</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>3.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Raft_Small_Weights.C_T_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are rescaled to <code class="docutils literal notranslate"><span class="pre">[-1.0,</span> <span class="pre">1.0]</span></code>.</p>
+</dd></dl>
+
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">raft_small</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="Optical flow is the task of predicting movement between two images, usually two consecutive fra..."><img alt="" src="../../_images/sphx_glr_plot_optical_flow_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_optical_flow.html#sphx-glr-auto-examples-others-plot-optical-flow-py"><span class="std std-ref">Optical Flow: Predicting movement with the RAFT model</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Optical Flow: Predicting movement with the RAFT model</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../../datasets.html" class="btn btn-neutral float-right" title="Datasets" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.optical_flow.raft_large.html" class="btn btn-neutral" title="raft_large" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">raft_small</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.googlenet.html b/0.16/models/generated/torchvision.models.quantization.googlenet.html
new file mode 100644
index 00000000000..27ec5b21bb0
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.googlenet.html
@@ -0,0 +1,806 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>googlenet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Quantized InceptionV3" href="../inception_quant.html" />
+    <link rel="prev" title="Quantized GoogLeNet" href="../googlenet_quant.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../googlenet_quant.html">Quantized GoogLeNet</a> &gt;</li>
+        
+      <li>googlenet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.googlenet.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="googlenet">
+<h1>googlenet<a class="headerlink" href="#googlenet" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.googlenet">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">googlenet</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.GoogLeNet_QuantizedWeights" title="torchvision.models.quantization.googlenet.GoogLeNet_QuantizedWeights"><span class="pre">GoogLeNet_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.googlenet.html#torchvision.models.GoogLeNet_Weights" title="torchvision.models.googlenet.GoogLeNet_Weights"><span class="pre">GoogLeNet_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableGoogLeNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/googlenet.html#googlenet"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.googlenet" title="Permalink to this definition">¶</a></dt>
+<dd><p>GoogLeNet (Inception v1) model architecture from <a class="reference external" href="http://arxiv.org/abs/1409.4842">Going Deeper with Convolutions</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.GoogLeNet_QuantizedWeights" title="torchvision.models.quantization.GoogLeNet_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.googlenet.html#torchvision.models.GoogLeNet_Weights" title="torchvision.models.GoogLeNet_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.GoogLeNet_QuantizedWeights" title="torchvision.models.quantization.GoogLeNet_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model. Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.QuantizableGoogLeNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/googlenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.GoogLeNet_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">GoogLeNet_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/googlenet.html#GoogLeNet_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.GoogLeNet_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>GoogLeNet_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+                These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+                weights listed below.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.826</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>89.404</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>6624904</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=15, width=15</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>GoogLeNet_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>1.50</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>12.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">GoogLeNet_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">GoogLeNet_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/googlenet.html#GoogLeNet_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>GoogLeNet_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.778</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>89.53</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>6624904</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=15, width=15</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#googlenet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.50</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>49.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">GoogLeNet_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../inception_quant.html" class="btn btn-neutral float-right" title="Quantized InceptionV3" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../googlenet_quant.html" class="btn btn-neutral" title="Quantized GoogLeNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">googlenet</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.inception_v3.html b/0.16/models/generated/torchvision.models.quantization.inception_v3.html
new file mode 100644
index 00000000000..e72a5df7209
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.inception_v3.html
@@ -0,0 +1,813 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>inception_v3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Quantized MobileNet V2" href="../mobilenetv2_quant.html" />
+    <link rel="prev" title="Quantized InceptionV3" href="../inception_quant.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../inception_quant.html">Quantized InceptionV3</a> &gt;</li>
+        
+      <li>inception_v3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.inception_v3.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="inception-v3">
+<h1>inception_v3<a class="headerlink" href="#inception-v3" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.inception_v3">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">inception_v3</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.Inception_V3_QuantizedWeights" title="torchvision.models.quantization.inception.Inception_V3_QuantizedWeights"><span class="pre">Inception_V3_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.inception_v3.html#torchvision.models.Inception_V3_Weights" title="torchvision.models.inception.Inception_V3_Weights"><span class="pre">Inception_V3_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableInception3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/inception.html#inception_v3"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.inception_v3" title="Permalink to this definition">¶</a></dt>
+<dd><p>Inception v3 model architecture from
+<a class="reference external" href="http://arxiv.org/abs/1512.00567">Rethinking the Inception Architecture for Computer Vision</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p><strong>Important</strong>: In contrast to the other models the inception_v3 expects tensors with a size of
+N x 3 x 299 x 299, so ensure your images are sized accordingly.</p>
+</div>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.Inception_V3_QuantizedWeights" title="torchvision.models.quantization.Inception_V3_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.inception_v3.html#torchvision.models.Inception_V3_Weights" title="torchvision.models.Inception_V3_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_Weights</span></code></a>, optional) – The pretrained
+weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.Inception_V3_QuantizedWeights" title="torchvision.models.quantization.Inception_V3_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr.
+Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model.
+Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.QuantizableInception3</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/inception.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.Inception_V3_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">Inception_V3_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/inception.html#Inception_V3_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.Inception_V3_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+                These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+                weights listed below.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.176</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.354</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>27161264</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=75, width=75</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>Inception_V3_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>5.71</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>23.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[342]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[299]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Inception_V3_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/inception.html#Inception_V3_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Inception_V3_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights are ported from the original paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.294</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.45</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>27161264</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=75, width=75</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#inception-v3">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>5.71</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>103.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Inception_V3_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[342]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[299]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../mobilenetv2_quant.html" class="btn btn-neutral float-right" title="Quantized MobileNet V2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../inception_quant.html" class="btn btn-neutral" title="Quantized InceptionV3" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">inception_v3</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html b/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
new file mode 100644
index 00000000000..77d309d6117
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
@@ -0,0 +1,840 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mobilenet_v2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Quantized MobileNet V3" href="../mobilenetv3_quant.html" />
+    <link rel="prev" title="Quantized MobileNet V2" href="../mobilenetv2_quant.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mobilenetv2_quant.html">Quantized MobileNet V2</a> &gt;</li>
+        
+      <li>mobilenet_v2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.mobilenet_v2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mobilenet-v2">
+<h1>mobilenet_v2<a class="headerlink" href="#mobilenet-v2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.mobilenet_v2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">mobilenet_v2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.MobileNet_V2_QuantizedWeights" title="torchvision.models.quantization.mobilenetv2.MobileNet_V2_QuantizedWeights"><span class="pre">MobileNet_V2_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.mobilenet_v2.html#torchvision.models.MobileNet_V2_Weights" title="torchvision.models.mobilenetv2.MobileNet_V2_Weights"><span class="pre">MobileNet_V2_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableMobileNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/mobilenetv2.html#mobilenet_v2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.mobilenet_v2" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a MobileNetV2 architecture from
+<a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear Bottlenecks</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.MobileNet_V2_QuantizedWeights" title="torchvision.models.quantization.MobileNet_V2_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.mobilenet_v2.html#torchvision.models.MobileNet_V2_Weights" title="torchvision.models.MobileNet_V2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.MobileNet_V2_QuantizedWeights" title="torchvision.models.quantization.MobileNet_V2_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, returns a quantized version of the model. Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.QuantizableMobileNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.MobileNet_V2_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">MobileNet_V2_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/mobilenetv2.html#MobileNet_V2_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.MobileNet_V2_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights.IMAGENET1K_QNNPACK_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_QNNPACK_V1'</span></code>.</p>
+<p><strong>MobileNet_V2_QuantizedWeights.IMAGENET1K_QNNPACK_V1</strong>:</p>
+<p>
+                These weights were produced by doing Quantization Aware Training (eager mode) on top of the unquantized
+                weights listed below.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>71.658</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.15</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>3504872</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>backend</p></td>
+<td><p>qnnpack</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv2">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>MobileNet_V2_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>3.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_QuantizedWeights.IMAGENET1K_QNNPACK_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MobileNet_V2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv2.html#MobileNet_V2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MobileNet_V2_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>71.878</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.286</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>3504872</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>13.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>MobileNet_V2_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>72.154</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.822</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>3504872</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>13.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V2_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../mobilenetv3_quant.html" class="btn btn-neutral float-right" title="Quantized MobileNet V3" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../mobilenetv2_quant.html" class="btn btn-neutral" title="Quantized MobileNet V2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mobilenet_v2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
new file mode 100644
index 00000000000..85e4abaa3d2
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
@@ -0,0 +1,841 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Quantized ResNet" href="../resnet_quant.html" />
+    <link rel="prev" title="Quantized MobileNet V3" href="../mobilenetv3_quant.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../mobilenetv3_quant.html">Quantized MobileNet V3</a> &gt;</li>
+        
+      <li>mobilenet_v3_large</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.mobilenet_v3_large.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mobilenet-v3-large">
+<h1>mobilenet_v3_large<a class="headerlink" href="#mobilenet-v3-large" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.mobilenet_v3_large">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">mobilenet_v3_large</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights" title="torchvision.models.quantization.mobilenetv3.MobileNet_V3_Large_QuantizedWeights"><span class="pre">MobileNet_V3_Large_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Large_Weights"><span class="pre">MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableMobileNetV3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/mobilenetv3.html#mobilenet_v3_large"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.mobilenet_v3_large" title="Permalink to this definition">¶</a></dt>
+<dd><p>MobileNetV3 (Large) model from
+<a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights" title="torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights" title="torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, return a quantized version of the model. Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv3.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">MobileNet_V3_Large_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/mobilenetv3.html#MobileNet_V3_Large_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.MobileNet_V3_Large_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights.IMAGENET1K_QNNPACK_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_QNNPACK_V1'</span></code>.</p>
+<p><strong>MobileNet_V3_Large_QuantizedWeights.IMAGENET1K_QNNPACK_V1</strong>:</p>
+<p>
+                These weights were produced by doing Quantization Aware Training (eager mode) on top of the unquantized
+                weights listed below.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>73.004</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.858</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5483032</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>backend</p></td>
+<td><p>qnnpack</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#qat-mobilenetv3">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>MobileNet_V3_Large_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>0.22</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>21.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_QuantizedWeights.IMAGENET1K_QNNPACK_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">MobileNet_V3_Large_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/mobilenetv3.html#MobileNet_V3_Large_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>MobileNet_V3_Large_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>74.042</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.34</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5483032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#mobilenetv3-large--small">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.22</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>21.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>MobileNet_V3_Large_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve marginally upon the results of the original paper by using a modified version of
+                TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.274</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.566</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5483032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuning">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.22</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>21.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../resnet_quant.html" class="btn btn-neutral float-right" title="Quantized ResNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../mobilenetv3_quant.html" class="btn btn-neutral" title="Quantized MobileNet V3" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mobilenet_v3_large</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.resnet18.html b/0.16/models/generated/torchvision.models.quantization.resnet18.html
new file mode 100644
index 00000000000..e84c3ffebc4
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.resnet18.html
@@ -0,0 +1,807 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnet18 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnet50" href="torchvision.models.quantization.resnet50.html" />
+    <link rel="prev" title="Quantized ResNet" href="../resnet_quant.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnet_quant.html">Quantized ResNet</a> &gt;</li>
+        
+      <li>resnet18</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.resnet18.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet18">
+<h1>resnet18<a class="headerlink" href="#resnet18" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.resnet18">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">resnet18</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ResNet18_QuantizedWeights" title="torchvision.models.quantization.resnet.ResNet18_QuantizedWeights"><span class="pre">ResNet18_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.resnet18.html#torchvision.models.ResNet18_Weights" title="torchvision.models.resnet.ResNet18_Weights"><span class="pre">ResNet18_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#resnet18"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.resnet18" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNet-18 model from
+<a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a></p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ResNet18_QuantizedWeights" title="torchvision.models.quantization.ResNet18_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.resnet18.html#torchvision.models.ResNet18_Weights" title="torchvision.models.ResNet18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.ResNet18_QuantizedWeights" title="torchvision.models.quantization.ResNet18_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model. Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.QuantizableResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ResNet18_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ResNet18_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#ResNet18_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ResNet18_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ResNet18_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.494</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>88.882</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>11689512</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ResNet18_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>1.81</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>11.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet18_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNet18_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNet18_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNet18_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.758</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>89.078</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>11689512</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.81</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>44.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.quantization.resnet50.html" class="btn btn-neutral float-right" title="resnet50" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../resnet_quant.html" class="btn btn-neutral" title="Quantized ResNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnet18</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.resnet50.html b/0.16/models/generated/torchvision.models.quantization.resnet50.html
new file mode 100644
index 00000000000..6c0a140a458
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.resnet50.html
@@ -0,0 +1,879 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnet50 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Quantized ResNeXt" href="../resnext_quant.html" />
+    <link rel="prev" title="resnet18" href="torchvision.models.quantization.resnet18.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnet_quant.html">Quantized ResNet</a> &gt;</li>
+        
+      <li>resnet50</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.resnet50.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet50">
+<h1>resnet50<a class="headerlink" href="#resnet50" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.resnet50">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">resnet50</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ResNet50_QuantizedWeights" title="torchvision.models.quantization.resnet.ResNet50_QuantizedWeights"><span class="pre">ResNet50_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#resnet50"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.resnet50" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNet-50 model from
+<a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a></p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ResNet50_QuantizedWeights" title="torchvision.models.quantization.ResNet50_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.ResNet50_QuantizedWeights" title="torchvision.models.quantization.ResNet50_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model. Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.QuantizableResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ResNet50_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ResNet50_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#ResNet50_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ResNet50_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.92</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.814</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25557032</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ResNet50_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>4.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>24.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V2</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.282</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.976</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25557032</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ResNet50_Weights.IMAGENET1K_V2</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>4.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>25.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet50_QuantizedWeights.IMAGENET1K_FBGEMM_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNet50_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNet50_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNet50_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.13</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.862</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25557032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>97.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNet50_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.858</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.434</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25557032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#issuecomment-1013906621">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>97.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../resnext_quant.html" class="btn btn-neutral float-right" title="Quantized ResNeXt" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.quantization.resnet18.html" class="btn btn-neutral" title="resnet18" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnet50</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html b/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
new file mode 100644
index 00000000000..8ac2c42fb34
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
@@ -0,0 +1,879 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnext101_32x8d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnext101_64x4d" href="torchvision.models.quantization.resnext101_64x4d.html" />
+    <link rel="prev" title="Quantized ResNeXt" href="../resnext_quant.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnext_quant.html">Quantized ResNeXt</a> &gt;</li>
+        
+      <li>resnext101_32x8d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.resnext101_32x8d.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnext101-32x8d">
+<h1>resnext101_32x8d<a class="headerlink" href="#resnext101-32x8d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.resnext101_32x8d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">resnext101_32x8d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights" title="torchvision.models.quantization.resnet.ResNeXt101_32X8D_QuantizedWeights"><span class="pre">ResNeXt101_32X8D_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.resnext101_32x8d.html#torchvision.models.ResNeXt101_32X8D_Weights" title="torchvision.models.resnet.ResNeXt101_32X8D_Weights"><span class="pre">ResNeXt101_32X8D_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#resnext101_32x8d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.resnext101_32x8d" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNeXt-101 32x8d model from
+<a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a></p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights" title="torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.resnext101_32x8d.html#torchvision.models.ResNeXt101_32X8D_Weights" title="torchvision.models.ResNeXt101_32X8D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_32X8D_QuantizedWeights</span></code> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model. Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.QuantizableResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ResNeXt101_32X8D_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#ResNeXt101_32X8D_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ResNeXt101_32X8D_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.986</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.48</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88791336</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ResNeXt101_32X8D_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>16.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>86.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V2</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.574</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.132</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88791336</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ResNeXt101_32X8D_Weights.IMAGENET1K_V2</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>16.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>86.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_QuantizedWeights.IMAGENET1K_FBGEMM_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNeXt101_32X8D_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNeXt101_32X8D_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNeXt101_32X8D_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>79.312</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.526</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88791336</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>16.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>339.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNeXt101_32X8D_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.834</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.228</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88791336</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>16.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>339.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.quantization.resnext101_64x4d.html" class="btn btn-neutral float-right" title="resnext101_64x4d" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../resnext_quant.html" class="btn btn-neutral" title="Quantized ResNeXt" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnext101_32x8d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html b/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
new file mode 100644
index 00000000000..50835f1de89
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
@@ -0,0 +1,809 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnext101_64x4d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Quantized ShuffleNet V2" href="../shufflenetv2_quant.html" />
+    <link rel="prev" title="resnext101_32x8d" href="torchvision.models.quantization.resnext101_32x8d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnext_quant.html">Quantized ResNeXt</a> &gt;</li>
+        
+      <li>resnext101_64x4d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.resnext101_64x4d.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnext101-64x4d">
+<h1>resnext101_64x4d<a class="headerlink" href="#resnext101-64x4d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.resnext101_64x4d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">resnext101_64x4d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights" title="torchvision.models.quantization.resnet.ResNeXt101_64X4D_QuantizedWeights"><span class="pre">ResNeXt101_64X4D_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.resnext101_64x4d.html#torchvision.models.ResNeXt101_64X4D_Weights" title="torchvision.models.resnet.ResNeXt101_64X4D_Weights"><span class="pre">ResNeXt101_64X4D_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#resnext101_64x4d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.resnext101_64x4d" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNeXt-101 64x4d model from
+<a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a></p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights" title="torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.resnext101_64x4d.html#torchvision.models.ResNeXt101_64X4D_Weights" title="torchvision.models.ResNeXt101_64X4D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_64X4D_QuantizedWeights</span></code> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model. Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.QuantizableResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ResNeXt101_64X4D_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/resnet.html#ResNeXt101_64X4D_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ResNeXt101_64X4D_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ResNeXt101_64X4D_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.898</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.326</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5935">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>83455272</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ResNeXt101_64X4D_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>15.46</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>81.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNeXt101_64X4D_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNeXt101_64X4D_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNeXt101_64X4D_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.246</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.454</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>83455272</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5935">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.46</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>319.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../shufflenetv2_quant.html" class="btn btn-neutral float-right" title="Quantized ShuffleNet V2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.quantization.resnext101_32x8d.html" class="btn btn-neutral" title="resnext101_32x8d" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnext101_64x4d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
new file mode 100644
index 00000000000..b9a2fdbc0e9
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
@@ -0,0 +1,808 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x0_5 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="shufflenet_v2_x1_0" href="torchvision.models.quantization.shufflenet_v2_x1_0.html" />
+    <link rel="prev" title="Quantized ShuffleNet V2" href="../shufflenetv2_quant.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2_quant.html">Quantized ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x0_5</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x0-5">
+<h1>shufflenet_v2_x0_5<a class="headerlink" href="#shufflenet-v2-x0-5" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.shufflenet_v2_x0_5">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x0_5</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights" title="torchvision.models.quantization.shufflenetv2.ShuffleNet_V2_X0_5_QuantizedWeights"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.ShuffleNet_V2_X0_5_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X0_5_Weights"><span class="pre">ShuffleNet_V2_X0_5_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#shufflenet_v2_x0_5"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.shufflenet_v2_x0_5" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 with 0.5x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.ShuffleNet_V2_X0_5_Weights" title="torchvision.models.ShuffleNet_V2_X0_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr.
+Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model.
+Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#ShuffleNet_V2_X0_5_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ShuffleNet_V2_X0_5_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X0_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>57.972</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>79.78</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>1366792</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>0.04</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>1.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X0_5_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X0_5_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch to reproduce closely the results of the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>60.552</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>81.746</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>1366792</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.04</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>5.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.quantization.shufflenet_v2_x1_0.html" class="btn btn-neutral float-right" title="shufflenet_v2_x1_0" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../shufflenetv2_quant.html" class="btn btn-neutral" title="Quantized ShuffleNet V2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x0_5</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
new file mode 100644
index 00000000000..b8c7debc0a2
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
@@ -0,0 +1,808 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x1_0 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="shufflenet_v2_x1_5" href="torchvision.models.quantization.shufflenet_v2_x1_5.html" />
+    <link rel="prev" title="shufflenet_v2_x0_5" href="torchvision.models.quantization.shufflenet_v2_x0_5.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2_quant.html">Quantized ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x1_0</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x1-0">
+<h1>shufflenet_v2_x1_0<a class="headerlink" href="#shufflenet-v2-x1-0" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.shufflenet_v2_x1_0">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x1_0</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights" title="torchvision.models.quantization.shufflenetv2.ShuffleNet_V2_X1_0_QuantizedWeights"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.ShuffleNet_V2_X1_0_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X1_0_Weights"><span class="pre">ShuffleNet_V2_X1_0_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#shufflenet_v2_x1_0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.shufflenet_v2_x1_0" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 with 1.0x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.ShuffleNet_V2_X1_0_Weights" title="torchvision.models.ShuffleNet_V2_X1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr.
+Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model.
+Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#ShuffleNet_V2_X1_0_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ShuffleNet_V2_X1_0_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X1_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>68.36</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>87.582</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#post-training-quantized-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>2278604</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>0.14</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>2.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X1_0_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X1_0_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch to reproduce closely the results of the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.362</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>88.316</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>2278604</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.14</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>8.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.quantization.shufflenet_v2_x1_5.html" class="btn btn-neutral float-right" title="shufflenet_v2_x1_5" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.quantization.shufflenet_v2_x0_5.html" class="btn btn-neutral" title="shufflenet_v2_x0_5" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x1_0</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
new file mode 100644
index 00000000000..bbd2efdf839
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
@@ -0,0 +1,810 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x1_5 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="shufflenet_v2_x2_0" href="torchvision.models.quantization.shufflenet_v2_x2_0.html" />
+    <link rel="prev" title="shufflenet_v2_x1_0" href="torchvision.models.quantization.shufflenet_v2_x1_0.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2_quant.html">Quantized ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x1_5</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x1-5">
+<h1>shufflenet_v2_x1_5<a class="headerlink" href="#shufflenet-v2-x1-5" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.shufflenet_v2_x1_5">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x1_5</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights" title="torchvision.models.quantization.shufflenetv2.ShuffleNet_V2_X1_5_QuantizedWeights"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.ShuffleNet_V2_X1_5_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X1_5_Weights"><span class="pre">ShuffleNet_V2_X1_5_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#shufflenet_v2_x1_5"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.shufflenet_v2_x1_5" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 with 1.5x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.ShuffleNet_V2_X1_5_Weights" title="torchvision.models.ShuffleNet_V2_X1_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr.
+Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model.
+Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#ShuffleNet_V2_X1_5_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ShuffleNet_V2_X1_5_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X1_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>72.052</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.7</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>3503624</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>3.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X1_5_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X1_5_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>72.996</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.086</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>3503624</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>13.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.quantization.shufflenet_v2_x2_0.html" class="btn btn-neutral float-right" title="shufflenet_v2_x2_0" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.quantization.shufflenet_v2_x1_0.html" class="btn btn-neutral" title="shufflenet_v2_x1_0" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x1_5</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
new file mode 100644
index 00000000000..8077f35d2f0
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
@@ -0,0 +1,810 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x2_0 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="DeepLabV3" href="../deeplabv3.html" />
+    <link rel="prev" title="shufflenet_v2_x1_5" href="torchvision.models.quantization.shufflenet_v2_x1_5.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2_quant.html">Quantized ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x2_0</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x2-0">
+<h1>shufflenet_v2_x2_0<a class="headerlink" href="#shufflenet-v2-x2-0" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.quantization.shufflenet_v2_x2_0">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x2_0</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><span class="pre">Union</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights" title="torchvision.models.quantization.shufflenetv2.ShuffleNet_V2_X2_0_QuantizedWeights"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights</span></a><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.ShuffleNet_V2_X2_0_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X2_0_Weights"><span class="pre">ShuffleNet_V2_X2_0_Weights</span></a><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">quantize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">QuantizableShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#shufflenet_v2_x2_0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.shufflenet_v2_x2_0" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 with 2.0x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Note that <code class="docutils literal notranslate"><span class="pre">quantize</span> <span class="pre">=</span> <span class="pre">True</span></code> returns a quantized model with 8 bit
+weights. Quantized models only support inference and run on CPUs.
+GPU inference is not yet supported.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights</span></code></a> or <a class="reference internal" href="torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.ShuffleNet_V2_X2_0_Weights" title="torchvision.models.ShuffleNet_V2_X2_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights</span></code></a>, optional) – The
+pretrained weights for the model. See
+<a class="reference internal" href="#torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights" title="torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr.
+Default is True.</p></li>
+<li><p><strong>quantize</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, return a quantized version of the model.
+Default is False.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.quantization.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/quantization/shufflenetv2.html#ShuffleNet_V2_X2_0_QuantizedWeights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.quantization.ShuffleNet_V2_X2_0_QuantizedWeights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_FBGEMM_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X2_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1</strong>:</p>
+<p>
+        These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized
+        weights listed below.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.354</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.488</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>backend</p></td>
+<td><p>fbgemm</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>7393996</p></td>
+</tr>
+<tr class="row-even"><td><p>unquantized</p></td>
+<td><p>ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1</p></td>
+</tr>
+<tr class="row-odd"><td><p>GIPS</p></td>
+<td><p>0.58</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>7.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_QuantizedWeights.IMAGENET1K_FBGEMM_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X2_0_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X2_0_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.23</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.006</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>7393996</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.58</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>28.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../deeplabv3.html" class="btn btn-neutral float-right" title="DeepLabV3" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.quantization.shufflenet_v2_x1_5.html" class="btn btn-neutral" title="shufflenet_v2_x1_5" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x2_0</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_x_16gf.html b/0.16/models/generated/torchvision.models.regnet_x_16gf.html
new file mode 100644
index 00000000000..1a6649d79d0
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_x_16gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_x_16gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_x_32gf" href="torchvision.models.regnet_x_32gf.html" />
+    <link rel="prev" title="regnet_x_8gf" href="torchvision.models.regnet_x_8gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_x_16gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_x_16gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-x-16gf">
+<h1>regnet_x_16gf<a class="headerlink" href="#regnet-x-16gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_x_16gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_x_16gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_X_16GF_Weights" title="torchvision.models.regnet.RegNet_X_16GF_Weights"><span class="pre">RegNet_X_16GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_x_16gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_x_16gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetX_16GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_X_16GF_Weights" title="torchvision.models.RegNet_X_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_X_16GF_Weights" title="torchvision.models.RegNet_X_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_X_16GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_X_16GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_X_16GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_X_16GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_X_16GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.058</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.944</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>54278536</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.94</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>207.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_X_16GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.716</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.196</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>54278536</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.94</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>207.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_16GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_x_32gf.html" class="btn btn-neutral float-right" title="regnet_x_32gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_x_8gf.html" class="btn btn-neutral" title="regnet_x_8gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_x_16gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html b/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
new file mode 100644
index 00000000000..fcddd6c99cf
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_x_1_6gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_x_3_2gf" href="torchvision.models.regnet_x_3_2gf.html" />
+    <link rel="prev" title="regnet_x_800mf" href="torchvision.models.regnet_x_800mf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_x_1_6gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_x_1_6gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-x-1-6gf">
+<h1>regnet_x_1_6gf<a class="headerlink" href="#regnet-x-1-6gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_x_1_6gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_x_1_6gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_X_1_6GF_Weights" title="torchvision.models.regnet.RegNet_X_1_6GF_Weights"><span class="pre">RegNet_X_1_6GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_x_1_6gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_x_1_6gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetX_1.6GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_X_1_6GF_Weights" title="torchvision.models.RegNet_X_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_X_1_6GF_Weights" title="torchvision.models.RegNet_X_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_X_1_6GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_X_1_6GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_X_1_6GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_X_1_6GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_X_1_6GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.04</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.44</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>9190136</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.60</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>35.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_X_1_6GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>79.668</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.922</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>9190136</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.60</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>35.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_1_6GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_x_3_2gf.html" class="btn btn-neutral float-right" title="regnet_x_3_2gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_x_800mf.html" class="btn btn-neutral" title="regnet_x_800mf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_x_1_6gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_x_32gf.html b/0.16/models/generated/torchvision.models.regnet_x_32gf.html
new file mode 100644
index 00000000000..e2dd42cd1ab
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_x_32gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_x_32gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="ResNet" href="../resnet.html" />
+    <link rel="prev" title="regnet_x_16gf" href="torchvision.models.regnet_x_16gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_x_32gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_x_32gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-x-32gf">
+<h1>regnet_x_32gf<a class="headerlink" href="#regnet-x-32gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_x_32gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_x_32gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_X_32GF_Weights" title="torchvision.models.regnet.RegNet_X_32GF_Weights"><span class="pre">RegNet_X_32GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_x_32gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_x_32gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetX_32GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_X_32GF_Weights" title="torchvision.models.RegNet_X_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_X_32GF_Weights" title="torchvision.models.RegNet_X_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_X_32GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_X_32GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_X_32GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_X_32GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_X_32GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.622</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.248</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>107811560</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>31.74</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>412.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_X_32GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.014</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.288</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>107811560</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>31.74</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>412.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_32GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../resnet.html" class="btn btn-neutral float-right" title="ResNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_x_16gf.html" class="btn btn-neutral" title="regnet_x_16gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_x_32gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html b/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
new file mode 100644
index 00000000000..0262ffd235b
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_x_3_2gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_x_8gf" href="torchvision.models.regnet_x_8gf.html" />
+    <link rel="prev" title="regnet_x_1_6gf" href="torchvision.models.regnet_x_1_6gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_x_3_2gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_x_3_2gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-x-3-2gf">
+<h1>regnet_x_3_2gf<a class="headerlink" href="#regnet-x-3-2gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_x_3_2gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_x_3_2gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_X_3_2GF_Weights" title="torchvision.models.regnet.RegNet_X_3_2GF_Weights"><span class="pre">RegNet_X_3_2GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_x_3_2gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_x_3_2gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetX_3.2GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_X_3_2GF_Weights" title="torchvision.models.RegNet_X_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_X_3_2GF_Weights" title="torchvision.models.RegNet_X_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_X_3_2GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_X_3_2GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_X_3_2GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_X_3_2GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_X_3_2GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.364</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.992</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>15296552</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>3.18</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>58.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_X_3_2GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.196</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.43</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>15296552</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>3.18</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>58.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_3_2GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_x_8gf.html" class="btn btn-neutral float-right" title="regnet_x_8gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_x_1_6gf.html" class="btn btn-neutral" title="regnet_x_1_6gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_x_3_2gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_x_400mf.html b/0.16/models/generated/torchvision.models.regnet_x_400mf.html
new file mode 100644
index 00000000000..14e43d5fb69
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_x_400mf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_x_400mf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_x_800mf" href="torchvision.models.regnet_x_800mf.html" />
+    <link rel="prev" title="regnet_y_128gf" href="torchvision.models.regnet_y_128gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_x_400mf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_x_400mf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-x-400mf">
+<h1>regnet_x_400mf<a class="headerlink" href="#regnet-x-400mf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_x_400mf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_x_400mf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_X_400MF_Weights" title="torchvision.models.regnet.RegNet_X_400MF_Weights"><span class="pre">RegNet_X_400MF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_x_400mf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_x_400mf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetX_400MF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_X_400MF_Weights" title="torchvision.models.RegNet_X_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_X_400MF_Weights" title="torchvision.models.RegNet_X_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_X_400MF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_X_400MF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_X_400MF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_X_400MF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_X_400MF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>72.834</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.95</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5495976</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>21.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_X_400MF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>74.864</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.322</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>5495976</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>21.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_400MF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_x_800mf.html" class="btn btn-neutral float-right" title="regnet_x_800mf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_128gf.html" class="btn btn-neutral" title="regnet_y_128gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_x_400mf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_x_800mf.html b/0.16/models/generated/torchvision.models.regnet_x_800mf.html
new file mode 100644
index 00000000000..d67962b5a8a
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_x_800mf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_x_800mf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_x_1_6gf" href="torchvision.models.regnet_x_1_6gf.html" />
+    <link rel="prev" title="regnet_x_400mf" href="torchvision.models.regnet_x_400mf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_x_800mf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_x_800mf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-x-800mf">
+<h1>regnet_x_800mf<a class="headerlink" href="#regnet-x-800mf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_x_800mf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_x_800mf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_X_800MF_Weights" title="torchvision.models.regnet.RegNet_X_800MF_Weights"><span class="pre">RegNet_X_800MF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_x_800mf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_x_800mf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetX_800MF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_X_800MF_Weights" title="torchvision.models.RegNet_X_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_X_800MF_Weights" title="torchvision.models.RegNet_X_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_X_800MF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_X_800MF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_X_800MF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_X_800MF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_X_800MF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.212</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.348</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>7259656</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.80</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>27.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_X_800MF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.522</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.826</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>7259656</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.80</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>27.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_800MF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_x_1_6gf.html" class="btn btn-neutral float-right" title="regnet_x_1_6gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_x_400mf.html" class="btn btn-neutral" title="regnet_x_400mf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_x_800mf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_x_8gf.html b/0.16/models/generated/torchvision.models.regnet_x_8gf.html
new file mode 100644
index 00000000000..2b014723b09
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_x_8gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_x_8gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_x_16gf" href="torchvision.models.regnet_x_16gf.html" />
+    <link rel="prev" title="regnet_x_3_2gf" href="torchvision.models.regnet_x_3_2gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_x_8gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_x_8gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-x-8gf">
+<h1>regnet_x_8gf<a class="headerlink" href="#regnet-x-8gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_x_8gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_x_8gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_X_8GF_Weights" title="torchvision.models.regnet.RegNet_X_8GF_Weights"><span class="pre">RegNet_X_8GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_x_8gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_x_8gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetX_8GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_X_8GF_Weights" title="torchvision.models.RegNet_X_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_X_8GF_Weights" title="torchvision.models.RegNet_X_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_X_8GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_X_8GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_X_8GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_X_8GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_X_8GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>79.344</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.686</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>39572648</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>8.00</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>151.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_X_8GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.682</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.678</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>39572648</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>8.00</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>151.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_X_8GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_x_16gf.html" class="btn btn-neutral float-right" title="regnet_x_16gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_x_3_2gf.html" class="btn btn-neutral" title="regnet_x_3_2gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_x_8gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_128gf.html b/0.16/models/generated/torchvision.models.regnet_y_128gf.html
new file mode 100644
index 00000000000..f8c266a2042
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_128gf.html
@@ -0,0 +1,791 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_128gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_x_400mf" href="torchvision.models.regnet_x_400mf.html" />
+    <link rel="prev" title="regnet_y_32gf" href="torchvision.models.regnet_y_32gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_128gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_128gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-128gf">
+<h1>regnet_y_128gf<a class="headerlink" href="#regnet-y-128gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_128gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_128gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_128GF_Weights" title="torchvision.models.regnet.RegNet_Y_128GF_Weights"><span class="pre">RegNet_Y_128GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_128gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_128gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_128GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_128GF_Weights" title="torchvision.models.RegNet_Y_128GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_128GF_Weights" title="torchvision.models.RegNet_Y_128GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_128GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_128GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_128GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_128GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_E2E_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_SWAG_E2E_V1'</span></code>.</p>
+<p><strong>RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_E2E_V1</strong>:</p>
+<p>
+                These weights are learnt via transfer learning by end-to-end fine-tuning the original
+                <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> weights on ImageNet-1K data.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>88.228</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>98.682</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>644812894</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>374.57</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>2461.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_E2E_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[384]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[384]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</strong>:</p>
+<p>
+                These weights are composed of the original frozen <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> trunk
+                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>86.068</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.844</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>644812894</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>127.52</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>2461.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_128GF_Weights.IMAGENET1K_SWAG_LINEAR_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[224]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_x_400mf.html" class="btn btn-neutral float-right" title="regnet_x_400mf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_32gf.html" class="btn btn-neutral" title="regnet_y_32gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_128gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_16gf.html b/0.16/models/generated/torchvision.models.regnet_y_16gf.html
new file mode 100644
index 00000000000..fc7ce99eedf
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_16gf.html
@@ -0,0 +1,855 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_16gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_y_32gf" href="torchvision.models.regnet_y_32gf.html" />
+    <link rel="prev" title="regnet_y_8gf" href="torchvision.models.regnet_y_8gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_16gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_16gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-16gf">
+<h1>regnet_y_16gf<a class="headerlink" href="#regnet-y-16gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_16gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_16gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_16GF_Weights" title="torchvision.models.regnet.RegNet_Y_16GF_Weights"><span class="pre">RegNet_Y_16GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_16gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_16gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_16GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_16GF_Weights" title="torchvision.models.RegNet_Y_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_16GF_Weights" title="torchvision.models.RegNet_Y_16GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_16GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_16GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_16GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_16GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_Y_16GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.424</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.24</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>83590140</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.91</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>319.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_16GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.886</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.328</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>83590140</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.91</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>319.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_E2E_V1</strong>:</p>
+<p>
+                These weights are learnt via transfer learning by end-to-end fine-tuning the original
+                <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> weights on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>86.012</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>98.054</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>83590140</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>46.73</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>319.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_E2E_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[384]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[384]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</strong>:</p>
+<p>
+                These weights are composed of the original frozen <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> trunk
+                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.976</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.244</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>83590140</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>15.91</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>319.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_16GF_Weights.IMAGENET1K_SWAG_LINEAR_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[224]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_y_32gf.html" class="btn btn-neutral float-right" title="regnet_y_32gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_8gf.html" class="btn btn-neutral" title="regnet_y_8gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_16gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html b/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
new file mode 100644
index 00000000000..7b3d1eac693
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_1_6gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_y_3_2gf" href="torchvision.models.regnet_y_3_2gf.html" />
+    <link rel="prev" title="regnet_y_800mf" href="torchvision.models.regnet_y_800mf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_1_6gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_1_6gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-1-6gf">
+<h1>regnet_y_1_6gf<a class="headerlink" href="#regnet-y-1-6gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_1_6gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_1_6gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_1_6GF_Weights" title="torchvision.models.regnet.RegNet_Y_1_6GF_Weights"><span class="pre">RegNet_Y_1_6GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_1_6gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_1_6gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_1.6GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_1_6GF_Weights" title="torchvision.models.RegNet_Y_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_1_6GF_Weights" title="torchvision.models.RegNet_Y_1_6GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_1_6GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_1_6GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_1_6GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_1_6GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_Y_1_6GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.95</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.966</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>11202430</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.61</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>43.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_1_6GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.876</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.444</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>11202430</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.61</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>43.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_1_6GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_y_3_2gf.html" class="btn btn-neutral float-right" title="regnet_y_3_2gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_800mf.html" class="btn btn-neutral" title="regnet_y_800mf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_1_6gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_32gf.html b/0.16/models/generated/torchvision.models.regnet_y_32gf.html
new file mode 100644
index 00000000000..d6bf6d78a1f
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_32gf.html
@@ -0,0 +1,855 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_32gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_y_128gf" href="torchvision.models.regnet_y_128gf.html" />
+    <link rel="prev" title="regnet_y_16gf" href="torchvision.models.regnet_y_16gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_32gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_32gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-32gf">
+<h1>regnet_y_32gf<a class="headerlink" href="#regnet-y-32gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_32gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_32gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_32GF_Weights" title="torchvision.models.regnet.RegNet_Y_32GF_Weights"><span class="pre">RegNet_Y_32GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_32gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_32gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_32GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_32GF_Weights" title="torchvision.models.RegNet_Y_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_32GF_Weights" title="torchvision.models.RegNet_Y_32GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_32GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_32GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_32GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_32GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_Y_32GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.878</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.34</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>145046770</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#large-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>32.28</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>554.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_32GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.368</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.498</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>145046770</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>32.28</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>554.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_E2E_V1</strong>:</p>
+<p>
+                These weights are learnt via transfer learning by end-to-end fine-tuning the original
+                <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> weights on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>86.838</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>98.362</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>145046770</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>94.83</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>554.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_E2E_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[384]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[384]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_LINEAR_V1</strong>:</p>
+<p>
+                These weights are composed of the original frozen <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> trunk
+                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>84.622</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.48</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>145046770</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>32.28</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>554.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_32GF_Weights.IMAGENET1K_SWAG_LINEAR_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[224]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_y_128gf.html" class="btn btn-neutral float-right" title="regnet_y_128gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_16gf.html" class="btn btn-neutral" title="regnet_y_16gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_32gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html b/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
new file mode 100644
index 00000000000..12a56a574f7
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_3_2gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_y_8gf" href="torchvision.models.regnet_y_8gf.html" />
+    <link rel="prev" title="regnet_y_1_6gf" href="torchvision.models.regnet_y_1_6gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_3_2gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_3_2gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-3-2gf">
+<h1>regnet_y_3_2gf<a class="headerlink" href="#regnet-y-3-2gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_3_2gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_3_2gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_3_2GF_Weights" title="torchvision.models.regnet.RegNet_Y_3_2GF_Weights"><span class="pre">RegNet_Y_3_2GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_3_2gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_3_2gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_3.2GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_3_2GF_Weights" title="torchvision.models.RegNet_Y_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_3_2GF_Weights" title="torchvision.models.RegNet_Y_3_2GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_3_2GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_3_2GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_3_2GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_3_2GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_Y_3_2GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.948</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.576</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>19436338</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>3.18</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>74.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_3_2GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.982</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.972</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>19436338</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>3.18</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>74.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_3_2GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_y_8gf.html" class="btn btn-neutral float-right" title="regnet_y_8gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_1_6gf.html" class="btn btn-neutral" title="regnet_y_1_6gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_3_2gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_400mf.html b/0.16/models/generated/torchvision.models.regnet_y_400mf.html
new file mode 100644
index 00000000000..2d21017007f
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_400mf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_400mf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_y_800mf" href="torchvision.models.regnet_y_800mf.html" />
+    <link rel="prev" title="RegNet" href="../regnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_400mf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_400mf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-400mf">
+<h1>regnet_y_400mf<a class="headerlink" href="#regnet-y-400mf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_400mf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_400mf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_400MF_Weights" title="torchvision.models.regnet.RegNet_Y_400MF_Weights"><span class="pre">RegNet_Y_400MF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_400mf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_400mf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_400MF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_400MF_Weights" title="torchvision.models.RegNet_Y_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_400MF_Weights" title="torchvision.models.RegNet_Y_400MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_400MF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_400MF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_400MF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_400MF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_Y_400MF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>74.046</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.716</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>4344144</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.40</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>16.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_400MF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.804</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.742</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>4344144</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.40</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>16.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_400MF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_y_800mf.html" class="btn btn-neutral float-right" title="regnet_y_800mf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../regnet.html" class="btn btn-neutral" title="RegNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_400mf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_800mf.html b/0.16/models/generated/torchvision.models.regnet_y_800mf.html
new file mode 100644
index 00000000000..992e9a5cdf8
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_800mf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_800mf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_y_1_6gf" href="torchvision.models.regnet_y_1_6gf.html" />
+    <link rel="prev" title="regnet_y_400mf" href="torchvision.models.regnet_y_400mf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_800mf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_800mf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-800mf">
+<h1>regnet_y_800mf<a class="headerlink" href="#regnet-y-800mf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_800mf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_800mf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_800MF_Weights" title="torchvision.models.regnet.RegNet_Y_800MF_Weights"><span class="pre">RegNet_Y_800MF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_800mf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_800mf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_800MF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_800MF_Weights" title="torchvision.models.RegNet_Y_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_800MF_Weights" title="torchvision.models.RegNet_Y_800MF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_800MF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_800MF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_800MF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_800MF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_Y_800MF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.42</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.136</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>6432512</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#small-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.83</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>24.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_800MF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.828</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.502</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>6432512</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.83</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>24.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_800MF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_y_1_6gf.html" class="btn btn-neutral float-right" title="regnet_y_1_6gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_400mf.html" class="btn btn-neutral" title="regnet_y_400mf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_800mf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.regnet_y_8gf.html b/0.16/models/generated/torchvision.models.regnet_y_8gf.html
new file mode 100644
index 00000000000..1e7cd5f62a7
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.regnet_y_8gf.html
@@ -0,0 +1,783 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>regnet_y_8gf &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="regnet_y_16gf" href="torchvision.models.regnet_y_16gf.html" />
+    <link rel="prev" title="regnet_y_3_2gf" href="torchvision.models.regnet_y_3_2gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../regnet.html">RegNet</a> &gt;</li>
+        
+      <li>regnet_y_8gf</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.regnet_y_8gf.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet-y-8gf">
+<h1>regnet_y_8gf<a class="headerlink" href="#regnet-y-8gf" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.regnet_y_8gf">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">regnet_y_8gf</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.RegNet_Y_8GF_Weights" title="torchvision.models.regnet.RegNet_Y_8GF_Weights"><span class="pre">RegNet_Y_8GF_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">RegNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#regnet_y_8gf"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.regnet_y_8gf" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a RegNetY_8GF architecture from
+<a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.RegNet_Y_8GF_Weights" title="torchvision.models.RegNet_Y_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights</span></code></a>, optional) – The pretrained weights to use.
+See <a class="reference internal" href="#torchvision.models.RegNet_Y_8GF_Weights" title="torchvision.models.RegNet_Y_8GF_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights</span></code></a> below for more details and possible values.
+By default, no pretrained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to either <code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> or
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.BlockParams</span></code> class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a>
+for more detail about the classes.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.RegNet_Y_8GF_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">RegNet_Y_8GF_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/regnet.html#RegNet_Y_8GF_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.RegNet_Y_8GF_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>RegNet_Y_8GF_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.032</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.048</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>39381472</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#medium-models">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>8.47</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>150.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>RegNet_Y_8GF_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.828</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.33</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>39381472</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>8.47</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>150.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">RegNet_Y_8GF_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.regnet_y_16gf.html" class="btn btn-neutral float-right" title="regnet_y_16gf" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.regnet_y_3_2gf.html" class="btn btn-neutral" title="regnet_y_3_2gf" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">regnet_y_8gf</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnet101.html b/0.16/models/generated/torchvision.models.resnet101.html
new file mode 100644
index 00000000000..1605c98fd62
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnet101.html
@@ -0,0 +1,790 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnet101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnet152" href="torchvision.models.resnet152.html" />
+    <link rel="prev" title="resnet50" href="torchvision.models.resnet50.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnet.html">ResNet</a> &gt;</li>
+        
+      <li>resnet101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnet101.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet101">
+<h1>resnet101<a class="headerlink" href="#resnet101" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnet101">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnet101</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNet101_Weights" title="torchvision.models.resnet.ResNet101_Weights"><span class="pre">ResNet101_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnet101"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnet101" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNet-101 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The bottleneck of TorchVision places the stride for downsampling to the second 3x3
+convolution while the original paper places it to the first 1x1 convolution.
+This variant improves the accuracy and is known as <a class="reference external" href="https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch">ResNet V1.5</a>.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNet101_Weights" title="torchvision.models.ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ResNet101_Weights" title="torchvision.models.ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNet101_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNet101_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNet101_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNet101_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet101_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet101_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNet101_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.374</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.546</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>44549160</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>7.80</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>170.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet101_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNet101_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet101_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.886</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.78</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>44549160</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>7.80</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>170.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet101_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.resnet152.html" class="btn btn-neutral float-right" title="resnet152" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.resnet50.html" class="btn btn-neutral" title="resnet50" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnet101</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnet152.html b/0.16/models/generated/torchvision.models.resnet152.html
new file mode 100644
index 00000000000..783b08e7fe5
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnet152.html
@@ -0,0 +1,790 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnet152 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="ResNeXt" href="../resnext.html" />
+    <link rel="prev" title="resnet101" href="torchvision.models.resnet101.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnet.html">ResNet</a> &gt;</li>
+        
+      <li>resnet152</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnet152.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet152">
+<h1>resnet152<a class="headerlink" href="#resnet152" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnet152">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnet152</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNet152_Weights" title="torchvision.models.resnet.ResNet152_Weights"><span class="pre">ResNet152_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnet152"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnet152" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNet-152 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The bottleneck of TorchVision places the stride for downsampling to the second 3x3
+convolution while the original paper places it to the first 1x1 convolution.
+This variant improves the accuracy and is known as <a class="reference external" href="https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch">ResNet V1.5</a>.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNet152_Weights" title="torchvision.models.ResNet152_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet152_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ResNet152_Weights" title="torchvision.models.ResNet152_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet152_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNet152_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNet152_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNet152_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNet152_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet152_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet152_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNet152_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.312</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.046</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>60192808</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>11.51</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>230.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet152_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNet152_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet152_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.284</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.002</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>60192808</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>11.51</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>230.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet152_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../resnext.html" class="btn btn-neutral float-right" title="ResNeXt" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.resnet101.html" class="btn btn-neutral" title="resnet101" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnet152</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnet18.html b/0.16/models/generated/torchvision.models.resnet18.html
new file mode 100644
index 00000000000..714f107e86b
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnet18.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnet18 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnet34" href="torchvision.models.resnet34.html" />
+    <link rel="prev" title="ResNet" href="../resnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnet.html">ResNet</a> &gt;</li>
+        
+      <li>resnet18</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnet18.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet18">
+<h1>resnet18<a class="headerlink" href="#resnet18" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnet18">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnet18</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNet18_Weights" title="torchvision.models.resnet.ResNet18_Weights"><span class="pre">ResNet18_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnet18"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnet18" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNet-18 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNet18_Weights" title="torchvision.models.ResNet18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ResNet18_Weights" title="torchvision.models.ResNet18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet18_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNet18_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNet18_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNet18_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNet18_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNet18_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.758</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>89.078</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>11689512</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>1.81</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>44.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet18_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">resnet18</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates torchscript support of the torchvision transforms on Tensor images."><img alt="" src="../../_images/sphx_glr_plot_scripted_tensor_transforms_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Torchscript support</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.resnet34.html" class="btn btn-neutral float-right" title="resnet34" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../resnet.html" class="btn btn-neutral" title="ResNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnet18</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnet34.html b/0.16/models/generated/torchvision.models.resnet34.html
new file mode 100644
index 00000000000..621443eae1f
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnet34.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnet34 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnet50" href="torchvision.models.resnet50.html" />
+    <link rel="prev" title="resnet18" href="torchvision.models.resnet18.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnet.html">ResNet</a> &gt;</li>
+        
+      <li>resnet34</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnet34.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet34">
+<h1>resnet34<a class="headerlink" href="#resnet34" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnet34">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnet34</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNet34_Weights" title="torchvision.models.resnet.ResNet34_Weights"><span class="pre">ResNet34_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnet34"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnet34" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNet-34 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNet34_Weights" title="torchvision.models.ResNet34_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet34_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ResNet34_Weights" title="torchvision.models.ResNet34_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet34_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNet34_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNet34_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNet34_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNet34_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet34_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet34_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNet34_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet34_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>73.314</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.42</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>21797672</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>3.66</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>83.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet34_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.resnet50.html" class="btn btn-neutral float-right" title="resnet50" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.resnet18.html" class="btn btn-neutral" title="resnet18" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnet34</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnet50.html b/0.16/models/generated/torchvision.models.resnet50.html
new file mode 100644
index 00000000000..e7da239fd34
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnet50.html
@@ -0,0 +1,790 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnet50 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnet101" href="torchvision.models.resnet101.html" />
+    <link rel="prev" title="resnet34" href="torchvision.models.resnet34.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnet.html">ResNet</a> &gt;</li>
+        
+      <li>resnet50</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnet50.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet50">
+<h1>resnet50<a class="headerlink" href="#resnet50" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnet50">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnet50</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnet50"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnet50" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNet-50 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The bottleneck of TorchVision places the stride for downsampling to the second 3x3
+convolution while the original paper places it to the first 1x1 convolution.
+This variant improves the accuracy and is known as <a class="reference external" href="https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch">ResNet V1.5</a>.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNet50_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNet50_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNet50_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNet50_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNet50_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.13</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.862</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25557032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnet">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>97.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNet50_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>80.858</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.434</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25557032</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#issuecomment-1013906621">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>97.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNet50_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.resnet101.html" class="btn btn-neutral float-right" title="resnet101" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.resnet34.html" class="btn btn-neutral" title="resnet34" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnet50</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnext101_32x8d.html b/0.16/models/generated/torchvision.models.resnext101_32x8d.html
new file mode 100644
index 00000000000..049c92567ab
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnext101_32x8d.html
@@ -0,0 +1,785 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnext101_32x8d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnext101_64x4d" href="torchvision.models.resnext101_64x4d.html" />
+    <link rel="prev" title="resnext50_32x4d" href="torchvision.models.resnext50_32x4d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnext.html">ResNeXt</a> &gt;</li>
+        
+      <li>resnext101_32x8d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnext101_32x8d.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnext101-32x8d">
+<h1>resnext101_32x8d<a class="headerlink" href="#resnext101-32x8d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnext101_32x8d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnext101_32x8d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNeXt101_32X8D_Weights" title="torchvision.models.resnet.ResNeXt101_32X8D_Weights"><span class="pre">ResNeXt101_32X8D_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnext101_32x8d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnext101_32x8d" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNeXt-101 32x8d model from
+<a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNeXt101_32X8D_Weights" title="torchvision.models.ResNeXt101_32X8D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ResNeXt101_32X8D_Weights" title="torchvision.models.ResNeXt101_32X8D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNeXt101_32X8D_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNeXt101_32X8D_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNeXt101_32X8D_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNeXt101_32X8D_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNeXt101_32X8D_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>79.312</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.526</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88791336</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>16.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>339.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNeXt101_32X8D_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.834</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.228</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88791336</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>16.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>339.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_32X8D_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.resnext101_64x4d.html" class="btn btn-neutral float-right" title="resnext101_64x4d" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.resnext50_32x4d.html" class="btn btn-neutral" title="resnext50_32x4d" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnext101_32x8d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnext101_64x4d.html b/0.16/models/generated/torchvision.models.resnext101_64x4d.html
new file mode 100644
index 00000000000..1da793d34f9
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnext101_64x4d.html
@@ -0,0 +1,754 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnext101_64x4d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="ShuffleNet V2" href="../shufflenetv2.html" />
+    <link rel="prev" title="resnext101_32x8d" href="torchvision.models.resnext101_32x8d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnext.html">ResNeXt</a> &gt;</li>
+        
+      <li>resnext101_64x4d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnext101_64x4d.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnext101-64x4d">
+<h1>resnext101_64x4d<a class="headerlink" href="#resnext101-64x4d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnext101_64x4d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnext101_64x4d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNeXt101_64X4D_Weights" title="torchvision.models.resnet.ResNeXt101_64X4D_Weights"><span class="pre">ResNeXt101_64X4D_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnext101_64x4d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnext101_64x4d" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNeXt-101 64x4d model from
+<a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNeXt101_64X4D_Weights" title="torchvision.models.ResNeXt101_64X4D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ResNeXt101_64X4D_Weights" title="torchvision.models.ResNeXt101_64X4D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNeXt101_64X4D_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNeXt101_64X4D_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNeXt101_64X4D_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNeXt101_64X4D_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNeXt101_64X4D_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.246</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.454</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>83455272</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5935">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.46</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>319.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt101_64X4D_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../shufflenetv2.html" class="btn btn-neutral float-right" title="ShuffleNet V2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.resnext101_32x8d.html" class="btn btn-neutral" title="resnext101_32x8d" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnext101_64x4d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.resnext50_32x4d.html b/0.16/models/generated/torchvision.models.resnext50_32x4d.html
new file mode 100644
index 00000000000..163a1d87940
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.resnext50_32x4d.html
@@ -0,0 +1,785 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>resnext50_32x4d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="resnext101_32x8d" href="torchvision.models.resnext101_32x8d.html" />
+    <link rel="prev" title="ResNeXt" href="../resnext.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../resnext.html">ResNeXt</a> &gt;</li>
+        
+      <li>resnext50_32x4d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.resnext50_32x4d.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnext50-32x4d">
+<h1>resnext50_32x4d<a class="headerlink" href="#resnext50-32x4d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.resnext50_32x4d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">resnext50_32x4d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ResNeXt50_32X4D_Weights" title="torchvision.models.resnet.ResNeXt50_32X4D_Weights"><span class="pre">ResNeXt50_32X4D_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#resnext50_32x4d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.resnext50_32x4d" title="Permalink to this definition">¶</a></dt>
+<dd><p>ResNeXt-50 32x4d model from
+<a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ResNeXt50_32X4D_Weights" title="torchvision.models.ResNeXt50_32X4D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<code class="xref py py-class docutils literal notranslate"><span class="pre">ResNext50_32X4D_Weights</span></code> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ResNeXt50_32X4D_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ResNeXt50_32X4D_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#ResNeXt50_32X4D_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ResNeXt50_32X4D_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ResNeXt50_32X4D_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>77.618</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.698</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25028904</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#resnext">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.23</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>95.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ResNeXt50_32X4D_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.198</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.34</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>25028904</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.23</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>95.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ResNeXt50_32X4D_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.resnext101_32x8d.html" class="btn btn-neutral float-right" title="resnext101_32x8d" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../resnext.html" class="btn btn-neutral" title="ResNeXt" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">resnext50_32x4d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
new file mode 100644
index 00000000000..9e2e3cc36fd
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
@@ -0,0 +1,757 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>deeplabv3_mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="deeplabv3_resnet50" href="torchvision.models.segmentation.deeplabv3_resnet50.html" />
+    <link rel="prev" title="DeepLabV3" href="../deeplabv3.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../deeplabv3.html">DeepLabV3</a> &gt;</li>
+        
+      <li>deeplabv3_mobilenet_v3_large</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="deeplabv3-mobilenet-v3-large">
+<h1>deeplabv3_mobilenet_v3_large<a class="headerlink" href="#deeplabv3-mobilenet-v3-large" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.deeplabv3_mobilenet_v3_large">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">deeplabv3_mobilenet_v3_large</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.deeplabv3.DeepLabV3_MobileNet_V3_Large_Weights"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aux_loss</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Large_Weights"><span class="pre">MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">DeepLabV3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/deeplabv3.html#deeplabv3_mobilenet_v3_large"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.deeplabv3_mobilenet_v3_large" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a DeepLabV3 model with a MobileNetV3-Large backbone.</p>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1706.05587">Rethinking Atrous Convolution for Semantic Image Segmentation</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>aux_loss</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, it uses an auxiliary loss</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a>, optional) – The pretrained weights
+for the backbone</p></li>
+<li><p><strong>**kwargs</strong> – unused</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/deeplabv3.html#DeepLabV3_MobileNet_V3_Large_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.DeepLabV3_MobileNet_V3_Large_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_WITH_VOC_LABELS_V1'</span></code>.</p>
+<p><strong>DeepLabV3_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</strong>:</p>
+<p>
+        These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC
+        dataset.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>miou (on COCO-val2017-VOC-labels)</p></td>
+<td><p>60.3</p></td>
+</tr>
+<tr class="row-even"><td><p>pixel_acc (on COCO-val2017-VOC-labels)</p></td>
+<td><p>91.2</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, aeroplane, bicycle, … (18 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>11029328</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_mobilenet_v3_large">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>10.45</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>42.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">DeepLabV3_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[520]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.segmentation.deeplabv3_resnet50.html" class="btn btn-neutral float-right" title="deeplabv3_resnet50" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../deeplabv3.html" class="btn btn-neutral" title="DeepLabV3" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">deeplabv3_mobilenet_v3_large</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
new file mode 100644
index 00000000000..c7cccfbab5d
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
@@ -0,0 +1,761 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>deeplabv3_resnet101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="FCN" href="../fcn.html" />
+    <link rel="prev" title="deeplabv3_resnet50" href="torchvision.models.segmentation.deeplabv3_resnet50.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../deeplabv3.html">DeepLabV3</a> &gt;</li>
+        
+      <li>deeplabv3_resnet101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="deeplabv3-resnet101">
+<h1>deeplabv3_resnet101<a class="headerlink" href="#deeplabv3-resnet101" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.deeplabv3_resnet101">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">deeplabv3_resnet101</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_ResNet101_Weights" title="torchvision.models.segmentation.deeplabv3.DeepLabV3_ResNet101_Weights"><span class="pre">DeepLabV3_ResNet101_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aux_loss</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights" title="torchvision.models.resnet.ResNet101_Weights"><span class="pre">ResNet101_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet101_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">DeepLabV3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/deeplabv3.html#deeplabv3_resnet101"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.deeplabv3_resnet101" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a DeepLabV3 model with a ResNet-101 backbone.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1706.05587">Rethinking Atrous Convolution for Semantic Image Segmentation</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_ResNet101_Weights" title="torchvision.models.segmentation.DeepLabV3_ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_ResNet101_Weights" title="torchvision.models.segmentation.DeepLabV3_ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>aux_loss</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, it uses an auxiliary loss</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights" title="torchvision.models.ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights</span></code></a>, optional) – The pretrained weights for the
+backbone</p></li>
+<li><p><strong>**kwargs</strong> – unused</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.DeepLabV3_ResNet101_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">DeepLabV3_ResNet101_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/deeplabv3.html#DeepLabV3_ResNet101_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.DeepLabV3_ResNet101_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_WITH_VOC_LABELS_V1'</span></code>.</p>
+<p><strong>DeepLabV3_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</strong>:</p>
+<p>
+        These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC
+        dataset.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>miou (on COCO-val2017-VOC-labels)</p></td>
+<td><p>67.4</p></td>
+</tr>
+<tr class="row-even"><td><p>pixel_acc (on COCO-val2017-VOC-labels)</p></td>
+<td><p>92.4</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, aeroplane, bicycle, … (18 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>60996202</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet101">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>258.74</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>233.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[520]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../fcn.html" class="btn btn-neutral float-right" title="FCN" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.segmentation.deeplabv3_resnet50.html" class="btn btn-neutral" title="deeplabv3_resnet50" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">deeplabv3_resnet101</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
new file mode 100644
index 00000000000..24c8888bca5
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
@@ -0,0 +1,761 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>deeplabv3_resnet50 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="deeplabv3_resnet101" href="torchvision.models.segmentation.deeplabv3_resnet101.html" />
+    <link rel="prev" title="deeplabv3_mobilenet_v3_large" href="torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../deeplabv3.html">DeepLabV3</a> &gt;</li>
+        
+      <li>deeplabv3_resnet50</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="deeplabv3-resnet50">
+<h1>deeplabv3_resnet50<a class="headerlink" href="#deeplabv3-resnet50" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.deeplabv3_resnet50">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">deeplabv3_resnet50</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_ResNet50_Weights" title="torchvision.models.segmentation.deeplabv3.DeepLabV3_ResNet50_Weights"><span class="pre">DeepLabV3_ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aux_loss</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">DeepLabV3</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/deeplabv3.html#deeplabv3_resnet50"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.deeplabv3_resnet50" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a DeepLabV3 model with a ResNet-50 backbone.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1706.05587">Rethinking Atrous Convolution for Semantic Image Segmentation</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_ResNet50_Weights" title="torchvision.models.segmentation.DeepLabV3_ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.segmentation.DeepLabV3_ResNet50_Weights" title="torchvision.models.segmentation.DeepLabV3_ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background)</p></li>
+<li><p><strong>aux_loss</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, it uses an auxiliary loss</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The pretrained weights for the
+backbone</p></li>
+<li><p><strong>**kwargs</strong> – unused</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.DeepLabV3_ResNet50_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">DeepLabV3_ResNet50_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/deeplabv3.html#DeepLabV3_ResNet50_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.DeepLabV3_ResNet50_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_WITH_VOC_LABELS_V1'</span></code>.</p>
+<p><strong>DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</strong>:</p>
+<p>
+        These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC
+        dataset.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>miou (on COCO-val2017-VOC-labels)</p></td>
+<td><p>66.4</p></td>
+</tr>
+<tr class="row-even"><td><p>pixel_acc (on COCO-val2017-VOC-labels)</p></td>
+<td><p>92.4</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, aeroplane, bicycle, … (18 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>42004074</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet50">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>178.72</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>160.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[520]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.segmentation.deeplabv3_resnet101.html" class="btn btn-neutral float-right" title="deeplabv3_resnet101" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html" class="btn btn-neutral" title="deeplabv3_mobilenet_v3_large" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">deeplabv3_resnet50</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
new file mode 100644
index 00000000000..9d7f23bc5ea
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
@@ -0,0 +1,763 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>fcn_resnet101 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="LRASPP" href="../lraspp.html" />
+    <link rel="prev" title="fcn_resnet50" href="torchvision.models.segmentation.fcn_resnet50.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../fcn.html">FCN</a> &gt;</li>
+        
+      <li>fcn_resnet101</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.segmentation.fcn_resnet101.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fcn-resnet101">
+<h1>fcn_resnet101<a class="headerlink" href="#fcn-resnet101" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.fcn_resnet101">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">fcn_resnet101</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.segmentation.FCN_ResNet101_Weights" title="torchvision.models.segmentation.fcn.FCN_ResNet101_Weights"><span class="pre">FCN_ResNet101_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aux_loss</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights" title="torchvision.models.resnet.ResNet101_Weights"><span class="pre">ResNet101_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet101_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">FCN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/fcn.html#fcn_resnet101"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.fcn_resnet101" title="Permalink to this definition">¶</a></dt>
+<dd><p>Fully-Convolutional Network model with a ResNet-101 backbone from the <a class="reference external" href="https://arxiv.org/abs/1411.4038">Fully Convolutional
+Networks for Semantic Segmentation</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.segmentation.FCN_ResNet101_Weights" title="torchvision.models.segmentation.FCN_ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.segmentation.FCN_ResNet101_Weights" title="torchvision.models.segmentation.FCN_ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background).</p></li>
+<li><p><strong>aux_loss</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, it uses an auxiliary loss.</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet101.html#torchvision.models.ResNet101_Weights" title="torchvision.models.ResNet101_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet101_Weights</span></code></a>, optional) – The pretrained
+weights for the backbone.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.segmentation.fcn.FCN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.FCN_ResNet101_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">FCN_ResNet101_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/fcn.html#FCN_ResNet101_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.FCN_ResNet101_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_WITH_VOC_LABELS_V1'</span></code>.</p>
+<p><strong>FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1</strong>:</p>
+<p>
+        These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC
+        dataset.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>miou (on COCO-val2017-VOC-labels)</p></td>
+<td><p>63.7</p></td>
+</tr>
+<tr class="row-even"><td><p>pixel_acc (on COCO-val2017-VOC-labels)</p></td>
+<td><p>91.9</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, aeroplane, bicycle, … (18 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>54314346</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#deeplabv3_resnet101">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>232.74</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>207.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">FCN_ResNet101_Weights.COCO_WITH_VOC_LABELS_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[520]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../lraspp.html" class="btn btn-neutral float-right" title="LRASPP" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.segmentation.fcn_resnet50.html" class="btn btn-neutral" title="fcn_resnet50" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">fcn_resnet101</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
new file mode 100644
index 00000000000..887a8e212c2
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
@@ -0,0 +1,767 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>fcn_resnet50 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="fcn_resnet101" href="torchvision.models.segmentation.fcn_resnet101.html" />
+    <link rel="prev" title="FCN" href="../fcn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../fcn.html">FCN</a> &gt;</li>
+        
+      <li>fcn_resnet50</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.segmentation.fcn_resnet50.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="fcn-resnet50">
+<h1>fcn_resnet50<a class="headerlink" href="#fcn-resnet50" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.fcn_resnet50">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">fcn_resnet50</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.segmentation.FCN_ResNet50_Weights" title="torchvision.models.segmentation.fcn.FCN_ResNet50_Weights"><span class="pre">FCN_ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aux_loss</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.resnet.ResNet50_Weights"><span class="pre">ResNet50_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">ResNet50_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">FCN</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/fcn.html#fcn_resnet50"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.fcn_resnet50" title="Permalink to this definition">¶</a></dt>
+<dd><p>Fully-Convolutional Network model with a ResNet-50 backbone from the <a class="reference external" href="https://arxiv.org/abs/1411.4038">Fully Convolutional
+Networks for Semantic Segmentation</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.segmentation.FCN_ResNet50_Weights" title="torchvision.models.segmentation.FCN_ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.segmentation.FCN_ResNet50_Weights" title="torchvision.models.segmentation.FCN_ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background).</p></li>
+<li><p><strong>aux_loss</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, it uses an auxiliary loss.</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.resnet50.html#torchvision.models.ResNet50_Weights" title="torchvision.models.ResNet50_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ResNet50_Weights</span></code></a>, optional) – The pretrained
+weights for the backbone.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.segmentation.fcn.FCN</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/fcn.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.FCN_ResNet50_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">FCN_ResNet50_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/fcn.html#FCN_ResNet50_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.FCN_ResNet50_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_WITH_VOC_LABELS_V1'</span></code>.</p>
+<p><strong>FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1</strong>:</p>
+<p>
+        These weights were trained on a subset of COCO, using only the 20 categories that are present in the Pascal VOC
+        dataset.
+    
+Also available as <code class="docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>miou (on COCO-val2017-VOC-labels)</p></td>
+<td><p>60.5</p></td>
+</tr>
+<tr class="row-even"><td><p>pixel_acc (on COCO-val2017-VOC-labels)</p></td>
+<td><p>91.4</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>__background__, aeroplane, bicycle, … (18 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>35322218</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#fcn_resnet50">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>152.72</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>135.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">FCN_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[520]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+<p>Examples using <code class="docutils literal notranslate"><span class="pre">fcn_resnet50</span></code>:</p>
+<div class="sphx-glr-thumbnails"><div class="sphx-glr-thumbcontainer" tooltip="This example illustrates some of the utilities that torchvision offers for visualizing images, ..."><img alt="" src="../../_images/sphx_glr_plot_visualization_utils_thumb.png" />
+<p><a class="reference internal" href="../../auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">Visualization utilities</span></a></p>
+  <div class="sphx-glr-thumbnail-title">Visualization utilities</div>
+</div></div></dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.segmentation.fcn_resnet101.html" class="btn btn-neutral float-right" title="fcn_resnet101" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../fcn.html" class="btn btn-neutral" title="FCN" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">fcn_resnet50</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
new file mode 100644
index 00000000000..5f4744e58ca
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
@@ -0,0 +1,763 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>lraspp_mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Faster R-CNN" href="../faster_rcnn.html" />
+    <link rel="prev" title="LRASPP" href="../lraspp.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../lraspp.html">LRASPP</a> &gt;</li>
+        
+      <li>lraspp_mobilenet_v3_large</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lraspp-mobilenet-v3-large">
+<h1>lraspp_mobilenet_v3_large<a class="headerlink" href="#lraspp-mobilenet-v3-large" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.lraspp_mobilenet_v3_large">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">lraspp_mobilenet_v3_large</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.lraspp.LRASPP_MobileNet_V3_Large_Weights"><span class="pre">LRASPP_MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><span class="pre">int</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights_backbone</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.mobilenetv3.MobileNet_V3_Large_Weights"><span class="pre">MobileNet_V3_Large_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">MobileNet_V3_Large_Weights.IMAGENET1K_V1</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">LRASPP</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/lraspp.html#lraspp_mobilenet_v3_large"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.lraspp_mobilenet_v3_large" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a Lite R-ASPP Network model with a MobileNetV3-Large backbone from
+<a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights" title="torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>num_classes</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><em>int</em></a><em>, </em><em>optional</em>) – number of output classes of the model (including the background).</p></li>
+<li><p><strong>aux_loss</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, it uses an auxiliary loss.</p></li>
+<li><p><strong>weights_backbone</strong> (<a class="reference internal" href="torchvision.models.mobilenet_v3_large.html#torchvision.models.MobileNet_V3_Large_Weights" title="torchvision.models.MobileNet_V3_Large_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MobileNet_V3_Large_Weights</span></code></a>, optional) – The pretrained
+weights for the backbone.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.segmentation.LRASPP</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/lraspp.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.segmentation.</span></span><span class="sig-name descname"><span class="pre">LRASPP_MobileNet_V3_Large_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/segmentation/lraspp.html#LRASPP_MobileNet_V3_Large_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.segmentation.LRASPP_MobileNet_V3_Large_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='COCO_WITH_VOC_LABELS_V1'</span></code>.</p>
+<p><strong>LRASPP_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1</strong>:</p>
+<p>
+                These weights were trained on a subset of COCO, using only the 20 categories that are present in the
+                Pascal VOC dataset.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>miou (on COCO-val2017-VOC-labels)</p></td>
+<td><p>57.9</p></td>
+</tr>
+<tr class="row-even"><td><p>pixel_acc (on COCO-val2017-VOC-labels)</p></td>
+<td><p>91.2</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>3221538</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>__background__, aeroplane, bicycle, … (18 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation#lraspp_mobilenet_v3_large">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>2.09</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>12.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">LRASPP_MobileNet_V3_Large_Weights.COCO_WITH_VOC_LABELS_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[520]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../faster_rcnn.html" class="btn btn-neutral float-right" title="Faster R-CNN" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../lraspp.html" class="btn btn-neutral" title="LRASPP" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">lraspp_mobilenet_v3_large</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
new file mode 100644
index 00000000000..56ee4b0ad5f
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x0_5 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="shufflenet_v2_x1_0" href="torchvision.models.shufflenet_v2_x1_0.html" />
+    <link rel="prev" title="ShuffleNet V2" href="../shufflenetv2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2.html">ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x0_5</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.shufflenet_v2_x0_5.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x0-5">
+<h1>shufflenet_v2_x0_5<a class="headerlink" href="#shufflenet-v2-x0-5" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.shufflenet_v2_x0_5">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x0_5</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X0_5_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X0_5_Weights"><span class="pre">ShuffleNet_V2_X0_5_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#shufflenet_v2_x0_5"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.shufflenet_v2_x0_5" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 architecture with 0.5x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X0_5_Weights" title="torchvision.models.ShuffleNet_V2_X0_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X0_5_Weights" title="torchvision.models.ShuffleNet_V2_X0_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.shufflenetv2.ShuffleNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ShuffleNet_V2_X0_5_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X0_5_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X0_5_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ShuffleNet_V2_X0_5_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch to reproduce closely the results of the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>60.552</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>81.746</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>1366792</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.04</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>5.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X0_5_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.shufflenet_v2_x1_0.html" class="btn btn-neutral float-right" title="shufflenet_v2_x1_0" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../shufflenetv2.html" class="btn btn-neutral" title="ShuffleNet V2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x0_5</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
new file mode 100644
index 00000000000..6f4a0a21cf2
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x1_0 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="shufflenet_v2_x1_5" href="torchvision.models.shufflenet_v2_x1_5.html" />
+    <link rel="prev" title="shufflenet_v2_x0_5" href="torchvision.models.shufflenet_v2_x0_5.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2.html">ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x1_0</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.shufflenet_v2_x1_0.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x1-0">
+<h1>shufflenet_v2_x1_0<a class="headerlink" href="#shufflenet-v2-x1-0" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.shufflenet_v2_x1_0">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x1_0</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X1_0_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X1_0_Weights"><span class="pre">ShuffleNet_V2_X1_0_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#shufflenet_v2_x1_0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.shufflenet_v2_x1_0" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 architecture with 1.0x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X1_0_Weights" title="torchvision.models.ShuffleNet_V2_X1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X1_0_Weights" title="torchvision.models.ShuffleNet_V2_X1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.shufflenetv2.ShuffleNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ShuffleNet_V2_X1_0_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X1_0_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X1_0_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ShuffleNet_V2_X1_0_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch to reproduce closely the results of the paper.
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.362</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>88.316</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/ericsun99/Shufflenet-v2-Pytorch">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>2278604</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.14</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>8.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_0_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.shufflenet_v2_x1_5.html" class="btn btn-neutral float-right" title="shufflenet_v2_x1_5" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.shufflenet_v2_x0_5.html" class="btn btn-neutral" title="shufflenet_v2_x0_5" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x1_0</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
new file mode 100644
index 00000000000..d17234956a9
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
@@ -0,0 +1,754 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x1_5 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="shufflenet_v2_x2_0" href="torchvision.models.shufflenet_v2_x2_0.html" />
+    <link rel="prev" title="shufflenet_v2_x1_0" href="torchvision.models.shufflenet_v2_x1_0.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2.html">ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x1_5</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.shufflenet_v2_x1_5.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x1-5">
+<h1>shufflenet_v2_x1_5<a class="headerlink" href="#shufflenet-v2-x1-5" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.shufflenet_v2_x1_5">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x1_5</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X1_5_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X1_5_Weights"><span class="pre">ShuffleNet_V2_X1_5_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#shufflenet_v2_x1_5"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.shufflenet_v2_x1_5" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 architecture with 1.5x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X1_5_Weights" title="torchvision.models.ShuffleNet_V2_X1_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X1_5_Weights" title="torchvision.models.ShuffleNet_V2_X1_5_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.shufflenetv2.ShuffleNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ShuffleNet_V2_X1_5_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X1_5_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X1_5_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ShuffleNet_V2_X1_5_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>72.996</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.086</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>3503624</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.30</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>13.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X1_5_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.shufflenet_v2_x2_0.html" class="btn btn-neutral float-right" title="shufflenet_v2_x2_0" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.shufflenet_v2_x1_0.html" class="btn btn-neutral" title="shufflenet_v2_x1_0" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x1_5</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
new file mode 100644
index 00000000000..542a6679cac
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
@@ -0,0 +1,754 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>shufflenet_v2_x2_0 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="SqueezeNet" href="../squeezenet.html" />
+    <link rel="prev" title="shufflenet_v2_x1_5" href="torchvision.models.shufflenet_v2_x1_5.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../shufflenetv2.html">ShuffleNet V2</a> &gt;</li>
+        
+      <li>shufflenet_v2_x2_0</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.shufflenet_v2_x2_0.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2-x2-0">
+<h1>shufflenet_v2_x2_0<a class="headerlink" href="#shufflenet-v2-x2-0" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.shufflenet_v2_x2_0">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">shufflenet_v2_x2_0</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X2_0_Weights" title="torchvision.models.shufflenetv2.ShuffleNet_V2_X2_0_Weights"><span class="pre">ShuffleNet_V2_X2_0_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ShuffleNetV2</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#shufflenet_v2_x2_0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.shufflenet_v2_x2_0" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a ShuffleNetV2 architecture with 2.0x output channels, as described in
+<a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X2_0_Weights" title="torchvision.models.ShuffleNet_V2_X2_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.ShuffleNet_V2_X2_0_Weights" title="torchvision.models.ShuffleNet_V2_X2_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.shufflenetv2.ShuffleNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ShuffleNet_V2_X2_0_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ShuffleNet_V2_X2_0_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/shufflenetv2.html#ShuffleNet_V2_X2_0_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ShuffleNet_V2_X2_0_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.23</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.006</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5906">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>7393996</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.58</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>28.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ShuffleNet_V2_X2_0_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../squeezenet.html" class="btn btn-neutral float-right" title="SqueezeNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.shufflenet_v2_x1_5.html" class="btn btn-neutral" title="shufflenet_v2_x1_5" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">shufflenet_v2_x2_0</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.squeezenet1_0.html b/0.16/models/generated/torchvision.models.squeezenet1_0.html
new file mode 100644
index 00000000000..2c76b3d6dda
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.squeezenet1_0.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>squeezenet1_0 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="squeezenet1_1" href="torchvision.models.squeezenet1_1.html" />
+    <link rel="prev" title="SqueezeNet" href="../squeezenet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../squeezenet.html">SqueezeNet</a> &gt;</li>
+        
+      <li>squeezenet1_0</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.squeezenet1_0.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="squeezenet1-0">
+<h1>squeezenet1_0<a class="headerlink" href="#squeezenet1-0" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.squeezenet1_0">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">squeezenet1_0</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.SqueezeNet1_0_Weights" title="torchvision.models.squeezenet.SqueezeNet1_0_Weights"><span class="pre">SqueezeNet1_0_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SqueezeNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/squeezenet.html#squeezenet1_0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.squeezenet1_0" title="Permalink to this definition">¶</a></dt>
+<dd><p>SqueezeNet model architecture from the <a class="reference external" href="https://arxiv.org/abs/1602.07360">SqueezeNet: AlexNet-level
+accuracy with 50x fewer parameters and &lt;0.5MB model size</a> paper.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.SqueezeNet1_0_Weights" title="torchvision.models.SqueezeNet1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.SqueezeNet1_0_Weights" title="torchvision.models.SqueezeNet1_0_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.squeezenet.SqueezeNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/squeezenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.SqueezeNet1_0_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">SqueezeNet1_0_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/squeezenet.html#SqueezeNet1_0_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.SqueezeNet1_0_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>SqueezeNet1_0_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>58.092</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>80.42</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/49#issuecomment-277560717">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=21, width=21</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>1248424</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.82</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>4.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">SqueezeNet1_0_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.squeezenet1_1.html" class="btn btn-neutral float-right" title="squeezenet1_1" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../squeezenet.html" class="btn btn-neutral" title="SqueezeNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">squeezenet1_0</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.squeezenet1_1.html b/0.16/models/generated/torchvision.models.squeezenet1_1.html
new file mode 100644
index 00000000000..638678b463e
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.squeezenet1_1.html
@@ -0,0 +1,753 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>squeezenet1_1 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="SwinTransformer" href="../swin_transformer.html" />
+    <link rel="prev" title="squeezenet1_0" href="torchvision.models.squeezenet1_0.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../squeezenet.html">SqueezeNet</a> &gt;</li>
+        
+      <li>squeezenet1_1</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.squeezenet1_1.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="squeezenet1-1">
+<h1>squeezenet1_1<a class="headerlink" href="#squeezenet1-1" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.squeezenet1_1">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">squeezenet1_1</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.SqueezeNet1_1_Weights" title="torchvision.models.squeezenet.SqueezeNet1_1_Weights"><span class="pre">SqueezeNet1_1_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SqueezeNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/squeezenet.html#squeezenet1_1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.squeezenet1_1" title="Permalink to this definition">¶</a></dt>
+<dd><p>SqueezeNet 1.1 model from the <a class="reference external" href="https://github.com/DeepScale/SqueezeNet/tree/master/SqueezeNet_v1.1">official SqueezeNet repo</a>.</p>
+<p>SqueezeNet 1.1 has 2.4x less computation and slightly fewer parameters
+than SqueezeNet 1.0, without sacrificing accuracy.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.SqueezeNet1_1_Weights" title="torchvision.models.SqueezeNet1_1_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.SqueezeNet1_1_Weights" title="torchvision.models.SqueezeNet1_1_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.squeezenet.SqueezeNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/squeezenet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.SqueezeNet1_1_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">SqueezeNet1_1_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/squeezenet.html#SqueezeNet1_1_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.SqueezeNet1_1_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>SqueezeNet1_1_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>58.178</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>80.624</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/49#issuecomment-277560717">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=17, width=17</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>1235496</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>0.35</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>4.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">SqueezeNet1_1_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../swin_transformer.html" class="btn btn-neutral float-right" title="SwinTransformer" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.squeezenet1_0.html" class="btn btn-neutral" title="squeezenet1_0" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">squeezenet1_1</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.swin_b.html b/0.16/models/generated/torchvision.models.swin_b.html
new file mode 100644
index 00000000000..6e02233596f
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.swin_b.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin_b &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="swin_v2_t" href="torchvision.models.swin_v2_t.html" />
+    <link rel="prev" title="swin_s" href="torchvision.models.swin_s.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../swin_transformer.html">SwinTransformer</a> &gt;</li>
+        
+      <li>swin_b</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.swin_b.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin-b">
+<h1>swin_b<a class="headerlink" href="#swin-b" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.swin_b">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">swin_b</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Swin_B_Weights" title="torchvision.models.swin_transformer.Swin_B_Weights"><span class="pre">Swin_B_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#swin_b"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.swin_b" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_base architecture from
+<a class="reference external" href="https://arxiv.org/abs/2103.14030">Swin Transformer: Hierarchical Vision Transformer using Shifted Windows</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Swin_B_Weights" title="torchvision.models.Swin_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_B_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Swin_B_Weights" title="torchvision.models.Swin_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_B_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Swin_B_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Swin_B_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#Swin_B_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Swin_B_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin_B_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin_B_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Swin_B_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a similar training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin_B_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.582</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.64</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>87768224</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.43</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>335.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin_B_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[238]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.swin_v2_t.html" class="btn btn-neutral float-right" title="swin_v2_t" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.swin_s.html" class="btn btn-neutral" title="swin_s" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin_b</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.swin_s.html b/0.16/models/generated/torchvision.models.swin_s.html
new file mode 100644
index 00000000000..500454c5310
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.swin_s.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin_s &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="swin_b" href="torchvision.models.swin_b.html" />
+    <link rel="prev" title="swin_t" href="torchvision.models.swin_t.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../swin_transformer.html">SwinTransformer</a> &gt;</li>
+        
+      <li>swin_s</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.swin_s.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin-s">
+<h1>swin_s<a class="headerlink" href="#swin-s" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.swin_s">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">swin_s</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Swin_S_Weights" title="torchvision.models.swin_transformer.Swin_S_Weights"><span class="pre">Swin_S_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#swin_s"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.swin_s" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_small architecture from
+<a class="reference external" href="https://arxiv.org/abs/2103.14030">Swin Transformer: Hierarchical Vision Transformer using Shifted Windows</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Swin_S_Weights" title="torchvision.models.Swin_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_S_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Swin_S_Weights" title="torchvision.models.Swin_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_S_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Swin_S_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Swin_S_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#Swin_S_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Swin_S_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin_S_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin_S_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Swin_S_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a similar training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin_S_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.196</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.36</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>49606258</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>8.74</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>189.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin_S_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[246]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.swin_b.html" class="btn btn-neutral float-right" title="swin_b" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.swin_t.html" class="btn btn-neutral" title="swin_t" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin_s</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.swin_t.html b/0.16/models/generated/torchvision.models.swin_t.html
new file mode 100644
index 00000000000..b47607277f5
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.swin_t.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin_t &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="swin_s" href="torchvision.models.swin_s.html" />
+    <link rel="prev" title="SwinTransformer" href="../swin_transformer.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../swin_transformer.html">SwinTransformer</a> &gt;</li>
+        
+      <li>swin_t</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.swin_t.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin-t">
+<h1>swin_t<a class="headerlink" href="#swin-t" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.swin_t">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">swin_t</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Swin_T_Weights" title="torchvision.models.swin_transformer.Swin_T_Weights"><span class="pre">Swin_T_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#swin_t"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.swin_t" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_tiny architecture from
+<a class="reference external" href="https://arxiv.org/abs/2103.14030">Swin Transformer: Hierarchical Vision Transformer using Shifted Windows</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Swin_T_Weights" title="torchvision.models.Swin_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_T_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Swin_T_Weights" title="torchvision.models.Swin_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_T_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Swin_T_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Swin_T_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#Swin_T_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Swin_T_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin_T_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin_T_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Swin_T_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a similar training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin_T_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.474</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.776</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>28288354</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.49</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>108.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin_T_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.swin_s.html" class="btn btn-neutral float-right" title="swin_s" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../swin_transformer.html" class="btn btn-neutral" title="SwinTransformer" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin_t</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.swin_v2_b.html b/0.16/models/generated/torchvision.models.swin_v2_b.html
new file mode 100644
index 00000000000..75525d2e552
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.swin_v2_b.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin_v2_b &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="VGG" href="../vgg.html" />
+    <link rel="prev" title="swin_v2_s" href="torchvision.models.swin_v2_s.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../swin_transformer.html">SwinTransformer</a> &gt;</li>
+        
+      <li>swin_v2_b</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.swin_v2_b.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin-v2-b">
+<h1>swin_v2_b<a class="headerlink" href="#swin-v2-b" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.swin_v2_b">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">swin_v2_b</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Swin_V2_B_Weights" title="torchvision.models.swin_transformer.Swin_V2_B_Weights"><span class="pre">Swin_V2_B_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#swin_v2_b"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.swin_v2_b" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_v2_base architecture from
+<a class="reference external" href="https://arxiv.org/abs/2111.09883">Swin Transformer V2: Scaling Up Capacity and Resolution</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Swin_V2_B_Weights" title="torchvision.models.Swin_V2_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_B_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Swin_V2_B_Weights" title="torchvision.models.Swin_V2_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_B_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Swin_V2_B_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Swin_V2_B_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#Swin_V2_B_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Swin_V2_B_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin_V2_B_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin_V2_B_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Swin_V2_B_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a similar training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin_V2_B_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>84.112</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.864</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>87930848</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=256, width=256</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>20.32</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>336.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin_V2_B_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[272]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[256]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../vgg.html" class="btn btn-neutral float-right" title="VGG" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.swin_v2_s.html" class="btn btn-neutral" title="swin_v2_s" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin_v2_b</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.swin_v2_s.html b/0.16/models/generated/torchvision.models.swin_v2_s.html
new file mode 100644
index 00000000000..6851291c2c7
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.swin_v2_s.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin_v2_s &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="swin_v2_b" href="torchvision.models.swin_v2_b.html" />
+    <link rel="prev" title="swin_v2_t" href="torchvision.models.swin_v2_t.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../swin_transformer.html">SwinTransformer</a> &gt;</li>
+        
+      <li>swin_v2_s</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.swin_v2_s.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin-v2-s">
+<h1>swin_v2_s<a class="headerlink" href="#swin-v2-s" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.swin_v2_s">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">swin_v2_s</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Swin_V2_S_Weights" title="torchvision.models.swin_transformer.Swin_V2_S_Weights"><span class="pre">Swin_V2_S_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#swin_v2_s"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.swin_v2_s" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_v2_small architecture from
+<a class="reference external" href="https://arxiv.org/abs/2111.09883">Swin Transformer V2: Scaling Up Capacity and Resolution</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Swin_V2_S_Weights" title="torchvision.models.Swin_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_S_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Swin_V2_S_Weights" title="torchvision.models.Swin_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_S_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Swin_V2_S_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Swin_V2_S_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#Swin_V2_S_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Swin_V2_S_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin_V2_S_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin_V2_S_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Swin_V2_S_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a similar training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin_V2_S_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>83.712</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.816</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>49737442</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=256, width=256</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>11.55</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>190.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin_V2_S_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[260]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[256]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.swin_v2_b.html" class="btn btn-neutral float-right" title="swin_v2_b" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.swin_v2_t.html" class="btn btn-neutral" title="swin_v2_t" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin_v2_s</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.swin_v2_t.html b/0.16/models/generated/torchvision.models.swin_v2_t.html
new file mode 100644
index 00000000000..fae33e8bc46
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.swin_v2_t.html
@@ -0,0 +1,752 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin_v2_t &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="swin_v2_s" href="torchvision.models.swin_v2_s.html" />
+    <link rel="prev" title="swin_b" href="torchvision.models.swin_b.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../swin_transformer.html">SwinTransformer</a> &gt;</li>
+        
+      <li>swin_v2_t</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.swin_v2_t.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin-v2-t">
+<h1>swin_v2_t<a class="headerlink" href="#swin-v2-t" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.swin_v2_t">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">swin_v2_t</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Swin_V2_T_Weights" title="torchvision.models.swin_transformer.Swin_V2_T_Weights"><span class="pre">Swin_V2_T_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#swin_v2_t"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.swin_v2_t" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_v2_tiny architecture from
+<a class="reference external" href="https://arxiv.org/abs/2111.09883">Swin Transformer V2: Scaling Up Capacity and Resolution</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Swin_V2_T_Weights" title="torchvision.models.Swin_V2_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_T_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Swin_V2_T_Weights" title="torchvision.models.Swin_V2_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin_V2_T_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Swin_V2_T_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Swin_V2_T_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/swin_transformer.html#Swin_V2_T_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Swin_V2_T_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin_V2_T_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin_V2_T_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Swin_V2_T_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a similar training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin_V2_T_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.072</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.132</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>28351570</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=256, width=256</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#swintransformer-v2">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>5.94</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>108.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin_V2_T_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[260]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[256]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.swin_v2_s.html" class="btn btn-neutral float-right" title="swin_v2_s" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.swin_b.html" class="btn btn-neutral" title="swin_b" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin_v2_t</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg11.html b/0.16/models/generated/torchvision.models.vgg11.html
new file mode 100644
index 00000000000..e6624c19add
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg11.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg11 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vgg11_bn" href="torchvision.models.vgg11_bn.html" />
+    <link rel="prev" title="VGG" href="../vgg.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg11</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg11.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg11">
+<h1>vgg11<a class="headerlink" href="#vgg11" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg11">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg11</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG11_Weights" title="torchvision.models.vgg.VGG11_Weights"><span class="pre">VGG11_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg11"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg11" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-11 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG11_Weights" title="torchvision.models.VGG11_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG11_Weights" title="torchvision.models.VGG11_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG11_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG11_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG11_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG11_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG11_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG11_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG11_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG11_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.02</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>88.628</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>132863336</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>7.61</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>506.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG11_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vgg11_bn.html" class="btn btn-neutral float-right" title="vgg11_bn" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../vgg.html" class="btn btn-neutral" title="VGG" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg11</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg11_bn.html b/0.16/models/generated/torchvision.models.vgg11_bn.html
new file mode 100644
index 00000000000..82b0ee1d9df
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg11_bn.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg11_bn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vgg13" href="torchvision.models.vgg13.html" />
+    <link rel="prev" title="vgg11" href="torchvision.models.vgg11.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg11_bn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg11_bn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg11-bn">
+<h1>vgg11_bn<a class="headerlink" href="#vgg11-bn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg11_bn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg11_bn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG11_BN_Weights" title="torchvision.models.vgg.VGG11_BN_Weights"><span class="pre">VGG11_BN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg11_bn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg11_bn" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-11-BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG11_BN_Weights" title="torchvision.models.VGG11_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_BN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG11_BN_Weights" title="torchvision.models.VGG11_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG11_BN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG11_BN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG11_BN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG11_BN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG11_BN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG11_BN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG11_BN_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG11_BN_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG11_BN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>70.37</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>89.81</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>132868840</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>7.61</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>506.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG11_BN_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vgg13.html" class="btn btn-neutral float-right" title="vgg13" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vgg11.html" class="btn btn-neutral" title="vgg11" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg11_bn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg13.html b/0.16/models/generated/torchvision.models.vgg13.html
new file mode 100644
index 00000000000..f7f765a07d0
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg13.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg13 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vgg13_bn" href="torchvision.models.vgg13_bn.html" />
+    <link rel="prev" title="vgg11_bn" href="torchvision.models.vgg11_bn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg13</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg13.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg13">
+<h1>vgg13<a class="headerlink" href="#vgg13" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg13">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg13</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG13_Weights" title="torchvision.models.vgg.VGG13_Weights"><span class="pre">VGG13_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg13"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg13" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-13 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG13_Weights" title="torchvision.models.VGG13_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG13_Weights" title="torchvision.models.VGG13_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG13_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG13_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG13_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG13_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG13_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG13_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG13_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG13_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>69.928</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>89.246</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>133047848</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>11.31</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>507.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG13_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vgg13_bn.html" class="btn btn-neutral float-right" title="vgg13_bn" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vgg11_bn.html" class="btn btn-neutral" title="vgg11_bn" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg13</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg13_bn.html b/0.16/models/generated/torchvision.models.vgg13_bn.html
new file mode 100644
index 00000000000..ec53c8e9d35
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg13_bn.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg13_bn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vgg16" href="torchvision.models.vgg16.html" />
+    <link rel="prev" title="vgg13" href="torchvision.models.vgg13.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg13_bn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg13_bn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg13-bn">
+<h1>vgg13_bn<a class="headerlink" href="#vgg13-bn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg13_bn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg13_bn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG13_BN_Weights" title="torchvision.models.vgg.VGG13_BN_Weights"><span class="pre">VGG13_BN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg13_bn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg13_bn" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-13-BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG13_BN_Weights" title="torchvision.models.VGG13_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_BN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG13_BN_Weights" title="torchvision.models.VGG13_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG13_BN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG13_BN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG13_BN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG13_BN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG13_BN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG13_BN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG13_BN_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG13_BN_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG13_BN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>71.586</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.374</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>133053736</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>11.31</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>507.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG13_BN_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vgg16.html" class="btn btn-neutral float-right" title="vgg16" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vgg13.html" class="btn btn-neutral" title="vgg13" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg13_bn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg16.html b/0.16/models/generated/torchvision.models.vgg16.html
new file mode 100644
index 00000000000..6128953812a
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg16.html
@@ -0,0 +1,785 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg16 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vgg16_bn" href="torchvision.models.vgg16_bn.html" />
+    <link rel="prev" title="vgg13_bn" href="torchvision.models.vgg13_bn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg16</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg16.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg16">
+<h1>vgg16<a class="headerlink" href="#vgg16" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg16">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg16</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG16_Weights" title="torchvision.models.vgg.VGG16_Weights"><span class="pre">VGG16_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg16"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg16" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-16 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG16_Weights" title="torchvision.models.VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG16_Weights" title="torchvision.models.VGG16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG16_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG16_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG16_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG16_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG16_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG16_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG16_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG16_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>71.592</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.382</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>138357544</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.47</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>527.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG16_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>VGG16_Weights.IMAGENET1K_FEATURES</strong>:</p>
+<p>
+                These weights can’t be used for classification because they are missing values in the <cite>classifier</cite>
+                module. Only the <cite>features</cite> module has valid values and can be used for feature extraction. The weights
+                were trained using the original input standardization method as described in the paper.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>nan</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>nan</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>None</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/amdegroot/ssd.pytorch#training-ssd">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>138357544</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.47</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>527.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG16_Weights.IMAGENET1K_FEATURES.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.48235,</span> <span class="pre">0.45882,</span> <span class="pre">0.40784]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.00392156862745098,</span> <span class="pre">0.00392156862745098,</span> <span class="pre">0.00392156862745098]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vgg16_bn.html" class="btn btn-neutral float-right" title="vgg16_bn" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vgg13_bn.html" class="btn btn-neutral" title="vgg13_bn" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg16</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg16_bn.html b/0.16/models/generated/torchvision.models.vgg16_bn.html
new file mode 100644
index 00000000000..fc2034ae6d4
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg16_bn.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg16_bn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vgg19" href="torchvision.models.vgg19.html" />
+    <link rel="prev" title="vgg16" href="torchvision.models.vgg16.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg16_bn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg16_bn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg16-bn">
+<h1>vgg16_bn<a class="headerlink" href="#vgg16-bn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg16_bn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg16_bn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG16_BN_Weights" title="torchvision.models.vgg.VGG16_BN_Weights"><span class="pre">VGG16_BN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg16_bn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg16_bn" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-16-BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG16_BN_Weights" title="torchvision.models.VGG16_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_BN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG16_BN_Weights" title="torchvision.models.VGG16_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG16_BN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG16_BN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG16_BN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG16_BN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG16_BN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG16_BN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG16_BN_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG16_BN_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG16_BN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>73.36</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.516</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>138365992</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.47</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>527.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG16_BN_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vgg19.html" class="btn btn-neutral float-right" title="vgg19" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vgg16.html" class="btn btn-neutral" title="vgg16" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg16_bn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg19.html b/0.16/models/generated/torchvision.models.vgg19.html
new file mode 100644
index 00000000000..a5d786d6af3
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg19.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg19 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vgg19_bn" href="torchvision.models.vgg19_bn.html" />
+    <link rel="prev" title="vgg16_bn" href="torchvision.models.vgg16_bn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg19</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg19.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg19">
+<h1>vgg19<a class="headerlink" href="#vgg19" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg19">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg19</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG19_Weights" title="torchvision.models.vgg.VGG19_Weights"><span class="pre">VGG19_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg19"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg19" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-19 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG19_Weights" title="torchvision.models.VGG19_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG19_Weights" title="torchvision.models.VGG19_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG19_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG19_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG19_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG19_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG19_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG19_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG19_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG19_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>72.376</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>90.876</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>143667240</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>19.63</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>548.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG19_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vgg19_bn.html" class="btn btn-neutral float-right" title="vgg19_bn" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vgg16_bn.html" class="btn btn-neutral" title="vgg16_bn" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg19</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vgg19_bn.html b/0.16/models/generated/torchvision.models.vgg19_bn.html
new file mode 100644
index 00000000000..f2839f4c5f1
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vgg19_bn.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vgg19_bn &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="VisionTransformer" href="../vision_transformer.html" />
+    <link rel="prev" title="vgg19" href="torchvision.models.vgg19.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vgg.html">VGG</a> &gt;</li>
+        
+      <li>vgg19_bn</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vgg19_bn.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg19-bn">
+<h1>vgg19_bn<a class="headerlink" href="#vgg19-bn" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vgg19_bn">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vgg19_bn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.VGG19_BN_Weights" title="torchvision.models.vgg.VGG19_BN_Weights"><span class="pre">VGG19_BN_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VGG</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#vgg19_bn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vgg19_bn" title="Permalink to this definition">¶</a></dt>
+<dd><p>VGG-19_BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.VGG19_BN_Weights" title="torchvision.models.VGG19_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_BN_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.VGG19_BN_Weights" title="torchvision.models.VGG19_BN_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">VGG19_BN_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.VGG19_BN_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">VGG19_BN_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vgg.html#VGG19_BN_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.VGG19_BN_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">VGG19_BN_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">VGG19_BN_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>VGG19_BN_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights were trained from scratch by using a simplified training recipe.
+Also available as <code class="docutils literal notranslate"><span class="pre">VGG19_BN_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>74.218</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>91.842</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=32, width=32</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#alexnet-and-vgg">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>143678248</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>19.63</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>548.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">VGG19_BN_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../vision_transformer.html" class="btn btn-neutral float-right" title="VisionTransformer" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vgg19.html" class="btn btn-neutral" title="vgg19" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vgg19_bn</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.mc3_18.html b/0.16/models/generated/torchvision.models.video.mc3_18.html
new file mode 100644
index 00000000000..8affd14c0eb
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.mc3_18.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mc3_18 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="r2plus1d_18" href="torchvision.models.video.r2plus1d_18.html" />
+    <link rel="prev" title="r3d_18" href="torchvision.models.video.r3d_18.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_resnet.html">Video ResNet</a> &gt;</li>
+        
+      <li>mc3_18</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.mc3_18.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mc3-18">
+<h1>mc3_18<a class="headerlink" href="#mc3-18" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.mc3_18">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">mc3_18</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.MC3_18_Weights" title="torchvision.models.video.resnet.MC3_18_Weights"><span class="pre">MC3_18_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VideoResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/resnet.html#mc3_18"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.mc3_18" title="Permalink to this definition">¶</a></dt>
+<dd><p>Construct 18 layer Mixed Convolution network as in</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1711.11248">A Closer Look at Spatiotemporal Convolutions for Action Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.MC3_18_Weights" title="torchvision.models.video.MC3_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MC3_18_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.MC3_18_Weights" title="torchvision.models.video.MC3_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MC3_18_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.resnet.VideoResNet</span></code> base class.
+Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.MC3_18_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">MC3_18_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/resnet.html#MC3_18_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.MC3_18_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MC3_18_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MC3_18_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>MC3_18_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights reproduce closely the accuracy of the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=5</cite>, and <cite>clip_len=16</cite>.
+Also available as <code class="docutils literal notranslate"><span class="pre">MC3_18_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>63.96</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>84.13</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>11695440</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>43.34</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>44.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MC3_18_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[128,</span> <span class="pre">171]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[112,</span> <span class="pre">112]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.43216,</span> <span class="pre">0.394666,</span> <span class="pre">0.37645]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.22803,</span> <span class="pre">0.22145,</span> <span class="pre">0.216989]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.video.r2plus1d_18.html" class="btn btn-neutral float-right" title="r2plus1d_18" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.video.r3d_18.html" class="btn btn-neutral" title="r3d_18" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mc3_18</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.mvit_v1_b.html b/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
new file mode 100644
index 00000000000..c010b4e7cd1
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
@@ -0,0 +1,759 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mvit_v1_b &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="mvit_v2_s" href="torchvision.models.video.mvit_v2_s.html" />
+    <link rel="prev" title="Video MViT" href="../video_mvit.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_mvit.html">Video MViT</a> &gt;</li>
+        
+      <li>mvit_v1_b</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.mvit_v1_b.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mvit-v1-b">
+<h1>mvit_v1_b<a class="headerlink" href="#mvit-v1-b" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.mvit_v1_b">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">mvit_v1_b</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.MViT_V1_B_Weights" title="torchvision.models.video.mvit.MViT_V1_B_Weights"><span class="pre">MViT_V1_B_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MViT</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/mvit.html#mvit_v1_b"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.mvit_v1_b" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a base MViTV1 architecture from
+<a class="reference external" href="https://arxiv.org/abs/2104.11227">Multiscale Vision Transformers</a>.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.MViT_V1_B_Weights" title="torchvision.models.video.MViT_V1_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V1_B_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.MViT_V1_B_Weights" title="torchvision.models.video.MViT_V1_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V1_B_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.MViT</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/mvit.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.MViT_V1_B_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">MViT_V1_B_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/mvit.html#MViT_V1_B_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.MViT_V1_B_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MViT_V1_B_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MViT_V1_B_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>MViT_V1_B_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights were ported from the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=7.5</cite>, <cite>clips_per_video=5</cite>, and <cite>clip_len=16</cite>
+Also available as <code class="docutils literal notranslate"><span class="pre">MViT_V1_B_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>78.477</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>93.582</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>min_temporal_size</p></td>
+<td><p>16</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/pytorchvideo/blob/main/docs/source/model_zoo.md">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>36610672</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>70.60</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>139.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MViT_V1_B_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224,</span> <span class="pre">224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.45,</span> <span class="pre">0.45,</span> <span class="pre">0.45]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.225,</span> <span class="pre">0.225,</span> <span class="pre">0.225]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.video.mvit_v2_s.html" class="btn btn-neutral float-right" title="mvit_v2_s" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../video_mvit.html" class="btn btn-neutral" title="Video MViT" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mvit_v1_b</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.mvit_v2_s.html b/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
new file mode 100644
index 00000000000..9f9cc588397
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
@@ -0,0 +1,761 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>mvit_v2_s &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Video ResNet" href="../video_resnet.html" />
+    <link rel="prev" title="mvit_v1_b" href="torchvision.models.video.mvit_v1_b.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_mvit.html">Video MViT</a> &gt;</li>
+        
+      <li>mvit_v2_s</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.mvit_v2_s.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mvit-v2-s">
+<h1>mvit_v2_s<a class="headerlink" href="#mvit-v2-s" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.mvit_v2_s">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">mvit_v2_s</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.MViT_V2_S_Weights" title="torchvision.models.video.mvit.MViT_V2_S_Weights"><span class="pre">MViT_V2_S_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">MViT</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/mvit.html#mvit_v2_s"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.mvit_v2_s" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a small MViTV2 architecture from
+<a class="reference external" href="https://arxiv.org/abs/2104.11227">Multiscale Vision Transformers</a> and
+<a class="reference external" href="https://arxiv.org/abs/2112.01526">MViTv2: Improved Multiscale Vision Transformers for Classification
+and Detection</a>.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.MViT_V2_S_Weights" title="torchvision.models.video.MViT_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V2_S_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.MViT_V2_S_Weights" title="torchvision.models.video.MViT_V2_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">MViT_V2_S_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.MViT</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/mvit.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.MViT_V2_S_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">MViT_V2_S_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/mvit.html#MViT_V2_S_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.MViT_V2_S_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">MViT_V2_S_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">MViT_V2_S_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>MViT_V2_S_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights were ported from the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=7.5</cite>, <cite>clips_per_video=5</cite>, and <cite>clip_len=16</cite>
+Also available as <code class="docutils literal notranslate"><span class="pre">MViT_V2_S_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>80.757</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>94.665</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>min_temporal_size</p></td>
+<td><p>16</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SlowFast/blob/main/MODEL_ZOO.md">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>34537744</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>64.22</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>131.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">MViT_V2_S_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224,</span> <span class="pre">224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.45,</span> <span class="pre">0.45,</span> <span class="pre">0.45]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.225,</span> <span class="pre">0.225,</span> <span class="pre">0.225]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../video_resnet.html" class="btn btn-neutral float-right" title="Video ResNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.video.mvit_v1_b.html" class="btn btn-neutral" title="mvit_v1_b" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">mvit_v2_s</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.r2plus1d_18.html b/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
new file mode 100644
index 00000000000..24f78e2b9b5
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>r2plus1d_18 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Video S3D" href="../video_s3d.html" />
+    <link rel="prev" title="mc3_18" href="torchvision.models.video.mc3_18.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_resnet.html">Video ResNet</a> &gt;</li>
+        
+      <li>r2plus1d_18</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.r2plus1d_18.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="r2plus1d-18">
+<h1>r2plus1d_18<a class="headerlink" href="#r2plus1d-18" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.r2plus1d_18">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">r2plus1d_18</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.R2Plus1D_18_Weights" title="torchvision.models.video.resnet.R2Plus1D_18_Weights"><span class="pre">R2Plus1D_18_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VideoResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/resnet.html#r2plus1d_18"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.r2plus1d_18" title="Permalink to this definition">¶</a></dt>
+<dd><p>Construct 18 layer deep R(2+1)D network as in</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1711.11248">A Closer Look at Spatiotemporal Convolutions for Action Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.R2Plus1D_18_Weights" title="torchvision.models.video.R2Plus1D_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.R2Plus1D_18_Weights" title="torchvision.models.video.R2Plus1D_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.resnet.VideoResNet</span></code> base class.
+Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.R2Plus1D_18_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">R2Plus1D_18_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/resnet.html#R2Plus1D_18_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.R2Plus1D_18_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>R2Plus1D_18_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights reproduce closely the accuracy of the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=5</cite>, and <cite>clip_len=16</cite>.
+Also available as <code class="docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>67.463</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>86.175</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>31505325</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>40.52</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>120.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">R2Plus1D_18_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[128,</span> <span class="pre">171]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[112,</span> <span class="pre">112]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.43216,</span> <span class="pre">0.394666,</span> <span class="pre">0.37645]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.22803,</span> <span class="pre">0.22145,</span> <span class="pre">0.216989]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../video_s3d.html" class="btn btn-neutral float-right" title="Video S3D" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.video.mc3_18.html" class="btn btn-neutral" title="mc3_18" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">r2plus1d_18</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.r3d_18.html b/0.16/models/generated/torchvision.models.video.r3d_18.html
new file mode 100644
index 00000000000..43b3101de2b
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.r3d_18.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>r3d_18 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="mc3_18" href="torchvision.models.video.mc3_18.html" />
+    <link rel="prev" title="Video ResNet" href="../video_resnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_resnet.html">Video ResNet</a> &gt;</li>
+        
+      <li>r3d_18</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.r3d_18.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="r3d-18">
+<h1>r3d_18<a class="headerlink" href="#r3d-18" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.r3d_18">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">r3d_18</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.R3D_18_Weights" title="torchvision.models.video.resnet.R3D_18_Weights"><span class="pre">R3D_18_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VideoResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/resnet.html#r3d_18"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.r3d_18" title="Permalink to this definition">¶</a></dt>
+<dd><p>Construct 18 layer Resnet3D model.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1711.11248">A Closer Look at Spatiotemporal Convolutions for Action Recognition</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.R3D_18_Weights" title="torchvision.models.video.R3D_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">R3D_18_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.R3D_18_Weights" title="torchvision.models.video.R3D_18_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">R3D_18_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.resnet.VideoResNet</span></code> base class.
+Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.R3D_18_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">R3D_18_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/resnet.html#R3D_18_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.R3D_18_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">R3D_18_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">R3D_18_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>R3D_18_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights reproduce closely the accuracy of the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=5</cite>, and <cite>clip_len=16</cite>.
+Also available as <code class="docutils literal notranslate"><span class="pre">R3D_18_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>63.2</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>83.479</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>33371472</p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>40.70</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>127.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">R3D_18_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[128,</span> <span class="pre">171]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[112,</span> <span class="pre">112]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.43216,</span> <span class="pre">0.394666,</span> <span class="pre">0.37645]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.22803,</span> <span class="pre">0.22145,</span> <span class="pre">0.216989]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.video.mc3_18.html" class="btn btn-neutral float-right" title="mc3_18" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../video_resnet.html" class="btn btn-neutral" title="Video ResNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">r3d_18</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.s3d.html b/0.16/models/generated/torchvision.models.video.s3d.html
new file mode 100644
index 00000000000..95c1abe3fef
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.s3d.html
@@ -0,0 +1,758 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>s3d &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Video SwinTransformer" href="../video_swin_transformer.html" />
+    <link rel="prev" title="Video S3D" href="../video_s3d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_s3d.html">Video S3D</a> &gt;</li>
+        
+      <li>s3d</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.s3d.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="s3d">
+<h1>s3d<a class="headerlink" href="#s3d" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.s3d">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">s3d</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.S3D_Weights" title="torchvision.models.video.s3d.S3D_Weights"><span class="pre">S3D_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">S3D</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/s3d.html#s3d"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.s3d" title="Permalink to this definition">¶</a></dt>
+<dd><p>Construct Separable 3D CNN model.</p>
+<p>Reference: <a class="reference external" href="https://arxiv.org/abs/1712.04851">Rethinking Spatiotemporal Feature Learning</a>.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.S3D_Weights" title="torchvision.models.video.S3D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">S3D_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.S3D_Weights" title="torchvision.models.video.S3D_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">S3D_Weights</span></code></a>
+below for more details, and possible values. By default, no
+pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.S3D</span></code> base class.
+Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/s3d.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.S3D_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">S3D_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/s3d.html#S3D_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.S3D_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">S3D_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">S3D_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>S3D_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights aim to approximate the accuracy of the paper. The accuracies are estimated on clip-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=1</cite>, and <cite>clip_len=128</cite>.
+Also available as <code class="docutils literal notranslate"><span class="pre">S3D_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>68.368</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>88.05</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>min_temporal_size</p></td>
+<td><p>14</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification#s3d">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>8320048</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>17.98</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>32.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">S3D_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256,</span> <span class="pre">256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224,</span> <span class="pre">224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.43216,</span> <span class="pre">0.394666,</span> <span class="pre">0.37645]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.22803,</span> <span class="pre">0.22145,</span> <span class="pre">0.216989]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../video_swin_transformer.html" class="btn btn-neutral float-right" title="Video SwinTransformer" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../video_s3d.html" class="btn btn-neutral" title="Video S3D" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">s3d</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_b.html b/0.16/models/generated/torchvision.models.video.swin3d_b.html
new file mode 100644
index 00000000000..a111edf32fd
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.swin3d_b.html
@@ -0,0 +1,789 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin3d_b &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="RAFT" href="../raft.html" />
+    <link rel="prev" title="swin3d_s" href="torchvision.models.video.swin3d_s.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_swin_transformer.html">Video SwinTransformer</a> &gt;</li>
+        
+      <li>swin3d_b</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.swin3d_b.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin3d-b">
+<h1>swin3d_b<a class="headerlink" href="#swin3d-b" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.swin3d_b">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">swin3d_b</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.Swin3D_B_Weights" title="torchvision.models.video.swin_transformer.Swin3D_B_Weights"><span class="pre">Swin3D_B_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer3d</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/swin_transformer.html#swin3d_b"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.swin3d_b" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_base architecture from
+<a class="reference external" href="https://arxiv.org/abs/2106.13230">Video Swin Transformer</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.Swin3D_B_Weights" title="torchvision.models.video.Swin3D_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_B_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.Swin3D_B_Weights" title="torchvision.models.video.Swin3D_B_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_B_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.Swin3D_B_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">Swin3D_B_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/swin_transformer.html#Swin3D_B_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.Swin3D_B_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin3D_B_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin3D_B_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>Swin3D_B_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights were ported from the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=12</cite>, and <cite>clip_len=32</cite>
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin3D_B_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>79.427</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>94.386</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_temporal_size</p></td>
+<td><p>1</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88048984</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>140.67</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>364.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin3D_B_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224,</span> <span class="pre">224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+<p><strong>Swin3D_B_Weights.KINETICS400_IMAGENET22K_V1</strong>:</p>
+<p>The weights were ported from the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=12</cite>, and <cite>clip_len=32</cite></p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>81.643</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>95.574</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_temporal_size</p></td>
+<td><p>1</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>88048984</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>140.67</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>364.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin3D_B_Weights.KINETICS400_IMAGENET22K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224,</span> <span class="pre">224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../raft.html" class="btn btn-neutral float-right" title="RAFT" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.video.swin3d_s.html" class="btn btn-neutral" title="swin3d_s" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin3d_b</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_s.html b/0.16/models/generated/torchvision.models.video.swin3d_s.html
new file mode 100644
index 00000000000..9db91e2e27a
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.swin3d_s.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin3d_s &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="swin3d_b" href="torchvision.models.video.swin3d_b.html" />
+    <link rel="prev" title="swin3d_t" href="torchvision.models.video.swin3d_t.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_swin_transformer.html">Video SwinTransformer</a> &gt;</li>
+        
+      <li>swin3d_s</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.swin3d_s.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin3d-s">
+<h1>swin3d_s<a class="headerlink" href="#swin3d-s" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.swin3d_s">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">swin3d_s</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.Swin3D_S_Weights" title="torchvision.models.video.swin_transformer.Swin3D_S_Weights"><span class="pre">Swin3D_S_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer3d</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/swin_transformer.html#swin3d_s"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.swin3d_s" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_small architecture from
+<a class="reference external" href="https://arxiv.org/abs/2106.13230">Video Swin Transformer</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.Swin3D_S_Weights" title="torchvision.models.video.Swin3D_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_S_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.Swin3D_S_Weights" title="torchvision.models.video.Swin3D_S_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_S_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.Swin3D_S_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">Swin3D_S_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/swin_transformer.html#Swin3D_S_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.Swin3D_S_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin3D_S_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin3D_S_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>Swin3D_S_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights were ported from the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=12</cite>, and <cite>clip_len=32</cite>
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin3D_S_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>79.521</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>94.158</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_temporal_size</p></td>
+<td><p>1</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>49816678</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>82.84</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>218.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin3D_S_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224,</span> <span class="pre">224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.video.swin3d_b.html" class="btn btn-neutral float-right" title="swin3d_b" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.video.swin3d_t.html" class="btn btn-neutral" title="swin3d_t" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin3d_s</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_t.html b/0.16/models/generated/torchvision.models.video.swin3d_t.html
new file mode 100644
index 00000000000..f19fb2c3527
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.video.swin3d_t.html
@@ -0,0 +1,755 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>swin3d_t &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="swin3d_s" href="torchvision.models.video.swin3d_s.html" />
+    <link rel="prev" title="Video SwinTransformer" href="../video_swin_transformer.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../video_swin_transformer.html">Video SwinTransformer</a> &gt;</li>
+        
+      <li>swin3d_t</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.video.swin3d_t.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swin3d-t">
+<h1>swin3d_t<a class="headerlink" href="#swin3d-t" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.video.swin3d_t">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">swin3d_t</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.video.Swin3D_T_Weights" title="torchvision.models.video.swin_transformer.Swin3D_T_Weights"><span class="pre">Swin3D_T_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">SwinTransformer3d</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/video/swin_transformer.html#swin3d_t"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.swin3d_t" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a swin_tiny architecture from
+<a class="reference external" href="https://arxiv.org/abs/2106.13230">Video Swin Transformer</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.video.Swin3D_T_Weights" title="torchvision.models.video.Swin3D_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_T_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.video.Swin3D_T_Weights" title="torchvision.models.video.Swin3D_T_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Swin3D_T_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.video.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.video.Swin3D_T_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.video.</span></span><span class="sig-name descname"><span class="pre">Swin3D_T_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/video/swin_transformer.html#Swin3D_T_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.video.Swin3D_T_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Swin3D_T_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Swin3D_T_Weights.KINETICS400_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='KINETICS400_V1'</span></code>.</p>
+<p><strong>Swin3D_T_Weights.KINETICS400_V1</strong>:</p>
+<p>The weights were ported from the paper. The accuracies are estimated on video-level with parameters <cite>frame_rate=15</cite>, <cite>clips_per_video=12</cite>, and <cite>clip_len=32</cite>
+Also available as <code class="docutils literal notranslate"><span class="pre">Swin3D_T_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on Kinetics-400)</p></td>
+<td><p>77.715</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on Kinetics-400)</p></td>
+<td><p>93.519</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>abseiling, air drumming, answering questions, … (397 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_temporal_size</p></td>
+<td><p>1</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/SwinTransformer/Video-Swin-Transformer#kinetics-400">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>28158070</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>43.88</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>121.5 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Swin3D_T_Weights.KINETICS400_V1.transforms</span></code> and perform the following preprocessing operations: Accepts batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(T,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> video frame <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The frames are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224,</span> <span class="pre">224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>. Finally the output dimensions are permuted to <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">T,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> tensors.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.video.swin3d_s.html" class="btn btn-neutral float-right" title="swin3d_s" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../video_swin_transformer.html" class="btn btn-neutral" title="Video SwinTransformer" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">swin3d_t</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vit_b_16.html b/0.16/models/generated/torchvision.models.vit_b_16.html
new file mode 100644
index 00000000000..42ca8c934bc
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vit_b_16.html
@@ -0,0 +1,823 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vit_b_16 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vit_b_32" href="torchvision.models.vit_b_32.html" />
+    <link rel="prev" title="VisionTransformer" href="../vision_transformer.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vision_transformer.html">VisionTransformer</a> &gt;</li>
+        
+      <li>vit_b_16</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vit_b_16.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vit-b-16">
+<h1>vit_b_16<a class="headerlink" href="#vit-b-16" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vit_b_16">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vit_b_16</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ViT_B_16_Weights" title="torchvision.models.vision_transformer.ViT_B_16_Weights"><span class="pre">ViT_B_16_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VisionTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#vit_b_16"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vit_b_16" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a vit_b_16 architecture from
+<a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ViT_B_16_Weights" title="torchvision.models.ViT_B_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ViT_B_16_Weights" title="torchvision.models.ViT_B_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_16_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vision_transformer.VisionTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ViT_B_16_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ViT_B_16_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#ViT_B_16_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ViT_B_16_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ViT_B_16_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ViT_B_16_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using a modified version of <a class="reference external" href="https://arxiv.org/abs/2012.12877">DeIT</a>’s training recipe.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ViT_B_16_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.072</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.318</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>86567656</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_b_16">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>17.56</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>330.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ViT_B_16_Weights.IMAGENET1K_SWAG_E2E_V1</strong>:</p>
+<p>
+                These weights are learnt via transfer learning by end-to-end fine-tuning the original
+                <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> weights on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>85.304</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.65</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>86859496</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=384, width=384</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>55.48</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>331.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_SWAG_E2E_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[384]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[384]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ViT_B_16_Weights.IMAGENET1K_SWAG_LINEAR_V1</strong>:</p>
+<p>
+                These weights are composed of the original frozen <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> trunk
+                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.886</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.18</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>86567656</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>17.56</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>330.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_B_16_Weights.IMAGENET1K_SWAG_LINEAR_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[224]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vit_b_32.html" class="btn btn-neutral float-right" title="vit_b_32" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../vision_transformer.html" class="btn btn-neutral" title="VisionTransformer" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vit_b_16</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vit_b_32.html b/0.16/models/generated/torchvision.models.vit_b_32.html
new file mode 100644
index 00000000000..c6af42e359e
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vit_b_32.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vit_b_32 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vit_l_16" href="torchvision.models.vit_l_16.html" />
+    <link rel="prev" title="vit_b_16" href="torchvision.models.vit_b_16.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vision_transformer.html">VisionTransformer</a> &gt;</li>
+        
+      <li>vit_b_32</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vit_b_32.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vit-b-32">
+<h1>vit_b_32<a class="headerlink" href="#vit-b-32" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vit_b_32">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vit_b_32</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ViT_B_32_Weights" title="torchvision.models.vision_transformer.ViT_B_32_Weights"><span class="pre">ViT_B_32_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VisionTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#vit_b_32"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vit_b_32" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a vit_b_32 architecture from
+<a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ViT_B_32_Weights" title="torchvision.models.ViT_B_32_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_32_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ViT_B_32_Weights" title="torchvision.models.ViT_B_32_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_B_32_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vision_transformer.VisionTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ViT_B_32_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ViT_B_32_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#ViT_B_32_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ViT_B_32_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ViT_B_32_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ViT_B_32_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ViT_B_32_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using a modified version of <a class="reference external" href="https://arxiv.org/abs/2012.12877">DeIT</a>’s training recipe.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ViT_B_32_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>75.912</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>92.466</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>88224232</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_b_32">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>4.41</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>336.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_B_32_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vit_l_16.html" class="btn btn-neutral float-right" title="vit_l_16" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vit_b_16.html" class="btn btn-neutral" title="vit_b_16" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vit_b_32</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vit_h_14.html b/0.16/models/generated/torchvision.models.vit_h_14.html
new file mode 100644
index 00000000000..2bd61ea9fa1
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vit_h_14.html
@@ -0,0 +1,791 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vit_h_14 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Wide ResNet" href="../wide_resnet.html" />
+    <link rel="prev" title="vit_l_32" href="torchvision.models.vit_l_32.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vision_transformer.html">VisionTransformer</a> &gt;</li>
+        
+      <li>vit_h_14</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vit_h_14.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vit-h-14">
+<h1>vit_h_14<a class="headerlink" href="#vit-h-14" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vit_h_14">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vit_h_14</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ViT_H_14_Weights" title="torchvision.models.vision_transformer.ViT_H_14_Weights"><span class="pre">ViT_H_14_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VisionTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#vit_h_14"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vit_h_14" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a vit_h_14 architecture from
+<a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ViT_H_14_Weights" title="torchvision.models.ViT_H_14_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_H_14_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ViT_H_14_Weights" title="torchvision.models.ViT_H_14_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_H_14_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vision_transformer.VisionTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ViT_H_14_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ViT_H_14_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#ViT_H_14_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ViT_H_14_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ViT_H_14_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ViT_H_14_Weights.IMAGENET1K_SWAG_E2E_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_SWAG_E2E_V1'</span></code>.</p>
+<p><strong>ViT_H_14_Weights.IMAGENET1K_SWAG_E2E_V1</strong>:</p>
+<p>
+                These weights are learnt via transfer learning by end-to-end fine-tuning the original
+                <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> weights on ImageNet-1K data.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ViT_H_14_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>88.552</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>98.694</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>633470440</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=518, width=518</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>1016.72</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>2416.6 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_H_14_Weights.IMAGENET1K_SWAG_E2E_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[518]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[518]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ViT_H_14_Weights.IMAGENET1K_SWAG_LINEAR_V1</strong>:</p>
+<p>
+                These weights are composed of the original frozen <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> trunk
+                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>85.708</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.73</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>632045800</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>167.29</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>2411.2 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_H_14_Weights.IMAGENET1K_SWAG_LINEAR_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[224]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../wide_resnet.html" class="btn btn-neutral float-right" title="Wide ResNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vit_l_32.html" class="btn btn-neutral" title="vit_l_32" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vit_h_14</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vit_l_16.html b/0.16/models/generated/torchvision.models.vit_l_16.html
new file mode 100644
index 00000000000..6b89ccc16b1
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vit_l_16.html
@@ -0,0 +1,824 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vit_l_16 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vit_l_32" href="torchvision.models.vit_l_32.html" />
+    <link rel="prev" title="vit_b_32" href="torchvision.models.vit_b_32.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vision_transformer.html">VisionTransformer</a> &gt;</li>
+        
+      <li>vit_l_16</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vit_l_16.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vit-l-16">
+<h1>vit_l_16<a class="headerlink" href="#vit-l-16" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vit_l_16">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vit_l_16</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ViT_L_16_Weights" title="torchvision.models.vision_transformer.ViT_L_16_Weights"><span class="pre">ViT_L_16_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VisionTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#vit_l_16"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vit_l_16" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a vit_l_16 architecture from
+<a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ViT_L_16_Weights" title="torchvision.models.ViT_L_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ViT_L_16_Weights" title="torchvision.models.ViT_L_16_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_16_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vision_transformer.VisionTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ViT_L_16_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ViT_L_16_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#ViT_L_16_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ViT_L_16_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ViT_L_16_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ViT_L_16_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using a modified version of TorchVision’s
+                <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ViT_L_16_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>79.662</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.638</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>304326632</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_l_16">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>61.55</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>1161.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[242]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ViT_L_16_Weights.IMAGENET1K_SWAG_E2E_V1</strong>:</p>
+<p>
+                These weights are learnt via transfer learning by end-to-end fine-tuning the original
+                <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> weights on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>88.064</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>98.512</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>305174504</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=512, width=512</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>361.99</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>1164.3 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_SWAG_E2E_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[512]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[512]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>ViT_L_16_Weights.IMAGENET1K_SWAG_LINEAR_V1</strong>:</p>
+<p>
+                These weights are composed of the original frozen <a class="reference external" href="https://arxiv.org/abs/2201.08371">SWAG</a> trunk
+                weights and a linear classifier learnt on top of them trained on ImageNet-1K data.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>85.146</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>97.422</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/5793">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>license</p></td>
+<td><p><a class="reference external" href="https://github.com/facebookresearch/SWAG/blob/main/LICENSE">link</a></p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>304326632</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>GFLOPS</p></td>
+<td><p>61.55</p></td>
+</tr>
+<tr class="row-odd"><td><p>File size</p></td>
+<td><p>1161.0 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_L_16_Weights.IMAGENET1K_SWAG_LINEAR_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[224]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BICUBIC</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vit_l_32.html" class="btn btn-neutral float-right" title="vit_l_32" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vit_b_32.html" class="btn btn-neutral" title="vit_b_32" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vit_l_16</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.vit_l_32.html b/0.16/models/generated/torchvision.models.vit_l_32.html
new file mode 100644
index 00000000000..bb88eb96d6d
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.vit_l_32.html
@@ -0,0 +1,751 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>vit_l_32 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="vit_h_14" href="torchvision.models.vit_h_14.html" />
+    <link rel="prev" title="vit_l_16" href="torchvision.models.vit_l_16.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../vision_transformer.html">VisionTransformer</a> &gt;</li>
+        
+      <li>vit_l_32</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.vit_l_32.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vit-l-32">
+<h1>vit_l_32<a class="headerlink" href="#vit-l-32" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.vit_l_32">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">vit_l_32</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.ViT_L_32_Weights" title="torchvision.models.vision_transformer.ViT_L_32_Weights"><span class="pre">ViT_L_32_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">VisionTransformer</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#vit_l_32"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.vit_l_32" title="Permalink to this definition">¶</a></dt>
+<dd><p>Constructs a vit_l_32 architecture from
+<a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.ViT_L_32_Weights" title="torchvision.models.ViT_L_32_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_32_Weights</span></code></a>, optional) – The pretrained
+weights to use. See <a class="reference internal" href="#torchvision.models.ViT_L_32_Weights" title="torchvision.models.ViT_L_32_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">ViT_L_32_Weights</span></code></a>
+below for more details and possible values. By default, no pre-trained weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.vision_transformer.VisionTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.ViT_L_32_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">ViT_L_32_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/vision_transformer.html#ViT_L_32_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.ViT_L_32_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">ViT_L_32_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">ViT_L_32_Weights.IMAGENET1K_V1</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>ViT_L_32_Weights.IMAGENET1K_V1</strong>:</p>
+<p>
+                These weights were trained from scratch by using a modified version of <a class="reference external" href="https://arxiv.org/abs/2012.12877">DeIT</a>’s training recipe.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">ViT_L_32_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>76.972</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>93.07</p></td>
+</tr>
+<tr class="row-odd"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-even"><td><p>num_params</p></td>
+<td><p>306535400</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=224, width=224</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification#vit_l_32">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>15.38</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>1169.4 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">ViT_L_32_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.vit_h_14.html" class="btn btn-neutral float-right" title="vit_h_14" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.vit_l_16.html" class="btn btn-neutral" title="vit_l_16" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">vit_l_32</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.wide_resnet101_2.html b/0.16/models/generated/torchvision.models.wide_resnet101_2.html
new file mode 100644
index 00000000000..1213161f866
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.wide_resnet101_2.html
@@ -0,0 +1,789 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>wide_resnet101_2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="Quantized GoogLeNet" href="../googlenet_quant.html" />
+    <link rel="prev" title="wide_resnet50_2" href="torchvision.models.wide_resnet50_2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../wide_resnet.html">Wide ResNet</a> &gt;</li>
+        
+      <li>wide_resnet101_2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.wide_resnet101_2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="wide-resnet101-2">
+<h1>wide_resnet101_2<a class="headerlink" href="#wide-resnet101-2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.wide_resnet101_2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">wide_resnet101_2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Wide_ResNet101_2_Weights" title="torchvision.models.resnet.Wide_ResNet101_2_Weights"><span class="pre">Wide_ResNet101_2_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#wide_resnet101_2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.wide_resnet101_2" title="Permalink to this definition">¶</a></dt>
+<dd><p>Wide ResNet-101-2 model from
+<a class="reference external" href="https://arxiv.org/abs/1605.07146">Wide Residual Networks</a>.</p>
+<p>The model is the same as ResNet except for the bottleneck number of channels
+which is twice larger in every block. The number of channels in outer 1x1
+convolutions is the same, e.g. last block in ResNet-101 has 2048-512-2048
+channels, and in Wide ResNet-101-2 has 2048-1024-2048.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Wide_ResNet101_2_Weights" title="torchvision.models.Wide_ResNet101_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Wide_ResNet101_2_Weights" title="torchvision.models.Wide_ResNet101_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Wide_ResNet101_2_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Wide_ResNet101_2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#Wide_ResNet101_2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Wide_ResNet101_2_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Wide_ResNet101_2_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.848</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.284</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>126886696</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/912#issue-445437439">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>22.75</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>242.9 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>Wide_ResNet101_2_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>82.51</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>96.02</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>126886696</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>22.75</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>484.7 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Wide_ResNet101_2_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../googlenet_quant.html" class="btn btn-neutral float-right" title="Quantized GoogLeNet" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="torchvision.models.wide_resnet50_2.html" class="btn btn-neutral" title="wide_resnet50_2" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">wide_resnet101_2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/generated/torchvision.models.wide_resnet50_2.html b/0.16/models/generated/torchvision.models.wide_resnet50_2.html
new file mode 100644
index 00000000000..0c256bdf335
--- /dev/null
+++ b/0.16/models/generated/torchvision.models.wide_resnet50_2.html
@@ -0,0 +1,789 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>wide_resnet50_2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="wide_resnet101_2" href="torchvision.models.wide_resnet101_2.html" />
+    <link rel="prev" title="Wide ResNet" href="../wide_resnet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+          <li><a href="../wide_resnet.html">Wide ResNet</a> &gt;</li>
+        
+      <li>wide_resnet50_2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../../_sources/models/generated/torchvision.models.wide_resnet50_2.rst.txt" rel="nofollow"><img src="../../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="wide-resnet50-2">
+<h1>wide_resnet50_2<a class="headerlink" href="#wide-resnet50-2" title="Permalink to this heading">¶</a></h1>
+<dl class="py function">
+<dt class="sig sig-object py" id="torchvision.models.wide_resnet50_2">
+<span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">wide_resnet50_2</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weights</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><span class="pre">Optional</span></a><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#torchvision.models.Wide_ResNet50_2_Weights" title="torchvision.models.resnet.Wide_ResNet50_2_Weights"><span class="pre">Wide_ResNet50_2_Weights</span></a><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">progress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><span class="pre">bool</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><span class="pre">Any</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">ResNet</span></span></span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#wide_resnet50_2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.wide_resnet50_2" title="Permalink to this definition">¶</a></dt>
+<dd><p>Wide ResNet-50-2 model from
+<a class="reference external" href="https://arxiv.org/abs/1605.07146">Wide Residual Networks</a>.</p>
+<p>The model is the same as ResNet except for the bottleneck number of channels
+which is twice larger in every block. The number of channels in outer 1x1
+convolutions is the same, e.g. last block in ResNet-50 has 2048-512-2048
+channels, and in Wide ResNet-50-2 has 2048-1024-2048.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><ul class="simple">
+<li><p><strong>weights</strong> (<a class="reference internal" href="#torchvision.models.Wide_ResNet50_2_Weights" title="torchvision.models.Wide_ResNet50_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights</span></code></a>, optional) – The
+pretrained weights to use. See
+<a class="reference internal" href="#torchvision.models.Wide_ResNet50_2_Weights" title="torchvision.models.Wide_ResNet50_2_Weights"><code class="xref py py-class docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights</span></code></a> below for
+more details, and possible values. By default, no pre-trained
+weights are used.</p></li>
+<li><p><strong>progress</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><em>bool</em></a><em>, </em><em>optional</em>) – If True, displays a progress bar of the
+download to stderr. Default is True.</p></li>
+<li><p><strong>**kwargs</strong> – parameters passed to the <code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source code</a>
+for more details about this class.</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py class">
+<dt class="sig sig-object py" id="torchvision.models.Wide_ResNet50_2_Weights">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchvision.models.</span></span><span class="sig-name descname"><span class="pre">Wide_ResNet50_2_Weights</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">value</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../_modules/torchvision/models/resnet.html#Wide_ResNet50_2_Weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchvision.models.Wide_ResNet50_2_Weights" title="Permalink to this definition">¶</a></dt>
+<dd><p>The model builder above accepts the following values as the <code class="docutils literal notranslate"><span class="pre">weights</span></code> parameter.
+<code class="docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.DEFAULT</span></code> is equivalent to <code class="docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.IMAGENET1K_V2</span></code>. You can also use strings, e.g. <code class="docutils literal notranslate"><span class="pre">weights='DEFAULT'</span></code> or <code class="docutils literal notranslate"><span class="pre">weights='IMAGENET1K_V1'</span></code>.</p>
+<p><strong>Wide_ResNet50_2_Weights.IMAGENET1K_V1</strong>:</p>
+<p>These weights reproduce closely the results of the paper using a simple training recipe.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>78.468</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>94.086</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>68883240</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/pull/912#issue-445437439">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>11.40</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>131.8 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.IMAGENET1K_V1.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[256]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+<p><strong>Wide_ResNet50_2_Weights.IMAGENET1K_V2</strong>:</p>
+<p>
+                These weights improve upon the results of the original paper by using TorchVision’s <a class="reference external" href="https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/">new training recipe</a>.
+            
+Also available as <code class="docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.DEFAULT</span></code>.</p>
+<table class="table-weights docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p>acc&#64;1 (on ImageNet-1K)</p></td>
+<td><p>81.602</p></td>
+</tr>
+<tr class="row-even"><td><p>acc&#64;5 (on ImageNet-1K)</p></td>
+<td><p>95.758</p></td>
+</tr>
+<tr class="row-odd"><td><p>min_size</p></td>
+<td><p>height=1, width=1</p></td>
+</tr>
+<tr class="row-even"><td><p>categories</p></td>
+<td><p>tench, goldfish, great white shark, … (997 omitted)</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_params</p></td>
+<td><p>68883240</p></td>
+</tr>
+<tr class="row-even"><td><p>recipe</p></td>
+<td><p><a class="reference external" href="https://github.com/pytorch/vision/issues/3995#new-recipe-with-fixres">link</a></p></td>
+</tr>
+<tr class="row-odd"><td><p>GFLOPS</p></td>
+<td><p>11.40</p></td>
+</tr>
+<tr class="row-even"><td><p>File size</p></td>
+<td><p>263.1 MB</p></td>
+</tr>
+</tbody>
+</table>
+<p>The inference transforms are available at <code class="docutils literal notranslate"><span class="pre">Wide_ResNet50_2_Weights.IMAGENET1K_V2.transforms</span></code> and perform the following preprocessing operations: Accepts <code class="docutils literal notranslate"><span class="pre">PIL.Image</span></code>, batched <code class="docutils literal notranslate"><span class="pre">(B,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and single <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> image <code class="docutils literal notranslate"><span class="pre">torch.Tensor</span></code> objects. The images are resized to <code class="docutils literal notranslate"><span class="pre">resize_size=[232]</span></code> using <code class="docutils literal notranslate"><span class="pre">interpolation=InterpolationMode.BILINEAR</span></code>, followed by a central crop of <code class="docutils literal notranslate"><span class="pre">crop_size=[224]</span></code>. Finally the values are first rescaled to <code class="docutils literal notranslate"><span class="pre">[0.0,</span> <span class="pre">1.0]</span></code> and then normalized using <code class="docutils literal notranslate"><span class="pre">mean=[0.485,</span> <span class="pre">0.456,</span> <span class="pre">0.406]</span></code> and <code class="docutils literal notranslate"><span class="pre">std=[0.229,</span> <span class="pre">0.224,</span> <span class="pre">0.225]</span></code>.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="torchvision.models.wide_resnet101_2.html" class="btn btn-neutral float-right" title="wide_resnet101_2" accesskey="n" rel="next">Next <img src="../../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="../wide_resnet.html" class="btn btn-neutral" title="Wide ResNet" accesskey="p" rel="prev"><img src="../../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">wide_resnet50_2</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
+         <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
+         <script src="../../_static/jquery.js"></script>
+         <script src="../../_static/underscore.js"></script>
+         <script src="../../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../../_static/doctools.js"></script>
+         <script src="../../_static/clipboard.min.js"></script>
+         <script src="../../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/googlenet.html b/0.16/models/googlenet.html
new file mode 100644
index 00000000000..3405f47f1d3
--- /dev/null
+++ b/0.16/models/googlenet.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>GoogLeNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="googlenet" href="generated/torchvision.models.googlenet.html" />
+    <link rel="prev" title="efficientnet_v2_l" href="generated/torchvision.models.efficientnet_v2_l.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>GoogLeNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/googlenet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="googlenet">
+<h1>GoogLeNet<a class="headerlink" href="#googlenet" title="Permalink to this heading">¶</a></h1>
+<p>The GoogleNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1409.4842">Going Deeper with Convolutions</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a GoogLeNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.googlenet.GoogLeNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/googlenet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.googlenet.html#torchvision.models.googlenet" title="torchvision.models.googlenet"><code class="xref py py-obj docutils literal notranslate"><span class="pre">googlenet</span></code></a>(*[, weights, progress])</p></td>
+<td><p>GoogLeNet (Inception v1) model architecture from <a class="reference external" href="http://arxiv.org/abs/1409.4842">Going Deeper with Convolutions</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.googlenet.html" class="btn btn-neutral float-right" title="googlenet" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.efficientnet_v2_l.html" class="btn btn-neutral" title="efficientnet_v2_l" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">GoogLeNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/googlenet_quant.html b/0.16/models/googlenet_quant.html
new file mode 100644
index 00000000000..eedf20c1516
--- /dev/null
+++ b/0.16/models/googlenet_quant.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Quantized GoogLeNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="googlenet" href="generated/torchvision.models.quantization.googlenet.html" />
+    <link rel="prev" title="wide_resnet101_2" href="generated/torchvision.models.wide_resnet101_2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Quantized GoogLeNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/googlenet_quant.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="quantized-googlenet">
+<h1>Quantized GoogLeNet<a class="headerlink" href="#quantized-googlenet" title="Permalink to this heading">¶</a></h1>
+<p>The Quantized GoogleNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1409.4842">Going Deeper with Convolutions</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a quantized GoogLeNet
+model, with or without pre-trained weights. All the model builders internally
+rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.googlenet.QuantizableGoogLeNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/googlenet.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.googlenet.html#torchvision.models.quantization.googlenet" title="torchvision.models.quantization.googlenet"><code class="xref py py-obj docutils literal notranslate"><span class="pre">googlenet</span></code></a>(*[, weights, progress, quantize])</p></td>
+<td><p>GoogLeNet (Inception v1) model architecture from <a class="reference external" href="http://arxiv.org/abs/1409.4842">Going Deeper with Convolutions</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.quantization.googlenet.html" class="btn btn-neutral float-right" title="googlenet" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.wide_resnet101_2.html" class="btn btn-neutral" title="wide_resnet101_2" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Quantized GoogLeNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/inception.html b/0.16/models/inception.html
new file mode 100644
index 00000000000..b2029c883e4
--- /dev/null
+++ b/0.16/models/inception.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Inception V3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="inception_v3" href="generated/torchvision.models.inception_v3.html" />
+    <link rel="prev" title="googlenet" href="generated/torchvision.models.googlenet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Inception V3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/inception.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="inception-v3">
+<h1>Inception V3<a class="headerlink" href="#inception-v3" title="Permalink to this heading">¶</a></h1>
+<p>The InceptionV3 model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.00567">Rethinking the Inception Architecture for
+Computer Vision</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an InceptionV3 model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.inception.Inception3</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/inception.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.inception_v3.html#torchvision.models.inception_v3" title="torchvision.models.inception_v3"><code class="xref py py-obj docutils literal notranslate"><span class="pre">inception_v3</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Inception v3 model architecture from <a class="reference external" href="http://arxiv.org/abs/1512.00567">Rethinking the Inception Architecture for Computer Vision</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.inception_v3.html" class="btn btn-neutral float-right" title="inception_v3" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.googlenet.html" class="btn btn-neutral" title="googlenet" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Inception V3</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/inception_quant.html b/0.16/models/inception_quant.html
new file mode 100644
index 00000000000..4e40d76572a
--- /dev/null
+++ b/0.16/models/inception_quant.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Quantized InceptionV3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="inception_v3" href="generated/torchvision.models.quantization.inception_v3.html" />
+    <link rel="prev" title="googlenet" href="generated/torchvision.models.quantization.googlenet.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Quantized InceptionV3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/inception_quant.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="quantized-inceptionv3">
+<h1>Quantized InceptionV3<a class="headerlink" href="#quantized-inceptionv3" title="Permalink to this heading">¶</a></h1>
+<p>The Quantized Inception model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.00567">Rethinking the Inception Architecture for
+Computer Vision</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a quantized Inception
+model, with or without pre-trained weights. All the model builders internally
+rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.inception.QuantizableInception3</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/inception.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.inception_v3.html#torchvision.models.quantization.inception_v3" title="torchvision.models.quantization.inception_v3"><code class="xref py py-obj docutils literal notranslate"><span class="pre">inception_v3</span></code></a>(*[, weights, progress, quantize])</p></td>
+<td><p>Inception v3 model architecture from <a class="reference external" href="http://arxiv.org/abs/1512.00567">Rethinking the Inception Architecture for Computer Vision</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.quantization.inception_v3.html" class="btn btn-neutral float-right" title="inception_v3" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.quantization.googlenet.html" class="btn btn-neutral" title="googlenet" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Quantized InceptionV3</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/keypoint_rcnn.html b/0.16/models/keypoint_rcnn.html
new file mode 100644
index 00000000000..5e5ed92d4ee
--- /dev/null
+++ b/0.16/models/keypoint_rcnn.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Keypoint R-CNN &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="keypointrcnn_resnet50_fpn" href="generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html" />
+    <link rel="prev" title="maskrcnn_resnet50_fpn_v2" href="generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Keypoint R-CNN</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/keypoint_rcnn.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="keypoint-r-cnn">
+<h1>Keypoint R-CNN<a class="headerlink" href="#keypoint-r-cnn" title="Permalink to this heading">¶</a></h1>
+<p>The Keypoint R-CNN model is based on the <a class="reference external" href="https://arxiv.org/abs/1703.06870">Mask R-CNN</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a Keypoint R-CNN model,
+with or without pre-trained weights. All the model builders internally rely on
+the <code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.KeypointRCNN</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/keypoint_rcnn.py">source
+code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html#torchvision.models.detection.keypointrcnn_resnet50_fpn" title="torchvision.models.detection.keypointrcnn_resnet50_fpn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">keypointrcnn_resnet50_fpn</span></code></a>(*[, weights, ...])</p></td>
+<td><p>Constructs a Keypoint R-CNN model with a ResNet-50-FPN backbone.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html" class="btn btn-neutral float-right" title="keypointrcnn_resnet50_fpn" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html" class="btn btn-neutral" title="maskrcnn_resnet50_fpn_v2" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Keypoint R-CNN</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/lraspp.html b/0.16/models/lraspp.html
new file mode 100644
index 00000000000..82c81e47e53
--- /dev/null
+++ b/0.16/models/lraspp.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>LRASPP &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="lraspp_mobilenet_v3_large" href="generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html" />
+    <link rel="prev" title="fcn_resnet101" href="generated/torchvision.models.segmentation.fcn_resnet101.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>LRASPP</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/lraspp.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="lraspp">
+<h1>LRASPP<a class="headerlink" href="#lraspp" title="Permalink to this heading">¶</a></h1>
+<p>The LRASPP model is based on the <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The segmentation module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a FCN model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.segmentation.LRASPP</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/segmentation/lraspp.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html#torchvision.models.segmentation.lraspp_mobilenet_v3_large" title="torchvision.models.segmentation.lraspp_mobilenet_v3_large"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lraspp_mobilenet_v3_large</span></code></a>(*[, weights, ...])</p></td>
+<td><p><p>Constructs a Lite R-ASPP Network model with a MobileNetV3-Large backbone from <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> paper.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html" class="btn btn-neutral float-right" title="lraspp_mobilenet_v3_large" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.segmentation.fcn_resnet101.html" class="btn btn-neutral" title="fcn_resnet101" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">LRASPP</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/mask_rcnn.html b/0.16/models/mask_rcnn.html
new file mode 100644
index 00000000000..638fcde9ef4
--- /dev/null
+++ b/0.16/models/mask_rcnn.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Mask R-CNN &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="maskrcnn_resnet50_fpn" href="generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html" />
+    <link rel="prev" title="ssdlite320_mobilenet_v3_large" href="generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Mask R-CNN</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/mask_rcnn.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mask-r-cnn">
+<h1>Mask R-CNN<a class="headerlink" href="#mask-r-cnn" title="Permalink to this heading">¶</a></h1>
+<p>The Mask R-CNN model is based on the <a class="reference external" href="https://arxiv.org/abs/1703.06870">Mask R-CNN</a>
+paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a Mask R-CNN model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.mask_rcnn.MaskRCNN</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/mask_rcnn.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html#torchvision.models.detection.maskrcnn_resnet50_fpn" title="torchvision.models.detection.maskrcnn_resnet50_fpn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">maskrcnn_resnet50_fpn</span></code></a>(*[, weights, ...])</p></td>
+<td><p><p>Mask R-CNN model with a ResNet-50-FPN backbone from the <a class="reference external" href="https://arxiv.org/abs/1703.06870">Mask R-CNN</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html#torchvision.models.detection.maskrcnn_resnet50_fpn_v2" title="torchvision.models.detection.maskrcnn_resnet50_fpn_v2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">maskrcnn_resnet50_fpn_v2</span></code></a>(*[, weights, ...])</p></td>
+<td><p>Improved Mask R-CNN model with a ResNet-50-FPN backbone from the <a class="reference external" href="https://arxiv.org/abs/2111.11429">Benchmarking Detection Transfer Learning with Vision Transformers</a> paper.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html" class="btn btn-neutral float-right" title="maskrcnn_resnet50_fpn" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html" class="btn btn-neutral" title="ssdlite320_mobilenet_v3_large" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Mask R-CNN</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/maxvit.html b/0.16/models/maxvit.html
new file mode 100644
index 00000000000..e38ea884371
--- /dev/null
+++ b/0.16/models/maxvit.html
@@ -0,0 +1,707 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MaxVit &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="maxvit_t" href="generated/torchvision.models.maxvit_t.html" />
+    <link rel="prev" title="inception_v3" href="generated/torchvision.models.inception_v3.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>MaxVit</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/maxvit.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="maxvit">
+<h1>MaxVit<a class="headerlink" href="#maxvit" title="Permalink to this heading">¶</a></h1>
+<p>The MaxVit transformer models are based on the <a class="reference external" href="https://arxiv.org/abs/2204.01697">MaxViT: Multi-Axis Vision Transformer</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an MaxVit model with and without pre-trained weights.
+All the model builders internally rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.maxvit.MaxVit</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/maxvit.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.maxvit_t.html#torchvision.models.maxvit_t" title="torchvision.models.maxvit_t"><code class="xref py py-obj docutils literal notranslate"><span class="pre">maxvit_t</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a maxvit_t architecture from <a class="reference external" href="https://arxiv.org/abs/2204.01697">MaxViT: Multi-Axis Vision Transformer</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.maxvit_t.html" class="btn btn-neutral float-right" title="maxvit_t" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.inception_v3.html" class="btn btn-neutral" title="inception_v3" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MaxVit</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/mnasnet.html b/0.16/models/mnasnet.html
new file mode 100644
index 00000000000..9b830dd5df6
--- /dev/null
+++ b/0.16/models/mnasnet.html
@@ -0,0 +1,720 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MNASNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="mnasnet0_5" href="generated/torchvision.models.mnasnet0_5.html" />
+    <link rel="prev" title="maxvit_t" href="generated/torchvision.models.maxvit_t.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>MNASNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/mnasnet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mnasnet">
+<h1>MNASNet<a class="headerlink" href="#mnasnet" title="Permalink to this heading">¶</a></h1>
+<p>The MNASNet model is based on the <a class="reference external" href="https://arxiv.org/pdf/1807.11626.pdf">MnasNet: Platform-Aware Neural Architecture
+Search for Mobile</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an MNASNet model.
+All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.mnasnet.MNASNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mnasnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.mnasnet0_5.html#torchvision.models.mnasnet0_5" title="torchvision.models.mnasnet0_5"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mnasnet0_5</span></code></a>(*[, weights, progress])</p></td>
+<td><p>MNASNet with depth multiplier of 0.5 from <a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.mnasnet0_75.html#torchvision.models.mnasnet0_75" title="torchvision.models.mnasnet0_75"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mnasnet0_75</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>MNASNet with depth multiplier of 0.75 from <a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.mnasnet1_0.html#torchvision.models.mnasnet1_0" title="torchvision.models.mnasnet1_0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mnasnet1_0</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>MNASNet with depth multiplier of 1.0 from <a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.mnasnet1_3.html#torchvision.models.mnasnet1_3" title="torchvision.models.mnasnet1_3"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mnasnet1_3</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>MNASNet with depth multiplier of 1.3 from <a class="reference external" href="https://arxiv.org/abs/1807.11626">MnasNet: Platform-Aware Neural Architecture Search for Mobile</a> paper.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.mnasnet0_5.html" class="btn btn-neutral float-right" title="mnasnet0_5" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.maxvit_t.html" class="btn btn-neutral" title="maxvit_t" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MNASNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/mobilenetv2.html b/0.16/models/mobilenetv2.html
new file mode 100644
index 00000000000..c04f4885d3c
--- /dev/null
+++ b/0.16/models/mobilenetv2.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MobileNet V2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="mobilenet_v2" href="generated/torchvision.models.mobilenet_v2.html" />
+    <link rel="prev" title="mnasnet1_3" href="generated/torchvision.models.mnasnet1_3.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>MobileNet V2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/mobilenetv2.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mobilenet-v2">
+<h1>MobileNet V2<a class="headerlink" href="#mobilenet-v2" title="Permalink to this heading">¶</a></h1>
+<p>The MobileNet V2 model is based on the <a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear
+Bottlenecks</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a MobileNetV2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.mobilenetv2.MobileNetV2</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv2.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.mobilenet_v2.html#torchvision.models.mobilenet_v2" title="torchvision.models.mobilenet_v2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mobilenet_v2</span></code></a>(*[, weights, progress])</p></td>
+<td><p>MobileNetV2 architecture from the <a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear Bottlenecks</a> paper.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.mobilenet_v2.html" class="btn btn-neutral float-right" title="mobilenet_v2" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.mnasnet1_3.html" class="btn btn-neutral" title="mnasnet1_3" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MobileNet V2</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/mobilenetv2_quant.html b/0.16/models/mobilenetv2_quant.html
new file mode 100644
index 00000000000..d717e50960e
--- /dev/null
+++ b/0.16/models/mobilenetv2_quant.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Quantized MobileNet V2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="mobilenet_v2" href="generated/torchvision.models.quantization.mobilenet_v2.html" />
+    <link rel="prev" title="inception_v3" href="generated/torchvision.models.quantization.inception_v3.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Quantized MobileNet V2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/mobilenetv2_quant.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="quantized-mobilenet-v2">
+<h1>Quantized MobileNet V2<a class="headerlink" href="#quantized-mobilenet-v2" title="Permalink to this heading">¶</a></h1>
+<p>The Quantized MobileNet V2 model is based on the <a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear
+Bottlenecks</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a quantized MobileNetV2
+model, with or without pre-trained weights. All the model builders internally
+rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.mobilenetv2.QuantizableMobileNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv2.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.mobilenet_v2.html#torchvision.models.quantization.mobilenet_v2" title="torchvision.models.quantization.mobilenet_v2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mobilenet_v2</span></code></a>(*[, weights, progress, quantize])</p></td>
+<td><p>Constructs a MobileNetV2 architecture from <a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear Bottlenecks</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.quantization.mobilenet_v2.html" class="btn btn-neutral float-right" title="mobilenet_v2" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.quantization.inception_v3.html" class="btn btn-neutral" title="inception_v3" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Quantized MobileNet V2</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/mobilenetv3.html b/0.16/models/mobilenetv3.html
new file mode 100644
index 00000000000..04727448f90
--- /dev/null
+++ b/0.16/models/mobilenetv3.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>MobileNet V3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="mobilenet_v3_large" href="generated/torchvision.models.mobilenet_v3_large.html" />
+    <link rel="prev" title="mobilenet_v2" href="generated/torchvision.models.mobilenet_v2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>MobileNet V3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/mobilenetv3.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="mobilenet-v3">
+<h1>MobileNet V3<a class="headerlink" href="#mobilenet-v3" title="Permalink to this heading">¶</a></h1>
+<p>The MobileNet V3 model is based on the <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a MobileNetV3 model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.mobilenetv3.MobileNetV3</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/mobilenetv3.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.mobilenet_v3_large.html#torchvision.models.mobilenet_v3_large" title="torchvision.models.mobilenet_v3_large"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mobilenet_v3_large</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a large MobileNetV3 architecture from <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.mobilenet_v3_small.html#torchvision.models.mobilenet_v3_small" title="torchvision.models.mobilenet_v3_small"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mobilenet_v3_small</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a small MobileNetV3 architecture from <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.mobilenet_v3_large.html" class="btn btn-neutral float-right" title="mobilenet_v3_large" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.mobilenet_v2.html" class="btn btn-neutral" title="mobilenet_v2" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">MobileNet V3</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/mobilenetv3_quant.html b/0.16/models/mobilenetv3_quant.html
new file mode 100644
index 00000000000..38acf7b9035
--- /dev/null
+++ b/0.16/models/mobilenetv3_quant.html
@@ -0,0 +1,707 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Quantized MobileNet V3 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="mobilenet_v3_large" href="generated/torchvision.models.quantization.mobilenet_v3_large.html" />
+    <link rel="prev" title="mobilenet_v2" href="generated/torchvision.models.quantization.mobilenet_v2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Quantized MobileNet V3</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/mobilenetv3_quant.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="quantized-mobilenet-v3">
+<h1>Quantized MobileNet V3<a class="headerlink" href="#quantized-mobilenet-v3" title="Permalink to this heading">¶</a></h1>
+<p>The Quantized MobileNet V3 model is based on the <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a quantized MobileNetV3
+model, with or without pre-trained weights. All the model builders internally
+rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.mobilenetv3.QuantizableMobileNetV3</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/mobilenetv3.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.mobilenet_v3_large.html#torchvision.models.quantization.mobilenet_v3_large" title="torchvision.models.quantization.mobilenet_v3_large"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mobilenet_v3_large</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>MobileNetV3 (Large) model from <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.quantization.mobilenet_v3_large.html" class="btn btn-neutral float-right" title="mobilenet_v3_large" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.quantization.mobilenet_v2.html" class="btn btn-neutral" title="mobilenet_v2" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Quantized MobileNet V3</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/raft.html b/0.16/models/raft.html
new file mode 100644
index 00000000000..f1576603381
--- /dev/null
+++ b/0.16/models/raft.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RAFT &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="raft_large" href="generated/torchvision.models.optical_flow.raft_large.html" />
+    <link rel="prev" title="swin3d_b" href="generated/torchvision.models.video.swin3d_b.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>RAFT</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/raft.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="raft">
+<h1>RAFT<a class="headerlink" href="#raft" title="Permalink to this heading">¶</a></h1>
+<p>The RAFT model is based on the <a class="reference external" href="https://arxiv.org/abs/2003.12039">RAFT: Recurrent All-Pairs Field Transforms for
+Optical Flow</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a RAFT model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.optical_flow.RAFT</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/optical_flow/raft.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.optical_flow.raft_large.html#torchvision.models.optical_flow.raft_large" title="torchvision.models.optical_flow.raft_large"><code class="xref py py-obj docutils literal notranslate"><span class="pre">raft_large</span></code></a>(*[, weights, progress])</p></td>
+<td><p>RAFT model from <a class="reference external" href="https://arxiv.org/abs/2003.12039">RAFT: Recurrent All Pairs Field Transforms for Optical Flow</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.optical_flow.raft_small.html#torchvision.models.optical_flow.raft_small" title="torchvision.models.optical_flow.raft_small"><code class="xref py py-obj docutils literal notranslate"><span class="pre">raft_small</span></code></a>(*[, weights, progress])</p></td>
+<td><p>RAFT &quot;small&quot; model from <a class="reference external" href="https://arxiv.org/abs/2003.12039">RAFT: Recurrent All Pairs Field Transforms for Optical Flow</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.optical_flow.raft_large.html" class="btn btn-neutral float-right" title="raft_large" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.video.swin3d_b.html" class="btn btn-neutral" title="swin3d_b" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RAFT</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/regnet.html b/0.16/models/regnet.html
new file mode 100644
index 00000000000..9b9e502f0cf
--- /dev/null
+++ b/0.16/models/regnet.html
@@ -0,0 +1,763 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RegNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="regnet_y_400mf" href="generated/torchvision.models.regnet_y_400mf.html" />
+    <link rel="prev" title="mobilenet_v3_small" href="generated/torchvision.models.mobilenet_v3_small.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>RegNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/regnet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="regnet">
+<h1>RegNet<a class="headerlink" href="#regnet" title="Permalink to this heading">¶</a></h1>
+<p>The RegNet model is based on the <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a RegNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.regnet.RegNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/regnet.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_400mf.html#torchvision.models.regnet_y_400mf" title="torchvision.models.regnet_y_400mf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_400mf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_400MF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_800mf.html#torchvision.models.regnet_y_800mf" title="torchvision.models.regnet_y_800mf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_800mf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_800MF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_1_6gf.html#torchvision.models.regnet_y_1_6gf" title="torchvision.models.regnet_y_1_6gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_1_6gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_1.6GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_3_2gf.html#torchvision.models.regnet_y_3_2gf" title="torchvision.models.regnet_y_3_2gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_3_2gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_3.2GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_8gf.html#torchvision.models.regnet_y_8gf" title="torchvision.models.regnet_y_8gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_8gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_8GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_16gf.html#torchvision.models.regnet_y_16gf" title="torchvision.models.regnet_y_16gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_16gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_16GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_32gf.html#torchvision.models.regnet_y_32gf" title="torchvision.models.regnet_y_32gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_32gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_32GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_y_128gf.html#torchvision.models.regnet_y_128gf" title="torchvision.models.regnet_y_128gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_y_128gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetY_128GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_x_400mf.html#torchvision.models.regnet_x_400mf" title="torchvision.models.regnet_x_400mf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_x_400mf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetX_400MF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_x_800mf.html#torchvision.models.regnet_x_800mf" title="torchvision.models.regnet_x_800mf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_x_800mf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetX_800MF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_x_1_6gf.html#torchvision.models.regnet_x_1_6gf" title="torchvision.models.regnet_x_1_6gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_x_1_6gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetX_1.6GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_x_3_2gf.html#torchvision.models.regnet_x_3_2gf" title="torchvision.models.regnet_x_3_2gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_x_3_2gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetX_3.2GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_x_8gf.html#torchvision.models.regnet_x_8gf" title="torchvision.models.regnet_x_8gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_x_8gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetX_8GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_x_16gf.html#torchvision.models.regnet_x_16gf" title="torchvision.models.regnet_x_16gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_x_16gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetX_16GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.regnet_x_32gf.html#torchvision.models.regnet_x_32gf" title="torchvision.models.regnet_x_32gf"><code class="xref py py-obj docutils literal notranslate"><span class="pre">regnet_x_32gf</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a RegNetX_32GF architecture from <a class="reference external" href="https://arxiv.org/abs/2003.13678">Designing Network Design Spaces</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.regnet_y_400mf.html" class="btn btn-neutral float-right" title="regnet_y_400mf" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.mobilenet_v3_small.html" class="btn btn-neutral" title="mobilenet_v3_small" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RegNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/resnet.html b/0.16/models/resnet.html
new file mode 100644
index 00000000000..aefdec9b675
--- /dev/null
+++ b/0.16/models/resnet.html
@@ -0,0 +1,725 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ResNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resnet18" href="generated/torchvision.models.resnet18.html" />
+    <link rel="prev" title="regnet_x_32gf" href="generated/torchvision.models.regnet_x_32gf.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>ResNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/resnet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnet">
+<h1>ResNet<a class="headerlink" href="#resnet" title="Permalink to this heading">¶</a></h1>
+<p>The ResNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a> paper.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The bottleneck of TorchVision places the stride for downsampling to the second 3x3
+convolution while the original paper places it to the first 1x1 convolution.
+This variant improves the accuracy and is known as <a class="reference external" href="https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch">ResNet V1.5</a>.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a ResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.resnet18.html#torchvision.models.resnet18" title="torchvision.models.resnet18"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnet18</span></code></a>(*[, weights, progress])</p></td>
+<td><p>ResNet-18 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.resnet34.html#torchvision.models.resnet34" title="torchvision.models.resnet34"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnet34</span></code></a>(*[, weights, progress])</p></td>
+<td><p>ResNet-34 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.resnet50.html#torchvision.models.resnet50" title="torchvision.models.resnet50"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnet50</span></code></a>(*[, weights, progress])</p></td>
+<td><p>ResNet-50 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.resnet101.html#torchvision.models.resnet101" title="torchvision.models.resnet101"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnet101</span></code></a>(*[, weights, progress])</p></td>
+<td><p>ResNet-101 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.resnet152.html#torchvision.models.resnet152" title="torchvision.models.resnet152"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnet152</span></code></a>(*[, weights, progress])</p></td>
+<td><p>ResNet-152 from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.resnet18.html" class="btn btn-neutral float-right" title="resnet18" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.regnet_x_32gf.html" class="btn btn-neutral" title="regnet_x_32gf" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ResNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/resnet_quant.html b/0.16/models/resnet_quant.html
new file mode 100644
index 00000000000..f659f099303
--- /dev/null
+++ b/0.16/models/resnet_quant.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Quantized ResNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resnet18" href="generated/torchvision.models.quantization.resnet18.html" />
+    <link rel="prev" title="mobilenet_v3_large" href="generated/torchvision.models.quantization.mobilenet_v3_large.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Quantized ResNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/resnet_quant.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="quantized-resnet">
+<h1>Quantized ResNet<a class="headerlink" href="#quantized-resnet" title="Permalink to this heading">¶</a></h1>
+<p>The Quantized ResNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a quantized ResNet
+model, with or without pre-trained weights. All the model builders internally
+rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.resnet.QuantizableResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.resnet18.html#torchvision.models.quantization.resnet18" title="torchvision.models.quantization.resnet18"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnet18</span></code></a>(*[, weights, progress, quantize])</p></td>
+<td><p><p>ResNet-18 model from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a></p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.resnet50.html#torchvision.models.quantization.resnet50" title="torchvision.models.quantization.resnet50"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnet50</span></code></a>(*[, weights, progress, quantize])</p></td>
+<td><p><p>ResNet-50 model from <a class="reference external" href="https://arxiv.org/abs/1512.03385">Deep Residual Learning for Image Recognition</a></p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.quantization.resnet18.html" class="btn btn-neutral float-right" title="resnet18" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.quantization.mobilenet_v3_large.html" class="btn btn-neutral" title="mobilenet_v3_large" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Quantized ResNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/resnext.html b/0.16/models/resnext.html
new file mode 100644
index 00000000000..ae8e2d6bd2a
--- /dev/null
+++ b/0.16/models/resnext.html
@@ -0,0 +1,716 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ResNeXt &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resnext50_32x4d" href="generated/torchvision.models.resnext50_32x4d.html" />
+    <link rel="prev" title="resnet152" href="generated/torchvision.models.resnet152.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>ResNeXt</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/resnext.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="resnext">
+<h1>ResNeXt<a class="headerlink" href="#resnext" title="Permalink to this heading">¶</a></h1>
+<p>The ResNext model is based on the <a class="reference external" href="https://arxiv.org/abs/1611.05431v2">Aggregated Residual Transformations for Deep Neural Networks</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a ResNext model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.resnext50_32x4d.html#torchvision.models.resnext50_32x4d" title="torchvision.models.resnext50_32x4d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnext50_32x4d</span></code></a>(*[, weights, progress])</p></td>
+<td><p>ResNeXt-50 32x4d model from <a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.resnext101_32x8d.html#torchvision.models.resnext101_32x8d" title="torchvision.models.resnext101_32x8d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnext101_32x8d</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>ResNeXt-101 32x8d model from <a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.resnext101_64x4d.html#torchvision.models.resnext101_64x4d" title="torchvision.models.resnext101_64x4d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnext101_64x4d</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>ResNeXt-101 64x4d model from <a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.resnext50_32x4d.html" class="btn btn-neutral float-right" title="resnext50_32x4d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.resnet152.html" class="btn btn-neutral" title="resnet152" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ResNeXt</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/resnext_quant.html b/0.16/models/resnext_quant.html
new file mode 100644
index 00000000000..126ccf24932
--- /dev/null
+++ b/0.16/models/resnext_quant.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Quantized ResNeXt &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="resnext101_32x8d" href="generated/torchvision.models.quantization.resnext101_32x8d.html" />
+    <link rel="prev" title="resnet50" href="generated/torchvision.models.quantization.resnet50.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Quantized ResNeXt</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/resnext_quant.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="quantized-resnext">
+<h1>Quantized ResNeXt<a class="headerlink" href="#quantized-resnext" title="Permalink to this heading">¶</a></h1>
+<p>The quantized ResNext model is based on the <a class="reference external" href="https://arxiv.org/abs/1611.05431v2">Aggregated Residual Transformations for Deep Neural Networks</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a quantized ResNeXt
+model, with or without pre-trained weights. All the model builders internally
+rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.resnet.QuantizableResNet</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/resnet.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.resnext101_32x8d.html#torchvision.models.quantization.resnext101_32x8d" title="torchvision.models.quantization.resnext101_32x8d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnext101_32x8d</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>ResNeXt-101 32x8d model from <a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.resnext101_64x4d.html#torchvision.models.quantization.resnext101_64x4d" title="torchvision.models.quantization.resnext101_64x4d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resnext101_64x4d</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p><p>ResNeXt-101 64x4d model from <a class="reference external" href="https://arxiv.org/abs/1611.05431">Aggregated Residual Transformation for Deep Neural Networks</a></p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.quantization.resnext101_32x8d.html" class="btn btn-neutral float-right" title="resnext101_32x8d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.quantization.resnet50.html" class="btn btn-neutral" title="resnet50" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Quantized ResNeXt</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/retinanet.html b/0.16/models/retinanet.html
new file mode 100644
index 00000000000..a934e0a91c6
--- /dev/null
+++ b/0.16/models/retinanet.html
@@ -0,0 +1,713 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>RetinaNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="retinanet_resnet50_fpn" href="generated/torchvision.models.detection.retinanet_resnet50_fpn.html" />
+    <link rel="prev" title="fcos_resnet50_fpn" href="generated/torchvision.models.detection.fcos_resnet50_fpn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>RetinaNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/retinanet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="retinanet">
+<h1>RetinaNet<a class="headerlink" href="#retinanet" title="Permalink to this heading">¶</a></h1>
+<p>The RetinaNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1708.02002">Focal Loss for Dense Object Detection</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a RetinaNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.retinanet.RetinaNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/retinanet.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.retinanet_resnet50_fpn.html#torchvision.models.detection.retinanet_resnet50_fpn" title="torchvision.models.detection.retinanet_resnet50_fpn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">retinanet_resnet50_fpn</span></code></a>(*[, weights, ...])</p></td>
+<td><p>Constructs a RetinaNet model with a ResNet-50-FPN backbone.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html#torchvision.models.detection.retinanet_resnet50_fpn_v2" title="torchvision.models.detection.retinanet_resnet50_fpn_v2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">retinanet_resnet50_fpn_v2</span></code></a>(*[, weights, ...])</p></td>
+<td><p>Constructs an improved RetinaNet model with a ResNet-50-FPN backbone.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.detection.retinanet_resnet50_fpn.html" class="btn btn-neutral float-right" title="retinanet_resnet50_fpn" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.detection.fcos_resnet50_fpn.html" class="btn btn-neutral" title="fcos_resnet50_fpn" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">RetinaNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/shufflenetv2.html b/0.16/models/shufflenetv2.html
new file mode 100644
index 00000000000..16f4eb2d1e2
--- /dev/null
+++ b/0.16/models/shufflenetv2.html
@@ -0,0 +1,717 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>ShuffleNet V2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="shufflenet_v2_x0_5" href="generated/torchvision.models.shufflenet_v2_x0_5.html" />
+    <link rel="prev" title="resnext101_64x4d" href="generated/torchvision.models.resnext101_64x4d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>ShuffleNet V2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/shufflenetv2.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="shufflenet-v2">
+<h1>ShuffleNet V2<a class="headerlink" href="#shufflenet-v2" title="Permalink to this heading">¶</a></h1>
+<p>The ShuffleNet V2 model is based on the <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient
+CNN Architecture Design</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a ShuffleNetV2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.shufflenetv2.ShuffleNetV2</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.shufflenet_v2_x0_5.html#torchvision.models.shufflenet_v2_x0_5" title="torchvision.models.shufflenet_v2_x0_5"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x0_5</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a ShuffleNetV2 architecture with 0.5x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.shufflenet_v2_x1_0.html#torchvision.models.shufflenet_v2_x1_0" title="torchvision.models.shufflenet_v2_x1_0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x1_0</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a ShuffleNetV2 architecture with 1.0x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.shufflenet_v2_x1_5" title="torchvision.models.shufflenet_v2_x1_5"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x1_5</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a ShuffleNetV2 architecture with 1.5x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.shufflenet_v2_x2_0.html#torchvision.models.shufflenet_v2_x2_0" title="torchvision.models.shufflenet_v2_x2_0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x2_0</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a ShuffleNetV2 architecture with 2.0x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.shufflenet_v2_x0_5.html" class="btn btn-neutral float-right" title="shufflenet_v2_x0_5" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.resnext101_64x4d.html" class="btn btn-neutral" title="resnext101_64x4d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">ShuffleNet V2</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/shufflenetv2_quant.html b/0.16/models/shufflenetv2_quant.html
new file mode 100644
index 00000000000..89e12b1aa12
--- /dev/null
+++ b/0.16/models/shufflenetv2_quant.html
@@ -0,0 +1,717 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Quantized ShuffleNet V2 &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="shufflenet_v2_x0_5" href="generated/torchvision.models.quantization.shufflenet_v2_x0_5.html" />
+    <link rel="prev" title="resnext101_64x4d" href="generated/torchvision.models.quantization.resnext101_64x4d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Quantized ShuffleNet V2</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/shufflenetv2_quant.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="quantized-shufflenet-v2">
+<h1>Quantized ShuffleNet V2<a class="headerlink" href="#quantized-shufflenet-v2" title="Permalink to this heading">¶</a></h1>
+<p>The Quantized ShuffleNet V2 model is based on the <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient
+CNN Architecture Design</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a quantized ShuffleNetV2
+model, with or without pre-trained weights. All the model builders internally rely
+on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.quantization.shufflenetv2.QuantizableShuffleNetV2</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/quantization/shufflenetv2.py">source code</a>
+for more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.shufflenet_v2_x0_5.html#torchvision.models.quantization.shufflenet_v2_x0_5" title="torchvision.models.quantization.shufflenet_v2_x0_5"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x0_5</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Constructs a ShuffleNetV2 with 0.5x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.shufflenet_v2_x1_0.html#torchvision.models.quantization.shufflenet_v2_x1_0" title="torchvision.models.quantization.shufflenet_v2_x1_0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x1_0</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Constructs a ShuffleNetV2 with 1.0x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.shufflenet_v2_x1_5.html#torchvision.models.quantization.shufflenet_v2_x1_5" title="torchvision.models.quantization.shufflenet_v2_x1_5"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x1_5</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Constructs a ShuffleNetV2 with 1.5x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.quantization.shufflenet_v2_x2_0.html#torchvision.models.quantization.shufflenet_v2_x2_0" title="torchvision.models.quantization.shufflenet_v2_x2_0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">shufflenet_v2_x2_0</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>Constructs a ShuffleNetV2 with 2.0x output channels, as described in <a class="reference external" href="https://arxiv.org/abs/1807.11164">ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.quantization.shufflenet_v2_x0_5.html" class="btn btn-neutral float-right" title="shufflenet_v2_x0_5" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.quantization.resnext101_64x4d.html" class="btn btn-neutral" title="resnext101_64x4d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Quantized ShuffleNet V2</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/squeezenet.html b/0.16/models/squeezenet.html
new file mode 100644
index 00000000000..ee01f6841e2
--- /dev/null
+++ b/0.16/models/squeezenet.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SqueezeNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="squeezenet1_0" href="generated/torchvision.models.squeezenet1_0.html" />
+    <link rel="prev" title="shufflenet_v2_x2_0" href="generated/torchvision.models.shufflenet_v2_x2_0.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>SqueezeNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/squeezenet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="squeezenet">
+<h1>SqueezeNet<a class="headerlink" href="#squeezenet" title="Permalink to this heading">¶</a></h1>
+<p>The SqueezeNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1602.07360">SqueezeNet: AlexNet-level accuracy with
+50x fewer parameters and &lt;0.5MB model size</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a SqueezeNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.squeezenet.SqueezeNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/squeezenet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.squeezenet1_0.html#torchvision.models.squeezenet1_0" title="torchvision.models.squeezenet1_0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">squeezenet1_0</span></code></a>(*[, weights, progress])</p></td>
+<td><p>SqueezeNet model architecture from the <a class="reference external" href="https://arxiv.org/abs/1602.07360">SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and &lt;0.5MB model size</a> paper.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.squeezenet1_1.html#torchvision.models.squeezenet1_1" title="torchvision.models.squeezenet1_1"><code class="xref py py-obj docutils literal notranslate"><span class="pre">squeezenet1_1</span></code></a>(*[, weights, progress])</p></td>
+<td><p>SqueezeNet 1.1 model from the <a class="reference external" href="https://github.com/DeepScale/SqueezeNet/tree/master/SqueezeNet_v1.1">official SqueezeNet repo</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.squeezenet1_0.html" class="btn btn-neutral float-right" title="squeezenet1_0" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.shufflenet_v2_x2_0.html" class="btn btn-neutral" title="shufflenet_v2_x2_0" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SqueezeNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/ssd.html b/0.16/models/ssd.html
new file mode 100644
index 00000000000..5baa11efb09
--- /dev/null
+++ b/0.16/models/ssd.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SSD &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ssd300_vgg16" href="generated/torchvision.models.detection.ssd300_vgg16.html" />
+    <link rel="prev" title="retinanet_resnet50_fpn_v2" href="generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>SSD</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/ssd.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ssd">
+<h1>SSD<a class="headerlink" href="#ssd" title="Permalink to this heading">¶</a></h1>
+<p>The SSD model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.02325">SSD: Single Shot MultiBox Detector</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a SSD model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.SSD</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.ssd300_vgg16.html#torchvision.models.detection.ssd300_vgg16" title="torchvision.models.detection.ssd300_vgg16"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ssd300_vgg16</span></code></a>(*[, weights, progress, ...])</p></td>
+<td><p>The SSD300 model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.02325">SSD: Single Shot MultiBox Detector</a> paper.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.detection.ssd300_vgg16.html" class="btn btn-neutral float-right" title="ssd300_vgg16" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html" class="btn btn-neutral" title="retinanet_resnet50_fpn_v2" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SSD</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/ssdlite.html b/0.16/models/ssdlite.html
new file mode 100644
index 00000000000..738834996a6
--- /dev/null
+++ b/0.16/models/ssdlite.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SSDlite &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="ssdlite320_mobilenet_v3_large" href="generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html" />
+    <link rel="prev" title="ssd300_vgg16" href="generated/torchvision.models.detection.ssd300_vgg16.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>SSDlite</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/ssdlite.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="ssdlite">
+<h1>SSDlite<a class="headerlink" href="#ssdlite" title="Permalink to this heading">¶</a></h1>
+<p>The SSDLite model is based on the <a class="reference external" href="https://arxiv.org/abs/1512.02325">SSD: Single Shot MultiBox Detector</a>, <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> and <cite>MobileNetV2: Inverted Residuals and Linear
+Bottlenecks &lt;https://arxiv.org/abs/1801.04381&gt;__</cite> papers.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The detection module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a SSD Lite model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.detection.ssd.SSD</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/detection/ssd.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html#torchvision.models.detection.ssdlite320_mobilenet_v3_large" title="torchvision.models.detection.ssdlite320_mobilenet_v3_large"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ssdlite320_mobilenet_v3_large</span></code></a>(*[, weights, ...])</p></td>
+<td><p>SSDlite model architecture with input size 320x320 and a MobileNetV3 Large backbone, as described at <a class="reference external" href="https://arxiv.org/abs/1905.02244">Searching for MobileNetV3</a> and <a class="reference external" href="https://arxiv.org/abs/1801.04381">MobileNetV2: Inverted Residuals and Linear Bottlenecks</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html" class="btn btn-neutral float-right" title="ssdlite320_mobilenet_v3_large" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.detection.ssd300_vgg16.html" class="btn btn-neutral" title="ssd300_vgg16" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SSDlite</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/swin_transformer.html b/0.16/models/swin_transformer.html
new file mode 100644
index 00000000000..ad430ea1cb6
--- /dev/null
+++ b/0.16/models/swin_transformer.html
@@ -0,0 +1,730 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>SwinTransformer &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="swin_t" href="generated/torchvision.models.swin_t.html" />
+    <link rel="prev" title="squeezenet1_1" href="generated/torchvision.models.squeezenet1_1.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>SwinTransformer</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/swin_transformer.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="swintransformer">
+<h1>SwinTransformer<a class="headerlink" href="#swintransformer" title="Permalink to this heading">¶</a></h1>
+<p>The SwinTransformer models are based on the <a class="reference external" href="https://arxiv.org/abs/2103.14030">Swin Transformer: Hierarchical Vision
+Transformer using Shifted Windows</a>
+paper.
+SwinTransformer V2 models are based on the <a class="reference external" href="https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Swin_Transformer_V2_Scaling_Up_Capacity_and_Resolution_CVPR_2022_paper.pdf">Swin Transformer V2: Scaling Up Capacity
+and Resolution</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an SwinTransformer model (original and V2) with and without pre-trained weights.
+All the model builders internally rely on the <code class="docutils literal notranslate"><span class="pre">torchvision.models.swin_transformer.SwinTransformer</span></code>
+base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/swin_transformer.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.swin_t.html#torchvision.models.swin_t" title="torchvision.models.swin_t"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin_t</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a swin_tiny architecture from <a class="reference external" href="https://arxiv.org/abs/2103.14030">Swin Transformer: Hierarchical Vision Transformer using Shifted Windows</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.swin_s.html#torchvision.models.swin_s" title="torchvision.models.swin_s"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin_s</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a swin_small architecture from <a class="reference external" href="https://arxiv.org/abs/2103.14030">Swin Transformer: Hierarchical Vision Transformer using Shifted Windows</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.swin_b.html#torchvision.models.swin_b" title="torchvision.models.swin_b"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin_b</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a swin_base architecture from <a class="reference external" href="https://arxiv.org/abs/2103.14030">Swin Transformer: Hierarchical Vision Transformer using Shifted Windows</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.swin_v2_t.html#torchvision.models.swin_v2_t" title="torchvision.models.swin_v2_t"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin_v2_t</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a swin_v2_tiny architecture from <a class="reference external" href="https://arxiv.org/abs/2111.09883">Swin Transformer V2: Scaling Up Capacity and Resolution</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.swin_v2_s.html#torchvision.models.swin_v2_s" title="torchvision.models.swin_v2_s"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin_v2_s</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a swin_v2_small architecture from <a class="reference external" href="https://arxiv.org/abs/2111.09883">Swin Transformer V2: Scaling Up Capacity and Resolution</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.swin_v2_b.html#torchvision.models.swin_v2_b" title="torchvision.models.swin_v2_b"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin_v2_b</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a swin_v2_base architecture from <a class="reference external" href="https://arxiv.org/abs/2111.09883">Swin Transformer V2: Scaling Up Capacity and Resolution</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.swin_t.html" class="btn btn-neutral float-right" title="swin_t" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.squeezenet1_1.html" class="btn btn-neutral" title="squeezenet1_1" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">SwinTransformer</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/vgg.html b/0.16/models/vgg.html
new file mode 100644
index 00000000000..22dfb73255b
--- /dev/null
+++ b/0.16/models/vgg.html
@@ -0,0 +1,728 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>VGG &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="vgg11" href="generated/torchvision.models.vgg11.html" />
+    <link rel="prev" title="swin_v2_b" href="generated/torchvision.models.swin_v2_b.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>VGG</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/vgg.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="vgg">
+<h1>VGG<a class="headerlink" href="#vgg" title="Permalink to this heading">¶</a></h1>
+<p>The VGG model is based on the <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale
+Image Recognition</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a VGG model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.vgg.VGG</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vgg.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.vgg11.html#torchvision.models.vgg11" title="torchvision.models.vgg11"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg11</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-11 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.vgg11_bn.html#torchvision.models.vgg11_bn" title="torchvision.models.vgg11_bn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg11_bn</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-11-BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.vgg13.html#torchvision.models.vgg13" title="torchvision.models.vgg13"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg13</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-13 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.vgg13_bn.html#torchvision.models.vgg13_bn" title="torchvision.models.vgg13_bn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg13_bn</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-13-BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.vgg16.html#torchvision.models.vgg16" title="torchvision.models.vgg16"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg16</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-16 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.vgg16_bn.html#torchvision.models.vgg16_bn" title="torchvision.models.vgg16_bn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg16_bn</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-16-BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.vgg19.html#torchvision.models.vgg19" title="torchvision.models.vgg19"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg19</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-19 from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.vgg19_bn.html#torchvision.models.vgg19_bn" title="torchvision.models.vgg19_bn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vgg19_bn</span></code></a>(*[, weights, progress])</p></td>
+<td><p>VGG-19_BN from <a class="reference external" href="https://arxiv.org/abs/1409.1556">Very Deep Convolutional Networks for Large-Scale Image Recognition</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.vgg11.html" class="btn btn-neutral float-right" title="vgg11" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.swin_v2_b.html" class="btn btn-neutral" title="swin_v2_b" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">VGG</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/video_mvit.html b/0.16/models/video_mvit.html
new file mode 100644
index 00000000000..0782ee4c3e3
--- /dev/null
+++ b/0.16/models/video_mvit.html
@@ -0,0 +1,711 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Video MViT &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="mvit_v1_b" href="generated/torchvision.models.video.mvit_v1_b.html" />
+    <link rel="prev" title="keypointrcnn_resnet50_fpn" href="generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Video MViT</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/video_mvit.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="video-mvit">
+<h1>Video MViT<a class="headerlink" href="#video-mvit" title="Permalink to this heading">¶</a></h1>
+<p>The MViT model is based on the
+<a class="reference external" href="https://arxiv.org/abs/2112.01526">MViTv2: Improved Multiscale Vision Transformers for Classification and Detection</a> and <a class="reference external" href="https://arxiv.org/abs/2104.11227">Multiscale Vision Transformers</a> papers.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a MViT v1 or v2 model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.video.MViT</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/mvit.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.video.mvit_v1_b.html#torchvision.models.video.mvit_v1_b" title="torchvision.models.video.mvit_v1_b"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mvit_v1_b</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a base MViTV1 architecture from <a class="reference external" href="https://arxiv.org/abs/2104.11227">Multiscale Vision Transformers</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.video.mvit_v2_s.html#torchvision.models.video.mvit_v2_s" title="torchvision.models.video.mvit_v2_s"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mvit_v2_s</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a small MViTV2 architecture from <a class="reference external" href="https://arxiv.org/abs/2104.11227">Multiscale Vision Transformers</a> and <a class="reference external" href="https://arxiv.org/abs/2112.01526">MViTv2: Improved Multiscale Vision Transformers for Classification and Detection</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.video.mvit_v1_b.html" class="btn btn-neutral float-right" title="mvit_v1_b" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html" class="btn btn-neutral" title="keypointrcnn_resnet50_fpn" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Video MViT</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/video_resnet.html b/0.16/models/video_resnet.html
new file mode 100644
index 00000000000..7af50e4bcdb
--- /dev/null
+++ b/0.16/models/video_resnet.html
@@ -0,0 +1,718 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Video ResNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="r3d_18" href="generated/torchvision.models.video.r3d_18.html" />
+    <link rel="prev" title="mvit_v2_s" href="generated/torchvision.models.video.mvit_v2_s.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Video ResNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/video_resnet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="video-resnet">
+<h1>Video ResNet<a class="headerlink" href="#video-resnet" title="Permalink to this heading">¶</a></h1>
+<p>The VideoResNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1711.11248">A Closer Look at Spatiotemporal
+Convolutions for Action Recognition</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a VideoResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.video.resnet.VideoResNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/resnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.video.r3d_18.html#torchvision.models.video.r3d_18" title="torchvision.models.video.r3d_18"><code class="xref py py-obj docutils literal notranslate"><span class="pre">r3d_18</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Construct 18 layer Resnet3D model.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.video.mc3_18.html#torchvision.models.video.mc3_18" title="torchvision.models.video.mc3_18"><code class="xref py py-obj docutils literal notranslate"><span class="pre">mc3_18</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Construct 18 layer Mixed Convolution network as in</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.video.r2plus1d_18.html#torchvision.models.video.r2plus1d_18" title="torchvision.models.video.r2plus1d_18"><code class="xref py py-obj docutils literal notranslate"><span class="pre">r2plus1d_18</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Construct 18 layer deep R(2+1)D network as in</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.video.r3d_18.html" class="btn btn-neutral float-right" title="r3d_18" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.video.mvit_v2_s.html" class="btn btn-neutral" title="mvit_v2_s" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Video ResNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/video_s3d.html b/0.16/models/video_s3d.html
new file mode 100644
index 00000000000..b6cc2bf1a3f
--- /dev/null
+++ b/0.16/models/video_s3d.html
@@ -0,0 +1,708 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Video S3D &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="s3d" href="generated/torchvision.models.video.s3d.html" />
+    <link rel="prev" title="r2plus1d_18" href="generated/torchvision.models.video.r2plus1d_18.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Video S3D</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/video_s3d.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="video-s3d">
+<h1>Video S3D<a class="headerlink" href="#video-s3d" title="Permalink to this heading">¶</a></h1>
+<p>The S3D model is based on the
+<a class="reference external" href="https://arxiv.org/abs/1712.04851">Rethinking Spatiotemporal Feature Learning: Speed-Accuracy Trade-offs in Video Classification</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate an S3D model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.video.S3D</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/s3d.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.video.s3d.html#torchvision.models.video.s3d" title="torchvision.models.video.s3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">s3d</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Construct Separable 3D CNN model.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.video.s3d.html" class="btn btn-neutral float-right" title="s3d" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.video.r2plus1d_18.html" class="btn btn-neutral" title="r2plus1d_18" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Video S3D</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/video_swin_transformer.html b/0.16/models/video_swin_transformer.html
new file mode 100644
index 00000000000..fb3e95cdeae
--- /dev/null
+++ b/0.16/models/video_swin_transformer.html
@@ -0,0 +1,719 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Video SwinTransformer &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="swin3d_t" href="generated/torchvision.models.video.swin3d_t.html" />
+    <link rel="prev" title="s3d" href="generated/torchvision.models.video.s3d.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Video SwinTransformer</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/video_swin_transformer.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="video-swintransformer">
+<h1>Video SwinTransformer<a class="headerlink" href="#video-swintransformer" title="Permalink to this heading">¶</a></h1>
+<p>The Video SwinTransformer model is based on the <a class="reference external" href="https://arxiv.org/abs/2106.13230">Video Swin Transformer</a> paper.</p>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>The video module is in Beta stage, and backward compatibility is not guaranteed.</p>
+</div>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a VideoResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.video.swin_transformer.SwinTransformer3d</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/video/swin_transformer.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.video.swin3d_t.html#torchvision.models.video.swin3d_t" title="torchvision.models.video.swin3d_t"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin3d_t</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Constructs a swin_tiny architecture from <a class="reference external" href="https://arxiv.org/abs/2106.13230">Video Swin Transformer</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.video.swin3d_s.html#torchvision.models.video.swin3d_s" title="torchvision.models.video.swin3d_s"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin3d_s</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a swin_small architecture from <a class="reference external" href="https://arxiv.org/abs/2106.13230">Video Swin Transformer</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.video.swin3d_b.html#torchvision.models.video.swin3d_b" title="torchvision.models.video.swin3d_b"><code class="xref py py-obj docutils literal notranslate"><span class="pre">swin3d_b</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a swin_base architecture from <a class="reference external" href="https://arxiv.org/abs/2106.13230">Video Swin Transformer</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.video.swin3d_t.html" class="btn btn-neutral float-right" title="swin3d_t" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.video.s3d.html" class="btn btn-neutral" title="s3d" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Video SwinTransformer</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/vision_transformer.html b/0.16/models/vision_transformer.html
new file mode 100644
index 00000000000..50b63e4e32a
--- /dev/null
+++ b/0.16/models/vision_transformer.html
@@ -0,0 +1,725 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>VisionTransformer &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="vit_b_16" href="generated/torchvision.models.vit_b_16.html" />
+    <link rel="prev" title="vgg19_bn" href="generated/torchvision.models.vgg19_bn.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>VisionTransformer</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/vision_transformer.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="visiontransformer">
+<h1>VisionTransformer<a class="headerlink" href="#visiontransformer" title="Permalink to this heading">¶</a></h1>
+<p>The VisionTransformer model is based on the <a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words:
+Transformers for Image Recognition at Scale</a> paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a VisionTransformer model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.vision_transformer.VisionTransformer</span></code> base class.
+Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/vision_transformer.py">source code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.vit_b_16.html#torchvision.models.vit_b_16" title="torchvision.models.vit_b_16"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vit_b_16</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a vit_b_16 architecture from <a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.vit_b_32.html#torchvision.models.vit_b_32" title="torchvision.models.vit_b_32"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vit_b_32</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a vit_b_32 architecture from <a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.vit_l_16.html#torchvision.models.vit_l_16" title="torchvision.models.vit_l_16"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vit_l_16</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a vit_l_16 architecture from <a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.vit_l_32.html#torchvision.models.vit_l_32" title="torchvision.models.vit_l_32"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vit_l_32</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a vit_l_32 architecture from <a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.vit_h_14.html#torchvision.models.vit_h_14" title="torchvision.models.vit_h_14"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vit_h_14</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Constructs a vit_h_14 architecture from <a class="reference external" href="https://arxiv.org/abs/2010.11929">An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.vit_b_16.html" class="btn btn-neutral float-right" title="vit_b_16" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.vgg19_bn.html" class="btn btn-neutral" title="vgg19_bn" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">VisionTransformer</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/models/wide_resnet.html b/0.16/models/wide_resnet.html
new file mode 100644
index 00000000000..45acb5d1394
--- /dev/null
+++ b/0.16/models/wide_resnet.html
@@ -0,0 +1,712 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Wide ResNet &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="../_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="wide_resnet50_2" href="generated/torchvision.models.wide_resnet50_2.html" />
+    <link rel="prev" title="vit_h_14" href="generated/torchvision.models.vit_h_14.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="../_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="../auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="../index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+          <li><a href="../models.html">Models and pre-trained weights</a> &gt;</li>
+        
+      <li>Wide ResNet</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="../_sources/models/wide_resnet.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="wide-resnet">
+<h1>Wide ResNet<a class="headerlink" href="#wide-resnet" title="Permalink to this heading">¶</a></h1>
+<p>The Wide ResNet model is based on the <a class="reference external" href="https://arxiv.org/abs/1605.07146">Wide Residual Networks</a>
+paper.</p>
+<section id="model-builders">
+<h2>Model builders<a class="headerlink" href="#model-builders" title="Permalink to this heading">¶</a></h2>
+<p>The following model builders can be used to instantiate a Wide ResNet model, with or
+without pre-trained weights. All the model builders internally rely on the
+<code class="docutils literal notranslate"><span class="pre">torchvision.models.resnet.ResNet</span></code> base class. Please refer to the <a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/models/resnet.py">source
+code</a> for
+more details about this class.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.models.wide_resnet50_2.html#torchvision.models.wide_resnet50_2" title="torchvision.models.wide_resnet50_2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">wide_resnet50_2</span></code></a>(*[, weights, progress])</p></td>
+<td><p>Wide ResNet-50-2 model from <a class="reference external" href="https://arxiv.org/abs/1605.07146">Wide Residual Networks</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.models.wide_resnet101_2.html#torchvision.models.wide_resnet101_2" title="torchvision.models.wide_resnet101_2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">wide_resnet101_2</span></code></a>(*[, weights, progress])</p></td>
+<td><p><p>Wide ResNet-101-2 model from <a class="reference external" href="https://arxiv.org/abs/1605.07146">Wide Residual Networks</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.models.wide_resnet50_2.html" class="btn btn-neutral float-right" title="wide_resnet50_2" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.models.vit_h_14.html" class="btn btn-neutral" title="vit_h_14" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Wide ResNet</a><ul>
+<li><a class="reference internal" href="#model-builders">Model builders</a></li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
+         <script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
+         <script src="../_static/jquery.js"></script>
+         <script src="../_static/underscore.js"></script>
+         <script src="../_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="../_static/doctools.js"></script>
+         <script src="../_static/clipboard.min.js"></script>
+         <script src="../_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="../_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="../_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/objects.inv b/0.16/objects.inv
new file mode 100644
index 0000000000000000000000000000000000000000..ab8893bf8f206bbacec857c221ab6c113119b8c1
GIT binary patch
literal 18596
zcmV*JKxV%qAX9K?X>NERX>N99Zgg*Qc_4OWa&u{KZXhxWBOp+6Z)#;@bUGkZZ*pU3
zc4>2IZ*B@BAXa5^b7^mGIv{OfX>K4WFfK7RE-))OVlgvfV`3=^BOp|0Wgv28ZDDC{
zWMy(7Z)PBLXlZjGW@&6?AZc?TV{dJ6a%FRKWn>_Ab7^j8AbM<RVlE1J+QofcciXnM
z?R$O&@7R4jOfAWFoITDsdn9(!`sT~E9C!D;aL^JgbB7{TlCrJ-`U{XC36cP@)&kx8
z(A0#m=9~Zl5FZObk!Guh$0(1|#M>SIxlPymn0ZHDl&snF>xW_+|Dy`?fAaiSMH~C|
zIxNDR75VG)%a8A_XGg+(6^D85&AbguSQZv+{qvYb1#iQB9RwRzL<QRhK`D_H59xYh
z5`uatvZH=0&Gm<6!7`R&Qc&AKvB_zG5`}TWRu8kI8738t5g47E4kXq5<P@I}ZUjbm
zObCQ3i$i?DYOUb-^t6D<Sj_I;VbW0=fKbWmK@?GO$XDTx<rwje5g47E2ryN;O3%Yx
z!2?lAQOpnw<mBM!E*P&%h5T|NQq<UCI9$>MC)_7Rc9_pl&2VD`MyIC*NOf^@fk*+5
zZO5Z(27p(8d`#mt%P_K4I}FF`d)~VXN|vf&0jGt`xD9`>K>qepj32&yy<FZPGO#iL
zp;F&IjN<0k#Rd9Xt60JD$?1TT?Xz_G4)q!;3m7gr0U(l(AFk&|vjsxDw1DAK#RExx
zxjlauWveW_FEA-k%n%IZv>@nsqahjqOk$)75Sg3^C`}wk$>!#P|C6IDxW*dKM^7Q1
zJb&N;*hW}2Lom?O1*eOn)sJj(^aGQc+6IbEP6vwU44;NGD6(4$05s_n%ktUL9F>IH
z2#iio2AJx=>(co{7-cysDXkfTfu0;BUDZM?F3?G;4N)v^$%0a*=^8f`B`sjM<TQXt
zez{p*zeh+f<3lj1vLOV&Ud%orG>hl}M5<y4{FjUOCzuja8Gum9vLO_gU&CUbg>jT)
z`iP1Z9G{vF8rg~iZqn=!B;l655ZyN1F!V58v4Y@#Wd#<Tn*fo?uLMYwR6WHxrn}J^
zfzipyfKq+J_F_LlI+UNBdaf{vDB;*EOb4Sj0V0#-Lura4LaFERA((XCKmtBOpQFck
zJxd&qQ<Bi6C*;JPP<UJuhaeSQcx<wS4HVlcDFC9+AAi3>jw)&sATl)t2pX2|VunmX
zX#>S}Ndt&z`TZ+KKaUPTB&)W=W502b?~l_L4nz>G*D<@>X9r%#kI_w%&YZwVP9cFl
zedL2y^t)=U;P~WJfyusYlV}sCm>zCz0z{^!0ZQ|HpJwaJ%gdW9R6o6Ahv6Wn2SIps
z{_YDREh0Jqk*XX5|Bx2xO}fVn!x|eXHd#NEC=ORFU!0y`5>eYgvB`-568(m4n0|va
z6Tdk%M<9x8mhb^NTQ8S4n2BZ848cH63W9F={-Rx82d1X6fnrnB0Vn#7lK<U~N0)2`
z|M2C*<u^>9Pg=lm$--eIH=j{1yJdU`CRsLw;QQCb+cy~1r2z<)tQ$me`zxxP@aO<U
zvSa}KibIH@C%T8-v&8X`Q-mXpl7hvUG?fM*RB{U76!=%7m?0S4k^-YE!bJ2_&qLJg
zg=&UiASVb;_x=1MZso8t0HIRl!zh+lON?wD9e_xdY=^)7c24oXRm~6#)P&&ZmTa?S
zNfF}Hq*~&598!iMEuZl{)Q+8n<M+#p57!_0xCMdV*g&zV*##vE(p?d)!Z^4W^S*kY
z5uf12VRuPE2pal)_#_Vuh+7FrR@Z;PN38Ih-4Q^<n)L%8JFjn9m;mYMcWHc7hhwal
zWTMZ@?|ZNfi<Nj)x_xLqc7K><gBw_iZts$h-%n)S7zGuib@2A_PS;lNaj2kC$5fy0
z918Y=g$mVq6!_?Mew(ubRE=?+*Qc$yf)<mp;!fz|PQc<$kmAOdZG4A?6Zk3-RI739
z>(j=epaC{moOTH469#4W5vD=Ys@g-U-L&hYx6<171H(jgHr)E?ZLqezye1K14`(>o
zUh1PZe~#@~HStFGVK&@%@8kBwk8b9NFjJ@j+#}@8Zd2XHSRcJz)wbu>q$6FWh8?Xw
zVUMYO!>mn*W+3cSHiiOrx*DZ4J2Q2gNPYCyPTT%egN}8v?r#3{;TtrKJ4p>(Y@elh
zSPXZ(`MBL^qub~V-1p%RCLT0<q<rM=rSYv-CVrNcJJJRl=6u{{w$WX9ChpDSDhTT>
zWcpoiK4Bl4eN&xDP`Q&uw_nJ|?&Gm+M1lx7Bk-Tu>CZ;kn;kH^%{o4M3yp0bkV!}7
zV{O<@?Njy|1NLi=i_yA_`*@#mt3SF?Mzv>Tu%*SvZ8sU+Cu88=6nS>n-;d&B_KS>c
za4~Q`Y}a=uv%&TZAGLX6Y{!RzHxjQRVTHzr-$#1ak8cey@h9yj2JIFPAHNU6x_`u=
z<wx!5J#<F*k#6GSKJG?-;j!IwB=<3n-Xosr2G02YN#lm=BOPSNy?ZUxJqKDJ>)2}X
z%r&v+`=sak>LZ*}&7QOdzHK7PXaA@NA9WDJ*y$0YwE8X1K4H_beFycZkWXj{pKRox
z0JLaU8!3SEH#F&QHrC(9X-jX`K&Dvdee2^rVy&*d1~F;vmOAvD^^rcU;|{qd{?d&!
zh@z+ncKi!v@PG9w{-^*4TZ5|ID%|$GR(*tDso4qE#3x)j3*pkK@jUgh9-$VOR+EOt
zBU9rz>SG-`Egq=`4ViRD4D!5~8(b)Tf>Wd2L(?EAzgz1ZAAPJtqQ#@q$iDA-Ec$S-
zLZkbmiHmKS2pBjW`Z$NcQBOt_xAcDrR>A>B@3Z3*ylCt$K}JFG?5OX;=wn<AM|>hp
zocl!h`So28eUyXXm{+5T_wga=`49RC4?nXzqJeL}x)-Cqz9W;5ax)t9jWY4(yS^8a
zkMI*RJ2e^j<g>vs#iw|t06bqzV$zktp~)w>7}@=$#%1W;kbIg;5Xjfbpyw&#Ux0hz
zG^6qD@v%NP7N;SThAiCGC)WkMVG4>YOmhCmHV+=l@811BsdCjSXdFxL8MNV1%h`qC
zE?oWK-^3oJa!|9*^pR&_iZ_kYYRci7h3hZr8sJjZw=m+~3PmCo<P?Q&d<bTl>Mbk6
zwVV~VcxVwAa>jnV6mk))_8ISI6+vEPEZk1Vvt}#)TB@=HDz-fv8}NU2Y=guVCN|hz
zf<@&PMo^ZZwJIhyScKNr!U$UBx}h5SxOzBRw+GC;;H`uEC<d34#=w}EG6t&bwHD)`
zWebM_EiC|`K@n{k@BD6eIUZjzHd0j)v#Q=|+eIH!=|>vc0rgWB6%6%`TMJ`)swkMs
z)mm6U5ml=(ut1^;j$)#R-SV0g=da(z?D;Ei#eZkf=Ar0pYN%rTtf!FXAvaG<p3XAP
zd73<aW&bXM_q-oLC0cQiggqC$YmhS--RCgdxbwvV1xC2shH>mnm<tpb;Z2krocRh4
zGD>&Bl3c9i%;qyH*;+s(B?OG>)Sc=SO7+&A>Me+Be&n30`O$c)KHT3&tB7~UM4f(r
zM5SXj45tk4ju4cyX(%zNoKHhJM^G-Np<Ey+kEfwLMo^wiLwSOrJe`K}6hZlB8p=0t
z%G-JHc@omcc@U#$zf48DMbR#&qRmHXe@WBLXXenG*>HfE+4?h=$%>VZj=?Rt9f%r?
z$MHts%hz{Ho}#0GTp71Xm?tFyjOdL!(;E=gEI4vx;s`LJg)@=mErrjw5$}(YX~O46
zZa-e`zNB}2Fe9rJm8it1Y7>-f;ntH@C>Ma<bcfifN8KT|_o}biMzH=nn4SLek;<8|
zjUasHAe(z2T`&n>047|_C*WJm!F0hSd;ys7_~_`1J0VBe0twG1Av{A6PLQEzu<ZOJ
zn9bek4gv*Ccsd2&DVQ#pgf9RSPEh|2;QIH+B=zqAuYZ3`QvVL{`uE2q_3r?$e}4p1
zR0w-@$i*Db?1=Y0=KnsspcCd|ny12fK3hXcPmlk7JO%0L@$)f=^yEk+)S0t{0a2Fn
zT*!)<Wj03AoXj0*PUaArGe?#)0LkLmjbw2QAUQd5BRLr#6)Ye2_xDXl`ri&itGq&z
zgVSf*5UWS*F#hQg+r~iy;*-gUQ7N8Jmg2cB#eeP@`-{1aGV=-qFwj)#W>e6Ow=2u1
zC<)$^d4%<yonYxiklP7voCpd#0WWiIxx|aw9f)S>N|S?VZd_@`Gwt@5UuLu5{Z}f>
z<7UI{?-Brs3xgGY;h2U6Bn=BE1}prjV;W8&X*hLau)@D}Ov7798s0iFjKfDaya92@
zrKkw*0$WCvB#XHl$zpCL`8AjwyO2C^dn?Ij$LxwB+4b2myJA~*{T{7Z&<+yFWWpgd
zXvPWwn3VT`r{EMYqm<HpqvupImE%Vp$LM3?6Sx3AELt(Ozk|&k#V7ltfL&jJ`V6z~
zKHY%@?kxk9Xqcb(=`JsD{~Dk~y^oAfc9H@6<BaMniw&XL!QG&`@85;WzGl;XkuF^J
zWtz?_aiOyMJ=H!p;|@?h(E|$TXa!VuI)SoNfU;A(vbPf`dkawZ7O70@jpY;Fv4CD*
zqssVeCzNNXYn}3%U06<`EN`7y-hx=>M=oix`O>YMT}mI%zb)II*~hD|bT_^l@_v=(
zZ7=P9mq4Y2;)K2H^RHh6%@X}$K9#(ZCR-&*FkHVcX{M6=H^?PP?YwTf!-ZC(`rd7h
z7<O?lr3@(R6FcnEC@Ee**bfO!mG&jff4t1g+;U`xSE(2FGOB=*t{LwTg`#Rso>^H5
zEvh6fw+zx=NEcAla(S^hItrX-(OR@TUkfz*v|2~N=cr(!EuPL<tD`8E@JJ7?e!~)o
zB1!#l`7ik8n(#wASw{RFnJ`5LxF_D)%=lA(VO>A{6QuLS66ugaRPrnpi_{%|iB6}r
z2E3qBl1lYuPNj|TSu@w6mk<T-1BmewWE|#CG!P?>2LY*=@J$oPkaf4ZYN`<ZcYopz
zyyLh6n9OObm+P2B^wFz~hmTJDX@@6nV<yM+y-?!FDMKV{+JfXVSTnuf=>@@C=w>v9
z8A|#Sk0l+4xUNZjsr1nkhfUdZ<=bg=ZG6oyAPHW?`Iy);?)TjX{`gr3yXuC*Df<YM
z6mk{;J6Yksz-+|?lPR#o>P5KbJaIBc-9c+V9l&=u9sqF~0aod3BBe8+(ivIld?KZD
zpwc;6>0%<K3!u^kS?TdaN{@j`kI70;CQ^C=RC+>IdODHPQ=rmQveGvbDSZP}`i81h
zf2xD;bgTpFBniCUFO%urg7toJ(7T*WZw}VGgzM#>Lc&)bbJkApbLPx8L$J-A+2#ng
zg)`d%!FKG-c8p*<ab`O~u$?-yog&!YIJ3Qhvx$b%mCt&H&Y-EAfN_gr%qL(R-(Pgk
z#4R~2V9!MsRzr?e#=}_9`z6gl2N2?6vGT;JxtlE9Jp_q3moy2tEO(TqgB+Drp4=<`
z#1J<|FzDc^{o<i1YGW%^6%Wf%S&f6&Cn^z&Fzhs@HrB}TPd<qebi|ZaRCBzO<*+E9
zdRi3pSgLV#9FFi_z-L#9OSh_mh$_B{*(iK<Xwt~pOFrfBCCKScBXR;tNe3POB#)vR
zJrT^O9SH_IcMPWL%>=UE0Cc^<3Da%^_i3k%!HzV8sam+JTG&-BK&tr5rW9V-`Ett3
zbs>2JV&U(BPGZ~a-)6iE$9U7Jp3_EQfgbZV$vic3N_u!Mn=7Y%*H9Ck=g0UtdOkIW
zp(plP^etXbxIltf3&AdX7!|wi2{gS|+)062#VPN!tim`D_y4}W4(}aL2=Bm&>y`mU
ze<;D}gu|f(drcVFjAn&DL0%AE1xC4Bc^l0_!W)DsCpWdug|1EFB<%L`_UArKis-NE
zhNX7+^jj6o*2JRUfrX6}J0_jc(xhU*!=sK>=TkAM9&lg4q^JHIyp$?PpP^$rI2vzK
zWvB!5OuDMm^Il4q{Y<@~E^$8G^m=Tk#%Xk%9KD1N%b9N-3(Wy=lPW?U3pb`s3@Bex
zn{BK!uI=Q7wbAo!{=)e-$K;#KB&h3mPB%Sj&KTcaltUN5o*Kiv>a+<0&9)*s6NqHd
zCS)h*emKrIOwi$&WF}&=E{1WwVA`SvnnghL6EO)};QDy}gl}Mq&Zj-8RmT9+Tl6RE
zF1mMNDe1HII{nGIo$g&>T>8d)ufB8lmEMI$ren1C>pORU>0NMWI%a3C?J|DgP^WPk
zX3S8bU%Mw!KpI^3>*0h-ki*I0@Wxe(%YbP(pqW?=YKSx(!c4A)o;UN!^JY%V8{XML
zjlum1%hVK;18!*@AJh92me?sK8(h<R^3rKNA*OZm_d0*+_c|xO*E#sTR=1PbuEO%k
zcVU6A-GZqLrjZxG^aW^v`kQ`y@>N^lTfJcFf@$OhFns}9py47RpMH}U{Ms;(z_V!t
zp1}m35d=>0g3lnY_YC)Pjn|6#^n1nNmzco>o=&3g6eRBytuB~GUI5b<pao9R_8mZN
z-yi6<?~iHPz5}f7`vc$h{V`43cYw8hf8g7`Kc;E>4zRZG4}9DAM=(kA0%}}<O^fUa
z5r$y!TSdE_8Yij%%q1$R3zhXQ!){;0i7Ehdl~u=u%YK7lw+G<Db%x+_%fT~P{>d*h
zoe6SfXy+^IM0E;g3S6lIFqd1sI)Lld!L?o;;PvX@TCWc9dUbHER|j~#Is}fT-Cox9
zt6B6<9B|#r$Ox2axJcA@xk(grbtyom`RY;M;qFnyRi{>s`nyVf(q*Nf`%nRjPUfzP
zPUbd6Cv%LVGgm=pHa%wuIg4XQIg4Y9oW(Ih&dHIZoRKxg0mGOa`{i1`%k5g2>%4$6
z_1AIv4)<{(uKya*I3wME0lT}nsiJfS%eBQ0>&c)c<hl}(@hz7XJFH5B7NRFudqyoc
zPq4y_T2$_RH0gApE5CEL*|Mal7Dz4_+kFo2ZWj|Co!=o?bAmX$dt6}8Yct4nzOdY&
z4%QZoilX1vJ*vr4IXbo<&%aKbL-(tMk~gi8iE^dYfys~0*YB2BS1(nz<1EZ~JEwx(
zyliAm^k1YZPSRS?67Al(hgEEsmc}Jz4xLU=>)_B>A(A!Cah@;mfzZ6~>4Xv5zu{||
zS`MZe<c+88)8J3@ZJ>Riuql|TG7Uc}r0*I?{&WSC_(@h1JDmhfF|59qn2)kQ{a4w&
z%K%O23pAF*9>)}Zsz}$=NxV^!4ycp(oFd&!C-E#rI+NP`N6FO)?kPILik9qhA`!yj
zbA**3`D8lDCveFQxjdgPm*?<Yo?CPI&pl&*mAfXF)t`=MRJDMkg+a@OQ8=4K;S8j3
z231IGW`|k!lbvbi$YkldJVi-xN1-SU?btaLJGWyORO|wPorOt$pJrQ@y<Uod_w<6O
zWR<Q2cF_@%mTr>raY;*xq#HL$#kiy!iX`4XI=(yJ7+YRnhXbc=>8vfcYFiSu-8gG2
ztlA*&RJ;IcU2k9BPvZ=^149mCC>$8Zo2<vp=C!z^XBNDtUkWAe*_j1*=F6W33>Ga;
zLM$pk5*Du6v9M*w0-PNSHz5`kAPJ|g*>P&ij#F@UoVp3Ir~pZL>zW;JZQ1b_oE>l7
zgjiJAB+MwCOr@6oG=j|Yhzj0h?+E^a2)?i*)0HnE%1?>lr*>qz@>7WNw?y!_c4WHp
zw^rrRnx(J5oG${M-xiRl#MrI`288T3x(RM)y8b7MPaM|$IJ5L)R7!2P5>ILC*i><^
z=U4GQpItcTYc}5rpfxOWuNM<AE`Vgq#f2MFzF0$OcyEt0IhGgEXXFXHgA*<%A<W@~
zH<J(+0K%%WQW_OpUEh{>1DGqk?@;3r4Gc`7%dk`1Kp@&9i^4M5^w))GDw`xYvcXzO
zrEw$k;Tp)Qaifz#+eYcXYWlMJ)anNPhTEl$exZD)H-P#o0+cSM5V?TpTR^4pPZzlm
z!9IH;mi7zllej=%@i95_`$<nV`~zHnI$gPPI9K7yRRFmLbBg+|jZe>7ct?sLbvcdH
z942+?BK2k(sRc~xjZLa}>c|wGfcJsRq!0hl-UVaxK79t9w8h*}+G1{zwwRNo{Tj@U
zouoZ*fKh2328^`Nt}isUz0l9p7y8-tg~qlQ`Wf&-t0GHkZPn8L(-{6@PR3u%t@xtw
z5b@h`H-aD2^CyPnPfX9B*p@&2-CzVB{f06-pV+VEM%V5P>$9c1{PlU7JkHl&d3}1v
z@gGa4ICY?|XE9lh^DM|b3oy?G6KzAK4Ti5;z;IvlKcB*^#NV>wAze>Fkfpo#ak}~e
z*4-L_P_e2}@C&Ty1q_ucdJO(!bwBIXAq%(B`YS7*((DIJzSb1UgiO*Tl;TQN-Wu%V
zBDl}eZ4f1#F1QmpmH+?L5r>DD){&)uvE+SNtR6%Y0<ByFGXw+n1;eSne7=Io=MhIx
zc=kujRT#7D^zvOCZGa8&&<?|a%oPmT)e@JGHa-LsDI5%bm8LO5uZRvnq^Pw(vuwLB
z7*J{z*@}jhX@kWT0Q(yvzI+fc{KZ?}4eg~{VWp?+hv!vPAbM&8GXw)N2^gv(T|I<(
z5v?xRu6O_@rE3I6$0h?xb0<CzY#k(9*nQVMn-SpjIS4Zr!cnEcIOK}=c3D=SP^HZ{
zaFp&*pgg`Aa1}?pfX8BaP^3Y$4F~&)%-|Z^5u$Xvi&?<}Y!dp0aEc&K^PJ36TARt%
z<&dU)V}=ozj~fw<4&z4o=WB2xJ60o!i}E5&R`{3IH-uB*UY6E+gp3c0sy?=Ukm;j_
zfn(!V8219ry|8gF#<?+NL@|P7>4E}1Hafkp#y7#P8sG$p9N}0-tv+$g1RER=WmaUd
zIQAV~Pc*jX{Ln<k)fN+gnQ|NEKgw3UY~H~{U9%DRC|_CT-F(`_8)$NGM9_jRXaN?q
z016rzx%-6iw0&a0prbSB_Ccd=-MHO=J{Vp3;Bk;?pk`ni>D2p#uC{&5Fw}wRy^i);
zcYOH5vhk@5pEA7wm@Cn!$%8D7g0SWXrB5{)NvIl2H_zc?<|0NWPK?w$Ia<XeYBxH%
zR3m9-LVU!0g?VCSNSm>3`pBwl#gQ%h(A1ndd`SQlgEk7X1dPA`k++Y2_4fDP#K67s
z(HCxh-FdXxro8`jpNebcQ9XvCsX0-3GlCD+_4ERaq-O}HFwbcC=(z^lWJy;|&WdWJ
z8{CDfA1vwjeWPT}p7jY4rln^I-n3cq&tJdWZ@xs&_9>i!2@n}EeM6xrQXW;DJ%UU@
zV~63OriDf(R(5tFA2j5D3*S{5N2>!h8!Kxo-#L%UvU5IcLlL!R2nO2Q!ce7gn*HCX
z;PG*(Y3(o^#Pkr!sHG3bx80>VLuOuS1Hdk)J}*VvdQlv9SeBkOo{##vu*4tY^5b$-
zhoI8gH%7D3Qi-Ph7(Fs{t5F+)(P@bR)BF<d^E?WZ_wgRrV5o*D7FvSnbks%z$C!nO
zT#O%ad01P)@rk(yCi)z1@75tI6_tq{IZ9ZVHPs8a;AjCLvC{I8L@(x?%1KF7N*gOU
zJ}pgfqN~f#H|Y&a@)XxhD0X&^aE-$b*APrv)<D3o!(>hA>MM2_4q8I6Wa;+Z`ak=;
zSU!Z=E@63&PTBy0#!F8wS-|~0ujG{UmWC)6r_>SXsGmU!<8BupD7h%?FdR;Kh#})+
z%2k^1@Vua;OED)fI;9d%eg2R}D@rA(452L8j&o~<U~tJgER|aYL$H1LbC2&*bWM>=
zF4>5s<RvvqHk7ngmN*`#l#vMCiV2RBHY(}aV!2$h6ifPRnni!{<}Hjr#u24G=>m{g
zUD8XUzifQ&kcqt<alA7{c1RqG?pwA-&R8mTK#nVx<vS)k5-1fyvBvYc<f9w0TYE#`
zPV<6gw9ivxi{*04UL@%?%Y~B>u>+%6<M~{Y=|=2UiU{0kfuA7inj)E;@|H;Hn5#I}
zWg3UXdRdzynOw3INqNh^G|`IO4N%Pqj4lbqQI{17m%!2lh)hffjN*nR=tfi8K(T51
z!6fO`<>$+K`v5!*jTwM}FvSX<U%knq$0&^DWT3b$6vs0#ComG<5S}{R7ilP`J^TJ7
zKGNG`+)V{2LE{zTQ#>aCZYNkpeK!lA?q>n^=>TcePO$rwNpHXmK1k769r7JYGQ7$f
zW!7$1Vf;EQr@w={Oq^y?&*5AB)qq&msy8%NCr{mEfG0DDrkilP4c!%#Fa-LCJ##1L
zAOz}MZ1!mS({2h{EQ%0yP03LQcx^|uZ5DW_<gj%R)uOj{KqbaZ<STZo(se9Y$&MG1
zT|q&SSB^u{nRH!yJW)jq(NJO>vjCZPrp(e4DH(L?T(#i^#EDbbV+ik~N9J0*8Vn5a
zM*8OjchYr<NZ!Fg3E-pmb;Q5RB92Ro)&;^!sH#J<-v+CPFiBYKDrE>rla#xa{w@$k
z!o$6s1t2n)_87tlz6*W~07IFmjzvV)1nm8z812Hf1LY3jRij4l&e&5)5tG=eSrm(H
zo<%bvr_!r9C;{qJI+YweyR8^y%(3ofEGJQ}UB4tM42phpC*ILUT`;9e=s}@mh$&qs
z@)7_;nJd%6MA95OSF70Y+Ti-GYX}tWH1C-Dez~1!4d6$R*kO?L&aD}#C3|%IB0qw@
z$d)}lK4GVgeajF?jc!wpPuaKv*!2Sw)Z66ZV|Te&cF({>7<a??^j#_7tv3*f=6x(a
zc}EF!%M3){a0`l$-W+1v-2%~}-$mk+c3D6+s(>^Nou7RCRtf8V4~UfhrV1aqt73ew
z21vo^<_q6x`vYdb2Slr8&xDWP%V6D*0g<BJ_25&sJpi^<fCPnWq>ykuE+AZkh%Gw<
zd^gX0)OG}f+@3$bPx5z%Y#Bgk>N>~!6o++y_dP<8*16qxa50Db(!*8uz0!S>TROz?
z9ia*R<?Z`=+g;@0YKCs%KG8`V=6#M(*7xc5Y5vY2$8oqUVh2I%sSI?(hKn<KHv3M#
z#TaL5xYoWyvQKgmhWI?gHTC?7eTqXdz+o9KNb7v;(;acap3Vq~L)TrO>P8Fn5r&J?
zy50J87gVs1FG6Bk5IWX6XHJ-JEL_>p_tPhOV!~WZ;mY)mo4$klB-~FFq_XF6=_CCi
z#~nE#3I=YIKGvnt;*SXup>cim3BHeZ&q|mY%9so^AhVCg9_N(vsTwQmEM^;%<SL0I
zqMcRMHAONxq>D^R^|Z&c5>J!rSb})saJze-?vpj|^}esqIYjXd5J0@d1E#vBNG6x}
zhi1i}|J0fx7+$ky9v)<Dml4v0W}{X*6jj+vSBlifa^er4mwFt82I3K0Isg#sz94)M
zDk})S(|KANdk81#cq_#Y!a+WLs%MMka%oS{yrfOosv(4haBfp$1;=-3vk+XaCq)@B
zAl$q{en%Ji(g=c1I}S=Ugs@Q0YU-MTn2;wnRVEN*>d{JFQzR4lBT1liJ7&oQf?&s=
zb=0%Pa=A3RL|*hkL|s!PlS}L9!0J+H@s!k~khBPJ-t)>8xw`y-^hGe?U^JjlM(WvO
zxg6gunv~GpLmXe!jqThT;5bRACJn3+e1wCH`o?HB_e@1Ile@Rr6Mq^LheLl17;tXF
z**iU31Xuc<ULmD1oDF{fPuCR5<k+$yDXAy_=m1Ep?iG`!;2|wiPL@Thizw%@iA8sa
zgU~=a5NK!)XC$1G(X+*Jt?3mNb2g6F$hl<E6r?V_F!0eeLksuhqPdq3yudR)$|9F4
z6#>CZIk9J84d){tqN7EC^Zswj8}vP~X_CeuHd?y?ZV{GY5*0kI)2fUXgv?I<%+h6u
z&KzxhNHK*lQO*-;%|Hw(X$e7^!5Hx83AJW02GXHL9ZMk3K4Y|5N7WX_Mdd+oxE+?H
zMWFGL>o85g_aqX3->~g2&BA!Ozso~$Ycw|bCLD-{pgwM=zp&=xSJtenb|4|Wbpao}
zFkoA90I6{Br`3<}p|cnkbwwJuoVC^SS+P3aVAK$g-O@RJ!2D?7;>f34=~f(IODS-1
zt_NL>v_GgucU198_;%2@SL(b&V4g1kRYdpl;leLuX&Inrf*m~Lt4`SUF1Tj$4waEb
z2h>6bSnAj^g+IkQ1!0{8ES|Q5#4tw=o{tC1qgN=*mJQ9sHdrBLBZPK(7&r|L7B)5y
z=ua^bNSKI(i>GZ%FwG#s=PSVqY3}ugnHun{kVRfngb`Oj+w0$V{ImN@0@aW=LFyK)
zVe9`WWt4q}+Vxi8Vlqk`VN;WC#RRq#G^%XPb{Sg<VxlwtU^EUG?T;$ju(EwQkdOY*
zcwmFPykV-`PLeRXa3WkrVLE9iN(|gS0dA!*J?IVZs*BmU;~Go=eJ8orZ3*a>h0#WO
zUBI2lUDptdDwhqcqs#P2_W+FM!HT<2UUG*n?qlSUw~(3_kueMB7^#$PbJ{gn;94(U
z+uGqxLpPIZSBX(lQ>Q^BXjAEMNf)C~zc%W-EsmlVWHD;#yG=)yI8n>07`20SQJ=gp
z3SEK)$|KfpSiQ#7XzgmFPh4(<Eo#Dqk#_C&S10XjpQB>9l5ljNpzpR(5VbiGpwzm3
z%%5Wa7-9d~xb`Stxt?Wb-gk!C>1LNnKc`K*@d>!G%dU%ZKw580k#%bnK!x)mX`>sD
zkQ<)tdWi?64Yw0PHZ0lYSlnNHXU7w)=Z#(Rz`?^OczD?TM@EGda)BJM@a|8s@=jR!
zAJ@-Q7*uyNls}Wt%+qZ@gKcs{=^MD*_ynIDyGIXF6YY2njh7C{69_9#<Jsd=U3-8o
zL}L<_lY!Nl$j<S`D3#{(jd3<$)n)5rxE_V7+;~3<)f$=_=i8<%a1f87>F%5w1*1tJ
z-8^MRqKP2fKV`z9383FXWq~8gq25I`j)oOOx{*qUfhBghm&%NU6yb0?l@SOlK@3%c
zJK1sMf5=X&gBA@UR+MavRgmDm84WvVGXoPPYCm4neY@a#+{+NY@UUu;&5c)rV5y1s
zi|J9Y<dj>+G!RHy@?B#F7$m*ZjbkHFNEr{!0#`t}R$~+lDTQ|Nl@0?*Ouqig0E46_
zT!J+Kz*3eTeT;m`+c+9t4E0_vBM?@C%k5kvP*@q%JGu-oNcxOzB4HW)U>SdpZX$YM
zXeqgKHe}$vb`xD4W=6t_aJ)cl91Sambd8t}14&H1OiYi0B&S{}rboe&r?mTR^(c7q
zi3*-Wq2s2cd(HGHcyjW^YGx$7h=Q~)k#xOU*KS|U-8ZWVL9Q+z(6hsgixl`My&xv&
zvv!~iG9G0%=C3Kok&akn_&~eA4VG3O=$_u5g%$TpnEyb*3Ii+Z%`JG#?$8K~{(8+G
zrTg}dx8b~;9A1Wpzo+fXv*Zuq?&T758?9N2dLN~Q6}Lb2MRMq{)<AG=r|5*OTaVUU
zQ0rjUR(*fKOM==;KH6}Yskfp&cUu7;)eMX^EN0p=JgA%b;DN<raQHfBMUb&#pCtiw
zfvg4X8IQ{A6OAaX|JJl>KDvh58cef}$Fe&1s<Dr&$D?h7k=poDsPAN?IeE&$9g090
z=u!7Y6z8wkS@=Ym@Hbjp$$qfIE{&1`&C?iKDRNeBnJ(AE${LL38X}C5-pA=vfa>yV
z?8XST;Sa{gBGDQF-q?%*=iwtm%WjQY=#ZC!!cWpQ^OJDP3sGLI{~3pOEEbjUo{9BI
z7L;eguDZ*V<wkq<Ty~@LpQB{Wo=f9(x)P?EpG#ZsgADV+WbG9^Ru+YE-h)ev2i|Vl
zP}Nu!__s1d54MZbB1m_2uRDFR0uL-CNfqLK<RSm`t}?ccc-|x%Z<{`{@`Qq?sCe)=
zhIjq$;|&j9vv@>P#&-K`mx?ngf-p&>2YU{aXE3gmUbjDbRej?1F!k!1@0B%wL|MMd
zq8%?xvRMk+(F{zS*}l?uBg$wM$;<aUJ~%5z)Rdo!AlyYT@+Nx4RecTKU0sec<mFOQ
z)qa$<2Ff)stXY*GJn=(RM5a8?H#~@~_9aLW)zxu~eP~kgiXE->-uW`P$Jo9JTP5qq
zly9yHH3OG9v)boHx&@USt7%33E3dh{5I<#6!MsDd&%E?0@gC=1fz{M}Zt<GhFZGC~
zru$QV_N}6dZS)-Q|MokKp4O@(WuG%oo(&`#+VXrJi-Pt!U~>l7(6#HA5B_HU>3vAn
zev$f&D@pJE!&Zg2W(6z1C_k&;Y)_KFm2Oxe+TI{9!i={xB-K5PDh0n(fnE(<wdrcI
z;3A_q=H(`{wd+uXO5mju<fZrU;!i2}{Q_SX?cDqL?l&_ZrbQbyGqq}WVa|fGda*oM
zhXsEudYza$wCcJD178~Wjlp~23Lrzhdr@P{dfwQ+Dt$YW_Wd>Jz1E3$jh5Aij2D?G
z;c7iGK&W~<3dpP0H?`k38lhVgUokmvLSJ>#FY~yQdheo64<xI8YNHQ3f+~lho9KP4
z!ac=G<Zam>;+az|nsVLkv?w?)S`#%d+KuA!UH+HXM&u9TB~<Hz`KwJB$1FQ^X-vIV
zo)?CGYun*2DWiMdH7twmpZhT3^(MTFJ54DhewBOwous^w7cu)EGYK5t2HdpA&3J7L
z+!Px`0K&Dw@#0g2$%=J{d?U~!1|u|3bQGvMG;6Bd`Q^uV*Rvx%{FsGo8mk8&3xTG~
zV<)=)uq;@{yW9w-+D1~<k5;ZQ=Ede=b~M9~HAa%Hc?QzWPfqc4+^9pxpY3#69OCn`
zwW<l504pPl+1)z~S!p17%D=4?Q6Uy+cij#F*ho?}&sMs0m7a&YLH`S3R`)>K5D4i-
z-Jl^^8@raa39|F<lOj9JXQ=AX7}bpRvyJZJ<^sXPW6Q3{&4m@Y`r~67i?e(X9I9QX
z<5LCMct3pkdbzwo5LO0qeyjZ3N^$e+;sRZ%6)Q>B23Q&QS-O0ON>ybcnW_gH%f}Db
z^P|}U!6GeWkFk2Nvix#;{w~T^S$JPys1!5Fr-WEZ!yil(BMhN5k!148P87#{<azVJ
z|H;v1L}Ojk4}P#K$sc$CwmPV0HFq0gC5@xik8E-D1H)L`NV@tF#&w3LI_of<0f^Ek
zK1!P%%~4dfQB77q+vyJcEuTMxQI?|!TeF(74S|tX<!f<)rmYQC;yTPqou+GC_fJ|#
zvg*Oc^2^Qg`aOc9j4x-C#Gf|&UoU2#5bz?pob%&<S;4<tyg$L@P-P$q%D-(CmtV{6
zY;lxhT4lvbPHeRSRz|U!h3+3HR+6y|05N`L1(uGRC|vx>N|aP>^c>STwMLSxdA8Ag
z!Zz}sl;-^tsA2yUC9H^6m<Fjfku>tjPE-^TiiO9oDfpiteBMu)qd(3)ONFxygtDKI
z*-v!rCkXbBz135wjlw7%ZCsx}{(go0g4HG!Q~d-XV(HE_xifClQk9QZuI2Zy7#bd3
zGVs4_;Mhv6)L~E+{t~U%e0aa-1H+@)F}eWi%q6?#7ps``k@pVJNp7tqbsJz~{I*S^
zO`KvH|Jp>l4#+1f(er(ptuHSxZ>~_EnvR{MRD*21SLg4(Am~K&ibeco1sCUM--yLg
z1Ve2jY3fHSR~)WbzBoO>aMd=FtA4a`{e~@qzbU2fH&DU5W(n_`u=R3zgBd`mX0=C1
z8)73}zL&Em2*$>yqH7-QT;EYF-)sEuR^$&~K3smoG`Z43GE@&%mYdJ0x%)D{WDtK^
z@xOmvynTZqC=Db-`L~tg_E%Ic@#qqs|78VV@!!QaAN|efSxU-opp`vJ3KnBXO9RPO
z{tctRr%o|b>451f!bG(0=OL<-q?$=mCB#Ph{rn?ts9hOUMAhF`isjW3L%^d;c>b3S
z{Px>9WmaA_lfGR_2!M2{`njNZRm+B24YabC&-m87W7jyewLv!C-!CsdTz}-BQ3PjW
zQ}H#AHZI{YU3-9t(?C@3nMce=a~!BrH6F(xHA<_gfNK`6zoctGrt~e88H75>%vCx(
zifE9zk#2m&W3mb`v#i;QtB}6P5E|8>;!#7aME}{b4TeX=@6d<<E6XlHk?fK#2Y)tG
zumqP7C4NPs1Xx+132CEuI7Sna@uCsjM=|WZH?b>%=Fx%=FIa7Kg~3v&$~dK?Fr4DC
z*?8i$&k%~rldTB1yBx`;SoVpPpUfIczf}0EYNOY!UHxpM02k^ux=Jx#q^mg465YgB
z2}a9v6$M@zl=v#gSa~R2-}2WQhq)Dhr6QBNzodYs8;@D3HT{Z~E^ag5@DJicAoh_2
z?LHEsY#m-DS@+AhTwzi98$ty&7A<A9fdgh0JTY<PV~*#mEqlCmu${k(h$5~`TE^;m
zoJy?ARMOyyRBS3TCBVRz?(*_@@cCMt-~3J-uLJBln=EU`mf<fJ2@A4>g;Bx+F5!ZS
zj%B4AMy*;%{}&}>j4Dfa@8fj!1I*MKNTN0X!n45gEc!f)F&^oGe09jeZM6Q%il;RD
z0b|sfcKn*VKaHu#(!W^pJ}g!b;-y3Lc3>vi`@tj3Up`;KxOhZ48~*uf3~wJ7QF)QX
zb$a<OjyAx$Ftn3wgYXgZt0j)KjV~v3mr#<cG>s7)BDz8$9>*A}W6OZ}Rb+`?J&a)E
zGN_w9fXqQ<O($dQ*|Pj|&)8q=!}BUC5UuRMOipO^gU6VQboEdkUwOfH#RHJHYgA98
zbpl5???m-w>mb?YFh0eg;!*;}c+%$}<Q+XAiHt27<dX$n9(xLBkOnP@46yJ-=^oC(
zBe(ecVFY^>N4shde7hGM#G^5kJ?r5QWBjSuSHgSZs6_UyBy%qmK^w$rp3}HXQ@w!l
zrx6wFvR_h`0O)72V^UAMC_!Uvk=z4<%IdzA<m`naXc1YhHPs8SW;O3I1$UUvooXRD
zl@Jis0!_6TP>uHRs$b&r$V1dy)fm>}purC~e#GlFSsRtuUUIq1U$i>aksu%uU7L!!
z8#Y1+{(v<fuJI4nE)TLa3PN}<MzyFI)DSz>4&0+r3_3)+@t#e_#K90jz769T^InE_
z_2geae2g5InZ2BZWFsyQyBP?`y*%1%(`X&sr{WY2c-0!(Rg}Z<G4iHEC+=v0_6_@1
zlDHT8%6ekpyZuJ&-?wrOOe(_aZwO71^0?*6B{;6et|n{(P^?#J9IXyi?#g<gpbEC@
zP;SA-2<Tn2?X}MU$}~<hc@rawv$d-yl9~VrYjw#zj<K|9ci-~|2vh%7KOCp-5<zbe
zY}YZkJ{%)Mdsn&F54DSE?h40{cP=9L>L)u{=gLA1;ox3GPYnSOH%IrPSUbn`dd;Jq
zPu_=!AyY0|^s2uBG`dR-QKY7u6g}N9c0ui3h8Pvys}jBD(axvdP={gcUxMc;0RYx;
zjboEH5qSI$JG!_nnDV_-?Bt|Y8vtWXx9`fEmX^)6O<4W`LaSu(6IfGqeG@@SduNlU
zhJcB8yZAui6?Pq35ddaYFO4FQ_iu{whT%{t=MQPLqU3yKsM1#<U}EeCsWt16f<8)B
z+z<6#KX}>?>IFYuHw;YA-%A8>9WYsT$g3bIH|^utvsL*KAIi$$*XD+t@&b3fj4ptx
zrTYt1&?Rur*kGs%xEz~29@u}&*2rGBVyDuU0Z`T}<*pe#_f_wL8E+5_*3sPbLQv7U
z-o@($LKR$B*OL$%AjMi$p#+0<qy@e=*EQ`ZNDE*Ou8O~`FB!qpFV&kxylxnj{#M>T
zKx_?Eb5({K4v<n^wu7Q?@7D3EzwI=<(}^x^(ng+jS3Llj+RKSiWZm13z4jSGN$8w+
zd*<+N`Arr*Mqw<wyuU|lWI6}t^;wbqaJ!Vzn|m>$tatW$gJ6m}Qis!Wg|F8Or0A+{
z^>q`~-skJ7Arv{{eqT2+_07Is`%Dp{-1F<Er+34zHwY%FBaI(z7yf$vP@FXUC{lm@
zuNMLbk*14)sS*YQyAU;`Gh7KwRMWo<*i!;<S~PduQZ*Saxb-wY;552qRGUNCtYMTP
zsp;ZbkAX*|gG*?s{LS^VUil}AxxI&$$lkqy)@vS7D!SLv5|s=uqV+mKI7O@o))DWC
z<gmj6At<ebbsL5jRaLHJziqEibr3g1Yef!!BZ|!(Dpwb!FD}Pcw1<1_P^WKPNhtqB
zF-Q2R4*ku-UId)!(B3gj<nP`t>?QD}gWgrcME?4cVQ&YS+BaP9OXcp*yX-JCF2;L)
zscHr{{Cb@roFdXhknyr#ZwH?T8eX?cl%QO+>t!sbk*}S;frHC-Jv9U;r+@h_Rg&)7
zU9TUClU7~DOBK{!!s}IkqsVj0OquB-UM~kvl<6+zB}(XA$LrNkI0+M!jA5xDmx#_)
zwRAb%OKLqe1S`j_ZSA$Iv^Mq8rHk>{b}=4dU5v*$#X+pG!ug1mW;8dkt|`G&i~|sz
zv^th}KP_P%hudBEzDult0fH1V2!{zG`i^x?3GQPh0N#VFHA6a)m4;=9cX%BtT7Vv>
z<5)KM6(f=8>ZSuAd%Fp+>cS_igwjg#mVblDoV@Ea_EZNu4n^{@Q|z#AcHETgY1gx*
zINJ#bQfELlM7!K^V}jRRV}*0P6K1ZycLod>_d5<*-lT@rQ5w}JU5cC7AVxG_Kh+TJ
z?#GR(9)Dd^g43S>K+Y3XCJ5I*Zb0?^>zWcA040E%51<Le$qldt#joM?Y$?uwEd<&h
zuxm<i32XtVD@a-dx?^AicV11sR(rlljh_jp>o(Z`6(tevzSy&+crdmQgfG5}A7f)q
za%CJ?)4Um57=lA%-<agn*np7xAk%m@Hb4%}jRp)C|HcNA;NsY`rFc2E5b1Y%1(C)b
z0aXw}0p8WJYfA98YyhYZmpTBl&t(fHj;-FZB8yfRQO;u%i^UM9qiygjMj~lJU}#Qs
z4sC!4PJun!p<^hTeoe1AQ+TMhV5!TO-_1|8fmuF?+iZD<lKfs&gpSv$5{?l>SrHgm
z)7)np7_tYg7J=?W+ra&A%3H}jv0;(Mstg$fQ$VY<1ZQ4uwhWV~;E|nXWwjvk5Fg!7
zFlD0SLYtdaOzRWQn*b0u#ml)hLpnMO!<01LDa~rnWBAnOO^SD{Zp~^wpjUNiNb=?G
zSmNEeEAxHEXcI81ZA0FK0O_gJG@o)|>gHCiMWB0@bGYwGB>ui(+g+N4@p6Bcmv?_-
zWtni~S@nirK@unPf45+fUEsC2)Crc!9pSZ@cwcxk48<K@i-z@xH-X@t;<ZRbzjzB4
z*)?8|OPyJL0SEaWDrtVz(M`UG;4sfQp|gA!jp{FNL?O7$_prq3d~hebfY4K$P;m8G
zfx}<WVx{}NiW1zDmdatMwU>+=3N@M3oeWkqhs_K<Y{!C(+|tmsX4MM11HVL&I}54@
zv<l)J`2^c5dMM}}78-lhCX22;ZJ&h!1HILvi$mXSF@i(daM8tkXy(|1d%}E*8A<x=
zcLTBNBGN>@qYr-<A)&egOybGk#o6H(>B;{5J=^O1Yuf&oaj3?>zY|IjNL_W9Y}-K(
zi@ft-1PAf?NA6JQ8B^CJCfTRZ#mU$v5~jfqmL-tuM4(9Zwt-F{UR~~%O_aS9I)t_B
zr_OapMGwtoQ^hzGbzemf6~DDY2Y}mM(M8CK%PAA=vFM_Zwpr*Qusbb!D5>-4qJ}^;
zm9+^$?G)wPb$7@BEuu`GYano07=b6+r`XP7hQX;uY<n5lI)B^gn_s#(`)Wd)TURsM
zK~9}G2Z2^<&A(&iKgxAi*C9v;NABoRgG{;}Wwz(9DH|R}tSb%R{S8_q(CnANa=>q%
z;c5|}OK&|3<%U!{$Cohwfuaxw(u<q_Whc0~1uv!zjpP#6?f6PsYj+E~xUg@b-1Tbb
z5ZB#itk`Es0Q^82ZDn7f8)#P`Zr(-F3WN62UIW3th+=UGMfilC^EH~**H{dGuxt3i
z4!bl;3LJf7r7$;PcG6tAP&z2@e?YMh%@rBLA4m1v$LSN|G9HbQoG4R+Mj5x^55{{t
z(HhFu7}WP=$j4EZJbYxhM70*OzBJDYk-P{tyfb}il?WBge<!I|L`BU0N5DkMnpL~T
z;_!}-Z7bU{RJNz$&sS|UNub0~l=FD<VjBL|gLGf)q*qb{E|!V{_FR6Y^Pkh=fo1v6
zyErWZj!@3>Rt2TX0)hXJ8q?>PzQnwgVV+vLJb9^P)M*N%w7v#$m5Pe!{p@{6<ez13
z=3bbry)vKjS9PVCN_>r13%`{S1$$dThgWK<7iNB`eibTr8S{DXBTCrXt1kytj2oqj
z`9FSw0sPFn`p-%Cj`;2`a{n%W_{xgfN-UWq>{(}4|I}`4{L22_7_J%HalMZVjb-y&
z{p#56b9i^VXl;|6>9y8B)t3B8QOoY{qgBKUjNVxNS?)=!Lku9@>*ci(hCsLT8tOhQ
zXMVC(l4x+0U*$H%GQhj`&%b_cAXZw_Rc&&v#7uK-)HJW-ep}X&NAGks7Zr$^Tr{p8
zKzwyc3hsIfI%wl|ov#VPDTu>tLzhrm_cWAYI1%OjKV%W5bx%YYh7(al$$_jv822QI
zAdrH(iqGaVEWct}Umw>5;J8oe+^2o+Q!MveI``W?_gfVA{D_=xZrbJM0YL78Ro8=8
z4WeidgKc^j#R4UGTvSykSj>-t`(1+9hk}>gOGwm7Q}==?1E$I;MH$OE*vWKnNCh0I
zSAU#FbD;gwJ7bP2=f7jc^Kbc7S>1<{Qv1P3g>|QQ!a`w$tP~r6Mk*I$*9lc?;o2J*
zUqr1_AvPh7XvP=djPY(jn3PjPQ{=K9uU=i${WghUr;Ij2j>*Pnb)K&mM@PZqW-~jb
zi&atNJcSa1mlQ_@6Q9rF_s2=28MYB^jsi-BI_L}Tj<9Nb=F;n|8%WSGbJH=?>6j68
z%-wX%bvot*9Sb)d3!RPyLC3M1j$@sUV}gzoHytNB9VY}Gr*1k<bvjN7I^MYHc%#$t
z2Cw6B9>h*c1b{}12qvl7x~kdg)oe*>a#uCEUQG^Cvq{rU%n0*ejiL0Ltsg<GQL<vY
za(ej@-H55SwJ&NL0;1oB&yP_NpwnF2)xE^(2}YTOqJMJ~pc|>yQqnfhIN~>S<~M!n
zHz@8|a73i$mK|pPj3Zu<i5ER$V99FxTj*YbYAeZ=5G*~hqiO0>5<8L|Gtx}&5J4jd
zKI5D2Y7dt5>2(&~6DPq9Fmh1SJQF36GRgp>NP_3mpZhQ=<O(4D6QM_;d@njkE*}pg
z5b;aW!E>qnMj&EKV8SCd3l=!1Hll-FO#{r%oCMY&jpKFsL=|&#G=Z3tBUQ}F5mC%@
zF+1YlqWSaL+Fg_arifEQokX4<KOavfvIHB5EJN`k^N0QYeO;S^XF{LM0w<?o2I0<9
z2|XzboK`Bdn>CJoGOg4TgH-oioll#qb3?Aqfw`KojWBr*W~ZBbtgw!8vnQtwLMjNH
zBm^1_0b0aj?w~;!_aulQq=LXnLZHzQphX-X9c}3%O6#78G7Kl;%t^$VM#LFGgu}}?
z8@!4$+>1B_vpHRWwC+fdVMr0DE*egU5>C+y0w)Q9MnixW;m|f5hHb+E-8LMY+J=Ls
zZ8+fDhJ#bvaL}|32YlObaB3S4nzrG9ZyOGQOVu1!On9{{hZuN&qQ<Z@Q~~-D$=D}z
zBJ0V#!+J8uu$~b~DTA^q^5U3Axj0rS7sm)n{Ls%~wC6g&n{Z6$I*gle<R*<yj)Z)&
zlG?l@qz(k>P@VH%H$L4HEbW7VYllN~43~tEn=l(*UpB6d8%4)gPYax0E3p1{u<uvN
zL|NKFs{Pgymuzg!9Wzh`>MNsUTAYRXZZ~zEGNM+JjX%TmJI^IeJbmOoV>Iz3k^8*R
z#8W};vp*A0_~d`?8T*UTR!Td@Qid9YRxoo>FdHfWHc(Gd65OGwODoA){tYL~No2V~
zmcz*k5?P^;0SgH9_Z4t5UiN4?DGU`#@rYv0sjPXIHAk@)RMw))3d>>P3W}~GrI9>C
zLHr443hojT8MjmT_!F3NLE&4pWD7VGF>#Ad;ue_1ol^KtTe4F)6ESh8oy47D68DzE
z_qHW_3uhuG?rkS=Z^4Nx4|v$F76E?RLA903)aBn8V&UfXIKwxpJ#$H4KcW>78rYt>
zssP%|RuI{I7w_}gn*0V7067S6128%k4iXj$`vOUvFV-|xZYaMqi#3Lif0ejKVwtP0
z+b^7W!qewY^m&Iq$I};1^hJjr_<845<1Qe4C#0Qh-Q*L3axte-F6JubVh*N!2xi9=
zN@3ShiYG8-Ow8E0ld&--V`E~*#+{6fp&9$c%U?S+6QHU4_VUD^UMN<|$&r&1VLlKd
z0vjD;9)exJzhC&{XY~||=J%lVsTf13VpKT4RSp<zJw$Ar_%v9}Z}ovNb=3Aej@Adz
zB&ZjoFQ@@+F`JWE42NTMey{L8$)N558P#N!2BG>v-U)sEZh3W8n~2SfCf??^_B60A
zU*+SjwG=<pX3w`#ll{_TT4z)JB8|SjTPiQDnSRr{PAb3F$H<5PZtu0{k0_ChuWk^h
z|MJFx7o(7ifwqMP0-$YS)z$zD$6gJV2$naJz5966L@w7gb<xWo>MP7P%6q9QyVfLE
zO!9$=F`4Sm{S<0|o{bpKD*M0tR^$B7V**qxxjj{jBY&$^h(|?zI~k}g((2Fhs<bMk
zTlUmw#v~0R_0y%tZf5Cpn9Yw;UG=OcFxT~nd8jm4UMMlehE#oQ@k?VN9vZ81+N(aE
zvU?YuA&SJ|p*L+{jmL+a8VDXLj}H9hmnLR;&e&Gp^0hxIM-F;lZljy7xv}P`KdTVk
z!e7PGk2;#)RP*KHqGxN*KYhdwsv$4SzJe9KAE`zb;Tlw>d!d0-eywoIE7sav(;WY{
zJNWev&GBRVn8}0S(!2k#RS~RtYb4HPE6;foXVf>R4T_)4XOUIrK9A!I^YTbHVeU8Q
zqzUtHWi0Qa-oX7|?RjrY^}IK6pqaRpm;c#(`WvHH>8>*p<BwH(zU@R0F3s{LMkBS3
z^N*jDSi3L_x8mNKr!ZUB*KQlmglkWRt<Hs8?eikt2FjU~C433#E09+rfQhrc+Q9!R
z)5p#BnTT?~FZLPpR_S)fQ_sI@QE~9nd+(6$Gp{;2UTDqp;vq}-n}^_mal__17@*+x
zIoPnG5C>QCoX>cNw^rT|_$`6|N+~s#;&gQ=9{3CElqycTVF6FBa`<+Gpq#c_N+=`y
zJhER!t}df2^Xge#mB5?s$1vVAuZ;LAXCd#XcOk1V+45DtUxj7dt!VvMQyyPfW2(m&
zE@hbZ_`-bm@EmO7Ea;F06&XNIo7evkP(Fw9MMS@j=u4<?<^RU{f--TPeNwvvG*3Ea
zc0s-{@}-e4jC{`Zt$Ox8^5=hgR~cJJD^V_ac0DqYd3=ObJa`<T?6+092uW1NcKd9X
z=Aw*;Ng{7Y&JpP|AwZL=*!zmTNcDC5lUE(c${V-Tt51{ls?%fvsmxbdwBvO`Hb6=R
zM<jFv1e(?w!ml9w7D6QVMhvE_C|<tb?ZgG<fW%eb;&2xs($d;MrI!{yw-8FJPyD;9
zOPIc4qkvIrYEmGY6<fcwt<)>Qo(PVL$Xbx{4G&_geW{~}whYn?4UHh;^NrtF`4!bU
z|Kjas6Iq6s)V4(m_{FoglnP|i_X6+-osU1X#+f<~u*#-ZC{Y3`>s~$kHGT4UGxC&)
zp+(tqNS_jq54H+&>e{XiEp@%F5i)g;vOO6|U2_uqD(0ORc=C#E^c?X2_B(XS6jNNH
zgrQ#=D(Tth%sbx~U!v!-m&5=6eFsg3^33K8nUJ0_CMn7@H!Yj^wkkc&|LJ|;{Wk%?
z6I2Zfyym>7pF6!GNkK?(zXMB7ytXHd&j<<QOBj(Uzf?h<43b?LHKz#aj^dctjttbD
z8A){)hM49Kh)D6J5+H^z_=bnv;t-pfZW~j#Cb{?V-ERQNxNoN-8TtKGRO7y(Yu-l2
zX}JDdyZ<#QVjSS3B>!EVdq6@)dil6C{k>gzK+4AZ^ti-L*_8)K@AT=hN?*294^pq~
z-P_agcXv?@X#w?B>8W|CE-5lKo$m%EH80Tv1*f6g6~)Vcxt=F53BxX^QzHJR-XuP8
znr@|2LjQ*DBR-LXek37@llL2;$>=o%<^3z71Z3=<Wh_~B@0!y5Ebjx8KX~mAuX=MH
zEri-0w_`71Eyj@Hz8qn-ci_A-O`i0d_MdQI!8$C$oE5x0Bn8X5s}~d2A0^LgMqk6I
z3_^wS3mMfLMpe_W3FDY$2mGv(T1zE{GL57bP?p_~=et^{MC2>}kGRWEEQ+moo#6c$
zSvyqRLY-Fy!mBpAdD*5q1gngNjnfdZGBuUMZ_vcF$2#>c%XAglepPs3Gw%AA!oN?%
zy>7emb~W}8K1Qj$+TK}9i|n_^m)=9TR-|}$`@&1OSn<On6Teiq_j$jvc=vx=q20RK
zC2a?k{rK&>GRe#p<FXT|^0Zw`@4NU@iP0b{-Orz7tjAfvb5^=v7>mo}%;&kpW8Ka?
zo)Ne6e|f(@M5_lcO_-NIr2BZyb7h~if4zE~iypdY%q)=evO6x~-Oj}Xd)v%;zIV-d
v?c}Y@nyp_wJ%mLNW-Ksem|IgK;QdhClhx+)?>x-oVQq&0>(&1QWA0|jli>sw

literal 0
HcmV?d00001

diff --git a/0.16/ops.html b/0.16/ops.html
new file mode 100644
index 00000000000..64b2e0a3bf0
--- /dev/null
+++ b/0.16/ops.html
@@ -0,0 +1,858 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Operators &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="batched_nms" href="generated/torchvision.ops.batched_nms.html" />
+    <link rel="prev" title="save_image" href="generated/torchvision.utils.save_image.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Operators</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/ops.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="operators">
+<span id="ops"></span><h1>Operators<a class="headerlink" href="#operators" title="Permalink to this heading">¶</a></h1>
+<p><code class="xref py py-mod docutils literal notranslate"><span class="pre">torchvision.ops</span></code> implements operators, losses and layers that are specific for Computer Vision.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>All operators have native support for TorchScript.</p>
+</div>
+<section id="detection-and-segmentation-operators">
+<h2>Detection and Segmentation Operators<a class="headerlink" href="#detection-and-segmentation-operators" title="Permalink to this heading">¶</a></h2>
+<p>The below operators perform pre-processing as well as post-processing required in object detection and segmentation models.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.batched_nms.html#torchvision.ops.batched_nms" title="torchvision.ops.batched_nms"><code class="xref py py-obj docutils literal notranslate"><span class="pre">batched_nms</span></code></a>(boxes, scores, idxs, iou_threshold)</p></td>
+<td><p>Performs non-maximum suppression in a batched fashion.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.masks_to_boxes.html#torchvision.ops.masks_to_boxes" title="torchvision.ops.masks_to_boxes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">masks_to_boxes</span></code></a>(masks)</p></td>
+<td><p>Compute the bounding boxes around the provided masks.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.nms.html#torchvision.ops.nms" title="torchvision.ops.nms"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nms</span></code></a>(boxes, scores, iou_threshold)</p></td>
+<td><p>Performs non-maximum suppression (NMS) on the boxes according to their intersection-over-union (IoU).</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.roi_align.html#torchvision.ops.roi_align" title="torchvision.ops.roi_align"><code class="xref py py-obj docutils literal notranslate"><span class="pre">roi_align</span></code></a>(input, boxes, output_size[, ...])</p></td>
+<td><p>Performs Region of Interest (RoI) Align operator with average pooling, as described in Mask R-CNN.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.roi_pool.html#torchvision.ops.roi_pool" title="torchvision.ops.roi_pool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">roi_pool</span></code></a>(input, boxes, output_size[, ...])</p></td>
+<td><p>Performs Region of Interest (RoI) Pool operator described in Fast R-CNN</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.ps_roi_align.html#torchvision.ops.ps_roi_align" title="torchvision.ops.ps_roi_align"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ps_roi_align</span></code></a>(input, boxes, output_size[, ...])</p></td>
+<td><p>Performs Position-Sensitive Region of Interest (RoI) Align operator mentioned in Light-Head R-CNN.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.ps_roi_pool.html#torchvision.ops.ps_roi_pool" title="torchvision.ops.ps_roi_pool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ps_roi_pool</span></code></a>(input, boxes, output_size[, ...])</p></td>
+<td><p>Performs Position-Sensitive Region of Interest (RoI) Pool operator described in R-FCN</p></td>
+</tr>
+</tbody>
+</table>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.FeaturePyramidNetwork.html#torchvision.ops.FeaturePyramidNetwork" title="torchvision.ops.FeaturePyramidNetwork"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FeaturePyramidNetwork</span></code></a>(in_channels_list, ...)</p></td>
+<td><p>Module that adds a FPN from on top of a set of feature maps.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.MultiScaleRoIAlign.html#torchvision.ops.MultiScaleRoIAlign" title="torchvision.ops.MultiScaleRoIAlign"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MultiScaleRoIAlign</span></code></a>(featmap_names, ...[, ...])</p></td>
+<td><p>Multi-scale RoIAlign pooling, which is useful for detection with or without FPN.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.RoIAlign.html#torchvision.ops.RoIAlign" title="torchvision.ops.RoIAlign"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RoIAlign</span></code></a>(output_size, spatial_scale, ...[, ...])</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.roi_align.html#torchvision.ops.roi_align" title="torchvision.ops.roi_align"><code class="xref py py-func docutils literal notranslate"><span class="pre">roi_align()</span></code></a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.RoIPool.html#torchvision.ops.RoIPool" title="torchvision.ops.RoIPool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RoIPool</span></code></a>(output_size, spatial_scale)</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.roi_pool.html#torchvision.ops.roi_pool" title="torchvision.ops.roi_pool"><code class="xref py py-func docutils literal notranslate"><span class="pre">roi_pool()</span></code></a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.PSRoIAlign.html#torchvision.ops.PSRoIAlign" title="torchvision.ops.PSRoIAlign"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PSRoIAlign</span></code></a>(output_size, spatial_scale, ...)</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.ps_roi_align.html#torchvision.ops.ps_roi_align" title="torchvision.ops.ps_roi_align"><code class="xref py py-func docutils literal notranslate"><span class="pre">ps_roi_align()</span></code></a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.PSRoIPool.html#torchvision.ops.PSRoIPool" title="torchvision.ops.PSRoIPool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PSRoIPool</span></code></a>(output_size, spatial_scale)</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.ps_roi_pool.html#torchvision.ops.ps_roi_pool" title="torchvision.ops.ps_roi_pool"><code class="xref py py-func docutils literal notranslate"><span class="pre">ps_roi_pool()</span></code></a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="box-operators">
+<h2>Box Operators<a class="headerlink" href="#box-operators" title="Permalink to this heading">¶</a></h2>
+<p>These utility functions perform various operations on bounding boxes.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.box_area.html#torchvision.ops.box_area" title="torchvision.ops.box_area"><code class="xref py py-obj docutils literal notranslate"><span class="pre">box_area</span></code></a>(boxes)</p></td>
+<td><p>Computes the area of a set of bounding boxes, which are specified by their (x1, y1, x2, y2) coordinates.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.box_convert.html#torchvision.ops.box_convert" title="torchvision.ops.box_convert"><code class="xref py py-obj docutils literal notranslate"><span class="pre">box_convert</span></code></a>(boxes, in_fmt, out_fmt)</p></td>
+<td><p>Converts boxes from given in_fmt to out_fmt.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.box_iou.html#torchvision.ops.box_iou" title="torchvision.ops.box_iou"><code class="xref py py-obj docutils literal notranslate"><span class="pre">box_iou</span></code></a>(boxes1, boxes2)</p></td>
+<td><p>Return intersection-over-union (Jaccard index) between two sets of boxes.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.clip_boxes_to_image.html#torchvision.ops.clip_boxes_to_image" title="torchvision.ops.clip_boxes_to_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">clip_boxes_to_image</span></code></a>(boxes, size)</p></td>
+<td><p>Clip boxes so that they lie inside an image of size <cite>size</cite>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.complete_box_iou.html#torchvision.ops.complete_box_iou" title="torchvision.ops.complete_box_iou"><code class="xref py py-obj docutils literal notranslate"><span class="pre">complete_box_iou</span></code></a>(boxes1, boxes2[, eps])</p></td>
+<td><p>Return complete intersection-over-union (Jaccard index) between two sets of boxes.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.distance_box_iou.html#torchvision.ops.distance_box_iou" title="torchvision.ops.distance_box_iou"><code class="xref py py-obj docutils literal notranslate"><span class="pre">distance_box_iou</span></code></a>(boxes1, boxes2[, eps])</p></td>
+<td><p>Return distance intersection-over-union (Jaccard index) between two sets of boxes.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.generalized_box_iou.html#torchvision.ops.generalized_box_iou" title="torchvision.ops.generalized_box_iou"><code class="xref py py-obj docutils literal notranslate"><span class="pre">generalized_box_iou</span></code></a>(boxes1, boxes2)</p></td>
+<td><p>Return generalized intersection-over-union (Jaccard index) between two sets of boxes.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.remove_small_boxes.html#torchvision.ops.remove_small_boxes" title="torchvision.ops.remove_small_boxes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">remove_small_boxes</span></code></a>(boxes, min_size)</p></td>
+<td><p>Remove boxes which contains at least one side smaller than min_size.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="losses">
+<h2>Losses<a class="headerlink" href="#losses" title="Permalink to this heading">¶</a></h2>
+<p>The following vision-specific loss functions are implemented:</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.complete_box_iou_loss.html#torchvision.ops.complete_box_iou_loss" title="torchvision.ops.complete_box_iou_loss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">complete_box_iou_loss</span></code></a>(boxes1, boxes2[, ...])</p></td>
+<td><p>Gradient-friendly IoU loss with an additional penalty that is non-zero when the boxes do not overlap.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.distance_box_iou_loss.html#torchvision.ops.distance_box_iou_loss" title="torchvision.ops.distance_box_iou_loss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">distance_box_iou_loss</span></code></a>(boxes1, boxes2[, ...])</p></td>
+<td><p>Gradient-friendly IoU loss with an additional penalty that is non-zero when the distance between boxes' centers isn't zero.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.generalized_box_iou_loss.html#torchvision.ops.generalized_box_iou_loss" title="torchvision.ops.generalized_box_iou_loss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">generalized_box_iou_loss</span></code></a>(boxes1, boxes2[, ...])</p></td>
+<td><p>Gradient-friendly IoU loss with an additional penalty that is non-zero when the boxes do not overlap and scales with the size of their smallest enclosing box.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.sigmoid_focal_loss.html#torchvision.ops.sigmoid_focal_loss" title="torchvision.ops.sigmoid_focal_loss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">sigmoid_focal_loss</span></code></a>(inputs, targets[, alpha, ...])</p></td>
+<td><p>Loss used in RetinaNet for dense detection: <a class="reference external" href="https://arxiv.org/abs/1708.02002">https://arxiv.org/abs/1708.02002</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="layers">
+<h2>Layers<a class="headerlink" href="#layers" title="Permalink to this heading">¶</a></h2>
+<p>TorchVision provides commonly used building blocks as layers:</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.Conv2dNormActivation.html#torchvision.ops.Conv2dNormActivation" title="torchvision.ops.Conv2dNormActivation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Conv2dNormActivation</span></code></a>(in_channels, ...)</p></td>
+<td><p>Configurable block used for Convolution2d-Normalization-Activation blocks.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.Conv3dNormActivation.html#torchvision.ops.Conv3dNormActivation" title="torchvision.ops.Conv3dNormActivation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Conv3dNormActivation</span></code></a>(in_channels, ...)</p></td>
+<td><p>Configurable block used for Convolution3d-Normalization-Activation blocks.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.DeformConv2d.html#torchvision.ops.DeformConv2d" title="torchvision.ops.DeformConv2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DeformConv2d</span></code></a>(in_channels, out_channels, ...)</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.deform_conv2d.html#torchvision.ops.deform_conv2d" title="torchvision.ops.deform_conv2d"><code class="xref py py-func docutils literal notranslate"><span class="pre">deform_conv2d()</span></code></a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.DropBlock2d.html#torchvision.ops.DropBlock2d" title="torchvision.ops.DropBlock2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DropBlock2d</span></code></a>(p, block_size[, inplace, eps])</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.drop_block2d.html#torchvision.ops.drop_block2d" title="torchvision.ops.drop_block2d"><code class="xref py py-func docutils literal notranslate"><span class="pre">drop_block2d()</span></code></a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.DropBlock3d.html#torchvision.ops.DropBlock3d" title="torchvision.ops.DropBlock3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DropBlock3d</span></code></a>(p, block_size[, inplace, eps])</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.drop_block3d.html#torchvision.ops.drop_block3d" title="torchvision.ops.drop_block3d"><code class="xref py py-func docutils literal notranslate"><span class="pre">drop_block3d()</span></code></a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.FrozenBatchNorm2d.html#torchvision.ops.FrozenBatchNorm2d" title="torchvision.ops.FrozenBatchNorm2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FrozenBatchNorm2d</span></code></a>(num_features[, eps])</p></td>
+<td><p>BatchNorm2d where the batch statistics and the affine parameters are fixed</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.MLP.html#torchvision.ops.MLP" title="torchvision.ops.MLP"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MLP</span></code></a>(in_channels, hidden_channels, ...)</p></td>
+<td><p>This block implements the multi-layer perceptron (MLP) module.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.Permute.html#torchvision.ops.Permute" title="torchvision.ops.Permute"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Permute</span></code></a>(dims)</p></td>
+<td><p>This module returns a view of the tensor input with its dimensions permuted.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.SqueezeExcitation.html#torchvision.ops.SqueezeExcitation" title="torchvision.ops.SqueezeExcitation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SqueezeExcitation</span></code></a>(input_channels, ...)</p></td>
+<td><p>This block implements the Squeeze-and-Excitation block from <a class="reference external" href="https://arxiv.org/abs/1709.01507">https://arxiv.org/abs/1709.01507</a> (see Fig.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.StochasticDepth.html#torchvision.ops.StochasticDepth" title="torchvision.ops.StochasticDepth"><code class="xref py py-obj docutils literal notranslate"><span class="pre">StochasticDepth</span></code></a>(p, mode)</p></td>
+<td><p>See <a class="reference internal" href="generated/torchvision.ops.stochastic_depth.html#torchvision.ops.stochastic_depth" title="torchvision.ops.stochastic_depth"><code class="xref py py-func docutils literal notranslate"><span class="pre">stochastic_depth()</span></code></a>.</p></td>
+</tr>
+</tbody>
+</table>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.deform_conv2d.html#torchvision.ops.deform_conv2d" title="torchvision.ops.deform_conv2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">deform_conv2d</span></code></a>(input, offset, weight[, bias, ...])</p></td>
+<td><p>Performs Deformable Convolution v2, described in <a class="reference external" href="https://arxiv.org/abs/1811.11168">Deformable ConvNets v2: More Deformable, Better Results</a> if <code class="xref py py-attr docutils literal notranslate"><span class="pre">mask</span></code> is not <code class="docutils literal notranslate"><span class="pre">None</span></code> and Performs Deformable Convolution, described in <a class="reference external" href="https://arxiv.org/abs/1703.06211">Deformable Convolutional Networks</a> if <code class="xref py py-attr docutils literal notranslate"><span class="pre">mask</span></code> is <code class="docutils literal notranslate"><span class="pre">None</span></code>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.drop_block2d.html#torchvision.ops.drop_block2d" title="torchvision.ops.drop_block2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">drop_block2d</span></code></a>(input, p, block_size[, ...])</p></td>
+<td><p>Implements DropBlock2d from <cite>&quot;DropBlock: A regularization method for convolutional networks&quot; &lt;https://arxiv.org/abs/1810.12890&gt;</cite>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.ops.drop_block3d.html#torchvision.ops.drop_block3d" title="torchvision.ops.drop_block3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">drop_block3d</span></code></a>(input, p, block_size[, ...])</p></td>
+<td><p>Implements DropBlock3d from <cite>&quot;DropBlock: A regularization method for convolutional networks&quot; &lt;https://arxiv.org/abs/1810.12890&gt;</cite>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.ops.stochastic_depth.html#torchvision.ops.stochastic_depth" title="torchvision.ops.stochastic_depth"><code class="xref py py-obj docutils literal notranslate"><span class="pre">stochastic_depth</span></code></a>(input, p, mode[, training])</p></td>
+<td><p>Implements the Stochastic Depth from <a class="reference external" href="https://arxiv.org/abs/1603.09382">&quot;Deep Networks with Stochastic Depth&quot;</a> used for randomly dropping residual branches of residual architectures.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.ops.batched_nms.html" class="btn btn-neutral float-right" title="batched_nms" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.utils.save_image.html" class="btn btn-neutral" title="save_image" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Operators</a><ul>
+<li><a class="reference internal" href="#detection-and-segmentation-operators">Detection and Segmentation Operators</a><ul>
+</ul>
+</li>
+<li><a class="reference internal" href="#box-operators">Box Operators</a></li>
+<li><a class="reference internal" href="#losses">Losses</a></li>
+<li><a class="reference internal" href="#layers">Layers</a><ul>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/py-modindex.html b/0.16/py-modindex.html
new file mode 100644
index 00000000000..6956c37f5f9
--- /dev/null
+++ b/0.16/py-modindex.html
@@ -0,0 +1,688 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Python Module Index &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+    <script>
+      DOCUMENTATION_OPTIONS.COLLAPSE_INDEX = true;
+    </script>
+
+
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Python Module Index</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+
+   <h1>Python Module Index</h1>
+
+   <div class="modindex-jumpbox">
+   <a href="#cap-t"><strong>t</strong></a>
+   </div>
+
+   <table class="indextable modindextable">
+     <tr class="pcap"><td></td><td>&#160;</td><td></td></tr>
+     <tr class="cap" id="cap-t"><td></td><td>
+       <strong>t</strong></td><td></td></tr>
+     <tr>
+       <td></td>
+       <td>
+       <a href="index.html#module-torchvision"><code class="xref">torchvision</code></a></td><td>
+       <em></em></td></tr>
+   </table>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/search.html b/0.16/search.html
new file mode 100644
index 00000000000..4d5bac9d6d2
--- /dev/null
+++ b/0.16/search.html
@@ -0,0 +1,683 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Search &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="#" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="#" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Search</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <noscript>
+  <div id="fallback" class="admonition warning">
+    <p class="last">
+      Please activate JavaScript to enable the search
+      functionality.
+    </p>
+  </div>
+  </noscript>
+
+  
+  <div id="search-results">
+  
+  </div>
+
+             </article>
+             
+            </div>
+            <footer>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+         <script src="_static/searchtools.js"></script>
+         <script src="_static/language_data.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script>
+  <script type="text/javascript" src="searchindex.js" defer></script>
+   
+
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/searchindex.js b/0.16/searchindex.js
new file mode 100644
index 00000000000..783305af9b9
--- /dev/null
+++ b/0.16/searchindex.js
@@ -0,0 +1 @@
+Search.setIndex({"docnames": ["auto_examples/index", "auto_examples/others/index", "auto_examples/others/plot_optical_flow", "auto_examples/others/plot_repurposing_annotations", "auto_examples/others/plot_scripted_tensor_transforms", "auto_examples/others/plot_video_api", "auto_examples/others/plot_visualization_utils", "auto_examples/others/sg_execution_times", "auto_examples/transforms/index", "auto_examples/transforms/plot_custom_transforms", "auto_examples/transforms/plot_custom_tv_tensors", "auto_examples/transforms/plot_cutmix_mixup", "auto_examples/transforms/plot_transforms_e2e", "auto_examples/transforms/plot_transforms_getting_started", "auto_examples/transforms/plot_transforms_illustrations", "auto_examples/transforms/plot_tv_tensors", "auto_examples/transforms/sg_execution_times", "datasets", "feature_extraction", "generated/classification_quant_table", "generated/classification_table", "generated/detection_keypoint_table", "generated/detection_table", "generated/instance_segmentation_table", "generated/segmentation_table", "generated/torchvision.datasets.CIFAR10", "generated/torchvision.datasets.CIFAR100", "generated/torchvision.datasets.CREStereo", "generated/torchvision.datasets.Caltech101", "generated/torchvision.datasets.Caltech256", "generated/torchvision.datasets.CarlaStereo", "generated/torchvision.datasets.CelebA", "generated/torchvision.datasets.Cityscapes", "generated/torchvision.datasets.CocoCaptions", "generated/torchvision.datasets.CocoDetection", "generated/torchvision.datasets.Country211", "generated/torchvision.datasets.DTD", "generated/torchvision.datasets.DatasetFolder", "generated/torchvision.datasets.EMNIST", "generated/torchvision.datasets.ETH3DStereo", "generated/torchvision.datasets.EuroSAT", "generated/torchvision.datasets.FER2013", "generated/torchvision.datasets.FGVCAircraft", "generated/torchvision.datasets.FakeData", "generated/torchvision.datasets.FallingThingsStereo", "generated/torchvision.datasets.FashionMNIST", "generated/torchvision.datasets.Flickr30k", "generated/torchvision.datasets.Flickr8k", "generated/torchvision.datasets.Flowers102", "generated/torchvision.datasets.FlyingChairs", "generated/torchvision.datasets.FlyingThings3D", "generated/torchvision.datasets.Food101", "generated/torchvision.datasets.GTSRB", "generated/torchvision.datasets.HD1K", "generated/torchvision.datasets.HMDB51", "generated/torchvision.datasets.INaturalist", "generated/torchvision.datasets.ImageFolder", "generated/torchvision.datasets.ImageNet", "generated/torchvision.datasets.InStereo2k", "generated/torchvision.datasets.KMNIST", "generated/torchvision.datasets.Kinetics", "generated/torchvision.datasets.Kitti", "generated/torchvision.datasets.Kitti2012Stereo", "generated/torchvision.datasets.Kitti2015Stereo", "generated/torchvision.datasets.KittiFlow", "generated/torchvision.datasets.LFWPairs", "generated/torchvision.datasets.LFWPeople", "generated/torchvision.datasets.LSUN", "generated/torchvision.datasets.MNIST", "generated/torchvision.datasets.Middlebury2014Stereo", "generated/torchvision.datasets.MovingMNIST", "generated/torchvision.datasets.Omniglot", "generated/torchvision.datasets.OxfordIIITPet", "generated/torchvision.datasets.PCAM", "generated/torchvision.datasets.PhotoTour", "generated/torchvision.datasets.Places365", "generated/torchvision.datasets.QMNIST", "generated/torchvision.datasets.RenderedSST2", "generated/torchvision.datasets.SBDataset", "generated/torchvision.datasets.SBU", "generated/torchvision.datasets.SEMEION", "generated/torchvision.datasets.STL10", "generated/torchvision.datasets.SUN397", "generated/torchvision.datasets.SVHN", "generated/torchvision.datasets.SceneFlowStereo", "generated/torchvision.datasets.Sintel", "generated/torchvision.datasets.SintelStereo", "generated/torchvision.datasets.StanfordCars", "generated/torchvision.datasets.UCF101", "generated/torchvision.datasets.USPS", "generated/torchvision.datasets.VOCDetection", "generated/torchvision.datasets.VOCSegmentation", "generated/torchvision.datasets.VisionDataset", "generated/torchvision.datasets.WIDERFace", "generated/torchvision.datasets.wrap_dataset_for_transforms_v2", "generated/torchvision.io.ImageReadMode", "generated/torchvision.io.VideoReader", "generated/torchvision.io.decode_image", "generated/torchvision.io.decode_jpeg", "generated/torchvision.io.decode_png", "generated/torchvision.io.encode_jpeg", "generated/torchvision.io.encode_png", "generated/torchvision.io.read_file", "generated/torchvision.io.read_image", "generated/torchvision.io.read_video", "generated/torchvision.io.read_video_timestamps", "generated/torchvision.io.write_file", "generated/torchvision.io.write_jpeg", "generated/torchvision.io.write_png", "generated/torchvision.io.write_video", "generated/torchvision.models.feature_extraction.create_feature_extractor", "generated/torchvision.models.feature_extraction.get_graph_node_names", "generated/torchvision.models.get_model", "generated/torchvision.models.get_model_weights", "generated/torchvision.models.get_weight", "generated/torchvision.models.list_models", "generated/torchvision.ops.Conv2dNormActivation", "generated/torchvision.ops.Conv3dNormActivation", "generated/torchvision.ops.DeformConv2d", "generated/torchvision.ops.DropBlock2d", "generated/torchvision.ops.DropBlock3d", "generated/torchvision.ops.FeaturePyramidNetwork", "generated/torchvision.ops.FrozenBatchNorm2d", "generated/torchvision.ops.MLP", "generated/torchvision.ops.MultiScaleRoIAlign", "generated/torchvision.ops.PSRoIAlign", "generated/torchvision.ops.PSRoIPool", "generated/torchvision.ops.Permute", "generated/torchvision.ops.RoIAlign", "generated/torchvision.ops.RoIPool", "generated/torchvision.ops.SqueezeExcitation", "generated/torchvision.ops.StochasticDepth", "generated/torchvision.ops.batched_nms", "generated/torchvision.ops.box_area", "generated/torchvision.ops.box_convert", "generated/torchvision.ops.box_iou", "generated/torchvision.ops.clip_boxes_to_image", "generated/torchvision.ops.complete_box_iou", "generated/torchvision.ops.complete_box_iou_loss", "generated/torchvision.ops.deform_conv2d", "generated/torchvision.ops.distance_box_iou", "generated/torchvision.ops.distance_box_iou_loss", "generated/torchvision.ops.drop_block2d", "generated/torchvision.ops.drop_block3d", "generated/torchvision.ops.generalized_box_iou", "generated/torchvision.ops.generalized_box_iou_loss", "generated/torchvision.ops.masks_to_boxes", "generated/torchvision.ops.nms", "generated/torchvision.ops.ps_roi_align", "generated/torchvision.ops.ps_roi_pool", "generated/torchvision.ops.remove_small_boxes", "generated/torchvision.ops.roi_align", "generated/torchvision.ops.roi_pool", "generated/torchvision.ops.sigmoid_focal_loss", "generated/torchvision.ops.stochastic_depth", "generated/torchvision.transforms.AugMix", "generated/torchvision.transforms.AutoAugment", "generated/torchvision.transforms.AutoAugmentPolicy", "generated/torchvision.transforms.CenterCrop", "generated/torchvision.transforms.ColorJitter", "generated/torchvision.transforms.Compose", "generated/torchvision.transforms.ConvertImageDtype", "generated/torchvision.transforms.ElasticTransform", "generated/torchvision.transforms.FiveCrop", "generated/torchvision.transforms.GaussianBlur", "generated/torchvision.transforms.Grayscale", "generated/torchvision.transforms.Lambda", "generated/torchvision.transforms.LinearTransformation", "generated/torchvision.transforms.Normalize", "generated/torchvision.transforms.PILToTensor", "generated/torchvision.transforms.Pad", "generated/torchvision.transforms.RandAugment", "generated/torchvision.transforms.RandomAdjustSharpness", "generated/torchvision.transforms.RandomAffine", "generated/torchvision.transforms.RandomApply", "generated/torchvision.transforms.RandomAutocontrast", "generated/torchvision.transforms.RandomChoice", "generated/torchvision.transforms.RandomCrop", "generated/torchvision.transforms.RandomEqualize", "generated/torchvision.transforms.RandomErasing", "generated/torchvision.transforms.RandomGrayscale", "generated/torchvision.transforms.RandomHorizontalFlip", "generated/torchvision.transforms.RandomInvert", "generated/torchvision.transforms.RandomOrder", "generated/torchvision.transforms.RandomPerspective", "generated/torchvision.transforms.RandomPosterize", "generated/torchvision.transforms.RandomResizedCrop", "generated/torchvision.transforms.RandomRotation", "generated/torchvision.transforms.RandomSolarize", "generated/torchvision.transforms.RandomVerticalFlip", "generated/torchvision.transforms.Resize", "generated/torchvision.transforms.TenCrop", "generated/torchvision.transforms.ToPILImage", "generated/torchvision.transforms.ToTensor", "generated/torchvision.transforms.TrivialAugmentWide", "generated/torchvision.transforms.functional.adjust_brightness", "generated/torchvision.transforms.functional.adjust_contrast", "generated/torchvision.transforms.functional.adjust_gamma", "generated/torchvision.transforms.functional.adjust_hue", "generated/torchvision.transforms.functional.adjust_saturation", "generated/torchvision.transforms.functional.adjust_sharpness", "generated/torchvision.transforms.functional.affine", "generated/torchvision.transforms.functional.autocontrast", "generated/torchvision.transforms.functional.center_crop", "generated/torchvision.transforms.functional.convert_image_dtype", "generated/torchvision.transforms.functional.crop", "generated/torchvision.transforms.functional.equalize", "generated/torchvision.transforms.functional.erase", "generated/torchvision.transforms.functional.five_crop", "generated/torchvision.transforms.functional.gaussian_blur", "generated/torchvision.transforms.functional.get_dimensions", "generated/torchvision.transforms.functional.get_image_num_channels", "generated/torchvision.transforms.functional.get_image_size", "generated/torchvision.transforms.functional.hflip", "generated/torchvision.transforms.functional.invert", "generated/torchvision.transforms.functional.normalize", "generated/torchvision.transforms.functional.pad", "generated/torchvision.transforms.functional.perspective", "generated/torchvision.transforms.functional.pil_to_tensor", "generated/torchvision.transforms.functional.posterize", "generated/torchvision.transforms.functional.resize", "generated/torchvision.transforms.functional.resized_crop", "generated/torchvision.transforms.functional.rgb_to_grayscale", "generated/torchvision.transforms.functional.rotate", "generated/torchvision.transforms.functional.solarize", "generated/torchvision.transforms.functional.ten_crop", "generated/torchvision.transforms.functional.to_grayscale", "generated/torchvision.transforms.functional.to_pil_image", "generated/torchvision.transforms.functional.to_tensor", "generated/torchvision.transforms.functional.vflip", "generated/torchvision.transforms.v2.AugMix", "generated/torchvision.transforms.v2.AutoAugment", "generated/torchvision.transforms.v2.CenterCrop", "generated/torchvision.transforms.v2.ClampBoundingBoxes", "generated/torchvision.transforms.v2.ColorJitter", "generated/torchvision.transforms.v2.Compose", "generated/torchvision.transforms.v2.ConvertBoundingBoxFormat", "generated/torchvision.transforms.v2.ConvertImageDtype", "generated/torchvision.transforms.v2.CutMix", "generated/torchvision.transforms.v2.ElasticTransform", "generated/torchvision.transforms.v2.FiveCrop", "generated/torchvision.transforms.v2.GaussianBlur", "generated/torchvision.transforms.v2.Grayscale", "generated/torchvision.transforms.v2.Lambda", "generated/torchvision.transforms.v2.LinearTransformation", "generated/torchvision.transforms.v2.MixUp", "generated/torchvision.transforms.v2.Normalize", "generated/torchvision.transforms.v2.PILToTensor", "generated/torchvision.transforms.v2.Pad", "generated/torchvision.transforms.v2.RandAugment", "generated/torchvision.transforms.v2.RandomAdjustSharpness", "generated/torchvision.transforms.v2.RandomAffine", "generated/torchvision.transforms.v2.RandomApply", "generated/torchvision.transforms.v2.RandomAutocontrast", "generated/torchvision.transforms.v2.RandomChannelPermutation", "generated/torchvision.transforms.v2.RandomChoice", "generated/torchvision.transforms.v2.RandomCrop", "generated/torchvision.transforms.v2.RandomEqualize", "generated/torchvision.transforms.v2.RandomErasing", "generated/torchvision.transforms.v2.RandomGrayscale", "generated/torchvision.transforms.v2.RandomHorizontalFlip", "generated/torchvision.transforms.v2.RandomInvert", "generated/torchvision.transforms.v2.RandomIoUCrop", "generated/torchvision.transforms.v2.RandomOrder", "generated/torchvision.transforms.v2.RandomPerspective", "generated/torchvision.transforms.v2.RandomPhotometricDistort", "generated/torchvision.transforms.v2.RandomPosterize", "generated/torchvision.transforms.v2.RandomResize", "generated/torchvision.transforms.v2.RandomResizedCrop", "generated/torchvision.transforms.v2.RandomRotation", "generated/torchvision.transforms.v2.RandomShortestSize", "generated/torchvision.transforms.v2.RandomSolarize", "generated/torchvision.transforms.v2.RandomVerticalFlip", "generated/torchvision.transforms.v2.RandomZoomOut", "generated/torchvision.transforms.v2.Resize", "generated/torchvision.transforms.v2.SanitizeBoundingBoxes", "generated/torchvision.transforms.v2.ScaleJitter", "generated/torchvision.transforms.v2.TenCrop", "generated/torchvision.transforms.v2.ToDtype", "generated/torchvision.transforms.v2.ToImage", "generated/torchvision.transforms.v2.ToPILImage", "generated/torchvision.transforms.v2.ToPureTensor", "generated/torchvision.transforms.v2.ToTensor", "generated/torchvision.transforms.v2.TrivialAugmentWide", "generated/torchvision.transforms.v2.UniformTemporalSubsample", "generated/torchvision.transforms.v2.functional.adjust_brightness", "generated/torchvision.transforms.v2.functional.adjust_contrast", "generated/torchvision.transforms.v2.functional.adjust_gamma", "generated/torchvision.transforms.v2.functional.adjust_hue", "generated/torchvision.transforms.v2.functional.adjust_saturation", "generated/torchvision.transforms.v2.functional.adjust_sharpness", "generated/torchvision.transforms.v2.functional.affine", "generated/torchvision.transforms.v2.functional.autocontrast", "generated/torchvision.transforms.v2.functional.center_crop", "generated/torchvision.transforms.v2.functional.clamp_bounding_boxes", "generated/torchvision.transforms.v2.functional.convert_bounding_box_format", "generated/torchvision.transforms.v2.functional.convert_image_dtype", "generated/torchvision.transforms.v2.functional.crop", "generated/torchvision.transforms.v2.functional.elastic", "generated/torchvision.transforms.v2.functional.equalize", "generated/torchvision.transforms.v2.functional.erase", "generated/torchvision.transforms.v2.functional.five_crop", "generated/torchvision.transforms.v2.functional.gaussian_blur", "generated/torchvision.transforms.v2.functional.horizontal_flip", "generated/torchvision.transforms.v2.functional.invert", "generated/torchvision.transforms.v2.functional.normalize", "generated/torchvision.transforms.v2.functional.pad", "generated/torchvision.transforms.v2.functional.permute_channels", "generated/torchvision.transforms.v2.functional.perspective", "generated/torchvision.transforms.v2.functional.pil_to_tensor", "generated/torchvision.transforms.v2.functional.posterize", "generated/torchvision.transforms.v2.functional.register_kernel", "generated/torchvision.transforms.v2.functional.resize", "generated/torchvision.transforms.v2.functional.resized_crop", "generated/torchvision.transforms.v2.functional.rgb_to_grayscale", "generated/torchvision.transforms.v2.functional.rotate", "generated/torchvision.transforms.v2.functional.solarize", "generated/torchvision.transforms.v2.functional.ten_crop", "generated/torchvision.transforms.v2.functional.to_dtype", "generated/torchvision.transforms.v2.functional.to_grayscale", "generated/torchvision.transforms.v2.functional.to_image", "generated/torchvision.transforms.v2.functional.to_pil_image", "generated/torchvision.transforms.v2.functional.to_tensor", "generated/torchvision.transforms.v2.functional.uniform_temporal_subsample", "generated/torchvision.transforms.v2.functional.vertical_flip", "generated/torchvision.tv_tensors.BoundingBoxFormat", "generated/torchvision.tv_tensors.BoundingBoxes", "generated/torchvision.tv_tensors.Image", "generated/torchvision.tv_tensors.Mask", "generated/torchvision.tv_tensors.TVTensor", "generated/torchvision.tv_tensors.Video", "generated/torchvision.tv_tensors.set_return_type", "generated/torchvision.tv_tensors.wrap", "generated/torchvision.utils.draw_bounding_boxes", "generated/torchvision.utils.draw_keypoints", "generated/torchvision.utils.draw_segmentation_masks", "generated/torchvision.utils.flow_to_image", "generated/torchvision.utils.make_grid", "generated/torchvision.utils.save_image", "generated/video_table", "index", "io", "models", "models/alexnet", "models/convnext", "models/deeplabv3", "models/densenet", "models/efficientnet", "models/efficientnetv2", "models/faster_rcnn", "models/fcn", "models/fcos", "models/generated/torchvision.models.alexnet", "models/generated/torchvision.models.convnext_base", "models/generated/torchvision.models.convnext_large", "models/generated/torchvision.models.convnext_small", "models/generated/torchvision.models.convnext_tiny", "models/generated/torchvision.models.densenet121", "models/generated/torchvision.models.densenet161", "models/generated/torchvision.models.densenet169", "models/generated/torchvision.models.densenet201", "models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn", "models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn", "models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn", "models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2", "models/generated/torchvision.models.detection.fcos_resnet50_fpn", "models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn", "models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn", "models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2", "models/generated/torchvision.models.detection.retinanet_resnet50_fpn", "models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2", "models/generated/torchvision.models.detection.ssd300_vgg16", "models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large", "models/generated/torchvision.models.efficientnet_b0", "models/generated/torchvision.models.efficientnet_b1", "models/generated/torchvision.models.efficientnet_b2", "models/generated/torchvision.models.efficientnet_b3", "models/generated/torchvision.models.efficientnet_b4", "models/generated/torchvision.models.efficientnet_b5", "models/generated/torchvision.models.efficientnet_b6", "models/generated/torchvision.models.efficientnet_b7", "models/generated/torchvision.models.efficientnet_v2_l", "models/generated/torchvision.models.efficientnet_v2_m", "models/generated/torchvision.models.efficientnet_v2_s", "models/generated/torchvision.models.googlenet", "models/generated/torchvision.models.inception_v3", "models/generated/torchvision.models.maxvit_t", "models/generated/torchvision.models.mnasnet0_5", "models/generated/torchvision.models.mnasnet0_75", "models/generated/torchvision.models.mnasnet1_0", "models/generated/torchvision.models.mnasnet1_3", "models/generated/torchvision.models.mobilenet_v2", "models/generated/torchvision.models.mobilenet_v3_large", "models/generated/torchvision.models.mobilenet_v3_small", "models/generated/torchvision.models.optical_flow.raft_large", "models/generated/torchvision.models.optical_flow.raft_small", "models/generated/torchvision.models.quantization.googlenet", "models/generated/torchvision.models.quantization.inception_v3", "models/generated/torchvision.models.quantization.mobilenet_v2", "models/generated/torchvision.models.quantization.mobilenet_v3_large", "models/generated/torchvision.models.quantization.resnet18", "models/generated/torchvision.models.quantization.resnet50", "models/generated/torchvision.models.quantization.resnext101_32x8d", "models/generated/torchvision.models.quantization.resnext101_64x4d", "models/generated/torchvision.models.quantization.shufflenet_v2_x0_5", "models/generated/torchvision.models.quantization.shufflenet_v2_x1_0", "models/generated/torchvision.models.quantization.shufflenet_v2_x1_5", "models/generated/torchvision.models.quantization.shufflenet_v2_x2_0", "models/generated/torchvision.models.regnet_x_16gf", "models/generated/torchvision.models.regnet_x_1_6gf", "models/generated/torchvision.models.regnet_x_32gf", "models/generated/torchvision.models.regnet_x_3_2gf", "models/generated/torchvision.models.regnet_x_400mf", "models/generated/torchvision.models.regnet_x_800mf", "models/generated/torchvision.models.regnet_x_8gf", "models/generated/torchvision.models.regnet_y_128gf", "models/generated/torchvision.models.regnet_y_16gf", "models/generated/torchvision.models.regnet_y_1_6gf", "models/generated/torchvision.models.regnet_y_32gf", "models/generated/torchvision.models.regnet_y_3_2gf", "models/generated/torchvision.models.regnet_y_400mf", "models/generated/torchvision.models.regnet_y_800mf", "models/generated/torchvision.models.regnet_y_8gf", "models/generated/torchvision.models.resnet101", "models/generated/torchvision.models.resnet152", "models/generated/torchvision.models.resnet18", "models/generated/torchvision.models.resnet34", "models/generated/torchvision.models.resnet50", "models/generated/torchvision.models.resnext101_32x8d", "models/generated/torchvision.models.resnext101_64x4d", "models/generated/torchvision.models.resnext50_32x4d", "models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large", "models/generated/torchvision.models.segmentation.deeplabv3_resnet101", "models/generated/torchvision.models.segmentation.deeplabv3_resnet50", "models/generated/torchvision.models.segmentation.fcn_resnet101", "models/generated/torchvision.models.segmentation.fcn_resnet50", "models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large", "models/generated/torchvision.models.shufflenet_v2_x0_5", "models/generated/torchvision.models.shufflenet_v2_x1_0", "models/generated/torchvision.models.shufflenet_v2_x1_5", "models/generated/torchvision.models.shufflenet_v2_x2_0", "models/generated/torchvision.models.squeezenet1_0", "models/generated/torchvision.models.squeezenet1_1", "models/generated/torchvision.models.swin_b", "models/generated/torchvision.models.swin_s", "models/generated/torchvision.models.swin_t", "models/generated/torchvision.models.swin_v2_b", "models/generated/torchvision.models.swin_v2_s", "models/generated/torchvision.models.swin_v2_t", "models/generated/torchvision.models.vgg11", "models/generated/torchvision.models.vgg11_bn", "models/generated/torchvision.models.vgg13", "models/generated/torchvision.models.vgg13_bn", "models/generated/torchvision.models.vgg16", "models/generated/torchvision.models.vgg16_bn", "models/generated/torchvision.models.vgg19", "models/generated/torchvision.models.vgg19_bn", "models/generated/torchvision.models.video.mc3_18", "models/generated/torchvision.models.video.mvit_v1_b", "models/generated/torchvision.models.video.mvit_v2_s", "models/generated/torchvision.models.video.r2plus1d_18", "models/generated/torchvision.models.video.r3d_18", "models/generated/torchvision.models.video.s3d", "models/generated/torchvision.models.video.swin3d_b", "models/generated/torchvision.models.video.swin3d_s", "models/generated/torchvision.models.video.swin3d_t", "models/generated/torchvision.models.vit_b_16", "models/generated/torchvision.models.vit_b_32", "models/generated/torchvision.models.vit_h_14", "models/generated/torchvision.models.vit_l_16", "models/generated/torchvision.models.vit_l_32", "models/generated/torchvision.models.wide_resnet101_2", "models/generated/torchvision.models.wide_resnet50_2", "models/googlenet", "models/googlenet_quant", "models/inception", "models/inception_quant", "models/keypoint_rcnn", "models/lraspp", "models/mask_rcnn", "models/maxvit", "models/mnasnet", "models/mobilenetv2", "models/mobilenetv2_quant", "models/mobilenetv3", "models/mobilenetv3_quant", "models/raft", "models/regnet", "models/resnet", "models/resnet_quant", "models/resnext", "models/resnext_quant", "models/retinanet", "models/shufflenetv2", "models/shufflenetv2_quant", "models/squeezenet", "models/ssd", "models/ssdlite", "models/swin_transformer", "models/vgg", "models/video_mvit", "models/video_resnet", "models/video_s3d", "models/video_swin_transformer", "models/vision_transformer", "models/wide_resnet", "ops", "training_references", "transforms", "tv_tensors", "utils"], "filenames": ["auto_examples/index.rst", "auto_examples/others/index.rst", "auto_examples/others/plot_optical_flow.rst", "auto_examples/others/plot_repurposing_annotations.rst", "auto_examples/others/plot_scripted_tensor_transforms.rst", "auto_examples/others/plot_video_api.rst", "auto_examples/others/plot_visualization_utils.rst", "auto_examples/others/sg_execution_times.rst", "auto_examples/transforms/index.rst", "auto_examples/transforms/plot_custom_transforms.rst", "auto_examples/transforms/plot_custom_tv_tensors.rst", "auto_examples/transforms/plot_cutmix_mixup.rst", "auto_examples/transforms/plot_transforms_e2e.rst", "auto_examples/transforms/plot_transforms_getting_started.rst", "auto_examples/transforms/plot_transforms_illustrations.rst", "auto_examples/transforms/plot_tv_tensors.rst", "auto_examples/transforms/sg_execution_times.rst", "datasets.rst", "feature_extraction.rst", "generated/classification_quant_table.rst", "generated/classification_table.rst", "generated/detection_keypoint_table.rst", "generated/detection_table.rst", "generated/instance_segmentation_table.rst", "generated/segmentation_table.rst", "generated/torchvision.datasets.CIFAR10.rst", "generated/torchvision.datasets.CIFAR100.rst", "generated/torchvision.datasets.CREStereo.rst", "generated/torchvision.datasets.Caltech101.rst", "generated/torchvision.datasets.Caltech256.rst", "generated/torchvision.datasets.CarlaStereo.rst", "generated/torchvision.datasets.CelebA.rst", "generated/torchvision.datasets.Cityscapes.rst", "generated/torchvision.datasets.CocoCaptions.rst", "generated/torchvision.datasets.CocoDetection.rst", "generated/torchvision.datasets.Country211.rst", "generated/torchvision.datasets.DTD.rst", "generated/torchvision.datasets.DatasetFolder.rst", "generated/torchvision.datasets.EMNIST.rst", "generated/torchvision.datasets.ETH3DStereo.rst", "generated/torchvision.datasets.EuroSAT.rst", "generated/torchvision.datasets.FER2013.rst", "generated/torchvision.datasets.FGVCAircraft.rst", "generated/torchvision.datasets.FakeData.rst", "generated/torchvision.datasets.FallingThingsStereo.rst", "generated/torchvision.datasets.FashionMNIST.rst", "generated/torchvision.datasets.Flickr30k.rst", "generated/torchvision.datasets.Flickr8k.rst", "generated/torchvision.datasets.Flowers102.rst", "generated/torchvision.datasets.FlyingChairs.rst", "generated/torchvision.datasets.FlyingThings3D.rst", "generated/torchvision.datasets.Food101.rst", "generated/torchvision.datasets.GTSRB.rst", "generated/torchvision.datasets.HD1K.rst", "generated/torchvision.datasets.HMDB51.rst", "generated/torchvision.datasets.INaturalist.rst", "generated/torchvision.datasets.ImageFolder.rst", "generated/torchvision.datasets.ImageNet.rst", "generated/torchvision.datasets.InStereo2k.rst", "generated/torchvision.datasets.KMNIST.rst", "generated/torchvision.datasets.Kinetics.rst", "generated/torchvision.datasets.Kitti.rst", "generated/torchvision.datasets.Kitti2012Stereo.rst", "generated/torchvision.datasets.Kitti2015Stereo.rst", "generated/torchvision.datasets.KittiFlow.rst", "generated/torchvision.datasets.LFWPairs.rst", "generated/torchvision.datasets.LFWPeople.rst", "generated/torchvision.datasets.LSUN.rst", "generated/torchvision.datasets.MNIST.rst", "generated/torchvision.datasets.Middlebury2014Stereo.rst", "generated/torchvision.datasets.MovingMNIST.rst", "generated/torchvision.datasets.Omniglot.rst", "generated/torchvision.datasets.OxfordIIITPet.rst", "generated/torchvision.datasets.PCAM.rst", "generated/torchvision.datasets.PhotoTour.rst", "generated/torchvision.datasets.Places365.rst", "generated/torchvision.datasets.QMNIST.rst", "generated/torchvision.datasets.RenderedSST2.rst", "generated/torchvision.datasets.SBDataset.rst", "generated/torchvision.datasets.SBU.rst", "generated/torchvision.datasets.SEMEION.rst", "generated/torchvision.datasets.STL10.rst", "generated/torchvision.datasets.SUN397.rst", "generated/torchvision.datasets.SVHN.rst", "generated/torchvision.datasets.SceneFlowStereo.rst", "generated/torchvision.datasets.Sintel.rst", "generated/torchvision.datasets.SintelStereo.rst", "generated/torchvision.datasets.StanfordCars.rst", "generated/torchvision.datasets.UCF101.rst", "generated/torchvision.datasets.USPS.rst", "generated/torchvision.datasets.VOCDetection.rst", "generated/torchvision.datasets.VOCSegmentation.rst", "generated/torchvision.datasets.VisionDataset.rst", "generated/torchvision.datasets.WIDERFace.rst", "generated/torchvision.datasets.wrap_dataset_for_transforms_v2.rst", "generated/torchvision.io.ImageReadMode.rst", "generated/torchvision.io.VideoReader.rst", "generated/torchvision.io.decode_image.rst", "generated/torchvision.io.decode_jpeg.rst", "generated/torchvision.io.decode_png.rst", "generated/torchvision.io.encode_jpeg.rst", "generated/torchvision.io.encode_png.rst", "generated/torchvision.io.read_file.rst", "generated/torchvision.io.read_image.rst", "generated/torchvision.io.read_video.rst", "generated/torchvision.io.read_video_timestamps.rst", "generated/torchvision.io.write_file.rst", "generated/torchvision.io.write_jpeg.rst", "generated/torchvision.io.write_png.rst", "generated/torchvision.io.write_video.rst", "generated/torchvision.models.feature_extraction.create_feature_extractor.rst", "generated/torchvision.models.feature_extraction.get_graph_node_names.rst", "generated/torchvision.models.get_model.rst", "generated/torchvision.models.get_model_weights.rst", "generated/torchvision.models.get_weight.rst", "generated/torchvision.models.list_models.rst", "generated/torchvision.ops.Conv2dNormActivation.rst", "generated/torchvision.ops.Conv3dNormActivation.rst", "generated/torchvision.ops.DeformConv2d.rst", "generated/torchvision.ops.DropBlock2d.rst", "generated/torchvision.ops.DropBlock3d.rst", "generated/torchvision.ops.FeaturePyramidNetwork.rst", "generated/torchvision.ops.FrozenBatchNorm2d.rst", "generated/torchvision.ops.MLP.rst", "generated/torchvision.ops.MultiScaleRoIAlign.rst", "generated/torchvision.ops.PSRoIAlign.rst", "generated/torchvision.ops.PSRoIPool.rst", "generated/torchvision.ops.Permute.rst", "generated/torchvision.ops.RoIAlign.rst", "generated/torchvision.ops.RoIPool.rst", "generated/torchvision.ops.SqueezeExcitation.rst", "generated/torchvision.ops.StochasticDepth.rst", "generated/torchvision.ops.batched_nms.rst", "generated/torchvision.ops.box_area.rst", "generated/torchvision.ops.box_convert.rst", "generated/torchvision.ops.box_iou.rst", "generated/torchvision.ops.clip_boxes_to_image.rst", "generated/torchvision.ops.complete_box_iou.rst", "generated/torchvision.ops.complete_box_iou_loss.rst", "generated/torchvision.ops.deform_conv2d.rst", "generated/torchvision.ops.distance_box_iou.rst", "generated/torchvision.ops.distance_box_iou_loss.rst", "generated/torchvision.ops.drop_block2d.rst", "generated/torchvision.ops.drop_block3d.rst", "generated/torchvision.ops.generalized_box_iou.rst", "generated/torchvision.ops.generalized_box_iou_loss.rst", "generated/torchvision.ops.masks_to_boxes.rst", "generated/torchvision.ops.nms.rst", "generated/torchvision.ops.ps_roi_align.rst", "generated/torchvision.ops.ps_roi_pool.rst", "generated/torchvision.ops.remove_small_boxes.rst", "generated/torchvision.ops.roi_align.rst", "generated/torchvision.ops.roi_pool.rst", "generated/torchvision.ops.sigmoid_focal_loss.rst", "generated/torchvision.ops.stochastic_depth.rst", "generated/torchvision.transforms.AugMix.rst", "generated/torchvision.transforms.AutoAugment.rst", "generated/torchvision.transforms.AutoAugmentPolicy.rst", "generated/torchvision.transforms.CenterCrop.rst", "generated/torchvision.transforms.ColorJitter.rst", "generated/torchvision.transforms.Compose.rst", "generated/torchvision.transforms.ConvertImageDtype.rst", "generated/torchvision.transforms.ElasticTransform.rst", "generated/torchvision.transforms.FiveCrop.rst", "generated/torchvision.transforms.GaussianBlur.rst", "generated/torchvision.transforms.Grayscale.rst", "generated/torchvision.transforms.Lambda.rst", "generated/torchvision.transforms.LinearTransformation.rst", "generated/torchvision.transforms.Normalize.rst", "generated/torchvision.transforms.PILToTensor.rst", "generated/torchvision.transforms.Pad.rst", "generated/torchvision.transforms.RandAugment.rst", "generated/torchvision.transforms.RandomAdjustSharpness.rst", "generated/torchvision.transforms.RandomAffine.rst", "generated/torchvision.transforms.RandomApply.rst", "generated/torchvision.transforms.RandomAutocontrast.rst", "generated/torchvision.transforms.RandomChoice.rst", "generated/torchvision.transforms.RandomCrop.rst", "generated/torchvision.transforms.RandomEqualize.rst", "generated/torchvision.transforms.RandomErasing.rst", "generated/torchvision.transforms.RandomGrayscale.rst", "generated/torchvision.transforms.RandomHorizontalFlip.rst", "generated/torchvision.transforms.RandomInvert.rst", "generated/torchvision.transforms.RandomOrder.rst", "generated/torchvision.transforms.RandomPerspective.rst", "generated/torchvision.transforms.RandomPosterize.rst", "generated/torchvision.transforms.RandomResizedCrop.rst", "generated/torchvision.transforms.RandomRotation.rst", "generated/torchvision.transforms.RandomSolarize.rst", "generated/torchvision.transforms.RandomVerticalFlip.rst", "generated/torchvision.transforms.Resize.rst", "generated/torchvision.transforms.TenCrop.rst", "generated/torchvision.transforms.ToPILImage.rst", "generated/torchvision.transforms.ToTensor.rst", "generated/torchvision.transforms.TrivialAugmentWide.rst", "generated/torchvision.transforms.functional.adjust_brightness.rst", "generated/torchvision.transforms.functional.adjust_contrast.rst", "generated/torchvision.transforms.functional.adjust_gamma.rst", "generated/torchvision.transforms.functional.adjust_hue.rst", "generated/torchvision.transforms.functional.adjust_saturation.rst", "generated/torchvision.transforms.functional.adjust_sharpness.rst", "generated/torchvision.transforms.functional.affine.rst", "generated/torchvision.transforms.functional.autocontrast.rst", "generated/torchvision.transforms.functional.center_crop.rst", "generated/torchvision.transforms.functional.convert_image_dtype.rst", "generated/torchvision.transforms.functional.crop.rst", "generated/torchvision.transforms.functional.equalize.rst", "generated/torchvision.transforms.functional.erase.rst", "generated/torchvision.transforms.functional.five_crop.rst", "generated/torchvision.transforms.functional.gaussian_blur.rst", "generated/torchvision.transforms.functional.get_dimensions.rst", "generated/torchvision.transforms.functional.get_image_num_channels.rst", "generated/torchvision.transforms.functional.get_image_size.rst", "generated/torchvision.transforms.functional.hflip.rst", "generated/torchvision.transforms.functional.invert.rst", "generated/torchvision.transforms.functional.normalize.rst", "generated/torchvision.transforms.functional.pad.rst", "generated/torchvision.transforms.functional.perspective.rst", "generated/torchvision.transforms.functional.pil_to_tensor.rst", "generated/torchvision.transforms.functional.posterize.rst", "generated/torchvision.transforms.functional.resize.rst", "generated/torchvision.transforms.functional.resized_crop.rst", "generated/torchvision.transforms.functional.rgb_to_grayscale.rst", "generated/torchvision.transforms.functional.rotate.rst", "generated/torchvision.transforms.functional.solarize.rst", "generated/torchvision.transforms.functional.ten_crop.rst", "generated/torchvision.transforms.functional.to_grayscale.rst", "generated/torchvision.transforms.functional.to_pil_image.rst", "generated/torchvision.transforms.functional.to_tensor.rst", "generated/torchvision.transforms.functional.vflip.rst", "generated/torchvision.transforms.v2.AugMix.rst", "generated/torchvision.transforms.v2.AutoAugment.rst", "generated/torchvision.transforms.v2.CenterCrop.rst", "generated/torchvision.transforms.v2.ClampBoundingBoxes.rst", "generated/torchvision.transforms.v2.ColorJitter.rst", "generated/torchvision.transforms.v2.Compose.rst", "generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.rst", "generated/torchvision.transforms.v2.ConvertImageDtype.rst", "generated/torchvision.transforms.v2.CutMix.rst", "generated/torchvision.transforms.v2.ElasticTransform.rst", "generated/torchvision.transforms.v2.FiveCrop.rst", "generated/torchvision.transforms.v2.GaussianBlur.rst", "generated/torchvision.transforms.v2.Grayscale.rst", "generated/torchvision.transforms.v2.Lambda.rst", "generated/torchvision.transforms.v2.LinearTransformation.rst", "generated/torchvision.transforms.v2.MixUp.rst", "generated/torchvision.transforms.v2.Normalize.rst", "generated/torchvision.transforms.v2.PILToTensor.rst", "generated/torchvision.transforms.v2.Pad.rst", "generated/torchvision.transforms.v2.RandAugment.rst", "generated/torchvision.transforms.v2.RandomAdjustSharpness.rst", "generated/torchvision.transforms.v2.RandomAffine.rst", "generated/torchvision.transforms.v2.RandomApply.rst", "generated/torchvision.transforms.v2.RandomAutocontrast.rst", "generated/torchvision.transforms.v2.RandomChannelPermutation.rst", "generated/torchvision.transforms.v2.RandomChoice.rst", "generated/torchvision.transforms.v2.RandomCrop.rst", "generated/torchvision.transforms.v2.RandomEqualize.rst", "generated/torchvision.transforms.v2.RandomErasing.rst", "generated/torchvision.transforms.v2.RandomGrayscale.rst", "generated/torchvision.transforms.v2.RandomHorizontalFlip.rst", "generated/torchvision.transforms.v2.RandomInvert.rst", "generated/torchvision.transforms.v2.RandomIoUCrop.rst", "generated/torchvision.transforms.v2.RandomOrder.rst", "generated/torchvision.transforms.v2.RandomPerspective.rst", "generated/torchvision.transforms.v2.RandomPhotometricDistort.rst", "generated/torchvision.transforms.v2.RandomPosterize.rst", "generated/torchvision.transforms.v2.RandomResize.rst", "generated/torchvision.transforms.v2.RandomResizedCrop.rst", "generated/torchvision.transforms.v2.RandomRotation.rst", "generated/torchvision.transforms.v2.RandomShortestSize.rst", "generated/torchvision.transforms.v2.RandomSolarize.rst", "generated/torchvision.transforms.v2.RandomVerticalFlip.rst", "generated/torchvision.transforms.v2.RandomZoomOut.rst", "generated/torchvision.transforms.v2.Resize.rst", "generated/torchvision.transforms.v2.SanitizeBoundingBoxes.rst", "generated/torchvision.transforms.v2.ScaleJitter.rst", "generated/torchvision.transforms.v2.TenCrop.rst", "generated/torchvision.transforms.v2.ToDtype.rst", "generated/torchvision.transforms.v2.ToImage.rst", "generated/torchvision.transforms.v2.ToPILImage.rst", "generated/torchvision.transforms.v2.ToPureTensor.rst", "generated/torchvision.transforms.v2.ToTensor.rst", "generated/torchvision.transforms.v2.TrivialAugmentWide.rst", "generated/torchvision.transforms.v2.UniformTemporalSubsample.rst", "generated/torchvision.transforms.v2.functional.adjust_brightness.rst", "generated/torchvision.transforms.v2.functional.adjust_contrast.rst", "generated/torchvision.transforms.v2.functional.adjust_gamma.rst", "generated/torchvision.transforms.v2.functional.adjust_hue.rst", "generated/torchvision.transforms.v2.functional.adjust_saturation.rst", "generated/torchvision.transforms.v2.functional.adjust_sharpness.rst", "generated/torchvision.transforms.v2.functional.affine.rst", "generated/torchvision.transforms.v2.functional.autocontrast.rst", "generated/torchvision.transforms.v2.functional.center_crop.rst", "generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.rst", "generated/torchvision.transforms.v2.functional.convert_bounding_box_format.rst", "generated/torchvision.transforms.v2.functional.convert_image_dtype.rst", "generated/torchvision.transforms.v2.functional.crop.rst", "generated/torchvision.transforms.v2.functional.elastic.rst", "generated/torchvision.transforms.v2.functional.equalize.rst", "generated/torchvision.transforms.v2.functional.erase.rst", "generated/torchvision.transforms.v2.functional.five_crop.rst", "generated/torchvision.transforms.v2.functional.gaussian_blur.rst", "generated/torchvision.transforms.v2.functional.horizontal_flip.rst", "generated/torchvision.transforms.v2.functional.invert.rst", "generated/torchvision.transforms.v2.functional.normalize.rst", "generated/torchvision.transforms.v2.functional.pad.rst", "generated/torchvision.transforms.v2.functional.permute_channels.rst", "generated/torchvision.transforms.v2.functional.perspective.rst", "generated/torchvision.transforms.v2.functional.pil_to_tensor.rst", "generated/torchvision.transforms.v2.functional.posterize.rst", "generated/torchvision.transforms.v2.functional.register_kernel.rst", "generated/torchvision.transforms.v2.functional.resize.rst", "generated/torchvision.transforms.v2.functional.resized_crop.rst", "generated/torchvision.transforms.v2.functional.rgb_to_grayscale.rst", "generated/torchvision.transforms.v2.functional.rotate.rst", "generated/torchvision.transforms.v2.functional.solarize.rst", "generated/torchvision.transforms.v2.functional.ten_crop.rst", "generated/torchvision.transforms.v2.functional.to_dtype.rst", "generated/torchvision.transforms.v2.functional.to_grayscale.rst", "generated/torchvision.transforms.v2.functional.to_image.rst", "generated/torchvision.transforms.v2.functional.to_pil_image.rst", "generated/torchvision.transforms.v2.functional.to_tensor.rst", "generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.rst", "generated/torchvision.transforms.v2.functional.vertical_flip.rst", "generated/torchvision.tv_tensors.BoundingBoxFormat.rst", "generated/torchvision.tv_tensors.BoundingBoxes.rst", "generated/torchvision.tv_tensors.Image.rst", "generated/torchvision.tv_tensors.Mask.rst", "generated/torchvision.tv_tensors.TVTensor.rst", "generated/torchvision.tv_tensors.Video.rst", "generated/torchvision.tv_tensors.set_return_type.rst", "generated/torchvision.tv_tensors.wrap.rst", "generated/torchvision.utils.draw_bounding_boxes.rst", "generated/torchvision.utils.draw_keypoints.rst", "generated/torchvision.utils.draw_segmentation_masks.rst", "generated/torchvision.utils.flow_to_image.rst", "generated/torchvision.utils.make_grid.rst", "generated/torchvision.utils.save_image.rst", "generated/video_table.rst", "index.rst", "io.rst", "models.rst", "models/alexnet.rst", "models/convnext.rst", "models/deeplabv3.rst", "models/densenet.rst", "models/efficientnet.rst", "models/efficientnetv2.rst", "models/faster_rcnn.rst", "models/fcn.rst", "models/fcos.rst", "models/generated/torchvision.models.alexnet.rst", "models/generated/torchvision.models.convnext_base.rst", "models/generated/torchvision.models.convnext_large.rst", "models/generated/torchvision.models.convnext_small.rst", "models/generated/torchvision.models.convnext_tiny.rst", "models/generated/torchvision.models.densenet121.rst", "models/generated/torchvision.models.densenet161.rst", "models/generated/torchvision.models.densenet169.rst", "models/generated/torchvision.models.densenet201.rst", "models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.rst", "models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.rst", "models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.rst", "models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.rst", "models/generated/torchvision.models.detection.fcos_resnet50_fpn.rst", "models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.rst", "models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.rst", "models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.rst", "models/generated/torchvision.models.detection.retinanet_resnet50_fpn.rst", "models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.rst", "models/generated/torchvision.models.detection.ssd300_vgg16.rst", "models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.rst", "models/generated/torchvision.models.efficientnet_b0.rst", "models/generated/torchvision.models.efficientnet_b1.rst", "models/generated/torchvision.models.efficientnet_b2.rst", "models/generated/torchvision.models.efficientnet_b3.rst", "models/generated/torchvision.models.efficientnet_b4.rst", "models/generated/torchvision.models.efficientnet_b5.rst", "models/generated/torchvision.models.efficientnet_b6.rst", "models/generated/torchvision.models.efficientnet_b7.rst", "models/generated/torchvision.models.efficientnet_v2_l.rst", "models/generated/torchvision.models.efficientnet_v2_m.rst", "models/generated/torchvision.models.efficientnet_v2_s.rst", "models/generated/torchvision.models.googlenet.rst", "models/generated/torchvision.models.inception_v3.rst", "models/generated/torchvision.models.maxvit_t.rst", "models/generated/torchvision.models.mnasnet0_5.rst", "models/generated/torchvision.models.mnasnet0_75.rst", "models/generated/torchvision.models.mnasnet1_0.rst", "models/generated/torchvision.models.mnasnet1_3.rst", "models/generated/torchvision.models.mobilenet_v2.rst", "models/generated/torchvision.models.mobilenet_v3_large.rst", "models/generated/torchvision.models.mobilenet_v3_small.rst", "models/generated/torchvision.models.optical_flow.raft_large.rst", "models/generated/torchvision.models.optical_flow.raft_small.rst", "models/generated/torchvision.models.quantization.googlenet.rst", "models/generated/torchvision.models.quantization.inception_v3.rst", "models/generated/torchvision.models.quantization.mobilenet_v2.rst", "models/generated/torchvision.models.quantization.mobilenet_v3_large.rst", "models/generated/torchvision.models.quantization.resnet18.rst", "models/generated/torchvision.models.quantization.resnet50.rst", "models/generated/torchvision.models.quantization.resnext101_32x8d.rst", "models/generated/torchvision.models.quantization.resnext101_64x4d.rst", "models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.rst", "models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.rst", "models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.rst", "models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.rst", "models/generated/torchvision.models.regnet_x_16gf.rst", "models/generated/torchvision.models.regnet_x_1_6gf.rst", "models/generated/torchvision.models.regnet_x_32gf.rst", "models/generated/torchvision.models.regnet_x_3_2gf.rst", "models/generated/torchvision.models.regnet_x_400mf.rst", "models/generated/torchvision.models.regnet_x_800mf.rst", "models/generated/torchvision.models.regnet_x_8gf.rst", "models/generated/torchvision.models.regnet_y_128gf.rst", "models/generated/torchvision.models.regnet_y_16gf.rst", "models/generated/torchvision.models.regnet_y_1_6gf.rst", "models/generated/torchvision.models.regnet_y_32gf.rst", "models/generated/torchvision.models.regnet_y_3_2gf.rst", "models/generated/torchvision.models.regnet_y_400mf.rst", "models/generated/torchvision.models.regnet_y_800mf.rst", "models/generated/torchvision.models.regnet_y_8gf.rst", "models/generated/torchvision.models.resnet101.rst", "models/generated/torchvision.models.resnet152.rst", "models/generated/torchvision.models.resnet18.rst", "models/generated/torchvision.models.resnet34.rst", "models/generated/torchvision.models.resnet50.rst", "models/generated/torchvision.models.resnext101_32x8d.rst", "models/generated/torchvision.models.resnext101_64x4d.rst", "models/generated/torchvision.models.resnext50_32x4d.rst", "models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.rst", "models/generated/torchvision.models.segmentation.deeplabv3_resnet101.rst", "models/generated/torchvision.models.segmentation.deeplabv3_resnet50.rst", "models/generated/torchvision.models.segmentation.fcn_resnet101.rst", "models/generated/torchvision.models.segmentation.fcn_resnet50.rst", "models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.rst", "models/generated/torchvision.models.shufflenet_v2_x0_5.rst", "models/generated/torchvision.models.shufflenet_v2_x1_0.rst", "models/generated/torchvision.models.shufflenet_v2_x1_5.rst", "models/generated/torchvision.models.shufflenet_v2_x2_0.rst", "models/generated/torchvision.models.squeezenet1_0.rst", "models/generated/torchvision.models.squeezenet1_1.rst", "models/generated/torchvision.models.swin_b.rst", "models/generated/torchvision.models.swin_s.rst", "models/generated/torchvision.models.swin_t.rst", "models/generated/torchvision.models.swin_v2_b.rst", "models/generated/torchvision.models.swin_v2_s.rst", "models/generated/torchvision.models.swin_v2_t.rst", "models/generated/torchvision.models.vgg11.rst", "models/generated/torchvision.models.vgg11_bn.rst", "models/generated/torchvision.models.vgg13.rst", "models/generated/torchvision.models.vgg13_bn.rst", "models/generated/torchvision.models.vgg16.rst", "models/generated/torchvision.models.vgg16_bn.rst", "models/generated/torchvision.models.vgg19.rst", "models/generated/torchvision.models.vgg19_bn.rst", "models/generated/torchvision.models.video.mc3_18.rst", "models/generated/torchvision.models.video.mvit_v1_b.rst", "models/generated/torchvision.models.video.mvit_v2_s.rst", "models/generated/torchvision.models.video.r2plus1d_18.rst", "models/generated/torchvision.models.video.r3d_18.rst", "models/generated/torchvision.models.video.s3d.rst", "models/generated/torchvision.models.video.swin3d_b.rst", "models/generated/torchvision.models.video.swin3d_s.rst", "models/generated/torchvision.models.video.swin3d_t.rst", "models/generated/torchvision.models.vit_b_16.rst", "models/generated/torchvision.models.vit_b_32.rst", "models/generated/torchvision.models.vit_h_14.rst", "models/generated/torchvision.models.vit_l_16.rst", "models/generated/torchvision.models.vit_l_32.rst", "models/generated/torchvision.models.wide_resnet101_2.rst", "models/generated/torchvision.models.wide_resnet50_2.rst", "models/googlenet.rst", "models/googlenet_quant.rst", "models/inception.rst", "models/inception_quant.rst", "models/keypoint_rcnn.rst", "models/lraspp.rst", "models/mask_rcnn.rst", "models/maxvit.rst", "models/mnasnet.rst", "models/mobilenetv2.rst", "models/mobilenetv2_quant.rst", "models/mobilenetv3.rst", "models/mobilenetv3_quant.rst", "models/raft.rst", "models/regnet.rst", "models/resnet.rst", "models/resnet_quant.rst", "models/resnext.rst", "models/resnext_quant.rst", "models/retinanet.rst", "models/shufflenetv2.rst", "models/shufflenetv2_quant.rst", "models/squeezenet.rst", "models/ssd.rst", "models/ssdlite.rst", "models/swin_transformer.rst", "models/vgg.rst", "models/video_mvit.rst", "models/video_resnet.rst", "models/video_s3d.rst", "models/video_swin_transformer.rst", "models/vision_transformer.rst", "models/wide_resnet.rst", "ops.rst", "training_references.rst", "transforms.rst", "tv_tensors.rst", "utils.rst"], "titles": ["Examples and tutorials", "Others", "Optical Flow: Predicting movement with the RAFT model", "Repurposing masks into bounding boxes", "Torchscript support", "Video API", "Visualization utilities", "Computation times", "Transforms", "How to write your own v2 transforms", "How to write your own TVTensor class", "How to use CutMix and MixUp", "Transforms v2: End-to-end object detection/segmentation example", "Getting started with transforms v2", "Illustration of transforms", "TVTensors FAQ", "Computation times", "Datasets", "Feature extraction for model inspection", "&lt;no title&gt;", "&lt;no title&gt;", "&lt;no title&gt;", "&lt;no title&gt;", "&lt;no title&gt;", "&lt;no title&gt;", "CIFAR10", "CIFAR100", "CREStereo", "Caltech101", "Caltech256", "CarlaStereo", "CelebA", "Cityscapes", "CocoCaptions", "CocoDetection", "Country211", "DTD", "DatasetFolder", "EMNIST", "ETH3DStereo", "EuroSAT", "FER2013", "FGVCAircraft", "FakeData", "FallingThingsStereo", "FashionMNIST", "Flickr30k", "Flickr8k", "Flowers102", "FlyingChairs", "FlyingThings3D", "Food101", "GTSRB", "HD1K", "HMDB51", "INaturalist", "ImageFolder", "ImageNet", "InStereo2k", "KMNIST", "Kinetics", "Kitti", "Kitti2012Stereo", "Kitti2015Stereo", "KittiFlow", "LFWPairs", "LFWPeople", "LSUN", "MNIST", "Middlebury2014Stereo", "MovingMNIST", "Omniglot", "OxfordIIITPet", "PCAM", "PhotoTour", "Places365", "QMNIST", "RenderedSST2", "SBDataset", "SBU", "SEMEION", "STL10", "SUN397", "SVHN", "SceneFlowStereo", "Sintel", "SintelStereo", "StanfordCars", "UCF101", "USPS", "VOCDetection", "VOCSegmentation", "VisionDataset", "WIDERFace", "wrap_dataset_for_transforms_v2", "ImageReadMode", "VideoReader", "decode_image", "decode_jpeg", "decode_png", "encode_jpeg", "encode_png", "read_file", "read_image", "read_video", "read_video_timestamps", "write_file", "write_jpeg", "write_png", "write_video", "create_feature_extractor", "get_graph_node_names", "get_model", "get_model_weights", "get_weight", "list_models", "Conv2dNormActivation", "Conv3dNormActivation", "DeformConv2d", "DropBlock2d", "DropBlock3d", "FeaturePyramidNetwork", "FrozenBatchNorm2d", "MLP", "MultiScaleRoIAlign", "PSRoIAlign", "PSRoIPool", "Permute", "RoIAlign", "RoIPool", "SqueezeExcitation", "StochasticDepth", "batched_nms", "box_area", "box_convert", "box_iou", "clip_boxes_to_image", "complete_box_iou", "complete_box_iou_loss", "deform_conv2d", "distance_box_iou", "distance_box_iou_loss", "drop_block2d", "drop_block3d", "generalized_box_iou", "generalized_box_iou_loss", "masks_to_boxes", "nms", "ps_roi_align", "ps_roi_pool", "remove_small_boxes", "roi_align", "roi_pool", "sigmoid_focal_loss", "stochastic_depth", "AugMix", "AutoAugment", "AutoAugmentPolicy", "CenterCrop", "ColorJitter", "Compose", "ConvertImageDtype", "ElasticTransform", "FiveCrop", "GaussianBlur", "Grayscale", "Lambda", "LinearTransformation", "Normalize", "PILToTensor", "Pad", "RandAugment", "RandomAdjustSharpness", "RandomAffine", "RandomApply", "RandomAutocontrast", "RandomChoice", "RandomCrop", "RandomEqualize", "RandomErasing", "RandomGrayscale", "RandomHorizontalFlip", "RandomInvert", "RandomOrder", "RandomPerspective", "RandomPosterize", "RandomResizedCrop", "RandomRotation", "RandomSolarize", "RandomVerticalFlip", "Resize", "TenCrop", "ToPILImage", "ToTensor", "TrivialAugmentWide", "adjust_brightness", "adjust_contrast", "adjust_gamma", "adjust_hue", "adjust_saturation", "adjust_sharpness", "affine", "autocontrast", "center_crop", "convert_image_dtype", "crop", "equalize", "erase", "five_crop", "gaussian_blur", "get_dimensions", "get_image_num_channels", "get_image_size", "hflip", "invert", "normalize", "pad", "perspective", "pil_to_tensor", "posterize", "resize", "resized_crop", "rgb_to_grayscale", "rotate", "solarize", "ten_crop", "to_grayscale", "to_pil_image", "to_tensor", "vflip", "AugMix", "AutoAugment", "CenterCrop", "ClampBoundingBoxes", "ColorJitter", "Compose", "ConvertBoundingBoxFormat", "ConvertImageDtype", "CutMix", "ElasticTransform", "FiveCrop", "GaussianBlur", "Grayscale", "Lambda", "LinearTransformation", "MixUp", "Normalize", "PILToTensor", "Pad", "RandAugment", "RandomAdjustSharpness", "RandomAffine", "RandomApply", "RandomAutocontrast", "RandomChannelPermutation", "RandomChoice", "RandomCrop", "RandomEqualize", "RandomErasing", "RandomGrayscale", "RandomHorizontalFlip", "RandomInvert", "RandomIoUCrop", "RandomOrder", "RandomPerspective", "RandomPhotometricDistort", "RandomPosterize", "RandomResize", "RandomResizedCrop", "RandomRotation", "RandomShortestSize", "RandomSolarize", "RandomVerticalFlip", "RandomZoomOut", "Resize", "SanitizeBoundingBoxes", "ScaleJitter", "TenCrop", "ToDtype", "ToImage", "ToPILImage", "ToPureTensor", "ToTensor", "TrivialAugmentWide", "UniformTemporalSubsample", "adjust_brightness", "adjust_contrast", "adjust_gamma", "adjust_hue", "adjust_saturation", "adjust_sharpness", "affine", "autocontrast", "center_crop", "clamp_bounding_boxes", "torchvision.transforms.v2.functional.convert_bounding_box_format", "convert_image_dtype", "crop", "elastic", "equalize", "erase", "five_crop", "gaussian_blur", "horizontal_flip", "invert", "normalize", "pad", "permute_channels", "perspective", "torchvision.transforms.v2.functional.pil_to_tensor", "posterize", "register_kernel", "resize", "resized_crop", "rgb_to_grayscale", "rotate", "solarize", "ten_crop", "torchvision.transforms.v2.functional.to_dtype", "to_grayscale", "torchvision.transforms.v2.functional.to_image", "torchvision.transforms.v2.functional.to_pil_image", "to_tensor", "uniform_temporal_subsample", "vertical_flip", "BoundingBoxFormat", "BoundingBoxes", "Image", "Mask", "TVTensor", "Video", "set_return_type", "wrap", "draw_bounding_boxes", "draw_keypoints", "draw_segmentation_masks", "flow_to_image", "make_grid", "save_image", "&lt;no title&gt;", "torchvision", "Decoding / Encoding images and videos", "Models and pre-trained weights", "AlexNet", "ConvNeXt", "DeepLabV3", "DenseNet", "EfficientNet", "EfficientNetV2", "Faster R-CNN", "FCN", "FCOS", "alexnet", "convnext_base", "convnext_large", "convnext_small", "convnext_tiny", "densenet121", "densenet161", "densenet169", "densenet201", "fasterrcnn_mobilenet_v3_large_320_fpn", "fasterrcnn_mobilenet_v3_large_fpn", "fasterrcnn_resnet50_fpn", "fasterrcnn_resnet50_fpn_v2", "fcos_resnet50_fpn", "keypointrcnn_resnet50_fpn", "maskrcnn_resnet50_fpn", "maskrcnn_resnet50_fpn_v2", "retinanet_resnet50_fpn", "retinanet_resnet50_fpn_v2", "ssd300_vgg16", "ssdlite320_mobilenet_v3_large", "efficientnet_b0", "efficientnet_b1", "efficientnet_b2", "efficientnet_b3", "efficientnet_b4", "efficientnet_b5", "efficientnet_b6", "efficientnet_b7", "efficientnet_v2_l", "efficientnet_v2_m", "efficientnet_v2_s", "googlenet", "inception_v3", "maxvit_t", "mnasnet0_5", "mnasnet0_75", "mnasnet1_0", "mnasnet1_3", "mobilenet_v2", "mobilenet_v3_large", "mobilenet_v3_small", "raft_large", "raft_small", "googlenet", "inception_v3", "mobilenet_v2", "mobilenet_v3_large", "resnet18", "resnet50", "resnext101_32x8d", "resnext101_64x4d", "shufflenet_v2_x0_5", "shufflenet_v2_x1_0", "shufflenet_v2_x1_5", "shufflenet_v2_x2_0", "regnet_x_16gf", "regnet_x_1_6gf", "regnet_x_32gf", "regnet_x_3_2gf", "regnet_x_400mf", "regnet_x_800mf", "regnet_x_8gf", "regnet_y_128gf", "regnet_y_16gf", "regnet_y_1_6gf", "regnet_y_32gf", "regnet_y_3_2gf", "regnet_y_400mf", "regnet_y_800mf", "regnet_y_8gf", "resnet101", "resnet152", "resnet18", "resnet34", "resnet50", "resnext101_32x8d", "resnext101_64x4d", "resnext50_32x4d", "deeplabv3_mobilenet_v3_large", "deeplabv3_resnet101", "deeplabv3_resnet50", "fcn_resnet101", "fcn_resnet50", "lraspp_mobilenet_v3_large", "shufflenet_v2_x0_5", "shufflenet_v2_x1_0", "shufflenet_v2_x1_5", "shufflenet_v2_x2_0", "squeezenet1_0", "squeezenet1_1", "swin_b", "swin_s", "swin_t", "swin_v2_b", "swin_v2_s", "swin_v2_t", "vgg11", "vgg11_bn", "vgg13", "vgg13_bn", "vgg16", "vgg16_bn", "vgg19", "vgg19_bn", "mc3_18", "mvit_v1_b", "mvit_v2_s", "r2plus1d_18", "r3d_18", "s3d", "swin3d_b", "swin3d_s", "swin3d_t", "vit_b_16", "vit_b_32", "vit_h_14", "vit_l_16", "vit_l_32", "wide_resnet101_2", "wide_resnet50_2", "GoogLeNet", "Quantized GoogLeNet", "Inception V3", "Quantized InceptionV3", "Keypoint R-CNN", "LRASPP", "Mask R-CNN", "MaxVit", "MNASNet", "MobileNet V2", "Quantized MobileNet V2", "MobileNet V3", "Quantized MobileNet V3", "RAFT", "RegNet", "ResNet", "Quantized ResNet", "ResNeXt", "Quantized ResNeXt", "RetinaNet", "ShuffleNet V2", "Quantized ShuffleNet V2", "SqueezeNet", "SSD", "SSDlite", "SwinTransformer", "VGG", "Video MViT", "Video ResNet", "Video S3D", "Video SwinTransformer", "VisionTransformer", "Wide ResNet", "Operators", "Training references", "Transforming and augmenting images", "TVTensors", "Utils"], "terms": {"get": [0, 2, 3, 5, 6, 8, 15, 16, 18, 32, 34, 61, 94, 103, 111, 112, 114, 121, 156, 159, 160, 168, 173, 177, 179, 181, 184, 186, 187, 231, 234, 235, 246, 251, 256, 258, 260, 264, 265, 268, 269, 275, 278, 326, 327, 328, 329, 330, 340, 508, 509], "start": [0, 2, 5, 6, 8, 12, 16, 34, 94, 103, 104, 160, 168, 177, 181, 186, 235, 246, 256, 260, 265, 268, 275, 278, 326, 327, 328, 329, 330, 340, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 509], "v2": [0, 4, 8, 10, 11, 14, 15, 16, 34, 77, 92, 94, 103, 112, 139, 160, 168, 177, 181, 186, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 310, 311, 312, 313, 314, 315, 316, 317, 319, 322, 323, 324, 326, 327, 328, 329, 330, 340, 342, 404, 405, 406, 407, 437, 438, 439, 440, 446, 447, 448, 498, 500, 509], "illustr": [0, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 16, 155, 156, 157, 158, 159, 162, 163, 164, 165, 170, 171, 172, 173, 174, 175, 177, 178, 181, 182, 184, 185, 186, 187, 188, 189, 190, 194, 200, 201, 202, 203, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 223, 224, 226, 229, 230, 231, 232, 234, 239, 240, 241, 242, 248, 249, 250, 251, 252, 253, 256, 257, 260, 261, 264, 266, 268, 269, 271, 272, 274, 283, 342], "end": [0, 2, 3, 4, 5, 6, 8, 9, 10, 11, 13, 14, 15, 16, 18, 34, 92, 94, 104, 112, 160, 181, 235, 260, 262, 265, 273, 275, 278, 279, 327, 329, 331, 394, 395, 415, 416, 418, 466, 468, 469, 508], "object": [0, 3, 6, 8, 13, 14, 15, 16, 18, 32, 34, 54, 61, 88, 92, 94, 96, 112, 121, 160, 162, 181, 235, 239, 260, 262, 265, 273, 275, 278, 279, 326, 327, 329, 338, 340, 349, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 492, 506, 508], "detect": [0, 5, 6, 8, 9, 16, 18, 34, 61, 74, 90, 92, 94, 96, 97, 112, 121, 124, 153, 160, 181, 235, 260, 262, 265, 273, 275, 278, 279, 327, 328, 329, 340, 349, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 459, 477, 479, 492, 496, 497, 500, 507, 508], "segment": [0, 5, 8, 16, 32, 34, 72, 78, 91, 92, 94, 112, 160, 181, 235, 260, 262, 265, 267, 273, 275, 276, 278, 279, 327, 328, 329, 335, 340, 345, 350, 363, 365, 366, 367, 369, 431, 432, 433, 434, 435, 436, 478, 507, 508], "how": [0, 2, 4, 5, 6, 8, 12, 13, 16, 18, 43, 92, 159, 160, 163, 168, 169, 172, 176, 181, 186, 191, 195, 196, 198, 199, 200, 234, 235, 238, 240, 245, 246, 247, 250, 255, 260, 265, 268, 275, 277, 278, 311, 326, 329, 332, 342, 367, 394, 395, 508], "us": [0, 3, 4, 5, 6, 8, 9, 10, 12, 14, 16, 17, 18, 27, 28, 31, 32, 34, 37, 38, 43, 44, 50, 54, 55, 56, 57, 60, 61, 62, 65, 66, 67, 69, 72, 75, 77, 78, 84, 85, 86, 88, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 103, 104, 110, 111, 112, 116, 117, 121, 123, 124, 146, 148, 151, 153, 154, 155, 156, 157, 158, 159, 160, 162, 163, 164, 165, 166, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 193, 194, 198, 200, 201, 202, 203, 204, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 226, 227, 229, 230, 231, 232, 234, 235, 237, 238, 239, 240, 241, 242, 243, 245, 246, 247, 248, 249, 250, 251, 252, 253, 255, 256, 257, 258, 260, 261, 262, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 282, 283, 296, 311, 322, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 340, 341, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506, 507, 509], "cutmix": [0, 8, 13, 16, 43, 92, 160, 168, 169, 176, 181, 186, 235, 245, 246, 247, 255, 260, 268, 278], "mixup": [0, 8, 13, 16, 43, 92, 160, 168, 169, 176, 181, 186, 235, 238, 246, 247, 255, 260, 268, 278], "write": [0, 8, 15, 16, 106, 109, 110, 160, 168, 181, 186, 235, 246, 260, 268, 311, 326, 329, 332, 341, 508], "your": [0, 2, 3, 4, 8, 11, 12, 15, 16, 17, 110, 148, 149, 151, 152, 159, 160, 163, 168, 181, 186, 191, 198, 208, 225, 234, 235, 240, 243, 246, 252, 260, 265, 268, 277, 311, 326, 329, 331, 332, 342, 385, 397, 508], "own": [0, 8, 12, 15, 16, 17, 110, 160, 168, 181, 186, 235, 243, 246, 252, 260, 268, 311, 326, 329, 332, 342, 508], "tvtensor": [0, 8, 9, 12, 16, 181, 232, 239, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 278, 311, 325, 326, 327, 328, 330, 331, 332, 340, 508], "faq": [0, 8, 10, 13, 16, 325, 326, 327, 328, 329, 330, 331, 332, 509], "class": [0, 2, 3, 4, 5, 6, 8, 9, 11, 12, 13, 15, 16, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 95, 96, 97, 98, 99, 103, 110, 113, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 153, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 296, 311, 322, 325, 326, 327, 328, 329, 330, 331, 332, 340, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "optic": [0, 1, 7, 49, 50, 53, 64, 77, 85, 96, 104, 220, 227, 336, 340, 394, 395, 486], "flow": [0, 1, 7, 18, 49, 50, 53, 64, 84, 85, 96, 104, 220, 227, 336, 340, 394, 395, 486], "predict": [0, 1, 4, 6, 7, 96, 104, 153, 220, 227, 336, 342, 361, 362, 363, 365, 366, 367, 369, 371, 372, 394, 395], "movement": [0, 1, 7, 96, 104, 220, 227, 336, 394, 395], "raft": [0, 1, 7, 96, 104, 220, 227, 336, 342, 394, 395], "model": [0, 1, 3, 4, 5, 7, 12, 15, 42, 77, 96, 104, 110, 111, 112, 113, 114, 115, 121, 163, 190, 191, 220, 227, 267, 274, 275, 331, 336, 340, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 506, 507, 508], "repurpos": [0, 1, 7, 103, 146, 204, 227, 333, 335, 363], "mask": [0, 1, 7, 12, 13, 15, 23, 27, 39, 49, 50, 53, 62, 63, 64, 85, 94, 103, 118, 139, 146, 151, 193, 204, 227, 239, 248, 251, 256, 264, 269, 273, 275, 278, 282, 333, 335, 340, 342, 363, 366, 367, 368, 477, 508], "bound": [0, 1, 7, 9, 12, 13, 15, 31, 103, 124, 133, 145, 146, 186, 204, 227, 232, 233, 236, 239, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 275, 276, 325, 326, 333, 335, 342, 363, 506, 508], "box": [0, 1, 7, 9, 12, 13, 15, 21, 22, 23, 31, 94, 103, 124, 132, 133, 134, 135, 136, 137, 138, 140, 141, 144, 145, 146, 147, 148, 149, 150, 151, 152, 203, 204, 205, 221, 227, 232, 233, 236, 239, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 275, 276, 325, 326, 333, 335, 340, 342, 363, 365, 366, 367, 369, 371, 508], "torchscript": [0, 1, 7, 103, 121, 160, 166, 169, 170, 176, 177, 181, 183, 186, 190, 192, 193, 201, 209, 216, 217, 218, 220, 223, 227, 228, 235, 243, 247, 248, 255, 256, 263, 268, 274, 279, 280, 282, 309, 321, 340, 341, 425, 506], "support": [0, 1, 2, 7, 10, 12, 13, 15, 35, 36, 41, 42, 48, 51, 52, 57, 60, 70, 72, 73, 77, 87, 90, 91, 94, 95, 96, 98, 103, 121, 134, 155, 156, 159, 160, 161, 162, 166, 167, 168, 169, 170, 171, 173, 176, 177, 179, 181, 183, 184, 186, 187, 190, 192, 193, 194, 197, 198, 201, 204, 207, 209, 215, 216, 217, 218, 220, 221, 222, 223, 226, 227, 228, 230, 231, 234, 235, 237, 239, 243, 244, 246, 247, 248, 249, 251, 255, 256, 258, 263, 264, 267, 268, 269, 270, 274, 276, 279, 280, 282, 283, 307, 309, 321, 340, 341, 342, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 425, 506], "video": [0, 1, 7, 12, 15, 54, 60, 70, 88, 94, 96, 104, 105, 109, 117, 160, 190, 230, 231, 232, 234, 239, 240, 242, 244, 246, 248, 249, 250, 251, 253, 254, 256, 257, 258, 259, 260, 261, 262, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 277, 278, 283, 284, 307, 326, 340, 457, 458, 459, 460, 461, 462, 463, 464, 465, 507, 508], "api": [0, 1, 2, 7, 9, 10, 13, 17, 33, 34, 94, 96, 160, 190, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 342], "visual": [0, 1, 3, 7, 14, 18, 42, 103, 227, 333, 334, 335, 337, 342, 363, 366, 367, 435, 510], "util": [0, 1, 2, 3, 5, 7, 11, 12, 17, 18, 60, 103, 111, 134, 227, 333, 334, 335, 336, 337, 338, 340, 342, 363, 366, 367, 435, 506, 508], "download": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 25, 26, 28, 29, 31, 33, 34, 35, 36, 38, 40, 42, 45, 46, 47, 48, 49, 51, 52, 55, 57, 59, 60, 61, 65, 66, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 88, 89, 90, 91, 93, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "all": [0, 2, 3, 4, 5, 6, 9, 12, 13, 14, 15, 17, 18, 28, 31, 32, 36, 37, 42, 51, 54, 55, 60, 70, 72, 76, 78, 88, 93, 94, 96, 110, 111, 115, 116, 117, 122, 124, 125, 126, 127, 128, 129, 130, 131, 132, 147, 155, 156, 161, 162, 170, 171, 173, 174, 177, 179, 184, 187, 188, 194, 201, 216, 217, 223, 224, 230, 231, 235, 239, 248, 249, 252, 255, 256, 258, 262, 263, 271, 275, 278, 281, 283, 329, 333, 335, 337, 341, 343, 344, 345, 346, 347, 348, 349, 350, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 394, 395, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506, 508], "python": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 104, 110, 159, 162, 164, 186, 201, 209, 217, 234, 239, 241, 255, 258, 262, 265, 268, 270, 273, 276, 326, 340], "sourc": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 340, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "code": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 35, 60, 96, 111, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 508], "auto_examples_python": 0, "zip": [0, 2, 4, 6, 12, 37, 71, 93], "jupyt": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15], "notebook": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15], "auto_examples_jupyt": 0, "galleri": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 342, 508], "gener": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 27, 28, 30, 37, 39, 43, 44, 49, 50, 56, 58, 60, 62, 63, 77, 84, 85, 144, 145, 159, 162, 198, 225, 234, 239, 265, 333, 335, 340, 367, 507, 508], "sphinx": [0, 2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15], "try": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 161, 204, 237, 275], "collab": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15], "go": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 384, 396, 473, 474, 508], "full": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 55, 70, 76, 94, 114, 240, 335], "exampl": [2, 3, 4, 6, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 27, 30, 32, 33, 34, 39, 43, 44, 49, 50, 53, 54, 58, 60, 62, 63, 64, 69, 76, 84, 85, 86, 88, 92, 94, 96, 103, 104, 110, 111, 112, 114, 121, 124, 139, 146, 148, 149, 151, 152, 153, 155, 156, 157, 158, 159, 160, 162, 163, 164, 165, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 200, 201, 202, 203, 204, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 223, 224, 226, 227, 229, 230, 231, 232, 234, 235, 238, 239, 240, 241, 242, 245, 246, 247, 248, 249, 250, 251, 252, 253, 255, 256, 257, 258, 260, 261, 262, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 283, 307, 311, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 341, 342, 361, 362, 363, 365, 366, 367, 369, 371, 372, 394, 395, 425, 435, 508], "task": [2, 3, 12, 13, 18, 267, 342, 508], "between": [2, 4, 12, 14, 48, 54, 60, 69, 88, 96, 98, 100, 101, 107, 108, 110, 111, 116, 117, 132, 135, 137, 139, 140, 141, 144, 147, 201, 262, 333, 334, 335, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 508], "two": [2, 3, 5, 6, 13, 17, 39, 54, 60, 77, 88, 94, 96, 111, 135, 137, 138, 140, 141, 144, 145, 151, 190, 217, 273, 326, 341], "imag": [2, 3, 4, 5, 9, 11, 12, 14, 15, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 61, 62, 63, 64, 65, 66, 67, 68, 69, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 89, 90, 91, 92, 93, 94, 95, 97, 98, 99, 100, 101, 103, 107, 108, 116, 124, 130, 136, 139, 148, 149, 151, 152, 155, 156, 158, 159, 160, 161, 162, 163, 164, 165, 167, 168, 169, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 237, 238, 239, 240, 241, 242, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 256, 257, 258, 259, 260, 261, 262, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 282, 283, 296, 307, 309, 320, 321, 326, 328, 331, 333, 334, 335, 336, 337, 338, 340, 342, 345, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472, 488, 489, 499, 504, 507], "usual": 2, "consecut": [2, 161, 204, 237, 238, 245], "frame": [2, 5, 54, 60, 70, 88, 96, 104, 105, 109, 284, 341, 457, 458, 459, 460, 461, 462, 463, 464, 465], "take": [2, 3, 10, 11, 15, 25, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 100, 101, 107, 108, 121, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 238, 245, 249, 252, 255, 263, 275, 283, 307, 341], "input": [2, 3, 4, 6, 15, 17, 18, 32, 33, 34, 61, 78, 90, 91, 97, 98, 99, 100, 101, 107, 108, 116, 117, 118, 119, 120, 121, 122, 123, 125, 126, 127, 128, 129, 130, 131, 134, 139, 142, 143, 148, 149, 151, 152, 153, 154, 155, 156, 159, 162, 163, 164, 165, 168, 170, 171, 173, 174, 177, 179, 180, 184, 186, 187, 190, 191, 192, 193, 194, 198, 201, 206, 207, 208, 215, 216, 217, 220, 221, 222, 223, 225, 227, 230, 231, 232, 234, 235, 237, 238, 239, 240, 241, 242, 245, 246, 248, 249, 250, 251, 252, 253, 255, 256, 257, 258, 259, 260, 262, 263, 264, 265, 266, 267, 268, 269, 270, 272, 273, 274, 275, 276, 277, 278, 280, 282, 283, 307, 321, 331, 333, 334, 335, 336, 340, 342, 363, 365, 366, 367, 369, 371, 372, 453, 509], "indic": [2, 6, 53, 64, 73, 132, 147, 150, 151, 238, 245, 275, 284, 307, 394, 395], "displac": [2, 162, 239, 298], "everi": [2, 5, 13, 15, 54, 60, 88, 96, 122, 125, 126, 127, 128, 129, 130, 131, 135, 137, 140, 144, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283, 341, 342, 471, 472], "singl": [2, 4, 6, 9, 10, 12, 13, 44, 96, 148, 149, 151, 152, 165, 170, 176, 177, 179, 186, 190, 201, 203, 209, 216, 217, 220, 222, 223, 226, 235, 243, 248, 252, 255, 256, 258, 262, 265, 268, 270, 273, 274, 333, 335, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 496, 497], "pixel": [2, 3, 6, 14, 15, 51, 75, 89, 148, 149, 151, 155, 156, 159, 162, 170, 171, 173, 175, 177, 179, 184, 187, 188, 192, 194, 198, 201, 202, 216, 217, 223, 224, 227, 230, 231, 234, 239, 248, 249, 251, 253, 256, 258, 264, 265, 269, 271, 273, 280, 283, 321, 337, 394, 395], "first": [2, 3, 5, 6, 13, 54, 60, 70, 88, 94, 96, 121, 135, 137, 138, 140, 141, 142, 143, 144, 145, 148, 149, 151, 152, 154, 201, 269, 275, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 488, 508], "map": [2, 18, 21, 22, 23, 37, 110, 119, 120, 121, 124, 148, 149, 151, 152, 161, 204, 206, 237, 239, 248, 251, 256, 264, 269, 273, 342, 508], "its": [2, 3, 4, 6, 10, 14, 15, 32, 33, 34, 37, 56, 57, 61, 75, 78, 90, 91, 100, 101, 114, 127, 142, 143, 154, 167, 179, 190, 202, 214, 219, 220, 224, 244, 258, 274, 342, 394, 395, 508], "correspond": [2, 3, 6, 12, 13, 15, 18, 35, 37, 61, 74, 94, 100, 101, 110, 121, 124, 130, 132, 148, 149, 151, 152, 162, 173, 184, 186, 187, 190, 201, 208, 217, 220, 221, 223, 225, 233, 239, 251, 262, 264, 267, 268, 269, 270, 274, 275, 276, 326, 336, 342, 394, 395, 507], "second": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 94, 96, 109, 135, 137, 138, 140, 141, 144, 145, 201, 238, 245, 275, 423, 424, 427, 488], "ar": [2, 3, 4, 5, 6, 9, 11, 12, 14, 17, 18, 27, 30, 31, 32, 33, 34, 37, 39, 42, 44, 46, 47, 48, 49, 50, 51, 53, 55, 56, 58, 60, 61, 62, 63, 64, 69, 71, 74, 75, 78, 84, 85, 86, 92, 93, 94, 96, 97, 98, 99, 103, 110, 111, 115, 116, 117, 121, 122, 124, 132, 133, 134, 135, 137, 138, 140, 141, 144, 145, 146, 147, 148, 149, 151, 152, 155, 156, 157, 159, 161, 162, 171, 173, 184, 186, 187, 188, 190, 192, 193, 194, 197, 198, 201, 204, 206, 217, 220, 221, 223, 224, 230, 231, 234, 235, 236, 237, 238, 239, 243, 245, 249, 251, 252, 264, 267, 268, 269, 270, 271, 274, 275, 276, 280, 282, 283, 284, 325, 327, 332, 333, 335, 337, 338, 340, 341, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 480, 498, 506, 507, 508, 509], "2": [2, 3, 4, 6, 10, 11, 12, 13, 14, 15, 18, 19, 20, 22, 24, 49, 50, 53, 54, 60, 62, 64, 85, 88, 96, 116, 117, 118, 124, 139, 153, 164, 170, 171, 172, 173, 177, 192, 195, 196, 199, 200, 216, 241, 248, 249, 250, 251, 256, 262, 276, 280, 307, 331, 334, 336, 337, 339, 342, 357, 361, 362, 363, 365, 368, 375, 376, 390, 394, 395, 400, 405, 407, 417, 431, 432, 436, 440, 442, 445, 460, 461, 468, 471, 472, 508], "h": [2, 5, 6, 9, 15, 27, 30, 39, 44, 49, 50, 53, 54, 58, 60, 62, 63, 64, 69, 70, 78, 84, 85, 86, 88, 104, 109, 122, 124, 134, 142, 143, 146, 148, 149, 151, 152, 155, 156, 158, 159, 163, 164, 165, 167, 169, 170, 171, 172, 173, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 207, 208, 209, 213, 214, 215, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 234, 239, 240, 241, 242, 244, 247, 248, 249, 250, 251, 253, 256, 257, 258, 259, 260, 261, 262, 264, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 277, 280, 282, 283, 284, 300, 333, 334, 335, 336, 337, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "w": [2, 6, 9, 15, 27, 30, 39, 44, 49, 50, 53, 54, 58, 60, 62, 63, 64, 69, 70, 78, 84, 85, 86, 88, 104, 109, 122, 124, 134, 142, 143, 146, 148, 149, 151, 152, 155, 156, 158, 159, 163, 164, 165, 167, 169, 170, 171, 172, 173, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 207, 208, 209, 213, 214, 215, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 234, 239, 240, 241, 242, 244, 247, 248, 249, 250, 251, 253, 256, 257, 258, 259, 260, 261, 262, 264, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 277, 280, 282, 283, 284, 300, 333, 334, 335, 336, 337, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "dimension": [2, 3, 96, 97, 98, 99, 100, 101, 103], "tensor": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 31, 54, 60, 70, 74, 88, 94, 96, 97, 98, 99, 100, 101, 102, 103, 104, 106, 107, 108, 109, 118, 119, 120, 121, 122, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 158, 159, 160, 161, 162, 163, 164, 165, 167, 168, 169, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 234, 235, 238, 239, 240, 241, 242, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 256, 257, 258, 259, 260, 261, 262, 264, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 279, 280, 281, 282, 283, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 323, 324, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508, 509], "where": [2, 3, 5, 6, 9, 12, 13, 15, 25, 26, 28, 29, 30, 31, 32, 33, 34, 35, 38, 40, 41, 43, 44, 45, 46, 47, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 70, 71, 74, 76, 78, 79, 80, 81, 83, 84, 86, 87, 88, 89, 90, 91, 93, 100, 101, 104, 107, 108, 109, 122, 132, 146, 148, 149, 151, 152, 155, 156, 158, 159, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 234, 238, 239, 241, 242, 245, 248, 249, 250, 251, 253, 256, 257, 259, 261, 264, 266, 269, 271, 273, 283, 284, 329, 334, 336, 337, 363, 365, 366, 367, 369, 371, 508], "axi": [2, 5, 173, 201, 251, 386, 480], "horizont": [2, 10, 14, 173, 181, 191, 201, 205, 213, 221, 225, 251, 260, 277], "vertic": [2, 14, 173, 189, 191, 201, 205, 221, 225, 229, 251, 272, 277], "The": [2, 3, 4, 5, 6, 9, 10, 11, 12, 14, 15, 18, 27, 30, 31, 32, 35, 36, 38, 39, 41, 42, 44, 48, 49, 50, 51, 52, 53, 54, 57, 58, 60, 62, 63, 64, 65, 66, 69, 70, 72, 73, 74, 75, 76, 77, 78, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 94, 96, 97, 98, 99, 103, 104, 111, 112, 113, 114, 115, 119, 120, 121, 123, 124, 127, 138, 141, 142, 143, 145, 146, 148, 149, 151, 152, 153, 154, 155, 159, 162, 171, 173, 184, 186, 187, 190, 194, 197, 198, 201, 206, 210, 211, 212, 217, 219, 220, 221, 223, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 307, 332, 333, 334, 335, 337, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506, 508, 510], "follow": [2, 3, 5, 6, 13, 14, 18, 27, 30, 37, 39, 44, 49, 50, 53, 54, 58, 60, 61, 62, 63, 64, 69, 84, 85, 86, 88, 93, 96, 124, 163, 167, 191, 197, 244, 262, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506, 508], "can": [2, 3, 4, 6, 9, 10, 11, 12, 13, 14, 17, 18, 28, 31, 32, 37, 50, 54, 55, 56, 60, 65, 66, 72, 73, 75, 76, 88, 90, 91, 94, 96, 104, 110, 111, 116, 117, 123, 124, 163, 164, 172, 182, 188, 190, 191, 195, 196, 197, 198, 199, 200, 202, 206, 209, 213, 214, 219, 220, 222, 224, 226, 229, 232, 238, 239, 240, 241, 245, 248, 250, 251, 256, 260, 261, 262, 264, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 326, 327, 328, 330, 331, 332, 333, 334, 335, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 508], "our": [2, 5, 6, 9, 10, 12, 13, 18, 239, 352, 508], "implement": [2, 4, 5, 10, 13, 14, 15, 17, 96, 123, 130, 142, 143, 151, 154, 193, 235, 238, 239, 243, 245, 252, 282, 340, 343, 352, 506, 508], "we": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 15, 18, 54, 60, 74, 83, 88, 89, 94, 96, 110, 111, 115, 151, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 331, 340, 341, 507, 508], "also": [2, 4, 5, 6, 10, 11, 12, 13, 14, 15, 17, 18, 28, 31, 32, 49, 55, 72, 90, 91, 94, 110, 190, 197, 220, 238, 239, 245, 248, 251, 256, 264, 265, 269, 273, 274, 275, 329, 333, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "see": [2, 5, 6, 10, 12, 14, 15, 18, 50, 85, 88, 94, 97, 98, 99, 103, 110, 111, 118, 119, 120, 125, 126, 128, 129, 130, 131, 160, 162, 163, 179, 190, 191, 193, 197, 198, 215, 218, 220, 227, 228, 235, 237, 238, 239, 240, 245, 258, 269, 274, 277, 278, 282, 286, 290, 291, 292, 293, 294, 295, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 323, 324, 327, 329, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "convert": [2, 6, 12, 13, 14, 94, 95, 97, 98, 99, 103, 134, 161, 165, 169, 180, 192, 193, 198, 204, 218, 222, 226, 227, 228, 236, 237, 242, 247, 259, 278, 279, 280, 281, 282, 309, 321, 332, 336, 508], "rgb": [2, 40, 62, 95, 97, 98, 99, 103, 155, 156, 171, 175, 178, 182, 185, 188, 192, 193, 194, 197, 202, 206, 214, 219, 222, 224, 226, 230, 231, 249, 253, 257, 261, 266, 271, 280, 282, 283, 333, 334, 335, 336], "import": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 33, 96, 138, 258, 341, 342, 385, 397, 508], "numpi": [2, 3, 6, 27, 30, 39, 44, 49, 50, 53, 58, 62, 63, 64, 69, 78, 84, 85, 86, 192, 193, 227, 228, 280, 282, 321], "np": [2, 3, 6, 193, 282], "torch": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 17, 18, 60, 70, 76, 94, 96, 98, 110, 111, 116, 117, 121, 123, 124, 130, 139, 155, 156, 158, 159, 160, 161, 163, 164, 165, 167, 168, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 181, 184, 185, 186, 187, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 208, 209, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 226, 230, 231, 232, 234, 235, 237, 239, 240, 242, 244, 246, 248, 249, 250, 251, 252, 253, 255, 256, 257, 258, 259, 260, 262, 263, 264, 266, 267, 268, 269, 270, 272, 273, 274, 276, 277, 278, 280, 282, 283, 296, 307, 318, 326, 327, 328, 329, 330, 331, 332, 336, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508, 509], "matplotlib": [2, 3, 4, 5, 6, 13, 14], "pyplot": [2, 3, 4, 5, 6, 13, 14], "plt": [2, 3, 4, 5, 6, 13, 14], "transform": [2, 3, 4, 5, 6, 10, 11, 15, 16, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 103, 112, 146, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 310, 311, 312, 313, 314, 315, 316, 317, 319, 322, 323, 324, 326, 327, 328, 329, 330, 331, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 480, 486, 490, 491, 498, 500, 503, 504, 507, 509], "function": [2, 3, 4, 6, 10, 12, 13, 15, 25, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 96, 105, 110, 113, 116, 117, 122, 123, 125, 126, 127, 128, 129, 130, 131, 138, 159, 160, 161, 166, 174, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 234, 235, 237, 243, 249, 252, 255, 263, 265, 275, 283, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 310, 311, 312, 313, 314, 315, 316, 317, 319, 322, 323, 324, 331, 340, 341, 342, 506], "f": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 18, 159, 193, 198, 234, 282, 307, 342], "rcparam": [2, 3, 4, 6, 13, 14], "savefig": [2, 3, 4, 6, 13, 14], "bbox": [2, 3, 4, 6, 9, 12, 13, 14, 15, 31, 61, 94, 326], "tight": [2, 3, 4, 6, 13, 14], "def": [2, 3, 4, 5, 6, 9, 10, 11, 18, 110, 240], "plot": [2, 3, 4, 6, 12, 13, 14, 333, 342, 367], "img": [2, 3, 6, 9, 11, 12, 13, 14, 15, 32, 33, 155, 156, 158, 159, 163, 164, 165, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 226, 229, 230, 231, 234, 249, 253, 256, 257, 258, 261, 266, 268, 271, 283, 326, 331, 333, 334, 335, 336, 342, 508], "imshow_kwarg": 2, "isinst": [2, 3, 6, 9, 13, 15], "0": [2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 19, 20, 22, 31, 32, 43, 54, 60, 65, 69, 81, 83, 88, 89, 96, 97, 98, 99, 101, 103, 104, 108, 110, 118, 123, 124, 132, 133, 135, 136, 137, 138, 139, 140, 141, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 155, 158, 159, 160, 162, 163, 164, 170, 172, 173, 174, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 191, 193, 195, 196, 198, 199, 200, 203, 205, 206, 208, 209, 216, 219, 221, 225, 230, 232, 234, 235, 238, 239, 240, 241, 245, 248, 250, 251, 252, 253, 256, 257, 258, 259, 260, 261, 262, 264, 265, 266, 268, 269, 271, 272, 273, 275, 276, 277, 282, 307, 333, 334, 335, 337, 341, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 495, 507, 508], "list": [2, 3, 6, 9, 12, 14, 18, 28, 31, 32, 37, 46, 47, 55, 61, 67, 72, 94, 105, 110, 111, 115, 121, 123, 124, 127, 128, 129, 148, 149, 151, 152, 155, 156, 159, 160, 163, 171, 173, 174, 176, 179, 183, 184, 186, 187, 194, 201, 203, 208, 209, 210, 212, 215, 216, 217, 220, 221, 223, 225, 234, 235, 251, 252, 255, 258, 262, 263, 264, 268, 269, 270, 291, 293, 298, 301, 302, 305, 306, 307, 308, 312, 313, 315, 317, 329, 333, 334, 335, 337, 338, 341, 363, 365, 366, 367, 369, 371, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 508], "make": [2, 5, 9, 10, 12, 13, 14, 15, 89, 92, 96, 98, 110, 111, 160, 168, 174, 179, 186, 187, 190, 197, 215, 220, 221, 223, 235, 246, 252, 258, 267, 268, 269, 270, 274, 276, 337, 508], "2d": [2, 163, 191], "grid": [2, 148, 151, 162, 239, 337, 338], "even": [2, 14, 15, 342], "s": [2, 4, 5, 6, 11, 12, 13, 15, 18, 90, 94, 110, 115, 238, 245, 269, 307, 353, 354, 355, 356, 374, 382, 383, 388, 390, 391, 392, 394, 395, 398, 399, 401, 402, 403, 406, 407, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 427, 428, 429, 430, 439, 440, 466, 467, 469, 470, 471, 472, 507], "just": [2, 6, 10, 11, 12, 15, 18, 508], "1": [2, 3, 4, 6, 9, 10, 11, 12, 13, 14, 15, 18, 19, 20, 21, 22, 27, 30, 31, 36, 39, 44, 54, 58, 60, 62, 63, 65, 69, 83, 84, 86, 88, 89, 98, 99, 100, 101, 107, 108, 110, 116, 117, 118, 121, 124, 130, 139, 148, 149, 151, 152, 153, 155, 156, 158, 159, 163, 164, 165, 168, 170, 171, 172, 173, 175, 177, 178, 179, 180, 182, 184, 185, 186, 188, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 206, 208, 209, 214, 216, 217, 219, 220, 222, 223, 224, 225, 226, 230, 231, 232, 234, 238, 240, 241, 242, 245, 246, 248, 249, 250, 251, 253, 255, 256, 257, 258, 259, 261, 262, 264, 265, 266, 268, 271, 273, 274, 275, 276, 277, 280, 282, 283, 287, 307, 314, 319, 333, 335, 337, 339, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "row": [2, 14, 142, 143, 154, 337], "num_row": 2, "len": [2, 3, 5, 6, 33, 179, 240, 258, 307, 363], "num_col": 2, "_": [2, 5, 10, 14, 18, 111, 342], "ax": [2, 3, 6], "subplot": [2, 3, 5, 6], "nrow": [2, 337], "ncol": [2, 3, 6], "squeez": [2, 3, 6, 130, 342], "fals": [2, 3, 4, 5, 6, 25, 26, 28, 29, 31, 35, 36, 40, 42, 45, 48, 51, 52, 55, 59, 60, 61, 65, 66, 68, 69, 70, 71, 72, 73, 74, 75, 77, 78, 81, 82, 83, 87, 89, 90, 91, 93, 96, 110, 111, 119, 120, 128, 142, 143, 151, 168, 177, 179, 186, 187, 190, 191, 207, 215, 220, 221, 223, 225, 246, 256, 258, 267, 268, 269, 270, 274, 276, 277, 278, 295, 300, 305, 315, 317, 318, 326, 327, 328, 330, 333, 337, 342, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "row_idx": 2, "enumer": [2, 3, 4, 5, 6, 18, 342], "col_idx": 2, "to_pil_imag": [2, 3, 6, 342], "cpu": [2, 4, 98, 147, 326, 327, 328, 330, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 508], "imshow": [2, 3, 5, 6], "asarrai": [2, 3, 6], "set": [2, 3, 6, 15, 18, 25, 28, 29, 32, 35, 51, 65, 66, 71, 76, 78, 82, 83, 96, 110, 111, 115, 121, 133, 135, 137, 138, 140, 141, 142, 143, 144, 145, 148, 149, 151, 152, 173, 179, 207, 235, 243, 251, 252, 258, 269, 331, 341, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 508], "xticklabel": [2, 3, 6], "yticklabel": [2, 3, 6], "xtick": [2, 3, 6], "ytick": [2, 3, 6], "tight_layout": 2, "read_video": [2, 341, 342], "altern": [2, 14, 15, 94], "one": [2, 3, 5, 6, 10, 11, 13, 18, 32, 38, 42, 55, 65, 66, 72, 75, 81, 96, 97, 98, 99, 100, 101, 102, 105, 106, 110, 122, 124, 125, 126, 127, 128, 129, 130, 131, 132, 142, 143, 147, 150, 154, 161, 174, 193, 230, 231, 235, 238, 245, 249, 252, 255, 263, 269, 275, 282, 283, 326, 340, 342, 343, 363, 365, 366, 367, 369, 371], "new": [2, 6, 10, 13, 110, 121, 342, 353, 354, 355, 356, 374, 382, 383, 388, 390, 391, 392, 398, 399, 401, 402, 403, 406, 407, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 427, 428, 429, 430, 439, 440, 469, 471, 472, 508], "videoread": [2, 5, 341], "built": [2, 10, 15, 35, 49, 50, 85, 275, 340], "from": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 25, 28, 29, 31, 35, 36, 37, 38, 40, 42, 45, 48, 49, 50, 51, 52, 54, 55, 56, 57, 59, 60, 61, 62, 63, 65, 66, 68, 69, 70, 71, 72, 73, 74, 76, 77, 78, 79, 80, 81, 82, 83, 87, 88, 89, 90, 91, 93, 96, 104, 110, 111, 115, 116, 117, 121, 122, 124, 130, 134, 142, 143, 148, 149, 151, 152, 154, 155, 156, 159, 161, 162, 167, 173, 176, 184, 187, 204, 225, 230, 231, 234, 236, 237, 244, 251, 255, 258, 262, 264, 265, 267, 269, 273, 284, 326, 327, 328, 330, 337, 338, 340, 343, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 458, 459, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "here": [2, 3, 5, 6, 9, 11, 12, 13, 18, 57, 88, 89, 110, 111, 340, 342, 394, 395], "free": [2, 194, 283, 365, 370], "pexel": 2, "com": [2, 4, 5, 12, 13, 14, 15, 94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 507], "credit": 2, "pavel": 2, "danilyuk": 2, "tempfil": [2, 4], "pathlib": [2, 4, 6, 12, 13, 14], "path": [2, 3, 4, 5, 6, 12, 13, 14, 17, 18, 33, 34, 37, 46, 47, 54, 56, 57, 75, 88, 96, 102, 103, 104, 105, 106, 107, 108, 109, 110, 338, 341], "urllib": 2, "request": [2, 110, 114, 333], "urlretriev": 2, "video_url": 2, "http": [2, 3, 4, 5, 6, 12, 13, 14, 15, 69, 74, 94, 130, 138, 141, 142, 143, 145, 153, 179, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 497, 507], "pytorch": [2, 3, 4, 5, 6, 11, 12, 13, 14, 15, 18, 83, 94, 147, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 342, 507], "org": [2, 3, 4, 6, 130, 138, 141, 142, 143, 145, 153, 179, 258, 497], "tutori": [2, 3, 12, 340, 394, 395, 508], "pexelscom_pavel_danilyuk_basketball_hd": 2, "mp4": [2, 5, 60], "video_path": [2, 5, 96, 341], "mkdtemp": 2, "basketbal": 2, "return": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 110, 111, 112, 113, 114, 115, 119, 120, 121, 124, 127, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 158, 159, 162, 163, 164, 165, 167, 168, 170, 171, 172, 173, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 231, 234, 238, 240, 241, 245, 251, 256, 258, 264, 268, 269, 275, 277, 282, 309, 321, 329, 331, 333, 334, 335, 336, 337, 340, 341, 363, 365, 366, 367, 369, 371, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "audio": [2, 5, 54, 60, 88, 94, 96, 104, 109], "metadata": [2, 5, 15, 96, 104, 281, 341], "associ": [2, 6, 13, 113, 275, 281], "In": [2, 3, 5, 6, 9, 10, 11, 12, 13, 15, 18, 37, 48, 78, 94, 96, 110, 111, 115, 139, 160, 163, 170, 174, 177, 186, 190, 191, 193, 201, 209, 216, 217, 220, 223, 235, 238, 245, 248, 252, 256, 262, 268, 269, 274, 282, 327, 333, 341, 367, 385, 397, 507, 508], "case": [2, 3, 5, 6, 9, 10, 12, 15, 37, 74, 78, 94, 108, 110, 111, 115, 116, 117, 139, 193, 275, 282, 331, 342, 361], "onli": [2, 3, 5, 6, 9, 10, 12, 13, 14, 15, 36, 37, 74, 94, 96, 98, 124, 155, 156, 160, 162, 170, 171, 173, 174, 177, 184, 186, 187, 190, 194, 201, 216, 217, 220, 221, 222, 223, 230, 231, 235, 239, 249, 251, 252, 264, 267, 268, 269, 270, 274, 276, 278, 281, 283, 326, 331, 342, 363, 365, 366, 367, 369, 371, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 431, 432, 433, 434, 435, 436, 453, 508], "need": [2, 4, 5, 6, 9, 10, 11, 12, 13, 15, 18, 28, 48, 49, 67, 78, 83, 87, 94, 110, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 238, 245, 249, 252, 255, 263, 275, 283, 340, 507, 508, 509], "pre": [2, 12, 81, 340, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506, 507], "select": [2, 5, 6, 18, 31, 36, 60, 76, 78, 81, 83, 90, 91, 94, 147, 154, 173, 179, 187, 251, 258, 269, 284, 341, 370], "pair": [2, 94, 238, 245, 334, 394, 395, 486, 508], "name": [2, 4, 5, 12, 18, 37, 55, 74, 86, 109, 110, 111, 112, 113, 114, 115, 121, 124, 340, 342, 507], "100": [2, 5, 6, 11, 14, 35, 42, 100, 107, 342], "101": [2, 28, 51, 402, 403, 423, 428, 429, 432, 434, 471], "150": [2, 6, 35, 422], "151": [2, 22, 342, 369, 414], "each": [2, 3, 5, 6, 14, 15, 18, 31, 35, 36, 37, 42, 43, 48, 51, 54, 60, 73, 76, 87, 88, 89, 96, 105, 118, 121, 124, 132, 133, 139, 147, 148, 149, 151, 152, 153, 168, 170, 177, 179, 185, 186, 215, 216, 219, 246, 248, 255, 256, 258, 265, 266, 268, 334, 335, 336, 337, 342, 363, 365, 366, 367, 369, 371, 394, 395, 508], "io": [2, 3, 4, 5, 6, 13, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 341, 342], "str": [2, 4, 6, 12, 13, 14, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 96, 98, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 121, 124, 131, 134, 138, 141, 145, 153, 154, 170, 177, 179, 186, 216, 220, 221, 230, 231, 235, 236, 239, 243, 248, 249, 251, 252, 256, 258, 264, 267, 268, 269, 270, 273, 274, 275, 276, 278, 280, 283, 306, 312, 313, 326, 327, 328, 330, 331, 333, 334, 335, 338, 340], "output_format": [2, 54, 60, 88, 94, 104, 342], "tchw": [2, 54, 60, 88, 94, 104, 342], "img1_batch": 2, "stack": [2, 4, 5, 6, 12, 116, 117, 123, 163, 191, 240], "img2_batch": 2, "vision": [2, 4, 5, 12, 13, 14, 15, 18, 69, 94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 342, 364, 368, 385, 386, 397, 443, 444, 445, 458, 459, 475, 476, 480, 498, 500, 506, 507, 508], "py": [2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 25, 65, 66, 71, 80], "161": [2, 358], "userwarn": [2, 5], "pts_unit": [2, 104, 105], "pt": [2, 5, 38, 96, 104, 105], "give": [2, 54, 60, 88, 110, 172, 195, 196, 198, 199, 200, 250, 508], "wrong": [2, 51], "result": [2, 4, 6, 14, 18, 36, 100, 101, 107, 108, 110, 115, 121, 124, 139, 148, 149, 151, 152, 162, 163, 170, 177, 190, 191, 216, 220, 239, 248, 256, 269, 274, 333, 352, 353, 354, 355, 356, 374, 382, 383, 386, 387, 389, 391, 392, 393, 398, 399, 400, 401, 402, 404, 405, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 430, 437, 438, 441, 442, 443, 444, 445, 446, 447, 448, 471, 472, 508], "pleas": [2, 9, 15, 78, 94, 110, 111, 160, 174, 201, 217, 222, 223, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 508], "sec": [2, 104, 105], "warn": [2, 5, 110, 111, 186, 190, 220, 221, 267, 268, 270, 274, 276, 312, 313, 342], "accept": [2, 5, 9, 10, 18, 162, 173, 184, 186, 187, 190, 201, 217, 220, 221, 223, 235, 239, 243, 251, 252, 264, 267, 268, 269, 270, 274, 276, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "resiz": [2, 5, 75, 186, 221, 267, 268, 270, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 371, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "them": [2, 3, 5, 6, 11, 13, 43, 75, 110, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283, 342, 415, 416, 418, 466, 468, 469, 507, 508], "ensur": [2, 3, 385, 397], "dimens": [2, 3, 6, 61, 102, 106, 123, 127, 138, 141, 142, 143, 145, 146, 148, 149, 151, 152, 154, 155, 156, 158, 159, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 210, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 233, 234, 239, 240, 241, 242, 248, 249, 250, 251, 253, 256, 257, 259, 260, 261, 262, 264, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 277, 283, 284, 457, 458, 459, 460, 461, 462, 463, 464, 465, 508], "divis": [2, 137, 138, 140, 141, 145], "8": [2, 3, 5, 6, 14, 20, 22, 23, 87, 121, 139, 185, 209, 219, 266, 335, 337, 339, 342, 355, 357, 361, 362, 367, 368, 383, 386, 387, 393, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 411, 414, 420, 421, 422, 427, 430, 438, 441, 444, 449, 453, 458, 472], "note": [2, 3, 6, 11, 12, 13, 18, 54, 60, 78, 83, 88, 105, 110, 111, 160, 187, 198, 222, 223, 235, 269, 327, 333, 342, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 508], "explicitli": [2, 96, 341], "antialia": [2, 4, 5, 9, 11, 13, 186, 190, 220, 221, 267, 268, 270, 274, 276, 312, 313, 508], "becaus": [2, 6, 10, 11, 12, 15, 18, 186, 190, 193, 220, 221, 267, 268, 270, 274, 276, 282, 340, 342, 453, 508], "thi": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 18, 26, 28, 35, 37, 38, 48, 51, 54, 55, 56, 57, 60, 67, 70, 71, 73, 76, 77, 78, 83, 84, 87, 88, 94, 96, 98, 110, 111, 116, 117, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 134, 138, 139, 141, 142, 143, 145, 147, 148, 149, 151, 152, 159, 160, 161, 163, 166, 167, 168, 169, 170, 174, 176, 177, 179, 183, 186, 188, 190, 191, 192, 193, 198, 204, 207, 208, 215, 216, 218, 220, 221, 222, 224, 225, 226, 227, 228, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 307, 309, 321, 327, 329, 331, 333, 338, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 423, 424, 425, 426, 427, 428, 429, 430, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 507, 508], "those": [2, 6, 9, 11, 13, 110, 508], "were": [2, 3, 6, 11, 12, 13, 15, 48, 51, 94, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 388, 390, 392, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 429, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 449, 450, 451, 452, 453, 454, 455, 456, 458, 459, 463, 464, 465, 466, 467, 469, 470], "train": [2, 3, 11, 18, 25, 26, 27, 31, 32, 35, 36, 38, 39, 41, 42, 45, 48, 49, 50, 51, 52, 53, 54, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 73, 74, 75, 76, 77, 78, 81, 83, 85, 86, 87, 88, 89, 90, 91, 93, 110, 111, 142, 143, 154, 186, 190, 220, 238, 267, 268, 274, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 508], "Then": [2, 18, 167, 244, 508], "bundl": [2, 18, 342], "weight": [2, 3, 4, 5, 6, 12, 19, 20, 21, 22, 23, 24, 113, 114, 139, 153, 339, 340, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 507], "order": [2, 5, 6, 13, 15, 18, 55, 60, 111, 121, 127, 132, 147, 159, 160, 174, 183, 206, 234, 235, 252, 262, 263, 367, 508], "preprocess": [2, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "rescal": [2, 51, 190, 220, 274, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "valu": [2, 3, 5, 6, 12, 14, 15, 18, 53, 64, 78, 81, 89, 94, 95, 96, 97, 98, 99, 103, 104, 105, 110, 114, 122, 123, 132, 135, 137, 140, 142, 143, 144, 147, 148, 151, 155, 156, 157, 159, 161, 162, 169, 170, 171, 172, 173, 175, 177, 178, 179, 181, 182, 184, 185, 186, 187, 188, 189, 190, 192, 193, 194, 198, 201, 204, 206, 207, 216, 217, 220, 221, 222, 223, 224, 226, 230, 231, 234, 236, 237, 239, 247, 248, 249, 250, 251, 253, 256, 257, 258, 260, 261, 262, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 278, 279, 280, 281, 282, 283, 307, 325, 326, 327, 328, 330, 333, 334, 335, 337, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "requir": [2, 6, 15, 18, 33, 34, 55, 57, 73, 90, 96, 156, 160, 174, 231, 235, 252, 262, 342, 363, 365, 366, 367, 369, 371, 506], "interv": [2, 155, 159, 173, 198, 230, 234, 251, 265, 267], "optical_flow": [2, 50, 394, 395, 486], "raft_large_weight": [2, 394], "default": [2, 3, 4, 5, 6, 11, 12, 15, 28, 31, 35, 36, 37, 39, 40, 41, 43, 48, 49, 50, 51, 52, 53, 54, 55, 56, 60, 61, 62, 63, 64, 65, 66, 69, 70, 72, 73, 74, 75, 76, 77, 84, 85, 87, 88, 90, 93, 94, 96, 97, 98, 99, 100, 101, 103, 104, 105, 107, 108, 110, 111, 116, 117, 121, 122, 123, 130, 137, 138, 139, 140, 141, 142, 143, 145, 148, 149, 151, 152, 153, 154, 155, 156, 162, 170, 171, 172, 173, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 194, 201, 207, 209, 215, 216, 217, 220, 221, 222, 223, 225, 226, 230, 231, 238, 239, 245, 248, 249, 250, 251, 253, 255, 256, 257, 258, 259, 260, 261, 262, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 280, 283, 326, 327, 328, 330, 331, 333, 335, 337, 340, 341, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "size": [2, 3, 4, 5, 6, 9, 11, 12, 13, 14, 15, 33, 43, 54, 60, 88, 110, 116, 117, 121, 122, 124, 136, 139, 142, 143, 145, 148, 149, 150, 151, 152, 158, 163, 164, 177, 186, 187, 190, 191, 203, 205, 207, 208, 209, 212, 215, 220, 221, 223, 225, 232, 240, 241, 256, 262, 267, 268, 269, 270, 273, 274, 276, 277, 284, 301, 312, 313, 317, 329, 333, 337, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 495, 508], "520": [2, 431, 432, 433, 434, 435, 436], "960": 2, "print": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 15, 18, 33, 110, 111, 121, 124, 139, 235, 243, 252, 341, 342], "shape": [2, 3, 6, 9, 11, 12, 13, 14, 15, 18, 27, 30, 31, 39, 44, 49, 50, 53, 58, 62, 63, 64, 69, 78, 84, 85, 86, 110, 121, 124, 139, 153, 155, 156, 158, 159, 163, 164, 165, 167, 169, 170, 171, 172, 173, 175, 177, 178, 180, 181, 184, 185, 186, 187, 189, 190, 191, 192, 193, 194, 201, 203, 205, 208, 209, 216, 217, 220, 221, 222, 223, 225, 230, 231, 232, 234, 238, 239, 240, 241, 242, 244, 245, 247, 248, 249, 250, 251, 253, 256, 257, 259, 260, 262, 264, 266, 267, 268, 269, 270, 272, 273, 274, 276, 277, 280, 282, 283, 284, 329, 333, 334, 335, 336, 337, 363, 365, 366, 367, 369, 371, 508], "dtype": [2, 3, 6, 11, 13, 15, 31, 161, 193, 204, 206, 237, 278, 282, 296, 318, 326, 327, 328, 330, 333, 334, 335, 336], "3": [2, 3, 4, 6, 9, 10, 11, 12, 13, 14, 15, 18, 19, 20, 22, 24, 30, 33, 43, 44, 49, 50, 54, 58, 60, 61, 84, 85, 88, 97, 98, 99, 100, 101, 103, 107, 108, 110, 116, 117, 124, 130, 139, 155, 156, 159, 165, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 182, 185, 188, 192, 194, 195, 196, 197, 198, 199, 200, 202, 206, 209, 214, 216, 219, 222, 224, 226, 230, 231, 234, 239, 240, 242, 248, 249, 250, 251, 252, 253, 256, 257, 258, 259, 261, 262, 264, 266, 269, 271, 273, 280, 283, 307, 329, 331, 334, 335, 336, 342, 359, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 385, 388, 390, 394, 395, 397, 398, 403, 404, 405, 406, 409, 411, 412, 419, 426, 429, 431, 437, 460, 464, 466, 469, 508], "float32": [2, 6, 11, 12, 13, 15, 161, 204, 237, 278, 282, 296, 318, 508], "raft_larg": 2, "which": [2, 3, 5, 6, 10, 12, 14, 15, 17, 18, 36, 38, 42, 44, 49, 50, 53, 54, 55, 60, 64, 73, 78, 83, 84, 85, 88, 92, 96, 98, 104, 105, 110, 111, 112, 113, 115, 116, 117, 119, 120, 121, 123, 124, 133, 134, 147, 150, 159, 167, 190, 202, 206, 220, 234, 244, 262, 274, 331, 340, 342, 471, 472, 507, 509], "same": [2, 4, 5, 6, 9, 12, 13, 14, 15, 18, 31, 56, 65, 96, 111, 124, 138, 139, 141, 145, 147, 153, 169, 187, 190, 197, 209, 218, 220, 221, 223, 225, 238, 245, 247, 255, 259, 269, 274, 275, 309, 331, 332, 337, 471, 472, 508], "architectur": [2, 27, 154, 340, 342, 343, 352, 353, 354, 355, 356, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 437, 438, 439, 440, 441, 443, 444, 445, 446, 447, 448, 458, 459, 463, 464, 465, 466, 467, 468, 469, 470, 475, 476, 481, 493, 494], "describ": [2, 6, 13, 36, 139, 149, 151, 152, 194, 283, 340, 372, 394, 404, 405, 406, 407, 437, 438, 439, 440, 453], "origin": [2, 6, 12, 14, 65, 66, 69, 74, 94, 121, 124, 148, 149, 151, 152, 167, 172, 173, 184, 186, 187, 195, 196, 198, 199, 200, 201, 217, 223, 244, 250, 251, 264, 268, 269, 273, 343, 352, 353, 354, 355, 356, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 391, 392, 393, 394, 395, 396, 397, 398, 399, 401, 402, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 427, 428, 430, 453, 466, 468, 469, 471, 472, 488, 498], "paper": [2, 27, 124, 238, 245, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 361, 362, 363, 364, 365, 366, 367, 368, 369, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 404, 405, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 430, 434, 435, 436, 437, 438, 441, 442, 443, 444, 445, 446, 447, 448, 453, 457, 458, 459, 460, 461, 462, 463, 464, 465, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "provid": [2, 3, 9, 17, 18, 51, 74, 84, 110, 111, 121, 146, 158, 163, 170, 177, 186, 191, 208, 216, 225, 232, 238, 240, 245, 248, 256, 268, 277, 341, 342, 506, 507, 508], "raft_smal": 2, "builder": [2, 112, 113, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "smaller": [2, 150, 158, 161, 177, 190, 197, 203, 204, 205, 220, 232, 237, 256, 274, 348, 381, 382, 383], "faster": [2, 3, 6, 12, 340, 342, 348, 361, 362, 363, 364, 381, 382, 383, 508], "run": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 67, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283, 340, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "sacrif": [2, 442], "bit": [2, 6, 9, 11, 13, 14, 185, 219, 266, 310, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 508], "accuraci": [2, 11, 342, 364, 368, 370, 423, 424, 427, 441, 442, 457, 458, 459, 460, 461, 462, 463, 464, 465, 488, 495, 502, 508], "If": [2, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 25, 28, 29, 30, 31, 35, 36, 37, 38, 40, 42, 44, 45, 48, 49, 50, 51, 52, 54, 55, 58, 59, 60, 61, 65, 66, 68, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 87, 89, 90, 91, 93, 94, 96, 98, 110, 116, 117, 123, 142, 143, 147, 148, 149, 151, 152, 155, 156, 158, 159, 161, 162, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 234, 237, 239, 240, 241, 242, 248, 249, 250, 251, 253, 255, 256, 257, 258, 259, 260, 261, 262, 264, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 280, 283, 307, 326, 327, 328, 330, 332, 333, 337, 338, 340, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "you": [2, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 17, 18, 49, 67, 73, 78, 94, 110, 148, 149, 151, 152, 159, 163, 186, 190, 191, 198, 220, 221, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 329, 331, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "gpu": [2, 147, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "lot": [2, 275, 508], "devic": [2, 4, 5, 15, 98, 240, 326, 327, 328, 329, 330, 340], "cuda": [2, 4, 96, 98, 508], "is_avail": [2, 4], "els": [2, 4, 32, 61, 173, 251], "progress": [2, 3, 4, 6, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "eval": [2, 4, 6, 18, 110, 111, 342, 361, 362, 363, 365, 366, 367, 369, 371, 372], "list_of_flow": 2, "type": [2, 5, 11, 12, 13, 15, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 88, 89, 90, 91, 93, 96, 100, 101, 104, 105, 111, 112, 113, 114, 115, 116, 117, 119, 120, 121, 123, 130, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 154, 155, 156, 158, 159, 161, 162, 164, 165, 167, 168, 169, 170, 171, 172, 173, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 192, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 234, 237, 239, 241, 243, 247, 248, 249, 251, 256, 258, 264, 266, 268, 269, 273, 274, 278, 280, 281, 283, 309, 311, 321, 326, 327, 328, 330, 331, 333, 334, 335, 336, 337, 340], "length": [2, 5, 6, 51, 158, 163, 170, 177, 179, 186, 190, 191, 201, 208, 209, 216, 217, 220, 223, 225, 232, 239, 240, 248, 251, 256, 258, 264, 268, 269, 273, 274, 277, 342], "number": [2, 3, 5, 6, 12, 14, 18, 33, 43, 54, 60, 70, 76, 88, 96, 100, 101, 104, 107, 108, 109, 116, 117, 121, 122, 123, 124, 130, 137, 138, 140, 141, 145, 146, 148, 149, 151, 152, 155, 156, 158, 159, 162, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 211, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 226, 229, 230, 231, 232, 234, 238, 239, 240, 241, 242, 245, 248, 249, 250, 251, 253, 256, 257, 258, 259, 260, 261, 262, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 277, 283, 284, 307, 335, 337, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436, 471, 472, 508], "iter": [2, 5, 96, 115, 147], "raft_large_c_t_skht_v2": 2, "ff5fadd5": 2, "pth": [2, 3, 4, 6], "root": [2, 3, 4, 5, 6, 12, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93], "cach": [2, 3, 4, 6, 342], "hub": [2, 3, 4, 6], "checkpoint": [2, 3, 4, 6, 366], "12": [2, 5, 6, 20, 342, 388, 396, 436, 463, 464, 465], "output": [2, 5, 6, 9, 12, 13, 15, 18, 28, 31, 32, 33, 54, 55, 60, 72, 88, 97, 98, 99, 100, 102, 103, 104, 106, 110, 116, 117, 121, 124, 134, 138, 139, 141, 145, 148, 149, 151, 152, 153, 158, 161, 163, 165, 168, 177, 186, 187, 190, 191, 203, 204, 205, 206, 208, 220, 221, 222, 223, 225, 226, 232, 236, 237, 240, 242, 246, 256, 259, 267, 268, 269, 273, 274, 277, 307, 331, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 404, 405, 406, 407, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 457, 458, 459, 460, 461, 462, 463, 464, 465, 508], "entri": [2, 6, 9, 32, 33, 34, 54, 60, 61, 78, 88, 90, 91, 114], "n": [2, 9, 12, 69, 89, 109, 122, 124, 132, 133, 134, 135, 136, 137, 138, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 154, 168, 246, 333, 334, 336, 363, 365, 366, 367, 369, 371, 385, 397, 508], "batch": [2, 4, 5, 6, 11, 12, 25, 122, 132, 142, 143, 148, 149, 151, 152, 154, 163, 191, 232, 238, 239, 240, 245, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 277, 337, 338, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "given": [2, 5, 6, 10, 13, 14, 27, 30, 37, 39, 44, 49, 50, 53, 54, 56, 57, 58, 60, 61, 62, 63, 64, 69, 75, 84, 85, 86, 87, 88, 110, 113, 134, 155, 156, 158, 159, 161, 162, 163, 164, 167, 168, 170, 171, 172, 173, 174, 175, 177, 178, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 201, 203, 204, 205, 207, 208, 209, 213, 216, 217, 220, 221, 223, 225, 229, 230, 231, 234, 236, 237, 239, 241, 244, 246, 248, 249, 250, 252, 253, 257, 260, 261, 262, 264, 266, 268, 271, 272, 274, 275, 283, 307, 333, 334, 335, 336, 338, 340, 342], "For": [2, 3, 4, 5, 6, 9, 10, 11, 12, 15, 17, 18, 51, 81, 92, 94, 110, 111, 148, 149, 151, 152, 170, 173, 177, 207, 216, 222, 232, 239, 240, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 277, 307, 340, 342, 363, 365, 366, 367, 369, 508], "more": [2, 5, 6, 9, 10, 11, 12, 13, 15, 18, 32, 74, 96, 97, 98, 99, 103, 110, 111, 139, 197, 198, 215, 218, 227, 228, 309, 321, 327, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 508, 509], "detail": [2, 6, 9, 11, 13, 15, 18, 27, 50, 85, 94, 110, 111, 197, 198, 215, 218, 227, 228, 238, 245, 291, 292, 293, 294, 295, 297, 298, 299, 300, 301, 302, 303, 305, 306, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 323, 324, 327, 329, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 508], "natur": [2, 5, 6], "refer": [2, 6, 10, 11, 13, 14, 110, 124, 138, 141, 145, 147, 193, 282, 332, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 509], "interest": [2, 6, 12, 74, 78, 124, 148, 149, 151, 152], "final": [2, 18, 50, 84, 85, 86, 186, 268, 337, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "thei": [2, 3, 5, 6, 9, 11, 12, 13, 15, 17, 18, 71, 94, 96, 110, 111, 121, 124, 132, 133, 136, 147, 238, 245, 332, 341, 342, 371, 394, 395, 453, 507, 508], "most": [2, 4, 6, 9, 11, 13, 15, 42, 60, 94, 170, 177, 216, 238, 245, 275, 342, 508, 509], "accur": 2, "ones": [2, 3, 6, 13, 75, 508], "so": [2, 3, 5, 6, 9, 10, 11, 12, 13, 15, 18, 54, 56, 60, 88, 94, 96, 136, 138, 141, 145, 190, 202, 220, 238, 245, 274, 342, 385, 397, 508], "retriev": [2, 18, 27, 30, 39, 44, 49, 50, 53, 58, 62, 63, 64, 69, 84, 85, 86, 341], "last": [2, 18, 70, 170, 177, 216, 248, 256, 471, 472, 508], "item": [2, 4, 5, 6, 12, 32, 61, 94, 110, 121, 342], "As": [2, 3, 6, 12, 13, 96, 190, 220, 274, 342], "abov": [2, 4, 5, 6, 9, 10, 13, 14, 15, 18, 33, 50, 85, 88, 94, 98, 110, 111, 188, 224, 271, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "unit": [2, 48, 104, 105], "normal": [2, 6, 9, 11, 13, 116, 117, 121, 123, 138, 159, 160, 179, 198, 234, 235, 255, 258, 265, 337, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "r": [2, 3, 6, 148, 149, 151, 152, 165, 170, 177, 179, 180, 216, 222, 226, 239, 248, 251, 256, 258, 264, 269, 273, 342, 361, 362, 363, 364, 366, 367, 368, 436, 460], "t": [2, 5, 6, 10, 11, 12, 13, 15, 54, 60, 70, 88, 89, 104, 109, 110, 111, 116, 117, 121, 123, 124, 141, 167, 186, 190, 220, 221, 244, 255, 267, 268, 270, 274, 276, 281, 284, 307, 329, 453, 457, 458, 459, 460, 461, 462, 463, 464, 465], "predicted_flow": 2, "min": [2, 6, 159, 164, 173, 187, 234, 241, 251, 265, 269, 276, 337], "max": [2, 6, 159, 164, 173, 187, 234, 241, 251, 265, 269, 337], "8997180461883545": 2, "6": [2, 5, 6, 14, 19, 20, 21, 23, 38, 98, 101, 108, 124, 142, 143, 339, 342, 359, 361, 362, 365, 366, 367, 372, 381, 387, 391, 396, 398, 399, 402, 403, 406, 408, 415, 419, 428, 439, 448, 452, 467, 468], "400400161743164": 2, "flow_to_imag": [2, 340], "an": [2, 3, 4, 6, 9, 12, 13, 14, 15, 18, 25, 28, 29, 31, 33, 34, 35, 37, 38, 40, 41, 42, 43, 45, 48, 51, 52, 54, 55, 56, 57, 59, 60, 65, 66, 67, 68, 70, 71, 74, 75, 76, 77, 78, 80, 81, 82, 83, 87, 88, 89, 90, 91, 92, 94, 96, 97, 100, 101, 106, 107, 108, 110, 112, 121, 122, 136, 138, 139, 141, 142, 143, 145, 147, 148, 149, 151, 152, 155, 156, 158, 159, 162, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 210, 211, 212, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 227, 229, 230, 231, 232, 234, 238, 239, 240, 241, 242, 245, 249, 250, 251, 253, 254, 256, 257, 259, 261, 262, 265, 266, 268, 271, 274, 277, 280, 283, 321, 331, 336, 338, 340, 342, 343, 347, 348, 364, 366, 368, 370, 381, 382, 383, 431, 432, 433, 434, 435, 436, 466, 467, 468, 469, 470, 475, 480, 481, 498, 502, 504, 507, 508, 509], "It": [2, 3, 10, 13, 14, 15, 18, 33, 34, 61, 92, 96, 110, 121, 124, 177, 186, 190, 220, 221, 238, 245, 256, 267, 268, 270, 274, 275, 276, 331, 340, 341, 342, 361, 362, 364, 394, 395], "direct": [2, 18, 198, 201, 203, 209, 336], "color": [2, 3, 6, 12, 14, 32, 51, 61, 73, 155, 159, 182, 185, 192, 198, 199, 214, 219, 222, 224, 227, 230, 261, 266, 280, 321, 333, 334, 335, 336, 342], "below": [2, 4, 6, 9, 10, 12, 13, 94, 110, 160, 163, 174, 190, 191, 220, 235, 240, 252, 262, 269, 274, 275, 277, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 506, 508], "similar": [2, 5, 6, 15, 17, 48, 110, 147, 361, 362, 363, 365, 366, 367, 369, 371, 372, 386, 394, 395, 443, 444, 445, 446, 447, 448, 507], "assum": [2, 5, 9, 124, 187, 192, 223, 269, 280], "move": 2, "properli": [2, 6, 13, 14, 15, 262], "abl": [2, 10], "ball": 2, "player": 2, "particular": [2, 3, 5, 18, 78, 110], "differ": [2, 3, 4, 6, 11, 12, 14, 15, 18, 37, 38, 42, 50, 60, 65, 69, 74, 78, 85, 110, 124, 132, 157, 171, 190, 194, 220, 249, 269, 274, 283, 342, 343, 345, 363, 365, 366, 367, 369, 371, 394, 395, 508], "left": [2, 44, 50, 58, 61, 84, 134, 170, 173, 177, 184, 187, 197, 201, 205, 207, 208, 216, 217, 220, 221, 223, 225, 248, 251, 256, 264, 269, 297, 313], "up": [2, 4, 15, 60, 273, 446, 447, 448, 498, 508], "flow_img": 2, "have": [2, 4, 6, 9, 10, 11, 12, 13, 17, 27, 30, 35, 39, 44, 48, 49, 50, 51, 53, 54, 58, 60, 62, 63, 64, 84, 85, 86, 88, 94, 124, 132, 138, 139, 141, 145, 147, 150, 155, 156, 158, 159, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 342, 363, 365, 366, 367, 369, 371, 506, 508], "been": [2, 6, 11, 87, 124, 132, 147], "want": [2, 5, 9, 10, 12, 18, 96, 115, 148, 149, 151, 152, 186, 190, 220, 221, 267, 268, 270, 274, 275, 276, 329, 340, 341, 363, 366, 367, 508], "img1": [2, 30, 49, 50, 53, 63, 64, 84, 85, 86], "shown": [2, 174, 252], "A": [2, 3, 6, 9, 12, 13, 18, 25, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 100, 101, 115, 142, 143, 145, 153, 155, 186, 218, 230, 232, 239, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 278, 309, 333, 334, 338, 344, 353, 354, 355, 356, 365, 457, 460, 461, 501, 508], "fun": 2, "wai": [2, 4, 5, 6, 10, 11, 13, 15, 56, 96, 342], "appli": [2, 4, 12, 115, 118, 132, 138, 139, 141, 142, 143, 145, 153, 154, 159, 166, 171, 173, 174, 176, 183, 186, 190, 201, 202, 206, 220, 221, 234, 238, 243, 245, 249, 251, 252, 255, 263, 267, 268, 270, 274, 276, 342, 508], "entir": [2, 6, 154, 187, 223, 269, 331, 342], "snippet": [2, 3], "comment": 2, "out": [2, 5, 6, 9, 10, 11, 12, 13, 15, 18, 110, 139, 168, 197, 215, 246, 273, 341, 342, 508], "being": [2, 3, 15, 74, 134, 142, 143, 154, 172, 175, 178, 181, 182, 184, 185, 188, 189, 190, 220, 250, 253, 255, 257, 260, 261, 264, 266, 271, 272, 274], "render": [2, 77], "machin": [2, 340], "without": [2, 9, 10, 12, 69, 124, 139, 170, 177, 193, 216, 248, 256, 282, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 442, 473, 474, 475, 476, 477, 478, 479, 480, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "would": [2, 3, 6, 9, 11, 15, 18, 110, 239, 341, 508], "too": 2, "long": [2, 5, 9, 340], "write_jpeg": 2, "i": [2, 3, 4, 5, 6, 9, 10, 17, 75, 109, 124, 142, 143, 148, 149, 151, 152, 154, 159, 160, 168, 174, 177, 179, 186, 190, 192, 193, 198, 207, 215, 220, 234, 235, 239, 246, 252, 256, 258, 268, 274, 280, 282, 300, 331, 340, 342, 363], "img2": [2, 49, 50, 53, 63, 64, 84, 85, 86], "instead": [2, 3, 4, 14, 37, 75, 94, 122, 125, 126, 127, 128, 129, 130, 131, 158, 161, 163, 170, 173, 174, 177, 186, 187, 191, 208, 216, 225, 230, 231, 232, 235, 237, 240, 249, 251, 252, 255, 256, 263, 268, 269, 275, 277, 282, 283, 296, 322, 338, 352, 507, 508], "individu": [2, 5, 11, 12, 109, 115, 238, 245, 508], "output_fold": 2, "tmp": 2, "updat": [2, 5, 508], "folder": [2, 5, 37, 54, 60, 61, 88, 93], "choic": [2, 5], "predicted_flow_": 2, "jpg": [2, 4, 6, 13, 14, 15, 27, 44, 342], "onc": [2, 11, 12, 14, 18, 111, 275, 508], "save": [2, 4, 5, 25, 28, 29, 65, 66, 107, 108, 109, 333, 338, 342], "ffmpeg": [2, 96, 340], "e": [2, 3, 4, 5, 9, 10, 12, 13, 17, 25, 28, 29, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41, 42, 43, 45, 46, 47, 48, 51, 52, 55, 56, 57, 59, 61, 65, 66, 67, 68, 70, 71, 72, 73, 75, 76, 77, 79, 80, 81, 82, 83, 87, 89, 90, 91, 92, 93, 94, 96, 109, 142, 143, 148, 149, 151, 152, 154, 160, 162, 168, 173, 174, 179, 184, 186, 187, 190, 192, 201, 207, 215, 217, 220, 221, 223, 232, 235, 236, 239, 246, 248, 251, 252, 256, 258, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 275, 276, 278, 280, 326, 331, 333, 334, 335, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "g": [2, 3, 5, 9, 12, 13, 25, 28, 29, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41, 42, 43, 45, 46, 47, 48, 51, 52, 55, 56, 57, 59, 61, 65, 66, 67, 68, 70, 71, 72, 73, 75, 76, 77, 79, 80, 81, 82, 83, 87, 89, 90, 91, 92, 93, 94, 96, 162, 165, 170, 173, 177, 179, 180, 184, 186, 187, 190, 201, 216, 217, 220, 221, 222, 223, 226, 232, 236, 239, 248, 251, 256, 258, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 275, 276, 278, 326, 333, 334, 335, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "image2": [2, 44, 65], "framer": [2, 5], "30": [2, 14, 15, 20, 42, 121, 342, 357, 374, 391, 398, 406, 439], "d": [2, 5, 6, 111, 143, 167, 244, 363, 460], "loop": [2, 11, 96, 163, 191], "total": [2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16], "time": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 18, 96, 104, 197, 220, 340, 349, 363, 508], "script": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 160, 174, 235, 252, 507, 508], "minut": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15], "10": [2, 5, 6, 7, 9, 13, 14, 15, 20, 24, 25, 31, 36, 42, 43, 54, 60, 70, 81, 83, 88, 89, 96, 98, 121, 139, 160, 191, 235, 333, 334, 335, 342, 378, 387, 431, 507, 508], "096": [2, 7], "plot_optical_flow": [2, 7], "ipynb": [2, 3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15], "oper": [3, 6, 10, 13, 14, 15, 18, 96, 97, 110, 111, 116, 117, 121, 123, 124, 142, 143, 148, 149, 151, 152, 155, 168, 179, 207, 215, 230, 246, 258, 265, 273, 326, 327, 328, 330, 331, 340, 341, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "avail": [3, 5, 13, 14, 18, 69, 74, 96, 97, 98, 99, 103, 111, 115, 157, 325, 340, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "torchvis": [3, 4, 5, 6, 9, 10, 11, 12, 13, 14, 15, 17, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 310, 311, 312, 313, 314, 315, 316, 317, 319, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506, 507, 508, 510], "op": [3, 15, 18, 94, 110, 111, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 506], "modul": [3, 4, 13, 14, 17, 18, 110, 111, 112, 115, 116, 117, 121, 122, 123, 125, 126, 127, 128, 129, 130, 131, 153, 161, 174, 230, 231, 235, 243, 249, 252, 255, 263, 275, 283, 342, 345, 349, 350, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 432, 433, 434, 435, 436, 453, 457, 458, 459, 460, 461, 462, 477, 478, 479, 492, 496, 497, 501, 503, 508, 510], "local": 3, "annot": [3, 9, 12, 28, 33, 34, 42, 46, 47, 55, 73, 78, 88, 93], "instanc": [3, 4, 10, 12, 13, 14, 18, 32, 94, 96, 110, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 276, 283, 326, 327, 334, 340, 341, 363, 365, 366, 367, 369], "panopt": 3, "method": [3, 5, 13, 17, 18, 37, 56, 92, 110, 112, 142, 143, 155, 156, 171, 222, 230, 231, 235, 243, 249, 252, 276, 341, 342, 453, 508], "os": [3, 5], "assets_directori": 3, "asset": [3, 4, 5, 6, 12, 13, 14, 15, 342], "show": [3, 4, 6, 14, 342, 508], "fix": [3, 54, 60, 88, 122, 164, 241, 363, 366, 367, 371], "detach": [3, 6, 15, 342], "like": [3, 6, 9, 10, 11, 12, 13, 14, 15, 18, 96, 111, 158, 162, 163, 173, 177, 186, 187, 190, 191, 208, 209, 220, 225, 232, 239, 240, 251, 256, 268, 269, 274, 277, 327, 328, 330, 332, 340, 341, 508], "commonli": [3, 48, 506], "defin": [3, 4, 5, 6, 10, 11, 12, 71, 81, 96, 122, 125, 126, 127, 128, 129, 130, 131, 148, 149, 151, 152, 155, 156, 161, 162, 166, 171, 173, 174, 184, 186, 187, 190, 194, 201, 209, 217, 220, 221, 223, 230, 231, 235, 236, 239, 243, 249, 251, 252, 255, 263, 264, 267, 268, 269, 270, 273, 274, 275, 276, 283, 329, 394, 395, 508], "packag": [3, 18, 67, 73, 341, 507, 508], "multi": [3, 11, 12, 74, 123, 124, 235, 243, 252, 342, 386, 480], "arrai": [3, 27, 30, 39, 44, 49, 50, 53, 58, 62, 63, 64, 69, 78, 84, 85, 86, 218, 309], "num_object": 3, "height": [3, 31, 124, 134, 136, 139, 148, 149, 151, 152, 184, 190, 203, 205, 207, 210, 212, 220, 221, 264, 274, 276, 297, 313, 326, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "width": [3, 6, 31, 124, 134, 136, 139, 148, 149, 151, 152, 184, 190, 203, 205, 207, 210, 212, 220, 221, 264, 274, 276, 297, 313, 326, 333, 334, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "exactli": [3, 11, 13, 54, 60, 88, 141, 148, 151, 161, 204, 237], "ha": [3, 5, 6, 11, 18, 33, 37, 38, 87, 159, 161, 186, 192, 193, 198, 204, 234, 237, 259, 265, 268, 269, 280, 282, 331, 442, 453, 471, 472], "224": [3, 4, 9, 11, 13, 18, 43, 110, 124, 160, 179, 235, 240, 258, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 458, 459, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "x": [3, 4, 5, 6, 18, 31, 73, 75, 96, 110, 121, 122, 124, 127, 148, 149, 151, 152, 167, 169, 173, 187, 192, 193, 201, 209, 244, 247, 251, 269, 280, 282, 333, 334, 337, 361, 362, 363, 365, 366, 367, 369, 371, 372, 385, 397], "four": [3, 14, 15, 116, 117, 163, 191, 208, 217, 225, 240, 277], "4": [3, 6, 11, 12, 14, 15, 17, 18, 19, 20, 22, 23, 24, 27, 30, 31, 39, 44, 49, 50, 53, 58, 61, 62, 63, 64, 69, 84, 85, 86, 110, 124, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 144, 145, 146, 147, 148, 149, 150, 151, 152, 170, 173, 177, 192, 216, 232, 239, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 280, 329, 333, 342, 356, 358, 360, 362, 363, 365, 366, 367, 368, 369, 371, 372, 377, 379, 394, 398, 401, 407, 424, 427, 430, 432, 433, 435, 440, 441, 442, 443, 445, 446, 461, 466, 467, 470, 508], "nice": 3, "properti": [3, 14, 329], "easili": [3, 4, 5, 6, 111], "solv": 3, "varieti": [3, 18], "masks_to_box": 3, "fasterrcnn": [3, 349, 361, 362, 363, 364], "retinanet": [3, 153, 342, 369, 370], "penfudan": 3, "read_imag": [3, 4, 6, 13, 342], "img_path": 3, "join": [3, 6], "fudanped00054": 3, "png": [3, 30, 39, 44, 56, 58, 62, 63, 69, 84, 86, 97, 99, 101, 103, 108, 333], "mask_path": 3, "fudanped00054_mask": 3, "repres": [3, 6, 18, 28, 31, 72, 111, 134, 333, 334, 335, 508], "float": [3, 5, 6, 60, 61, 96, 104, 105, 109, 119, 120, 122, 123, 125, 126, 128, 129, 131, 132, 137, 138, 140, 141, 142, 143, 145, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 159, 160, 161, 162, 164, 171, 172, 173, 174, 175, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 192, 194, 195, 196, 197, 198, 199, 200, 201, 204, 209, 215, 216, 217, 223, 224, 230, 231, 234, 235, 237, 238, 239, 241, 245, 246, 248, 249, 250, 251, 252, 253, 255, 256, 257, 258, 259, 260, 261, 262, 264, 265, 266, 268, 269, 271, 272, 273, 275, 276, 280, 283, 285, 286, 287, 288, 289, 290, 291, 298, 302, 305, 306, 308, 315, 316, 335, 336, 337, 508], "point": [3, 6, 13, 18, 28, 31, 54, 60, 74, 88, 96, 104, 138, 148, 151, 161, 204, 237, 333, 394, 395, 508], "encod": [3, 5, 96, 238, 245, 340], "background": [3, 6, 33, 71, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "notic": [3, 13, 340], "spatial": [3, 139, 146, 267, 270, 273], "match": [3, 74, 76, 115, 190, 220, 236, 238, 245, 274, 307, 508], "498": [3, 20, 342, 418], "533": 3, "uint8": [3, 6, 13, 15, 60, 96, 97, 98, 99, 102, 103, 106, 109, 155, 156, 171, 185, 193, 194, 206, 219, 230, 231, 249, 266, 282, 283, 333, 334, 335, 336, 508], "uniqu": [3, 5, 96], "id": [3, 5, 6, 12, 55, 96], "obj_id": 3, "remov": [3, 5, 6, 10, 12, 18, 110, 115, 147, 150, 262, 275, 281, 282, 342], "split": [3, 14, 31, 32, 35, 36, 38, 39, 41, 42, 48, 49, 50, 51, 52, 53, 54, 57, 58, 60, 61, 62, 63, 64, 65, 66, 69, 70, 72, 73, 75, 77, 78, 81, 83, 85, 87, 88, 93, 139], "boolean": [3, 6, 53, 64, 69, 76, 177, 179, 256, 258], "work": [3, 5, 9, 11, 13, 15, 18, 159, 160, 174, 198, 230, 231, 234, 235, 249, 252, 265, 275, 283, 361, 362, 364], "well": [3, 6, 12, 15, 17, 51, 161, 162, 173, 184, 186, 187, 190, 201, 204, 217, 220, 221, 223, 237, 239, 251, 264, 267, 268, 269, 270, 274, 275, 276, 327, 328, 506], "int": [3, 5, 13, 18, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 94, 96, 100, 101, 104, 105, 107, 108, 110, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 130, 136, 139, 142, 143, 148, 149, 151, 152, 155, 156, 158, 163, 164, 165, 170, 171, 173, 177, 179, 184, 185, 186, 190, 191, 192, 194, 201, 203, 205, 207, 208, 209, 210, 211, 212, 216, 217, 219, 220, 221, 222, 223, 225, 226, 230, 231, 232, 238, 239, 240, 241, 242, 245, 248, 249, 251, 256, 258, 262, 264, 266, 267, 268, 269, 270, 273, 274, 276, 277, 280, 283, 284, 291, 293, 294, 297, 298, 300, 301, 302, 306, 307, 308, 310, 312, 313, 314, 315, 317, 319, 323, 326, 327, 328, 329, 330, 333, 334, 335, 337, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "none": [3, 5, 6, 12, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 96, 104, 106, 109, 110, 111, 115, 116, 117, 118, 121, 123, 128, 129, 138, 139, 141, 145, 151, 152, 153, 155, 156, 159, 171, 173, 176, 177, 179, 186, 187, 190, 192, 194, 201, 209, 217, 220, 221, 223, 227, 230, 231, 234, 239, 248, 249, 251, 255, 256, 258, 262, 264, 267, 268, 269, 270, 273, 274, 275, 276, 278, 280, 283, 291, 294, 295, 298, 302, 306, 308, 312, 315, 321, 326, 327, 328, 330, 333, 334, 335, 337, 338, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "now": [3, 4, 5, 6, 9, 10, 11, 12, 13, 18, 94, 342], "denot": [3, 155, 205, 221, 230, 284, 334, 335], "peopl": [3, 6], "other": [3, 6, 12, 13, 14, 15, 49, 50, 60, 85, 94, 186, 190, 193, 220, 221, 222, 267, 268, 270, 274, 276, 278, 282, 333, 338, 341, 385, 397], "equal": [3, 14, 15, 69, 178, 188, 190, 220, 224, 257, 271, 274], "belong": [3, 6, 36, 193, 282], "let": [3, 4, 5, 6, 9, 11, 12, 13, 18], "draw_segmentation_mask": [3, 6, 340], "draw": [3, 6, 333, 334, 335], "drawn_mask": 3, "append": [3, 5, 60, 96, 110, 111, 363], "alpha": [3, 6, 14, 61, 153, 155, 159, 162, 197, 198, 230, 234, 238, 239, 245, 335], "blue": [3, 6, 33], "To": [3, 5, 6, 10, 12, 18, 54, 60, 88, 159, 234, 235, 243, 252, 265, 342], "xmin": [3, 6, 333], "ymin": [3, 6, 333], "xmax": [3, 6, 333], "ymax": [3, 6, 333], "format": [3, 6, 9, 13, 15, 28, 48, 54, 60, 69, 78, 83, 87, 88, 89, 94, 96, 97, 98, 99, 103, 104, 109, 124, 132, 133, 134, 135, 136, 137, 138, 140, 141, 144, 145, 146, 147, 148, 149, 150, 151, 152, 182, 188, 195, 196, 197, 198, 199, 200, 202, 206, 213, 214, 219, 224, 229, 236, 261, 271, 294, 325, 326, 332, 333, 334, 338, 342, 363, 365, 366, 367, 369, 371, 508], "96": [3, 5, 19, 20, 339, 342, 353, 354, 355, 356, 376, 377, 378, 379, 380, 383, 386, 402, 403, 408, 410, 416, 418, 422, 424, 428, 429, 443, 444, 446, 447, 448, 457, 466, 471], "134": [3, 22, 23, 342, 363, 367], "181": 3, "417": 3, "286": [3, 20, 342, 391, 398], "113": 3, "357": [3, 6], "331": [3, 466], "363": 3, "120": [3, 460], "436": 3, "328": [3, 6, 20, 342, 416], "These": [3, 6, 11, 13, 14, 15, 161, 204, 237, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472, 506, 507, 508], "veri": [3, 10, 13, 48, 342, 449, 450, 451, 452, 453, 454, 455, 456, 499, 508], "draw_bounding_box": [3, 6, 340, 342], "drawn_box": 3, "red": [3, 333, 334, 335, 342], "directli": [3, 5, 6, 15, 110, 111, 342, 508, 509], "demo": [3, 6], "cnn": [3, 6, 124, 148, 151, 152, 342, 361, 362, 363, 364, 366, 367, 368, 404, 405, 406, 407, 437, 438, 439, 440, 462, 493, 494], "load": [3, 4, 6, 17, 28, 33, 37, 48, 56, 57, 67, 74, 75, 76, 78, 81, 83, 87, 95, 340, 342], "fasterrcnn_resnet50_fpn": [3, 6, 361, 362, 364], "fasterrcnn_resnet50_fpn_weight": [3, 6, 22, 342, 363], "tranform": [3, 13, 508], "target": [3, 5, 11, 12, 13, 17, 25, 26, 28, 29, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41, 42, 43, 45, 46, 47, 48, 51, 52, 55, 56, 57, 59, 61, 65, 66, 67, 68, 71, 72, 73, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 89, 90, 91, 92, 93, 94, 124, 153, 163, 191, 193, 208, 225, 240, 276, 277, 282, 363, 365, 366, 367, 369, 371], "label": [3, 4, 6, 9, 11, 12, 13, 31, 51, 54, 60, 72, 73, 83, 88, 89, 92, 94, 153, 238, 240, 245, 275, 333, 342, 363, 365, 366, 367, 369, 371, 431, 432, 433, 434, 435, 436], "int64": [3, 11, 132, 147, 161, 204, 237, 278], "detection_output": 3, "unsqueez": [3, 342], "fasterrcnn_resnet50_fpn_coco": 3, "258fb6c6": 3, "With": [3, 96], "becom": [3, 202, 269], "simpl": [3, 6, 10, 11, 13, 155, 230, 276, 365, 391, 392, 393, 398, 399, 400, 401, 402, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 430, 441, 442, 471, 472], "One": [3, 11, 18, 31, 55, 67, 81, 83, 93, 343, 351, 352, 365], "similarli": [3, 6, 12, 96, 361, 362, 364], "re": [3, 4, 5, 6, 9, 10, 12, 13, 14, 15, 39, 110, 235, 243, 252, 329, 508], "purpos": [3, 5, 12, 18, 51], "segmentationtodetectiondataset": 3, "data": [3, 10, 11, 14, 15, 17, 30, 31, 32, 33, 34, 35, 36, 37, 41, 42, 48, 49, 51, 52, 54, 56, 60, 69, 70, 72, 73, 75, 77, 78, 82, 83, 87, 88, 89, 94, 96, 102, 106, 121, 155, 156, 161, 167, 171, 179, 192, 194, 204, 227, 230, 231, 233, 237, 239, 244, 248, 249, 251, 256, 258, 262, 264, 267, 269, 273, 276, 280, 283, 321, 326, 327, 328, 330, 415, 416, 418, 466, 468, 469, 508], "__init__": [3, 4, 5, 18, 110], "self": [3, 4, 5, 9, 18, 37, 110, 121, 240, 329], "file": [3, 5, 7, 16, 28, 33, 34, 37, 46, 47, 48, 49, 54, 55, 56, 57, 67, 71, 75, 76, 78, 87, 88, 89, 96, 100, 101, 102, 104, 105, 106, 107, 108, 109, 333, 338, 341, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "sort": [3, 5, 132, 147], "align": [3, 128, 148, 151], "listdir": 3, "pngimag": 3, "pedmask": 3, "__getitem__": [3, 13, 17, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93], "idx": [3, 6, 36, 41, 42, 48, 51, 54, 60, 70, 72, 73, 77, 82, 87, 88, 121, 132, 342], "convert_image_dtyp": 3, "507": [3, 7, 451, 452], "plot_repurposing_annot": [3, 7], "nn": [4, 6, 11, 13, 18, 110, 111, 112, 116, 117, 123, 130, 160, 174, 235, 252, 255, 263, 508], "v1": [4, 9, 12, 13, 340, 384, 396, 423, 424, 427, 488, 500], "manual_se": [4, 12, 13, 14], "helper": [4, 12, 13, 14], "github": [4, 5, 9, 12, 13, 14, 15, 30, 94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 507], "tree": [4, 6, 12, 13, 14, 27, 90, 507], "main": [4, 5, 12, 13, 14, 18, 340], "sy": 4, "assets_path": 4, "compos": [4, 5, 9, 11, 12, 13, 163, 179, 191, 240, 258, 282, 415, 416, 418, 466, 468, 469, 508], "sequenti": [4, 160, 171, 235, 249, 508], "dog1": [4, 6, 342], "dog2": [4, 6], "randomcrop": [4, 13, 25, 28, 29, 32, 35, 36, 37, 38, 40, 41, 42, 43, 45, 48, 51, 52, 55, 56, 57, 59, 67, 68, 70, 71, 72, 73, 75, 76, 77, 79, 80, 81, 82, 83, 87, 89, 90, 91, 92, 93, 267, 293, 297, 508], "randomhorizontalflip": [4, 9, 10, 11, 12, 13, 179, 258, 303, 508], "p": [4, 9, 10, 11, 12, 13, 14, 119, 120, 131, 142, 143, 154, 172, 174, 175, 176, 178, 179, 180, 181, 182, 184, 185, 188, 189, 193, 206, 250, 252, 253, 255, 257, 258, 259, 260, 261, 264, 265, 266, 271, 272, 273, 282, 508], "scripted_transform": [4, 160, 174, 235, 252, 508], "jit": [4, 160, 174, 235, 252, 508], "namespac": [4, 12, 13, 508], "recommend": [4, 5, 10, 13, 275, 278, 331, 340], "call": [4, 5, 10, 11, 12, 14, 15, 96, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 238, 245, 249, 252, 255, 263, 275, 283, 338, 342, 508], "ll": [4, 6, 11, 12, 13, 18, 148, 149, 151, 152, 508], "actual": [4, 5, 12, 15, 60, 508], "equival": [4, 121, 186, 190, 220, 221, 267, 268, 270, 274, 276, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "mai": [4, 6, 10, 12, 13, 15, 18, 94, 110, 163, 190, 191, 208, 220, 225, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 333, 340, 342, 363, 365, 366, 367, 369, 508], "lead": [4, 155, 156, 158, 159, 161, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 234, 237, 239, 240, 241, 242, 248, 249, 250, 251, 253, 256, 257, 259, 260, 261, 262, 264, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 277, 283, 342, 508], "slightli": [4, 11, 13, 18, 94, 190, 220, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 343, 442, 508], "eager": [4, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 508], "execut": [4, 7, 16, 18, 111, 508], "due": [4, 269, 508], "realli": [4, 13, 186, 190, 220, 221, 267, 268, 270, 274, 276, 508], "avoid": [4, 15, 177, 256, 275, 331, 508], "surpris": [4, 15, 508], "combin": [4, 5, 14, 36, 92, 394, 508], "forward": [4, 15, 18, 110, 118, 119, 120, 121, 122, 124, 125, 126, 127, 128, 129, 130, 131, 155, 156, 158, 159, 161, 162, 163, 164, 165, 167, 168, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 230, 231, 235, 240, 249, 252, 255, 263, 275, 283, 331], "pass": [4, 5, 6, 9, 10, 11, 12, 15, 17, 18, 37, 50, 81, 84, 85, 86, 96, 109, 110, 111, 112, 115, 121, 122, 125, 126, 127, 128, 129, 130, 131, 148, 149, 151, 152, 159, 161, 164, 167, 173, 174, 177, 179, 186, 187, 230, 231, 234, 235, 241, 244, 249, 251, 252, 255, 256, 258, 263, 268, 269, 275, 278, 283, 331, 332, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "while": [4, 6, 9, 13, 14, 94, 95, 122, 125, 126, 127, 128, 129, 130, 131, 161, 172, 174, 192, 195, 196, 197, 198, 199, 200, 201, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 423, 424, 427, 488, 507, 508], "obtain": [4, 124, 367], "predictor": 4, "imagenet": [4, 11, 13, 14, 17, 156, 157, 231, 342, 343, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472, 508], "resnet18": [4, 110, 111], "resnet18_weight": [4, 20, 342, 400, 425], "super": [4, 5, 18, 55, 110], "true": [4, 5, 9, 11, 12, 13, 15, 17, 25, 26, 28, 29, 31, 35, 36, 38, 40, 42, 45, 48, 51, 52, 54, 55, 59, 60, 61, 65, 66, 68, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 88, 89, 90, 91, 93, 96, 116, 117, 118, 123, 142, 143, 151, 154, 155, 186, 187, 190, 220, 221, 223, 230, 237, 267, 268, 269, 270, 274, 276, 278, 282, 333, 337, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "no_grad": [4, 18], "y_pred": 4, "argmax": [4, 6, 342], "dim": [4, 6, 127, 329, 342], "non": [4, 6, 132, 138, 141, 145, 147, 159, 172, 177, 195, 196, 197, 198, 200, 206, 234, 250, 265], "multipl": [4, 5, 6, 14, 17, 18, 32, 60, 96, 147, 326], "scripted_predictor": 4, "res_script": 4, "f37072fd": 4, "verifi": 4, "json": [4, 12, 32, 33, 34, 44], "open": [4, 9, 14, 15, 340], "imagenet_class_index": 4, "labels_fil": 4, "pred": 4, "pred_script": 4, "assert": [4, 9, 15, 342], "dog": [4, 6, 56, 342], "n02113023": 4, "pembrok": 4, "n02106662": 4, "german_shepherd": 4, "sinc": [4, 5, 6, 10, 12, 74, 94, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 177, 204, 230, 231, 235, 237, 249, 252, 255, 256, 263, 269, 275, 283], "dump": 4, "disk": [4, 37], "namedtemporaryfil": 4, "dumped_scripted_predictor": 4, "res_scripted_dump": 4, "683": [4, 7], "plot_scripted_tensor_transform": [4, 7], "some": [5, 6, 9, 13, 14, 15, 18, 51, 78, 94, 110, 119, 120, 121, 124, 192, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 342, 508], "offer": [5, 6, 342, 507], "togeth": [5, 15, 18, 110, 160, 235, 267, 508], "test": [5, 25, 31, 32, 35, 36, 38, 39, 41, 42, 48, 50, 51, 52, 53, 54, 58, 60, 61, 62, 63, 64, 65, 66, 67, 69, 70, 72, 73, 76, 77, 78, 81, 83, 85, 87, 88, 90, 91, 93, 163, 191, 340, 341, 342, 394], "sake": 5, "argument": [5, 13, 17, 38, 60, 76, 110, 111, 138, 141, 145, 184, 264, 338, 341, 508], "creat": [5, 6, 10, 12, 13, 17, 18, 25, 38, 45, 54, 59, 68, 71, 88, 89, 96, 110, 124, 164, 206, 241, 273, 342], "stream": [5, 33, 96, 109, 341], "chosen": [5, 48, 159, 164, 234, 241, 265], "statist": [5, 122], "wuzgd7c1pwa": 5, "kinet": [5, 94, 342, 457, 458, 459, 460, 461, 462, 463, 464, 465], "400": [5, 6, 60, 342, 361, 362, 363, 365, 366, 367, 369, 371, 372, 457, 458, 459, 460, 461, 462, 463, 464, 465], "264": 5, "mpeg": 5, "avc": 5, "part": [5, 78, 96, 340, 507, 508], "avc1": 5, "fp": [5, 109, 338, 341], "29": [5, 20, 342, 357, 358, 359, 360, 468], "97": [5, 20, 342, 381, 382, 401, 415, 416, 418, 427, 466, 468, 469], "aac": [5, 109], "mp4a": 5, "rate": [5, 60, 96, 105, 109, 341], "48k": 5, "hz": 5, "download_url": 5, "set_video_backend": [5, 96, 340], "video_read": [5, 96, 340], "blob": 5, "raw": [5, 38, 45, 59, 61, 68, 76, 97, 98, 99, 100, 101], "githubusercont": 5, "7": [5, 6, 19, 20, 22, 24, 137, 138, 140, 141, 145, 262, 339, 342, 355, 358, 359, 363, 364, 380, 383, 384, 386, 395, 396, 400, 402, 406, 407, 422, 423, 425, 428, 434, 442, 447, 449, 450, 457, 458, 459, 471], "11": [5, 6, 19, 20, 24, 98, 339, 342, 363, 366, 367, 400, 424, 447, 449, 450, 451, 452, 472], "14": [5, 6, 7, 16, 19, 20, 110, 342, 366, 405, 438, 462], "18": [5, 6, 20, 342, 388, 394, 400, 411, 419, 425, 431, 432, 433, 434, 435, 436, 457, 460, 461, 466], "22": [5, 19, 20, 22, 339, 342, 361, 392, 399, 459, 471], "25": [5, 19, 20, 22, 153, 342, 371, 395, 401], "5": [5, 6, 11, 13, 14, 18, 19, 20, 22, 24, 54, 60, 88, 96, 121, 122, 124, 139, 143, 148, 149, 151, 152, 159, 162, 163, 172, 174, 175, 178, 179, 181, 182, 184, 185, 188, 189, 198, 209, 234, 239, 240, 250, 252, 253, 257, 258, 260, 261, 262, 264, 265, 266, 271, 272, 273, 329, 331, 339, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 363, 364, 365, 366, 367, 368, 369, 370, 371, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 433, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 488, 508], "33": [5, 20, 179, 258, 339, 342, 381, 382, 383, 422], "36": [5, 19, 20, 22, 339, 342, 353, 354, 359, 369, 405, 444, 454], "40": [5, 15, 31, 48, 262, 339, 342, 420, 460, 461, 472], "44": [5, 6, 20, 23, 342, 400, 409, 425, 457], "47": [5, 20, 23, 342, 368, 376, 395, 422, 453, 454], "9": [5, 6, 14, 20, 23, 24, 81, 89, 101, 108, 171, 249, 262, 342, 367, 378, 385, 389, 393, 397, 413, 434, 436, 450, 454, 459, 471], "51": [5, 20, 342, 389, 424, 471], "55": [5, 20, 342, 447, 466, 469], "58": [5, 19, 20, 22, 23, 342, 368, 372, 382, 407, 411, 440, 441, 442], "62": 5, "66": [5, 20, 24, 342, 395, 426, 433], "70": [5, 9, 14, 20, 42, 339, 342, 450, 458, 461], "73": [5, 19, 20, 342, 358, 389, 399, 416, 426, 454, 468], "77": [5, 19, 20, 339, 342, 358, 360, 373, 385, 397, 409, 413, 417, 423, 430, 465], "81": [5, 19, 20, 339, 342, 400, 403, 404, 411, 414, 419, 423, 425, 430, 437, 445, 450, 463, 466, 472], "84": [5, 20, 339, 342, 353, 354, 379, 380, 383, 418, 446, 457, 464], "88": [5, 19, 20, 339, 342, 361, 362, 363, 364, 365, 367, 368, 369, 370, 371, 372, 400, 405, 415, 438, 449, 462, 465, 468, 469], "92": [5, 19, 20, 21, 24, 342, 359, 366, 392, 399, 401, 407, 412, 413, 420, 427, 432, 433, 467], "95": [5, 20, 339, 342, 375, 401, 410, 411, 412, 414, 416, 417, 418, 419, 422, 423, 427, 430, 445, 463, 466, 472], "99": [5, 20, 342, 469], "fashion": [5, 45, 132], "string": [5, 10, 13, 25, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 96, 110, 138, 141, 145, 153, 235, 236, 243, 252, 333, 334, 335, 338, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "form": [5, 37, 51], "stream_typ": [5, 96, 341], "stream_id": [5, 96], "constructor": [5, 341, 342], "auto": [5, 14, 96], "discov": 5, "firstli": 5, "get_metadata": [5, 96, 341], "durat": [5, 96, 341, 342], "9109": 5, "97002997002997": 5, "48000": [5, 109], "subtitl": 5, "cc": 5, "current": [5, 6, 12, 78, 94, 96, 121, 186, 190, 220, 221, 267, 268, 270, 274, 276, 340, 341, 342, 507], "includ": [5, 78, 96, 115, 116, 117, 155, 230, 275, 340, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "descriptor": [5, 18, 74, 96], "consist": [5, 48, 49, 50, 62, 82, 85, 96, 186, 190, 220, 221, 267, 268, 270, 274, 276, 340], "determin": [5, 13, 96, 192, 280, 307, 338, 342], "contain": [5, 6, 18, 42, 51, 54, 69, 76, 77, 78, 87, 88, 94, 96, 97, 98, 99, 100, 101, 104, 109, 110, 121, 135, 137, 140, 144, 146, 148, 149, 150, 151, 152, 184, 217, 264, 326, 332, 333, 334, 335, 337, 342, 363, 365, 366, 367, 369, 371, 508, 510], "user": [5, 10, 13, 15, 18, 84, 94, 96, 110, 111, 166, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 509], "access": [5, 18, 84, 94, 96, 342], "decod": [5, 96, 97, 98, 99, 105, 340], "read": [5, 6, 10, 13, 15, 37, 95, 96, 97, 98, 99, 102, 103, 104, 111, 341], "By": [5, 13, 15, 110, 111, 116, 117, 207, 238, 245, 275, 333, 335, 337, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "next": [5, 13, 54, 60, 88, 96], "dict": [5, 6, 9, 11, 12, 13, 18, 37, 54, 60, 88, 93, 94, 96, 104, 109, 110, 111, 121, 124, 230, 231, 239, 248, 249, 251, 256, 264, 269, 273, 275, 278, 283, 341, 363, 365, 366, 367, 369, 371, 508], "field": [5, 104, 363, 365, 366, 367, 369, 371, 394, 395, 486], "timestamp": [5, 96, 105], "set_current_stream": [5, 96, 341], "ptss": 5, "present": [5, 12, 54, 60, 75, 88, 104, 105, 147, 341, 342, 431, 432, 433, 434, 435, 436], "five": 5, "approx_nf": 5, "approx": 5, "datapoint": 5, "expect": [5, 6, 9, 11, 12, 13, 27, 30, 39, 44, 49, 50, 53, 58, 61, 62, 63, 64, 83, 84, 85, 86, 93, 94, 121, 122, 132, 133, 134, 135, 137, 138, 140, 141, 144, 145, 147, 151, 155, 156, 158, 159, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 342, 363, 365, 366, 367, 369, 371, 385, 397], "021332999999999998": 5, "042667": 5, "064": [5, 20, 342, 469], "08533299999999999": 5, "511": 5, "523200": 5, "523264": 5, "But": [5, 13, 18], "what": [5, 6, 10, 11, 12, 76, 186, 190, 220, 221, 267, 268, 270, 274, 276, 508], "certain": [5, 18], "That": [5, 6, 13, 14], "done": [5, 10, 12, 177, 233, 238, 245, 256, 508], "seek": [5, 15, 96], "fact": [5, 13], "reli": [5, 10, 12, 13, 15, 98, 265, 343, 344, 345, 346, 347, 348, 349, 350, 351, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 507, 508], "leverag": [5, 12], "itertool": [5, 96], "simplifi": [5, 12, 342, 352, 449, 450, 451, 452, 453, 454, 455, 456], "process": [5, 10, 12, 14, 155, 230, 363, 365, 366, 367, 369, 371, 506], "ten": [5, 225], "islic": [5, 96], "Or": [5, 508], "2nd": [5, 18], "5th": 5, "takewhil": [5, 96], "correct": [5, 197, 342], "lambda": [5, 12, 96, 160, 163, 174, 191, 235, 252], "90": [5, 19, 20, 342, 388, 391, 398, 399, 406, 412, 452, 453, 455], "89": [5, 19, 20, 342, 384, 396, 400, 425, 450, 451], "91008991008991": 5, "256": [5, 9, 10, 13, 15, 18, 29, 75, 89, 110, 124, 240, 307, 352, 357, 358, 359, 360, 373, 374, 384, 387, 389, 391, 392, 393, 396, 398, 399, 400, 401, 402, 404, 405, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 430, 437, 438, 441, 442, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 458, 459, 462, 463, 464, 465, 466, 467, 470, 471, 472, 508], "340": [5, 6, 13], "exist": [5, 13, 25, 28, 29, 38, 40, 41, 45, 59, 65, 66, 68, 70, 71, 79, 80, 81, 96, 186, 190, 220, 221, 267, 268, 270, 274, 276, 342, 508], "example_read_video": 5, "video_object": 5, "read_audio": 5, "inf": 5, "rais": [5, 37, 60, 75, 94, 110, 161, 177, 204, 237, 256, 307], "valueerror": [5, 37, 307], "should": [5, 10, 12, 13, 18, 36, 37, 54, 60, 88, 96, 110, 122, 125, 126, 127, 128, 129, 130, 131, 138, 139, 141, 145, 148, 149, 151, 152, 155, 156, 159, 161, 170, 171, 174, 177, 180, 185, 193, 194, 198, 216, 219, 230, 231, 234, 235, 243, 248, 249, 252, 255, 256, 259, 263, 265, 266, 275, 282, 283, 326, 327, 328, 330, 333, 334, 335, 338, 340, 363, 365, 366, 367, 369, 371], "larger": [5, 96, 150, 161, 197, 204, 237, 284, 471, 472], "than": [5, 6, 10, 12, 18, 32, 74, 96, 110, 147, 150, 158, 177, 190, 197, 203, 205, 220, 232, 256, 274, 284, 337, 340, 442, 508], "got": [5, 13, 15], "video_fram": 5, "empti": [5, 31, 37, 72, 329], "video_pt": 5, "audio_fram": 5, "audio_pt": 5, "cat": [5, 56], "327": [5, 6, 73], "vf": 5, "af": 5, "info": [5, 6, 104, 509], "meta": [5, 6, 10, 31, 33, 34, 36, 41, 42, 48, 51, 52, 54, 60, 72, 73, 75, 77, 78, 82, 88, 233, 342], "cool": 5, "principl": 5, "suggest": [5, 15], "makedir": 5, "exist_ok": 5, "ratrace_wave_f_nm_np1_fr_goo_37": 5, "avi": [5, 60, 342], "sox5ya1l24a": 5, "v_soccerjuggling_g23_c01": [5, 342], "v_soccerjuggling_g24_c01": 5, "24": [5, 20, 22, 342, 370, 382, 390, 401, 416, 421], "37": [5, 20, 22, 23, 72, 342, 360, 363, 364, 367, 380, 383, 450], "49": [5, 20, 22, 339, 342, 362, 384, 387, 396, 445], "74": [5, 20, 24, 342, 357, 361, 362, 377, 392, 399, 410, 412, 419, 420, 432, 434, 444, 456], "87": [5, 19, 20, 342, 353, 387, 393, 405], "17": [5, 6, 15, 20, 186, 190, 220, 221, 267, 268, 270, 274, 276, 339, 342, 394, 442, 462, 466, 508], "23": [5, 6, 20, 342, 397, 407, 430, 440], "35": [5, 20, 22, 24, 342, 375, 409, 442], "46": [5, 19, 20, 22, 23, 342, 356, 364, 403, 416, 429], "52": [5, 20, 339, 342, 356, 415, 460], "64": [5, 14, 20, 110, 121, 124, 339, 342, 443, 459], "76": [5, 20, 342, 360, 390, 401, 407, 421, 427, 440, 470], "93": [5, 19, 20, 339, 342, 358, 360, 373, 385, 390, 397, 407, 409, 411, 413, 417, 421, 423, 430, 440, 458, 465, 470], "19": [5, 20, 22, 342, 379, 394, 455, 456], "32": [5, 14, 20, 22, 124, 342, 353, 354, 355, 356, 362, 418, 446, 449, 450, 451, 452, 453, 454, 455, 456, 462, 463, 464, 465, 508], "38": [5, 20, 22, 23, 342, 363, 367, 470], "45": [5, 20, 24, 342, 385, 397, 431, 458, 459], "83": [5, 19, 20, 339, 342, 355, 357, 376, 377, 378, 386, 403, 410, 416, 418, 421, 426, 429, 443, 444, 447, 461], "15": [5, 6, 13, 19, 20, 54, 60, 88, 342, 353, 384, 396, 398, 403, 408, 416, 429, 443, 453, 454, 457, 460, 461, 462, 463, 464, 465, 470, 508], "21": [5, 6, 20, 22, 342, 365, 372, 388, 392, 399, 412, 441], "26": 5, "31": [5, 20, 171, 194, 249, 283, 339, 342, 375, 389, 410, 451, 452], "42": [5, 20, 21, 24, 342, 366, 421, 426, 431, 441], "57": [5, 19, 20, 24, 342, 404, 415, 436], "63": [5, 20, 24, 339, 342, 352, 434, 455, 456, 457, 461], "68": [5, 19, 20, 339, 342, 355, 405, 462], "78": [5, 19, 20, 339, 342, 374, 402, 404, 411, 419, 421, 423, 424, 458, 471, 472], "94": [5, 19, 20, 339, 342, 374, 394, 401, 402, 408, 409, 414, 418, 419, 421, 424, 428, 448, 459, 463, 464, 469, 471, 472], "housekeep": 5, "random": [5, 14, 43, 124, 159, 162, 164, 173, 177, 179, 183, 184, 186, 187, 234, 239, 241, 251, 256, 258, 262, 263, 264, 267, 268, 269, 333, 335, 342, 508], "make_dataset": [5, 37], "_find_class": 5, "dir": [5, 33, 37], "scandir": 5, "is_dir": 5, "class_to_idx": [5, 37, 342], "cls_name": 5, "get_sampl": 5, "extens": [5, 37, 60, 338], "basic": 5, "structur": [5, 11, 12, 13, 14, 27, 30, 37, 39, 44, 49, 50, 53, 58, 60, 61, 62, 63, 64, 84, 85, 86, 93, 341, 508], "folderdataset": 5, "add": [5, 10, 11, 18, 121], "paramet": [5, 6, 11, 12, 15, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 127, 130, 132, 133, 134, 135, 136, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 248, 249, 250, 251, 252, 253, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 280, 283, 284, 307, 309, 321, 326, 327, 328, 330, 331, 332, 333, 334, 335, 336, 337, 338, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 495, 508], "clip_len": [5, 457, 458, 459, 460, 461, 462, 463, 464, 465], "clip": [5, 54, 60, 88, 136, 342, 462], "frame_transform": 5, "video_transform": 5, "sequenc": [5, 12, 72, 155, 156, 158, 162, 163, 164, 168, 170, 171, 173, 174, 177, 179, 184, 186, 187, 190, 191, 194, 201, 203, 208, 209, 215, 216, 217, 220, 221, 223, 225, 230, 231, 232, 234, 235, 239, 240, 241, 246, 248, 249, 251, 252, 255, 256, 258, 263, 264, 268, 269, 270, 273, 274, 277, 283], "epoch": [5, 366], "iterabledataset": 5, "allow": [5, 37, 96, 124, 190, 220, 274, 342], "oversampl": 5, "randomdataset": 5, "epoch_s": 5, "16": [5, 6, 15, 19, 20, 87, 121, 124, 342, 389, 394, 402, 420, 428, 453, 454, 457, 458, 459, 460, 461], "tempor": [5, 284], "jitter": [5, 14, 159, 234, 265, 276], "__iter__": 5, "rang": [5, 6, 14, 83, 153, 159, 161, 164, 173, 179, 184, 186, 187, 192, 193, 204, 234, 237, 241, 251, 258, 264, 268, 269, 276, 278, 280, 282, 337, 363, 365, 366, 367, 369, 371], "vid": [5, 342], "buffer": [5, 96, 100, 101], "max_seek": 5, "uniform": [5, 206], "current_pt": 5, "yield": 5, "dataload": [5, 12, 17, 76, 508], "112": [5, 20, 342, 382, 446, 457, 460, 461], "loader": [5, 37, 56, 57, 75, 333], "batch_siz": [5, 6, 11, 12, 17, 118, 139, 238, 240, 245], "tensors": 5, "1603": 5, "randomresizedcrop": [5, 9, 11, 13, 221, 313, 508], "etc": [5, 109, 226, 232, 236, 239, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 342, 508], "chang": [5, 9, 12, 13, 14, 36, 94, 159, 186, 190, 220, 221, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 342, 363, 365, 366, 367, 369, 371, 508], "v0": [5, 186, 190, 220, 221, 267, 268, 270, 274, 276, 342, 507], "across": [5, 6, 33, 340, 342], "pil": [5, 12, 13, 14, 15, 25, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 61, 62, 63, 64, 65, 66, 67, 68, 69, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 89, 90, 91, 92, 93, 94, 155, 156, 158, 159, 160, 161, 162, 163, 164, 165, 167, 168, 169, 170, 171, 172, 173, 174, 175, 177, 178, 179, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 234, 235, 237, 239, 244, 246, 247, 249, 251, 252, 253, 256, 257, 258, 261, 264, 266, 267, 268, 269, 270, 271, 274, 276, 279, 280, 282, 283, 307, 309, 321, 327, 328, 333, 334, 335, 340, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472, 508], "backend": [5, 12, 96, 186, 190, 220, 221, 267, 268, 270, 274, 276, 340, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 508], "suppress": [5, 110, 111, 132, 147], "futur": [5, 282, 508], "mean": [5, 9, 11, 13, 14, 24, 96, 124, 138, 141, 145, 153, 155, 156, 158, 159, 163, 164, 165, 168, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 190, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 213, 214, 215, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 234, 241, 242, 246, 249, 250, 253, 257, 259, 261, 266, 271, 278, 283, 305, 335, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "still": [5, 9, 11, 12, 13, 15, 51, 186, 190, 220, 221, 267, 268, 270, 274, 276, 508], "antialias": [5, 186, 190, 220, 221, 267, 268, 270, 274, 276], "infer": [5, 124, 326, 327, 328, 330, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "2933978369482158": 5, "7918684185786267": 5, "373791147055652": 5, "9307513195312067": 5, "6426948895065263": 5, "7999999999999998": 5, "3023": 5, "9049": 5, "437767": 5, "170833": 5, "figur": 5, "figsiz": 5, "permut": [5, 254, 265, 307, 457, 458, 459, 460, 461, 462, 463, 464, 465], "off": [5, 12, 159, 234, 269, 502], "cleanup": 5, "shutil": 5, "rmtree": 5, "277": [5, 6, 7], "plot_video_api": [5, 7], "fig": [6, 130], "make_grid": [6, 338, 340], "dog1_int": 6, "dog2_int": 6, "dog_list": 6, "font": [6, 333], "50": [6, 9, 14, 18, 20, 21, 35, 87, 162, 239, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 384, 396, 401, 427, 430, 433, 435, 472], "200": 6, "210": [6, 13], "350": 6, "430": 6, "yellow": 6, "produc": [6, 14, 116, 117, 162, 239, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "215": 6, "9767": 6, "171": [6, 457, 460, 461], "1661": 6, "402": [6, 20, 342, 393], "0078": 6, "378": 6, "7391": 6, "344": [6, 15, 20, 342, 414], "6341": 6, "172": 6, "6735": 6, "6114": 6, "220": 6, "1435": 6, "153": 6, "1306": 6, "185": [6, 87], "5567": 6, "9223": 6, "254": [6, 380], "7014": 6, "grad_fn": [6, 15], "stackbackward0": 6, "score": [6, 9, 132, 147, 342, 363, 365, 366, 367, 369, 371], "9989": 6, "0701": 6, "0611": 6, "indexbackward0": 6, "5964": 6, "132": [6, 19, 20, 342, 402, 448], "4331": 6, "449": 6, "9359": 6, "493": [6, 12], "0222": 6, "225": [6, 11, 13, 160, 179, 235, 258, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 458, 459, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "8182": 6, "124": 6, "6292": 6, "467": 6, "2861": 6, "492": 6, "2621": 6, "5248": 6, "135": [6, 435], "4171": 6, "420": 6, "9786": 6, "479": [6, 339, 342, 461], "2225": 6, "9980": 6, "0879": 6, "0671": 6, "greater": [6, 147, 190, 220, 274], "threshold": [6, 14, 188, 224, 262, 271, 316, 367], "score_threshold": 6, "dogs_with_box": 6, "dog_int": 6, "treat": [6, 13, 110, 508], "independ": [6, 194, 283, 394, 395], "fcn": [6, 149, 342, 434, 435, 478], "resnet": [6, 18, 110, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 400, 401, 423, 424, 425, 426, 427, 428, 429, 430, 432, 433, 434, 435, 457, 460, 461, 471, 472, 490, 491], "fcn_resnet50": [6, 342], "look": [6, 9, 10, 11, 12, 13, 15, 457, 460, 461, 501, 508], "fcn_resnet50_weight": [6, 24, 342, 435], "resize_s": [6, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "fcn_resnet50_coco": 6, "1167a1af": 6, "500": [6, 361, 362, 363, 365, 366, 367, 369, 371, 372], "089669704437256": 6, "858257293701172": 6, "num_class": [6, 11, 18, 43, 60, 78, 238, 245, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "softmax": [6, 342], "after": [6, 96, 115, 119, 120, 121, 142, 143, 148, 149, 151, 152, 177, 190, 220, 256, 262, 274, 275, 508], "interpret": [6, 96, 104, 158, 163, 177, 179, 186, 191, 208, 225, 232, 240, 256, 258, 268, 277], "probabl": [6, 14, 15, 69, 123, 142, 143, 154, 155, 172, 174, 175, 178, 179, 180, 181, 182, 184, 185, 186, 188, 189, 190, 220, 221, 230, 250, 252, 253, 255, 257, 258, 259, 260, 261, 264, 265, 266, 267, 268, 270, 271, 272, 273, 274, 276, 329], "boat": 6, "sem_class_to_idx": 6, "cl": [6, 342], "categori": [6, 28, 48, 51, 55, 67, 72, 82, 111, 132, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "normalized_mask": [6, 342], "dog_and_boat_mask": 6, "img_idx": 6, "confid": 6, "about": [6, 11, 13, 15, 18, 192, 280, 341, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 508], "much": [6, 13, 15, 96, 159, 172, 195, 196, 198, 199, 200, 234, 250, 265, 508], "top": [6, 18, 33, 110, 116, 117, 121, 123, 134, 170, 177, 184, 205, 208, 216, 217, 221, 225, 248, 256, 264, 297, 313, 335, 340, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 415, 416, 418, 466, 468, 469, 507], "do": [6, 9, 11, 12, 54, 60, 88, 94, 96, 116, 117, 123, 138, 142, 143, 145, 163, 186, 190, 191, 220, 221, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 342, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 507, 508, 509], "class_dim": 6, "boolean_dog_mask": 6, "m": [6, 9, 18, 121, 124, 135, 137, 140, 144, 382], "bool": [6, 25, 26, 28, 29, 31, 35, 36, 37, 38, 40, 42, 45, 48, 51, 52, 54, 55, 56, 59, 60, 61, 65, 66, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 88, 89, 90, 91, 93, 96, 110, 111, 116, 117, 118, 119, 120, 123, 128, 142, 143, 151, 154, 155, 168, 186, 187, 190, 191, 207, 215, 220, 221, 223, 225, 230, 246, 256, 258, 267, 268, 269, 270, 274, 276, 277, 278, 295, 300, 305, 312, 313, 315, 317, 318, 326, 327, 328, 330, 333, 335, 337, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "line": [6, 235, 243, 252, 334], "cryptic": 6, "queri": 6, "gotten": 6, "preserv": [6, 12, 14, 15, 94, 342, 508], "dogs_with_mask": 6, "per": [6, 109, 148, 151, 202, 278, 326, 341], "rememb": 6, "mani": [6, 15, 17, 115, 340, 507], "ask": [6, 12], "c": [6, 54, 60, 70, 83, 88, 100, 101, 104, 107, 108, 109, 122, 142, 143, 148, 149, 151, 152, 163, 164, 167, 169, 191, 192, 193, 207, 215, 232, 239, 240, 241, 244, 247, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 277, 280, 282, 284, 333, 334, 335, 337, 340, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 507, 508], "involv": [6, 14, 15], "dog1_mask": 6, "dog1_all_classes_mask": 6, "arang": [6, 13], "dog_with_all_mask": 6, "drawn": [6, 334, 335], "think": 6, "had": [6, 9, 10, 13, 508], "anoth": [6, 111, 147], "among": 6, "seen": [6, 13], "index": [6, 12, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 132, 135, 137, 140, 144, 148, 149, 151, 152, 307, 340], "juggl": 6, "all_classes_mask": 6, "swap": 6, "swapax": 6, "significantli": 6, "analyz": 6, "rcnn": [6, 12], "don": [6, 10, 12, 13, 110, 186, 190, 220, 221, 267, 268, 270, 274, 276, 329], "person": [6, 31, 66, 340, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372], "extra": [6, 83, 121, 235, 243, 252, 275], "keypointrcnn_resnet50_fpn": 6, "maskrcnn_resnet50_fpn": [6, 368], "maskrcnn_resnet50_fpn_weight": [6, 23, 342, 367], "maskrcnn_resnet50_fpn_coco": 6, "bf2d0c1e": 6, "219": 6, "7444": 6, "168": 6, "1722": 6, "7378": 6, "384": [6, 20, 342, 377, 383, 415, 416, 418, 466], "0263": 6, "343": 6, "9716": 6, "2287": 6, "358": 6, "3447": 6, "222": 6, "6263": 6, "301": 6, "0303": 6, "192": [6, 14], "6917": 6, "313": 6, "8879": 6, "232": [6, 24, 342, 353, 354, 388, 390, 391, 392, 398, 399, 401, 402, 403, 406, 407, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 427, 428, 429, 430, 434, 439, 440, 445, 471, 472], "3154": 6, "9987": 6, "7187": 6, "6525": 6, "unsqueezebackward0": 6, "6767": 6, "137": [6, 21, 342, 366], "9018": 6, "446": 6, "5324": 6, "487": 6, "3429": 6, "0000": 6, "288": [6, 20, 342, 375, 410], "0053": 6, "489": 6, "9292": 6, "490": 6, "2352": 6, "9978": 6, "0697": 6, "break": [6, 11, 94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340], "down": [6, 12, 18, 110], "vari": [6, 12, 342], "organ": [6, 42], "kei": [6, 9, 11, 12, 18, 61, 94, 110, 275, 278], "num_inst": [6, 334], "element": [6, 12, 13, 54, 60, 88, 116, 117, 132, 135, 137, 139, 140, 142, 143, 144, 147, 148, 149, 151, 152, 153, 170, 177, 216, 248, 256, 275, 307], "quit": 6, "saw": 6, "dog1_output": 6, "9999862909317017": 6, "interestingli": 6, "ahead": [6, 340], "could": [6, 10, 13, 15, 18, 508], "choos": [6, 164, 241], "proba_threshold": 6, "dog1_bool_mask": 6, "There": [6, 13, 15, 98, 111, 326, 342], "seem": [6, 177, 256], "confus": 6, "close": [6, 352, 386, 387, 389, 391, 398, 400, 401, 402, 404, 405, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 430, 437, 438, 441, 442, 443, 444, 445, 446, 447, 448, 457, 460, 461, 471, 472, 508], "help": [6, 9, 10, 190, 220, 274, 508], "relev": [6, 12, 110, 111, 341], "clearli": 6, "good": [6, 13, 508], "when": [6, 12, 13, 14, 15, 18, 69, 76, 110, 111, 138, 141, 145, 147, 161, 170, 177, 190, 193, 204, 216, 220, 237, 239, 248, 251, 256, 264, 269, 273, 274, 282, 284, 508], "75": [6, 14, 19, 20, 100, 107, 186, 268, 342, 359, 380, 385, 388, 392, 397, 399, 401, 407, 413, 420, 467, 471], "boolean_mask": 6, "lower": [6, 147, 186, 268, 341, 509], "bench": 6, "wa": [6, 12, 13, 35, 37, 77, 202, 206, 275, 331, 338, 342, 343, 352, 394], "draw_keypoint": [6, 340], "keypointrcnn": [6, 366, 477], "keypointrcnn_resnet50_fpn_weight": [6, 21, 342, 366], "person_int": 6, "person1": 6, "person_float": 6, "keypointrcnn_resnet50_fpn_coco": 6, "fc266e95": 6, "3751": 6, "177": [6, 368], "9242": 6, "6354": 6, "574": [6, 19, 339, 342, 402, 463], "7064": 6, "3625": 6, "180": [6, 14, 201], "7574": 6, "290": 6, "1061": 6, "390": 6, "7958": 6, "9998": 6, "1070": 6, "208": [6, 382], "0176": 6, "214": 6, "2408": 6, "207": [6, 408, 434], "0375": 6, "197": [6, 20, 342], "8246": 6, "6392": 6, "211": [6, 394], "8398": 6, "178": [6, 20, 24, 342, 433, 442], "6378": 6, "217": 6, "8425": 6, "221": 6, "2086": 6, "253": 6, "8590": 6, "160": [6, 433], "6502": 6, "269": 6, "4662": 6, "243": 6, "9929": 6, "304": [6, 20, 342, 466], "2822": 6, "138": [6, 20, 342, 358], "4654": 6, "8935": 6, "5698": 6, "8990": 6, "4551": 6, "374": [6, 20, 342, 415, 423, 452], "5144": 6, "226": [6, 366], "375": 6, "7150": 6, "370": [6, 13], "3125": 6, "8082": 6, "455": 6, "5516": 6, "273": 6, "9723": 6, "448": 6, "9486": 6, "193": 6, "6275": 6, "546": [6, 20, 342, 423], "1932": 6, "3727": 6, "545": 6, "5930": 6, "8327": 6, "6636": 6, "2343": 6, "4622": 6, "198": [6, 20, 342, 430], "2590": 6, "209": 6, "8627": 6, "4310": 6, "4628": 6, "5134": 6, "218": [6, 20, 342, 456, 464], "2642": 6, "7997": 6, "251": 6, "8704": 6, "162": 6, "3579": 6, "2736": 6, "245": 6, "5289": 6, "6800": 6, "4238": 6, "330": [6, 466], "4848": 6, "278": 6, "4382": 6, "346": 6, "0876": 6, "3826": 6, "8929": 6, "233": [6, 352, 432], "5618": 6, "368": [6, 20, 339, 342, 418, 462], "2917": 6, "7832": 6, "367": 6, "6916": 6, "289": 6, "8069": 6, "4897": 6, "389": 6, "8956": 6, "281": 6, "4300": 6, "349": 6, "0882": 6, "0294": 6, "copyslic": 6, "keypoints_scor": 6, "0164": 6, "6672": 6, "8312": 6, "6510": 6, "2053": 6, "8280": 6, "1136": 6, "2084": 6, "1901": 6, "13": [6, 20, 339, 342, 372, 391, 398, 401, 406, 427, 439, 451, 452, 457], "8453": 6, "7090": 6, "5852": 6, "5005": 6, "3378": 6, "3700": 6, "2987": 6, "4479": 6, "9326": 6, "8158": 6, "9053": 6, "9368": 6, "9585": 6, "4240": 6, "8328": 6, "4227": 6, "2907": 6, "1066": 6, "1019": 6, "1822": 6, "3057": 6, "9904": 6, "7409": 6, "7874": 6, "9329": 6, "dictionari": [6, 9, 12, 13, 18, 37, 61, 90, 94, 96, 109, 110, 111, 239, 248, 251, 256, 264, 269, 273, 341, 363, 365, 366, 367, 369, 371], "keypoint_scor": 6, "kpt": 6, "recogn": 6, "surfboard": 6, "realiz": 6, "enough": [6, 10, 187, 223, 269, 275], "filter": [6, 35, 115], "detect_threshold": 6, "great": [6, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472], "y": [6, 31, 69, 173, 187, 201, 209, 251, 269, 334, 366], "coordin": [6, 14, 15, 35, 94, 124, 133, 148, 149, 151, 152, 207, 236, 275, 325, 333], "visibl": [6, 366], "radiu": [6, 334], "appear": [6, 14, 18], "circl": 6, "over": [6, 33, 135, 137, 138, 140, 141, 144, 145, 147, 161, 163, 191, 204, 237, 337, 394, 395], "coco": [6, 12, 33, 34, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "coco_keypoint": 6, "nose": [6, 366], "left_ey": [6, 366], "right_ey": [6, 366], "left_ear": 6, "right_ear": 6, "left_should": 6, "right_should": 6, "left_elbow": 6, "right_elbow": 6, "left_wrist": 6, "right_wrist": 6, "left_hip": 6, "right_hip": 6, "left_kne": 6, "right_kne": 6, "left_ankl": 6, "right_ankl": 6, "especi": [6, 18, 508], "pose": [6, 48], "action": [6, 54, 60, 88, 457, 460, 461, 501], "recognit": [6, 18, 52, 54, 60, 77, 82, 88, 400, 401, 423, 424, 425, 426, 427, 449, 450, 451, 452, 453, 454, 455, 456, 457, 460, 461, 466, 467, 468, 469, 470, 488, 489, 499, 501, 504], "connect": [6, 116, 117, 334, 346, 357, 358, 359, 360], "observ": [6, 18], "reveal": 6, "construct": [6, 13, 94, 239, 326, 327, 328, 330, 362, 364, 365, 366, 369, 370, 381, 382, 383, 386, 392, 393, 398, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 431, 432, 433, 436, 437, 438, 439, 440, 443, 444, 445, 446, 447, 448, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470], "human": 6, "skeleton": 6, "connect_skeleton": 6, "555": [6, 7], "plot_visualization_util": [6, 7], "00": [7, 12, 16, 414], "34": [7, 20, 22, 23, 339, 342, 354, 367, 371, 392, 399, 418, 426, 430, 457], "118": [7, 20, 342, 386], "auto_examples_oth": 7, "mb": [7, 16, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "05": [7, 122, 265, 339, 342, 462], "02": [7, 20, 179, 258, 342, 449, 471], "01": [7, 394], "guid": [9, 10, 15], "explain": [9, 10, 11], "compat": [9, 12, 13, 76, 83, 92, 94, 96, 98, 278, 340, 341, 342, 345, 349, 350, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 432, 433, 434, 435, 436, 457, 458, 459, 460, 461, 462, 477, 478, 479, 492, 496, 497, 501, 503, 507, 508], "tv_tensor": [9, 10, 12, 13, 94, 236, 239, 240, 248, 251, 256, 262, 264, 269, 273, 278, 281, 307, 311, 325, 326, 327, 328, 329, 330, 331, 332, 508], "alreadi": [9, 10, 11, 13, 15, 25, 28, 29, 31, 35, 36, 38, 40, 42, 45, 48, 51, 52, 55, 59, 60, 61, 65, 66, 68, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 89, 90, 91, 93, 94, 508], "know": [9, 10, 12, 13, 186, 190, 220, 221, 267, 268, 270, 274, 276], "classif": [9, 11, 48, 57, 73, 75, 77, 94, 153, 340, 343, 352, 363, 365, 366, 367, 369, 371, 453, 459, 500, 502, 507, 508], "typic": [9, 11, 12, 13, 15, 109, 340, 508], "hard": [9, 153], "mycustomtransform": 9, "new_img": 9, "new_label": 9, "custom": [9, 10, 12, 13, 37, 56, 235, 243, 252, 311, 329, 340, 342, 508], "ani": [9, 10, 12, 13, 15, 25, 26, 28, 29, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41, 42, 43, 45, 46, 47, 48, 51, 52, 54, 55, 56, 57, 59, 60, 61, 65, 66, 67, 68, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 88, 89, 90, 91, 93, 94, 96, 104, 109, 110, 111, 112, 115, 158, 172, 186, 190, 195, 196, 200, 203, 205, 218, 220, 221, 226, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 309, 322, 326, 327, 328, 330, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "complet": [9, 137, 138, 198, 340], "sampl": [9, 11, 12, 13, 17, 27, 30, 31, 32, 33, 34, 35, 36, 37, 39, 40, 41, 42, 44, 48, 51, 52, 54, 57, 58, 60, 61, 62, 63, 69, 72, 73, 75, 77, 78, 81, 82, 84, 86, 88, 90, 91, 94, 109, 124, 148, 151, 155, 173, 230, 238, 240, 245, 251, 267, 273, 284, 326, 370, 508], "alwai": [9, 10, 12, 15, 18, 186, 190, 220, 221, 267, 268, 270, 274, 276, 331, 338, 508], "though": [9, 508], "std": [9, 11, 13, 168, 215, 246, 305, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "rand": [9, 10, 11, 15, 110, 121, 124, 139, 240, 307, 331, 361, 362, 363, 365, 366, 367, 369, 371, 372], "boundingbox": [9, 10, 12, 13, 15, 94, 232, 239, 248, 251, 256, 260, 262, 264, 267, 268, 269, 270, 272, 273, 274, 276, 332, 340, 508], "20": [9, 12, 15, 20, 78, 90, 121, 342, 373, 394, 431, 432, 433, 434, 435, 436, 446], "xyxi": [9, 13, 15, 94, 134, 236, 325, 508], "canvas_s": [9, 13, 15, 233, 294, 326, 332, 508], "out_img": [9, 13], "out_bbox": 9, "out_label": 9, "boundingboxformat": [9, 15, 236, 294, 295, 326, 340], "nout_bbox": 9, "136": [9, 20, 342, 371, 421], "173": 9, "sure": [9, 10, 12, 14, 15, 98, 160, 174, 235, 252], "familiar": [9, 18], "yourself": 9, "section": [9, 13, 73, 508], "ok": 9, "flexibl": 9, "possibl": [9, 15, 69, 78, 186, 190, 220, 221, 236, 267, 268, 270, 274, 276, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "limit": [9, 340], "featur": [9, 110, 111, 119, 120, 121, 122, 124, 148, 149, 151, 152, 238, 340, 453, 462, 502, 508], "builtin": 9, "tupl": [9, 11, 12, 13, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 94, 104, 105, 111, 116, 117, 124, 136, 139, 148, 149, 151, 152, 156, 159, 163, 164, 170, 173, 174, 177, 179, 184, 186, 191, 208, 216, 225, 231, 234, 239, 240, 241, 248, 251, 252, 256, 258, 264, 265, 268, 269, 270, 273, 275, 276, 277, 294, 301, 317, 326, 329, 333, 334, 335, 337, 508], "nest": [9, 13], "structured_input": 9, "something_that_will_be_ignor": 9, "hello": 9, "structured_output": 9, "246": [9, 20, 342, 403, 429, 444, 451], "186": [9, 20, 342, 374], "206": 9, "reproduc": [9, 352, 386, 387, 389, 391, 398, 400, 401, 402, 404, 405, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 430, 437, 438, 441, 442, 443, 444, 445, 446, 447, 448, 457, 460, 461, 471, 472], "behavior": [9, 15, 147, 342, 363, 365, 366, 367, 369, 371], "invit": 9, "adapt": [9, 37, 148, 151, 370], "brief": 9, "core": [9, 10, 13, 507, 508], "logic": [9, 12, 13], "unpack": 9, "flat": 9, "pytre": 9, "decis": 9, "made": [9, 110, 158, 163, 177, 186, 191, 192, 208, 225, 232, 240, 256, 268, 277, 280, 508], "base": [9, 10, 14, 43, 57, 74, 92, 121, 155, 156, 162, 171, 197, 230, 231, 239, 249, 276, 284, 329, 340, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 423, 424, 425, 426, 427, 428, 429, 430, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "subclass": [9, 12, 13, 15, 17, 26, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283, 326, 327, 328, 330, 332, 509], "plu": [9, 191, 225, 277], "check": [9, 12, 37, 56, 210, 211, 212, 342, 508], "potenti": 9, "repack": 9, "public": [9, 10, 342, 508], "dev": [9, 111], "face": [9, 93], "tool": 9, "achiev": [9, 14, 96, 110], "someth": [9, 15], "valuabl": 9, "issu": [9, 15, 94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284], "repo": [9, 27, 30, 442], "006": [9, 12, 16, 20, 342, 407, 440], "plot_custom_transform": [9, 16], "intend": [10, 12, 15], "advanc": [10, 13, 15, 96, 509], "downstream": [10, 18], "librari": [10, 98, 333, 342], "maintain": [10, 18, 220, 340], "befor": [10, 11, 13, 54, 57, 60, 88, 98, 124, 159, 186, 198, 234, 265, 268, 275, 342, 371], "continu": 10, "inherit": [10, 56], "cover": [10, 12, 13, 33, 238, 245], "elabor": 10, "carri": 10, "mytvtensor": 10, "my_dp": 10, "perform": [10, 14, 15, 18, 96, 97, 121, 122, 124, 125, 126, 127, 128, 129, 130, 131, 132, 139, 147, 148, 149, 151, 152, 161, 164, 167, 174, 179, 184, 190, 197, 209, 217, 218, 220, 230, 231, 235, 241, 244, 249, 252, 255, 258, 263, 264, 273, 274, 275, 276, 283, 309, 340, 341, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 506], "hook": [10, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "via": [10, 15, 96, 110, 124, 134, 342, 370, 415, 416, 418, 466, 468, 469], "register_kernel": 10, "flip": [10, 12, 14, 181, 189, 191, 213, 225, 229, 260, 272, 277], "regist": [10, 112, 113, 115, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283, 311], "hflip": [10, 14], "tv_tensor_cl": [10, 311], "hflip_my_tv_tensor": 10, "arg": [10, 17], "kwarg": [10, 38, 57, 76, 332, 338, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "wrap": [10, 18, 94, 110, 111, 240, 340], "understand": [10, 15, 82], "why": 10, "ignor": [10, 13, 70, 122, 125, 126, 127, 128, 129, 130, 131, 153, 161, 174, 186, 190, 220, 221, 230, 231, 235, 239, 249, 252, 255, 263, 267, 268, 270, 274, 275, 276, 283, 332], "itself": [10, 18, 88], "And": [10, 12, 508], "intern": [10, 15, 54, 88, 96, 343, 344, 345, 346, 347, 348, 349, 350, 351, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "cannot": [10, 161, 204, 237], "reason": [10, 15, 111, 186, 190, 220, 221, 267, 268, 270, 274, 276], "therefor": [10, 190, 198, 220, 274], "backward": [10, 96, 98, 340, 341, 342, 345, 349, 350, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 432, 433, 434, 435, 436, 457, 458, 459, 460, 461, 462, 477, 478, 479, 492, 496, 497, 501, 503, 507, 508], "guarante": [10, 18, 96, 98, 147, 238, 245, 341, 342, 345, 349, 350, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 432, 433, 434, 435, 436, 457, 458, 459, 460, 461, 462, 477, 478, 479, 492, 496, 497, 501, 503, 507], "won": [10, 116, 117, 123, 124], "renam": 10, "proper": [10, 12], "deprec": [10, 237, 282, 296, 342, 508], "cycl": 10, "howev": [10, 14, 15, 83, 269, 340], "imagin": 10, "version": [10, 14, 18, 25, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 96, 98, 110, 111, 151, 164, 165, 180, 191, 209, 222, 225, 226, 277, 342, 353, 354, 355, 356, 374, 382, 383, 391, 392, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 466, 467, 469, 470, 507, 508], "inplac": [10, 15, 116, 117, 119, 120, 123, 142, 143, 168, 179, 207, 215, 246, 258, 295, 300, 305], "noqa": 10, "fail": [10, 12, 342], "doesn": [10, 15, 121, 186, 190, 220, 221, 255, 267, 268, 270, 274, 276, 281, 307], "signatur": 10, "technic": 10, "ad": [10, 94, 116, 117, 121, 122, 142, 143, 162, 239, 508], "004": [10, 16, 19, 342, 399], "plot_custom_tv_tensor": [10, 16], "popular": [11, 94, 340], "augment": [11, 12, 13, 155, 156, 171, 179, 194, 230, 231, 249, 258, 267, 276, 283, 340], "strategi": [11, 156, 231, 238], "improv": [11, 155, 230, 340, 353, 354, 355, 356, 364, 368, 370, 374, 382, 383, 391, 392, 393, 398, 399, 401, 402, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 427, 428, 430, 459, 471, 472, 488, 500, 508], "rest": 11, "dataset": [11, 15, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 157, 163, 191, 194, 208, 225, 240, 275, 277, 283, 340, 342, 394, 395, 431, 432, 433, 434, 435, 436, 507, 508], "fakedata": 11, "preproc": 11, "piltotensor": [11, 31, 33, 34, 46, 47, 61, 160, 163, 179, 191, 218, 235, 258, 309], "todtyp": [11, 12, 13, 237, 282, 318, 508], "scale": [11, 12, 13, 14, 31, 48, 124, 145, 148, 149, 151, 152, 161, 169, 173, 179, 186, 190, 193, 201, 204, 237, 247, 251, 258, 262, 268, 273, 276, 278, 279, 281, 282, 291, 318, 337, 347, 373, 374, 375, 376, 377, 378, 379, 380, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 498, 499, 504, 508], "485": [11, 13, 160, 179, 235, 258, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "456": [11, 13, 20, 160, 179, 235, 258, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "406": [11, 13, 160, 179, 235, 258, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "229": [11, 13, 160, 179, 235, 258, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "1000": [11, 43], "67": [11, 20, 24, 339, 342, 387, 393, 432, 460, 463], "thing": [11, 12, 13, 508], "neither": 11, "nor": 11, "later": [11, 13, 262], "refresh": 11, "weren": 11, "shuffl": [11, 17, 238, 245], "simplest": 11, "right": [11, 44, 50, 58, 84, 134, 170, 177, 184, 197, 208, 216, 217, 220, 225, 248, 256, 264, 342], "cutmix_or_mixup": 11, "randomchoic": 11, "went": 11, "loss": [11, 15, 83, 138, 141, 145, 153, 340, 363, 365, 366, 367, 369, 371, 431, 432, 433, 434, 435, 436, 492], "cross_entropi": 11, "disadvantag": 11, "doe": [11, 27, 39, 55, 62, 63, 159, 160, 161, 166, 167, 168, 169, 174, 176, 179, 183, 192, 193, 198, 204, 207, 215, 218, 226, 227, 228, 234, 235, 237, 243, 244, 246, 247, 252, 255, 258, 263, 265, 279, 280, 282, 309, 321, 340, 341, 508], "advantag": [11, 508], "doc": [11, 13, 15], "learn": [11, 13, 15, 156, 157, 231, 340, 364, 368, 400, 401, 415, 416, 418, 423, 424, 425, 426, 427, 462, 466, 468, 469, 488, 489, 502, 507, 508], "default_col": 11, "collate_fn": [11, 12], "num_work": [11, 17, 54, 60, 88, 508], "No": [11, 138, 141, 145, 153, 342], "far": [11, 394, 395], "ve": [11, 12, 13], "magic": 11, "common": [11, 17, 238, 245, 340, 508], "document": [11, 18, 110, 111, 338, 340, 342, 508], "labels_gett": [11, 238, 245, 275], "callabl": [11, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 113, 116, 117, 121, 123, 130, 235, 238, 243, 245, 252, 255, 263, 275, 372], "randint": [11, 13, 363, 508], "some_other_kei": 11, "through": [11, 13, 14, 15, 110, 111, 162, 239, 340], "155": [11, 16], "plot_cutmix_mixup": [11, 16], "nativ": [12, 15, 340, 506], "enabl": [12, 96], "jointli": 12, "showcas": [12, 15], "everyth": [12, 13, 15], "semant": [12, 13, 32, 78, 94, 221, 340, 345, 350, 431, 432, 433, 434, 435], "fake": [12, 43], "practic": [12, 13, 171, 249, 269, 341, 404, 405, 406, 407, 437, 438, 439, 440, 493, 494], "replac": [12, 94, 278], "images_path": 12, "annotations_path": 12, "cocodetect": [12, 13, 94], "memori": [12, 96, 98, 341, 508], "dict_kei": 12, "iscrowd": 12, "image_id": [12, 94], "category_id": [12, 55], "author": [12, 71, 74, 394, 395], "overcom": 12, "wrap_dataset_for_transforms_v2": [12, 13], "target_kei": [12, 94], "_bounding_box": [12, 13], "_mask": 12, "specifi": [12, 18, 28, 31, 32, 38, 54, 55, 60, 72, 76, 88, 98, 110, 121, 124, 133, 138, 141, 145, 170, 177, 186, 201, 205, 216, 248, 256, 268, 278, 333, 337, 340, 372], "kind": [12, 508], "drop": [12, 54, 60, 88, 119, 120, 142, 143, 154], "unncessari": 12, "previou": [12, 96], "shouldn": 12, "slow": 12, "unnecessarili": 12, "baselin": 12, "handl": [12, 54, 88, 508], "boud": 12, "toimag": [12, 282, 320, 508], "randomphotometricdistort": [12, 13], "randomzoomout": 12, "fill": [12, 155, 156, 162, 170, 171, 173, 177, 184, 187, 194, 201, 216, 217, 223, 230, 231, 239, 248, 249, 251, 256, 264, 269, 273, 283, 291, 298, 306, 308, 315, 333], "123": [12, 37, 56, 365], "117": 12, "104": 12, "randomioucrop": [12, 275], "sanitizeboundingbox": [12, 13, 262], "few": [12, 13, 15, 18, 342], "worth": [12, 466, 467, 468, 469, 470, 504], "isn": [12, 15, 141], "strictli": 12, "necessari": [12, 92, 342], "degener": [12, 275], "place": [12, 13, 15, 57, 75, 116, 117, 123, 142, 143, 168, 207, 215, 246, 269, 423, 424, 427, 488], "least": [12, 150], "pipelin": [12, 13, 262, 275, 331, 508], "particularli": 12, "critic": [12, 275, 342], "distort": [12, 184, 264, 265], "zoom": [12, 273], "accordingli": [12, 13, 31, 81, 83, 161, 193, 204, 237, 282, 385, 397], "further": [12, 110], "ado": 12, "data_load": [12, 17], "collat": [12, 508], "tri": 12, "get_model": [12, 342], "maskrcnn_resnet50_fpn_v2": 12, "weights_backbon": [12, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "loss_dict": 12, "put": [12, 18, 25, 28, 29, 31, 35, 36, 38, 40, 42, 45, 48, 51, 52, 55, 59, 61, 65, 66, 68, 70, 71, 72, 73, 74, 76, 77, 78, 79, 80, 81, 82, 83, 87, 89, 90, 91, 93, 307], "loss_val": 12, "3f": 12, "512": [12, 13, 15, 20, 51, 124, 342, 469, 471, 472], "409": 12, "loss_classifi": 12, "721": 12, "loss_box_reg": 12, "loss_mask": 12, "734": [12, 20, 342, 387], "loss_object": 12, "691": 12, "loss_rpn_box_reg": 12, "036": 12, "find": [12, 15, 18, 37, 262, 275], "disclaim": [12, 13], "complex": 12, "afraid": 12, "keep": [12, 18, 173, 185, 201, 219, 251, 266], "813": [12, 16], "plot_transforms_e2": [12, 16], "setup": 13, "astronaut": [13, 14, 15], "behav": [13, 15], "regular": [13, 15, 142, 143, 238], "instanti": [13, 112, 343, 344, 345, 346, 347, 348, 349, 350, 351, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "care": [13, 18, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "Such": 13, "pretti": 13, "convent": [13, 18, 110, 111, 238, 245, 340], "explor": 13, "switch": [13, 342], "easi": [13, 153], "fulli": [13, 94, 341, 350, 351, 365, 434, 435, 508], "beyond": [13, 245, 508], "briefli": 13, "bare": 13, "510": 13, "275": 13, "130": [13, 342, 369], "345": 13, "425": 13, "out_box": 13, "focus": 13, "question": [13, 457, 458, 459, 460, 461, 462, 463, 464, 465], "answer": [13, 457, 458, 459, 460, 461, 462, 463, 464, 465], "feel": [13, 15], "plain": [13, 15, 307], "sum": [13, 15, 138, 141, 145, 153, 255], "img_dp": 13, "25087958": 13, "dispatch": [13, 15, 509], "appropri": [13, 15, 97, 342, 509], "who": 13, "both": [13, 14, 15, 37, 39, 44, 50, 62, 63, 84, 85, 86, 92, 104, 111, 135, 137, 138, 140, 141, 144, 145, 150, 170, 177, 198, 203, 209, 216, 235, 243, 248, 252, 256, 363, 365, 366, 367, 369, 371, 508], "arbitrari": [13, 153, 154, 155, 156, 158, 159, 163, 164, 165, 170, 171, 172, 173, 175, 177, 178, 180, 181, 182, 184, 185, 186, 187, 188, 189, 191, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 206, 208, 209, 213, 214, 216, 217, 219, 220, 221, 222, 223, 224, 225, 229, 230, 231, 232, 234, 239, 240, 241, 242, 248, 249, 250, 251, 253, 256, 257, 259, 260, 261, 262, 264, 266, 267, 268, 269, 270, 271, 272, 273, 274, 276, 277, 283, 508], "arbitrarili": 13, "anyth": [13, 18], "target_dict": 13, "this_is_ignor": 13, "definit": [13, 342], "out_target": 13, "back": [13, 161, 198, 204, 237], "mention": [13, 148], "foreign": 13, "simpli": 13, "debug": 13, "safe": 13, "skip": 13, "pure": [13, 15, 281, 327, 331, 508], "specif": [13, 18, 94, 110, 278, 341, 342, 394, 395, 506, 508], "inde": [13, 141], "haven": 13, "yet": [13, 121, 194, 283, 340, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 507], "whether": [13, 69, 76, 97, 110, 111, 116, 117, 123, 186, 190, 220, 221, 267, 268, 270, 274, 276, 278, 326, 327, 328, 330, 342, 508], "depth": [13, 44, 121, 154, 155, 192, 227, 230, 280, 321, 387, 388, 389, 390], "wise": 13, "travers": 13, "happen": [13, 340], "although": [13, 110, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283, 326, 340, 508], "roughli": [13, 18, 87], "speak": 13, "must": [13, 96, 98, 100, 101, 107, 108, 110, 124, 148, 149, 151, 152, 198, 206, 262, 342], "depend": [13, 18, 96, 190, 192, 220, 274, 280, 336, 342, 363, 365, 366, 367, 369, 371], "datatset": 13, "predat": 13, "forc": 13, "my_transform": 13, "convers": [13, 159, 161, 198, 204, 234, 237, 265, 278], "At": [13, 341], "sub": [13, 18, 110], "step": [13, 18, 54, 60, 88, 342, 394], "either": [13, 14, 37, 39, 44, 49, 50, 53, 54, 58, 60, 62, 63, 64, 69, 85, 86, 88, 104, 105, 110, 262, 394, 395, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422], "838": [13, 16, 20, 342, 374, 418], "plot_transforms_getting_start": [13, 16], "variou": [14, 95, 96, 97, 98, 99, 103, 506, 510], "seed": [14, 43], "orig_img": [14, 155], "alter": 14, "orient": 14, "posit": [14, 77, 118, 139, 148, 149, 153, 198], "mathemat": 14, "desir": [14, 15, 18, 97, 98, 99, 103, 110, 127, 155, 156, 158, 161, 162, 163, 165, 171, 173, 177, 184, 186, 187, 190, 191, 194, 201, 204, 208, 217, 220, 221, 223, 225, 230, 231, 232, 237, 239, 240, 242, 249, 251, 256, 264, 267, 268, 269, 270, 274, 276, 277, 283, 326, 327, 328, 330], "border": [14, 170, 177, 216, 248, 256], "padded_img": 14, "resized_img": 14, "center_crop": 14, "crop": [14, 152, 158, 163, 177, 186, 191, 203, 208, 221, 225, 232, 240, 256, 262, 268, 277, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "center": [14, 134, 139, 141, 158, 167, 173, 187, 201, 203, 208, 223, 225, 232, 244, 251, 269, 291, 315], "five_crop": 14, "corner": [14, 74, 134, 163, 173, 187, 191, 201, 205, 207, 208, 217, 221, 223, 225, 240, 251, 269, 277], "central": [14, 138, 163, 191, 208, 225, 240, 277, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "top_left": 14, "top_right": 14, "bottom_left": 14, "bottom_right": 14, "perspect": [14, 184, 264], "perspective_transform": 14, "distortion_scal": [14, 184, 264], "perspective_img": 14, "rotat": [14, 173, 187, 201, 251, 269], "angl": [14, 187, 201, 223, 269, 291, 315], "degre": [14, 173, 184, 187, 201, 223, 251, 264, 269], "rotated_img": 14, "affin": [14, 122, 173, 251], "affine_transfom": 14, "translat": [14, 173, 187, 201, 223, 251, 269, 291], "affine_img": 14, "elastic_transform": 14, "morpholog": [14, 162, 239], "water": [14, 162, 239], "effect": [14, 161, 162, 204, 237, 239, 269, 273, 331], "250": [14, 51], "transformed_img": 14, "locat": [14, 30, 32, 44, 58, 61, 62, 63, 84, 86, 177, 205, 256, 334, 366], "cropper": 14, "128": [14, 22, 342, 365, 394, 395, 457, 460, 461, 462], "resized_crop": 14, "resize_cropp": 14, "modifi": [14, 275, 353, 354, 355, 356, 374, 382, 383, 391, 392, 398, 399, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 466, 467, 469, 470], "bright": [14, 33, 155, 159, 195, 230, 234, 265, 285], "contrast": [14, 155, 159, 196, 202, 230, 234, 265, 385, 397], "tone": 14, "except": [14, 177, 256, 340, 342, 471, 472], "to_grayscal": [14, 222], "gray_img": 14, "cmap": 14, "grai": [14, 95, 98, 196], "satur": [14, 159, 199, 234, 265, 289], "hue": [14, 159, 198, 234, 265, 288], "jittered_img": 14, "gaussian_blur": 14, "gaussian": [14, 74, 164, 209, 241], "blur": [14, 164, 172, 200, 209, 241, 250], "blurrer": 14, "kernel_s": [14, 116, 117, 118, 164, 209, 241, 302], "sigma": [14, 130, 162, 164, 209, 239, 241, 302], "blurred_img": 14, "invert": [14, 182, 188, 224, 261, 271, 372, 391, 398, 482, 483, 497], "invertered_img": 14, "poster": [14, 185, 266], "reduc": [14, 171, 185, 219, 249, 266], "channel": [14, 18, 54, 60, 88, 100, 101, 104, 107, 108, 109, 116, 117, 121, 123, 124, 130, 159, 165, 168, 170, 177, 179, 180, 185, 192, 197, 198, 202, 210, 211, 215, 216, 219, 222, 226, 234, 239, 242, 246, 248, 251, 254, 256, 258, 259, 264, 265, 266, 269, 273, 280, 307, 404, 405, 406, 407, 437, 438, 439, 440, 471, 472, 508], "posterized_img": 14, "solar": [14, 188, 271], "solarized_img": 14, "adjust_sharp": 14, "adjust": [14, 172, 192, 195, 196, 197, 198, 199, 200, 250, 265, 280, 285, 287, 288, 289], "sharp": [14, 155, 172, 200, 230, 250], "sharpness_adjust": 14, "sharpness_factor": [14, 172, 200, 250, 290], "sharpened_img": 14, "autocontrast": [14, 175, 253], "autocontrasted_img": 14, "histogram": [14, 178, 206, 257], "equalized_img": 14, "automat": [14, 60, 255], "polici": [14, 156, 157, 231, 508], "autoaugmentpolici": [14, 156, 231], "cifar10": [14, 26, 157, 508], "svhn": [14, 157, 508], "row_titl": 14, "strength": [14, 162, 239], "interpol": [14, 60, 148, 151, 155, 156, 159, 162, 171, 173, 184, 186, 187, 190, 194, 198, 201, 217, 220, 221, 223, 230, 231, 234, 239, 249, 251, 264, 265, 267, 268, 269, 270, 274, 276, 283, 284, 291, 298, 308, 312, 313, 315, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "chanc": 14, "hflipper": 14, "vflip": 14, "vflipper": 14, "applier": 14, "772": [14, 16], "plot_transforms_illustr": [14, 16], "introduc": [15, 343, 352], "audienc": 15, "unless": [15, 186, 190, 220, 221, 267, 268, 270, 274, 276, 329, 332], "fairli": 15, "low": [15, 39, 361, 508], "level": [15, 18, 42, 110, 121, 124, 341, 441, 457, 458, 459, 460, 461, 462, 463, 464, 465, 495, 508, 509], "topic": 15, "worri": 15, "effici": [15, 404, 405, 406, 407, 437, 438, 439, 440, 493, 494], "zero": [15, 137, 138, 139, 140, 141, 142, 143, 145, 154, 167, 244], "copi": [15, 18, 218, 276, 309], "data_ptr": 15, "under": [15, 74, 111, 112, 113, 265, 509], "hood": [15, 265, 509], "correctli": 15, "gotcha": 15, "turn": [15, 110, 159, 234, 326, 327, 328, 330], "creation": [15, 54, 60, 88], "requires_grad": [15, 326, 327, 328, 330], "float_imag": 15, "aliasbackward0": 15, "addit": [15, 18, 48, 69, 94, 138, 141, 145, 341], "alongsid": 15, "495": 15, "new_bbox": 15, "overrid": [15, 37, 92], "shift": [15, 151, 173, 198, 251, 337, 443, 444, 445, 498], "affect": [15, 186, 190, 220, 221, 267, 268, 270, 274, 276, 331], "set_return_typ": [15, 340], "global": [15, 331], "config": [15, 112], "whole": [15, 105, 161, 204, 237], "program": [15, 331], "context": [15, 331], "manag": [15, 331], "caveat": 15, "__torch_function__": [15, 331], "protocol": 15, "induc": 15, "small": [15, 75, 137, 138, 140, 141, 145, 239, 355, 393, 395, 459], "overhead": [15, 331], "matter": 15, "problem": 15, "better": [15, 139, 151], "anywai": 15, "sens": [15, 186, 190, 220, 221, 267, 268, 270, 274, 276], "prefer": [15, 190, 220, 274, 508], "logit": 15, "behaviour": [15, 331, 342], "activ": [15, 110, 116, 117, 123, 130, 340], "feedback": [15, 94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 508], "reach": 15, "7319": 15, "unwrap": 15, "rule": 15, "clone": 15, "requires_grad_": 15, "retain": 15, "obj": 15, "add_": 15, "new_imag": 15, "mul_": 15, "share": 15, "underli": [15, 218, 309], "008": [15, 16, 20, 342, 376, 379], "plot_tv_tensor": [15, 16], "597": 16, "auto_examples_transform": 16, "08": [16, 20, 186, 268, 342, 381], "04": [16, 19, 20, 342, 404, 409, 437], "build": [17, 18, 340, 506], "__len__": [17, 92], "henc": 17, "parallel": [17, 173, 201, 251, 343, 352], "multiprocess": [17, 60], "worker": [17, 60], "imagenet_data": 17, "imagenet_root": 17, "nthread": 17, "almost": 17, "target_transform": [17, 25, 26, 28, 29, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41, 42, 43, 45, 46, 47, 48, 51, 52, 55, 56, 57, 59, 61, 65, 66, 67, 68, 71, 72, 73, 75, 76, 77, 79, 80, 81, 82, 83, 87, 89, 90, 91, 92, 93], "respect": [17, 18, 31, 33, 34, 36, 41, 42, 48, 51, 52, 54, 60, 72, 73, 75, 76, 77, 78, 82, 88, 123, 124, 147, 155, 156, 162, 170, 171, 173, 177, 179, 184, 186, 187, 194, 198, 201, 216, 217, 223, 230, 231, 239, 248, 249, 251, 256, 258, 264, 268, 269, 273, 283, 333, 508], "feature_extract": [18, 110, 111], "tap": 18, "intermedi": [18, 110], "applic": [18, 111, 162, 167, 239, 244], "comput": [18, 110, 121, 122, 124, 125, 126, 127, 128, 129, 130, 131, 133, 146, 148, 151, 161, 167, 174, 209, 230, 231, 235, 244, 249, 252, 255, 263, 275, 283, 337, 340, 385, 394, 395, 397, 442, 475, 476, 506, 508], "facial": 18, "network": [18, 121, 124, 139, 142, 143, 154, 186, 190, 220, 268, 274, 343, 346, 347, 349, 350, 352, 357, 358, 359, 360, 363, 373, 374, 375, 376, 377, 378, 379, 380, 402, 403, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 428, 429, 430, 434, 435, 436, 449, 450, 451, 452, 453, 454, 455, 456, 457, 460, 471, 472, 487, 490, 491, 499, 505], "mind": 18, "hierarchi": [18, 42, 110], "pyramid": [18, 121, 124], "head": [18, 148], "create_feature_extractor": [18, 111], "symbol": 18, "trace": [18, 110, 111], "graphic": 18, "represent": [18, 96, 121, 235, 243, 252], "graph": [18, 110, 111], "node": [18, 73, 110, 111], "redund": 18, "fx": [18, 110, 111], "explan": 18, "procedur": [18, 81], "inner": 18, "separ": [18, 110, 337, 462], "walk": [18, 110], "leaf": [18, 110, 111], "layer4": 18, "relu": [18, 110, 116, 117, 123, 130], "block": [18, 116, 117, 123, 130, 142, 143, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 471, 472, 506], "4th": [18, 33], "layer": [18, 116, 117, 119, 120, 121, 123, 340, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 457, 460, 461], "finer": [18, 42], "truncat": [18, 61], "shortcut": 18, "train_nod": [18, 111], "get_graph_node_nam": 18, "pertain": 18, "relu_2": 18, "repeat": [18, 111, 147, 170, 177, 216, 248, 256], "postfix": [18, 111], "disambigu": 18, "mayb": 18, "three": [18, 42, 54, 60, 77, 88, 94], "add_1": 18, "add_2": 18, "counter": [18, 111, 223], "within": [18, 30, 44, 48, 49, 50, 58, 84, 85, 96, 111, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "scope": [18, 331], "parent": [18, 110], "resid": 18, "might": [18, 54, 60, 78, 88, 161, 190, 204, 220, 237, 274, 507, 508], "maskrcnn": [18, 367, 368, 479], "resnet50": [18, 342], "mask_rcnn": [18, 367, 368, 479], "backbone_util": 18, "lastlevelmaxpool": 18, "feature_pyramid_network": 18, "featurepyramidnetwork": 18, "assist": 18, "design": [18, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 437, 438, 439, 440, 487, 493, 494], "extractor": 18, "eval_nod": [18, 111], "mode": [18, 32, 78, 95, 97, 98, 99, 103, 110, 111, 131, 154, 155, 156, 159, 170, 171, 175, 177, 178, 182, 185, 186, 188, 190, 192, 193, 194, 197, 198, 201, 202, 206, 209, 214, 216, 217, 219, 220, 221, 223, 224, 226, 227, 230, 231, 234, 248, 249, 253, 256, 257, 261, 266, 267, 268, 270, 271, 274, 276, 280, 282, 283, 321, 342, 363, 365, 366, 367, 369, 371, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 508], "control": [18, 162, 184, 239, 264, 341], "return_nod": [18, 110], "node_nam": 18, "layer1": [18, 110], "layer2": 18, "layer3": [18, 110], "pick": [18, 176, 238, 245, 255], "descend": 18, "tip": [18, 110, 508], "consult": 18, "confirm": 18, "whose": [18, 76, 275, 394, 395], "backbon": [18, 345, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "attach": 18, "fpn": [18, 121, 124, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370], "resnet50withfpn": 18, "bodi": 18, "k": [18, 54, 60, 88, 104, 110, 121, 148, 149, 150, 151, 152, 334, 366], "v": [18, 110, 121, 179, 207, 258, 300, 366], "dry": 18, "inp": 18, "randn": 18, "in_channels_list": [18, 121], "o": 18, "out_channel": [18, 116, 117, 118, 121, 139], "extra_block": [18, 121], "91": [18, 20, 24, 342, 357, 363, 389, 392, 399, 406, 416, 420, 426, 431, 434, 435, 436, 439, 454, 456], "acc": [19, 20, 24, 339, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "param": [19, 20, 21, 22, 23, 24, 137, 156, 173, 177, 179, 186, 231, 251, 256, 258, 268, 339, 342], "gip": [19, 342, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "recip": [19, 20, 21, 22, 23, 24, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283, 339, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "googlenet_quantizedweight": [19, 342, 396], "imagenet1k_fbgemm_v1": [19, 342, 396, 397, 400, 401, 402, 403, 404, 405, 406, 407], "69": [19, 20, 342, 374, 384, 396, 400, 405, 425, 438, 449, 451], "826": [19, 20, 342, 396, 413], "404": [19, 342, 396], "6m": [19, 20, 22, 339, 342], "link": [19, 20, 21, 22, 23, 24, 30, 50, 85, 339, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "inception_v3_quantizedweight": [19, 342, 397], "176": [19, 342, 397], "354": [19, 342, 397, 407], "27": [19, 20, 342, 378, 413], "2m": [19, 20, 22, 24, 339, 342], "71": [19, 20, 342, 352, 385, 388, 391, 397, 398, 452, 453], "mobilenet_v2_quantizedweight": [19, 342, 398], "imagenet1k_qnnpack_v1": [19, 342, 398, 399], "658": [19, 342, 398], "5m": [19, 20, 339, 342], "mobilenet_v3_large_quantizedweight": [19, 342, 399], "858": [19, 20, 342, 399, 401, 427], "resnext101_32x8d_quantizedweight": [19, 342, 402], "986": [19, 342, 402], "48": [19, 20, 342, 402, 418, 466], "8m": [19, 20, 22, 339, 342], "41": [19, 20, 22, 23, 342, 368, 370, 402, 412, 428, 467], "imagenet1k_fbgemm_v2": [19, 342, 401, 402], "82": [19, 20, 339, 342, 356, 376, 383, 402, 403, 408, 416, 422, 424, 428, 441, 448, 464, 471], "resnext101_64x4d_quantizedweight": [19, 342, 403], "898": [19, 342, 403], "326": [19, 342, 403], "resnet18_quantizedweight": [19, 342, 400], "494": [19, 342, 400], "882": [19, 342, 400], "7m": [19, 20, 22, 339, 342], "resnet50_quantizedweight": [19, 342, 401], "814": [19, 342, 401], "09": [19, 20, 24, 342, 375, 401, 427, 436], "80": [19, 20, 339, 342, 375, 401, 408, 410, 413, 416, 417, 418, 422, 423, 427, 441, 442, 459], "282": [19, 342, 401], "976": [19, 20, 342, 354, 401, 416], "shufflenet_v2_x0_5_quantizedweight": [19, 342, 404], "972": [19, 20, 342, 357, 404, 419, 470], "79": [19, 20, 339, 342, 352, 374, 402, 404, 409, 414, 428, 463, 464, 469], "4m": [19, 20, 22, 23, 339, 342], "shufflenet_v2_x1_0_quantizedweight": [19, 342, 405], "582": [19, 20, 339, 342, 405, 443, 458], "3m": [19, 20, 22, 24, 339, 342], "shufflenet_v2_x1_5_quantizedweight": [19, 342, 406], "72": [19, 20, 22, 24, 342, 361, 391, 398, 406, 412, 433, 435, 439, 455, 468], "052": [19, 342, 406], "shufflenet_v2_x2_0_quantizedweight": [19, 342, 407], "488": [19, 342, 407], "gflop": [20, 21, 22, 23, 24, 339, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "alexnet_weight": [20, 342, 352], "imagenet1k_v1": [20, 114, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 365, 366, 367, 369, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 469, 470, 471, 472], "56": [20, 110, 342, 352, 358, 381, 386, 466], "522": [20, 342, 352, 390, 413], "066": [20, 342, 352], "61": [20, 21, 24, 342, 366, 417, 449, 450, 469], "1m": [20, 21, 342], "convnext_base_weight": [20, 342, 353], "062": [20, 342, 353], "convnext_large_weight": [20, 342, 354], "414": [20, 342, 354], "convnext_small_weight": [20, 342, 355], "616": [20, 342, 355], "65": [20, 21, 342, 355, 366, 466], "convnext_tiny_weight": [20, 342, 356], "146": [20, 342, 356, 370, 469], "28": [20, 339, 342, 407, 418, 440], "densenet121_weight": [20, 342, 357], "434": [20, 342, 357, 401, 427], "0m": [20, 22, 24, 339, 342], "densenet161_weight": [20, 342, 358], "densenet169_weight": [20, 342, 359], "806": [20, 342, 359], "densenet201_weight": [20, 342, 360], "896": [20, 342, 360], "efficientnet_b0_weight": [20, 342, 373], "692": [20, 342, 373], "532": [20, 342, 373], "39": [20, 22, 342, 365, 373, 377], "efficientnet_b1_weight": [20, 342, 374], "642": [20, 342, 374], "imagenet1k_v2": [20, 342, 374, 391, 392, 398, 399, 401, 402, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 427, 428, 430, 471, 472], "934": [20, 342, 374], "efficientnet_b2_weight": [20, 342, 375], "608": [20, 342, 375], "efficientnet_b3_weight": [20, 342, 376], "054": [20, 342, 376, 416], "efficientnet_b4_weight": [20, 342, 377], "594": [20, 342, 377], "efficientnet_b5_weight": [20, 342, 378], "444": [20, 77, 342, 378, 417], "628": [20, 342, 378, 449], "efficientnet_b6_weight": [20, 342, 379], "916": [20, 342, 379], "43": [20, 22, 339, 342, 411, 417, 443, 457, 465], "07": [20, 137, 138, 140, 141, 145, 342, 379, 470], "efficientnet_b7_weight": [20, 342, 380], "122": [20, 342, 380], "908": [20, 342, 380], "efficientnet_v2_l_weight": [20, 342, 381], "85": [20, 342, 381, 382, 466, 468, 469], "808": [20, 342, 381], "788": [20, 342, 381], "efficientnet_v2_m_weight": [20, 342, 382], "156": [20, 342, 382], "54": [20, 21, 22, 24, 342, 359, 366, 369], "efficientnet_v2_s_weight": [20, 342, 383], "228": [20, 342, 383, 402, 415, 428], "878": [20, 342, 383, 391, 398, 418], "googlenet_weight": [20, 342, 384, 396], "778": [20, 342, 384, 396], "53": [20, 342, 384, 390, 396], "inception_v3_weight": [20, 342, 385, 397], "294": [20, 342, 385, 397], "mnasnet0_5_weight": [20, 342, 387], "mnasnet0_75_weight": [20, 342, 388], "496": [20, 342, 388], "mnasnet1_0_weight": [20, 342, 389], "mnasnet1_3_weight": [20, 342, 390], "506": [20, 342, 390, 449, 450], "maxvit_t_weight": [20, 342, 386], "722": [20, 342, 386], "9m": [20, 342], "mobilenet_v2_weight": [20, 342, 391, 398], "154": [20, 342, 391, 398], "822": [20, 342, 391, 398], "mobilenet_v3_large_weight": [20, 342, 361, 362, 372, 392, 399, 431, 436], "042": [20, 342, 392, 399], "274": [20, 342, 392, 399], "566": [20, 342, 392, 399], "mobilenet_v3_small_weight": [20, 342, 393], "668": [20, 342, 393, 409], "06": [20, 119, 120, 142, 143, 342, 393], "regnet_x_16gf_weight": [20, 342, 408], "058": [20, 342, 408], "944": [20, 342, 408], "716": [20, 342, 408, 420], "196": [20, 87, 342, 408, 411, 444], "regnet_x_1_6gf_weight": [20, 342, 409], "922": [20, 342, 409], "regnet_x_32gf_weight": [20, 342, 410], "622": [20, 342, 410, 418], "248": [20, 342, 410], "107": [20, 342], "014": [20, 342, 410], "regnet_x_3_2gf_weight": [20, 342, 411], "364": [20, 342, 411, 463], "992": [20, 342, 411], "regnet_x_400mf_weight": [20, 342, 412], "834": [20, 342, 402, 412, 428], "864": [20, 342, 412, 446], "322": [20, 342, 412], "regnet_x_800mf_weight": [20, 342, 413], "212": [20, 342, 413], "348": [20, 342, 413], "regnet_x_8gf_weight": [20, 342, 414], "686": [20, 342, 414], "682": [20, 342, 414, 415], "678": [20, 342, 414], "regnet_y_128gf_weight": [20, 342, 415], "imagenet1k_swag_e2e_v1": [20, 342, 415, 416, 418, 466, 468, 469], "98": [20, 339, 342, 415, 416, 418, 462, 468, 469], "644": [20, 342], "imagenet1k_swag_linear_v1": [20, 342, 415, 416, 418, 466, 468, 469], "86": [20, 22, 339, 342, 371, 402, 415, 416, 418, 460], "068": [20, 342, 415], "844": [20, 342, 415], "127": [20, 239, 248, 251, 256, 264, 269, 273, 342, 415, 461], "regnet_y_16gf_weight": [20, 342, 416], "424": [20, 342, 416], "886": [20, 342, 416, 423, 466], "012": [20, 342, 416], "244": [20, 342, 416], "regnet_y_1_6gf_weight": [20, 342, 417], "966": [20, 342, 417], "876": [20, 342, 417, 455], "regnet_y_32gf_weight": [20, 342, 418], "145": [20, 342], "362": [20, 342, 405, 418, 438], "regnet_y_3_2gf_weight": [20, 342, 419], "948": [20, 342, 419], "576": [20, 342, 419], "982": [20, 342, 419], "regnet_y_400mf_weight": [20, 342, 420], "046": [20, 342, 420, 424], "804": [20, 342, 420], "742": [20, 342, 420], "regnet_y_800mf_weight": [20, 342, 421], "828": [20, 342, 421, 422], "502": [20, 342, 421], "regnet_y_8gf_weight": [20, 342, 422], "032": [20, 342, 422], "048": [20, 342, 422], "resnext101_32x8d_weight": [20, 342, 402, 428], "312": [20, 342, 402, 424, 428], "526": [20, 342, 402, 428], "resnext101_64x4d_weight": [20, 342, 403, 429], "454": [20, 342, 381, 403, 429], "resnext50_32x4d_weight": [20, 342, 430], "618": [20, 342, 430], "698": [20, 342, 430], "resnet101_weight": [20, 342, 423, 432, 434], "resnet152_weight": [20, 342, 424], "60": [20, 24, 342, 404, 409, 431, 435, 437, 458], "284": [20, 342, 424, 471], "002": [20, 342, 424], "758": [20, 342, 400, 425, 472], "078": [20, 342, 400, 425], "resnet34_weight": [20, 342, 426], "314": [20, 342, 426], "resnet50_weight": [20, 114, 342, 363, 364, 365, 366, 367, 368, 369, 370, 401, 427, 433, 435], "862": [20, 342, 401, 427], "shufflenet_v2_x0_5_weight": [20, 342, 404, 437], "552": [20, 342, 404, 437, 468], "746": [20, 342, 404, 437], "shufflenet_v2_x1_0_weight": [20, 342, 405, 438], "316": [20, 342, 405, 438], "shufflenet_v2_x1_5_weight": [20, 342, 406, 439], "996": [20, 342, 406, 439], "086": [20, 342, 406, 439, 472], "shufflenet_v2_x2_0_weight": [20, 342, 407, 440], "squeezenet1_0_weight": [20, 342, 441], "092": [20, 342, 441], "squeezenet1_1_weight": [20, 342, 442], "624": [20, 342, 442], "swin_b_weight": [20, 342, 443], "swin_s_weight": [20, 342, 444], "swin_t_weight": [20, 342, 445], "474": [20, 342, 445], "776": [20, 342, 445], "swin_v2_b_weight": [20, 342, 446], "swin_v2_s_weight": [20, 342, 447], "712": [20, 342, 447], "816": [20, 342, 447], "swin_v2_t_weight": [20, 342, 448], "072": [20, 342, 448, 466], "vgg11_bn_weight": [20, 342, 450], "vgg11_weight": [20, 342, 449], "vgg13_bn_weight": [20, 342, 452], "586": [20, 342, 452], "133": [20, 21, 342, 366], "vgg13_weight": [20, 342, 451], "928": [20, 342, 451], "vgg16_bn_weight": [20, 342, 454], "516": [20, 342, 454], "vgg16_weight": [20, 342, 371, 453], "592": [20, 342, 453], "382": [20, 342, 453], "imagenet1k_featur": [20, 342, 371, 453], "nan": [20, 342, 453], "vgg19_bn_weight": [20, 342, 456], "842": [20, 342, 456], "143": [20, 342], "vgg19_weight": [20, 342, 455], "376": [20, 342, 455], "vit_b_16_weight": [20, 342, 466], "318": [20, 342, 466], "vit_b_32_weight": [20, 342, 467], "912": [20, 77, 342, 467], "466": [20, 342, 467], "vit_h_14_weight": [20, 342, 468], "694": [20, 342, 468], "633": [20, 342], "1016": [20, 342, 468], "708": [20, 342, 468], "632": [20, 342], "167": [20, 342, 364, 468], "vit_l_16_weight": [20, 342, 469], "662": [20, 342, 469], "638": [20, 342, 469], "305": [20, 342], "361": [20, 342, 469], "422": [20, 342, 469], "vit_l_32_weight": [20, 342, 470], "306": [20, 342], "wide_resnet101_2_weight": [20, 342, 471], "848": [20, 342, 471], "126": [20, 342], "wide_resnet50_2_weight": [20, 342, 472], "468": [20, 342, 472], "602": [20, 342, 472], "keypoint": [21, 334, 340, 366], "coco_legaci": [21, 342, 366], "59": [21, 342], "coco_v1": [21, 22, 23, 342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372], "fcos_resnet50_fpn_weight": [22, 342, 365], "fasterrcnn_mobilenet_v3_large_320_fpn_weight": [22, 342, 361], "fasterrcnn_mobilenet_v3_large_fpn_weight": [22, 342, 362], "fasterrcnn_resnet50_fpn_v2_weight": [22, 342, 364], "280": [22, 342, 364], "retinanet_resnet50_fpn_v2_weight": [22, 342, 370], "152": [22, 24, 342, 370, 424, 435], "retinanet_resnet50_fpn_weight": [22, 342, 369], "ssd300_vgg16_weight": [22, 342, 371], "ssdlite320_mobilenet_v3_large_weight": [22, 342, 372], "maskrcnn_resnet50_fpn_v2_weight": [23, 342, 368], "333": [23, 342, 368], "iou": [24, 132, 135, 137, 138, 140, 141, 144, 145, 147, 262, 342], "pixelwis": [24, 342], "deeplabv3_mobilenet_v3_large_weight": [24, 342, 431], "coco_with_voc_labels_v1": [24, 342, 431, 432, 433, 434, 435, 436], "deeplabv3_resnet101_weight": [24, 342, 432], "258": [24, 48, 342, 432], "deeplabv3_resnet50_weight": [24, 342, 433], "fcn_resnet101_weight": [24, 342, 434], "lraspp_mobilenet_v3_large_weight": [24, 342, 436], "option": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 96, 97, 98, 99, 103, 104, 105, 109, 110, 111, 115, 116, 117, 118, 121, 123, 124, 130, 137, 138, 139, 140, 141, 145, 153, 155, 156, 159, 168, 171, 173, 177, 179, 186, 187, 190, 192, 194, 201, 207, 209, 215, 217, 220, 221, 223, 227, 230, 231, 234, 238, 239, 245, 246, 248, 249, 251, 255, 256, 258, 260, 262, 264, 265, 267, 268, 269, 270, 272, 273, 274, 275, 276, 277, 278, 280, 283, 291, 294, 295, 298, 302, 306, 308, 312, 313, 315, 321, 326, 327, 328, 330, 333, 334, 335, 337, 338, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 508], "directori": [25, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 333, 342], "cifar": 25, "otherwis": [25, 32, 38, 45, 54, 59, 68, 71, 88, 89, 96, 116, 117, 123, 186, 190, 220, 221, 267, 268, 270, 274, 276, 326, 327, 328, 330, 332], "internet": [25, 28, 29, 31, 35, 36, 38, 40, 42, 45, 48, 51, 52, 55, 59, 61, 65, 66, 68, 70, 71, 72, 73, 74, 76, 77, 78, 79, 80, 81, 82, 83, 87, 89, 90, 91, 93], "again": [25, 28, 29, 31, 35, 36, 38, 40, 42, 45, 48, 51, 52, 55, 59, 61, 65, 66, 68, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 89, 90, 91, 93], "special": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 94, 508], "member": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93], "synthet": 27, "offici": [27, 60, 442], "img1_left": 27, "img1_right": 27, "disp": [27, 69], "img2_left": 27, "img2_right": 27, "shapenet": 27, "reflect": [27, 170, 177, 216, 248, 256], "hole": 27, "ndarrai": [27, 30, 39, 44, 49, 50, 53, 58, 62, 63, 64, 69, 84, 85, 86, 192, 193, 227, 228, 279, 280, 282, 320, 321], "img_left": [27, 30, 39, 44, 58, 62, 63, 69, 84, 86], "img_right": [27, 30, 39, 44, 58, 62, 63, 69, 84, 86], "dispar": [27, 30, 39, 44, 58, 62, 63, 69, 84, 86], "valid_mask": [27, 30, 39, 44, 58, 62, 63, 69, 84, 86], "implicitli": [27, 39, 62, 63, 69, 508], "valid": [27, 31, 32, 35, 37, 39, 49, 50, 53, 56, 62, 63, 64, 77, 85, 307, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 453, 508], "target_typ": [28, 31, 32, 55, 72, 94], "union": [28, 31, 32, 49, 50, 53, 55, 64, 67, 72, 74, 85, 104, 110, 113, 115, 116, 117, 124, 128, 129, 135, 137, 138, 140, 141, 144, 145, 147, 151, 152, 159, 186, 216, 220, 221, 230, 231, 232, 234, 236, 239, 240, 241, 248, 249, 251, 252, 256, 264, 267, 268, 269, 270, 273, 274, 275, 276, 277, 278, 283, 291, 298, 306, 308, 312, 313, 315, 320, 326, 327, 328, 329, 330, 333, 334, 335, 337, 338, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "caltech": [28, 29], "scipi": [28, 48, 78, 83, 87], "mat": [28, 48, 78, 83, 87], "hand": 28, "outlin": 28, "carla": 30, "simul": 30, "crestereo": 30, "highr": 30, "trainingf": 30, "scene1": [30, 39, 44, 58, 69, 84, 86], "img0": 30, "disp0gt": [30, 39], "pfm": [30, 39, 69, 84], "disp1gt": 30, "calib": [30, 39, 62, 63, 69], "txt": [30, 39, 49, 69], "scene2": [30, 39, 44, 58, 69, 84, 86], "attr": 31, "larg": [31, 48, 187, 223, 239, 269, 276, 327, 354, 361, 362, 372, 392, 399, 431, 436, 449, 450, 451, 452, 453, 454, 455, 456, 499, 507], "celebfac": 31, "attribut": [31, 341, 342], "ident": [31, 65, 66, 162, 239], "landmark": 31, "binari": [31, 73, 76, 153, 340, 367], "lefteye_x": 31, "lefteye_i": 31, "righteye_x": 31, "righteye_i": 31, "nose_x": 31, "nose_i": 31, "leftmouth_x": 31, "leftmouth_i": 31, "rightmouth_x": 31, "rightmouth_i": 31, "fine": [32, 94, 96, 394, 415, 416, 418, 466, 468, 469], "leftimg8bit": 32, "gtfine": 32, "gtcoars": 32, "val": [32, 36, 42, 48, 49, 57, 60, 67, 73, 75, 77, 78, 90, 91, 93], "train_extra": 32, "qualiti": [32, 100, 107], "coars": 32, "polygon": 32, "smnt": 32, "inst": 32, "col": 32, "poli": 32, "annfil": [33, 34], "ms": [33, 34], "caption": [33, 46, 47, 79], "instal": [33, 34, 67, 73, 342, 507], "dset": 33, "cap": 33, "82783": 33, "3l": 33, "427l": 33, "640l": 33, "u": 33, "plane": 33, "emit": [33, 342], "smoke": 33, "fly": 33, "mountain": 33, "dart": 33, "sky": 33, "behind": [33, 340], "snow": 33, "leav": [33, 94, 269], "contrail": 33, "snowi": 33, "overheard": 33, "distanc": [33, 138, 139, 140, 141], "view": [33, 39, 69, 74, 127, 163, 191], "plume": 33, "openai": 35, "yfcc100m": 35, "gp": 35, "iso": 35, "3166": 35, "countri": 35, "balanc": [35, 38, 153], "class_index": [35, 40, 43, 57], "partit": 36, "textur": 36, "thu": [36, 51, 94, 159, 198, 234, 265, 269, 508], "regardless": [36, 98], "is_valid_fil": [37, 56], "find_class": 37, "corrupt": [37, 56, 75], "class_x": 37, "xxx": [37, 56], "ext": 37, "xxy": [37, 56], "xxz": [37, 56], "class_i": 37, "nsdf3": [37, 56], "asd932_": [37, 56], "overridden": [37, 56, 122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "consid": [37, 54, 60, 88, 138, 222, 237], "subset": [37, 342, 431, 432, 433, 434, 435, 436], "filenotfounderror": 37, "static": [37, 156, 159, 164, 173, 177, 179, 184, 186, 187, 231, 234, 241, 251, 256, 258, 264, 268, 269], "path_to_sampl": 37, "compress": [37, 93, 101, 108], "found": [37, 333, 342, 508], "idx3": [38, 45, 59, 68], "ubyt": [38, 45, 59, 68], "t10k": [38, 45, 59, 68], "byclass": 38, "bymerg": 38, "letter": 38, "digit": [38, 76, 83], "mnist": [38, 45, 59, 76], "eth3d": 39, "two_view_train": 39, "im1": [39, 69], "im0": 39, "camera": [39, 50], "two_view_training_gt": 39, "mask0nocc": 39, "two_view_test": 39, "scene": [39, 62, 63, 69, 82, 84], "trainval": [42, 72, 90, 91], "annotation_level": 42, "variant": [42, 44, 84, 342, 423, 424, 427, 488], "fgvc": 42, "aircraft": 42, "000": [42, 51], "airplan": 42, "coarser": 42, "boe": 42, "737": 42, "700": [42, 60], "collaps": 42, "indistinguish": 42, "compris": 42, "famili": [42, 55, 342], "manufactur": 42, "image_s": [43, 124], "random_offset": 43, "randomli": [43, 119, 120, 142, 143, 154, 159, 162, 164, 172, 173, 174, 175, 176, 178, 179, 180, 181, 182, 184, 185, 186, 188, 189, 234, 239, 241, 251, 252, 254, 255, 258, 259, 265, 267, 270, 273, 508], "offset": [43, 118, 139, 162, 177, 239, 256], "fallingth": 44, "dir1": 44, "_object_set": 44, "_camera_set": 44, "image1": [44, 65], "mix": [44, 457, 508], "ann_fil": [46, 47], "entiti": [46, 47], "oxford": [48, 72], "102": 48, "flower": 48, "occur": 48, "kingdom": [48, 55], "light": [48, 148], "variat": 48, "sever": [48, 155, 160, 230, 235], "flyingchairs_train_v": 49, "page": [49, 508], "00001_flow": 49, "flo": 49, "00001_img1": 49, "ppm": 49, "00001_img2": 49, "valid_flow_mask": [49, 50, 53, 64, 85], "kittiflow": [49, 50, 85, 394], "pass_nam": [50, 84, 85, 86], "clean": [50, 51, 84, 85, 86, 394], "frames_cleanpass": [50, 84], "frames_finalpass": [50, 84], "intel": [50, 340], "food": 51, "challeng": [51, 75], "manual": 51, "review": 51, "750": 51, "On": [51, 342, 507], "amount": [51, 198, 337], "nois": 51, "come": [51, 54, 60, 88], "mostli": [51, 69, 510], "intens": [51, 197, 198], "sometim": [51, 340, 394, 395], "maximum": [51, 132, 147, 161, 164, 173, 190, 204, 220, 237, 241, 251, 262, 267, 270, 273, 274, 276], "side": [51, 116, 117, 150, 170, 177, 216, 248, 256], "german": 52, "traffic": 52, "sign": 52, "benchmark": [52, 62, 63, 364, 368], "hd1k_challeng": 53, "image_2": [53, 61, 63, 64], "hd1k_flow_gt": 53, "flow_occ": [53, 64], "hd1k_input": 53, "annotation_path": [54, 88], "frames_per_clip": [54, 60, 88], "step_between_clip": [54, 60, 88], "frame_r": [54, 60, 88, 457, 458, 459, 460, 461, 462, 463, 464, 465], "fold": [54, 81, 88], "_precomputed_metadata": [54, 60, 88], "_video_width": [54, 60, 88], "_video_height": [54, 60, 88], "_video_min_dimens": [54, 60, 88], "_audio_sampl": [54, 60, 88], "thwc": [54, 60, 88, 94, 104], "collect": [54, 60, 88, 94], "videoclip": [54, 60, 88], "txhxwxc": [54, 60, 88], "l": [54, 60, 88, 104, 148, 149, 151, 152, 155, 156, 171, 175, 178, 182, 185, 188, 193, 194, 202, 206, 214, 219, 224, 230, 231, 249, 253, 257, 261, 266, 271, 282, 283, 381], "2021_train": 55, "store": [55, 83, 89, 98, 153, 161, 204, 237], "2017": 55, "2018": 55, "2019": 55, "2021_train_mini": 55, "2021_valid": 55, "2021": 55, "speci": 55, "animalia": 55, "phylum": 55, "arthropoda": 55, "insecta": 55, "coleoptera": 55, "clerida": 55, "genu": 55, "trichod": 55, "numer": [55, 122, 142, 143, 269], "amphibian": 55, "category_nam": [55, 342], "category_typ": 55, "default_load": [56, 75], "arrang": 56, "datasetfold": 56, "2012": [57, 62, 78, 90, 91], "ilsvrc2012_devkit_t12": 57, "tar": [57, 79], "gz": [57, 79], "ilsvrc2012_img_train": 57, "ilsvrc2012_img_v": 57, "left_disp": 58, "right_disp": 58, "kuzushiji": 59, "num_download_work": 60, "_audio_channel": 60, "_legaci": 60, "600": [60, 363, 380], "class1": 60, "clip1": 60, "clip2": 60, "clip3": 60, "class2": 60, "clipx": 60, "omit": [60, 94, 187, 223, 269, 326, 327, 328, 330, 338, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "speed": [60, 502], "runtimeerror": [60, 75, 161, 204, 237], "archiv": [60, 75], "extract": [60, 73, 75, 110, 111, 115, 340, 453], "label_2": 61, "occlud": 61, "rotation_i": 61, "kitti": [62, 63, 64, 94, 394, 395], "stereo": [62, 63, 69, 74, 86], "evalu": [62, 63, 71, 77, 81, 342, 363, 365, 366, 367, 369, 371], "2015": [62, 63, 64], "kitti2012": 62, "colored_0": 62, "1_10": 62, "2_10": 62, "colored_1": 62, "disp_noc": 62, "kitti2015": 63, "image_3": 63, "disp_occ_0": 63, "disp_occ_1": 63, "10fold": [65, 66], "image_set": [65, 66, 78, 90, 91], "funnel": [65, 66], "lfw": [65, 66], "deepfunnel": [65, 66], "randomrot": [65, 66, 315], "indent": 65, "lmdb": 67, "pip": [67, 73], "databas": 67, "bedroom_train": 67, "church_outdoor_train": 67, "calibr": 69, "perfect": 69, "use_ambient_view": 69, "publicli": 69, "middleburi": 69, "2014": 69, "edu": [69, 74], "scenes2014": 69, "ambient": 69, "subdirectori": 69, "middlebury2014": 69, "imperfect": 69, "im": [69, 342], "im1l": 69, "sd": 69, "middleburri": 69, "expos": 69, "lightn": 69, "uncalibr": 69, "split_ratio": 70, "mnist_test_seq": 70, "npy": 70, "ratio": [70, 124, 138, 179, 186, 220, 258, 262, 268], "terminolog": 71, "charact": [71, 77], "iiit": 72, "pet": 72, "trimap": 72, "patchcamelyon": 73, "680": 73, "96px": 73, "histopatholog": 73, "scan": 73, "lymph": 73, "presenc": 73, "metastat": 73, "tissu": 73, "h5py": 73, "newer": 74, "state": [74, 194, 283], "suitabl": [74, 96], "harri": 74, "patch": 74, "centr": [74, 134], "real": [74, 197, 349, 363], "rather": [74, 110, 337, 508], "project": [74, 340], "3d": [74, 462], "old": [74, 342], "cs": 74, "washington": 74, "htm": 74, "data1": 74, "data2": 74, "standard": [75, 81, 164, 168, 209, 215, 241, 246, 342, 453], "high": [75, 341, 362], "resolut": [75, 121, 342, 361, 362, 446, 447, 448, 498], "compon": [75, 205, 221], "devkit": 75, "subdir": 76, "test10k": 76, "test50k": 76, "nist": 76, "60k": 76, "10k": 76, "50k": 76, "remain": 76, "accord": [76, 94, 147, 190, 220, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 307], "sai": 76, "vector": [76, 162, 167, 239, 244], "inform": [76, 97, 98, 99, 103, 110, 235, 243, 252, 340, 341, 508], "decid": 76, "sst2": 77, "capabl": 77, "sentenc": 77, "standford": 77, "sentiment": 77, "treebank": 77, "neg": [77, 153, 155, 159, 172, 195, 196, 197, 198, 200, 230, 234, 250, 265], "divid": 77, "6920": 77, "3610": 77, "3310": 77, "872": 77, "428": 77, "1821": 77, "909": 77, "boundari": 78, "sbd": 78, "11355": 78, "taken": [78, 148, 149, 151, 152, 326, 327, 328, 330], "pascal": [78, 90, 91, 342, 431, 432, 433, 434, 435, 436], "voc": [78, 90, 91, 342, 431, 432, 433, 434, 435, 436], "2011": 78, "voc2012": 78, "train_nov": 78, "exclud": [78, 115], "photo": 79, "tarbal": 79, "sbucaptionedphotodataset": 79, "stl10_binari": 81, "unlabel": 81, "1k": [81, 342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472], "5k": 81, "sun": 82, "397": [82, 457, 458, 459, 460, 461, 462, 463, 464, 465], "108": [82, 445, 448], "754": [82, 354], "assign": [83, 332], "flyingthings3d": [84, 394, 395], "interfac": 84, "monkaa": 84, "drive": 84, "sceneflow": 84, "scene_1": 85, "scene_2": 85, "sintel": [86, 394, 395], "final_left": 86, "final_right": 86, "occlus": 86, "outoffram": 86, "whilst": [86, 341], "stanford": 87, "car": 87, "144": 87, "041": 87, "pil_imag": 87, "class_id": [87, 342], "websit": 88, "docstr": 88, "instruct": 88, "num_lin": 89, "li": 89, "255": [89, 97, 98, 99, 103, 193, 197, 206, 282, 333, 334, 335, 374, 508], "bz2": 89, "year": [90, 91], "2007": [90, 91], "alphabet": 90, "xml": 90, "mutual": 92, "exclus": 92, "wider_face_split": 93, "wider_train": 93, "wider_v": 93, "wider_test": 93, "beta": [94, 96, 98, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 286, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 308, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 340, 341, 342, 345, 349, 350, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 432, 433, 434, 435, 436, 457, 458, 459, 460, 461, 462, 477, 478, 479, 492, 496, 497, 501, 503, 508], "usag": [94, 238, 245, 311, 508], "stage": [94, 96, 98, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 340, 341, 342, 345, 349, 350, 351, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 432, 433, 434, 435, 436, 457, 458, 459, 460, 461, 462, 477, 478, 479, 492, 496, 497, 501, 503], "disrupt": [94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 508], "submit": [94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 508], "6753": [94, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284], "furthermor": 94, "wrapper": 94, "configur": [94, 112, 116, 117, 273], "encount": 94, "error": [94, 110, 161, 204, 237, 239, 394, 395], "prompt": 94, "descript": 94, "vocdetect": 94, "celeba": 94, "oxfordiiitpet": 94, "cityscap": 94, "widerfac": 94, "vocsegment": 94, "grain": [94, 96], "unchang": [95, 97, 98, 99, 103, 180], "grayscal": [95, 97, 98, 99, 103, 108, 180, 206, 214, 222, 224, 226, 259, 314, 319], "gray_alpha": 95, "transpar": [95, 159, 197, 198, 234, 335], "rgb_alpha": 95, "src": 96, "num_thread": 96, "pyav": [96, 109, 340], "2s": 96, "path_to_a_test_video": 96, "reader": [96, 341], "conjunct": 96, "insid": [96, 136], "addition": 96, "fluent": 96, "byte": [96, 97, 98, 99, 100, 101, 102], "media": 96, "thread": [96, 341], "codec": [96, 109], "multithread": 96, "heurist": [96, 124, 275], "time_": 96, "keyframes_onli": 96, "keyfram": 96, "precis": [96, 269], "exact": [96, 147, 239], "success": 96, "imagereadmod": [97, 98, 99, 103], "jpeg": [97, 98, 100, 103, 107], "image_channel": [97, 98, 99, 103], "image_height": [97, 98, 99, 100, 101, 103, 107, 108], "image_width": [97, 98, 99, 100, 101, 103, 107, 108], "nvjpeg": 98, "leak": 98, "chw": [100, 101, 107, 108], "layout": [100, 101, 107, 108], "content": [100, 101, 102, 106], "int8": [100, 101, 107, 108, 342], "compression_level": [101, 108], "factor": [101, 108, 124, 138, 148, 149, 151, 152, 153, 172, 173, 195, 196, 199, 200, 250, 251, 262, 273, 508], "filenam": [104, 105, 106, 107, 108, 109, 333, 338], "start_pt": 104, "fraction": [104, 105, 173, 251], "end_pt": 104, "afram": 104, "video_fp": [104, 105], "audio_fp": [104, 109], "vframe": 104, "written": [106, 394, 395], "hw": 108, "video_arrai": 109, "video_codec": 109, "libx264": 109, "audio_arrai": 109, "audio_codec": 109, "audio_opt": 109, "4d": [109, 163, 191, 337], "h264": 109, "44100": 109, "mp3": 109, "train_return_nod": 110, "eval_return_nod": 110, "tracer_kwarg": [110, 111], "suppress_diff_warn": [110, 111], "graphmodul": 110, "unus": [110, 431, 432, 433], "subhead": [110, 111], "Not": [110, 111], "traceabl": 110, "massag": 110, "cooper": 110, "exhaust": 110, "problemat": 110, "leaf_modul": [110, 111], "hold": [110, 187, 223, 269], "likewis": [110, 148, 151], "autowrap_funct": 110, "inbuilt": 110, "dure": [110, 363, 365, 366, 367, 369, 371], "partial": 110, "keyword": [110, 111, 124], "nodepathtrac": [110, 111], "onto": [110, 111], "tracer": [110, 111], "autowrap_modul": [110, 111], "math": [110, 111], "_get_leaf_modules_for_op": [110, 111], "discrep": [110, 111], "feat1": [110, 124], "feat2": [110, 121, 124], "leaf_funct": 110, "typeerror": 110, "leafmodul": 110, "mymodul": 110, "conv": [110, 116, 117], "conv2d": 110, "submodul": 111, "fall": 111, "twice": [111, 471, 472], "eventu": 111, "initi": 112, "weightsenum": [113, 114], "enum": [113, 114, 155, 156, 162, 171, 173, 184, 186, 187, 190, 194, 201, 217, 220, 221, 223, 230, 231, 236, 239, 249, 251, 264, 267, 268, 269, 270, 274, 276, 283, 342], "weights_enum": [113, 342], "moduletyp": 115, "fnmatch": 115, "unix": 115, "shell": 115, "style": 115, "wildcard": 115, "include_filt": 115, "in_channel": [116, 117, 118, 123, 139], "stride": [116, 117, 118, 139, 423, 424, 427, 488, 508], "pad": [116, 117, 118, 139, 158, 177, 203, 205, 232, 256, 273, 337], "group": [116, 117, 118, 139], "norm_lay": [116, 117, 121, 123, 372], "batchnorm": [116, 117], "batchnorm2d": [116, 122], "activation_lay": [116, 117, 123], "dilat": [116, 117, 118, 139], "bia": [116, 117, 118, 123, 139], "convolution2d": 116, "convolut": [116, 117, 118, 139, 142, 143, 343, 345, 346, 347, 350, 351, 352, 357, 358, 359, 360, 365, 373, 374, 375, 376, 377, 378, 379, 380, 384, 396, 423, 424, 427, 431, 432, 433, 434, 435, 449, 450, 451, 452, 453, 454, 455, 456, 457, 460, 461, 471, 472, 473, 474, 488, 499, 501], "convolv": [116, 117], "kernel": [116, 117, 118, 139, 164, 209, 241, 311, 340, 509], "calcul": [116, 117, 164, 241], "norm": [116, 117, 123, 394, 395], "space": [116, 117, 139, 159, 171, 192, 198, 222, 227, 234, 249, 265, 280, 321, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 487], "bias": [116, 117], "batchnorm3d": 117, "convolution3d": 117, "deform_conv2d": 118, "in_height": [118, 139], "in_width": [118, 139], "offset_group": [118, 139], "kernel_height": [118, 139], "kernel_width": [118, 139], "out_height": [118, 139], "out_width": [118, 139], "block_siz": [119, 120, 142, 143], "ep": [119, 120, 122, 137, 138, 140, 141, 142, 143, 145, 394, 395], "1e": [119, 120, 122, 137, 138, 140, 141, 142, 143, 145], "drop_block2d": 119, "area": [119, 120, 133, 138, 155, 156, 162, 171, 173, 179, 184, 186, 187, 194, 201, 217, 223, 230, 231, 249, 258, 268, 283], "dropblock": [119, 120, 142, 143], "drop_block3d": 120, "extrafpnblock": 121, "suppos": [121, 167, 244], "increas": [121, 172, 195, 196, 200, 250], "ordereddict": [121, 124], "dummi": 121, "feat0": 121, "feat3": [121, 124], "highest": 121, "get_result_from_inner_block": 121, "inner_block": 121, "get_result_from_layer_block": 121, "layer_block": 121, "num_featur": 122, "denomin": [122, 142, 143], "stabil": [122, 142, 143], "afterward": [122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "former": [122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "latter": [122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "silent": [122, 125, 126, 127, 128, 129, 130, 131, 161, 174, 230, 231, 235, 249, 252, 255, 263, 275, 283], "hidden_channel": 123, "dropout": 123, "perceptron": 123, "hidden": 123, "linear": [123, 206, 372, 391, 398, 415, 416, 418, 466, 468, 469, 482, 483, 497], "featmap_nam": 124, "output_s": [124, 125, 126, 128, 129, 148, 149, 151, 152, 177, 203, 256, 293], "sampling_ratio": [124, 125, 128, 148, 151], "canonical_scal": 124, "canonical_level": 124, "roialign": 124, "pool": [124, 148, 149, 151, 152], "eq": [124, 130], "k0": 124, "region": [124, 148, 149, 151, 152, 179, 197, 207, 258, 349, 363], "levelmapp": 124, "image_shap": 124, "x1": [124, 132, 133, 134, 135, 136, 137, 138, 140, 141, 144, 145, 146, 147, 148, 149, 150, 151, 152, 275, 363, 365, 366, 367, 369, 371], "y1": [124, 132, 133, 134, 135, 136, 137, 138, 140, 141, 144, 145, 146, 147, 148, 149, 150, 151, 152, 363, 365, 366, 367, 369, 371], "x2": [124, 132, 133, 134, 135, 136, 137, 138, 140, 141, 144, 145, 146, 147, 148, 149, 150, 151, 152, 275, 363, 365, 366, 367, 369, 371], "y2": [124, 132, 133, 134, 135, 136, 137, 138, 140, 141, 144, 145, 146, 147, 148, 149, 150, 151, 152, 363, 365, 366, 367, 369, 371], "satisfi": [124, 147, 148, 149, 151, 152], "fed": 124, "spatial_scal": [125, 126, 128, 129, 148, 149, 151, 152], "ps_roi_align": 125, "roi": [125, 126, 128, 129, 148, 149, 151, 152], "ps_roi_pool": 126, "roi_align": 128, "roi_pool": 129, "input_channel": 130, "squeeze_channel": 130, "scale_activ": 130, "sigmoid": 130, "excit": 130, "arxiv": [130, 138, 141, 142, 143, 145, 153, 179, 258, 497], "ab": [130, 138, 141, 142, 143, 145, 153, 179, 258, 497], "1709": 130, "01507": 130, "delta": 130, "stochastic_depth": 131, "iou_threshold": [132, 147], "nm": 132, "discard": [132, 147], "overlap": [132, 138, 141, 145, 147, 262], "kept": [132, 147], "decreas": [132, 147, 342], "in_fmt": 134, "out_fmt": 134, "bottom": [134, 170, 177, 184, 208, 216, 217, 225, 248, 256, 264], "xywh": [134, 236, 325], "cxcywh": [134, 236, 325], "cx": 134, "cy": 134, "boxes1": [135, 137, 138, 140, 141, 144, 145], "boxes2": [135, 137, 138, 140, 141, 144, 145], "intersect": [135, 137, 138, 140, 141, 144, 145, 147], "jaccard": [135, 137, 140, 144, 262], "nxm": [135, 137, 140, 144], "matrix": [135, 137, 140, 144, 167, 244], "pairwis": [135, 137, 140, 144], "lie": [136, 164, 241], "prevent": [137, 138, 140, 141, 145], "reduct": [138, 141, 145, 153], "gradient": [138, 141, 145], "friendli": [138, 141, 145], "penalti": [138, 141, 145], "geometr": 138, "aspect": [138, 179, 186, 220, 258, 262, 268], "symmetr": [138, 141, 145, 170, 177, 216, 248, 256], "interchang": [138, 141, 145, 327], "averag": [138, 141, 145, 151, 153, 394, 395], "zhaohui": [138, 141], "zheng": [138, 141], "et": [138, 141, 145, 179, 258], "al": [138, 141, 145, 179, 258], "1911": [138, 141], "08287": [138, 141], "deform": 139, "convnet": [139, 344, 353, 354, 355, 356], "around": [139, 146, 187, 223, 269], "batch_sz": 139, "out_h": 139, "out_w": 139, "kh": 139, "kw": 139, "dropblock2d": 142, "1810": [142, 143], "12890": [142, 143], "dropblock3d": 143, "smallest": 145, "enclos": 145, "hamid": 145, "rezatofighi": 145, "metric": [145, 394, 395], "regress": [145, 363, 365, 366, 367, 369, 371], "1902": 145, "09630": 145, "higher": 147, "criterion": 147, "argsort": 147, "sensit": [148, 149, 508], "column": [148, 149, 151, 152, 167, 244], "bin": [148, 149, 151], "224x224": [148, 149, 151, 152], "112x112": [148, 149, 151, 152], "5x": [148, 149, 151, 152, 404, 406, 437, 439], "ceil": [148, 151], "roi_width": [148, 151], "output_width": [148, 151, 267, 273], "min_siz": [150, 267, 270, 275, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "minimum": [150, 164, 241, 262, 267, 270, 273, 276], "quantiz": [151, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "legaci": [151, 186, 190, 220, 221, 267, 268, 270, 274, 276], "neighbor": [151, 284], "detectron2": 151, "fast": 152, "gamma": [153, 197, 287], "dens": [153, 346, 357, 358, 359, 360, 369, 492], "1708": [153, 179, 258], "02002": 153, "vs": 153, "expon": 153, "p_t": 153, "stochast": [154, 155, 230], "deep": [154, 218, 309, 343, 352, 400, 401, 402, 403, 423, 424, 425, 426, 427, 428, 429, 430, 449, 450, 451, 452, 453, 454, 455, 456, 460, 488, 489, 490, 491, 499], "residu": [154, 372, 391, 398, 400, 401, 402, 403, 423, 424, 425, 426, 427, 428, 429, 430, 471, 472, 482, 483, 488, 489, 490, 491, 497, 505], "branch": 154, "mixture_width": [155, 230], "chain_depth": [155, 230], "all_op": [155, 230], "interpolationmod": [155, 156, 162, 171, 173, 184, 186, 187, 190, 194, 201, 217, 220, 221, 223, 230, 231, 239, 249, 251, 264, 267, 268, 269, 270, 274, 276, 283, 291, 298, 308, 312, 313, 315, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "bilinear": [155, 156, 162, 171, 173, 184, 186, 187, 190, 194, 201, 217, 220, 221, 223, 230, 231, 239, 249, 251, 264, 267, 268, 269, 270, 274, 276, 283, 298, 308, 312, 313, 352, 353, 354, 355, 356, 357, 358, 359, 360, 374, 382, 383, 384, 385, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 469, 470, 471, 472, 508], "robust": [155, 230, 340], "uncertainti": [155, 230], "chain": [155, 230], "hyperparamet": [155, 230, 238, 245], "distribut": [155, 206, 230, 238, 245, 340], "nearest": [155, 156, 162, 171, 173, 184, 186, 187, 190, 194, 201, 217, 220, 221, 223, 230, 231, 239, 249, 251, 264, 267, 268, 269, 270, 274, 276, 283, 284, 291, 315], "outsid": [155, 156, 162, 171, 173, 184, 187, 194, 201, 217, 223, 230, 231, 249, 275, 283], "band": [155, 156, 162, 171, 173, 184, 187, 194, 201, 217, 223, 230, 231, 249, 283], "get_param": [156, 159, 164, 173, 177, 179, 184, 186, 187, 231, 234, 241, 251, 256, 258, 264, 268, 269, 508], "transform_num": [156, 231], "autoaug": [157, 508], "along": [158, 159, 203, 205, 232, 234], "edg": [158, 170, 177, 186, 190, 203, 205, 216, 220, 232, 248, 256, 268, 274], "squar": [158, 163, 167, 177, 186, 191, 208, 209, 225, 232, 240, 244, 256, 268, 277], "brightness_factor": [159, 195, 234, 265, 285], "uniformli": [159, 164, 234, 241, 265, 284], "contrast_factor": [159, 196, 234, 265, 286], "saturation_factor": [159, 199, 234, 265, 289], "hue_factor": [159, 198, 234, 265, 288], "hsv": [159, 198, 226, 234, 265], "centercrop": [160, 235, 508], "convertimagedtyp": [160, 179, 235, 258, 278], "scriptabl": [160, 174, 235, 252], "dpython": [161, 204, 237, 326, 327, 328, 330], "integ": [161, 162, 173, 184, 186, 187, 190, 201, 204, 209, 217, 220, 221, 223, 237, 239, 251, 264, 267, 268, 269, 270, 274, 276, 334, 508], "forth": [161, 204, 237], "mismatch": [161, 163, 191, 204, 208, 225, 237, 240, 277], "cast": [161, 204, 237], "int32": [161, 204, 237], "float64": [161, 204, 237], "overflow": [161, 204, 237], "elast": [162, 239], "smooth": [162, 239], "grid_sampl": 162, "magnitud": [162, 171, 194, 239, 249, 283], "pillow": [162, 173, 184, 186, 187, 190, 201, 217, 220, 221, 223, 239, 251, 264, 267, 268, 269, 270, 274, 276], "constant": [162, 170, 173, 177, 184, 186, 187, 190, 197, 201, 216, 217, 220, 221, 223, 239, 248, 251, 256, 264, 267, 268, 269, 270, 273, 274, 276, 306], "deal": [163, 191, 240, 277], "5d": [163, 170, 177, 191, 216], "bs": [163, 191], "ncrop": [163, 191], "fuse": [163, 191], "result_avg": [163, 191], "avg": [163, 191], "deviat": [164, 168, 209, 215, 241, 246], "sigma_min": [164, 241], "sigma_max": [164, 241], "num_output_channel": [165, 222, 226, 242, 314, 319], "b": [165, 170, 173, 177, 179, 180, 215, 216, 222, 226, 239, 248, 251, 256, 258, 264, 269, 273, 337, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "lambd": [166, 243], "transformation_matrix": [167, 244], "mean_vector": [167, 244], "offlin": [167, 244], "flatten": [167, 244], "subtract": [167, 244], "dot": [167, 244], "product": [167, 244], "reshap": [167, 244], "whiten": [167, 244], "covari": [167, 244], "mm": [167, 244], "svd": [167, 244], "act": [168, 215, 246], "mutat": [168, 215, 246], "padding_mod": [170, 177, 216, 239, 248, 251, 256, 264, 269, 273, 306], "num_op": [171, 249], "num_magnitude_bin": [171, 194, 249, 283], "autom": [171, 249], "search": [171, 249, 333, 372, 387, 388, 389, 390, 392, 393, 399, 436, 478, 481, 484, 485, 497], "sharpen": [172, 250], "shear": [173, 201, 251, 291], "invari": [173, 201, 251], "deactiv": [173, 251], "absolut": [173, 251, 333], "img_width": [173, 251], "dx": [173, 251], "img_height": [173, 251], "dy": [173, 251], "Will": [173, 251], "upper": [173, 186, 187, 201, 207, 223, 251, 268, 269], "scale_rang": [173, 251, 276], "img_siz": [173, 251], "modulelist": [174, 252], "colorjitt": [174, 252, 265], "pad_if_need": [177, 256], "j": [177, 179, 186, 207, 256, 258, 268, 300], "rectangl": [179, 258], "eras": [179, 258], "zhong": [179, 258], "04896": [179, 258], "proport": [179, 258], "against": [179, 258], "3333333333333333": [186, 268], "portion": [186, 268], "popularli": [186, 268], "incept": [186, 268, 342, 384, 385, 396, 397, 476], "nearest_exact": [186, 190, 220, 221, 267, 268, 270, 274, 276], "bicub": [186, 190, 220, 221, 267, 268, 270, 274, 276, 373, 374, 375, 376, 377, 378, 379, 380, 381, 386, 415, 416, 418, 443, 444, 445, 446, 447, 448, 466, 468, 469, 508], "aren": [186, 190, 220, 221, 267, 268, 270, 274, 276], "expand": [187, 223, 269, 315], "expans": [187, 223, 269], "flag": [187, 223, 269, 331, 340], "max_siz": [190, 220, 267, 270, 274, 312], "downsampl": [190, 220, 274, 423, 424, 427, 488], "signific": [190, 220, 274, 508], "serv": [190, 220, 274], "closer": [190, 220, 274, 457, 460, 461, 501], "longer": [190, 220, 274], "overrul": [190, 220, 274], "shorter": [190, 220, 274], "vertical_flip": [191, 225, 277, 317], "assumpt": [192, 239, 280], "rgba": [192, 193, 280, 282], "la": [192, 193, 280, 282], "short": [192, 280], "floattensor": [193, 282, 363, 365, 366, 367, 369, 371], "ycbcr": [193, 282], "cmyk": [193, 282], "trivialaug": [194, 283], "wide": [194, 283, 342, 471, 472], "tune": [194, 283, 361, 394, 415, 416, 418, 466, 468, 469], "art": [194, 283], "black": [195, 199, 202], "solid": 196, "gain": [197, 287], "known": [197, 423, 424, 427, 488], "power": 197, "law": 197, "equat": 197, "i_": 197, "text": [197, 220], "frac": [197, 220], "shadow": 197, "darker": 197, "dark": 197, "lighter": 197, "multipli": [197, 387, 388, 389, 390], "cyclic": 198, "revers": 198, "complementari": 198, "white": [199, 202, 352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472], "enhanc": [199, 364, 368, 370], "clockwis": [201, 223], "post": [201, 363, 365, 366, 367, 369, 371, 396, 397, 400, 401, 402, 403, 404, 405, 406, 407, 506], "overal": 201, "maxim": 202, "remap": 202, "lowest": 202, "lightest": 202, "tl": [208, 225, 508], "tr": [208, 225], "bl": [208, 225], "br": [208, 225], "kx": 209, "ky": 209, "ksize": 209, "sigma_x": 209, "sigma_i": 209, "startpoint": [217, 308], "endpoint": [217, 308], "pic": [218, 227, 228, 309, 321], "notabl": 221, "meth": 222, "tl_flip": 225, "tr_flip": 225, "bl_flip": 225, "br_flip": 225, "center_flip": 225, "lab": 226, "topilimag": [227, 321], "totensor": 228, "clamp": 233, "extra_repr": [235, 243, 252], "eg": 236, "strong": [238, 276, 365], "classifi": [238, 340, 415, 416, 418, 453, 466, 468, 469], "localiz": 238, "meant": [238, 245, 508], "determinist": [238, 245], "hot": [238, 245], "identifi": [238, 245, 275], "scenario": [238, 245], "imgs_batch": [238, 245], "labels_batch": [238, 245], "randomperspect": [239, 308], "approxim": [239, 462], "invers": 239, "inverse_grid": 239, "batchmulticrop": 240, "images_or_video": 240, "image_or_video": 240, "gausssianblur": 241, "empir": [245, 508], "risk": 245, "minim": [245, 262], "liter": [248, 256], "min_scal": 262, "max_scal": 262, "min_aspect_ratio": 262, "max_aspect_ratio": 262, "sampler_opt": 262, "trial": 262, "ssd": [262, 265, 273, 342, 371, 372, 497], "shot": [262, 265, 273, 371, 496, 497], "multibox": [262, 265, 273, 371, 496, 497], "detector": [262, 265, 273, 365, 371, 496, 497], "immedi": 262, "875": 265, "125": 265, "uniform_sampl": [267, 273], "output_height": [267, 273], "theori": 269, "compar": [269, 341, 508], "best": [269, 342, 508], "side_rang": 273, "input_width": 273, "input_height": 273, "invalid": 275, "clampboundingbox": [275, 294], "undesir": 275, "insensit": [275, 331], "target_s": 276, "past": 276, "fivecrop": [277, 301], "catch": 278, "releas": [282, 340, 507, 508], "num_sampl": [284, 323], "subsampl": 284, "equispac": 284, "inpt": [285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 310, 312, 313, 314, 315, 316, 317, 318, 319, 320, 322, 323, 324], "randomautocontrast": [286, 292], "randomadjustsharp": 290, "randomaffin": 291, "old_format": 295, "new_format": 295, "convertboundingboxformat": 295, "to_dtyp": [296, 322], "elastictransform": 298, "randomequ": 299, "randomeras": 300, "gaussianblur": 302, "randominvert": 304, "rgb_imag": 307, "bgr_imag": 307, "permutate_channel": 307, "\u00ecnpt": 307, "coeffici": 308, "randomposter": 310, "decor": 311, "randomsolar": 316, "tencrop": 317, "depreact": 322, "to_imag": 322, "uniformtemporalsubsampl": 323, "randomverticalflip": 324, "as_tensor": [326, 327, 328, 330], "autograd": [326, 327, 328, 330], "record": [326, 327, 328, 330], "Is": 329, "ndim": 329, "alia": [329, 342], "return_typ": 331, "topuretensor": 331, "restrict": 331, "wrappe": 332, "font_siz": [333, 342], "word": [333, 466, 467, 468, 469, 470, 504], "ff00ff": [333, 334, 335], "240": [333, 334, 335, 374], "157": [333, 334, 335], "truetyp": 333, "window": [333, 443, 444, 445, 498], "system": 333, "maco": 333, "num_mask": 335, "value_rang": 337, "scale_each": 337, "pad_valu": 337, "mini": [337, 338], "displai": [337, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "binaryio": 338, "mc3_18_weight": [339, 342, 457], "kinetics400_v1": [339, 342, 457, 458, 459, 460, 461, 462, 463, 464, 465], "mvit_v1_b_weight": [339, 342, 458], "477": [339, 342, 458], "mvit_v2_s_weight": [339, 342, 459], "757": [339, 342, 459], "665": [339, 342, 459], "r2plus1d_18_weight": [339, 342, 460], "463": [339, 342, 460], "175": [339, 342, 460], "r3d_18_weight": [339, 342, 461], "s3d_weight": [339, 342, 462], "swin3d_b_weight": [339, 342, 463], "427": [339, 342, 463], "386": [339, 342, 463], "140": [339, 342, 463], "kinetics400_imagenet22k_v1": [339, 342, 463], "643": [339, 342, 463], "swin3d_s_weight": [339, 342, 464], "521": [339, 342, 464], "158": [339, 342, 464], "swin3d_t_weight": [339, 342, 465], "715": [339, 342, 465], "519": [339, 342, 465], "framework": 340, "statu": 340, "stabl": [340, 507, 508], "term": [340, 342, 508], "major": 340, "gap": [340, 370], "tag": 340, "coverag": 340, "commit": 340, "prototyp": 340, "pypi": 340, "conda": [340, 507], "earli": [340, 366], "consider": 340, "save_imag": 340, "inspect": [340, 341], "get_image_backend": 340, "get_video_backend": 340, "set_image_backend": 340, "accimag": 340, "ipp": 340, "3rd": 340, "parti": 340, "bind": 340, "perhap": 340, "less": [340, 442], "disabl": 340, "latest": [340, 507], "compil": [340, 508], "torchaudio": 340, "torchtext": 340, "torchelast": 340, "torchserv": 340, "xla": 340, "alloc": 341, "moment": 341, "often": 341, "reader_md": 341, "whichev": 341, "subpackag": 342, "address": 342, "torch_hom": 342, "environ": [342, 508], "variabl": [342, 508], "load_state_dict_from_url": 342, "licens": [342, 415, 416, 418, 466, 468, 469], "condit": 342, "deriv": [342, 508], "respons": 342, "permiss": 342, "serial": 342, "state_dict": 342, "contrari": 342, "scriptmodul": 342, "older": [342, 507], "histor": 342, "migrat": 342, "straightforward": 342, "pretrain": [342, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "incorrect": 342, "img_transform": 342, "mechan": 342, "all_model": 342, "list_model": 342, "classification_model": 342, "m1": 342, "mobilenet_v3_larg": 342, "m2": 342, "quantized_mobilenet_v3_larg": 342, "fetch": 342, "get_weight": 342, "get_model_weight": 342, "weights_enum2": 342, "weight_enum": 342, "alexnet": [342, 441, 495], "convnext": [342, 353, 354, 355, 356], "densenet": [342, 357, 358, 359, 360], "efficientnet": [342, 348, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383], "efficientnetv2": [342, 381, 382, 383], "googlenet": 342, "v3": [342, 385, 397], "maxvit": [342, 386], "mnasnet": [342, 387, 388, 389, 390], "mobilenet": [342, 392, 393], "regnet": [342, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422], "resnext": [342, 402, 403, 428, 429, 430], "shufflenet": [342, 404, 405, 406, 407, 437, 438, 439, 440], "squeezenet": [342, 352, 441, 442], "swintransform": [342, 443, 444, 445, 446, 447, 448, 463, 464, 465], "vgg": [342, 449, 450, 451, 452, 453, 454, 455, 456], "visiontransform": [342, 466, 467, 468, 469, 470], "encode_jpeg": 342, "grace_hopper_517x606": 342, "1f": 342, "report": [342, 394, 395], "inceptionv3": [342, 475], "deeplabv3": [342, 431, 432, 433], "lraspp": [342, 436], "val2017": [342, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "fco": [342, 365], "ssdlite": [342, 372], "fasterrcnn_resnet50_fpn_v2": 342, "box_score_thresh": 342, "keypoint_nam": [342, 366], "mvit": [342, 458, 459], "s3d": 342, "r3d_18": 342, "shorten": 342, "neural": [343, 347, 352, 373, 374, 375, 376, 377, 378, 379, 380, 387, 388, 389, 390, 402, 403, 428, 429, 430, 481, 490, 491], "weird": [343, 352], "trick": [343, 352], "2020": [344, 353, 354, 355, 356], "rethink": [345, 347, 373, 374, 375, 376, 377, 378, 379, 380, 385, 397, 431, 432, 433, 462, 475, 476, 502], "atrou": [345, 431, 432, 433], "toward": [349, 363], "propos": [349, 363], "faster_rcnn": [349, 361, 362, 363, 364], "bar": [352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "stderr": [352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "num_param": [352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "61100840": 352, "tench": [352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472], "goldfish": [352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472], "shark": [352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472], "997": [352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 468, 469, 470, 471, 472], "crop_siz": [352, 353, 354, 355, 356, 357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472], "upon": [353, 354, 355, 356, 374, 382, 383, 391, 392, 393, 398, 399, 401, 402, 408, 409, 410, 411, 412, 413, 414, 416, 417, 418, 419, 420, 421, 422, 423, 424, 427, 428, 430, 471, 472], "88591464": 353, "338": 353, "197767336": 354, "50223688": 355, "191": 355, "230": [355, 424], "tini": 356, "28589128": 356, "109": 356, "236": 356, "121": [357, 465], "port": [357, 358, 359, 360, 373, 374, 375, 376, 377, 378, 379, 380, 381, 384, 385, 394, 395, 396, 397, 458, 459, 463, 464, 465], "luatorch": [357, 358, 359, 360], "7978856": 357, "28681000": 358, "110": 358, "169": [359, 367], "14149480": 359, "201": 360, "20013928": 360, "trainable_backbone_lay": [361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372], "mobilenetv3": [361, 362, 372, 392, 393, 399, 431, 436, 478, 484, 485, 497], "mobil": [361, 387, 388, 389, 390, 481], "300": [361, 362, 363, 365, 366, 367, 369, 371, 376], "trainabl": [361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372], "frozen": [361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 415, 416, 418, 466, 468, 469], "box_map": [361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372], "__background__": [361, 362, 363, 364, 365, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "bicycl": [361, 362, 363, 364, 365, 367, 368, 369, 370, 371, 372, 431, 432, 433, 434, 435, 436], "19386354": [361, 362], "ground": [363, 365, 366, 367, 369, 371], "truth": [363, 365, 366, 367, 369, 371], "int64tensor": [363, 365, 366, 367, 369, 371], "rpn": [363, 366, 367], "export": [363, 366, 367], "onnx": [363, 366, 367], "1200": 363, "opset_vers": [363, 366, 367], "41755286": 363, "159": 363, "transfer": [364, 368, 415, 416, 418, 466, 468, 469], "boost": [364, 368, 370], "43712278": 364, "anchor": [365, 370], "32269600": 365, "num_keypoint": 366, "keypoint_rcnn": 366, "kp_map": 366, "59137258": 366, "uint8tensor": 367, "soft": 367, "mask_map": [367, 368], "44401393": 367, "46359409": 368, "focal": [369, 492], "34014999": 369, "bridg": 370, "retinanet_resnet50_fpn": 370, "38198935": 370, "ssd300": 371, "35641826": 371, "320x320": 372, "mobilenetv2": [372, 391, 398, 482, 483, 497], "bottleneck": [372, 391, 398, 423, 424, 427, 471, 472, 482, 483, 488, 497], "ssd300_vgg16": 372, "320": [372, 376], "3440060": 372, "b0": 373, "5288548": 373, "b1": 374, "7794184": 374, "b2": 375, "9109994": 375, "b3": 376, "12233232": 376, "b4": 377, "19341616": 377, "380": 377, "b5": 378, "30389784": 378, "116": 378, "b6": 379, "43040704": 379, "165": 379, "528": 379, "b7": 380, "66347960": 380, "118515272": 381, "480": [381, 382], "54139356": 382, "21458488": 383, "deeper": [384, 396, 473, 474], "6624904": [384, 396], "inception3": [385, 475], "299": [385, 397], "27161264": [385, 397], "103": [385, 397], "342": [385, 397], "30919624": 386, "platform": [387, 388, 389, 390, 481], "awar": [387, 388, 389, 390, 398, 399, 481], "2218512": 387, "scratch": [388, 390, 392, 394, 395, 399, 403, 404, 405, 406, 407, 429, 437, 438, 439, 440, 449, 450, 451, 452, 453, 454, 455, 456, 466, 467, 469, 470], "3170208": 388, "4383312": 389, "6282256": 390, "3504872": [391, 398], "5483032": [392, 399], "margin": [392, 399], "2542856": 393, "recurr": [394, 395, 486], "c_t_skht_v2": 394, "c_t_v1": [394, 395], "per_image_ep": [394, 395], "fl": [394, 395], "f1": [394, 395], "leaderboard": [394, 395], "3px": [394, 395], "flyingchair": [394, 395], "cleanpass": [394, 395], "4411": 394, "finalpass": [394, 395], "7894": 394, "0172": 394, "fl_all": [394, 395], "4506": 394, "5257536": 394, "c_t_v2": [394, 395], "3822": 394, "7161": 394, "5118": 394, "0679": 394, "c_t_skht_v1": 394, "hd1k": 394, "819": 394, "067": 394, "c_t_skht_k_v1": 394, "c_t_skht_k_v2": 394, "raft_small_weight": 395, "1231": 395, "279": 395, "6557": 395, "2801": 395, "990162": 395, "9901": 395, "2831": 395, "5978": 395, "2369": 395, "quantizablegooglenet": [396, 474], "unquant": [396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407], "fbgemm": [396, 397, 400, 401, 402, 403, 404, 405, 406, 407], "quantizableinception3": [397, 476], "quantizablemobilenetv2": [398, 483], "qnnpack": [398, 399], "quantizablemobilenetv3": [399, 485], "quantizableresnet": [400, 401, 402, 403, 489, 491], "11689512": [400, 425], "25557032": [401, 427], "32x8d": [402, 428], "aggreg": [402, 403, 428, 429, 430, 490, 491], "resnet101_32x8d_quantizedweight": 402, "88791336": [402, 428], "339": [402, 428], "64x4d": [403, 429], "resnet101_64x4d_quantizedweight": 403, "83455272": [403, 429], "319": [403, 416, 429], "quantizableshufflenetv2": [404, 405, 406, 407, 494], "shufflenetv2": [404, 405, 406, 407, 437, 438, 439, 440, 493, 494], "guidelin": [404, 405, 406, 407, 437, 438, 439, 440, 493, 494, 508], "1366792": [404, 437], "0x": [405, 407, 438, 440], "2278604": [405, 438], "3503624": [406, 439], "7393996": [407, 440], "regnetx_16gf": 408, "blockparam": [408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422], "54278536": 408, "regnetx_1": 409, "6gf": [409, 417], "9190136": 409, "regnetx_32gf": 410, "107811560": 410, "412": 410, "regnetx_3": 411, "2gf": [411, 419], "15296552": 411, "regnetx_400mf": 412, "5495976": 412, "regnetx_800mf": 413, "7259656": 413, "regnetx_8gf": 414, "39572648": 414, "regnety_128gf": 415, "learnt": [415, 416, 418, 466, 468, 469], "swag": [415, 416, 418, 466, 468, 469], "644812894": 415, "2461": 415, "trunk": [415, 416, 418, 466, 468, 469], "regnety_16gf": 416, "83590140": 416, "regnety_1": 417, "11202430": 417, "regnety_32gf": 418, "145046770": 418, "554": 418, "regnety_3": 419, "19436338": 419, "regnety_400mf": 420, "4344144": 420, "regnety_800mf": 421, "6432512": 421, "regnety_8gf": 422, "39381472": 422, "3x3": [423, 424, 427, 488], "1x1": [423, 424, 427, 471, 472, 488], "44549160": 423, "170": 423, "60192808": 424, "21797672": 426, "32x4d": 430, "25028904": 430, "aux_loss": [431, 432, 433, 434, 435, 436], "auxiliari": [431, 432, 433, 434, 435, 436], "miou": [431, 432, 433, 434, 435, 436], "pixel_acc": [431, 432, 433, 434, 435, 436], "aeroplan": [431, 432, 433, 434, 435, 436], "11029328": 431, "60996202": 432, "42004074": 433, "54314346": 434, "35322218": 435, "lite": [436, 497], "aspp": 436, "3221538": 436, "50x": [441, 495], "fewer": [441, 442, 495], "5mb": [441, 495], "1248424": 441, "4x": 442, "1235496": 442, "swin_bas": [443, 463], "swin": [443, 444, 445, 446, 447, 448, 463, 464, 465, 498, 503], "hierarch": [443, 444, 445, 498], "swin_transform": [443, 444, 445, 446, 447, 448, 463, 464, 465, 498, 503], "87768224": 443, "335": 443, "238": 443, "swin_smal": [444, 464], "49606258": 444, "189": 444, "swin_tini": [445, 465], "28288354": 445, "swin_v2_bas": 446, "capac": [446, 447, 448, 498], "87930848": 446, "336": [446, 467], "272": 446, "swin_v2_smal": 447, "49737442": 447, "190": 447, "260": [447, 448], "swin_v2_tini": 448, "28351570": 448, "132863336": 449, "bn": [450, 452, 454], "132868840": 450, "133047848": 451, "133053736": 452, "138357544": 453, "527": [453, 454], "miss": 453, "48235": 453, "45882": 453, "40784": 453, "00392156862745098": 453, "138365992": 454, "143667240": 455, "548": [455, 456], "19_bn": 456, "143678248": 456, "videoresnet": [457, 460, 461, 501, 503], "spatiotempor": [457, 460, 461, 462, 501, 502], "estim": [457, 458, 459, 460, 461, 462, 463, 464, 465], "clips_per_video": [457, 458, 459, 460, 461, 462, 463, 464, 465], "abseil": [457, 458, 459, 460, 461, 462, 463, 464, 465], "air": [457, 458, 459, 460, 461, 462, 463, 464, 465], "drum": [457, 458, 459, 460, 461, 462, 463, 464, 465], "11695440": 457, "43216": [457, 460, 461, 462], "394666": [457, 460, 461, 462], "37645": [457, 460, 461, 462], "22803": [457, 460, 461, 462], "22145": [457, 460, 461, 462], "216989": [457, 460, 461, 462], "mvitv1": 458, "multiscal": [458, 459, 500], "min_temporal_s": [458, 459, 462, 463, 464, 465], "36610672": 458, "139": 458, "mvitv2": [459, 500], "34537744": 459, "131": [459, 472], "31505325": 460, "resnet3d": 461, "33371472": 461, "aim": 462, "8320048": 462, "swintransformer3d": [463, 464, 465, 503], "88048984": 463, "49816678": 464, "28158070": 465, "16x16": [466, 467, 468, 469, 470, 504], "vision_transform": [466, 467, 468, 469, 470, 504], "deit": [466, 467, 470], "86567656": 466, "86859496": 466, "88224232": 467, "633470440": 468, "518": 468, "2416": 468, "632045800": 468, "2411": 468, "304326632": 469, "1161": 469, "242": [469, 471], "305174504": 469, "1164": 469, "306535400": 470, "1169": 470, "outer": [471, 472], "2048": [471, 472], "1024": [471, 472], "126886696": 471, "484": 471, "68883240": 472, "263": 472, "1801": 497, "04381": 497, "__": 497, "trade": 502, "nightli": 507, "safer": 507, "output_dict": 508, "experienc": 508, "encourag": 508, "brows": 508, "max_dtyp": 508, "largest": 508, "dr": 508, "march": 508, "2023": 508, "neglig": 508, "plan": 508, "tend": 508, "experi": 508, "chase": 508, "talk": 508, "benefit": 508, "image_tensor": 508, "Their": 508, "counterpart": 508, "slighlti": 508, "parametr": 508, "resize_bounding_box": 508, "resized_crop_mask": 508, "underscor": 508, "bewar": 508, "vice": 508, "versa": 508, "techniqu": 508, "studi": 508, "standalon": 508, "manipul": 509, "introduct": 509}, "objects": {"": [[340, 0, 0, "-", "torchvision"]], "torchvision.datasets": [[25, 1, 1, "", "CIFAR10"], [26, 1, 1, "", "CIFAR100"], [27, 1, 1, "", "CREStereo"], [28, 1, 1, "", "Caltech101"], [29, 1, 1, "", "Caltech256"], [30, 1, 1, "", "CarlaStereo"], [31, 1, 1, "", "CelebA"], [32, 1, 1, "", "Cityscapes"], [33, 1, 1, "", "CocoCaptions"], [34, 1, 1, "", "CocoDetection"], [35, 1, 1, "", "Country211"], [36, 1, 1, "", "DTD"], [37, 1, 1, "", "DatasetFolder"], [38, 1, 1, "", "EMNIST"], [39, 1, 1, "", "ETH3DStereo"], [40, 1, 1, "", "EuroSAT"], [41, 1, 1, "", "FER2013"], [42, 1, 1, "", "FGVCAircraft"], [43, 1, 1, "", "FakeData"], [44, 1, 1, "", "FallingThingsStereo"], [45, 1, 1, "", "FashionMNIST"], [46, 1, 1, "", "Flickr30k"], [47, 1, 1, "", "Flickr8k"], [48, 1, 1, "", "Flowers102"], [49, 1, 1, "", "FlyingChairs"], [50, 1, 1, "", "FlyingThings3D"], [51, 1, 1, "", "Food101"], [52, 1, 1, "", "GTSRB"], [53, 1, 1, "", "HD1K"], [54, 1, 1, "", "HMDB51"], [55, 1, 1, "", "INaturalist"], [56, 1, 1, "", "ImageFolder"], [57, 1, 1, "", "ImageNet"], [58, 1, 1, "", "InStereo2k"], [59, 1, 1, "", "KMNIST"], [60, 1, 1, "", "Kinetics"], [61, 1, 1, "", "Kitti"], [62, 1, 1, "", "Kitti2012Stereo"], [63, 1, 1, "", "Kitti2015Stereo"], [64, 1, 1, "", "KittiFlow"], [65, 1, 1, "", "LFWPairs"], [66, 1, 1, "", "LFWPeople"], [67, 1, 1, "", "LSUN"], [68, 1, 1, "", "MNIST"], [69, 1, 1, "", "Middlebury2014Stereo"], [70, 1, 1, "", "MovingMNIST"], [71, 1, 1, "", "Omniglot"], [72, 1, 1, "", "OxfordIIITPet"], [73, 1, 1, "", "PCAM"], [74, 1, 1, "", "PhotoTour"], [75, 1, 1, "", "Places365"], [76, 1, 1, "", "QMNIST"], [77, 1, 1, "", "RenderedSST2"], [78, 1, 1, "", "SBDataset"], [79, 1, 1, "", "SBU"], [80, 1, 1, "", "SEMEION"], [81, 1, 1, "", "STL10"], [82, 1, 1, "", "SUN397"], [83, 1, 1, "", "SVHN"], [84, 1, 1, "", "SceneFlowStereo"], [85, 1, 1, "", "Sintel"], [86, 1, 1, "", "SintelStereo"], [87, 1, 1, "", "StanfordCars"], [88, 1, 1, "", "UCF101"], [89, 1, 1, "", "USPS"], [90, 1, 1, "", "VOCDetection"], [91, 1, 1, "", "VOCSegmentation"], [92, 1, 1, "", "VisionDataset"], [93, 1, 1, "", "WIDERFace"], [94, 3, 1, "", "wrap_dataset_for_transforms_v2"]], "torchvision.datasets.CIFAR10": [[25, 2, 1, "", "__getitem__"]], "torchvision.datasets.CIFAR100": [[26, 2, 1, "", "__getitem__"]], "torchvision.datasets.CREStereo": [[27, 2, 1, "", "__getitem__"]], "torchvision.datasets.Caltech101": [[28, 2, 1, "", "__getitem__"]], "torchvision.datasets.Caltech256": [[29, 2, 1, "", "__getitem__"]], "torchvision.datasets.CarlaStereo": [[30, 2, 1, "", "__getitem__"]], "torchvision.datasets.CelebA": [[31, 2, 1, "", "__getitem__"]], "torchvision.datasets.Cityscapes": [[32, 2, 1, "", "__getitem__"]], "torchvision.datasets.CocoCaptions": [[33, 2, 1, "", "__getitem__"]], "torchvision.datasets.CocoDetection": [[34, 2, 1, "", "__getitem__"]], "torchvision.datasets.Country211": [[35, 2, 1, "", "__getitem__"]], "torchvision.datasets.DTD": [[36, 2, 1, "", "__getitem__"]], "torchvision.datasets.DatasetFolder": [[37, 2, 1, "", "find_classes"], [37, 2, 1, "", "make_dataset"]], "torchvision.datasets.EMNIST": [[38, 2, 1, "", "__getitem__"]], "torchvision.datasets.ETH3DStereo": [[39, 2, 1, "", "__getitem__"]], "torchvision.datasets.EuroSAT": [[40, 2, 1, "", "__getitem__"]], "torchvision.datasets.FER2013": [[41, 2, 1, "", "__getitem__"]], "torchvision.datasets.FGVCAircraft": [[42, 2, 1, "", "__getitem__"]], "torchvision.datasets.FakeData": [[43, 2, 1, "", "__getitem__"]], "torchvision.datasets.FallingThingsStereo": [[44, 2, 1, "", "__getitem__"]], "torchvision.datasets.FashionMNIST": [[45, 2, 1, "", "__getitem__"]], "torchvision.datasets.Flickr30k": [[46, 2, 1, "", "__getitem__"]], "torchvision.datasets.Flickr8k": [[47, 2, 1, "", "__getitem__"]], "torchvision.datasets.Flowers102": [[48, 2, 1, "", "__getitem__"]], "torchvision.datasets.FlyingChairs": [[49, 2, 1, "", "__getitem__"]], "torchvision.datasets.FlyingThings3D": [[50, 2, 1, "", "__getitem__"]], "torchvision.datasets.Food101": [[51, 2, 1, "", "__getitem__"]], "torchvision.datasets.GTSRB": [[52, 2, 1, "", "__getitem__"]], "torchvision.datasets.HD1K": [[53, 2, 1, "", "__getitem__"]], "torchvision.datasets.HMDB51": [[54, 2, 1, "", "__getitem__"]], "torchvision.datasets.INaturalist": [[55, 2, 1, "", "__getitem__"], [55, 2, 1, "", "category_name"]], "torchvision.datasets.ImageNet": [[57, 2, 1, "", "__getitem__"]], "torchvision.datasets.InStereo2k": [[58, 2, 1, "", "__getitem__"]], "torchvision.datasets.KMNIST": [[59, 2, 1, "", "__getitem__"]], "torchvision.datasets.Kinetics": [[60, 2, 1, "", "__getitem__"]], "torchvision.datasets.Kitti": [[61, 2, 1, "", "__getitem__"]], "torchvision.datasets.Kitti2012Stereo": [[62, 2, 1, "", "__getitem__"]], "torchvision.datasets.Kitti2015Stereo": [[63, 2, 1, "", "__getitem__"]], "torchvision.datasets.KittiFlow": [[64, 2, 1, "", "__getitem__"]], "torchvision.datasets.LFWPairs": [[65, 2, 1, "", "__getitem__"]], "torchvision.datasets.LFWPeople": [[66, 2, 1, "", "__getitem__"]], "torchvision.datasets.LSUN": [[67, 2, 1, "", "__getitem__"]], "torchvision.datasets.MNIST": [[68, 2, 1, "", "__getitem__"]], "torchvision.datasets.Middlebury2014Stereo": [[69, 2, 1, "", "__getitem__"]], "torchvision.datasets.MovingMNIST": [[70, 2, 1, "", "__getitem__"]], "torchvision.datasets.Omniglot": [[71, 2, 1, "", "__getitem__"]], "torchvision.datasets.OxfordIIITPet": [[72, 2, 1, "", "__getitem__"]], "torchvision.datasets.PCAM": [[73, 2, 1, "", "__getitem__"]], "torchvision.datasets.PhotoTour": [[74, 2, 1, "", "__getitem__"]], "torchvision.datasets.Places365": [[75, 2, 1, "", "__getitem__"]], "torchvision.datasets.QMNIST": [[76, 2, 1, "", "__getitem__"]], "torchvision.datasets.RenderedSST2": [[77, 2, 1, "", "__getitem__"]], "torchvision.datasets.SBDataset": [[78, 2, 1, "", "__getitem__"]], "torchvision.datasets.SBU": [[79, 2, 1, "", "__getitem__"]], "torchvision.datasets.SEMEION": [[80, 2, 1, "", "__getitem__"]], "torchvision.datasets.STL10": [[81, 2, 1, "", "__getitem__"]], "torchvision.datasets.SUN397": [[82, 2, 1, "", "__getitem__"]], "torchvision.datasets.SVHN": [[83, 2, 1, "", "__getitem__"]], "torchvision.datasets.SceneFlowStereo": [[84, 2, 1, "", "__getitem__"]], "torchvision.datasets.Sintel": [[85, 2, 1, "", "__getitem__"]], "torchvision.datasets.SintelStereo": [[86, 2, 1, "", "__getitem__"]], "torchvision.datasets.StanfordCars": [[87, 2, 1, "", "__getitem__"]], "torchvision.datasets.UCF101": [[88, 2, 1, "", "__getitem__"]], "torchvision.datasets.USPS": [[89, 2, 1, "", "__getitem__"]], "torchvision.datasets.VOCDetection": [[90, 2, 1, "", "__getitem__"]], "torchvision.datasets.VOCSegmentation": [[91, 2, 1, "", "__getitem__"]], "torchvision.datasets.WIDERFace": [[93, 2, 1, "", "__getitem__"]], "torchvision": [[340, 3, 1, "", "get_image_backend"], [340, 3, 1, "", "get_video_backend"], [340, 3, 1, "", "set_image_backend"], [340, 3, 1, "", "set_video_backend"]], "torchvision.io": [[95, 1, 1, "", "ImageReadMode"], [96, 1, 1, "", "VideoReader"], [97, 3, 1, "", "decode_image"], [98, 3, 1, "", "decode_jpeg"], [99, 3, 1, "", "decode_png"], [100, 3, 1, "", "encode_jpeg"], [101, 3, 1, "", "encode_png"], [102, 3, 1, "", "read_file"], [103, 3, 1, "", "read_image"], [104, 3, 1, "", "read_video"], [105, 3, 1, "", "read_video_timestamps"], [106, 3, 1, "", "write_file"], [107, 3, 1, "", "write_jpeg"], [108, 3, 1, "", "write_png"], [109, 3, 1, "", "write_video"]], "torchvision.io.VideoReader": [[96, 2, 1, "", "get_metadata"], [96, 2, 1, "", "seek"], [96, 2, 1, "", "set_current_stream"]], "torchvision.models": [[352, 1, 1, "", "AlexNet_Weights"], [353, 1, 1, "", "ConvNeXt_Base_Weights"], [354, 1, 1, "", "ConvNeXt_Large_Weights"], [355, 1, 1, "", "ConvNeXt_Small_Weights"], [356, 1, 1, "", "ConvNeXt_Tiny_Weights"], [357, 1, 1, "", "DenseNet121_Weights"], [358, 1, 1, "", "DenseNet161_Weights"], [359, 1, 1, "", "DenseNet169_Weights"], [360, 1, 1, "", "DenseNet201_Weights"], [373, 1, 1, "", "EfficientNet_B0_Weights"], [374, 1, 1, "", "EfficientNet_B1_Weights"], [375, 1, 1, "", "EfficientNet_B2_Weights"], [376, 1, 1, "", "EfficientNet_B3_Weights"], [377, 1, 1, "", "EfficientNet_B4_Weights"], [378, 1, 1, "", "EfficientNet_B5_Weights"], [379, 1, 1, "", "EfficientNet_B6_Weights"], [380, 1, 1, "", "EfficientNet_B7_Weights"], [381, 1, 1, "", "EfficientNet_V2_L_Weights"], [382, 1, 1, "", "EfficientNet_V2_M_Weights"], [383, 1, 1, "", "EfficientNet_V2_S_Weights"], [384, 1, 1, "", "GoogLeNet_Weights"], [385, 1, 1, "", "Inception_V3_Weights"], [387, 1, 1, "", "MNASNet0_5_Weights"], [388, 1, 1, "", "MNASNet0_75_Weights"], [389, 1, 1, "", "MNASNet1_0_Weights"], [390, 1, 1, "", "MNASNet1_3_Weights"], [386, 1, 1, "", "MaxVit_T_Weights"], [391, 1, 1, "", "MobileNet_V2_Weights"], [392, 1, 1, "", "MobileNet_V3_Large_Weights"], [393, 1, 1, "", "MobileNet_V3_Small_Weights"], [408, 1, 1, "", "RegNet_X_16GF_Weights"], [409, 1, 1, "", "RegNet_X_1_6GF_Weights"], [410, 1, 1, "", "RegNet_X_32GF_Weights"], [411, 1, 1, "", "RegNet_X_3_2GF_Weights"], [412, 1, 1, "", "RegNet_X_400MF_Weights"], [413, 1, 1, "", "RegNet_X_800MF_Weights"], [414, 1, 1, "", "RegNet_X_8GF_Weights"], [415, 1, 1, "", "RegNet_Y_128GF_Weights"], [416, 1, 1, "", "RegNet_Y_16GF_Weights"], [417, 1, 1, "", "RegNet_Y_1_6GF_Weights"], [418, 1, 1, "", "RegNet_Y_32GF_Weights"], [419, 1, 1, "", "RegNet_Y_3_2GF_Weights"], [420, 1, 1, "", "RegNet_Y_400MF_Weights"], [421, 1, 1, "", "RegNet_Y_800MF_Weights"], [422, 1, 1, "", "RegNet_Y_8GF_Weights"], [428, 1, 1, "", "ResNeXt101_32X8D_Weights"], [429, 1, 1, "", "ResNeXt101_64X4D_Weights"], [430, 1, 1, "", "ResNeXt50_32X4D_Weights"], [423, 1, 1, "", "ResNet101_Weights"], [424, 1, 1, "", "ResNet152_Weights"], [425, 1, 1, "", "ResNet18_Weights"], [426, 1, 1, "", "ResNet34_Weights"], [427, 1, 1, "", "ResNet50_Weights"], [437, 1, 1, "", "ShuffleNet_V2_X0_5_Weights"], [438, 1, 1, "", "ShuffleNet_V2_X1_0_Weights"], [439, 1, 1, "", "ShuffleNet_V2_X1_5_Weights"], [440, 1, 1, "", "ShuffleNet_V2_X2_0_Weights"], [441, 1, 1, "", "SqueezeNet1_0_Weights"], [442, 1, 1, "", "SqueezeNet1_1_Weights"], [443, 1, 1, "", "Swin_B_Weights"], [444, 1, 1, "", "Swin_S_Weights"], [445, 1, 1, "", "Swin_T_Weights"], [446, 1, 1, "", "Swin_V2_B_Weights"], [447, 1, 1, "", "Swin_V2_S_Weights"], [448, 1, 1, "", "Swin_V2_T_Weights"], [450, 1, 1, "", "VGG11_BN_Weights"], [449, 1, 1, "", "VGG11_Weights"], [452, 1, 1, "", "VGG13_BN_Weights"], [451, 1, 1, "", "VGG13_Weights"], [454, 1, 1, "", "VGG16_BN_Weights"], [453, 1, 1, "", "VGG16_Weights"], [456, 1, 1, "", "VGG19_BN_Weights"], [455, 1, 1, "", "VGG19_Weights"], [466, 1, 1, "", "ViT_B_16_Weights"], [467, 1, 1, "", "ViT_B_32_Weights"], [468, 1, 1, "", "ViT_H_14_Weights"], [469, 1, 1, "", "ViT_L_16_Weights"], [470, 1, 1, "", "ViT_L_32_Weights"], [471, 1, 1, "", "Wide_ResNet101_2_Weights"], [472, 1, 1, "", "Wide_ResNet50_2_Weights"], [352, 3, 1, "", "alexnet"], [353, 3, 1, "", "convnext_base"], [354, 3, 1, "", "convnext_large"], [355, 3, 1, "", "convnext_small"], [356, 3, 1, "", "convnext_tiny"], [357, 3, 1, "", "densenet121"], [358, 3, 1, "", "densenet161"], [359, 3, 1, "", "densenet169"], [360, 3, 1, "", "densenet201"], [373, 3, 1, "", "efficientnet_b0"], [374, 3, 1, "", "efficientnet_b1"], [375, 3, 1, "", "efficientnet_b2"], [376, 3, 1, "", "efficientnet_b3"], [377, 3, 1, "", "efficientnet_b4"], [378, 3, 1, "", "efficientnet_b5"], [379, 3, 1, "", "efficientnet_b6"], [380, 3, 1, "", "efficientnet_b7"], [381, 3, 1, "", "efficientnet_v2_l"], [382, 3, 1, "", "efficientnet_v2_m"], [383, 3, 1, "", "efficientnet_v2_s"], [112, 3, 1, "", "get_model"], [113, 3, 1, "", "get_model_weights"], [114, 3, 1, "", "get_weight"], [384, 3, 1, "", "googlenet"], [385, 3, 1, "", "inception_v3"], [115, 3, 1, "", "list_models"], [386, 3, 1, "", "maxvit_t"], [387, 3, 1, "", "mnasnet0_5"], [388, 3, 1, "", "mnasnet0_75"], [389, 3, 1, "", "mnasnet1_0"], [390, 3, 1, "", "mnasnet1_3"], [391, 3, 1, "", "mobilenet_v2"], [392, 3, 1, "", "mobilenet_v3_large"], [393, 3, 1, "", "mobilenet_v3_small"], [408, 3, 1, "", "regnet_x_16gf"], [409, 3, 1, "", "regnet_x_1_6gf"], [410, 3, 1, "", "regnet_x_32gf"], [411, 3, 1, "", "regnet_x_3_2gf"], [412, 3, 1, "", "regnet_x_400mf"], [413, 3, 1, "", "regnet_x_800mf"], [414, 3, 1, "", "regnet_x_8gf"], [415, 3, 1, "", "regnet_y_128gf"], [416, 3, 1, "", "regnet_y_16gf"], [417, 3, 1, "", "regnet_y_1_6gf"], [418, 3, 1, "", "regnet_y_32gf"], [419, 3, 1, "", "regnet_y_3_2gf"], [420, 3, 1, "", "regnet_y_400mf"], [421, 3, 1, "", "regnet_y_800mf"], [422, 3, 1, "", "regnet_y_8gf"], [423, 3, 1, "", "resnet101"], [424, 3, 1, "", "resnet152"], [425, 3, 1, "", "resnet18"], [426, 3, 1, "", "resnet34"], [427, 3, 1, "", "resnet50"], [428, 3, 1, "", "resnext101_32x8d"], [429, 3, 1, "", "resnext101_64x4d"], [430, 3, 1, "", "resnext50_32x4d"], [437, 3, 1, "", "shufflenet_v2_x0_5"], [438, 3, 1, "", "shufflenet_v2_x1_0"], [439, 3, 1, "", "shufflenet_v2_x1_5"], [440, 3, 1, "", "shufflenet_v2_x2_0"], [441, 3, 1, "", "squeezenet1_0"], [442, 3, 1, "", "squeezenet1_1"], [443, 3, 1, "", "swin_b"], [444, 3, 1, "", "swin_s"], [445, 3, 1, "", "swin_t"], [446, 3, 1, "", "swin_v2_b"], [447, 3, 1, "", "swin_v2_s"], [448, 3, 1, "", "swin_v2_t"], [449, 3, 1, "", "vgg11"], [450, 3, 1, "", "vgg11_bn"], [451, 3, 1, "", "vgg13"], [452, 3, 1, "", "vgg13_bn"], [453, 3, 1, "", "vgg16"], [454, 3, 1, "", "vgg16_bn"], [455, 3, 1, "", "vgg19"], [456, 3, 1, "", "vgg19_bn"], [466, 3, 1, "", "vit_b_16"], [467, 3, 1, "", "vit_b_32"], [468, 3, 1, "", "vit_h_14"], [469, 3, 1, "", "vit_l_16"], [470, 3, 1, "", "vit_l_32"], [471, 3, 1, "", "wide_resnet101_2"], [472, 3, 1, "", "wide_resnet50_2"]], "torchvision.models.detection": [[365, 1, 1, "", "FCOS_ResNet50_FPN_Weights"], [361, 1, 1, "", "FasterRCNN_MobileNet_V3_Large_320_FPN_Weights"], [362, 1, 1, "", "FasterRCNN_MobileNet_V3_Large_FPN_Weights"], [364, 1, 1, "", "FasterRCNN_ResNet50_FPN_V2_Weights"], [363, 1, 1, "", "FasterRCNN_ResNet50_FPN_Weights"], [366, 1, 1, "", "KeypointRCNN_ResNet50_FPN_Weights"], [368, 1, 1, "", "MaskRCNN_ResNet50_FPN_V2_Weights"], [367, 1, 1, "", "MaskRCNN_ResNet50_FPN_Weights"], [370, 1, 1, "", "RetinaNet_ResNet50_FPN_V2_Weights"], [369, 1, 1, "", "RetinaNet_ResNet50_FPN_Weights"], [371, 1, 1, "", "SSD300_VGG16_Weights"], [372, 1, 1, "", "SSDLite320_MobileNet_V3_Large_Weights"], [361, 3, 1, "", "fasterrcnn_mobilenet_v3_large_320_fpn"], [362, 3, 1, "", "fasterrcnn_mobilenet_v3_large_fpn"], [363, 3, 1, "", "fasterrcnn_resnet50_fpn"], [364, 3, 1, "", "fasterrcnn_resnet50_fpn_v2"], [365, 3, 1, "", "fcos_resnet50_fpn"], [366, 3, 1, "", "keypointrcnn_resnet50_fpn"], [367, 3, 1, "", "maskrcnn_resnet50_fpn"], [368, 3, 1, "", "maskrcnn_resnet50_fpn_v2"], [369, 3, 1, "", "retinanet_resnet50_fpn"], [370, 3, 1, "", "retinanet_resnet50_fpn_v2"], [371, 3, 1, "", "ssd300_vgg16"], [372, 3, 1, "", "ssdlite320_mobilenet_v3_large"]], "torchvision.models.feature_extraction": [[110, 3, 1, "", "create_feature_extractor"], [111, 3, 1, "", "get_graph_node_names"]], "torchvision.models.optical_flow": [[394, 1, 1, "", "Raft_Large_Weights"], [395, 1, 1, "", "Raft_Small_Weights"], [394, 3, 1, "", "raft_large"], [395, 3, 1, "", "raft_small"]], "torchvision.models.quantization": [[396, 1, 1, "", "GoogLeNet_QuantizedWeights"], [397, 1, 1, "", "Inception_V3_QuantizedWeights"], [398, 1, 1, "", "MobileNet_V2_QuantizedWeights"], [399, 1, 1, "", "MobileNet_V3_Large_QuantizedWeights"], [402, 1, 1, "", "ResNeXt101_32X8D_QuantizedWeights"], [403, 1, 1, "", "ResNeXt101_64X4D_QuantizedWeights"], [400, 1, 1, "", "ResNet18_QuantizedWeights"], [401, 1, 1, "", "ResNet50_QuantizedWeights"], [404, 1, 1, "", "ShuffleNet_V2_X0_5_QuantizedWeights"], [405, 1, 1, "", "ShuffleNet_V2_X1_0_QuantizedWeights"], [406, 1, 1, "", "ShuffleNet_V2_X1_5_QuantizedWeights"], [407, 1, 1, "", "ShuffleNet_V2_X2_0_QuantizedWeights"], [396, 3, 1, "", "googlenet"], [397, 3, 1, "", "inception_v3"], [398, 3, 1, "", "mobilenet_v2"], [399, 3, 1, "", "mobilenet_v3_large"], [400, 3, 1, "", "resnet18"], [401, 3, 1, "", "resnet50"], [402, 3, 1, "", "resnext101_32x8d"], [403, 3, 1, "", "resnext101_64x4d"], [404, 3, 1, "", "shufflenet_v2_x0_5"], [405, 3, 1, "", "shufflenet_v2_x1_0"], [406, 3, 1, "", "shufflenet_v2_x1_5"], [407, 3, 1, "", "shufflenet_v2_x2_0"]], "torchvision.models.segmentation": [[431, 1, 1, "", "DeepLabV3_MobileNet_V3_Large_Weights"], [432, 1, 1, "", "DeepLabV3_ResNet101_Weights"], [433, 1, 1, "", "DeepLabV3_ResNet50_Weights"], [434, 1, 1, "", "FCN_ResNet101_Weights"], [435, 1, 1, "", "FCN_ResNet50_Weights"], [436, 1, 1, "", "LRASPP_MobileNet_V3_Large_Weights"], [431, 3, 1, "", "deeplabv3_mobilenet_v3_large"], [432, 3, 1, "", "deeplabv3_resnet101"], [433, 3, 1, "", "deeplabv3_resnet50"], [434, 3, 1, "", "fcn_resnet101"], [435, 3, 1, "", "fcn_resnet50"], [436, 3, 1, "", "lraspp_mobilenet_v3_large"]], "torchvision.models.video": [[457, 1, 1, "", "MC3_18_Weights"], [458, 1, 1, "", "MViT_V1_B_Weights"], [459, 1, 1, "", "MViT_V2_S_Weights"], [460, 1, 1, "", "R2Plus1D_18_Weights"], [461, 1, 1, "", "R3D_18_Weights"], [462, 1, 1, "", "S3D_Weights"], [463, 1, 1, "", "Swin3D_B_Weights"], [464, 1, 1, "", "Swin3D_S_Weights"], [465, 1, 1, "", "Swin3D_T_Weights"], [457, 3, 1, "", "mc3_18"], [458, 3, 1, "", "mvit_v1_b"], [459, 3, 1, "", "mvit_v2_s"], [460, 3, 1, "", "r2plus1d_18"], [461, 3, 1, "", "r3d_18"], [462, 3, 1, "", "s3d"], [463, 3, 1, "", "swin3d_b"], [464, 3, 1, "", "swin3d_s"], [465, 3, 1, "", "swin3d_t"]], "torchvision.ops": [[116, 1, 1, "", "Conv2dNormActivation"], [117, 1, 1, "", "Conv3dNormActivation"], [118, 1, 1, "", "DeformConv2d"], [119, 1, 1, "", "DropBlock2d"], [120, 1, 1, "", "DropBlock3d"], [121, 1, 1, "", "FeaturePyramidNetwork"], [122, 1, 1, "", "FrozenBatchNorm2d"], [123, 1, 1, "", "MLP"], [124, 1, 1, "", "MultiScaleRoIAlign"], [125, 1, 1, "", "PSRoIAlign"], [126, 1, 1, "", "PSRoIPool"], [127, 1, 1, "", "Permute"], [128, 1, 1, "", "RoIAlign"], [129, 1, 1, "", "RoIPool"], [130, 1, 1, "", "SqueezeExcitation"], [131, 1, 1, "", "StochasticDepth"], [132, 3, 1, "", "batched_nms"], [133, 3, 1, "", "box_area"], [134, 3, 1, "", "box_convert"], [135, 3, 1, "", "box_iou"], [136, 3, 1, "", "clip_boxes_to_image"], [137, 3, 1, "", "complete_box_iou"], [138, 3, 1, "", "complete_box_iou_loss"], [139, 3, 1, "", "deform_conv2d"], [140, 3, 1, "", "distance_box_iou"], [141, 3, 1, "", "distance_box_iou_loss"], [142, 3, 1, "", "drop_block2d"], [143, 3, 1, "", "drop_block3d"], [144, 3, 1, "", "generalized_box_iou"], [145, 3, 1, "", "generalized_box_iou_loss"], [146, 3, 1, "", "masks_to_boxes"], [147, 3, 1, "", "nms"], [148, 3, 1, "", "ps_roi_align"], [149, 3, 1, "", "ps_roi_pool"], [150, 3, 1, "", "remove_small_boxes"], [151, 3, 1, "", "roi_align"], [152, 3, 1, "", "roi_pool"], [153, 3, 1, "", "sigmoid_focal_loss"], [154, 3, 1, "", "stochastic_depth"]], "torchvision.ops.DeformConv2d": [[118, 2, 1, "", "forward"]], "torchvision.ops.DropBlock2d": [[119, 2, 1, "", "forward"]], "torchvision.ops.DropBlock3d": [[120, 2, 1, "", "forward"]], "torchvision.ops.FeaturePyramidNetwork": [[121, 2, 1, "", "forward"], [121, 2, 1, "", "get_result_from_inner_blocks"], [121, 2, 1, "", "get_result_from_layer_blocks"]], "torchvision.ops.FrozenBatchNorm2d": [[122, 2, 1, "", "forward"]], "torchvision.ops.MultiScaleRoIAlign": [[124, 2, 1, "", "forward"]], "torchvision.ops.PSRoIAlign": [[125, 2, 1, "", "forward"]], "torchvision.ops.PSRoIPool": [[126, 2, 1, "", "forward"]], "torchvision.ops.Permute": [[127, 2, 1, "", "forward"]], "torchvision.ops.RoIAlign": [[128, 2, 1, "", "forward"]], "torchvision.ops.RoIPool": [[129, 2, 1, "", "forward"]], "torchvision.ops.SqueezeExcitation": [[130, 2, 1, "", "forward"]], "torchvision.ops.StochasticDepth": [[131, 2, 1, "", "forward"]], "torchvision.transforms": [[155, 1, 1, "", "AugMix"], [156, 1, 1, "", "AutoAugment"], [157, 1, 1, "", "AutoAugmentPolicy"], [158, 1, 1, "", "CenterCrop"], [159, 1, 1, "", "ColorJitter"], [160, 1, 1, "", "Compose"], [161, 1, 1, "", "ConvertImageDtype"], [162, 1, 1, "", "ElasticTransform"], [163, 1, 1, "", "FiveCrop"], [164, 1, 1, "", "GaussianBlur"], [165, 1, 1, "", "Grayscale"], [166, 1, 1, "", "Lambda"], [167, 1, 1, "", "LinearTransformation"], [168, 1, 1, "", "Normalize"], [169, 1, 1, "", "PILToTensor"], [170, 1, 1, "", "Pad"], [171, 1, 1, "", "RandAugment"], [172, 1, 1, "", "RandomAdjustSharpness"], [173, 1, 1, "", "RandomAffine"], [174, 1, 1, "", "RandomApply"], [175, 1, 1, "", "RandomAutocontrast"], [176, 1, 1, "", "RandomChoice"], [177, 1, 1, "", "RandomCrop"], [178, 1, 1, "", "RandomEqualize"], [179, 1, 1, "", "RandomErasing"], [180, 1, 1, "", "RandomGrayscale"], [181, 1, 1, "", "RandomHorizontalFlip"], [182, 1, 1, "", "RandomInvert"], [183, 1, 1, "", "RandomOrder"], [184, 1, 1, "", "RandomPerspective"], [185, 1, 1, "", "RandomPosterize"], [186, 1, 1, "", "RandomResizedCrop"], [187, 1, 1, "", "RandomRotation"], [188, 1, 1, "", "RandomSolarize"], [189, 1, 1, "", "RandomVerticalFlip"], [190, 1, 1, "", "Resize"], [191, 1, 1, "", "TenCrop"], [192, 1, 1, "", "ToPILImage"], [193, 1, 1, "", "ToTensor"], [194, 1, 1, "", "TrivialAugmentWide"]], "torchvision.transforms.AugMix": [[155, 2, 1, "", "forward"]], "torchvision.transforms.AutoAugment": [[156, 2, 1, "", "forward"], [156, 2, 1, "", "get_params"]], "torchvision.transforms.CenterCrop": [[158, 2, 1, "", "forward"]], "torchvision.transforms.ColorJitter": [[159, 2, 1, "", "forward"], [159, 2, 1, "", "get_params"]], "torchvision.transforms.ConvertImageDtype": [[161, 2, 1, "", "forward"]], "torchvision.transforms.ElasticTransform": [[162, 2, 1, "", "forward"]], "torchvision.transforms.FiveCrop": [[163, 2, 1, "", "forward"]], "torchvision.transforms.GaussianBlur": [[164, 2, 1, "", "forward"], [164, 2, 1, "", "get_params"]], "torchvision.transforms.Grayscale": [[165, 2, 1, "", "forward"]], "torchvision.transforms.LinearTransformation": [[167, 2, 1, "", "forward"]], "torchvision.transforms.Normalize": [[168, 2, 1, "", "forward"]], "torchvision.transforms.Pad": [[170, 2, 1, "", "forward"]], "torchvision.transforms.RandAugment": [[171, 2, 1, "", "forward"]], "torchvision.transforms.RandomAdjustSharpness": [[172, 2, 1, "", "forward"]], "torchvision.transforms.RandomAffine": [[173, 2, 1, "", "forward"], [173, 2, 1, "", "get_params"]], "torchvision.transforms.RandomApply": [[174, 2, 1, "", "forward"]], "torchvision.transforms.RandomAutocontrast": [[175, 2, 1, "", "forward"]], "torchvision.transforms.RandomCrop": [[177, 2, 1, "", "forward"], [177, 2, 1, "", "get_params"]], "torchvision.transforms.RandomEqualize": [[178, 2, 1, "", "forward"]], "torchvision.transforms.RandomErasing": [[179, 2, 1, "", "forward"], [179, 2, 1, "", "get_params"]], "torchvision.transforms.RandomGrayscale": [[180, 2, 1, "", "forward"]], "torchvision.transforms.RandomHorizontalFlip": [[181, 2, 1, "", "forward"]], "torchvision.transforms.RandomInvert": [[182, 2, 1, "", "forward"]], "torchvision.transforms.RandomPerspective": [[184, 2, 1, "", "forward"], [184, 2, 1, "", "get_params"]], "torchvision.transforms.RandomPosterize": [[185, 2, 1, "", "forward"]], "torchvision.transforms.RandomResizedCrop": [[186, 2, 1, "", "forward"], [186, 2, 1, "", "get_params"]], "torchvision.transforms.RandomRotation": [[187, 2, 1, "", "forward"], [187, 2, 1, "", "get_params"]], "torchvision.transforms.RandomSolarize": [[188, 2, 1, "", "forward"]], "torchvision.transforms.RandomVerticalFlip": [[189, 2, 1, "", "forward"]], "torchvision.transforms.Resize": [[190, 2, 1, "", "forward"]], "torchvision.transforms.TenCrop": [[191, 2, 1, "", "forward"]], "torchvision.transforms.TrivialAugmentWide": [[194, 2, 1, "", "forward"]], "torchvision.transforms.functional": [[195, 3, 1, "", "adjust_brightness"], [196, 3, 1, "", "adjust_contrast"], [197, 3, 1, "", "adjust_gamma"], [198, 3, 1, "", "adjust_hue"], [199, 3, 1, "", "adjust_saturation"], [200, 3, 1, "", "adjust_sharpness"], [201, 3, 1, "", "affine"], [202, 3, 1, "", "autocontrast"], [203, 3, 1, "", "center_crop"], [204, 3, 1, "", "convert_image_dtype"], [205, 3, 1, "", "crop"], [206, 3, 1, "", "equalize"], [207, 3, 1, "", "erase"], [208, 3, 1, "", "five_crop"], [209, 3, 1, "", "gaussian_blur"], [210, 3, 1, "", "get_dimensions"], [211, 3, 1, "", "get_image_num_channels"], [212, 3, 1, "", "get_image_size"], [213, 3, 1, "", "hflip"], [214, 3, 1, "", "invert"], [215, 3, 1, "", "normalize"], [216, 3, 1, "", "pad"], [217, 3, 1, "", "perspective"], [218, 3, 1, "", "pil_to_tensor"], [219, 3, 1, "", "posterize"], [220, 3, 1, "", "resize"], [221, 3, 1, "", "resized_crop"], [222, 3, 1, "", "rgb_to_grayscale"], [223, 3, 1, "", "rotate"], [224, 3, 1, "", "solarize"], [225, 3, 1, "", "ten_crop"], [226, 3, 1, "", "to_grayscale"], [227, 3, 1, "", "to_pil_image"], [228, 3, 1, "", "to_tensor"], [229, 3, 1, "", "vflip"]], "torchvision.transforms.v2": [[230, 1, 1, "", "AugMix"], [231, 1, 1, "", "AutoAugment"], [232, 1, 1, "", "CenterCrop"], [233, 1, 1, "", "ClampBoundingBoxes"], [234, 1, 1, "", "ColorJitter"], [235, 1, 1, "", "Compose"], [236, 1, 1, "", "ConvertBoundingBoxFormat"], [237, 1, 1, "", "ConvertImageDtype"], [238, 1, 1, "", "CutMix"], [239, 1, 1, "", "ElasticTransform"], [240, 1, 1, "", "FiveCrop"], [241, 1, 1, "", "GaussianBlur"], [242, 1, 1, "", "Grayscale"], [243, 1, 1, "", "Lambda"], [244, 1, 1, "", "LinearTransformation"], [245, 1, 1, "", "MixUp"], [246, 1, 1, "", "Normalize"], [247, 1, 1, "", "PILToTensor"], [248, 1, 1, "", "Pad"], [249, 1, 1, "", "RandAugment"], [250, 1, 1, "", "RandomAdjustSharpness"], [251, 1, 1, "", "RandomAffine"], [252, 1, 1, "", "RandomApply"], [253, 1, 1, "", "RandomAutocontrast"], [254, 1, 1, "", "RandomChannelPermutation"], [255, 1, 1, "", "RandomChoice"], [256, 1, 1, "", "RandomCrop"], [257, 1, 1, "", "RandomEqualize"], [258, 1, 1, "", "RandomErasing"], [259, 1, 1, "", "RandomGrayscale"], [260, 1, 1, "", "RandomHorizontalFlip"], [261, 1, 1, "", "RandomInvert"], [262, 1, 1, "", "RandomIoUCrop"], [263, 1, 1, "", "RandomOrder"], [264, 1, 1, "", "RandomPerspective"], [265, 1, 1, "", "RandomPhotometricDistort"], [266, 1, 1, "", "RandomPosterize"], [267, 1, 1, "", "RandomResize"], [268, 1, 1, "", "RandomResizedCrop"], [269, 1, 1, "", "RandomRotation"], [270, 1, 1, "", "RandomShortestSize"], [271, 1, 1, "", "RandomSolarize"], [272, 1, 1, "", "RandomVerticalFlip"], [273, 1, 1, "", "RandomZoomOut"], [274, 1, 1, "", "Resize"], [275, 1, 1, "", "SanitizeBoundingBoxes"], [276, 1, 1, "", "ScaleJitter"], [277, 1, 1, "", "TenCrop"], [278, 1, 1, "", "ToDtype"], [279, 1, 1, "", "ToImage"], [280, 1, 1, "", "ToPILImage"], [281, 1, 1, "", "ToPureTensor"], [282, 1, 1, "", "ToTensor"], [283, 1, 1, "", "TrivialAugmentWide"], [284, 1, 1, "", "UniformTemporalSubsample"]], "torchvision.transforms.v2.AugMix": [[230, 2, 1, "", "forward"]], "torchvision.transforms.v2.AutoAugment": [[231, 2, 1, "", "forward"], [231, 2, 1, "", "get_params"]], "torchvision.transforms.v2.ColorJitter": [[234, 2, 1, "", "get_params"]], "torchvision.transforms.v2.Compose": [[235, 2, 1, "", "extra_repr"], [235, 2, 1, "", "forward"]], "torchvision.transforms.v2.GaussianBlur": [[241, 2, 1, "", "get_params"]], "torchvision.transforms.v2.Lambda": [[243, 2, 1, "", "extra_repr"]], "torchvision.transforms.v2.RandAugment": [[249, 2, 1, "", "forward"]], "torchvision.transforms.v2.RandomAffine": [[251, 2, 1, "", "get_params"]], "torchvision.transforms.v2.RandomApply": [[252, 2, 1, "", "extra_repr"], [252, 2, 1, "", "forward"]], "torchvision.transforms.v2.RandomChoice": [[255, 2, 1, "", "forward"]], "torchvision.transforms.v2.RandomCrop": [[256, 2, 1, "", "get_params"]], "torchvision.transforms.v2.RandomErasing": [[258, 2, 1, "", "get_params"]], "torchvision.transforms.v2.RandomOrder": [[263, 2, 1, "", "forward"]], "torchvision.transforms.v2.RandomPerspective": [[264, 2, 1, "", "get_params"]], "torchvision.transforms.v2.RandomResizedCrop": [[268, 2, 1, "", "get_params"]], "torchvision.transforms.v2.RandomRotation": [[269, 2, 1, "", "get_params"]], "torchvision.transforms.v2.SanitizeBoundingBoxes": [[275, 2, 1, "", "forward"]], "torchvision.transforms.v2.TrivialAugmentWide": [[283, 2, 1, "", "forward"]], "torchvision.transforms.v2.functional": [[285, 3, 1, "", "adjust_brightness"], [286, 3, 1, "", "adjust_contrast"], [287, 3, 1, "", "adjust_gamma"], [288, 3, 1, "", "adjust_hue"], [289, 3, 1, "", "adjust_saturation"], [290, 3, 1, "", "adjust_sharpness"], [291, 3, 1, "", "affine"], [292, 3, 1, "", "autocontrast"], [293, 3, 1, "", "center_crop"], [294, 3, 1, "", "clamp_bounding_boxes"], [295, 3, 1, "", "convert_bounding_box_format"], [296, 1, 1, "", "convert_image_dtype"], [297, 3, 1, "", "crop"], [298, 3, 1, "", "elastic"], [299, 3, 1, "", "equalize"], [300, 3, 1, "", "erase"], [301, 3, 1, "", "five_crop"], [302, 3, 1, "", "gaussian_blur"], [303, 3, 1, "", "horizontal_flip"], [304, 3, 1, "", "invert"], [305, 3, 1, "", "normalize"], [306, 3, 1, "", "pad"], [307, 3, 1, "", "permute_channels"], [308, 3, 1, "", "perspective"], [309, 3, 1, "", "pil_to_tensor"], [310, 3, 1, "", "posterize"], [311, 3, 1, "", "register_kernel"], [312, 3, 1, "", "resize"], [313, 3, 1, "", "resized_crop"], [314, 3, 1, "", "rgb_to_grayscale"], [315, 3, 1, "", "rotate"], [316, 3, 1, "", "solarize"], [317, 3, 1, "", "ten_crop"], [318, 3, 1, "", "to_dtype"], [319, 3, 1, "", "to_grayscale"], [320, 3, 1, "", "to_image"], [321, 3, 1, "", "to_pil_image"], [322, 1, 1, "", "to_tensor"], [323, 3, 1, "", "uniform_temporal_subsample"], [324, 3, 1, "", "vertical_flip"]], "torchvision.tv_tensors": [[325, 1, 1, "", "BoundingBoxFormat"], [326, 1, 1, "", "BoundingBoxes"], [327, 1, 1, "", "Image"], [328, 1, 1, "", "Mask"], [329, 1, 1, "", "TVTensor"], [330, 1, 1, "", "Video"], [331, 1, 1, "", "set_return_type"], [332, 1, 1, "", "wrap"]], "torchvision.tv_tensors.TVTensor": [[329, 4, 1, "", "device"], [329, 4, 1, "", "ndim"], [329, 4, 1, "", "shape"]], "torchvision.utils": [[333, 3, 1, "", "draw_bounding_boxes"], [334, 3, 1, "", "draw_keypoints"], [335, 3, 1, "", "draw_segmentation_masks"], [336, 3, 1, "", "flow_to_image"], [337, 3, 1, "", "make_grid"], [338, 3, 1, "", "save_image"]]}, "objtypes": {"0": "py:module", "1": "py:class", "2": "py:method", "3": "py:function", "4": "py:property"}, "objnames": {"0": ["py", "module", "Python module"], "1": ["py", "class", "Python class"], "2": ["py", "method", "Python method"], "3": ["py", "function", "Python function"], "4": ["py", "property", "Python property"]}, "titleterms": {"exampl": [0, 5, 12, 340], "tutori": 0, "transform": [0, 8, 9, 12, 13, 14, 17, 295, 309, 318, 320, 321, 508], "other": [0, 1, 508], "optic": [2, 17, 342], "flow": [2, 17, 342], "predict": [2, 17], "movement": 2, "raft": [2, 486], "model": [2, 6, 18, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "read": 2, "video": [2, 5, 13, 17, 330, 341, 342, 500, 501, 502, 503], "us": [2, 11, 13, 15, 342, 508], "torchvis": [2, 295, 309, 318, 320, 321, 340], "estim": 2, "visual": [2, 5, 6], "bonu": 2, "creat": [2, 9], "gif": 2, "repurpos": 3, "mask": [3, 6, 328, 479], "bound": [3, 6], "box": [3, 6, 506], "convert": 3, "segment": [3, 6, 12, 13, 17, 342, 506], "dataset": [3, 5, 12, 13, 17], "detect": [3, 12, 13, 17, 342, 506], "torchscript": [4, 508], "support": [4, 9, 508], "api": [5, 18, 341, 508], "1": 5, "introduct": 5, "build": 5, "new": 5, "object": [5, 12, 342], "examin": 5, "properti": 5, "2": 5, "sampl": 5, "read_video": [5, 104], "function": [5, 11, 295, 309, 318, 320, 321, 508], "3": 5, "an": 5, "randomli": [5, 14], "can": [5, 15], "appli": [5, 14], "train": [5, 12, 340, 342, 507], "kinetics400": 5, "4": 5, "data": [5, 12], "util": [6, 510], "grid": 6, "imag": [6, 13, 17, 327, 341, 508], "semant": [6, 342], "instanc": [6, 342], "keypoint": [6, 342, 477], "comput": [7, 16], "time": [7, 16], "how": [9, 10, 11, 15], "write": [9, 10], "your": [9, 10, 13], "own": [9, 10, 13], "v2": [9, 12, 13, 17, 295, 309, 318, 320, 321, 482, 483, 493, 494, 508], "just": [9, 13], "nn": 9, "modul": 9, "overrid": 9, "forward": [9, 10], "method": 9, "arbitrari": 9, "input": [9, 11, 13, 508], "structur": 9, "tvtensor": [10, 13, 15, 329, 509], "class": [10, 17, 508], "paramet": 10, "ensur": 10, "futur": 10, "compat": 10, "kernel": [10, 508], "cutmix": [11, 238, 508], "mixup": [11, 245, 508], "pre": [11, 342], "process": 11, "pipelin": 11, "where": 11, "after": 11, "dataload": 11, "As": 11, "part": 11, "collat": 11, "non": 11, "standard": 11, "format": 11, "end": 12, "prepar": 12, "load": 12, "loop": 12, "refer": [12, 18, 340, 507, 508], "get": 13, "start": [13, 508], "The": 13, "basic": 13, "i": [13, 15, 508], "want": [13, 15], "do": [13, 15], "classif": [13, 17, 342], "what": [13, 15], "ar": [13, 15], "pass": 13, "intercompat": 13, "built": [13, 17], "illustr": 14, "geometr": 14, "pad": [14, 170, 216, 248, 306], "resiz": [14, 190, 220, 274, 312, 508], "centercrop": [14, 158, 232], "fivecrop": [14, 163, 240], "randomperspect": [14, 184, 264], "randomrot": [14, 187, 269], "randomaffin": [14, 173, 251], "elastictransform": [14, 162, 239], "randomcrop": [14, 177, 256], "randomresizedcrop": [14, 186, 268], "photometr": 14, "grayscal": [14, 165, 242], "colorjitt": [14, 159, 234], "gaussianblur": [14, 164, 241], "randominvert": [14, 182, 261], "randomposter": [14, 185, 266], "randomsolar": [14, 188, 271], "randomadjustsharp": [14, 172, 250], "randomautocontrast": [14, 175, 253], "randomequ": [14, 178, 257], "augment": [14, 508], "autoaug": [14, 156, 231], "randaug": [14, 171, 249], "trivialaugmentwid": [14, 194, 283], "augmix": [14, 155, 230], "randomhorizontalflip": [14, 181, 260], "randomverticalflip": [14, 189, 272], "randomappli": [14, 174, 252], "faq": 15, "construct": 15, "constructor": 15, "tv_tensor": 15, "wrap": [15, 332], "had": 15, "now": 15, "have": 15, "tensor": 15, "help": 15, "But": 15, "back": 15, "why": 15, "thi": 15, "happen": 15, "except": 15, "stereo": 17, "match": 17, "pair": 17, "caption": 17, "base": 17, "custom": 17, "featur": 18, "extract": 18, "inspect": 18, "cifar10": 25, "cifar100": 26, "crestereo": 27, "caltech101": 28, "caltech256": 29, "carlastereo": 30, "celeba": 31, "cityscap": 32, "cococapt": 33, "cocodetect": 34, "country211": 35, "dtd": 36, "datasetfold": 37, "emnist": 38, "eth3dstereo": 39, "eurosat": 40, "fer2013": 41, "fgvcaircraft": 42, "fakedata": 43, "fallingthingsstereo": 44, "fashionmnist": 45, "flickr30k": 46, "flickr8k": 47, "flowers102": 48, "flyingchair": 49, "flyingthings3d": 50, "food101": 51, "gtsrb": 52, "hd1k": 53, "hmdb51": 54, "inaturalist": 55, "imagefold": 56, "imagenet": 57, "instereo2k": 58, "kmnist": 59, "kinet": 60, "kitti": 61, "kitti2012stereo": 62, "kitti2015stereo": 63, "kittiflow": 64, "lfwpair": 65, "lfwpeopl": 66, "lsun": 67, "mnist": 68, "middlebury2014stereo": 69, "movingmnist": 70, "omniglot": 71, "oxfordiiitpet": 72, "pcam": 73, "phototour": 74, "places365": 75, "qmnist": 76, "renderedsst2": 77, "sbdataset": 78, "sbu": 79, "semeion": 80, "stl10": 81, "sun397": 82, "svhn": 83, "sceneflowstereo": 84, "sintel": 85, "sintelstereo": 86, "stanfordcar": 87, "ucf101": 88, "usp": 89, "vocdetect": 90, "vocsegment": 91, "visiondataset": 92, "widerfac": 93, "wrap_dataset_for_transforms_v2": 94, "imagereadmod": 95, "videoread": 96, "decode_imag": 97, "decode_jpeg": 98, "decode_png": 99, "encode_jpeg": 100, "encode_png": 101, "read_fil": 102, "read_imag": 103, "read_video_timestamp": 105, "write_fil": 106, "write_jpeg": 107, "write_png": 108, "write_video": 109, "create_feature_extractor": 110, "get_graph_node_nam": 111, "get_model": 112, "get_model_weight": 113, "get_weight": 114, "list_model": 115, "conv2dnormactiv": 116, "conv3dnormactiv": 117, "deformconv2d": 118, "dropblock2d": 119, "dropblock3d": 120, "featurepyramidnetwork": 121, "frozenbatchnorm2d": 122, "mlp": 123, "multiscaleroialign": 124, "psroialign": 125, "psroipool": 126, "permut": 127, "roialign": 128, "roipool": 129, "squeezeexcit": 130, "stochasticdepth": 131, "batched_nm": 132, "box_area": 133, "box_convert": 134, "box_iou": 135, "clip_boxes_to_imag": 136, "complete_box_i": 137, "complete_box_iou_loss": 138, "deform_conv2d": 139, "distance_box_i": 140, "distance_box_iou_loss": 141, "drop_block2d": 142, "drop_block3d": 143, "generalized_box_i": 144, "generalized_box_iou_loss": 145, "masks_to_box": 146, "nm": 147, "ps_roi_align": 148, "ps_roi_pool": 149, "remove_small_box": 150, "roi_align": 151, "roi_pool": 152, "sigmoid_focal_loss": 153, "stochastic_depth": 154, "autoaugmentpolici": 157, "compos": [160, 235], "convertimagedtyp": [161, 237], "lambda": [166, 243], "lineartransform": [167, 244], "normal": [168, 215, 246, 305], "piltotensor": [169, 247], "randomchoic": [176, 255], "randomeras": [179, 258], "randomgrayscal": [180, 259], "randomord": [183, 263], "tencrop": [191, 277], "topilimag": [192, 280], "totensor": [193, 282], "adjust_bright": [195, 285], "adjust_contrast": [196, 286], "adjust_gamma": [197, 287], "adjust_hu": [198, 288], "adjust_satur": [199, 289], "adjust_sharp": [200, 290], "affin": [201, 291], "autocontrast": [202, 292], "center_crop": [203, 293], "convert_image_dtyp": [204, 296], "crop": [205, 297, 508], "equal": [206, 299], "eras": [207, 300], "five_crop": [208, 301], "gaussian_blur": [209, 302], "get_dimens": 210, "get_image_num_channel": 211, "get_image_s": 212, "hflip": 213, "invert": [214, 304], "perspect": [217, 308], "pil_to_tensor": [218, 309], "poster": [219, 310], "resized_crop": [221, 313], "rgb_to_grayscal": [222, 314], "rotat": [223, 315], "solar": [224, 316], "ten_crop": [225, 317], "to_grayscal": [226, 319], "to_pil_imag": [227, 321], "to_tensor": [228, 322], "vflip": 229, "clampboundingbox": 233, "convertboundingboxformat": 236, "randomchannelpermut": 254, "randomioucrop": 262, "randomphotometricdistort": 265, "randomres": 267, "randomshortests": 270, "randomzoomout": 273, "sanitizeboundingbox": 275, "scalejitt": 276, "todtyp": 278, "toimag": 279, "topuretensor": 281, "uniformtemporalsubsampl": 284, "clamp_bounding_box": 294, "convert_bounding_box_format": 295, "elast": 298, "horizontal_flip": 303, "permute_channel": 307, "register_kernel": 311, "to_dtyp": 318, "to_imag": 320, "uniform_temporal_subsampl": 323, "vertical_flip": 324, "boundingboxformat": 325, "boundingbox": 326, "set_return_typ": 331, "draw_bounding_box": 333, "draw_keypoint": 334, "draw_segmentation_mask": 335, "flow_to_imag": 336, "make_grid": 337, "save_imag": 338, "packag": 340, "pytorch": 340, "librari": 340, "indic": 340, "decod": 341, "encod": 341, "fine": 341, "grain": 341, "weight": 342, "gener": 342, "inform": 342, "initi": 342, "list": 342, "retriev": 342, "avail": 342, "from": 342, "hub": 342, "tabl": 342, "all": 342, "quantiz": [342, 474, 476, 483, 485, 489, 491, 494], "person": 342, "alexnet": [343, 352], "builder": [343, 344, 345, 346, 347, 348, 349, 350, 351, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505], "convnext": 344, "deeplabv3": 345, "densenet": 346, "efficientnet": 347, "efficientnetv2": 348, "faster": 349, "r": [349, 477, 479], "cnn": [349, 477, 479], "fcn": 350, "fco": 351, "convnext_bas": 353, "convnext_larg": 354, "convnext_smal": 355, "convnext_tini": 356, "densenet121": 357, "densenet161": 358, "densenet169": 359, "densenet201": 360, "fasterrcnn_mobilenet_v3_large_320_fpn": 361, "fasterrcnn_mobilenet_v3_large_fpn": 362, "fasterrcnn_resnet50_fpn": 363, "fasterrcnn_resnet50_fpn_v2": 364, "fcos_resnet50_fpn": 365, "keypointrcnn_resnet50_fpn": 366, "maskrcnn_resnet50_fpn": 367, "maskrcnn_resnet50_fpn_v2": 368, "retinanet_resnet50_fpn": 369, "retinanet_resnet50_fpn_v2": 370, "ssd300_vgg16": 371, "ssdlite320_mobilenet_v3_larg": 372, "efficientnet_b0": 373, "efficientnet_b1": 374, "efficientnet_b2": 375, "efficientnet_b3": 376, "efficientnet_b4": 377, "efficientnet_b5": 378, "efficientnet_b6": 379, "efficientnet_b7": 380, "efficientnet_v2_l": 381, "efficientnet_v2_m": 382, "efficientnet_v2_": 383, "googlenet": [384, 396, 473, 474], "inception_v3": [385, 397], "maxvit_t": 386, "mnasnet0_5": 387, "mnasnet0_75": 388, "mnasnet1_0": 389, "mnasnet1_3": 390, "mobilenet_v2": [391, 398], "mobilenet_v3_larg": [392, 399], "mobilenet_v3_smal": 393, "raft_larg": 394, "raft_smal": 395, "resnet18": [400, 425], "resnet50": [401, 427], "resnext101_32x8d": [402, 428], "resnext101_64x4d": [403, 429], "shufflenet_v2_x0_5": [404, 437], "shufflenet_v2_x1_0": [405, 438], "shufflenet_v2_x1_5": [406, 439], "shufflenet_v2_x2_0": [407, 440], "regnet_x_16gf": 408, "regnet_x_1_6gf": 409, "regnet_x_32gf": 410, "regnet_x_3_2gf": 411, "regnet_x_400mf": 412, "regnet_x_800mf": 413, "regnet_x_8gf": 414, "regnet_y_128gf": 415, "regnet_y_16gf": 416, "regnet_y_1_6gf": 417, "regnet_y_32gf": 418, "regnet_y_3_2gf": 419, "regnet_y_400mf": 420, "regnet_y_800mf": 421, "regnet_y_8gf": 422, "resnet101": 423, "resnet152": 424, "resnet34": 426, "resnext50_32x4d": 430, "deeplabv3_mobilenet_v3_larg": 431, "deeplabv3_resnet101": 432, "deeplabv3_resnet50": 433, "fcn_resnet101": 434, "fcn_resnet50": 435, "lraspp_mobilenet_v3_larg": 436, "squeezenet1_0": 441, "squeezenet1_1": 442, "swin_b": 443, "swin_": 444, "swin_t": 445, "swin_v2_b": 446, "swin_v2_": 447, "swin_v2_t": 448, "vgg11": 449, "vgg11_bn": 450, "vgg13": 451, "vgg13_bn": 452, "vgg16": 453, "vgg16_bn": 454, "vgg19": 455, "vgg19_bn": 456, "mc3_18": 457, "mvit_v1_b": 458, "mvit_v2_": 459, "r2plus1d_18": 460, "r3d_18": 461, "s3d": [462, 502], "swin3d_b": 463, "swin3d_": 464, "swin3d_t": 465, "vit_b_16": 466, "vit_b_32": 467, "vit_h_14": 468, "vit_l_16": 469, "vit_l_32": 470, "wide_resnet101_2": 471, "wide_resnet50_2": 472, "incept": 475, "v3": [475, 484, 485], "inceptionv3": 476, "lraspp": 478, "maxvit": 480, "mnasnet": 481, "mobilenet": [482, 483, 484, 485], "regnet": 487, "resnet": [488, 489, 501, 505], "resnext": [490, 491], "retinanet": 492, "shufflenet": [493, 494], "squeezenet": 495, "ssd": 496, "ssdlite": 497, "swintransform": [498, 503], "vgg": 499, "mvit": 500, "visiontransform": 504, "wide": 505, "oper": 506, "loss": 506, "layer": 506, "here": 508, "type": 508, "convent": 508, "dtype": 508, "expect": 508, "valu": 508, "rang": 508, "v1": 508, "which": 508, "one": 508, "should": 508, "perform": 508, "consider": 508, "recommend": 508, "geometri": 508, "color": 508, "composit": 508, "miscellan": 508, "convers": 508, "auto": 508, "develop": 508, "tool": 508}, "envversion": {"sphinx.domains.c": 2, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 6, "sphinx.domains.index": 1, "sphinx.domains.javascript": 2, "sphinx.domains.math": 2, "sphinx.domains.python": 3, "sphinx.domains.rst": 2, "sphinx.domains.std": 2, "sphinx.ext.intersphinx": 1, "sphinx.ext.todo": 2, "sphinx.ext.viewcode": 1, "sphinx": 56}})
\ No newline at end of file
diff --git a/0.16/training_references.html b/0.16/training_references.html
new file mode 100644
index 00000000000..ea6a452a208
--- /dev/null
+++ b/0.16/training_references.html
@@ -0,0 +1,706 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Training references &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="prev" title="Visualization utilities" href="auto_examples/others/plot_visualization_utils.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Training references</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/training_references.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="training-references">
+<h1>Training references<a class="headerlink" href="#training-references" title="Permalink to this heading">¶</a></h1>
+<p>On top of the many models, datasets, and image transforms, Torchvision also
+provides training reference scripts. These are the scripts that we use to train
+the <a class="reference internal" href="models.html#models"><span class="std std-ref">models</span></a> which are then available with pre-trained weights.</p>
+<p>These scripts are not part of the core package and are instead available <a class="reference external" href="https://github.com/pytorch/vision/tree/main/references">on
+GitHub</a>. We currently
+provide references for
+<a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/classification">classification</a>,
+<a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/detection">detection</a>,
+<a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/segmentation">segmentation</a>,
+<a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/similarity">similarity learning</a>,
+and <a class="reference external" href="https://github.com/pytorch/vision/tree/main/references/video_classification">video classification</a>.</p>
+<p>While these scripts are largely stable, they do not offer backward compatibility
+guarantees.</p>
+<p>In general, these scripts rely on the latest (not yet released) pytorch version
+or the latest torchvision version. This means that to use them, <strong>you might need
+to install the latest pytorch and torchvision versions</strong>, with e.g.:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">conda</span> <span class="n">install</span> <span class="n">pytorch</span> <span class="n">torchvision</span> <span class="o">-</span><span class="n">c</span> <span class="n">pytorch</span><span class="o">-</span><span class="n">nightly</span>
+</pre></div>
+</div>
+<p>If you need to rely on an older stable version of pytorch or torchvision, e.g.
+torchvision 0.10, then it’s safer to use the scripts from that corresponding
+release on GitHub, namely
+<a class="reference external" href="https://github.com/pytorch/vision/tree/v0.10.0/references">https://github.com/pytorch/vision/tree/v0.10.0/references</a>.</p>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+      
+        <a href="auto_examples/others/plot_visualization_utils.html" class="btn btn-neutral" title="Visualization utilities" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Training references</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/transforms.html b/0.16/transforms.html
new file mode 100644
index 00000000000..4d726a501fa
--- /dev/null
+++ b/0.16/transforms.html
@@ -0,0 +1,1648 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Transforming and augmenting images &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="Resize" href="generated/torchvision.transforms.v2.Resize.html" />
+    <link rel="prev" title="torchvision" href="index.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Transforming and augmenting images</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/transforms.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="transforming-and-augmenting-images">
+<span id="transforms"></span><h1>Transforming and augmenting images<a class="headerlink" href="#transforming-and-augmenting-images" title="Permalink to this heading">¶</a></h1>
+<p>Torchvision supports common computer vision transformations in the
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms</span></code> and <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> modules. Transforms
+can be used to transform or augment data for training or inference of different
+tasks (image classification, detection, segmentation, video classification).</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Image Classification</span>
+<span class="kn">import</span> <span class="nn">torch</span>
+<span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+
+<span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="mi">32</span><span class="p">,</span> <span class="mi">32</span>
+<span class="n">img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+
+<span class="n">transforms</span> <span class="o">=</span> <span class="n">v2</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">RandomResizedCrop</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">RandomHorizontalFlip</span><span class="p">(</span><span class="n">p</span><span class="o">=</span><span class="mf">0.5</span><span class="p">),</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">ToDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">Normalize</span><span class="p">(</span><span class="n">mean</span><span class="o">=</span><span class="p">[</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">],</span> <span class="n">std</span><span class="o">=</span><span class="p">[</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">]),</span>
+<span class="p">])</span>
+<span class="n">img</span> <span class="o">=</span> <span class="n">transforms</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Detection (re-using imports and transforms from above)</span>
+<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">tv_tensors</span>
+
+<span class="n">img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">256</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">)</span>
+<span class="n">boxes</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">H</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">4</span><span class="p">))</span>
+<span class="n">boxes</span><span class="p">[:,</span> <span class="mi">2</span><span class="p">:]</span> <span class="o">+=</span> <span class="n">boxes</span><span class="p">[:,</span> <span class="p">:</span><span class="mi">2</span><span class="p">]</span>
+<span class="n">boxes</span> <span class="o">=</span> <span class="n">tv_tensors</span><span class="o">.</span><span class="n">BoundingBoxes</span><span class="p">(</span><span class="n">boxes</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;XYXY&quot;</span><span class="p">,</span> <span class="n">canvas_size</span><span class="o">=</span><span class="p">(</span><span class="n">H</span><span class="p">,</span> <span class="n">W</span><span class="p">))</span>
+
+<span class="c1"># The same transforms can be used!</span>
+<span class="n">img</span><span class="p">,</span> <span class="n">boxes</span> <span class="o">=</span> <span class="n">transforms</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="n">boxes</span><span class="p">)</span>
+<span class="c1"># And you can pass arbitrary input structures</span>
+<span class="n">output_dict</span> <span class="o">=</span> <span class="n">transforms</span><span class="p">({</span><span class="s2">&quot;image&quot;</span><span class="p">:</span> <span class="n">img</span><span class="p">,</span> <span class="s2">&quot;boxes&quot;</span><span class="p">:</span> <span class="n">boxes</span><span class="p">})</span>
+</pre></div>
+</div>
+<p>Transforms are typically passed as the <code class="docutils literal notranslate"><span class="pre">transform</span></code> or <code class="docutils literal notranslate"><span class="pre">transforms</span></code> argument
+to the <a class="reference internal" href="datasets.html#datasets"><span class="std std-ref">Datasets</span></a>.</p>
+<section id="start-here">
+<h2>Start here<a class="headerlink" href="#start-here" title="Permalink to this heading">¶</a></h2>
+<p>Whether you’re new to Torchvision transforms, or you’re already experienced with
+them, we encourage you to start with
+<a class="reference internal" href="auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a> in
+order to learn more about what can be done with the new v2 transforms.</p>
+<p>Then, browse the sections in below this page for general information and
+performance tips. The available transforms and functionals are listed in the
+<a class="reference internal" href="#v2-api-ref"><span class="std std-ref">API reference</span></a>.</p>
+<p>More information and tutorials can also be found in our <a class="reference internal" href="auto_examples/index.html#gallery"><span class="std std-ref">example gallery</span></a>, e.g. <a class="reference internal" href="auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a>
+or <a class="reference internal" href="auto_examples/transforms/plot_custom_transforms.html#sphx-glr-auto-examples-transforms-plot-custom-transforms-py"><span class="std std-ref">How to write your own v2 transforms</span></a>.</p>
+</section>
+<section id="supported-input-types-and-conventions">
+<span id="conventions"></span><h2>Supported input types and conventions<a class="headerlink" href="#supported-input-types-and-conventions" title="Permalink to this heading">¶</a></h2>
+<p>Most transformations accept both <a class="reference external" href="https://pillow.readthedocs.io">PIL</a> images
+and tensor inputs. Both CPU and CUDA tensors are supported.
+The result of both backends (PIL or Tensors) should be very
+close. In general, we recommend relying on the tensor backend <a class="reference internal" href="#transforms-perf"><span class="std std-ref">for
+performance</span></a>.  The <a class="reference internal" href="#conversion-transforms"><span class="std std-ref">conversion transforms</span></a> may be used to convert to and from PIL images, or for
+converting dtypes and ranges.</p>
+<p>Tensor image are expected to be of shape <code class="docutils literal notranslate"><span class="pre">(C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code>, where <code class="docutils literal notranslate"><span class="pre">C</span></code> is the
+number of channels, and <code class="docutils literal notranslate"><span class="pre">H</span></code> and <code class="docutils literal notranslate"><span class="pre">W</span></code> refer to height and width. Most
+transforms support batched tensor input. A batch of Tensor images is a tensor of
+shape <code class="docutils literal notranslate"><span class="pre">(N,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code>, where <code class="docutils literal notranslate"><span class="pre">N</span></code> is a number of images in the batch. The
+<a class="reference internal" href="#v1-or-v2"><span class="std std-ref">v2</span></a> transforms generally accept an arbitrary number of leading
+dimensions <code class="docutils literal notranslate"><span class="pre">(...,</span> <span class="pre">C,</span> <span class="pre">H,</span> <span class="pre">W)</span></code> and can handle batched images or batched videos.</p>
+<section id="dtype-and-expected-value-range">
+<span id="range-and-dtype"></span><h3>Dtype and expected value range<a class="headerlink" href="#dtype-and-expected-value-range" title="Permalink to this heading">¶</a></h3>
+<p>The expected range of the values of a tensor image is implicitly defined by
+the tensor dtype. Tensor images with a float dtype are expected to have
+values in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">1]</span></code>. Tensor images with an integer dtype are expected to
+have values in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">MAX_DTYPE]</span></code> where <code class="docutils literal notranslate"><span class="pre">MAX_DTYPE</span></code> is the largest value
+that can be represented in that dtype. Typically, images of dtype
+<code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code> are expected to have values in <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">255]</span></code>.</p>
+<p>Use <a class="reference internal" href="generated/torchvision.transforms.v2.ToDtype.html#torchvision.transforms.v2.ToDtype" title="torchvision.transforms.v2.ToDtype"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToDtype</span></code></a> to convert both the dtype and
+range of the inputs.</p>
+</section>
+</section>
+<section id="v1-or-v2-which-one-should-i-use">
+<span id="v1-or-v2"></span><h2>V1 or V2? Which one should I use?<a class="headerlink" href="#v1-or-v2-which-one-should-i-use" title="Permalink to this heading">¶</a></h2>
+<p><strong>TL;DR</strong> We recommending using the <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> transforms
+instead of those in <code class="docutils literal notranslate"><span class="pre">torchvision.transforms</span></code>. They’re faster and they can do
+more things. Just change the import and you should be good to go.</p>
+<p>In Torchvision 0.15 (March 2023), we released a new set of transforms available
+in the <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code> namespace. These transforms have a lot of
+advantages compared to the v1 ones (in <code class="docutils literal notranslate"><span class="pre">torchvision.transforms</span></code>):</p>
+<ul class="simple">
+<li><p>They can transform images <strong>but also</strong> bounding boxes, masks, or videos. This
+provides support for tasks beyond image classification: detection, segmentation,
+video classification, etc. See
+<a class="reference internal" href="auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a>
+and <a class="reference internal" href="auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py"><span class="std std-ref">Transforms v2: End-to-end object detection/segmentation example</span></a>.</p></li>
+<li><p>They support more transforms like <a class="reference internal" href="generated/torchvision.transforms.v2.CutMix.html#torchvision.transforms.v2.CutMix" title="torchvision.transforms.v2.CutMix"><code class="xref py py-class docutils literal notranslate"><span class="pre">CutMix</span></code></a>
+and <a class="reference internal" href="generated/torchvision.transforms.v2.MixUp.html#torchvision.transforms.v2.MixUp" title="torchvision.transforms.v2.MixUp"><code class="xref py py-class docutils literal notranslate"><span class="pre">MixUp</span></code></a>. See
+<a class="reference internal" href="auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a>.</p></li>
+<li><p>They’re <a class="reference internal" href="#transforms-perf"><span class="std std-ref">faster</span></a>.</p></li>
+<li><p>They support arbitrary input structures (dicts, lists, tuples, etc.).</p></li>
+<li><p>Future improvements and features will be added to the v2 transforms only.</p></li>
+</ul>
+<p>These transforms are <strong>fully backward compatible</strong> with the v1 ones, so if
+you’re already using tranforms from <code class="docutils literal notranslate"><span class="pre">torchvision.transforms</span></code>, all you need to
+do to is to update the import to <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code>. In terms of
+output, there might be negligible differences due to implementation differences.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>The v2 transforms are still BETA, but at this point we do not expect
+disruptive changes to be made to their public APIs. We’re planning to make
+them fully stable in version 0.17. Please submit any feedback you may have
+<a class="reference external" href="https://github.com/pytorch/vision/issues/6753">here</a>.</p>
+</div>
+</section>
+<section id="performance-considerations">
+<span id="transforms-perf"></span><h2>Performance considerations<a class="headerlink" href="#performance-considerations" title="Permalink to this heading">¶</a></h2>
+<p>We recommend the following guidelines to get the best performance out of the
+transforms:</p>
+<ul class="simple">
+<li><p>Rely on the v2 transforms from <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2</span></code></p></li>
+<li><p>Use tensors instead of PIL images</p></li>
+<li><p>Use <code class="docutils literal notranslate"><span class="pre">torch.uint8</span></code> dtype, especially for resizing</p></li>
+<li><p>Resize with bilinear or bicubic mode</p></li>
+</ul>
+<p>This is what a typical transform pipeline could look like:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision.transforms</span> <span class="kn">import</span> <span class="n">v2</span>
+<span class="n">transforms</span> <span class="o">=</span> <span class="n">v2</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">ToImage</span><span class="p">(),</span>  <span class="c1"># Convert to tensor, only needed if you had a PIL image</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">ToDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">uint8</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>  <span class="c1"># optional, most input are already uint8 at this point</span>
+    <span class="c1"># ...</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">RandomResizedCrop</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="p">(</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">antialias</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>  <span class="c1"># Or Resize(antialias=True)</span>
+    <span class="c1"># ...</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">ToDtype</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>  <span class="c1"># Normalize expects float input</span>
+    <span class="n">v2</span><span class="o">.</span><span class="n">Normalize</span><span class="p">(</span><span class="n">mean</span><span class="o">=</span><span class="p">[</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">],</span> <span class="n">std</span><span class="o">=</span><span class="p">[</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">]),</span>
+<span class="p">])</span>
+</pre></div>
+</div>
+<p>The above should give you the best performance in a typical training environment
+that relies on the <a class="reference external" href="https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.utils.data.DataLoader</span></code></a> with <code class="docutils literal notranslate"><span class="pre">num_workers</span> <span class="pre">&gt;</span>
+<span class="pre">0</span></code>.</p>
+<p>Transforms tend to be sensitive to the input strides / memory format. Some
+transforms will be faster with channels-first images while others prefer
+channels-last. Like <code class="docutils literal notranslate"><span class="pre">torch</span></code> operators, most transforms will preserve the
+memory format of the input, but this may not always be respected due to
+implementation details. You may want to experiment a bit if you’re chasing the
+very best performance.  Using <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.compile.html#torch.compile" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.compile()</span></code></a> on individual transforms may
+also help factoring out the memory format variable (e.g. on
+<a class="reference internal" href="generated/torchvision.transforms.v2.Normalize.html#torchvision.transforms.v2.Normalize" title="torchvision.transforms.v2.Normalize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Normalize</span></code></a>). Note that we’re talking about
+<strong>memory format</strong>, not <a class="reference internal" href="#conventions"><span class="std std-ref">tensor shape</span></a>.</p>
+<p>Note that resize transforms like <a class="reference internal" href="generated/torchvision.transforms.v2.Resize.html#torchvision.transforms.v2.Resize" title="torchvision.transforms.v2.Resize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Resize</span></code></a>
+and <a class="reference internal" href="generated/torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.v2.RandomResizedCrop" title="torchvision.transforms.v2.RandomResizedCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code></a> typically prefer
+channels-last input and tend <strong>not</strong> to benefit from <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.compile.html#torch.compile" title="(in PyTorch v2.0)"><code class="xref py py-func docutils literal notranslate"><span class="pre">torch.compile()</span></code></a> at
+this time.</p>
+</section>
+<section id="transform-classes-functionals-and-kernels">
+<span id="functional-transforms"></span><h2>Transform classes, functionals, and kernels<a class="headerlink" href="#transform-classes-functionals-and-kernels" title="Permalink to this heading">¶</a></h2>
+<p>Transforms are available as classes like
+<a class="reference internal" href="generated/torchvision.transforms.v2.Resize.html#torchvision.transforms.v2.Resize" title="torchvision.transforms.v2.Resize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Resize</span></code></a>, but also as functionals like
+<a class="reference internal" href="generated/torchvision.transforms.v2.functional.resize.html#torchvision.transforms.v2.functional.resize" title="torchvision.transforms.v2.functional.resize"><code class="xref py py-func docutils literal notranslate"><span class="pre">resize()</span></code></a> in the
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2.functional</span></code> namespace.
+This is very much like the <a class="reference external" href="https://pytorch.org/docs/stable/nn.html#module-torch.nn" title="(in PyTorch v2.0)"><code class="xref py py-mod docutils literal notranslate"><span class="pre">torch.nn</span></code></a> package which defines both classes
+and functional equivalents in <code class="xref py py-mod docutils literal notranslate"><span class="pre">torch.nn.functional</span></code>.</p>
+<p>The functionals support PIL images, pure tensors, or <a class="reference internal" href="tv_tensors.html#tv-tensors"><span class="std std-ref">TVTensors</span></a>, e.g. both <code class="docutils literal notranslate"><span class="pre">resize(image_tensor)</span></code> and <code class="docutils literal notranslate"><span class="pre">resize(boxes)</span></code> are
+valid.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Random transforms like <a class="reference internal" href="generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop" title="torchvision.transforms.v2.RandomCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomCrop</span></code></a> will
+randomly sample some parameter each time they’re called. Their functional
+counterpart (<a class="reference internal" href="generated/torchvision.transforms.v2.functional.crop.html#torchvision.transforms.v2.functional.crop" title="torchvision.transforms.v2.functional.crop"><code class="xref py py-func docutils literal notranslate"><span class="pre">crop()</span></code></a>) does not do
+any kind of random sampling and thus have a slighlty different
+parametrization. The <code class="docutils literal notranslate"><span class="pre">get_params()</span></code> class method of the transforms class
+can be used to perform parameter sampling when using the functional APIs.</p>
+</div>
+<p>The <code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2.functional</span></code> namespace also contains what we
+call the “kernels”. These are the low-level functions that implement the
+core functionalities for specific types, e.g. <code class="docutils literal notranslate"><span class="pre">resize_bounding_boxes</span></code> or
+<code class="docutils literal notranslate"><span class="pre">`resized_crop_mask</span></code>. They are public, although not documented. Check the
+<a class="reference external" href="https://github.com/pytorch/vision/blob/main/torchvision/transforms/v2/functional/__init__.py">code</a>
+to see which ones are available (note that those starting with a leading
+underscore are <strong>not</strong> public!). Kernels are only really useful if you want
+<a class="reference internal" href="#transforms-torchscript"><span class="std std-ref">torchscript support</span></a> for types like bounding
+boxes or masks.</p>
+</section>
+<section id="torchscript-support">
+<span id="transforms-torchscript"></span><h2>Torchscript support<a class="headerlink" href="#torchscript-support" title="Permalink to this heading">¶</a></h2>
+<p>Most transform classes and functionals support torchscript. For composing
+transforms, use <a class="reference external" href="https://pytorch.org/docs/stable/generated/torch.nn.Sequential.html#torch.nn.Sequential" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.Sequential</span></code></a> instead of
+<a class="reference internal" href="generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose" title="torchvision.transforms.v2.Compose"><code class="xref py py-class docutils literal notranslate"><span class="pre">Compose</span></code></a>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
+    <span class="n">CenterCrop</span><span class="p">(</span><span class="mi">10</span><span class="p">),</span>
+    <span class="n">Normalize</span><span class="p">((</span><span class="mf">0.485</span><span class="p">,</span> <span class="mf">0.456</span><span class="p">,</span> <span class="mf">0.406</span><span class="p">),</span> <span class="p">(</span><span class="mf">0.229</span><span class="p">,</span> <span class="mf">0.224</span><span class="p">,</span> <span class="mf">0.225</span><span class="p">)),</span>
+<span class="p">)</span>
+<span class="n">scripted_transforms</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">script</span><span class="p">(</span><span class="n">transforms</span><span class="p">)</span>
+</pre></div>
+</div>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>v2 transforms support torchscript, but if you call <code class="docutils literal notranslate"><span class="pre">torch.jit.script()</span></code> on
+a v2 <strong>class</strong> transform, you’ll actually end up with its (scripted) v1
+equivalent.  This may lead to slightly different results between the
+scripted and eager executions due to implementation differences between v1
+and v2.</p>
+<p>If you really need torchscript support for the v2 transforms, we recommend
+scripting the <strong>functionals</strong> from the
+<code class="docutils literal notranslate"><span class="pre">torchvision.transforms.v2.functional</span></code> namespace to avoid surprises.</p>
+</div>
+<p>Also note that the functionals only support torchscript for pure tensors, which
+are always treated as images. If you need torchscript support for other types
+like bounding boxes or masks, you can rely on the <a class="reference internal" href="#functional-transforms"><span class="std std-ref">low-level kernels</span></a>.</p>
+<p>For any custom transformations to be used with <code class="docutils literal notranslate"><span class="pre">torch.jit.script</span></code>, they should
+be derived from <code class="docutils literal notranslate"><span class="pre">torch.nn.Module</span></code>.</p>
+<p>See also: <a class="reference internal" href="auto_examples/others/plot_scripted_tensor_transforms.html#sphx-glr-auto-examples-others-plot-scripted-tensor-transforms-py"><span class="std std-ref">Torchscript support</span></a>.</p>
+</section>
+<section id="v2-api-reference-recommended">
+<span id="v2-api-ref"></span><h2>V2 API reference - Recommended<a class="headerlink" href="#v2-api-reference-recommended" title="Permalink to this heading">¶</a></h2>
+<section id="geometry">
+<h3>Geometry<a class="headerlink" href="#geometry" title="Permalink to this heading">¶</a></h3>
+<section id="resizing">
+<h4>Resizing<a class="headerlink" href="#resizing" title="Permalink to this heading">¶</a></h4>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.Resize.html#torchvision.transforms.v2.Resize" title="torchvision.transforms.v2.Resize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.Resize</span></code></a>(size[, interpolation, max_size, ...])</p></td>
+<td><p>[BETA] Resize the input to the given size.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ScaleJitter.html#torchvision.transforms.v2.ScaleJitter" title="torchvision.transforms.v2.ScaleJitter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ScaleJitter</span></code></a>(target_size[, scale_range, ...])</p></td>
+<td><p>[BETA] Perform Large Scale Jitter on the input according to <a class="reference external" href="https://arxiv.org/abs/2012.07177">&quot;Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation&quot;</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomShortestSize.html#torchvision.transforms.v2.RandomShortestSize" title="torchvision.transforms.v2.RandomShortestSize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomShortestSize</span></code></a>(min_size[, max_size, ...])</p></td>
+<td><p>[BETA] Randomly resize the input.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomResize.html#torchvision.transforms.v2.RandomResize" title="torchvision.transforms.v2.RandomResize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomResize</span></code></a>(min_size, max_size[, ...])</p></td>
+<td><p>[BETA] Randomly resize the input.</p></td>
+</tr>
+</tbody>
+</table>
+<p>Functionals</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.resize.html#torchvision.transforms.v2.functional.resize" title="torchvision.transforms.v2.functional.resize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.resize</span></code></a>(inpt, size[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.Resize.html#torchvision.transforms.v2.Resize" title="torchvision.transforms.v2.Resize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Resize</span></code></a> for details.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="cropping">
+<h4>Cropping<a class="headerlink" href="#cropping" title="Permalink to this heading">¶</a></h4>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop" title="torchvision.transforms.v2.RandomCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomCrop</span></code></a>(size[, padding, ...])</p></td>
+<td><p>[BETA] Crop the input at a random location.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.v2.RandomResizedCrop" title="torchvision.transforms.v2.RandomResizedCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomResizedCrop</span></code></a>(size[, scale, ratio, ...])</p></td>
+<td><p>[BETA] Crop a random portion of the input and resize it to a given size.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomIoUCrop.html#torchvision.transforms.v2.RandomIoUCrop" title="torchvision.transforms.v2.RandomIoUCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomIoUCrop</span></code></a>([min_scale, max_scale, ...])</p></td>
+<td><p>[BETA] Random IoU crop transformation from <a class="reference external" href="https://arxiv.org/abs/1512.02325">&quot;SSD: Single Shot MultiBox Detector&quot;</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.CenterCrop.html#torchvision.transforms.v2.CenterCrop" title="torchvision.transforms.v2.CenterCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.CenterCrop</span></code></a>(size)</p></td>
+<td><p>[BETA] Crop the input at the center.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.FiveCrop.html#torchvision.transforms.v2.FiveCrop" title="torchvision.transforms.v2.FiveCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.FiveCrop</span></code></a>(size)</p></td>
+<td><p>[BETA] Crop the image or video into four corners and the central crop.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.TenCrop.html#torchvision.transforms.v2.TenCrop" title="torchvision.transforms.v2.TenCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.TenCrop</span></code></a>(size[, vertical_flip])</p></td>
+<td><p>[BETA] Crop the image or video into four corners and the central crop plus the flipped version of these (horizontal flipping is used by default).</p></td>
+</tr>
+</tbody>
+</table>
+<p>Functionals</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.crop.html#torchvision.transforms.v2.functional.crop" title="torchvision.transforms.v2.functional.crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.crop</span></code></a>(inpt, top, left, height, ...)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop" title="torchvision.transforms.v2.RandomCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomCrop</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.resized_crop.html#torchvision.transforms.v2.functional.resized_crop" title="torchvision.transforms.v2.functional.resized_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.resized_crop</span></code></a>(inpt, top, left, ...)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomResizedCrop.html#torchvision.transforms.v2.RandomResizedCrop" title="torchvision.transforms.v2.RandomResizedCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.ten_crop.html#torchvision.transforms.v2.functional.ten_crop" title="torchvision.transforms.v2.functional.ten_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.ten_crop</span></code></a>(inpt, size[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.TenCrop.html#torchvision.transforms.v2.TenCrop" title="torchvision.transforms.v2.TenCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">TenCrop</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.center_crop.html#torchvision.transforms.v2.functional.center_crop" title="torchvision.transforms.v2.functional.center_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.center_crop</span></code></a>(inpt, output_size)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomCrop.html#torchvision.transforms.v2.RandomCrop" title="torchvision.transforms.v2.RandomCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomCrop</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.five_crop.html#torchvision.transforms.v2.functional.five_crop" title="torchvision.transforms.v2.functional.five_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.five_crop</span></code></a>(inpt, size)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.FiveCrop.html#torchvision.transforms.v2.FiveCrop" title="torchvision.transforms.v2.FiveCrop"><code class="xref py py-class docutils literal notranslate"><span class="pre">FiveCrop</span></code></a> for details.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="others">
+<h4>Others<a class="headerlink" href="#others" title="Permalink to this heading">¶</a></h4>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomHorizontalFlip.html#torchvision.transforms.v2.RandomHorizontalFlip" title="torchvision.transforms.v2.RandomHorizontalFlip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomHorizontalFlip</span></code></a>([p])</p></td>
+<td><p>[BETA] Horizontally flip the input with a given probability.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomVerticalFlip.html#torchvision.transforms.v2.RandomVerticalFlip" title="torchvision.transforms.v2.RandomVerticalFlip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomVerticalFlip</span></code></a>([p])</p></td>
+<td><p>[BETA] Vertically flip the input with a given probability.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.Pad.html#torchvision.transforms.v2.Pad" title="torchvision.transforms.v2.Pad"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.Pad</span></code></a>(padding[, fill, padding_mode])</p></td>
+<td><p>[BETA] Pad the input on all sides with the given &quot;pad&quot; value.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomZoomOut.html#torchvision.transforms.v2.RandomZoomOut" title="torchvision.transforms.v2.RandomZoomOut"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomZoomOut</span></code></a>([fill, side_range, p])</p></td>
+<td><p><p>[BETA] &quot;Zoom out&quot; transformation from <a class="reference external" href="https://arxiv.org/abs/1512.02325">&quot;SSD: Single Shot MultiBox Detector&quot;</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomRotation.html#torchvision.transforms.v2.RandomRotation" title="torchvision.transforms.v2.RandomRotation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomRotation</span></code></a>(degrees[, interpolation, ...])</p></td>
+<td><p>[BETA] Rotate the input by angle.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomAffine.html#torchvision.transforms.v2.RandomAffine" title="torchvision.transforms.v2.RandomAffine"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomAffine</span></code></a>(degrees[, translate, scale, ...])</p></td>
+<td><p>[BETA] Random affine transformation the input keeping center invariant.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomPerspective.html#torchvision.transforms.v2.RandomPerspective" title="torchvision.transforms.v2.RandomPerspective"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomPerspective</span></code></a>([distortion_scale, p, ...])</p></td>
+<td><p>[BETA] Perform a random perspective transformation of the input with a given probability.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ElasticTransform.html#torchvision.transforms.v2.ElasticTransform" title="torchvision.transforms.v2.ElasticTransform"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ElasticTransform</span></code></a>([alpha, sigma, ...])</p></td>
+<td><p>[BETA] Transform the input with elastic transformations.</p></td>
+</tr>
+</tbody>
+</table>
+<p>Functionals</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.horizontal_flip.html#torchvision.transforms.v2.functional.horizontal_flip" title="torchvision.transforms.v2.functional.horizontal_flip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.horizontal_flip</span></code></a>(inpt)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomHorizontalFlip.html#torchvision.transforms.v2.RandomHorizontalFlip" title="torchvision.transforms.v2.RandomHorizontalFlip"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomHorizontalFlip</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.vertical_flip.html#torchvision.transforms.v2.functional.vertical_flip" title="torchvision.transforms.v2.functional.vertical_flip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.vertical_flip</span></code></a>(inpt)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomVerticalFlip.html#torchvision.transforms.v2.RandomVerticalFlip" title="torchvision.transforms.v2.RandomVerticalFlip"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomVerticalFlip</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.pad.html#torchvision.transforms.v2.functional.pad" title="torchvision.transforms.v2.functional.pad"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.pad</span></code></a>(inpt, padding[, fill, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.Pad.html#torchvision.transforms.v2.Pad" title="torchvision.transforms.v2.Pad"><code class="xref py py-class docutils literal notranslate"><span class="pre">Pad</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.rotate.html#torchvision.transforms.v2.functional.rotate" title="torchvision.transforms.v2.functional.rotate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.rotate</span></code></a>(inpt, angle[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomRotation.html#torchvision.transforms.v2.RandomRotation" title="torchvision.transforms.v2.RandomRotation"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomRotation</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.affine.html#torchvision.transforms.v2.functional.affine" title="torchvision.transforms.v2.functional.affine"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.affine</span></code></a>(inpt, angle, translate, ...)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomAffine.html#torchvision.transforms.v2.RandomAffine" title="torchvision.transforms.v2.RandomAffine"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAffine</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.perspective.html#torchvision.transforms.v2.functional.perspective" title="torchvision.transforms.v2.functional.perspective"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.perspective</span></code></a>(inpt, startpoints, ...)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomPerspective.html#torchvision.transforms.v2.RandomPerspective" title="torchvision.transforms.v2.RandomPerspective"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomPerspective</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.elastic.html#torchvision.transforms.v2.functional.elastic" title="torchvision.transforms.v2.functional.elastic"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.elastic</span></code></a>(inpt, displacement[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.ElasticTransform.html#torchvision.transforms.v2.ElasticTransform" title="torchvision.transforms.v2.ElasticTransform"><code class="xref py py-class docutils literal notranslate"><span class="pre">ElasticTransform</span></code></a> for details.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+<section id="color">
+<h3>Color<a class="headerlink" href="#color" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ColorJitter.html#torchvision.transforms.v2.ColorJitter" title="torchvision.transforms.v2.ColorJitter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ColorJitter</span></code></a>([brightness, contrast, ...])</p></td>
+<td><p>[BETA] Randomly change the brightness, contrast, saturation and hue of an image or video.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomChannelPermutation.html#torchvision.transforms.v2.RandomChannelPermutation" title="torchvision.transforms.v2.RandomChannelPermutation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomChannelPermutation</span></code></a>()</p></td>
+<td><p>[BETA] Randomly permute the channels of an image or video</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomPhotometricDistort.html#torchvision.transforms.v2.RandomPhotometricDistort" title="torchvision.transforms.v2.RandomPhotometricDistort"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomPhotometricDistort</span></code></a>([brightness, ...])</p></td>
+<td><p>[BETA] Randomly distorts the image or video as used in <a class="reference external" href="https://arxiv.org/abs/1512.02325">SSD: Single Shot MultiBox Detector</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.Grayscale.html#torchvision.transforms.v2.Grayscale" title="torchvision.transforms.v2.Grayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.Grayscale</span></code></a>([num_output_channels])</p></td>
+<td><p>[BETA] Convert images or videos to grayscale.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomGrayscale.html#torchvision.transforms.v2.RandomGrayscale" title="torchvision.transforms.v2.RandomGrayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomGrayscale</span></code></a>([p])</p></td>
+<td><p>[BETA] Randomly convert image or videos to grayscale with a probability of p (default 0.1).</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.GaussianBlur.html#torchvision.transforms.v2.GaussianBlur" title="torchvision.transforms.v2.GaussianBlur"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.GaussianBlur</span></code></a>(kernel_size[, sigma])</p></td>
+<td><p>[BETA] Blurs image with randomly chosen Gaussian blur.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomInvert.html#torchvision.transforms.v2.RandomInvert" title="torchvision.transforms.v2.RandomInvert"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomInvert</span></code></a>([p])</p></td>
+<td><p>[BETA] Inverts the colors of the given image or video with a given probability.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomPosterize.html#torchvision.transforms.v2.RandomPosterize" title="torchvision.transforms.v2.RandomPosterize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomPosterize</span></code></a>(bits[, p])</p></td>
+<td><p>[BETA] Posterize the image or video with a given probability by reducing the number of bits for each color channel.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomSolarize.html#torchvision.transforms.v2.RandomSolarize" title="torchvision.transforms.v2.RandomSolarize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomSolarize</span></code></a>(threshold[, p])</p></td>
+<td><p>[BETA] Solarize the image or video with a given probability by inverting all pixel values above a threshold.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomAdjustSharpness.html#torchvision.transforms.v2.RandomAdjustSharpness" title="torchvision.transforms.v2.RandomAdjustSharpness"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomAdjustSharpness</span></code></a>(sharpness_factor[, p])</p></td>
+<td><p>[BETA] Adjust the sharpness of the image or video with a given probability.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomAutocontrast.html#torchvision.transforms.v2.RandomAutocontrast" title="torchvision.transforms.v2.RandomAutocontrast"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomAutocontrast</span></code></a>([p])</p></td>
+<td><p>[BETA] Autocontrast the pixels of the given image or video with a given probability.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomEqualize.html#torchvision.transforms.v2.RandomEqualize" title="torchvision.transforms.v2.RandomEqualize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomEqualize</span></code></a>([p])</p></td>
+<td><p>[BETA] Equalize the histogram of the given image or video with a given probability.</p></td>
+</tr>
+</tbody>
+</table>
+<p>Functionals</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.permute_channels.html#torchvision.transforms.v2.functional.permute_channels" title="torchvision.transforms.v2.functional.permute_channels"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.permute_channels</span></code></a>(inpt, permutation)</p></td>
+<td><p>Permute the channels of the input according to the given permutation.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html#torchvision.transforms.v2.functional.rgb_to_grayscale" title="torchvision.transforms.v2.functional.rgb_to_grayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.rgb_to_grayscale</span></code></a>(inpt[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.Grayscale.html#torchvision.transforms.v2.Grayscale" title="torchvision.transforms.v2.Grayscale"><code class="xref py py-class docutils literal notranslate"><span class="pre">Grayscale</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.to_grayscale.html#torchvision.transforms.v2.functional.to_grayscale" title="torchvision.transforms.v2.functional.to_grayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.to_grayscale</span></code></a>(inpt[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.Grayscale.html#torchvision.transforms.v2.Grayscale" title="torchvision.transforms.v2.Grayscale"><code class="xref py py-class docutils literal notranslate"><span class="pre">Grayscale</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.gaussian_blur.html#torchvision.transforms.v2.functional.gaussian_blur" title="torchvision.transforms.v2.functional.gaussian_blur"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.gaussian_blur</span></code></a>(inpt, kernel_size)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.GaussianBlur.html#torchvision.transforms.v2.GaussianBlur" title="torchvision.transforms.v2.GaussianBlur"><code class="xref py py-class docutils literal notranslate"><span class="pre">GaussianBlur</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.invert.html#torchvision.transforms.v2.functional.invert" title="torchvision.transforms.v2.functional.invert"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.invert</span></code></a>(inpt)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomInvert.html#torchvision.transforms.v2.RandomInvert" title="torchvision.transforms.v2.RandomInvert"><code class="xref py py-func docutils literal notranslate"><span class="pre">RandomInvert()</span></code></a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.posterize.html#torchvision.transforms.v2.functional.posterize" title="torchvision.transforms.v2.functional.posterize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.posterize</span></code></a>(inpt, bits)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomPosterize.html#torchvision.transforms.v2.RandomPosterize" title="torchvision.transforms.v2.RandomPosterize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomPosterize</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.solarize.html#torchvision.transforms.v2.functional.solarize" title="torchvision.transforms.v2.functional.solarize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.solarize</span></code></a>(inpt, threshold)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomSolarize.html#torchvision.transforms.v2.RandomSolarize" title="torchvision.transforms.v2.RandomSolarize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomSolarize</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.adjust_sharpness.html#torchvision.transforms.v2.functional.adjust_sharpness" title="torchvision.transforms.v2.functional.adjust_sharpness"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.adjust_sharpness</span></code></a>(inpt, ...)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness" title="torchvision.transforms.RandomAdjustSharpness"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAdjustSharpness</span></code></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.autocontrast.html#torchvision.transforms.v2.functional.autocontrast" title="torchvision.transforms.v2.functional.autocontrast"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.autocontrast</span></code></a>(inpt)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomAutocontrast.html#torchvision.transforms.v2.RandomAutocontrast" title="torchvision.transforms.v2.RandomAutocontrast"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.adjust_contrast.html#torchvision.transforms.v2.functional.adjust_contrast" title="torchvision.transforms.v2.functional.adjust_contrast"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.adjust_contrast</span></code></a>(inpt, ...)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast" title="torchvision.transforms.RandomAutocontrast"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.equalize.html#torchvision.transforms.v2.functional.equalize" title="torchvision.transforms.v2.functional.equalize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.equalize</span></code></a>(inpt)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.RandomEqualize.html#torchvision.transforms.v2.RandomEqualize" title="torchvision.transforms.v2.RandomEqualize"><code class="xref py py-class docutils literal notranslate"><span class="pre">RandomEqualize</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.adjust_brightness.html#torchvision.transforms.v2.functional.adjust_brightness" title="torchvision.transforms.v2.functional.adjust_brightness"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.adjust_brightness</span></code></a>(inpt, ...)</p></td>
+<td><p>Adjust brightness.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.adjust_saturation.html#torchvision.transforms.v2.functional.adjust_saturation" title="torchvision.transforms.v2.functional.adjust_saturation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.adjust_saturation</span></code></a>(inpt, ...)</p></td>
+<td><p>Adjust saturation.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.adjust_hue.html#torchvision.transforms.v2.functional.adjust_hue" title="torchvision.transforms.v2.functional.adjust_hue"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.adjust_hue</span></code></a>(inpt, hue_factor)</p></td>
+<td><p>Adjust hue</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.adjust_gamma.html#torchvision.transforms.v2.functional.adjust_gamma" title="torchvision.transforms.v2.functional.adjust_gamma"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.adjust_gamma</span></code></a>(inpt, gamma[, gain])</p></td>
+<td><p>Adjust gamma.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="composition">
+<h3>Composition<a class="headerlink" href="#composition" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.Compose.html#torchvision.transforms.v2.Compose" title="torchvision.transforms.v2.Compose"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.Compose</span></code></a>(transforms)</p></td>
+<td><p>[BETA] Composes several transforms together.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomApply.html#torchvision.transforms.v2.RandomApply" title="torchvision.transforms.v2.RandomApply"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomApply</span></code></a>(transforms[, p])</p></td>
+<td><p>[BETA] Apply randomly a list of transformations with a given probability.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomChoice.html#torchvision.transforms.v2.RandomChoice" title="torchvision.transforms.v2.RandomChoice"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomChoice</span></code></a>(transforms[, p])</p></td>
+<td><p>[BETA] Apply single transformation randomly picked from a list.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomOrder.html#torchvision.transforms.v2.RandomOrder" title="torchvision.transforms.v2.RandomOrder"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomOrder</span></code></a>(transforms)</p></td>
+<td><p>[BETA] Apply a list of transformations in a random order.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="miscellaneous">
+<h3>Miscellaneous<a class="headerlink" href="#miscellaneous" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.LinearTransformation.html#torchvision.transforms.v2.LinearTransformation" title="torchvision.transforms.v2.LinearTransformation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.LinearTransformation</span></code></a>(...)</p></td>
+<td><p>[BETA] Transform a tensor image or video with a square transformation matrix and a mean_vector computed offline.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.Normalize.html#torchvision.transforms.v2.Normalize" title="torchvision.transforms.v2.Normalize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.Normalize</span></code></a>(mean, std[, inplace])</p></td>
+<td><p>[BETA] Normalize a tensor image or video with mean and standard deviation.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandomErasing.html#torchvision.transforms.v2.RandomErasing" title="torchvision.transforms.v2.RandomErasing"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandomErasing</span></code></a>([p, scale, ratio, value, ...])</p></td>
+<td><p>[BETA] Randomly select a rectangle region in the input image or video and erase its pixels.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.Lambda.html#torchvision.transforms.v2.Lambda" title="torchvision.transforms.v2.Lambda"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.Lambda</span></code></a>(lambd, *types)</p></td>
+<td><p>[BETA] Apply a user-defined function as a transform.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html#torchvision.transforms.v2.SanitizeBoundingBoxes" title="torchvision.transforms.v2.SanitizeBoundingBoxes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.SanitizeBoundingBoxes</span></code></a>([min_size, ...])</p></td>
+<td><p>[BETA] Remove degenerate/invalid bounding boxes and their corresponding labels and masks.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ClampBoundingBoxes.html#torchvision.transforms.v2.ClampBoundingBoxes" title="torchvision.transforms.v2.ClampBoundingBoxes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ClampBoundingBoxes</span></code></a>()</p></td>
+<td><p>[BETA] Clamp bounding boxes to their corresponding image dimensions.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.UniformTemporalSubsample.html#torchvision.transforms.v2.UniformTemporalSubsample" title="torchvision.transforms.v2.UniformTemporalSubsample"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.UniformTemporalSubsample</span></code></a>(num_samples)</p></td>
+<td><p>[BETA] Uniformly subsample <code class="docutils literal notranslate"><span class="pre">num_samples</span></code> indices from the temporal dimension of the video.</p></td>
+</tr>
+</tbody>
+</table>
+<p>Functionals</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.normalize.html#torchvision.transforms.v2.functional.normalize" title="torchvision.transforms.v2.functional.normalize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.normalize</span></code></a>(inpt, mean, std[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.Normalize.html#torchvision.transforms.v2.Normalize" title="torchvision.transforms.v2.Normalize"><code class="xref py py-class docutils literal notranslate"><span class="pre">Normalize</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.erase.html#torchvision.transforms.v2.functional.erase" title="torchvision.transforms.v2.functional.erase"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.erase</span></code></a>(inpt, i, j, h, w, v[, ...])</p></td>
+<td><p>[BETA] See <code class="xref py py-class docutils literal notranslate"><span class="pre">RandomErase</span></code> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html#torchvision.transforms.v2.functional.clamp_bounding_boxes" title="torchvision.transforms.v2.functional.clamp_bounding_boxes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.clamp_bounding_boxes</span></code></a>(inpt[, ...])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.ClampBoundingBoxes.html#torchvision.transforms.v2.ClampBoundingBoxes" title="torchvision.transforms.v2.ClampBoundingBoxes"><code class="xref py py-func docutils literal notranslate"><span class="pre">ClampBoundingBoxes()</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html#torchvision.transforms.v2.functional.uniform_temporal_subsample" title="torchvision.transforms.v2.functional.uniform_temporal_subsample"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.uniform_temporal_subsample</span></code></a>(...)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.UniformTemporalSubsample.html#torchvision.transforms.v2.UniformTemporalSubsample" title="torchvision.transforms.v2.UniformTemporalSubsample"><code class="xref py py-class docutils literal notranslate"><span class="pre">UniformTemporalSubsample</span></code></a> for details.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="conversion">
+<span id="conversion-transforms"></span><h3>Conversion<a class="headerlink" href="#conversion" title="Permalink to this heading">¶</a></h3>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Beware, some of these conversion transforms below will scale the values
+while performing the conversion, while some may not do any scaling. By
+scaling, we mean e.g. that a <code class="docutils literal notranslate"><span class="pre">uint8</span></code> -&gt; <code class="docutils literal notranslate"><span class="pre">float32</span></code> would map the [0,
+255] range into [0, 1] (and vice-versa). See <a class="reference internal" href="#range-and-dtype"><span class="std std-ref">Dtype and expected value range</span></a>.</p>
+</div>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ToImage.html#torchvision.transforms.v2.ToImage" title="torchvision.transforms.v2.ToImage"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ToImage</span></code></a>()</p></td>
+<td><p>[BETA] Convert a tensor, ndarray, or PIL Image to <a class="reference internal" href="generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-class docutils literal notranslate"><span class="pre">Image</span></code></a> ; this does not scale values.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ToPureTensor.html#torchvision.transforms.v2.ToPureTensor" title="torchvision.transforms.v2.ToPureTensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ToPureTensor</span></code></a>()</p></td>
+<td><p>[BETA] Convert all tv_tensors to pure tensors, removing associated metadata (if any).</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.PILToTensor.html#torchvision.transforms.v2.PILToTensor" title="torchvision.transforms.v2.PILToTensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.PILToTensor</span></code></a>()</p></td>
+<td><p>[BETA] Convert a PIL Image to a tensor of the same type - this does not scale values.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ToPILImage.html#torchvision.transforms.v2.ToPILImage" title="torchvision.transforms.v2.ToPILImage"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ToPILImage</span></code></a>([mode])</p></td>
+<td><p>[BETA] Convert a tensor or an ndarray to PIL Image</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ToDtype.html#torchvision.transforms.v2.ToDtype" title="torchvision.transforms.v2.ToDtype"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ToDtype</span></code></a>(dtype[, scale])</p></td>
+<td><p>[BETA] Converts the input to a specific dtype, optionally scaling the values for images or videos.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html#torchvision.transforms.v2.ConvertBoundingBoxFormat" title="torchvision.transforms.v2.ConvertBoundingBoxFormat"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ConvertBoundingBoxFormat</span></code></a>(format)</p></td>
+<td><p>[BETA] Convert bounding box coordinates to the given <code class="docutils literal notranslate"><span class="pre">format</span></code>, eg from &quot;CXCYWH&quot; to &quot;XYXY&quot;.</p></td>
+</tr>
+</tbody>
+</table>
+<p>functionals</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.to_image.html#torchvision.transforms.v2.functional.to_image" title="torchvision.transforms.v2.functional.to_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.to_image</span></code></a>(inpt)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.ToImage.html#torchvision.transforms.v2.ToImage" title="torchvision.transforms.v2.ToImage"><code class="xref py py-class docutils literal notranslate"><span class="pre">ToImage</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.pil_to_tensor.html#torchvision.transforms.v2.functional.pil_to_tensor" title="torchvision.transforms.v2.functional.pil_to_tensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.pil_to_tensor</span></code></a>(pic)</p></td>
+<td><p>Convert a <code class="docutils literal notranslate"><span class="pre">PIL</span> <span class="pre">Image</span></code> to a tensor of the same type.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.to_pil_image.html#torchvision.transforms.v2.functional.to_pil_image" title="torchvision.transforms.v2.functional.to_pil_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.to_pil_image</span></code></a>(pic[, mode])</p></td>
+<td><p>Convert a tensor or an ndarray to PIL Image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.to_dtype.html#torchvision.transforms.v2.functional.to_dtype" title="torchvision.transforms.v2.functional.to_dtype"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.to_dtype</span></code></a>(inpt[, dtype, scale])</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.ToDtype.html#torchvision.transforms.v2.ToDtype" title="torchvision.transforms.v2.ToDtype"><code class="xref py py-func docutils literal notranslate"><span class="pre">ToDtype()</span></code></a> for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html#torchvision.transforms.v2.functional.convert_bounding_box_format" title="torchvision.transforms.v2.functional.convert_bounding_box_format"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.convert_bounding_box_format</span></code></a>(inpt)</p></td>
+<td><p>[BETA] See <a class="reference internal" href="generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html#torchvision.transforms.v2.ConvertBoundingBoxFormat" title="torchvision.transforms.v2.ConvertBoundingBoxFormat"><code class="xref py py-func docutils literal notranslate"><span class="pre">ConvertBoundingBoxFormat()</span></code></a> for details.</p></td>
+</tr>
+</tbody>
+</table>
+<p>Deprecated</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ToTensor.html#torchvision.transforms.v2.ToTensor" title="torchvision.transforms.v2.ToTensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ToTensor</span></code></a>()</p></td>
+<td><p>[BETA] [DEPRECATED] Use <code class="docutils literal notranslate"><span class="pre">v2.Compose([v2.ToImage(),</span> <span class="pre">v2.ToDtype(torch.float32,</span> <span class="pre">scale=True)])</span></code> instead.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.to_tensor.html#torchvision.transforms.v2.functional.to_tensor" title="torchvision.transforms.v2.functional.to_tensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.to_tensor</span></code></a>(inpt)</p></td>
+<td><p>[BETA] [DEPREACTED] Use to_image() and to_dtype() instead.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.ConvertImageDtype.html#torchvision.transforms.v2.ConvertImageDtype" title="torchvision.transforms.v2.ConvertImageDtype"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.ConvertImageDtype</span></code></a>([dtype])</p></td>
+<td><p>[BETA] [DEPRECATED] Use <code class="docutils literal notranslate"><span class="pre">v2.ToDtype(dtype,</span> <span class="pre">scale=True)</span></code> instead.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.convert_image_dtype.html#torchvision.transforms.v2.functional.convert_image_dtype" title="torchvision.transforms.v2.functional.convert_image_dtype"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.convert_image_dtype</span></code></a>(image[, dtype])</p></td>
+<td><p>[BETA] [DEPRECATED] Use to_dtype() instead.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="auto-augmentation">
+<h3>Auto-Augmentation<a class="headerlink" href="#auto-augmentation" title="Permalink to this heading">¶</a></h3>
+<p><a class="reference external" href="https://arxiv.org/pdf/1805.09501.pdf">AutoAugment</a> is a common Data Augmentation technique that can improve the accuracy of Image Classification models.
+Though the data augmentation policies are directly linked to their trained dataset, empirical studies show that
+ImageNet policies provide significant improvements when applied to other datasets.
+In TorchVision we implemented 3 policies learned on the following datasets: ImageNet, CIFAR10 and SVHN.
+The new transform can be used standalone or mixed-and-matched with existing transforms:</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.AutoAugment.html#torchvision.transforms.v2.AutoAugment" title="torchvision.transforms.v2.AutoAugment"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.AutoAugment</span></code></a>([policy, interpolation, fill])</p></td>
+<td><p>[BETA] AutoAugment data augmentation method based on <a class="reference external" href="https://arxiv.org/pdf/1805.09501.pdf">&quot;AutoAugment: Learning Augmentation Strategies from Data&quot;</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.RandAugment.html#torchvision.transforms.v2.RandAugment" title="torchvision.transforms.v2.RandAugment"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.RandAugment</span></code></a>([num_ops, magnitude, ...])</p></td>
+<td><p>[BETA] RandAugment data augmentation method based on <a class="reference external" href="https://arxiv.org/abs/1909.13719">&quot;RandAugment: Practical automated data augmentation with a reduced search space&quot;</a>.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.TrivialAugmentWide.html#torchvision.transforms.v2.TrivialAugmentWide" title="torchvision.transforms.v2.TrivialAugmentWide"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.TrivialAugmentWide</span></code></a>([num_magnitude_bins, ...])</p></td>
+<td><p>[BETA] Dataset-independent data-augmentation with TrivialAugment Wide, as described in <a class="reference external" href="https://arxiv.org/abs/2103.10158">&quot;TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation&quot;</a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.AugMix.html#torchvision.transforms.v2.AugMix" title="torchvision.transforms.v2.AugMix"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.AugMix</span></code></a>([severity, mixture_width, ...])</p></td>
+<td><p>[BETA] AugMix data augmentation method based on <a class="reference external" href="https://arxiv.org/abs/1912.02781">&quot;AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty&quot;</a>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="cutmix-mixup">
+<h3>CutMix - MixUp<a class="headerlink" href="#cutmix-mixup" title="Permalink to this heading">¶</a></h3>
+<p>CutMix and MixUp are special transforms that
+are meant to be used on batches rather than on individual images, because they
+are combining pairs of images together. These can be used after the dataloader
+(once the samples are batched), or part of a collation function. See
+<a class="reference internal" href="auto_examples/transforms/plot_cutmix_mixup.html#sphx-glr-auto-examples-transforms-plot-cutmix-mixup-py"><span class="std std-ref">How to use CutMix and MixUp</span></a> for detailed usage examples.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.CutMix.html#torchvision.transforms.v2.CutMix" title="torchvision.transforms.v2.CutMix"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.CutMix</span></code></a>(*[, alpha, labels_getter])</p></td>
+<td><p>[BETA] Apply CutMix to the provided batch of images and labels.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.MixUp.html#torchvision.transforms.v2.MixUp" title="torchvision.transforms.v2.MixUp"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.MixUp</span></code></a>(*[, alpha, labels_getter])</p></td>
+<td><p>[BETA] Apply MixUp to the provided batch of images and labels.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="developer-tools">
+<h3>Developer tools<a class="headerlink" href="#developer-tools" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.v2.functional.register_kernel.html#torchvision.transforms.v2.functional.register_kernel" title="torchvision.transforms.v2.functional.register_kernel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">v2.functional.register_kernel</span></code></a>(functional, ...)</p></td>
+<td><p>[BETA] Decorate a kernel to register it for a functional and a (custom) tv_tensor type.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+<section id="v1-api-reference">
+<h2>V1 API Reference<a class="headerlink" href="#v1-api-reference" title="Permalink to this heading">¶</a></h2>
+<section id="id3">
+<h3>Geometry<a class="headerlink" href="#id3" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.Resize.html#torchvision.transforms.Resize" title="torchvision.transforms.Resize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Resize</span></code></a>(size[, interpolation, max_size, ...])</p></td>
+<td><p>Resize the input image to the given size.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomCrop.html#torchvision.transforms.RandomCrop" title="torchvision.transforms.RandomCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomCrop</span></code></a>(size[, padding, pad_if_needed, ...])</p></td>
+<td><p>Crop the given image at a random location.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomResizedCrop.html#torchvision.transforms.RandomResizedCrop" title="torchvision.transforms.RandomResizedCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomResizedCrop</span></code></a>(size[, scale, ratio, ...])</p></td>
+<td><p>Crop a random portion of image and resize it to a given size.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.CenterCrop.html#torchvision.transforms.CenterCrop" title="torchvision.transforms.CenterCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CenterCrop</span></code></a>(size)</p></td>
+<td><p>Crops the given image at the center.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.FiveCrop.html#torchvision.transforms.FiveCrop" title="torchvision.transforms.FiveCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FiveCrop</span></code></a>(size)</p></td>
+<td><p>Crop the given image into four corners and the central crop.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.TenCrop.html#torchvision.transforms.TenCrop" title="torchvision.transforms.TenCrop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TenCrop</span></code></a>(size[, vertical_flip])</p></td>
+<td><p>Crop the given image into four corners and the central crop plus the flipped version of these (horizontal flipping is used by default).</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.Pad.html#torchvision.transforms.Pad" title="torchvision.transforms.Pad"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Pad</span></code></a>(padding[, fill, padding_mode])</p></td>
+<td><p>Pad the given image on all sides with the given &quot;pad&quot; value.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomRotation.html#torchvision.transforms.RandomRotation" title="torchvision.transforms.RandomRotation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomRotation</span></code></a>(degrees[, interpolation, ...])</p></td>
+<td><p>Rotate the image by angle.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomAffine.html#torchvision.transforms.RandomAffine" title="torchvision.transforms.RandomAffine"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomAffine</span></code></a>(degrees[, translate, scale, ...])</p></td>
+<td><p>Random affine transformation of the image keeping center invariant.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomPerspective.html#torchvision.transforms.RandomPerspective" title="torchvision.transforms.RandomPerspective"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomPerspective</span></code></a>([distortion_scale, p, ...])</p></td>
+<td><p>Performs a random perspective transformation of the given image with a given probability.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.ElasticTransform.html#torchvision.transforms.ElasticTransform" title="torchvision.transforms.ElasticTransform"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ElasticTransform</span></code></a>([alpha, sigma, ...])</p></td>
+<td><p>Transform a tensor image with elastic transformations.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomHorizontalFlip.html#torchvision.transforms.RandomHorizontalFlip" title="torchvision.transforms.RandomHorizontalFlip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomHorizontalFlip</span></code></a>([p])</p></td>
+<td><p>Horizontally flip the given image randomly with a given probability.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomVerticalFlip.html#torchvision.transforms.RandomVerticalFlip" title="torchvision.transforms.RandomVerticalFlip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomVerticalFlip</span></code></a>([p])</p></td>
+<td><p>Vertically flip the given image randomly with a given probability.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="id4">
+<h3>Color<a class="headerlink" href="#id4" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.ColorJitter.html#torchvision.transforms.ColorJitter" title="torchvision.transforms.ColorJitter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ColorJitter</span></code></a>([brightness, contrast, ...])</p></td>
+<td><p>Randomly change the brightness, contrast, saturation and hue of an image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.Grayscale.html#torchvision.transforms.Grayscale" title="torchvision.transforms.Grayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Grayscale</span></code></a>([num_output_channels])</p></td>
+<td><p>Convert image to grayscale.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomGrayscale.html#torchvision.transforms.RandomGrayscale" title="torchvision.transforms.RandomGrayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomGrayscale</span></code></a>([p])</p></td>
+<td><p>Randomly convert image to grayscale with a probability of p (default 0.1).</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.GaussianBlur.html#torchvision.transforms.GaussianBlur" title="torchvision.transforms.GaussianBlur"><code class="xref py py-obj docutils literal notranslate"><span class="pre">GaussianBlur</span></code></a>(kernel_size[, sigma])</p></td>
+<td><p>Blurs image with randomly chosen Gaussian blur.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomInvert.html#torchvision.transforms.RandomInvert" title="torchvision.transforms.RandomInvert"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomInvert</span></code></a>([p])</p></td>
+<td><p>Inverts the colors of the given image randomly with a given probability.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomPosterize.html#torchvision.transforms.RandomPosterize" title="torchvision.transforms.RandomPosterize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomPosterize</span></code></a>(bits[, p])</p></td>
+<td><p>Posterize the image randomly with a given probability by reducing the number of bits for each color channel.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomSolarize.html#torchvision.transforms.RandomSolarize" title="torchvision.transforms.RandomSolarize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomSolarize</span></code></a>(threshold[, p])</p></td>
+<td><p>Solarize the image randomly with a given probability by inverting all pixel values above a threshold.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomAdjustSharpness.html#torchvision.transforms.RandomAdjustSharpness" title="torchvision.transforms.RandomAdjustSharpness"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomAdjustSharpness</span></code></a>(sharpness_factor[, p])</p></td>
+<td><p>Adjust the sharpness of the image randomly with a given probability.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomAutocontrast.html#torchvision.transforms.RandomAutocontrast" title="torchvision.transforms.RandomAutocontrast"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomAutocontrast</span></code></a>([p])</p></td>
+<td><p>Autocontrast the pixels of the given image randomly with a given probability.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomEqualize.html#torchvision.transforms.RandomEqualize" title="torchvision.transforms.RandomEqualize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomEqualize</span></code></a>([p])</p></td>
+<td><p>Equalize the histogram of the given image randomly with a given probability.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="id5">
+<h3>Composition<a class="headerlink" href="#id5" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.Compose.html#torchvision.transforms.Compose" title="torchvision.transforms.Compose"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Compose</span></code></a>(transforms)</p></td>
+<td><p>Composes several transforms together.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomApply.html#torchvision.transforms.RandomApply" title="torchvision.transforms.RandomApply"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomApply</span></code></a>(transforms[, p])</p></td>
+<td><p>Apply randomly a list of transformations with a given probability.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomChoice.html#torchvision.transforms.RandomChoice" title="torchvision.transforms.RandomChoice"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomChoice</span></code></a>(transforms[, p])</p></td>
+<td><p>Apply single transformation randomly picked from a list.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomOrder.html#torchvision.transforms.RandomOrder" title="torchvision.transforms.RandomOrder"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomOrder</span></code></a>(transforms)</p></td>
+<td><p>Apply a list of transformations in a random order.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="id6">
+<h3>Miscellaneous<a class="headerlink" href="#id6" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.LinearTransformation.html#torchvision.transforms.LinearTransformation" title="torchvision.transforms.LinearTransformation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LinearTransformation</span></code></a>(transformation_matrix, ...)</p></td>
+<td><p>Transform a tensor image with a square transformation matrix and a mean_vector computed offline.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.Normalize.html#torchvision.transforms.Normalize" title="torchvision.transforms.Normalize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Normalize</span></code></a>(mean, std[, inplace])</p></td>
+<td><p>Normalize a tensor image with mean and standard deviation.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandomErasing.html#torchvision.transforms.RandomErasing" title="torchvision.transforms.RandomErasing"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandomErasing</span></code></a>([p, scale, ratio, value, inplace])</p></td>
+<td><p>Randomly selects a rectangle region in a torch.Tensor image and erases its pixels.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.Lambda.html#torchvision.transforms.Lambda" title="torchvision.transforms.Lambda"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Lambda</span></code></a>(lambd)</p></td>
+<td><p>Apply a user-defined lambda as a transform.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="id7">
+<h3>Conversion<a class="headerlink" href="#id7" title="Permalink to this heading">¶</a></h3>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>Beware, some of these conversion transforms below will scale the values
+while performing the conversion, while some may not do any scaling. By
+scaling, we mean e.g. that a <code class="docutils literal notranslate"><span class="pre">uint8</span></code> -&gt; <code class="docutils literal notranslate"><span class="pre">float32</span></code> would map the [0,
+255] range into [0, 1] (and vice-versa). See <a class="reference internal" href="#range-and-dtype"><span class="std std-ref">Dtype and expected value range</span></a>.</p>
+</div>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.ToPILImage.html#torchvision.transforms.ToPILImage" title="torchvision.transforms.ToPILImage"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ToPILImage</span></code></a>([mode])</p></td>
+<td><p>Convert a tensor or an ndarray to PIL Image</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.ToTensor.html#torchvision.transforms.ToTensor" title="torchvision.transforms.ToTensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ToTensor</span></code></a>()</p></td>
+<td><p>Convert a PIL Image or ndarray to tensor and scale the values accordingly.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.PILToTensor.html#torchvision.transforms.PILToTensor" title="torchvision.transforms.PILToTensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PILToTensor</span></code></a>()</p></td>
+<td><p>Convert a PIL Image to a tensor of the same type - this does not scale values.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.ConvertImageDtype.html#torchvision.transforms.ConvertImageDtype" title="torchvision.transforms.ConvertImageDtype"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ConvertImageDtype</span></code></a>(dtype)</p></td>
+<td><p>Convert a tensor image to the given <code class="docutils literal notranslate"><span class="pre">dtype</span></code> and scale the values accordingly.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="id8">
+<h3>Auto-Augmentation<a class="headerlink" href="#id8" title="Permalink to this heading">¶</a></h3>
+<p><a class="reference external" href="https://arxiv.org/pdf/1805.09501.pdf">AutoAugment</a> is a common Data Augmentation technique that can improve the accuracy of Image Classification models.
+Though the data augmentation policies are directly linked to their trained dataset, empirical studies show that
+ImageNet policies provide significant improvements when applied to other datasets.
+In TorchVision we implemented 3 policies learned on the following datasets: ImageNet, CIFAR10 and SVHN.
+The new transform can be used standalone or mixed-and-matched with existing transforms:</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.AutoAugmentPolicy.html#torchvision.transforms.AutoAugmentPolicy" title="torchvision.transforms.AutoAugmentPolicy"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AutoAugmentPolicy</span></code></a>(value)</p></td>
+<td><p>AutoAugment policies learned on different datasets.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.AutoAugment.html#torchvision.transforms.AutoAugment" title="torchvision.transforms.AutoAugment"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AutoAugment</span></code></a>([policy, interpolation, fill])</p></td>
+<td><p><p>AutoAugment data augmentation method based on <a class="reference external" href="https://arxiv.org/pdf/1805.09501.pdf">&quot;AutoAugment: Learning Augmentation Strategies from Data&quot;</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.RandAugment.html#torchvision.transforms.RandAugment" title="torchvision.transforms.RandAugment"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RandAugment</span></code></a>([num_ops, magnitude, ...])</p></td>
+<td><p><p>RandAugment data augmentation method based on <a class="reference external" href="https://arxiv.org/abs/1909.13719">&quot;RandAugment: Practical automated data augmentation with a reduced search space&quot;</a>.</p>
+</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.TrivialAugmentWide.html#torchvision.transforms.TrivialAugmentWide" title="torchvision.transforms.TrivialAugmentWide"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TrivialAugmentWide</span></code></a>([num_magnitude_bins, ...])</p></td>
+<td><p><p>Dataset-independent data-augmentation with TrivialAugment Wide, as described in <a class="reference external" href="https://arxiv.org/abs/2103.10158">&quot;TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation&quot;</a>.</p>
+</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.AugMix.html#torchvision.transforms.AugMix" title="torchvision.transforms.AugMix"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AugMix</span></code></a>([severity, mixture_width, ...])</p></td>
+<td><p><p>AugMix data augmentation method based on <a class="reference external" href="https://arxiv.org/abs/1912.02781">&quot;AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty&quot;</a>.</p>
+</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+<section id="id14">
+<h3>Functional Transforms<a class="headerlink" href="#id14" title="Permalink to this heading">¶</a></h3>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.adjust_brightness.html#torchvision.transforms.functional.adjust_brightness" title="torchvision.transforms.functional.adjust_brightness"><code class="xref py py-obj docutils literal notranslate"><span class="pre">adjust_brightness</span></code></a>(img, brightness_factor)</p></td>
+<td><p>Adjust brightness of an image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.adjust_contrast.html#torchvision.transforms.functional.adjust_contrast" title="torchvision.transforms.functional.adjust_contrast"><code class="xref py py-obj docutils literal notranslate"><span class="pre">adjust_contrast</span></code></a>(img, contrast_factor)</p></td>
+<td><p>Adjust contrast of an image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.adjust_gamma.html#torchvision.transforms.functional.adjust_gamma" title="torchvision.transforms.functional.adjust_gamma"><code class="xref py py-obj docutils literal notranslate"><span class="pre">adjust_gamma</span></code></a>(img, gamma[, gain])</p></td>
+<td><p>Perform gamma correction on an image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.adjust_hue.html#torchvision.transforms.functional.adjust_hue" title="torchvision.transforms.functional.adjust_hue"><code class="xref py py-obj docutils literal notranslate"><span class="pre">adjust_hue</span></code></a>(img, hue_factor)</p></td>
+<td><p>Adjust hue of an image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.adjust_saturation.html#torchvision.transforms.functional.adjust_saturation" title="torchvision.transforms.functional.adjust_saturation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">adjust_saturation</span></code></a>(img, saturation_factor)</p></td>
+<td><p>Adjust color saturation of an image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.adjust_sharpness.html#torchvision.transforms.functional.adjust_sharpness" title="torchvision.transforms.functional.adjust_sharpness"><code class="xref py py-obj docutils literal notranslate"><span class="pre">adjust_sharpness</span></code></a>(img, sharpness_factor)</p></td>
+<td><p>Adjust the sharpness of an image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.affine.html#torchvision.transforms.functional.affine" title="torchvision.transforms.functional.affine"><code class="xref py py-obj docutils literal notranslate"><span class="pre">affine</span></code></a>(img, angle, translate, scale, shear)</p></td>
+<td><p>Apply affine transformation on the image keeping image center invariant.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.autocontrast.html#torchvision.transforms.functional.autocontrast" title="torchvision.transforms.functional.autocontrast"><code class="xref py py-obj docutils literal notranslate"><span class="pre">autocontrast</span></code></a>(img)</p></td>
+<td><p>Maximize contrast of an image by remapping its pixels per channel so that the lowest becomes black and the lightest becomes white.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.center_crop.html#torchvision.transforms.functional.center_crop" title="torchvision.transforms.functional.center_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">center_crop</span></code></a>(img, output_size)</p></td>
+<td><p>Crops the given image at the center.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.convert_image_dtype.html#torchvision.transforms.functional.convert_image_dtype" title="torchvision.transforms.functional.convert_image_dtype"><code class="xref py py-obj docutils literal notranslate"><span class="pre">convert_image_dtype</span></code></a>(image[, dtype])</p></td>
+<td><p>Convert a tensor image to the given <code class="docutils literal notranslate"><span class="pre">dtype</span></code> and scale the values accordingly This function does not support PIL Image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.crop.html#torchvision.transforms.functional.crop" title="torchvision.transforms.functional.crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">crop</span></code></a>(img, top, left, height, width)</p></td>
+<td><p>Crop the given image at specified location and output size.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.equalize.html#torchvision.transforms.functional.equalize" title="torchvision.transforms.functional.equalize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">equalize</span></code></a>(img)</p></td>
+<td><p>Equalize the histogram of an image by applying a non-linear mapping to the input in order to create a uniform distribution of grayscale values in the output.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.erase.html#torchvision.transforms.functional.erase" title="torchvision.transforms.functional.erase"><code class="xref py py-obj docutils literal notranslate"><span class="pre">erase</span></code></a>(img, i, j, h, w, v[, inplace])</p></td>
+<td><p>Erase the input Tensor Image with given value.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.five_crop.html#torchvision.transforms.functional.five_crop" title="torchvision.transforms.functional.five_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">five_crop</span></code></a>(img, size)</p></td>
+<td><p>Crop the given image into four corners and the central crop.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.gaussian_blur.html#torchvision.transforms.functional.gaussian_blur" title="torchvision.transforms.functional.gaussian_blur"><code class="xref py py-obj docutils literal notranslate"><span class="pre">gaussian_blur</span></code></a>(img, kernel_size[, sigma])</p></td>
+<td><p>Performs Gaussian blurring on the image by given kernel.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.get_dimensions.html#torchvision.transforms.functional.get_dimensions" title="torchvision.transforms.functional.get_dimensions"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_dimensions</span></code></a>(img)</p></td>
+<td><p>Returns the dimensions of an image as [channels, height, width].</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.get_image_num_channels.html#torchvision.transforms.functional.get_image_num_channels" title="torchvision.transforms.functional.get_image_num_channels"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_image_num_channels</span></code></a>(img)</p></td>
+<td><p>Returns the number of channels of an image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.get_image_size.html#torchvision.transforms.functional.get_image_size" title="torchvision.transforms.functional.get_image_size"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_image_size</span></code></a>(img)</p></td>
+<td><p>Returns the size of an image as [width, height].</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.hflip.html#torchvision.transforms.functional.hflip" title="torchvision.transforms.functional.hflip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">hflip</span></code></a>(img)</p></td>
+<td><p>Horizontally flip the given image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.invert.html#torchvision.transforms.functional.invert" title="torchvision.transforms.functional.invert"><code class="xref py py-obj docutils literal notranslate"><span class="pre">invert</span></code></a>(img)</p></td>
+<td><p>Invert the colors of an RGB/grayscale image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.normalize.html#torchvision.transforms.functional.normalize" title="torchvision.transforms.functional.normalize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">normalize</span></code></a>(tensor, mean, std[, inplace])</p></td>
+<td><p>Normalize a float tensor image with mean and standard deviation.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.pad.html#torchvision.transforms.functional.pad" title="torchvision.transforms.functional.pad"><code class="xref py py-obj docutils literal notranslate"><span class="pre">pad</span></code></a>(img, padding[, fill, padding_mode])</p></td>
+<td><p>Pad the given image on all sides with the given &quot;pad&quot; value.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.perspective.html#torchvision.transforms.functional.perspective" title="torchvision.transforms.functional.perspective"><code class="xref py py-obj docutils literal notranslate"><span class="pre">perspective</span></code></a>(img, startpoints, endpoints[, ...])</p></td>
+<td><p>Perform perspective transform of the given image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.pil_to_tensor.html#torchvision.transforms.functional.pil_to_tensor" title="torchvision.transforms.functional.pil_to_tensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">pil_to_tensor</span></code></a>(pic)</p></td>
+<td><p>Convert a <code class="docutils literal notranslate"><span class="pre">PIL</span> <span class="pre">Image</span></code> to a tensor of the same type.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.posterize.html#torchvision.transforms.functional.posterize" title="torchvision.transforms.functional.posterize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">posterize</span></code></a>(img, bits)</p></td>
+<td><p>Posterize an image by reducing the number of bits for each color channel.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.resize.html#torchvision.transforms.functional.resize" title="torchvision.transforms.functional.resize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resize</span></code></a>(img, size[, interpolation, max_size, ...])</p></td>
+<td><p>Resize the input image to the given size.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.resized_crop.html#torchvision.transforms.functional.resized_crop" title="torchvision.transforms.functional.resized_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">resized_crop</span></code></a>(img, top, left, height, width, size)</p></td>
+<td><p>Crop the given image and resize it to desired size.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.rgb_to_grayscale.html#torchvision.transforms.functional.rgb_to_grayscale" title="torchvision.transforms.functional.rgb_to_grayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">rgb_to_grayscale</span></code></a>(img[, num_output_channels])</p></td>
+<td><p>Convert RGB image to grayscale version of image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.rotate.html#torchvision.transforms.functional.rotate" title="torchvision.transforms.functional.rotate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">rotate</span></code></a>(img, angle[, interpolation, expand, ...])</p></td>
+<td><p>Rotate the image by angle.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.solarize.html#torchvision.transforms.functional.solarize" title="torchvision.transforms.functional.solarize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">solarize</span></code></a>(img, threshold)</p></td>
+<td><p>Solarize an RGB/grayscale image by inverting all pixel values above a threshold.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.ten_crop.html#torchvision.transforms.functional.ten_crop" title="torchvision.transforms.functional.ten_crop"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ten_crop</span></code></a>(img, size[, vertical_flip])</p></td>
+<td><p>Generate ten cropped images from the given image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.to_grayscale.html#torchvision.transforms.functional.to_grayscale" title="torchvision.transforms.functional.to_grayscale"><code class="xref py py-obj docutils literal notranslate"><span class="pre">to_grayscale</span></code></a>(img[, num_output_channels])</p></td>
+<td><p>Convert PIL image of any mode (RGB, HSV, LAB, etc) to grayscale version of image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.to_pil_image.html#torchvision.transforms.functional.to_pil_image" title="torchvision.transforms.functional.to_pil_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">to_pil_image</span></code></a>(pic[, mode])</p></td>
+<td><p>Convert a tensor or an ndarray to PIL Image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.to_tensor.html#torchvision.transforms.functional.to_tensor" title="torchvision.transforms.functional.to_tensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">to_tensor</span></code></a>(pic)</p></td>
+<td><p>Convert a <code class="docutils literal notranslate"><span class="pre">PIL</span> <span class="pre">Image</span></code> or <code class="docutils literal notranslate"><span class="pre">numpy.ndarray</span></code> to tensor.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.transforms.functional.vflip.html#torchvision.transforms.functional.vflip" title="torchvision.transforms.functional.vflip"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vflip</span></code></a>(img)</p></td>
+<td><p>Vertically flip the given image.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+</section>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.transforms.v2.Resize.html" class="btn btn-neutral float-right" title="Resize" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="index.html" class="btn btn-neutral" title="torchvision" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Transforming and augmenting images</a><ul>
+<li><a class="reference internal" href="#start-here">Start here</a></li>
+<li><a class="reference internal" href="#supported-input-types-and-conventions">Supported input types and conventions</a><ul>
+<li><a class="reference internal" href="#dtype-and-expected-value-range">Dtype and expected value range</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#v1-or-v2-which-one-should-i-use">V1 or V2? Which one should I use?</a></li>
+<li><a class="reference internal" href="#performance-considerations">Performance considerations</a></li>
+<li><a class="reference internal" href="#transform-classes-functionals-and-kernels">Transform classes, functionals, and kernels</a></li>
+<li><a class="reference internal" href="#torchscript-support">Torchscript support</a></li>
+<li><a class="reference internal" href="#v2-api-reference-recommended">V2 API reference - Recommended</a><ul>
+<li><a class="reference internal" href="#geometry">Geometry</a><ul>
+<li><a class="reference internal" href="#resizing">Resizing</a><ul>
+</ul>
+</li>
+<li><a class="reference internal" href="#cropping">Cropping</a><ul>
+</ul>
+</li>
+<li><a class="reference internal" href="#others">Others</a><ul>
+</ul>
+</li>
+</ul>
+</li>
+<li><a class="reference internal" href="#color">Color</a><ul>
+</ul>
+</li>
+<li><a class="reference internal" href="#composition">Composition</a></li>
+<li><a class="reference internal" href="#miscellaneous">Miscellaneous</a><ul>
+</ul>
+</li>
+<li><a class="reference internal" href="#conversion">Conversion</a><ul>
+</ul>
+</li>
+<li><a class="reference internal" href="#auto-augmentation">Auto-Augmentation</a></li>
+<li><a class="reference internal" href="#cutmix-mixup">CutMix - MixUp</a></li>
+<li><a class="reference internal" href="#developer-tools">Developer tools</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#v1-api-reference">V1 API Reference</a><ul>
+<li><a class="reference internal" href="#id3">Geometry</a></li>
+<li><a class="reference internal" href="#id4">Color</a></li>
+<li><a class="reference internal" href="#id5">Composition</a></li>
+<li><a class="reference internal" href="#id6">Miscellaneous</a></li>
+<li><a class="reference internal" href="#id7">Conversion</a></li>
+<li><a class="reference internal" href="#id8">Auto-Augmentation</a></li>
+<li><a class="reference internal" href="#id14">Functional Transforms</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/tv_tensors.html b/0.16/tv_tensors.html
new file mode 100644
index 00000000000..a459475c705
--- /dev/null
+++ b/0.16/tv_tensors.html
@@ -0,0 +1,721 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>TVTensors &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="Image" href="generated/torchvision.tv_tensors.Image.html" />
+    <link rel="prev" title="vflip" href="generated/torchvision.transforms.functional.vflip.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils.html">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>TVTensors</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/tv_tensors.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="tvtensors">
+<span id="tv-tensors"></span><h1>TVTensors<a class="headerlink" href="#tvtensors" title="Permalink to this heading">¶</a></h1>
+<p>TVTensors are <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclasses which the v2 <a class="reference internal" href="transforms.html#transforms"><span class="std std-ref">transforms</span></a> use under the hood to dispatch their inputs to the appropriate
+lower-level kernels. Most users do not need to manipulate TVTensors directly.</p>
+<p>Refer to
+<a class="reference internal" href="auto_examples/transforms/plot_transforms_getting_started.html#sphx-glr-auto-examples-transforms-plot-transforms-getting-started-py"><span class="std std-ref">Getting started with transforms v2</span></a> for
+an introduction to TVTensors, or
+<a class="reference internal" href="auto_examples/transforms/plot_tv_tensors.html#sphx-glr-auto-examples-transforms-plot-tv-tensors-py"><span class="std std-ref">TVTensors FAQ</span></a> for more advanced
+info.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.Image.html#torchvision.tv_tensors.Image" title="torchvision.tv_tensors.Image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Image</span></code></a>(data, *[, dtype, device, requires_grad])</p></td>
+<td><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for images.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.Video.html#torchvision.tv_tensors.Video" title="torchvision.tv_tensors.Video"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Video</span></code></a>(data, *[, dtype, device, requires_grad])</p></td>
+<td><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for videos.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.BoundingBoxFormat.html#torchvision.tv_tensors.BoundingBoxFormat" title="torchvision.tv_tensors.BoundingBoxFormat"><code class="xref py py-obj docutils literal notranslate"><span class="pre">BoundingBoxFormat</span></code></a>(value)</p></td>
+<td><p>[BETA] Coordinate format of a bounding box.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.BoundingBoxes.html#torchvision.tv_tensors.BoundingBoxes" title="torchvision.tv_tensors.BoundingBoxes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">BoundingBoxes</span></code></a>(data, *, format, canvas_size)</p></td>
+<td><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for bounding boxes.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.Mask.html#torchvision.tv_tensors.Mask" title="torchvision.tv_tensors.Mask"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Mask</span></code></a>(data, *[, dtype, device, requires_grad])</p></td>
+<td><p>[BETA] <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> subclass for segmentation and detection masks.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TVTensor</span></code></a></p></td>
+<td><p>[Beta] Base class for all TVTensors.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.set_return_type.html#torchvision.tv_tensors.set_return_type" title="torchvision.tv_tensors.set_return_type"><code class="xref py py-obj docutils literal notranslate"><span class="pre">set_return_type</span></code></a>(return_type)</p></td>
+<td><p>[BETA] Set the return type of torch operations on <a class="reference internal" href="generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.tv_tensors.wrap.html#torchvision.tv_tensors.wrap" title="torchvision.tv_tensors.wrap"><code class="xref py py-obj docutils literal notranslate"><span class="pre">wrap</span></code></a>(wrappee, *, like, **kwargs)</p></td>
+<td><p>[BETA] Convert a <a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">torch.Tensor</span></code></a> (<code class="docutils literal notranslate"><span class="pre">wrappee</span></code>) into the same <a class="reference internal" href="generated/torchvision.tv_tensors.TVTensor.html#torchvision.tv_tensors.TVTensor" title="torchvision.tv_tensors.TVTensor"><code class="xref py py-class docutils literal notranslate"><span class="pre">TVTensor</span></code></a> subclass as <code class="docutils literal notranslate"><span class="pre">like</span></code>.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.tv_tensors.Image.html" class="btn btn-neutral float-right" title="Image" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.transforms.functional.vflip.html" class="btn btn-neutral" title="vflip" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">TVTensors</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/0.16/utils.html b/0.16/utils.html
new file mode 100644
index 00000000000..0da0eb38478
--- /dev/null
+++ b/0.16/utils.html
@@ -0,0 +1,710 @@
+
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  <meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>Utils &mdash; Torchvision main documentation</title>
+  
+
+  
+  
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
+  <link rel="stylesheet" href="_static/pygments.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
+  <link rel="stylesheet" href="_static/copybutton.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-binder.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-dataframe.css" type="text/css" />
+  <link rel="stylesheet" href="_static/sg_gallery-rendered-html.css" type="text/css" />
+  <link rel="stylesheet" href="_static/css/custom_torchvision.css" type="text/css" />
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="draw_bounding_boxes" href="generated/torchvision.utils.draw_bounding_boxes.html" />
+    <link rel="prev" title="wrap_dataset_for_transforms_v2" href="generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html" />
+  <!-- Google Tag Manager -->
+    <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
+    new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
+    j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
+    'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
+    })(window,document,'script','dataLayer','GTM-T8XT4PS');</script>
+    <!-- End Google Tag Manager -->
+  
+
+  
+  <script src="_static/js/modernizr.min.js"></script>
+
+  <!-- Preload the theme fonts -->
+
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+
+<!-- Preload the katex fonts -->
+
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
+  <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
+</head>
+
+<div class="container-fluid header-holder tutorials-header" id="header-holder">
+  <div class="container">
+    <div class="header-container">
+      <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="active docs-active">
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-orange-arrow">
+                Docs
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
+                  <span class="dropdown-title">PyTorch</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/audio/stable/index.html">
+                  <span class="dropdown-title">torchaudio</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/text/stable/index.html">
+                  <span class="dropdown-title">torchtext</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/vision/stable/index.html">
+                  <span class="dropdown-title">torchvision</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torcharrow">
+                  <span class="dropdown-title">torcharrow</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/data">
+                  <span class="dropdown-title">TorchData</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchrec">
+                  <span class="dropdown-title">TorchRec</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/serve/">
+                  <span class="dropdown-title">TorchServe</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/torchx/">
+                  <span class="dropdown-title">TorchX</span>
+                  <p></p>
+                </a>
+                <a class="doc-dropdown-option nav-dropdown-item" href="https://pytorch.org/xla">
+                  <span class="dropdown-title">PyTorch on XLA Devices</span>
+                  <p></p>
+                </a>
+            </div>
+          </li>
+
+          <li>
+            <div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
+              <a class="resource-option with-down-arrow">
+                Resources
+              </a>
+              <div class="resources-dropdown-menu">
+                <a class="nav-dropdown-item" href="https://pytorch.org/features">
+                  <span class="dropdown-title">About</span>
+                  <p>Learn about PyTorch’s features and capabilities</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/foundation">
+                  <span class="dropdown-title">PyTorch Foundation</span>
+                  <p>Learn about the PyTorch foundation</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
+                  <span class="dropdown-title">Community</span>
+                  <p>Join the PyTorch developer community to contribute, learn, and get your questions answered.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
+                  <span class="dropdown-title">Community Stories</span>
+                  <p>Learn how our community solves real, everyday machine learning problems with PyTorch.</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/resources">
+                  <span class="dropdown-title">Developer Resources</span>
+                  <p>Find resources and get questions answered</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/events">
+                  <span class="dropdown-title">Events</span>
+                  <p>Find events, webinars, and podcasts</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
+                  <span class="dropdown-title">Forums</span>
+                  <p>A place to discuss PyTorch code, issues, install, research</p>
+                </a>
+                <a class="nav-dropdown-item" href="https://pytorch.org/hub">
+                  <span class="dropdown-title">Models (Beta)</span>
+                  <p>Discover, publish, and reuse pre-trained models</p>
+                </a>
+              </div>
+            </div>
+          </li>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">GitHub</a>
+          </li>
+        </ul>
+      </div>
+
+      <a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
+    </div>
+  </div>
+</div>
+
+<body class="pytorch-body">
+
+   
+
+    
+
+    <div class="table-of-contents-link-wrapper">
+      <span>Table of Contents</span>
+      <a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
+    </div>
+
+    <nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
+      <div class="pytorch-side-scroll">
+        <div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
+          <div class="pytorch-left-menu-search">
+            
+    <div class="version">
+      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+    </div>
+    
+
+
+  
+
+
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="search.html" method="get">
+    <input type="text" name="q" placeholder="Search Docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>
+
+          </div>
+
+          
+            
+            
+              
+            
+            
+              <p class="caption" role="heading"><span class="caption-text">Package Reference</span></p>
+<ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="transforms.html">Transforming and augmenting images</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tv_tensors.html">TVTensors</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models.html">Models and pre-trained weights</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets.html">Datasets</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">Utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="ops.html">Operators</a></li>
+<li class="toctree-l1"><a class="reference internal" href="io.html">Decoding / Encoding images and videos</a></li>
+<li class="toctree-l1"><a class="reference internal" href="feature_extraction.html">Feature extraction for model inspection</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">Examples and training references</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_examples/index.html">Examples and tutorials</a></li>
+<li class="toctree-l1"><a class="reference internal" href="training_references.html">Training references</a></li>
+</ul>
+<p class="caption" role="heading"><span class="caption-text">PyTorch Libraries</span></p>
+<ul>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/docs">PyTorch</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/vision">torchvision</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">TorchElastic</a></li>
+<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/serve">TorchServe</a></li>
+<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
+</ul>
+
+            
+          
+        </div>
+      </div>
+    </nav>
+
+    <div class="pytorch-container">
+      <div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
+        <div class="pytorch-breadcrumbs-wrapper">
+          
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+
+  <ul class="pytorch-breadcrumbs">
+    
+      <li>
+        <a href="index.html">
+          
+            Docs
+          
+        </a> &gt;
+      </li>
+
+        
+      <li>Utils</li>
+    
+    
+      <li class="pytorch-breadcrumbs-aside">
+        
+            
+            <a href="_sources/utils.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
+          
+        
+      </li>
+    
+  </ul>
+
+  
+</div>
+        </div>
+
+        <div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
+          Shortcuts
+        </div>
+      </div>
+
+      <section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
+        <div class="pytorch-content-left">
+
+        
+          <!-- Google Tag Manager (noscript) -->
+          <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-T8XT4PS"
+          height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
+          <!-- End Google Tag Manager (noscript) -->
+          
+          <div class="rst-content">
+          
+            <div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
+             <article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
+              
+  <section id="utils">
+<span id="id1"></span><h1>Utils<a class="headerlink" href="#utils" title="Permalink to this heading">¶</a></h1>
+<p>The <code class="docutils literal notranslate"><span class="pre">torchvision.utils</span></code> module contains various utilities, mostly <a class="reference internal" href="auto_examples/others/plot_visualization_utils.html#sphx-glr-auto-examples-others-plot-visualization-utils-py"><span class="std std-ref">for
+visualization</span></a>.</p>
+<table class="autosummary longtable docutils align-default">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.utils.draw_bounding_boxes.html#torchvision.utils.draw_bounding_boxes" title="torchvision.utils.draw_bounding_boxes"><code class="xref py py-obj docutils literal notranslate"><span class="pre">draw_bounding_boxes</span></code></a>(image, boxes[, labels, ...])</p></td>
+<td><p>Draws bounding boxes on given image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.utils.draw_segmentation_masks.html#torchvision.utils.draw_segmentation_masks" title="torchvision.utils.draw_segmentation_masks"><code class="xref py py-obj docutils literal notranslate"><span class="pre">draw_segmentation_masks</span></code></a>(image, masks[, ...])</p></td>
+<td><p>Draws segmentation masks on given RGB image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.utils.draw_keypoints.html#torchvision.utils.draw_keypoints" title="torchvision.utils.draw_keypoints"><code class="xref py py-obj docutils literal notranslate"><span class="pre">draw_keypoints</span></code></a>(image, keypoints[, ...])</p></td>
+<td><p>Draws Keypoints on given RGB image.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.utils.flow_to_image.html#torchvision.utils.flow_to_image" title="torchvision.utils.flow_to_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">flow_to_image</span></code></a>(flow)</p></td>
+<td><p>Converts a flow to an RGB image.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="generated/torchvision.utils.make_grid.html#torchvision.utils.make_grid" title="torchvision.utils.make_grid"><code class="xref py py-obj docutils literal notranslate"><span class="pre">make_grid</span></code></a>(tensor[, nrow, padding, ...])</p></td>
+<td><p>Make a grid of images.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="generated/torchvision.utils.save_image.html#torchvision.utils.save_image" title="torchvision.utils.save_image"><code class="xref py py-obj docutils literal notranslate"><span class="pre">save_image</span></code></a>(tensor, fp[, format])</p></td>
+<td><p>Save a given Tensor into an image file.</p></td>
+</tr>
+</tbody>
+</table>
+</section>
+
+
+             </article>
+             
+            </div>
+            <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="generated/torchvision.utils.draw_bounding_boxes.html" class="btn btn-neutral float-right" title="draw_bounding_boxes" accesskey="n" rel="next">Next <img src="_static/images/chevron-right-orange.svg" class="next-page"></a>
+      
+      
+        <a href="generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html" class="btn btn-neutral" title="wrap_dataset_for_transforms_v2" accesskey="p" rel="prev"><img src="_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
+      
+    </div>
+  
+
+  
+
+    <hr>
+
+  
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2017-present, Torch Contributors.
+
+    </p>
+  </div>
+    
+      <div>
+        Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
+      </div>
+     
+
+</footer>
+
+          </div>
+        </div>
+
+        <div class="pytorch-content-right" id="pytorch-content-right">
+          <div class="pytorch-right-menu" id="pytorch-right-menu">
+            <div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
+              <ul>
+<li><a class="reference internal" href="#">Utils</a></li>
+</ul>
+
+            </div>
+          </div>
+        </div>
+      </section>
+    </div>
+
+  
+
+
+  
+
+     
+       <script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
+         <script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
+         <script src="_static/jquery.js"></script>
+         <script src="_static/underscore.js"></script>
+         <script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
+         <script src="_static/doctools.js"></script>
+         <script src="_static/clipboard.min.js"></script>
+         <script src="_static/copybutton.js"></script>
+     
+
+  
+
+  <script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
+  <script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
+  <script type="text/javascript" src="_static/js/theme.js"></script>
+
+  <script type="text/javascript">
+      jQuery(function () {
+          SphinxRtdTheme.Navigation.enable(true);
+      });
+  </script> 
+
+  <!-- Begin Footer -->
+
+  <div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
+    <div class="container">
+      <div class="row">
+        <div class="col-md-4 text-center">
+          <h2>Docs</h2>
+          <p>Access comprehensive developer documentation for PyTorch</p>
+          <a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Tutorials</h2>
+          <p>Get in-depth tutorials for beginners and advanced developers</p>
+          <a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
+        </div>
+
+        <div class="col-md-4 text-center">
+          <h2>Resources</h2>
+          <p>Find development resources and get your questions answered</p>
+          <a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
+        </div>
+      </div>
+    </div>
+  </div>
+
+  <footer class="site-footer">
+    <div class="container footer-container">
+      <div class="footer-logo-wrapper">
+        <a href="https://pytorch.org/" class="footer-logo"></a>
+      </div>
+
+      <div class="footer-links-wrapper">
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
+            <li><a href="https://pytorch.org/get-started">Get Started</a></li>
+            <li><a href="https://pytorch.org/features">Features</a></li>
+            <li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
+            <li><a href="https://pytorch.org/blog/">Blog</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
+            <li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
+            <li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
+            <li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
+            <li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
+            <li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
+          </ul>
+        </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">Stay up to date</li>
+            <li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
+            <li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
+            <li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
+            <li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
+          </ul>  
+          </div>
+
+        <div class="footer-links-col">
+          <ul>
+            <li class="list-title">PyTorch Podcasts</li>
+            <li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
+            <li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
+            <li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
+            <li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
+          </ul>
+         </div>
+        </div>
+        
+        <div class="privacy-policy">
+          <ul>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
+            <li class="privacy-policy-links">|</li>
+            <li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
+          </ul>
+        </div>
+        <div class="copyright">
+        <p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
+          For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
+          <a href="www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
+          project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
+          please see <a href="www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
+      </div>
+     </div>
+
+  </footer>
+
+  <div class="cookie-banner-wrapper">
+  <div class="container">
+    <p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
+    <img class="close-button" src="_static/images/pytorch-x.svg">
+  </div>
+</div>
+
+  <!-- End Footer -->
+
+  <!-- Begin Mobile Menu -->
+
+  <div class="mobile-main-menu">
+    <div class="container-fluid">
+      <div class="container">
+        <div class="mobile-main-menu-header-container">
+          <a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
+          <a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
+        </div>
+      </div>
+    </div>
+
+    <div class="mobile-main-menu-links-container">
+      <div class="main-menu">
+        <ul>
+          <li>
+            <a href="https://pytorch.org/get-started">Get Started</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/ecosystem">Ecosystem</a>
+          </li>
+            
+          <li>
+            <a href="https://pytorch.org/mobile">Mobile</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/blog/">Blog</a>
+          </li>
+
+          <li>
+            <a href="https://pytorch.org/tutorials">Tutorials</a>
+          </li>
+
+          <li class="resources-mobile-menu-title" class="active">
+            Docs
+          </li>
+
+          <ul class="resources-mobile-menu-items">
+            <li>
+              <a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/audio/stable/index.html">torchaudio</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/text/stable/index.html">torchtext</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/vision/stable/index.html">torchvision</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torcharrow">torcharrow</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/data">TorchData</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchrec">TorchRec</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/serve/">TorchServe</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/torchx/">TorchX</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/xla">PyTorch on XLA Devices</a>
+            </li>
+          </ul>
+
+          <li class="resources-mobile-menu-title">
+            Resources
+          </li>
+            
+           <ul class="resources-mobile-menu-items">
+
+            <li>
+              <a href="https://pytorch.org/features">About</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/foundation">PyTorch Foundation</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/#community-module">Community</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/community-stories">Community Stories</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/resources">Developer Resources</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/events">Events</a>
+            </li>
+
+            <li>
+              <a href="https://discuss.pytorch.org/">Forums</a>
+            </li>
+
+            <li>
+              <a href="https://pytorch.org/hub">Models (Beta)</a>
+            </li>
+          </ul>
+
+          <li>
+            <a href="https://github.com/pytorch/pytorch">Github</a>
+          </li>
+        </ul>
+      </div>
+    </div>
+  </div>
+
+  <!-- End Mobile Menu -->
+
+  <script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>
+
+  <script type="text/javascript">
+    $(document).ready(function() {
+      mobileMenu.bind();
+      mobileTOC.bind();
+      pytorchAnchors.bind();
+      sideMenus.bind();
+      scrollToAnchor.bind();
+      highlightNavigation.bind();
+      mainMenuDropdown.bind();
+      filterTags.bind();
+
+      // Add class to links that have code blocks, since we cannot create links in code blocks
+      $("article.pytorch-article a span.pre").each(function(e) {
+        $(this).closest("a").addClass("has-code");
+      });
+    })
+  </script>
+</body>
+</html>
\ No newline at end of file

From 7c2ecab7073eb44b38c63910e8f3328190564004 Mon Sep 17 00:00:00 2001
From: Nicolas Hug <contact@nicolas-hug.com>
Date: Wed, 4 Oct 2023 17:43:18 +0100
Subject: [PATCH 2/5] main documentation -> 0.16 documentation

---
 0.16/_modules/index.html                                        | 2 +-
 0.16/_modules/torchvision.html                                  | 2 +-
 0.16/_modules/torchvision/datasets/_optical_flow.html           | 2 +-
 0.16/_modules/torchvision/datasets/_stereo_matching.html        | 2 +-
 0.16/_modules/torchvision/datasets/caltech.html                 | 2 +-
 0.16/_modules/torchvision/datasets/celeba.html                  | 2 +-
 0.16/_modules/torchvision/datasets/cifar.html                   | 2 +-
 0.16/_modules/torchvision/datasets/cityscapes.html              | 2 +-
 0.16/_modules/torchvision/datasets/coco.html                    | 2 +-
 0.16/_modules/torchvision/datasets/country211.html              | 2 +-
 0.16/_modules/torchvision/datasets/dtd.html                     | 2 +-
 0.16/_modules/torchvision/datasets/eurosat.html                 | 2 +-
 0.16/_modules/torchvision/datasets/fakedata.html                | 2 +-
 0.16/_modules/torchvision/datasets/fer2013.html                 | 2 +-
 0.16/_modules/torchvision/datasets/fgvc_aircraft.html           | 2 +-
 0.16/_modules/torchvision/datasets/flickr.html                  | 2 +-
 0.16/_modules/torchvision/datasets/flowers102.html              | 2 +-
 0.16/_modules/torchvision/datasets/folder.html                  | 2 +-
 0.16/_modules/torchvision/datasets/food101.html                 | 2 +-
 0.16/_modules/torchvision/datasets/gtsrb.html                   | 2 +-
 0.16/_modules/torchvision/datasets/hmdb51.html                  | 2 +-
 0.16/_modules/torchvision/datasets/imagenet.html                | 2 +-
 0.16/_modules/torchvision/datasets/inaturalist.html             | 2 +-
 0.16/_modules/torchvision/datasets/kinetics.html                | 2 +-
 0.16/_modules/torchvision/datasets/kitti.html                   | 2 +-
 0.16/_modules/torchvision/datasets/lfw.html                     | 2 +-
 0.16/_modules/torchvision/datasets/lsun.html                    | 2 +-
 0.16/_modules/torchvision/datasets/mnist.html                   | 2 +-
 0.16/_modules/torchvision/datasets/moving_mnist.html            | 2 +-
 0.16/_modules/torchvision/datasets/omniglot.html                | 2 +-
 0.16/_modules/torchvision/datasets/oxford_iiit_pet.html         | 2 +-
 0.16/_modules/torchvision/datasets/pcam.html                    | 2 +-
 0.16/_modules/torchvision/datasets/phototour.html               | 2 +-
 0.16/_modules/torchvision/datasets/places365.html               | 2 +-
 0.16/_modules/torchvision/datasets/rendered_sst2.html           | 2 +-
 0.16/_modules/torchvision/datasets/sbd.html                     | 2 +-
 0.16/_modules/torchvision/datasets/sbu.html                     | 2 +-
 0.16/_modules/torchvision/datasets/semeion.html                 | 2 +-
 0.16/_modules/torchvision/datasets/stanford_cars.html           | 2 +-
 0.16/_modules/torchvision/datasets/stl10.html                   | 2 +-
 0.16/_modules/torchvision/datasets/sun397.html                  | 2 +-
 0.16/_modules/torchvision/datasets/svhn.html                    | 2 +-
 0.16/_modules/torchvision/datasets/ucf101.html                  | 2 +-
 0.16/_modules/torchvision/datasets/usps.html                    | 2 +-
 0.16/_modules/torchvision/datasets/vision.html                  | 2 +-
 0.16/_modules/torchvision/datasets/voc.html                     | 2 +-
 0.16/_modules/torchvision/datasets/widerface.html               | 2 +-
 0.16/_modules/torchvision/io/image.html                         | 2 +-
 0.16/_modules/torchvision/io/video.html                         | 2 +-
 0.16/_modules/torchvision/io/video_reader.html                  | 2 +-
 0.16/_modules/torchvision/models/_api.html                      | 2 +-
 0.16/_modules/torchvision/models/alexnet.html                   | 2 +-
 0.16/_modules/torchvision/models/convnext.html                  | 2 +-
 0.16/_modules/torchvision/models/densenet.html                  | 2 +-
 0.16/_modules/torchvision/models/detection/faster_rcnn.html     | 2 +-
 0.16/_modules/torchvision/models/detection/fcos.html            | 2 +-
 0.16/_modules/torchvision/models/detection/keypoint_rcnn.html   | 2 +-
 0.16/_modules/torchvision/models/detection/mask_rcnn.html       | 2 +-
 0.16/_modules/torchvision/models/detection/retinanet.html       | 2 +-
 0.16/_modules/torchvision/models/detection/ssd.html             | 2 +-
 0.16/_modules/torchvision/models/detection/ssdlite.html         | 2 +-
 0.16/_modules/torchvision/models/efficientnet.html              | 2 +-
 0.16/_modules/torchvision/models/feature_extraction.html        | 2 +-
 0.16/_modules/torchvision/models/googlenet.html                 | 2 +-
 0.16/_modules/torchvision/models/inception.html                 | 2 +-
 0.16/_modules/torchvision/models/maxvit.html                    | 2 +-
 0.16/_modules/torchvision/models/mnasnet.html                   | 2 +-
 0.16/_modules/torchvision/models/mobilenetv2.html               | 2 +-
 0.16/_modules/torchvision/models/mobilenetv3.html               | 2 +-
 0.16/_modules/torchvision/models/optical_flow/raft.html         | 2 +-
 0.16/_modules/torchvision/models/quantization/googlenet.html    | 2 +-
 0.16/_modules/torchvision/models/quantization/inception.html    | 2 +-
 0.16/_modules/torchvision/models/quantization/mobilenetv2.html  | 2 +-
 0.16/_modules/torchvision/models/quantization/mobilenetv3.html  | 2 +-
 0.16/_modules/torchvision/models/quantization/resnet.html       | 2 +-
 0.16/_modules/torchvision/models/quantization/shufflenetv2.html | 2 +-
 0.16/_modules/torchvision/models/regnet.html                    | 2 +-
 0.16/_modules/torchvision/models/resnet.html                    | 2 +-
 0.16/_modules/torchvision/models/segmentation/deeplabv3.html    | 2 +-
 0.16/_modules/torchvision/models/segmentation/fcn.html          | 2 +-
 0.16/_modules/torchvision/models/segmentation/lraspp.html       | 2 +-
 0.16/_modules/torchvision/models/shufflenetv2.html              | 2 +-
 0.16/_modules/torchvision/models/squeezenet.html                | 2 +-
 0.16/_modules/torchvision/models/swin_transformer.html          | 2 +-
 0.16/_modules/torchvision/models/vgg.html                       | 2 +-
 0.16/_modules/torchvision/models/video/mvit.html                | 2 +-
 0.16/_modules/torchvision/models/video/resnet.html              | 2 +-
 0.16/_modules/torchvision/models/video/s3d.html                 | 2 +-
 0.16/_modules/torchvision/models/video/swin_transformer.html    | 2 +-
 0.16/_modules/torchvision/models/vision_transformer.html        | 2 +-
 0.16/_modules/torchvision/ops/boxes.html                        | 2 +-
 0.16/_modules/torchvision/ops/ciou_loss.html                    | 2 +-
 0.16/_modules/torchvision/ops/deform_conv.html                  | 2 +-
 0.16/_modules/torchvision/ops/diou_loss.html                    | 2 +-
 0.16/_modules/torchvision/ops/drop_block.html                   | 2 +-
 0.16/_modules/torchvision/ops/feature_pyramid_network.html      | 2 +-
 0.16/_modules/torchvision/ops/focal_loss.html                   | 2 +-
 0.16/_modules/torchvision/ops/giou_loss.html                    | 2 +-
 0.16/_modules/torchvision/ops/misc.html                         | 2 +-
 0.16/_modules/torchvision/ops/poolers.html                      | 2 +-
 0.16/_modules/torchvision/ops/ps_roi_align.html                 | 2 +-
 0.16/_modules/torchvision/ops/ps_roi_pool.html                  | 2 +-
 0.16/_modules/torchvision/ops/roi_align.html                    | 2 +-
 0.16/_modules/torchvision/ops/roi_pool.html                     | 2 +-
 0.16/_modules/torchvision/ops/stochastic_depth.html             | 2 +-
 0.16/_modules/torchvision/transforms/autoaugment.html           | 2 +-
 0.16/_modules/torchvision/transforms/functional.html            | 2 +-
 0.16/_modules/torchvision/transforms/transforms.html            | 2 +-
 0.16/_modules/torchvision/transforms/v2/_augment.html           | 2 +-
 0.16/_modules/torchvision/transforms/v2/_auto_augment.html      | 2 +-
 0.16/_modules/torchvision/transforms/v2/_color.html             | 2 +-
 0.16/_modules/torchvision/transforms/v2/_container.html         | 2 +-
 0.16/_modules/torchvision/transforms/v2/_deprecated.html        | 2 +-
 0.16/_modules/torchvision/transforms/v2/_geometry.html          | 2 +-
 0.16/_modules/torchvision/transforms/v2/_meta.html              | 2 +-
 0.16/_modules/torchvision/transforms/v2/_misc.html              | 2 +-
 0.16/_modules/torchvision/transforms/v2/_temporal.html          | 2 +-
 0.16/_modules/torchvision/transforms/v2/_type_conversion.html   | 2 +-
 .../_modules/torchvision/transforms/v2/functional/_augment.html | 2 +-
 0.16/_modules/torchvision/transforms/v2/functional/_color.html  | 2 +-
 .../torchvision/transforms/v2/functional/_deprecated.html       | 2 +-
 .../torchvision/transforms/v2/functional/_geometry.html         | 2 +-
 0.16/_modules/torchvision/transforms/v2/functional/_meta.html   | 2 +-
 0.16/_modules/torchvision/transforms/v2/functional/_misc.html   | 2 +-
 .../torchvision/transforms/v2/functional/_temporal.html         | 2 +-
 .../torchvision/transforms/v2/functional/_type_conversion.html  | 2 +-
 0.16/_modules/torchvision/transforms/v2/functional/_utils.html  | 2 +-
 0.16/_modules/torchvision/tv_tensors.html                       | 2 +-
 0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html       | 2 +-
 0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html      | 2 +-
 0.16/_modules/torchvision/tv_tensors/_image.html                | 2 +-
 0.16/_modules/torchvision/tv_tensors/_mask.html                 | 2 +-
 .../torchvision/tv_tensors/_torch_function_helpers.html         | 2 +-
 0.16/_modules/torchvision/tv_tensors/_tv_tensor.html            | 2 +-
 0.16/_modules/torchvision/tv_tensors/_video.html                | 2 +-
 0.16/_modules/torchvision/utils.html                            | 2 +-
 0.16/auto_examples/index.html                                   | 2 +-
 0.16/auto_examples/others/index.html                            | 2 +-
 0.16/auto_examples/others/plot_optical_flow.html                | 2 +-
 0.16/auto_examples/others/plot_repurposing_annotations.html     | 2 +-
 0.16/auto_examples/others/plot_scripted_tensor_transforms.html  | 2 +-
 0.16/auto_examples/others/plot_video_api.html                   | 2 +-
 0.16/auto_examples/others/plot_visualization_utils.html         | 2 +-
 0.16/auto_examples/others/sg_execution_times.html               | 2 +-
 0.16/auto_examples/transforms/index.html                        | 2 +-
 0.16/auto_examples/transforms/plot_custom_transforms.html       | 2 +-
 0.16/auto_examples/transforms/plot_custom_tv_tensors.html       | 2 +-
 0.16/auto_examples/transforms/plot_cutmix_mixup.html            | 2 +-
 0.16/auto_examples/transforms/plot_transforms_e2e.html          | 2 +-
 .../transforms/plot_transforms_getting_started.html             | 2 +-
 .../auto_examples/transforms/plot_transforms_illustrations.html | 2 +-
 0.16/auto_examples/transforms/plot_tv_tensors.html              | 2 +-
 0.16/auto_examples/transforms/sg_execution_times.html           | 2 +-
 0.16/datasets.html                                              | 2 +-
 0.16/feature_extraction.html                                    | 2 +-
 0.16/generated/classification_quant_table.html                  | 2 +-
 0.16/generated/classification_table.html                        | 2 +-
 0.16/generated/detection_keypoint_table.html                    | 2 +-
 0.16/generated/detection_table.html                             | 2 +-
 0.16/generated/instance_segmentation_table.html                 | 2 +-
 0.16/generated/segmentation_table.html                          | 2 +-
 0.16/generated/torchvision.datasets.CIFAR10.html                | 2 +-
 0.16/generated/torchvision.datasets.CIFAR100.html               | 2 +-
 0.16/generated/torchvision.datasets.CREStereo.html              | 2 +-
 0.16/generated/torchvision.datasets.Caltech101.html             | 2 +-
 0.16/generated/torchvision.datasets.Caltech256.html             | 2 +-
 0.16/generated/torchvision.datasets.CarlaStereo.html            | 2 +-
 0.16/generated/torchvision.datasets.CelebA.html                 | 2 +-
 0.16/generated/torchvision.datasets.Cityscapes.html             | 2 +-
 0.16/generated/torchvision.datasets.CocoCaptions.html           | 2 +-
 0.16/generated/torchvision.datasets.CocoDetection.html          | 2 +-
 0.16/generated/torchvision.datasets.Country211.html             | 2 +-
 0.16/generated/torchvision.datasets.DTD.html                    | 2 +-
 0.16/generated/torchvision.datasets.DatasetFolder.html          | 2 +-
 0.16/generated/torchvision.datasets.EMNIST.html                 | 2 +-
 0.16/generated/torchvision.datasets.ETH3DStereo.html            | 2 +-
 0.16/generated/torchvision.datasets.EuroSAT.html                | 2 +-
 0.16/generated/torchvision.datasets.FER2013.html                | 2 +-
 0.16/generated/torchvision.datasets.FGVCAircraft.html           | 2 +-
 0.16/generated/torchvision.datasets.FakeData.html               | 2 +-
 0.16/generated/torchvision.datasets.FallingThingsStereo.html    | 2 +-
 0.16/generated/torchvision.datasets.FashionMNIST.html           | 2 +-
 0.16/generated/torchvision.datasets.Flickr30k.html              | 2 +-
 0.16/generated/torchvision.datasets.Flickr8k.html               | 2 +-
 0.16/generated/torchvision.datasets.Flowers102.html             | 2 +-
 0.16/generated/torchvision.datasets.FlyingChairs.html           | 2 +-
 0.16/generated/torchvision.datasets.FlyingThings3D.html         | 2 +-
 0.16/generated/torchvision.datasets.Food101.html                | 2 +-
 0.16/generated/torchvision.datasets.GTSRB.html                  | 2 +-
 0.16/generated/torchvision.datasets.HD1K.html                   | 2 +-
 0.16/generated/torchvision.datasets.HMDB51.html                 | 2 +-
 0.16/generated/torchvision.datasets.INaturalist.html            | 2 +-
 0.16/generated/torchvision.datasets.ImageFolder.html            | 2 +-
 0.16/generated/torchvision.datasets.ImageNet.html               | 2 +-
 0.16/generated/torchvision.datasets.InStereo2k.html             | 2 +-
 0.16/generated/torchvision.datasets.KMNIST.html                 | 2 +-
 0.16/generated/torchvision.datasets.Kinetics.html               | 2 +-
 0.16/generated/torchvision.datasets.Kitti.html                  | 2 +-
 0.16/generated/torchvision.datasets.Kitti2012Stereo.html        | 2 +-
 0.16/generated/torchvision.datasets.Kitti2015Stereo.html        | 2 +-
 0.16/generated/torchvision.datasets.KittiFlow.html              | 2 +-
 0.16/generated/torchvision.datasets.LFWPairs.html               | 2 +-
 0.16/generated/torchvision.datasets.LFWPeople.html              | 2 +-
 0.16/generated/torchvision.datasets.LSUN.html                   | 2 +-
 0.16/generated/torchvision.datasets.MNIST.html                  | 2 +-
 0.16/generated/torchvision.datasets.Middlebury2014Stereo.html   | 2 +-
 0.16/generated/torchvision.datasets.MovingMNIST.html            | 2 +-
 0.16/generated/torchvision.datasets.Omniglot.html               | 2 +-
 0.16/generated/torchvision.datasets.OxfordIIITPet.html          | 2 +-
 0.16/generated/torchvision.datasets.PCAM.html                   | 2 +-
 0.16/generated/torchvision.datasets.PhotoTour.html              | 2 +-
 0.16/generated/torchvision.datasets.Places365.html              | 2 +-
 0.16/generated/torchvision.datasets.QMNIST.html                 | 2 +-
 0.16/generated/torchvision.datasets.RenderedSST2.html           | 2 +-
 0.16/generated/torchvision.datasets.SBDataset.html              | 2 +-
 0.16/generated/torchvision.datasets.SBU.html                    | 2 +-
 0.16/generated/torchvision.datasets.SEMEION.html                | 2 +-
 0.16/generated/torchvision.datasets.STL10.html                  | 2 +-
 0.16/generated/torchvision.datasets.SUN397.html                 | 2 +-
 0.16/generated/torchvision.datasets.SVHN.html                   | 2 +-
 0.16/generated/torchvision.datasets.SceneFlowStereo.html        | 2 +-
 0.16/generated/torchvision.datasets.Sintel.html                 | 2 +-
 0.16/generated/torchvision.datasets.SintelStereo.html           | 2 +-
 0.16/generated/torchvision.datasets.StanfordCars.html           | 2 +-
 0.16/generated/torchvision.datasets.UCF101.html                 | 2 +-
 0.16/generated/torchvision.datasets.USPS.html                   | 2 +-
 0.16/generated/torchvision.datasets.VOCDetection.html           | 2 +-
 0.16/generated/torchvision.datasets.VOCSegmentation.html        | 2 +-
 0.16/generated/torchvision.datasets.VisionDataset.html          | 2 +-
 0.16/generated/torchvision.datasets.WIDERFace.html              | 2 +-
 .../torchvision.datasets.wrap_dataset_for_transforms_v2.html    | 2 +-
 0.16/generated/torchvision.io.ImageReadMode.html                | 2 +-
 0.16/generated/torchvision.io.VideoReader.html                  | 2 +-
 0.16/generated/torchvision.io.decode_image.html                 | 2 +-
 0.16/generated/torchvision.io.decode_jpeg.html                  | 2 +-
 0.16/generated/torchvision.io.decode_png.html                   | 2 +-
 0.16/generated/torchvision.io.encode_jpeg.html                  | 2 +-
 0.16/generated/torchvision.io.encode_png.html                   | 2 +-
 0.16/generated/torchvision.io.read_file.html                    | 2 +-
 0.16/generated/torchvision.io.read_image.html                   | 2 +-
 0.16/generated/torchvision.io.read_video.html                   | 2 +-
 0.16/generated/torchvision.io.read_video_timestamps.html        | 2 +-
 0.16/generated/torchvision.io.write_file.html                   | 2 +-
 0.16/generated/torchvision.io.write_jpeg.html                   | 2 +-
 0.16/generated/torchvision.io.write_png.html                    | 2 +-
 0.16/generated/torchvision.io.write_video.html                  | 2 +-
 ...sion.models.feature_extraction.create_feature_extractor.html | 2 +-
 ...chvision.models.feature_extraction.get_graph_node_names.html | 2 +-
 0.16/generated/torchvision.models.get_model.html                | 2 +-
 0.16/generated/torchvision.models.get_model_weights.html        | 2 +-
 0.16/generated/torchvision.models.get_weight.html               | 2 +-
 0.16/generated/torchvision.models.list_models.html              | 2 +-
 0.16/generated/torchvision.ops.Conv2dNormActivation.html        | 2 +-
 0.16/generated/torchvision.ops.Conv3dNormActivation.html        | 2 +-
 0.16/generated/torchvision.ops.DeformConv2d.html                | 2 +-
 0.16/generated/torchvision.ops.DropBlock2d.html                 | 2 +-
 0.16/generated/torchvision.ops.DropBlock3d.html                 | 2 +-
 0.16/generated/torchvision.ops.FeaturePyramidNetwork.html       | 2 +-
 0.16/generated/torchvision.ops.FrozenBatchNorm2d.html           | 2 +-
 0.16/generated/torchvision.ops.MLP.html                         | 2 +-
 0.16/generated/torchvision.ops.MultiScaleRoIAlign.html          | 2 +-
 0.16/generated/torchvision.ops.PSRoIAlign.html                  | 2 +-
 0.16/generated/torchvision.ops.PSRoIPool.html                   | 2 +-
 0.16/generated/torchvision.ops.Permute.html                     | 2 +-
 0.16/generated/torchvision.ops.RoIAlign.html                    | 2 +-
 0.16/generated/torchvision.ops.RoIPool.html                     | 2 +-
 0.16/generated/torchvision.ops.SqueezeExcitation.html           | 2 +-
 0.16/generated/torchvision.ops.StochasticDepth.html             | 2 +-
 0.16/generated/torchvision.ops.batched_nms.html                 | 2 +-
 0.16/generated/torchvision.ops.box_area.html                    | 2 +-
 0.16/generated/torchvision.ops.box_convert.html                 | 2 +-
 0.16/generated/torchvision.ops.box_iou.html                     | 2 +-
 0.16/generated/torchvision.ops.clip_boxes_to_image.html         | 2 +-
 0.16/generated/torchvision.ops.complete_box_iou.html            | 2 +-
 0.16/generated/torchvision.ops.complete_box_iou_loss.html       | 2 +-
 0.16/generated/torchvision.ops.deform_conv2d.html               | 2 +-
 0.16/generated/torchvision.ops.distance_box_iou.html            | 2 +-
 0.16/generated/torchvision.ops.distance_box_iou_loss.html       | 2 +-
 0.16/generated/torchvision.ops.drop_block2d.html                | 2 +-
 0.16/generated/torchvision.ops.drop_block3d.html                | 2 +-
 0.16/generated/torchvision.ops.generalized_box_iou.html         | 2 +-
 0.16/generated/torchvision.ops.generalized_box_iou_loss.html    | 2 +-
 0.16/generated/torchvision.ops.masks_to_boxes.html              | 2 +-
 0.16/generated/torchvision.ops.nms.html                         | 2 +-
 0.16/generated/torchvision.ops.ps_roi_align.html                | 2 +-
 0.16/generated/torchvision.ops.ps_roi_pool.html                 | 2 +-
 0.16/generated/torchvision.ops.remove_small_boxes.html          | 2 +-
 0.16/generated/torchvision.ops.roi_align.html                   | 2 +-
 0.16/generated/torchvision.ops.roi_pool.html                    | 2 +-
 0.16/generated/torchvision.ops.sigmoid_focal_loss.html          | 2 +-
 0.16/generated/torchvision.ops.stochastic_depth.html            | 2 +-
 0.16/generated/torchvision.transforms.AugMix.html               | 2 +-
 0.16/generated/torchvision.transforms.AutoAugment.html          | 2 +-
 0.16/generated/torchvision.transforms.AutoAugmentPolicy.html    | 2 +-
 0.16/generated/torchvision.transforms.CenterCrop.html           | 2 +-
 0.16/generated/torchvision.transforms.ColorJitter.html          | 2 +-
 0.16/generated/torchvision.transforms.Compose.html              | 2 +-
 0.16/generated/torchvision.transforms.ConvertImageDtype.html    | 2 +-
 0.16/generated/torchvision.transforms.ElasticTransform.html     | 2 +-
 0.16/generated/torchvision.transforms.FiveCrop.html             | 2 +-
 0.16/generated/torchvision.transforms.GaussianBlur.html         | 2 +-
 0.16/generated/torchvision.transforms.Grayscale.html            | 2 +-
 0.16/generated/torchvision.transforms.Lambda.html               | 2 +-
 0.16/generated/torchvision.transforms.LinearTransformation.html | 2 +-
 0.16/generated/torchvision.transforms.Normalize.html            | 2 +-
 0.16/generated/torchvision.transforms.PILToTensor.html          | 2 +-
 0.16/generated/torchvision.transforms.Pad.html                  | 2 +-
 0.16/generated/torchvision.transforms.RandAugment.html          | 2 +-
 .../generated/torchvision.transforms.RandomAdjustSharpness.html | 2 +-
 0.16/generated/torchvision.transforms.RandomAffine.html         | 2 +-
 0.16/generated/torchvision.transforms.RandomApply.html          | 2 +-
 0.16/generated/torchvision.transforms.RandomAutocontrast.html   | 2 +-
 0.16/generated/torchvision.transforms.RandomChoice.html         | 2 +-
 0.16/generated/torchvision.transforms.RandomCrop.html           | 2 +-
 0.16/generated/torchvision.transforms.RandomEqualize.html       | 2 +-
 0.16/generated/torchvision.transforms.RandomErasing.html        | 2 +-
 0.16/generated/torchvision.transforms.RandomGrayscale.html      | 2 +-
 0.16/generated/torchvision.transforms.RandomHorizontalFlip.html | 2 +-
 0.16/generated/torchvision.transforms.RandomInvert.html         | 2 +-
 0.16/generated/torchvision.transforms.RandomOrder.html          | 2 +-
 0.16/generated/torchvision.transforms.RandomPerspective.html    | 2 +-
 0.16/generated/torchvision.transforms.RandomPosterize.html      | 2 +-
 0.16/generated/torchvision.transforms.RandomResizedCrop.html    | 2 +-
 0.16/generated/torchvision.transforms.RandomRotation.html       | 2 +-
 0.16/generated/torchvision.transforms.RandomSolarize.html       | 2 +-
 0.16/generated/torchvision.transforms.RandomVerticalFlip.html   | 2 +-
 0.16/generated/torchvision.transforms.Resize.html               | 2 +-
 0.16/generated/torchvision.transforms.TenCrop.html              | 2 +-
 0.16/generated/torchvision.transforms.ToPILImage.html           | 2 +-
 0.16/generated/torchvision.transforms.ToTensor.html             | 2 +-
 0.16/generated/torchvision.transforms.TrivialAugmentWide.html   | 2 +-
 .../torchvision.transforms.functional.adjust_brightness.html    | 2 +-
 .../torchvision.transforms.functional.adjust_contrast.html      | 2 +-
 .../torchvision.transforms.functional.adjust_gamma.html         | 2 +-
 .../generated/torchvision.transforms.functional.adjust_hue.html | 2 +-
 .../torchvision.transforms.functional.adjust_saturation.html    | 2 +-
 .../torchvision.transforms.functional.adjust_sharpness.html     | 2 +-
 0.16/generated/torchvision.transforms.functional.affine.html    | 2 +-
 .../torchvision.transforms.functional.autocontrast.html         | 2 +-
 .../torchvision.transforms.functional.center_crop.html          | 2 +-
 .../torchvision.transforms.functional.convert_image_dtype.html  | 2 +-
 0.16/generated/torchvision.transforms.functional.crop.html      | 2 +-
 0.16/generated/torchvision.transforms.functional.equalize.html  | 2 +-
 0.16/generated/torchvision.transforms.functional.erase.html     | 2 +-
 0.16/generated/torchvision.transforms.functional.five_crop.html | 2 +-
 .../torchvision.transforms.functional.gaussian_blur.html        | 2 +-
 .../torchvision.transforms.functional.get_dimensions.html       | 2 +-
 ...orchvision.transforms.functional.get_image_num_channels.html | 2 +-
 .../torchvision.transforms.functional.get_image_size.html       | 2 +-
 0.16/generated/torchvision.transforms.functional.hflip.html     | 2 +-
 0.16/generated/torchvision.transforms.functional.invert.html    | 2 +-
 0.16/generated/torchvision.transforms.functional.normalize.html | 2 +-
 0.16/generated/torchvision.transforms.functional.pad.html       | 2 +-
 .../torchvision.transforms.functional.perspective.html          | 2 +-
 .../torchvision.transforms.functional.pil_to_tensor.html        | 2 +-
 0.16/generated/torchvision.transforms.functional.posterize.html | 2 +-
 0.16/generated/torchvision.transforms.functional.resize.html    | 2 +-
 .../torchvision.transforms.functional.resized_crop.html         | 2 +-
 .../torchvision.transforms.functional.rgb_to_grayscale.html     | 2 +-
 0.16/generated/torchvision.transforms.functional.rotate.html    | 2 +-
 0.16/generated/torchvision.transforms.functional.solarize.html  | 2 +-
 0.16/generated/torchvision.transforms.functional.ten_crop.html  | 2 +-
 .../torchvision.transforms.functional.to_grayscale.html         | 2 +-
 .../torchvision.transforms.functional.to_pil_image.html         | 2 +-
 0.16/generated/torchvision.transforms.functional.to_tensor.html | 2 +-
 0.16/generated/torchvision.transforms.functional.vflip.html     | 2 +-
 0.16/generated/torchvision.transforms.v2.AugMix.html            | 2 +-
 0.16/generated/torchvision.transforms.v2.AutoAugment.html       | 2 +-
 0.16/generated/torchvision.transforms.v2.CenterCrop.html        | 2 +-
 .../generated/torchvision.transforms.v2.ClampBoundingBoxes.html | 2 +-
 0.16/generated/torchvision.transforms.v2.ColorJitter.html       | 2 +-
 0.16/generated/torchvision.transforms.v2.Compose.html           | 2 +-
 .../torchvision.transforms.v2.ConvertBoundingBoxFormat.html     | 2 +-
 0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html | 2 +-
 0.16/generated/torchvision.transforms.v2.CutMix.html            | 2 +-
 0.16/generated/torchvision.transforms.v2.ElasticTransform.html  | 2 +-
 0.16/generated/torchvision.transforms.v2.FiveCrop.html          | 2 +-
 0.16/generated/torchvision.transforms.v2.GaussianBlur.html      | 2 +-
 0.16/generated/torchvision.transforms.v2.Grayscale.html         | 2 +-
 0.16/generated/torchvision.transforms.v2.Lambda.html            | 2 +-
 .../torchvision.transforms.v2.LinearTransformation.html         | 2 +-
 0.16/generated/torchvision.transforms.v2.MixUp.html             | 2 +-
 0.16/generated/torchvision.transforms.v2.Normalize.html         | 2 +-
 0.16/generated/torchvision.transforms.v2.PILToTensor.html       | 2 +-
 0.16/generated/torchvision.transforms.v2.Pad.html               | 2 +-
 0.16/generated/torchvision.transforms.v2.RandAugment.html       | 2 +-
 .../torchvision.transforms.v2.RandomAdjustSharpness.html        | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomAffine.html      | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomApply.html       | 2 +-
 .../generated/torchvision.transforms.v2.RandomAutocontrast.html | 2 +-
 .../torchvision.transforms.v2.RandomChannelPermutation.html     | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomChoice.html      | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomCrop.html        | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomEqualize.html    | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomErasing.html     | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomGrayscale.html   | 2 +-
 .../torchvision.transforms.v2.RandomHorizontalFlip.html         | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomInvert.html      | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html     | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomOrder.html       | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomPerspective.html | 2 +-
 .../torchvision.transforms.v2.RandomPhotometricDistort.html     | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomPosterize.html   | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomResize.html      | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomRotation.html    | 2 +-
 .../generated/torchvision.transforms.v2.RandomShortestSize.html | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomSolarize.html    | 2 +-
 .../generated/torchvision.transforms.v2.RandomVerticalFlip.html | 2 +-
 0.16/generated/torchvision.transforms.v2.RandomZoomOut.html     | 2 +-
 0.16/generated/torchvision.transforms.v2.Resize.html            | 2 +-
 .../torchvision.transforms.v2.SanitizeBoundingBoxes.html        | 2 +-
 0.16/generated/torchvision.transforms.v2.ScaleJitter.html       | 2 +-
 0.16/generated/torchvision.transforms.v2.TenCrop.html           | 2 +-
 0.16/generated/torchvision.transforms.v2.ToDtype.html           | 2 +-
 0.16/generated/torchvision.transforms.v2.ToImage.html           | 2 +-
 0.16/generated/torchvision.transforms.v2.ToPILImage.html        | 2 +-
 0.16/generated/torchvision.transforms.v2.ToPureTensor.html      | 2 +-
 0.16/generated/torchvision.transforms.v2.ToTensor.html          | 2 +-
 .../generated/torchvision.transforms.v2.TrivialAugmentWide.html | 2 +-
 .../torchvision.transforms.v2.UniformTemporalSubsample.html     | 2 +-
 .../torchvision.transforms.v2.functional.adjust_brightness.html | 2 +-
 .../torchvision.transforms.v2.functional.adjust_contrast.html   | 2 +-
 .../torchvision.transforms.v2.functional.adjust_gamma.html      | 2 +-
 .../torchvision.transforms.v2.functional.adjust_hue.html        | 2 +-
 .../torchvision.transforms.v2.functional.adjust_saturation.html | 2 +-
 .../torchvision.transforms.v2.functional.adjust_sharpness.html  | 2 +-
 0.16/generated/torchvision.transforms.v2.functional.affine.html | 2 +-
 .../torchvision.transforms.v2.functional.autocontrast.html      | 2 +-
 .../torchvision.transforms.v2.functional.center_crop.html       | 2 +-
 ...rchvision.transforms.v2.functional.clamp_bounding_boxes.html | 2 +-
 ...on.transforms.v2.functional.convert_bounding_box_format.html | 2 +-
 ...orchvision.transforms.v2.functional.convert_image_dtype.html | 2 +-
 0.16/generated/torchvision.transforms.v2.functional.crop.html   | 2 +-
 .../generated/torchvision.transforms.v2.functional.elastic.html | 2 +-
 .../torchvision.transforms.v2.functional.equalize.html          | 2 +-
 0.16/generated/torchvision.transforms.v2.functional.erase.html  | 2 +-
 .../torchvision.transforms.v2.functional.five_crop.html         | 2 +-
 .../torchvision.transforms.v2.functional.gaussian_blur.html     | 2 +-
 .../torchvision.transforms.v2.functional.horizontal_flip.html   | 2 +-
 0.16/generated/torchvision.transforms.v2.functional.invert.html | 2 +-
 .../torchvision.transforms.v2.functional.normalize.html         | 2 +-
 0.16/generated/torchvision.transforms.v2.functional.pad.html    | 2 +-
 .../torchvision.transforms.v2.functional.permute_channels.html  | 2 +-
 .../torchvision.transforms.v2.functional.perspective.html       | 2 +-
 .../torchvision.transforms.v2.functional.pil_to_tensor.html     | 2 +-
 .../torchvision.transforms.v2.functional.posterize.html         | 2 +-
 .../torchvision.transforms.v2.functional.register_kernel.html   | 2 +-
 0.16/generated/torchvision.transforms.v2.functional.resize.html | 2 +-
 .../torchvision.transforms.v2.functional.resized_crop.html      | 2 +-
 .../torchvision.transforms.v2.functional.rgb_to_grayscale.html  | 2 +-
 0.16/generated/torchvision.transforms.v2.functional.rotate.html | 2 +-
 .../torchvision.transforms.v2.functional.solarize.html          | 2 +-
 .../torchvision.transforms.v2.functional.ten_crop.html          | 2 +-
 .../torchvision.transforms.v2.functional.to_dtype.html          | 2 +-
 .../torchvision.transforms.v2.functional.to_grayscale.html      | 2 +-
 .../torchvision.transforms.v2.functional.to_image.html          | 2 +-
 .../torchvision.transforms.v2.functional.to_pil_image.html      | 2 +-
 .../torchvision.transforms.v2.functional.to_tensor.html         | 2 +-
 ...ion.transforms.v2.functional.uniform_temporal_subsample.html | 2 +-
 .../torchvision.transforms.v2.functional.vertical_flip.html     | 2 +-
 0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html    | 2 +-
 0.16/generated/torchvision.tv_tensors.BoundingBoxes.html        | 2 +-
 0.16/generated/torchvision.tv_tensors.Image.html                | 2 +-
 0.16/generated/torchvision.tv_tensors.Mask.html                 | 2 +-
 0.16/generated/torchvision.tv_tensors.TVTensor.html             | 2 +-
 0.16/generated/torchvision.tv_tensors.Video.html                | 2 +-
 0.16/generated/torchvision.tv_tensors.set_return_type.html      | 2 +-
 0.16/generated/torchvision.tv_tensors.wrap.html                 | 2 +-
 0.16/generated/torchvision.utils.draw_bounding_boxes.html       | 2 +-
 0.16/generated/torchvision.utils.draw_keypoints.html            | 2 +-
 0.16/generated/torchvision.utils.draw_segmentation_masks.html   | 2 +-
 0.16/generated/torchvision.utils.flow_to_image.html             | 2 +-
 0.16/generated/torchvision.utils.make_grid.html                 | 2 +-
 0.16/generated/torchvision.utils.save_image.html                | 2 +-
 0.16/generated/video_table.html                                 | 2 +-
 0.16/genindex.html                                              | 2 +-
 0.16/index.html                                                 | 2 +-
 0.16/io.html                                                    | 2 +-
 0.16/models.html                                                | 2 +-
 0.16/models/alexnet.html                                        | 2 +-
 0.16/models/convnext.html                                       | 2 +-
 0.16/models/deeplabv3.html                                      | 2 +-
 0.16/models/densenet.html                                       | 2 +-
 0.16/models/efficientnet.html                                   | 2 +-
 0.16/models/efficientnetv2.html                                 | 2 +-
 0.16/models/faster_rcnn.html                                    | 2 +-
 0.16/models/fcn.html                                            | 2 +-
 0.16/models/fcos.html                                           | 2 +-
 0.16/models/generated/torchvision.models.alexnet.html           | 2 +-
 0.16/models/generated/torchvision.models.convnext_base.html     | 2 +-
 0.16/models/generated/torchvision.models.convnext_large.html    | 2 +-
 0.16/models/generated/torchvision.models.convnext_small.html    | 2 +-
 0.16/models/generated/torchvision.models.convnext_tiny.html     | 2 +-
 0.16/models/generated/torchvision.models.densenet121.html       | 2 +-
 0.16/models/generated/torchvision.models.densenet161.html       | 2 +-
 0.16/models/generated/torchvision.models.densenet169.html       | 2 +-
 0.16/models/generated/torchvision.models.densenet201.html       | 2 +-
 ....models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html | 2 +-
 ...sion.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html | 2 +-
 .../torchvision.models.detection.fasterrcnn_resnet50_fpn.html   | 2 +-
 ...torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html | 2 +-
 .../torchvision.models.detection.fcos_resnet50_fpn.html         | 2 +-
 .../torchvision.models.detection.keypointrcnn_resnet50_fpn.html | 2 +-
 .../torchvision.models.detection.maskrcnn_resnet50_fpn.html     | 2 +-
 .../torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html  | 2 +-
 .../torchvision.models.detection.retinanet_resnet50_fpn.html    | 2 +-
 .../torchvision.models.detection.retinanet_resnet50_fpn_v2.html | 2 +-
 .../generated/torchvision.models.detection.ssd300_vgg16.html    | 2 +-
 ...chvision.models.detection.ssdlite320_mobilenet_v3_large.html | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b0.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b1.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b2.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b3.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b4.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b5.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b6.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_b7.html   | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_v2_l.html | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_v2_m.html | 2 +-
 0.16/models/generated/torchvision.models.efficientnet_v2_s.html | 2 +-
 0.16/models/generated/torchvision.models.googlenet.html         | 2 +-
 0.16/models/generated/torchvision.models.inception_v3.html      | 2 +-
 0.16/models/generated/torchvision.models.maxvit_t.html          | 2 +-
 0.16/models/generated/torchvision.models.mnasnet0_5.html        | 2 +-
 0.16/models/generated/torchvision.models.mnasnet0_75.html       | 2 +-
 0.16/models/generated/torchvision.models.mnasnet1_0.html        | 2 +-
 0.16/models/generated/torchvision.models.mnasnet1_3.html        | 2 +-
 0.16/models/generated/torchvision.models.mobilenet_v2.html      | 2 +-
 .../models/generated/torchvision.models.mobilenet_v3_large.html | 2 +-
 .../models/generated/torchvision.models.mobilenet_v3_small.html | 2 +-
 .../generated/torchvision.models.optical_flow.raft_large.html   | 2 +-
 .../generated/torchvision.models.optical_flow.raft_small.html   | 2 +-
 .../generated/torchvision.models.quantization.googlenet.html    | 2 +-
 .../generated/torchvision.models.quantization.inception_v3.html | 2 +-
 .../generated/torchvision.models.quantization.mobilenet_v2.html | 2 +-
 .../torchvision.models.quantization.mobilenet_v3_large.html     | 2 +-
 .../generated/torchvision.models.quantization.resnet18.html     | 2 +-
 .../generated/torchvision.models.quantization.resnet50.html     | 2 +-
 .../torchvision.models.quantization.resnext101_32x8d.html       | 2 +-
 .../torchvision.models.quantization.resnext101_64x4d.html       | 2 +-
 .../torchvision.models.quantization.shufflenet_v2_x0_5.html     | 2 +-
 .../torchvision.models.quantization.shufflenet_v2_x1_0.html     | 2 +-
 .../torchvision.models.quantization.shufflenet_v2_x1_5.html     | 2 +-
 .../torchvision.models.quantization.shufflenet_v2_x2_0.html     | 2 +-
 0.16/models/generated/torchvision.models.regnet_x_16gf.html     | 2 +-
 0.16/models/generated/torchvision.models.regnet_x_1_6gf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_x_32gf.html     | 2 +-
 0.16/models/generated/torchvision.models.regnet_x_3_2gf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_x_400mf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_x_800mf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_x_8gf.html      | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_128gf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_16gf.html     | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_1_6gf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_32gf.html     | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_3_2gf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_400mf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_800mf.html    | 2 +-
 0.16/models/generated/torchvision.models.regnet_y_8gf.html      | 2 +-
 0.16/models/generated/torchvision.models.resnet101.html         | 2 +-
 0.16/models/generated/torchvision.models.resnet152.html         | 2 +-
 0.16/models/generated/torchvision.models.resnet18.html          | 2 +-
 0.16/models/generated/torchvision.models.resnet34.html          | 2 +-
 0.16/models/generated/torchvision.models.resnet50.html          | 2 +-
 0.16/models/generated/torchvision.models.resnext101_32x8d.html  | 2 +-
 0.16/models/generated/torchvision.models.resnext101_64x4d.html  | 2 +-
 0.16/models/generated/torchvision.models.resnext50_32x4d.html   | 2 +-
 ...vision.models.segmentation.deeplabv3_mobilenet_v3_large.html | 2 +-
 .../torchvision.models.segmentation.deeplabv3_resnet101.html    | 2 +-
 .../torchvision.models.segmentation.deeplabv3_resnet50.html     | 2 +-
 .../torchvision.models.segmentation.fcn_resnet101.html          | 2 +-
 .../generated/torchvision.models.segmentation.fcn_resnet50.html | 2 +-
 ...rchvision.models.segmentation.lraspp_mobilenet_v3_large.html | 2 +-
 .../models/generated/torchvision.models.shufflenet_v2_x0_5.html | 2 +-
 .../models/generated/torchvision.models.shufflenet_v2_x1_0.html | 2 +-
 .../models/generated/torchvision.models.shufflenet_v2_x1_5.html | 2 +-
 .../models/generated/torchvision.models.shufflenet_v2_x2_0.html | 2 +-
 0.16/models/generated/torchvision.models.squeezenet1_0.html     | 2 +-
 0.16/models/generated/torchvision.models.squeezenet1_1.html     | 2 +-
 0.16/models/generated/torchvision.models.swin_b.html            | 2 +-
 0.16/models/generated/torchvision.models.swin_s.html            | 2 +-
 0.16/models/generated/torchvision.models.swin_t.html            | 2 +-
 0.16/models/generated/torchvision.models.swin_v2_b.html         | 2 +-
 0.16/models/generated/torchvision.models.swin_v2_s.html         | 2 +-
 0.16/models/generated/torchvision.models.swin_v2_t.html         | 2 +-
 0.16/models/generated/torchvision.models.vgg11.html             | 2 +-
 0.16/models/generated/torchvision.models.vgg11_bn.html          | 2 +-
 0.16/models/generated/torchvision.models.vgg13.html             | 2 +-
 0.16/models/generated/torchvision.models.vgg13_bn.html          | 2 +-
 0.16/models/generated/torchvision.models.vgg16.html             | 2 +-
 0.16/models/generated/torchvision.models.vgg16_bn.html          | 2 +-
 0.16/models/generated/torchvision.models.vgg19.html             | 2 +-
 0.16/models/generated/torchvision.models.vgg19_bn.html          | 2 +-
 0.16/models/generated/torchvision.models.video.mc3_18.html      | 2 +-
 0.16/models/generated/torchvision.models.video.mvit_v1_b.html   | 2 +-
 0.16/models/generated/torchvision.models.video.mvit_v2_s.html   | 2 +-
 0.16/models/generated/torchvision.models.video.r2plus1d_18.html | 2 +-
 0.16/models/generated/torchvision.models.video.r3d_18.html      | 2 +-
 0.16/models/generated/torchvision.models.video.s3d.html         | 2 +-
 0.16/models/generated/torchvision.models.video.swin3d_b.html    | 2 +-
 0.16/models/generated/torchvision.models.video.swin3d_s.html    | 2 +-
 0.16/models/generated/torchvision.models.video.swin3d_t.html    | 2 +-
 0.16/models/generated/torchvision.models.vit_b_16.html          | 2 +-
 0.16/models/generated/torchvision.models.vit_b_32.html          | 2 +-
 0.16/models/generated/torchvision.models.vit_h_14.html          | 2 +-
 0.16/models/generated/torchvision.models.vit_l_16.html          | 2 +-
 0.16/models/generated/torchvision.models.vit_l_32.html          | 2 +-
 0.16/models/generated/torchvision.models.wide_resnet101_2.html  | 2 +-
 0.16/models/generated/torchvision.models.wide_resnet50_2.html   | 2 +-
 0.16/models/googlenet.html                                      | 2 +-
 0.16/models/googlenet_quant.html                                | 2 +-
 0.16/models/inception.html                                      | 2 +-
 0.16/models/inception_quant.html                                | 2 +-
 0.16/models/keypoint_rcnn.html                                  | 2 +-
 0.16/models/lraspp.html                                         | 2 +-
 0.16/models/mask_rcnn.html                                      | 2 +-
 0.16/models/maxvit.html                                         | 2 +-
 0.16/models/mnasnet.html                                        | 2 +-
 0.16/models/mobilenetv2.html                                    | 2 +-
 0.16/models/mobilenetv2_quant.html                              | 2 +-
 0.16/models/mobilenetv3.html                                    | 2 +-
 0.16/models/mobilenetv3_quant.html                              | 2 +-
 0.16/models/raft.html                                           | 2 +-
 0.16/models/regnet.html                                         | 2 +-
 0.16/models/resnet.html                                         | 2 +-
 0.16/models/resnet_quant.html                                   | 2 +-
 0.16/models/resnext.html                                        | 2 +-
 0.16/models/resnext_quant.html                                  | 2 +-
 0.16/models/retinanet.html                                      | 2 +-
 0.16/models/shufflenetv2.html                                   | 2 +-
 0.16/models/shufflenetv2_quant.html                             | 2 +-
 0.16/models/squeezenet.html                                     | 2 +-
 0.16/models/ssd.html                                            | 2 +-
 0.16/models/ssdlite.html                                        | 2 +-
 0.16/models/swin_transformer.html                               | 2 +-
 0.16/models/vgg.html                                            | 2 +-
 0.16/models/video_mvit.html                                     | 2 +-
 0.16/models/video_resnet.html                                   | 2 +-
 0.16/models/video_s3d.html                                      | 2 +-
 0.16/models/video_swin_transformer.html                         | 2 +-
 0.16/models/vision_transformer.html                             | 2 +-
 0.16/models/wide_resnet.html                                    | 2 +-
 0.16/ops.html                                                   | 2 +-
 0.16/py-modindex.html                                           | 2 +-
 0.16/search.html                                                | 2 +-
 0.16/training_references.html                                   | 2 +-
 0.16/transforms.html                                            | 2 +-
 0.16/tv_tensors.html                                            | 2 +-
 0.16/utils.html                                                 | 2 +-
 650 files changed, 650 insertions(+), 650 deletions(-)

diff --git a/0.16/_modules/index.html b/0.16/_modules/index.html
index 19e831020c9..b46f8605d2a 100644
--- a/0.16/_modules/index.html
+++ b/0.16/_modules/index.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Overview: module code &mdash; Torchvision main documentation</title>
+  <title>Overview: module code &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision.html b/0.16/_modules/torchvision.html
index 062e1be30dd..6bb9a84cb83 100644
--- a/0.16/_modules/torchvision.html
+++ b/0.16/_modules/torchvision.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision &mdash; Torchvision main documentation</title>
+  <title>torchvision &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/_optical_flow.html b/0.16/_modules/torchvision/datasets/_optical_flow.html
index cea3d728c3d..a1b2e904968 100644
--- a/0.16/_modules/torchvision/datasets/_optical_flow.html
+++ b/0.16/_modules/torchvision/datasets/_optical_flow.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets._optical_flow &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets._optical_flow &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/_stereo_matching.html b/0.16/_modules/torchvision/datasets/_stereo_matching.html
index 9930f07a3a6..2b74e331e6f 100644
--- a/0.16/_modules/torchvision/datasets/_stereo_matching.html
+++ b/0.16/_modules/torchvision/datasets/_stereo_matching.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets._stereo_matching &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets._stereo_matching &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/caltech.html b/0.16/_modules/torchvision/datasets/caltech.html
index 28d1f009775..9806c9cb7a4 100644
--- a/0.16/_modules/torchvision/datasets/caltech.html
+++ b/0.16/_modules/torchvision/datasets/caltech.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.caltech &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.caltech &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/celeba.html b/0.16/_modules/torchvision/datasets/celeba.html
index 112779d3fe0..abbfeb4f9d8 100644
--- a/0.16/_modules/torchvision/datasets/celeba.html
+++ b/0.16/_modules/torchvision/datasets/celeba.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.celeba &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.celeba &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/cifar.html b/0.16/_modules/torchvision/datasets/cifar.html
index fd991ca9ad2..94d4fb8f45a 100644
--- a/0.16/_modules/torchvision/datasets/cifar.html
+++ b/0.16/_modules/torchvision/datasets/cifar.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.cifar &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.cifar &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/cityscapes.html b/0.16/_modules/torchvision/datasets/cityscapes.html
index 2c7a4dfbe46..19ff5de0170 100644
--- a/0.16/_modules/torchvision/datasets/cityscapes.html
+++ b/0.16/_modules/torchvision/datasets/cityscapes.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.cityscapes &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.cityscapes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/coco.html b/0.16/_modules/torchvision/datasets/coco.html
index 20790b2b53a..7e8d3f3916f 100644
--- a/0.16/_modules/torchvision/datasets/coco.html
+++ b/0.16/_modules/torchvision/datasets/coco.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.coco &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.coco &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/country211.html b/0.16/_modules/torchvision/datasets/country211.html
index 02953139322..3974110fadb 100644
--- a/0.16/_modules/torchvision/datasets/country211.html
+++ b/0.16/_modules/torchvision/datasets/country211.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.country211 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.country211 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/dtd.html b/0.16/_modules/torchvision/datasets/dtd.html
index 29393635451..9b7bd8d07c4 100644
--- a/0.16/_modules/torchvision/datasets/dtd.html
+++ b/0.16/_modules/torchvision/datasets/dtd.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.dtd &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.dtd &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/eurosat.html b/0.16/_modules/torchvision/datasets/eurosat.html
index 3c270c93435..4b688af2cb2 100644
--- a/0.16/_modules/torchvision/datasets/eurosat.html
+++ b/0.16/_modules/torchvision/datasets/eurosat.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.eurosat &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.eurosat &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/fakedata.html b/0.16/_modules/torchvision/datasets/fakedata.html
index 21012c8272d..a87ceac3e38 100644
--- a/0.16/_modules/torchvision/datasets/fakedata.html
+++ b/0.16/_modules/torchvision/datasets/fakedata.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.fakedata &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.fakedata &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/fer2013.html b/0.16/_modules/torchvision/datasets/fer2013.html
index bd368498a1d..e85f381f5a3 100644
--- a/0.16/_modules/torchvision/datasets/fer2013.html
+++ b/0.16/_modules/torchvision/datasets/fer2013.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.fer2013 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.fer2013 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/fgvc_aircraft.html b/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
index e8aaa40b758..c7ce59f1e0b 100644
--- a/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
+++ b/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.fgvc_aircraft &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.fgvc_aircraft &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/flickr.html b/0.16/_modules/torchvision/datasets/flickr.html
index 36b4fc2ba7f..38f229040ac 100644
--- a/0.16/_modules/torchvision/datasets/flickr.html
+++ b/0.16/_modules/torchvision/datasets/flickr.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.flickr &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.flickr &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/flowers102.html b/0.16/_modules/torchvision/datasets/flowers102.html
index e786a1673e8..d4832c29e95 100644
--- a/0.16/_modules/torchvision/datasets/flowers102.html
+++ b/0.16/_modules/torchvision/datasets/flowers102.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.flowers102 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.flowers102 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/folder.html b/0.16/_modules/torchvision/datasets/folder.html
index 701163743e0..b2a1a2e1d50 100644
--- a/0.16/_modules/torchvision/datasets/folder.html
+++ b/0.16/_modules/torchvision/datasets/folder.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.folder &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.folder &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/food101.html b/0.16/_modules/torchvision/datasets/food101.html
index abcf0a5fc30..205c5792aa7 100644
--- a/0.16/_modules/torchvision/datasets/food101.html
+++ b/0.16/_modules/torchvision/datasets/food101.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.food101 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.food101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/gtsrb.html b/0.16/_modules/torchvision/datasets/gtsrb.html
index 75d1b2c02ad..41d3986c306 100644
--- a/0.16/_modules/torchvision/datasets/gtsrb.html
+++ b/0.16/_modules/torchvision/datasets/gtsrb.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.gtsrb &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.gtsrb &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/hmdb51.html b/0.16/_modules/torchvision/datasets/hmdb51.html
index c16c147ceef..ac46e953395 100644
--- a/0.16/_modules/torchvision/datasets/hmdb51.html
+++ b/0.16/_modules/torchvision/datasets/hmdb51.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.hmdb51 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.hmdb51 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/imagenet.html b/0.16/_modules/torchvision/datasets/imagenet.html
index e4db71fa2b7..c87fe52465f 100644
--- a/0.16/_modules/torchvision/datasets/imagenet.html
+++ b/0.16/_modules/torchvision/datasets/imagenet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.imagenet &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.imagenet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/inaturalist.html b/0.16/_modules/torchvision/datasets/inaturalist.html
index 1953205206c..cb4a896ae34 100644
--- a/0.16/_modules/torchvision/datasets/inaturalist.html
+++ b/0.16/_modules/torchvision/datasets/inaturalist.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.inaturalist &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.inaturalist &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/kinetics.html b/0.16/_modules/torchvision/datasets/kinetics.html
index f95faf23bd9..9c3e13a5d28 100644
--- a/0.16/_modules/torchvision/datasets/kinetics.html
+++ b/0.16/_modules/torchvision/datasets/kinetics.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.kinetics &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.kinetics &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/kitti.html b/0.16/_modules/torchvision/datasets/kitti.html
index bbc4f2c4f79..cff517c4c61 100644
--- a/0.16/_modules/torchvision/datasets/kitti.html
+++ b/0.16/_modules/torchvision/datasets/kitti.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.kitti &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.kitti &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/lfw.html b/0.16/_modules/torchvision/datasets/lfw.html
index ec7948db64d..0b2b0b6d4ed 100644
--- a/0.16/_modules/torchvision/datasets/lfw.html
+++ b/0.16/_modules/torchvision/datasets/lfw.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.lfw &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.lfw &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/lsun.html b/0.16/_modules/torchvision/datasets/lsun.html
index 0c84756b4d6..620cf680144 100644
--- a/0.16/_modules/torchvision/datasets/lsun.html
+++ b/0.16/_modules/torchvision/datasets/lsun.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.lsun &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.lsun &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/mnist.html b/0.16/_modules/torchvision/datasets/mnist.html
index f7d461d157c..7f749e8666d 100644
--- a/0.16/_modules/torchvision/datasets/mnist.html
+++ b/0.16/_modules/torchvision/datasets/mnist.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.mnist &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.mnist &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/moving_mnist.html b/0.16/_modules/torchvision/datasets/moving_mnist.html
index a606375a723..bb0a65c4b3c 100644
--- a/0.16/_modules/torchvision/datasets/moving_mnist.html
+++ b/0.16/_modules/torchvision/datasets/moving_mnist.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.moving_mnist &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.moving_mnist &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/omniglot.html b/0.16/_modules/torchvision/datasets/omniglot.html
index e7651c15dbe..29d2b0e361f 100644
--- a/0.16/_modules/torchvision/datasets/omniglot.html
+++ b/0.16/_modules/torchvision/datasets/omniglot.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.omniglot &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.omniglot &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html b/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
index f7fd048bb00..081170449fb 100644
--- a/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
+++ b/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.oxford_iiit_pet &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.oxford_iiit_pet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/pcam.html b/0.16/_modules/torchvision/datasets/pcam.html
index 2e9ac21fc89..a348fa29f92 100644
--- a/0.16/_modules/torchvision/datasets/pcam.html
+++ b/0.16/_modules/torchvision/datasets/pcam.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.pcam &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.pcam &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/phototour.html b/0.16/_modules/torchvision/datasets/phototour.html
index 470e6b6dc8a..fb96c141c06 100644
--- a/0.16/_modules/torchvision/datasets/phototour.html
+++ b/0.16/_modules/torchvision/datasets/phototour.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.phototour &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.phototour &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/places365.html b/0.16/_modules/torchvision/datasets/places365.html
index 0818d529945..86730d25ba6 100644
--- a/0.16/_modules/torchvision/datasets/places365.html
+++ b/0.16/_modules/torchvision/datasets/places365.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.places365 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.places365 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/rendered_sst2.html b/0.16/_modules/torchvision/datasets/rendered_sst2.html
index 2afb77f7d44..1df536a560f 100644
--- a/0.16/_modules/torchvision/datasets/rendered_sst2.html
+++ b/0.16/_modules/torchvision/datasets/rendered_sst2.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.rendered_sst2 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.rendered_sst2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/sbd.html b/0.16/_modules/torchvision/datasets/sbd.html
index 98721d5627e..6cea539dcd4 100644
--- a/0.16/_modules/torchvision/datasets/sbd.html
+++ b/0.16/_modules/torchvision/datasets/sbd.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.sbd &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.sbd &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/sbu.html b/0.16/_modules/torchvision/datasets/sbu.html
index 5e9df7bd2bb..ebf51754f3d 100644
--- a/0.16/_modules/torchvision/datasets/sbu.html
+++ b/0.16/_modules/torchvision/datasets/sbu.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.sbu &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.sbu &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/semeion.html b/0.16/_modules/torchvision/datasets/semeion.html
index a81802673b2..dd67e2dc2f5 100644
--- a/0.16/_modules/torchvision/datasets/semeion.html
+++ b/0.16/_modules/torchvision/datasets/semeion.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.semeion &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.semeion &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/stanford_cars.html b/0.16/_modules/torchvision/datasets/stanford_cars.html
index 73cf9ef13a5..a7dbbc9f78d 100644
--- a/0.16/_modules/torchvision/datasets/stanford_cars.html
+++ b/0.16/_modules/torchvision/datasets/stanford_cars.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.stanford_cars &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.stanford_cars &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/stl10.html b/0.16/_modules/torchvision/datasets/stl10.html
index 2385a7b94c5..4893576d91d 100644
--- a/0.16/_modules/torchvision/datasets/stl10.html
+++ b/0.16/_modules/torchvision/datasets/stl10.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.stl10 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.stl10 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/sun397.html b/0.16/_modules/torchvision/datasets/sun397.html
index bb88ca440e5..d30d8024744 100644
--- a/0.16/_modules/torchvision/datasets/sun397.html
+++ b/0.16/_modules/torchvision/datasets/sun397.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.sun397 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.sun397 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/svhn.html b/0.16/_modules/torchvision/datasets/svhn.html
index 6557942cb98..a76e45206a1 100644
--- a/0.16/_modules/torchvision/datasets/svhn.html
+++ b/0.16/_modules/torchvision/datasets/svhn.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.svhn &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.svhn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/ucf101.html b/0.16/_modules/torchvision/datasets/ucf101.html
index aefca6692a2..a8ebf4e1080 100644
--- a/0.16/_modules/torchvision/datasets/ucf101.html
+++ b/0.16/_modules/torchvision/datasets/ucf101.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.ucf101 &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.ucf101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/usps.html b/0.16/_modules/torchvision/datasets/usps.html
index 4f729294c69..24f7f1400fe 100644
--- a/0.16/_modules/torchvision/datasets/usps.html
+++ b/0.16/_modules/torchvision/datasets/usps.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.usps &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.usps &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/vision.html b/0.16/_modules/torchvision/datasets/vision.html
index 60beb10a328..aeab9b7a7fe 100644
--- a/0.16/_modules/torchvision/datasets/vision.html
+++ b/0.16/_modules/torchvision/datasets/vision.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.vision &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.vision &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/voc.html b/0.16/_modules/torchvision/datasets/voc.html
index 1f3efb77f98..440910e19c6 100644
--- a/0.16/_modules/torchvision/datasets/voc.html
+++ b/0.16/_modules/torchvision/datasets/voc.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.voc &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.voc &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/datasets/widerface.html b/0.16/_modules/torchvision/datasets/widerface.html
index 67d29ce8f97..74bd6f1e7e8 100644
--- a/0.16/_modules/torchvision/datasets/widerface.html
+++ b/0.16/_modules/torchvision/datasets/widerface.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.datasets.widerface &mdash; Torchvision main documentation</title>
+  <title>torchvision.datasets.widerface &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/io/image.html b/0.16/_modules/torchvision/io/image.html
index f209ef54b85..846a5c22d6f 100644
--- a/0.16/_modules/torchvision/io/image.html
+++ b/0.16/_modules/torchvision/io/image.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.io.image &mdash; Torchvision main documentation</title>
+  <title>torchvision.io.image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/io/video.html b/0.16/_modules/torchvision/io/video.html
index ca8c5f40c56..b7144d0a22d 100644
--- a/0.16/_modules/torchvision/io/video.html
+++ b/0.16/_modules/torchvision/io/video.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.io.video &mdash; Torchvision main documentation</title>
+  <title>torchvision.io.video &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/io/video_reader.html b/0.16/_modules/torchvision/io/video_reader.html
index 5ee4254f7d8..06f12157ece 100644
--- a/0.16/_modules/torchvision/io/video_reader.html
+++ b/0.16/_modules/torchvision/io/video_reader.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.io.video_reader &mdash; Torchvision main documentation</title>
+  <title>torchvision.io.video_reader &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/_api.html b/0.16/_modules/torchvision/models/_api.html
index dcc0ccf0f0c..427ca9cef3e 100644
--- a/0.16/_modules/torchvision/models/_api.html
+++ b/0.16/_modules/torchvision/models/_api.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models._api &mdash; Torchvision main documentation</title>
+  <title>torchvision.models._api &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/alexnet.html b/0.16/_modules/torchvision/models/alexnet.html
index 40ecd368134..28801ab45ca 100644
--- a/0.16/_modules/torchvision/models/alexnet.html
+++ b/0.16/_modules/torchvision/models/alexnet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.alexnet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.alexnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/convnext.html b/0.16/_modules/torchvision/models/convnext.html
index eb432c007bd..d868af7f129 100644
--- a/0.16/_modules/torchvision/models/convnext.html
+++ b/0.16/_modules/torchvision/models/convnext.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.convnext &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.convnext &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/densenet.html b/0.16/_modules/torchvision/models/densenet.html
index 5705c581fa7..557181a3485 100644
--- a/0.16/_modules/torchvision/models/densenet.html
+++ b/0.16/_modules/torchvision/models/densenet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.densenet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.densenet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/detection/faster_rcnn.html b/0.16/_modules/torchvision/models/detection/faster_rcnn.html
index 8d2e8acb1db..674c841c090 100644
--- a/0.16/_modules/torchvision/models/detection/faster_rcnn.html
+++ b/0.16/_modules/torchvision/models/detection/faster_rcnn.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.detection.faster_rcnn &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.detection.faster_rcnn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/detection/fcos.html b/0.16/_modules/torchvision/models/detection/fcos.html
index ca11dbff4dd..5e2d881560f 100644
--- a/0.16/_modules/torchvision/models/detection/fcos.html
+++ b/0.16/_modules/torchvision/models/detection/fcos.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.detection.fcos &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.detection.fcos &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html b/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
index d397c36fc83..a56e9116fb0 100644
--- a/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
+++ b/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.detection.keypoint_rcnn &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.detection.keypoint_rcnn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/detection/mask_rcnn.html b/0.16/_modules/torchvision/models/detection/mask_rcnn.html
index 143e1dff660..2f6b6bf2152 100644
--- a/0.16/_modules/torchvision/models/detection/mask_rcnn.html
+++ b/0.16/_modules/torchvision/models/detection/mask_rcnn.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.detection.mask_rcnn &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.detection.mask_rcnn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/detection/retinanet.html b/0.16/_modules/torchvision/models/detection/retinanet.html
index badf85db7eb..f111582260e 100644
--- a/0.16/_modules/torchvision/models/detection/retinanet.html
+++ b/0.16/_modules/torchvision/models/detection/retinanet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.detection.retinanet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.detection.retinanet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/detection/ssd.html b/0.16/_modules/torchvision/models/detection/ssd.html
index 6c3e468dbd9..46768273270 100644
--- a/0.16/_modules/torchvision/models/detection/ssd.html
+++ b/0.16/_modules/torchvision/models/detection/ssd.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.detection.ssd &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.detection.ssd &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/detection/ssdlite.html b/0.16/_modules/torchvision/models/detection/ssdlite.html
index bb6e3dfd8c3..c3f99923b36 100644
--- a/0.16/_modules/torchvision/models/detection/ssdlite.html
+++ b/0.16/_modules/torchvision/models/detection/ssdlite.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.detection.ssdlite &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.detection.ssdlite &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/efficientnet.html b/0.16/_modules/torchvision/models/efficientnet.html
index f025a3bfd6e..82ce6575f06 100644
--- a/0.16/_modules/torchvision/models/efficientnet.html
+++ b/0.16/_modules/torchvision/models/efficientnet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.efficientnet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.efficientnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/feature_extraction.html b/0.16/_modules/torchvision/models/feature_extraction.html
index bb776a4450d..3a91421c9a9 100644
--- a/0.16/_modules/torchvision/models/feature_extraction.html
+++ b/0.16/_modules/torchvision/models/feature_extraction.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.feature_extraction &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.feature_extraction &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/googlenet.html b/0.16/_modules/torchvision/models/googlenet.html
index b5e30188e41..434e91846f9 100644
--- a/0.16/_modules/torchvision/models/googlenet.html
+++ b/0.16/_modules/torchvision/models/googlenet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.googlenet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.googlenet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/inception.html b/0.16/_modules/torchvision/models/inception.html
index 6f1dbb5efc3..c09ebd4422d 100644
--- a/0.16/_modules/torchvision/models/inception.html
+++ b/0.16/_modules/torchvision/models/inception.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.inception &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.inception &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/maxvit.html b/0.16/_modules/torchvision/models/maxvit.html
index 4a44dfa8be6..06f20c4dd70 100644
--- a/0.16/_modules/torchvision/models/maxvit.html
+++ b/0.16/_modules/torchvision/models/maxvit.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.maxvit &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.maxvit &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/mnasnet.html b/0.16/_modules/torchvision/models/mnasnet.html
index b30ce88b745..80dafb35a9c 100644
--- a/0.16/_modules/torchvision/models/mnasnet.html
+++ b/0.16/_modules/torchvision/models/mnasnet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.mnasnet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.mnasnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/mobilenetv2.html b/0.16/_modules/torchvision/models/mobilenetv2.html
index 1842262536a..04f2a378b6e 100644
--- a/0.16/_modules/torchvision/models/mobilenetv2.html
+++ b/0.16/_modules/torchvision/models/mobilenetv2.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.mobilenetv2 &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.mobilenetv2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/mobilenetv3.html b/0.16/_modules/torchvision/models/mobilenetv3.html
index aef5fec9a5a..24f0339a21a 100644
--- a/0.16/_modules/torchvision/models/mobilenetv3.html
+++ b/0.16/_modules/torchvision/models/mobilenetv3.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.mobilenetv3 &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.mobilenetv3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/optical_flow/raft.html b/0.16/_modules/torchvision/models/optical_flow/raft.html
index 7ef1211b4b1..22aa287546e 100644
--- a/0.16/_modules/torchvision/models/optical_flow/raft.html
+++ b/0.16/_modules/torchvision/models/optical_flow/raft.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.optical_flow.raft &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.optical_flow.raft &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/quantization/googlenet.html b/0.16/_modules/torchvision/models/quantization/googlenet.html
index fb72b77ddb9..174885ba9ee 100644
--- a/0.16/_modules/torchvision/models/quantization/googlenet.html
+++ b/0.16/_modules/torchvision/models/quantization/googlenet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.quantization.googlenet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.quantization.googlenet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/quantization/inception.html b/0.16/_modules/torchvision/models/quantization/inception.html
index 95543e950e2..ae2cfdd8bcb 100644
--- a/0.16/_modules/torchvision/models/quantization/inception.html
+++ b/0.16/_modules/torchvision/models/quantization/inception.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.quantization.inception &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.quantization.inception &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/quantization/mobilenetv2.html b/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
index 255f4921153..d54f6bb32e4 100644
--- a/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
+++ b/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.quantization.mobilenetv2 &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.quantization.mobilenetv2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/quantization/mobilenetv3.html b/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
index 60c9ba8f8b3..c987ec1a3b4 100644
--- a/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
+++ b/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.quantization.mobilenetv3 &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.quantization.mobilenetv3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/quantization/resnet.html b/0.16/_modules/torchvision/models/quantization/resnet.html
index 846b097e67a..78b97ea5143 100644
--- a/0.16/_modules/torchvision/models/quantization/resnet.html
+++ b/0.16/_modules/torchvision/models/quantization/resnet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.quantization.resnet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.quantization.resnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/quantization/shufflenetv2.html b/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
index 04db9ba72be..f6e6f7de5ac 100644
--- a/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
+++ b/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.quantization.shufflenetv2 &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.quantization.shufflenetv2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/regnet.html b/0.16/_modules/torchvision/models/regnet.html
index e089328b13e..6e873d47cba 100644
--- a/0.16/_modules/torchvision/models/regnet.html
+++ b/0.16/_modules/torchvision/models/regnet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.regnet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.regnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/resnet.html b/0.16/_modules/torchvision/models/resnet.html
index 469bca1f32e..2fddfff96c8 100644
--- a/0.16/_modules/torchvision/models/resnet.html
+++ b/0.16/_modules/torchvision/models/resnet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.resnet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.resnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/segmentation/deeplabv3.html b/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
index 904ec4ec9c6..09be854330d 100644
--- a/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
+++ b/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.segmentation.deeplabv3 &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.segmentation.deeplabv3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/segmentation/fcn.html b/0.16/_modules/torchvision/models/segmentation/fcn.html
index e4a133f020b..f91d1efc553 100644
--- a/0.16/_modules/torchvision/models/segmentation/fcn.html
+++ b/0.16/_modules/torchvision/models/segmentation/fcn.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.segmentation.fcn &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.segmentation.fcn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/segmentation/lraspp.html b/0.16/_modules/torchvision/models/segmentation/lraspp.html
index 2f269ec0128..095d30ec7f2 100644
--- a/0.16/_modules/torchvision/models/segmentation/lraspp.html
+++ b/0.16/_modules/torchvision/models/segmentation/lraspp.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.segmentation.lraspp &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.segmentation.lraspp &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/shufflenetv2.html b/0.16/_modules/torchvision/models/shufflenetv2.html
index e58affd8fbd..7c0dadca7c8 100644
--- a/0.16/_modules/torchvision/models/shufflenetv2.html
+++ b/0.16/_modules/torchvision/models/shufflenetv2.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.shufflenetv2 &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.shufflenetv2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/squeezenet.html b/0.16/_modules/torchvision/models/squeezenet.html
index 2fc54675f2e..9ca784dd038 100644
--- a/0.16/_modules/torchvision/models/squeezenet.html
+++ b/0.16/_modules/torchvision/models/squeezenet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.squeezenet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.squeezenet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/swin_transformer.html b/0.16/_modules/torchvision/models/swin_transformer.html
index 57f2f66d25f..a47b4ac9896 100644
--- a/0.16/_modules/torchvision/models/swin_transformer.html
+++ b/0.16/_modules/torchvision/models/swin_transformer.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.swin_transformer &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.swin_transformer &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/vgg.html b/0.16/_modules/torchvision/models/vgg.html
index 1351eac9c43..56af2b05b26 100644
--- a/0.16/_modules/torchvision/models/vgg.html
+++ b/0.16/_modules/torchvision/models/vgg.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.vgg &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.vgg &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/video/mvit.html b/0.16/_modules/torchvision/models/video/mvit.html
index d7fbf90b374..0a0ad19e3b1 100644
--- a/0.16/_modules/torchvision/models/video/mvit.html
+++ b/0.16/_modules/torchvision/models/video/mvit.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.video.mvit &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.video.mvit &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/video/resnet.html b/0.16/_modules/torchvision/models/video/resnet.html
index 92ee1189093..877eec9fa4c 100644
--- a/0.16/_modules/torchvision/models/video/resnet.html
+++ b/0.16/_modules/torchvision/models/video/resnet.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.video.resnet &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.video.resnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/video/s3d.html b/0.16/_modules/torchvision/models/video/s3d.html
index 8d8d9d16cda..d4dd2cf2290 100644
--- a/0.16/_modules/torchvision/models/video/s3d.html
+++ b/0.16/_modules/torchvision/models/video/s3d.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.video.s3d &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.video.s3d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/video/swin_transformer.html b/0.16/_modules/torchvision/models/video/swin_transformer.html
index a0faaf227ec..b8150796187 100644
--- a/0.16/_modules/torchvision/models/video/swin_transformer.html
+++ b/0.16/_modules/torchvision/models/video/swin_transformer.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.video.swin_transformer &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.video.swin_transformer &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/models/vision_transformer.html b/0.16/_modules/torchvision/models/vision_transformer.html
index fc4c1d51823..9ac371a66f7 100644
--- a/0.16/_modules/torchvision/models/vision_transformer.html
+++ b/0.16/_modules/torchvision/models/vision_transformer.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.models.vision_transformer &mdash; Torchvision main documentation</title>
+  <title>torchvision.models.vision_transformer &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/boxes.html b/0.16/_modules/torchvision/ops/boxes.html
index 92d6567e277..03995e6ff55 100644
--- a/0.16/_modules/torchvision/ops/boxes.html
+++ b/0.16/_modules/torchvision/ops/boxes.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.boxes &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.boxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/ciou_loss.html b/0.16/_modules/torchvision/ops/ciou_loss.html
index f9f1ecb1145..954c0f54059 100644
--- a/0.16/_modules/torchvision/ops/ciou_loss.html
+++ b/0.16/_modules/torchvision/ops/ciou_loss.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.ciou_loss &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.ciou_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/deform_conv.html b/0.16/_modules/torchvision/ops/deform_conv.html
index c2949659bc9..a1a66b1ddcf 100644
--- a/0.16/_modules/torchvision/ops/deform_conv.html
+++ b/0.16/_modules/torchvision/ops/deform_conv.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.deform_conv &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.deform_conv &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/diou_loss.html b/0.16/_modules/torchvision/ops/diou_loss.html
index b970a9c1b74..74edead1a95 100644
--- a/0.16/_modules/torchvision/ops/diou_loss.html
+++ b/0.16/_modules/torchvision/ops/diou_loss.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.diou_loss &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.diou_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/drop_block.html b/0.16/_modules/torchvision/ops/drop_block.html
index 778f4819ac8..14c159919cc 100644
--- a/0.16/_modules/torchvision/ops/drop_block.html
+++ b/0.16/_modules/torchvision/ops/drop_block.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.drop_block &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.drop_block &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/feature_pyramid_network.html b/0.16/_modules/torchvision/ops/feature_pyramid_network.html
index 6e55bab930a..6f6be47dad4 100644
--- a/0.16/_modules/torchvision/ops/feature_pyramid_network.html
+++ b/0.16/_modules/torchvision/ops/feature_pyramid_network.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.feature_pyramid_network &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.feature_pyramid_network &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/focal_loss.html b/0.16/_modules/torchvision/ops/focal_loss.html
index b21aedf25e1..a596b740e0a 100644
--- a/0.16/_modules/torchvision/ops/focal_loss.html
+++ b/0.16/_modules/torchvision/ops/focal_loss.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.focal_loss &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.focal_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/giou_loss.html b/0.16/_modules/torchvision/ops/giou_loss.html
index 90dd1dfcb27..3dbc8ea1e22 100644
--- a/0.16/_modules/torchvision/ops/giou_loss.html
+++ b/0.16/_modules/torchvision/ops/giou_loss.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.giou_loss &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.giou_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/misc.html b/0.16/_modules/torchvision/ops/misc.html
index 3370851f775..673a8106fa5 100644
--- a/0.16/_modules/torchvision/ops/misc.html
+++ b/0.16/_modules/torchvision/ops/misc.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.misc &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.misc &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/poolers.html b/0.16/_modules/torchvision/ops/poolers.html
index fa08c7510bf..760114e5862 100644
--- a/0.16/_modules/torchvision/ops/poolers.html
+++ b/0.16/_modules/torchvision/ops/poolers.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.poolers &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.poolers &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/ps_roi_align.html b/0.16/_modules/torchvision/ops/ps_roi_align.html
index ebd85e4513e..30704d941ce 100644
--- a/0.16/_modules/torchvision/ops/ps_roi_align.html
+++ b/0.16/_modules/torchvision/ops/ps_roi_align.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.ps_roi_align &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.ps_roi_align &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/ps_roi_pool.html b/0.16/_modules/torchvision/ops/ps_roi_pool.html
index cba744ee505..f19d2461023 100644
--- a/0.16/_modules/torchvision/ops/ps_roi_pool.html
+++ b/0.16/_modules/torchvision/ops/ps_roi_pool.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.ps_roi_pool &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.ps_roi_pool &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/roi_align.html b/0.16/_modules/torchvision/ops/roi_align.html
index 900ed45524f..da03c86541c 100644
--- a/0.16/_modules/torchvision/ops/roi_align.html
+++ b/0.16/_modules/torchvision/ops/roi_align.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.roi_align &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.roi_align &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/roi_pool.html b/0.16/_modules/torchvision/ops/roi_pool.html
index 23862fc70b8..985389adfe4 100644
--- a/0.16/_modules/torchvision/ops/roi_pool.html
+++ b/0.16/_modules/torchvision/ops/roi_pool.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.roi_pool &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.roi_pool &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/ops/stochastic_depth.html b/0.16/_modules/torchvision/ops/stochastic_depth.html
index 65745e2ecaf..c11421050b7 100644
--- a/0.16/_modules/torchvision/ops/stochastic_depth.html
+++ b/0.16/_modules/torchvision/ops/stochastic_depth.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.ops.stochastic_depth &mdash; Torchvision main documentation</title>
+  <title>torchvision.ops.stochastic_depth &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/autoaugment.html b/0.16/_modules/torchvision/transforms/autoaugment.html
index 3fceb59a8ba..3289088e495 100644
--- a/0.16/_modules/torchvision/transforms/autoaugment.html
+++ b/0.16/_modules/torchvision/transforms/autoaugment.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.autoaugment &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.autoaugment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/functional.html b/0.16/_modules/torchvision/transforms/functional.html
index 1ff0cf9e05e..8f47497ed53 100644
--- a/0.16/_modules/torchvision/transforms/functional.html
+++ b/0.16/_modules/torchvision/transforms/functional.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.functional &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.functional &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/transforms.html b/0.16/_modules/torchvision/transforms/transforms.html
index f4d87015c33..78dd6d39024 100644
--- a/0.16/_modules/torchvision/transforms/transforms.html
+++ b/0.16/_modules/torchvision/transforms/transforms.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.transforms &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.transforms &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_augment.html b/0.16/_modules/torchvision/transforms/v2/_augment.html
index 3a47e87f898..da7bc499536 100644
--- a/0.16/_modules/torchvision/transforms/v2/_augment.html
+++ b/0.16/_modules/torchvision/transforms/v2/_augment.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._augment &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._augment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_auto_augment.html b/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
index 338ebfdc46d..5313f54fb57 100644
--- a/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
+++ b/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._auto_augment &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._auto_augment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_color.html b/0.16/_modules/torchvision/transforms/v2/_color.html
index c4da81c3969..83ab292fc1b 100644
--- a/0.16/_modules/torchvision/transforms/v2/_color.html
+++ b/0.16/_modules/torchvision/transforms/v2/_color.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._color &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._color &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_container.html b/0.16/_modules/torchvision/transforms/v2/_container.html
index 06264bcad9e..adc05acca65 100644
--- a/0.16/_modules/torchvision/transforms/v2/_container.html
+++ b/0.16/_modules/torchvision/transforms/v2/_container.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._container &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._container &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_deprecated.html b/0.16/_modules/torchvision/transforms/v2/_deprecated.html
index 5f35389ce4f..00dd42168ba 100644
--- a/0.16/_modules/torchvision/transforms/v2/_deprecated.html
+++ b/0.16/_modules/torchvision/transforms/v2/_deprecated.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._deprecated &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._deprecated &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_geometry.html b/0.16/_modules/torchvision/transforms/v2/_geometry.html
index f93af774b48..4cf36db1748 100644
--- a/0.16/_modules/torchvision/transforms/v2/_geometry.html
+++ b/0.16/_modules/torchvision/transforms/v2/_geometry.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._geometry &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._geometry &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_meta.html b/0.16/_modules/torchvision/transforms/v2/_meta.html
index 59bdfa6bca2..08a4da98e33 100644
--- a/0.16/_modules/torchvision/transforms/v2/_meta.html
+++ b/0.16/_modules/torchvision/transforms/v2/_meta.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._meta &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._meta &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_misc.html b/0.16/_modules/torchvision/transforms/v2/_misc.html
index 2f8067d84c9..7ad1c9cdf6f 100644
--- a/0.16/_modules/torchvision/transforms/v2/_misc.html
+++ b/0.16/_modules/torchvision/transforms/v2/_misc.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._misc &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._misc &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_temporal.html b/0.16/_modules/torchvision/transforms/v2/_temporal.html
index 4e22b8a2b2e..1305955d813 100644
--- a/0.16/_modules/torchvision/transforms/v2/_temporal.html
+++ b/0.16/_modules/torchvision/transforms/v2/_temporal.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._temporal &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._temporal &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/_type_conversion.html b/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
index 0b5af12a030..5a12f1dc053 100644
--- a/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
+++ b/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2._type_conversion &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2._type_conversion &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_augment.html b/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
index 8ded01f03b9..7ff068bea14 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._augment &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._augment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_color.html b/0.16/_modules/torchvision/transforms/v2/functional/_color.html
index ec8f12ab3b7..7f6f3daa166 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_color.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_color.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._color &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._color &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html b/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
index 29881743f27..97d49f332c5 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._deprecated &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._deprecated &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html b/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
index 26d15534520..1f8315ce3e3 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._geometry &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._geometry &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_meta.html b/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
index 1dc9c533959..4f46ab01fb1 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._meta &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._meta &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_misc.html b/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
index d93715faee1..72a6c571c39 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._misc &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._misc &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html b/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
index 618f64b4bbe..11fb9afb9e9 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._temporal &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._temporal &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html b/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
index 818c8305b11..610f62a7c45 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._type_conversion &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._type_conversion &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_utils.html b/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
index e4eec91ffbf..880a2b700dd 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional._utils &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional._utils &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors.html b/0.16/_modules/torchvision/tv_tensors.html
index e5ee49dcdcc..83caa48bfd1 100644
--- a/0.16/_modules/torchvision/tv_tensors.html
+++ b/0.16/_modules/torchvision/tv_tensors.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html b/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
index 58f61e69b5e..ffde8aa72d7 100644
--- a/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
+++ b/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors._bounding_boxes &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors._bounding_boxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html b/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
index 995ef0e5832..638d4450651 100644
--- a/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
+++ b/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors._dataset_wrapper &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors._dataset_wrapper &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors/_image.html b/0.16/_modules/torchvision/tv_tensors/_image.html
index 23238465a31..f393eceab21 100644
--- a/0.16/_modules/torchvision/tv_tensors/_image.html
+++ b/0.16/_modules/torchvision/tv_tensors/_image.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors._image &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors._image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors/_mask.html b/0.16/_modules/torchvision/tv_tensors/_mask.html
index 40368e3532f..2b1a1ceea70 100644
--- a/0.16/_modules/torchvision/tv_tensors/_mask.html
+++ b/0.16/_modules/torchvision/tv_tensors/_mask.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors._mask &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors._mask &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html b/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
index 37c46dcf7cd..00faede710b 100644
--- a/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
+++ b/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors._torch_function_helpers &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors._torch_function_helpers &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html b/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
index f721b955a42..bf6cd60bca6 100644
--- a/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
+++ b/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors._tv_tensor &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors._tv_tensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/tv_tensors/_video.html b/0.16/_modules/torchvision/tv_tensors/_video.html
index 251273c2af4..ab603b4e50a 100644
--- a/0.16/_modules/torchvision/tv_tensors/_video.html
+++ b/0.16/_modules/torchvision/tv_tensors/_video.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.tv_tensors._video &mdash; Torchvision main documentation</title>
+  <title>torchvision.tv_tensors._video &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/_modules/torchvision/utils.html b/0.16/_modules/torchvision/utils.html
index 20e21b95987..1ce0cb9be10 100644
--- a/0.16/_modules/torchvision/utils.html
+++ b/0.16/_modules/torchvision/utils.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.utils &mdash; Torchvision main documentation</title>
+  <title>torchvision.utils &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/index.html b/0.16/auto_examples/index.html
index 0b44a607aa9..647e2944c91 100644
--- a/0.16/auto_examples/index.html
+++ b/0.16/auto_examples/index.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Examples and tutorials &mdash; Torchvision main documentation</title>
+  <title>Examples and tutorials &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/others/index.html b/0.16/auto_examples/others/index.html
index f62686b0620..b58a5431270 100644
--- a/0.16/auto_examples/others/index.html
+++ b/0.16/auto_examples/others/index.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Others &mdash; Torchvision main documentation</title>
+  <title>Others &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/others/plot_optical_flow.html b/0.16/auto_examples/others/plot_optical_flow.html
index 202bd31696c..7e5409be75b 100644
--- a/0.16/auto_examples/others/plot_optical_flow.html
+++ b/0.16/auto_examples/others/plot_optical_flow.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Optical Flow: Predicting movement with the RAFT model &mdash; Torchvision main documentation</title>
+  <title>Optical Flow: Predicting movement with the RAFT model &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/others/plot_repurposing_annotations.html b/0.16/auto_examples/others/plot_repurposing_annotations.html
index be6e095fe8d..3e99c57bb85 100644
--- a/0.16/auto_examples/others/plot_repurposing_annotations.html
+++ b/0.16/auto_examples/others/plot_repurposing_annotations.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Repurposing masks into bounding boxes &mdash; Torchvision main documentation</title>
+  <title>Repurposing masks into bounding boxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/others/plot_scripted_tensor_transforms.html b/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
index fe3a5825cfc..50757817e61 100644
--- a/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
+++ b/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Torchscript support &mdash; Torchvision main documentation</title>
+  <title>Torchscript support &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/others/plot_video_api.html b/0.16/auto_examples/others/plot_video_api.html
index 72e59903f59..840a8323dde 100644
--- a/0.16/auto_examples/others/plot_video_api.html
+++ b/0.16/auto_examples/others/plot_video_api.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Video API &mdash; Torchvision main documentation</title>
+  <title>Video API &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/others/plot_visualization_utils.html b/0.16/auto_examples/others/plot_visualization_utils.html
index e2508627d97..f768af7c544 100644
--- a/0.16/auto_examples/others/plot_visualization_utils.html
+++ b/0.16/auto_examples/others/plot_visualization_utils.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Visualization utilities &mdash; Torchvision main documentation</title>
+  <title>Visualization utilities &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/others/sg_execution_times.html b/0.16/auto_examples/others/sg_execution_times.html
index e8ca1632633..245e01a0cfe 100644
--- a/0.16/auto_examples/others/sg_execution_times.html
+++ b/0.16/auto_examples/others/sg_execution_times.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Computation times &mdash; Torchvision main documentation</title>
+  <title>Computation times &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/index.html b/0.16/auto_examples/transforms/index.html
index b664ea1e085..b74d03d2a5a 100644
--- a/0.16/auto_examples/transforms/index.html
+++ b/0.16/auto_examples/transforms/index.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Transforms &mdash; Torchvision main documentation</title>
+  <title>Transforms &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/plot_custom_transforms.html b/0.16/auto_examples/transforms/plot_custom_transforms.html
index 4e0b9b6287d..556338a28a9 100644
--- a/0.16/auto_examples/transforms/plot_custom_transforms.html
+++ b/0.16/auto_examples/transforms/plot_custom_transforms.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to write your own v2 transforms &mdash; Torchvision main documentation</title>
+  <title>How to write your own v2 transforms &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/plot_custom_tv_tensors.html b/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
index 22cc922c0e2..f6273a3e416 100644
--- a/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
+++ b/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to write your own TVTensor class &mdash; Torchvision main documentation</title>
+  <title>How to write your own TVTensor class &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/plot_cutmix_mixup.html b/0.16/auto_examples/transforms/plot_cutmix_mixup.html
index c9df80ab8db..acc6c61cc22 100644
--- a/0.16/auto_examples/transforms/plot_cutmix_mixup.html
+++ b/0.16/auto_examples/transforms/plot_cutmix_mixup.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>How to use CutMix and MixUp &mdash; Torchvision main documentation</title>
+  <title>How to use CutMix and MixUp &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/plot_transforms_e2e.html b/0.16/auto_examples/transforms/plot_transforms_e2e.html
index 1e839ea3279..4448b5ae8ca 100644
--- a/0.16/auto_examples/transforms/plot_transforms_e2e.html
+++ b/0.16/auto_examples/transforms/plot_transforms_e2e.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Transforms v2: End-to-end object detection/segmentation example &mdash; Torchvision main documentation</title>
+  <title>Transforms v2: End-to-end object detection/segmentation example &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/plot_transforms_getting_started.html b/0.16/auto_examples/transforms/plot_transforms_getting_started.html
index e365e69810f..17bab393c67 100644
--- a/0.16/auto_examples/transforms/plot_transforms_getting_started.html
+++ b/0.16/auto_examples/transforms/plot_transforms_getting_started.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Getting started with transforms v2 &mdash; Torchvision main documentation</title>
+  <title>Getting started with transforms v2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/plot_transforms_illustrations.html b/0.16/auto_examples/transforms/plot_transforms_illustrations.html
index 6a48616c0f5..45e02e2eb4d 100644
--- a/0.16/auto_examples/transforms/plot_transforms_illustrations.html
+++ b/0.16/auto_examples/transforms/plot_transforms_illustrations.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Illustration of transforms &mdash; Torchvision main documentation</title>
+  <title>Illustration of transforms &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/plot_tv_tensors.html b/0.16/auto_examples/transforms/plot_tv_tensors.html
index a7505414c78..f6c3242d4d3 100644
--- a/0.16/auto_examples/transforms/plot_tv_tensors.html
+++ b/0.16/auto_examples/transforms/plot_tv_tensors.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TVTensors FAQ &mdash; Torchvision main documentation</title>
+  <title>TVTensors FAQ &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/auto_examples/transforms/sg_execution_times.html b/0.16/auto_examples/transforms/sg_execution_times.html
index 214f282b485..ed6ce19bfa3 100644
--- a/0.16/auto_examples/transforms/sg_execution_times.html
+++ b/0.16/auto_examples/transforms/sg_execution_times.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Computation times &mdash; Torchvision main documentation</title>
+  <title>Computation times &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/datasets.html b/0.16/datasets.html
index 32f76faaeb9..0c96b30f3df 100644
--- a/0.16/datasets.html
+++ b/0.16/datasets.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Datasets &mdash; Torchvision main documentation</title>
+  <title>Datasets &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/feature_extraction.html b/0.16/feature_extraction.html
index be882368842..cede89aba1b 100644
--- a/0.16/feature_extraction.html
+++ b/0.16/feature_extraction.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Feature extraction for model inspection &mdash; Torchvision main documentation</title>
+  <title>Feature extraction for model inspection &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/classification_quant_table.html b/0.16/generated/classification_quant_table.html
index e7ef8185ceb..63d73f48fa4 100644
--- a/0.16/generated/classification_quant_table.html
+++ b/0.16/generated/classification_quant_table.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  <title>&lt;no title&gt; &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/classification_table.html b/0.16/generated/classification_table.html
index b37d604106d..c87bad351a4 100644
--- a/0.16/generated/classification_table.html
+++ b/0.16/generated/classification_table.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  <title>&lt;no title&gt; &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/detection_keypoint_table.html b/0.16/generated/detection_keypoint_table.html
index 2e4bd1b4bac..ad1b56d6d53 100644
--- a/0.16/generated/detection_keypoint_table.html
+++ b/0.16/generated/detection_keypoint_table.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  <title>&lt;no title&gt; &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/detection_table.html b/0.16/generated/detection_table.html
index a02f48d55f6..0c81a16c731 100644
--- a/0.16/generated/detection_table.html
+++ b/0.16/generated/detection_table.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  <title>&lt;no title&gt; &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/instance_segmentation_table.html b/0.16/generated/instance_segmentation_table.html
index dcba209ec91..efad0be4c23 100644
--- a/0.16/generated/instance_segmentation_table.html
+++ b/0.16/generated/instance_segmentation_table.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  <title>&lt;no title&gt; &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/segmentation_table.html b/0.16/generated/segmentation_table.html
index 3fc2327d25d..0536fb49873 100644
--- a/0.16/generated/segmentation_table.html
+++ b/0.16/generated/segmentation_table.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  <title>&lt;no title&gt; &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.CIFAR10.html b/0.16/generated/torchvision.datasets.CIFAR10.html
index 6ac4a17e911..722e960e213 100644
--- a/0.16/generated/torchvision.datasets.CIFAR10.html
+++ b/0.16/generated/torchvision.datasets.CIFAR10.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CIFAR10 &mdash; Torchvision main documentation</title>
+  <title>CIFAR10 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.CIFAR100.html b/0.16/generated/torchvision.datasets.CIFAR100.html
index 73b3f353699..b6940f7c6ef 100644
--- a/0.16/generated/torchvision.datasets.CIFAR100.html
+++ b/0.16/generated/torchvision.datasets.CIFAR100.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CIFAR100 &mdash; Torchvision main documentation</title>
+  <title>CIFAR100 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.CREStereo.html b/0.16/generated/torchvision.datasets.CREStereo.html
index fa97ff81b12..c32cc8d50fd 100644
--- a/0.16/generated/torchvision.datasets.CREStereo.html
+++ b/0.16/generated/torchvision.datasets.CREStereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CREStereo &mdash; Torchvision main documentation</title>
+  <title>CREStereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Caltech101.html b/0.16/generated/torchvision.datasets.Caltech101.html
index b3d5b5fcdf6..adf46abfda8 100644
--- a/0.16/generated/torchvision.datasets.Caltech101.html
+++ b/0.16/generated/torchvision.datasets.Caltech101.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Caltech101 &mdash; Torchvision main documentation</title>
+  <title>Caltech101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Caltech256.html b/0.16/generated/torchvision.datasets.Caltech256.html
index 8169c77606d..d449ce98cdc 100644
--- a/0.16/generated/torchvision.datasets.Caltech256.html
+++ b/0.16/generated/torchvision.datasets.Caltech256.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Caltech256 &mdash; Torchvision main documentation</title>
+  <title>Caltech256 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.CarlaStereo.html b/0.16/generated/torchvision.datasets.CarlaStereo.html
index b3b01e604c9..77e009f21df 100644
--- a/0.16/generated/torchvision.datasets.CarlaStereo.html
+++ b/0.16/generated/torchvision.datasets.CarlaStereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CarlaStereo &mdash; Torchvision main documentation</title>
+  <title>CarlaStereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.CelebA.html b/0.16/generated/torchvision.datasets.CelebA.html
index 60bf1f1492b..e144fd1048d 100644
--- a/0.16/generated/torchvision.datasets.CelebA.html
+++ b/0.16/generated/torchvision.datasets.CelebA.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CelebA &mdash; Torchvision main documentation</title>
+  <title>CelebA &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Cityscapes.html b/0.16/generated/torchvision.datasets.Cityscapes.html
index 939bdb2a1b4..cf1dffef071 100644
--- a/0.16/generated/torchvision.datasets.Cityscapes.html
+++ b/0.16/generated/torchvision.datasets.Cityscapes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Cityscapes &mdash; Torchvision main documentation</title>
+  <title>Cityscapes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.CocoCaptions.html b/0.16/generated/torchvision.datasets.CocoCaptions.html
index 4de2ad4753a..b8fccf2b472 100644
--- a/0.16/generated/torchvision.datasets.CocoCaptions.html
+++ b/0.16/generated/torchvision.datasets.CocoCaptions.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CocoCaptions &mdash; Torchvision main documentation</title>
+  <title>CocoCaptions &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.CocoDetection.html b/0.16/generated/torchvision.datasets.CocoDetection.html
index a256636df2c..c93eee57ea0 100644
--- a/0.16/generated/torchvision.datasets.CocoDetection.html
+++ b/0.16/generated/torchvision.datasets.CocoDetection.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CocoDetection &mdash; Torchvision main documentation</title>
+  <title>CocoDetection &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Country211.html b/0.16/generated/torchvision.datasets.Country211.html
index ac47a052194..44555bffffa 100644
--- a/0.16/generated/torchvision.datasets.Country211.html
+++ b/0.16/generated/torchvision.datasets.Country211.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Country211 &mdash; Torchvision main documentation</title>
+  <title>Country211 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.DTD.html b/0.16/generated/torchvision.datasets.DTD.html
index 4a912a1011f..c98c24bddf0 100644
--- a/0.16/generated/torchvision.datasets.DTD.html
+++ b/0.16/generated/torchvision.datasets.DTD.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>DTD &mdash; Torchvision main documentation</title>
+  <title>DTD &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.DatasetFolder.html b/0.16/generated/torchvision.datasets.DatasetFolder.html
index 335e9935651..42578d2421b 100644
--- a/0.16/generated/torchvision.datasets.DatasetFolder.html
+++ b/0.16/generated/torchvision.datasets.DatasetFolder.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>DatasetFolder &mdash; Torchvision main documentation</title>
+  <title>DatasetFolder &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.EMNIST.html b/0.16/generated/torchvision.datasets.EMNIST.html
index f85c2edd715..5149608135c 100644
--- a/0.16/generated/torchvision.datasets.EMNIST.html
+++ b/0.16/generated/torchvision.datasets.EMNIST.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>EMNIST &mdash; Torchvision main documentation</title>
+  <title>EMNIST &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.ETH3DStereo.html b/0.16/generated/torchvision.datasets.ETH3DStereo.html
index 43fe81c0f78..c19c496de1c 100644
--- a/0.16/generated/torchvision.datasets.ETH3DStereo.html
+++ b/0.16/generated/torchvision.datasets.ETH3DStereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ETH3DStereo &mdash; Torchvision main documentation</title>
+  <title>ETH3DStereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.EuroSAT.html b/0.16/generated/torchvision.datasets.EuroSAT.html
index 37e8c632a52..97cdfc674c5 100644
--- a/0.16/generated/torchvision.datasets.EuroSAT.html
+++ b/0.16/generated/torchvision.datasets.EuroSAT.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>EuroSAT &mdash; Torchvision main documentation</title>
+  <title>EuroSAT &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.FER2013.html b/0.16/generated/torchvision.datasets.FER2013.html
index 9d2f7ae8b46..bdf645bee7f 100644
--- a/0.16/generated/torchvision.datasets.FER2013.html
+++ b/0.16/generated/torchvision.datasets.FER2013.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FER2013 &mdash; Torchvision main documentation</title>
+  <title>FER2013 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.FGVCAircraft.html b/0.16/generated/torchvision.datasets.FGVCAircraft.html
index df685fe873a..f5477c6d3b3 100644
--- a/0.16/generated/torchvision.datasets.FGVCAircraft.html
+++ b/0.16/generated/torchvision.datasets.FGVCAircraft.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FGVCAircraft &mdash; Torchvision main documentation</title>
+  <title>FGVCAircraft &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.FakeData.html b/0.16/generated/torchvision.datasets.FakeData.html
index 878828b8cf7..d27c24ec903 100644
--- a/0.16/generated/torchvision.datasets.FakeData.html
+++ b/0.16/generated/torchvision.datasets.FakeData.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FakeData &mdash; Torchvision main documentation</title>
+  <title>FakeData &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.FallingThingsStereo.html b/0.16/generated/torchvision.datasets.FallingThingsStereo.html
index 8f4278df4db..b440e4acc65 100644
--- a/0.16/generated/torchvision.datasets.FallingThingsStereo.html
+++ b/0.16/generated/torchvision.datasets.FallingThingsStereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FallingThingsStereo &mdash; Torchvision main documentation</title>
+  <title>FallingThingsStereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.FashionMNIST.html b/0.16/generated/torchvision.datasets.FashionMNIST.html
index b52756054bd..df34d6d39ca 100644
--- a/0.16/generated/torchvision.datasets.FashionMNIST.html
+++ b/0.16/generated/torchvision.datasets.FashionMNIST.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FashionMNIST &mdash; Torchvision main documentation</title>
+  <title>FashionMNIST &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Flickr30k.html b/0.16/generated/torchvision.datasets.Flickr30k.html
index 75f02fdc47d..2f61be961d9 100644
--- a/0.16/generated/torchvision.datasets.Flickr30k.html
+++ b/0.16/generated/torchvision.datasets.Flickr30k.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Flickr30k &mdash; Torchvision main documentation</title>
+  <title>Flickr30k &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Flickr8k.html b/0.16/generated/torchvision.datasets.Flickr8k.html
index faa3085a6d0..79b658ae5a5 100644
--- a/0.16/generated/torchvision.datasets.Flickr8k.html
+++ b/0.16/generated/torchvision.datasets.Flickr8k.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Flickr8k &mdash; Torchvision main documentation</title>
+  <title>Flickr8k &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Flowers102.html b/0.16/generated/torchvision.datasets.Flowers102.html
index c4848dbc28d..55964a86b49 100644
--- a/0.16/generated/torchvision.datasets.Flowers102.html
+++ b/0.16/generated/torchvision.datasets.Flowers102.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Flowers102 &mdash; Torchvision main documentation</title>
+  <title>Flowers102 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.FlyingChairs.html b/0.16/generated/torchvision.datasets.FlyingChairs.html
index 234a1bdd1fc..05fee62aa32 100644
--- a/0.16/generated/torchvision.datasets.FlyingChairs.html
+++ b/0.16/generated/torchvision.datasets.FlyingChairs.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FlyingChairs &mdash; Torchvision main documentation</title>
+  <title>FlyingChairs &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.FlyingThings3D.html b/0.16/generated/torchvision.datasets.FlyingThings3D.html
index 84a9b6e20a7..2feb977e669 100644
--- a/0.16/generated/torchvision.datasets.FlyingThings3D.html
+++ b/0.16/generated/torchvision.datasets.FlyingThings3D.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FlyingThings3D &mdash; Torchvision main documentation</title>
+  <title>FlyingThings3D &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Food101.html b/0.16/generated/torchvision.datasets.Food101.html
index d76519566cd..ceb17761652 100644
--- a/0.16/generated/torchvision.datasets.Food101.html
+++ b/0.16/generated/torchvision.datasets.Food101.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Food101 &mdash; Torchvision main documentation</title>
+  <title>Food101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.GTSRB.html b/0.16/generated/torchvision.datasets.GTSRB.html
index e0949ced80b..3d40152ef1d 100644
--- a/0.16/generated/torchvision.datasets.GTSRB.html
+++ b/0.16/generated/torchvision.datasets.GTSRB.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>GTSRB &mdash; Torchvision main documentation</title>
+  <title>GTSRB &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.HD1K.html b/0.16/generated/torchvision.datasets.HD1K.html
index f90a0317b64..ef9c28850f5 100644
--- a/0.16/generated/torchvision.datasets.HD1K.html
+++ b/0.16/generated/torchvision.datasets.HD1K.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>HD1K &mdash; Torchvision main documentation</title>
+  <title>HD1K &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.HMDB51.html b/0.16/generated/torchvision.datasets.HMDB51.html
index bf78d8c65ff..d3f06d442aa 100644
--- a/0.16/generated/torchvision.datasets.HMDB51.html
+++ b/0.16/generated/torchvision.datasets.HMDB51.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>HMDB51 &mdash; Torchvision main documentation</title>
+  <title>HMDB51 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.INaturalist.html b/0.16/generated/torchvision.datasets.INaturalist.html
index c00c70a592d..bbe4462f98f 100644
--- a/0.16/generated/torchvision.datasets.INaturalist.html
+++ b/0.16/generated/torchvision.datasets.INaturalist.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>INaturalist &mdash; Torchvision main documentation</title>
+  <title>INaturalist &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.ImageFolder.html b/0.16/generated/torchvision.datasets.ImageFolder.html
index 020b0a1366a..b3d6fd97393 100644
--- a/0.16/generated/torchvision.datasets.ImageFolder.html
+++ b/0.16/generated/torchvision.datasets.ImageFolder.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ImageFolder &mdash; Torchvision main documentation</title>
+  <title>ImageFolder &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.ImageNet.html b/0.16/generated/torchvision.datasets.ImageNet.html
index ea02e4954b5..dfbdc687846 100644
--- a/0.16/generated/torchvision.datasets.ImageNet.html
+++ b/0.16/generated/torchvision.datasets.ImageNet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ImageNet &mdash; Torchvision main documentation</title>
+  <title>ImageNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.InStereo2k.html b/0.16/generated/torchvision.datasets.InStereo2k.html
index cc55fb535f3..d4f5fa7c5b1 100644
--- a/0.16/generated/torchvision.datasets.InStereo2k.html
+++ b/0.16/generated/torchvision.datasets.InStereo2k.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>InStereo2k &mdash; Torchvision main documentation</title>
+  <title>InStereo2k &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.KMNIST.html b/0.16/generated/torchvision.datasets.KMNIST.html
index 824b1e71a55..7aafe8876c5 100644
--- a/0.16/generated/torchvision.datasets.KMNIST.html
+++ b/0.16/generated/torchvision.datasets.KMNIST.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>KMNIST &mdash; Torchvision main documentation</title>
+  <title>KMNIST &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Kinetics.html b/0.16/generated/torchvision.datasets.Kinetics.html
index 16ad9323afb..fd613b08919 100644
--- a/0.16/generated/torchvision.datasets.Kinetics.html
+++ b/0.16/generated/torchvision.datasets.Kinetics.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Kinetics &mdash; Torchvision main documentation</title>
+  <title>Kinetics &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Kitti.html b/0.16/generated/torchvision.datasets.Kitti.html
index a141be050c0..758905eb2a4 100644
--- a/0.16/generated/torchvision.datasets.Kitti.html
+++ b/0.16/generated/torchvision.datasets.Kitti.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Kitti &mdash; Torchvision main documentation</title>
+  <title>Kitti &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Kitti2012Stereo.html b/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
index db8cb559a03..b5ad87b4cff 100644
--- a/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
+++ b/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Kitti2012Stereo &mdash; Torchvision main documentation</title>
+  <title>Kitti2012Stereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Kitti2015Stereo.html b/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
index 0240118af6f..7afc75041ad 100644
--- a/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
+++ b/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Kitti2015Stereo &mdash; Torchvision main documentation</title>
+  <title>Kitti2015Stereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.KittiFlow.html b/0.16/generated/torchvision.datasets.KittiFlow.html
index ffff8d80473..8ecf5ba2afa 100644
--- a/0.16/generated/torchvision.datasets.KittiFlow.html
+++ b/0.16/generated/torchvision.datasets.KittiFlow.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>KittiFlow &mdash; Torchvision main documentation</title>
+  <title>KittiFlow &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.LFWPairs.html b/0.16/generated/torchvision.datasets.LFWPairs.html
index 77953b0ad7d..b18c1ebd898 100644
--- a/0.16/generated/torchvision.datasets.LFWPairs.html
+++ b/0.16/generated/torchvision.datasets.LFWPairs.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>LFWPairs &mdash; Torchvision main documentation</title>
+  <title>LFWPairs &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.LFWPeople.html b/0.16/generated/torchvision.datasets.LFWPeople.html
index b7c4930198f..590c967e7dd 100644
--- a/0.16/generated/torchvision.datasets.LFWPeople.html
+++ b/0.16/generated/torchvision.datasets.LFWPeople.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>LFWPeople &mdash; Torchvision main documentation</title>
+  <title>LFWPeople &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.LSUN.html b/0.16/generated/torchvision.datasets.LSUN.html
index 3a3a10d6a67..bea16e91a83 100644
--- a/0.16/generated/torchvision.datasets.LSUN.html
+++ b/0.16/generated/torchvision.datasets.LSUN.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>LSUN &mdash; Torchvision main documentation</title>
+  <title>LSUN &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.MNIST.html b/0.16/generated/torchvision.datasets.MNIST.html
index 2614af13161..4cc06db2623 100644
--- a/0.16/generated/torchvision.datasets.MNIST.html
+++ b/0.16/generated/torchvision.datasets.MNIST.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MNIST &mdash; Torchvision main documentation</title>
+  <title>MNIST &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html b/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
index 8d93d3c043e..3793b670b11 100644
--- a/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
+++ b/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Middlebury2014Stereo &mdash; Torchvision main documentation</title>
+  <title>Middlebury2014Stereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.MovingMNIST.html b/0.16/generated/torchvision.datasets.MovingMNIST.html
index cc6651d1e26..a3f8a777407 100644
--- a/0.16/generated/torchvision.datasets.MovingMNIST.html
+++ b/0.16/generated/torchvision.datasets.MovingMNIST.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MovingMNIST &mdash; Torchvision main documentation</title>
+  <title>MovingMNIST &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Omniglot.html b/0.16/generated/torchvision.datasets.Omniglot.html
index dfcd7049ff0..60349a8d2d4 100644
--- a/0.16/generated/torchvision.datasets.Omniglot.html
+++ b/0.16/generated/torchvision.datasets.Omniglot.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Omniglot &mdash; Torchvision main documentation</title>
+  <title>Omniglot &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.OxfordIIITPet.html b/0.16/generated/torchvision.datasets.OxfordIIITPet.html
index 6c570848ba4..eb68787226c 100644
--- a/0.16/generated/torchvision.datasets.OxfordIIITPet.html
+++ b/0.16/generated/torchvision.datasets.OxfordIIITPet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>OxfordIIITPet &mdash; Torchvision main documentation</title>
+  <title>OxfordIIITPet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.PCAM.html b/0.16/generated/torchvision.datasets.PCAM.html
index 964a4600ca4..b7969ad6670 100644
--- a/0.16/generated/torchvision.datasets.PCAM.html
+++ b/0.16/generated/torchvision.datasets.PCAM.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>PCAM &mdash; Torchvision main documentation</title>
+  <title>PCAM &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.PhotoTour.html b/0.16/generated/torchvision.datasets.PhotoTour.html
index 3851a5e8d11..62bd154e410 100644
--- a/0.16/generated/torchvision.datasets.PhotoTour.html
+++ b/0.16/generated/torchvision.datasets.PhotoTour.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>PhotoTour &mdash; Torchvision main documentation</title>
+  <title>PhotoTour &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Places365.html b/0.16/generated/torchvision.datasets.Places365.html
index 8d44bf41456..f506d3fe9c7 100644
--- a/0.16/generated/torchvision.datasets.Places365.html
+++ b/0.16/generated/torchvision.datasets.Places365.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Places365 &mdash; Torchvision main documentation</title>
+  <title>Places365 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.QMNIST.html b/0.16/generated/torchvision.datasets.QMNIST.html
index 66ac98e9170..fd0d141c6e2 100644
--- a/0.16/generated/torchvision.datasets.QMNIST.html
+++ b/0.16/generated/torchvision.datasets.QMNIST.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>QMNIST &mdash; Torchvision main documentation</title>
+  <title>QMNIST &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.RenderedSST2.html b/0.16/generated/torchvision.datasets.RenderedSST2.html
index 2d73bfb0ecb..795623cef9f 100644
--- a/0.16/generated/torchvision.datasets.RenderedSST2.html
+++ b/0.16/generated/torchvision.datasets.RenderedSST2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RenderedSST2 &mdash; Torchvision main documentation</title>
+  <title>RenderedSST2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.SBDataset.html b/0.16/generated/torchvision.datasets.SBDataset.html
index 25ebb5b28d3..76d234848f6 100644
--- a/0.16/generated/torchvision.datasets.SBDataset.html
+++ b/0.16/generated/torchvision.datasets.SBDataset.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SBDataset &mdash; Torchvision main documentation</title>
+  <title>SBDataset &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.SBU.html b/0.16/generated/torchvision.datasets.SBU.html
index 82ed892acfd..8f484ef458f 100644
--- a/0.16/generated/torchvision.datasets.SBU.html
+++ b/0.16/generated/torchvision.datasets.SBU.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SBU &mdash; Torchvision main documentation</title>
+  <title>SBU &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.SEMEION.html b/0.16/generated/torchvision.datasets.SEMEION.html
index 1da59be1584..50443dfea49 100644
--- a/0.16/generated/torchvision.datasets.SEMEION.html
+++ b/0.16/generated/torchvision.datasets.SEMEION.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SEMEION &mdash; Torchvision main documentation</title>
+  <title>SEMEION &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.STL10.html b/0.16/generated/torchvision.datasets.STL10.html
index 2e353a32219..992c83018be 100644
--- a/0.16/generated/torchvision.datasets.STL10.html
+++ b/0.16/generated/torchvision.datasets.STL10.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>STL10 &mdash; Torchvision main documentation</title>
+  <title>STL10 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.SUN397.html b/0.16/generated/torchvision.datasets.SUN397.html
index 6a3f04f5141..e8a5a27cb78 100644
--- a/0.16/generated/torchvision.datasets.SUN397.html
+++ b/0.16/generated/torchvision.datasets.SUN397.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SUN397 &mdash; Torchvision main documentation</title>
+  <title>SUN397 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.SVHN.html b/0.16/generated/torchvision.datasets.SVHN.html
index e00f97b40c3..d0fe0a8a178 100644
--- a/0.16/generated/torchvision.datasets.SVHN.html
+++ b/0.16/generated/torchvision.datasets.SVHN.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SVHN &mdash; Torchvision main documentation</title>
+  <title>SVHN &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.SceneFlowStereo.html b/0.16/generated/torchvision.datasets.SceneFlowStereo.html
index 91414dacec5..5b5ca5138b4 100644
--- a/0.16/generated/torchvision.datasets.SceneFlowStereo.html
+++ b/0.16/generated/torchvision.datasets.SceneFlowStereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SceneFlowStereo &mdash; Torchvision main documentation</title>
+  <title>SceneFlowStereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.Sintel.html b/0.16/generated/torchvision.datasets.Sintel.html
index ddf4d2be19c..dafc35162a3 100644
--- a/0.16/generated/torchvision.datasets.Sintel.html
+++ b/0.16/generated/torchvision.datasets.Sintel.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Sintel &mdash; Torchvision main documentation</title>
+  <title>Sintel &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.SintelStereo.html b/0.16/generated/torchvision.datasets.SintelStereo.html
index fdcff746521..4c9c650de75 100644
--- a/0.16/generated/torchvision.datasets.SintelStereo.html
+++ b/0.16/generated/torchvision.datasets.SintelStereo.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SintelStereo &mdash; Torchvision main documentation</title>
+  <title>SintelStereo &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.StanfordCars.html b/0.16/generated/torchvision.datasets.StanfordCars.html
index f049b422ab9..254acf23e62 100644
--- a/0.16/generated/torchvision.datasets.StanfordCars.html
+++ b/0.16/generated/torchvision.datasets.StanfordCars.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>StanfordCars &mdash; Torchvision main documentation</title>
+  <title>StanfordCars &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.UCF101.html b/0.16/generated/torchvision.datasets.UCF101.html
index 7c93a606e06..5bc9d032c65 100644
--- a/0.16/generated/torchvision.datasets.UCF101.html
+++ b/0.16/generated/torchvision.datasets.UCF101.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>UCF101 &mdash; Torchvision main documentation</title>
+  <title>UCF101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.USPS.html b/0.16/generated/torchvision.datasets.USPS.html
index b7832ac8e73..8cb037c53c5 100644
--- a/0.16/generated/torchvision.datasets.USPS.html
+++ b/0.16/generated/torchvision.datasets.USPS.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>USPS &mdash; Torchvision main documentation</title>
+  <title>USPS &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.VOCDetection.html b/0.16/generated/torchvision.datasets.VOCDetection.html
index 2bb5787e40f..884bc1819ad 100644
--- a/0.16/generated/torchvision.datasets.VOCDetection.html
+++ b/0.16/generated/torchvision.datasets.VOCDetection.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>VOCDetection &mdash; Torchvision main documentation</title>
+  <title>VOCDetection &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.VOCSegmentation.html b/0.16/generated/torchvision.datasets.VOCSegmentation.html
index 13d51c6d93a..4a54e2c0424 100644
--- a/0.16/generated/torchvision.datasets.VOCSegmentation.html
+++ b/0.16/generated/torchvision.datasets.VOCSegmentation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>VOCSegmentation &mdash; Torchvision main documentation</title>
+  <title>VOCSegmentation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.VisionDataset.html b/0.16/generated/torchvision.datasets.VisionDataset.html
index 7ee924321fd..769ad4163c0 100644
--- a/0.16/generated/torchvision.datasets.VisionDataset.html
+++ b/0.16/generated/torchvision.datasets.VisionDataset.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>VisionDataset &mdash; Torchvision main documentation</title>
+  <title>VisionDataset &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.WIDERFace.html b/0.16/generated/torchvision.datasets.WIDERFace.html
index 153640a2767..e5adef2d4f8 100644
--- a/0.16/generated/torchvision.datasets.WIDERFace.html
+++ b/0.16/generated/torchvision.datasets.WIDERFace.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>WIDERFace &mdash; Torchvision main documentation</title>
+  <title>WIDERFace &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html b/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
index 3e42acc468b..1b24f496578 100644
--- a/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
+++ b/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>wrap_dataset_for_transforms_v2 &mdash; Torchvision main documentation</title>
+  <title>wrap_dataset_for_transforms_v2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.ImageReadMode.html b/0.16/generated/torchvision.io.ImageReadMode.html
index f05acafc5fd..7065bc36e39 100644
--- a/0.16/generated/torchvision.io.ImageReadMode.html
+++ b/0.16/generated/torchvision.io.ImageReadMode.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ImageReadMode &mdash; Torchvision main documentation</title>
+  <title>ImageReadMode &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.VideoReader.html b/0.16/generated/torchvision.io.VideoReader.html
index bf3349b7027..bd282ce64a5 100644
--- a/0.16/generated/torchvision.io.VideoReader.html
+++ b/0.16/generated/torchvision.io.VideoReader.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>VideoReader &mdash; Torchvision main documentation</title>
+  <title>VideoReader &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.decode_image.html b/0.16/generated/torchvision.io.decode_image.html
index 0c2118019d0..bc9a356acc0 100644
--- a/0.16/generated/torchvision.io.decode_image.html
+++ b/0.16/generated/torchvision.io.decode_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>decode_image &mdash; Torchvision main documentation</title>
+  <title>decode_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.decode_jpeg.html b/0.16/generated/torchvision.io.decode_jpeg.html
index b8c30f547b6..df8256a413f 100644
--- a/0.16/generated/torchvision.io.decode_jpeg.html
+++ b/0.16/generated/torchvision.io.decode_jpeg.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>decode_jpeg &mdash; Torchvision main documentation</title>
+  <title>decode_jpeg &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.decode_png.html b/0.16/generated/torchvision.io.decode_png.html
index 3c5e422660b..ffc7504cd1e 100644
--- a/0.16/generated/torchvision.io.decode_png.html
+++ b/0.16/generated/torchvision.io.decode_png.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>decode_png &mdash; Torchvision main documentation</title>
+  <title>decode_png &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.encode_jpeg.html b/0.16/generated/torchvision.io.encode_jpeg.html
index 734a4dc2d86..7658efd058b 100644
--- a/0.16/generated/torchvision.io.encode_jpeg.html
+++ b/0.16/generated/torchvision.io.encode_jpeg.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>encode_jpeg &mdash; Torchvision main documentation</title>
+  <title>encode_jpeg &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.encode_png.html b/0.16/generated/torchvision.io.encode_png.html
index 091ba3bc79d..9660c1fb4c2 100644
--- a/0.16/generated/torchvision.io.encode_png.html
+++ b/0.16/generated/torchvision.io.encode_png.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>encode_png &mdash; Torchvision main documentation</title>
+  <title>encode_png &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.read_file.html b/0.16/generated/torchvision.io.read_file.html
index 5e27d0bac92..ca7e78d407c 100644
--- a/0.16/generated/torchvision.io.read_file.html
+++ b/0.16/generated/torchvision.io.read_file.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>read_file &mdash; Torchvision main documentation</title>
+  <title>read_file &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.read_image.html b/0.16/generated/torchvision.io.read_image.html
index 2d1a503d055..c446ba25371 100644
--- a/0.16/generated/torchvision.io.read_image.html
+++ b/0.16/generated/torchvision.io.read_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>read_image &mdash; Torchvision main documentation</title>
+  <title>read_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.read_video.html b/0.16/generated/torchvision.io.read_video.html
index 6bb5875d884..96e3cf6024b 100644
--- a/0.16/generated/torchvision.io.read_video.html
+++ b/0.16/generated/torchvision.io.read_video.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>read_video &mdash; Torchvision main documentation</title>
+  <title>read_video &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.read_video_timestamps.html b/0.16/generated/torchvision.io.read_video_timestamps.html
index 132ae9a6082..ce5fcfd18fb 100644
--- a/0.16/generated/torchvision.io.read_video_timestamps.html
+++ b/0.16/generated/torchvision.io.read_video_timestamps.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>read_video_timestamps &mdash; Torchvision main documentation</title>
+  <title>read_video_timestamps &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.write_file.html b/0.16/generated/torchvision.io.write_file.html
index abae8b2a7c6..2ffa619b9ff 100644
--- a/0.16/generated/torchvision.io.write_file.html
+++ b/0.16/generated/torchvision.io.write_file.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>write_file &mdash; Torchvision main documentation</title>
+  <title>write_file &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.write_jpeg.html b/0.16/generated/torchvision.io.write_jpeg.html
index 9d90a8147fe..5fc545faf4d 100644
--- a/0.16/generated/torchvision.io.write_jpeg.html
+++ b/0.16/generated/torchvision.io.write_jpeg.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>write_jpeg &mdash; Torchvision main documentation</title>
+  <title>write_jpeg &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.write_png.html b/0.16/generated/torchvision.io.write_png.html
index 7c9ead9d9d0..786f38c431f 100644
--- a/0.16/generated/torchvision.io.write_png.html
+++ b/0.16/generated/torchvision.io.write_png.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>write_png &mdash; Torchvision main documentation</title>
+  <title>write_png &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.io.write_video.html b/0.16/generated/torchvision.io.write_video.html
index cb28b16ce4d..08d6960a69f 100644
--- a/0.16/generated/torchvision.io.write_video.html
+++ b/0.16/generated/torchvision.io.write_video.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>write_video &mdash; Torchvision main documentation</title>
+  <title>write_video &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html b/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
index 3e9cd57dcba..b26dc7d961e 100644
--- a/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
+++ b/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>create_feature_extractor &mdash; Torchvision main documentation</title>
+  <title>create_feature_extractor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html b/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
index c6e07f8ae2c..5b6c74a9399 100644
--- a/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
+++ b/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>get_graph_node_names &mdash; Torchvision main documentation</title>
+  <title>get_graph_node_names &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.models.get_model.html b/0.16/generated/torchvision.models.get_model.html
index 6e3b44d88a7..2a836fa4014 100644
--- a/0.16/generated/torchvision.models.get_model.html
+++ b/0.16/generated/torchvision.models.get_model.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>get_model &mdash; Torchvision main documentation</title>
+  <title>get_model &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.models.get_model_weights.html b/0.16/generated/torchvision.models.get_model_weights.html
index cd1fea6ff4e..31ec89e4858 100644
--- a/0.16/generated/torchvision.models.get_model_weights.html
+++ b/0.16/generated/torchvision.models.get_model_weights.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>get_model_weights &mdash; Torchvision main documentation</title>
+  <title>get_model_weights &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.models.get_weight.html b/0.16/generated/torchvision.models.get_weight.html
index 0ed2839d831..8b5ba263a37 100644
--- a/0.16/generated/torchvision.models.get_weight.html
+++ b/0.16/generated/torchvision.models.get_weight.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>get_weight &mdash; Torchvision main documentation</title>
+  <title>get_weight &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.models.list_models.html b/0.16/generated/torchvision.models.list_models.html
index efe02d02c29..fe4e509ae95 100644
--- a/0.16/generated/torchvision.models.list_models.html
+++ b/0.16/generated/torchvision.models.list_models.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>list_models &mdash; Torchvision main documentation</title>
+  <title>list_models &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.Conv2dNormActivation.html b/0.16/generated/torchvision.ops.Conv2dNormActivation.html
index d15274935a4..e6ffc320f8a 100644
--- a/0.16/generated/torchvision.ops.Conv2dNormActivation.html
+++ b/0.16/generated/torchvision.ops.Conv2dNormActivation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Conv2dNormActivation &mdash; Torchvision main documentation</title>
+  <title>Conv2dNormActivation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.Conv3dNormActivation.html b/0.16/generated/torchvision.ops.Conv3dNormActivation.html
index 4f77e20bd81..a5bd457f180 100644
--- a/0.16/generated/torchvision.ops.Conv3dNormActivation.html
+++ b/0.16/generated/torchvision.ops.Conv3dNormActivation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Conv3dNormActivation &mdash; Torchvision main documentation</title>
+  <title>Conv3dNormActivation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.DeformConv2d.html b/0.16/generated/torchvision.ops.DeformConv2d.html
index dcf27776e33..fb7ca7e8930 100644
--- a/0.16/generated/torchvision.ops.DeformConv2d.html
+++ b/0.16/generated/torchvision.ops.DeformConv2d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>DeformConv2d &mdash; Torchvision main documentation</title>
+  <title>DeformConv2d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.DropBlock2d.html b/0.16/generated/torchvision.ops.DropBlock2d.html
index 91a06b0fd94..f1d1c44904a 100644
--- a/0.16/generated/torchvision.ops.DropBlock2d.html
+++ b/0.16/generated/torchvision.ops.DropBlock2d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>DropBlock2d &mdash; Torchvision main documentation</title>
+  <title>DropBlock2d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.DropBlock3d.html b/0.16/generated/torchvision.ops.DropBlock3d.html
index cd44591a1ec..929ae1b7ea1 100644
--- a/0.16/generated/torchvision.ops.DropBlock3d.html
+++ b/0.16/generated/torchvision.ops.DropBlock3d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>DropBlock3d &mdash; Torchvision main documentation</title>
+  <title>DropBlock3d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html b/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
index fc019a3a548..940aa008aa4 100644
--- a/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
+++ b/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FeaturePyramidNetwork &mdash; Torchvision main documentation</title>
+  <title>FeaturePyramidNetwork &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html b/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
index 95632534a2d..39e584c43f1 100644
--- a/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
+++ b/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FrozenBatchNorm2d &mdash; Torchvision main documentation</title>
+  <title>FrozenBatchNorm2d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.MLP.html b/0.16/generated/torchvision.ops.MLP.html
index 507415ffebf..5b885ed3556 100644
--- a/0.16/generated/torchvision.ops.MLP.html
+++ b/0.16/generated/torchvision.ops.MLP.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MLP &mdash; Torchvision main documentation</title>
+  <title>MLP &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html b/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
index 2b11819319e..d3ca072b438 100644
--- a/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
+++ b/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MultiScaleRoIAlign &mdash; Torchvision main documentation</title>
+  <title>MultiScaleRoIAlign &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.PSRoIAlign.html b/0.16/generated/torchvision.ops.PSRoIAlign.html
index c6150accef0..d1f90023131 100644
--- a/0.16/generated/torchvision.ops.PSRoIAlign.html
+++ b/0.16/generated/torchvision.ops.PSRoIAlign.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>PSRoIAlign &mdash; Torchvision main documentation</title>
+  <title>PSRoIAlign &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.PSRoIPool.html b/0.16/generated/torchvision.ops.PSRoIPool.html
index 93103ae7e3d..ab43e98c3ac 100644
--- a/0.16/generated/torchvision.ops.PSRoIPool.html
+++ b/0.16/generated/torchvision.ops.PSRoIPool.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>PSRoIPool &mdash; Torchvision main documentation</title>
+  <title>PSRoIPool &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.Permute.html b/0.16/generated/torchvision.ops.Permute.html
index 0e9601a1835..05ae6e88a9c 100644
--- a/0.16/generated/torchvision.ops.Permute.html
+++ b/0.16/generated/torchvision.ops.Permute.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Permute &mdash; Torchvision main documentation</title>
+  <title>Permute &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.RoIAlign.html b/0.16/generated/torchvision.ops.RoIAlign.html
index 4986169ed94..110c9fbb9ba 100644
--- a/0.16/generated/torchvision.ops.RoIAlign.html
+++ b/0.16/generated/torchvision.ops.RoIAlign.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RoIAlign &mdash; Torchvision main documentation</title>
+  <title>RoIAlign &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.RoIPool.html b/0.16/generated/torchvision.ops.RoIPool.html
index e7d4df78a54..60643f5f02b 100644
--- a/0.16/generated/torchvision.ops.RoIPool.html
+++ b/0.16/generated/torchvision.ops.RoIPool.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RoIPool &mdash; Torchvision main documentation</title>
+  <title>RoIPool &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.SqueezeExcitation.html b/0.16/generated/torchvision.ops.SqueezeExcitation.html
index 371445f8893..89ada5ef725 100644
--- a/0.16/generated/torchvision.ops.SqueezeExcitation.html
+++ b/0.16/generated/torchvision.ops.SqueezeExcitation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SqueezeExcitation &mdash; Torchvision main documentation</title>
+  <title>SqueezeExcitation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.StochasticDepth.html b/0.16/generated/torchvision.ops.StochasticDepth.html
index 368046d3ac7..271bbf135c6 100644
--- a/0.16/generated/torchvision.ops.StochasticDepth.html
+++ b/0.16/generated/torchvision.ops.StochasticDepth.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>StochasticDepth &mdash; Torchvision main documentation</title>
+  <title>StochasticDepth &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.batched_nms.html b/0.16/generated/torchvision.ops.batched_nms.html
index 041c14fc230..0d3e32c1b7a 100644
--- a/0.16/generated/torchvision.ops.batched_nms.html
+++ b/0.16/generated/torchvision.ops.batched_nms.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>batched_nms &mdash; Torchvision main documentation</title>
+  <title>batched_nms &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.box_area.html b/0.16/generated/torchvision.ops.box_area.html
index c1ff6c7d1a8..324708929ad 100644
--- a/0.16/generated/torchvision.ops.box_area.html
+++ b/0.16/generated/torchvision.ops.box_area.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>box_area &mdash; Torchvision main documentation</title>
+  <title>box_area &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.box_convert.html b/0.16/generated/torchvision.ops.box_convert.html
index c394b1aba92..1db03a96da0 100644
--- a/0.16/generated/torchvision.ops.box_convert.html
+++ b/0.16/generated/torchvision.ops.box_convert.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>box_convert &mdash; Torchvision main documentation</title>
+  <title>box_convert &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.box_iou.html b/0.16/generated/torchvision.ops.box_iou.html
index 1fc77fe5be5..3e774addf0f 100644
--- a/0.16/generated/torchvision.ops.box_iou.html
+++ b/0.16/generated/torchvision.ops.box_iou.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>box_iou &mdash; Torchvision main documentation</title>
+  <title>box_iou &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.clip_boxes_to_image.html b/0.16/generated/torchvision.ops.clip_boxes_to_image.html
index 91b0fac2476..0005bafed09 100644
--- a/0.16/generated/torchvision.ops.clip_boxes_to_image.html
+++ b/0.16/generated/torchvision.ops.clip_boxes_to_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>clip_boxes_to_image &mdash; Torchvision main documentation</title>
+  <title>clip_boxes_to_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.complete_box_iou.html b/0.16/generated/torchvision.ops.complete_box_iou.html
index d32103c55ad..839e1cc7474 100644
--- a/0.16/generated/torchvision.ops.complete_box_iou.html
+++ b/0.16/generated/torchvision.ops.complete_box_iou.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>complete_box_iou &mdash; Torchvision main documentation</title>
+  <title>complete_box_iou &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.complete_box_iou_loss.html b/0.16/generated/torchvision.ops.complete_box_iou_loss.html
index d1936b44d11..b48ffdc3846 100644
--- a/0.16/generated/torchvision.ops.complete_box_iou_loss.html
+++ b/0.16/generated/torchvision.ops.complete_box_iou_loss.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>complete_box_iou_loss &mdash; Torchvision main documentation</title>
+  <title>complete_box_iou_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.deform_conv2d.html b/0.16/generated/torchvision.ops.deform_conv2d.html
index 48ad785c7e4..8da88f23b41 100644
--- a/0.16/generated/torchvision.ops.deform_conv2d.html
+++ b/0.16/generated/torchvision.ops.deform_conv2d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>deform_conv2d &mdash; Torchvision main documentation</title>
+  <title>deform_conv2d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.distance_box_iou.html b/0.16/generated/torchvision.ops.distance_box_iou.html
index ae30fbcec40..ac5df337ed8 100644
--- a/0.16/generated/torchvision.ops.distance_box_iou.html
+++ b/0.16/generated/torchvision.ops.distance_box_iou.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>distance_box_iou &mdash; Torchvision main documentation</title>
+  <title>distance_box_iou &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.distance_box_iou_loss.html b/0.16/generated/torchvision.ops.distance_box_iou_loss.html
index 7e826e79d77..37c3894bdda 100644
--- a/0.16/generated/torchvision.ops.distance_box_iou_loss.html
+++ b/0.16/generated/torchvision.ops.distance_box_iou_loss.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>distance_box_iou_loss &mdash; Torchvision main documentation</title>
+  <title>distance_box_iou_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.drop_block2d.html b/0.16/generated/torchvision.ops.drop_block2d.html
index b486fb5e5f2..94661f239ac 100644
--- a/0.16/generated/torchvision.ops.drop_block2d.html
+++ b/0.16/generated/torchvision.ops.drop_block2d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>drop_block2d &mdash; Torchvision main documentation</title>
+  <title>drop_block2d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.drop_block3d.html b/0.16/generated/torchvision.ops.drop_block3d.html
index d27de613a1d..0f013eb72c9 100644
--- a/0.16/generated/torchvision.ops.drop_block3d.html
+++ b/0.16/generated/torchvision.ops.drop_block3d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>drop_block3d &mdash; Torchvision main documentation</title>
+  <title>drop_block3d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.generalized_box_iou.html b/0.16/generated/torchvision.ops.generalized_box_iou.html
index c95c66f15cd..74b297196eb 100644
--- a/0.16/generated/torchvision.ops.generalized_box_iou.html
+++ b/0.16/generated/torchvision.ops.generalized_box_iou.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>generalized_box_iou &mdash; Torchvision main documentation</title>
+  <title>generalized_box_iou &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.generalized_box_iou_loss.html b/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
index 06e90c953b4..9bcfca3dbf6 100644
--- a/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
+++ b/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>generalized_box_iou_loss &mdash; Torchvision main documentation</title>
+  <title>generalized_box_iou_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.masks_to_boxes.html b/0.16/generated/torchvision.ops.masks_to_boxes.html
index 75f8c3b8fec..2e264b2bdff 100644
--- a/0.16/generated/torchvision.ops.masks_to_boxes.html
+++ b/0.16/generated/torchvision.ops.masks_to_boxes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>masks_to_boxes &mdash; Torchvision main documentation</title>
+  <title>masks_to_boxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.nms.html b/0.16/generated/torchvision.ops.nms.html
index 2c7e30ce6d2..fb542d16c2e 100644
--- a/0.16/generated/torchvision.ops.nms.html
+++ b/0.16/generated/torchvision.ops.nms.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>nms &mdash; Torchvision main documentation</title>
+  <title>nms &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.ps_roi_align.html b/0.16/generated/torchvision.ops.ps_roi_align.html
index 17d7f5a049b..90d0482a841 100644
--- a/0.16/generated/torchvision.ops.ps_roi_align.html
+++ b/0.16/generated/torchvision.ops.ps_roi_align.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ps_roi_align &mdash; Torchvision main documentation</title>
+  <title>ps_roi_align &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.ps_roi_pool.html b/0.16/generated/torchvision.ops.ps_roi_pool.html
index ad06dc9ea4a..05a047c92f3 100644
--- a/0.16/generated/torchvision.ops.ps_roi_pool.html
+++ b/0.16/generated/torchvision.ops.ps_roi_pool.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ps_roi_pool &mdash; Torchvision main documentation</title>
+  <title>ps_roi_pool &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.remove_small_boxes.html b/0.16/generated/torchvision.ops.remove_small_boxes.html
index 10993509bd6..8d1de78acf4 100644
--- a/0.16/generated/torchvision.ops.remove_small_boxes.html
+++ b/0.16/generated/torchvision.ops.remove_small_boxes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>remove_small_boxes &mdash; Torchvision main documentation</title>
+  <title>remove_small_boxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.roi_align.html b/0.16/generated/torchvision.ops.roi_align.html
index ad60cd32e40..513ef0f3456 100644
--- a/0.16/generated/torchvision.ops.roi_align.html
+++ b/0.16/generated/torchvision.ops.roi_align.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>roi_align &mdash; Torchvision main documentation</title>
+  <title>roi_align &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.roi_pool.html b/0.16/generated/torchvision.ops.roi_pool.html
index 5c6a79656f2..072db500e4e 100644
--- a/0.16/generated/torchvision.ops.roi_pool.html
+++ b/0.16/generated/torchvision.ops.roi_pool.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>roi_pool &mdash; Torchvision main documentation</title>
+  <title>roi_pool &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.sigmoid_focal_loss.html b/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
index 1e14abf0771..972cb222ce9 100644
--- a/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
+++ b/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>sigmoid_focal_loss &mdash; Torchvision main documentation</title>
+  <title>sigmoid_focal_loss &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.ops.stochastic_depth.html b/0.16/generated/torchvision.ops.stochastic_depth.html
index 957599ccf9c..e78d79fe87f 100644
--- a/0.16/generated/torchvision.ops.stochastic_depth.html
+++ b/0.16/generated/torchvision.ops.stochastic_depth.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>stochastic_depth &mdash; Torchvision main documentation</title>
+  <title>stochastic_depth &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.AugMix.html b/0.16/generated/torchvision.transforms.AugMix.html
index b6954e152b6..40ba03e58ae 100644
--- a/0.16/generated/torchvision.transforms.AugMix.html
+++ b/0.16/generated/torchvision.transforms.AugMix.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>AugMix &mdash; Torchvision main documentation</title>
+  <title>AugMix &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.AutoAugment.html b/0.16/generated/torchvision.transforms.AutoAugment.html
index e10d5f37cc8..5a62377b8a4 100644
--- a/0.16/generated/torchvision.transforms.AutoAugment.html
+++ b/0.16/generated/torchvision.transforms.AutoAugment.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>AutoAugment &mdash; Torchvision main documentation</title>
+  <title>AutoAugment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html b/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
index 6bf3d128367..e4b58e04c14 100644
--- a/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
+++ b/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>AutoAugmentPolicy &mdash; Torchvision main documentation</title>
+  <title>AutoAugmentPolicy &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.CenterCrop.html b/0.16/generated/torchvision.transforms.CenterCrop.html
index 1ff70a8709c..7d4f4fa948a 100644
--- a/0.16/generated/torchvision.transforms.CenterCrop.html
+++ b/0.16/generated/torchvision.transforms.CenterCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CenterCrop &mdash; Torchvision main documentation</title>
+  <title>CenterCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.ColorJitter.html b/0.16/generated/torchvision.transforms.ColorJitter.html
index a168182fc9a..6a6a1d4ba89 100644
--- a/0.16/generated/torchvision.transforms.ColorJitter.html
+++ b/0.16/generated/torchvision.transforms.ColorJitter.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ColorJitter &mdash; Torchvision main documentation</title>
+  <title>ColorJitter &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.Compose.html b/0.16/generated/torchvision.transforms.Compose.html
index ff9dc550ba2..0e0b7aa56ff 100644
--- a/0.16/generated/torchvision.transforms.Compose.html
+++ b/0.16/generated/torchvision.transforms.Compose.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compose &mdash; Torchvision main documentation</title>
+  <title>Compose &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.ConvertImageDtype.html b/0.16/generated/torchvision.transforms.ConvertImageDtype.html
index 61a87942dc7..0a1198fd79a 100644
--- a/0.16/generated/torchvision.transforms.ConvertImageDtype.html
+++ b/0.16/generated/torchvision.transforms.ConvertImageDtype.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ConvertImageDtype &mdash; Torchvision main documentation</title>
+  <title>ConvertImageDtype &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.ElasticTransform.html b/0.16/generated/torchvision.transforms.ElasticTransform.html
index d142ca2d181..299824e53a9 100644
--- a/0.16/generated/torchvision.transforms.ElasticTransform.html
+++ b/0.16/generated/torchvision.transforms.ElasticTransform.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ElasticTransform &mdash; Torchvision main documentation</title>
+  <title>ElasticTransform &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.FiveCrop.html b/0.16/generated/torchvision.transforms.FiveCrop.html
index 2b601c8b8dd..23136e7c3d6 100644
--- a/0.16/generated/torchvision.transforms.FiveCrop.html
+++ b/0.16/generated/torchvision.transforms.FiveCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FiveCrop &mdash; Torchvision main documentation</title>
+  <title>FiveCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.GaussianBlur.html b/0.16/generated/torchvision.transforms.GaussianBlur.html
index 2ce48578078..b4bb8b2ceb2 100644
--- a/0.16/generated/torchvision.transforms.GaussianBlur.html
+++ b/0.16/generated/torchvision.transforms.GaussianBlur.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>GaussianBlur &mdash; Torchvision main documentation</title>
+  <title>GaussianBlur &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.Grayscale.html b/0.16/generated/torchvision.transforms.Grayscale.html
index cbfa68d1ebe..ee580052264 100644
--- a/0.16/generated/torchvision.transforms.Grayscale.html
+++ b/0.16/generated/torchvision.transforms.Grayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Grayscale &mdash; Torchvision main documentation</title>
+  <title>Grayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.Lambda.html b/0.16/generated/torchvision.transforms.Lambda.html
index 552e35e0271..d1b1392c803 100644
--- a/0.16/generated/torchvision.transforms.Lambda.html
+++ b/0.16/generated/torchvision.transforms.Lambda.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Lambda &mdash; Torchvision main documentation</title>
+  <title>Lambda &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.LinearTransformation.html b/0.16/generated/torchvision.transforms.LinearTransformation.html
index 670bb6a4a05..03950d65f51 100644
--- a/0.16/generated/torchvision.transforms.LinearTransformation.html
+++ b/0.16/generated/torchvision.transforms.LinearTransformation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>LinearTransformation &mdash; Torchvision main documentation</title>
+  <title>LinearTransformation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.Normalize.html b/0.16/generated/torchvision.transforms.Normalize.html
index d16cc57d9da..b1699ae2a6b 100644
--- a/0.16/generated/torchvision.transforms.Normalize.html
+++ b/0.16/generated/torchvision.transforms.Normalize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Normalize &mdash; Torchvision main documentation</title>
+  <title>Normalize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.PILToTensor.html b/0.16/generated/torchvision.transforms.PILToTensor.html
index 58d5e490713..bf010508d80 100644
--- a/0.16/generated/torchvision.transforms.PILToTensor.html
+++ b/0.16/generated/torchvision.transforms.PILToTensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>PILToTensor &mdash; Torchvision main documentation</title>
+  <title>PILToTensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.Pad.html b/0.16/generated/torchvision.transforms.Pad.html
index e6b941168d7..9ac225fb4c0 100644
--- a/0.16/generated/torchvision.transforms.Pad.html
+++ b/0.16/generated/torchvision.transforms.Pad.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Pad &mdash; Torchvision main documentation</title>
+  <title>Pad &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandAugment.html b/0.16/generated/torchvision.transforms.RandAugment.html
index d075f588062..cf00191a65c 100644
--- a/0.16/generated/torchvision.transforms.RandAugment.html
+++ b/0.16/generated/torchvision.transforms.RandAugment.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandAugment &mdash; Torchvision main documentation</title>
+  <title>RandAugment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html b/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
index fb2aead4b3a..b6c2fbed2af 100644
--- a/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
+++ b/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomAdjustSharpness &mdash; Torchvision main documentation</title>
+  <title>RandomAdjustSharpness &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomAffine.html b/0.16/generated/torchvision.transforms.RandomAffine.html
index 2b4bd491f16..37cc7999a66 100644
--- a/0.16/generated/torchvision.transforms.RandomAffine.html
+++ b/0.16/generated/torchvision.transforms.RandomAffine.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomAffine &mdash; Torchvision main documentation</title>
+  <title>RandomAffine &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomApply.html b/0.16/generated/torchvision.transforms.RandomApply.html
index e6d65d8c51b..c2a2a34b0a0 100644
--- a/0.16/generated/torchvision.transforms.RandomApply.html
+++ b/0.16/generated/torchvision.transforms.RandomApply.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomApply &mdash; Torchvision main documentation</title>
+  <title>RandomApply &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomAutocontrast.html b/0.16/generated/torchvision.transforms.RandomAutocontrast.html
index c0acd76745e..5cae86d3788 100644
--- a/0.16/generated/torchvision.transforms.RandomAutocontrast.html
+++ b/0.16/generated/torchvision.transforms.RandomAutocontrast.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomAutocontrast &mdash; Torchvision main documentation</title>
+  <title>RandomAutocontrast &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomChoice.html b/0.16/generated/torchvision.transforms.RandomChoice.html
index 63e6cd25f9b..41c466fdefd 100644
--- a/0.16/generated/torchvision.transforms.RandomChoice.html
+++ b/0.16/generated/torchvision.transforms.RandomChoice.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomChoice &mdash; Torchvision main documentation</title>
+  <title>RandomChoice &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomCrop.html b/0.16/generated/torchvision.transforms.RandomCrop.html
index 4fad394dcb2..66c3a10069d 100644
--- a/0.16/generated/torchvision.transforms.RandomCrop.html
+++ b/0.16/generated/torchvision.transforms.RandomCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomCrop &mdash; Torchvision main documentation</title>
+  <title>RandomCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomEqualize.html b/0.16/generated/torchvision.transforms.RandomEqualize.html
index dd76d823d6d..bcfbd2f1d97 100644
--- a/0.16/generated/torchvision.transforms.RandomEqualize.html
+++ b/0.16/generated/torchvision.transforms.RandomEqualize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomEqualize &mdash; Torchvision main documentation</title>
+  <title>RandomEqualize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomErasing.html b/0.16/generated/torchvision.transforms.RandomErasing.html
index 8bafeaf7e02..d78b11c281e 100644
--- a/0.16/generated/torchvision.transforms.RandomErasing.html
+++ b/0.16/generated/torchvision.transforms.RandomErasing.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomErasing &mdash; Torchvision main documentation</title>
+  <title>RandomErasing &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomGrayscale.html b/0.16/generated/torchvision.transforms.RandomGrayscale.html
index d7e3726d972..5af06784588 100644
--- a/0.16/generated/torchvision.transforms.RandomGrayscale.html
+++ b/0.16/generated/torchvision.transforms.RandomGrayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomGrayscale &mdash; Torchvision main documentation</title>
+  <title>RandomGrayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html b/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
index 9ad1632b198..c0d81173558 100644
--- a/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
+++ b/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomHorizontalFlip &mdash; Torchvision main documentation</title>
+  <title>RandomHorizontalFlip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomInvert.html b/0.16/generated/torchvision.transforms.RandomInvert.html
index 21886fd0ca1..d097a6a22d0 100644
--- a/0.16/generated/torchvision.transforms.RandomInvert.html
+++ b/0.16/generated/torchvision.transforms.RandomInvert.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomInvert &mdash; Torchvision main documentation</title>
+  <title>RandomInvert &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomOrder.html b/0.16/generated/torchvision.transforms.RandomOrder.html
index 9d847a6075b..69c533e24ec 100644
--- a/0.16/generated/torchvision.transforms.RandomOrder.html
+++ b/0.16/generated/torchvision.transforms.RandomOrder.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomOrder &mdash; Torchvision main documentation</title>
+  <title>RandomOrder &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomPerspective.html b/0.16/generated/torchvision.transforms.RandomPerspective.html
index d3123acc6a2..27105947acb 100644
--- a/0.16/generated/torchvision.transforms.RandomPerspective.html
+++ b/0.16/generated/torchvision.transforms.RandomPerspective.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomPerspective &mdash; Torchvision main documentation</title>
+  <title>RandomPerspective &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomPosterize.html b/0.16/generated/torchvision.transforms.RandomPosterize.html
index dd82bbdc747..24327b17a2a 100644
--- a/0.16/generated/torchvision.transforms.RandomPosterize.html
+++ b/0.16/generated/torchvision.transforms.RandomPosterize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomPosterize &mdash; Torchvision main documentation</title>
+  <title>RandomPosterize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomResizedCrop.html b/0.16/generated/torchvision.transforms.RandomResizedCrop.html
index 4ba310c3730..9125197b6f9 100644
--- a/0.16/generated/torchvision.transforms.RandomResizedCrop.html
+++ b/0.16/generated/torchvision.transforms.RandomResizedCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomResizedCrop &mdash; Torchvision main documentation</title>
+  <title>RandomResizedCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomRotation.html b/0.16/generated/torchvision.transforms.RandomRotation.html
index 7225ddde6a4..499bf30afd6 100644
--- a/0.16/generated/torchvision.transforms.RandomRotation.html
+++ b/0.16/generated/torchvision.transforms.RandomRotation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomRotation &mdash; Torchvision main documentation</title>
+  <title>RandomRotation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomSolarize.html b/0.16/generated/torchvision.transforms.RandomSolarize.html
index 874660927c1..809723fa7f8 100644
--- a/0.16/generated/torchvision.transforms.RandomSolarize.html
+++ b/0.16/generated/torchvision.transforms.RandomSolarize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomSolarize &mdash; Torchvision main documentation</title>
+  <title>RandomSolarize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.RandomVerticalFlip.html b/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
index e3f4b128f5b..58a0b892630 100644
--- a/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
+++ b/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomVerticalFlip &mdash; Torchvision main documentation</title>
+  <title>RandomVerticalFlip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.Resize.html b/0.16/generated/torchvision.transforms.Resize.html
index 3ecd987856f..a053c5aa53c 100644
--- a/0.16/generated/torchvision.transforms.Resize.html
+++ b/0.16/generated/torchvision.transforms.Resize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Resize &mdash; Torchvision main documentation</title>
+  <title>Resize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.TenCrop.html b/0.16/generated/torchvision.transforms.TenCrop.html
index 8b8cf350dcd..d3f087380f6 100644
--- a/0.16/generated/torchvision.transforms.TenCrop.html
+++ b/0.16/generated/torchvision.transforms.TenCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TenCrop &mdash; Torchvision main documentation</title>
+  <title>TenCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.ToPILImage.html b/0.16/generated/torchvision.transforms.ToPILImage.html
index db9c487adcd..eafba217d2e 100644
--- a/0.16/generated/torchvision.transforms.ToPILImage.html
+++ b/0.16/generated/torchvision.transforms.ToPILImage.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ToPILImage &mdash; Torchvision main documentation</title>
+  <title>ToPILImage &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.ToTensor.html b/0.16/generated/torchvision.transforms.ToTensor.html
index 43c4007a488..ae18f539807 100644
--- a/0.16/generated/torchvision.transforms.ToTensor.html
+++ b/0.16/generated/torchvision.transforms.ToTensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ToTensor &mdash; Torchvision main documentation</title>
+  <title>ToTensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.TrivialAugmentWide.html b/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
index 373249d0eb5..23607a9aac9 100644
--- a/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
+++ b/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TrivialAugmentWide &mdash; Torchvision main documentation</title>
+  <title>TrivialAugmentWide &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_brightness.html b/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
index c28975ab15e..77942c2662b 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_brightness &mdash; Torchvision main documentation</title>
+  <title>adjust_brightness &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_contrast.html b/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
index 6a9eb7c86a4..f730dd9d3ad 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_contrast &mdash; Torchvision main documentation</title>
+  <title>adjust_contrast &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_gamma.html b/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
index 192eca1875a..d3ff1b1cdfb 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_gamma &mdash; Torchvision main documentation</title>
+  <title>adjust_gamma &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_hue.html b/0.16/generated/torchvision.transforms.functional.adjust_hue.html
index f85476af428..e100be1f010 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_hue.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_hue.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_hue &mdash; Torchvision main documentation</title>
+  <title>adjust_hue &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_saturation.html b/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
index 4eb7681d79d..5dab767c142 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_saturation &mdash; Torchvision main documentation</title>
+  <title>adjust_saturation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html b/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
index 1077e33364e..32b0f69a829 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_sharpness &mdash; Torchvision main documentation</title>
+  <title>adjust_sharpness &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.affine.html b/0.16/generated/torchvision.transforms.functional.affine.html
index 60b11012ac6..e3a591776c0 100644
--- a/0.16/generated/torchvision.transforms.functional.affine.html
+++ b/0.16/generated/torchvision.transforms.functional.affine.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>affine &mdash; Torchvision main documentation</title>
+  <title>affine &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.autocontrast.html b/0.16/generated/torchvision.transforms.functional.autocontrast.html
index 3664f1a7244..c0156023e10 100644
--- a/0.16/generated/torchvision.transforms.functional.autocontrast.html
+++ b/0.16/generated/torchvision.transforms.functional.autocontrast.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>autocontrast &mdash; Torchvision main documentation</title>
+  <title>autocontrast &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.center_crop.html b/0.16/generated/torchvision.transforms.functional.center_crop.html
index 42b25606f2c..b2edfaa9a91 100644
--- a/0.16/generated/torchvision.transforms.functional.center_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.center_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>center_crop &mdash; Torchvision main documentation</title>
+  <title>center_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html b/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
index ac273d202c4..c37ac10f617 100644
--- a/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
+++ b/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>convert_image_dtype &mdash; Torchvision main documentation</title>
+  <title>convert_image_dtype &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.crop.html b/0.16/generated/torchvision.transforms.functional.crop.html
index f60da230509..109a7e5f78a 100644
--- a/0.16/generated/torchvision.transforms.functional.crop.html
+++ b/0.16/generated/torchvision.transforms.functional.crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>crop &mdash; Torchvision main documentation</title>
+  <title>crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.equalize.html b/0.16/generated/torchvision.transforms.functional.equalize.html
index 0ca3f3979ae..92b31463db8 100644
--- a/0.16/generated/torchvision.transforms.functional.equalize.html
+++ b/0.16/generated/torchvision.transforms.functional.equalize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>equalize &mdash; Torchvision main documentation</title>
+  <title>equalize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.erase.html b/0.16/generated/torchvision.transforms.functional.erase.html
index 72b95c39e06..7425e393ad9 100644
--- a/0.16/generated/torchvision.transforms.functional.erase.html
+++ b/0.16/generated/torchvision.transforms.functional.erase.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>erase &mdash; Torchvision main documentation</title>
+  <title>erase &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.five_crop.html b/0.16/generated/torchvision.transforms.functional.five_crop.html
index 1e596a638bc..5fbe85cdd5f 100644
--- a/0.16/generated/torchvision.transforms.functional.five_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.five_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>five_crop &mdash; Torchvision main documentation</title>
+  <title>five_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.gaussian_blur.html b/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
index 477eac916f2..b5ba2b91eb5 100644
--- a/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
+++ b/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>gaussian_blur &mdash; Torchvision main documentation</title>
+  <title>gaussian_blur &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.get_dimensions.html b/0.16/generated/torchvision.transforms.functional.get_dimensions.html
index 1282b5bf1b3..01d69c7b2de 100644
--- a/0.16/generated/torchvision.transforms.functional.get_dimensions.html
+++ b/0.16/generated/torchvision.transforms.functional.get_dimensions.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>get_dimensions &mdash; Torchvision main documentation</title>
+  <title>get_dimensions &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html b/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
index 0ea57f22264..12124024a0a 100644
--- a/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
+++ b/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>get_image_num_channels &mdash; Torchvision main documentation</title>
+  <title>get_image_num_channels &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.get_image_size.html b/0.16/generated/torchvision.transforms.functional.get_image_size.html
index 238f78b825c..ac21645e4b3 100644
--- a/0.16/generated/torchvision.transforms.functional.get_image_size.html
+++ b/0.16/generated/torchvision.transforms.functional.get_image_size.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>get_image_size &mdash; Torchvision main documentation</title>
+  <title>get_image_size &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.hflip.html b/0.16/generated/torchvision.transforms.functional.hflip.html
index b3b3151dea1..354f3682dcf 100644
--- a/0.16/generated/torchvision.transforms.functional.hflip.html
+++ b/0.16/generated/torchvision.transforms.functional.hflip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>hflip &mdash; Torchvision main documentation</title>
+  <title>hflip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.invert.html b/0.16/generated/torchvision.transforms.functional.invert.html
index a06fd54b78e..fe67087d4d0 100644
--- a/0.16/generated/torchvision.transforms.functional.invert.html
+++ b/0.16/generated/torchvision.transforms.functional.invert.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>invert &mdash; Torchvision main documentation</title>
+  <title>invert &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.normalize.html b/0.16/generated/torchvision.transforms.functional.normalize.html
index 8f3de5bd2c9..51a81bd06a5 100644
--- a/0.16/generated/torchvision.transforms.functional.normalize.html
+++ b/0.16/generated/torchvision.transforms.functional.normalize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>normalize &mdash; Torchvision main documentation</title>
+  <title>normalize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.pad.html b/0.16/generated/torchvision.transforms.functional.pad.html
index b194313f5a2..696014523bb 100644
--- a/0.16/generated/torchvision.transforms.functional.pad.html
+++ b/0.16/generated/torchvision.transforms.functional.pad.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>pad &mdash; Torchvision main documentation</title>
+  <title>pad &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.perspective.html b/0.16/generated/torchvision.transforms.functional.perspective.html
index 56d51646bad..412db98ab4a 100644
--- a/0.16/generated/torchvision.transforms.functional.perspective.html
+++ b/0.16/generated/torchvision.transforms.functional.perspective.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>perspective &mdash; Torchvision main documentation</title>
+  <title>perspective &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html b/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
index 44594c0faa3..a73f0063d1c 100644
--- a/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
+++ b/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>pil_to_tensor &mdash; Torchvision main documentation</title>
+  <title>pil_to_tensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.posterize.html b/0.16/generated/torchvision.transforms.functional.posterize.html
index 780a01da39d..f0029553f74 100644
--- a/0.16/generated/torchvision.transforms.functional.posterize.html
+++ b/0.16/generated/torchvision.transforms.functional.posterize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>posterize &mdash; Torchvision main documentation</title>
+  <title>posterize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.resize.html b/0.16/generated/torchvision.transforms.functional.resize.html
index a60352465a9..2ecf87f277c 100644
--- a/0.16/generated/torchvision.transforms.functional.resize.html
+++ b/0.16/generated/torchvision.transforms.functional.resize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resize &mdash; Torchvision main documentation</title>
+  <title>resize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.resized_crop.html b/0.16/generated/torchvision.transforms.functional.resized_crop.html
index 1a016c01a67..57de342d2b3 100644
--- a/0.16/generated/torchvision.transforms.functional.resized_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.resized_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resized_crop &mdash; Torchvision main documentation</title>
+  <title>resized_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html b/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
index 6574ce0b24f..05c62db9b06 100644
--- a/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>rgb_to_grayscale &mdash; Torchvision main documentation</title>
+  <title>rgb_to_grayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.rotate.html b/0.16/generated/torchvision.transforms.functional.rotate.html
index 5129213add5..f03b59096b2 100644
--- a/0.16/generated/torchvision.transforms.functional.rotate.html
+++ b/0.16/generated/torchvision.transforms.functional.rotate.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>rotate &mdash; Torchvision main documentation</title>
+  <title>rotate &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.solarize.html b/0.16/generated/torchvision.transforms.functional.solarize.html
index 1edebfb2672..c2f8acd1b9f 100644
--- a/0.16/generated/torchvision.transforms.functional.solarize.html
+++ b/0.16/generated/torchvision.transforms.functional.solarize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>solarize &mdash; Torchvision main documentation</title>
+  <title>solarize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.ten_crop.html b/0.16/generated/torchvision.transforms.functional.ten_crop.html
index 1ecfe453f73..9d57dad2413 100644
--- a/0.16/generated/torchvision.transforms.functional.ten_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.ten_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ten_crop &mdash; Torchvision main documentation</title>
+  <title>ten_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.to_grayscale.html b/0.16/generated/torchvision.transforms.functional.to_grayscale.html
index 0830b755b47..375d5568cf0 100644
--- a/0.16/generated/torchvision.transforms.functional.to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.functional.to_grayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>to_grayscale &mdash; Torchvision main documentation</title>
+  <title>to_grayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.to_pil_image.html b/0.16/generated/torchvision.transforms.functional.to_pil_image.html
index 2d896b3aa00..93c57ffb7e5 100644
--- a/0.16/generated/torchvision.transforms.functional.to_pil_image.html
+++ b/0.16/generated/torchvision.transforms.functional.to_pil_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>to_pil_image &mdash; Torchvision main documentation</title>
+  <title>to_pil_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.to_tensor.html b/0.16/generated/torchvision.transforms.functional.to_tensor.html
index dac36254aed..c869692f50c 100644
--- a/0.16/generated/torchvision.transforms.functional.to_tensor.html
+++ b/0.16/generated/torchvision.transforms.functional.to_tensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>to_tensor &mdash; Torchvision main documentation</title>
+  <title>to_tensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.functional.vflip.html b/0.16/generated/torchvision.transforms.functional.vflip.html
index 7a89fed2977..28eda1da93c 100644
--- a/0.16/generated/torchvision.transforms.functional.vflip.html
+++ b/0.16/generated/torchvision.transforms.functional.vflip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vflip &mdash; Torchvision main documentation</title>
+  <title>vflip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.AugMix.html b/0.16/generated/torchvision.transforms.v2.AugMix.html
index a0125765ae9..8483b176794 100644
--- a/0.16/generated/torchvision.transforms.v2.AugMix.html
+++ b/0.16/generated/torchvision.transforms.v2.AugMix.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>AugMix &mdash; Torchvision main documentation</title>
+  <title>AugMix &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.AutoAugment.html b/0.16/generated/torchvision.transforms.v2.AutoAugment.html
index 11a38ca11fc..96ce67633f6 100644
--- a/0.16/generated/torchvision.transforms.v2.AutoAugment.html
+++ b/0.16/generated/torchvision.transforms.v2.AutoAugment.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>AutoAugment &mdash; Torchvision main documentation</title>
+  <title>AutoAugment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.CenterCrop.html b/0.16/generated/torchvision.transforms.v2.CenterCrop.html
index 1865db23374..bb2b038fdd9 100644
--- a/0.16/generated/torchvision.transforms.v2.CenterCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.CenterCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CenterCrop &mdash; Torchvision main documentation</title>
+  <title>CenterCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html b/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
index 08ad02a21ba..7dc73291142 100644
--- a/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
+++ b/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ClampBoundingBoxes &mdash; Torchvision main documentation</title>
+  <title>ClampBoundingBoxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ColorJitter.html b/0.16/generated/torchvision.transforms.v2.ColorJitter.html
index b8fecd96f9f..66bf46f56a0 100644
--- a/0.16/generated/torchvision.transforms.v2.ColorJitter.html
+++ b/0.16/generated/torchvision.transforms.v2.ColorJitter.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ColorJitter &mdash; Torchvision main documentation</title>
+  <title>ColorJitter &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.Compose.html b/0.16/generated/torchvision.transforms.v2.Compose.html
index 7b5af6e35da..f7f58b23fdd 100644
--- a/0.16/generated/torchvision.transforms.v2.Compose.html
+++ b/0.16/generated/torchvision.transforms.v2.Compose.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Compose &mdash; Torchvision main documentation</title>
+  <title>Compose &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html b/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
index 47071cbb7f9..74354dd438d 100644
--- a/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
+++ b/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ConvertBoundingBoxFormat &mdash; Torchvision main documentation</title>
+  <title>ConvertBoundingBoxFormat &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html b/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
index 814faddc525..1a2409a3b72 100644
--- a/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
+++ b/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ConvertImageDtype &mdash; Torchvision main documentation</title>
+  <title>ConvertImageDtype &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.CutMix.html b/0.16/generated/torchvision.transforms.v2.CutMix.html
index a5cd872ad42..f7299eafa6f 100644
--- a/0.16/generated/torchvision.transforms.v2.CutMix.html
+++ b/0.16/generated/torchvision.transforms.v2.CutMix.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>CutMix &mdash; Torchvision main documentation</title>
+  <title>CutMix &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ElasticTransform.html b/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
index a1955501e34..ba787862cef 100644
--- a/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
+++ b/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ElasticTransform &mdash; Torchvision main documentation</title>
+  <title>ElasticTransform &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.FiveCrop.html b/0.16/generated/torchvision.transforms.v2.FiveCrop.html
index 7df16e39c8a..3fc0a6f7950 100644
--- a/0.16/generated/torchvision.transforms.v2.FiveCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.FiveCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FiveCrop &mdash; Torchvision main documentation</title>
+  <title>FiveCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.GaussianBlur.html b/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
index 2ff8e460d54..3c07aea9185 100644
--- a/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
+++ b/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>GaussianBlur &mdash; Torchvision main documentation</title>
+  <title>GaussianBlur &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.Grayscale.html b/0.16/generated/torchvision.transforms.v2.Grayscale.html
index 7ac1315ba89..43fa98362ce 100644
--- a/0.16/generated/torchvision.transforms.v2.Grayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.Grayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Grayscale &mdash; Torchvision main documentation</title>
+  <title>Grayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.Lambda.html b/0.16/generated/torchvision.transforms.v2.Lambda.html
index dc5014a37a0..b3a7290d4a5 100644
--- a/0.16/generated/torchvision.transforms.v2.Lambda.html
+++ b/0.16/generated/torchvision.transforms.v2.Lambda.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Lambda &mdash; Torchvision main documentation</title>
+  <title>Lambda &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.LinearTransformation.html b/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
index 492171e78f0..808a8c2b713 100644
--- a/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
+++ b/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>LinearTransformation &mdash; Torchvision main documentation</title>
+  <title>LinearTransformation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.MixUp.html b/0.16/generated/torchvision.transforms.v2.MixUp.html
index 4fc305e0fce..75242664689 100644
--- a/0.16/generated/torchvision.transforms.v2.MixUp.html
+++ b/0.16/generated/torchvision.transforms.v2.MixUp.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MixUp &mdash; Torchvision main documentation</title>
+  <title>MixUp &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.Normalize.html b/0.16/generated/torchvision.transforms.v2.Normalize.html
index 05f2c819ba3..8913d9fd486 100644
--- a/0.16/generated/torchvision.transforms.v2.Normalize.html
+++ b/0.16/generated/torchvision.transforms.v2.Normalize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Normalize &mdash; Torchvision main documentation</title>
+  <title>Normalize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.PILToTensor.html b/0.16/generated/torchvision.transforms.v2.PILToTensor.html
index 3ade4353499..772004bc8bc 100644
--- a/0.16/generated/torchvision.transforms.v2.PILToTensor.html
+++ b/0.16/generated/torchvision.transforms.v2.PILToTensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>PILToTensor &mdash; Torchvision main documentation</title>
+  <title>PILToTensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.Pad.html b/0.16/generated/torchvision.transforms.v2.Pad.html
index 7349e7c3565..b8c28aba5cb 100644
--- a/0.16/generated/torchvision.transforms.v2.Pad.html
+++ b/0.16/generated/torchvision.transforms.v2.Pad.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Pad &mdash; Torchvision main documentation</title>
+  <title>Pad &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandAugment.html b/0.16/generated/torchvision.transforms.v2.RandAugment.html
index 03c4d840cfe..c6f1254531e 100644
--- a/0.16/generated/torchvision.transforms.v2.RandAugment.html
+++ b/0.16/generated/torchvision.transforms.v2.RandAugment.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandAugment &mdash; Torchvision main documentation</title>
+  <title>RandAugment &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html b/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
index 3a8d1f471d6..c53f659d5c2 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomAdjustSharpness &mdash; Torchvision main documentation</title>
+  <title>RandomAdjustSharpness &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAffine.html b/0.16/generated/torchvision.transforms.v2.RandomAffine.html
index b351530718a..9894b83363a 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomAffine.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomAffine.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomAffine &mdash; Torchvision main documentation</title>
+  <title>RandomAffine &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomApply.html b/0.16/generated/torchvision.transforms.v2.RandomApply.html
index 870ddc483ce..d0c2fa417da 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomApply.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomApply.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomApply &mdash; Torchvision main documentation</title>
+  <title>RandomApply &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html b/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
index bfb126d1278..e930f5b0a24 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomAutocontrast &mdash; Torchvision main documentation</title>
+  <title>RandomAutocontrast &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html b/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
index 3fc07dc4e6c..2771a2c1e6e 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomChannelPermutation &mdash; Torchvision main documentation</title>
+  <title>RandomChannelPermutation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomChoice.html b/0.16/generated/torchvision.transforms.v2.RandomChoice.html
index d18eda41254..86fd43aed8a 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomChoice.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomChoice.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomChoice &mdash; Torchvision main documentation</title>
+  <title>RandomChoice &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomCrop.html b/0.16/generated/torchvision.transforms.v2.RandomCrop.html
index ac34512925e..0dccb5acaf4 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomCrop &mdash; Torchvision main documentation</title>
+  <title>RandomCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomEqualize.html b/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
index 5f7bb8d2c04..f380d7d7828 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomEqualize &mdash; Torchvision main documentation</title>
+  <title>RandomEqualize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomErasing.html b/0.16/generated/torchvision.transforms.v2.RandomErasing.html
index bcf717eadce..fd0823659a2 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomErasing.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomErasing.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomErasing &mdash; Torchvision main documentation</title>
+  <title>RandomErasing &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html b/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
index 9cabe2183fa..7dab1db0a2e 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomGrayscale &mdash; Torchvision main documentation</title>
+  <title>RandomGrayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html b/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
index 5e348c1f69d..1c6345c998e 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomHorizontalFlip &mdash; Torchvision main documentation</title>
+  <title>RandomHorizontalFlip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomInvert.html b/0.16/generated/torchvision.transforms.v2.RandomInvert.html
index 2b66b1ea44a..f86816a5a22 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomInvert.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomInvert.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomInvert &mdash; Torchvision main documentation</title>
+  <title>RandomInvert &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html b/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
index aa64bfc0a2a..c0a79c62e76 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomIoUCrop &mdash; Torchvision main documentation</title>
+  <title>RandomIoUCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomOrder.html b/0.16/generated/torchvision.transforms.v2.RandomOrder.html
index 0b1c0130f01..6ac5b838e05 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomOrder.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomOrder.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomOrder &mdash; Torchvision main documentation</title>
+  <title>RandomOrder &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPerspective.html b/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
index c37caf83bf6..71660112572 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomPerspective &mdash; Torchvision main documentation</title>
+  <title>RandomPerspective &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html b/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
index d2a9401ea50..ba75f996e7c 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomPhotometricDistort &mdash; Torchvision main documentation</title>
+  <title>RandomPhotometricDistort &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPosterize.html b/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
index 0d7a853fe0e..5a3541ce226 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomPosterize &mdash; Torchvision main documentation</title>
+  <title>RandomPosterize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomResize.html b/0.16/generated/torchvision.transforms.v2.RandomResize.html
index ee97e377a4f..e52095ef784 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomResize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomResize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomResize &mdash; Torchvision main documentation</title>
+  <title>RandomResize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html b/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
index 8c939d1d5f6..e7a53c0ead6 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomResizedCrop &mdash; Torchvision main documentation</title>
+  <title>RandomResizedCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomRotation.html b/0.16/generated/torchvision.transforms.v2.RandomRotation.html
index 727d781763b..773fc5496e8 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomRotation.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomRotation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomRotation &mdash; Torchvision main documentation</title>
+  <title>RandomRotation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html b/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
index f02e32122b3..1594851f7d5 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomShortestSize &mdash; Torchvision main documentation</title>
+  <title>RandomShortestSize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomSolarize.html b/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
index 38733e5d058..4df12f8e338 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomSolarize &mdash; Torchvision main documentation</title>
+  <title>RandomSolarize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html b/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
index bc5c42a8983..4d5ae885420 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomVerticalFlip &mdash; Torchvision main documentation</title>
+  <title>RandomVerticalFlip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html b/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
index 0b25d42e5a0..09a58ed6b5a 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RandomZoomOut &mdash; Torchvision main documentation</title>
+  <title>RandomZoomOut &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.Resize.html b/0.16/generated/torchvision.transforms.v2.Resize.html
index 079a2853e80..9b15667a83a 100644
--- a/0.16/generated/torchvision.transforms.v2.Resize.html
+++ b/0.16/generated/torchvision.transforms.v2.Resize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Resize &mdash; Torchvision main documentation</title>
+  <title>Resize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html b/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
index 5531bbe3ea7..d765a024e31 100644
--- a/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
+++ b/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SanitizeBoundingBoxes &mdash; Torchvision main documentation</title>
+  <title>SanitizeBoundingBoxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ScaleJitter.html b/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
index 290b4b6f567..1d36317d161 100644
--- a/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
+++ b/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ScaleJitter &mdash; Torchvision main documentation</title>
+  <title>ScaleJitter &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.TenCrop.html b/0.16/generated/torchvision.transforms.v2.TenCrop.html
index fd21fa0bda4..f330580bd18 100644
--- a/0.16/generated/torchvision.transforms.v2.TenCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.TenCrop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TenCrop &mdash; Torchvision main documentation</title>
+  <title>TenCrop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ToDtype.html b/0.16/generated/torchvision.transforms.v2.ToDtype.html
index 911bdf01e7f..d1a6b488609 100644
--- a/0.16/generated/torchvision.transforms.v2.ToDtype.html
+++ b/0.16/generated/torchvision.transforms.v2.ToDtype.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ToDtype &mdash; Torchvision main documentation</title>
+  <title>ToDtype &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ToImage.html b/0.16/generated/torchvision.transforms.v2.ToImage.html
index 728483fb4eb..2e5ddb7bb18 100644
--- a/0.16/generated/torchvision.transforms.v2.ToImage.html
+++ b/0.16/generated/torchvision.transforms.v2.ToImage.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ToImage &mdash; Torchvision main documentation</title>
+  <title>ToImage &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ToPILImage.html b/0.16/generated/torchvision.transforms.v2.ToPILImage.html
index a6716625373..adcd01e6f09 100644
--- a/0.16/generated/torchvision.transforms.v2.ToPILImage.html
+++ b/0.16/generated/torchvision.transforms.v2.ToPILImage.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ToPILImage &mdash; Torchvision main documentation</title>
+  <title>ToPILImage &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ToPureTensor.html b/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
index 292449f67af..6a12f6f807c 100644
--- a/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
+++ b/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ToPureTensor &mdash; Torchvision main documentation</title>
+  <title>ToPureTensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.ToTensor.html b/0.16/generated/torchvision.transforms.v2.ToTensor.html
index 08f955f02c0..eca883c502a 100644
--- a/0.16/generated/torchvision.transforms.v2.ToTensor.html
+++ b/0.16/generated/torchvision.transforms.v2.ToTensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ToTensor &mdash; Torchvision main documentation</title>
+  <title>ToTensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html b/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
index 71f44bda931..d699509629a 100644
--- a/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
+++ b/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TrivialAugmentWide &mdash; Torchvision main documentation</title>
+  <title>TrivialAugmentWide &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html b/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
index ecdbf967a48..b71a5c15974 100644
--- a/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
+++ b/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>UniformTemporalSubsample &mdash; Torchvision main documentation</title>
+  <title>UniformTemporalSubsample &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
index 7859ca63cc7..82419d15340 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_brightness &mdash; Torchvision main documentation</title>
+  <title>adjust_brightness &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
index a2cb45ac8a4..1be982a4c2d 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_contrast &mdash; Torchvision main documentation</title>
+  <title>adjust_contrast &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
index 170ad700e7a..aaf09c2df82 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_gamma &mdash; Torchvision main documentation</title>
+  <title>adjust_gamma &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
index f234ebdf1fb..d1711f1a442 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_hue &mdash; Torchvision main documentation</title>
+  <title>adjust_hue &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
index 44ab51d5d5d..848f8ee0569 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_saturation &mdash; Torchvision main documentation</title>
+  <title>adjust_saturation &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
index 586bfa373ad..e60944264e6 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>adjust_sharpness &mdash; Torchvision main documentation</title>
+  <title>adjust_sharpness &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.affine.html b/0.16/generated/torchvision.transforms.v2.functional.affine.html
index 3b6b663cbfc..8d2e009f59c 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.affine.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.affine.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>affine &mdash; Torchvision main documentation</title>
+  <title>affine &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html b/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
index 990ee955795..42a9c1d57e4 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>autocontrast &mdash; Torchvision main documentation</title>
+  <title>autocontrast &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.center_crop.html b/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
index b2eefc80731..a0ffff13860 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>center_crop &mdash; Torchvision main documentation</title>
+  <title>center_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html b/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
index b42462901c4..eb2c2cd0792 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>clamp_bounding_boxes &mdash; Torchvision main documentation</title>
+  <title>clamp_bounding_boxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html b/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
index 189c66b1c64..545ec23ba1f 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional.convert_bounding_box_format &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional.convert_bounding_box_format &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html b/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
index b6cd1a5c82a..0bc2089585a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>convert_image_dtype &mdash; Torchvision main documentation</title>
+  <title>convert_image_dtype &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.crop.html b/0.16/generated/torchvision.transforms.v2.functional.crop.html
index 2940942dbdc..94f90985120 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>crop &mdash; Torchvision main documentation</title>
+  <title>crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.elastic.html b/0.16/generated/torchvision.transforms.v2.functional.elastic.html
index b9f52f1ff56..3a71e0d0638 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.elastic.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.elastic.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>elastic &mdash; Torchvision main documentation</title>
+  <title>elastic &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.equalize.html b/0.16/generated/torchvision.transforms.v2.functional.equalize.html
index 41fec5b6988..a59f96135dd 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.equalize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.equalize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>equalize &mdash; Torchvision main documentation</title>
+  <title>equalize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.erase.html b/0.16/generated/torchvision.transforms.v2.functional.erase.html
index 2f7b0e04709..53403d48730 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.erase.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.erase.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>erase &mdash; Torchvision main documentation</title>
+  <title>erase &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.five_crop.html b/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
index 697024aa2ae..8ee38c909fc 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>five_crop &mdash; Torchvision main documentation</title>
+  <title>five_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html b/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
index ce2ba4b8b12..87bc6de2aad 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>gaussian_blur &mdash; Torchvision main documentation</title>
+  <title>gaussian_blur &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html b/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
index 01d6ff4a5ed..a91a765c6e9 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>horizontal_flip &mdash; Torchvision main documentation</title>
+  <title>horizontal_flip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.invert.html b/0.16/generated/torchvision.transforms.v2.functional.invert.html
index 932eaba8d19..063485348ef 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.invert.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.invert.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>invert &mdash; Torchvision main documentation</title>
+  <title>invert &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.normalize.html b/0.16/generated/torchvision.transforms.v2.functional.normalize.html
index f14bab49ef6..e2c04bb0036 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.normalize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.normalize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>normalize &mdash; Torchvision main documentation</title>
+  <title>normalize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.pad.html b/0.16/generated/torchvision.transforms.v2.functional.pad.html
index cab9a1996d7..63b76e53b2e 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.pad.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.pad.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>pad &mdash; Torchvision main documentation</title>
+  <title>pad &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html b/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
index 2e9ac522184..f8f39cbd603 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>permute_channels &mdash; Torchvision main documentation</title>
+  <title>permute_channels &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.perspective.html b/0.16/generated/torchvision.transforms.v2.functional.perspective.html
index 1d533b775df..ce2df88a957 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.perspective.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.perspective.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>perspective &mdash; Torchvision main documentation</title>
+  <title>perspective &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html b/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
index 05e05aafad1..97d240f7071 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional.pil_to_tensor &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional.pil_to_tensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.posterize.html b/0.16/generated/torchvision.transforms.v2.functional.posterize.html
index 2243f1a40bd..2c84d20b182 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.posterize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.posterize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>posterize &mdash; Torchvision main documentation</title>
+  <title>posterize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html b/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
index 85871e7d60e..b6dab73e388 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>register_kernel &mdash; Torchvision main documentation</title>
+  <title>register_kernel &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.resize.html b/0.16/generated/torchvision.transforms.v2.functional.resize.html
index 29f03b5cc46..b1d116abdbf 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.resize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.resize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resize &mdash; Torchvision main documentation</title>
+  <title>resize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html b/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
index cea28709089..1ff98ba562a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resized_crop &mdash; Torchvision main documentation</title>
+  <title>resized_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html b/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
index cef77c76cbd..7cd6d943f21 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>rgb_to_grayscale &mdash; Torchvision main documentation</title>
+  <title>rgb_to_grayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.rotate.html b/0.16/generated/torchvision.transforms.v2.functional.rotate.html
index b5c02f231c0..c6c4f642424 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.rotate.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.rotate.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>rotate &mdash; Torchvision main documentation</title>
+  <title>rotate &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.solarize.html b/0.16/generated/torchvision.transforms.v2.functional.solarize.html
index 2fd0c34a963..0d5de7c198c 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.solarize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.solarize.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>solarize &mdash; Torchvision main documentation</title>
+  <title>solarize &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html b/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
index dcab89504de..2af6b66a27c 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ten_crop &mdash; Torchvision main documentation</title>
+  <title>ten_crop &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html b/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
index 54e59455168..b40acca4531 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional.to_dtype &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional.to_dtype &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html b/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
index 81f198bda7d..396f44d036e 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>to_grayscale &mdash; Torchvision main documentation</title>
+  <title>to_grayscale &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_image.html b/0.16/generated/torchvision.transforms.v2.functional.to_image.html
index 217985c24ea..5037ea2a35a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_image.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional.to_image &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional.to_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html b/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
index 369f49d57ef..c14952f489b 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision.transforms.v2.functional.to_pil_image &mdash; Torchvision main documentation</title>
+  <title>torchvision.transforms.v2.functional.to_pil_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html b/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
index 6a2705e0309..e089ac34e8f 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>to_tensor &mdash; Torchvision main documentation</title>
+  <title>to_tensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html b/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
index 17ffddc5938..26062711fbc 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>uniform_temporal_subsample &mdash; Torchvision main documentation</title>
+  <title>uniform_temporal_subsample &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html b/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
index d170a1cfb16..9bec53a3497 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vertical_flip &mdash; Torchvision main documentation</title>
+  <title>vertical_flip &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html b/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
index efb56f0bc27..e3ce955c129 100644
--- a/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
+++ b/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>BoundingBoxFormat &mdash; Torchvision main documentation</title>
+  <title>BoundingBoxFormat &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html b/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
index 7992cdd5052..0292fcccff3 100644
--- a/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
+++ b/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>BoundingBoxes &mdash; Torchvision main documentation</title>
+  <title>BoundingBoxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.Image.html b/0.16/generated/torchvision.tv_tensors.Image.html
index 279dbef4116..72b5f0e18c3 100644
--- a/0.16/generated/torchvision.tv_tensors.Image.html
+++ b/0.16/generated/torchvision.tv_tensors.Image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Image &mdash; Torchvision main documentation</title>
+  <title>Image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.Mask.html b/0.16/generated/torchvision.tv_tensors.Mask.html
index b13e11e12d4..d77942ff318 100644
--- a/0.16/generated/torchvision.tv_tensors.Mask.html
+++ b/0.16/generated/torchvision.tv_tensors.Mask.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Mask &mdash; Torchvision main documentation</title>
+  <title>Mask &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.TVTensor.html b/0.16/generated/torchvision.tv_tensors.TVTensor.html
index f465e89d2f1..9a7f2fda18c 100644
--- a/0.16/generated/torchvision.tv_tensors.TVTensor.html
+++ b/0.16/generated/torchvision.tv_tensors.TVTensor.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TVTensor &mdash; Torchvision main documentation</title>
+  <title>TVTensor &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.Video.html b/0.16/generated/torchvision.tv_tensors.Video.html
index ae00fafdf10..58e6348c3be 100644
--- a/0.16/generated/torchvision.tv_tensors.Video.html
+++ b/0.16/generated/torchvision.tv_tensors.Video.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Video &mdash; Torchvision main documentation</title>
+  <title>Video &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.set_return_type.html b/0.16/generated/torchvision.tv_tensors.set_return_type.html
index b78e233b3cc..920c6baf63f 100644
--- a/0.16/generated/torchvision.tv_tensors.set_return_type.html
+++ b/0.16/generated/torchvision.tv_tensors.set_return_type.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>set_return_type &mdash; Torchvision main documentation</title>
+  <title>set_return_type &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.tv_tensors.wrap.html b/0.16/generated/torchvision.tv_tensors.wrap.html
index e6975091ef7..6615945fde9 100644
--- a/0.16/generated/torchvision.tv_tensors.wrap.html
+++ b/0.16/generated/torchvision.tv_tensors.wrap.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>wrap &mdash; Torchvision main documentation</title>
+  <title>wrap &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.utils.draw_bounding_boxes.html b/0.16/generated/torchvision.utils.draw_bounding_boxes.html
index dc764271ab8..8af44a05dd3 100644
--- a/0.16/generated/torchvision.utils.draw_bounding_boxes.html
+++ b/0.16/generated/torchvision.utils.draw_bounding_boxes.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>draw_bounding_boxes &mdash; Torchvision main documentation</title>
+  <title>draw_bounding_boxes &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.utils.draw_keypoints.html b/0.16/generated/torchvision.utils.draw_keypoints.html
index b448e653428..16b385699a1 100644
--- a/0.16/generated/torchvision.utils.draw_keypoints.html
+++ b/0.16/generated/torchvision.utils.draw_keypoints.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>draw_keypoints &mdash; Torchvision main documentation</title>
+  <title>draw_keypoints &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.utils.draw_segmentation_masks.html b/0.16/generated/torchvision.utils.draw_segmentation_masks.html
index d48f6c49d7d..dafdf93bfd7 100644
--- a/0.16/generated/torchvision.utils.draw_segmentation_masks.html
+++ b/0.16/generated/torchvision.utils.draw_segmentation_masks.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>draw_segmentation_masks &mdash; Torchvision main documentation</title>
+  <title>draw_segmentation_masks &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.utils.flow_to_image.html b/0.16/generated/torchvision.utils.flow_to_image.html
index 1bb6cdb719e..bca6fec3764 100644
--- a/0.16/generated/torchvision.utils.flow_to_image.html
+++ b/0.16/generated/torchvision.utils.flow_to_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>flow_to_image &mdash; Torchvision main documentation</title>
+  <title>flow_to_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.utils.make_grid.html b/0.16/generated/torchvision.utils.make_grid.html
index 57af3cee14a..53a3a1fb966 100644
--- a/0.16/generated/torchvision.utils.make_grid.html
+++ b/0.16/generated/torchvision.utils.make_grid.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>make_grid &mdash; Torchvision main documentation</title>
+  <title>make_grid &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/torchvision.utils.save_image.html b/0.16/generated/torchvision.utils.save_image.html
index 644b9b4d920..d26918d1049 100644
--- a/0.16/generated/torchvision.utils.save_image.html
+++ b/0.16/generated/torchvision.utils.save_image.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>save_image &mdash; Torchvision main documentation</title>
+  <title>save_image &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/generated/video_table.html b/0.16/generated/video_table.html
index e7b9a8ddb0c..8369937e8ce 100644
--- a/0.16/generated/video_table.html
+++ b/0.16/generated/video_table.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>&lt;no title&gt; &mdash; Torchvision main documentation</title>
+  <title>&lt;no title&gt; &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/genindex.html b/0.16/genindex.html
index 35a609885af..e5ba553d1d7 100644
--- a/0.16/genindex.html
+++ b/0.16/genindex.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Index &mdash; Torchvision main documentation</title>
+  <title>Index &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/index.html b/0.16/index.html
index 6258bb7771e..9138f681381 100644
--- a/0.16/index.html
+++ b/0.16/index.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>torchvision &mdash; Torchvision main documentation</title>
+  <title>torchvision &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/io.html b/0.16/io.html
index 28c27092f28..d4524e27d17 100644
--- a/0.16/io.html
+++ b/0.16/io.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Decoding / Encoding images and videos &mdash; Torchvision main documentation</title>
+  <title>Decoding / Encoding images and videos &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models.html b/0.16/models.html
index 0b3afffef3e..0c27afadb28 100644
--- a/0.16/models.html
+++ b/0.16/models.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Models and pre-trained weights &mdash; Torchvision main documentation</title>
+  <title>Models and pre-trained weights &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/alexnet.html b/0.16/models/alexnet.html
index 5d602c7e172..a7edec8a02c 100644
--- a/0.16/models/alexnet.html
+++ b/0.16/models/alexnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>AlexNet &mdash; Torchvision main documentation</title>
+  <title>AlexNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/convnext.html b/0.16/models/convnext.html
index d238ff0ed70..13e2700db0c 100644
--- a/0.16/models/convnext.html
+++ b/0.16/models/convnext.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ConvNeXt &mdash; Torchvision main documentation</title>
+  <title>ConvNeXt &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/deeplabv3.html b/0.16/models/deeplabv3.html
index d53b85c6014..381916ef56a 100644
--- a/0.16/models/deeplabv3.html
+++ b/0.16/models/deeplabv3.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>DeepLabV3 &mdash; Torchvision main documentation</title>
+  <title>DeepLabV3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/densenet.html b/0.16/models/densenet.html
index 70322ce77cd..e9cb95b32e6 100644
--- a/0.16/models/densenet.html
+++ b/0.16/models/densenet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>DenseNet &mdash; Torchvision main documentation</title>
+  <title>DenseNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/efficientnet.html b/0.16/models/efficientnet.html
index f23391b0fa2..b3c36bc93fa 100644
--- a/0.16/models/efficientnet.html
+++ b/0.16/models/efficientnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>EfficientNet &mdash; Torchvision main documentation</title>
+  <title>EfficientNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/efficientnetv2.html b/0.16/models/efficientnetv2.html
index 9965b412435..45fab004c55 100644
--- a/0.16/models/efficientnetv2.html
+++ b/0.16/models/efficientnetv2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>EfficientNetV2 &mdash; Torchvision main documentation</title>
+  <title>EfficientNetV2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/faster_rcnn.html b/0.16/models/faster_rcnn.html
index a88b8386bd9..f639a595dcd 100644
--- a/0.16/models/faster_rcnn.html
+++ b/0.16/models/faster_rcnn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Faster R-CNN &mdash; Torchvision main documentation</title>
+  <title>Faster R-CNN &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/fcn.html b/0.16/models/fcn.html
index ad6f73656a8..d0f5024e94c 100644
--- a/0.16/models/fcn.html
+++ b/0.16/models/fcn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FCN &mdash; Torchvision main documentation</title>
+  <title>FCN &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/fcos.html b/0.16/models/fcos.html
index f4aa37a20db..5cbcb7827ac 100644
--- a/0.16/models/fcos.html
+++ b/0.16/models/fcos.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>FCOS &mdash; Torchvision main documentation</title>
+  <title>FCOS &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.alexnet.html b/0.16/models/generated/torchvision.models.alexnet.html
index 76a8c2df1ae..7cfdd307271 100644
--- a/0.16/models/generated/torchvision.models.alexnet.html
+++ b/0.16/models/generated/torchvision.models.alexnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>alexnet &mdash; Torchvision main documentation</title>
+  <title>alexnet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.convnext_base.html b/0.16/models/generated/torchvision.models.convnext_base.html
index 3c975ed1841..784110af317 100644
--- a/0.16/models/generated/torchvision.models.convnext_base.html
+++ b/0.16/models/generated/torchvision.models.convnext_base.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>convnext_base &mdash; Torchvision main documentation</title>
+  <title>convnext_base &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.convnext_large.html b/0.16/models/generated/torchvision.models.convnext_large.html
index 938d36696b5..7a4d4dbb2d9 100644
--- a/0.16/models/generated/torchvision.models.convnext_large.html
+++ b/0.16/models/generated/torchvision.models.convnext_large.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>convnext_large &mdash; Torchvision main documentation</title>
+  <title>convnext_large &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.convnext_small.html b/0.16/models/generated/torchvision.models.convnext_small.html
index ae0add8b00a..de8ae26149a 100644
--- a/0.16/models/generated/torchvision.models.convnext_small.html
+++ b/0.16/models/generated/torchvision.models.convnext_small.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>convnext_small &mdash; Torchvision main documentation</title>
+  <title>convnext_small &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.convnext_tiny.html b/0.16/models/generated/torchvision.models.convnext_tiny.html
index 4bc21bca1e6..c644597c772 100644
--- a/0.16/models/generated/torchvision.models.convnext_tiny.html
+++ b/0.16/models/generated/torchvision.models.convnext_tiny.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>convnext_tiny &mdash; Torchvision main documentation</title>
+  <title>convnext_tiny &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.densenet121.html b/0.16/models/generated/torchvision.models.densenet121.html
index 98d04556aab..bb1b1157eeb 100644
--- a/0.16/models/generated/torchvision.models.densenet121.html
+++ b/0.16/models/generated/torchvision.models.densenet121.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>densenet121 &mdash; Torchvision main documentation</title>
+  <title>densenet121 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.densenet161.html b/0.16/models/generated/torchvision.models.densenet161.html
index 9c77e1e6447..246093881ec 100644
--- a/0.16/models/generated/torchvision.models.densenet161.html
+++ b/0.16/models/generated/torchvision.models.densenet161.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>densenet161 &mdash; Torchvision main documentation</title>
+  <title>densenet161 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.densenet169.html b/0.16/models/generated/torchvision.models.densenet169.html
index 5e0dab7049f..e6ec34243a2 100644
--- a/0.16/models/generated/torchvision.models.densenet169.html
+++ b/0.16/models/generated/torchvision.models.densenet169.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>densenet169 &mdash; Torchvision main documentation</title>
+  <title>densenet169 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.densenet201.html b/0.16/models/generated/torchvision.models.densenet201.html
index 53bbc9b508c..0c032a38397 100644
--- a/0.16/models/generated/torchvision.models.densenet201.html
+++ b/0.16/models/generated/torchvision.models.densenet201.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>densenet201 &mdash; Torchvision main documentation</title>
+  <title>densenet201 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
index 068b5d0724d..c8c8f8df827 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>fasterrcnn_mobilenet_v3_large_320_fpn &mdash; Torchvision main documentation</title>
+  <title>fasterrcnn_mobilenet_v3_large_320_fpn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
index 5ddc71f1fa3..2a6a6ed1202 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>fasterrcnn_mobilenet_v3_large_fpn &mdash; Torchvision main documentation</title>
+  <title>fasterrcnn_mobilenet_v3_large_fpn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
index a373428e761..6eac0a7455d 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>fasterrcnn_resnet50_fpn &mdash; Torchvision main documentation</title>
+  <title>fasterrcnn_resnet50_fpn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
index fcbe7a58b4e..a544f5eb65d 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>fasterrcnn_resnet50_fpn_v2 &mdash; Torchvision main documentation</title>
+  <title>fasterrcnn_resnet50_fpn_v2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
index 3a6f140e51c..0348483dff2 100644
--- a/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>fcos_resnet50_fpn &mdash; Torchvision main documentation</title>
+  <title>fcos_resnet50_fpn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
index 8f2e1c0c9cb..814d0ae1a59 100644
--- a/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>keypointrcnn_resnet50_fpn &mdash; Torchvision main documentation</title>
+  <title>keypointrcnn_resnet50_fpn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
index 3e5638191e2..780fae52150 100644
--- a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>maskrcnn_resnet50_fpn &mdash; Torchvision main documentation</title>
+  <title>maskrcnn_resnet50_fpn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
index 6d72b182f1d..43443ca242d 100644
--- a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
+++ b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>maskrcnn_resnet50_fpn_v2 &mdash; Torchvision main documentation</title>
+  <title>maskrcnn_resnet50_fpn_v2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
index 60d1c1c6e58..9b14ae8937a 100644
--- a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>retinanet_resnet50_fpn &mdash; Torchvision main documentation</title>
+  <title>retinanet_resnet50_fpn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
index bbaba0410c1..7d4d6211050 100644
--- a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
+++ b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>retinanet_resnet50_fpn_v2 &mdash; Torchvision main documentation</title>
+  <title>retinanet_resnet50_fpn_v2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html b/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
index 92c7ca87fe2..9462093e3dd 100644
--- a/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
+++ b/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ssd300_vgg16 &mdash; Torchvision main documentation</title>
+  <title>ssd300_vgg16 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
index 825217fb5a3..2d0fce77d52 100644
--- a/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ssdlite320_mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  <title>ssdlite320_mobilenet_v3_large &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b0.html b/0.16/models/generated/torchvision.models.efficientnet_b0.html
index 23592fe4489..9ec12afd016 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b0.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b0.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b0 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b0 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b1.html b/0.16/models/generated/torchvision.models.efficientnet_b1.html
index 972db29db91..77e28fdf533 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b1.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b1.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b1 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b1 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b2.html b/0.16/models/generated/torchvision.models.efficientnet_b2.html
index cb35698d450..e22abde1c4c 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b2.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b2 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b3.html b/0.16/models/generated/torchvision.models.efficientnet_b3.html
index 9ac0385fcfd..5bfabbf8936 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b3.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b3.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b3 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b4.html b/0.16/models/generated/torchvision.models.efficientnet_b4.html
index 3128de93ce4..90e29f85ebc 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b4.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b4.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b4 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b4 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b5.html b/0.16/models/generated/torchvision.models.efficientnet_b5.html
index 8107162669e..aa74d2068f5 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b5.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b5.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b5 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b5 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b6.html b/0.16/models/generated/torchvision.models.efficientnet_b6.html
index 5d7d2cbd7c0..49a6cbe4c24 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b6.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b6.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b6 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b6 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b7.html b/0.16/models/generated/torchvision.models.efficientnet_b7.html
index ac224dd95fb..d73e186511a 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b7.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b7.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_b7 &mdash; Torchvision main documentation</title>
+  <title>efficientnet_b7 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_l.html b/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
index 57376ca9987..0216545b44d 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_v2_l &mdash; Torchvision main documentation</title>
+  <title>efficientnet_v2_l &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_m.html b/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
index 4483f561da9..5d6a14d6d37 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_v2_m &mdash; Torchvision main documentation</title>
+  <title>efficientnet_v2_m &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_s.html b/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
index 07cc266ad41..e503cabec46 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>efficientnet_v2_s &mdash; Torchvision main documentation</title>
+  <title>efficientnet_v2_s &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.googlenet.html b/0.16/models/generated/torchvision.models.googlenet.html
index 2c715dcf2c7..47de7f67533 100644
--- a/0.16/models/generated/torchvision.models.googlenet.html
+++ b/0.16/models/generated/torchvision.models.googlenet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>googlenet &mdash; Torchvision main documentation</title>
+  <title>googlenet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.inception_v3.html b/0.16/models/generated/torchvision.models.inception_v3.html
index bb8ff078c27..4f08b38ff3b 100644
--- a/0.16/models/generated/torchvision.models.inception_v3.html
+++ b/0.16/models/generated/torchvision.models.inception_v3.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>inception_v3 &mdash; Torchvision main documentation</title>
+  <title>inception_v3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.maxvit_t.html b/0.16/models/generated/torchvision.models.maxvit_t.html
index 2a4c2b7f870..4f4c297aebb 100644
--- a/0.16/models/generated/torchvision.models.maxvit_t.html
+++ b/0.16/models/generated/torchvision.models.maxvit_t.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>maxvit_t &mdash; Torchvision main documentation</title>
+  <title>maxvit_t &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.mnasnet0_5.html b/0.16/models/generated/torchvision.models.mnasnet0_5.html
index a4a411891b1..7f6f03d6fd5 100644
--- a/0.16/models/generated/torchvision.models.mnasnet0_5.html
+++ b/0.16/models/generated/torchvision.models.mnasnet0_5.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mnasnet0_5 &mdash; Torchvision main documentation</title>
+  <title>mnasnet0_5 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.mnasnet0_75.html b/0.16/models/generated/torchvision.models.mnasnet0_75.html
index fc1a3e7ee27..2bcd4471f50 100644
--- a/0.16/models/generated/torchvision.models.mnasnet0_75.html
+++ b/0.16/models/generated/torchvision.models.mnasnet0_75.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mnasnet0_75 &mdash; Torchvision main documentation</title>
+  <title>mnasnet0_75 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.mnasnet1_0.html b/0.16/models/generated/torchvision.models.mnasnet1_0.html
index c5807c9abd4..6a2e10bddc5 100644
--- a/0.16/models/generated/torchvision.models.mnasnet1_0.html
+++ b/0.16/models/generated/torchvision.models.mnasnet1_0.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mnasnet1_0 &mdash; Torchvision main documentation</title>
+  <title>mnasnet1_0 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.mnasnet1_3.html b/0.16/models/generated/torchvision.models.mnasnet1_3.html
index ced32261751..8133f52f9f7 100644
--- a/0.16/models/generated/torchvision.models.mnasnet1_3.html
+++ b/0.16/models/generated/torchvision.models.mnasnet1_3.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mnasnet1_3 &mdash; Torchvision main documentation</title>
+  <title>mnasnet1_3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v2.html b/0.16/models/generated/torchvision.models.mobilenet_v2.html
index f187953bdc8..6fb7e93c69f 100644
--- a/0.16/models/generated/torchvision.models.mobilenet_v2.html
+++ b/0.16/models/generated/torchvision.models.mobilenet_v2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mobilenet_v2 &mdash; Torchvision main documentation</title>
+  <title>mobilenet_v2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
index b0fdaa37a7f..e0b4fe38460 100644
--- a/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  <title>mobilenet_v3_large &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v3_small.html b/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
index 99e8584f00d..cddd4549433 100644
--- a/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
+++ b/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mobilenet_v3_small &mdash; Torchvision main documentation</title>
+  <title>mobilenet_v3_small &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.optical_flow.raft_large.html b/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
index e3d39568419..186b3f02ba7 100644
--- a/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
+++ b/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>raft_large &mdash; Torchvision main documentation</title>
+  <title>raft_large &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.optical_flow.raft_small.html b/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
index 6ffd1339bf3..f1e1e3a8a40 100644
--- a/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
+++ b/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>raft_small &mdash; Torchvision main documentation</title>
+  <title>raft_small &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.googlenet.html b/0.16/models/generated/torchvision.models.quantization.googlenet.html
index 27ec5b21bb0..8efc2459fc9 100644
--- a/0.16/models/generated/torchvision.models.quantization.googlenet.html
+++ b/0.16/models/generated/torchvision.models.quantization.googlenet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>googlenet &mdash; Torchvision main documentation</title>
+  <title>googlenet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.inception_v3.html b/0.16/models/generated/torchvision.models.quantization.inception_v3.html
index e72a5df7209..77f16251fdf 100644
--- a/0.16/models/generated/torchvision.models.quantization.inception_v3.html
+++ b/0.16/models/generated/torchvision.models.quantization.inception_v3.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>inception_v3 &mdash; Torchvision main documentation</title>
+  <title>inception_v3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html b/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
index 77d309d6117..d3b55b0b5ce 100644
--- a/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
+++ b/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mobilenet_v2 &mdash; Torchvision main documentation</title>
+  <title>mobilenet_v2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
index 85e4abaa3d2..6f7750a68e6 100644
--- a/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  <title>mobilenet_v3_large &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.resnet18.html b/0.16/models/generated/torchvision.models.quantization.resnet18.html
index e84c3ffebc4..880a9f84ac3 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnet18.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnet18.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnet18 &mdash; Torchvision main documentation</title>
+  <title>resnet18 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.resnet50.html b/0.16/models/generated/torchvision.models.quantization.resnet50.html
index 6c0a140a458..d98a86f82dd 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnet50.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnet50.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnet50 &mdash; Torchvision main documentation</title>
+  <title>resnet50 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html b/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
index 8ac2c42fb34..73556f7b371 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnext101_32x8d &mdash; Torchvision main documentation</title>
+  <title>resnext101_32x8d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html b/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
index 50835f1de89..0ccb4f69bb9 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnext101_64x4d &mdash; Torchvision main documentation</title>
+  <title>resnext101_64x4d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
index b9a2fdbc0e9..1d8cc82e255 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x0_5 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x0_5 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
index b8c7debc0a2..c090a6b96d9 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x1_0 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x1_0 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
index bbd2efdf839..d65ef816df3 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x1_5 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x1_5 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
index 8077f35d2f0..8def6dfd7dd 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x2_0 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x2_0 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_x_16gf.html b/0.16/models/generated/torchvision.models.regnet_x_16gf.html
index 1a6649d79d0..5974897e5ca 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_16gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_16gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_x_16gf &mdash; Torchvision main documentation</title>
+  <title>regnet_x_16gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html b/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
index fcddd6c99cf..669988d37c6 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_x_1_6gf &mdash; Torchvision main documentation</title>
+  <title>regnet_x_1_6gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_x_32gf.html b/0.16/models/generated/torchvision.models.regnet_x_32gf.html
index e2dd42cd1ab..b230001b8e0 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_32gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_32gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_x_32gf &mdash; Torchvision main documentation</title>
+  <title>regnet_x_32gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html b/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
index 0262ffd235b..a6ac0644093 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_x_3_2gf &mdash; Torchvision main documentation</title>
+  <title>regnet_x_3_2gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_x_400mf.html b/0.16/models/generated/torchvision.models.regnet_x_400mf.html
index 14e43d5fb69..135864ebb30 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_400mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_400mf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_x_400mf &mdash; Torchvision main documentation</title>
+  <title>regnet_x_400mf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_x_800mf.html b/0.16/models/generated/torchvision.models.regnet_x_800mf.html
index d67962b5a8a..40fb5985662 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_800mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_800mf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_x_800mf &mdash; Torchvision main documentation</title>
+  <title>regnet_x_800mf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_x_8gf.html b/0.16/models/generated/torchvision.models.regnet_x_8gf.html
index 2b014723b09..a8dab23d246 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_8gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_8gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_x_8gf &mdash; Torchvision main documentation</title>
+  <title>regnet_x_8gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_128gf.html b/0.16/models/generated/torchvision.models.regnet_y_128gf.html
index f8c266a2042..194ccb553c0 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_128gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_128gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_128gf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_128gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_16gf.html b/0.16/models/generated/torchvision.models.regnet_y_16gf.html
index fc7ce99eedf..161019b716c 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_16gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_16gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_16gf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_16gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html b/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
index 7b3d1eac693..f8cc39d8bff 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_1_6gf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_1_6gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_32gf.html b/0.16/models/generated/torchvision.models.regnet_y_32gf.html
index d6bf6d78a1f..482d52dba39 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_32gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_32gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_32gf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_32gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html b/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
index 12a56a574f7..226b13a947e 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_3_2gf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_3_2gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_400mf.html b/0.16/models/generated/torchvision.models.regnet_y_400mf.html
index 2d21017007f..716394ec8d3 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_400mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_400mf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_400mf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_400mf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_800mf.html b/0.16/models/generated/torchvision.models.regnet_y_800mf.html
index 992e9a5cdf8..eaa1af855b8 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_800mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_800mf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_800mf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_800mf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.regnet_y_8gf.html b/0.16/models/generated/torchvision.models.regnet_y_8gf.html
index 1e7cd5f62a7..dc4161be285 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_8gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_8gf.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>regnet_y_8gf &mdash; Torchvision main documentation</title>
+  <title>regnet_y_8gf &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnet101.html b/0.16/models/generated/torchvision.models.resnet101.html
index 1605c98fd62..df7f0e01237 100644
--- a/0.16/models/generated/torchvision.models.resnet101.html
+++ b/0.16/models/generated/torchvision.models.resnet101.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnet101 &mdash; Torchvision main documentation</title>
+  <title>resnet101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnet152.html b/0.16/models/generated/torchvision.models.resnet152.html
index 783b08e7fe5..53b474e7b59 100644
--- a/0.16/models/generated/torchvision.models.resnet152.html
+++ b/0.16/models/generated/torchvision.models.resnet152.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnet152 &mdash; Torchvision main documentation</title>
+  <title>resnet152 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnet18.html b/0.16/models/generated/torchvision.models.resnet18.html
index 714f107e86b..95470b49960 100644
--- a/0.16/models/generated/torchvision.models.resnet18.html
+++ b/0.16/models/generated/torchvision.models.resnet18.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnet18 &mdash; Torchvision main documentation</title>
+  <title>resnet18 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnet34.html b/0.16/models/generated/torchvision.models.resnet34.html
index 621443eae1f..17363719208 100644
--- a/0.16/models/generated/torchvision.models.resnet34.html
+++ b/0.16/models/generated/torchvision.models.resnet34.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnet34 &mdash; Torchvision main documentation</title>
+  <title>resnet34 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnet50.html b/0.16/models/generated/torchvision.models.resnet50.html
index e7da239fd34..e3a42f1830b 100644
--- a/0.16/models/generated/torchvision.models.resnet50.html
+++ b/0.16/models/generated/torchvision.models.resnet50.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnet50 &mdash; Torchvision main documentation</title>
+  <title>resnet50 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnext101_32x8d.html b/0.16/models/generated/torchvision.models.resnext101_32x8d.html
index 049c92567ab..bfc941c429d 100644
--- a/0.16/models/generated/torchvision.models.resnext101_32x8d.html
+++ b/0.16/models/generated/torchvision.models.resnext101_32x8d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnext101_32x8d &mdash; Torchvision main documentation</title>
+  <title>resnext101_32x8d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnext101_64x4d.html b/0.16/models/generated/torchvision.models.resnext101_64x4d.html
index 1da793d34f9..78fd699494e 100644
--- a/0.16/models/generated/torchvision.models.resnext101_64x4d.html
+++ b/0.16/models/generated/torchvision.models.resnext101_64x4d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnext101_64x4d &mdash; Torchvision main documentation</title>
+  <title>resnext101_64x4d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.resnext50_32x4d.html b/0.16/models/generated/torchvision.models.resnext50_32x4d.html
index 163a1d87940..2ede22a0e2a 100644
--- a/0.16/models/generated/torchvision.models.resnext50_32x4d.html
+++ b/0.16/models/generated/torchvision.models.resnext50_32x4d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>resnext50_32x4d &mdash; Torchvision main documentation</title>
+  <title>resnext50_32x4d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
index 9e2e3cc36fd..e16a18392db 100644
--- a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>deeplabv3_mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  <title>deeplabv3_mobilenet_v3_large &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
index c7cccfbab5d..b821f6f3170 100644
--- a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>deeplabv3_resnet101 &mdash; Torchvision main documentation</title>
+  <title>deeplabv3_resnet101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
index 24c8888bca5..8f33be90f5a 100644
--- a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>deeplabv3_resnet50 &mdash; Torchvision main documentation</title>
+  <title>deeplabv3_resnet50 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
index 9d7f23bc5ea..d9f3edb0305 100644
--- a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
+++ b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>fcn_resnet101 &mdash; Torchvision main documentation</title>
+  <title>fcn_resnet101 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
index 887a8e212c2..f6afd02b1b5 100644
--- a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
+++ b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>fcn_resnet50 &mdash; Torchvision main documentation</title>
+  <title>fcn_resnet50 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
index 5f4744e58ca..3819f3de113 100644
--- a/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>lraspp_mobilenet_v3_large &mdash; Torchvision main documentation</title>
+  <title>lraspp_mobilenet_v3_large &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
index 56ee4b0ad5f..c3bd8fbe5b3 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x0_5 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x0_5 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
index 6f4a0a21cf2..af710e7675d 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x1_0 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x1_0 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
index d17234956a9..8da3a711130 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x1_5 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x1_5 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
index 542a6679cac..490f8a69151 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>shufflenet_v2_x2_0 &mdash; Torchvision main documentation</title>
+  <title>shufflenet_v2_x2_0 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.squeezenet1_0.html b/0.16/models/generated/torchvision.models.squeezenet1_0.html
index 2c76b3d6dda..c67aeae745b 100644
--- a/0.16/models/generated/torchvision.models.squeezenet1_0.html
+++ b/0.16/models/generated/torchvision.models.squeezenet1_0.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>squeezenet1_0 &mdash; Torchvision main documentation</title>
+  <title>squeezenet1_0 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.squeezenet1_1.html b/0.16/models/generated/torchvision.models.squeezenet1_1.html
index 638678b463e..7667930155e 100644
--- a/0.16/models/generated/torchvision.models.squeezenet1_1.html
+++ b/0.16/models/generated/torchvision.models.squeezenet1_1.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>squeezenet1_1 &mdash; Torchvision main documentation</title>
+  <title>squeezenet1_1 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.swin_b.html b/0.16/models/generated/torchvision.models.swin_b.html
index 6e02233596f..47c0e8c7a75 100644
--- a/0.16/models/generated/torchvision.models.swin_b.html
+++ b/0.16/models/generated/torchvision.models.swin_b.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin_b &mdash; Torchvision main documentation</title>
+  <title>swin_b &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.swin_s.html b/0.16/models/generated/torchvision.models.swin_s.html
index 500454c5310..3b27bd73176 100644
--- a/0.16/models/generated/torchvision.models.swin_s.html
+++ b/0.16/models/generated/torchvision.models.swin_s.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin_s &mdash; Torchvision main documentation</title>
+  <title>swin_s &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.swin_t.html b/0.16/models/generated/torchvision.models.swin_t.html
index b47607277f5..2d0f3e1dfd1 100644
--- a/0.16/models/generated/torchvision.models.swin_t.html
+++ b/0.16/models/generated/torchvision.models.swin_t.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin_t &mdash; Torchvision main documentation</title>
+  <title>swin_t &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.swin_v2_b.html b/0.16/models/generated/torchvision.models.swin_v2_b.html
index 75525d2e552..3f12f321553 100644
--- a/0.16/models/generated/torchvision.models.swin_v2_b.html
+++ b/0.16/models/generated/torchvision.models.swin_v2_b.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin_v2_b &mdash; Torchvision main documentation</title>
+  <title>swin_v2_b &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.swin_v2_s.html b/0.16/models/generated/torchvision.models.swin_v2_s.html
index 6851291c2c7..3b74f9fa3ff 100644
--- a/0.16/models/generated/torchvision.models.swin_v2_s.html
+++ b/0.16/models/generated/torchvision.models.swin_v2_s.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin_v2_s &mdash; Torchvision main documentation</title>
+  <title>swin_v2_s &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.swin_v2_t.html b/0.16/models/generated/torchvision.models.swin_v2_t.html
index fae33e8bc46..c43048e0feb 100644
--- a/0.16/models/generated/torchvision.models.swin_v2_t.html
+++ b/0.16/models/generated/torchvision.models.swin_v2_t.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin_v2_t &mdash; Torchvision main documentation</title>
+  <title>swin_v2_t &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg11.html b/0.16/models/generated/torchvision.models.vgg11.html
index e6624c19add..750ebb203f4 100644
--- a/0.16/models/generated/torchvision.models.vgg11.html
+++ b/0.16/models/generated/torchvision.models.vgg11.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg11 &mdash; Torchvision main documentation</title>
+  <title>vgg11 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg11_bn.html b/0.16/models/generated/torchvision.models.vgg11_bn.html
index 82b0ee1d9df..f6a8c475de7 100644
--- a/0.16/models/generated/torchvision.models.vgg11_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg11_bn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg11_bn &mdash; Torchvision main documentation</title>
+  <title>vgg11_bn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg13.html b/0.16/models/generated/torchvision.models.vgg13.html
index f7f765a07d0..62488ef6a94 100644
--- a/0.16/models/generated/torchvision.models.vgg13.html
+++ b/0.16/models/generated/torchvision.models.vgg13.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg13 &mdash; Torchvision main documentation</title>
+  <title>vgg13 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg13_bn.html b/0.16/models/generated/torchvision.models.vgg13_bn.html
index ec53c8e9d35..93ca2101183 100644
--- a/0.16/models/generated/torchvision.models.vgg13_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg13_bn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg13_bn &mdash; Torchvision main documentation</title>
+  <title>vgg13_bn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg16.html b/0.16/models/generated/torchvision.models.vgg16.html
index 6128953812a..7ccefbfbfce 100644
--- a/0.16/models/generated/torchvision.models.vgg16.html
+++ b/0.16/models/generated/torchvision.models.vgg16.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg16 &mdash; Torchvision main documentation</title>
+  <title>vgg16 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg16_bn.html b/0.16/models/generated/torchvision.models.vgg16_bn.html
index fc2034ae6d4..bf7f0d43ed6 100644
--- a/0.16/models/generated/torchvision.models.vgg16_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg16_bn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg16_bn &mdash; Torchvision main documentation</title>
+  <title>vgg16_bn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg19.html b/0.16/models/generated/torchvision.models.vgg19.html
index a5d786d6af3..683bce4c7be 100644
--- a/0.16/models/generated/torchvision.models.vgg19.html
+++ b/0.16/models/generated/torchvision.models.vgg19.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg19 &mdash; Torchvision main documentation</title>
+  <title>vgg19 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vgg19_bn.html b/0.16/models/generated/torchvision.models.vgg19_bn.html
index f2839f4c5f1..28ab8cf57e2 100644
--- a/0.16/models/generated/torchvision.models.vgg19_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg19_bn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vgg19_bn &mdash; Torchvision main documentation</title>
+  <title>vgg19_bn &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.mc3_18.html b/0.16/models/generated/torchvision.models.video.mc3_18.html
index 8affd14c0eb..576892f2268 100644
--- a/0.16/models/generated/torchvision.models.video.mc3_18.html
+++ b/0.16/models/generated/torchvision.models.video.mc3_18.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mc3_18 &mdash; Torchvision main documentation</title>
+  <title>mc3_18 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.mvit_v1_b.html b/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
index c010b4e7cd1..16ebe20bcbe 100644
--- a/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
+++ b/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mvit_v1_b &mdash; Torchvision main documentation</title>
+  <title>mvit_v1_b &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.mvit_v2_s.html b/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
index 9f9cc588397..71d519176fa 100644
--- a/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
+++ b/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>mvit_v2_s &mdash; Torchvision main documentation</title>
+  <title>mvit_v2_s &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.r2plus1d_18.html b/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
index 24f78e2b9b5..d3a5e826bd3 100644
--- a/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
+++ b/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>r2plus1d_18 &mdash; Torchvision main documentation</title>
+  <title>r2plus1d_18 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.r3d_18.html b/0.16/models/generated/torchvision.models.video.r3d_18.html
index 43b3101de2b..0a1194f2f2d 100644
--- a/0.16/models/generated/torchvision.models.video.r3d_18.html
+++ b/0.16/models/generated/torchvision.models.video.r3d_18.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>r3d_18 &mdash; Torchvision main documentation</title>
+  <title>r3d_18 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.s3d.html b/0.16/models/generated/torchvision.models.video.s3d.html
index 95c1abe3fef..2dbcf53a4ee 100644
--- a/0.16/models/generated/torchvision.models.video.s3d.html
+++ b/0.16/models/generated/torchvision.models.video.s3d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>s3d &mdash; Torchvision main documentation</title>
+  <title>s3d &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_b.html b/0.16/models/generated/torchvision.models.video.swin3d_b.html
index a111edf32fd..80d3cc900c5 100644
--- a/0.16/models/generated/torchvision.models.video.swin3d_b.html
+++ b/0.16/models/generated/torchvision.models.video.swin3d_b.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin3d_b &mdash; Torchvision main documentation</title>
+  <title>swin3d_b &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_s.html b/0.16/models/generated/torchvision.models.video.swin3d_s.html
index 9db91e2e27a..77f9aca2c63 100644
--- a/0.16/models/generated/torchvision.models.video.swin3d_s.html
+++ b/0.16/models/generated/torchvision.models.video.swin3d_s.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin3d_s &mdash; Torchvision main documentation</title>
+  <title>swin3d_s &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_t.html b/0.16/models/generated/torchvision.models.video.swin3d_t.html
index f19fb2c3527..59799876633 100644
--- a/0.16/models/generated/torchvision.models.video.swin3d_t.html
+++ b/0.16/models/generated/torchvision.models.video.swin3d_t.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>swin3d_t &mdash; Torchvision main documentation</title>
+  <title>swin3d_t &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vit_b_16.html b/0.16/models/generated/torchvision.models.vit_b_16.html
index 42ca8c934bc..e72966c9cb3 100644
--- a/0.16/models/generated/torchvision.models.vit_b_16.html
+++ b/0.16/models/generated/torchvision.models.vit_b_16.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vit_b_16 &mdash; Torchvision main documentation</title>
+  <title>vit_b_16 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vit_b_32.html b/0.16/models/generated/torchvision.models.vit_b_32.html
index c6af42e359e..4c93fd474bf 100644
--- a/0.16/models/generated/torchvision.models.vit_b_32.html
+++ b/0.16/models/generated/torchvision.models.vit_b_32.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vit_b_32 &mdash; Torchvision main documentation</title>
+  <title>vit_b_32 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vit_h_14.html b/0.16/models/generated/torchvision.models.vit_h_14.html
index 2bd61ea9fa1..ae1e2f206f4 100644
--- a/0.16/models/generated/torchvision.models.vit_h_14.html
+++ b/0.16/models/generated/torchvision.models.vit_h_14.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vit_h_14 &mdash; Torchvision main documentation</title>
+  <title>vit_h_14 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vit_l_16.html b/0.16/models/generated/torchvision.models.vit_l_16.html
index 6b89ccc16b1..6e54825fed1 100644
--- a/0.16/models/generated/torchvision.models.vit_l_16.html
+++ b/0.16/models/generated/torchvision.models.vit_l_16.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vit_l_16 &mdash; Torchvision main documentation</title>
+  <title>vit_l_16 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.vit_l_32.html b/0.16/models/generated/torchvision.models.vit_l_32.html
index bb88eb96d6d..6e0e331944a 100644
--- a/0.16/models/generated/torchvision.models.vit_l_32.html
+++ b/0.16/models/generated/torchvision.models.vit_l_32.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>vit_l_32 &mdash; Torchvision main documentation</title>
+  <title>vit_l_32 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.wide_resnet101_2.html b/0.16/models/generated/torchvision.models.wide_resnet101_2.html
index 1213161f866..a602c07edd0 100644
--- a/0.16/models/generated/torchvision.models.wide_resnet101_2.html
+++ b/0.16/models/generated/torchvision.models.wide_resnet101_2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>wide_resnet101_2 &mdash; Torchvision main documentation</title>
+  <title>wide_resnet101_2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/generated/torchvision.models.wide_resnet50_2.html b/0.16/models/generated/torchvision.models.wide_resnet50_2.html
index 0c256bdf335..365d27b7e2b 100644
--- a/0.16/models/generated/torchvision.models.wide_resnet50_2.html
+++ b/0.16/models/generated/torchvision.models.wide_resnet50_2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>wide_resnet50_2 &mdash; Torchvision main documentation</title>
+  <title>wide_resnet50_2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/googlenet.html b/0.16/models/googlenet.html
index 3405f47f1d3..e99f0841b09 100644
--- a/0.16/models/googlenet.html
+++ b/0.16/models/googlenet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>GoogLeNet &mdash; Torchvision main documentation</title>
+  <title>GoogLeNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/googlenet_quant.html b/0.16/models/googlenet_quant.html
index eedf20c1516..0fd40edb99f 100644
--- a/0.16/models/googlenet_quant.html
+++ b/0.16/models/googlenet_quant.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Quantized GoogLeNet &mdash; Torchvision main documentation</title>
+  <title>Quantized GoogLeNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/inception.html b/0.16/models/inception.html
index b2029c883e4..186ada24c52 100644
--- a/0.16/models/inception.html
+++ b/0.16/models/inception.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Inception V3 &mdash; Torchvision main documentation</title>
+  <title>Inception V3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/inception_quant.html b/0.16/models/inception_quant.html
index 4e40d76572a..e9d5a0c27ff 100644
--- a/0.16/models/inception_quant.html
+++ b/0.16/models/inception_quant.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Quantized InceptionV3 &mdash; Torchvision main documentation</title>
+  <title>Quantized InceptionV3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/keypoint_rcnn.html b/0.16/models/keypoint_rcnn.html
index 5e5ed92d4ee..5d2794691d8 100644
--- a/0.16/models/keypoint_rcnn.html
+++ b/0.16/models/keypoint_rcnn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Keypoint R-CNN &mdash; Torchvision main documentation</title>
+  <title>Keypoint R-CNN &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/lraspp.html b/0.16/models/lraspp.html
index 82c81e47e53..95ef22ed2da 100644
--- a/0.16/models/lraspp.html
+++ b/0.16/models/lraspp.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>LRASPP &mdash; Torchvision main documentation</title>
+  <title>LRASPP &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/mask_rcnn.html b/0.16/models/mask_rcnn.html
index 638fcde9ef4..dc4cedbe79c 100644
--- a/0.16/models/mask_rcnn.html
+++ b/0.16/models/mask_rcnn.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Mask R-CNN &mdash; Torchvision main documentation</title>
+  <title>Mask R-CNN &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/maxvit.html b/0.16/models/maxvit.html
index e38ea884371..580995c932a 100644
--- a/0.16/models/maxvit.html
+++ b/0.16/models/maxvit.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MaxVit &mdash; Torchvision main documentation</title>
+  <title>MaxVit &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/mnasnet.html b/0.16/models/mnasnet.html
index 9b830dd5df6..c4e3ea5bd4c 100644
--- a/0.16/models/mnasnet.html
+++ b/0.16/models/mnasnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MNASNet &mdash; Torchvision main documentation</title>
+  <title>MNASNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/mobilenetv2.html b/0.16/models/mobilenetv2.html
index c04f4885d3c..55cc7ceccf9 100644
--- a/0.16/models/mobilenetv2.html
+++ b/0.16/models/mobilenetv2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MobileNet V2 &mdash; Torchvision main documentation</title>
+  <title>MobileNet V2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/mobilenetv2_quant.html b/0.16/models/mobilenetv2_quant.html
index d717e50960e..87f7fdf4f45 100644
--- a/0.16/models/mobilenetv2_quant.html
+++ b/0.16/models/mobilenetv2_quant.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Quantized MobileNet V2 &mdash; Torchvision main documentation</title>
+  <title>Quantized MobileNet V2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/mobilenetv3.html b/0.16/models/mobilenetv3.html
index 04727448f90..74d6d2a63ba 100644
--- a/0.16/models/mobilenetv3.html
+++ b/0.16/models/mobilenetv3.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>MobileNet V3 &mdash; Torchvision main documentation</title>
+  <title>MobileNet V3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/mobilenetv3_quant.html b/0.16/models/mobilenetv3_quant.html
index 38acf7b9035..e756c277879 100644
--- a/0.16/models/mobilenetv3_quant.html
+++ b/0.16/models/mobilenetv3_quant.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Quantized MobileNet V3 &mdash; Torchvision main documentation</title>
+  <title>Quantized MobileNet V3 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/raft.html b/0.16/models/raft.html
index f1576603381..b55b616e8ab 100644
--- a/0.16/models/raft.html
+++ b/0.16/models/raft.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RAFT &mdash; Torchvision main documentation</title>
+  <title>RAFT &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/regnet.html b/0.16/models/regnet.html
index 9b9e502f0cf..eaa6c4143ba 100644
--- a/0.16/models/regnet.html
+++ b/0.16/models/regnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RegNet &mdash; Torchvision main documentation</title>
+  <title>RegNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/resnet.html b/0.16/models/resnet.html
index aefdec9b675..009a23cf157 100644
--- a/0.16/models/resnet.html
+++ b/0.16/models/resnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ResNet &mdash; Torchvision main documentation</title>
+  <title>ResNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/resnet_quant.html b/0.16/models/resnet_quant.html
index f659f099303..d4b65917178 100644
--- a/0.16/models/resnet_quant.html
+++ b/0.16/models/resnet_quant.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Quantized ResNet &mdash; Torchvision main documentation</title>
+  <title>Quantized ResNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/resnext.html b/0.16/models/resnext.html
index ae8e2d6bd2a..243e239f728 100644
--- a/0.16/models/resnext.html
+++ b/0.16/models/resnext.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ResNeXt &mdash; Torchvision main documentation</title>
+  <title>ResNeXt &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/resnext_quant.html b/0.16/models/resnext_quant.html
index 126ccf24932..116d4cb321b 100644
--- a/0.16/models/resnext_quant.html
+++ b/0.16/models/resnext_quant.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Quantized ResNeXt &mdash; Torchvision main documentation</title>
+  <title>Quantized ResNeXt &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/retinanet.html b/0.16/models/retinanet.html
index a934e0a91c6..4368b29b773 100644
--- a/0.16/models/retinanet.html
+++ b/0.16/models/retinanet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>RetinaNet &mdash; Torchvision main documentation</title>
+  <title>RetinaNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/shufflenetv2.html b/0.16/models/shufflenetv2.html
index 16f4eb2d1e2..2c76e10f24a 100644
--- a/0.16/models/shufflenetv2.html
+++ b/0.16/models/shufflenetv2.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>ShuffleNet V2 &mdash; Torchvision main documentation</title>
+  <title>ShuffleNet V2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/shufflenetv2_quant.html b/0.16/models/shufflenetv2_quant.html
index 89e12b1aa12..27ad8df1efa 100644
--- a/0.16/models/shufflenetv2_quant.html
+++ b/0.16/models/shufflenetv2_quant.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Quantized ShuffleNet V2 &mdash; Torchvision main documentation</title>
+  <title>Quantized ShuffleNet V2 &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/squeezenet.html b/0.16/models/squeezenet.html
index ee01f6841e2..ae9f17db3b7 100644
--- a/0.16/models/squeezenet.html
+++ b/0.16/models/squeezenet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SqueezeNet &mdash; Torchvision main documentation</title>
+  <title>SqueezeNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/ssd.html b/0.16/models/ssd.html
index 5baa11efb09..f67d5b52c25 100644
--- a/0.16/models/ssd.html
+++ b/0.16/models/ssd.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SSD &mdash; Torchvision main documentation</title>
+  <title>SSD &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/ssdlite.html b/0.16/models/ssdlite.html
index 738834996a6..0fdd12eb8b3 100644
--- a/0.16/models/ssdlite.html
+++ b/0.16/models/ssdlite.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SSDlite &mdash; Torchvision main documentation</title>
+  <title>SSDlite &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/swin_transformer.html b/0.16/models/swin_transformer.html
index ad430ea1cb6..8495903bd0a 100644
--- a/0.16/models/swin_transformer.html
+++ b/0.16/models/swin_transformer.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>SwinTransformer &mdash; Torchvision main documentation</title>
+  <title>SwinTransformer &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/vgg.html b/0.16/models/vgg.html
index 22dfb73255b..ec427d8e528 100644
--- a/0.16/models/vgg.html
+++ b/0.16/models/vgg.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>VGG &mdash; Torchvision main documentation</title>
+  <title>VGG &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/video_mvit.html b/0.16/models/video_mvit.html
index 0782ee4c3e3..99d0943910a 100644
--- a/0.16/models/video_mvit.html
+++ b/0.16/models/video_mvit.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Video MViT &mdash; Torchvision main documentation</title>
+  <title>Video MViT &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/video_resnet.html b/0.16/models/video_resnet.html
index 7af50e4bcdb..6d9d315729c 100644
--- a/0.16/models/video_resnet.html
+++ b/0.16/models/video_resnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Video ResNet &mdash; Torchvision main documentation</title>
+  <title>Video ResNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/video_s3d.html b/0.16/models/video_s3d.html
index b6cc2bf1a3f..345b91759a4 100644
--- a/0.16/models/video_s3d.html
+++ b/0.16/models/video_s3d.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Video S3D &mdash; Torchvision main documentation</title>
+  <title>Video S3D &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/video_swin_transformer.html b/0.16/models/video_swin_transformer.html
index fb3e95cdeae..e1c1d2beff8 100644
--- a/0.16/models/video_swin_transformer.html
+++ b/0.16/models/video_swin_transformer.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Video SwinTransformer &mdash; Torchvision main documentation</title>
+  <title>Video SwinTransformer &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/vision_transformer.html b/0.16/models/vision_transformer.html
index 50b63e4e32a..c12e81368b8 100644
--- a/0.16/models/vision_transformer.html
+++ b/0.16/models/vision_transformer.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>VisionTransformer &mdash; Torchvision main documentation</title>
+  <title>VisionTransformer &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/models/wide_resnet.html b/0.16/models/wide_resnet.html
index 45acb5d1394..0175a2e6177 100644
--- a/0.16/models/wide_resnet.html
+++ b/0.16/models/wide_resnet.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Wide ResNet &mdash; Torchvision main documentation</title>
+  <title>Wide ResNet &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/ops.html b/0.16/ops.html
index 64b2e0a3bf0..58828994c34 100644
--- a/0.16/ops.html
+++ b/0.16/ops.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Operators &mdash; Torchvision main documentation</title>
+  <title>Operators &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/py-modindex.html b/0.16/py-modindex.html
index 6956c37f5f9..77d9d99b879 100644
--- a/0.16/py-modindex.html
+++ b/0.16/py-modindex.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Python Module Index &mdash; Torchvision main documentation</title>
+  <title>Python Module Index &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/search.html b/0.16/search.html
index 4d5bac9d6d2..3ba23eca75b 100644
--- a/0.16/search.html
+++ b/0.16/search.html
@@ -9,7 +9,7 @@
   
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Search &mdash; Torchvision main documentation</title>
+  <title>Search &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/training_references.html b/0.16/training_references.html
index ea6a452a208..25aa52c78e8 100644
--- a/0.16/training_references.html
+++ b/0.16/training_references.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Training references &mdash; Torchvision main documentation</title>
+  <title>Training references &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/transforms.html b/0.16/transforms.html
index 4d726a501fa..ffd16a0f583 100644
--- a/0.16/transforms.html
+++ b/0.16/transforms.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Transforming and augmenting images &mdash; Torchvision main documentation</title>
+  <title>Transforming and augmenting images &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/tv_tensors.html b/0.16/tv_tensors.html
index a459475c705..28a0f69549e 100644
--- a/0.16/tv_tensors.html
+++ b/0.16/tv_tensors.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>TVTensors &mdash; Torchvision main documentation</title>
+  <title>TVTensors &mdash; Torchvision 0.16 documentation</title>
   
 
   
diff --git a/0.16/utils.html b/0.16/utils.html
index 0da0eb38478..6686f32f0a0 100644
--- a/0.16/utils.html
+++ b/0.16/utils.html
@@ -10,7 +10,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
   
-  <title>Utils &mdash; Torchvision main documentation</title>
+  <title>Utils &mdash; Torchvision 0.16 documentation</title>
   
 
   

From 5977476b7b44e7009810056355d799b8d0cde0d6 Mon Sep 17 00:00:00 2001
From: Nicolas Hug <contact@nicolas-hug.com>
Date: Wed, 4 Oct 2023 17:44:54 +0100
Subject: [PATCH 3/5] main (0.16.0+9b13bcb) -> 0.16

---
 0.16/_modules/index.html                      |   2 +-
 0.16/_modules/torchvision.html                |   2 +-
 .../torchvision/datasets/_optical_flow.html   |   2 +-
 .../datasets/_stereo_matching.html            |   2 +-
 .../torchvision/datasets/caltech.html         |   2 +-
 .../_modules/torchvision/datasets/celeba.html |   2 +-
 0.16/_modules/torchvision/datasets/cifar.html |   2 +-
 .../torchvision/datasets/cityscapes.html      |   2 +-
 0.16/_modules/torchvision/datasets/coco.html  |   2 +-
 .../torchvision/datasets/country211.html      |   2 +-
 0.16/_modules/torchvision/datasets/dtd.html   |   2 +-
 .../torchvision/datasets/eurosat.html         |   2 +-
 .../torchvision/datasets/fakedata.html        |   2 +-
 .../torchvision/datasets/fer2013.html         |   2 +-
 .../torchvision/datasets/fgvc_aircraft.html   |   2 +-
 .../_modules/torchvision/datasets/flickr.html |   2 +-
 .../torchvision/datasets/flowers102.html      |   2 +-
 .../_modules/torchvision/datasets/folder.html |   2 +-
 .../torchvision/datasets/food101.html         |   2 +-
 0.16/_modules/torchvision/datasets/gtsrb.html |   2 +-
 .../_modules/torchvision/datasets/hmdb51.html |   2 +-
 .../torchvision/datasets/imagenet.html        |   2 +-
 .../torchvision/datasets/inaturalist.html     |   2 +-
 .../torchvision/datasets/kinetics.html        |   2 +-
 0.16/_modules/torchvision/datasets/kitti.html |   2 +-
 0.16/_modules/torchvision/datasets/lfw.html   |   2 +-
 0.16/_modules/torchvision/datasets/lsun.html  |   2 +-
 0.16/_modules/torchvision/datasets/mnist.html |   2 +-
 .../torchvision/datasets/moving_mnist.html    |   2 +-
 .../torchvision/datasets/omniglot.html        |   2 +-
 .../torchvision/datasets/oxford_iiit_pet.html |   2 +-
 0.16/_modules/torchvision/datasets/pcam.html  |   2 +-
 .../torchvision/datasets/phototour.html       |   2 +-
 .../torchvision/datasets/places365.html       |   2 +-
 .../torchvision/datasets/rendered_sst2.html   |   2 +-
 0.16/_modules/torchvision/datasets/sbd.html   |   2 +-
 0.16/_modules/torchvision/datasets/sbu.html   |   2 +-
 .../torchvision/datasets/semeion.html         |   2 +-
 .../torchvision/datasets/stanford_cars.html   |   2 +-
 0.16/_modules/torchvision/datasets/stl10.html |   2 +-
 .../_modules/torchvision/datasets/sun397.html |   2 +-
 0.16/_modules/torchvision/datasets/svhn.html  |   2 +-
 .../_modules/torchvision/datasets/ucf101.html |   2 +-
 0.16/_modules/torchvision/datasets/usps.html  |   2 +-
 .../_modules/torchvision/datasets/vision.html |   2 +-
 0.16/_modules/torchvision/datasets/voc.html   |   2 +-
 .../torchvision/datasets/widerface.html       |   2 +-
 0.16/_modules/torchvision/io/image.html       |   2 +-
 0.16/_modules/torchvision/io/video.html       |   2 +-
 .../_modules/torchvision/io/video_reader.html |   2 +-
 0.16/_modules/torchvision/models/_api.html    |   2 +-
 0.16/_modules/torchvision/models/alexnet.html |   2 +-
 .../_modules/torchvision/models/convnext.html |   2 +-
 .../_modules/torchvision/models/densenet.html |   2 +-
 .../models/detection/faster_rcnn.html         |   2 +-
 .../torchvision/models/detection/fcos.html    |   2 +-
 .../models/detection/keypoint_rcnn.html       |   2 +-
 .../models/detection/mask_rcnn.html           |   2 +-
 .../models/detection/retinanet.html           |   2 +-
 .../torchvision/models/detection/ssd.html     |   2 +-
 .../torchvision/models/detection/ssdlite.html |   2 +-
 .../torchvision/models/efficientnet.html      |   2 +-
 .../models/feature_extraction.html            |   2 +-
 .../torchvision/models/googlenet.html         |   2 +-
 .../torchvision/models/inception.html         |   2 +-
 0.16/_modules/torchvision/models/maxvit.html  |   2 +-
 0.16/_modules/torchvision/models/mnasnet.html |   2 +-
 .../torchvision/models/mobilenetv2.html       |   2 +-
 .../torchvision/models/mobilenetv3.html       |   2 +-
 .../torchvision/models/optical_flow/raft.html |   2 +-
 .../models/quantization/googlenet.html        |   2 +-
 .../models/quantization/inception.html        |   2 +-
 .../models/quantization/mobilenetv2.html      |   2 +-
 .../models/quantization/mobilenetv3.html      |   2 +-
 .../models/quantization/resnet.html           |   2 +-
 .../models/quantization/shufflenetv2.html     |   2 +-
 0.16/_modules/torchvision/models/regnet.html  |   2 +-
 0.16/_modules/torchvision/models/resnet.html  |   2 +-
 .../models/segmentation/deeplabv3.html        |   2 +-
 .../torchvision/models/segmentation/fcn.html  |   2 +-
 .../models/segmentation/lraspp.html           |   2 +-
 .../torchvision/models/shufflenetv2.html      |   2 +-
 .../torchvision/models/squeezenet.html        |   2 +-
 .../torchvision/models/swin_transformer.html  |   2 +-
 0.16/_modules/torchvision/models/vgg.html     |   2 +-
 .../torchvision/models/video/mvit.html        |   2 +-
 .../torchvision/models/video/resnet.html      |   2 +-
 .../torchvision/models/video/s3d.html         |   2 +-
 .../models/video/swin_transformer.html        |   2 +-
 .../models/vision_transformer.html            |   2 +-
 0.16/_modules/torchvision/ops/boxes.html      |   2 +-
 0.16/_modules/torchvision/ops/ciou_loss.html  |   2 +-
 .../_modules/torchvision/ops/deform_conv.html |   2 +-
 0.16/_modules/torchvision/ops/diou_loss.html  |   2 +-
 0.16/_modules/torchvision/ops/drop_block.html |   2 +-
 .../ops/feature_pyramid_network.html          |   2 +-
 0.16/_modules/torchvision/ops/focal_loss.html |   2 +-
 0.16/_modules/torchvision/ops/giou_loss.html  |   2 +-
 0.16/_modules/torchvision/ops/misc.html       |   2 +-
 0.16/_modules/torchvision/ops/poolers.html    |   2 +-
 .../torchvision/ops/ps_roi_align.html         |   2 +-
 .../_modules/torchvision/ops/ps_roi_pool.html |   2 +-
 0.16/_modules/torchvision/ops/roi_align.html  |   2 +-
 0.16/_modules/torchvision/ops/roi_pool.html   |   2 +-
 .../torchvision/ops/stochastic_depth.html     |   2 +-
 .../torchvision/transforms/autoaugment.html   |   2 +-
 .../torchvision/transforms/functional.html    |   2 +-
 .../torchvision/transforms/transforms.html    |   2 +-
 .../torchvision/transforms/v2/_augment.html   |   2 +-
 .../transforms/v2/_auto_augment.html          |   2 +-
 .../torchvision/transforms/v2/_color.html     |   2 +-
 .../torchvision/transforms/v2/_container.html |   2 +-
 .../transforms/v2/_deprecated.html            |   2 +-
 .../torchvision/transforms/v2/_geometry.html  |   2 +-
 .../torchvision/transforms/v2/_meta.html      |   2 +-
 .../torchvision/transforms/v2/_misc.html      |   2 +-
 .../torchvision/transforms/v2/_temporal.html  |   2 +-
 .../transforms/v2/_type_conversion.html       |   2 +-
 .../transforms/v2/functional/_augment.html    |   2 +-
 .../transforms/v2/functional/_color.html      |   2 +-
 .../transforms/v2/functional/_deprecated.html |   2 +-
 .../transforms/v2/functional/_geometry.html   |   2 +-
 .../transforms/v2/functional/_meta.html       |   2 +-
 .../transforms/v2/functional/_misc.html       |   2 +-
 .../transforms/v2/functional/_temporal.html   |   2 +-
 .../v2/functional/_type_conversion.html       |   2 +-
 .../transforms/v2/functional/_utils.html      |   2 +-
 0.16/_modules/torchvision/tv_tensors.html     |   2 +-
 .../tv_tensors/_bounding_boxes.html           |   2 +-
 .../tv_tensors/_dataset_wrapper.html          |   2 +-
 .../torchvision/tv_tensors/_image.html        |   2 +-
 .../torchvision/tv_tensors/_mask.html         |   2 +-
 .../tv_tensors/_torch_function_helpers.html   |   2 +-
 .../torchvision/tv_tensors/_tv_tensor.html    |   2 +-
 .../torchvision/tv_tensors/_video.html        |   2 +-
 0.16/_modules/torchvision/utils.html          |   2 +-
 0.16/auto_examples/index.html                 |   2 +-
 0.16/auto_examples/others/index.html          |   2 +-
 .../others/plot_optical_flow.html             |   2 +-
 .../others/plot_repurposing_annotations.html  |   2 +-
 .../plot_scripted_tensor_transforms.html      |   2 +-
 0.16/auto_examples/others/plot_video_api.html |   2 +-
 .../others/plot_visualization_utils.html      |   2 +-
 .../others/sg_execution_times.html            |   2 +-
 0.16/auto_examples/transforms/index.html      |   2 +-
 .../transforms/plot_custom_transforms.html    |   2 +-
 .../transforms/plot_custom_tv_tensors.html    |   2 +-
 .../transforms/plot_cutmix_mixup.html         |   2 +-
 .../transforms/plot_transforms_e2e.html       |   2 +-
 .../plot_transforms_getting_started.html      |   2 +-
 .../plot_transforms_illustrations.html        |   2 +-
 .../transforms/plot_tv_tensors.html           |   2 +-
 .../transforms/sg_execution_times.html        |   2 +-
 0.16/datasets.html                            |   2 +-
 0.16/feature_extraction.html                  |   2 +-
 .../generated/classification_quant_table.html |   2 +-
 0.16/generated/classification_table.html      |   2 +-
 0.16/generated/detection_keypoint_table.html  |   2 +-
 0.16/generated/detection_table.html           |   2 +-
 .../instance_segmentation_table.html          |   2 +-
 0.16/generated/segmentation_table.html        |   2 +-
 .../torchvision.datasets.CIFAR10.html         |   2 +-
 .../torchvision.datasets.CIFAR100.html        |   2 +-
 .../torchvision.datasets.CREStereo.html       |   2 +-
 .../torchvision.datasets.Caltech101.html      |   2 +-
 .../torchvision.datasets.Caltech256.html      |   2 +-
 .../torchvision.datasets.CarlaStereo.html     |   2 +-
 .../torchvision.datasets.CelebA.html          |   2 +-
 .../torchvision.datasets.Cityscapes.html      |   2 +-
 .../torchvision.datasets.CocoCaptions.html    |   2 +-
 .../torchvision.datasets.CocoDetection.html   |   2 +-
 .../torchvision.datasets.Country211.html      |   2 +-
 0.16/generated/torchvision.datasets.DTD.html  |   2 +-
 .../torchvision.datasets.DatasetFolder.html   |   2 +-
 .../torchvision.datasets.EMNIST.html          |   2 +-
 .../torchvision.datasets.ETH3DStereo.html     |   2 +-
 .../torchvision.datasets.EuroSAT.html         |   2 +-
 .../torchvision.datasets.FER2013.html         |   2 +-
 .../torchvision.datasets.FGVCAircraft.html    |   2 +-
 .../torchvision.datasets.FakeData.html        |   2 +-
 ...chvision.datasets.FallingThingsStereo.html |   2 +-
 .../torchvision.datasets.FashionMNIST.html    |   2 +-
 .../torchvision.datasets.Flickr30k.html       |   2 +-
 .../torchvision.datasets.Flickr8k.html        |   2 +-
 .../torchvision.datasets.Flowers102.html      |   2 +-
 .../torchvision.datasets.FlyingChairs.html    |   2 +-
 .../torchvision.datasets.FlyingThings3D.html  |   2 +-
 .../torchvision.datasets.Food101.html         |   2 +-
 .../generated/torchvision.datasets.GTSRB.html |   2 +-
 0.16/generated/torchvision.datasets.HD1K.html |   2 +-
 .../torchvision.datasets.HMDB51.html          |   2 +-
 .../torchvision.datasets.INaturalist.html     |   2 +-
 .../torchvision.datasets.ImageFolder.html     |   2 +-
 .../torchvision.datasets.ImageNet.html        |   2 +-
 .../torchvision.datasets.InStereo2k.html      |   2 +-
 .../torchvision.datasets.KMNIST.html          |   2 +-
 .../torchvision.datasets.Kinetics.html        |   2 +-
 .../generated/torchvision.datasets.Kitti.html |   2 +-
 .../torchvision.datasets.Kitti2012Stereo.html |   2 +-
 .../torchvision.datasets.Kitti2015Stereo.html |   2 +-
 .../torchvision.datasets.KittiFlow.html       |   2 +-
 .../torchvision.datasets.LFWPairs.html        |   2 +-
 .../torchvision.datasets.LFWPeople.html       |   2 +-
 0.16/generated/torchvision.datasets.LSUN.html |   2 +-
 .../generated/torchvision.datasets.MNIST.html |   2 +-
 ...hvision.datasets.Middlebury2014Stereo.html |   2 +-
 .../torchvision.datasets.MovingMNIST.html     |   2 +-
 .../torchvision.datasets.Omniglot.html        |   2 +-
 .../torchvision.datasets.OxfordIIITPet.html   |   2 +-
 0.16/generated/torchvision.datasets.PCAM.html |   2 +-
 .../torchvision.datasets.PhotoTour.html       |   2 +-
 .../torchvision.datasets.Places365.html       |   2 +-
 .../torchvision.datasets.QMNIST.html          |   2 +-
 .../torchvision.datasets.RenderedSST2.html    |   2 +-
 .../torchvision.datasets.SBDataset.html       |   2 +-
 0.16/generated/torchvision.datasets.SBU.html  |   2 +-
 .../torchvision.datasets.SEMEION.html         |   2 +-
 .../generated/torchvision.datasets.STL10.html |   2 +-
 .../torchvision.datasets.SUN397.html          |   2 +-
 0.16/generated/torchvision.datasets.SVHN.html |   2 +-
 .../torchvision.datasets.SceneFlowStereo.html |   2 +-
 .../torchvision.datasets.Sintel.html          |   2 +-
 .../torchvision.datasets.SintelStereo.html    |   2 +-
 .../torchvision.datasets.StanfordCars.html    |   2 +-
 .../torchvision.datasets.UCF101.html          |   2 +-
 0.16/generated/torchvision.datasets.USPS.html |   2 +-
 .../torchvision.datasets.VOCDetection.html    |   2 +-
 .../torchvision.datasets.VOCSegmentation.html |   2 +-
 .../torchvision.datasets.VisionDataset.html   |   2 +-
 .../torchvision.datasets.WIDERFace.html       |   2 +-
 ...tasets.wrap_dataset_for_transforms_v2.html |   2 +-
 .../torchvision.io.ImageReadMode.html         |   2 +-
 .../generated/torchvision.io.VideoReader.html |   2 +-
 .../torchvision.io.decode_image.html          |   2 +-
 .../generated/torchvision.io.decode_jpeg.html |   2 +-
 0.16/generated/torchvision.io.decode_png.html |   2 +-
 .../generated/torchvision.io.encode_jpeg.html |   2 +-
 0.16/generated/torchvision.io.encode_png.html |   2 +-
 0.16/generated/torchvision.io.read_file.html  |   2 +-
 0.16/generated/torchvision.io.read_image.html |   2 +-
 0.16/generated/torchvision.io.read_video.html |   2 +-
 .../torchvision.io.read_video_timestamps.html |   2 +-
 0.16/generated/torchvision.io.write_file.html |   2 +-
 0.16/generated/torchvision.io.write_jpeg.html |   2 +-
 0.16/generated/torchvision.io.write_png.html  |   2 +-
 .../generated/torchvision.io.write_video.html |   2 +-
 ...e_extraction.create_feature_extractor.html |   2 +-
 ...ature_extraction.get_graph_node_names.html |   2 +-
 .../torchvision.models.get_model.html         |   2 +-
 .../torchvision.models.get_model_weights.html |   2 +-
 .../torchvision.models.get_weight.html        |   2 +-
 .../torchvision.models.list_models.html       |   2 +-
 .../torchvision.ops.Conv2dNormActivation.html |   2 +-
 .../torchvision.ops.Conv3dNormActivation.html |   2 +-
 .../torchvision.ops.DeformConv2d.html         |   2 +-
 .../torchvision.ops.DropBlock2d.html          |   2 +-
 .../torchvision.ops.DropBlock3d.html          |   2 +-
 ...torchvision.ops.FeaturePyramidNetwork.html |   2 +-
 .../torchvision.ops.FrozenBatchNorm2d.html    |   2 +-
 0.16/generated/torchvision.ops.MLP.html       |   2 +-
 .../torchvision.ops.MultiScaleRoIAlign.html   |   2 +-
 .../generated/torchvision.ops.PSRoIAlign.html |   2 +-
 0.16/generated/torchvision.ops.PSRoIPool.html |   2 +-
 0.16/generated/torchvision.ops.Permute.html   |   2 +-
 0.16/generated/torchvision.ops.RoIAlign.html  |   2 +-
 0.16/generated/torchvision.ops.RoIPool.html   |   2 +-
 .../torchvision.ops.SqueezeExcitation.html    |   2 +-
 .../torchvision.ops.StochasticDepth.html      |   2 +-
 .../torchvision.ops.batched_nms.html          |   2 +-
 0.16/generated/torchvision.ops.box_area.html  |   2 +-
 .../torchvision.ops.box_convert.html          |   2 +-
 0.16/generated/torchvision.ops.box_iou.html   |   2 +-
 .../torchvision.ops.clip_boxes_to_image.html  |   2 +-
 .../torchvision.ops.complete_box_iou.html     |   2 +-
 ...torchvision.ops.complete_box_iou_loss.html |   2 +-
 .../torchvision.ops.deform_conv2d.html        |   2 +-
 .../torchvision.ops.distance_box_iou.html     |   2 +-
 ...torchvision.ops.distance_box_iou_loss.html |   2 +-
 .../torchvision.ops.drop_block2d.html         |   2 +-
 .../torchvision.ops.drop_block3d.html         |   2 +-
 .../torchvision.ops.generalized_box_iou.html  |   2 +-
 ...chvision.ops.generalized_box_iou_loss.html |   2 +-
 .../torchvision.ops.masks_to_boxes.html       |   2 +-
 0.16/generated/torchvision.ops.nms.html       |   2 +-
 .../torchvision.ops.ps_roi_align.html         |   2 +-
 .../torchvision.ops.ps_roi_pool.html          |   2 +-
 .../torchvision.ops.remove_small_boxes.html   |   2 +-
 0.16/generated/torchvision.ops.roi_align.html |   2 +-
 0.16/generated/torchvision.ops.roi_pool.html  |   2 +-
 .../torchvision.ops.sigmoid_focal_loss.html   |   2 +-
 .../torchvision.ops.stochastic_depth.html     |   2 +-
 .../torchvision.transforms.AugMix.html        |   2 +-
 .../torchvision.transforms.AutoAugment.html   |   2 +-
 ...chvision.transforms.AutoAugmentPolicy.html |   2 +-
 .../torchvision.transforms.CenterCrop.html    |   2 +-
 .../torchvision.transforms.ColorJitter.html   |   2 +-
 .../torchvision.transforms.Compose.html       |   2 +-
 ...chvision.transforms.ConvertImageDtype.html |   2 +-
 ...rchvision.transforms.ElasticTransform.html |   2 +-
 .../torchvision.transforms.FiveCrop.html      |   2 +-
 .../torchvision.transforms.GaussianBlur.html  |   2 +-
 .../torchvision.transforms.Grayscale.html     |   2 +-
 .../torchvision.transforms.Lambda.html        |   2 +-
 ...ision.transforms.LinearTransformation.html |   2 +-
 .../torchvision.transforms.Normalize.html     |   2 +-
 .../torchvision.transforms.PILToTensor.html   |   2 +-
 .../generated/torchvision.transforms.Pad.html |   2 +-
 .../torchvision.transforms.RandAugment.html   |   2 +-
 ...sion.transforms.RandomAdjustSharpness.html |   2 +-
 .../torchvision.transforms.RandomAffine.html  |   2 +-
 .../torchvision.transforms.RandomApply.html   |   2 +-
 ...hvision.transforms.RandomAutocontrast.html |   2 +-
 .../torchvision.transforms.RandomChoice.html  |   2 +-
 .../torchvision.transforms.RandomCrop.html    |   2 +-
 ...torchvision.transforms.RandomEqualize.html |   2 +-
 .../torchvision.transforms.RandomErasing.html |   2 +-
 ...orchvision.transforms.RandomGrayscale.html |   2 +-
 ...ision.transforms.RandomHorizontalFlip.html |   2 +-
 .../torchvision.transforms.RandomInvert.html  |   2 +-
 .../torchvision.transforms.RandomOrder.html   |   2 +-
 ...chvision.transforms.RandomPerspective.html |   2 +-
 ...orchvision.transforms.RandomPosterize.html |   2 +-
 ...chvision.transforms.RandomResizedCrop.html |   2 +-
 ...torchvision.transforms.RandomRotation.html |   2 +-
 ...torchvision.transforms.RandomSolarize.html |   2 +-
 ...hvision.transforms.RandomVerticalFlip.html |   2 +-
 .../torchvision.transforms.Resize.html        |   2 +-
 .../torchvision.transforms.TenCrop.html       |   2 +-
 .../torchvision.transforms.ToPILImage.html    |   2 +-
 .../torchvision.transforms.ToTensor.html      |   2 +-
 ...hvision.transforms.TrivialAugmentWide.html |   2 +-
 ...ansforms.functional.adjust_brightness.html |   2 +-
 ...transforms.functional.adjust_contrast.html |   2 +-
 ...on.transforms.functional.adjust_gamma.html |   2 +-
 ...sion.transforms.functional.adjust_hue.html |   2 +-
 ...ansforms.functional.adjust_saturation.html |   2 +-
 ...ransforms.functional.adjust_sharpness.html |   2 +-
 ...chvision.transforms.functional.affine.html |   2 +-
 ...on.transforms.functional.autocontrast.html |   2 +-
 ...ion.transforms.functional.center_crop.html |   2 +-
 ...sforms.functional.convert_image_dtype.html |   2 +-
 ...orchvision.transforms.functional.crop.html |   2 +-
 ...vision.transforms.functional.equalize.html |   2 +-
 ...rchvision.transforms.functional.erase.html |   2 +-
 ...ision.transforms.functional.five_crop.html |   2 +-
 ...n.transforms.functional.gaussian_blur.html |   2 +-
 ....transforms.functional.get_dimensions.html |   2 +-
 ...rms.functional.get_image_num_channels.html |   2 +-
 ....transforms.functional.get_image_size.html |   2 +-
 ...rchvision.transforms.functional.hflip.html |   2 +-
 ...chvision.transforms.functional.invert.html |   2 +-
 ...ision.transforms.functional.normalize.html |   2 +-
 ...torchvision.transforms.functional.pad.html |   2 +-
 ...ion.transforms.functional.perspective.html |   2 +-
 ...n.transforms.functional.pil_to_tensor.html |   2 +-
 ...ision.transforms.functional.posterize.html |   2 +-
 ...chvision.transforms.functional.resize.html |   2 +-
 ...on.transforms.functional.resized_crop.html |   2 +-
 ...ransforms.functional.rgb_to_grayscale.html |   2 +-
 ...chvision.transforms.functional.rotate.html |   2 +-
 ...vision.transforms.functional.solarize.html |   2 +-
 ...vision.transforms.functional.ten_crop.html |   2 +-
 ...on.transforms.functional.to_grayscale.html |   2 +-
 ...on.transforms.functional.to_pil_image.html |   2 +-
 ...ision.transforms.functional.to_tensor.html |   2 +-
 ...rchvision.transforms.functional.vflip.html |   2 +-
 .../torchvision.transforms.v2.AugMix.html     |   2 +-
 ...torchvision.transforms.v2.AutoAugment.html |   2 +-
 .../torchvision.transforms.v2.CenterCrop.html |   2 +-
 ...sion.transforms.v2.ClampBoundingBoxes.html |   2 +-
 ...torchvision.transforms.v2.ColorJitter.html |   2 +-
 .../torchvision.transforms.v2.Compose.html    |   2 +-
 ...ransforms.v2.ConvertBoundingBoxFormat.html |   2 +-
 ...ision.transforms.v2.ConvertImageDtype.html |   2 +-
 .../torchvision.transforms.v2.CutMix.html     |   2 +-
 ...vision.transforms.v2.ElasticTransform.html |   2 +-
 .../torchvision.transforms.v2.FiveCrop.html   |   2 +-
 ...orchvision.transforms.v2.GaussianBlur.html |   2 +-
 .../torchvision.transforms.v2.Grayscale.html  |   2 +-
 .../torchvision.transforms.v2.Lambda.html     |   2 +-
 ...on.transforms.v2.LinearTransformation.html |   2 +-
 .../torchvision.transforms.v2.MixUp.html      |   2 +-
 .../torchvision.transforms.v2.Normalize.html  |   2 +-
 ...torchvision.transforms.v2.PILToTensor.html |   2 +-
 .../torchvision.transforms.v2.Pad.html        |   2 +-
 ...torchvision.transforms.v2.RandAugment.html |   2 +-
 ...n.transforms.v2.RandomAdjustSharpness.html |   2 +-
 ...orchvision.transforms.v2.RandomAffine.html |   2 +-
 ...torchvision.transforms.v2.RandomApply.html |   2 +-
 ...sion.transforms.v2.RandomAutocontrast.html |   2 +-
 ...ransforms.v2.RandomChannelPermutation.html |   2 +-
 ...orchvision.transforms.v2.RandomChoice.html |   2 +-
 .../torchvision.transforms.v2.RandomCrop.html |   2 +-
 ...chvision.transforms.v2.RandomEqualize.html |   2 +-
 ...rchvision.transforms.v2.RandomErasing.html |   2 +-
 ...hvision.transforms.v2.RandomGrayscale.html |   2 +-
 ...on.transforms.v2.RandomHorizontalFlip.html |   2 +-
 ...orchvision.transforms.v2.RandomInvert.html |   2 +-
 ...rchvision.transforms.v2.RandomIoUCrop.html |   2 +-
 ...torchvision.transforms.v2.RandomOrder.html |   2 +-
 ...ision.transforms.v2.RandomPerspective.html |   2 +-
 ...ransforms.v2.RandomPhotometricDistort.html |   2 +-
 ...hvision.transforms.v2.RandomPosterize.html |   2 +-
 ...orchvision.transforms.v2.RandomResize.html |   2 +-
 ...ision.transforms.v2.RandomResizedCrop.html |   2 +-
 ...chvision.transforms.v2.RandomRotation.html |   2 +-
 ...sion.transforms.v2.RandomShortestSize.html |   2 +-
 ...chvision.transforms.v2.RandomSolarize.html |   2 +-
 ...sion.transforms.v2.RandomVerticalFlip.html |   2 +-
 ...rchvision.transforms.v2.RandomZoomOut.html |   2 +-
 .../torchvision.transforms.v2.Resize.html     |   2 +-
 ...n.transforms.v2.SanitizeBoundingBoxes.html |   2 +-
 ...torchvision.transforms.v2.ScaleJitter.html |   2 +-
 .../torchvision.transforms.v2.TenCrop.html    |   2 +-
 .../torchvision.transforms.v2.ToDtype.html    |   2 +-
 .../torchvision.transforms.v2.ToImage.html    |   2 +-
 .../torchvision.transforms.v2.ToPILImage.html |   2 +-
 ...orchvision.transforms.v2.ToPureTensor.html |   2 +-
 .../torchvision.transforms.v2.ToTensor.html   |   2 +-
 ...sion.transforms.v2.TrivialAugmentWide.html |   2 +-
 ...ransforms.v2.UniformTemporalSubsample.html |   2 +-
 ...forms.v2.functional.adjust_brightness.html |   2 +-
 ...nsforms.v2.functional.adjust_contrast.html |   2 +-
 ...transforms.v2.functional.adjust_gamma.html |   2 +-
 ...n.transforms.v2.functional.adjust_hue.html |   2 +-
 ...forms.v2.functional.adjust_saturation.html |   2 +-
 ...sforms.v2.functional.adjust_sharpness.html |   2 +-
 ...ision.transforms.v2.functional.affine.html |   2 +-
 ...transforms.v2.functional.autocontrast.html |   2 +-
 ....transforms.v2.functional.center_crop.html |   2 +-
 ...ms.v2.functional.clamp_bounding_boxes.html |   2 +-
 ...unctional.convert_bounding_box_format.html |   2 +-
 ...rms.v2.functional.convert_image_dtype.html |   2 +-
 ...hvision.transforms.v2.functional.crop.html |   2 +-
 ...sion.transforms.v2.functional.elastic.html |   2 +-
 ...ion.transforms.v2.functional.equalize.html |   2 +-
 ...vision.transforms.v2.functional.erase.html |   2 +-
 ...on.transforms.v2.functional.five_crop.html |   2 +-
 ...ransforms.v2.functional.gaussian_blur.html |   2 +-
 ...nsforms.v2.functional.horizontal_flip.html |   2 +-
 ...ision.transforms.v2.functional.invert.html |   2 +-
 ...on.transforms.v2.functional.normalize.html |   2 +-
 ...chvision.transforms.v2.functional.pad.html |   2 +-
 ...sforms.v2.functional.permute_channels.html |   2 +-
 ....transforms.v2.functional.perspective.html |   2 +-
 ...ransforms.v2.functional.pil_to_tensor.html |   2 +-
 ...on.transforms.v2.functional.posterize.html |   2 +-
 ...nsforms.v2.functional.register_kernel.html |   2 +-
 ...ision.transforms.v2.functional.resize.html |   2 +-
 ...transforms.v2.functional.resized_crop.html |   2 +-
 ...sforms.v2.functional.rgb_to_grayscale.html |   2 +-
 ...ision.transforms.v2.functional.rotate.html |   2 +-
 ...ion.transforms.v2.functional.solarize.html |   2 +-
 ...ion.transforms.v2.functional.ten_crop.html |   2 +-
 ...ion.transforms.v2.functional.to_dtype.html |   2 +-
 ...transforms.v2.functional.to_grayscale.html |   2 +-
 ...ion.transforms.v2.functional.to_image.html |   2 +-
 ...transforms.v2.functional.to_pil_image.html |   2 +-
 ...on.transforms.v2.functional.to_tensor.html |   2 +-
 ...functional.uniform_temporal_subsample.html |   2 +-
 ...ransforms.v2.functional.vertical_flip.html |   2 +-
 ...chvision.tv_tensors.BoundingBoxFormat.html |   2 +-
 .../torchvision.tv_tensors.BoundingBoxes.html |   2 +-
 .../torchvision.tv_tensors.Image.html         |   2 +-
 .../torchvision.tv_tensors.Mask.html          |   2 +-
 .../torchvision.tv_tensors.TVTensor.html      |   2 +-
 .../torchvision.tv_tensors.Video.html         |   2 +-
 ...orchvision.tv_tensors.set_return_type.html |   2 +-
 .../torchvision.tv_tensors.wrap.html          |   2 +-
 ...torchvision.utils.draw_bounding_boxes.html |   2 +-
 .../torchvision.utils.draw_keypoints.html     |   2 +-
 ...hvision.utils.draw_segmentation_masks.html |   2 +-
 .../torchvision.utils.flow_to_image.html      |   2 +-
 .../torchvision.utils.make_grid.html          |   2 +-
 .../torchvision.utils.save_image.html         |   2 +-
 0.16/generated/video_table.html               |   2 +-
 0.16/genindex.html                            |   2 +-
 0.16/index.html                               |   2 +-
 0.16/io.html                                  |   2 +-
 0.16/models.html                              |   2 +-
 0.16/models/alexnet.html                      |   2 +-
 0.16/models/convnext.html                     |   2 +-
 0.16/models/deeplabv3.html                    |   2 +-
 0.16/models/densenet.html                     |   2 +-
 0.16/models/efficientnet.html                 |   2 +-
 0.16/models/efficientnetv2.html               |   2 +-
 0.16/models/faster_rcnn.html                  |   2 +-
 0.16/models/fcn.html                          |   2 +-
 0.16/models/fcos.html                         |   2 +-
 .../generated/torchvision.models.alexnet.html |   2 +-
 .../torchvision.models.convnext_base.html     |   2 +-
 .../torchvision.models.convnext_large.html    |   2 +-
 .../torchvision.models.convnext_small.html    |   2 +-
 .../torchvision.models.convnext_tiny.html     |   2 +-
 .../torchvision.models.densenet121.html       |   2 +-
 .../torchvision.models.densenet161.html       |   2 +-
 .../torchvision.models.densenet169.html       |   2 +-
 .../torchvision.models.densenet201.html       |   2 +-
 ...fasterrcnn_mobilenet_v3_large_320_fpn.html |   2 +-
 ...ion.fasterrcnn_mobilenet_v3_large_fpn.html |   2 +-
 ...els.detection.fasterrcnn_resnet50_fpn.html |   2 +-
 ....detection.fasterrcnn_resnet50_fpn_v2.html |   2 +-
 ...on.models.detection.fcos_resnet50_fpn.html |   2 +-
 ...s.detection.keypointrcnn_resnet50_fpn.html |   2 +-
 ...odels.detection.maskrcnn_resnet50_fpn.html |   2 +-
 ...ls.detection.maskrcnn_resnet50_fpn_v2.html |   2 +-
 ...dels.detection.retinanet_resnet50_fpn.html |   2 +-
 ...s.detection.retinanet_resnet50_fpn_v2.html |   2 +-
 ...hvision.models.detection.ssd300_vgg16.html |   2 +-
 ...tection.ssdlite320_mobilenet_v3_large.html |   2 +-
 .../torchvision.models.efficientnet_b0.html   |   2 +-
 .../torchvision.models.efficientnet_b1.html   |   2 +-
 .../torchvision.models.efficientnet_b2.html   |   2 +-
 .../torchvision.models.efficientnet_b3.html   |   2 +-
 .../torchvision.models.efficientnet_b4.html   |   2 +-
 .../torchvision.models.efficientnet_b5.html   |   2 +-
 .../torchvision.models.efficientnet_b6.html   |   2 +-
 .../torchvision.models.efficientnet_b7.html   |   2 +-
 .../torchvision.models.efficientnet_v2_l.html |   2 +-
 .../torchvision.models.efficientnet_v2_m.html |   2 +-
 .../torchvision.models.efficientnet_v2_s.html |   2 +-
 .../torchvision.models.googlenet.html         |   2 +-
 .../torchvision.models.inception_v3.html      |   2 +-
 .../torchvision.models.maxvit_t.html          |   2 +-
 .../torchvision.models.mnasnet0_5.html        |   2 +-
 .../torchvision.models.mnasnet0_75.html       |   2 +-
 .../torchvision.models.mnasnet1_0.html        |   2 +-
 .../torchvision.models.mnasnet1_3.html        |   2 +-
 .../torchvision.models.mobilenet_v2.html      |   2 +-
 ...torchvision.models.mobilenet_v3_large.html |   2 +-
 ...torchvision.models.mobilenet_v3_small.html |   2 +-
 ...vision.models.optical_flow.raft_large.html |   2 +-
 ...vision.models.optical_flow.raft_small.html |   2 +-
 ...hvision.models.quantization.googlenet.html |   2 +-
 ...sion.models.quantization.inception_v3.html |   2 +-
 ...sion.models.quantization.mobilenet_v2.html |   2 +-
 ...odels.quantization.mobilenet_v3_large.html |   2 +-
 ...chvision.models.quantization.resnet18.html |   2 +-
 ...chvision.models.quantization.resnet50.html |   2 +-
 ....models.quantization.resnext101_32x8d.html |   2 +-
 ....models.quantization.resnext101_64x4d.html |   2 +-
 ...odels.quantization.shufflenet_v2_x0_5.html |   2 +-
 ...odels.quantization.shufflenet_v2_x1_0.html |   2 +-
 ...odels.quantization.shufflenet_v2_x1_5.html |   2 +-
 ...odels.quantization.shufflenet_v2_x2_0.html |   2 +-
 .../torchvision.models.regnet_x_16gf.html     |   2 +-
 .../torchvision.models.regnet_x_1_6gf.html    |   2 +-
 .../torchvision.models.regnet_x_32gf.html     |   2 +-
 .../torchvision.models.regnet_x_3_2gf.html    |   2 +-
 .../torchvision.models.regnet_x_400mf.html    |   2 +-
 .../torchvision.models.regnet_x_800mf.html    |   2 +-
 .../torchvision.models.regnet_x_8gf.html      |   2 +-
 .../torchvision.models.regnet_y_128gf.html    |   2 +-
 .../torchvision.models.regnet_y_16gf.html     |   2 +-
 .../torchvision.models.regnet_y_1_6gf.html    |   2 +-
 .../torchvision.models.regnet_y_32gf.html     |   2 +-
 .../torchvision.models.regnet_y_3_2gf.html    |   2 +-
 .../torchvision.models.regnet_y_400mf.html    |   2 +-
 .../torchvision.models.regnet_y_800mf.html    |   2 +-
 .../torchvision.models.regnet_y_8gf.html      |   2 +-
 .../torchvision.models.resnet101.html         |   2 +-
 .../torchvision.models.resnet152.html         |   2 +-
 .../torchvision.models.resnet18.html          |   2 +-
 .../torchvision.models.resnet34.html          |   2 +-
 .../torchvision.models.resnet50.html          |   2 +-
 .../torchvision.models.resnext101_32x8d.html  |   2 +-
 .../torchvision.models.resnext101_64x4d.html  |   2 +-
 .../torchvision.models.resnext50_32x4d.html   |   2 +-
 ...entation.deeplabv3_mobilenet_v3_large.html |   2 +-
 ...dels.segmentation.deeplabv3_resnet101.html |   2 +-
 ...odels.segmentation.deeplabv3_resnet50.html |   2 +-
 ...ion.models.segmentation.fcn_resnet101.html |   2 +-
 ...sion.models.segmentation.fcn_resnet50.html |   2 +-
 ...egmentation.lraspp_mobilenet_v3_large.html |   2 +-
 ...torchvision.models.shufflenet_v2_x0_5.html |   2 +-
 ...torchvision.models.shufflenet_v2_x1_0.html |   2 +-
 ...torchvision.models.shufflenet_v2_x1_5.html |   2 +-
 ...torchvision.models.shufflenet_v2_x2_0.html |   2 +-
 .../torchvision.models.squeezenet1_0.html     |   2 +-
 .../torchvision.models.squeezenet1_1.html     |   2 +-
 .../generated/torchvision.models.swin_b.html  |   2 +-
 .../generated/torchvision.models.swin_s.html  |   2 +-
 .../generated/torchvision.models.swin_t.html  |   2 +-
 .../torchvision.models.swin_v2_b.html         |   2 +-
 .../torchvision.models.swin_v2_s.html         |   2 +-
 .../torchvision.models.swin_v2_t.html         |   2 +-
 .../generated/torchvision.models.vgg11.html   |   2 +-
 .../torchvision.models.vgg11_bn.html          |   2 +-
 .../generated/torchvision.models.vgg13.html   |   2 +-
 .../torchvision.models.vgg13_bn.html          |   2 +-
 .../generated/torchvision.models.vgg16.html   |   2 +-
 .../torchvision.models.vgg16_bn.html          |   2 +-
 .../generated/torchvision.models.vgg19.html   |   2 +-
 .../torchvision.models.vgg19_bn.html          |   2 +-
 .../torchvision.models.video.mc3_18.html      |   2 +-
 .../torchvision.models.video.mvit_v1_b.html   |   2 +-
 .../torchvision.models.video.mvit_v2_s.html   |   2 +-
 .../torchvision.models.video.r2plus1d_18.html |   2 +-
 .../torchvision.models.video.r3d_18.html      |   2 +-
 .../torchvision.models.video.s3d.html         |   2 +-
 .../torchvision.models.video.swin3d_b.html    |   2 +-
 .../torchvision.models.video.swin3d_s.html    |   2 +-
 .../torchvision.models.video.swin3d_t.html    |   2 +-
 .../torchvision.models.vit_b_16.html          |   2 +-
 .../torchvision.models.vit_b_32.html          |   2 +-
 .../torchvision.models.vit_h_14.html          |   2 +-
 .../torchvision.models.vit_l_16.html          |   2 +-
 .../torchvision.models.vit_l_32.html          |   2 +-
 .../torchvision.models.wide_resnet101_2.html  |   2 +-
 .../torchvision.models.wide_resnet50_2.html   |   2 +-
 0.16/models/googlenet.html                    |   2 +-
 0.16/models/googlenet_quant.html              |   2 +-
 0.16/models/inception.html                    |   2 +-
 0.16/models/inception_quant.html              |   2 +-
 0.16/models/keypoint_rcnn.html                |   2 +-
 0.16/models/lraspp.html                       |   2 +-
 0.16/models/mask_rcnn.html                    |   2 +-
 0.16/models/maxvit.html                       |   2 +-
 0.16/models/mnasnet.html                      |   2 +-
 0.16/models/mobilenetv2.html                  |   2 +-
 0.16/models/mobilenetv2_quant.html            |   2 +-
 0.16/models/mobilenetv3.html                  |   2 +-
 0.16/models/mobilenetv3_quant.html            |   2 +-
 0.16/models/raft.html                         |   2 +-
 0.16/models/regnet.html                       |   2 +-
 0.16/models/resnet.html                       |   2 +-
 0.16/models/resnet_quant.html                 |   2 +-
 0.16/models/resnext.html                      |   2 +-
 0.16/models/resnext_quant.html                |   2 +-
 0.16/models/retinanet.html                    |   2 +-
 0.16/models/shufflenetv2.html                 |   2 +-
 0.16/models/shufflenetv2_quant.html           |   2 +-
 0.16/models/squeezenet.html                   |   2 +-
 0.16/models/ssd.html                          |   2 +-
 0.16/models/ssdlite.html                      |   2 +-
 0.16/models/swin_transformer.html             |   2 +-
 0.16/models/vgg.html                          |   2 +-
 0.16/models/video_mvit.html                   |   2 +-
 0.16/models/video_resnet.html                 |   2 +-
 0.16/models/video_s3d.html                    |   2 +-
 0.16/models/video_swin_transformer.html       |   2 +-
 0.16/models/vision_transformer.html           |   2 +-
 0.16/models/wide_resnet.html                  |   2 +-
 0.16/objects.inv                              | Bin 18596 -> 18579 bytes
 0.16/ops.html                                 |   2 +-
 0.16/py-modindex.html                         |   2 +-
 0.16/search.html                              |   2 +-
 0.16/training_references.html                 |   2 +-
 0.16/transforms.html                          |   2 +-
 0.16/tv_tensors.html                          |   2 +-
 0.16/utils.html                               |   2 +-
 651 files changed, 650 insertions(+), 650 deletions(-)

diff --git a/0.16/_modules/index.html b/0.16/_modules/index.html
index b46f8605d2a..9ae8c6061cf 100644
--- a/0.16/_modules/index.html
+++ b/0.16/_modules/index.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision.html b/0.16/_modules/torchvision.html
index 6bb9a84cb83..31ab88e7acb 100644
--- a/0.16/_modules/torchvision.html
+++ b/0.16/_modules/torchvision.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/_optical_flow.html b/0.16/_modules/torchvision/datasets/_optical_flow.html
index a1b2e904968..ba7215695b4 100644
--- a/0.16/_modules/torchvision/datasets/_optical_flow.html
+++ b/0.16/_modules/torchvision/datasets/_optical_flow.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/_stereo_matching.html b/0.16/_modules/torchvision/datasets/_stereo_matching.html
index 2b74e331e6f..5da3cb9be98 100644
--- a/0.16/_modules/torchvision/datasets/_stereo_matching.html
+++ b/0.16/_modules/torchvision/datasets/_stereo_matching.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/caltech.html b/0.16/_modules/torchvision/datasets/caltech.html
index 9806c9cb7a4..d71d0a208cf 100644
--- a/0.16/_modules/torchvision/datasets/caltech.html
+++ b/0.16/_modules/torchvision/datasets/caltech.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/celeba.html b/0.16/_modules/torchvision/datasets/celeba.html
index abbfeb4f9d8..0170e246a59 100644
--- a/0.16/_modules/torchvision/datasets/celeba.html
+++ b/0.16/_modules/torchvision/datasets/celeba.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/cifar.html b/0.16/_modules/torchvision/datasets/cifar.html
index 94d4fb8f45a..d865cf848a5 100644
--- a/0.16/_modules/torchvision/datasets/cifar.html
+++ b/0.16/_modules/torchvision/datasets/cifar.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/cityscapes.html b/0.16/_modules/torchvision/datasets/cityscapes.html
index 19ff5de0170..acf3baf711d 100644
--- a/0.16/_modules/torchvision/datasets/cityscapes.html
+++ b/0.16/_modules/torchvision/datasets/cityscapes.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/coco.html b/0.16/_modules/torchvision/datasets/coco.html
index 7e8d3f3916f..ed14c2920bb 100644
--- a/0.16/_modules/torchvision/datasets/coco.html
+++ b/0.16/_modules/torchvision/datasets/coco.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/country211.html b/0.16/_modules/torchvision/datasets/country211.html
index 3974110fadb..78127f1ce23 100644
--- a/0.16/_modules/torchvision/datasets/country211.html
+++ b/0.16/_modules/torchvision/datasets/country211.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/dtd.html b/0.16/_modules/torchvision/datasets/dtd.html
index 9b7bd8d07c4..b807c325d6c 100644
--- a/0.16/_modules/torchvision/datasets/dtd.html
+++ b/0.16/_modules/torchvision/datasets/dtd.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/eurosat.html b/0.16/_modules/torchvision/datasets/eurosat.html
index 4b688af2cb2..c3462bb629e 100644
--- a/0.16/_modules/torchvision/datasets/eurosat.html
+++ b/0.16/_modules/torchvision/datasets/eurosat.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/fakedata.html b/0.16/_modules/torchvision/datasets/fakedata.html
index a87ceac3e38..e708958abc5 100644
--- a/0.16/_modules/torchvision/datasets/fakedata.html
+++ b/0.16/_modules/torchvision/datasets/fakedata.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/fer2013.html b/0.16/_modules/torchvision/datasets/fer2013.html
index e85f381f5a3..643f4ad750d 100644
--- a/0.16/_modules/torchvision/datasets/fer2013.html
+++ b/0.16/_modules/torchvision/datasets/fer2013.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/fgvc_aircraft.html b/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
index c7ce59f1e0b..81da480411a 100644
--- a/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
+++ b/0.16/_modules/torchvision/datasets/fgvc_aircraft.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/flickr.html b/0.16/_modules/torchvision/datasets/flickr.html
index 38f229040ac..ec877f7b826 100644
--- a/0.16/_modules/torchvision/datasets/flickr.html
+++ b/0.16/_modules/torchvision/datasets/flickr.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/flowers102.html b/0.16/_modules/torchvision/datasets/flowers102.html
index d4832c29e95..c6f2fefb884 100644
--- a/0.16/_modules/torchvision/datasets/flowers102.html
+++ b/0.16/_modules/torchvision/datasets/flowers102.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/folder.html b/0.16/_modules/torchvision/datasets/folder.html
index b2a1a2e1d50..24c7c0fe5bd 100644
--- a/0.16/_modules/torchvision/datasets/folder.html
+++ b/0.16/_modules/torchvision/datasets/folder.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/food101.html b/0.16/_modules/torchvision/datasets/food101.html
index 205c5792aa7..00f7bc8c4ba 100644
--- a/0.16/_modules/torchvision/datasets/food101.html
+++ b/0.16/_modules/torchvision/datasets/food101.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/gtsrb.html b/0.16/_modules/torchvision/datasets/gtsrb.html
index 41d3986c306..7ea265edb13 100644
--- a/0.16/_modules/torchvision/datasets/gtsrb.html
+++ b/0.16/_modules/torchvision/datasets/gtsrb.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/hmdb51.html b/0.16/_modules/torchvision/datasets/hmdb51.html
index ac46e953395..07cccacf953 100644
--- a/0.16/_modules/torchvision/datasets/hmdb51.html
+++ b/0.16/_modules/torchvision/datasets/hmdb51.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/imagenet.html b/0.16/_modules/torchvision/datasets/imagenet.html
index c87fe52465f..cbf7b0a0da8 100644
--- a/0.16/_modules/torchvision/datasets/imagenet.html
+++ b/0.16/_modules/torchvision/datasets/imagenet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/inaturalist.html b/0.16/_modules/torchvision/datasets/inaturalist.html
index cb4a896ae34..023763a47c3 100644
--- a/0.16/_modules/torchvision/datasets/inaturalist.html
+++ b/0.16/_modules/torchvision/datasets/inaturalist.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/kinetics.html b/0.16/_modules/torchvision/datasets/kinetics.html
index 9c3e13a5d28..40917fc5c2a 100644
--- a/0.16/_modules/torchvision/datasets/kinetics.html
+++ b/0.16/_modules/torchvision/datasets/kinetics.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/kitti.html b/0.16/_modules/torchvision/datasets/kitti.html
index cff517c4c61..91e4530138b 100644
--- a/0.16/_modules/torchvision/datasets/kitti.html
+++ b/0.16/_modules/torchvision/datasets/kitti.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/lfw.html b/0.16/_modules/torchvision/datasets/lfw.html
index 0b2b0b6d4ed..fc46c1a8e19 100644
--- a/0.16/_modules/torchvision/datasets/lfw.html
+++ b/0.16/_modules/torchvision/datasets/lfw.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/lsun.html b/0.16/_modules/torchvision/datasets/lsun.html
index 620cf680144..4ef0737f143 100644
--- a/0.16/_modules/torchvision/datasets/lsun.html
+++ b/0.16/_modules/torchvision/datasets/lsun.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/mnist.html b/0.16/_modules/torchvision/datasets/mnist.html
index 7f749e8666d..0cd48c55676 100644
--- a/0.16/_modules/torchvision/datasets/mnist.html
+++ b/0.16/_modules/torchvision/datasets/mnist.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/moving_mnist.html b/0.16/_modules/torchvision/datasets/moving_mnist.html
index bb0a65c4b3c..11f11e1f837 100644
--- a/0.16/_modules/torchvision/datasets/moving_mnist.html
+++ b/0.16/_modules/torchvision/datasets/moving_mnist.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/omniglot.html b/0.16/_modules/torchvision/datasets/omniglot.html
index 29d2b0e361f..971d85975af 100644
--- a/0.16/_modules/torchvision/datasets/omniglot.html
+++ b/0.16/_modules/torchvision/datasets/omniglot.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html b/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
index 081170449fb..e531e8e245d 100644
--- a/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
+++ b/0.16/_modules/torchvision/datasets/oxford_iiit_pet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/pcam.html b/0.16/_modules/torchvision/datasets/pcam.html
index a348fa29f92..4535e146ef8 100644
--- a/0.16/_modules/torchvision/datasets/pcam.html
+++ b/0.16/_modules/torchvision/datasets/pcam.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/phototour.html b/0.16/_modules/torchvision/datasets/phototour.html
index fb96c141c06..46b7368006f 100644
--- a/0.16/_modules/torchvision/datasets/phototour.html
+++ b/0.16/_modules/torchvision/datasets/phototour.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/places365.html b/0.16/_modules/torchvision/datasets/places365.html
index 86730d25ba6..c838d4aea94 100644
--- a/0.16/_modules/torchvision/datasets/places365.html
+++ b/0.16/_modules/torchvision/datasets/places365.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/rendered_sst2.html b/0.16/_modules/torchvision/datasets/rendered_sst2.html
index 1df536a560f..f35c4b095ca 100644
--- a/0.16/_modules/torchvision/datasets/rendered_sst2.html
+++ b/0.16/_modules/torchvision/datasets/rendered_sst2.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/sbd.html b/0.16/_modules/torchvision/datasets/sbd.html
index 6cea539dcd4..740e357404f 100644
--- a/0.16/_modules/torchvision/datasets/sbd.html
+++ b/0.16/_modules/torchvision/datasets/sbd.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/sbu.html b/0.16/_modules/torchvision/datasets/sbu.html
index ebf51754f3d..2cd065d9c6f 100644
--- a/0.16/_modules/torchvision/datasets/sbu.html
+++ b/0.16/_modules/torchvision/datasets/sbu.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/semeion.html b/0.16/_modules/torchvision/datasets/semeion.html
index dd67e2dc2f5..a2bdc17c11d 100644
--- a/0.16/_modules/torchvision/datasets/semeion.html
+++ b/0.16/_modules/torchvision/datasets/semeion.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/stanford_cars.html b/0.16/_modules/torchvision/datasets/stanford_cars.html
index a7dbbc9f78d..8707ab7691b 100644
--- a/0.16/_modules/torchvision/datasets/stanford_cars.html
+++ b/0.16/_modules/torchvision/datasets/stanford_cars.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/stl10.html b/0.16/_modules/torchvision/datasets/stl10.html
index 4893576d91d..cc83b6a30c2 100644
--- a/0.16/_modules/torchvision/datasets/stl10.html
+++ b/0.16/_modules/torchvision/datasets/stl10.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/sun397.html b/0.16/_modules/torchvision/datasets/sun397.html
index d30d8024744..7f2235556cb 100644
--- a/0.16/_modules/torchvision/datasets/sun397.html
+++ b/0.16/_modules/torchvision/datasets/sun397.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/svhn.html b/0.16/_modules/torchvision/datasets/svhn.html
index a76e45206a1..980240d5968 100644
--- a/0.16/_modules/torchvision/datasets/svhn.html
+++ b/0.16/_modules/torchvision/datasets/svhn.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/ucf101.html b/0.16/_modules/torchvision/datasets/ucf101.html
index a8ebf4e1080..90234671dfe 100644
--- a/0.16/_modules/torchvision/datasets/ucf101.html
+++ b/0.16/_modules/torchvision/datasets/ucf101.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/usps.html b/0.16/_modules/torchvision/datasets/usps.html
index 24f7f1400fe..b3a5bda62ae 100644
--- a/0.16/_modules/torchvision/datasets/usps.html
+++ b/0.16/_modules/torchvision/datasets/usps.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/vision.html b/0.16/_modules/torchvision/datasets/vision.html
index aeab9b7a7fe..43c1f25e212 100644
--- a/0.16/_modules/torchvision/datasets/vision.html
+++ b/0.16/_modules/torchvision/datasets/vision.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/voc.html b/0.16/_modules/torchvision/datasets/voc.html
index 440910e19c6..a9acfb4275b 100644
--- a/0.16/_modules/torchvision/datasets/voc.html
+++ b/0.16/_modules/torchvision/datasets/voc.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/datasets/widerface.html b/0.16/_modules/torchvision/datasets/widerface.html
index 74bd6f1e7e8..9d03e893c82 100644
--- a/0.16/_modules/torchvision/datasets/widerface.html
+++ b/0.16/_modules/torchvision/datasets/widerface.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/io/image.html b/0.16/_modules/torchvision/io/image.html
index 846a5c22d6f..afa1edc8a3b 100644
--- a/0.16/_modules/torchvision/io/image.html
+++ b/0.16/_modules/torchvision/io/image.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/io/video.html b/0.16/_modules/torchvision/io/video.html
index b7144d0a22d..8474591287f 100644
--- a/0.16/_modules/torchvision/io/video.html
+++ b/0.16/_modules/torchvision/io/video.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/io/video_reader.html b/0.16/_modules/torchvision/io/video_reader.html
index 06f12157ece..7766a12bb19 100644
--- a/0.16/_modules/torchvision/io/video_reader.html
+++ b/0.16/_modules/torchvision/io/video_reader.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/_api.html b/0.16/_modules/torchvision/models/_api.html
index 427ca9cef3e..aedbab4c204 100644
--- a/0.16/_modules/torchvision/models/_api.html
+++ b/0.16/_modules/torchvision/models/_api.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/alexnet.html b/0.16/_modules/torchvision/models/alexnet.html
index 28801ab45ca..2cf7d7dc4e0 100644
--- a/0.16/_modules/torchvision/models/alexnet.html
+++ b/0.16/_modules/torchvision/models/alexnet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/convnext.html b/0.16/_modules/torchvision/models/convnext.html
index d868af7f129..76ba8015466 100644
--- a/0.16/_modules/torchvision/models/convnext.html
+++ b/0.16/_modules/torchvision/models/convnext.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/densenet.html b/0.16/_modules/torchvision/models/densenet.html
index 557181a3485..48a1640434d 100644
--- a/0.16/_modules/torchvision/models/densenet.html
+++ b/0.16/_modules/torchvision/models/densenet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/detection/faster_rcnn.html b/0.16/_modules/torchvision/models/detection/faster_rcnn.html
index 674c841c090..9c0c130a7ae 100644
--- a/0.16/_modules/torchvision/models/detection/faster_rcnn.html
+++ b/0.16/_modules/torchvision/models/detection/faster_rcnn.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/detection/fcos.html b/0.16/_modules/torchvision/models/detection/fcos.html
index 5e2d881560f..041f55cfdc8 100644
--- a/0.16/_modules/torchvision/models/detection/fcos.html
+++ b/0.16/_modules/torchvision/models/detection/fcos.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html b/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
index a56e9116fb0..b65dffd5177 100644
--- a/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
+++ b/0.16/_modules/torchvision/models/detection/keypoint_rcnn.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/detection/mask_rcnn.html b/0.16/_modules/torchvision/models/detection/mask_rcnn.html
index 2f6b6bf2152..f5951ea88f6 100644
--- a/0.16/_modules/torchvision/models/detection/mask_rcnn.html
+++ b/0.16/_modules/torchvision/models/detection/mask_rcnn.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/detection/retinanet.html b/0.16/_modules/torchvision/models/detection/retinanet.html
index f111582260e..6f42d52e9aa 100644
--- a/0.16/_modules/torchvision/models/detection/retinanet.html
+++ b/0.16/_modules/torchvision/models/detection/retinanet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/detection/ssd.html b/0.16/_modules/torchvision/models/detection/ssd.html
index 46768273270..d13ebe41e86 100644
--- a/0.16/_modules/torchvision/models/detection/ssd.html
+++ b/0.16/_modules/torchvision/models/detection/ssd.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/detection/ssdlite.html b/0.16/_modules/torchvision/models/detection/ssdlite.html
index c3f99923b36..64dab8a9051 100644
--- a/0.16/_modules/torchvision/models/detection/ssdlite.html
+++ b/0.16/_modules/torchvision/models/detection/ssdlite.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/efficientnet.html b/0.16/_modules/torchvision/models/efficientnet.html
index 82ce6575f06..619d1d12fa8 100644
--- a/0.16/_modules/torchvision/models/efficientnet.html
+++ b/0.16/_modules/torchvision/models/efficientnet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/feature_extraction.html b/0.16/_modules/torchvision/models/feature_extraction.html
index 3a91421c9a9..ea86aeb5633 100644
--- a/0.16/_modules/torchvision/models/feature_extraction.html
+++ b/0.16/_modules/torchvision/models/feature_extraction.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/googlenet.html b/0.16/_modules/torchvision/models/googlenet.html
index 434e91846f9..0102ee9ae66 100644
--- a/0.16/_modules/torchvision/models/googlenet.html
+++ b/0.16/_modules/torchvision/models/googlenet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/inception.html b/0.16/_modules/torchvision/models/inception.html
index c09ebd4422d..ee04e102987 100644
--- a/0.16/_modules/torchvision/models/inception.html
+++ b/0.16/_modules/torchvision/models/inception.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/maxvit.html b/0.16/_modules/torchvision/models/maxvit.html
index 06f20c4dd70..4c608d8983b 100644
--- a/0.16/_modules/torchvision/models/maxvit.html
+++ b/0.16/_modules/torchvision/models/maxvit.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/mnasnet.html b/0.16/_modules/torchvision/models/mnasnet.html
index 80dafb35a9c..c4a1a911a0a 100644
--- a/0.16/_modules/torchvision/models/mnasnet.html
+++ b/0.16/_modules/torchvision/models/mnasnet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/mobilenetv2.html b/0.16/_modules/torchvision/models/mobilenetv2.html
index 04f2a378b6e..86ff89b2404 100644
--- a/0.16/_modules/torchvision/models/mobilenetv2.html
+++ b/0.16/_modules/torchvision/models/mobilenetv2.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/mobilenetv3.html b/0.16/_modules/torchvision/models/mobilenetv3.html
index 24f0339a21a..f868d92b337 100644
--- a/0.16/_modules/torchvision/models/mobilenetv3.html
+++ b/0.16/_modules/torchvision/models/mobilenetv3.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/optical_flow/raft.html b/0.16/_modules/torchvision/models/optical_flow/raft.html
index 22aa287546e..700d074b620 100644
--- a/0.16/_modules/torchvision/models/optical_flow/raft.html
+++ b/0.16/_modules/torchvision/models/optical_flow/raft.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/quantization/googlenet.html b/0.16/_modules/torchvision/models/quantization/googlenet.html
index 174885ba9ee..2c9957d5ed6 100644
--- a/0.16/_modules/torchvision/models/quantization/googlenet.html
+++ b/0.16/_modules/torchvision/models/quantization/googlenet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/quantization/inception.html b/0.16/_modules/torchvision/models/quantization/inception.html
index ae2cfdd8bcb..f0b6a7c091d 100644
--- a/0.16/_modules/torchvision/models/quantization/inception.html
+++ b/0.16/_modules/torchvision/models/quantization/inception.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/quantization/mobilenetv2.html b/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
index d54f6bb32e4..d0aefd10c0f 100644
--- a/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
+++ b/0.16/_modules/torchvision/models/quantization/mobilenetv2.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/quantization/mobilenetv3.html b/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
index c987ec1a3b4..0356b8226b6 100644
--- a/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
+++ b/0.16/_modules/torchvision/models/quantization/mobilenetv3.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/quantization/resnet.html b/0.16/_modules/torchvision/models/quantization/resnet.html
index 78b97ea5143..2c43cb394fb 100644
--- a/0.16/_modules/torchvision/models/quantization/resnet.html
+++ b/0.16/_modules/torchvision/models/quantization/resnet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/quantization/shufflenetv2.html b/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
index f6e6f7de5ac..dcadeab5ed9 100644
--- a/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
+++ b/0.16/_modules/torchvision/models/quantization/shufflenetv2.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/regnet.html b/0.16/_modules/torchvision/models/regnet.html
index 6e873d47cba..91500593ff5 100644
--- a/0.16/_modules/torchvision/models/regnet.html
+++ b/0.16/_modules/torchvision/models/regnet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/resnet.html b/0.16/_modules/torchvision/models/resnet.html
index 2fddfff96c8..e135e46a397 100644
--- a/0.16/_modules/torchvision/models/resnet.html
+++ b/0.16/_modules/torchvision/models/resnet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/segmentation/deeplabv3.html b/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
index 09be854330d..8bcca8671a1 100644
--- a/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
+++ b/0.16/_modules/torchvision/models/segmentation/deeplabv3.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/segmentation/fcn.html b/0.16/_modules/torchvision/models/segmentation/fcn.html
index f91d1efc553..b5246b9bbba 100644
--- a/0.16/_modules/torchvision/models/segmentation/fcn.html
+++ b/0.16/_modules/torchvision/models/segmentation/fcn.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/segmentation/lraspp.html b/0.16/_modules/torchvision/models/segmentation/lraspp.html
index 095d30ec7f2..618720aa2f5 100644
--- a/0.16/_modules/torchvision/models/segmentation/lraspp.html
+++ b/0.16/_modules/torchvision/models/segmentation/lraspp.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/shufflenetv2.html b/0.16/_modules/torchvision/models/shufflenetv2.html
index 7c0dadca7c8..14ea29aa06b 100644
--- a/0.16/_modules/torchvision/models/shufflenetv2.html
+++ b/0.16/_modules/torchvision/models/shufflenetv2.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/squeezenet.html b/0.16/_modules/torchvision/models/squeezenet.html
index 9ca784dd038..add56164262 100644
--- a/0.16/_modules/torchvision/models/squeezenet.html
+++ b/0.16/_modules/torchvision/models/squeezenet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/swin_transformer.html b/0.16/_modules/torchvision/models/swin_transformer.html
index a47b4ac9896..7d47d4597f4 100644
--- a/0.16/_modules/torchvision/models/swin_transformer.html
+++ b/0.16/_modules/torchvision/models/swin_transformer.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/vgg.html b/0.16/_modules/torchvision/models/vgg.html
index 56af2b05b26..2506db1df71 100644
--- a/0.16/_modules/torchvision/models/vgg.html
+++ b/0.16/_modules/torchvision/models/vgg.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/video/mvit.html b/0.16/_modules/torchvision/models/video/mvit.html
index 0a0ad19e3b1..11edbd7591c 100644
--- a/0.16/_modules/torchvision/models/video/mvit.html
+++ b/0.16/_modules/torchvision/models/video/mvit.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/video/resnet.html b/0.16/_modules/torchvision/models/video/resnet.html
index 877eec9fa4c..3eae17e4d06 100644
--- a/0.16/_modules/torchvision/models/video/resnet.html
+++ b/0.16/_modules/torchvision/models/video/resnet.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/video/s3d.html b/0.16/_modules/torchvision/models/video/s3d.html
index d4dd2cf2290..b1c9c5b758e 100644
--- a/0.16/_modules/torchvision/models/video/s3d.html
+++ b/0.16/_modules/torchvision/models/video/s3d.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/video/swin_transformer.html b/0.16/_modules/torchvision/models/video/swin_transformer.html
index b8150796187..b0684b92eaa 100644
--- a/0.16/_modules/torchvision/models/video/swin_transformer.html
+++ b/0.16/_modules/torchvision/models/video/swin_transformer.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/models/vision_transformer.html b/0.16/_modules/torchvision/models/vision_transformer.html
index 9ac371a66f7..a9874bed538 100644
--- a/0.16/_modules/torchvision/models/vision_transformer.html
+++ b/0.16/_modules/torchvision/models/vision_transformer.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/boxes.html b/0.16/_modules/torchvision/ops/boxes.html
index 03995e6ff55..a30cacca3d1 100644
--- a/0.16/_modules/torchvision/ops/boxes.html
+++ b/0.16/_modules/torchvision/ops/boxes.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/ciou_loss.html b/0.16/_modules/torchvision/ops/ciou_loss.html
index 954c0f54059..7d42c1e647a 100644
--- a/0.16/_modules/torchvision/ops/ciou_loss.html
+++ b/0.16/_modules/torchvision/ops/ciou_loss.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/deform_conv.html b/0.16/_modules/torchvision/ops/deform_conv.html
index a1a66b1ddcf..05bc954b067 100644
--- a/0.16/_modules/torchvision/ops/deform_conv.html
+++ b/0.16/_modules/torchvision/ops/deform_conv.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/diou_loss.html b/0.16/_modules/torchvision/ops/diou_loss.html
index 74edead1a95..5c521c7955f 100644
--- a/0.16/_modules/torchvision/ops/diou_loss.html
+++ b/0.16/_modules/torchvision/ops/diou_loss.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/drop_block.html b/0.16/_modules/torchvision/ops/drop_block.html
index 14c159919cc..482b91f390e 100644
--- a/0.16/_modules/torchvision/ops/drop_block.html
+++ b/0.16/_modules/torchvision/ops/drop_block.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/feature_pyramid_network.html b/0.16/_modules/torchvision/ops/feature_pyramid_network.html
index 6f6be47dad4..ecfcd21100f 100644
--- a/0.16/_modules/torchvision/ops/feature_pyramid_network.html
+++ b/0.16/_modules/torchvision/ops/feature_pyramid_network.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/focal_loss.html b/0.16/_modules/torchvision/ops/focal_loss.html
index a596b740e0a..d0a0b8ae610 100644
--- a/0.16/_modules/torchvision/ops/focal_loss.html
+++ b/0.16/_modules/torchvision/ops/focal_loss.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/giou_loss.html b/0.16/_modules/torchvision/ops/giou_loss.html
index 3dbc8ea1e22..3fa9cb5e36f 100644
--- a/0.16/_modules/torchvision/ops/giou_loss.html
+++ b/0.16/_modules/torchvision/ops/giou_loss.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/misc.html b/0.16/_modules/torchvision/ops/misc.html
index 673a8106fa5..6282bf9e60e 100644
--- a/0.16/_modules/torchvision/ops/misc.html
+++ b/0.16/_modules/torchvision/ops/misc.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/poolers.html b/0.16/_modules/torchvision/ops/poolers.html
index 760114e5862..75d8c6c00dd 100644
--- a/0.16/_modules/torchvision/ops/poolers.html
+++ b/0.16/_modules/torchvision/ops/poolers.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/ps_roi_align.html b/0.16/_modules/torchvision/ops/ps_roi_align.html
index 30704d941ce..448b3ef7b30 100644
--- a/0.16/_modules/torchvision/ops/ps_roi_align.html
+++ b/0.16/_modules/torchvision/ops/ps_roi_align.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/ps_roi_pool.html b/0.16/_modules/torchvision/ops/ps_roi_pool.html
index f19d2461023..f1aac891545 100644
--- a/0.16/_modules/torchvision/ops/ps_roi_pool.html
+++ b/0.16/_modules/torchvision/ops/ps_roi_pool.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/roi_align.html b/0.16/_modules/torchvision/ops/roi_align.html
index da03c86541c..23674255e2e 100644
--- a/0.16/_modules/torchvision/ops/roi_align.html
+++ b/0.16/_modules/torchvision/ops/roi_align.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/roi_pool.html b/0.16/_modules/torchvision/ops/roi_pool.html
index 985389adfe4..c1af5224659 100644
--- a/0.16/_modules/torchvision/ops/roi_pool.html
+++ b/0.16/_modules/torchvision/ops/roi_pool.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/ops/stochastic_depth.html b/0.16/_modules/torchvision/ops/stochastic_depth.html
index c11421050b7..b04823970a8 100644
--- a/0.16/_modules/torchvision/ops/stochastic_depth.html
+++ b/0.16/_modules/torchvision/ops/stochastic_depth.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/autoaugment.html b/0.16/_modules/torchvision/transforms/autoaugment.html
index 3289088e495..526770533fd 100644
--- a/0.16/_modules/torchvision/transforms/autoaugment.html
+++ b/0.16/_modules/torchvision/transforms/autoaugment.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/functional.html b/0.16/_modules/torchvision/transforms/functional.html
index 8f47497ed53..7afaaf35cc0 100644
--- a/0.16/_modules/torchvision/transforms/functional.html
+++ b/0.16/_modules/torchvision/transforms/functional.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/transforms.html b/0.16/_modules/torchvision/transforms/transforms.html
index 78dd6d39024..31aef1d09b9 100644
--- a/0.16/_modules/torchvision/transforms/transforms.html
+++ b/0.16/_modules/torchvision/transforms/transforms.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_augment.html b/0.16/_modules/torchvision/transforms/v2/_augment.html
index da7bc499536..e68b8c44e2b 100644
--- a/0.16/_modules/torchvision/transforms/v2/_augment.html
+++ b/0.16/_modules/torchvision/transforms/v2/_augment.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_auto_augment.html b/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
index 5313f54fb57..e1c9ed99b09 100644
--- a/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
+++ b/0.16/_modules/torchvision/transforms/v2/_auto_augment.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_color.html b/0.16/_modules/torchvision/transforms/v2/_color.html
index 83ab292fc1b..e026dea9145 100644
--- a/0.16/_modules/torchvision/transforms/v2/_color.html
+++ b/0.16/_modules/torchvision/transforms/v2/_color.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_container.html b/0.16/_modules/torchvision/transforms/v2/_container.html
index adc05acca65..dd148a0c067 100644
--- a/0.16/_modules/torchvision/transforms/v2/_container.html
+++ b/0.16/_modules/torchvision/transforms/v2/_container.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_deprecated.html b/0.16/_modules/torchvision/transforms/v2/_deprecated.html
index 00dd42168ba..85e1371e868 100644
--- a/0.16/_modules/torchvision/transforms/v2/_deprecated.html
+++ b/0.16/_modules/torchvision/transforms/v2/_deprecated.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_geometry.html b/0.16/_modules/torchvision/transforms/v2/_geometry.html
index 4cf36db1748..45667e7598b 100644
--- a/0.16/_modules/torchvision/transforms/v2/_geometry.html
+++ b/0.16/_modules/torchvision/transforms/v2/_geometry.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_meta.html b/0.16/_modules/torchvision/transforms/v2/_meta.html
index 08a4da98e33..9bb8dfcf2d0 100644
--- a/0.16/_modules/torchvision/transforms/v2/_meta.html
+++ b/0.16/_modules/torchvision/transforms/v2/_meta.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_misc.html b/0.16/_modules/torchvision/transforms/v2/_misc.html
index 7ad1c9cdf6f..abeaae03f96 100644
--- a/0.16/_modules/torchvision/transforms/v2/_misc.html
+++ b/0.16/_modules/torchvision/transforms/v2/_misc.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_temporal.html b/0.16/_modules/torchvision/transforms/v2/_temporal.html
index 1305955d813..f2412834be8 100644
--- a/0.16/_modules/torchvision/transforms/v2/_temporal.html
+++ b/0.16/_modules/torchvision/transforms/v2/_temporal.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/_type_conversion.html b/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
index 5a12f1dc053..78e87bba3f1 100644
--- a/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
+++ b/0.16/_modules/torchvision/transforms/v2/_type_conversion.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_augment.html b/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
index 7ff068bea14..be6d5ce30ab 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_augment.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_color.html b/0.16/_modules/torchvision/transforms/v2/functional/_color.html
index 7f6f3daa166..0e8adae4af2 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_color.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_color.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html b/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
index 97d49f332c5..a4cb7eeb6aa 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_deprecated.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html b/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
index 1f8315ce3e3..9cea27220ec 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_geometry.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_meta.html b/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
index 4f46ab01fb1..c623d262bd6 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_meta.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_misc.html b/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
index 72a6c571c39..b49482659ca 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_misc.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html b/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
index 11fb9afb9e9..9867f7ce527 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_temporal.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html b/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
index 610f62a7c45..28d151c8baa 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_type_conversion.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/transforms/v2/functional/_utils.html b/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
index 880a2b700dd..f8c745cd011 100644
--- a/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
+++ b/0.16/_modules/torchvision/transforms/v2/functional/_utils.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors.html b/0.16/_modules/torchvision/tv_tensors.html
index 83caa48bfd1..21d4f9facfa 100644
--- a/0.16/_modules/torchvision/tv_tensors.html
+++ b/0.16/_modules/torchvision/tv_tensors.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html b/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
index ffde8aa72d7..f640e753c47 100644
--- a/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
+++ b/0.16/_modules/torchvision/tv_tensors/_bounding_boxes.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html b/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
index 638d4450651..51a19e5287a 100644
--- a/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
+++ b/0.16/_modules/torchvision/tv_tensors/_dataset_wrapper.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors/_image.html b/0.16/_modules/torchvision/tv_tensors/_image.html
index f393eceab21..47325365880 100644
--- a/0.16/_modules/torchvision/tv_tensors/_image.html
+++ b/0.16/_modules/torchvision/tv_tensors/_image.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors/_mask.html b/0.16/_modules/torchvision/tv_tensors/_mask.html
index 2b1a1ceea70..601ddcef21d 100644
--- a/0.16/_modules/torchvision/tv_tensors/_mask.html
+++ b/0.16/_modules/torchvision/tv_tensors/_mask.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html b/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
index 00faede710b..4a126740484 100644
--- a/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
+++ b/0.16/_modules/torchvision/tv_tensors/_torch_function_helpers.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html b/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
index bf6cd60bca6..b8567975314 100644
--- a/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
+++ b/0.16/_modules/torchvision/tv_tensors/_tv_tensor.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/tv_tensors/_video.html b/0.16/_modules/torchvision/tv_tensors/_video.html
index ab603b4e50a..b7c92de539b 100644
--- a/0.16/_modules/torchvision/tv_tensors/_video.html
+++ b/0.16/_modules/torchvision/tv_tensors/_video.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/_modules/torchvision/utils.html b/0.16/_modules/torchvision/utils.html
index 1ce0cb9be10..5b53b11d412 100644
--- a/0.16/_modules/torchvision/utils.html
+++ b/0.16/_modules/torchvision/utils.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/index.html b/0.16/auto_examples/index.html
index 647e2944c91..b567e851346 100644
--- a/0.16/auto_examples/index.html
+++ b/0.16/auto_examples/index.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/others/index.html b/0.16/auto_examples/others/index.html
index b58a5431270..ca8fbd13184 100644
--- a/0.16/auto_examples/others/index.html
+++ b/0.16/auto_examples/others/index.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/others/plot_optical_flow.html b/0.16/auto_examples/others/plot_optical_flow.html
index 7e5409be75b..4b7a53f6d4c 100644
--- a/0.16/auto_examples/others/plot_optical_flow.html
+++ b/0.16/auto_examples/others/plot_optical_flow.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/others/plot_repurposing_annotations.html b/0.16/auto_examples/others/plot_repurposing_annotations.html
index 3e99c57bb85..ead46b54451 100644
--- a/0.16/auto_examples/others/plot_repurposing_annotations.html
+++ b/0.16/auto_examples/others/plot_repurposing_annotations.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/others/plot_scripted_tensor_transforms.html b/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
index 50757817e61..d30a878f686 100644
--- a/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
+++ b/0.16/auto_examples/others/plot_scripted_tensor_transforms.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/others/plot_video_api.html b/0.16/auto_examples/others/plot_video_api.html
index 840a8323dde..976b1ab96f1 100644
--- a/0.16/auto_examples/others/plot_video_api.html
+++ b/0.16/auto_examples/others/plot_video_api.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/others/plot_visualization_utils.html b/0.16/auto_examples/others/plot_visualization_utils.html
index f768af7c544..b1eecaecccc 100644
--- a/0.16/auto_examples/others/plot_visualization_utils.html
+++ b/0.16/auto_examples/others/plot_visualization_utils.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/others/sg_execution_times.html b/0.16/auto_examples/others/sg_execution_times.html
index 245e01a0cfe..7859f00e169 100644
--- a/0.16/auto_examples/others/sg_execution_times.html
+++ b/0.16/auto_examples/others/sg_execution_times.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/index.html b/0.16/auto_examples/transforms/index.html
index b74d03d2a5a..057a9bf41fb 100644
--- a/0.16/auto_examples/transforms/index.html
+++ b/0.16/auto_examples/transforms/index.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/plot_custom_transforms.html b/0.16/auto_examples/transforms/plot_custom_transforms.html
index 556338a28a9..7381b3564e7 100644
--- a/0.16/auto_examples/transforms/plot_custom_transforms.html
+++ b/0.16/auto_examples/transforms/plot_custom_transforms.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/plot_custom_tv_tensors.html b/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
index f6273a3e416..df691317880 100644
--- a/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
+++ b/0.16/auto_examples/transforms/plot_custom_tv_tensors.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/plot_cutmix_mixup.html b/0.16/auto_examples/transforms/plot_cutmix_mixup.html
index acc6c61cc22..dbb4ee4946d 100644
--- a/0.16/auto_examples/transforms/plot_cutmix_mixup.html
+++ b/0.16/auto_examples/transforms/plot_cutmix_mixup.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/plot_transforms_e2e.html b/0.16/auto_examples/transforms/plot_transforms_e2e.html
index 4448b5ae8ca..ebe7c64c12a 100644
--- a/0.16/auto_examples/transforms/plot_transforms_e2e.html
+++ b/0.16/auto_examples/transforms/plot_transforms_e2e.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/plot_transforms_getting_started.html b/0.16/auto_examples/transforms/plot_transforms_getting_started.html
index 17bab393c67..7055304df9d 100644
--- a/0.16/auto_examples/transforms/plot_transforms_getting_started.html
+++ b/0.16/auto_examples/transforms/plot_transforms_getting_started.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/plot_transforms_illustrations.html b/0.16/auto_examples/transforms/plot_transforms_illustrations.html
index 45e02e2eb4d..458c89f2666 100644
--- a/0.16/auto_examples/transforms/plot_transforms_illustrations.html
+++ b/0.16/auto_examples/transforms/plot_transforms_illustrations.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/plot_tv_tensors.html b/0.16/auto_examples/transforms/plot_tv_tensors.html
index f6c3242d4d3..ac4e0a8043d 100644
--- a/0.16/auto_examples/transforms/plot_tv_tensors.html
+++ b/0.16/auto_examples/transforms/plot_tv_tensors.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/auto_examples/transforms/sg_execution_times.html b/0.16/auto_examples/transforms/sg_execution_times.html
index ed6ce19bfa3..7864857a5cf 100644
--- a/0.16/auto_examples/transforms/sg_execution_times.html
+++ b/0.16/auto_examples/transforms/sg_execution_times.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/datasets.html b/0.16/datasets.html
index 0c96b30f3df..fd430de5a6f 100644
--- a/0.16/datasets.html
+++ b/0.16/datasets.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/feature_extraction.html b/0.16/feature_extraction.html
index cede89aba1b..6eed013c339 100644
--- a/0.16/feature_extraction.html
+++ b/0.16/feature_extraction.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/classification_quant_table.html b/0.16/generated/classification_quant_table.html
index 63d73f48fa4..060b6bbcbb9 100644
--- a/0.16/generated/classification_quant_table.html
+++ b/0.16/generated/classification_quant_table.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/classification_table.html b/0.16/generated/classification_table.html
index c87bad351a4..3a45c9b2eef 100644
--- a/0.16/generated/classification_table.html
+++ b/0.16/generated/classification_table.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/detection_keypoint_table.html b/0.16/generated/detection_keypoint_table.html
index ad1b56d6d53..2330ca1f642 100644
--- a/0.16/generated/detection_keypoint_table.html
+++ b/0.16/generated/detection_keypoint_table.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/detection_table.html b/0.16/generated/detection_table.html
index 0c81a16c731..25b3b4425dd 100644
--- a/0.16/generated/detection_table.html
+++ b/0.16/generated/detection_table.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/instance_segmentation_table.html b/0.16/generated/instance_segmentation_table.html
index efad0be4c23..5535bae5db2 100644
--- a/0.16/generated/instance_segmentation_table.html
+++ b/0.16/generated/instance_segmentation_table.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/segmentation_table.html b/0.16/generated/segmentation_table.html
index 0536fb49873..65bca4dfbfa 100644
--- a/0.16/generated/segmentation_table.html
+++ b/0.16/generated/segmentation_table.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.CIFAR10.html b/0.16/generated/torchvision.datasets.CIFAR10.html
index 722e960e213..3a5af1ba1a5 100644
--- a/0.16/generated/torchvision.datasets.CIFAR10.html
+++ b/0.16/generated/torchvision.datasets.CIFAR10.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.CIFAR100.html b/0.16/generated/torchvision.datasets.CIFAR100.html
index b6940f7c6ef..15cc2209b61 100644
--- a/0.16/generated/torchvision.datasets.CIFAR100.html
+++ b/0.16/generated/torchvision.datasets.CIFAR100.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.CREStereo.html b/0.16/generated/torchvision.datasets.CREStereo.html
index c32cc8d50fd..b6d6f859ee5 100644
--- a/0.16/generated/torchvision.datasets.CREStereo.html
+++ b/0.16/generated/torchvision.datasets.CREStereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Caltech101.html b/0.16/generated/torchvision.datasets.Caltech101.html
index adf46abfda8..49c8b206899 100644
--- a/0.16/generated/torchvision.datasets.Caltech101.html
+++ b/0.16/generated/torchvision.datasets.Caltech101.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Caltech256.html b/0.16/generated/torchvision.datasets.Caltech256.html
index d449ce98cdc..8179f3b2d39 100644
--- a/0.16/generated/torchvision.datasets.Caltech256.html
+++ b/0.16/generated/torchvision.datasets.Caltech256.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.CarlaStereo.html b/0.16/generated/torchvision.datasets.CarlaStereo.html
index 77e009f21df..18b62f05869 100644
--- a/0.16/generated/torchvision.datasets.CarlaStereo.html
+++ b/0.16/generated/torchvision.datasets.CarlaStereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.CelebA.html b/0.16/generated/torchvision.datasets.CelebA.html
index e144fd1048d..6b69bd89a94 100644
--- a/0.16/generated/torchvision.datasets.CelebA.html
+++ b/0.16/generated/torchvision.datasets.CelebA.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Cityscapes.html b/0.16/generated/torchvision.datasets.Cityscapes.html
index cf1dffef071..8396ddc2c19 100644
--- a/0.16/generated/torchvision.datasets.Cityscapes.html
+++ b/0.16/generated/torchvision.datasets.Cityscapes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.CocoCaptions.html b/0.16/generated/torchvision.datasets.CocoCaptions.html
index b8fccf2b472..ada9e43c839 100644
--- a/0.16/generated/torchvision.datasets.CocoCaptions.html
+++ b/0.16/generated/torchvision.datasets.CocoCaptions.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.CocoDetection.html b/0.16/generated/torchvision.datasets.CocoDetection.html
index c93eee57ea0..fa564a1a78c 100644
--- a/0.16/generated/torchvision.datasets.CocoDetection.html
+++ b/0.16/generated/torchvision.datasets.CocoDetection.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Country211.html b/0.16/generated/torchvision.datasets.Country211.html
index 44555bffffa..cedf4553317 100644
--- a/0.16/generated/torchvision.datasets.Country211.html
+++ b/0.16/generated/torchvision.datasets.Country211.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.DTD.html b/0.16/generated/torchvision.datasets.DTD.html
index c98c24bddf0..034309bfa1e 100644
--- a/0.16/generated/torchvision.datasets.DTD.html
+++ b/0.16/generated/torchvision.datasets.DTD.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.DatasetFolder.html b/0.16/generated/torchvision.datasets.DatasetFolder.html
index 42578d2421b..a02f525e38b 100644
--- a/0.16/generated/torchvision.datasets.DatasetFolder.html
+++ b/0.16/generated/torchvision.datasets.DatasetFolder.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.EMNIST.html b/0.16/generated/torchvision.datasets.EMNIST.html
index 5149608135c..e4fb1062fee 100644
--- a/0.16/generated/torchvision.datasets.EMNIST.html
+++ b/0.16/generated/torchvision.datasets.EMNIST.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.ETH3DStereo.html b/0.16/generated/torchvision.datasets.ETH3DStereo.html
index c19c496de1c..4325ea8722f 100644
--- a/0.16/generated/torchvision.datasets.ETH3DStereo.html
+++ b/0.16/generated/torchvision.datasets.ETH3DStereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.EuroSAT.html b/0.16/generated/torchvision.datasets.EuroSAT.html
index 97cdfc674c5..dd59644464c 100644
--- a/0.16/generated/torchvision.datasets.EuroSAT.html
+++ b/0.16/generated/torchvision.datasets.EuroSAT.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.FER2013.html b/0.16/generated/torchvision.datasets.FER2013.html
index bdf645bee7f..6ca417d55e6 100644
--- a/0.16/generated/torchvision.datasets.FER2013.html
+++ b/0.16/generated/torchvision.datasets.FER2013.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.FGVCAircraft.html b/0.16/generated/torchvision.datasets.FGVCAircraft.html
index f5477c6d3b3..4cbfa00b11f 100644
--- a/0.16/generated/torchvision.datasets.FGVCAircraft.html
+++ b/0.16/generated/torchvision.datasets.FGVCAircraft.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.FakeData.html b/0.16/generated/torchvision.datasets.FakeData.html
index d27c24ec903..396d661df81 100644
--- a/0.16/generated/torchvision.datasets.FakeData.html
+++ b/0.16/generated/torchvision.datasets.FakeData.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.FallingThingsStereo.html b/0.16/generated/torchvision.datasets.FallingThingsStereo.html
index b440e4acc65..6650a771547 100644
--- a/0.16/generated/torchvision.datasets.FallingThingsStereo.html
+++ b/0.16/generated/torchvision.datasets.FallingThingsStereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.FashionMNIST.html b/0.16/generated/torchvision.datasets.FashionMNIST.html
index df34d6d39ca..092f68131e5 100644
--- a/0.16/generated/torchvision.datasets.FashionMNIST.html
+++ b/0.16/generated/torchvision.datasets.FashionMNIST.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Flickr30k.html b/0.16/generated/torchvision.datasets.Flickr30k.html
index 2f61be961d9..e897620b176 100644
--- a/0.16/generated/torchvision.datasets.Flickr30k.html
+++ b/0.16/generated/torchvision.datasets.Flickr30k.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Flickr8k.html b/0.16/generated/torchvision.datasets.Flickr8k.html
index 79b658ae5a5..b382055a658 100644
--- a/0.16/generated/torchvision.datasets.Flickr8k.html
+++ b/0.16/generated/torchvision.datasets.Flickr8k.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Flowers102.html b/0.16/generated/torchvision.datasets.Flowers102.html
index 55964a86b49..0b590029380 100644
--- a/0.16/generated/torchvision.datasets.Flowers102.html
+++ b/0.16/generated/torchvision.datasets.Flowers102.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.FlyingChairs.html b/0.16/generated/torchvision.datasets.FlyingChairs.html
index 05fee62aa32..b416c775073 100644
--- a/0.16/generated/torchvision.datasets.FlyingChairs.html
+++ b/0.16/generated/torchvision.datasets.FlyingChairs.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.FlyingThings3D.html b/0.16/generated/torchvision.datasets.FlyingThings3D.html
index 2feb977e669..b4f3ed0cd9b 100644
--- a/0.16/generated/torchvision.datasets.FlyingThings3D.html
+++ b/0.16/generated/torchvision.datasets.FlyingThings3D.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Food101.html b/0.16/generated/torchvision.datasets.Food101.html
index ceb17761652..8173bc5c3ce 100644
--- a/0.16/generated/torchvision.datasets.Food101.html
+++ b/0.16/generated/torchvision.datasets.Food101.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.GTSRB.html b/0.16/generated/torchvision.datasets.GTSRB.html
index 3d40152ef1d..9e70d8fefb0 100644
--- a/0.16/generated/torchvision.datasets.GTSRB.html
+++ b/0.16/generated/torchvision.datasets.GTSRB.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.HD1K.html b/0.16/generated/torchvision.datasets.HD1K.html
index ef9c28850f5..e51180dc5f9 100644
--- a/0.16/generated/torchvision.datasets.HD1K.html
+++ b/0.16/generated/torchvision.datasets.HD1K.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.HMDB51.html b/0.16/generated/torchvision.datasets.HMDB51.html
index d3f06d442aa..7ab2a015a5f 100644
--- a/0.16/generated/torchvision.datasets.HMDB51.html
+++ b/0.16/generated/torchvision.datasets.HMDB51.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.INaturalist.html b/0.16/generated/torchvision.datasets.INaturalist.html
index bbe4462f98f..60a9a86c701 100644
--- a/0.16/generated/torchvision.datasets.INaturalist.html
+++ b/0.16/generated/torchvision.datasets.INaturalist.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.ImageFolder.html b/0.16/generated/torchvision.datasets.ImageFolder.html
index b3d6fd97393..b70fae6cb04 100644
--- a/0.16/generated/torchvision.datasets.ImageFolder.html
+++ b/0.16/generated/torchvision.datasets.ImageFolder.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.ImageNet.html b/0.16/generated/torchvision.datasets.ImageNet.html
index dfbdc687846..2aa7758c30d 100644
--- a/0.16/generated/torchvision.datasets.ImageNet.html
+++ b/0.16/generated/torchvision.datasets.ImageNet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.InStereo2k.html b/0.16/generated/torchvision.datasets.InStereo2k.html
index d4f5fa7c5b1..9cf4bfed354 100644
--- a/0.16/generated/torchvision.datasets.InStereo2k.html
+++ b/0.16/generated/torchvision.datasets.InStereo2k.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.KMNIST.html b/0.16/generated/torchvision.datasets.KMNIST.html
index 7aafe8876c5..40ebafdd151 100644
--- a/0.16/generated/torchvision.datasets.KMNIST.html
+++ b/0.16/generated/torchvision.datasets.KMNIST.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Kinetics.html b/0.16/generated/torchvision.datasets.Kinetics.html
index fd613b08919..a6efe04c7db 100644
--- a/0.16/generated/torchvision.datasets.Kinetics.html
+++ b/0.16/generated/torchvision.datasets.Kinetics.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Kitti.html b/0.16/generated/torchvision.datasets.Kitti.html
index 758905eb2a4..370ba2894b8 100644
--- a/0.16/generated/torchvision.datasets.Kitti.html
+++ b/0.16/generated/torchvision.datasets.Kitti.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Kitti2012Stereo.html b/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
index b5ad87b4cff..6338912cc8b 100644
--- a/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
+++ b/0.16/generated/torchvision.datasets.Kitti2012Stereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Kitti2015Stereo.html b/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
index 7afc75041ad..4318ddc47be 100644
--- a/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
+++ b/0.16/generated/torchvision.datasets.Kitti2015Stereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.KittiFlow.html b/0.16/generated/torchvision.datasets.KittiFlow.html
index 8ecf5ba2afa..e02ec7d97df 100644
--- a/0.16/generated/torchvision.datasets.KittiFlow.html
+++ b/0.16/generated/torchvision.datasets.KittiFlow.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.LFWPairs.html b/0.16/generated/torchvision.datasets.LFWPairs.html
index b18c1ebd898..92ae6d62228 100644
--- a/0.16/generated/torchvision.datasets.LFWPairs.html
+++ b/0.16/generated/torchvision.datasets.LFWPairs.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.LFWPeople.html b/0.16/generated/torchvision.datasets.LFWPeople.html
index 590c967e7dd..2691d8cbb81 100644
--- a/0.16/generated/torchvision.datasets.LFWPeople.html
+++ b/0.16/generated/torchvision.datasets.LFWPeople.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.LSUN.html b/0.16/generated/torchvision.datasets.LSUN.html
index bea16e91a83..e1b69941a21 100644
--- a/0.16/generated/torchvision.datasets.LSUN.html
+++ b/0.16/generated/torchvision.datasets.LSUN.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.MNIST.html b/0.16/generated/torchvision.datasets.MNIST.html
index 4cc06db2623..996106e1f54 100644
--- a/0.16/generated/torchvision.datasets.MNIST.html
+++ b/0.16/generated/torchvision.datasets.MNIST.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html b/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
index 3793b670b11..2c0bdc78dc4 100644
--- a/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
+++ b/0.16/generated/torchvision.datasets.Middlebury2014Stereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.MovingMNIST.html b/0.16/generated/torchvision.datasets.MovingMNIST.html
index a3f8a777407..30b77580247 100644
--- a/0.16/generated/torchvision.datasets.MovingMNIST.html
+++ b/0.16/generated/torchvision.datasets.MovingMNIST.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Omniglot.html b/0.16/generated/torchvision.datasets.Omniglot.html
index 60349a8d2d4..17133aa9786 100644
--- a/0.16/generated/torchvision.datasets.Omniglot.html
+++ b/0.16/generated/torchvision.datasets.Omniglot.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.OxfordIIITPet.html b/0.16/generated/torchvision.datasets.OxfordIIITPet.html
index eb68787226c..404e48f25bf 100644
--- a/0.16/generated/torchvision.datasets.OxfordIIITPet.html
+++ b/0.16/generated/torchvision.datasets.OxfordIIITPet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.PCAM.html b/0.16/generated/torchvision.datasets.PCAM.html
index b7969ad6670..fbde2602270 100644
--- a/0.16/generated/torchvision.datasets.PCAM.html
+++ b/0.16/generated/torchvision.datasets.PCAM.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.PhotoTour.html b/0.16/generated/torchvision.datasets.PhotoTour.html
index 62bd154e410..961d7be35b9 100644
--- a/0.16/generated/torchvision.datasets.PhotoTour.html
+++ b/0.16/generated/torchvision.datasets.PhotoTour.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Places365.html b/0.16/generated/torchvision.datasets.Places365.html
index f506d3fe9c7..c89731ec07f 100644
--- a/0.16/generated/torchvision.datasets.Places365.html
+++ b/0.16/generated/torchvision.datasets.Places365.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.QMNIST.html b/0.16/generated/torchvision.datasets.QMNIST.html
index fd0d141c6e2..c24afe2f271 100644
--- a/0.16/generated/torchvision.datasets.QMNIST.html
+++ b/0.16/generated/torchvision.datasets.QMNIST.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.RenderedSST2.html b/0.16/generated/torchvision.datasets.RenderedSST2.html
index 795623cef9f..605c35606ab 100644
--- a/0.16/generated/torchvision.datasets.RenderedSST2.html
+++ b/0.16/generated/torchvision.datasets.RenderedSST2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.SBDataset.html b/0.16/generated/torchvision.datasets.SBDataset.html
index 76d234848f6..9e487611fff 100644
--- a/0.16/generated/torchvision.datasets.SBDataset.html
+++ b/0.16/generated/torchvision.datasets.SBDataset.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.SBU.html b/0.16/generated/torchvision.datasets.SBU.html
index 8f484ef458f..a1029602d2a 100644
--- a/0.16/generated/torchvision.datasets.SBU.html
+++ b/0.16/generated/torchvision.datasets.SBU.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.SEMEION.html b/0.16/generated/torchvision.datasets.SEMEION.html
index 50443dfea49..e56028d6754 100644
--- a/0.16/generated/torchvision.datasets.SEMEION.html
+++ b/0.16/generated/torchvision.datasets.SEMEION.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.STL10.html b/0.16/generated/torchvision.datasets.STL10.html
index 992c83018be..ff75a38a5af 100644
--- a/0.16/generated/torchvision.datasets.STL10.html
+++ b/0.16/generated/torchvision.datasets.STL10.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.SUN397.html b/0.16/generated/torchvision.datasets.SUN397.html
index e8a5a27cb78..4bcb648091a 100644
--- a/0.16/generated/torchvision.datasets.SUN397.html
+++ b/0.16/generated/torchvision.datasets.SUN397.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.SVHN.html b/0.16/generated/torchvision.datasets.SVHN.html
index d0fe0a8a178..a61bf066ed1 100644
--- a/0.16/generated/torchvision.datasets.SVHN.html
+++ b/0.16/generated/torchvision.datasets.SVHN.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.SceneFlowStereo.html b/0.16/generated/torchvision.datasets.SceneFlowStereo.html
index 5b5ca5138b4..0f40ddc25e3 100644
--- a/0.16/generated/torchvision.datasets.SceneFlowStereo.html
+++ b/0.16/generated/torchvision.datasets.SceneFlowStereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.Sintel.html b/0.16/generated/torchvision.datasets.Sintel.html
index dafc35162a3..62ffd874d2a 100644
--- a/0.16/generated/torchvision.datasets.Sintel.html
+++ b/0.16/generated/torchvision.datasets.Sintel.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.SintelStereo.html b/0.16/generated/torchvision.datasets.SintelStereo.html
index 4c9c650de75..86c8243f5b6 100644
--- a/0.16/generated/torchvision.datasets.SintelStereo.html
+++ b/0.16/generated/torchvision.datasets.SintelStereo.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.StanfordCars.html b/0.16/generated/torchvision.datasets.StanfordCars.html
index 254acf23e62..a5aad507236 100644
--- a/0.16/generated/torchvision.datasets.StanfordCars.html
+++ b/0.16/generated/torchvision.datasets.StanfordCars.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.UCF101.html b/0.16/generated/torchvision.datasets.UCF101.html
index 5bc9d032c65..21cd370ed75 100644
--- a/0.16/generated/torchvision.datasets.UCF101.html
+++ b/0.16/generated/torchvision.datasets.UCF101.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.USPS.html b/0.16/generated/torchvision.datasets.USPS.html
index 8cb037c53c5..0325b9a17cb 100644
--- a/0.16/generated/torchvision.datasets.USPS.html
+++ b/0.16/generated/torchvision.datasets.USPS.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.VOCDetection.html b/0.16/generated/torchvision.datasets.VOCDetection.html
index 884bc1819ad..d15d3a2d5cd 100644
--- a/0.16/generated/torchvision.datasets.VOCDetection.html
+++ b/0.16/generated/torchvision.datasets.VOCDetection.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.VOCSegmentation.html b/0.16/generated/torchvision.datasets.VOCSegmentation.html
index 4a54e2c0424..e3f29f983bf 100644
--- a/0.16/generated/torchvision.datasets.VOCSegmentation.html
+++ b/0.16/generated/torchvision.datasets.VOCSegmentation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.VisionDataset.html b/0.16/generated/torchvision.datasets.VisionDataset.html
index 769ad4163c0..592fdb0b3e9 100644
--- a/0.16/generated/torchvision.datasets.VisionDataset.html
+++ b/0.16/generated/torchvision.datasets.VisionDataset.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.WIDERFace.html b/0.16/generated/torchvision.datasets.WIDERFace.html
index e5adef2d4f8..408323ab083 100644
--- a/0.16/generated/torchvision.datasets.WIDERFace.html
+++ b/0.16/generated/torchvision.datasets.WIDERFace.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html b/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
index 1b24f496578..15c2656c68e 100644
--- a/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
+++ b/0.16/generated/torchvision.datasets.wrap_dataset_for_transforms_v2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.ImageReadMode.html b/0.16/generated/torchvision.io.ImageReadMode.html
index 7065bc36e39..5008c824c34 100644
--- a/0.16/generated/torchvision.io.ImageReadMode.html
+++ b/0.16/generated/torchvision.io.ImageReadMode.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.VideoReader.html b/0.16/generated/torchvision.io.VideoReader.html
index bd282ce64a5..6339452ed87 100644
--- a/0.16/generated/torchvision.io.VideoReader.html
+++ b/0.16/generated/torchvision.io.VideoReader.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.decode_image.html b/0.16/generated/torchvision.io.decode_image.html
index bc9a356acc0..0a1fa156f57 100644
--- a/0.16/generated/torchvision.io.decode_image.html
+++ b/0.16/generated/torchvision.io.decode_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.decode_jpeg.html b/0.16/generated/torchvision.io.decode_jpeg.html
index df8256a413f..139b0113dbd 100644
--- a/0.16/generated/torchvision.io.decode_jpeg.html
+++ b/0.16/generated/torchvision.io.decode_jpeg.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.decode_png.html b/0.16/generated/torchvision.io.decode_png.html
index ffc7504cd1e..70ee24ef4af 100644
--- a/0.16/generated/torchvision.io.decode_png.html
+++ b/0.16/generated/torchvision.io.decode_png.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.encode_jpeg.html b/0.16/generated/torchvision.io.encode_jpeg.html
index 7658efd058b..8d37df57e53 100644
--- a/0.16/generated/torchvision.io.encode_jpeg.html
+++ b/0.16/generated/torchvision.io.encode_jpeg.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.encode_png.html b/0.16/generated/torchvision.io.encode_png.html
index 9660c1fb4c2..fef13382b84 100644
--- a/0.16/generated/torchvision.io.encode_png.html
+++ b/0.16/generated/torchvision.io.encode_png.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.read_file.html b/0.16/generated/torchvision.io.read_file.html
index ca7e78d407c..1e665d7294e 100644
--- a/0.16/generated/torchvision.io.read_file.html
+++ b/0.16/generated/torchvision.io.read_file.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.read_image.html b/0.16/generated/torchvision.io.read_image.html
index c446ba25371..12ebe814279 100644
--- a/0.16/generated/torchvision.io.read_image.html
+++ b/0.16/generated/torchvision.io.read_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.read_video.html b/0.16/generated/torchvision.io.read_video.html
index 96e3cf6024b..ab96208a3fe 100644
--- a/0.16/generated/torchvision.io.read_video.html
+++ b/0.16/generated/torchvision.io.read_video.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.read_video_timestamps.html b/0.16/generated/torchvision.io.read_video_timestamps.html
index ce5fcfd18fb..fd9756d27c6 100644
--- a/0.16/generated/torchvision.io.read_video_timestamps.html
+++ b/0.16/generated/torchvision.io.read_video_timestamps.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.write_file.html b/0.16/generated/torchvision.io.write_file.html
index 2ffa619b9ff..2948ef4155a 100644
--- a/0.16/generated/torchvision.io.write_file.html
+++ b/0.16/generated/torchvision.io.write_file.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.write_jpeg.html b/0.16/generated/torchvision.io.write_jpeg.html
index 5fc545faf4d..d7e59014dcf 100644
--- a/0.16/generated/torchvision.io.write_jpeg.html
+++ b/0.16/generated/torchvision.io.write_jpeg.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.write_png.html b/0.16/generated/torchvision.io.write_png.html
index 786f38c431f..2c9285a5ac5 100644
--- a/0.16/generated/torchvision.io.write_png.html
+++ b/0.16/generated/torchvision.io.write_png.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.io.write_video.html b/0.16/generated/torchvision.io.write_video.html
index 08d6960a69f..e3cb8dab414 100644
--- a/0.16/generated/torchvision.io.write_video.html
+++ b/0.16/generated/torchvision.io.write_video.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html b/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
index b26dc7d961e..1f5decca338 100644
--- a/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
+++ b/0.16/generated/torchvision.models.feature_extraction.create_feature_extractor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html b/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
index 5b6c74a9399..d03ee0dbeee 100644
--- a/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
+++ b/0.16/generated/torchvision.models.feature_extraction.get_graph_node_names.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.models.get_model.html b/0.16/generated/torchvision.models.get_model.html
index 2a836fa4014..ddba1add773 100644
--- a/0.16/generated/torchvision.models.get_model.html
+++ b/0.16/generated/torchvision.models.get_model.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.models.get_model_weights.html b/0.16/generated/torchvision.models.get_model_weights.html
index 31ec89e4858..7e9e58b57ac 100644
--- a/0.16/generated/torchvision.models.get_model_weights.html
+++ b/0.16/generated/torchvision.models.get_model_weights.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.models.get_weight.html b/0.16/generated/torchvision.models.get_weight.html
index 8b5ba263a37..a2b9e0fb460 100644
--- a/0.16/generated/torchvision.models.get_weight.html
+++ b/0.16/generated/torchvision.models.get_weight.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.models.list_models.html b/0.16/generated/torchvision.models.list_models.html
index fe4e509ae95..a67768778df 100644
--- a/0.16/generated/torchvision.models.list_models.html
+++ b/0.16/generated/torchvision.models.list_models.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.Conv2dNormActivation.html b/0.16/generated/torchvision.ops.Conv2dNormActivation.html
index e6ffc320f8a..5df223fc908 100644
--- a/0.16/generated/torchvision.ops.Conv2dNormActivation.html
+++ b/0.16/generated/torchvision.ops.Conv2dNormActivation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.Conv3dNormActivation.html b/0.16/generated/torchvision.ops.Conv3dNormActivation.html
index a5bd457f180..765a12b335d 100644
--- a/0.16/generated/torchvision.ops.Conv3dNormActivation.html
+++ b/0.16/generated/torchvision.ops.Conv3dNormActivation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.DeformConv2d.html b/0.16/generated/torchvision.ops.DeformConv2d.html
index fb7ca7e8930..3ace61c179d 100644
--- a/0.16/generated/torchvision.ops.DeformConv2d.html
+++ b/0.16/generated/torchvision.ops.DeformConv2d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.DropBlock2d.html b/0.16/generated/torchvision.ops.DropBlock2d.html
index f1d1c44904a..b506b8d18e7 100644
--- a/0.16/generated/torchvision.ops.DropBlock2d.html
+++ b/0.16/generated/torchvision.ops.DropBlock2d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.DropBlock3d.html b/0.16/generated/torchvision.ops.DropBlock3d.html
index 929ae1b7ea1..129c40001e3 100644
--- a/0.16/generated/torchvision.ops.DropBlock3d.html
+++ b/0.16/generated/torchvision.ops.DropBlock3d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html b/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
index 940aa008aa4..b6d01cf59a9 100644
--- a/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
+++ b/0.16/generated/torchvision.ops.FeaturePyramidNetwork.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html b/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
index 39e584c43f1..5ca7bb96a20 100644
--- a/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
+++ b/0.16/generated/torchvision.ops.FrozenBatchNorm2d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.MLP.html b/0.16/generated/torchvision.ops.MLP.html
index 5b885ed3556..50667b3788c 100644
--- a/0.16/generated/torchvision.ops.MLP.html
+++ b/0.16/generated/torchvision.ops.MLP.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html b/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
index d3ca072b438..86cc286c6c4 100644
--- a/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
+++ b/0.16/generated/torchvision.ops.MultiScaleRoIAlign.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.PSRoIAlign.html b/0.16/generated/torchvision.ops.PSRoIAlign.html
index d1f90023131..424ded0c9da 100644
--- a/0.16/generated/torchvision.ops.PSRoIAlign.html
+++ b/0.16/generated/torchvision.ops.PSRoIAlign.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.PSRoIPool.html b/0.16/generated/torchvision.ops.PSRoIPool.html
index ab43e98c3ac..fda9b34af0e 100644
--- a/0.16/generated/torchvision.ops.PSRoIPool.html
+++ b/0.16/generated/torchvision.ops.PSRoIPool.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.Permute.html b/0.16/generated/torchvision.ops.Permute.html
index 05ae6e88a9c..336c11194c3 100644
--- a/0.16/generated/torchvision.ops.Permute.html
+++ b/0.16/generated/torchvision.ops.Permute.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.RoIAlign.html b/0.16/generated/torchvision.ops.RoIAlign.html
index 110c9fbb9ba..11a2c8efe70 100644
--- a/0.16/generated/torchvision.ops.RoIAlign.html
+++ b/0.16/generated/torchvision.ops.RoIAlign.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.RoIPool.html b/0.16/generated/torchvision.ops.RoIPool.html
index 60643f5f02b..7b8b5dd4bff 100644
--- a/0.16/generated/torchvision.ops.RoIPool.html
+++ b/0.16/generated/torchvision.ops.RoIPool.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.SqueezeExcitation.html b/0.16/generated/torchvision.ops.SqueezeExcitation.html
index 89ada5ef725..2569b601a47 100644
--- a/0.16/generated/torchvision.ops.SqueezeExcitation.html
+++ b/0.16/generated/torchvision.ops.SqueezeExcitation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.StochasticDepth.html b/0.16/generated/torchvision.ops.StochasticDepth.html
index 271bbf135c6..f9e135a4462 100644
--- a/0.16/generated/torchvision.ops.StochasticDepth.html
+++ b/0.16/generated/torchvision.ops.StochasticDepth.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.batched_nms.html b/0.16/generated/torchvision.ops.batched_nms.html
index 0d3e32c1b7a..fdba8a4ebca 100644
--- a/0.16/generated/torchvision.ops.batched_nms.html
+++ b/0.16/generated/torchvision.ops.batched_nms.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.box_area.html b/0.16/generated/torchvision.ops.box_area.html
index 324708929ad..0c8a3eaec77 100644
--- a/0.16/generated/torchvision.ops.box_area.html
+++ b/0.16/generated/torchvision.ops.box_area.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.box_convert.html b/0.16/generated/torchvision.ops.box_convert.html
index 1db03a96da0..f4bf35366dd 100644
--- a/0.16/generated/torchvision.ops.box_convert.html
+++ b/0.16/generated/torchvision.ops.box_convert.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.box_iou.html b/0.16/generated/torchvision.ops.box_iou.html
index 3e774addf0f..a163d19a891 100644
--- a/0.16/generated/torchvision.ops.box_iou.html
+++ b/0.16/generated/torchvision.ops.box_iou.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.clip_boxes_to_image.html b/0.16/generated/torchvision.ops.clip_boxes_to_image.html
index 0005bafed09..36deeab5e08 100644
--- a/0.16/generated/torchvision.ops.clip_boxes_to_image.html
+++ b/0.16/generated/torchvision.ops.clip_boxes_to_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.complete_box_iou.html b/0.16/generated/torchvision.ops.complete_box_iou.html
index 839e1cc7474..9c144be8c1b 100644
--- a/0.16/generated/torchvision.ops.complete_box_iou.html
+++ b/0.16/generated/torchvision.ops.complete_box_iou.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.complete_box_iou_loss.html b/0.16/generated/torchvision.ops.complete_box_iou_loss.html
index b48ffdc3846..8962e6a5e5c 100644
--- a/0.16/generated/torchvision.ops.complete_box_iou_loss.html
+++ b/0.16/generated/torchvision.ops.complete_box_iou_loss.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.deform_conv2d.html b/0.16/generated/torchvision.ops.deform_conv2d.html
index 8da88f23b41..504fa656c8f 100644
--- a/0.16/generated/torchvision.ops.deform_conv2d.html
+++ b/0.16/generated/torchvision.ops.deform_conv2d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.distance_box_iou.html b/0.16/generated/torchvision.ops.distance_box_iou.html
index ac5df337ed8..9059d72dff6 100644
--- a/0.16/generated/torchvision.ops.distance_box_iou.html
+++ b/0.16/generated/torchvision.ops.distance_box_iou.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.distance_box_iou_loss.html b/0.16/generated/torchvision.ops.distance_box_iou_loss.html
index 37c3894bdda..c6ffea00507 100644
--- a/0.16/generated/torchvision.ops.distance_box_iou_loss.html
+++ b/0.16/generated/torchvision.ops.distance_box_iou_loss.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.drop_block2d.html b/0.16/generated/torchvision.ops.drop_block2d.html
index 94661f239ac..1dfdc3b285c 100644
--- a/0.16/generated/torchvision.ops.drop_block2d.html
+++ b/0.16/generated/torchvision.ops.drop_block2d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.drop_block3d.html b/0.16/generated/torchvision.ops.drop_block3d.html
index 0f013eb72c9..beedc1d63d4 100644
--- a/0.16/generated/torchvision.ops.drop_block3d.html
+++ b/0.16/generated/torchvision.ops.drop_block3d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.generalized_box_iou.html b/0.16/generated/torchvision.ops.generalized_box_iou.html
index 74b297196eb..245aa36561d 100644
--- a/0.16/generated/torchvision.ops.generalized_box_iou.html
+++ b/0.16/generated/torchvision.ops.generalized_box_iou.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.generalized_box_iou_loss.html b/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
index 9bcfca3dbf6..4e2b6dd5d2c 100644
--- a/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
+++ b/0.16/generated/torchvision.ops.generalized_box_iou_loss.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.masks_to_boxes.html b/0.16/generated/torchvision.ops.masks_to_boxes.html
index 2e264b2bdff..dbd0d67e112 100644
--- a/0.16/generated/torchvision.ops.masks_to_boxes.html
+++ b/0.16/generated/torchvision.ops.masks_to_boxes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.nms.html b/0.16/generated/torchvision.ops.nms.html
index fb542d16c2e..051b196cf5f 100644
--- a/0.16/generated/torchvision.ops.nms.html
+++ b/0.16/generated/torchvision.ops.nms.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.ps_roi_align.html b/0.16/generated/torchvision.ops.ps_roi_align.html
index 90d0482a841..0398a741578 100644
--- a/0.16/generated/torchvision.ops.ps_roi_align.html
+++ b/0.16/generated/torchvision.ops.ps_roi_align.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.ps_roi_pool.html b/0.16/generated/torchvision.ops.ps_roi_pool.html
index 05a047c92f3..2374aec62cd 100644
--- a/0.16/generated/torchvision.ops.ps_roi_pool.html
+++ b/0.16/generated/torchvision.ops.ps_roi_pool.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.remove_small_boxes.html b/0.16/generated/torchvision.ops.remove_small_boxes.html
index 8d1de78acf4..991e9a9a0de 100644
--- a/0.16/generated/torchvision.ops.remove_small_boxes.html
+++ b/0.16/generated/torchvision.ops.remove_small_boxes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.roi_align.html b/0.16/generated/torchvision.ops.roi_align.html
index 513ef0f3456..ab39e1ffb16 100644
--- a/0.16/generated/torchvision.ops.roi_align.html
+++ b/0.16/generated/torchvision.ops.roi_align.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.roi_pool.html b/0.16/generated/torchvision.ops.roi_pool.html
index 072db500e4e..d3f52b18e6f 100644
--- a/0.16/generated/torchvision.ops.roi_pool.html
+++ b/0.16/generated/torchvision.ops.roi_pool.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.sigmoid_focal_loss.html b/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
index 972cb222ce9..de3296d5b36 100644
--- a/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
+++ b/0.16/generated/torchvision.ops.sigmoid_focal_loss.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.ops.stochastic_depth.html b/0.16/generated/torchvision.ops.stochastic_depth.html
index e78d79fe87f..0f01f724f9c 100644
--- a/0.16/generated/torchvision.ops.stochastic_depth.html
+++ b/0.16/generated/torchvision.ops.stochastic_depth.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.AugMix.html b/0.16/generated/torchvision.transforms.AugMix.html
index 40ba03e58ae..8e5e63d83df 100644
--- a/0.16/generated/torchvision.transforms.AugMix.html
+++ b/0.16/generated/torchvision.transforms.AugMix.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.AutoAugment.html b/0.16/generated/torchvision.transforms.AutoAugment.html
index 5a62377b8a4..f1520e8f8f9 100644
--- a/0.16/generated/torchvision.transforms.AutoAugment.html
+++ b/0.16/generated/torchvision.transforms.AutoAugment.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html b/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
index e4b58e04c14..1cb0bfbb485 100644
--- a/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
+++ b/0.16/generated/torchvision.transforms.AutoAugmentPolicy.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.CenterCrop.html b/0.16/generated/torchvision.transforms.CenterCrop.html
index 7d4f4fa948a..c8a566dc1b6 100644
--- a/0.16/generated/torchvision.transforms.CenterCrop.html
+++ b/0.16/generated/torchvision.transforms.CenterCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.ColorJitter.html b/0.16/generated/torchvision.transforms.ColorJitter.html
index 6a6a1d4ba89..5543b261298 100644
--- a/0.16/generated/torchvision.transforms.ColorJitter.html
+++ b/0.16/generated/torchvision.transforms.ColorJitter.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.Compose.html b/0.16/generated/torchvision.transforms.Compose.html
index 0e0b7aa56ff..1bb6f5a9cab 100644
--- a/0.16/generated/torchvision.transforms.Compose.html
+++ b/0.16/generated/torchvision.transforms.Compose.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.ConvertImageDtype.html b/0.16/generated/torchvision.transforms.ConvertImageDtype.html
index 0a1198fd79a..cd1dc03842a 100644
--- a/0.16/generated/torchvision.transforms.ConvertImageDtype.html
+++ b/0.16/generated/torchvision.transforms.ConvertImageDtype.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.ElasticTransform.html b/0.16/generated/torchvision.transforms.ElasticTransform.html
index 299824e53a9..86db3a39121 100644
--- a/0.16/generated/torchvision.transforms.ElasticTransform.html
+++ b/0.16/generated/torchvision.transforms.ElasticTransform.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.FiveCrop.html b/0.16/generated/torchvision.transforms.FiveCrop.html
index 23136e7c3d6..4883bfe9d23 100644
--- a/0.16/generated/torchvision.transforms.FiveCrop.html
+++ b/0.16/generated/torchvision.transforms.FiveCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.GaussianBlur.html b/0.16/generated/torchvision.transforms.GaussianBlur.html
index b4bb8b2ceb2..a4ce3b37610 100644
--- a/0.16/generated/torchvision.transforms.GaussianBlur.html
+++ b/0.16/generated/torchvision.transforms.GaussianBlur.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.Grayscale.html b/0.16/generated/torchvision.transforms.Grayscale.html
index ee580052264..065dc2093c2 100644
--- a/0.16/generated/torchvision.transforms.Grayscale.html
+++ b/0.16/generated/torchvision.transforms.Grayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.Lambda.html b/0.16/generated/torchvision.transforms.Lambda.html
index d1b1392c803..916e96f1a5e 100644
--- a/0.16/generated/torchvision.transforms.Lambda.html
+++ b/0.16/generated/torchvision.transforms.Lambda.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.LinearTransformation.html b/0.16/generated/torchvision.transforms.LinearTransformation.html
index 03950d65f51..3ff8505e8ee 100644
--- a/0.16/generated/torchvision.transforms.LinearTransformation.html
+++ b/0.16/generated/torchvision.transforms.LinearTransformation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.Normalize.html b/0.16/generated/torchvision.transforms.Normalize.html
index b1699ae2a6b..d9adfc376be 100644
--- a/0.16/generated/torchvision.transforms.Normalize.html
+++ b/0.16/generated/torchvision.transforms.Normalize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.PILToTensor.html b/0.16/generated/torchvision.transforms.PILToTensor.html
index bf010508d80..d76c70f9004 100644
--- a/0.16/generated/torchvision.transforms.PILToTensor.html
+++ b/0.16/generated/torchvision.transforms.PILToTensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.Pad.html b/0.16/generated/torchvision.transforms.Pad.html
index 9ac225fb4c0..663053834e5 100644
--- a/0.16/generated/torchvision.transforms.Pad.html
+++ b/0.16/generated/torchvision.transforms.Pad.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandAugment.html b/0.16/generated/torchvision.transforms.RandAugment.html
index cf00191a65c..111e6848489 100644
--- a/0.16/generated/torchvision.transforms.RandAugment.html
+++ b/0.16/generated/torchvision.transforms.RandAugment.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html b/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
index b6c2fbed2af..254fc4aa652 100644
--- a/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
+++ b/0.16/generated/torchvision.transforms.RandomAdjustSharpness.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomAffine.html b/0.16/generated/torchvision.transforms.RandomAffine.html
index 37cc7999a66..0c2102603ab 100644
--- a/0.16/generated/torchvision.transforms.RandomAffine.html
+++ b/0.16/generated/torchvision.transforms.RandomAffine.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomApply.html b/0.16/generated/torchvision.transforms.RandomApply.html
index c2a2a34b0a0..864e3ed6d8e 100644
--- a/0.16/generated/torchvision.transforms.RandomApply.html
+++ b/0.16/generated/torchvision.transforms.RandomApply.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomAutocontrast.html b/0.16/generated/torchvision.transforms.RandomAutocontrast.html
index 5cae86d3788..75b159471f0 100644
--- a/0.16/generated/torchvision.transforms.RandomAutocontrast.html
+++ b/0.16/generated/torchvision.transforms.RandomAutocontrast.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomChoice.html b/0.16/generated/torchvision.transforms.RandomChoice.html
index 41c466fdefd..31d1d245f1b 100644
--- a/0.16/generated/torchvision.transforms.RandomChoice.html
+++ b/0.16/generated/torchvision.transforms.RandomChoice.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomCrop.html b/0.16/generated/torchvision.transforms.RandomCrop.html
index 66c3a10069d..fda0eb47051 100644
--- a/0.16/generated/torchvision.transforms.RandomCrop.html
+++ b/0.16/generated/torchvision.transforms.RandomCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomEqualize.html b/0.16/generated/torchvision.transforms.RandomEqualize.html
index bcfbd2f1d97..ed85bb30759 100644
--- a/0.16/generated/torchvision.transforms.RandomEqualize.html
+++ b/0.16/generated/torchvision.transforms.RandomEqualize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomErasing.html b/0.16/generated/torchvision.transforms.RandomErasing.html
index d78b11c281e..f84165d2753 100644
--- a/0.16/generated/torchvision.transforms.RandomErasing.html
+++ b/0.16/generated/torchvision.transforms.RandomErasing.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomGrayscale.html b/0.16/generated/torchvision.transforms.RandomGrayscale.html
index 5af06784588..ce1c368c3fe 100644
--- a/0.16/generated/torchvision.transforms.RandomGrayscale.html
+++ b/0.16/generated/torchvision.transforms.RandomGrayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html b/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
index c0d81173558..68873bf2903 100644
--- a/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
+++ b/0.16/generated/torchvision.transforms.RandomHorizontalFlip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomInvert.html b/0.16/generated/torchvision.transforms.RandomInvert.html
index d097a6a22d0..a735fc246c7 100644
--- a/0.16/generated/torchvision.transforms.RandomInvert.html
+++ b/0.16/generated/torchvision.transforms.RandomInvert.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomOrder.html b/0.16/generated/torchvision.transforms.RandomOrder.html
index 69c533e24ec..5c647704c9c 100644
--- a/0.16/generated/torchvision.transforms.RandomOrder.html
+++ b/0.16/generated/torchvision.transforms.RandomOrder.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomPerspective.html b/0.16/generated/torchvision.transforms.RandomPerspective.html
index 27105947acb..429e712d7dc 100644
--- a/0.16/generated/torchvision.transforms.RandomPerspective.html
+++ b/0.16/generated/torchvision.transforms.RandomPerspective.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomPosterize.html b/0.16/generated/torchvision.transforms.RandomPosterize.html
index 24327b17a2a..c2ddfed9429 100644
--- a/0.16/generated/torchvision.transforms.RandomPosterize.html
+++ b/0.16/generated/torchvision.transforms.RandomPosterize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomResizedCrop.html b/0.16/generated/torchvision.transforms.RandomResizedCrop.html
index 9125197b6f9..daa053ca95d 100644
--- a/0.16/generated/torchvision.transforms.RandomResizedCrop.html
+++ b/0.16/generated/torchvision.transforms.RandomResizedCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomRotation.html b/0.16/generated/torchvision.transforms.RandomRotation.html
index 499bf30afd6..e336fa05394 100644
--- a/0.16/generated/torchvision.transforms.RandomRotation.html
+++ b/0.16/generated/torchvision.transforms.RandomRotation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomSolarize.html b/0.16/generated/torchvision.transforms.RandomSolarize.html
index 809723fa7f8..bdf3dfabd4d 100644
--- a/0.16/generated/torchvision.transforms.RandomSolarize.html
+++ b/0.16/generated/torchvision.transforms.RandomSolarize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.RandomVerticalFlip.html b/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
index 58a0b892630..88c15a87bbd 100644
--- a/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
+++ b/0.16/generated/torchvision.transforms.RandomVerticalFlip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.Resize.html b/0.16/generated/torchvision.transforms.Resize.html
index a053c5aa53c..bef273ab87b 100644
--- a/0.16/generated/torchvision.transforms.Resize.html
+++ b/0.16/generated/torchvision.transforms.Resize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.TenCrop.html b/0.16/generated/torchvision.transforms.TenCrop.html
index d3f087380f6..7d424d7245e 100644
--- a/0.16/generated/torchvision.transforms.TenCrop.html
+++ b/0.16/generated/torchvision.transforms.TenCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.ToPILImage.html b/0.16/generated/torchvision.transforms.ToPILImage.html
index eafba217d2e..5a1809a8e2c 100644
--- a/0.16/generated/torchvision.transforms.ToPILImage.html
+++ b/0.16/generated/torchvision.transforms.ToPILImage.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.ToTensor.html b/0.16/generated/torchvision.transforms.ToTensor.html
index ae18f539807..9693202eb78 100644
--- a/0.16/generated/torchvision.transforms.ToTensor.html
+++ b/0.16/generated/torchvision.transforms.ToTensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.TrivialAugmentWide.html b/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
index 23607a9aac9..c400ca17ebb 100644
--- a/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
+++ b/0.16/generated/torchvision.transforms.TrivialAugmentWide.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_brightness.html b/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
index 77942c2662b..c27975dbf4e 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_brightness.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_contrast.html b/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
index f730dd9d3ad..bc9a86fa7a3 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_contrast.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_gamma.html b/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
index d3ff1b1cdfb..2c402911530 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_gamma.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_hue.html b/0.16/generated/torchvision.transforms.functional.adjust_hue.html
index e100be1f010..893d9215a66 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_hue.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_hue.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_saturation.html b/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
index 5dab767c142..59680c860e1 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_saturation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html b/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
index 32b0f69a829..d9db4f2bfd1 100644
--- a/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
+++ b/0.16/generated/torchvision.transforms.functional.adjust_sharpness.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.affine.html b/0.16/generated/torchvision.transforms.functional.affine.html
index e3a591776c0..e92d86189ba 100644
--- a/0.16/generated/torchvision.transforms.functional.affine.html
+++ b/0.16/generated/torchvision.transforms.functional.affine.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.autocontrast.html b/0.16/generated/torchvision.transforms.functional.autocontrast.html
index c0156023e10..2215241685a 100644
--- a/0.16/generated/torchvision.transforms.functional.autocontrast.html
+++ b/0.16/generated/torchvision.transforms.functional.autocontrast.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.center_crop.html b/0.16/generated/torchvision.transforms.functional.center_crop.html
index b2edfaa9a91..65c0c135699 100644
--- a/0.16/generated/torchvision.transforms.functional.center_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.center_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html b/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
index c37ac10f617..389d059fd91 100644
--- a/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
+++ b/0.16/generated/torchvision.transforms.functional.convert_image_dtype.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.crop.html b/0.16/generated/torchvision.transforms.functional.crop.html
index 109a7e5f78a..305d9929193 100644
--- a/0.16/generated/torchvision.transforms.functional.crop.html
+++ b/0.16/generated/torchvision.transforms.functional.crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.equalize.html b/0.16/generated/torchvision.transforms.functional.equalize.html
index 92b31463db8..b4571662870 100644
--- a/0.16/generated/torchvision.transforms.functional.equalize.html
+++ b/0.16/generated/torchvision.transforms.functional.equalize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.erase.html b/0.16/generated/torchvision.transforms.functional.erase.html
index 7425e393ad9..e1de7dc5471 100644
--- a/0.16/generated/torchvision.transforms.functional.erase.html
+++ b/0.16/generated/torchvision.transforms.functional.erase.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.five_crop.html b/0.16/generated/torchvision.transforms.functional.five_crop.html
index 5fbe85cdd5f..13e5a8179eb 100644
--- a/0.16/generated/torchvision.transforms.functional.five_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.five_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.gaussian_blur.html b/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
index b5ba2b91eb5..ce52b134065 100644
--- a/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
+++ b/0.16/generated/torchvision.transforms.functional.gaussian_blur.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.get_dimensions.html b/0.16/generated/torchvision.transforms.functional.get_dimensions.html
index 01d69c7b2de..afef57f3bc2 100644
--- a/0.16/generated/torchvision.transforms.functional.get_dimensions.html
+++ b/0.16/generated/torchvision.transforms.functional.get_dimensions.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html b/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
index 12124024a0a..e013f7ea116 100644
--- a/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
+++ b/0.16/generated/torchvision.transforms.functional.get_image_num_channels.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.get_image_size.html b/0.16/generated/torchvision.transforms.functional.get_image_size.html
index ac21645e4b3..4b48ea1ef7e 100644
--- a/0.16/generated/torchvision.transforms.functional.get_image_size.html
+++ b/0.16/generated/torchvision.transforms.functional.get_image_size.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.hflip.html b/0.16/generated/torchvision.transforms.functional.hflip.html
index 354f3682dcf..92ed4097246 100644
--- a/0.16/generated/torchvision.transforms.functional.hflip.html
+++ b/0.16/generated/torchvision.transforms.functional.hflip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.invert.html b/0.16/generated/torchvision.transforms.functional.invert.html
index fe67087d4d0..70d627df002 100644
--- a/0.16/generated/torchvision.transforms.functional.invert.html
+++ b/0.16/generated/torchvision.transforms.functional.invert.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.normalize.html b/0.16/generated/torchvision.transforms.functional.normalize.html
index 51a81bd06a5..857f6532f96 100644
--- a/0.16/generated/torchvision.transforms.functional.normalize.html
+++ b/0.16/generated/torchvision.transforms.functional.normalize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.pad.html b/0.16/generated/torchvision.transforms.functional.pad.html
index 696014523bb..8fe71d55983 100644
--- a/0.16/generated/torchvision.transforms.functional.pad.html
+++ b/0.16/generated/torchvision.transforms.functional.pad.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.perspective.html b/0.16/generated/torchvision.transforms.functional.perspective.html
index 412db98ab4a..58a6ef9ab88 100644
--- a/0.16/generated/torchvision.transforms.functional.perspective.html
+++ b/0.16/generated/torchvision.transforms.functional.perspective.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html b/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
index a73f0063d1c..6edbcfdcb84 100644
--- a/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
+++ b/0.16/generated/torchvision.transforms.functional.pil_to_tensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.posterize.html b/0.16/generated/torchvision.transforms.functional.posterize.html
index f0029553f74..8c85e3ecde7 100644
--- a/0.16/generated/torchvision.transforms.functional.posterize.html
+++ b/0.16/generated/torchvision.transforms.functional.posterize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.resize.html b/0.16/generated/torchvision.transforms.functional.resize.html
index 2ecf87f277c..960b0bf3da2 100644
--- a/0.16/generated/torchvision.transforms.functional.resize.html
+++ b/0.16/generated/torchvision.transforms.functional.resize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.resized_crop.html b/0.16/generated/torchvision.transforms.functional.resized_crop.html
index 57de342d2b3..0fbf416f79c 100644
--- a/0.16/generated/torchvision.transforms.functional.resized_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.resized_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html b/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
index 05c62db9b06..fbf0dd25fc7 100644
--- a/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.functional.rgb_to_grayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.rotate.html b/0.16/generated/torchvision.transforms.functional.rotate.html
index f03b59096b2..e6c12534d79 100644
--- a/0.16/generated/torchvision.transforms.functional.rotate.html
+++ b/0.16/generated/torchvision.transforms.functional.rotate.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.solarize.html b/0.16/generated/torchvision.transforms.functional.solarize.html
index c2f8acd1b9f..30c6621f30b 100644
--- a/0.16/generated/torchvision.transforms.functional.solarize.html
+++ b/0.16/generated/torchvision.transforms.functional.solarize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.ten_crop.html b/0.16/generated/torchvision.transforms.functional.ten_crop.html
index 9d57dad2413..715634f3329 100644
--- a/0.16/generated/torchvision.transforms.functional.ten_crop.html
+++ b/0.16/generated/torchvision.transforms.functional.ten_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.to_grayscale.html b/0.16/generated/torchvision.transforms.functional.to_grayscale.html
index 375d5568cf0..bd2565b62ba 100644
--- a/0.16/generated/torchvision.transforms.functional.to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.functional.to_grayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.to_pil_image.html b/0.16/generated/torchvision.transforms.functional.to_pil_image.html
index 93c57ffb7e5..d4dda56fb90 100644
--- a/0.16/generated/torchvision.transforms.functional.to_pil_image.html
+++ b/0.16/generated/torchvision.transforms.functional.to_pil_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.to_tensor.html b/0.16/generated/torchvision.transforms.functional.to_tensor.html
index c869692f50c..916bf24c4ca 100644
--- a/0.16/generated/torchvision.transforms.functional.to_tensor.html
+++ b/0.16/generated/torchvision.transforms.functional.to_tensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.functional.vflip.html b/0.16/generated/torchvision.transforms.functional.vflip.html
index 28eda1da93c..686af9b9e55 100644
--- a/0.16/generated/torchvision.transforms.functional.vflip.html
+++ b/0.16/generated/torchvision.transforms.functional.vflip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.AugMix.html b/0.16/generated/torchvision.transforms.v2.AugMix.html
index 8483b176794..bf60aa197f7 100644
--- a/0.16/generated/torchvision.transforms.v2.AugMix.html
+++ b/0.16/generated/torchvision.transforms.v2.AugMix.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.AutoAugment.html b/0.16/generated/torchvision.transforms.v2.AutoAugment.html
index 96ce67633f6..6a5dc6cf490 100644
--- a/0.16/generated/torchvision.transforms.v2.AutoAugment.html
+++ b/0.16/generated/torchvision.transforms.v2.AutoAugment.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.CenterCrop.html b/0.16/generated/torchvision.transforms.v2.CenterCrop.html
index bb2b038fdd9..576196d06e7 100644
--- a/0.16/generated/torchvision.transforms.v2.CenterCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.CenterCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html b/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
index 7dc73291142..1061f7fd92b 100644
--- a/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
+++ b/0.16/generated/torchvision.transforms.v2.ClampBoundingBoxes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ColorJitter.html b/0.16/generated/torchvision.transforms.v2.ColorJitter.html
index 66bf46f56a0..831757b31ad 100644
--- a/0.16/generated/torchvision.transforms.v2.ColorJitter.html
+++ b/0.16/generated/torchvision.transforms.v2.ColorJitter.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.Compose.html b/0.16/generated/torchvision.transforms.v2.Compose.html
index f7f58b23fdd..fcb28284c41 100644
--- a/0.16/generated/torchvision.transforms.v2.Compose.html
+++ b/0.16/generated/torchvision.transforms.v2.Compose.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html b/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
index 74354dd438d..8d3a8c90bc5 100644
--- a/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
+++ b/0.16/generated/torchvision.transforms.v2.ConvertBoundingBoxFormat.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html b/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
index 1a2409a3b72..1a9cd8c8960 100644
--- a/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
+++ b/0.16/generated/torchvision.transforms.v2.ConvertImageDtype.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.CutMix.html b/0.16/generated/torchvision.transforms.v2.CutMix.html
index f7299eafa6f..061ee0c494b 100644
--- a/0.16/generated/torchvision.transforms.v2.CutMix.html
+++ b/0.16/generated/torchvision.transforms.v2.CutMix.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ElasticTransform.html b/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
index ba787862cef..77d4f83c596 100644
--- a/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
+++ b/0.16/generated/torchvision.transforms.v2.ElasticTransform.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.FiveCrop.html b/0.16/generated/torchvision.transforms.v2.FiveCrop.html
index 3fc0a6f7950..230d4df7ca8 100644
--- a/0.16/generated/torchvision.transforms.v2.FiveCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.FiveCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.GaussianBlur.html b/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
index 3c07aea9185..f76717234d3 100644
--- a/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
+++ b/0.16/generated/torchvision.transforms.v2.GaussianBlur.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.Grayscale.html b/0.16/generated/torchvision.transforms.v2.Grayscale.html
index 43fa98362ce..e7058bb3027 100644
--- a/0.16/generated/torchvision.transforms.v2.Grayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.Grayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.Lambda.html b/0.16/generated/torchvision.transforms.v2.Lambda.html
index b3a7290d4a5..a9255610acd 100644
--- a/0.16/generated/torchvision.transforms.v2.Lambda.html
+++ b/0.16/generated/torchvision.transforms.v2.Lambda.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.LinearTransformation.html b/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
index 808a8c2b713..c21e3df1d6b 100644
--- a/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
+++ b/0.16/generated/torchvision.transforms.v2.LinearTransformation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.MixUp.html b/0.16/generated/torchvision.transforms.v2.MixUp.html
index 75242664689..ac1e7007a02 100644
--- a/0.16/generated/torchvision.transforms.v2.MixUp.html
+++ b/0.16/generated/torchvision.transforms.v2.MixUp.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.Normalize.html b/0.16/generated/torchvision.transforms.v2.Normalize.html
index 8913d9fd486..4bb6df04ec1 100644
--- a/0.16/generated/torchvision.transforms.v2.Normalize.html
+++ b/0.16/generated/torchvision.transforms.v2.Normalize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.PILToTensor.html b/0.16/generated/torchvision.transforms.v2.PILToTensor.html
index 772004bc8bc..6dd56230afd 100644
--- a/0.16/generated/torchvision.transforms.v2.PILToTensor.html
+++ b/0.16/generated/torchvision.transforms.v2.PILToTensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.Pad.html b/0.16/generated/torchvision.transforms.v2.Pad.html
index b8c28aba5cb..8d95327aa4a 100644
--- a/0.16/generated/torchvision.transforms.v2.Pad.html
+++ b/0.16/generated/torchvision.transforms.v2.Pad.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandAugment.html b/0.16/generated/torchvision.transforms.v2.RandAugment.html
index c6f1254531e..8975cd94803 100644
--- a/0.16/generated/torchvision.transforms.v2.RandAugment.html
+++ b/0.16/generated/torchvision.transforms.v2.RandAugment.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html b/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
index c53f659d5c2..f99fd57015b 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomAdjustSharpness.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAffine.html b/0.16/generated/torchvision.transforms.v2.RandomAffine.html
index 9894b83363a..a2ea21d522e 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomAffine.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomAffine.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomApply.html b/0.16/generated/torchvision.transforms.v2.RandomApply.html
index d0c2fa417da..857697776cc 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomApply.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomApply.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html b/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
index e930f5b0a24..98b9cb7d5ee 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomAutocontrast.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html b/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
index 2771a2c1e6e..3891530ff72 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomChannelPermutation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomChoice.html b/0.16/generated/torchvision.transforms.v2.RandomChoice.html
index 86fd43aed8a..2bc5811c68b 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomChoice.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomChoice.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomCrop.html b/0.16/generated/torchvision.transforms.v2.RandomCrop.html
index 0dccb5acaf4..7c4df96db00 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomEqualize.html b/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
index f380d7d7828..fbe896f9dee 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomEqualize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomErasing.html b/0.16/generated/torchvision.transforms.v2.RandomErasing.html
index fd0823659a2..4dcd7c6aa37 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomErasing.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomErasing.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html b/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
index 7dab1db0a2e..57311fc6680 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomGrayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html b/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
index 1c6345c998e..37c156b55f8 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomHorizontalFlip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomInvert.html b/0.16/generated/torchvision.transforms.v2.RandomInvert.html
index f86816a5a22..a790f43b4d3 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomInvert.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomInvert.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html b/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
index c0a79c62e76..99066789a45 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomIoUCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomOrder.html b/0.16/generated/torchvision.transforms.v2.RandomOrder.html
index 6ac5b838e05..5116f504616 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomOrder.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomOrder.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPerspective.html b/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
index 71660112572..7e3b4ff93d3 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomPerspective.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html b/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
index ba75f996e7c..fc315d756f4 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomPhotometricDistort.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomPosterize.html b/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
index 5a3541ce226..d7ba9dfca9c 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomPosterize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomResize.html b/0.16/generated/torchvision.transforms.v2.RandomResize.html
index e52095ef784..95de94529cb 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomResize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomResize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html b/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
index e7a53c0ead6..4ddb508c75a 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomResizedCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomRotation.html b/0.16/generated/torchvision.transforms.v2.RandomRotation.html
index 773fc5496e8..8a9b861bd8d 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomRotation.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomRotation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html b/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
index 1594851f7d5..01014bb1234 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomShortestSize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomSolarize.html b/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
index 4df12f8e338..ba65ba85324 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomSolarize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html b/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
index 4d5ae885420..5163dac978e 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomVerticalFlip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html b/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
index 09a58ed6b5a..ed1c5f68858 100644
--- a/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
+++ b/0.16/generated/torchvision.transforms.v2.RandomZoomOut.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.Resize.html b/0.16/generated/torchvision.transforms.v2.Resize.html
index 9b15667a83a..62d8421773c 100644
--- a/0.16/generated/torchvision.transforms.v2.Resize.html
+++ b/0.16/generated/torchvision.transforms.v2.Resize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html b/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
index d765a024e31..0f93f97a9fb 100644
--- a/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
+++ b/0.16/generated/torchvision.transforms.v2.SanitizeBoundingBoxes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ScaleJitter.html b/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
index 1d36317d161..d0a87900539 100644
--- a/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
+++ b/0.16/generated/torchvision.transforms.v2.ScaleJitter.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.TenCrop.html b/0.16/generated/torchvision.transforms.v2.TenCrop.html
index f330580bd18..7629310bc6e 100644
--- a/0.16/generated/torchvision.transforms.v2.TenCrop.html
+++ b/0.16/generated/torchvision.transforms.v2.TenCrop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ToDtype.html b/0.16/generated/torchvision.transforms.v2.ToDtype.html
index d1a6b488609..7cc17bbb7e4 100644
--- a/0.16/generated/torchvision.transforms.v2.ToDtype.html
+++ b/0.16/generated/torchvision.transforms.v2.ToDtype.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ToImage.html b/0.16/generated/torchvision.transforms.v2.ToImage.html
index 2e5ddb7bb18..d5c2da3b465 100644
--- a/0.16/generated/torchvision.transforms.v2.ToImage.html
+++ b/0.16/generated/torchvision.transforms.v2.ToImage.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ToPILImage.html b/0.16/generated/torchvision.transforms.v2.ToPILImage.html
index adcd01e6f09..d106ef6c41d 100644
--- a/0.16/generated/torchvision.transforms.v2.ToPILImage.html
+++ b/0.16/generated/torchvision.transforms.v2.ToPILImage.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ToPureTensor.html b/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
index 6a12f6f807c..27b1e9e7056 100644
--- a/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
+++ b/0.16/generated/torchvision.transforms.v2.ToPureTensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.ToTensor.html b/0.16/generated/torchvision.transforms.v2.ToTensor.html
index eca883c502a..545e124f4a1 100644
--- a/0.16/generated/torchvision.transforms.v2.ToTensor.html
+++ b/0.16/generated/torchvision.transforms.v2.ToTensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html b/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
index d699509629a..3f970881c7e 100644
--- a/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
+++ b/0.16/generated/torchvision.transforms.v2.TrivialAugmentWide.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html b/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
index b71a5c15974..55f088675a7 100644
--- a/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
+++ b/0.16/generated/torchvision.transforms.v2.UniformTemporalSubsample.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
index 82419d15340..ef9fdccc329 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_brightness.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
index 1be982a4c2d..0f87cf3565a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_contrast.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
index aaf09c2df82..51e56ec240c 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_gamma.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
index d1711f1a442..08e24d58d97 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_hue.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
index 848f8ee0569..cd9d1f2f02b 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_saturation.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html b/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
index e60944264e6..5de3218859a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.adjust_sharpness.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.affine.html b/0.16/generated/torchvision.transforms.v2.functional.affine.html
index 8d2e009f59c..d4525ce032f 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.affine.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.affine.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html b/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
index 42a9c1d57e4..941b931b259 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.autocontrast.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.center_crop.html b/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
index a0ffff13860..f5dd5ce6352 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.center_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html b/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
index eb2c2cd0792..07d24cbcf0d 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.clamp_bounding_boxes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html b/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
index 545ec23ba1f..95e8b814f7b 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.convert_bounding_box_format.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html b/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
index 0bc2089585a..8b291cf2278 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.convert_image_dtype.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.crop.html b/0.16/generated/torchvision.transforms.v2.functional.crop.html
index 94f90985120..4e8ced82350 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.elastic.html b/0.16/generated/torchvision.transforms.v2.functional.elastic.html
index 3a71e0d0638..a9285513137 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.elastic.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.elastic.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.equalize.html b/0.16/generated/torchvision.transforms.v2.functional.equalize.html
index a59f96135dd..5efd4570818 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.equalize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.equalize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.erase.html b/0.16/generated/torchvision.transforms.v2.functional.erase.html
index 53403d48730..b9fab73900d 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.erase.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.erase.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.five_crop.html b/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
index 8ee38c909fc..3a3c6ed70bf 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.five_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html b/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
index 87bc6de2aad..58a4669b93f 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.gaussian_blur.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html b/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
index a91a765c6e9..17c1de45367 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.horizontal_flip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.invert.html b/0.16/generated/torchvision.transforms.v2.functional.invert.html
index 063485348ef..059bf37395e 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.invert.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.invert.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.normalize.html b/0.16/generated/torchvision.transforms.v2.functional.normalize.html
index e2c04bb0036..e8b7e124c08 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.normalize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.normalize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.pad.html b/0.16/generated/torchvision.transforms.v2.functional.pad.html
index 63b76e53b2e..67bdd82e7c6 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.pad.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.pad.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html b/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
index f8f39cbd603..cd1adb9885a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.permute_channels.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.perspective.html b/0.16/generated/torchvision.transforms.v2.functional.perspective.html
index ce2df88a957..e36022e5631 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.perspective.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.perspective.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html b/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
index 97d240f7071..be234431f90 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.pil_to_tensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.posterize.html b/0.16/generated/torchvision.transforms.v2.functional.posterize.html
index 2c84d20b182..fae503edff9 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.posterize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.posterize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html b/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
index b6dab73e388..af0355cb435 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.register_kernel.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.resize.html b/0.16/generated/torchvision.transforms.v2.functional.resize.html
index b1d116abdbf..6fa8b68313a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.resize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.resize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html b/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
index 1ff98ba562a..6115f7e1e61 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.resized_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html b/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
index 7cd6d943f21..6123358619b 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.rgb_to_grayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.rotate.html b/0.16/generated/torchvision.transforms.v2.functional.rotate.html
index c6c4f642424..3dde02eda6e 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.rotate.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.rotate.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.solarize.html b/0.16/generated/torchvision.transforms.v2.functional.solarize.html
index 0d5de7c198c..13e706814e5 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.solarize.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.solarize.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html b/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
index 2af6b66a27c..41bb903917c 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.ten_crop.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html b/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
index b40acca4531..fa5761dc8da 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_dtype.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html b/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
index 396f44d036e..027d60a9c6d 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_grayscale.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_image.html b/0.16/generated/torchvision.transforms.v2.functional.to_image.html
index 5037ea2a35a..efe8d4614d6 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_image.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html b/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
index c14952f489b..4fabccafb8a 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_pil_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html b/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
index e089ac34e8f..7872bfe5568 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.to_tensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html b/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
index 26062711fbc..752d500ee02 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.uniform_temporal_subsample.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html b/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
index 9bec53a3497..22da63d1ebe 100644
--- a/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
+++ b/0.16/generated/torchvision.transforms.v2.functional.vertical_flip.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html b/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
index e3ce955c129..cc83b7c0318 100644
--- a/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
+++ b/0.16/generated/torchvision.tv_tensors.BoundingBoxFormat.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html b/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
index 0292fcccff3..ffeefb21968 100644
--- a/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
+++ b/0.16/generated/torchvision.tv_tensors.BoundingBoxes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.Image.html b/0.16/generated/torchvision.tv_tensors.Image.html
index 72b5f0e18c3..7443bdd8891 100644
--- a/0.16/generated/torchvision.tv_tensors.Image.html
+++ b/0.16/generated/torchvision.tv_tensors.Image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.Mask.html b/0.16/generated/torchvision.tv_tensors.Mask.html
index d77942ff318..977b2701ae1 100644
--- a/0.16/generated/torchvision.tv_tensors.Mask.html
+++ b/0.16/generated/torchvision.tv_tensors.Mask.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.TVTensor.html b/0.16/generated/torchvision.tv_tensors.TVTensor.html
index 9a7f2fda18c..a68a2a94056 100644
--- a/0.16/generated/torchvision.tv_tensors.TVTensor.html
+++ b/0.16/generated/torchvision.tv_tensors.TVTensor.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.Video.html b/0.16/generated/torchvision.tv_tensors.Video.html
index 58e6348c3be..61b74c139e6 100644
--- a/0.16/generated/torchvision.tv_tensors.Video.html
+++ b/0.16/generated/torchvision.tv_tensors.Video.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.set_return_type.html b/0.16/generated/torchvision.tv_tensors.set_return_type.html
index 920c6baf63f..96af0b94804 100644
--- a/0.16/generated/torchvision.tv_tensors.set_return_type.html
+++ b/0.16/generated/torchvision.tv_tensors.set_return_type.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.tv_tensors.wrap.html b/0.16/generated/torchvision.tv_tensors.wrap.html
index 6615945fde9..37b56dff56e 100644
--- a/0.16/generated/torchvision.tv_tensors.wrap.html
+++ b/0.16/generated/torchvision.tv_tensors.wrap.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.utils.draw_bounding_boxes.html b/0.16/generated/torchvision.utils.draw_bounding_boxes.html
index 8af44a05dd3..a55dcd99b51 100644
--- a/0.16/generated/torchvision.utils.draw_bounding_boxes.html
+++ b/0.16/generated/torchvision.utils.draw_bounding_boxes.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.utils.draw_keypoints.html b/0.16/generated/torchvision.utils.draw_keypoints.html
index 16b385699a1..5e981323606 100644
--- a/0.16/generated/torchvision.utils.draw_keypoints.html
+++ b/0.16/generated/torchvision.utils.draw_keypoints.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.utils.draw_segmentation_masks.html b/0.16/generated/torchvision.utils.draw_segmentation_masks.html
index dafdf93bfd7..35231b791ac 100644
--- a/0.16/generated/torchvision.utils.draw_segmentation_masks.html
+++ b/0.16/generated/torchvision.utils.draw_segmentation_masks.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.utils.flow_to_image.html b/0.16/generated/torchvision.utils.flow_to_image.html
index bca6fec3764..0ef9a1e1732 100644
--- a/0.16/generated/torchvision.utils.flow_to_image.html
+++ b/0.16/generated/torchvision.utils.flow_to_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.utils.make_grid.html b/0.16/generated/torchvision.utils.make_grid.html
index 53a3a1fb966..266e2adbefc 100644
--- a/0.16/generated/torchvision.utils.make_grid.html
+++ b/0.16/generated/torchvision.utils.make_grid.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/torchvision.utils.save_image.html b/0.16/generated/torchvision.utils.save_image.html
index d26918d1049..67bb2c5d68d 100644
--- a/0.16/generated/torchvision.utils.save_image.html
+++ b/0.16/generated/torchvision.utils.save_image.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/generated/video_table.html b/0.16/generated/video_table.html
index 8369937e8ce..61b243a6f4d 100644
--- a/0.16/generated/video_table.html
+++ b/0.16/generated/video_table.html
@@ -218,7 +218,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/genindex.html b/0.16/genindex.html
index e5ba553d1d7..f0d56b44cbf 100644
--- a/0.16/genindex.html
+++ b/0.16/genindex.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/index.html b/0.16/index.html
index 9138f681381..3ccc6676bb2 100644
--- a/0.16/index.html
+++ b/0.16/index.html
@@ -219,7 +219,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/io.html b/0.16/io.html
index d4524e27d17..3ddbf10e234 100644
--- a/0.16/io.html
+++ b/0.16/io.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models.html b/0.16/models.html
index 0c27afadb28..fdef7e376be 100644
--- a/0.16/models.html
+++ b/0.16/models.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/alexnet.html b/0.16/models/alexnet.html
index a7edec8a02c..acd6f04952b 100644
--- a/0.16/models/alexnet.html
+++ b/0.16/models/alexnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/convnext.html b/0.16/models/convnext.html
index 13e2700db0c..be9e8404093 100644
--- a/0.16/models/convnext.html
+++ b/0.16/models/convnext.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/deeplabv3.html b/0.16/models/deeplabv3.html
index 381916ef56a..a4adf3f7bb0 100644
--- a/0.16/models/deeplabv3.html
+++ b/0.16/models/deeplabv3.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/densenet.html b/0.16/models/densenet.html
index e9cb95b32e6..423bf08ea71 100644
--- a/0.16/models/densenet.html
+++ b/0.16/models/densenet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/efficientnet.html b/0.16/models/efficientnet.html
index b3c36bc93fa..25ae773f27a 100644
--- a/0.16/models/efficientnet.html
+++ b/0.16/models/efficientnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/efficientnetv2.html b/0.16/models/efficientnetv2.html
index 45fab004c55..dd3b8fe10dd 100644
--- a/0.16/models/efficientnetv2.html
+++ b/0.16/models/efficientnetv2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/faster_rcnn.html b/0.16/models/faster_rcnn.html
index f639a595dcd..cd99f031d91 100644
--- a/0.16/models/faster_rcnn.html
+++ b/0.16/models/faster_rcnn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/fcn.html b/0.16/models/fcn.html
index d0f5024e94c..063f7afcf9b 100644
--- a/0.16/models/fcn.html
+++ b/0.16/models/fcn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/fcos.html b/0.16/models/fcos.html
index 5cbcb7827ac..00392ece5f8 100644
--- a/0.16/models/fcos.html
+++ b/0.16/models/fcos.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.alexnet.html b/0.16/models/generated/torchvision.models.alexnet.html
index 7cfdd307271..ffc577f8134 100644
--- a/0.16/models/generated/torchvision.models.alexnet.html
+++ b/0.16/models/generated/torchvision.models.alexnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.convnext_base.html b/0.16/models/generated/torchvision.models.convnext_base.html
index 784110af317..3287734b9d2 100644
--- a/0.16/models/generated/torchvision.models.convnext_base.html
+++ b/0.16/models/generated/torchvision.models.convnext_base.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.convnext_large.html b/0.16/models/generated/torchvision.models.convnext_large.html
index 7a4d4dbb2d9..ff22cd0cef3 100644
--- a/0.16/models/generated/torchvision.models.convnext_large.html
+++ b/0.16/models/generated/torchvision.models.convnext_large.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.convnext_small.html b/0.16/models/generated/torchvision.models.convnext_small.html
index de8ae26149a..a02aa88677c 100644
--- a/0.16/models/generated/torchvision.models.convnext_small.html
+++ b/0.16/models/generated/torchvision.models.convnext_small.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.convnext_tiny.html b/0.16/models/generated/torchvision.models.convnext_tiny.html
index c644597c772..15c8f32a9b1 100644
--- a/0.16/models/generated/torchvision.models.convnext_tiny.html
+++ b/0.16/models/generated/torchvision.models.convnext_tiny.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.densenet121.html b/0.16/models/generated/torchvision.models.densenet121.html
index bb1b1157eeb..b10a441f62e 100644
--- a/0.16/models/generated/torchvision.models.densenet121.html
+++ b/0.16/models/generated/torchvision.models.densenet121.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.densenet161.html b/0.16/models/generated/torchvision.models.densenet161.html
index 246093881ec..33901bc385a 100644
--- a/0.16/models/generated/torchvision.models.densenet161.html
+++ b/0.16/models/generated/torchvision.models.densenet161.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.densenet169.html b/0.16/models/generated/torchvision.models.densenet169.html
index e6ec34243a2..2a2b0c401e3 100644
--- a/0.16/models/generated/torchvision.models.densenet169.html
+++ b/0.16/models/generated/torchvision.models.densenet169.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.densenet201.html b/0.16/models/generated/torchvision.models.densenet201.html
index 0c032a38397..90fac52429f 100644
--- a/0.16/models/generated/torchvision.models.densenet201.html
+++ b/0.16/models/generated/torchvision.models.densenet201.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
index c8c8f8df827..784181891e6 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_320_fpn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
index 2a6a6ed1202..8d01d1b4a6b 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_mobilenet_v3_large_fpn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
index 6eac0a7455d..139cabe326e 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
index a544f5eb65d..c4e5d9a99e6 100644
--- a/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
+++ b/0.16/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
index 0348483dff2..368d6b9e5ae 100644
--- a/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.fcos_resnet50_fpn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
index 814d0ae1a59..89e8461615f 100644
--- a/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.keypointrcnn_resnet50_fpn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
index 780fae52150..6ab814e2a4a 100644
--- a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
index 43443ca242d..bb6a501a937 100644
--- a/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
+++ b/0.16/models/generated/torchvision.models.detection.maskrcnn_resnet50_fpn_v2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
index 9b14ae8937a..16fd9b8158f 100644
--- a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
+++ b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
index 7d4d6211050..fd869f725ad 100644
--- a/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
+++ b/0.16/models/generated/torchvision.models.detection.retinanet_resnet50_fpn_v2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html b/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
index 9462093e3dd..723b29c1526 100644
--- a/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
+++ b/0.16/models/generated/torchvision.models.detection.ssd300_vgg16.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
index 2d0fce77d52..111dd425bcd 100644
--- a/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.detection.ssdlite320_mobilenet_v3_large.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b0.html b/0.16/models/generated/torchvision.models.efficientnet_b0.html
index 9ec12afd016..a904ff190e3 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b0.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b0.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b1.html b/0.16/models/generated/torchvision.models.efficientnet_b1.html
index 77e28fdf533..9376e5b5b16 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b1.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b1.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b2.html b/0.16/models/generated/torchvision.models.efficientnet_b2.html
index e22abde1c4c..02b04e325a8 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b2.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b3.html b/0.16/models/generated/torchvision.models.efficientnet_b3.html
index 5bfabbf8936..14899868d88 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b3.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b3.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b4.html b/0.16/models/generated/torchvision.models.efficientnet_b4.html
index 90e29f85ebc..ab8d2f3020e 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b4.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b4.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b5.html b/0.16/models/generated/torchvision.models.efficientnet_b5.html
index aa74d2068f5..f4dd74afc1e 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b5.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b5.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b6.html b/0.16/models/generated/torchvision.models.efficientnet_b6.html
index 49a6cbe4c24..16ea947ce8e 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b6.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b6.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_b7.html b/0.16/models/generated/torchvision.models.efficientnet_b7.html
index d73e186511a..9af8b69b0c3 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_b7.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_b7.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_l.html b/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
index 0216545b44d..d1dccb51af0 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_l.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_m.html b/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
index 5d6a14d6d37..c69a6c879f0 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_m.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.efficientnet_v2_s.html b/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
index e503cabec46..e9f83b2cc87 100644
--- a/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
+++ b/0.16/models/generated/torchvision.models.efficientnet_v2_s.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.googlenet.html b/0.16/models/generated/torchvision.models.googlenet.html
index 47de7f67533..80b7cc71ba2 100644
--- a/0.16/models/generated/torchvision.models.googlenet.html
+++ b/0.16/models/generated/torchvision.models.googlenet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.inception_v3.html b/0.16/models/generated/torchvision.models.inception_v3.html
index 4f08b38ff3b..8672320af07 100644
--- a/0.16/models/generated/torchvision.models.inception_v3.html
+++ b/0.16/models/generated/torchvision.models.inception_v3.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.maxvit_t.html b/0.16/models/generated/torchvision.models.maxvit_t.html
index 4f4c297aebb..2e99831aef7 100644
--- a/0.16/models/generated/torchvision.models.maxvit_t.html
+++ b/0.16/models/generated/torchvision.models.maxvit_t.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.mnasnet0_5.html b/0.16/models/generated/torchvision.models.mnasnet0_5.html
index 7f6f03d6fd5..df3629bfa85 100644
--- a/0.16/models/generated/torchvision.models.mnasnet0_5.html
+++ b/0.16/models/generated/torchvision.models.mnasnet0_5.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.mnasnet0_75.html b/0.16/models/generated/torchvision.models.mnasnet0_75.html
index 2bcd4471f50..b76c628b9b3 100644
--- a/0.16/models/generated/torchvision.models.mnasnet0_75.html
+++ b/0.16/models/generated/torchvision.models.mnasnet0_75.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.mnasnet1_0.html b/0.16/models/generated/torchvision.models.mnasnet1_0.html
index 6a2e10bddc5..4e361a37df3 100644
--- a/0.16/models/generated/torchvision.models.mnasnet1_0.html
+++ b/0.16/models/generated/torchvision.models.mnasnet1_0.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.mnasnet1_3.html b/0.16/models/generated/torchvision.models.mnasnet1_3.html
index 8133f52f9f7..719c6b7f928 100644
--- a/0.16/models/generated/torchvision.models.mnasnet1_3.html
+++ b/0.16/models/generated/torchvision.models.mnasnet1_3.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v2.html b/0.16/models/generated/torchvision.models.mobilenet_v2.html
index 6fb7e93c69f..0d4d7cefe16 100644
--- a/0.16/models/generated/torchvision.models.mobilenet_v2.html
+++ b/0.16/models/generated/torchvision.models.mobilenet_v2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
index e0b4fe38460..a3e19187c2f 100644
--- a/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.mobilenet_v3_large.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.mobilenet_v3_small.html b/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
index cddd4549433..cd0ac006c93 100644
--- a/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
+++ b/0.16/models/generated/torchvision.models.mobilenet_v3_small.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.optical_flow.raft_large.html b/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
index 186b3f02ba7..1ca983f2a16 100644
--- a/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
+++ b/0.16/models/generated/torchvision.models.optical_flow.raft_large.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.optical_flow.raft_small.html b/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
index f1e1e3a8a40..552da3b1877 100644
--- a/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
+++ b/0.16/models/generated/torchvision.models.optical_flow.raft_small.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.googlenet.html b/0.16/models/generated/torchvision.models.quantization.googlenet.html
index 8efc2459fc9..65b94d1613c 100644
--- a/0.16/models/generated/torchvision.models.quantization.googlenet.html
+++ b/0.16/models/generated/torchvision.models.quantization.googlenet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.inception_v3.html b/0.16/models/generated/torchvision.models.quantization.inception_v3.html
index 77f16251fdf..e7b1bc08751 100644
--- a/0.16/models/generated/torchvision.models.quantization.inception_v3.html
+++ b/0.16/models/generated/torchvision.models.quantization.inception_v3.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html b/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
index d3b55b0b5ce..ea66eae4a22 100644
--- a/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
+++ b/0.16/models/generated/torchvision.models.quantization.mobilenet_v2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
index 6f7750a68e6..86ec88b9d3c 100644
--- a/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.quantization.mobilenet_v3_large.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.resnet18.html b/0.16/models/generated/torchvision.models.quantization.resnet18.html
index 880a9f84ac3..b0cd08c885d 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnet18.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnet18.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.resnet50.html b/0.16/models/generated/torchvision.models.quantization.resnet50.html
index d98a86f82dd..faf7c31939d 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnet50.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnet50.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html b/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
index 73556f7b371..93badb753fd 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnext101_32x8d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html b/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
index 0ccb4f69bb9..417139b890a 100644
--- a/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
+++ b/0.16/models/generated/torchvision.models.quantization.resnext101_64x4d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
index 1d8cc82e255..b93263727cd 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x0_5.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
index c090a6b96d9..3929d69cefc 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_0.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
index d65ef816df3..d1bbced3456 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x1_5.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
index 8def6dfd7dd..694b81c0c87 100644
--- a/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
+++ b/0.16/models/generated/torchvision.models.quantization.shufflenet_v2_x2_0.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_x_16gf.html b/0.16/models/generated/torchvision.models.regnet_x_16gf.html
index 5974897e5ca..781f5ce19ed 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_16gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_16gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html b/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
index 669988d37c6..07234081bdb 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_1_6gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_x_32gf.html b/0.16/models/generated/torchvision.models.regnet_x_32gf.html
index b230001b8e0..3032506b248 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_32gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_32gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html b/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
index a6ac0644093..98bc4274de4 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_3_2gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_x_400mf.html b/0.16/models/generated/torchvision.models.regnet_x_400mf.html
index 135864ebb30..a92517a2bd8 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_400mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_400mf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_x_800mf.html b/0.16/models/generated/torchvision.models.regnet_x_800mf.html
index 40fb5985662..6b8e12cd4ef 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_800mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_800mf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_x_8gf.html b/0.16/models/generated/torchvision.models.regnet_x_8gf.html
index a8dab23d246..fbc3aaf9542 100644
--- a/0.16/models/generated/torchvision.models.regnet_x_8gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_x_8gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_128gf.html b/0.16/models/generated/torchvision.models.regnet_y_128gf.html
index 194ccb553c0..926ffa4baa1 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_128gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_128gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_16gf.html b/0.16/models/generated/torchvision.models.regnet_y_16gf.html
index 161019b716c..7dc004dd896 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_16gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_16gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html b/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
index f8cc39d8bff..af59230d595 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_1_6gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_32gf.html b/0.16/models/generated/torchvision.models.regnet_y_32gf.html
index 482d52dba39..d3b4cd5d348 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_32gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_32gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html b/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
index 226b13a947e..118a879fbf8 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_3_2gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_400mf.html b/0.16/models/generated/torchvision.models.regnet_y_400mf.html
index 716394ec8d3..9c216550356 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_400mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_400mf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_800mf.html b/0.16/models/generated/torchvision.models.regnet_y_800mf.html
index eaa1af855b8..687c5ca9447 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_800mf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_800mf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.regnet_y_8gf.html b/0.16/models/generated/torchvision.models.regnet_y_8gf.html
index dc4161be285..7a07caa319d 100644
--- a/0.16/models/generated/torchvision.models.regnet_y_8gf.html
+++ b/0.16/models/generated/torchvision.models.regnet_y_8gf.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnet101.html b/0.16/models/generated/torchvision.models.resnet101.html
index df7f0e01237..03db7e38de5 100644
--- a/0.16/models/generated/torchvision.models.resnet101.html
+++ b/0.16/models/generated/torchvision.models.resnet101.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnet152.html b/0.16/models/generated/torchvision.models.resnet152.html
index 53b474e7b59..8eff7eaabf6 100644
--- a/0.16/models/generated/torchvision.models.resnet152.html
+++ b/0.16/models/generated/torchvision.models.resnet152.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnet18.html b/0.16/models/generated/torchvision.models.resnet18.html
index 95470b49960..ff2e35f06e2 100644
--- a/0.16/models/generated/torchvision.models.resnet18.html
+++ b/0.16/models/generated/torchvision.models.resnet18.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnet34.html b/0.16/models/generated/torchvision.models.resnet34.html
index 17363719208..716afd4d9e0 100644
--- a/0.16/models/generated/torchvision.models.resnet34.html
+++ b/0.16/models/generated/torchvision.models.resnet34.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnet50.html b/0.16/models/generated/torchvision.models.resnet50.html
index e3a42f1830b..d72d2d3d329 100644
--- a/0.16/models/generated/torchvision.models.resnet50.html
+++ b/0.16/models/generated/torchvision.models.resnet50.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnext101_32x8d.html b/0.16/models/generated/torchvision.models.resnext101_32x8d.html
index bfc941c429d..44784910882 100644
--- a/0.16/models/generated/torchvision.models.resnext101_32x8d.html
+++ b/0.16/models/generated/torchvision.models.resnext101_32x8d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnext101_64x4d.html b/0.16/models/generated/torchvision.models.resnext101_64x4d.html
index 78fd699494e..5c3a99845c9 100644
--- a/0.16/models/generated/torchvision.models.resnext101_64x4d.html
+++ b/0.16/models/generated/torchvision.models.resnext101_64x4d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.resnext50_32x4d.html b/0.16/models/generated/torchvision.models.resnext50_32x4d.html
index 2ede22a0e2a..5b327a543de 100644
--- a/0.16/models/generated/torchvision.models.resnext50_32x4d.html
+++ b/0.16/models/generated/torchvision.models.resnext50_32x4d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
index e16a18392db..80f2291c08a 100644
--- a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_mobilenet_v3_large.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
index b821f6f3170..0a069de36f6 100644
--- a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet101.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
index 8f33be90f5a..3d46033b7a9 100644
--- a/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
+++ b/0.16/models/generated/torchvision.models.segmentation.deeplabv3_resnet50.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
index d9f3edb0305..9818bf77359 100644
--- a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
+++ b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet101.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
index f6afd02b1b5..e4dcb8412f8 100644
--- a/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
+++ b/0.16/models/generated/torchvision.models.segmentation.fcn_resnet50.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html b/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
index 3819f3de113..5e98dbfb1f1 100644
--- a/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
+++ b/0.16/models/generated/torchvision.models.segmentation.lraspp_mobilenet_v3_large.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
index c3bd8fbe5b3..8dd9aeba0ed 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x0_5.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
index af710e7675d..caa770a4a54 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_0.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
index 8da3a711130..2e6b4151b7b 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x1_5.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html b/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
index 490f8a69151..de594289e42 100644
--- a/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
+++ b/0.16/models/generated/torchvision.models.shufflenet_v2_x2_0.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.squeezenet1_0.html b/0.16/models/generated/torchvision.models.squeezenet1_0.html
index c67aeae745b..494d556d9b6 100644
--- a/0.16/models/generated/torchvision.models.squeezenet1_0.html
+++ b/0.16/models/generated/torchvision.models.squeezenet1_0.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.squeezenet1_1.html b/0.16/models/generated/torchvision.models.squeezenet1_1.html
index 7667930155e..9d254fb2209 100644
--- a/0.16/models/generated/torchvision.models.squeezenet1_1.html
+++ b/0.16/models/generated/torchvision.models.squeezenet1_1.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.swin_b.html b/0.16/models/generated/torchvision.models.swin_b.html
index 47c0e8c7a75..482efdd690e 100644
--- a/0.16/models/generated/torchvision.models.swin_b.html
+++ b/0.16/models/generated/torchvision.models.swin_b.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.swin_s.html b/0.16/models/generated/torchvision.models.swin_s.html
index 3b27bd73176..3c9694fe684 100644
--- a/0.16/models/generated/torchvision.models.swin_s.html
+++ b/0.16/models/generated/torchvision.models.swin_s.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.swin_t.html b/0.16/models/generated/torchvision.models.swin_t.html
index 2d0f3e1dfd1..1f00b4493e5 100644
--- a/0.16/models/generated/torchvision.models.swin_t.html
+++ b/0.16/models/generated/torchvision.models.swin_t.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.swin_v2_b.html b/0.16/models/generated/torchvision.models.swin_v2_b.html
index 3f12f321553..bf2b33a53d8 100644
--- a/0.16/models/generated/torchvision.models.swin_v2_b.html
+++ b/0.16/models/generated/torchvision.models.swin_v2_b.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.swin_v2_s.html b/0.16/models/generated/torchvision.models.swin_v2_s.html
index 3b74f9fa3ff..822c848f68b 100644
--- a/0.16/models/generated/torchvision.models.swin_v2_s.html
+++ b/0.16/models/generated/torchvision.models.swin_v2_s.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.swin_v2_t.html b/0.16/models/generated/torchvision.models.swin_v2_t.html
index c43048e0feb..d68c9a62956 100644
--- a/0.16/models/generated/torchvision.models.swin_v2_t.html
+++ b/0.16/models/generated/torchvision.models.swin_v2_t.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg11.html b/0.16/models/generated/torchvision.models.vgg11.html
index 750ebb203f4..a446bd1e852 100644
--- a/0.16/models/generated/torchvision.models.vgg11.html
+++ b/0.16/models/generated/torchvision.models.vgg11.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg11_bn.html b/0.16/models/generated/torchvision.models.vgg11_bn.html
index f6a8c475de7..6fa15e5c4ef 100644
--- a/0.16/models/generated/torchvision.models.vgg11_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg11_bn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg13.html b/0.16/models/generated/torchvision.models.vgg13.html
index 62488ef6a94..825662923f5 100644
--- a/0.16/models/generated/torchvision.models.vgg13.html
+++ b/0.16/models/generated/torchvision.models.vgg13.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg13_bn.html b/0.16/models/generated/torchvision.models.vgg13_bn.html
index 93ca2101183..55c290166ed 100644
--- a/0.16/models/generated/torchvision.models.vgg13_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg13_bn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg16.html b/0.16/models/generated/torchvision.models.vgg16.html
index 7ccefbfbfce..36749ab32cf 100644
--- a/0.16/models/generated/torchvision.models.vgg16.html
+++ b/0.16/models/generated/torchvision.models.vgg16.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg16_bn.html b/0.16/models/generated/torchvision.models.vgg16_bn.html
index bf7f0d43ed6..3d3c5dfbf84 100644
--- a/0.16/models/generated/torchvision.models.vgg16_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg16_bn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg19.html b/0.16/models/generated/torchvision.models.vgg19.html
index 683bce4c7be..b2ea205cf72 100644
--- a/0.16/models/generated/torchvision.models.vgg19.html
+++ b/0.16/models/generated/torchvision.models.vgg19.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vgg19_bn.html b/0.16/models/generated/torchvision.models.vgg19_bn.html
index 28ab8cf57e2..5a2c76c5205 100644
--- a/0.16/models/generated/torchvision.models.vgg19_bn.html
+++ b/0.16/models/generated/torchvision.models.vgg19_bn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.mc3_18.html b/0.16/models/generated/torchvision.models.video.mc3_18.html
index 576892f2268..cba5d09b80a 100644
--- a/0.16/models/generated/torchvision.models.video.mc3_18.html
+++ b/0.16/models/generated/torchvision.models.video.mc3_18.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.mvit_v1_b.html b/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
index 16ebe20bcbe..566376a8b13 100644
--- a/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
+++ b/0.16/models/generated/torchvision.models.video.mvit_v1_b.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.mvit_v2_s.html b/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
index 71d519176fa..863b7184647 100644
--- a/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
+++ b/0.16/models/generated/torchvision.models.video.mvit_v2_s.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.r2plus1d_18.html b/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
index d3a5e826bd3..bd096642efa 100644
--- a/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
+++ b/0.16/models/generated/torchvision.models.video.r2plus1d_18.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.r3d_18.html b/0.16/models/generated/torchvision.models.video.r3d_18.html
index 0a1194f2f2d..35e101db465 100644
--- a/0.16/models/generated/torchvision.models.video.r3d_18.html
+++ b/0.16/models/generated/torchvision.models.video.r3d_18.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.s3d.html b/0.16/models/generated/torchvision.models.video.s3d.html
index 2dbcf53a4ee..0046f7b97b6 100644
--- a/0.16/models/generated/torchvision.models.video.s3d.html
+++ b/0.16/models/generated/torchvision.models.video.s3d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_b.html b/0.16/models/generated/torchvision.models.video.swin3d_b.html
index 80d3cc900c5..d7c369583bb 100644
--- a/0.16/models/generated/torchvision.models.video.swin3d_b.html
+++ b/0.16/models/generated/torchvision.models.video.swin3d_b.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_s.html b/0.16/models/generated/torchvision.models.video.swin3d_s.html
index 77f9aca2c63..79f1727d338 100644
--- a/0.16/models/generated/torchvision.models.video.swin3d_s.html
+++ b/0.16/models/generated/torchvision.models.video.swin3d_s.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.video.swin3d_t.html b/0.16/models/generated/torchvision.models.video.swin3d_t.html
index 59799876633..431673efcd4 100644
--- a/0.16/models/generated/torchvision.models.video.swin3d_t.html
+++ b/0.16/models/generated/torchvision.models.video.swin3d_t.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vit_b_16.html b/0.16/models/generated/torchvision.models.vit_b_16.html
index e72966c9cb3..93e1eef7df3 100644
--- a/0.16/models/generated/torchvision.models.vit_b_16.html
+++ b/0.16/models/generated/torchvision.models.vit_b_16.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vit_b_32.html b/0.16/models/generated/torchvision.models.vit_b_32.html
index 4c93fd474bf..6dc4c38f9df 100644
--- a/0.16/models/generated/torchvision.models.vit_b_32.html
+++ b/0.16/models/generated/torchvision.models.vit_b_32.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vit_h_14.html b/0.16/models/generated/torchvision.models.vit_h_14.html
index ae1e2f206f4..b95a05df934 100644
--- a/0.16/models/generated/torchvision.models.vit_h_14.html
+++ b/0.16/models/generated/torchvision.models.vit_h_14.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vit_l_16.html b/0.16/models/generated/torchvision.models.vit_l_16.html
index 6e54825fed1..ccda29e2500 100644
--- a/0.16/models/generated/torchvision.models.vit_l_16.html
+++ b/0.16/models/generated/torchvision.models.vit_l_16.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.vit_l_32.html b/0.16/models/generated/torchvision.models.vit_l_32.html
index 6e0e331944a..078f127122e 100644
--- a/0.16/models/generated/torchvision.models.vit_l_32.html
+++ b/0.16/models/generated/torchvision.models.vit_l_32.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.wide_resnet101_2.html b/0.16/models/generated/torchvision.models.wide_resnet101_2.html
index a602c07edd0..0f34adc8df0 100644
--- a/0.16/models/generated/torchvision.models.wide_resnet101_2.html
+++ b/0.16/models/generated/torchvision.models.wide_resnet101_2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/generated/torchvision.models.wide_resnet50_2.html b/0.16/models/generated/torchvision.models.wide_resnet50_2.html
index 365d27b7e2b..5d27324beb3 100644
--- a/0.16/models/generated/torchvision.models.wide_resnet50_2.html
+++ b/0.16/models/generated/torchvision.models.wide_resnet50_2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/googlenet.html b/0.16/models/googlenet.html
index e99f0841b09..2c2a3fbc82f 100644
--- a/0.16/models/googlenet.html
+++ b/0.16/models/googlenet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/googlenet_quant.html b/0.16/models/googlenet_quant.html
index 0fd40edb99f..0449f7096bd 100644
--- a/0.16/models/googlenet_quant.html
+++ b/0.16/models/googlenet_quant.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/inception.html b/0.16/models/inception.html
index 186ada24c52..d0ccc0eb95e 100644
--- a/0.16/models/inception.html
+++ b/0.16/models/inception.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/inception_quant.html b/0.16/models/inception_quant.html
index e9d5a0c27ff..997306e08bf 100644
--- a/0.16/models/inception_quant.html
+++ b/0.16/models/inception_quant.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/keypoint_rcnn.html b/0.16/models/keypoint_rcnn.html
index 5d2794691d8..fa3772c8845 100644
--- a/0.16/models/keypoint_rcnn.html
+++ b/0.16/models/keypoint_rcnn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/lraspp.html b/0.16/models/lraspp.html
index 95ef22ed2da..9c115ec6d99 100644
--- a/0.16/models/lraspp.html
+++ b/0.16/models/lraspp.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/mask_rcnn.html b/0.16/models/mask_rcnn.html
index dc4cedbe79c..ff8f931adb7 100644
--- a/0.16/models/mask_rcnn.html
+++ b/0.16/models/mask_rcnn.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/maxvit.html b/0.16/models/maxvit.html
index 580995c932a..c8ae6a0b762 100644
--- a/0.16/models/maxvit.html
+++ b/0.16/models/maxvit.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/mnasnet.html b/0.16/models/mnasnet.html
index c4e3ea5bd4c..135b699b46f 100644
--- a/0.16/models/mnasnet.html
+++ b/0.16/models/mnasnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/mobilenetv2.html b/0.16/models/mobilenetv2.html
index 55cc7ceccf9..0eb1453b982 100644
--- a/0.16/models/mobilenetv2.html
+++ b/0.16/models/mobilenetv2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/mobilenetv2_quant.html b/0.16/models/mobilenetv2_quant.html
index 87f7fdf4f45..ea66322c0af 100644
--- a/0.16/models/mobilenetv2_quant.html
+++ b/0.16/models/mobilenetv2_quant.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/mobilenetv3.html b/0.16/models/mobilenetv3.html
index 74d6d2a63ba..857c4940ca9 100644
--- a/0.16/models/mobilenetv3.html
+++ b/0.16/models/mobilenetv3.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/mobilenetv3_quant.html b/0.16/models/mobilenetv3_quant.html
index e756c277879..6dfea7a8780 100644
--- a/0.16/models/mobilenetv3_quant.html
+++ b/0.16/models/mobilenetv3_quant.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/raft.html b/0.16/models/raft.html
index b55b616e8ab..5f91219b258 100644
--- a/0.16/models/raft.html
+++ b/0.16/models/raft.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/regnet.html b/0.16/models/regnet.html
index eaa6c4143ba..0769b53e120 100644
--- a/0.16/models/regnet.html
+++ b/0.16/models/regnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/resnet.html b/0.16/models/resnet.html
index 009a23cf157..d5f10e21f6b 100644
--- a/0.16/models/resnet.html
+++ b/0.16/models/resnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/resnet_quant.html b/0.16/models/resnet_quant.html
index d4b65917178..630eac62b02 100644
--- a/0.16/models/resnet_quant.html
+++ b/0.16/models/resnet_quant.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/resnext.html b/0.16/models/resnext.html
index 243e239f728..ead132a993d 100644
--- a/0.16/models/resnext.html
+++ b/0.16/models/resnext.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/resnext_quant.html b/0.16/models/resnext_quant.html
index 116d4cb321b..6fa545fbb18 100644
--- a/0.16/models/resnext_quant.html
+++ b/0.16/models/resnext_quant.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/retinanet.html b/0.16/models/retinanet.html
index 4368b29b773..21ec4aeeb86 100644
--- a/0.16/models/retinanet.html
+++ b/0.16/models/retinanet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/shufflenetv2.html b/0.16/models/shufflenetv2.html
index 2c76e10f24a..68ae264eba3 100644
--- a/0.16/models/shufflenetv2.html
+++ b/0.16/models/shufflenetv2.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/shufflenetv2_quant.html b/0.16/models/shufflenetv2_quant.html
index 27ad8df1efa..a1109757467 100644
--- a/0.16/models/shufflenetv2_quant.html
+++ b/0.16/models/shufflenetv2_quant.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/squeezenet.html b/0.16/models/squeezenet.html
index ae9f17db3b7..98ee87b6aba 100644
--- a/0.16/models/squeezenet.html
+++ b/0.16/models/squeezenet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/ssd.html b/0.16/models/ssd.html
index f67d5b52c25..63defa4321d 100644
--- a/0.16/models/ssd.html
+++ b/0.16/models/ssd.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/ssdlite.html b/0.16/models/ssdlite.html
index 0fdd12eb8b3..13fcd1957e6 100644
--- a/0.16/models/ssdlite.html
+++ b/0.16/models/ssdlite.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/swin_transformer.html b/0.16/models/swin_transformer.html
index 8495903bd0a..daee4fe38cd 100644
--- a/0.16/models/swin_transformer.html
+++ b/0.16/models/swin_transformer.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/vgg.html b/0.16/models/vgg.html
index ec427d8e528..0b441ea574e 100644
--- a/0.16/models/vgg.html
+++ b/0.16/models/vgg.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/video_mvit.html b/0.16/models/video_mvit.html
index 99d0943910a..da5a95c0357 100644
--- a/0.16/models/video_mvit.html
+++ b/0.16/models/video_mvit.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/video_resnet.html b/0.16/models/video_resnet.html
index 6d9d315729c..318fa1620b7 100644
--- a/0.16/models/video_resnet.html
+++ b/0.16/models/video_resnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/video_s3d.html b/0.16/models/video_s3d.html
index 345b91759a4..c1ce8c0a2b5 100644
--- a/0.16/models/video_s3d.html
+++ b/0.16/models/video_s3d.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/video_swin_transformer.html b/0.16/models/video_swin_transformer.html
index e1c1d2beff8..649eb7b2b76 100644
--- a/0.16/models/video_swin_transformer.html
+++ b/0.16/models/video_swin_transformer.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/vision_transformer.html b/0.16/models/vision_transformer.html
index c12e81368b8..cf3c805e696 100644
--- a/0.16/models/vision_transformer.html
+++ b/0.16/models/vision_transformer.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/models/wide_resnet.html b/0.16/models/wide_resnet.html
index 0175a2e6177..41ff8321b22 100644
--- a/0.16/models/wide_resnet.html
+++ b/0.16/models/wide_resnet.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/objects.inv b/0.16/objects.inv
index ab8893bf8f206bbacec857c221ab6c113119b8c1..2426f83ea41e828a37c333a5939c59f6e1bb4513 100644
GIT binary patch
delta 17
YcmZ27k#X`w#tHT;26~2O8$$v-05$IgkpKVy

delta 34
pcmbO{k#WgH#tHVKxrv#13K|A_hGu#O+LlR%#!1OZnj4)1JOI5f3RVCB

diff --git a/0.16/ops.html b/0.16/ops.html
index 58828994c34..c59a164c7cb 100644
--- a/0.16/ops.html
+++ b/0.16/ops.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/py-modindex.html b/0.16/py-modindex.html
index 77d9d99b879..02447132c04 100644
--- a/0.16/py-modindex.html
+++ b/0.16/py-modindex.html
@@ -224,7 +224,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/search.html b/0.16/search.html
index 3ba23eca75b..c08767e67ae 100644
--- a/0.16/search.html
+++ b/0.16/search.html
@@ -217,7 +217,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/training_references.html b/0.16/training_references.html
index 25aa52c78e8..b410245ec57 100644
--- a/0.16/training_references.html
+++ b/0.16/training_references.html
@@ -219,7 +219,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/transforms.html b/0.16/transforms.html
index ffd16a0f583..64f91678598 100644
--- a/0.16/transforms.html
+++ b/0.16/transforms.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/tv_tensors.html b/0.16/tv_tensors.html
index 28a0f69549e..b57526302ac 100644
--- a/0.16/tv_tensors.html
+++ b/0.16/tv_tensors.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 
diff --git a/0.16/utils.html b/0.16/utils.html
index 6686f32f0a0..4928fbbea41 100644
--- a/0.16/utils.html
+++ b/0.16/utils.html
@@ -220,7 +220,7 @@
           <div class="pytorch-left-menu-search">
             
     <div class="version">
-      <a href='https://pytorch.org/vision/versions.html'>main (0.16.0+9b13bcb) &#x25BC</a>
+      <a href='https://pytorch.org/vision/versions.html'>0.16 &#x25BC</a>
     </div>
     
 

From 1e56a6c674a737ee1edd4060a836583f86a921f4 Mon Sep 17 00:00:00 2001
From: Nicolas Hug <contact@nicolas-hug.com>
Date: Wed, 4 Oct 2023 17:45:16 +0100
Subject: [PATCH 4/5] update stable symlink

---
 stable | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/stable b/stable
index 20adfac7f5a..a59ac460838 120000
--- a/stable
+++ b/stable
@@ -1 +1 @@
-0.15
\ No newline at end of file
+0.16
\ No newline at end of file

From 79b84762a3263aab48e842b5a3398c86caa7d2f3 Mon Sep 17 00:00:00 2001
From: Nicolas Hug <contact@nicolas-hug.com>
Date: Wed, 4 Oct 2023 17:45:50 +0100
Subject: [PATCH 5/5] update versions.html

---
 versions.html | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/versions.html b/versions.html
index 99360511392..c2d6f356b21 100644
--- a/versions.html
+++ b/versions.html
@@ -21,7 +21,10 @@ <h1> PyTorch Documentation </h1>
           <a class="reference internal" href="main/">main (unstable)</a>
         </li>
         <li class="toctree-l1">
-          <a class="reference internal" href="0.15/">v0.15 (stable release)</a>
+          <a class="reference internal" href="0.16/">v0.16 (stable release)</a>
+        </li>
+        <li class="toctree-l1">
+          <a class="reference internal" href="0.15/">v0.15</a>
         </li>
         <li class="toctree-l1">
           <a class="reference internal" href="0.14/">v0.14</a>